Can you please elaborate on your actual requirement? Maybe it's because I'm not a native English speaker, but never heard of such a thing as a "confirmed dimension" (al least for versions until 10). Did you mean "conformed dimension" instead?
Check this document on data modelling regarding QlikView by QlikTech's Dan English. In short and not very technically indeed, a conformed dimension is a term used in the data warehouse design to refer a subset of data where all the elements have the same structure, although they can come from different tables in the data source, usually denormalized. For example, "Customer" would be a conformed dimension when you can query the Customer table and get all the information you need (date, group, name, family...) In the data source, however, the "Customer" group of attributes and values will be made up of several different tables (Customer "Basics", Customer "Regions", Customer "Groups", Customer "Methods of payment", Customer "Categories", Customer "Bank Account Information" and so on). This other tables are usually highly normalized.