Maximal cliques based method for detecting and evaluating learning communities in social networks

• Published in: Future generation computer systems 2022-01, Vol.126, p.1-14
• Main Authors: Adraoui, Meriem, Retbi, Asmaâ, Idrissi, Mohammed Khalidi, Bennani, Samir
• Format: Article
• Language: English
• Subjects: Centrality; Cliques; Collaborative assessment; Evaluation; Homophily; Learning community
• ISSN: 0167-739X; 1872-7115
• DOI: 10.1016/j.future.2021.07.034

Massive data is generated due to the intensive use of social networks by learners. Thus, various applications in different domains such as education are conducted to understand the relationships among actors. Generally, learners create learning communities according to their levels, interests, and skills. A large number of approaches have been proposed to detect these communities in social networks. In this study, we propose a new approach to detect and evaluate learning communities in order to help teachers and administrators understand learners’ needs and improve the educational process. This approach is an improved version of our previous model. Hence, the main objective of this article is to lower the execution time of our previous approach and to improve the evaluation process. Our approach mainly consists of two phases. The first phase is developed to discover the learning community using the maximal clique concept, while the second one is devoted to learning community evaluation based on the interactions among learners and their socio-economic characteristics. The performance of our model is tested on four real-world networks: Seventh graders, UC Irvine, UK Faculty, and Forum discussion using the modularity and the silhouette measures. Two types of learning communities have been identified: safe communities and at-risk communities. The experimental results showed that our approach is highly reliable and efficient for discovering and evaluating learning communities in social networks compared to other approaches, as well as it has a low temporal complexity. •A new algorithm for detecting and evaluating learning communities is proposed.•The proposed evaluation process combines both static and dynamic evaluation.•A Safely Centrality measure is introduced to compute the learners’ success degree.•The proposed method is able to identify safe and at-risk communities.•The conducted experiments have proven the efficiency of the proposed method.