cluster: Cluster the recurrence embedding via HDBSCAN

In timradtke/recur: Naive Embedding of Many Time Series via Recurrence Plots

Description

This function extends the embedding by an HDBSCAN clustering. Note that the embedding item is extended by two columns cluster and membership_prob of the as optimal identified cut in the hierarchical clustering. The full HDBSCAN object returned by dbscan::hdbscan() is returned as well.

Usage

cluster(object, min_points = 10)

Arguments

object	The recurrence embedding object returned by embed().
min_points	The min_points argument of dbscan::hdbscan().

References

Leland McInnes, John Healy, Steve Astels. The hdbscan Clustering Library.

Ricardo J.G.B. Campello, Davoud Moulavi, Joerg Sander (2013). Density-Based Clustering Based on Hierarchical Density Estimates, PAKDD 2013, Part II, LNAI 7819, pp. 160-172.

See Also

measure(), embed(), dbscan::hdbscan()

Examples

times <- recur::synthetic
measured <- measure(data = times, size = "triangle", shape = "wide")
embedded <- embed(object = measured)
embedded_and_clustered <- cluster(embedded, min_points = 25)

plot(embedded_and_clustered$embedding$x, 
     embedded_and_clustered$embedding$y,
     col = 1 + embedded_and_clustered$embedding$cluster)
     
plot(embedded_and_clustered$hdbscan)

timradtke/recur documentation built on June 15, 2020, 12:41 a.m.