I
A
E
S
I
n
t
e
r
n
at
io
n
al
Jou
r
n
al
of
A
r
t
if
ic
ia
l
I
n
t
e
ll
ig
e
n
c
e
(
I
J
-
AI
)
V
ol
.
14
, N
o.
5
,
O
c
to
be
r
2025
, pp.
3771
~
3780
I
S
S
N
:
2252
-
8938
,
D
O
I
:
10.11591/
ij
a
i.
v
14
.i
5
.pp
3771
-
3780
3771
Jou
r
n
al
h
om
e
page
:
ht
tp
:
//
ij
ai
.
ia
e
s
c
or
e
.c
om
D
at
a
-
d
r
i
ve
n
c
l
u
st
e
r
i
n
g an
d
p
r
e
d
i
c
t
i
on
of
h
i
gh
sc
h
ool
gr
ad
u
at
i
on
r
at
e
s i
n
In
d
on
e
si
a (
2015
-
2023)
u
si
n
g
m
ac
h
i
n
e
l
e
ar
n
i
n
g
M
u
h
am
m
ad
S
al
m
an
A
r
r
os
yi
d
, M
ar
z
u
k
i,
Wi
d
ih
as
t
u
t
i,
H
ar
yan
t
o, M
ar
ia
A
n
ge
li
n
a F
r
an
s
is
k
a M
b
ar
i
D
e
pa
r
t
m
e
nt
of
E
duc
a
t
i
ona
l
R
e
s
e
a
r
c
h a
nd E
va
l
ua
t
i
on,
G
r
a
dua
t
e
S
c
hool
,
U
ni
ve
r
s
i
t
a
s
N
e
ge
r
i
Y
ogya
ka
r
t
a
,
Y
ogya
ka
r
t
a
,
I
ndone
s
i
a
A
r
t
ic
le
I
n
f
o
A
B
S
T
R
A
C
T
A
r
ti
c
le
h
is
to
r
y
:
R
e
c
e
iv
e
d
J
a
n 7, 2025
R
e
vi
s
e
d
J
un 24, 2025
A
c
c
e
pt
e
d
A
ug 6, 2025
This
study
aims
to
analyze
the
graduation
rate
of
senior
high
school
education
in
34
Indonesi
an
provinces
during
the
period
2015
-
202
3
and
identify
patterns
of
educationa
l
disparities
between
regions.
To
achie
ve
the
objectives,
this
study
applies
a
neural
network
to
predict
ed
ucation
completi
on
patterns
based
on
histori
cal
data,
then
the
predicti
on
resu
lts
are
analyzed
using
K
-
means
clustering
technique
utilizing
the
elbow
met
hod
to
select
the
ideal
number
of
clusters.
The
clustering
results
show
three
categ
ories
of
provinces
based
on
education
completion
rates:
high,
m
edium,
and
low.
The
provinces
with
high
completion
rates,
generally,
sup
ported
with
good
education
infrastructure
and
effective
policies,
while
the
m
edium
category
faces
challenges
in
resource
distri
bution
,
but
still
pote
ntiall
y
improve.
In
contrast,
the
low
categor
y
suffers
from
limited
access,
geographical
constraints
,
and
socio
-
economic
disparit
ies.
This
re
search
contribu
tes to
education
policy
-
making by off
ering a ma
chine lea
rning
-
based
appr
oach
to
understanding
education
disparities
between
regions.
Th
e
new
insight
offere
d
by
this
study
lies
in
the
integration
of
neural
network
and
K
-
means cluste
ring in mapping
education
completion r
ates to suppo
rt str
ategies
for improvi
ng access and qu
ality of ed
ucation in
Indonesia.
K
e
y
w
o
r
d
s
:
E
duc
a
ti
on
E
duc
a
ti
on dis
pa
r
it
y
G
r
a
dua
ti
on r
a
te
K
-
m
e
a
ns
c
lu
s
te
r
in
g
M
a
c
hi
ne
l
e
a
r
ni
ng
N
e
ur
a
l
ne
twor
k
This is an
open
acce
ss artic
le unde
r the
CC BY
-
SA
license.
C
or
r
e
s
pon
di
n
g A
u
th
or
:
M
uha
m
m
a
d S
a
lm
a
n A
r
r
os
yi
d
D
e
pa
r
tm
e
nt
of
E
duc
a
ti
ona
l
R
e
s
e
a
r
c
h a
nd E
va
lu
a
ti
on,
G
r
a
dua
te
S
c
hool
,
U
ni
ve
r
s
it
a
s
N
e
ge
r
i
Y
ogya
ka
r
ta
St
. C
ol
om
bo N
o.1 Ka
r
a
ngm
a
la
ng Y
ogya
ka
r
ta
55281
,
I
ndone
s
ia
E
m
a
il
:
m
uha
m
m
a
d0039pa
s
c
a
.2023@
s
tu
de
nt
.uny.
a
c
.i
d
1.
I
N
T
R
O
D
U
C
T
I
O
N
E
duc
a
ti
on
is
th
e
m
a
in
f
ounda
ti
on
of
th
e
I
ndone
s
ia
n
c
ount
r
y
in
c
r
e
a
ti
ng
a
s
m
a
r
t
a
nd
c
iv
il
iz
e
d
s
oc
ie
ty
[
1]
.
H
ow
e
ve
r
,
th
is
e
duc
a
ti
on
s
e
c
to
r
is
s
ti
ll
pl
a
gue
d
by
c
o
m
pl
e
x
pr
obl
e
m
s
,
e
s
pe
c
ia
ll
y
th
e
in
e
qui
ta
bl
e
di
s
tr
ib
ut
io
n
of
e
duc
a
ti
on,
w
hi
c
h
hi
nde
r
s
a
c
c
e
s
s
to
qu
a
li
ty
e
duc
a
ti
on
f
or
a
ll
le
ve
ls
of
s
oc
ie
ty
[
2]
,
[
3]
.
B
a
s
e
d
on
L
a
w
N
um
be
r
21
of
2003.
A
c
c
or
di
ng
to
th
e
N
a
ti
ona
l
E
duc
a
ti
on
S
ys
te
m
,
e
duc
a
ti
on
r
e
f
e
r
s
to
a
pur
pos
e
f
ul
a
nd
s
ys
te
m
a
ti
c
a
ll
y
de
s
ig
ne
d
e
f
f
or
t
to
bui
ld
a
n
e
nvi
r
onm
e
nt
a
nd
le
a
r
ni
ng
pr
oc
e
s
s
th
a
t
e
nc
our
a
ge
s
le
a
r
ne
r
s
to
be
a
c
ti
ve
a
nd
in
te
r
a
c
t
in
de
ve
lo
pi
ng
th
e
ir
pot
e
nt
ia
l.
T
he
pur
pos
e
of
th
is
e
duc
a
ti
on
is
to
f
o
r
m
s
tu
de
nt
s
to
de
ve
lo
p
s
pi
r
it
ua
l
s
tr
e
ngt
h,
s
e
lf
-
r
e
gul
a
ti
on,
in
te
ll
ig
e
nc
e
,
c
om
m
e
nda
bl
e
c
ha
r
a
c
te
r
,
a
nd
c
om
pe
te
n
c
ie
s
e
s
s
e
nt
ia
l
f
or
one
s
e
lf
, t
he
c
om
m
uni
ty
, t
he
na
ti
on, a
nd t
he
c
ount
r
y
[
4]
, [
5]
.
I
nc
r
e
a
s
in
g
c
om
pl
e
xi
ty
of
gl
oba
l
c
ha
ll
e
nge
s
[
6]
,
s
uc
h
a
s
c
li
m
a
te
c
ha
nge
,
pove
r
ty
,
a
nd
s
oc
ia
l
in
e
qua
li
ty
[
7]
.
T
he
r
e
f
or
e
e
duc
a
ti
on
ha
s
a
s
tr
a
te
gi
c
r
ol
e
a
s
th
e
m
a
in
in
s
tr
um
e
nt
in
s
ha
pi
ng
a
nd
in
c
r
e
a
s
in
g
th
e
c
a
pa
c
it
y
of
hum
a
n
r
e
s
our
c
e
s
a
nd
e
m
pow
e
r
in
g
th
e
c
om
m
uni
ty
to
c
ont
r
ib
ut
e
to
bui
ld
in
g
a
nd
r
e
a
li
z
in
g
s
oc
i
a
l
w
e
lf
a
r
e
w
it
hout
in
e
qua
li
ty
[
8]
.
T
hr
ough
e
duc
a
ti
on,
hum
a
ns
not
onl
y
a
c
qui
r
e
knowle
dge
,
but
a
ls
o
s
ki
ll
s
,
c
r
it
ic
a
l
th
in
ki
ng a
nd a
w
a
r
e
ne
s
s
t
o a
da
pt
iv
e
ly
f
a
c
e
t
he
gl
oba
l
dy
na
m
ic
s
.
T
he
s
e
ni
or
hi
gh s
c
hool
e
du
c
a
ti
on
le
v
e
l
is
a
n
im
por
ta
nt
a
nd
de
c
i
s
iv
e
pha
s
e
in
th
e
e
du
c
a
ti
ona
l
jo
ur
ne
y
of
s
tu
de
nt
s
[
9]
.
S
e
ni
or
hi
gh
s
c
hool
is
a
c
r
uc
ia
l
s
te
p
f
or
s
tu
d
e
n
ts
to
w
a
r
ds
hi
ghe
r
e
duc
a
ti
on
a
nd
e
m
pl
oym
e
nt
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
, V
ol
.
14
, N
o.
5
,
O
c
to
be
r
20
25
:
3771
-
3780
3772
[
10]
, [
11
]
, T
he
r
e
f
or
e
, s
tu
de
nt
s
a
t
th
e
s
e
ni
or
hi
gh s
c
hool
l
e
ve
l
m
us
t
be
e
qui
ppe
d w
it
h ba
s
ic
knowle
dge
, c
r
it
ic
a
l
th
in
ki
ng
s
ki
ll
s
,
a
nd
c
ogni
ti
ve
a
nd
a
f
f
e
c
ti
ve
s
ki
ll
s
to
f
a
c
e
va
r
i
ous
c
ha
ll
e
nge
s
a
f
te
r
th
e
y
c
om
pl
e
te
th
e
s
e
ni
or
hi
gh
s
c
hool
le
ve
l
[
12]
.
C
ont
e
xt
ua
ll
y,
it
is
im
por
ta
nt
to
e
n
s
ur
e
th
a
t
s
e
ni
or
s
e
c
onda
r
y
e
duc
a
ti
on
is
e
qua
ll
y
a
c
c
e
s
s
ib
le
to
a
ll
pe
opl
e
in
I
ndone
s
ia
n
pr
ovi
nc
e
s
w
it
h
no
di
s
pa
r
it
ie
s
in
f
a
c
il
it
ie
s
,
qua
li
ty
of
te
a
c
hi
ng
a
nd
le
a
r
ni
ng,
or
le
a
r
ni
ng
oppor
tu
ni
ti
e
s
[
13
]
.
T
he
da
ta
f
r
om
th
e
C
e
nt
r
a
l
B
ur
e
a
u
of
S
ta
ti
s
ti
c
s
or
B
a
da
n
P
us
a
t
S
ta
ti
s
ti
k
(
B
P
S
)
s
how
s
th
a
t
th
e
c
om
pl
e
ti
on
r
a
te
of
s
e
ni
or
hi
gh
s
c
hool
e
duc
a
ti
on
in
34
pr
ovi
nc
e
s
dur
in
g
2015
-
2023
e
xpe
r
ie
nc
e
d
s
ig
ni
f
ic
a
nt
di
s
pa
r
it
ie
s
.
G
e
ogr
a
phi
c
a
l
a
nd
s
oc
io
-
e
c
onomi
c
f
a
c
to
r
s
a
r
e
th
e
m
a
in
obs
ta
c
le
s
,
e
s
p
e
c
ia
ll
y
in
r
e
m
ot
e
,
bor
de
r
a
nd
unde
r
de
ve
lo
pe
d
a
r
e
a
s
.
T
hi
s
di
s
pa
r
it
y
a
f
f
e
c
ts
th
e
qua
li
ty
a
nd
qua
nt
it
y of
hi
gh s
c
hool
gr
a
dua
te
s
, a
nd ha
m
pe
r
s
e
qui
ta
bl
e
huma
n r
e
s
our
c
e
de
ve
lo
pm
e
nt
i
n I
ndone
s
ia
[
14]
.
E
d
uc
a
t
io
n
e
qu
it
y
is
no
t
o
nl
y
s
e
r
ve
d
a
s
a
n
e
f
f
o
r
t
t
o
e
ns
u
r
e
ju
s
ti
c
e
f
o
r
a
ll
le
a
r
ne
r
s
,
b
ut
a
ls
o
a
f
u
nda
m
e
n
ta
l
e
le
m
e
n
t
in
c
r
e
a
t
in
g
ba
la
nc
e
d
hum
a
n
r
e
s
o
u
r
c
e
de
ve
lo
pm
e
nt
t
hr
oug
ho
ut
I
ndo
ne
s
i
a
.
T
h
is
in
e
q
ua
li
ty
in
e
d
uc
a
t
io
na
l
,
s
o
c
ia
l
a
nd
e
c
on
om
ic
a
c
c
e
s
s
ha
s
a
s
ig
n
i
f
ic
a
nt
ne
ga
t
iv
e
i
m
pa
c
t
on
th
e
q
ua
l
it
y
a
nd
qua
n
ti
ty
o
f
hi
gh
s
c
ho
ol
g
r
a
d
ua
t
e
s
[
1
5]
.
B
y
e
ns
u
r
i
ng
s
us
ta
in
a
b
le
e
d
uc
a
t
io
n
e
qu
it
y,
e
ve
r
y
p
r
o
vi
n
c
e
in
I
n
done
s
ia
ha
s
th
e
p
ot
e
nt
ia
l
t
o
pr
odu
c
e
hi
g
h
s
c
ho
ol
g
r
a
d
ua
t
e
s
w
h
o
n
ot
o
nl
y
e
x
c
e
l
i
n
q
ua
l
it
y,
bu
t
a
ls
o
in
c
r
e
a
s
e
in
q
ua
n
ti
ty
.
T
hi
s
e
q
ui
ty
w
il
l
c
o
nt
r
ib
u
te
t
o
th
e
s
t
r
e
n
g
th
e
n
in
g
t
he
lo
c
a
l
e
c
on
om
y
in
e
a
c
h
p
r
o
vi
nc
e
a
nd
s
upp
o
r
t
na
t
io
na
l
e
c
o
no
m
ic
g
r
o
w
th
h
ol
is
t
ic
a
l
ly
.
I
n
a
dd
it
i
on,
th
e
g
r
a
dua
te
s
tu
d
e
nt
s
w
il
l
ha
ve
a
de
q
ua
t
e
c
om
p
e
te
nc
i
e
s
t
o
f
a
c
e
t
he
d
yna
m
i
c
s
a
n
d
c
ha
l
le
n
ge
s
of
g
lo
ba
l
c
o
m
pe
t
it
io
n
,
th
us
be
in
g
a
b
le
to
c
o
nt
r
i
bu
te
s
ig
n
i
f
ic
a
nt
ly
t
o
na
ti
on
bu
il
di
ng
.
A
lo
ng w
it
h t
he
r
a
pi
d
e
nha
nc
e
m
e
nt
of
t
e
c
hnol
ogy, the
a
ppl
ic
a
ti
o
n of
m
a
c
hi
ne
l
e
a
r
ni
ng i
s
i
nc
r
e
a
s
in
gl
y
w
id
e
s
pr
e
a
d
in
va
r
io
us
s
e
c
to
r
s
a
nd
a
ppl
ic
a
ti
on
s
[
16]
.
M
a
c
hi
ne
le
a
r
ni
ng
is
a
br
a
nc
h
of
a
r
ti
f
ic
ia
l
in
te
ll
ig
e
nc
e
(
A
I
)
th
a
t
e
na
bl
e
s
in
f
or
m
a
ti
on
s
ys
te
m
s
to
a
ut
om
a
ti
c
a
ll
y
le
a
r
n
pa
tt
e
r
ns
,
r
e
la
ti
ons
hi
ps
,
a
nd
c
ha
r
a
c
te
r
is
ti
c
s
in
da
ta
w
it
hout
r
e
qui
r
in
g
e
xpl
ic
it
p
r
ogr
a
m
m
in
g
in
s
tr
uc
ti
ons
[
17]
.
W
it
h
th
e
s
e
c
a
pa
bi
li
ti
e
s
,
m
a
c
hi
ne
le
a
r
ni
ng
is
a
bl
e
to
ge
ne
r
a
te
ne
w
knowle
dge
,
tr
a
in
a
lg
or
it
hm
s
,
id
e
nt
if
y
r
e
la
ti
ons
hi
ps
,
a
nd
r
e
c
ogni
z
e
hi
dde
n
p
a
tt
e
r
ns
th
a
t
ha
ve
not
be
e
n pr
e
vi
ous
ly
de
te
c
te
d
[
18]
.
T
he
pa
tt
e
r
ns
a
nd r
e
la
ti
ons
hi
ps
di
s
c
ove
r
e
d t
hr
ough thi
s
pr
oc
e
s
s
c
a
n be
us
e
d
to
a
na
ly
z
e
ne
w
a
nd
unknown
da
ta
,
e
na
bl
in
g
m
or
e
a
c
c
ur
a
te
pr
e
di
c
ti
ons
a
nd
s
uppor
ti
ng
e
f
f
ic
ie
nt
a
nd
a
da
pt
iv
e
pr
oc
e
s
s
opt
im
iz
a
ti
on
[
19]
.
T
hi
s
s
tu
dy
a
im
s
to
a
na
ly
z
e
th
e
gr
a
dua
ti
on
r
a
te
s
of
hi
gh
s
c
h
ool
s
tu
de
nt
s
in
va
r
io
us
pr
ovi
nc
e
s
in
I
ndone
s
ia
th
r
ough
a
m
a
c
hi
ne
le
a
r
ni
ng
a
ppr
oa
c
h,
by
in
te
gr
a
ti
ng
th
e
ne
ur
a
l
ne
twor
k
m
e
th
od
to
pr
e
di
c
t
gr
a
dua
ti
on
r
a
te
s
a
nd
th
e
K
-
m
e
a
n
s
c
lu
s
te
r
in
g
a
lg
or
it
hm
to
gr
oup
pr
ovi
nc
e
s
ba
s
e
d
on
c
om
pl
e
ti
on
p
a
tt
e
r
ns
dur
in
g
th
e
pe
r
io
d
2015
-
2023.
K
-
m
e
a
ns
is
kno
w
n
a
s
one
of
t
he
ol
de
s
t
a
nd
m
os
t
w
id
e
ly
u
s
e
d
p
a
r
ti
ti
oni
ng
m
e
th
ods
[
20]
.
T
hi
s
a
lg
or
it
hm
ha
s
be
e
n
th
e
obj
e
c
t
of
e
xt
e
ns
iv
e
s
tu
dy
w
it
h
va
r
io
us
de
ve
lo
pm
e
nt
s
in
th
e
li
te
r
a
tu
r
e
a
nd
a
ppl
ie
d
in
v
a
r
io
us
s
ub
s
ta
nt
iv
e
f
ie
ld
s
[
21]
.
T
hi
s
m
e
th
od
a
ll
ow
s
th
e
pr
oc
e
s
s
of
gr
oupi
ng
da
ta
ba
s
e
d
on
th
e
s
im
il
a
r
it
y
of
c
e
r
ta
in
c
ha
r
a
c
te
r
is
ti
c
s
[
22]
.
T
hi
s
s
tu
dy
pr
ovi
de
s
a
m
or
e
s
ys
te
m
a
ti
c
pi
c
tu
r
e
of
th
e
pa
tt
e
r
n
of
e
duc
a
ti
on
e
qui
ty
a
nd
in
e
qua
li
ty
in
e
a
c
h
pr
ovi
nc
e
.
T
h
e
f
in
di
ngs
a
r
e
e
xpe
c
te
d
to
s
e
r
ve
a
s
a
s
c
ie
nt
if
ic
ba
s
is
in
f
or
m
ul
a
ti
ng
s
tr
a
te
gi
c
pol
ic
ie
s
to
s
uppor
t
e
qui
ta
bl
e
a
c
c
e
s
s
to
e
duc
a
ti
on
a
nd
im
pr
ove
th
e
qua
li
ty
of
s
e
ni
or
s
e
c
ond
a
r
y
s
c
hool
gr
a
dua
t
e
s
.
I
n
a
ddi
ti
on,
th
e
r
e
s
ul
ts
o
f
th
is
s
tu
dy
a
ls
o
c
ont
r
ib
ut
e
to
s
tr
e
ngt
he
ni
ng
na
ti
ona
l
c
om
pe
ti
ti
ve
ne
s
s
th
r
ough
th
e
de
ve
lo
pm
e
nt
of
hum
a
n
r
e
s
our
c
e
s
th
a
t
a
r
e
m
or
e
a
da
pt
iv
e
a
nd
c
om
pe
ti
ti
ve
i
n t
he
m
id
s
t
of
gl
oba
l
dyna
m
ic
s
.
2.
M
E
T
H
O
D
T
hi
s
s
tu
dy a
na
ly
z
e
s
t
he
c
om
pl
e
ti
on r
a
te
of
s
e
ni
or
hi
gh s
c
hool
e
duc
a
ti
on i
n 34 pr
ovi
nc
e
s
i
n I
ndone
s
ia
in
t
he
pe
r
io
d 2015
-
2023 us
in
g da
ta
f
r
om
B
P
S
.
T
hi
s
s
tu
dy i
s
c
onduc
te
d i
n t
w
o pha
s
e
s
, t
he
a
r
e
da
ta
pr
e
pa
r
a
ti
on
s
ta
ge
a
nd
th
e
c
lu
s
te
r
in
g
s
ta
ge
,
a
s
s
how
n
in
F
ig
ur
e
1.
T
he
da
ta
pr
e
pa
r
a
ti
on
s
ta
ge
is
c
a
r
r
ie
d
out
in
s
e
ve
r
a
l
s
ta
ge
s
,
n
a
m
e
ly
c
ol
le
c
ti
ng
r
a
w
da
ta
,
c
le
a
ni
ng
da
ta
to
e
li
m
in
a
te
e
r
r
or
s
or
ir
r
e
le
va
nt
da
ta
,
f
il
te
r
in
g
da
ta
,
c
om
bi
ni
ng
a
ll
da
ta
,
a
nd
c
onv
e
r
ti
ng
da
ta
to
s
ui
t
th
e
n
e
e
ds
of
th
e
a
na
ly
s
is
.
F
ur
th
e
r
da
ta
a
na
ly
s
is
w
a
s
c
onduc
te
d
by
c
om
bi
ni
ng
m
a
c
hi
ne
le
a
r
ni
ng
ne
ur
a
l
ne
twor
k
m
e
th
od
a
nd
K
-
m
e
a
ns
c
lu
s
te
r
in
g
,
to
a
c
hi
e
v
e
a
th
or
ough
c
om
pr
e
he
ns
io
n
of
th
e
pa
tt
e
r
n
of
gr
a
dua
ti
on
or
c
om
pl
e
ti
on
of
e
duc
a
ti
on
a
t
th
e
hi
gh
s
c
hool
le
ve
l.
T
hi
s
in
te
gr
a
te
d
a
ppr
oa
c
h
pr
ovi
de
s
a
r
obus
t
f
r
a
m
e
w
or
k
f
or
unde
r
s
ta
ndi
ng
r
e
gi
ona
l
di
f
f
e
r
e
nc
e
s
a
nd
pa
tt
e
r
ns
of
e
duc
a
ti
ona
l
out
c
om
e
s
,
e
s
pe
c
ia
ll
y
hi
gh
s
c
hool
c
om
pl
e
ti
on
r
a
te
s
in
I
ndone
s
ia
.
F
ig
ur
e
1
s
how
s
th
e
s
ta
ge
s
of
da
ta
a
na
ly
s
i
s
.
2.1.
D
at
a
T
hi
s
r
e
s
e
a
r
c
h
ut
il
iz
e
s
a
da
ta
s
e
t
c
ont
a
in
in
g
da
ta
on
th
e
pe
r
c
e
nt
a
ge
of
c
om
pl
e
ti
on
o
r
gr
a
dua
ti
on
r
a
te
s
of
s
e
ni
or
hi
gh
s
c
hool
s
tu
de
nt
s
in
34
p
r
ovi
nc
e
s
in
I
ndone
s
ia
,
pr
e
s
e
nt
e
d
in
T
a
bl
e
1.
T
hi
s
da
ta
is
s
our
c
e
d
f
r
om
th
e
B
P
S
a
nd
c
ove
r
s
th
e
pe
r
io
d
2015
-
2023.
T
he
in
it
ia
l
s
ta
ge
w
a
s
c
a
r
r
ie
d
out
by
pr
oc
e
s
s
in
g
a
nd
pr
e
pa
r
in
g
th
e
da
ta
us
in
g
R
S
tu
di
o
s
of
twa
r
e
be
f
or
e
f
ur
th
e
r
a
na
ly
s
is
w
a
s
c
a
r
r
ie
d
out
by
pe
r
f
or
m
in
g
s
e
ve
r
a
l
s
ta
ge
s
,
na
m
e
ly
c
ol
le
c
ti
ng
r
a
w
da
ta
,
c
le
a
ni
ng
da
ta
to
e
li
m
in
a
te
e
r
r
or
s
or
ir
r
e
le
va
nt
da
ta
,
f
il
te
r
in
g
da
ta
,
c
om
bi
ni
ng
a
ll
da
ta
,
a
nd
c
onve
r
ti
ng
da
ta
to
s
ui
t
th
e
ne
e
ds
of
th
e
a
na
ly
s
is
.
T
he
a
na
ly
s
is
w
a
s
c
onduc
te
d
to
id
e
nt
if
y
gr
a
dua
ti
on
pa
tt
e
r
ns
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
D
at
a
-
dr
iv
e
n c
lu
s
te
r
in
g and pr
e
di
c
ti
on of hi
gh s
c
hool
gr
aduati
o
n r
at
e
s
i
n …
(
M
uhamm
ad Salman A
r
r
o
s
y
id
)
3773
a
nd e
duc
a
ti
ona
l
tr
e
nds
i
n e
a
c
h pr
ovi
nc
e
,
w
hi
c
h c
ont
r
ib
ut
e
s
t
o a
de
ta
il
e
d a
na
ly
s
is
of
t
he
de
te
r
m
in
a
nt
s
a
f
f
e
c
ti
ng
s
tu
de
nt
s
s
uc
c
e
s
s
r
a
te
in
c
om
pl
e
ti
ng
s
e
ni
or
s
e
c
onda
r
y
e
du
c
a
ti
o
n.
T
he
f
ol
lo
w
in
g
is
th
e
da
ta
on
th
e
pe
r
c
e
nt
a
g
e
(
%
)
of
s
e
ni
or
hi
gh s
c
hool
c
om
pl
e
ti
on r
a
te
s
f
r
om
pr
ovi
nc
e
s
i
n I
n
done
s
ia
f
r
om
2015 to 2023 i
n T
a
bl
e
1.
F
ig
ur
e
1. F
lo
w
c
ha
r
t
of
da
ta
a
na
ly
s
is
T
a
bl
e
1. S
tu
de
nt
pa
s
s
pe
r
c
e
nt
a
g
e
da
ta
P
r
ovi
nc
e
2015
2016
2017
2018
2019
2020
2021
2022
2023
A
c
e
h
68.16
74.46
70.64
70.68
69.96
70.07
74.36
70.67
74.46
S
um
a
t
e
r
a
U
t
a
r
a
59.54
69.69
67.16
68.34
65.21
70.39
72.81
77.16
74.43
S
um
a
t
e
r
a
B
a
r
a
t
58.04
64.97
60.8
65.34
60.32
67.11
70.06
65.96
68.64
R
i
a
u
57.28
62.12
61.9
63.71
58.78
66.62
68.94
66.91
67.79
J
a
m
bi
49.05
60.5
58.27
66.06
56.87
63.66
64.51
65.85
66.62
S
um
a
t
e
r
a
S
e
l
a
t
a
n
48.9
55.37
54.15
63.94
58.23
65.42
67.2
67.07
64.81
B
e
ngkul
u
55.94
64.31
62.57
58.86
61.47
62.73
62.46
64.88
63.41
L
a
m
pung
40.6
47.62
48.75
54.89
54.87
57.59
60.09
62.42
64.54
K
e
p. B
a
ngka
B
e
l
i
t
ung
43.46
53.84
51.55
55.01
53.84
56.74
63.98
66.87
68.96
K
e
p. R
i
a
u
65.28
75.93
83.55
82.86
78.14
78.65
81.07
73.93
78.97
J
a
ka
r
t
a
74.1
74.74
78.25
83.48
84.35
85.67
84.98
87.71
88.1
J
a
w
a
B
a
r
a
t
48.53
55.03
48.32
61.04
57.46
63.56
64.89
67.05
66.47
J
a
w
a
T
e
nga
h
43.86
44.59
51.52
55.62
49.79
55.82
59.9
58.75
58.35
D
I
Y
ogya
ka
r
t
a
80.77
79.95
85.53
81.96
84.54
87.99
90.12
87.92
89.69
J
a
w
a
T
i
m
ur
52.04
55.13
59.9
62.48
57.74
63.53
66.33
66.87
68.65
B
a
nt
e
n
52.95
60.83
59.87
67.54
56.94
64.24
66.9
66.02
70.07
B
a
l
i
69.08
73.65
74.62
78.67
64.52
74.88
75.86
76.59
76.51
N
us
a
T
e
ngga
r
a
B
a
r
a
t
51.83
55.01
59.1
52.6
57.6
64.66
65.71
61
63.66
N
us
a
T
e
ngga
r
a
T
i
m
ur
37.78
48.95
41.44
43.41
43.85
50.65
44.88
38.47
43.46
K
a
l
i
m
a
nt
a
n B
a
r
a
t
35.69
35.69
42.7
47.66
49.29
55.23
54.27
58.4
55.58
K
a
l
i
m
a
nt
a
n T
e
nga
h
47.28
52.42
56.48
53.47
50.01
60.77
61.04
61.88
63.93
K
a
l
i
m
a
nt
a
n
S
e
l
a
t
a
n
44.85
52.91
56.75
61.09
59.52
63.05
63.59
67.81
68.35
K
a
l
i
m
a
nt
a
n T
i
m
ur
67.56
66.76
67.72
68.73
64.74
71.63
74.26
74
73.63
K
a
l
i
m
a
nt
a
n U
t
a
r
a
47.64
58.6
57.43
58.22
61.1
67.77
62.3
54.8
59.5
S
ul
a
w
e
s
i
U
t
a
r
a
55.5
72.33
67.46
70.02
67.58
73.79
68.56
66.66
67.57
S
ul
a
w
e
s
i
T
e
nga
h
45.84
61.79
62.73
53.84
52
57.68
61.16
53.73
55.69
S
ul
a
w
e
s
i
S
e
l
a
t
a
n
50.85
59.56
63.82
56.86
60.97
66.22
69.43
68.32
67.41
S
ul
a
w
e
s
i
T
e
ngga
r
a
61.52
67.12
67.75
67.67
64.26
68.28
70.65
65.97
68.28
G
or
ont
a
l
o
44.67
50.79
55.3
52.39
50.87
55.35
53.73
45.12
46.19
S
ul
a
w
e
s
i
B
a
r
a
t
39.29
53.45
56.17
37.65
48.2
56.6
56.22
55.18
54.79
M
a
l
uku
58.59
72.87
73.58
66.42
67.82
70.55
68.12
72.08
75.01
M
a
l
uku U
t
a
r
a
57.12
64.87
65.14
60.07
59.13
66.52
66.95
67.1
64.61
P
a
pua
B
a
r
a
t
55.24
56.12
62.81
60.47
50.95
61.49
59.08
57.07
59.99
P
a
pua
28.23
35.69
33.82
29.56
27.44
30.92
32.95
39.01
39.5
2.2.
N
e
u
r
al
n
e
t
w
or
k
N
e
ur
a
l
n
e
twor
k
is
us
e
d
to
pr
e
di
c
t
tr
e
nd
s
in
e
duc
a
ti
on
c
om
pl
e
ti
on
r
a
te
s
ba
s
e
d
on
hi
s
to
r
ic
a
l
da
ta
. T
hi
s
m
e
th
od
w
a
s
c
hos
e
n
be
c
a
u
s
e
it
ha
s
th
e
a
bi
li
ty
to
c
a
pt
ur
e
n
on
-
li
ne
a
r
pa
tt
e
r
ns
a
nd
c
om
pl
e
x
r
e
la
ti
ons
hi
ps
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
, V
ol
.
14
, N
o.
5
,
O
c
to
be
r
20
25
:
3771
-
3780
3774
be
twe
e
n
va
r
ia
bl
e
s
.
T
h
e
ne
ur
a
l
ne
twor
k
m
ode
l
i
s
tr
a
in
e
d
u
s
in
g
a
da
ta
s
e
t
th
a
t
in
c
lu
de
s
e
duc
a
ti
on
in
di
c
a
to
r
s
,
s
uc
h
a
s
gr
a
dua
ti
on
r
a
te
s
in
e
a
c
h
pr
ovi
nc
e
.
T
he
m
ode
l
us
e
d
is
a
m
ul
ti
-
la
ye
r
pe
r
c
e
pt
r
on
(
M
L
P
)
w
it
h
s
e
ve
r
a
l
hi
dde
n
la
ye
r
s
opt
im
iz
e
d
to
c
a
pt
ur
e
c
om
pl
e
x
r
e
la
ti
ons
hi
ps
be
tw
e
e
n
va
r
ia
bl
e
s
.
T
he
pr
e
di
c
ti
on
r
e
s
ul
ts
f
r
om
th
is
ne
ur
a
l
ne
twor
k
pr
ovi
de
s
a
n
e
s
ti
m
a
ti
on
of
th
e
e
du
c
a
ti
on
gr
a
dua
ti
on
or
c
om
pl
e
ti
on
r
a
te
,
w
hi
c
h
us
e
d
a
s
th
e
ba
s
is
f
or
th
e
c
lu
s
te
r
in
g
pr
oc
e
s
s
us
in
g
K
-
m
e
a
n
s
.
T
he
out
put
of
th
is
m
ode
l
a
ll
ow
s
f
or
m
or
e
s
pe
c
if
ic
a
na
ly
s
is
in
c
lu
s
te
r
in
g
pr
ovi
nc
e
s
ba
s
e
d
on
th
e
e
duc
a
ti
ona
l
pa
tt
e
r
n
s
id
e
nt
if
ie
d
th
r
ough
th
e
pr
e
di
c
ti
ons
ge
ne
r
a
te
d. T
o
a
s
s
e
s
s
th
e
a
c
c
ur
a
c
y
a
nd
e
f
f
e
c
ti
ve
ne
s
s
of
th
e
ne
ur
a
l
ne
twor
k
m
ode
l
,
va
li
da
ti
on
w
a
s
c
onduc
te
d
us
in
g
f
our
m
a
in
m
e
tr
ic
s
,
na
m
e
ly
m
e
a
n
s
qua
r
e
d
e
r
r
or
(
M
S
E
)
,
r
oot
m
e
a
n
s
qua
r
e
d
e
r
r
or
(
R
M
S
E
)
,
m
e
a
n
a
bs
ol
ut
e
e
r
r
or
(
M
A
E
)
,
a
nd R
-
s
qua
r
e
d (
R
2
)
[
23]
.
2.2.1.
M
e
an
s
q
u
ar
e
d
e
r
r
or
M
S
E
m
e
a
s
ur
e
s
th
e
a
ve
r
a
g
e
of
th
e
s
qua
r
e
d
di
f
f
e
r
e
nc
e
be
tw
e
e
n
t
he
a
c
tu
a
l
v
a
lu
e
(
Yi
)
a
nd
th
e
pr
e
di
c
te
d
va
lu
e
(
yi
)
. T
he
s
m
a
ll
e
r
t
he
M
S
E
va
lu
e
, t
he
be
tt
e
r
m
ode
l
pr
e
di
c
ts
t
he
da
ta
w
it
h m
in
im
a
l
e
r
r
or
.
=
∑
=
−
(
1)
2.2.2.
R
oot
m
e
an
s
q
u
ar
e
d
e
r
r
o
r
R
M
S
E
i
s
a
n e
r
r
or
m
e
a
s
ur
e
t
ha
t
qua
nt
if
ie
s
t
he
va
r
ia
ti
on be
twe
e
n f
or
e
c
a
s
te
d a
nd t
r
ue
va
lu
e
s
. R
M
S
E
i
s
de
te
r
m
in
e
d
by
a
ppl
yi
ng
th
e
s
qua
r
e
r
oot
to
th
e
M
S
E
.
A
s
m
a
ll
e
r
R
M
S
E
va
lu
e
in
di
c
a
te
s
th
a
t
th
e
m
ode
l
h
a
s
a
lo
w
pr
e
di
c
ti
on e
r
r
or
r
a
te
.
=
√
(
2)
2.2.3.
M
e
an
ab
s
ol
u
t
e
e
r
r
or
M
A
E
is
a
m
e
a
s
ur
e
of
th
e
a
ve
r
a
ge
a
b
s
ol
ut
e
e
r
r
or
of
th
e
a
c
tu
a
l
va
lu
e
(
Yi
)
a
nd
th
e
pr
e
di
c
te
d
va
lu
e
(
yi
)
.
M
A
E
m
e
a
s
ur
e
s
how
la
r
ge
th
e
a
ve
r
a
g
e
di
f
f
e
r
e
nc
e
is
b
e
twe
e
n
th
e
pr
e
di
c
te
d
va
lu
e
a
nd
th
e
a
c
tu
a
l
va
lu
e
,
r
e
ga
r
dl
e
s
s
of
th
e
di
r
e
c
ti
on
of
th
e
e
r
r
o
r
(
pos
it
iv
e
o
r
ne
ga
ti
ve
)
.
T
he
s
m
a
ll
e
r
th
e
M
A
E
va
lu
e
,
th
e
be
tt
e
r
th
e
m
ode
l
is
a
t
m
a
ki
ng pr
e
di
c
ti
ons
.
=
∑
|
=
−
|
(
3)
2.2.4.
R
-
s
q
u
ar
e
d
R
2
is
a
s
ta
ti
s
ti
c
a
l
m
e
tr
ic
u
s
e
d
to
a
s
s
e
s
s
th
e
e
xt
e
nt
to
w
hi
c
h
th
e
m
ode
l
c
a
n
e
xpl
a
in
va
r
ia
bi
li
ty
in
th
e
a
c
tu
a
l
da
ta
(
Yi
)
.
R
²
va
lu
e
s
f
a
ll
w
it
hi
n
th
e
r
a
nge
of
0
to
1,
w
he
r
e
a
va
lu
e
a
ppr
oa
c
hi
ng
1
in
di
c
a
te
s
th
a
t
th
e
m
ode
l
is
ge
tt
in
g be
tt
e
r
a
t
e
xpl
a
in
in
g da
ta
va
r
ia
ti
on.
=
−
∑
(
−
)
∑
(
−
)
(
4)
2.3.
K
-
m
e
an
s
c
lu
s
t
e
r
in
g
K
-
m
e
a
ns
c
lu
s
te
r
in
g
s
e
r
ve
s
t
o
c
la
s
s
if
y
p
r
o
vi
n
c
e
s
ba
s
e
d
on
s
i
m
il
a
r
e
d
uc
a
t
io
na
l
c
ha
r
a
c
te
r
is
t
ic
s
.
T
h
e
K
-
m
e
a
ns
c
l
us
te
r
in
g
m
e
th
od
is
us
e
d
be
c
a
us
e
o
f
i
ts
a
bi
li
ty
t
o
e
f
f
ic
ie
n
tl
y
gr
oup
p
r
o
vi
nc
e
s
ba
s
e
d
on
th
e
s
im
il
a
r
i
ty
o
f
in
it
ia
l
c
e
n
tr
oi
d
va
l
ue
s
o
f
e
du
c
a
t
io
na
l
c
ha
r
a
c
te
r
is
ti
c
s
[
24
]
,
[
25
]
,
be
c
a
us
e
it
is
a
bl
e
t
o
e
va
l
ua
te
c
ha
n
ge
s
in
th
e
to
ta
l
w
i
th
i
n
-
c
l
us
te
r
s
u
m
o
f
s
qu
a
r
e
s
(
W
S
S
)
va
lu
e
a
ga
i
ns
t
v
a
r
io
us
nu
m
be
r
s
o
f
c
lu
s
t
e
r
s
.
T
he
e
lb
o
w
m
e
th
od
is
us
e
d
t
o
de
te
r
m
in
e
th
e
o
pt
im
a
l
n
um
be
r
o
f
c
lu
s
t
e
r
s
by
ba
la
nc
in
g
in
t
r
a
-
c
lu
s
te
r
v
a
r
ia
b
il
it
y
a
nd
op
ti
m
a
l
m
ode
l
c
om
pl
e
xi
ty
,
w
he
n
t
he
de
c
r
e
a
s
e
in
W
S
S
s
ta
r
ts
t
o
s
l
ow
do
w
n
s
ig
ni
f
ic
a
n
tl
y.
T
he
s
e
le
c
ti
on
of
t
he
nu
m
be
r
o
f
c
lu
s
te
r
s
a
im
s
to
a
c
h
ie
ve
a
ba
la
n
c
e
po
i
nt
be
twe
e
n
c
lu
s
te
r
i
ng
a
c
c
u
r
a
c
y
a
nd
m
ode
l
c
om
p
le
xi
ty
,
s
o
th
a
t
t
he
a
n
a
l
ys
is
r
e
s
u
lt
s
a
r
e
m
o
r
e
r
e
p
r
e
s
e
n
ta
t
iv
e
a
nd
t
he
da
ta
c
a
n
be
i
nt
e
r
pr
e
te
d
p
r
op
e
r
ly
in
t
o
f
i
xe
d
g
r
o
ups
[
26
]
.
2.4.
R
e
le
van
c
e
of
n
e
u
r
al
n
e
t
w
or
k
an
d
K
-
m
e
an
s
c
lu
s
t
e
r
in
g
T
hi
s
da
ta
a
na
ly
s
is
c
om
bi
ne
s
ne
u
r
a
l
ne
two
r
k
a
nd
K
-
m
e
a
ns
c
lu
s
te
r
in
g
a
s
th
e
y
ha
ve
s
ig
ni
f
i
c
a
n
t
r
e
l
e
va
n
c
e
in
e
duc
a
ti
on
a
na
ly
s
is
.
N
e
u
r
a
l
n
e
two
r
k
is
us
e
d
to
m
o
de
l
th
e
c
o
m
p
le
x
r
e
la
t
io
ns
h
ip
s
be
t
w
e
e
n
va
r
io
us
e
d
uc
a
t
io
na
l
f
a
c
to
r
s
,
w
i
th
it
s
a
bi
li
ty
to
c
a
p
tu
r
e
n
on
-
l
in
e
a
r
pa
t
te
r
ns
a
ll
ow
i
ng
f
o
r
m
o
r
e
a
c
c
u
r
a
te
pr
e
di
c
ti
on
o
f
e
du
c
a
t
io
na
l
c
o
m
p
le
ti
o
n
r
a
te
s
.
T
he
m
od
e
l
a
p
pl
ie
d
is
a
M
L
P
w
it
h
op
ti
m
iz
e
d
hi
dde
n
la
y
e
r
s
t
o
im
p
r
ove
p
r
e
di
c
ti
on
a
c
c
ur
a
c
y
.
A
f
te
r
ob
ta
in
i
ng
pr
e
di
c
ti
on
r
e
s
ul
ts
f
r
o
m
th
e
ne
u
r
a
l
ne
two
r
k
.
K
-
m
e
a
ns
c
lu
s
te
r
in
g
m
e
th
o
d
is
a
p
pl
ie
d
t
o
gr
ou
p
pr
ov
in
c
e
s
b
a
s
e
d
o
n
s
im
il
a
r
e
d
uc
a
t
io
na
l
c
ha
r
a
c
te
r
is
t
ic
s
[
27
]
.
T
hi
s
a
pp
r
oa
c
h
a
l
lo
w
s
th
e
id
e
nt
if
ic
a
ti
on
o
f
p
r
o
vi
nc
ia
l
c
l
us
te
r
s
w
it
h
ho
m
og
e
ne
ous
pa
tt
e
r
ns
o
f
e
duc
a
t
io
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
D
at
a
-
dr
iv
e
n c
lu
s
te
r
in
g and pr
e
di
c
ti
on of hi
gh s
c
hool
gr
aduati
o
n r
at
e
s
i
n …
(
M
uhamm
ad Salman A
r
r
o
s
y
id
)
3775
c
om
p
le
ti
on
[
2
8
]
.
S
o
t
ha
t
i
t
c
a
n
be
t
he
ba
s
is
in
f
o
r
m
u
la
ti
ng
e
d
uc
a
t
io
na
l
po
li
c
ie
s
th
a
t
a
r
e
m
or
e
ta
r
ge
te
d
[
2
9]
.
N
e
u
r
a
l
n
e
tw
or
k
a
c
ts
a
s
a
pr
e
d
ic
t
io
n
to
o
l
t
ha
t
c
a
pt
u
r
e
s
c
om
p
l
e
x
r
e
la
t
io
ns
hi
ps
in
th
e
da
ta
,
w
hi
le
K
-
m
e
a
ns
c
lu
s
te
r
in
g
gr
ou
ps
p
r
ovi
nc
e
s
ba
s
e
d
on
th
e
pr
e
di
c
ti
on
r
e
s
u
lt
s
,
a
l
lo
w
in
g
f
o
r
m
o
r
e
s
ys
te
m
a
t
ic
a
nd
i
n
-
de
p
th
a
na
l
ys
is
[
30
]
.
T
h
e
c
om
bi
na
t
io
n
o
f
th
e
s
e
tw
o
m
e
th
o
ds
no
t
o
nl
y
im
pr
ove
s
th
e
a
c
c
u
r
a
c
y
i
n
p
r
o
je
c
ti
n
g
e
duc
a
ti
on
t
r
e
n
ds
,
b
ut
a
ls
o
pr
ov
id
e
s
a
c
le
a
r
e
r
m
a
p
pi
ng
o
f
a
r
e
a
s
ba
s
e
d
on
e
d
uc
a
ti
o
n
c
ha
r
a
c
te
r
is
t
ic
s
,
th
us
s
upp
or
ti
ng
da
ta
-
d
r
iv
e
n
de
c
is
i
on
-
m
a
ki
ng
m
o
r
e
op
ti
m
a
ll
y.
3.
R
E
S
U
L
T
S
A
N
D
D
I
S
C
U
S
S
I
O
N
T
hi
s
s
tu
dy
a
ppl
ie
s
th
e
n
e
ur
a
l
ne
twor
k
a
nd
K
-
m
e
a
ns
c
lu
s
te
r
in
g
m
e
th
ods
to
a
na
ly
z
e
th
e
gr
a
dua
ti
on
r
a
te
of
s
e
ni
or
hi
gh
s
c
hool
e
duc
a
ti
on
in
34
pr
ovi
nc
e
s
of
I
ndone
s
ia
dur
in
g
th
e
pe
r
io
d
2015
-
2023.
T
he
da
ta
us
e
d
is
s
our
c
e
d
f
r
om
th
e
B
P
S
a
nd
in
c
lu
de
s
th
e
m
a
in
va
r
ia
bl
e
s
in
th
e
f
or
m
o
f
th
e
pe
r
c
e
nt
a
ge
of
e
duc
a
ti
on
c
om
pl
e
ti
on
e
a
c
h
ye
a
r
.
T
h
e
s
e
va
r
ia
bl
e
s
a
r
e
us
e
d
a
s
a
r
e
pr
e
s
e
nt
a
ti
on
of
th
e
le
ve
l
of
e
duc
a
ti
ona
l
s
uc
c
e
s
s
in
va
r
io
us
pr
ovi
nc
e
s
t
hr
oughout t
he
s
tu
dy pe
r
io
d.
A
t
th
e
in
it
ia
l
s
ta
ge
of
th
e
a
n
a
ly
s
is
,
a
da
ta
nor
m
a
li
z
a
ti
on
pr
oc
e
s
s
w
a
s
c
a
r
r
ie
d
out
us
in
g
th
e
Z
-
s
c
or
e
s
c
a
li
ng
m
e
th
od.
T
hi
s
s
te
p
a
im
s
to
e
qua
li
z
e
th
e
s
c
a
le
be
twe
e
n
va
r
ia
bl
e
s
s
o
th
a
t
no
va
r
ia
bl
e
ha
s
a
dom
in
a
nt
in
f
lu
e
nc
e
on
c
lu
s
t
e
r
in
g
r
e
s
ul
ts
.
T
he
s
c
a
li
ng
pr
oc
e
s
s
is
us
e
d
in
th
e
ne
ur
a
l
ne
twor
k
to
im
pr
ove
th
e
s
t
a
bi
li
ty
of
m
ode
l
le
a
r
ni
ng
a
nd e
ns
ur
e
th
e
a
c
ti
va
ti
on
f
unc
ti
on
w
or
ks
opt
im
a
ll
y
[
31]
.
I
n
a
ddi
ti
on,
s
c
a
li
ng
is
a
ls
o
a
ppl
ie
d
to
K
-
m
e
a
ns
c
lu
s
te
r
in
g
s
o
th
a
t
th
e
c
lu
s
te
r
in
g
of
pr
ovi
nc
e
s
ba
s
e
d
on
th
e
pa
tt
e
r
n
of
e
duc
a
ti
on
c
om
pl
e
ti
on
r
a
te
s
c
a
n
be
done
obj
e
c
ti
ve
ly
, t
a
ki
ng i
nt
o a
c
c
ount
t
he
e
qua
l
di
s
tr
ib
ut
io
n of
t
he
da
ta
[
32]
.
T
he
r
e
s
ul
ts
o
f
th
e
ne
u
r
a
l
ne
two
r
k
a
na
ly
s
is
a
r
e
v
is
ua
li
z
e
d
th
r
o
ugh
a
m
o
de
l
s
t
r
uc
tu
r
e
th
a
t
d
e
s
c
r
ib
e
s
th
e
r
e
l
a
t
io
ns
hi
p
be
twe
e
n
t
he
i
npu
t
da
ta
in
t
he
f
o
r
m
of
a
nn
ua
l
gr
a
dua
ti
on
s
c
or
e
s
(
2
01
5
-
20
23
)
a
n
d
th
e
ou
tp
ut
in
th
e
f
or
m
of
g
r
a
d
ua
t
io
n
.
T
he
w
e
ig
ht
s
be
twe
e
n
ne
u
r
ons
th
a
t
r
e
pr
e
s
e
nt
t
he
im
pa
c
t
o
f
in
d
iv
id
ua
l
va
r
ia
bl
e
s
on
t
he
p
r
e
di
c
ti
ve
o
ut
c
om
e
r
e
s
ul
ts
.
T
he
f
ol
lo
w
i
ng
a
r
e
th
e
r
e
s
u
lt
s
o
f
th
e
ne
u
r
a
l
ne
tw
or
k
v
is
ua
li
z
a
t
io
n
us
i
ng
R
S
t
ud
io
in
F
ig
ur
e
2
.
F
ig
ur
e
2. N
e
ur
a
l
ne
twor
k r
e
s
ul
ts
B
a
s
e
d
on
th
e
ne
ur
a
l
ne
twor
k
r
e
s
ul
ts
a
bove
,
th
e
m
ode
l
c
ons
is
ts
of
two
hi
dde
n
la
ye
r
s
th
a
t
pr
ogr
e
s
s
iv
e
ly
pr
oc
e
s
s
i
nput
da
ta
i
n t
he
f
or
m
of
a
nnua
l
gr
a
dua
ti
on r
a
te
s
(
2015
-
2023)
i
nt
o
out
put
i
n t
he
f
or
m
o
f
gr
a
dua
ti
on r
a
te
s
. T
he
w
e
ig
ht
s
be
twe
e
n ne
ur
ons
, a
s
s
how
n i
n F
ig
ur
e
2, r
e
f
le
c
t
th
e
l
e
ve
l
of
c
ont
r
ib
ut
io
n
of
e
a
c
h
in
put
va
r
ia
bl
e
to
th
e
pr
e
di
c
ti
on
r
e
s
ul
t
[
33]
.
T
h
e
vi
s
u
a
li
z
e
d
n
e
ur
on
a
c
ti
va
ti
ons
s
how
ho
w
th
e
m
ode
l
le
a
r
ns
c
om
pl
e
x
pa
tt
e
r
ns
of
r
e
la
ti
ons
hi
ps
,
in
c
lu
di
ng
non
-
li
ne
a
r
pa
tt
e
r
ns
th
a
t
c
a
nnot
be
c
a
pt
ur
e
d
by
c
onve
nt
io
na
l
a
na
ly
s
is
m
e
th
ods
.
T
he
im
pl
e
m
e
nt
e
d
n
e
ur
a
l
ne
twor
k
m
ode
l
s
how
s
a
hi
gh
le
ve
l
of
a
c
c
ur
a
c
y
in
pr
e
di
c
ti
ng
e
duc
a
ti
on
c
om
pl
e
ti
on
pa
tt
e
r
ns
.
T
he
m
od
e
l
va
li
da
ti
on
r
e
s
ul
ts
w
e
r
e
c
onduc
te
d
us
in
g
f
our
m
a
in
e
va
lu
a
ti
on
m
e
tr
ic
s
, na
m
e
ly
t
he
M
S
E
of
0.0001936, whic
h i
ndi
c
a
te
s
a
ve
r
y
s
m
a
ll
pr
e
di
c
ti
on e
r
r
or
r
a
te
a
f
te
r
262 it
e
r
a
ti
ons
.
I
n
a
ddi
ti
on,
th
e
R
M
S
E
o
f
0.0139
a
nd
th
e
r
e
s
ul
t
of
th
e
M
A
E
of
0.0100
in
di
c
a
te
th
a
t
th
e
m
ode
l
ha
s
a
good
pe
r
f
or
m
a
nc
e
in
r
e
duc
in
g
th
e
pr
e
di
c
ti
on
e
r
r
or
.
F
ur
th
e
r
m
or
e
,
th
e
R
²
va
lu
e
of
0.662
pr
ovi
de
s
in
f
or
m
a
ti
on
th
a
t
th
e
m
ode
l
is
a
bl
e
to
e
xpl
a
in
66.2%
of
da
ta
v
a
r
ia
bi
li
ty
,
w
hi
le
t
he
r
e
s
t
w
hi
c
h
in
f
lu
e
nc
e
d
by
ot
he
r
f
a
c
to
r
s
not
in
c
lu
de
d
in
th
e
m
ode
l.
A
ddi
ti
ona
ll
y,
th
e
ove
r
a
ll
e
r
r
or
va
lu
e
of
0
.003331 li
s
te
d i
n t
he
a
na
ly
s
is
r
e
s
ul
ts
c
onf
ir
m
s
th
e
s
ta
bi
li
ty
of
th
e
m
ode
l
dur
in
g
th
e
tr
a
in
in
g
pr
oc
e
s
s
. T
hus
,
th
e
ne
ur
a
l
ne
twor
k
va
li
da
ti
on
r
e
s
ul
ts
c
onf
ir
m
th
a
t
Erro
r:
0
.0
0
3
3
3
1
St
e
p
s
:
262
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
, V
ol
.
14
, N
o.
5
,
O
c
to
be
r
20
25
:
3771
-
3780
3776
th
e
m
ode
l
c
a
n
e
f
f
e
c
ti
ve
ly
id
e
nt
if
y
th
e
r
e
la
ti
ons
hi
p
be
twe
e
n
hi
s
to
r
ic
a
l
va
r
ia
bl
e
s
a
nd
e
duc
a
ti
on
c
om
pl
e
ti
on
r
a
te
s
, t
hus
s
uppor
ti
ng t
he
va
li
di
ty
of
t
he
m
ode
l
in
m
a
ki
ng pr
e
di
c
ti
ons
.
T
he
ne
xt
s
te
p
in
th
i
s
r
e
s
e
a
r
c
h
a
im
s
to
id
e
nt
if
y
th
e
opt
im
a
l
c
lu
s
te
r
c
ount
ba
s
e
d
on
th
e
ne
ur
a
l
ne
twor
k
pr
e
di
c
ti
on
r
e
s
ul
ts
us
e
d
in
pe
r
f
or
m
in
g
K
-
m
e
a
ns
c
lu
s
te
r
in
g
.
C
lu
s
te
r
a
na
ly
s
i
s
is
a
m
e
th
od
in
un
s
upe
r
vi
s
e
d
le
a
r
ni
ng
th
a
t
s
e
e
ks
to
di
vi
de
da
ta
in
to
s
pe
c
if
ic
gr
oups
,
e
ns
ur
in
g
th
a
t
e
le
m
e
nt
s
w
it
hi
n
a
c
lu
s
te
r
ha
ve
hom
oge
ne
ous
pr
ope
r
ti
e
s
,
w
hi
le
di
f
f
e
r
e
nc
e
s
be
tw
e
e
n
c
lu
s
te
r
s
a
r
e
m
a
de
a
s
c
le
a
r
a
s
po
s
s
ib
le
[
34]
.
T
o
a
c
hi
e
ve
th
is
goa
l,
th
e
e
lb
ow
m
e
th
od
is
a
ppl
ie
d,
a
vi
s
ua
l
a
ppr
oa
c
h
th
a
t
i
ll
us
tr
a
te
s
th
e
r
e
la
ti
ons
hi
p
be
twe
e
n
th
e
num
be
r
of
c
lu
s
te
r
s
a
nd
th
e
to
ta
l
W
S
S
va
lu
e
.
T
he
gr
a
ph
s
ge
ne
r
a
te
d
by
th
e
e
lb
ow
m
e
th
od
pr
ovi
de
gui
da
n
c
e
in
id
e
nt
if
yi
ng
th
e
m
os
t
s
ta
ti
s
ti
c
a
ll
y
a
nd
in
te
r
pr
e
ta
ti
ve
ly
opt
im
a
l
num
be
r
of
c
lu
s
te
r
s
.
F
ig
ur
e
3
s
how
s
th
e
num
be
r
of
c
lu
s
te
r
s
ge
ne
r
a
te
d.
F
ig
ur
e
3.
O
pt
im
a
l
num
be
r
of
c
lu
s
te
r
s
B
a
s
e
d
on
th
e
r
e
s
ul
ti
ng
e
lb
ow
m
e
th
od
g
r
a
ph,
it
c
a
n
be
s
e
e
n
th
a
t
th
e
to
ta
l
W
S
S
va
lu
e
de
c
r
e
a
s
e
s
s
ha
r
pl
y
f
r
om
k=
1k,
k=
2k
,
a
nd
k=
3k.
A
f
te
r
k=
3k,
th
e
de
c
r
e
a
s
e
in
W
S
S
va
lu
e
b
e
c
om
e
s
s
lo
w
e
r
a
nd
in
s
ig
ni
f
ic
a
nt
.
T
hi
s
pa
tt
e
r
n
in
di
c
a
te
s
th
a
t
th
e
us
e
of
th
r
e
e
c
lu
s
te
r
s
is
th
e
opt
im
a
l
c
hoi
c
e
,
a
s
th
e
num
be
r
of
a
ddi
ti
ona
l
c
lu
s
te
r
s
a
f
te
r
3
k
doe
s
not
pr
ovi
de
a
s
ig
ni
f
ic
a
nt
im
pr
ove
m
e
nt
in
r
e
duc
in
g
w
it
hi
n
-
c
lu
s
te
r
va
r
ia
ti
on.
T
he
r
e
f
or
e
,
th
e
opt
im
a
l
num
be
r
of
c
lu
s
te
r
s
c
hos
e
n
in
K
-
m
e
a
n
s
c
lu
s
t
e
r
in
g
a
na
ly
s
i
s
i
s
th
r
e
e
c
lu
s
te
r
s
,
w
hi
c
h
pr
ovi
de
s
a
ba
la
nc
e
be
tw
e
e
n
th
e
m
ode
l'
s
pe
r
f
or
m
a
nc
e
in
r
e
pr
e
s
e
nt
in
g
da
ta
va
r
ia
ti
on
a
nd
th
e
c
om
pl
e
xi
ty
of
th
e
m
ode
l
f
or
m
e
d.
K
-
m
e
a
ns
a
lg
or
it
hm
is
a
n
a
lg
or
it
hm
w
it
h
pa
r
ti
ti
oni
ng
[
3
5]
,
be
c
a
us
e
K
-
m
e
a
ns
m
e
th
o
d
r
e
qu
i
r
e
s
de
te
r
m
in
in
g
t
h
e
nu
m
be
r
o
f
c
l
us
te
r
s
a
t
th
e
be
g
in
ni
ng
o
f
t
he
p
r
o
c
e
s
s
,
th
e
a
lg
o
r
i
th
m
s
ta
r
ts
b
y
s
e
t
ti
ng
a
n
in
it
ia
l
c
e
nt
r
o
id
va
l
ue
t
ha
t
w
i
ll
s
e
r
v
e
a
s
t
he
c
e
nt
e
r
p
oi
nt
f
o
r
e
a
c
h
g
r
oup
[
3
6]
.
T
he
K
-
m
e
a
ns
a
lg
o
r
i
th
m
r
e
qu
ir
e
s
a
de
f
in
it
e
nu
m
be
r
of
c
lu
s
te
r
s
t
o
be
de
t
e
r
m
i
ne
d
be
f
o
r
e
th
e
c
lu
s
te
r
i
ng
pr
oc
e
s
s
c
a
n
be
pe
r
f
or
m
e
d
[
37
]
,
be
c
a
us
e
th
e
i
ni
ti
a
l
p
os
i
ti
ons
o
f
t
he
c
lu
s
t
e
r
c
e
nt
e
r
s
c
a
n
va
r
y
,
t
hi
s
m
a
y
l
e
a
d
to
i
nc
o
ns
is
te
n
t
c
l
us
te
r
in
g
ou
tc
o
m
e
s
f
or
th
e
da
ta
.
T
h
is
c
lu
s
t
e
r
in
g
a
i
m
s
to
pr
e
s
e
n
t
a
n
ua
n
c
e
d
pe
r
s
pe
c
ti
v
e
of
t
he
d
if
f
e
r
e
nc
e
s
a
nd
s
im
il
a
r
it
ie
s
be
twe
e
n
pr
ovi
nc
e
s
r
e
la
te
d
t
o h
ig
h s
c
hoo
l
gr
a
du
a
t
io
n
r
a
te
s
, w
hi
c
h
c
a
n
b
e
us
e
d t
o
und
e
r
s
ta
n
d
r
e
gi
o
na
l
d
yna
m
ic
s
m
o
r
e
c
om
p
r
e
he
ns
iv
e
ly
.
T
he
f
o
ll
ow
in
g
is
th
e
r
e
s
u
lt
o
f
c
l
us
te
r
in
g
p
r
ovi
nc
e
s
us
in
g
t
he
K
-
m
e
a
ns
a
lg
or
it
hm
s
ho
w
n
in
F
i
gu
r
e
4.
B
a
s
e
d
o
n
F
ig
u
r
e
4,
w
hi
c
h
p
r
e
s
e
nt
s
th
e
ou
tc
o
m
e
s
o
f
t
he
c
lu
s
te
r
in
g
a
na
ly
s
is
pe
r
f
o
r
m
e
d
w
it
h
t
he
K
-
m
e
a
ns
a
lg
or
it
hm
c
lu
s
te
r
i
ng
m
e
t
hod
,
t
he
t
h
r
e
e
c
lu
s
t
e
r
s
f
o
r
m
e
d
a
r
e
v
is
ua
li
z
e
d
w
it
h
di
f
f
e
r
e
n
t
s
y
m
bo
ls
a
nd
c
ol
o
r
s
to
f
a
c
il
it
a
te
in
te
r
p
r
e
t
a
t
io
n
.
T
h
is
v
is
ua
li
z
a
t
io
n
s
how
s
th
e
di
s
tr
ib
ut
io
n
o
f
p
r
o
vi
nc
e
s
b
a
s
e
d
on
th
e
le
v
e
l
of
h
ig
h
s
c
ho
ol
c
o
m
p
le
t
io
n
in
e
a
c
h
p
r
ov
in
c
e
.
D
i
m
e
ns
io
n
1
(
9
0.5
%
)
a
nd
d
im
e
ns
i
on
2
(
4.
5
%
)
r
e
p
r
e
s
e
nt
t
he
pr
in
c
ip
a
l
c
o
m
po
ne
n
ts
r
e
s
ul
ti
ng
f
r
om
d
im
e
ns
i
on
r
e
d
uc
t
io
n
us
in
g
th
e
p
r
in
c
ip
a
l
c
om
po
ne
n
t
a
na
ly
s
is
(
P
C
A
)
m
e
t
hod
,
w
hi
c
h
c
ov
e
r
s
a
to
ta
l
o
f
95
%
o
f
th
e
da
ta
va
r
ia
bi
li
ty
.
E
a
c
h
po
in
t
o
n
th
e
g
r
a
p
h
r
e
pr
e
s
e
nt
s
t
he
pos
it
io
n
of
e
a
c
h
p
r
ovi
nc
e
in
th
e
di
m
e
ns
io
n
s
pa
c
e
.
T
he
f
o
ll
ow
i
ng
is
a
n
in
te
r
p
r
e
ta
t
io
n
o
f
t
he
da
ta
b
a
s
e
d
on
t
he
c
lu
s
te
r
s
f
o
r
m
e
d.
C
lu
s
t
e
r
one
(
r
e
d
)
,
w
hi
c
h
c
o
ns
is
ts
o
f
m
os
t
p
r
o
vi
nc
e
s
s
uc
h
a
s
C
e
nt
r
a
l
J
a
va
,
E
a
s
t
J
a
va
,
a
n
d
N
o
r
t
h
S
u
m
a
t
r
a
,
ha
s
m
ode
r
a
te
le
v
e
ls
o
f
e
duc
a
ti
on
c
o
m
p
le
t
io
n.
P
r
o
vi
n
c
e
s
in
th
is
c
lu
s
te
r
f
a
c
e
a
nu
m
be
r
o
f
c
ha
l
le
n
ge
s
,
i
nc
l
ud
in
g
v
a
r
ia
t
io
ns
in
e
d
uc
a
ti
o
n
a
c
c
e
s
s
i
bi
li
ty
,
i
ne
qu
a
l
it
y
in
in
f
r
a
s
t
r
uc
t
ur
e
,
l
im
it
e
d
r
e
s
o
u
r
c
e
s
a
nd
t
he
ne
e
d
to
im
p
r
ove
t
e
a
c
h
in
g
qua
li
ty
.
N
e
ve
r
th
e
le
s
s
,
t
hi
s
c
l
us
te
r
ha
s
a
g
r
e
a
t
o
ppo
r
tu
n
it
y
to
im
pr
ove
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
D
at
a
-
dr
iv
e
n c
lu
s
te
r
in
g and pr
e
di
c
ti
on of hi
gh s
c
hool
gr
aduati
o
n r
at
e
s
i
n …
(
M
uhamm
ad Salman A
r
r
o
s
y
id
)
3777
e
duc
a
ti
on
p
e
r
f
o
r
m
a
nc
e
t
hr
oug
h
m
or
e
ta
r
g
e
te
d
po
li
c
ie
s
a
n
d
in
f
r
a
s
t
r
uc
t
ur
e
de
v
e
lo
pm
e
nt
th
a
t
s
u
pp
or
ts
e
qui
ta
bl
e
a
c
c
e
s
s
t
o
e
d
uc
a
ti
on.
T
he
s
e
c
ond
c
lu
s
te
r
(
gr
e
e
n)
in
c
lu
de
s
pr
ovi
nc
e
s
w
it
h
lo
w
e
duc
a
t
io
n
c
om
pl
e
ti
on
r
a
te
s
,
s
uc
h
a
s
P
a
pua
,
E
a
s
t
N
us
a
T
e
ngga
r
a
(
N
T
T
)
,
a
nd
W
e
s
t
S
ul
a
w
e
s
i,
w
hi
c
h
f
a
c
e
s
ig
ni
f
ic
a
nt
c
ha
ll
e
nge
s
s
uc
h
a
s
di
f
f
ic
ul
t
ge
ogr
a
phi
c
a
l
a
c
c
e
s
s
,
la
c
k
of
e
duc
a
ti
on
f
a
c
il
it
ie
s
a
nd
s
ta
r
k
s
oc
io
-
e
c
onomi
c
di
s
pa
r
it
ie
s
.
T
he
s
e
c
ha
ll
e
ng
e
s
r
e
qui
r
e
s
tr
a
te
gi
c
in
te
r
ve
nt
io
ns
s
uc
h
a
s
im
pr
ovi
ng
e
duc
a
ti
o
n
a
c
c
e
s
s
ib
il
it
y
in
r
e
m
ot
e
a
r
e
a
s
,
de
v
e
lo
pi
ng
s
uppor
ti
ve
in
f
r
a
s
tr
uc
tu
r
e
a
nd
r
e
duc
in
g
di
s
pa
r
it
ie
s
be
tw
e
e
n
r
e
gi
ons
.
T
he
c
lu
s
te
r
f
in
di
ngs
pr
ovi
de
im
por
ta
nt
in
s
ig
ht
s
in
to
th
e
va
r
ia
ti
ons
in
e
duc
a
ti
on
pe
r
f
or
m
a
nc
e
in
I
nd
one
s
ia
,
w
hi
c
h
c
a
n
in
f
or
m
th
e
f
or
m
ul
a
ti
on
of
da
ta
-
dr
iv
e
n e
duc
a
ti
on poli
c
ie
s
t
o i
m
pr
ove
t
he
qua
li
ty
a
nd e
qui
t
y of
e
duc
a
ti
on na
ti
onw
id
e
.
T
he
th
ir
d
c
lu
s
te
r
(
bl
ue
)
in
c
lu
de
s
pr
ovi
nc
e
s
w
it
h
hi
gh
e
duc
a
ti
on
c
om
pl
e
ti
on
r
a
te
s
,
s
u
c
h
a
s
D
K
I
J
a
ka
r
ta
a
nd
D
I
Y
ogya
ka
r
ta
.
P
r
ovi
nc
e
s
in
th
is
c
lu
s
t
e
r
ha
ve
e
xc
e
ll
e
nt
a
c
c
e
s
s
to
e
duc
a
ti
on,
a
de
qu
a
te
in
f
r
a
s
tr
uc
tu
r
e
a
nd
e
f
f
e
c
ti
ve
im
pl
e
m
e
nt
a
ti
on
of
e
duc
a
ti
on
pol
ic
ie
s
.
T
h
e
s
e
f
a
c
to
r
s
e
na
bl
e
th
e
pr
ovi
nc
e
s
in
th
is
c
lu
s
te
r
t
o c
ons
is
te
nt
ly
a
c
hi
e
ve
e
du
c
a
ti
on pe
r
f
or
m
a
nc
e
a
bove
t
he
na
ti
ona
l
a
ve
r
a
ge
.
F
ig
ur
e
4. V
is
ua
li
z
a
ti
on of
K
-
m
e
a
ns
a
lg
or
it
hm
r
e
s
ul
ts
4.
C
O
N
C
L
U
S
I
O
N
T
hi
s
s
tu
dy
e
xa
m
in
e
s
th
e
gr
a
dua
ti
on
r
a
te
of
s
e
ni
or
hi
gh
s
c
hool
e
duc
a
ti
on
in
34
pr
ovi
nc
e
s
in
I
ndone
s
ia
f
r
om
2015
-
2023
by
ut
il
iz
in
g
a
c
om
bi
na
ti
on
of
ne
ur
a
l
ne
twor
k
a
nd
K
-
m
e
a
ns
c
lu
s
te
r
in
g
m
e
th
ods
.
T
he
ne
ur
a
l
ne
twor
k
m
ode
l
s
how
e
d
s
upe
r
io
r
a
bi
li
ty
in
pr
e
di
c
ti
ng
e
duc
a
ti
on
c
om
pl
e
ti
on
r
a
te
s
w
it
h
a
M
S
E
of
0.0001936,
r
e
f
le
c
ti
ng
it
s
a
bi
li
ty
to
c
a
pt
ur
e
non
-
li
ne
a
r
r
e
la
ti
ons
hi
ps
f
r
om
hi
s
to
r
ic
a
l
da
ta
.
T
he
r
e
s
ul
ti
ng
pr
e
di
c
ti
ons
w
e
r
e
th
e
n
us
e
d
in
a
K
-
m
e
a
ns
c
lu
s
te
r
in
g
a
na
ly
s
i
s
to
gr
oup
th
e
pr
ovi
nc
e
s
in
to
th
r
e
e
m
a
in
c
lu
s
te
r
s
:
hi
gh,
m
e
di
um
,
a
nd
lo
w
c
om
pl
e
ti
on
r
a
te
s
.
T
he
c
om
bi
ne
d
a
ppr
oa
c
h
of
ne
ur
a
l
ne
twor
k
a
nd
K
-
m
e
a
ns
c
lu
s
t
e
r
in
g
pr
ovi
de
s
a
c
om
pr
e
he
ns
iv
e
pi
c
tu
r
e
of
e
duc
a
ti
on
di
s
pa
r
it
ie
s
in
I
ndone
s
ia
.
T
he
f
in
di
ngs
of
th
is
s
tu
dy
of
f
e
r
a
s
tr
ong
ba
s
is
f
or
da
ta
-
dr
iv
e
n
pol
ic
y
m
a
ki
ng
to
im
pr
ove
e
duc
a
ti
on
qua
li
ty
a
nd
r
e
duc
e
di
s
pa
r
it
ie
s
a
m
ong
pr
ovi
nc
e
s
in
I
ndone
s
ia
.
F
ut
ur
e
r
e
s
e
a
r
c
h
by
in
c
lu
di
ng
a
ddi
ti
ona
l
va
r
ia
bl
e
s
is
e
xpe
c
te
d
to
br
oa
d
e
n
th
e
a
na
ly
s
is
a
nd s
tr
e
ngt
he
n poli
c
y r
e
c
om
m
e
nda
ti
ons
t
ha
t
a
r
e
m
or
e
t
a
r
ge
te
d.
A
C
K
N
O
WL
E
D
G
E
M
E
N
T
S
T
hi
s
r
e
s
e
a
r
c
h
w
oul
d
not
ha
ve
be
e
n
po
s
s
ib
le
w
it
hout
th
e
s
uppor
t
of
va
r
io
us
pa
r
ti
e
s
.
O
ur
a
ppr
e
c
ia
ti
on
goe
s
to
th
e
C
e
nt
r
a
l
B
ur
e
a
u
of
S
ta
ti
s
ti
c
s
or
B
a
da
n
P
us
a
t
S
ta
ti
s
ti
k
(
B
P
S
)
f
o
r
pr
ovi
d
in
g
th
e
e
duc
a
ti
on
da
ta
o
n
w
hi
c
h t
hi
s
r
e
s
e
a
r
c
h i
s
ba
s
e
d, a
nd t
o our
i
ns
ti
tu
ti
on f
or
t
he
f
a
c
il
it
ie
s
a
nd oppor
tu
ni
ti
e
s
pr
ovi
de
d. W
e
a
ls
o t
ha
nk
our
f
e
ll
ow
r
e
s
e
a
r
c
he
r
s
f
or
th
e
ir
va
lu
a
bl
e
in
put
s
,
w
hi
c
h
e
nr
ic
he
d
th
e
qua
li
ty
of
th
is
s
tu
dy.
S
pe
c
ia
l
th
a
nk
s
go
to
P
r
of
.
E
r
na
A
ndr
iy
a
nt
i,
S
.S
., M
.H
um
., P
h.D
.,
f
or
he
r
i
ns
ig
ht
f
ul
g
ui
da
nc
e
, s
ugge
s
ti
on
s
, a
nd c
ont
in
uous
s
uppor
t,
w
hi
c
h gr
e
a
tl
y he
lp
e
d i
n t
he
c
om
pl
e
ti
on of
t
hi
s
a
r
ti
c
le
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
, V
ol
.
14
, N
o.
5
,
O
c
to
be
r
20
25
:
3771
-
3780
3778
F
U
N
D
I
N
G
I
N
F
O
R
M
A
T
I
O
N
T
hi
s
r
e
s
e
a
r
c
h
w
a
s
f
unde
d
by
th
e
C
e
nt
e
r
f
or
F
in
a
nc
in
g
a
n
d
A
s
s
e
s
s
m
e
nt
of
H
ig
he
r
E
duc
a
ti
on
(
P
P
A
P
T
)
, M
in
is
tr
y of
E
duc
a
ti
on, C
ul
tu
r
e
, R
e
s
e
a
r
c
h a
nd T
e
c
hn
ol
ogy (
K
e
m
di
kbudr
is
te
k)
, t
hr
ough the
de
c
is
io
n
of
th
e
H
e
a
d
of
th
e
C
e
nt
e
r
f
or
E
duc
a
ti
on
F
in
a
nc
in
g,
K
e
m
di
kbudr
is
te
k,
N
o.
00571/B
P
P
T
/B
P
I
.06/
9/
2023.
A
ddi
ti
ona
l
s
uppor
t
w
a
s
pr
ovi
de
d
by
th
e
E
duc
a
ti
on
F
und
M
a
na
ge
m
e
nt
I
ns
ti
tu
ti
on
(
L
P
D
P
)
,
M
in
is
tr
y
of
F
in
a
nc
e
of
th
e
R
e
publ
ic
of
I
ndone
s
ia
,
in
th
e
f
o
r
m
of
a
s
c
hol
a
r
s
hi
p
a
w
a
r
de
d
to
th
e
f
if
th
a
ut
hor
.
T
he
a
ut
hor
s
a
ls
o a
c
knowle
dg
e
N
us
a
N
ip
a
U
ni
ve
r
s
it
y M
a
um
e
r
e
f
or
s
uppor
ti
ng t
he
f
if
th
a
ut
hor
'
s
a
c
a
de
m
ic
c
a
r
e
e
r
.
A
U
T
H
O
R
C
O
N
T
R
I
B
U
T
I
O
N
S
S
T
A
T
E
M
E
N
T
T
hi
s
jo
ur
na
l
us
e
s
th
e
C
ont
r
ib
ut
or
R
ol
e
s
T
a
xonomy
(
C
R
e
di
T
)
to
r
e
c
ogni
z
e
in
di
vi
dua
l
a
ut
hor
c
ont
r
ib
ut
io
ns
, r
e
duc
e
a
ut
hor
s
hi
p di
s
put
e
s
,
a
nd f
a
c
il
it
a
te
c
ol
la
bo
r
a
ti
on.
N
am
e
o
f
A
u
t
h
or
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
M
uha
m
m
a
d S
a
lm
a
n
A
r
r
os
yi
d
✓
✓
✓
✓
✓
✓
✓
✓
✓
M
a
r
z
uki
✓
✓
✓
✓
✓
✓
W
id
ih
a
s
tu
ti
✓
✓
✓
✓
✓
✓
✓
H
a
r
ya
nt
o
✓
✓
✓
✓
✓
✓
M
a
r
ia
A
nge
li
na
F
r
a
ns
is
ka
M
ba
r
i
✓
✓
✓
✓
✓
✓
✓
✓
C
:
C
onc
e
pt
ua
l
i
z
a
t
i
on
M
:
M
e
t
hodol
ogy
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
da
t
i
on
Fo
:
Fo
r
m
a
l
a
na
l
ys
i
s
I
:
I
nve
s
t
i
ga
t
i
on
R
:
R
e
s
our
c
e
s
D
:
D
a
t
a
C
ur
a
t
i
on
O
:
W
r
i
t
i
ng
-
O
r
i
gi
na
l
D
r
a
f
t
E
:
W
r
i
t
i
ng
-
R
e
vi
e
w
&
E
di
t
i
ng
Vi
:
Vi
s
ua
l
i
z
a
t
i
on
Su
:
Su
pe
r
vi
s
i
on
P
:
P
r
oj
e
c
t
a
dm
i
ni
s
t
r
a
t
i
on
Fu
:
Fu
ndi
ng a
c
qui
s
i
t
i
on
C
O
N
F
L
I
C
T
O
F
I
N
T
E
R
E
S
T
S
T
A
T
E
M
E
N
T
A
ut
hor
s
s
ta
te
no c
onf
li
c
t
of
i
nt
e
r
e
s
t.
I
N
F
O
R
M
E
D
C
O
N
S
E
N
T
W
e
ha
ve
obt
a
in
e
d i
nf
or
m
e
d c
ons
e
nt
f
r
om
a
ll
i
ndi
vi
dua
ls
i
nc
lu
de
d i
n t
hi
s
s
tu
dy.
E
T
H
I
C
A
L
A
P
P
R
O
V
A
L
T
he
r
e
s
e
a
r
c
h
r
e
la
te
d
to
hum
a
n
u
s
e
h
a
s
be
e
n
c
om
pl
ie
d
w
it
h
a
ll
th
e
r
e
le
va
nt
na
ti
ona
l
r
e
gul
a
ti
ons
a
nd
in
s
ti
tu
ti
ona
l
pol
ic
ie
s
in
a
c
c
or
da
nc
e
w
it
h
th
e
te
ne
ts
of
th
e
H
e
ls
i
nki
D
e
c
la
r
a
ti
on
a
nd
ha
s
be
e
n
a
ppr
ove
d
by
th
e
a
ut
hor
s
'
i
ns
ti
tu
ti
ona
l
r
e
vi
e
w
boa
r
d or
e
qui
va
le
nt
c
om
m
it
te
e
.
D
A
T
A
A
V
A
I
L
A
B
I
L
I
T
Y
T
he
da
ta
th
a
t
s
uppor
t
th
e
f
in
di
ngs
of
th
is
s
tu
dy
a
r
e
ope
nl
y
a
va
il
a
bl
e
in
B
adan
P
us
at
St
at
is
ti
k
or
C
e
nt
r
a
l
B
ur
e
a
u
of
S
ta
ti
s
ti
c
s
a
t
ht
tp
s
:/
/ww
w
.bp
s
.go.i
d/
id
/s
ta
ti
s
ti
c
s
-
ta
bl
e
/2
/M
T
k4
M
S
M
y/
ti
ngka
t
-
pe
nye
le
s
a
ia
n
-
pe
ndi
di
ka
n
-
m
e
nur
ut
-
je
nj
a
ng
-
pe
ndi
di
ka
n
-
da
n
-
w
il
a
ya
h.ht
m
l.
R
E
F
E
R
E
N
C
E
S
[
1]
M
.
C
.
R
.
-
J
i
m
é
ne
z
,
R
.
M
.
-
J
i
m
é
ne
z
,
A
.
L
.
-
G
ut
i
é
r
r
e
z
,
a
nd
E
.
G
-
M
a
r
t
í
,
“
S
t
ude
nt
s
’
a
t
t
i
t
ude
:
ke
y
t
o
unde
r
s
t
a
ndi
ng
t
he
i
m
pr
ove
m
e
nt
of
t
he
i
r
a
c
a
de
m
i
c
r
e
s
ul
t
s
i
n
a
f
l
i
ppe
d
c
l
a
s
s
r
oom
e
nvi
r
onm
e
nt
,”
I
nt
e
r
nat
i
onal
J
our
nal
of
M
anage
m
e
nt
E
duc
at
i
on
,
vol
.
20,
no.
2,
pp. 1
–
16, 2022, doi
:
10.1016/
j
.i
j
m
e
.2022.100635.
[
2]
C
.
C
ha
bbot
t
a
nd
M
.
S
i
nc
l
a
i
r
,
“
S
D
G
4
a
nd
t
he
C
O
V
I
D
-
19
e
m
e
r
ge
nc
y:
t
e
xt
b
ooks
,
t
ut
or
i
ng,
a
nd
t
e
a
c
he
r
s
,
”
P
r
os
p
e
c
t
s
,
vol
.
49
,
no. 1
–
2, pp. 51
–
57, 2020, doi
:
10.1007/
s
11125
-
020
-
09485
-
y.
[
3]
E
.
F
.
S
.
R
i
ni
,
F
.
T
.
A
l
di
l
a
,
a
nd
R
.
P
.
W
i
r
a
yuda
,
“
A
s
t
udy
of
s
t
ude
nt
l
e
a
r
ni
ng
di
s
c
i
pl
i
ne
i
n
s
e
ni
or
hi
gh
s
c
hool
,”
J
u
r
nal
I
l
m
i
ah
I
l
m
u
T
e
r
apan U
ni
v
e
r
s
i
t
as
J
am
bi
, vol
. 7, no. 1, pp. 33
–
37, 2023, doi
:
10.22437/
j
i
i
t
uj
.v7i
1.26698.
[
4]
M
.
S
.
A
r
r
os
yi
d,
F
.
F
a
khr
uddi
n,
a
nd
Z
.
E
l
m
uba
r
ok,
“
D
e
ve
l
opi
ng
a
n
a
t
t
i
t
ude
a
s
s
e
s
s
m
e
nt
s
i
ns
t
r
um
e
nt
a
nd
P
a
nc
a
s
i
l
a
s
t
ude
nt
pr
of
i
l
e
s
i
n i
s
l
a
m
i
c
r
e
l
i
gi
ous
e
duc
a
t
i
on s
ubj
e
c
t
,
”
J
our
nal
of
R
e
s
e
a
r
c
h and E
duc
at
i
onal
R
e
s
e
ar
c
h E
v
al
uat
i
on
, vol
. 11, no. 2, 2022.
[
5]
M
.
H
uda
,
A
.
S
udr
a
j
a
t
,
R
.
M
uh
a
m
a
t
,
K
.
S
.
M
.
T
e
h,
a
nd
B
.
J
a
l
a
l
,
“
S
t
r
e
ngt
he
ni
ng
di
vi
ne
va
l
ue
s
f
or
s
e
l
f
-
r
e
gul
a
t
i
on
i
n
r
e
l
i
gi
os
i
t
y:
i
ns
i
ght
s
f
r
om
T
a
w
a
kkul
(
t
r
us
t
i
n
G
od)
,”
I
nt
e
r
nat
i
onal
J
our
nal
of
E
t
hi
c
s
and
Sy
s
t
e
m
s
,
vol
.
35,
no.
3,
pp.
323
–
344,
2019,
doi
:
10.1108/
i
j
oe
s
-
02
-
2018
-
0025.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
D
at
a
-
dr
iv
e
n c
lu
s
te
r
in
g and pr
e
di
c
ti
on of hi
gh s
c
hool
gr
aduati
o
n r
at
e
s
i
n …
(
M
uhamm
ad Salman A
r
r
o
s
y
id
)
3779
[
6]
A
.
K
.
-
D
i
nya
,
“
M
a
na
gi
ng
c
ha
l
l
e
nge
s
of
i
nc
r
e
a
s
i
ng
c
om
pl
e
xi
t
y
i
n
s
u
s
t
a
i
na
bi
l
i
t
y,”
E
c
oc
y
c
l
e
s
,
vol
.
6,
no.
2,
pp.
49
–
53,
2020,
doi
:
10.19040/
e
c
oc
yc
l
e
s
.v6i
2.159.
[
7]
H
e
r
i
be
r
t
a
, Z
ul
f
a
ne
t
t
i
, a
nd R
. S
e
t
i
a
w
a
t
i
, “
E
xpl
or
i
ng t
he
i
m
pa
c
t
of
a
ge
a
nd m
ot
i
v
a
t
i
on on s
e
l
f
-
de
t
e
r
m
i
na
t
i
on:
i
ns
i
ght
s
f
r
om
i
nf
or
m
a
l
s
e
c
t
or
m
ot
he
r
s
,”
J
ur
nal
I
l
m
i
ah I
l
m
u T
e
r
apan U
ni
v
e
r
s
i
t
as
J
am
bi
, vol
. 8, no. 1, p
p. 306
–
321, 2024, doi
:
10.22437/
j
i
i
t
uj
.v8i
1.32880.
[
8]
J
.
J
e
ong,
D
.
H
ong,
a
nd
J
.
C
ha
ng,
“
A
na
l
ys
i
s
i
n
R
e
publ
i
c
of
K
or
e
a
:
l
ong
-
t
e
r
m
r
e
s
e
a
r
c
h
c
onduc
t
e
d,”
Sy
s
t
e
m
,
vol
.
11,
pp.
1
–
20,
2023.
[
9]
J
.
L
.
F
ua
,
I
.
S
.
W
e
kke
,
Z
.
S
a
ba
r
a
,
a
nd
R
.
U
.
N
ur
l
i
l
a
,
“
D
e
ve
l
opm
e
nt
of
e
nvi
r
onm
e
nt
a
l
c
a
r
e
a
t
t
i
t
ude
of
s
t
ude
nt
s
t
hr
ough
r
e
l
i
gi
o
n
e
duc
a
t
i
on
a
ppr
oa
c
h
i
n
I
ndone
s
i
a
,”
I
O
P
C
onf
e
r
e
nc
e
Se
r
i
e
s
:
E
ar
t
h
and
E
nv
i
r
onm
e
nt
al
Sc
i
e
nc
e
,
vol
.
175,
no.
1,
pp.
1
–
7,
2018,
doi
:
10.1088/
1755
-
1315/
175/
1/
012229.
[
10]
D
.
P
ur
ohi
t
,
M
.
J
a
ys
w
a
l
,
a
nd
A
.
M
udul
i
,
“
F
a
c
t
or
s
i
nf
l
ue
nc
i
ng
gr
a
du
a
t
e
j
ob
c
hoi
c
e
–
a
s
ys
t
e
m
a
t
i
c
l
i
t
e
r
a
t
ur
e
r
e
vi
e
w
,”
E
ur
ope
a
n
J
our
nal
of
T
r
ai
ni
ng and D
e
v
e
l
opm
e
nt
, vol
. 45, no. 4
–
5, pp. 381
–
401, 2021, doi
:
10.1108/
E
J
T
D
-
06
-
2020
-
0101.
[
11]
N
.
C
.
H
a
i
,
N
.
H
.
T
ha
nh,
T
.
M
.
C
ha
u,
T
.
V
.
S
a
ng,
a
nd
V
.
H
.
D
ong,
“
F
a
c
t
or
s
a
f
f
e
c
t
i
ng
t
he
de
c
i
s
i
on
t
o
c
hoos
e
a
uni
ve
r
s
i
t
y
of
hi
gh
s
c
hool
s
t
ude
nt
s
:
a
s
t
udy i
n
a
n G
i
a
ng
P
r
ovi
nc
e
, V
i
e
t
na
m
,”
I
nt
e
r
nat
i
onal
J
our
nal
of
E
v
al
uat
i
on and R
e
s
e
a
r
c
h i
n E
duc
at
i
on
, vol
. 12
,
no. 1, pp. 535
–
545, 2023, doi
:
10.11591/
i
j
e
r
e
.v12i
1.22971.
[
12]
C
.
C
a
r
r
i
c
o,
H
.
M
.
M
a
t
us
ovi
c
h,
a
nd
M
.
C
.
P
a
r
e
t
t
i
,
“
A
qua
l
i
t
a
t
i
ve
a
na
l
ys
i
s
of
c
a
r
e
e
r
c
hoi
c
e
pa
t
hw
a
ys
of
c
ol
l
e
ge
-
or
i
e
nt
e
d
r
ur
a
l
c
e
nt
r
a
l
a
ppa
l
a
c
hi
a
n
hi
gh
s
c
hool
s
t
ude
nt
s
,”
J
ou
r
nal
of
C
ar
e
e
r
D
e
v
e
l
opm
e
nt
,
vol
.
46,
no.
2,
pp.
94
–
111,
2019,
doi
:
10.1177/
0894845317725603.
[
13]
A
.
P
onc
i
ni
,
“
S
t
a
nda
r
ds
s
e
t
t
i
ng
i
n
r
e
l
i
gi
ous
e
du
c
a
t
i
on:
a
ddr
e
s
s
i
ng
t
he
qua
l
i
t
y
of
t
e
a
c
hi
ng
a
nd
a
s
s
e
s
s
m
e
nt
pr
a
c
t
i
c
e
s
,”
R
e
l
i
gi
ons
,
vol
. 14, no. 3, pp. 1
–
14, 2023, doi
:
10.3390/
r
e
l
14030315.
[
14]
Z
.
M
a
,
F
.
O
.
A
gy
e
m
a
n,
A
.
K
.
S
a
m
pe
ne
,
M
.
L
i
,
a
nd
S
.
B
ot
c
hw
a
y,
“
S
e
ni
or
hi
gh
s
c
hool
t
e
a
c
he
r
s
’
pr
e
f
e
r
e
nc
e
f
or
pr
of
e
s
s
i
ona
l
de
ve
l
opm
e
nt
:
a
c
a
s
e
s
t
udy
i
n
t
he
c
e
nt
r
a
l
r
e
gi
on
of
G
ha
na
unde
r
t
he
c
ur
r
e
nt
f
r
e
e
s
e
ni
or
hi
gh
s
c
hool
pol
i
c
y,”
SA
G
E
O
p
e
n
,
vol
.
13
,
no. 3, pp. 1
–
14, 2023, doi
:
10.1177/
21582440231185116.
[
15]
S
. I
r
i
a
nt
i
a
nd P
. P
r
a
s
e
t
yoput
r
a
, “
R
ur
a
l
–
ur
ba
n di
s
pa
r
i
t
i
e
s
i
n a
c
c
e
s
s
t
o i
m
pr
ove
d s
a
ni
t
a
t
i
on i
n I
ndone
s
i
a
:
a
de
c
om
pos
i
t
i
on a
ppr
oa
c
h,
”
SA
G
E
O
pe
n
, vol
. 11, no. 3, 2021, doi
:
10.1177/
21582440211029920.
[
16]
N
.
K
ühl
,
M
.
G
out
i
e
r
,
R
.
H
i
r
t
,
a
nd
G
.
S
a
t
z
ge
r
,
“
M
a
c
hi
ne
l
e
a
r
ni
ng
i
n
a
r
t
i
f
i
c
i
a
l
i
nt
e
l
l
i
ge
nc
e
:
t
ow
a
r
ds
a
c
om
m
on
unde
r
s
t
a
ndi
ng,”
i
n
P
r
oc
e
e
di
ngs
of
t
he
A
nnual
H
aw
ai
i
I
nt
e
r
nat
i
onal
C
onf
e
r
e
nc
e
on
Sy
s
t
e
m
Sc
i
e
nc
e
s
,
2019,
pp.
5236
–
5245
,
doi
:
10.24251/
H
I
C
S
S
.2019.630.
[
17]
I
.
H
.
S
a
r
ke
r
,
“
M
a
c
hi
ne
l
e
a
r
ni
ng:
a
l
gor
i
t
hm
s
, r
e
a
l
-
w
or
l
d
a
ppl
i
c
a
t
i
ons
a
nd
r
e
s
e
a
r
c
h
di
r
e
c
t
i
ons
,”
SN
C
om
put
e
r
Sc
i
e
n
c
e
,
vol
.
2,
no.
3,
2021, doi
:
10.1007/
s
42979
-
021
-
00592
-
x.
[
18]
R
.
C
.
F
ong,
W
.
J
.
S
c
he
i
r
e
r
,
a
nd
D
.
D
.
C
ox,
“
U
s
i
ng
hum
a
n
br
a
i
n
a
c
t
i
vi
t
y
t
o
g
ui
de
m
a
c
hi
ne
l
e
a
r
ni
ng,”
Sc
i
e
nt
i
f
i
c
R
e
por
t
s
,
vol
.
8,
no. 1, pp. 1
–
10, 2018, doi
:
10.1038/
s
41598
-
018
-
23618
-
6.
[
19]
M
.
L
.
H
o
a
ng
a
nd
N
.
D
e
l
m
ont
e
,
“
K
-
c
e
nt
r
oi
d
c
onve
r
ge
nc
e
c
l
us
t
e
r
i
ng
i
de
nt
i
f
i
c
a
t
i
on
i
n
one
-
l
a
be
l
pe
r
t
ype
f
or
di
s
e
a
s
e
pr
e
di
c
t
i
on,”
I
A
E
S
I
nt
e
r
nat
i
onal
J
our
nal
of
A
r
t
i
f
i
c
i
al
I
nt
e
l
l
i
ge
nc
e
, vol
. 13, no. 1, pp. 1149
–
1159, 2024, doi
:
10.11591/
i
j
a
i
.v13.i
1.pp1149
-
1159.
[
20]
J
.
M
a
c
Q
ue
e
n,
“
S
om
e
m
e
t
hods
f
or
c
l
a
s
s
i
f
i
c
a
t
i
on
a
nd
a
n
a
l
ys
i
s
of
m
ul
t
i
va
r
i
a
t
e
obs
e
r
va
t
i
ons
,”
P
r
oc
e
e
di
ng
s
of
t
he
f
i
f
t
h
B
e
r
k
e
l
e
y
s
y
m
pos
i
um
on m
at
he
m
at
i
c
al
s
t
at
i
s
t
i
c
s
and p
r
obabi
l
i
t
y
, vol
. 1, no. 14, pp. 281
–
2
97, 1967.
[
21]
M
.
A
l
ha
w
a
r
a
t
a
nd
M
.
H
e
ga
z
i
,
“
R
e
vi
s
i
t
i
ng
K
-
m
e
a
ns
a
nd
t
opi
c
m
ode
l
i
ng,
a
c
o
m
pa
r
i
s
on
s
t
udy
t
o
c
l
us
t
e
r
A
r
a
bi
c
doc
um
e
nt
s
,”
I
E
E
E
A
c
c
e
s
s
, vol
. 6, pp. 42740
–
42749, 2018, doi
:
10.1109/
A
C
C
E
S
S
.2018.2852648.
[
22]
A
.
T
s
i
t
s
ul
i
n,
J
.
P
a
l
ow
i
t
c
h,
B
.
P
e
r
oz
z
i
,
a
nd
E
.
M
ül
l
e
r
,
“
G
r
a
ph
c
l
u
s
t
e
r
i
ng
w
i
t
h
gr
a
ph
ne
ur
a
l
ne
t
w
or
ks
,”
J
our
nal
of
M
ac
hi
n
e
L
e
ar
ni
ng R
e
s
e
ar
c
h
, vol
. 24, pp. 1
–
21, 2023.
[
23]
C
. S
t
yl
i
a
nouda
ki
, I
. T
r
i
c
ha
ki
s
, a
nd G
.
P
. K
a
r
a
t
z
a
s
, “
M
ode
l
i
ng gr
oundw
a
t
e
r
ni
t
r
a
t
e
c
ont
a
m
i
na
t
i
on us
i
ng
a
r
t
i
f
i
c
i
a
l
ne
ur
a
l
ne
t
w
o
r
ks
,”
W
at
e
r
, vol
. 14, no. 7, pp. 1
–
15, 2022, doi
:
10.3390/
w
14071173.
[
24]
Y
. D
ua
n, Q
.
L
i
u,
a
nd S
. X
i
a
,
“
A
n
i
m
pr
ove
d i
ni
t
i
a
l
i
z
a
t
i
on c
e
nt
e
r
k
-
m
e
a
n
s
c
l
u
s
t
e
r
i
ng a
l
gor
i
t
hm
ba
s
e
d on di
s
t
a
nc
e
a
nd de
n
s
i
t
y,”
A
I
P
C
onf
e
r
e
nc
e
P
r
oc
e
e
di
ngs
, vol
. 1955, no. A
ugu
s
t
, 2018, doi
:
10.1063/
1.5033710.
[
25]
Y
. M
e
ng,
J
. L
i
a
ng,
F
. C
a
o, a
nd
Y
. H
e
, “
A
ne
w
di
s
t
a
nc
e
w
i
t
h d
e
r
i
va
t
i
ve
i
nf
or
m
a
t
i
on f
or
f
unc
t
i
ona
l
K
-
m
e
a
ns
c
l
us
t
e
r
i
ng a
l
gor
i
t
hm
,”
I
nf
or
m
at
i
on Sc
i
e
nc
e
s
, vol
. 463
–
464, pp. 166
–
185, 2018, doi
:
10.1016/
j
.i
ns
.201
8.06.035.
[
26]
W
.
Y
a
nbo,
L
.
L
i
,
P
.
X
i
nf
u,
a
nd
F
.
E
npe
ng,
“
L
oa
d
f
or
e
c
a
s
t
i
ng
ba
s
e
d
on
i
m
pr
ove
d
K
-
m
e
a
ns
c
l
us
t
e
r
i
ng
a
l
gor
i
t
hm
,”
C
hi
na
I
nt
e
r
nat
i
onal
C
onf
e
r
e
nc
e
on E
l
e
c
t
r
i
c
i
t
y
D
i
s
t
r
i
but
i
on, C
I
C
E
D
, pp. 2751
–
2755, 2
018, doi
:
10.1109/
C
I
C
E
D
.2018.8592023.
[
27]
T
.
S
.
X
u,
H
.
D
.
C
hi
a
ng,
G
.
Y
.
L
i
u,
a
nd
C
.
W
.
T
a
n,
“
H
i
e
r
a
r
c
hi
c
a
l
K
-
m
e
a
ns
m
e
t
hod
f
or
c
l
us
t
e
r
i
ng
l
a
r
ge
-
s
c
a
l
e
a
dva
nc
e
d
m
e
t
e
r
i
ng
i
nf
r
a
s
t
r
uc
t
ur
e
da
t
a
,”
I
E
E
E
T
r
ans
ac
t
i
ons
on
P
ow
e
r
D
e
l
i
v
e
r
y
,
vol
.
32,
no.
2,
pp.
609
–
616,
2017,
doi
:
10.1109/
T
P
W
R
D
.2015.2479941.
[
28]
K
.
P
.
S
i
na
ga
a
nd
M
.
S
.
Y
a
ng,
“
U
ns
upe
r
vi
s
e
d
K
-
m
e
a
ns
c
l
us
t
e
r
i
ng
a
l
gor
i
t
hm
,”
I
E
E
E
A
c
c
e
s
s
,
vol
.
8,
pp.
80716
–
80727,
2020,
doi
:
10.1109/
A
C
C
E
S
S
.2020.2988796.
[
29]
J
.
F
.
L
u,
J
.
B
.
T
a
ng,
Z
.
M
.
T
a
ng,
a
nd
J
.
Y
.
Y
a
ng,
“
H
i
e
r
a
r
c
hi
c
a
l
i
ni
t
i
a
l
i
z
a
t
i
on
a
ppr
oa
c
h
f
or
K
-
M
e
a
ns
c
l
us
t
e
r
i
ng,”
P
at
t
e
r
n
R
e
c
ogni
t
i
on L
e
t
t
e
r
s
, vol
. 29, no. 6, pp. 787
–
795, 2008, doi
:
10.1016/
j
.pa
t
r
e
c
.2007.12.009.
[
30]
A
r
ni
t
a
,
F
.
M
a
r
pa
ung,
F
.
R
a
m
a
dha
ni
,
a
nd
D
.
D
i
na
t
a
,
“
M
a
c
hi
ne
l
e
a
r
ni
ng
f
or
m
a
ppi
ng
a
nd
f
or
e
c
a
s
t
i
ng
pove
r
t
y
i
n
nor
t
h
s
um
a
t
e
r
a
:
a
da
t
a
-
dr
i
ve
n a
ppr
oa
c
h,”
Sai
ns
M
al
ay
s
i
ana
, vol
. 53, no. 7, pp. 1715
–
1728, 2024,
doi
:
10.17576/
j
s
m
-
2024
-
5307
-
18.
[
31]
K
.
L
.
D
u,
“
C
l
us
t
e
r
i
ng:
a
ne
ur
a
l
ne
t
w
or
k
a
pp
r
oa
c
h,”
N
e
ur
al
N
e
t
w
or
k
s
,
vol
.
23,
no.
1,
pp.
89
–
107,
2010,
doi
:
10.1016/
j
.ne
une
t
.2009.08.007.
[
32]
M
.
S
hut
a
yw
i
a
nd
N
.
N
.
K
a
c
houi
e
,
“
S
i
l
houe
t
t
e
a
na
l
ys
i
s
f
or
pe
r
f
or
m
a
nc
e
e
va
l
ua
t
i
on
i
n
m
a
c
hi
ne
l
e
a
r
ni
ng
w
i
t
h
a
ppl
i
c
a
t
i
ons
t
o
c
l
us
t
e
r
i
ng,”
E
nt
r
opy
, vol
. 23, no. 6, pp. 1
–
17, 2021, doi
:
10.3390/
e
23060759.
[
33]
M
.
C
ho,
K
.
A
.
-
V
a
hi
d,
S
.
A
dya
,
a
nd
M
.
R
a
s
t
e
ga
r
i
,
“
D
i
f
f
e
r
e
nt
i
a
bl
e
K
-
m
e
a
ns
c
l
us
t
e
r
i
ng
l
a
ye
r
f
or
ne
ur
a
l
ne
t
w
or
k
c
om
pr
e
s
s
i
on,”
i
n
I
C
L
R
2022
-
10t
h I
n
t
e
r
nat
i
onal
C
onf
e
r
e
nc
e
on L
e
a
r
ni
ng R
e
pr
e
s
e
nt
at
i
ons
, 2022
, pp. 1
–
19.
[
34]
V
.
M
e
l
nykov
a
nd
S
.
M
i
c
ha
e
l
,
“
C
l
us
t
e
r
i
ng
l
a
r
ge
da
t
a
s
e
t
s
by
m
e
r
gi
ng
k
-
m
e
a
ns
s
ol
ut
i
ons
,”
J
our
nal
of
C
l
as
s
i
f
i
c
at
i
on
,
vol
.
37,
no.
1,
pp. 97
–
123, 2020, doi
:
10.1007/
s
00357
-
019
-
09314
-
8.
[
35]
P
.
G
a
r
i
ka
pa
t
i
,
K
.
B
a
l
a
m
ur
uga
n,
T
.
P
.
L
a
t
c
houm
i
,
a
nd
R
.
M
a
l
ka
pur
a
m
,
“
A
c
l
us
t
e
r
-
pr
of
i
l
e
c
om
pa
r
a
t
i
ve
s
t
udy
on
m
a
c
hi
ni
n
g
A
l
S
i
7/
63%
of
S
i
C
hybr
i
d
c
om
pos
i
t
e
u
s
i
ng
a
ggl
om
e
r
a
t
i
ve
hi
e
r
a
r
c
hi
c
a
l
c
l
us
t
e
r
i
ng
a
nd
K
-
m
e
a
n
s
,”
Si
l
i
c
on
,
vol
.
13,
no.
4
,
pp. 961
–
972, 2021, doi
:
10.1007/
s
12633
-
020
-
00447
-
9.
[
36]
M
.
A
.
S
ya
kur
,
B
.
K
.
K
hot
i
m
a
h,
E
.
M
.
S
.
R
oc
hm
a
n,
a
nd
B
.
D
.
S
a
t
ot
o,
“
I
nt
e
gr
a
t
i
on
K
-
m
e
a
ns
c
l
us
t
e
r
i
ng
m
e
t
hod
a
nd
e
l
bow
m
e
t
hod
f
or
i
de
nt
i
f
i
c
a
t
i
on
o
f
t
he
be
s
t
c
us
t
om
e
r
pr
of
i
l
e
c
l
us
t
e
r
,”
I
O
P
C
onf
e
r
e
nc
e
Se
r
i
e
s
:
M
at
e
r
i
al
s
Sc
i
e
nc
e
and
E
ngi
ne
e
r
i
ng
,
vol
.
336,
no. 1, 2018, doi
:
10.1088/
1757
-
899X
/
336/
1/
012017.
[
37]
J
. K
a
uf
f
m
a
nn, M
. E
s
de
r
s
,
L
.
R
uf
f
, G
.
M
ont
a
von, W
.
S
a
m
e
k, a
nd
K
.
R
. M
ul
l
e
r
,
“
F
r
om
c
l
us
t
e
r
i
ng t
o c
l
us
t
e
r
e
xpl
a
na
t
i
ons
vi
a
ne
ur
a
l
ne
t
w
or
ks
,”
I
E
E
E
T
r
ans
ac
t
i
ons
on
N
e
u
r
al
N
e
t
w
or
k
s
and
L
e
a
r
ni
ng
Sy
s
t
e
m
s
,
vol
.
35,
no.
2,
pp.
1926
–
1940,
2024,
doi
:
10.1109/
T
N
N
L
S
.2022.3185901.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
, V
ol
.
14
, N
o.
5
,
O
c
to
be
r
20
25
:
3771
-
3780
3780
B
I
O
G
R
A
P
H
I
E
S
O
F
A
U
T
H
O
R
S
Muhammad
Salman
Arrosyid
,
S.Pd
.
,
M.Pd
.
obt
ain
ed
a
m
as
t
er'
s
d
egr
ee
i
n
Edu
cat
io
nal
R
ese
ar
ch
an
d E
va
lua
ti
on
at
Un
ive
rs
ita
s N
eg
er
i S
ema
ra
ng
, I
ndo
nes
ia
an
d a
b
ach
el
or'
s
deg
ree
f
rom
U
IN Ra
den
I
nta
n Lam
pu
ng,
L
amp
un
g, In
don
es
ia.
N
ow a st
ude
nt in
th
e Edu
ca
tio
na
l
Res
ear
ch
a
nd
E
va
lua
ti
on
D
oct
or
al
S
tud
y
Pr
og
ram
,
Un
iv
ers
it
as
N
ege
ri
Yog
yak
ar
ta
P
ost
gr
adu
at
e
Sch
ool
.
He
ca
n b
e
co
nta
ct
ed
at
em
ai
l:
mu
ham
ma
d00
39
pas
ca
.2
023
@
st
ud
ent
.un
y.
ac.
id
.
Prof.
Dr.
Marzuki,
M.Ag
.
completed
his
undergraduat
e
studies
at
the
Faculty
of
Tarbiyah,
UIN
Sunan
Kalijaga,
Yogyakarta,
in
1990.
He
earned
his
m
aster'
s
degree
in
Islamic
Studies
from
the
Gradu
ate
School
of
UIN
Syarif
Hidaya
tullah,
Jakar
ta,
in
1997,
and
completed
his
doctoral
degree
at
the
same
institution
in
2007.
In
202
1,
he
was
appointed
as
a
p
rofessor
in
the
field
of
Islamic
Religiou
s
Education
at
the
Facu
lty
of
Social
Sciences,
Universitas Negeri Yogy
akarta
. He can be contac
ted at email: marzuki@
uny.ac.id.
Dr.
Widihastuti,
S.Pd.,
M.Pd
.
o
btained
a
bachelor'
s
degree
in
E
ducation
from
UNY
in
1998.
Graduated
with
a
master'
s
degree
from
UNY
Postgra
duate
Program
in
2007.
Obtained
a
doctorate
in
Education
Research
and
Evaluation
from
the
UNY
Postgraduat
e
Progra
m in 201
4. She
can b
e con
tacte
d
at
email:
widihast
uti@
uny.ac.id
.
Dr.
Drs.
Ir.
Haryanto,
M.Pd.,
M.T
.
is
a
lecturer
at
the
Faculty
o
f
Engineering
and
Research
and
Educational
Evaluation
,
Universitas
Negeri
Yo
gyakarta
.
His
research
interests
focus
on
artificia
l
intelligence
control,
education
resear
ch,
and
technica
l
and
vocational education. He ca
n be contacted a
t
email:
haryanto@
uny.ac.id
.
Maria
Angelina
Fransiska
Mbari,
M.Pd
.
became
a
doctoral
student
in
the
Education
Research
and
Evaluation
Study
Program
at
the
Postgradua
te
School
of
Universitas
Negeri
Yogyakarta
.
She
currently
works
as
a
permanent
lecture
r
in
the
Primary
School
Teacher
Education
Study
Program
at
Universitas
Nusa
Nipa,
Maumere.
She
can
be
contacted
at
email:
anjelinaan
selmus@
gmail.co
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.