Analysis And Research Of Mental Health Status Based On The Data Of Students Network Behavior

Nowadays,the topic about students 'mental health is getting more and more attention from the society.For example,the crimes and suicides caused by university students' psychological abnormalities have also caused frequent discussion in public opinion.At present,most students are inadequate in their understanding of mental diseases,and even have an attitude of neglecting or not paying attention,so that these students with psychological abnormalities cannot be found in time and effective treatment,so can colleges find it in time? These students and intervention are the top priority in student management.With the development of data mining technology,we have solved the problem of constructing data analysis models in terms of model analysis.This article builds a psychological state prediction model based on the students 'online data collected on the university campus,combined with the psychological assessment scale indicators,used to capture the students' psychological state information,so as to help psychological counselors to make timely psychological intervention.This paper analyzes the MBTI?PHQ-9 and GAD-7 assessment scale data,expounds the psychological meaning of each assessment dimension,and measures the correlation between the dimensions through the Pearson correlation coefficient,and finds that depression,anxiety,and internal and external personality tendencies have a strong correlation with each other.Considering that depression and anxiety belong to the category of psychological abnormalities,personality internal and external tendencies It is also the psychological dimension that is most easily used to divide the population and can be used to assist in the analysis of abnormal mental states.Therefore,the psychological state prediction model in this paper uses depression,anxiety,and internal and external tendencies as label data.Then through the use of logistic regression analysis,information entropy and GA algorithm(genetic algorithm)to construct the feature dimension of the network behavior data,and finally combined with the label data of mental state to obtain a sample data set for model experiments,and designed the inclusion Model experiment on the classification of internal and external personality tendency,anxiety and depression.In the experimental process,in order to reflect the difference in effect between different models,multiple types of mathematical models were constructed for horizontal comparison;at the same time,in order to reflect the parameter impact of the same type of model,a grid search algorithm was used to construct different parameter combinations To compare the differences vertically,and finally through experiments: In the 1303 samples,the average accuracy of the multi-group experiments of the internal and external personality tendency classification models can reach 0.75;in the 1433 samples,the average accuracy of the multi-group experiments of the anxiety and depression binary classification models can reach more than 0.80.The above experiments show that the method of predicting the psychological state of students through the network behavior data is feasible.It can be used to build a mathematical classification model to grasp the students' psychological state in real time,to warn students with abnormal psychological state,and to help school counselors Psychological intervention and prevention.