If the area under the curve (AUC) is .702, what does this imply? Can we consider this as having a good diagnostic accuracy?
Is ROC curve being used only for screening tests or it can also be used for assessment/diagnostic tests as well?
thanks!
Please see table 12 in the attached article, it gives some metrics on how to interpret the "goodness" of a ROC curves AOC.
Please have a look at this link provided below, it gives more understanding about ROC and AUC Curve and how to evaluate the performance of the model for diagnosis.
In, general the value of AUC in range of 0 to 1. If the AUC is more close to 1 the model predicts the target more precisely. If the AUC is more close to 0 the more predicts the results badly.
https://towardsdatascience.com/understanding-auc-roc-curve-68b2303cc9c5
- Badges
- Science method