A hybrid machine learning framework for personalized electronic service recommendations

Al Hussaini, Mohammed Abdulla Salim and AL Hinai, Turki Abdullah Masood and Al Husaini, Yousuf Nasser and Habaebi, Mohamed Hadi (2026) A hybrid machine learning framework for personalized electronic service recommendations. In: 10th International Conference on Mechatronics Engineering (ICOM’26), 08-09 September 2026, IIUM.

PDF - Published Version Restricted to Repository staff only Download (1MB) | Request a copy

Abstract

The rapid growth of digital services in higher education has increased the demand for intelligent decisionsupport systems capable of providing personalized services to students. This study presents an ensemble-based machine learning approach for predicting students' academic status and generating personalized electronic service recommendations using demographic, socioeconomic, and academic performance data. The experiments were conducted on a publicly available dataset containing 4,424 undergraduate students described by 36 features and three target classes: Graduate, Enrolled, and Dropout. A comprehensive preprocessing pipeline, including data cleaning, label encoding, feature scaling, and feature selection, was applied. The most influential features were identified using Mutual Information, Random Forest Importance, and SHAP, where Curricular Units 2nd Semester (Approved), Curricular Units 2nd Semester (Grade), and Curricular Units 1st Semester (Approved) were ranked as the most significant predictors. Six machine learning models, namely Random Forest, XGBoost, LightGBM, CatBoost, Support Vector Machine (SVM), and Multi-Layer Perceptron (MLP), were evaluated. Among the individual models, XGBoost achieved the best performance with an accuracy of 77.1% and an F1-score of 76.3%. In addition, hybrid ensemble models based on soft voting and stacking strategies achieved the highest precision of approximately 78% while maintaining an overall accuracy and F1-score close to 77%. Finally, a recommendation engine translated prediction outcomes into personalized electronic services, demonstrating the practical applicability of the proposed approach. The experimental results indicate that the proposed framework can effectively support academic decision-making and enhance personalized student services in higher education.

Item Type:	Proceeding Paper (Plenary Papers)
Uncontrolled Keywords:	Hybrid Machine Learning, Electronic Service Recommendation, Student Performance Prediction, Ensemble Learning, Feature Selection.
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7800 Electronics. Computer engineering. Computer hardware. Photoelectronic devices > TK7885 Computer engineering
Kulliyyahs/Centres/Divisions/Institutes (Can select more than one option. Press CONTROL button):	Kulliyyah of Engineering > Department of Electrical and Computer Engineering Kulliyyah of Engineering
Depositing User:	Dr. Mohamed Hadi Habaebi
Date Deposited:	05 Nov 2026 16:22
Last Update:	05 Nov 2026 16:22
Queue Number:	2026-10-Q6056
URI:	http://irep.iium.edu.my/id/eprint/132507
Indexed In:	Google Scholar

Actions (login required)

View Item