Student clustering and learning atmosphere analysis based on trajectory data

Abstract

With the popularity of mobile Internet and the rapid development of data mining, we can collect users' behavior trajectory and find regularity from a large amount of data. In this paper, we use the WiCloud system to obtain students' trajectory information, and propose a spatial-temporal clustering method based on K-means algorithm, which clusters students into three types: study-oriented, enclosed type and active type. A method for analyzing students' learning atmosphere based on trajectory data is also proposed, with which we get different trends of campus students' learning atmosphere in different time periods among different types of users.