Mining fuzzy brsowing patterns from web log

Web applications have made a significant impact on computer system.Among them mining useful information from web logs is is important and challenging research topics. In this paper, the time durations of the paged browsed. are computed from web logs. Since the time durations are numeric, fuzzy concepts are used here to process them and to form linguistic terms. Besides different pages in web sites may have different interest to users, and some factors (i.e. the design of web pages) may affect users´ browsing interest. So interest measures of web pages are evaluated by managers according to the judgment factors, and comprehensive fuzzy evaluation methods are used to evaluate the interest measure of each page. Then a new fuzzy web-mining algorithm is proposed, which can process to mine fuzzy browsing interest patterns with linguistic variables from web logs effectively. An example is given to clearly illustrate the proposed approach.