I have a dataset that contains data that needs to be preserved for e.g. clustering purposes, but whose concrete values do not matter. For example, order numbers or serial numbers with lots of accompanying measurement data. I would like to anonymize the serial numbers and order numbers, but still be able to discern measurement values that belong to the same order number or serial number. How can this be done in Orange3, preferrably w/o using Python? (In Knime Analytics there is a building block for just that.)

  • $\begingroup$ Maybe Edit Domain widget can help you. $\endgroup$ – K3---rnc Mar 23 '20 at 10:49
  • $\begingroup$ I checked it, but unfortunately I see no way to efficiently (or at all) reinterpret a string column as numeric with automated (or configurable) replacement of string values by, say, integers. $\endgroup$ – CKO Mar 23 '20 at 18:53

Your Answer

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

Browse other questions tagged or ask your own question.