Package index • dbscan

Clustering Algorithms

Cluster data with DBSCAN, HDBSCAN, OPTICS, shared nearest neighbor, and Jarvis-Patrick algorithms.

dbscan() is.corepoint() predict(<dbscan_fast>): Density-based Spatial Clustering of Applications with Noise (DBSCAN)
hdbscan() print(<hdbscan>) plot(<hdbscan>) coredist() mrdist() predict(<hdbscan>): Hierarchical DBSCAN (HDBSCAN)
optics() print(<optics>) plot(<optics>) as.reachability(<optics>) as.dendrogram(<optics>) extractDBSCAN() extractXi() predict(<optics>): Ordering Points to Identify the Clustering Structure (OPTICS)
sNNclust(): Shared Nearest Neighbor Clustering
jpclust(): Jarvis-Patrick Clustering
extractFOSC(): Framework for the Optimal Extraction of Clusters from Hierarchies
ncluster() nnoise(): Number of Clusters, Noise Points, and Observations

Find k-nearest, fixed-radius, and shared nearest neighbors and work with nearest-neighbor graphs.

kNN() sort(<kNN>) adjacencylist(<kNN>) print(<kNN>): Find the k Nearest Neighbors
frNN() sort(<frNN>) adjacencylist(<frNN>) print(<frNN>): Find the Fixed Radius Nearest Neighbors
sNN() sort(<sNN>) print(<sNN>): Find Shared Nearest Neighbors
kNNdist() kNNdistplot(): Calculate and Plot k-Nearest Neighbor Distances
adjacencylist() sort(<NN>) plot(<NN>): NN — Nearest Neighbors Superclass
comps(): Find Connected Components in a Nearest-neighbor Graph

Calculate local and hierarchical outlier scores and estimate local point density.

lof(): Local Outlier Factor Score
glosh(): Global-Local Outlier Score from Hierarchies
pointdensity(): Calculate Local Density at Each Data Point
kNNdist() kNNdistplot(): Calculate and Plot k-Nearest Neighbor Distances

Evaluate density-based clusterings with the Density-Based Clustering Validation index.

Plot clusters and work with reachability plots and cluster hierarchies.

hullplot() clplot(): Plot Clusters
print(<reachability>) plot(<reachability>) as.reachability(): Reachability Distances
as.dendrogram(): Coercions to Dendrogram

Turn clustering objects into tidy tibbles and augment the original data.

tidy() augment() glance(): Turn an dbscan clustering object into a tidy tibble

Example and benchmark data sets for density-based clustering.

moons: Moons Data
DS3: DS3: Spatial data with arbitrary shapes
DBCV_datasets Dataset_1 Dataset_2 Dataset_3 Dataset_4: DBCV Paper Datasets