Efficient algorithms for discovering high-utility patterns with strong frequency affinities

作者：

Highlights：

• Introduce efficient algorithms for mining discriminative high utility patterns.

• Present an affinity-utility list (AUL) structure.

• Present novel pruning strategies: EAFCS, LA-Prune, d-Prune, and UR-Prune.

• The proposed algorithms significantly outperform the state-of-the-art algorithm.

摘要

•Introduce efficient algorithms for mining discriminative high utility patterns.•Present an affinity-utility list (AUL) structure.•Present novel pruning strategies: EAFCS, LA-Prune, d-Prune, and UR-Prune.•The proposed algorithms significantly outperform the state-of-the-art algorithm.

论文关键词：High utility pattern mining,Frequency affinity,Discriminative,DHUP-Miner,DHUP-Miner*,Parallel algorithm

论文评审过程：Received 1 September 2020, Revised 4 November 2020, Accepted 4 December 2020, Available online 9 December 2020, Version of Record 5 January 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.114464