DBSCAN (Density-Based Spatial Clustering of Applications with Noise) is a density-based clustering base technique. It can find clusters of various forms and sizes in a big amount of data that includes noise and outliers.
The goal of DBSCAN is to locate data point neighbourhoods that exceed a specific density criterion. The radius of the neighbourhood (eps) and the minimum number of neighbors/data points (minPts) within the radius of the neighbourhood define the density threshold.
DBSCAN is a clustering method used in machine learning to distinguish high-density clusters from low-density clusters.
DBSCAN has two parameters: (eps) and the minimum number of points needed to build a dense zone (minPts). It begins with a random starting spot that has never been visited before. The -neighborhood of this point is retrieved, and if it contains enough points, a cluster is formed.
Advantages. DBSCAN does not require the number of clusters to be specified in advance. DBSCAN works effectively with clusters of any shape. DBSCAN understands noise and is resistant to outliers.