Itemset Mining on Indexed Data Blocks

This paper presents a novel index, called I-Forest, to support data mining activities on evolving databases, whose content is periodically updated through insertion (or deletion) of data blocks. I-Forest allows the extraction of itemsets from transactional databases such as transactional data from large retail chains. Item, support and time constraints may be enforced during the extraction phase. The proposed index is a covering index that represents transactional blocks in a succinct form and allows different kinds of analysis (e.g., analyze quarterly data). During the creation phase no support constraint is enforced. Thus, the index provides a complete representation of the evolving data. The I-Forest index has been implemented Into the Post-greSQL open source DBMS and exploits its physical level access methods. Experiments have been run for both sparse and dense data distributions. The execution time of the frequent itemset extraction task exploiting the index is always comparable with and for low support threshold faster than the Prefix-Tree algorithm accessing static data on at file