Construction of Psychological Crisis Assessment Model Based on Machine Learning

摘要

The assessment of psychological health is an important part of the construction of smart city. In order to establish psychological crisis discrimination criterion and analyze the contribution of each factor, this study uses machine learning algorithms based on big scale data collected during the construction of smart cities. Firstly, the psychological crisis level is discriminated by using ISOMAP algorithm and K-means algorithm. Then the logistic regression is modeled to estimate the influence coefficient of each factor. Finally, the above models were used to assess each individual's psychological states. The results show that the classification performance of machine learning algorithm is better than that of the traditional norm. Moreover, the coefficients obtained by logistic regression can be used to represent the contribution of various factors to individual's psychological state.