摘要
基于模糊相似量的模式识别是通过样本数据与标本数据之间的相似度程度来判断该样本属于哪一种模式,而交叉熵(cross Entropy)是Shannon信息论中一个重要概念,主要用于度量两个概率分布问的差异性信息.在模糊相似度量和交叉熵的基础上,提出了基于交叉熵的模糊相似度量,并对交叉熵中存在的分母为零的情况做了改进,通过在模式识别中的应用,验证了这种方法的可行性及有效性.
Based on fuzzy similarity measure, the pattern recognition determines that the sample belongs to a model by the degree of similarity of the sample data and samples of data. And the cross - entropy is an important concept in Shannon's information theory, mainly for measuring the difference between two probability distribution of the informa- tion. We proposed the fuzzy similarity measure of cross - entropy on the basis of cross - entropy and the similarity measure of fuzzy sets. As a result the problem that the denominator of cross - entropy is zero is improved. The final adoption of pattern recognition in applying this method demonstrated its feasibility and effectiveness. Besides, the method of the fuzzy similarity measure on the basis of cross entropy to pattern recognition is simple, easy to understand.
出处
《绍兴文理学院学报》
2008年第9期23-25,共3页
Journal of Shaoxing University
关键词
交叉熵
模糊相似性测度
模式识别
cross entropy
fuzzy similarity measures
pattern recognition