User-Centric Mining of Association Rules

Abstract. Association rule mining is an important mining function be-cause of its completeness. It has been even used to generate classiers and to segment database. However the current model of association rule mining poses problems regarding exibility while handling evolv-ing databases. The expensive process of massive rule discovery discour-ages frequent application of mining algorithm for exploratory mining. We present a methodology to reuse computation and provide an environment that encourages fair and exible exploration of database through win-dows. The environment allows multiple data miners to simultaneously explore the database without causing interference either to the database operation or among each other