Towards a flexible interactive Web usage mining system

With the rapid development of Internet technology, the World Wide Web is becoming one of the most important, media for disseminating, collecting and sharing, information. The enormous amount of web usage data, which has been collected and accumulated over years, has provided a very exciting arena for data mining.;Web usage mining generally refers to applying data mining techniques to web usage data and can provide in-depth analyses of users' navigation behaviours. There have been many sophisticated web usage mining systems which are dedicated to analyzing web usage data in the e-commerce field. However, their utilities are greatly restricted by the rigid pre-defined web usage mining process. In this thesis, we present a prototype of an open-structure interactive web usage mining system. With this system, not only useful models can be easily integrated by developers to expand the functionalities, but also various data mining experiments can be designed by decision makers interactively to discover patterns of interest.