Design And Implementation Of Psychological Knowledge Service System Based On Knowledge Graph

In the field of mental health,the number of people with different degrees of psychological problems is increasing year by year,and the mental health conditions in China are difficult to meet the needs of patients.Moreover,the traditional prediction method of psychological problems relies on psychological questionnaire survey,which has the defects of lagging behind and low recovery rate and effective rate.Therefore,how to use artificial intelligence and other computer technology to effectively organize the knowledge in the field of psychology,form a complete psychological problem prediction scheme,help patients find psychological problems early,and help schools and companies learn about the psychological problems of students and employees,is an urgent problem to be solved.The main research work of this thesis includes the following parts:1.This thesis proposes construction method of domain knowledge graph of psychology.The traditional domain knowledge mapping method is usually applied in the field of strong knowledge concentration,but it is ineffective in the field of psychology with high complexity,strong relevance.Based on the characteristics of psychology knowledge,this thesis puts forward a "four step" construction method of psychology knowledge graph.The graph is constructed through the mode layer construction,knowledge collection and knowledge fusion,and the knowledge storage is realized through the neo4j database,which provides data basis for the application of psychology knowledge graph.2.This thesis proposes the prediction scheme of psychological problems based on the psychological knowledge graph.Based on users’social media data and psychological knowledge graph data,this scheme predicts users’ psychological problem through three main steps:word vector generation,word vector retrieval and correlation calculation.Based on the structural advantages of knowledge graph,the scheme adopts the truncated random walk strategy of deepwalk to strengthen the initial corpus,and proposes the Target-aware attention mechanism to optimize the weight allocation of word vector.Experiments show that the two optimization strategies can effectively improve the prediction effect of psychological problems.3.The knowledge service system based on the domain knowledge graph of psychology is designed and implemented.Based on the self built psychological domain knowledge graph,this thesis uses BS architecture to build each functional module of the system,and forms a knowledge service system with the functions of query display,knowledge base addition,question answering and prediction of psychological problems.The feasibility and integrity of the system are proved by functional test and test method based on postman tool.