With so many people connected online, it has never been easier for people to access crowd sharing resources online. Airbnb is one of those services, allowing everyday people to provide short-leases on their home to practically anyone in the world. As people tend to get a lot more space and amenities renting out an Airbnb than a traditional hotel, it’s a no brainer these types of crowd sharing services are picking up in popularity faster than no other. However, with home owners in charge of deciding the prices of their lease, rather than a huge monopolistic company controlling the prices everyone pays, is there a reason to believe that there is a trend involved in how prices are determined or is it pure random? This article will attempt to explore this question by building a supervised machine learning predictive model for Airbnb listing prices through analyzing tens of thousands of Airbnb listing data gathered throughout Paris, France.
This project was developed using Python 3.6.2 and the supervised machine learning predictive model was built using scikit-learn. View the source code on Jupyter Notebook.
The full report can be viewed in this GitHub Repository or following this link.
This was an open ended assignment for UCSD CSE 158 (Web Mining and Recommender Systems) and was posted with permission from Professor McAuley.