Seok Won Lee

AAAI Conference 1999 Conference Paper

An Integrated Shell and Methodology for Rapid Development of Knowledge-Based Agents

• Gheorghe Tecuci
• Mihai Boicu
• Kathryn Wright
• Seok Won Lee
• Dorin Marcu
• Michael Bowman
• George Mason University

This paper introduces the concept of learning agent shell as a new class of tools for rapid development of practical endto-end knowledge-based agents, by domain experts, with limited assistance from knowledge engineers. A learning agent shell consists of a learning and knowledge acquisition engine as well as an inference engine and supports building an agent with a knowledge base consisting of an ontology and a set of problem solving rules. The paper describes a specific learning agent shell and its associated agent building methodology. The process of developing an agent relies on importing ontologies from existing repositories of knowledge, and on teaching the agent how to perform various tasks, in a way that resembles how an expert would teach a human apprentice when solving problems in cooperation. The shell and methodology represent a practical integration of knowledge representation, knowledge acquisition, learning and problem solving. This work is illustrated with an example of developing a hierarchical non-linear planning agent.

AAAI Conference 1999 Short Paper

Knowledge Base Revision through Exception-Driven Discovery and Learning

• Seok Won Lee
• Gheorghe Tecuci
• George Mason University

We are enhancing Disciple by developing a mixed-initiative multistrategy approach to KB revision that will result in an extended and domain-adapted ontology, as well as a set of rules with fewer (if any) exceptions. We are developing two classes of KB revision methods, a class of local methods and a class of global methods.

AAAI Conference 1999 Conference Paper

The Disciple Integrated Shell and Methodology for Rapid Development of Knowledge-Based Agents

• Mihai Boicu
• Kathryn Wright
• Dorin Marcu
• Seok Won Lee
• Michael Bowman
• Gheorghe Tecuci
• George Mason University

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.

AAAI Conference 1998 Short Paper

Building Agents from Shared Ontologies through Apprenticeship Multistrategy Learning

• Kathryn Wright
• Seok Won Lee