تقرير
A Skew-Sensitive Evaluation Framework for Imbalanced Data Classification
| العنوان: | A Skew-Sensitive Evaluation Framework for Imbalanced Data Classification |
|---|---|
| المؤلفون: | Du, Min, Tatbul, Nesime, Rivers, Brian, Gupta, Akhilesh Kumar, Hu, Lucas, Wang, Wei, Marcus, Ryan, Zhou, Shengtian, Lee, Insup, Gottschlich, Justin |
| سنة النشر: | 2020 |
| المجموعة: | Computer Science |
| مصطلحات موضوعية: | Computer Science - Machine Learning, Computer Science - Artificial Intelligence |
| الوصف: | Class distribution skews in imbalanced datasets may lead to models with prediction bias towards majority classes, making fair assessment of classifiers a challenging task. Metrics such as Balanced Accuracy are commonly used to evaluate a classifier's prediction performance under such scenarios. However, these metrics fall short when classes vary in importance. In this paper, we propose a simple and general-purpose evaluation framework for imbalanced data classification that is sensitive to arbitrary skews in class cardinalities and importances. Experiments with several state-of-the-art classifiers tested on real-world datasets from three different domains show the effectiveness of our framework - not only in evaluating and ranking classifiers, but also training them. Comment: 17 pages, Data-centric Machine Learning Research (DMLR) Workshop at ICML 2023 |
| نوع الوثيقة: | Working Paper |
| URL الوصول: | http://arxiv.org/abs/2010.05995 |
| رقم الأكسشن: | edsarx.2010.05995 |
| قاعدة البيانات: | arXiv |
| الوصف غير متاح. |