I'm trying to run clustering only with categorical variables. Since Kmeans is applicable only for Numeric data, are there any clustering techniques available?
I have 30 variables like zipcode, age group, hobbies, preferred channel, marital status, credit risk (low, medium, high), education status, etc. If I convert each of these variable in to dummies and run kmeans, I would be having 90 columns (30*3 - assuming each variable has 4 factors). Is this correct?