Abstract
In common binary classification scenarios, the presence of both positive and negative examples in training datais needed to build an efficient classifier. Unfortunately, in many domains, this requirement is not satisfied andonly one class of examples is available. To cope with this setting, classification algorithms have been introducedthat learn from Positive and Unlabeled (PU) data. Originally, these approaches were exploited in the context ofdocument classification. Only few works address the PU problem for categorical datasets. Nevertheless, theavailable algorithms are mainly based on Naive Bayes classifiers. In this work we present a new distance basedPU learning approach for categorical data: Pulce. Our framework takes advantage of the intrinsic relationshipsbetween attribute values and exceeds the independence assumption made by Naive Bayes. Pulce, in fact,leverages on the statistical properties of the data to learn a distance metric employed during the classificationtask. We extensively validate our approach over real world datasets and demonstrate that our strategy obtainsstatistically significant improvements w.r.t. state-of-the-art competitors.