摘要
概率转换是不确定推理中的有效决策方法之一。首先,对现有概率转换公式进行分析,给出两类统一的概率转换公式。针对现有概率转换公式的给出过程较为主观,概率转换方法缺乏严密性的缺陷,提出基于反馈证据冲突度的概率转换方法。再利用遗传算法,求解最小反馈证据冲突度的转换概率。最后,将本文概率转换方法与现有的概率转换方法进行比较,结合仿真分析表明所提方法的有效性。
Probability transformation is one of the most effective decision-making methods in the field of uncertainty reasoning.In this research,the existing probability transformation formulas are analyzed,and two kinds of unified probability transformation formulas are then introduced.In order to make up for the deficiencies that the present process of giving the probability transformation formulas is subjective and the probability transformation methods are imprecise,aprobability transformation method based on the conflict degree of feedback evidence is thus proposed.Furthermore,the genetic algorithm is used to solve the probability transformation formula of the minimum feedback conflicts.Results of simulation analysis show that the proposed method is more reasonable and effective compared with the existing methods.
作者
关欣
刘海桥
衣晓
赵静
GUAN Xin;LIU Haiqiao;YI Xiao;ZHAO Jing(Department of Electronic and Information Engineering,Navel Aeronautical Engineering Academy,Yantai 264001,China)
出处
《系统工程与电子技术》
EI
CSCD
北大核心
2018年第7期1436-1442,共7页
Systems Engineering and Electronics
基金
国家自然科学基金重大科研计划(91538201)
国家自然科学基金(61671463)
国防科技卓越青年人才基金
泰山学者工程专项经费(ts201712072)资助课题
关键词
概率转换
不确定推理
决策规则
反馈证据冲突度
遗传算法
probability transformation
uncertainty reasoning
decision-making rule
conflict degree of feedback evidence
genetic algorithm
