Fig. 6From: Successful classification of cocaine dependence using brain imaging: a generalizable machine learning approach10-fold cross-validation accuracies (the accuracy values used to select best parameter and voxel sets). These result are obtained with Support Vector Machines (SVM), polynomial kernel, f(x) = 0.007(Φ(x). Φ(x') + 10)4, coefficient R = 1.0, and penalty parameter C = 15. The 10-fold cross validation accuracy peaks with 1500 voxels in 30 clustersBack to article page