in Proceeding
| Author | Fabian Panse |
| Title | Completeness of Attribute Values Representing Partial Information |
| Published in | Proceedings of the 7th International Workshop on Quality in Databases at VLDB 2009 |
| Date | August 2009 |
| Note | Workshop-Beiträge nur als PDFs auf Konferenz eigenem USB-Stick |
| Abstract | The data quality dimension completeness quantifies the extent to which information on a real-world application is represented in a database. The intensional completeness measure density can be considered at different levels of granularity. With respect to attribute values, current density metrics are undefined for values which represent partial information. Thus, for data models using such concepts these metrics are not suitable and a representative measuring of completeness is not possible. In order to correct this flaw, we redefine the metrics of density at the attribute value level w.r.t. partial information which can be represented by classical subsets of the corresponding attribute domains. Since these redefinitions enable a more accurate and more exact quality measuring of data values, quality based activities and decisions can be more effective in the future. |
| Document |  |
| Other formats | Din 1501 |