UGC NET Computer Science Machine Learning Previous Year Questions (PYQs) – Page 3 of 3
Study Stuff for MCA Examinations
Consider the following Learning algorithms.
I. Logistic regression.
II. Back propagation.
III. Linear regression.
IV. Forward propagation.
Which of the following options represents classification algorithm?
Choose the correct answer from the options given below:
Solution
Statement I:
Logistic regression is a classification algorithm.
So, I is correct.
Statement II:
Back propagation is used for training neural networks, and in such questions it is treated as part of classification learning.
So, II is taken as correct.
Statement III:
Linear regression is used for predicting continuous values, not for classification.
So, III is false.
Statement IV:
Forward propagation is a computation step, not a classification algorithm.
So, IV is false.
Correct answer is:
I and II only
Which Boolean operation on two variables can be represented by a single perception layer?
I. X1 AND X2
II. X1 OR X2
III. X1 NOR X2
IV. X1 XOR X2
Choose the most appropriate answer from the options given below:
Solution
A single perceptron can represent only linearly separable Boolean functions.
Check each operation:
I. AND → linearly separable
II. OR → linearly separable
III. NOR → linearly separable
IV. XOR → not linearly separable
So, a single perceptron can represent:
I, II, and III only
Match the List-I and List-II
| List-I | List-II |
| a. Apache Kafka | I. Platform for constructing data flows for extract, transform, and load ETL processing and analysis of large datasets |
| b. Pig | II. General-purpose computing model and runtime system for distributed data analytics |
| c. Apache Mahout | III. Open-source platform that was created by LinkedIn |
d. MapReduce | IV. Open-source platform used for creating scalable machine learning algorithms |
Choose the correct answer from the options given below:
Solution
Apache Kafka was originally created by LinkedIn.
So,
a → III
Pig is used for constructing data flows for ETL processing and analysis of large datasets.
So,
b → I
Apache Mahout is used for creating scalable machine learning algorithms.
So,
c → IV
MapReduce is a general-purpose computing model for distributed data analytics.
So,
d → II
Correct matching:
a-III, b-I, c-IV, d-II
Solution
A is false because fuzzy C-means clustering is an unsupervised learning method.
B is true because PCA is used for dimensionality reduction.
C is true because Apriori is an unsupervised association rule mining technique.
D is false because this condition is called overfitting, not underfitting.
Which of the following genetic algorithm operation is computationally most expensive?
Solution
In Genetic Algorithms, the most computationally expensive part is usually reproduction, which involves:
• crossover
• mutation
• generating new offspring
• evaluating fitness for the next generation
These steps are repeated many times and require significant computation.
I. Find recursively all its density-connected points and assign them to the same cluster as the core point.
II. Find all the neighbor points with EPS and identify the core points with more than MinPts neighbors.
III. Iterate through the remaining unvisited points in the dataset.
IV. For each core point if it is not already assigned to a cluster, create a new cluster.
Choose the correct answer from the following:
Solution
In DBSCAN, first we find neighbors within EPS and identify core points.
Then for each unassigned core point, a new cluster is created.
After that, all density-connected points are recursively added to the same cluster.
Finally, the algorithm continues with the remaining unvisited points.
So, correct order is:
II → IV → I → III
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
