AttributeSelectedClassifier: === Attribute Selection on all input data === Search Method: Attribute ranking. Attribute Evaluator (supervised, Class (nominal): 11 binaryClass): Gain Ratio feature evaluator Ranked attributes: 0.2453 2 oz2 0.2038 1 oz1 0.0808 4 oz4 0 9 oz9 0 3 oz3 0 10 oz10 0 6 oz6 0 7 oz7 0 8 oz8 0 5 oz5 Selected attributes: 2,1,4,9,3,10,6,7,8,5 : 10 Header of reduced data: @relation 'fri_c3_250_10-weka.filters.unsupervised.attribute.Remove-V-R2,1,4,9,3,10,6-8,5,11' @attribute oz2 numeric @attribute oz1 numeric @attribute oz4 numeric @attribute oz9 numeric @attribute oz3 numeric @attribute oz10 numeric @attribute oz6 numeric @attribute oz7 numeric @attribute oz8 numeric @attribute oz5 numeric @attribute binaryClass {P,N} @data Classifier Model Loss function: Hinge loss (SVM) Squared loss (linear regression) binaryClass = -0.2818 (normalized) oz2 + -1.0127 (normalized) oz1 + 3.7234 (normalized) oz4 + 0.101 (normalized) oz9 + -5.5907 (normalized) oz3 + -0.0556 (normalized) oz10 + 1.2873 (normalized) oz6 + -0.253 (normalized) oz7 + -0.3788 (normalized) oz8 + 2.9763 (normalized) oz5 + 0.04