My research interest focuses on Natural Language Processing and Machine Learning for Text Mining. Some of the specific areas in which I am currently working on are Question Answering, Automatic Summarization, Document Clustering and, specially, Adaptive Information Extraction, which was the general topic of my PhD. thesis (An Information Extraction System Portable to New Domains [PDF]). Within this framework, I am interested in developing Machine Learning techniques to learn knowledge from large collections of document, knowledge useful for robustly dealing with standard and non standard text, such as semistructured Web pages and speech transcripts. Most of these efforts are funded by the CHIL european project (Computers in the Human Interaction Loop).