Clustering and Mapping Web Sites. For Displaying Implicit Associations and Visualizing Networks

This paper describes the implementation of an association analysis method to process Web site data. The association analysis makes use of Web site links in order to produce a representation of sites association structure. A description of the association network is the goal to reach. The data refers to 791 university sites from the 15 countries belonging to the European Union. The purpose is to perform the analysis of implicit associations discovered from this sample. In order to realize it, we propose to translate the coword analysis and applying it as co-site analysis. Given a set of Web sites, two sites are considered to be associated when they are both present as outgoing links in other sites. We expose the association coefficient, the clustering and mapping techniques that are used. Their results are also exposed. In this context, the problem of information society indicators will be discussed. The subject of this paper was presented as slide presentation titled: "Clustering and Mapping European University Web Sites Sample for Displaying Associations and Visualizing Networks" and cosigned by Xavier Polanco, Moses A. Boudourides, Dominique Besagni and Ivana Roche. ETK-NTTS Conference, Crete, June 18-22, 2001. ETK = Exchange of Technology and Know-How; NTTS = New Techniques and Technologies for Statistics. International Conference supported by EUROSTAT and Joint Research Centre: Institute for Systems, Informatics and Safety of the European Commission. See: