Nane moshomt Shana

Roto. NO yESKCXO43
Section G52 technigua
n Data mining- Concapts crd
Subject Ds (5thSen)
Bramch AssIGNMENTd
Section-A

Spatial.clata minirgoles to the boocew extoac

An sptial elatienshihs ard mte
onp Knooledgethot Cee. not specilîcctl sored
Seastirg pattexm yon the othee hand teo
Ima. spatal datebase fextaCtion

the occLÛbomce o/ an ewent

KroDledoe about
algonthms hae
aminimumn hesnod
clustee gthe o t i e s aeethe
lon pott*
daa
a

LLOm I tha otliers

trot aee not clusteseed
em aclstexgthy coeeiax auiayom athe
Cleste

AnoSone the c i e
onitemsiSimble m
asocit t o
based.
Qos antecodomt
tooltes Im the
Iniszules hawe forexamble
nd ome item n the CDnsegemt
aCustOMe buy diabed g them the
theule am exQmple
to also buy beee' 19
ace liRy
associcetionule.
gwntset it
Ans A Set iten kcalle
theesheld VvOlue tor
minimwm
Sctlspia a coidemoestremgth 6 easw
Subot ond cenilomceiSuþboxt eees
by Subbot und
 latbet et ule tu
thue In þuctice mesa givem
nsMaket basRet dotainngte
Dettee undertamdiroaustone putchaing
pattems Gt invelves amalasing laegeclot s
StCh as buxchase histo toveal pOau
gouing
ru traqvt hctten mining has am .data
atpicationa rngrg pom pattem
Cleaming to þatten-bed classtcattoncles
teringolotliex cr excoptiem Oralyis
Ans Actame þottean mining fes to the þroces
ciscouextg Cornplon crd non -tivial pottemby
aJLOiatier or xelatiomshib8 mlarge amdl
Complex datasetz for eg- Segluemtia. Pttern
mining
Quae
AoData mining ls based n þattem identiHc
aton om dataowlable Im amsystemieb
mining lz based on pattem idomtification
web tdol Toolo sd by Cata mining a e
machine laaxning algortindTools wed of
uoeb minrg are Page Ramt,Saypacte lng
bivcive clusteing wost t thu obet
aggeomeative clsteingstod by censic nge
big
escing al the clatapojrt imto a ton 1nto
clster ama later cm spitting
Smallel heterogeeo w clus tes CetínDuly
 dala
SuuloS o u auio video mining hlas major
suEveillace macine discDuexy gedua
in,enteete°nmemt ad oot
ART
AR B
Quall metho clus
Ans ieaarchial clustesirg Isa hat Ceeates a
texamalysis mdata rnig
hieaxchial epresemttion he clsters Im
adataset The method tota by treting each
point asa seeeate clos¿r cndthen
dato
itexatively conbines theclest entil a sko
bþing Coterion ls xechedThe es uet hieax
asechial clustering lB 2e iReBbuctue
adendogram wtuch llusttes thu
Called hecliesteg.
hicachal relationshb ameg

Ans Aeocition ule ming algorithim Buchas

pioxi oFp-guooth ace LAed tofdteusnk
aOciCtionlesThe
dtern 2et andgente
Eealgorithim oak by iteotivelyt geratig
he
Canolidote itenmsetE amd ouming2 p t
that donot metthe minimm
ttheshobl: onCe the eqent tem ets a e
loumd uciattan ules cam be geneated

no toasactiah that cetin

itom Bet and Cmteclet atloe
 A grosth algeithirn olata iw

hLaxge datauets Jt qemeseates feequert j te

4eicienty by Censtucting anFP- te
datr ChuctuNe. tct
Hatset
Dottems
and, mining tro t jor reeuoirg

cLbbrouch ealated. to k-mon

ottoning adata EetIto k gous arcuter
gnk-medi'ods clustoing each cluste ls leatre
ented by Gne thodata boirt l m uclster
Those pointa cenameol ceuste madiod.
for eq- youhaue Eutormo dato mgt
wart to ceott bet o Simlia customes a o
ran taget eoch goub uoith dyHeent be
mazkorg
Ans This omat lz knoum as Vetical data form
by espeing thetial dato fott byscam
ning the. cloto set onceThe suptot cownt
G amtemset lz simbt tre lorrgth the
T|DS2t the iternet.

fART C
 ttometg maxinum
(supbt=3) CcDY
(suppt-) B,DY (suppont
) = fB0Y
(suppert4) Brcy (Support
-4)
(Suppest
4) A,
cy(support=3) LA,B,Ccyy
ersetSamne (subþot
) A,Dy (suppot-3)
)
itemsetz clOsest
minns set-oth item freqemnt
I2/3,5) C:y(ocCue
n
lyay3) (occus
lm A:3 ten’ each Supbot
of
5
D Tyo0 4
Byc Ay Ta0o
CD B, A, T200
A,B,CH
atOm Licst 2
Any
TD SN
5 30, ,20, Ca-f10,
12. j=ay4,2y, Thxeyae
|4
2 B5
d atho
CT
cleste
C| Ce 6
balded
to
Point Cluster
Doßya]-J/-a) Datroirt
Ue
istale Eleoam
C2=3 c=u htialcoesto
 feqont
clDsed ttemset
maxmuum itemset Bcy
Arw st of itemy
A,Bc
AjCDt
Tyoo ABit
T5o0 B,CE
T600
&et calculate, the &ubbct Per each tterm
Saport A=/%>66-6
C= 5/623.33/
D=R/6)33:337:
E4)6=) 6662
Now, idamtl tha esguent temms based m u
66-67),c(Susbot =z3.33 )
A (Supot=
E (Supnt -66-67) that meets Cr excoed the
A, c)houesppot vales
30/ thucesheld.
Guut
An The agglomerative clusterirg lsthe mstcom
non tybe ot hieachal clusteng wsed to qroub
cbiecta n clustexs baed am tne sinmiloty.Nesting ).
4ts olso Rnoon as AGiNEs (agglornanative
Tre algorthim stats by treotg each otiect
singloton clste AROmeathe cleetering
sa ncutoUs dlomains guch a i o
i8 widole useo
B0clalsieres and lmage Lemomttonwitin
heatarchial stuctiwees
to diCOve
d ata and to ldarntty amigul clesterA o
doto poirta