Flexible Parsing.

Abstract

When people use language spontaneously, they often do not adhere strictly to commonly accepted standards of grammaticality. The primary objective of this project is to develop flexible computer parsing techniques which can deal with the various kinds of ungrammaticalities that arise, both on the lexical and the phrase level. The progress towards this goal covered by this report includes: 1) The initial development of the FlexP flexible parser based on pattern-matching techniques; 2) Review of the initial design choices for FlexP in the light of this evaluation, leading to the formulation of the construction-specific approach tp parsing, and its preliminary evaluation for applied natural language processing through the experimental parsers CASPAR and DYPAR; 3) Application of the construction-specific approach to flexible parsing to the parsing of an artificial command language in the parser for the Cousin command interface, a graceful interface for the Unix operating system; 4) Investigation of control structures that would allow the integration of multiple diverse parsing strategies into a single parsing system in an extensible manner; 5) Development of a taxonomy of grammatical deviations and recovery strategies for dealing with them; 6) Design and implementation of an initial version of MULTIPAR, the large-scale robust restricted-domain parser mentioned above that employs multiple construction-specific parsing strategies; and 7) Application of the flexible parsing techniques developed under previous parts of the contract to speech input.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jun 30, 1986
Accession Number
ADA185595

Entities

People

  • Philip J. Hayes

Organizations

  • Carnegie Mellon University

Tags

Communities of Interest

  • Cyber
  • Materials and Manufacturing Processes
  • Weapons Technologies

DTIC Thesaurus Topics

  • Artificial Intelligence
  • Automated Speech Recognition
  • Classification
  • Cognitive Science
  • Computational Linguistics
  • Computer Science
  • Computers
  • Databases
  • Grammars
  • Identification
  • Language
  • Linguistics
  • Natural Language Processing
  • Notation
  • Operating Systems
  • Standards
  • User Interface

Readers

  • Computational Linguistics
  • Systems Analysis and Design

Technology Areas

  • AI & ML
  • AI & ML - Machine Translation