ABSTRACT
The underlying hypothesis is that there is a sufficient amount of information in the initial database to carry out only one possible data generalization. The main issue of generalization is to find the characteristics allowing us to abstract the data. This chapter summarizes the different information/knowledge categories that can be considered and some possible directions for their representation and use in a generalization process. Most recently, however, this approach also appears to be worthy to develop further for model generalization in the scale range between 1:1 000 and 1:25 000 as it is related to such important fields of application as cadastre, town planning and regional planning, etc. The methods developed should then be adapted for cartographic generalization starting with standard map series. At this stage, cartographic modelling such as displacement is to be performed interactively.
