This talk will be an overview of the roles adaptivity and personalization have played in computer tutoring systems and examples of how these concepts have been applied for both within course supports in edX and course selection at UC Berkeley.
The instructor's own domain model of subject matter informs the organization of a course's syllabus; however, significant differences can exist between the ideal domain model and the ideal pedagogical model for a given audience. In this talk, we'll show how data from previous offerings of a course can be used as a type of syllabus refinement, creating an adaptive course sequencing by way of personalized suggestions to each learner based on her path [
1]. I will also cover how big data approaches are being adopted at UC Berkeley to surface personalized information to students about courses and progressing towards goal-based course sequence recommendation [
2].