The simplest and easiest to grasp of all, classification accuracy refers to the percentage of correct predictions. In order to calculate accuracy, we divide the number of correct predictions by the total number of instances:

In order for accuracy to hold any substantial value, the dataset must contain an equal number of instances belonging to each class. If the dataset is unbalanced, accuracy will be affected. For example, if a dataset consists of 90% class A and 10% class B, a model that predicts each instance as class A will have 90% accuracy, although it will hold zero predictive power.