Positive-Unlabeled Classification under Class Prior Shift and Asymmetric Error
This addresses practical limitations in PU classification for scenarios where test data distributions and error penalties vary, offering incremental improvements to existing methods.
The paper tackles the problem of binary classification from positive and unlabeled data under class prior shift and asymmetric error, showing that these scenarios are equivalent and proposing two frameworks (risk minimization and density ratio estimation) to handle them, with experiments on benchmark datasets demonstrating their effectiveness.
Bottlenecks of binary classification from positive and unlabeled data (PU classification) are the requirements that given unlabeled patterns are drawn from the test marginal distribution, and the penalty of the false positive error is identical to the false negative error. However, such requirements are often not fulfilled in practice. In this paper, we generalize PU classification to the class prior shift and asymmetric error scenarios. Based on the analysis of the Bayes optimal classifier, we show that given a test class prior, PU classification under class prior shift is equivalent to PU classification with asymmetric error. Then, we propose two different frameworks to handle these problems, namely, a risk minimization framework and density ratio estimation framework. Finally, we demonstrate the effectiveness of the proposed frameworks and compare both frameworks through experiments using benchmark datasets.