SineStream: Improving the Readability of Streamgraphs by Minimizing Sine Illusion Effects

Area of Science:

• Computer Science
• Information Visualization

Background:

• Streamgraphs are widely used for visualizing time-series data.
• Human perception of streamgraphs can be distorted by sine illusion effects, impacting readability.

Purpose of the Study:

• To introduce SineStream, a novel streamgraph variant designed to minimize sine illusion effects.
• To provide a perceptual foundation for streamgraph design by linking readability to illusion reduction.

Main Methods:

• Re-interpreting baseline computation and layer ordering algorithms to reduce sine illusions.
• Implementing a Gaussian weight for baseline computation to penalize thickness changes.
• Developing a hierarchical clustering algorithm for layer ordering based on three design requirements.

Main Results:

• Quantitative experiments and user studies validate SineStream's effectiveness.
• SineStream demonstrates improved readability and aesthetics compared to existing methods.

Conclusions:

• SineStream offers a perceptually-grounded approach to streamgraph design.
• The proposed methods enhance the clarity and visual appeal of streamgraph visualizations.