基于主题关系的中文短文本图模型实体消歧

计算机工程与科学 ›› 2023, Vol. 45 ›› Issue (01): 154-162.

基于主题关系的中文短文本图模型实体消歧

马瑛超，张晓滨

（西安工程大学计算机科学学院，陕西西安 710048）

收稿日期:2021-01-19 修回日期:2021-06-10 接受日期:2023-01-25 出版日期:2023-01-25 发布日期:2023-01-25
基金资助:
陕西省自然科学基金（2019JQ-849）；西安工程大学研究生创新基金（chx2021028）

Entity disambiguation of Chinese short text using graph model based on topic relations

MA Ying-chao，ZHANG Xiao-bin

（School of Computer Science,Xi’an Polytechnic University,Xi’an 710048,China）

Received:2021-01-19 Revised:2021-06-10 Accepted:2023-01-25 Online:2023-01-25 Published:2023-01-25

摘要/Abstract

摘要： 实体消歧作为知识库构建、信息检索等应用的重要支撑技术，在自然语言处理领域有着重要的作用。然而在短文本环境中，对实体的上下文特征进行建模的传统消歧方式很难提取到足够多用以消歧的特征。针对短文本的特点，提出一种基于实体主题关系的中文短文本图模型消歧方法，首先，通过TextRank算法对知识库信息构建的语料库进行主题推断，并使用主题推断的结果作为实体间关系的表示；然后，结合基于BERT的语义匹配模型给出的消歧评分对待消歧文本构建消歧网络图；最终，通过搜索排序得出最后的消歧结果。使用CCKS2020短文本实体链接任务提供的数据集对所提方法进行评测，实验结果表明，该方法对短文本的实体消歧效果优于其他方法，能有效解决在缺乏知识库实体关系情况下的中文短文本实体消歧问题。

关键词: 实体消歧, 图模型, 主题推断, TextRank

Abstract: As an important supporting technology for applications such as knowledge base construction and information retrieval, entity disambiguation plays an important role in the field of Natural Language Processing (NLP). However, in the short text environment, it is difficult for entity disambiguation to extract sufficient context features for disambiguation. Aiming at the characteristics of short texts, this paper proposes a disambiguation method of graph models based on entity topic relations. This method uses TextRank algorithm to infer the topic of corpus constructed by knowledge base information, and uses the result of topic inference as the representation of relationship between entities. By combining the disambiguation score given by the semantic matching model based on BERT, the disambiguation network graph is constructed, and the final disambiguation result is obtained through search and sorting. The data set provided in the short text entity link task of CCKS2020 is used to evaluate the method. The experimental results show that the proposed method is better than other entity linking methods in entity disambiguation of short text, and can effectively solve the entity disambiguation problem of Chinese short text.

Key words: entity disambiguation, graph model, topic inference, TextRank

马瑛超, 张晓滨. 基于主题关系的中文短文本图模型实体消歧[J]. 计算机工程与科学, 2023, 45(01): 154-162.

MA Ying-chao, ZHANG Xiao-bin. Entity disambiguation of Chinese short text using graph model based on topic relations[J]. Computer Engineering & Science, 2023, 45(01): 154-162.

参考文献［18］

［1］	Fan Peng-cheng,Shen Ying-han,Xu Hong-bo,et al.Joint entity disambiguation with entity knowledge description［J］.Journal of Chinese Information Processing,2020,34(7):42-49.(in Chinese)
［2］	Yang J Q,Li Y J,Gao C J,et al.Entity disambiguation with context awareness in user-generated short texts［J］.Expert Systems with Applications,2020,160:113652.
［3］	Wen Ping-mei, Ye Zhi-wei,Ding Wen-jian,et al.Developments of named entity disambiguation［J］.Data Analysis and Knowledge Discovery,2020,4(9):15-25.(in Chinese)
［4］	Devlin J, Chang M W,Lee K,et al.BERT:Pre-training of deep bidirectional transformers for language understanding［J］.arXiv:1810.04805v2,2019.
［5］	Peters M E,Neumann M,Iyyer M,et al.Deep contextualized word representations［C］∥Proc of 2018 Conference of the North American Chapter of the Association for Computational Linguistics,2018:2227-2237.
［6］	Radford A,Narasimhan K,Salimans T,et al.Improving language understanding by generative pre-training［EB/OL］.［2020-04-23］.https://s3-us-west-2.amazonaw s.com/openai-assets/research-covers/language-unsuperv ised/language_understanding_paper.pdf.
［7］	He Z Y,Liu S J,Li M,et al.Learning entity representation for entity disambiguation［C］∥Proc of the 51st Annual Meeting of the Association for Computational Linguistics,2013:30-34.
［8］	Francis-Landau M, Durrett G,Klein D.Capturing semantic similarity for entity linking with convolutional neural networks［EB/OL］.［2019-06-25］.https://arxiv.org/pdf/1604.00734.pdf.
［9］	Phong L,Ivan T.Improving entity linking by modeling latent relations between mentions［C］∥Proc of the 56th Annual Meeting of the Association for Computational Linguistics,2018:1-12.
［10］	Raiman J,Raiman O.DeepType:Multilingual entity linking by neural type system evolution［C］∥Proc of the 32nd AAAI Conference on Artificial Intelligence,2018:5406-5413.
［11］	Huang Bin.Random walk based name disambiguation algorithm in social networks［J］.Application Research of Computers,2015,32(12):3650-3653.(in Chinese)
［12］	Ma N Y, Liu X,Gao Y L.Entity linking based on graph model and semantic representation［C］∥Proc of International Conference on Knowledge Science,Engineering and Management,2019:561-571.
［13］	Kartsaklis D, Pilehvar M T,Collier N.Mapping text to knowledge graph entities using multi-sense LSTMs［C］∥Proc of the 2018 Conference on Empirical Methods in Natural Language Processing,2018:1959-1970.
［14］	Wang Rui,Li Bi-cheng,Du Wen-qian.Entity disambiguation based on context word vector and topic models［J］.Journal of Chinese Information Processing,2019,33(11):46-56.(in Chinese)
［15］	Mihalcea R, Tarau P.TextRank:Bringing order into texts［C］∥Proc of Conference on Empirical Methods in Natural Language Processing,2004:108-116.
［16］	Blei D M,Ng A Y,Jordan M I.Latent dirichlet allocation［J］.Journal of Machine Learning Research,2003,3:993-1022.
［17］	Zhang Y L,Eick C F.Tracking events in twitter by combin- ing a LDA-based approach and a density-contour clustering approach［J］.International Journal of Semantic Computing,2019,13(1):87-110.
［18］	Guo Qing.Chinese text keyword extraction algorithm based on graph and LDA［D］.Beijing:Beijing University of Posts and Telecommunications,2019.(in Chinese)
［19］	Kim Y.Convolutional neural networks for sentence classification［C］∥Proc of 2014 Conference on Empirical Methods in Natural Language Processing,2014:1746-1751.
［20］	Lai S W,Xu L H,Liu K,et al.Recurrent convolutional neural networks for text classification［C］∥Proc of the 29th AAAI Conference on Artificial Intelligence,2015:2267-2273.
	附中文参考文献：
［1］	范鹏程,沈英汉,许洪波,等.融合实体知识描述的实体联合消歧方法［J］.中文信息学报,2020,34(7):42-49.
［3］	温萍梅,叶志炜,丁文健,等.命名实体消歧研究进展综述［J］.数据分析与知识发现,2020,4(9):15-25.
［11］	黄斌. 社会网络中基于随机游走的名称消歧算法［J］. 计算机应用研究,2015,32(12):3650-3653.
［14］	王瑞,李弼程,杜文倩.基于上下文词向量和主题模型的实体消歧方法［J］.中文信息学报,2019,33(11):46-56.
［18］	郭庆.基于图与LDA的中文文本关键词提取算法［D］.北京:北京邮电大学,2019.

[1]	王会举, 李孟萱, 黄卫卫, 周秋怡. 基于隐马尔可夫模型的多真值发现算法[J]. 计算机工程与科学, 2021, 43(03): 518-524.
[2]	蒋亚芳, 严馨, 徐广义, 周枫, 邓忠莹, . 融合多信息句子图模型的多文档摘要抽取[J]. 计算机工程与科学, 2020, 42(03): 535-542.
[3]	陈勇1,袁梦霆2,李清安2. 分区内存体系结构片选指令优化方法[J]. J4, 2016, 38(02): 195-201.
[4]	卢正鼎郭洁刘芳. 一种基于超图模型的客户行为分析方法[J]. J4, 2006, 28(11): 54-55.
[5]	熊志辉[1] 李思昆[1] 陈吉华[1] 张鲁峰[2]. 基于平台的SoC系统建模方法研究[J]. J4, 2005, 27(8): 58-59.