Advanced document analysis in an intelligent information system

This paper discusses the use of advanced document analysis techniques in an intelligent information system that aids analysts in the health domain. The HINTS operational prototype provides analysts with tools to manage, analyse and assimilate a large quantity of electronic documents. Document metadata is automatically extracted from documents by: 1. A Fact Extractor that identifies key domain concepts; 2. The InTEXT Analyser, based on the NDA (Artificially Intelligent Document Analyser) technology. It provides summary and keyword information. The storage of document catalogue information is in XML, and dynamic views and searches of the data are served out to web clients and to a Geospatial Java application.