How does DBSCAN Clustering work, and in what cases is it useful?
Density-based clustering approaches, such as DBSCAN, tend to perform better than partitioning methods like K-Means when clusters are non-globular
dbscan
How is clustering affected by high-dimensional data, and how can the quality of clusters generated be improved in such cases?
One problem of performing clustering in high-dimensional data is that common distance metrics, such as Euclidean distance, do not perform as well.