"Mining events from the literature for bioinformatics applications" by S. Ananiadou, P. Thompson, and R. Nawaz; with Martin Vesely as coordinator

  • Authors:

  • Sophia Ananiadou;Paul Thompson;Raheel Nawaz

  • Affiliations:

  • National Centre for Text Mining, School of Computer Science, University of Manchester, UK;National Centre for Text Mining, School of Computer Science, University of Manchester, UK;National Centre for Text Mining, School of Computer Science, University of Manchester, UK

  • Venue:
  • ACM SIGWEB Newsletter
  • Year:
  • 2013

Quantified Score

Hi-index 0.00

Visualization

Abstract

The ever-increasing rate at which scientific articles are being published means that text mining (TM) is becoming a necessary technology to allow information relevant to a user's search to be isolated from the potential mountain of irrelevant information. Whilst the extraction of named entities, such as genes, proteins and phenotypes, is a well-studied topic, researchers are usually interested in discovering information about specific types of biomedical reactions in which these entities are involved. In order to facilitate efficient searching for such reactions, TM systems need to account for the fact that various types of relationships and links exist between entities in texts. In this article, we describe how the identification of such relationships, together with interpretative information from their textual contexts, can help to create structured representations of biomedical reactions (called events) from unstructured text. We detail the various challenges of extracting events from text, and explain how various tools, resources and infrastructures can help in the development of event extraction systems. Finally, we describe some concrete applications that make use of event extraction technology, i.e., semantic search systems and linking biological pathways with textual evidence.