Developing an Intelligent Educational Agent with Disciple

Disciple is an apprenticeship, multistrategy learning approach for developing intelligent agents where an expert teaches the agent to perform domain-specific tasks in a way that resembles how the expert would teach an apprentice, by giving the agent examples and explanations, and by supervising and correcting its behavior. The Disciple approach is currently implemented in the Disciple Learning Agent Shell. We make the claim that Disciple can naturally be used by an educator to build certain types of educational agents. The educator will directly teach the Disciple agent how to perform certain educational tasks and then the agent can interact with the students to perform such tasks. This paper presents the Disciple approach and its application to building an educational agent that generates history tests for students. These tests provide intelligent feedback to the student in the form of hints, answer and explanations, and assist in the assessment of students’ understanding and use of higher-order thinking skills.