Flow Map Layout

Stanford University

InfoVis 2005

Cartographers have long used flow maps to show the movement of objects from one location to another, such as the number of people in a migration, the amount of goods being traded, or the number of packets in a network. The advantage of flow maps is that they reduce visual clutter by merging edges. Most flow maps are drawn by hand and there are few computer algorithms available. We present a method for generating flow maps using hierarchical clustering given a set of nodes, positions, and flow data between the nodes. Our techniques are inspired by graph layout algorithms that minimize edge crossings and distort node positions while maintaining their relative position to one another. We demonstrate our technique by producing flow maps for network traffic, census data, and trade data.

Source Code and Demo

The code for the Flow Map Layout paper is available.

Paper

Adobe Acrobat PDF (2 MB)

Slides

Figure 1: A flow map of migration from California from 1995-2000, generated automatically by our system using edge routing but no layout adjustment.

Figure 2.  A map of the top 10 states that migrate to California and New York, showing that New York attracts more people from the East Coast and that California attracts people from more geographic regions.

Figure 3. The designer of this visualization used our system to produce a flow map which then they mapped to a globe. It appears in the UK Interdependence Report.