The Restriction Language for Computer Grammars of Natural Language

Over the past few years, a number of systems
for the computer analysis of natural language 
sentences have been based on augmented context-free
grammars: a context-free grammar which defines a 
set of parse trees for a sentence, plus a group of restrictions
to which a tree must conform in order 
to be a valid sentence analysis.  As the coverage of the
grammar is increased, an efficient representation 
becomes essential for further development.  This paper
presents a programming language designed specifically 
for the compact and perspicuous statement of restrictions
of a natural language grammar.  It is based 
on ten years' experience parsing text sentences with
the comprehensive English grammar of the N. 
Linguistic String Project, and embodies in its syntax
and routines the relations which were found to 
be useful and adequate for computerized natural language
analysis.  The language is used in the current 
implementation of the Linguistic String Parser.

CACM July 1975

Sager, N.
Grishman, R.

natural language, parsing, grammar, programming languages

3.42 3.79 4.22

CA750703 JB January 9, 1978  10:17 AM

2581	4	2739
2739	4	2739
2795	4	2739
2921	4	2739
3087	4	2739
1989	5	2739
2739	5	2739
2739	5	2739
2739	5	2739