Model Evaluation and Selection in ML

Note: False positive is also known as type one error whereas false negative is known as type two error. Confusion Matrix: The matrix of all combinations of the predicted label and a true label is called a confusion matrix. Precision: Precision is an evaluation metric that reflects the situation and obtained by dividing the number of true positives by the sum of true positives and false positives. Note: To increase precision we must either increase the number of true positives the classifier predicts or negative instances in positive class. F1 Score: Combining precision and recall into a single number.