A Semantic Expansion-Based Joint Model for Answer Ranking in Chinese Question Answering Systems

Wenxiu Xie; Leung-Pun Wong; Lap-Kei Lee; Oliver Au; Tianyong Hao

doi:10.1007/978-3-030-42835-8_3

A Semantic Expansion-Based Joint Model for Answer Ranking in Chinese Question Answering Systems

Wenxiu Xie
, Leung-Pun Wong
, Lap-Kei Lee
, Oliver Au
, Tianyong Hao^*

^*Corresponding author for this work

Department of Linguistics and Translation

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

Abstract

Answer ranking is one of essential steps in open domain question answering systems. The ranking of the retrieved answers directly affects user satisfaction. This paper proposes a new joint model for answer ranking by leveraging context semantic features, which balances both question-answer similarities and answer ranking scores. A publicly available dataset containing 40,000 Chinese questions and 369,919 corresponding answer passages from Sogou Lab is used for experiments. Evaluation on the joint model shows a Precison@1 of 72.6%, which outperforms the state-of-the-art baseline methods.

Original language	English
Title of host publication	Information Retrieval Technology - 15th Asia Information Retrieval Societies Conference, AIRS 2019, Proceedings
Editors	Fu Lee Wang, Haoran Xie, Wai Lam
Publisher	Springer
Pages	22-33
ISBN (Electronic)	9783030428358
ISBN (Print)	9783030428341
DOIs	https://doi.org/10.1007/978-3-030-42835-8_3
Publication status	Published - Nov 2019
Event	15th Asia Information Retrieval Societies Conference (AIRS 2019) - Hong Kong, China Duration: 7 Nov 2019 → 9 Nov 2019

Publication series

Name	Lecture Notes in Computer Science
Volume	12004
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	15th Asia Information Retrieval Societies Conference (AIRS 2019)
Place	China
City	Hong Kong
Period	7/11/19 → 9/11/19

Research Keywords

Answer ranking
Joint model
Synonyms
Word2vec

Access Output

10.1007/978-3-030-42835-8_3

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

Xie, W., Wong, L.-P., Lee, L.-K., Au, O., & Hao, T. (2019). A Semantic Expansion-Based Joint Model for Answer Ranking in Chinese Question Answering Systems. In F. L. Wang, H. Xie, & W. Lam (Eds.), Information Retrieval Technology - 15th Asia Information Retrieval Societies Conference, AIRS 2019, Proceedings (pp. 22-33). (Lecture Notes in Computer Science; Vol. 12004). Springer . https://doi.org/10.1007/978-3-030-42835-8_3

Xie, Wenxiu ; Wong, Leung-Pun ; Lee, Lap-Kei et al. / A Semantic Expansion-Based Joint Model for Answer Ranking in Chinese Question Answering Systems. Information Retrieval Technology - 15th Asia Information Retrieval Societies Conference, AIRS 2019, Proceedings. editor / Fu Lee Wang ; Haoran Xie ; Wai Lam. Springer , 2019. pp. 22-33 (Lecture Notes in Computer Science).

@inproceedings{6a2ec3edfb0a45ec8f8e0ff465432921,

title = "A Semantic Expansion-Based Joint Model for Answer Ranking in Chinese Question Answering Systems",

abstract = "Answer ranking is one of essential steps in open domain question answering systems. The ranking of the retrieved answers directly affects user satisfaction. This paper proposes a new joint model for answer ranking by leveraging context semantic features, which balances both question-answer similarities and answer ranking scores. A publicly available dataset containing 40,000 Chinese questions and 369,919 corresponding answer passages from Sogou Lab is used for experiments. Evaluation on the joint model shows a Precison@1 of 72.6\%, which outperforms the state-of-the-art baseline methods.",

keywords = "Answer ranking, Joint model, Synonyms, Word2vec",

author = "Wenxiu Xie and Leung-Pun Wong and Lap-Kei Lee and Oliver Au and Tianyong Hao",

year = "2019",

month = nov,

doi = "10.1007/978-3-030-42835-8\_3",

language = "English",

isbn = "9783030428341",

series = "Lecture Notes in Computer Science",

publisher = "Springer ",

pages = "22--33",

editor = "Wang, \{Fu Lee\} and Haoran Xie and Wai Lam",

booktitle = "Information Retrieval Technology - 15th Asia Information Retrieval Societies Conference, AIRS 2019, Proceedings",

note = "15th Asia Information Retrieval Societies Conference (AIRS 2019) ; Conference date: 07-11-2019 Through 09-11-2019",

}

Xie, W, Wong, L-P, Lee, L-K, Au, O & Hao, T 2019, A Semantic Expansion-Based Joint Model for Answer Ranking in Chinese Question Answering Systems. in FL Wang, H Xie & W Lam (eds), Information Retrieval Technology - 15th Asia Information Retrieval Societies Conference, AIRS 2019, Proceedings. Lecture Notes in Computer Science, vol. 12004, Springer , pp. 22-33, 15th Asia Information Retrieval Societies Conference (AIRS 2019), Hong Kong, China, 7/11/19. https://doi.org/10.1007/978-3-030-42835-8_3

A Semantic Expansion-Based Joint Model for Answer Ranking in Chinese Question Answering Systems. / Xie, Wenxiu; Wong, Leung-Pun; Lee, Lap-Kei et al.
Information Retrieval Technology - 15th Asia Information Retrieval Societies Conference, AIRS 2019, Proceedings. ed. / Fu Lee Wang; Haoran Xie; Wai Lam. Springer , 2019. p. 22-33 (Lecture Notes in Computer Science; Vol. 12004).

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - A Semantic Expansion-Based Joint Model for Answer Ranking in Chinese Question Answering Systems

AU - Xie, Wenxiu

AU - Wong, Leung-Pun

AU - Lee, Lap-Kei

AU - Au, Oliver

AU - Hao, Tianyong

PY - 2019/11

Y1 - 2019/11

N2 - Answer ranking is one of essential steps in open domain question answering systems. The ranking of the retrieved answers directly affects user satisfaction. This paper proposes a new joint model for answer ranking by leveraging context semantic features, which balances both question-answer similarities and answer ranking scores. A publicly available dataset containing 40,000 Chinese questions and 369,919 corresponding answer passages from Sogou Lab is used for experiments. Evaluation on the joint model shows a Precison@1 of 72.6%, which outperforms the state-of-the-art baseline methods.

AB - Answer ranking is one of essential steps in open domain question answering systems. The ranking of the retrieved answers directly affects user satisfaction. This paper proposes a new joint model for answer ranking by leveraging context semantic features, which balances both question-answer similarities and answer ranking scores. A publicly available dataset containing 40,000 Chinese questions and 369,919 corresponding answer passages from Sogou Lab is used for experiments. Evaluation on the joint model shows a Precison@1 of 72.6%, which outperforms the state-of-the-art baseline methods.

KW - Answer ranking

KW - Joint model

KW - Synonyms

KW - Word2vec

UR - https://www.scopus.com/pages/publications/85082387671

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85082387671&origin=recordpage

U2 - 10.1007/978-3-030-42835-8_3

DO - 10.1007/978-3-030-42835-8_3

M3 - RGC 32 - Refereed conference paper (with host publication)

SN - 9783030428341

T3 - Lecture Notes in Computer Science

SP - 22

EP - 33

BT - Information Retrieval Technology - 15th Asia Information Retrieval Societies Conference, AIRS 2019, Proceedings

A2 - Wang, Fu Lee

A2 - Xie, Haoran

A2 - Lam, Wai

PB - Springer

T2 - 15th Asia Information Retrieval Societies Conference (AIRS 2019)

Y2 - 7 November 2019 through 9 November 2019

ER -

Xie W, Wong LP, Lee LK, Au O, Hao T. A Semantic Expansion-Based Joint Model for Answer Ranking in Chinese Question Answering Systems. In Wang FL, Xie H, Lam W, editors, Information Retrieval Technology - 15th Asia Information Retrieval Societies Conference, AIRS 2019, Proceedings. Springer . 2019. p. 22-33. (Lecture Notes in Computer Science). doi: 10.1007/978-3-030-42835-8_3

A Semantic Expansion-Based Joint Model for Answer Ranking in Chinese Question Answering Systems

Abstract

Publication series

Conference

Research Keywords

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this