Mining Incrementally Closed Itemsets over Data Stream with the Technique of Batch-Update

摘要

Currently incremental mining techniques can be divided into two groups: direct-update technique and batch-update technique. Mining closed item sets is one of the core tasks of data mining. In addition, advances in hardware technology and information technology have created huge data streams in recent years. Therefore, mining incrementally closed item sets over data streams with the batch-update technique is necessary. Incremental algorithms are always associated with an intermediate structure such as tree, lattice, table... In the previous study, the author proposed an intermediate structure which is a linear list called constructive set. In this paper, an incremental mining algorithm based on the constructive with the batch-update technique is proposed in order to mine data streams.