图依赖研究与应用综述

图依赖是用于解决图数据的数据一致性问题的数据质量规则。基于图依赖提升数据一致性的过程通常分为图依赖定义与形式化、图依赖自动挖掘、基于图依赖的数据一致性提升三步。介绍了针对数据一致性的图依赖理论，并根据拓展类型将图依赖分为基于结构约束拓展、基于语义约束拓展和基于外部约束拓展的图依赖；综述并对比了从图数据中自动挖掘图依赖及其拓展的算法；分析了应用图依赖提高数据一致性的研究现状；总结了当前研究中仍存在的问题，并依据问题展望了图依赖在数据质量领域的应用前景。

Survey on research and application of graph dependencies

Graph dependency is a data quality rule based on functional dependency, which is proposed to solve data consistency problem in graphs. The process improving data consistency based on graph dependencies was partitioned into graph dependencies formalization step, graph dependencies discovery step and graph dependencies based data inconsistent promotion step. This paper introduced graph dependency theory for data consistency, divided graph dependencies into three categories according to the type of expansion, such as semantic constraint expansion based graph dependencies, structural constraint expansion based graph dependencies, external constraint expansion based graph dependencies. Then this paper summarized and compared existing automatic discovery algorithms of graph dependencies and their extensions. It analyzed the research status for improving data consistency based on graph dependencies. At last, it summarized the existing problems of present researches, and gave application prospect of graph dependencies in data quality area based on the previous problems.