Learning to rank non-factoid answers: Comment selection in Web forums

Kateryna Tymoshenko, Daniele Bonadiman, Alessandro Moschitti

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

9 Citations (Scopus)

Abstract

Recent initiatives in IR community have shown the importance of going beyond factoid Question Answering (QA) in order to design useful real-world applications. Questions asking for descriptions or explanations are much more difficult to be solved, e.g., the machine learning models cannot focus on specific answer words or their lexical type. Thus, researchers have started to explore powerful methods for feature engineering. Two of the most promising methods are convolution tree kernels (CTKs) and convolutional neural networks (CNNs) as they have been shown to obtain high performance in the task of answer sentence selection in factoid QA. In this paper, we design state-of-the-art models for non-factoid QA also carried out on noisy data. In particular, we study and compare models for comment selection in a community QA (cQA) scenario, where the majority of questions regard descriptions or explanations. To deal with such complex task, we incorporate relational information holding between questions and comments as well as domain-specific features into both convolutional models above. Our experiments on a cQA corpus show that both CTK and CNN achieve the state of the art, also according to a direct comparison with the results obtained by the best systems of the SemEval cQA challenge.

Original languageEnglish
Title of host publicationCIKM 2016 - Proceedings of the 2016 ACM Conference on Information and Knowledge Management
PublisherAssociation for Computing Machinery
Pages2049-2052
Number of pages4
Volume24-28-October-2016
ISBN (Electronic)9781450340731
DOIs
Publication statusPublished - 24 Oct 2016
Event25th ACM International Conference on Information and Knowledge Management, CIKM 2016 - Indianapolis, United States
Duration: 24 Oct 201628 Oct 2016

Other

Other25th ACM International Conference on Information and Knowledge Management, CIKM 2016
CountryUnited States
CityIndianapolis
Period24/10/1628/10/16

Fingerprint

Learning to rank
Question answering
World Wide Web
Kernel
Convolution
Neural networks
Scenarios
Experiment
Learning model
High performance
Machine learning

ASJC Scopus subject areas

  • Business, Management and Accounting(all)
  • Decision Sciences(all)

Cite this

Tymoshenko, K., Bonadiman, D., & Moschitti, A. (2016). Learning to rank non-factoid answers: Comment selection in Web forums. In CIKM 2016 - Proceedings of the 2016 ACM Conference on Information and Knowledge Management (Vol. 24-28-October-2016, pp. 2049-2052). Association for Computing Machinery. https://doi.org/10.1145/2983323.2983906

Learning to rank non-factoid answers : Comment selection in Web forums. / Tymoshenko, Kateryna; Bonadiman, Daniele; Moschitti, Alessandro.

CIKM 2016 - Proceedings of the 2016 ACM Conference on Information and Knowledge Management. Vol. 24-28-October-2016 Association for Computing Machinery, 2016. p. 2049-2052.

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

Tymoshenko, K, Bonadiman, D & Moschitti, A 2016, Learning to rank non-factoid answers: Comment selection in Web forums. in CIKM 2016 - Proceedings of the 2016 ACM Conference on Information and Knowledge Management. vol. 24-28-October-2016, Association for Computing Machinery, pp. 2049-2052, 25th ACM International Conference on Information and Knowledge Management, CIKM 2016, Indianapolis, United States, 24/10/16. https://doi.org/10.1145/2983323.2983906
Tymoshenko K, Bonadiman D, Moschitti A. Learning to rank non-factoid answers: Comment selection in Web forums. In CIKM 2016 - Proceedings of the 2016 ACM Conference on Information and Knowledge Management. Vol. 24-28-October-2016. Association for Computing Machinery. 2016. p. 2049-2052 https://doi.org/10.1145/2983323.2983906
Tymoshenko, Kateryna ; Bonadiman, Daniele ; Moschitti, Alessandro. / Learning to rank non-factoid answers : Comment selection in Web forums. CIKM 2016 - Proceedings of the 2016 ACM Conference on Information and Knowledge Management. Vol. 24-28-October-2016 Association for Computing Machinery, 2016. pp. 2049-2052
@inproceedings{1a471bcb9aad40008763429ac96e0a36,
title = "Learning to rank non-factoid answers: Comment selection in Web forums",
abstract = "Recent initiatives in IR community have shown the importance of going beyond factoid Question Answering (QA) in order to design useful real-world applications. Questions asking for descriptions or explanations are much more difficult to be solved, e.g., the machine learning models cannot focus on specific answer words or their lexical type. Thus, researchers have started to explore powerful methods for feature engineering. Two of the most promising methods are convolution tree kernels (CTKs) and convolutional neural networks (CNNs) as they have been shown to obtain high performance in the task of answer sentence selection in factoid QA. In this paper, we design state-of-the-art models for non-factoid QA also carried out on noisy data. In particular, we study and compare models for comment selection in a community QA (cQA) scenario, where the majority of questions regard descriptions or explanations. To deal with such complex task, we incorporate relational information holding between questions and comments as well as domain-specific features into both convolutional models above. Our experiments on a cQA corpus show that both CTK and CNN achieve the state of the art, also according to a direct comparison with the results obtained by the best systems of the SemEval cQA challenge.",
author = "Kateryna Tymoshenko and Daniele Bonadiman and Alessandro Moschitti",
year = "2016",
month = "10",
day = "24",
doi = "10.1145/2983323.2983906",
language = "English",
volume = "24-28-October-2016",
pages = "2049--2052",
booktitle = "CIKM 2016 - Proceedings of the 2016 ACM Conference on Information and Knowledge Management",
publisher = "Association for Computing Machinery",

}

TY - GEN

T1 - Learning to rank non-factoid answers

T2 - Comment selection in Web forums

AU - Tymoshenko, Kateryna

AU - Bonadiman, Daniele

AU - Moschitti, Alessandro

PY - 2016/10/24

Y1 - 2016/10/24

N2 - Recent initiatives in IR community have shown the importance of going beyond factoid Question Answering (QA) in order to design useful real-world applications. Questions asking for descriptions or explanations are much more difficult to be solved, e.g., the machine learning models cannot focus on specific answer words or their lexical type. Thus, researchers have started to explore powerful methods for feature engineering. Two of the most promising methods are convolution tree kernels (CTKs) and convolutional neural networks (CNNs) as they have been shown to obtain high performance in the task of answer sentence selection in factoid QA. In this paper, we design state-of-the-art models for non-factoid QA also carried out on noisy data. In particular, we study and compare models for comment selection in a community QA (cQA) scenario, where the majority of questions regard descriptions or explanations. To deal with such complex task, we incorporate relational information holding between questions and comments as well as domain-specific features into both convolutional models above. Our experiments on a cQA corpus show that both CTK and CNN achieve the state of the art, also according to a direct comparison with the results obtained by the best systems of the SemEval cQA challenge.

AB - Recent initiatives in IR community have shown the importance of going beyond factoid Question Answering (QA) in order to design useful real-world applications. Questions asking for descriptions or explanations are much more difficult to be solved, e.g., the machine learning models cannot focus on specific answer words or their lexical type. Thus, researchers have started to explore powerful methods for feature engineering. Two of the most promising methods are convolution tree kernels (CTKs) and convolutional neural networks (CNNs) as they have been shown to obtain high performance in the task of answer sentence selection in factoid QA. In this paper, we design state-of-the-art models for non-factoid QA also carried out on noisy data. In particular, we study and compare models for comment selection in a community QA (cQA) scenario, where the majority of questions regard descriptions or explanations. To deal with such complex task, we incorporate relational information holding between questions and comments as well as domain-specific features into both convolutional models above. Our experiments on a cQA corpus show that both CTK and CNN achieve the state of the art, also according to a direct comparison with the results obtained by the best systems of the SemEval cQA challenge.

UR - http://www.scopus.com/inward/record.url?scp=84996565478&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84996565478&partnerID=8YFLogxK

U2 - 10.1145/2983323.2983906

DO - 10.1145/2983323.2983906

M3 - Conference contribution

AN - SCOPUS:84996565478

VL - 24-28-October-2016

SP - 2049

EP - 2052

BT - CIKM 2016 - Proceedings of the 2016 ACM Conference on Information and Knowledge Management

PB - Association for Computing Machinery

ER -