KEViN: A Knowledge Enhanced Validity and Novelty Classifier for Arguments

Ameer Saadat-Yazdi; Xue Li; Sandrine Chausson; Vaishak Belle; Björn Ross; Jeff Z Pan; Nadin Kokciyan

KEViN: A Knowledge Enhanced Validity and Novelty Classifier for Arguments

Ameer Saadat-Yazdi, Xue Li^*, Sandrine Chausson, Vaishak Belle, Björn Ross, Jeff Z Pan, Nadin Kokciyan

^*Corresponding author for this work

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

Abstract / Description of output

The ArgMining 2022 Shared Task is concerned with predicting the validity and novelty of an inference for a given premise and conclusion pair. We propose two feed-forward network based models (KEViN1 and KEViN2), which combine features generated from several pretrained transformers and the WikiData knowledge graph. The transformers are used to predict entailment and semantic similarity, while WikiData is used to provide a semantic measure between concepts in the premise-conclusion pair. Our proposed models show significant improvement over RoBERTa, with KEViN1 outperforming KEViN2 and obtaining second rank on both subtasks (A and B) of the ArgMining 2022 Shared Task.

Original language	English
Title of host publication	Proceedings of 9th Workshop on Argument Mining
Editors	Gabriella Lapesa, Jodi Schneider, Yohan Jo, Sougata Saha
Publisher	International Conference on Computational Linguistics
Pages	104-110
Number of pages	7
Publication status	Published - 12 Oct 2022
Event	The 9th Workshop on Argument Mining, 2022 - Gyeongju, Korea, Republic of Duration: 12 Oct 2022 → 17 Oct 2022 Conference number: 9

Publication series

Name	International Conference on Computational Linguistics - Proceedings of 9th Workshop on Argument Mining
Publisher	International Conference on Computational Linguistics
Number	14
Volume	29
ISSN (Electronic)	2591-2093

Workshop

Workshop	The 9th Workshop on Argument Mining, 2022
Abbreviated title	ArgMining 2022
Country/Territory	Korea, Republic of
City	Gyeongju
Period	12/10/22 → 17/10/22

Access to Document

KEViN_SAADAT-YAZDI_DOA05092022_VOR_CC-BYFinal published version, 435 KBLicence: Creative Commons: Attribution (CC-BY)

https://aclanthology.org/2022.argmining-1.9/Licence: Creative Commons: Attribution (CC-BY)

Argument Mining for Argumentation-based Decision Making
Kokciyan, N. & Saadat-Yazdi, A.
Non-EU industry, commerce and public corporations
1/10/21 → 30/04/25
Project: Research

Cite this

Saadat-Yazdi, A., Li, X., Chausson, S., Belle, V., Ross, B., Pan, J. Z., & Kokciyan, N. (2022). KEViN: A Knowledge Enhanced Validity and Novelty Classifier for Arguments. In G. Lapesa, J. Schneider, Y. Jo, & S. Saha (Eds.), Proceedings of 9th Workshop on Argument Mining (pp. 104-110). (International Conference on Computational Linguistics - Proceedings of 9th Workshop on Argument Mining; Vol. 29, No. 14). International Conference on Computational Linguistics. https://aclanthology.org/2022.argmining-1.9/

Saadat-Yazdi, Ameer ; Li, Xue ; Chausson, Sandrine et al. / KEViN: A Knowledge Enhanced Validity and Novelty Classifier for Arguments. Proceedings of 9th Workshop on Argument Mining. editor / Gabriella Lapesa ; Jodi Schneider ; Yohan Jo ; Sougata Saha. International Conference on Computational Linguistics, 2022. pp. 104-110 (International Conference on Computational Linguistics - Proceedings of 9th Workshop on Argument Mining; 14).

@inproceedings{d5c4c7f11a414bc0ae08e47f28a16faf,

title = "KEViN: A Knowledge Enhanced Validity and Novelty Classifier for Arguments",

abstract = "The ArgMining 2022 Shared Task is concerned with predicting the validity and novelty of an inference for a given premise and conclusion pair. We propose two feed-forward network based models (KEViN1 and KEViN2), which combine features generated from several pretrained transformers and the WikiData knowledge graph. The transformers are used to predict entailment and semantic similarity, while WikiData is used to provide a semantic measure between concepts in the premise-conclusion pair. Our proposed models show significant improvement over RoBERTa, with KEViN1 outperforming KEViN2 and obtaining second rank on both subtasks (A and B) of the ArgMining 2022 Shared Task.",

author = "Ameer Saadat-Yazdi and Xue Li and Sandrine Chausson and Vaishak Belle and Bj{\"o}rn Ross and Pan, {Jeff Z} and Nadin Kokciyan",

year = "2022",

month = oct,

day = "12",

language = "English",

series = "International Conference on Computational Linguistics - Proceedings of 9th Workshop on Argument Mining",

publisher = "International Conference on Computational Linguistics",

number = "14",

pages = "104--110",

editor = "Gabriella Lapesa and Jodi Schneider and Yohan Jo and Sougata Saha",

booktitle = "Proceedings of 9th Workshop on Argument Mining",

note = "The 9th Workshop on Argument Mining, 2022, ArgMining 2022 ; Conference date: 12-10-2022 Through 17-10-2022",

}

Saadat-Yazdi, A, Li, X, Chausson, S, Belle, V , Ross, B , Pan, JZ & Kokciyan, N 2022, KEViN: A Knowledge Enhanced Validity and Novelty Classifier for Arguments. in G Lapesa, J Schneider, Y Jo & S Saha (eds), Proceedings of 9th Workshop on Argument Mining. International Conference on Computational Linguistics - Proceedings of 9th Workshop on Argument Mining, no. 14, vol. 29, International Conference on Computational Linguistics, pp. 104-110, The 9th Workshop on Argument Mining, 2022, Gyeongju, Korea, Republic of, 12/10/22. <https://aclanthology.org/2022.argmining-1.9/>

KEViN: A Knowledge Enhanced Validity and Novelty Classifier for Arguments. / Saadat-Yazdi, Ameer; Li, Xue; Chausson, Sandrine et al.
Proceedings of 9th Workshop on Argument Mining. ed. / Gabriella Lapesa; Jodi Schneider; Yohan Jo; Sougata Saha. International Conference on Computational Linguistics, 2022. p. 104-110 (International Conference on Computational Linguistics - Proceedings of 9th Workshop on Argument Mining; Vol. 29, No. 14).

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

TY - GEN

T1 - KEViN: A Knowledge Enhanced Validity and Novelty Classifier for Arguments

AU - Saadat-Yazdi, Ameer

AU - Li, Xue

AU - Chausson, Sandrine

AU - Belle, Vaishak

AU - Ross, Björn

AU - Pan, Jeff Z

AU - Kokciyan, Nadin

N1 - Conference code: 9

PY - 2022/10/12

Y1 - 2022/10/12

N2 - The ArgMining 2022 Shared Task is concerned with predicting the validity and novelty of an inference for a given premise and conclusion pair. We propose two feed-forward network based models (KEViN1 and KEViN2), which combine features generated from several pretrained transformers and the WikiData knowledge graph. The transformers are used to predict entailment and semantic similarity, while WikiData is used to provide a semantic measure between concepts in the premise-conclusion pair. Our proposed models show significant improvement over RoBERTa, with KEViN1 outperforming KEViN2 and obtaining second rank on both subtasks (A and B) of the ArgMining 2022 Shared Task.

AB - The ArgMining 2022 Shared Task is concerned with predicting the validity and novelty of an inference for a given premise and conclusion pair. We propose two feed-forward network based models (KEViN1 and KEViN2), which combine features generated from several pretrained transformers and the WikiData knowledge graph. The transformers are used to predict entailment and semantic similarity, while WikiData is used to provide a semantic measure between concepts in the premise-conclusion pair. Our proposed models show significant improvement over RoBERTa, with KEViN1 outperforming KEViN2 and obtaining second rank on both subtasks (A and B) of the ArgMining 2022 Shared Task.

M3 - Conference contribution

T3 - International Conference on Computational Linguistics - Proceedings of 9th Workshop on Argument Mining

SP - 104

EP - 110

BT - Proceedings of 9th Workshop on Argument Mining

A2 - Lapesa, Gabriella

A2 - Schneider, Jodi

A2 - Jo, Yohan

A2 - Saha, Sougata

PB - International Conference on Computational Linguistics

T2 - The 9th Workshop on Argument Mining, 2022

Y2 - 12 October 2022 through 17 October 2022

ER -

Saadat-Yazdi A, Li X, Chausson S, Belle V , Ross B , Pan JZ et al. KEViN: A Knowledge Enhanced Validity and Novelty Classifier for Arguments. In Lapesa G, Schneider J, Jo Y, Saha S, editors, Proceedings of 9th Workshop on Argument Mining. International Conference on Computational Linguistics. 2022. p. 104-110. (International Conference on Computational Linguistics - Proceedings of 9th Workshop on Argument Mining; 14).

KEViN: A Knowledge Enhanced Validity and Novelty Classifier for Arguments

Abstract / Description of output

Publication series

Workshop

Access to Document

Fingerprint

Projects

Argument Mining for Argumentation-based Decision Making

Cite this