频繁项集挖掘研究前沿及展望  

Research frontier and prospect of frequent itemset mining

作　　者：张晴 谭旭[2] 吕欣 ZHANG Qing;TAN Xu;LV Xin(College of Systems Engineering,National University of Defense Technology,Changsha,Hunan,China 410073;COME Center,Shenzhen Institute of Information Technology,Shenzhen,Guangdong,China 518172)

机构地区：[1]国防科技大学系统工程学院,湖南长沙410073 [2]深圳信息职业技术学院素质赋能中心,广东深圳518172

出　　处：《深圳信息职业技术学院学报》2024年第1期1-14,共14页Journal of Shenzhen Institute of Information Technology

基　　金：广东省普通高校创新团队及特色创新项目(项目编号:2020KCXTD040,2020KTSCX302)。

摘　　要：频繁项集挖掘是数据挖掘领域的核心任务之一,其目标是发现在数据库中频繁出现的模式。这些模式对于关联规则、分类、异常检测等多个数据挖掘任务都具有重要作用。由于随着项集大小的增加,项集的组合数量呈指数级增长,导致计算复杂性急剧上升,研究人员一直在努力开发高效的算法来解决这一问题。面向频繁项集挖掘的算法、紧凑表示和前沿应用,深入探讨不同技术的的工作原理、优势和局限性,从而对这一领域的研究现状进行全面总结。最后,进一步探讨了该领域的前沿发展趋势,指出计算效率、基于约束的频繁项集挖掘、模式的可解释性以及算法在不同领域的创新应用等未来潜在研究方向。Frequent itemset mining is one of the core tasks in the field of data mining,aiming to discover patterns that frequently occur in a database.These patterns play a crucial role in various data mining tasks such as association rule discovery,classification,and anomaly detection and so on.As the size of the itemset increases,the number of combinations of itemset increases exponentially,leading to a sharp increase in computational complexity.Researchers have been working hard to develop efficient algorithms to solve this problem.This study focuses on algorithms,compact representations,and cutting-edge applications for frequent itemset mining,exploring the working principles,advantages,and limitations of different technologies in depth,in order to comprehensively summarize the research status in this field.Finally,this study further discusses the frontier development trend in this field,and points out the future potential research directions,such as computational efficiency,constraint-based frequent itemset mining,interpretability of patterns and innovative applications of algorithms in different fields.

关 键 词：频繁项集 数据挖掘 模式增长 关联规则 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]