Accuracy confidence boundaries?

izxi / Learning

9 stars 3 forks source link

Accuracy confidence boundaries? #3

Open izxi opened 6 years ago

izxi commented 6 years ago

We have performed a supervised classification on a dataset containing 100 test set instances. Eighty of the test set instances were correctly classified. Which of the following represents the 95% test set accuracy confidence boundaries?

Select one:

[ ] A. 70% and 90%
[x] B. 72% and 88%
[ ] C. 78% and 82%
[ ] D. 76% and 84%

izxi commented 6 years ago

confidence-intervals

johnnieng commented 5 years ago

B. 72% and 88% *

iaminebriki commented 2 years ago

Confidence Interval Example Data: acc=80% => err=20%=0.2 ; const=1.96 (95%) ; n=100 Rule: " err +/- const sqrt( (err (1 - err)) / n) " Application: " 0.2 +/- 1.96 sqrt( (0.2 (1 - 0.2)) / 100) " ......................... " = 0.2 +/- 1.96 sqrt(0.16 / 100) " ......................... " = 0.2 +/- 1.96 0.04 " ......................... " = 0.2 +/- 0.0784 " Results: " err=[ 0.2784 : 0.1216 ] => acc=[0.7216 : 0.8784] => [ 72.16% : 87.84% ] => [ 72% : 88% ] to be more precise! "