结合因果强度和扩充元组的突发公共事件事理图谱构建

doi:10.3778/j.issn.1002-8331.2402-0138

摘要/Abstract

摘要： 突发公共事件对社会有严重的危害，通过对突发公共事件文本进行分析，可以辅助建立社会预警机制、提高突发公共事件的应急治理效率。由此，提出结合因果强度和扩充元组的突发公共事件事理图谱构建方法。在语料选取方面，选取新闻文本结合政策文本，保证语料的时效性和专业性。在事理图谱构建环节，基于直接因果关系词，结合语言学实现因果词扩充，根据扩充后的句法模式结合因果强度抽取因果事件句，并基于扩充后的语义元组实现事件抽取，对抽取得到的事件进行泛化和对齐，提高适用性。实验证明，该方法可以更有效地提取文本中的事件及其因果关系，据此构建的事理图谱能够体现国家应急预案文件中的治理思路，从而为辅助相关决策提供参考价值。

关键词: 事理图谱, 突发公共事件, BERT, 因果强度, 语义元组

Abstract: Public emergencies do serious harm to society, and analyzing the text of public emergencies can assist in the establishment of a social early-warning mechanism and improve the efficiency of emergency management of public emergencies. As a result, this paper proposes the construction method of public emergency event knowledge graph by combining causal intensity and expanded tuple. In terms of corpus selection, news texts and policy texts are selected to ensure the timeliness and specialization of the corpus. In the construction of event knowledge graph, based on the direct causal words, it combines with linguistics to realize causal word expansion. According to the expanded syntactic pattern combined with causal intensity, the paper extracts causal event sentences, realizes the event extraction based on the expanded semantic tuple, and finally generalizes and aligns the extracted events to improve the applicability. The experiment proves that the method can extract the events and their causal relations in the text more effectively, and the event knowledge graph constructed accordingly can reflect the governance ideas in the national emergency plan documents, thus providing reference value for assisting relevant decision-making.

Key words: event knowledge graph, public emergencies, BERT, causal intensity, semantic tuple

任俊玲, 戴景劢. 结合因果强度和扩充元组的突发公共事件事理图谱构建[J]. 计算机工程与应用, 2025, 61(11): 195-203.

REN Junling, DAI Jingmai. Construction of Event Knowledge Graph of Public Emergencies Combining Causal Intensity and Expanded Tuples[J]. Computer Engineering and Applications, 2025, 61(11): 195-203.

参考文献

[1] 刘雅姝, 栾宇, 周红磊, 等. 基于事理图谱的重大突发事件动态演变研究[J]. 图书情报工作, 2022, 66(10): 143-151.
LIU Y S, LUAN Y, ZHOU H L, et al. Research on the dynamic evolution of major emergencies based on event knowledge graph[J]. Library and Information Service, 2022, 66(10): 143-151.
[2] 王淑君, 郑文静, 王琦琦, 等. 基于政策工具的突发公共卫生事件风险防范化解保障政策文本量化分析[J]. 疾病监测, 2023, 38(1): 106-111.
WANG S J, ZHENG W J, WANG Q Q, et al. Quantitative analysis on supportive policy documents for risk prevention and mitigation of public health emergency based on policy instruments-oriented content[J]. Disease Surveillance, 2023, 38(1): 106-111.
[3] 盛东方, 尹航. 基于政策文本计算的突发公共事件下中小企业扶持政策供需匹配研究—以新冠肺炎疫情为例[J]. 现代情报, 2020, 40(8): 10-19.
SHENG D F, YIN H. Research on the matching of supply and demand of supporting policies for minor enterprises in public emergencies based on policy text computing—take novel coronavirus pneumonia as an example[J]. Journal of Modern Information, 2020, 40(8): 10-19.
[4] 张敬炟, 赵锐, 郭贺, 等. SARS以来我国突发公共卫生事件应对法律文本的政策工具量化分析[J]. 中国卫生事业管理, 2022, 39(7): 527-532.
ZHANG J D, ZHAO R, GUO H, et al. Quantitative analysis of policy tools for coping with legal texts for public health emergencies in China since SARSEvent[J]. Chinese Health Service Management, 2022, 39(7): 527-532.
[5] 王晰巍, 王小天, 李玥琪. 重大突发事件网络舆情UGC的事理图谱构建研究——以自然灾害7·20河南暴雨为例[J]. 图书情报工作, 2022, 66(16): 13-23.
WANG X W, WANG X T, LI Y Q. Research on the construction of the event evolution graph of UGC of network public opinions for major emergencies—taking the natural disaster 7·20 torrential rain in henan as an example[J]. Library and Information Service, 2022, 66(16): 13-23.
[6] 于凯, 杨富义. 基于事理图谱的突发事件网络舆情演化模型构建[J]. 上海理工大学学报, 2023, 45(1): 27-35.
YU K, YANG F Y. Construction of an evolution model of emergency network public opinion based on event knowledge graph[J]. Journal of University of Shanghai for Science and Technology, 2023, 45(1): 27-35.
[7] 朱艺娜, 曹阳, 钟靖越, 等. 事件抽取技术研究综述[J]. 计算机科学, 2022, 49(12): 264-273.
ZHU Y N, CAO Y, ZHONG J Y, et al. Survey on event extraction technology[J]. Computer Science, 2022, 49(12): 264-273.
[8] CHIEU H L, NG H T. A maximum entropy approach to information extraction from semi-structured and free text[C]//Proceedings of the 8th National Conference on Artificial Intelligence, 2002: 786-791.
[9] LLORENS H, ESTELA S, BORJA N. TimeML events recognition and classification: learning CRF models with semantic roles[C]//Proceedings of the 23rd International Conference on Computational Linguistics, 2010: 725-733.
[10] LI Q, JI H, HUANG L. Joint event extraction via structured prediction with global features[C]//Proceedings of the Meeting of the Association for Computational Linguistics, 2013: 73-82.
[11] DING N, LI Z, LIU Z, et al. Event detection with trigger-aware lattice neural network[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, 2019: 347-356.
[12] ZENG Y, YANG H, FENG Y, et al. A convolution BiLSTM neural network model for Chinese event extraction[C]//Proceedings of the Natural Language Understanding and Intelligent Applications, 2016: 275-287.
[13] SATYAPANICH T, FERRARO F, FININ T. CASIE: extracting cybersecurity event information from text[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2020: 8749-8757.
[14] 伍依兰. “才”字因果句[J]. 语言研究, 2023, 43(4): 35-44.
WU Y L. Complex sentences with Cái(才)[J]. Studies in Language and Linguistics, 2023, 43(4): 35-44.
[15] 刘畅. 表因果关系的关联词语研究[D]. 长春: 东北师范大学, 2011.
LIU C. The research of associated words which are said causal relationship[D]. Changchun: Northeast Normal University, 2011.
[16] 任康瑞, 任玲萱, 王中杰. 基于事理图谱的元宇宙发展路径分析[J]. 计算机仿真, 2023, 40(10): 9-12.
REN K R, REN L X, WANG Z J. Analysis on development path of metaverse based on event evolutionary graph[J]. Computer Simulation, 2023, 40(10): 9-12.
[17] 赵森栋, 刘挺. 因果关系及其在社会媒体上的应用研究综述[J]. 软件学报, 2014, 25(12): 2733-2752.
ZHAO S D, LIU T. Causality and its applications in social media: a survey[J]. Journal of Software, 2014, 25(12): 2733-2752.
[18] 李章超, 何琳, 喻雪寒. 基于事理图谱的典籍内容知识组织与应用——以《左传》为例[J]. 图书馆论坛, 2024, 44(4): 125-137.
LI Z C, HE L, YU X H. Research on contextual knowledge organization and application of classics based on event knowledge graph—take ZuoZhuan as example[J]. Library Tribune, 2024, 44(4): 125-137.
[19] 裘江南. 汉语文本中突发事件因果关系抽取方法研究[D]. 大连: 大连理工大学, 2012.
QIU J N. Research on emergency causality extraction from chinese corpus[D]. Dalian: Dalian University of Technology, 2012.
[20] 杨纪星, 杨波, 朱剑林, 等. 金融领域事件因果关系发现及事理图谱构建与应用[J]. 中文信息学报, 2023, 37(7): 131-142.
YANG J X, YANG B, ZHU J L, et al. Event causality extraction, eventic graph construction and application in financial domain[J]. Journal of Chinese Information Processing, 2023, 37(7): 131-142.
[21] 孙传明, 陈熙, 王萍. 基于在线评论的非遗纪录片受众需求事理图谱构建研究[J]. 情报科学, 2023, 41(9): 107-114.
SUN C M, CHEN X, WANG P. The construction of audience needs event logic graph of intangible cultural heritage documentary based on online reviews[J]. Information Science, 2023, 41(9): 107-114.
[22] 储泽祥, 陶伏平. 汉语因果复句的关联标记模式与“联系项居中原则”[J]. 中国语文, 2008, (5): 410-422.
CHU Z X, TAO F P. Associative marking patterns in Chinese causal complex clauses and the "principle of centering the linking term"[J]. Studies of the Chinese Language, 2008, (5): 410-422.
[23] 高文成, 吴超异. 基线/阐释视角下英汉因果关系复合句认知对比研究[J]. 解放军外国语学院学报, 2023, 46(6): 61-69.
GAO W C, WU C Y. A comparative study on the cognition of English-Chinese causal compound sentences from the baseline/explanation perspective[J]. Journal of PLA University of Foreign Languages, 2023, 46(6): 61-69.
[24] 曾子明, 李青青, 孙守强, 等. 面向突发公共卫生事件网络舆情的事理图谱构建及演化分析[J]. 情报理论与实践, 2023, 46(8): 147-155.
ZENG Z M, LI Q Q, SUN S Q, et al. Construction of the event evolution graph and analysis of evolution situation of network public opinion for public health emergencies[J]. Information Studies: Theory & Application, 2023, 46(8): 147-155.
[25] 郎君, 刘挺, 张会鹏, 等. LTP: 语言技术平台[C]//第三届学生计算语言学研讨会论文集, 2006: 79-83.
LANG J, LIU T, ZHANG H P, et al. LTP: language technology platform[C]//Proceedings of the 3rd Student Symposium on Computational Linguistics, 2006: 79-83.
[26] DEVLIN J, CHANG M W, LEE K, et al. BERT: pre-training of deep bidirectional transformers for language understanding[J]. arXiv:1810.04805, 2018.