Baby SRL

Modeling early language acquisition

Michael Connor, Yael Gertner, Cynthia L Fisher, Dan Roth

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

A fundamental task in sentence comprehension is to assign semantic roles to sentence constituents. The structure-mapping account proposes that children start with a shallow structural analysis of sentences: children treat the number of nouns in the sentence as a cue to its semantic predicateargument structure, and represent language experience in an abstract format that permits rapid generalization to new verbs. In this paper, we tested the consequences of these representational assumptions via experiments with a system for automatic semantic role labeling (SRL), trained on a sample of child-directed speech. When the SRL was presented with representations of sentence structure consisting simply of an ordered set of nouns, it mimicked experimental findings with toddlers, including a striking error found in children. Adding features representing the position of the verb increased accuracy and eliminated the error. We show the SRL system can use incremental knowledge gain to switch from error-prone noun order features to a more accurate representation, demonstrating a possible mechanism for this process in child development.

Original languageEnglish (US)
Title of host publicationCoNLL 2008 - Proceedings of the Twelfth Conference on Computational Natural Language Learning
Pages81-88
Number of pages8
StatePublished - Dec 1 2008
Event12th Conference on Computational Natural Language Learning, CoNLL 2008 - Manchester, United Kingdom
Duration: Aug 16 2008Aug 17 2008

Publication series

NameCoNLL 2008 - Proceedings of the Twelfth Conference on Computational Natural Language Learning

Other

Other12th Conference on Computational Natural Language Learning, CoNLL 2008
CountryUnited Kingdom
CityManchester
Period8/16/088/17/08

Fingerprint

language acquisition
baby
Labeling
Semantics
semantics
structural analysis
Structural analysis
comprehension
Switches
experiment
language
experience
Experiments

ASJC Scopus subject areas

  • Artificial Intelligence
  • Human-Computer Interaction
  • Linguistics and Language

Cite this

Connor, M., Gertner, Y., Fisher, C. L., & Roth, D. (2008). Baby SRL: Modeling early language acquisition. In CoNLL 2008 - Proceedings of the Twelfth Conference on Computational Natural Language Learning (pp. 81-88). (CoNLL 2008 - Proceedings of the Twelfth Conference on Computational Natural Language Learning).

Baby SRL : Modeling early language acquisition. / Connor, Michael; Gertner, Yael; Fisher, Cynthia L; Roth, Dan.

CoNLL 2008 - Proceedings of the Twelfth Conference on Computational Natural Language Learning. 2008. p. 81-88 (CoNLL 2008 - Proceedings of the Twelfth Conference on Computational Natural Language Learning).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Connor, M, Gertner, Y, Fisher, CL & Roth, D 2008, Baby SRL: Modeling early language acquisition. in CoNLL 2008 - Proceedings of the Twelfth Conference on Computational Natural Language Learning. CoNLL 2008 - Proceedings of the Twelfth Conference on Computational Natural Language Learning, pp. 81-88, 12th Conference on Computational Natural Language Learning, CoNLL 2008, Manchester, United Kingdom, 8/16/08.
Connor M, Gertner Y, Fisher CL, Roth D. Baby SRL: Modeling early language acquisition. In CoNLL 2008 - Proceedings of the Twelfth Conference on Computational Natural Language Learning. 2008. p. 81-88. (CoNLL 2008 - Proceedings of the Twelfth Conference on Computational Natural Language Learning).
Connor, Michael ; Gertner, Yael ; Fisher, Cynthia L ; Roth, Dan. / Baby SRL : Modeling early language acquisition. CoNLL 2008 - Proceedings of the Twelfth Conference on Computational Natural Language Learning. 2008. pp. 81-88 (CoNLL 2008 - Proceedings of the Twelfth Conference on Computational Natural Language Learning).
@inproceedings{6383bbf262ac4fc1a86e935bc46eafa8,
title = "Baby SRL: Modeling early language acquisition",
abstract = "A fundamental task in sentence comprehension is to assign semantic roles to sentence constituents. The structure-mapping account proposes that children start with a shallow structural analysis of sentences: children treat the number of nouns in the sentence as a cue to its semantic predicateargument structure, and represent language experience in an abstract format that permits rapid generalization to new verbs. In this paper, we tested the consequences of these representational assumptions via experiments with a system for automatic semantic role labeling (SRL), trained on a sample of child-directed speech. When the SRL was presented with representations of sentence structure consisting simply of an ordered set of nouns, it mimicked experimental findings with toddlers, including a striking error found in children. Adding features representing the position of the verb increased accuracy and eliminated the error. We show the SRL system can use incremental knowledge gain to switch from error-prone noun order features to a more accurate representation, demonstrating a possible mechanism for this process in child development.",
author = "Michael Connor and Yael Gertner and Fisher, {Cynthia L} and Dan Roth",
year = "2008",
month = "12",
day = "1",
language = "English (US)",
isbn = "1905593481",
series = "CoNLL 2008 - Proceedings of the Twelfth Conference on Computational Natural Language Learning",
pages = "81--88",
booktitle = "CoNLL 2008 - Proceedings of the Twelfth Conference on Computational Natural Language Learning",

}

TY - GEN

T1 - Baby SRL

T2 - Modeling early language acquisition

AU - Connor, Michael

AU - Gertner, Yael

AU - Fisher, Cynthia L

AU - Roth, Dan

PY - 2008/12/1

Y1 - 2008/12/1

N2 - A fundamental task in sentence comprehension is to assign semantic roles to sentence constituents. The structure-mapping account proposes that children start with a shallow structural analysis of sentences: children treat the number of nouns in the sentence as a cue to its semantic predicateargument structure, and represent language experience in an abstract format that permits rapid generalization to new verbs. In this paper, we tested the consequences of these representational assumptions via experiments with a system for automatic semantic role labeling (SRL), trained on a sample of child-directed speech. When the SRL was presented with representations of sentence structure consisting simply of an ordered set of nouns, it mimicked experimental findings with toddlers, including a striking error found in children. Adding features representing the position of the verb increased accuracy and eliminated the error. We show the SRL system can use incremental knowledge gain to switch from error-prone noun order features to a more accurate representation, demonstrating a possible mechanism for this process in child development.

AB - A fundamental task in sentence comprehension is to assign semantic roles to sentence constituents. The structure-mapping account proposes that children start with a shallow structural analysis of sentences: children treat the number of nouns in the sentence as a cue to its semantic predicateargument structure, and represent language experience in an abstract format that permits rapid generalization to new verbs. In this paper, we tested the consequences of these representational assumptions via experiments with a system for automatic semantic role labeling (SRL), trained on a sample of child-directed speech. When the SRL was presented with representations of sentence structure consisting simply of an ordered set of nouns, it mimicked experimental findings with toddlers, including a striking error found in children. Adding features representing the position of the verb increased accuracy and eliminated the error. We show the SRL system can use incremental knowledge gain to switch from error-prone noun order features to a more accurate representation, demonstrating a possible mechanism for this process in child development.

UR - http://www.scopus.com/inward/record.url?scp=79960119964&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79960119964&partnerID=8YFLogxK

M3 - Conference contribution

SN - 1905593481

SN - 9781905593484

T3 - CoNLL 2008 - Proceedings of the Twelfth Conference on Computational Natural Language Learning

SP - 81

EP - 88

BT - CoNLL 2008 - Proceedings of the Twelfth Conference on Computational Natural Language Learning

ER -