peer_feedback.md

Presentation Peer Review

Here's presentation feedback from other students, organized by time slot.

Because we skipped empty time slots, you may want to check the feedback under your scheduled timeslot and also at the timeslot that corresponds to when you actually presented (e.g., if you presented 10th, check the 10th timeslot). Sorry for the confusion -- this was Nick's fault.

Tuesday

09:02

• They spoke very clearly.
• They explained there plots very well.
• They made very good visualizations
• Excellent graphics.
• Explaining the possible reason behind the spikes in the data for UFO sightings.
• They had very good plots that displayed interesting results clearly.
• They explained their data sources very well.
• Explaining the results
• Explained their methods briefly Many different interesting graphics Word frequency analysis Well-timed presentation
• they show a lot pictures and plots
• Good background for slides, used a good color scheme (viridis) to please colorblind bois.
• The visualizations are very helpful, and the order of the slides are well organized.
• They made lots of good graphs, for example the density map.
• projected information well
• Data visualization
• A very thorough overview of what their project is about
• Interesting topic, map,
• The topic is interesting and the PPT is easy to understand.
• Very clear when explaining
• Good Visualization especially the visualization based on map
• Their ppt is easily to understand and the plots are readability
• Interesting finding about July 4th, and good explanations about this finding
• Lots of graphics
• The group chose a very interesting topics. Their plots are very nice and easy to understand.
• Good job here and I was attracted by their topic. Their work is in detailed. The plot are nicely formatted.
• Their graphs are very clear and easy to read.
• solid visuals
• Graphs were clear, I liked that you correlated spikes with holidays or other events.
• They used loud voices so that they can be heard from the back.
• Each speaker had good volume and spoke clearly
• Nice graphs
• Lots of plots
• Top: UFP sightings. - Data source is clear.Background is clear Plot is clear and descriptive , well explained Interesting topics related locations of UFO sightings and population
• Interesting topic, insightful analysis, nice visualization
• A lot of clear graphs. Very well prepared not notes.
• Original topic and nice visualization
• Nice map visualization, linear regression, use of plots. (UFO)
• They used maps to visualize the density of UFO sightings across the country and use different colors to display.
• Clear plot to show locations of UFO sightings.
• • The visualization is very clear and fit the topic.

• Presentation is easy to understand.

• They have a clear layout of PPT and good amount of text on the slides also their have a variety of visualizations to show the summary of UFO sightseeing
• Humor about UFO's and Forth of July made nice crowd interaction. The graphs were also presented and explained nicely, and talking about other factors that contribute to 'UFO' sightings was informative.
• clear visualization in slides
• Their plots were clear

Improvements

• Maybe more details into the problem

• Speak louder maybe? That could just be an issue with the classroom.

• They should bring in other data and bring do more complex analysis (which it seems like they plan on doing with Sci Fi movie releases)

• Further develop analysis in report. Great job!

• For the scatter plot I think that you should move the year into the title and add the unit measurement for population at the bottom.

• I was in the front so I could hear them, but I feel like they could've talked a little louder.

• I think they could speak more confidently.

• Good

• Why does the location of the UFO reporting center affect the number of claimed sightings?

• the plots could be more specific and just tell us some important plots

• Maybe an interactive UFO visualization would be hella cool.

• I sat in the back and couldn't hear everything they said, so they could be louder. For the project, when they ask for feedback, they could maybe put the questions on the slides on which they are looking for feedback.

• I think they did a good job.

• shorten the presentation (ran over time)

• Include other data?

• Too much information&plots within 4 minutes

• speaking voice is a little low.

• They could adjust the scale of the plots

• Could use more number figures instead of all plots.

• They may explore some deeper questions about the data related jobs

• Projecting the voice

• I think their slides are good, and improvement is not needed. However, their voices are low; it's hard to hear them. Maybe, they can speak louder next time.

• They may show more data statistics and investigate more on other factors that affect UFO seeing.

• They could speak a little louder

• too much text, could not see some words

• The presentation was brief and I'm curious as to what other data or analysis you are going to be working on (next steps)

• Use colors that have better contrast. The green is a bit hard to tell apart.

• Go more into detail about their data sources

• Move the legend to a better location

• Could use a summary or clearer direction

• more eye contact with audience.

• They could have spent less time on each slide, and didn't have to talk about every stat

• voice is bit low

• Could consider the development of camera, satellite and Photoshop as factors

• I think the overall presentation is really good.

• They need to improve their graphs. Specifically, they should change the x-axis of scientific notation to regular form, or the audience/grader will not read readily.

• Show more details about words with high frequency.

• They might need to improve more analysis using statistics.

• The results they show in class are good but kind of scattered. It would make more sense if they can somehow connect the pieces of findings.

• Being louder, during the presentation is the only area I feel could be imporved.

• spent a little bit too long on changing slides

• I liked the plots but it would be better if they make more.

09:06

• The graphs were well presented.
• Relevant topic with insightful analysis.
• Had very descriptive graphs
• This group was well rehearsed and I liked one group member's joke
• Good explanation
• Analyzed trends by states AND job types Exploratory Data Analysis
• the data set is huge
• Pertinent topic to my life, clear simple graphics
• Good flow
• great audience interaction when talking about switching major to a popular major.
• scraping data,
• Connect the indeed data and H1B data
• Graphs are representing and clear.
• This group used a wide variety of data sources.
• Slides are well designed and clear
• Nice work too. I am also interested by their topic. Their data summary and plots are carefully made. However, we hope the description to more clear.
• Graphs were easy to read
• Their figures were simple and each had clear meaning
• Nice introduction
• Good breakdown of data
• Topic is very attractive
• Not available
• Interesting and related topic.
• The topic is interesting and the plots did answering the question raised.
• interesting presentation topic

Improvements

• Not to read from slides.
• Sometimes difficult to hear presenters.
• Their first slides had a lot of text that was difficult to read while trying to listen at the same time.
• I couldn't hear them too clearly so I think they could speak louder
• Could be a little shorter
• Delivery felt a little rushed
• more clear speaking
• It doesn't seem like an original idea, lots of words on some of the slides, less monotone would be nice, maybe more websites scraped
• Maybe try to connect to more data
• Too much content on the second slide, could be more condensed.
• you could collect more data
• There are too many words in the ppt. And the explanation for the graph is not clear.
• They could speak a little louder.
• The focus of their project is too specific and they can extend their findings by looking for more perspectives in their topic. (e.g. number of students graduated from each major)
• Showing locations in maps may be better
• The group add more content on how different major students that final become data scientist and show how competitive the data scientist job is against other occupations.
• to many words on the powerpoint speak more slowly
• Speak louder and more clearly
• Speak louder
• Could use more explanation of charts, fewer animations
• Too many words in powerpoint. Eye contact and voice
• Not Available
• Maybe explain more about difference showed in plots.
• I am not sure if this group provide information about the data source. Maybe they should talk more about where they get the data.
• lack of eye contact with audience

09:10

• Speaking clearly.
• The topic is very useful for everybody in the room.
• They considered many different aspects and used multiple data sources
• Good presentation with informative slides, esp. last two slides.
• The graphs looked well made and easy to read. They conveyed their purposes well.
• Good explanation
• using programming
• easy to read slides; very comprehensive and detailed; cute bits of humor
• Good Graph, lots of useful info.
• The fonts are large enough for everyone to see. The graphs are also meaningful in terms of delivering the content of the project. The project topic is also helpful for us future job seekers.
• explain graphs well
• I liked some of the analysis a lot: you guys aptly connected the most common words to current events that happened in the 60's (space race), vs "woman-director" frequency indicating the success of feminism in the 21st century
• They did great by circling the most representing/meaningful plot on the slides
• map, word cloud, matrix scatterplot,
• The data analysis is very well.
• Very detailed analysis of each graph
• Very relevant topic, clean graphics
• Their plots are very well organized and easy to understand.
• The theme is intriguing and they did a good job on finding the theme trend over the time period. I like their word frequency plot very much.
• The topic they are looking at is very useful
• delivered information well
• I liked the abundance of graphs and data, good analysis
• They introduce their topics and motivations well; Their graph is clear.
• Useful topic, detailed analysis
• interesting topic
• The research was pretty comprehensive
• Help information. It gives suggestion for the skills one needs to learn. Nice bar plots. Well organized job market for STEM majors
• Interesting topic
• The figures in the slides is very clear and easy to understand. Words inside the figure is also big enough to read for audience in the back. Give a big picture about the data in the beginning which makes help audience understand the project.
• The data visualization is very cool.
• I really like their topic since it is very related to stats students and I think they did a great job on summarizing their findings from the data.
• Breaking down the information about the project was done very nicely, and one of the presenters had great projection of their voice (Patrick I believe, if my notes are correct)
• As I want to do data analysis job in the future, I am very interested in this topic. They includes the basic information that I want to know.
• slides are in good organization and clear structure
• Their topic is helpful to statistics major student. Plots are easy to understand. Slides are well-designed.
• Their topic was very interesting to us.
• Grouped bar is a good choice to analyze programming languages by different levels of job positions.
• This group did pretty good in combing two sources of data: one is from web scraping and H1B salary dataset. And they answered some basic questions of finding a data science related job, such as the major, language and salary.

Improvements

• Too many small graphs.
• Maybe differentiate the keywords between each different type of job for Data.
• Order the bar graphs for Senior, Entry, and Internship in order of experience level (Senior, then Entry, then Internship or the reverse)
• Explanation of keyword data was confusing.
• For the graph on language requirements and job positions, I think it could be helpful to also group languages against job types to see how popular each language is in general.
• Could be a little louder.
• plots could have more detail
• speak a little louder; potentially too much info presented all at once (a little difficult to digest)
• tell us more about how to improve their analysis.
• The pace was a little fast, and they could speak louder.
• explain more about the background information (H-B1 form?)
• Spoke very quickly--could slow the pace of presentation and simplify graphics (see Tufte's principle of the data-to-ink ratio)
• For the part about what kind of movie was trending during a period of time, it would be more clear if they could have using a plot instead of indicating the artists.
• There are so many graphs. Also, for word cloud graph, I do not really understand the explanation of the word.
• Too much text on one slide, speaking a bit too fast
• Work on projecting voice (speak louder)
• They put too much information on the first slide. It's overwhelming and hard to read.
• I think they can explore more on other factor affect the theme changes over years, such as human capital, society concerns, fashion and etc.
• On the first two pages of their slides, they have included to many words so the slides seems a little crowded.
• have less text and more visuals
• Volume - could be louder for those in the back.
• Too many words on the first two slides. They can put important information or key word instead of whole paragraphs
• Slides are quite wordy; question could have been answered better.
• graph small and not very clear, no eye contact
• Could reduce some words on the slides for presentation
• I think overall it is on the right track.
• Make results more clear.
• Too many words in the slide, which makes audience feel overwhelmed. Maybe try to used bullet point to capture the key features.
• Maybe they should include more detailed graphs about the data.
• I am sort of confused is the findings in their PPT only apply to international students? And also I think it is a good idea to consider sample sizes when they choose the sample states since population may have an effect on the number of job vacancies.
• There was a bit too much information on the first slide, making It hard to grab the major/main points from it. The other members could be louder when presenting.
• I think there should be a reason that companies prefer CS and engineering to statistics and I think it is a question which could be explore more.
• bad time control
• They should talk more about what they will do next.
• It would be better if there's more info about data analyst.
• The color for bar graph of STEM majors ranked is too bright, I think a color scheme of low brightness is better.
• They could dig more on the data from web scraping, for example do some NLP analysis on the description of the job requirements so as to answer more detailed questions on what are the job marketing exact focused.

09:14

• The word cloud between two different eras of film was very insightful.
• They made good conclusions from their word clouds
• Prevalence of doctors was a good exploratory analysis paired with diabetes analysis.
• I think the findings and analysis were interesting.
• I liked how they broke down productive actors and directors historically by year
• Their data analysis seems extensive and well done.
• Analyzed movie trends by country Analyzed keywords in different decades Explained trends well
• They use word cloud.
• a lot plots and picture to lead us
• creative visuals to display information; good use of extra graphics to direct audience attention to appropriate spots on charts
• The word maps are very cool
• nice and clear graphs
• They did well in pointing out the key words in the word cloud, and they limited the number of words in the word cloud to make it clear.
• Their graphs about trending words.
• nice visuals
• great visual on the key words
• They did well in going through every plot in details.
• data analysis, graph comparison
• The topic is interesting.
• Great explanations and body languages
• Very nice way of showing the word frequency in movie industry.
• They clarify their topic and what they want to do about their project.
• Explanations about the word cloud findings and related actors over time
• thorough analysis on the word maps
• Good analysis
• They had nice map plot and word clouds.
• Two of the speakers is introducing the presentation in a low volume and cannot really hear it from back. However, the plots are carefully made and their factors are good for analysis.
• Their slides are simple and easy to read.
• conveyed information well
• I liked the word cloud and pictures, great way to visualize the content
• They did a great job of explaining their findings in a way that we can understand
• Very vibrant and colorful figures
• I like how they saw the trends throughout time.
• They have a lot of content and most of the graph looks good. They also clearly state their content.
• Different types of plots
• Interesting topic, detailed analysis
• very specific and professional
• Density plot on map, trends, wordcloud (Movies)
• They used WordCloud to search keywords for movie trends They showed historical productive actor/actresses vividly
• Interesting topic.
• The group choose great graphs to present. For example, the word cloud provides a great visualization for the key words in the data set. In addition, the group also provides a lot of images for director for reference.
• Good visualization
• I like their visualizations and the keywords that they presents I think they are very representative.
• I like the word clouds and the breakdown of them
• The plot about time and genres is good.
• neat talk
• They got some interesting results shows the top keyword of movie type in some years.
• Their topic was very interesting especially the actors.
• It is clear to use word cloud to show key words.
• They provided a great figure in illustrating the trends of genres of the popular movies and NLP analysis for key words of the movies.

Improvements

• The faceting maybe isn't such a good idea given the limited amount of time.
• More types of visualization and analysis
• Somewhat difficult to hear presenters.
• I think the graphs were kind of cluttered and hard to read. I think the last two slides could be better represented by point plots where you label the top score in each of the different categories as the director/actor's name.
• They looked at the slides, not at the audience. Could have talked slower in order to be better understood
• One member didn't seem practiced and read off the slide alot.
• Face the audience when speaking instead of reading off the projector screen
• Maybe they can speak in louder voice and face to audience!
• plots is not clear to see
• some awkward slide formatting, space could have been used better; speak a little louder; what does flowchart with directors represent?
• A little too many words on slides, graphics text a bit too small and hard to read, spoke way too softly for me to hear from the middle of the room
• improve more info from keywords you found from dataset, can you make model to predict future movie trend?
• They could face the audience more when they are explaining the word clouds. Also they didn't coordinate well enough when changing the slides (sometimes they end up going back and forth). For the project, the timeline that shows the movie heroes, it's not informative (or intuitive) in terms of data analysis.
• I think they can speak louder during the presentation.
• speak louder
• Can add analysis of the yearly change of the key words in movie market.
• They could be louder. They should spending more time talking about what their project is(objective, motivation)
• All the graphs look similar.
• The scatter plots are a little confusing to me.
• Some plots are very small, can't see the details in the plot. Presenter can speak louder, it is very hard to hear in the back row.
• They may need more analysis plots shown on their presentation ppt to explain their data analysis work in python
• Facing the audience
• speaking too fast and too softly
• first graphic was a little hard to see, but the trends were clear. Make the charts bigger
• Maybe they could speak slower and split the line plots into two slides so that they are bigger to and easy to read.
• Hope they can add more explanation details in introducing their report. The topic is more academic, but it should also easy for normal people to undertand.
• It was difficult to hear parts of their presentation, they could speak louder
• spent too much time on a plot, be quick and to the point
• Moved quickly, text/graphs on slide 3 were too small to read
• This group can speak slower and also face the audience more as well as not put so many graphs on one slide
• The project narrative is a little unclear, however this is probably due to the difficulty faced in hearing the presenters. Speak a little louder.
• Plots were a little confusing.
• In one of the slide, there are two many graphs on that, it is hard for people to see so many graph at one time. They can choose some typical graph.
• Plots can be hard to read
• They could have avoided to read from slides, spent a long time on each slide
• Voice is very small, reading
• Should spend more time on stuff other than explaining key words. No further goals.
• They may be more clearly state their mission statement and describe their data source in report.
• Maybe more clear results.
• Some of the graphs are too small. It's hard for audience to see the words in the back. One possible solution might be try to avoid putting multiple slides. And some of the words does not fit well with the background since they are all black. Maybe changing the color of the words to make it more obvious for audience to read.
• After the presentation, I am still confused about the topic. Maybe they should add more background information about this topic.
• For the historical productive actor, I think there might be more things they can do such as what makes those actor popular? Is there a type of movie that they tend to play?
• The slides seemed really busy, in that the were at time a bit much to look at. The facet grid graph was good, but also a bit much on one slide
• I think it is better to analyze more obvious words in the word cloud.
• plots are hard to read due to small size label
• The topic goal is not clear. Some plots seems dispensable.
• I think their presentation was good enough
• The scale of color for counts of movie is not clear. It is necessary to use the better scale to map the color scheme.
• They could improve more if they compare the trends of genres or keywords of the movies for different countries so as to provide more interesting results on the preference of the movie market in each region.

09:18

• Graphs were very good
• Looking at different factors that may not seem relevant to Diabetes such as income, pollution, etc was very interesting, especially the income.
• They considered many different factors in their analysis and used lots of data
• Identified very interesting disparity between diabetes rate and income. Had a very good summary slide.
• Their plots were well made and needed little explanation to understand
• Explained methods Summarized what they've done and what they plan to do
• Good use of density plot!
• the topic is interesting disease and pollution picture is good
• They had an interesting topic with the data well plot out.
• good gesturing and use of space in front of classroom
• The overall fluency is great. We can see they have practiced before presentation. Good summary as well.
• Good choice of graphs, just wish I could read them smh
• introduction is clear, showing us the topic and factors of their problems.
• This group had a summary slide. I think this is really helpful in terms of getting feedback from peers.
• The graph about diabetes rate.
• nice visuals
• This group did a lot of research and statistical analysis. The presentation was very focused and the topic was clear and concise. There was a clear purpose and conclusion to the project.
• Excellent. Beautiful scatter plot.
• They were loud and clear. Great indication of importance on the plot. Great reference from outside sources to support the last plot.
• The topic is clear, and I can see this group has strong background on their topic.
• The topic is very interesting and great explainations
• Clear summary on their goal, method, result, and conclusion.
• They have a clear division of labor, each of them explain one plot of their ppt.
• Clear questions and strategies to answer them
• Having description of plots on the slide itself
• Graphics were clear
• They had a lot of statistical facts and very well-defined goals.
• The group number are introducing their project clearly. The plots are good-shaped and include useful details. we found that the statistics and the trend they get are clear.
• solid visuals
• Very insightful plots, I liked the use of scatter plots. Nice summary, I liked hearing about what you plan on doing.
• made use of bullet points
• Interesting topic (health and environmental factors) and synthesis of multiple types and sources of data. Not an easy topic
• Nice scatterplot
• Lots of different features used in analysis
• Topic: data science jobs. The 1st plot is clear and persuasive. The explanation is clear and understandable.
• Meaningful topic, detailed analysis and presentation
• Voice is loud and clear
• Nice visualization
• well prepared presentation nice plots California health disparity
• They analyzed public health issues from a statistical view (e.g., odds for having doctor in cities) and this topic helps general people well.
• Interesting topic. Clear results.
• Display the questions they are going to present at the beginning which gives a big picture to audience. Give clear explanation to the plots.
• The visual aid of this presentation is very interesting and cool. The explanation of the topic is clear.
• clear slides,organization and structure of content
• I have no feedback for what was done well by this group. See reasons in improvement category
• The density plot work well.
• good interaction with audience
• They shows some plots and explain well.
• They explained their finds very well.
• Their question (What patterns exist between external factors and health measures) is interesting.
• Very good topic in finding the relationships between each specific disease and income/pollutions/doctor density of the region.

Improvements

• Speak more clearly
• Speaking louder, especially in a large room.
• Looks at the different potential independent variables for one disease
• Very quiet and hard to understand, could have spoken louder.
• They spoke too quietly so they should speak louder
• Slides were fairly bland
• Maybe they can speak in louder voice! The density plot should be adjusted to the center of the plot!
• give more detail on your data sets
• They could speak a bit more loudly
• speak louder; many of the same plots repeated
• Group members can speak louder and slowly since I can not hear clearly at the back. The size of text on ppt is also too small.
• 1st speaker way too soft, slides bare bones with lots of words, can't read text on graphs (that's inexcusable, it's easy as set_context('talk')
• bigger plot, cannot see on the back
• They could put less graphs on one slide so that the words are larger.
• Maybe they can make their graphs larger in the slides.
• speak louder
• I would've liked to see the distribution of the data throughout the state, and not just with the top and lowest 5 cities to better understand if location affects or is correlated to the results.
• 1. Please speaker louder. 2. text in figure is too small. 3. should explain more on concept(eg. score)
• The first plot is very confusing to look at. The x-axis could be more clear and organized for audience to look at.
• I believe they are good enough.
• This group could explain the the stack plots a little more
• The speakers can speak louder. Can't hear them from the back row. The label of the plots can be larger so that it's easier to see. Can explain the jargon better.
• They only use bar plot and scatter plot to explain all their analysis results, it's a little bit single I think.
• Explanations of the scatter plot
• speaking too softly, more explanation on the contour/scatter graphs
• I felt like the message of the presentation wasn't clear. Hypothesis and conclusion could be better stated.
• Their plots are small and hard to read, and some of the bars had different colors which made it confusing. I thought different colors represent different categories at first, but I didn't see any legends, so I assumed the colors didn't represent anything.
• They may focus more on creating data summary with different factor that affect the nutritions inside a food product.
• less text on slides
• Volume - difficult to hear.
• speak a bit louder don't use to many side by side graphs Try not to read off the powerpoint
• They should've spoken a little louder.
• Maybe speak a little louder
• Scope maybe too broad (all diseases)
• The voice is a little bit low to hear.
• Presenters could have spoken louder and clearer.
• Graphs are kind of messy
• Could consider race and gender as factors
• speak louder.
• Their plots should be modified. Specifically, axis labels and titles should be enlarged. Although they are too many intervals in the plot, they should consider hiding some interval labels so that the plot can be easier to read.
• Maybe try to split the graphs into multiple slides. Putting multiple plots for the same slides makes it hard for audience to see from the back, especially the words in the coordinates.
• I don't get the meaning of the first plot.
• The group did a great job and they know what to do next. Maybe try to take a look at other disease and see what patterns do these disease have in common?
• Being louder! I could barely hear the presentation at all, and missed much of what the project was about
• I wonder where it is possible to explain how Chronic Obstructive Pulmonary Disease is linked to smoking.
• too many words on slides and hard to read
• Some members’ voices are too small. Students who sit behind can barely heard what they say.
• There were many plots that used same method.
• It is not clear about the scatter plot graphs (contour maps), so should give more explanation or detailed legends about these graphs
• They could improve their visualization, eg. using google map to show the disease rate disparity at different regions in CA. Also, I recommend they only focus on one kind of disease and consider more factors associated.

09:22

• the spoke very clearly
• They explained the trend in their time series in a way that makes sense.
• Very interesting topic with data from many kinds of sources (history on Wikipedia, ingredient lists, and numeric nutrition data)
• I think that their presentation idea was very interesting in terms of the idea that they analyzed. Their heat map of ingredients worked well
• Interesting topic, good transitions, great analysis. I liked how you identified the top ingredients and analyzed them, including their historical data.
• I think that the additional research into some of the anomalies was well done and helped clarify the graphs.
• Their introduction had a nice narrative.
• Their presentation was well rehearsed
• Slide design and explanations
• Ingredient frequency analysis Explored possible explanations for trends
• They make a clear and organized presentation.
• the best introduction until this time, clear reason and logic, virtual plots on infants fomula is interesting
• Their graphs brought light to an interesting disparity in ingredients.
• good flow to presentation in beginning; good transitions between slides and graphics (especially between stacked bar charts)
• Attractive topic! Graphs are clearly represented with detailed explanations.
• Good speech volume! :) Nice pictures/animations on the same slides
• graph well support their ideas
• They spoke loud enough for me to hear. And the plots are really helpful when they use boxes and zoom into parts of the graphs to give insights.
• Good visualization and clear expression.
• They relate the results they found with the news.
• very well descripted
• I liked the deeper domain knowledge exercised here. Good job of communicating analysis of FDA standards.
• It was an interesting topic, and the group spoke clearly and was organized. The group also mentioned the next step in their research that they will explore before completing the project.
• Excellent (good introduction, interesting topic, speak loudly)
• Loud and clear.
• great introduction, web scraping, multiple data resources, depth
• The PPT is well written.
• the presentation was conveyed clearly with its main point
• The topic is very interesting, and the results are representing.
• Good introduction. The speakers are audible. Very interesting topic.
• Their plots shown on the ppt are diverse and easy to understand
• Nice introduction comparison; clear tasks; loud and clear
• Presentation was very well-prepared and had good flow
• They had very interesting and educational topic.
• The topic is interesting and their goals of analysis are meaningful. The speaker voice could be higher. The introduction of one speaker is not quite clear. The plots are good.
• They spoke loud enough for me to hear all parts of the presentation.
• conveyed information well
• Good volume! I like the context and establishment of topic. Interesting topic as well, I like that it's relevant to the presenters' majors. I like the summary as well.
• Results are significant, and the variables are suitable.
• I really like the use of pictures This group spoke at a reasonable pace
• Interesting problem and good introduction that presents the primary question.
• Spoke clearly and had a clear narrative. Interesting plots with good spoken explanations. Great job.
• They talked about a relevant topic.
• They make the summary clearly, and they say thank you to the audience.
• Innovative data source (Walmart)
• Topic: infant food - The plot is clean and easy to understand. Very interesting topic as the food is related to infant products. Good control of time.
• Very meaningful topic and insightful topic, using a lot of data sources (API, web scrapping)
• Interesting topic
• Nice flow
• fancy barplot. dot plot Nutrition of baby milk
• They analyzed their professional knowledge very in detail and summarized the result clearly.
• The topic is interesting. They talked about how they get the data.
• Being loud at projecting their voices!! So many groups I could not hear. I thought the break downs of the information displayed (nutrition facts and graphs) and the summary of the presentation was great.
• clear illustrations to get across key concepts
• They have nice conclusions.

Improvements

• maybe clearer graphs, less texts, more picture
• The slide titled "product percentage of top 30 common ingredients" was kind of hard to look at, especially with the larger words.
• Include more deep analysis of brands with outlier ingredients and things like that
• would like to see them do more going forward in analyzing the data they have
• Unreadable x-axis on Nutrition Fact graph.
• I think the labels for the Composition of nutrients in infant formula are hard to read. Maybe consider breaking it up into subsets to easily see the bar for each category.
• For the scatterplot of product percentage of top 30 common ingredients, they should have labeled each point to make the graph easier to read.
• I they could spend more time on each image, it seemed too quick paced
• Good
• Slides bland, but interactive Why only Walmart?
• The plot has an unadjusted x bar for the Composition of nutrients in infant formula plot. It seems like they have a continuous instead of discrete x axis.
• more introduction to the data sets
• They sounded nervous, possibly rehearse their lines more
• good speaking volume; difficult to read x-axis on stacked bar charts;
• The introduction part is too long compared and they should talk more about data analysis part.
• one slide looked cluttered and unprofessional, and a graph had some unreadable text with too much information on it. Didn't answer my question well lol, but maybe that's my fault
• show us what are inside their dataset
• They could avoid using too much overlapping of pictures on the slides - just put them in separate slides if they were to cover the whole slides. For the graphs, avoid stretching so that the words are not distorted.
• Maybe they can change the graphs about nutrients to show x scales.
• lot of information on a slide to show
• Charts were hit or miss. Consider interactivity to decrease the number of xticks you need in charts with large amounts of brands or nutrients.
• It was difficult to read the axis labels on the plot for "Composition of nutrients in infant formula (all)" and "Composition of nutrients in infant formula in water".
• The axis sticker in their figure should be more clear (1st figure).
• They seem like were not prepared when talking about the line plot.
• The plot of composition of nutrients is a little hard to read( the white lines for example
• This group is not familiar with their material.
• labeling on the points in the plot could be more clear.
• I think it would be better if the font the bigger and using more viewable colors.
• Some plots' label is too small. Didn't introduce jargon clearly.
• They may explore more questions about their topic
• First graph x-axis not clear what it is
• Some of it was a bit fast. Slow down so we can digest the topics
• Some of the plots had too many info on them and look very overwhelming.
• Some of the word frequency plots can be more clear. One of the speaker should be more familiar to the topic.
• Some of their graphs are difficult to understand for someone who does not have much knowledge about this topic.
• less text, more visuals
• Moved quickly in the beginning, but it was still easy to understand.
• The plots are not clear enough, and the label is vague. They can only look at the big difference in the data.
• Maybe choose a few ingredients to focus on the graph don't put pictures on top of graphs
• More explanation of how they got the data.
• Some slides were a little cluttered (not too bad though).
• They could try to speak louder
• Their first graph is not clear and there are too many elements in the second plot. they can focus on some important elements and show them on graph.
• Prefer more statistic plots
• I really like this presentation :)
• graph not so clear and not very fluent, prepare more
• Keep doing the good job, can apply some statistical models.
• The graphs may be confusing and puzzling to the audience (x-axis label too small, percentage axis should *100(%), or it will be proportion). They need to adjust their plots so that they will not lose many points in final report.
• Maybe more background information is needed.
• Unfortunately, I don't have any areas that I can suggest for the group to improve on.
• word color is too close to background so it is hard to read
• Label in x-axis of nurtrition fact graph is small and not clear, so I think make it bigger is better.

09:26

• well defined problem
• The time zone and posting time slides were very well thoughtout.
• They gathered lots of interesting data about each post (and the last slide)
• Really liked how they drew a connection between the proficiency of players and how toxic they were to the gaming community.
• I think the graphs were clear and easy to comprehend.
• Slides had a good overall design
• Their plots and methods were well documented
• Slide Design
• Interesting topic! Explained goals
• It seems like they spend a lot of time on data collecting and their data come from multiple sources! Good job! Interesting photo of Nick!!!haha
• gane popular is interesting and the intro logic is good
• made an interesting parallel between the level of player and quanitified their level of toxicity
• A+ group name; nice theme to slides; interesting project
• They represented their project with nice graphs but they should improve their consistency and fluency.
• ppt is attractive.
• They listed out their project goals in detail.
• Fresh topic.
• good context
• Intriguing topic! Good application of domain knowledge & web scraping methods.
• Clear objective and methodology. Cool name.
• Good (interesting topic)
• interesting topic
• Interesting topic, multiple data resources,
• The data resource is reliable and the PPT is well written.
• the topic is creative
• Very clear and organizing .
• Plots are well made. It has big labels and clear legend.
• Their last plot is impressive!hhh
• Diverse visualization plots
• Slides look cool;
• Interesting topic
• The topics are intriguing. The step of their analysis is clear. The plots and introduction are in a clear way.
• solid visuals
• Interesting topic and analysis. I liked that you mentioned the differences in what you expected vs what you observed. I like the plots where you compared the time.
• Interesting topic, and use the NLP well.
• they summarized the data well
• Excellent looking slides.
• Very clear slides
• They have a eye contact with us.
• Found trends in noisy data
• Topic: toxic word - Clear explanation about the project goal and motivation.
• Interesting topic, have some insightful visualizations
• cool and clear graphs
• Original topic
• Complete interpretation on intro and goals Nice slides theme applied Nice bar plots. Games toxicity
• Their slides are very visually attractive and interesting. They introduced their tasks and target questions completely and clearly.
• Clear results
• Interesting topic; Listed out all the games they are going to discuss. The graphs are informative.
• The way they talked about the background is very clear and interesting.
• I though the approach/process of the project was explained well. The tier graph was also informative (results vs expectations)
• presentation is in clear structure
• Their topic about game is very interesting and their goals in each parts are specific.
• The topic was good.
• Clear explanation about the description of the topic.
• Showing different graphs simultaneously was helpful for understanding the trends.

Improvements

• the presentation colors were not very good

• In the slides, the dark text on dark background was kind of hard to read. Also maybe explain what rank is in terms of the game.

• Using more of that data (how good a player is other than ranking, etc)

• i thought their bar graphs could have been more clear on what they were showing

• I think that the time postings by region isn't that helpful because of time regions. Maybe consider adjusting for that and seeing if there is a difference?

• However, some slides were difficult to read because the text color and background color were too similar. Presenter read notes off hand.

• They didnt seemed like they had rehearsed

• Looking at the slides less, increase volume

• Some group members need much more rehearsal

• They may avoid giggling and communication between group members during presentation.

• the 'big 3' explaining

• Some more data to back up their claims. One of their slides dealt with the times these players were most frequent but I wasn't able to hear how relevant it actually was. (may have also been my poor hearing)

• text was difficult to read on slides; a lot of information to digest with last few bar charts on a single slide

• The power point's background is too dark colorful and the text size is too small. They could improve it by changing the background image and make the text size greater.

• show us more findings, not only the frequencies of interactions.

• The word cloud (and some other slides with dark color of fonts) were hard to see. They also switched back and forth when talking.

• The graphics are kind of messy including the x axis and the format of the plot. Maybe try to use another kind of plot.

• speak louder and more clear, text color made it a bit hard to read

• Slides were dark/unreadable in some cases, and the contrast with the "Interesting Results" slide was jarring (plus that graph was pretty spare). Check out Tufte's principle of the data-to-ink ratio

• The color scheme on the word cloud was difficult to view. For the "Toxic Words Freq of Players' Tiers" plot, I wasn't clear on how many words they studied or for how long they studied players' behavior and comments.

• 1. Slides are not well-organized. 2.cannot see text in slides clearly(similar color as background). 3.the male is not familiar with his slides. 4. should explore more. Their analysis are not enough and are too simple.

• include the unit of frequency in the chart. How many times a game or an hour?

• the information about game (like the rank) may be hard to understand for people who never played the game, the post

• Can include more questions.

• could practice more to be more fluent

• It's safer to avoid using pie plots.

• The slide can be designed better. It is very hard to see some of the words. The speaker can speak up a little bit. Can't hear the speak. They can be more prepared as there are some stops in their presentation. There is a very obvious outlier in one of the plot and the group could elaborate more on the outlier.

• The color and design of their ppt it's not good, which is hard to read. Also, they don't have enough plot to explain their results.

• Toxic Words Freq might be better to shown as scatter plot without line since name is categorical variable

• some slides too dark to see; not even analysis on word map; last plot not sure what the difference is just from the label without verbal explanation

• Could be better prepared and talk more about the graphics and analysis

• certain plots can be reformed to provide more details.

• spend more time explaining visuals

• Difficult to read slide text, could fix colors, size, formatting.

• I don't understand the terms in the reports. More explanation please.

• use colors that have a better contrast use a loud voice

• It was difficult to hear the presenters.

• They should practice their presentation more

• I can not see the first graph clearly, also one of them should not put his hands in the pocket. He'd better use some gesture instead of just putting hand in pockets. They should not laugh during the presentation.

• Plot backgrounds could be clearer

• • Unclear data support.Prefer more data visualization, such as statistics plots, data analyze.

• They could have rehearsed it so they would not forget what to say

• add in more content

• A little more definition on 'toxic words'

• Can rehearse some more.

• One group member may prepare a bit more since this person did not speak fluently.

• The words and the background color are very similar, which makes me hard to read the slides from the back. It would be better just changes the color to white.

• The plot of "popular games use more storage" is a little bit weird.

• Being louder and practicing more would help improve the delivery and flow of the presentation

• one of the speaker moved too much and it was distracting

• They can do some analysis about how many people like trash talk in game by sex by age.

• The presentation needs to be more well organized.

• I think they need to give more statistic graphs to analyze their data sets.

• Font should be lighter on dark backgrounds. I also would have appreciated more motivation on why they chose to analyze certain variables, like time active.

09:30

• Explanations were clear, and everything made sense with their train of thought.
• Analysis on many levels of depth (the whole platform, by year/genre, and on individual games)
• I liked how they drew a correllation between the storage that games required and the popularity of them (measured by scale). Their analysis on reviews and players and when they peaked was very interesting.
• Interesting topic and good analysis!
• First presenter was compelling and had a good style of presenting. Overall, examined interesting ideas in their topic.
• Slide Design
• Explained goals and methods clearly Explained their questions
• They use different kind of plots to help visualize the results!
• explain on the topic and plots
• They stream rolled their way through the presentation - they know their stuff!
• creative flowchart in the beginning
• Group members used various kinds of plot to fully represent their topic. Also, they talked very natural by using eye contact and body language in the presentation.
• Very nice theme/font and topic. Very cool learning about the toxic word frequency,, and I loved the double histogram plot: it had a gorgeous color scheme
• graph shows the proportions clearly. logic is clear.
• Use different method to access the data.
• Their PPT is attractive and their topic is interesting
• well descripted
• Analysis was spot on and in depth for the questions asked, and visualized neatly overall.
• The group did a very good job in presenting their topic and introducing it to an audience as if they have no prior knowledge of the topic. They were clear on explaining their methodology and hypothesis.
• Excellent (beautiful slides, interesting topic, interesting analysis)
• Nice visualizations on the Max players.
• great visualization and analysis
• The statistic model is good and interesting
• Plots are in various forms and clear
• They speak loud and clear.
• Very interesting topic and the group did a good job introducing the topic. The data source is very reliable. Speakers are audible.
• Their plots shown in the ppt is multiple and diverse, they also give good explaination of their project.
• loud and clear
• Good presenting skills, graphics clear and good analysis
• The method of analysis are in a clear way. Plots are useful for introduction. Good work.
• They spoke loud enough and confidently through the presentation
• explained visualizations well
• Great graphs, very clear and easy to understand. Great analysis!
• Good data resource, interesting results
• I really enjoyed this presentation I liked how you defined any jargon used I also liked how you moved sideways rather than turning your back to the audience
• Interesting question of whether reviews drive plays or plays drive reviews.
• Great introduction to the project objectives.
• Very clear speaking
• The first speaker really have a good eye contact with audience, and her speaking speed is really appropriate.
• topic: Gift/game.Clear explanation about motivation. Enough data analysis support. Good time control.
• Interesting topic, insightful analysis
• speaking clearly
• Nice introduction and explanation of motivations
• Game catrgories explained very well. Consider the storage requirement, good. Steam Games popularity
• They used wonderful graphs to show their tasks step by step Graphs are clear and easy to read (informative).
• The flow is smooth. Introduce the pipeline for the project, which gives me a clear idea about the project and what they are doing. The volume is voice is good enough to pass to the back.
• The visualization is cool.
• They have great visualizations and interesting topic
• There are diversity kinds of graphs.
• did analysis deeply, especially machine learning methodology
• Their topic is very interesting and they shows some meaningful plots.
• The ppt was very clear and exquisite
• The topic was really interesting.
• Clear description for data processing
• Analyzing the storage requirement of the popular games is a very interesting point (but the graph could improve with an 'increasing value' x axis).
• They described what they were skeptical about.

Improvements

• The time series plot of the PUBG players and twitch viewers would be easier to read if the ticks were larger as well as the plot in general. Don't take that as a negative though, as the plot still made a lot of sense.
• Graphs that more clearly show their conclusions from what they found (namely the storage slide)
• I would like to know more on where they're going after this
• Order x-axis of GB size graph ascending or descending, not mixed
• Nothing that I can think of.
• Bringing the results together
• Why just focused on Steam?
• For the plot popular games use more storage, it seems like they have an overlap histogram. Is it normal? Maybe change this plot a little bit?
• summary is too much
• I didn't see much they could've improved on
• pie charts are bad
• Their plots should be improved by adding x and y axis and make the plot larger.
• Some text too dark, one speaker too soft to hear. Couldn't read word map, it was hella dark.
• show us the dataset to let us have a better understanding
• The backgroound color of the slide makes it hard to read. The plot is simple.
• Maybe they can introduce their dataset
• graph text was hard to read
• I'd have liked to see a more focused overall question. Analysis answers a bunch of interesting minor questions but I had trouble picking out a well-developed central theme other than "we wanted to analyze Steam."
• I was a little unclear on what their final conclusion was for the study.
• 1. avoid pie plot 2. the x axis sticker for the first figure is strange, but they didn't explain it.
• improve the visual on the storage of games.
• Too many words in one ppt. Two graphs about clustering have the same information.
• axis labels could be made larger
• This group can explain the the clustering dendrogram a little more.
• Some plots are not clearly explained. The label of some plots can be enlarged for visibility.
• It is better if they can use a bigger data set
• storage graph x-axis labelling is confusing
• Can't think of anything
• Some of the plots can be more clear.
• They can explain some of their graphs by putting explanations next to the graphs on the slides.
• was hard to see text
• Would be interesting to see a graph of changing genres over time instead of 2 pie charts.
• The plot has too many information, and may use more variables. They can find more resources.
• Color on pie charts should be the same for the same category. A little hard to interpret frequency analysis plot.
• The figures were a little challenging to digest.
• I think maybe they could have to separate the bar graphs
• I can not see the graph 'max players' vary well. There are too many elements in this graph. They can choose some elements that are more important and put them in graphs.
• Lack transition between topics
• They could have avoided to use pie charts as they don't bring a lot of insights
• too many words and bit messy, graph is not very clear
• I think overall it's good.
• They may change their axis label size and change it easier for the audience to read (000 to thousands).
• Maybe convert the pie plot to the bar plot. It's hard to compare when the percentages are very similar in pie plot. In some of the plots, the characters are too small to see from the back.
• Maybe add more pictures to introduce the background rather than using a lot of words in the PPT.
• The topic that the group choose is what makes a game popular however based on the content present they seem to go on the opposite direction(they looked at the features of popular game) maybe try to take a look at a few popular game and look into the specific features of these game to see if there is anything in common.
• I think there could be analysis about why other games decreased a lot after 2014. There should be information about how and why the favorite games type change online.
• explanation of plots on slides is not clear
• They can do some analysis about what keywords are usually mentioned in popular games.
• The presenters can be more confident.
• They only talked about #1 game but it would be nice if they show top5 or top10 games.
• The visualization for frequency vs storage is not clear since some bars are overlapped. My idea is to use separated bar charts in the same figure.
• First, they could analyze more on the attributes of the top games, eg. looking at the duration of their popularity, and see if the duration changes over their release time. Second, I recommend they investigate the type or popularity of the top games in different regions, so as to figure out the preferences of different game markets.
• The graphs were a little confusing to understand.

09:34

• very interesting graphs

• Their exploratory work I would say is very good. Not to mention that the topic is still very relevant to us being in Northern California.

• Interesting types of analysis with clustering

• Beautiful graphs. Insightful analysis not seen in class. Good job creating and testing hypotheses.

• Talked loudly and was easy to understand

• Their more nonstandard plots made for a more interesting presentation

• Explained questions and importance of their project Interactive visualization

• Good use of algorithms!

• SQL data is strong evidence

• Their twitter scraping was cool.

• really nice-looking plots; seem knowledgeable about their data and focus for their project

• Their plots seemed very attractive to me and they did a good job in the organization.

• Very cool that they got their data from STEAM!!!! One of the bar graphs looked nice

• interesting graph from twitter.

• There were a lot of unique ways to present the results; they also undertook a machine learning method (the LDA).

• Interesting topic. Good choice of data source. Good perspective to analyze on.

• Successfully fit a model

• nice unique plots

• Ambitious use of machine learning & statistical methods.

• Interesting topic and thorough statistical analysis. Clear on hypothesis and results. Organized presentation.

• Excellent(excellent analysis, use fancy methods)

• Applied great method to analyze the words

• multiple data resources, dendrogram

• word cloud graph is so good

• The plots applied some great statistical models

• The tree plot is very cool but it is also hard to see due to the small font size. The statistical methods used are very suitable and interesting.

• They use different methods to analysis their data and their plots looks professional.

• Diverse techniques for in-depth analysis

• description of work and graphs on second slide; great clustering analysis (different clustering methods)

• very interesting analysis

• They had very nice plots and clear and straight-forward presentation.

• Clear and good introduction. The plots are good for knowing the material.

• solid visuals

• I like the inclusion of context, lots of preliminary work. Interesting hierarchical clustering dendrogram, k-means clustering and interactive LDA, it's clear that lots of work went into this. Great summary as well.

• Good topic and clear presentation

• Nice application of machine learning algorithms. Cool interactive topic visualization.

• Group spoke clearly. The latter figures were interesting.

• Clear speaking

• Their speaking speed is appropriate.

• topic: camp fire. - Good data support, such as hierarchical clustering dendrogram, K-means clustering modeling, LDA modeling,

• colorful graphs

• Exploitation of Machine Learning methods; topic can be related

• Adequate informations Clustering really good. Did predictions. Caliifornia camp fire

• Their graphs are very easy to read and clear. They state hypothesis clearly. They use machine learning techniques as well, which is great!

• Clustering plot.

• The explanation is clear and the topic is interesting.

• They have good visualizations and great choice of topic.

• Presentation of the data and findings was done well. The follow up on, LDA or CDA (I cant read what I wrote sorry!) was nice

• slides are cure and attractive

• The presentation was fluent.

• They had a lot of visualations

• It is interesting to use k-means clustering to analyze their data sets.

• They used interesting models, like hierarchical clustering dendrogram, and methods to analyze the data.

Improvements

• Write the title as the main point of the slide
• The dendogram was kind of hard to understand initially.
• More explanation as to why this analysis matters
• Make cluster graph readable without in-person explanation.
• Slides had too much text, hard to read while also listening. Some presenters only looked at the laptop, not at the audience.
• At the same time the plot were interesting they needed more explanation
• Mentioned hypotheses in conclusion, but I didn't hear them before
• They can add some data analysis plot like histogram, density plot or bar plot.
• more introduction on data set and topic
• their slides were a little cluttered. the dendogram is also a bit difficult to visualize
• too much text on slides; focus would be better placed on visuals; include legend for hierarchical clustering dendrogram
• They put too many things in one slide. They can improve it by separating the slides。
• First two plots could have been higher resolution, they used a dreaded pie chart :0, some of the later plots had way too small text
• but k plot is difficult to understand.
• The dendrogram looks really cool but we could see none of the words on the x-axis label.
• The figures are not clear. The x-axis is hard to read.
• Maybe they can improve their hierarchical plot because it is not really clear.
• speak louder, lots of text in a slide
• Dendrogram looks impressive but is pretty much completely opaque. As someone who doesn't know what a dendrogram is, what did this communicate to me? Other charts/methods were similarly confusing.
• The hierarchical clustering dendogram was difficult to read.
• 1. cannot see those cluster figures clearly 2.should explain more on LDA and Hierarchical Clustering Dendrogram.
• Adding bootstrap on the branches of dendrogram
• Maybe let us see the upper branches.
• The search item is limited. Data scientist can not represent all data related job
• too much text and the plots are a bit too complex to be explained in a short time
• The speakers can speak up a little bit. The font size of their plots should be bigger. The team should explain the jargon they used more clearly.
• They need more explanations to their plots.
• Shapes in k-means clustering is a little bit hard to differentiated
• dendrogram x-axis words are unreadable (not sure what it's showing)
• I wasn't sure what the big words meant and it didn't seem clear what the results of the analysis were. State results clearly
• The words on the dendogram are so tiny and hard to read. Maybe, they could make it bigger.
• Focus more on people choosing to be data scientists, it will be good topic to go through.
• too much text
• Started out slightly quiet but great overall
• Too many information on all the plots, so the pattern of plots is unclear. They can filter the results.
• Maybe do two different k-means clustering plots for 1. twitter and 2. news so it's more clear that there's more overlap in news perspectives.
• It was difficult to understand the first two clustering and k-means plots.
• The bracket looking graph should be broken down into its most important parts so it's clearer?
• When introduce the what work they do, they'd better not put the graph of result in that slides. Also, there are too many elements in their graph, it is not a good choose of graph.
• The hierarchical clustering might be a little bit hard to read.
• does not cover a lot content , go deeper
• The hierarchical clustering dendrogram is not very readable. It also needs some interpretations.
• Nice. Keep doing it.
• The main problem is still visualization. The dendrogram is difficult to read as the texts are too small. They need to adjust the size of texts and labels.
• For the project, they might need to add more statistical analysis.
• The presented content are good start for further explorations. Maybe the group can try to narrow down the scope and try to have a more specific objective to focus on
• Being louder while presenting, though this did improve as the presentation went on
• The ppt can be more organized
• I think some of the plots were unnecessary.
• Label for the x-axis of dendrogram is not clear, it is a good choice to rotate them with 90 degrees
• It was sometimes a little unclear what their plots, like the dendrogram showed.

09:38

• I think your use of machine learning and statistics was bold and well done.
• They seemed to flow well together and were practiced.
• Explaining
• language using
• They did a good job on making the power point and they answered people's question clearly.
• Dendogram was interesting, never saw it before.
• ppt is well organized, voice of speakers is loud and clear.
• Their topic is attractive.
• Ahhhhhhhhhhh finally a group that removed spines & tick labels yessssssssss
• 1. interesting picture 2. interesting topic
• Youtube data is interesting and reliable.
• The group introduction is good and they did their analysis in reasonable and clear way. The exploration factors are good to discuss.
• Funny graphics
• Not available
• Intersting topic.

Improvements

• The Hierarchical Dendogram was very hard to read. It seems like the text was color coded as well but did not include a legend?

Maybe explain why you used certain methods to graph different things.

• They could speak a little louder.
• Good
• topic clearly
• They can improve it by adding more explanations on each section.
• too much text on slides, crammed too many plots on one slide, can't read anything on the dendogram
• add more data set not only include indeed.com
• although, it's good to title-capitalize titles and axis lables
• 1. should use more website. 2. should not have more job title instead of only "data scientist". For example, "data analyst"
• The explanation is not so clear.
• Hope they can made more explanations of their plots.
• Funny graphics, but I think it should be taken out?
• Not available
• There could be more information about where to apply
• Add more plots.
• there were too many plots on one slide and it could be combined

09:42

• very clear slides
• They brought up all the relevant needs for the data science fields. Looking at soft skills and other technologies was very helpful to know.
• Doing their own analysis from job postings and also looking at Kaggle data (also Eric Andre)
• Was very interesting how they found the number of mentions of junior positions on job websites, especially since the number was so low.
• Great slides. Excellent job of telling a story with data.
• Explaining and Slide Design
• Entertaining slides Explained results
• They make a very interesting presentation with a lot of meme.
• topic about web has some good data
• this was very relevant material! major buzz words to get a good job are always welcome
• I appreciate the memes; good use of graphics to get main point across; easy to read/understand visuals
• I can see that they fully investigated their topic(Trending categories in YouTube) with graphs and facts. Also, good transition between group members.
• use nltk technique that we learned in class
• Good sense of humor in using memes. The graph labels are also too small, but they highlighted the results in large font.
• The information from this project is helpful.
• Their topic is attractive.
• interesting topic
• The m.a.k did a good job researching to the key skills that the data science job require as well as the showing the keywords associated with data science.
• Spoke clearly and presentation was concise.
• 1. interesting topic 2. analysis trending for each month, which is very useful. 3. they analyzed the title, which is very interesting.
• Easy to get their significant results.
• good analysis
• Big data (52 G)
• important texts are obvious to see and remain in short
• Funny pictures in the sides. Plots are interesting and very informative. The presenter answered the question clearly
• Their topic is related to students and have practical significance
• Nice summary
• interesting visuals to convey the topic
• memes. Helpful topic and clear graphics.
• They had clear and straightforward presentation. They had nice plots and word clouds.
• The topic and the introduction is good. Their work is careful and method is clear.
• conveyed information well
• I like the memes. Great slides and graphs, easy to read.
• Plots are good.
• explained their findings well
• Good graphs
• Their slide is clear and have some interesting graphs. Their speaking speed is appropriate.
• topic: data science job. -Good data support. Bar plots and cloud words are clear.
• various data resource
• The topic and the structure was clear. Memes included
• I liked the word cloud.
• Very humorous and with interesting memes to attract the audience.
• Related topic.
• They did a good job of researching something that relates to most people taking this class.
• Interesting topic. Have a lot of meme figure to catch the audience attentions. The slides are clean and informative.
• Talked about the data source clearly and they add the interactive visualization.
• The humor of the presentation was great, the first slide was funny and a nice 'ice-breaker'. Including information/analysis of soft skills was informative and done well
• topic was interesting
• Their topic is very helpful for statistics student.
• The ppt was good.
• The titles provided a concise summary of information about the main point of each slide.

Improvements

• more intro
• The ticks were kind of difficult to read, but the plots still made sense.
• I would like to see more analysis as to other positions that aren't explicitly data science that may have more entry level options (such as analyst roles)
• Would be interesting to see the trend in job hires who have an advanced degree
• What was the hypothesis
• Somewhat basic topic...maybe delve deeper into some questions
• They could apply some algorithms in the project.
• the bar graph is wierd
• not much to improve I think.
• would've been interesting to see how data science compares to other jobs in the market
• Some slides looked a little bit messy and not clear. They can improve it by removing some unnecessary elements.
• voice is too low
• Their next steps were unclear.
• Try to do more about the data, like clustering, classification etc.
• speak louder
• This group can look more into the frequencies of the job that are related to data science, and compare the salaries between data science jobs vs. non-data science job.
• Methodology was unclear. In the future, it would have been better to replace the memes with a slide that explained the process used in the analysis rather than jumping straight into the results.
• 1. what's the future meaning of the analysis? 2. should explore more on the "title", instead of just uppercase/lowercase.
• Add the relative percentage of the languages that are required.
• I do not understand why they need the web app.
• some text are not related to the topic, plots and questions being solved are a bit simple.
• The plot label can be bigger.
• They only give the bar plots to illustrate their results
• How to define "must" and "important"? Some explanations might be better.
• words a bit too small on the graphs
• Could work on presenting skills. Know exactly what to talk about beforehand
• Maybe, they could put less memes, so their presentation could be more formal.
• Add more explanations for their plot in the report, since their topic is kind of academic.
• more visuals
• More interesting variables, may combine salary with other socioeconomic factors to get the life expense.
• Try make the font bigger and use at most 2 graphs per slide Don't use your phone as notecards make more eye contact
• Should speak louder
• I do not know what the last graph means. The key word like support, forward and act. confuse me. I do not how these words contribute to the topic. So they should give a better explanation, or remove some words that are not meaningful to the topic.
• prefer more data support as two many bar plots
• too many words in the ppt, make graphs more clearly
• Maybe change the black background if you want, it was a little hard to read. Maybe go in depth what you mean by the communication, support, management. To what degree does 80% of D.S. with a Graduate Degrees from your data apply to others (maybe from a certain area)? Is there more information regarding the sample observation group (80% of graduates) you can provide to make it overall clearer?
• Visualization. Graphs (percentage axis, may multiply 100(%)) . Make sure that all axis have correct labels and units.
• Explore more.
• There could be more information about the types of jobs we can do
• Voice are a bit too low, hard to hear from the back.
• May be do more statistical analysis.
• Speaking louder is an something to improve
• kept changing app In order to show plots but it was distracting
• They can do some analysis about the percentage of different track of statistics student can work in the field of statistics.
• The presentation can be more clear.
• Maybe they could explore other ways of analyzing the data.

09:46

• good graphs
• Interesting analysis on clickbait
• Key words and related events graph was a great idea
• I liked how they compared popular videos for different time periods, good summary slide
• They had interesting plots
• Explained goals and purpose clearly
• They make a good use of NLP and word cloud. This two analysis ways are very fit for this research topic.
• dark web data sets is interesting
• found a good relationship between words
• interesting topic; good use of data collection
• They presented the project with professional tones. Also, they combined text and graphs together to make each section very clear. They also add animation on the graphs!
• ppt is well organized, logic is clear, numbers of interactive graphs
• Their flow of presentation was good. The introduction explained well enough for people who don't know how the Youtube works.
• Interesting topic.
• easy to follow presentation
• The Miners did a good job researching the factor that could possibly affect the trending video.
• The group gave relevant background information to introduce the topic and purpose of the study to the audience.
• Good flow.
• Good visualizations
• The explanation are clear.
• the presentation was conveyed clearly with its main idea
• Very interesting question. The topic is closely related to our everyday life. The plots are made of interesting variables and many proper categories are made to clarify their topic.
• They catch the project which most people may interested in.
• In-depth analysis of text
• including words and visuals on one slide
• Interesting topic and good analysis
• The group chose very interesting topics.
• The introduction is good and their exploring method is clear. One of the speaker is not that familiar to the material.
• visuals were good
• I like the establishment of context and general information. I like that you related words with events.
• Good topic about Youtube and NPL is suitable.
• Clear public speaking
• Related to their jobs as consultants
• Topic: youtube - Clear explanation about project motivation. Good data support: word cloud, clean data plots, word frequency.
• Interesting topic
• Cool interactive graph
• Comprehensive research
• I like how the word cloud showed the dates relevant to it. I like the bar graph of the capitalized or not titles since it was easy to understand and not too busy.
• They introduce YouTube site interesting, keywords and related events (graph clearly)
• Interesting topic.
• Good job on looking at what type of videos we should make.
• Introduce the background information about the data set, which makes it clear for what they are trying to do. Provided related images to the dataset, which can help audience understand the result better.
• They used statistics!
• I liked what I could hear, which if my notes serve me well, talking about trading by month was nice
• The explanation of trending categories each month works well.
• summary was clear
• They present very clear about what what the goals of the project and the reasons why they decided this topic
• The ppt was very good.
• Nice explanation for explaining why video trend is Important.
• They did pretty good in the following aspects: first, they connect the high frequency key words to the related events; second, they analyze the difference in views of videos with capitalized titles and non capitalized ones. Both are very interesting.

Improvements

• speak more clearly
• Use more compact visualizations that are easier to draw conclusions from
• More analysis of what indicators affect video popularity
• Too much text on slides, hard to read and listen at the same time
• Some members looked like they were reading off the slides.
• N-gram word analysis could be more useful here for analyzing video content trends rather than keywords
• The facet plots are a little bit too crowded.
• focus questing may too many
• speak a bit more loudly
• word cloud was difficult to read; plots seemed to have very similar patterns to each other; would've been nice to see more analysis as to why; some plots didn't feel like they were saying much, look more into follow-up questions
• Some group members looked very nervous and talked too softly. They can be improved by practicing more before presentation.
• Confused about their dataset
• For the monthly graph, what they could do is to put them into an interactive graph to minimize the number of graphs. Also their summary was a bit too short and not meaningful enough.
• Slides are kind of messy. Try to make some animation graphics since the project is analyze data in time series.
• plots text too small to read
• Maybe they can make more eye contacts with the audience.
• The results were unclear in the presentation. There was a large number of plots with little explanation.
• Show us the significant results for most popular topics in each month. How about the title with non-capital and capital at the same time?
• Maybe more analysis
• One group member is not familiar with the presentation material.
• plots are presented too much in one slide
• The speaker should speak louder. Should make the introduction more concise. Summary words acquired from other sources. The plots should be made bigger.
• They show too many plots in one slice, which is hard to see and understand
• Does title capitalization really matter? How about their content or meaning?
• one group member is speaking too softly; not enough analysis done
• graphics were a little hard to see. Make them bigger or show fewer graphics
• I think their topic is too broad. Maybe, they can narrow it down to some specific trends/ areas.
• Creating more plots and dig into how themes are important of movies overall.
• less text on the slides, too much information
• How did you gather the data for trending categories/words?
• Capitalization seems not significant. May be more analysis about text.
• Should speak slower
• Improve plot clarity on slides
• Some of the members might not be prepared well enough to present without notes.
• Spent less time on each slide, slide could have been more concise.
• The word cloud was not very clear on the screen. Some word pairs seemed to be separated.
• For the keywords and related events, maybe have ‘super bowl’ as one word as well as ‘iphone x’. Maybe add in another category to fill the capitalized or not in title plot and see if you find anything interesting.
• Word Cloud graph and bar plots by graph are somehow vague or hard to read. Make sure that text are clear and larger in size.
• Trending categories plots are not easy to read.
• It would be nice to know more information about clickbait.
• Hard to see the plots when many of them are in the same slides. Maybe split out the graphs to different slides.
• Add more plots to the report.
• Some members spoke very quietly and it was hard to hear them. Looking at the slide a bit less could improve as well
• The bar chart about average views is kind of weird. It is better to use a table instead of a bar chart.
• data size used in the project seems small and incredible
• Some plots they provided are too small to see clearly.
• The presenters can be more confident.
• It is not clear that why use different color in word cloud, please give more detailed legends to explain it.
• They may improve by analyzing the impact of length of videos to count of views to figure out the best length of videos in each category so as to give better recommendations to Youtubers.

09:50

• I like the interactive graphs

• Interesting topic and good visualizations

• Thought their kaggle algorithm to determine drug prevelant countries was very interesting.

• Very interesting topic and excellent analysis. I liked the analysis of the users at the end of your report.

• Presenters had notecards, but still made good eye contact, good interactive plots, explained why some data was missing

• The interactive plot was a nice difference between other groups

• Design and Plots

• Interesting topic Interactive visualizations Explained methods clearly Successfully categorized words

• Very interesting topic!!! Good use of algorithms! Also, they make an interactive plot which is good!

• the virtual picture is clealy

• the interactive bar chart was cool. was very easy to see how the agora marketplace changed over time

• I liked the interactive bar chart

• Their talked with good organization.

• interesting topic

• The interactive plots are nice, but the scale seems too different from each other.

• Good visualization makes the trend clear to read.

• Coolest topic so far. Obviously had to go to some lengths to get data on such an arcane subject. Oh sick interactive visuals as well.

• Obeasts CT did a good job introducing the dark web, which is an interesting topic to look into. Also, the interactive plots is very creative.

• Interesting visuals and interactive graphs. Cool topic and analysis.

• 1. very interesting topic 2. use lots of statistical method to do analysis. 3. excellent interactive visualization

• Great visualization and methods.

• They provided detailed description about what their project is about and define what dar web is. They did well in using bullet points to represent the main points. Great animation on the slides.

• Interactive visualizations

• The topic is really interesting.

• interactive visualization was done well.

• Very interesting topic and a concise introduction. Very nice animated plot.

• They use map to show the frequency of drug with the location which is easy to understand.

• Interactive visualization

• explanation of topic; words and visuals on one slide; cool interactive visualization

• Liked the topic, nice presentation and visuals

• They had very nice interactive visualization.

• Doing good and clearlyb represented

• Good establishment of context. Good use of visualizations to show time.

• I liked the use of interactive plots

• Nice interactive visualization and interesting topic and data source.

• Good graphics

• Interesting topic.

• topic: agora/dark web. Interesting topic. Good use of interactive plots, such as pygal packages.

• Interactive plots

• I like how you guys plot it through time and were able to asses how it changed over time.

• They use pictures to attract people. They use machine learning techniques and interactive visualizations, which is very great and wonderful.

• Interactive plot.

• I like the onion in the beginning @ jiemin. Also, good job on the interactive plot

• Give background information about the project. The interactive plot are very informative.

• Clear data visualization.

• One meber spoke loudly enough for me to hear at least part of the presentation.

• topic was fun

• The presentation went well.

• Interesting interaction visualizations

• The interactive visualizations provided a clear way of understanding a lot of data.

Improvements

• speak more clearly
• They could have gotten more detailed as far as what is sold and the trends over time
• very difficult to hear in the back of the class. Also, would like to know more about where they plan on taking the remainder of their project
• What is agora? What is one-hot encoding?
• Maybe just include one interactive plot, so you don't go over time
• They could of practiced more, it seemed a little clunky
• Bringing the results together.
• More analysis of what caused these trends
• For one of the plot, it seems like they only got left part of the plot but left the right part of plot blank. they may adjust the plot a little bit.
• presenting skill
• difficult to hear
• choropleth could have more variation to it, maybe look into more specific questions to observe and in turn represent in choropleth form
• They could spend more time talking about the data analysis since their results were too few. Additionally, one group member were unfamiliar with her topic.
• more graphs on statistic findings
• There weren't too many connections between the results.
• The map can be improved by the color and the size
• For the visuals, having a lot of x factors approached unreadable.
• So far, I think this group is doing a job analyzing the data.
• The hypothesis was a little unclear.
• 1. speak loudly
• Can explain the chart better.
• The bar plot could be more clear by organizing the x-axis
• Why is "grams" leads to drugs
• The amount of data is small.
• explanations are a bit confusing
• Some speakers should speak up a little bit. The bar plot would be clearer if it is dodges instead of stacked.
• They use too many bar plots, which is single and the word is too small to read
• For the interactive visualization of the bar graph, bars not at the bottom are hard to be compared across time
• make the plots bigger (hard to see the words on the axes)
• Not sure what your analysis/conclusions were. state them clearly
• Maybe, they could speak louder. It was hard to hear them.
• Look into more details of our dataset and give more explanations of the plot.
• Could speak a bit louder.
• Bold important points maybe include just one interactive graph
• Would love to hear more about how the data was collected!
• Speak slower!
• Could have gotten real data not from Kaggle
• Prefer more plots to support the topic. Might not prepare well enough to present without notes.
• More eye contacts with the audience and reduce the number of times shifting between tabs.
• Since the U.S. had more with the drugs frequency, will you be going deeper and see which states (maybe) or region had which drugs?
• Probably they use more kinds of plots (scatter, or etc)
• Maybe more explanation of difference
• Maybe explain more when presenting the interactive plot. The plot is changing to fast to catch information.
• Changing the table to plots.
• Please speak louder, I could not hear the presentation very well and missed many main points
• The presenters can be more confident.
• I think they should give more statistic analysis to explain their data.
• They could include a clear summary of the main takeaway at the end.

09:54

• Interesting concept in looking at the difference between critics and the general audience
• interesting how they found trends in the manner in which critics and audiences both rate movies. Also no correllation between ticket sales and ratings makes sense, as some of the oscar movies (critically acclaimed) are not high grossing
• I liked the statistics that was added to this project.
• summarized findings in small amount of text
• They had alot of interesting information and an intriguing starting point
• Clear objective
• Scraped data from multiple sources Good statistical analysis
• Very interesting topic.
• the regression model is strong
• They had a lot of followup to their project
• good exploration into outside factors that could affect scores
• Nice and attractive graphs. Their slides are well organized. Group members also talked with confidence.
• I liked the interactive graph, and the changing colors was interesting
• factors of winning teams are reasonable.
• Using web scraping to get the data.
• good r2 comparison
• Great findings! Who would have expected that ticket sales have little correlation to reviews?
• The group did good job to use statistical data to find out the correlation between movie tickets and movies
• Presentation of topic and hypothesis as well as further analysis to be completed.
• 1.combine multiple data source
• Multiple data source
• They provided great example when describing their project instead of going through text main points
• multiple data resources
• The data resource is reliable.
• the statistical numbers are appealing
• The slides are simple and easy to understand.
• Very interesting introduction. Nice summary page. The presenter presented explained the audience question very well.
• They have explicit topic.
• using statistical methods (linear regression)
• Timely topic
• They had very nice plots and used statistical methods that they learned from other class.
• The theme is good and the introduction can be more clear.
• information was conveyed well
• I like the inclusion of other ideas.
• Good data resource and suitable plot. Speed is good.
• Provided small examples
• It was very organized
• It is a good idea to look at two different kind of score at the begining.
• Everything
• topic: rating. - - good explanation about motivations and project goals.
• Interesting take on movie analysis
• Clear objectives
• I like how you displayed the percentage with the images in the first slide. I thought it was interesting how ticket sales have no relation to ratings.
• Bollinger bands. Sharpe Ratios. The goal is clear. It has asset forcasting later on. Stock Market.
• They had attractive pictures and introduced tasks clearly, interesting topics
• using statistics to figure out the correlation between movies and ticket sales
• Having clear idea about what to present in the presentation. Great explanation about the intuition and the goal of the project.
• I like how they start with the difference in the ratings on different websites and they have clear idea about what to do next.
• I though the break down of the tasks involved with the project was good, as well as explaining what they were doing.
• Their topic is meaningful.
• They were clear about what they did.
• The topic(difference between audience and critic preferences) is meaningful.
• They did well in comparing the scores of recent movies from audience and critics.
• The powerpoint was uncluttered and clear.

Improvements

• They could have looked at multiple years looked at trends over time
• Again, would like to know where they are going next
• For the linear regression scatter plot I think you should set the y limit so we can get a better look at the fitted line. I think you should also report the sample size with your statistics as well.
• It didn't seem like the group practiced very much, read off notes and wasn't very loud, didn't support findings with good graphs
• Their presentation was a little bare, and put together quickly
• Explaining the plots
• Slides bland Presentation skills could use work
• They may add more plots and do more analysis!! It seems like they don't explore deep enough for this topic.
• data set describe
• the slides were a little bland and there wasn't much analysis
• why just observe seven movies? look into more!
• They can explore their topic more deeply.
• The graph could have had bolded, bigger, and rotated text so we could read it better
• graphs are little mass.
• The data size is relatively small, maybe try to scrape or find some api to access more data about movues.The plot is simple, and lack of explanation of the target.
• speak louder
• Too many of your great results begged to be plotted. I'm sure this isn't the only time you'll be hearing this.
• This group can do more analysis for the plots that they got.
• Plots and results were unclear.
• 1. should have more plots 2. the female student is not familiar with the slides. 3. it seems they only finish small part of their project.
• Make the charts better. I didn't get much time to look at last couple charts.
• The plots could be bigger to see and could provide more details, go through more about what the plots represent.
• Could have a better analysis. Should not read your notes all the time
• The material is not enough.
• one of the presenters need to explain more instead of reading the slide
• Can talk about which 7 movies had large differences on ratings, and see if there is a connect.
• The female speaker can speak up a little bit. The team can make more analysis on their data. Make more visualizations.
• They don't show enough analysis plots to explain their results and most of their slices only have simple sentences
• include some graphics for findings; more explanation on the graphs; more analysis/questions to look at
• Talk about the statistics and the graphics at the same time so we have a visual guide to the statistics
• The girl did not explain the the linear regression plots and elaborate more on what they tend to do next.
• Add more explanations and details for the plot.
• less visualizations
• Could put findings next to graphs
• The topic of difference between audience and critics is too small. They should look at more factors of movies such as directors, actors and so on.
• Maybe try to have more concise bullet point Try to not read so much off of notecards
• The graphs were a bit unclear though, try to split them up!
• They can also pay more attention to other elements other than the review. For the linear regression model, I do not seeing the meaning of that. They can just use the dot plot.
• Flawless
• • might not prepared well enough to present without notes. Prefer more factors and data analysis to support the conclusion.
• They should have practiced the presentation beforehand so they wouldn't need from flashcard; could have included more analysis
• Some claims may need more evidence: low correlation between the ticket sales and the ratings. This is a little counter-intuitive.
• I’m not sure if you have them or not, but maybe try to find some categories that best relates/correlates to ratings? This information might be helpful to movie makers.
• Don't be nervous when you are talking. You can do this!
• They may introduce their methods of analysis clearly, and their graphs labels should be larger .
• More visuals would be nice for the results
• The visualization are a bit simple. Maybe introduce other visualization such as word cloud, bar plot, etc
• I notice that there some outliers in the scatterplot maybe these outliers are worth to look into
• Reading note less and looking up could use improvement as well as being louder, one member was too quite
• plots were not clear
• Their slides are not well-designed and the group members seems not rehearsed.
• Need more plots and visualizations.
• Provide more statistics analysis like text mining and more visualizations for results.
• Two of their findings, which are "7 movies had differences over 30, 25 over 20" and "Audiences rated higher on controversial movies", lack the corresponding graph to illustrate. They could make some nice graph to show the distribution of the scores' differences.
• They could have included more plots, rather than just text with statistics.

09:58

• Like the topic that they chose. Liked how they drew a correlation between the height and weight and success against the team.
• Great topic! I liked the different segments that each of you led.
• I think that the research questions were very interesting and informative.
• Looked at interesting factors that could affect which teams beat the warriors, I liked the foul differential analysis portion.
• Spoke well together and and kept pace, and explanations of the plots were good
• Clear goal and plots.
• Interesting topic and goal Focused on multiple approaches
• data source is credible
• Thorough presentation, although they didn't factor in injuries. Also the Warriors are lazy during the regular season - theyve even said it themselves.
• Good choice of topic: gun shooting. Also, they presented the topic according to timeline, group of people, etc.
• Text was big enough, not too much text/bullet points on slides so I appreciate that
• The topic was interesting, and there was a new perspective: seeing how the referees were biased.
• Meaningful aspect to analyze on.
• good flow
• Great attention-grabbing title & overall theme. Interesting questions about referees & home vs away games
• With the limited amount of actual data, this group did a good job digging more data from referee.
• Clear methodology, topic, and conclusion.
• 1. interesting topic(beat warriors) 2.good flow of presentation 3.strategies analysis is good.
• Nice catch on the refereeing analysis. I LOVE IT.
• Very loud and clear voice. Great use of graphic for representation. Very interesting perspective looking at the referees.
• Interesting topic, plots comparison
• The data resource is reliable. (stock price)
• The focuses are clear
• They have a good understanding of their topic and explore some detail questions related with the topic.
• Knowledgeable about the topic
• loud and clear
• good presentation skills
• The theme is good and their analysis method is also reasonable. One of the speakers is bit unclear when doing the introduction and overall the analysis is good.
• Their topic is very interesting and creative. They looked at multiple aspects including players and strategies. They also spoke loud and clear.
• information was delivered well
• Good, clear focus, easy to follow.
• Good structure of the project. Referee is an interesting factor.
• Very good presentation overall
• Well-structured presentation with clear topics (player characteristics, strategies, refereeing). Spoke at a good volume.
• They investigate the Referee, and it is an interesting aspect.
• topic: how to beat the golden State warriors. -- - Good data support about players’ weight, height, and game strategy:two points, three shots, and rebound, faul.
• Interesting topic, nice ML models
• Interesting topic!
• interesting topic, clear interpretations
• I like how you guys all had basketball in common and the topic is pretty interesting.
• Sufficient analysis for surpassing Warriors in various areas. Have a clear goal path. Beat Warriors.
• Use clear table to list best teams, graphs are clear as well This topic is attractive for me as well (GSW)
• Interesting topic.
• The fouls difference was nice. The negative was pretty interesting
• Attractive title and topic. Clean slides. Fonts in the plots are also big enough to see from the back.
• I really like their graphs, they are clear and to the point
• A nice and loud introduction. I though the slides were clean and simple, easy to read. Great break down of analysis I though as well.
• The groups in a dot plots can be distinguished clearly.
• The topic is very interesting and they have a very clear focus.
• The topic was really fun. I enjoyed it a lot.
• They had a lot of details about the basketball team.
• The legend of graph is clear and helpful to understand the meaning of marks.
• Very interesting analyze of the impacts of refereeing and home/away on the Warriors' loss. They show a clear pattern of the bias through bar plot.
• The topic was more focused and detailed due to the smaller scale.

Improvements

• Would like to see some more advanced statistics (shooting percentage, turnovers, etc....)
• Continue to develop your analysis.
• I think setting the y-limits for the height and weight chart would be helpful for comparing the two and seeing how height and weight changed over time.
• Talked loud and was easy to understand.
• The plots could be labeled better, I didn't fully understand what they were conveying
• Bringing the results together
• Graphics were a little confusing Glossed over what they are still working on
• more plots on analysis
• This is a difficult topic to cover, as you cant quantify something like heart in sports. Teams that play the warriors generally play harder because they want to dethrone the champions. I'm not sure how to fix this however..
• This group can be improved by talking more loudly and making the size of plot larger. I can barely hear their presentation in the back.
• Barebones slides, could use some color. Graph was very boring and hard to read, it also needs color and bigger labels.
• The weight vs. height is not so obvious for seeing where the Warriors are.
• Data size is comparatively small.
• explain plots more
• Scatter plots were a little opaque. The questions they answered were hit or miss (height vs weight?)
• The group can explain more about the strategies for winning
• Not enough time spent on each plot.
• 1. should combine the change of players in warriors.
• The 3-point and 2-point shooting percentage chart is kinda unconvincing.
• Too much content within 4 minutes.
• I guess you are using the average stats of warriors for plots of 2/3 points, and I think an average stat is not convincing to be compared with when other data are in a single match.
• Group member are not familiar with material.(Too many emmm) The graph on the ppt is wrong.
• plot's forms need to be more diverse
• Their data is limited.
• Explanations for people unfamiliar with the terms
• legend labels need to be bigger; some 'other' teams are high in the graph, maybe find what those teams are as well; more description on each graph visually
• Introduction was a bit fast, I wasn't sure what your hypothesis was. State hypothesis and what your presentation is trying to answer
• The explanation can be added to normal people who doesn't understand the theme quite much.
• They could resize the font of slides or include more information, some slides seem a little empty.
• more visuals
• Inclusion of next steps
• More about details of players. May be include which players other team can employ to beat warriors.
• use team names not team id's
• Scatterplots were not very easy to read quickly.
• Actually, base on my knowledge, I do not think that they should investigate the weight and height. Also, since their topic is how to beat warriors, I think they should also invest on the assistant, the salary and the coach.
• • Prefer different kinds of plot supports and more statistic supports.
• I really like this presentation :)
• graph can be more clear
• maybe a little more introduction for people who don't know sports much
• You said the data was limited, maybe try to find more data sources to be able to answer more questions or data from another team (with some relation) and compare the two.
• Can add more analysis in different directions. I think it will be a very good consulting paper.
• They may enlarge their legend size of symbols The graph about number of games may be used by histogram or density plot/dot plot
• More explanation of results
• Have you thought about anallyzing the different players?
• Maybe give some background information for some terminology in the project. For example "rebound", free throws etc...
• The group mentioned in the plan that they want to create a logistic regression model that is a good idea but just pay attention to the sample size and the assumptions
• Some members could look up more and at the crowd
• The table about best teams against the Warriors is kind of confused. It may be better to draw a table with some data.
• too much work on terminology explanation and too little project result shown
• Some results they shows may not very useful such as the weight and height of players.
• The presenters can be more fluent
• I think it was good enough.
• The size of plots are not suitable enough since some plots are overlapped, so rescale the x-axis is a good choice.
• They could do better in providing a more clear way of visualization to compare the strategies of Two/Three points shots, Rebound and Free Throws between Warriors and other best teams other than dot plots, which I cannot quite see the clear patterns.
• Maybe they could explore other factors other than referee, like player performance.

10:02

• Addressed a relevant topic
• They spoke confidently and didnt hesitate
• Explainations
• Meaningful topic Good explanation of what they still plan to do
• Use various of plot(map plot, line plot, etc.)
• ourself
• Knew what they needed to expand on
• interesting and relevant topic
• Nice slides with graphs.
• Very good topic, good use of text and images on slides with minimal clutter
• The goals of the plots are clear
• easy to read text on plots
• This group did a nice graphs about the amount of incidents with respect to different cities.
• Clear topic and plan for further analysis.
• 1. interesting topic(gun)
• Very interesting topic as it relates to the recent event in Davis. They were focusing on the main points.
• good analysis
• the topic is appealing and the idea of the focus is clear.
• The topic is very relevant, which is a major issue we concern today.
• The introduction is very relevant to our community. There are a lot of visualizations, which makes the presentation very straightforward.
• They give a good introduction to their topic
• Topic linked to current incident
• trend graphs are clear
• interesting topic and clear graphics
• They had very interesting topics and nice plots.
• The presentation is doing good and introduction is clear.
• had great visualizations
• Good, clear plots.
• Gun violence is a good topic.
• good presentation overall
• I like the relevance of the topic
• topic: Gun violence - - interesting topic about gun violence.
• Meaningful topic
• I'm loving this topic! Thank you for choosing it.
• Good topic. It has many parts to show. Interesting. Gun Violence
• They introduce their data sets clearly and tell the audience their expectations clearly
• Nice graph about the total number of incidents by city.
• This is a really good topic since there are many aspect and directions you can choose.
• Simple clean slides
• The topic was chose well as it is one of the most important problems which need to be solved.
• Their topic is very meaningful since a police officer just got shot in Davis.
• The topic is good.
• The topic was very interesting because most of us knew the incident.
• The next step(Find the streets have most shooting incidents) they want to do is meaningful.
• They did a clear visualization to show the differences of number of gun incidents for different time and locations across California.

Improvements

• Some plots were cut off, could have talked louder and made more eye contact with the audience
• Some of the plots were hard to read because the data was so similar, they could add labels to show differentiation.
• Increasing volume, bringing the results together
• Explore what might cause these trends
• They may speak in louder voice
• ourself
• Talk more clearly, I couldn't hear at the back of the room.
• be wary of potential bias when it comes to looking at gun-related data, make sure you are aware of where the info is coming from
• They can practice more before presentation!!! Some group members looked very nervous and did not talk with fluency and consistency.
• Some plots have way too small and hard to read text. Also, the presentation was too long and became boring. :(
• Try to use some animation to show the trend.
• speak louder
• The group members can speak louder and clearly.
• I would have liked to see a percentage of gun violence incidents per city population to better compare cities of different population sizes.
• 1. speak louder 2. does the number of incident related to the number of guns? 3.poor analysis
• Could have spoken louder. The plot are good representation, but it would be better if x-axis and y-axis switch to show the trend between different months.
• maybe more data(like from twitter)
• the map graphs could be elaborated more by the team instead of showing them from the original source.
• It was hard to hear from the back, the group could speak up a little.
• The plot is not completely displayed in the presentation and the labels can be larger. Some speakers need to speak up. Some plots' axis scale can be adjusted to make them more informative.
• They need to speak louder so that audience can listen to them clearly.
• Month and day of weeks by time order might be more intuitive
• speaking too softly; graph is cut off on the first graph slide (need to explain more on the graphs); location y-axis relabel
• the analysis could be more clear by going into more detail about why certain trends appear
• Maybe, they could memorize what they wanted to say and not reading from the slides.
• Some of the box plot can be merged into one and colored by the content
• speak up more
• Volume of speakers, difficult to hear. Expand time-related trends to reflect a broader period of time
• The number of incidents by day of week and by month is not so interesting. They can use more data resource to combine more factors such as income.
• use a louder voice
• Speak louder!
• Could analyze data better (consider population, etc)
• • might not prepared well enough to present without notes. Unclear data index
• Could have more polished visualization, could spend less time on each slide
• Graphs could be more clear: the count of incidence during what time period?
• Can you guys elaborate on why it may be that from after July and a few months is the time period with the most gun violence? Maybe try to scrape some articles and see what people have been saying about this.
• Try to make the presentation more powerful and impressive.
• They may change legend names. The axis label is not correct for year. Year cannot be 2015.0,2016.0, etc. Graphs may be a bit crowded .
• What is the relationship between the months and the number of incidents?
• As you mentioned in the plan, take a closer look at specific city is a good idea, maybe also take a look at the ethnicity for each city?
• Louder! I couldn't hear the presentation. Also not looking at the laptop as much could be improved
• The bar chart about Time Related Trends is better to be a line chart which can see changes over time more clearly.
• They didn't talk too much about their topic. It is a little confused that what result they are trying to get.
• one of the presenter's voice was really low, I barely heard him.
• I think time is not very important.
• I think the bar chart is not a good choice to show the trend, the histogram or line chart may be better.
• They could do better if they can provide a more detailed explanation on why the total number of incidents stopped increasing at the year 2014, and keep constant since that. This sharp change looks interesting.

10:06

• had interesting graphs
• They had alot of complex information they presented and explained well
• Cool topic and plots
• expected value model is reasonable
• Interesting topic: cosmetic -- it relates closely to our lives.
• Good topic, nice font and color scheme
• nice graphs
• They made good graphs
• nice statistics
• Excellent slides & charts. Seriously, clean & good aesthetic sense across the board (maximized information, minimized chartjunk)
• Thorough research and analysis.
• 1. really cool analysis(predict stock price)
• Multiple data sources and multiple methods applied. Looks like a comprehensive analysis.
• They stated clear about what their project is and how they scrape their data. They provided very descriptive explanation beside plots.
• good topic
• the plots applied some advanced statistical models and the topic is appealing
• The topic is very interesting and good eye contacts.
• The plots are interesting and clear with visible labels. The speakers attempted to explain the methods they used but it is still very hard to be understood. The speakers talked about their limitations.
• They have good explanation to their topic and they also have some fancy plots in their ppt.
• Speakers are knowledgeable about the topic and meaning of the indexes
• including descriptions next to plots
• seemed like very good, complex statistics that might translate
• The group's introduction is good. The research is in reasonable method.
• Their topic is very interesting and there seems to be a lot to investigate in. Their graphs looks complicated and professional.
• visualizations were good
• Good introduction to things you've done.
• Good to analyze stock market and choose suitable companies.
• created nice graphs
• Cool graphs!
• They use lot of models and many good method like time series and optimization.
• Different approach to a hard problem (finance)
• topic: stock market. - - Good data source.
• They have clear and professional graphs about stock market.
• Interesting rainbow plot for the efficient frontier
• The ppt is very clear and they also include statistical analysis.
• The first member was loud and clear.
• The ppt was good.
• The topic is interesting
• I liked that they described the definitions of the statistical and financial terms and concepts throughout the presentation.

Improvements

• They seemed like they needed to practice more
• Time management
• They can add more graphs on the slides. Otherwise, it is a little bit boring sometimes.
• Way too much and text too big on one slides, too many visuals lol. They need to manually set labels, I saw that they clearly didn't do that for one graph.
• speech is not that clear and organized.
• I think their project is good enough...
• Presentation could use more background information to support the hypothesis and analysis.
• 1. should be more familiar with slides 2.does your analysis really useful if I use it to do invest? (should have more predict thing)
• Presentation could be better. Could try to make it better in report.
• They could have going through more main points about their discoveries. They spent too much time on one point.
• Could have more preparation for the presentation.
• This group could explain how to get the optimal portfolio.
• There are a lot of jargon that needs to be explained.
• More explanations on how the LSTM model works, why it is better, and what its limitations are
• work on pacing of the presentation and preparation, perhaps write down what you're going to say before hand
• Some of the plots can be made more clear with more explanantions.
• They could speak a little louder. For someone not familiar with the topic, it could be a little difficult to understand.
• less filler words
• Could use some context. Lots of text on some slides.
• try to define more some of the jargon
• When doing the time series analysis, I think they do not need to give a boundary to the time series except they are predicting something.
• Not well prepared for the presentation. Unclear plots and explanation.
• A little stumble; spent a little to long on introducing concepts
• What do all of those test mean? Can you explain to someone who doesn't understand machine learning?
• More explanation about the topic.
• Practice public speaking, being louder and not reading off the laptop
• It need to be better organized.
• The explanation about how to use LSTM model to forecast is not clear, please provide more details.
• They could have simplified the presentation a bit, maybe they didn't have to include every concept.

10:10

• the efficient frontier plots is interesting and amazing
• I like the dark theme with white text contrast. I also like how they used their stock market knowledge with what they learned in the class. Apart from some reservations, their graphs were unique and informative.
• thoughtful ideas, factors, nice graphs, good logic
• Their topic and ideas are good
• good samples of data
• Visually appealing plots of ingredients...
• Great breakdown on different market strategies based on the categories of products.
• Thorough analysis and research.
• 1. interesting topic 2.nice word cloud 3. nice heat plot 4.use machine learning
• Great job in cleaning the data after scrapping.
• They stated clear about what their project is about and their motivation and research perspective. They provided great description to support their plots.
• Interesting topic, plot comparison, good analysis, web scraping
• The statistic analysis is good.
• the presentation was conveyed clearly and fluently, and there were plenty of graphs in various forms.
• Good explanations.
• The speakers did a good job explaining their plots. The plots they used to present their language processing result is very interesting. This project included a wide variety of statistical methods.
• They use model fitting in their project.
• Diverse analysis including machine learning
• including a sample of what the web you're scraping from looks like; using statistical models
• The theme is good and the introduction is clear.
• went over, had to leave
• Cosmetics is a good topic. Questions are interesting. Good model.
• Cool use of topic modeling and cosine similarity.
• Clear slides.
• It is a good idea to investigate the tag of each brand. Using machine learning is a good idea.
• topic: cosmetics. - Attractive topic about cosmetics. Good data source. Web scraping. Good data plots.
• Interesting topic, detailed and well-rounded analysis
• Comprehensive research
• I like the choice in topic.
• Did characteristics for different brands. Analysis for makeups on different parts on body. linear regression for price. cosmetics
• Not available
• The group has already done a lot of work and they have pretty good results now
• Did a good job looking at the crowd, and lots of analysis.
• There are diversity kinds of graphs.
• The ppt is great. They did a very good job on everything in the presentation.
• They had very good visualizations.
• The topic is very interesting and I like their choice that using heat-map to measure similarity in consumers’ reviews and producers’ description since this method is clear to find correlations.
• The comparison between eye and face products was interesting.

Improvements

• predicting presenting
• Text too small on slides, too much text with graphs squished to the side. Speakers need to speak faster and in less of a monotone - unfortunately because of the delivery it made an interesting topic seem boring.
• voice is low, speak louder.
• They can make more eye contacts with audience
• reduce text in a slide
• ...but axis labels and units were not clear.
• Some of the plots are a little bit confusing since there are many colors involve and not that many patterns shown in the plots
• Spend more time discussing plots and analysis.
• 1. the axis sticker should me more clearly
• Make the significant results more appealing.
• They could have included less plots as they were going too fast between plots with each slide.
• The color chosen for the several visualization is not good because people can't really find the value by color.
• The graph is hard to understand.
• some plots need to be integrated into one plot, too much text, data size is a bit small
• The plots are a little bit overwhelming, it's hard to follow with.
• The speakers need to speak up. The plots are too small that I cannot see much information from the plot.
• They show too many plots in one slice which is hard to read and bigger data set may need to them.
• More in-depth analysis of the following steps
• voices too soft; more explanation on the graphs; ingredients should also include which ingredients (hopefully in your report)
• They can improve their report by adding more explanations of their plots
• went over, had to leave
• Categories are too big, so it may be more suitable to split into small categories for face. The plots are too small.
• Speak a little louder.
• Speak Louder !!
• Some of the graph have too many elements, and it is hard for people to see. I think is a big problem of their project. They can keep the most import thing on the graph to make the graph more clear.
• -Voice is low and a little bit hard to hear. Prefer more factors such as ingredient .
• Presentation could have been more concise
• For some of the top (and similar brands for the products for either eye, face, or body) maybe look further into them and see if you get any further interesting results. From one of your plots, natural products was up high, maybe investigate further what about natural products makes it noticeable high over others? Maybe look into some of the most healthy ingredients or what not and assess which products uses them most. Are some products overly priced, under priced in terms of healthiest (or something else in particular) ingredients. Can you recommend some products for people with different categories of demand/issue type?
• I think it can add some special ingredients.
• Not available
• In the model fitting section, there are a number of outliers that seem to skew the result, maybe try to see if they can be removed?
• Speaking louder for some of the members. The 3rd presenter was audible, but some of the other could speak up a bit more.
• I think it is also useful to analyze how to increase the sales volume of the makeup products. It could be done by analyze factors which might affect sales volume of the products.
• Maybe it can be more concise.
• I think it was good enough.
• The color of bar chart for tag frequency is too bright, please choose a low brightness color scheme.
• They could have described some of their plots in a bit more detail.

10:14

• Cool rainbow graphs and boxplots.
• Using web scraping to get the data. Plenty of plots.
• went over, had to leave
• They spoke well!
• Not Available
• I liked that you guys had hypotheses before going into your research

Improvements

• Talked too softly, which is especially bad as people were leaving and making noise. Crammed way too much on each slide and at one point showed us some of their pandas data frame which I feel is unprofessional and messy.
• Too much infomation on a single slides which can be somehow confusing. And some plots are hard to read.
• went over, had to leave
• Their graph labels were a bit unclear.
• Not Available
• Have you checked the assumptions before fitting a linear regression model to your data?

Thursday

09:02

• Clear on their project goal
• Their topic was really interesting and the heat maps looked really nice.
• The group had nice visualizations although hard to see some scales, and spoke clearly indicating they practiced quite well.
• introduce the the detailed background. the topic is pretty new and interesting.
• Very interesting topic, nice to provide background
• Google Wannabee group did a good job looking at the social impacts of their project.
• Good topic
• Very interesting data source, topic, and question. Good maps.
• They have the choropleth map which makes the distribution more clear.
• Graphs we're clear, topic was something I never thought of
• interesting topic
• Nice visuals. Your maps look really nice.
• I liked their graphics and the topic was interesting
• Really good maps that showcase their data. Clearly stated what they would further explore and how.
• The group did well on the combining the heatmap with the maps to explain the changes of the income of the New York area with the effect of Uber.
• Your map was well done.
• The heat map looks really nice.
• the maps show clear results
• Engaging topics and analysis
• Interesting, relevant topic and good visuals (maps)
• Well-made and clear maps to shows the popularity of the ride-hailing industry
• Interesting Topic, I like the heat map it would be awesome to have an animation
• Well-communicated and relevant goals, questions and methods
• Really nice income map Good solid argument about Lyft and Uber
• Speaking loudly, and the slides were clean and readable and easy to digest
• I think what the group did well is have a good background and through the lmited time explain it to the audience so its easy to understand
• Clear voice projection and flow
• They did great in choosing to compare the number of pick ups of taxi at different years with Uber&Lyft in low-income area to illustrate the impact of the new mobility on the improvement of mobility of low income people.
• Good topic and everyone's speech is clear

Improvements

• Maybe a few word to describe the map
• Could use more factors for analysis
• Some slides had little information other than a title, consider combining these next time? But overall, good.
• lack of data pictures to show more
• Still in the intro and I can tell they will likely run out of time, maybe rehearse and time more? I can't understand the results, hard to understand and compare across slides
• They could explain how they got their heat maps.
• what is the line of low income? Is it different over different states?
• Need to explain yellow vs. green taxi meaning. Could've shown how the data supported their hypothesis more clearly. It wasn't immediately clear from the heat maps (on different slides as well).
• There only analyze the graphs of yellow taxi. I think they need to contain more information.
• Future work seems more interesting
• prediction part need normal distribution to support the whole model
• Did you compare for other cities in NY? It would be interesting to see if there is a difference for other cities.
• Speak a little more loudly so the people in the back can hear
• maybe speak a little louder
• The group could improve on giving more data analysis on their data such as the distribution of the income using a bar plot or the change of the income through a line plot.
• I think it would be better to make the map interactive to see the actual proportion of people using a service in each area.
• I think an animated diagram would be better to show the change.
• the text in the graphs are hard to see and also consider adding a legend for the first heat map
• Different coloring for the heat map
• Expand the study to maybe a couple more geographical regions to improve generalizability
• More audience engagement
• needed to utilize data from 2014-2017
• Motivation and reasoning for picking the topic not super clear at the start Slide showing green pickup between 2013 and 2018 shows as "yellow pickup" for the 2013 map. I'm not sure if this was an error or if there was some explanation behind this.
• Less reading off the laptop (second speaker). The number of picks ups slide/graph could have been explained better
• Struggled with eye connection and voice projection
• Large graph hard to see. Legends could be larger in particular for yellow/green taxi pick up
• They made great geo-heatmap, but could be better if they make the legends consistent in magnitude, so that the color on graph would clearly indicate the change of pick ups.
• The three sets of the plot in the result are a little bit hard to compare with each other.

09:06

• This group did a good job in presenting the frequencies of words that different genders used during Valentine's Day.
• Project is well thought
• Data visualizations were nice, I really liked the common words by gender visualization based on male and female. Clear and concise presentation by everyone.
• Interesting to know about the peak occurring around President Day instead of Valentine.
• Overall, the group put together a cohesive and clear presentation. I really was interested as to how they went about presenting.
• topic is pretty interesting about the valentines day. the trending pictures is a good idea, and they made several ones for different topics.
• Slides do a good job of not including too much text, so we can focus on what you're saying
• They did a good job with the graphs and showed that the peak of discussion was on the 18th, which was very clear to see.
• The topic is able to utilize most knowledge we've learnt this quarter
• Creative topic!
• The distribution of post are clear enough to show the information.
• the conclusion is interesting
• Interesting topic. I like you exploratory data analysis.
• The topic is interesting and one of the presenters was very good at giving their presentation.
• Very interesting and relatable topic. Show what they want to explore further and had great visuals for the top 5 words each gender uses.
• The group did well on the combining the heatmap with the maps to explain the changes of the income of the New York area with the effect of Uber.
• I liked the topic. I also liked the background info on reddit posts.
• The presentation was very straight forward and easy to follow
• The topic is really interesting and the slides are well made.
• the focus questions and goals were easy to comprehend
• Visualize effect of the word clouds
• Fun topic, many types of analyses (scatter plots, time trend, word clouds)
• visualization is cool
• Nice layout on the slides
• clear plots
• Good slide design, and pacing (the right amount of information)
• Nice aesthetic & presentation! I especially liked the line plot that showed spikes before and after V-day
• I think the finding that people are posting about older people in their lives is interesting.
• Clean slides and straightforward analysis. The first speaker did a good job being out and way from the podium
• I liked their visuals as they were simple, but easy to read
• They did great job in collecting data from posts and picking up the information on age of pairs from the posts. Really interesting results on age distributions of pairs.
• It is good topic but i didnt found something interesting.

Improvements

• Maybe this group can analyze their data deeper
• Barley hear what they said
• You could explore more about with other big holidays of the year, or based on season.
• The WordCloud is a bit difficult to tell which words are most frequently used.
• The only little issue that distracted me from the presentation, was the low voice levels at certain times. Nonetheless, great presentation!
• the topic could be deeper, such how their reaserch can affect the market place.
• Often I can't hear you and I'm in the front row
• They could speak a bit louder.
• It will be better to address some specific questions
• Could've spoken a little louder. Scatterplot looked very overplotted. Perhaps you could do some topic modeling or other NLP techniques to try to group/categorize posts?
• They have less explanation on ppt. They need to have more words on ppt.
• the differences between uber and lyft are not distinctive
• I think it would nice to see analysis of reddit posts in other months.
• spoke too quietly, project their voices loudly. Look at the audience to engage eye contact
• The group could improve on the statistical analysis on the data
• I think that for the word maps, it would be more helpful to have counts with the words so we can better see the difference in the top words between men and women.
• In the "age" graph there was an age outlier around 80+, they could set the x limit to omit this to make the overall graph clearer for a presentation.
• Maybe it is better to take away the stop-words before making the word map. I wonder whether it is possible to set the color of the text according to the type of word(verb, noun, adj..)
• try making the points in the scatter plots a bit larger. the cloud maps show pretty much the same results
• Further analysis of relationships
• Identify actual topics being discussed, not just words
• did not include statistic analysis.
• In the age breakdown. What is the relationship between the two individuals? Maybe add more illustration here.
• maybe they filter the word cloud plots better to show more relevant words and distinction between females and males
• Why do you want to look at Christmas data? The words by gender analysis doesn't show many differences.
• The word clouds were misinterpreted. It should have been said that there was virtually no difference between male and female posts. Possibly mis-collected data?
• plot of ages of people posting could be clearer with colors maybe? The word maps are kind of hard to make sense of, especially when you're comparing the male and female words. They said that women used the word "feel" more often than men, but they appear to be about the same size. Maybe a barplot would show the differences more clearly.
• Some members could speak louder
• They could have added more information on the slides as though they talked about it, it was kind of hard to hear
• They could further improve their analysis of NLP by using more strategies we learnt in class, such as similarity measurement for posts of male and female, or different age group of couples.
• They can bring more deep than that. Such as, which gender post more than others and why.

09:10

• Good combination with R to graph the distribution of jobs related to data science in California.
• Very nice graphic
• Great presentation slides, clean with great data graphics paired with succinct verbal explanations. Really useful and interest project topic. Overall very clear and insightful findings.
• The map of USA was good to show which places were the best to work at.
• The group chose an interesting topic, and I believe they explored some interesting questions pertaining to the topic.
• the maps they made about the percentage or ratio around the US are pretty clear. the topic is really relative to the students who take this class.
• Nice to see a word cloud that is different in some way, good job of using not too much text on slides, so that we can focus on what you're saying, great job of speaking clearly
• I liked how they looked at population/ job opportunities. Their topic was very relevant to our class.
• The topic is useful for all students in this class
• Nice job scaling to population/job for map of states.
• They talked about the method they used to scrap data.
• They made fancy maps to show job opportunities
• Data related job, they answered questions about a current topic which is good and their site for scraping is a trusted website which means their data generation is cool
• the analysis of age pair is interesting to mention that the relationship is not only romantic one but also relationship in the family
• Your maps are very nice and it helps to understand the analysis.
• very interesting to note how the different techniques vary from data science and data analyst jobs
• Their slides provided a lot great information. I really enjoyed their topic and liked the colors of their graphics.
• Relatable because I myself didn't necessarily know what differentiated a data scientist from a data analyst. Good graphs that clearly showed what different skills were required for each job
• This group did well nature language processing and getting the more frequent words.
• I liked the word-frequency plots for job-titles and common languages.

I also liked the research question

• Maps on California specific was very useful in determining the differences between cities such as LA, SD, etc. Comparing graphs between data analyst and data scientists side-by-side is very useful.
• The maps comparison between jobs number and jog pressure is impressing. The measure of friendliness is interesting. Overall the diagrams look really clear and nice.
• the plots were nicely chosen to answer your focus questions
• Analyzing the heat map of job opportunities, great use of ggplot
• Great analyses, well explained, good questions, inferences below the plots was helpful
• Many statistic analysis.
• clear flow on the presentation
• they are very detailed with their analysis that they looked into different aspects that could affect data science related jobs
• Interesting topic, I like the degree requirements analysis at the end
• -I liked the map of job opportunities per population in each state.
• I really like that they take population of state into account and take a look at the ratio the result is very insightful
• Speaking loudly
• There are diverse types of graphs.
• Overall, they had good energy so it was an engaging topic
• Clear slides and voice projection
• They did pretty great in using geo-maps to show the disparity of data related jobs in the United States and California.
• I like this good. Clear power point and very useful information for us.

Improvements

• Maybe this group can combine at least two factors together in a plot to analyze more
• Pretty good overall
• Could speak more slowly.
• Is the map proportionate to the population size?
• The group spoke overall good, but fast at times.
• i think there is no need to introduce too much about how to get data souce
• I didn't understand how you determined if a job was entry level or not, assuming this was in the data or derived?
• They could have spent more time talking about how they got their data set.
• Many job titles is not really matches the job functions
• Interpretation of California job location map seemed incorrect - it looked sparse because the jobs were clustered by city.
• Some of the graphs are blurry. I don't know if it's because they are in the ppt.
• It is a little bit hard to see the points on their map.
• California data could have been more clearer
• For wordcloud, it will be better to combine verb of different tenses to make the word map more meaningful.
• I am skeptical that masters have a lower salary when compared to undergraduates. It would be nice to see why this is the case.
• would have like analysis from more than just california. Are there jobs overseas?
• Overall great presentation.
• Don't have to include an outline of what they are going to talk about in the presentation.
• This group could improve on giving models or
• Based on the map, I don't think it needs to be that in depth (with street lines etc) since you are only using city coordinates.
• California maps slide could be formatted a little better.
• Maybe adding salary difference between people with different efucational background is better.
• hard to see clear results from the rendered maps-- consider zooming in on the areas with actual results
• Further explanation of why masters make less than undergraduate
• Heat map color scheme can be made more discriminatory--right now colors sort of blend in
• nothing, well done
• Could make the distribution better
• I wouldn't use a outline for such a short presentation. The analysis seem disjointed, try to find one overarching story
• -Spelled the word "scraped" wrong -Ratio of people competing for one job does not depend on the number of people who specialize in that area. Maybe consider splitting it by industry/people with an applicable degree. -Also maybe consider taking cost of living/salary into account -For the data scientist vs. data analyst plot, maybe have them combined so it's easier to compare which keywords are more important for each.
• In the presentation you mentioned the jobs availability in CA as well but I am not sure how it is connected to or builds on the objective of the topic. Maybe try to clarify more in the final report but other than that everything is very impressive.
• Less reading off the laptop and looking up more (second speaker)
• I think it is useful to more about the relation between degree and data job, like why the job have different number of people in different degrees. It can be done by analyzing job requirements.
• Could have shown more visuals to support
• Job opportunities map colors are too similar, making it hard to see. May also consider looking into detail about what disciplines are asked for in addition to degree level
• They could improve by introducing some NLP analysis skills to further investigate the descriptions of the jobs.
• They are good.

09:14

• This group did a good job in combining at least two variables in one plot, which allowed them to analyze deeper.
• Clear project goal
• Explanation of project timeline, goals, current progress vey clear.
• Good analysis between price and production
• Visualizations were effective in supporting the claims that were made throughout the presentation. I really like how they made clear of interesting points in their plots by doing further analysis.
• the topic is pretty new, since tariffs were focused last year.
• Thank you for explaining tariffs; nice a lot of very timely topic! A lot of nice and relatively straight-forward plots.
• They did a good job explaining the definitions of the terms needed in the beginning.
• Interesting topic and bold to take it on!
• There are enough explanation before going to the points.
• Their PPT is very clear about what they are talking about
• The graphs for the AI/steel industry we're very clear
• It is good to analysis not only the top five places with the most job opportunities but also how competitive of each job. Interesting information concluded.
• Nice topic and you explained the background very well. I also like how you did graphs for different tariffs.
• Really liked their topic. It's relevant and nuanced
• 1. focus outliers/change points
• The graphs were really nice!
• Had really good graphs
• This group did well on plotting the density plot of the map and using statistic analysis on the data such as box plot.
• I liked how you looked into data from different presidents.
• Background slide is very clear and straight forward. Good mention of limitations near the end (conclusions can only currently explain the short-term).
• Plots are lableled nicely to show the effect of tax.
• Analyzing the tariffs on the U.S
• good introduction good visualisation
• Good topic and good charts.
• Good summary and conclusion
• Topic is interesting
• -The findings on a short-term improvement were really interesting, and I liked the plots -good conclusion
• Being loud and away from podium was nicely done. I like the explanation of tariff, simple and straightforward. This really was key to presenting any findings, I thought.
• Introduced the topic, ina really great way as it would be hard to understand otherwise
• Clear introduction and description of steps. Liked the arrow to help identify when the tariff was imposed.
• Good formating of the powerpoint.really detailed

Improvements

• Maybe they can try out different plots beside the lineplots.
• Lack of visualization
• Could analyze more tariffs on other goods over the years with other major countries the U.S. trades with (such as oil in the Middle East)
• Have you thought about combining the plots about employment together?
• Eye contact / voice levels could be better for a better presentation as it was a bit distracting and more difficult to understand, but I really think they did a good job.
• they focus on the steel industry, but i think there could be more about how tariffs impact the government policy.
• A lot of text on slides, no way we can read and listen at the same time; the time first series you show do not show the EVENT of the tariffs being put in place, I can't hear the third speaker and I'm in the first row.
• They didn't mention anything about where they got their data.
• More caution about interpreting inherently noisy economic data.
• They are mostly reading the slides. They need to give more information which are not only on the slides.
• Too much data on the website
• The ratio should be more specific because out-state population might also compete for the job in state too.
• How do you know if the tariffs resulted in the changes? Did you consider other factors?
• Would like to see some indicator of how the tariffs are affection the market (company values)
• 1. should put less text on slides, we don't have enough time to read all of them
• Speak a little bit louder
• Could speak a little louder because its hard to hear from the back. Could write less words on their slides.
• I would suggest them to improve on things such as fitting a model for the data science jobs
• I think that the price-production graph was nice to see the relationship between the two. But I think that the difference in y-axis makes it hard to interpret.
• One member was a little hard to hear (given I was near the back).
• Maybe adding other categories from different fields.
• Mention about other countries but partly spoke about other countries
• less words in the powerpoint
• Try analyze more products, not just steel, aluminum and soybeans.
• Could use a better guide to show tariff time
• One of the group member is nervous.
• -could have explained what a tariff is better -For your plots that show a time progression, consider having a vertical line showing where the tariff started (plots in the second part had this) -Make sure to speak loudly and clearly
• Some of the members could speaker louder, and look up at the crowd more
• Their powerpoint slides were not too good too much information and graphs were hard to see
• Inconsistent animations. Also, if year is in the title of the graph, don't need x axis labels to say "Jan-18", for example
• I feel they are good.

09:18

• This group did a good job in collecting data from different sources
• Really dig thru the dataset
• Good collection of data.
• The group chose an interesting topic, and was one of the groups that communicated very clearly. As a result, I was able to follow along with claims and visualizations.
• the oscar topic was catching my eyes. the analyse of percentage of race is interesting.
• Timely topic, interesting, good questions, I think I've seen some of these questions before like actor age in a stats textbook somewhere, although dated data of course. Interesting challenge dealing with category names.
• The oscars group did a good job clearly identifying their goals and questions.
• The group have a big topic, but it is good that they addressed couple small questions, which makes the research more meaningful.
• Their gragh are good looking
• They stated their project goals and the place that they gathered their data at the beginning.
• Their topic is interesting.
• Good data collection
• interesting topic
• Interesting topic and you guys explained the background pretty well.
• Data analysis presented well
• Their visualizations and insights into the relation between ethnicities and oscar success was interesting
• 1. interesting topic
• Very interesting topic discussed.
• Really good job with the data collection and sued many sources in roder to gather the data they did.
• This group did well on the choosing the topics and giving plots on the economics and tariff especially on the change of employment and the tariff.
• Good talking about the background information of the Oscars and specifically laying out goals & questions. Well explained graphs graphs that are harder to read presentation format.
• The thought about he last plot is interesting.
• goals were clearly defined and the plots support the motivation
• Analyzing the trend of previous and past oscars awards ever nominated
• very through explanation about the topic
• -Good guiding questions -good explanation of the problems you encountered
• Explanation of the project and slides was done well, and I liked some of the charts/graphs
• Everything
• Good voice projection, clear explanation of topic. Also, graph of ethnicity breakdown clear.

Improvements

• For the graph part, maybe try to facet different categories since there are some plots involve too much information
• Be more prepare on presenting
• What will you be using the children column for? The graph for names is a bit hard to read.
• Maybe increase the font of the labels for the plots, it was difficult to read despite the enlargement of the visualization on the projector.
• too much about data gathering as a four minutes presentation. the picture is too hard to see clearly. better to use line instead of dots.
• Plot definitely hard to read; ran out of time, maybe rehearse and time more?
• Doing a better job timing their presentation.
• It will be hard to get a specific conclusion, better to study the trend
• Their presentation is over the time limit
• They spent too much time on introducing data gathering. They need to put more time on analyzing their results.
• Their oscar category names graph is hard to read. Maybe they can change the type of graph they use.
• Presentation could have been rehearsed more
• data collecting part needs to be improved
• Did you consider what are the demographics of all actors and actresses?
• Conclusion/trend not clear
• Thought they could have done a deeper dive into specific categories
• 1. cannot see the plot clearly(title, legend, x/y lable) 2. time control
• Could have subsetted their groups more so the graphs were easier to look at
• Visuals were good but were a little confusing and I did't understand them.
• I would suggest the group to work more on the data collecting and could draw the data on a map.
• First three slides are a bit plain and could be compressed together.
• Making the legend and title of the plot bigger to show the plot clearer.
• i also agree that the scatter plot graph was hard to read. There were simply too many variables -- is there a way to lump some of the variables together so the graph would be less overwhelming
• Don't assume the audience, even though many people know of the Oscars
• more visualization, labels in the graph are not clear.
• make the chart better.
• -Too much background information,; not enough time to present results -instead of organizing data by the show number, could be more insightful to organize by year
• Less reading off the laptop for some members
• None they were perfect
• Some graphs would be better split into parts. Also, be careful about word choice. Certain verbal statements could be taken as offensive

09:22

• This group did a good job in graphing illustrations
• Eye contact and body language
• Interesting data visualizations.
• Good look at the mean salary.
• I like the added images used to emphasis the topic being discussed. It made the presentation more interesting and clear. I also like how clear and cohesive the group spoke as a whole.
• a lot of analysis to show. they chose great data and appropriate time period. they also focus on the salary transfer.
• Cool oiriginal topic, good job using sparse text so we can listen at same time, and clearly explaining your project goals. Challenging to combine data sources and deal with variations, cool!
• They had some of the best graphs, and I like how they included 6-7 graphs of different points of views (i.e. looking at UC Davis, comparing different levels of education).
• Their presentation is fluent.
• They used a lot of graphs and explain well on presentation.
• They get good data set.
• The data generated graphs are clear,
• the topic of nominees by race is really good. But it will be good to dive into this topic more and give some opinions about it.
• Nice visualizations.
• Data presented very well visually
• interesting how the relation in salaries from different professors is highly variable. Salaries of full professors increasing at a positive rate
• multiple statistics methods and figure
• Interesting topic. I loved the variety of graphs
• The spoke loud enough to be heard clearly. Their slides are also straightforward and clear.
• really good graphs that clearly showed their results.
• This group did well on mining the data and
• Show conclusive plots to show correlation.
• I liked the data sources you used. The add a lot of credibility to your project.
• Extra pictures to fill up background slides are visually appealing. Supplemented a table from a graph was a good idea.
• Adjust the salary according to 2017 is thoughtful.
• It was easy to follow the slides and understand the goals of the project.
• Analyzing information from two different sources
• Very interesting topic on academic publications and salary, clean plots, valid inferences
• good introduction good visualization
• Very nicely done on the analysis so far.
• Graphs nice to look at
• This group has very well-made and well-designed plots that are easy to understand
• Relevant question to audience & good job finding hidden API ;)
• -Good motivating questions -salary plot is SUPER messy and hard to read, but you made up for this with later plots -interesting results on difference between assistant and associate professors -good that you included next steps
• Speaking loudly, and explaining the H-index (tough by the end I forgot what it was)
• There are diverse type of charts.
• Interesting topic for professor salaries and publications.
• I think what they did good was exlaining their decisions and step by step process
• Very clear explanation of steps (how you got the data, etc.). Was able to answer someone's question about outliers well.
• They use a Venn diagram to show the conclusion.

Improvements

• This group can improve the plots by making them easier to read
• Don’t sound exciting in their own project
• Could explain more the process of getting the data
• The line plot was difficult to read. Maybe it would be better to just keep the boxplot
• The salaries graph was really hard to read. But this issue was addressed once the group showed other graphs that were easier to interpret.
• the analysis of each position professor salary analyse is kind of mess. maybe change it instead of the lines to show
• Seems like department or field of study may be a lurking variable, investigate this more?
• Speaking a bit louder
• I think they didn't give the propriate answer to audience's question
• The graphs of salary are kind of messy. I think they can use other type of graphs.
• Their salaries line plot is unclear, maybe they should change the graph types or just plot top salaries.
• The topic was available online everywhere, so something novel would have been cool
• wish the picture to be more clear and readable.
• The first graph is a little bit clustered, I like the box plot better. Can you also clarify what is the difference in job descriptions? I feel like that it is common knowledge that full professors would earn more salary.
• Conclusion or trend of data could be more explicit
• Would have been cool if they can break it down from academia as a whole to more details on specific disciplines
• 1. the line plots are messy 2.totally wrong answer to the question.
• Great presentation!
• Some if their graphs are difficult to read.
• Not much. Overall really good presentation
• I would suggest to sue subplots instead of plotting on the data on a single figure
• Add more data sources for validation of output.
• Some graphs were a little cluttered but explained well to compensate for it.
• Maybe fit a model to predict the salary with number of publications, citation rate and other possible variables. remove the outliers in the plot.
• The plot containing the salaries of professor is difficult to understand due to numerous amount of lines. Maybe consider integrating a a button menu where you can select the lines you want to see and mute the ones you don't.
• Difficulty reading the line graphs, consider adjusting the size, alpha level
• Need to expand data sources- maybe data from all the UCs
• some overploting in the graph
• charts could be better.
• Graph was cut off
• It would be interesting to see if there is any gender-bias or department specific factors associated professor salaries
• Overplotted UC Davis Salaries line plot. Consider using density/contour plot, or setting some kind of alpha on the lines.
• -you could try to fit a regression model to the publication plot
• Reading off the laptop and slide less
• The line charts about salaries is kind of confused. It may be better to use a bar charts with different ranges of salaries.
• Have you looked into other types of factors that may affect the professor salaries? For the individual salaries, maybe pick a few interesting cases and focus deeper into those. Also, I think the line plot for each professor can be a bit much.
• I think they could have founded more data still
• Maybe you could use facets for the topic diversity graph since the origin was pretty cluttered. Also, don't know if the graph with individual professors is useful.
• some plots are hard to understand and their voice really low when they present.

09:26

• The group did a good job in collecting information from different new sites
• PPT looks nice
• Great job explaining the data graphs-- they're really insightful and looked cool! Everyone spoke really well-- very thorough and good-paced.
• Interesting to see words associated with political bias
• Selected one of the more interesting topics. They got the audience interested with an initial question, and the did not simply read off the slides, rather they seemed to have practice and understand the problem they had at hand and how they wanted to go about answering it.
• they asked questions to get students involve. they introduced steps how they get the data.
• Did a great job creating interest with their topic & including classroom engagement.
• Good topic to use web scraping, it is comprehensive
• Their topic and content of the project is great.
• Good take on a potentially fraught topic. Thanks for not using word clouds!
• I like that they interacted with audience at the beginning.
• PPT style was cool, relevant topic right now, NLP is a plus for me
• i like the way to compare the rate of salary increase of different teaching levels. I am also interested in the result of the next step about how number of publication effects the salary.
• Good explanation of your project. I also like your visualizations.
• Good project flow with introduction and clear data analysis steps
• really cool topic. presentation very clear. Liked how they analyzed n-grams
• 1.interact with audience 2.interesting topic
• Great presentation skills. Interesting idea and the slides were well put together
• They interacted with the audience which is very good. Their slides are also very good, colorful but not difficult to read.
• Interesting topic that is very relatable in this political climate. Good job fo recognizing what else that have to do in order to answer their question further.
• This group did well on the gathering the data and making the scatter plot as well as finding the trend of the salaries.
• Showed process of implementation of idea very well in presentation.
• Background leading to project question was well stated. Clear step-by-step process.
• one-two-three word phrases are impressing.
• GREAT PRESENTATION
• Nice analysis on a difficult topic
• good introduction
• Scrapping multiple news source to analyze.
• Well-made and colorful slides
• Good topic and reasonable method
• Confident presentation. Thanks for fitting in some audience participation, and interesting topic
• Ben Millam's notecards were quirky and Cyrus is a hunk
• -good captivating introduction -good explanation of your process -really good use of NLP concepts we learned -good that you included next steps -good enthusiasm -good thorough answer to the audience question
• Speaking loud and interacting with the crowd
• Great visuals
• Great interactive introduction. Clear voice projection and hang gestures.
• Very thoughtful and interesting results for democrat and republican.

Improvements

• Maybe the group can try out other kinds of plot
• Well done overall
• Nothing!
• More plots about other things and not just words
• Sometimes there was a lot on the powerpoint, and it was went through quickly.
• i am afraid that there are not that much data to collect to do deeper analysis.
• Explain how they obtainied their graphs.
• it will be too limit to study the trend if you guys only list top ten results
• I think they do really good job
• I have no criticism, great job!
• They talked about too much professional statistical things which will make audience who do not have that kind of knowledge confused.(I am not sure whether we need to assume the audience be 'statistical'.) It's better if they make their presentation easier to be understand.
• The topic is really good but the work done is almost done in the class itself. So something the group did themselves would have been better
• salaries plot needs to be improved to make the plot more readable
• How did you calculate the political bias score? That wasn't clear to me.
• Sources chosen were already biased towards a certain conclusion
• Would have liked a relative frequency count. Also, they should get some 'moderate' news sites in their as well as clearly partisan ones
• 1.should spend more time on the result.
• maybe use a different variety of plots?
• Maybe they can try not to read off a note next time
• Overall good
• I would suggest the group to clear the line plot and using subplots
• Could add more plots or other graphs to show correlation.
• Could distribute more time to going in depth on results than process
• Explain the choice of key phrases. How are they related to being biased?
• y'all did great
• More background on bias for international students not familiar with the problem
• need more statistical analysis
• They could have been more explicit with how they calculate the political spectrum scores and they could also weight their scores by frequency of each key phrase.
• There are too many word in one ppt slide.
• More even distribution of speaking within group
• -kind of weird that one of you was all the way across the room
• Reading off note cards less
• improve on eye contect and voice projection
• For trigram graph, x axis labels could be rotated a bit to make it less cluttered
• The power point has too many colors, easy to lose the track when I read it. suggest having a clear power point background

09:30

• Like how they shape the word clouds

• Outstanding job introducing and explaining the project. I loved the pipeline process explanation.

• Good research about the most popular types of food

• Visualizations were effective in supporting claims. Although it was hard to hear at times, what I did hear clearly was insightful. Overall, interesting presentation and topic.

• they choose a good way to show the distribution of the restaurant. I like the picture they make different states shape picture. they not only focus on the restaurant but also on what element makes customers satisfied.

• Cool result to see the peak times bimodal around meal times!

• The yelp group did a good job analyzing the types of restaurants from different views (like by state or by cuisine).

• good topic to use web api

• Interesting to take on "perspective" of both diners and owners.

• I like the common words part of their ppt. Those picture have the shape of each country or cities.

• Yelp analysis, good clear graphs, words mentioned in different states were a good addition

• I really like your word clouds. The shapes look really nice.

• Word cloud data presented well in the form of each state

• multiple methods.

• Great topic! I like how they used the word maps that were shaped to the states

• Good use of visuals and very relatable topic. I liked the graph of all the most popular words that were used to describe restaurants

• The topic is interesting and the nature language processing

• Had interesting visualizations, implemented a model to go along with it to make the data very useful.

• The data you used is really credible.

• Well-explained results from graphs made. Shaping word clouds by different states is a very nice touch.

• word map according to the shape of the state.

• Goals and focus questions for this project are clear

• Analyzing information using regression

• approach the project in consumers and producers (restaurant-owners) perspectives

• Interesting topic, I like how you framed the importance in terms of helping restaurant owners improve their ratings

• Presenting the word clouds, I liked them

• They had a good topic and were performing with energy which made it easier to focus on since class is so early.

• Word cloud for different states was interesting

• Nice plot and very clear

Improvements

• Speak a little louder
• Could incorporate a sentiment analysis.
• It would be better to use a flat picture of the map
• I think it would be nice to change how you first barplot is laid out. I think you cna organize it either alphabetically or organize it in ascending or descending order.
• instead of using map to show distribution, also can use other more numerical way.
• I see R's notation used for the regression model, like the tilde ~, not sure if this is universal outside of R?
• Speaking louder
• Is the conclusion meaningful? Why not study the city development and type of restaurant?
• The restaurant distribution map didn't look right or informative. Why no dots over SD, LA, or SF? Ratings didn't look normally distributed.
• It would be better if they discuss data gathering.
• Rating is readily available on Yelp, so I could not understand the linear regression model, wish there was more in that
• For the map, I think it would be more clear if it wasn't a satellite image. It would also be nice to explore if people in different cities rate differently. For instance, do people in California tend to rate restaurants better?
• Model to predict rating seemed too simple as they are just summing different aspects of restaurant.
• The map is hard to see(change the type of map), too much word cloud.
• Maybe include where they got their data from. did they get it from the yelp app or a different location
• Could speak a little louder
• This group could improve on making more plots such as frequency plots
• Maybe add another data source to validate Yelp API predictions.
• Look at the difference in popularity of restaurants in different cities? Do types of restaurants differ in large cities vs smaller cities.

I think you should also try weighting the rating by the number of reviews. Or faceting rating by the amount of reviews and seeing fi there is a difference.

• Might want to consider circling important words in the word clouds (for presentation purposes only)
• the distribution of restaurants across America can be plotted on a black and white map to be more obvious.
• There were slightly too many word clouds that makes it hard to see the comparison across all the states. Consider employing comparisons with only 1 or 2 plots.
• Maybe include a quick overview of the description of the restaurants
• restaurant data are not representative. they are clustered around state capitals
• Too much information for the time.
• Speaking louder!!
• They could have added more visuals in powerpoint
• Restaurant distribution map doesn't need to be so detailed. Some grammar mistakes throughout and the graph did not show that the ratings were normally distributed
• Increase the volume when they presentation. I was little bit hard to hear what they say.

09:34

• Interesting project
• The different factors analyzed were interesting. It was nice that you compared key words visualizations based on some states.
• Good topic with nice plots
• The visual appearance of the powerpoint was my favorite. I found it to be an interesting theme that went well with the topic. The group addressed the little things that mattered like this that made the presentation interesting.
• focus a lot of life ladder. the correlation matrix show clear.
• Great title slide and graphics, simple text so we can focus on what you're saying, also an ambitious topic to tackle!
• They did a good job making their presentation interesting.
• The topic is interesting
• Interesting data and topic.
• They discuss the data sources.
• Happiness is all that matters, statistical methods were clear and easy to comprehend. Sources of data are good, correlation matrix was good, 157 countries is a vast job which is good
• Good explanation of your project.
• Clear trends of data shown and significant analysis presented
• i thought their use of regression in the presentation was cool. We haven't really worked w/ parametrized models in this class so it showed a willingness to explore outside the box.
• 1. beautiful map 2.multiple type of plot
• their plots were amazing!
• Visuals clearly showed their results so far
• Our group
• Very conclusive results and well rounded data sources.
• All slide visuals are nice. Potentially hard to read graphs were explained well.
• The slides are well made and the topic is interesting.
• Good use of regression and correlation matrix to look for factors that may influence an individual's happiness
• -really nice linear regression
• Explanation of the life ladder was informative. That is explanation of a key element of the project was done well
• good presenters had a great flow
• Clear slides and diversity of plots

Improvements

• No eye contact
• You could possible look into reviews on google.
• How did you determine and find the Life Ladder to plot
• The plots were hard to read, maybe increase the font.
• as they said, the data set is small.
• I can barely hear your third speaker and I'm in the front row. On the correlation matrix, red is often associated with negative vs blue positive (although this may shift based on culture in which presenting), so you might consider making the positive correlations blue and negatives read, I initially misread your first two matrices before I noticed this.
• They could have ran through their presentation a bit more .
• It will be really hard to get a meaningful conclusion for the current topic, the factors are not persuasive.
• Unclear what data points are for case study of China. Perhaps you can use the Gini Coefficient to measure inequality?
• The two correlation matrix are not easy to understand. I think it would be better if they used other understandable graphs to show their data on presentation.
• Case study could have been USA.
• For the graph, It would be nice to see the country names instead of the codes.
• Life ladder may not be the best measure of happiness.
• I thought the heat map could have been clearer
• should explain more on what is life ladder.
• project their voices a little bit louder. maybe try to stand in away from the laptop and make eye contact with the audience
• Maybe speak up a little more
• Our group
• Could define each of their metrics more in depth.
• Might want to speak a little louder
• Consider different countries and how to balance the related difference which could possibly affect the result.
• Have y'all consider leisure as a factor of happiness?
• -could speak a little louder -kind of hard to tell the difference on the life ladder map; consider using a multicolored gradient -dark red and dark blue look pretty similar on tile plot -could use more explanation of how you used the factors at the beginning to create your life ladder scale
• Reading of the laptop less and speaking louder for some members
• graphs too small to see
• Grammar mistakes on many slides. Voice projection could be improved. Life ladder regression plot very confusing (missing axis labels)

09:38

• Beautiful presentation slides and really clear explanations
• interesting comparison between cars
• The group spoke clearly and maintained solid eye contact throughout the presentation.
• The last several plots about the percentage with extra features is pretty coo. and they made plots overlap each other.
• Good comparison plots, interesting to see the transmission type change over time. Also I like the price difference over years.
• Their slides did not have a lot of text compared to a lot of the other groups!
• used most knowledge we've learnt
• Really cool interactive visualizations.
• The graphs are colorful.
• eBay car analysis, graphs were clear
• the sources of the data is interesting.
• Good analysis. I like how you guys found good conclusions and related them to each other. Your interactive visualizations also look really good.
• Interactive data visualization part was very well done and unique
• Liked the links that they made between the increase in electric cars and the last ten years of changing technology
• very beautiful interactive visualization.
• Their slides were very well put together: easy to read and follow along. The different types of plots was really neat.
• Used very intricate visuals that I haven't seen any other groups use so far and showcased the data very well.
• The group did well on building the model of the measuring the life ladder especially on choosing the different variables that represents the life ladder
• Had really cool interactive plots to show conclusion.
• I think a lot of the graphs were easy to read and interpret.

The interactive visualizations were well done.

• Graphs were well-explained. Slide-to-slide transition was fluid.
• The interactive diagram is quite impressing.
• The interactive plots were extremely nice and well done!
• cool visitation
• They have very well-made interactive plots
• Pretty Graphics!
• • I liked the box plots

• REAL GOOD interactive visualization

• Posture and use of hands for one some of the presenters was done well
• I like how they listed their data sources, slides were neat and straight to the point
• Good diversity of plots and voice projection in general
• They had a lot fancy and beautiful plots

Improvements

• Could explore the factors that contribute to low happiness scale
• can you relate the electric cars to the brands of the car?
• I think for the bar plots you can organize them in ascending order.
• i was wondering for car, is that good idea to grab data from ebay instead of the carmax, which the enterprise only focus on car?
• You're telling us project goals, but might be good to also state these on a slide instead of narrating a lot off of just the title slide. Is there a way to combine your category comparisons to make further conclusions?
• Explain how old their data is/ where it is from.
• better to study that for different income level families
• Don't use a pie/donut chart. Perhaps you could look at what types of cars "hold value" best.
• They don't have an introduction on ppt. They go straight to the graphs.
• The topic and the graphs were redundant. Could have gone into car miles, condition to get a better metric to judge which brand of cars is better
• could add more factors that affect the happiness to make the prediction more reliable.
• For the fuel type graph, it would be nice to see a bar chart instead because the pie chart is not as easy to understand.
• Could have done further analysis into how different features of cars affected resale market
• Wouldn't have used a pie chart, was difficult to see the market breakdown from it
• The data is not solid, because there are more websites for second-hand car.
• could explain the interactive plots in a bit more detail
• Could speak a little louder
• I would suggest the group to explain more on the formula of their model.
• Conclusion could have more well articulated.
• I think looking at seeing which cars held their price better over time would be worth looking into.
• Might have been better to present a little less information
• Maybe classify cars into categories bigger than just brand, Japanese brands, luxuary types, etc.
• I think there are previous discourses for this topic already. I thought it was slightly hard to understand what this project would add to the existing discussions.
• more statistic analysis
• They can group different car brands into more general categories, such as luxury, economic, SUVs
• You seem to have many figures representing the same data, it would be better to choose the one that best shows your main point.
• • the plot of automatic vs manual cars is kind of confusing, since you have the pie chart on top of the line plot; instead of having raw number of cars, maybe just show the proportion of automatic to manual
• Speaking louder and reading off the laptop less
• Talked a little louder
• Labels could be larger on some graphs. Why do people buy used Porsches (why is this luxury car often purchased secondhand)
• Not really show us the interesting things they found from eBay. and the source is limited. Try craigslist or other websites to find more information to compare with.

09:42

• Good use of sparse slide text so we have the opportunity to focus on what you're saying, I like the procedure 'pipeline' sequence steps with bullet points. Good narrative of finding something curious in the data (in the scatterplot) and following up with subsequent analysis.
• They did a good job showing the data by using maps, and their graduation rate by zipcode that had dots with different size.
• Addressed specific questions
• Their plot is good looking
• Wow, awesome job using so many data sources. Would love to see a combination of all the cities.
• Public school performance, data collection includes all categories, within the city and among cities is a good thing to analyse, the data sources are good,
• interesting interactive graph
• I like the map of the graduation rate.
• Good analysis of data taking into account relevant factors
• 1. beautiful plot
• great plots I like how you focused on 3 major cities
• Interesting topic. Good graphs and graphics that showcased their results
• The group did well on statistically analyzing the data of the used cars using boxplot and line plot to find the trend of the used car market and the interactive figure.
• Had enough data sources for analysis.
• I think that research and follow up questions were very interesting.
• Presenting graduation rate of NYC in a map was a good idea. Plots were explained well.
• The comparison between different cities.
• Chose an interesting topic of income vs. education
• very interesting topics, nice visualization
• I like this group has well-designed plot
• Topic is of personal interest to me! Nice use of contrast (blue/orange) in slide aesthetic (not strictly data science but all presentation matters!) Interesting nonlinear results in the Graduation Rate by ZIP Code Income Across Cities visualization. Linear trends are so boring sometimes.
• The group has a really clear idea of what they are going to next and there is variety in the visualizations they used.
• Explain the 'gaps' in the graphs. (the large white space)
• Slide animations helpful
• a very clear procedure of analysis

Improvements

• I can't hear the first speaker and I'm in the first row.
• Couldn't hear the first speaker at all :/
• Why picked those 3 cities? By randomly? Maybe it will be better to pick top income level cities or education level cities.
• They spoke too quitely, it is hard to hear
• One person needed to speak louder.
• Should have been audible, graphs could have been explained with data behind it and it could have been more cities than NYC.
• None, I think it would be nice to see the analysis of the other potential factors.
• Speaking skills could be improved as it was difficult to hear the members
• 1. speak louder
• speak louder
• For the most part spoke loudly but one person couldn't really here
• I would suggest the group to add models using the data
• Could relate data sources to result more better.
• I think there were too many categories for the graduation rate by type of school. I think it would have been easier to look at it if you faceted the graph or made it interactive so you could pick the different categories.
• Tan titles a little hard on the eyes (of the graduation rate to NYC map).
• Maybe classify schools into different categories: middle , high school.
• Better explanation of the plots, more analysis
• maybe make the label of the graph more clear
• they could add trend line on their plots so it is easier to see the trend
• Legend in Chicago: Graduation Rate by ZIP Code Income was overplotted and confusing
• • really need to speak louder

• plot of graduation rate by income is pretty hard to read; consider splitting information into different plots

• I really like the visualizations the group made, they are colorful and informative just make sure in the final report add a little explanations/inferences for each one
• Speaking louder, I could not hear some of the speaker
• Voice projection could be improved and a legend for the graduation distribution rate would be helpful
• I think graduation is not only related to income. They can analysis more other things than income to evaluate the public school performance, such as the teacher's salary, student's tuition fee etc.

09:46

• it is a useful topic
• Amsterdam Airbnb, dataset can be trusted, comprehensive database for the 3 year period
• the comparison of generation rate of school type is reasonable and informative.
• Good exploratory data analysis.
• Good analysis of trends and discussion of outliers
• Liked their breakdown of users by year which clearly showed growth over the years. Also liked how they highlighted the difference between prices for airbnb and hotels
• 1. focus on unusual place in map 2.beautiful slides
• I like the idea, AirB&B is very popular. Presenters were ver passionate about their topic! The pictures of Amsterdam were a nice addition
• They interacted with audience. Their shaped their word map as a house which is creative.
• Really liked the frequency of words visual and they chose a good topic that could be done for other countries too.
• This group did well on plotting the data on the map and
• Provided really good plots and conclusive data.
• Side-by-side bar graphs by year was a good idea. Using black dots in the maps to indicate maximums was smart to make those points better seen. Extra details, (pictures to help label parts) is a nice touch.
• The slides are well made.
• Effective use of visuals. It was great that you explained the abnormalities.
• Nice case study, simple but useful analysis
• very interesting topic, good introduction
• They have clear and well-organized slides. I particularly liked their heat map which also show where are the top 10 most expensive airbnb houses
• I like the heat maps, right amount of information for the given time
• Nice catch analyzing the anomalous absent area around the Amsterdam college.
• Crowd interaction and most of the members speaking loudly
• I like the plots overall.
• Clear format and voice projection. Also good summary

Improvements

• what is specific related to airbnb? what is the difference with researching the housing price?
• Could have compared with other websites, the results shown like average price is available already on Airbnb so the metrics for analysing could have been better, I felt the heat map covered the entire Amsterdam
• I could barely hear their voice and I am confused about how graduation rate was plotted on the map? Based on public school location?
• You said that the airbnb prices are lower for hotels. I think it would be nice to see a graph comparing the airbnb prices vs hotels.
• Relevance of word cloud of description was not clear
• thought the heat map could have been clearer
• 1. word cloud is beautiful but not clear, should use bar plot.
• Eye contact and voice projection.
• They could speak a little louder.
• Could you maybe compare airbnb in Amsterdam to another country?
• I would suggest the group to do more statistical analysis on the data
• Could have asked more questions to solve a deeper problem on the dataset.
• Maps plot could use a legend.
• Maybe look into the rating of each apartment and other factors, does that affect the price?
• instead of a word cloud, consider frequency distribution plots instead
• Would have been nice to see their interactive viz
• more statistical analysis
• I think they are explore more about what features contribute the most to prices, beside location
• Too many animations on the slides are distracting
• I would have liked to see Seasonal Demand as an overlaid kernel density plot as well, to show distribution rather than count.
• The third member could speak loud, was hard to hear
• Small change: maybe update the x and y axis for the Seasonal Demand average price per bedroom over 2015-2018 plot.
• Animations didn't help and would like explanation of why you chose Amsterdam

09:50

• good topic, good to learn
• Interesting "by sex" plots - surprising results! Interesting further directions.
• California road safety
• Concise and nice PPT.
• Compared relevant features well
• interesting topic
• nice touch with the visuals on the slides.
• Good graphs
• This group did well on analyzing the off-set data and using the natural language processing to get the high frequency words.
• Answered a meaningful question, laid out objectives and showed conclusive plots.
• The goals of the project was well looked into.
• Side-by-side graphs is a good tool to emphasize comparisons. Outlined next steps clearly.
• The classification of people injured by group implies the type of the accidents.
• Goals were logical and some achieved.
• cool visualization
• Good pacing at the beginning,
• The introduction and statistics at the start were a nice attention grabber
• I found the questions to be interesting.
• Introduction slides very useful. Clear about future exploration

Improvements

• better to subset car types, like bus and track
• One person needed to speak much louder.
• Could be audible, basic graphs, what to do next is more interesting than the work presented.
• It will be good to state why Amsterdam specifically.
• Slides were formatted incorrectly and speakers could barely be heard
• should leave more time on result, I cannot read all four plots in 10 second.
• speak louder
• Could speak a little louder.
• I would suggest the group to clean the density plot
• Could state conclusion and next steps more clearly.
• I think you could better name your bar graphs because it makes the graph seem like people are being killed by having sex.

I think it would also be helpful to look at how the data changes over the years as well.

• One member was a little hard to hear.
• Maybe modify the number killed or injured based on the total population of the area to which areas are more dangerous than others, or look into the location where the accidents happened to analyze why.
• The histograms should have the same axis if they are to be used for comparison.
• more statistical analysis less words in the powerpoint
• More eye contact with audience, fewer words on slides
• Speaking louder, I was hard to hear, and reading off the laptop less
• Maybe fill in the bar plot with some other category so we can see what number amount within the bar plot makes that amount. Also maybe add in the year to your plots so it will be clearer what your information pertains to. Have you looked into maybe weather and what months or season does traffic accidents happen the most? Just an idea that may be a factor.
• Some grammar mistakes and inconsistent formatting. Scales are different for some of the graphs, which made it hard to compare.

09:54

• Pediatrician density and preventable emergency department visits in CA counties. Got to know asthma data is not there before 2012
• Nice presentation. I think the motivation for your project is really meaningful.
• Rationale and goal of project presented clearly
• Very relevant topic. Very clear, informative topic. Really liked her use of analysis to predict relations between the conditions and doctors
• plenty of analysis
• great information
• She knew what she was talking about. Also chose a very interesting topic that needs to be researched more
• This group did well on the analyzing the data of the traffic accidents
• Provided a very good overview of the process of the project.
• I think the motivations, research, and constraints were well explained.
• Good speaker. Explained the scope of the data, which is good. Important information in specific parts of a data frame highlighted which is a good idea.
• The introduction is thorough and clear.
• Methodology was well explained.
• good introduction and through analysis
• Uses her expertise and knowledge
• Interesting topic, good job with time mangement
• • really good explanation of why lack of access to a pediatrician is a problem; I can tell you care about it

• good explanation of your method; good decision to focus on one area
• good plots of your results
• good that you included next steps

• Speaking loudly and being out from the podium
• I like how you chose a topic you are familiar with to be able to focus on what may be important.
• Great voice projection and hand gestures. Clear explanation of importance of topic

Improvements

• Too much data on the PPT, looked more like a health PPT rather than a statistical PPT, the data plotted is already in the data, so it seems to be a bit too easy
• graph needs to be more diverse and conclude more details.
• None, I do think the heat maps are a good idea for the next steps.
• Explain certain research specific words to be better understood by audience
• Would have liked to see more visualizations to understand the relation between the data.
• too much text
• maybe slow down when talking, was a little difficult to follow along but had really great information
• Speaking really fast, so she could slow down a bit to understand better. Overall knew what she was talking about.
• I would suggest to generate plots that has more categories that involved with the dataset.
• Could have provided plots and data to provide evidence to conclusion.
• I think the asthma ed visit by rate was too busy. I think you should minimize the number of groups by regrouping them or facetting the plot.
• Might want to allocate more time on results than methods
• Maybe a heat map to aid the visualization.
• It would be great if there were more integration of visuals to support your claims.
• more visualization too many words in the powerpoint
• Too many tests on the slides (use less raw output)
• Hard to near, perhaps a microphone could help for longer presentations
• Making explanation short/more concise. Also having less busy slides
• I think doing the heat maps like you said would be helpful.
• Not sure you need the plot with all the counties. I think the plot with just the top and bottom 5 suffices

09:58

• Explained data analysis very well
• multiple plots, which are beautiful.
• great information
• Did a good job gathering their data and they showcased it well. Interesting topic.
• This group did well on the model building and analysis of the data.
• Problem statement was on its on slide which is a good way to emphasize. Background and scope was clear, and graphs were explained properly.
• Incorporating different aspects into the yelp analysis is interesting.
• Fairly detailed analysis, explained well
• though analysis of the method used clear visualization
• This group uses a variety of data sources
• I like the use of NLP to compare between good and bad reviews
• Being loud and away from podium. Also cross collection of the health inception information was convey in a great manner
• Clear explanation of data acquisition and steps taken

Improvements

• Presentation was too long
• the text on map is hard to see. should use arrow instead.
• could provide more plots
• Overall good presentation, but just had to time it better
• The group could improve on cleaning the line plot
• Might want to consider having word clouds in the same format. I think I might've saw "the" in one of the word clouds. Might want to introduce more stopwords.
• Maybe analyze on a single - restaurant level instead of a area-level to see what are the influencing factors.
• Should have worked on making sure they finished on time
• less words in the power point can use visualization instead.
• Self-reported food poison can be biased and not representative. Maybe they can use hospital food poisoning records
• too much information for the given time
• Read off the laptop less
• Dont know if a different color for each district is needed. Positive word cloud hard to see. Maybe order bar plot for iwaspoisoned by frequency

10:02

• Interesting project topic chosen
• Their ability to show the areas where terrorists attacks have increased over the years is very insightful.
• multiple plots, which are beautiful.
• they had nice plots
• Their graphs are very clear, easy to read, and representative.
• They had good graphs and plots
• This group did well on the finding the regression of the rating and prices.
• Interesting topic.
• Word map with points is a good way to represent general location. Graphs were clear to read.
• The plots are well made. The regional analysis is interesting.
• Showing the global maps
• I think the topic is interesting. I like how you included up to 1970 data.
• Liked the map they made and showed what populations their project was on
• Nice animations.

Improvements

• Different aspects of data could be analyzed more in depth
• Would like to see if there is a metric for attempted terrorist attack that was foiled. If so, do areas that seem to have a heightened fear of attacks actually have justification for said fear?
• don't nervous :)
• make eye contact & speak louder
• They can try to avoid reading from notes and speak a little louder
• Could speak up a little more
• This group could improve on natural language processing
• Graphs and facts could support a more practical conclusion.
• Word cloud resized, which made it blurry. Might want to speak a bit louder.
• Maybe speak louder.
• For road safety, the group only uses counts in all their plots. This can be problematic because California has a large population, it would be interesting to use relative measures and compare them across the country.
• Could practice more, and speak louder
• I'm not sure if you have it included or not, but maybe include the specific year or year ranges onto the plots. Maybe try to analyze over which months have the most number of attacks and can you determine why?
• talked too fast
• Map for attacks hard to see. Also maybe consider interactive visualization with slider. Recommend ordering bar plots by frequency and tilting some axis labels

10:06

• The group did well on plotting data on the map and building an interesting mode
• Always nice to mention limitations. Explaining interesting points of graphs well.
• The machine learning process is interesting.
• Lots of plots
• Clear and well-organized slide and interesting slides on machine learning techniques
• Explaining the limitations of the project/data and showing confidence interval on the graphs was nice
• I like the topic of Gun Violence. Thanks for it.
• everyone had a clear speech

Improvements

• I would suggest this group to add more on the statistical analysis of the data.
• Might wanted to have used arrows from google presentation rather than drawn in exclamation points.
• Maybe explore a little bit more to improve the accuracy.
• Speak more clearly, combine multiple plots into 1
• They only used self-report perceptions of police violence. This can be biased. It would be interesting to see administrative data, which is less biased
• Read a bit less of note cards and the laptop
• Maybe try to add in other types of demographic?
• Too many elements in a bar chat and hard to see which is important. maybe they can just limit a few important elements to compare with.

10:10

• na
• For the group on terrorist attack, I like that they break down the terrorist attacks by years and showed a significant increase in attacks after 2010

Improvements

• na
• Their maps are not well-labelled. This can confusing

10:14

• na
• Interesting topic, many plots, tried to build a random forest model

Improvements

• na
• Did not finish on time, so needed to present better given the time.
• I do not know the purpose of this project. Why do I need to know Police Activity in LA?