My research is currently supported by two European projects:

"GraphRules: Rule Discovery, Exploration and Visualization of Collaborative Graph Structures" (2010-2014) Marie-Curie International Reintegration Grant > Our goal here is design algorithms that can extract useful knowledge from static and evolving graph structures. We are also considering applications of these these techniques in data visualization, data streams and recommendation systems. [more...]

"Exact Mining from InExact Data" (2011-2016) European Research Council (ERC) Starting Grant > We are interested in designing data transformations (such as anonymization, compression, right-protection, etc.) that do hinder the "mining capacity" of a dataset. How can we guarantee that mining the original or the transformed dataset would give identical results for a set of mining and learning algorithms? Are there any commonalities between the various algorithms and transformations? [more...]