mlgraph function

Multilevel graph

A function to create and manipulate multilevel graphs

mlgraph(net, layout = c("circ", "force", "stress", "rand", "conc", "bip"), main = NULL, 
        seed = NULL, maxiter = 100, directed = TRUE, alpha = c(1, 1, 1), scope, collRecip, 
        undRecip, showLbs, showAtts, cex.main, coord, clu, cex, lwd, pch, lty, bwd, bwd2, 
        att, bg, mar, pos, asp, ecol, vcol, vcol0, col, lbat, swp, loops, swp2, mirrorX, 
        mirrorY, mirrorD, mirrorL, lbs, mirrorV, mirrorH, rot, hds, scl, vedist, ffamily, 
        fstyle, fsize, fcol, valued, modes, elv, lng, nr, ...)

Arguments

• net: a "Multilevel" class object or a three dimensional array with clustering information

• layout: the visualization layout:

  • circ circular
  • force force-directed
  • stress stress-majorization
  • rand random
  • conc concentric
  • bip as bipartite graph

• main: (optional) title of the plot

• seed: (optional) random seed number for the vertices' initial coordinates. Ignored except for force, stress and rand

• maxiter: (optional) maximum number of iterations in layout algorithms. Ignored except for force, stress and rand

• directed: (logical) whether or not the graph is directed or undirected

• alpha: vector (vertex, edge, bg) with the alpha color transparency

• scope: (optional) scope of the graph (see details)

• collRecip: (optional and logical) whether or not collapse reciprocated edges in the undirected graph

• undRecip: (optional and logical) whether or not plot reciprocated edges as undirected

• showLbs: (optional and logical) whether or not to show the vertex labels

• showAtts: (optional and logical) whether or not to show the vertex attribute labels

• cex.main: (optional) size of the plot's title

• coord: (optional) data frame with the coordinates of the vertices. If coordinates are given then the layout option is ignored

• clu: (optional) clustering of the vertices as a list of vectors with integers or NULL (see details)

• cex: (optional) size of the vertices

• lwd: (optional) width of the edges; ignored if valued is set to TRUE

• pch: (optional) symbol representing the vertices

• lty: (optional) shape of the edges

• bwd: (optional) width of the bundle edges. Ranges from 0 (edges collapsed) to the default 1 (depending on the vertices' size), and for valued a value greater than one is possible

• bwd2: (optional) width of the bundle loop edges.

• att: (optional) a vector or an array representing the vertex attributes

• bg: (optional) background color of the plot

• mar: (optional) margins of the plot

• pos: (optional) position of the vertices' labels (0 means ``at the center of the vertex'')

• asp: (optional) aspect ratio of the plot

• ecol: (optional) color of the edges

• vcol: (optional) color of the vertices

• vcol0: (optional) color of the vertices' contour (only works for pch 21 through 25

• col: (optional) alias for vcol

• lbat: (optional) labels for the vertex attributes

• swp: (optional and logical) whether or not to swap the bundle patterns

• loops: (optional, logical, and experimental) plot graph loops?

• swp2: (optional and logical) whether or not to swap reciprocals

• mirrorX: (optional) mirror of the $X$ axis

• mirrorY: (optional) mirror of the $Y$ axis

• mirrorD: (optional) mirror reflection across diagonal $Y=X$

• mirrorL: (optional) mirror reflection across diagonal $Y=-X$

• lbs: (optional) vertex labels

• mirrorV: same as mirrorX

• mirrorH: same as mirrorY

• rot: (optional) clockwise rotation of the graph in degrees

• hds: (optional and experimental) arcs' head scale

• scl: (optional and experimental) numerical scalar ($x$ and $y$) or vector ($x$, $y$) of the graph's scale

• vedist: (optional and experimental) a real number with vertex - edge distance

• ffamily: the font family

• fstyle: the font style

• fsize: the font size

• fcol: the font color

• valued: (optional and logical) whether the graph is depicyed as valued or not

• modes: (optional) a vector indicating which matrices are domains and which codomains (works only with a "Multilevel" class object)

• elv: (experimental) control loops 1

• lng: (experimental) control loops 2

• nr: integer or NULL with the number of radii for conc layout (see details)

• ...: Additional argument items (see e.g. par)

Details

Multilevel graphs serve to represent networks with different ``levels'' such as different domains in the network structure. A characteristic of multilevel networks is the existence of ties within and across domains.

Since this function can handle a large number of arguments, these can be stored as a list object that is passed through the scope option. In this case, a vector made of lists and scalars or combinations of these is accepted.

The bundle width specified by bwd and bwd2 ranges from 0 (edges collapsed) to the default 1 (depending on the vertices' size). For the valued option, a number greater than one is possible.

In a multilevel structure, argument clu is to class network members and it is possible to class all members of the domain or co-domain into a single class by setting the vector to NULL. Similarly, NULL in argument nr for the conc layout implies the use of two radii, one for each domain.

Returns

A plot of the multilevel graph structure for the network

Author(s)

Antonio Rivero Ostoic

Note

Multilevel graphs depend on multilevel class objects

See Also

mlvl, multigraph, bmgraph, frcd, stsm, conc

Examples

## Not run:
# create network data as arrays
arr <- round( replace( array(runif(18), c(3,3,2)), array(runif(18),
       c(3,3,2))>.5, 3 ) )
arr2 <- round( replace( array(runif(18), c(3,3,2)), array(runif(18),
       c(3,3,2))>.5, 3 ) )

# create multilevel class object and plot multilevel graph
require(multiplex)
mlvl(arr, arr2) |> 
  mlgraph()
## End(**Not run**)