�ݺ�ߣ

Computing Accuracy Precision And Recall

•

11 likes•11,986 views

Nicolas Bettenburg

Accuracy measures the percentage of correct predictions out of the total number of predictions. Precision measures the percentage of positive predictions that were actually correct. Recall measures the percentage of positive cases that were correctly identified.

Calculating Accuracy, Precision and Recall
PREDICTION
true false

true
REALITY a b
false
c d

(a+d)
accuracy: “what percent of the prediction were correct?”
(a+b+c+d)

(a)
precision: “what percent of positive predictions were correct?”
(a+c)

(a)
recall: “what percent of positive cases were caught?”
(a+b)
Samstag, 19. Januar 2008 1

More Related Content

Computing Accuracy Precision And Recall

1. Calculating Accuracy, Precision and Recall PREDICTION true false true REALITY a b false c d (a+d) accuracy: “what percent of the prediction were correct?” (a+b+c+d) (a) precision: “what percent of positive predictions were correct?” (a+c) (a) recall: “what percent of positive cases were caught?” (a+b) Samstag, 19. Januar 2008 1