Can someone explain what each of these mean? both in simple terms and in terms of TP, TN, FP, FN?
Also are there any other common metrics that I am missing?
Topic f1score score accuracy machine-learning
Category Data Science
Lets say you are creating a model to predict whether a person has a disease or not.
True Positives (TP): Model predicts yes (they have the disease), and they do have the disease.
True Negatives (TN): Model predicts no, and they don't have the disease.
False Positives (FP): Model predicts yes, but they don't actually have the disease. (Also known as a "Type I error.")
False Negatives (FN): Model predicts no, but they actually do have the disease. (Also known as a "Type II error.")
If you look at confusion metric below
Accuracy: Overall, how often is the classifier correct?
(TP+TN)/total = (100+50)/165 = 0.91
True Positive Rate: When it's actually yes, how often does it predict yes?
TP/actual yes = 100/105 = 0.95
also known as "Sensitivity" or "Recall"
Precision: Model predicts yes, how often is it correct?
TP/predicted yes = 100/110 = 0.91
F1 Score = 2 * (Precision * Recall)/ Precision + Recall
= 2 * .91 * .95 / (0.91+.95)
Other terms which people usually use are :
AUC - Area Under Curve
Cohen's Kappa
Geeks Mental is a community that publishes articles and tutorials about Web, Android, Data Science, new techniques and Linux security.