(Question Anstues
Nam thefrst c hatbot,
Ans: NatunalLanguage Pocesing arNLe, I Hhe
Sub-feld of ALthat foesed on enabllng
Compute to cndenstand and o e s human
language
3 wnat f
fnsi n NLeadictenan s a iist of all thewni
ae epeated îaaiffent docments, Heg
are alriHen ust oncea kle Creathhg
aconat doqa mean by Carpus?
AnsTe dem ced for teohole tetual deta frm
all t e docments altagthe g Rnean as CaRPOs
[Link] the tyees of data cied for NLP agelcathns?
Natnal Language Proceng ar NLe, takas ih the
data of nathal lanquage Dn the fom ofiarntten
dailliea and operates on ths
6 What do qau mean by Text Nocmalsaton
seveal steps a te kann
Domalisaho,
cmpleity to a leer Rael
edece the tet machine to po ceas .
that it is easy for the
So
Senttmce Segentaton
OR
Segmertation
explaia the proess of Sentence
Seamentatfon, te whole caps
ns Under Sentence acn sennce is ta Ran
sentences.
f6 Aded into helecarp
aditfent dataSone tha
g2ts Soduced to sentence.
do au meon by
Tokenisation
$Ohat sentme
has AttehSeqmantng tue sentnces, eacn
or spca
temwsed for ans coord o numbe
ina sentence Onda
chara cte
toRensahon, evey coord, number and Spe cal
characte s considdaed sepesate ls ad each sf
then s n a a sepenste
example, L am a bag
Define stopaonds ohgare stopoords, specal
fa tha carpus but ào nat add aon value to Tt. Toab
PEe Ne
sun copw mght hae
2 a r d a lot of +mes Natdapends
specaalchcachters andlornumbehs
not
hesheLld keotam Tn tor
wat
for ft1s, fo i
Examplean, a n d a e a
How s utomatie
utomahe Summasahan s e l e t far
Ans:
ofaNes Ttemor bleg pos, whiie oodng
ultiple sones ond maximising
dndancfr
Contet obtnned
the divenity of
ie theapplicatong of TEIDE
amd geore of adocument
b) Togi Medellog:lt hadpsin predicting tee topic for a
Corps
olnforeation Ratieval steny To extract tha inpor tant
informatonaet of acerpu
a)soord fheing Helps in e maring the nneceskary
worda out afa text body
naTnegaal ofasentineat amalys to lentif
Cxpressed. Componies sentinant onalst
e
to fAmhiyepfafams and sentiment on\eto
help hancAndastanduohat ccusto es tainb
abot thair products aaA seices and aueall
fodicators of tha ir eputoton
Text clasøEatan
Ars Text clacsifcation mapes t poseibleto asgn
predefed cateqeres to a docimant and argai
Rt to halo uou ffad te îaformatia go nNad
soma actites For esample,an
applfcatan of text categorsatan is spam
Awhat do 4oa mean by Bag f cwocds Algonthm
oR
Ans Bag of Words is an NLP madecohih helgs în
extrachg featwus ot of thhe tex ohihcaa
be help ful în machne leaning algorth
ord and constuct tha
vocabulany for th
1S-Rte thg steps toInnplement the Baa of Wna
Algarin.
ns a) Tet Normalis athon Callect deta
6) Create
and popne
Dictonac Maka a list of all the uniu
worda occina inthagcorps
înHa
)ea Doaumemt Vectors: For each document
hau many hma
corpuz find out
fron the u r s lie
tha sord occwud
cards has
of
A) Create doqument ectars for al tha dgcwunnens
f oceene Vs
l6. Daw a qaaphical preg emtaho
value of d f e e t wrda fthe Cocpu
Step words
Occurence
Fresant Words
Rarevatuatte ord
Yalue
(9Give an example of the Pallofng
a)Mulhele neanings ofa c d
" Thehed ca 2oamed past his ose
cas
Pro bablytalaing about the color of tha
"Hs face tunns red af consumina tae mediine
’ls hhayîna anallergi Seactan or s ha nat
to bear the taste of tnata adicie?
able
b) Pefect Syntax no Meaning
Chickens feed extraagantly hiethemoan dainee
Inis statement îs corect gammatcall but doe
Pege No
HhismaRe amy senseln Human langg
pesfect balance of synta a d semanthcs
Bonportant for betten undehs tanding
Complicated fora
haumn lanaage
Cenpudes to cmdehstnd Eaphin
A la aemulthple characteitics of the hnan
laanguae that miqht be cabg for t a hunmGn to
andehstand out detett for te compAten t
undenstand Ther ae Jule in huuman
laguag
werbs cte AcaocdCan bea
nonat one tme and an adeche Somaothas
Cneate dieAculty whil proccang
by Coonpu tes
Penfect Syntax, no meaning
DocliLandon is te capital of he uh
Doc 2.3London is Rnsn for îts choppignd hitorical Cultuse
Doc3Lbndon has mamg places cf nteet
Dec4Ihammee inen flos toagh London
DicTD NRI
Lendens, thacap+talof QKRnewn, for its,and,
she2pfe [Link] lhehas, man places,
intenetToancs, neh, flaus,thaugh
Pge No
ctur table for all dc wnets
&ngtthatheginamcopu aormlsed
We
to
Numb
is
1
ama
place
aPesformaIl tha stes of text normalfsahon onthe
folleig cerpus
hea th chacttbots for teatng
GheS
be ma Ring heathehatbots nel
Doc 3: Health chatbots cannot Seplace hu man
Cone(lars no. Yay X nelA4
Dnte
6) Lernmg hzatjon
chatbots Chatbt
treatíng
Lownsellors
hreat
CounGellor
[Link]
AnNohavacbala iofacoroL does n o t m s
some before arodiaf ten ext normalhathn ab
to nninin vocabu a
lnGennming, h a f f f b o r d s ane gamoreo
ormal6atfoneget the hadLced
Se, aft
936 onats docuwme vecto?c
A docunamt vector s a table sheetlue iqe
ce cana nowe wrie tha foeauamey of that
dfent ho5s o each docment. In th
94 0RE-enentiate be tweam thefollacoikq
natbots, haalth, nous,catto, Cneate NLP
sechatotstreatig, weealth, e,
Case men acom Conveting
tD )
ill, fo
cam, hamoed) Stopords
be to
nambes
wsordssecaal
chanactos
and StoeRamoving )
eA,4, Yag
ocs,, cqnsellas, human eplace cannot
haalthchatbots,
chattaats Health, nos, Ceate,
king cwel chatbots,
and
shneecan,
theahg
fo, NP cse Qe,
haaltth wse can,
tor,
echatbots, kenisahon Ta 2)
A4YXL@tel Yay a)
now llorsCownse
hman helae cannot chatbots )Health
tbotsa hea makiog
Hh be
tsand ate Cre NLPto wse
stresS
tentn forchatbats a)wecan
heath
Segmentaton Sentence Ana)
Dess
NoPagn
Pg No,
lemmaththon
Semming maes Seothot
does ta he into
nt
Stemmed Lemna th
accont if te
eanin
not.
longe fima
to etecate than
afie Hence,t R fasteh
|Steming
and Lemmat2ahon a a ee
Batth Stemming
f o r t h e e moal of affixe
Scie Bot Smat Bat
b
lkeane ecsy t make Thet a fleble and
pouseful
Seript shich pogam-databas and othe
hesouces dinctq
Linaited RunctSolty Lwtde funhanalt
Mostiy te ahe fre ond Tka le asn wtn more
easyto integate t a data
mesa aging plat form.
[Link] Sentemce.
entence the of shructuneof
a
gammahal tofes Ra
mantcs Syntetue
Derte: