AI helps improve academic performance: Study

A new study through a personalized AI-powered robot to interact with students has revealed the positive impact of integrating artificial intelligence (AI) and robotics into education at elementary school level.

The groundbreaking project was led by Dr. Jinane Mounsef, Chair of the Rochester Institute of Technology (RIT) of Dubai’s Electrical Engineering and Computing Sciences Department, resulting in an average 8% improvement in academic performance, versus outcomes from human teaching.

Combining cognitive and emotional assessment to deliver tailored educational experiences, the tutoring system was tested with a group of grade four school students studying mathematics. In addition to measuring the speed and accuracy of a student’s performance, the robot conducted real-time facial expression analysis to determine the learner’s state of mind. The metrics were combined to classify students into three categories: Proficient Students (Prof.S), Meeting-Expectations Students (MES), and Developing Students (DVS). These classifications were then used to deliver customized learning content, motivational messages, and constructive feedback to the students.

The robot, known as Duet, employed powerful machine learning algorithms and the ROS framework to predict a student’s proficiency level with 100% accuracy through indicators such as test scores, task completion time, and emotional engagement. It was then able to adjust the challenges and learning materials dynamically to suit each student’s needs. The process also incorporated feedback mechanisms for participants to share their thoughts on the experience of learning through robotics and AI.

Explaining the rationale and findings of the research, Dr. Mounsef said, “We wanted to explore whether personalization enhances the effectiveness of a robotic tutor in improving learning outcomes. To do this, we considered how engagement and academic performance were impacted by the personalization of robotic tutoring and how the results compared to a human teacher, which serves as a benchmark for evaluating the system’s impact. Through post-diagnostic exams we found that the experimental group of students using the AI-robot system showed a significant improvement rate over the control group.This demonstrates that such systems can provide a powerful tool to improve efficiency and augment education outcomes.”

The research was the subject of an academic paper entitled, ‘CARE: towards customized assistive robot-based education’, authored by Dr. Mounsef, alongside RIT Dubai graduate student, Nafisa Maaz, and Assistant Professor at Lebanese American University Beirut, Dr. Noel Maalouf. The paper was published by Frontiers in Robotics and AI, a world-leading, multidisciplinary scientific journal.

Building on the project, Dr. Mounsef is now preparing to collaborate with colleagues at RIT New York to develop an AI-powered social robot.