Q1. Explain three types of fuzzy inference system (FIS) in detail.

FIS: Fuzzy rule-based systems, fuzzy models, and fuzzy expert systems are
generally known as fuzzy inference systems. The key unit of a fuzzy logic system
is FIS. The primary work of this system is decision making. FIS uses "IF ...
THEN" rules along with connectors "OR" or "AND" for making necessary
decision rules. The input to FlS may be fuzzy or crisp, but the output from FIS is
always a fuzzy set. When FlS is used as a controller, it is necessary to have crisp
output.
Three types of FIS:
1. Mamdani FIS:
Ebashim Mamdani proposed this system in the year 1975 to control a
stream engine and boiler combination by synthesizing a set of fuzzy rules
obtained from people working on the system. In this case, the output
membership function are expected to the fuzzy sets. After aggregation
process, each output variable contains fuzzy set, hence defuzzification is
important at the output stage. The followmg steps have to be followed to
compute the output from this FIS:
Step 1: Determine a set of fuzzy rules.
Step 2: Make the inputs fuzzzy using input membership function.
Step 3: Combine the fuzzified inputs according to the fuzzy rules for establishing
a rule strength.
Step 4: Determine the consequent of the rule by combining the rule strength and
the output membership function.
Step 5: Combine all the consequents to get an output distribution.
Step 6: Finally, a defuzzified output distribution is obtained.
The fuzzy rules are formed using "IF-THEN" statements and "AND/OR''
connectives. The consequence of the rule can be obtained in two steps:
i.
ii.

by computing the rule strength completely using the fuzzified inputs

from the fuzzy combination;
by clipping the output membership function at the rule strength.

The outputs of all the fuzzy rules are combined to obtain one fuzzy output
distribucion. From FIS, it is desired to get only one crisp output. This crisp
output may be obtained from defuzzification process. The common
techniques of defuzzification used are center of mass and mean of
maximum.

 Consider a two-input Mamdani FIS with two rules. The model fuzzifies the
two inputs by finding the intersecrion of two crisp input values with rhe
input membership function. The minimum operation is used to compute
the fuzzy input "and" for combining the two fuzzified inputs to obtain a
rule strength. The output membership function is clipped at the rule
strength. Finally, the maximum operator is used to compute the fuzzy
output "or" for combining rhe output of the two rules. This process is
illusrrated in Figure 12-2.
2. Sugeno Fuzzy Model:
The format of the fuzzy rule of a Sugeno fuzzy model is given by,
IF x is A and y is B THEN z = f(x,y)
Where AB are fuzzy sets in the antecedents and z = f(x,y) is a crisp function
in the consequent. Generally, f(x,y) is a polynomial in the input variable x
and y, If f(x,y) is a first-order polynomial, we get first-order Sugeno fuzzy
model. If f is a constant, we get zero-order Sugeno fuzzy model. A zero-

order Sugeno fuzzy model is functionally equivalent to a radial basis

function network under certain minor constrains.

The main steps of the fuzzy inference process namely,

1. Fuzzyfying the inputs;
2. Applying the fuzzy operator
are exact the same. The main difference between Mamdanis and Sugenos
method is that Sugeno output membership function are either linear or
constant.
The rule format of Sugeno form is given by
If 3 = x and 5 = y then output is z = ax + by + c
For a Sugeno model of zero order, the output level z is a constant. The
operation of a Sugeno rule is as shown in figure 12-3.
Sugeno's method can act as an interpolating supervisor for multiple linear
comrollers, which are to be applied, because of the linear dependence of
each rule on the input variables of a system. A Sugeno model is suited for
smooth interpolation of linear gains that would be applied across the input
space and for modeling nonlinear systems by interpolating between
multiple linear models. The Sugeno system uses adaptive techniques for
constructing fuzzy models. The adaptive techniques are used to customize
the membership functions.

3. Tsukamoto fuzzy model:

In the Tsukamoto fuzzy models, the consequent of each fuzzy if-then rule
is represented by a fuzzy set with a monotomical MF, as shown in figure
below. As a result, the inferred output of each rule is defined as a numeric
value induced by the rule firing strength. The overall output is taken as the
weighted average of each rules output. Figure below illustrates the
reasoning procesdure for a two-input two-rule system.
Since each rule infers a numeric output, the Tsukamoto fuzzy model
aggregates each rules output by the method of weighted average and thus
avoids the time-consuming process of defuzzification. However, the
Tsukamoto fuzzy model is not used often since it is not as transparent as
either the Mamdani or Sugeno fuzzy models. Since the reasoning method
of the Tsukamoto fuzzy model does not follow strictly the compositional
rule of compositional rule of inference, the output is always crisp even
when the inputs are fuzzy.

Q2. Comparison between Biological Neuron and Artificial Neuron.

Criteria
Biological Neuron
Artifical Neuron
The cycle time of execution in The cycle time of execution in
Speed
the biological neuron is of few the biological neuron is of few
milliseconds.
nanoseconds.

Processing

Biological neuron can perform

massive parallel operations
simultaneously but it process
slower than ANN.

Size and The total number of neurons in

complexity the brain is about 1011 and the
total number of interconnections
is about 1015. Hence, it can be
noted that the complexity of the
brain is comparatively higher,
i.e. computational work takes
places not only in the brain cell
body, but also in axon, synapse,
etc.
Therefore
size
and
complexity of Biological neuron
is more than that of an atrificial
neuron.
The biological neuron stores the
Storage
information
in
its
capacity
(memory) interconnections or in synapse
strength.
Tolerance The bilogical neuron processes
fault tolerant capability.
The control mechanism is
Control
mechanism complex compared to ANN
because there is no control unit
for monitoring in the brain. The
strength of neuron in the brain
depends on the active chemicals
present and whether neuron
connections are strong or weak
as a result of structure layer
rather than individual synapses.

The artificial neuron can also

perform
several
parallel
operations simultaneously but,
in general, the artificial neuron
network process is faster than
that of brain.
The size and complexity of an
ANN is based on the chosen
application and the network
designer.
The
size
and
complexity of ANN is less than
that of an biological neuron.

In an ANN stores information in

its
contiguous
memory
locations.
The ANN has no fault tolerance.
The control mechanism of an
ANN is very simple compated
to that of biological neuron
because ANN modeled using
computer using a computer,
there is a control unit present in
CPU, which can transfer and
control precise scalar values
from unit to unit.

Q3. Explain any four defuzzification methods with sutable example.

1. Max-Membership Principle:
This method is also known as height method and is limited to peak output
functions. This method is given by the algebraic expression

Max-Membership defuzzification method

2. Centroid Method:
This method is also known as center of mass, center of area or center of
gravity method. It is the most commonly used defuzzification method. The
defuzzified output x* is defined as

This method is illustrated in Figure below.

Centroid defuzzification method

3. Weighted Average Method:

This method is valid for symmetrical output membership function only.
Each membership function is weighted by its maximum membership value.
The output in this case is given by,

 Where denites algebraic sum.

4. Mean-Max Membership:
This method is also known as the middle of the maxima. This is closely
related to max-membership method, except that the locations of the
maximum membership can be non-unique. This is illustrated in figure
below,

The defuzzified output is given by

Where a and b are shown in the figure.

Q4. What is leraning in neural networks? Compare different learning rules.

Learning: Learning or training is a process by means of which a neural network
adapts itself to a stimulus by making proper parameter adjustments resulting in
the production of desired response.
Q5. Explain different membership functions.
1. Triangular
2. Trapezoidal
3. Gaussian
4. Singleton
1. Triangular:
Defined by its lower limit a, its upper limit b, and the modal value m, so
that a < m < b. We call the value b - m margin when it is equal to the value
m a.

2. Singleton:
It takes the value 0 in all the universe of discourse except in the point m,
where it takes the value 1. It is the representation of a crisp value.

3. Trapezoid Function:
Defined by its lower limit a and its upper limit d, and the lower and upper
limits of its nucleus, b and c respectively.

4. Gaussian Function:
This is the typical Gauss bell, defined by its midvalue m and the value of
> 0. The smaller is, the narrower the bell.

Q6. Explain with neat diagram Supervised Learning and Unsupervised

Learning.
1. Supervised Learning:
The learning here is performed with the help of a teacher. Let us take the
example of the learning process of a small child. The child doesn't know
how to read/write. He/she is being taught by the parenrs at home and by
the reacher in school. The children are trained and molded to recognize rhe
alphabets, numerals, etc. Their each and every action is supervised by a
teacher. Acrually, a child works on the basis of the output that he/She has
to produce. All these real-time events involve supervised learning
methodology. Similarly, in ANNs following the supervised learning, each
input vector requires a corresponding target vector, which represents the
desired output. The input vector along with the target vector is called
training pair. The network here is informed precisely about what should be
emitted as output.
The block diagram of figure 2-12 depicts the working of a supervised
learning network.

 During training, the input vector is presented to the network, which results
in an output vecror. This output vector is the actual output vector. Then the
actual output vector is compared with the desired (target) output vector. If
there exists a difference berween the two output vectors then an error signal
is generated by the network. This error signal is used for adjustment of
weights until the actual output matches the desired (target) output. In this
type of training, a supervisor or teacher is required for error minimization.
Henxe the network trained by this method is said to be using supervised
training methodology. In supervised learning, it is assumed that the correct
target output values are known for each input pattern.
2. Unsupervised Learning:
The learning here is performed without the help of a teacher. Consider the
learning process of a tadpole, it learns by itself, that is, a child fish learns
to swim by itself, it is not taught by its mother. Thus, its learning process
is independent and is not supervised by a teacher. In ANNs following
unsupervised learning, the input vectors of similar type are grouped
without the use of training data t specify how a membership of each group
looks or to which group a number belongs. In each training process, the
network receives rhe input patterns and organizes these patterns to form
clusters. When a new input panern is applied, the neural network gives an
output response indicating the class to which the input pattern belongs. If
for an input, a pattern class cannot be found then a new class is generated
The block diagram of unsupervised learning is shown in Figure 2-13.

From Figure 213 it is clear that there is no feedback from the environment to
inform what the outputs should be or whether the outputs are correct. In this case,
the network must itself discover patterns, regularities, features or categories from
the input data and relations for the input data over the output. While discovering
all these features, the network undergoes change in its parameters. IThis process
is called self-organizing in which exact clusters will be formed by discovering
similarities and dis-similarities among the objects.