Revision of Fusion Task from Tue, 10/27/2009 - 10:39

Short description

Goal of the fusion task is to explore strategies for fusing data of visual and textual retrieval systems for an optimal multi-modal retrieval performance. We plan to make limited training data available, so data on a very similar database and also the ground truth. All data will be in the trec_eval form so use of trec_eval is necessary for the testing. trec_eval is available from here.
Registration for the fusion task is free of charge and can be done here.

Contact person

• Jayashre Kaplathy-Cramer, OHSU, Portland, OR, USA, kalpathy@ohsu.edu
• Henning Müller, University of Applied Sciences Western Switzerland, Sierre, henning.mueller@sim.hcuge.ch

Important dates

15.11.2009 Release of training data
30.11.2009 Release of test data
15.1.2010 Paper submission of main conference
15.1.2009 Submission of results without training data
31.1.2009 Release of ground truth data
(31.3.2009 Submission of results with ground truth data, as comparison with other results)
30.04.2009 Deadline for overview papers for contests
23.8.-26.8.2010 ICPR Conference

Data sets

Interested participants will received the best purely visual and the best purely textual runs of the 2009 ImageCLEF medical retrieval task and then need to combine the runs in a way to optimize the performance.

Format of the supplied runs and of the results to submit is the trec_eval format.

Performance measures

All performance measures of trec_eva will be made available.

Lead measure will be mean average precision (MAP) but early precision (P@10, P@30) will also be used for rankings.

How to register for the task

We plan to set up a registration interface by the end of October so interested participants can register automatically for this task.