Using a projection-based approach to mine frequent inter-transaction patterns |
| |
Authors: | Chun-Sheng Wang Kuo-Chung Chu |
| |
Affiliation: | 1. Data Science Research Center, Yazd University, Yazd, 89195-741, Iran;2. Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, 518055, China;1. Rutgers, The State University of New Jersey, Department of Accounting and Information Systems, One Washington Park, Newark, NJ 07102-3122, United States of America;2. Southwestern University of Finance and Economics, Chengdu 611130, China |
| |
Abstract: | In this paper, we propose an algorithm called PITP-Miner that utilizes a projection based approach to mine frequent inter-transaction patterns efficiently. The algorithm only searches for local frequent items in a projected database that stores potential local inter-transaction items and partitions the database into a set of smaller databases recursively. In addition, two pruning strategies are designed to further condense the partitioned databases and thus accelerate the algorithm. Our experiment results demonstrate that the proposed PITP-Miner algorithm outperforms the ITP-Miner and FITI algorithms in most cases. |
| |
Keywords: | |
本文献已被 ScienceDirect 等数据库收录! |
|