摘要
恐怖主义受到世界各国的普遍关注,众多学者在事件模型上进行了大量探索,但现有事件模型的恐怖主义概念层次特征不够突出,时序、空间和语义关系相对缺失,且知识图谱模式层的概念和关系类型固定,难以满足事件的多样化描述信息,因此,亟需以事件为中心并结合时空和语义特征构建事件表示模型。在分析事件组成要素的基础上,结合知识图谱前沿技术,提出并设计了模式、事件、数据三层结构的恐怖主义事件表示模型。无须在模式层扩充概念和关系类型,即可在数据层实现事件类型和描述信息的拓展,在事件层实现时序、空间和语义关系表示,并在原型系统中以宏观概览、时序回溯和语义展示3个场景为例,验证了模型的可行性和有效性。
Objectives: Terrorism has received widespread attention from all over the world, and researchers have explored construction of event model. However, there are less temporal, spatial and semantic relationships in event models currently, and the diversified expression of event pose challenges to construct an event model. Therefore, it is urgent to construct a terrorism event model combining temporal, spatial and semantic features.Methods: Based on the analysis of event components, we propose a three-layer terrorism event model which is proposed and designed with the advanced technology of knowledge graph. Results:Without expanding the concepts and relationships, the multi-source heterogeneous data is integrated to represent the conceptual hierarchy of event as well as the temporal, spatial and semantic relationships.Conclusions: In the prototype system, the feasibility and effectiveness of the model is confirmed with three cases:Macro overview, temporal backtracking and semantic display.
作者
刘俊楠
刘海砚
陈晓慧
郭漩
郭文月
朱新铭
赵清波
李佳
LIU Junnan;LIU Haiyan;CHEN Xiaohui;GUO Xuan;GUO Wenyue;ZHU Xinming;ZHAO Qingbo;LI Jia(School of Data and Target Engineering,Information Engineering University,Zhengzhou 450001,China;School of Geo-Science and Technology,Zhengzhou University,Zhengzhou 450052,China;Joint Laboratory of Eco-Meteorology,Chinese Academy of Meteorological Sciences,Zhengzhou University,Zhengzhou 450052,China;Institute of Geospatial Information,Information Engineering University,Zhengzhou 450001,China)
出处
《武汉大学学报(信息科学版)》
EI
CAS
CSCD
北大核心
2022年第2期313-322,共10页
Geomatics and Information Science of Wuhan University
基金
国家自然科学基金(41801313)
河南省自然科学基金(182300410005)。
关键词
恐怖主义
事件模型
知识图谱
时空关系
语义关系
terrorism
event model
knowledge graph
spatiotemporal relationship
semantic relationship