Evolving Decision Strategies for Computational Intelligence Agents

Authors: Roman Neruda, M. Šlapák
Publication year: 2012
Publication type: Konferenční příspěvek (mezinárodní konf.)
Periodical: Intelligent Computing Theories and Applications
Projects: Learning of functional relationships from high-dimensional data
Bibliographic citation: NERUDA, Roman, ŠLAPÁK, M. Evolving Decision Strategies for Computational Intelligence Agents. In: HUANG, D.S., MA, J., JO, K.H., GROMIHA, M.M., eds. Intelligent Computing Theories and Applications. Berlin: Springer, 2012, S. 213-220. Lecture Notes in Artificial Intelligence, 7390. ISBN 978-3-642-31575-6. ISSN 0302-9743
Dostupné z: doi:10.1007/978-3-642-31576-3_28
DOI: 10.1007/978-3-642-31576-3_28
ID: 384818

Abstract

An adaptive control system for computational intelligence agent within a data mining multi-agent system is presented. As opposed to other approaches concerning a fixed control mechanism, the presented approach is based on evolutionary trained decission trees. This leads to control approach created adaptively based on data tasks the agent encounters during its adaptive phase. A pilot implementation within a JADE-based data mining system illustrates the suitability of such approach.