Pattern recognition using interval-valued intuitionistic fuzzy set and its similarity degree

Pattern recognition under fuzzy environments is an interesting and important research topic which has been receiving more and more attention in recent years. Aiming at this kind of pattern recognition problems, fuzzy theories have been applied to the field widely and effectively. Especially interval-valued intuitionistic fuzzy sets (IVIFSs) can give not only a membership degree, but also a non-membership degree, which is more or less independent. Meanwhile the membership degree and non-membership degree are denoted by an interval which makes the IVIFSs can represent the dynamic character of features. Therefore in this paper, depending on IVIFSs and corresponding similarity degree (or distance measure) we construct a kind of novel pattern recognition approach. This approach chooses different weight for each feature according to its dissimilarity with other features. Thus the approach can show the corresponding influence and importance of different features. Finally, we utilize concrete examples to validate the proposed approach.