Microcluster-Based Incremental Ensemble Learning for Noisy, Nonstationary Data Streams

Figure 4

Three different scenarios: (a) incoming instance’s label is the same as the nearest microcluster’s label; (b) incoming instance’s label varies from the nearest microcluster’s label, which represents noisy instance; and (c) incoming instance as a new concept does not drop into the nearest microcluster and its label is different from the nearest microcluster’s label. Note: the color represents the class label, the rectangle suggests the incoming instance, and the circle represents the nearest microcluster.

(a)

(b)

(c)