Skip to contents

Louvain community-detection clustering

Source: R/cluster_louvain.R
cluster_louvain.Rd

Builds a shared nearest-neighbour graph from the input data and applies the Louvain algorithm (igraph::cluster_louvain()) to detect communities.

Usage

cluster_louvain(data, vars = NULL, k = 15L, resolution = 1)

Arguments

data

matrix or data.frame. Rows are observations, columns are variables.

vars

character vector or NULL. Column names to use for clustering. If NULL (default), all columns are used.

k

integer(1). Number of nearest neighbours used to build the kNN graph. Default 15L.

resolution

numeric(1). Resolution parameter passed to igraph::cluster_louvain(). Higher values yield more (smaller) clusters. Default 1.

Value

A character vector of cluster labels, one per row of data. Labels are of the form "1", "2", etc.

Details

The function:

  1. Scales the selected columns to zero mean and unit variance.

  2. Finds the k nearest neighbours for each observation using Euclidean distance (via FNN::get.knn()).

  3. Constructs a shared-nearest-neighbour (SNN) graph where the edge weight between two observations equals the number of shared neighbours.

  4. Applies Louvain community detection on the SNN graph.

Examples

set.seed(1)
mat <- matrix(
  c(rnorm(50, -3), rnorm(50, 3), rnorm(100, 0)),
  ncol = 2,
  dimnames = list(NULL, c("v1", "v2"))
)
cl <- cluster_louvain(mat, k = 10)
table(cl)
#> cl
#>  1  2  3  4  5  6  7 
#> 18 12 13  7 22 15 13