Information Technology Reference
In-Depth Information
2.2 Cognitive Resonance Model
An information system which analyses data based on the individual features
characteristic for the specific type of data holds in its base the indispensable
knowledge, which, during the analysis processes conducted, becomes the basis for
generating the system's expectations. These expectations are generated automati-
cally using the expert knowledge bases collected in the system. At the same time,
the information system executes an analysis aimed at identifying and indicating
features characteristic for the analysed data sets. As a result of combining signifi-
cant characteristic features of the analysed data with expectations generated using
the knowledge held in the system concerning the analysed semantic content,
cognitive resonance occurs (Figure 2.10.) [87], [97], [114].
Fig. 2.10. The cognitive resonance phenomenon
Cognitive resonance becomes the cornerstone of the process of understanding
(as such) the analysed data sets. This understanding occurs as follows. The stream
of expectations generated by certain hypothetic semantic content (meaning) of
data and the stream of features characteristic for the specific data set are com-
pared, as a result of which certain pairs of expectations and features identified in
the analysed data can gain in importance (become significant) or, conversely, lose
importance (become insignificant). This comparison, by causing cognitive reso-
nance, leads to confirming one of the possible hypotheses (in the case of data
whose content can be understood), or conversely, shows that the inconsistency of
features and expectations cannot be eliminated. The first case means that the
analysis process conducted was successful, the second means that the attempt to
automatically understand data failed [92], [97].
Information systems using cognitive resonance for data analysis are based on
methods defining structural reasoning techniques for matching patterns [89], [92].
The structure of the image being analysed is compared to the structure of data
constituting the pattern during the analysis process. The comparison is made pos-
sible by using strings of derivation rules which enable the pattern to be generated
unambiguously. These rules, referred to as productions, are defined in a specially
introduced grammar, which in turn defines a certain formal language. Data thus
recognised is assigned to the class to which the pattern representing it belongs.
The cognitive analysis utilised in cognitive information systems very frequently
Search WWH ::




Custom Search