EnggTree.

com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com
UNIT-I
SUPERVISE D LEARNING1

REGRESeSloN

Ragression find Cor relabisns bebw eom

dopendent and ndapeudant Vanables Tf the desired

vaT jable
CUtput Consists of ene or mor Conbinvs

then tha baake allad eabosSTGN

ContinouS
Kagossten algortm halp prodict
Martat trends, ueathor
Vamiabe Sodh as hsuee prias
pattens oi a d as pricos ect

Data points

inea easesie

Indepen dent Vamatla

ofstabistial
t a seb
Ko esion QNalysis
mdbhods uged for Dha esbmabiouu relationshi
wodo 0 fbbure
bobuan Vamaba Omd foo modelling the
relabionshp betuceo

Downloaded from EnggTree.com

 EnggTree.com
LINEA RE6IRESSION MoDELS3
34abistical mebhod tHat
t at
regossin is
a
¥ ingar bebwen
ard sdy velaticnAp
bo &ummavi2e
allsus
qpambidabive Vaviables .

600 Cmtmoous

madol
Jinar egsass im
*The objecti vos
f a

s bo tin a rdabtons hip betuween ho inppt

VaT Iable
Vanata8 omd a
tavget
s Tegarad as bhe
One
O e variabla, denced y
deperdoot Vaiabo
Jespons, OUttome o

*The othor denojed x Cs regosed

VOApONBe PRadcfor explanabony 07 indapendant

p32duco
Vornabe
Conbinwous
Vamable
modlos Pedic a
KagTession poedio

sales made bha day 0T

Such bhe Thas we
imagine
kets
Tempaature of a
Poin
tRa ue

wrtb do focinimg
i a ine daba point
OmotRe
to add
omt
hawe change
existing
existi9
heed bo
to fit it
bot
modol
Jabla (cogsa), pdictiony
clasci-fiatsr predicts abep
modol Conbinoos- Voled fncbims classi-fiotiomr is Considarod
be Soperviuecd lornirq
Downloaded from EnggTree.com
 EnggTree.com
REGRESSIaN UNE
LE ves tho aworage Tolabiorship bebuneon

bioo Vaiats mabhomebial fot m

FoD buo0 aiatas x avd y, bhare Ase aluwaug

too na of
Reasessim Jina ofX on y gves tha bast estimate fr tha
Vale of x fs an a tby
where
x - intecopt
b sope of tRe n e
Depandant Vawiabia

Y Tndapendant Vavbie

Karessim ino y on X
tha bestestimate tha
LE givas
Specific gihe Valas ox
Vae o y

Y= atbx

4 = y -Indesapt
b o p e of tRo lina
y Depondont Voble

inda pendon VaraKa

UsIn Sqpore mefod we a a able dbo

COnehnct a bost
Fttig bo Saber duagram potns Ond
Downloaded
a T e gfrom
essEnggTree.com
i e eWatien n 1he m
Bhon forwolate
 EnggTree.com

Bas tebM
Wo

fCx,w
W
Inpot
vetbor

a+bbX

y 4bCx - )
art a d suenca2
aalye is Da bka
0 of daba.
Cnes bD pattern
of fittima Straiaht
Ihe Vanabe o ntevst
In Jinaar Tagesion madd
o thor Vavables
oThed Vamablas
dapendlent Vamaba) is prediched ao k

Chaa eapabien If y denotes

usirg
CCndependent Vorahas using
are
Varale and X X k
the dependen is b h o
assUmptidn
the
Varabe thon fs
n dapendent daba Sample
in a
imet
at
the Valu of Y the naa 2Ajabin
dolerminod

lonstanis and
Whore the
ha betas ame

epsilons are tndeaperden and identialy dishmbotod

Downloaded from EnggTree.com
 Momal andm Vamables wh
EnggTree.com zero

As dn spldt Point The esoy betwoen t a

prodicled Vale md dhha achial Valoas s

Sapanea bo 9et a "Som

SUm o Sauared Eorors (SSE)
Tho Sp.lit point eDS a coce tho Vamalos ase Cornpand
dba Vanabla poin ielding Tho Jouat SSE IS

Chosen ha
ooot nda s upt point Thi's pro03S

IS
ecunsavily Con tinued

Advamtages
naar Taroiciom mbda
Toulmm a
Than mebhoo Such as
Usualy hueh fasler

hwsal nobwoyks
mo tmple ahd
YarsesSIon
wwodols
inaar
Jo plemenb
Yeawire m
LE AST S60ARE
The wetho d of lamsb Swanas ts abrob

Ostimabm parTOmeters Dy winimizirg Da Swared

dtscrepametng bebuweon obseve dalba bm b o
romd Cnd
petted Val n dohe ofhes

Citerien Stateg 1at

The aast quae
h SOM
Som o Downloaded
Sawane from
o EnggTree.com
s minimuom
 EnggTree.com
The Anst Sapase Soobiona yields yt) whose
butpots to be in TRa
Som bo4 but do QnUTe

Tcw Co,
How dao ine based bn tha
o Soch

daba polns obsonvad? Suppase a

imagimy ine of
ya+bx bebuan Jina
1fo J n
olisBaa
magine a Verti
a e daba point E - Y-ECY)

doviabt othe data

Tho eso i
Thon
Jno.
wagivan Jina, Teresien
povn and b
bast Vaues of
a
Uohab is bha
Such
thab
minimio0 Sum o
a b

FCy)atbX

Deviatien doos hot Jae good popes Hog os mpotatlom

Tho USe
DownloadedSwae o clouiation ?
from EnggTree.com
 Lot us ek EnggTree.com (
a b thab Cn munim2e ohe Som
Sqpored douiabiews. s Mabtod is Collod loaat
Sanave8
Laast Sqpare mebod winimi bha Sm¢

Capars oferos Such a a 4 b are Calla oost

Sajpore atimakre. ie atimahs of pavalmokeps d 4 B

poocess } etting peroumer astimaas (q.a46)

1callod @timobio Least 9apame mduad
Ctiwabon watbhod ofodenaay laast SaparesoLs)
Disadwant asps of lenst apare
Lack robustness oubuers
Vnsuitable fod leasB Sauare
Cortatn dabasebs
olasifitabion Machera
Machene
to
Correspon ds
Deu'sion bondan
Solotio

MvLTIPLE RERESSIDN :
predtc bha
btha
Usel to
Rogreesie amalsis
Vogpbns03 ovm a Se of
Value Ohe or movm
also be usadr estimabe
CE COm
prodi chos. Preddchrs
assocbbion tbobuuan ttha
the Jinaan
be tonbtnoous 0
o

Tasponse Pre dicht am

Categoicod a
wixtuse of
orDownloaded botth
from EnggTree.com
 EnggTree.com vamably affec
S dhe wwlbipla mdopendont
Dhe asponve Va noble, Then the anakysis Coll for a
wodel diffesent tat used ha Sind
Predichod Vawablo n asitabien whoo MOno a
ndepondont facod (Variabla) affecs tha oxoma
oaprotS a wwlbiple ragossion wodel is Used
use
is efemad a3 wlbipla unar Jeqosiem
This
wodel
wwlbivaiate aast Squoe fitting.

Y
Ohera is bha yowdom e
CATe
Bi,io
Co-efficen
Dirferouce bebuueen Simpla Raaression and MOUtiple
Pqsession
Sumplo Raqsessien
Mwtbiplo Raqsussisn
Owe dapendant Vaiala
y One dependont Vaade YpreAicle
prodickcl from e e irdapedortoo set
of independevt
a

Variado Variates Cxi, Xr)

One pusjow Cooltheient
On regvessien Coeffiejent fo ach
mdependent vaiade

Downloaded from EnggTree.com

 BA YESIAN UNEAR REGRESSJON
EnggTree.com

Bayesiam JUtnca neomesGion aWuosa Usebl machamsm

dal nefficient. daka, 05 pood disriboted data
T a0ous Uces put O prior on bhe
Coefhuiornts nd bho Oise 90 thab

absonca of daba the pnos Com aka dwr .A prio

ts a dishibubvon n paramabed.
Could Fip tha CDin an infinik
wber of bimas, inaing Chs bia uoauld be aa
douo of Jonaq wbers »

Houoever whab if we old enly filp

Wou We
the
he oin a
uomd fov of bimeg?
three haadds
u
tab a co s bau'sed we sauo

thoo ips, m event hab appens 6ha oot of

these
ean timeas oith onbi ased Loina? Thay Oerttt
a toin bias ofP
daba, inlemmna
bestimate
# Baye San webihods allnos ue
bo
wodo pasameter bo onsoeb modofoeosts and

Cendoct wode Comper8ons Pauestam Jaanning algonths

Com oleombe expliit obabilihs os hupotthases
Use O Simple idea
bdea
Bayesiam classifie*s
thab tha braiia daboa me u ilmed bo lelale
obseaad
obsenad Downloaded from EnggTree.com
porbobidly ach dase based eu fetue
 EnggTree.com
Valwes
When he Bayesa clasi fies is Uc fo ndasified
0
daba, 't Use Jho Obsone probabilhia bo predict tAa

most
Jikay
ilay alass fo the naw eabues

Eadh obsee tuining RRampla Com naewoh

decraase OT incraase bha 2estinatad probabilihy hat aa
that

potficeis S Core c

Bayes iom wbhods Com acLommodale nupothaes

0 Nouo istamcas Con
bhab woke boobabilistic pradictions
be clasipied by Cembining
tta edicti ons af nudlbiple
thair pobabifie
ypofhas os, Weiguted by
Ohore Payegia mabhods puue
m Cass
ve
ntcocfable , tey Cam pr ide a standaray
Compotabiealy
onidh othor

ophima ducisiou waleimg ogaimst

be moasUBed
methods Con
prachica
classifiers
USes f Bayesiam
in eat-basad
caasiHalieu foir finding
Usad
os jone wail iltering
Spom

Medical diagnogig

Nobwork Sewsih such Qs delecting ilega

wbvctiey.

Downloaded from EnggTree.com

Bauic EnggTree.com
pradwou o plemenbi ng Bayaian Ainaa
Ragrssion
Spe priorc s ba woda pammater
Creale a modal wappi'ng a Sraimdm9 hpuls

bo 1fe soumime butputs

Have a Markov chain Mon Carlo MCMC)

dgoorthm drauo mpl foorn tha posteriod dishibtia
for HRe pasameers.

Desant
G1Ta dien t
a t orclor opimzafon
descant
G adcent
minimom of a fmchon
find ocal
algonthm. To a

das canb, eua faks Seps poporti onal to

USing qrodibnt fha fmcbi on at
tha oddent of
he hegabrve
of
tho CUrsen point

G1adiank
dascenb
Populas oVory
beause
2bion oroblems
Scalo Ophimi
hamde
hamdlu black.
bloack box
Imploment, COm
ls easy
ncbiens Ond ech iembiom is choo
chaap
wil 9iva bho slope of
The 9Tadient

Cunve tho X Omd is directhion wiu

pom aM n Creage mcbien s0 w
Downloaded from EnggTree.com
Com EnggTree.com
nma Opposi irectien bo

bha mcbion Value

+= f *)
The A>o C a Swall wwber hab oT T2
algontim t wak &wallumps.

imitai m af
617odient Desant

dascenb is alativey eloo close

Gnodient
minimwm teehnital Sasympbotic dak o

inleio h» momy otRas mobhods .

CODergenhee ts

Corditienad CoTvey pooblams

For poosty
ag
i2gsas a
radient dastont lmgreaai mo
ghoxfast
thogova to the
Cyradient poinsnaas
oiTectien a polnt
muwmom Polnt

Weakt Grradert

Ineremana

wwniw m Ccast

perivafhe ofogt
Downloaded from EnggTree.com
 EnggTree.com
T we ove bow ards haabiva qTadieni or

wwaH so bha gadiens f bhe meble ab btaz

CUmenb pobnt will gve bha Joas minimu
o bhe mcien
when eod wove bouwads a posdive
qradient or bouards ha qsodieut
of bhe mebion
Bhe wront poink, w
o et fha loal mayimom
O he fmbiom
entiYe psdcedoso s Rwwn a Gadinr
asOnt onich also noun as stpeat doSont The

objectve o vcivoa Yediot descenl algorrthm

s to winimize the Cost etto UCi teratiou
Calwlato h first osdor donivative of the

fmcbi en kD Conpole bhe gTodiont O7 elope of The

ie
ove auwa rom the direction of the Giadient
which Mms dope nereased trem the umont ponl

b alpha Hma, Whene alpa is doinad ag leamirg

Rose is a tonmg paramaled n the bpHm2ationv

proas wwch holps bo dectda ta langt of tha

Seps

Downloaded from EnggTree.com

WWoskima Smdiont EnggTree.com
Desuon

.mX+C
Ohee repoen bhe lope of tha Line
ino
amd c epsoaenls bhe ntertaps m Tha -y Ax i3
Axig

Staruira poin
oss

Value of welght
point of onYer ence
Whese the Cost mctiom is at
s miÜmum
The Aope Steapeo ab a Stestinq
Point
Point o6 osbitanH point but enever houd paramdeTS ane

Generaed bhon eleepnass qradooll redoa and at

the Jowest point, 1 approaches bhe loest poin

Ohcch is Called a point of Conveaente

Laaromg Rote TA ic
delinco as bhe Cep Size
takon
o Teach dhe minmwm 07 uoet
povnt. This Is hepi caly
a Cmall vale bab is Olrted o d opclaled bosed bosed
ow e bokowlor D ta cost metim TP to
Co
laasning
Tabe i i TQsuke in
Jarqes Seps but it
also laads Downloaded from EnggTree.com
to isks of torshootung o wunimom
 he ame Jbime O Jouw 2ootg
EnggTree.com rate shows
Ab
bhe wall ep si2es LOnich Compoumuse UUerall eficien
Uesall elticienu

ut ives bha advowaae of woe preus) ou

Types of radinb Desonb

Pateh Grradient Descent

2.tochastfc qrodtenb Destant
Mini Bath Strodienb Deacent
3

Bautch Grodient Descent

It is tD Hnd t o es6os fot each

point he aumimg Let and opdae a madol

afier evalovatng all scnem exampes. T+ s

Rwouwnn as
traimimg epoch
epoch.

Stvchaa He qadient DesCant

StochasHe qadient Dascarn is a tYpe

OF. qacient Degcont Thab n g tma tsorsm exoumpa

per iteratin . Tt is mona

icient bo orage
dotasets
Mini batch 6vadient Dezont
Combiroh
peston je The
Mini ba4Ch radient
choshic
both botChv
qadient doscont and gto
chtaseks
gradtonE dostont. T+ divids tha hramni n
IntD Small batch siz0sfrom bhon
Downloaded pesfoams The update
EnggTree.com
 EnggTree.com
UNEAR CLASSIACATON MODELS

A
A olassificbien algortAm a b makes

its claacifiaben basa brvLncas predich mcbi e n

Cenbindna a sek werns with bho
featne Vectos.
dacisie bapd on
bageon
4dls cassiiaaia

of Tha Charaderistic
a Jnao Combinati
Vaun af
noas clogsifier monga into
Lmagio eharacerisHa a dakino a
all the

Pastcol ar cloas

Pisciminabi ve mcion
is de
inea Discriminant Analgsie CLDA)
dimensimalihy raducbien
fechmques
USec
Che
Cowmimenly
bo Sloe loxe t too
h o too
thou
-

- dosS
dois

in m a c h i n e aaandng
is Rueon as Nodma
dossificatior poo blams T+
piecrmùnamt Analysis CNDA) or Disesiminamt Funchic

Anausis(DEA)
inao Discminanb amalysis C nO
0
huost popolar
Ppolar dimarsisvelihy Teducbieu techniqw used
for Soponvisad elaasifaon poocblems in wachine

eancng.T is also Cns ido rec a Pe-poO3Sing

step bd wodeDownloaded from EnggTree.com

Wna diforerc ML ave appliatiav
 EnggTree.com
inear Disciminant amalusIS is Osed a
dimansi evalihy edocbte technique in Wachia kasnng

0Sing Ohich w Cam as il tsomsfunm a2-D adA

Bhe dimonsi enal plan
-D9sap nto a

Lobs ons des au exampla Ohere we hawe

bwo olosse n a -p plane Jawing an X-yaxis

and aa to
wn
clossi fam efficien y As ue kae
alroad Seen n dha above example fab LDA erabe
drauo a
Stsaiht ina tol Com Complady
Separate he hoo classas of daba poins
Here LDA USeg OM axic to Crale
O
O heuw aris
m -Y
by Sapambirg am using a
Saught
uine amd Projech mga dala Ond he0 oxis
Henca w Con max imiz2 tto separation
Dotweeh tRoge dagses andedwce 2 - D plana Inho
one dimens itrnal A

Downloaded from EnggTree.com

 EnggTree.com ha buoo
patlen ass iabion For eg, I
claases wrth wabipla aaues and naod Jro eparate
sing a
them officienty when e lassihy am
Sivgo featue, Then it way shaw overappima

OO0 O 00 o)o O0O

Overlappina
To Ovos Cone cbo Oes loppiag &se io The

classifitobiom proue, u ws.b Cn caase tha nDmbe

of faahmas ag
bo clasciy boo
Let's assoma Jae
boo Ses of daka poinfs
dillesant clo&ses
houim
Ih a 2 - dimerSimad plano
O

O
O
O
O

Downloaded from EnggTree.com

X
 To C2ate O EnggTree.com
ax is nean Dismmirant

Avalysis USeR oQ0ooi Citeia

Tt marimies dictatoe bobuoen ms

Joha
buoo clasos

TE minimLs TRo Vaviana oIthin

ha
individwa ologs
Ln wosds Can SayTkot tho nau
othes
Tha
Oxxis wil increae tho geparation bohueen
Tom onto t h
data Points hoo cassea and plot

hew ax ig.

LOGISTIc eEeRESSION
hogistic TesessIon ic On oftfo nost popolas
which Coma Undo tha
Machina aTwma olaprthns,
Sopewis leanimA techntaue
Tt uSe fus psedieting The

dapendant Vomabe usima a

nven
Cet Indapendant Varsiada

Thevere ho OotComo must be a

dis csete Vale Tt Con be erthor

y os No O , ne or floe ectbot In Slead

the adYoalue a9 04I t qivs The
of givin Downloaded from EnggTree.com
ivos the

probabi lisic Valuws Whoch s leho een 0q 1

 EnggTree.com
¥ hogisHc TRapOgcim s s a fos souimg
Pod Bom
the claseif cabin

Tn Jgistic yagosion, ingkezad of tinga

2pasei on Jina, we ft an s skapad Logistic
Unction, Which pedics boo maximom VolnO)

Tn logistic eqsagsiDY imtead of ftfiq

a1sion ina, we fan skaped logistic
fun ction Whech psediek
bhoo maxIOM

Valua (o Cud )
waehine laarmine
a Signifiant

alapvrthm because abilihy to pvuida

bha
Contimous
elaasi data usig
pobobilitie and
noLD

OMd discsele dobasets

obsenNatio-
TH m be used to claasihy tfo

hpes o daka aud cam asily

Usim dilfesent
variablos usad a
dolenmihe
wost fective
claseicotien
the
s-Cowe

0 5
e r skold Valwe
Y-0.8
Downloaded from EnggTree.com 7
LOGISTIC FONCTION EnggTree.com

The eig moid fmcbien 0 hogistic fonction

Use to wap da prodicted Valuas Jto pobabilitieg
TE
mae any. a Valun into
Vawe anotfor
wihin
Tame of o adI
Te Value Agisic eessin wst

bebuen awd whtch Camuot go beyond

this imit so ib bms a Cwove Jibe

dapendenb Vamablo swoud be Caepno

The

n hatuna
Tha Indapendont Vamiata ghoold hot raue

wwllbi -CoUinaanhy

Loqistic Raqoasion Equatt on

Y bo + bi + b, 72 + bgZg+ +bnn

Ln
dbgishe Pareasio Con be bebueen

and y 8o o this Jots divida tha aboue

aobion
yC
o r y-o and infiny fr -

Downloaded from EnggTree.com

 EnggTree.com
Bot vaad mge behuean-CinfimA to +nfinihy

than ake Joqaithnm of bhe apation it wiu

becona

bot bz,+b22,+ ba 734bnn

Type o Logjstic a9soasiun
Binomial
n nts
rogqsossion toe Co e ny
buoo possiba ypes tta dapendanb Vam'ablos
Such oos pess os fail e

Moltinomtal
sogsi) Tho2 Can be 3 o
wose po&sibla wodosed hpas af fha dopendau
Vannatla Soch as Cok o dog or
aap
0rdinal: In dois sasesiom Thoae Com be 2 od
wone Pogsibaa 0'sdaed hpes odapendant Vamakes

Such a J a , anadiom, 0s hah" .

GtENARAT*VE. ModEL
Genesobive wodola a a
cloas of
Statisttal wodola that Rnesae auo doba
nstan@g
Downloaded from EnggTree.com
 Those EnggTree.com (
wwodolg Ome USed iv wsopenvise maciyo
caswng Jo perfonm tasks Such aa
pobabi0iy amol
Jteelihood gtinaiou odolling dota Pounls
poins amd
and
distingoisha beboem clas&es usimg Thaze
pobabiliieg

ienerobive wodolg o on tha Baeg

DheOTevw fird tha oin probabi0ih
*Slemesabive wodo dascrbe dhoo daba is

herabed vira pocbabiluiskic modals.

hoy psediet PCyla), tta obabiliy of ygien x,
Calcolatona ta PCR1), tha prmbab fxand
NAVIE BAYES CLASIFIER
agorTRm 8 a opesvise
Navia Bayes
tonch 1s
based n Bas The cwom
aarsniR algositfw
cloasftcobien poubloms8
paubloms
wavsed for Caluma
0
n ext classfiaban
wano Usa
dimonsjon l tpaimtmg
Tob in clodos high
dakago
ho ast
holps in boildr
Thas Coun
or modole
alaorithm
machine aasnig

make qoice Downloaded

poedifi onfrom EnggTree.com
 EnggTree.com
*4 s a probobilistie clossifes tohuch moan

pseucs on Te oasis of fha obcbil of

obled
Soma Populas ekampo4 Naive Baye
Algoothm a spam HatimSentimartal avalysia
Omd closifyi avticlag

T+ is callad Nouve Decause o + Qssumal Thof

Te OCeUseNe
The ocuserne of a Corfain eatne iS indapeudont of

the OCumeo o other cahmes.ch as if fha

fosit is idanified on tF bages of olos, Sape amd

Taste,thon
thon red. SPpherical and Susce faoit is etoqriaa
a& aM appla

ete2 ach cahna emdividoay Conto botes

appla withowt
idanhhy i t i s an

doperdi1 On ch dfhes

Bayes The orew

L is als0 now Bayes Pole

Gu which is used to dojeomno

Bayes
ofa, hupotfais Loifh pood
tho Poobabi
Conolti oral
dapends on The
paobabilty. Downloaded from EnggTree.com
 13
The foToa for EnggTree.com
Bayes hesrem is given as

PCAlB p(Ble) pA)

P (B)
wheve
PCA B is postenior poo babily, probabiluky of
hypothais A bho obsenvad ewnt

pCela)s a iholikod prbabili probabiliy of the

owidenta iven hat the pobob tha

hupothasis is e
psnbabily of
pobabliy,
DCAis a poior
befre
evidena2
pobabiiy
o
af
ypotResis pobabilhy,
Margima
PC is
evidonee
eheRtive
nd LSmMivahve.
Dferan bebwean
Model
Discimivative mode
1evarabie oda
dutcomi nate behuan
his
TE anesata
denens eird of
doka
doka iostacos

S1onoabve
wodol DQVolwg wakas psactichlas
distmbotion O bage on londihonal
Cound The
tD dohvm a prombiui
aa dousget poobsbii and is eithor
for a qiven example sed foY clossifiadion
Downloaded from EnggTree.com
Taoossiom
 EnggTree.com Disommirative wodols
x enerabie modol Caphne
Caphwo dthe Corditigral
bhe sint probeubil
plx) ¢
toe pobabiity ylx
PCk) t
abels
A distmminarve wadal
includos
A eneTabiva madol *A
tho queshi
gn

dishiboion a t o daba
vmoas
the
hw ualH ofhabher a qiveh
Hsel, and lells yau instamce is ioy ad
a ven examnpo Is
how
Jjt telles you
wkaly a Jobla is apP
to the instana
eneotive modals ane uged
machna Tus moda
Cr supendi sed *
taak
Leaing to pestom particolasy ued
Sucn ikolihod
as prbabili hy a
fos Sopeise
O9 fivvathem

Eg: Gussaina, Nowe Bas E9 ogishc Paoscssitn

SuppoeT UECTDR2 MACHNE

* Soppovt Machivas CM ate

webhods tokidh
Sex Sopeov ise Jasning
coon fe da dodeset and ad fo cloxff ation.

SUM alascifhies doaved

eAcdictial laasnina Tkamyl Vapk ard
Downloaded from EnggTree.com
ChorVonehkis.
 (1
EnggTree.com
Smpug Speakima COm hine of an SUM mode

sepseeentina examp0s Points in Spaa

mapped so t o ach of exampos of To eparale

osses are dividied b aap Thaf is wide

paasibla
Exampa oo Bod Pecisitn Bovndories

SyM Ae psimaiy hoo-Class doasiftess

usith The distinch aharacterisht ot tha0 au'
D fin the optima hperpbna Swch ttat the
eypecte qehesalt2oablen eoso 1S nimimi2ed

Instead
o direct mimimizgma ha
Qmpyriansle Galculate the taimima data
SUMs pertm Stnca risk
minimi2atiso to achweve
ood nerali' 2atien

The empinica ris is tha auexage

Joss of an astimator o7ainie o
Set data
olTauon
Om P

jdaa of Tisk uzobio

ot vl MmeUne
the
oeima for Pertomaue of
is disk,b t
to atdl Seasch
The estimato Tob r
minimirL s es
P
dtstrbuher-
Downloaded from EnggTree.com
 EnggTree.com
age Masgin eoasifie
sa
wachine looning
Sma
base
iS Vachs
a
is ko Pind
method wore Classea t o t
betneen
docisievn bovdo Cy poinf inn
is maximally au
paiwmg dolaa

D
cals 2

Clas1

Two cass poble

dox2
O

class A

Bo deuision bovnda of SUM

of 10uiimg examplas, exh markd
0 belev
J Dne boo classeg an SVM
aloprth bbilcs mwlel bhat pre dects whaTher a
Downloaded from EnggTree.com
hew
exampls fall ino one clos bo h othor
 EnggTree.com
Beavsee (Ne dont dictn bo tion p
Unskad minm20 empinal nse bwr a
oOutmima
oemimg
data s et dvaion from p.|hes enesad gormma
Tachnie is Calla empncal Tick minini2abron

Gjoee Deaisio ond a

Th dousion Domda houlo be @s fas

auday bhe daka of boTk cbases as

Poseiba
ha data points le e close
be
f comdas
bomday, Te cloasifter
ieab
a
mae esros
Corsistent but is more

On naus nsfancos y tha dusmbution

Enee Closs itiers Jha
pefere
maximi2e The imal distauna of daka points

the Sepaatos

expecte

Snost ContidonCa :Epska

Rise

Empirica rieb
Downloaded from EnggTree.com
Smal oroparihy Set of ch.Large
 EnggTree.com

cass 2

Cags 2

G100d dociion bodont

ha
ab behooo daba points and
Classifes bomdany
d nmom
*Tha MaTa
distewmce ha deusien
Sampla
bomdosy

Maratn o ha epaasas 3

distona beueeM Coppot Veckos.

Marmin C) lwll

Maxiwa warqu dassiftes ts a classifie in

F ftob mximi23 Tke

wangin
The is good accordig to
Maximizima
jntditi o and PAC thaco
dmpltog TkatDownloaded from EnggTree.com
only Suppot VechoS matHer otkor atning
examPB Ocne ignaable
 EnggTree.com
Pope6ties SVM

Use Simaa whtch Sobdiv idas

Pepano
Space into al pacas ne ohidh
is copied by Classs and 1o ottos b
Clas 2
h e maximiz2 Tho agin douision
Optimiaaim
bamda Osing qadsatic

tind optima kyperplana

techntaea, ohch

Abity
andle Jar feahse Spatas

Ovestittioa Com be
Confollod
by soft
appprach
Mavai
SVM Appliatons
Loccassy in any
SUM bean Osed
pooblea
P
Text(md hypexlext) tzosatom

coasiftatim
ma

Pioinoomata poteim clascifition ntes

loas fiation
Haw wtlen charsaoteY
Detemiratiou CPAM ewai
Downloaded from EnggTree.com
 EnggTree.com
imtatons q SM

I gensiive tu oise

t a in Tho
Te
bianee Qitabion SUM

Choio kesna
Another immtati ens s Speed ad sza
daeign foo olticlass UM
*Tle opR

is alsoo a esease h asea

Coseifier

SOFT MAR6tINN

ya dimona Dna pookoma

eassifitin, Somebima 5he data
ommen n

Seasabla.
Ceas
n e r a Case they a

hot Ome even i We mahb prefera

They a2 ,

e
That bettes sepasadas the bolk o
data ohila norimg a Peo weiro olse
dowmens
ohas ttasounum Set is hot
JDneas oparaBo 9 ack Varabas Cowm be
addad t alloo misclassi piatien ef ditftiaut or
wois @rame eeulbina waogdrv aallad sct
A Sof Ma aOlOs f e o Vanatas
Downloaded
o o e s ind tko wovin from EnggTree.com
Ober TKa penplama
 DECISION TREE EnggTree.com
A odeuiston twe Simpla apoasentcdja-
r casifyirg examplos Pecsion d agnimg is
Ohe weet Succaastol fechmquas for Sopnvise
Closcificatien loasnu
docisiom alyei's a dorision a
e tro visally a explucitby opsalen
dortsieus ad doisien waking As bfo a m a
poas, it O tDae-Llka woda daulsina
loat ooes
Bad nod
A
a
abel, dokeminad
of saiming exumpls Yaching
bwalony ote

Each }eomene nada is a o s i on

odt aCcovdWmg o
eatud boonchas

Omguess

Deusien is a wabtuod fos

aupperimahmga distoete valaad tas mche
Tha Jaarnad meho depasenteby dausio

A oama2 desion oee Com also be2

de- depoasendeed O Set of - R a n aolas
Downloaded from EnggTree.com
 EnggTree.com of Ro M t
Veeielo lamimg
mabhoda
widoly se amd pacttcal
Endochi ve ienen
s bab us- to hoisy dada ad Copabla

ouis e t i ve expooas ion

laasi
eesl e 0amemga mebhod athaa A

Complalety eppoaesiva potRasis

DECIsIoN TREE REPRESENTSTnON

Boild a desSion bo olaesi xamplam

Magabie
instanco OF Concop+
postie

Jaaf node Jras assotfeel

ach

oth i4 a attibote.
wi th
hode has aecob t e d
ach Joa
io cloasitation oT -) os ascoctcded with
ncda
Eaceh athmboe
pocsbo
Vales o e
of the
m whtch Tho arC
at tho nodo

dinecte

domoleg ates on
emaul noda

Ottoboe .Bomch poageutg

au oOf come o
Downloaded from EnggTree.com or ClaacS
the test leo hoda epreeeus cloas abolbdistibwtieu
 A EnggTree.com
desion a Loo chard
Shce, wee cach hoda donotes a tet bn an
Cttriboe Valun., each brandh progaus am oofona
The has awd tdoe oanos depousen cdoaces o?
class dutshibotins
DE CISION TREE
ALGCRITHM
9enovete detciou roe oom Tka tralming
topla of date parhtion D

wpot
Daa pastinon CO)
Attbute List
Atribue gelection matiad
ALGroR ITHM
Craate a noda CN

pla in D aTe al ame closs tRon

RetoTn nodo Cn) as a laa hda labead lth tha
tho
Class C
>T attribue st S mphy Ton Tetum N as a
Teum
o hodo abelo otn fha
waoity class in D
Appl attibote Seto chim mathod CD attnboe uS) to
ind 4ho
beetSplcttira CrHeitn.
Labe Nodo N oith
plittiog attnbote
Downloaded from EnggTree.com
 EnggTree.com
T Sptti g attnbo 1 discnafe Valnd
Cwd Wo ltiway
Splis allavd
Then ottnbuk cl >atobye st plittimg attobue
Fos (each svtuome i osplirtting Coherion)
Let D, be the set of data oplas in D Clatclyirg
butuome
D is emphy the attach a
Jaat Jabelad with
wadon class in D bo Nodo NS

>Else attach e node sehrad by b anarate

Deexei c tae CD, attibute wstto Noda N

End offor loop

Retm N

Dewsin agrosoHon Covists of hwo phases one

s see2 Conewchhon prontD
To ree Conschen phoae all bhe raintog
examplog Ore at ke doot.patitien oxanplas
Ters Kdy
based on Celected athibotes

n taa
prnina
7phase, The idontifiaHnd
bemova) of bsomehes at oflocf noice er
Out liers
Downloaded from EnggTree.com
 EnggTree.com
Advoniaaps
Rolas Simplo and to wdershemd

Deusion toea Com homdla bote nomina

nomenal atmbules
Deusim ave apable f homolemg dataek
Thod may awe
aue
esss

has dataek tat

Capeble of omdl
wa Jhane wusi na Vales

Dectsin Cons idemad o

ebe
Wonposame tic wathod

Deusion ane Sel explanboy

Disod Vamtaags'
Mos he algpithma euire thar Ra

faTae otthboje oil awe nly discoee Vaues

Some po ble os2 dipicols to sole bs.

XOR.
Deusin sc
appo poiae o
sti matiom ase hase gpa is pedtct
Valuz of a Conhnd ooc athi bote

Downloaded from EnggTree.com

 EnggTree.com
RANDOM FORESIS
Komdem foveas S fewwmous Aslem omur

Set o ola dotod baloras to bha Sopersispd

tima to pwoun watod.

LH ay be ed s both cloasification
Omd vession issuesIo ML

s basa otay the Cotapt of

poas Of Cobimina
ensene shoalying that's a

qmd
woltiple cloifies to Solve a Copla pobla
Tke ede!
ehaane
uhamca The eral pe rbsmanle of
a classifier fhat
RCn do m fore2t
timberr on do
i ie
rssse
a
timber on
choi a
Choie
ma F
IhcorposOtes taka f
taes h ONebco
fha ovebta
tRa iven datae+ amd
Omd
SObses of of t R a
fhe preoici vV acusa
to Impove

datase}

Rand oen Forosts Algpvrthm oarg

Ramdor orast Dorks ndboo Sectim

first 1s o Coeape the Damde wocdlad by

Coneiu N etoetion toeas ad Second is

to ma prodcti one o ach oe Csaate

tivst Spqmant

Downloaded from EnggTree.com

 EnggTree.com n
Sele ct Tamdom K statisti c poins om tho

Schoolimg se
2Poild Seloction taas a9souiated uw ith
The elaate
tnfosmotim points
3 choase coida
uariey op N for ebatiom
toees
ascoLiate wh te Selocked inosmatio,
whtch we Wwt to build

4 peat slep ud 2

S For neLo foachoss

Coae +ta paici ons a

ea ch choica bka heo ecoso

fackoss ho fe
a t oins mast peDpla
Calegua
Votos

PA Inatance

LOSA
dee 2
Tclges.

Mapshy Vot)
Downloaded from EnggTree.com
Fira) cp2.S
 EnggTree.com
Appttah'on of Rando forezt+:

Banbmg > Bankem 20ma I noral use

This agonthm
rbheidortificoti on of
clomge
Medicine ? with tha aasistante ts
ce dis os des toack and rishs
o the drsorder ma be decmizd
Land use We m pestetve
Compadabl amd itn Tha ald of
Ts algpnthm
Marbati09 Marbetig Hendancies Con be
eComi2ed b usaoe of s algothm

Advamao 0f Pandorm Forast

Capabe of wawagigage datebcek
ith Jta demansionalihy
enancos acwoac Vho veasicn
Omd resals te bvosqt Ho tovdo

Psadvamtaa
Artevga Celom toest n ba

OSe forr bth closs

YgoaseOragpun
0e3ponaibi t i a i4 isw
Downloaded fromext appooprice
EnggTree.com os
aposdom obtqati ons
 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com

Downloaded from EnggTree.com

 EnggTree.com
UNITV
MACH INE
DESIer N AND ANALYSIS OF
LEARING1 ExPERIMENTS

MACHINE EAANG UFE CYCLE:

wachine leasning L) Wwodal mavagmat

vd the dolieng of highy performin model is ag
inportaut as ste inbiall boild af tha madol by
Checsing igh datosc
The Conapts asond modol retaining,
vmodel vrslorumg, medol daployment avd modol
ae the basia fos Wacuino Jaanina
monitong
operabiong tta holos tta data scence team
deliver models
Jeihy perbmaibg
The VSe of wacive lkaming tas ireaveL
Subeiontaly in enterpise dakeu awalysHa Scenaos

bo ertvact valvcble nsighs fom business

data
Hena it veny mporbant bo bae au
is
moclel,
elosyslem to build bhe maclel, boild
chase bat
Cempute performamce mebics and
perfomig wodel
The model waintenane plauys a Cntical ole
Once ohe mdal is deployed
Downloaded inbo producbin
from EnggTree.com
 waintonno EnggTree.com

ncwdes Wudol op to dao wl

thorve
Cavose of tino

Machne
poCes that Covons nht forn Shnco data
Dha
ldertfttotitn to ocol doolepranb, odal bploant
cd wodol wainteyamte. A.b
Cutire activitos fall tdon bw0 brood oteros
Mdal dondoprnant and ML modal
3uch as ML
bperotions .
has
ilke Mache 'ne Joanng iecp
fulnaing pdaos.
Podrness gual TdenbirfKabivo
ML Paotlorm framg
Data colo cti ondala pepoasig
feabune evqimaariog)
Madel doolop mant (Tracnimy, honimg, evaloahon
Medol doplogmant (Tnfesonee pautietion)
Mudol monionng
Busimaze qual
hane
An organizatim (oneidenig ML Shoobd
Downloaded from EnggTree.com
 EnggTree.com
Valie to be gaiet by Solbiog poblamy.
be abe to woasne bucimogS Vahe

qinst Speetfic bucimas Checties rd

Citeia
ML poo bem famig:
In this phase, tie bocinass
is fscmad podom
wachine daanng probem:
what Ls bhsend ad ohat shold be predicted
Cenoon labo oX
Delesm n
tenet Vanabla)
predict an
perormance and metna most be
optiizd is a key step in ttuis phae
DATA PRcESS IN G:

Training
acurabe ML wodel Tauine
data pooLsi3 to Convert data into a Ugabla
fusmat
Data poosig steps indoda Colleefion
prepasiag data and tahne aginaiog
the procoss cf oreing
ovd
tenshg, ertactirg
selectig vahablas fpfrarn
DEpU MENT r data
MODEL DEVSopMENT:
After a model is tso'ne, ned
evalwvoated and Valiolated we cau
Downloaded from EnggTree.com oeplry the
 EnggTree.com
made Dhto the prodcbior. we can mabo predich os
and infeseress aqimat to modol .
MbDEL DEECopME N1 :
ouolopnads Consids modal
boudng, traming, toning awmd enlovat tn
Meolal building inclodas seotirng a
Pipelne tat aulomatog Tho build ,aim
taqimg Qud prodoeh tov
envom manig

Bueimass
gal
wonitoihg
fanig

Data
procossir9

Machise aarnng i uyee pooas

Downloaded from EnggTree.com
 EnggTree.com

MON) TDRING:
Model moni toring Sen
tor modal is waiwtanng a dosred

detecti Dn
pefraname gh ealy
igitatiou
Gu1DELNES MACHINE LEARNING ExPERIMENTS
AlM OF THE SUDY :
glOhat are bie cbjecbives (eg
asessing bru
erpected enos of an algostm an a partiolad
poobiers, ect)
SELECTTON OF TtE DegpoNSE VAR\ABE:
Ohat should e e
MeasUne (eg esor, preasi on ard
e,
qaliy
CHOI CE OF PACTO2s AND Comdexiy, ec
LEJELS:

ain f tha ShayC tocors

Gre hyperpormoers (hon
-Be algoith is fix and ant o fird bast
hyperpararoBes,if we ane compmng algoit my
Joasming algoitom is a tack)
Choio of eypeomenta deaign.
*Use factonal dasigo unoss Le
tthat Downloaded from EnggTree.com
ttüa fockss do not inferact
 EnggTree.com
Rplicati en omber deperds dolset
Si20 t n be ba pt Small ohan Tha
dataset lorge
* Avoid using Small dabnseks hich loadi
Vaiarea and 1
and
dferencos oil not be sgnifant
eguts not be Conclosive
pertrming fhe epenimant :
*Doings a feo tral ans
Jamdom settimg to check at d is
as orpected, bejose doing the fechosal
expenmant
Al Tulbs shool be
epsdiocblo
Statisticol Qyalysis a tie data : Conclasi
be due to choco
jet shold vot
Conclosion and ecDromandations

trmyauty Conclosi is
hoed fors futkes expe mon tation Thone
tiat Dur Concosion
be orong espectalley i f tio cata is Srall
and noisy ohen DUr expectabibns
Downloaded ane not mat, it is mest
from EnggTree.com
helpol bo invstigae ohy they ane hot
DIASET PPE AON:EnggTree.com
Machine aaring it. abot Jaasrig
Some
popesti of data sot ard aplyng
nous dola.
(ommon 50ctico In rrachira
a
to esnlivato n algont ig to 3pt
dara at
raiming set (ohich ue Jaam data ooere
and Cne all a
tet Those
tætig
Casing
poperties
set, bn ohich

YIh tha troiniang data, data ane

gsian The Jabals. TTh tos data, data label
Unknoon but not iven The tranimg
dataset Con iss of 'ing eXAmpes
ylhe Yeal aim of Sopenv ise arnig
on tost dala Thab iis mot

bnon dosing loartning

valuog or the
po0rnotes at minini2e the las mcha
on ha
trauiming dola ie mot recussarily
the best poli Downloaded from EnggTree.com
 EnggTree.com
The hainkg
the to°nèmg Sam ple
lho tat erospY ig tha expected predichtn
im clopendont tet Sanplo
is

mot
*pobem
a god
trcng
estimaoY for tha tat

* raimnng e Can be Tedeo by mabi

tho hypothesis mone ens itie
Seng itive to tscu'mumg dale
and poo
bot this
enerali2ation
oad tior
Fting
Set of Rxamples used
Tioriming set '
whore
A
the target Value is

Test Set:
to asegs

pertnmanes of a cagsifie
Mever used duing bhe troining
tost s e
pooase
Poruideg Unbaige estimae of tha
ererali atin eoret tha euooadge aboot
# Traiming data ls
1he daba Sovce V ohi ch ue ue to Cnct cassitir
Downloaded from EnggTree.com
 EnggTree.com

dotaset a haimng set is inplo rno nted la

boild p a mudol Lohile a test
les (or validotie) o
is Vaidate ta modol buil
x Data pons in tha rauming seto} o
exolocad fum tha tast ( Val'datim) set
a

set, a
runng Set, Valdatory Set (Sonme pocpla bse
ingtead) in each derotibn, 0r dividd into
in.
Set ,a Val'daiy Set and a tegt st
traung
ach iteration.
Macina daasne'
baialy iy
to prediet tie test data
to create mode)
So wa Tha traunubg duta fit The moda!
avnd teatng ata to Gest it .
The modes generate e to preict
the reullo Unknon oeh s amed as Tho
test cet
CROsS VALIDAT1ON CCU) AND RESAMPLI NGt
gValidation
Uged
bechqe in wachine zamng
o yet the eonr rate of the M ML
model Which Downloaded
Can be from EnggTree.com
onsideved as Qose to the
 EnggTree.com
yate popoati cn
*I he cale Voloma Jore enn'ga
be epagentive of tha populaion yu nay
hot hoa the Validation Aechnia
Machine loaming, wodal hlidatio
olemad Jbo bhe protDS ohore
traimed wodal
dabcusot
Ls Qvalovoded Ubh a
tatig
*The tas tig data seb a separate
portion of the data Set frm aiCh
the Set is desved
the
lhe vain porpose of osig
data et to tet dha narali zatin
iy of a toaulmed
abily odel.

Data set
Traming ostig Holdot mtfkod
Cooss Vadation

Data pesmitng

Downloaded from EnggTree.com

Teatiro
 EnggTree.com
Cross Valudati on a techuqe. for eosvoir
ML
Sobeeks
mads
hrainng
data
Sereal ML

Use css - ldotien to doteck Ontthrg

gnovalie patterm.
In general ML Drol2a dasvong vwodols fronn
daba, wrth the cçm of acheivtng some 2ird of
dasned bohauian ea,pAicton or coscifiaton
But <bhis uerie tosk s broken dacn into
omeyr
of peeal cases. when trraining is dene
that data was nambe an be sed bo. tast the
performamce ol tha laanad mdal on
This tie basic jdaa fo a oole coas
of model ealivatieu athods callad cos
Validatbien

*Typs of CoDss Validatioy mbhods ana hold t.

k-fold and loave- Dvo -oot .
The holdot mathod is the sëmplt kind af
no wo
CoOSs Vaidabion . Tha data cet iS Sepamted
ses, agd tte troining set and he testing dako
Set
*The
fncbieu appoimate tits a facbisu Usiny
Downloaded from EnggTree.com
the' haindng set only
 EnggTree.com
K-ld way bo impsne
over dhe haldat waticd.
dabaseb is divldad into k Sobsets
and the holdot mattuod is epeated k timno,
* Ech time of tie k Subcets is us ag tha
test selt and tthe othar k-l Suhseks a put togttar

to fosm
t Than he
trials is Computeol.
Leae - one -out C0es Validaion is k-fold
Validation abon to i4s ogil ex feseme
oitn k eqel te nombor
o N ta ombor of
of data poins
Set
WWoans thak NSeparate imes
data
the mction apox wak ie tsouned n he
execpt tor oe foint ad predichiY is wacd or
avd a pedi'cibny
Ttat point
k. fold Chos Validation:
k- fold cv is where
given cola set
Split ito a k womber of
Sectiors folds
ohere each old is used as testing et at Som
Downloaded from EnggTree.com
poBnt
 EnggTree.com

,Lets toba the Scanansics o s-fold Choss

lalidabior (K=s) Hene tha daloset 1s Splat Into
5 o ds
tha first inesa cbion the firs+ fold is
Used to teet tie modol and bta Test osed Jo trau
the modal
*Dh ta Seeond
the
iteabhion- 2ud d is used
tattng set Ohde me Yost Sesve as ta
trauming Set This proas As
each fold of tte s fulds Dapeate
as been ased
unbil
the testmg dta sat
k -fold \hlidation, is petormed
per tho follaoing stepe
partition the ogimal huning dota et into
K equal Sobsets,Fach Sobgzt is alled aa fold
plds be naned as A,fe.. k.

gkoep tthe fold fi os Validatior Sot and

Keap al remaining k-i folds in bhe (oS
Validabion trou'mng set.
#Estiate the acuay af tor machihe
Jeaming modd by wragig the qcera cles
Downloaded from EnggTree.com
 EnggTree.com
deied n l tha kases of CousSS Valldatio
ohe k- fold Ooss Valdabion mebhod,all
origionad taunig daka set
Dsd fors bottu trantng ay oel as alidai on.
g Also each enhy is used fo Valiaion vst
Once

Fod 4
lods
od s
t Tiau'ming sel
he acvontage of
watrs Jass aw the this-mathod,
daa qet
is aab the
dividad .
ve data polng gete to be
onto. and gek to be in a in tegt a

Sek
eracty
k - times
iaming
|he Varionce of tne
js educa
The disodvautage of 1e wotnod is ttat
tte raummg wothod
mcthed hasdro be cm Surabch
9ich maans it talas k timos ag mwoh
Downloaded from EnggTree.com
 EnggTree.com

Compojebien o .
4

A vardaut b t s wthod is to
divide te
h daln nto aa tes) and traing
Set K
dueeek
hrag times
Tndependanty checse hew
Jang eoch
set is and do wany tri als Yat oorca
OVeY

Boolshapping:
It is a nethod off ample sese tat
is wch mae
heral. "an cos aidatio
ha idaa is bo use 1o obseed
Sanple to astimate tfa populat un dishbuhon
amplas Cau be drauon fono tha
*Then
csti mated popolatior ad Tha Sampling
dishibotioy o ay ype o estimato an itsel be
stimoded.
boctstrap is ,faxibe and
* The
paur Statistial tool. fat an be used to quautty
ayeroaked wlth afiven estimato
or Satisial lsarng wmetfod
Downloaded from EnggTree.com
 EnggTree.com
*Por example iH ca an ostt mate
of t e Stendad e r of a Coeffiaent oY
Cofidene inleal or tcat Coelhcien+ .

ppose at olsh to Invag t a fixd

Sum of menay
into JOo mara'a oests
thal
yeled ehne of x and y raepecthi valy
ohene Xend y ave uaetitieg
we wil ovest a fnetion, a afracton
mon inin X and oi invegt t
remaining
wish bo oose to minçmize the
Jtotal isk, o Varian, o! Ünes tnenb
¥ n obier ooTds, ue wwan Jo rinimize
Var (ax + (l -a) Y)
Ohe Can show thab te Valwe thab
winimizos is bay,

where
Vor),oyVar (y) and oy= Crv)
Downloaded from EnggTree.com
 But tho Valcs aEnggTree.com
o,0y ad dxy ao onkwon
Car Cormpole. osnatQs )

Aata set
*y vsmg daba
ant thak Contaim,
measUomonts br X and
Can hen
minimbo ur Inenont using
dhe aonta of ouY

To esttmate Oho etndard daviati on o oo ue

epaatad the pO@AS of Simolatio (oo pairad cbseovationg
of xand y esthi matimg d looo fines

DbBainod ipoo esimae for d

ohich
We
thasebyCall di, d2,... o00
FoT the simolatins the
pamles wc

he Vale of d is o6 .
tne
The wane Over all ,o0O egti natog fos d is

rel

Coge to
doviatio of the esti mateg
Downloaded from EnggTree.com
 EnggTree.com

lopo

|000-)

oO83

Vay good idaa of ta

gies
Cumacy f : SEC&)0-o83
o a rondn
Sample fom th Spoabing
pöpütabin, USOold er pect
d approsimatd 0-08

There trso fns ofaf bodstvoping

ohich dife primanly in dhs the population
estimated

NONPARAMETRe B0otS TRAP :

Jhe won parramehic booshrap a Sample

of tha
the. Sana ci2 the data
as s takon Bor
fom
the datba wa Teplocemant
measune 10 Samples, we
Create ha mple by replicatig
Some o tietee Samplas ttat We haue

Sesn and omittig othos

Downloaded from EnggTree.com
alrady
 SEMIPARAMETRIC BootTRAP:
EnggTree.com
(9

the
The
Jem
resaunpling bolatmp Can
that
y The Seniipaametic
the bcotstap agsumas ta
population imcdg otfer temg
Similas ko oha
a Smooted
dsenad Sampla by Soupling fom
version of tte omole h'shogo
Senple
be done.
Vany siapy by frct tabig a Soumpla toith
eplacement fro the Obseed ample ad than
a ddimg noise .

PARAMETPIC Root ST2AD:

paTametic bocsapping assUmes thatte
data Cones fro buouon distribufion rt
UuenUOn paTametes
estimate the parameterg fm
tha daba Jae aud tan'
tthe es tiwated
'shibution to mlae the
Soumplas
MEASURIN
CLASSIFIERPERAURMANCE
*A binaa assifiahion a
ttat agsins ls aa mtad
dgs do a bho bas iS
Downloaded from EnggTree.com
is
dosursphion
 EnggTree.com

The pevbosmana o a bina clasitiey an

be cssessed
by abuating is
tost Set uoith knoon
coit mOLn abols in
br Confsj sn mattx torth ochal Cas93. i Conbingeny
Tows and precte classes in Colomra
MeasoTOs pestornanca hacd to etishy
Severa Ctena!
must
Cohavertiy ophao The
aspect of performarta of interest
most be intotie enogh
becomo toidaly sed,
ome ae. tonickenty reproted by
wide
Vogearchors, Quablimg Cowwoinibg
nabling
Concosù ons o be doauon:
moct be Computabievally
bractoe, bo mabch he apid qrouth in Scal
wodom dala Coocisu.
fol
hay.massbeSumple to epot as
womber
Single homber kor each mobthd - daaet

Combuabiou
Downloaded from EnggTree.com
perfoswemco metics fo binay Classifi cati oy
EnggTree.com

dsignad to (astnad tadoefts bohaan fou

fdametal populabieu quauttie True
false positives, toe vagpti a oud fabe posi tie
*The evalatbi en
vaqpkives
measunes in docifiation
potems dainad Bom a matix witt tha
a
wmbey
Cosecty
for each class, vaed Couhsi
aud inceNoctty Cbsifed
wotnx
The Confvgion watix fr a binay
Closcifialior oldom is Shoon belos
Trve ass prediced cbac
egatie postive
False napbive Tre positive
Negative False positiue

A Cosien matix conbains abaut cal

poadited lassifications dtne by a

classificati System
pesomamea f och yste is
data Matx
eon fusion watx iss also calod ag
Downloaded from EnggTree.com
 EnggTree.com
(
false posiive
Ohich ame
Examples prediced as positive,
dbho class
Fase
vagtie
vagatie :
Exomple8 pedice as hagabie,olose
toe Cos is positi ve
Tieje positves:
Exoms Cassectty psedlced as pertaing
to the positive Cass

Tre hagaties:
Exonples ase coseety.pirediced as
belonings bo ta hegptie class -
measne wat
wost Usedin
tn racHco.
he eralvabi on
is the acunag rate
Acunaey ate Tne nogabie tt Te
Te positive

false vasatis +. Fale pocitive +Taue

\hegabie + e
ACweAcy AND ROc CURVES : psitie

Bimay classifi atbion y acura

mem'e qpanty tthe two ype f. Comect
predietiong Cud Hpes of eoos.
Downloaded from EnggTree.com
 23
EnggTree.com
mees are acwtacy AcC), preesi ohon
Aypial
Tecal feee postive ate, f- wmeane:
* bach mene a diffesont
The predicatie mode!.
Ospect

(Meagnas the fractin Df

Comect predicHons
stua Psaction af
acha positieg among these examglas Ttot ase
predietocd ás positie
Recall moagUnes
positiveg psedited as positie
f-meagUTe haUnic
wean af precision ana /teca.

Operating haractenstia (Roc)

Rocaier
Gapha hae dong baon Used in Sigmal delector
tobo dapic ttha hoadecfe botwean
dit rate aurd fale aym fate8 DVer
chanal
pacent yeas haue eau Cun lhreele
in the
boc qraphe n tte wachie
Downloaded from EnggTree.com
 EnggTree.com
An ppc
plots dve pusitive rate
yaxis false positive ate oh X-axis

Simyle Cou tingeng table Correspords

Simge point 0m an poc plot
The porfromance o a Rano an be
Cussessed
|n a drauoig
plot, e a n as
a plecauoise imear Cme
an

Tthe Cune Stark in (o, ) finishas in

C,) and is wonitoriatly dorraasi
bota ares

Roc Curve the e positive

rate Ds
plo tied
in fmcbieu o
felsoPositive
rate Cl00 Speeitiiy for dfferent tot-off
poins of a parameter

Fach point Cuve

eprasents a
Senaitivihy Speuhciny paiy cenegendy
a particular deu'silon threshdd .
The are

a weasne of ho uel a paramaer

dishngulsh belucn thcse
Downloaded from EnggTree.com
Segmants
 EnggTree.com
*An ROC Cue s ConveX if the lopas Cnd

mont
olmgtmualy hon - ushen

A
wo o Concauiadjacant
wona y Sogmens oith
Cvave
increasim
than
Slopes, tnd'ates Loally worse
andau raneing
90ald bete
raubing pertoace by Jonlng
Seaments imvoLued in oha Con Cawihy
Tus
Creabtng a loarsey elassifeN
PRE ceION AND RECALL:
Relevamce Subjecti e hotion
Diffenent way dtfer about bhe
TeleraMee non- Yolavenca of porttealaf
docomeuk dho qen quasti imns
Yesponse o
IR Sygem Searchos is doesment Colechien and
ems Odorod esponses
Is Called Tei Set of
Downloaded from EnggTree.com
 EnggTree.com 26
A better eanch geilds a better raukod
and better ranbed sts hep tha gerr
imhomatbBoy raod

This Set oP eords jm tha

database. oiich holevant bhe Search
topic
GsSUmed to be eithet
Recorcs
olevat or imelevant
ay not
achal Yetnval get
sel o volovan
match dote
perkety
reords.
Set

Set o
Yelevant
ifemsio
the dateba

\nelonl
items
reiewed

rolaOm
-s EnggTree.com
Downloaded from
ems- reved
Relewamt Tlems- ot vehve
 Recall EnggTree.com (21
is bha ratio
nomer of
velevont Yeeords Yatived to tohe total
womber elevam the
dalabase
s vsoaly
percenta
A A = wmber of
o ralevo
Teds eied
B Nmber of reluvan
Recai A.
A
* (00. reords not
A tB Teeved
Preesin Yabioof dota nmber
of rolewat YeLords heeved tthe total
wmber I'elsvent and elevan reords
rehived . expressed

pestenta A: wmber of
relevat
Yeods ratneved
C- Nmbey of i
YO LoYds roeevElyawe
Downloaded from EnggTree.com
 (
EnggTree.com

preslon = A
A+c

As Yecall pracision deoma,

and daes tRa peclsi n
F- Measme

ts a measna d et
aturay and
i's daj'med eighbad hamone
mean Yeaal
tta test

The P- aasne F- Soe

mest Singa hombe"

mease in Tofonation ehieval, Naal argage
poocssimg Qnd Maawie laamworg
4-meagune Comes rom Bnformabi on
Ratival CiR) Lohare Pecale is

ohich relavant dooments ara ehiee

recalled
by bot i+ S
by Tn
posi tie
eleeleve
Rade CTpe)
Senaitiity
Downloaded from EnggTree.com
 EnggTree.com

is tha
Y Precision
etieved docomens fraaueny
porecdichiong ase
Whi ch
ralsvant
Comeet and s a fon of
also kuown positie predicti ve
Vase Cppv) oT pasithe Aoraey (Tea)
0s lmtended Combine these i nto a
singe meousne af
of Search epecsvenes
2 pNeeisjon x ocall
prousion + Peealu
MOUTICASS CLASSIFICATION
is Machie laonming
tosk ttat
casifiakon.
Consists of more than huoo
Classes otputs
Each
of N deffenen
hraiming poimt belongs
belongs to Una
Closses
The goal is
Laich, given a
fnchion
data point, oil Conecty
pradict class Jho tha
point
kalanygs
Downloaded from EnggTree.com
 EnggTree.com

botMean
gtned Averagi
aNe Yoe poe LisiDo (MAp) ts also
alled
peusim at Seen ralevent docomank
Dt daermne preu'sion at each paih wben
neoo relevemds doomant yt
retrieed.
Avera of preusion Valo cbtalnay
docoment each bino a relwent
cscyment is ehieued.
MAP |
N

ohee
6 NUmber of elavont dolomOnt
for
NeOmber aquenes
P(aea
doe mans
MOutCLASS UASSIFICATION TECHNIQUES:
truig
aiffenent clos point balogs
Downloaded from EnggTree.com
 EnggTree.com
The gol Cenatvet a fachi
which guen a he dala polnt oi

Cosecty prodicjed ta class o coh'ch

whichh
tto neo point balorgs
The oli -clogs clousifiaim pvoblan
Yefers t assign'mg ch f tee obsevations

0nto ona o k classes.

A Connmon combine pair
otse is
Compoisim Jy vohmg
Conshek a rule
beheon nan

Gaechmg
oo -
closs uoit te most
loss.deeisions r

\oing otedime
Degutnas Just peiy sise daesiong, it
predes class dabel.
A xXX

BinaryDownloaded
casahfrom EnggTree.com clasciioin
 EnggTree.com
One is Al (OvA)

*For ttu's appoach wwe Taqina

cassifiers, where k th
N2 k
bimaytraloned oito posihve examples
Caseifer ls
and vagatia

ekomple belngig to te offer k-| clases

Ohen an Unknun Qxampe

the classiAes prodsng te awim
naxio
Consldeed ta Loinner and
Dotput s
tthis class (abd is assinad to tat

example
Eror Comechg Doptpat Cedomg
yEoros Coechng code appro aches
o Conmbime bimamy cossi Aers tm
way that You exploit de
Corelation and Commeet

Downloaded from EnggTree.com

 EnggTree.com
E- Test

When Smalt Sample si zo 130) 1s

Considelod the tests a Ye InappliCa ble
because the aSS umptipnS We Nacle Por
Sample tests, oo ot hold goocd Por
lavg
Small Samples.
Small Samples It ao not Possi bla

ko aBre,

i) hat tie random Sampling dsti bubicn

Statistics normal
a

Sample Values are syfielenty case

r) The
bo Popation values bo Caleula be Te
estimale.

Thus entisely
to dea! with pob lemo of Small Sanples. Bat
Ohe Ghold note that tae methods and teorg
q Sma ll Samples applicoble bo lange
Barnples but it Coveise

when
Sample &ine Sma |), as to ften
ease n paetie, tu Conttal linit
(leorem loes not
not appy ne must ton
èmpuse 8\icter aMunptcons on teu poplatioy
to quwe latstial valdik to te kestprocacha
asunpbio
Fáom bwich Tlu Sample b balken has
Downloaded from EnggTree.com
normal Aobabiiby dostibution bo bogis wits
 EnggTree.com
* Dogee fheodom (di): ey degue freedom
we mean numbr classes bo whih
he value Can be
nignod asb taily Dr
or af
withot Asthidios Or
Ainiations plasd.
For ercanple ashe to
whose todal 5o. cleorly
we at feelonn
e
Mubes eSoy lo, 23, g but tu fourt5
rmter, fireol ine ttu botal
6o To- C1o +23+ )
7)i o . Thus'we e quen
a estkiction, hare bue qreedom asclaylcon

sf paeesbnm (dy) 1s denotad by

dagrea Nn-ki
~Cnu) r of and it jo given by
where mbe 4 lases and k:nmber
G indpenolent Constsains.
Mean
t- Test por gingle lome tfiom a
when te Gample Values eart distubetios
nor mal distwbtion
was worled ut by W. Gose tt
called JE a
One t- ohibutan
Lngortcnotely ee es not
litaibutions gor lauh
Ttere are feent t- tere a cwtain
I n:
diyunt Value the t- dshibution
t- distbuticn but it n: 43
Say trat te Vavable
Va
aliHle dit fuent we
t- s hibuton
Downloaded from EnggTree.com
wt n
* Sappose
SipleEnggTree.com
random anpla
n rs drawn FAOm Population. if te epulaton
Paom which
tnple ks takan Pollow8
normal elsibution
random Vaiable

Folows tudeot's t. Disthibudion with n- ( deges

Pheclom
The Sample ean and the Sanple ghandard
dovicutien &.
umber Ree
choices lyt efter a ample Rtatutiu Suh as
is calcalated when you wse a b- lsiibasion so
Cstmabe a pppulation een,
Rqual bo onu los than
df. = n

I::Population s normal attougo thà alumpton

Can be relae
a. Random
St int erest.
Sample wo drauon Faom tuu Poplation
Based Lompari Son ealeulabecd Value
tuito tine thohe tial ' Vaue Prom he table,
we conclucle:

Downloaded from EnggTree.com

 EnggTree.com
&hape Steclen t's t stibudion

Prepertis Students t- Distaibudiorn lffent

difeent Por
4. vhe t- olistibudion s
deqtees fheeotom.
rte t- utibuton Cented at o and
2
dyrnmetnis about O.
3. The tolal lnler te
te leyt onel tie ea bo tae rogh t

magnibede ineeases t e gtaph

A3 te
Cpproae but hever

area tue
de taiks a tre t- lisibution is
tn tee tal[ te norma)
largor fan te area

utribution.
tte t- dstibuteon N olkpendent
6 qte 3hape
lnereales tiee cshibton
AS Sampe Siza
becomes Opprorimately nomal.
Tee Stand ard lewiaton } reater
than .
8
tre t- drstribufion
tails
Je
a hf greator than the area in the 6ails Of
Tuo Standayol norha listibution, betaue we

}ndsodueng further
an
estimate ot
Vaniabilety
q. e men, medlian , and mode fue t- ostaibution
Downloaded from EnggTree.com
eual to
 populor is ho) p
Downloaded from EnggTree.com
o-efhcient Comelabin The
lation! Coe foy Test
Woma)
960 |:
|:962
O42 2
213|
251)
o25 totreodg lagnee
af
n
wonmal) te
ed dishobuon
Qre fofrocdan
r
to t a
dogsecs
f ous Vard too Vales Critical y
Valtag Citca) T-
numbeY&.
lore Lawg tue by claser
bo
get
values increOog
Sixe
n Sampe tte as
urve. olensity ma
l no
te toloser get luve
y teireasos In Sample tu'
EnggTree.com
 33)
EnggTree.com
Shonat h of tto volahnship
desenbi
beean wo Vanad
CoYrolation
The Co-efdent tho clopo of
behemn wo Vana Ha
Yeopeslon
Vamabos aue bean andardiod
when bote
t a r moang aud aividing
Ssbhihgstandasod moang Or doation
\oy bolwaen
Comolation
rovg
mias one
plus
Casinptive
tohenSpecial dishhbutional assUmptiong
Shatisic, ho aboot t e Voaniates
wada
to be
whi ch
it is Calculated

Comelatien oefaens
fest fo
Rrmola
t r

with dagness
State moll and alemate hypotesis
Ho
Downloaded from EnggTree.com
Ha p +0
 EnggTree.com
Hee p Is tta popuation Comelatior
Lo-oftiedent
Stote tho Significanto davel.
the fest eatisttes of Corobhn
to-efHelent oith
aboe - dahod
above
brma
do's ion Use
Vale approach or tha p- Vale approach
Fnalty Stete tfe Conlgion
Me emar's Test:

It is a hon- paDamene tost

paired nomial data
Used for fmdimg a

chang ppportov for paired data

Compare tha
losciflers Dn N ideny om Sirgle tes
Sel

This testto Comp On

the porfomanca of bo caacihers
Lame test set
Downloaded from EnggTree.com
 EnggTree.com
This test
4
WomeY ohich A and B
wabe reehons

Me NOmarlg test applid to

taldes wtth ratchad pairs
conbigeny
Subjecs tro
datcm'a Lo haastta
warqial traayenu
tha and cdom
are oal.
malin ascUmptiong o test
The tfnoe

hominal aato
most
Juoo Oud one indopendent
ategones aud
toitt oo ConnoCted oops
Variabio

Tha grups im the dapendonb

Vatiabla be wohaly dosie

be sando Semple
k-flod Cv pained t Test:
we Use k- fold Validation to get
K
Tratnng Validoion set pais
Downloaded from EnggTree.com
 EnggTree.com
To traln the to classifi Cahsw

algoni ralming sots Ti;

K and test

Vali'dation Sek V;

Tha percantagu of
Vali'dation Gets
clossiers (On

oordod as p and p?
buo obgcifiahisn alguitims
daue Some eor rate, Hhon we expect
Jhoue Same
ttthem o
eruvalently,te differenca of tfair maansis
Tates
in
,The difperera
ld i as Ri =P-P tu's a paied
test
that is or each both aloithns
bhe Sone
Validaim Ses

k
(Pi-m)
Downloaded from EnggTree.com
K -)
 EnggTree.com
*Undar coha hyoóllais ttok o,
have a Shatist t tat is 4- di'sibute
oith

Thos tte K-fold

veyecls
hyoteate ttat hwo Coasihcaton
lgonthns hoe ttha Sona emoy rate at
Biniicance toel
tta interval C-t K'ja K-)
we want o ast whaatter te
first aleosth Jhas dos ttan tha
Se cond maed a sidad hipofkasis
test :
Ho
the tegt
rejace, bur ca'
first
tal Ono bas
signhiant
Less is Sogprted
Downloaded from EnggTree.com
 EnggTree.com

ttat eah tegtstS

Advamteg
independon+
But bha
traeng set
oerlap
Tha
obtaimig a tood shat of to
moont of aiati on thab cooUld be
bbsene each tsang Set weno,

Completey imdoperelons f pnawia trau'mirg

tete

Vanana in
the t statisti ray
Somattas Uhalosastimaded, te maan he
eshimated
regu in t Valos
large

Downloaded from EnggTree.com