Ph.D. Research Proposal Information Extraction for Legal Documents

Information Extraction technique, which is a process of converting unstructured text into structured data, has been used in a variety range of applications such as traffic monitoring systems, biomedical systems and more general open knowledge bases for open language learning. However, the use of information extraction in legal area is still quite limited and suffers difficulties. For example, the existing system AustLII, which is developed and maintained by University of Technology Sydney and University of New South Wales, uses traditional information retrieval approach to organise all the legal documents and the system only supports keywords search. Another example, the LexML Web project in Brazil also applies similar approach. Although more sophisticated techniques such as case based reasoning, semantic network and ontology based approaches are introduced, not many of them are actually used in an integrated application. Hence, the aim of this study would be to come up with new design that applies information extraction techniques in order to construct a more accurate legal text management system. The new system is designed to manage laws as well as legal cases for both legal study and use. Comparing with AustLII, the new system will also focus on Australia laws but provide more accurate search results because of the use of information extraction methods. Australasian Legal Information Institute