Abstract

Search engines must become more customized and accurate to meet the demands of the growing Web. We propose the use of filtering techniques to enhance the ability of a search engine to deliver relevant results. We have designed and implemented a working collaborative search engine, called Foible. We demonstrate the advantages of a hybrid search/recommender/collaborative system by performing an extensive user study. Our conclusions show that users prefer intelligent search engines incorporating our new techniques over traditional means of performing searches.

Downloads:

• Report: Compressed postscript (474 Kbytes)
• Slides: Powerpoint (134 Kbytes)

See also:

• REFER research group Web page