Pt. I. Semantics --
pt. II. Syntax --
pt. III. Representational adequacy --
pt. IV. Computational psycholinguistics. Cover
Half Title
Title Page
Copyright Page
Original Title Page
Original Copyright Page
Dedication
Table of Contents
List of Contributors
Preface
1 Connectionist Natural Language Processing
Introduction
Overview of chapters
Acknowledgements
References
PART I SEMANTICS
Introduction
2 Distributed Symbol Discovery through Symbol Recirculation: Toward Natural Language Processing in Distributed Connectionist Networks
Introduction
Natural language processing: Constraints from the task domain
Dynamic vs. static symbol representations
Symbol recirculation. Encoding semantic networks in DUAL: A distributed connectionist architectureOther symbol recirculation methods
Open problems
Variable binding research and symbol formation
Summary and conclusions
Acknowledgements
References
3 Representing Meaning Using Microfeatures
Introduction
Microfeature representations in PARROT
Implementation of the microfeature concept within PARROT
Examples
Discussion and next steps
Acknowledgements
References
Appendix I: Outline of the PARROT system
Appendix II: Example entries from the lexicon
4 Noun Phrase Analysis with Connectionist Networks. IntroductionThe domain
Learning level: Learning semantic prepositional relationships
Integration level: Integration of semantic and syntactic constraints
A case study for the disambiguation of noun phrases
Discussion
Conclusion
Acknowledgements
References
5 Parallel Constraint Satisfaction as a Comprehension Mechanism
Introduction
Sentence comprehension
Story comprehension
Conclusions
Acknowledgements
References
Appendix I: Input and output representations
PART II SYNTAX
Introduction
References
6 Self-correcting Connectionist Parsing
Introduction: Constrained chaos. AgreementCounting
Constituent motion
Missing constituents
Conclusions
Acknowledgements
References
7 A Net-linguistic "Earley" Parser
Introduction
The basic characteristics of the parser
The representation of parse-information
The Earley parse-list algorithm
Our approach
References
Appendix
PART III REPRESENTATIONAL ADEQUACY
Introduction
Reference
8 The Demons and the Beast-Modular and Nodular Kinds of Knowledge
Introduction and summary
Structure and habits-The knowledge and the power
A model that learns some morphology. Evidence for nodes unseen-some models that learn to read aloudThe study of statistically available information
Conclusion-Behavioural strategies and mental structures
Acknowledgements
References
9 Representational Adequacy and the Case for a Hybrid Connectionist/Marker-parsing Model
Introduction
Representational adequacy
Autonomous semantic networks
ASNs and representational adequacy
Discussion
Conclusion
Acknowledgements
References
10 A Step Toward Sub-symbolic Language Models without Linguistic Representations
Introduction
Basic observations about language
An implementation.