Research On The Fuzzy Clustering Analysis Of Data About College Students’ Mental Health

It is surveyed that young college students are regarded as one of the greatest occurrence crowds among various types of mental disorders in China. It’s common that mental or psychological diseases lead to suspension of schooling, drop-out, autonomy, suicide, or even crime. College students represents the new mental outlook of the country as a whole, whose mental health has a great relation with the quality of the national construction personnel. In a word, it is very necessary and urgent to strengthen and attach more importance to the psychological health education of college students.The application of cluster analysis in the college students’mental and psychological health data shows that the psychological characteristics, the relationship between the various factors that affect the mental health, and a reference for the development of college psychological health education. That FCM based on information entropy weighted attribute (Reformed FCM algorithm)is proposed on the basis of summarizing the advantages and disadvantages of Fuzzy C-mean clustering algorithm, which is mainly used to overcome it solver-dependence, local convergence, and not high requirement on initial clustering. That the clustering results, with the application of FCM based on information entropy weighted attribute in college mental health data analysis, is more satisfactory than that of FCM. Meanwhile, the results in the school psychological self testing table SCL-9 and UPI personality factor are consistent with that of the reformed FCM. The reformed FCM reduces the effect of initialization on the clustering algorithm and solves the problem of local convergence and reduces the number of iterations so as to improve the efficiency of clustering. Meanwhile, the class formation hidden in the mental health data is found, which fully reflects the influence of mental health factors on the clustering result and proves the practicability of the algorithm.The innovations and achievements of this thesis are as follows:(1) Based on the fuzzy C-mean clustering algorithm, the information entropy theory is introduced, and the class merging algorithm and the attribute weighting method are introduced to improve the algorithm. Firstly, use information entropy to initialize cluster centers in order to determine the number of cluster centers, and then with the idea of merger, the large clusters or arbitrary shaped clusters are divided into clusters, and then the clusters are merged according to the class merging, so as to verify or adjust the position of the initial cluster center and get the final cluster center. Finally, the concept of attribute weighting is introduced to avoid the problem of local minimum in the iteration and solve the problem of local convergence so as to achieve the global minimum.The improved algorithm greatly improves the operation efficiency of the algorithm and reduces the error caused by the initial clustering algorithm.(2) With the application of FCM based on information entropy weighted attribute in college mental health data analysis, the class formation hidden in the mental health data is discovered, which fully reflects the influence of mental health factors on the clustering result and proves the practicability of the algorithm and provides reference for the prevention and intervention of mental disorders in Colleges and universities.