-- Exploration on the Boston Airbnb Data
This is a project with a theme of "becoming a data driven Airbnb host. We explore the Airbnb data assuming that we are an Airbnb host in Boston. Topics include:
- Descriptive analysis on the listings and the hosts data
- Aspect-opinion analysis on the texts data including decriptions about listings and guests' comments
- Gradient boosting models to predict future earnings using the current information about the listings and the hosts
- Analysis on variables that have large impact on the future earnings
Data: https://www.kaggle.com/airbnb/boston
The 2 notebooks in this repo can also be found here --> Technical Blog (on Kaggle):
Data exploration
Predict future earnings
For results and insights --> Non-technical Blog: Become a data-driven Airbnb host
Supporting file:
ReviewMiner is a tool for analyzing text data I built in another project. I haven't included a documentation yet but there are some example notebooks in the repo
Libaraies used: pandas, numpy, matplotlib, datetime, reviewminer, seaborn, sklearn