Using clustering and visualization for refining the results of a WWW image search engine
Sougata Mukherjea, Kyoji Hirata, Yoshinori Hara · 1998
Search engines are useful because they allow the user to find information of interest from the World-Wide Web.We have developed a WWW search engine called AMORE that allows users to find images on the Web.However, like textual search engines, AMORE sometimes retrieves a lot of results.Showing the retrieved images as pages of scrolled list, like other search engines, is not very userfriendly.Therefore, we have developed a query result visualization environment that allows the user to organize the search results in a Java applet using various techniques; for example, the user can use text and image clustering.This paper presents the architecture of the system and discusses the text and image clustering algorithms.Examples are used to show the usefulness of the techniques.