Analysis on Melbourne AirBnB Open Data

In this Project, we are going to explore the Airbnb data of Melbourne which includes 3 files:
- listings.csv
- calendar.csv
- reviews.csv

listings.csv: consists of details of all the listings in Melbourne including their price, accommodates, ratings, number of reviews, summary, name, owner name, Description, host Id and many other columns describing details of listings. This file contains 22,909 rows x 106 columns i.e. 2,428,354 values.
calendar.csv: calendar.csv consists of details of listings and its availability and its price. This file contains 526,414 rowsx6 columns i.e. 3,158,484 values.
reviews.csv:* reviews.csv consists of reviews for each listing in Melbourne. This file contains 1,048,575 rows x 7 columns i.e. 7,340,025 values.

We did the project as 2 parts:

• Part-1: Predicting the Yield using NLP
• Part-2: Analysis and Visualization of the data.
  • Analysis of the factors that affect the prices of listings
  • Analysis to recommend the users on where to invest a property in Melbourne
  • Analysis on the Seasonal Pattern of Prices
  • Analysis of reviews and their relationship with the prices
  • Host analysis and recommendation for prices

Technologies/Libraries Used

• Python 3
  • pandas
  • scikit-lean
• Jupyter Notebook

Additional libraries for visualization:

• pyLDA library
• Bokeh JS

Installation

You can install the above packages using pip:

pip install pandas
pip install -U scikit-learn
pip install bokeh
pip install pyldavis

Note: In the Analysis to recommend the users on where to invest a property in Melbourne, please fill your own Google API key near the api_key variable.