计算机科学 ›› 2019, Vol. 46 ›› Issue (10): 215-221.doi: 10.11896/jsjkx.190400071

• 人工智能 • 上一篇    下一篇

基于多角度注意力机制的单一事实知识库问答方法

罗达, 苏锦钿, 李鹏飞   

  1. (华南理工大学计算机科学与工程学院 广州511400)
  • 收稿日期:2019-04-11 修回日期:2019-06-28 出版日期:2019-10-15 发布日期:2019-10-21
  • 通讯作者: 苏锦钿(1980-),男,博士,副教授,CCF会员,主要研究方向为机器学习和自然语言处理,E-mail:sujd@scut.edu.cn。
  • 作者简介:罗达(1994-),男,硕士生,主要研究方向为机器学习和自然语言处理,E-mail:472447327@qq.com;李鹏飞(1993-),男,硕士生,主要研究方向为机器学习、自然语言处理。
  • 基金资助:
    本文受广东省自然科学基金(2015A030310318),广东省科技厅应用型科技研发专项资金项目(20168010124010),广东省医学科学技术研究基金项目(A2015065)资助。

Multi-view Attentional Approach to Single-fact Knowledge-based Question Answering

LUO Da, SU Jin-dian, LI Peng-fei   

  1. (School of Computer Science and Engineering,South China University of Technology,Guangzhou 511400,China)
  • Received:2019-04-11 Revised:2019-06-28 Online:2019-10-15 Published:2019-10-21

摘要: 近年来,基于知识库的问答受到了广泛的关注,成为了一个重要的自然语言处理任务。在基于知识库的问答任务中,简单问题是指能够通过知识库的单一事实进行回答的问题。针对简单问题的回答,现有的解决方法主要是将问题和知识库事实映射到同一向量空间中,然后通过计算问题和事实之间的相似度来得到答案,但这种方法会损失原始单词的部分语义交互信息。针对该问题,文中提出了一种基于多角度注意力机制的关系检测模型,从多个角度对问题和知识库关系的相关性进行了建模,从而保留了更多的原始交互信息,并提高了模型的准确率。此外,为了减小噪音的影响并提高实体识别的准确率,在实体链接过程中提出结合基于语言模型的动态词向量和单词的词性特征对问题进行表征。实验结果表明,所提方法在基于FB2M和FB5M的SimpleQuestions数据集上分别获得了78.9%和78.3%的准确率,能够很好地反映问题与知识库关系之间的语义相关性,并提升了单一事实知识库问答的准确率。

关键词: 深度学习, 问答, 知识库, 注意力机制, 自然语言处理

Abstract: Knowledge base question answering (KB-QA) has received extensive attention in recent years,and becomes an important natural language processing task.In the knowledge base question answering task,simple question refers to the question that can be answered by a single-fact of the knowledge base.For this task,the existing approaches mainly map the question and the KB fact into a common vector space and calculate their similarity to get the answer.But this approach would lose part of the semantic interaction information of the original words.To solve this problem,a multi-view attention-based relation detection approach was proposed,which aims to model the correlation between the question and the KB relation from multiple perspectives,and preserves more original interaction information so as to improve the accuracy of the approach.In addition,in order to alleviate the impacts of noisy data and improve the accuracy of entity recognition,this paper also encoded the question by combining the dynamic word vectors based on the language model and the part-of-speech feature of the word during the process of entity linking.Finally,experiments conducted on the SimpleQuestions dataset based on FB2M and FB5M achieve the accuracy results of 78.9% and 78.3%,respectively,which illustrative the effectiveness of the proposed approach for reflecting the semantic correlation between the question and the KB relation,and reflect the improvements of the accuracy for single-fact KBQA.

Key words: Attention mechanism, Deep learning, Knowledge base, Natural language processing, Question answering

中图分类号: 

  • TP183
[1]SUCHANEK F M,KASNECI G,WEIKUM G.Yago:a core of semantic knowledge[C]//Proceedings of the 16th international conference on World Wide Web.New York,NY:ACM,2007:697-706.
[2]BOLLACKER K,EVANS C,PARITOSH P,et al.Freebase:a collaboratively created graph database for structuring human knowledge[C]//Proceedings of the 2008 ACM SIGMOD international conference on Management of data.New York,NY:ACM,2008:1247-1250.
[3]LEHMANN J,ISELE R,JAKOB M,et al.DBpedia-a large-scale,multilingual knowledge base extracted from Wikipedia[J].Semantic Web,2015,6(2):167-195.
[4]BERANT J,LIANG P.Semantic parsing via paraphrasing[C]//Proceedings of the 52nd Annual Meeting of ACL (Volume 1:Long Papers).Stroudsburg,PA:ACL,2014:1415-1425.
[5]REDDY S,LAPATA M,STEEDMAN M.Large-scale semantic parsing without question-answer pairs[J].Transactions of the Association of Computational Linguistics,2014,2(1):377-392.
[6]BERANT J,CHOU A,FROSTIG R,et al.Semantic parsing on freebase from question-answer pairs[C]//Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing.2013:1533-1544.
[7]BORDES A,CHOPRA S,WESTON J.Question Answering with Subgraph Embeddings[C]//Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing,A meeting of SIGDAT,a Special Interest Group of the ACL.Doha,Qatar,View at Publisher View at Google Scholar,2014:615-620.
[8]BORDES A,USUNIER N,CHOPRA S,et al.Large-scale simple question answering with memory networks[J].arXiv:1506.02075,2015.
[9]HE X,GOLUB D.Character-Level Question Answering with Attention[C]//Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing.Stroudsburg,PA:ALL,2016:1598-1607.
[10]HAKIMOV S,JEBBARA S,CIMIANO P.Evaluating Architectural Choices for Deep Learning Approaches for Question Answering Over Knowledge Bases[C]//ICSC.IEEE,2019:110-113.
[11]DAI Z,LI L,XU W.CFO:Conditional Focused Neural Question Answering with Large-scale Knowledge Bases[C]//Procee-dings of the 54th Annual Meeting of the Association for Computational Linguistics.Stroudsburg,PA:ACL,2016:800-810.
[12]YIN W,YU M,XIANG B,et al.Simple Question Answering by Attentive Convolutional Neural Network[C]//COLING 2016.New York:ACM,2016:1746-1756.
[13]YIH W T,CHANG M W,HE X,et al.Semantic Parsing via Staged Query Graph Generation:Question Answering with Knowledge Base[C]//Meeting of the Association for Computational Linguistics & the International Joint Conference on Natural Language Processing.Stroudsburg,PA:ALL,2015:1321-1331.
[14]YANG Y,CHANG M W.S-MART:Novel Tree-based Struc-tured Learning Algorithms Applied to Tweet Entity Linking[C]//ACL 2015.Stroudsburg,PA:ACL,2015:504-513.
[15]QU Y,LIU J,KANG L,et al.Question Answering over Freebase via Attentive RNN with Similarity Matrix based CNN[J].arXiv:1804.03317,2018.
[16]YU Y,HASAN K S,YU M,et al.Knowledge Base Relation Detection via Multi-View Matching[C]//ADBIS (Short Papers and Workshops).Berlin:Springer,2018:286-294.
[17]WANG Z,HAMZA W,FLORIAN R.Bilateral Multi-Perspective Matching for Natural Language Sentences[C]//IJCAI 2017.Cambridge,MA:MIT,2017:4144-4150.
[18]ZHANG H,XU G,LIANG X,et al.An Attention-Based Word-Level Interaction Model:Relation Detection for Knowledge Base Question Answering[J].IEEE Access,2018:1-1.
[19]BERANT J,CHOU A,FROSTIG R,et al.Semantic parsing on freebase from question-answer pairs[C]//Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing.Stroudsburg,PA:ALL,2013:1533-1544.
[20]HAO Y,LIU H,HE S,et al.Pattern-revising Enhanced Simple Question Answering over Knowledge Bases[C]//Proceedings of the 27th International Conference on Computational Linguistics.New York:ACM,2018:3272-3282.
[21]YU M,YIN W,HASAN K S,et al.Improved neural relation detection for knowledge base question answering[C]//Proceedings of the 55th Annual Meeting of the Association for ComputationalLinguistics.Stroudsburg,PA:ACL,2017:571-581.
[1] 饶志双, 贾真, 张凡, 李天瑞.
基于Key-Value关联记忆网络的知识图谱问答方法
Key-Value Relational Memory Networks for Question Answering over Knowledge Graph
计算机科学, 2022, 49(9): 202-207. https://doi.org/10.11896/jsjkx.220300277
[2] 汤凌韬, 王迪, 张鲁飞, 刘盛云.
基于安全多方计算和差分隐私的联邦学习方案
Federated Learning Scheme Based on Secure Multi-party Computation and Differential Privacy
计算机科学, 2022, 49(9): 297-305. https://doi.org/10.11896/jsjkx.210800108
[3] 周芳泉, 成卫青.
基于全局增强图神经网络的序列推荐
Sequence Recommendation Based on Global Enhanced Graph Neural Network
计算机科学, 2022, 49(9): 55-63. https://doi.org/10.11896/jsjkx.210700085
[4] 吕晓锋, 赵书良, 高恒达, 武永亮, 张宝奇.
基于异质信息网的短文本特征扩充方法
Short Texts Feautre Enrichment Method Based on Heterogeneous Information Network
计算机科学, 2022, 49(9): 92-100. https://doi.org/10.11896/jsjkx.210700241
[5] 戴禹, 许林峰.
基于文本行匹配的跨图文本阅读方法
Cross-image Text Reading Method Based on Text Line Matching
计算机科学, 2022, 49(9): 139-145. https://doi.org/10.11896/jsjkx.220600032
[6] 周乐员, 张剑华, 袁甜甜, 陈胜勇.
多层注意力机制融合的序列到序列中国连续手语识别和翻译
Sequence-to-Sequence Chinese Continuous Sign Language Recognition and Translation with Multi- layer Attention Mechanism Fusion
计算机科学, 2022, 49(9): 155-161. https://doi.org/10.11896/jsjkx.210800026
[7] 徐涌鑫, 赵俊峰, 王亚沙, 谢冰, 杨恺.
时序知识图谱表示学习
Temporal Knowledge Graph Representation Learning
计算机科学, 2022, 49(9): 162-171. https://doi.org/10.11896/jsjkx.220500204
[8] 熊丽琴, 曹雷, 赖俊, 陈希亮.
基于值分解的多智能体深度强化学习综述
Overview of Multi-agent Deep Reinforcement Learning Based on Value Factorization
计算机科学, 2022, 49(9): 172-182. https://doi.org/10.11896/jsjkx.210800112
[9] 王剑, 彭雨琦, 赵宇斐, 杨健.
基于深度学习的社交网络舆情信息抽取方法综述
Survey of Social Network Public Opinion Information Extraction Based on Deep Learning
计算机科学, 2022, 49(8): 279-293. https://doi.org/10.11896/jsjkx.220300099
[10] 郝志荣, 陈龙, 黄嘉成.
面向文本分类的类别区分式通用对抗攻击方法
Class Discriminative Universal Adversarial Attack for Text Classification
计算机科学, 2022, 49(8): 323-329. https://doi.org/10.11896/jsjkx.220200077
[11] 姜梦函, 李邵梅, 郑洪浩, 张建朋.
基于改进位置编码的谣言检测模型
Rumor Detection Model Based on Improved Position Embedding
计算机科学, 2022, 49(8): 330-335. https://doi.org/10.11896/jsjkx.210600046
[12] 朱承璋, 黄嘉儿, 肖亚龙, 王晗, 邹北骥.
基于注意力机制的医学影像深度哈希检索算法
Deep Hash Retrieval Algorithm for Medical Images Based on Attention Mechanism
计算机科学, 2022, 49(8): 113-119. https://doi.org/10.11896/jsjkx.210700153
[13] 孙奇, 吉根林, 张杰.
基于非局部注意力生成对抗网络的视频异常事件检测方法
Non-local Attention Based Generative Adversarial Network for Video Abnormal Event Detection
计算机科学, 2022, 49(8): 172-177. https://doi.org/10.11896/jsjkx.210600061
[14] 闫佳丹, 贾彩燕.
基于双图神经网络信息融合的文本分类方法
Text Classification Method Based on Information Fusion of Dual-graph Neural Network
计算机科学, 2022, 49(8): 230-236. https://doi.org/10.11896/jsjkx.210600042
[15] 汪鸣, 彭舰, 黄飞虎.
基于多时间尺度时空图网络的交通流量预测模型
Multi-time Scale Spatial-Temporal Graph Neural Network for Traffic Flow Prediction
计算机科学, 2022, 49(8): 40-48. https://doi.org/10.11896/jsjkx.220100188
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!