G2S:基于语义块的知识图谱问答语义解析

doi:10.12263/DZXB.20200157

摘要/Abstract

摘要： 问题意图理解是知识图谱问答的主要任务之一，语义解析是当前理解问题意图的主流方法.其主要挑战是如何充分利用知识图谱上下文理解问句中的隐含实体或关系，以及时间、排序和聚合等复杂约束条件等意图.为了应对这些挑战，本文提出了一种基于语义块的知识图谱问答语义解析框架——Graph-to-Segment，框架中的语义解析模型结合了基于规则的准确度和基于深度学习的覆盖度，实现了问题到语义块序列的解析和语义查询图的构造.框架将问题意图使用基于语义块的语义查询图表示，将问题的语义解析建模为语义块序列生成任务，采用编码器-解码器神经网络模型实现问题到语义块序列的解析，然后通过语义块组装形成语义查询图.同时，结合知识图谱中的上下文信息，模型使用图神经网络学习问题的表示，改进隐含实体或关系的语义解析效果.在两个知识图谱问答数据集上的实验表明，模型性能达到了良好的效果.

关键词: 知识图谱, 问答, 语义解析

Abstract: Question understanding is one of the important tasks of question answering over knowledge graph,where semantic parsing is the mainstream approach for understanding question utterance.The most significant challenge in this task is to understand the implicit entities,relations and the utterances of complex constraints such as time,ordinal,and aggregation in the question with the context of knowledge graph.In this paper,we propose graph-to-segment,a semantic segments based semantic parsing framework for question answering over knowledge graph.Our semantic parsing model integrates both rule-based and neural-based approaches to parse the semantic segment sequences and constructs the semantic query graphs with high accuracy and coverage.These semantic segment-based semantic query graphs,which consist of the semantic segments,are used to represent the utterance of questions.Question semantic parsing is modeled as a sequence generation task,where an encoder-decoder neural network is used to generate the semantic segments from natural language questions.Additionally,with the context information of knowledge graph,a graph neural network is used to learn the representation of questions to improve the effect of semantic parsing on implicit entities or relations.Experimental results show that our model achieves good performance on the two datasets.

Key words: knowledge graph, question answering, semantic parser

中图分类号:

TP183

高留杰, 赵文, 张君福, 姜波. G2S:基于语义块的知识图谱问答语义解析[J]. 电子学报, 2021, 49(6): 1132-1141.

GAO Liu-jie, ZHAO Wen, ZHANG Jun-fu, JIANG Bo. G2S: Semantic Segment Based Semantic Parsing for Question Answering over Knowledge Graph[J]. Acta Electronica Sinica, 2021, 49(6): 1132-1141.

参考文献

[1] Nilesh Chakraborty,Denis Lukovnikov,Gaurav Maheshwari,et al.Introduction to Neural Network based Approaches for Question Answering over Knowledge Graphs[EB/OL].http://export.arxiv.org/abs/1907.09361, 2019.
[2] Antoine Bordes,Sumit Chopra,Jason Weston.Question answering with subgraph embeddings[A].Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing[C].Doha,Qatar:EMNLP,2014.615-620.
[3] Li Dong,Furu Wei,Ming Zhou,et al.Question answering over freebase with multi-column convolutional neural networks[A].53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing[C].Beijing,China:ACL,2015.260-269.
[4] Denis Lukovnikov,Asja Fischer,Jens Lehmann,et al.Neural network-based question answering over knowledge graphs on word and character level[A].WWW'17 Proceedings of the 26th International Conference on World Wide Web[C].Perth,Australia:IW3C2,2017.1211-1220.
[5] Sen Hu,Lei Zou,Jeffrey Xu Yu,et al.Answering natural language questions by subgraph matching over knowledge graphs[J].IEEE Transactions on Knowledge and Data Engineering,2018,30(5),824-837.
[6] Yuanzhe Zhang,Kang Liu,Shizhu He,et al.Question Answering over Knowledge Base with Neural Attention Combining Global Knowledge Information[EB/OL].https://arxiv.org/abs/1606.00979,2016.
[7] Antoine Bordes,Nicolas Usunier,Alberto Garcia-Duran,et al.Translating embeddings for modeling multi-relational data[A].Advances in Neural Information Processing Systems 26[C].Lake Tahoe,Nevada:NIPS,2013.2787-2795.
[8] Zihang Dai,Lei Li,Wei Xu.CFO-conditional focused neural question answering with large-scale knowledge bases[A].Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics[C].Berlin,Germany:ACL,2016.800-810.
[9] Gaurav Maheshwari,Priyansh Trivedi,Denis Lukovnikov,et al.Learning to rank query graphs for complex question answering over knowledge graphs[A].The Semantic Web-ISWC 2019[C].Springer,Cham:ISWC, 2019.487-504.
[10] Wen-tauYih,Ming-Wei Chang,Xiaodong He,et al.Semantic parsing via staged query graph generation-question answering with knowledge base[A].Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing[C].Beijing,China:ACL,2015.1321-1331.
[11] Jun-Wei Bao,Nan Duan,Zhao Yan,et al.Constraint-based question answering with knowledge graph[A].Proceedings of COLING 2016,the 26th International Conference on Computational Linguistics:Technical Papers[C].Osaka,Japan:COLING,2016.2503-2514.
[12] Hongye Tan,Honghong Zhao,Ru Li,Bei Liu.A pipeline approach to free-description question answering in chinese gaokao reading comprehension[J].Chinese Journal of Electronics,2019,28(1),113-119.
[13] Shizhu He,Yuanzhe Zhang,Kang Liu,et al.CASIA@V2:A mln-based question answering system over linked data[A].Question Answering over Linked Data (QALD-4)[C].Sheffield,United Kingdom:CLEF,2014.1249-1259.
[14] Saeedeh Shekarpour,Edgard Marx,Axel-Cyrille Ngonga Ngomo,et al.SINA:Semantic interpretation of user queries for question answering on interlinked data[J].Journal of Web Semantics,2015,30:39-51.
[15] Jie Liu,Wei Li,Liming Luo,et al.Linked open data query based on natural language[J].Chinese Journal of Electronics,2017,26(2),230-235.
[16] Jonathan Berant,Andrew Chou,Roy Frostig,et al.Semantic parsing on freebase from question-answer pairs[A].Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing[C].Seattle,Washington,USA:EMNLP,2013.1533-1544.
[17] Percy Liang,Michael I Jordan,Dan Klein.Learning dependency-based compositional semantics[J].Computational Linguistics,2013,39(2),389-446.
[18] Luke S Zettlemoyer,Michael Collins.Learning to map sentences to logical form:Structured classification with probabilistic categorial grammars[A].UAI'05 Proceedings of the Twenty-First Conference on Uncertainty in Artificial Intelligence[C].Arlington,Virginia,USA:AUAI,2005.658-666.
[19] Luke Zettlemoyer,Michael Collins.Online learning of relaxed CCG grammars for parsing to logical form[A].Proceedings of the 2007 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning[C].Prague,Czech Republic:EMNLP-CoNLL,2007.678-687.
[20] Jianpeng Cheng,Siva Reddy,Vijay A Saraswat,et al.Learning an executable neural semantic parser[J].Computational Linguistics,2019, 45(1):59-94.
[21] Robin Jia,Percy Liang.Data recombination for neural semantic parsing[A].Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics[C].Berlin,Germany:ACL,2016.12-22.
[22] Chunyang Xiao,Marc Dymetman,Claire Gardent.Sequence-based structured prediction for semantic parsing[A].Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics[C].Berlin,Germany:ACL,2016.1341-1350.
[23] Li Dong,Mirella Lapata.Language to logical form with neural attention[A].Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics[C].Berlin,Germany:ACL,2016.33-43.
[24] Sen Hu,Lei Zou,Xinbo Zhang.A state-transition framework to answer complex questions over knowledge base[A].Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing,Brussels[C].Brussels,Belgium:ACL,2018.2098-2108
[25] Bo Chen,Le Sun,Xianpei Han.Sequence-to-action-end-to-end semantic graph generation for semantic parsing[A].Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics[C].Melbourne,Australia:ACL,2018.766-777.
[26] Pengcheng Yin,Graham Neubig.TRANX:A transition-based neural abstract syntax parser for semantic parsing and code generation[A].Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing:System Demonstrations[C].Brussels,Belgium:2018,EMNLP.7-12.
[27] D Punjani,K Singh,A Both,et al.Template-based question answering over linked geospatial data[A].Proceedings of the 12th workshop on geographic information retrieval[C].New York,USA:ACM,2018.7-16.
[28] Zheng Weiguo,Yu Jeffrey Xu,Lei Zou,et al.Question answering over knowledge graphs-question understanding via template decomposition[J].PVLDB,2018,11(11),1373-1386.
[29] Kun Xu,Lingfei Wu,Zhiguo Wang,et al.Graph2Seq:Graph to Sequence Learning with Attention-based Neural Networks[EB/OL].https://arxiv.org/abs/1804.00823, 2018.
[30] George A.Miller.Wordnet:A lexical database for english[J].Communications of the ACM,1995,38(11),39-41.
[31] Yi Yang,Ming-Wei Chang.S-MART:Novel tree-based structured learning algorithms applied to tweet entity linking[A].Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing[C].Beijing,China:ACL,2015.504-513.
[32] Dzmitry Bahdanau,Kyunghyun Cho,Yoshua Bengio.Neural machine translation by jointly learning to align and translate[A].3rd International Conference on Learning Representations[C].San Diego,United States:ICLR,2015.
[33] Ilya Sutskever,Oriol Vinyals,Quoc V Le.Sequence to sequence learning with neural networks[A].27th International Conference on Neural Information Processing Systems[C].Montreal,Quebec,Canada:NIPS,2014.3104-3112.
[34] Nitish Srivastava,Geoffrey Hinton,Alex Krizhevsky,et al.Dropout:A simple way to prevent neural networks from overfitting[J].Journal of Machine Learning Research,2014,15(1):1929-1958.
[35] Tom Kwiatkowski,Luke S Zettlemoyer,Sharon Goldwater,et al.Lexical generalization in CCG grammar induction for semantic parsing[A].Proceedings of the 2011 Conference on Empirical Methods in Natural Language Processing[C].Edinburgh,Scotland,UK:EMNLP,2011.1512-1523.
[36] Kai Zhao,Hany Hassan,Michael Auli.Learning translation models from monolingual continuous representations[A].Proceedings of the 2015 Conference of the North American Chapter of the Association for Computational Linguistics:Human Language Technologies[C].Denver,Colorado:NAACL,2015.1527-1536.
[37] Maxim Rabinovich,Mitchell Stern,Dan Klein.Abstractsyntax networks for code generation and semantic parsing[A].Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics[C].Vancouver,Canada:ACL,2017.1139-49.
[38] Zhaopeng Tu,Zhengdong Lu,Yang Liu,et al.Modelingcoverage for neural machine translation[A].Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics[C].Berlin,Germany:ACL,2016.76-85.
[39] Bo Chen,Le Sun,Xianpei Han,et al.Sentence rewriting for semantic parsing[A].Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics[C].Berlin,Germany:ACL,2019.766-777.
[40] Li Dong,Jonathan Mallinson,Siva Reddy,et al.Learning to paraphrase for question answering[A].Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing[C].Copenhagen,Denmark:EMNLP,2017.875-886.
[41] Yushi Wang,Jonathan Berant,Percy Liang.Building asemantic parser overnight[A].Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing[C].Beijing,China:ACL,2015.1332-1342.