Towards Developing Effective Fact Extractors

Abstract

DSTO has a program of research into automated text processing. Part of this research has led to the development of a prototype information extraction system known as the DSTO Fact Extractor System. This system can be used to extract interesting information from free text documents. Part of applying the DSTO technology involves a skilled user developing a set of one or more fact extractors that control the behaviour of the information extraction engine. These fact extractors are developed with the aid of an integrated development environment known as the Fact Extractor Workbench. This report uses a range of examples to discuss the issues that must be considered when developing fact extractors.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jun 01, 2005
Accession Number
ADA437663

Entities

People

  • Greg Chase
  • Jyotsna Das
  • Scott Davis

Organizations

  • Defence Science and Technology Group

Tags

Communities of Interest

  • C4I
  • Energy and Power Technologies
  • Space
  • Weapons Technologies

DTIC Thesaurus Topics

  • Accuracy
  • Case Studies
  • Command And Control
  • Commerce
  • Control Systems
  • Databases
  • Electronic Mail
  • Information Science
  • Information Systems
  • Internet
  • Mobile Phones
  • Named Entity Recognition
  • Network Protocols
  • Personnel Management
  • Text Processing
  • Warfare
  • Websites

Fields of Study

  • Computer science

Readers

  • Database Systems and Applications
  • Neural Network Machine Learning.
  • Theoretical Analysis.

Technology Areas

  • AI & ML
  • AI & ML - Information Retrieval
  • AI & ML - Machine Learning Algorithms