A Learning Agent for Knowledge Extraction from an Active Semantic Network

This paper outlines the development of a learning retrieval agent. Task of this agent is to extract knowledge of the Active Semantic Network in respect to user-requests. Based on a reinforcement learning approach, the agent learns to interpret the user-s intention. Especially, the learning algorithm focuses on the retrieval of complex long distant relations. Increasing its learnt knowledge with every request-result-evaluation sequence, the agent enhances his capability in finding the intended information.

• Simon Thiel
• Stavros Dalakakis
• Dieter Roller

• Reinforcement learning
• learning retrieval agent
• search in semantic networks.

[1] Dalakakis, S., Stoyanov, E., Roller, D.: A Retrieval Agent Architecture
for Rapid Product Development. In: Perspectives from Europe and Asia
on Engineering Design and Manufacture, EASED 2004, X.-T. Yan, Ch-
Y. Jiang, N. P. Juster, (eds.), Kluwer Academic Publishers, 2004, pp. 41-
58.
[2] Dalakakis, S.; Diederich, M.; Roller, D.; Warschat, J.:
Multiagentensystem zur Wissenskommunikation im Bereich der
Produktentstehung-Rapid Product Development. In:
Wirtschaftsinformatik 2005 / Ferstl, O. K.; Sinz, E. J.; Eckert, S.;
Isselhorst, T. (Eds.). Heidelberg: Physica-Verlag, 2005. ISBN 3-7908-
1574-8 pp. 1621-1640.
[3] Stuart J. Russell and Peter Norvig. Artificial Intelligence: a modern
approach. Prentice Hall, 1995.
[4] Richard S. Sutton und Andrew G. Barto. Reinforcement Learning: An
Introduction. Bradford Books, 1998.
[5] Tom M. Mitchel. Machine Learning. McGraw-Hill, 1997.
[6] I. Kreuz, D. Roller: Reinforcement Learning and Forgetting for
knowledge based Configuration, Artificial Intelligence and Computer
Science, 83-121, Nova Science Publishers, Inc., 2005.