Predicting Airbnb Rental Prices Using Multiple Feature Modalities

TL;DR

This paper presents a multi-modal approach to predict Airbnb rental prices by integrating geolocation, temporal, visual, and natural language features to improve accuracy and utility for hosts and guests.

Contribution

It introduces a novel multi-modal feature integration method for Airbnb price prediction, combining diverse data types for enhanced accuracy.

Findings

Achieved improved price prediction accuracy over baseline models

Demonstrated the effectiveness of multi-modal features in rental price estimation

Provided insights into key drivers influencing Airbnb rental prices

Abstract

Figuring out the price of a listed Airbnb rental is an important and difficult task for both the host and the customer. For the former, it can enable them to set a reasonable price without compromising on their profits. For the customer, it helps understand the key drivers for price and also provides them with similarly priced places. This price prediction regression task can also have multiple downstream uses, such as in recommendation of similar rentals based on price. We propose to use geolocation, temporal, visual and natural language features to create a reliable and accurate price prediction algorithm.