基于偏序集的数据清洗规则链自动生成方法

针对数据清洗中规则间逻辑冲突频发和出错率高的问题,提出一种基于偏序集的规则链自动生成方法。通过分层组合的数据清洗框架自顶向下对规则进行分类处理,采用偏序集和哈斯图自动生成每个层级的逻辑正确和一致的规则链,并设计出对应的生成算法和自动清洗算法。以扶贫领域数据为例进行实验,结果表明该方法使数据清洗效率有一定提升,清洗结果出错率明显降低,检验了方法的科学性和有效性。

Automatic generation of data cleaning rule chain based on poset

In order to solve the problem of frequent logical conflicts and high error rate between rules in data cleaning,this paper proposed an automatic rule chain generation method based on partial order set.It classified and processed the rules from top to bottom by the hierarchical data cleaning framework.It automatically generated the rules chain of each level by using partial order set and Hasse diagram,and designed the corresponding generation algorithm and automatic cleaning algorithm.Taking the data of poverty alleviation as an example,the results show that the proposed method can improve the efficiency of data cleaning,reduce the error rate of cleaning results,and test the scientificity and effectiveness of the method.