The decision theoretic rough set model was introduced in the 90’s in order to loosen the restrictions of conventional rough approximations. Following this the conventional rough c-means was extended to the decision theoretic rough set context by Li et al. 2014. However, the Euclidean distance used as the similarity measure in this paper had the property of separability and to rectify this problem Kernel measures were used to develop the Kernel based decision theoretic rough C-means by Ryan et al. in 2016. As it is known, the hybrid models are more efficient than the individual models this approach was further extended and the Kernel based decision theoretic rough Fuzzy C-means was introduced by them recently in 2016. As a model of uncertainty intuitionistic fuzzy sets are more general than the fuzzy sets, So, we use the intuitionistic fuzzy sets instead of fuzzy sets and introduce a Kernel based Decision theoretic rough intuitionistic Fuzzy C-means in this paper. To provide variety and measure the effects, we have selected three of the most popular kernels; the Radial Basis, the Gaussian and the hyperbolic tangent kernels in our model. For the experimentation purpose we use three datasets namely the Iris, the wine and the glass data sets from the UCI repository. The efficiency measuring indices DB and D are used for evaluating the relative efficiencies of this algorithm and the other algorithms in this direction.Our results show that the proposed model provides improved results than the other two models. Some diagrams are presented to show the results visually
View more info for "An Analysis of Decision Theoretic Kernalized Rough Intuitionistic Fuzzy C-Means"
|Journal||Data powered by TypesetAdvances in Intelligent Systems and Computing Proceedings of Sixth International Conference on Soft Computing for Problem Solving|
|Publisher||Data powered by TypesetSpringer Singapore|