Skip to main content
- Numeric representation of uncertainty. Our group works with different formalism to represent uncertainty: probability theory, possibility theory, imprecise probabilities, among others, as well as in the concepts and operators which are necessary to manage these formalisms.
- Probabilistic graphical models. Dependence graphs (Bayesian networks, Markov networks, influence diagrams) as basic mechanisms to represent knowledge quantitatively and qualitatively and making decisions. Algorithms that perform inference and learning tasks on these structures. Applications.
- Information retrieval. Development of information retrieval systems for documents. Structured information retrieval. Document categorization and text mining.
- Recommender systems. Development of models that allow personalized recommendations, based on both content-based and collaborative information.
- Detection of genetic factors in complex diseases from genome-wide analysis. Machine learning approaches to predict individual risk to complex diseases. Data management and visualization of gene-disease association patterns and individual profiles of disease risk.