Hideaki Takeda's Publication

• R. Ichise, H. Takeda and S. Honiden: Automated Alignment of Multiple Internet Directories, in Poster Proceedings, The Tenth International World Wide Web Conference, pp. 194–195, HonKong (2001). (PDF)

  Directory services are tools for making useful information more accessible, but individual internet directories in directory services are of limited use in finding user-relevant web pages. In this paper, we propose a method for aligning URL information from one internet directory to another. This method can discover an appropriate position in a directory for a web page which is not shown in that directory but shown in another directory. By using this method, one can extend a favorite directory with information on other directory services. Discovery of alignment is based on categorization similarity in the concept hierarchies of the two internet directories. We adopted the K statistic method to measure the similarity syntactically. We conducted an experiment using real- world internet directories. The results of this experiment show that the proposed method is promising, i.e., it can classify unknown web pages into appropriate categories within an internet directory.