CatBoost

CatBoost is a gradient boosting library that handles categorical features automatically and reduces overfitting through ordered boosting (Prokhorenkova et al., 2018). It is particularly useful when working with tabular data that contains categorical variables and is known for its ease of use and robust performance.

Key Features

Native handling of categorical variables without preprocessing.
Ordered boosting to reduce overfitting.
Excellent default hyperparameters for quick model building.
Supports GPU acceleration for faster training.
Integrates with Python, R, and other major ML frameworks.

Official Documentation

CatBoost Documentation

Notes

In this study, CatBoost was selected for its ability to handle categorical data effectively, minimizing preprocessing efforts and improving model generalization.