基于关联规则和聚类分析的数据缺失补偿方法  

A Data Missing Compensation Method Based on Association Rules and Cluster Analysis

作　　者：李庐 LI Lu(Dean's Office,Anhui University of Finance and Economics,Bengbu 233030,Anhui,China)

机构地区：[1]安徽财经大学教务处,安徽蚌埠233030

出　　处：《山西师范大学学报(自然科学版)》2025年第1期71-75,共5页Journal of Shanxi Normal University(Natural Science Edition)

摘　　要：在常规数据缺失的补偿中,一般会根据数据的特征进行提取,查找数据的缺失部分,并对其进行补偿,但这往往会导致补偿数据与原有数据的差别较大,因此提出基于关联规则和聚类分析的数据缺失补偿方法.在数据补偿方法的研究中,首先通过对缺失数据的处理,来分析数据信息与类别,然后对数据整体进行单元格的划分,分析数据的重要特征,然后根据关联规则的方式,运用FP-growth算法构建数据间的相似度,通过相似度收集数据信息,最后结合聚类分析,选择最优的补偿数据.在实验中,对所设计方法的补偿效果进行模拟,实验结果体现出在三组数据中都具有较好的补偿效果,可以运用到实际的数据补偿中.In the compensation of conventional data loss,it is generally based on the characteristics of the data to extract,search for missing parts of the data,and compensate for them.However,this often leads to significant differences between the compensated data and the original data.Therefore,a data loss compensation method based on association rules and cluster analysis is proposed.In the research of data compensation methods,the missing data is first processed to analyze data information and categories.Then,the overall data is divided into cells to analyze important features of the data.Then,based on association rules,the FP growth algorithm is used to construct similarity between the data.Data information is collected through similarity,and finally,cluster analysis is used to select the optimal compensation data.In the experiment,the compensation effect of the designed method was simulated,and the experimental results showed that it had good compensation effect in all three sets of data,which can be applied to actual data compensation.

关 键 词：关联规则 聚类分析 数据缺失 数据补偿 

分 类 号：G642[文化科学—高等教育学]