Improving Recall for Automatic Information Extraction: Final Status Report

Abstract

This report describes the results of the SBIR Phase I research project on improving recall for automatic information extraction (IE), carried out by Onyx Consulting, Inc. Current IE systems suffer from a number of limitations, especially with respect to the types and amount of knowledge that they bring to bear on the process of extraction. The current effort works on improving the coverage and quality of proper name recognition but also significantly on enhancing the syntactic and semantic knowledge used in extraction. In addition, we also work on resolving multiple references to the same entity in the text. Preliminary testing suggests that the knowledge-intensive methods contribute to enhancing the recall in information extraction.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Dec 05, 2000
Accession Number
ADA385283

Entities

People

  • Charles Watts
  • James Cowie
  • Sergei Nirenburg

Tags

Communities of Interest

  • Weapons Technologies

DTIC Thesaurus Topics

  • Acquisition
  • Army Aviation
  • Artificial Intelligence
  • Automatic
  • Commerce
  • Contracts
  • Electronic Commerce
  • Internet
  • Language
  • Models
  • Natural Language Processing
  • Ontologies
  • Recognition
  • Software Development
  • Text Processing
  • United States
  • Web Browsers

Readers

  • Computational Linguistics
  • Organizational Process Management (OPM).
  • Systems Analysis and Design

Technology Areas

  • AI & ML
  • AI & ML - Information Retrieval