Classification Report

This report summarizes baseline classification results for employee attrition prediction.

Model Comparison

model	accuracy	precision	recall	f1	roc_auc	pr_auc
logistic_regression	0.795455	0.410714	0.657143	0.505495	0.803552	0.57677
xgboost	0.840909	0.5	0.142857	0.222222	0.719691	0.376778
random_forest	0.840909	0.5	0.0571429	0.102564	0.718687	0.35309

Notes

• Accuracy is not the main metric because the target is imbalanced.
• PR AUC, recall and F1-score are especially relevant for attrition detection.
• Logistic Regression is used as the interpretable baseline.
• Random Forest and XGBoost are used as nonlinear tabular baselines.