6. Data acquisition phase

6.1 Acquisition

The data may be located on different sites from the one where the EDC is carried out. They may be stored in a variety of architectures: in relational databases, in data warehouses, on the Web, or in specialized databases (images, digital libraries, genomic databases). They may be structured or unstructured, and may include tabular or textual data, images, sounds or video sequences. In EDC, the analyst, whether knowledge engineer or statistician, must have a relatively well-defined problem. He doesn't embark on EDC without having some idea of the objectives of his operation and the informational and technological resources at his disposal. For example, he may want to understand why some of his customers have turned to a competitor, or he may want to know more about his...