Dynamic DBSCAN-GM clustering algorithm
Abir Smiti, Zied Elouedi · 2015
Clustering algorithms are being the core topic of many fields of study in Computational Intelligence and Informatics. Their objective is to determine the critical grouping in a set of unlabeled data. Lot of clustering works engages input number of clusters which is severe to find out. Additionally, the majority is not forceful enough towards noisy data. On the contrary, the clustering method DBSCAN-GM, which is the merger of DBSCAN and Gaussian-means, can solve these problems. However, it is not dynamic, it is not suitable for the frequently change databases. In this paper, we present an extended version of DBSCAN-GM called Dynamic DBSCAN-GM (DDG) to handle incremental databases which evolve over time.