0
$\begingroup$

Here is my question in my assignment:

You have built a classification model with 90% accuracy but your client is not happy because False Positive rate was very high then what will you do?

This is the question..nothing is given in the background

$\endgroup$
  • $\begingroup$ You have an imbalanced data set, therefore you would deal with it before building a classifier, as the classifier would inevitably be biased in that case. $\endgroup$ – user2974951 Sep 25 '18 at 6:29
  • $\begingroup$ 90% accuracy for an imbalanced data set simply requires the imbalance to be calculated and the accuracy to be adjusted (increased) accordingly. $\endgroup$ – Michael G. Jun 5 at 13:37
  • $\begingroup$ any feedback on my answer? $\endgroup$ – Francesco Pegoraro Jun 5 at 15:10
2
$\begingroup$

I think the only general solution would be to: increase the threshold of the model confidence.

For example you are doing binary classification of dog in images: Dog = 1, No Dog = 0

Generally a model (like Neural Network) would output the probability of the image being 1: if it's > 0.5 then predicts 1 else 0. Increasing the confidence to 0.7 would decrease the False Positive.

$\endgroup$
  • 1
    $\begingroup$ Restating @Francesco’s answer - you can use the ROC curve to select a desired false alarm rate. You’ll have to recode the classification output using this threshold after outputting the probability values. $\endgroup$ – HEITZ Sep 24 '18 at 22:13
  • $\begingroup$ @HEITZ Yes true! $\endgroup$ – Mugdha Bhatnagar Sep 26 '18 at 10:44
  • $\begingroup$ Do you plan to accept the answer? $\endgroup$ – Francesco Pegoraro Sep 26 '18 at 11:21
0
$\begingroup$

This is likely to be caused by an imbalanced dataset. It means that some observations are less numerous, therefore your model is not learning enough about them. A possible solution might be: use Mini-Batch Gradient Descent optimization, and build the mini-batches in a way that the number of observations is balanced across all the classes. This would attribute greater weight to the observations that are less frequent.

$\endgroup$

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.