Not logged in.
Quick Search - Contribution
Contribution Details
| Type | Conference or Workshop Paper |
| Scope | Discipline-based scholarship |
| Published in Proceedings | Yes |
| Title | The binormal assumption on precision-recall curves |
| Authors |
|
| Presentation Type | paper |
| Item Subtype | Original Work |
| Refereed | Yes |
| Status | Published in final form |
| Page Range | 4263 - 4266 |
| Event Title | Proceedings of the 20th International Conference on Pattern Recognition |
| Event Type | conference |
| Event Location | Istanbul, Turkey |
| Event Start Date | August 22 - 2010 |
| Event End Date | August 25 - 2010 |
| Place of Publication | Istanbul, Turkey |
| Publisher | IEEE Computer Society |
| Abstract Text | The precision-recall curve (PRC) has become a widespread conceptual basis for assessing classification performance. The curve relates the positive predictive value of a classifier to its true positive rate and often provides a useful alternative to the well-known receiver operating characteristic (ROC). The empirical PRC, however, turns out to be a highly imprecise estimate of the true curve, especially in the case of a small sample size and class imbalance in favour of negative examples. Ironically, this situation tends to occur precisely in those applications where the curve would be most useful, e.g., in anomaly detection or information retrieval. Here, we propose to estimate the PRC on the basis of a simple distributional assumption about the decision values that generalizes the established binormal model for estimating smooth ROC curves. Using simulations, we show that our approach outperforms empirical estimates, and that an account of the class imbalance is crucial for obtaining unbiased PRC estimates. |
| Export |
BibTeX
EP3 XML (ZORA)
|