Proportional Aggregation in Hierarchical Data Visualization

Area of Science:

• Information Visualization
• Human-Computer Interaction
• Data Analysis

Background:

• Comparing data attributes within and between hierarchical levels is challenging.
• Existing implicit hierarchy visualization methods primarily use additive aggregation.
• Proportional aggregation for hierarchical data visualization is underexplored.

Purpose of the Study:

• Introduce stalactite plots, a novel visualization technique for proportional aggregation in hierarchical data.
• Enable visual comparison of data attributes across different hierarchy levels.
• Evaluate the effectiveness and usability of stalactite plots.

Main Methods:

• Developed stalactite plots, a visualization technique encoding proportional aggregation.
• Conducted empirical evaluations with user studies (N=148, N=50).
• Compared stalactite plots against established hierarchical visualization methods.

Main Results:

• Stalactite plots, with explanation, are understood comparably to existing techniques.
• Users demonstrated faster and more accurate performance with stalactite plots on large datasets.
• The technique effectively supports visual value comparison in hierarchical data.

Conclusions:

• Stalactite plots provide a viable alternative for visualizing hierarchical data with proportional aggregation.
• The method enhances data attribute comparison, particularly for large and complex datasets.
• Further research can explore applications and optimizations for stalactite plots.