Exploring Flow, Factors, and Outcomes of Temporal Event Sequences with the Outflow Visualization.

Area of Science:

• Data Visualization
• Computational Science
• Health Informatics

Background:

• Event sequence data is prevalent across domains like electronic medical records (EMRs) and sports analytics.
• These sequences often lead to critical outcomes, necessitating methods to understand their progression.
• Existing methods may not adequately capture the complexity of aggregated event pathways and their associated outcomes.

Purpose of the Study:

• To introduce Outflow, a novel visualization technique for analyzing event sequence data.
• To enable the aggregation and visualization of event sequences into progression pathways.
• To facilitate the exploration of outcomes and external factors linked to specific pathway transitions.

Main Methods:

• Developed the Outflow visualization technique to aggregate multiple event sequences.
• Designed Outflow to display aggregate pathways, including timing and cardinality of event states.
• Integrated functionality to summarize pathway outcomes and explore correlating external factors.

Main Results:

• User study with twelve participants demonstrated ease of learning Outflow with minimal training.
• Participants performed various analytical tasks accurately and rapidly using the Outflow technique.
• Outflow successfully supported the exploration of event pathways, outcomes, and external correlations.

Conclusions:

• Outflow is an effective and user-friendly visualization tool for analyzing complex event sequence data.
• The technique aids in understanding how different event pathways correlate with specific outcomes.
• Outflow shows promise for applications in domains requiring analysis of sequential events and their results.