The Effectiveness of Interactive Visualization Techniques for Time Navigation of Dynamic Graphs on Large Displays

Area of Science:

• Computer Science
• Information Visualization
• Network Science

Background:

• Visualizing dynamic networks over extended periods presents challenges due to evolving nodes and edges.
• Existing visualization methods like small multiples and interactive animation may not optimally support comparing disparate network states.

Purpose of the Study:

• To evaluate the effectiveness of different interactive visualization techniques for analyzing dynamic networks.
• To compare traditional methods (small multiples, animation) against interactive timeslicing for specific analytical tasks.

Main Methods:

• Conducted laboratory experiments involving participants performing network analysis tasks on a large touch display.
• Compared user performance across three visualization techniques: small multiples, interactive animation, and interactive timeslicing.
• Participants completed 24 trials across three distinct analytical tasks.

Main Results:

• Interactive timeslicing demonstrated significant benefits for comparing network states at distant points in time.
• Interactive timeslicing showed less advantage when analyzing contiguous or closely spaced time intervals.
• Task performance varied across techniques depending on the temporal relationship between network states.

Conclusions:

• Interactive timeslicing is a valuable tool for specific dynamic network analysis tasks, particularly when comparing non-adjacent time points.
• The choice of visualization technique should consider the nature of the temporal comparison required for dynamic network analysis.
• Further research can explore optimizing interactive timeslicing for analyzing temporal continuity in dynamic networks.