Second Contextual Suggestion Track
Text REtrieval Conference (TREC 2013)
November 19 - 22, Gaithersburg MD, USA
Submissions Due: July 15, 2013
* Call for Participants
The Contextual Suggestion track considers a complex information need, highly dependent on context, which nonetheless can be simply expressed. The 2012 and 2013 tracks focus on one situation: a user with a mobile device with limited interaction but some sort of user profile; who is in a strange town; and who is looking for something to do. There is no explicit query: the implicit query is "here I am, what should I do?".
As input to the task, participants will be given a set of profiles, a set of geographical contexts, and a set of example suggestions. Each profile corresponds to a single user, and indicates that user's preference with respect to each example suggestion. Each context corresponds to a particular city, for example, Gaithersburg, Maryland. Each example suggestion includes a title, description, and an associated URL. One example suggestion might be to have a beer at the Dogfish Head Alehouse, and the profile might include a negative preference with respect to that suggestion.
For each profile/context pairing, participants should return a ranked list of up to 50 ranked suggestions. Each suggestion should be appropriate to the profile (based on the user's preferences) and the context (according to the location). The description of the suggestion may be tailored to reflect the preferences of that user. Profiles correspond to the stated preferences of real individuals, who will return to judge proposed suggestions. For the purposes of this experiment, you can assume users are of legal drinking age at the location specified by the context, has up to five hours available to follow a suggestion, and has access to appropriate transportation (e.g., a car).
The 2013 track covers several aspects:
- Filtering of suggestions based on context.
- Ranking of suggestions based on profiles.
- Personalization of suggestion descriptions.
* We Need Your Help!
Help us define the best methods of suggesting attractions:
- Sign up now!
- Start investigating the problem when contexts become available in early May and profiles become available in early June.
- Submit suggestions by July 15.
- Participate in the track workshop in November.
* Further details
See the Contextual Suggestion Track guidelines at https://sites.google.com/site/treccontext/trec-2013-guidelines
Sign up for TREC'13 at http://trec.nist.gov/pubs/call2013.html
Mailing list: Send a mail message to listproc (at) nist.gov such that the body consists of the line subscribe trec-context <FirstName> <LastName>
Text REtrieval Conference (TREC 2013)
November 19 - 22, Gaithersburg MD, USA
Submissions Due: July 15, 2013
* Call for Participants
The Contextual Suggestion track considers a complex information need, highly dependent on context, which nonetheless can be simply expressed. The 2012 and 2013 tracks focus on one situation: a user with a mobile device with limited interaction but some sort of user profile; who is in a strange town; and who is looking for something to do. There is no explicit query: the implicit query is "here I am, what should I do?".
As input to the task, participants will be given a set of profiles, a set of geographical contexts, and a set of example suggestions. Each profile corresponds to a single user, and indicates that user's preference with respect to each example suggestion. Each context corresponds to a particular city, for example, Gaithersburg, Maryland. Each example suggestion includes a title, description, and an associated URL. One example suggestion might be to have a beer at the Dogfish Head Alehouse, and the profile might include a negative preference with respect to that suggestion.
For each profile/context pairing, participants should return a ranked list of up to 50 ranked suggestions. Each suggestion should be appropriate to the profile (based on the user's preferences) and the context (according to the location). The description of the suggestion may be tailored to reflect the preferences of that user. Profiles correspond to the stated preferences of real individuals, who will return to judge proposed suggestions. For the purposes of this experiment, you can assume users are of legal drinking age at the location specified by the context, has up to five hours available to follow a suggestion, and has access to appropriate transportation (e.g., a car).
The 2013 track covers several aspects:
- Filtering of suggestions based on context.
- Ranking of suggestions based on profiles.
- Personalization of suggestion descriptions.
* We Need Your Help!
Help us define the best methods of suggesting attractions:
- Sign up now!
- Start investigating the problem when contexts become available in early May and profiles become available in early June.
- Submit suggestions by July 15.
- Participate in the track workshop in November.
* Further details
See the Contextual Suggestion Track guidelines at https://sites.google.com/site/treccontext/trec-2013-guidelines
Sign up for TREC'13 at http://trec.nist.gov/pubs/call2013.html
Mailing list: Send a mail message to listproc (at) nist.gov such that the body consists of the line subscribe trec-context <FirstName> <LastName>
No comments:
Post a Comment