i am working with ID3 algorithm, and i know that classic ID3 basically can handle missing data. But i am trying to code this algorithm, so what should i do if there is missing attribute's value in training dataset? How can i classify the data point that has missing attribute's value?
$\begingroup$
$\endgroup$
Add a comment
|
$\begingroup$
$\endgroup$
1
One approach is to use imputation to estimate a value for the missing attribute, based on other attributes.
See also https://en.wikipedia.org/wiki/Missing_data#Techniques_of_dealing_with_missing_data and https://stats.stackexchange.com/questions/tagged/data-imputation and https://stats.stackexchange.com/questions/tagged/missing-data and https://stats.stackexchange.com/q/208845/2921.
