Extracting commonsense properties from embeddings with limited human guidance

Yiben Yang; Lawrence A Birnbaum; Jiping Wang; Douglas C Downey

doi:10.18653/v1/p18-2102

Extracting commonsense properties from embeddings with limited human guidance

Yiben Yang, Lawrence A Birnbaum, Jiping Wang, Douglas C Downey

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

11 Scopus citations

Abstract

Intelligent systems require common sense, but automatically extracting this knowledge from text can be difficult. We propose and assess methods for extracting one type of commonsense knowledge, object-property comparisons, from pre-trained embeddings. In experiments, we show that our approach exceeds the accuracy of previous work but requires substantially less hand-annotated knowledge. Further, we show that an active learning approach that synthesizes common-sense queries can boost accuracy.

Original language	English (US)
Title of host publication	ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Short Papers)
Publisher	Association for Computational Linguistics (ACL)
Pages	644-649
Number of pages	6
ISBN (Electronic)	9781948087346
DOIs	https://doi.org/10.18653/v1/p18-2102
State	Published - 2018
Event	56th Annual Meeting of the Association for Computational Linguistics, ACL 2018 - Melbourne, Australia Duration: Jul 15 2018 → Jul 20 2018

Publication series

Name	ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers)
Volume	2

Conference

Conference	56th Annual Meeting of the Association for Computational Linguistics, ACL 2018
Country/Territory	Australia
City	Melbourne
Period	7/15/18 → 7/20/18

ASJC Scopus subject areas

Software
Computational Theory and Mathematics

Access to Document

10.18653/v1/p18-2102

Cite this

Yang, Y., Birnbaum, L. A., Wang, J., & Downey, D. C. (2018). Extracting commonsense properties from embeddings with limited human guidance. In ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Short Papers) (pp. 644-649). (ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers); Vol. 2). Association for Computational Linguistics (ACL). https://doi.org/10.18653/v1/p18-2102

Yang, Yiben ; Birnbaum, Lawrence A ; Wang, Jiping et al. / Extracting commonsense properties from embeddings with limited human guidance. ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Short Papers). Association for Computational Linguistics (ACL), 2018. pp. 644-649 (ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers)).

@inproceedings{6fc1d7a0d621480aa365329aa06b4b25,

title = "Extracting commonsense properties from embeddings with limited human guidance",

abstract = "Intelligent systems require common sense, but automatically extracting this knowledge from text can be difficult. We propose and assess methods for extracting one type of commonsense knowledge, object-property comparisons, from pre-trained embeddings. In experiments, we show that our approach exceeds the accuracy of previous work but requires substantially less hand-annotated knowledge. Further, we show that an active learning approach that synthesizes common-sense queries can boost accuracy.",

author = "Yiben Yang and Birnbaum, {Lawrence A} and Jiping Wang and Downey, {Douglas C}",

note = "Funding Information: This work was supported in part by NSF Grant IIS-1351029. We thank the anonymous reviewers for helpful comments. Publisher Copyright: {\textcopyright} 2018 Association for Computational Linguistics; 56th Annual Meeting of the Association for Computational Linguistics, ACL 2018 ; Conference date: 15-07-2018 Through 20-07-2018",

year = "2018",

doi = "10.18653/v1/p18-2102",

language = "English (US)",

series = "ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers)",

publisher = "Association for Computational Linguistics (ACL)",

pages = "644--649",

booktitle = "ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Short Papers)",

}

Yang, Y, Birnbaum, LA , Wang, J & Downey, DC 2018, Extracting commonsense properties from embeddings with limited human guidance. in ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Short Papers). ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers), vol. 2, Association for Computational Linguistics (ACL), pp. 644-649, 56th Annual Meeting of the Association for Computational Linguistics, ACL 2018, Melbourne, Australia, 7/15/18. https://doi.org/10.18653/v1/p18-2102

Extracting commonsense properties from embeddings with limited human guidance. / Yang, Yiben; Birnbaum, Lawrence A ; Wang, Jiping et al.

ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Short Papers). Association for Computational Linguistics (ACL), 2018. p. 644-649 (ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers); Vol. 2).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Extracting commonsense properties from embeddings with limited human guidance

AU - Yang, Yiben

AU - Birnbaum, Lawrence A

AU - Wang, Jiping

AU - Downey, Douglas C

N1 - Funding Information: This work was supported in part by NSF Grant IIS-1351029. We thank the anonymous reviewers for helpful comments. Publisher Copyright: © 2018 Association for Computational Linguistics

PY - 2018

Y1 - 2018

N2 - Intelligent systems require common sense, but automatically extracting this knowledge from text can be difficult. We propose and assess methods for extracting one type of commonsense knowledge, object-property comparisons, from pre-trained embeddings. In experiments, we show that our approach exceeds the accuracy of previous work but requires substantially less hand-annotated knowledge. Further, we show that an active learning approach that synthesizes common-sense queries can boost accuracy.

AB - Intelligent systems require common sense, but automatically extracting this knowledge from text can be difficult. We propose and assess methods for extracting one type of commonsense knowledge, object-property comparisons, from pre-trained embeddings. In experiments, we show that our approach exceeds the accuracy of previous work but requires substantially less hand-annotated knowledge. Further, we show that an active learning approach that synthesizes common-sense queries can boost accuracy.

UR - http://www.scopus.com/inward/record.url?scp=85063155035&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85063155035&partnerID=8YFLogxK

U2 - 10.18653/v1/p18-2102

DO - 10.18653/v1/p18-2102

M3 - Conference contribution

AN - SCOPUS:85063155035

T3 - ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers)

SP - 644

EP - 649

BT - ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Short Papers)

PB - Association for Computational Linguistics (ACL)

T2 - 56th Annual Meeting of the Association for Computational Linguistics, ACL 2018

Y2 - 15 July 2018 through 20 July 2018

ER -

Yang Y, Birnbaum LA , Wang J , Downey DC. Extracting commonsense properties from embeddings with limited human guidance. In ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Short Papers). Association for Computational Linguistics (ACL). 2018. p. 644-649. (ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers)). doi: 10.18653/v1/p18-2102

Extracting commonsense properties from embeddings with limited human guidance

Abstract

Publication series

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this