基于远程监督的多因子人物关系抽取模型

doi:10.11959/j.issn.1000-436x.2018118

通信学报 ›› 2018, Vol. 39 ›› Issue (7): 103-112.doi: 10.11959/j.issn.1000-436x.2018118

基于远程监督的多因子人物关系抽取模型

黄杨琛¹(),贾焰¹,甘亮¹,徐菁¹,黄九鸣¹,赫中翮²

¹ 国防科技大学计算机学院，湖南长沙 410073
² 湖南星汉数智科技有限公司知识图谱研发部，湖南长沙 410205

修回日期:2018-06-21 出版日期:2018-07-01 发布日期:2018-08-08
作者简介:黄杨琛（1991-），女，湖南衡阳人，国防科技大学博士生，主要研究方向为数据挖掘、自然语言处理等。|贾焰（1960-），女，四川成都人，国防科技大学教授、博士生导师，主要研究方向为社交网络分析、信息安全等。|甘亮（1977-），男，江西樟树人，国防科技大学讲师，主要研究方向为知识工程、专家系统等。|徐菁（1989-），女，山东济南人，国防科技大学博士生，主要研究方向为知识工程、文本挖掘等。|黄九鸣（1981-），男，福建安溪人，国防科技大学讲师，主要研究方向为社交网络分析、信息安全等。|赫中翮（1991-），男，山西大同人，湖南星汉数智科技有限公司工程师，主要研究方向为信息抽取。
基金资助:
国家重点研究发展计划基金资助项目(2016QY03D0601);国家重点研究发展计划基金资助项目(2016QY03D0603);国家自然科学基金资助项目(61502517);湖南省重点研发计划基金资助项目(2018GK2056)

Multi-factor person entity relation extraction model based on distant supervision

Yangchen HUANG¹(),Yan JIA¹,Liang GAN¹,Jing XU¹,Jiuming HUANG¹,Zhonghe HE²

¹ College of Computer,National University of Defense Technology,Changsha 410073,China
² KB R＆D department,Hunan Singhand Intelligent Data Technology Co.,Ltd.,Changsha 410205,China

Revised:2018-06-21 Online:2018-07-01 Published:2018-08-08
Supported by:
The National Key Research and Development Program of China(2016QY03D0601);The National Key Research and Development Program of China(2016QY03D0603);The National Natural Science Foundation of China(61502517);Key Research and Development Plan of Hunan Province(2018GK2056)

摘要/Abstract

摘要：

针对远程监督的基本假设过强容易引入噪声数据的问题，提出了一种可以对远程监督自动生成的训练数据去噪的人物实体关系抽取模型。在训练数据生成阶段，通过多示例学习的思想和基于 TF-IDF 的关系指示词发现的方法对远程监督产生的数据进行去噪处理，使训练数据达到人工标注质量。在模型分类器中，提出采用词法特征和句法特征相结合的多因子特征作为关系特征向量用于分类器的学习。在大规模真实数据集上的实验结果表明，所提模型结果优于同类型的关系抽取方法。

关键词: 关系抽取, 人物关系, 远程监督, 机器学习, 自然语言处理

Abstract:

Aiming at the problem that the basic assumption of distant supervision was too strong and easy to produce noise data,a model of the person entity relation extraction which could automatically filter the training data generated by distant supervision was proposed.For training data generation,the data produced by distant supervision would be filtered by multiple instance learning and the method of TF-IDF-based relation keyword detecting,which tried to make the training data has the manual annotation quality.Furthermore,the model combined lexical and syntactic features to extract the effective relation feature vector from two angles of words and semantics for classifier.The experiment results on large scale real-world datasets show that the proposed model outperforms other relation extraction methods which based on distant supervision.

Key words: relation extraction, person entity relation, distant supervision, machine learning, natural language processing

中图分类号:

TP391

黄杨琛,贾焰,甘亮,徐菁,黄九鸣,赫中翮. 基于远程监督的多因子人物关系抽取模型[J]. 通信学报, 2018, 39(7): 103-112.

Yangchen HUANG,Yan JIA,Liang GAN,Jing XU,Jiuming HUANG,Zhonghe HE. Multi-factor person entity relation extraction model based on distant supervision[J]. Journal on Communications, 2018, 39(7): 103-112.

图/表 11

图1

图2

图3

表1

图4

表2

图5

图6

表3

表4

表5

参考文献 29

[1]	CHINCHOR N , MARSH E . Muc-7 information extraction task definition[C]// The Seventh Message Understanding Conference (MUC-7),Appendices. 1998: 359-367.
[2]	CHE W , LIU T , LI S . Automatic entity relation extraction[J]. Journal of Chinese Information Processing, 2005,19(2): 1-6.
[3]	刘绍毓, 李弼程, 郭志刚 ,等. 实体关系抽取研究综述[J]. 信息工程大学学报, 2016(5): 541-547.
	LIU S Y , LI B C , GUO Z G ,et al. Review of entity relation extraction[J]. Journal of Information Engineering University, 2016(5): 541-547.
[4]	APPELT D E , HOBBS J R , BEAR J ,et al. SRI International FASTUS system:MUC-6 test results and analysis[C]// The 6th Conference On Message Understanding. 1995: 237-248.
[5]	YANGARBER R , GRISHMAN R . NYU:description of the proteus/PET system as used for MUC-7 ST[C]// Seventh Message Understanding Conference (MUC-7). 1998.
[6]	AONE C , RAMOS S M . REES:a large-scale relation and event extraction system[C]// The Sixth Conference On Applied Natural Language Processing. 2000: 76-83.
[7]	KAMBHATLA N , . Combining lexical,syntactic,and semantic features with maximum entropy models for extracting relations[C]// The ACL 2004 On Interactive Poster And Demonstration Sessions. 2004:22.
[8]	GUO D Z , JIAN S , JIE Z ,et al. Exploring various knowledge in relation extraction[C]// The 43rd Annual Meeting on Association for Computational Linguistics. 2005: 427-434.
[9]	宁海燕 . 实体关系自动抽取技术的比较研究[D]. 哈尔滨:哈尔滨工业大学, 2010.
	NING H Y . Comparative study of automatic entity relation extraction[D]. Harbin:Harbin Institute of Technology, 2010.
[10]	WANG L , CAO Z , MELO D G ,et al. Relation classification via multi-level attention CNNS[C]// The 54th Annual Meeting of the Association for Computational Linguistics. 2016: 1298-1307.
[11]	KUMAR S . A survey of deep learning methods for relation extraction[J]. arXiv preprint,arXiv:1705.03645, 2017.
[12]	ZHOU P , SHI W , TIAN J ,et al. Attention-based bidirectional long short-term memory networks for relation classification[C]// The 54th Annual Meeting of the Association for Computational Linguistics. 2016: 207-212.
[13]	CARLSON A , BETTERIDGE J , KISIEL B ,et al. Toward an architecture for never-ending language learning[C]// AAAI. 2010:3.
[14]	刘锦文, 许静, 张利萍 ,等. 基于标签传播和主动学习的人物社会关系抽取[J]. 计算机工程, 2017,34(2): 234-240.
	LIU J W , XU J , ZHANG L P ,et al. Personal social relation extraction based on label propagation and active learning[J]. Computer Engineering, 2017,34(2): 234-240.
[15]	秦兵, 刘安安, 刘挺 . 无指导的中文开放式实体关系抽取[J]. 计算机研究与发展, 2015,52(5): 1029-1035.
	QIN B , LIU A A , LIU T . Unsupervised Chinese open entity relation extraction[J]. Journal of Computer Research and Development, 2015,52(5): 1029-1035.
[16]	GASHTEOVSKI K , GEMULLA R , DEL C L . MinIE:minimizing facts in open information extraction[C]// The 2017 Conference on Empirical Methods in Natural Language Processing. 2017: 2630-2640.
[17]	BOLLACKER K , EVANS C , PARITOSH P ,et al. Freebase:a collaboratively created graph database for structuring human knowledge[C]// The 2008 ACM SIGMOD International Conference on Management of Data. 2008: 1247-1250.
[18]	AUER S , BIZER C , KOBILAROV G ,et al. Dbpedia:a nucleus for a web of open data[M]// The Semantic Web. 2007: 722-735.
[19]	SUCHANEK F M , KASNECI G , WEIKUM G . Yago:a core of semantic knowledge[C]// The 16th International Conference on World Wide Web. 2007: 697-706.
[20]	MINTZ M , BILLS S , SNOW R ,et al. Distant supervision for relation extraction without labeled data[C]// The Joint Conference of the 47th Annual Meeting of the ACL and The 4th International Joint Conference on Natural Language Processing of the AFNLP. 2009: 1003-1011.
[21]	JI G , LIU K , HE S ,et al. Distant supervision for relation extraction with sentence-level attention and entity descriptions[C]// AAAI. 2017: 3060-3066.
[22]	PERSHINA M , MIN B , XU W ,et al. Infusion of labeled data into distant supervision for relation extraction[C]// The 52nd Annual Meeting of the Association for Computational Linguistics. 2014: 732-738.
[23]	潘云, 布勒布丽汗·伊沙巴依, 杨静 ,等. 利用中文在线资源的远程监督人物关系抽取[J]. 小型微型计算机系统, 2015,4:11.
	PAN Y , BULEHULIHAN.YISHABAY , YANG J ,et al. Distant supervised personal relation extraction using chinese online resource[J]. Journal of Chinese Mini-Micro Computer Systems, 2015,4:11.
[24]	黄蓓静, 贺樑, 杨静 . 远程监督人物关系抽取中的去噪研究[J]. 计算机应用与软件, 2017,34(7): 11-18.
	HUANG B J , HE L , YANG J . Research on noise reduction in distant supervised personal relation extraction[J]. Computer Application and Software, 2017,34(7): 11-18.
[25]	CRAVEN M , KUMLIEN J . Constructing biological knowledge bases by extracting information from text sources[C]// ISMB. 1999: 77-86.
[26]	SURDEANU M , TIBSHIRANI J , NALLAPATI R ,et al. Multi-instance multi-label learning for relation extraction[C]// The 2012 Joint Conference On Empirical Methods In Natural Language Processing And Computational Natural Language Learning. 2012: 455-465.
[27]	HOFFMANN R , ZHANG C , LING X ,et al. Knowledge-based weak supervision for information extraction of overlapping relations[C]// The 49th Annual Meeting of the Association for Computational Linguistics:Human Language Technologies. 2011: 541-550.
[28]	胡宝顺, 王大玲, 于戈 ,等. 基于句法结构特征分析及分类技术的答案提取算法[J]. 计算机学报, 2008,31(4): 662-676.
	HU B S , WANG D L , YU G ,et al. An answer extraction algorithm based on syntax structure feature parsing and classification[J]. Chinese Journal of Computers, 2008,31(4): 662-676.
[29]	郭喜跃, 何婷婷, 胡小华 ,等. 基于句法语义特征的中文实体关系抽取[J]. 中文信息学报, 2014,28(6): 183-189.
	GUO X Y , HE T T , HU X H ,et al. Chinese named entity relation extraction based on syntactic and semantic features[J]. Journal of Chinese Information Processing, 2014,28(6): 183-189.

关系类型	编号	知识库中的三元组/个	训练集/个	测试集/个
夫妻	1	96 593	22 176	5 543
父子	2	75 304	19 938	4 984
母子	3	52 047	16 182	4 046
兄弟	4	80 691	10 947	2 735
姐妹	5	20 785	14 432	3 611

n	数目/个
1	238
2	296
3	317
4	331
5	342

特征	P	R	F₁
dis	61.3%	57.5%	59.3%
+order	65.1%	60.4%	62.2%
+vn	66.4%	60.7%	63.4%
+parsing_r	72.6%	64.1%	68.1%
+p_dis	77.6%	65.8%	71.2%
+context	79.1%	66.9%	72.5%

新关系	P	R	F₁
朋友	76.4%	63.1%	69.1%
同事	73.7%	61.2%	66.8%

研究方法	P	R	F₁
标签传播模型	69.3%	—	—
模式聚类模型	74.2%	60.6%	64.6%
本文模型	79.1%	66.9%	72.5%

基于远程监督的多因子人物关系抽取模型

Multi-factor person entity relation extraction model based on distant supervision

在线阅读

PDF下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 11

参考文献 29

相关文章 15

Metrics

推荐阅读 0

[1]	戴千一, 张斌, 郭松, 徐开勇. 基于多分类器集成的区块链网络层异常流量检测方法[J]. 通信学报, 2023, 44(3): 66-80.
[2]	何高峰, 魏千峰, 肖咸财, 朱海婷, 徐丙凤. 支持数据隐私保护的恶意加密流量检测确认方法[J]. 通信学报, 2022, 43(2): 156-170.
[3]	冯智斌, 徐煜华, 杜智勇, 刘鑫, 李文, 韩昊, 张晓博. 对抗智能干扰的主动防御技术[J]. 通信学报, 2022, 43(10): 42-54.
[4]	陆彦辉, 柳寒, 李航, 朱光旭. 基于多鉴别器生成对抗网络的时间序列生成模型[J]. 通信学报, 2022, 43(10): 167-176.
[5]	梅锴, 赵海涛, 刘潇然, 刘军, 熊俊, 任保全, 魏急波. 高效的基于数据与模型的信道估计算法[J]. 通信学报, 2022, 43(1): 59-70.
[6]	彭长根, 高婷, 刘惠篮, 丁红发. 面向机器学习模型的基于PCA的成员推理攻击[J]. 通信学报, 2022, 43(1): 149-160.
[7]	邹福泰, 谭越, 王林, 蒋永康. 基于生成对抗网络的僵尸网络检测[J]. 通信学报, 2021, 42(7): 95-106.
[8]	刘留, 张建华, 樊圆圆, 于力, 张嘉驰. 机器学习在信道建模中的应用综述[J]. 通信学报, 2021, 42(2): 134-153.
[9]	伏玉笋,杨根科. 人工智能在移动通信中的应用：挑战与实践[J]. 通信学报, 2020, 41(9): 190-201.
[10]	陈铁明,金成强,吕明琪,朱添田. 基于样本增强的网络恶意流量智能检测方法[J]. 通信学报, 2020, 41(6): 128-138.
[11]	韩春雨,张永铮,张玉. Fast-flucos：基于DNS流量的Fast-flux恶意域名检测方法[J]. 通信学报, 2020, 41(5): 37-47.
[12]	刘敖迪, 杜学绘, 王娜, 乔蕊. 基于深度学习的ABAC访问控制策略自动化生成技术[J]. 通信学报, 2020, 41(12): 8-20.
[13]	周鑫,何晓新,郑昌文. 基于图像深度学习的无线电信号识别[J]. 通信学报, 2019, 40(7): 114-125.
[14]	杜学绘,林杨东,孙奕. 基于混合特征的恶意PDF文档检测[J]. 通信学报, 2019, 40(2): 118-128.
[15]	李青,钟将,李立力,李琪. 基于预训练机制的自修正复杂语义分析方法[J]. 通信学报, 2019, 40(12): 41-50.