efficient Map- Reduce policy

Abstract- Cloud computing has become a feasible mainstream solution for data processing, storage and distribution. It assures on demand, scalable, pay-as-you-go compute and storage capacity. To analyze such huge data on clouds, it is very important to research data mining approach based on cloud computing model from both theoretical and practical views. There are large amount of data in cloud database or any other cloud file systems, then apply mining on that data to extract knowledge. Data mining is the process of analyzing data from different perspective and shortening it into useful information. For that Naïve Bayes and support vector machine algorithms are used, which are classification algorithms. In this paper both algorithms are used with MapReduce policy and get the high accuracy, efficiency, high performance. Hadoop is an open source implementation of Map Reduce which can achieve better performance