基于关联规则的条件函数依赖发现及数据修复  

Dependency discovery and data recovery of conditional functions based on association rules

作　　者：张春生[1] 图雅[1] 翁慧[1] 李艳[1] 魏国利[1] 

机构地区：[1]内蒙古民族大学计算机科学与技术学院,内蒙古通辽028043

出　　处：《计算机应用研究》2016年第2期384-387,共4页Application Research of Computers

基　　金：国家自然科学基金资助项目(81460656);内蒙古自然科学基金资助项目(2012MS0913);通辽市与内蒙古民族大学合作项目(SXZD2012021)

摘　　要：大数据时代,数据的来源复杂,数据质量存在严重问题,有些数据不准确、缺失或存在错误。不正确的数据严重影响了数据挖掘的质量,给决策造成重大的影响。关于缺失数据的修复方法很多,其中条件函数依赖就是一个有效的方法,在如何发现条件函数依赖方面已经有了很多研究成果。提出一种应用关联规则构建条件函数依赖的方法,由于关联规则通过数据挖掘得到,具有一定的隐蔽性,不是一般的方法能够发现,所以,由关联规则构建的条件函数依赖具有一定的应用价值,给出了构建方法,并通过实验证明其有效性。In the era of big data, the data sources are complex, the data qualities are problematic, and some data is not accu- rate, missing or involves error. Incorrect data has seriously affected the quality of data mining, resuhing in a significant impact on decision making. There are many repair methods about the missing data. In them conditional function dependency is an ef- fective one and many research findings on how to find conditional function dependencies have been found. This paper presen- ted a method of constructing conditional function dependencies based on association rules. The association rules were gotten by data mining and they had some hidden features and could not be found by usual ways. Thus the conditional function dependen- cy based on association rules has a certain value of application. The paper gave a construction method and experiments prove its effectiveness.

关 键 词：关联规则 条件函数依赖 一致性 数据修复 大数据 缺失数据 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]