Wei Ai Part of $1.5M NSF Grant to Improve Middle-Grade Math Education with AI

Teachers are increasingly turning to online platforms and social media sites to supplement their curricula, either for content enrichment or to make lessons more interactive and culturally relevant.

Previous research shows that a vast majority of educators use search engines like Google and platforms like TeachersPayTeachers and Pinterest to source lesson materials. Now—with the rise of advanced chatbots—a recent survey shows that 40 percent of teachers also use ChatGPT on a weekly basis.

Wei Ai, an assistant professor in the College of Information Studies with an appointment in the University of Maryland Institute for Advanced Computer Studies (UMIACS), is co-PI of a project that’s using machine learning to help middle grade math teachers design effective lesson plans that support positive learning outcomes.

“As teachers increasingly turn to the web for resources, our research is pivotal in addressing information overload and guiding them toward high-quality lesson plans,” says Ai. “The recent rise of large foundational models such as ChatGPT only intensifies the urgency of this initiative, as it enables the machine learning community to develop their models using only the best in educational content.”

Supported by a $1.5 million grant from the National Science Foundation (NSF), the multi-institutional project includes researchers from the University of Washington, University of Nebraska-Lincoln and Duquesne University.

The team will identify ways to assess the quality of open-source mathematics lesson plans by integrating cutting-edge machine learning techniques with knowledge of effective mathematics education and human feedback. Then they will develop machine learning algorithms to check a lesson plan’s content rigor, and how well it engages students with language and special education needs.

Jing Liu, an assistant professor of education policy at UMD and co-PI on the project, points out how teachers’ input and voices will be centered throughout the process, meaning that machine learning techniques will be used in a highly responsible manner.

Ai, who is also a member of the Computational Linguistics and Information Processing (CLIP) Lab, says that the project exemplifies the value and promise of using machine learning to address societal issues and advance education research in an ethical way.

The NSF-funded team is focusing on fifth through eighth grade math because that’s when the subject starts to become more complex, and kids face unique challenges in terms of their physical and social-emotional development, says lead PI Min Sun, a professor of education at the University of Washington.

She adds that the project will democratize access to quality, inclusive and tailored learning materials, while providing instructors with more time to interact with students.

“Beyond scientific contributions, we are addressing equity issues because junior teachers spend more time on lesson planning, as do teachers serving historically marginalized students and communities,” says Sun. “It’s critical that teachers’ lesson plans effectively support students with a wide range of academic performance levels, language and cultural backgrounds.”

Findings from the project will be publicly available through the AmplifyLearn.AI Center, a cross-institution collaboration led by Sun that explores AI research in education, EdTech product development, and data science training and outreach.

Liu and Ai are collaborating on a separate UMD project that’s using machine learning to analyze K–12 math teachers’ instruction and provide them with feedback to improve, with a focus on equity and fairness inside the classroom.

That project, which also includes Professor of Electrical and Computer Engineering Carol Espy-Wilson, is supported by a $1 million team project grant from UMD’s Grand Challenges Grants program.

This article was based on news releases from Maryland Today and the University of Washington.