Automatic construction of Web directory using hyperlink and anchor text

This paper proposes a technique for automatically constructing Web directories from several sites. To construct the hierarchical structure of the directories, the technique finds Web pages with a super-sub relation, which are connected by hyperlinks, and replaces the relation with a super-sub hierarchical relation between directories. The technique constructs hierarchical directories by iterating the integration of directories. As a result of an experiment using five Web sites, it was possible to construct hierarchical directories containing Web pages from several sites.