International Journal of Advanced Research in Computer and Communication Engineering

Abstract- To increase flexibility and pupil engagement, blended learning combines traditional classroom settings with web-based learning. Despite its growing popularity, a quantitative assessment of its efficacy in comparison to conventional classroom teaching approach is still necessary. Using offline test data collected from both traditional and blended learning setups, the present study analyzes and forecasts student performance using supervised machine learning (ML) methods. Students' final test marks from a variety of disciplines are included in the data collection process, and each record is labeled with the learning discipline mode. The classification accuracy of several machine learning models, such as Decision Tree, Random Forest, Support Vector Machine, and Logistic Regression, is compared and evaluated by using the process of training and testing. Standard measures including accuracy, precision, recall, and F1-score are used to assess and analyze every model's performance. Comparative analysis presents the best method for modeling academic achievement and machine learning's predictive capacity in educational data analyzing. An algorithmic approach for measuring and evaluating instructional modes supports data-driven educational system approach design in computational learning analytics. This study uses selected and limited but important performance data to reveal how machine learning can maximize blended learning outcomes.

Keywords: Machine Learning, Machine Learning Algorithms, ML Model performance, Blended Learning, Education Technology, Open-Source Tools for learning and Student Performance.

Downloads: | DOI: 10.17148/IJARCCE.2026.15128