recommender systems - chi 2003: new horizons · understand recommender systems and their...
TRANSCRIPT
Recommender Systems: Interfaces and Technology
Copyright 2003 John Riedl and Joseph A. Konstan
April 7, 2003
CHI 2003 1
CHI 2003 1
Recommender Systems: Interfaces and Technology
Joseph A. KonstanJohn Riedl
University of Minnesota{konstan,riedl}@cs.umn.edu
http://www.cs.umn.edu/Research/GroupLens
CHI 2003 2
The Problem: Overload
Recommender Systems: Interfaces and Technology
Copyright 2003 John Riedl and Joseph A. Konstan
April 7, 2003
CHI 2003 2
CHI 2003 3
Recommenders
Tools to help identify worthwhile stuffFiltering interfaces
E-mail filters, clipping servicesRecommendation interfaces
Suggestion lists, “top-n,” offers and promotions
Prediction interfacesEvaluate candidates, predicted ratings
CHI 2003 4
Goals
When you leave, you should …Understand recommender systems and their applicationKnow enough about recommender systems technology to evaluate application ideasBe able to design and critique recommender application designsSee where recommender systems have been, and where they are going
Recommender Systems: Interfaces and Technology
Copyright 2003 John Riedl and Joseph A. Konstan
April 7, 2003
CHI 2003 3
CHI 2003 5
OutlineIntroductionRecommender Systems Application SpaceMovieLens Case StudyRecommender AlgorithmsEight Principles and Case StudiesDesigning Recommender ApplicationsPrivacy IssuesCommercial Tool Survey
CHI 2003 6
Who are We?
John RiedlCollaborative and distributed systems
Joe KonstanHuman-computer interaction
GroupLens ResearchNet PerceptionsWord of Mouse
Recommender Systems: Interfaces and Technology
Copyright 2003 John Riedl and Joseph A. Konstan
April 7, 2003
CHI 2003 4
CHI 2003 7
Collaborative Filtering
?
TargetTargetCustomerCustomer
Weighted Sum
3
CHI 2003 8
Using MVM
Recommender Systems: Interfaces and Technology
Copyright 2003 John Riedl and Joseph A. Konstan
April 7, 2003
CHI 2003 5
CHI 2003 9
Launch 1
CHI 2003 10
From The Laboratory
Mitchell’s Calendar ApprenticeLearns rules of how to scheduleTakes over as confidence and user permit
Effective InterrogationValue-of-information analysisWhich items to ask about?
High popularity? High Entropy?Value to individual? Value to Community?System-Driven? User-Driven?
Recommender Systems: Interfaces and Technology
Copyright 2003 John Riedl and Joseph A. Konstan
April 7, 2003
CHI 2003 6
CHI 2003 11
Principles LearnedBe a Customer Agent
Listen, Learn, and UseAnticipate PitfallsBring “Inside” Opportunities and InfoMake the Match
Box Products, Not PeopleIndividuals, not DemographicsEvolving PersonalizationReal-Time Updates
CHI 2003 12
Privacy Issues
Same as E-commerce, plusExtra sensitivity of profile data
E.g., Tacit’s dual profiles
Honesty/openness vs. edited content
Recommender Systems: Interfaces and Technology
Copyright 2003 John Riedl and Joseph A. Konstan
April 7, 2003
CHI 2003 7
CHI 2003 13
Entrée
CHI 2003 14
Social Nav Example: Groceries
Recommender Systems: Interfaces and Technology
Copyright 2003 John Riedl and Joseph A. Konstan
April 7, 2003
CHI 2003 8
CHI 2003 15
GroupLens
Filterbots Apart
CHI 2003 16
Slashdot article about Geocities
Recommender Systems: Interfaces and Technology
Copyright 2003 John Riedl and Joseph A. Konstan
April 7, 2003
CHI 2003 9
CHI 2003 17
Matchmaker: Seeker Features
CHI 2003 18
Principles Learned
Use Communities to Create ContentEditorial process is value addedFree is better than paying for it
customers trust what they produceHelp customers find interesting information from other customers
Turn Communities Into ContentHelp customers find interesting other peopleEncourage interactionYour customers may be the most interesting thing about you
Recommender Systems: Interfaces and Technology
Copyright 2003 John Riedl and Joseph A. Konstan
April 7, 2003
CHI 2003 10
CHI 2003 19
Recommendations Unplugged
What movieShould I
see?
What good movies
are close by?
What DVDShould I
buy?
Tell me what I should
See!
Wireless PDA
Voice
WML
Experimental questions•How do users interact?•What usage patterns?•What happens as users gain experience?
•How do different modalities compare?
•How does usage compare with web?
Avant Go
CHI 2003 20
Discussion