I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
pu
t
er
E
ng
ineering
(
I
J
E
CE
)
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
25
,
p
p
.
2977
~
2
9
8
5
I
SS
N:
2088
-
8
7
0
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijece.
v
15
i
3
.
pp
2
9
7
7
-
2
9
8
5
2977
J
o
ur
na
l
ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m
Cla
ss
ificatio
n of
c
usto
mer com
pla
i
nts on so
cia
l medi
a
f
o
r
e
-
co
mm
erce
in
In
do
nesia
Achm
a
d Riz
k
i A
dita
m
a
,
Alf
a
n F
a
rizk
i Wic
a
k
s
o
no
F
a
c
u
l
t
y
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
U
n
i
v
e
r
si
t
a
s
I
n
d
o
n
e
si
a
,
D
e
p
o
k
,
I
n
d
o
n
e
s
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
u
n
1
3
,
2
0
2
4
R
ev
is
ed
Dec
2
8
,
2
0
2
4
Acc
ep
ted
J
an
1
6
,
2
0
2
5
Th
e
e
-
c
o
m
m
e
rc
e
in
d
u
str
y
in
I
n
d
o
n
e
sia
h
a
s
e
x
p
e
rien
c
e
d
ra
p
i
d
g
ro
wt
h
,
e
sp
e
c
ially
d
u
r
in
g
th
e
COV
ID
-
1
9
p
a
n
d
e
m
ic,
wh
ich
a
c
c
e
lera
ted
th
e
sh
ift
to
o
n
li
n
e
p
latfo
rm
s.
T
h
e
m
a
rk
e
t
is
e
x
p
e
c
ted
to
g
r
o
w
b
y
1
0
5
.
5
%
fr
o
m
2
0
2
5
t
o
20
30
d
u
e
to
i
n
c
re
a
se
d
in
tern
e
t
a
n
d
sm
a
rtp
h
o
n
e
u
se
.
As
e
-
c
o
m
m
e
rc
e
e
x
p
a
n
d
s,
c
o
m
p
a
n
ies
m
u
st i
m
p
ro
v
e
h
o
w t
h
e
y
h
a
n
d
le cu
sto
m
e
r
c
o
m
p
lain
ts t
o
b
u
il
d
tru
st
a
n
d
l
o
y
a
lt
y
.
S
o
c
ial
m
e
d
ia
is
a
c
ru
c
ial
c
h
a
n
n
e
l
f
o
r
c
u
st
o
m
e
r
in
tera
c
ti
o
n
s,
b
u
t
it
a
lso
i
n
c
lu
d
e
s
n
o
n
-
c
o
m
p
lain
t
m
e
ss
a
g
e
s
li
k
e
p
o
siti
v
e
c
o
m
m
e
n
t
s,
g
e
n
e
ra
l
que
stio
n
s,
a
n
d
sp
a
m
s
t
h
a
t
n
e
e
d
t
o
b
e
fi
lt
e
re
d
o
u
t.
Th
is
re
se
a
rc
h
p
ro
p
o
se
s
a
m
a
c
h
in
e
lea
rn
in
g
m
o
d
e
l
to
a
u
t
o
m
a
ti
c
a
ll
y
c
las
sify
so
c
ial
m
e
d
ia
i
n
tera
c
ti
o
n
s
in
to
c
o
m
p
lain
ts
a
n
d
n
o
n
-
c
o
m
p
lain
ts,
fo
c
u
si
n
g
o
n
I
n
d
o
n
e
sia
n
-
lan
g
u
a
g
e
c
o
n
ten
t.
Th
e
m
o
d
e
li
n
g
p
ro
c
e
ss
u
ti
li
z
e
d
1
0
,
6
0
0
d
a
ta
p
o
i
n
ts
c
o
l
lec
ted
fro
m
so
c
ial
m
e
d
ia
X.
Th
e
b
e
st
m
o
d
e
l,
a
b
id
irec
ti
o
n
a
l
e
n
c
o
d
e
r
re
p
re
se
n
t
a
ti
o
n
fro
m
tran
sfo
rm
e
rs
(BERT
)
b
a
se
d
c
las
s
ifi
e
r,
a
c
h
iev
e
d
a
n
F
1
-
sc
o
re
o
f
9
8
.
3
%
.
Th
e
M
c
Ne
m
a
r
tes
t
re
v
e
a
led
sig
n
ifi
c
a
n
t
p
e
rfo
rm
a
n
c
e
d
iffere
n
c
e
s
b
e
twe
e
n
se
v
e
ra
l
m
o
d
e
ls,
with
t
h
e
BERT
-
b
a
s
e
d
m
o
d
e
l
o
u
tp
e
rfo
rm
i
n
g
o
th
e
rs.
Th
is
d
e
m
o
n
stra
tes
th
a
t
it
is
h
ig
h
l
y
e
ffe
c
ti
v
e
in
d
isti
n
g
u
ish
i
n
g
b
e
twe
e
n
c
o
m
p
lain
ts
a
n
d
n
o
n
-
c
o
m
p
lai
n
ts,
m
a
k
in
g
i
t
a
v
a
lu
a
b
le
to
o
l
fo
r
e
n
h
a
n
c
in
g
c
u
sto
m
e
r
se
rv
ice
in
In
d
o
n
e
sia
'
s e
-
c
o
m
m
e
r
c
e
se
c
to
r.
K
ey
w
o
r
d
s
:
B
id
ir
ec
tio
n
al
en
co
d
er
r
ep
r
esen
tatio
n
s
f
r
o
m
tr
an
s
f
o
r
m
er
s
E
-
co
m
m
er
ce
Ma
ch
in
e
lear
n
in
g
So
cial
m
ed
ia
T
ex
t c
lass
if
icatio
n
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
:
Ach
m
ad
R
izk
i A
d
itam
a
Ma
s
ter
o
f
I
n
f
o
r
m
atio
n
T
ec
h
n
o
lo
g
y
,
Un
iv
e
r
s
itas
I
n
d
o
n
esia
Facu
lty
o
f
C
o
m
p
u
ter
Scien
ce
B
u
ild
in
g
,
UI
C
am
p
u
s
Dep
o
k
1
6
4
2
4
,
I
n
d
o
n
esia
E
m
ail: a
ch
m
ad
.
r
izk
i
0
2
@
u
i.a
c.
id
1.
I
NT
RO
D
UCT
I
O
N
T
h
e
e
-
c
o
m
m
er
ce
in
d
u
s
tr
y
i
n
I
n
d
o
n
esia
h
as
ex
p
er
ien
ce
d
s
u
b
s
tan
tial
g
r
o
wth
in
r
ec
en
t
y
ea
r
s
,
p
ar
ticu
lar
ly
d
u
e
to
th
e
C
OVI
D
-
1
9
p
an
d
e
m
ic,
wh
ic
h
f
o
r
ce
d
m
an
y
b
u
s
in
ess
es
to
m
o
v
e
o
n
lin
e
[
1
]
.
T
h
e
r
ap
id
g
r
o
wth
o
f
e
-
c
o
m
m
er
ce
h
as
b
ee
n
a
k
ey
f
ac
to
r
in
d
r
iv
in
g
th
e
d
ig
ital
tr
an
s
f
o
r
m
atio
n
o
f
th
e
n
atio
n
al
ec
o
n
o
m
y
[
2
]
.
T
h
e
m
ar
k
et
is
an
ticip
ated
to
g
r
o
w
b
y
ab
o
u
t
1
0
5
.
5
%
f
r
o
m
2
0
2
5
to
2
0
3
0
,
d
r
iv
e
n
p
r
i
m
ar
ily
b
y
th
e
r
is
in
g
n
u
m
b
er
o
f
i
n
ter
n
et
an
d
s
m
ar
tp
h
o
n
e
u
s
er
s
,
f
ac
ilit
atin
g
ea
s
ier
o
n
lin
e
s
h
o
p
p
in
g
[
3
]
.
As
th
e
e
-
co
m
m
er
ce
s
ec
to
r
ex
p
an
d
s
,
co
m
p
a
n
ies
m
u
s
t
im
p
r
o
v
e
th
eir
cu
s
to
m
er
h
an
d
lin
g
p
r
o
ce
s
s
es.
Qu
ick
an
d
ef
f
ec
tiv
e
r
esp
o
n
s
es
to
cu
s
to
m
er
co
m
p
lain
ts
ar
e
c
r
u
cial
f
o
r
en
h
an
cin
g
s
atis
f
ied
an
d
lo
y
alty
[
4
]
.
A
n
in
teg
r
ated
cu
s
to
m
er
r
elatio
n
s
h
ip
m
an
ag
e
m
en
t
s
y
s
tem
ca
n
s
ig
n
if
ican
tly
b
o
o
s
t
cu
s
to
m
e
r
s
atis
f
ac
tio
n
[
5
]
.
On
e
ef
f
ec
tiv
e
s
tr
ateg
y
is
u
tili
zin
g
m
u
ltip
le
ch
a
n
n
els,
in
clu
d
in
g
s
o
cial
m
ed
ia,
t
o
e
n
g
ag
e
with
cu
s
to
m
er
s
.
So
cial
m
ed
ia
en
ab
les
cu
s
to
m
er
s
to
q
u
ick
ly
f
ile
co
m
p
lain
ts
an
d
r
ec
eiv
e
r
esp
o
n
s
es,
h
elp
in
g
co
m
p
an
ies
m
an
ag
e
th
ese
in
ter
ac
tio
n
s
ef
f
icien
tly
.
Mo
r
e
o
v
er
,
s
o
cial
m
ed
ia
ac
tiv
ities
ca
n
s
ig
n
if
ican
tly
im
p
ac
t
cu
s
to
m
er
s
atis
f
ac
tio
n
an
d
en
g
ag
em
e
n
t,
th
er
eb
y
en
h
an
ci
n
g
b
r
an
d
lo
y
alty
an
d
p
u
b
lic
p
er
ce
p
tio
n
[
6
]
.
Ho
wev
er
,
s
o
cia
l
m
ed
ia
in
ter
ac
tio
n
s
ar
e
n
o
t
lim
ited
to
co
m
p
lain
ts
;
th
ey
also
in
clu
d
e
p
o
s
itiv
e
r
em
ar
k
s
,
g
en
er
al
q
u
er
ies,
an
d
s
p
am
,
wh
ich
m
u
s
t
b
e
f
ilter
ed
o
u
t
[
7
]
.
C
o
m
p
an
ies
n
ee
d
to
d
if
f
er
en
tiate
ac
tu
al
co
m
p
lain
ts
f
r
o
m
o
th
er
in
ter
ac
tio
n
s
to
en
s
u
r
e
tim
ely
cu
s
to
m
er
f
ee
d
b
ac
k
m
a
n
ag
em
e
n
t,
th
er
eb
y
im
p
r
o
v
in
g
cu
s
to
m
e
r
s
atis
f
ac
tio
n
an
d
o
p
er
atio
n
al
e
f
f
icien
cy
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
25
:
2
9
7
7
-
2
9
8
5
2978
I
n
ad
d
r
ess
in
g
th
ese
ch
allen
g
es,
m
ac
h
in
e
lear
n
in
g
ca
n
h
elp
b
y
au
to
m
atica
lly
s
o
r
tin
g
in
ter
a
ctio
n
s
in
to
co
m
p
lain
ts
an
d
n
o
n
-
c
o
m
p
lain
ts
.
T
h
is
m
ak
es
th
e
co
m
p
an
y
'
s
o
p
er
atio
n
s
m
o
r
e
ef
f
icien
t,
as
r
eso
u
r
ce
s
a
r
e
o
n
ly
u
s
ed
f
o
r
is
s
u
es
n
ee
d
in
g
s
p
ec
ial
atten
tio
n
.
W
e
ar
g
u
e
th
at
th
is
im
p
r
o
v
es
cu
s
to
m
e
r
in
te
r
ac
tio
n
m
an
a
g
em
en
t
an
d
s
atis
f
ac
tio
n
,
s
u
p
p
o
r
tin
g
th
e
g
r
o
wth
o
f
e
-
co
m
m
er
ce
in
I
n
d
o
n
esia.
Desp
ite
th
is
p
o
ten
tial,
th
er
e
ar
e
s
o
m
e
ch
allen
g
es,
in
clu
d
in
g
tech
n
ical,
o
p
er
atio
n
al,
an
d
h
u
m
an
r
eso
u
r
ce
is
s
u
es.
T
h
er
ef
o
r
e,
it
is
im
p
o
r
tan
t
to
u
n
d
er
s
tan
d
h
o
w
th
e
y
cu
r
r
en
tly
m
an
ag
e
s
o
cial
m
ed
ia
in
ter
ac
tio
n
s
an
d
wh
ic
h
o
n
es n
ee
d
f
u
r
th
er
h
an
d
lin
g
.
R
ec
en
t
s
tu
d
ies
s
h
o
w
th
at
m
ac
h
in
e
lear
n
in
g
is
ef
f
ec
tiv
e
f
o
r
class
if
y
in
g
cu
s
to
m
er
co
m
p
l
ain
ts
.
Key
tech
n
iq
u
es
in
clu
d
e
o
b
tain
in
g
n
u
m
e
r
ical
r
ep
r
esen
tatio
n
s
o
f
tex
t
d
ata
u
s
in
g
b
id
i
r
ec
tio
n
al
en
co
d
er
r
ep
r
esen
tatio
n
s
f
r
o
m
tr
a
n
s
f
o
r
m
er
s
(
B
E
R
T
)
[
8
]
.
Fo
r
th
e
I
n
d
o
n
esian
lan
g
u
ag
e,
alter
n
ativ
e
m
o
d
els
o
r
l
o
ca
lized
v
er
s
io
n
s
o
f
B
E
R
T
m
ay
b
e
n
ec
ess
ar
y
.
An
o
th
er
ap
p
r
o
ac
h
is
g
en
er
atin
g
n
u
m
e
r
ical
tex
t
r
ep
r
e
s
en
tatio
n
s
b
ased
o
n
ter
m
f
r
eq
u
en
c
y
-
in
v
er
s
e
d
o
cu
m
en
t
f
r
eq
u
e
n
cy
(
T
F
-
I
DF)
,
wh
ich
ca
n
b
e
co
m
b
in
ed
with
v
ar
io
u
s
alg
o
r
ith
m
s
[
9
]
.
Ma
ch
in
e
lear
n
i
n
g
m
eth
o
d
s
,
in
clu
d
in
g
s
u
p
p
o
r
t
v
ec
t
o
r
m
ac
h
i
n
e
(
SVM)
,
r
an
d
o
m
f
o
r
est
(
R
F),
ex
tr
e
m
e
g
r
ad
ien
t
b
o
o
s
tin
g
(
XGBo
o
s
t)
,
n
eu
r
al
n
etwo
r
k
s
(
NN)
,
an
d
B
E
R
T
,
h
av
e
b
ee
n
e
f
f
ec
tiv
e
i
n
class
if
y
in
g
d
if
f
e
r
en
t
d
ata
class
es
[
1
0
]
–
[
1
2
]
.
Mo
d
el
ev
alu
atio
n
is
cr
itical
f
o
r
en
s
u
r
i
n
g
r
eliab
le
p
r
ed
ictio
n
s
.
K
-
f
o
ld
cr
o
s
s
-
v
alid
atio
n
in
v
o
lv
es
d
iv
id
i
n
g
th
e
d
ataset
in
to
k
-
f
o
ld
s
,
u
s
in
g
ea
ch
f
o
l
d
f
o
r
test
in
g
wh
ile
tr
ain
in
g
o
n
th
e
o
th
er
s
,
an
d
r
ep
ea
tin
g
th
is
p
r
o
ce
s
s
k
tim
e
s
,
p
r
o
v
id
i
n
g
s
tab
le
p
e
r
f
o
r
m
an
ce
esti
m
atio
n
[
1
3
]
.
A
co
n
f
u
s
i
o
n
m
atr
ix
ass
ess
e
s
m
o
d
el
p
er
f
o
r
m
an
ce
u
s
in
g
m
e
tr
ics
s
u
ch
as
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
an
d
F1
-
s
co
r
e
[
1
4
]
.
Ad
d
itio
n
ally
,
th
e
Mc
Nem
ar
test
ca
n
co
m
p
ar
e
th
e
p
er
f
o
r
m
a
n
ce
o
f
two
m
o
d
els
o
n
th
e
s
am
e
d
ataset
to
d
eter
m
in
e
if
d
if
f
er
en
ce
s
in
p
r
ed
ictio
n
e
r
r
o
r
s
ar
e
s
tatis
tical
ly
s
ig
n
if
ican
t
[
1
5
]
.
T
h
is
r
esear
ch
p
r
o
p
o
s
es
d
ev
elo
p
in
g
a
m
ac
h
in
e
lear
n
in
g
m
o
d
el
to
au
to
m
atica
lly
class
if
y
s
o
cial
m
ed
ia
in
ter
ac
tio
n
s
as
co
m
p
lain
ts
o
r
n
o
n
-
c
o
m
p
lain
ts
,
en
h
a
n
cin
g
r
eso
u
r
ce
u
s
e
ef
f
ec
tiv
en
ess
an
d
s
o
cial
m
ed
ia
in
ter
ac
tio
n
m
an
a
g
em
en
t.
T
ec
h
n
iq
u
es
s
u
ch
as
B
E
R
T
an
d
T
F
-
I
DF
h
av
e
p
r
o
v
en
ef
f
ec
tiv
e
with
tex
t
d
ata
in
p
r
ev
io
u
s
s
tu
d
ies.
Alg
o
r
ith
m
s
l
ik
e
SVM,
R
F,
XGBo
o
s
t,
NN,
an
d
B
E
R
T
h
a
v
e
d
em
o
n
s
tr
ated
ef
f
icac
y
in
v
ar
i
o
u
s
class
if
icatio
n
task
s
.
E
v
alu
atin
g
th
ese
m
o
d
els
u
s
in
g
k
-
f
o
ld
c
r
o
s
s
-
v
alid
atio
n
,
co
n
f
u
s
io
n
m
a
tr
ix
,
an
d
Mc
Nem
ar
test
en
s
u
r
es
r
eliab
le
p
r
ed
ictio
n
s
.
T
h
e
k
ey
r
esear
ch
q
u
esti
o
n
is
:
“T
o
wh
at
ex
ten
t
ca
n
c
u
s
to
m
er
co
m
p
lain
t
in
ter
ac
tio
n
s
b
e
au
to
m
atica
lly
d
etec
ted
f
r
o
m
s
o
cial
m
ed
ia?”
T
h
is
m
o
d
el
aim
s
to
en
h
an
ce
o
p
er
atio
n
al
ef
f
icien
cy
an
d
s
u
p
p
o
r
t th
e
g
r
o
wth
o
f
I
n
d
o
n
esia'
s
e
-
co
m
m
er
c
e
s
ec
to
r
.
2.
M
E
T
H
O
D
T
h
e
s
tu
d
y
r
esear
ch
d
esig
n
ap
p
lies
th
e
cr
o
s
s
-
in
d
u
s
tr
y
s
tan
d
ar
d
p
r
o
ce
s
s
f
o
r
d
ata
m
in
in
g
(
C
R
I
SP
-
DM
)
f
r
am
ewo
r
k
,
ex
clu
d
in
g
th
e
d
ep
lo
y
m
en
t
p
h
ase.
C
R
I
SP
-
DM
i
s
a
g
en
er
al
r
ef
er
en
ce
f
o
r
d
ata
m
in
in
g
,
ex
p
lain
i
n
g
th
e
d
if
f
er
e
n
t
s
tag
es
o
f
th
e
d
at
a
m
in
in
g
p
r
o
ject
life
cy
cle.
T
h
is
m
o
d
el
is
k
n
o
wn
f
o
r
its
r
el
iab
ilit
y
,
s
im
p
licity
,
co
m
p
atib
ilit
y
,
f
lex
ib
ilit
y
in
its
ap
p
licatio
n
,
iter
ativ
e
p
r
o
ce
s
s
es,
an
d
ef
f
icien
t
tim
e
co
n
s
u
m
p
tio
n
[
1
6
]
.
T
h
is
s
tu
d
y
in
v
o
lv
es
s
ev
er
al
s
tag
es:
b
u
s
in
ess
u
n
d
er
s
tan
d
in
g
,
d
ata
u
n
d
er
s
tan
d
in
g
,
d
ata
p
r
ep
ar
ati
o
n
,
m
o
d
elin
g
,
an
d
ev
alu
atio
n
.
I
n
itially
,
it
is
c
r
u
cial
to
u
n
d
er
s
tan
d
th
e
p
r
o
b
lem
s
th
is
b
u
s
in
ess
f
ac
es
in
id
e
n
tify
in
g
an
d
a
d
d
r
ess
in
g
cu
s
to
m
er
in
ter
ac
tio
n
s
.
I
n
t
h
e
d
ata
u
n
d
er
s
tan
d
in
g
p
h
ase,
s
o
c
ial
m
ed
ia
d
ata
is
m
a
n
u
ally
an
n
o
tated
.
Du
r
in
g
th
e
d
ata
p
r
ep
ar
atio
n
p
h
ase,
d
ata
cl
ea
n
in
g
in
v
o
lv
es
r
em
o
v
i
n
g
s
p
e
cial
ch
ar
ac
ter
s
,
co
n
v
er
tin
g
tex
t
to
lo
wer
ca
s
e,
an
d
elim
in
atin
g
s
to
p
wo
r
d
s
.
T
h
e
clea
n
ed
tex
t
is
th
en
u
s
ed
f
o
r
f
ea
tu
r
e
ex
tr
ac
tio
n
an
d
co
n
v
er
ted
in
to
n
u
m
er
ic
v
ec
to
r
s
.
T
h
ese
v
ec
to
r
s
ar
e
u
s
ed
in
th
e
m
o
d
elin
g
s
tag
e
f
o
r
class
if
icatio
n
.
T
h
e
d
ata
is
s
p
li
t
in
to
5
-
f
o
ld
cr
o
s
s
-
v
alid
atio
n
an
d
class
if
ied
u
s
in
g
alg
o
r
ith
m
s
s
u
ch
as
SV
M,
R
F
,
XGBo
o
s
t,
N
N,
an
d
B
E
R
T
.
Mo
d
el
p
er
f
o
r
m
an
ce
is
ev
alu
ated
u
s
in
g
a
c
o
n
f
u
s
io
n
m
atr
ix
an
d
m
etr
ics
-
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
an
d
F1
-
m
ea
s
u
r
e
-
ap
p
ly
in
g
th
e
Mc
Nem
ar
test
to
d
eter
m
in
e
w
h
ich
m
o
d
el
b
est cla
s
s
if
ies cu
s
t
o
m
er
co
m
p
lain
ts
.
2
.
1
.
B
us
ines
s
un
dersta
nd
i
ng
T
h
er
e
ar
e
s
ev
er
al
lim
itatio
n
s
en
co
u
n
ter
e
d
in
im
p
lem
en
tin
g
s
o
cial
m
ed
ia
as
a
ch
a
n
n
el
f
o
r
cu
s
to
m
er
in
ter
ac
tio
n
s
.
T
h
ese
lim
itatio
n
s
in
clu
d
e
tech
n
ical
asp
ec
ts
,
o
p
e
r
atio
n
al
ch
allen
g
es,
an
d
h
u
m
a
n
r
eso
u
r
ce
s
th
at
ar
e
n
o
t
y
et
f
u
lly
o
p
tim
ized
to
lev
er
ag
e
th
e
f
u
ll
p
o
ten
tial
o
f
s
o
c
ial
m
ed
ia.
T
h
er
ef
o
r
e,
it
is
cr
u
cial
to
u
n
d
er
s
tan
d
h
o
w
th
e
co
m
p
an
y
cu
r
r
en
tly
id
en
tifie
s
an
d
m
an
a
g
es c
u
s
to
m
e
r
in
ter
ac
tio
n
s
o
cc
u
r
r
in
g
o
n
s
o
c
ial
m
ed
ia.
2
.
2
.
Da
t
a
un
dersta
nd
ing
T
h
r
o
u
g
h
th
e
d
ata
co
llectio
n
p
r
o
ce
s
s
,
th
is
s
tu
d
y
u
tili
ze
s
i
n
ter
ac
tio
n
d
ata
f
r
o
m
o
n
e
o
f
I
n
d
o
n
esia
e
-
co
m
m
er
ce
ac
c
o
u
n
ts
o
n
s
o
ci
al
m
ed
ia
X.
Data
co
llectio
n
was
co
n
d
u
cted
u
s
in
g
a
cr
awlin
g
m
eth
o
d
.
T
h
e
d
ata
was c
o
llected
f
r
o
m
in
ter
ac
tio
n
s
ex
clu
d
in
g
r
ep
lies
.
2
.
3
.
Da
t
a
prepa
ra
t
io
n
I
n
th
is
s
tag
e,
ir
r
elev
an
t
c
h
ar
a
cter
s
an
d
n
u
m
b
er
s
ar
e
f
ilter
e
d
o
u
t,
f
o
llo
wed
b
y
co
n
v
er
tin
g
all
v
alid
ch
ar
ac
ter
s
to
l
o
wer
ca
s
e
an
d
r
e
m
o
v
in
g
s
to
p
w
o
r
d
s
.
T
h
e
g
o
al
is
to
p
r
o
d
u
ce
clea
n
tex
t
r
ea
d
y
f
o
r
f
u
r
th
e
r
an
al
y
s
is
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
C
la
s
s
i
fica
tio
n
o
f c
u
s
to
mer
co
mp
la
in
ts
o
n
s
o
cia
l m
ed
ia
f
o
r
…
(
A
ch
ma
d
R
iz
ki
A
d
ita
ma
)
2979
[
1
7
]
.
T
h
is
is
cr
u
cial
as
s
o
cial
m
ed
ia
d
ata
o
f
ten
in
clu
d
es
r
an
d
o
m
te
x
ts
with
m
ea
n
in
g
less
wo
r
d
s
,
ch
a
r
ac
ter
s
,
o
r
em
o
jis
.
C
h
ar
ac
ter
f
ilter
in
g
r
e
m
o
v
es
m
ea
n
in
g
less
ch
ar
ac
ter
s
o
r
em
o
jis
an
d
u
n
r
elate
d
p
u
n
ct
u
atio
n
m
ar
k
s
,
u
s
in
g
r
eg
ex
to
elim
in
ate
all
p
u
n
ct
u
at
io
n
m
ar
k
s
.
C
ase
f
o
ld
in
g
ch
a
n
g
es
all
u
p
p
er
ca
s
e
letter
s
to
lo
wer
ca
s
e
to
u
n
if
y
th
e
tex
t
u
s
in
g
th
e
lo
wer
f
u
n
ctio
n
.
Sto
p
wo
r
d
s
ar
e
r
em
o
v
ed
u
s
in
g
a
r
u
le
-
b
ased
m
eth
o
d
with
a
lis
t
o
f
ir
r
elev
an
t
co
m
m
o
n
wo
r
d
s
.
Featu
r
e
ex
tr
a
ctio
n
th
en
tr
an
s
f
o
r
m
s
th
is
clea
n
ed
d
ata
in
to
r
ep
r
esen
tatio
n
s
s
u
itab
le
f
o
r
m
ac
h
in
e
lear
n
in
g
al
g
o
r
ith
m
s
b
y
id
en
tif
y
in
g
a
n
d
s
elec
tin
g
th
e
m
o
s
t
r
e
lev
an
t
attr
ib
u
tes
f
r
o
m
th
e
r
aw
d
ata.
T
h
is
e
n
h
an
ce
s
m
o
d
el
p
er
f
o
r
m
an
ce
b
y
co
n
v
er
tin
g
tex
t
d
ata
in
to
s
im
p
ler
n
u
m
e
r
ical
f
ea
tu
r
es,
allo
win
g
m
ac
h
i
n
e
lear
n
in
g
m
o
d
els
to
p
r
o
ce
s
s
an
d
an
aly
z
e
th
e
d
ata
m
o
r
e
ef
f
ec
tiv
ely
[
1
8
]
.
T
wo
m
eth
o
d
s
u
s
ed
f
o
r
f
ea
tu
r
e
ex
tr
ac
tio
n
ar
e
TF
-
I
DF a
n
d
B
E
R
T
.
TF
-
I
DF
m
ea
s
u
r
es
h
o
w
im
p
o
r
tan
t
a
wo
r
d
is
in
a
d
o
cu
m
en
t
co
m
p
ar
ed
t
o
o
th
er
d
o
cu
m
e
n
ts
.
T
er
m
f
r
eq
u
e
n
cy
T
F
c
o
u
n
ts
h
o
w
o
f
te
n
a
wo
r
d
ap
p
ea
r
s
,
wh
ile
in
v
er
s
e
d
o
cu
m
en
t
f
r
eq
u
en
cy
(
I
DF)
r
ed
u
ce
s
th
e
weig
h
t
o
f
co
m
m
o
n
wo
r
d
s
ap
p
ea
r
in
g
in
m
an
y
d
o
c
u
m
en
ts
.
T
h
e
co
m
b
in
atio
n
o
f
T
F
a
n
d
I
DF
g
iv
es
h
ig
h
er
s
co
r
es
to
s
ig
n
if
ican
t
wo
r
d
s
in
a
s
p
ec
if
ic
d
o
cu
m
en
t
co
n
te
x
t
[
1
9
]
.
W
h
ile
T
F
-
I
DF
is
p
o
p
u
lar
in
tr
ad
itio
n
al
m
ac
h
in
e
lear
n
in
g
,
B
E
R
T
is
an
ad
v
an
ce
d
m
o
d
el
b
ased
o
n
n
eu
r
al
n
etw
o
r
k
s
f
o
r
au
t
o
m
atic
f
ea
tu
r
e
ex
tr
ac
tio
n
f
r
o
m
tex
tu
a
l
d
o
cu
m
e
n
ts
[
2
0
]
.
T
r
ad
itio
n
al
m
eth
o
d
s
lik
e
T
F
-
I
DF
p
r
o
ce
s
s
tex
t
in
a
u
n
id
ir
ec
tio
n
al
m
an
n
e
r
,
eith
er
lef
t
t
o
r
ig
h
t
o
r
r
ig
h
t
t
o
lef
t,
p
o
ten
tially
m
is
s
in
g
im
p
o
r
tan
t
c
o
n
tex
tu
al
i
n
f
o
r
m
atio
n
.
B
E
R
T
u
n
d
e
r
s
tan
d
s
th
e
co
n
tex
t
o
f
wo
r
d
s
b
id
ir
ec
tio
n
ally
,
allo
win
g
f
o
r
d
ee
p
er
a
n
d
m
o
r
e
ac
cu
r
ate
tex
t
co
m
p
r
e
h
en
s
io
n
.
B
E
R
T
u
s
es
a
tr
an
s
f
o
r
m
er
ar
ch
itectu
r
e
with
an
atten
tio
n
m
ec
h
an
is
m
to
co
n
tex
t
u
alize
wo
r
d
s
b
ased
o
n
all
o
th
e
r
wo
r
d
s
in
th
e
s
en
ten
ce
an
d
ca
n
b
e
u
s
ed
f
o
r
v
ar
io
u
s
tex
t m
in
in
g
ap
p
licatio
n
s
,
in
clu
d
in
g
t
ex
t c
lass
if
icatio
n
[
2
1
]
.
2
.
4
.
Cla
s
s
if
ica
t
io
n
m
o
delin
g
T
h
e
n
e
x
t
s
tag
e,
class
if
icatio
n
m
o
d
elin
g
in
v
o
lv
es
ca
teg
o
r
izin
g
tex
t
in
to
p
r
ed
ef
in
ed
class
es
o
r
lab
els
u
s
in
g
th
e
ex
tr
ac
ted
f
ea
tu
r
es.
T
ex
t
class
if
icatio
n
is
a
tech
n
iq
u
e
in
tex
t
m
in
in
g
th
at
aim
s
to
ca
teg
o
r
ize
tex
t
in
to
p
r
ed
ef
in
e
d
class
es
o
r
lab
els
[
2
2
]
.
I
t
u
s
es
f
ea
tu
r
es
ex
tr
ac
ted
f
r
o
m
th
e
tex
t
t
o
lear
n
an
d
id
en
tify
p
atter
n
s
f
o
r
class
if
icatio
n
.
T
h
is
p
r
o
ce
s
s
cl
ass
if
ies
d
o
cu
m
en
ts
in
to
p
r
ed
e
f
in
ed
class
es
o
r
lab
els
b
ased
o
n
p
atter
n
s
f
o
r
m
ed
f
r
o
m
p
r
e
v
io
u
s
d
ata
[
1
9
]
.
I
n
th
i
s
s
tu
d
y
,
we
s
et
u
p
f
i
v
e
d
i
f
f
er
e
n
t
m
o
d
els
to
f
in
d
th
e
b
est
o
n
e,
s
u
ch
as
SVM,
R
F,
XGBo
o
s
t,
NN,
an
d
tr
an
s
f
o
r
m
e
r
-
b
ased
class
if
ier
.
SVM
is
a
p
o
p
u
lar
tech
n
i
q
u
e
f
o
r
class
if
y
in
g
d
o
cu
m
en
ts
u
s
in
g
d
is
cr
im
in
ativ
e
class
if
ier
s
.
SVM
wo
r
k
s
b
y
f
i
n
d
in
g
th
e
h
y
p
er
p
la
n
e
th
a
t
b
est
s
ep
ar
ates
d
ata
in
to
class
es.
I
n
itially
u
s
ed
f
o
r
b
in
a
r
y
class
if
icatio
n
task
s
,
SVMs
h
av
e
n
o
w
b
ee
n
ex
te
n
d
e
d
to
m
u
lti
-
class
p
r
o
b
lem
s
.
T
h
e
y
ar
e
v
er
s
atile
an
d
ca
n
b
e
ap
p
l
ied
in
v
ar
io
u
s
d
ata
m
in
in
g
ar
ea
s
,
in
cl
u
d
in
g
tex
t,
im
ag
es,
an
d
v
i
d
eo
s
[
1
3
]
.
R
F
is
an
en
s
em
b
le
lear
n
i
n
g
m
eth
o
d
th
at
co
m
b
in
es
m
u
ltip
le
d
ec
is
io
n
tr
ee
s
to
im
p
r
o
v
e
class
if
icatio
n
ac
cu
r
ac
y
.
E
ac
h
tr
ee
in
th
e
f
o
r
est
is
tr
ain
ed
u
s
in
g
a
r
an
d
o
m
ly
s
elec
ted
s
u
b
s
et
o
f
th
e
d
ata.
T
h
e
f
in
al
p
r
ed
ictio
n
is
m
ad
e
b
y
av
er
a
g
in
g
th
e
p
r
ed
ictio
n
s
o
f
all
th
e
tr
ee
s
.
T
h
is
m
eth
o
d
r
e
d
u
ce
s
o
v
er
f
itti
n
g
a
n
d
im
p
r
o
v
es th
e
m
o
d
el'
s
ab
ilit
y
to
g
en
er
alize
[
2
3
]
.
XGBo
o
s
t
is
an
ad
v
an
ce
d
e
n
s
e
m
b
le
m
et
h
o
d
th
at
u
s
es
b
o
o
s
ti
n
g
t
o
im
p
r
o
v
e
cl
ass
i
f
ic
ati
o
n
p
e
r
f
o
r
m
a
n
c
e.
I
t
b
u
il
d
s
tr
ee
s
s
eq
u
e
n
tial
ly
,
wit
h
ea
ch
tr
ee
f
i
x
i
n
g
th
e
e
r
r
o
r
s
o
f
t
h
e
p
r
e
v
i
o
u
s
o
n
e
.
I
t
is
k
n
o
w
n
f
o
r
i
ts
h
i
g
h
ef
f
ic
ie
n
c
y
a
n
d
a
cc
u
r
ac
y
,
m
a
k
in
g
i
t
a
p
o
p
u
l
ar
c
h
o
ic
e
f
o
r
m
an
y
m
a
c
h
i
n
e
le
ar
n
i
n
g
co
m
p
et
iti
o
n
s
[
2
4
]
.
NN
is
d
esi
g
n
e
d
t
o
l
ea
r
n
f
r
o
m
d
a
ta
t
h
r
o
u
g
h
m
u
lt
i
p
le
i
n
t
e
r
c
o
n
n
ec
t
ed
l
ay
e
r
s
o
f
n
o
d
es
o
r
n
e
u
r
o
n
s
.
E
ac
h
la
y
e
r
p
r
o
ce
s
s
es
th
e
i
n
p
u
t
d
ata
a
n
d
p
ass
es
it
t
o
th
e
n
e
x
t
l
a
y
e
r
,
g
r
a
d
u
al
ly
e
x
t
r
ac
ti
n
g
h
i
g
h
e
r
-
le
v
el
f
ea
t
u
r
es
.
Neu
r
a
l
n
et
wo
r
k
c
an
ac
c
o
m
m
o
d
at
e
c
o
m
p
l
ex
r
el
ati
o
n
s
h
ip
s
i
n
d
at
a,
m
ak
in
g
t
h
e
m
s
u
ita
b
le
f
o
r
a
wi
d
e
r
a
n
g
e
o
f
cl
ass
if
i
ca
t
io
n
tas
k
s
[
1
3
]
.
T
r
an
s
f
o
r
m
e
r
is
o
n
e
o
f
n
eu
r
al
n
etwo
r
k
ar
c
h
itectu
r
e
d
esig
n
e
d
to
h
an
d
le
s
eq
u
en
tial
d
ata,
s
u
ch
as
tex
t,
m
o
r
e
ef
f
ec
tiv
ely
.
I
t
im
p
lem
en
ts
a
m
ec
h
an
is
m
ca
lled
s
elf
-
att
en
tio
n
to
weig
h
th
e
im
p
o
r
tan
c
e
o
f
u
n
iq
u
e
wo
r
d
s
in
a
s
en
ten
ce
,
allo
win
g
th
e
m
o
d
el
to
u
n
d
er
s
tan
d
c
o
n
tex
t
b
e
tter
.
B
E
R
T
is
an
en
co
d
er
p
ar
t
o
f
th
e
tr
an
s
f
o
r
m
er
m
o
d
el
th
at
p
r
o
ce
s
s
es
tex
t
b
id
ir
ec
tio
n
ally
,
p
r
o
v
id
in
g
an
ac
cu
r
ate
v
ec
to
r
r
ep
r
esen
tatio
n
f
o
r
a
tex
tu
al
d
o
cu
m
en
t
[
2
0
]
,
[
2
5
]
.
A
d
d
itio
n
ally
,
we
u
s
e
th
e
p
r
e
-
tr
ain
ed
B
E
R
T
B
ah
asa
(
in
d
o
b
er
t
-
b
ase
-
p
1
)
ca
n
b
e
u
s
ed
n
o
t
o
n
ly
f
o
r
f
ea
tu
r
e
ex
tr
ac
tio
n
b
u
t
also
as
an
en
d
-
to
-
e
n
d
class
if
icatio
n
m
o
d
el
b
y
ad
d
in
g
a
f
ee
d
-
f
o
r
war
d
n
eu
r
al
n
etwo
r
k
(
FF
N)
lay
er
o
n
to
p
o
f
B
E
R
T
[
2
6
]
.
E
ac
h
m
o
d
elin
g
alg
o
r
ith
m
is
r
u
n
f
iv
e
tim
es
u
s
in
g
th
e
k
-
f
o
ld
m
eth
o
d
.
T
h
is
m
eth
o
d
is
u
s
ed
to
en
s
u
r
e
th
at
th
e
m
o
d
el'
s
p
er
f
o
r
m
an
ce
is
co
n
s
is
ten
t
an
d
r
eliab
le
b
y
av
er
ag
in
g
th
e
r
esu
lts
o
v
e
r
m
u
ltip
le
tr
ain
in
g
a
n
d
test
in
g
s
p
lits
,
r
ed
u
cin
g
th
e
v
ar
ian
ce
ass
o
ciate
d
with
a
s
in
g
le
tr
ain
-
test
s
p
lit
[
2
7
]
.
T
h
e
en
tire
d
ataset
is
d
iv
id
ed
in
to
f
iv
e
v
er
s
io
n
s
o
f
tr
ain
in
g
a
n
d
test
in
g
d
ata.
I
n
t
h
e
f
ir
s
t
iter
atio
n
,
th
e
f
ir
s
t
⅕
o
f
th
e
d
ata
w
ill
b
e
th
e
test
d
ata
an
d
th
e
r
est
will
b
e
th
e
tr
ain
i
n
g
d
ata.
T
h
is
p
r
o
ce
s
s
r
ep
ea
ts
u
n
til
th
e
last
⅕
o
f
th
e
d
ata
b
e
co
m
es
th
e
test
d
ata
an
d
th
e
r
est
is
tr
ain
in
g
d
ata.
T
h
e
p
r
ed
ictio
n
r
esu
lts
f
r
o
m
t
h
e
test
d
ata
in
ea
ch
iter
atio
n
f
o
r
ea
ch
m
o
d
el
a
r
e
co
m
b
in
ed
an
d
e
v
alu
ated
b
ased
o
n
th
e
o
r
ig
in
al
d
ataset.
2
.
5
.
E
v
a
lua
t
io
n
T
h
e
ev
al
u
atio
n
o
f
class
if
icatio
n
m
o
d
els
u
tili
ze
s
th
e
c
o
n
f
u
s
io
n
m
atr
ix
m
eth
o
d
,
a
ta
b
le
u
s
ed
to
ass
ess
th
e
p
er
f
o
r
m
an
ce
o
f
class
if
icatio
n
m
o
d
els
b
y
d
is
p
lay
in
g
p
r
e
d
ictio
n
s
in
r
o
ws
an
d
ac
tu
al
d
e
cisi
o
n
s
in
co
lu
m
n
s
,
co
n
s
is
tin
g
o
f
tr
u
e
p
o
s
itiv
e
(
T
P),
tr
u
e
n
eg
ativ
e
(
T
N)
,
f
alse
p
o
s
itiv
e
(
FP
)
,
an
d
f
alse
n
eg
ativ
e
(
FN)
.
C
o
m
m
o
n
ly
u
s
ed
m
etr
ics
in
clu
d
e
ac
c
u
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
an
d
F1
-
s
co
r
e.
T
h
ese
e
v
alu
atio
n
m
etr
ic
s
co
m
p
lem
en
t
ea
c
h
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
25
:
2
9
7
7
-
2
9
8
5
2980
o
th
er
to
p
r
o
v
id
e
a
co
m
p
r
eh
e
n
s
iv
e
p
ictu
r
e
o
f
th
e
m
o
d
el'
s
p
e
r
f
o
r
m
a
n
ce
.
Acc
u
r
ac
y
g
iv
es
a
n
o
v
e
r
all
v
iew
(
1
)
,
p
r
ec
is
io
n
in
f
o
r
m
s
ab
o
u
t
th
e
ac
cu
r
ac
y
o
f
p
o
s
itiv
e
p
r
ed
icti
o
n
s
(
2
)
,
r
ec
all
m
ea
s
u
r
es
th
e
m
o
d
el'
s
ab
ilit
y
t
o
id
en
tify
tr
u
e
p
o
s
itiv
es (
3
)
,
an
d
F1
-
s
co
r
e
co
m
b
in
es p
r
ec
is
io
n
an
d
r
ec
all
f
o
r
an
o
v
er
all
e
v
alu
atio
n
(
4
)
.
=
(
+
)
(
+
+
+
)
(
1
)
=
(
+
)
(
2
)
=
(
+
)
(
3
)
1
=
(
(
2
∗
∗
)
(
+
)
)
(
4
)
Ad
d
itio
n
ally
,
Mc
Nem
ar
test
is
a
s
tatis
tical
m
eth
o
d
u
s
ed
to
co
m
p
ar
e
th
e
p
e
r
f
o
r
m
an
ce
o
f
two
class
if
icatio
n
m
o
d
els
o
n
th
e
s
am
e
d
ataset.
A
th
r
esh
o
ld
o
f
0
.
0
5
is
co
m
m
o
n
ly
u
s
ed
to
d
eter
m
in
e
s
tatis
tical
s
ig
n
if
ican
ce
in
h
y
p
o
t
h
esis
tes
tin
g
,
in
clu
d
in
g
Mc
Nem
ar
test
,
wh
er
e
a
p
-
v
alu
e
o
f
less
th
an
0
.
0
5
in
d
icate
s
a
s
tatis
t
ically
s
ig
n
if
ican
t
d
if
f
er
en
ce
b
etwe
en
m
o
d
el
p
er
f
o
r
m
an
ce
s
.
T
h
is
m
eth
o
d
is
s
u
i
tab
le
wh
en
d
ata
is
ca
teg
o
r
ized
as
co
r
r
ec
t
o
r
in
c
o
r
r
ec
t
p
r
ed
ictio
n
s
b
y
b
o
th
m
o
d
els.
Mc
Nem
ar
test
ca
lcu
la
tes
th
e
d
if
f
e
r
en
ce
b
etwe
en
th
e
n
u
m
b
er
o
f
er
r
o
r
s
m
ad
e
b
y
b
o
th
m
o
d
els
a
n
d
test
s
f
o
r
s
tatis
tical
s
ig
n
if
ican
ce
,
h
elp
in
g
to
d
eter
m
in
e
if
th
e
d
if
f
e
r
en
ce
in
p
r
ed
ictiv
e
er
r
o
r
s
b
etwe
en
t
h
e
two
m
o
d
els is
s
ig
n
if
ican
t.
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
Sev
er
al
s
tag
es we
r
e
co
n
d
u
cted
to
ex
am
in
e
th
e
class
if
icatio
n
o
f
cu
s
to
m
er
c
o
m
p
lain
ts
o
n
s
o
c
ial
m
ed
ia.
I
n
itially
,
cu
s
to
m
er
in
ter
ac
tio
n
d
ata
f
r
o
m
s
o
cial
m
ed
ia
was
co
llected
an
d
m
an
u
ally
lab
eled
.
T
h
e
p
r
ep
r
o
ce
s
s
in
g
s
tag
e
in
v
o
lv
ed
clea
n
in
g
th
e
d
ata
b
y
r
em
o
v
in
g
ir
r
ele
v
an
t
ch
ar
ac
ter
s
,
co
n
v
e
r
tin
g
all
tex
t
to
lo
wer
ca
s
e,
an
d
elim
in
atin
g
s
to
p
wo
r
d
s
.
C
lean
ed
tex
t
was
th
en
co
n
v
er
ted
i
n
to
n
u
m
e
r
ical
v
ec
to
r
s
f
o
r
f
ea
tu
r
e
ex
tr
ac
tio
n
u
s
in
g
TF
-
I
DF
an
d
B
E
R
T
.
T
h
ese
v
ec
to
r
s
wer
e
u
s
ed
in
th
e
class
if
icatio
n
m
o
d
elin
g
s
tag
e,
w
h
er
e
th
e
d
ata
was
s
p
lit
u
s
in
g
5
-
f
o
ld
cr
o
s
s
-
v
alid
atio
n
a
n
d
class
if
ied
u
s
in
g
f
iv
e
d
if
f
e
r
e
n
t m
o
d
els:
SVM,
R
F,
XG
B
o
o
s
t,
NN,
an
d
B
E
R
T
.
E
ac
h
m
o
d
el
was
r
u
n
m
u
ltip
le
tim
es
u
s
in
g
th
e
k
-
f
o
ld
m
et
h
o
d
f
o
r
co
n
s
is
ten
cy
an
d
r
eliab
ili
ty
.
Fin
ally
,
m
o
d
el
p
er
f
o
r
m
an
ce
was e
v
alu
ated
u
s
in
g
a
co
n
f
u
s
io
n
m
atr
ix
an
d
m
e
tr
ics lik
e
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
F1
-
m
ea
s
u
r
e,
an
d
th
e
Mc
Nem
ar
test
to
d
eter
m
in
e
th
e
b
est m
o
d
el
f
o
r
class
i
f
y
in
g
c
u
s
to
m
er
co
m
p
lain
ts
.
3
.
1
.
Da
t
a
co
llect
io
n
T
h
is
s
tu
d
y
u
s
es
in
ter
ac
tio
n
d
ata
f
r
o
m
two
s
o
cial
m
ed
ia
ac
co
u
n
ts
o
n
X,
b
o
t
h
b
el
o
n
g
in
g
to
a
s
in
g
le
I
n
d
o
n
esian
e
-
co
m
m
er
ce
co
m
p
an
y
.
On
e
ac
co
u
n
t
is
d
ed
icate
d
to
h
an
d
lin
g
co
m
p
lain
ts
(
ac
co
u
n
t
A)
,
wh
ile
th
e
o
th
er
s
er
v
es
as
t
h
e
g
e
n
er
al
c
o
m
p
an
y
ac
co
u
n
t
(
ac
c
o
u
n
t
B
)
.
Data
co
llectio
n
was
co
n
d
u
cte
d
u
s
in
g
a
cr
awlin
g
m
eth
o
d
,
f
o
cu
s
in
g
o
n
I
n
d
o
n
esian
-
lan
g
u
a
g
e
co
n
ten
t d
u
e
to
its
r
elev
an
ce
to
th
e
I
n
d
o
n
esian
e
-
co
m
m
er
ce
s
ec
to
r
.
T
h
e
r
esu
lts
w
er
e
m
an
u
all
y
an
n
o
tated
b
ased
o
n
p
r
ed
ef
i
n
ed
cr
iter
ia
to
en
s
u
r
e
c
o
n
s
is
ten
t
lab
elin
g
o
f
co
m
p
lain
ts
an
d
n
o
n
-
c
o
m
p
lain
ts
.
T
h
e
an
n
o
tatio
n
p
r
o
ce
s
s
in
v
o
lv
ed
th
r
ee
t
r
ain
ed
a
n
n
o
tato
r
s
,
wh
o
ar
e
ty
p
ically
r
esp
o
n
s
ib
le
f
o
r
h
an
d
lin
g
an
d
p
r
o
ce
s
s
in
g
cu
s
to
m
er
co
m
p
lai
n
ts
,
en
s
u
r
in
g
th
at
th
eir
ex
p
er
t
is
e
was
ap
p
lied
in
ac
cu
r
ately
lab
elin
g
th
e
s
o
cial
m
ed
ia
in
ter
ac
tio
n
s
.
I
n
ter
-
an
n
o
tato
r
ag
r
ee
m
en
t
was
ca
lcu
lated
to
e
n
s
u
r
e
r
eliab
ilit
y
an
d
co
n
s
is
ten
cy
in
th
e
lab
elin
g
p
r
o
ce
s
s
.
Du
r
in
g
th
e
co
llectio
n
p
er
io
d
,
1
0
,
7
9
2
in
ter
ac
tio
n
s
wer
e
g
ath
er
ed
,
o
f
wh
ic
h
6
,
2
2
0
wer
e
lab
eled
as
co
m
p
lain
ts
an
d
4
,
5
7
2
as
n
o
n
-
co
m
p
lain
ts
.
A
f
ter
f
ilter
in
g
u
s
in
g
wh
itelis
ted
k
ey
wo
r
d
s
,
th
e
f
in
al
d
ataset
co
m
p
r
is
ed
6
,
2
0
4
c
o
m
p
lain
ts
an
d
4
,
3
9
5
n
o
n
-
co
m
p
lain
ts
,
as
s
h
o
wn
in
T
ab
le
1
.
T
ab
le
1
.
Data
d
is
tr
ib
u
tio
n
I
n
t
e
r
a
c
t
i
o
n
C
o
m
p
l
a
i
n
t
N
o
n
-
c
o
m
p
l
a
i
n
t
M
e
n
t
i
o
n
i
n
g
a
c
c
o
u
n
t
A
2
,
6
4
9
7
8
9
M
e
n
t
i
o
n
i
n
g
a
c
c
o
u
n
t
B
1
,
6
9
2
3
,
6
9
3
M
e
n
t
i
o
n
i
n
g
b
o
t
h
a
c
c
o
u
n
t
1
,
8
7
0
89
3.
2
.
M
o
del
perf
o
rm
a
nce
B
ased
o
n
th
e
r
esu
lts
in
T
ab
le
s
2
an
d
3
,
it c
an
b
e
co
n
cl
u
d
ed
th
at
B
E
R
T
m
o
d
el
with
a
FF
N
cl
ass
if
ier
is
th
e
b
est
-
p
er
f
o
r
m
in
g
m
o
d
el.
T
h
is
m
o
d
el
ac
h
iev
es
an
im
p
r
ess
iv
e
o
v
er
all
F1
-
s
co
r
e
o
f
9
7
.
6
9
%
an
d
an
ac
cu
r
ac
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
C
la
s
s
i
fica
tio
n
o
f c
u
s
to
mer
co
mp
la
in
ts
o
n
s
o
cia
l m
ed
ia
f
o
r
…
(
A
ch
ma
d
R
iz
ki
A
d
ita
ma
)
2981
o
f
9
7
.
7
5
%,
h
ig
h
lig
h
tin
g
its
ef
f
ec
tiv
en
ess
in
b
o
th
f
ea
tu
r
e
e
x
tr
ac
tio
n
an
d
class
if
icatio
n
task
s
.
Fo
r
th
e
c
o
m
p
lain
t
class
,
B
E
R
T
-
FF
N
ac
h
iev
es
an
F1
-
s
co
r
e
o
f
9
8
.
0
8
%,
s
ig
n
if
ic
an
tly
o
u
tp
e
r
f
o
r
m
in
g
o
th
er
m
o
d
els,
d
em
o
n
s
tr
atin
g
its
s
u
p
er
io
r
ab
ilit
y
to
ac
c
u
r
atel
y
id
en
tif
y
an
d
class
if
y
co
m
p
la
in
t
in
s
tan
ce
s
.
I
n
th
e
n
o
n
-
co
m
p
lain
t
class
,
B
E
R
T
-
FF
N
al
s
o
ac
h
iev
es
an
F1
-
s
co
r
e
o
f
9
7
.
2
9
%,
in
d
icatin
g
a
b
a
lan
ce
d
p
er
f
o
r
m
an
ce
ac
r
o
s
s
d
if
f
er
en
t
class
es
an
d
en
s
u
r
in
g
h
i
g
h
p
r
ec
is
io
n
an
d
r
e
ca
ll r
ates.
C
o
m
p
a
r
a
ti
v
el
y
,
o
t
h
e
r
m
o
d
els
s
u
c
h
as
T
F
-
I
DF
w
it
h
S
VM
,
R
F
,
N
N,
a
n
d
X
GB
o
o
s
t
s
h
o
w
c
o
m
m
e
n
d
a
b
le
p
e
r
f
o
r
m
a
n
c
e
b
u
t
f
al
l
s
h
o
r
t
o
f
B
E
R
T
-
FF
N.
T
F
-
I
DF
wit
h
SV
M,
f
o
r
i
n
s
t
an
ce
,
ac
h
i
e
v
es
a
n
o
v
e
r
al
l
F1
-
s
c
o
r
e
o
f
9
3
.
0
2
%
an
d
a
n
ac
c
u
r
ac
y
o
f
9
3
.
2
8
%,
w
h
i
le
T
F
-
I
DF
wit
h
R
F
s
c
o
r
es
a
n
o
v
e
r
a
ll
F1
-
s
co
r
e
o
f
9
1
.
8
6
%
an
d
a
n
ac
c
u
r
a
cy
o
f
9
2
.
2
0
%
.
Similar
l
y
,
T
F
-
I
DF
with
NN
r
ec
o
r
d
s
a
n
o
v
er
all
F1
-
s
co
r
e
o
f
9
1
.
5
6
%
an
d
an
ac
c
u
r
ac
y
o
f
9
1
.
8
5
%,
an
d
T
F
-
I
DF
with
XG
B
o
o
s
t
ac
h
iev
es
an
o
v
er
all
F1
-
s
co
r
e
o
f
9
2
.
0
5
%
an
d
an
ac
cu
r
ac
y
o
f
9
2
.
3
0
%.
T
h
e
s
ize
o
f
th
e
d
ata
u
s
ed
f
o
r
t
r
ain
i
n
g
an
d
ev
alu
atio
n
is
co
n
s
is
ten
t a
cr
o
s
s
m
o
d
els.
Fro
m
Mc
Nem
ar
test
r
esu
lts
,
s
ig
n
if
ican
t
d
if
f
er
en
ce
s
wer
e
f
o
u
n
d
b
etwe
en
s
ev
er
al
m
o
d
el
p
air
s
,
in
d
icatin
g
th
eir
p
er
f
o
r
m
a
n
ce
d
if
f
er
en
ce
s
.
T
h
e
th
r
esh
o
ld
o
f
0
.
0
5
is
co
m
m
o
n
l
y
u
s
ed
to
d
eter
m
in
e
s
tatis
tical
s
ig
n
if
ican
ce
in
h
y
p
o
th
esis
test
in
g
,
in
clu
d
in
g
Mc
Nem
ar
te
s
t,
wh
er
e
a
p
-
v
alu
e
less
th
a
n
0
.
0
5
i
n
d
icate
s
a
s
tatis
t
ically
s
ig
n
if
ican
t
d
if
f
er
e
n
ce
b
etwe
en
m
o
d
el
p
er
f
o
r
m
a
n
ce
s
[
2
8
]
.
T
h
e
Mc
Nem
ar
test
r
esu
lts
ca
n
b
e
s
ee
n
in
T
ab
le
4
.
T
ab
le
2
.
Mo
d
el’
s
p
er
f
o
r
m
an
ce
f
o
r
ea
c
h
class
F
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
C
l
a
s
si
f
i
e
r
C
o
m
p
l
a
i
n
t
N
o
n
-
c
o
m
p
l
a
i
n
t
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F1
-
sc
o
r
e
S
i
z
e
d
a
t
a
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F1
-
sc
o
r
e
S
i
z
e
d
a
t
a
TF
-
I
D
F
S
V
M
9
2
.
5
9
%
9
6
.
2
3
%
9
4
.
3
7
%
6
,
2
0
4
9
4
.
3
7
%
8
9
.
1
3
%
9
1
.
6
7
%
4
,
3
9
6
TF
-
I
D
F
RF
9
0
.
9
3
%
9
6
.
2
8
%
9
3
.
5
3
%
6
,
2
0
4
9
4
.
2
7
%
8
6
.
4
5
%
9
0
.
1
9
%
4
,
3
9
6
TF
-
I
D
F
NN
9
1
.
9
3
%
9
4
.
3
6
%
9
3
.
1
3
%
6
,
2
0
4
9
1
.
7
3
%
8
8
.
3
1
%
8
9
.
9
9
%
4
,
3
9
6
TF
-
I
D
F
X
G
B
o
o
st
9
2
.
8
0
%
9
4
.
1
5
%
9
3
.
4
7
%
6
,
2
0
4
9
1
.
5
7
%
8
9
.
7
0
%
9
0
.
6
2
%
4
,
3
9
6
B
ER
T
FFN
9
8
.
0
4
%
9
8
.
1
3
%
9
8
.
0
8
%
6
,
2
0
4
9
7
.
3
6
%
9
7
.
2
3
%
9
7
.
2
9
%
4
,
3
9
6
T
ab
le
3
.
Ov
e
r
all
m
o
d
el’
s
p
e
r
f
o
r
m
an
ce
F
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
C
l
a
s
si
f
i
e
r
A
l
l
c
l
a
s
s
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F1
-
sc
o
r
e
A
c
c
u
r
a
c
y
S
i
z
e
d
a
t
a
TF
-
I
D
F
S
V
M
93
.
4
8
%
92
.
6
8
%
93
.
0
2
%
93
.
2
8
%
10
,
600
TF
-
I
D
F
RF
92
.
6
0
%
91
.
3
6
%
91
.
8
6
%
92
.
2
0
%
10
,
600
TF
-
I
D
F
NN
91
.
8
3
%
91
.
3
3
%
91
.
5
6
%
91
.
8
5
%
10
,
600
TF
-
I
D
F
X
G
B
o
o
st
92
.
1
9
%
91
.
9
2
%
92
.
0
5
%
92
.
3
0
%
10
,
600
B
ER
T
FFN
97
.
7
7
%
97
.
6
8
%
97
.
6
9
%
97
.
7
5
%
10
,
600
T
ab
le
4
.
Mc
Nem
ar
test
r
esu
lt
M
o
d
e
l
p
a
i
r
p
-
v
a
l
u
e
S
V
M
a
n
d
R
F
<
0
.
0
5
S
V
M
a
n
d
N
N
<
0
.
0
5
S
V
M
a
n
d
X
G
B
o
o
st
<
0
.
0
5
S
V
M
a
n
d
B
ER
T
<
0
.
0
5
R
F
a
n
d
N
N
0
.
2
1
R
F
a
n
d
X
G
B
o
o
s
t
0
.
6
2
R
F
a
n
d
B
ER
T
<
0
.
0
5
N
N
a
n
d
X
G
B
o
o
st
0
.
1
1
N
N
a
n
d
B
ER
T
<
0
.
0
5
X
G
B
o
o
st
a
n
d
B
E
R
T
<
0
.
0
5
Sig
n
if
ican
t
d
if
f
e
r
en
ce
s
(
p
-
v
al
u
e
<
0
.
0
5
)
wer
e
o
b
s
er
v
ed
b
e
twee
n
s
ev
er
al
m
o
d
el
p
ai
r
s
:
SVM
an
d
r
an
d
o
m
f
o
r
est
,
SVM
an
d
NN,
SVM
an
d
XGBo
o
s
t,
SVM
a
n
d
FF
N,
r
an
d
o
m
f
o
r
est
an
d
F
FN,
NN
an
d
FF
N,
an
d
XGBo
o
s
t
an
d
FF
N.
T
h
is
m
ea
n
s
th
ese
m
o
d
el
p
air
s
d
if
f
e
r
s
ig
n
if
ican
tly
in
p
er
f
o
r
m
an
ce
.
On
th
e
o
t
h
er
h
a
n
d
,
n
o
s
ig
n
if
ican
t
d
if
f
er
e
n
ce
s
(
p
-
v
alu
e
>
0
.
0
5
)
wer
e
f
o
u
n
d
b
et
wee
n
r
an
d
o
m
f
o
r
est
an
d
NN,
r
an
d
o
m
f
o
r
est
an
d
XGBo
o
s
t,
an
d
NN
a
n
d
XGBo
o
s
t.
T
h
is
in
d
icate
s
th
at
th
ese
m
o
d
el
p
air
s
p
er
f
o
r
m
s
im
ilar
ly
in
class
if
y
in
g
th
e
d
ata.
I
n
o
th
er
w
o
r
d
s
,
th
e
ef
f
e
ctiv
en
ess
o
f
th
ese
m
o
d
els
in
class
if
icatio
n
task
s
is
co
m
p
ar
ab
le,
an
d
ch
o
o
s
in
g
b
etwe
en
th
em
m
a
y
d
ep
e
n
d
o
n
f
ac
to
r
s
lik
e
co
m
p
u
tatio
n
al
ef
f
icien
cy
o
r
ea
s
e
o
f
im
p
lem
e
n
tatio
n
.
3.
3
.
M
o
del
im
plem
ent
a
t
io
n
B
ased
o
n
th
e
b
est
class
if
icatio
n
m
o
d
el,
th
e
im
p
lem
en
tatio
n
ca
n
u
s
e
p
r
o
b
a
b
ilit
y
v
al
u
es,
co
n
f
id
en
ce
s
co
r
es,
an
d
th
r
esh
o
ld
s
to
d
ec
i
d
e
if
th
e
au
to
m
ated
r
esu
lts
ar
e
r
eliab
le
o
r
n
ee
d
m
an
u
al
r
e
v
iew.
T
h
is
h
elp
s
to
r
ed
u
ce
FP
,
FN,
an
d
im
p
r
o
v
e
o
v
er
all
ac
cu
r
ac
y
a
n
d
p
r
ec
is
io
n
.
As
s
h
o
wn
in
Fig
u
r
e
1
,
th
e
er
r
o
r
d
is
tr
ib
u
tio
n
f
o
r
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
25
:
2
9
7
7
-
2
9
8
5
2982
FP
is
v
is
u
alize
d
u
s
in
g
k
er
n
el
d
en
s
ity
esti
m
atio
n
(
KDE
)
.
Fig
u
r
e
2
illu
s
tr
ates
th
e
KDE
v
i
s
u
aliza
tio
n
f
o
r
FN
.
T
h
ese
v
is
u
aliza
tio
n
s
m
ak
e
it
ea
s
y
to
s
ee
wh
er
e
th
e
m
o
d
el
wo
r
k
s
well
an
d
wh
e
r
e
it
n
ee
d
s
im
p
r
o
v
em
en
t,
h
ig
h
lig
h
tin
g
ar
ea
s
with
h
ig
h
er
r
o
r
d
e
n
s
ity
an
d
id
e
n
tify
in
g
p
at
ter
n
s
th
at
m
ig
h
t r
e
q
u
ir
e
f
u
r
th
e
r
tu
n
in
g
.
T
h
e
KDE
d
iag
r
am
ab
o
v
e
h
e
lp
s
id
en
tify
wh
er
e
t
h
e
p
r
o
b
a
b
ilit
y
o
f
th
e
m
o
d
el
is
lik
ely
to
m
ak
e
m
is
tak
es,
allo
win
g
f
o
r
b
etter
d
ec
is
io
n
-
m
ak
in
g
wh
e
n
r
ev
ie
win
g
u
n
ce
r
tain
p
r
e
d
ictio
n
s
.
T
h
is
ap
p
r
o
ac
h
ca
n
im
p
r
o
v
e
th
e
o
v
er
all
p
er
f
o
r
m
a
n
ce
o
f
th
e
class
if
icatio
n
s
y
s
tem
.
T
ab
le
5
s
h
o
ws
th
e
s
tatis
ti
ca
l
m
ea
s
u
r
es
o
f
th
e
m
o
d
el
f
o
r
p
r
e
d
ictin
g
th
e
c
h
an
c
es o
f
FP
an
d
FN.
Fig
u
r
e
1
.
KDE
f
o
r
FP
b
ased
o
n
p
r
o
b
ab
ilit
y
Fig
u
r
e
2
.
KDE
f
o
r
FN b
ased
o
n
p
r
o
b
ab
ilit
y
T
ab
le
5
.
Statis
tical
v
alu
es
o
f
FP
an
d
FN
FP
FN
mea
n
0
.
8
9
0
.
8
9
mi
n
0
.
5
1
0
.
5
8
p
2
5
0
.
8
6
0
.
8
9
p
5
0
0
.
9
6
0
.
9
6
p
7
5
0
.
9
8
0
.
9
7
max
1
.
0
0
1
.
0
0
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
C
la
s
s
i
fica
tio
n
o
f c
u
s
to
mer
co
mp
la
in
ts
o
n
s
o
cia
l m
ed
ia
f
o
r
…
(
A
ch
ma
d
R
iz
ki
A
d
ita
ma
)
2983
T
o
d
eter
m
in
e
t
h
e
o
p
tim
al
p
r
o
b
ab
ilit
y
th
r
esh
o
ld
,
an
an
aly
s
is
is
n
ee
d
ed
to
b
alan
c
e
th
e
tr
ad
e
-
o
f
f
s
b
etwe
en
d
if
f
e
r
en
t
ty
p
es
o
f
e
r
r
o
r
s
an
d
th
e
b
e
n
ef
its
o
f
T
P
an
d
T
N
o
u
tco
m
es.
Fo
r
co
m
p
lain
ts
,
b
ased
o
n
th
e
s
tatis
t
ical
v
alu
es,
a
h
ig
h
er
th
r
esh
o
ld
,
ar
o
u
n
d
th
e
m
e
d
ian
FN
v
alu
e
o
f
0
.
9
6
o
r
h
ig
h
e
r
,
is
r
ec
o
m
m
en
d
e
d
to
m
in
im
ize
FN.
Fo
r
n
o
n
-
co
m
p
lain
ts
,
lo
wer
in
g
th
e
p
r
o
b
ab
ili
ty
th
r
esh
o
ld
h
elp
s
in
co
r
r
ec
tl
y
id
en
tify
in
g
n
o
n
-
co
m
p
lain
ts
,
th
er
e
b
y
r
ed
u
ci
n
g
FP
an
d
in
cr
ea
s
in
g
T
N.
A
lo
w
er
th
r
esh
o
l
d
,
a
r
o
u
n
d
t
h
e
2
5
th
p
er
ce
n
tile
FP
v
alu
e
o
f
0
.
8
6
o
r
lo
wer
,
is
r
ec
o
m
m
en
d
ed
to
m
i
n
im
ize
FP
.
T
h
ese
th
r
esh
o
ld
s
en
s
u
r
e
a
b
a
lan
ce
d
tr
ad
e
-
o
f
f
b
etwe
en
d
if
f
er
en
t
ty
p
es
o
f
er
r
o
r
s
an
d
e
n
h
an
ce
th
e
o
v
er
all
p
er
f
o
r
m
a
n
ce
o
f
th
e
c
lass
if
icatio
n
m
o
d
el.
Ad
d
itio
n
ally
,
f
in
e
-
tu
n
i
n
g
th
ese
th
r
esh
o
ld
s
b
ased
o
n
th
e
s
p
ec
if
ic
co
n
tex
t
an
d
th
e
im
p
o
r
tan
ce
o
f
d
if
f
er
e
n
t
er
r
o
r
ty
p
es
ca
n
f
u
r
th
er
im
p
r
o
v
e
th
e
m
o
d
el'
s
ac
cu
r
ac
y
an
d
r
eliab
ilit
y
.
R
eg
u
lar
ly
r
ea
s
s
ess
in
g
an
d
ad
j
u
s
tin
g
th
e
th
r
esh
o
ld
s
as
n
ew
d
ata
b
ec
o
m
es
av
ailab
le
will
h
elp
m
ain
tain
o
p
tim
al
p
e
r
f
o
r
m
an
c
e
o
v
er
tim
e.
Fu
r
th
er
m
o
r
e,
i
n
co
r
p
o
r
atin
g
d
o
m
ain
ex
p
er
ti
s
e
in
s
ettin
g
th
ese
th
r
esh
o
ld
s
ca
n
p
r
o
v
i
d
e
v
alu
a
b
le
in
s
ig
h
ts
,
en
s
u
r
in
g
th
e
m
o
d
el
alig
n
s
with
p
r
ac
tical
b
u
s
in
ess
n
ee
d
s
an
d
o
b
jectiv
es.
I
n
im
p
lem
e
n
tin
g
th
e
m
o
d
el
i
n
r
ea
l
-
wo
r
ld
s
ettin
g
s
,
t
h
e
in
teg
r
a
tio
n
with
ex
is
tin
g
cu
s
to
m
er
r
el
atio
n
s
h
ip
m
an
ag
em
en
t
(
C
R
M)
s
y
s
tem
s
is
f
ea
s
ib
le
th
r
o
u
g
h
ap
p
licatio
n
p
r
o
g
r
am
m
in
g
in
ter
f
ac
e
(
API
)
.
T
h
is
in
teg
r
atio
n
f
ac
ilit
ates
r
ea
l
-
tim
e
co
m
p
lain
t
id
en
tific
atio
n
an
d
r
esp
o
n
s
e,
en
ab
lin
g
co
m
p
an
ies
to
ad
d
r
ess
cu
s
to
m
er
is
s
u
es
p
r
o
m
p
tly
a
n
d
ef
f
ec
tiv
ely
.
T
h
e
m
o
d
el
is
ex
p
ec
ted
to
p
r
o
ce
s
s
in
p
u
t
tex
t
in
th
e
f
o
r
m
o
f
s
en
te
n
ce
s
o
b
tain
ed
f
r
o
m
v
ar
io
u
s
s
o
cial
m
ed
ia
p
latf
o
r
m
s
,
allo
win
g
f
o
r
a
th
o
r
o
u
g
h
class
if
icatio
n
o
f
in
ter
ac
tio
n
s
as
e
ith
er
co
m
p
lain
ts
o
r
non
-
c
o
m
p
lain
ts
.
T
h
e
d
esire
d
o
u
tp
u
t
f
r
o
m
th
is
m
o
d
el
is
a
co
n
f
id
en
ce
s
co
r
e
f
o
r
ea
ch
o
f
th
e
d
ef
in
e
d
lab
e
ls
,
wh
ich
q
u
an
tifie
s
th
e
m
o
d
el’
s
ce
r
tai
n
ty
r
e
g
ar
d
in
g
its
class
if
icatio
n
.
B
ased
o
n
t
h
ese
co
n
f
id
en
c
e
s
co
r
es,
th
e
C
R
M
s
y
s
tem
ca
n
m
ak
e
in
f
o
r
m
ed
d
ec
is
io
n
s
o
n
wh
eth
er
to
ac
t
o
n
th
e
p
r
ed
ictio
n
s
,
ad
h
e
r
in
g
t
o
th
e
r
ec
o
m
m
en
d
ed
th
r
esh
o
ld
f
o
r
class
if
icatio
n
ac
cu
r
ac
y
.
I
f
t
h
e
co
n
f
id
e
n
ce
s
co
r
e
f
alls
b
elo
w
th
is
p
r
ed
eter
m
in
ed
th
r
esh
o
ld
,
th
e
in
ter
ac
tio
n
will
b
e
r
o
u
te
d
to
t
h
e
cu
r
r
e
n
t
an
n
o
tato
r
s
f
o
r
f
u
r
t
h
er
r
ev
iew,
in
d
icatin
g
th
at
th
e
m
o
d
el
h
as
n
o
t
y
et
r
ea
ch
ed
th
e
n
ec
ess
ar
y
lev
el
o
f
ac
cu
r
ac
y
to
class
if
y
th
e
in
ter
a
ctio
n
r
eliab
ly
.
T
h
is
p
r
o
ce
s
s
n
o
t o
n
ly
en
h
a
n
ce
s
th
e
ef
f
icien
cy
o
f
c
u
s
to
m
er
s
er
v
ic
e
o
p
er
atio
n
s
b
u
t
also
en
s
u
r
es
th
at
co
m
p
lex
o
r
am
b
ig
u
o
u
s
ca
s
es
ar
e
h
an
d
led
ap
p
r
o
p
r
iately
b
y
h
u
m
an
ex
p
er
ts
,
m
ain
tain
in
g
a
h
ig
h
s
tan
d
ar
d
o
f
c
u
s
to
m
er
ca
r
e.
4.
CO
NCLU
SI
O
N
Fro
m
th
e
a
b
o
v
e
r
esear
ch
a
n
d
an
aly
s
is
,
th
e
f
o
llo
win
g
co
n
cl
u
s
io
n
s
ca
n
b
e
d
r
awn
.
Usi
n
g
p
r
e
-
tr
ain
ed
B
E
R
T
B
ah
a
s
a
(
in
d
o
o
r
-
b
ase
-
p
1
)
as
a
f
ea
tu
r
e
e
x
tr
ac
to
r
an
d
c
lass
if
ier
g
iv
es
th
e
b
est
m
o
d
el,
wh
ich
ac
h
iev
ed
a
n
F1
-
s
co
r
e
o
f
9
8
,
3
%.
T
h
e
b
est
m
o
d
el
ca
n
d
etec
t
ac
tu
al
c
o
m
p
lain
t
an
d
n
o
n
-
co
m
p
lain
t
i
n
ter
ac
tio
n
s
f
r
o
m
th
e
tr
ain
in
g
d
ataset
ef
f
ec
tiv
ely
.
T
h
e
r
esu
lt
o
f
th
e
Mc
Nem
ar
test
in
d
icate
s
th
at
SVM
b
e
h
av
ed
s
ig
n
if
ican
tly
d
if
f
er
en
tly
co
m
p
ar
ed
to
r
a
n
d
o
m
f
o
r
est
,
NN,
XGBo
o
s
t,
a
n
d
B
E
R
T
.
Similar
ly
,
r
an
d
o
m
f
o
r
est
,
NN,
an
d
XGBo
o
s
t
s
h
o
wed
an
im
m
en
s
e
d
if
f
er
en
ce
f
r
o
m
B
E
R
T
.
Me
an
wh
ile,
th
e
p
er
f
o
r
m
an
ce
s
f
o
r
r
an
d
o
m
f
o
r
est
an
d
NN,
r
an
d
o
m
f
o
r
est
an
d
XGBo
o
s
t,
an
d
NN
a
n
d
XGBo
o
s
t
d
id
n
o
t
h
av
e
a
s
tatis
tical
d
if
f
er
en
ce
in
p
e
r
f
o
r
m
an
ce
,
s
h
o
win
g
a
s
im
ilar
ity
in
th
eir
p
r
o
ce
s
s
.
Fro
m
f
u
r
th
er
a
n
aly
s
is
,
m
o
s
tly
th
e
er
r
o
r
co
m
es
f
r
o
m
FP
an
d
FN
clas
s
if
icatio
n
.
So
m
e
o
f
co
m
p
lain
d
ata
is
b
ein
g
class
if
ied
as
n
o
n
-
co
m
p
lain
t,
a
n
d
v
ice
v
er
s
a.
T
h
is
h
ap
p
en
s
d
u
e
to
t
h
e
lack
o
f
a
d
eq
u
ate
co
n
tex
t
to
k
n
o
w
th
e
in
ten
t
o
f
th
e
in
ter
ac
t
io
n
.
Mo
r
eo
v
er
,
th
er
e
ar
e
a
f
e
w
ca
s
es
o
f
u
n
clea
r
lab
elin
g
in
th
e
d
ataset
b
ec
au
s
e
co
m
m
o
n
wo
r
d
s
in
th
e
o
p
p
o
s
ite
lab
el
ex
is
t.
T
o
tack
le
th
is
in
th
e
im
p
lem
en
tatio
n
p
h
ase,
p
r
o
b
ab
ilit
y
v
alu
es
ca
n
b
e
u
s
ed
to
d
eter
m
in
e
w
h
eth
e
r
th
e
au
to
m
ated
r
esu
lts
s
h
o
u
l
d
b
e
tr
u
s
ted
o
r
r
e
q
u
ir
e
d
a
m
an
u
al
r
ev
iew.
T
h
is
en
s
u
r
es
th
at
th
e
class
if
icatio
n
is
d
o
n
e
p
r
o
p
er
ly
a
n
d
m
in
im
i
ze
s
th
e
p
o
ten
tial
f
o
r
er
r
o
r
s
.
I
t
is
es
s
en
tial
to
f
in
d
s
u
ch
a
th
r
esh
o
ld
f
o
r
th
e
b
ala
n
ce
o
f
er
r
o
r
s
an
d
th
e
b
e
n
ef
it
to
T
P
an
d
T
N.
Ho
wev
er
,
f
o
r
co
m
p
lain
ts
,
a
h
ig
h
er
th
r
esh
o
ld
o
f
a
r
o
u
n
d
0
.
9
6
s
h
o
u
ld
m
in
im
ize
FN,
wh
ile
f
o
r
n
o
n
-
co
m
p
lain
ts
,
a
lo
wer
th
r
esh
o
ld
o
f
ar
o
u
n
d
0
.
8
6
m
ak
es FP lo
wer
.
T
h
ese
th
r
esh
o
ld
s
co
m
p
r
o
m
is
e
d
if
f
er
e
n
t ty
p
es o
f
er
r
o
r
s
an
d
b
o
o
s
t o
v
e
r
all
p
er
f
o
r
m
an
ce
.
T
h
e
s
u
g
g
esti
o
n
f
o
r
f
u
r
th
e
r
r
es
ea
r
ch
in
clu
d
es
ad
d
r
ess
in
g
m
o
r
e
f
ea
tu
r
es
to
s
u
p
p
o
r
t
t
h
e
class
if
icatio
n
o
f
co
m
p
lain
ts
an
d
n
o
n
-
c
o
m
p
lain
t
s
,
as
u
s
er
s
o
n
s
o
cial
m
ed
ia
ca
n
d
is
cu
s
s
a
wid
e
v
ar
iety
o
f
to
p
ics.
E
x
p
a
n
d
in
g
th
e
class
if
icatio
n
to
in
clu
d
e
o
t
h
er
ty
p
es
o
f
in
te
r
ac
tio
n
s
,
s
u
ch
as
s
u
g
g
esti
o
n
s
f
o
r
im
p
r
o
v
em
e
n
ts
o
r
co
m
p
lim
en
ts
o
n
f
ea
tu
r
es,
wo
u
ld
p
r
o
v
id
e
a
m
o
r
e
co
m
p
r
eh
e
n
s
iv
e
u
n
d
er
s
tan
d
in
g
o
f
u
s
er
i
n
ter
ac
t
io
n
s
.
T
o
im
p
r
o
v
e
g
en
er
aliza
b
ilit
y
,
f
u
tu
r
e
r
esear
ch
s
h
o
u
ld
u
s
e
d
ata
f
r
o
m
m
u
ltip
le
co
m
p
an
ies
with
in
I
n
d
o
n
esia
'
s
e
-
co
m
m
er
ce
.
I
m
p
lem
en
tin
g
th
ese
s
u
g
g
esti
o
n
s
will
en
h
an
ce
th
e
m
o
d
el'
s
ef
f
ec
tiv
en
ess
an
d
im
p
r
o
v
e
class
if
icatio
n
ac
cu
r
ac
y
ac
r
o
s
s
v
ar
io
u
s
ty
p
es
o
f
s
o
cial
m
ed
ia
in
ter
ac
tio
n
s
,
lead
in
g
to
d
ee
p
er
in
s
ig
h
ts
.
T
h
e
class
if
icatio
n
ca
teg
o
r
ies
ca
n
b
e
d
ef
in
e
d
b
ased
o
n
th
e
p
r
im
a
r
y
p
r
o
b
lem
s
o
r
n
ee
d
s
id
en
tifie
d
in
th
e
ca
s
e
s
tu
d
y
.
ACK
NO
WL
E
DG
E
M
E
NT
S
T
h
e
au
th
o
r
d
ec
lar
es
th
at
th
er
e
ar
e
n
o
co
n
f
licts
o
f
in
ter
est.
T
h
is
r
esear
ch
d
id
n
o
t
r
ec
eiv
e
an
y
s
p
ec
if
ic
g
r
an
t
o
r
f
in
an
cial
s
u
p
p
o
r
t
f
r
o
m
f
u
n
d
in
g
a
g
en
cies
in
th
e
p
u
b
lic,
co
m
m
er
cial,
o
r
n
o
t
-
f
o
r
-
p
r
o
f
it
s
ec
to
r
s
.
All
d
ata
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
25
:
2
9
7
7
-
2
9
8
5
2984
an
d
m
ater
ials
u
s
ed
in
th
is
s
tu
d
y
wer
e
o
b
tain
ed
with
o
u
t
an
y
ex
ter
n
al
f
in
an
cial
in
f
lu
en
ce
o
r
b
ias,
en
s
u
r
i
n
g
th
e
in
teg
r
ity
an
d
im
p
ar
tiality
o
f
th
e
r
esear
ch
f
in
d
in
g
s
.
ACK
NO
WL
E
DG
M
E
N
T
S
T
h
e
au
th
o
r
d
ec
lar
es
th
at
th
e
r
e
ar
e
n
o
c
o
n
f
licts
o
f
in
ter
est.
T
h
is
r
esear
ch
d
i
d
n
o
t
r
ec
eiv
e
a
n
y
s
p
ec
if
ic
g
r
an
t
o
r
f
in
an
cial
s
u
p
p
o
r
t
f
r
o
m
f
u
n
d
in
g
a
g
en
cies
in
th
e
p
u
b
lic,
co
m
m
er
cial,
o
r
n
o
t
-
f
o
r
-
p
r
o
f
it
s
ec
to
r
s
.
All
d
ata
an
d
m
ater
ials
u
s
ed
in
th
is
s
tu
d
y
wer
e
o
b
tain
ed
with
o
u
t
an
y
ex
ter
n
al
f
in
an
cial
in
f
lu
en
ce
o
r
b
ias,
en
s
u
r
i
n
g
th
e
in
teg
r
ity
an
d
im
p
ar
tiality
o
f
th
e
r
esear
ch
f
in
d
in
g
s
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
T
h
e
au
th
o
r
s
tates th
at
n
o
f
u
n
d
i
n
g
was in
v
o
lv
e
d
f
o
r
th
is
r
esear
ch
.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
T
h
is
jo
u
r
n
al
u
s
es
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT)
to
r
ec
o
g
n
ize
in
d
iv
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
llab
o
r
atio
n
.
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
Ach
m
ad
R
izk
i
Ad
itam
a
Alf
an
Far
izk
i
W
icak
s
o
n
o
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
T
h
e
au
th
o
r
d
ec
lar
es th
at
th
e
r
e
ar
e
n
o
c
o
n
f
licts
o
f
in
ter
est.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
d
ata
th
at
s
u
p
p
o
r
t
th
e
f
in
d
in
g
s
o
f
th
is
s
tu
d
y
ar
e
av
aila
b
le
o
n
r
eq
u
est
f
r
o
m
th
e
co
r
r
esp
o
n
d
in
g
au
th
o
r
,
A
R
A.
T
h
e
d
ata,
wh
ich
co
n
tain
in
f
o
r
m
atio
n
th
at
co
u
ld
co
m
p
r
o
m
is
e
th
e
p
r
iv
ac
y
o
f
r
esear
c
h
p
ar
ticip
an
ts
,
ar
e
n
o
t p
u
b
licly
a
v
ailab
le
d
u
e
to
ce
r
tain
r
estrictio
n
s
.
RE
F
E
R
E
NC
E
S
[
1
]
S
.
D
.
N
e
g
a
r
a
a
n
d
E
.
S
.
S
o
e
si
l
o
w
a
t
i
,
“
E
-
c
o
mm
e
r
c
e
i
n
I
n
d
o
n
e
si
a
:
i
m
p
r
e
ssi
v
e
g
r
o
w
t
h
b
u
t
f
a
c
i
n
g
seri
o
u
s
c
h
a
l
l
e
n
g
e
s,
”
I
S
EA
S
Pe
rsp
e
c
t
i
v
e
,
n
o
.
1
0
2
,
p
p
.
1
–
1
4
,
2
0
2
1
.
[
2
]
A
.
F
a
s
h
o
l
a
a
n
d
F
.
K
u
su
ma,
"
E
-
C
o
m
merce
D
e
v
e
l
o
p
me
n
t
f
o
r
t
h
e
D
i
g
i
t
a
l
Ec
o
n
o
my
i
n
I
n
d
o
n
e
s
i
a
,
"
Ac
t
i
v
a
Y
u
r
i
s:
J
u
r
n
a
l
H
u
k
u
m
,
v
o
l
.
4
,
A
u
g
.
2
0
2
4
,
d
o
i
:
1
0
.
2
5
2
7
3
/
a
y
.
v
4
i
2
.
2
0
8
4
2
.
[
3
]
M
o
r
d
o
r
i
n
t
e
l
l
i
g
e
n
c
e
,
“
I
n
d
o
n
e
s
i
a
e
-
c
o
mm
e
r
c
e
m
a
r
k
e
t
s
i
z
e
&
sh
a
r
e
a
n
a
l
y
si
s
-
g
r
o
w
t
h
t
r
e
n
d
s
&
f
o
r
e
c
a
s
t
s
(
2
0
2
5
-
2
0
3
0
)
,
”
Mo
r
d
o
r
i
n
t
e
l
l
i
g
e
n
c
e
,
2
0
2
4
.
A
c
c
e
ss
e
d
:
F
e
b
.
14
,
2
0
2
5
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s:
/
/
w
w
w
.
mo
r
d
o
r
i
n
t
e
l
l
i
g
e
n
c
e
.
c
o
m
/
i
n
d
u
s
t
r
y
-
r
e
p
o
r
t
s/
i
n
d
o
n
e
si
a
-
e
c
o
mm
e
r
c
e
-
mar
k
e
t
[
4
]
T.
B
a
y
ı
r
a
n
d
S
.
B
o
z
y
i
ğ
i
t
,
"
C
o
n
s
u
m
e
r
c
o
m
p
l
a
i
n
t
s
m
a
n
a
g
e
me
n
t
i
n
t
h
e
d
i
g
i
t
a
l
e
r
a
,
"
i
n
Pro
c
e
e
d
i
n
g
s
o
f
t
h
e
2
0
2
3
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
C
o
n
su
m
e
r B
e
h
a
v
i
o
r
a
n
d
M
a
rk
e
t
i
n
g
,
D
e
c
.
2
0
2
3
,
p
p
.
5
7
-
9
0
,
d
o
i
:
1
0
.
4
0
1
8
/
9
7
9
-
8
-
3
6
9
3
-
0
4
2
8
-
0
.
c
h
0
0
3
.
[
5
]
D
.
S
.
M
.
R
.
K
u
r
n
i
a
w
a
n
,
"
T
h
e
e
f
f
e
c
t
o
f
c
o
m
p
l
a
i
n
t
h
a
n
d
l
i
n
g
o
n
c
u
st
o
mer
l
o
y
a
l
t
y
a
n
d
i
t
s
i
m
p
a
c
t
o
n
c
u
st
o
mer
s
a
t
i
sf
a
c
t
i
o
n
(
s
t
u
d
y
o
n
c
o
n
su
mers
o
f
P
T
T
e
l
k
o
m
W
i
t
e
l
o
f
W
e
st
K
a
l
i
m
a
n
t
a
n
)
,
"
Ea
st
A
f
ri
c
a
n
S
c
h
o
l
a
rs
J
o
u
r
n
a
l
o
f
E
c
o
n
o
m
i
c
s,
B
u
si
n
e
ss
a
n
d
Ma
n
a
g
e
m
e
n
t
,
v
o
l
.
6
,
n
o
.
1
,
p
p
.
1
1
-
1
6
,
J
a
n
.
2
0
2
3
,
d
o
i
:
1
0
.
3
6
3
4
9
/
e
a
s
j
e
b
m.
2
0
2
3
.
v
0
6
i
0
1
.
0
0
2
.
[
6
]
K
.
J
a
mi
l
,
L
.
D
u
n
n
a
n
,
R
.
F
.
G
u
l
,
M
.
U
.
S
h
e
h
z
a
d
,
S
.
H
.
M
.
G
i
l
l
a
n
i
,
a
n
d
F
.
H
.
A
w
a
n
,
“
R
o
l
e
o
f
so
c
i
a
l
me
d
i
a
m
a
r
k
e
t
i
n
g
a
c
t
i
v
i
t
i
e
s
i
n
i
n
f
l
u
e
n
c
i
n
g
c
u
s
t
o
mer
i
n
t
e
n
t
i
o
n
s:
A
p
e
r
sp
e
c
t
i
v
e
o
f
a
n
e
w
e
mer
g
i
n
g
e
r
a
,
”
Fr
o
n
t
i
e
rs
i
n
Ps
y
c
h
o
l
o
g
y
,
v
o
l
.
1
2
,
J
a
n
.
2
0
2
2
,
d
o
i
:
1
0
.
3
3
8
9
/
f
p
s
y
g
.
2
0
2
1
.
8
0
8
5
2
5
.
[
7
]
S
h
o
p
i
f
y
,
“
O
m
n
i
c
h
a
n
n
e
l
e
c
o
m
merce
:
w
h
a
t
i
t
i
s
a
n
d
h
o
w
t
o
i
mp
l
e
m
e
n
t
i
t
,
”
S
h
o
p
i
f
y
,
2
0
2
4
.
h
t
t
p
s
:
/
/
w
w
w
.
sh
o
p
i
f
y
.
c
o
m
/
e
n
t
e
r
p
r
i
s
e
/
b
l
o
g
/
o
m
n
i
c
h
a
n
n
e
l
-
e
c
o
mm
e
r
c
e
(
a
c
c
e
ss
e
d
M
a
r
.
2
4
,
2
0
2
4
)
.
[
8
]
A
.
R
.
M
a
n
j
r
e
k
a
r
,
S
.
S
.
G
o
k
h
a
l
e
,
a
n
d
R
.
A
.
W
i
l
so
n
,
“
D
e
t
e
c
t
i
o
n
o
f
a
n
t
i
-
h
u
m
a
n
r
i
g
h
t
s
d
i
sc
o
u
r
se
f
r
o
m
C
o
l
o
m
b
i
a
n
s
o
c
i
a
l
me
d
i
a
c
o
n
v
e
r
sa
t
i
o
n
s
u
si
n
g
a
d
v
a
n
c
e
d
t
r
a
n
sf
o
r
mer
m
o
d
e
l
s
,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
-
2
2
n
d
I
EEE
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
Ma
c
h
i
n
e
L
e
a
rn
i
n
g
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
,
I
C
MLA
2
0
2
3
,
D
e
c
.
2
0
2
3
,
p
p
.
1
5
0
1
–
1
5
0
7
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
M
LA
5
8
9
7
7
.
2
0
2
3
.
0
0
2
2
6
.
[
9
]
M
.
K
h
a
l
i
l
a
n
d
M
.
A
z
z
e
h
,
“
Tr
u
t
h
s
e
e
k
e
r
o
f
t
h
e
l
a
r
g
e
s
t
s
o
c
i
a
l
me
d
i
a
c
o
n
t
e
n
t
u
si
n
g
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
ms
,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
-
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
C
la
s
s
i
fica
tio
n
o
f c
u
s
to
mer
co
mp
la
in
ts
o
n
s
o
cia
l m
ed
ia
f
o
r
…
(
A
ch
ma
d
R
iz
ki
A
d
ita
ma
)
2985
2
2
n
d
I
EE
E
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
,
I
C
MLA
2
0
2
3
,
D
e
c
.
2
0
2
3
,
p
p
.
1
6
0
6
–
1
6
1
0
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
M
LA
5
8
9
7
7
.
2
0
2
3
.
0
0
2
4
3
.
[
1
0
]
J.
H
u
ss
e
y
,
K
.
S
t
o
n
e
,
a
n
d
T.
C
a
mp
,
“
P
o
si
t
i
v
e
a
n
d
u
n
l
a
b
e
l
e
d
l
e
a
r
n
i
n
g
f
o
r
mo
b
i
l
e
a
p
p
l
i
c
a
t
i
o
n
t
r
a
f
f
i
c
c
l
a
ssi
f
i
c
a
t
i
o
n
,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
-
I
EEE
Mi
l
i
t
a
r
y
C
o
m
m
u
n
i
c
a
t
i
o
n
s
C
o
n
f
e
r
e
n
c
e
MIL
C
O
M
,
N
o
v
.
2
0
2
2
,
v
o
l
.
2
0
2
2
-
N
o
v
e
m
,
p
p
.
2
5
–
3
0
,
d
o
i
:
1
0
.
1
1
0
9
/
M
I
LC
O
M
5
5
1
3
5
.
2
0
2
2
.
1
0
0
1
7
6
9
9
.
[
1
1
]
N
.
Th
a
i
r
A
l
i
,
K
.
F
a
l
i
h
H
a
ssa
n
,
M
.
N
a
j
i
m
A
b
d
u
l
l
a
h
,
a
n
d
Z.
S
a
l
a
m
A
l
-
H
c
h
i
m
y
,
“
T
h
e
a
p
p
l
i
c
a
t
i
o
n
o
f
r
a
n
d
o
m
f
o
r
e
st
t
o
t
h
e
c
l
a
ss
i
f
i
c
a
t
i
o
n
o
f
f
a
k
e
n
e
w
s,
”
BI
O
We
b
o
f
C
o
n
f
e
re
n
c
e
s
,
v
o
l
.
9
7
,
p
.
4
9
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
5
1
/
b
i
o
c
o
n
f
/
2
0
2
4
9
7
0
0
0
4
9
.
[
1
2
]
K
.
P
u
r
w
a
n
d
a
r
i
,
R
.
B
.
P
e
r
d
a
n
a
,
J
.
W
.
C
.
S
i
g
a
l
i
n
g
g
i
n
g
,
R
.
R
a
h
u
t
o
mo
,
a
n
d
B
.
P
a
r
d
a
m
e
a
n
,
“
A
u
t
o
m
a
t
i
c
sm
a
r
t
c
r
a
w
l
i
n
g
o
n
Tw
i
t
t
e
r
f
o
r
w
e
a
t
h
e
r
i
n
f
o
r
ma
t
i
o
n
i
n
I
n
d
o
n
e
s
i
a
,
”
Pr
o
c
e
d
i
a
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
2
2
7
,
p
p
.
7
9
5
–
8
0
4
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
r
o
c
s.2
0
2
3
.
1
0
.
5
8
5
.
[
1
3
]
C
.
M
.
B
i
sh
o
p
,
Pa
t
t
e
rn
re
c
o
g
n
i
t
i
o
n
a
n
d
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
.
S
p
r
i
n
g
e
r
N
e
w
Y
o
r
k
,
2
0
0
6
.
[
1
4
]
P.
-
N
.
Ta
n
,
M
.
S
t
e
i
n
b
a
c
h
,
a
n
d
V
.
K
u
m
a
r
,
I
n
t
r
o
d
u
c
t
i
o
n
t
o
d
a
t
a
m
i
n
i
n
g
,
2
n
d
e
d
.
P
e
a
r
s
o
n
,
2
0
2
0
.
[
1
5
]
S
.
R
a
s
c
h
k
a
,
“
M
o
d
e
l
e
v
a
l
u
a
t
i
o
n
,
m
o
d
e
l
se
l
e
c
t
i
o
n
,
a
n
d
a
l
g
o
r
i
t
h
m
se
l
e
c
t
i
o
n
i
n
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
,
”
a
r
Xi
v
:
1
8
1
1
.
1
2
8
0
8
,
N
o
v
.
2
0
1
8
,
d
o
i
:
1
0
.
4
8
5
5
0
/
a
r
X
i
v
.
1
8
1
1
.
1
2
8
0
8
.
[
1
6
]
P
.
C
h
a
p
ma
n
e
t
a
l
.
,
“
C
R
I
S
P
-
D
M
1
.
0
:
S
t
e
p
-
by
-
st
e
p
d
a
t
a
mi
n
i
n
g
g
u
i
d
e
,
”
S
P
S
S
i
n
c
,
v
o
l
.
7
8
,
p
p
.
1
–
7
8
,
2
0
0
0
.
[
1
7
]
S
.
W
a
n
k
h
e
d
e
,
R
.
P
a
t
i
l
,
S
.
S
o
n
a
w
a
n
e
,
a
n
d
P
.
A
.
S
a
v
e
,
“
D
a
t
a
p
r
e
p
r
o
c
e
ssi
n
g
f
o
r
e
f
f
i
c
i
e
n
t
se
n
t
i
me
n
t
a
l
a
n
a
l
y
s
i
s,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
I
n
v
e
n
t
i
v
e
C
o
m
m
u
n
i
c
a
t
i
o
n
a
n
d
C
o
m
p
u
t
a
t
i
o
n
a
l
T
e
c
h
n
o
l
o
g
i
e
s,
I
C
I
C
C
T
2
0
1
8
,
2
0
1
8
,
p
p
.
7
2
3
–
7
2
6
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
I
C
C
T.
2
0
1
8
.
8
4
7
3
2
7
7
.
[
1
8
]
T.
B
.
B
r
o
w
n
e
t
a
l
.
,
“
La
n
g
u
a
g
e
m
o
d
e
l
s
a
r
e
f
e
w
-
s
h
o
t
l
e
a
r
n
e
r
s
,
”
a
rX
i
v
:
2
0
0
5
.
1
4
1
6
5
,
M
a
y
2
0
2
0
.
[
1
9
]
M
d
,
“
B
l
e
n
d
i
n
g
w
e
i
g
h
t
e
d
TF
-
I
D
F
B
ER
T
f
o
r
i
m
p
r
o
v
i
n
g
s
e
ma
n
t
i
c
sea
r
c
h
,
”
i
n
I
C
ARC
2
0
2
2
-
2
n
d
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
Ad
v
a
n
c
e
d
Re
s
e
a
r
c
h
i
n
C
o
m
p
u
t
i
n
g
:
T
o
w
a
r
d
s
a
D
i
g
i
t
a
l
l
y
Em
p
o
w
e
re
d
S
o
c
i
e
t
y
,
F
e
b
.
2
0
2
2
,
p
p
.
1
5
4
–
1
5
9
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
A
R
C
5
4
4
8
9
.
2
0
2
2
.
9
7
5
3
8
7
5
.
[
2
0
]
J.
D
e
v
l
i
n
,
M
.
-
W
.
C
h
a
n
g
,
K
.
L
e
e
,
a
n
d
K
.
T
o
u
t
a
n
o
v
a
,
“
B
ER
T
:
p
r
e
-
t
r
a
i
n
i
n
g
o
f
d
e
e
p
b
i
d
i
r
e
c
t
i
o
n
a
l
t
r
a
n
sf
o
r
m
e
r
s
f
o
r
l
a
n
g
u
a
g
e
u
n
d
e
r
s
t
a
n
d
i
n
g
,
”
a
rX
i
v
:
1
8
1
0
.
0
4
8
0
5
,
O
c
t
.
2
0
1
8
.
[
2
1
]
H
.
F
a
n
a
n
d
Y
.
Q
i
n
,
“
R
e
s
e
a
r
c
h
o
n
t
e
x
t
c
l
a
ss
i
f
i
c
a
t
i
o
n
b
a
se
d
o
n
i
mp
r
o
v
e
d
TF
-
I
D
F
a
l
g
o
r
i
t
h
m,”
i
n
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
2
0
1
8
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
N
e
t
w
o
rk,
C
o
m
m
u
n
i
c
a
t
i
o
n
,
C
o
m
p
u
t
e
r
E
n
g
i
n
e
e
ri
n
g
(
N
C
C
E
2
0
1
8
)
,
2
0
1
8
,
p
p
.
5
0
1
–
5
0
6
,
d
o
i
:
1
0
.
2
9
9
1
/
n
c
c
e
-
1
8
.
2
0
1
8
.
7
9
.
[
2
2
]
A
.
S
i
n
h
a
,
M
.
N
.
B
.
J.
N
a
sk
a
r
,
M
.
P
a
n
d
e
y
,
a
n
d
S
.
S
.
R
a
u
t
a
r
a
y
,
“
Te
x
t
c
l
a
ssi
f
i
c
a
t
i
o
n
u
si
n
g
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s
:
c
o
mp
a
r
a
t
i
v
e
a
n
a
l
y
si
s
,
”
i
n
2
0
2
2
O
I
T
S
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
I
n
f
o
rm
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
(
O
C
I
T
)
,
D
e
c
.
2
0
2
2
,
p
p
.
1
0
2
–
1
0
7
,
d
o
i
:
1
0
.
1
1
0
9
/
O
C
I
T5
6
7
6
3
.
2
0
2
2
.
0
0
0
2
9
.
[
2
3
]
H
.
A
.
S
a
l
m
a
n
,
A
.
K
a
l
a
k
e
c
h
,
a
n
d
A
.
S
t
e
i
t
i
,
“
R
a
n
d
o
m
f
o
r
e
st
a
l
g
o
r
i
t
h
m
o
v
e
r
v
i
e
w
,
”
B
a
b
y
l
o
n
i
a
n
J
o
u
r
n
a
l
o
f
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
,
v
o
l
.
2
0
2
4
,
p
p
.
6
9
–
7
9
,
Ju
n
.
2
0
2
4
,
d
o
i
:
1
0
.
5
8
4
9
6
/
B
J
M
L
/
2
0
2
4
/
0
0
7
.
[
2
4
]
T.
C
h
e
n
a
n
d
C
.
G
u
e
s
t
r
i
n
,
“
X
g
b
o
o
s
t
:
a
sca
l
a
b
l
e
t
r
e
e
b
o
o
st
i
n
g
s
y
st
e
m,
”
i
n
Pro
c
e
e
d
i
n
g
s
o
f
t
h
e
2
2
n
d
A
C
M
S
I
G
K
D
D
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
K
n
o
w
l
e
d
g
e
D
i
s
c
o
v
e
r
y
a
n
d
D
a
t
a
Mi
n
i
n
g
,
A
u
g
.
2
0
1
6
,
p
p
.
7
8
5
–
7
9
4
,
d
o
i
:
1
0
.
1
1
4
5
/
2
9
3
9
6
7
2
.
2
9
3
9
7
8
5
.
[
2
5
]
A
.
V
a
sw
a
n
i
e
t
a
l
.
,
“
A
t
t
e
n
t
i
o
n
i
s
a
l
l
y
o
u
n
e
e
d
,
”
A
d
v
a
n
c
e
s
i
n
n
e
u
r
a
l
i
n
f
o
rm
a
t
i
o
n
p
r
o
c
e
ssi
n
g
s
y
st
e
m
s
,
v
o
l
.
3
0
,
2
0
1
7
.
[
2
6
]
F
.
K
o
t
o
,
A
.
R
a
h
i
mi
,
J.
H
.
La
u
,
a
n
d
T.
B
a
l
d
w
i
n
,
“
I
n
d
o
LE
M
a
n
d
I
n
d
o
B
ER
T:
A
b
e
n
c
h
mark
d
a
t
a
se
t
a
n
d
p
r
e
-
t
r
a
i
n
e
d
l
a
n
g
u
a
g
e
m
o
d
e
l
f
o
r
I
n
d
o
n
e
s
i
a
n
N
LP
,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
2
8
t
h
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
C
o
m
p
u
t
a
t
i
o
n
a
l
L
i
n
g
u
i
st
i
c
s
,
2
0
2
0
,
p
p
.
7
5
7
–
7
7
0
,
d
o
i
:
1
0
.
1
8
6
5
3
/
v
1
/
2
0
2
0
.
c
o
l
i
n
g
-
m
a
i
n
.
6
6
.
[
2
7
]
D
.
B
e
r
r
a
r
,
“
C
r
o
ss
-
v
a
l
i
d
a
t
i
o
n
,
”
i
n
E
n
c
y
c
l
o
p
e
d
i
a
o
f
Bi
o
i
n
f
o
rm
a
t
i
c
s
a
n
d
C
o
m
p
u
t
a
t
i
o
n
a
l
Bi
o
l
o
g
y
,
E
l
se
v
i
e
r
,
2
0
1
9
,
p
p
.
5
4
2
–
5
4
5
.
[
2
8
]
G
.
D
i
L
e
o
a
n
d
F
.
S
a
r
d
a
n
e
l
l
i
,
“
S
t
a
t
i
st
i
c
a
l
s
i
g
n
i
f
i
c
a
n
c
e
:
p
v
a
l
u
e
,
0
.
0
5
t
h
r
e
sh
o
l
d
,
a
n
d
a
p
p
l
i
c
a
t
i
o
n
s
t
o
r
a
d
i
o
m
i
c
s
—
r
e
a
s
o
n
s
f
o
r
a
c
o
n
ser
v
a
t
i
v
e
a
p
p
r
o
a
c
h
,
”
E
u
r
o
p
e
a
n
R
a
d
i
o
l
o
g
y
Ex
p
e
r
i
m
e
n
t
a
l
,
v
o
l
.
4
,
n
o
.
1
,
M
a
r
.
2
0
2
0
,
d
o
i
:
1
0
.
1
1
8
6
/
s4
1
7
4
7
-
020
-
0
1
4
5
-
y.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Ac
h
m
a
d
Riz
k
i
A
d
ita
m
a
is
a
m
a
ste
r
o
f
i
n
fo
rm
a
ti
o
n
tec
h
n
o
l
o
g
y
st
u
d
e
n
t
i
n
t
h
e
F
a
c
u
lt
y
o
f
C
o
m
p
u
ter
S
c
ien
c
e
a
t
Un
iv
e
rsitas
In
d
o
n
e
sia
.
He
o
b
tain
e
d
th
e
ir
b
a
c
h
e
lo
r'
s
d
e
g
re
e
i
n
c
o
m
p
u
ter
sc
ien
c
e
fr
o
m
th
e
F
a
c
u
lt
y
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
t
Un
i
v
e
r
sitas
Bra
wijay
a
.
In
a
d
d
i
ti
o
n
to
b
e
i
n
g
a
n
a
c
ti
v
e
st
u
d
e
n
t,
h
e
is
wo
rk
i
n
g
p
ro
fe
ss
io
n
a
ll
y
in
so
ftwa
re
e
n
g
in
e
e
rin
g
a
t
o
n
e
o
f
t
h
e
lea
d
in
g
tec
h
c
o
m
p
a
n
ies
in
I
n
d
o
n
e
sia
.
His
i
n
tere
st
in
m
o
d
e
rn
b
a
c
k
e
n
d
d
e
v
e
l
o
p
m
e
n
t,
m
icro
se
rv
ice
s
a
rc
h
it
e
c
tu
re
,
sc
a
lab
le
sy
ste
m
d
e
sig
n
a
n
d
e
m
e
rg
in
g
te
c
h
n
o
l
o
g
ies
su
c
h
a
s
m
a
c
h
in
e
lea
rn
in
g
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
a
c
h
m
a
d
.
rizk
i0
2
@u
i
.
a
c
.
id
.
Alfa
n
Fa
r
izki
Wica
k
so
n
o
se
rv
e
s
a
s
a
d
irec
to
r
a
t
Un
i
v
e
rsitas
In
d
o
n
e
sia
Ce
n
ter
f
o
r
Leg
a
l
In
fo
rm
a
ti
c
s,
w
h
e
re
h
is
e
x
p
e
rti
se
sig
n
ifi
c
a
n
tl
y
c
o
n
tr
ib
u
tes
t
o
t
h
e
c
e
n
ter’s
re
se
a
rc
h
a
n
d
d
e
v
e
lo
p
m
e
n
t
i
n
it
iati
v
e
s.
Wi
th
a
P
h
.
D.
fro
m
th
e
Un
i
v
e
rsity
o
f
M
e
lb
o
u
r
n
e
,
a
n
M
.
S
c
.
fro
m
th
e
Ko
re
a
Ad
v
a
n
c
e
d
I
n
stit
u
te
o
f
S
c
i
e
n
c
e
a
n
d
Tec
h
n
o
l
o
g
y
,
a
n
d
a
Ba
c
h
e
lo
r
o
f
E
n
g
i
n
e
e
rin
g
(S
.
T
.
)
fro
m
In
st
it
u
t
Tek
n
o
lo
g
i
Ba
n
d
u
n
g
.
His
re
se
a
rc
h
i
n
tere
sts
li
e
a
t
t
h
e
in
ters
e
c
ti
o
n
o
f
in
f
o
rm
a
ti
o
n
re
tri
e
v
a
l
a
n
d
re
p
re
se
n
tatio
n
lea
rn
i
n
g
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
a
lfan
@c
s.u
i.
a
c
.
id
.
Evaluation Warning : The document was created with Spire.PDF for Python.