ABSTRACT
To model a continuous geographic phenomenon, first a set of data points needs to be collected from the geographical space. There are two stages in this process: sampling and measurement. Sampling refers to the distribution of measurement places, in order to have a representative sample data set of the phenomena. In this context, the uncertainty arises not only from the inaccuracy of measurements but also from the lack of representativity of sample points and from the choice of the model to be used. The uncertainty will be included in membership functions of fuzzy numbers, which will be used to express attribute values (see Figure 5.1). Therefore, the sample set will by given by
f (xi , z˜i ), i = 1, 2, . . . , N , (5.1) where f (xi ) ∈ Rn is the position of sample z˜i ∈ F(R).
