Incremental Maintenance of Generalized Multi-supported Association Rules under Transaction Update and Taxonomy Evolution

Abstract

Mining generalized association rules has been recognized as a very important topic in data mining. Earlier work on mining generalized association rules ignores the fact that the taxonomy of items would be changed while new transactions are continuously added into the database. In our previous paper, we have proposed a method to solve this problem with uniform minimum support; however, a uniform minimum support assumption would obstruct the discovery of associations on some high value or new items that are more interesting but much less supported than general trends. In this paper, we examine this problem and propose a novel algorithm, called MMAITTE, which can incrementally update the discovered generalized association rules with multiple minimum supports when the taxonomy of items is evolved with incremental transactions. Experimental results show that our algorithm can maintain its performance even in large amounts of incremental transactions and high degree of taxonomy evolution, and is faster than applying the contemporary generalized association mining algorithms to the whole updated database.