A Tool for Integrating Log and Video Data for Exploratory Analysis and Model Generation

Victor Girotto, Elissa Thomas, Cecil Lozano, Kasia Muldner, Winslow Burleson, Erin Walker

Analysis of students’ log data to understand their process as they solve problems is an essential part of educational technology research. Models of correct and buggy student behavior can be generated from this log data and used as a basis for intelligent feedback. Another important technique for understanding problem-solving process is video protocol analysis, but historically, this has not been well integrated with log data. In this paper, we describe a tool to 1) facilitate the annotation of log data with information from video data, and 2) automatically generate models of student problem-solving process that include both video and log data. We demonstrate the utility of the tool with analysis of student use of a teachable robot system for geometry.

The final publication is available at Springer via https://doi.org/10.1007/978-3-319-07221-0_9.