A Hybrid Method for High-Utility Itemsets Mining in Large High-Dimensional Data

摘要

Existing algorithms for high-utility itemsets mining are column enumeration based, adopting an Apriori-like candidate set generation-and-test approach, and thus are inadequate in datasets with high dimensions or long patterns. To solve the problem, this paper proposed a hybrid model and a row enumeration-based algorithm, i.e., Inter-transaction, to discover high-utility itemsets from two directions: an existing algorithm can be used to seek short high-utility itemsets from the bottom, while Inter-transaction can be used to seek long high-utility itemsets from the top. Inter-transaction makes full use of the characteristic that there are few common items between or among long transactions. By intersecting relevant transactions, the new algorithm can identify long high-utility itemsets, without extending short itemsets step by step. In addition, we also developed new pruning strategies and an optimization technique to improve the performance of Inter-transaction.