Feature: Allow unsupervised random forest for distance calculation. (a data preparation step)

[jacekkotowski]

As in article http://gradientdescending.com/unsupervised-random-forest-example/ unsupervised random forest can be used for distance calculation. Perhaps it can be used as a preparation step for the data. Such a solution is very convenient, since Random Forest will take care for normalization of data and categorical variables. Used with PAM it shows better separation of species in iris dataset than clustering with distances calculated the standard way.

[EmilHvitfeldt]

Hello @jacekkotowski Thank you for this suggestion!

I will get around to doing some more {tidyclust} development, hopefully in the next couple of months, and will see if we can add this then.