SHREC 2007-Protein Challenge

The participants will have to provide one ranked list for each query object.
The performance will be measured as described in [9]: For each query exists a set of highly relevant items (same SCOP fold) and a set of marginally relevant items (same SCOP class). The submitted ranked lists are turned into a gain vector by replacing item IDs by their relevance scores. A non-relevant retrieved item corresponds to relevance score 0. Further, marginally relevant has score 1 and highly relevant score 2.
For each individual submitted query result list, the following performance measures are calculated: Average Precision, First Tier, Second Tier, Precision, Recall, (all these are evaluated for only highly relevant items, and all relevant items), Average Dynamic Recall, Cumulated Gain, Normalized Cumulated Gain, Discounted Cumulated Gain, Normalized Discounted Cumulated Gain (all for the first 5, 10, 25, 50, and 100 ranked items), Average (Discounted) Cumulated Gain vs. Recall %, (Discounted) Cumulated Gain vs. Rank [1,100], and Normalized (Discounted) Cumulated Gain vs. Rank [1,100]. Additional to the query lists, the participants are asked to provide an abstract of the applied method and optionally the executable program or source code.

This homepage was created by Maja Temerinac for the SHREC 2007 Protein Challenge.
For more information, please contact:
Maja Temerinac temerina(at)informatik.uni-freiburg.de or
Marco Reisert reisert(at)informatik.uni-freiburg.de