Research on the Intelligent Construction of College Management System under the Diversification of College Students' Education and Management

Abstract

In this paper, the trajectory data of college students are collected for preprocessing and then populated using the mean trajectory filling algorithm. The student campus trajectory prediction model is constructed to study the activity characteristics of college students in different functional areas. Three features, namely, travel distance, travel mode selection ratio, and pedestrian flow on campus main roads, are extracted to analyze the regularity of trajectory data. The variation of the number of devices in different functional areas is used as the trajectory data of college students, and the prediction model is used to visualize the clustering. 80% of the students' average daily travel distance is less than 12km, and 20% of the students' travel distance is between 12km and 22km. more than 92% of the students choose walking as their travel mode. The foot traffic on each main road has a stronger correlation with the location of their hostel area and the location of their activities. The number of students' equipment has a clear pattern of change in different functional areas. The activities in the teaching area are much higher than the other functional areas in the time period of 6:30-12:30. The frequency distribution of different functional areas in clusters can be used as a basis for predicting the behavioral trajectories of students with different characteristics and realizing intelligent student management in colleges and universities.