Information Technology Reference
In-Depth Information
27.2
Fuzzy Cognitive Maps
Fuzzy Cognitive Maps (FCMs) belong to Soft Computing approaches that are in-
troduced to create advanced modeling systems aiming to resemble human-like rea-
soning. FCMs have successfully been applied to a wide range of problems in many
engineering application domains, mainly to model complex systems and develop
advanced diagnosis and/or decision support systems. Human knowledge and expe-
rience is reflected in the creation procedure and the infrastructure of FCMs, mak-
ing them suitable for modeling the decision-making and reasoning approach in a
human-like manner. Especially in the medical field, the decision-making procedure
is often crucial and must be achieved in a timely manner.
Fuzzy Cognitive Maps with their modifications integrate aspects of fuzzy logic,
neural networks, semantic networks, expert systems and they are usually supple-
mented with other soft and hard computing methodologies. An FCM is illustrated
as a causal graphical representation consisting of interrelated concepts [19]. FCMs
are fuzzy signed directed graphs permitting feedback, where the weighted edge
w
ij
from causal concept
C
i
to affected concept
C
j
describes the degree with which the
first concept influences the latter, as is illustrated in Fig. 27.1. FCMs are char-
acterized as fuzzy feedback models of causality, where the weighted interconnec-
tions among concepts of the FCMs present causality among concepts and creating
an interconnected network of interrelated entities, like an abstract mental model.
Feedback interconnections are permitted along with if-then inferencing; that per-
mits FCMs to model complex nonlinear dynamic systems. FCMs have the ability to
include hidden nonlinear dynamics.
Fig. 27.1
The Fuzzy Cognitive Map model
