Quiz 1 review
Evaluating Classifiers Reading: T. Fawcett paper, link on class website, Sections 1-4 Optional reading: Davis and Goadrich paper, link on class website
“Confusion matrix” for a given class c ActualPredicted (or “classified”) True False (in class c)(not in class c) True (in class c) TruePositiveFalseNegative False (not in class c)FalsePositiveTrueNegative Evaluating classification algorithms
Accuracy: Fraction of correct answers out of all problems Precision: Fraction of true positives out of all predicted positives: Recall: Fraction of true positives out of all actual positives:
Trading off precision against recall w1w1 w2w2 w 64 o w0w0 +1 x1x1 x2x2 x 64 How can we improve precision (at the expense of recall) with a fixed classifier?
True False (“8”)(“0”) True (“8”) 4010 False (“0”)30120 True False (“8”)(“0”) True (“8”) ?? False (“0”)?? Old, with threshold of 0 New, with threshold of: -∞ Example 1: Assume 200 sample digits, of which 50 have class “8” Precision? Recall? Actual Predicted
True False (“8”)(“0”) True (“8”) 4010 False (“0”)30120 True False (“8”)(“0”) True (“8”) ?? False (“0”)?? Old, with threshold of 0 New, with threshold of: +∞ Precision? Recall? Actual Predicted Example 2: Assume 200 sample digits, of which 50 have class “8”
Results of classifier ThresholdAccuracyPrecisionRecall ∞ Creating a Precision/Recall Curve
9 (“sensitivity”) (1 “specificity”)
10
Results of classifier ThresholdAccuracyTPRFPR ∞ Creating a ROC Curve
Precision/Recall versus ROC curves 12
13
14
15