A new concordant partial AUC and partial c statistic for imbalanced data in the evaluation of machine learning algorithms
André M. Carrington, Paul W. Fieguth, Hammad Qazi, Andreas Holzinger, Helen H. Chen, Franz Mayr, Douglas G. Manuel · 2020
Read A new concordant partial AUC and partial c statistic for imbalanced data in the evaluation of machine learning algorithms by André M. Carrington, Paul W. Fieguth, Hammad Qazi, Andreas Holzinger, Helen H. Chen, Franz Mayr, Douglas G. Manuel (2020) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/openalex-w2997833216.txt.
Description
André Carrington, Paul Fieguth, Hammad Qazi, and colleagues published this methods paper in BMC Medical Informatics and Decision Making in 2020. It proposes new scores for judging machine learning classifiers on skewed data. Standard scores can mislead when one class is far rarer than the other, as in disease screening.
The usual tool, the area under the ROC curve, mixes useful and useless parts of the curve on such data. Existing partial versions drop information about true negatives. The authors derive a concordant partial area and a matching partial c statistic, and prove the pair agree with each other and rebuild the full measures. They test the scores on classic examples and on two breast cancer datasets.
Catalog record
| CREATOR | André M. Carrington, Paul W. Fieguth, Hammad Qazi, Andreas Holzinger, Helen H. Chen, Franz Mayr, Douglas G. Manuel |
|---|---|
| YEAR | 2020 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 47 KB |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | OpenAlex |