Density-based clustering: DBSCAN and OPTICS (dbscan conventions)

DbscanClusters: Ester et al. (1996), clusters grown from core points in index order, 0 for noise, as dbscan::dbscan. OpticsOrdering: Ankerst et al. (1999) ordering, reachability and core distances, ties to the higher index as dbscan::optics.

Usage

DbscanClusters(X, eps, min_pts = 5, border_points = TRUE)

OpticsOrdering(X, eps = Inf, min_pts = 5)

Arguments

X: Data matrix.
eps: Radius.
min_pts: Minimum points (itself included).
border_points: Assign border points.

Value

List.

References

Ester, M., Kriegel, H.-P., Sander, J. and Xu, X. (1996). A density-based algorithm for discovering clusters in large spatial databases with noise. KDD-96, 226-231.

Examples

X <- rbind(c(0, 0), c(0, 1), c(1, 0), c(9, 9), c(9, 8), c(8, 9), c(5, 5))
DbscanClusters(X, 1.5, 3)$cluster
#> [1] 1 1 1 2 2 2 0
OpticsOrdering(rbind(c(0, 0), c(0, 1), c(0, 3), c(10, 0)), 5, 2)$order
#> [1] 1 2 3 4