engleski

Data Quality Improvement through Horizontal Fragmentation in Data Warehouses

In this paper we propose the procedure forhorizontal fragmentation of data warehouse tables andshow how it affects data warehouse’s data quality, namely, completeness and timeliness data qualitydimensions. Horizontal fragmentation is suitable whendata from multiple, more or less independent, organizational units are stored in the same fact table. Insuch cases, proposed procedure enables horizontal dataseparation based on the configured attributes thusisolating one institution’s data quality problems fromanother. The proposed procedure relies on metadata andis generic and applicable to any data warehouse.

Completeness; Data Warehouse; ETL; Horizontal Fragmentation; Timeliness

nije evidentirano