Predicting University Student Academic Achievement Using the C4.5 Decision Tree Based on Socioeconomic Factors

Abstract

Student academic achievement is an important indicator in assessing the success of higher education; however, it is often influenced by external factors such as socio-economic background, including parents’ occupation and income. Differences in these conditions can lead to disparities in academic performance if not properly identified. This study aims to identify student academic achievement based on socio-economic background using a data mining approach with the C4.5 algorithm. The research process includes data collection through observation, interviews, questionnaires, and documentation, followed by data preprocessing, attribute selection, and classification model development using the Orange application on 510 student records with variables such as gender, Grade Point Average (GPA), parents’ occupation, and parents’ income. The results indicate that the C4.5 algorithm is capable of generating a decision tree model that identifies patterns between socio-economic factors and academic achievement, with GPA as the most dominant attribute. This model can be utilized as a basis for academic policy-making, such as providing educational assistance and academic support programs, and may also serve as a reference for future research to achieve more accurate and comprehensive analyses.