Equipment Simulation for Language Understanding,

Abstract

This work is part of PROTEUS (the PROtotype TExt Understanding System), currently under development. The objective is to understand short natural language texts. Our texts at present are CASualty REPorts (CASREPs) which describe failures of equipment installed on Navy ships. We consider the task of analyzing reports regarding the failure, diagnosis and repair of equipment. We show that a general knowledge of equipment is not sufficient for a full understanding of such reports. As an alternative, we describe a system - PROTEUS - which relies on a detailed simulation model to support language understanding. We describe the structure of the model and emphasize features specifically required for language understanding. We show how this model can be used in analyzing and determining the referents for complex noun phrases. We point out the importance of identifying the implicit causal relations in the text and show how the simulation capabilities of the model support this task. Finally, we indicate how PROTEUS may be extended to facilitate the entry of new equipment models and to assist in equipment diagnosis.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Sep 01, 1987
Accession Number
ADA188639

Entities

People

  • Ralph David Grishman
  • Tomasz Ksiezyk

Organizations

  • New York University

Tags

DTIC Thesaurus Topics

  • Air Compressors
  • Air Pressure
  • Artificial Intelligence
  • Cognitive Science
  • Compressors
  • Computer Science
  • Couplings
  • Drive Shafts
  • Gas Turbines
  • High Pressure
  • Language
  • Natural Language Processing
  • Natural Languages
  • New York
  • Ontologies
  • Reasoning
  • Simulations

Readers

  • Computational Linguistics
  • Software Engineering
  • Theoretical Analysis.