It is a table that is used to describe the classification model performance on a set of test data for which the true values are known. It is very simple to understand but only the related terms can be confusing. Confusion Matrix allows us to find the measure recall, accuracy, precision, etc. It visualizes the accuracy of a classifier by comparing the actual and predicted classes. The binary confusion matrix is composed of squares:
True Positive (TP) : It predicts values correctly predicted as actual positive.
True Negative (TN) : It predicts values correctly predicted as an actual negative.
False Positive (FP) : It predicts values incorrectly predicted as actual positive.
False Negative (FN) : It predicts values correctly predicted as negative.
