IRIS: Intelligent Recommender for Instructors and Students -- Completing Personalized Assessment Loop
Assessment plays an important role in learning for it providing instructors and students feedbacks on their teaching or learning effectiveness. This project will build a personalized education system, called IRIS, to recommend learning materials to be specific to the assessment outcomes and at the appropriate knowledge level for the student. Through automatically mining the available data related to two high-enrollment courses (undergraduate Java Programming course and graduate level Interactive System Design course), which include textbooks, syllabus, slides, online tutorial videos, and the assessments built in the courses, IRIS employs domain-specific concept-embedding-based representations for the core learning material, assessments, and relevant remedial online materials. IRIS also observes students’ activities in their learning and in their assessments to infer students’ knowledge level on certain concepts. Thus, based on each student individual learning behavior, assessment performance, and knowledge level, IRIS makes truly personalized learning recommendations. This project extends the key technologies and tools developed as part of the NSF project led by Drs Brusilovsky and He. and utilizes materials and data collected and organized by instructors over the years. At the same time, IRIS is capable of harvesting automatically online materials from various MOOC sites, independent tutorials, YouTube videos, and academic articles in digital libraries. IRIS is generic enough to be used in any undergraduate and graduate courses.
Project related publications:
Thaker, K., Huang, Y., Brusilovsky, P. and He, D. (2018 ) Dynamic Knowledge Modeling with Heterogeneous Activities for Adaptive Textbooks. In: Proceedings of 11th International Conference on Educational Data Mining, July 15 - 18, 2018, Buffalo NY. pp. 592-595
Thaker, K., He, D. and Brusilovsky, P. (2018) Concept Enhanced Content Representation For Linking Educational Resources. In Proceedings of International Conference on Web Intelligence, Dec 4 - 6. 2018, Santiago, Chile