On-Shelf Utility Mining of Sequence Data

Utility mining has emerged as an important and interesting topic owing to its wide application considerable popularity. However, conventional utility methods have a bias toward items that longer on-shelf time they greater chance generate high utility. To eliminate the bias, problem of (OSUM) is introduced. In this article, we focus on task OSUM sequence data, where sequential database divided into several partitions according periods are associated with utilities periods. address problem, propose two methods, data (OSUMS) OSUMS + , extract high-utility patterns. For further efficiency, also design strategies reduce search space avoid redundant calculation upper bounds prefix extension ( TPEU ) reduced TRSU ). addition, novel structures developed for facilitating utilities. Substantial experimental results certain real synthetic datasets show outperform state-of-the-art algorithm. conclusion, may consume large amount memory unsuitable cases limited memory, while wider real-life applications efficiency.