sparseAHC: Agglomerative hierarchical clustering for sparse similarity...

In jimhester/sparseAHC: Agglomerative Hierarchical Clustering for Sparse Similarity Matrices.

Description

Agglomerative hierarchical clustering for sparse similarity matrices.

Usage

sparseAHC(S, linkage = c("average", "single", "complete"),
  check.symmetry = TRUE)

Arguments

S	sparse similarity matrix of type dgCMatrix
linkage	linkage criteria. It determines the similarity (or originally distance) between sets of nodes. See details.
check.symmetry	logical. If TRUE, tests if the input similarity matrix is symmetric.

Details

Let A and B be two disjoint subsets of points (nodes of the graph). “Complete linkage” is min S(a,b) for a \in A and b \in B, “single linkage” is max S(a,b) for a \in A and b \in B and “Average linkage” is 1/(|A|.|B|) * sum_{a \in A} sum_{b \in B} S(a,b).

Value

dendrogram of type hclust

Examples

library("Matrix")
library("igraph")

A <- Matrix(0, nrow = 6, ncol = 6, sparse = TRUE)
A[1,2] <- 2; A[1,5] <- 3; A[1,4] <- 2.5; A[2,5] <- 4
A[3,4] <- 2; A[3,5] <- 6; A[4,5] <- 5; A[5,6] <- 4.6
A <- A + t(A)

G <- graph.adjacency(A, mode = "undirected", weighted=TRUE)
plot(G, edge.label=E(G)$weight, vertex.label=V(G)-1)

H <- sparseAHC(A, "average", TRUE)
plot(H)

jimhester/sparseAHC documentation built on May 19, 2019, 10:33 a.m.