3D clustering of streaming data

A density-based spatial clustering algorithm is used to find clusters in fast streaming data. Clusters are color-coded and visualized in 3D. The thresholds for including elements in a cluster can be set so that many small or a few large clusters can be identified, depending on the analysis intent.

Colors encode the membership of elements (crypto USD markets) to a cluster, where elements of a cluster share the same color, except for gray elements that are not part of any cluster.

The most important parameter to control the size and quantity of clusters is the Cluster Radius: To generally find more small clusters, reduce it; to generally find fewer large clusters, increase it.

For the axis data, the use of log-transformation is recommended. If the same data is used for 2 axes, a 2D plane is created; if the same data is used for 3 axes, a 1D line is created.