Effective in version 10.2.1, you can discover similar columns based on column names, column patterns, unique values, and value frequencies in a resource.
Previously, the Similarity Discovery system resource identified similar columns in the source data.
For more information about column similarity, see the