A Survey of Advancements in DBSCAN Clustering Algorithms for Big Data

Cluster analysis is an unsupervised machine learning job of grouping objects based on some similarity measure. Among clustering algorithms, DBSCAN (Density Based Spatial Clustering of Application with Noise) contributes to unsupervised machine learning by enabling the clustering of datasets with varying densities, shapes, and sizes. DBSCAN does not require the predefinition of the number of clusters and is able to recognize noiseless arbitrary clusters by using two parameters, minPts and eps. This paper reviews the different DBSCAN algorithms for big data clustering and provides a detailed comparison among the algorithms.