Training intelligent agents in the semantic web era: The golf advisor agent

By I.N. Athanasiadis
In Proc. of the 2007 IEEE/WIC/ACM International Conferences on Web Intelligence and Intelligent Agent Technology - Workshops , Silicon Valey, California, USA, 2007.

Abstract:
Cover image Agent training techniques study methods to embed empirical, inductive knowledge representations into intelligent agents, in dynamic, recursive or semi-automated ways, expressed in forms that can be used for agent reasoning. This paper investigates how data-driven rule-sets can be transcribed into ontologies, and how semantic web technologies as OWL can be used for representing inductive systems for agent decision-making. The method presented avoids the transliteration of data-driven knowledge into conventional if-then-else systems, rather demonstrates how inferencing through description logics and Semantic Web inference engines can be incorporated into the training process of agents that manipulate categorical and/or numerical data.

Read also... Article (in pdf) - ADMI-07 - IAT-2007