抽象的な

Data Mining with Rough Set Using Map- Reduce

Prachi Patil

A colossal data mining and knowledge discovery exhibits a great challenge with the volume of data growing at an unpredicted rate. Different techniques used to retrieve meaningful data. Rough set is one of them. This method is based on lower approximation, upper approximation. Existing method calculates rough set approximation in serial way. Therefore we propose a parallel method. Map-Reduce has developed to manage many large-scale computation. Recently introduced Map-Reduce technique has received much consideration from both scientific community and industry for its applicability in big data analysis. The effective computation of approximation is essential step in improving the performance of rough set. For mining the massive data, parallel computing modes, algorithms and different methods get used in research fields. In this paper, we have explained a parallel method for computing rough set. Using map-reduce we can achieve the same. Because of map-reduce we can generate rules and abstract attributes of massive data.