My main research areas are artificial intelligence and machine learning. In particular, I have been working on Statistical Relational Learning (SRL), which aims at developing machine learning algorithms for complex data having a relational structure, such as graphs or trees. Such data are typically employed in a variety of fields, like molecular biology, social networks analysis, natural language parsing, and many others. Traditional machine learning algorithms, which typically consider the examples as independent, had to be extended in order to handle relational and structured data. SRL methodologies usually combine the powerful and expressive formalism of first-order logic for data description, with graphical models, which are able to handle uncertainty in data. In this context, I have mainly worked on the development of new SRL algorithms, with applications in the fields of bioinformatics, computer vision and traffic forecasting.
Currently, I am working on argumentation mining, a very challenging problem at the intersection of natural language processing, computational linguistics
and machine learning. The goal is to extract arguments from unstructured textual documents, so as to construct a system capable of digesting information,
reasoning and debating around any topic, in the spirit of IBM's Debater. More details
on our work on argumentation mining are available at this page.
See also my publications page to get access to my papers, and my software page for
some code, data sets and links to web servers I have developed.