ABSTRACT
The term batch generalization is used for the process in which a computer gets an input dataset and returns an output dataset using algorithms, rales, or constraints without the intervention of humans. On-the-fly map generalization does not produce a second dataset, as this would introduce redundant data. In order to avoid gaps when not selecting small area features, a special structure is proposed: the GAP-tree. The linearized GAP-tree is reflected by the sort by clause of the query. The use of the GAP-tree will ensure that the map does not contain gaps when omitting the less important area features. The gap introduced by leaving out one area feature must be filled again. The best results will be obtained by filling the gap with neighbouring features. The polygonal area partitioning is usually stored in a topological data structure with nodes, edges, and faces. The process for producing the area partitioning hierarchy assumes such a topological data structure.
