Data Quality Rules: General Attribute Dependencies
Data represent attributes of real world objects (e.g., people, sales) whose characteristics are interrelated and whose behavior is complex and restricted by numerous logical constraints. These constraints can always be translated into data relationships and used to test data quality.
In the last three articles of this series, we discussed several specific views of the data for complex real world objects. We first investigated the time dimension and the data quality rules arising from the dynamic relationships in the data. We then looked at the lifecycles of state-dependent objects and the rules governing the state-transition data. These investigations, however, still leave unexplored numerous miscellaneous attribute dependencies in the data describing real world objects.
Finding such general attribute dependency rules is more difficult; yet this is key to the success of any data quality assessment project. In this issue, we discuss various strategies and techniques that help us succeed in this challenging task.