These guys are ripe for acquisition. Clusterizer, who operates the technology behind the iBoogie clustering meta-search engine, has a terrific description of the clustering process and the technology behind it. It’s an excellent way to help understand exactly how term dicsovery for topics is done in real-time during a web search.
Clustering is a major part of web retrieval technology, as it can help disambiguate (make clear) the relevancy of a particular topic to a specific query. Although the major search engines (Google, Yahoo!, MSN, Ask) do not display clustered results, they do use similiar technologies to help uncover the meanings of user entered searches.
Some good discussion about clustering technology is in the Search Technology Forum at SEOChat – here and here.
