Top Banner
State of RecSys Recap of the 2012 ACM Conference on Recommender Systems @alansaid October 23 2012 RecSys Meetup, Berlin
18

State of RecSys: Recap of RecSys 2012

May 08, 2015

Download

Technology

Alan Said

Recap of some of the papers and presentations at RecSys 2012. Given at the Berlin RecSys Meetup
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: State of RecSys: Recap of RecSys 2012

State of RecSysRecap of the 2012 ACM Conference on

Recommender Systems

@alansaid October 23 2012RecSys Meetup, Berlin

Page 2: State of RecSys: Recap of RecSys 2012

BackgroundWorkshops etc.● SIGIR '99, '01● ECAI '06● Recommenders Summer School @ Strands '06 "1st RecSys"

ACM Conference on Recommender System'07 - Minnesota 46%'08 - Lausanne 31%'09 - New York 43%'10 - Barcelona 19%'11 - Chicago 20%'12 - Dublin 20%

2013 - Hong KongOctober 12 - 16

Page 3: State of RecSys: Recap of RecSys 2012

Background cnt'd

"Break" point - 2006

Page 4: State of RecSys: Recap of RecSys 2012

What? Who?

8 Workshops4 Tutorials2 Keynotes4 Invited Talks24 Papers21 Posters10 Demos2 Challenges

NetflixLinkedInFoursquareStumbleUponFacebookYahoo!MicrosoftThe Echo NestAcademia

33% industry

Page 5: State of RecSys: Recap of RecSys 2012

Topics (selection)

Human Factors

Social

Preference

Context

Big Data

Evaluation

Optimization, Accuracy, Interaction

Rating effort vs. AccuracyCremonesi et al.http://dx.doi.org/10.1145/2365952.2365963

Page 6: State of RecSys: Recap of RecSys 2012

Topics

Human Factors

Social

Preference

Context

Big Data

Evaluation

Control, Trends, Social Networks/Feeds

Inspectability and Control in Social RecommendersKnijnenburg et al.http://dx.doi.org/10.1145/2365952.2365966

http://www.slideshare.net/usabart/inspectability-and-control-in-social-recommenders

Page 7: State of RecSys: Recap of RecSys 2012

Topics

Human Factors

Social

Preference

Context

Big Data

Evaluation

Implicit Feedback, Noisy ratings, Ranking

How Many Bits Per Rating?Kluver et al.http://dx.doi.org/10.1145/2365952.2365974

https://www.dropbox.com/sh/e6l0tmdwrsgyhpi/wXw4GfhcNx/presentation.pdf

Page 8: State of RecSys: Recap of RecSys 2012

Topics

Human Factors

Social

Preference

Context

Big Data

Evaluation

Cold start, semantic analysis, quality assessment

Finding a needle in a haystack of reviewsLevi et al.http://dx.doi.org/10.1145/2365952.2365977

http://www.slideshare.net/OssiMokryn/cold-start-context-aware-hotel-recommender-system

Page 9: State of RecSys: Recap of RecSys 2012

Topics

Human Factors

Social

Preference

Context

Big Data

Evaluation

Big data, frameworks, algorithms, cost, requirements

Approaches used and problems faced

@xamat

@_krisjack

@kamikaze_bhasin

@plamere

http://www.slideshare.net/KrisJack/mendeley-suggest-engineering-a-personalised-article-recommender-system

Page 10: State of RecSys: Recap of RecSys 2012

Topics

Human Factors

Social

Preference

Context

Big Data

Evaluation

Big data, frameworks, algorithms, cost, requirements

Approaches used and problems faced

@xamat

@plamere

http://www.slideshare.net/xamat/building-largescale-realworld-recommender-systems-recsys2012-tutorial

Page 11: State of RecSys: Recap of RecSys 2012

Topics

Human Factors

Social

Preference

Context

Big Data

Evaluation

Big data, frameworks, algorithms, cost, requirements

Approaches used and problems faced

@xamat

@kamikaze_bhasinhttp://www.slideshare.net/anmolbhasin/beyond-ratings-andfollowers-recsys-2012

Page 12: State of RecSys: Recap of RecSys 2012

Topics

Human Factors

Social

Preference

Context

Big Data

Evaluation

Big data, frameworks, algorithms, cost, requirements

Approaches used and problems faced

@xamat

@_krisjack

@kamikaze_bhasin

@plamere

http://www.slideshare.net/plamere/ive-got-10-million-songs-in-my-pocket-now-what

http://www.slideshare.net/KrisJack/mendeley-suggest-engineering-a-personalised-article-recommender-system

http://www.slideshare.net/xamat/building-largescale-realworld-recommender-systems-recsys2012-tutorial

http://www.slideshare.net/anmolbhasin/beyond-ratings-andfollowers-recsys-2012

Page 13: State of RecSys: Recap of RecSys 2012

Topics

Human Factors

Social

Preference

Context

Big Data

Evaluation

top-n, popularity, user-centricity, experiments

Online Controlled ExperimentsKohavi

Ranking Rating prediction

@ronnyk

Page 14: State of RecSys: Recap of RecSys 2012

Topics

Human Factors

Social

Preference

Context

Big Data

Evaluation

top-n, popularity, user-centricity, experiments

Online Controlled ExperimentsKohavi

Ranking Rating prediction

@ronnyk

http://www.exp-platform.com/Pages/2012RecSys.aspx

Page 15: State of RecSys: Recap of RecSys 2012

Popular vs. Unpopular topics

User-centricityRankingBig DataReal Systems Live EvaluationQuality of data

Rating predictionTagging(Movies)

Page 16: State of RecSys: Recap of RecSys 2012

RecSys 2013

Website:http://recsys.hosting.acm.org

Deadlines:Probably April 2013

Page 17: State of RecSys: Recap of RecSys 2012

on twitter

@neal_lathiaCambridge

@xamatNetflix

@elehack@LenskitRS

@usabartUC Irvine

@dtunkelangLinkedIn

@zenogantner@MyMediaLite

@abelloginUAM

@peterpawasPitt.

@denisparraPitt.

@ocelmaGracenote

@recsyswikiRecommender Systems Wiki

@alexk_zTelefonica R&D

@danielequerciaCambridge

@_krisjackMendeley

@plamereThe Echo Nest

#recsys @recsysdeGerman RecSys stuff

Page 18: State of RecSys: Recap of RecSys 2012

Thanks!Questions?