In this project, we prepared a dataset of apartment rentals by cleaning and engineering features, including binary indicators for amenities and average prices by city and state. We scaled numerical features and trained an XGBoost regression model to predict rental prices, evaluating its performance using metrics like R², MAE, and RMSE.
To enhance usability, we created a dynamic dashboard with widgets for city and state selections, allowing real-time calculations of average prices and predictions. A summary template presents key insights, including average prices and prediction errors, making it accessible for stakeholders while ensuring the dataset retains critical features for analysis.