Communications of the ACM
AI Enables Teachers to Rapidly Develop Intelligent Tutoring Systems
Researchers in Carnegie Mellon University created an artificial intelligence-based technique to help educators rapidly develop intelligent tutoring systems.
Credit: Carnegie Mellon University
Carnegie Mellon University (CMU) researchers have created an artificial intelligence-based technique to enable educators to rapidly develop intelligent computerized tutoring systems.
The teachers can teach the computer by demonstrating several ways to solve problems in a topic, correcting the system if it responds erroneously.
The system can not only solve the problems as it was trained to do, but also can generalize to solve all other problems in the topic in ways that the teacher did not demonstrate.
The method employs a machine learning algorithm that models how students learn, and a user-friendly teaching interface that uses a "show-and-correct" process that is significantly easier than programming.
CMU's Ken Koedinger said the technique may enable teachers to produce a 30-minute lesson in about half an hour, which he described as "a grand vision" among intelligent tutor developers.
From Carnegie Mellon University Human-Computer Interaction Institute
View Full Article
Abstracts Copyright © 2020 SmithBucklin, Washington, DC, USA
No entries found