Study on category classification of conversation document in psychological counseling with machine learning

Abstract

The beginner counselors have difficulty doing to turns interests for the cognitive characteristic and the internal problems by the client, and are using frequency closed-ended question to confirm the interpretation created in ones mind for the client. Therefore, there is the opportunity for education and training which called the supervision to improve the counseling skill of beginner counselor by expert counselors. However, these documents of the verbatim record in the counseling used in the supervision are large-scale and complex, the expert counselors are very difficult to extract the characteristics and situation of the conversation. As appropriate method to visualize each reaction of the client for each question by beginner counselor, we have developed a system for visualizing the flow of conversation in counseling. However, the expert counselor as the system user requires to correct the initial classification result manually, and the work burden is large, because the accuracy of the category classification of conversation document is very lowin the current system. To improve this problem, we have implemented on the category classification method for text data of conversation document with SVM (Support Vector Machine) as machine learning technique. In addition, we have compared and evaluated with the result of the initial classification in the current system. As these results, we have shown that the accuracy rate of the classification method with SVM become higher than the result in the current system.