DIVI: Dynamically Interactive Visualization

Area of Science:

• Computer Science
• Human-Computer Interaction
• Data Visualization

Background:

• Static visualizations often limit dynamic data exploration and cross-tool analysis.
• Existing interactive systems typically require explicit specification of interactions.

Purpose of the Study:

• To introduce Dynamically Interactive Visualization (DIVI), a novel framework for orchestrating interactions within and across static visualizations.
• To enable dynamic interaction without prior specification, facilitating rapid data exploration.

Main Methods:

• DIVI deconstructs Scalable Vector Graphics (SVG) charts at runtime to infer chart components, scales, and data fields.
• It enumerates candidate transformations for linking views and introduces a taxonomy of standard interactions.
• The system reconstructs chart elements and coordinates user input to decouple interaction from specification logic.

Main Results:

• DIVI successfully infers chart structure and data relationships from static SVG charts.
• The framework supports dynamic, composed interactions across different chart types and analysis goals.
• A usability study with 13 participants demonstrated DIVI's effectiveness for rapid data exploration.

Conclusions:

• DIVI offers a powerful new approach to creating dynamic and interactive data visualizations.
• The framework's ability to infer and compose interactions significantly enhances data exploration capabilities.
• DIVI advances the field of interactive data visualization by enabling flexible, specification-free interactions.