Utilizing Clustering Analysis and Machine Learning Techniques for Providing Student Learning Recommendations

This research endeavors to scrutinize the influence of courses on students' final year project (FYP) scores and prognosticate FYP scores by applying methodologies such as clustering analysis, decision trees, logistic regression, and support vector machines. By scrutinizing the correlation between students' course grades and FYP scores, the objective is to gain deeper insights into the impact of diverse courses on FYP performance and offer tailored learning recommendations. Initially, clustering analysis is employed to categorize students into distinct groups, discerning the potential impact of specific course combinations on FYP outcomes. Subsequently, machine learning algorithms, including decision trees, logistic regression, and support vector machines, are leveraged to forecast FYP scores based on students' course grades. Through these analyses, customized learning recommendations and guidance can be provided to augment students' FYP performance, fostering their academic advancement and maturation. The findings of this research are pivotal for educational institutions and individual students, contributing to the enhancement of curriculum design and pedagogical methodologies, thereby elevating the standard of education and enriching students' learning encounters. Furthermore, drawing upon assessment data spanning over 16 years and encompassing more than 2,000 non-CS majors at MPU, we delve into the factors influencing student learning quality and proffer timely learning recommendations based on amassed student subject grades and FYP scores.