EEd Seminar: Recommendation system to support Work-Integrated Learning experiences in Engineering

Details

Jinnie Shin, Ph.D.
Assistant Professor
Research and Evaluation Methodology
School of Human Development and Organizational Studies in Education
College of Education University of Florida

Join us via Zoom: https://ufl.zoom.us/j/93266326094

Abstract:
This talk will focus on the design and development of a transparent and interpretable AI-driven recommender system that optimizes Work-Integrated Learning (WIL) experiences for Engineering students. WIL is essential in engineering education, bridging academic learning with industry practice, yet traditional approaches lack personalization. Our system leverages transformer models to predict core tasks in real-world engineering internship postings, ensuring tailored recommendations that align with students’ academic backgrounds, technical skills, and professional development needs. Using a hybrid content-based approach, the system personalizes WIL experiences by analyzing students’ competencies, program structures, and key industry tasks. In this study, more than 200 undergraduate engineering students participated, where our system demonstrated high predictive accuracy (R² = 0.767–0.783) in recommending experiences that foster both technical and professional growth, while enhancing self-efficacy in confidence, leadership, identity, and commitment to engineering careers. The system’s performance improved as students engaged in multiple WIL opportunities, achieving optimal results when recommending up to four tailored experiences. By integrating transformer-based models for task prediction and emphasizing transparency and interpretability, this system provides a scalable, evidence-based solution to enhance WIL outcomes. Our findings offer valuable insights for educators and program designers, demonstrating how AI can support student learning, skill development, and career readiness in STEM fields.

Bio
Jinnie Shin, Ph.D., is an Assistant Professor of Research and Evaluation Methodology in the School of Human Development and Organizational Studies in Education within the College of Education at the University of Florida. She has expertise in application of theory-based natural language processing and learning analytics in education research. In her work, she has focused on investigating how to bridge the gap between psychometric analysis and artificial intelligence in education research. Dr. Shin has experience with various international industry projects with the Medical Council of Canada, American College Testing, and the New Zealand Qualifications Authority, which focus on providing effective solutions to complex education problems using deep learning and natural language processing research.