Data ScienceXGBoostScikit-LearnRegression
Airbnb Prediction
Listing Price Estimation via Ensemble Tree Methods
Project Dossier
A data science project focused on predicting Airbnb listing prices based on various features such as location, room type, and availability. The goal was to build a model that helps hosts price their listings competitively.
Modeling Approach
Ensemble Learning
Leveraged XGBoost (Extreme Gradient Boosting) and Random Forest regressors. XGBoost provided superior performance by sequentially correcting errors of previous trees, while Random Forest offered robustness against overfitting.
Feature Engineering
Extensive preprocessing was performed using Pandas and NumPy, including handling missing values, encoding categorical variables, and normalizing numerical features to prepare the dataset for training.
Key Stats
Ensemble
Models
Price
Target
Sklearn
Stack
Actions