COMP9417: Machine Learning and Data Mining

This may be used to summarise the predictive performance of a classifier on test data. It is commonly encountered in a two-class format, but can be generated for any number of classes. Suppose we have a two-class problem with classes referred to as positive and negative. A single prediction by a classifier can have four outcomes which are displayed in the confusion matrix of Table 1. Predicted Class positive negative Actual positive true positive false negative Class negative false positive true negative Table 1: Two-class confusion matrix (also known as a 2× 2 contingency table).