Short Bio

Currently, I’m focused on extracting relevant and useful information from large repositories and streams of data, using supervised, unsupervised and semi-supervised machine learning algorithms. In real-world scenarios this problem is often exacerbated by the presence of incomplete data and by changes (drifts) in the concepts of interest. In this context, a line of research, being pursued, consists of developing incremental learning algorithms capable of readily update their models while dealing with concept drifts. Another line of research consists of developing parallel Graphics Processing Unit (GPU) implementations of machine learning algorithms with the objective of decreasing substantially the time required to execute them, providing the means to study larger datasets. Moreover, new strategies for handling missing data directly are also being considered.