SOLUTIONS FOR DATA QUALITY IN DATA WAREHOUSES APPLICATIONS
Supp/2008 , p440..447
Author(s):
Mirela OANCEA
Valentin DUMITRU
Vasile GORGAN
Keywords:
Data mining, data quality, data warehouse, business intelligence, ETL
Abstract:
According to a study performed by Gartner Research, over 50% of the BI deployments have problems related to their acceptance or rejection because of the poor quality of the data in the phase of data warehouse's feeding. Lots of the current data warehouse deployment projects are centered on the final analysis stage and lose sight or treat superficially the process of regularization and cleanup of the data in the data warehouse. Using the data affected by the "lack of quality" produced by the duplicate, incomplete or wrong records lead to a waste of time and incorrect results. The purpose of this article is to examine the techniques that improve the quality of data, to analyze the connection between their quality and the success of the projects of deployment of a data warehouse. For this, we will examine, among others, the possibility to use data mining techniques to assure a coherent data warehouse with irredundant and unpolluted data with errors.
Download:
Back
|
|