MACHINE LEARNING FOR PREDICTING STUDENTS' ACADEMIC ACHIEVEMENT BASED ON LEARNING STYLE AND ACADEMIC RESULTS

Abstract

Abstract: The objective of this study is to explore the viability of utilizing machine learning methods in forecasting students' academic success by examining their unique approaches to learning. The data was obtained by administering a questionnaire that employed the Motivated Strategies for Learning Questionnaires (MSLQ), along with the academic records of Semester 4 students from Jabatan Teknologi Maklumat dan Komunikasi (JTMK) Kuching Sarawak. After pre-processing the data and devising relevant features, we proceeded to train and test our machine-learning model. By employing Linear Regression, Decision Tree Regressor, Random Forest Regressor, Lasso, and Ridge, we were able to accurately anticipate students' performance based on their learning styles. Our study shows that the Random Forest and the Ridge have the most accuracy in predicting students' performance with MSE = 0.11, MAE = 0.22, and RMSE 0.32 for both models. The findings demonstrate the potential of machine learning models in accurately forecasting academic achievement, thereby offering valuable insights to educators who wish to personalize their teaching methods and intervention strategies. In summary, this study underscores the capacity of machine learning techniques in education to optimize learning outcomes and ensure the academic success of students.