U.G.C. NET Exam. December 2022 Paper II (COMPUTER SCIENCE) And Applicatation.

Given that
TCP segment of size = 1400 bytes
IP header size = 20 bytes
At R packet will be Fragmented
MTU = 800 + 20 (Header)
Packet size = segment size + IP Header Size
= 1400 + 20
= 1420 bytes
Hence Packet's Data (1400) will be divided in two packets = (800 + 600)
First Fragment = (800 + 20) // Data + Header
Second Fragment = (600 + 20) bytes
A. Two fragments are created at R and IP datagram size carrying the second fragment is 620 bytes. It is correct.
B. It is second fragment is lost then R resend the fragment with IP Identification value of 0x1234. It is incorrect As if any of the fragments is lost then the entire packet/TCP segment has to be retransmitted.
C. If the second fragment lost then P required to resend the entire TCP segment. It is correct. If any of the fragments is lost then the entire packet /TCP segment has to be retransmitted this one of the drawbacks of fragmentation.
D. TCP destination port can be determined by analyzing the second fragment only. It is incorrect as Destination port number can be identified by looking at any packet or fragment not just only from second.
So correct option is (b).

Silly Window Syndrome is a problem that arises due to poor implementation of TCP.
Cause 1 : Sender window transmitting one byte of Data repeatedly– This problem is solved by Nagla's algorithm.
Cause 2 : Receiver window accepting one byte of data repeatedly–
This problem is solved by problem is clark's solution or clark''s algorithm.
So option (1) is correct answer.

Genetic Algorthms–An AI system that mimics the evolutionary, survival of the fittest process to generate increasing to better solutions to a problem. Genetic Algoritms use three concepts of evolution : Selection or the survival of the fittest.

Learning algoritm
A learning algorithm is a set of instructions used in machine learning that allows a computer program to imitate the way a human gets better at characterising some type of information. (A) Logistic regression–It is used for binary classification problems. It used to obtain an odds ratio in the presence of more than one variable.
(B) Back propagation–It is the essence of neural net training. It is the method of fine tuning the weights of a neural net based on the error rate obtained in the previous iteration. It is a standard method of training artificial neural networks.

A single perceptron can represent any linearly separable function. AND, OR and NOR are linearly separable in 2D plane and hence a single perception (single perception layer) can represent them. So option (c) is correct.

Planar Graph
(A) 4 color Theorem–"The chromatic number of a planar graph is no greater than 4."
A (iii) Planar Graph → 4 colorable
(B) Bipartite Graph–A bipartite graph is possible if the graph coloring possible using two colors such that vertices in a set are colored with the same color.
B (iv) Bipartite Graph → 2-Colorable.
PERT (Nature of modal) PERT is a probabilities modal i.e. it takes into account uncertainties involved in the estimation of time of a Job or an activity. It uses three estimates of the activity time optimistic of the activity time optimistic pessimistic and most likely. Thus the expected duration of each activity is probabilistic and expected duration indicates that there is 50% probability.
(i) (C) PERT→ Probabilistic modal.
CPM (Nature of Model) CPM is a deterministic model i.e. does not take into account the uncertainties involved in the estimation of time for execution of a job or an activity.
(D) (ii) Deterministic model→CPM
So option (d) is correct answer.

Direct Addressing Mode
In direct addressing mode, the data to be operated is available inside a memory location and that memory location is directly specified as an operand. The operand is directly available in the instruction itself Example. STA F FFFH
(A) (iii) Direct addressing → STA F FFFH
(B) Indirect Addressing–In register indirect addressing mode, the data to be operated is available inside a memory location and that memory location is indirectly specified by register pair.
Example
MOV A, M (Move the contents of the memory location pointed by the H–L pair to the accumulator. (B) (iv) Indirect Addressing mode → MOV AM
(C) Register Addressing–In register addressing mode the data to be operated is available inside the register (S) and register (S) is are operands. Therefore the operation is performed within various registers of the microprocessor.
Example
: MOV A, B (move the contents of register B to register A)
C (ii) Register Addressing → Move A, B
(D) Immediate Addressing–Immediate addressing mode the source operand is always data. If the data is 8 bit, then the instruction will be 2 bytes, if the data is 16 bit then the instruction will be 3 bytes.
D (I) MVIA, FFH →Immediate
Addressing
So option (c) is correct.

A. (iii) If the implementation of generated or derived classes differ only through a parameter we have to use templates.
B. (i) If the actual types of objects used cannot be known at compile time, then we have to use class hierarchy.
C (ii) If inline operations are essential and templates used then we have improved runtime efficiency.
D. (iv) To gain access to differing instances for derived classes through base we have to use explicit casting.
So option (b) is correct answer

A. (iii) Apache Kafka is an open-source platform that was created by Linkedin.
B. Pig is a general-purpose computing model and runtime system for distributed data analytics.
C. (iv) Apache Mahout is an open-source platform used for creating scalable machine learning algorithms.
D. (ii) Mapreduce is a general-purpose computing model and runtime system for distributed data analytics.
So option (c) is correct answer.

IPC helps achieve these things