Semantic analysis and classification method for customer enquiries in telecommunication services

摘要

A variety of services have recently been provided that depend on highly developed networks and personal equipment. With these advances, connecting this equipment has become increasingly more complicated. Customer enquiries about problems such as no-connection to an Internet/phone service will increase, and telecom operators will require the ability to understand such situations and act as quickly as possible. Therefore, it is important to analyze failure trends and establish, in advance, effective coping processes for complex problems conveyed in customer enquiries. We present a method for analyzing and classifying customer enquiries efficiently and precisely from the structural type of description in ontologies. Moreover, our method can reconstruct semantic content efficiently by extracting related terms through analysis and classification. This method is based on a dependency parsing and co-occurrence technique to enable classification of a large amount of unstructured data into patterns because customer enquiries are generally stored as unstructured textual data. The validity of the method is evaluated and the method to determine threshold values is developed by using a large amount of customer enquiries in the real operational field.