Integrating question classification and deep learning for improved answer selection

Harish Tayyar Madabushi; Mark Lee; John Barnden

Integrating question classification and deep learning for improved answer selection

Harish Tayyar Madabushi, Mark Lee, John Barnden

Computer Science

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

20 Citations (Scopus)

94 Downloads (Pure)

Abstract

We present a system for Answer Selection that integrates fine-grained Question Classification with a Deep Learning model designed for Answer Selection. We detail the necessary changes to the Question Classification taxonomy and system, the creation of a new Entity Identification system and methods of highlighting entities to achieve this objective. Our experiments show that Question Classes are a strong signal to Deep Learning models for Answer Selection, and enable us to outperform the current state of the art in all variations of our experiments except one. In the best configuration, our MRR and MAP scores outperform the current state of the art by between 3 and 5 points on both versions of the TREC Answer Selection test set, a standard dataset for this task.

Original language	English
Title of host publication	Proceedings of the 27th International Conference on Computational Linguistics
Editors	Emily M. Bender, Leon Derczynski, Pierre Isabelle
Publisher	Association for Computational Linguistics, ACL
Pages	3283-3294
Number of pages	12
ISBN (Electronic)	9781948087506
Publication status	Published - 21 Aug 2018
Event	The 27th International Conference on Computational Linguistics - Santa Fe Community Convention Center , Santa Fe, New-Mexico, United States Duration: 20 Aug 2018 → 26 Aug 2018

Conference

Conference	The 27th International Conference on Computational Linguistics
Abbreviated title	COLING 2018
Country/Territory	United States
City	Santa Fe, New-Mexico
Period	20/08/18 → 26/08/18

Bibliographical note

Publisher Copyright:
© 2018 COLING 2018 - 27th International Conference on Computational Linguistics, Proceedings. All rights reserved.

ASJC Scopus subject areas

Language and Linguistics
Computational Theory and Mathematics
Linguistics and Language

Access to Document

Madabushi_et_al_Integrating_question_classification_COLING_2018Final published version, 176 KBLicence: Creative Commons: Attribution (CC BY)

https://aclweb.org/anthology/papers/C/C18/C18-1278/Licence: Unspecified

Cite this

Tayyar Madabushi, H., Lee, M., & Barnden, J. (2018). Integrating question classification and deep learning for improved answer selection. In E. M. Bender, L. Derczynski, & P. Isabelle (Eds.), Proceedings of the 27th International Conference on Computational Linguistics (pp. 3283-3294). Association for Computational Linguistics, ACL. https://aclweb.org/anthology/papers/C/C18/C18-1278/

@inproceedings{0a16d2f73f5246c6a614db441cdd7c15,

title = "Integrating question classification and deep learning for improved answer selection",

abstract = "We present a system for Answer Selection that integrates fine-grained Question Classification with a Deep Learning model designed for Answer Selection. We detail the necessary changes to the Question Classification taxonomy and system, the creation of a new Entity Identification system and methods of highlighting entities to achieve this objective. Our experiments show that Question Classes are a strong signal to Deep Learning models for Answer Selection, and enable us to outperform the current state of the art in all variations of our experiments except one. In the best configuration, our MRR and MAP scores outperform the current state of the art by between 3 and 5 points on both versions of the TREC Answer Selection test set, a standard dataset for this task.",

author = "{Tayyar Madabushi}, Harish and Mark Lee and John Barnden",

note = "Publisher Copyright: {\textcopyright} 2018 COLING 2018 - 27th International Conference on Computational Linguistics, Proceedings. All rights reserved.; The 27th International Conference on Computational Linguistics , COLING 2018 ; Conference date: 20-08-2018 Through 26-08-2018",

year = "2018",

month = aug,

day = "21",

language = "English",

pages = "3283--3294",

editor = "Bender, {Emily M.} and Leon Derczynski and Pierre Isabelle",

booktitle = "Proceedings of the 27th International Conference on Computational Linguistics",

publisher = "Association for Computational Linguistics, ACL",

}

Tayyar Madabushi, H, Lee, M & Barnden, J 2018, Integrating question classification and deep learning for improved answer selection. in EM Bender, L Derczynski & P Isabelle (eds), Proceedings of the 27th International Conference on Computational Linguistics. Association for Computational Linguistics, ACL, pp. 3283-3294, The 27th International Conference on Computational Linguistics , Santa Fe, New-Mexico, United States, 20/08/18. <https://aclweb.org/anthology/papers/C/C18/C18-1278/>

Integrating question classification and deep learning for improved answer selection. / Tayyar Madabushi, Harish; Lee, Mark; Barnden, John.
Proceedings of the 27th International Conference on Computational Linguistics. ed. / Emily M. Bender; Leon Derczynski; Pierre Isabelle. Association for Computational Linguistics, ACL, 2018. p. 3283-3294.

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

TY - GEN

T1 - Integrating question classification and deep learning for improved answer selection

AU - Tayyar Madabushi, Harish

AU - Lee, Mark

AU - Barnden, John

PY - 2018/8/21

Y1 - 2018/8/21

N2 - We present a system for Answer Selection that integrates fine-grained Question Classification with a Deep Learning model designed for Answer Selection. We detail the necessary changes to the Question Classification taxonomy and system, the creation of a new Entity Identification system and methods of highlighting entities to achieve this objective. Our experiments show that Question Classes are a strong signal to Deep Learning models for Answer Selection, and enable us to outperform the current state of the art in all variations of our experiments except one. In the best configuration, our MRR and MAP scores outperform the current state of the art by between 3 and 5 points on both versions of the TREC Answer Selection test set, a standard dataset for this task.

AB - We present a system for Answer Selection that integrates fine-grained Question Classification with a Deep Learning model designed for Answer Selection. We detail the necessary changes to the Question Classification taxonomy and system, the creation of a new Entity Identification system and methods of highlighting entities to achieve this objective. Our experiments show that Question Classes are a strong signal to Deep Learning models for Answer Selection, and enable us to outperform the current state of the art in all variations of our experiments except one. In the best configuration, our MRR and MAP scores outperform the current state of the art by between 3 and 5 points on both versions of the TREC Answer Selection test set, a standard dataset for this task.

UR - http://www.scopus.com/inward/record.url?scp=85119453372&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85119453372

SP - 3283

EP - 3294

BT - Proceedings of the 27th International Conference on Computational Linguistics

A2 - Bender, Emily M.

A2 - Derczynski, Leon

A2 - Isabelle, Pierre

PB - Association for Computational Linguistics, ACL

T2 - The 27th International Conference on Computational Linguistics

Y2 - 20 August 2018 through 26 August 2018

ER -

Integrating question classification and deep learning for improved answer selection

Abstract

Conference

Bibliographical note

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this