基于函数依赖与条件约束的数据修复方法

随着经济与信息技术的发展,在许多应用中均产生大量数据.然而,受硬件设备、人工操作、多源数据集成等诸多因素的影响,在这些应用之中往往存在较为严重的数据质量问题,特别是不一致性问题,从而无法有效管理数据.因此,首要的任务就是开发新型数据清洗技术来提升数据质量,以支持后续的数据管理与分析.现有工作主要研究基于函数依赖的数据修复技术,即以函数依赖来描述数据一致性约束,通过变更数据库中部分元组的属性值（而非增加/删除元组）来使得整个数据库遵循函数依赖集合.从一致性约束描述的角度来看,函数依赖并非是唯一的表达方式,还存在其他表达方式,例如硬约束、数量约束、等值约束、非等值约束等.然而,随着一致性约束种类的增加,其处理难度也远比仅有函数依赖的场景要困难.本文考虑以函数依赖与其他一致性约束共同表述数据库的一致性约束,并在此基础上设计数据修复算法,从而提升数据质量.实验结果表明,本文所提方法的执行效率较高.

Functional Dependency and Conditional Constraint Based Data Repair

Along with the development of economy and information technology, a large amount of data is produced in many applications. However, due to the influence of some factors, such as hardware equipments, manual operations, and multi-source data integration, serious data quality issues arise, including data inconsistency, which makes it more challenging to manage data effectively. Hence, it is urgent to develop new data cleaning technology to improve data quality to support further data management and analysis. Currently, the existing work mainly focuses on the situation where functional dependencies are used to describe data inconsistency. Once some violations are detected, some tuples must be changed to suit for the functional dependency set via update, neither insert nor delete. Besides functional dependency, there also exist other kinds of constraints, such as the hard constraint, quantity constraint, equivalent constraint, non-equivalent constraint, etc. However, it becomes more difficult when more kinds of inconsistent conditions are involved. In this paper, we consider the general scenario where functional dependencies and other constraints co-exist. We design the corresponding data repair algorithm to improve the data quality effectively. Experimental results show that the proposed method behaves effectively and efficiently.