I
nd
o
ne
s
ia
n J
o
urna
l o
f
E
lect
rica
l En
g
ineering
a
nd
Co
m
pu
t
er
Science
Vo
l.
38
,
No
.
2
,
Ma
y
20
25
,
p
p
.
1
170
~1
1
8
0
I
SS
N:
2
502
-
4
7
52
,
DOI
: 1
0
.
1
1
5
9
1
/ijee
cs
.v
38.
i
2
.
p
p
1
1
7
0
-
1
1
8
0
1170
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ee
cs
.
ia
esco
r
e.
co
m
Wo
rd embed
ding
f
o
r cont
ex
tual si
mila
rity using
cos
ine
sim
ila
rity
Yess
y
Asri
1
,
Dw
ina
K
us
wa
r
da
ni
2
,
Am
a
nd
a
At
ik
a
Sa
ri
2
,
At
ik
a
h Rif
da
h Ans
y
a
ri
1
1
F
a
c
u
l
t
y
o
f
En
e
r
g
y
a
n
d
T
e
l
e
ma
t
i
c
s,
S
t
u
d
y
P
r
o
g
r
a
m I
n
f
o
r
ma
t
i
o
n
S
y
st
e
ms,
I
n
st
i
t
u
t
e
Te
c
h
n
o
l
o
g
y
o
f
P
LN
,
Ja
k
a
r
t
a
,
I
n
d
o
n
e
s
i
a
2
F
a
c
u
l
t
y
o
f
En
e
r
g
y
a
n
d
T
e
l
e
ma
t
i
c
s,
S
t
u
d
y
P
r
o
g
r
a
m I
n
f
o
r
ma
t
i
c
s E
n
g
i
n
e
e
r
i
n
g
,
I
n
st
i
t
u
t
e
Te
c
h
n
o
l
o
g
y
o
f
P
LN
,
Ja
k
a
r
t
a
,
I
n
d
o
n
e
si
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Mar
13
,
2
0
2
4
R
ev
is
ed
No
v
5
,
2
0
2
4
Acc
ep
ted
No
v
11
,
2
0
2
4
P
e
rsp
e
c
ti
v
e
s
o
n
tec
h
n
o
lo
g
y
o
ften
h
a
v
e
sim
il
a
rit
ies
in
c
e
rtain
c
o
n
tex
t
s,
su
c
h
a
s
in
fo
rm
a
ti
o
n
sy
ste
m
s
a
n
d
in
f
o
r
m
a
ti
c
s
e
n
g
in
e
e
rin
g
.
T
h
e
so
u
rc
e
o
f
o
p
in
io
n
d
a
ta
c
o
m
e
s
fro
m
th
e
Q
u
o
ra
a
p
p
l
ica
ti
o
n
,
with
a
re
tri
e
v
a
l
li
m
it
o
f
th
e
las
t
5
y
e
a
rs.
Th
is
re
se
a
rc
h
a
ims
t
o
imp
lem
e
n
t
In
d
o
-
b
id
i
re
c
ti
o
n
a
l
e
n
c
o
d
e
r
re
p
re
se
n
tatio
n
s
fr
o
m
tran
sf
o
rm
e
rs
(
BERT
)
,
a
v
a
rian
t
o
f
t
h
e
BE
RT
m
o
d
e
l
o
p
ti
m
ize
d
f
o
r
In
d
o
n
e
s
ian
lan
g
u
a
g
e
,
in
t
h
e
c
o
n
tex
t
o
f
in
f
o
rm
a
ti
o
n
sy
ste
m
(IS
)
a
n
d
i
n
fo
rm
a
ti
o
n
tec
h
n
o
l
o
g
y
(IT
)
to
p
ic
c
las
sifica
ti
o
n
with
4
1
4
o
ri
g
in
a
l
d
a
ta,
wh
ich
,
a
fter
b
e
in
g
a
u
g
m
e
n
te
d
u
si
n
g
t
h
e
sy
n
o
n
y
m
re
p
lac
e
m
e
n
t
m
e
t
h
o
d
,
Th
e
g
e
n
e
ra
ted
d
a
ta
b
e
c
o
m
e
s
8
2
8
.
Da
ta
a
u
g
m
e
n
tatio
n
a
ims
t
o
e
v
a
lu
a
te
th
e
p
e
rfo
rm
a
n
c
e
o
f
m
o
d
e
ls
b
y
u
si
n
g
s
y
n
o
n
y
m
s
a
n
d
re
a
rra
n
g
in
g
tex
t
wh
i
le
m
a
in
tain
in
g
m
e
a
n
in
g
a
n
d
stru
c
t
u
r
e
.
Th
e
a
p
p
r
o
a
c
h
u
se
d
is t
o
lab
e
l
t
h
e
o
p
in
i
o
n
tex
t
b
a
se
d
o
n
t
h
e
c
o
si
n
e
sim
il
a
rit
y
c
a
lcu
latio
n
o
f
th
e
e
m
b
e
d
d
in
g
t
o
k
e
n
fro
m
th
e
In
d
o
B
ERT
m
o
d
e
l.
T
h
e
n
,
t
h
e
In
d
o
BERT
m
o
d
e
l
is
a
p
p
li
e
d
to
c
las
sify
th
e
re
v
iew
s.
Th
e
e
x
p
e
rime
n
tal
re
su
lt
s
s
h
o
w
th
a
t
th
e
a
p
p
r
o
a
c
h
o
f
u
sin
g
In
d
o
BERT
t
o
c
las
sify
S
I
a
n
d
IT
to
p
ics
b
a
se
d
o
n
c
o
n
tex
tu
a
l
sim
il
a
rit
y
a
c
h
iev
e
s
9
0
%
a
c
c
u
ra
c
y
b
a
se
d
o
n
th
e
c
o
n
fu
si
o
n
m
a
tri
x
.
Th
e
se
p
o
sit
iv
e
re
su
lt
s
sh
o
w
t
h
e
g
re
a
t
p
o
ten
ti
a
l
o
f
u
si
n
g
tran
sfo
rm
e
r
-
b
a
se
d
lan
g
u
a
g
e
m
o
d
e
ls,
su
c
h
a
s
In
d
o
B
ERT
,
to
su
p
p
o
r
t
t
h
e
a
n
a
ly
sis
o
f
c
o
m
m
e
n
ts
a
n
d
re
late
d
t
o
p
ics
i
n
In
d
o
n
e
sia
n
.
K
ey
w
o
r
d
s
:
Au
g
m
en
ted
d
ata
C
o
n
tex
tu
al
s
im
ilar
ity
C
o
s
in
e
s
im
ilar
ity
I
n
d
o
B
E
R
T
W
o
r
d
em
b
ed
d
in
g
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
:
Yess
y
Asri
Facu
lty
o
f
E
n
e
r
g
y
a
n
d
T
elem
a
tics
,
Stu
d
y
Pro
g
r
am
I
n
f
o
r
m
ati
o
n
Sy
s
tem
s
,
I
n
s
titu
te
T
ec
h
n
o
l
o
g
y
PLN
J
ak
ar
ta
1
1
7
5
0
,
I
n
d
o
n
esia
E
m
ail: y
ess
y
asri@
i
tp
ln
.
ac
.
id
1.
I
NT
RO
D
UCT
I
O
N
A
to
p
ic
in
th
e
f
o
r
m
o
f
an
o
p
in
io
n
b
ec
o
m
es
a
r
ef
er
en
ce
in
d
eter
m
in
in
g
ce
r
tain
in
ter
ests
,
g
o
als,
o
r
o
th
er
th
in
g
s
.
Ho
wev
er
,
it
o
f
te
n
s
ee
m
s
co
m
p
licated
an
d
co
n
f
u
s
in
g
wh
en
to
p
ics
h
av
e
d
if
f
er
en
t
ca
teg
o
r
ies
in
th
e
s
am
e
f
ield
.
E
s
p
ec
ially
in
th
e
tech
n
o
lo
g
y
ca
teg
o
r
y
,
wh
ich
h
a
s
s
im
ilar
itie
s
b
etwe
en
s
ev
er
al
p
ar
ts
,
s
u
ch
as
th
e
to
p
ic
o
f
in
f
o
r
m
atics
en
g
in
ee
r
i
n
g
a
n
d
t
h
e
to
p
ic
o
f
in
f
o
r
m
ati
o
n
s
y
s
tem
s
,
b
o
th
t
o
p
ics
h
a
v
e
m
an
y
en
th
u
s
iast
s
,
b
u
t f
r
o
m
e
n
th
u
s
iast
s
s
u
ch
as p
r
o
s
p
ec
tiv
e
s
tu
d
en
ts
wh
o
h
av
e
an
in
ter
est,
s
o
m
e
o
f
th
em
e
x
p
e
r
ien
ce
co
n
f
u
s
io
n
in
u
n
d
er
s
tan
d
i
n
g
th
e
d
if
f
e
r
en
ce
s
an
d
ap
p
r
o
ac
h
es
o
r
s
im
ilar
ities
b
etwe
en
th
e
to
p
ics
o
f
in
f
o
r
m
atics
en
g
in
ee
r
in
g
an
d
in
f
o
r
m
atio
n
s
y
s
tem
s
.
So
,
th
ey
ten
d
to
h
a
v
e
v
ar
i
o
u
s
q
u
e
s
tio
n
s
co
v
er
in
g
asp
ec
ts
o
f
b
o
t
h
to
p
ics
b
y
wan
tin
g
an
o
p
in
i
o
n
o
r
o
p
i
n
io
n
o
n
a
p
latf
o
r
m
as a
s
o
u
r
ce
o
f
i
n
f
o
r
m
atio
n
.
I
n
tex
t
d
ata,
it
d
o
es
n
o
t
o
n
ly
h
av
e
o
n
e
o
r
two
s
en
ten
ce
s
b
u
t
ca
n
f
o
r
m
a
lo
n
g
p
ar
a
g
r
a
p
h
.
A
f
ield
b
ec
o
m
es
th
e
to
p
ic
o
f
a
s
er
ies
o
f
s
en
ten
ce
s
.
Ho
wev
er
,
d
esp
it
e
h
av
in
g
th
e
s
am
e
f
ield
,
t
h
er
e
will
b
e
d
if
f
er
e
n
ce
s
in
ca
te
g
o
r
ies
[
1
]
.
T
h
is
ca
n
b
e
an
aly
ze
d
ac
c
o
r
d
i
n
g
to
th
e
co
n
tex
t.
A
n
u
m
b
e
r
o
f
d
o
c
u
m
en
ts
ar
e
g
iv
en
ter
m
f
r
eq
u
e
n
cy
-
in
v
er
s
e
d
o
cu
m
en
t
f
r
eq
u
e
n
cy
(
TF
-
I
DF
)
an
d
co
s
in
e
s
im
ilar
ity
m
o
d
el
ca
lcu
la
tio
n
s
to
s
ee
th
eir
s
im
ilar
ity
[
2
]
.
I
n
th
is
s
co
p
e,
c
o
n
tex
t
is
th
e
m
ain
co
n
ce
r
n
,
s
o
a
m
o
d
el
is
n
ee
d
ed
th
at
ca
n
s
e
e
th
e
en
tire
c
o
n
tex
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
Wo
r
d
emb
ed
d
in
g
f
o
r
co
n
textu
a
l simil
a
r
ity
u
s
in
g
co
s
in
e
s
imil
a
r
ity
(
Yes
s
y
A
s
r
i
)
1171
o
f
s
en
ten
ce
co
n
s
tr
u
ctio
n
.
B
id
ir
ec
tio
n
al
en
co
d
e
r
r
ep
r
esen
t
atio
n
s
f
r
o
m
t
r
an
s
f
o
r
m
e
r
s
(
B
E
R
T
)
is
o
n
e
o
f
th
e
tech
n
iq
u
es
th
at
ca
n
b
e
u
s
ed
i
n
th
is
p
r
o
ce
s
s
.
B
E
R
T
m
o
d
els
h
av
e
p
r
o
v
en
ef
f
ec
tiv
e
in
p
er
f
o
r
m
in
g
lan
g
u
ag
e
p
r
o
ce
s
s
in
g
task
s
s
u
ch
as
s
em
an
tic
tex
t
[
3
]
an
d
s
y
n
tax
[
4
]
.
I
n
ter
m
s
o
f
lan
g
u
a
g
e,
I
n
d
o
B
E
R
T
is
a
B
E
R
T
d
ev
elo
p
m
e
n
t
th
at
ca
n
b
e
u
s
ed
f
o
r
I
n
d
o
n
esian
d
ata.
I
n
ad
d
it
io
n
to
I
n
d
o
B
E
R
T
in
d
icato
r
s
,
d
ata
au
g
m
en
tatio
n
tech
n
iq
u
es
a
r
e
also
p
er
f
o
r
m
e
d
with
t
h
e
aim
o
f
e
x
p
an
d
in
g
th
e
d
ata
t
o
h
av
e
ce
r
tain
v
ar
iatio
n
s
.
Me
r
g
in
g
is
p
er
f
o
r
m
ed
,
f
o
llo
we
d
b
y
s
eq
u
e
n
ce
r
an
d
o
m
izatio
n
,
b
ef
o
r
e
d
iv
id
in
g
th
e
tex
t
in
to
two
co
lu
m
n
s
f
o
r
tex
t
s
im
ilar
ity
ca
lcu
latio
n
u
s
in
g
co
s
in
e
s
im
ilar
ity
b
y
u
tili
zin
g
wo
r
d
em
b
e
d
d
in
g
f
r
o
m
I
n
d
o
B
E
R
T
.
A
s
i
m
p
le
tex
t
s
im
ilar
ity
lab
el
is
0
m
ea
n
in
g
n
o
t similar
an
d
1
m
ea
n
in
g
s
im
ilar
[
5
]
.
I
n
th
e
er
a
o
f
d
ig
italizatio
n
,
th
e
r
e
ar
e
a
n
u
m
b
e
r
o
f
p
latf
o
r
m
s
a
s
in
f
o
r
m
atio
n
-
s
o
u
r
cin
g
to
o
ls
.
Qu
o
r
a
is
a
q
u
esti
o
n
-
an
d
-
an
s
wer
ap
p
f
o
u
n
d
ed
b
y
Ad
am
D
’
An
g
elo
a
n
d
a
co
llea
g
u
e.
T
h
is
ap
p
lic
atio
n
was
in
itially
o
n
ly
av
ailab
le
in
th
e
E
n
g
lis
h
v
er
s
io
n
;
u
n
til
Ap
r
il
2
0
1
8
,
Qu
o
r
a
was
p
r
esen
t
in
th
e
I
n
d
o
n
esian
v
er
s
io
n
(
h
ttp
s
://www.
id
n
tim
es.c
o
m
/tech
/tre
n
d
/m
a
h
d
a
-
len
a/
k
eu
n
g
g
u
lan
-
ap
lik
asi
-
q
u
o
r
a
-
c1
c
2
)
.
R
ep
o
r
tin
g
f
r
o
m
s
tatis
t
ics
b
y
th
e
d
ir
ec
to
r
o
f
p
r
o
d
u
ct
m
an
ag
em
e
n
t,
Qu
o
r
a
is
a
web
s
ite
an
d
ap
p
licatio
n
-
b
ased
p
latf
o
r
m
with
a
n
ac
h
iev
em
en
t
o
f
3
0
0
m
illi
o
n
v
is
ito
r
s
in
cr
ea
tin
g
a
s
e
t
o
f
q
u
esti
o
n
s
f
r
o
m
v
ar
io
u
s
to
p
ics
p
o
s
ted
b
y
u
s
er
s
an
d
an
s
wer
s
f
r
o
m
o
th
e
r
u
s
er
s
wh
o
h
av
e
g
r
ea
ter
in
s
ig
h
t
in
to
a
to
p
ic
[
6
]
,
[
7
]
.
Ho
wev
e
r
,
p
r
o
s
p
e
ctiv
e
s
tu
d
en
ts
o
f
ten
h
av
e
d
if
f
icu
lty
f
ilter
in
g
an
d
an
aly
zin
g
Qu
o
r
a
u
s
er
o
p
i
n
io
n
s
th
at
ar
e
r
elev
a
n
t
to
th
eir
n
ee
d
s
.
I
n
a
d
d
itio
n
,
s
o
m
etim
es
wo
r
k
er
s
in
t
h
e
m
a
r
k
etin
g
f
ield
h
a
v
e
a
litt
le
h
ass
le
wh
en
id
e
n
tify
in
g
an
d
f
ilter
i
n
g
th
e
in
f
o
r
m
atio
n
o
b
tain
ed
,
s
u
ch
as
in
f
o
r
m
atio
n
ab
o
u
t
t
h
e
m
o
s
t
r
elev
an
t
q
u
esti
o
n
to
p
ics
with
in
f
o
r
m
ativ
e
o
p
i
n
io
n
s
r
elate
d
to
t
h
e
d
if
f
er
en
ce
b
etwe
en
in
f
o
r
m
atics
en
g
in
ee
r
in
g
an
d
in
f
o
r
m
atio
n
s
y
s
tem
s
,
s
o
th
at
la
ter
t
h
e
d
ata
ca
n
b
e
u
s
ed
as
a
r
ef
er
en
ce
in
p
r
o
m
o
tio
n
al
m
ater
ials
an
d
o
th
er
m
ar
k
etin
g
m
att
er
s
.
T
h
er
ef
o
r
e,
a
n
ef
f
ec
tiv
e
m
eth
o
d
is
n
ee
d
ed
to
an
aly
ze
th
e
o
p
in
io
n
s
o
f
Qu
o
r
a
u
s
er
s
to
h
elp
in
r
ea
d
in
g
t
o
p
i
cs
in
th
e
f
ield
o
f
tec
h
n
o
lo
g
y
,
esp
ec
ially
b
etwe
en
th
e
two
to
p
ics
,
n
am
ely
in
f
o
r
m
atics
en
g
in
ee
r
in
g
an
d
in
f
o
r
m
at
io
n
s
y
s
tem
s
.
On
e
tech
n
iq
u
e
th
at
ca
n
b
e
u
s
ed
is
a
T
r
an
s
f
o
r
m
e
r
s
-
b
ased
n
atu
r
al
l
an
g
u
ag
e
p
r
o
ce
s
s
in
g
(
NL
P)
te
ch
n
iq
u
e,
s
u
c
h
as
B
E
R
T
[
8
]
.
Su
ch
m
o
d
els
h
av
e
p
r
o
v
e
n
to
b
e
v
er
y
ef
f
ec
tiv
e
in
p
er
f
o
r
m
in
g
lan
g
u
ag
e
p
r
o
c
ess
in
g
task
s
s
u
ch
as
ca
p
tu
r
in
g
s
y
n
tax
[
4
]
an
d
s
en
ten
ce
-
s
em
an
tic
tex
t
[
3
]
.
On
a
lan
g
u
ag
e
b
asis
,
B
E
R
T
h
as
b
ee
n
d
ev
elo
p
ed
u
s
in
g
th
e
I
n
d
o
B
E
R
T
m
o
d
el,
wh
ich
is
u
s
ed
f
o
r
I
n
d
o
n
esian
l
an
g
u
ag
e
d
ata.
C
las
s
if
icatio
n
b
y
im
p
le
m
en
tin
g
th
e
I
n
d
o
B
E
R
T
m
o
d
el
in
em
b
ed
d
in
g
I
n
d
o
b
en
c
h
m
ar
k
o
r
I
n
d
o
B
E
R
T
h
as
an
ac
c
u
r
ac
y
lev
el
o
f
8
7
%
with
o
n
lin
e
ar
ticle
co
n
ten
t
d
ata
in
r
esear
ch
in
t
h
e
b
u
ild
in
g
o
f
i
n
f
o
r
m
atics,
tech
n
o
lo
g
y
,
an
d
s
cien
ce
(
B
I
T
S)
jo
u
r
n
al,
“
C
lick
b
ait
C
las
s
if
icatio
n
Mo
d
el
o
n
On
lin
e
Ne
ws
with
Sem
an
tic
Similar
ity
C
alcu
latio
n
B
etwe
en
New
s
T
itle
an
d
C
o
n
ten
t
”
[
1
]
.
I
n
t
h
is
s
tu
d
y
,
th
e
au
th
o
r
s
p
er
f
o
r
m
ed
tex
t
s
im
i
lar
ity
an
aly
s
i
tech
n
iq
u
es
u
s
in
g
th
e
I
n
d
o
B
E
R
T
m
o
d
el
to
class
if
y
SI
an
d
I
T
to
p
ics
in
c
o
n
tex
tu
al
s
im
ilar
ity
-
b
ased
o
p
in
io
n
s
.
T
h
e
d
if
f
er
e
n
ce
in
th
is
s
tu
d
y
lies
in
th
e
la
b
elin
g
u
s
ed
;
in
th
e
jo
u
r
n
al,
n
o
lab
elin
g
is
d
o
n
e,
wh
ile
in
th
is
s
tu
d
y
,
lab
elin
g
is
d
o
n
e
u
s
i
n
g
co
s
in
e
ca
lcu
latio
n
s
b
y
u
tili
zin
g
em
b
e
d
d
in
g
t
o
k
en
s
in
co
n
tex
t
u
al
ca
lcu
latio
n
s
.
An
o
th
er
d
if
f
er
e
n
ce
lies
in
th
e
m
ea
s
u
r
em
en
t
o
f
s
im
ilar
ity
b
ein
g
lim
ited
to
t
h
e
titl
e
an
d
co
n
ten
t
o
f
o
n
e
n
ews
s
to
r
y
,
n
o
t
th
e
w
h
o
le
n
ews
s
to
r
y
,
an
d
u
s
in
g
s
em
an
ti
cs,
wh
ile
th
is
r
e
s
ea
r
ch
d
o
es
a
s
h
u
f
f
le
to
ca
lcu
late
th
e
s
im
ilar
ity
o
f
s
en
ten
ce
s
r
an
d
o
m
ly
in
co
n
tex
t so
t
h
at
th
e
ac
cu
r
ac
y
o
f
th
e
m
o
d
el
ca
n
b
e
s
e
en
.
2.
RE
L
AT
E
D
R
E
SE
ARCH
I
n
d
o
B
E
R
T
was
d
ev
elo
p
ed
an
d
tr
ain
ed
s
p
ec
if
ically
f
o
r
th
e
I
n
d
o
n
esian
lan
g
u
ag
e
s
o
th
at
it
ca
n
p
r
o
v
id
e
m
o
r
e
ac
c
u
r
ate
r
esu
lts
in
an
aly
zin
g
tex
t
in
I
n
d
o
n
esian
.
I
n
a
s
tu
d
y
en
titl
ed
“
click
b
ait
clas
s
if
icatio
n
m
o
d
el
o
n
o
n
lin
e
n
ews
with
s
em
an
tic
s
im
ilar
ity
ca
lcu
latio
n
b
etwe
en
n
ews
titl
e
an
d
co
n
ten
t
”
[
1
]
,
i
n
th
e
class
if
icatio
n
tech
n
iq
u
e
p
e
r
f
o
r
m
ed
,
it
was
f
o
u
n
d
th
at
I
n
d
o
B
E
R
T
h
ad
an
ac
cu
r
ac
y
r
ate
in
s
em
an
tic
s
im
ilar
ity
o
f
8
7
%.
I
n
a
s
tu
d
y
en
titl
ed
“
i
d
en
tific
atio
n
o
f
tex
t
s
im
ilar
ity
u
s
in
g
class
in
d
ex
in
g
b
ased
an
d
co
s
in
e
s
im
ilar
ity
f
o
r
co
m
p
lain
t
d
o
cu
m
e
n
t
class
if
icatio
n
”
[
5
]
,
u
s
in
g
tex
t
s
im
il
ar
ity
id
en
tific
atio
n
tech
n
iq
u
es
u
s
in
g
class
in
d
ex
in
g
-
b
ased
an
d
co
s
in
e
s
im
ilar
ity
m
eth
o
d
s
to
class
if
y
co
m
p
lain
t
d
o
cu
m
en
ts
,
th
e
ac
cu
r
ac
y
o
f
th
e
r
esear
ch
was
8
4
.
1
2
%.
T
h
e
n
ex
t
r
esear
ch
,
en
titl
ed
“
I
n
d
o
n
esian
n
ews
class
if
icati
o
n
u
s
in
g
I
n
d
o
B
E
R
T
”
[
9
]
,
co
n
d
u
cts
n
ews
r
ec
o
m
m
en
d
atio
n
s
b
ased
o
n
r
ec
o
m
m
en
d
atio
n
s
b
y
co
m
p
ar
i
n
g
th
e
I
n
d
o
B
E
R
T
m
o
d
el
with
XGB.
T
h
e
h
ig
h
est
ac
cu
r
ac
y
wh
e
n
im
p
lem
en
ti
n
g
I
n
d
o
B
E
R
T
is
9
4
.
5
%.
Fro
m
s
ev
er
al
s
tu
d
ies
th
at
b
ec
o
m
e
r
e
f
er
en
ce
s
f
o
r
th
is
r
esear
ch
,
th
e
au
th
o
r
u
s
es
th
e
I
n
d
o
B
E
R
T
m
o
d
el
in
m
ea
s
u
r
in
g
tex
t
s
im
ilar
ity
w
ith
th
e
c
o
s
in
e
m
eth
o
d
o
n
co
n
t
ex
tu
al
s
im
ilar
ity
u
s
in
g
th
e
Py
th
o
n
p
r
o
g
r
am
m
i
n
g
lan
g
u
ag
e.
I
t
is
k
n
o
wn
t
h
at
th
e
im
p
lem
en
tatio
n
o
f
th
e
I
n
d
o
B
E
R
T
m
o
d
el
in
an
aly
zi
n
g
I
n
d
o
n
esian
-
b
ased
d
ata
h
as
h
ig
h
ac
cu
r
ac
y
wh
en
f
in
e
-
tu
n
in
g
.
I
n
ad
d
itio
n
,
c
o
s
in
e
s
im
ilar
ity
,
as
a
v
ec
to
r
ca
lcu
la
tio
n
f
o
r
clu
s
ter
in
g
b
ef
o
r
e
p
er
f
o
r
m
in
g
s
em
an
tic
s
im
ilar
ity
,
is
u
s
ed
t
o
m
ea
s
u
r
e
tex
t
s
im
ilar
ity
in
lan
g
u
ag
e
s
tr
u
ctu
r
e
a
n
d
c
o
n
tex
t.
T
h
er
e
ar
e
s
im
ilar
ities
f
r
o
m
s
ev
er
al
p
r
e
v
io
u
s
s
tu
d
ies,
n
a
m
ely
u
s
in
g
t
h
e
I
n
d
o
B
E
R
T
m
o
d
el
f
o
r
class
if
icatio
n
.
T
h
e
d
if
f
e
r
en
ce
lies
in
th
e
u
s
e
o
f
co
n
tex
tu
al
s
im
ilar
ity
,
wh
i
ch
p
r
o
d
u
ce
s
lab
elin
g
f
r
o
m
th
e
r
esu
lts
o
f
co
s
in
e
ca
lcu
latio
n
s
in
tex
t
s
im
ilar
ity
an
aly
s
is
.
I
n
th
is
r
esear
ch
lead
s
to
tex
t
s
im
ilar
ity
an
aly
s
is
f
o
r
n
ew
d
ata
co
n
tain
in
g
o
p
i
n
io
n
s
with
two
t
o
p
ics,
n
am
ely
th
e
to
p
ic
o
f
in
f
o
r
m
atics
en
g
in
ee
r
i
n
g
a
n
d
th
e
to
p
ic
o
f
in
f
o
r
m
atio
n
s
y
s
tem
s
,
b
y
u
tili
zin
g
th
e
au
g
m
en
tati
o
n
p
r
o
ce
s
s
to
ex
p
an
d
th
e
d
ataset
wh
ich
th
en
s
h
u
f
f
les
th
e
d
ata
to
p
r
o
d
u
ce
a
r
an
d
o
m
s
eq
u
e
n
ce
as
a
r
ef
e
r
e
n
ce
th
at
th
e
d
ata
is
n
o
t
th
e
s
am
e
as
th
e
r
esu
lts
o
f
a
u
g
m
en
tat
io
n
wh
en
s
ep
ar
atin
g
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
38
,
No
.
2
,
May
20
25
:
1
1
7
0
-
1
1
8
0
1172
in
to
two
co
lu
m
n
s
with
th
e
co
l
u
m
n
n
am
es tex
t 1
an
d
tex
t 2
,
c
o
s
in
e
s
im
ilar
i
ty
f
o
r
th
e
p
r
o
ce
s
s
o
f
ca
lcu
latin
g
tex
t
s
im
ilar
ity
with
th
e
a
im
o
f
lab
elin
g
wh
ich
u
tili
ze
s
em
b
ed
d
in
g
to
k
en
s
f
r
o
m
I
n
d
o
B
E
R
T
,
im
p
lem
en
tatio
n
o
f
th
e
I
n
d
o
B
E
R
T
m
o
d
el
in
to
p
ic
clas
s
if
icatio
n
,
an
d
b
i
n
ar
y
class
co
n
f
u
s
io
n
m
atr
ix
as m
o
d
el
test
in
g
.
B
ased
o
n
[
8
]
,
s
em
an
tic
la
b
elin
g
is
th
e
p
r
o
ce
s
s
o
f
m
a
p
p
in
g
attr
ib
u
tes
in
d
ata
s
o
u
r
ce
s
t
o
o
n
to
lo
g
y
class
es
as
an
im
p
o
r
tan
t
s
tep
wh
en
in
teg
r
atin
g
h
eter
o
g
en
e
o
u
s
d
ata.
I
n
th
e
r
esear
ch
“
Sem
an
tic
L
ab
elin
g
:
A
Do
m
ain
-
I
n
d
ep
en
d
en
t
Ap
p
r
o
a
ch
,
”
s
im
ilar
ity
m
etr
ics
ar
e
a
p
p
r
o
ac
h
ed
as
a
co
m
p
ar
is
o
n
f
ea
tu
r
e
f
o
r
lab
ele
d
d
o
m
ain
d
ata.
I
t
is
ex
p
lain
ed
th
at
in
s
em
an
tic
lab
elin
g
,
attr
ib
u
te
v
alu
es
h
av
e
an
im
p
o
r
tan
t
r
o
le
in
id
en
tific
atio
n
with
th
e
s
am
e
s
em
an
tic
ty
p
e.
T
h
e
s
im
ilar
ity
ap
p
r
o
ac
h
ca
r
r
ie
d
o
u
t
in
th
e
s
tu
d
y
h
as
d
if
f
e
r
en
t
m
etr
ics,
in
clu
d
in
g
J
ac
ca
r
d
s
im
ilar
ity
as a
m
o
d
if
ic
atio
n
f
o
r
n
u
m
e
r
ical
v
alu
es a
n
d
T
F
-
I
DF f
o
r
tex
tu
al
d
ata.
C
o
s
in
e
s
im
ilar
ity
is
a
co
m
m
o
n
m
eth
o
d
t
o
p
e
r
f
o
r
m
d
ata
s
im
ilar
ity
,
as
in
th
e
jo
u
r
n
al
“
i
m
p
r
o
v
i
n
g
p
atien
t
clu
s
ter
in
g
b
y
in
c
o
r
p
o
r
atin
g
s
tr
u
ctu
r
ed
lab
el
r
elatio
n
s
h
i
p
s
in
s
im
ilar
ity
m
ea
s
u
r
es
”
[
9
]
,
wh
ich
u
s
es
co
s
in
e
s
im
ilar
ity
to
cla
s
s
if
y
p
atien
t
s
im
ilar
ity
.
T
h
e
u
s
e
o
f
I
n
d
o
B
E
R
T
is
d
o
n
e
wh
en
th
e
av
ailab
le
d
ata
u
s
es
I
n
d
o
n
esian
[
1
0
]
.
T
h
is
is
b
ec
au
s
e
I
n
d
o
B
E
R
T
is
s
p
ec
ially
tr
ain
ed
f
o
r
I
n
d
o
n
esian
,
as sh
o
wn
i
n
F
i
g
u
r
e
1
[
11
]
.
F
i
g
u
r
e
1
.
I
n
d
o
NL
U
b
en
ch
m
ar
k
F
i
g
u
r
e
1
s
h
o
ws
th
e
ty
p
e
o
f
i
n
d
o
b
e
n
ch
m
ar
k
.
T
h
e
p
ar
am
ete
r
s
in
th
e
r
esear
ch
ad
j
u
s
t
th
e
m
o
d
el
ty
p
e
an
d
d
ata
s
ize.
I
n
th
is
r
esear
ch
,
lab
elin
g
is
d
o
n
e
with
th
e
tex
t
s
im
ilar
ity
m
eth
o
d
,
n
am
el
y
co
s
in
e
s
im
ilar
ity
,
wh
ich
u
tili
ze
s
wo
r
d
em
b
e
d
d
in
g
f
r
o
m
th
e
in
d
o
B
E
R
T
m
o
d
el
an
d
th
en
p
e
r
f
o
r
m
s
class
if
icatio
n
b
ased
o
n
co
n
tex
tu
al
s
im
ilar
ity
[1
2
]
,
[
1
3
]
u
s
in
g
th
e
m
o
d
el.
Pre
v
io
u
s
ly
,
th
e
d
ata
will
b
e
ex
p
an
d
ed
u
s
in
g
th
e
d
ata
au
g
m
en
tatio
n
m
eth
o
d
,
n
am
el
y
s
y
n
o
n
y
m
r
e
p
lace
m
en
t,
to
p
er
f
o
r
m
v
ar
iatio
n
s
s
o
as
to
e
m
p
h
asize
th
e
m
o
d
el
p
r
o
v
id
i
n
g
wo
r
d
em
b
e
d
d
in
g
an
d
co
s
in
e
s
im
ilar
ity
,
ca
lcu
l
atin
g
s
im
ilar
it
y
in
d
ata
th
at
h
as b
ee
n
v
ar
ied
.
2
.
1
.
T
ex
t
s
im
ila
rit
y
T
ex
t similar
ity
is
th
e
m
ea
s
u
r
em
en
t o
f
tex
t similar
ity
,
wh
ich
is
th
e
b
asis
o
f
NL
P
task
s
.
T
ex
t similar
ity
is
d
ef
in
ed
as
th
e
s
im
ilar
ity
b
e
twee
n
two
tex
ts
.
No
t
o
n
ly
t
h
a
t,
tex
t
s
im
ilar
ity
als
o
co
n
s
id
er
s
a
b
r
o
a
d
er
c
o
n
tex
t
p
er
s
p
ec
tiv
e
in
an
aly
zin
g
th
e
s
em
an
tic
p
r
o
p
er
ties
o
f
two
wo
r
d
s
[1
4
]
.
T
h
e
m
eth
o
d
o
f
m
ea
s
u
r
in
g
tex
t
s
im
ilar
ity
in
v
o
lv
es two
asp
ec
ts
,
in
clu
d
in
g
:
2
.
1
.
1
.
T
ex
t
d
is
t
a
nce
T
h
er
e
ar
e
t
h
r
ee
way
s
o
f
m
ea
s
u
r
in
g
te
x
t
d
is
tan
ce
b
ased
o
n
l
en
g
th
,
d
is
tr
ib
u
tio
n
,
a
n
d
s
em
an
tic
o
b
jects,
o
n
e
o
f
wh
ic
h
is
co
s
in
e
d
is
tan
ce
.
T
h
e
co
s
in
e
m
ea
s
u
r
em
e
n
t
m
ea
s
u
r
es
th
e
co
s
in
e
an
g
le
b
et
wee
n
th
e
two
tex
ts
.
J
u
d
g
in
g
f
r
o
m
th
e
co
s
in
e
o
f
0
°
b
ein
g
1
an
d
th
e
co
s
i
n
e
o
f
9
0
°
b
ein
g
0
,
th
e
s
im
ilar
ity
v
alu
e
lies
in
th
e
n
u
m
b
er
s
-
1
to
1
,
wh
er
e
th
e
co
s
in
e
m
ea
s
u
r
e
is
r
elate
d
to
o
r
ien
tatio
n
.
A
s
os
in
th
e
f
o
llo
win
g
f
o
r
m
u
la
[1
4
]
:
=
.
|
|
|
|
.
|
|
|
|
∑
∗
=
1
√
∑
(
)
2
−
1
∗
√
∑
(
)
2
−
1
2
.
1
.
2
.
T
ex
t
r
epre
s
ent
a
t
io
n
T
ex
t
r
ep
r
esen
tatio
n
p
er
f
o
r
m
s
ca
lcu
latio
n
s
d
ir
ec
tly
as
n
u
m
e
r
ical
f
ea
tu
r
es
th
at
ar
e
s
im
ilar
in
lex
ical
an
d
s
em
an
tic
way
s
.
L
ex
ical
s
im
ilar
ity
is
d
o
n
e
th
r
o
u
g
h
d
if
f
er
en
t
m
ea
s
u
r
em
en
ts
,
wh
ile
s
em
an
tic
s
im
ilar
ity
is
in
tr
o
d
u
ce
d
th
r
o
u
g
h
s
tr
in
g
-
b
ased
,
co
r
p
u
s
-
b
ased
,
s
em
an
ti
c
tex
t
m
atch
in
g
,
an
d
g
r
ap
h
s
tr
u
ctu
r
e
-
b
ased
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
Wo
r
d
emb
ed
d
in
g
f
o
r
co
n
textu
a
l simil
a
r
ity
u
s
in
g
co
s
in
e
s
imil
a
r
ity
(
Yes
s
y
A
s
r
i
)
1173
m
eth
o
d
s
[1
4
]
.
T
ex
t
s
im
ilar
ity
r
esear
ch
o
f
ten
u
s
es
th
e
co
s
in
e
s
im
ilar
ity
f
o
r
m
u
la
b
ec
au
s
e
it
p
r
o
v
id
es
in
tu
itiv
e
in
ter
p
r
etatio
n
s
an
d
v
alu
es,
wh
ich
r
an
g
e
b
etwe
en
-
1
an
d
1
.
T
h
e
f
o
r
m
u
la
is
s
ca
le
-
co
n
s
is
ten
t,
wh
er
e
v
alu
es
clo
s
e
to
1
ar
e
in
ter
p
r
eted
as
a
h
ig
h
d
eg
r
ee
o
f
s
im
ilar
ity
,
wh
ile
v
alu
es
clo
s
e
to
-
1
in
d
icate
d
is
s
im
ilar
ity
.
I
n
th
e
co
n
tex
t
o
f
tex
t
s
im
ilar
ity
r
ese
ar
ch
o
n
d
o
c
u
m
en
t
d
ata,
co
s
in
e
s
im
ila
r
ity
al
s
o
s
h
o
ws
r
o
b
u
s
t
n
ess
to
d
im
en
s
io
n
al
d
if
f
er
en
ce
s
,
d
em
o
n
s
tr
atin
g
th
e
f
lex
ib
le
n
atu
r
e
th
at
m
ak
es it c
o
m
m
o
n
l
y
u
s
ed
.
C
o
s
in
e
s
im
ilar
ity
ca
n
b
e
c
o
n
s
i
d
er
ed
a
tex
t
s
im
ilar
ity
ca
lcu
la
tio
n
tech
n
iq
u
e
in
th
e
f
r
am
ewo
r
k
o
f
tex
t
r
ep
r
esen
tatio
n
,
with
a
f
o
c
u
s
o
n
th
e
c
ateg
o
r
y
o
f
s
em
an
tic
te
x
t
m
atch
in
g
to
ass
ess
th
e
s
im
ilar
ity
b
etwe
en
tex
t
an
d
d
o
cu
m
en
ts
.
I
n
ad
d
r
ess
in
g
th
e
co
m
p
lex
ity
o
f
s
en
ten
c
e
m
ea
n
in
g
an
d
v
ec
to
r
r
ep
r
es
en
tatio
n
s
th
at
tak
e
in
to
ac
co
u
n
t
in
ter
-
wo
r
d
an
d
co
n
tex
tu
al
r
elatio
n
s
h
ip
p
att
er
n
s
,
th
e
B
E
R
T
m
o
d
el
is
a
r
elev
an
t
ch
o
ice.
T
h
e
co
m
b
in
atio
n
o
f
co
s
in
e
s
im
ilar
ity
ca
lcu
latio
n
with
th
e
u
s
e
o
f
em
b
ed
d
i
n
g
to
k
en
s
f
r
o
m
th
e
I
n
d
o
B
E
R
T
m
o
d
el
is
ch
o
s
en
as
a
m
eth
o
d
f
o
r
co
n
tex
tu
al
s
im
ilar
ity
-
b
as
ed
lab
elin
g
,
co
n
s
id
er
in
g
th
e
co
m
p
lex
ity
o
f
d
ata
ar
is
in
g
f
r
o
m
th
e
r
elatio
n
s
h
ip
b
et
wee
n
s
en
ten
ce
s
to
f
o
r
m
p
ar
a
g
r
ap
h
s
.
2
.
2
.
B
idi
re
ct
io
na
l e
nco
der
re
presenta
t
io
ns
f
ro
m
t
ra
ns
f
o
r
m
er
s
(
B
E
RT
)
B
E
R
T
i
s
th
e
latest
NL
P
alg
o
r
ith
m
d
ev
el
o
p
ed
b
y
Go
o
g
le
.
I
t
was
f
ir
s
t
in
tr
o
d
u
ce
d
b
y
Go
o
g
le
AI
r
esear
ch
er
s
in
2
0
1
8
.
B
E
R
T
u
tili
ze
s
th
e
tr
an
s
f
o
r
m
er
m
o
d
el
in
lear
n
in
g
co
n
tex
tu
al
r
elatio
n
s
h
ip
s
b
etwe
en
wo
r
d
s
in
a
tex
t,
wh
er
e
th
e
tr
an
s
f
o
r
m
er
h
as
two
m
ec
h
an
is
m
s
,
n
a
m
ely
en
co
d
e
r
an
d
d
ec
o
d
e
r
.
H
o
wev
er
,
B
E
R
T
o
n
ly
r
eq
u
ir
es
an
en
c
o
d
er
.
B
E
R
T
u
s
es
a
b
id
ir
ec
tio
n
al
ap
p
r
o
ac
h
an
d
p
er
f
o
r
m
s
s
eq
u
en
tial
r
ea
d
in
g
o
f
tex
t
in
p
u
ts
,
allo
win
g
th
e
m
o
d
el
to
lear
n
t
h
e
co
n
te
x
t
o
f
wo
r
d
s
b
ased
o
n
th
e
s
u
r
r
o
u
n
d
in
g
wo
r
d
s
.
I
n
th
e
en
co
d
er
in
p
u
t,
th
e
s
eq
u
en
ce
o
f
to
k
en
s
will
b
e
em
b
ed
d
e
d
in
to
a
v
ec
to
r
,
wh
ich
will
th
en
b
e
p
ass
ed
o
n
to
th
e
n
eu
r
al
n
etwo
r
k
an
d
o
u
tp
u
t
v
ec
to
r
an
d
g
e
n
er
ated
ac
co
r
d
in
g
to
th
e
in
p
u
t
[1
5
]
.
Fig
u
r
e
2
s
h
o
ws o
f
B
E
R
T
ar
ch
itect
u
r
e.
Fig
u
r
e
2
s
h
o
ws
B
E
R
T
u
tili
ze
s
t
h
e
T
r
an
s
f
o
r
m
er
ar
c
h
itectu
r
e
to
lear
n
co
n
tex
tu
al
r
e
latio
n
s
h
ip
s
b
etwe
en
wo
r
d
s
in
a
tex
t.
T
r
a
n
s
f
o
r
m
er
s
h
av
e
two
m
ec
h
a
n
is
m
s
:
an
en
c
o
d
er
an
d
a
d
ec
o
d
e
r
.
Ho
wev
er
,
B
E
R
T
o
n
ly
h
as a
n
e
n
co
d
e
r
m
ec
h
a
n
is
m
th
at
tak
es a
b
id
ir
ec
tio
n
al
a
p
p
r
o
ac
h
a
n
d
r
ea
d
s
tex
t in
p
u
t se
q
u
en
ti
ally
,
allo
win
g
th
e
m
o
d
el
to
lear
n
c
o
n
tex
t
b
a
s
ed
o
n
s
u
r
r
o
u
n
d
in
g
wo
r
d
s
.
B
E
R
T
h
as
B
E
R
T
-
b
ase
with
as
m
an
y
as
1
2
en
c
o
d
er
lay
er
s
,
7
6
8
h
id
d
en
n
o
d
es,
1
2
atten
tio
n
h
ea
d
s
,
an
d
ab
o
u
t
1
1
0
,
0
0
0
,
0
0
0
p
ar
a
m
eter
s
,
an
d
B
E
R
T
-
lar
g
e
with
as
m
an
y
as 2
4
en
co
d
er
lay
er
s
,
1
0
2
4
h
id
d
e
n
n
o
d
es,
1
6
atten
tio
n
h
ea
d
s
,
an
d
a
b
o
u
t
3
4
0
,
0
0
0
,
0
0
0
p
ar
am
eter
s
[
3
]
.
T
o
k
en
em
b
e
d
d
in
g
in
th
e
co
n
t
ex
t
o
f
B
E
R
T
r
ef
er
s
to
th
e
n
u
m
er
ical
v
ec
to
r
r
ep
r
esen
tatio
n
o
f
a
to
k
en
g
en
er
ated
b
y
a
B
E
R
T
m
o
d
el.
B
E
R
T
is
o
n
e
o
f
th
e
t
r
an
s
f
o
r
m
er
a
r
ch
itectu
r
es
th
at
h
as
p
r
o
v
e
n
to
b
e
v
er
y
ef
f
ec
tiv
e
in
n
at
u
r
al
lan
g
u
ag
e
u
n
d
er
s
tan
d
i
n
g
task
s
,
s
u
ch
as
q
u
esti
o
n
u
n
d
er
s
tan
d
in
g
,
lan
g
u
ag
e
tr
an
s
latio
n
,
an
d
o
th
er
task
s
.
Fig
u
r
e
3
s
h
o
ws
o
f
em
b
ed
d
in
g
to
k
e
n
B
E
R
T
[1
6
]
.
So
m
e
d
etailed
p
o
in
ts
ab
o
u
t
t
o
k
en
e
m
b
ed
d
in
g
in
B
E
R
T
in
clu
d
e
to
k
en
izatio
n
,
e
m
b
ed
d
in
g
lay
e
r
s
,
p
o
s
itio
n
em
b
ed
d
in
g
s
,
s
eg
m
en
t
e
m
b
ed
d
in
g
s
,
f
in
e
-
tu
n
in
g
,
an
d
b
id
ir
ec
tio
n
al
co
n
tex
t.
E
m
b
ed
d
in
g
to
k
en
s
in
B
E
R
T
p
r
o
v
id
es
a
r
ich
an
d
co
n
tex
t
u
al
r
ep
r
esen
tatio
n
f
o
r
ea
c
h
wo
r
d
o
r
s
u
b
wo
r
d
in
th
e
te
x
t,
a
llo
win
g
th
e
m
o
d
el
to
b
etter
co
p
e
with
n
atu
r
al
lan
g
u
ag
e
u
n
d
e
r
s
tan
d
in
g
task
s
.
Fig
u
r
e
2
.
B
E
R
T
a
r
ch
itectu
r
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
38
,
No
.
2
,
May
20
25
:
1
1
7
0
-
1
1
8
0
1174
Fig
u
r
e
3
.
E
m
b
ed
d
in
g
t
o
k
en
B
E
R
T
3.
M
E
T
H
O
D
Fig
u
r
e
4
illu
s
tr
ates
th
e
wo
r
k
f
l
o
w
m
eth
o
d
o
lo
g
y
f
o
r
em
b
ed
d
i
n
g
wo
r
d
f
o
r
co
n
tex
t
u
al
s
im
ilar
ity
u
s
in
g
co
s
in
e
s
im
ilar
ity
.
I
n
th
e
f
ir
s
t
s
tag
e,
in
f
o
r
m
atio
n
is
s
ea
r
ch
ed
o
n
o
p
in
io
n
d
ata
o
n
th
e
t
o
p
ic
o
f
in
f
o
r
m
atics
en
g
in
ee
r
in
g
an
d
in
f
o
r
m
atio
n
s
y
s
tem
s
o
n
th
e
Qu
o
r
a
p
latf
o
r
m
.
Data
r
etr
ie
v
al
in
th
is
s
tu
d
y
u
s
es
a
s
am
p
lin
g
tech
n
iq
u
e
ca
lled
s
im
p
le
r
an
d
o
m
s
am
p
lin
g
,
wh
er
e
d
ata
i
s
tak
en
f
r
o
m
th
e
two
r
eq
u
i
r
ed
to
p
ics,
n
am
ely
in
f
o
r
m
atics
en
g
i
n
ee
r
in
g
an
d
i
n
f
o
r
m
atio
n
s
y
s
tem
s
,
an
d
f
r
o
m
b
o
th
to
p
ics
will
h
av
e
a
n
u
m
b
er
o
f
d
ata
f
r
o
m
t
h
e
two
to
p
ics
in
th
e
s
am
e
d
ata
s
o
th
at
it
o
n
ly
h
as
o
n
e
o
p
in
io
n
c
o
lu
m
n
th
at
will
b
e
u
s
ed
in
th
e
n
ex
t
s
tag
e
o
f
d
ata
p
r
ep
ar
atio
n
.
Fu
r
t
h
er
m
o
r
e,
th
e
d
ata
p
r
ep
ar
atio
n
p
r
o
ce
s
s
in
clu
d
es
web
s
cr
ap
i
n
g
,
d
ata
p
r
e
-
p
r
o
ce
s
s
in
g
,
d
ata
lab
elin
g
,
an
d
d
ata
s
p
litt
in
g
f
o
r
m
o
d
elin
g
u
s
ed
in
t
h
e
r
esear
ch
.
I
n
o
n
e
o
f
s
tep
,
th
e
r
e
h
a
v
e
s
tep
f
o
r
th
em
s
elf
.
Fig
u
r
e
5
s
h
o
ws
th
e
d
ata
p
r
e
p
ar
atio
n
wo
r
k
f
lo
w
an
d
F
ig
u
r
e
6
s
h
o
ws
ab
o
u
t
d
ata
p
r
ep
r
o
ce
s
s
in
g
wo
r
k
f
lo
w[
1
7
]
.
Fig
u
r
e
6
s
h
o
ws
th
e
p
r
e
-
p
r
o
ce
s
s
in
g
d
ata
wo
r
k
f
lo
w
[
1
8
]
.
Af
ter
o
b
tain
in
g
o
p
in
io
n
d
ata
f
r
o
m
t
h
e
two
t
o
p
ics
th
at
will
b
e
th
e
o
b
ject
o
f
r
esear
ch
,
t
h
en
,
d
ata
p
r
e
-
p
r
o
ce
s
s
in
g
in
clu
d
es
ca
s
e
f
o
ld
in
g
,
d
elete
d
u
n
iq
u
e
ch
ar
ac
ter
s
,
s
lan
g
wo
r
d
s
,
an
d
to
k
e
n
izatio
n
b
y
ad
d
i
n
g
a
s
tep
to
r
em
o
v
e
n
u
m
b
er
s
th
at
ar
e
n
o
lo
n
g
er
n
ee
d
ed
.
Fig
u
r
e
4
.
T
h
e
r
esear
ch
w
o
r
k
f
l
o
w
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
Wo
r
d
emb
ed
d
in
g
f
o
r
co
n
textu
a
l simil
a
r
ity
u
s
in
g
co
s
in
e
s
imil
a
r
ity
(
Yes
s
y
A
s
r
i
)
1175
Fig
u
r
e
5
.
Data
p
r
ep
ar
atio
n
wo
r
k
f
lo
w
Fig
u
r
e
6
.
T
h
e
p
r
e
-
p
r
o
ce
s
s
in
g
d
ata
wo
r
k
f
l
o
w
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
Data
au
g
m
en
tatio
n
is
a
co
llec
tio
n
o
f
alg
o
r
ith
m
s
th
at
cr
ea
te
s
y
n
th
etic
d
ata
b
y
m
a
k
in
g
s
m
a
ll
ch
an
g
es
to
ex
is
tin
g
d
ata,
aim
in
g
to
ex
p
an
d
th
e
am
o
u
n
t
o
f
tr
ain
in
g
d
ata
in
d
ee
p
n
eu
r
al
n
etwo
r
k
lear
n
in
g
[1
4
]
–
[1
6
]
.
T
h
is
te
ch
n
iq
u
e
is
u
s
ef
u
l
f
o
r
o
b
s
er
v
in
g
m
o
d
el
f
ailu
r
es
an
d
im
p
r
o
v
i
n
g
th
eir
p
er
f
o
r
m
a
n
ce
.
Data
au
g
m
en
tatio
n
is
an
im
p
o
r
tan
t
s
tep
in
m
o
d
el
tr
a
in
in
g
,
h
el
p
in
g
t
o
o
v
er
co
m
e
th
e
lim
itatio
n
s
o
f
lim
ited
d
ata.
I
t
is
also
co
n
s
id
er
ed
a
co
s
t
-
ef
f
icien
t
way
to
in
cr
ea
s
e
d
ata
s
ize,
r
ed
u
ce
tr
ain
in
g
er
r
o
r
s
,
an
d
p
r
o
d
u
ce
m
o
r
e
ac
cu
r
ate
p
r
ed
ictio
n
s
.
Ma
n
y
m
ac
h
in
e
lear
n
in
g
p
r
o
jects r
ely
o
n
d
ata
a
u
g
m
e
n
tatio
n
a
s
th
eir
cr
itical
s
u
cc
ess
f
ac
to
r
[1
8
]
.
T
h
e
r
ef
o
r
e,
a
d
ata
au
g
m
en
tatio
n
m
eth
o
d
is
p
er
f
o
r
m
e
d
to
p
er
f
o
r
m
s
en
ten
ce
v
ar
iatio
n
u
s
in
g
s
y
n
o
n
y
m
r
ep
lace
m
en
t.
I
n
d
o
in
g
s
o
,
a
r
a
n
d
o
m
wo
r
d
will
b
e
s
ea
r
ch
ed
an
d
ch
an
g
e
d
in
to
an
o
th
er
w
o
r
d
th
at
h
as
th
e
s
am
e
m
ea
n
in
g
.
Ho
wev
er
,
th
e
p
r
o
ce
s
s
d
o
es
n
o
t
in
v
o
lv
e
w
o
r
d
s
th
at
a
r
e
ca
te
g
o
r
ized
as
s
to
p
wo
r
d
s
.
I
n
th
e
ch
an
g
ed
wo
r
d
,
it
is
p
o
s
s
ib
le
f
o
r
th
e
em
b
ed
d
in
g
wo
r
d
to
h
a
v
e
a
d
if
f
er
en
t
n
u
m
b
er
v
al
u
e
f
r
o
m
th
e
ac
tu
al
wo
r
d
.
T
h
e
r
esu
lts
will
b
e
co
m
b
in
ed
with
th
e
d
ata
b
ef
o
r
e
th
e
ch
a
n
g
es
ar
e
m
ad
e,
an
d
to
ad
d
to
th
e
im
p
r
o
v
em
en
t
,
r
an
d
o
m
iza
tio
n
o
f
th
e
o
r
d
e
r
wi
ll
b
e
ca
r
r
ied
o
u
t
s
o
th
at
t
h
e
d
i
v
is
io
n
o
f
th
e
two
c
o
lu
m
n
s
wil
l
n
o
t
h
a
v
e
th
e
s
am
e
s
en
ten
ce
,
alth
o
u
g
h
it is
p
o
s
s
ib
le
to
h
av
e
s
im
ilar
p
o
s
itio
n
s
in
r
an
d
o
m
izatio
n
.
I
n
th
e
am
o
u
n
t o
f
4
1
4
d
ata,
it is
s
til
l r
elativ
ely
s
m
all
f
o
r
th
e
u
s
e
o
f
a
s
er
ies o
f
d
ee
p
lear
n
in
g
p
r
o
ce
s
s
es;
th
er
ef
o
r
e,
d
ata
au
g
m
en
tatio
n
i
s
ca
r
r
ied
o
u
t
u
s
in
g
th
e
s
y
n
o
n
y
m
r
ep
lace
m
en
t
tech
n
iq
u
e,
wh
i
ch
tak
es
o
n
e
wo
r
d
at
r
an
d
o
m
a
n
d
r
ep
lace
s
it with
th
e
s
am
e
s
en
ten
ce
wh
er
e
th
e
s
en
ten
ce
is
in
th
e
eq
u
atio
n
co
n
tain
ed
in
W
o
r
d
Net.
T
h
e
p
r
o
ce
s
s
ca
n
b
e
s
ee
n
in
Fig
u
r
e
7
,
with
t
h
e
r
esu
lt in
Fig
u
r
e
8
.
Fig
u
r
e
7
.
So
u
r
ce
co
d
e
o
f
s
y
n
o
n
im
r
ep
lacm
en
t
Fig
u
r
e
8
.
R
esu
lt o
f
s
eg
m
e
n
tatio
n
d
ata
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
38
,
No
.
2
,
May
20
25
:
1
1
7
0
-
1
1
8
0
1176
Af
ter
au
g
m
en
tin
g
th
e
d
ata
s
o
th
at
it
h
as
8
2
8
s
en
ten
ce
s
wh
er
e
th
e
d
ata
is
in
d
ir
ec
tly
s
eq
u
en
tial
wh
en
p
er
f
o
r
m
in
g
th
e
m
er
g
e
co
m
m
an
d
,
r
an
d
o
m
izatio
n
is
p
er
f
o
r
m
ed
o
n
th
e
d
ata
to
av
o
id
s
im
ilar
ities
b
etwe
en
s
en
ten
ce
s
f
r
o
m
th
e
o
r
ig
i
n
al
d
ata
an
d
s
en
ten
ce
s
f
r
o
m
th
e
a
u
g
m
en
ted
d
ata
t
h
at
will
b
e
ca
l
cu
lated
.
Fig
u
r
e
9
is
th
e
r
esu
lt
o
f
th
e
d
ata
s
h
u
f
f
le.
B
ef
o
r
e
p
er
f
o
r
m
in
g
ca
lc
u
latio
n
s
u
s
in
g
I
n
d
o
B
E
R
T
o
n
m
ea
s
u
r
in
g
tex
t
s
im
ilar
ity
with
co
s
in
e,
th
e
d
ata
is
d
i
v
id
e
d
in
to
two
c
o
lu
m
n
s
,
n
a
m
ely
te
x
t 1
an
d
tex
t 2
,
wh
ich
ca
n
b
e
s
ee
n
in
Fig
u
r
e
1
0
.
Fig
u
r
e
9
.
R
esu
lt o
f
s
h
u
f
f
le
d
at
a
Fig
u
r
e
1
0
.
Div
is
io
n
r
esu
lts
in
to
two
co
lu
m
n
s
T
h
e
n
e
x
t
s
tep
is
d
ata
la
b
elin
g
b
ased
o
n
co
s
in
e
s
im
ilar
ity
ca
lcu
latio
n
s
.
C
o
s
in
e
s
im
ilar
ity
p
r
o
v
id
es
lab
elin
g
b
y
ca
lcu
latin
g
th
e
s
im
ilar
ity
o
f
o
p
in
io
n
d
ata
o
n
th
e
in
f
o
r
m
atio
n
en
g
in
ee
r
in
g
lab
el
an
d
th
e
in
f
o
r
m
atio
n
s
y
s
tem
s
lab
el
af
t
er
ca
lcu
latin
g
t
h
e
s
im
ilar
ity
b
etwe
en
th
e
two
tex
t
co
lu
m
n
s
.
T
h
en
t
h
e
lab
elin
g
u
s
es
th
e
av
er
ag
e
v
alu
e
as
a
th
r
esh
o
ld
to
p
r
o
d
u
ce
b
alan
ce
d
d
ata,
b
ec
au
s
e
if
y
o
u
u
s
e
a
th
r
esh
o
ld
b
ased
o
n
th
e
h
ig
h
est
o
r
lo
west
v
alu
e
,
it
wi
ll
r
esu
lt
i
n
a
to
tal
o
f
o
n
e
lab
el
h
av
in
g
h
u
n
d
r
ed
s
m
o
r
e
d
at
a
th
an
o
th
er
lab
els,
wh
ich
o
n
ly
h
av
e
ten
s
o
f
th
o
u
s
an
d
s
o
f
d
ata
p
o
in
ts
.
T
h
e
lab
eli
n
g
s
tates
th
at
0
i
s
n
o
t
s
im
ilar
,
wh
ile
1
is
s
im
ilar
.
Fig
u
r
e
1
1
s
h
o
ws o
f
th
e
lab
elli
n
g
p
r
o
ce
s
s
.
Fig
u
r
e
1
1
.
L
ab
ellin
g
p
r
o
ce
s
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
Wo
r
d
emb
ed
d
in
g
f
o
r
co
n
textu
a
l simil
a
r
ity
u
s
in
g
co
s
in
e
s
imil
a
r
ity
(
Yes
s
y
A
s
r
i
)
1177
Data
lab
elin
g
is
b
ased
o
n
c
o
s
in
e
s
im
ilar
ity
ca
lcu
latio
n
s
,
wh
i
ch
ca
n
b
e
s
ee
n
in
Fig
u
r
es
1
2
an
d
1
3
f
o
r
th
e
r
esu
lts
o
f
s
ea
r
ch
in
g
f
o
r
th
e
h
ig
h
est,
lo
west,
an
d
av
e
r
ag
e
v
alu
es.
T
h
en
th
e
la
b
elin
g
u
s
e
s
th
e
av
er
a
g
e
v
alu
e
as
a
th
r
esh
o
ld
to
p
r
o
d
u
ce
b
ala
n
ce
d
d
ata,
b
e
ca
u
s
e
if
y
o
u
u
s
e
a
th
r
esh
o
ld
b
ased
o
n
th
e
h
ig
h
est
o
r
lo
west
v
alu
e,
it
will
r
esu
lt
in
a
to
tal
o
f
o
n
e
lab
el
h
av
in
g
h
u
n
d
r
ed
s
m
o
r
e
d
ata
th
an
o
th
er
la
b
els,
wh
ich
o
n
ly
h
av
e
te
n
s
o
f
th
o
u
s
an
d
s
o
f
d
ata
p
o
in
ts
.
T
h
e
lab
elin
g
s
tates
th
at
0
is
n
o
t
s
im
ilar
,
wh
ile
1
is
s
im
ilar
.
T
h
e
lab
elin
g
r
esu
lts
ca
n
b
e
s
ee
n
in
F
ig
u
r
e
1
4
.
Fig
u
r
e
1
2
.
C
o
s
in
e
s
im
ilar
ity
r
esu
lt
Fig
u
r
e
1
3
.
Sear
ch
f
o
r
h
ig
h
est,
lo
west
,
an
d
av
er
a
g
e
v
alu
es o
f
co
s
in
e
v
alu
es
Fig
u
r
e
14
.
Data
lab
elin
g
r
esu
lt
s
Af
ter
co
s
in
e
s
im
ilar
ity
m
ea
s
u
r
em
en
t
u
tili
zin
g
I
n
d
o
B
E
R
T
wo
r
d
em
b
ed
d
in
g
as
lab
elin
g
,
th
e
I
n
d
o
B
E
R
T
m
o
d
el
is
u
s
ed
to
p
er
f
o
r
m
co
n
tex
tu
al
s
im
ilar
ity
-
b
ased
class
if
icat
io
n
.
I
n
th
is
s
tu
d
y
,
th
e
d
ata
s
h
ar
in
g
in
d
ex
was
8
0
:1
0
:1
0
,
with
tr
ain
in
g
d
ata
b
ein
g
8
0
%,
wh
ile
v
al
id
atio
n
d
ata
an
d
test
in
g
d
ata
w
er
e
1
0
%
ea
ch
.
T
h
e
f
lo
w
o
f
th
e
I
n
d
o
B
E
R
T
m
o
d
el
r
eq
u
ir
es
tr
ain
in
g
d
ata,
an
d
v
a
lid
atio
n
d
ata
will
b
e
at
th
e
B
e
r
tTo
k
en
izer
m
o
d
el
s
tag
e
b
y
h
av
in
g
v
ar
io
u
s
p
ar
am
eter
s
th
at
s
u
p
p
o
r
t
th
e
p
r
o
ce
s
s
,
in
clu
d
in
g
m
ax
_
len
g
th
1
2
8
,
b
atch
_
s
ize
3
2
,
ep
o
ch
5
,
an
d
lear
n
in
g
r
ate
1
e
-
5
.
M
o
d
el
s
ize
an
d
t
h
e
n
u
m
b
er
o
f
p
a
r
am
eter
s
g
iv
e
a
n
in
d
icatio
n
o
f
th
e
co
m
p
lex
ity
an
d
m
em
o
r
y
r
e
q
u
i
r
em
en
ts
o
f
a
m
o
d
el.
Mo
d
els
with
a
lar
g
e
n
u
m
b
er
o
f
p
ar
am
eter
s
ten
d
to
r
eq
u
ir
e
m
o
r
e
co
m
p
u
tatio
n
al
r
eso
u
r
ce
s
an
d
m
em
o
r
y
.
T
h
is
in
f
o
r
m
atio
n
is
u
s
ef
u
l
f
o
r
ev
alu
atin
g
w
h
eth
er
th
e
m
o
d
el
is
s
u
itab
le
f
o
r
u
s
e
o
n
r
eso
u
r
ce
-
co
n
s
tr
ain
ed
d
ev
ices.
B
y
u
n
d
e
r
s
tan
d
in
g
th
e
s
ize
an
d
co
m
p
l
ex
ity
o
f
th
e
m
o
d
el,
d
ec
is
io
n
s
r
eg
ar
d
in
g
th
e
im
p
l
em
en
tatio
n
a
n
d
u
s
e
o
f
th
e
m
o
d
el
ca
n
b
e
b
etter
co
n
s
id
er
e
d
,
esp
ec
ially
i
n
th
e
co
n
tex
t o
f
r
eso
u
r
ce
co
n
s
tr
ain
ts
o
n
th
e
d
ev
ice.
T
h
e
n
ex
t
s
tag
e
is
f
in
e
-
tu
n
in
g
,
ad
j
u
s
tin
g
th
e
f
ea
tu
r
e
r
ep
r
esen
tatio
n
th
at
alr
ea
d
y
ex
is
ts
in
B
E
R
T
[1
9
]
–
[
21
]
ac
c
o
r
d
i
n
g
to
th
e
ch
a
r
ac
ter
is
t
ics
o
f
th
e
d
ata
s
et.
Af
ter
g
o
in
g
th
r
o
u
g
h
th
e
m
o
d
el
tr
ain
i
n
g
s
tag
e
u
s
in
g
tr
ain
in
g
d
ata
a
n
d
v
alid
atio
n
d
ata,
th
e
n
ex
t
s
tep
is
to
test
th
e
m
o
d
el
with
test
in
g
d
ata.
W
h
en
m
ak
in
g
p
r
ed
ictio
n
s
o
n
test
in
g
d
ata,
t
h
e
m
o
d
el
will
p
r
o
d
u
ce
a
p
er
ce
n
tag
e
p
r
o
b
ab
ilit
y
o
f
co
n
f
id
en
ce
r
elate
d
to
th
e
p
r
ed
ictio
n
s
m
ad
e.
T
h
e
h
ig
h
er
th
e
p
r
o
b
ab
ilit
y
g
iv
en
b
y
th
e
m
o
d
el,
th
e
m
o
r
e
c
o
n
f
i
d
en
t
t
h
e
m
o
d
el
is
i
n
its
p
r
ed
ictio
n
r
esu
lts
.
T
h
is
p
r
o
ce
s
s
p
r
o
v
id
es
in
s
ig
h
t
in
to
t
h
e
m
o
d
el
’
s
lev
el
o
f
co
n
f
id
en
ce
i
n
th
e
p
r
ed
ictio
n
r
esu
lts
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
38
,
No
.
2
,
May
20
25
:
1
1
7
0
-
1
1
8
0
1178
g
iv
en
o
n
d
ata
t
h
a
t
was
n
o
t
u
s
ed
d
u
r
in
g
th
e
tr
ain
in
g
an
d
v
al
id
atio
n
p
r
o
ce
s
s
es.
T
h
er
e
is
a
p
r
o
b
a
b
ilit
y
co
lu
m
n
f
o
r
th
e
p
e
r
ce
n
tag
e
o
f
co
n
f
id
e
n
ce
th
at
t
h
e
m
o
d
el
p
lace
s
th
e
tex
t
in
th
at
ca
teg
o
r
y
.
T
h
e
co
n
f
u
s
io
n
m
atr
ix
is
a
tech
n
iq
u
e
th
at
is
o
f
ten
u
s
ed
in
th
e
class
if
icatio
n
o
f
m
o
d
el
r
esu
lts
b
ased
o
n
s
tim
u
lated
o
b
jects
[
22
]
,
[
2
3
]
.
T
h
e
co
n
f
u
s
io
n
m
atr
i
x
p
er
f
o
r
m
ed
is
a
ty
p
e
o
f
b
in
ar
y
class
co
n
f
u
s
io
n
m
atr
ix
.
Fig
u
r
e
1
5
is
an
im
ag
e
o
f
th
e
p
lace
m
en
t
o
f
th
e
b
in
a
r
y
class
co
n
f
u
s
io
n
m
atr
ix
[
6
]
,
[
16
]
,
[
22
]
,
[
2
4
]
,
[
2
5
]
,
a
n
d
Fig
u
r
e
1
6
is
th
e
r
esu
lt
o
f
th
e
co
n
f
u
s
io
n
m
atr
ix
f
r
o
m
t
h
e
r
e
s
ea
r
ch
.
T
h
e
ev
alu
atio
n
o
f
th
e
m
o
d
el
in
th
is
r
esear
ch
u
s
es
a
co
n
f
u
s
io
n
m
atr
ix
,
wh
ich
will
ca
lcu
late
ac
cu
r
a
cy
v
alu
es
in
th
e
tex
t
s
im
ilar
ity
an
aly
s
is
f
o
r
th
e
class
if
icatio
n
th
at
h
as
b
ee
n
ca
r
r
ie
d
o
u
t.
Fro
m
t
h
e
co
n
f
u
s
io
n
m
at
r
ix
,
th
e
ac
cu
r
ac
y
v
al
u
e
o
f
I
n
d
o
B
E
R
T
is
9
0
%.
Fig
u
r
e
1
5
.
C
o
n
f
u
s
s
io
n
m
atr
ix
Fig
u
r
e
16
.
C
o
n
f
u
s
s
io
n
m
atr
ix
r
esu
lt
5.
CO
NCLU
SI
O
N
B
ased
o
n
th
e
r
esu
lts
o
f
th
e
r
esear
ch
an
d
d
is
cu
s
s
io
n
,
it
ca
n
b
e
c
o
n
clu
d
ed
th
at
th
e
d
ata
s
im
ilar
ity
an
aly
s
is
p
r
o
ce
s
s
co
n
s
is
ts
o
f
d
a
ta
co
llectio
n
,
d
ata
p
r
e
-
p
r
o
ce
s
s
in
g
,
d
ata
au
g
m
e
n
tatio
n
,
ca
lcu
l
atin
g
th
e
s
im
ilar
ity
o
f
tex
t
in
th
e
d
ata,
lab
elin
g
t
h
e
d
ata
ac
co
r
d
in
g
t
o
th
e
av
e
r
ag
e
v
alu
e
o
f
th
e
r
esu
lts
o
f
th
e
co
s
in
e
s
im
ilar
ity
ca
lcu
latio
n
,
wh
ich
will
b
ec
o
m
e
th
e
th
r
esh
o
ld
,
T
h
e
ca
lcu
la
tio
n
o
f
tex
t
s
im
ilar
ity
b
ec
o
m
e
s
lab
elin
g
with
two
lab
els,
n
am
ely
lab
el
0
is
n
o
t
s
im
ilar
wh
ile
lab
el
1
i
s
a
s
i
m
ilar
lab
el.
I
n
th
e
I
n
d
o
B
E
R
T
m
o
d
el,
lab
el
0
is
r
ep
r
esen
ted
as
an
I
S
to
p
ic
cl
ass
if
icatio
n
an
d
lab
el
1
as
an
I
T
lab
el
class
if
icatio
n
.
I
n
th
is
r
esear
ch
,
it
was
co
n
clu
d
e
d
th
at
th
e
I
n
d
o
B
E
R
T
m
o
d
el
h
a
d
ca
lcu
latio
n
ac
c
u
r
a
cy
r
esu
lts
u
s
in
g
a
co
n
f
u
s
io
n
m
atr
ix
wo
r
th
9
0
%
in
ca
lcu
latio
n
s
an
d
u
n
d
er
s
tan
d
in
g
o
f
co
n
te
x
tu
al
tex
t
b
ec
au
s
e
it
h
ad
b
ee
n
tr
ain
ed
s
p
ec
if
ically
f
o
r
d
ata
i
n
I
n
d
o
n
esian
.
I
n
o
r
d
er
to
p
er
f
ec
t
th
e
r
esu
ltin
g
tex
t
s
im
ilar
ity
an
aly
s
is
,
th
er
e
ar
e
s
ev
er
al
s
u
g
g
esti
o
n
s
,
in
clu
d
in
g
ad
d
in
g
G
o
o
g
le
T
r
an
s
late
to
t
h
e
p
r
o
g
r
am
m
i
n
g
b
ef
o
r
e
d
o
in
g
s
la
n
g
wo
r
d
s
f
o
r
s
ev
er
al
w
o
r
d
s
in
s
en
ten
ce
s
th
at
u
s
e
f
o
r
eig
n
lan
g
u
ag
es
s
o
t
h
at
it
ca
n
im
p
r
o
v
e
c
o
n
tex
t
ca
lcu
latio
n
s
,
an
d
it
is
h
o
p
ed
th
at
d
if
f
er
en
t
I
n
d
o
b
en
ch
m
a
r
k
s
will
b
e
u
s
ed
,
s
u
ch
as
I
n
d
o
B
E
R
T
L
ite
o
r
I
n
d
o
B
E
R
T
L
ar
g
e
in
ty
p
e
in
d
o
b
e
n
ch
m
ar
k
-
p
1
o
r
t
y
p
e
in
d
o
b
e
n
ch
m
ar
k
-
p
2
.
ACK
NO
WL
E
DG
E
M
E
NT
S
Ou
r
d
ee
p
est
g
r
atitu
d
e
g
o
es
to
th
e
I
n
s
titu
te
f
o
r
C
o
m
m
u
n
ity
S
er
v
ice
R
esear
ch
(
L
PP
M)
o
f
th
e
I
n
s
titu
te
T
ec
h
n
o
lo
g
y
o
f
PLN
f
o
r
f
u
n
d
in
g
th
e
co
m
m
u
n
ity
s
er
v
ice
ac
tiv
ities
f
o
r
f
is
ca
l y
ea
r
2
0
2
3
.
RE
F
E
R
E
NC
E
S
[
1
]
H
.
A
.
A
h
m
a
d
i
a
n
d
A
.
C
h
o
w
a
n
d
a
,
“
C
l
i
c
k
b
a
i
t
c
l
a
ssi
f
i
c
a
t
i
o
n
m
o
d
e
l
o
n
o
n
l
i
n
e
n
e
w
s
w
i
t
h
s
e
ma
n
t
i
c
si
mi
l
a
r
i
t
y
c
a
l
c
u
l
a
t
i
o
n
b
e
t
w
e
e
n
n
e
w
s
t
i
t
l
e
a
n
d
c
o
n
t
e
n
t
,
”
B
u
i
l
d
i
n
g
o
f
I
n
f
o
rm
a
t
i
c
s
,
T
e
c
h
n
o
l
o
g
y
a
n
d
S
c
i
e
n
c
e
(
BI
T
S
)
,
v
o
l
.
4
,
n
o
.
4
,
M
a
r
.
2
0
2
3
,
d
o
i
:
1
0
.
4
7
0
6
5
/
b
i
t
s.
v
4
i
4
.
3
0
3
0
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
Wo
r
d
emb
ed
d
in
g
f
o
r
co
n
textu
a
l simil
a
r
ity
u
s
in
g
co
s
in
e
s
imil
a
r
ity
(
Yes
s
y
A
s
r
i
)
1179
[
2
]
L.
S
a
h
u
a
n
d
B
.
R
.
M
o
h
a
n
,
“
A
n
i
mp
r
o
v
e
d
K
-
mea
n
s
a
l
g
o
r
i
t
h
m
u
si
n
g
m
o
d
i
f
i
e
d
c
o
si
n
e
d
i
st
a
n
c
e
mea
s
u
r
e
f
o
r
d
o
c
u
me
n
t
c
l
u
s
t
e
r
i
n
g
u
si
n
g
M
a
h
o
u
t
w
i
t
h
H
a
d
o
o
p
,
”
2
0
1
4
9
t
h
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
d
u
st
ri
a
l
a
n
d
I
n
f
o
rm
a
t
i
o
n
S
y
s
t
e
m
s
(
I
C
I
I
S
)
,
G
w
a
l
i
o
r
,
I
n
d
i
a
,
2
0
1
4
,
p
p
.
1
-
5
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
I
I
N
F
S
.
2
0
1
4
.
7
0
3
6
6
6
1
.
[
3
]
J.
D
e
v
l
i
n
,
M
.
-
W
.
C
h
a
n
g
,
K
.
Le
e
,
K
.
T.
G
o
o
g
l
e
,
a
n
d
A
.
I
.
La
n
g
u
a
g
e
,
“
B
E
R
T:
p
re
-
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
mers
f
o
r
l
a
n
g
u
a
g
e
u
n
d
e
r
st
a
n
d
i
n
g
,
”
N
a
a
c
l
-
Hl
t
2
0
1
9
,
p
p
.
4
1
7
1
–
4
1
8
6
,
2
0
1
8
.
[
4
]
G
.
Jawa
h
a
r
,
B
.
S
a
g
o
t
,
a
n
d
D
.
S
e
d
d
a
h
,
“
W
h
a
t
d
o
e
s
B
E
R
T
l
e
a
r
n
a
b
o
u
t
t
h
e
st
r
u
c
t
u
r
e
o
f
l
a
n
g
u
a
g
e
?
,
”
i
n
A
C
L
2
0
1
9
-
5
7
t
h
An
n
u
a
l
Me
e
t
i
n
g
o
f
t
h
e
As
so
c
i
a
t
i
o
n
f
o
r
C
o
m
p
u
t
a
t
i
o
n
a
l
L
i
n
g
u
i
s
t
i
c
s,
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
C
o
n
f
e
re
n
c
e
,
2
0
2
0
,
p
p
.
3
6
5
1
–
3
6
5
7
,
d
o
i
:
1
0
.
1
8
6
5
3
/
v
1
/
p
1
9
-
1
3
5
6
.
[
5
]
J.
Y
e
,
“
I
mp
r
o
v
e
d
c
o
s
i
n
e
si
m
i
l
a
r
i
t
y
mea
s
u
r
e
s
o
f
si
mp
l
i
f
i
e
d
n
e
u
t
r
o
s
o
p
h
i
c
se
t
s
f
o
r
me
d
i
c
a
l
d
i
a
g
n
o
ses
,
”
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
i
n
Me
d
i
c
i
n
e
,
v
o
l
.
6
3
,
n
o
.
3
,
p
p
.
1
7
1
–
1
7
9
,
2
0
1
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
r
t
me
d
.
2
0
1
4
.
1
2
.
0
0
7
.
[
6
]
Z.
I
mt
i
a
z
,
M
.
U
m
e
r
,
M
.
A
h
m
a
d
,
S
.
U
l
l
a
h
,
G
.
S
.
C
h
o
i
,
a
n
d
A
.
M
e
h
m
o
o
d
,
“
D
u
p
l
i
c
a
t
e
q
u
e
s
t
i
o
n
s
p
a
i
r
d
e
t
e
c
t
i
o
n
u
s
i
n
g
s
i
a
m
e
se
M
a
LS
TM
,
”
I
E
EE
Ac
c
e
ss
,
v
o
l
.
8
,
p
p
.
2
1
9
3
2
–
2
1
9
4
2
,
2
0
2
0
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
0
.
2
9
6
9
0
4
1
.
[
7
]
E.
U
z
u
n
,
“
A
n
o
v
e
l
w
e
b
scra
p
i
n
g
a
p
p
r
o
a
c
h
u
si
n
g
t
h
e
a
d
d
i
t
i
o
n
a
l
i
n
f
o
r
mat
i
o
n
o
b
t
a
i
n
e
d
f
r
o
m
w
e
b
p
a
g
e
s,
”
i
n
I
EEE
A
c
c
e
ss
,
v
o
l
.
8
,
p
p
.
6
1
7
2
6
-
6
1
7
4
0
,
2
0
2
0
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
0
.
2
9
8
4
5
0
3
.
[
8
]
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
,
”
i
n
3
1
st
C
o
n
f
e
re
n
c
e
o
n
N
e
u
r
a
l
I
n
f
o
rm
a
t
i
o
n
Pr
o
c
e
ss
i
n
g
S
y
st
e
m
s
(
N
I
PS
2
0
1
7
)
,
2
0
1
7
,
p
p
.
5
9
9
9
–
6
0
0
9
.
[
9
]
B
.
Ju
a
r
t
o
,
“
I
n
t
e
r
n
a
t
i
o
n
a
l
j
o
u
r
n
a
l
o
f
i
n
t
e
l
l
i
g
e
n
t
sy
s
t
e
ms
a
n
d
a
p
p
l
i
c
a
t
i
o
n
s
i
n
e
n
g
i
n
e
e
r
i
n
g
I
n
d
o
n
e
si
a
n
n
e
w
s
c
l
a
ssi
f
i
c
a
t
i
o
n
u
si
n
g
I
n
d
o
B
ER
T
,
”
O
ri
g
i
n
a
l
Re
se
a
r
c
h
P
a
p
e
r
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
I
n
t
e
l
l
i
g
e
n
t
S
y
st
e
m
s
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
i
n
E
n
g
i
n
e
e
ri
n
g
I
J
I
S
AE
,
v
o
l
.
2
0
2
3
,
n
o
.
2
,
2
0
2
3
.
[
10
]
L.
W
u
e
t
a
l
.
,
“
W
o
r
d
mo
v
e
r
’
s
e
m
b
e
d
d
i
n
g
:
f
r
o
m
w
o
r
d
2
v
e
c
t
o
d
o
c
u
me
n
t
e
mb
e
d
d
i
n
g
,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
2
0
1
8
C
o
n
f
e
r
e
n
c
e
o
n
Em
p
i
r
i
c
a
l
M
e
t
h
o
d
s
i
n
N
a
t
u
r
a
l
L
a
n
g
u
a
g
e
Pr
o
c
e
ssi
n
g
,
EM
N
L
P
2
0
1
8
,
2
0
1
8
,
p
p
.
4
5
2
4
–
4
5
3
4
,
d
o
i
:
1
0
.
1
8
6
5
3
/
v
1
/
d
1
8
-
1
4
8
2
.
[
11
]
M
.
P
h
a
m,
S
.
A
l
s
e
,
C
.
A
.
K
n
o
b
l
o
c
k
,
a
n
d
P
.
S
z
e
k
e
l
y
,
S
e
m
a
n
t
i
c
l
a
b
e
l
i
n
g
:
A
d
o
m
a
i
n
-
i
n
d
e
p
e
n
d
e
n
t
a
p
p
r
o
a
c
h
,
v
o
l
.
9
9
8
1
.
C
h
a
m:
S
p
r
i
n
g
e
r
I
n
t
e
r
n
a
t
i
o
n
a
l
P
u
b
l
i
s
h
i
n
g
,
2
0
1
6
.
[1
2
]
J.
L
a
mb
e
r
t
,
A
.
-
L
.
Le
u
t
e
n
e
g
g
e
r
,
A
.
B
a
u
d
o
t
,
a
n
d
A
.
-
S
.
Ja
n
n
o
t
,
“
I
mp
r
o
v
i
n
g
p
a
t
i
e
n
t
c
l
u
st
e
r
i
n
g
b
y
i
n
c
o
r
p
o
r
a
t
i
n
g
s
t
r
u
c
t
u
r
e
d
l
a
b
e
l
r
e
l
a
t
i
o
n
s
h
i
p
s
i
n
s
i
mi
l
a
r
i
t
y
me
a
s
u
r
e
s,
”
m
e
d
R
x
i
v
,
2
0
2
3
.
[1
3
]
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
m
a
r
k
d
a
t
a
s
e
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
si
a
n
N
LP,
”
i
n
C
O
L
I
N
G
2
0
2
0
-
2
8
t
h
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
m
p
u
t
a
t
i
o
n
a
l
L
i
n
g
u
i
s
t
i
c
s
,
P
ro
c
e
e
d
i
n
g
s
o
f
t
h
e
C
o
n
f
e
re
n
c
e
,
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
-
mai
n
.
6
6
.
[1
4
]
J.
W
a
n
g
a
n
d
Y
.
D
o
n
g
,
“
M
e
a
su
r
e
me
n
t
o
f
t
e
x
t
s
i
mi
l
a
r
i
t
y
:
a
s
u
r
v
e
y
,
”
I
n
f
o
r
m
a
t
i
o
n
(
S
w
i
t
z
e
rl
a
n
d
)
,
v
o
l
.
1
1
,
n
o
.
9
,
p
p
.
1
–
1
7
,
2
0
2
0
,
d
o
i
:
1
0
.
3
3
9
0
/
i
n
f
o
1
1
0
9
0
4
2
1
.
[1
5
]
D
.
R
o
t
h
ma
n
,
“
Tr
a
n
sf
o
r
mers
f
o
r
n
a
t
u
r
a
l
l
a
n
g
u
a
g
e
p
r
o
c
e
ss
i
n
g
:
B
u
i
l
d
,
t
r
a
i
n
,
a
n
d
f
i
n
e
-
t
u
n
i
n
g
d
e
e
p
n
e
u
r
a
l
n
e
t
w
o
r
k
a
r
c
h
i
t
e
c
t
u
r
e
s
f
o
r
N
LP w
i
t
h
P
y
t
h
o
n
,
P
y
To
r
c
h
,
T
e
n
s
o
r
F
l
o
w
,
B
E
R
T,
a
n
d
G
P
T
-
3,
”
P
a
c
k
t
P
u
b
l
i
s
h
i
n
g
,
2
0
2
2
.
[1
6
]
B
.
W
i
l
i
e
e
t
a
l
.
,
“
I
n
d
o
N
LU
:
B
e
n
c
h
mar
k
a
n
d
r
e
s
o
u
r
c
e
s
f
o
r
e
v
a
l
u
a
t
i
n
g
i
n
d
o
n
e
si
a
n
n
a
t
u
r
a
l
l
a
n
g
u
a
g
e
u
n
d
e
r
s
t
a
n
d
i
n
g
,
”
2
0
2
0
.
[
1
7
]
U
.
S
h
a
f
i
q
u
e
a
n
d
H
.
Q
a
i
ser
,
“
A
c
o
mp
a
r
a
t
i
v
e
s
t
u
d
y
o
f
d
a
t
a
m
i
n
i
n
g
p
r
o
c
e
ss
m
o
d
e
l
s
(
K
D
D
,
C
R
I
S
P
-
D
M
a
n
d
S
E
M
M
A
)
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
I
n
n
o
v
a
t
i
o
n
a
n
d
S
c
i
e
n
t
i
f
i
c
R
e
se
a
rc
h
,
v
o
l
.
1
2
,
n
o
.
1
,
p
p
.
2
1
7
-
2
2
2
,
2
0
1
4
.
[1
8
]
J.
M
.
W
u
,
Y
.
B
e
l
i
n
k
o
v
,
H
.
S
a
j
j
a
d
,
N
.
D
u
r
r
a
n
i
,
F
.
D
a
l
v
i
,
a
n
d
J.
G
l
a
ss,
“
S
i
mi
l
a
r
i
t
y
a
n
a
l
y
si
s
o
f
c
o
n
t
e
x
t
u
a
l
w
o
r
d
r
e
p
r
e
s
e
n
t
a
t
i
o
n
mo
d
e
l
s
,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
A
n
n
u
a
l
M
e
e
t
i
n
g
o
f
t
h
e
A
sso
c
i
a
t
i
o
n
f
o
r
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
.
4
6
3
8
–
4
6
5
5
,
d
o
i
:
1
0
.
1
8
6
5
3
/
v
1
/
2
0
2
0
.
a
c
l
-
m
a
i
n
.
4
2
2
.
[1
9
]
C
.
B
i
e
m
a
n
n
a
n
d
M
.
R
i
e
d
l
,
“
Te
x
t
:
N
o
w
i
n
2
D
!
a
f
r
a
mew
o
r
k
f
o
r
l
e
x
i
c
a
l
e
x
p
a
n
s
i
o
n
w
i
t
h
c
o
n
t
e
x
t
u
a
l
si
mi
l
a
r
i
t
y
,
”
J
o
u
rn
a
l
o
f
L
a
n
g
u
a
g
e
Mo
d
e
l
l
i
n
g
,
v
o
l
.
1
,
n
o
.
1
,
2
0
1
3
,
d
o
i
:
1
0
.
1
5
3
9
8
/
j
l
m
.
v
1
i
1
.
6
0
.
[
20
]
C
.
S
h
o
r
t
e
n
,
T.
M
.
K
h
o
s
h
g
o
f
t
a
a
r
,
a
n
d
B
.
F
u
r
h
t
,
“
Te
x
t
d
a
t
a
a
u
g
me
n
t
a
t
i
o
n
f
o
r
d
e
e
p
l
e
a
r
n
i
n
g
,
”
J
o
u
r
n
a
l
o
f
B
i
g
D
a
t
a
,
v
o
l
.
8
,
n
o
.
1
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
8
6
/
s4
0
5
3
7
-
0
2
1
-
0
0
4
9
2
-
0.
[
21
]
D
.
H
a
b
a
,
“
D
a
t
a
a
u
g
m
e
n
t
a
t
i
o
n
w
i
t
h
p
y
t
h
o
n
e
n
h
a
n
c
e
d
e
e
p
l
e
a
r
n
i
n
g
a
c
c
u
r
a
c
y
w
i
t
h
d
a
t
a
a
u
g
me
n
t
a
t
i
o
n
me
t
h
o
d
s fo
r
i
mag
e
,
t
e
x
t
,
a
u
d
i
o
,
a
n
d
t
a
b
u
l
a
r
d
a
t
a
,
”
2
0
2
3
.
[
22
]
J.
W
e
i
a
n
d
K
.
Zo
u
,
“
ED
A
:
Ea
s
y
d
a
t
a
a
u
g
me
n
t
a
t
i
o
n
t
e
c
h
n
i
q
u
e
s
f
o
r
b
o
o
s
t
i
n
g
p
e
r
f
o
r
ma
n
c
e
o
n
t
e
x
t
c
l
a
ss
i
f
i
c
a
t
i
o
n
t
a
s
k
s,
”
i
n
E
MN
L
P
-
I
J
C
N
L
P
2
0
1
9
-
2
0
1
9
C
o
n
f
e
re
n
c
e
o
n
Em
p
i
r
i
c
a
l
Me
t
h
o
d
s
i
n
N
a
t
u
ra
l
L
a
n
g
u
a
g
e
Pr
o
c
e
ss
i
n
g
a
n
d
9
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
i
n
t
C
o
n
f
e
r
e
n
c
e
o
n
N
a
t
u
r
a
l
L
a
n
g
u
a
g
e
Pr
o
c
e
ss
i
n
g
,
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
C
o
n
f
e
r
e
n
c
e
,
2
0
1
9
,
p
p
.
6
3
8
2
–
6
3
8
8
,
d
o
i
:
1
0
.
1
8
6
5
3
/
v
1
/
d
1
9
-
1
6
7
0
.
[2
3
]
M
.
B
a
y
e
r
,
M
.
A
.
K
a
u
f
h
o
l
d
,
B
.
B
u
c
h
h
o
l
d
,
M
.
K
e
l
l
e
r
,
J.
D
a
l
l
m
e
y
e
r
,
a
n
d
C
.
R
e
u
t
e
r
,
“
D
a
t
a
a
u
g
m
e
n
t
a
t
i
o
n
i
n
n
a
t
u
r
a
l
l
a
n
g
u
a
g
e
p
r
o
c
e
ss
i
n
g
:
A
n
o
v
e
l
t
e
x
t
g
e
n
e
r
a
t
i
o
n
a
p
p
r
o
a
c
h
f
o
r
l
o
n
g
a
n
d
s
h
o
r
t
t
e
x
t
c
l
a
ssi
f
i
e
r
s,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
Ma
c
h
i
n
e
L
e
a
r
n
i
n
g
a
n
d
C
y
b
e
r
n
e
t
i
c
s
,
v
o
l
.
1
4
,
n
o
.
1
,
p
p
.
1
3
5
–
1
5
0
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
0
7
/
s1
3
0
4
2
-
0
2
2
-
0
1
5
5
3
-
3.
[2
4
]
M
.
H
a
s
n
a
i
n
e
t
a
l
.
,
“
Ev
a
l
u
a
t
i
n
g
t
r
u
s
t
p
r
e
d
i
c
t
i
o
n
a
n
d
c
o
n
f
u
si
o
n
mat
r
i
x
m
e
a
su
r
e
s
f
o
r
w
e
b
s
e
r
v
i
c
e
s
r
a
n
k
i
n
g
,
”
I
EE
E
A
c
c
e
s
s
,
v
o
l
.
8
,
p
p
.
9
0
8
4
7
–
9
0
8
6
1
,
2
0
2
0
.
[2
5
]
A
.
Lu
q
u
e
,
A
.
C
a
r
r
a
sc
o
,
A
.
M
a
r
t
í
n
,
a
n
d
A
.
d
e
l
a
s
H
e
r
a
s,
“
T
h
e
i
m
p
a
c
t
o
f
c
l
a
ss
i
m
b
a
l
a
n
c
e
i
n
c
l
a
ssi
f
i
c
a
t
i
o
n
p
e
r
f
o
r
ma
n
c
e
me
t
r
i
c
s
b
a
s
e
d
o
n
t
h
e
b
i
n
a
r
y
c
o
n
f
u
s
i
o
n
mat
r
i
x
,
”
P
a
t
t
e
r
n
R
e
c
o
g
n
i
t
i
o
n
,
v
o
l
.
9
1
,
p
p
.
2
1
6
–
2
3
1
,
2
0
1
9
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
a
t
c
o
g
.
2
0
1
9
.
0
2
.
0
2
3
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Ye
ss
y
As
r
i
sh
e
wa
s
b
o
r
n
o
n
Oc
to
b
e
r
1
3
,
1
9
7
6
,
i
n
P
a
d
a
n
g
c
it
y
.
S
h
e
is
t
h
e
first
o
f
th
re
e
c
h
il
d
re
n
,
b
o
r
n
to
fa
t
h
e
r
As
ri
M
a
’
a
rif
a
n
d
m
o
th
e
r
E
d
in
a
r.
S
t
a
rti
n
g
e
d
u
c
a
ti
o
n
i
n
th
e
S
1
De
p
a
rtme
n
t
o
f
I
n
fo
rm
a
ti
c
s
En
g
in
e
e
rin
g
,
t
h
e
n
c
o
n
t
in
u
in
g
h
e
r
M
a
ste
rs
e
d
u
c
a
ti
o
n
i
n
t
h
e
De
p
a
rtme
n
t
o
f
In
fo
rm
a
ti
o
n
S
y
ste
m
s,
a
n
d
c
u
rre
n
tl
y
,
sh
e
is
c
o
n
ti
n
u
i
n
g
h
is
d
o
c
t
o
ra
l
e
d
u
c
a
ti
o
n
in
th
e
De
p
a
rtme
n
t
o
f
In
fo
rm
a
ti
o
n
T
e
c
h
n
o
l
o
g
y
.
Cu
rre
n
t
ly
,
sh
e
se
rv
e
s
a
s
He
a
d
o
f
th
e
In
fo
rm
a
ti
o
n
S
y
ste
m
s
S
tu
d
y
P
r
o
g
ra
m
a
t
t
h
e
P
LN
In
stit
u
te
o
f
Tec
h
n
o
lo
g
y
(IT
P
LN)
in
Ja
k
a
rta
a
n
d
is
a
lso
a
lec
tu
re
r
a
t
se
v
e
ra
l
we
ll
-
k
n
o
w
n
u
n
iv
e
rsiti
e
s
in
Ja
k
a
rta.
S
h
e
is
a
lso
c
u
rre
n
tl
y
a
c
ti
v
e
a
s
th
e
Ch
ief
Ed
it
o
r
o
f
t
h
e
ITP
LN
KILAT
Jo
u
rn
a
l,
is
a
c
ti
v
e
a
s
a
n
i
n
stru
c
to
r
a
t
se
v
e
ra
l
train
in
g
in
stit
u
ti
o
n
s
,
a
n
d
is
a
lso
c
u
rre
n
tl
y
a
n
a
c
ti
v
e
BNSP
a
ss
e
ss
o
r.
Wi
th
p
e
rs
e
v
e
ra
n
c
e
a
n
d
h
i
g
h
m
o
ti
v
a
ti
o
n
to
c
o
n
ti
n
u
e
lea
rn
i
n
g
a
n
d
tr
y
i
n
g
,
u
n
ti
l
n
o
w
th
e
a
u
th
o
r
h
a
s
p
ro
d
u
c
e
d
m
a
n
y
writi
n
g
s
i
n
se
v
e
ra
l
S
INTA
-
a
c
c
re
d
it
e
d
n
a
ti
o
n
a
l
jo
u
rn
a
ls
a
n
d
in
d
e
x
e
d
i
n
tern
a
ti
o
n
a
l
jo
u
r
n
a
ls
o
n
in
tellec
tu
a
l
p
r
o
p
e
rt
y
rig
h
ts
(Ha
KI),
c
a
rried
o
u
t
m
a
n
y
c
o
m
m
u
n
it
y
se
rv
ice
a
c
ti
v
it
ies
(P
k
M
),
a
n
d
re
c
e
iv
e
d
se
v
e
ra
l
a
wa
rd
s.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
y
e
ss
y
a
sri@it
p
l
n
.
a
c
.
id
.
Evaluation Warning : The document was created with Spire.PDF for Python.