Challenge and Workshop on Context-aware Movie Recommendation

This is a multi-part message in MIME format...

------------=_1272694900-2478-147
Content-Language: en-US
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

- Apologies for cross-posting
- Please forward to your students/colleagues, who might consider participat=
ing


CALL FOR PAPERS AND PARTICIPATION

CAMRa-2010 :: Challenge and Workshop on Context-aware Movie Recommendation
Joint event with CARS-2010 :: RecSys-2010 :: Barcelona :: September 30, 2010
   http://www.dai-labor.de/camra2010/


General Information:
--------------------
The majority of existing recommendation approaches does not take into accou=
nt contextual information, such as time, location, or weather. This challen=
ge aims to tackle the practical issue of context-aware movie recommendation=
. Two new movie ratings datasets, one from Moviepilot and one from Filmtips=
et, are released for the challenge. The datasets contain a number of contex=
tual features, typically not found in standard collaborative filtering data=
sets. The participating teams are requested to the additional contextual fe=
atures to generate context-aware recommendations.=20

The challenge focuses on classification accuracy metrics. The participants =
are invited to submit papers focusing on the challenge to the workshop, whi=
ch will be conducted in conjunction with RecSys-2010. The winners of the ch=
allenge will be invited to submit extended versions of their papers to the =
follow-up special issue on context-aware recommendations (not confirmed yet=
).


Challenge:
----------
The challenge consists of three tracks: one track is common for both datase=
ts and a separate for each dataset. Participating teams are invited to take=
 part in any track. The number of tracks a team is participating in is not =
limited. Both datasets are anonymized to protect the users of each service.=
 Participants are expected to use the provided datasets; the use of externa=
l information sources, like IMDB, Wikipedia, or NetFlix, is not allowed.

The challenge tracks are:
* Weekly Recommendation. Focuses on the temporal dimension of context. The =
participants are asked to recommend movies for two different weeks: the Chr=
istmas week and the week leading to the Oscars ceremony. For this track, an=
y of the two datasets can be used. The algorithms used for each dataset may=
 differ.
* Moviepilot Track. Focuses on the mood of movies. The participants are ask=
ed to recommend a list of movies for a selection of users based on a given =
mood.
* Filmtipset Track. Focuses on the social connections between users. The pa=
rticipants are asked to recommend a list of movies for a selection of users=
 based on the friend relations in the social network.

The evaluations should address the following metrics: MAP, P@5, P@10, and A=
UC. The performance of the teams will be published on the leaderboard. Addi=
tional datasets will be released for the final evaluation to identify the w=
inners of each track. An online evaluation with real users will be conducte=
d in some tracks.


Datasets:
---------
There are four datasets available:  Moviepilot_week, Filmtipset_week, Movie=
pilot_mood, Filmtipset_social. To access the datasets, please send an inqui=
ry stating which dataset(s) you request, your affiliation, and which track(=
s) you intend to participate in to [email protected] The request will=
 help the organizers to estimate the number of participants.


Call for Papers:
----------------
The participants are invited to submit papers focusing on the challenge and=
 algorithms evaluated using the released datasets. The submissions are limi=
ted to 8 pages in the ACM SIG proceedings format. Each submission should be=
 accompanied by a 1-page summary of the algorithm. The papers will be revie=
wed by the Program Committee based on significance, technical soundness, an=
d presentation clarity. The 1-page summaries will be reviewed by the Expert=
 Panel based on creativity, originality, and scalability.

Paper submissions and reviews will be handled electronically through the CA=
MRa page in EasyChair (will be available later). The authors of high-qualit=
y papers will be invited to submit the extended versions of their work to a=
 planned journal on context-awareness in recommender systems.

Important dates:
* Paper submission: July 1, 2010
* Notification: July 22, 2010
* Camera-ready submission: August 16, 2010
* 1-pager summary submission: August 16, 2010
* Feedback from the Expert Panel: September 10, 2010
* Winners announced: RecSys dinner
* Workshop: September 30, 2010


Organizers and Committees:
--------------------------
General Chairs ([email protected])
* Shlomo Berkovsky, CSIRO
* Ernesto William De Luca, DAI Lab/Technische Universit=E4t Berlin
* Alan Said, DAI Lab/Technische Universit=E4t Berlin

Industrial Chairs
* Jannis Hermanns, Moviepilot
* Magnus Hoem, Filmtipset/Entertainity

Program Committee
* Sahin Albayrak, DAI-Laboratory/Technische Universit=E4t Berlin, Germany
* Liliana Ardissono, Universit=E0 degli Studi di Torino, Italy
* Lora Aroyo, Free University Amsterdam, The Netherlands
* Stephan Baumann, DFKI, Germany
* Dominik Benz, University of Kassel, Germany
* Toine Bogers, Royal School of Library Information Science, Denmark
* Li Chen, Hong Kong Baptist University, China
* Jill Freyne, CSIRO, Australia
* Ido Guy, IBM Research, Israel
* Tom Heath, Talis, UK
* Robert J=E4schke, University of Kassel, Germany
* J=E9r=F4me Kunegis, DAI-Laboratory/Technische Universit=E4t Berlin, Germa=
ny
* Leandro Balby Marinho, Universit=E4t Hildesheim, Germany
* Jon Sanders, Netflix, USA
* Rene Schult, Otto-von-Guericke-Magdeburg, Germany
* Bracha Shapira, Ben Gurion University, Israel
* Myra Spiliopoulou, Otto-von-Guericke-Magdeburg, Germany
* Domonkos Tikk, Budapest University of Technology and Economics, Hungary
* Robert Wetzker, Technische Universit=E4t Berlin, Germany

Expert Panel
* Gedas Adomavicius, University of Minnesota, USA
* Sarabjot Singh Anand, University of Warwick, UK
* Derek Bridge, University College Cork, Ireland
* Anthony Jameson, DFKI, Germany
* Joseph Konstan, University of Minnesota, USA
* Yehuda Koren, Yahoo! Research, Israel
* Francisco Martin, Strands, Spain
* Bamshad Mobasher, DePaul University, USA
* Francesco Ricci, Free University of Bozen-Bolzano, Italy
* Barry Smyth, University College Dublin, Ireland
* Alex Tuzhilin, New York University, USA
* Qiang Yang, Hong Kong University of Science and Technology, China


Further Information
-------------------
* Webpage: http://www.dai-labor.de/camra2010/
* Contact: [email protected]

------------=_1272694900-2478-147
Content-Type: text/plain; name="message-footer.txt"
Content-Disposition: inline; filename="message-footer.txt"
Content-Transfer-Encoding: 8bit

- User Modeling mailing list -

Please, use the following address to post messages to the mailing list:
[email protected]

Web page of the mailing list:
https://www.di.unito.it/listserver/info/um (secure web site)


------------=_1272694900-2478-147--