Aggregation geoconcepts for generalization maps, appropriate logical consistency and semantic accuracy

In recent years, the number of fixed and mobile devices using automated methods of generalization map dramatically increased. This raises the question of the quality of digital imaging models map. Generalization should not only ensure certain standards for the derived scale of the map, but also maintain a certain level of semantics. To conduct an objective comparison of the different results of generalization, it is necessary to introduce a measure of the accuracy of semantic aggregation areas and geoconcepts. An approach that shows how this could be used to compare the results of generalization is presented. Also, the methods for determining the relationship between $n$ geoconcepts to be distributed over $N$ units so that in each unit, there were geoconcepts most similar to each other on the set of $m$ selected descriptive features. The function $F$ is introduced — a similarity geoconcepts function of jointly $k$ selected features, normalized to the maximum range of the characteristic values on the set of $n$ geoconcepts.