clusterings

Specifies column names to use for data clustering. Up to four top-level columns are allowed, and should be specified in descending priority order.