Find out how ICT can support biomedical and clinical researchFind out more. From Clever cars to clever farms... Embedded Systems
Document Analysis and Understanding

The information revolution has led to an the explosive growth of unstructured documents available for analysis. Some examples include
  • Web Pages
  • E-mails 
  • Scanned Books
  • Blogs 
  • News feeds.
Another area where abundant information is available is from the e-government initiative. Such data glut leads to the need to understand this data and transform it into knowledge.

The DonAU project aims to use state of the art non-parametric statistics and natural language processing for large scale document analysis. We plan to develop tools that applicable to NLP in general and to intelligence analysis in particular. Our industry partner in this project is a local SME, The Distillery Pty. Ltd. (homepage), who specialize in software for law enforcement and intelligence services in the context of document analysis.

The theoretical tools used in this context are kernel methods, Conditional Random Fields (CRF's), graphical models, fast string comparison methods, kernels on structured data, and low-dimensional data representation. These tools will be implemented within our statistical inference toolbox codenamed ELEFANT. Many of these algorithms will be made available publicly when they are developed and stabilized.

 

Contact 1: Dr S V N Vishwanathan
Phone: +61 2 6125 8657
Email: SVN.Vishwanathan -at- nicta.com.au

Contact 2: Dr Alex Smola
Phone: +61 2 6125 8652
Email: Alex.Smola -at- nicta.com.au