Manuscript received August 13, 2024; revised September 4, 2024; accepted September 18, 2024; published November 12, 2024
Abstract—Collaborative and competitive learning is essential in educational development, enhancing students’ social and emotional competencies critical for academic and personal success. This study explores the integration of the Gale-Shapley algorithm, initially designed for the stable marriage problem, to optimize student pairings in collaborative and competitive learning environments. The objective is to maximize the effectiveness of Social and Emotional Learning (SEL) interventions by fostering productive social interactions and essential skill development. We propose a modified version of the Gale-Shapley algorithm to handle a single list of students, utilizing compatibility scores based on inverse Euclidean distance, Jaccard similarity, and cosine similarity. Our methodology includes generating synthetic datasets to simulate various educational contexts and evaluate the algorithm’s performance. Results demonstrate the algorithm’s efficiency in forming stable and effective pairs, significantly enhancing the learning experience. This innovative approach aligns with SEL guidelines and contemporary educational requirements, offering a robust, personalized, and dynamic learning framework. Future research should focus on empirical validations in diverse educational settings to confirm the algorithm’s effectiveness and scalability.
Keywords—collaborative learning, competitive learning, Gale-Shapley algorithm, social and emotional learning, student pairing, educational optimization
Cite: Luiz Carlos Pinheiro Junior, Everton Gomede, and Leonardo de Souza Mendes, "Optimizing Social and Emotional Learning through Modified Gale-Shapley Algorithm for Collaborative and Competitive Education," International Journal of Information and Education Technology vol. 14, no. 11, pp. 1482-1492, 2024.