Orange Forum • View topic - Verbose feature output when clustering or classifying

Verbose feature output when clustering or classifying

A place to ask questions about methods in Orange and how they are used and other general support.

Verbose feature output when clustering or classifying

Postby x0ner » Thu May 03, 2012 4:34

When running one of the clustering algorithms on a dataset the user is shown the end result (data in the respective cluster), but it is unclear what features helped to determine this. For instance, what if there was one feature in particular that brought that whole cluster together. Is there a way to display this as the process is happening?

Also along the same lines is verbose output for classification. If I test my classifiers against my training set and get about a 90% match then it would be helpful to know the items not labeled properly, so more investigation could be done. If the classification was done properly, then which features happened to assist in that decision and in what order.

Return to Questions & Support