Sommaire de la publication

Sommaire de la publication

Gervais, M., Y. BĂ©dard, M. Levesque, E. Bernier, R. Devillers, 2009, Data Quality Issues and Geographic Knowledge Discovery, Dans: Geographic Data Mining and Knowledge Discovery, Chap. 5, pp. 99-115

Abstract

Geographical data warehouses contain data coming from multiple sources potentially
collected at different times and using different techniques. One of the most important
concerns about geographical data warehouses is the quality or reliability of the data
used for knowledge discovery, decision making, and, finally, action. In fact, this is
the ultimate objective aimed by using this type of database. On the other hand, with
increasing maturity and the proliferation of data warehouses and related applications
(e.g., OLAP, data mining, and dashboards), a recent survey indicated that for the
second year in a row, data quality has become the first concern for companies using
these technologies (Knightsbridge 2006). Similarly, a recent survey of Canadian
decision-makers using spatial data has identified data quality as the third most important
obstacle in increasing the use of spatial data (Environics Research Group 2006).
Thus, while data quality has become the number one concern for users of non-spatial
data warehouses, it is also recognized as an emerging issue for spatial data (Sonnen
2007, Sanderson 2007) and the quality of spatial datacubes is being investigated seriously
within university laboratories. In this context, the concept of data quality is
making its way into the realm of geographic knowledge discovery, leading us to think
in terms of risks for the users, for the developers, and for the suppliers of data, especially
in terms of prevention mechanisms and possible legal consequences.