Rule discovery algorithms that are discovered in data mining helps generating an array of rules and patterns. Sometimes it also exceeded the size of the existing database and only a fraction of that proves to be useful for the users. In the process of knowledge discovery it is important to interpret the discovered rules and patterns. When there is a huge number of rules and patterns it become almost difficult to choose and analyze the most interesting among all .
For example it might not be a good idea to provide the user with an association rules list, raked by their support and confidence. This might not also be a good way of organizing the set of methods and rules and on another it can also overwhelm the users. It is also not important that all the rules and methods are interesting and it depends on a variety of reasons.
A useful data mining system must able to generate methods and rules feseability thus providing flexible tools for the purpose of rule selection. In the association of rule mining various approaches for the processing of the discovered rules are discussed before. Another approach is also made for grouping of similar rules that goes well for a moderate quantity of rules. Clusters can be created in case of too many numbers of rules and method.
A flexible approach allows to identify the rules that have special values for the users. It is done through union queries of data or templates. Moreover, this approach is just perfect for complementing the grouping rule approach. By the concept of inductive database the importance if data mining has been highlighted. It also allows the clients to query about the pattern and rules as well as about the data and the models extracted from it.