A New Support Value Method Filtering Based on Object Support Partition for Soft Reduction

Abstract

Soft set time complexity is become really a problem when the numbers of parameters are increased. In order to solve time complexity problem, it necessary to reduce the boundary of optimal soft set growth and due to this the time cost can be enhanced. Several soft set methods are determining the soft set reduction but in performing the reduction it spends more time to produce the result and this happens because the false candidate sets are a part of solution. So, if the boundary of candidate reduction is narrowed then the reduction process will speed up. In this paper, we proposed a new method which reducing the boundary of candidate reduction using Lipschitz constant and wavelet discrete transform to eliminate large false sets from the solution. In Lipschitz constant function the value of candidate implies are determined, where based on wavelet WDT the false sets which is not in the form of implies also can be determined. The proposed method remove an inconsistency noise from the soft set in a pre-processing filtering based on if then method which help to classify further reduction in short time. It found that by using Lipschitz constant function and wavelet discrete transform the reduction time can be enhanced several times comparing to previous reduction methods. The result indicates that Lipschitz constant function and wavelet WDT algorithm. It complements each other to determine candidate soft set reduction.