Airbnb Dynamic Pricing Using Machine Learning

Airbnb is a leading home and apartment-sharing company which have large number of listings. However, how to set an optimal price is a big challenge. While it also involves a large number of data. Thus, we consider dynamic pricing with machine learning techniques to solve this problem. This paper investigates the dynamic pricing on Airbnb using machine learning-based models. Linear regression, random forest, K-nearest neighborhood, AdaBoost Classifier, and Naïve Bayes are trained and tuned on the open dataset of Airbnb listings from New York in 2022. The main contribution of our paper is to find the most suitable price in Airbnb. The resulting models are compared based on R-squared (R2) and root mean square error (RMSE). Experiments show that the random forest model achieves an R2 of 0.997 and an RMSE of 0.038, which is the best model among these five models. The R2 of Naïve Bayes classifier is also ideal, but its RMSE is different from random forest.