Evaluating service quality in insurance customer complaint handling throught text categorization

The paper presents the findings of an industry-based study in the utility of text categorization. The purpose of the study is to explore new approach to evaluate service quality of customer complaint handling. The industrial research setting is a large China insurance company. The text categorization methodologies are used in this research including nature language processing and machine learning. In addition, a multiple filtration algorithm using synonyms (near-synonym) dictionary, term frequency and Chi-square statistic is adopted in feature selection. The experimental result shows that text categorization can be used to evaluate the service quality by text data from customer complaint handling. The method can solve automatic evaluation problems in customer complaint handling management.