The objective of this research is to predict the nightly price of an airbnb. Our focus is a dataset of 74,111 airbnbs in New York City, Los Angeles, San Francisco, Washington DC, Boston, and Chicago. We wanted to see if the general price per night could be predicted, based on different features of the given airbnb. To accomplish this goal, we created and analyzed seven regression models: linear regres-sion, random forest, decision tree, ADAboost bagging with decision tree as base estimator, ADAboost bagging with linear regrgession as base estimator, bagging with decision tree as base estimator, and bagging with linear regression as base estimator. We experimented with different scaling and feature selection methods with these models.
Repository contains the DataMiningFinal.ipynb colaboratory python file, Airbnb.pdf report, and Airbnb Model Presentation.pdf.