Fig. 7From: Identifying pre-outbreak signals of hand, foot and mouth disease based on landscape dynamic network markerThe performance comparisons of L-DNM-based and machine-learning-based methods. It can be seen that the L-DNM-based surveillance system performs better than the logistic regression or SVM using only the records of Tokyo clinic visits. The AUC of L-DNM is 0.8831, while that of support vector machine is 0.8441 and that of logistic regression is 0.8396Back to article page