The aim of the article is to present our current works on measuring the impact of data quality on the quality of extracted association rules. The first part of the article reviews previous work on data quality. These quality issues are specially relevant for qualifying and improving the quality of the knowledge discovered into database systems ; the second part presents a methodology for controlling data quality in the process of knowledge discovery in databases. Our approach consists in the fusion of data quality indicators in order to add meta-information on discovered knowledge quality and provides several advantages for the qualification and validation of extracted rules.
|Number of pages||7|
|Journal||Revue d'Intelligence Artificielle|
|Publication status||Published - 1 Dec 2003|
- Data quality
- Knowledge quality
ASJC Scopus subject areas
- Artificial Intelligence