FinalProject_group046.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "2WdpPLpIa_4X"
   },
   "source": [
    "# COGS 108 - YouTube 'How To & Style' Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "tIfGncZKboNb"
   },
   "source": [
    "# Overview - Group #46\n",
    "\n",
    "For our project, we decided to explore four different areas revolving around the \"How To & Style\" category for YouTube videos. In these four different areas, we explored four properties of a video and tried to answer our research question. We explored the the quantity of tags used, popular tags used, title sentiment, and upload date of each trending YouTube videos.\n",
    "\n",
    "These four different sections are named as \n",
    "1.   Quantity of Tags vs Views\n",
    "2.   Usage of Popular Tags vs Views\n",
    "3.   Title Sentiment vs Views\n",
    "4.   Upload Date vs Time to Trending \n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "pjdzQ0xOfMX1"
   },
   "source": [
    "# Names and IDs\n",
    "\n",
    "## Names\n",
    "* Dat Nguyen \n",
    "\n",
    "* Tianheng Ma \n",
    "\n",
    "* Justin Nguyen \n",
    "\n",
    "* Kasidech Tantipanichaphan \n",
    "\n",
    "* Samuel Lim \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "-K2V4LxFgNAx"
   },
   "source": [
    "# Research Question\n",
    "\n",
    "- What combination of number of tags, usage of popular tags, method of phrasing titles, and upload dates should How-to and Style content creators employ to maximize views and minimize time to reach trending?\n",
    "\n",
    "## Sub Research Questions\n",
    "1.   Quantity of Tags vs Views: How does the quantity of tags used in a video affect its number of views?\n",
    "2.   Usage of Popular Tags vs Views: How does the usage  of popular tags used in a trending video affect its number of views?\n",
    "3.   Title Sentiment vs Views: How does the title sentiment of a video affect its number of views?\n",
    "4.   Upload Date vs Time to Trending: How does the upload date of a video affect its time to be on trending? \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "h5V_ZuuJugJ4"
   },
   "source": [
    "# Background and Prior Work\n",
    "* Where did we find the dataset\n",
    " * We found the dataset off of Kaggle, we were perusing through their already published datasets.\n",
    " \n",
    "* Why do we choose to use YouTube datasets\n",
    " * We chose to use YouTube datasets because everyone today uses YouTube. For those that want to delve into the world of content creation, specifically into How-to and Style content, making great content that balance revenue and growth would be of paramount priority. \n",
    " * There’s a lot of data to use.\n",
    " \n",
    "* Why did we choose to use How To & Style category \n",
    " * There were many factors that we put into selecting this category. As we were sifting through the many categories with many observations in our dataset(1000+), we found that a lot of them were just uninteresting (News and Politics) or required a certain amount of prestige (Entertainment, in the case of it being filled with movie trailers). Therefore, we narrowed it down into How To & Style because it revolved around the creativity of the content creator (what kind of nifty tools they could make out of household products, etc). As such, this category would require certain levels of creativity and skill to thrive, but will clickbait offset this?\n",
    " \n",
    "* What we currently know about the topic\n",
    " * We don't really know much, other than recreational use of YouTube. We aren't content creators nor exposed to this type of content.\n",
    " \n",
    "## Previous Work (include links):\n",
    "1) https://www.kaggle.com/donyoe/exploring-youtube-trending-statistics-eda/report\n",
    " * Observed a high correlation between views and likes\n",
    " * Made word clouds of titles and tags of youtube videos. Found that titles and tags related to “music”, “new”, “iphone”, and “episode” showed up consistently in popular videos\n",
    " * Includes visualizations to depict trends in: views vs likes, likes vs comments, analysis of sentimentality in video descriptions. Notices positive correlations between views, likes, and comments. Sentimentality in descriptions tends to be positive in majority\n",
    " \n",
    "2) https://towardsdatascience.com/strategies-to-be-a-successful-youtuber-extensive-youtube-exploratory-analysis-part-2-2-6862cd4f40be\n",
    " * Observed how to be a successful youtuber. Found out that making models per topic would be better. \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "PEqfuhk-vNKG"
   },
   "source": [
    "# Hypothesis\n",
    "* Hypothesis: Titles that have positive sentiment will induce more interaction, thus yielding more views. Additionally, videos that are uploaded in the weekends are faster to reach trending. Furthermore, we hypothesize that the more number of tags there are and more popular tags are used, the more views as well.\n",
    "\n",
    "\n",
    "## Reasons:\n",
    "For titles that have positive sentiment, we believe that titles with more amicable words will tend to have more views (so as to not give off a negative connotation before the viewer even views the video). To us, this makes sense because videos with negative sentiment are generally looked down upon.\n",
    "\n",
    "For videos that are uploaded in the weekends, people tend to have more free time, thus more people are going to watch them so they will reach trending quicker.\n",
    "\n",
    "For videos that have more tags labelled, we believe they will be recommended to more viewers, thus they will get more views.\n",
    "\n",
    "For popular tags used, we believe that under the How-to and Style section, a lot of makeup gurus tend to publish their videos under this category. With that style of video being the most popular, we expect a great number of videos to represent that tag. Additionally, food creation videos (recipe videos) are also very popular. Thus, food tags, in a similar fashion to make up tags, will be highly represented."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "h5X3sV-tvRZs"
   },
   "source": [
    "# Dataset(s)\n",
    "\n",
    "* Dataset Name: Trending Youtube Video Statistics\n",
    "* Variables Included : Tags, Publish Date, Trending Date, Views, Titles\n",
    "* Link to the dataset: https://www.kaggle.com/datasnaek/youtube-new/version/114\n",
    "* Number of observations: 4,000 (specific to the How-to & Style category)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "GLRs-zGZG0AA"
   },
   "source": [
    "# Setup"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "tYc-qlkXasZH"
   },
   "source": [
    "These are the step for general set-up across all of the analyses we have done. It allows for use of our general dataset, in addition to using np/pd/matplotlib."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "Xg-V3NnHG1jM"
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "j2nqomNR6u0q"
   },
   "source": [
    "**We have 4 different sections throughout this notebook. Each section pertains to a different topic that we worked on. For that reason, we have 4 different set-ups, 4 different data cleaning sections, and 4 different data analysis sections. **"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "3m1AJVmbbWqp"
   },
   "source": [
    "### Create a General Dataframe Called \"youtube\"\n",
    "\n",
    "We will only keep the columns 'title', 'tags', 'views', 'trending_date', and 'publish_time'. We also drop any rows with null values.\n",
    "\n",
    "Note : We clean the data here because the other set-ups rely on 'youtube'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 195
    },
    "colab_type": "code",
    "id": "u3w9O1AH_XZ_",
    "outputId": "cc9ce636-d94d-4a70-c717-e2322ffcaebe"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>title</th>\n",
       "      <th>tags</th>\n",
       "      <th>views</th>\n",
       "      <th>trending_date</th>\n",
       "      <th>publish_time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>What $4,800 Will Get You In NYC | Sweet Digs H...</td>\n",
       "      <td>refinery29|\"refinery 29\"|\"r29\"|\"r29 video\"|\"vi...</td>\n",
       "      <td>145921</td>\n",
       "      <td>17.14.11</td>\n",
       "      <td>2017-11-12T16:00:01.000Z</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>A Smart... MUG?! - Take apart a Heated Thermos!</td>\n",
       "      <td>Smart mug|\"Heated thermos\"|\"tech\"|\"gift idea\"|...</td>\n",
       "      <td>120727</td>\n",
       "      <td>17.14.11</td>\n",
       "      <td>2017-11-13T16:00:03.000Z</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>iPhone X vs Makeup Transformation (Face ID TEST)</td>\n",
       "      <td>iphone x|\"iphonex makeup\"|\"iphonex test\"|\"new ...</td>\n",
       "      <td>1456472</td>\n",
       "      <td>17.14.11</td>\n",
       "      <td>2017-11-12T00:45:54.000Z</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>John Lewis Christmas Ad 2017 - #MozTheMonster</td>\n",
       "      <td>christmas|\"john lewis christmas\"|\"john lewis\"|...</td>\n",
       "      <td>7224515</td>\n",
       "      <td>17.14.11</td>\n",
       "      <td>2017-11-10T07:38:29.000Z</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>A VERY #EXTRA EBAY SHOPPING HAUL!</td>\n",
       "      <td>ebay|\"haul\"|\"cheap\"|\"challenge\"|\"extra\"|\"amber...</td>\n",
       "      <td>484185</td>\n",
       "      <td>17.14.11</td>\n",
       "      <td>2017-11-11T02:42:23.000Z</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                               title  \\\n",
       "0  What $4,800 Will Get You In NYC | Sweet Digs H...   \n",
       "1    A Smart... MUG?! - Take apart a Heated Thermos!   \n",
       "2   iPhone X vs Makeup Transformation (Face ID TEST)   \n",
       "3      John Lewis Christmas Ad 2017 - #MozTheMonster   \n",
       "4                  A VERY #EXTRA EBAY SHOPPING HAUL!   \n",
       "\n",
       "                                                tags    views trending_date  \\\n",
       "0  refinery29|\"refinery 29\"|\"r29\"|\"r29 video\"|\"vi...   145921      17.14.11   \n",
       "1  Smart mug|\"Heated thermos\"|\"tech\"|\"gift idea\"|...   120727      17.14.11   \n",
       "2  iphone x|\"iphonex makeup\"|\"iphonex test\"|\"new ...  1456472      17.14.11   \n",
       "3  christmas|\"john lewis christmas\"|\"john lewis\"|...  7224515      17.14.11   \n",
       "4  ebay|\"haul\"|\"cheap\"|\"challenge\"|\"extra\"|\"amber...   484185      17.14.11   \n",
       "\n",
       "               publish_time  \n",
       "0  2017-11-12T16:00:01.000Z  \n",
       "1  2017-11-13T16:00:03.000Z  \n",
       "2  2017-11-12T00:45:54.000Z  \n",
       "3  2017-11-10T07:38:29.000Z  \n",
       "4  2017-11-11T02:42:23.000Z  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#take in csv file and set category to How-to and Style\n",
    "youtube = pd.read_csv(\"USvideos.csv\")\n",
    "youtube = youtube[youtube['category_id']==26]\n",
    "\n",
    "# clean out columns that aren't relevant and drop nulls\n",
    "youtube = youtube[['title','tags','views','trending_date','publish_time']]\n",
    "youtube.reset_index(drop=True, inplace=True)\n",
    "youtube = youtube.dropna()\n",
    "youtube.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "sKTpDR_V0yGJ"
   },
   "source": [
    "## Quantity of Tags vs Views Set-up"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "BzEmA7rnWNRW"
   },
   "source": [
    "### Process Tags Into Our Dataframe\n",
    "\n",
    "We store tags of each video in a list, which we keep in the variable 'new_tags'.\n",
    "\n",
    "We store the number of tags of each video in a list, which we keep in the variable 'count_tags'\n",
    "\n",
    "Finally, we add the number of each tag into youtube['count_tags'] and update the string of tags with the list of tags in youtube['tags']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 511
    },
    "colab_type": "code",
    "id": "DXq7rAsgWMPG",
    "outputId": "4462e17b-b55a-452a-f087-6817d7e63bf4"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>title</th>\n",
       "      <th>tags</th>\n",
       "      <th>views</th>\n",
       "      <th>trending_date</th>\n",
       "      <th>publish_time</th>\n",
       "      <th>count_tags</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>What $4,800 Will Get You In NYC | Sweet Digs H...</td>\n",
       "      <td>[refinery29, \"refinery 29\", \"r29\", \"r29 video\"...</td>\n",
       "      <td>145921</td>\n",
       "      <td>17.14.11</td>\n",
       "      <td>2017-11-12T16:00:01.000Z</td>\n",
       "      <td>35</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>A Smart... MUG?! - Take apart a Heated Thermos!</td>\n",
       "      <td>[Smart mug, \"Heated thermos\", \"tech\", \"gift id...</td>\n",
       "      <td>120727</td>\n",
       "      <td>17.14.11</td>\n",
       "      <td>2017-11-13T16:00:03.000Z</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>iPhone X vs Makeup Transformation (Face ID TEST)</td>\n",
       "      <td>[iphone x, \"iphonex makeup\", \"iphonex test\", \"...</td>\n",
       "      <td>1456472</td>\n",
       "      <td>17.14.11</td>\n",
       "      <td>2017-11-12T00:45:54.000Z</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>John Lewis Christmas Ad 2017 - #MozTheMonster</td>\n",
       "      <td>[christmas, \"john lewis christmas\", \"john lewi...</td>\n",
       "      <td>7224515</td>\n",
       "      <td>17.14.11</td>\n",
       "      <td>2017-11-10T07:38:29.000Z</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>A VERY #EXTRA EBAY SHOPPING HAUL!</td>\n",
       "      <td>[ebay, \"haul\", \"cheap\", \"challenge\", \"extra\", ...</td>\n",
       "      <td>484185</td>\n",
       "      <td>17.14.11</td>\n",
       "      <td>2017-11-11T02:42:23.000Z</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>EXTREME GOLDEN TAN TRANSFORMATION</td>\n",
       "      <td>[Extreme Tan, \"Tanning\", \"Tanning Bed\", \"How t...</td>\n",
       "      <td>300617</td>\n",
       "      <td>17.14.11</td>\n",
       "      <td>2017-11-10T20:40:44.000Z</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Inside the Cheese-Aging Caves 30 Feet Under Br...</td>\n",
       "      <td>[cave, \"cheese\", \"goat\", \"rind\", \"sheep's milk...</td>\n",
       "      <td>92523</td>\n",
       "      <td>17.14.11</td>\n",
       "      <td>2017-11-10T16:01:04.000Z</td>\n",
       "      <td>28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>ICE PRIMER? DOES IT WORK?!</td>\n",
       "      <td>[ice primer, \"beauty hack\", \"weird beauty hack...</td>\n",
       "      <td>103927</td>\n",
       "      <td>17.14.11</td>\n",
       "      <td>2017-11-11T05:23:36.000Z</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>How To Dry a Shirt in 30 Seconds</td>\n",
       "      <td>[how to dry a shirt in 30 seconds, \"how to dry...</td>\n",
       "      <td>2063667</td>\n",
       "      <td>17.14.11</td>\n",
       "      <td>2017-11-09T18:36:48.000Z</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Surprise After Surprise | October Vlog</td>\n",
       "      <td>[jenn im, \"jenn im vlog\", \"imjennim\", \"october...</td>\n",
       "      <td>247319</td>\n",
       "      <td>17.14.11</td>\n",
       "      <td>2017-11-10T18:42:55.000Z</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                               title  \\\n",
       "0  What $4,800 Will Get You In NYC | Sweet Digs H...   \n",
       "1    A Smart... MUG?! - Take apart a Heated Thermos!   \n",
       "2   iPhone X vs Makeup Transformation (Face ID TEST)   \n",
       "3      John Lewis Christmas Ad 2017 - #MozTheMonster   \n",
       "4                  A VERY #EXTRA EBAY SHOPPING HAUL!   \n",
       "5                  EXTREME GOLDEN TAN TRANSFORMATION   \n",
       "6  Inside the Cheese-Aging Caves 30 Feet Under Br...   \n",
       "7                         ICE PRIMER? DOES IT WORK?!   \n",
       "8                   How To Dry a Shirt in 30 Seconds   \n",
       "9             Surprise After Surprise | October Vlog   \n",
       "\n",
       "                                                tags    views trending_date  \\\n",
       "0  [refinery29, \"refinery 29\", \"r29\", \"r29 video\"...   145921      17.14.11   \n",
       "1  [Smart mug, \"Heated thermos\", \"tech\", \"gift id...   120727      17.14.11   \n",
       "2  [iphone x, \"iphonex makeup\", \"iphonex test\", \"...  1456472      17.14.11   \n",
       "3  [christmas, \"john lewis christmas\", \"john lewi...  7224515      17.14.11   \n",
       "4  [ebay, \"haul\", \"cheap\", \"challenge\", \"extra\", ...   484185      17.14.11   \n",
       "5  [Extreme Tan, \"Tanning\", \"Tanning Bed\", \"How t...   300617      17.14.11   \n",
       "6  [cave, \"cheese\", \"goat\", \"rind\", \"sheep's milk...    92523      17.14.11   \n",
       "7  [ice primer, \"beauty hack\", \"weird beauty hack...   103927      17.14.11   \n",
       "8  [how to dry a shirt in 30 seconds, \"how to dry...  2063667      17.14.11   \n",
       "9  [jenn im, \"jenn im vlog\", \"imjennim\", \"october...   247319      17.14.11   \n",
       "\n",
       "               publish_time  count_tags  \n",
       "0  2017-11-12T16:00:01.000Z          35  \n",
       "1  2017-11-13T16:00:03.000Z          18  \n",
       "2  2017-11-12T00:45:54.000Z          17  \n",
       "3  2017-11-10T07:38:29.000Z           9  \n",
       "4  2017-11-11T02:42:23.000Z          12  \n",
       "5  2017-11-10T20:40:44.000Z          13  \n",
       "6  2017-11-10T16:01:04.000Z          28  \n",
       "7  2017-11-11T05:23:36.000Z           5  \n",
       "8  2017-11-09T18:36:48.000Z           9  \n",
       "9  2017-11-10T18:42:55.000Z           9  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "quantTagsDF = youtube.copy()\n",
    "\n",
    "countTagsSeries = quantTagsDF['tags']\n",
    "tags_new = list(countTagsSeries)\n",
    "\n",
    "new_tags = []\n",
    "count_tags = []\n",
    "\n",
    "# Split words in tags into a list of tags\n",
    "for s in tags_new:\n",
    "    s = s.split('|')\n",
    "    new_tags.append(s)\n",
    "\n",
    "# Make list of how many tags are in each video\n",
    "for i in new_tags:\n",
    "    count_tags.append(len(i))\n",
    "\n",
    "# Prepare count_tags to be added into dataframe\n",
    "count_tag = np.array(count_tags, dtype=np.int64)\n",
    "count_tag = pd.Series(count_tag)\n",
    "\n",
    "# Add count_tag to dataframe\n",
    "quantTagsDF['count_tags'] = count_tag.values\n",
    "quantTagsDF['tags'] = new_tags\n",
    "\n",
    "quantTagsDF.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "kheYTKVG0y_e"
   },
   "source": [
    "## Usage of Popular Tags vs Views Set-up"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "vsXJJV-_XThp"
   },
   "source": [
    "### To get the 10 most popular tags.\n",
    "\n",
    "We use value_counts method to count the number of mentioning tags and put it in dataframe (this step is done later in the data cleaning section)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "hFlBGmO5XCsd"
   },
   "outputs": [],
   "source": [
    "usePopTagsDF = youtube.copy()\n",
    "\n",
    "tags = usePopTagsDF['tags'].apply(pd.Series).stack()\n",
    "tags_df = usePopTagsDF['tags'].apply(pd.Series).stack()\n",
    "pop_tags = usePopTagsDF['tags'].apply(tuple).value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "IiWRqO3d02CG"
   },
   "source": [
    "## Title Sentiment vs Views Set-up\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "dCHWtjKSYoAY"
   },
   "source": [
    "Here we are performing a sentiment analysis on the titles of all the videos on our data set in order to determine if there is a relationship between the amount of views and its overall sentiment score. We used the Vader Lexicon from the NLTK package."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 52
    },
    "colab_type": "code",
    "id": "EilxZfrSYp9T",
    "outputId": "92562826-c096-40ad-ea03-11fea2d1351e"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package vader_lexicon to\n",
      "[nltk_data]     /Users/tma2017/nltk_data...\n",
      "[nltk_data]   Package vader_lexicon is already up-to-date!\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/tma2017/anaconda3/lib/python3.6/site-packages/nltk/twitter/__init__.py:20: UserWarning: The twython library has not been installed. Some functionality from the twitter package will not be available.\n",
      "  warnings.warn(\"The twython library has not been installed. \"\n"
     ]
    }
   ],
   "source": [
    "\n",
    "import nltk\n",
    "from nltk.sentiment.vader import SentimentIntensityAnalyzer\n",
    "\n",
    "titleSentDF = youtube.copy()\n",
    "\n",
    "nltk.download('vader_lexicon')\n",
    "analyser = SentimentIntensityAnalyzer()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "pNVB0xr9Ytnz"
   },
   "source": [
    "\n",
    "We write a function in order to allow us to view different subsets of our Youtube video dataframe. This will be helpful when we compare extremely viral videos versus the rest of the dataframe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "J5ZbiILWYu-c"
   },
   "outputs": [],
   "source": [
    "def title_sentiments(df):\n",
    "    title_sentiment = []\n",
    "    title_sentiment_ind = []\n",
    "\n",
    "    titles = df['title']\n",
    "\n",
    "    zero_sentiment = 0\n",
    "\n",
    "    # Add only non-zero sentiment analysis of titles into 'title_sentiment'\n",
    "    for i, title in zip(range(len(titles)), titles):\n",
    "        sentiment = analyser.polarity_scores(title)\n",
    "        \n",
    "        if sentiment['compound'] != 0:\n",
    "            title_sentiment.append(sentiment['compound'])\n",
    "            title_sentiment_ind.append(i)\n",
    "        if sentiment['compound'] == 0:\n",
    "            zero_sentiment += 1\n",
    "            \n",
    "    return title_sentiment, title_sentiment_ind, zero_sentiment"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "6JURUCgc05Su"
   },
   "source": [
    "## Upload Date vs Time to Trending Set-up"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "OiwL1672ep7Q"
   },
   "source": [
    "In this section, we are performing an analysis on the correlation of upload dates and trending dates and see how an average of how long it would take a content creator to get onto trending."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 326
    },
    "colab_type": "code",
    "id": "XMOPF0_qclqG",
    "outputId": "4e1e1803-a10b-48a3-cfdc-9e6db30a342e"
   },
   "outputs": [],
   "source": [
    "import datetime\n",
    "import researchpy as rp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "mPuxlAS0a4Ks"
   },
   "outputs": [],
   "source": [
    "uploadDF = youtube.copy() #We just wanted a simple way to copy over the DF without ruining the original one."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "0i2jQZSWB2RO"
   },
   "source": [
    "# Data Cleaning\n",
    "\n",
    "\"What methods did you use to analyze your data? Briefly explain what steps you had to take before you were able to use the datasets you chose to answer your question of interest.\"\n",
    "\n",
    "- Taking averages and t-testing. OLS regression and sentiment analysis.\n",
    "\n",
    "\"How clean is the data?\"\n",
    "\n",
    "- The data had extraneous columns beforehand\n",
    "\n",
    "\"What did you have to do to get the data into a usable format?\"\n",
    "\n",
    "- It was already 'usable' but just to make it neater and more concise, we eliminated some columns\n",
    "\n",
    "\"What pre-processing steps were required for your methods (for example checking data distributions and performing any transformations that may be required)\"\n",
    "\n",
    "- We didn't have to really make any transformations. To check data distributions, we printed out certain intervals or criteria so that we can see discrepancies."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "TaP76OTo0-aX"
   },
   "source": [
    "## Quantity of Tags vs Views Data Cleaning"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "g_fZst0cI9RZ"
   },
   "source": [
    "We only retain the columns of title, tag, views, and count_tags because they are the only ones relevant to this section.\n",
    "We drop duplicates by keeping only the last one. We do this because we do not want the same video to have the same tags double-counted. We get the last one because we want the video with the most views.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "oOnFPxA47mbk"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>title</th>\n",
       "      <th>tags</th>\n",
       "      <th>views</th>\n",
       "      <th>count_tags</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Three meals that cost me $1.50 each</td>\n",
       "      <td>[brothers green eats, \"budget cooking\", \"cooki...</td>\n",
       "      <td>77630</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>Chapter 28: I Love My Baby But, I Miss My Dad</td>\n",
       "      <td>[Whitney Port, \"I Love My Baby But\", \"Grief\", ...</td>\n",
       "      <td>55398</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>FACE SURGERY WITH MAKEUP</td>\n",
       "      <td>[makeup, \"makeup tutorial\", \"surgery\", \"transf...</td>\n",
       "      <td>471266</td>\n",
       "      <td>28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>Chocolate Orange Autumn Cupcakes | Cupcake Jemma</td>\n",
       "      <td>[cupcakes, \"cupcake\", \"cup cake\", \"cake\", \"cak...</td>\n",
       "      <td>122669</td>\n",
       "      <td>36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>My Every Day Autumn Makeup | Zoella</td>\n",
       "      <td>[zoe sugg, \"zoe\", \"sugg\", \"zoella\", \"beauty\", ...</td>\n",
       "      <td>1390440</td>\n",
       "      <td>22</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                               title  \\\n",
       "12               Three meals that cost me $1.50 each   \n",
       "13     Chapter 28: I Love My Baby But, I Miss My Dad   \n",
       "14                          FACE SURGERY WITH MAKEUP   \n",
       "15  Chocolate Orange Autumn Cupcakes | Cupcake Jemma   \n",
       "16               My Every Day Autumn Makeup | Zoella   \n",
       "\n",
       "                                                 tags    views  count_tags  \n",
       "12  [brothers green eats, \"budget cooking\", \"cooki...    77630          19  \n",
       "13  [Whitney Port, \"I Love My Baby But\", \"Grief\", ...    55398           6  \n",
       "14  [makeup, \"makeup tutorial\", \"surgery\", \"transf...   471266          28  \n",
       "15  [cupcakes, \"cupcake\", \"cup cake\", \"cake\", \"cak...   122669          36  \n",
       "16  [zoe sugg, \"zoe\", \"sugg\", \"zoella\", \"beauty\", ...  1390440          22  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "quantTagsDF = quantTagsDF[['title','tags','views','count_tags']]\n",
    "quantTagsDF = quantTagsDF.dropna()\n",
    "quantTagsDF.drop_duplicates(subset = 'title', keep = 'last',inplace = True)\n",
    "quantTagsDF.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "WiYJnpfH0-aa"
   },
   "source": [
    "## Usage of Popular Tags vs Views Data Cleaning"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "pfnAvUi4H8bc"
   },
   "source": [
    "For popular tags, we drop the duplicates of the counts subset. \n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "qL_1cCPs4Qkb"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>counts</th>\n",
       "      <th>tags</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>31</td>\n",
       "      <td>(c, u, p, c, a, k, e, s, |, \", h, o, w,  , t, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>27</td>\n",
       "      <td>(t, h, e,  , g, r, e, e, n,  , m, a, k, e, u, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>25</td>\n",
       "      <td>([, n, o, n, e, ])</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>23</td>\n",
       "      <td>(o, l, i, v, i, a,  , j, a, d, e, |, \", o, l, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>21</td>\n",
       "      <td>(j, a, c, l, y, n, h, i, l, l, 1, |, \", j, a, ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   counts                                               tags\n",
       "0      31  (c, u, p, c, a, k, e, s, |, \", h, o, w,  , t, ...\n",
       "2      27  (t, h, e,  , g, r, e, e, n,  , m, a, k, e, u, ...\n",
       "4      25                                 ([, n, o, n, e, ])\n",
       "7      23  (o, l, i, v, i, a,  , j, a, d, e, |, \", o, l, ...\n",
       "8      21  (j, a, c, l, y, n, h, i, l, l, 1, |, \", j, a, ..."
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pop_tagsDF = pd.DataFrame({'tags':pop_tags.index, 'counts':pop_tags.values})\n",
    "pop_tagsDF = pop_tagsDF.drop_duplicates(subset=['counts']) #Drop duplicates\n",
    "pop_tagsDF.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "tnrm177V0-ab"
   },
   "source": [
    "## Title Sentiment vs Views Data Cleaning\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "3UdQnW1zY1sj"
   },
   "source": [
    "We make sure to drop any duplicate rows that are a result of a video that trends for multiple days. While this information may be helpful when examining how long it takes for a video to become popular after being published, for the purposes of sentiment analysis it only adds more noise to the data.\n",
    "\n",
    "We also print out the first 50 titles that return non-zero sentiments using NLTK vader."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "yYIjbhjQY45F"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sentiment score of first 50 titles with non-zero sentiment scores:\n",
      "[0.5574, -0.7088, 0.4939, 0.4588, 0.2342, -0.2732, 0.7964, 0.0516, 0.2263, -0.3612, 0.4588, -0.4404, -0.1779, 0.4926, 0.7579, -0.128, 0.5106, 0.5859, -0.4619, 0.8225, 0.6239, 0.3818, 0.7096, -0.4767, 0.4728, 0.5719, 0.4019, 0.658, 0.2584, 0.4215, -0.34, 0.4019, 0.5859, -0.3612, 0.2023, -0.4215, 0.5106, 0.4939, 0.5719, 0.4278, 0.4404, 0.3182, 0.6739, -0.7604, 0.5859, 0.0989, 0.4404, 0.5859, 0.508, 0.6037]\n",
      "\n",
      "Index of the first 50 non-zero sentiment titles\n",
      "[1, 5, 7, 11, 23, 24, 25, 26, 27, 30, 32, 33, 34, 35, 38, 40, 45, 47, 51, 54, 56, 57, 58, 62, 63, 64, 65, 67, 69, 73, 74, 77, 80, 81, 82, 83, 84, 86, 87, 88, 89, 90, 91, 94, 95, 97, 103, 105, 106, 107]\n",
      "\n",
      "Number of sentiment videos: 280\n",
      "Number of Zero sentiment videos: 322\n"
     ]
    }
   ],
   "source": [
    "# Drop duplicate titles, we are only interested in the last title that was trending\n",
    "titleSentDF.drop_duplicates(subset='title', keep='last',inplace=True)\n",
    "\n",
    "title_sentiment, title_sentiment_ind, zero_sentiment = title_sentiments(titleSentDF)\n",
    "\n",
    "# Print out test results\n",
    "print('Sentiment score of first 50 titles with non-zero sentiment scores:')\n",
    "print(title_sentiment[:50])\n",
    "print('\\nIndex of the first 50 non-zero sentiment titles')\n",
    "print(title_sentiment_ind[:50])\n",
    "print('\\nNumber of sentiment videos: ' + str(len(title_sentiment_ind)))\n",
    "print(\"Number of Zero sentiment videos: \" + str(zero_sentiment))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "XnWb2Rlb0-ad"
   },
   "source": [
    "## Upload Date vs Time to Trending Data Cleaning"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To clean this data, we want to make sure our columns are in the same format. Therefore, we rewrite our strings so that they're following the same format."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "rfc_Hq6bbiNB"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>publish_time</th>\n",
       "      <th>publish_to_trending</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017-11-12</td>\n",
       "      <td>2 days</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017-11-13</td>\n",
       "      <td>1 days</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2017-11-12</td>\n",
       "      <td>2 days</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2017-11-10</td>\n",
       "      <td>4 days</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2017-11-11</td>\n",
       "      <td>3 days</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2017-11-10</td>\n",
       "      <td>4 days</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2017-11-10</td>\n",
       "      <td>4 days</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2017-11-11</td>\n",
       "      <td>3 days</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2017-11-09</td>\n",
       "      <td>5 days</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2017-11-10</td>\n",
       "      <td>4 days</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  publish_time publish_to_trending\n",
       "0   2017-11-12              2 days\n",
       "1   2017-11-13              1 days\n",
       "2   2017-11-12              2 days\n",
       "3   2017-11-10              4 days\n",
       "4   2017-11-11              3 days\n",
       "5   2017-11-10              4 days\n",
       "6   2017-11-10              4 days\n",
       "7   2017-11-11              3 days\n",
       "8   2017-11-09              5 days\n",
       "9   2017-11-10              4 days"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "uploadDF = uploadDF[['publish_time', 'trending_date']] #for this analysis, these are the only two relevant columns\n",
    "\n",
    "uploadDF['publish_time'] = uploadDF['publish_time'].str[:10].astype(str) # to better parse our data, we convert them to str\n",
    "\n",
    "uploadDF['trending_date'] = '20' + uploadDF['trending_date'] # modify trending_date so it is in the same format as publish_time\n",
    "uploadDF['trending_date'] = uploadDF['trending_date'].str.replace('.','-')\n",
    "\n",
    "uploadDF['trending_date'] = uploadDF['trending_date'].str[0:4]+uploadDF['trending_date'].str[7:10]+uploadDF['trending_date'].str[4:7]\n",
    "uploadDF.head()\n",
    "\n",
    "uploadDF['trending_date'] = pd.to_datetime(uploadDF['trending_date']) #convert both of the columns to 'datetimes'\n",
    "uploadDF['publish_time'] = pd.to_datetime(uploadDF['publish_time'])\n",
    "\n",
    "uploadDF['publish_to_trending'] = uploadDF['trending_date'] - uploadDF['publish_time'] #subtract the two times to get the amt to get to trending\n",
    "\n",
    "uploadDF = uploadDF.drop(columns=[\"trending_date\"]) #just drop trending date, irrelevant to us now\n",
    "\n",
    "uploadDF.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "Nd0j5JU3-KW8"
   },
   "source": [
    "# Data Analysis and Results\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "k_Kdppoz1FM6"
   },
   "source": [
    "## Quantity of Tags vs Views Data Analysis and Results"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this section, we focus on investigating whether the __total number of tags__ used in a video will affect the number of __views__ it get.\n",
    "\n",
    "In order to better see how the __count_tags__ column correlates with the __views__ column, we decided to graph the data in a scatter plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "FVH_LhJ_1QwX"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0,0.5,'num_views')"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1a14b38860>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# get the counts of the tags for each video\n",
    "counts = list(quantTagsDF['count_tags'])\n",
    "num_views = list(quantTagsDF['views'])\n",
    "\n",
    "plt.scatter(counts,num_views)\n",
    "plt.xlabel(\"counts\")\n",
    "plt.ylabel(\"num_views\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here we can see that there are two extremely large value, which indicates two videos that have extremely large views. Since these two datapoints are too far away from the rest of the datapoints, they do not represent the overall correlation between __count_tags__ and __views__ column. \n",
    "\n",
    "Therefore, to better to picture the the overall correlation between these two columns, we will remove the outliers from the datasets by removing all datapoints that are above 10 million views"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0,0.5,'Views')"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1c1f061ef0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Creating a new dataframe that only includes datapoints that have less or equal to 10 million views\n",
    "youtube_10m = quantTagsDF[quantTagsDF['views']<=10000000]\n",
    "counts = list(youtube_10m['count_tags'])\n",
    "num_views = list(youtube_10m['views'])\n",
    "\n",
    "# Plot again\n",
    "plt.scatter(counts,num_views)\n",
    "plt.xlabel(\"Number of Tags\")\n",
    "plt.ylabel(\"Views\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It's difficult to see any trend from this scatter plot. Therefore, we will build a linear model to try to predict our result. \n",
    "\n",
    "According to our hypothesis, we believed that the number of views is positively correlated with the number of tags, since videos with more tags are likely to see by more people. If our model shows that the two values are negatively correlated or not correlated, then it will prove our hypothesis is false. \n",
    "\n",
    "In the following graph, we will plot our prediction on top of our scatter plot to visualize see how well the model fits the data \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1c1f061ac8>]"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1c1f061a20>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the scatter plot again\n",
    "plt.scatter(counts,num_views)\n",
    "plt.xlabel(\"Number of Tags\")\n",
    "plt.ylabel(\"Views\")\n",
    "\n",
    "# Use ployfit to get the best fit line, and show it in the graph\n",
    "a1, b1 = np.polyfit(youtube_10m['count_tags'], youtube_10m['views'], 1)\n",
    "count_tags_range = np.arange(0., 70.)\n",
    "pred_views = a1 * count_tags_range + b1\n",
    "plt.plot(count_tags_range, pred_views, c  ='r')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As we can see from the graph, there doesn't seem to be like a strong correlation between the number of tags and views. To check if our intuition is correct, we will use scipy's linegress method to get the r value, which indicates how well are the two correlated. \n",
    "\n",
    "The r value ranges from -1 to 1. If the r value is -1, it means our data shows a perfectly negative relationship; if the r value is is 1, it means our data shows a perfectly postive relationship; if the r value is is 0, it means our data shows no relationship."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "r value is:  -0.0012811657995536235\n"
     ]
    }
   ],
   "source": [
    "import scipy.stats\n",
    "from scipy.stats import linregress\n",
    "\n",
    "slope, intercept, r_value, p_value, std_err = scipy.stats.linregress(youtube_10m['count_tags'], youtube_10m['views'])\n",
    "\n",
    "print(\"r value is: \",r_value)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the result, we can see that the r value is close to 0, which indicates that there is not a clear relationship between the number of tags used and the views of a video. \n",
    "\n",
    "Therefore, our hypothesis regarding the relationship between the number of tags used and the views is proven to be false"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "VSEYytP11FM_"
   },
   "source": [
    "## Usage of Popular Tags vs Views Data Analysis and Results"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "UnVq6r3rJlXd"
   },
   "source": [
    "For the popular tags, we create the top 10 counts from the dataframe. Then, we collerated it to the number of views of each tags. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "ZCO0leLJ1Ktd"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'[': 41588638,\n",
       " 'c': 3903763073,\n",
       " 'f': 3515759998,\n",
       " 'j': 1545645901,\n",
       " 'k': 3741729499,\n",
       " 'l': 4002388036,\n",
       " 'o': 4048073780,\n",
       " 's': 4022986226,\n",
       " 't': 3994705928,\n",
       " 'z': 1380854589}"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pop_tagsDF = pop_tagsDF.head(10)\n",
    "\n",
    "dict_popTags = {}\n",
    "for elem,count in zip(pop_tagsDF['tags'], pop_tagsDF['counts']) : #zip the tags and count of how many videos they're in\n",
    "    \n",
    "    dict_popTags[elem[0]] = count\n",
    "\n",
    "list_popTags = list(dict_popTags.keys()) #only extract the tags\n",
    "\n",
    "dict_popViews = {}\n",
    "\n",
    "for popTag in list_popTags:\n",
    "    \n",
    "    currViews = 0\n",
    "    for tags,views in zip(usePopTagsDF['tags'],usePopTagsDF['views']):\n",
    "        if(popTag in tags):\n",
    "            currViews = currViews + views #accumulate the views\n",
    "    \n",
    "    dict_popViews[popTag] = currViews\n",
    "    \n",
    "dict_popViews "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "BSWxm8KcYALm"
   },
   "source": [
    "Here, we calculate the mean of the views for the 10 most popular tags. We plan to use this to plot it against the views we get for each popular tag, and see if using a more popular tag will nececessarily mean that it garners more views.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "a4lyAZOA4iha"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'[': 1663545.52,\n",
       " 'c': 125927841.06451613,\n",
       " 'f': 219734999.875,\n",
       " 'j': 73602185.76190476,\n",
       " 'k': 207873861.05555555,\n",
       " 'l': 210652001.89473686,\n",
       " 'o': 176003207.82608697,\n",
       " 's': 201149311.3,\n",
       " 't': 147952071.4074074,\n",
       " 'z': 81226740.52941176}"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dict_popMeans = {}\n",
    "for tag in dict_popViews:\n",
    "    dict_popMeans[tag] = dict_popViews[tag] / dict_popTags[tag]\n",
    "\n",
    "    \n",
    "dict_popMeans "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "xDgT4C3LYZZS"
   },
   "source": [
    "Next, we plot the tags versus the mean views. The y-axis resembles the mean views, while the x-axis were the top 10 tags. Out of those tags, the most popular one is beauty."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "P3wbCK_L4k4w"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1c1f4f7b00>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "data = [['beauty', 5539308.064516129], ['Business Insider', 3573679.111111111], ['ysac', 620388.0], ['olivia jade', 523903.82608695654], ['DIY', 1071556.0], ['spongebob', 784910.3], ['lifestyle', 2926548.947368421], ['kim kardashian', 2576792.777777778], ['kinetic sand', 1193684.8823529412], ['cook', 577825.5]]\n",
    "df_pop = pd.DataFrame(data, columns = ['Tags', 'Mean Views'])\n",
    "ax = df_pop.plot.bar(x='Tags', y='Mean Views', rot=90) #plot\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "_YFVYgqt1FNC"
   },
   "source": [
    "## Title Sentiment vs Views Data Analysis and Results\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "ljKx8KceZAqO"
   },
   "source": [
    "First, we graph all of the video titles that returned a non-zero sentiment (i.e. the NLTK toolkit was able to return a sentiment score) against its number of views."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "QUg0LsbSY-0W"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0,0.5,'Number of views')"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXwAAAERCAYAAAB4jRxOAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvNQv5yAAAIABJREFUeJzt3XmcHHWd//HXp3uOZDIhM5MZIJALFFG5QjJCAgIKXisreLECuj9RJOKioq7rgrLq/jxQ9+eFu6sC6y2JCqsLKx7cgoQjE5IQQELAhIRAMiSZZI5Mz3T35/dHVU96Jj09NZOuufr9fDzmMd11frqq+vP91re+VW3ujoiITH6JsQ5ARERGhxK+iEiZUMIXESkTSvgiImVCCV9EpEwo4YuIlIlxl/DN7Admtt3M1kWY9ptmtjr8W29mbaMRo4jIRGTjrR++mZ0OdAA/cfdjhzHfR4AT3f39sQUnIjKBjbsavrv/CdiZP8zMXmJmvzezFjO718xeXmDWC4BloxKkiMgEVDHWAUR0LXCpuz9lZicD/wmcmRtpZvOAI4A7xyg+EZFxb9wnfDOrBU4BfmVmucHVAyY7H7jR3TOjGZuIyEQy7hM+QbNTm7svKDLN+cBloxSPiMiENO7a8Ady9z3AX83sPAALnJAbb2ZHA/XAijEKUURkQhh3Cd/MlhEk76PNbIuZXQy8G7jYzNYAjwHn5s1yAbDcx1t3IxGRcWbcdcsUEZF4jLsavoiIxGNcXbRtbGz0+fPnj3UYIiITRktLy4vu3hRl2nGV8OfPn8/KlSvHOgwRkQnDzDZFnVZNOiIiZUIJX0SkTCjhi4iUCSV8EZEyoYQvIlImlPBFRMqEEr6IjCvZrNPankJPASi9cdUPX0TKWzbrXHDdA7Rs2sWiefUsu2QxiYQNPaNEohq+iIwbOzp7aNm0i3TWadm0ix2dPWMd0qSihC8i40ZjbRWL5tVTkTAWzaunsbZqrEOaVNSkIyLjhpmx7JLF7OjsobG2irxfuZMSUMIXkXElkTCapg/8FVMpBTXpiIiUCSV8EZEyoYQvIlImlPBFRMqEEr6ISJlQwhcRKRNK+CIiZUIJX0SkTMR645WZbQTagQyQdvfmONcnIiKDG407bV/r7i+OwnpERKQINemIiJSJuBO+A380sxYzW1poAjNbamYrzWxla2trzOGIiJSvuBP+qe6+EPgb4DIzO33gBO5+rbs3u3tzU1NTzOGIiJSvWBO+u28N/28Hfg2cFOf6RERkcLElfDObZmbTc6+BNwDr4lqfiIgUF2cvnUOAX4c/YFAB3ODuv49xfSIiUkRsCd/dnwFOiGv5IiIyPOqWKSJSJpTwRUTKhBK+iEiZUMIXESkTSvgiImVCCV9EpEwo4YuIlAklfBGRMqGELyJSJpTwRUTKhBK+iEiZUMIXESkTSvgiImVCCV9EpEwo4YuIlAklfBGRMqGELyJSJpTwRUTKhBK+iEiZUMIXESkTSvgiImVCCV9EpEwo4YuIlAklfBGRMqGELyJSJpTwRUTKhBK+iEiZUMIXESkTsSd8M0ua2SNm9r9xr0tERAY3GjX8y4EnRmE9IiJSRKwJ38xmA2cD18e5HhERGVrcNfxvAZ8CsoNNYGZLzWylma1sbW2NORwRkfIVW8I3s78Ftrt7S7Hp3P1ad2929+ampqa4whERKXtx1vBPBc4xs43AcuBMM/tZjOsTEZEiYkv47n6lu8929/nA+cCd7v6euNYnIiLFqR++iEiZqBiNlbj73cDdo7EuEREpTDV8EZEyMWTCN7NTzWxa+Po9ZvYNM5sXf2giIlJKUWr43wW6zOwEgj71m4CfxBqViIiUXJSEn3Z3B84Fvu3u3wamxxuWiIiUWpSLtu1mdiXwHuB0M0sClfGGJSIipRalhv8uIAVc7O4vAIcD/xZrVCIiUnJRavhvBm5x96cA3P1Z1IYvIjLhREn484H3hD1zWoB7gT+5+5o4AxMRkdIasknH3T/r7mcCxwL3Af8ErIo7MBERKa0ha/hmdhXBg9BqgUeATxLU8kVEZAKJ0qTzdiAN/Ba4B3jA3btjjUpEREouSpPOQuAs4CHg9cCjZnZf3IGJiEhpRWnSORY4DTgDaAY2oyYdEZEJJ0qTzleBPwHXAA+7e2+8IYmISByGTPjufraZTQXmKtmLiExcUZ6W+RZgNfD78P0CM7s57sBERKS0ojxa4fPASUAbgLuvJrgZS0REJpCoT8vcHXskIiISqygXbdeZ2YVA0syOAj4K3B9vWCIiUmpRavgfAY4heGLmMmAP8LE4gxIRkdKL0kunC/hM+CciIhPUoAnfzL7l7h8zs1sAHzje3c+JNTIRESmpYjX8n4b//99oBCIiIvEaNOG7e0v4sgG41d1ToxOSiIjEIcpF23OA9Wb2UzM728yi9OwREZFxJsrTMt8HvBT4FXAh8LSZXR93YCIiUlqRauvu3mtmvyO4eDsVOBf4QJyBiYhIaUV5ls6bzOxHwAbgncD1wKwI800xs4fMbI2ZPWZm/3rA0YqIyIhFqeFfBCwHPjjMC7cp4Ex37zCzSuA+M/uduz8wgjhFROQARbnx6vyRLNjdHegI31aGf/v15xcRkdERpZfOiJlZ0sxWA9uB29z9wQLTLDWzlWa2srW1Nc5wRETKWqwJ390z7r4AmA2cFP5c4sBprnX3ZndvbmpqijMcEZGyNmjCN7M7wv9fPdCVuHsbcDfwpgNdloiIjEyxNvxZZnYGcI6ZLQcsf6S7ryq2YDNrAnrdvS38icTXEfw+roiIjIFiCf+zwBUEzTHfGDDOgTOHWPYs4MdmliQ4k/ilu//vSAMVEZEDU+xZOjcCN5rZv7j7F4a7YHdfC5x4IMGJiEjpROmW+QUzOwc4PRx0t2rqIiITT5Q7ba8GLgceD/8uD4eJiMgEEuVO27OBBe6eBTCzHwOPAFfGGZiIiJRW1H74dXmvZ8QRiIiIxCtKDf9q4BEzu4uga+bpqHYvIjLhRLlou8zM7gZeRZDw/9ndX4g7MBERKa2oz8N/Hrg55lhERCRGsT5LR0RExg8lfBGRMlE04ZtZwszWjVYwIiISn6IJP+x7v8bM5o5SPCIiEpMoF21nAY+Z2UNAZ26gu58TW1QiIlJyURK+fnxcRGQSiNIP/x4zmwcc5e63m1kNkIw/NBERKaUoD0+7BLgR+H446HDgN3EGJSIipRelW+ZlwKnAHgB3fwo4OM6gRESk9KIk/JS79+TemFkFwS9eiYjIBBIl4d9jZp8GpprZ64FfAbfEG5aIiJRalIR/BdAKPAp8ELgVuCrOoEREpPSi9NLJhj968iBBU86T7q4mHRGRCWbIhG9mZwPfA54meDzyEWb2QXf/XdzBiYhI6US58errwGvdfQOAmb0E+C2ghC8iMoFEacPfnkv2oWeA7THFIyIiMRm0hm9mbw9fPmZmtwK/JGjDPw94eBRiExGREirWpPOWvNfbgDPC161AfWwRyZjJZp0dnT001lZhZmMdjoiU2KAJ393fN5qByNjKZp0LrnuAlk27WDSvnmWXLCaRUNIXmUyi9NI5AvgIMD9/ej0eeXLZ0dlDy6ZdpLNOy6Zd7OjsoWl69ViHJSIlFKWXzm+A/yK4uzYbbzgyVhprq1g0r76vht9YWzXWIYlIiUVJ+N3ufs1wF2xmc4CfAIcSFBTXuvu3h7scGR1mxrJLFqsNX2QSi5Lwv21mnwP+CKRyA9191RDzpYF/dPdVZjYdaDGz29z98ZGHK3FKJEzNOCKTWJSEfxzw98CZ7GvS8fD9oNz9eeD58HW7mT1B8Cx9JXwRkTEQJeG/DTgy/xHJw2Vm84ETCZ7HM3DcUmApwNy5+q10EZG4RLnTdg1QN9IVmFktcBPwMXffM3C8u1/r7s3u3tzU1DTS1YiIyBCi1PAPAf5iZg/Tvw1/yG6ZZlZJkOx/7u7/PeIoRUTkgEVJ+J8byYIt6ObxX8AT7v6NkSxDRERKJ8rz8O8Z4bJPJbjY+6iZrQ6Hfdrdbx3h8kRE5ABEudO2nX2/YVsFVAKd7n5Qsfnc/T6C5+eLiMg4EKWGPz3/vZm9FTgptohERCQWUXrp9OPuv2GIPvgiIjL+RGnSeXve2wTQzL4mHhERmSCi9NLJfy5+GtgInBtLNCIiEpsobfh6Lr6IyCRQ7CcOP1tkPnf3L8QQj4iIxKRYDb+zwLBpwMXATEAJX0RkAin2E4dfz70OH298OfA+YDnw9cHmExGR8aloG76ZNQCfAN4N/BhY6O67RiMwEREprWJt+P8GvB24FjjO3TtGLSoRESm5Yjde/SNwGHAVsNXM9oR/7Wa232OORURkfCvWhj/su3BFRGT8UlIXESkTSvgiImVCCV9EpEwo4YuIlAklfBGRMqGELyJSJpTwRUTKhBK+iEiZUMIXESkTSvgiImVCCV9EpEwo4YuIlAklfBGRMqGELyJSJpTwRUTKRGwJ38x+YGbbzWxdXOsQEZHo4qzh/wh4U4zLFxGRYYgt4bv7n4CdcS1fRESGZ8zb8M1sqZmtNLOVra2tYx2OiMikNeYJ392vdfdmd29uamoa63BEZAxks05rewp3H+tQJrVBf8RcRGQ0ZLPOBdc9QMumXSyaV8+ySxaTSNhYhzUpjXkNX0TK247OHlo27SKddVo27WJHZ89YhzRpxdktcxmwAjjazLaY2cVxrUtEJq7G2ioWzaunImEsmldPY23VWIc0acXWpOPuF8S1bIlHNuvs6OyhsbYKM51Sy+gwM5ZdsljH3ihQG74AakeVsZVIGE3Tq8c6jElPbfgCqB1VpBwo4QugdlSRcqAmHQHUjipSDpTwpY/aUUUmNzXpiIiUCSV8EZEyoYQvQ9JzTkQmB7XhTwBjeUOU+ueLTB6q4Y9zuYS75Oo7OP/aB8hmR7eWPV775+usQ2T4lPDHubFOuI21VSycW0fSYOHcunHRPz/uQlCFiUxWSvjj3FjfEBXkPAML/sZDDoyzEBzrMyqROKkNf5wb6xuidnT2sOrZXWSyzqowuY51X/1cIZi7rlDKQrBQYTLWn1ekVJTwJ4CxvCEqzuQ6UnEWguPx84qUio2ndsrm5mZfuXLlWIchA5TbY5MP9POW2/aSsWVmLe7eHGVa1fBjNhm+/OX2yIUD+bwDu7H+/OKT2bW3d0Lvf5k8lPBjNF76sE+GQmeiyL8GsHLTLv7u2hWs3bJ7VPb/WOxnHVsTi3rpxGisu1SCep2MtvxeVcfPnsGazW2jsv+H2s9xdDUdjWNLXWRLSwk/RmPdpRLGR6FTTnIXlFdceRY3XbqE5vkNJdv/xZJfsf2czTrnX7uCxV++nXd9f0XJEnPcx1bUAkWFQnRq0onRWHepBPU6gbFrdijl/s8l7dx+XL50Sb/moWL7eVt7Nw9t3AXAQxt30dqR4pCDpoz8g4Uaaio5bvaMviarUh9bUbrIjpdm04lCCT9mw7kAGEdiGizplEvb62gnhCAxP0DLs2FivmRxSS54t3akiibtYvv5H362qt+yDuTT546bhppKLrz+QdZubuOEOXXc8IGTh3UcRTn+olRWSn3fxGT/XijhjxNxJqaBhU451YpG+0aq1vYUD23cCcBDf91Ja3uKQ2YceG164N7Jvc9PUIUqFzs6e1j73O6+9yfOrStJD6TjZ89gzZbdZBzWbtnNzq7e/Y6xwRJn1OMvyhlSKc9gy+F7oTb8cWI029oH9iRZv629aPtnfhvpRGsvHe3rKANzUqkqiU3TqzlpfgPJhHHSEQ00Ta+O1MbdWFtF87x6kgnjxLl13HTpkhHXXHd09rBy407SWWf1s20cP3tGwe06VFzDOf5yhZiZFTwOgb5rJsuXLj6gWnk5XO9SDX+cGKqmUspTzdy6Vm7aRU1VkrOvuZfm+Q39ajQDT91bNu1i4dx6wFn1bNuEqQGN9nWUpunVvGpeHS2b2lg0b+S16YHMjOVL+3+OFztSQ569lPLzN9RUMrUqSUcqQ011kl9espi27vR+TUjrt7WzcuNOMk7BuIIH8tXT8uy+42/RvAa+c+GJHBwm94Hya98L59YBxqpn99XEi23nqN+dcrjeNTkS/tKl0NsLiQQkk/3/Fxo22uMGDMti7E5lqJs+BUsmIZnEEgmWvXoGba+aTv30amzjxr55s5bg0hseYfXWdo6bU891F51EorKi8PrCh5wVk0sC67e1c/Y19/Z9MVvbUyQS1i/J9526Z52WZ3eB+6Bf5FIpdTvqaN445g5mCSxhWCIRvi/Nsgd+jmIJaqimnpF4sbOHjlQGgI5Uhp17e/tdR8gl5ZWbdlFTXUFXKl0wcQaVecezTnt3GoCHNu5kyZfvoHn+/hekYf/aN2Zkss7KjTtZv62dow+dPmRBMVQlZTx0sojb5Ej4994LXV2QyUA2u+9//utMBs97b9ksY/XoxwRQP8jwhkGGX5s/4JNDrMBsyAIpkUhwdDLJg3vTdGehsqICv95oz0CqsoIvZCBjRtYSTKmuoLPXmTKlkqwlaO/NMm1qFY0r/q1vuZ5I0uNQVV2JHUCB6Wbcsm4bz7f30DSjhjcefxjTplYFBeMIC98sRkfamV5ThVUMUlCWqLDfuTfNxr9spNaNp57sYOfzL6WxrqZwwXyA3OGa80/EjL5mD4ivLXqw6wg5uaScyTpdPRl++9HTCibi4IF8bWTD90kzMu5kGbwXUX7htnBuHZjRsnEnNdUVnP2d+2jO+5zZrNPakeqLL8o1nPwCslBPoGLXIyZSATE5Ev4TTww5ScEvgbFfoVCooIg6LtubZndnioOqEuzpTFE3JYm595uurSPFJ5avgkyGSnO+9rZjmVGdLLouz2T4r3s2sGVHJ/Pqp3DRyXMwd7LpNMse2MjzO7uYM6OK8xYeTmLA+orFbtksM9NpUj1pPJvlj2u3Yp4l6VnqqhJ07O2hrjrBXqC9K0WlgaczJJJZaiBYVm8vnsmw4YU9dHf3UFuZYH7DVCyTKbzeCNv2jT1pEtksCc9S8fts8R0bQQI46ICXEk0j8FD+gG8WmTiRwPMKARuiYPFkkgwGiQTJygq27k7RmXGmVFXCwdPxRIK0GyQTfGprO2kLltN7w0zSDjVTKvcVeAXW44kkqaxTPaWqYKHdZMZ/PL6d59t7OXjGFBq/+Ce6ss7UKVWQSDA1A19Y9wLP7Ukxq2EaL/vJOjqzMC1/vYkEjYkE/7R5I8/s6mbOzGm84dhZ/Pu9G3GMTCLJlP/ZA7VT+q3bEgmWzTf2HJ5gxrQUnkiy+dBK/vGmR+nFSD2fZPcxCWZMq+afblrLmhc6cTNeObuev5ma4NHn23nlnHoa27ZDe3K/M+6LfrKSVZv3cMK8Bn669BQSFcE0WWfQR2X4gHEToolzPF18i/Phaa3tKZZcfQfprFORMFZceVZJT/P7nc5WJelKpfdrFwdwD7vt9fWnjnahqVBNorU9xeKr7yCTdZIJ44ED+Ez5cR176DRWb+0oOn3S4P4rzyJhhrtzylfu3G/b5l8H2NkV7XkymUyW876/gtWb28hd60viLJh9EI9v3sWiOTP4wbsXctrX7mDv3h6mVye55+Onk/Asuzq6aRhQyO7c083537uPbCZLFc5PL2rma797nCe37uaVh9Tyxbe8ArJZ9nR2M6N6/wJ6YIGU7U1zzW1Psql1D0fOrOGy044ggfdNl01n+OG9T/Pcjk7m1VXz9yfP6VcIZ9Np9nb3UlNhkMnwm1Vb2N7Wyazp1bzlmIOD9RdYt6cz/OkvL7CjI0Uym6VhagXtnd2YOxWe5bSXNLBuSxudXSlmTEmSyGTp7O5hRlWCvd29eDZLJVmOO7Q2iCfvrDeTzpD0LC/s6qK3p5cpSaNpWmVQaGezeCaDZ4Lp96bCZXkWc8eyGZLuJPzAC+bxKm1Bs6qb4ckkaQxLJplaXUlbKkMmHDdzxlRIJEhUVAxZeO83buZM+PWvRxTfuHl4mpm9Cfg2kASud/evxLm+Ykp9QWZgAs4/nc21SxY6hYzSThj1NLGhppKaqiTt3WlqqpI01FRGjtmd/ZJxLq6f/PnpIRP+iXNm8JFlj7Bq0y4WzqtnweyDWPXsbk6cM4PG2irS6Sznhc+RKVQAFvqM2axz4fUPsmbLbk6YXUdFAh7ZvJvjZ9exestuMlbBiq1dtOzoYZtXwZQq9gDrs9V8/ubH+/V9zxWy9e7UL9zTt98zxx/PTbe+SPqQJtYljMuPW8hHlz1Cy6aeSLW01j3dfGv1HdAUvP+7d55FU21132fZ0dHDl5+7g8z8oBB+8wfPYua0qv0ugC86vJ5rLjiRj3/5jr5lL/nM6/qOlYHb58X2FBd96XZy1bNkwjj+8Bk8srkNgJqqBF0nBEk3YXDzh0+lLiyM3/GdP/eto6YqyapPv45ndnZSX1PJR5evZtWmNo6bfRCrN+/rvnn/P7+WZDIB7lz28xZWPrtvXCFJIOtZEu4ks9m+M8VEWBj8/iNLuOKXa3jiud2YB4XPdy88gX/46ap902WzvPKQGr7xjuPZ05kim0nz1d8+xtPbOqipMHp70tRUGKlUL+ZZjm6cyt+fPIcf3/dXntvRwVGN08hmMjzb2h7E4VmOaJjK5a89ku/d+RTPvtjBEfVTuOTU+ZDN0tGVwrIZyGT59u1P9s1zaG0VO/d0MXvGFN5xwixuXrWZbW17ObS2ghf3dGOZDEngXQtn8fCT23lxTzeHTKvA3Nm1Zy9N0yo5/cgGzIfRWjBKzUGxJXwzSwL/Abwe2AI8bGY3u/vjca1ziHhKetfjwFO5gT1fBrtgBcUvIhZaNpDXQ6G+rzfDzq5eulJB4dKVSu/XF3qw5eZ6OeR6SeQn45nTqrjt8W2DfvZjZk3niRfayWA8EhZwLZt2MaUiQRb4y7YOUqkM77xuBeue2wPQVwCu3LiTHZ09zJxW1Xc2dPzsGdz4wSUkk4l+heajz+3m3k+9lt17e3lJYw0Lv3RHX8G2cM4MkgYZD8406qZU9Ov7vn1PN8lkom8/5+93oF/Bb8DKcJ0rI1yI3u+o8f7762fvP4mayiTtqTQ1lUnqplT0jT9u9gzWbm7ru+jd0ZXqt6gq8wL7Ktjf9VMqMOhL+McdfhD/+e4TefVX7yLj0NWzr4Y9pTLB34ZJ/lXz65lSAeEuoKsnw8Iv/ZGu3v5n9vnJHuCi61ewfkf3oNthoAyAJfAE9CSS+40/64fr2JtOQO2+q1fn3LIV6g7tN93TGbjzN8/Tmfs8ycPhsMLrvB/44WNA/TFQD7cT7B+fGYw/9rCD+NaHT2VHZy9ff/IOMrOCQvi0c0/l8zc/xkM7ghvZjj98Os+cdgIdqTQJo+/MMmFwxqfO4q1hgT5zWiUXXPdg376+aOliXh9WnAY7yx1v4qzhnwRscPdnAMxsOXAuMCYJH0rXW2Owm3lyiWU4TRhRlg37Ljw9tHEnp1x9B83zG7jhAyfTPL8h0lnLYL0cBp6NAKxv7Rp0OU+80E7WYe3mNo6fU8ejW3bzilnTeTQvub/1e3/mLy/sO0PIfYlqqitoqKkM+nOHSfaRZ9s47/sruPHSU/a7MHf5L1azKkyUnd37Crand3SRCb+UGYdde3v7xfihn7fw6HN7+tXY8/d7fgGQzfqwzpJy/eFzZxNm1m+7PtXaQXtYCLen0jzV2tE3fu2W3Zwwp67vUQQ3r3mh37J/eP8mPvaGo/vtq9z+fsVhB5HfaLJm824uX76aRfMa+gruzlSaYw47qG9fADy8cRfHH1bL2vCMraYqSVdPpuhnBIaV7PMN9pievenoTcedPSNvHjph9gxWbwkKr6mVCcD6nQnjztnX3Ed+NGufa+97nR9/1uHDN6xi+dIlfcfPwEpj7oK5u0+ILp1xJvzDgc1577cAJw+cyMyWAksB5s6dG2M4pTNY81B+YhlpwTLYsnNnD5nsvm6RO7t6I5+1FOzlMMjZyMsOmc5fXtj3JXhp0zT+uqOLRXPrAPr64d/wgZPZ2dVLQ00FC75wO+3daaZVJ3kyL9m//NDavvddPZm+gvD42TN45NmgOWLN5rb9Cs38GtPaLbtZMHdfopw5rf+XqbG2mpPm1++rRYd3gA7WKyN/P+3s6ol8lgT794fP7Zvc/hoY28xp/fdnbps11lbR09PDt+/a0Dfth86Y329f5e/vx7fuobY62dct0sP98OcrziRh1lfJmDmtknd8934eyauxrwuTfQK44+On8fpv3du3nLHw6Te9jC//fn3f+9qqBB15Sb7CoFj58JLGGp55sQuDfoXgiXPq+O57FnLqV+4k40FzYK4Sk9vHmejlDhBs4/xjaLBK40Tp0hlnwi/0iffb3O5+LWGvw+bm5vFzBbmIOHfuYMtedsliXuxI8eEbVvUl3Nz4KIXLwOUWasPPreuWy07hZf/yB5ygdv67j76atu5Mv/kGrvuRq17PhtYOjjp4Ghde/xArN+7khDl1/OqDi7nw+of6FWBmxo0fXMJ531/Bms1tNM9v2K/QHFhjyk+UQL9a9sEHTWH50iUFT7uHqmk11lZFPkvKKXbGUCi2gfszN291dTVrr3odP7h/Ix86Yz7V1dX99tXA/f2z95/EhtYOPvs/63hkc1D45d+olFvujZeewju+d39fAWnsK6QPrath9b+8gfXb22moqeDMr99DV68zrSrBw1ecxT0bdnDWy2ey6Et30Z5KM60qydEH17B6SzsnzjmI9du7+s5g8iUTxsI5M3h4U9t+42qqEn1NTq+aV8fFpx3Jd+56pm/5D3zq1Rz7xXv6pl991WvoTCf77r51z3Lq1+4m68Hx+PvLT6OtO8OM6iRv//4KHt+6hxPm7LuLuND+bJ7fsF9z6zXnL+BDP1/FmvA6U2cqQ01Vgr29WWqqkuztyQyrtj4hfijI3WP5A5YAf8h7fyVwZbF5Fi1a5FJcJpP17Xu6PZvNxrqeVCrt9z/V6ul0etjzDoxxsJiH+izFxo903HDXMxKlXF6hZUVZfv40xabv7c34E1t3eyaTGXR4/vy9vRl/7Lk2f6Gty7u7e/3+p1q9p6e3b3x7R8qvu3uD793b7S+07fVtu/d6Op3pe52LYeB6d+zu8k/ftMa7uroKfp7Bjseo2yc3LJ3OFDw2c8MH/o/7e1ZBRBT4AAAIuUlEQVQKwEqPmJdj65ZpZhXAeuAs4DngYeBCd39ssHn0m7YiIsMzLrplunvazD4M/IGg19YPiiV7ERGJV6z98N39VuDWONchIiLR6PHIIiJlQglfRKRMKOGLiJQJJXwRkTIxrp6WaWatwKYRzt4IvFjCcEpFcQ2P4hoexTU8kzGuee7eFGXCcZXwD4SZrYzaF3U0Ka7hUVzDo7iGp9zjUpOOiEiZUMIXESkTkynhXzv0JGNCcQ2P4hoexTU8ZR3XpGnDFxGR4iZTDV9ERIpQwhcRKRMTKuGb2Xlm9piZZc1s0C5MZvYmM3vSzDaY2RV5w48wswfN7Ckz+4WZleR3yMyswcxuC5d7m5nVF5jmtWa2Ou+v28zeGo77kZn9NW/cgtGKK5wuk7fum/OGj+X2WmBmK8L9vdbM3pU3rqTba7DjJW98dfj5N4TbY37euCvD4U+a2RsPJI4RxPUJM3s83D53mNm8vHEF9+koxXWRmbXmrf8DeePeG+73p8zsvaMc1zfzYlpvZm1542LZXmb2AzPbbmbrBhlvZnZNGPNaM1uYN6702yrqg/PHwx/wCuBo4G6geZBpksDTwJFAFbAGeGU47pfA+eHr7wEfKlFcXwOuCF9fAXx1iOkbgJ1ATfj+R8A7Y9hekeICOgYZPmbbC3gZcFT4+jDgeaCu1Nur2PGSN80/AN8LX58P/CJ8/cpw+mrgiHA5yVGM67V5x9CHcnEV26ejFNdFwL8XmLcBeCb8Xx++rh+tuAZM/xGCR7bHvb1OBxYC6wYZ/2bgdwS/ELgYeDDObTWhavju/oS7PznEZH0/nu7uPcBy4FwzM+BM4MZwuh8Dby1RaOeGy4u63HcCv3P3wX8tvDSGG1efsd5e7r7e3Z8KX28FtgOR7iYcpoLHS5F4bwTOCrfPucByd0+5+1+BDeHyRiUud78r7xh6AJhdonUfUFxFvBG4zd13uvsu4DbgTWMU1wXAshKte1Du/ieCyt1gzgV+4oEHgDozm0VM22pCJfyICv14+uHATKDN3dMDhpfCIe7+PED4/+Ahpj+f/Q+2L4WndN80s1L9MGbUuKaY2UozeyDXzMQ42l5mdhJBre3pvMGl2l6DHS8Fpwm3x26C7RNl3jjjyncxQU0xp9A+Hc243hHunxvNbM4w540zLsKmryOAO/MGx7W9hjJY3LFsq1h/AGUkzOx24NACoz7j7v8TZREFhnmR4QccV9RlhMuZBRxH8EtgOVcCLxAktWuBfwb+7yjGNdfdt5rZkcCdZvYosKfAdGO1vX4KvNfds+HgEW+vQqsoMGzg54zlmBpC5GWb2XuAZuCMvMH77VN3f7rQ/DHEdQuwzN1TZnYpwdnRmRHnjTOunPOBG909kzcsru01lFE9tsZdwnf31x3gIrYAc/Lezwa2EjyYqM7MKsJaWm74AcdlZtvMbJa7Px8mqO1FFvV3wK/dvTdv2c+HL1Nm9kPgk6MZV9hkgrs/Y2Z3AycCNzHG28vMDgJ+C1wVnu7mlj3i7VXAYMdLoWm2WPBbzTMITtOjzBtnXJjZ6wgK0TPcPZUbPsg+LUUCGzIud9+R9/Y64Kt5875mwLx3lyCmSHHlOR+4LH9AjNtrKIPFHcu2moxNOg8DR1nQw6SKYOfe7MGVkLsI2s8B3gtEOWOI4uZweVGWu1/bYZj0cu3mbwUKXtGPIy4zq881iZhZI3Aq8PhYb69w3/2aoH3zVwPGlXJ7FTxeisT7TuDOcPvcDJxvQS+eI4CjgIcOIJZhxWVmJwLfB85x9+15wwvu01GMa1be23OAJ8LXfwDeEMZXD7yB/me6scYVxnY0wUXQFXnD4txeQ7kZ+D9hb53FwO6wQhPPtorjynRcf8DbCEq+FLAN+EM4/DDg1rzp3gysJyihP5M3/EiCL+QG4FdAdYnimgncATwV/m8IhzcD1+dNNx94DkgMmP9O4FGCxPUzoHa04gJOCde9Jvx/8XjYXsB7gF5gdd7fgji2V6HjhaCJ6Jzw9ZTw828It8eRefN+JpzvSeBvSny8DxXX7eH3ILd9bh5qn45SXFcDj4Xrvwt4ed687w+34wbgfaMZV/j+88BXBswX2/YiqNw9Hx7LWwiutVwKXBqON+A/wpgfJa/3YRzbSo9WEBEpE5OxSUdERApQwhcRKRNK+CIiZUIJX0SkTCjhi4iUCSV8mRTM7DO278maq83s5BEsY4GZvTnv/TlW4KmLpWRmrzGzU+Jch0jOuLvTVmS4zGwJ8LfAQg9u528keOzCcC0guBfgVgB3v5kCN++U2GuADuD+mNcjon74MvGZ2dsJbkx5y4Dhi4BvALUEj9a4yIPHOdwNPEjweOE6gpthHiS4wWUqwc1xV4evm939w2b2I2Av8HJgHvA+grtvlxA80vaicJ1vAP6V4JHJT4dxdZjZRoJnyrwFqATOA7oJnnKZAVqBj7j7vaXdOiL7qElHJoM/AnMs+FGL/zSzM8ysEvgOwXPzFwE/AL6UN0+Fu58EfAz4nAeP1P0swTPlF7j7Lwqsp57gIWAfJ3hA2DeBY4DjwuagRuAq4HXuvhBYCXwib/4Xw+HfBT7p7hsJfmfgm+E6lewlVmrSkQkvrEEvAk4jqLX/AvgicCxwW/DIHZIEt7jn/Hf4v4XgkRdR3OLuHj5NdJu7PwpgZo+Fy5hN8KMofw7XWUXeM1sGrPPt0T+hSGko4cuk4MGjbu8G7g4T8mXAY+6+ZJBZck+WzBD9e5CbJ5v3Ove+IlzWbe5+QQnXKVIyatKRCc/Mjjazo/IGLSB4QmNTeEEXM6s0s2OGWFQ7MP0AQnkAONXMXhqus8bMXhbzOkUiU8KXyaAW+LGFP+hN0KzyWYJHGX/VzNYQPE1yqO6PdwGvDLt1vmuIaffj7q0Ev+e6LIzjAYKLvMXcArwtXOdpw12nyHCol46ISJlQDV9EpEwo4YuIlAklfBGRMqGELyJSJpTwRUTKhBK+iEiZUMIXESkT/x+v0W0+F88+fwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x110f9aeb8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "views = list(titleSentDF['views'])\n",
    "views_sent = []\n",
    "\n",
    "for i in title_sentiment_ind:\n",
    "    views_sent.append(views[i])\n",
    "\n",
    "a_views, b_views = np.polyfit(title_sentiment, views_sent,1 )  \n",
    "sentiment_range = np.linspace(-1, 1, 10)\n",
    "pred_views = a_views*sentiment_range + b_views\n",
    "\n",
    "plt.scatter(title_sentiment, views_sent, s=5)\n",
    "plt.plot(sentiment_range, pred_views, c = 'red')\n",
    "plt.xlabel('Sentiment')\n",
    "plt.ylabel('Number of views')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "rvqfxmS5ZEj9"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0,0.5,'Frequency')"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x110f63358>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(title_sentiment, bins=20)\n",
    "plt.xlabel('Sentiment')\n",
    "plt.ylabel('Frequency')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "F8SPF2QIZHEk"
   },
   "source": [
    "Looking at the above two graphs, by inspection we have a feeling that there is not a strong correlation between sentiment and number of views. Representing the sentiment with a histogram, we see how the title sentiments are bimodal. More videos tend to populate the -0.50 and 0.50 areas of sentimentality. We also notice that there are some videos with large numbers of views (i.e. > 5,000,000 views).\n",
    "\n",
    "Next, we are going to run an OLS regression model to calculate some statistics and run statistical tests."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "rpiGB4ndZJnB"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/tma2017/anaconda3/lib/python3.6/site-packages/statsmodels/compat/pandas.py:56: FutureWarning: The pandas.core.datetools module is deprecated and will be removed in a future version. Please use the pandas.tseries module instead.\n",
      "  from pandas.core import datetools\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:        title_sentiment   R-squared:                       0.002\n",
      "Model:                            OLS   Adj. R-squared:                 -0.001\n",
      "Method:                 Least Squares   F-statistic:                    0.6447\n",
      "Date:                Tue, 11 Jun 2019   Prob (F-statistic):              0.423\n",
      "Time:                        18:16:20   Log-Likelihood:                -168.26\n",
      "No. Observations:                 280   AIC:                             340.5\n",
      "Df Residuals:                     278   BIC:                             347.8\n",
      "Df Model:                           1                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==============================================================================\n",
      "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "Intercept      0.2278      0.028      8.229      0.000       0.173       0.282\n",
      "views_sent -5.871e-09   7.31e-09     -0.803      0.423   -2.03e-08    8.52e-09\n",
      "==============================================================================\n",
      "Omnibus:                       38.681   Durbin-Watson:                   2.081\n",
      "Prob(Omnibus):                  0.000   Jarque-Bera (JB):               27.234\n",
      "Skew:                          -0.648   Prob(JB):                     1.22e-06\n",
      "Kurtosis:                       2.192   Cond. No.                     3.96e+06\n",
      "==============================================================================\n",
      "\n",
      "Warnings:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
      "[2] The condition number is large, 3.96e+06. This might indicate that there are\n",
      "strong multicollinearity or other numerical problems.\n"
     ]
    }
   ],
   "source": [
    "import patsy\n",
    "import statsmodels.api as sm\n",
    "\n",
    "data = {'title_sentiment': title_sentiment, 'views_sent': views_sent}\n",
    "sentiment_df = pd.DataFrame(data=data)\n",
    "\n",
    "outcome_1, predictors_1 = patsy.dmatrices('title_sentiment ~ views_sent', data=sentiment_df)\n",
    "mod_1 = sm.OLS(outcome_1, predictors_1)\n",
    "res_1 = mod_1.fit()\n",
    "\n",
    "print(res_1.summary())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "lttiw4HhZNep"
   },
   "source": [
    "\n",
    "From the above result, using an alpha of 0.01, we conclude that we don't have statistically significant evidence (i.e. Prob(F-stat) >> 0.01) that a youtube video's title sentiment is correlated with its number of views.\n",
    "\n",
    "However, what if this is due to those 'viral' videos (i.e. views > 5,000,000) that we mentioned above? In the next part we run seperate tests on videos with >5,000,000 views and <5,000,000 views."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "W4FyYF-VZSPw"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0,0.5,'Number of Views')"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1c1e6128d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "views_sent_norm = []\n",
    "\n",
    "youtube_norm = titleSentDF[titleSentDF['views'] < 5_000_000]\n",
    "\n",
    "views = list(youtube_norm['views'])\n",
    "\n",
    "title_sentiment_norm, title_sentiment_ind_norm, zero_sentiment_norm = title_sentiments(youtube_norm)\n",
    "\n",
    "for i in title_sentiment_ind_norm:\n",
    "    views_sent_norm.append(views[i])\n",
    "    \n",
    "a_norm, b_norm = np.polyfit(title_sentiment_norm, views_sent_norm, 1)\n",
    "sent_norm_range = np.linspace(-1, 1, 10)\n",
    "views_norm_pred = a_norm*sent_norm_range + b_norm\n",
    "\n",
    "\n",
    "plt.scatter(title_sentiment_norm, views_sent_norm, s=5, c='g')\n",
    "plt.plot(sent_norm_range, views_norm_pred, c = 'red')\n",
    "plt.xlabel('Sentiment')\n",
    "plt.ylabel('Number of Views')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "65DZ5PePZVZT"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0,0.5,'Frequency')"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x110f9a518>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(title_sentiment_norm, bins=20, color='green')\n",
    "plt.xlabel('Sentiment')\n",
    "plt.ylabel('Frequency')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "eFPqI6oyZWMr"
   },
   "source": [
    "In this scatterplot, the videos whose view counts exceed 5,000,000 are left out in order to observe the data clusters in better detail. When we fit a linear model, we observe a negative correlation between title sentiment and number of views, which is unexpected. Next we will see if this fit is statistically significant."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "VOeGKse9ZY9T"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:        title_sentiment   R-squared:                       0.016\n",
      "Model:                            OLS   Adj. R-squared:                  0.012\n",
      "Method:                 Least Squares   F-statistic:                     4.367\n",
      "Date:                Tue, 11 Jun 2019   Prob (F-statistic):             0.0376\n",
      "Time:                        18:16:21   Log-Likelihood:                -157.72\n",
      "No. Observations:                 271   AIC:                             319.4\n",
      "Df Residuals:                     269   BIC:                             326.6\n",
      "Df Model:                           1                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==============================================================================\n",
      "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "Intercept      0.2796      0.034      8.146      0.000       0.212       0.347\n",
      "views_sent -6.543e-08   3.13e-08     -2.090      0.038   -1.27e-07   -3.79e-09\n",
      "==============================================================================\n",
      "Omnibus:                       31.229   Durbin-Watson:                   2.092\n",
      "Prob(Omnibus):                  0.000   Jarque-Bera (JB):               27.305\n",
      "Skew:                          -0.695   Prob(JB):                     1.18e-06\n",
      "Kurtosis:                       2.304   Cond. No.                     1.43e+06\n",
      "==============================================================================\n",
      "\n",
      "Warnings:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
      "[2] The condition number is large, 1.43e+06. This might indicate that there are\n",
      "strong multicollinearity or other numerical problems.\n"
     ]
    }
   ],
   "source": [
    "data = {'title_sentiment': title_sentiment_norm, 'views_sent': views_sent_norm}\n",
    "sentiment_df_norm = pd.DataFrame(data=data)\n",
    "\n",
    "outcome_1, predictors_1 = patsy.dmatrices('title_sentiment ~ views_sent', data=sentiment_df_norm)\n",
    "mod_1 = sm.OLS(outcome_1, predictors_1)\n",
    "res_1 = mod_1.fit()\n",
    "\n",
    "print(res_1.summary())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "jbmQ2dJ9ZdMa"
   },
   "source": [
    "Once again, even when we exclude 'viral' videos we don't have statistically significant evidence that sentiment is correlated with the number of views a video gets.\n",
    "\n",
    "Lastly, let's see if there is a correlation between sentiment and number of views in 'viral' videos."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "k-plHQOGZf-L"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0,0.5,'Number of Views')"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1107ab208>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "views_sent_viral = []\n",
    "\n",
    "youtube_viral = titleSentDF[titleSentDF['views'] > 5_000_000]\n",
    "\n",
    "views = list(youtube_viral['views'])\n",
    "\n",
    "title_sentiment_viral, title_sentiment_ind_viral, zero_sentiment_viral = title_sentiments(youtube_viral)\n",
    "\n",
    "for i in title_sentiment_ind_viral:\n",
    "    views_sent_viral.append(views[i])\n",
    "\n",
    "a_views_viral, b_views_viral = np.polyfit(title_sentiment_viral, views_sent_viral, 1)\n",
    "title_sent_viral_range = np.linspace(-1, 1, 10)\n",
    "views_viral_pred = a_views_viral*title_sent_viral_range + b_views_viral\n",
    "\n",
    "plt.scatter(title_sentiment_viral, views_sent_viral, s=5, c='r')\n",
    "plt.plot(title_sent_viral_range, views_viral_pred, c = 'r')\n",
    "plt.xlabel('Sentiment')\n",
    "plt.ylabel('Number of Views')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "ewK1JPfyZiV7"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0,0.5,'Frequency')"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1c1e69df60>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(title_sentiment_viral, bins=20, color='red')\n",
    "plt.xlabel('Sentiment')\n",
    "plt.ylabel('Frequency')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "thsyU75ECxri"
   },
   "source": [
    "Now we observe a positive correlation when we fit a linear model on the data of just the 'viral' videos. Next we will run another test to see if this correlation is statistically significant."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "tnyqYlVGZkgl"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:        title_sentiment   R-squared:                       0.328\n",
      "Model:                            OLS   Adj. R-squared:                  0.232\n",
      "Method:                 Least Squares   F-statistic:                     3.422\n",
      "Date:                Tue, 11 Jun 2019   Prob (F-statistic):              0.107\n",
      "Time:                        18:16:22   Log-Likelihood:                -3.4625\n",
      "No. Observations:                   9   AIC:                             10.93\n",
      "Df Residuals:                       7   BIC:                             11.32\n",
      "Df Model:                           1                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==============================================================================\n",
      "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "Intercept     -0.3742      0.179     -2.092      0.075      -0.797       0.049\n",
      "views_sent  1.635e-08   8.84e-09      1.850      0.107   -4.55e-09    3.72e-08\n",
      "==============================================================================\n",
      "Omnibus:                        4.692   Durbin-Watson:                   3.127\n",
      "Prob(Omnibus):                  0.096   Jarque-Bera (JB):                1.609\n",
      "Skew:                           1.018   Prob(JB):                        0.447\n",
      "Kurtosis:                       3.381   Cond. No.                     2.69e+07\n",
      "==============================================================================\n",
      "\n",
      "Warnings:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
      "[2] The condition number is large, 2.69e+07. This might indicate that there are\n",
      "strong multicollinearity or other numerical problems.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/tma2017/anaconda3/lib/python3.6/site-packages/scipy/stats/stats.py:1390: UserWarning: kurtosistest only valid for n>=20 ... continuing anyway, n=9\n",
      "  \"anyway, n=%i\" % int(n))\n"
     ]
    }
   ],
   "source": [
    "data = {'title_sentiment': title_sentiment_viral, 'views_sent': views_sent_viral}\n",
    "sentiment_df_viral = pd.DataFrame(data=data)\n",
    "\n",
    "outcome_1, predictors_1 = patsy.dmatrices('title_sentiment ~ views_sent', data=sentiment_df_viral)\n",
    "mod_1 = sm.OLS(outcome_1, predictors_1)\n",
    "res_1 = mod_1.fit()\n",
    "\n",
    "print(res_1.summary())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "8nfIOcbRZnQO"
   },
   "source": [
    "At this point, even amongst 'viral' videos it is no surprise that there is no correlation between sentiment and number of views."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "la1zenyVZn12"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0,0.5,'Number of Views')"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x110f9aa58>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(title_sentiment_norm, views_sent_norm, s=5, c='g')\n",
    "plt.scatter(title_sentiment_viral, views_sent_viral, s=5, c='r')\n",
    "plt.xlabel('Sentiment')\n",
    "plt.ylabel('Number of Views')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "RvrH8KtZZqI1"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0,0.5,'Frequency')"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1c1eb606a0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(title_sentiment_norm, bins=20, color='green')\n",
    "plt.hist(title_sentiment_viral, bins=20, color='red')\n",
    "plt.xlabel('Sentiment')\n",
    "plt.ylabel('Frequency')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "Z491SADR1FNF"
   },
   "source": [
    "## Upload Date vs Time to Trending Data Analysis and Results"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "m4k9LqlJdABO"
   },
   "source": [
    "In this section, our goal was to find out how can the upload date of the video affect its time to reach trending on YouTube. Specifically, we wish to explore whether uploading a video on a weekend or a weekday made a difference in terms of minimizing its time to reach trending.\n",
    "\n",
    "First, we split the dataframe into two parts based on whether the upload time of the video is a weekend or not. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "zbLxKfwEdLrM"
   },
   "outputs": [],
   "source": [
    "# uploadDF_weekends contains all the videos that have upload time in the weekends\n",
    "uploadDF_weekends = uploadDF[uploadDF['publish_time'].dt.dayofweek >= 5] \n",
    "# uploadDF_weekdays contains all the videos that have upload time in the weekdays\n",
    "uploadDF_weekdays = uploadDF[uploadDF['publish_time'].dt.dayofweek < 5]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "Dxq-IwqLdgBR"
   },
   "source": [
    "Now, we wanted to find the average amount of days it would take for a video to reach trending from both the weekend and weekdays."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "0u0_be8Edkdo"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average time to trending\n",
      "weekends:  6 days 13:30:17.070600\n",
      "weekdays:  8 days 07:32:40.462934\n",
      "\n",
      "\n",
      "Standard Deviations\n",
      "weekends:  4 days 15:17:34.906387\n",
      "weekdays:  31 days 22:07:38.284176\n"
     ]
    }
   ],
   "source": [
    "mean_weekends= uploadDF_weekends[\"publish_to_trending\"].mean()\n",
    "mean_weekdays = uploadDF_weekdays[\"publish_to_trending\"].mean()\n",
    "print('Average time to trending')\n",
    "print(\"weekends: \",mean_weekends)\n",
    "print(\"weekdays: \",mean_weekdays)\n",
    "\n",
    "print('\\n')\n",
    "\n",
    "sd_weekends = uploadDF_weekends[\"publish_to_trending\"].std()\n",
    "sd_weekdays = uploadDF_weekdays[\"publish_to_trending\"].std()\n",
    "print(\"Standard Deviations\")\n",
    "print(\"weekends: \",sd_weekends)\n",
    "print(\"weekdays: \",sd_weekdays)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "49eWgyvtPLBF"
   },
   "source": [
    "To visualize our findings, we graphed the data in a bar chart"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "1xlvtiOgPO4R"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1c1ed2e278>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualizing the result\n",
    "objects = ('Weekends', 'Weekdays')\n",
    "y_pos = np.arange(len(objects))\n",
    "performance = [6,8]\n",
    "\n",
    "plt.bar(y_pos, performance, align='center', alpha=0.5)\n",
    "plt.xticks(y_pos, objects)\n",
    "plt.ylabel('Days')\n",
    "plt.title('Average Time to go from publish to trending')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "AIqTzQpXPURp"
   },
   "source": [
    "From the result above, we hypothesize that the average time to reach trending when uploading in weekends is shorter than that in weekdays. To further check our hypothesis, we wanted to apply a t-test to make sure that our results make sense."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "BdrXQDZ0d0cl"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/tma2017/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:1: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  \"\"\"Entry point for launching an IPython kernel.\n",
      "/Users/tma2017/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:2: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  \n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Independent t-test</th>\n",
       "      <th>results</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Difference (weekends_ptt - weekdays_ptt) =</td>\n",
       "      <td>-1.7517</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Degrees of freedom =</td>\n",
       "      <td>4144.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>t =</td>\n",
       "      <td>-1.6850</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Two side test p value =</td>\n",
       "      <td>0.0921</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Mean of weekends_ptt &gt; mean of weekdays_ptt p ...</td>\n",
       "      <td>0.0460</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Mean of weekends_ptt &lt; mean of weekdays_ptt p ...</td>\n",
       "      <td>0.9540</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Cohen's d =</td>\n",
       "      <td>-0.0623</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Hedge's g =</td>\n",
       "      <td>-0.0623</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Glass's delta =</td>\n",
       "      <td>-0.3777</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>r =</td>\n",
       "      <td>0.0262</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                  Independent t-test    results\n",
       "0        Difference (weekends_ptt - weekdays_ptt) =     -1.7517\n",
       "1                              Degrees of freedom =   4144.0000\n",
       "2                                               t =     -1.6850\n",
       "3                           Two side test p value =      0.0921\n",
       "4  Mean of weekends_ptt > mean of weekdays_ptt p ...     0.0460\n",
       "5  Mean of weekends_ptt < mean of weekdays_ptt p ...     0.9540\n",
       "6                                       Cohen's d =     -0.0623\n",
       "7                                       Hedge's g =     -0.0623\n",
       "8                                   Glass's delta =     -0.3777\n",
       "9                                               r =      0.0262"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "uploadDF_weekends[\"weekends_ptt\"] = uploadDF_weekends[\"publish_to_trending\"].dt.days\n",
    "uploadDF_weekdays[\"weekdays_ptt\"] = uploadDF_weekdays[\"publish_to_trending\"].dt.days\n",
    "\n",
    "descriptives, results = rp.ttest(uploadDF_weekends[\"weekends_ptt\"],uploadDF_weekdays[\"weekdays_ptt\"])\n",
    "results"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "0w3fYTEnd-BH"
   },
   "source": [
    "From the t-test above, we can see that the probability that our hypothesis is true is calculated to be 95.4%, which confirms our hypothesis. \n",
    "\n",
    "For a more specific analysis, we wanted to take a look at each individual day and see if it was any different from the data we got from generalizing the days of the weeks into two categories."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "Ckw288J3eSpL"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average time to reach trending\n",
      "Mon:  6 days 22:06:42.484472\n",
      "Tue:  8 days 21:10:56.859971\n",
      "Wed:  10 days 04:56:35.121951\n",
      "Thu:  7 days 14:59:13.911806\n",
      "Fri:  7 days 04:24:35.618374\n",
      "Sat:  5 days 23:26:03.064833\n",
      "Sun:  7 days 05:46:54.545454\n"
     ]
    }
   ],
   "source": [
    "uploadDF_mon = uploadDF[uploadDF['publish_time'].dt.dayofweek ==0]\n",
    "uploadDF_tue = uploadDF[uploadDF['publish_time'].dt.dayofweek ==1]\n",
    "uploadDF_wed = uploadDF[uploadDF['publish_time'].dt.dayofweek ==2]\n",
    "uploadDF_thu = uploadDF[uploadDF['publish_time'].dt.dayofweek ==3]\n",
    "uploadDF_fri = uploadDF[uploadDF['publish_time'].dt.dayofweek ==4]\n",
    "uploadDF_sat = uploadDF[uploadDF['publish_time'].dt.dayofweek ==5]\n",
    "uploadDF_sun = uploadDF[uploadDF['publish_time'].dt.dayofweek ==6]\n",
    "\n",
    "mean_mon = uploadDF_mon[\"publish_to_trending\"].mean()\n",
    "mean_tue = uploadDF_tue[\"publish_to_trending\"].mean()\n",
    "mean_wed = uploadDF_wed[\"publish_to_trending\"].mean()\n",
    "mean_thu = uploadDF_thu[\"publish_to_trending\"].mean()\n",
    "mean_fri = uploadDF_fri[\"publish_to_trending\"].mean()\n",
    "mean_sat = uploadDF_sat[\"publish_to_trending\"].mean()\n",
    "mean_sun = uploadDF_sun[\"publish_to_trending\"].mean()\n",
    "\n",
    "print(\"Average time to reach trending\")\n",
    "print(\"Mon: \",mean_mon)\n",
    "print(\"Tue: \",mean_tue)\n",
    "print(\"Wed: \",mean_wed)\n",
    "print(\"Thu: \",mean_thu)\n",
    "print(\"Fri: \",mean_fri)\n",
    "print(\"Sat: \",mean_sat)\n",
    "print(\"Sun: \",mean_sun)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "6VPUSKc3Pe3h"
   },
   "source": [
    "To visualize our findings, we graphed the data in a bar chart"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "4TOs5PRuGeF9"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1c1f06f2e8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "objects = ('Mon', 'Tue', 'Wed','Thu','Fri','Sat','Sun')\n",
    "y_pos = np.arange(len(objects))\n",
    "performance = [6,8,10,7,7,5,7]\n",
    "\n",
    "plt.bar(y_pos, performance, align='center', alpha=0.5)\n",
    "plt.xticks(y_pos, objects)\n",
    "plt.ylabel('Days')\n",
    "plt.title('Average Time to go from publish to trending')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the result above, we can see that uploading on Wednesday would help the video reach trending the quickest"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "bq1ScUHY8y6C"
   },
   "source": [
    "# Ethical Considerations and Privacy \n",
    "\n",
    "In general, because the dataset is viewable among the public on Kaggle, therefore it doesn't really violate the personal information of any individual. We didn't include any private video information to perform our analysis.\n",
    "\n",
    "The person who is negatively affected by your finding could be the owners. Even though they wouldn't mind sharing their videos, however, we are using their privacy informations regarding their video to perform our analysis, without their permission. The content that we observe such as how to & style. Our findings could negatively affect the variability of content found in the How To & Style category.  \n",
    "\n",
    "People who could make money off of our findings include potential content creators and companies that sponsor YouTube / these content creators. It is not necessarily an exploiting way to generate money, but some content can be deemed immoral (by viewers and other content creators.) But no personal information is going to be leaked because there is no such thing in our dataset.\n",
    "\n",
    "Also, the dataset could contain some biases that could create results that are unrepresentative of actual Youtube trends. Currently, we found that there is no specific group that is the most successful in this genre. Nevertheless, it is hard to see that this dataset is biased, as it seems that it was created using automated processes. Upon further research, the dataset has no mention of race, gender, occupation, or religious preference. Though, this does not change the fact that the creator may have altered their dataset. Therefore, the data that we've reported could be biased.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "JZ1xz8CFCFOt"
   },
   "source": [
    "# Conclusion and Discussion \n",
    "\n",
    "As a reminder, our hypothesis was : \"Titles that have positive sentiment will induce more interaction, thus yielding more views. Additionally, popular videos are more likely to be uploads that are on the weekends. Furthermore, we hypothesize that the more number of tags there are, the more views as well, in addition to using more popular tags.\"\n",
    "\n",
    "In the end, we discovered that for the date of upload, our hypothesis was correct. The upload date played a role in views attained because the average days to trending were lower on the weekends rather than the weekdays. As a result, it was faster to reach trending if one were to upload on a weekend and thus reap more views. We took this approach because we believed that people had more free time on the weekends to spend leisurely on YouTube. As a result, more leisure time would be correlated with more YouTube usage. Thus, more binge-watching videos on the weekends.\n",
    "\n",
    "\n",
    "The number of tags was also a similar concept. Since some content creators had more tags than other, they gained more exposure thanks to people exploring those tags. Therefore, the more exposure their video got via crossing with other tags increased their viewership. \n",
    "\n",
    "To further elaborate on the tags, we found out that the top ten popular tags were beauty, business insider, ysac, olivia jade, DIY, spongebob, lifesttyle, kim kardashian, kinetic sand, and cook from this dataset. With the most popular tag related to beauty. And the least popular tag related to cook. This is based on the number of views of those tags.\n",
    "\n",
    "Finally, for the sentiment analysis, in general, there tended to be more videos with positive sentiments than negative ones. However, based on our findings in this section, we conclude that there is no strong, univariate correlation between a youtube video's sentiment score (based on the NLTK vader's sentiment analysis) and its number of views. We did notice that videos tend to be bimodal around the +0.50 and -0.50 sentiment marks, although this has nothing to do with the popularity of the videos themselves. Additionally, there also was a tendency for viral videos (videos with over 5,000,000 views) to lean towards the negative sentiment side, although this could be due to a low viral video population.\n",
    "\n",
    "In general, one limitation of our project is that our study only inlcudes 4000 videos, whereas there are millions of trending on YouTube. This dataset only includes the videos in 2017-2018, which does not represent the YouTube throughout the years. If we have more observations, we could have a more powerful analysis. "
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "Final Project",
   "provenance": [],
   "toc_visible": true,
   "version": "0.3.2"
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}