UMass Amherst to commercialize math tutoring software: Preparing the intelligent tutor MathSpring …
IMAGE: Students in grades 5 through 10 using MathSpring software can choose from a number of different screen companions as they work through mathematics problems. The tutor evaluates student skills and… view more
Credit: UMass Amherst
AMHERST, Mass. – University of Massachusetts Amherst computer scientist Beverly Woolf, an international leader in intelligent tutoring systems and expert in science and mathematics learning, recently received a one-year, $199,944 grant from the National Science Foundation to commercialize the intelligent tutor known as MathSpring for e-learning in mathematics.
In 2013, Woolf was named a Presidential Innovation Fellow by President Barack Obama in recognition of her leadership in designing software tutors that respond to a student’s mood and personal learning pace, for example, to dramatically improve lesson effectiveness. Her work combines artificial intelligence, computer network technology and multimedia features in digital tutoring software for teaching mathematics according to an individual student’s needs.
Woolf, along with assistant professor Ivon Aroyo at Worcester Polytechnic Institute and staff at UMass Amherst, created intelligent and emotionally perceptive teaching software for students in grades 5 through 10 that features friendly animated characters who interact one-on-one with users like a personal trainer. The tutor uses artificial intelligence techniques to evaluate student skills and knowledge in real time, then adjusts its responses to offer personalized strategies that address knowledge gaps and provide advice for tackling difficult problems. The characters also speak to students about the importance of perseverance and effort.
Woolf, who calls herself a change agent in education technology, says, “No online tutor today responds by analyzing both student knowledge and behavior. Our MathSpring project addresses the high failure rate of K-12 students to learn mathematics. It applies theoretical understanding of the individual student’s knowledge and mood, which helps to guide each tutor response. These interactions are designed to move students away from boredom or disengagement, and will have the capability to select from among as many as 700 problems in the system.”
An earlier version of Woolf’s software improved student performance on standardized test scores by an average 10 percent, a critical difference for low-achieving and other students who often struggle with math. It has been used worldwide to improve students’ early relationship with mathematics, keeping later career options open.
Woolf and colleagues say the potential economic impact of translating the MathSpring technology to the marketplace will make a positive contribution to the growth rate of e-learning within the next five years and to U.S. competitiveness in this field.
“Since the annual U.S. education expenditure for K-12 is approximately $625 billion, a large potential exists for making both a commercial and social impact in this space,” they note. The researchers plan is to design the tutoring software for both Android and Apple’s iOS platforms and hope to identify groups of schools as long-term partners.
UMass Amherst graduate students and programmers involved in MathSpring will benefit from taking part in innovation and technology translation experiences, Woolf points out.
The MathSpring project will partner with CarneyLabs, a performance acceleration company, to guide commercial aspects of the translation from research to commercial reality and Virginia Advanced Studies Strategies, a non-profit company that works with the Virginia Department of Education, to provide a test environment, Woolf says.
Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.
Source: UMass Amherst to commercialize math tutoring software: Preparing the intelligent tutor MathSpring …
Via: Google Alerts for AI