Time Series Visualizations

themer riverThere are countless other visualizations that show time series data. These include everything from charts to maps. One unique time series visualization (closely related to area charts) is ThemeRiverâ„¢.

The Theme River visualization was developed at Pacific Northwest Laboratories to display time-related patterns, trends and relationships across a large collection of documents. Since it’s a text visualization, the ThemeRiver is particularly good for speeches, news articles and other recurring text-based subjects.

The goal of a ThemeRiver is to identify trends and patterns by showing the viewer thematic changes over time. Viewers are able to interact with the ThemeRiver to explore the information and to discover trends.

Themes in a ThemeRiver are represented by a river that flows left to right through time, widening or narrowing depending on the collective strength of a selected theme. Individual themes are shown as colored currents that flow within the river, also widening or narrowing to show their strength at any particular point.

In this way, the ThemeRiver is a bit like a histogram, since the ThemeRiver uses variations in width to represent variations in strength. But the continuous flow adds a dimension that makes the visualization more useful and easier to understand, since currents maintain their integrity over time. In other words, if a theme disappears and then reoccurs, it reappears in the same position in the river relative to other themes, making it easier to interpret.

Another famous visualization that shows time series data is Charles Joseph Minard's map of Napoleon's March to Moscow.

Touted as "the best statistical graphic ever drawn," the map portrays the losses suffered by Napoleon's army in the Russian campaign of 1812. Beginning at the Polish-Russian border, the thick band shows the size of the army at each position. The path of Napoleon's retreat from Moscow in the bitterly cold winter is depicted by the dark lower band, which is tied to temperature and time scales.