The Research And Application Of Web Access Information Based On UIMA Architecture

Web Data Mining(WDM) is a very hot research topic which combines two activated research areas: Data Mining and World Wide Web. WDM can extract interesting, potential, useful, novel and hidden patterns from web documents and the users'activities on web. As a branch of WDM, Web Usage Mining(WUM) has been gaining a lot of attention because of its potential commercial benefits. The knowledge, obtained through the WUM can not only direct the web users'navigations but also assist the design of the web site.At first we introduced the reason, definition and classification of Web Data Mining, and then introduced the significance and difficulties of Web Data Mining. Web Data Mining is the combination of Web technology and data mining. Web Usage Mining usually consist of two process: data extract and data mining. We gave the process model and elaborated every steps of web usage mining thoroughly.UIMA (Unstructured Information Management Architecture) is a architecture of unstructured information, which integrate many powerful function such as words deal and information search. The data source of web usage information is huge unstructured information, so we use UIMA responsible for data extract and present its architecture principle.In data extract, the process model of WUM is brought forward and interpreted in brief. According to the format of the web log file, apart from all other phrases of WUM, the data pre-processing is emphasized, including the user's undefined request filtering algorithm, web frame page filtering algorithm and user session partition methods and their precision measures. After the path completion phase, the result format of each phase is presented.In data mining, Web Fuzzy Clustering (WFC) and its model(WFCM) are put...