When it comes to quickly and accurately understanding behavior trends among undergraduates, teachers may find that sophisticated analytics become useful. There has been a shift in the management of student affairs from anecdotal, qualitative information to data-driven, quantitative analysis, thanks in large part to the emergence and advancement of big data techniques that enable educators to analyze the behavior patterns of learners in a timely and effective manner, particularly to find the groups of learners that need to be concentrated on in a timely fashion. Using the clustering technique of data mining, the authors of this research suggest a Behavioral Characteristics Extraction Model (BCEM) for studying student networking habits on campus. According to the findings, there are 400 pupils who make heavy use of the Internet and are spread over four distinct categories. There is an impact on these students' academic performance and outcomes. Data mining of student network behavior was conducted for this study, which may be utilized as a real-world example of data mining in student affairs management and so offers solid empirical backing for the advancement of that field.