Machine Learning
Accuracy, Precision, Recall, F1 score, ... are metrics to evaluate which types of models?
Classification models.
Card 1 of 259. Answer: Classification models.
Classification models.
Card 1 of 259. Answer: Classification models.
About · Quizzes · History of AI · To suggest new questions or report an error send me a dm.
Classification models. Accuracy, precision, recall and F1 all compare predicted class labels with true labels, so they apply to any model that outputs a discrete class. They come from the counts in a confusion matrix: true positives (TP), false positives (FP), true negatives (TN) and false negatives (FN).
Accuracy = (TP + TN) / total. Precision = TP / (TP + FP), the share of predicted positives that are correct. Recall = TP / (TP + FN), the share of actual positives found. F1 = 2 · precision · recall / (precision + recall), their harmonic mean. Example: if 10 of 1,000 transactions are fraud, a model that always predicts "not fraud" scores 99% accuracy but 0 recall.
Regression models use other metrics, such as MAE, RMSE and R². A common follow-up is multi-class problems: compute each metric per class, then combine with macro, micro or weighted averaging. Interviewers also check that you know precision and recall depend on the decision threshold.