Predicting Airbnb Prices: A Study of Key Features Using Machine Learning

Short-term rentals have become more popular in the housing market, driven by platforms like Airbnb, which have largely transformed how people engage with vacation stays and temporary housing. This shift reflects a growing demand to understand key drivers for rental rates. In order to examine various factors influencing Airbnb rental costs, this study analyzes 74,111 Airbnb public data sets. The number of bedrooms, guest capacity, bathrooms, and review score ratings were determined to be the main predictors of the rental price of the property through the use of linear regression and decision tree modeling. The linear regression model achieved an R of 0.3567, the decision tree model improved the prediction with an R of 0.4384, while the random forest model further strengthened the predictability with an R of o.4605. The findings suggest that rental prices may increase significantly by adding more rooms to accommodate more guests and maintaining a higher review score. These findings offer crucial information for policymakers to comprehend the economic effect of short-term rentals, as well as insightful advice for homeowners looking to maximize listings and travelers making well-informed decisions.