A Heuristic Data Sampling Approach for Association Rule Classification under a Big Data Environment

摘要

This study proposes a heuristic data sampling approach to solve the problem of mining big data without changing the association rule classification method. The proposed heuristic data sampling approach consists of two parts. The first part is a heuristic sampling method applied in the initial phase, which samples representative data from a big data set with important, discriminative attributes. Then the second part deals with the incremental big data problem. Merging the sampled data from both the preliminary and incremental data sets and their classifiers, it is possible to apply them to verify the combined classifier.