2026-05-11T23:44:53.698Z http://ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai

ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai:3:9930 2025-01-03T14:40:07Z 3

Leveraging Metaheuristic Optimized Machine Learning Classifiers to Determine Employee Satisfaction, Chapter in AIS Algorithms for Intelligent Systems: ICMSLE 2024: International Conference on Multi-Strategy Learning Environment, Springer 2024 http://ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai/record/3/9930 https://link.springer.com/chapter/10.1007/978-981-97-1488-9_26 L. Babic L. Jovanovic A. Petrovic M. Zivkovic T. Zivkovic N. Bacanin The scholarly exploration of employee satisfaction has intensified following the identification of its correlation with work performance. Numerous intricate factors contribute significantly, and their nuanced influence varies at the individual level. While certain factors may be readily discernible, achieving equilibrium proves challenging. This study aims to meticulously probe the applicability of cutting-edge artificial intelligence (AI) algorithms in gauging the multifaceted dimensions of employee satisfaction. Given the substantial reliance of AI algorithms on judicious hyperparameter selection, a modified iteration of the recently introduced sinh cosh optimizer is proposed for fine-tuning hyperparameters within the highly regarded extreme gradient boosting (XGBoost) methodology. The resulting outcomes have undergone rigorous scrutiny and comprehensive comparative analysis alongside other contemporary optimization techniques, illuminating the distinctive efficacy of the introduced approach. The novel methodology yielded commendable results when assessed against a publicly accessible dataset on employee satisfaction, establishing its robustness in diverse settings. The optimal models are further subjected to meticulous evaluation with the best-constructed model outperforming models optimized by the original algorithm as well as those created by other evaluated optimizer algorithms. bookPart Springer, Singapore 337 352 10.1007/978-981-97-1488-9_26 AIS Algorithms for Intelligent Systems: ICMSLE 2024: Proceedings of International Conference on Multi-Strategy Learning Environment