MetaCart Sign in to MyCiteSeerX

Include Citations | Advanced Search | Help

Disambiguated Search | Include Citations | Advanced Search | Help

Learning the Past Tense of English Verbs: The Symbolic Pattern Associator vs. Connectionist Models (1994) [32 citations — 2 self]

by Charles Ling ,  Charles X. Ling
Journal of Artificial Intelligence Research
Add To MetaCart

Abstract:

Learning the past tense of English verbs --- a seemingly minor aspect of language acquisition --- has generated heated debates since 1986, and has become a landmark task for testing the adequacy of cognitive modeling. Several artificial neural networks (ANNs) have been implemented, and a challenge for better symbolic models has been posed. In this paper, we present a general-purpose Symbolic Pattern Associator (SPA) based upon the decision-tree learning algorithm ID3. We conduct extensive head-to-head comparisons on the generalization ability between ANN models and the SPA under different representations. We conclude that the SPA generalizes the past tense of unseen verbs better than ANN models by a wide margin, and we offer insights as to why this should be the case. We also discuss a new default strategy for decision-tree learning algorithms. 1. Introduction Learning the past tense of English verbs, a seemingly minor aspect of language acquisition, has generated heated debates since...

Citations

3356 C4.5: Programs for Machine Learning – Quinlan - 1993
2526 Induction of decision trees – Quinlan - 1986
508 Neural networks and the bias/variance dilemma – Geman, Bienenstock, et al. - 1992
429 Connectionism and cognitive architecture: A critical analysis – Fodor - 1988
257 On learning the past tense of English verbs – Rumelhart - 1986
189 On language and connectionism: Analysis of a parallel distributed processing model of language acquisition – Pinker, Prince - 1988
97 Rules of language – Pinker - 1991
92 Symbolic and neural learning algorithms: An experimental comparison – Shavlik, Mooney, et al. - 1991
85 U-shaped learning and frequency effects in a multi-layered perceptron: Implications for child language acquisition – Plunkett, Marchman - 1991
78 Statistical Aspects of Neural Networks – Ripley - 1993
74 The CHILDES project: Tools for analyzing talk – MacWhinney - 2000
68 Error-correcting output codes: a general method for improving multiclass inductive learning programs – Dietterich, Bakiri - 1991
52 Generalizations of regular and irregular morphological patterns – Prasada - 1993
47 A Comparative Study of ID3 and Backpropagation for English Text-to-speech Mapping – Dietterich, Hild, et al. - 1990
38 Answering the connectionist challenge: A symbolic model of learning the past tense of English verbs – Ling - 1993
35 The relation between linguistic structure and associative theories of language learning: A constructive critique of some connectionist learning models – Lachter, Bever - 1988
10 Implementations are not conceptualizations: Revising the verb model – MacWhinney, Leinbach - 1991
6 Connections and symbols: closing the gap – MacWhinney - 1993
3 Beyond rules and exceptions: A connectionist modeling approach to inflectional morphology – Daugherty - 1993
3 U-shaped learning and frequency e ects in a multilayered perceptron: Implications for child language acquisition – Plunkett - 1991
2 Using a recurrent net to learn the past tense – Cottrell - 1991
1 Comparison of symbolic, statistical and neural network classifiers – Feng, King, et al. - 1992
1 Learning the Past Tense: Symbolic vs Connectionist Models – Ling, Cherwenka - 1993
View or Download | Add to My Collection | Correct Errors