Pedro was recognized “for his foundational research in data stream analysis, cost-sensitive classification, adversarial learning, and Markov logic networks.”
Pedro carried out some of the earliest research on mining data streams – his VFML toolkit is one of the best open-source resources for stream mining. Another key contribution was the MetaCost algorithm, perhaps the most widely used algorithm for cost-sensitive classification. He was a pioneer in social network mining, where he defined the influence maximization problem and proposed the first algorithms for it. Another area that he pioneered is adversarial learning – important in areas such as spam filtering, fraud detection and counter-terrorism, where the people being modeled by the learning system modify their behavior adversarially in response to the system. He also pioneered the use of machine learning methods in information integration. Most recently Pedro has led the field of statistical multi-relational learning, which is essential for a mature science of knowledge discovery; he proposed Markov logic networks as a means to unify first-order logic and probabilistic graphical models, a formalism that forms the basis of research by many different groups, and of the open-source Alchemy system.
Read the KDD press release here.