Information Technology Reference
In-Depth Information
Optimization of fl uid bed granulation and tablet
compression process. ANNs were used to
analyze the infl uence of the atomization pressure
(of the fl uid bed granulation process) and
tableting compression force on the tablet
hardness and drug release profi le.
MLP
Djuris et al.,
2012
Abbreviations: MLP - Multilayered Perceptron, GRNN - Generalized Regression Neural
Network, EANN - Ensemble ANN, RBFNN - Radial Basis Function Neural Network, EDNN
- Elman Dynamic Neural Network, GMDNN - Gamma memory Dynamic Neural Network,
OLDNN - One Layer Dynamic Neural Network, HNN - Hierarchical Neural Network.
5.2 Fuzzy logic
5.2.1 Introduction
The origin of fuzzy logic is a fuzzy set theory (Zadeh, 1965). It has long
been recognized that many real-life problems cannot be presented in a
binary manner, using discrete terms. Fuzzy logic is similar to human
thinking and reasoning. It is based upon certain features, such as
temperature, that do not only have two values (low or high, presence or
absence); in fact, it has many different levels to describe it more precisely.
Fuzzy logic is the science of reasoning, thinking, and inference that
recognizes and uses the real world phenomenon that everything is a
matter of degree (Sproule et al., 2002). If binary logic is considered as a
black-and-white approach (crisp, sharp solutions are available), than
fuzzy logic comes from shades of gray.
5.2.2 Theory
Fuzzy logic is a multi-valued logic that allows a range of truth-values
between 0 (completely false) and 1 (completely true) (Klenner et al., 2010).
Therefore, in binary logic, values are limited to two states: 0 (false) and 1
(true). Fuzzy logic accounts for both existence and quantifi es the presence
of a feature (Kosko and Isaka, 1993). By using the fuzzy approach, the
transition between terms can be gradual, and the binary all-or-none
options become the extreme ends of a continuum (Sproule et al., 2002).
It has been accentuated that neuro-fuzzy systems are predictive data
mining tools that provide sets of linguistic rules, allowing for easier
understanding of analyzed processes (Mansa et al., 2008). Fuzzy logic
Search WWH ::
Custom Search