Dynamic Detection and Analysis Strategy of Students’ Abnormal Behaviors Based on LOF-GMM Algorithm

This study focuses on the field of vocational education and aims to use the LOF-GMM algorithm to achieve dynamic detection of students’ abnormal behaviors. Based on the detection results, effective analysis strategies are proposed. By collecting multi-dimensional data of students, this algorithm is utilized to accurately identify the abnormal points in learning behaviors and performances. Furthermore, an in-depth analysis of the causes behind the abnormal behaviors is conducted, providing a powerful basis for personalized education, optimization of teaching content, construction of decision support systems, etc. This is intended to promote the improvement of students’ comprehensive qualities and contribute to the highquality development of vocational education.