Program
The workshop consists of an opening keynote by Prof. Joe Konstan, oral presentations of research papers (12+3 minutes), the presentation of position papers as posters during the coffee break, and a concluding open discussion.
The proceedings of the workshop are available at CEUR-WS.org (Volume 2462)
Keynote
Recommending for Impact: Intentions, Algorithms, and Metrics (slides)
Abstract: What does it mean to recommend for impact? Where are there tensions between the interests of the recommender operator and the interests of the user? Can we measure the impact of a recommendation, or a recommender algorithm? And how can we optimize algorithms to improve impact? This talk takes a long look at the field of recommender systems, including periods of significant impact-focused developments and periods of technical progress made by looking at other objectives. Building on that past, we look at promising directions for impact-focused recommender systems research.
|
|
Joseph A. Konstan is Distinguished McKnight
University Professor, Distinguished University Teaching
Professor, and the College of Science and Engineering's
Associate Dean for Research at the University of
Minnesota. His research addresses a variety of
human-computer interaction issues, including
personalization (particularly through recommender
systems), eliciting on-line participation, and designing
computer systems to improve public health. He is probably
best known for his work in collaborative filtering
recommenders (the GroupLens project, which won the ACM
Software Systems Award). |
Detailed Schedule
14:00 - 14:10 Opening
14:10 - 15:00 Keynote (Joe Konstan)
15:00 - 15:30 Paper Session 1
15:30 - 16:00 Coffee Break / Poster Presentations
16:00 - 17:00 Paper Session 2
17:00 - 17:30 Open Discussion and Workshop Closing
Position Papers
Best Paper
The best paper award sponsored by Intuit goes to:
Recommendations for Long-Term Profit Optimization by Patrick Hosein, Inzamam Rahaman, Keanu Nichols and Kiran Maharaj
Background
Research in the area of recommender systems is largely focused
on helping individual users finding items they are interested
in. This is usually done by learning to rank the recommendable
items based on their assumed relevance for each user. The
implicit underlying goal of a such system is to affect users in
different positive ways, e.g., by making their search and
decision processes easier or by helping them discover new
things.
Recommender systems can, however, also have other more
directly-measurable impacts, e.g., such that go beyond the
individual user or the short term influence. A recommender
system on a news platform, for example, can lead to a shift in
the reading patterns of the entire user base. Similarly, on
e-commerce platforms, it has been shown that a recommender can
induce significant changes in the purchase behavior of
consumers, leading, for example, to generally higher sales
diversity across the site. On the other hand, recommender
systems usually serve certain business goals and can have an
impact not only on the customers, e.g., by stimulating higher
engagement on a media streaming platform or a social network,
but also direct and indirect affect sales, revenue or conversion
and churn rates.
Goals of the Workshop
The research literature that considers such more direct
measurements of impact of recommender systems on the various
stakeholders is comparably scarce and scattered. With the
proposed workshop, we pursue different goals.
- First, the workshop will serve as a platform where
researchers can present their latest works in which they
analyzed different forms of impact of recommenders. We
consider both papers where impact on individual users was
measured (e.g., more healthy eating habits that were
stimulated by a food recommender or a more efficient choice
process), papers that highlight effects on a community or a
society as a whole, and papers that demonstrate effects in
terms of business value.
- Second, the goal of the interactive session is to explore
new ways how the impact of recommender systems can be measured
within academic settings, i.e., were some impacts can only be
analyzed based on simulations or alternative computational
measures.
- Third, the workshop shall serve as an instrument to raise
awareness in the community regarding the importance of
impact-oriented research. This aspect in our view is
particularly important as more and more research works
indicate that optimizing for the most accurate prediction not
necessarily leads to the best recommendations in terms of the
users' quality perception or the desired effects of a
recommender.
Topics
The topics of interest include, e.g.,
- Field studies on the impact and business value of
recommender systems
- Offline evaluation protocols and measures to assess the
impact or recommenders
- User studies on the effects of recommenders on users, e.g.,
on decision making
- Simulation-based approaches to impact assessment
- Price- and profit-awareness of recommender systems
- Network effects of recommender systems
- Multi-stakeholder recommender systems
- Beyond accuracy optimization methods for recommender systems
(e.g., business metrics, user satisfaction)
- Long-term implications of deployed recommender systems
Submission and Publication
Submission types are long research papers reporting on complete
research (8 pages plus references), short papers reporting on work
in progress (4 pages plus references), and position papers (up to
two pages). Submissions must be formatted according to the
conference guidelines (
ACM
SIG Proceedings Template) and submitted via
EasyChair.
The review process is single-blind, i.e., please include
author names in the papers. All papers will be peer reviewed by
the workshop's program committee.
Accepted papers will be published within formal proceedings at
CEUR-WS.org.
The best paper will be awarded with a Google Home (can be
replaced with an equivalent prize upon request).
Important dates
- Paper submission deadline: July 1st, 2019
- Author notification: July 29th, 2019
- Camera-ready version deadline: August 27th,2019
- ImpactRS'19 Workshop: September 19th, 2019
- RecSys conference: September 16-20th, 2019
Organization
Workshop Co-Organizers
- Oren Sar Shalom: Oren is a principal
data scientist at Intuit AI, where he conducts applied machine
learning research in the fields of recommender systems and NLP
- Dietmar Jannach: Dietmar is a professor
of computer science at the University of Klagenfurt, Austria,
with a special research focus on recommender systems.
- Ido Guy: Ido is a Global Director of
Applied Research at eBay. Prior to that he was a Principal
Research Engineer at Yahoo! Labs and a Senior Technical Staff
Engineer and Manager at IBM Research.
Program Committee
- Gediminas Adomavicius, University of Minnesota
- Christine Bauer, University of Innsbruck
- Joeran Beel, Trinity College Dublin
- Pablo Castells, Universidad Autónoma de Madrid
- Paolo Cremonesi, Politecnico di Milano
- Michael Ekstrand, Boise State University
- Alexander Felfernig, TU Graz
- Maurizio Ferrari Dacrema, Politecnico di Milano
- Werner Geyer, IBM T. J. Watson Research
- Michael Jugovac, TU Dortmund
- Surya Kallumadi, The Home Depot
- Iman Kamehkhosh, TU Dortmund
- Gal Lavee, Microsoft
- Slava Novgorodov, eBay Research
- Massimo Quadrana, Pandora
- Filip Radlinski, Google
- Adi Shalev, Intuit
- Harald Steck, Netflix
- Markus Zanker, Free University of Bozen-Bolzano
- Yong Zheng, Illinois Institute of Technology
- Alex Zhicharevich, Tel Aviv University
Contact
If you have questions regarding the workshop, do not hesitate
to contact the workshop chairs: impactrs19@ainf.at