Computing and Information Systems - Theses

Permanent URI for this collection

Search Results

Now showing 1 - 1 of 1
  • Item
    Thumbnail Image
    Developing systems for gene normalisation
    Goudey, Benjamin ( 2007-10)
    The rapid growth of biomedical literature has attracted interest from the text mining community to develop methods to help manage the ever-increasing amounts of data. Initiatives such as the BioCreative challenge (Hirschman et al. 2005b) have created standard corpora and tasks in which to evaluate a variety of systems in a common framework. One such task is gene normalisation, in which the problems of synonymy and polysemy in gene name identification are overcome by mapping each mention back to a unique identifier, unambiguously identifying that gene. This task is one of the foundations required for any kind of text mining system working with biomedical literature, where we must be very certain of which genes are being discussed in the text. (For complete abstract open document)