Rekayasa Fitur dan Ensemble Stacking untuk Meningkatkan Akurasi Klasifikasi Kinerja Karyawan Bebas Data Leakage Berbasis Rekam Jejak Digital
Keywords:
klasifikasi kinerja karyawan, ensemble stacking, rekayasa fitur, data leakage, permutation importanceAbstract
Klasifikasi kinerja karyawan secara otomatis menjadi kebutuhan penting dalam manajemen sumber daya manusia berbasis data, namun banyak model prediktif pada domain ini rentan terhadap masalah data leakage yang menghasilkan estimasi akurasi yang menyesatkan. Penelitian ini bertujuan membangun model klasifikasi kinerja karyawan yang bebas data leakage dengan memanfaatkan rekayasa fitur dan ensemble stacking berbasis rekam jejak digital (jam kerja, email, rapat, tidur, stres, dan variabel terkait lainnya). Metodologi penelitian meliputi audit dan mitigasi dua sumber leakage (variabel proksi Performance_Score dan baris data sintetis), penggabungan kelas Good dan Average menjadi kelas Menengah berdasarkan bukti empiris Cohen's d, penambahan empat fitur rekayasa baru (Productivity_Index, Skill_Support_Gap, Training_Efficiency, dan Balance_Stress_Ratio) di samping delapan fitur sebelumnya, serta pembangunan model stacking yang menggabungkan Random Forest, Gradient Boosting, dan Extra Trees dengan meta-learner Logistic Regression yang dioptimasi melalui RandomizedSearchCV. Hasil penelitian menunjukkan model stacking mencapai akurasi 0,9487 dan F1-score macro 0,7288, dengan recall macro tertinggi (0,8867) di antara seluruh model yang diuji, mengungguli Decision Tree, Random Forest, dan Gradient Boosting dalam mendeteksi kelas minoritas Low dan Excellent. Analisis permutation importance menunjukkan Physical_Activity_Steps, Base_Salary, dan Sleep_Duration sebagai fitur paling berpengaruh. Penelitian ini menyimpulkan bahwa penggabungan kelas yang dilandasi bukti statistik dan rekayasa fitur tambahan mampu meningkatkan performa klasifikasi secara sah tanpa mengorbankan validitas evaluasi model.
References
Konar, K., Das, S., Das, S., & Misra, S. (2025). Employee attrition prediction using Bayesian optimized stacked ensemble learning and explainable AI. SN Computer Science, 6(6), Article 672. https://doi.org/10.1007/s42979-025-04204-w
[2] Chung, D., Yun, J., Lee, J., & Jeon, Y. (2023). Predictive model of employee attrition based on stacking ensemble learning. Expert Systems with Applications, 215, Article 119364. https://doi.org/10.1016/j.eswa.2022.119364
[3] Akasheh, M. A., Malik, E. F., Hujran, O., & Zaki, N. (2024). A decade of research on machine learning techniques for predicting employee turnover: A systematic literature review. Expert Systems with Applications, 238, Article 121794. https://doi.org/10.1016/j.eswa.2023.121794
[4] Raza, A., Munir, K., Almutairi, M., Younas, F., & Fareed, M. M. S. (2022). Predicting employee attrition using machine learning approaches. Applied Sciences, 12(13), Article 6424. https://doi.org/10.3390/app12136424
[5] Park, J., Feng, Y., & Jeong, S.-P. (2024). Developing an advanced prediction model for new employee turnover intention utilizing machine learning techniques. Scientific Reports, 14(1), Article 1221. https://doi.org/10.1038/s41598-023-50593-4
[6] Yahia, N. B., Hlel, J., & Colomo-Palacios, R. (2021). From big data to deep data to support people analytics for employee attrition prediction. IEEE Access, 9, 60447–60458. https://doi.org/10.1109/ACCESS.2021.3074559
[7] Lazzari, M., Alvarez, J. M., & Ruggieri, S. (2022). Predicting and explaining employee turnover intention. International Journal of Data Science and Analytics, 14(3), 279–292. https://doi.org/10.1007/s41060-022-00329-w
[8] Duan, L., Paknejad, J., & Kim, H. (2025). Employee attrition prediction with convolutional neural network and synthetic minority over-sampling technique. Journal of Business Analytics, 8(1), 24–35. https://doi.org/10.1080/2573234X.2024.2399772
[9] Al Akasheh, M., Hujran, O., Faisal Malik, E., & Zaki, N. (2024). Enhancing the prediction of employee turnover with knowledge graphs and explainable AI. IEEE Access, 12, 77041–77053. https://doi.org/10.1109/ACCESS.2024.3404829
[10] Sasse, L., Nicolaisen-Sobesky, E., Dukart, J., Eickhoff, S. B., Götz, M., Hamdan, S., Komeyer, V., Kulkarni, A., Lahnakoski, J. M., Love, B. C., Raimondo, F., & Patil, K. R. (2025). Overview of leakage scenarios in supervised machine learning. Journal of Big Data, 12, Article 135. https://doi.org/10.1186/s40537-025-01193-8
[11] Huang, H.-N., Chen, H.-M., Lin, W.-W., Huang, C.-J., Chen, Y.-C., Wang, Y.-H., & Yang, C.-T. (2023). Employing feature engineering strategies to improve the performance of machine learning algorithms on echocardiogram dataset. Digital Health, 9, 1–17. https://doi.org/10.1177/20552076231207589
[12] Nemade, B. P., Bharadi, V., Alegavi, S. S., & Marakarkandy, B. (2023). A comprehensive review: SMOTE-based oversampling methods for imbalanced classification techniques, evaluation, and result comparisons. International Journal of Intelligent Systems and Applications in Engineering, 11(9s), 790–803.
[13] Wang, X., & Zhi, J. (2021). A machine learning-based analytical framework for employee turnover prediction. Journal of Management Analytics, 8(3), 351–370. https://doi.org/10.1080/23270012.2021.1961318
[14] Thanri, Y. Y., Safrizal, S., Riza, B. S., Iriani, J., & Lutfiana, T. (2026). Application of decision tree algorithm with hyperparameter tuning for income level classification based on demographic and occupational factors. IT Journal, 14(1), 86–100.
[15] Thanri, Y. Y., Iriani, J., Gowasa, A., & Zaidi, L. (2025). Model evaluasi kinerja karyawan berbasis rekam jejak digital menggunakan machine learning. CSRID (Computer Science Research and Its Development Journal), 17(3), 421–435.
[16] Biswas, A. K., Seethalakshmi, R., Mariappan, P., & Bhattacharjee, D. (2023). An ensemble learning model for predicting the intention to quit among employees using classification algorithms. Decision Analytics Journal, 9, Article 100335. https://doi.org/10.1016/j.dajour.2023.100335
[17] Alsheref, F. K., Fattoh, I. E., & Ead, W. M. (2022). Automated prediction of employee attrition using ensemble model based on machine learning algorithms. Computational Intelligence and Neuroscience, 2022, 1–9. https://doi.org/10.1155/2022/7728668
[18] Qutub, A., Al-Mehmadi, A., Al-Hssan, M., Aljohani, R., & Alghamdi, H. S. (2021). Prediction of employee attrition using machine learning and ensemble methods. International Journal of Machine Learning and Computing, 11(2), 110–114. https://doi.org/10.18178/ijmlc.2021.11.2.1022
[19] Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32. https://doi.org/10.1023/A:1010933404324
[20] Friedman, J. H. (2001). Greedy function approximation: A gradient boosting machine. The Annals of Statistics, 29(5), 1189–1232. https://doi.org/10.1214/aos/1013203451
[21] Wolpert, D. H. (1992). Stacked generalization. Neural Networks, 5(2), 241–259. https://doi.org/10.1016/S0893-6080(05)80023-1
[22] Bischl, B., Binder, M., Lang, M., Pielok, T., Richter, J., Coors, S., Thomas, J., Ullmann, T., Becker, M., Boulesteix, A.-L., Deng, D., & Lindauer, M. (2023). Hyperparameter optimization: Foundations, algorithms, best practices, and open challenges. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 13(2), Article e1484. https://doi.org/10.1002/widm.1484
[23] Massaoudi, M., Refaat, S. S., Chihi, I., Trabelsi, M., Oueslati, F. S., & Abu-Rub, H. (2021). A novel stacked generalization ensemble-based hybrid LGBM-XGB-MLP model for short-term load forecasting. Energy, 214, Article 118874. https://doi.org/10.1016/j.energy.2020.118874
[24] Talebi, S. (2025). Machine learning approaches for predicting employee turnover: A systematic review. Engineering Reports, 7(8), e70298. https://doi.org/10.1002/eng2.70298
[25] Susanto, P., Hoque, M. E., Jannat, T., Emely, B., Zona, M. A., & Islam, M. A. (2022). Work-life balance, job satisfaction, and job performance of SMEs employees: The moderating role of family-supportive supervisor behaviors. Frontiers in Psychology, 13, Article 906876. https://doi.org/10.3389/fpsyg.2022.906876
[26] Chawla, N. V., Bowyer, K. W., Hall, L. O., & Kegelmeyer, W. P. (2002). SMOTE: Synthetic minority over-sampling technique. Journal of Artificial Intelligence Research, 16, 321–357. https://doi.org/10.1613/jair.953
