1
$\begingroup$

I am new to Orange and it seems very useful to do some EDA. One question I have is can we use a widget to modify some values in the dataset? For example, several numerical columns have some values of 0 which does not make sense and should be a "missing data". I do not want to ignore the line as some other column data are important. Also not all 0 should be replaced with "missing data" as for some column, a 0 is a valid data. Can we do that with a widget? I could not figure it.

Thank you for the help!

| improve this question | |
$\endgroup$
0
$\begingroup$

You can accomplish this using a Feature Constructor widget with some conditional logic. For example, if looking at the Titanic survival data set, some fares are marked $0. This does not make sense since tickets have a non-zero value. You can create a new variable called fare_new where valid fares are copied over but 0 values are replaced with missing values. See screenshot below:

enter image description here

| improve this answer | |
$\endgroup$

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.