计算机科学 ›› 2016, Vol. 43 ›› Issue (12): 130-134.doi: 10.11896/j.issn.1002-137X.2016.12.023
薛安荣,毛文渊,王孟頔,陈泉浈
XUE An-rong, MAO Wen-yuan, WANG Meng-di and CHEN Quan-zhen
摘要： 传统的恐怖行为预测算法没有考虑到组织会改变其行为策略,而CAPE算法根据组织背景的改变预测行为变化,但其只能根据变化表中存在的背景变化预测行为。为了能根据任意背景变化预测恐怖行为,针对恐怖数据高维小样本的特点,提出了一种 利用贝叶斯方法在改进的变化表上预测组织行为的算法。利用贝叶斯方法可快速有效地解决高维小样本分类问题的特性,在改进的变化表上实现对组织行为的预测,从而提高了预测精度和计算效率。此外,考虑到背景的变化会在时间序列上对组织行为产生持续的影响,因此在不同时间滞差下,利用加权的贝叶斯方法预测组织行为。MAROB数据集上多个组织数据的实验结果也表明,所提算法在准确率及时间复杂度上优于CAPE算法。
| [1] Serra E,Subrahmanian V S.A Survey of Quantitative Models of Terror Group Behavior and an Analysis of Strategic Disclosure of Behavioral Models[J].IEEE Transactions on Computational Social Systems,2014,1(1):66-88 [2] Subrahmanian V S.Handbook of Computaional Approaches toCounterterrorism[M].New York:Springer Press,2013:99-268 [3] Xue An-rong,Wang Wei,Zhang Ming-cai.Terrorist Organization Behavior Prediction Algorithm Based on Context Subspace[M]∥Advanced Data Mining and Applications.Springer Berlin Heidelberg,2011:332-345 [4] Li Xiao-chen,Mao Wen-ji,Zeng D,et al.Performance evaluation of machine learning methods in cultural modeling[J].Journal of Computer Science and Technology,2009,24(6):1010-1017 [5] Subrahmanian V S.Cultural Modeling in real time[J].Science,2007,317(5844):1509-1510 [6] Khuller S,Martinez V,Nau D,et al.Finding Most ProbableWorlds of Probabilistic Logic Programs[C]∥1st International Conference on Scalable Uncertainty Management.Washington DC:IEEE Press,2007:45-57 [7] Subrahmanian V S,Albanese M,Martinez M V,et al.CARA:a cultural adversarial reasoning architecture[J].IEEE Intel Syst,2007,2(2):12-16 [8] Martinez M V,Simari G I,Sliva A,et al.CONVEX:context vectors as a similarity-based paradigm for forecasting group beha-viors[J].IEEE Intel Syst,2008,23(4):51-57 [9] Martinez M V,Sliva A,Simari G I,et al.CAPE:automatically predicting changes in group behavior[C]∥Mathematical Metho-ds in Counterterrorism.Springer Vienna,2009:253-269 [10] Victor A,Pate A,Wilkenfeld J.Minorities at risk organizationbehavior data and codebook version 9[EB/OL].http://www.cidcm.umd.edu/mar [11] Guo Hong-yu.Research on weighted feature selection algorithm based on information entropy theory[J].Computer Engineering and Applications,2013,49(10):140-146(in Chinese) 郭红钰.基于信息熵理论的特征权重算法研究[J].计算机工程与应用,2013,49(10):140-146 [12] Liu Hua-wen.Research on feature selection algorithm based on information entropy[D].Changchun:Jilin University,2010(in Chinese) 刘华文.基于信息熵的特征选择算法研究[D].长春:吉林大学,2010 [13] He Qing,Li Ning,Luo Wen-juan,et al.Summary of machinelearning algorithms in large data[J].Pattern Recognition and Artificial Intelligence,2014,27(4):327-336(in Chinese) 何清,李宁,罗文娟,等.大数据下的机器学习算法综述[J].模式识别与人工智能,2014,27(4):327-336 [14] Han Jia-wei,Kamber M.数据挖掘:概念与技术(3rd ed)[M].范明,等译.北京:机械工业出版社,2012 |
| No related articles found! |
|
||
首页