Knowledge Memory Based LSTM Model for Answer Selection

Weijie An; Qin Chen; Yan Yang; Liang He

doi:10.1007/978-3-319-70096-0_4

Knowledge Memory Based LSTM Model for Answer Selection

Weijie An, Qin Chen, Yan Yang, Liang He

School of Computer Science and Technology

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

2 Scopus citations

Abstract

Recurrent neural networks (RNN) have shown great success in answer selection task in recent years. Although the attention mechanism has been widely used to enhance the information interaction between questions and answers, knowledge is still the gap between their representations. In this paper, we propose a knowledge memory based RNN model, which incorporates the knowledge learned from the data sets into the question representations. Experiments on two benchmark data sets show the great advantages of our proposed model over that without the knowledge memory. Furthermore, our model outperforms most of the recent progress in question answering.

Original language	English
Title of host publication	Neural Information Processing - 24th International Conference, ICONIP 2017, Proceedings
Editors	Dongbin Zhao, El-Sayed M. El-Alfy, Derong Liu, Shengli Xie, Yuanqing Li
Publisher	Springer Verlag
Pages	34-42
Number of pages	9
ISBN (Print)	9783319700953
DOIs	https://doi.org/10.1007/978-3-319-70096-0_4
State	Published - 2017
Event	24th International Conference on Neural Information Processing, ICONIP 2017 - Guangzhou, China Duration: 14 Nov 2017 → 18 Nov 2017

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	10635 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	24th International Conference on Neural Information Processing, ICONIP 2017
Country/Territory	China
City	Guangzhou
Period	14/11/17 → 18/11/17

Keywords

Answer selection
Deep learning
Knowledge memory

Access to Document

10.1007/978-3-319-70096-0_4

Cite this

An, W., Chen, Q., Yang, Y., & He, L. (2017). Knowledge Memory Based LSTM Model for Answer Selection. In D. Zhao, E.-S. M. El-Alfy, D. Liu, S. Xie, & Y. Li (Eds.), Neural Information Processing - 24th International Conference, ICONIP 2017, Proceedings (pp. 34-42). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10635 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-70096-0_4

An, Weijie ; Chen, Qin ; Yang, Yan et al. / Knowledge Memory Based LSTM Model for Answer Selection. Neural Information Processing - 24th International Conference, ICONIP 2017, Proceedings. editor / Dongbin Zhao ; El-Sayed M. El-Alfy ; Derong Liu ; Shengli Xie ; Yuanqing Li. Springer Verlag, 2017. pp. 34-42 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{fe45231fe7b74339b7c5429746511ea3,

title = "Knowledge Memory Based LSTM Model for Answer Selection",

abstract = "Recurrent neural networks (RNN) have shown great success in answer selection task in recent years. Although the attention mechanism has been widely used to enhance the information interaction between questions and answers, knowledge is still the gap between their representations. In this paper, we propose a knowledge memory based RNN model, which incorporates the knowledge learned from the data sets into the question representations. Experiments on two benchmark data sets show the great advantages of our proposed model over that without the knowledge memory. Furthermore, our model outperforms most of the recent progress in question answering.",

keywords = "Answer selection, Deep learning, Knowledge memory",

author = "Weijie An and Qin Chen and Yan Yang and Liang He",

note = "Publisher Copyright: {\textcopyright} 2017, Springer International Publishing AG.; 24th International Conference on Neural Information Processing, ICONIP 2017 ; Conference date: 14-11-2017 Through 18-11-2017",

year = "2017",

doi = "10.1007/978-3-319-70096-0\_4",

language = "英语",

isbn = "9783319700953",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "34--42",

editor = "Dongbin Zhao and El-Alfy, \{El-Sayed M.\} and Derong Liu and Shengli Xie and Yuanqing Li",

booktitle = "Neural Information Processing - 24th International Conference, ICONIP 2017, Proceedings",

address = "德国",

}

An, W, Chen, Q, Yang, Y & He, L 2017, Knowledge Memory Based LSTM Model for Answer Selection. in D Zhao, E-SM El-Alfy, D Liu, S Xie & Y Li (eds), Neural Information Processing - 24th International Conference, ICONIP 2017, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10635 LNCS, Springer Verlag, pp. 34-42, 24th International Conference on Neural Information Processing, ICONIP 2017, Guangzhou, China, 14/11/17. https://doi.org/10.1007/978-3-319-70096-0_4

Knowledge Memory Based LSTM Model for Answer Selection. / An, Weijie; Chen, Qin; Yang, Yan et al.
Neural Information Processing - 24th International Conference, ICONIP 2017, Proceedings. ed. / Dongbin Zhao; El-Sayed M. El-Alfy; Derong Liu; Shengli Xie; Yuanqing Li. Springer Verlag, 2017. p. 34-42 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10635 LNCS).

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

TY - GEN

T1 - Knowledge Memory Based LSTM Model for Answer Selection

AU - An, Weijie

AU - Chen, Qin

AU - Yang, Yan

AU - He, Liang

PY - 2017

Y1 - 2017

N2 - Recurrent neural networks (RNN) have shown great success in answer selection task in recent years. Although the attention mechanism has been widely used to enhance the information interaction between questions and answers, knowledge is still the gap between their representations. In this paper, we propose a knowledge memory based RNN model, which incorporates the knowledge learned from the data sets into the question representations. Experiments on two benchmark data sets show the great advantages of our proposed model over that without the knowledge memory. Furthermore, our model outperforms most of the recent progress in question answering.

AB - Recurrent neural networks (RNN) have shown great success in answer selection task in recent years. Although the attention mechanism has been widely used to enhance the information interaction between questions and answers, knowledge is still the gap between their representations. In this paper, we propose a knowledge memory based RNN model, which incorporates the knowledge learned from the data sets into the question representations. Experiments on two benchmark data sets show the great advantages of our proposed model over that without the knowledge memory. Furthermore, our model outperforms most of the recent progress in question answering.

KW - Answer selection

KW - Deep learning

KW - Knowledge memory

UR - https://www.scopus.com/pages/publications/85035090764

U2 - 10.1007/978-3-319-70096-0_4

DO - 10.1007/978-3-319-70096-0_4

M3 - 会议稿件

AN - SCOPUS:85035090764

SN - 9783319700953

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 34

EP - 42

BT - Neural Information Processing - 24th International Conference, ICONIP 2017, Proceedings

A2 - Zhao, Dongbin

A2 - El-Alfy, El-Sayed M.

A2 - Liu, Derong

A2 - Xie, Shengli

A2 - Li, Yuanqing

PB - Springer Verlag

T2 - 24th International Conference on Neural Information Processing, ICONIP 2017

Y2 - 14 November 2017 through 18 November 2017

ER -

An W, Chen Q, Yang Y, He L. Knowledge Memory Based LSTM Model for Answer Selection. In Zhao D, El-Alfy ESM, Liu D, Xie S, Li Y, editors, Neural Information Processing - 24th International Conference, ICONIP 2017, Proceedings. Springer Verlag. 2017. p. 34-42. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-319-70096-0_4

Knowledge Memory Based LSTM Model for Answer Selection

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this