From Uncertainty to Precision: Enhancing Binary Classifier Performance through Calibration
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910326943907840 |
|---|---|
| author | Machado, Agathe Fernandes Charpentier, Arthur Flachaire, Emmanuel Gallic, Ewen Hu, François |
| author_facet | Machado, Agathe Fernandes Charpentier, Arthur Flachaire, Emmanuel Gallic, Ewen Hu, François |
| contents | The assessment of binary classifier performance traditionally centers on discriminative ability using metrics, such as accuracy. However, these metrics often disregard the model's inherent uncertainty, especially when dealing with sensitive decision-making domains, such as finance or healthcare. Given that model-predicted scores are commonly seen as event probabilities, calibration is crucial for accurate interpretation. In our study, we analyze the sensitivity of various calibration measures to score distortions and introduce a refined metric, the Local Calibration Score. Comparing recalibration methods, we advocate for local regressions, emphasizing their dual role as effective recalibration tools and facilitators of smoother visualizations. We apply these findings in a real-world scenario using Random Forest classifier and regressor to predict credit default while simultaneously measuring calibration during performance optimization. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_07790 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | From Uncertainty to Precision: Enhancing Binary Classifier Performance through Calibration Machado, Agathe Fernandes Charpentier, Arthur Flachaire, Emmanuel Gallic, Ewen Hu, François Machine Learning The assessment of binary classifier performance traditionally centers on discriminative ability using metrics, such as accuracy. However, these metrics often disregard the model's inherent uncertainty, especially when dealing with sensitive decision-making domains, such as finance or healthcare. Given that model-predicted scores are commonly seen as event probabilities, calibration is crucial for accurate interpretation. In our study, we analyze the sensitivity of various calibration measures to score distortions and introduce a refined metric, the Local Calibration Score. Comparing recalibration methods, we advocate for local regressions, emphasizing their dual role as effective recalibration tools and facilitators of smoother visualizations. We apply these findings in a real-world scenario using Random Forest classifier and regressor to predict credit default while simultaneously measuring calibration during performance optimization. |
| title | From Uncertainty to Precision: Enhancing Binary Classifier Performance through Calibration |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2402.07790 |