Animated Narrative Visualization for Video Clickstream Data
We propose to use animation and data storytelling to present insights in video clickstream data. Two novel designs, non-linear time mapping and foreshadowing, are introduced to make the presentation more engaging and interesting.
Visualization of Massive Open Online Courses (MOOC) Data
The aim of this project is to serve MOOC instructors, instructional designers, institutional curriculum leaders and learning scientists by developing an open framework that combines: 1) Analytical methods for learning behavior analysis and predictive analytics; and 2) Visualization interfaces for understanding the huge amount of data collected by MOOC platforms and the analytical results.
Visual Analytics on Streaming Social Media Data
We design and develop StreamExplorer, a visual analytic system that encompasses an online event detection method and a tailored GPU-assisted Self-Organizing Map method, to facilitate the visual analysis, tracking, and comparison of a social stream (e.g., Twitter) at macroscopic, mesoscopic, and microscopic levels.