摘要
Apriori算法是一种常用的关联规则分析算法,通过该算法分析,可以得到同一类型的几个事务之间的关联关系,进而对其关联的后续事务进行预测。但该算法也存在不足,为了解决该算法的问题,通过分析了Apriori关联规则算法挖掘过程,然后提出了利用事先剪枝策略进行算法的改进,通过使用Matlab对改进的算法进行了评测,同时使用了U检验思想衡量标准的算法对评测的结果进行信度测算。通过评测发现,改进后的算法提高了产生关联规则的效率及规则集合的精准性。
Apriori algorithm is a commonly used association rule analysis algorithm.Through the analysis of this algorithm,the association relationship between several transactions of the same type can be obtained,and then the subsequent transactions related to it can be predicted.However,the algorithm also has shortcomings.In order to solve the problem of the algorithm,the mining process of the Apriori association rule algorithm is analyzed,and then the improvement of the algorithm is proposed by using the prior pruning strategy.The improved algorithm is evaluated by using Matlab,and the U test is used to measure the standard.The algorithm measures the reliability of the evaluation results.Through the evaluation,it is found that the improved algorithm improves the efficiency of generating association rules and the accuracy of the rule set.
作者
邓毅
邓晓慧
DENG Yi;DENG Xiaohui(College of Artificial Intelligence,Chongqing Creation Vocational College,Yongchuan Chongqing 402160,China;Business College,Chongqing City Vocational College,Yongchuan Chongqing 402160,China)
出处
《四川职业技术学院学报》
2021年第5期148-153,共6页
Journal of Sichuan Vocational and Technical College
关键词
剪枝
关联规则
U检验
信度
pruning
association rules
U-test
reliability
