Proceedings of the Workshop on BioNLP Shared Task - BioNLP '09

Abstract

The need for automatic processing of the rapidly increasing body of publications describing results in molecular biology continues to drive efforts in Biomedical natural language processing (BioNLP). Until recently, domain efforts have largely concentrated on foundational tasks such as entity recognition and relatively simple information extraction targets such as interacting entity pairs. By contrast, biological research increasingly aims to create detailed descriptions of complex processes in biological systems. To respond to the needs of such research, it is necessary to develop BioNLP methods that are able to process more fine-grained representations. The BioNLP'09 Shared Task is the first community-wide step in this direction. Shared tasks have a strong tradition in the BioNLP community. The TREC Genomics, KDD cup, JNLBPA, LLL and BioCreative tasks have focused the efforts of the community on timely challenges in the domain, both establishing the capabilities and problem points of current systems as well as advancing the state of the art in various areas of biomedical text mining. These are also the goals of the present shared task. The focus on a rich representation of extracted information in the BioNLP'09 Shared Task can also be seen as natural continuation of the succession of previous tasks. The BioNLP'09 Shared Task targets event extraction following a model similar to those currently applied in the wider IE community. Corpus resources supporting this type of representation have only recently become available in the domain, and the task thus represents a novel challenge to much of the community. Despite its novelty and relatively complex task settings, the BioNLP'09 Shared Task met with an enthusiastic response from the community: more than 40 teams registered their preliminary interest, and 24 teams submitted final results. Thanks to the efforts of the participants and the shared task program committee in reviewing, we have the pleasure of presenting these proceedings of 19 manuscripts accepted for presentation at the shared task session of the workshop.

Related Papers

• → Text Mining: Natural Language techniques and Text Mining applications(1998)148 cited
• → Improving Biomedical Information Extraction with Word Embeddings Trained on Closed-Domain Corpora(2019)13 cited
• → Review on Event Extraction for BioNLP with a Survey(2023)2 cited
• → A biomedical events extracted approach based on phrase structure tree(2017)2 cited
• Design and Development of Integrated Biomedical Ontology for Information Extraction from Medline Abstracts(2013)