The Recommendation system reference article from the English Wikipedia on 24-Apr-2004 (provided by Fixed Reference: snapshots of Wikipedia from wikipedia.org)

Recommendation system

Recomendation systems retrieve information on the basis of users' tastes. This process relies on taste vectors and has become popular in movie recomendation systems and for telephone sales in which a caller who might buy bath towels will cause a computer to prompt the sales representitive to ask "Whould you like a set of shower curtain rings?" Such systems can predict user wants with suprising accuracy and increase sales by as much as 25%.

In movie review applications the user assay of films and reviews are used to compile numerous other users profiles weighted by degree of similarity into an expanded taste a profile. Those films which come up consistantly proove to be reliable recomendations because the elements of the movies and matching aspects of human taste tend to apply across individuals.

The utility of such of systems remains poorly explored but is being tested by google.com [[1] and other companies to deliver user-specific information.

Taste data can create personal marketing systems where the individual controls what is marketed to themselves and is particularly useful in fuzzy areas such as musical tastes in which there is no known parameter of ranking or searching. However, taste systems can also be applied to semantic data.

See also: Collective intelligence, Consensus, E-Consensus, Personalized marketing

This is the "Recommendation system" reference article from the English Wikipedia. All text is available under the terms of the GNU Free Documentation License. See also our Disclaimer.