EUVIMEDEUVIMED entdecken
Open Access · CC BY-SA 4.0

Behavioral Absenteeism Pattern Classification for Early Employee Intervention Using Machine Learning

Yuricha Yuricha

bit-Tech · 10.08.2026 · 9 (1) · 1578-1589

DOI: 10.32877/bt.v9i1.4151 ↗

Kurz gefasst

Fachartikel in der Zeitschrift bit-Tech, online erschienen am 10.08.2026. Verfasst von Yuricha Yuricha. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Employee absenteeism is commonly evaluated using cumulative indicators, such as absence frequency and duration, which provide limited insight into behavioral absenteeism patterns reflected in employee attendance over time. This study develops a supervised machine learning framework for classifying behavioral absenteeism patterns into Routine Monitoring and Early HR Review categories to support early employee intervention. Daily attendance records of 100 employees (24,800 observations) collected between June 2025 and May 2026 were transformed into employee-level behavioral profiles using three interpretable indicators: the Weekend Extension Index (WEI), Holiday Proximity Index (HPI), and Attendance Entropy. The Bradford Factor was used only to construct the target classes and was excluded from the predictor variables. Five classification algorithms, namely Dummy Classifier, Logistic Regression, Support Vector Machine, XGBoost, and Random Forest, were comparatively evaluated using repeated stratified cross-validation and an independent hold-out test. Random Forest achieved the strongest overall classification performance, obtaining an Accuracy of 0.85, Balanced Accuracy of 0.84, Precision of 0.83, Recall of 0.91, F1-score of 0.87, ROC-AUC of 0.96, and PR-AUC of 0.97. Permutation importance analysis further identified Attendance Entropy as the most influential behavioral indicator. These findings demonstrate that behavioral absenteeism patterns can be represented using a compact set of interpretable behavioral indicators and effectively classified to support early Human Resource review and timely employee intervention.

Zitieren

APA

Yuricha, Y. (2026). Behavioral Absenteeism Pattern Classification for Early Employee Intervention Using Machine Learning. bit-Tech, 9(1), 1578-1589. https://doi.org/10.32877/bt.v9i1.4151

Vancouver

Yuricha Y. Behavioral Absenteeism Pattern Classification for Early Employee Intervention Using Machine Learning. bit-Tech. 2026;9(1):1578-1589. doi:10.32877/bt.v9i1.4151

BibTeX
@article{Yuricha2026,
  author = {Yuricha, Yuricha},
  title = {Behavioral Absenteeism Pattern Classification for Early Employee Intervention Using Machine Learning},
  journal = {bit-Tech},
  year = {2026},
  volume = {9},
  number = {1},
  pages = {1578--1589},
  doi = {10.32877/bt.v9i1.4151}
}
Kostenlos starten →