A Relative Examination on Clustering Techniques: Agglomerative, K-Means, Affinity Propagation and DBSCAN

Abstract

Clustering is a procedure of grouping a collection of certain objects into a relevant sub-group. Each sub-group is called as a cluster, which guides users to comprehend the collections in a data set. It is an unsupervised learning technique where each dispute of this type deals with discovering a structure during the accumulation of unlabeled data. Statistics, Pattern Recognition, Machine learning are some of the active research in the theme of Clustering techniques. A Large and Multivariate database is built upon excellent data mining tools in the analysis of clustering. Many types of clustering techniques are— Hierarchical, Partitioning, Density–based, Model based, Grid–based, and Soft-Computing techniques. In this paper a comparative study is done on Agglomerative Hierarchical, K-Means, Affinity Propagation and DBSCAN Clustering and its Techniques.