Please use this identifier to cite or link to this item: http://arks.princeton.edu/ark:/88435/dsp01cv43p0225
 Title: Technology and Pedagogy: Using Big Data to Enhance Student Learning Authors: Brinton, Christopher Greg Advisors: Chiang, Mung Contributors: Electrical Engineering Department Keywords: Big DataData MiningLearning TechnologiesMachine LearningSocial LearningSocial Networks Subjects: Electrical engineeringEducational technology Issue Date: 2016 Publisher: Princeton, NJ : Princeton University Abstract: The big data revolution'' has penetrated many fields, from network monitoring to online retail. Education and learning are quickly becoming part of it, too, because today, course delivery platforms can collect unprecedented amounts of behavioral data about students as they interact with learning content online. This data includes, for example, each click made while watching a lecture video, while submitting an answer to a quiz question, or while posting a question on a discussion forum. The ability to capture this data presents novel opportunities to study the complex process by which learning occurs, and also raises interesting research questions around how behavioral data can be leveraged to improve the quality of each student's learning experience, especially as online learning is scaled in size at the apparent expense of efficacy. In this thesis, I detail three research thrusts we have undertaken in using big data to study learning and enhance pedagogy. First is Learning Data Analytics (LDA), in which we have developed new methods for representing student video-watching behaviors as compact sequences, extracted recurring patterns from these sequences and showed how certain ones are significantly correlated with performance, and used the results in the design of behavior-based, early detection algorithms for performance prediction. Second is Social Learning Networks (SLN), in which we have proposed a new model for social learning that combines the topical and structural aspects of discussions, used this model to determine the efficiency of existing discussions, and designed algorithms to encourage SLN formation around a more optimal state. Third is Integrated and Individualized Courses (IIC), in which we have developed two new learning technology systems - a student-facing, course delivery platform and an instructor-facing, analytics dashboard - that build models based on behavior, individualize the content delivered to students based on these models, and visualize certain components of the models to instructors. I will also discuss the extensions we are exploring in terms of additional data capture, data analytics, algorithms, system design, and user trials by deploying IIC in various learning scenarios. URI: http://arks.princeton.edu/ark:/88435/dsp01cv43p0225 Alternate format: The Mudd Manuscript Library retains one bound copy of each dissertation. Search for these copies in the library's main catalog: http://catalog.princeton.edu/ Type of Material: Academic dissertations (Ph.D.) Language: en Appears in Collections: Electrical Engineering