You are on page 1of 10

BTCS- AI and Soft Computing

BTCS AI and Soft Computing


Unit I
Soft Computing: Introduction of soft computing, soft computing vs. hard computing, various types of soft
computing techniques, applications of soft computing.
Artificial Intelligence : Introduction, Various types of production systems, characteristics of production
systems, breadth first search, depth first search techniques, other Search Techniques like hill Climbing, Best
first Search, A* algorithm, AO* Algorithms and various types of control strategies. Knowledge
representation issues, Prepositional and predicate logic, monotonic and non monotonic reasoning, forward
Reasoning, backward reasoning, Weak & Strong Slot & filler structures, NLP.
Unit II
Neural Network: Structure and Function of a single neuron: Biological neuron, artificial neuron, definition
of ANN, Taxonomy of neural net, Difference between ANN and human brain, characteristics and
applications of ANN, single layer network, Perceptron training algorithm, Linear separability, Widrow &
Hebb;s learning rule/Delta rule, ADALINE, MADALINE, AI v/s ANN. Introduction of MLP, different
activation functions, Error back propagation algorithm, derivation of BBPA, momentum, limitation,
characteristics and application of EBPA.
Unit III
Counter propagation network, architecture, functioning & characteristics of counter Propagation network,
Hopfield/ Recurrent network, configuration, stability constraints, associative memory, and characteristics,
limitations and applications. Hopfield v/s Boltzman machine. Adaptive Resonance Theory: Architecture,
classifications, Implementation and training. Associative Memory.
Unit IV
Fuzzy Logic: Fuzzy set theory, Fuzzy set versus crisp set, Crisp relation & fuzzy relations, Fuzzy systems:
crisp logic, fuzzy logic, introduction & features of membership functions, Fuzzy rule base system : fuzzy
propositions, formation, decomposition & aggregation of fuzzy rules, fuzzy reasoning, fuzzy inference
systems, fuzzy decision making & Applications of fuzzy logic.
Unit V
Genetic algorithm : Fundamentals, basic concepts, working principle, encoding, fitness function,
reproduction, Genetic modeling: Inheritance operator, cross over, inversion & deletion, mutation operator,
Bitwise operator, Generational Cycle, Convergence of GA, Applications & advances in GA, Differences &
similarities between GA & other traditional methods.
References :
1. S, Rajasekaran & G.A. Vijayalakshmi Pai, Neural Networks, Fuzzy Logic & Genetic
Algorithms, Synthesis & applications, PHI Publication.
2. S.N. Sivanandam & S.N. Deepa, Principles of Soft Computing, Wiley Publications
Rich E and Knight K, Artificial Intelligence, TMH, New Delhi.
3. Bose, Neural Network fundamental with Graph , Algo.& Appl, TMH
4. Kosko: Neural Network & Fuzzy System, PHI Publication
5. Klir & Yuan ,Fuzzy sets & Fuzzy Logic: Theory & Appli.,PHI Pub.
6. Hagen, Neural Network Design, Cengage Learning

LECTURE PLAN

Subject Title : Soft Computing

Session : 2014

Lect
No.

Topics to be covered

* REFERENCE

R2:8
Subject Code : BTCS-802 Unit-I Semester : VIII
Department : Computer Science
Branch : Computer Science and
1.
Introduction of soft computing,
Engineering R2:8, R2:8
soft computing vs. hard
computing
2

various types of soft


computing techniques,
applications of soft computing

R2:3

Introduction of Artificial
Intelligence, Various types of
production systems

R1:1, R1:30

characteristics of production
systems

R1:43, R1:32

breadth first search, depth first


search techniques
other Search Techniques like
hill Climbing, Best first Search
A* algorithm, AO* Algorithms
and various types of control
strategies

R1:32, R1:52

6
7
8

Knowledge representation
issues, Prepositional and
predicate logic

monotonic and non monotonic


reasoning
forward Reasoning backward
reasoning
Weak & Strong Slot & filler
structures, NLP

10
11

R1:57, R1:59
R1:67

R1:79, R1:98
R1:147, R1:134
R1:188,R1:285
R2:8

Unit-II
12
13

Neural Network Structure and


Function of a single neuron
Biological neuron, artificial
neuron, definition of ANN

14

Taxonomy of neural net,


Difference between ANN and
human brain

15

characteristics and
applications of ANN, single
layer network
Perceptron training algorithm,
Linear separability

16
17
18

Widrow & Hebb;s learning


rule/Delta rule
ADALINE, MADALINE AI v/s
ANN

19
20
21

Introduction of MLP, different


activation functions
Error back propagation
algorithm, derivation of BBPA
momentum, limitation,

R2:11
R2:12
R2:24

R2:14,R2:3
R2:49
R2:29,R231
R2:57, R2:60
R2:64
R2:64

REFERENCES
S. No.

TITLE

AUTHOR

Rich E and Knight

R1

Artificial Intelligence

R2

Principles of Soft Computing

R3

Neural Networks, Fuzzy Logic & Genetic


Algorithms, Synthesis & applications

S.N. Sivanandam
& S.N. Deepa

TMH, New Delhi.

Wiley Publications

S, Rajasekaran &
G.A.
Vijayalakshmi Pai

UNIT WISE BLOW UP


Unit-1

PUBLISHER / EDITION

PHI Publication

The term soft computing was proposed by the inventor of fuzzy logic, Lotfi A. Zadeh. He describes it as
follows: Soft computing is a collection of methodologies that aim to exploit the tolerance for imprecision
and uncertainty to achieve tractability, robustness, and low solution cost. Its principal constituents are fuzzy
logic, neurocomputing, and probabilistic reasoning. Soft computing is likely to play an increasingly
important role in many application areas, including software engineering. The role model for soft computing
is the human mind.
Soft computing is not precisely defined. It consists of distinct concepts and techniques which aim to
overcome the difficulties encountered in real world problems. These problems result from the fact that our
world seems to be imprecise, uncertain and difficult to categorize. For example, the uncertainty in a
measured quantity is due to inherent variations in the measurement process itself.
The uncertainty in a result is due to the combined and accumulated effects of these measurement
uncertainties which were used in the calculation of that result.
Artificial Intelligence fills the gap between the scientists of human behaviour and the computer scientists.
One thing we know about computers is that given a set of rules written in the programming language of the
computer the rules will always be obeyed exactly. So human scientists can test their theories about human
behaviour by converting their rules to computer programs and seeing if the behaviour of the computer in
executing these programs is like the natural behaviour of a human being, or at least that small subset of
human behaviour they are studying. Computer scientists can look at modelling human behaviour as a
challenge to their programming abilities: if a person can do something, can we write a computer program
which does the same thing?
The name Artificial Intelligence is rather an emotive one, though it has become established now and is
unlikely to be changed. It may be interesting to debate whether human beings are just very complex meat
machines or whether it would be theoretically possible to create a real artificial intelligence, that is a
computer program which really does behave as a full human being, but studying Artificial Intelligence does
not involve committing oneself to such beliefs. In fact, one thing that Artificial Intelligence has quite firmly
established is that human beings are more complex than some rather nave analyses have suggested.
Predictions that human-like artificial intelligences are soon to be built are often just media sensationalism,
and have so far proved wrong.
Unit-2
There is huge amount of data available in Information Industry. This data is of no use until converted into
useful information. Analysing this huge amount of data and extracting useful information from it is
necessary.
The extraction of information is not the only process we need to perform; it also involves other processes
such as Data Cleaning, Data Integration, Data Transformation, Data Mining, Pattern Evaluation and Data
Presentation.
Once all these processes are over, we are now position to use this information in many applications such as
Fraud Detection, Market Analysis, Production Control, Science Exploration etc.

Unit-3
The study of neural networks started by the publication of McCulloch and Pitts. The singlelayer networks,
with threshold activation functions, were introduced by Rosenblatt. These types of networks were called
perceptrons. In the 1960s it was experimentally shown that perceptrons could solve many problems, but
many problems which did not seem to be more difficult could not be solved. These limitations of one-layer
perceptron were mathematically shown by Minsky and Papert in their book Perceptron. The result of this
publication was that the neural networks lost their interestingness for almost two decades. In the mid-1980s,
back-propagation algorithm was reported by Rumelhart, Hinton, and Williams, which revived the study of
neural networks. The significance of this new algorithm was that multilayer networks could be trained by
using it.
Neural network makes an attempt to simulate human brain. The simulating is based on the present
knowledge of brain function, and this knowledge is even at its best primitive. So, it is not absolutely wrong
to claim that artificial neural networks probably have no close relationship to operation of human brains. The
operation of brain is believed to be based on simple basic elements called neurons which are connected to
each other with transmission lines called axons and receptive lines called dendrites. The learning may be
based on two mechanisms: the creation of new connections, and the modification of connections. Each
neuron has an activation level which, in contrast to Boolean logic, ranges between some minimum and
maximum value.
Unit-4
Fuzzy set theory was developed by Lotfi A. Zadeh, professor for computer science at the University of
California in Berkeley, to provide a mathematical tool for dealing with the concepts used in natural language
(linguistic variables). Fuzzy Logic is basically a multivalued logic that allows intermediate values to be
defined between conventional evaluations.
However, the story of fuzzy logic started much earlier. To devise a concise theory of logic, and later
mathematics, Aristotle posited the so-calledLaws of Thought. One of these, theLaw of the Excluded
Middle, states that every proposition must either be True (T) or False (F). Even when Parminedes proposed
the first version of this law (around 400 Before Christ) there were strong and immediate objections: for
example, Heraclitus proposed that things could be simultaneously True and not True. It was Plato who laid
the foundation for what would become fuzzy logic, indicating that there was a third region (beyond T and F)
where these oppositestumbled about. A systematic alternative to the bi-valued logic of Aristotle was first
proposed by ukasiewicz around 1920, when he described a three-valued logic, along with the mathematics
to accompany it. The third value he proposed can best be translated as the term possible, and he assigned it
a numeric value between T and F. Eventually, he proposed an entire notation and axiomatic system from
which he hoped to derive modern mathematics.
Unit-5

Genetic algorithm starts off with population of randomly generated chromosomes, each representing a
candidate solution to the concrete problem being solved and advances towards better chromosomes by
applying genetic operators based on the genetic processes occurring in nature. So far, GAs had a great
measure of success in search and optimization problems due to their robust ability to exploit the information
accumulated about an initially unknown search space. Particularly GAs specialize in large, complex and
poorly understood search spaces where classic tools are inappropriate, inefficient or time consuming.
As mentioned, the GA's basic idea is to maintain a population of chromosomes. This population evolves
over time through a successive iteration process of competition and controlled variation. Each state of
population is called generation. Associated with each chromosome at every generation is a fitness value,
which indicates the quality of the solution, represented by the chromosome values. Based upon these fitness
values, the selection of the chromosomes, which form the new generation, takes place. Like in nature, the
new chromosomes are created using genetic operators such as crossover and mutation.
Question Bank
Unit-1
1. Compare and contrast hard and soft computing. Write Applications of Soft Computing. [RGPV
2013]
2. Write Neuro-Fuzzy and Soft Computing characteristics. [RGPV 2012]
3. Explain the signification of hidden layer. How it is useful in pattern recognition and control problem?
[RGPV 2012]
4. State the applications of artificial neural networks and explain any two in details. [RGPV 2012]
5. Write short note on AI vs ANN. [RGPV 2014]
Unit-2
1. Explain unsupervised learning mechanism in contrast with a supervised learning mechanism. [RGPV
2013]
2. Write short note on ADALINE and MEDALINE. [RGPV 2012]
3. Differentiate between biological neuron and artificial neuron on the basis of structure and function of
a single neuron. [RGPV 2013]
4. Explain backpropagation algorithm for MLP with a neat signal flow graph. [RGPV 2012]
5. Train a perceptron network for learning a binary OR gate function. Work out two complete iterations.
[RGPV 2012]
Unit-3
1. Draw the architecture of Adaptive Resonance Theory. [RGPV 2014]
2. What is the other name of Auto associative net? Content addressable nature. [RGPV 2014]
3. Briefly describe the architecture of an ARTZ network. [RGPV 2013]
4. Full CPN is more efficient than the forward only CPN; Justify. [RGPV 2013]
5.

Give the Limitations and Applications of Hopfield and Boltzman. [RGPV 2013]

Unit-4
1. Define a fuzzy set and explain the concept of a fuzzy number. What is the significance of fuzziness?
[RGPV 2014]
2. Define Support. The support of the fuzzy set A is the set of all points x in X such that
A(x) > 0. Support(n) = {x | A(x) > 0 }[RGPV 2013]
3. Define Core. A core of a fuzzy set A is the set of all points x in X such that A(x) = 1.
Core(A) = {x | A(x) =1[RGPV 2012]
4. Define normality. A fuzzy set A is normal if its core is non empty. In otherwords, we can always find
a point xX such that A(x) =1. [RGPV 2012]
5. Define fuzzy sington. A fuzzy set whose support is a single point in X with A(x) =1 is called a
fuzzy singleton. [RGPV 2014]
Unit-5
1. What is genetic algorithm? Genetic algorithms are search algorithms based on the mechanics of

natural selection and natural genetics. [RGPV 2013]


2. What is the theme of research on genetic algorithms? The central theme of research on genetic

algorithms has been robustness, the balance between efficiency and efficacy necessary for survival in
many different environments. [RGPV 2014]
3. What is schema? A schema is a similarity template describing a subset of strings with similarities at certain
string positions. [RGPV 2013]
4. What is cross over? Cross over is a recombination operator. [RGPV 2014]
5. What are the types of cross over? Single site cross over, two point cross over, multipoint cross over and
uniform cross over. [RGPV 2013]

Old Question Paper

You might also like