
136
Complex Networks: An Algorithmic Perspective
X
X
X
noise point
border point
core point
Eps=1
MinPts=5
1
Figure 7.12: Core, border and noise points in density-based clustering
Border points: Points that are on the border of a cluster
Noise points: Points that are neither core nor border points.
Figure 7.12 displays these three types of data points as a core point and a border
point of a cluster and noise point. The MinPts is equal to 5 and the Eps is unity in
this example.
Before describing the representative density-based algorithm called the density-
based spatial clustering of applications with noise (DBSCAN), we will have the fol-
lowing definitions. ...