Web document classification for Mass-Customized online service in e-CRM based on Fuzzy logic

초록

Internet technology enables companies to capture new customers, track their performances, online behavior, and customize communications, products, services, and price. The problem of customer analysis, customer interactions, and the optimization of performance metrics in electronic customer relationship management (e-CRM) can be better analyzed by either data-mining (DM), optimization methods, or combined the approaches. Most existing methods are based on a model that assumes a fixed-size of keywords or key terms with predefined a set of categories. This assumption is not realistic in large and diverse document collections such as World Wide Web. We propose a new approach to obtain the category keywords set with un-predefined number of categories on the basis of the training sets of Web documents using Fuzzy-Graph neural network and also Fuzzy association to classify test documents into a set of obtained categories. Finally with evolutionary rule we achieve the new sets of keywords and training documents to update the category keywords set and document classification.