Thapar University 2006 M.C.A AI & Applications - Question Paper

Thapar Institute of Engineering & Technology
MCA (3rd Year)
Final Term exam
CA021 (AI & Applications)

THAPAR INSTITUTE OF ENGINEERING & TECHONOLOGY SCHOOL OF MATHEMATICS & COMPUTER APPLICATIONS End Semester Examination-] 5^th December, 2006 AI & Applications(CA-3gl) ,MCA-Vth Semester M.M:36    Time: 3 Hour

Note: Attempt any five questions. First five questions will be checked. Attempt all parts of a question at only one place.

1.    (a) Use binary resolution to prove that there is a green object if we are given

If pushable objects are blue, then nonpushable ones are green.

All objectare either blue or green but not both.

If there is/nonpushable object, then all pushable ones are blue.

ObjectOl is pushable

0bject02 is not pushable    (4.5)

(b) Prove that composition of substitution set is associative .Is it commutative also. Give example to support your answer.    (2.7)

2.    (a) The following rewrite rules can be used to replace the numeral on the lefthand side with string of

numeral on the right.

6 -> 3,3 6 -> 4,2 4->3,1

       Consider the problem of using these rules to transform the

4->2,2 3->2,1 2->1,1

numeral 6 into a string of 1 .Illustrate how algorithm AO* works by using it to solve this problem. Assume that the cost of k-connector is k units and that the value of h function at node labeled by numeral 1 is zero and at node labeled by {n\n * 1 is n.    (4.5)

(b) Write an algorithm to perform breath first search of a problem graph. Make sure your algorithm works properly when single node is generated at more than one level.    (2.7)

3.    (a) How you will choose better heuristic for finding minimal path among two A* heuristic h_x,h₂

.Elaborate with an example.    (2.0)

(b) consider the sliding puzzle problem consists of two red tiles and two green tiles and an empty space. The puzzle has two legal moves A tile may move into an adjacent empty location.

A tile can move over an another tile into an empty location.

Apply best first search with data driven approach using heuristic function as the number of tiles out of place. The starting configuration is | r | g | [ r | g I

The goal configuration is    | g i o I I R I r I

4. (a)Convert the facts of following Knowledge Base into conceptual graphs.

(VX)food{x) like(john, X)    (VX)(yY\eat(X, Y) a not _ kiUedby(X)] -> food{Y)

S n |

(VX )eai(bill, X) -> eat (sue, X)    eat(bill, peanut) a not _ killedby(peanut)

Infer from the derived graphs the fact peanut is a food. What operations on graphs you needed to infer this fact.

(b) Differentiate between script and frame structure.    (2.2)

5. (a) Both case grammar and conceptual dependency provide representation of sentences in which

noun phrase and described in termed of their semantic relationships to the verb. In what ways the two approaches similar. In what ways they are different. Is one a more general version of other. As an example compare the representation of the sentence    (3.0)

John break the window with a hammer.

(P.T.O)

Derive an ATN for the given parse tree.

Prove DeMorgans Laws for fuzzy sets analytically.

Assume that the variables x, y, and z all take on values in the interval [0,5], and that the following membership functions and rules are defined: low(t) = 1 - (t/5 ) high(t) = t / 5

rule 1: if x is low and y is low then z is low rule 2: if x is low and y is high then z is high rule 3: if x is high and y is low then z is high rule 4: if x is high and y is high then z is low Using product inferencing what should be value of z for x=4.2 and y=2.8 .Use Mean of Maxima for defuzzification.    (3.2)

Consider the fuzzy set A and B defined on interval [0,6] of real number, by the membership

functionp~(x) = //₅(x) = 2~^x .Determine the expression for (i)L4 -5) x + \

(ii) (2 n A^c) (iii) (a X b)    (1.5)

(a)    What is meant by topology of artificial neural network . Give a few basic topological structures of artificial neural network?    (2.2)

(b)    Determine the weights of network with 4 inputs and 2 outputs using perceptron learning law with

/(x) =- for the following input-output pairs, {(ill 00], [l l]), [(0110),(00)]5 where 77 = 0.01 ar^A \ + e~^x

.(use initial weights to compute weights in both the cases). (3.0)

0 1 -1 -1 -3 1 0 1 1 -1 -1 1 0 3 1 -1 1 3 0 1 -3 -1 1 1 0

Assuming that the

threshold and input of each unit to be zero ,Compute the energy at the state [-11111] (2.0)

Attachment: thapar-university-2006-m-c-a--26297-1.pdf

Earning:   Approval pending.