引用本文: | 肖卫东,周城,孙扬,等.Parallel Sets的改进及其在全球恐怖袭击数据分析中的应用[J].国防科技大学学报,2011,33(1):115-119.[点击复制] |
XIAO Weidong,ZHOU Cheng,SUN Yang,et al.Improvement of Parallel Sets and Its Application in Analyzing Global Terrorism Database[J].Journal of National University of Defense Technology,2011,33(1):115-119[点击复制] |
|
|
|
本文已被:浏览 7089次 下载 5606次 |
Parallel Sets的改进及其在全球恐怖袭击数据分析中的应用 |
肖卫东, 周城, 孙扬, 葛斌, 汤大权 |
(国防科技大学 C4ISR技术国防科技重点实验室,湖南 长沙 410073)
|
摘要: |
随着恐怖主义愈演愈烈,“反恐”成为当今世界各国军事安全部门的中心任务。使用分类型可视化工具Parallel Sets分析国际恐怖主义数据库中多属性分类值间的关系,揭示数据库中的隐性信息,并针对Parallel Sets任意排列分类值产生较多交叉的不足,提出带降势的启发式分类值布局算法,自动优化分类值布局顺序,减轻视图中的可视混乱,降势策略可以减少参与计算的分类值数目。实验结果表明,改进的Parallel Sets可清晰展现国际恐怖主义数据库中各分类值间的关联,从而辅助用户分析不同恐怖组织的行为特征等信息;带降势的启发式分类值布局算法简单高效,适用于数据量较大、分类值较多的数据集。 |
关键词: 平行集 边交叉问题 恐怖主义 可视化分析 降势 |
DOI: |
投稿日期:2010-05-18 |
基金项目:国家自然科学基金资助项目(60903225);国防科技大学优秀研究生创新基金资助项目(B080503) |
|
Improvement of Parallel Sets and Its Application in Analyzing Global Terrorism Database |
XIAO Weidong, ZHOU Cheng, SUN Yang, GE Bin, TANG Daquan |
(C4ISR Technology Key LAB, National Univ. of Defense Technology, Changsha 410073, China)
|
Abstract: |
With terrorism aggravating, anti-terrorism has been a main task for national military security departments around the world. The current study utilized categorical data visualization, Parallel Sets, to analyze the relations among the multi-categories in Global Terrorism Database, aimed to uncover the implicit information within the data set. To alleviate the deficiency of excessive edge crossing brought by random layout of categorical values, the research proposed a heuristic layout algorithm based on average heuristic with cardinality reduction, which optimized the layout order of categories and the visual clutter is eased so that the cardinality reduction strategies can reduce the numbers of categories involved in computation. The experimental results demonstrate that the improved parallel sets can clearly express the association among
the multi-categories in Global Terrorism Database, thereby assist users in analyzing the information of various terrorist organizations, such as the behavior characteristics. Furthermore, the average-based heuristic with cardinality reduction is simple and highly efficient, which is suitable for large data sets with many categorical attributes. |
Keywords: parallel sets edge crossing terrorism visual analytics cardinality reduction |
|
|
|
|
|