Being accurate is not enough: How accuracy metrics have hurt recommender systems

Sean M. McNee, John Riedl, Joseph A. Konstan

Research output: Chapter in Book/Report/Conference proceedingConference contribution

613 Scopus citations

Abstract

Recommender systems have shown great potential to help users find interesting and relevant items from within a large information space. Most research up to this point has focused on improving the accuracy of recommender systems. We believe that not only has this narrow focus been misguided, but has even been detrimental to the field. The recommendations that are most accurate according to the standard metrics are sometimes not the recommendations that are most useful to users. In this paper, we propose informal arguments that the recommender community should move beyond the conventional accuracy metrics and their associated experimental methodologies. We propose new user-centric directions for evaluating recommender systems.

Original languageEnglish (US)
Title of host publicationCHI'06 Extended Abstracts on Human Factors in Computing Systems, CHI EA'06
Pages1097-1101
Number of pages5
DOIs
StatePublished - 2006
EventConference on Human Factors in Computing Systems, CHI EA 2006 - Montreal, QC, Canada
Duration: Apr 22 2006Apr 27 2006

Publication series

NameConference on Human Factors in Computing Systems - Proceedings

Other

OtherConference on Human Factors in Computing Systems, CHI EA 2006
Country/TerritoryCanada
CityMontreal, QC
Period4/22/064/27/06

Keywords

  • Collaborative filtering
  • Metrics
  • Personalization
  • Recommender systems

Fingerprint

Dive into the research topics of 'Being accurate is not enough: How accuracy metrics have hurt recommender systems'. Together they form a unique fingerprint.

Cite this