An Evolutionary Method for Associative Local Distribution Rule Mining

摘要

A method for rule mining for continuous value prediction has been proposed using a graph structure based evolutionary computation technique. The method extracts the rules named associative local distribution rule whose consequent part has a narrow distribution of continuous value. A set of associative local distribution rules is applied to the continuous value prediction. The experimental results showed that the method can bring us useful rules for the continuous value prediction. In addition, two cases of contrast rules are defined based on the associative local distribution rules. The performances of the contrast rule extraction were evaluated and the results showed that the proposed method has a potential to realize contrast analysis between two datasets.