Christopher Span, Ph.D. Dean & Distinguished Professor Rutgers GSE | Rutgers Graduate School of Education
Dr. Dake Zhang, an Associate Professor of Special Education, is engaged in several research projects aimed at developing AI-based tools for improving the education of students with mathematical difficulties. Zhang is leading the MathNet project in collaboration with a Computer Science professor from Rutgers and an Educational Assessment professor from the University of Washington, Seattle. This project is focused on developing AI techniques for auto-grading and classifying students' problem-solving strategies and errors.
Zhang explained, "We currently focus on students’ hand-written solutions to represent a fraction with a number line, with special attention to the imbalanced data representing the uncommon strategies and error types among students with math difficulties." The team analyzed a large dataset from ASSISTment, evaluating large language models, including ChatGPT-4o, finding their computer vision components inadequate for interpreting handwriting. Zhang and her collaborators trained a visual processing model using expert-annotated images and are now working on a "visual translator" to improve text conversion for deeper diagnosis.
Besides MathNet, Zhang is involved in a grant to build AI platforms offering individualized accommodations and scaffolding for students facing difficulties in mathematics. Another project focuses on designing visual representations for word problems to aid college students, especially those with challenges in civil engineering, in solving word problems effectively.
Discussing the motivation behind her research, Zhang stated, "My research uses visual representations, such as geometry, number lines, diagrams for word problems, and graphs, to help struggling students solve mathematics problems." She believes that integrating AI and data science in STEM education can enhance educators' ability to answer complex research questions and provide better support for diverse student needs. Zhang emphasized the importance of computer vision in AI development and has attracted collaborators eager to innovate in this field.
Zhang employs a quantitative research approach, focusing on cognitive processing, which she describes as "the mental processes involved in acquiring, processing, and storing information, enabling learning, memory, reasoning, and problem-solving functions." Her goal is to identify barriers that students with mathematical difficulties face and develop effective interventions, such as schematic diagrams to facilitate problem-solving.
In the future, Zhang and her collaborators plan to make their research data open source to aid other AI researchers. She is also working on creating teacher-friendly interfaces for ongoing projects and aims to secure additional funding. Teaming up with Dean Span and Professor Janice Gobert, Zhang is drafting a large-scale research proposal on AI and STEM education, showing her commitment to advancing this important field.
Alerts Sign-up