An approach to achieve high efficiency for large volume data processing using multiple clustering algorithms

Abstract

Now a days data is increasing exponentially daily in terms of velocity, variety and volume which is also known as Big data. When the dataset has small number of dimensions, limited number of clusters and less number of data points the existing traditional clustering al- gorithms will give the expected results. As we know this is the Big Data age, with large volume of data sets through the traditional clus- tering algorithms we will not be able to get expected results. So there is a need to develop a new approach which gives better accuracy and computational time for large volume of data processing. The Proposed new System Architecture is a combination of canopy, Kmeans and RK sorting algorithm through Map Reduce Hadoop frame work platform. The analysis shows that the large volume of data processing will take less computational time and higher accuracy, and the RK sorting does not require swapping of elements and stack spaces.