Learn R Programming

motifcluster

An R package for motif-based spectral clustering of weighted directed networks.

Introduction

The motifcluster package provides implementations of motif-based spectral clustering of weighted directed networks in R. These provide the capability for:

Building motif adjacency matrices
Sampling random weighted directed networks
Spectral embedding with motif adjacency matrices
Motif-based spectral clustering

The methods are all designed to run quickly on large sparse networks, and are easy to install and use. These methods are based on those described in [Underwood, Elliott and Cucuringu, 2020], which is available at arXiv:2004.01293.

Installation

install.packages("motifcluster")

Dependencies

igraph
Matrix
RSpectra

Documentation

Documentation for the motifcluster package is available in the doc directory.

Vignette

An instructional vignette for the motifcluster package is available in the vignettes directory.

Author

William George Underwood, Princeton University

Copy Link

Version

Install

install.packages('motifcluster')

Monthly Downloads

271

Version

0.2.3

License

GPL-3

Issues

Pull Requests

Stars

Forks

Repository

https://github.com/wgunderwood/motifcluster

Maintainer

William George Underwood

Last Published

November 18th, 2022

Functions in motifcluster (0.2.3)

Perform the motif adjacency matrix calculations for motif M11

get_motif_names

Get common motif names

Perform the motif adjacency matrix calculations for motif M12

random_sparse_matrix

Build a random sparse matrix

run_motif_embedding

Run motif embedding

Sample a bipartite stochastic block model (BSBM)

run_laplace_embedding

Run Laplace embedding

Perform the motif adjacency matrix calculations for motif Ms

run_motif_clustering

Run motif-based clustering

kmeans++ clustering

Sample a directed stochastic block model (DSBM)

Perform the motif adjacency matrix calculations for motif M10

Perform the motif adjacency matrix calculations for motif M5

Perform the motif adjacency matrix calculations for motif M3

Perform the motif adjacency matrix calculations for motif M2

Perform the motif adjacency matrix calculations for motif M6

Perform the motif adjacency matrix calculations for motif M13

get_largest_component

Get largest connected component

Perform the motif adjacency matrix calculations for motif Mexpa

Perform the motif adjacency matrix calculations for motif M4

Perform the motif adjacency matrix calculations for motif Mcoll

Perform the motif adjacency matrix calculations for motif M1

Perform the motif adjacency matrix calculations for motif M9

Perform the motif adjacency matrix calculations for motif M7

Perform the motif adjacency matrix calculations for motif M8

Perform the motif adjacency matrix calculations for motif Md

Build double-edge adjacency matrix

build_motif_adjacency_matrix

Build a motif adjacency matrix

Compute a right-multiplication with the ones matrix

Build product adjacency matrix

Build identity matrix

Build missing-edge indicator matrix

demonstration_graph

Generate a small graph for demonstrations

Build directed indicator matrix

Build sparse adjacency matrix

Build double-edge indicator matrix

Set diagonal entries to zero and sparsify

Compute first few eigenvalues and eigenvectors of a matrix

Build single-edge adjacency matrix

Compute a left-multiplication with the ones matrix

cluster_spectrum

Get cluster assignments from spectrum using k-means++

Build vertex-distinct indicator matrix

Build edge-and-diagonal indicator matrix

Build single-edge indicator matrix

build_laplacian

Build a Laplacian matrix