Data Security Mining Method for Social Media Users’ Mental Health Status Test Based on Machine Learning Algorithm

Users who frequently use social media can easily lead to changes in their mental health status. In order to accurately predict users’ mental health status, a data mining method for testing social media users’ mental health status is designed based on machine learning algorithms. Perform clustering processing on the test data, calculate the centroid of each cluster and the sum of squared errors of the data set, and obtain the clustering result of the test data; extract the characteristics of social media users’ mental health status, and calculate the abnormal score of the object's mental health status, Use this to distinguish the user's psychological state; build a data mining model based on machine learning algorithms, build a one-dimensional convolutional neural network framework, record the activation functions of the convolutional layer and the pooling layer, and obtain the output value of the fully connected layer; design social media user psychology Health state test data analysis algorithm to obtain a safe mining method for social media users’ mental health state test data. Obtain the best period of social media users’ mental health status observation through experiments. It can be seen from the experimental data that the correlation of the mental health state mining results obtained by this method is higher than 0.91, and the prediction accuracy is higher than 92%, which improves the effectiveness of mental health state prediction. The best observation period is one month to two months before the expected time.