Question

Question 1 2 pts Sensitivity and Specificity are plotted on an ROC Curve. True False Question 2 2 pts To obtain an honest estimate of future classification error, we use the classification matrix that is computed from validation data test data training data

Answer 1

It is true that sensitivity and specificity are plotted on Roc Curve . The x-axis shows t-specificity (False positive fraction) The 4-axis shows sensitivity (True positive fraction) a 2: "Validation set " is the independent eet used to evaluate the performance of a fully specied classifier. Hence, to obtain an honest estimate of future classification error, we used the class. matrix that is computed from - validation data. @3: The classification matrix produces four outcomes * True positive a true negative - False positive - False negative True classification rate = TP + TN TPATN+FNPFP Misclassification mate FPT IN TPT TNTENT FP Hence true Matrix first au. Let complete the confusion I Predicted Actual 1 IO | 920 Foted 32 30 88 952 Pred. Total fraudelant = 88 Pred Total Non-fraud. = 952 Error rate : 32758 = 952+ 8 8 90 - 10.08651ity 1040