The RecSys 2007 conference program has both invited and peer-reviewed components.

Session Schedule

Friday, October 19

• 9:00 am - Conference Opening and Keynote
• 10:30 am - Break
• 11:00 am - Paper Session: Privacy and Trust
• 12:45 pm - Lunch (on your own)
• 2:15 pm - Panel: Where should we be investing most in research and practice to increase value of recommenders?
• 4:00 pm - Poster and Demo Session (starting with one-minute madness presentations
• 6:00 pm - Conference Dinner

Saturday, October 20

• 8:30 am - Paper Session: Algorithms: Collaborative Filtering
• 10:14 am - Break
• 10:45 am - Panel: Appraisal of Recommender Systems
• 12:30 pm - Lunch (details TBA)
• 2:00 pm - Paper Session: User Issues in Recommender Systems
• 3:45 pm - Break
• 4:00 pm - Paper Session: Algorithms: Learning
• 5:45 pm - Conference Ends

Keynote Speaker: Krishna Bharat

Computing with News The Internet has transformed the world by bring information into our lives ubiquitously. Information when delivered in a timely, relevant, and personalized fashion can be powerful. In the past it has been the prerogative of the news industry to educate minds, entertain readers, and empower citizens with information. However, the net has taken on this role more broadly. The union of computing and journalism was both logical and natural and has yielded a plethora of online news channels. Also, it has brought us value added citizen reporting in the form of blogs and recommendation networks. While connecting people to news and to social networks is very important, this is not the only way in which computing can help journalism. Computers can analyze vast news collections in an efficient and scalable way. They can extract information, induce structure, and transform and categorize text in ways that makes news easy to browse and search for both readers and journalists. This is a ripe area of research with tangible social benefits. If we can apply computers to make news more efficient to produce, distribute and absorb and fundamentally more truthful, it can better inform the actions taken by both individuals and nations. Bio Krishna Bharat is a Principal Scientist at Google Inc, Mountain View, California working on web and news search. He graduated with a Ph.D. in Computer Science from Georgia Tech in 1996. Before joining Google in 1999, he was a member of the research staff at DEC Systems Research Center in Palo Alto, CA. Krishna has served on the program committees of UIST and the World Wide Web Conference and has been a reviewer for the WWW Conference, UIST, SIGCHI, SIGIR and TOCHI. He has taught tutorials on Web-IR at SIGCHI and SIGIR. Krishna is the creator of Google News which won the 2003 Webby Award in the news category. Also, he received the 2003 World Technology Award for Media & Journalism. In 2004 he founded Google's R&D operations in India and served as the center's first director until 2006.

Invited Industry Panels -- NEW!

Building on a highlight of the Recommenders06 event in Bilbao, our practice/industry committee has assembled two exciting panels:

Friday Afternoon Panel: Where should we be investing most in research and practice to increase value of recommenders?

• Moderator: Todd Beaupre, Yahoo, Inc.
• Joaquin Delgado, CTO, Lending Club Corp.
• Jason Herskowitz, VP of Consumer Products, MyStrands
• Kartik Hosanagar, Assistant Professor, Wharton School of Business, University of Pennsylvania
• David Jennings, DJ Alchemi LLC
• Zac Johnson, Product Manager, All Media Guide, Inc.

Saturday Morning Panel: Appraisal of Recommender Systems

• Moderator: Kaushal Kurapati, Ask.com
• Delip Andra, CEO, Minekey, Inc.
• Jennifer Consalvo, Director of Personalization, AOL
• Greg Linden, Founder, Findory, Inc.
• Shail Patel, Platform Leader, Unilever Corporate Research
• Neel Sundaresan, Director, eBay Research Labs
• Tim Vogel, Chief Scientist, Aggregate Knowledge, Inc

Research Papers

RecSys 2007 has selected the following full-length research papers for publication and presentation at the conference:

Privacy and Trust

• Private Distributed Collaborative Filtering using Estimated Concordance Measures by Neal Lathia, Stephen Hailes, and Licia Capra, University College London
• Enhancing Privacy and Preserving Accuracy of a Distributed Collaborative Filtering by Shlomo Berkovsky (University of Haifa), Yaniv Eytani (University of Illinois at Urbana-Champaign), Tsvi Kuflik (University of Haifa), and Francesco Ricci (Free University of Bozen-Bolzano)
• Trust-aware Recommender Systems, by Paolo Massa and Paolo Avesani, IRST/FBK
• Making Recommender Systems Provably Manipulation-Resistant Through Influence Limits by Paul Resnick and Rahul Sami, University of Michigan

Algorithms: Collaborative Filtering

• Distributed Collaborative Filtering with Domain Specialization by Shlomo Berkovsky (University of Haifa), Tsvi Kuflik (University of Haifa), and Francesco Ricci (Free University of Bozen-Bolzano)
• Complex-network theoretic clustering for identifying groups of similar listeners in P2P systems by Amelie Anglade (Queen Mary College University of London), Marco Tiemann (Philips Research Europe), and Fabio Vignoli (Philips Research Europe)
• Robust Collaborative Filtering by Bhaskar Mehta, L3S
• A Recursive Prediction Algorithm for Collaborative Filtering Recommender Systems by Jiyong Zhang and Pearl Pu, EPFL

User Issues in Recommender Systems

• Supporting Product Selection with Query Editing Recommendations by Derek G Bridge (University College Cork) and Francesco Ricci (Free University of Bozen-Bolzano)
• Incorporating User Control into Recommender Systems Based on Naive Bayesian Classification by Verus Pronk, Wim Verhaegh, Adolf Proidl, and Marco Tiemann, Philips Research Laboratories
• Replaying Live-User Interactions in the Off-Line Evaluation of Critique-based Mobile Recommendations by Quang Nhat Nguyen and Francesco Ricci, Free University of Bozen-Bolzano
• Conversational Recommenders with Adaptive Suggestions by Paolo Viappiani, Pearl Pu, and Boi Faltings, EPFL

Algorithms: Learning

• Addressing uncertainty in implicit preferences by Sandra Gadanho and Nicolas Lhuillier, Motorola Labs
• Robustness of Collaborative Recommendation Based On Association Rule Mining by JJ Sandvig, Bamshad Mobasher, and Robin Burke, DePaul University
• Usage-Based Web Recommendations: A Reinforcement Learning Approach by Nima Taghipour, Ahmad Kardan, and Saeed Shiry Ghidary, Amirkabir University of Technology
• Improving New User Recommendations with Rule-based Induction on Cold User Data by An-Te Nguyen (University of Natural Sciences), Nathalie Denos (Laboratoire LIG), and Catherine Berrut (Laboratoire LIG)

Research Short Papers

RecSys 2007 has selected the following short research papers for publication and for poster presentation at the conference:

• A Probabilistic Model for Item-Based Recommender Systems by Ming Li (Unilever Corporate Research), Benjamin Dias (Unilever Corporate Research), Wael El-Deredy (University of Manchester), Lisboa Paulo (Liverpool John Moores University)
• A Recommender System for On-line Course Enrolment: An Initial Study by Michael P O'Mahony and Barry Smyth, University College Dublin.
• Case Amazon: Ratings and Reviews as Part of Recommendations by Juha Leino and Kari-Jouko Räihä, University of Tampere
• Comparing and Evaluating Information Retrieval Algorithms for News Recommendation by Toine Bogers and Antal Van den Bosch, University of Tilburg
• Dependence-based Collaborative Active Learning by Neil Rouben, Tokyo Institute of Technology
• Eigentaste 3.0: Constant-Time Adaptability with Item-Item Clustering by Tavi Nathanson and Ken Goldberg, University of California, Berkeley
• Effective Explanations of Recommendations: User-Centered Design by Nava Tintarev and Judith Masthoff, University of Aberdeen
• Evaluating Information Presentation Strategies for Spoken Recommendations by Andi Winterboer and Johanna D. Moore, University of Edinburgh
• Leveraging Aggregate Ratings for Better Recommendations by Akhmed Umyarov and Alexander Tuzhilin
• Supporting Social Recommendations with Activity-Balanced Clustering by F. Maxwell Harper, Shildad Sen, and Dan Frankowski, University of Minnesota
• The Evaluation of a Hybrid Critiquing System with Preference-based Recommendations Organization by Li Chen and Pearl Pu, EPFL
• The KeepUP Recommender System by Andrew Webster and Julita Vassileva, University of Saskatchewan
• Towards Ensemble Learning for Hybrid Music RecommendationMarco Tiemann and Steffen Pauws, Philips Research
• Toward the Exploitation of Social Access Patterns for Recommendation by Jill Freyne (University College Dublin), Maurice Coyle (University College Dublin), Rosta Farzan (University of Pittsburgh)

Practice/Industry Track Abstracts

• Demo: TechLens – A Researcher’s Desktop by Nishikant Kapoor, Jilin Chen, John T. Butler, Gary C. Fouty, James A. Stemper, John Riedl, and Joseph A. Konstan, University of Minnesota
• Challenge: The Challenges of Recommending Digital Selves in Physical Spaces by Joseph F. McCarthy, Nokia Research Center

Doctoral Symposium

The following five student submissions were selected to participate in the RecSys 2007 doctoral symposium. While the symposium itself is a private event for these students and their faculty mentors, their work will be displayed at the poster and demo session on Friday, October 19th.

• Justin Donaldson (Indiana University) A Hybrid Social-Acoustic Recommendation System for Popular Music
• Xin Fu (University of North Carolina) Evaluating Sources of Implicit Feedback in Web Searches
• Fabiana Lorenzi (Federal University of Rio Grande do Sul) A Multiagent Knowledge-based Recommender Approach with Truth
• Mike Radmacher (Johann Wolfgang Goethe University) Elicitation of Profile Attributes by Transparent Communication
• Nava Tintarev (University of Aberdeen) Explanation of Recommendations
• Riina Vuorikari (K.U. Leuven) Can Social Information Retrieval Enhance the Discovery and Reuse of Digital Educational Content?