一种基于ISODATA聚类和改进相似度的证据推理方法

李新德; 王丰羽

doi:10.16383/j.aas.2015.c140543

一种基于ISODATA聚类和改进相似度的证据推理方法

doi: 10.16383/j.aas.2015.c140543

李新德^,,
王丰羽

1.
东南大学自动化学院复杂工程测量与控制教育部重点实验室南京 210096

基金项目:

国家自然科学基金 (60804063, 61175091), 江苏省 “青蓝工程” 资助计划, 航空基金 (20140169002), 江苏省 “六大高峰人才” 资助计划,江苏高校优势学科建设工程资助项目资助

详细信息

作者简介:
王丰羽东南大学自动化学院硕士研究生.主要研究方向为信息融合和不确定推理. E-mail: treexiaohei@163.com

通讯作者:
李新德东南大学自动化学院副教授.主要研究方向为智能机器人, 人机交互, 机器感知, 信息融合, 不确定推理和机器视觉.本文通信作者. E-mail: xindeli@seu.edu.cn

计量
- 文章访问数: 1575
- HTML全文浏览量: 70
- PDF下载量: 943
- 被引次数: 0
出版历程
- 收稿日期: 2014-07-18
- 修回日期: 2014-09-27
- 刊出日期: 2015-03-20

A Method of Evidence Reasoning Based on ISODATA Clustering and Improved Similarity Measure

LI Xin-De^,,
WANG Feng-Yu

1.
Key Laboratory of Measurement and Control of Complex Systems of Engineering of Ministry of Education, School of Automation, Southeast University, Nanjing 210096

Funds:

Supported by National Natural Science Foundation of China (60804063, 61175091), Qing Lan Project of Jiangsu Province, Aeronautical Science Foundation of China (20140169002), Six Major Top-talent Plan of Jiangsu Province, and A Project Funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions

摘要

摘要: 针对智能信息处理中Dempster组合规则不能处理高度冲突的问题, 从内、外证据不确定性分析的角度深入揭示了证据冲突产生的原因, 即证据的冲突性不仅仅根源于证据间的矛盾, 也与证据自身的不确定性密切相关, 提出了一种同时考虑证据自冲突和外部冲突的相似性测度, 然后利用新测度计算证据的众信度, 对证据源进行修正;与此同时, 根据原始证据间的聚类特性, 利用迭代自组织数据分析技术(Iterative selforganizing data analysis techniques algorithm, ISODATA)聚类方法进行聚类, 然后利用Dempster组合规则合成每一聚类中所有证据为证据代表, 并综合众信度和证据在该聚类的频度计算可靠度, 最后, 利用统一组合规则合成证据代表.并通过大量的算例, 同其他方法和自身改进前后进行深入比较, 优势比较明显, 有效地解决了冲突证据合成出现的问题.
- 证据推理 /
- 冲突 /
- 聚类 /
- 相似测度 /
- 组合规则
Abstract: For the problem that the highly conflictive evidences can not be processed by Dempster rule in intelligent information processing, the reason of conflict occurrence is further revealed by analysis of uncertainty among evidences and evidences themselves. A similarity measure by considering self-conflict and external conflict of evidence together is proposed and used to compute the commonality of evidence. All initial evidences can be revised according to the commonality. At the same time, according to the clustering characteristic of initial evidences, they are clustered by applying ISODATA (Iterative selforganizing data analysis techniques algorithm), and then, all evidences in each cluster are combined as a representative of evidences by applying Dempster rule. The reliability of the representative of evidence is computed by considering the credibility and frequency of occurrence of the evidence in a cluster together. Finally, these representatives of evidences are combined by using the unified combination rule. Several numerical examples are given to deeply compare this method with others and itself before improving its similarity measure. The result shows that the new method has a distinct advantage over others and can effectively solve the problem of combination of conflictive evidences.
- Evidence reasoning /
- conflict /
- clustering /
- similarity measure /
- combination rule

HTML全文

参考文献(37)

[1]	Shafer G. A Mathematical Theory of Evidence. Princeton: Princeton University Press, 1976.
[2]	[2] Zadeh L A. Review of books: a mathematical theory of evidence. AI Magazine, 1984, 5(3): 81-83
[3]	[3] Haenni R. Are alternatives to Dempster's rule of combination real alternatives. Information Fusion, 2002, 3(4): 237-239
[4]	Bian Zhao-Qi, Zhang Xue-Gong. Pattern Recognition. Beijing: Tsinghua University Press, 2000. (边肇祺, 张学工. 模式识别. 北京: 清华大学出版社, 2000.)
[5]	[5] Martin A, Osswald C. Toward a combination rule to deal with partial conflict and specificity in belief functions theory. In: Proceedings of the 10th International Conference on Information Fusion. Quebec, Canada: IEEE, 2007. 1-8
[6]	[6] Lefevre E, Colot O, Vannoorenberghe P. Belief function combination and conflict management. Information Fusion, 2002, 3(2): 149-162
[7]	[7] Yager R R. On the D-S framework and new combination rules. Information Sciences, 1987, 41(2): 93-138
[8]	[8] Liu W R. Analyzing the degree of conflict among belief functions. Artificial Intelligence, 2006, 170(11): 909-924
[9]	Wang Dong, Li Qi, Jiang Wen, Xu Xiao-Bin, Deng Yong. New method to combine conflict evidence based on pignistic probability distance. Infrared and Laser Engineering, 2009, 38(1): 149-154 (王栋, 李齐, 蒋雯, 徐晓滨, 邓勇. 基于pignistic概率距离的冲突证据合成方法. 红外与激光工程, 2009, 38(1): 149-154)
[10]	Deng Yong, Wang Dong, Li Qi, Zhang Ya-Juan. A new method to analyze evidence conflict. Control Theory Applications, 2011, 28(6): 839-844(邓勇, 王栋, 李齐, 章雅娟. 一种新的证据冲突分析方法. 控制理论与应用, 2011, 28(6): 839-844)
[11]	Quan Wen, Wang Xiao-Dan, Wang Jian, Zhang Yu-Xi. New combination rule of DST based on local conflict distribution strategy. Acta Electronica Sinica, 2012, 40(9): 1880-1884(权文, 王晓丹, 王坚, 张玉玺. 一种基于局部冲突分配的DST组合规则. 电子学报, 2012, 40(9): 1880-1884)
[12]	He Bing, Hu Hong-Li. A modified DS evidence combination strategy. Acta Aeronautica et Astronautica Sinica, 2003, 24(6): 559-562(何兵, 胡红丽. 一种修正的DS 证据融合策略. 航空学报, 2003, 24(6): 559-562)
[13]	Yang J B, Xu D L. Evidential reasoning rule for evidence combination. Artificial Intelligence, 2013, 205: 1-29
[14]	Smets P. Analyzing the combination of conflicting belief functions. Information Fusion, 2007, 8(4): 387-412
[15]	Han De-Qiang, Han Chong-Zhao, Deng Yong, Yang Yi. Weighted combination of conflicting evidence based on evidence variance. Acta Electronica Sinica, 2011, 39(3A): 153-157(韩德强, 韩崇昭, 邓勇, 杨艺. 基于证据方差的加权证据组合. 电子学报, 2011, 39(3A): 153-157)
[16]	Hu Li-Fang, Guan Xin, Deng Yong, He You. Cause-analysis for conflicting evidences in the generalized power space. Control Theory and Applications, 2011, 28(12): 1717-1722(胡丽芳, 关欣, 邓勇, 何友. 广义幂集空间中证据冲突的原因分析. 控制理论与应用, 2011, 28(12): 1717-1722)
[17]	Smarandache F, Dezert J. Advances and Applications of DSmT for Information Fusion. Rehoboth, USA: American Research Press, Vol.2, 2006
[18]	Xiong Yan-Ming, Yang Zhan-Ping, Qu Xin-Fen. Novel combination method of conflict evidence based on evidential model modification. Control and Decision, 2011, 26(6): 883-887(熊彦铭, 杨战平, 屈新芬. 基于模型修正的冲突证据组合新方法. 控制与决策, 2011, 26(6): 883-887)
[19]	Lu Zheng-Cai, Qin Zheng. General framework for evidence combination and its approach to highly conflicting evidence fusion. Journal of Tsinghua University (Science and Technology), 2011, 51(11): 1701-1705, 1716(卢正才, 覃征. 证据合成的一般框架及高度冲突证据合成方法. 清华大学学报(自然科学版), 2011, 51(11): 1701-1705, 1716)
[20]	Li Wen-Li, Guo Kai-Hong. Combination rules of D-S evidence theory and conflict problem. Systems Engineering-Theory and Practice, 2010, 30(8): 1422-1432(李文立, 郭凯红. D-S 证据理论合成规则及冲突问题. 系统工程理论与实践, 2010, 30(8): 1422-1432)
[21]	Wang Lian-Feng, Song Jian-She, Zhu Yu, Cao Ji-Ping. Evidence combination based on fuzzy clustering analysis. Systems Engineering and Electronics, 2013, 35(1): 113-119(王连锋, 宋建社, 朱昱, 曹继平. 基于模糊聚类分析的证据组合. 系统工程与电子技术, 2013, 35(1): 113-119)
[22]	Quan Wen, Wang Xiao-Dan, Shi Zhao-Hui. Conflict evidence combination strategy based on evidence classification. Journal of Air Force Engineering University (Natural Science), 2012, 13(3): 80-84(权文, 王晓丹, 史朝辉. 基于证据分类的冲突证据合成方法. 空军工程大学学报(自然科学版), 2012, 13(3): 80-84)
[23]	Wang Jin-Hua, Wu Di, Cao Jie, Li Jun. Weighted combination of conflicting evidence based on evidence classification. Computer Science, 2013, 40(1): 247-250(王进花, 吴迪, 曹洁, 李军. 基于证据分类的加权冲突证据组合. 计算机科学, 2013, 40(1): 247-250)
[24]	Jousselme A L, Maupin P. Distances in evidence theory: comprehensive survey and generalizations. International Journal of Approximate Reasoning, 2012, 53(2): 118-145
[25]	Jousselme A L, Grenier D, Boss . A new distance between two bodies of evidence. Information Fusion, 2001, 2(2): 91-101
[26]	Li X, Dezert J, Smarandache F. Evidence supporting measure of similarity for reducing the complexity in information fusion. Information Sciences, 2011, 181(10): 1818-1835
[27]	Smets P. Data fusion in the transferable belief model. In: Proceedings of the 3rd International Conference on Information Fusion. Paris, France: IEEE, 2000. 21-33
[28]	Wen C L, Wang Y C, Xu X B. Fuzzy information fusion algorithm of fault diagnosis based on similarity measure of evidence. In: Proceedings of the 5th international symposium on Neural Networks: Advances in Neural Networks, Part II. Berlin, Heidelberg: Springer, 2008. 506-515
[29]	Liu Z, Dezert J, Pan Q, Mercier G. Combination of sources of evidence with different discounting factors based on a new dissimilarity measure. Decision Support Systems, 2011, 52(1): 133-141
[30]	Gu Zhao-Sheng. The Research and Application of Compositive Mathematical Methods to Water Quality Analysis, Simulation and Forecast in Reservoirs and Lakes [Ph.D. dissertation], Jilin University, China, 2006. (谷照升. 水库湖泊水质分析、模拟与预测的综合数学方法及其应用[博士学位论文]. 吉林大学, 中国, 2006.)
[31]	Sun Quan, Ye Xiu-Qing, Gu Wei-Kang. A new combination rules of evidence theory. Acta Electronica Sinica, 2000, 28(8): 117-119(孙全, 叶秀清, 顾伟康. 一种新的基于证据理论的合成公式. 电子学报, 2000, 28(8): 117-119)
[32]	Deng Yong, Shi Wen-Kang. A modified combination rule of evidence theory. Journal of Shanghai Jiaotong University, 2003, 37(8): 1275-1278(邓勇, 施文康. 一种改进的证据推理组合规则. 上海交通大学学报, 2003, 37(8): 1275-1278)
[33]	Li Bi-Cheng, Wang Bo, Wei Jun, Qian Ceng-Bo, Huang Yu-Qi. An efficient combination rule of evidence theory. Journal of Data Acquisition and Processing, 2002, 17(1): 33-36(李弼程, 王波, 魏俊, 钱曾波, 黄玉琪. 一种有效的证据理论合成公式. 数据采集与处理, 2002, 17(1): 33-36)
[34]	Li Xin-De, Dezert J, Huang Xin-Han, Meng Zheng-Da, Wu Xue-Jian. A fast approximate reasoning method in hierarchical DSmT (A). Acta Electronica Sinica, 2010, 38(11): 2566-2572(李新德, Dezert J, 黄心汉, 孟正大, 吴雪建. 一种快速分层递阶DSmT近似推理融合方法(A). 电子学报, 2010, 38(11): 2566-2572)
[35]	Li Xin-De, Yang Wei-Dong, Wu Xue-Jian, Dezert J. A fast approximate reasoning method in hierarchical DSmT (B). Acta Electronica Sinica, 2011, 39(3A): 31-36(李新德, 杨伟东, 吴雪建, Dezert J. 一种快速分层递阶DSmT近似推理融合方法(B). 电子学报, 2011, 39(3A): 31-36)
[36]	Li Xin-De, Yang Wei-Dong, Dezert J. An airplane image target's Multi-feature fusion recognition method. Acta Automatica Sinica, 2012, 38(8): 1298-1307(李新德, 杨伟东, Dezert J. 一种飞机图像目标多特征信息融合识别方法. 自动化学报, 2012, 38(8): 1298-1307)
[37]	Li Xin-De, Pan Jin-Dong, Dezert J. A target recognition algorithm of sequential aircrafts based on DSmT and HMM. Acta Automatica Sinica, 2014, 40(12): 2862-2876 (李新德, 潘锦东, Dezert J. 一种基于DSmT和HMM的序列飞机目标识别算法. 自动化学报, 2014, 40(12): 2862-2876)

施引文献

资源附件(0)

访问统计

点击查看大图

计量

文章访问数: 1575
HTML全文浏览量: 70
PDF下载量: 943
被引次数: 0

姓名
邮箱
手机号码
标题
留言内容
验证码

留言板

一种基于ISODATA聚类和改进相似度的证据推理方法

doi: 10.16383/j.aas.2015.c140543

作者简介:
王丰羽东南大学自动化学院硕士研究生.主要研究方向为信息融合和不确定推理. E-mail: treexiaohei@163.com

通讯作者:
李新德东南大学自动化学院副教授.主要研究方向为智能机器人, 人机交互, 机器感知, 信息融合, 不确定推理和机器视觉.本文通信作者. E-mail: xindeli@seu.edu.cn

计量

A Method of Evidence Reasoning Based on ISODATA Clustering and Improved Similarity Measure

计量

目录

留言板

一种基于ISODATA聚类和改进相似度的证据推理方法

doi: 10.16383/j.aas.2015.c140543

作者简介: 王丰羽 东南大学自动化学院硕士研究生.主要研究方向为信息融合和不确定推理. E-mail: treexiaohei@163.com

通讯作者: 李新德 东南大学自动化学院副教授.主要研究方向为智能机器人, 人机交互, 机器感知, 信息融合, 不确定推理和机器视觉.本文通信作者. E-mail: xindeli@seu.edu.cn

计量

出版历程

A Method of Evidence Reasoning Based on ISODATA Clustering and Improved Similarity Measure

计量

出版历程

目录

作者简介:
王丰羽东南大学自动化学院硕士研究生.主要研究方向为信息融合和不确定推理. E-mail: treexiaohei@163.com

通讯作者:
李新德东南大学自动化学院副教授.主要研究方向为智能机器人, 人机交互, 机器感知, 信息融合, 不确定推理和机器视觉.本文通信作者. E-mail: xindeli@seu.edu.cn