Leaders: Diego Molla and Robert Dale
Every day new information is available in text, images, sounds, and other media that are difficult to search with current technology. Existing text retrieval systems effectively treat documents as unstructured bags of words. Yet it is clear that human processors of information make use of a much deeper understanding of text than these systems exhibit. Humans cannot compete with machines in terms of quantity, but their abilities far exceed those of machines when it comes to quality.
This priority research area focuses on how we can improve search technology by integrating more knowledge about the processing and understanding of language. This knowledge comes from a range of disciplines represented within HCSNet. What aspects of natural language processing are most effective to find information and present it to the user? How can current linguistics theories help us find better results? Are there insights from the cognitive sciences that can tell us how to build better tools for finding information? How do we extend these technologies when the data we are concerned with includes images, audio and video as well as text?
Events
- 27-28 July 2009: Hands-on Workshop on Mashups at Macquarie University, NSW
- 16-17 April 2009: NGS09 at Jemby-Rinjah Eco Lodge, Blue Mountains, NSW
- 5 December 2008: Search and Information Extraction from Audio Data at the HCSNet SummerFest, Sydney
- 30 November 2008: Search on Biomedical Information at Brisbane
- 13 November 2008: NGS08 at RMIT University
- 16-17 May 2008: Joint HCSNet-EII Workshop on Interactive and Ubiquitous Information Access at NICTA ATP, Sydney
- 19-20 July 2007: NGS07 at CSIRO-Macquarie University
- 21-22 September 2006: NGS06 at University of Melbourne