基于两维识别框架的证据分组合成研究 本期目录 >>
Title: Grouping method for characteristics-based combination of evidences under 2-dimensional frame of discernment
作者 朱卫东;王凯;吴勇;孙宜博
Author(s): Zhu weidong; Wang kai; Wu yong; Sun yibo
摘要: 基本可信度数值之间的关系有时不能反映证据之间实际的联系,仅仅依靠这种数值关系来修正证据可能会出现不符实际的结论,同时也会降低合成结果的可解释性。针对该问题,面向具体决策问题提出了一种基于证据源特征的证据分组合成的方法:在经典证据框架的基础上增加了反映证据源特征的第二维信息框架,并基于第二维框架确定证据分组标准和证据修正系数,利用不同的合成规则对证据组内部和外部的信度进行修正和合成。最后,以证券市场上多位证券分析师预测信息融合的实际算例,验证了该方法的科学性和合理性。
Abstract: Academicians have done many intensive researches of the Dempster-Shafer Evidence Theory, abstractly and practically, and these researches have lead to very enriched and credible research findings, and these findings have contributed the impetus functions of the evidence theory development. Because the relationship between the basic reliable numerical values sometimes cannot reflect the actual connection between the evidences, it can also lead to neglecting of the correlation between economic and technical evidence. Therefore, if just relying on the relationship between the numerical values to modify the evidence, can lead to the erroneous results, meanwhile, it can also decrease the interpretability and expansibility of results. In order to solve this problem, a new and concrete solution of the issue, which is based on evidence source feature, is proposed: on account of the classical evidence framework which the second dimension information frame is added, and by using the second dimension framework the evidence grouping criterion and evidence correction factor is confirmed, where the different synthetic rules is used to compose the internal and external credibility of the evidence group. The second dimensional evidence framework includes not only the support of the evidence source to the focal element, but also the characteristics of the source itself, which can increase the information content of the framework. Above all, according to the issues of information fusion that analysts of security market has forecast, the securities analysts’ information concerning stock rating and earnings forecast are used to construct a two-dimension progressive evidence framework. Using the method of evidence grouping to fuse the information, which is based on the two dimensional recognition framework, and comparing of the results which is found by using the traditional methods of information fusion, classical Dumpster’s synthesis method, and average evidence synthesis method, the proposed method can manage the conflicting information harmoniously; it can also improve the accuracy of information fusion.
关键词: D-S证据理论;两维证据框架;证据分组合成方法;证据源特征
Keywords: Dempster-Shafer evidence theory; two-dimensional frames of discernment; grouping method for combining evidence; characteristics of evidence
基金项目: 国家自然科学基金项目;国家自然科学基金项目;安徽省软科学研究计划项目;安徽省自然科学基金项目
发表期数: 2018年 第3期
中图分类号: 文献标识码: 文章编号:
参考文献/References:

[1] Ben Abdallah N, Mouhous-Voyneau N, Denœux T. Combining statistical and expert evidence using belief functions: Application to centennial sea level estimation taking into account climate change[J]. International Journal of Approximate Reasoning. 2014, 55(1): 341-354.

[2] Yanga G, Yangc J, Liud W, et al. Cross-efficiency Aggregation in DEA Models using the Evidential-Reasoning Approach[J]. European Journal of Operational Research. 2013.

[3] Xu D. An introduction and survey of the evidential reasoning approach for multiple criteria decision analysis[J]. Annals of Operations Research. 2012, 195(1): 163-187.

[4] Beynon M J. The Role of the DS/AHP in Identifying Inter-Group Alliances and Majority Rule Within Group Decision Making[J]. Group Decision and Negotiation. 2006, 15(1): 21-42.

[5] Feng T, Mi J, Zhang S. Belief functions on general intuitionistic fuzzy information systems[J]. Information Sciences. 2014.

[6] 齐照辉,刘雪梅,梁伟. 基于证据理论的导弹对抗防御雷达作战效能评估方法[J]. 系统工程理论与实践. 2010(01): 173-177.

[7] Shafer G. A mathematical theory of evidence[M]. Princeton:Princeton university press, 1976.

[8] 何友,胡丽芳,关欣,等. 一种度量广义基本概率赋值冲突的方法[J]. 中国科学:信息科学. 2011(08): 989-997.

[9] Yang J, Xu D. Evidential reasoning rule for evidence combination[J]. Artificial Intelligence. 2013, 205: 1-29.

[10] 周剑,朱耀琴,柴旭东,等. 基于云模型与证据理论的共识分析方法[J]. 系统工程理论与实践. 2012(12): 2756-2763.

[11] Smets P. Decision making in the TBM: the necessity of the pignistic transformation[J]. International Journal of Approximate Reasoning. 2005, 38(2): 133-147.

[12] Smarandache F. Unification of fusion theories (UFT)[J]. Advances and Challenges in Multisensor Data and Information Processing. 2007, 8: 114.

[13] 杨国梁,李晓轩,孟溦. 基于区间数证据推理方法的用户满意度调查[J]. 管理工程学报. 2012(01): 27-34.

[14] Murphy C K. Combining belief functions when evidence conflicts[J]. Decision support systems. 2000, 29(1): 1-9.

[15] 杨善林,朱卫东,任明仑. 基于可变参数优化的相关证据合成方法研究[J]. 管理科学学报. 2003(05): 12-16.

[16] Yang J B, Wang Y M, Xu D L, et al. The evidential reasoning approach for MADA under both probabilistic and fuzzy uncertainties[J]. European Journal of Operational Research. 2006, 171(1): 309-343.

[17] Xu D, Yang J, Wang Y. The evidential reasoning approach for multi-attribute decision analysis under interval uncertainty[J]. European Journal of Operational Research. 2006, 174(3): 1914-1943.

[18] Yang J, Xu D. Evidential reasoning rule for evidence combination[J]. Artificial Intelligence. 2013, 205: 1-29.

[19] 陈圣群,王应明. 证据的分组合成法[J]. 控制与决策. 2013(04): 574-578.

[20] 朱卫东,张洪涛,张晨,等. 基于两维语义的科学基金立项评估方法[J]. 系统工程理论与实践. 2012(12): 2697-2703.

[21] 朱卫东,周光中,杨善林. 基于二维语言评价信息的群体决策方法[J]. 系统工程. 2009(02): 113-118.

[22] 张洪涛,朱卫东,王慧,等. 多维框架证据推理的科研项目立项评估方法[J]. 科研管理. 2013(06): 122-128.

[23] Liu W. Analyzing the degree of conflict among belief functions[J]. Artificial Intelligence. 2006, 170(11): 909-924.

[24] Ristic B, Smets P. The TBM global distance measure for the association of uncertain combat ID declarations[J]. Information Fusion. 2006, 7(3): 276-284.

[25] 曹胜,朱红军. 王婆贩瓜:券商自营业务与分析师乐观性[J]. 管理世界. 2011(07): 20-30.

[26] Affleck Graves J, Davis L R, Mendenhall R R. Forecasts of earnings per share: Possible sources of analyst superiority and bias[J]. Contemporary Accounting Research. 1990, 6(2): 501-517.

[27] 黄静,董秀良. 证券分析师业绩预测和投资评级准确性实证分析[J]. 数理统计与管理. 2006(06): 742-749.

[28] 王蕾. 证券分析师荐股评级准确性研究[D]. 合肥工业大学, 2013.