Research

mrebollo MIguel Rebollo comunica2 complex networksI belong to the Informatics Technology-Artificial Intelligence Group (GTI-IA). These are research interests and the projects I am working in.

Interests

The topic of my Ph.D. Thesis in AI was temporal knowledge representation and reasoning, directed by Vicent Botti and Eva Onaindía. The general idea was (i) to propose a temporal model to specify the beliefs of an intelligent agent, and (ii) to create predictable methods to apply temporal reasoning in a strict real-time environment. The dissertation is titled Real-time Imagination.

I continue working in the multiagent area, applied to reach agreements on autonomous entities and the dynamics of emergent, distributed complex systems. In the Publications section, you can see the main results achieved during these years.

Currently, I have completed a master in Physics of Complex Systems and I am finishing my Ph.D. in this new area. It is entitled Generalization of Consensus Processes in Complex Networks,  directed by Rosa M. Benito, from the Complex System Group (GSC) in the Polytechnic University of Madrid (UPM). I am focused on complex networks and chaotic systems and this is what I apply to the social network analysis. It is particularly interesting when both, humans and artificial entities (agents) share the same environment and interact with each other. In that way, agents benefit from the knowledge of the people and people take advantage of the calculus capacity and analysis of the intelligent agents.

And so the game went on. Our friend was absolutely correct: nobody from the group needed more than five links in the chain to reach, just by using the method of acquaintance, any inhabitant of our Planet

Chain Links, Frigyes Karinthy, 1929

Other research lines

From the application of network analysis and artificial intelligence to other areas, other projects arise.

uTool: it began as a tool to extract tweets and it has received more and more functionalities for network analysis, In includes also geolocated information and sentiment analysis among others

Race analysis. The behavior of the participants in races can’t be explained as an isolated experience and the rest of runners in the neighborhood affect to the performance. To consider the race as a dynamic complex networks gives an insight to patterns that emerge and are common for all the races independently on the distance or the type of race (asphalt or trail):

Sentiment and topic analysis. of interactions in Twitter and other social networks.