Improving Social Emotional Learning through AI-powered Tutoring
Emotion has a substantial influence on learning, attention, memory, and problem solving. This talk describes artificial intelligence tools that detect emotion, provide one-on-one assistance during online learning, and adjust their responses to each student's learning level.
Problems are modulated and responses generated based on motivating students and advancing learning. Inferences about student knowledge are derived from predictive learning models based on students' faces, gestures, and emotions to make decisions about the next best intervention (e.g., provide video, reduce difficulty of math problem).
These tutoring systems can:
- Perform continuous assessment and they can sleuth the students' responses to address student performance and emotion in real-time.
- Create computer-generated synthetic partners and personalized video content. For example, the tools will shortly be able to generate special computational characters. These generated characters may perhaps work more efficiently than do human tutors.
- Create avatars to answer student queries, provide explanations, assist in problem solving and teach about social awareness and relationship skills.
Cross-modal generative AI will provide ChatGPT-4 responses including hints, explanations, data-driven insights and recommendations.
Understanding the role that emotions play in learning and teaching has supported the design and deployment of more effective online tools and learning experiences. This talk will identify the differential impact of AI tutors on student achievement, self-efficacy, and social emotional learning.
Date and Time
Location
Hosts
Registration
- Date: 05 Dec 2024
- Time: 08:00 PM to 09:30 PM
- All times are (UTC-05:00) Eastern Time (US & Canada)
-
Add Event to Calendar
- 35 Olden St
- Princeton, New Jersey
- United States 08544
- Building: Computer Science Building
- Room Number: 105
Speakers
Bev Woolf
Improving Social Emotional Learning through AI-powered Tutoring
Biography:
Dr. Beverly Woolf is a Research Professor in the College of Information and Computer Sciences, UMass-Amherst. Her team has worked with hundreds of students and dozens of teachers to evaluate online K12 teaching platforms. The team has developed tutors for education and industry and in a variety of disciplines (e.g., chemistry, psychology, and mathematics). Dr. Woolf published the book "Building Intelligent Interactive Tutors," over 250 articles and was lead author on the NSF report Roadmap to Education Technology in which forty experts and visionaries identified the next big computing ideas for education. Dr. Woolf is a Fellow of the Association for the Advancement of Artificial Intelligence and served as PI on numerous NSF and U.S. Dept. of Education awards.
