Orange Forum • View topic - Ratio/Interval Vs. Continuous Attribute Types

Ratio/Interval Vs. Continuous Attribute Types

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

Ratio/Interval Vs. Continuous Attribute Types

Postby PaintingInAir » Mon Nov 25, 2013 8:01

I'm currently taking a data mining class, and have been using Orange for most of my data mining calculations. My current project is on clustering, and I've been using Orange as usual. In class I've learned about various attribute types such as Ratio, Interval, Binary, Nominal, and Ordinal. I know that Ratio and Interval are continuous attribute types, and Binary, Nominal, and Ordinal are discrete attribute types. What I'm wondering is how Orange differentiates between these different continuous and discrete attribute types when it only allows you to classify attributes as discrete or continuous.

If I am unable to specify the attribute type to Orange then I'll need to know a little about how the distances are calculated so that I can justify it in my report. I've tried looking through the source code, but I'm having trouble finding where the calculations are done.

Return to Questions & Support