论文部分内容阅读
离散信息在专家系统、模式识别、决策分析等领域普遍存在,为了解决这类信息融合问题,提出一种离散证据推理方法.首先,将每个离散证据拆分成一类单点值证据;然后,以冲突最小化为目标修正类内证据,并采用证据推理进行组合;最后,以同样的方法对类间证据进行修正与组合.所提出方法不仅可以解决离散证据的内外部冲突问题,而且能够克服运算量过大的问题.算例分析表明了所提出的方法是合理且有效的.
Discrete information is widely used in expert systems, pattern recognition, decision analysis and other fields. In order to solve this kind of information fusion problem, a discrete evidence reasoning method is proposed. First, each discrete evidence is split into a single type of point evidence. Finally, using the same method to correct and combine the evidences between classes, the proposed method can not only solve the problem of internal and external conflicts of discrete evidence, but also overcome Computational complexity is too large.An example analysis shows that the proposed method is reasonable and effective.