Perbandingan Logistic Regression, Random Forest, XGBoost untuk Prediksi Employee Attrition dengan SMOTE-Pipeline dan SHAP
DOI:
https://doi.org/10.52436/1.jpti.2416Kata Kunci:
Pengunduran Diri Karyawan, Feature Importance, Logistic Regression, Random Forest, SHAP, XGBoostAbstrak
Tingkat keluar-masuk karyawan (employee attrition) merupakan tantangan utama dalam manajemen sumber daya manusia karena berdampak langsung pada biaya operasional dan stabilitas organisasi. Penelitian ini bertujuan membandingkan performa tiga algoritma machine learning yaitu Logistic Regression, Random Forest, dan XGBoost dalam memprediksi attrition karyawan berdasarkan 1.057 data ulasan karyawan di Indonesia (employee reviews) beserta atribut skor kepuasan kerja. Ketidakseimbangan kelas pada label target ditangani menggunakan Synthetic Minority Over-sampling Technique (SMOTE) yang digabungkan dengan StandardScaler dalam satu pipeline untuk mencegah kebocoran data selama validasi silang. Optimasi hyperparameter dilakukan dengan RandomizedSearchCV dan validasi 5-fold stratified, serta pengujian stabilitas melalui repeated stratified k-fold sebanyak 50 kali pengulangan. Evaluasi pada data uji dilakukan dengan metrik Accuracy, Precision, Recall, F1-Score, Matthews Correlation Coefficient, dan ROC-AUC. Hasil menunjukkan XGBoost memberikan ROC-AUC tertinggi (0,598), diikuti Random Forest (0,573) dan Logistic Regression (0,564). Analisis feature importance dan SHAP (SHapley Additive exPlanations) secara konsisten mengidentifikasi fitur firm, years_working, overall, dan management sebagai faktor paling dominan dalam keputusan attrition. Nilai SHAP yang relatif kecil menunjukkan sinyal prediktif dari data yang tersedia masih lemah, sehingga akurasi model belum optimal meskipun pemodelan telah dilakukan secara sistematis. Temuan ini merekomendasikan penguatan program retensi pada aspek kepuasan kerja, manajemen, dan masa kerja karyawan. Secara implikatif, penelitian ini berkontribusi pada bidang pendidikan dan teknologi dengan menunjukkan penerapan praktis teknik pembelajaran mesin (SMOTE-Pipeline dan SHAP) yang dapat diadopsi sebagai bahan ajar analitik data dan sistem pendukung keputusan berbasis kecerdasan buatan di lingkungan pendidikan vokasi maupun industri.
Unduhan
Referensi
R. Qin, X. Qi, Y. Yuan, and B. Alatas, “Employee Turnover Prediction Research of Human Resource Management on Machine Learning Algorithms and Big Data Analysis,” Journal of Organizational and End User Computing, vol. 38, no. 1, Jan. 2026, doi: 10.4018/JOEUC.399146.
H. Talebi, A. K. Bardsiri, and V. K. Bardsiri, “Developing a hybrid machine learning model for employee turnover prediction: Integrating LightGBM and genetic algorithms,” Journal of Open Innovation: Technology, Market, and Complexity, vol. 11, no. 2, Jun. 2025, doi: 10.1016/j.joitmc.2025.100557.
A. Raza, K. Munir, M. Almutairi, F. Younas, and M. M. S. Fareed, “Predicting Employee Attrition Using Machine Learning Approaches,” Applied Sciences (Switzerland), vol. 12, no. 13, Jul. 2022, doi: 10.3390/app12136424.
Badan Pusat Statistik, “Pencari Kerja Terdaftar, Lowongan Kerja Terdaftar, dan Penempatan/Pemenuhan Tenaga Kerja Menurut Provinsi dan Jenis Kelamin, 2024,” BPS-Statistics Indonesia. Accessed: Jul. 29, 2026. [Online]. Available: https://www.bps.go.id/en/statistics-table/3/VEU5VVVERlVWM0JwYTNvdk1ISkpWR3R1VUhaVmR6MDkjMw==/pencari-kerja-terdaftar--lowongan-kerja-terdaftar--dan-penempatan-pemenuhan-tenaga-kerja-menurut-provinsi-dan-jenis-kelamin--2024.html?year=2024
C. Zhang and W. Han, “Ensembles of decision trees and gradientbased learning for employee turnover rate prediction,” PeerJ Comput. Sci., vol. 10, 2024, doi: 10.7717/PEERJ-CS.2387.
K. Konar, S. Das, S. Das, and S. Misra, “Employee Attrition Prediction Using Bayesian Optimized Stacked Ensemble Learning and Explainable AI,” SN Comput. Sci., vol. 6, no. 6, Aug. 2025, doi: 10.1007/s42979-025-04204-w.
S. Barman et al., “An Explainable Machine Learning Framework for Prediction of Employee Attrition,” in Lecture Notes in Networks and Systems, Springer Science and Business Media Deutschland GmbH, 2025, pp. 615–627. doi: 10.1007/978-981-96-2721-9_40.
Q. Ke and Y. Xu, “Integrating Textual Features with Survival Analysis for Predicting Employee Turnover,” Behavioral Sciences, vol. 16, no. 2, Feb. 2026, doi: 10.3390/bs16020174.
W. Wu and S. Fukui, “Using Human Resources Data to Predict Turnover of Community Mental Health Employees: Prediction and Interpretation of Machine Learning Methods,” Int. J. Ment. Health Nurs., vol. 33, no. 6, pp. 2180–2192, Dec. 2024, doi: 10.1111/inm.13387.
M. I. Alyousef, H. W. Khan, and M. U. Sattar, “A Hybrid Predictive Model for Employee Turnover: Integrating Ensemble Learning and Feature-Driven Insights from IBM HR Analytics,” Information (Switzerland), vol. 17, no. 2, Feb. 2026, doi: 10.3390/info17020208.
Y. Li, Y. Yang, P. Song, L. Duan, and R. Ren, “An improved SMOTE algorithm for enhanced imbalanced data classification by expanding sample generation space,” Sci. Rep., vol. 15, no. 1, Dec. 2025, doi: 10.1038/s41598-025-09506-w.
J. Park, Y. Feng, and S. P. Jeong, “Developing an advanced prediction model for new employee turnover intention utilizing machine learning techniques,” Sci. Rep., vol. 14, no. 1, Dec. 2024, doi: 10.1038/s41598-023-50593-4.
W. Li, “A transformer-based deep learning framework to predict employee attrition,” PeerJ Comput. Sci., vol. 9, 2023, doi: 10.7717/peerj-cs.1570.
?. T. Baydili and B. Tasci, “Predicting Employee Attrition: XAI-Powered Models for Managerial Decision-Making,” Systems, vol. 13, no. 7, Jul. 2025, doi: 10.3390/systems13070583.
S. Manafi Varkiani, F. Pattarin, T. Fabbri, and G. Fantoni, “Predicting employee attrition and explaining its determinants,” Expert Syst. Appl., vol. 272, May 2025, doi: 10.1016/j.eswa.2025.126575.
G. Marín Díaz, J. J. Galán Hernández, and J. L. Galdón Salvador, “Analyzing Employee Attrition Using Explainable AI for Strategic HR Decision-Making,” Mathematics, vol. 11, no. 22, Nov. 2023, doi: 10.3390/math11224677.
