I'd like to know if there are some methods to correct a specific misclassified instance of interest (e.g. in a Logistic Regression or Random Forest).
Like maybe increasing the error for that particular observation or some other approach.
Thanks.
5
-
$\begingroup$ Have a look at boosting. $\endgroup$– user2974951Mar 13 at 9:27
-
$\begingroup$ I'm familiar with that but is it possible to boost one particular instance? $\endgroup$– MetricianMar 13 at 9:36
-
$\begingroup$ No. You can use class weight in case of classification (if you want to go down that road). $\endgroup$– user2974951Mar 13 at 9:41
-
$\begingroup$ Random Forest and Logistic Regression objectives are very different. With logistic regression, you can adapt the objective for what you are looking for. Not in Random Forest which is entropy based $\endgroup$– qmeeusMar 13 at 11:18
-
$\begingroup$ I see. So other than that are there no other known methods to correct one misclassified instance? $\endgroup$– MetricianMar 14 at 14:51
Add a comment
|