UGC NET Computer Science Machine Learning Previous Year Questions (PYQs) – Page 1 of 3
Study Stuff for MCA Examinations
Choose the correct option describing the features of Artificial neural network
I. It is essentially machine learning algorithm.
II. It is useful when solving the problems for which the data set is very large.
III. They are able to extract features without input from the programmer.
IV. These are systems modeled on the human brain and nervous system
Choose the correct answer from the options given below:
Solution
Statement I:
Artificial Neural Network is a machine learning technique.
So, statement I is correct.
Statement II:
ANN is useful for large datasets because it can learn complex patterns from data.
So, statement II is correct.
Statement III:
Neural networks can automatically learn important features from data without manual feature programming.
So, statement III is correct.
Statement IV:
ANN is inspired by the human brain and nervous system.
So, statement IV is correct.
Therefore, all statements are correct.
Solution
A Perceptron is a single-layer feed-forward neural network used for linear classification.
Solution
Backpropagation computes error at the output layer (sink) and propagates the error backwards through hidden layers toward the input layer (source) to update weights.
If two variables $V_1$ and $V_2$ are used for clustering, then consider the following statements for $k$ means clustering with $k=3$.
Statement I: If $V_1$ and $V_2$ have correlation of $1$, the cluster centroid will be in straight line.
Statement II: If $V_1$ and $V_2$ have correlation of $0$, the cluster centroid will be in straight line.
In the light of the above statements, choose the correct answer from the options given below:
Solution
If $V_1$ and $V_2$ have correlation $1$, then they have perfect positive linear relationship.
So, all data points lie on a straight line.
In $k$ means clustering, centroids are calculated as the mean of points in each cluster.
Mean of points lying on a straight line also lies on the same straight line.
Therefore, the cluster centroids will be in a straight line.
Statement II is false.
If $V_1$ and $V_2$ have correlation $0$, then there is no linear relationship between them.
So, the data points are not necessarily on a straight line.
Therefore, the cluster centroids will not necessarily be in a straight line.
$K$-mean clustering algorithm has clustered the given $8$ observations into $3$ clusters after $1^{st}$ iteration as follows:
$C1:{(3,3),(5,5),(7,7)}$
$C2:{(0,6),(6,0),(3,0)}$
$C3:{(8,8),(4,4)}$
What will be the Manhattan distance for observation $(4,4)$ from cluster centroid $C1$ in second iteration?
Solution
Cluster $C1$ contains the points
$(3,3),(5,5),(7,7)$
Centroid of $C1$ is
$\left(\dfrac{3+5+7}{3},\dfrac{3+5+7}{3}\right)$
$=(5,5)$
Now, Manhattan distance between $(4,4)$ and $(5,5)$ is
$|4-5|+|4-5|$
$=1+1$
$=2$
If we convert a decision tree to a set of logical rules, then:
Solution
Solution
Which of the following is an example of unsupervised neural network?
Solution
A self-organizing feature map is an unsupervised neural network.
It groups similar input patterns without using labelled output data.
Back-propagation is generally supervised learning.
The value of the derivative of the Sigmoid function given by
$f(x)=\dfrac{1}{1+e^{-2x}}$
at $x=0$ is:
Solution
Given,
$f(x)=\dfrac{1}{1+e^{-2x}}$
Differentiate with respect to $x$:
$f'(x)=\dfrac{2e^{-2x}}{(1+e^{-2x})^2}$
At $x=0$,
$e^{-2(0)}=e^0=1$
So,
$f'(0)=\dfrac{2}{(1+1)^2}$
$=\dfrac{2}{4}$
$=\dfrac{1}{2}$
Solution
Reinforcement learning is generally formalized using a Markov decision process.
In a Markov decision process, the agent interacts with an environment using states and actions.
The agent knows the set of possible states and the set of possible actions.
Therefore, the blanks are:
Markov decision process, states
UGC NET Computer Science
Online Test Series, Information About Examination,
Syllabus, Notification
and More.
View More
UGC NET Computer Science
Online Test Series, Information About Examination,
Syllabus, Notification
and More.
View More
