Given a set of (non-linear) ranked values for a single attribute, would it be acceptable to:
(1) First apply a maximizing function to these values and then
(2) Apply a spatial clustering algorithm to the fitted values?
I'm essentially looking for a ordinal-based spatial clustering approach. Any comments/suggestions, or links to programming modules, would be very much appreciated.
Are you talking about optimisation ( ie. maximizing the objective function using genetic algorithm) and data clustering (k-means, fuzzy, NNT)?
Hi Habibi Mohamed, thanks for your question. I'll be a bit more specific, I'm looking for an approach to spatially cluster a attribute that is of ordinal type. If it was ratio data, all would be well. I was thinking about applying a function to data, and then apply a spatial clustering approach thereafter to model fit values. However, this was just me being a bit fustrated with not finding a programming package in python or R that can do the job. Of course I may not be using the right search words on the web to find what I want. Any advice would be very much appreciated.
- Badges
- Science topic
- Similar topics
- Computer Science
- Program