I
n
d
on
e
s
ian
Jou
r
n
al
o
f
E
lec
t
r
ica
l
E
n
gin
e
e
r
in
g
a
n
d
Com
p
u
t
e
r
S
c
ience
Vo
l
.
3
7
,
N
o
.
1
,
J
a
n
ua
r
y
20
2
5
,
pp.
3
1
5
~
324
I
S
S
N:
2
502
-
4
7
52
,
DO
I
:
10
.
11591/i
j
e
e
cs
.v
3
7
.
i
1
.
pp
3
1
5
-
324
315
Jou
r
n
al
h
o
m
e
page
:
ht
tp:
//
ij
e
e
cs
.
iaes
c
or
e
.
c
om
In
t
e
gr
at
in
g E
L
E
C
T
R
A
an
d
B
E
R
T
m
o
d
e
ls
in
t
r
an
sf
or
m
e
r
-
b
ase
d
m
e
n
t
al
h
e
al
t
h
c
ar
e
c
h
at
b
ot
Ju
n
t
a
Z
e
n
iar
j
a
1
,
2
,
Cin
an
t
ya
P
ar
a
m
it
a
1
,
2
,
E
gia
R
os
i
S
u
b
h
iyak
t
o
1
,
2
,
S
in
d
h
u
Rak
as
iwi
1
,
2
,
Gu
r
u
h
F
aj
ar
S
h
id
ik
1
,
2
,
P
u
l
u
n
g
Nur
t
an
t
io
And
o
n
o
1
,
2
,
Anam
ar
ij
a
Ju
r
c
e
v
S
avice
vic
3
1
F
a
c
ul
t
y
of
C
omput
e
r
S
c
i
e
nc
e
, U
ni
ve
r
s
it
a
s
D
ia
n N
us
w
a
nt
o
r
o
, S
e
ma
r
a
ng, I
nd
o
n
e
s
ia
2
R
e
s
e
a
r
c
h C
e
nt
e
r
f
o
r
I
nt
e
ll
ig
e
nt
D
is
tr
ib
ut
e
d S
ur
ve
il
la
n
c
e
a
nd S
e
c
ur
it
y
(
I
D
S
S
)
, U
ni
ve
r
s
it
a
s
D
ia
n N
us
w
a
nt
o
r
o
, S
e
ma
r
a
ng, I
nd
o
n
e
s
ia
3
D
e
pa
r
tm
e
nt
of
H
e
a
lt
h S
tu
di
e
s
, P
ubl
ic
H
e
a
lt
h I
ns
ti
tu
t
e
of
S
pl
it
-
D
a
lm
a
ti
a
, S
c
h
oo
l
of
M
e
di
c
in
e
, U
ni
ve
r
s
it
y
of
S
pl
it
,
S
pl
it
,
C
r
o
a
ti
a
Ar
t
ic
l
e
I
n
f
o
AB
S
T
RA
CT
A
r
ti
c
le
h
is
tor
y
:
R
e
c
e
i
ve
d
M
a
y
2
,
202
4
R
e
vi
s
e
d
S
e
p
9
,
202
4
A
c
c
e
pt
e
d
S
e
p
29
,
202
4
O
v
e
r
t
h
e
l
as
t
d
ec
a
d
e,
t
h
e
s
u
r
g
e
i
n
men
t
al
h
e
al
t
h
d
i
s
o
r
d
e
rs
h
as
n
ece
s
s
i
t
at
e
d
i
n
n
o
v
at
i
v
e
s
u
p
p
o
rt
me
t
h
o
d
s
,
n
o
t
ab
l
y
art
i
fi
ci
al
i
n
t
el
l
i
g
en
t
(
AI
)
c
h
at
b
o
t
s
.
T
h
e
s
e
c
h
at
b
o
t
s
p
ro
v
i
d
e
p
ro
m
p
t
,
t
ai
l
o
r
ed
co
n
v
e
rs
at
i
o
n
s
,
b
ec
o
mi
n
g
c
r
u
c
i
a
l
i
n
me
n
t
al
h
e
al
t
h
s
u
p
p
o
rt
.
T
h
i
s
art
i
cl
e
d
e
l
v
e
s
i
n
t
o
t
h
e
u
s
e
o
f
s
o
p
h
i
s
t
i
c
at
ed
mo
d
e
l
s
l
i
k
e
co
n
v
o
l
u
t
i
o
n
a
l
n
e
u
ra
l
n
e
t
w
o
rk
(
CN
N
)
,
l
o
n
g
-
s
h
o
rt
t
e
r
m
memo
r
y
(
L
ST
M
)
,
e
ffi
ci
e
n
t
l
y
l
e
arn
i
n
g
an
e
n
co
d
e
r
t
h
at
cl
as
s
i
fi
e
s
t
o
k
en
r
e
p
l
a
cemen
t
s
acc
u
rat
e
l
y
(
E
L
E
CT
RA
)
,
an
d
b
i
d
i
r
e
c
t
i
o
n
a
l
e
n
co
d
e
r
re
p
re
s
e
n
t
at
i
o
n
o
f
t
ran
s
fo
rme
rs
(
B
E
RT
)
i
n
d
ev
e
l
o
p
i
n
g
e
ff
ec
t
i
v
e
men
t
al
h
e
a
l
t
h
ch
at
b
o
t
s
.
D
e
s
p
i
t
e
t
h
e
i
r
i
m
p
o
rt
an
ce
f
o
r
em
o
t
i
o
n
a
l
as
s
i
s
t
an
ce
,
t
h
e
s
e
ch
at
b
o
t
s
s
t
ru
g
g
l
e
w
i
t
h
p
re
ci
s
e
an
d
r
el
e
v
an
t
r
e
s
p
o
n
s
e
s
t
o
c
o
m
p
l
e
x
men
t
al
h
e
al
t
h
i
s
s
u
e
s
.
BE
RT
,
w
h
i
l
e
s
t
ro
n
g
i
n
co
n
t
e
x
t
u
al
u
n
d
e
rs
t
an
d
i
n
g
,
l
a
ck
s
i
n
r
e
s
p
o
n
s
e
g
en
e
rat
i
o
n
.
Co
n
v
e
rs
e
l
y
,
E
L
E
CT
RA
s
h
o
w
s
p
ro
mi
s
e
i
n
t
e
x
t
c
r
e
at
i
o
n
b
u
t
i
s
n
o
t
f
u
l
l
y
e
x
p
l
o
i
t
e
d
i
n
me
n
t
al
h
e
al
t
h
co
n
t
e
x
t
s
.
T
h
e
art
i
c
l
e
i
n
v
e
s
t
i
g
at
e
s
me
r
g
i
n
g
E
L
E
CT
RA
a
n
d
B
E
RT
t
o
i
m
p
ro
v
e
ch
at
b
o
t
e
ffi
ci
e
n
cy
i
n
men
t
al
h
e
a
l
t
h
s
i
t
u
at
i
o
n
s
.
B
y
l
e
v
e
ra
g
i
n
g
a
n
e
x
t
e
n
s
i
v
e
me
n
t
al
h
e
al
t
h
d
i
al
o
g
u
e
d
at
as
e
t
,
t
h
i
s
i
n
t
eg
rat
i
o
n
s
u
b
s
t
an
t
i
al
l
y
en
h
a
n
ce
d
ch
at
b
o
t
p
re
ci
s
i
o
n
,
s
u
rp
as
s
i
n
g
9
9
%
acc
u
ra
cy
i
n
me
n
t
al
h
e
al
t
h
r
e
s
p
o
n
s
e
s
.
T
h
i
s
d
ev
e
l
o
p
men
t
i
s
a
s
i
g
n
i
f
i
c
an
t
s
t
ri
d
e
i
n
ad
v
a
n
c
i
n
g
A
I
c
h
at
b
o
t
i
n
t
e
ra
c
t
i
o
n
s
a
n
d
t
h
e
i
r
c
o
n
t
ri
b
u
t
i
o
n
t
o
me
n
t
al
h
e
al
t
h
s
u
p
p
o
rt
.
K
e
y
w
o
r
d
s
:
B
E
R
T
C
ha
t
b
ot
E
L
E
C
T
R
A
M
e
n
t
a
l
h
e
a
l
t
h
T
r
a
n
s
f
o
r
m
e
r
Th
i
s
i
s
a
n
o
p
en
a
c
ces
s
a
r
t
i
c
l
e
u
n
d
e
r
t
h
e
CC
B
Y
-
SA
l
i
cen
s
e.
C
or
r
e
s
pon
din
g
A
u
th
or
:
C
i
n
a
n
t
y
a
P
a
r
a
m
i
t
a
F
a
c
u
l
t
y
o
f
C
o
m
put
e
r
S
c
i
e
n
c
e
,
Uni
ve
r
s
i
t
a
s
D
i
a
n
Nu
s
wa
n
t
or
o
S
e
m
a
r
a
n
g,
I
n
do
n
e
s
i
a
E
m
a
i
l
:
c
i
na
n
t
y
a
.
pa
r
a
m
i
t
a
@
ds
n
.
d
i
n
u
s
.
a
c
.
i
d
1.
I
NT
RODU
C
T
I
ON
I
n
r
e
c
e
n
t
y
e
a
r
s
,
t
h
e
c
r
i
t
i
c
a
l
im
po
r
t
a
n
c
e
o
f
m
e
n
ta
l
he
a
l
t
h
c
a
r
e
h
a
s
be
e
n
i
n
c
r
e
a
s
i
ng
l
y
r
e
c
o
gni
z
e
d,
r
e
f
l
e
c
t
i
n
g
b
o
t
h
i
t
s
pr
o
f
o
un
d
i
m
pa
c
t
o
n
i
n
d
i
v
i
dua
ls
a
n
d
i
t
s
s
u
b
s
t
a
n
t
i
a
l
c
o
s
t
to
t
h
e
gl
o
b
a
l
e
c
o
n
o
m
y
.
S
t
udi
e
s
r
e
v
e
a
l
t
h
a
t
a
s
i
g
ni
f
i
c
a
n
t
n
u
m
b
e
r
o
f
i
n
d
i
v
i
dua
l
s
wh
o
s
uc
c
u
m
b
e
d
to
s
ui
c
i
de
h
a
d
i
n
t
e
r
a
c
t
e
d
wi
t
h
he
a
l
t
h
c
a
r
e
pr
o
vi
de
r
s
s
h
o
r
t
l
y
b
e
f
o
r
e
t
h
e
i
r
de
a
t
h
,
un
de
r
s
c
o
r
i
n
g
t
h
e
c
h
a
ll
e
n
ge
s
i
n
de
t
e
c
t
i
n
g
a
n
d
a
ddr
e
s
s
i
n
g
s
u
i
c
i
da
l
i
de
a
t
i
o
n
,
pa
r
t
i
c
u
l
a
r
l
y
i
n
t
h
o
s
e
wh
o
vi
e
w
s
uc
h
d
i
s
c
us
s
i
o
n
s
a
s
t
a
b
o
o
[
1]
.
Gl
o
b
a
ll
y
,
m
e
n
t
a
l
i
ll
ne
s
s
e
s
,
i
n
c
l
ud
i
ng
c
o
m
m
o
n
c
o
n
d
i
t
i
o
ns
l
i
ke
a
nxi
e
t
y
a
n
d
de
pr
e
s
s
i
o
n
a
s
we
l
l
a
s
m
o
r
e
s
e
v
e
r
e
d
i
s
o
r
v
de
r
s
,
a
f
f
e
c
t
c
l
o
s
e
to
a
bi
ll
i
o
n
pe
o
pl
e
,
de
gr
a
d
i
n
g
t
h
e
i
r
qua
l
i
t
y
o
f
l
if
e
a
n
d
s
o
c
i
e
t
a
l
c
o
n
t
r
i
b
ut
i
o
n
.
T
hi
s
r
e
s
u
l
t
s
i
n
a
n
e
s
t
i
m
a
t
e
d
e
c
o
n
o
m
i
c
b
ur
de
n
o
f
USD
2.
5
t
r
i
l
li
o
n
a
nn
ua
ll
y
,
pr
o
j
e
c
t
e
d
to
r
i
s
e
to
USD
6
tr
i
ll
i
o
n
by
2030
[
2]
.
De
s
pi
t
e
t
h
e
gr
o
wi
n
g
pr
e
v
a
l
e
n
c
e
o
f
ps
y
c
h
o
l
o
g
i
c
a
l
d
i
s
t
r
e
s
s
,
a
v
a
s
t
m
a
j
o
r
i
t
y
o
f
t
h
o
s
e
s
u
f
f
e
r
i
n
g
do
n
o
t
r
e
c
e
i
v
e
a
d
e
qua
t
e
tr
e
a
t
m
e
n
t
due
to
l
im
i
t
e
d
a
c
c
e
s
s
to
m
e
n
t
a
l
h
e
a
l
t
h
c
o
un
s
e
li
ng.
He
a
l
t
h
c
o
a
c
hi
ng
h
a
s
e
m
e
r
ge
d
a
s
a
pr
o
m
i
s
i
ng
m
e
t
h
o
d
f
o
r
s
uppor
t
i
n
g
t
h
o
s
e
w
i
t
h
c
h
r
o
ni
c
c
o
n
d
i
t
i
o
ns
,
b
ut
t
h
e
s
c
a
r
c
i
t
y
o
f
pr
a
c
t
i
t
i
o
n
e
r
s
l
im
i
t
s
i
t
s
r
e
a
c
h
.
M
e
a
n
w
hil
e
,
t
h
e
e
x
pa
n
d
i
ng
v
o
l
u
m
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2
5
0
2
-
4
7
52
In
do
n
e
s
i
a
n
J
E
l
e
c
E
n
g
&
C
o
m
p
S
c
i
,
Vo
l
.
3
7
,
N
o
.
1
,
J
a
n
ua
r
y
20
2
5
:
3
1
5
-
324
316
o
f
c
l
i
n
i
c
a
l
i
nf
o
r
m
a
t
i
o
n
i
n
h
e
a
l
t
h
c
a
r
e
pr
e
s
e
n
t
s
n
e
w
o
pp
o
r
t
uni
t
i
e
s
f
o
r
l
e
ve
r
a
g
i
n
g
t
e
c
hn
o
l
o
g
y
a
n
d
c
o
m
put
e
r
s
c
i
e
n
c
e
t
o
i
m
pr
o
v
e
m
e
n
t
a
l
h
e
a
l
t
h
c
a
r
e
a
c
c
e
s
s
i
b
il
i
t
y
a
n
d
e
f
f
e
c
t
i
ve
n
e
s
s
.
A
r
t
i
f
i
c
i
a
l
i
n
t
e
ll
i
ge
n
c
e
(
A
I
)
r
e
pr
e
s
e
n
t
s
a
gr
o
un
db
r
e
a
k
i
n
g
s
hi
f
t
i
n
he
a
l
t
h
c
a
r
e
,
b
r
o
ugh
t
a
b
o
ut
by
t
h
e
c
o
n
v
e
r
ge
n
c
e
o
f
e
x
p
a
n
d
i
n
g
he
a
l
t
h
da
t
a
a
n
d
a
d
v
a
n
c
e
d
a
n
a
ly
t
i
c
a
l
t
e
c
hni
que
s
.
T
hi
s
t
e
c
hn
o
l
o
g
y
i
s
p
i
v
o
t
a
l
in
mi
r
r
o
r
i
n
g
h
u
m
a
n
c
o
gni
t
i
v
e
pr
o
c
e
s
s
e
s
a
n
d
m
a
r
ks
a
s
i
g
nif
i
c
a
n
t
t
r
a
n
s
f
o
r
m
a
t
i
o
n
i
n
h
e
a
l
t
h
c
a
r
e
dy
na
mi
c
s
.
F
ur
t
h
e
r
m
o
r
e
,
A
I
o
f
f
e
r
s
uni
qu
e
s
o
l
ut
i
o
n
s
to
o
v
e
r
c
om
e
s
o
c
i
a
l
s
t
i
g
m
a
,
a
n
o
t
a
bl
e
b
a
r
r
i
e
r
i
n
a
c
c
e
s
s
i
ng
he
a
l
t
h
c
a
r
e
.
Am
o
n
g
i
t
s
d
i
ve
r
s
e
a
pp
l
i
c
a
t
i
o
ns
,
A
I
-
dr
i
v
e
n
he
a
l
t
h
c
h
a
t
b
o
t
s
h
a
v
e
b
e
e
n
i
ns
t
r
u
m
e
n
t
a
l
i
n
pr
o
m
o
t
i
n
g
phy
s
i
c
a
l
a
c
t
i
vi
t
y
,
h
e
a
l
t
hy
e
a
t
i
n
g
h
a
bi
t
s
,
a
n
d
b
r
o
a
de
ni
ng
pa
t
i
e
n
t
a
c
c
e
s
s
t
o
e
s
s
e
n
t
i
a
l
h
e
a
l
t
h
i
nf
o
r
m
a
t
i
o
n
.
T
h
e
s
e
c
h
a
t
b
o
t
s
,
s
o
phi
s
t
i
c
a
t
e
d
c
o
m
put
e
r
p
r
o
gr
a
m
s
de
s
i
g
ne
d
f
o
r
s
i
mu
l
a
t
i
n
g
i
n
t
e
r
a
c
t
i
v
e
h
u
m
a
n
c
o
n
v
e
r
s
a
t
i
o
n
vi
a
t
h
e
i
n
t
e
r
n
e
t
,
a
r
e
e
m
e
r
g
i
ng
a
s
va
l
ua
bl
e
too
l
s
i
n
e
nh
a
n
c
i
ng
pa
t
i
e
n
t
e
n
ga
ge
m
e
n
t
a
n
d
h
e
a
l
t
h
c
a
r
e
de
l
i
ve
r
y
.
R
e
c
e
n
t
r
e
s
e
a
r
c
h
i
n
AI
pr
i
m
a
r
i
ly
f
o
c
u
s
e
s
o
n
de
v
e
l
o
p
i
n
g
m
o
de
l
s
a
n
d
a
l
go
r
i
t
hm
s
,
s
pe
c
if
i
c
a
ll
y
a
r
t
i
f
i
c
i
a
l
n
e
ur
a
l
n
e
t
wo
r
ks
[
3]
,
de
e
p
n
e
ur
a
l
n
e
t
wo
r
k
[
4
]
,
c
o
n
v
o
l
ut
i
o
n
a
l
n
e
ur
a
l
n
e
t
wo
r
k
(
C
NN
)
[
5
]
,
l
o
n
g
s
ho
r
t
t
e
r
m
m
e
m
o
r
y
(
L
S
T
M
)
[
6]
a
n
d
ot
h
e
r
de
e
p
l
e
a
r
ni
ng
a
l
g
o
r
i
t
hm
s
[
7]
,
[
8
]
.
A
I
r
e
f
e
r
s
to
t
h
e
c
a
pa
bil
i
t
y
o
f
m
a
c
hi
ne
s
to
pe
r
f
o
r
m
t
a
s
ks
t
h
a
t
t
y
p
i
c
a
l
ly
r
e
qu
i
r
e
h
u
m
a
n
i
n
t
e
l
li
ge
n
c
e
,
s
uc
h
a
s
v
o
i
c
e
r
e
c
o
gni
t
i
o
n
,
n
a
t
ur
a
l
l
a
n
gua
g
e
un
de
r
s
t
a
n
d
i
ng,
a
n
d
de
c
i
s
i
o
n
-
m
a
k
i
ng
[
9]
.
R
e
b
e
l
o
e
t
al.
[
10]
h
a
v
e
s
h
o
wn
t
h
e
e
f
f
e
c
t
s
o
f
A
I
o
n
m
e
n
t
a
l
h
e
a
l
t
h
t
a
s
ks
.
T
h
e
y
i
nv
o
l
ve
s
e
v
e
r
a
l
m
e
n
t
a
l
h
e
a
l
t
h
c
a
r
e
wo
r
ke
r
s
s
uc
h
a
s
ps
y
c
h
o
l
o
gi
s
t
s
,
t
h
e
r
a
p
i
s
t
s
,
a
n
d
ps
y
c
hi
a
t
r
i
s
t
s
,
i
nc
l
ud
i
ng
t
h
e
t
a
s
ks
t
h
e
y
c
a
r
r
y
o
ut
s
uc
h
a
s
a
s
s
e
s
s
m
e
n
t
,
t
h
e
r
a
p
y
,
m
e
d
i
c
a
l
pr
e
s
c
r
i
pt
i
o
ns
,
do
c
um
e
n
t
a
ti
o
n
,
a
n
d
m
o
ni
t
o
r
i
n
g.
T
h
e
a
bi
li
t
y
t
o
c
o
m
pr
e
h
e
n
d
n
a
t
ur
a
l
l
a
ngua
ge
i
s
n
o
t
l
i
mi
t
e
d
t
o
h
u
m
a
n
c
o
m
m
u
ni
c
a
t
i
o
n
b
ut
e
x
t
e
n
d
s
to
c
o
m
m
u
ni
c
a
t
i
o
n
b
e
t
we
e
n
h
u
m
a
n
s
a
n
d
m
a
c
hi
ne
s
,
b
r
i
dg
e
d
by
n
a
t
ur
a
l
l
a
n
gua
ge
pr
o
c
e
s
s
i
ng
(
NL
P
)
.
T
h
e
t
e
c
h
n
o
l
o
g
y
e
m
p
l
o
y
e
d
by
NL
P
ha
s
s
i
g
ni
f
i
c
a
n
t
l
y
a
dv
a
nc
e
d
i
n
r
e
c
e
n
t
y
e
a
r
s
,
e
nh
a
n
c
i
ng
b
o
t
h
v
o
i
c
e
a
n
d
t
e
x
t
-
b
a
s
e
d
i
n
t
e
r
a
c
t
i
o
n
s
t
o
a
s
s
i
s
t
h
u
m
a
ns
.
S
o
m
e
r
e
s
e
a
r
c
h
e
r
s
a
r
e
de
d
i
c
a
t
e
d
to
f
i
nd
i
ng
m
o
de
l
s
w
i
t
h
hi
g
h
e
r
a
c
c
ur
a
c
y
,
c
a
p
i
t
a
l
i
z
i
ng
o
n
b
r
e
a
kt
h
r
o
ugh
s
i
n
ne
ur
a
l
n
e
t
wo
r
ks
th
a
t
b
r
i
n
g
n
o
v
e
l
i
de
a
s
t
o
t
h
e
f
i
e
l
d
s
o
f
A
I
a
n
d
NL
P
,
r
e
s
u
l
t
i
n
g
i
n
t
h
e
e
m
e
r
ge
n
c
e
o
f
v
a
r
i
o
us
t
e
c
hni
que
s
a
n
d
m
e
t
h
o
ds
.
F
u
e
t
a
l.
[
11]
p
r
e
s
e
n
t
e
d
r
e
pr
e
s
e
n
t
a
t
i
v
e
wor
ks
a
n
d
c
a
t
e
g
o
r
i
z
e
d
e
xi
s
t
i
n
g
d
i
a
l
o
gue
m
o
de
l
s
i
n
t
o
t
h
r
e
e
t
y
pe
s
,
a
n
a
l
y
s
i
ng
tr
e
n
ds
i
n
o
pe
n
-
do
m
a
i
n
d
i
a
l
o
gue
de
v
e
l
o
p
m
e
n
t
a
n
d
s
u
m
m
a
r
i
z
i
ng
o
bj
e
c
t
i
v
e
s
i
n
t
w
o
a
s
pe
c
t
s
:
i
nf
o
r
m
a
t
i
v
e
a
n
d
c
o
n
t
r
o
l
l
e
d.
M
a
h
e
t
al.
[
12]
u
t
i
li
z
e
d
t
e
x
t
c
o
n
t
e
n
t
a
s
pa
r
t
o
f
NL
P
a
n
d
A
I
to
a
s
s
e
s
s
t
h
e
c
onn
e
c
t
i
o
n
o
f
t
h
e
i
n
t
e
r
n
e
t
o
f
t
hi
n
gs
(
I
oT
s
)
wi
t
h
h
u
m
a
n
n
e
e
ds
.
C
ha
t
b
ot
s
,
a
l
s
o
kn
o
wn
a
s
a
r
t
i
f
i
c
i
a
l
c
o
nv
e
r
s
a
t
i
o
n
e
n
t
i
t
i
e
s
,
c
a
n
b
e
de
v
e
l
o
p
e
d
us
i
n
g
m
a
c
hi
ne
l
e
a
r
ni
ng
a
n
d
pa
tt
e
r
n
m
a
t
c
hi
ng,
w
i
t
h
t
h
r
e
e
t
y
pe
s
b
a
s
e
d
o
n
r
e
s
po
n
s
e
ge
n
e
r
a
t
i
o
n
m
e
t
h
o
ds
:
r
ul
e
-
b
a
s
e
d,
r
e
t
r
i
e
v
a
l
-
b
a
s
e
d,
a
n
d
ge
n
e
r
a
t
i
v
e
[
13]
.
P
a
n
de
y
a
n
d
S
ha
r
m
a
[
14]
c
om
pa
r
e
d
r
e
t
r
i
e
v
a
l
-
b
a
s
e
d
a
n
d
ge
n
e
r
a
t
i
v
e
-
b
a
s
e
d
c
h
a
t
b
o
t
s
f
o
r
m
e
n
t
a
l
h
e
a
l
t
h
,
f
i
nd
i
ng
t
h
a
t
ge
n
e
r
a
t
i
v
e
m
o
de
l
s
w
i
t
h
e
n
c
o
de
r
-
de
c
o
de
r
a
r
c
hi
t
e
c
t
ur
e
a
c
hi
e
v
e
d
t
h
e
hi
g
he
s
t
a
c
c
ur
a
c
y
a
t
94.
45%
.
S
i
m
a
r
m
a
t
a
e
t
al.
[
15
]
e
x
p
l
o
r
e
d
v
a
r
i
o
us
t
r
a
n
s
f
o
r
m
e
r
m
o
de
l
s
f
o
r
i
n
t
e
n
t
c
l
a
s
s
if
i
c
a
t
i
o
n
,
n
ot
i
n
g
bi
d
i
r
e
c
t
i
o
n
a
l
e
n
c
ode
r
r
e
p
r
e
s
e
n
ta
t
i
o
n
o
f
tr
a
n
s
f
or
m
e
r
s
(
B
E
R
T
)
wi
t
h
L
S
T
M
'
s
9
4
.
4
7
%
a
c
c
u
r
a
c
y
.
Y
u
e
t
a
l
.
[
1
6
]
de
v
e
l
o
p
e
d
a
B
E
R
T
-
b
a
s
e
d
c
h
a
t
b
ot
f
or
f
i
n
a
n
c
i
a
l
c
us
to
m
e
r
s
e
r
v
i
c
e
c
a
pa
bl
e
o
f
m
a
n
a
gi
n
g
c
o
m
p
l
e
x
qu
e
r
i
e
s
a
nd
p
e
r
f
or
m
i
n
g
a
u
to
m
a
t
i
c
s
pe
l
l
i
n
g
c
or
r
e
c
t
i
o
n
.
F
a
t
i
m
a
e
t
a
l
.
[
1
7
]
us
e
d
a
C
OV
I
D
-
19
r
um
or
s
da
tas
e
t
f
or
v
e
r
a
c
i
t
y
,
s
tan
c
e
,
a
n
d
s
e
n
t
i
m
e
n
t
a
n
a
l
y
s
i
s
,
a
c
hi
e
vi
n
g
h
i
gh
a
c
c
u
r
a
c
y
w
i
t
h
L
S
T
M
+
C
NN
m
o
d
e
l
s
.
T
h
e
s
tudy
a
l
s
o
hi
ghl
i
gh
t
s
th
e
f
us
i
o
n
o
f
e
f
f
i
c
i
e
n
t
l
y
l
e
a
r
n
i
n
g
a
n
e
n
c
ode
r
t
h
a
t
c
l
a
s
s
i
f
i
e
s
tok
e
n
r
e
pl
a
c
e
m
e
n
t
s
a
c
c
u
r
a
tel
y
(
E
L
E
C
T
R
A
)
a
n
d
L
S
T
M
m
ode
l
s
f
or
q
ue
s
t
i
o
n
c
l
a
s
s
i
f
i
c
a
t
i
o
n
,
c
o
m
bi
ni
n
g
E
L
E
C
T
R
A
'
s
l
a
n
gu
a
ge
un
de
r
s
tan
di
n
g
w
i
t
h
L
S
T
M
'
s
s
e
qu
e
n
t
i
a
l
l
e
a
r
ni
n
g
a
bi
l
i
t
i
e
s
[
1
8
]
,
[
19
]
.
C
e
n
tr
a
l
to
t
hi
s
r
e
s
e
a
r
c
h
i
s
a
c
o
m
pr
e
h
e
n
s
i
v
e
da
tas
e
t
f
or
tr
a
i
ni
n
g
c
h
a
t
b
ots
i
n
m
e
n
t
a
l
h
e
a
l
t
h
c
o
n
v
e
r
s
a
t
i
o
n
s
,
e
m
p
l
o
y
i
n
g
A
I
m
o
d
e
l
s
l
i
ke
B
E
R
T
,
C
NN
,
L
S
T
M
,
a
n
d
E
L
E
C
T
R
A
to
e
n
ha
n
c
e
th
e
c
h
a
t
b
ot’
s
a
bi
l
i
t
y
to
un
de
r
s
tan
d
a
n
d
r
e
s
p
o
n
d
to
h
um
a
n
e
m
ot
i
o
n
s
a
n
d
l
i
n
gui
s
t
i
c
s
ub
t
l
e
t
i
e
s
[
20
]
,
[
21
]
.
A
c
r
i
t
i
c
a
l
c
h
a
ll
e
n
g
e
f
o
r
m
e
n
t
a
l
h
e
a
l
t
h
c
a
r
e
c
h
a
t
b
o
t
s
i
s
a
c
hi
e
vi
ng
hi
g
h
a
c
c
ur
a
c
y
i
n
un
d
e
r
s
t
a
n
d
i
ng
a
n
d
r
e
s
po
n
d
i
n
g
t
o
us
e
r
i
n
put
s
.
De
s
p
i
t
e
a
dv
a
n
c
e
m
e
n
t
s
w
i
t
h
m
o
de
l
s
l
i
ke
E
L
E
C
T
R
A
a
n
d
B
E
R
T
,
t
h
e
i
r
i
n
t
e
gr
a
t
i
o
n
i
n
t
o
t
h
e
T
r
a
n
s
f
o
r
m
e
r
f
r
a
m
e
wo
r
k
f
o
r
n
ua
n
c
e
d
m
e
nt
a
l
h
e
a
l
t
h
d
i
a
l
o
gue
s
r
e
m
a
i
ns
s
u
b
o
pt
i
m
a
l
.
T
hi
s
a
c
c
ur
a
c
y
i
s
vi
t
a
l
f
o
r
e
f
f
e
c
t
i
v
e
s
uppo
r
t
b
ut
f
a
c
e
s
i
s
s
ue
s
s
u
c
h
a
s
da
t
a
p
r
i
v
a
c
y
,
d
i
ve
r
s
e
da
t
a
s
e
t
s
,
a
n
d
c
ul
t
ur
a
l
ly
s
e
n
s
i
t
i
ve
r
e
s
po
n
s
e
s
.
E
v
a
l
ua
t
e
d
us
i
n
g
m
e
t
r
i
c
s
l
i
ke
a
c
c
ur
a
c
y
,
pr
e
c
i
s
i
o
n
,
r
e
c
a
l
l
,
a
n
d
F
1
s
c
o
r
e
,
tr
a
n
s
f
o
r
m
e
r
m
o
de
l
s
o
u
t
pe
r
f
o
r
m
ot
h
e
r
s
b
ut
s
t
i
l
l
r
e
qu
i
r
e
e
t
hi
c
a
l
c
o
ns
i
d
e
r
a
t
i
o
ns
a
n
d
r
e
s
po
ns
i
bl
e
r
e
gu
l
a
t
i
o
n
.
B
a
l
c
o
m
b
e
[
22]
hi
g
hli
g
h
t
s
A
I
c
h
a
t
b
ot
s
'
de
c
a
de
-
l
o
n
g
r
o
l
e
i
n
i
m
pr
o
v
i
ng
m
e
n
t
a
l
h
e
a
l
t
h
c
a
r
e
,
t
h
o
ugh
t
h
e
y
l
a
c
k
pe
r
s
o
n
a
l
e
m
pa
t
hy
.
Hu
m
a
n
-
A
I
c
o
l
l
a
b
o
r
a
t
i
o
n
c
a
n
i
n
t
e
gr
a
t
e
h
u
m
a
n
va
l
ue
s
i
n
t
o
A
I
,
i
m
pa
c
t
i
n
g
va
r
i
o
us
s
e
c
t
or
s
l
i
ke
de
pr
e
s
s
i
o
n
,
a
nxi
e
t
y
,
a
n
d
A
DH
D
s
uppo
r
t.
Ho
we
v
e
r
,
f
ur
t
h
e
r
r
e
s
e
a
r
c
h
i
s
n
e
e
de
d
f
o
r
c
r
i
s
i
s
s
uppo
r
t
qua
l
i
t
y
.
C
a
l
da
r
i
n
i
e
t
al.
[
23]
r
e
vi
e
w
s
t
h
e
c
o
m
p
l
e
xi
t
i
e
s
o
f
e
v
a
l
ua
t
i
n
g
c
h
a
t
b
o
t
s
due
to
di
ve
r
s
e
c
o
nv
e
r
s
a
t
i
o
n
a
l
go
a
l
s
a
n
d
do
m
a
i
n
-
s
pe
c
i
f
i
c
da
t
a
,
e
m
p
h
a
s
i
z
i
ng
t
h
e
ga
p
b
e
t
we
e
n
i
n
dus
t
r
y
pr
a
c
t
i
c
e
s
a
n
d
a
c
a
de
m
i
c
a
d
v
a
nc
e
m
e
n
t
s
a
n
d
t
h
e
n
e
e
d
f
o
r
a
c
o
m
m
o
n
e
v
a
l
u
a
t
i
o
n
f
r
a
m
e
wo
r
k
a
n
d
b
e
tt
e
r
da
t
a
s
e
t
s
.
R
a
h
a
l
i
a
n
d
A
k
hl
o
u
f
i
[
24]
e
x
p
l
o
r
e
s
T
r
a
n
s
f
o
r
m
e
r
-
b
a
s
e
d
m
o
de
l
s
i
n
NL
P
,
n
o
t
i
n
g
t
h
e
i
r
s
i
g
nif
i
c
a
n
t
i
m
pa
c
t
a
n
d
s
ugge
s
t
i
n
g
i
m
pr
o
v
e
m
e
n
t
s
t
h
r
o
ugh
i
n
t
e
r
m
e
d
i
a
t
e
l
a
y
e
r
s
.
Kh
a
n
e
t
al.
[
25
]
r
e
s
e
a
r
c
h
o
n
a
tr
a
n
s
f
o
r
m
e
r
-
b
a
s
e
d
m
o
de
l
f
o
r
po
l
i
t
e
n
e
s
s
pr
e
d
i
c
t
i
o
n
i
n
c
o
nv
e
r
s
a
t
i
o
ns
hi
g
hli
g
h
t
s
i
t
s
s
upe
r
i
o
r
pe
r
f
o
r
m
a
nc
e
b
ut
n
o
t
e
s
i
t
s
l
i
mi
t
a
t
i
o
n
t
o
E
n
g
l
i
s
h
da
t
a
s
e
t
s
,
i
n
d
i
c
a
t
i
n
g
a
n
e
e
d
f
o
r
f
ut
ur
e
r
e
s
e
a
r
c
h
o
n
m
u
l
t
i
li
ngu
a
l
o
r
c
o
de
-
m
i
xe
d
da
t
a
.
T
h
e
a
u
t
h
or
s
s
ug
ge
s
t
th
a
t
th
e
p
r
op
o
s
e
d
m
ode
l
i
s
e
f
f
e
c
t
i
v
e
f
or
po
l
i
ten
e
s
s
p
r
e
di
c
t
i
o
n
i
n
E
n
gl
i
s
h
tex
t
s
b
u
t
l
a
c
ks
i
m
por
tan
t
f
e
a
tu
r
e
s
.
F
u
r
t
h
e
r
r
e
s
e
a
r
c
h
wi
l
l
i
n
c
or
p
or
a
te
t
h
e
s
e
f
e
a
tu
r
e
s
a
nd
e
x
pl
or
e
i
t
s
a
da
p
ta
t
i
o
n
to
m
u
l
t
i
-
l
i
n
gua
l
a
n
d
c
ode
m
i
x
e
d
d
a
ta
.
T
h
e
m
ode
l
h
a
s
n
ot
u
n
de
r
g
o
n
e
e
v
a
l
ua
t
i
o
n
f
or
th
e
d
e
tec
t
i
o
n
o
f
h
a
t
e
s
pe
e
c
h
a
n
d
o
f
f
e
n
s
i
v
e
c
o
n
ten
t.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
do
n
e
s
i
a
n
J
E
l
e
c
E
n
g
&
C
o
m
p
S
c
i
I
S
S
N:
2
5
0
2
-
4
7
52
I
ntegr
ati
ng
E
L
E
C
T
R
A
and
B
E
R
T
mode
ls
in
tr
ans
for
me
r
-
bas
e
d
me
ntal
…
(
J
unta
Z
e
niar
ja
)
317
C
h
a
t
b
ots
l
i
ke
C
h
a
tG
P
T
a
r
e
A
I
-
p
ow
e
r
e
d
p
r
og
r
a
m
s
de
s
i
gn
e
d
to
r
e
pl
i
c
a
t
e
h
um
a
n
c
o
n
v
e
r
s
a
t
i
o
n
a
n
d
p
e
r
f
or
m
tas
ks
s
u
c
h
a
s
a
n
s
we
r
i
n
g
q
ue
s
t
i
o
n
s
a
n
d
c
o
n
t
r
o
l
l
i
n
g
s
m
a
r
t
h
o
m
e
de
vi
c
e
s
.
A
s
tu
dy
by
S
udh
e
e
s
h
e
t
a
l
.
[
2
6
]
a
n
d
a
te
a
m
f
r
o
m
Un
i
v
e
r
s
i
da
d
I
n
te
r
n
a
c
i
o
n
a
l
a
n
a
l
yz
e
d
s
e
n
t
i
m
e
n
t
s
a
n
d
topi
c
s
i
n
C
h
a
t
G
P
T
-
b
a
s
e
d
twe
e
ts
us
i
n
g
l
a
t
e
n
t
di
r
i
c
hl
e
t
a
l
l
o
c
a
t
i
o
n
(
L
DA
)
f
or
topi
c
m
ode
l
in
g
a
n
d
t
h
e
B
E
R
T
m
o
de
l
f
or
s
e
n
t
i
m
e
n
t
a
n
a
l
y
s
i
s
,
a
c
hi
e
vi
n
g
9
6
.
49
%
a
c
c
u
r
a
c
y
.
T
h
e
r
e
s
e
a
r
c
h
hi
ghl
i
g
h
t
s
C
h
a
t
G
P
T
'
s
a
bi
l
i
t
y
to
e
n
h
a
n
c
e
us
e
r
e
x
pe
r
i
e
n
c
e
t
h
r
ough
n
a
t
u
r
a
l
l
a
n
g
u
a
g
e
p
r
oc
e
s
s
i
n
g
,
wi
t
h
B
E
R
T
e
f
f
i
c
i
e
n
t
l
y
c
a
teg
or
i
z
i
n
g
e
m
ot
i
o
n
s
i
n
tw
e
e
ts
.
De
s
pi
te
t
h
e
m
a
j
or
i
t
y
o
f
s
e
n
t
i
m
e
n
t
s
b
e
i
n
g
p
o
s
i
t
i
v
e
,
s
o
m
e
c
r
i
t
i
c
i
s
m
s
e
xi
s
t.
T
h
e
c
h
a
t
b
ot
i
n
dus
tr
y
i
s
p
r
o
j
e
c
ted
to
r
e
a
c
h
$3
.
62
bi
l
l
i
o
n
by
20
3
0
,
w
i
t
h
p
oten
t
i
a
l
e
du
c
a
t
i
o
n
a
l
b
e
n
e
f
i
t
s
.
T
h
e
s
tu
dy
e
m
ph
a
s
i
z
e
s
t
h
e
i
m
p
o
r
tan
c
e
o
f
publ
i
c
p
e
r
c
e
p
t
i
o
n
,
i
n
di
c
a
t
i
ng
f
u
tur
e
r
e
s
e
a
r
c
h
w
i
l
l
e
x
pa
n
d
da
ta
c
o
l
l
e
c
t
i
o
n
f
r
o
m
s
o
c
i
a
l
m
e
d
i
a
to
un
de
r
s
tan
d
s
e
n
t
i
m
e
n
t
s
b
e
tt
e
r
a
n
d
de
v
e
l
o
p
m
a
c
hi
n
e
l
e
a
r
ni
n
g
a
pp
r
oa
c
h
e
s
f
or
r
e
s
p
o
n
s
i
bl
e
de
p
l
o
y
m
e
n
t
o
f
C
h
a
tG
P
T
.
Al
r
u
q
i
a
n
d
Al
z
a
h
r
a
ni
[
27
]
r
e
s
e
a
r
c
h
i
n
“
E
v
a
l
ua
t
i
o
n
o
f
a
n
A
r
a
bi
c
c
h
a
t
b
ot
b
a
s
e
d
o
n
e
x
tr
a
c
t
i
v
e
qu
e
s
t
i
o
n
-
a
n
s
w
e
r
i
n
g
tr
a
n
s
f
e
r
l
e
a
r
ni
n
g
a
n
d
l
a
n
gua
ge
tr
a
n
s
f
or
m
e
r
s
”
de
m
o
n
s
tr
a
tes
th
e
A
r
a
E
l
e
c
tr
a
-
S
QuA
D
m
ode
l
's
s
up
e
r
i
o
r
pe
r
f
o
r
m
a
n
c
e
i
n
i
m
pr
o
v
i
n
g
t
h
e
a
c
c
u
r
a
c
y
a
n
d
r
e
l
e
v
a
n
c
e
o
f
A
r
a
bi
c
c
h
a
t
b
ots
.
Us
i
n
g
v
a
r
i
ous
A
r
a
bi
c
QA
d
a
tas
e
ts
a
n
d
tr
a
n
s
f
e
r
l
e
a
r
ni
n
g
tec
h
ni
que
s
,
t
h
e
s
tu
d
y
a
dd
r
e
s
s
e
s
t
h
e
c
o
m
p
l
e
xi
t
y
o
f
A
r
a
bi
c
c
h
a
t
b
ots
a
n
d
hi
ghl
i
g
h
t
s
th
e
n
e
e
d
f
or
b
e
tt
e
r
c
o
m
pu
ta
t
i
o
n
a
l
r
e
s
ou
r
c
e
s
a
n
d
p
r
e
p
r
oc
e
s
s
i
n
g
m
e
t
h
ods
.
T
h
e
r
e
s
e
a
r
c
h
e
m
ph
a
s
i
z
e
s
t
h
e
i
m
por
tan
c
e
o
f
c
o
n
f
i
de
n
c
e
m
e
tr
i
c
s
a
n
d
s
ug
ge
s
ts
f
u
tu
r
e
w
or
k
s
h
o
ul
d
f
o
c
us
o
n
i
m
pr
o
v
i
n
g
c
o
h
e
r
e
n
c
e
,
qua
l
i
t
y
,
a
n
d
a
u
to
m
a
t
e
d
p
r
oc
e
s
s
e
s
.
A
d
d
i
t
i
o
n
a
l
l
y
,
F
a
n
g
a
n
d
e
t
a
l
.
[
2
8
]
pr
op
os
e
a
n
e
f
f
e
c
t
i
v
e
E
L
E
C
T
R
A
-
b
a
s
e
d
pi
pe
l
i
n
e
m
od
e
l
f
o
r
s
e
n
t
i
m
e
n
t
a
n
a
l
y
s
i
s
o
f
tou
r
i
s
t
a
tt
r
a
c
t
i
o
n
r
e
vi
e
ws
,
s
h
o
w
c
a
s
i
n
g
hi
gh
e
f
f
i
c
i
e
n
c
y
a
n
d
a
dv
oc
a
t
i
n
g
de
e
p
l
e
a
r
ni
n
g
m
ode
l
s
f
or
e
n
h
a
n
c
e
d
a
n
a
l
y
s
i
s
.
E
v
a
l
ua
t
i
o
n
m
e
tr
i
c
s
s
u
c
h
a
s
pr
e
c
i
s
i
o
n
,
r
e
c
a
l
l
,
a
n
d
F
1
-
s
c
or
e
hi
ghl
i
gh
t
th
e
m
ode
l
'
s
s
uc
c
e
s
s
,
s
ug
ge
s
t
i
n
g
i
t
s
a
p
p
l
i
c
a
t
i
o
n
to
oth
e
r
do
m
a
i
n
s
.
I
n
"
C
o
m
b
a
t
i
n
g
f
a
k
e
n
e
ws
w
i
t
h
tr
a
n
s
f
or
m
e
r
s
,
"
K
a
s
n
e
s
i
s
e
t
al
.
[
2
9
]
d
i
s
c
us
s
e
s
us
i
n
g
l
a
n
gua
ge
m
o
d
e
l
s
to
a
s
s
e
s
s
t
h
e
s
ubj
e
c
t
i
vi
t
y
a
n
d
s
tan
c
e
o
f
s
oc
i
a
l
m
e
d
i
a
p
os
t
s
,
wi
t
h
R
o
B
E
R
T
a
ou
tp
e
r
f
or
m
i
n
g
E
L
E
C
T
R
A
i
n
s
tan
c
e
d
e
tec
t
i
o
n
.
T
h
e
s
tudy
hi
ghl
i
gh
t
s
th
e
c
h
a
l
l
e
n
ge
s
o
f
v
e
r
i
f
y
i
n
g
s
oc
i
a
l
m
e
d
i
a
i
nf
or
m
a
t
i
o
n
a
n
d
s
u
gge
s
ts
f
u
t
u
r
e
r
e
s
e
a
r
c
h
o
n
tr
e
e
-
b
a
s
e
d
s
tr
uc
tu
r
e
s
,
f
r
e
e
z
i
n
g
m
ode
l
p
a
r
a
m
e
te
r
s
,
a
n
d
g
r
a
ph
n
e
u
r
a
l
n
e
tw
or
ks
,
c
o
n
c
l
ud
i
n
g
t
h
a
t
m
i
s
i
nf
or
m
a
t
i
o
n
i
s
o
f
ten
l
i
n
ke
d
to
s
ubj
e
c
t
i
v
e
l
a
n
gua
ge
.
T
h
i
s
s
tudy
i
nv
e
s
t
i
ga
t
e
d
th
e
i
n
t
e
g
r
a
t
i
o
n
o
f
a
dv
a
n
c
e
d
NL
P
m
ode
l
s
l
i
ke
E
L
E
C
T
R
A
a
n
d
B
E
R
T
i
n
to
m
e
n
t
a
l
h
e
a
l
t
h
c
a
r
e
c
h
a
t
b
ots
.
W
hi
l
e
e
a
r
l
i
e
r
s
tu
d
i
e
s
h
a
v
e
e
x
pl
or
e
d
th
e
de
v
e
l
o
pm
e
n
t
o
f
A
I
-
d
r
i
v
e
n
c
h
a
t
b
ots
f
or
m
e
n
t
a
l
h
e
a
l
t
h
a
n
d
t
h
e
i
r
p
oten
t
i
a
l
i
n
h
a
n
d
l
i
n
g
m
e
n
t
a
l
h
e
a
l
t
h
di
a
l
ogue
s
,
th
e
y
h
a
v
e
n
ot
e
x
pl
i
c
i
t
l
y
a
d
d
r
e
s
s
e
d
t
h
e
c
h
a
l
l
e
n
ge
s
r
e
l
a
ted
to
c
ul
tu
r
a
l
s
e
n
s
i
t
i
vi
t
y
,
e
t
h
i
c
a
l
c
o
n
s
i
de
r
a
t
i
o
n
s
,
a
n
d
t
h
e
a
p
p
l
i
c
a
t
i
o
n
o
f
t
h
e
s
e
m
od
e
l
s
to
m
u
l
t
i
l
i
n
gua
l
or
c
ode
-
m
i
x
e
d
da
tas
e
t
s
.
T
h
e
i
n
teg
r
a
t
i
o
n
o
f
a
dv
a
n
c
e
d
NL
P
m
ode
l
s
l
i
ke
E
L
E
C
T
R
A
a
n
d
B
E
R
T
i
n
to
m
e
n
t
a
l
h
e
a
l
t
h
c
a
r
e
c
h
a
t
b
o
t
s
i
s
a
r
a
pi
d
l
y
de
v
e
l
o
pi
n
g
a
r
e
a
o
f
r
e
s
e
a
r
c
h
t
h
a
t
p
r
o
m
i
s
e
s
to
e
n
h
a
n
c
e
t
h
e
e
f
f
e
c
t
i
v
e
n
e
s
s
a
n
d
e
f
f
i
c
i
e
n
c
y
o
f
t
h
e
s
e
s
y
s
te
m
s
.
T
h
e
l
i
te
r
a
tu
r
e
r
e
v
e
a
l
s
a
c
l
e
a
r
tr
a
j
e
c
tor
y
f
r
o
m
s
i
m
p
l
e
r
u
l
e
-
b
a
s
e
d
s
y
s
tem
s
to
c
o
m
pl
e
x
A
I
-
d
r
i
v
e
n
s
o
l
u
t
i
o
n
s
,
un
de
r
s
c
or
i
n
g
th
e
p
oten
t
i
a
l
o
f
t
h
e
s
e
m
o
de
l
s
i
n
h
a
n
dl
i
n
g
t
h
e
n
ua
n
c
e
s
o
f
m
e
n
t
a
l
h
e
a
l
t
h
di
a
l
ogu
e
s
.
A
s
th
e
s
e
tec
h
n
o
l
og
i
e
s
e
v
o
l
v
e
,
i
t
i
s
i
m
pe
r
a
t
i
v
e
to
b
a
l
a
n
c
e
t
h
e
i
r
tec
h
ni
c
a
l
a
dv
a
n
c
e
m
e
n
t
s
w
i
t
h
e
t
h
i
c
a
l
c
o
n
s
i
de
r
a
t
i
o
n
s
,
e
n
s
u
r
i
n
g
t
h
e
y
s
e
r
v
e
a
s
a
be
n
e
f
i
c
i
a
l
too
l
i
n
m
e
n
t
a
l
h
e
a
l
t
h
c
a
r
e
.
F
u
tu
r
e
r
e
s
e
a
r
c
h
s
h
ou
l
d
c
o
n
t
i
n
ue
to
e
x
p
l
or
e
t
h
i
s
i
n
t
e
g
r
a
t
i
o
n
,
f
o
c
us
i
n
g
o
n
i
m
pr
o
v
i
n
g
t
h
e
e
m
ot
i
o
n
a
l
i
n
t
e
l
l
i
ge
n
c
e
a
n
d
r
e
s
p
o
n
s
i
v
e
n
e
s
s
o
f
c
h
a
t
b
ots
to
b
e
tt
e
r
a
d
d
r
e
s
s
t
h
e
di
v
e
r
s
e
n
e
e
ds
o
f
m
e
n
t
a
l
h
e
a
l
t
h
c
a
r
e
.
2.
M
E
T
HO
D
T
hi
s
r
e
s
e
a
r
c
h
f
o
l
l
o
we
d
a
s
t
r
uc
t
u
r
e
d
a
ppr
o
a
c
h
,
a
s
de
p
i
c
t
e
d
i
n
F
i
g
ur
e
1,
be
g
i
nni
ng
w
i
t
h
da
t
a
c
o
l
l
e
c
t
i
o
n
a
n
d
pr
o
gr
e
s
s
i
n
g
t
h
r
o
ugh
da
t
a
s
p
l
i
t
t
i
n
g
f
o
r
t
r
a
i
nin
g,
t
e
s
t
i
n
g,
a
n
d
v
a
li
da
t
i
o
n
pha
s
e
s
.
T
h
e
m
o
de
l
i
n
g
ph
a
s
e
ut
i
li
z
e
d
f
i
ve
d
i
f
f
e
r
e
n
t
m
o
de
l
s
:
B
E
R
T
,
C
NN
,
L
S
T
M
,
E
L
E
C
T
R
A
,
a
n
d
a
hy
br
i
d
o
f
E
L
E
C
T
R
A
a
nd
B
E
R
T
.
E
v
a
l
ua
t
i
o
n
o
f
t
h
e
s
e
m
o
de
l
s
wa
s
pe
r
f
o
r
m
e
d
us
i
ng
m
e
t
r
i
c
s
l
i
ke
a
c
c
ur
a
c
y
,
pr
e
c
i
s
i
o
n
,
r
e
c
a
l
l
,
a
n
d
F
1
s
c
or
e
,
u
l
t
i
m
a
t
e
l
y
de
t
e
r
m
i
n
i
ng
t
h
e
c
l
a
s
s
if
i
c
a
t
i
o
n
a
c
c
ur
a
c
y.
F
i
g
ur
e
2,
i
s
e
x
p
l
a
i
ne
d
i
n
s
e
c
t
i
o
ns
2.
1
t
h
r
o
ugh
2.
4
o
f
t
h
e
d
i
s
c
us
s
i
o
n
.
F
i
gur
e
1
.
S
h
o
ws
t
h
e
f
l
o
wc
ha
r
t
o
f
t
h
e
A
I
-
b
a
s
e
d
m
o
de
l
s
a
n
d
e
x
pe
r
im
e
n
t
a
l
m
e
t
h
o
ds
a
pp
l
i
e
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2
5
0
2
-
4
7
52
In
do
n
e
s
i
a
n
J
E
l
e
c
E
n
g
&
C
o
m
p
S
c
i
,
Vo
l
.
3
7
,
N
o
.
1
,
J
a
n
ua
r
y
20
2
5
:
3
1
5
-
324
318
F
i
gur
e
2
.
E
L
E
C
T
R
A
a
n
d
B
E
R
T
i
n
t
r
a
n
s
f
o
r
m
e
r
m
o
de
l
2
.
1.
E
x
p
e
r
im
e
n
t
s
e
t
u
p
T
o
c
o
n
duc
t
o
ur
t
r
a
i
ni
ng,
we
l
e
ve
r
a
ge
Goo
gl
e
C
o
l
a
b
P
r
o,
e
qui
ppe
d
w
i
t
h
a
s
i
n
g
l
e
NV
I
DI
A
P
100/T
4
GPU
wi
t
h
a
c
a
pa
c
i
t
y
o
f
up
to
15
GB
a
n
d
R
A
M
r
e
a
c
hi
ng
8
GB
.
T
h
e
pr
e
-
t
r
a
i
n
e
d
l
a
n
gu
a
ge
m
o
de
l
s
u
t
i
li
z
e
d
e
n
c
o
m
pa
s
s
E
L
E
C
T
R
A
ba
s
e
m
o
de
l
a
n
d
B
E
R
T
e
m
p
l
o
yi
ng
t
h
e
T
r
a
n
s
f
o
r
m
e
r
a
l
go
r
i
t
hm
,
a
l
o
n
g
w
i
t
h
t
h
e
C
NN
.
S
ub
s
e
qu
e
n
t
l
y
,
we
a
pp
ly
t
h
e
E
L
E
C
T
R
A
b
a
s
e
m
o
de
l
w
i
t
h
L
S
T
M
a
s
t
h
e
a
l
go
r
i
t
hm
,
E
L
E
C
T
R
A
+
B
E
R
T
o
p
t
i
mi
z
a
t
i
o
n
i
nv
o
l
v
e
s
t
h
e
u
s
e
o
f
t
h
e
A
da
m
W
o
p
t
i
mi
z
e
r
,
s
e
tt
i
n
g
t
h
e
l
e
a
r
ni
ng
r
a
t
e
to
1r
=
5e
-
5.
2
.
2
.
Dat
as
e
t
T
hi
s
r
e
s
e
a
r
c
h
ut
i
li
z
e
s
s
e
c
o
n
d
a
r
y
da
t
a
,
c
o
m
pr
i
s
in
g
t
wo
t
y
pe
s
o
f
I
n
t
e
n
t
s
.
j
s
o
n
da
t
a
s
e
t
s
.
T
h
e
f
i
r
s
t
da
t
a
s
e
t,
I
n
t
e
n
t
s
.
j
s
o
n
(
E
n
g)
,
i
s
s
o
ur
c
e
d
publ
i
c
ly
a
n
d
i
n
c
l
ude
s
a
t
tr
i
b
ut
e
s
s
uc
h
a
s
t
a
g
(
)
,
pa
tt
e
r
n
s
(
)
,
a
n
d
r
e
s
po
n
s
e
s
(
)
.
C
o
n
v
e
r
s
e
ly
,
t
h
e
s
e
c
o
n
d
da
t
a
s
e
t
,
I
n
t
e
n
t
s
.
j
s
o
n
(
I
n
d)
,
po
s
s
e
s
s
e
s
s
i
mi
l
a
r
a
tt
r
i
b
ut
e
s
b
ut
i
s
a
c
qu
i
r
e
d
t
h
r
o
ugh
p
r
i
va
t
e
s
o
ur
c
e
s
.
T
h
e
s
e
a
tt
r
i
b
ut
e
s
pl
a
y
a
c
r
uc
i
a
l
r
o
l
e
i
n
b
ot
h
da
t
a
s
e
t
s
.
T
h
e
t
a
g
(
)
f
u
n
c
t
i
o
n
i
s
de
s
i
g
ne
d
to
c
l
a
s
s
if
y
t
h
e
i
n
t
e
n
t
or
pur
p
o
s
e
o
f
a
que
s
t
i
o
n
o
r
s
t
a
t
e
m
e
n
t
.
P
a
tt
e
r
n
s
(
)
,
c
o
n
s
i
s
t
i
n
g
o
f
t
e
x
t
e
x
a
m
p
l
e
s
o
r
ph
r
a
s
e
s
,
a
r
e
c
r
a
f
t
e
d
to
r
e
pr
e
s
e
n
t
t
h
e
i
n
t
e
n
t
a
s
s
o
c
i
a
t
e
d
wi
t
h
a
s
pe
c
i
f
i
c
t
a
g
a
n
d
a
r
e
us
e
d
a
s
t
r
a
i
ni
ng
i
n
put
s
f
o
r
th
e
NL
P
m
o
de
l
.
T
hi
s
a
ll
o
ws
t
h
e
m
o
de
l
t
o
r
e
c
o
gn
i
z
e
a
n
d
a
c
c
ur
a
t
e
l
y
m
a
p
s
i
mi
l
a
r
i
n
put
s
to
t
h
e
c
or
r
e
s
po
n
d
i
n
g
t
a
g
.
On
t
h
e
ot
h
e
r
h
a
n
d,
r
e
s
po
n
s
e
(
)
i
s
t
h
e
o
ut
pu
t
pr
o
vi
de
d
by
t
he
c
h
a
t
b
o
t
wh
e
n
a
pa
t
t
e
r
n
i
s
i
d
e
n
t
i
f
i
e
d
w
i
t
h
a
pa
r
t
i
c
u
l
a
r
t
a
g.
T
hi
s
r
e
s
po
ns
e
i
s
t
y
p
i
c
a
ll
y
pr
e
de
f
i
ne
d
by
t
h
e
c
ha
t
b
ot
'
s
c
r
e
a
to
r
a
n
d
i
s
ut
i
li
z
e
d
t
o
de
l
i
ve
r
a
n
a
ppr
o
pr
i
a
t
e
r
e
p
l
y
t
o
t
h
e
us
e
r
.
2
.
3
.
S
p
l
it
d
at
a
F
o
l
l
o
w
i
ng
t
hi
s
pr
o
c
e
s
s
,
da
t
a
s
p
l
i
t
t
i
n
g
i
s
e
x
e
c
ut
e
d,
wh
e
r
e
i
n
t
h
e
da
t
a
i
s
m
e
t
h
o
d
i
c
a
l
ly
d
i
v
i
de
d
i
n
t
o
d
i
s
t
i
n
c
t
s
e
t
s
,
t
y
p
i
c
a
ll
y
c
o
m
pr
i
s
i
n
g
t
r
a
i
ni
ng,
t
e
s
t
i
n
g,
a
n
d
va
l
i
d
a
t
i
o
n
s
e
t
s
.
I
n
t
h
e
pr
o
vi
de
d
c
o
n
t
e
x
t
,
t
h
i
s
p
r
o
c
e
dur
e
i
s
s
pe
c
if
i
c
a
ll
y
a
pp
l
i
e
d
to
t
h
e
'
I
n
t
e
n
t
s
.
E
n
g
'
a
n
d
'
I
n
t
e
n
t
s
.
I
n
d
'
da
t
a
s
e
t
s
.
S
uc
h
s
e
g
m
e
n
t
a
t
i
o
n
o
f
da
t
a
i
s
o
f
pa
r
a
m
o
un
t
i
m
po
r
t
a
n
c
e
f
o
r
t
h
e
e
f
f
e
c
t
i
ve
tr
a
i
ni
ng
o
f
m
o
de
l
s
,
a
s
i
t
a
l
l
o
w
s
t
h
e
m
to
b
e
f
i
ne
-
t
un
e
d
o
n
di
ve
r
s
e
da
t
a
s
a
m
p
l
e
s
a
n
d
e
n
s
ur
e
s
t
h
e
i
r
r
o
b
us
t
pe
r
f
o
r
m
a
n
c
e
o
n
n
e
w,
u
n
s
e
e
n
da
t
a
,
t
h
e
r
e
by
e
nh
a
n
c
i
ng
t
h
e
o
v
e
r
a
l
l
e
f
f
i
c
a
c
y
o
f
t
h
e
m
o
de
l
i
ng
pr
o
c
e
s
s
.
2
.
4
.
M
od
e
l
in
g
M
o
de
l
i
n
g
i
s
a
s
t
a
ge
wh
e
r
e
de
e
p
l
e
a
r
ni
ng
m
o
de
l
s
a
r
e
i
m
p
l
e
m
e
n
t
e
d
to
a
c
hi
e
v
e
t
h
e
r
e
s
e
a
r
c
h
o
bj
e
c
t
i
v
e
,
n
a
m
e
ly
t
h
e
a
c
c
ur
a
c
y
o
f
t
h
e
d
a
t
a
s
e
t
f
o
r
i
n
t
e
n
t
s
d
a
t
a
.
T
hi
s
e
x
pe
r
i
m
e
n
t
i
nv
o
l
v
e
s
f
i
ve
pr
e
d
i
c
t
i
v
e
a
l
g
o
r
i
t
hm
s
:
B
E
R
T
,
C
NN
,
L
S
T
M
,
E
L
E
C
T
R
A
,
a
n
d
a
c
o
m
bi
na
t
i
o
n
o
f
E
L
E
C
T
R
A
a
n
d
B
E
R
T
.
T
h
e
s
e
f
i
ve
m
o
de
l
s
c
a
n
h
a
n
d
l
e
c
a
s
e
s
o
f
s
e
n
t
i
m
e
n
t
a
n
a
ly
s
i
s
,
t
e
x
t
s
um
m
a
r
i
z
a
t
i
o
n
,
o
r
que
s
t
i
o
n
un
de
r
s
t
a
n
d
i
ng,
wh
e
r
e
a
c
c
ur
a
c
y
i
s
m
e
a
s
ur
e
d
b
y
c
o
m
pa
r
i
ng
t
h
e
m
o
de
l
'
s
pr
e
d
i
c
t
i
o
n
s
w
i
t
h
a
c
t
ua
l
a
n
s
we
r
s
i
n
t
h
e
f
o
r
m
o
f
t
e
x
t
da
t
a
.
2.
4.
1.
B
id
ir
e
c
t
ion
al
e
n
c
od
e
r
r
e
p
r
e
s
e
n
t
at
ion
f
r
o
m
t
r
an
s
f
o
r
m
e
r
s
B
E
R
T
e
m
p
l
o
y
s
a
pr
e
-
t
r
a
i
ni
ng
m
e
t
h
o
d
c
a
l
l
e
d
a
m
a
s
ke
d
l
a
n
gua
ge
m
o
de
l
(
M
L
M
)
,
i
n
w
hi
c
h
a
s
pe
c
if
i
c
f
r
a
c
t
i
o
n
o
f
i
n
put
to
ke
n
s
i
s
o
b
s
c
ur
e
d.
T
h
e
o
bj
e
c
t
i
ve
o
f
t
h
e
m
o
de
l
i
s
to
m
a
ke
pr
e
d
i
c
t
i
o
ns
r
e
ga
r
di
n
g
t
h
e
s
e
c
o
n
c
e
a
l
e
d
to
ke
n
s
by
c
o
n
s
i
de
r
i
ng
t
h
e
i
r
c
o
n
t
e
x
t
ua
l
inf
o
r
m
a
t
i
o
n
.
I
n
m
a
t
h
e
m
a
t
i
c
a
l
t
e
r
m
s
,
wh
e
n
pr
o
vi
de
d
w
i
t
h
a
n
i
nput
s
e
que
n
c
e
X,
t
h
e
l
o
s
s
L
i
s
c
o
m
put
e
d
a
s
-
l
o
g
P
(
xi
|
X
-
i
;
θ)
,
wi
t
h
θ
r
e
pr
e
s
e
n
t
i
n
g
t
h
e
m
o
de
l
's
pa
r
a
m
e
t
e
r
s
.
F
i
g
ur
e
3
s
h
o
ws
t
h
e
a
r
c
hi
t
e
c
t
ur
e
o
f
B
E
R
T
.
2.
4.
2.
Convol
u
t
ion
al
n
e
u
r
al
n
e
t
wo
r
k
T
h
e
de
v
e
l
o
p
m
e
n
t
o
f
a
C
NN
m
o
de
l
k
i
c
ks
o
f
f
w
i
t
h
m
e
t
i
c
u
l
o
us
da
t
a
pr
e
pr
o
c
e
s
s
i
n
g,
wh
e
r
e
to
ke
ni
z
a
t
i
o
n
a
n
d
pa
dd
i
n
g
a
r
e
c
r
uc
i
a
l
f
o
r
s
t
a
n
da
r
d
i
z
i
ng
i
n
put
l
e
n
gt
h
s
.
Us
i
n
g
T
e
ns
o
r
F
l
o
w's
T
F
T
o
ke
ni
z
e
r
,
t
h
e
da
t
a
i
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
do
n
e
s
i
a
n
J
E
l
e
c
E
n
g
&
C
o
m
p
S
c
i
I
S
S
N:
2
5
0
2
-
4
7
52
I
ntegr
ati
ng
E
L
E
C
T
R
A
and
B
E
R
T
mode
ls
in
tr
ans
for
me
r
-
bas
e
d
me
ntal
…
(
J
unta
Z
e
niar
ja
)
319
to
ke
ni
z
e
d,
a
n
d
n
u
m
e
r
i
c
a
l
s
e
que
n
c
e
s
a
r
e
pa
dde
d
to
e
n
s
ur
e
c
o
n
s
i
s
t
e
n
c
y
.
L
a
b
e
l
s
a
r
e
c
o
nv
e
r
t
e
d
to
n
u
m
e
r
i
c
a
l
f
o
r
m
a
n
d
e
n
c
o
de
d
i
n
t
o
bi
n
a
r
y
m
a
t
r
i
c
e
s
vi
a
o
n
e
-
h
o
t
e
n
c
o
di
n
g
us
i
ng
a
l
a
be
l
e
n
c
o
de
r
.
T
h
e
da
t
a
s
e
t
i
s
t
h
e
n
d
i
vi
de
d
i
n
t
o
80%
f
o
r
tr
a
i
ni
n
g
a
n
d
20%
f
o
r
t
e
s
t
i
n
g,
e
n
s
ur
i
ng
a
ba
l
a
n
c
e
d
a
ppr
o
a
c
h
t
o
m
o
de
l
t
r
a
i
ni
ng
a
n
d
v
a
li
da
t
i
o
n
.
T
h
e
C
NN
a
r
c
hi
t
e
c
t
ur
e
i
s
c
o
m
po
s
e
d
o
f
m
u
l
t
i
p
l
e
l
a
y
e
r
s
:
a
n
e
m
b
e
dd
i
ng
l
a
y
e
r
f
o
r
v
e
c
to
r
t
r
a
n
s
f
o
r
m
a
t
i
o
n
,
c
o
nv
o
l
ut
i
o
na
l
a
n
d
po
o
l
i
ng
l
a
y
e
r
s
f
o
r
f
e
a
t
ur
e
e
x
t
r
a
c
t
i
o
n
,
de
ns
e
l
a
y
e
r
s
f
o
r
f
e
a
t
ur
e
c
onn
e
c
t
i
o
n
,
a
n
d
a
n
o
u
t
pu
t
l
a
y
e
r
to
ge
n
e
r
a
t
e
i
n
t
e
n
t
pr
o
b
a
bi
li
t
i
e
s
.
T
h
e
m
o
de
l
i
s
t
r
a
i
n
e
d
w
i
t
h
t
h
e
A
da
m
o
pt
i
mi
z
e
r
,
e
m
p
l
o
yi
ng
c
a
t
e
g
o
r
i
c
a
l
c
r
o
s
s
e
n
t
r
o
py
a
s
t
h
e
l
o
s
s
f
u
n
c
t
i
o
n
a
n
d
a
c
c
ur
a
c
y
a
s
t
h
e
pe
r
f
o
r
m
a
n
c
e
m
e
t
r
i
c
.
T
hi
s
we
l
l
-
s
t
r
uc
t
u
r
e
d
m
e
t
h
o
do
l
o
g
y
e
n
s
ur
e
s
t
h
e
C
NN
m
o
de
l
pr
o
c
e
s
s
e
s
a
nd
c
l
a
s
s
i
f
i
e
s
da
t
a
wi
t
h
hi
g
h
pr
e
c
i
s
i
o
n
a
n
d
e
f
f
i
c
i
e
n
c
y.
F
i
gur
e
3
.
B
E
R
T
m
o
de
l
a
r
c
hi
t
e
c
t
ur
e
2.
4.
3.
L
on
g
s
h
or
t
-
t
e
r
m
m
e
m
o
r
y
L
S
T
M
i
s
a
s
o
phi
s
t
i
c
a
t
e
d
t
y
pe
o
f
a
r
t
i
f
i
c
i
a
l
r
e
c
ur
r
e
nt
n
e
ur
a
l
n
e
t
wor
k
(
R
NN
)
a
r
c
hi
t
e
c
t
ur
e
us
e
d
i
n
de
e
p
l
e
a
r
ni
ng,
n
o
t
a
bl
e
f
o
r
i
t
s
f
e
e
d
b
a
c
k
c
o
n
n
e
c
t
i
o
ns
t
h
a
t
s
e
t
i
t
a
pa
r
t
f
r
o
m
t
r
a
di
t
i
o
n
a
l
n
e
ur
a
l
n
e
t
wo
r
ks
.
A
n
L
S
T
M
c
e
l
l
i
s
c
o
m
po
s
e
d
o
f
s
e
v
e
r
a
l
ke
y
c
o
m
po
n
e
n
t
s
:
t
h
e
C
e
l
l
S
t
a
t
e
(
C
_t)
,
I
n
put
Ga
t
e
(
i
_t
)
,
F
o
r
ge
t
Ga
t
e
(
f
_t
)
,
a
n
d
Out
pu
t
Ga
t
e
(
o
_
t
)
[
30
]
.
Af
t
e
r
t
h
e
e
m
b
e
dd
i
ng
pr
o
c
e
s
s
,
L
S
T
M
n
e
t
w
o
r
ks
e
x
t
r
a
c
t
s
e
que
n
t
i
a
l
de
pe
n
de
n
c
i
e
s
f
r
o
m
t
h
e
i
nput
t
e
x
t
.
Gi
ve
n
t
h
e
i
r
a
bi
li
t
y
t
o
pr
o
c
e
s
s
s
e
que
n
c
e
s
,
L
S
T
M
s
a
r
e
pa
r
t
i
c
u
l
a
r
ly
a
de
pt
a
t
h
a
n
d
li
ng
que
s
t
i
o
n
s
t
h
a
t
t
y
p
i
c
a
ll
y
s
t
a
r
t
wi
t
h
"
w
h"
wo
r
ds
l
i
ke
"
w
h
o
,
"
"
wha
t
,
"
o
r
"
wh
e
r
e
,
"
a
n
d
e
n
d
w
i
t
h
a
s
u
bj
e
c
t
or
o
bj
e
c
t
.
R
e
c
o
gni
z
i
ng
t
hi
s
s
t
r
uc
t
u
r
e
h
e
l
p
s
i
n
ga
i
n
i
ng
de
e
pe
r
i
n
s
i
g
h
t
s
i
n
t
o
th
e
que
s
t
i
o
ns
'
un
de
r
lyi
ng
pa
tt
e
r
n
s
.
2.
4.
4.
E
f
f
icie
n
t
l
y
l
e
a
r
n
in
g
an
e
n
c
od
e
r
t
h
at
c
l
as
s
i
f
ies
t
ok
e
n
r
e
p
l
ac
e
m
e
n
t
s
ac
c
u
r
at
e
l
y
E
L
E
C
T
R
A
e
m
p
l
o
y
s
a
d
i
s
c
r
im
i
na
t
i
v
e
tr
a
i
ni
ng
m
e
t
ho
d
wh
e
r
e
t
h
e
m
o
de
l
l
e
a
r
ns
to
di
s
t
i
n
gu
i
s
h
b
e
t
we
e
n
"
r
e
a
l
"
a
n
d
"
f
a
k
e
"
to
ke
n
s
w
i
t
hi
n
a
s
e
n
t
e
nc
e
.
Gi
ve
n
a
n
i
n
put
X=
[
x
1,
x
2,
.
.
.
,
x
n
]
,
a
ge
n
e
r
a
to
r
G
p
r
o
p
o
s
e
s
s
ubs
t
i
t
u
t
e
s
xi
f
o
r
m
a
s
ke
d
to
ke
n
s
,
whi
l
e
a
di
s
c
r
i
mi
n
a
t
o
r
D
e
v
a
l
ua
t
e
s
t
h
e
pr
o
b
a
bil
i
t
y
P
(
D(
xi
)
=
1|
X)
th
a
t
e
a
c
h
to
ke
n
i
s
ge
n
u
i
ne
.
T
he
o
bj
e
c
t
i
v
e
i
s
t
o
m
i
n
i
mi
z
e
-
l
o
g(
D(
xi
)
)
f
o
r
r
e
a
l
to
ke
n
s
a
n
d
-
l
o
g(
1
-
D(
xi
)
)
f
o
r
f
a
k
e
to
ke
n
s
,
r
e
f
i
n
i
ng
t
h
e
m
o
de
l
's
a
bil
i
t
y
t
o
i
de
n
t
i
f
y
a
ut
he
n
t
i
c
ve
r
s
us
i
n
a
ut
h
e
n
t
i
c
e
l
e
m
e
n
t
s
i
n
t
h
e
t
e
x
t
.
F
i
g
ur
e
4
s
ho
ws
t
h
e
a
r
c
hi
t
e
c
t
ur
e
o
f
E
L
E
C
T
R
A
.
F
i
gur
e
4
.
E
L
E
C
T
R
A
m
o
de
l
a
r
c
hi
t
e
c
t
ur
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2
5
0
2
-
4
7
52
In
do
n
e
s
i
a
n
J
E
l
e
c
E
n
g
&
C
o
m
p
S
c
i
,
Vo
l
.
3
7
,
N
o
.
1
,
J
a
n
ua
r
y
20
2
5
:
3
1
5
-
324
320
2.
4.
5.
Com
b
in
at
ion
of
E
L
E
CT
RA
an
d
B
E
RT
I
n
t
h
e
r
e
a
l
m
o
f
n
a
t
ur
a
l
l
a
n
gua
ge
pr
o
c
e
s
s
i
n
g,
tr
a
n
s
f
o
r
m
e
r
m
o
de
l
s
l
i
ke
E
L
E
C
T
R
A
a
n
d
B
E
R
T
a
r
e
p
i
v
o
t
a
l
f
o
r
a
n
a
ly
z
i
ng
i
n
t
e
n
t
da
t
a
s
e
t
s
c
o
m
po
s
e
d
o
f
t
a
gs
,
pa
tt
e
r
n
s
,
a
n
d
r
e
s
po
n
s
e
s
.
T
h
e
pr
o
c
e
s
s
s
t
a
r
t
s
by
o
r
ga
ni
z
i
ng
l
a
b
e
l
s
a
n
d
s
e
n
t
e
n
c
e
pa
tt
e
r
n
s
i
n
to
l
i
s
t
s
,
wi
t
h
t
a
gs
a
c
t
i
n
g
a
s
l
a
be
l
s
a
n
d
pa
tt
e
r
n
s
s
to
r
i
n
g
s
e
n
t
e
n
c
e
s
t
r
uc
t
ur
e
s
.
Us
i
n
g
S
k
l
e
a
r
n
,
t
h
e
da
t
a
i
s
d
i
vi
de
d
in
t
o
80
%
f
o
r
t
r
a
i
ni
ng
a
n
d
20%
f
o
r
t
e
s
t
i
n
g.
E
L
E
C
T
R
A's
to
ke
ni
z
e
r
,
pa
r
t
i
c
u
l
a
r
l
y
go
o
gl
e
/e
l
e
c
t
r
a
-
s
m
a
ll
-
ge
n
e
r
a
to
r
,
c
r
e
a
t
e
s
t
e
x
t
r
e
s
e
m
bli
ng
t
h
e
o
r
i
g
i
na
l
,
w
hi
c
h
i
s
t
h
e
n
e
v
a
l
ua
t
e
d
by
G
o
o
gl
e
/
E
L
E
C
T
R
A
-
s
m
a
ll
-
d
i
s
c
r
im
i
n
a
to
r
,
l
e
v
e
r
a
g
i
ng
t
r
a
n
s
f
o
r
m
e
r
s
f
o
r
e
f
f
i
c
i
e
n
t
a
n
d
a
c
c
ur
a
t
e
m
o
de
li
ng.
T
h
e
ge
n
e
r
a
to
r
m
a
s
k
s
a
n
d
a
l
t
e
r
s
wo
r
d
s
to
pr
o
duc
e
c
o
n
t
e
x
t
ua
l
ly
r
e
l
e
v
a
n
t
s
e
n
t
e
n
c
e
s
,
whi
c
h
a
r
e
e
n
c
o
de
d
to
a
i
d
t
h
e
m
o
de
l
's
un
d
e
r
s
t
a
n
d
i
ng,
w
i
t
h
a
d
j
us
t
m
e
n
t
s
f
o
r
s
e
n
t
e
n
c
e
l
e
n
gt
h
a
n
d
pa
dd
i
ng.
T
h
e
E
L
E
C
T
R
A
d
i
s
c
r
i
mi
na
t
o
r
tr
a
i
ns
o
n
t
hi
s
da
t
a
,
a
s
s
e
s
s
i
ng
i
f
s
e
n
t
e
n
c
e
s
a
r
e
ge
n
e
r
a
t
e
d
or
or
i
g
i
na
l
,
upda
t
i
n
g
t
h
e
m
a
s
k
ge
n
e
r
a
t
o
r
a
c
c
o
r
di
n
g
ly
.
Dur
i
n
g
t
r
a
i
ni
ng,
t
h
e
A
da
m
W
o
pt
i
m
i
z
e
r
f
i
ne
-
t
un
e
s
t
h
e
m
o
de
l
us
i
ng
B
E
R
T
pr
i
n
c
i
p
l
e
s
.
B
E
R
T
e
m
p
l
o
y
s
M
A
S
K
a
n
d
n
e
x
t
s
e
n
t
e
n
c
e
pr
e
d
i
c
t
i
o
n
s
(
N
S
P
)
to
un
de
r
s
t
a
n
d
i
n
t
e
r
-
s
e
n
t
e
n
c
e
r
e
l
a
t
i
o
n
s
hi
ps
,
f
o
c
us
i
n
g
o
n
wo
r
d
r
e
l
e
v
a
n
c
e
a
n
d
m
e
a
ni
ng
to
de
c
i
de
i
f
s
e
n
t
e
n
c
e
s
n
e
e
d
r
e
ge
n
e
r
a
t
i
o
n
b
a
s
e
d
o
n
t
h
e
d
i
s
c
r
im
i
na
t
or
'
s
j
u
dg
m
e
n
t
.
T
hi
s
s
o
phi
s
t
i
c
a
t
e
d
m
e
t
h
o
d
e
nh
a
n
c
e
s
t
h
e
m
o
de
l
'
s
n
a
t
ur
a
l
l
a
n
gua
g
e
un
de
r
s
t
a
n
d
i
ng
a
n
d
r
e
s
po
n
s
e
c
a
pa
bi
li
t
i
e
s
.
2.
5.
E
val
u
at
ion
Af
t
e
r
s
p
l
i
t
t
i
n
g
a
n
d
f
i
ne
-
t
uni
n
g
t
h
e
m
o
de
l
,
t
h
e
e
v
a
l
ua
t
i
o
n
o
f
i
t
s
pe
r
f
o
r
m
a
nc
e
w
i
ll
b
e
c
o
n
duc
t
e
d
t
h
r
o
ugh
t
h
e
a
n
a
ly
s
i
s
o
f
o
ut
pu
t
r
e
s
ul
t
s
.
T
hi
s
a
s
s
e
s
s
m
e
n
t
w
i
ll
i
nv
o
l
ve
t
h
e
ut
i
l
i
z
a
t
i
o
n
o
f
a
C
o
nf
u
s
i
o
n
M
a
t
r
i
x
a
n
d
v
a
r
i
o
us
m
e
t
r
i
c
s
s
uc
h
a
s
a
c
c
ur
a
c
y
,
pr
e
c
i
s
i
o
n
,
r
e
c
a
l
l
,
a
n
d
F
1
-
s
c
o
r
e
,
a
i
m
i
ng
to
c
o
m
pr
e
he
n
s
i
ve
ly
g
a
uge
t
h
e
m
o
de
l
's
e
f
f
e
c
t
i
v
e
n
e
s
s
a
c
r
o
s
s
a
l
l
t
a
s
ks
.
A
c
c
ur
a
c
y
:
c
a
l
c
u
l
a
t
e
s
t
h
e
pr
o
p
or
t
i
o
n
o
f
a
c
c
ur
a
t
e
l
y
p
r
e
d
i
c
t
e
d
s
a
m
p
l
e
s
o
u
t
o
f
t
h
e
e
n
t
i
r
e
s
a
m
p
l
e
s
e
t
.
=
(
(
+
)
)
/
(
(
+
+
+
)
)
(
1)
P
r
e
c
i
s
i
o
n
:
r
e
pr
e
s
e
n
t
s
t
h
e
p
o
r
t
i
o
n
o
f
po
s
i
t
i
ve
i
de
n
t
i
f
i
c
a
t
i
o
ns
t
h
a
t
we
r
e
a
c
c
ur
a
t
e
l
y
m
a
de
.
=
/
(
(
+
)
)
(
2)
R
e
c
a
ll
:
d
e
m
o
ns
t
r
a
t
e
s
t
h
e
pr
o
p
or
t
i
o
n
o
f
t
r
ue
p
o
s
i
t
i
ve
s
t
h
a
t
we
r
e
c
o
r
r
e
c
t
l
y
r
e
c
o
gni
z
e
d.
=
/
(
(
+
)
)
(
3)
F
1
s
c
o
r
e
:
t
hi
s
b
a
l
a
n
c
e
d
m
e
t
r
i
c
f
u
n
c
t
i
o
n
s
a
s
t
h
e
h
a
r
m
o
ni
c
m
e
a
n
b
e
t
we
e
n
pr
e
c
i
s
i
o
n
a
n
d
r
e
c
a
ll
.
1
=
2
×
(
×
)
/
(
(
+
)
)
(
4)
W
h
e
r
e
T
P
s
t
a
n
ds
f
o
r
tr
ue
p
o
s
i
t
i
v
e
,
T
N
f
o
r
t
r
ue
n
e
ga
t
i
v
e
,
F
P
f
o
r
f
a
l
s
e
po
s
i
t
i
v
e
,
a
n
d
F
N
f
o
r
f
a
l
s
e
n
e
ga
t
i
v
e
.
3.
RE
S
UL
T
S
AN
D
DI
S
CU
S
S
I
ON
T
h
e
T
a
bl
e
1
pr
e
s
e
n
t
s
t
h
e
pe
r
f
o
r
m
a
nc
e
o
f
va
r
i
o
us
de
e
p
l
e
a
r
ni
ng
m
o
de
l
s
f
o
r
i
n
t
e
n
t
c
l
a
s
s
i
f
i
c
a
t
i
o
n
o
n
E
n
g
l
i
s
h
a
n
d
I
n
do
n
e
s
i
a
n
da
t
a
s
e
t
s
.
T
h
e
m
e
t
r
i
c
s
i
n
c
l
ude
a
c
c
ur
a
c
y
(
A
c
c
)
,
F
1
-
s
c
o
r
e
(
F
1)
,
pr
e
c
i
s
i
o
n
(
P
r
e
)
,
a
n
d
r
e
c
a
l
l
(
R
e
c
)
.
T
h
e
m
o
de
l
s
e
v
a
l
ua
t
e
d
a
r
e
B
E
R
T
,
C
NN
,
L
S
T
M
,
E
L
E
C
T
R
A
,
a
n
d
a
c
o
m
bi
na
t
i
o
n
o
f
E
L
E
C
T
R
A
a
n
d
B
E
R
T
.
T
a
bl
e
1.
R
e
s
u
l
t
s
o
f
a
c
c
ur
a
c
y
p
e
r
f
o
r
m
a
n
c
e
o
f
de
e
p
l
e
a
r
ni
ng
m
o
de
l
s
f
o
r
i
n
t
e
n
t
E
n
g
l
i
s
h
,
a
n
d
i
n
t
e
n
t
I
n
do
n
e
s
i
a
n
da
t
a
s
e
t
s
M
o
de
ls
D
a
ta
s
e
t
A
c
c
(
%
)
F
1 (
%
)
P
r
e
(
%
)
R
e
c
(
%
)
B
E
R
T
I
nt
e
nt
E
ngl
is
h
76.92
77.15
84.48
76.92
I
nt
e
nt
I
nd
o
ne
s
ia
n
96.79
96.55
97.38
96.79
C
N
N
I
nt
e
nt
E
ngl
is
h
71.79
69.87
79.57
71.79
I
nt
e
nt
I
nd
o
ne
s
ia
n
89.94
89.66
93.15
89.94
L
S
T
M
I
nt
e
nt
E
ngl
is
h
64.10
60.91
73.68
64.10
I
nt
e
nt
I
nd
o
ne
s
ia
n
90.55
90.17
93.07
90.55
E
L
E
C
T
R
A
I
nt
e
nt
E
ngl
is
h
98.92
53.61
55.31
57.44
I
nt
e
nt
I
nd
o
ne
s
ia
n
99.66
48.23
49.12
50.81
E
L
E
C
T
R
A
+
B
E
R
T
I
nt
e
nt
E
ngl
is
h
99.46
43.90
42.19
48.93
I
nt
e
nt
I
nd
o
ne
s
ia
n
99.66
48.74
49.66
51.63
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
do
n
e
s
i
a
n
J
E
l
e
c
E
n
g
&
C
o
m
p
S
c
i
I
S
S
N:
2
5
0
2
-
4
7
52
I
ntegr
ati
ng
E
L
E
C
T
R
A
and
B
E
R
T
mode
ls
in
tr
ans
for
me
r
-
bas
e
d
me
ntal
…
(
J
unta
Z
e
niar
ja
)
321
T
h
e
pe
r
f
o
r
m
a
n
c
e
r
e
s
u
l
t
s
o
f
va
r
i
o
us
de
e
p
l
e
a
r
ni
ng
m
o
de
l
s
f
o
r
i
n
t
e
n
t
c
l
a
s
s
i
f
i
c
a
t
i
o
n
s
h
o
w
t
h
a
t
B
E
R
T
de
m
o
ns
t
r
a
t
e
s
s
tr
o
n
g
a
n
d
b
a
l
a
nc
e
d
pe
r
f
o
r
m
a
nc
e
,
a
c
hi
e
vi
ng
76.
92%
a
c
c
ur
a
c
y
f
o
r
E
n
g
l
i
s
h
i
n
t
e
n
t
s
a
n
d
96.
79%
f
o
r
I
n
do
n
e
s
i
a
n
i
n
t
e
n
t
s
,
wi
t
h
hi
g
h
F
1
-
s
c
o
r
e
s
,
pr
e
c
i
s
i
o
n
,
a
n
d
r
e
c
a
ll
,
pa
r
t
i
c
u
l
a
r
ly
e
xc
e
ll
i
ng
i
n
t
h
e
I
ndo
n
e
s
i
a
n
da
t
a
s
e
t
wi
t
h
a
n
F
1
-
s
c
o
r
e
o
f
96.
55%
,
p
r
e
c
i
s
i
o
n
of
97.
38%
,
a
n
d
r
e
c
a
l
l
o
f
96.
79%
.
C
NN
,
whi
l
e
pe
r
f
o
r
m
i
ng
r
e
a
s
o
n
a
bly
w
e
l
l
,
f
a
ll
s
s
h
o
r
t
o
f
B
E
R
T
'
s
e
f
f
e
c
t
i
v
e
n
e
s
s
,
a
c
hi
e
vi
ng
71.
79%
a
c
c
ur
a
c
y
f
o
r
E
n
g
l
i
s
h
i
n
t
e
n
t
s
a
n
d
89.
94%
f
o
r
I
n
do
n
e
s
i
a
n
i
n
t
e
n
t
s
,
w
i
t
h
s
li
g
h
t
l
y
l
o
w
e
r
F
1
-
s
c
o
r
e
s
,
pr
e
c
i
s
i
o
n
,
a
n
d
r
e
c
a
ll
.
L
S
T
M
h
a
s
t
h
e
l
o
we
s
t
pe
r
f
o
r
m
a
n
c
e
f
o
r
E
n
g
l
i
s
h
i
n
t
e
n
t
s
a
t
64.
10%
a
c
c
ur
a
c
y
b
ut
i
m
pr
o
v
e
s
s
i
g
nif
i
c
a
n
t
l
y
f
o
r
I
n
do
n
e
s
i
a
n
i
n
t
e
n
t
s
w
i
t
h
90.
55%
a
c
c
ur
a
c
y
a
n
d
b
a
l
a
n
c
e
d
m
e
t
r
i
c
s
.
E
L
E
C
T
R
A
s
h
o
ws
ve
r
y
hi
g
h
a
c
c
ur
a
c
y
f
o
r
b
ot
h
E
n
g
l
i
s
h
(
98.
92%
)
a
n
d
I
n
do
n
e
s
i
a
n
(
99.
66%
)
i
n
t
e
n
t
s
,
b
ut
wi
t
h
n
o
t
a
bl
y
l
o
w
e
r
F
1
-
s
c
o
r
e
s
,
p
r
e
c
i
s
i
o
n
,
a
n
d
r
e
c
a
ll
,
pa
r
t
i
c
u
l
a
r
ly
f
o
r
E
n
g
l
i
s
h
i
n
t
e
n
t
s
(
F
1
-
s
c
o
r
e
o
f
53.
61%
,
pr
e
c
i
s
i
o
n
o
f
55.
31%
,
a
n
d
r
e
c
a
l
l
o
f
57.
44%
)
,
i
n
d
i
c
a
t
i
n
g
pot
e
n
t
i
a
l
i
s
s
ue
s
l
i
ke
c
l
a
s
s
im
ba
l
a
n
c
e
o
r
m
i
s
c
l
a
s
s
if
i
c
a
t
i
o
n
.
T
h
e
c
o
m
bi
na
t
i
o
n
o
f
E
L
E
C
T
R
A
a
n
d
B
E
R
T
a
c
hi
e
v
e
s
t
h
e
hi
g
h
e
s
t
a
c
c
ur
a
c
y
f
o
r
b
o
t
h
E
n
g
l
i
s
h
(
99.
46%
)
a
n
d
I
n
do
n
e
s
i
a
n
(
99.
66%
)
i
n
t
e
n
t
s
,
b
ut
wi
t
h
l
o
w
F
1
-
s
c
o
r
e
s
,
pr
e
c
i
s
i
o
n
,
a
nd
r
e
c
a
l
l
,
s
ugge
s
t
i
n
g
po
s
s
i
bl
e
o
v
e
r
f
i
t
t
i
n
g
o
r
ot
h
e
r
i
s
s
ue
s
.
T
his
un
de
r
s
c
o
r
e
s
t
h
e
n
e
e
d
t
o
c
o
n
s
i
de
r
a
l
l
pe
r
f
o
r
m
a
n
c
e
m
e
t
r
i
c
s
,
n
o
t
j
u
s
t
a
c
c
ur
a
c
y
,
wh
e
n
e
v
a
l
ua
t
i
n
g
m
o
de
l
pe
r
f
o
r
m
a
n
c
e
.
B
a
s
e
d
o
n
t
h
e
T
a
bl
e
2,
t
h
e
pe
r
f
o
r
m
a
n
c
e
o
f
de
e
p
l
e
a
r
ni
ng
m
o
de
l
s
f
o
r
i
n
t
e
n
t
c
l
a
s
s
i
f
i
c
a
t
i
o
n
s
h
o
ws
s
i
g
nif
i
c
a
n
t
i
m
pr
o
v
e
m
e
n
t
a
c
r
o
s
s
e
po
c
h
s
f
o
r
b
ot
h
E
n
g
l
i
s
h
a
n
d
I
n
do
n
e
s
i
a
n
da
t
a
s
e
t
s
.
A
t
50
e
p
o
c
h
s
,
th
e
m
o
de
l
e
xhi
b
i
t
s
l
o
w
a
c
c
ur
a
c
y
a
n
d
F
1
-
s
c
o
r
e
s
,
wi
t
h
E
n
g
l
is
h
i
n
t
e
n
t
s
a
t
34.
59%
a
c
c
ur
a
c
y
a
n
d
10.
02%
F
1
-
s
c
o
r
e
,
a
n
d
I
n
do
n
e
s
i
a
n
i
n
t
e
n
t
s
a
t
21.
38
%
a
c
c
ur
a
c
y
a
n
d
4.
35%
F
1
-
s
c
o
r
e
.
T
h
e
s
e
r
e
s
u
l
t
s
s
ugge
s
t
t
h
a
t
t
h
e
m
o
de
l
i
s
un
de
r
t
r
a
i
ne
d
a
n
d
u
n
a
bl
e
t
o
e
f
f
e
c
t
i
v
e
ly
c
l
a
s
s
if
y
i
n
t
e
n
t
s
a
t
t
hi
s
s
t
a
ge
.
T
a
bl
e
2.
R
e
s
u
l
t
s
o
f
a
c
c
ur
a
c
y
p
e
r
f
o
r
m
a
n
c
e
o
f
de
e
p
l
e
a
r
ni
ng
m
o
de
l
s
f
o
r
i
n
t
e
n
t
E
n
g
l
i
s
h
,
a
n
d
i
n
t
e
n
t
I
n
do
n
e
s
i
a
n
da
t
a
s
e
t
s
wi
t
h
d
i
f
f
e
r
e
n
t
E
p
o
c
h
E
poc
h
D
a
ta
s
e
t
A
c
c
(
%
)
F
1 (
%
)
P
r
e
(
%
)
R
e
c
(
%
)
50
I
nt
e
nt
E
ngl
is
h
34.59
10.02
9.34
12.76
I
nt
e
nt
I
nd
o
ne
s
ia
n
21.38
4.35
3.82
8.69
100
I
nt
e
nt
E
ngl
is
h
49.19
13.09
11.87
17.02
I
nt
e
nt
I
nd
o
ne
s
ia
n
85.43
30.32
29.37
35.05
150
I
nt
e
nt
E
ngl
is
h
81.62
34.62
37.58
36.17
I
nt
e
nt
I
nd
o
ne
s
ia
n
98.03
43.29
44.33
45.38
200
I
nt
e
nt
E
ngl
is
h
93.51
38.15
40.14
42.55
I
nt
e
nt
I
nd
o
ne
s
ia
n
99.18
45.18
45.24
48.09
250
I
nt
e
nt
E
ngl
is
h
97.84
44.32
43.97
48.93
I
nt
e
nt
I
nd
o
ne
s
ia
n
99.52
49.17
50.73
51.35
300
I
nt
e
nt
E
ngl
is
h
99.46
43.90
42.19
48.93
I
nt
e
nt
I
nd
o
ne
s
ia
n
99.66
49.94
50.02
52.71
B
y
100
e
po
c
h
s
,
t
h
e
m
o
de
l
s
h
o
ws
n
o
t
i
c
e
a
bl
e
im
p
r
o
v
e
m
e
n
t
,
wi
t
h
t
h
e
I
n
do
n
e
s
i
a
n
da
t
a
s
e
t
a
c
hi
e
vi
n
g
85.
43%
a
c
c
ur
a
c
y
a
n
d
a
30.
32%
F
1
-
s
c
o
r
e
,
a
n
d
t
he
E
n
g
li
s
h
da
t
a
s
e
t
a
c
hi
e
vi
ng
49.
19%
a
c
c
ur
a
c
y
a
n
d
13.
09%
F1
-
s
c
o
r
e
.
A
t
150
e
po
c
h
s
,
pe
r
f
o
r
m
a
n
c
e
f
ur
t
h
e
r
i
n
c
r
e
a
s
e
s
,
w
i
t
h
t
h
e
E
n
g
li
s
h
da
t
a
s
e
t
r
e
a
c
hi
ng
81.
62%
a
c
c
ur
a
c
y
a
n
d
a
34.
62%
F
1
-
s
c
o
r
e
,
a
n
d
t
h
e
I
n
do
n
e
s
i
a
n
da
t
a
s
e
t
a
c
hi
e
vi
ng
98.
03%
a
c
c
ur
a
c
y
a
n
d
a
43.
29%
F
1
-
s
c
o
r
e
.
T
hi
s
t
r
e
n
d
c
o
n
t
i
n
ue
s
a
t
200
e
p
o
c
h
s
,
w
i
t
h
t
h
e
E
n
g
l
i
s
h
d
a
t
a
s
e
t
a
t
93
.
51%
a
c
c
ur
a
c
y
a
n
d
a
38.
15%
F
1
-
s
c
o
r
e
,
a
n
d
t
h
e
I
n
do
n
e
s
i
a
n
da
t
a
s
e
t
a
t
99.
18%
a
c
c
ur
a
c
y
a
n
d
a
45.
18%
F
1
-
s
c
o
r
e
.
B
y
250
e
po
c
hs
,
t
h
e
m
o
de
l
r
e
a
c
he
s
hi
g
h
pe
r
f
o
r
m
a
n
c
e
,
wi
t
h
t
h
e
E
n
g
l
i
s
h
da
t
a
s
e
t
a
t
97
.
84%
a
c
c
ur
a
c
y
a
n
d
a
44.
32%
F
1
-
s
c
o
r
e
,
a
n
d
t
h
e
I
n
do
ne
s
i
a
n
da
t
a
s
e
t
a
t
99
.
52%
a
c
c
ur
a
c
y
a
n
d
a
49.
17%
F
1
-
s
c
o
r
e
.
A
t
3
00
e
po
c
h
s
,
t
h
e
hi
g
h
e
s
t
pe
r
f
o
r
m
a
n
c
e
i
s
o
b
s
e
r
v
e
d,
w
i
t
h
t
h
e
E
n
g
l
i
s
h
da
t
a
s
e
t
a
c
hi
e
vi
ng
99.
46%
a
c
c
ur
a
c
y
a
n
d
a
43.
90%
F
1
-
s
c
o
r
e
,
a
n
d
t
h
e
I
n
do
n
e
s
i
a
n
da
t
a
s
e
t
a
c
hi
e
vi
ng
99.
66%
a
c
c
ur
a
c
y
a
n
d
a
49.
94%
F
1
-
s
c
o
r
e
.
T
h
e
s
e
r
e
s
u
l
t
s
i
n
d
i
c
a
t
e
t
h
a
t
i
n
c
r
e
a
s
i
ng
t
h
e
n
u
m
be
r
o
f
t
r
a
i
ni
ng
e
po
c
h
s
s
i
g
nif
i
c
a
n
t
l
y
e
nh
a
n
c
e
s
t
h
e
m
o
de
l
'
s
pe
r
f
o
r
m
a
n
c
e
f
o
r
i
n
t
e
n
t
c
l
a
s
s
if
i
c
a
t
i
o
n
,
w
i
t
h
t
h
e
m
o
s
t
s
ub
s
t
a
n
t
i
a
l
ga
i
ns
o
b
s
e
r
v
e
d
up
to
250
e
po
c
h
s
,
b
e
y
o
n
d
w
hi
c
h
a
c
c
ur
a
c
y
i
m
pr
o
v
e
m
e
n
t
s
a
r
e
m
a
r
g
i
n
a
l
.
T
h
e
r
e
s
u
l
t
s
v
a
li
da
t
e
t
h
e
i
ni
t
i
a
l
hy
po
t
h
e
s
i
s
t
h
a
t
i
nc
r
e
a
s
i
ng
t
h
e
n
u
m
be
r
o
f
t
r
a
i
ni
ng
e
po
c
hs
e
nh
a
n
c
e
s
m
o
de
l
pe
r
f
o
r
m
a
n
c
e
s
i
g
ni
f
i
c
a
n
t
l
y
.
A
t
50
e
po
c
h
s
,
a
c
c
ur
a
c
y
wa
s
34.
59%
f
o
r
E
n
g
l
i
s
h
a
n
d
21.
38%
f
o
r
I
n
do
n
e
s
i
a
n
i
n
t
e
n
t
s
,
b
ut
by
300
e
p
o
c
h
s
,
a
c
c
ur
a
c
y
i
m
pr
o
v
e
d
to
99.
46
%
f
o
r
E
n
g
l
i
s
h
a
n
d
99.
66%
f
o
r
I
n
do
n
e
s
ian
.
T
hi
s
c
o
nf
i
r
m
s
s
u
b
s
t
a
n
t
i
a
l
ga
i
ns
up
t
o
250
e
p
o
c
h
s
.
T
h
e
hy
po
t
h
e
s
i
s
t
h
a
t
t
r
a
n
s
f
o
r
m
e
r
-
b
a
s
e
d
m
o
de
l
s
li
ke
B
E
R
T
a
n
d
E
L
E
C
T
R
A
wo
u
l
d
o
u
t
pe
r
f
o
r
m
o
t
h
e
r
s
wa
s
pa
r
t
l
y
c
o
nf
i
r
m
e
d
;
E
L
E
C
T
R
A
a
c
hi
e
v
e
d
hi
g
h
a
c
c
ur
a
c
y
(
98.
92%
f
o
r
E
n
g
l
i
s
h
a
n
d
99.
66%
f
o
r
I
n
do
n
e
s
i
a
n
)
b
ut
h
a
d
l
o
we
r
F
1
-
s
c
o
r
e
s
,
pr
e
c
i
s
i
o
n
,
a
n
d
r
e
c
a
ll
,
i
nd
i
c
a
t
i
n
g
po
s
s
i
bl
e
b
a
l
a
n
c
e
i
s
s
u
e
s
.
B
E
R
T
s
h
o
we
d
s
t
r
o
n
g
pe
r
f
o
r
m
a
n
c
e
,
e
s
pe
c
i
a
l
l
y
w
i
t
h
I
n
do
n
e
s
i
a
n
i
n
t
e
n
t
s
(
96.
79%
a
c
c
ur
a
c
y
a
n
d
96.
55%
F
1
-
s
c
or
e
)
.
T
h
e
c
o
m
bi
na
t
i
o
n
o
f
E
L
E
C
T
R
A
a
n
d
B
E
R
T
a
c
hi
e
v
e
d
t
h
e
hi
g
h
e
s
t
a
c
c
ur
a
c
y
b
ut
wi
t
h
l
o
we
r
F1
-
s
c
o
r
e
s
,
s
ugge
s
t
i
n
g
pot
e
n
t
i
a
l
o
v
e
r
f
i
t
t
i
n
g.
T
h
e
f
in
d
i
ngs
a
l
s
o
e
m
p
h
a
s
i
z
e
t
h
a
t
r
e
l
yi
ng
s
o
l
e
ly
o
n
a
c
c
ur
a
c
y
c
a
n
b
e
m
i
s
l
e
a
d
i
ng,
a
n
d
m
u
l
t
i
p
l
e
m
e
t
r
i
c
s
a
r
e
c
r
uc
i
a
l
f
o
r
a
b
a
l
a
n
c
e
d
e
v
a
l
ua
t
i
o
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2
5
0
2
-
4
7
52
In
do
n
e
s
i
a
n
J
E
l
e
c
E
n
g
&
C
o
m
p
S
c
i
,
Vo
l
.
3
7
,
N
o
.
1
,
J
a
n
ua
r
y
20
2
5
:
3
1
5
-
324
322
4.
CONC
L
USI
ON
T
h
e
r
e
s
u
l
t
s
o
f
t
hi
s
r
e
s
e
a
r
c
h
hi
g
hli
g
h
t
s
i
g
ni
f
i
c
a
n
t
a
dv
a
n
c
e
m
e
n
t
s
i
n
t
h
e
h
e
a
l
t
h
s
e
c
t
o
r
a
n
d
r
e
l
a
t
e
d
s
c
i
e
n
t
i
f
i
c
c
o
m
m
u
ni
t
i
e
s
by
de
m
o
n
s
t
r
a
t
i
n
g
t
h
a
t
c
h
a
tb
o
t
s
us
i
n
g
t
h
e
E
L
E
C
T
R
A
a
n
d
B
E
R
T
t
r
a
n
s
f
o
r
m
e
r
m
e
t
h
o
ds
c
a
n
pr
o
vi
de
hi
g
hly
a
c
c
ur
a
t
e
,
e
f
f
e
c
t
i
ve
,
a
n
d
e
m
pa
t
h
e
t
i
c
s
uppo
r
t
f
o
r
m
e
n
t
a
l
he
a
l
t
h
i
s
s
ue
s
.
W
i
t
h
a
n
i
m
pr
e
s
s
i
ve
a
c
c
ur
a
c
y
o
f
99.
66%
a
c
r
o
s
s
E
n
g
l
i
s
h
a
n
d
I
n
do
ne
s
i
a
n
da
t
a
s
e
t
s
,
t
h
e
s
e
c
ha
t
b
ot
s
o
f
f
e
r
v
a
l
ua
bl
e
24/7
s
up
po
r
t
a
n
d
t
i
m
e
ly
r
e
s
po
n
s
e
s
,
e
nh
a
n
c
i
ng
a
c
c
e
s
s
i
b
il
i
t
y
t
o
m
e
n
t
a
l
he
a
l
t
h
r
e
s
o
ur
c
e
s
a
n
d
pot
e
n
t
i
a
ll
y
r
e
duc
i
ng
t
h
e
bur
de
n
o
n
h
u
m
a
n
pr
o
f
e
s
s
i
o
na
l
s
.
T
hi
s
a
d
va
n
c
e
m
e
n
t
i
s
pa
r
t
i
c
u
l
a
r
ly
b
e
n
e
f
i
c
i
a
l
i
n
e
m
e
r
ge
n
c
y
s
i
t
ua
t
i
o
n
s
,
w
h
e
r
e
im
m
e
d
i
a
t
e
a
s
s
i
s
t
a
n
c
e
i
s
c
r
i
t
i
c
a
l
.
M
o
r
e
o
v
e
r
,
t
h
e
s
t
udy
unde
r
s
c
o
r
e
s
t
h
e
p
ot
e
n
t
i
a
l
o
f
i
n
t
e
gr
a
t
i
n
g
c
utt
i
n
g
-
e
dge
A
I
t
e
c
h
n
o
l
o
g
i
e
s
i
n
t
o
m
e
n
t
a
l
h
e
a
l
t
h
c
a
r
e
,
c
o
n
t
r
i
b
ut
i
n
g
to
a
de
e
pe
r
un
de
r
s
t
a
n
d
i
ng
o
f
t
h
e
i
r
pr
a
c
t
i
c
a
l
a
pp
li
c
a
t
i
o
ns
a
n
d
f
o
s
t
e
r
i
n
g
f
ur
t
h
e
r
r
e
s
e
a
r
c
h
i
n
t
h
e
f
i
e
l
d.
T
h
e
f
i
nd
i
ng
s
s
ugge
s
t
t
h
a
t
t
h
e
s
e
A
I
m
o
de
l
s
c
a
n
c
o
m
p
l
e
m
e
n
t
t
r
a
di
t
i
o
na
l
m
e
n
t
a
l
h
e
a
l
t
h
s
e
r
vi
c
e
s
by
pr
o
vi
d
i
ng
pr
e
l
im
i
na
r
y
s
uppo
r
t
a
n
d
di
r
e
c
t
i
n
g
us
e
r
s
to
pr
o
f
e
s
s
i
o
n
a
l
h
e
l
p
wh
e
n
n
e
e
de
d,
t
h
e
r
e
by
e
n
r
i
c
hi
ng
t
h
e
o
v
e
r
a
l
l
m
e
n
t
a
l
h
e
a
l
t
h
c
a
r
e
l
a
n
ds
c
a
pe
.
T
hi
s
s
t
ud
y
de
m
o
n
s
t
r
a
t
e
d
t
h
e
e
f
f
e
c
t
i
v
e
n
e
s
s
o
f
us
i
n
g
E
L
E
C
T
R
A
a
n
d
B
E
R
T
t
r
a
n
s
f
o
r
m
e
r
m
e
t
h
o
ds
i
n
m
e
n
t
a
l
h
e
a
l
t
h
c
h
a
t
b
o
t
s
,
a
c
hi
e
vi
ng
hi
g
h
a
c
c
ur
a
c
y
a
n
d
pr
o
vi
d
i
ng
v
a
l
ua
bl
e
s
uppo
r
t
i
n
E
n
g
l
i
s
h
a
n
d
I
n
d
o
n
e
s
i
a
n
.
Ho
we
v
e
r
,
f
ur
t
h
e
r
r
e
s
e
a
r
c
h
i
s
ne
e
de
d
t
o
a
ddr
e
s
s
l
im
i
t
a
t
i
o
n
s
s
u
c
h
a
s
t
h
e
m
o
de
l
s
'
p
e
r
f
o
r
m
a
n
c
e
a
c
r
o
s
s
o
t
h
e
r
l
a
n
gua
ge
s
a
n
d
c
u
l
t
ur
a
l
ly
d
i
v
e
r
s
e
c
o
n
t
e
x
t
s
.
A
dd
i
t
i
o
n
a
ll
y
,
e
t
hi
c
a
l
c
o
n
c
e
r
ns
,
i
n
c
l
ud
i
ng
da
t
a
p
r
i
v
a
c
y
a
n
d
t
h
e
a
bs
e
n
c
e
o
f
h
u
m
a
n
e
m
p
a
t
hy
i
n
A
I
s
o
l
ut
i
o
n
s
,
we
r
e
n
o
t
di
s
c
us
s
e
d.
F
ut
ur
e
r
e
s
e
a
r
c
h
c
o
ul
d
e
x
p
l
o
r
e
t
h
e
a
da
pt
a
t
i
o
n
o
f
t
h
e
s
e
m
o
de
l
s
to
a
ddi
t
i
o
na
l
l
a
n
gua
ge
s
a
n
d
c
u
l
t
ur
a
l
l
y
d
i
ve
r
s
e
c
o
n
t
e
x
t
s
to
b
r
o
a
de
n
t
h
e
i
r
a
pp
l
i
c
a
bil
i
t
y
.
A
dd
i
t
i
o
n
a
ll
y
,
i
nve
s
t
i
ga
t
i
n
g
wa
y
s
t
o
e
n
h
a
n
c
e
t
h
e
e
t
hi
c
a
l
a
s
pe
c
t
s
o
f
A
I
-
dr
i
v
e
n
s
o
l
ut
i
o
ns
,
s
uc
h
a
s
i
m
pr
o
vi
ng
da
t
a
p
r
i
va
c
y
a
n
d
i
n
t
e
gr
a
t
i
n
g
e
m
pa
t
h
e
t
i
c
r
e
s
po
ns
e
s
,
c
o
ul
d
a
ddr
e
s
s
c
ur
r
e
n
t
l
im
i
t
a
t
i
o
n
s
.
AC
K
NOWL
E
DGE
M
E
NT
S
T
h
e
a
ut
h
o
r
w
o
ul
d
li
ke
t
o
e
x
pr
e
s
s
s
i
nc
e
r
e
gr
a
t
i
t
ude
to
t
h
e
R
e
s
e
a
r
c
h
a
n
d
C
o
m
m
u
ni
t
y
S
e
r
vi
c
e
I
n
s
t
i
t
ut
e
(
L
P
P
M
)
a
t
Uni
ve
r
s
i
t
a
s
D
i
a
n
Nus
w
a
n
to
r
o
f
o
r
t
h
e
i
r
i
nva
l
ua
bl
e
s
uppo
r
t
.
T
hi
s
r
e
s
e
a
r
c
h
wa
s
f
u
n
de
d
by
t
h
e
B
a
s
i
c
R
e
s
e
a
r
c
h
Gr
a
n
t
S
c
h
e
m
e
u
n
de
r
c
o
n
t
r
a
c
t
n
o.
109/A
.
38
-
0
4/UDN
-
09/XI
/2023.
RE
F
E
R
E
NC
E
S
[
1]
A
.
L
i,
X
.
H
ua
ng,
D
.
J
ia
o
,
B
.
O
’
D
e
a
,
T
.
Z
hu,
a
nd
H
.
C
hr
is
t
e
ns
e
n,
“
A
n
a
na
l
y
s
is
of
s
ti
gma
a
nd
s
ui
c
id
e
li
t
e
r
a
c
y
in
r
e
s
p
o
ns
e
s
to
s
ui
c
id
e
s
br
o
a
d
c
a
s
t
o
n s
o
c
ia
l
me
di
a
,”
A
s
ia
-
P
ac
if
ic
P
s
y
c
hi
at
r
y
, vo
l.
10, n
o
. 1, M
a
r
. 2018, d
o
i:
10.1111/app
y
.12314.
[
2]
“
M
e
nt
a
l
he
a
lt
h
a
c
ti
o
n
p
la
n
2013
-
2020,”
W
or
ld
H
e
al
th
O
r
gani
z
at
io
n
.
[
O
nl
in
e
]
.
A
v
a
il
a
bl
e
:
ht
tp
:/
/ww
w
.w
ho
.i
nt
/
ir
is
/h
a
ndl
e
/1
0665
/8
9966.
[
3]
J
.
M
.
G
ór
r
i
z
e
t
al
.
,
“
C
o
mput
a
ti
o
na
l
a
ppr
o
a
c
h
e
s
to
e
x
p
la
in
a
bl
e
a
r
ti
f
ic
ia
l
in
te
ll
ig
e
n
c
e
:
a
dv
a
nc
e
s
in
th
e
o
r
y
,
a
ppl
ic
a
ti
o
ns
a
nd
tr
e
n
ds
,”
I
nf
or
m
at
io
n F
us
io
n
, vo
l.
100, p. 101945, D
e
c
. 2023, d
o
i:
10.10
16/
j.
in
f
f
us
.2023.101945.
[
4]
M
.
K
.
R
us
ia
a
nd
D
.
K
.
S
in
gh,
“
A
c
o
l
o
r
-
te
xt
ur
e
-
ba
s
e
d
d
e
e
p
n
e
u
r
a
l
ne
tw
o
r
k
t
e
c
hni
qu
e
t
o
d
e
te
c
t
f
a
c
e
s
poof
in
g
a
tt
a
c
ks
,”
C
y
be
r
ne
ti
c
s
and I
nf
or
m
at
io
n T
e
c
hnol
ogi
e
s
, v
ol
. 22, n
o
. 3, pp. 127
–
145, S
e
p
. 2022, do
i:
10.2478/
c
a
it
-
2022
-
0032.
[
5]
B
.
P
.
B
a
bu
a
nd
S
.
J
.
N
a
r
a
y
a
na
n,
“
O
ne
-
vs
-
a
ll
c
o
n
vo
lu
ti
o
na
l
ne
ur
a
l
ne
tw
or
ks
f
o
r
s
y
nt
h
e
ti
c
a
pe
r
tu
r
e
r
a
da
r
ta
r
ge
t
r
e
c
o
g
ni
t
i
o
n,”
C
y
be
r
ne
ti
c
s
and I
n
f
or
m
at
io
n T
e
c
hnol
ogi
e
s
, v
ol
. 22, n
o
. 3, pp. 1
79
–
197, S
e
p. 2022, d
o
i:
10.2478/
c
a
it
-
2022
-
0035.
[
6]
V
.
B
a
r
o
t
a
nd
V
.
K
a
pa
di
a
,
“
L
o
ng
s
h
o
r
t
t
e
r
m
m
e
m
or
y
n
e
ur
a
l
ne
tw
o
r
k
-
ba
s
e
d
m
o
d
e
l
c
o
ns
tr
u
c
ti
o
n
a
nd
f
n
e
-
tu
ni
ng
f
or
a
ir
qua
li
t
y
pa
r
a
me
t
e
r
s
pr
e
di
c
ti
o
n,”
C
y
be
r
ne
ti
c
s
and
I
nf
o
r
m
at
io
n
T
e
c
hnol
ogi
e
s
,
vo
l.
22,
n
o
.
1,
pp.
171
–
189,
M
a
r
.
2022,
d
o
i:
10.2478/
c
a
it
-
2022
-
0011.
[
7]
L
.
K
uma
r
a
nd
D
.
K
.
S
in
gh,
“
H
a
r
dw
a
r
e
r
e
s
po
ns
e
a
nd
pe
r
f
o
r
ma
nc
e
a
na
l
y
s
is
of
mul
ti
c
or
e
c
o
mput
in
g
s
y
s
t
e
ms
f
or
de
e
p
le
a
r
ni
ng
a
lg
o
r
it
h
ms
,”
C
y
be
r
n
e
ti
c
s
and I
n
f
or
m
at
io
n T
e
c
hnol
ogi
e
s
, v
o
l.
2
2, no
. 3, pp. 68
–
81, S
e
p. 2022, d
o
i:
10.2478/
c
a
it
-
2022
-
0028.
[
8]
S
.
V
a
ndha
na
a
nd
J
.
A
nur
a
dha
,
“
S
pa
ti
a
l
a
nd
t
e
mp
o
r
a
l
v
a
r
ia
t
io
ns
o
n
a
ir
qua
li
t
y
pr
e
di
c
ti
o
n
us
in
g
d
e
e
p
le
a
r
ni
ng
t
e
c
hni
qu
e
s
,”
C
y
be
r
ne
ti
c
s
an
d I
n
f
or
m
at
io
n T
e
c
hnol
ogi
e
s
, v
ol
. 23, n
o
. 4, pp. 2
13
–
232, Nov
. 2023, d
o
i:
10.2478/
c
a
it
-
2023
-
0045.
[
9]
M
.
S
o
or
i,
B
.
A
r
e
z
oo
,
a
nd
R
.
D
a
s
tr
e
s
,
“
A
r
ti
f
i
c
ia
l
in
t
e
ll
ig
e
n
c
e
,
m
a
c
hi
ne
l
e
a
r
ni
ng
a
nd
de
e
p
le
a
r
ni
ng
in
a
d
v
a
nc
e
d
r
o
b
o
ti
c
s
,
a
r
e
v
i
e
w
,”
C
ogni
ti
v
e
R
obot
ic
s
, vo
l.
3, pp. 54
–
70, 2023, d
o
i:
10.1016/j
.
c
o
gr
.2023.04.001.
[
10]
A
.
D
.
R
e
be
l
o
,
D
.
E
.
V
e
r
b
oo
m,
N
.
R
.
do
s
S
a
nt
o
s
,
a
nd
J
.
W
.
de
G
r
a
a
f
,
“
T
h
e
im
pa
c
t
of
a
r
ti
f
i
c
ia
l
in
te
l
li
ge
n
c
e
o
n
th
e
ta
s
ks
of
m
e
n
ta
l
he
a
lt
hc
a
r
e
w
o
r
k
e
r
s
:
A
s
c
o
pi
ng
r
e
vi
e
w
,”
C
om
put
e
r
s
in
H
um
an
B
e
hav
io
r
:
A
r
ti
f
ic
ia
l
H
um
ans
,
vo
l.
1,
n
o
.
2,
p.
100008,
A
ug.
20
23,
do
i:
10.1016/j
.
c
hba
h.2023.100008.
[
11]
T
.
F
u,
S
.
G
a
o
,
X
.
Z
ha
o
,
J
.
W
e
n,
a
nd
R
.
Y
a
n,
“
L
e
a
r
ni
ng
t
o
w
a
r
ds
c
o
n
ve
r
s
a
ti
o
na
l
A
I
:
a
s
ur
ve
y
,”
A
I
O
pe
n
,
vol
.
3,
pp.
14
–
28,
2
022,
do
i:
10.1016/j
.a
i
o
p
e
n.2022.02.001.
[
12]
P
.
M
.
M
a
h,
I
.
S
ka
ln
a
,
a
nd J
. M
uz
a
m,
“
N
a
tu
r
a
l
la
ngu
a
ge
pr
oc
e
s
s
in
g
a
nd a
r
ti
f
i
c
ia
l
in
t
e
ll
ig
e
n
c
e
f
o
r
e
nt
e
r
pr
is
e
ma
na
ge
m
e
nt
in
th
e
e
r
a
of
I
ndus
tr
y
4.0,”
A
ppl
ie
d Sc
ie
nc
e
s
, vo
l.
12, n
o
. 18, p. 9207, S
e
p
. 2022, do
i:
10.3390/app121892
07.
[
13]
E
.
A
da
mo
p
o
ul
o
u
a
nd
L
.
M
o
us
s
ia
de
s
,
“
C
ha
tb
o
ts
:
h
is
to
r
y
,
t
e
c
hn
o
l
o
g
y
,
a
nd
a
ppl
ic
a
ti
o
ns
,”
M
ac
hi
ne
L
e
ar
ni
ng
w
it
h
A
ppl
ic
at
io
ns
, vo
l.
2, p. 100006, De
c
. 2020, d
oi
:
10.1016/j
.ml
w
a
.2020.100006.
[
14]
S
.
P
a
nde
y
a
nd
S
.
S
ha
r
ma
,
“
A
c
o
mpa
r
a
ti
v
e
s
tu
d
y
of
r
e
t
r
i
e
va
l
-
ba
s
e
d
a
n
d
ge
ne
r
a
ti
v
e
-
ba
s
e
d
c
ha
tb
ot
s
us
in
g
de
e
p
l
e
a
r
ni
ng
a
nd
ma
c
hi
ne
le
a
r
n
in
g
,”
H
e
al
th
c
ar
e
A
nal
y
ti
c
s
, v
o
l.
3, p. 100198, N
ov
. 2023, d
o
i:
10.1016/j
.h
e
a
lt
h.2023.100198.
[
15]
R
.
S
im
a
r
ma
ta
,
J
.
K
r
is
ta
nt
o
,
a
nd
A
.
C
ho
w
a
nda
,
“
U
ti
li
z
in
g
tr
a
ns
f
o
r
m
e
r
-
ba
s
e
d
d
e
e
p
l
e
a
r
ni
ng
f
or
i
nt
e
nt
c
la
s
s
if
ic
a
ti
o
n
o
n
t
e
x
t,
”
J
our
nal
of
T
he
o
r
e
ti
c
al
and A
ppl
ie
d I
nf
o
r
m
at
io
n T
e
c
hnol
ogy
, v
ol
. 101, no
. 13, pp. 5078
–
5084, 2023.
[
16]
S
.
Y
u,
Y
.
C
he
n,
a
nd
H
.
Z
a
id
i,
“
A
V
A
:
a
f
in
a
n
c
ia
l
s
e
r
v
i
c
e
c
ha
tb
o
t
ba
s
e
d
o
n
d
e
e
p
bi
di
r
e
c
ti
o
na
l
t
r
a
n
s
f
or
me
r
s
,”
F
r
ont
ie
r
s
in
A
p
pl
ie
d
M
at
he
m
at
ic
s
and Stat
is
ti
c
s
, v
o
l.
7, A
ug. 2021, d
o
i:
10.3389/
f
a
ms
.2021.604842.
[
17]
R
.
F
a
ti
ma
e
t
al
.
,
“
A
na
tu
r
a
l
la
ngua
ge
pr
o
c
e
s
s
in
g
(
N
L
P
)
e
v
a
lu
a
t
io
n
o
n
C
O
V
I
D
-
19
r
umo
u
r
da
ta
s
e
t
us
in
g
de
e
p
le
a
r
ni
ng
t
e
c
hn
iq
u
e
s
,”
C
om
put
at
io
nal
I
nt
e
ll
ig
e
nc
e
and N
e
ur
o
s
c
ie
n
ce
, v
o
l.
2022, pp. 1
–
17, S
e
p. 2022, d
o
i:
10.1155/2022/
6561622.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
do
n
e
s
i
a
n
J
E
l
e
c
E
n
g
&
C
o
m
p
S
c
i
I
S
S
N:
2
5
0
2
-
4
7
52
I
ntegr
ati
ng
E
L
E
C
T
R
A
and
B
E
R
T
mode
ls
in
tr
ans
for
me
r
-
bas
e
d
me
ntal
…
(
J
unta
Z
e
niar
ja
)
323
[
18]
K
.
C
la
r
k,
M
.
T
.
L
u
o
ng,
Q
.
V
.
L
e
,
a
nd
C
.
D
.
M
a
nni
ng,
“
E
le
c
tr
a
:
pr
e
-
tr
a
in
in
g
te
x
t
e
n
c
o
d
e
r
s
a
s
di
s
c
r
im
in
a
to
r
s
r
a
th
e
r
t
ha
n
ge
n
e
r
a
t
o
r
s
,”
8t
h I
nt
e
r
nat
io
nal
C
onf
e
r
e
nc
e
on L
e
a
r
ni
ng
R
e
pr
e
s
e
nt
at
io
ns
, I
C
L
R
2020
, 2020.
[
19]
S
.
A
bur
a
s
s
,
O
.
D
o
r
gha
m,
a
nd
M
.
A
.
R
umm
a
n,
“
A
n
e
ns
e
mbl
e
a
ppr
o
a
c
h
to
qu
e
s
ti
o
n
c
la
s
s
if
i
c
a
ti
o
n:
in
t
e
gr
a
ti
ng
e
le
c
tr
a
tr
a
ns
f
o
r
me
r
,
G
l
o
V
e
,
a
nd
L
S
T
M
,”
I
nt
e
r
nat
io
nal
J
our
nal
of
A
dv
anc
e
d
C
om
p
ut
e
r
Sc
ie
nc
e
and
A
ppl
ic
at
io
ns
,
v
o
l.
15,
n
o
.
1,
pp.
507
–
514,
20
24,
do
i:
10.14569/
I
J
A
C
S
A
.2024.0150148.
[
20]
S
.
S
a
r
ka
r
,
M
.
G
a
u
r
,
L
.
K
.
C
h
e
n,
M
.
G
a
r
g,
a
nd
B
.
S
r
i
v
a
s
ta
v
a
,
“
A
r
e
v
i
e
w
of
th
e
e
x
pl
a
in
a
bi
li
t
y
a
nd
s
a
f
e
t
y
of
c
o
n
ve
r
s
a
ti
o
na
l
a
ge
n
ts
f
or
me
nt
a
l
h
e
a
lt
h
t
o
id
e
nt
i
f
y
a
ve
nu
e
s
f
o
r
im
pr
ove
m
e
nt
,”
Fr
ont
ie
r
s
in
A
r
ti
f
ic
ia
l
I
nt
e
ll
ig
e
nc
e
,
v
ol
.
6,
O
c
t.
2023,
do
i:
10.3389/f
r
a
i.
2023.1229805.
[
21]
A
.
V
a
s
w
a
ni
e
t
al
.
,
“
A
tt
e
nt
io
n
is
a
ll
y
o
u
ne
e
d,”
A
dv
anc
e
s
in
N
e
ur
al
I
nf
or
m
at
io
n
P
r
oc
e
s
s
in
g
Sy
s
te
m
s
,
v
o
l.
2017
-
D
e
c
e
m,
n
o
.
N
ip
s
,
pp. 5999
–
6009, 2017.
[
22]
L
.
B
a
lc
o
mb
e
,
“
A
I
c
ha
tb
o
ts
in
d
ig
it
a
l
m
e
nt
a
l
he
a
lt
h,
”
I
nf
or
m
at
ic
s
,
v
o
l.
10,
n
o
.
4,
p.
82,
O
c
t.
2023,
do
i:
10.3390/i
n
f
o
r
ma
ti
c
s
10040082.
[
23]
G
.
C
a
ld
a
r
in
i,
S
.
J
a
f
,
a
nd
K
.
M
c
G
a
r
r
y
,
“
A
li
t
e
r
a
tu
r
e
s
ur
ve
y
of
r
e
c
e
nt
a
dv
a
n
c
e
s
in
c
ha
tb
o
ts
,”
I
nf
or
m
at
io
n
,
vo
l.
13,
n
o
.
1,
p.
41,
J
a
n.
2022, do
i:
10.3390/i
n
f
o13010041.
[
24]
A
.
R
a
ha
li
a
nd
M
.
A
.
A
khl
o
u
f
i,
“
E
nd
-
to
-
e
nd
tr
a
ns
f
o
r
m
e
r
-
ba
s
e
d
m
o
de
ls
in
t
e
x
tu
a
l
-
ba
s
e
d
N
L
P
,”
AI
,
vo
l.
4,
n
o
.
1,
pp.
54
–
110,
J
a
n.
2023, do
i:
10.3390/ai4010004.
[
25]
S
. K
ha
n
e
t
al
.
, “
T
r
a
ns
f
or
me
r
a
r
c
hi
t
e
c
tu
r
e
-
ba
s
e
d
tr
a
ns
f
e
r
l
e
a
r
ni
n
g
f
o
r
p
o
li
te
n
e
s
s
pr
e
di
c
ti
o
n i
n
c
o
n
ve
r
s
a
ti
o
n,”
Sus
ta
in
abi
li
ty
,
vo
l.
15,
no
. 14, p. 10828, J
ul
. 2023, d
o
i:
10.3390/s
u151410828.
[
26]
R
.
S
udhe
e
s
h
e
t
al
.
,
“
A
na
l
y
z
in
g
s
e
nt
im
e
n
ts
r
e
ga
r
d
in
g
C
ha
tG
P
T
us
in
g
nove
l
B
E
R
T
:
a
ma
c
hi
n
e
l
e
a
r
ni
ng
a
p
pr
o
a
c
h,”
I
nf
or
m
at
io
n
,
vo
l.
14, n
o
. 9, p. 474, Aug. 202
3, d
o
i:
10.3390/i
n
f
o14090474.
[
27]
T
.
N
.
A
lr
uqi
a
nd
S
.
M
.
A
lz
a
hr
a
ni
,
“
E
v
a
lu
a
ti
o
n
of
a
n
A
r
a
bi
c
c
h
a
tb
o
t
ba
s
e
d
o
n
e
xt
r
a
c
ti
v
e
qu
e
s
ti
o
n
-
a
ns
w
e
r
in
g
tr
a
ns
f
e
r
le
a
r
ni
ng
a
nd
la
ngua
ge
t
r
a
ns
f
or
m
e
r
s
,”
AI
,
vo
l.
4, n
o
. 3,
pp. 667
–
691, Aug. 20
23, do
i:
10.3390/ai4030035.
[
28]
H
.
F
a
ng,
G
.
X
u,
Y
.
L
o
ng,
a
nd
W
.
T
a
ng,
“
A
n
e
f
f
e
c
t
i
v
e
E
L
E
C
T
R
A
-
ba
s
e
d
p
ip
e
l
in
e
f
o
r
s
e
nt
im
e
nt
a
na
l
y
s
is
of
t
o
ur
is
t
a
tt
r
a
c
t
io
n
r
e
v
i
e
w
s
,”
A
ppl
ie
d Sc
ie
nc
e
s
(
Sw
it
z
e
r
la
nd)
, vo
l.
12, n
o
. 21, p. 10
881, Oc
t.
2022, d
o
i:
10.3390/app122110
881.
[
29]
P
.
K
a
s
ne
s
is
,
L
.
T
ouma
ni
di
s
,
a
nd
C
.
Z
.
P
a
t
r
ik
a
ki
s
,
“
C
o
mba
ti
n
g
f
a
ke
n
e
w
s
w
it
h
t
r
a
ns
f
o
r
me
r
s
:
a
c
o
mpa
r
a
ti
ve
a
na
l
y
s
is
of
s
ta
nc
e
de
t
e
c
t
i
o
n a
nd s
ubj
e
c
ti
v
it
y
a
na
l
y
s
is
,”
I
nf
or
m
at
io
n (
Sw
it
z
e
r
la
nd)
, vo
l.
12, n
o
. 10, p. 409, O
c
t.
2021, d
o
i:
10.3390/i
n
f
o12100409.
[
30]
H
.
R
.
S
a
y
e
gh,
W
.
D
o
ng,
a
nd
A
.
M
.
A
l
-
ma
da
ni
,
“
E
nha
nc
e
d
in
tr
us
io
n
d
e
t
e
c
ti
o
n
w
it
h
L
S
T
M
-
ba
s
e
d
m
o
d
e
l,
f
e
a
tu
r
e
s
e
l
e
c
ti
o
n,
a
nd
S
M
O
T
E
f
o
r
i
mba
la
n
c
e
d d
a
ta
,”
A
ppl
ie
d Sc
ie
nc
e
s
(
Sw
it
z
e
r
la
nd)
, vo
l.
14, n
o
. 2, p. 479, J
a
n. 2024, doi:
10.3390/a
pp14020479.
B
I
OG
RA
P
HI
E
S
OF
AU
T
HO
RS
J
un
ta
Zeni
a
rja
i
s
a
p
e
r
m
a
n
e
n
t
l
ec
t
u
r
e
r
at
t
h
e
D
e
p
art
men
t
o
f
I
n
f
o
r
m
at
i
c
s
E
n
g
i
n
ee
r
i
n
g
,
Fa
cu
l
t
y
o
f
Co
m
p
u
t
e
r
S
ci
e
n
ce,
U
n
i
v
e
rs
i
t
as
D
i
an
N
u
s
w
an
t
o
r
o
,
I
n
d
o
n
e
s
i
a.
H
e
rece
i
v
e
d
t
w
o
m
as
t
e
r's
d
eg
ree
s
i
n
In
fo
r
m
at
i
c
s
E
n
g
i
n
ee
ri
n
g
fro
m
U
n
i
v
e
rs
i
t
i
T
e
k
n
i
k
al
Ma
l
a
y
s
i
a
Me
l
a
k
a
(U
T
e
M)
i
n
Mal
a
y
s
i
a
a
n
d
U
n
i
v
e
rs
i
t
as
D
i
an
N
u
s
w
an
t
o
r
o
i
n
I
n
d
o
n
e
s
i
a.
T
h
e
m
a
i
n
re
s
e
ar
ch
i
n
t
e
r
e
s
t
s
are
d
at
a
m
i
n
i
n
g
,
m
a
c
h
i
n
e
l
e
arn
i
n
g
,
d
ee
p
l
e
arn
i
n
g
,
i
n
f
o
r
m
at
i
o
n
r
e
t
ri
e
v
a
l
,
an
d
g
eo
g
rap
h
i
c
i
n
f
o
r
m
at
i
o
n
s
y
s
t
em
s
.
H
i
s
cu
rr
e
n
t
r
e
s
e
arch
c
o
v
e
rs
a
w
i
d
e
ra
n
g
e
o
f
D
at
a
Mi
n
i
n
g
an
d
D
at
a
S
c
i
e
n
ce
ap
p
l
i
c
at
i
o
n
s
,
s
u
ch
as
p
red
i
c
t
i
n
g
s
t
u
d
en
t
g
rad
u
at
i
o
n
,
s
e
ar
c
h
e
n
g
i
n
e
s
fo
r
k
i
d
s
,
t
h
e
s
i
s
d
o
c
u
men
t
c
l
as
s
i
fi
c
at
i
o
n
,
s
e
n
t
i
men
t
an
al
y
s
i
s
,
a
n
d
g
e
o
g
rap
h
i
c
i
n
fo
r
m
at
i
o
n
s
y
s
t
em
s
b
as
e
d
o
n
i
n
f
o
r
m
at
i
o
n
r
e
t
ri
ev
al
(G
eo
g
rap
h
i
c
In
fo
r
m
at
i
o
n
R
e
t
ri
e
v
a
l
).
H
e
c
an
b
e
c
o
n
t
ac
t
ed
at
em
a
i
l
:
j
u
n
t
a@
d
s
n
.
d
i
n
u
s
.
a
c
.
i
d
.
C
i
n
a
n
ty
a
P
a
ra
m
i
ta
i
s
a
p
e
r
m
an
en
t
l
ec
t
u
r
e
r
at
t
h
e
D
e
p
art
men
t
o
f
In
fo
r
m
at
i
c
s
E
n
g
i
n
ee
r
i
n
g
,
Fa
cu
l
t
y
o
f
C
o
m
p
u
t
e
r
S
c
i
en
ce
,
U
n
i
v
e
rs
i
t
as
D
i
an
N
u
s
w
a
n
t
o
ro
,
I
n
d
o
n
e
s
i
a.
S
h
e
rece
i
v
e
d
t
h
e
b
ach
e
l
o
r's
d
e
g
r
ee
i
n
c
o
m
p
u
t
e
r
s
ci
e
n
ce
fro
m
U
n
i
v
e
rs
i
t
as
D
i
a
n
N
u
s
w
a
n
t
o
ro
,
Sem
aran
g
,
In
d
o
n
e
s
i
a,
i
n
2
0
1
1
,
t
h
e
m
as
t
e
r's
d
eg
r
ee
i
n
E
l
ec
t
ri
c
al
a
n
d
C
o
m
p
u
t
e
r
E
n
g
i
n
ee
r
i
n
g
fro
m
So
u
t
h
Ch
i
n
a
U
n
i
v
e
rs
i
t
y
o
f
T
ech
n
o
l
o
g
y
,
G
u
an
g
zh
o
u
Ch
i
n
a,
i
n
2
0
1
4
.
T
h
e
m
ai
n
r
e
s
e
ar
ch
i
n
t
e
r
e
s
t
s
are
I
T
e
s
s
en
t
i
al
s
,
cy
b
e
rs
ec
u
r
i
t
y
,
d
at
a
m
i
n
i
n
g
,
m
a
c
h
i
n
e
l
e
arn
i
n
g
,
a
n
d
d
ee
p
l
e
arn
i
n
g
.
Sh
e
c
o
n
t
ri
b
u
t
i
n
g
as
a
t
rai
n
e
r
fo
r
v
ari
o
u
s
ac
a
d
em
i
e
s
u
n
d
e
r
Mi
n
i
s
t
r
y
o
f
C
o
mm
u
n
i
c
at
i
o
n
s
an
d
I
n
f
o
r
m
at
i
c
s
,
c
o
v
e
ri
n
g
I
T
e
s
s
en
t
i
al
s
,
cy
b
e
rs
ec
u
r
i
t
y
,
a
n
d
d
at
a
an
al
y
s
t
.
S
h
e
c
an
b
e
co
n
t
a
c
t
ed
at
em
ai
l
:
c
i
n
an
t
y
a.
p
ara
m
i
t
a@
d
s
n
.
d
i
n
u
s
.
a
c
.
i
d
.
Eg
i
a
R
o
s
i
Su
bhi
y
a
k
to
co
m
p
l
e
t
e
d
h
i
s
Ba
c
h
e
l
o
r's
d
e
g
r
ee
i
n
Co
m
p
u
t
e
r
S
c
i
en
ce
an
d
E
n
g
i
n
ee
r
i
n
g
i
n
2
0
1
1
at
t
h
e
U
n
i
v
e
rs
i
t
as
K
o
m
p
u
t
e
r
I
n
d
o
n
e
s
i
a,
Ban
d
u
n
g
,
W
e
s
t
J
av
a.
H
e
p
u
rs
u
e
d
h
e
r
Mas
t
e
r's
d
e
g
r
ee
i
n
C
o
m
p
u
t
e
r
S
ci
en
ce
a
t
D
i
an
N
u
s
w
an
t
o
r
o
U
n
i
v
e
rs
i
t
y
an
d
U
n
i
v
e
rs
i
t
i
T
e
k
n
i
k
al
Mal
a
y
s
i
a
M
e
l
a
k
a
(
U
T
e
M)
i
n
2
0
1
4
.
Si
n
ce
2
0
1
4
,
h
e
h
as
b
een
w
o
r
k
i
n
g
as
a
l
ec
t
u
r
e
r
at
D
i
a
n
N
u
s
w
a
n
t
o
ro
U
n
i
v
e
rs
i
t
y
,
S
em
aran
g
,
i
n
t
h
e
Co
m
p
u
t
e
r
S
ci
e
n
ce
p
ro
g
ra
m.
H
e
r
G
o
o
g
l
e
S
ch
o
l
ar
H
-
I
n
d
e
x
i
s
1
2
,
i
1
0
-
I
n
d
e
x
i
s
1
5
,
an
d
Sc
o
p
u
s
H
-
I
n
d
e
x
i
s
3
.
H
e
c
an
b
e
co
n
t
a
c
t
ed
at
em
ai
l
:
eg
i
a
@
d
s
n
.
d
i
n
u
s
.
ac.
i
d
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2
5
0
2
-
4
7
52
In
do
n
e
s
i
a
n
J
E
l
e
c
E
n
g
&
C
o
m
p
S
c
i
,
Vo
l
.
3
7
,
N
o
.
1
,
J
a
n
ua
r
y
20
2
5
:
3
1
5
-
324
324
Si
n
d
hu
R
a
k
a
s
i
w
i
co
m
p
l
e
t
ed
h
e
r
Ba
c
h
el
o
r's
d
e
g
r
ee
i
n
Co
m
p
u
t
e
r
S
ci
en
ce
a
n
d
E
n
g
i
n
ee
r
i
n
g
i
n
2
0
1
0
a
t
U
n
i
v
e
rs
i
t
as
D
i
an
N
u
s
w
an
t
o
r
o
,
Sem
aran
g
,
C
e
n
t
ral
J
av
a
.
Sh
e
p
u
rs
u
e
d
h
e
r
Mas
t
e
r's
d
eg
r
ee
i
n
Co
m
p
u
t
e
r
S
ci
e
n
ce
at
U
n
i
v
e
rs
i
t
as
D
i
an
N
u
s
w
an
t
o
r
o
,
i
n
2
0
1
2
.
Fro
m
2
0
1
3
t
o
2
0
2
3
,
s
h
e
w
o
r
k
e
d
as
a
l
ec
t
u
r
e
r
at
t
h
e
U
n
i
v
e
rs
i
t
as
Sai
n
s
d
a
n
T
e
k
n
o
l
o
g
i
K
o
m
p
u
t
e
r
,
Sem
aran
g
.
Si
n
ce
2
0
2
3
,
s
h
e
h
as
b
ee
n
s
e
r
v
i
n
g
as
a
l
ec
t
u
re
r
at
D
i
a
n
N
u
s
w
a
n
t
o
ro
U
n
i
v
e
rs
i
t
y
,
Sem
aran
g
,
i
n
t
h
e
Co
m
p
u
t
e
r
S
c
i
en
ce
p
ro
g
ra
m
.
H
e
r
G
o
o
g
l
e
S
ch
o
l
ar
H
-
In
d
e
x
i
s
5
,
an
d
h
e
r
i
1
0
-
I
n
d
e
x
i
s
4
.
S
h
e
c
an
b
e
co
n
t
a
c
t
ed
at
em
a
i
l
:
s
i
n
d
h
u
.
ra
k
as
i
w
i
@
d
s
n
.
d
i
n
u
s
.
a
c
.
i
d
.
Gu
r
uh
F
a
ja
r
Shi
di
k
(Mem
b
e
r
,
I
E
E
E
)
w
as
b
o
rn
i
n
W
e
s
t
Bo
r
n
e
o
,
I
n
d
o
n
e
s
i
a,
i
n
Fe
b
ru
ar
y
1
9
8
7
.
H
e
recei
v
ed
t
h
e
b
ac
h
el
o
r's
d
eg
r
ee
i
n
co
m
p
u
t
e
r
s
c
i
en
ce
fro
m
U
n
i
v
e
rs
i
t
as
D
i
an
N
u
s
w
an
t
o
r
o
,
S
em
aran
g
,
I
n
d
o
n
e
s
i
a,
i
n
2
0
0
9
,
t
h
e
m
as
t
er's
d
e
g
r
ee
i
n
co
m
p
u
t
e
r
s
ci
e
n
ce
fro
m
T
e
ch
n
i
c
al
Ma
l
a
y
s
i
a
M
el
a
k
a
U
n
i
v
e
rs
i
t
y
,
Ma
l
a
y
s
i
a,
i
n
2
0
1
1
,
an
d
t
h
e
Ph
.
D
.
d
e
g
r
ee
fro
m
U
n
i
v
e
rs
i
t
as
G
ad
j
ah
Mad
a,
Y
o
g
y
a
k
art
a,
In
d
o
n
e
s
i
a,
i
n
2
0
1
6
.
H
e
i
s
cu
rr
e
n
t
l
y
a
n
A
s
s
o
c
i
at
e
Pro
fe
s
s
o
r
w
i
t
h
U
n
i
v
e
rs
i
t
as
D
i
an
N
u
s
w
a
n
t
o
ro
.
H
i
s
r
e
s
e
ar
c
h
i
n
t
e
r
e
s
t
s
i
n
c
l
u
d
e
c
l
o
u
d
c
o
m
p
u
t
i
n
g
,
w
i
r
e
l
e
s
s
c
o
mmu
n
i
c
a
t
i
o
n
,
m
a
c
h
i
n
e
l
e
arn
i
n
g
,
a
n
d
art
i
f
i
c
i
a
l
i
n
t
el
l
i
g
en
c
e.
H
e
c
a
n
b
e
co
n
t
a
c
t
ed
at
em
ai
l
:
g
u
ru
h
.
faj
ar
@
r
e
s
e
ar
ch
.
d
i
n
u
s
.
a
c
.
i
d
.
P
u
l
un
g
N
u
rta
n
ti
o
A
n
do
n
o
rece
i
v
e
d
t
h
e
b
ach
e
l
o
r’s
d
e
g
r
ee
fr
o
m
T
ri
s
ak
t
i
U
n
i
v
e
rs
i
t
y
,
J
ak
art
a,
I
n
d
o
n
e
s
i
a,
i
n
2
0
0
6
,
t
h
e
m
as
t
e
r
’s
d
e
g
r
ee
fr
o
m
D
i
an
N
u
s
w
a
n
t
o
ro
U
n
i
v
e
rs
i
t
y
,
S
em
ara
n
g
,
I
n
d
o
n
e
s
i
a,
i
n
2
0
0
9
,
an
d
t
h
e
P
h
.
D
.
d
eg
r
ee
fr
o
m
t
h
e
I
n
s
t
i
t
u
t
T
ek
n
o
l
o
g
i
Se
p
u
l
u
h
N
o
v
em
b
e
r
(
I
T
S),
S
u
rab
a
y
a,
In
d
o
n
e
s
i
a,
i
n
2
0
1
6
.
H
e
cu
rr
e
n
t
l
y
w
o
rk
s
as
a
L
ec
t
u
r
e
r
a
n
d
a
R
e
s
e
ar
c
h
e
r
at
t
h
e
Fa
cu
l
t
y
o
f
C
o
m
p
u
t
e
r
S
c
i
en
ce,
D
i
a
n
N
u
s
w
an
t
o
r
o
U
n
i
v
e
rs
i
t
y
.
H
i
s
r
e
s
e
ar
c
h
i
n
t
e
r
e
s
t
s
i
n
cl
u
d
e
i
m
ag
e
s
ec
u
r
i
t
y
,
c
o
m
p
u
t
e
r
v
i
s
i
o
n
,
an
d
3
D
i
m
ag
e
r
ec
o
n
s
t
ru
c
t
i
o
n
.
H
e
h
as
au
t
h
o
r
e
d
o
r
co
au
t
h
o
r
e
d
mo
r
e
t
h
an
4
3
re
fe
r
eed
j
o
u
rn
a
l
s
an
d
c
o
n
f
e
r
e
n
ce
p
ap
e
rs
i
n
d
e
x
ed
b
y
Sco
p
u
s
.
H
e
c
a
n
b
e
c
o
n
t
ac
t
e
d
at
em
ai
l
:
pu
l
u
n
g
@
d
s
n
.
d
i
n
u
s
.
ac
.
i
d
.
A
n
a
m
a
ri
ja
J
u
rcev
Sa
v
i
cev
i
c
i
s
med
i
c
a
l
d
o
c
t
o
r,
s
p
e
c
i
a
l
i
s
t
o
f
e
p
i
d
emi
o
l
o
g
y
,
p
ri
m
ari
u
s
.
S
h
e
h
as
b
ee
n
w
o
rk
i
n
g
at
T
e
a
c
h
i
n
g
P
u
b
l
i
c
H
e
a
l
t
h
In
s
t
i
t
u
t
e
o
f
Sp
l
i
t
an
d
D
al
m
at
i
a
Co
u
n
t
y
as
a
h
e
ad
o
f
S
c
i
en
t
i
fi
c
U
n
i
t
a
n
d
h
e
a
d
o
f
D
e
p
art
me
n
t
f
o
r
E
p
i
d
emi
o
l
o
g
y
o
f
R
e
s
p
i
rat
o
r
y
I
n
f
ec
t
i
o
n
s
.
A
s
a
fi
el
d
e
p
i
d
em
i
o
l
o
g
i
s
t
,
s
h
e
h
as
b
een
d
ee
p
l
y
i
n
v
o
l
v
e
d
i
n
d
i
ffe
r
en
t
o
u
t
b
re
ak
c
o
n
t
ro
l
a
c
t
i
v
i
t
i
e
s
.
A
t
u
n
i
v
e
rs
i
t
y
l
ev
e
l
,
s
h
e
i
s
e
x
p
e
ri
en
ce
d
t
e
a
ch
e
r
,
r
e
s
e
ar
ch
e
r,
men
t
o
r
an
d
re
v
i
ew
e
r
,
em
p
l
o
y
ed
at
S
c
h
o
o
l
o
f
M
ed
i
ci
n
e
U
n
i
v
e
rs
i
t
y
o
f
Sp
l
i
t
an
d
D
e
p
art
me
n
t
o
f
H
e
al
t
h
St
u
d
i
e
s
U
n
i
v
e
rs
i
t
y
o
f
Sp
l
i
t
w
h
e
r
e
s
h
e
ru
n
s
D
e
p
art
me
n
t
fo
r
Pre
v
en
t
i
v
e
M
ed
i
ci
n
e.
T
h
e
m
a
i
n
re
s
e
ar
ch
i
n
t
e
re
s
t
s
are
d
i
s
e
as
e
p
r
e
v
en
t
i
o
n
an
d
c
o
n
t
ro
l
as
w
e
l
l
as
h
e
al
t
h
p
ro
mo
t
i
o
n
,
i
n
cl
u
d
i
n
g
l
i
f
e
s
t
y
l
e
med
i
ci
n
e
.
A
u
t
h
o
r
o
f
n
u
me
ro
u
s
p
ap
e
rs
,
ed
i
t
o
r
o
f
u
n
i
v
e
rs
i
t
y
t
e
x
t
b
o
o
k
s
an
d
au
t
h
o
rs
o
f
s
e
v
e
ral
b
o
o
k
s
an
d
p
u
b
l
i
c
at
i
o
n
s
fo
r
h
e
al
t
h
ed
u
c
at
i
o
n
.
Rece
n
t
l
y
,
s
h
e
h
as
b
ee
n
i
n
t
e
r
e
s
t
e
d
i
n
ad
v
an
t
ag
e
s
o
f
art
i
fi
ci
al
i
n
t
el
l
i
g
en
ce
i
n
t
h
e
h
e
al
t
h
s
e
r
v
i
ce
.
Fu
l
l
mem
b
e
r
o
f
Cro
at
i
a
n
A
c
ad
emy
o
f
M
e
d
i
c
al
S
ci
e
n
ce
s
.
S
h
e
c
a
n
b
e
co
n
t
a
c
t
ed
at
em
ai
l
:
an
a
m
ari
j
a.
j
u
r
cev
.
s
av
i
cev
i
c@
n
z
j
z
-
s
p
l
i
t
.
h
r
.
Evaluation Warning : The document was created with Spire.PDF for Python.