ROC-Boosting: A Feature Selection Method for Health Identification Using Tongue Image

Figure 6

Comparative results of the negative/positive ratio and conditions, as well as the solely AUC value condition. Although the AUC value on the right image (0.727) is slightly larger than the left image (0.723), the solely AUC value condition obtains this superiority, because it is inclined to predict all subjects as with disease. When the number of positive and negative subjects is unbalanced, prediction tending to the major class is the most common phenomenon.

(a)

(b)