lOMoARcPSD|22950144

Sixth Semester University QP

Artificial Intelligence (SRM Institute of Science and Technology)

Scan to open on Studocu

Studocu is not sponsored or endorsed by any college or university

Downloaded by Sam Cruise ([email protected])
 lOMoARcPSD|22950144

Reg. No. o65

B.Tech. DEGREE EXAMINATION,MAY 2023
Sixth &. Seventh Semester

18CSC303J- DATABASE MANAGEMENTSYSTEMS

(For the candidatesadmittedduring the academicyear 2018-20!9to 2021-2022)
Note:
handed
Part - A should be answered in OMR sheet within first 40 minutes and OMR sheet should be
over to hall invigilatorat the cnd of 40h minute.
() Part - B & Part - Cshould be answeredin answer booklet.
Max. Marks: 100
Time: 3 hours
Marks BI CO PO

PART -A (20 x 1= 20 Marks)

Answer ALL Questions 1
of
1. An unsophisticated user whointeract with the system by invoking one
the applicationprograms that have been written previously.
(A) Applicationprogrammers (B) Naive user
(C) Specializedusers (D) Maintenanceuser
2 1
and the data
2. Which manages the allocation of space on disk storage
structuresused to representinformationstored on disk.
(A) File manager (B) Buffermanager
(C) Transactionmanager (D) Integritymanager
1 2 2
actuallystored.
3. The lowest level of abstractiondescribeshow the data are
(A) Logicallevel (B) View level
(C) Data independence (D) Physical level
1 2
one to define tables, integrity
4. SQL provides a rich language that allows
constraints,assertions
(A) Applicationprograms (B) Design language
(C) Data manipulationlanguage (D) Data definitionlanguage
2 3 2
form a primary key.
5. An entityset that does not have sufficientattributesto
(A) Strong entity set (B) Weak entity set
(C) Cardinalityset (D) Relationshipset
1 2 3 2

6. Which of the followingis a derived attribute?

(B) Phone number
(A) Street name
(C) City (D) Age
2 3
lower level entities created by
7. An important property of the higher and
specializationand generalization.
(A) Cardinality (B) Transform
(C) Schema (D) Attributeinheritance
3 2
values called the
8. For each attributeof a relation, there is a set of permitted
of that attribute.
(A) Domain (B) Tuple
(C) Instance (D) Schema
25MF6&7-18CSC303J
Page 1 of 4

Downloaded by Sam Cruise ([email protected])

 preventdald
lOMoARcPSD|22950144

databaseto
9. Which of the following best describes the process Of parsing in query usedin
4
18. Which of the following is same data?
access to the (B) Blocking
processing? caused by concurrent
(A) Checking the syntax of the (B) Translatngthe SQL query into (A) Locking (D) Chustering 6 2
SQL query an internal representation 2
(C) Indexing schedule
(C) Executingthe SQL query (D) Optimizingthe SQL query thatthe
schedulein database that ensures schedule.
is a property of in the
10. Identify the statementlhat best describesa view in a database. 1 4 2
19. oftransactions
is equivalentto same serial execution (B) Conflict
(A) A view is a type of index (B) Aview isa stored prOcedure (A) Ordering (D) Deadlock
(C) Aview is a virtual table that is (D) A view a type of table that (C) Serializability 1 6 2
based on the resuit of a select Contains data cannot
locks but it
can release
statement
20. In two phase locking protocol a transaction
acquire any new locks in thís phase.
11. Transaction control language commands such as are used to 2 4
(A) Shrinking (B) Growing
manage transactions in a database. (D) Serial
(C) Time stamp B co P0
(A) COMMIT, ROLLBACK and (B) EXIT and CLOSE Marks
SAVE POINT PART-B (5 x4 =20 Marks)
(C) LOCK and TIME (D) GRANT and REVOKE Answer ANY FIVE Questions
2 1 1
constraint is a constraint that ensures that a column or set of 1 management system. For eacn
21. List three responsibilitiesof a database
12. A 2 4

responsibility
columns satisfiesa specifiedcondition. responsibility,explain the problems that would arise if the
(A) Participation (B) Referential were not discharged.
(C) Count (D) Check 2 3
candidate key and
22. Explain the distinctions among the terms primary key,
13. A relationis saidto be in 2NF if itis INF and if all of itsnon-keyattributes 2 5
superkey.
are dependenton the entireprimarykey. 3 2
weak entity and
(A) Fully functionally (B) Closure 23. Provide an example to illustratethe relationshipbetween a
(C) Decomposition (D) Join its owner entity.
4 3 1 2
in the database
14. Which of the following operators in relational algebra is used to combine 24. Examine why it is important to include a Jogical level
2 5 2

two relations and return a relation that contain all the rows from both the architectureand what role this level plays in data abstraction.
relations,with duplicatesremoved? 4 4
be used
(A) Union (B) Intersection 25. Define check constraintsand provide an example of how they can
(C) Difference (D) Prolog to enforceddata integrity.
4 2 5 2
I 2 5 26. Define first normal form in DBMS with an example.
15. The refers to the set of all functional dependencies that can be
4 6 3
inferred from the given set of dependencies.
(B) Redundancy 27. Discuss the propertiesof a transactionin a DBMS.
(A) Isolation BL Co PO
(C) Inverse (D) Closure PART -C(5x 12 = 60 Marks)
Marks

2 5 2 Answer ALL Questions

normalization in database
16. Which of the following is a primary goal of 12 3 1 2
trained, developers
design? 28. a. Since many database system users are not computer
(A) To reduce data redundancy (B) To improvedata access speed hide the complexity from users through several levels of
abstraction
number of (D) To reduces the number of simplify users interactionwith the system. Illustratethe
various levels of
(C) To increase the
attributes relations abstractionwith an analogy, to the concept of data types in programming
1 6 1 languages.
transactiononce committed willpersist even in
I7. ThiS propertyensures that a
the event of a system failure. (OR) 12 3

(B) Consistency file processing system has a

b. Keeping organizational information in athem
(A) Atomicity in detail to understand the
(C) Isolation (D) Durability number of major disadvantages.Explain
purpose of databasesystems. 25MF6&7-18CSC303J

Page 3 of 4
25MF6&7-18CSC303J
2e 2 of 4 Downloaded by Sam Cruise ([email protected])
 lOMoARcPSD|22950144

most database features.Some aspects of 12

4 3 2

29, a. The basic ER Concepts can model

expressed by certain extensionsto the basic
adatabase may be more aptly of specialization and
ER model. Discuss the extended ER features
generalizationwith examples.
(OR) 12 4 3
b. A university has multiple departments, each of which offers several
multiple
courses. Each course is taught by a professor who may teach
university
courses. Based on this scenario, construct an ER model for the
and explain its various entities and relationships.Also show the mapping
cardinalities.
12 3 4 3
30. a. Show howdevelopers can write their own functions and procedures,store
them in the database and then invoke them from SØL statements.
Demonstratewith syntax and examples.
(OR) 12 3 4 3
b. Demonstratethe steps involvedin processingaquery with examples.
12 4 5 2
31. a. Define tuple relationalcalculus andstate the formal syntax of the
expressionform with examples.
(OR) 12 4 5 2
b. Discuss the conceps of functional dependency that are required for a
databaseto be in third normal form. Ilustratethem with properexamples.
12 2 6
32. a. Explain the two phase locking protocol and define how does it provide
solutionstoconcurrencyrelatedproblems.
(OR) 12 2 6 3
b. Discuss the basic concepts of serializabilityin transactions and suggest
methods for testing a given schedule.

25MF6&7-18CSC303J
Page 4 of 4

Downloaded by Sam Cruise ([email protected])

 lOMoARcPSD|22950144

Reg. No.

B.Tech. DEGREE EXAMINATION,MAY 2023

Sixth Semestcr

18CSC304J -COMPLIERDESIGN
(or the candidatesadmittedduring the academicyear
Note: 2018-2019 to 2021-2022)
() Iar- A Should be answercd in OMR sheet within first 40 minutes and
over to hall invigilatorat the end of OMR sheet should be handed
(ii) Part - B& Part -C should be 40h minute.
answered in answer booklet.
Time: 3 hours
Max. Marks: 100

PART - A(20 x 1= 20 Marks) Marks PI Co PO

Answer ALL Questions

1. What is the minimum number of statesto
recognizethe language 1 1 1 1

L=w|we(0+1+2)*?
(A) 1 (B) 3
(C) 4 (D) 2
2. Which of the following is correct sequence of finite state automata in terms 1 1 1 1
of flexibility?
(A) NFA, DFA, e-NFA (B) DFA, NFA, e-NFA
(C) e-NFA, DFA, NFA (D) e-NFA, NFA, DFA
3. Entries in symbol table are made in 1 1 I
(A) Semanticanalysis (B) Lexical and syntax analysis
(C) Lexicalanalysis (D) Code generation
1 1 1 1
4. Any languageacceptedby a finite automationis
(A) Only context free (B) Only context sensitive
(C) Only regular (D) Both regular and contextfree
1 2 2
5. In which parsing, the parser constructsthe parse tree from the start symbol
and transformsit into inputsymbol
(A) Bottom-upparser (B) Top-down parser
(C) Both bottom-up andtop-down (D) LR parser
parser
1 2 1
6. Which of the followingis true?
(A) * has higher precedencethan + (B) - has higher precedencethan *
(C) + and * have same precedence (D) + has higher precedencethan *
1 1 2 2
7. A context free grammar is not closed under
(A) Dot operation (B) Union operation
(C) Concatenation (D) Iteration
1 1 2 1
8. Parsing is also known as
(A) Lexical analysis (B) Syntax analysis
(C) Semanticanalysis (D) Code generation
26MF6-18CSC304J
Page 1 of 4

Downloaded by Sam Cruise ([email protected])

 lOMoARcPSD|22950144

20. Which of the followingsymbol table implementationmakes efficient use of 1 S 2

9 A bottom-upparser generates
(A) Right most derivation (B) Right most derivation in reverse memory?
(C) Left most derivation (D) Left most derivation in reverse (A) List (B) Search tree
(C) Hash table (D) Self-organizinglist
1 3 1
10. What is the similaritybetween LR, LALR and SLR?
(A) Use same algorithm, but (B) Same parsing table, but different
different parsing table algorithm PART -B (5 x 4 = 20 Marks) Marks BL Co PO

(C) Their parsing tables and (D) Both parsing tables and Answer ANY FIVE Questions
algorithm are similar but uses algorithm are different
top-down approach 21. Discuss in detail on compilerconstructiontools. 2 1 2

1 3 1 4 3 1 3
11. Which is nota shift reduce parser action? 22. Construct a deterministic finite automata that accepts numbers that are
(A) Shift (B) Goto divisibleby five.
(C) Reduce (D) Accept
2 3 2 3
1 3
23.i. Build a grammar withoutleft recursionfor the following
12. LR parser construct a type of derivation. A> ABd | Aala
(A) MMD (B) RMD B -’ Be|b
(C) LMD (D) CLR
2 3 2 3
1 ii. Build the grammarby doing left factoringfor the following grammar
13. is a tool that depicts the structure of basic blocks,helps to see the 4 1
A’ aAB|aA|a
flow of values flowing among basic blocks. B’ bB |b
(A) DAG (B) CAG
(C) SAG (D) PAG 4 2 2 2
24. Explain the parsingtechniqueswith a hierarchicaldiagram.
14. In algebraic expression simplification,a=a+l can simply be replaced by 4
4 2 3 3
25. Discuss the rules involvedin constructingthe SLR parsingtable.
(A) a (B) INCa A 3 4 3
(C) DECa (D) MUL a 26. Build the syntax tree and DAG for the followingexpression
(a*b) +(c- d)'(a b)
15. Which of the following is not a form of internediaterepresentation? 1 4

(A) Abstractsyntax tree (B) 3-addresscode 27. Discuss in detail on stack storage allocation. 4 2

(C) Directed cyclic graph (D) Reverse polish notation

Marks BL CO
16. Code generator uses function to determine the status of available 1 4 1 PART -C (5× 12 =60 Marks)
registersand the locationof name values. Answer AlLL Questions
(A) setReg (B) cinReg 12 2 1
(C) pfReg (D) getReg 28. a. Discuss in detail on phases of compiler and write down the output of each
phase for the expressiona:= b+c*50.
1 5 1
17. Which optimizationtechniqueis used to reduce the multiplejumps?
(A) Latter optimizationtechnique (B) Peepholeoptimizationtechnique (OR)
12 3
(C) Local optimizationtechnique (D) Code optimizationtechnique b. ConstructDFA for the regular expression ba(a + b) ab using first pos, last
pos and follow pos.
18. The graph that shows basic blocks and their successorrelationshipis called
3
(A) Dag (B) Flow graph 29. a. Constructpredictiveparsing table for the grammar, 12

(C) Control graph (D) Hamiltoniangraph E’E+T|T

T’T*F|F
19. Which is not part of runtimememory subdivisions? F’(E)| id
(A) Stack (B) Heap And parse the input id + id * id
(C) Static data (D) Access link (OR)

Page 2 of4 26MF6-18CS

26MF6-18CSC304J Page 3 of 4

Downloaded by Sam Cruise ([email protected])

 lOMoARcPSD|22950144

2 3 2 3
His father has brought
b. Rahul is applying for his dream job in Indian army. apply in any of the
an advertiscmentabout the eligibilitycriteria. Rahul can
greater than 174 cm.
two categories.The first categorysays height should be
should be greater than 174 cm
The second category says mandatory heighteducational qualificationcan be
and optional educationalqualification.The (UG) or Post
Tenth (T) or Higher Secondary (HS) or Under Graduate and common
Graduate (PG). Design CFG and eliminate left recursion
prefixes whereverapplicable.
(Grammarconstruction: 6 Marks
Left recursion : 2 Marks
Left Factoring: 4 Marks]
12 3 3 3
30. a. Considerthe following grammar
S’ AS|b
A’ SA|a
Construct the SLR parse table for the grammar. Show the actions of the
parser for the input string abab.

(OR)
12 3 3 3
b. Constructa canonicalparsingtable for the grammar given below
E’E+T
F’(E)
E’T
F’ id
T’T* F
T’F
And parse any string derived from the grammar.
12 3 4 3
31. a. Generate the three address code statement and construct the DAG
representationfor the expression I = a + a (b+c)-(b+c)'d.

(OR)
12 2 4 2
b. Discuss in detail on three address code representation with appropriate
examples.
12 2 5 2
32. a. Explain in detail about displays. Also explain how varibie length data is
handledby compilers.
(OR)
12 2 5 2
b. Discuss in detailabout storageallocationstrategieswith ablock diagram.

Page 4 of 4 26MF6-18CSC304J

Downloaded by Sam Cruise ([email protected])

 lOMoARcPSD|22950144

Reg. No.

B.Tech. DEGREE EXAMINATION,MAY 2023

Fifth and Sixth Semester

18ECO107T- FIBER OPTICS AND OPTOELECTRONICS

(For the candidates admitedfromthe ocademicyear 2018-20i9to 2021-2022)
Note:
(i) Part - A shouldbe answered in OMR shect within first 40 minutes and OMR sheet should be handed
over to hall invigilatorat the end of 40 minute.
(ü) Part - B & Part -Cshould be answered in answer bookiet.

Time: 3 hours Max. Marks: 100

Marks B CO PO
PART -A(20 x 1 - 20 Marks)
AnswerALLQuestions .2
1. The expressionfor the refractiveindex 'n' is given by
(A) n= (B) n=
(C) n= CV (D) n=Ya
1 2 1.2
2. The numerical aperature of a coaxial cable with core and cladding indices
given by 2.33 and 1.4 respectivelyis
(A) 3.73 (B) 0.83
(C) 3.46 (D) L.86
2 1
3. Identify the device that distributes light from a main fiber to one or more
branch fibers.
(A) Optical Coupler (B) Optical Slice
(C) Optical Connector (D) Optical Isolator
1 2 12
4. Numerical aperature in optical fiber is used to describe
(A) Light SpreadingAbility (B) Light GatheringAbility
(C) Light Leakage Abiiity (D) Light output from External
Shield

2 I,3
5. In singie mode fibers, the dominant dispersionmechanism is
(A) Inter modai Dispersion (B) Intra modal Dispersion
(C) Modal Delay (D) FrequencyDistribution
2 13
6. Inter modal dispersionoccurringin a large amount in multi-modestep index
fiber results in
(A) Propagationof the Fiber (B) Attenuationof Waves
(C) Pulse Broadeningat Output (D) Compressionthrough Fiber

7 dispersionarises due to the variations of the refractive index 2 1,3

of the core materialas a function of wave length

(A) Material (B) Waveguide
(C) Polarization (D) Scattering

Paye 1 of 4
27ME5,6-19ECOtO7T

Downloaded by Sam Cruise ([email protected])

 lOMoARcPSD|22950144

2 1.3
8. The macroscopic bending losses show an exponential increase due to 18. The sensitivityofa PIN HBT photo receiver is proportionalto
in radius of curvature (B is Bit Error Rate)
(A}) Increase (B) Decrease (A) B (B) B?
(C) No Change (D) Stability (C) B (D) B
1 1 3
9. Stimulatedemission by recombinationof injected carriers is done in 19. The Mach Zehnder interferometeris a modulatorswitchingdevice based on 1 6 12

(A) SemiconductorInjectionLaser (B) Gas Laser

(C) Chemical Laser (D) Dye Laser (A) AcustoOptic Effect (B) Magneto Optic Effect
(C) Electro Optic Effect (D) Incandescence
10. A device that convertslight energy in to electricalenergy is calleda 3 1.2
(A) LED (B) Photo Detector 20. Monolithicintegrationcan be achieved in configuration 5 13
(C) Optical Source (D) Laser (A) Verticalor Horizontal (B) Vertical and Horizonta!
11 is defined as the ratio of the photons emitted from the LED to 1 3 1.2
(C) Circularor Planar (D) Circular and Pianar
the number of internallygenerated photons PART - B (5 x 4 = 20 Marks)
(A) External Quantum Efficiency (B) Internal Quantum Efficiency Answer ANY FIVE Questions
Marks BL CO PO

(C) Emission Efficiency (D) Photo Efficiency 21. A step index fiber has an acceptance angle of 20° in air and a relative 4 3 12

refractiveindex differenceof 3%. Estimate the NumericalAperature(NA)

12. The conditionin which the simulatedemission willexceed absorpiiononly 1 3 1,2 and the critical angle at the Core Cladding Interface.
if the populationof excited states is greater than that of the ground states
known as 1
3 2 1,3
22. For silicathe fictive temperatureTr is 1400K, the isothermalcompensability
(A) PopulationInversion (B) StimulatedEmission B, is 6.8 x 10:12 cm'/dyn= 6.8 x 10-'m²Nand the photo elastic coefficient
(C) Pumping (D) Photonic Eficiency loss at a 1.3um wavelengthwhere x =1.450.
13. in a Raman - Nath modulator,the acusto optic grating is 1 4 ,3 4 2
(A) So thin that it behaves almost (B) So thick that it behaves like 23. State the phenomenonof Cathodo luminescence.Mention the applications 3 1,2
3
of Cathodo luminescence.
like a Plane Transmission Bragg's Crystal Grating
Grating 24. Elaborateon the types of Optical Amplifiers. 4 4 4 12
(C) Analogous to a concave (D) Quite ComplicatedGrating
Rowlands Grading 25. Comment on the reasons on why it is difficult to integrate Lasers with 4 2 6 1,2
associated Electrons
14. The change in the refractive index of a medium due to presence of sound 4 1.3

wave is called as
26. Using Snell's Law define the relationshipat interfacebetween two different 2 I 1.2
(A) Photo EmissiveEffect (B) AcousticOptic Effect media.
(C) Electro Optic Effect (D) Magneto Optic Efect
27. i. Describe on the Principle of Photo detection. 2 2
15. Pumping is done in order to achieve 1 4 1.3

(A) Steady State Condition (B) PopulationInversion ii. Differentiatebetween PIN and Avalanche Photodiode. 2 2 3 1.2
(C) Equilibrium (D) Photon Emission 3
PART -C(5x 12 =60 Marks)
16 are operated at a current beyond normal lasingthreshold | 4 1.3 Answer ALL Questions Marks RI CO PO

current 28. a. i. Derive the expressionfor acceptangleangle and Numerical aperature. 10 3 1 1.2

(A) Semiconductoroptical (B) Erbium Doped Fiber Amplifier

Amplifier ii. Differentiatebetween step index and graded index fibers. 2 2 1 1,2
(C) Raman Fiber Amplifier (D) BrillouinFiber Amplifier (OR)
b. i. With the aid of neatdiagram, elaborateon the Fiber OpticalCommunication 10 2 1 12
17. In system.
integrationdiscretedevices on separate functional blocks are 6 1,2
connected using electronicor optical interconnects
(A) Mornclithic (B) Hybrid ii. When aray of light moves from medium i to medium 2 with different 2 2

(C) Massive refractiveindices,explain the conditionfor total internal reflection.

(D) Large Scale Array
Pase 2 of 4
27MES,6-18ECO107T Page 3 of 4 27MF3.6-18ECO1077

Downloaded by Sam Cruise ([email protected])

 lOMoARcPSD|22950144

mathematical
relevant
dispersion with
in detail about material
-9. a. . Explain
2 13
2 2
express1Ons.
extrinsicabsorptionlosses.
betwcenintrinsicand
. Differentiate 10 2 2 13
losses.
(OR) Microbending
bendingand
diagrams,explainaboutMacro 2
2 2 1,3
b. i. Wihneat dispersion?
velocity? What is its impacton 3 1.2
ii. What is group
10 2
Avalanche
and Operationof
3
the Construction
sketches,Explain
30. a. i. With neat
Photodiode 2 2 3 1.2
3
incandescence
Luminescenceand
ii. Differentiatebetween
3 1.2
(OR) of Light ,3
quantum efficiencyand Optical Power
Derive the expressionfor
b. i. Emitting Diode (LED). 2 3 1.2
3

ü. Define: Responsivity. 12 2 4 3
working of
with neat diagrams, the construction and
31. a Explain in detail Optic Modulator.
LongitudinalElectro
4 1.3
(OR) 12 2

diagrams, explain in about Acousto Optic Modulator.

detail
b. With neat 12 2 5 1.2

with neat diagram, the operationofMach ZehnderInterferometer.

32. a. Expiain
(OR) Circuit. 12 2 5 1.3

of an OptoelectronicTransmitter
b. Describethe fabricationprocess

Page 4 of4 27MF5,6-18ECO107T

Downloaded by Sam Cruise ([email protected])

 lOMoARcPSD|22950144

Reg. No.
RAROoR662o o65
B.Tech. DEGREE EXAMINATION,MAY 2023
Sixth Semester

18CSC305J - ARTIFICIAL INTELLIGENCE

(For the candidatesadmittedduringthe academicyear 2018-2019to 2021-2022)
Note:
) Part - A should be answered in OMR sheet within first 40 minutes and OMR sheet should be handed
over to hall invigilatorat the cnd of 40h minute.
(i) Part -B&Part-C should be answered in answer booklet.
Time: 3 hours Max. Marks: 100

Marks BL CO PO
PART - A (20 × 1 = 20 Marks)
Answer ALL Questions 1
1. Which tool is used for describing the judgment or common sense part of
problem solving?
(A) Heuristic (B) Critical
(C) Value based (D) Analytical
1 1 1
2. Artificialintelligenceis defined as
(A) Making machine intelligent (B) Putting more memory to
computer
(C) Programming with your (D) Transferring your intelligence
intelligence into computers
1 1 1
3. A problem solvingapproachworks well for
(A) 8-puzzleproblem (B) 8-queenproblem
(C) Finding a optimal path from a (D) Robot navigation
given source to a destination
2 1
4. Turing test is used to check
(A) The intelligentof humans (B) Intelligentof machines
(C) Both intelligentof humans and (D) It can't check intelligence but
machines check the speed
5. A problem is a search space defined by oneof these 1 I 2 1

(A) Initial state (B) Last state

(C) Intermediatestate (D) Final state
2 2 2
6. Which of the followinguses a priorityqueue?
(A) Uniformedsearch (B) Depth first search
(C) Best first search (D) Iterativedeepening
1 2 2 1
7. Heuristicis used in
(A) Informedsearch (B) Uninformedsearch
(C) Brute force (D) Blind search
2 2 2
8. The time complexityfor breadth first search is
(A) O(6) (B) O (bd)
(C) 0 (d) (D) O(n)
Page 1 of4 30MF6-18CSC305J

Downloaded by Sam Cruise ([email protected])

 lOMoARcPSD|22950144

9. Conradictionin propositionallogic representsthe truth value of

sentence? compound 1 3
20. In fuzzy expert system conversion to crisp value is done by
(A) Always true 1 5 12
(B) Always false (A) Inferencemechanism (B) Composition
(C) Some are true, some are false
(D) Can't be infered (C) Fuzzification (D) Defuzzification
10. A productionrule consist of
1 1 3 1
(A) Set of rules (B) PART-B (5 x 4 = 20 Marks)
Sequence of steps
(C) Set of rules and sequence of (D) Arbitrary Answer ANY FIVE Questions
Marks BL Co Po

representation to
steps problem 4 1 1 2
21. Defne ihe ri!i wing
11. What will happen if two literals are identical? i) Al
1 1 3 1
(A) Remains the same (B) Added as three (ii) State space search problem
(C) Reduced to one (D) One variable less 4 I ! 1
22. What is turing test?
12. Which is not a property of representation of knowledge? 1 1 3 1 4 2 3 1
(A) Representationverification (B) Representationadequacy 23. What is forward chaining? Explain it with an example.
(C) Inferentialadequacy (D) Inferentialefficiency 4 2 4 1
24. Discuss about the learning.Give some examples.
13. What are the major aspects which combines AI planning problem? 1 1 4 1 2 2 2
(A) Search dLogic (B) Logic dKnowiedge based 25. State the differences between BFS and DFS.
4

system 4 3 5 2
(C) FOL dLogic (D) Knowledgebased system 26. Illustratehow knowledgeis representedin fuzzy based expert system.
4 2 3
14. Unsupervisedlearning is one in which 2 4 12 27. How is predicate logic helpful in knowledge representationand state the
(A) Input outputpairs given (B) Learning is done automatically syntax of first order predicatelogic?
(C) Leaming is done in semi (D) Only inputs are given
supervisedmanner Marks BI CO P0
PART - C(5x12 =60 Maris)
15. One of the main challengesof NLP is 2 4 1 Answer ALL Questions
(A) Handling ambiguity of (B) Handlingtokenization 1 2
sentences 28. a. Construct productionrules for the situationwhere there are two jugs, one 12 3

(C) HandlingPOS-tagging (D) Linguistics with capacityof 4 liters and the other with capacityof 3 liters,both without
measurements.Imagine there is a pump that can be used to draw water in
16. How many types of quantifiersare availablein AI? 1 4 1 any quantityfind 2 liters of water in a 4 liters jug.
(A) 6 (B) 2
(C) 3 (OR)
(D) 4 b. There are three missionaries and 3 cannibals stand on the left bank of a 12 3 1 2

river. A boat is available which c¡n take maximum 2. At any point of time
17. Pruning is used for 1 1 5 12
number of missionariesshould not be out numbered by cannibals which is
(A) Traverse the tree from 1left to (B) Top down search fatal. Make a planto safely take all to the right bank. Represent the above
right problemby state space search problem.
(C) Reduce the search space (D) Botom up search Representinitial state
1
(ii) Goal state operators
18. The main component of the expert systems are 5 12
(iii) Action plan
(A) Inferenceengine (B) Knowiedgebase (iv) Find the entire solution
(C) Inference engine and (D) Meta data
knowledgebase 29. a.1. state A algorithmand explain it with appropriateexample. 2 2 4

19. General games involve 1

(A) Single agent (B) Multi agent ii. Discuss about hil climbing problem. 4 2 2

(C) Neither single-agentnor multi- (D) Only single-agent and multi

agent agent (OR)
2ue 2 of 4
30MF6-18CSC305J Page 3 of 4 30MF6-18CSC305J
Downloaded by Sam Cruise ([email protected])
 lOMoARcPSD|22950144

b.i. Find the solution for the following 8-puzzleproblem using best first search 3 2 4
Initial state Goalstate
2 3 3
4 6
7S 8 7

6
i. State the simulated anncaling algorithm. Explain how it is used in 2 2

optimizationproblems.

30. a. Explain the knowledge representation using predicate and propositional

12 2 3 4

logic with a unificationalgorithm.

(OR)
b.i. What is semantic network? Explain it with an example. 6 2 3 2

i. RepresentWUMBUS word problem in FOL.

6 2 3 2

12 3
31. a. Write short notes on the followingconceptswith an example. 4

() Reinforcementlearning
(ii) Adaptivelearning
(i1) Multi agent based learning
(iv) Ensemble learning
(OR)
12 2 4
b. Describethe componentsof planningin detail.
12 2 5 1
32. a. Describethe frame-basedexpert systems functioningprinciplesin detail.
(OR) 4 5 2
12
b. Write about alpha-betapruningprocedurewith an example.

Page 4 of 4 30MF6-18CSC305J
Downloaded by Sam Cruise ([email protected])
 lOMoARcPSD|22950144

Reg. No.

B.Tech. DEGREE EXAMINATION,MAY 2023

Sixth & Seventh Semester

18CSE481T-APPLIED MACHINE
(For the candidatesadmittedduring the academic yearLEARNING
Note: 2018-2019to 2021-2022)
() Part - A should be answered in OMR sheet within first 40 minutes
and OMR sheet should be handed
over to hall invigilatorat the end of 40h minute.
(i) Part - B & Part - Cshould be answered in answer
booklet.
Time: 3 hours
Max. Marks: 100

PART-A (20 x 1 =20 Marks) Marks BI CO PO

Answer ALL Questions

1. A group of related documents against which information retrieval 1 1 1 1

employed is called
(A) Text database (B) Corpus
(C) Index collection (D) Repository
2. is used to identify the category of each word in a sentence. 1 1 1
(A) POS tagging (B) Stemming
(C) Tokenization (D) Lemmatization
3. In filtrationprocess what happens 1

(A) Special charactersare removed (B) Stop words are removed

(C) Stemmingis performed (D) Bottom up parser is applied
4. Identity the correct order for preprocessingin NLP. 1 1

(A) Tokenization ’ Stemming ’ (B) Lemmatization’ Tokenization

Lemmatization ’ Stemming
(C) Stemming ’ Tokenization ’ (D) Tokenization- Lemmatization
Lemmatization ’ Stemming

5. means understandingvoice by the computer and performing any 1 2 2 2

required task.
(A) Speech translation (B) Speakeridentification
(C) Voice recognition (D) Automatic speech recognition
1 2 2 2
6. Which is not a type of speech recognitionmodel?
(A) Speakerindependent (B) Speakerdependent
(C) Speaker isolation (D) Speakeradaptive
1 2 2
7. is a method used to convert analog signals to digital signals.
(A) Pulse code modulation (B) Sampling
(C) Coding (D) Quantization

Page 1 of 4 31MF6&7-18CSE481T

Downloaded by Sam Cruise ([email protected])

 lOMoARcPSD|22950144

8. allows us topredict a sequence of unknownvariables froma set 1 1 2 2 18. With respect to PCA, we assume that the signal of interestis 1 1 4 3

of observed variables. (A) Along the direction with the (B) Along the direction with the
(A) Gaussianmixture model (B) Naive Bayes largest average smallestvariance
(C) Hidden Markov Model (D) Bayes theorem (C) Along the direction with the (D) Along the direction of one of
largestvariance the naïve base vectors
1 2 3 2
9 is used to understandhow the value of some variable changes over 2 4 3
time. 19. When do we need kernel PCA?
(A) Graph analysis (B) Time series analysis (A) When we need to capture non- (B) When there are kernels in the
(C) Linear regression (D) Logisticregression linearity in the data data
reduce the
10. Which is not a component of time series data? 1 2 3 2
(C) When any two features are (D) When we need to
highly correlated number of dimensions to m/2,
(A) Trend (B) Seasonality where m is the original number
(C) Complexity (D) Cyclicity of features
1 2 4 3
11. Which is not a method to convert non-stationary time series data into 1 2 3 2 20. The face recognitionsystem is based on
stationarytime series data? (A) Strong artificial intelligence (B) Weak artificial intelligence
(A) Translation (B) Detrending approach approach
(C) Differencing (D) Transformation (C) Cogitiveartificialintelligence (D) Applied artificial intelligence
approach approach
12. Wheat crops badly damaged on account of rains is 2 3 2
Marks BL CO PO
(A) Cyclical movement (B) Random movement PART -B(5 x4= 20 Marks)
(C) Seculartrend (D) Seasonal movement Answer ANY FIVE Questions
3 1
13. is an image processingtechniquefor finding the boundariesof an 1 1 4 2 21. Define term frequency and inverse document frequency. Find the term
object in the given image. frequency and inverse documentfrequency for the following example.
(A) Pattern recognition (B) Edge detection Sentence l: Good Boy
(C) Feature extraction (D) Image morphology Sentence 2: Good Girl
Sentence 3: Boy Girl Good
2 4
14. If the histogram of same images with different contrast, are different, then 4 3
what is the relationbetween the histogramequalizedimages? 22. What are the provisions available for text cleaning? How it can be done
(A) They look visually different (B) They look visually similar to using the availablepython libraries?
from one another one another
4 1 2 2
(C) They look like just the input (D) There is no relationbetween the 23. List the applicationsof speech recognition.
inage histogramequalizedimages 3 2 2

2 4 3
24. How music can be synthesizedusing availablepythonlibraries?
15. Which of thefollowingfact is the true for an image? 4 3 3
(A) An image is the multiplication (B) It is the subtraction of the 25. Compare stationaryand non-stationarytimeseries data.
of the illumination and illumination component from 4 3 4 4

reflectancecomponent the reflectancecomponent 26. Explain how images are stored in the computer.
(C) It is the addition of (D) It is the subtraction of the 4 2 4 4
illumination and reflectance reflectancecomponentfrom the 27. What are the needs of biometricsin face recognition?
component illuminationcomponent Marks RI CO PO
PART-C(5x 12= 60 Marks)
2 3
16. What is defined by the total number of pixels within the region? Answer ALL Questions
(A) Brightness (B) Intensity 12 4 1

(C) Perimeter (D) Area 28.a. Explain the importanceof "Bag of words" in NLP. Apply "Bag of words"
for the following sentences.
1 2 4 3
17. PCA is a technique for Sentence 1: "Welcometo Great learning,now start learning"
(A) Data augmentation (B) Dimensionalityreduction Sentence2: "Learningis a good practice"
(C) Variancenormalization (D) Decomposition
(OR)
31MF6&7-18CSE481T Page 3 of4 31MF6&7-18CSE481T
Page 2 of 4

Downloaded by Sam Cruise ([email protected])

 lOMoARcPSD|22950144

b. Define the role of #hashtags, how twitter trends works? What are the 12 4 1

technical aspects behind #hashtags or keyword trending? Write a detail

analysis for any #hashtag.
29. a. Explain with python code how audio signals can be generated with custom 12 4 2 2

parameters.

(OR)
12 4 2 2
b. Describehow a speech recognizercan be build using python code.
12 4 3 3
30. a. Explain how to extractstatisticsform time series data using python code.
(OR)
b. Explain Hidden Markov model for sequential data usingpythoncode. 12 4 3 3

12 3 4 4
31. a. Elaboratethe ways to extractthe edges from an image.

(OR) 4
12 3 4
b. Discuss Scale InvariantFeatureTransform(SIFT) in detail.
12 3 4 4
32. a. Explainbuildinga face detectorusing Haar cascade.
(OR)
3 4 4
b. Discuss in detail about principal component analysis (PCA) in face 12
recognition.

Page 4 of 4 31MF6&7-18CSE481T

Downloaded by Sam Cruise ([email protected])

 lOMoARcPSD|22950144

Reg. No. O65

B.Tech. DEGREE EXAMINATION,JUNE 2023

Fifth & Sixth Semester

18CSE479T- STATISTICALMACHINE LEARNING

2021-2022)
(For the candidatesadmittedduring the academicyear 2018-2019to
Note:
and OMR sheet should be handed over
) Part - A should be answered in OMR sheet within first 40 minutes
to hall invigilatorat the end of 40th minute.
(1 Part - B &Part -C should be answered in answer booklet.
Max. Marks: 100
Time: 3 hours
Marks BL CO PO

PART -A (20 x 1= 20 Marks)

Answer ALL Questions 1 2
1. Identifywhich of the followingis an unsupervisedtask? hand-written digits
(A) Predicting if a new edible item (B) Grouping of
is sweet or spicy based on the from their image
information of the ingredients,
their quantities, and level for
many other similardishes Predictingif a patienthas diabetes
(C) Predictingif a user would like to (D)
listen to newly released song or or not based on historicalmnedical
not based on historical data records

1 3 1 1
ANOVA?
2. Which one of the followingis used for
(A) Tocompare 2 or more gTOups (B) To determine the correlation
between two variable
a normal distribution
(C) To evaluate overall significant (D) To test
of aregressionmodel
1 2
find the best fit line for data in
3. Which of the following methods do we use to
linearregression? (B) Least square erTor
(A) Maximumlikelihood
(C) Logarithmicloss (D) R-squarevalue
1 2 1 1

4. Choose the valid categoricalvariablegiven below

(A) Gender (B) Object
(C) Number (D) Alphabet
Y=B + B,X+E,(B,B,) 1 2 2 2
5. In the mathematicalequation of linear regress ion
refers to
(A) (slope,X-intercept) (B) (Y-intercept,slope)
(C) (Slope, Y-intercept) (D) (X-intercept,slope)
2 2 2
6. What's the penalty term for the Lasso regression?
(A) The absolute sum of the (B) The square of the magnitude of
coefficients the coefficients
the coefficients
(C) The square root of the (D) The sum of
magnitudeof the coefficients 01JF5&6-18CSE479T
Page 1 of 4

Downloaded by Sam Cruise ([email protected])

 lOMoARcPSD|22950144

7. Identifywhich of the followingreduce over fitting in decisiontrees. 2 2 2

(A) Enforce a maximum number of (B) Enforce a minimumn depth for the 18. Which of the following will be Manhattan distance between the two data 1 2 5
samplesin leafnodes tree and pruning
points A(8, 3) and B(4, 3)?
(C) Make sure cach leaf node is one (D) Enforce a maximum depth for the (A) 1 (B) 4
pure class tree and pruning (C) 2 (D) 8
1 1 2
8. When handling training data with missing attribute values, handling 2 2 2 19. ldentify the reason, for feature scaling is an importantstep before applying the
continuousattributesand improvingcomputationaleficiencyare issues in k-mean algorithm.
(A) Unsupervisedlearning (B) Decisiontree learming (A) In distance calculation, it will (B) You always, gee the same
(C) Supervisedlearning (D) Predictivelearning give the same weight for all clusters, if you use of don't use
1
features feature scaling
2 2 3
9. In Bayes, theorem class conditionalprobabilityis called as (C) InManhatan, distance it is an (D) In distance calculation, it will
(A) Evidence (B) Likelihood important step, but in Euclidean give differentweight
(C) Prior (D) Posterior distance,it is not
1 1 3 3 1 5 5
10. The Euclidean distance between two a set of numerical attribute is called as 20. Which of the followingalgorithmis most sensitiveto outliers?
1

(A) K-means clusteringalgorithm (B) K-mediansclusteringalgorithm

(A) Closeness (B) Validationdata (C) K-medoidsclusteringalgorithm (D) K-mods clusteringalgorithm
(C) Error rate (D) Testing data
Marks BL Co PO
1 PART - B(5x4 =20 Marks)
11. In optimalvalue of 'K' trainingpart consideredas 3 3
AnswerANY FIVE Questions
(A) P (B) P-1
(C) 1-P (D) 1 4 4 1 1
21. Examine the importance of machine learningalgorithm.
1 1 3 3 2 1 1
12. Which one of the followingis used in dimensionalityreductionalgorithm? 22. ComparedescriptionstatisticsVs inferential statistics.
4

(A) Linearregression (B) Logisticregression

2 2 2
(C) Principal Component Analysis (D) Support Vector Machine (SVM) 23. Short note on maximum likelihoodestimate(MLE) and entropy.
4

(PCA) 4 3 2 2
2 4 4
24. Apply precision,recall to any model and F1 score.
13. Which ofthe followingalgorithmcan be slow to run on very large dataset? 4 1 3 3
(A) Linear Discriminant Analysis (B) PCM 25. State Bayes theorem with conditional probability.
(LDA) 4 2 4 4
(C) Linear regression (D) Gradient descent 26. List out theadvantagesof supportvectormachine.
1 2 4 4 4 5 5 5
14. SVM's are less effectivewhen 27. Justify the importantof dimensionalityreductionin machinelearning.
(A) The data is noisy and contain (B) The data is linearly separable Marks BL CO PO0
overlappingpoints PART -C (5x 12 = 60 Marks)
(C) The data is clean and ready to (D) The data is non-linearlyseparable AnswerALL Questions
used
1 5 5 12 4 1 1
15. ANN create its own organization of information it receives during learning 28. a. A clinical psychologisthas run between subjects experiment comparing two
time is called treatments for depression and client centered therapy against a control
(A) Adaptivelearning (B) Self-organization condition, subjects were randomly assigned to the experimental condition.
(C) What-ifanalysis (D) Supervisedlearning After 12 weeks, the subject depression scores were measured using CESD
1 1 4 4
depressionscale. The data are summarizedas follows
16. In ANN information flow in unidirectional. n mean Sd
(A) Backward propagation (B) Feed forward Control 40 21.4 4.5
(C) Neural network (D) Hidden layer
CBT 40 16.9 5.5
2 5 5
17. PCA reduces the dimensionby findinga few CCT 40 19.1 5.8
(A) Pentagonallinearcombination (B) Hexagonallinear combination
(C) Octagonallinear combination (D) Orthogonallinear combination Use one-wayANOVA with x =.01 for the test. Fcritical value =4.79.
01JF5&6-18CSE479T
Page 2 of 4 01.JFS&6-18CSE479T Page 3 of4
Downloaded by Sam Cruise ([email protected])
 lOMoARcPSD|22950144

(OR) 12 4
b. Examine in detail about supervisedmachine lcarning algorithm.
12 2 2
29.a. Explain when to use ridge and larso regressionwith suitable example.

(OR) 12 5 2 2
b. Find root mean square error for the following data sample using lincar
regressionmodel. y = mx +c, where, =0.43 and c =0.72.
X 1 2 3 4 5
y 2 1.5 2.75 3.25
12 4 3 3
30. a. The following table gives the dataset about stolen vehicle. Using Naive Bayes
classifier classify the new data {Color = Red, Type = SUV, Origin =
Domestic),concludethe result.
Color Type Origin Stolen
Red Sports Domestic Yes
Red Sports Domestic No
Red Sports Domestic Yes
Yellow Sports Domestic No
Yellow Sports Imported Yes
Yellow SUV Imported No
Yellow SUV Imported Yes
Yellow SUV Domestic No
Red SUV Imported No
Red Sports Imported Yes.

(OR) 10 4 3 3
b.i. ExplainK-nearestneighbouralgorithmwith an example.
4 3 3

ii. List out the disadvantagesof K-NN.

12 5 4 4
different kernal function with an
31. a. Define support vector machine. Explain
example.
(OR) 12 3 4 4
artificialneural network.
b. Constructand explain with neat diagram of
12 4 5 5
approaches. How it differ from
32. a. Discuss the concepts of clustering
classification?

(OR) 12 3 5
(7, 8).
b. Considerthe two dimensionalpatterns (2, 1),(3, 5), (4,3), (5,6), (6, 7),
Computethe principalcomponentusing PCA algorithm.

01JF5&6-18CSE479T
Page 4 of4

folh
Downloaded by Sam Cruise ([email protected])