| Summary: | 碩士 === 國立清華大學 === 工業工程與工程管理學系 === 92 === Owing to the drastic development of the Internet technology, the problems of information overload is more and more serious. Under such circumstance, the Internet users cannot efficiently and effectively retrieve the information that meets their requirements over the Internet. Recently, in the customer centric market, customer relationship management (CRM) has become a critical issue for business operation. To obtain long-term relationship with the customers, the enterprises have to capture the customer demands, to establish a personalized relationship with customer and to intelligently provide information/document based on user requirements.
In order to explore a knowledge clustering and service model for the organizations to efficiently manage the domain documents, three algorithms namely document similarity analysis (DSA), document clustering algorithm (DCA), and document publication algorithm (DPA) are proposed in this research. The document similarity analysis is developed based on the key attributes of documents (including the document keywords, providers and categories) while the document clustering methodology is proposed by application of document similarity and the K-means approach. In addition, according to the browse history of users and document clusters, a publication algorithm of documents is utilized to automatically provide the documents to the target users. A web-based prototype system of the knowledge clustering and service model will also be implemented to ensure the applicability of the model. Effectiveness of the model is also by a demonstration case. As a whole, the knowledge management model proposed in this research can assist the organizations to intelligently and efficiently manage the documents and to realize the objective of one-to-one marketing.
|
|---|
