Improving Lower and Upper Approximation of Dominance based Rough Set Approach

Abstract

This is era of information and on daily basis we have to deal with the huge amount of data. Various tools and techniques are available for processing such huge volumes of data and Rough Set Theory (RST) is one of the most prominent tools for this purpose. However, it does not consider the preference order between the values of the attributes. Dominance Based Rough Set Approach (DRSA) is the one that provides dominance relation in this regard. In DRSA the lower and upper approximations are two measures that form basis of the most of the algorithms based on DRSA. However, computing these approximations is computationally so expensive that the algorithms using these measures may suffer serious performance bottlenecks while dealing with datasets beyond smaller size. In this paper, we have proposed a new approach to compute these measures. The proposed algorithm directly calculates approximations without considering the objects that do not play any role in approximations. The proposed approach was compared with the conventional method using ten benchmark datasets from UCI. Results have shown that the proposed approach significantly reduces the execution time. Average reduction in execution time was found to be 80%. The proposed approach also reduces the complexity (Big-O). This approach also reduces memory consumption up to 75%.