Information Technology Reference
In-Depth Information
and reasoning processes on the data being interpreted. The main subject of this publi-
cation is to present a selected class of cognitive categorisation systems which support
analyses of data recorded in the form of images and vectors. Cognitive categorisation
systems operate by executing a particular type of thought, cognitive and reasoning
processes which take place in the human mind and which ultimately lead to making an
in-depth description of the analysis and reasoning process.
The most important element in this analysis and reasoning process is that it occurs
both in the human cognitive and thinking process and in the system's information and
reasoning process that conducts the in-depth interpretation and analysis of data. It
should be added that this process is based on cognitive resonance which occurs during
the examination process [16], [28], and which forms the starting point for the process
of data understanding consisting in extracting the semantic information and the mean-
ing contained in the analysed type of data that makes reasoning possible [27].
Cognitive analysis and cognitive resonance is an attempt to compare and distin-
guish certain similarities and differences between the set of analysed data and the set
represented by a knowledge base.
The set of data containing the analysed data group is subjected to a process of
broadly-understood analysis, which means analysing the form, content, meaning,
shape and the like. This analysis makes it possible to extract certain significant fea-
tures of the analysed data. At the same time, the set of knowledge collected (pos-
sessed) by the system is used to generate certain expectations as to the substantive
content of the analysed data. These expectations are then compared to the features of
the analysed data extracted during the analysis process.
When the features and expectations are compared, cognitive resonance occurs; its
essence is that it indicates the similarities that appear between the analysed dataset
and the generated set of expectations about the possible results of the knowledge
acquired by the system. These similarities are revealed during the comparative analy-
sis conducted by the system, in the course of which the analysed data is subjected to
the phenomenon of understanding.
The reasoning process which forms the result of the understanding process is an
indispensable factor for the correct data analysis. If it did not occur, it would become
impossible to forecast and reason as to the future of the phenomenon being studied.
So conducting the analysis without the reasoning process could actually lead to im-
poverishing the entire analysis process, as it would be limited only to understanding
the reasons why the analysed phenomenon occurred, but without a chance of deter-
mining its further development.
2 Intelligent Cognitive Information Systems
So-called intelligent information systems were first distinguished within the entire set
of information systems in late 1990s. The word 'intelligent' is understood in this case
mainly as referring to the ability of these systems to signal their readiness to answer
the formulated question in a not fully determined environment, in conditions of uncer-
tainty, when the right behaviour cannot be determined algorithmically. At the same
time these systems make success the most likely. In this sense, intelligence is devel-
oped on many (intelligence) levels, determined by the computing and storage capacity
Search WWH ::
Custom Search