Erato: Cooperative Data Story Editing via Fact Interpolation

Area of Science:

• Human-computer interaction
• Data visualization
• Information visualization

Background:

• Visual data stories are effective for communicating complex insights.
• Creating data stories requires interdisciplinary skills (data analysis, design).
• Existing methods for data story generation are complex and time-consuming.

Purpose of the Study:

• Introduce Erato, a human-machine cooperative system for data story editing.
• Enable users to generate insightful and fluent data stories with computer assistance.
• Simplify the creation process of visual data stories.

Main Methods:

• Erato system utilizes user-provided keyframes to define story structure and topic.
• A novel interpolation algorithm generates intermediate frames for smooth transitions.
• Evaluations included a Turing test, user study, performance validation, and expert interviews.

Main Results:

• The Erato system significantly improves the efficiency of data story creation.
• The proposed interpolation technique generates coherent and fluent story content.
• User evaluations confirmed the effectiveness and usefulness of the Erato system.

Conclusions:

• Erato offers an efficient approach to data story generation through human-machine cooperation.
• The system lowers the barrier for creating high-quality visual data stories.
• Future work can explore further enhancements in automated narrative generation.