I have focused my research on  information extraction, retrieval and organization from text documents, using several machine learning (probabilistic and statistical) approaches. 

In these years, I have analyzed different aspects of text/web mining moving from classical text problems as supervised and unsupervised learning, or new representation for text document to more computational linguistic tasks, as language evolution and statistical machine translation. 

I have been part of the European project SMART, where I have applied machine learning techniques to statistical machine translation problems, and I have been also involved in a media analysis project aimed at modeling the mediasphere based on text mining and cross-language analysis techniques.

My current research is centered on thesaurus indexing, machine learning, multilingual multi class document classification,  news analysis and  SMT techniques applied to news  domain.

My last works involve learning curves analysis of a SMT system, confidence estimation in Machine Translation, self-learning for SMT system.