A Robust Density-Based Clustering Approach Using DBCURE –MapReduce Techniques

Clustering is the process of grouping similar data into clusters and dissimilar data into different clusters. Density-based clustering is a useful clustering approach such as DBSCAN and OPTICS. The increasing volume of data and varying size of data sets lead the clustering process challenging. So that we propose a parallel framework of clustering with advanced approach called MapReduce. We develop a new algorithm DBCURE which is modified from DBSCAN. Next we propose a newly developed parallel algorithm called DBCURE-MR that is DBCURE algorithm is parallelized with MapReduce framework. This proposed method finds clusters in parallel manner efficiently.