-2
$\begingroup$

I’m developing a mutual information based classifier, the output of the classifier is {{0,4,5,6},{1,2,3},{7},{8},{9}} and the correct decision output is {{0,1,2,4,5,6,9},{3},{7,8}}. How to compute the accuracy of this classifier according to the concept of True positive, true negative, false positive and false negative found on https://en.wikipedia.org/wiki/Evaluation_of_binary_classifiers

Thanks

$\endgroup$
  • 2
    $\begingroup$ Welcome to CS.SE! What approaches have you considered? Help us help you, by showing us what approaches you've already considered and why you rejected them. What does it mean to say that the output is {{0,4,5,6},{1,2,3},{7},{8},{9}}? Do you really not know the correct class for each instance in your training set? If not, that's the first thing you should fix. Also, the title you have chosen is not well suited to representing your question. Please take some time to improve it; we have collected some advice here. Thank you! $\endgroup$ – D.W. Jul 19 '16 at 0:05
  • 1
    $\begingroup$ Welcome to Computer Science! The title you have chosen is not well suited to representing your question. Please take some time to improve it; we have collected some advice here. Thank you! $\endgroup$ – Raphael Jul 19 '16 at 10:01
0
$\begingroup$

The basic idea would be to take every pair {(0,1), (0, 2), (0, 3), ... ,(8, 9)}. For each one, ask whether it is correctly classified as in the same group or not.

In your example:

If you take the pair (0, 1), this would be a false negative because it is incorrectly classified in different groups. Then if you move on to the pair (0, 4), this would be a true positive because they are correctly classified in the same group. Then later when you take the pair (1, 3), this would be a false positive because they are incorrectly classified int he same group. If you take the pair (7, 9), this would be a true negative because they are correctly classified into different groups.

You can classify each pair as either being a TP, TN, FP, or FN then apply the formulas you found on the wikipedia page!

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.