Evaluation Metrics in Imbalanced Domains
Most of the studies in imbalanced domains mainly concentrate on two-class problem
as multi-class problem can be simplified to two-class problem. By convention, the
class label of the minority class is positive, and the class label of the majority class is
negative. Table 1 illustrates a confusion matrix of a two-class problem. The first column
of the table is the actual class label of the examples, and the first row presents
their predicted class label. TP and TN denote the number of positive and negative
examples that are classified correctly, while FN and FP denote the number of misclassified
positive and negative examples respectively.