Study Of Knowledge Navigation In Customers' Complaints Of Mobile Communication Services

With the rapid development of China's economy and communication technologies, more and more Chinese people are prone to carry ceil phones, and to improve the mobile service level, the Mobile Communication Corporations in China provide more types of services now than before. As a result, more and more customers' complaining information are piled up. In such a situation, how to use the customers' complaining information to enhance the corporations' operation efficiency and improve the corporations' service quality become a big issue to be seriously considered. At present, many Mobile Communication Corporations adopt the idea of knowledge management by constructing a knowledge management platform to manage knowledge and improve the competitive ability.From the perspective of knowledge classification, this paper presents an approach by constructing the knowledge map in the domain of Customers' Services of Mobile Communication Corporations in China to navigate explicit and tacit knowledge in two ways: building Topic Map (TM) based explicit knowledge navigation model, which includes domain TM construction, semantic based topic expansion algorithm, and vector space model (VSM) based similarity calculation; building Social Network Analysis based tacit knowledge navigation model, which includes the Balanced Score Card (BSC) based expert navigation method, multi-relational expert network based expert navigating route search method and relevant expert navigation method, and SociometryPro based criterions of expert network, as well as the core expert search method.Key work of this paper can be summarized in the following three parts:(1) Domain TM construction. Extract topics from customers' complaining documents; analyze the associations between topics; and append the occurrences back to the corresponding topics. In doing so, the explicit knowledge inside the corporations can be well managed and navigated by domain TM.(2) Present a semantic based topic expansion method, which not only decreases the time complexity, but also enhances the recall result and does some reasoning work in some extent. Also similarities between queries and complaining documents are calculated based on VSM, and better precision result is obtained. In doing so, the explicit knowledge inside the corporations can be well searched. (3) Propose a Balanced Score Card (BSC) based expert search method aiming at searching experts under the consideration of differences between experts' abilities; present a multi-relation based expert navigation algorithm in order to navigate experts in some fuzzy networks and provide relevant expert search; use SociometryPro to analyze the status of expert network, find out core expert in networks, and propose some approaches to improve tacit knowledge navigation inside the corporations.The experiment is carried out with 500 pieces of customers' complaining documents. A domain knowledge navigation system is created, and roles for updating and maintaining the system are also presented. Results show that TM based explicit knowledge navigation and information retrieval and SNA based tacit knowledge navigation method has a good performance on knowledge navigation and knowledge management. Therefore, the customers' complaining information is well used to enhance the corporations' operation efficiency and improve the service quality. Finally, the core competition of the corporations will be enhanced as well.