摘要
针对数据集中的关联规则挖掘问题,提出一种基于改进量子粒子群优化(improved quantum particle swarm optimization,IQPSO)算法的关联规则挖掘方法。首先,将数据实例以量子比特形式表示,构建一个基于量子进化算法(quantum evolutionary algorithm,QEA)的关联规则挖掘基础框架。然后,在该基础框架上,采用新的量子角度更新公式,即使用QPSO代替QEA实现关联规则挖掘。最后,为了进一步提高QPSO算法的收敛性能,融入变异机制和动态惯性权重对其进行改进,加快其收敛速度和跳出局部最优的能力。在UCI和课程成绩数据集上的实验结果表明,提出的算法能够快速且有效地挖掘出关联规则,相比其他几种算法,挖掘到的关联规则价值更高。
In view of the existing problems of the association rule mining in big data set,an association rule mining method based on improved quantum particle swarm optimization(IQPSO)algorithm was proposed in this paper.Firstly,the data instances were represented in the form of quantum bits,and a basic framework of association rule mining based on quantum evolutionary algorithm(QEA)was constructed.Then,QPSO was used instead of QEA to construct a new quantum angle updating formula to realize association rule mining.Finally,mutation mechanism and dynamic inertia weight were integrated to improve QPSO so as to accelerate convergence speed and the ability to jump out of local optimum.The experimental results on UCI and student test score data sets show that the proposed method can mine the association rules quickly and effectively and the fitness value of association rules mined by this algorithm is higher than those mined by other algorithms.
作者
吴嵘
张姣玲
刘小兰
WU Rong;ZHANG Jiaoling;LIU Xiaolan(School of Information and Automation,Guangdong Polytechnic of Science and Trade,Guangzhou,Guangdong 510640,China;School of Mathematics and Systems Science,Guangdong Polytechnic Normal University,Guangzhou,Guangdong 510665,China;School of Mathematics,South China University of Technology,Guangzhou,Guangdong 510641,China)
出处
《山东科技大学学报(自然科学版)》
CAS
北大核心
2020年第2期95-104,共10页
Journal of Shandong University of Science and Technology(Natural Science)
基金
国家自然科学基金项目(11801097)
关键词
关联规则挖掘
量子粒子群优化
变异机制
动态惯性权重
量子进化算法
association rule mining
quantum-behaved particle swarm optimization
mutation mechanism
dynamic inertia weight
quantum evolutionary algorithm