The Treemap Concept
The “Treemap” was introduced by Ben Shneiderman on the University of Maryland within the early 1990s¹. Simply put, it’s an efficient way of displaying hierarchical data as a set of nested rectangles. Although the concept is easy, the arrangement of the rectangles is subject to an aesthetic preference, and various arrangement algorithms have been developed to reinforce the looks of the ultimate layout.
Treemap Mechanics
Given a hierarchy, a Treemap represents each branch within the hierarchy as a rectangle, which is then tiled with smaller rectangles representing sub-branches. The space in a Treemap is split in line with a selected attribute of the information (often size or value), and every rectangle’s area corresponds to the attribute’s magnitude, making it easy to match different parts of the hierarchy.
To account for the arrangement of the rectangles, listed below are a few of the common algorithms that govern a Treemap’s construction and supreme appearance:
- Squarified Treemaps² – creates rectangles as near squares as possible by adjusting the aspect ratio of the rectangles
- Strip Treemaps³ – lays out the rectangles in strips, either horizontally or vertically, based on the information’s hierarchy
- Slice-and-Dice⁴ – alternates between horizontal and vertical divisions which is easy but can create elongated rectangles
Treemap Features
- Proportions – the dimensions of every rectangle is proportional to the information point it represents, allowing quick identification of larger and smaller items
- Lines and Colours – clever usage of borders, border size and color, and buffers can delineate hierarchy levels while container colours are sometimes used to represent different dimensions of the information
- Spatial Efficiency – Treemaps…
