I was introduced to a beta search engine, Clusty that offers a technology to ‘cluster’ search results. If you do a search for ‘Dune’ you can then determine which search cluster best describes what you were seeking. Here are cluster results for ‘Dune’ that allows you to filter the results.

⇨Movie (35)
⇨Dune Buggy (33)
⇨Game (19)
⇨Sand Dune (18)
⇨Science, Fiction (19)
⇨Kevin J. Anderson (10)
⇨Frank Herbert’s Dune (7)
⇨Dune Chronicles (6)
⇨Beach (7)
⇨Mud (3)

Why Clustering?

The problem of not being able to find what users are looking for has now transformed into one of finding too much information. Today’s search engines burden the user with the task of interpreting hundreds or thousands of documents to glean the information they need.

Categorization of all this data has been the obvious solution to enable users to better deal with this “information overload”. Traditional approaches to categorization involving taxonomies, however, are too expensive, time-consuming and complex for most organizations.