Think of a teacher who inspired you, made you think in new ways, or even made you want to do your algebra homework. There was probably more to their influence than just being a great person, like how they listened carefully to what students were saying in class and incorporated that into their lesson to boost engagement and understanding.

The technical term for this skill mastered by the best instructors is “assimilation”, and it is not easy to teach it to teachers or to assess how well they use it. The traditional way of assessing it and other teaching practices combines annual classroom observation and expert grading in an infrequent, staff-intensive, and highly subjective process. But a project led by the University of Maryland could one day provide frequent, automated measurements of teacher performance.

Known as M-Powering Teaching, or MPT, the system was selected last month as one of 30 educational technology projects to win $250,000 in funding from the nonprofit Learning Agency. education-focused supported by Schmidt Futures, the Bill and Melinda Gates Foundation, and other high profile philanthropies.

MPT uses natural language processing, a branch of machine learning and artificial intelligence, to analyze how math teachers teach and interact with students, with the goal of providing a near-instantaneous feedback stream. It wouldn’t replace human ratings or feedback to teachers, and the system has plenty of human checks and balances, said MPT core team member Jing Liu, an assistant professor of education policy focused on how the field intersects with data science.

But when fully operational, it has the potential to be faster and more accurate than humans can when measuring performance.

“What we do is combine theory and insights from teaching, learning and linguistics, and use an automated process that generates useful information from a transcript of the class,” Liu said. “We can measure, for example, how often in the classroom a teacher picks up on students’ ideas, or when the teacher asks questions, are they closed yes-no questions or open-ended questions that can generate useful talk?”

The technology was developed based on thousands of hours of archived math teaching recordings housed at Harvard University – where Professor Heather Hill, Liu’s MPT partner, studies teaching quality and training programs – and at the University of Michigan. The third core team member, Dora Demszky, is an incoming assistant professor of educational data science at Stanford University who studies natural language processing.

It will soon be tested in a school district in Utah, and MPT should eventually be integrated into a teaching improvement application set up by another partner, the start-up TeachFX, which currently offers information based on metrics on conversation patterns in classrooms, like how long teachers talk before a student can type a word on the edge.

“Dr. Liu’s work on MPT is a great example of innovation in education,” said Kimberly Griffin, Dean of UMD’s College of Education. will undoubtedly help advance the field of learning engineering.”

Teaching math is just the beginning, Liu said. “Improving math skills is a national priority and the foundation of STEM education, so it’s a natural thing to focus on,” he said. “But we’re starting to expand it to apply to language arts, and it can be useful for many subjects and disciplines in the future.”