T
E
L
K
O
M
N
I
K
A
T
elec
o
m
m
un
ica
t
io
n,
Co
m
pu
t
ing
,
E
lect
ro
nics
a
nd
Co
ntr
o
l
Vo
l.
19
,
No
.
1
,
Feb
r
u
ar
y
2
0
2
1
,
p
p
.
1
4
6
~
1
5
4
I
SS
N:
1
6
9
3
-
6
9
3
0
,
ac
cr
ed
ited
First Gr
ad
e
b
y
Kem
en
r
is
tek
d
i
k
ti,
Dec
r
ee
No
: 2
1
/E/KPT
/2
0
1
8
DOI
: 1
0
.
1
2
9
2
8
/TE
L
KOM
NI
K
A.
v
1
9
i1
.
1
7
4
1
7
146
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//jo
u
r
n
a
l.u
a
d
.
a
c.
id
/in
d
e
x
.
p
h
p
/TELK
OM
N
I
K
A
Sentimen
ts ana
ly
sis
of customer
sa
tisfaction
in
publi
c services
using
K
-
nea
rest
n
eig
hbo
rs
a
lg
o
rithm and
na
tural la
ng
ua
g
e
pro
cess
ing
a
ppro
a
ch
E
lik
H
a
ri
M
uk
t
a
f
in
,
P
ra
m
o
n
o
,
K
us
rini
M
a
g
ister o
f
I
n
fo
rm
a
ti
c
s,
Un
iv
e
rsi
tas
AMIKOM
Yo
g
y
a
k
a
rta
,
In
d
o
n
e
sia
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
u
l 2
2
,
2
0
2
0
R
ev
is
ed
Au
g
7
,
2
0
2
0
Acc
ep
ted
Sep
5
,
2
0
2
0
Cu
sto
m
e
r
sa
ti
sfa
c
ti
o
n
is
v
e
ry
imp
o
rtan
t
f
o
r
p
u
b
li
c
se
rv
ice
p
r
o
v
id
e
rs
,
c
u
sto
m
e
r
sa
ti
sfa
c
ti
o
n
c
a
n
b
e
d
e
li
v
e
re
d
wit
h
a
su
rv
e
y
a
p
p
l
ica
ti
o
n
o
r
writi
n
g
c
ri
ti
c
ism
th
a
t
c
a
n
b
e
u
se
d
t
o
e
v
a
lu
a
te
a
n
d
imp
r
o
v
e
se
rv
ice
.
Un
f
o
rtu
n
a
tely
,
th
e
re
a
re
o
n
ly
a
fe
w
c
u
sto
m
e
rs
wh
o
a
re
will
in
g
to
g
iv
e
a
n
a
ss
e
ss
m
e
n
t
.
Th
e
su
rv
e
y
a
p
p
li
c
a
ti
o
n
c
a
n
n
o
t
re
p
re
se
n
t
th
e
o
v
e
ra
ll
fe
e
l
in
g
o
f
t
h
e
c
u
sto
m
e
r,
s
o
i
t
is
n
e
c
e
ss
a
ry
to
a
n
a
ly
z
e
th
e
c
o
n
te
n
t
o
f
th
e
c
o
n
v
e
rsa
ti
o
n
b
e
twe
e
n
t
h
e
c
u
st
o
m
e
r
a
n
d
th
e
se
rv
ice
p
e
rso
n
n
e
l
to
d
e
term
in
e
th
e
le
v
e
l
o
f
c
u
sto
m
e
r
sa
ti
sfa
c
ti
o
n
.
I
n
sm
a
ll
a
m
o
u
n
ts
,
it
c
a
n
b
e
d
o
n
e
m
a
n
u
a
ll
y
,
b
u
t
i
n
lar
g
e
q
u
a
n
ti
ti
e
s
i
t
is
m
o
re
e
ffe
c
ti
v
e
to
u
se
th
e
sy
s
tem
.
A
so
lu
ti
o
n
is
n
e
e
d
e
d
i
n
th
e
fo
rm
o
f
a
sy
ste
m
th
a
t
c
o
n
v
e
rts
v
o
ice
c
o
n
v
e
rsa
ti
o
n
s
i
n
to
tex
t
a
n
d
a
n
a
ly
z
e
s
c
u
sto
m
e
r
sa
ti
sfa
c
ti
o
n
to
o
b
tai
n
in
fo
rm
a
ti
o
n
f
o
r
e
v
a
lu
a
ti
o
n
a
n
d
imp
ro
v
e
m
e
n
t
o
f
se
rv
ice
s.
T
h
is
re
se
a
rc
h
u
se
s
K
-
n
e
a
re
st
n
e
ig
h
b
o
rs
(KN
N)
a
n
d
term
fre
q
u
e
n
cy
-
i
n
v
e
rse
d
o
c
u
m
e
n
t
fre
q
u
e
n
c
y
(
TF
-
IDF
)
a
l
g
o
ri
th
m
wit
h
n
a
tu
ra
l
lan
g
u
a
g
e
p
r
o
c
e
ss
in
g
(
NLP
)
a
p
p
r
o
a
c
h
to
c
las
sify
c
o
n
v
e
rsa
ti
o
n
s
in
to
2
c
las
se
s,
"
sa
ti
sfie
d
"
a
n
d
"
d
issa
ti
sfie
d
".
Th
e
re
su
lt
s
o
f
th
is
stu
d
y
re
c
e
iv
e
d
7
4
.
0
0
%
a
c
c
u
ra
c
y
,
7
6
.
0
0
%
p
re
c
isio
n
a
n
d
7
3
.
0
8
%
re
c
a
ll
.
In
c
o
n
v
e
rsa
ti
o
n
s
wi
th
t
h
e
lab
e
l
"
sa
ti
sfie
d
"
sh
o
ws
c
u
sto
m
e
rs
sa
ti
sfie
d
with
th
e
se
rv
ice
a
n
d
fu
lfi
ll
m
e
n
t
o
f
c
u
sto
m
e
r
d
e
sire
s,
wh
il
e
in
c
o
n
v
e
rsa
ti
o
n
s
wit
h
t
h
e
lab
e
l
"
n
o
t
sa
ti
sfie
d
"
c
u
st
o
m
e
rs are
les
s sa
ti
sfie
d
with
t
h
e
wa
it
in
g
ti
m
e
.
K
ey
w
o
r
d
s
:
K
-
n
ea
r
est n
eig
h
b
o
r
s
Natu
r
al
lan
g
u
ag
e
p
r
o
ce
s
s
in
g
Pu
b
lic
s
er
v
ice
Sen
tim
en
t a
n
aly
s
is
Sp
ee
ch
r
ec
o
g
n
itio
n
s
y
s
tem
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r
th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
E
lik
Har
i M
u
k
taf
in
Ma
g
is
ter
o
f
I
n
f
o
r
m
atics
Un
iv
er
s
itas
AM
I
KOM
Yo
g
y
ak
ar
ta
R
in
g
R
o
ad
Utar
a
St.,
Ng
r
in
g
in
,
C
o
n
d
o
n
g
ca
tu
r
,
Dis
tr
ict.
Dep
o
k
,
Slem
an
R
eg
en
c
y
Sp
ec
ial
R
eg
io
n
o
f
Yo
g
y
ak
ar
ta
5
5
2
8
1
,
I
n
d
o
n
esia
E
m
ail:
elix
r
o
p
as@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
Pu
b
lic
s
atis
f
ac
tio
n
is
a
d
eter
m
in
in
g
f
ac
to
r
in
th
e
q
u
ality
o
f
p
u
b
lic
s
er
v
ices,
s
o
ev
er
y
o
r
g
an
iz
atio
n
th
at
p
r
o
v
id
es
p
u
b
lic
s
er
v
ices
is
ex
p
ec
ted
to
b
e
a
b
le
to
p
r
o
v
i
d
e
s
atis
f
ac
tio
n
to
cu
s
to
m
er
s
[
1
]
.
T
o
f
in
d
o
u
t
cu
s
to
m
e
r
s
atis
f
ac
tio
n
,
th
e
r
ig
h
t
m
ea
s
u
r
in
g
in
s
tr
u
m
en
t
is
n
ee
d
ed
[
2
]
.
C
u
s
to
m
er
s
atis
f
ac
tio
n
is
u
s
u
ally
m
ea
s
u
r
ed
b
y
a
s
u
r
v
ey
ap
p
licatio
n
o
r
cr
itical
wr
itin
g
th
at
ca
n
b
e
u
s
ed
to
ev
a
lu
ate
an
d
im
p
r
o
v
e
s
er
v
ice.
Un
f
o
r
tu
n
ately
,
t
h
er
e
ar
e
o
n
ly
a
f
ew
cu
s
to
m
er
s
wh
o
ar
e
willin
g
to
g
iv
e
an
ass
es
s
m
en
t.
L
o
w
co
n
d
itio
n
s
o
f
p
u
b
lic
s
er
v
ices
ar
e
ch
ar
ac
ter
ized
b
y
p
u
b
lic
s
er
v
ices
th
at
ar
e
n
o
t
tr
an
s
p
a
r
en
t,
d
is
cr
im
in
ato
r
y
,
co
m
p
licated
,
an
d
c
o
r
r
u
p
t.
T
h
e
ap
p
licatio
n
o
f
p
u
b
lic
s
er
v
ic
e
p
o
licy
in
n
o
v
atio
n
s
b
ec
o
m
es a
n
ec
ess
ity
in
r
ea
lizin
g
q
u
ality
p
u
b
lic
s
er
v
ices
[
3
]
.
T
h
e
s
u
r
v
ey
ap
p
licatio
n
ca
n
n
o
t
r
ep
r
esen
t th
e
o
v
e
r
all
f
ee
lin
g
o
f
th
e
cu
s
to
m
er
,
it is
n
ec
ess
ar
y
to
an
aly
ze
th
e
co
n
te
n
ts
o
f
th
e
c
o
n
v
e
r
s
atio
n
b
etwe
en
th
e
cu
s
to
m
er
an
d
th
e
s
er
v
ice
p
er
s
o
n
n
el
t
o
d
eter
m
in
e
th
e
lev
el
o
f
cu
s
to
m
er
s
atis
f
ac
tio
n
an
d
ab
s
o
r
b
an
d
u
n
d
er
s
tan
d
c
u
s
to
m
er
d
esire
s
[
4
]
.
An
o
th
er
p
u
r
p
o
s
e
is
to
p
r
ev
en
t
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
S
en
timen
ts
a
n
a
lysi
s
o
f c
u
s
to
mer sa
tis
fa
ctio
n
in
p
u
b
lic
s
ervices u
s
in
g
… (
E
lik
Ha
r
i Mu
kta
fin
)
147
d
is
cr
im
in
ato
r
y
o
r
co
r
r
u
p
t
p
r
ac
tices
in
th
e
d
eliv
er
y
o
f
s
er
v
ic
es,
s
o
th
at
th
e
y
ca
n
d
eliv
er
s
e
r
v
ices
th
at
ar
e
f
air
,
d
em
o
cr
atic
an
d
f
r
ie
n
d
ly
[
5
]
.
A
n
aly
s
is
o
f
s
er
v
ice
co
n
v
er
s
atio
n
s
in
s
m
all
am
o
u
n
ts
ca
n
b
e
d
o
n
e
m
an
u
ally
,
b
u
t
in
it is
m
o
r
e
ef
f
ec
tiv
e
to
u
s
e
th
e
s
y
s
tem
in
lar
g
e
q
u
an
titi
es.
R
esear
ch
er
s
h
av
e
d
e
v
elo
p
e
d
a
p
r
o
d
u
ct
ca
lled
"
K
a
ta
K
ita
"
wh
ich
s
er
v
es
to
s
u
p
p
o
r
t
p
u
b
li
c
s
er
v
ices.
"
K
a
ta
K
ita
"
h
as th
e
f
ea
tu
r
e
o
f
tr
an
s
latin
g
co
n
v
er
s
atio
n
s
in
to
tex
t a
n
d
d
is
p
lay
in
g
th
e
m
o
n
t
h
e
s
cr
ee
n
t
o
h
el
p
th
e
d
ea
f
u
n
d
er
s
tan
d
s
er
v
ice
p
er
s
o
n
n
el
an
d
s
en
tim
en
t
an
al
y
s
is
f
ea
tu
r
es
to
ass
ess
cu
s
to
m
er
s
atis
f
ac
tio
n
f
r
o
m
th
e
co
n
v
er
s
atio
n
.
Sen
tim
en
t
An
al
y
s
is
is
u
s
ed
to
f
in
d
o
u
t
th
e
o
p
i
n
io
n
ab
o
u
t
a
to
p
ic
wh
ich
is
as
p
o
s
itiv
e,
n
eg
ativ
e,
an
d
n
eu
tr
al
s
en
tim
en
ts
[
6
]
.
T
h
e
p
u
r
p
o
s
e
o
f
th
is
s
tu
d
y
is
to
ex
am
in
e
th
e
ef
f
ec
tiv
en
ess
o
f
"
K
a
ta
K
ita
"
to
co
n
d
u
ct
s
en
tim
en
t a
n
aly
s
is
o
n
p
u
b
lic
s
er
v
ice
co
n
v
er
s
atio
n
s
in
o
n
e
o
f
th
e
h
o
s
p
itals
in
Yo
g
y
a
k
ar
ta,
I
n
d
o
n
esia.
C
o
n
v
er
s
atio
n
s
ar
e
r
ec
o
r
d
ed
a
n
d
co
n
v
e
r
ted
to
tex
t
u
s
in
g
Sp
ee
ch
R
ec
o
g
n
itio
n
Sy
s
tem
tech
n
o
lo
g
y
.
Sp
ee
ch
r
ec
o
g
n
itio
n
s
y
s
tem
is
th
e
p
r
o
ce
s
s
o
f
co
n
v
er
tin
g
s
p
ee
c
h
s
ig
n
als
in
to
wo
r
d
s
eq
u
en
ce
s
u
s
in
g
an
alg
o
r
ith
m
im
p
lem
en
ted
as
a
co
m
p
u
ter
p
r
o
g
r
am
[
7
]
.
T
h
e
te
x
t
o
f
th
e
co
n
v
er
s
atio
n
is
p
r
o
ce
s
s
ed
u
s
in
g
m
ac
h
in
e
lear
n
in
g
to
b
e
lab
eled
au
to
m
atica
lly
with
o
u
t
th
e
n
ee
d
to
wait
f
o
r
in
p
u
t
f
r
o
m
cu
s
to
m
er
s
,
s
o
th
at
th
e
p
r
o
ce
s
s
o
f
ev
alu
atin
g
an
d
im
p
r
o
v
in
g
s
er
v
ices c
an
r
u
n
ef
f
ec
tiv
ely
.
Pre
v
io
u
s
r
esear
ch
b
y
L
u
tf
i
a
n
d
Per
m
atasar
i
[
8
]
an
aly
ze
d
th
e
m
ar
k
etp
lace
to
f
in
d
o
u
t
p
o
s
itiv
e
o
r
n
eg
ativ
e
u
s
er
s
en
tim
en
t
with
an
ap
p
r
o
a
c
h
u
tili
zin
g
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
with
a
n
ac
c
u
r
ac
y
o
f
9
3
.
4
2
%.
I
n
a
s
tu
d
y
co
n
d
u
cted
b
y
Ak
h
m
ad
Dev
iy
an
to
an
d
W
ah
y
u
d
i
[
9
]
p
r
esen
ted
a
s
en
tim
en
t
an
aly
s
is
o
f
o
p
i
n
io
n
d
ata
o
n
T
witter
u
s
er
s
en
tim
en
t
an
aly
s
is
o
n
th
e
t
o
p
ic
o
f
lo
ca
l
elec
ti
o
n
s
in
th
e
J
ak
ar
ta
o
f
I
n
d
o
n
esia
in
2
0
1
7
with
an
ac
cu
r
ac
y
o
f
6
7
.
2
%.
I
n
a
n
o
th
er
s
tu
d
y
b
y
No
r
m
an
Ken
d
al
[
1
0
]
s
h
o
wed
th
e
u
s
e
o
f
n
atu
r
al
la
n
g
u
ag
e
p
r
o
ce
s
s
in
g
(
NL
P)
ca
n
im
p
r
o
v
e
th
e
p
r
ed
ict
io
n
o
f
f
ash
io
n
p
r
o
d
u
ct
ca
teg
o
r
ies o
r
s
u
b
ca
teg
o
r
ies o
f
p
r
o
d
u
c
t titles
.
T
h
e
p
u
r
p
o
s
e
o
f
t
h
is
s
tu
d
y
is
t
o
an
aly
ze
th
e
s
en
tim
en
ts
o
f
p
u
b
lic
s
er
v
ice
c
o
n
v
er
s
atio
n
in
o
n
e
o
f
th
e
h
o
s
p
itals
in
Yo
g
y
a
k
ar
ta,
I
n
d
o
n
esia
u
s
in
g
th
e
u
s
es
K
-
n
ea
r
es
t
n
eig
h
b
o
r
s
(
KNN)
alg
o
r
ith
m
b
y
weig
h
tin
g
ter
m
f
r
eq
u
e
n
cy
-
in
v
er
s
e
d
o
c
u
m
en
t
f
r
eq
u
en
c
y
(
T
F
-
I
DF)
,
an
d
c
lass
if
y
in
to
two
class
es
"
s
ati
s
f
ied
"
an
d
"
d
is
s
atis
f
ied
".
T
h
en
f
in
d
o
u
t
attr
ib
u
tes
wh
ic
h
ar
e
th
e
f
o
cu
s
ed
o
f
"
s
atis
f
ied
"
an
d
"
d
is
s
atis
f
ied
"
ass
e
s
s
m
en
ts
,
s
o
th
at
s
er
v
ice
p
r
o
v
id
e
r
s
ca
n
p
r
o
p
er
l
y
e
v
alu
at
e
an
d
im
p
r
o
v
e
th
e
s
y
s
tem
[
1
1
]
.
Ma
ch
in
e
lear
n
in
g
is
an
ef
f
ic
ien
t
way
t
o
e
x
tr
ac
t
k
n
o
wled
g
e
f
r
o
m
la
r
g
e
am
o
u
n
t
s
o
f
d
ata
[
1
2
]
.
T
h
e
p
r
o
b
lem
in
an
aly
zin
g
p
u
b
lic
s
er
v
ice
c
o
n
v
e
r
s
atio
n
s
is
th
e
lar
g
e
n
u
m
b
er
o
f
n
o
n
-
s
tan
d
ar
d
wo
r
d
s
th
at
ar
e
d
if
f
icu
lt
to
u
n
d
er
s
tan
d
b
y
th
e
s
y
s
tem
.
T
h
e
NL
P
ap
p
r
o
ac
h
is
n
ee
d
ed
t
o
im
p
r
o
v
e
th
e
la
n
g
u
a
g
e
o
f
th
e
c
o
n
v
er
s
atio
n
s
o
th
at
it
is
m
o
r
e
ea
s
ily
u
n
d
er
s
to
o
d
b
y
th
e
s
y
s
tem
.
W
ith
o
u
t
NL
P,
m
ac
h
in
e
lear
n
in
g
ca
n
n
o
t m
a
k
e
m
ea
n
in
g
f
u
l p
r
o
g
r
ess
[
1
3
]
.
2.
RE
S
E
ARCH
M
E
T
H
O
D
T
h
e
f
lo
w
o
f
r
esear
ch
s
en
tim
en
t
an
aly
s
is
o
f
co
n
v
er
s
atio
n
s
in
p
u
b
lic
s
er
v
ices
u
s
in
g
th
e
KNN
an
d
TF
-
ID
F
alg
o
r
ith
m
b
y
u
s
in
g
th
e
NL
P
ap
p
r
o
ac
h
is
ex
p
lain
e
d
in
Fig
u
r
e
1
.
Th
e
in
itial
s
tep
o
f
th
e
s
tu
d
y
was
th
e
co
llectio
n
o
f
r
ec
o
r
d
ed
c
o
n
v
e
r
s
atio
n
s
o
f
f
r
o
n
t
d
esk
s
er
v
ices a
t o
n
e
o
f
th
e
h
o
s
p
itals
in
Yo
g
y
a
k
ar
ta,
I
n
d
o
n
esia o
n
J
u
ly
1
3
,
2
0
2
0
.
C
o
n
v
er
s
atio
n
r
ec
o
r
d
i
n
g
s
wer
e
co
n
v
er
ted
in
t
o
tex
t
u
s
in
g
th
e
"
K
a
ta
K
ita
"
ap
p
licatio
n
with
t
h
e
s
p
ee
ch
r
ec
o
g
n
itio
n
s
y
s
te
m
f
r
o
m
Go
o
g
le
s
p
ee
c
h
-
to
-
tex
t
[
1
4
]
.
T
h
e
n
ex
t
s
tep
,
tex
t
was
s
to
r
ed
in
a
My
SQL
d
atab
ase
s
o
th
at
it
is
ea
s
ily
m
an
ag
ed
[
1
5
]
.
T
h
e
wo
r
k
f
lo
w
o
f
th
e
s
p
ee
ch
to
tex
t
s
y
s
tem
in
th
e
"
K
a
ta
K
ita
”
ap
p
licatio
n
is
ex
p
lain
e
d
in
Fig
u
r
e
2
.
Fig
u
r
e
1
.
R
esear
ch
s
tag
es
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
19
,
No
.
1
,
Feb
r
u
ar
y
2
0
2
1
:
1
4
6
-
1
5
4
148
Fig
u
r
e
2
.
T
h
e
p
r
o
ce
s
s
o
f
ch
an
g
in
g
th
e
v
o
ice
o
f
co
n
v
er
s
atio
n
in
to
tex
t
T
h
e
n
ex
t
s
tep
,
co
n
v
er
s
atio
n
t
ex
t
was
th
en
p
u
r
ch
ased
m
an
u
ally
b
y
t
h
r
ee
r
esp
o
n
d
e
n
ts
,
co
n
s
is
tin
g
o
f
cu
s
to
m
er
s
,
p
u
b
lic
r
elatio
n
s
an
d
m
ed
ical
s
tu
d
en
ts
.
T
h
e
d
ataset
o
f
2
5
0
co
n
v
er
s
atio
n
s
was
lab
eled
in
two
class
e
s
,
"
s
atis
f
ied
"
an
d
"
d
is
s
atis
f
ied
".
T
h
e
last
s
tep
,
d
ataset
is
s
u
b
j
ec
ted
to
a
p
r
ep
r
o
ce
s
s
in
g
p
r
o
c
ess
u
s
in
g
th
e
NL
P
ap
p
r
o
ac
h
.
Pre
p
r
o
ce
s
s
in
g
d
ata
is
th
e
p
r
o
ce
s
s
o
f
clea
n
in
g
an
d
p
r
ep
ar
in
g
d
ata
f
o
r
an
aly
s
is
[
1
6
]
.
T
h
e
p
u
r
p
o
s
e
o
f
p
r
ep
r
o
ce
s
s
in
g
is
to
co
r
r
ec
t
th
e
lan
g
u
ag
e
i
n
th
e
co
n
v
er
s
atio
n
,
b
ec
au
s
e
in
th
e
co
n
v
er
s
atio
n
f
o
u
n
d
m
an
y
u
s
es
o
f
non
-
s
tan
d
a
r
d
wo
r
d
s
an
d
lo
ca
l
lan
g
u
ag
es.
I
n
th
e
ca
s
e
o
f
tex
t
class
if
icatio
n
,
m
an
y
p
r
ep
r
o
ce
s
s
in
g
tech
n
iq
u
es c
an
b
e
u
s
ed
[
1
7
]
.
T
h
e
wo
r
k
f
lo
w
o
f
th
e
p
r
ep
r
o
ce
s
s
in
g
s
tag
e
with
t
h
e
NL
P a
p
p
r
o
ac
h
is
ex
p
lain
e
d
in
Fig
u
r
e
3
.
Fig
u
r
e
3
.
T
h
e
p
r
e
p
r
o
ce
s
s
in
g
s
t
ag
e
u
s
es th
e
NL
P a
p
p
r
o
ac
h
NL
P
is
a
b
r
eak
th
r
o
u
g
h
in
tr
an
s
cen
d
in
g
la
n
g
u
ag
e
b
ar
r
ier
s
[
1
6
]
.
NL
P
is
a
b
r
an
ch
o
f
Ar
tific
ia
l
I
n
tellig
en
ce
th
at fo
cu
s
es
o
n
p
r
o
cess
in
g
n
at
u
r
al l
an
g
u
a
g
e eff
ectiv
ely
an
d
accu
r
ately
as
h
u
m
an
s
d
o
[
1
8
]
.
T
h
e
NL
P
f
eatu
r
es
u
s
ed
i
n
th
is
s
tu
d
y
ar
e a
s
f
o
ll
o
ws
:
−
L
o
wer
ca
s
e
f
o
ld
in
g
L
o
wer
ca
s
e
f
o
ld
i
n
g
is
c
h
an
g
i
n
g
all
to
k
en
s
to
lo
wer
ca
s
e
lette
r
s
[
1
9
]
.
C
o
n
v
e
r
s
atio
n
“
S
ela
m
a
t
p
a
g
i
ka
k
a
d
a
ya
n
g
b
is
a
d
ib
a
n
tu
”
will
b
e
ch
an
g
e
d
to
“
s
ela
ma
t
p
a
g
i
ka
k
a
d
a
ya
n
g
b
is
a
d
ib
a
n
tu
”
,
alp
h
ab
et
“
S
”
in
th
e
f
o
r
m
o
f
ca
p
ital is ch
an
g
ed
in
to
”
s
”
in
s
m
all
f
o
r
m
.
−
W
o
r
d
n
o
r
m
alize
r
T
h
e
wo
r
d
n
o
r
m
alize
r
f
ea
tu
r
e
is
u
s
ed
to
co
r
r
ec
t
wo
r
d
s
in
s
en
ten
ce
s
,
s
o
th
at
a
g
o
o
d
an
d
co
r
r
ec
t
s
en
ten
ce
is
p
r
o
d
u
ce
d
ac
co
r
d
i
n
g
to
g
r
a
m
m
ar
r
u
les
[
2
0
]
.
C
o
n
v
er
s
atio
n
“
s
ela
ma
t
p
a
g
i
k
a
k
a
d
a
ya
n
g
b
is
a
d
ib
a
n
tu
”
th
e
n
af
ter
th
e
W
o
r
d
No
r
m
alize
r
p
r
o
ce
s
s
b
ec
o
m
es
“
s
a
la
m
ka
ka
k
a
d
a
ya
n
g
b
is
a
d
ib
a
n
t
u
”
.
T
er
m
“
s
ela
ma
t
p
a
g
i
”
ch
an
g
ed
in
to
“
s
a
la
m
”
an
d
ter
m
“
ka
k
”
ch
a
n
g
ed
i
n
to
“
ka
k
a
k
”
.
−
Stem
m
in
g
T
h
e
s
tem
m
in
g
f
ea
tu
r
e
f
u
n
ctio
n
s
is
to
r
ed
u
ce
wo
r
d
s
to
b
asic
wo
r
d
s
b
y
elim
in
atin
g
th
e
af
f
i
x
es
th
at
ex
is
t
in
th
o
s
e
wo
r
d
s
[
2
1
]
.
Fo
r
ex
am
p
le
,
s
en
ten
ce
s
“
s
a
la
m
ka
ka
k
a
d
a
ya
n
g
b
is
a
d
ib
a
n
tu
”
c
h
an
g
e
d
in
to
“
s
a
la
m
ka
k
a
k
ada
ya
n
g
b
is
a
b
a
n
tu
”
,
ter
m
“
d
ib
a
n
tu
”
c
h
an
g
e
d
to
b
asic w
o
r
d
s
“
b
a
n
tu
”.
−
Sto
p
wo
r
d
r
em
o
v
al
T
h
e
s
to
p
wo
r
d
r
em
o
v
al
f
ea
tu
r
e
wo
r
k
s
to
elim
in
ate
wo
r
d
s
th
at
o
f
ten
ap
p
ea
r
in
n
atu
r
al
lan
g
u
ag
e
b
u
t
h
av
e
v
er
y
litt
le
m
ea
n
in
g
[
2
2
]
.
Fo
r
ex
am
p
le
s
en
ten
ce
s
“
s
a
la
m
ka
ka
k
a
d
a
ya
n
g
b
is
a
b
a
n
tu
”
ch
an
g
ed
in
to
“
s
a
la
m
ka
ka
k
b
is
a
b
a
n
tu
”
,
ter
m
“
a
d
a
”
an
d
“
ya
n
g
”
r
e
m
o
v
e
d
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
S
en
timen
ts
a
n
a
lysi
s
o
f c
u
s
to
mer sa
tis
fa
ctio
n
in
p
u
b
lic
s
ervices u
s
in
g
… (
E
lik
Ha
r
i Mu
kta
fin
)
149
T
er
m
f
r
eq
u
en
cy
is
a
m
eth
o
d
f
o
r
f
in
d
in
g
th
e
weig
h
t
o
f
a
d
o
cu
m
en
t
b
y
lo
o
k
in
g
f
o
r
th
e
n
u
m
b
er
o
f
o
cc
u
r
r
e
n
ce
s
o
f
ter
m
s
in
t
h
e
d
o
cu
m
en
t.
T
h
e
m
o
r
e
o
f
ten
th
e
ter
m
ap
p
ea
r
s
,
it
will
af
f
ec
t
th
e
am
o
u
n
t
o
f
weig
h
t
an
d
th
e
s
u
itab
ilit
y
v
alu
e
o
f
th
e
d
o
cu
m
en
t.
I
n
v
e
r
s
e
Do
cu
m
en
t
Fre
q
u
en
cy
is
a
m
eth
o
d
f
o
r
ca
lcu
la
tin
g
th
e
d
is
tr
ib
u
tio
n
o
f
ter
m
s
in
a
d
o
cu
m
e
n
t
[
2
3
]
.
T
h
e
T
F
-
I
DF m
eth
o
d
allo
ws
d
o
cu
m
en
ts
to
b
e
class
if
ied
in
to
two
class
es
(
p
o
s
itiv
e
an
d
n
e
g
ativ
e)
[
2
4
]
.
TF
-
I
DF c
alcu
latio
n
u
s
in
g
(
1
)
.
,
=
,
×
log
(
)
(
1
)
,
is
th
e
weig
h
t
o
f
t
h
e
ter
m
(
t
y
)
of
do
c
u
m
en
t
(
d
x
)
.
W
h
ile
,
is
th
e
n
u
m
b
e
r
o
f
o
cc
u
r
r
en
ce
s
of
te
r
m
(
t
y
)
in
do
c
u
m
e
n
t
(
d
x
)
.
N
is
th
e
n
u
m
b
er
o
f
d
o
cu
m
en
ts
in
th
e
d
atab
a
s
e
an
d
df
x
is
th
e
n
u
m
b
e
r
o
f
d
o
cu
m
en
ts
co
n
tain
in
g
th
e
ter
m
(
t
y
)
,
t
h
er
e
is
at
least o
n
e
wo
r
d
,
ter
m
(
t
y
).
3.
RE
SU
L
T
S
A
ND
AN
AL
Y
SI
S
3
.
1
.
Serv
ice
c
o
nv
er
s
a
t
io
n
t
e
x
t
C
o
n
v
er
s
atio
n
tex
t u
s
ed
is
o
b
ta
in
ed
f
r
o
m
th
e
ac
tiv
ity
o
f
ch
an
g
in
g
co
n
v
er
s
atio
n
s
in
to
tex
t.
T
h
is
s
er
v
ice
co
n
v
er
s
atio
n
was
tak
en
in
J
u
l
y
2
0
2
0
at
a
h
o
s
p
ital
in
Yo
g
y
a
k
ar
ta,
I
n
d
o
n
esia.
E
x
am
p
les
o
f
co
n
v
er
s
atio
n
s
ar
e
s
h
o
wn
in
T
ab
le
1
.
B
ased
o
n
T
a
b
le
1
,
C
u
s
to
m
er
-
1
was
r
ed
ee
m
in
g
d
r
u
g
s
at
th
e
h
o
s
p
ital
p
h
ar
m
ac
y
an
d
is
s
atis
f
ied
with
th
e
s
er
v
ice
h
e
r
ec
eiv
e
d
.
C
u
s
to
m
er
-
2
was
p
ay
i
n
g
m
e
d
ical
ex
p
e
n
s
es
at
th
e
ca
s
h
ier
,
b
u
t
h
e
h
a
d
b
ee
n
tr
an
s
f
er
r
ed
to
an
o
t
h
er
ca
s
h
ier
f
o
r
u
s
in
g
B
a
d
a
n
P
en
ye
len
g
g
a
r
a
Ja
min
a
n
S
o
s
ia
l
K
eten
a
g
a
ke
r
j
a
a
n
(
B
PJ
S
)
with
o
u
t
s
ay
in
g
an
ap
o
lo
g
y
,
s
o
th
e
c
u
s
t
o
m
er
f
elt
d
is
s
atis
f
ied
.
C
u
s
t
o
m
er
-
3
was
talk
in
g
to
d
o
cto
r
s
with
a
m
ix
tu
r
e
o
f
J
av
an
ese
wo
r
d
s
ab
o
u
t
th
e
p
r
o
ce
s
s
o
f
r
ec
o
v
er
in
g
h
an
d
s
,
a
n
d
f
ee
lin
g
s
atis
f
ied
with
th
e
s
er
v
ices
p
r
o
v
id
ed
.
C
u
s
to
m
er
-
4
was
d
is
s
atis
f
ied
w
h
en
h
e
was
in
a
q
u
eu
e
p
o
s
i
tio
n
,
an
d
h
e
m
u
s
t
leav
e
th
e
h
o
s
p
ital
tem
p
o
r
ar
ily
,
w
h
ile
th
e
q
u
e
u
e
o
r
d
er
will
r
etu
r
n
f
r
o
m
th
e
b
eg
in
n
in
g
i
f
th
e
cu
s
to
m
er
d
id
n
o
t
r
etu
r
n
wh
e
n
h
e
was
ca
lled
.
C
u
s
to
m
er
-
5
was
in
a
ch
ild
'
s
p
o
ly
f
o
r
im
m
u
n
izin
g
h
is
ch
ild
an
d
h
e
was
f
ee
lin
g
s
atis
f
ied
b
ec
au
s
e
th
e
d
o
cto
r
ca
n
ca
lm
th
e
cu
s
to
m
er
.
I
n
f
o
r
m
atio
n
f
r
o
m
th
is
co
n
v
er
s
atio
n
ca
n
b
e
p
r
o
ce
s
s
ed
to
b
e
a
f
u
tu
r
e
im
p
r
o
v
em
en
t
.
T
ab
le
1
.
E
x
am
p
le
co
n
v
er
s
atio
n
No
C
u
s
t
o
mer
C
o
n
v
e
r
sa
t
i
o
n
La
b
e
l
1
C
u
s
t
o
mer
-
1
Assa
l
a
m
u
a
l
a
i
k
u
m
a
d
a
y
a
n
g
b
i
s
a
sa
y
a
b
a
n
t
u
k
a
k
s
a
y
a
i
n
g
i
n
m
e
n
e
b
u
s
res
e
p
i
n
i
m
b
a
k
s
i
l
a
h
k
a
n
d
u
d
u
k
d
u
l
u
k
a
k
s
a
y
a
a
k
a
n
m
e
n
y
i
a
p
k
a
n
n
y
a
i
n
i
o
b
a
t
n
y
a
k
a
k
y
a
n
g
t
a
b
l
e
t
h
i
j
a
u
d
i
m
i
n
u
m
4
5
m
e
n
i
t
seb
e
l
u
m
m
a
k
a
n
t
i
g
a
k
a
l
i
s
e
h
a
r
i
y
a
n
g
k
a
p
su
l
b
i
r
u
d
a
n
p
u
t
i
h
m
a
si
n
g
m
a
s
i
n
g
sa
t
u
s
e
t
e
l
a
h
m
a
k
a
n
t
i
g
a
k
a
l
i
se
h
a
ri
a
d
a
y
a
n
g
b
i
s
a
d
i
b
a
n
t
u
l
a
g
i
k
a
k
t
i
d
a
k
t
e
ri
m
a
k
a
s
i
h
m
b
a
k
S
a
t
i
sf
i
e
d
2
C
u
s
t
o
mer
-
2
P
e
rm
i
s
i
b
u
sa
y
a
i
n
g
i
n
m
e
l
a
k
u
k
a
n
p
e
m
b
a
y
a
r
a
n
a
t
a
s
n
a
m
a
w
i
n
d
a
ri
i
b
u
m
e
n
g
g
u
n
a
k
a
n
b
p
j
s
k
e
se
h
a
t
a
n
a
t
a
u
re
g
u
l
e
r
b
u
p
a
k
a
i
b
p
j
s
b
u
u
n
t
u
k
b
p
j
s k
e
se
h
a
t
a
n
b
i
s
a
d
i
l
a
y
a
n
i
d
i
k
a
si
r
se
b
e
l
a
h
y
a
b
u
si
n
i
k
h
u
s
u
s
y
a
n
g
r
e
g
u
l
e
r
D
i
ssat
i
sf
i
e
d
3
C
u
s
t
o
mer
-
3
G
i
m
a
n
a
t
a
n
g
a
n
sa
y
a
d
o
k
t
a
n
g
a
n
s
a
m
p
e
a
n
su
d
a
h
m
e
m
b
a
i
k
p
a
k
k
a
p
a
n
b
i
s
a
l
e
p
a
s
g
i
p
s
n
y
a
d
o
k
g
i
p
s
d
a
p
a
t
d
i
l
e
p
a
s s
e
k
i
t
a
r
d
u
a
m
i
n
g
g
u
l
a
g
i
p
a
k
b
a
p
a
k
d
i
b
a
n
y
a
k
i
n
m
a
k
a
n
i
k
a
n
i
k
a
n
a
n
y
a
p
a
k
b
i
a
r
c
e
p
e
t
p
u
l
i
h
t
a
n
g
a
n
n
y
a
s
i
a
p
d
o
k
t
e
ri
m
a
k
a
si
h
S
a
t
i
sf
i
e
d
4
C
u
s
t
o
mer
-
4
Pa
k
i
y
a
m
b
a
k
a
n
t
r
i
a
n
n
y
a
k
a
n
m
a
si
h
l
a
m
a
k
a
l
o
s
a
y
a
t
i
n
g
a
l
d
u
l
u
b
o
l
e
h
t
i
d
a
k
b
o
l
e
h
m
b
a
k
t
a
p
i
n
a
n
t
i
k
a
l
o
t
i
b
a
u
r
u
t
a
n
m
b
a
k
n
y
a
g
a
k
a
d
a
n
a
n
t
i
h
a
r
u
s m
e
n
g
u
l
a
n
g
d
a
ri
a
w
a
l
y
a
m
b
a
k
k
o
k
g
i
t
u
p
a
k
h
a
r
u
s
n
y
a
k
a
l
o
k
e
l
e
w
a
t
y
a
d
i
s
u
su
l
k
a
n
s
a
j
a
t
o
p
a
k
m
a
a
f
m
b
a
k
t
i
d
a
k
b
i
s
a
D
i
ssat
i
sf
i
e
d
5
C
u
s
t
o
mer
-
5
S
i
a
n
g
b
u
n
d
a
b
e
n
a
r
a
d
e
k
n
y
a
b
e
r
n
a
m
a
f
a
h
m
i
b
e
n
a
r
d
o
k
u
n
t
u
k
i
m
u
n
i
sa
s
i
y
a
i
y
a
d
o
k
a
p
a
k
a
h
n
a
n
t
i
b
a
d
a
n
y
a
p
a
n
a
s
d
o
k
b
u
n
d
a
g
a
k
p
e
rl
u
k
a
w
a
t
i
r
n
a
n
t
i
p
a
n
a
s
n
y
a
c
u
m
a
n
se
b
e
n
t
a
r k
o
k
b
i
s
a
d
i
k
o
m
p
res
b
i
a
r
c
e
p
e
t
re
d
a
p
a
n
a
s
n
y
a
t
e
ri
m
a
k
a
si
h
d
o
k
S
a
t
i
sf
i
e
d
3
.
2
.
Co
nv
er
s
a
t
io
n
a
na
ly
s
is
T
h
e
d
ata
u
s
ed
2
5
0
c
o
n
v
e
r
s
atio
n
s
,
wh
ich
wer
e
lab
eled
b
y
3
co
r
r
esp
o
n
d
en
ts
an
d
r
esu
lt
ed
in
1
2
5
"
s
atis
f
ied
"
co
n
v
er
s
atio
n
s
an
d
1
2
5
"
d
is
s
atis
f
ied
"
co
n
v
er
s
atio
n
s
.
T
h
e
c
o
n
v
e
r
s
atio
n
is
m
ad
e
i
n
to
a
d
ataset
with
a
co
m
p
o
s
itio
n
o
f
2
0
0
tr
ain
i
n
g
d
ata
an
d
5
0
test
in
g
d
ata,
th
e
n
p
r
ep
r
o
ce
s
s
in
g
u
s
in
g
th
e
NL
P
ap
p
r
o
ac
h
.
At
th
e
p
r
ep
r
o
ce
s
s
in
g
s
tag
e
th
e
wo
r
d
n
o
r
m
alize
r
,
s
tem
m
in
g
an
d
s
to
p
wo
r
d
r
em
o
v
al
f
ea
tu
r
es
ar
e
ap
p
lied
f
o
r
ea
ch
co
n
v
er
s
atio
n
.
T
ex
t
n
o
r
m
aliza
tio
n
is
a
ch
allen
g
e
to
f
in
d
th
e
s
am
e
wo
r
d
s
with
d
if
f
er
en
t
wo
r
d
v
a
r
iatio
n
s
[
2
5
]
.
W
o
r
d
n
o
r
m
alize
r
ca
n
h
an
d
le
v
a
r
iatio
n
s
o
f
wr
itin
g
wo
r
d
s
w
h
ich
h
av
e
th
e
s
am
e
m
ea
n
in
g
s
o
th
at
th
ey
ar
e
co
u
n
ted
as
a
s
in
g
le
ter
m
in
t
h
e
ca
lcu
latio
n
o
f
th
e
T
F
-
I
DF
al
g
o
r
ith
m
.
E
x
a
m
p
les
o
f
v
ar
iatio
n
s
in
wr
itin
g
wo
r
d
s
wh
ich
h
av
e
th
e
s
am
e
m
ea
n
in
g
ar
e
s
h
o
wn
i
n
T
ab
le
2
.
I
n
T
ab
le
2
s
h
o
ws
v
ar
i
atio
n
s
o
f
wo
r
d
s
wh
ich
h
av
e
th
e
s
am
e
m
ea
n
in
g
.
T
er
m
“
pak
”
,
“
bu
”
,
“
d
o
k
”
,
“
ka
k
”
,
“
d
ik
”
,
“
ma
s
”
an
d
“
mb
a
k
”
u
s
ed
to
r
ef
er
th
e
o
t
h
er
p
er
s
o
n
,
s
o
th
at
it
ca
n
b
e
r
ep
lace
d
with
ter
m
“
s
o
d
a
r
a
”.
T
er
m
“
s
ela
ma
t
s
o
r
e
”
,
“
s
ela
ma
t
s
ia
n
g
”
an
d
“
a
s
s
a
la
mu
a
la
iku
m
”
u
s
ed
t
o
g
i
v
e
g
r
ee
tin
g
s
s
o
th
at
it c
an
b
e
r
ep
l
ac
ed
with
ter
m
“
s
a
la
m
”.
T
h
e
n
ex
t stag
e
is
th
e
p
r
o
ce
s
s
o
f
s
tem
m
in
g
,
to
elim
in
ate
th
e
p
r
ef
ix
es,
in
s
er
ts
an
d
wo
r
d
s
u
f
f
ix
es so
th
at
it b
ec
o
m
es th
e
b
asic f
o
r
m
.
Ste
m
m
in
g
is
an
im
p
o
r
tan
t te
ch
n
i
q
u
e
in
NL
P f
o
r
ef
f
icien
t
an
d
ef
f
ec
tiv
e
in
f
o
r
m
atio
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
19
,
No
.
1
,
Feb
r
u
ar
y
2
0
2
1
:
1
4
6
-
1
5
4
150
r
etr
iev
al
[
2
6
]
.
An
ex
a
m
p
le
o
f
s
tem
m
in
g
ap
p
licatio
n
is
s
h
o
wn
in
T
ab
le
3
.
I
n
T
a
b
le
3
ter
m
s
“
d
ip
eriks
a
”
,
“
d
ip
eriks
a
ka
n
”
d
an
“
mem
eriks
a
”
o
m
itted
p
r
ef
ix
es,
i
n
s
er
ts
an
d
en
d
in
g
s
will
b
e
t
h
e
s
am
e
b
asic
ter
m
is
“
p
eriks
a
”.
T
er
m
s
“
men
g
a
n
t
a
r
”
an
d
“
d
ia
n
ta
r
”
af
ter
g
o
in
g
th
r
o
u
g
h
th
e
s
tem
m
in
g
p
r
o
ce
s
s
,
it
b
ec
o
m
es
t
h
e
s
am
e
b
asic
ter
m
is
“
a
n
ta
r
”.
Stem
m
in
g
m
ak
es a
wo
r
d
in
to
its
b
asic f
o
r
m
an
d
b
ec
o
m
es th
e
s
am
e
ter
m
.
T
ab
le
2
.
Var
iatio
n
s
o
f
wo
r
d
s
with
th
e
s
am
e
m
ea
n
in
g
No
Te
r
m
on
c
o
n
v
e
r
s
a
t
i
o
n
W
o
r
d
n
o
r
ma
l
i
z
e
r
r
e
s
u
l
t
s
1
Pa
k
sa
u
d
a
ra
2
Bu
sa
u
d
a
ra
3
D
o
k
sa
u
d
a
ra
4
K
a
k
sa
u
d
a
ra
5
D
i
k
sa
u
d
a
ra
6
Ma
s
sa
u
d
a
ra
7
Mb
a
k
sa
u
d
a
ra
8
S
e
l
a
m
a
t
S
o
re
sa
l
a
m
9
S
e
l
a
m
a
t
S
i
a
n
g
sa
l
a
m
10
Assa
l
a
m
u
a
l
a
i
k
u
m
sa
l
a
m
T
ab
le
3
.
T
h
e
r
esu
lts
o
f
ap
p
ly
in
g
s
tem
m
in
g
No
Te
r
m
o
n
c
o
n
v
e
r
s
a
t
i
o
n
S
t
e
mm
i
n
g
r
e
s
u
l
t
s
1
D
i
p
e
r
i
k
s
a
p
e
r
i
k
s
a
2
D
i
p
e
r
i
k
s
a
k
a
n
p
e
r
i
k
s
a
3
Me
m
e
r
i
k
s
a
p
e
r
i
k
s
a
4
Me
n
g
a
n
t
a
r
a
n
t
a
r
5
D
i
a
n
t
a
r
a
n
t
a
r
6
D
i
a
m
b
i
l
a
m
b
i
l
7
Me
n
g
a
m
b
i
l
a
m
b
i
l
8
D
i
t
u
n
g
g
u
t
u
n
g
g
u
9
Me
n
u
n
g
g
u
t
u
n
g
g
u
10
m
e
n
d
a
f
t
a
r
d
a
f
t
a
r
T
h
e
n
e
x
t
s
tep
is
th
e
s
to
p
wo
r
d
r
em
o
v
al
p
r
o
ce
s
s
to
r
em
o
v
e
t
h
e
s
to
p
wo
r
d
f
r
o
m
th
e
c
o
n
v
e
r
s
atio
n
.
T
h
e
s
to
p
wo
r
d
lis
t
wh
ich
is
u
s
ed
is
m
ad
e
b
y
its
elf
,
r
e
f
er
s
to
th
e
c
o
n
tex
t
o
f
w
o
r
d
s
wh
ich
a
r
e
o
f
t
en
u
s
ed
in
h
o
s
p
ital
s
er
v
ices
an
d
n
u
m
b
er
s
.
Nu
m
b
e
r
h
av
e
n
o
ef
f
ec
t
o
n
s
e
n
tim
en
t
an
aly
s
is
,
an
d
r
em
o
v
in
g
th
em
c
an
r
ed
u
ce
n
o
is
e
an
d
in
cr
ea
s
e
ef
f
icien
c
y
[
2
7
]
.
An
e
x
am
p
le
o
f
a
s
to
p
wo
r
d
lis
t
u
s
e
d
is
s
h
o
wn
in
T
ab
le
4
.
C
an
b
e
s
ee
n
in
T
ab
le
4
,
th
e
ter
m
"
ya
n
g
"
is
o
n
e
o
f
th
e
m
o
s
t
f
r
eq
u
en
tly
s
to
p
p
ed
wo
r
d
lis
ts
,
wh
ic
h
is
2
0
4
tim
es,
an
d
th
e
ter
m
"
di
"
wh
ich
ap
p
ea
r
s
1
0
1
tim
es.
Sto
p
wo
r
d
s
ca
n
b
e
r
e
m
o
v
e
d
b
ec
a
u
s
e
th
ey
u
s
u
ally
a
p
p
ea
r
in
lar
g
e
n
u
m
b
er
s
an
d
d
o
n
o
t
h
av
e
m
ea
n
in
g
as
a
s
in
g
le
ter
m
.
T
h
e
o
v
er
all
r
esu
lts
o
f
t
h
e
p
r
ep
r
o
ce
s
s
in
g
p
r
o
ce
s
s
ca
n
b
e
s
ee
n
in
T
ab
le
5
.
T
ab
le
5
s
h
o
ws
co
n
v
er
s
atio
n
s
wh
ich
h
av
e
p
ass
ed
p
r
ep
r
o
ce
s
s
in
g
u
s
in
g
th
e
NL
P
ap
p
r
o
ac
h
with
t
h
e
wo
r
d
n
o
r
m
alize
,
s
tem
m
in
g
an
d
s
to
p
wo
r
d
r
em
o
v
al
f
ea
tu
r
es.
On
e
in
d
icato
r
o
f
s
atis
f
ied
cu
s
to
m
er
s
is
to
s
ay
"
t
erima
ka
s
ih
"
at
th
e
en
d
o
f
th
e
c
o
n
v
e
r
s
atio
n
wh
ich
in
d
icate
s
th
at
cu
s
to
m
er
s
f
ee
l
h
elp
ed
an
d
ar
e
s
atis
f
ied
with
th
e
s
er
v
ices
p
r
o
v
i
d
ed
.
I
n
co
n
v
er
s
atio
n
n
u
m
b
er
2
,
th
er
e
is
a
m
is
tak
e
o
f
th
e
c
u
s
to
m
er
b
ec
a
u
s
e
o
f
th
e
wr
o
n
g
c
h
o
ice
o
f
th
e
ca
s
h
ier
,
b
u
t
cu
s
to
m
e
r
s
d
is
ap
p
o
in
ted
ca
u
s
e
is
th
at
n
o
wo
r
d
s
o
f
f
o
r
g
iv
en
ess
m
ak
e
c
u
s
to
m
er
s
f
ee
l
co
m
f
o
r
ta
b
le.
Ap
o
lo
g
y
d
ea
ls
with
t
h
e
r
elatio
n
s
h
ip
b
etwe
en
in
d
iv
i
d
u
als
in
v
o
lv
ed
i
n
th
e
wr
o
n
g
s
itu
atio
n
[
2
8
]
.
T
h
e
ch
o
ice
an
d
wo
r
d
u
s
ag
e
in
p
u
b
lic
s
er
v
ices
af
f
ec
t
th
e
lev
el
o
f
c
u
s
to
m
er
s
atis
f
ac
tio
n
,
alth
o
u
g
h
s
er
v
ice
p
er
s
o
n
n
el
m
u
s
t
b
e
s
tr
ict
with
th
e
r
u
les,
with
th
e
r
ig
h
t c
h
o
ice
o
f
w
o
r
d
s
ca
n
m
ain
tain
c
u
s
to
m
er
f
ee
lin
g
s
.
T
ab
le
4.
E
x
am
p
le
s
to
p
wo
r
d
lis
t
No
S
t
o
p
w
o
r
d
l
i
s
t
F
r
e
q
u
e
n
c
y
o
f
a
p
p
e
a
r
a
n
c
e
1
y
a
n
g
2
0
4
2
di
10
1
3
t
a
p
i
9
2
4
d
a
n
7
3
5
ya
6
1
6
j
u
g
a
6
0
7
j
a
d
i
3
6
8
u
n
t
u
k
3
0
9
d
e
n
g
a
n
3
0
10
ke
2
3
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
S
en
timen
ts
a
n
a
lysi
s
o
f c
u
s
to
mer sa
tis
fa
ctio
n
in
p
u
b
lic
s
ervices u
s
in
g
… (
E
lik
Ha
r
i Mu
kta
fin
)
151
T
ab
le
5.
Pre
p
r
o
ce
s
s
in
g
r
esu
lts
No
C
o
n
v
e
r
sa
t
i
o
n
P
r
e
p
r
o
c
e
ss
i
n
g
r
e
su
l
t
s
W
o
r
d
N
o
r
ma
l
i
z
e
r
S
t
e
mm
i
n
g
S
t
o
p
w
o
r
d
R
e
m
o
v
a
l
1
Assa
l
a
m
u
a
l
a
i
k
u
m
a
d
a
y
a
n
g
b
i
s
a
s
a
y
a
b
a
n
t
u
k
a
k
sa
y
a
i
n
g
i
n
m
e
n
e
b
u
s
r
e
se
p
i
n
i
m
b
a
k
si
l
a
h
k
a
n
d
u
d
u
k
d
u
l
u
k
a
k
s
a
y
a
a
k
a
n
m
e
n
y
i
a
p
k
a
n
n
y
a
i
n
i
o
b
a
t
n
y
a
k
a
k
y
a
n
g
t
a
b
l
e
t
h
i
j
a
u
d
i
m
i
n
u
m
4
5
m
e
n
i
t
se
b
e
l
u
m
m
a
k
a
n
t
i
g
a
k
a
l
i
se
h
a
r
i
y
a
n
g
k
a
p
su
l
b
i
r
u
d
a
n
p
u
t
i
h
m
a
si
n
g
m
a
si
n
g
sa
t
u
se
t
e
l
a
h
m
a
k
a
n
t
i
g
a
k
a
l
i
s
e
h
a
r
i
a
d
a
y
a
n
g
b
i
s
a
d
i
b
a
n
t
u
l
a
g
i
k
a
k
t
i
d
a
k
t
e
ri
m
a
k
a
s
i
h
m
b
a
k
sa
l
a
m
a
d
a
y
a
n
g
b
i
s
a
s
a
y
a
b
a
n
t
u
o
b
j
e
k
s
a
y
a
i
n
g
i
n
m
e
n
e
b
u
s
r
e
se
p
i
n
i
o
b
j
e
k
si
l
a
h
k
a
n
d
u
d
u
k
d
u
l
u
o
b
j
e
k
s
a
y
a
a
k
a
n
m
e
n
y
i
a
p
k
a
n
n
y
a
i
n
i
o
b
a
t
n
y
a
o
b
j
e
k
y
a
n
g
t
a
b
l
e
t
h
i
j
a
u
d
i
m
i
n
u
m
4
5
m
e
n
i
t
se
b
e
l
u
m
m
a
k
a
n
t
i
g
a
k
a
l
i
se
h
a
r
i
y
a
n
g
k
a
p
su
l
b
i
r
u
d
a
n
p
u
t
i
h
m
a
si
n
g
m
a
si
n
g
sa
t
u
se
t
e
l
a
h
m
a
k
a
n
t
i
g
a
k
a
l
i
s
e
h
a
r
i
a
d
a
y
a
n
g
b
i
s
a
d
i
b
a
n
t
u
l
a
g
i
o
b
j
e
k
t
i
d
a
k
t
e
r
i
m
a
k
a
s
i
h
o
b
j
e
k
sa
l
a
m
a
d
a
y
a
n
g
b
i
s
a
s
a
y
a
b
a
n
t
u
o
b
j
e
k
s
a
y
a
i
n
g
i
n
t
e
b
u
s
rese
p
i
n
i
o
b
j
e
k
si
l
a
h
d
u
d
u
k
d
u
l
u
o
b
j
e
k
sa
y
a
a
k
a
n
si
a
p
i
n
i
o
b
a
t
o
b
j
e
k
y
a
n
g
t
a
b
l
e
t
h
i
j
a
u
m
i
n
u
m
4
5
m
e
n
i
t
b
e
l
u
m
m
a
k
a
n
t
i
g
a
k
a
l
i
h
a
ri
y
a
n
g
k
a
p
su
l
b
i
r
u
d
a
n
p
u
t
i
h
m
a
si
n
g
m
a
si
n
g
sa
t
u
t
e
l
a
h
m
a
k
a
n
t
i
g
a
k
a
l
i
h
a
ri
a
d
a
y
a
n
g
b
i
sa
b
a
n
t
u
l
a
g
i
o
b
j
e
k
t
i
d
a
k
t
e
ri
m
a
k
a
si
h
o
b
j
e
k
sa
l
a
m
a
d
a
b
i
sa
sa
y
a
b
a
n
t
u
o
b
j
e
k
s
a
y
a
t
e
b
u
s
res
e
p
o
b
j
e
k
si
l
a
h
d
u
d
u
k
o
b
j
e
k
sa
y
a
s
i
a
p
o
b
a
t
o
b
j
e
k
t
a
b
l
e
t
h
i
j
a
u
m
i
n
u
m
b
e
l
u
m
m
a
k
a
n
k
a
p
su
l
b
i
r
u
p
u
t
i
h
m
a
k
a
n
a
d
a
b
i
s
a
b
a
n
t
u
o
b
j
e
k
t
i
d
a
k
t
e
ri
m
a
k
a
s
i
h
o
b
j
e
k
2
P
e
rm
i
s
i
b
u
sa
y
a
i
n
g
i
n
m
e
l
a
k
u
k
a
n
p
e
m
b
a
y
a
r
a
n
a
t
a
s
n
a
m
a
w
i
n
d
a
r
i
i
b
u
m
e
n
g
g
u
n
a
k
a
n
b
p
j
s
k
e
s
e
h
a
t
a
n
a
t
a
u
re
g
u
l
e
r
b
u
p
a
k
a
i
b
p
j
s
b
u
u
n
t
u
k
b
p
j
s
k
e
s
e
h
a
t
a
n
b
i
s
a
d
i
l
a
y
a
n
i
d
i
k
a
s
i
r s
e
b
e
l
a
h
y
a
b
u
si
n
i
k
h
u
su
s y
a
n
g
re
g
u
l
e
r
sa
l
a
m
o
b
j
e
k
s
a
y
a
i
n
g
i
n
m
e
l
a
k
u
k
a
n
p
e
m
b
a
y
a
r
a
n
a
t
a
s
n
a
m
a
w
i
n
d
a
r
i
o
b
j
e
k
m
e
n
g
g
u
n
a
k
a
n
b
p
j
s
k
e
s
e
h
a
t
a
n
a
t
a
u
re
g
u
l
e
r
o
b
j
e
k
p
a
k
a
i
b
p
j
s
o
b
j
e
k
u
n
t
u
k
b
p
j
s
k
e
s
e
h
a
t
a
n
b
i
s
a
d
i
l
a
y
a
n
i
d
i
k
a
s
i
r s
e
b
e
l
a
h
y
a
o
b
j
e
k
si
n
i
k
h
u
su
s
y
a
n
g
re
g
u
l
e
r
sa
l
a
m
o
b
j
e
k
s
a
y
a
i
n
g
i
n
l
a
k
u
b
a
y
a
r
a
t
a
s
n
a
m
a
w
i
n
d
a
r
i
o
b
j
e
k
g
u
n
a
b
p
j
s s
e
h
a
t
a
t
a
u
reg
u
l
e
r
o
b
j
e
k
p
a
k
a
i
b
p
j
s
o
b
j
e
k
u
n
t
u
k
b
p
j
s s
e
h
a
t
b
i
s
a
l
a
y
a
n
d
i
k
a
si
r
b
e
l
a
h
y
a
o
b
j
e
k
si
n
i
k
h
u
su
s y
a
n
g
re
g
u
l
e
r
sa
l
a
m
o
b
j
e
k
s
a
y
a
l
a
k
u
b
a
y
a
r
n
a
m
a
w
i
n
d
a
ri
o
b
j
e
k
g
u
n
a
b
p
j
s s
e
h
a
t
reg
u
l
e
r
o
b
j
e
k
p
a
k
a
i
b
p
j
s
o
b
j
e
k
b
p
j
s s
e
h
a
t
b
i
s
a
l
a
y
a
n
k
a
si
r
b
e
l
a
h
o
b
j
e
k
k
h
u
su
s
re
g
u
l
e
r
3
G
i
m
a
n
a
t
a
n
g
a
n
sa
y
a
d
o
k
t
a
n
g
a
n
s
a
m
p
e
a
n
s
u
d
a
h
m
e
m
b
a
i
k
p
a
k
k
a
p
a
n
b
i
s
a
l
e
p
a
s
g
i
p
s
n
y
a
d
o
k
g
i
p
s
d
a
p
a
t
d
i
l
e
p
a
s
sek
i
t
a
r
d
u
a
m
i
n
g
g
u
l
a
g
i
p
a
k
b
a
p
a
k
d
i
b
a
n
y
a
k
i
n
m
a
k
a
n
i
k
a
n
i
k
a
n
a
n
y
a
p
a
k
b
i
a
r
c
e
p
e
t
p
u
l
i
h
t
a
n
g
a
n
n
y
a
si
a
p
d
o
k
t
e
r
i
m
a
k
a
s
i
h
b
a
g
a
i
m
a
n
a
t
a
n
g
a
n
s
a
y
a
o
b
j
e
k
t
a
n
g
a
n
o
b
j
e
k
s
u
d
a
h
m
e
m
b
a
i
k
o
b
j
e
k
k
a
p
a
n
b
i
sa
l
e
p
a
s
g
i
p
s
n
y
a
o
b
j
e
k
g
i
p
s
d
a
p
a
t
d
i
l
e
p
a
s s
e
k
i
t
a
r
d
u
a
m
i
n
g
g
u
l
a
g
i
o
b
j
e
k
o
b
j
e
k
d
i
b
a
n
y
a
k
i
n
m
a
k
a
n
i
k
a
n
i
k
a
n
a
n
y
a
o
b
j
e
k
b
i
a
r
c
e
p
e
t
p
u
l
i
h
t
a
n
g
a
n
n
y
a
si
a
p
o
b
j
e
k
t
e
ri
m
a
k
a
s
i
h
b
a
g
a
i
m
a
n
a
t
a
n
g
a
n
s
a
y
a
o
b
j
e
k
t
a
n
g
a
n
o
b
j
e
k
s
u
d
a
h
b
a
i
k
o
b
j
e
k
k
a
p
a
n
b
i
s
a
l
e
p
a
s
g
i
p
s
o
b
j
e
k
g
i
p
s
d
a
p
a
t
l
e
p
a
s
sek
i
t
a
r
d
u
a
m
i
n
g
g
u
l
a
g
i
o
b
j
e
k
o
b
j
e
k
b
a
n
y
a
k
m
a
k
a
n
i
k
a
n
i
k
a
n
y
a
o
b
j
e
k
b
i
a
r c
e
p
e
t
p
u
l
i
h
t
a
n
g
a
n
s
i
a
p
o
b
j
e
k
t
e
r
i
m
a
k
a
s
i
h
t
a
n
g
a
n
s
a
y
a
o
b
j
e
k
t
a
n
g
a
n
o
b
j
e
k
s
u
d
a
h
b
a
i
k
o
b
j
e
k
k
a
p
a
n
b
i
s
a
l
e
p
a
s
g
i
p
s
o
b
j
e
k
g
i
p
s
d
a
p
a
t
l
e
p
a
s s
e
k
i
t
a
r
o
b
j
e
k
o
b
j
e
k
b
a
n
y
a
k
m
a
k
a
n
i
k
a
n
i
k
a
n
o
b
j
e
k
c
e
p
e
t
p
u
l
i
h
t
a
n
g
a
n
s
i
a
p
o
b
j
e
k
t
e
ri
m
a
k
a
s
i
h
4
Pa
k
i
y
a
m
b
a
k
a
n
t
r
i
a
n
n
y
a
k
a
n
m
a
si
h
l
a
m
a
k
a
l
o
sa
y
a
t
i
n
g
a
l
d
u
l
u
b
o
l
e
h
t
i
d
a
k
b
o
l
e
h
m
b
a
k
t
a
p
i
n
a
n
t
i
k
a
l
o
t
i
b
a
u
ru
t
a
n
m
b
a
k
n
y
a
g
a
k
a
d
a
n
a
n
t
i
h
a
ru
s
m
e
n
g
u
l
a
n
g
d
a
r
i
a
w
a
l
y
a
m
b
a
k
k
o
k
g
i
t
u
p
a
k
h
a
r
u
sn
y
a
k
a
l
o
k
e
l
e
w
a
t
y
a
d
i
s
u
s
u
l
k
a
n
s
a
j
a
t
o
p
a
k
m
a
a
f
m
b
a
k
t
i
d
a
k
b
i
s
a
o
b
j
e
k
i
y
a
o
b
j
e
k
a
n
t
ri
a
n
n
y
a
k
a
n
m
a
si
h
l
a
m
a
k
a
l
o
sa
y
a
t
i
n
g
a
l
d
u
l
u
b
o
l
e
h
t
i
d
a
k
b
o
l
e
h
o
b
j
e
k
t
a
p
i
n
a
n
t
i
k
a
l
o
t
i
b
a
u
ru
t
a
n
o
b
j
e
k
t
i
d
a
k
a
d
a
n
a
n
t
i
h
a
r
u
s
m
e
n
g
u
l
a
n
g
d
a
r
i
a
w
a
l
y
a
o
b
j
e
k
k
o
k
g
i
t
u
o
b
j
e
k
h
a
ru
s
n
y
a
k
a
l
o
k
e
l
e
w
a
t
y
a
d
i
s
u
s
u
l
k
a
n
s
a
j
a
t
o
o
b
j
e
k
m
a
a
f
o
b
j
e
k
t
i
d
a
k
b
i
s
a
o
b
j
e
k
i
y
a
o
b
j
e
k
a
n
t
ri
k
a
n
m
a
si
h
l
a
m
a
k
a
l
o
sa
y
a
t
i
n
g
a
l
d
u
l
u
b
o
l
e
h
t
i
d
a
k
b
o
l
e
h
o
b
j
e
k
t
a
p
i
n
a
n
t
i
k
a
l
o
t
i
b
a
u
ru
t
o
b
j
e
k
t
i
d
a
k
a
d
a
n
a
n
t
i
h
a
r
u
s
u
l
a
n
g
d
a
ri
a
w
a
l
y
a
o
b
j
e
k
k
o
k
g
i
t
u
o
b
j
e
k
h
a
r
u
s k
a
l
o
l
e
w
a
t
y
a
s
u
s
u
l
sa
j
a
t
o
o
b
j
e
k
m
a
a
f
o
b
j
e
k
t
i
d
a
k
b
i
sa
o
b
j
e
k
i
y
a
o
b
j
e
k
a
n
t
ri
m
a
si
h
l
a
m
a
s
a
y
a
t
i
n
g
a
l
b
o
l
e
h
t
i
d
a
k
b
o
l
e
h
o
b
j
e
k
t
a
p
i
k
a
l
o
t
i
b
a
u
r
u
t
o
b
j
e
k
t
i
d
a
k
a
d
a
h
a
ru
s
u
l
a
n
g
a
w
a
l
o
b
j
e
k
g
i
t
u
o
b
j
e
k
h
a
r
u
s
l
e
w
a
t
s
u
s
u
l
sa
j
a
o
b
j
e
k
m
a
a
f
o
b
j
e
k
t
i
d
a
k
b
i
s
a
5
S
i
a
n
g
b
u
n
d
a
b
e
n
a
r
a
d
e
k
n
y
a
b
e
r
n
a
m
a
f
a
h
m
i
b
e
n
a
r
d
o
k
u
n
t
u
k
i
m
u
n
i
s
a
si
y
a
i
y
a
d
o
k
a
p
a
k
a
h
n
a
n
t
i
b
a
d
a
n
y
a
p
a
n
a
s
d
o
k
b
u
n
d
a
g
a
k
p
e
r
l
u
k
a
w
a
t
i
r
n
a
n
t
i
p
a
n
a
sn
y
a
c
u
m
a
n
s
e
b
e
n
t
a
r k
o
k
b
i
s
a
d
i
k
o
m
p
res
b
i
a
r c
e
p
e
t
re
d
a
p
a
n
a
sn
y
a
t
e
ri
m
a
k
a
s
i
h
d
o
k
si
a
n
g
o
b
j
e
k
b
e
n
a
r
a
d
e
k
n
y
a
b
e
r
n
a
m
a
f
a
h
m
i
b
e
n
a
r
o
b
j
e
k
u
n
t
u
k
i
m
u
n
i
s
a
s
i
y
a
i
y
a
o
b
j
e
k
a
p
a
k
a
h
n
a
n
t
i
b
a
d
a
n
y
a
p
a
n
a
s
o
b
j
e
k
o
b
j
e
k
t
i
d
a
k
p
e
r
l
u
k
a
w
a
t
i
r
n
a
n
t
i
p
a
n
a
s
n
y
a
c
u
m
a
n
se
b
e
n
t
a
r
k
o
k
b
i
s
a
d
i
k
o
m
p
res
b
i
a
r
c
e
p
e
t
red
a
p
a
n
a
sn
y
a
t
e
ri
m
a
k
a
s
i
h
o
b
j
e
k
si
a
n
g
o
b
j
e
k
b
e
n
a
r
a
d
e
k
n
a
m
a
f
a
h
m
i
b
e
n
a
r
o
b
j
e
k
u
n
t
u
k
i
m
u
n
i
s
a
si
y
a
i
y
a
o
b
j
e
k
a
p
a
n
a
n
t
i
b
a
d
a
n
p
a
n
a
s
o
b
j
e
k
o
b
j
e
k
t
i
d
a
k
p
e
r
l
u
k
a
w
a
t
i
r
n
a
n
t
i
p
a
n
a
s
c
u
m
a
n
se
b
e
n
t
a
r
k
o
k
b
i
s
a
k
o
m
p
res
b
i
a
r
c
e
p
e
t
red
a
p
a
n
a
s
t
e
r
i
m
a
k
a
si
h
o
b
j
e
k
si
a
n
g
o
b
j
e
k
b
e
n
a
r
a
d
e
k
n
a
m
a
f
a
h
m
i
b
e
n
a
r
o
b
j
e
k
i
m
u
n
i
s
a
si
o
b
j
e
k
a
p
a
b
a
d
a
n
p
a
n
a
s
o
b
j
e
k
o
b
j
e
k
t
i
d
a
k
p
e
r
l
u
k
a
w
a
t
i
r
n
a
n
t
i
p
a
n
a
s
c
u
m
a
n
seb
e
n
t
a
r
b
i
s
a
k
o
m
p
res
c
e
p
e
t
r
e
d
a
p
a
n
a
s
t
e
ri
m
a
k
a
s
i
h
o
b
j
e
k
3
.
3
.
K
NN
c
a
lcula
t
io
n
us
ing
TF
-
I
DF
T
h
is
s
tu
d
y
u
s
es
th
e
KNN
alg
o
r
ith
m
with
k
=3
an
d
th
e
ter
m
weig
h
tin
g
u
s
es
T
F
-
I
DF
.
TF
-
I
DF
r
ef
lect
s
th
e
im
p
o
r
tan
ce
o
f
a
wo
r
d
in
a
tex
t
d
o
cu
m
en
t
[
2
9
]
.
T
h
e
T
F
-
I
DF
ca
lcu
latio
n
is
ap
p
lied
to
2
0
0
tr
ain
i
n
g
d
ata
an
d
5
0
test
in
g
d
ata
af
ter
p
r
ep
r
o
ce
s
s
in
g
u
s
in
g
th
e
NL
P
ap
p
r
o
ac
h
.
T
h
e
n
ex
t
s
tep
is
ca
lcu
latin
g
th
e
n
u
m
b
er
o
f
m
atch
es
lab
elin
g
r
esu
lts
f
r
o
m
th
e
p
r
e
d
i
ctio
n
o
f
5
0
test
in
g
d
ata
co
m
p
a
r
ed
with
m
an
u
al
lab
elin
g
b
y
c
o
r
r
esp
o
n
d
en
ts
u
s
in
g
co
n
f
u
s
io
n
m
atr
i
x
.
T
h
e
co
n
f
u
s
io
n
m
atr
ix
d
iv
id
es
th
e
r
esu
lts
o
f
th
e
p
r
ed
ictio
n
s
in
to
f
o
u
r
c
ateg
o
r
ies,
th
ey
ar
e
;
1)
tr
u
e
p
o
s
i
tiv
e
(
T
P)
s
h
o
ws
th
e
am
o
u
n
t
o
f
d
ata
with
p
o
s
itiv
e
c
lass
es
an
d
tr
u
e
p
r
ed
ictiv
e
r
esu
l
ts
,
2)
tr
u
e
n
eg
ativ
e
(
T
N)
s
h
o
ws
th
e
am
o
u
n
t
o
f
d
ata
with
n
eg
ativ
e
class
e
s
an
d
tr
u
e
p
r
ed
ictiv
e
r
esu
lts
,
3
)
f
alse
p
o
s
itiv
e
(
FP
)
s
h
o
ws
th
e
n
u
m
b
er
o
f
d
ata
with
p
o
s
itiv
e
class
e
s
an
d
f
alse
,
an
d
4
)
f
alse
n
eg
ativ
e
p
r
ed
ictio
n
r
esu
lts
(
FN)
s
h
o
ws
th
e
n
u
m
b
er
o
f
d
ata
with
n
eg
ativ
e
class
an
d
wr
o
n
g
p
r
ed
ictio
n
r
esu
lts
.
Pre
d
ictio
n
r
esu
lts
ar
e
s
h
o
wn
in
T
ab
le
6
.
T
h
e
r
esu
lts
o
f
t
h
e
co
n
f
u
s
io
n
m
atr
ix
ar
e
p
er
f
o
r
m
ed
ca
lcu
latio
n
s
to
g
et
th
e
v
alu
e
o
f
ac
cu
r
ac
y
,
p
r
ec
is
io
n
an
d
r
e
ca
ll.
Acc
u
r
ac
y
is
th
e
p
r
o
p
o
r
tio
n
o
f
th
e
to
tal
n
u
m
b
e
r
o
f
co
r
r
ec
t p
r
ed
ictio
n
s
co
m
p
a
r
ed
to
th
e
to
tal
n
u
m
b
e
r
o
f
s
am
p
les,
ac
c
u
r
ac
y
is
ca
lc
u
lated
u
s
in
g
(
2
)
[
3
0
]
.
P
r
e
c
i
s
i
o
n
i
s
t
h
e
p
r
o
p
o
r
t
i
o
n
o
f
p
o
s
i
t
i
v
e
s
a
m
p
l
e
s
w
h
i
c
h
a
r
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
19
,
No
.
1
,
Feb
r
u
ar
y
2
0
2
1
:
1
4
6
-
1
5
4
152
c
o
r
r
e
c
t
l
y
c
l
a
s
s
i
f
i
e
d
a
g
a
i
n
s
t
t
h
e
p
o
s
i
t
i
v
e
t
o
t
a
l
n
u
m
b
e
r
o
f
s
a
m
p
l
e
s
,
p
r
e
c
i
s
i
o
n
i
s
c
a
l
c
u
l
a
t
e
d
u
s
i
n
g
(
3
)
.
A
r
e
c
a
l
l
i
s
a
p
o
s
i
t
i
v
e
s
a
m
p
l
e
t
h
a
t
i
s
c
l
a
s
s
i
f
i
e
d
c
o
r
r
e
c
t
l
y
t
o
t
h
e
t
o
t
a
l
n
u
m
b
e
r
o
f
p
o
s
i
t
i
v
e
s
a
m
p
l
e
s
,
r
e
c
a
l
l
i
s
c
a
l
c
u
l
a
t
e
d
u
s
i
n
g
(
4
)
[
3
1
]
.
=
+
+
+
+
(
2
)
=
+
(
3
)
=
+
(
4
)
In
(
2
-
4
)
a
r
e
ap
p
lied
to
th
e
test
d
ata
test
in
g
r
esu
lts
an
d
th
e
r
es
u
lts
ar
e
as sh
o
wn
in
T
ab
le
7
.
T
h
e
m
ea
s
u
r
em
e
n
t
r
esu
lts
in
T
a
b
le
7
s
h
o
w
t
h
at
lab
elin
g
u
s
in
g
KNN
an
d
T
F
-
I
DF
with
th
e
NL
P
ap
p
r
o
ac
h
p
r
o
d
u
ce
s
an
ac
cu
r
ac
y
7
4
.
0
0
%,
p
r
ec
is
io
n
7
6
.
0
0
%
an
d
r
ec
all
7
3
.
0
8
%.
Ov
er
all
th
e
test
r
esu
l
ts
s
h
o
w
in
g
p
o
s
itiv
e
r
esu
lts
s
h
o
w
m
ac
h
in
e
lear
n
i
n
g
h
as
g
o
o
d
p
er
f
o
r
m
an
ce
.
Pre
v
i
o
u
s
r
esear
ch
[
9
]
u
s
in
g
KNN
an
d
T
F
-
I
DF
alg
o
r
ith
m
s
in
cla
s
s
if
y
in
g
s
en
tim
en
ts
o
n
T
witter
s
o
cial
m
ed
ia
o
b
tain
e
d
an
ac
c
u
r
ac
y
o
f
6
7
.
2
%
,
c
o
m
p
ar
ed
t
o
th
is
s
tu
d
y
o
b
tain
ed
a
h
ig
h
er
ac
cu
r
ac
y
o
f
7
4
.
0
0
%.
Hig
h
e
r
ac
cu
r
ac
y
v
alu
es
in
d
icate
th
at
th
e
u
s
e
o
f
NL
P
f
o
r
p
r
ep
r
o
ce
s
s
in
g
p
r
o
v
id
es a
n
in
cr
ea
s
e
in
lab
elin
g
ac
cu
r
ac
y
.
T
ab
le
6
.
Hasil
C
o
n
f
u
s
io
n
m
at
r
ix
P
r
e
d
i
c
t
i
o
n
r
e
su
l
t
s
N
u
mb
e
r
o
f
d
o
c
u
m
e
n
t
s
Tr
u
e
P
o
s
i
t
i
f
(
TP)
19
Tr
u
e
N
e
g
a
t
i
f
(
TN
)
18
F
a
l
se
P
o
s
i
t
i
f
(
F
P
)
6
F
a
l
se
N
e
g
a
t
i
f
(
F
N
)
7
T
ab
le
7
.
L
a
b
elin
g
test
r
esu
lts
M
e
a
su
r
e
me
n
t
M
e
a
su
r
e
me
n
t
r
e
su
l
t
s
A
c
c
u
r
a
c
y
7
4
.
0
0
%
P
r
e
c
i
s
i
o
n
76
.
0
0
%
R
e
c
a
l
l
7
3
.
08
%
3
.
4
.
T
er
m
o
cc
urre
nce
f
re
qu
ency
T
er
m
s
wh
ich
h
as
a
h
ig
h
f
r
eq
u
en
cy
o
f
o
cc
u
r
r
en
ce
in
c
o
n
v
e
r
s
atio
n
ca
n
d
escr
ib
e
m
a
r
k
et
ac
c
ep
tan
ce
o
f
s
er
v
ice
q
u
ality
.
T
h
e
ter
m
s
o
f
t
en
u
s
ed
in
co
n
v
er
s
atio
n
s
ar
e
in
d
icate
d
b
y
wo
r
d
co
u
d
cr
ea
ted
with
th
e
“W
o
r
d
ar
t
”
ap
p
licatio
n
i
n
Fig
u
r
e
4
an
d
Fig
u
r
e
5
.
I
n
c
o
n
v
e
r
s
atio
n
s
w
ith
th
e
lab
el
"
s
atis
f
ied
",
th
e
ter
m
wh
ich
h
as
th
e
h
i
g
h
est
f
r
eq
u
e
n
cy
o
f
o
cc
u
r
r
en
ce
is
th
e
ter
m
“
a
ka
n
”
in
1
4
3
tim
es
,
ter
m
“
in
g
in
”
in
9
8
tim
es
,
ter
m
“
a
p
a
ka
h
”
in
7
8
tim
es
,
ter
m
“
b
a
ik
”
i
n
5
4
tim
es
,
ter
m
“
teri
ma
”
in
5
2
tim
es
an
d
ter
m
“
ka
s
ih
”
in
5
1
tim
es
.
T
er
m
“
a
p
a
ka
h
”
an
d
“
in
g
in
”
r
ef
er
to
c
u
s
to
m
er
q
u
esti
o
n
s
an
d
d
esire
s
,
wh
ile
ter
m
s
“
b
a
ik
”,
“
teri
ma
”
an
d
“
ka
s
ih
”
r
ef
e
r
s
to
c
u
s
to
m
er
s
atis
f
ac
tio
n
b
ec
au
s
e
q
u
esti
o
n
s
o
r
d
esire
s
ca
n
b
e
r
ea
lized
.
I
n
co
n
v
er
s
atio
n
s
with
t
h
e
lab
el
"
s
atis
f
ied
",
th
e
ter
m
th
at
h
as
t
h
e
h
ig
h
est
f
r
eq
u
en
cy
o
f
o
cc
u
r
r
en
ce
is
th
e
ter
m
“
tid
a
k
-
b
is
a
”
in
1
3
8
tim
es
,
ter
m
“
la
ma
”
i
n
1
0
5
tim
es
,
ter
m
“
a
p
a
”
in
7
5
tim
es
,
te
r
m
“
B
P
JS
”
in
7
0
tim
es
an
d
ter
m
“
kira
”
in
55
tim
es
.
T
er
m
“
apa
”
a
n
d
“
tid
a
k
-
b
is
a
”
s
h
o
ws
cu
s
to
m
er
s
d
o
n
o
t
g
et
wh
a
t
is
ex
p
ec
ted
.
T
er
m
“
la
ma
”
an
d
“
kira
”
s
h
o
ws
th
at
th
e
cu
s
to
m
er
is
n
o
t
co
m
f
o
r
ta
b
le
waitin
g
.
T
er
m
“
B
P
JS
”
s
h
o
ws
th
e
u
s
e
o
f
B
P
J
S
f
o
r
p
a
y
m
en
t
o
f
s
er
v
ice
f
ee
s
.
Fig
u
r
e
4.
W
o
r
d
c
lo
u
d
ap
p
ea
r
a
n
ce
o
f
wo
r
d
s
in
a
s
atis
f
ied
co
n
v
er
s
atio
n
Fig
u
r
e
5.
W
o
r
d
clo
u
d
th
e
ap
p
e
ar
an
ce
o
f
wo
r
d
s
in
a
co
n
v
er
s
atio
n
is
n
o
t satis
f
ied
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
S
en
timen
ts
a
n
a
lysi
s
o
f c
u
s
to
mer sa
tis
fa
ctio
n
in
p
u
b
lic
s
ervices u
s
in
g
… (
E
lik
Ha
r
i Mu
kta
fin
)
153
4.
CO
NCLU
SI
O
N
Sen
tim
en
t
an
aly
s
is
o
n
p
u
b
lic
s
er
v
ice
co
n
v
e
r
s
atio
n
s
ca
n
p
r
o
v
id
e
d
ee
p
er
in
f
o
r
m
atio
n
ab
o
u
t
cu
s
to
m
er
r
atin
g
s
o
n
s
er
v
ices
p
r
o
v
id
ed
.
Utilizatio
n
o
f
th
e
s
p
ee
ch
r
e
co
g
n
itio
n
s
y
s
tem
tech
n
o
lo
g
y
ca
n
co
n
v
er
t
v
o
ice
co
n
v
er
s
atio
n
s
in
to
tex
t,
th
u
s
h
elp
in
g
th
e
p
r
o
ce
s
s
o
f
an
aly
zin
g
cu
s
to
m
er
s
atis
f
ac
tio
n
.
C
lass
if
icatio
n
u
s
in
g
th
e
KNN
an
d
T
F
-
I
DF
alg
o
r
ith
m
with
o
p
tim
izatio
n
u
s
in
g
th
e
NL
P
ap
p
r
o
ac
h
at
th
e
p
r
e
p
r
o
ce
s
s
in
g
s
tag
e
ca
n
r
esu
lt
in
b
etter
lab
elin
g
o
f
th
e
c
o
n
v
e
r
s
atio
n
d
ata.
T
esti
n
g
with
co
n
f
u
s
io
n
m
atr
ix
p
r
o
d
u
ce
s
a
v
alu
e
7
4
.
0
0
%
o
f
ac
cu
r
ac
y
,
7
6
.
0
0
%
o
f
p
r
ec
is
io
n
an
d
7
3
.
0
8
%
o
f
r
ec
all.
T
h
is
r
esear
ch
f
o
u
n
d
th
at
ter
m
s
ar
e
d
o
m
in
an
t
in
co
n
v
er
s
atio
n
“
s
atis
f
ied
”
is
ter
m
“
a
ka
n
”
,
“
in
g
in
”
,
“
a
p
a
ka
h
”
,
“
b
a
ik
”
,
“
teri
ma
”
an
d
“
ka
s
ih
”
wh
ic
h
s
h
o
ws
cu
s
to
m
er
s
atis
f
ac
tio
n
o
n
th
e
s
e
r
v
ice
r
ec
eiv
ed
an
d
cu
s
to
m
er
d
esire
s
th
at
ca
n
b
e
r
ea
lized
.
W
h
er
ea
s
th
e
d
o
m
in
an
t
ter
m
s
in
th
e
co
n
v
er
s
atio
n
“
d
is
s
ati
s
f
ied
”
is
ter
m
“
tid
a
k
-
b
is
a
”
,
“
l
a
ma
”
,
“
apa
”
,
“
B
P
JS
”
an
d
“
kira
”
s
h
o
ws
cu
s
to
m
er
d
is
ap
p
o
in
tm
en
t
at
th
e
s
er
v
ice
r
ec
eiv
ed
.
T
h
is
r
esear
ch
is
a
s
er
ies
o
f
s
tu
d
ies
o
n
th
e
p
r
o
d
u
ct
"
K
a
ta
K
ita
"
to
b
e
ap
p
lied
in
p
u
b
lic
s
er
v
ices
th
a
t
f
u
n
ctio
n
t
o
r
ec
o
r
d
co
n
v
er
s
atio
n
s
in
th
e
f
o
r
m
o
f
tex
t,
lab
elin
g
a
u
to
lear
n
i
n
g
b
as
ed
o
n
m
ac
h
in
e
lear
n
in
g
a
n
d
d
ig
g
in
g
i
n
f
o
r
m
atio
n
.
I
n
f
o
r
m
atio
n
ex
tr
ac
ted
f
r
o
m
s
er
v
ice
co
n
v
er
s
atio
n
s
ca
n
b
e
u
s
ed
as
m
ater
ial
f
o
r
ev
alu
atin
g
an
d
d
ev
elo
p
in
g
s
er
v
ice
q
u
ality
.
Fu
r
t
h
er
r
esear
c
h
ca
n
b
e
ca
r
r
ied
o
u
t
o
n
o
th
e
r
f
ea
tu
r
es
o
f
"
K
a
ta
K
ita
"
p
r
o
d
u
cts
o
r
with
a
v
a
r
iety
o
f
d
atasets
f
r
o
m
d
if
f
er
e
n
t
p
u
b
lic
s
er
v
ice
s
o
u
r
ce
s
.
ACK
NO
WL
E
DG
E
M
E
NT
S
R
esear
ch
er
s
wo
u
ld
lik
e
to
t
h
a
n
k
th
e
Ma
g
is
ter
o
f
I
n
f
o
r
m
atic
s
,
Un
iv
er
s
itas
AM
I
KOM
Yo
g
y
ak
ar
ta
f
o
r
p
r
o
v
id
i
n
g
s
u
p
p
o
r
t a
n
d
g
u
id
an
c
e
to
th
is
r
esear
ch
.
RE
F
E
R
E
NC
E
S
[1
]
Ba
ra
ta,
Ate
p
,
a
n
d
A
d
y
a
,
"
Da
sa
r
-
d
a
sa
r
p
e
lay
a
n
a
n
p
rima
,"
Ja
k
a
rta:
P
T
.
El
e
x
me
d
ia
C
o
mp
u
ti
n
d
o
,
2
0
0
3
.
[2
]
I.
Og
ra
jen
še
k
a
n
d
I.
G
a
l,
“
Th
e
c
o
n
c
e
p
t
a
n
d
a
ss
e
ss
m
e
n
t
o
f
c
u
sto
m
e
r
sa
ti
sfa
c
ti
o
n
,
”
M
o
d
e
rn
An
a
ly
sis
o
f
Cu
sto
me
r
S
u
rv
e
y
s: W
it
h
Ap
p
li
c
a
t
io
n
s Us
in
g
R
,
p
p
.
1
0
7
-
1
2
7
,
2
0
1
1
.
[3
]
D.
Alb
u
r
y
,
“
F
o
ste
rin
g
In
n
o
v
a
ti
o
n
in
P
u
b
l
ic S
e
rv
ice
s,”
Pu
b
li
c
M
o
n
e
y
M
a
n
a
g
.
,
v
o
l
.
2
5
,
p
p
.
5
1
-
5
6
,
F
e
b
.
2
0
0
5
.
[4
]
T.
M
.
Hid
a
y
a
t,
“
In
tere
st
o
f
P
u
b
li
c
Ad
m
in
istrati
o
n
,
Ac
c
e
lera
ted
Im
p
ro
v
e
m
e
n
t
o
f
P
u
b
li
c
S
e
r
v
ice
Qu
a
li
ty
,
(
in
Ba
h
a
sa
:
Bu
n
g
a
Ra
m
p
a
i
Ad
m
in
istras
i
P
u
b
li
k
,
Ak
se
lera
si
P
e
n
i
n
g
k
a
tan
Ku
a
li
tas
P
e
lay
a
n
a
n
P
u
b
li
k
)
,
”
Bu
n
g
a
Ra
m
p
a
i
Ad
min
istr
a
si P
u
b
li
k
.
p
p
.
5
1
–
6
3
,
2
0
1
6
.
[5
]
M
.
Cu
i
,
T.
Wan
g
,
Z.
P
a
n
,
a
n
d
L.
Ni,
“
De
fin
it
io
n
o
f
P
e
o
p
le
wit
h
Im
p
e
d
ime
n
ts
a
n
d
Un
i
v
e
rsa
li
ty
Ev
a
lu
a
ti
o
n
o
f
P
u
b
l
ic
S
e
rv
ice
in
Air
p
o
rt
Tra
v
e
l
S
c
e
n
a
ri
o
s,”
I
n
ter
n
a
t
io
n
a
l
Co
n
fer
e
n
c
e
o
n
Hu
ma
n
-
C
o
mp
u
ter
In
ter
a
c
ti
o
n
,
p
p
.
6
2
3
-
6
3
9
,
2
0
2
0
.
[6
]
Y.
Wan
g
,
K.
Kim
,
B.
Lee
,
a
n
d
H
.
Y.
Yo
u
n
,
“
Wo
rd
c
lu
ste
rin
g
b
a
se
d
o
n
P
OS
fe
a
tu
re
f
o
r
e
fficie
n
t
t
witt
e
r
se
n
ti
m
e
n
t
a
n
a
ly
sis,”
Hu
ma
n
-
c
e
n
tric Co
m
p
u
t
.
In
f.
S
c
i.
,
v
o
l.
8
,
n
o
.
1
,
p
.
1
-
1
7
,
2
0
1
8
.
[7
]
N.
S
h
a
i
k
h
a
n
d
R.
De
sh
m
u
k
h
,
“
S
p
e
e
c
h
Re
c
o
g
n
it
i
o
n
S
y
ste
m
-
A
Re
v
i
e
w,”
IOS
R
J
.
C
o
mp
u
t.
E
n
g
.
,
v
o
l
.
1
8
,
p
p
.
1
-
9
,
2
0
1
6
.
[8
]
A.
A.
Lu
tfi
,
A.
E.
P
e
rm
a
n
a
sa
ri,
a
n
d
S
.
F
a
u
z
iati,
“
S
e
n
ti
m
e
n
t
An
a
ly
si
s
in
th
e
S
a
les
Re
v
iew
o
f
In
d
o
n
e
sia
n
M
a
rk
e
tp
lac
e
b
y
Uti
li
z
in
g
S
u
p
p
o
rt
Ve
c
to
r
M
a
c
h
in
e
,
”
J
.
I
n
f.
S
y
st.
E
n
g
.
Bu
s
.
In
tel
l
.
,
v
o
l.
4
,
n
o
.
1
,
p
p
.
5
7
-
64
,
2
0
1
8
.
[9
]
A.
De
v
iy
a
n
to
a
n
d
M
.
D.
R
.
Wah
y
u
d
i,
“
Ap
p
li
c
a
ti
o
n
o
f
S
e
n
ti
m
e
n
t
An
a
ly
sis
t
o
Twit
ter
Us
e
rs
Us
in
g
th
e
K
-
Ne
a
re
st
Ne
ig
h
b
o
r
M
e
t
h
o
d
(i
n
Ba
h
a
sa
:
P
e
n
e
ra
p
a
n
An
a
li
sis
S
e
n
t
ime
n
P
a
d
a
P
e
n
g
g
u
n
a
Twit
ter
M
e
n
g
g
u
n
a
k
a
n
M
e
to
d
e
K
-
Ne
a
re
st
Ne
ig
h
b
o
r
)
,
”
J
IS
KA
(J
u
rn
a
l
I
n
fo
r
m.
S
u
n
a
n
K
a
li
j
a
g
a
)
,
v
o
l.
3
,
n
o
.
1
,
p
p
.
1
-
1
3
,
2
0
1
8
.
[1
0
]
K.
No
rm
a
n
,
Z.
Li
,
Y
.
T.
Oh
,
G
.
Go
lwa
la,
S
.
S
u
n
d
a
ra
m
,
a
n
d
J.
Alleb
a
c
h
,
“
Ap
p
li
c
a
ti
o
n
o
f
n
a
tu
ra
l
lan
g
u
a
g
e
p
ro
c
e
ss
in
g
to
a
n
o
n
li
n
e
fa
sh
i
o
n
m
a
rk
e
tp
lac
e
,
”
IS
T
I
n
t.
S
y
mp
.
El
e
c
tro
n
.
Ima
g
in
g
S
c
i.
T
e
c
h
n
o
l.
,
p
p
.
1
-
5
,
2
0
1
8
.
[1
1
]
N.
Kim
a
n
d
S
.
H
o
n
g
,
“
Tex
t
m
i
n
i
n
g
f
o
r
t
h
e
e
v
a
l
u
a
ti
o
n
o
f
p
u
b
li
c
se
rv
ice
s:
th
e
c
a
se
o
f
a
p
u
b
li
c
b
i
k
e
-
sh
a
rin
g
s
y
ste
m
,
”
S
e
rv
.
Bu
s.
,
v
o
l.
1
4
,
p
p
.
3
1
5
-
3
3
1
,
2
0
2
0
.
[1
2
]
B.
Ch
e
e
k
a
ti
a
n
d
S
.
P
a
d
a
la,
“
Ev
a
lu
a
ti
n
g
t
h
e
P
ro
g
re
ss
iv
e
P
e
rfo
r
m
a
n
c
e
o
f
M
a
c
h
in
e
Lea
rn
in
g
Te
c
h
n
iq
u
e
s
o
n
E
-
c
o
m
m
e
rc
e
D
a
ta,”
S
ma
rt Co
mp
u
ti
n
g
a
n
d
I
n
fo
rm
a
t
ics
,
p
p
.
9
5
-
1
0
7
,
2
0
1
8
.
[1
3
]
M
.
T.
Kh
a
n
,
M
.
Du
rra
n
i,
A.
Ali
,
I.
In
a
y
a
t,
S
.
Kh
a
li
d
,
a
n
d
K.
H.
K
h
a
n
,
“
S
e
n
ti
m
e
n
t
a
n
a
l
y
sis
a
n
d
th
e
c
o
m
p
lex
n
a
tu
ra
l
lan
g
u
a
g
e
,
”
Co
mp
lex
A
d
a
p
t.
S
y
st.
M
o
d
e
l.
,
v
o
l
.
4
,
n
o
.
1
,
2
0
1
6
.
[1
4
]
B.
Ia
n
c
u
,
“
E
v
a
lu
a
ti
n
g
G
o
o
g
le
S
p
e
e
c
h
-
to
-
Tex
t
API’s
P
e
rfo
rm
a
n
c
e
fo
r
Ro
m
a
n
ian
e
-
Lea
rn
in
g
Re
so
u
rc
e
s,”
In
fo
rm
.
Eco
n
.
,
v
o
l.
2
3
,
n
o
.
1
/2
0
1
9
,
p
p
.
1
7
-
2
5
,
2
0
1
9
.
[1
5
]
A.
S
o
l
ich
in
,
“
M
y
S
q
l
5
:
F
r
o
m
Be
g
in
n
e
r
to
Ad
v
a
n
c
e
d
(i
n
Ba
h
a
sa
:
Da
ri
P
e
m
u
la
Hin
g
g
a
M
a
h
ir
)
,
”
U
n
iv.
B
u
d
i
L
u
h
u
r,
J
a
k
a
rta
,
Ja
n
.
2
0
1
0
.
[1
6
]
M
.
M
h
a
tre,
D.
P
h
o
n
d
e
k
a
r,
P
.
Ka
d
a
m
,
A.
Ch
a
wa
th
e
,
a
n
d
K.
G
h
a
g
,
“
Dim
e
n
sio
n
a
li
t
y
re
d
u
c
ti
o
n
f
o
r
se
n
t
ime
n
t
a
n
a
ly
sis
u
sin
g
p
re
-
p
r
o
c
e
ss
in
g
tec
h
n
iq
u
e
s,”
2
0
1
7
In
ter
n
a
ti
o
n
a
l
C
o
n
fer
e
n
c
e
o
n
Co
mp
u
ti
n
g
M
e
th
o
d
o
lo
g
ies
a
n
d
Co
mm
u
n
ica
ti
o
n
(ICCM
C)
,
p
p
.
1
6
-
2
1
,
2
0
1
7
.
[1
7
]
Z.
Jia
n
q
ian
g
a
n
d
G
.
Xia
o
li
n
,
“
Co
m
p
a
riso
n
Re
se
a
rc
h
o
n
Tex
t
P
re
-
p
r
o
c
e
ss
in
g
M
e
th
o
d
s
o
n
Twit
ter
S
e
n
t
ime
n
t
An
a
ly
sis,”
IEE
E
Acc
e
ss
,
v
o
l
.
5
,
p
p
.
2
8
7
0
-
2
8
7
9
,
2
0
1
7
.
[1
8
]
G
.
Ch
o
wd
h
a
r
y
,
“
Na
tu
ra
l
Lan
g
u
a
g
e
P
r
o
c
e
ss
in
g
,
”
An
n
u
a
l
Rev
iew
o
f
I
n
fo
rm
a
ti
o
n
S
c
ien
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l.
37
,
n
o
.
1
,
p
p
.
51
-
8
9
2
0
2
0
.
[1
9
]
L.
A.
M
u
l
len
,
K.
Be
n
o
it
,
O
.
Ke
y
e
s,
D.
S
e
li
v
a
n
o
v
,
a
n
d
J.
Ar
n
o
l
d
,
“
F
a
st,
Co
n
siste
n
t
T
o
k
e
n
iza
ti
o
n
o
f
N
a
tu
ra
l
Lan
g
u
a
g
e
Tex
t,
”
J
.
Op
e
n
S
o
u
rc
e
S
o
ft
w.
,
v
o
l
.
3
,
n
o
.
2
3
,
p
.
6
5
5
,
2
0
1
8
,
d
o
i:
1
0
.
2
1
1
0
5
/j
o
ss
.
0
0
6
5
5
.
[2
0
]
A.
N.
Ro
h
m
a
n
,
E.
Uta
m
i,
a
n
d
S
.
Ra
h
a
rjo
,
“
De
tek
si
Ko
n
d
isi
Emo
si
p
a
d
a
M
e
d
ia
S
o
sia
l
M
e
n
g
g
u
n
a
k
a
n
P
e
n
d
e
k
a
ta
n
Lek
sik
o
n
d
a
n
Na
tu
ra
l
Lan
g
u
a
g
e
P
ro
c
e
ss
in
g
,
”
Eks
p
lo
ra
In
f
o
rm
.
,
v
o
l.
9
,
n
o
.
1
,
p
p
.
7
0
-
7
6
,
2
0
1
9
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
19
,
No
.
1
,
Feb
r
u
ar
y
2
0
2
1
:
1
4
6
-
1
5
4
154
[2
1
]
B.
P
.
P
a
n
d
e
a
n
d
H.
S
.
Dh
a
m
i,
“
Ap
p
li
c
a
ti
o
n
o
f
Na
tu
ra
l
Lan
g
u
a
g
e
P
ro
c
e
ss
in
g
To
o
ls
in
S
tem
m
in
g
,
”
In
t.
J
.
Co
mp
u
t
.
Ap
p
l
.
,
v
o
l.
2
7
,
n
o
.
6
,
p
p
.
1
4
-
1
9
,
2
0
1
1
.
[2
2
]
A.
P
ra
k
a
sh
a
n
d
U.
Ku
m
a
r,
“
Au
t
h
e
n
ti
c
a
ti
o
n
P
ro
t
o
c
o
ls
a
n
d
Tec
h
n
i
q
u
e
s:
A
S
u
rv
e
y
,
"
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
C
o
mp
u
ter
S
c
ien
c
e
s a
n
d
En
g
i
n
e
e
rin
g
,
v
o
l.
6
,
n
o
.
6
,
p
p
.
1
0
1
4
-
1
0
2
0
,
2
0
1
9
.
[2
3
]
I.
Riad
i
,
S
.
S
u
n
a
rd
i,
a
n
d
P
.
Wi
d
i
a
n
d
a
n
a
,
“
M
o
b
il
e
F
o
re
n
sic
s
f
o
r
C
y
b
e
rb
u
ll
y
in
g
De
tec
ti
o
n
u
sin
g
Term
F
re
q
u
e
n
c
y
-
In
v
e
rse
Do
c
u
m
e
n
t
F
re
q
u
e
n
c
y
(TF
-
IDF),
”
J
.
Ilm.
T
e
k
.
E
lek
tro
Ko
m
p
u
t.
d
a
n
In
fo
rm
.
,
v
o
l.
5
,
p
p
.
6
8
-
7
6
,
2
0
1
9
.
[2
4
]
R.
M
e
li
ta,
V.
Am
riza
l,
H.
S
u
se
n
o
,
a
n
d
T
.
Dirjam
,
“
Ap
p
li
c
a
ti
o
n
o
f
T
e
rm
F
re
q
u
e
n
c
y
I
n
v
e
rse
Do
c
u
m
e
n
t
F
re
q
u
e
n
c
y
(Tf
-
Id
f)
a
n
d
C
o
sin
e
S
imilari
ty
M
e
t
h
o
d
s
i
n
In
f
o
rm
a
ti
o
n
Re
tri
e
v
a
l
S
y
s
tem
s
to
F
in
d
Web
-
Ba
se
d
Ha
d
it
h
s
(Ca
se
S
tu
d
y
:
Ha
d
it
s
S
a
h
ih
Bu
k
h
a
ri
-
M
u
slim
)
,
(i
n
Ba
h
a
sa
:
P
e
n
e
ra
p
a
n
M
e
t
o
d
e
Ter
m
F
re
q
u
e
n
c
y
I
n
v
e
rse
Do
c
u
m
e
n
t
F
re
q
u
e
n
c
y
(Tf
-
Id
f)
d
a
n
C
o
sin
e
S
imilarit
y
p
a
d
a
S
istem
Tem
u
Ke
m
b
a
li
In
fo
rm
a
si
u
n
t
u
k
M
e
n
g
e
tah
u
i
S
y
a
ra
h
Ha
d
i
t
s
Be
rb
a
sis
Web
(S
tu
d
i
Ka
su
s: Ha
d
it
s
S
h
a
h
ih
B
u
k
h
a
ri
-
M
u
slim
)
)
,
”
J
.
T
e
k
.
In
f
o
rm
.
,
v
o
l.
1
1
,
n
o
.
2
,
p
p
.
1
4
9
-
1
6
4
,
2
0
1
8
.
[2
5
]
V.
Ja
h
jah
,
R.
Kh
o
u
r
y
,
a
n
d
L.
Lam
o
n
tag
n
e
,
"
W
o
rd
N
o
rm
a
li
z
a
ti
o
n
Us
in
g
P
h
o
n
e
ti
c
S
ig
n
a
tu
re
s
,
"
C
o
n
fer
e
n
c
e
:
Ca
n
a
d
ia
n
Co
n
fer
e
n
c
e
o
n
Arti
fi
c
i
a
l
I
n
telli
g
e
n
c
e
,
2
0
1
6
.
[2
6
]
J.
As
ian
,
H.
E.
Wi
ll
iam
s,
a
n
d
S
.
M
.
M
.
Tah
a
g
h
o
g
h
i
,
“
S
tem
m
in
g
I
n
d
o
n
e
sia
n
,
”
Co
n
f.
Res
.
Pra
c
t.
In
f
.
T
e
c
h
n
o
l
.
S
e
r.
,
v
o
l.
3
8
,
n
o
.
Ja
n
u
a
ry
,
p
p
.
3
0
7
-
3
1
4
,
2
0
0
5
.
[2
7
]
M
.
Kh
a
d
e
r,
e
t
a
l.
,
“
Th
e
Eff
e
c
ts
o
f
Na
tu
ra
l
Lan
g
u
a
g
e
P
ro
c
e
ss
in
g
o
n
Big
Da
ta
An
a
ly
sis:
S
e
n
ti
m
e
n
t
An
a
ly
sis
Ca
se
S
tu
d
y
,
”
2
0
1
8
I
n
ter
n
a
ti
o
n
a
l
Ara
b
Co
n
fer
e
n
c
e
o
n
In
f
o
rm
a
ti
o
n
T
e
c
h
n
o
lo
g
y
(ACIT
)
,
p
p
.
1
-
7
,
2
0
1
8
.
[2
8
]
E.
M
a
rth
a
n
ty
a
n
d
I.
Les
tari,
“
T
h
e
Cu
lt
u
re
o
f
Ap
o
lo
g
y
a
t
W
o
rk
in
Ja
p
a
n
e
se
Dra
m
a
s:
A
S
o
c
io
li
n
g
u
isti
c
Ov
e
rv
iew
,
(in
Ba
h
a
sa
:
Bu
d
a
y
a
P
e
rm
in
taa
n
M
a
a
f
Di
Tem
p
a
t
Ke
rja
Da
lam
Dra
m
a
Je
p
a
n
g
:
Ti
n
jau
a
n
S
o
sio
li
n
g
u
isti
k
),
”
J
u
rn
a
l
Ba
h
a
sa
,
S
a
str
a
d
a
n
Bu
d
a
y
a
J
e
p
a
n
g
,
v
o
l.
8
,
n
o
.
2
,
p
p
.
8
2
-
1
0
0
2
0
1
9
.
[2
9
]
A.
Tri
p
a
th
y
,
A.
Ag
ra
wa
l,
a
n
d
S
.
K.
Ra
th
,
“
Clas
sifica
ti
o
n
o
f
se
n
ti
m
e
n
t
re
v
iew
s
u
sin
g
n
-
g
ra
m
m
a
c
h
in
e
lea
rn
i
n
g
a
p
p
ro
a
c
h
,
”
Exp
e
rt S
y
st.
Ap
p
l.
,
v
o
l
.
5
7
,
p
p
.
1
1
7
-
1
2
6
,
2
0
1
6
.
[3
0
]
A.
K.
S
a
n
tra
a
n
d
C.
J.
C
h
risty
,
“
G
e
n
e
ti
c
Alg
o
r
it
h
m
a
n
d
Co
n
fu
sio
n
M
a
tri
x
f
o
r
D
o
c
u
m
e
n
t
Cl
u
ste
rin
g
1
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Co
m
p
u
ter
S
c
ien
c
e
,
v
o
l
.
9
,
n
o
.
1
,
p
p
.
3
2
2
-
3
2
8
,
2
0
1
2
.
[3
1
]
A.
Th
a
rwa
t,
“
Clas
sifica
ti
o
n
As
se
ss
m
e
n
t
M
e
th
o
d
s,”
Ap
p
l.
Co
mp
u
t.
I
n
fo
rm
a
t
ics
,
n
o
.
Ja
n
u
a
ry
,
p
p
.
1
-
1
3
,
2
0
1
8
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Eli
k
H
a
r
i
Mu
k
t
a
fin
,
b
o
rn
i
n
S
ra
g
e
n
,
Ja
wa
Ten
g
a
h
,
I
n
d
o
n
e
sia
,
2
S
e
p
tem
b
e
r
1
9
9
2
.
He
h
o
ld
s
a
Ba
c
h
e
lo
r
o
f
I
n
fo
rm
a
ti
c
s
(
S
.
Ko
m
)
fro
m
S
TM
IK
AMIKOM
Yo
g
y
a
k
a
rta
in
2
0
1
5
.
C
u
rre
n
tl
y
th
e
a
u
th
o
r
is
c
a
rry
in
g
o
u
t
stu
d
ies
in
th
e
field
o
f
M
a
g
ister
o
f
In
f
o
rm
a
ti
c
s,
Un
iv
e
rsitas
AMIKOM
Yo
g
y
a
k
a
rta
.
E
m
a
il
a
d
d
re
ss
:
e
li
k
.
m
u
k
tafi
n
@s
tu
d
e
n
ts.am
ik
o
m
.
a
c
.
i
d
/
e
li
x
ro
p
a
s@
g
m
a
il
.
c
o
m
Pra
m
o
n
o
,
b
o
r
n
in
Bo
y
o
lali
,
Ja
w
a
Ten
g
a
h
,
I
n
d
o
n
e
sia
2
3
J
u
li
1
9
8
6
,
h
e
H
o
ld
s
a
Ba
c
h
e
l
o
r
o
f
In
fo
rm
a
ti
c
s
(S
.
K
o
m
)
F
r
o
m
S
T
M
I
K
DU
TA
BAN
G
S
A
S
u
ra
k
a
rta
in
2
0
1
2
.
C
u
rre
n
tl
y
t
h
e
a
u
th
o
r
is
c
a
rry
in
g
o
u
t
stu
d
ies
in
t
h
e
field
o
f
M
a
g
ister
o
f
In
fo
rm
a
t
ics
,
Un
iv
e
rsitas
AMIKOM
Yo
g
y
a
k
a
rta
.
E
m
a
il
a
d
d
re
ss
:
p
ra
m
o
n
o
.
2
3
@s
t
u
d
e
n
ts.am
ik
o
m
.
a
c
.
id
o
r
p
ra
m
o
n
o
2
3
0
7
@g
m
a
il
.
c
o
m
K
u
sr
in
i
is
a
n
a
ss
o
c
iate
p
r
o
fe
ss
o
r
fr
o
m
Un
i
v
e
rsitas
AMIKOM
Yo
g
y
a
k
a
rta
In
d
o
n
e
sia
.
S
h
e
fin
ish
e
d
h
e
r
d
o
c
t
o
ra
l
p
r
o
g
ra
m
fro
m
Un
iv
e
rsitas
G
a
d
jah
M
a
d
a
Y
o
g
y
a
k
a
rta
In
d
o
n
e
sia
in
2
0
1
0
.
S
h
e
is
in
tere
ste
d
in
e
x
p
lo
ri
n
g
m
a
n
y
th
in
g
s
a
b
o
u
t
m
a
c
h
in
e
lea
rn
in
g
a
n
d
o
t
h
e
r
a
rti
ficia
l
in
t
e
ll
i
g
e
n
c
e
field
.
S
h
e
a
lso
l
o
v
e
s i
n
d
o
i
n
g
re
se
a
rc
h
a
b
o
u
t
d
e
c
isio
n
s
u
p
p
o
rt
sy
ste
m
a
n
d
d
a
tab
a
se
.
S
h
e
is me
m
b
e
r
o
f
t
h
e
IEE
E
a
n
d
I
EE
E
S
y
ste
m
s,
M
a
n
,
a
n
d
C
y
b
e
rn
e
t
ics
S
o
c
iety
Evaluation Warning : The document was created with Spire.PDF for Python.