Module 1

Introduction to Intelligent Systems

and Soft Computing
Some Useful Abbreviations will be used throughout this course:
AI : Artificial Intelligence,
SC : Soft Computing,
HC : Hard Computing ,
FL : Fuzzy Logic,
FLC : Fuzzy Logic Control
FIS : Fuzzy Inference System
F : Membership Function
PB : Positive Big,
PS : Positive Small,
NB : Negative Big,
NS : Negative Small,
NN : Neural Networks,
ANN : Artificial Neural Networks,
SL : Supervised Learning,
USL : Un-Supervised Learning,
LMS : Least Mean Square,
BP : Back-Propagation,
ANFIS : Adaptive Neuro-Fuzzy Inference System,
EC : Evolutionary Computation,
GA : Genetic Algorithms,
RBN : Radial Basis Network , … etc
Some Useful Terms will be used throughout this course:
Intelligence , Emulate , Mimic , Reasoning , Rational , Thinking ,
Computations , Decision making , Fuzzy , Linguistic , Evaluations,
Imprecision, Uncertainty, Precisely, partial truth, Approximation
, Crisp , Stochastic , Ambiguous , Vague , Evolutionary ,
Probabilistic , Inspired , Tolerance , Continuum , Mapping ,
Generalization , Fuzzification , Defuzzification , Inference ,
Randomized , Neuron , Pattern-Recognition , Aggregation ,
Reinforced , interconnected , Weights , etc ...
What is Artificial Intelligence (AI ) ?
 Making computers and machines to emulate human
thinking, like decision making , learning abilities and
reasoning ... ?
 The art of creating machines to perform functions that
require intelligence as it is performed by people ?

 The study of computations that make it possible to

realize, think, and act rationally?
What is Artificial Intelligence (AI ) ?

Systems that Systems that

THOUGHTS think think
like humans rationally

Systems that Systems that

BEHAVIOUR act act
like humans rationally

HUMAN RATIONAL
Problem Solving (Computation)
 Problem Solving Techniques:
The different types of computing methods in solving real-world problems
can make a big difference in terms of : Time, Cost and Efficiency.
Generally, Computing can be classified by two types, namely:
Hard-Computing and Soft-Computing.
 Some Differences between Hard and Soft Computings

Hard Computing Soft computing

Requires a precisely stated analytical Tolerant to imprecision, uncertainty,
models partial truth, and approximation

Based on binary logic, crisp systems, Fuzzy logic, Neural networks,

numerical analysis, crisp software Genetic Algorithms, Probability
Reasoning.
Programs are to be written in advance Evolve their own programs

Two values logic (0 , 1) Multi valued logic ( between 0 to 1)

Exact input data Uncertain and noisy data
Strictly sequential Parallel computations
Produce precise answers Produce approximate answers
Deterministic Stochastic
 What is Soft-Computing (SC ) ?
Soft computing is the use of approximate calculations to provide
imprecise but usable solutions to complex computational problems.
The SC enables solutions for problems that may be either
unsolvable or very time-consuming to be solved with traditional
computing techniques .
With the human mind as a role model, soft computing is tolerant of
partial truths, uncertainty, imprecision and approximation, unlike
traditional computing models. The tolerance of soft computing
allows researchers to approach some problems that cannot be
solved using normal computing techniques.
History of Soft-Computing :
 Goals of Soft-Computing :
❑ The main goal of soft computing is to develop
intelligent machines to provide solutions to real-
world problems, which are not modeled, or too
difficult to be explained and solved mathematically.

❑ It also aims to utilize the tolerance for

Approximation, Uncertainty, Imprecision, and
Partial Truth in order to achieve close similarity with
human-like decision-making and thinking.
 Constituents Soft-Computing :
The main constituents of SC are:
❖ Fuzzy Logic (FL)
❖ Neural Network (NN)
❖ Evolutionary Computation (EC) - based on the origin
of the species
• Genetic Algorithm
• Ant Colony Optimizations
❖ Machine Learning (ML)
Soft-Computing Development History :
Fuzzy Logic (FL) :
 Fuzzy Logic (FL) :
The term fuzzy refers to things which are not clear or are vague.
In the real world many times we encounter a situation when we can’t
determine whether the state is true or false, therefore, fuzzy logic
provides a very valuable flexibility for reasoning. In this way, we can
consider the inaccuracies and uncertainties of any situation.
In Boolean system truth value, 1.0 represents absolute truth value and 0.0
represents absolute false value. But in the fuzzy system, there is no logic
for absolute truth and absolute false values. But in fuzzy logic, there is
intermediate value too represent which is partially true and partially false.
Fuzzy Logic (FL) :
 What is Fuzzy Logic ?

Tap Water Temperature Example:

Fuzzy Logic (FL) :
 Why Fuzzy Logic is Useful?
Automatic Car Braking System Example:
1) Using Traditional Logic:

Case 1:

Case 2:
 Why Fuzzy Logic is Useful?
2) Using Fuzzy Logic:

Case 1:

Case 2:
 Fuzzy Logic Architectures :

Crisp Crisp
value value

Fuzzy Fuzzy
value value

FUZZIFICATION: It is used to convert inputs i.e. crisp numbers into fuzzy

sets. Crisp inputs are basically the exact inputs measured by sensors and
passed into the control system for processing, such as temperature,
pressure, etc.

RULE BASE: It contains the set of rules and the IF-THEN conditions
provided by the experts to govern the decision making system, on the basis
of linguistic information.
 Fuzzy Logic Architectures :

Crisp Crisp
value value

Fuzzy Fuzzy
value value

INFERENCE ENGINE: It determines the matching degree of the current

fuzzy input with respect to each rule and decides which rules are to be fired
according to the input field. Next, the fired rules are combined to form the
control actions.
DEFUZZIFICATION: It is used to convert the fuzzy sets obtained by
inference engine into a crisp value. There are several defuzzification
methods available and the best suited one is used with a specific expert
system to reduce the error.
Fuzzy Sets :
The classical example in fuzzy sets is Tall men. As shown in
next table : (The elements of the fuzzy set “Tall men” for all
men, and their degrees of membership depend on their
height).
✓ For Crisp set the question asks : Is the man is tall ?
❑ The answer is: (Tall men) are above 180 cm, and (not Tall
men) are below 180 cm.

✓ Fuzzy set asks the question: How tall is the man?

❑ The answer is : The linguistic variable Tall is having partial
membership in the fuzzy set, Tall.
Crisp Set vs. Fuzzy Set :
Crisp Set vs. Fuzzy Set :

180
 The same Fuzzy-Set can be interpreted differently
depending on the given situation or conditions

0.85

0.15
Artificial Neural Networks (ANN) :
(An extremely simplified model of the brain)
 Human brain consists of densely interconnected simple
processing elements known as neurons
( More than 10 billion Neurons , and 60 trillion Synapses in human cortex )
 Artificial Neural Networks (ANN) :

Biological
Neuron

Artificial
Neuron Model
• ANNs are based on a simplified model of the neurons and their operation
• ANNs usually learn from experience – repeated presentation of example
problems with their corresponding solutions
• After learning the ANN is able to solve problems, even with newish inputs
• The learning phase may or may not involve human intervention (supervised
vs unsupervised learning)
Learning Methods of ANNs :
Architectures of ANNs :
 Some Applications of Soft-Computing :
➢ Handwriting Recognition, Pattern Recognition, and Speech
Recognition.
➢ Automotive systems and Manufacturing processes.
➢ Image processing and data mining and compression
➢ Consumer appliances such as: Washing Machines, Air-Cons,
Refrigerators.
➢ Food preparation appliances like Rice cookers and Microwaves.
➢ Robotics and Automation systems.
➢ Application of soft computing to power systems .
➢ Decision-support systems.
➢ And many more different applications
Example of a Fuzzy Logic System for an air conditioning system to adjusts the
temperature of air conditioner by comparing the room temperature and the
target temperature value.

Rule No. Condition Action

1 IF temperature is (Cold OR Very_Cold) AND target is Warm THEN Heat ON
2 IF temperature is (Hot OR Very_Hot) AND target is Warm THEN Cool ON
3 IF (temperature is Warm) AND (target is Warm) THEN No_Change
Example: Memorized bit patterns and pattern recognition
examples using ANN system.
Example: A Neural character recognizer and a knowledge base
cooperate in responding to 3 handwritten characters
that form a word “dog”.