Identifying potential scam listings on Airbnb

By Medium - 2021-03-17

Description

I recently worked on a project using data on Airbnb listings in London. The initial aim of my project was to use machine learning to predict rental prices based on features of the property, its…

Summary

  • Introduction I recently worked on a project using data on Airbnb listings in London.
  • However hosts can easily bypass the cap by re-listing properties with different addresses or photos or by creating multiple host accounts, to avoid algorithms designed to detect duplicates.
  • WIRED) and a network graph from the data (Image by author) Typically you would not expect to find a small group of hosts with listings regularly reviewed by the same small group of reviewers.
  • Conclusion The WIRED article uncovered how the business of hosting on Airbnb (and other similar platforms) is becoming increasingly systemised with businesses able to make significantly more from short-term lets than long-term lets, if they can bypass the 90-night limit.

 

Topics

  1. Machine_Learning (0.32)
  2. Backend (0.24)
  3. Database (0.21)

Similar Articles

By Towards Data Science - 2021-01-27

A Medium publication sharing concepts, ideas, and codes.

What was the data approach to breaking The FinCEN Files?

By diginomica - 2020-09-24

How do you work backwards from thousands of disconnected PDFs to the 100,000 suspicious financial transactions they’re reporting on? A combination of hard human work, OCR, data analysis and graph.

15 Essential Steps To Build Reliable Data Pipelines

By Medium - 2020-12-01

If I learned anything from working as a data engineer, it is that practically any data pipeline fails at some point. Broken connection, broken dependencies, data arriving too late, or some external…