| cograph-package | R Documentation |
cograph analyses and plots social, transition and complex networks.
It accepts adjacency matrices, edge lists, igraph, statnet network,
qgraph and tna objects without conversion. It plots them with base R
(splot) or grid graphics (soplot), reshapes
them with a family of wrangling verbs (network_wrangling),
and returns most of its analyses as tidy data frames.
centrality: node centrality measures, one column per
measure (see list_centralities)
network_summary: network-level statistics, one row
per network
communities: community detection, one row per node
motifs and subgraphs: triad census and
motif instances with significance tests
robustness and vulnerability: network
robustness under node or edge removal
disparity_filter: backbone extraction
summarize_clusters and supra_adjacency:
multi-cluster and multilayer networks
splot: Plot a network with base R graphics
soplot: Plot a network with grid graphics
plot_tna, tplot: TNA-style plots with
qgraph-compatible arguments
cograph: Create a network object for the builder functions
sn_layout: Apply layout algorithms
sn_nodes: Customize node aesthetics
sn_edges: Customize edge aesthetics
sn_theme: Apply visual themes
sn_render: Render to device
sn_ggplot: Convert to ggplot2 object
Built-in layouts include the following; list_layouts lists
every registered name:
circle, oval: Nodes arranged on a circle or ellipse
spring: Fruchterman-Reingold force-directed layout
gephi_fr: Gephi-style Fruchterman-Reingold layout
groups: Group-based circular layout
target: levels by graph distance from a focal node
saqr: rows of a transition flow between Start and End
nodes
grid, star, bipartite, random
custom: User-provided coordinates
Built-in themes include:
classic: Traditional network visualization style
colorblind: Accessible color scheme
gray: Grayscale theme
dark: Dark background theme
minimal: Clean, minimal style
viridis: Viridis-based color theme
nature: Nature-inspired color theme
In the analytic functions an edge weight is a strength. A higher weight means a stronger connection, such as a larger transition probability or a stronger correlation. This follows the convention of qgraph and tna.
Path-based measures such as betweenness, closeness, harmonic centrality and
eccentricity can convert weights to distances as 1 / weight^alpha.
The invert_weights argument of centrality controls
this conversion. Its default is TRUE for tna objects and
FALSE for other inputs, which matches igraph and sna. The
alpha argument (default 1) sets the exponent.
Measures that do not use paths, such as degree, strength, eigenvector
centrality, PageRank and transitivity, use the weights as given. Functions
with a weights argument, such as shortest_paths,
compute unweighted results when weights = NA. Individual help pages
document exceptions.
Maintainer: Sonsoles López-Pernas sonsoles.lopez@uef.fi [copyright holder]
Authors:
Mohammed Saqr [copyright holder]
Useful links:
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.