<![CDATA[This study presents EduTrack, a profiling application that uses a Random Forest classifier to identify Indonesian high-school students’ needs for additional Math tutoring. The dataset consists of students’ chapter-wise Math scores, processed with Pandas and Scikit-learn and stored via SQLAlchemy. The backend is implemented in Flask, while the frontend employs Bootstrap with Chart.js for charts and DataTables for tabular display. Dummy evaluation yields model performance around 90% accuracy, with precision 88%, recall 91%, and F1-score 89% (Table 1, Figure 2). Evaluation metric formulas (precision = TP/(TP+FP), recall = TP/(TP+FN), F1 = 2 * precision * recall / (precision + recall)) are included for clarity. EduTrack is designed not only as a predictive tool, but also as a practical decision-support system for teachers. By visualizing student performance at the chapter level, the application enables educators to identify learning gaps more intuitively and implement timely interventions. This helps shift teaching strategies from reactive to proactive, ultimately supporting personalized learning and improving academic outcomes.]]>
Copyrights © 2026
