Feature selection based on cluster assumption in PU learning
This addresses feature selection for data mining in scenarios with limited labeled data, such as real-world PU tasks, but is incremental as it builds on existing cluster assumption ideas.
The paper tackles feature selection in positive-unlabeled (PU) learning, where only a few positive labels are available, by proposing FSCPU, a method based on the cluster assumption, and shows it achieves competitive performance compared to 10 conventional algorithms on open datasets.
Feature selection is essential for efficient data mining and sometimes encounters the positive-unlabeled (PU) learning scenario, where only a few positive labels are available, while most data remains unlabeled. In certain real-world PU learning tasks, data subjected to adequate feature selection often form clusters with concentrated positive labels. Conventional feature selection methods that treat unlabeled data as negative may fail to capture the statistical characteristics of positive data in such scenarios, leading to suboptimal performance. To address this, we propose a novel feature selection method based on the cluster assumption in PU learning, called FSCPU. FSCPU formulates the feature selection problem as a binary optimization task, with an objective function explicitly designed to incorporate the cluster assumption in the PU learning setting. Experiments on synthetic datasets demonstrate the effectiveness of FSCPU across various data conditions. Moreover, comparisons with 10 conventional algorithms on three open datasets show that FSCPU achieves competitive performance in downstream classification tasks, even when the cluster assumption does not strictly hold.