<![CDATA[Effective collaboration in mathematics learning is essential for developing students' critical thinking and problem-solving skills; however, identifying patterns that lead to successful group collaboration remains challenging. This study aims explicitly to identify and classify the patterns of student group assignment completion in the Logic and Sets course using the Naïve Bayes algorithm. Survey data from 65 mathematics education students were analyzed using a quantitative approach and machine learning techniques. Attributes such as group size, task completion time, participation, contribution strategies, and communication effectiveness were collected via structured questionnaires. Data analysis involved preprocessing, model training using Naïve Bayes, and validation through accuracy and posterior probability analysis. Results indicated that the Naïve Bayes model accurately distinguished groups with very good (A) and fairly good (B) performance, achieving 84.62% accuracy. Groups achieving an A grade typically featured balanced participation and open communication strategies, whereas groups graded B exhibited uneven participation and passive members. This research significantly contributes by demonstrating how data-driven predictive analytics can support instructors in monitoring and enhancing collaborative learning processes in mathematics courses. Future research could further refine predictive accuracy by incorporating additional factors such as leadership style and collaborative technologies, potentially integrating the model into learning management systems for real-time evaluation and intervention.]]>
Copyrights © 2025
