摘要
可视分析领导人出访事件的时空特征对量化分析国际关系具有重要意义。通过采集全球领导人出访的相关新闻文本数据,利用LDA模型抽取新闻主题,设计领导人出访事件数据模型,采用时间序列统计分析方法对出访事件的时间特征进行可视分析,得出领导人出访事件的时间规律、出访主题的演变规律及出访主题之间的相关性;对出访事件的空间特征进行可视分析,得出全球大国博弈的热点地区分布情况和部分国家的外交政策;结合复杂网络模型,使用地理-网络关系图对国家双边关系进行可视分析,得出国家之间的双边关系概览情况。
Visual analysis of the temporal and spatial characteristics of leaders’visits is of great significance to the quantitative analysis of international relations.In this paper,by collecting the relevant news text data of global leaders’visits,and using LDA model to extract news topics,we design a data model of leaders’visits.It uses time series statistical analysis method to visually analyze the time characteristics of the visits,and gets the time rule of leaders’visits,the evolution rule of the visits and the correlation between the visits.Based on the visual analysis of the spatial characteristics of the visiting events,the distribution of the hot spots of the global power game and the foreign policies of some countries are obtained.Combined with the complex network model,the geographic network diagram is used to carry out the visual analysis of the bilateral relations between countries,and the overview of the bilateral relations between countries is obtained.
作者
刘建湘
刘海砚
刘一萱
李佳
康磊
LIU Jianxiang;LIU Haiyan;LIU Yixuan;LI Jia;KANG Lei(University of Information Engineering,Zhengzhou 450001,China;Unit 96669,Beijing 100000,China)
出处
《信息工程大学学报》
2020年第4期482-489,共8页
Journal of Information Engineering University
基金
国家自然科学基金资助项目(41801313)。
关键词
领导人出访
事件
LDA模型
时间序列统计分析
复杂网络
时空特征
leader visit
event
LDA
time series statistical analysis
complex network
temporal and spatial characteristics