On Model Evaluation under Non-constant Class Imbalance
This addresses a practical issue for researchers and practitioners in machine learning who rely on accurate evaluation metrics for imbalanced classification problems, though it is incremental in refining existing evaluation methods.
The paper tackles the problem of evaluating classifiers when test dataset class imbalance differs from real-world imbalance, showing that this discrepancy can lead to overly optimistic results and affect classifier rankings. It demonstrates that using subsampling to match real-world imbalance is unnecessary and can cause significant errors in performance estimates.
Many real-world classification problems are significantly class-imbalanced to detriment of the class of interest. The standard set of proper evaluation metrics is well-known but the usual assumption is that the test dataset imbalance equals the real-world imbalance. In practice, this assumption is often broken for various reasons. The reported results are then often too optimistic and may lead to wrong conclusions about industrial impact and suitability of proposed techniques. We introduce methods focusing on evaluation under non-constant class imbalance. We show that not only the absolute values of commonly used metrics, but even the order of classifiers in relation to the evaluation metric used is affected by the change of the imbalance rate. Finally, we demonstrate that using subsampling in order to get a test dataset with class imbalance equal to the one observed in the wild is not necessary, and eventually can lead to significant errors in classifier's performance estimate.