Treemap

A treemap is a space-filling visualization for hierarchical data that uses nested rectangles to represent the structure and quantitative magnitudes of a dataset. Each node in the hierarchy is a rectangle, with the area proportional to a numeric value such as size, budget, or frequency. The rectangles are subdivided within their parent rectangle, producing a compact, visual summary of both hierarchy and composition. Color can be used to encode an additional attribute, such as category or another numeric measure, while position often reflects the tree structure or a chosen ordering.

Treemaps were introduced by Ben Shneiderman in 1991 as a way to visualize large hierarchical datasets in a compact form. Since then, various layout algorithms have been developed, including slice-and-dice and squarified treemaps. The squarified approach, designed to produce rectangles with more uniform aspect ratios, improves readability by making elements easier to compare.

Common applications include visualizing disk usage and file systems, project budgets, market or portfolio data, software metrics, and other hierarchical data where each node carries a weight. Treemaps enable quick assessment of how a whole is composed and how parts compare in size, at a glance.

Advantages of treemaps include efficient use of screen space, immediate visual comparison of component sizes, and the ability to convey both hierarchy and quantitative information simultaneously. Limitations include reduced readability for very large numbers of small rectangles, potential misinterpretation of area versus color, and sensitivity to data ordering or layout choice. Interactive features such as zooming, panning, tooltips, and breadcrumb trails are common to help navigate deeper levels. Variants include circular, mosaic, and Voronoi treemaps, each with different trade-offs in readability and aesthetics.