Commonly used software tools produce conflicting and overly-optimistic AUPRC values-Reference-Cited by-同舟云学术

Commonly used software tools produce conflicting and overly-optimistic AUPRC values

Published:2024-02-07 Issue: Volume: Page:
ISSN:
Container-title:
language:
Short-container-title:

Author:

Chen Wenyu,Miao Chen,Zhang Zhenghao,Fung Cathy Sin-Hang,Wang Ran,Chen Yizhen,Qian Yan,Cheng Lixin^ORCID,Yip Kevin Y.^ORCID,Tsui Stephen Kwok-Wing,Cao Qin

Abstract

AbstractThe precision-recall curve (PRC) and the area under it (AUPRC) are useful for quantifying classification performance. They are commonly used in situations with imbalanced classes, such as cancer diagnosis and cell type annotation. We evaluated 10 popular tools for plotting PRC and computing AUPRC, which were collectively used in>3,000 published studies. We found the AUPRC values computed by the tools rank classifiers differently and some tools produce overly-optimistic results.

Publisher

Cold Spring Harbor Laboratory

Reference52 articles.

1. The Precision-Recall Plot Is More Informative than the ROC Plot When Evaluating Binary Classifiers on Imbalanced Datasets

2. Jesse Davis and Mark Goadrich . The relationship between precision-recall and ROC curves. In Proceedings of the 23rd International Conference on Machine Learning, pages 233–240, 2006.

3. Learning from imbalanced data;IEEE Transactions on Knowledge and Data Engineering,2009

4. Ryan Lichtnwalter and Nitesh V Chawla . Link prediction: fair and effective evaluation. In 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, pages 376–383. IEEE, 2012.

5. Reconstruction of enhancer–target networks in 935 samples of human primary cells, tissues and cell lines

Cited by 1 articles. 订阅此论文施引文献订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献

1. Commonly used software tools produce conflicting and overly-optimistic AUPRC values;Genome Biology;2024-05-13