分布式数据库中关系数据正负关联规则挖掘  

Mining Positive and Negative Association Rules of Relational Data in Distributed Database

作　　者：吴爱华[1] 陈出新 WU Ai-hua;CHEN Chu-xin(South China Institute of Software Engineering,Guangzhou University,Guangdong Guangzhou 510990,China)

机构地区：[1]广州大学华软软件学院,广东广州510990

出　　处：《计算机仿真》2021年第9期344-347,352,共5页Computer Simulation

摘　　要：针对传统分布式数据库中关系数据正负关联规则挖掘的准确度较低、挖掘效率较低等问题,提出一种新的分布式数据库中关系数据正负关联规则挖掘方法。在关联规则基本概念和性质分析基础上,利用多级支持度从频繁项集中生成正关联规则,结合根据频繁项集和非频繁项集生成负关联规则,通过最小支持度合理设置相关置信度,引入不同权重值于各数据库中,实现分布式数据库中关系数据正负关联规则的挖掘。仿真结果表明,以上算法可有效识别结果规则集中的负关联规则和弱关联规则,确保数据库中关联数据挖掘更加准确;在不同最小支持度或不同事务数条件下,挖掘速度较快,提升了挖掘效率。This paper puts forward a new mining method of positive and negative association rules of relational data in distributed database in order to improve the accuracy and efficiency of mining the positive and negative association rules of relational data in distributed database.According to the concept and nature of association rules, from frequent item set, multi-level support was adopted to generate positive association rules.Frequent and non-frequent item-sets were introduced to generate negative association rules.Based on the minimum support, the correlation confidence was set reasonably.Different weight values were applied to each database.Eventually, mining the positive and negative association rules of relational data in distributed database was realized.The simulation results show that the algorithm is sensitive to the negative association rules and weak association rules in the result rule set, which ensures that the mining of association data in database is more accurate.The algorithm has fast mining speed and high mining efficiency under different minimum support or different transaction numbers.

关 键 词：分布式 数据库 正负关联规则 最小支持度 非频繁项集 置信度 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]