Oracle Data Mining requires a case table in single-record case format, with each record in a separate row. What if some or all of your data is in multi-record case format, with each record in several rows? What if you want one attribute to represent a series or collection of values, such as a student's test scores or the products purchased by a customer?
This kind of one-to-many relationship is usually implemented as a join between tables. For example, you might join your customer table to a sales table and thus associate a list of products purchased with each customer.
Oracle Data Mining supports dimensioned data through nested columns. To include dimensioned data in your case table, create a view and cast the joined data to one of the Data Mining nested table types. Each row in the nested column consists of an attribute name/value pair. Oracle Data Mining internally processes each nested row as a separate attribute.