Eigentaste: A Constant Time Collaborative Filtering Algorithm

  • Authors:

  • Ken Goldberg;Theresa Roeder;Dhruv Gupta;Chris Perkins

  • Affiliations:

  • IEOR and EECS Departments, University of California, Berkeley, CA, USA. goldberg@ieor.berkeley.edu;IEOR Department, University of California, Berkeley, CA, USA;IEOR Department, University of California, Berkeley, CA, USA;IEOR Department, University of California, Berkeley, CA, USA

  • Venue:
  • Information Retrieval
  • Year:
  • 2001

Quantified Score

Hi-index 0.02

Visualization

Abstract

Eigentaste is a collaborative filtering algorithm that uses iuniversal queries to elicit real-valued user ratings on a common set of items and applies principal component analysis (PCA) to the resulting dense subset of the ratings matrix. PCA facilitates dimensionality reduction for offline clustering of users and rapid computation of recommendations. For a database of in users, standard nearest-neighbor techniques require iO(in) processing time to compute recommendations, whereas Eigentaste requires iO(1) (constant) time. We compare Eigentaste to alternative algorithms using data from iJester, an online joke recommending system.Jester has collected approximately 2,500,000 ratings from 57,000 users. We use the Normalized Mean Absolute Error (NMAE) measure to compare performance of different algorithms. In the Appendix we use Uniform and Normal distribution models to derive analytic estimates of NMAE when predictions are random. On the Jester dataset, Eigentaste computes recommendations two orders of magnitude faster with no loss of accuracy. Jester is online at: http://eigentaste.berkeley.edu