Analysis of DBSCAN Algorithm in Determining Epsilon Parameters Numerical Data Clustering

Abstract

DBSCAN algorithm (Density-Based Spatial Clustering of Applications with Noise) is one of the numerical based clustering algorithms, numerical data is used as the test for this algorithm. The DBSCAN algorithm has the disadvantage of being difficult to determine the appropriate Epsilon value in order to obtain good clustering results. In the DBSCAN algorithm, the value of epsilon is calculated based on a lot of data from the entire data that is captured. In this study a modification of the DBSCAN (Density -Based Spatial Clustering of Applications with Noise) algorithm was carried out by determining the value of epsilon, the results obtained in the study of Euclidean Distance obtained better than the results obtained from the DBSCAN.