Skip to main content

Table 1 Classification performance of support vector machines based on different kernels.

From: Supervised prediction of drug-induced nephrotoxicity based on interleukin-6 and -8 expression levels

  Linear Polynomial Sigmoid RBF
Balanced accuracy (%) 74.6 75.8 75.7 81.6
Sensitivity (%) 63.9 67.7 70.8 78.7
Specificity (%) 85.3 83.8 80.7 84.2
  1. (RBF=radial basis function. The degree of polynomial kernel is three. The values were estimated using a 3-fold cross validation procedure with 10 random trials, and averaged across three batches of HPTCs.)
\