基于层级池化序列匹配的知识图谱复杂问答最优查询图选择方法

TP391; 在处理知识图谱复杂问答任务时,传统的查询图语义解析方法需要在排序阶段对大量结构复杂的候选查询图进行语义编码,用以获得各自多维特征表示.然而,在编码过程中采用的全局最大或平均池化操作通常存在对代表性特征提取能力不足的问题.针对以上问题,提出一种基于层级池化序列匹配的最优查询图选择方法.在实现候选查询图的交互建模过程中,同时采用层级池化滑动窗口技术分层提取问句和查询图序列对的局部显著性特征与全局语义特征,使得到的特征向量更好地用于候选查询图的语义匹配打分.所提方法在两个流行的复杂问答数据集MetaQA和WebQuestionsSP上开展广泛实验.实验结果表明:引入层级池化操作能够有效...

Full description

Saved in:
Bibliographic Details
Published in系统工程与电子技术 Vol. 46; no. 8; pp. 2686 - 2695
Main Authors 王冬, 周思航, 黄健, 张中杰
Format Journal Article
LanguageChinese
Published 国防科技大学智能科学学院,湖南长沙 410073 01.08.2024
Subjects
层级池化
complex question answering over knowledge graph
知识图谱复杂问答
查询图语义解析
hierarchical pooling
semantic parsing for query graph
interactive encoding
交互编码
Online AccessGet full text
ISSN1001-506X
DOI10.12305/j.issn.1001-506X.2024.08.16

Cover

Abstract TP391; 在处理知识图谱复杂问答任务时,传统的查询图语义解析方法需要在排序阶段对大量结构复杂的候选查询图进行语义编码,用以获得各自多维特征表示.然而,在编码过程中采用的全局最大或平均池化操作通常存在对代表性特征提取能力不足的问题.针对以上问题,提出一种基于层级池化序列匹配的最优查询图选择方法.在实现候选查询图的交互建模过程中,同时采用层级池化滑动窗口技术分层提取问句和查询图序列对的局部显著性特征与全局语义特征,使得到的特征向量更好地用于候选查询图的语义匹配打分.所提方法在两个流行的复杂问答数据集MetaQA和WebQuestionsSP上开展广泛实验.实验结果表明:引入层级池化操作能够有效提取复杂查询图序列的代表性语义特征,增强原有排序模型的交互编码能力,有助于进一步提升知识图谱复杂问答系统的性能.
AbstractList TP391; 在处理知识图谱复杂问答任务时,传统的查询图语义解析方法需要在排序阶段对大量结构复杂的候选查询图进行语义编码,用以获得各自多维特征表示.然而,在编码过程中采用的全局最大或平均池化操作通常存在对代表性特征提取能力不足的问题.针对以上问题,提出一种基于层级池化序列匹配的最优查询图选择方法.在实现候选查询图的交互建模过程中,同时采用层级池化滑动窗口技术分层提取问句和查询图序列对的局部显著性特征与全局语义特征,使得到的特征向量更好地用于候选查询图的语义匹配打分.所提方法在两个流行的复杂问答数据集MetaQA和WebQuestionsSP上开展广泛实验.实验结果表明:引入层级池化操作能够有效提取复杂查询图序列的代表性语义特征,增强原有排序模型的交互编码能力,有助于进一步提升知识图谱复杂问答系统的性能.
Abstract_FL When dealing with complex question answering task over knowledge graph,traditional semantic parsing method for query graphs requires encoding massive candidate query graphs with complex structures in the ranking stage to obtain their respective multi-dimensional feature representations.However,the global maximum or average pooling operation used during the encoding process often suffers from insufficient extracting capability for representative feature.To address the aforementioned problem,an optimal selection method for query graphs based on hierarchical pooling sequence matching is proposed.Meanwhile,sliding window technique based on hierarchical pooling is adopted to hierarchically extract local salient features and global semantic features of question and query graph sequence pairs during the interactive modeling of candidate query graphs,making the resulting feature vectors better used for semantic matching scoring of candidate query graphs.The proposed method is extensively evaluated on two popular complex question answering datasets,MetaQA and WebQuestionsSP.Experiment results show that by introducing hierarchical pooling operation,representative semantic features of complex query graph sequences can be effectively extracted,and the interactive encoding capability of the original ranking model can be enhanced,which helps further improve the performance of complex question answering systems over knowledge graph.
Author 黄健
周思航
张中杰
王冬
AuthorAffiliation 国防科技大学智能科学学院,湖南长沙 410073
AuthorAffiliation_xml – name: 国防科技大学智能科学学院,湖南长沙 410073
Author_FL ZHANG Zhongjie
ZHOU Sihang
WANG Dong
HUANG Jian
Author_FL_xml – sequence: 1
  fullname: WANG Dong
– sequence: 2
  fullname: ZHOU Sihang
– sequence: 3
  fullname: HUANG Jian
– sequence: 4
  fullname: ZHANG Zhongjie
Author_xml – sequence: 1
  fullname: 王冬
– sequence: 2
  fullname: 周思航
– sequence: 3
  fullname: 黄健
– sequence: 4
  fullname: 张中杰
BookMark eNo9kMtKw0AYhWdRwVr7Fq6ExH9mMkm6lOINCm4quCuZZFIaJAWjeFmVoqLY6kZb8UJFFHHhpbhqEJ-mM2nfwhbF1YHzwXfgTKFUWA0FQjMYdEwosLlAr0RRqGMArDEw13UCxNDB1rGZQun_ehJlo6jCgWFqMbCMNCrKTtyPz2S3nsTPqnsvGy0Zn8vjtmz0hofN5Pog6TwN3o_kzffgoysfm-quPmy_Ja8X6rbW_7pSY_owosPaiTp9Ua2e-rycRhO-sxGJ7F9m0NriQjG_rBVWl1by8wUtwsBMjTCTU18QMDxi5rBDHE5dN0ddbluCcdMTFmMUe4zaArgPwnK5J0Y8Ry1hEEwzaPbXu-OEvhOWS0F1ezMcLZZ2t8runrcfROMfwAZs0h-5pXaI
ClassificationCodes TP391
ContentType Journal Article
Copyright Copyright © Wanfang Data Co. Ltd. All Rights Reserved.
Copyright_xml – notice: Copyright © Wanfang Data Co. Ltd. All Rights Reserved.
DBID 2B.
4A8
92I
93N
PSX
TCJ
DOI 10.12305/j.issn.1001-506X.2024.08.16
DatabaseName Wanfang Data Journals - Hong Kong
WANFANG Data Centre
Wanfang Data Journals
万方数据期刊 - 香港版
China Online Journals (COJ)
China Online Journals (COJ)
DatabaseTitleList
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
DocumentTitle_FL Hierarchical pooling sequence matching based optimal selection method of query graph for complex question answering over knowledge graph
EndPage 2695
ExternalDocumentID xtgcydzjs202408016
GrantInformation_xml – fundername: 国家自然科学基金
  funderid: (62006237)
GroupedDBID -0Y
2B.
4A8
5XA
5XJ
92E
92I
93N
ABJNI
ACGFS
ALMA_UNASSIGNED_HOLDINGS
CCEZO
CUBFJ
CW9
PSX
TCJ
TGP
U1G
U5S
ID FETCH-LOGICAL-s1056-256b3fe204d2691a2ab3cc93cb87e5b6de75531d538e0bf0e7cbde93c937e4213
ISSN 1001-506X
IngestDate Thu May 29 04:00:31 EDT 2025
IsPeerReviewed false
IsScholarly true
Issue 8
Keywords 层级池化
complex question answering over knowledge graph
知识图谱复杂问答
查询图语义解析
hierarchical pooling
semantic parsing for query graph
interactive encoding
交互编码
Language Chinese
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-s1056-256b3fe204d2691a2ab3cc93cb87e5b6de75531d538e0bf0e7cbde93c937e4213
PageCount 10
ParticipantIDs wanfang_journals_xtgcydzjs202408016
PublicationCentury 2000
PublicationDate 2024-08-01
PublicationDateYYYYMMDD 2024-08-01
PublicationDate_xml – month: 08
  year: 2024
  text: 2024-08-01
  day: 01
PublicationDecade 2020
PublicationTitle 系统工程与电子技术
PublicationTitle_FL Systems Engineering and Electronics
PublicationYear 2024
Publisher 国防科技大学智能科学学院,湖南长沙 410073
Publisher_xml – name: 国防科技大学智能科学学院,湖南长沙 410073
SSID ssib051375074
ssib002263377
ssib001102898
ssib057620160
ssib023168126
ssib023646287
ssj0042237
Score 2.4447308
Snippet TP391; 在处理知识图谱复杂问答任务时,传统的查询图语义解析方法需要在排序阶段对大量结构复杂的候选查询图进行语义编码,用以获得各自多维特征表示.然而,在编码过程中采用...
SourceID wanfang
SourceType Aggregation Database
StartPage 2686
Title 基于层级池化序列匹配的知识图谱复杂问答最优查询图选择方法
URI https://d.wanfangdata.com.cn/periodical/xtgcydzjs202408016
Volume 46
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVEBS
  databaseName: Inspec with Full Text
  issn: 1001-506X
  databaseCode: ADMLS
  dateStart: 20180801
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: https://www.ebsco.com/products/research-databases/inspec-full-text
  omitProxy: false
  ssIdentifier: ssib057620160
  providerName: EBSCOhost
link http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnR3LahRBsNEERA_iE98ETJ9k47x6pvvYsztLEOPFBLyFeSXiYQWzguYUgopiohdNxAcRUcSDj-Api_g1md3kL6yqnZ0ZjEj0MvRUV1XXo3u6augHY6OGGxoqDlVtxrHdmuNaSS2ETzGMK8dOpJda_cN0Jq6441POpWsCT7yp7i5pR2Px_B_3lfyPVwEGfsVdsv_g2YIpAKAM_oUneBieu_IxDwRXTe5rHjj4lAFCfJNLiwceQrTHAxch2sAqWefKJRxAbhJEcuXlVb7igeISyg0kV4DjUKHJteCB5LrJJZErn_sBQnwDmQNEO0TlctWg1hWy1QGS6wZXDlXVcV0FigpiSIJUOGurwlkhplSII32uqQCS-1Twbd6_MnMQVZOyNvd9KvjIFnX0iDk8JTLBdiWZyEN5fEFig2wGtaJJNhJSF6udERcokFqg6uVBlSSriayR2EC1QQ2wptYlikJh0IZADLiiSgw20EYulB5Yzjeqf2Esp1gD2B83AwM1yL5gfXIz-BgkKXUgZ2gvV0-7xFtRLwEJbSIvqAqcoqAQGZxh1cnWKu8x4F3sKPAKsjepygLMCw4ugLErcxuunhMGXRxZTH75_9_-IJfVmcwdHFGev_Ydu2PGhRRW0JSLbYwVbYyhkfBwXPO3g84pdLrTno3vJvM35iw6Yg_Sjr1s2IJ52Rhiw7oxcflqGdJjBFz5JQDpgm2Xe6ctvHjNLFMIvA_BtcqUQ5g2xMRliiOwERN_CfSjNQfCY7qAaSD4PjY6UOviX5SiPX6tmbA1WwlHJw-xg3keOaL7H4XDbM_89SPsQOV00aNsMlvrbHaeZOuLvc7H7vrbbGkl6zzNHq5mSxvb95d7L-_11j5sfX2Qvfq59W09e7_cfbO4vfql9_lZ9_XC5o8XXax9B7XbC4-6jz91Vza6358fY1PNYLI-XssvUanNQerk1iClieyZ1DKcBHxohlYY2XGs7DiCL7GI3CT1BMzDCQQ-qRHNGKkXR0kK9ZC3pI5l2sfZUOtmKz3BRtJYGBC92sJOYsiiQqm8EPgknopEHBvmSXY-t8l0_pGcm97p6VO7wjrN9pfD7Awbat-6nZ6F8L8dnct7yC_m9Mqt
linkProvider EBSCOhost
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=%E5%9F%BA%E4%BA%8E%E5%B1%82%E7%BA%A7%E6%B1%A0%E5%8C%96%E5%BA%8F%E5%88%97%E5%8C%B9%E9%85%8D%E7%9A%84%E7%9F%A5%E8%AF%86%E5%9B%BE%E8%B0%B1%E5%A4%8D%E6%9D%82%E9%97%AE%E7%AD%94%E6%9C%80%E4%BC%98%E6%9F%A5%E8%AF%A2%E5%9B%BE%E9%80%89%E6%8B%A9%E6%96%B9%E6%B3%95&rft.jtitle=%E7%B3%BB%E7%BB%9F%E5%B7%A5%E7%A8%8B%E4%B8%8E%E7%94%B5%E5%AD%90%E6%8A%80%E6%9C%AF&rft.au=%E7%8E%8B%E5%86%AC&rft.au=%E5%91%A8%E6%80%9D%E8%88%AA&rft.au=%E9%BB%84%E5%81%A5&rft.au=%E5%BC%A0%E4%B8%AD%E6%9D%B0&rft.date=2024-08-01&rft.pub=%E5%9B%BD%E9%98%B2%E7%A7%91%E6%8A%80%E5%A4%A7%E5%AD%A6%E6%99%BA%E8%83%BD%E7%A7%91%E5%AD%A6%E5%AD%A6%E9%99%A2%2C%E6%B9%96%E5%8D%97%E9%95%BF%E6%B2%99+410073&rft.issn=1001-506X&rft.volume=46&rft.issue=8&rft.spage=2686&rft.epage=2695&rft_id=info:doi/10.12305%2Fj.issn.1001-506X.2024.08.16&rft.externalDocID=xtgcydzjs202408016
thumbnail_s http://utb.summon.serialssolutions.com/2.0.0/image/custom?url=http%3A%2F%2Fwww.wanfangdata.com.cn%2Fimages%2FPeriodicalImages%2Fxtgcydzjs%2Fxtgcydzjs.jpg