Research
My research interests are in Artificial Intelligence, mainly in Knowledge Discovery in Databases and Multi-Agent Systems .
My current focus is in the following sub-areas:
Spatial and Spatio-temporal Data Mining
The collection of moving object data (GPS, cellular phone, etc.) is becoming more and more common,
and therefore there is an increasing need for the efficient analysis of
these data in different application domains. The objectives in this topic include the development of new data
mining methods and algorithms to extract knowledge from this kind of data.
Artificial Intelligence applied to Computer Games
We are investigating the application of extended behavior networks to the control of an agent
in the game Unreal Tournament. Extended Behavior Networks (EBNs) are a class of action selection
architectures capable of selecting a good set of actions for complex agents situated in continuous
and dynamic environments. They have been successfully applied to the Robocup, but not
in computer games. We verify the quality of the action selection mechanism and its correctness in a
serie of experiments.
Another ongoing research is investigating the use of data mining techniques to prevent fraud in computer games.
Constructivist Artificial Agents
On the last decade, Artificial Intelligence has seen the emergence of new approaches and methods, and also a
search for new theoretical conceptions, coming from a healthy contact with other disciplines such as biology,
psychology and neurosciences. The exploration of these AI frontiers can make it possible to overcome
some current hard research problems. Among these new ways of conceiving AI there is the Constructivist Artificial
Intelligence, which comprises works on this science that refer to the Constructivist Psychological Theory.
According to Piaget, main author of this psychological paradigm, the human being is born with a few cognitive
structures, but this initial knowledge, joined with basic learning functions, enables the subject to build new
cognitive structures through the active interaction with the environment. Intelligence is the set of mechanisms
that allows the intellectual development, and this development provides the subject adaptation to the
environment. We are developing an agent learning architecture based on the constructivist approach.
We mainly focus on the learning method and the type of environmental regularities that can be discovered.