galoislattice

A R package to create Galois lattices (Concept Lattice) for two-mode networks and analyze them. By using a reduced label for directed galois lattices, the positional dominance of the original network can be extracted.

This package depends on the R package igraph.

Installation

You can install galoislattice from github with:

# install.packages("devtools")
devtools::install_github("Siliegia/galoislattice")

Example

Here you can see how to use galoislattice on a two-mode igraph object. As an example we use the well known southern women dataset southernwomenG, which is in the package.

data("southernwomenG")

# create a Galois Lattice with a reduced label
Galois <- do_galois_lattice(southernwomenG, label = "reduced")

# plot the Galois Lattice with the according layout
plot(Galois, layout = galois_layout(Galois))

# create a Galois Lattice to extract positional dominance
G <- do_galois_lattice(southernwomenG, label = "reduced", directed = TRUE)
# extracting the dominance relations for the women
Tree <- do_dominance_tree(G, nodes = V(southernwomenG)$name[V(southernwomenG)$type==0])
plot(Tree)

The do_galois_lattice function returns an igraph object, which is the Galois lattice formed from the two-mode network. galois_layout will return the level of positional dominance on the y-axis and do_dominance_tree extracts the positional dominance from the directed reduced labeled Galois Lattice.