<![CDATA[This study compares the effectiveness of the Random Forest and Decision Tree algorithms in predicting students' academic performance based on learning activities. The data used included reading scores, writing scores, math scores, and demographic variables such as gender, race/ethnicity, parental level of education, lunch, and test preparation course. The research was carried out through the stages of data cleaning, training and test data sharing, model training, and evaluation using confusion matrix and accuracy, precision, recall, and F1-score metrics. The results show that Random Forest performs best with 97% accuracy, surpassing Decision Tree which has 94% accuracy. The feature importance analysis revealed that cognitive ability—especially in the reading score, writing score, and math score features—had the greatest influence on prediction results. These findings confirm that the Random Forest model is more reliable and effective as a prediction tool in the academic decision support system to detect the potential for decline in student achievement early.]]>
Copyrights © 2025
