IDHUP: Incremental Discovery of High Utility Pattern

<p>As a sub-problem of pattern discovery, utility-oriented mining has recently emerged as focus researchers’ attention and offers broad application prospects. Considering the dynamic characteristics input databases, incremental utility methods have been proposed, aiming to discover implicit information/ patterns whose importance/utility is not less than user-specified threshold from databases. However, due explosive growth search space, most existing perform unsatisfactorily under low threshold, so there still room for improvement in terms running efficiency pruning capacity. Motivated by this, we provide an effective efficient method called IDHUP designing indexed partitioned list structure employing four strategies. With proposed data structure, can only dynamically update values but also avoid visiting non-occurred patterns. Moreover, further exclude ineligible unnecessary exploration, put forward remaining reducing strategy three other revised Experiments on various datasets demonstrated that designed algorithm best performance time compared state-of-the-art algorithms.</p> <p> </p>