AttributeSelectedClassifier: === Attribute Selection on all input data === Search Method: Best first. Start set: no attributes Search direction: forward Stale search after 5 node expansions Total number of subsets evaluated: 386 Merit of best subset found: 0.208 Attribute Subset Evaluator (supervised, Class (nominal): 51 binaryClass): CFS Subset Evaluator Including locally predictive attributes Selected attributes: 1,2,4 : 3 oz1 oz2 oz4 Header of reduced data: @relation 'fri_c0_250_50-weka.filters.unsupervised.attribute.Remove-V-R1-2,4,51' @attribute oz1 numeric @attribute oz2 numeric @attribute oz4 numeric @attribute binaryClass {P,N} @data Classifier Model Loss function: Hinge loss (SVM) Squared loss (linear regression) binaryClass = 3.3312 (normalized) oz1 + 3.2593 (normalized) oz2 + 4.4586 (normalized) oz4 - 5.77