Mining High Utility Itemsets in Large High Dimensional Data

Existing algorithms for utility mining are inadequate on datasets with high dimensions or long patterns. This paper proposes a hybrid method, which is composed of a row enumeration algorithm (i.e., inter-transaction) and a column enumeration algorithm (i.e., two-phase), to discover high utility itemsets from two directions: Two-phase seeks short high utility itemsets from the bottom, while inter-transaction seeks long high utility itemsets from the top. In addition, optimization technique is adopted to improve the performance of computing the intersection of transactions. Experiments on synthetic data show that the hybrid method achieves high performance in large high dimensional datasets.