Exploring the World Wide Web with Self-Organizing Map
Christopher C. Yang, Hsinchun Chen, Kay Hong · 2002
Information overload is a critical problem in World Wide Web. Kohonen's self-organizing map (SOM) has been proven to be a promising browsing tool for the Web. The SOM algorithm automatically categorizes a large Internet information space into manageable sub-spaces. However, as the amount of information increases, it is expected to increase the size of the map accordingly in order to accommodate the important concepts in the information space. It results in increasing of visual load of the SOM. Large pool of information is packed closely together on a limited size of displaying window, where local details are difficult to be clearly seen. In this paper, we investigate the fisheye views and fractal views to support the visualization of SOM in order to explore the World Wide Web.