A rough association rule is applicable for knowledge discovery

The traditional association rule which should be fixed in order to avoid both that only trivial rules are retained and also that interesting rules are not discarded. In fact, the situations which use the relative comparison to express are more complete than to use the absolute comparison. Through relative comparison we proposes a new approach for mining association rule, which has the ability to handle the uncertainty in the classing process, so that we can reduce information loss and enhance the result of data mining. In this paper, the new approach can be applied in find association rules, which has the ability to handle the uncertainty in the classing process and suitable for all data types.