[1]余 鑫,吴根秀*,蔡奥丽.基于双重中位数绝对偏差的证据加权组合方法[J].江西师范大学学报(自然科学版),2022,(05):523-532.[doi:10.16357/j.cnki.issn1000-5862.2022.05.13]
　YU Xin,WU Genxiu*,CAI Aoli.The Improved Evidence Weighted Combination Method Based on Double Median Absolute Detection[J].Journal of Jiangxi Normal University:Natural Science Edition,2022,(05):523-532.[doi:10.16357/j.cnki.issn1000-5862.2022.05.13]
点击复制

基于双重中位数绝对偏差的证据加权组合方法()
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:
2022年05期
页码:
523-532
栏目:
数学与应用数学
出版日期:
2022-09-25

文章信息/Info

Title:
The Improved Evidence Weighted Combination Method Based on Double Median Absolute Detection
文章编号:
1000-5862(2022)05-0523-10
作者:
余 鑫; 吴根秀*; 蔡奥丽
(江西师范大学数学与统计学院,江西 南昌 330022)
Author(s):
YU Xin; WU Genxiu*; CAI Aoli
(College of Mathematics and Statistics,Jiangxi Normal University,Nanchang Jiangxi 330022,China)
关键词:
D-S证据理论;  中位数绝对偏差;  高度冲突
Keywords:
Dempster-Shafer evidence theory;  median absolute deviation;  high conflict
分类号:
TP 391
DOI:
10.16357/j.cnki.issn1000-5862.2022.05.13
文献标志码:
A
摘要:
Dempster-Shafer证据理论目前已被广泛应用于大数据时代的各行各业,但是当该理论应用在高度冲突的证据源进行融合时,往往会产生一些有悖直觉的结果,具有一定的局限性.为了解决这一问题,该文提出了一种基于双重中位数绝对偏差(MAD)检测和一种新的证据加权组合改进方法.首先通过MAD算法检测出异常证据,再使用这组证据的平均值对异常证据进行修正,然后使用新的证据加权组合方法对证据加权平均后得到最终结果,最后通过随机模拟实验和具体的算例实验,并与其他几种经典的证据合成规则进行比较.仿真结果表明该文提出的方法能有效地解决高度冲突的证据融合问题.
Abstract:
Dempster-shafer evidence theory has been applicable in all walks of life of digital marketing in a large scale,but when it is applied to the integration of highly conflicting evidence sources,it often produces some counterintuitive results and has certain limitations.In order to solve this problem,the improved method based on double median absolute deviation detection and a new improved method of evidence weighted combination evidence weighted combination is proposed.Firstly,the abnormal evidence is detected by MAD algorithm,then the average value of this group of evidence is used to correct the abnormal evidence,and the new evidence weighting combination is applied to weight average the evidence to get the final count.Through random simulation experiments and example experiments,and compared with several other classical evidence synthesis rules,the simulation results show that the synthesis rules in this paper can effectively solve the problem of highly conflicting evidence fusion.

参考文献/References:

[1] DEMPSTER A P.Upper and lower probabilities induced by a multivalued mapping[J].Annals of Mathematical Statistics,1967,38(2):325-339.
[2] SHAFER G.A mathematical theory of evidence[M].Princeton:Princeton University Press,1976.
[3] 赵玲.多传感器信息融合技术及其应用[J].红外,2021,42(1):21-26.
[4] LEYS C,LEY C,KLEIN O,et al.Detecting outliers:do not use standard deviation around the mean,use absolute deviation around the median[J].Journal of Experimental Social Psychology,2013,49(4):764-766.
[5] FAN Chengli,SONG Yafei,LEi Lei,et al.Evidence reasoning for temporal uncertain information based on relative reliability evaluation[J].Expert Systems with Applications,2018,113:264-276.
[6] LI Junwei,XIE Baolin,JIN Yong,et al.Weighted conflict evidence combination method based on Hellinger distance and the belief entropy[J].IEEE Access,2020,8:225507-225521.
[7] XIONG Leihui,SU Xiaoyan,QIAN Hong.Conflicting evidence combination from the perspective of networks[J].Information Sciences,2021,580(2):408-418.
[8] 周国宪,伍星,刘韬.基于多传感器的神经网络和D-S证据理论在故障诊断中的应用[J].测试技术学报,2017,31(4):290-297.
[9] 姬晓飞,石宇辰,王昱,等.D-S理论多分类器融合的光学遥感图像多目标识别[J].电子测量与仪器学报,2020,34(5):127-132.
[10] ZADEH L A.The concept of a linguistic variable and its application to approximate reasoning:Part 1[J].Information Sciences,1975,8(3):199-249.
[11] YAGER R R.On the dempster-shafer framework and new combination rules[J].Information Sciences,1987,41(2):93-137.
[12] 孙全,叶秀清,顾伟康.一种新的基于证据理论的合成公式[J].电子学报,2000,28(8):117-119.
[13] 吴根秀.冲突证据组合方法[J].计算机工程,2005,31(9):151-154.
[14] MURPHY C K.Combining belief functions when evidence conflicts[J].Decision Support Systems,2000,29(1):1-9.
[15] 邓勇,施文康,朱振福.一种有效处理冲突证据的组合方法[J].红外与毫米波学报,2004,23(1):27-32.
[16] JOUSSELME A L,GRENIER D,BOSSÉ É.A new distance between two bodies of evidence[J].Information Fusion,2001,2(2):91-101.
[17] 刘康,何明浩,韩俊,等.基于本征向量和Jousselme距离的高冲突证据融合方法[J].系统工程与电子技术,2022,44(7):2175-2180.
[18] 严志军,陶洋.基于证据距离和不确定度的冲突数据融合算法[J].计算机工程与应用,2018,54(10):54-58,104.
[19] 佘玉萍.基于中位数的双MAD的离群值检测方法[J].廊坊师范学院学报(自然科学版),2016,16(2):25-27.
[20] 龚循强,张方泽,鲁铁定,等.基于中位数绝对偏差的异常训练样本探测方法[J].激光与光电子学进展,2020,57(23):173-178.
[21] HAMPEL F R.The influence curve and its role in robust estimation[J].Journal of the American Statistical Association,1974,69(346):383-393.
[22] HAMPEL F R,ROUSSEUW P J,RONCHETTI E M,et al.Robust statistics:the approach based on influence functions[J].Journal of the Royal Statistical Society Series D(The Statistician),1986,35(5):1-18.
[23] LIU Zhunga,LIU Yu,DEZERT J,et al.Evidence combination based on credal belief redistribution for pattern classification[J].IEEE Transactions on Fuzzy Systems,2019,28(4):618-631.
[24] LIU Jing,CHANG Chaowen,Zhang Yuchen,et al.A multisource situation information fusion method based on dynamic evidence combination[J].Mathematical Problems in Engineering,2021,2021:8834917.
[25] ZHANG Weiquan,DENG Yong.Combining conflicting evidence using the DEMATEL method[J].Soft Computing,2019,23(17):8207-8216.

相似文献/References:

[1]赖邦城,吴根秀.基于信息熵的证据融合方法[J].江西师范大学学报(自然科学版),2012,(05):519.
　LAI Bang-cheng,WU Gen-xiu.The Evidence Combination Method Based on Information Entropy[J].Journal of Jiangxi Normal University:Natural Science Edition,2012,(05):519.
[2]黄 梅,吴根秀,刘邱云,等.一种基于大焦元分解的信任函数逼近方法[J].江西师范大学学报(自然科学版),2016,40(03):285.
　HUANG Mei,WU Genxiu,LIU Qiuyun,et al.The Approximation Method of Belief Function Based on Decomposing Large Focal Elements[J].Journal of Jiangxi Normal University:Natural Science Edition,2016,40(05):285.

备注/Memo

备注/Memo:
收稿日期:2022-07-01
基金项目:国家自然科学基金(61876074)资助项目.
通信作者:吴根秀(1965—),女,江西南丰人,教授,主要从事不确定性推理与信息融合研究.E-mail:wgx_nc@sina.com
更新日期/Last Update: 2022-09-25