Overview of the infectious diseases (ID) task of BioNLP Shared Task 2011

  • Authors:
  • Sampo Pyysalo;Tomoko Ohta;Rafal Rak;Dan Sullivan;Chunhong Mao;Chunxia Wang;Bruno Sobral;Jun'ichi Tsujii;Sophia Ananiadou

  • Affiliations:
  • University of Tokyo, Tokyo, Japan;University of Tokyo, Tokyo, Japan;University of Manchester, Manchester, UK;Virginia Bioinformatics Institute, Virginia Tech, Blacksburg, Virginia;Virginia Bioinformatics Institute, Virginia Tech, Blacksburg, Virginia;Virginia Bioinformatics Institute, Virginia Tech, Blacksburg, Virginia;Virginia Bioinformatics Institute, Virginia Tech, Blacksburg, Virginia;Microsoft Research Asia, Beijing, China;University of Manchester, Manchester, UK

  • Venue:
  • BioNLP Shared Task '11 Proceedings of the BioNLP Shared Task 2011 Workshop
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper presents the preparation, resources, results and analysis of the Infectious Diseases (ID) information extraction task, a main task of the BioNLP Shared Task 2011. The ID task represents an application and extension of the BioNLP'09 shared task event extraction approach to full papers on infectious diseases. Seven teams submitted final results to the task, with the highest-performing system achieving 56% F-score in the full task, comparable to state-of-the-art performance in the established BioNLP'09 task. The results indicate that event extraction methods generalize well to new domains and full-text publications and are applicable to the extraction of events relevant to the molecular mechanisms of infectious diseases.