Answers are grounded in this paper and cite their sources.
Analysis compares Random Forest, Naïve Bayes, and Support Vector Machine for attendance-based sustainability, suggesting effective algorithm choice.
Sinau Bareng is one of the educational programs initiated by the Dinas Pendidikan Kota Surabaya, focusing on collaborative learning activities for school-aged children. The sustainability of Sinau Bareng activities relies heavily on participant attendance at each location. A key issue that arises is the absence of participants (tutors and students) in several locations, which prevents the activities from being carried out effectively. Therefore, it is necessary to predict the sustainability of activities at each location based on participant attendance data. This prediction aims to support decisionmaking in selecting and continuing appropriate learning locations. This study compares three algorithms Support Vector Machine (SVM), Random Forest, and Naïve Bayes to determine whichis more suitable for predicting sustainability at each location. The research process follows the CRISP-DM framework, which consists of six stages: business understanding, data understanding, data preparation, modeling, evaluation, and deployment. The dataset used includes attendance records categorized by tutors and students over a period of time, with each record labeled according to its sustainability status. The evaluation results indicate that the best-performing model is the Random Forest algorithm using a 4/5 K-Fold validation split, achieving an accuracy of 93.02%, precision of 86.53%, recall of 93.02%, F1-score of 89.66%, and AUC of 90.48%.. These results suggest that Random Forest is highly effective for handling attendance-based sustainability predictions in community-driven education programs.
No takes yet. Share an insight, caveat, or question.
Rabbaanii et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: