My main research interest is in developing querying processing strategies to support non-traditional applications. I have conceived a new strategy, named cherry picking, which analyses data dependency among attribute values to dynamically decide on tuple routing through expensive predicates. This problem is typical in many scientific applications that need to combine results from the evaluation of scientific programs. We are currently investigating how can one abstract data out of the scientific models that compute them.

We have also explored queries that compute inter and extrapolations. A new operator named "Orbit" evaluates cyclic routing of tuples through steps of the inter-extrapolation process. An adaptive execution model support the computation over large samples and within large iterations.

A second axe of research concerns applying database techniques to the Semantic Web. We are investigating the role of functional dependencies on ontologies and how can it be introduced into OWL-DL. In addition, I've being applying database query processing techniques to enhance reasoning tasks. Most of my research has produced software prototypes to be used by the targeted community. The most recent one can be found at (http:// codims dot epfl dot ch). CoDIMSD is a parallel query processing system configured for running Semantic Web service discovery queries.

My main research interest is in developing querying processing strategies to support non-traditional applications. I have conceived a new strategy, named cherry picking, which analyses data dependency among attribute values to dynamically decide on tuple routing through expensive predicates. This problem is typical in many scientific applications that need to combine results from the evaluation of scientific programs. We are currently investigating how can one abstract data out of the scientific models that compute them.

We have also explored queries that compute inter and extrapolations. A new operator named "Orbit" evaluates cyclic routing of tuples through steps of the inter-extrapolation process. An adaptive execution model support the computation over large samples and within large iterations.

A second axe of research concerns applying database techniques to the Semantic Web. We are investigating the role of functional dependencies on ontologies and how can it be introduced into OWL-DL. In addition, I've being applying database query processing techniques to enhance reasoning tasks. Most of my research has produced software prototypes to be used by the targeted community. The most recent one can be found at (http:// codims dot epfl dot ch). CoDIMSD is a parallel query processing system configured for running Semantic Web service discovery queries.