An intelligent tutoring system (ITS) is a computer system that aims to provide instant and customized instruction or feedback to learners.
The goal of this tutoring system is to provide access to high-quality education to each student. To achieve this goal, the system replicates the demonstrated benefits of one-to-one, personalized tutoring.
It works — to an extent.
Intelligent tutoring systems have proven effective in teaching specific subjects such as grammar and algebra.
In fact, authorities have incorporated a type of ITS, the Cognitive Tutor, into the mathematics curricula in some U.S. high schools. Furthermore, it’s producing improved student learning outcomes in standardized tests.
So, how does ITS work?
How Intelligent Tutoring Systems Works
It begins with using an AI method to teach the computer several ways to solve a problem in a topic. These include multicolumn addition as well as correcting the computer if it responds incorrectly.
After learning to solve the problem in the way that it was taught, the system then generalizes to address other issues in the topic. In so doing, its approach to problems could differ from the teacher’s.
A Ph.D. student at Carnegie Mellon University‘s Human-Computer Interaction Institute, Daniel Weitekamp, said:
“A student might learn one way to do a problem, and that would be sufficient. But a tutoring system needs to learn every kind of way to solve a problem.”
Unfortunately, creating an ITS has never been easy. The process is complicated and time-consuming, taking as much as 200 hours of development for each hour of tutored instruction.
As a result, it’s a task that has always been reserved for programmers instead of teachers. Now, the researchers at CMU have developed a method that eases the process significantly.
Speeding Up the Development of Intelligent Tutors
The new method uses machine learning to simulate how students learn.
Weitekamp developed a user-friendly interface for the machine learning engine that employs a “show and correct” process. As you may have guessed, that’s easier than programming the whole system.
That means teachers — instead of AI programmers — will be able to build computerized lessons. So, teachers that prefer a specific notation in Chemistry or a method of teaching addition can include it in their intelligent tutoring system.
The lead author of the study, Professor Ken Koedinger, noted other ways that teachers can benefit from building their systems.
“The machine learning system often stumbles in the same places that students do,” Koedinger explained. “As you’re teaching the computer, we can imagine a teacher may get new insights about what’s hard to learn because the machine has trouble learning it.”
Comments (0)
Most Recent