Home

About

Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network
FAQHELPSITE MAPCONTACT US


  Advanced Search  

 
COGNITIVE SCIENCE Vol 23 (4) 1999, pp. 615634 ISSN 0364-0213 Copyright 1999 Cognitive Science Society, Inc. All rights of reproduction in any form reserved.
 

Summary: COGNITIVE SCIENCE Vol 23 (4) 1999, pp. 615­634 ISSN 0364-0213
Copyright © 1999 Cognitive Science Society, Inc. All rights of reproduction in any form reserved.
Connectionist Sentence Processing in
Perspective
MARK STEEDMAN
University of Pennsylvania and University of Edinburgh
The emphasis in the connectionist sentence-processing literature on distrib-
uted representation and emergence of grammar from such systems can
easily obscure the often close relations between connectionist and symbolist
systems. This paper argues that the Simple Recurrent Network (SRN) models
proposed by Jordan (1989) and Elman (1990) are more directly related to
stochastic Part-of-Speech (POS) Taggers than to parsers or grammars as
such, while auto-associative memory models of the kind pioneered by
Longuet­Higgins, Willshaw, Pollack and others may be useful for grammar
induction from a network-based conceptual structure as well as for structure-
building. These observations suggest some interesting new directions for
specifically connectionist sentence processing research, including more ef-
ficient representations for finite state machines, and acquisition devices
based on a distinctively connectionist basis for grounded symbolist concep-
tual structure.

  

Source: Alishahi, Afra - Department of Computational Linguistics and Phonetics, Universität des Saarlandes

 

Collections: Computer Technologies and Information Sciences