graphBirth: Perform a cluster birth operation (split an existing cluster)

In rayisaacalan/BASTION: Bayesian Additive Spanning Tree's Input / Output For Non-Parametric Regression

Perform a cluster birth operation (split an existing cluster)

Description

graphBirth takes in a spanning forest graph with k disconnected components corresponding to the vector of cluster assignments 'membership' and splits an existing cluster. Optionally, the integer corresponding to one of the clusters can be specified as the one to split. If it is not specified, a cluster is uniformly selected at random. In the specified or selected cluster, an edge is uniformly randomly selected and cut, turning the one cluster into two.

Usage

graphBirth(graph, membership, clust = NULL)

Arguments

graph	An object of class 'graph' from the igraph package
membership	A vector of integers of length N with k unique integers (k < N) which map each vertex to a cluster
clust	(Optional) Integer, the cluster to split. Must be between 1 and k. By default, a cluster will be uniformly randomly selected

Value

A list containing two elements:

graph	The input graph with 1 fewer active edge
membership	A vector of integers of length N with k + 1 unique integers which map each vertex to a cluster
new_clust_ids	Vertex keys of the vertices belonging to the new cluster
old_clust_ids	Vertex keys of the vertices from the cluster which was split excluding the ones from new_clust_ids

References

Luo, Z.T. (*), Sang, H. and Mallick, B.K. (2021), BAST: Bayesian Additive Regression Spanning Trees for Complex Constrained Domain

Luo, Z.T. (*), Sang, H. and Mallick, B.K. (2021), A Bayesian Contiguous Partitioning Method for Learning Clustered Latent Variables, Journal of Machine Learning Research, 22, 1-52.

See Also

constructClusters, graphDeath, graphChange, graphHyper

Examples

set.seed(1234)
coords = data.frame(lon = rnorm(50), lat = rnorm(50))
g = constructGraph(coords, 5)
clust_out = constructClusters(g, 6, minclust = 5)
plot(clust_out$spanning_forest,
     layout = as.matrix(coords),
     vertex.color = clust_out$membership,
     edge.arrow.mode = 0)
g_7_clusters = graphBirth(clust_out$spanning_forest,
                          clust_out$membership,
                          4)
plot(g_7_clusters$graph,
     layout = as.matrix(coords),
     vertex.color = g_7_clusters$membership,
     edge.arrow.mode = 0)

rayisaacalan/BASTION documentation built on April 27, 2023, 2:06 p.m.