A Learning Agent Shell and Methodology for Developing Intelligent Agents

Disciple is a learning agent shell and methodology for efficient development of personal agents. The process of developing an agent with Disciple relies on importing ontologies from existing repositories of knowledge and on teaching the agent to perform various tasks in a way that resembles how an expert would teach an apprentice, by giving the agent examples and explanations, and by supervising and correcting its behavior. The paper presents briefly the architecture of Disciple, the process of developing a Disciple agent, and various agents developed with Disciple. Then it discusses several general issues in the design and development of intelligent agents, and how they are dealt with in Disciple.