Pros and Cons of DBSCAN

 Pros:

• Does not require to specify number of clusters beforehand.

• Performs well with arbitrary shapes clusters.

• DBSCAN is robust to outliers and able to detect the outliers.

Cons:

• In some cases, determining an appropriate distance of neighborhood (eps) is not easy and it requires domain knowledge.

• If clusters are very different in terms of in-cluster densities, DBSCAN is not well suited to define clusters. The characteristics of clusters are defined by the combination of eps-minPts parameters. Since we pass in one eps-minPts combination to the algorithm, it cannot generalize well to clusters with much different densities.