A Framework for Data Cleaning in Data Warehouses

It is a persistent challenge to achieve a high quality of data in data warehouses. Data cleaning is a crucial task for such a challenge. To deal with this challenge, a set of methods and tools has been developed. However, there are still at least two questions needed to be answered: How to improve the efficiency while performing data cleaning? How to improve the degree of automation when performing data cleaning? This paper challenges these two questions by presenting a novel framework, which provides an approach to managing data cleaning in data warehouses by focusing on the use of data quality dimensions, and decoupling a cleaning process into several sub-processes. Initial test run of the processes in the framework demonstrates that the approach presented is efficient and scalable for data cleaning in data warehouses.