Multi-view Positive and Unlabeled Learning

摘要: Learning with Positive and Unlabeled instances (PU learning) arises widely in information retrieval applications. To address the unavailability issue of negative instances, most existing PU learning approaches require to either identify a reliable set from unlabeled data or estimate probability densities as an intermediate step. However, inaccurate negative-instance identication poor density estimation may severely degrade overall performance nal predictive model. this end, we propose novel method based on ratio without constructing any sets estimating densities. further boost performance, extend our proposed multi-view manner by utilizing multiple heterogeneous sources. Extensive experimental studies demonstrate eectiveness methods, especially when positive labeled are limited.