, POSSE S/ON OF MOBI LES IN EXAM IS UFM PRACT ICE.
Enrolm ent No. l1Q ,o UQ QCg
Jaypee Institut e of Informa tion Techno logy, Noida
Tl Examin ation, Odd Semeste r 2023 (B. Tech.)
Course Title: 21B12C S417 Maximum Time: 1 Hr
Course Code: Machin e Learnin g an~ Big Data Maximu m Marks: 20
co J: Identify the characteristics of datasets and use of machine learning techniques.
CO2: Demonstrate online learning methods for big data applications.
C03: Select and implement machine learning techniques and computing environment
that
are suitable for the application under consideration.
C04: Implement parallel learning algorithms using OpenMP/CUDA/OpenCL.
COS: Evaluation and validate different problems associated with big data characte
ristics for
high dimensionality, and in scalability issues.
Important Instruction: Attempt all questions.
J).1-:-Write an algorithm for prefix sum on a d-dimensional hypercube. Explain it also
with 3-dimesnional hypercube. Marks 4 (CO 4~ Apply)
~ r i t e the odd-even transportation sort algorithm. Sort the following list using it:
3 2 3 8 5 .6 4 1 7 IO Marks 4 (CO 4, Apply)
~ u n t the total number of times each word appears in a collection of documents
using the MapReduce. Suppose we have the following three text document:,
Marks 4 (COS, Evaluate)
Document 1: "Hello world"
Document 2: "Hello again"
Document 3: "World, it is a small world''
~ the application worktlow of YARN in Hadoop for the execution of a job submitted
by the client. Explain each step in detail with the help of an example.
Marks 4 (COS, Evaluate)
QS: Find the MFLOPS for executing the given code where the extracting data from
~mo ry[M ] takes the 256ns and from registers[R] take 32ns. The number of inputs
is
1000.
Marks 4 (CO 4, Apply)
for (i =l; i <= n; i ++)
{
for G= 1; j <=i 2; j ++)
{
for (k= 1; k<= n/2; k++)
{
' Load RO, [M]
•
Load RI, [M]
:::,d :t,' [~~ R3] 'f
Add RO, [RO, R5]
Load R7, [M]
Mui R8, [RO, R7]
Add R7, [R7, RS]
Write [M], RO
Write (M], R7
}
}
}
Name \J/}J}k~v
~SSESI ON OF MOBILE S IN EXAM IS UFM PRACTICE.
~ Enrollme nt No. _Q\D 4Q~
2
Jaypcc Institute of Information Technolo gy, Noida
Test I Examination ODD SEM 2023
B. Tech. 7th Semester
Course Name: Machine Learning and Big Data Maximum Time: 1 Hr
Course Code: 21B12CS 417 Maximum Marks: 20
-
C431-12. I Identify the characteristics of data sets and use of machine learning technique.
----·
C43 l-12.2 Demonstrate onlinc learning method for big data applications.
----·--· - --- -·------··--- --- -· -------
C431-J2.3 Implement machine lcar~ing technique and computing environment that are suitable for the application
under
consideration.
C431-12.4 Implement parallel learning algorithms using OpenMP/CUDA/OpenCL.
C431-12.5 Evaluate and validate different problems associated with big data.
Note: Attempt all the questions in order.
~uppo se you are building a decision tree to classify whether a student will pass or fail an examinati
on based on
feature: the number of hours they studied. You have a dataset with the following information.
Student Hours Pass
·-
A 2 NO
-
8 1 NO
- ·-··
C 3 YES
----· -----
D 4 YES
E 2 NO
F 3 YES
Suppose in the first split your dataset is divided into two subsets as
Subset I Subset I
Student Hours Pass Student Hours Pass
- ·-----
A 2 NO D 4 YES
-·
B I NO E 2 NO
C 3 YES F 3 YES
-·
Using Entropy find the information gain.( COi, Understan d Level) fSMarks)
/'.oz(suppose we have two classes, Class A and Class B with the following data points:
Class A: (1,2) ,(2,3), (3,3)
Class B: (3, 1) ,(4,2), (5,3)
We want to find the equation of the decision boundary(hyperplane) that separate theses two classes using
linear SVM.
Assume point (1,2) is the Support vector of Class A and (3,1) is for Class B.. ( cot, Understan d Level I
fSMarksJ
~ptim ize the function f(x, y)=x 2 +y2 to find the minimum point using Newton method for initial
guess xo=2 and
y 0=3 ( CO2, Apply Level J ISMarks) .
P.T.O
--
• 7V EX AM IS UF M PR AC TIC E.
pO SSE SIO N OF MO BIL ES l
Enr ollm ent No.__fk)1DYO '18
Nam e ,J/[Link] Jb t;
'
Jay pee I ns t 1·tut e of Inf orm atio n Tec hno log y' No ida
End term Exa min atio n, OD D 202 3
• [Link] h. VII Sem est er
Course Title: Machine arning and Big Ma xim um Time: 2 hrs
Data
Course Code: 21B 127 417 Ma xim um Marks: 35
After co111pletio,,lr the cou rse, students
sha ll be able to:
C43 J-12. I ~enti-fy the characteristics of datasets and
use of machine learning techniques.
C43 l-l2. 2 Demonstrate online leamirtg methods for
Big Data applications.
C43 l-12.3 Select and implement machine learning techn
iques and computing environment that are
the application unde r consideration. • suitable for
C431-12.4 Implement parallel learning algorithms using
OpenMP/CUDA/OpenCL.
C431-12.5 Evaluate and validate different prob'lems
associated with big data characteristics for
dimensionality, and in scalability issues. high
Note: Atte mpt all que stio ns.
- ~ Find the entropy JI (Pas sed I GPA
) and also the information gain for the sam
the dataset below: e spli t given for
The decision tree pred icts if student pass
ed machine learning (Ye s or No) , based on
GPA (High, Med ium , Low ), and whether prev ious
or not they studied.
GP A Stu died Passed
L F F
L T T
M F F
M T T
H F T
H T T
[C4 31-1 2.1, Und erst and , 3+2 Ma rks ]
~Apply the Bitonic Merge Sort to sort the following sequence:
J
3, 5, 8, 9, 10, 12, 14, 20, 95, 90, 60, 40, 35,
23, 18, 0
✓.Apply
[C4 31-1 2.4, App ly, 5 Ma rks ]
the New ton' s seco nd orde r optimization
method on the following quadratic function
f(x) = 2x 2 -6x +5 to find the min imu m of :
this function. Find the value of f(x) afte r
iteration. Take initial gue ss xo = 2. the seco nd
[C431-12.2, Apply, 5 Ma rks]
~ d the total storage capacity and effective storage capacity for the following problem:
1
You are working with a Hadoop cluster that con_sis~s of 5 data nodes, each with a storage
capacity of 2TB. The cluster uses the Hadoop D1str1buted File System (HDFS), and you have
a large dataset of 20TB. [C431-12.5, Evaluate, 2 Marks]
Note: Consider HDFS replication factor as 3.
~termine the Single Value Decomposition of A, ULVT, where
A= G 2
3
[C431-12.3, Apply, 8 Marks]
~Explain the word/class embedding with the help of an example. Why it is preferred over one-
hot encoding? [C431-12.3, Apply, 3+2 Marks]
oc{Explain the Query-Synthesis method in active learning with Genei;ative Adversarial Network.
Briefly .elaborate it with the help of diagram. (C431-12.3, Apply, 3+2 Marks)
····•~***********************************************************************
