An Algorithm of Association Rules Mining Based on Restricted Conditional Probability Distribution

There are excessive and disordered rules generated by traditional approaches of association rule mining, many of which are redundant, so that they are difficult for users to understand and make use of. Agrawal et al pointed out the bottleneck of transaction number increase association rules according to the index increase. To solve this problem, a new method was represented, which is based on restricted conditional probability distribution to get a condensed rules set by removing redundant rules. Our set of rules is more meaningful, more concise and users are interested in than others. Especially, the number of rules in rules-set has been reduced greatly. We find that it is an effective method of association rules mining from examples, finally poses future research.