Improving DBSCAN using Parallel Computing Approaches

Abstract

Abstract

Data mining has become the buzz word of computational research due to its enormous economical and social significance. Clustering is subset of data mining operations and is a type of learning using observation techniques. Clustering uses on unsupervised learning model and does not require any training data to generate the clustering model. Clustering enables grouping of similar and dissimilar type of data in separate groups. DBSCAN is on emerging and powerful clustering algorithm and has gained widespread popularity in recent times. This work is aimed at reducing the time complexity of DBSCAN algorithm by parallel computing technology. Parallel computing techniques have been implemented using multiple cores with and without code information.

Keywords: Data mining, clustering, DBSCAN, parallel computing

Cite this Article

Anupam Kumari, Vishal Shrivastava. Improving DBSCAN using Parallel Computing Approaches. Recent Trends in Parallel Computing. 2019; 6(2): 20–26p.