Based on the functions of ad hoc and query-driven system

the essentials-five constraints for multidimension data mining are discussed. The constraints and the association rules are interpreted using an example of data mining query

and it is expressed with DMQL. Association rules handling is discussed. It produces efficient data mining that dimension/level and rule constraints are used together in mining association rules. It is interesting whether a rule constraint can be pushed deeply into the hierarchy to facilitate mining at the current level of abstraction and deeper levels. In the end

this paper provides a OLAM architecture

its composition

function and benefits are presented.