Parse fitting and prose fixing: getting a hold on ill-formedness

  • Authors:
  • K. Jensen;G. E. Heidorn;L. A. Miller;Y. Ravin

  • Affiliations:
  • IBM Thomas J. Watson Research Center, Yorktown Heights, New York;IBM Thomas J. Watson Research Center, Yorktown Heights, New York;IBM Thomas J. Watson Research Center, Yorktown Heights, New York;IBM Thomas J. Watson Research Center, Yorktown Heights, New York

  • Venue:
  • Computational Linguistics - Special issue on ill-formed input
  • Year:
  • 1983

Quantified Score

Hi-index 0.02

Visualization

Abstract

Processing syntactically ill-formed language is an important mission of the EPISTLE system, Ill-formed input is treated by this system in various ways. Misspellings are highlighted by a standard spelling checker; syntactic errors are detected and corrections are suggested; and stylistic infelicities are called to the user's attention.Central to the EPISTLE processing strategy is its technique of fitted parsing. When the rules of a conventional syntactic grammar are unable to produce a parse for an input string, this technique can be used to produce a reasonable approximate parse that can serve as input to the remaining stages of processing.This paper first describes the fitting process and gives examples of ill-formed language situations where it is called into play. We then show how a fitted parse allows EPISTLE to carry on its text-critiquing mission where conventional grammars would fail either because of input problems or because of limitations in the grammars themselves. Some inherent difficulties of the fitting technique are also discussed. In addition, we explore how style critiquing relates to the handling of ill-formed input, and how a fitted parse can be used in style checking.