Airbnb Price Prediction

“Airbnb Office” by Open Grid Scheduler / Grid Engine is marked with CC0 1.0.

Using Pandas, sk-learn, and XGBoost, I trained a series of models to predict Airbnb listing price for a sample of listings in European cities.

The bulk of the work was put into EDA, feature engineering and feature selection.

Nine models were tested, including linear regression, random forest, and a neural network.

All in all, the best model was a Random Forest with an RMSE of 84.

Here are some charts from that project: