[1]王兆辉,刘邱云*,吴根秀,等.基于Pareto法则的BPA概率转换[J].江西师范大学学报(自然科学版),2023,(03):287-295.[doi:10.16357/j.cnki.issn1000-5862.2023.03.09]
 WANG Zhaohui,LIU Qiuyun*,WU Genxiu,et al.The Probability Transformation of BPA Based on Pareto Principle[J].Journal of Jiangxi Normal University:Natural Science Edition,2023,(03):287-295.[doi:10.16357/j.cnki.issn1000-5862.2023.03.09]
点击复制

基于Pareto法则的BPA概率转换()
分享到:

《江西师范大学学报》(自然科学版)[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2023年03期
页码:
287-295
栏目:
出版日期:
2023-05-25

文章信息/Info

Title:
The Probability Transformation of BPA Based on Pareto Principle
文章编号:
1000-5862(2023)03-0287-09
作者:
王兆辉刘邱云*吴根秀朱鸿祥
(江西师范大学数学与统计学院,江西 南昌,330022)
Author(s):
WANG Zhaohui LIU Qiuyun*WU Genxiu ZHU Hongxiang
(School of Mathematics and Statistics, Jiangxi Normal University, Nanchang Jiangxi 330022,China)
关键词:
基本概率分配函数 概率转换 帕累托法则 决策
Keywords:
basic probability assignment function probability transformation Pareto principle decision-making
分类号:
TP 391
DOI:
10.16357/j.cnki.issn1000-5862.2023.03.09
文献标志码:
A
摘要:
基于帕累托(Pareto)法则,该文认为复杂焦元信度的分配应该依赖于其单子命题的信度,信度大的更能决定复杂焦元的分配,信度为零的也能影响复杂焦元的分配.将复杂焦元分成2类结构:单子命题没有零值和单子命题有零值.对于前者,找出了复杂焦元的帕累托元素,只在帕累托元素上按其信度权重进行分配,这既避免了一些反常的情况又能更加突出信度大的单子命题.对于后者,采用忽略一部分或平均的办法来进行分配; 设置了2个参数,既能控制焦元分配的冒险程度,又能控制转换的效果.提出了一种能够根据给定冒险程度来计算基本概率分配函数概率转换的新方法.最后通过实例分析,验证了该方法是有效的.
Abstract:
Based on the Pareto Principle,it is believed that the distribution of belief value of complex focal elements should depend on the belief value of its singleton propositions.A larger belief value can better determine the distribution of complex focal elements,and a zero trust value can also affect the distribution of complex focal elements.The complex focal elements are divided into two types of structures,one is that the singleton proposition has no zero value,and the other is that the singleton proposition has zero value.For the former,the Pareto elements of the complex focal elements are found,and only the Pareto elements are allocated according to their belief value weights,which avoids some abnormal situations and highlights the singleton propositions with large belief values.For the latter,the method of ignoring a part or averaging is used to allocate,two parameters are set, which can control the risk degree of focal element allocation and adjust the effect of transformation.The new method is proposed to calculate the probability transition of basic probability assignment function according to the given risk degree. Finally, the effectiveness of the method is verified by example analysis.

参考文献/References:

[1] SHAFER G.A mathematical theory of evidence [M].Princeton: Princeton University Press, 1976.
[2] 韩德强,杨艺,韩崇昭.DS证据理论研究进展及相关问题探讨 [J].控制与决策,2014,29(1): 1-11.
[3] SMETS P, KENNES R.The transferable belief model [J].Artifial Intelligence,1994,66(2):191-234.
[4] SUDANO J J.Pignistic probability transforms for mixes of low- and high-probability events [J].Computer Science,2015(1):23-27.
[5] SUDANO J J.The system probability information content(PIC)relationship to contributing components, combining independent multi-source beliefs, hybrid and pedigree pignistic probabilities [C]∥Proceedings of the Fifth International Conference on Information Fusion,July 8-11,2022,Loews Annapolis Hotel, Annapolis, Maryland. New York: IEEE, 2002,2: 1277-1283.
[6] SUDANO J J.Equivalence between belief theories and Naive Bayesian fusion for systems with independent evidential data: part II, the example [C]∥Proceedings of the Sixth International Conference of Information Fusion,July 8-11,2003,Radisson Hotel,Cairns,Queensland. New York: IEEE,2003,2:1357-1364.
[7] SUDANO J J.Yet another paradigm illustrating evidence fusion(YAPIEF)[C]∥Proceedings of the 9th International Conference on Information Fusion,July 10-13,2006, Florence.New York: IEEE,2006:301783.
[8] 潘巍,王阳生,杨宏戟. Pignistic概率转换算法设计 [J].计算机工程, 2005,31(4):20-22,25.
[9] HAN Deqiang,DEZERT J,HAN Chongzhao,et al.Is entropy enough to evaluate the probability transformation approach of belief function? [C]∥Proceedings of the 13th International Conference on Information Fusion,July 26-29,2010, The University of Edinburgh,Edinburgh,Scotland. New York:IEEE,2010,4:37-43.
[10] 程子成,吴根秀,宋姝婷.基于融合信息熵性质的信任函数概率逼近 [J].江西师范大学学报(自然科学版), 2014, 38(5): 534-538.
[11] DEZERT J,SMARANDACHE F.A new probabilistic transformation of belief mass assignment [C]∥Proceedings of the 11th International Conference on Information Fusion, June 30-July 3,2008,University of Cologne, Karlsruhe,North Rhine-Westphalia. New York: IEEE,2008:1410-1417.
[12] SUDANO J J.Belief fusion, pignistic probabilities, and information content in fusing tracking attributes [C]∥Proceedings of the IEEE 2004 Radar Conference,April 26-29,2004, Wyndham Philadelphia at Franklin Plaza, Philadelphia, Pennsylvania.New York: IEEE, 2004:218-224.
[13] JIANG Wen,HUANG Chan,DENG Xinyang.A new probability transformation method based on a correlation coefficient of belief functions [J].International Journal of Intelligent Systems,2019,34(6):1337-1347.
[14] HUANG Chongru,MI Xiangjun,KANG Bingyi.Basic probability assignment to probability distribution function based on the Shapley value approach [J].International Journal of Intelligent Systems,2021,36(8):4210- 4236.
[15] DENG Zhan,WANG Jianyu.A novel decision probability transformation method based on belief interval [J].Knowledge-Based Systems,2020,208:106427.
[16] 周千里,邓勇.基于量子演化的信度函数概率转换 [J].航空学报,2022,43(S1):182-192.
[17] CHEN Luyuan, DENG Yong, CHEONG Kanghao.Probability transformation of mass function:a weighted network method based on the ordered visibility graph [J].Engineering Applications of Artificial Intelligence,2021,105:104438.

备注/Memo

备注/Memo:
收稿日期:2022-11-23
基金项目:国家自然科学基金(61876074)资助项目.
通信作者:刘邱云(1976—),女,江西铜鼓人,讲师,主要从事不确定性推理与信息融合方面的研究.E-mail:lqyxinxiang@126.com
更新日期/Last Update: 2023-05-25