Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{193732,
        author = {K. R. Rohith Kumar and S Rajasekhar and Vuppu Sri Sucharitha and Bommisetti Nethan and G Narasimha Reddy and Naruboyina Vijay Kumar},
        title = {EMPLOYEE ATTRITION PREDICTION USING MACHINE LEARNING},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {10},
        pages = {1807-1813},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=193732},
        abstract = {The increasing adoption of machine learning techniques by organizational decision-makers has encouraged researchers to investigate their applicability in addressing critical workforce challenges. Employee attrition, particularly the loss of skilled and experienced personnel, remains a significant concern for modern organizations. This study examines the effectiveness of machine learning models in predicting employee attrition using a synthetic dataset provided by IBM Watson. Three experimental scenarios were designed to evaluate model performance. In the first scenario, machine learning algorithms Support Vector Machine (with multiple kernel functions), Random Forest, and K-Nearest Neighbors were trained on the original class-imbalanced dataset. The second scenario addressed class imbalance through the Adaptive Synthetic Sampling (ADASYN) technique, followed by retraining the same models on the balanced data. The third scenario applied manual undersampling to achieve class balance. Experimental results indicate that the ADASYN-balanced dataset combined with the K-Nearest Neighbors algorithm (K = 3) produced the best performance, achieving an F1-score . Additionally, feature selection techniques integrated with the Random Forest model yielded an F1-score  while reducing the feature set from 29 to 12 attributes. The findings demonstrate the importance of data balancing and feature optimization in enhancing employee attrition prediction.},
        keywords = {hand gesture recognition, deep learning, convolutional neural networks, CNN, human-computer interaction, sign language, image processing, real-time recognition, gesture classification, assistive technology},
        month = {March},
        }