MovieLens

From Wikipedia, the free encyclopedia
Jump to: navigation, search

MovieLens is a recommender system and virtual community website that recommends films for its users to watch, based on their film preferences and using collaborative filtering. The website is a project of GroupLens Research, a research lab in the Department of Computer Science and Engineering at the University of Minnesota which created MovieLens in 1997.[1] The website's ultimate goal is to gather research data on personalized recommendations systems.[2]

History[edit]

GroupLens, the creator of MovieLens, also formed a commercial venture called Net Perceptions in May 1996, which had clients including E! Online and Amazon.com. E! Online used Net Perceptions' services to create the recommendations system for Moviefinder.com,[2] while Amazon.com used the company's technology to form its early recommendations engine for consumer purchases.[3]

Recommendations[edit]

The website bases its recommendations on what the user provides to the website, including films the user likes, and what films other users with similar tastes prefer.[1] When a user joins the website, they are given several randomly chosen movies and told to rate them from one to five stars, five being the best. The system then compares the user's ratings to those of other users with similar tastes, and then accordingly recommends films that the user has not yet seen. When a user looks for a specific film on the website, the system returns a prediction of what it believes the user will rate the film after watching it.[4] The website suggests that users rate as many films that they have seen as possible so that the recommendations given will be more accurate, since the system would have a better sample of the user's film tastes.[2]

Features[edit]

In addition to movie recommendations, MovieLens also provides information on individual films, such as lists of actors and directors, which are used to increase the film recommendations system's accuracy.[2]

Reception[edit]

By September 1997, the website had over 50,000 users.[2] When the Akron Beacon Journal's Paula Schleis tried out the website, she was surprised at how accurate the website was in terms of recommending new films for her to watch based on her film tastes.[4]

References[edit]

  1. ^ a b Schofield, Jack (2003-05-22). "Land of Gnod". The Guardian (London). 
  2. ^ a b c d e Ojeda-Zapata, Julio (1997-09-15). "New Site Personalizes Movie Reviews". St. Paul Pioneer Press. p. 3E. 
  3. ^ Booth, Michael (2005-01-30). "How do computers know so much about us?". The Denver Post. p. F01. 
  4. ^ a b Schleis, Paula (2000-11-13). "Site Lets Everybody be a Critic". Akron Beacon Journal. p. D2. 

External links[edit]