基于粒子群和K-均值聚类算法的学生心理分析方法研究

Study on students' psychological analysis based on particle swarm optimization and K-mean clustering algorithm

随着高校学生人数的不断增多,评估与采集学生信息剧增,传统的学生评估与管理方式已不能满足需求。针对此问题,文中在研究了现有学生心理状态评估方法的基础上提出一种基于粒子群和K-均值聚类算法的学生心理分析方法。该方法弥补了全局优化中K-均值聚类算法的不足,结合粒子群算法,对学生的综合情况进行分析研究,从而为老师与学生的交互提供了良好的平台。最后通过比较所提算法与K-均值聚类算法、人工评估及基于遗传算法的K-均值聚类算法的学生心理分析结果,从结果中可看出文中所提的基于粒子群和K-均值聚类算法对学生心理分析评估更加客观与全面,有利于教师提高工作效率,及时发现问题,且增强与学生间的沟通。

As the number of students in colleges and universities is increasing, the information of evaluating and eolleeting students is inereasing, the traditional methods of student assessment and management are unable to meet the needs. In order to solve this problem, this paper puts forward a student psyehoanalysis method based on partiele swarm optimization and K-mean elustering algorithm on the basis of studying the existing method of evaluating the students＇ psyehologieal state. This method makes up the shortage of the K-mean elustering algorithm in global optimization, and eombines the partiele swarm optimization algorithm to analyze the eomprehensive situation of the students, thus providing a good platform for the interaetion between teaehers and students. At the end of the paper, by eomparing the results of the students＇ psyehologieal analysis of the proposed algorithm, the K-mean elustering algorithm, the artifieial evaluation and the K-mean elustering algorithm based on the genetie algorithm, it can be seen from the results that the partiele swarm optimization and the K-mean elustering algorithm proposed in this paper are more objective and eomprehensive for the students＇ psyehologieal analysis, which is benefieial to the teaeher to improve their work effieieney, to find problems in time and to enhanee eommunieation with students.