Our TEAM took part in the BioNLP 2009 Shared task on Event Extraction, organised by the Genia group. The main aim of the challenge was the extraction of bio-events from the literature, focusing particularly on molecular events involving proteins and genes. The shared task was designed to address a semantically rich information extraction problem as a whole, divided into three subtasks. We took part in Task 1: Event detection and characterisation.
There were more than 40 teams who expressed initial interest in participating in the Challenge. Final submissions were received from 24 teams who completed task 1, and six teams each of tasks 2 and 3. The results for our TEAM show that we were right in the middle of the table (12th out of 24 teams), with good results on non-regulation events (8th for the F-score).
- Detailed ranking results (our pseudonym is UNIMAN)
- F. Sarafraz, J. Eales, R. Mohammadi, J. Dickerson, D. Robertson, G. Nenadic: Biomedical Event Detection using Rules, Conditional Random Fields and Parse Tree Distances , Proceedings of the BioNLP shared task 2009 (in press)