layout_tbl_graph_treemap: Calculate nodes as rectangles subdividing that of their parent

Description

A treemap is a space filling hierarchical layout that maps nodes to rectangles. The rectangles of the children of a node is packed into the rectangle of the node so that the size of a rectangle is a function of the size of the children. The size of the leaf nodes can be mapped arbitrarily (defaults to 1). Many different algorithms exists for dividing a rectangle into smaller bits, some optimizing the aspect ratio and some focusing on the ordering of the rectangles. See details for more discussions on this. The treemap layout was first developed by Ben Shneiderman for visualizing disk usage in the early '90 and has seen many improvements since.

Usage

layout_tbl_graph_treemap(
  graph,
  algorithm = "split",
  weight = NULL,
  circular = FALSE,
  sort.by = NULL,
  direction = "out",
  height = 1,
  width = 1
)

Value

A data.frame with the columns x, y, width, height, leaf, depth, circular as well as any information stored as node variables in the tbl_graph object.

Arguments

graph: A tbl_graph object
algorithm: The name of the tiling algorithm to use. Defaults to 'split'
weight: An optional node variable to use as weight. Will only affect the weight of leaf nodes as the weight of non-leaf nodes are derived from their children.
circular: Logical. Should the layout be transformed to a circular representation. Ignored.
sort.by: The name of a node variables to sort the nodes by.
direction: The direction of the tree in the graph. 'out' (default) means that parents point towards their children, while 'in' means that children point towards their parent.
height: The height of the bounding rectangle
width: The width of the bounding rectangle

Details

Different approaches to dividing the rectangles in a treemap exists; all with their strengths and weaknesses. Currently only the split algorithm is implemented which strikes a good balance between aspect ratio and order preservation, but other, more well-known, algorithms such as squarify and slice-and-dice will eventually be implemented.

Algorithms

Split (default)

The Split algorithm was developed by Bjorn Engdahl in order to address the downsides of both the original slice-and-dice algorithm (poor aspect ratio) and the popular squarify algorithm (no ordering of nodes). It works by finding the best cut in the ordered list of children in terms of making sure that the two rectangles associated with the split will have optimal aspect ratio.

References

Engdahl, B. (2005). Ordered and unordered treemap algorithms and their applications on handheld devices. Master's Degree Project.