Skip to content

A platform that enables users make faster and more informed Airbnb decisions.

Notifications You must be signed in to change notification settings

spapadopoulos/AirbnBoost

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

45 Commits
 
 
 
 
 
 

Repository files navigation

AirbnBoost

AirbnBoost is an online platform powered by machine learning that enables users to make faster and better informed Airbnb decisions.

Features

  • See how each listing is priced compared to similar listings based on AirbnBoost's price prediction algorithm. Find great deals based on the deviation between the pricing algorithm prediction and the actual listing price.
  • Select listings that match your preferences without the need to read through full descriptions.
  • Understand the urban environment surrounding a listing, such as proximity to subway or neighborhood noise levels.

Tutorial

Here is a step-by-step navigation at AirbnBoost UI.

Implementation

The implementation has three main components:

  • 01 - data_preprocessing.ipynb, a notebook that geolocates and spatially joins Airbnb listing data with urban data sets and processes them before they are used as inputs in the modeling part.
  • 02 - data_modeling.ipynb, a notebook that trains the two models behind AirbnBoost. An LDA model extracts the topics from the listing descriptions. XGBoost is the algorithm behind the pricing model.
  • app.py, is a Flask web app deployed on Heroku cloud.

License

Copyright © 2019 Sokratis Papadopoulos. All rights reserved.

About

A platform that enables users make faster and more informed Airbnb decisions.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published