一种用于异常数据流挖掘的改进Apriori算法研究  

Research on an improved Apriori algorithm for mining abnormal data streams

作　　者：金兰 陈荆亮[2] JIN Lan;CHEN Jing-liang(College Information Science and Engineering,Wuchang Shouyi University,Wuhan Hubei 430064,China;School of Computer Science,Hubei University of Technology,Wuhan Hubei 430068,China)

机构地区：[1]武昌首义学院信息科学与工程学院,湖北武汉430064 [2]湖北工业大学计算机学院,湖北武汉430068

出　　处：《计算机仿真》2025年第1期480-484,共5页Computer Simulation

基　　金：2021年湖北高校一流本科课程(2021SA01);湖北高校省级教学研究项目(2022286);湖北省高等学校科技创新计划重点项目(D20201402)。

摘　　要：Apriori算法需要多次扫描数据库,算法运行时间长,不利于实际应用,导致现阶段的异常流数据挖掘存在准确度低、时间长等问题,为此提出一种用于异常数据流挖掘的改进Apriori算法。通过压缩矩阵改进Apriori算法,并通过改进Apriori算法获取数据流的数据特征量;通过局部离群因子方法剔除数据流中的远距离聚类离群点,以提高数据流聚类效果;通过改进K-means算法的最大最小距离方法完成异常数据流挖掘。实验结果表明,所提方法的异常数据流挖掘方法准确度更高、挖掘效率更快、效果更好,更适合于实际应用。At present,the Apriori algorithm requires multiple scans of the database,which is not conducive to practical application.In order to address the low accuracy and long running time of the Apriori algorithm in mining anomaly data streams,this paper proposed an improved Apriori algorithm.Firstly,we improved the Apriori algorithm by compressing the matrix.Then,we used the improved algorithm to obtain the feature quantity of the data stream.Moreover,we used the local outlier factor to eliminate distant clustering outliers in the data stream,thus improving the clustering effect of the data stream.Finally,we completed the mining for abnormal data flow by the maximum and minimum distance method of the improved K-means algorithm.Experimental results show that the proposed method has higher accuracy,faster mining efficiency,and better performance in mining abnormal data streams,so it is more suitable for practical application.

关 键 词：异常数据流挖掘 局部离群因子 压缩矩阵 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]