Employee Appraisal Prediction Using Classification Models

This paper presents a study that leverages machine learning techniques to determine employee performance using available data in organizations. The objective is to detect the key causes of attrition and minimize them through data analysis. The methodology involves data acquisition, conditioning, visualization, and classification using classification models like Random Forest, Support Vector Machine, K-Nearest Neighbor, Decision Tree, and Naive Bayes algorithms in Python. The results are evaluated using an accuracy score and confusion matrix, revealing that the Decision Tree XGBoost algorithm achieved the highest accuracy at 82.7%. However, Naive Bayes and Support Vector Machine models performed better in classifying true positives. This research provides valuable insights for organizations seeking to utilize data to understand and address attrition issues.