An efficient algorithm for fuzzy Web-mining

In this paper, Web mining in a fuzzy environment is devoted. Browsing time staying on a Web page is considered and characterized as a fuzzy variable. Thus, the frequent preferred paths with fuzzy expected values can be gained. With the comparison of the fuzzy expected values, the measure of the interest of people for different Web pages is clear. In order to find more completely frequent fuzzy preferred paths, an efficient algorithm based on the frequent link and access tree(FLAAT) is designed, in which the access tree is traversed using a top-down strategy, and to avoid the loss of useful information, the frequent link is searched to find such nodes that may be neglected, and then the access tree is searched again to find other frequent preferred paths. The gained frequent preferred paths with fuzzy expected values more completely disclose the interest of users. Finally, an example is provided to clearly illustrate the proposed approach. And the results show that our algorithm achieves significant performance improvement over previous work.