信息资源管理学报 ›› 2022, Vol. 12 ›› Issue (3): 137-151.doi: 10.13365/j.jirm.2022.03.137

所属专题: 数字经济时代信息技术在应急管理中的理论与实践

• 专刊-数字经济时代信息技术在应急管理中的理论与实践 • 上一篇    下一篇

面向应急管理的金融突发事件事理知识图谱构建与分析研究

刘政昊1,2,3 曾曦1  张志剑1,2,3   

  1. 1.武汉大学信息管理学院,武汉,430072; 
    2.武汉大学大数据研究院,武汉,430072; 
    3.武汉大学信息资源研究中心,武汉,430072
  • 出版日期:2022-05-26 发布日期:2022-06-26
  • 作者简介:刘政昊(通讯作者),博士生,研究方向为金融知识组织与舆情分析,Email:zhenghaoliu@whu.edu.cn;曾曦,硕士生,研究方向为文本挖掘与政策计量;张志剑,博士生,研究方向为知识图谱。
  • 基金资助:
    本文系国家自然科学基金重大研究计划“大数据驱动的管理与决策研究”重点支持项目“基于知识关联的金融大数据价值分析、发现及协同创造机制”(91646206);科技创新2030-“新一代人工智能”重大项目课题(2020AAA0108505)的研究成果之一。

Research on Construction and Analysis of Financial Emergencies Knowledge Graph with Event Logical for Emergency Management

Liu Zhenghao1,2,3 Zeng Xi1 Zhang Zhijian1,2,3   

  1. 1.School of Information Management, Wuhan University, Wuhan, 430072; 
    2.Institute of Big Data, Wuhan University, Wuhan, 430072, China;
    3.Center for Studies of Information Resources, Wuhan University, Wuhan,430072
  • Online:2022-05-26 Published:2022-06-26

摘要: 金融突发事件的爆发和演化不利于金融市场的稳定,并加剧了系统性风险发生的可能。为了及时分析和应对金融突发事件并建立健全应急管理机制,提出构建金融事理知识图谱用以揭示突发事件的演化机理和辅助应急管理与决策。研究基于知识驱动的事件关联与演化视角,从海量财经新闻报道中选取高质量金融突发事件文本数据,通过本体设计、事件抽取、事件泛化等步骤构建了具有199736个事件和知识实体、444950个关系的双层结构事理知识图谱,并通过包商银行破产案例分析了突发事件演变逻辑。 研究结果表明,金融突发事件事理知识图谱可以较为深入全面地揭示金融突发事件的演化逻辑和风险传导规律,并为金融突发事件的应急管理提供科学有效的分析思路和现实依据。

关键词: 金融突发事件, 事理知识图谱, 事件演化, 应急管理, 金融风险

Abstract: The outbreak and evolution of financial emergencies are not conducive to the stability of financial markets, and exacerbate the possibility of systemic risks. In order to timely analyze and respond to financial emergencies and improve the emergency management mechanism, it is proposed to build a knowledge graph with event logical(KGEL)of financial emergencies to reveal the evolution mechanism of emergencies and assist emergency management and decision-making. Based on the perspective of knowledge-driven event association and evolution, the study selects high-quality financial emergency text data from a large number of financial news reports and constructs a two-tier structure KGEL with 199,736 events and knowledge entities and 444,950 relationships through ontology design, event extraction and event generalization. It also analyzes the evolution logic of emergencies through cases of Baoshang Bank, to provide corresponding suggestions for the improvement scheme and countermeasures of emergency management capability based on the graph. The study results show that the KGEL of financial emergencies can deeply and comprehensively reveal the evolution logic and risk transmission law of financial emergencies, and provide scientific and effective analysis ideas for emergency management of financial emergencies.

Key words: Financial Emergency, Knowledge Graph with Event Logical, Event Evolution, Emergency Management, Financial Risk

中图分类号: