Scribd
Upload
30 views3 pages

Department of Computer Software Engineering: N-W.F.P. University of Engineering & Technology, Mardan Campus, Pakistan

The document discusses an artificial intelligence course final term paper. It includes questions about Prolog code to draw a family tree, possible variable values in queries, whether queries over a database will succeed or fail, comparing human and neural network nervous systems, the McCulloch-Pitts neuron model, applications of fuzzy logic controllers, and using the ID3 algorithm over a medical diagnosis training set.

Uploaded by

Muhammad Ishaq
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
30 views3 pages

Department of Computer Software Engineering: N-W.F.P. University of Engineering & Technology, Mardan Campus, Pakistan

The document discusses an artificial intelligence course final term paper. It includes questions about Prolog code to draw a family tree, possible variable values in queries, whether queries over a database will succeed or fail, comparing human and neural network nervous systems, the McCulloch-Pitts neuron model, applications of fuzzy logic controllers, and using the ID3 algorithm over a medical diagnosis training set.

Uploaded by

Muhammad Ishaq
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
You are on page 1/ 3

Department of Computer Software Engineering

N-W.F.P. University of Engineering & Technology,


Mardan Cam!s, Pa"istan
Artificial Intelligence-I
Final-Term Paper (Fall 2010)
atc!" #II Seme$ter" #
th

Total %ar&$" #$ Time Allowe'" % ho!rs
(ote" Attempt all )ue$tion$*
+ , 1" 1-
a)
arent&"im,holly'.
arent&margaret,"im'.
arent&margaret,"ent'.
arent&esther,margaret'.
arent&her(ert,margaret'.
arent&her(ert,)ean'.
ancestor&*,+' ,- arent&*,+'.
ancestor&*,+' ,-
arent&-,+',
ancestor&*,-'.
.ra/ the res!lting family tree from the a(ove rolog code.
0nd /rite all the ossi(le val!es of varia(les in the 1!eries
(elo/.
i' 2- arent&P, )ean'.
ii' 2- arent&margaret, *', arent&*, holly'.
iii' 2- arent&margaret, C'.
iv' 2- arent&P, "im', arent&3P, P'.
v' 2- arent&esther, C',
arent&C, 3C',
arent&3C, 33C'.
vi' 2- ancestor&0, holly'.
('
(ig&(ear'.
(ig&elehant'.
small&cat'.
(ro/n&(ear'.
(lac"&cat'.
gray&elehant'.
dar"&-' ,-
(lac"&-'.
Write all the ossi(le val!es of varia(les in the 1!eries (elo/.
i) 2- dar"&*', (ig&*'.
ii) 2- (ig&*', dar"&*'.
1
c)
3iven the data(ase (elo/, st!dy the 1!eries !nderneath it.
4ndicate /hether yo! thin" a artic!lar 1!ery /ill s!cceed or fail
(y ans/er yes or no.
li"es&)ohn,mary'.
li"es&)ohn,trains'.
li"es&eter,fast5cars'.
li"es&Person6,Person%',-
ho((y&Person6,7o((y',
ho((y&Person%,7o((y'.
ho((y&)ohn,trainsotting'.
ho((y&tim,sailing'.
ho((y&helen,trainsotting'.
ho((y&simon,sailing'.
i' .- li&e$(/o!n0train$)*
ii' .- li&e$(!elen0/o!n)*
iii' .- li&e$(tim0!elen)*
iv' .- li&e$(/o!n0!elen)*
+ , 2 1
a' Comare and contrast (et/een h!man nervo!s system and
0NN.
+ , 2 10
a' .iagrammatically ill!strate and disc!ss McC!lloch-Pitts
Ne!ron Model.
(' 0NN o()ective in general is (alance (et/een memori8ation
and generali8ation. Why2 E9lain (riefly /ith the hel of an
e9amle.
+, - 10
a' Mention some of the alications of f!88y logic controller in
real time /orld.
(' What are the se1!ences of stes ta"en in f!88y inference
rocess2 E9lain (riefly.
+ , 3 10
:!ose /e have some domain "no/ledge; "eeing a simlistic
aroach, /e say that three attri(!tes are necessary and s!fficient to
declare a erson as healthy or sic". These three attri(!tes are,
Temerat!re &T', <lood Press!re &<P' and 7eart =ate &7='. 0ny
atient coming into the hosital can have three val!es for T, <P and
7=, 7igh &7', Normal &N' and >o/ &>'. <ased on these val!es, the
erson is to (e classified as :ic" &:?'. :? is a <oolean concet, :? @ 6
means the erson is sic" and :? @ $ means erson is healthy. :o the
concet to (e learnt (y the system is of :ic".
a' .etermine the si8e of the instance and concet sace.
(' =ed!ce the hyothesis sace !sing lang!age (ias 0N..
2
c' For the follo/ing training set .. 0ly F4N.-: algorithm.
D T P 45 S6
*6 N 7 7 6
*% 7 N > 6
*A N N N $
*B 7 > N 6
3

You might also like