The disciple integrated shell and methodology for rapid development of knowledge-based agents

The Disciple Learning Agent Shell (Disciple-LAS) is an integrated set of modules for rapid development of practical end-to-end knowledge-based agents, by domain experts, with limited assistance from knowledge engineers. Disciple-LAS and its associated agent building methodology are presented in (Tecuci et al. 1999). Therefore, in this paper, we introduce two very different agents developed with Disciple-LAS, to show its applicability to a wide range of domains. Then we introduce the different modules that are part of Disciple-LAS, and present their use in the agent building process. Finally we summarize the solutions proposed by the Disciple approach to some of the issues that have been found to be limiting factors in developing knowledge-based agents.