Knowledge Discovery and Knowledge Visualization

Past experiences have shown that there is a strong connection between Knowledge Discovery in Databases and Knowledge Visualization. This connection is twofold. On the one hand, Visualization can serve as a powerful tool for identifying interesting patterns and relationships in data in an intuitive way. On the other hand, large amounts of data can often not be visualized directly, as the resulting graphical representation gets much to complex to be captured by human user. Therefore, intelligent methods are needed to extract and aggregate the essential information contained in the database first. The resulting structures can then be visualized, to give the user an intuitive access to the whole data set. Following this observation, we argue that in many domains only the combination of modern Knowledge Discovery and Knowledge Visualization techniques will lead to successful solutions. As to exemplify and support this thesis, several state-of-theart approaches in the field of Knowledge Discovery in large databases will be presented. These approaches derive from practical applications and therefore offer an interesting insight in "real life" Knowledge Discovery. Additional to this more general overview of the connection of Knowledge Discovery and Visualization, the special relevance of this connection to large databases of documents will be pointed out, as this is a central issue in the development of digital archives and portals.