Investigating the factors influencing the performance of social conditioning in the network environment is the core issue for improving academic performance. Through the search of existing literature, the paper analyzes the main factors that influence social conditioning learning in current research, and through the questionnaire survey and in-depth processing of the raw data, the advanced behavioral indicators related to learning are obtained and analyzed by Spearman correlation coefficient and fuzzy modeling in machine learning. The results showed that the twelve dimensions of motivation regulation, trust building, efficacy management, cognitive strategy, time management, goal setting, task strategy, peer support, team assessment, help seeking, environment construction, and team supervision were significantly related to group performance, with team supervision having a significant negative relationship with group performance. In addition, trust building, team supervision and environment construction were the main factors for online social learning, effectiveness management, task strategy, peer support and help-seeking were the secondary factors, while motivation regulation, cognitive strategies, goal setting and team assessment had little impact on the final performance. The findings have some implications for the optimization of social conditioning learning support services and the improvement of social conditioning learning performance.
