Obtaining Approximation using Map Reduce by Comparing Inter-Tables

Abstract

Size of the data increasing day by day because of digital world at unpredictable rate. Basically size of raw data is increasing so deal with such rough data is the challenging task and we need to acquire knowledge from such a colossal data. Number of techniques are used to retrieve knowledge from raw data like genetic algorithm, fuzzy sets and rough set. Rough set is very popular method and basically depends upon approximation i.e. lower approximation, upper approximation, boundary region. Hence, the effective computation of approximation is important pace in improving the performance of rough set. There are number of ways to calculate this approximation. In our system we have calculated rough approximation independent of each other. This can be achieved by dividing input dataset, so that we can also reduce number of comparison. The division of dataset based on decision attribute in the dataset. In our paper, we have explained a new method for computing rough set approximation. Using map-reduce we can deal with massive data and able to compute rough approximation for massive dataset.