PREDICTING AIRBNB PRICES USING MACHINE LEARNING ALGORITHMS
DOI:
https://doi.org/10.24867/12BE37CabriloKeywords:
Machine learning, Airbnb, Price prediction, Regression, ClassificationAbstract
In this paper, the price of Airbnb accommodation was predicted using multiple machine learning algorithms. The dataset was downloaded from the insideairbnb.com website. The price prediction was based on the values of the 62 attributes, which describe the accommodation, and on the sentiment of the user reviews. Sentiment of the each user review was calculated and the average value of the review sentiment was determined for each accommodation. Multiple exploratory data analysis techniques and feature selection algorithms were applied. Both regression and classification algorithms were used. Following algorithms were selected: linear regression, LASSO regression, ridge regression, support vector regression, Naive Bayes classification, Random Forest classification SVM classification.
References
[2] E. Tankg, K. Sangani, „Neighborhood and Price Prediction for San Francisco Airbnb Listings ”, Stanford, 2015
[3] H. Yu, J, Wu, „Real Estate Price Prediction with Regression and Classification”, Stanford, 2016
