Educational Data Mining

Computer-based learning systems can now keep detailed logs of user-system interactions, including key clicks, eye-tracking, and video data, opening up new opportunities to study how students learn with technology. Educational Data Mining (EDM; Romero, Ventura, Pechenizkiy, & Baker, 2010) is concerned with developing, researching, and applying computerized methods to detect patterns in large collections of educational data – patterns that would otherwise be hard or impossible to analyze due to the enormous volume of data they exist within. Data of interest is not restricted to interactions of individual students with an educational system (e.g., navigation behavior, input to quizzes and interactive exercises) but might also include data from collaborating students (e.g., text chat), administrative data (e.g., school, school district, teacher), and demographic data (e.g., gender, age, school grades). Data on student affect (e.g., motivation, emotional states) has also been a focus, which can be inferred from physiological sensors (e.g., facial expression, seat posture and perspiration). EDM uses methods and tools from the broader field of Data Mining (Witten & Frank, 2005), a sub-field of Computer Science and Artificial Intelligence that has been used for purposes as diverse as credit card fraud detection, analysis of gene sequences in bioinformatics, or the analysis of purchasing behaviors of customers. Distinguishing EDM features are its particular focus on educational data and problems, both theoretical (e.g., investigating a learning hypothesis) and practical (e.g., improving a learning tool). Furthermore, EDM makes a methodological contribution by developing and researching data mining techniques for educational applications. Typical steps in an EDM project include data acquisition, data preprocessing (e.g., data “cleaning”), data mining, and validation of results.