The document discusses a study that used a combined approach of social network analysis and educational data mining to examine the impact of communication networks, behavior networks, and the combination of these networks on students' academic performance. The study included 139 students from two universities. A questionnaire was used to collect data on the students' networks, computer experience, self-efficacy, and academic performance. The results showed that gender was highly prioritized in all three models for predicting academic performance. Communication networks had the highest confidence, accuracy, and precision in impacting predictions of academic performance, followed by the combination of networks and then behavior networks alone.