Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal Jurnal Pengabdian Masyarakat dan Riset Pendidikan
Faoziyah, Nindy
Unknown Affiliation
Humanities Education Languange, Linguistic, Communication & Media Mathematics Other

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Stay or Leave? Predicting Employee Retention With Hybrid Deep Learning Models: Penelitian Desvia , Yessica Fara; Azhar, Wafiqah Yasmin; Supriyadi, Supriyadi; Sihombing, J. Rolles Herwin; Faoziyah, Nindy
Jurnal Pengabdian Masyarakat dan Riset Pendidikan Vol. 4 No. 4 (2026): Jurnal Pengabdian Masyarakat dan Riset Pendidikan Volume 4 Nomor 4 Tahun 2026
Publisher : Lembaga Penelitian dan Pengabdian Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31004/jerkin.v4i4.6345

Abstract

Employee retention is a major challenge for organizations in the digital era because high turnover impacts productivity, operational costs, and organizational performance. This study proposes a Hybrid Deep Learning model based on XGBoost and Deep Neural Network (DNN) to predict employee retention using the HR_comma_sep dataset. This approach combines tree-based machine learning and deep learning to capture nonlinear relationships and complex decision patterns. Data preprocessing is performed through feature scaling and categorical encoding before model training. The hybrid architecture is built by integrating the probability outputs of XGBoost and DNN in the meta-classification layer. Evaluation using Accuracy, Precision, Recall, F1-Score, and AUC-ROC shows that the hybrid model has better prediction and generalization performance than conventional methods. SHAP Explainability is used to identify the main factors influencing turnover, namely job satisfaction, average monthly working hours, and length of service. This model can help organizations develop proactive HR management strategies.
Co-Authors Azhar, Wafiqah Yasmin Desvia , Yessica Fara Sihombing, J. Rolles Herwin Supriyadi Supriyadi