Stratiline: A visualization system based on stratified storyline

Abstract

In recent years, storyline visualization has garnered considerable attention from the visualization research community. However, previous studies have given little focus to representing the locations of scene and addressing visual clutter issues, especially with larger datasets. In response to this gap, we propose an innovative visual analysis method named Stratiline (short for stratified storyline), which emphasizes multiperspective story data exploration and overview+detail analysis for large-scale datasets. Stratiline introduces a novel framework for calculating the significance of locations, actors, and scenes, providing a mechanism that incorporates user adjustments into the calculation framework to enable multiperspective exploration. Based on this calculation framework, Stratiline offers multiple coordinated views that collaboratively present different perspectives of the story while facilitating rich interactions. Specifically, Stratiline includes time-range drill-down features for overview+detail analysis, while the Storyline View allows for detailed analysis, and the Scene View provides an overview of the entire narrative to help maintain the mental map. The effectiveness of Stratiline is validated through comparative analyses against contemporary storyline designs. Carefully designed case studies illustrate Stratiline’s capability for multiperspective story exploration and large-scale dataset analysis. Quantitative evaluations affirm the stability of our sorting algorithms, which are crucial for time-range drill-down analysis.