An XGBoost-Based Learning Analytics Framework for Predicting College Students' Information Literacy Performance

Abstract

Information literacy is an essential competency for students in modern higher education, enabling effective access, evaluation, and application of digital information. This study proposes a machine learning-based framework to predict students' information literacy performance using learning behaviour data. The collected dataset is preprocessed through data cleaning, normalization, and feature selection to improve model performance. Several classification algorithms, including Decision Tree, K-Nearest Neighbour (KNN), Naïve Bayes, Neural Network, Random Forest, and XGBoost, are implemented and evaluated using performance metrics such as accuracy, precision, recall, and F1-score. Experimental results demonstrate that the XGBoost model outperforms the other classifiers by providing higher prediction accuracy and better generalization. The proposed framework supports early identification of students requiring academic intervention and assists educators in developing personalized learning strategies. This approach contributes to intelligent learning analytics and promotes datadriven educational decision-making, ultimately enhancing students' information literacy and overall academic performance.