Rule Mining for Dynamic Databases

摘要

Association rules identify associations among data items and were introduced in [1]. A detailed discussion on association rules can be found in [2], [8]. One important step in Association rule mining is to find frequent itemsets. Most of the algorithms to find frequent itemsets deal with the static databases. There are very few algorithms that deal with dynamic(incremental) databases. The most classical algorithm to find frequent itemsets in dynamic database is Borders algorithm. But the Borders algorithm is suitable for centralized databases. This paper presents a modified version of the Borders algorithm, called Distributed Borders, which is suitable for Distributed Dynamic databases.