Agent Academy

Agent Academy’s objective was to develop an integrated environment for embedding and improving intelligence in newly created agents through the use of data mining techniques. The Agent Academy platform functioned as a training facility that supported both:

• the creation of intelligent agents with limited initial referencing capabilities, and
• the training of these agents in order to augment their intelligence efficiently, according to user specifications and preferences.

The agent training process exploited data mining algorithms, which were applied on prior agent behavior data, that were monitored by the Agent Academy platform.

While with CERTH, I was involved in the research, design and development of the Agent Academy platform, its demonstration for sensor data fusion related to air quality. I also coordinated financial reporting of the whole consortium.

Related publications:

• I. N. Athanasiadis, P. A. Mitkas, Social influence and water conservation: An agent-based approach, IEEE Computing in Science and Engineering, 7:65-70, 2005, doi:10.1109/mcse.2005.21.
• I. N. Athanasiadis, P. A. Mitkas, An agent-based intelligent environmental monitoring system, Management of Environmental Quality, 15:238-249, 2004, doi:10.1108/14777830410531216.
• P. A. Mitkas, D. Kehagias, A. L. Symeonidis, I. N. Athanasiadis, A framework for constructing multi-agent applications and training intelligent agents, Lecture Notes in Computer Science (Agent Oriented Software Engineering IV), vol. 2935, pg. 96-109, 2004, Springer, doi:10.1007/978-3-540-24620-6_7.