The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B4. Beijing 2008
211
parameters, to define weighting measures and to decrease
uncertainties resulting from the importance determination.
3. CONTEXT AND FUZZY APPROACH
3.1 Context Understanding
It is possible consider context as follows.
• context as the reflection of object or phenomena using
different interpretation through the system of cognition:
perception, conception and interpretation,
• context as the reflection of selected facts is concerned with
validity of statements and the system of argumentation,
• context as the reflection concerning validity of statement
using knowledge generating system - knowledge based
context
Using context it is possible to derive new quality of information
that can be used to support decision. The integral part of control
GIS [9] is the modelling where the information layers from real,
artificial and virtual world are composed together to select
optimal solution or verify given hypothesis or assumptions.
The contextual design of spatial data and further development
of geo-information technologies, image processing techniques
and the possibilities of object history modelling together with
the geographical networks environment will provide quite new
and considerably wider possibilities of using GIS.
Good message is that the increase of interactions and
correlations implies decrease of differences between objects.
From the point of uncertainty and contextual modelling the
wide context (temporal, spatial, local, objective, attribute
oriented, ...) influents in the sense of decreasing overall
uncertainty of the results - the similar effect like the use of
sequence of images in case of noise suppression - image
reconstruction.
3.2 Fuzzy Approach
GIS architecture is open to incorporate new requirements of
knowledge-based analysis and modelling, namely in connection
with web designed spatial databases and temporal oriented
approaches. The example of context implementation is
composed classifier where the fuzzy elements are implemented.
Composed classifier is composition of component classifiers,
which predictions are connecting by combining classifier,
unlike others simple classifiers. There are several architectures
for possible combination of classifiers.
The analysis of geographical data tends to tasks that are solved
using defined sequences of operations and algorithms where the
Figure 1. Outcome of the Fuzzy C- means
Figure 2. Outcome of the Gustafson-Kessel Algorithm
set of conditions is provided per parts. The amount of
cumulative functions allows to capture the uncertainty and
imprecision. The solution of zoning and tolerance extents where
the fuzzy approaches are used the slivers are generated and
together with overlay function arise the complexity of results.
Figure 3. Overlay function and its consequences
4. CONCLUSIONS
In this paper, the problem of uncertainty of multisource data and
muticriterial analysis is addressed and the use of wide spatial
and temporal context and expert knowledge integration to
decrease the uncertainty is discussed. The running development
of information technologies, image processing techniques and
knowledge-based databases, together with the geographical
networks environment, will provide quite new and considerably
wider possibilities of using GIS. Our decisions are becoming
increasingly dependent on understanding of complex relations
and phenomena in the world around and GIS technology is able
to incorporate new requirements.