arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2001.05571cs.LGstat.ML
收藏

非常数类不平衡下的模型评价

On Model Evaluation under Non-constant Class Imbalance

Cisco Systems, Inc.(思科系统公司);Czech Technical University in Prague(布拉格捷克理工大学)

Jan Brabec, Tomáš Komárek, Vojtěch Franc, Lukáš Machlica

摘要:

许多现实世界的分类问题是显着的类不平衡,损害类的利益。适当的评估指标的标准集是众所周知的,但通常的假设是测试数据集的不平衡等于现实世界的不平衡。在实践中,这种假设往往因为各种原因而被打破。报告的结果往往过于乐观,并可能导致关于工业影响和拟议技术的适用性的错误结论。我们介绍的方法专注于评估下的非恒定类不平衡。我们发现,不仅是常用的度量的绝对值,但即使是分类器的顺序相对于所使用的评估度量的不平衡率的变化的影响。最后,我们证明,使用二次抽样,以获得一个测试数据集类不平衡等于在野外观察到的是没有必要的,并最终可能导致分类器的性能估计的显着错误。

英文摘要:

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.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑