Student behavior analysis method based on machine learning

The invention discloses a student behavior analysis method based on machine learning. The student behavior analysis method comprises the following four steps: S1, carrying out data acquisition: uploading system data of a campus management system such as a one-card management system and a library system to a database; S2, carrying out data standardization: performing normalization, normalization and normalization on the acquired data to obtain standard data; S3, carrying out clustering model training: sending the processed data into a clustering model for training to obtain clustering results of various behaviors of the students; and S4, carrying out abnormal behavior judgment: performing abnormal analysis on various behaviors of the students through an SVM algorithm. According to the student behavior analysis method based on machine learning provided by the invention, student behaviors can be subdivided through the campus data of the students, and whether the student behaviors are abnormal behaviors or not is