基于包含度的Vague集相似度量

在模糊模式识别中经常要根据最大相似度原理来分辨待测样品属于哪种模式．由于现有的vague集相似度量公式都是基于距离测度的，因此只要vague集间距离测度一样，它们就无法分辨，因此非常有必要寻找其它的相似度量计算方法．首先将模糊集上的包含度概念扩展到Vague集上，指出Vague集相似度量可以由包含度诱导出，然后给出一组新的Vague集相似度量计算公式．数值算例证明它们是有效的，最后将它们与现有方法进行比较，发现它们各有所长．

Similarity Measures of Vague Sets Based on Inclusion Grades

In fuzzy pattern recognition, it is often to classify the sample according to the principle of maximum membership degree. However, most of the existing methods are based on distance measures, they can not work well when the vague sets have equal distance to the sample. It is very necessary to find out other methods to calculate the degree of similarity between them. At first, the concept of inclusion grades is generalized to vague sets in this paper. It is proved that the similarity measure of vague sets can be induced by corresponding vague inclusion grades. Some new similarity measures are proposed. Numerical examples show that they are effectual. Compared with existing methods, it is found that they have their strong point respectively.