<![CDATA[The quality of education is highly dependent on the institution's ability to manage and analyze student data effectively. This study aims to apply the C4.5 algorithm in predicting student graduation at SDN Leuwiranji 05 based on historical data, such as average report card grades, number of absences (Alpha), and sick data. The methodology used refers to the CRISP-DM (Cross Industry Standard Process for Data Mining) model, starting from business understanding, data understanding, data preparation, modeling, evaluation, presentation. The described dataset consists of 108 student data with the results of 77 students graduating and 31 failing. The initial entropy calculation result is 0.8648, and the highest gain value is obtained from the "Average Grade" (0.293) and "Alpha" (0.3186) attributes, while the "Sick" attribute has the lowest gain value (0.0378). This indicates that academic achievement and student attendance levels are the main factors that influence graduation. This study also produces a web-based prediction system that has been tested using the Black Box method and shows results in accordance with expectations. Thus, the implementation of the C4.5 algorithm has proven effective in supporting data- based decision-making at the elementary school level and opens up opportunities for further development in the field of technology-based elementary education.]]>
Copyrights © 2025
