Indonesi
an
Journa
l
of El
ect
ri
cal Engineer
ing
an
d
Comp
ut
er
Scie
nce
Vo
l.
23
,
No.
3
,
Septem
ber
20
21
,
pp.
1654
~
1662
IS
S
N: 25
02
-
4752, DO
I: 10
.11
591/ijeecs
.v
23
.i
3
.
pp
1654
-
1662
1654
Journ
al h
om
e
page
:
http:
//
ij
eecs.i
aesc
or
e.c
om
Machin
e le
ar
nin
g depl
oy
ment f
or arms dy
namics p
attern
recogniti
on in So
uth
ea
st
Asia re
gion
Z
ul Indra
1
, A
z
ha
ri
Set
iaw
an
2
, Y
e
ssi J
usm
an
3
, Ari
sm
an
Ad
n
an
4
1
Depa
rtment of I
nform
at
ic
s E
ng
i
nee
ring
,
Abdurr
ab
Univer
si
t
y
,
In
donesia
2
Depa
rtment of I
nte
rna
ti
ona
l
R
elati
ons,
Abdurrab
Univer
sit
y
,
Indo
nesia
3
Depa
rtment of
El
e
ct
ri
ca
l
Eng
in
ee
ring
,
Muham
m
adiy
ah
Yog
y
a
kar
ta Unive
rsi
t
y
,
Indone
sia
4
Depa
rtment of
Mathe
m
at
i
cs,
R
i
au
Univer
si
t
y
,
In
donesia
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
Oct
27
,
2020
Re
vised
Ju
l
28
,
2021
Accepte
d
Aug
4
,
2021
Finding
the
m
os
t
signifi
c
ant
dete
rm
ina
nt
var
ia
b
le
of
arms
dy
nami
c
is
high
l
y
req
uire
d
due
to
strategic
polic
ie
s
form
ula
t
ions
and
power
m
appi
ng
for
ac
ad
emics
and
poli
c
y
m
ake
rs
.
Mac
hine
l
e
arn
i
ng
is
stil
l
n
ew
or
under
-
discussed
among
the
stud
y
of
poli
tics
and
inte
rna
ti
on
al
re
la
t
io
ns
.
Exi
sting
li
te
r
at
ur
e
hav
e
m
uch
foc
us
on
using
adva
nc
e
d
quant
i
ta
t
ive
m
et
hods
b
y
apply
ing
v
ari
ous
t
y
p
es
of
reg
r
ession
anal
y
s
is.
Th
is
stud
y
ana
l
y
z
e
d
the
a
rm
s
d
y
nami
c
in
Southea
st
As
ia
coun
tri
es
al
ong
with
it
s
som
e
strat
egi
c
par
tn
ers
su
ch
as
Unite
d
Stat
es,
China,
Russ
ia
,
South
Korea
,
and
Jap
an
b
y
using
‘Dec
ision
T
ree
’
of
m
ac
hin
e
l
e
arn
ing
al
gori
th
m
.
Thi
s
stud
y
conduc
t
ed
a
m
ac
hine
learni
n
g
ana
l
y
s
is
on
5
5
var
ia
bl
e
it
ems
which
is
cl
assifie
d
int
o
8
cl
asses
of
var
ia
b
le
s
videlicet
d
ef
ense
budge
t
,
a
r
m
s
tra
de
expor
ts
,
arms
tra
de
imports,
poli
t
ic
a
l
posture,
ec
ono
m
ic
posture,
se
cur
ity
postur
e
a
nd
def
ens
e
priori
t
y
,
n
at
ion
a
l
ca
pab
il
i
t
y
,
an
d
dire
ct
cont
a
ct,.
Th
e
result
s
suggest
three
findi
ngs:
(1)
st
ate
who
per
ce
iv
es
m
ari
ti
m
e
as
str
ategic
dr
ive
rs
and
f
orc
es
wil
l
see
k
m
ore
power
for
i
ts
m
ari
tim
e
def
ense
pos
ture
which
is
tr
ansla
t
ed
t
o
def
ense
budget,
(2)
big
size
count
rie
s
te
nd
to
be
an
arms
expor
t
er
coun
t
r
y
,
and
(3)
stat
e’s
e
ner
g
y
d
epe
nde
n
ce
ofte
n
l
ea
ds
t
o
a
highe
r
volu
m
e
of
arm
s
tra
nsfers b
et
wee
n
count
r
ie
s.
Ke
yw
or
ds:
Ar
m
s d
ynam
ics
Decisi
on tree
Ma
chine
le
a
rn
i
ng
Patt
ern
recog
ni
ti
on
Pr
e
processin
g
Southeast
Asia
This
is an
open
acc
ess arti
cl
e
un
der
the
CC
B
Y
-
SA
l
ic
ense
.
Corres
pond
in
g
Aut
h
or
:
Azh
a
ri Seti
awa
n
Dep
a
rtm
ent o
f In
te
r
natio
nal R
el
at
ion
s
Abd
urrab U
nive
rsity
73 Ri
au
R
oad,
Peka
nb
a
r
u,
Ri
a
u 282
91, In
done
sia
Em
a
il
: azhari.seti
awan
@
univ
r
ab.
ac.i
d
1.
INTROD
U
CTION
Ma
chine
l
earn
i
ng
is
co
ns
ide
re
d
as
on
e
of
the
m
os
t
essenti
al
su
bject
s
in
sci
entifi
c
stud
ie
s
nowa
days
wh
ic
h
the
nam
e
firstly
was
co
ined
in
1959
[
1]
,
[2
]
.
It
is
a
fie
ld
of
stu
dy
for
com
pu
te
r
sci
en
ce
and
i
nfor
m
at
ion
te
chnolo
gy
w
hi
ch
is
f
ocu
se
d
on
al
gorithm
s
and
sta
ti
sti
cal
m
od
e
ls.
Ma
chine
le
ar
ning
al
gorithm
s
are
use
d
in
a
wide
va
riet
y
of
a
pp
li
cat
ions
li
ke
em
ail
filt
e
rin
g
a
nd
com
pu
te
r
visio
n.
It
is
cl
os
el
y
relat
ed
to
com
pu
ta
ti
on
al
sta
ti
sti
cs,
wh
ic
h
f
oc
us
es
on
m
aking
predic
ti
on
s
us
in
g
c
om
pu
te
rs
[3]
.
Ma
chine
le
ar
ni
ng
as
the
s
ub
set
of
‘ar
ti
fici
al
intel
li
gen
ce’
is
use
d
to
e
ffec
ti
ve
ly
per
f
or
m
a
sp
eci
fic
ta
s
k
without
us
in
g
ex
plici
t
instru
ct
ions,
rely
ing
on
patt
ern
s
an
d
i
nf
e
re
nce
instea
d
[4]
.
The
patte
r
ns
and
infe
re
nces
are
util
iz
ed
f
or
va
rio
us
predi
ct
ive
analy
ti
c
s
wh
ic
h
is
essenti
al
fo
r
s
ocial
sci
ences,
especial
ly
po
li
ti
cal
sci
ence
and
inter
nat
ion
al
relat
ion
s
.
Ther
e
are
m
any
pr
obabili
ti
es,
ty
pe
of
inte
racti
ons,
an
d
dy
nam
i
cs
in
the
stu
di
es
to
form
ula
te
decisi
on
m
akin
g
.
Howe
ver,
re
ga
rd
i
ng
t
he
po
l
it
ic
al
analy
sis,
the
re
ha
ve
be
en
nu
m
ero
us
and
va
rio
us
researc
hes
t
ha
t
us
ed
adv
a
nce
d
sta
ti
sti
cal
m
et
ho
ds
bu
t
sti
ll
rar
e
in
us
i
ng
m
achine
le
arn
i
ng
as
t
he
m
ai
n
par
t
of
the
resea
rc
h.
Ba
se
d
Evaluation Warning : The document was created with Spire.PDF for Python.
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci
IS
S
N:
25
02
-
4752
Mac
hin
e le
ar
ni
ng d
e
plo
y
me
nt
for
ar
ms
d
y
na
mics
pa
tt
ern
…
(
Zu
l I
ndr
a
)
1655
on
inter
natio
na
l
facu
lt
y
s
urve
y
by
te
achi
ng
,
resea
rch
&
i
nt
ern
at
io
nal
pol
ic
y
(TRIP)
on
2359
prof
e
ssor
s
an
d
ind
ivi
du
al
who
are
af
fili
at
ed
with
poli
ti
cal
sci
ence
unit
,
pr
of
essi
onal
or
poli
cy
school
a
nd
resea
rch
i
nst
it
ute
world
wide
s
ho
wed
t
he
m
e
thods
w
hich
is
m
os
t
em
plo
ye
d
by
academ
ic
s
fr
om
var
io
us
in
sti
tuti
on
s
in
th
e
stud
y
of
i
nter
nationa
l
relat
io
ns
a
re:
qual
it
at
ive
an
al
ysi
s
(64.60
%
),
po
li
cy
a
naly
sis
(16.3
0%),
qu
a
ntit
at
ive
an
al
ysi
s
(6.80%
),
oth
er
(5.20%
),
pure
theo
ry
(
3.30%),
le
gal
or
et
hical
analy
sis
(2.50%
),
f
orm
al
m
od
el
li
ng
(
0.80%)
,
and
e
xp
e
rim
e
ntal
(0
.
50%)
[5]
.
The
qu
a
nt
it
at
ive
analy
s
is
was
the
3
rd
m
os
t
e
m
plo
ye
d
in
the
stud
y
of
internati
onal
re
la
ti
on
s.
T
he
refor
e
,
it
can
be
c
on
cl
ud
e
d
t
hat
the
stu
dy
for
i
m
ple
m
entat
ion
of
m
achine
le
arn
i
ng
al
gorithm
is still
u
nder
-
disc
usse
d
i
n
the
stu
dy
o
f
inter
natio
nal r
el
a
ti
ons.
This
a
rgum
ent
is
sup
ported
by
fa
ct
s
obt
ai
ned
from
the
co
nducted
li
te
ratur
e
stu
dies
relat
ed
t
o
researc
h
on
qu
antit
at
ive
poli
ti
cs
an
d
inter
nat
ion
al
relat
ion
s
hip
in
t
he
la
st
10
ye
ars
[
6]
-
[
15]
.
Durin
g
t
he
la
st
10
ye
ars,
m
os
t
of
the
stud
ie
s
trie
d
to
fin
d
an
infe
ren
ti
al
insigh
t
su
c
h
as
pr
ob
a
bili
ty
,
causati
on
,
an
d
ef
fe
ct
s
by
us
in
g
certai
n
m
od
el
s
of
regr
ession.
All
of
tho
s
e
pr
e
vious
r
esearc
hes
ha
ve
no
t
ap
plied
pa
tt
ern
recog
niti
on
f
or
the
analy
sis
y
et
.
He
nce,
t
his
stu
dy
is
ai
m
ed
to
co
nduct
resear
ch
on
t
he
dynam
ic
s
of
t
he
a
rm
s
trade
in
Southeast
Asia
us
in
g
m
achine
le
arn
in
g
al
go
r
it
h
m
s
in
order
to
ta
ckle
the
is
su
e
of
this
rese
arch
ga
p.
T
hi
s
stud
y
tr
ie
s
to
e
xp
l
ore
the
patte
rn
s
of
arm
s
dynam
ic
s
in
Southeast
A
sia
re
gion
c
ountries
fo
c
us
i
ng
on
sever
a
l
prom
inent
var
i
ables,
nam
el
y
(1)
eco
nom
ic
po
st
ur
e
,
(
2)
poli
ti
cal
po
sture
,
(3)
co
ntig
uiti
es
,
an
d
(
4)
de
fen
se
po
st
ur
e
a
nd
secur
it
y
pr
i
or
it
iz
at
ion
.
Th
e
a
naly
sis
gen
e
ra
te
d
a
m
od
el
/p
at
te
rn
of
arm
s
trade
dyna
m
i
cs
by
app
ly
in
g ‘D
eci
sion Tree
’
al
go
rithm
.
2.
RESEA
R
CH MET
HO
D
As
sta
te
d
i
n
intr
oductio
n
se
ct
ion
,
t
his
stu
dy
is
aim
ed
to
an
al
yz
e
the
arm
s
trade
dy
nam
ic
s
in
Southeast
Asia
reg
io
n
by
usi
ng
the
al
gorithm
s
of
m
achine
le
arn
i
ng.
H
ow
e
ve
r,
this
stu
dy
di
scov
e
red
t
hat
there
are
th
ree
m
ai
n
chall
enges
to
a
chieve
t
his
go
a
l
are
(i)
ho
w
to
identify
a
nd
c
ollec
t
the
ap
pr
opriat
e
data,
(ii)
ho
w
to
prepa
re
t
he
colle
ct
ed
data
to
be
a
ble
proc
essed
by
m
achine
le
ar
ning
an
d
(iii
)
wh
at
is
t
he
m
os
t
fit
m
a
chin
e
le
arn
in
g
al
gorithm
to
be
em
plo
ye
d
t
o
a
naly
ze the
data.
This
stu
dy
is
div
ide
d
int
o
three
sta
ges
na
m
el
y
data
collecti
on
,
data
pr
e
-
proce
ssin
g
a
nd
m
achine
le
arn
in
g
de
pl
oym
ent
to
ta
ckle
these
chall
en
ges.
Ba
se
d
on
the
li
te
ratur
e
r
eview
that
has
been
ca
rr
ie
d
ou
t,
it
was
f
ound
that
there
are
se
ve
ral
cat
egories
of
data
that
ha
ve
to
be
ob
ta
in
ed
duri
ng
data
colle
ct
ion
sta
ge
.
The
necessa
ry
data
are
the
n
el
a
borated
t
o
fi
nd
pa
tt
ern
s
of
de
fe
ns
e
budget
,
th
e
num
ber
of
i
m
po
rts
an
d
e
xport
s
of
a
rm
s
fo
r
c
ount
ries
in
the
Sout
heast
Asia
reg
i
on
by
us
in
g
m
achine
le
ar
ning
al
gorithm
s
.
The
m
achine
le
arn
i
ng
al
gorithm
cho
s
en
in
t
his
stu
dy
is
a
decisi
on
tree
al
gorithm
since
it
can
ge
ner
at
e
a
m
od
el
of
decisi
on
ru
l
es
that
are
easy
to
under
sta
nd
[16],
[17]
.
I
n
ad
diti
on,
this
al
gorithm
is
ver
y
us
e
fu
l
in
the
data
exp
lo
rati
on
pr
ocess
[18],
[
19
]
.
It
is
reall
y
us
efu
l
al
gorithm
to
fi
nd
t
he
hi
dd
e
n
relat
ion
s
hip
s
be
tween
in
put
va
riables
an
d
a
ta
rg
et
var
ia
ble.
He
nc
e,
decisi
on
tre
e
is
an
a
ppr
opr
ia
te
al
go
rithm
to
get
m
od
el
from
a
set
of
data
colle
ct
ion
since
it
com
bin
es
data
exp
l
or
at
io
n
an
d
m
od
el
ing
.
T
o
accom
m
od
at
e
the
us
e
of
thi
s
decisi
on
tree
al
gorithm
,
this
study
pro
po
ses
a
novel
al
gorithm
for
data
pre
-
processin
g
t
o
en
able
the
c
ollec
te
d
d
at
a
ca
n
be
processe
d
by
this
chosen
alg
or
it
hm
. Th
e overall
arch
it
ect
ure
of
researc
h
m
et
hod
to
con
du
ct
t
hi
s stud
y i
s
il
lustrate
d
Fi
gure
1
.
D
a
t
a
C
o
l
l
e
c
t
i
o
n
o
f
a
r
m
s
d
y
n
a
m
i
c
s
i
n
S
o
u
t
h
e
a
s
t
A
s
i
a
P
r
o
p
o
s
e
d
P
r
e
p
r
o
c
e
s
s
i
n
g
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
D
e
p
l
o
y
m
e
n
t
N
a
t
i
o
n
a
l
c
a
p
a
b
i
l
i
t
i
e
s
s
e
c
u
r
i
t
y
p
r
i
o
r
i
t
i
z
a
t
i
o
n
&
d
e
f
e
n
s
e
p
o
s
t
u
r
e
e
c
o
n
o
m
i
c
p
o
s
t
u
r
e
p
o
l
i
t
i
c
a
l
p
o
s
t
u
r
e
s
t
a
t
e
’
s
c
o
n
t
i
g
u
i
t
i
e
s
O
b
t
a
i
n
K
n
o
w
l
e
d
g
e
/
P
a
t
t
e
r
n
u
s
i
n
g
D
e
c
i
s
i
o
n
T
r
e
e
C
o
n
v
e
r
t
n
u
m
e
r
i
c
d
a
t
a
i
n
t
o
l
a
b
e
l
u
s
i
n
g
p
r
o
p
o
s
e
d
p
r
e
p
r
o
c
e
s
s
i
n
g
s
t
a
r
t
e
n
d
Figure
1
.
Re
se
arch
f
ram
ewo
r
k
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2502
-
4752
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci,
Vo
l.
23
, N
o.
3
,
Se
ptem
ber
20
21
:
16
54
-
16
62
1656
2.1.
Data
c
ollec
tion st
age
As
in
gen
e
ral
s
tud
ie
s
relat
ed
t
o
m
achine
le
arn
in
g
w
hich
is
us
e
d
f
or
data
analy
ti
cs,
the
early
sta
ge
for
this
stu
dy
is
da
ta
colle
ct
ion
.
As
pr
e
viously
m
entioned,
the
first
c
halle
nge
to
co
nduct
t
his
stu
dy
is
how
t
o
colle
ct
the
ap
pro
pr
ia
te
dat
a
f
or
a
rm
s
dyna
m
ic
s
in
Southe
ast
Asia
w
hich
is
then
analy
z
ed
by
us
i
ng
m
achine
le
arn
in
g
al
gori
thm
.
Ba
sed
on
the
co
nducted
l
it
eratur
e
stu
dy,
it
is
rev
eal
ed
that
there
a
re
two
fun
dam
ental
data
reg
a
rd
i
ng
t
he
a
rm
s d
yna
m
ic
s,
nam
ely total
e
xport
s and tota
l im
po
rt
s of arm
s.
The
total
ex
po
rts and im
p
or
ts
of
arm
s
trade
al
ong
with
t
he
def
e
ns
e
budge
t
data
will
be
treat
ed
as
th
e
outp
ut
va
ria
bles
in
t
his
st
ud
y.
Fu
rt
her
m
or
e,
t
his
stu
dy
ha
s
i
den
ti
fie
d
5
gr
oups
of
in
put
va
riables
that
great
ly
influ
ence
the
arm
s
dynam
ic
s
nam
ely
po
li
ti
cal
po
st
ur
e
,
ec
onom
ic
po
st
ur
e,
secu
rity
po
st
ure,
natio
nal
ca
pab
il
it
ie
s,
an
d
direct
c
on
ti
guit
y.
Data
colle
ct
ion
to
ok
place
by
ob
serv
i
ng
data
f
ro
m
1960
t
o
2018
sig
nifica
nt
va
riables
f
or
al
l
co
untr
es
in
the
Southeast
Asia
reg
i
on
a
dded
sever
al
c
ountr
i
es
that
ha
ve
i
m
po
rtant
r
ole
in
ASEA
N,
na
m
el
y
Russia,
Japan,
China,
and
Sou
th Ko
rea.
2.2.
Data
pre
processin
g st
age
The
nex
t
sta
ge
in
this
st
ud
y
i
s
data
pr
e
-
pr
oc
essing.
As
pr
e
viously
m
entioned,
the
seco
nd
c
halle
ng
e
in
this
resea
rc
h
is
how
t
o
pr
epar
e
t
he
data
that
has
been
c
ollec
te
d
duri
ng
the
data
colle
ct
ion
sta
ge
t
o
be
able
be
processe
d
with
m
achine
l
earn
i
ng
al
gorit
hm
s.
The
us
e
of
decisi
on
tre
e
al
gorithm
raises
pr
ob
le
m
s
in
te
rm
s
of
data
su
it
abi
li
ty
since
the
colle
ct
ed
datas
et
is
nu
m
erica
l
data.
In
ot
he
r
ha
nd,
the
da
ta
req
ui
red
by
the
decisi
on
tree
al
gorithm
to
cond
uct
data
cl
assifi
cat
ion
is
cat
egorical
dat
a.
To
ta
c
kle
this
issue
,
this
stud
y
pro
po
ses
a
no
vel
al
gorithm
fo
r
data
pre
pro
cessi
ng
so
t
hat
nu
m
erical
data
can
be
analy
zed
usi
ng
a
de
ci
sion
tree.
The
m
ain
idea
of
this
novel
al
go
rit
hm
is
to
con
ve
rt
nu
m
erical
data
into
cat
egorical
data
ba
sed
on
aver
a
ging the dat
a
[20]
. T
he pr
opos
e
d
al
gor
it
h
m
w
il
l cal
cu
la
te
the av
era
ge
d
at
a for
eac
h
var
ia
ble
then
c
onve
rt
each
data
for
that
va
riable
in
to
seve
ral
cat
egories.
T
he
fl
owcha
rt
of
pr
opose
d
pr
e
proc
essing
f
or
in
put
an
d
ou
t
pu
t
va
riable
s can
b
e
seen
in
Fi
gure
2
a
nd
3
.
Figure
2. Pr
e
pr
ocessin
g f
or
ou
tpu
t
var
ia
ble
2.3.
Ma
c
hine
le
arnin
g
de
ployme
nt st
age
The
final
sta
ge
of
this
stud
y
is
analy
zi
ng
dat
a
us
ing
m
achine
le
arn
in
g.
Ba
sed
on
the
cond
ucted
stu
dy
li
te
ratur
e,
the
m
os
t
app
ropr
i
at
e
al
go
rithm
t
o
achie
ve
the
go
al
f
or
this
stud
y
is
the
de
c
isi
on
tree
al
go
rithm
.
Decisi
on
tree a
lgorit
hm
is
reall
y
us
ef
ul
al
gori
thm
to
find
the
hidde
n
relat
ion
sh
i
ps
b
et
wee
n
in
put
va
riabl
es
an
d
an
outp
ut
va
riable.
I
n
ad
diti
on,
this
al
gori
thm
can
gen
er
at
e
a
m
od
el
of
decisi
on
r
ul
es
that
are
easy
to
unde
rstan
d.
Th
e
m
ai
n
con
cept
o
f
a d
eci
sion
t
ree is to
con
ve
rt d
at
a into a
decisi
on
tree and d
eci
sio
n
ru
le
s.
This
decisi
on
tree
c
on
sist
s
of
se
ve
ral
ty
pes
of
no
des,
nam
el
y
ro
ot
no
des,
i
nter
nal
no
des
a
nd
le
af
node
s.
T
hi
s
node
is a re
pr
ese
ntat
ion
of the
d
at
a
var
ia
bles in
t
he
d
at
aset
.
Evaluation Warning : The document was created with Spire.PDF for Python.
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci
IS
S
N:
25
02
-
4752
Mac
hin
e le
ar
ni
ng d
e
plo
y
me
nt
for
ar
ms
d
y
na
mics
pa
tt
ern
…
(
Zu
l I
ndr
a
)
1657
Figure
3. Pr
e
pr
ocessin
g f
or
i
nput
va
riable
3.
RESU
LT
S
AND
DI
SCUS
S
ION
3.1.
Data
c
ollec
tion a
nd pre
processin
g
The
s
ucces
s
of
the
data
c
ollec
ti
on
sta
ge
is
de
te
rm
ined
by
th
e
qual
it
y
of
the
data
c
ollec
te
d
to
s
uppor
t
the
res
ults
of
data
a
naly
sis
at
the
m
achine
le
arn
i
ng
depl
oy
m
ent
sta
ge.
The
res
ults
of
pr
e
processi
ng
a
re
determ
ined
fro
m
the
su
ccess
of
the p
r
opos
e
d
prep
ro
ce
ssin
g
al
gorithm
to
conve
rt
nu
m
erical
data
in
the
dataset
into
cat
eg
or
ic
a
l
data.
Th
rou
ghout
the
data
colle
ct
ion
sta
ge
s,
t
his
stu
dy
suc
ceeded
i
n
ide
ntifyi
ng
t
he
va
riable
s
need
e
d
t
o
a
naly
ze
arm
s
trade
patte
r
ns
.
The
r
e
are
30
in
pu
t
var
ia
bles
that
hav
e
bee
n
i
dent
ifie
d
an
d
retri
eved
a
s
the
infl
uen
ti
al
var
ia
bles
that
aff
ect
the
patte
rn
s
of
a
rm
s
dynam
ic
s.
The
de
ta
il
s
of
these
var
ia
b
le
s
can
be
seen
in Ta
ble 1.
Table
1.
Id
e
nti
fied
var
ia
bles
of d
at
aset
Ty
p
e
Categ
o
ry
Data Sou
rce
Ind
icato
r
Inp
u
t
Variable
Po
litical Pos
tu
re
[
2
1
]
,
[
2
2
]
Po
lity
I
V
Project
Po
litical S
y
ste
m
,
Reg
i
m
e Dur
ab
ility
,
State
Fr
ag
ilit
y
Ind
ex
,
Ef
f
ectiv
en
ess
Sco
re,
Legiti
m
a
cy
Sco
re,
S
ecurit
y
E
f
f
ectiv
en
ess
,
Secu
rit
y
L
eg
iti
m
ac
y
,
Po
litical E
f
f
ectiv
en
ess
,
Po
litical L
eg
iti
m
ac
y
,
Econ
o
m
ic
Ef
f
ectiv
en
ess
,
Econ
o
m
i
c L
eg
iti
m
acy
,
Social
Ef
f
ectiv
en
ess
and
Social Legiti
m
ac
y
Econ
o
m
i
c Pos
tu
re
[
2
3
],
[
2
4
]
W
o
rldb
an
k
GDP (
Gros
s Do
m
e
stic Prod
u
ct),
GDP
per Capi
ta,
GDP
Growth
and
Inf
latio
n
Secu
rity Prioriti
zat
io
n
an
d
Defen
se Po
stur
e
[
2
3
]
,
[
2
5
],
[
2
6
]
SIPRI
Militar
y
Exp
en
d
itu
re
(%
GDP)
,
Milita
ry
Exp
en
d
itu
re
(C
o
n
stan
t$
m
)
an
d
Militar
y
P
erso
n
n
el
Natio
n
al Cap
ab
ilities
[
2
7
]
Co
rr
elates o
f
W
ar
Pr
o
ject
Ir
o
n
and
Steel
P
rod
u
ctio
n
(
1
0
0
0
to
n
s)
,
Pri
m
a
ry Energ
y
Co
n
su
m
p
tio
n
(10
0
0
co
alto
n
s),
To
tal Pop
u
latio
n
(
1
0
0
0
),
Ur
b
an
Pop
u
lati
o
n
(
1
0
0
0
),
Co
m
p
o
site
Ind
ex
of
Nation
al Cap
ab
ility,
an
d
High
Tech
Exp
o
rt
(US$)
Direct
Co
n
tig
u
ity
[
2
8
]
Co
rr
elates o
f
W
ar
Pr
o
ject
Total Nu
m
b
e
r
o
f
Dir
ect Co
n
tig
u
ities, T
o
tal
Nu
m
b
er
of
Dir
ect
Co
n
tig
u
ities b
y
L
an
d
and
T
o
tal Nu
m
b
er
o
f
Dir
ect Co
n
ti
g
u
ities b
y
Sea
Ou
p
u
t
Variable
Def
en
se Bu
d
g
et
SIPRI
Def
en
se Bu
d
g
et
Total o
f
Ar
m
s
E
x
p
o
rt
SIPRI
Air
d
ef
en
se
syste
m
,
Nu
m
b
e
r
o
f
Ar
m
s
Exp
o
rt
f
o
r
Aircr
af
t
,
Ar
m
o
red
v
eh
icles, Ar
till
er
y
,
M
iss
iles,
Nav
al w
eapo
n
s, Sens
o
rs,
S
h
ip
s an
d
Others
Total o
f
Ar
m
s
I
m
p
o
rt
SIPRI
Air
d
ef
en
se syste
m
,
Nu
m
b
e
r
o
f
Ar
m
s
I
m
p
o
rt
f
o
r
Ai
rcr
af
t
,
Ar
m
o
red
v
eh
icles, Naval w
e
ap
o
n
s
,
Artiller
y
,
E
n
g
in
es, M
iss
iles, S
en
so
rs,
Sh
ip
s
an
d
Others
3.2.
Ma
c
hine
le
arnin
g
de
ployme
nt
As
e
xp
la
ine
d
in
the
m
et
ho
d
sect
ion
,
t
he
a
lgorit
hm
cho
s
en
f
or
data
a
na
ly
sis
is
the
decisi
on
tree
al
gorithm
.
This
al
gorithm
pr
oduces
a
n
arm
s
tradin
g
patte
r
n
in
the
f
or
m
of
decisi
on
r
ul
es
by
disc
ov
e
ri
ng
t
he
root
var
ia
bles
that
gr
eat
ly
inf
luence
the
patte
rn
of
the
decisi
on
r
ule
f
or
e
ach
outp
ut
vari
able.
H
ow
e
ve
r,
thi
s
stud
y
has
t
hr
ee
outp
ut
va
riabl
es,
na
m
ely
defense
budget
,
t
ot
al
arm
s
exp
ort
an
d
total
a
rm
s
i
m
po
rts.
Ba
se
d
on
the
res
ults
of
data
co
nversi
on
at
the
pr
e
pro
cessi
ng
sta
ge,
the
val
ues
of
t
hese
th
ree
data
have
bee
n
set
to
be
two
outp
ut
val
ues,
nam
el
y
hig
h
a
nd
lo
w.
T
he
fi
rst
ste
p
t
o
c
onduct
dat
a
analy
sis
by
us
in
g
a
decisi
on
tree
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2502
-
4752
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci,
Vo
l.
23
, N
o.
3
,
Se
ptem
ber
20
21
:
16
54
-
16
62
1658
al
gorithm
is
by
cal
culat
ing
the
total
entrop
y
fo
r
al
l
data
in
the
dataset
.
On
ce
the
tota
l
entropy
has
been
ob
ta
ine
d,
the
ne
xt
proce
ss
is
t
o
determ
ine
th
e
no
de
of
the
decisi
on
tree
.
The
e
xpla
natio
n
of
the
decisi
on
tree
al
gorithm
an
al
ysi
s r
esults
for e
a
ch ou
t
pu
t
va
riable
.
3.2.1.
De
fense
b
ud
get
The
decisi
on
tr
ee
for
patte
r
n
of
the
de
fen
se
budget
var
ia
bl
e
is
pr
ese
nted
i
n
Fig
ur
e
4.
T
he
patte
rn
of
a
r
m
s
dy
n
a
m
i
c
s
f
o
r
d
e
f
e
n
s
e
b
u
d
g
e
t
o
u
t
p
u
t
v
a
r
i
a
b
l
e
s
c
a
n
b
e
de
t
e
r
m
i
ne
d
b
a
s
e
d
o
n
t
h
e
d
a
t
a
p
r
e
s
e
n
t
e
d
i
n
F
i
gu
r
e
4
.
The deci
sio
n r
ules fo
r
Fi
gure
4
a
re
descr
i
bed
as s
how
n
in
.
C
o
n
t
s
e
a
E
n
t
r
o
p
y
=
0
.
1
4
0
1
7
8
G
D
P
P
e
r
c
a
p
i
t
a
E
n
t
r
o
p
y
=
0
.
1
9
9
7
0
S
f
i
E
n
t
r
o
p
y
=
0
.
2
2
2
7
7
D
u
r
a
b
i
l
i
t
y
E
n
t
r
o
p
y
=
0
.
1
0
6
9
4
L
o
w
P
o
s
i
t
i
v
e
H
i
g
h
P
o
s
i
t
i
v
e
M
i
d
l
e
L
o
w
P
o
s
i
t
i
v
e
H
i
g
h
P
o
s
i
t
i
v
e
M
i
d
l
e
L
o
w
(
+
0
,
-
3
9
)
L
o
w
(
+
9
,
-
4
2
)
H
i
g
h
(
+
1
0
,
-
7
)
L
o
w
P
o
s
i
t
i
v
e
H
i
g
h
P
o
s
i
t
i
v
e
M
i
d
l
e
L
o
w
(
+
1
8
,
-
7
3
)
H
i
g
h
(
+
4
6
,
-
1
)
H
i
g
h
(
+
2
1
2
,
-
6
4
)
L
o
w
P
o
s
i
t
i
v
e
H
i
g
h
P
o
s
i
t
i
v
e
M
i
d
l
e
H
i
g
h
(
+
1
4
,
-
8
)
H
i
g
h
(
+
1
2
9
,
-
7
5
)
L
o
w
(
+
5
,
-
1
2
9
)
Figure
4. Decis
ion
t
ree fo
r def
ense
budget
goal
IF
(
Co
ntsea
= No
ne AND
GD
P
Per
ca
p
ita
= H
ig
h P
o
s
itive)
OR (Co
ntsea
=
L
o
w Pos
itive AND
Sfi
=
Low
P
o
sitive
)
OR
(Co
ntsea
= L
o
w Pos
itive AND
Sfi
=
H
ig
h
Pos
itive)
OR (Co
ntsea
=
H
ig
h P
o
sitive AND
Dura
b
ility = M
id
d
le)
OR
(Co
ntsea
=
H
ig
h
Pos
itive AN
D Dura
b
ility =
Low
P
o
sitive)
TH
EN
De
fense B
ud
g
et = H
ig
h
ELSE De
fense Bu
d
g
et = L
o
w
3.2.2.
T
otal
of
arms
e
xpor
t
The
decisi
on
r
ules
for
Fig
ur
e
5.
T
he
decisi
on
tree
f
or
patte
rn
of
the
arm
s
expo
rt
var
ia
ble
is
pr
esente
d
in Figu
re
6
.
T
P
o
p
1
0
0
0
E
n
t
r
o
p
y
=
0
.
2
5
0
9
2
C
o
n
t
L
a
n
d
E
n
t
r
o
p
y
=
0
.
4
2
1
2
8
M
i
d
d
l
e
H
i
g
h
L
o
w
L
o
w
L
o
w
P
o
s
i
t
i
v
e
H
i
g
h
P
o
s
i
t
i
v
e
N
o
n
e
L
o
w
(
+
0
,
-
4
0
)
H
i
g
h
(
+
5
6
,
-
0
)
H
i
g
h
(
+
7
1
,
-
2
5
)
L
o
w
Figure
5. Decis
ion
t
ree fo
r
ar
m
s ex
po
rt
goal
Evaluation Warning : The document was created with Spire.PDF for Python.
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci
IS
S
N:
25
02
-
4752
Mac
hin
e le
ar
ni
ng d
e
plo
y
me
nt
for
ar
ms
d
y
na
mics
pa
tt
ern
…
(
Zu
l I
ndr
a
)
1659
IF
(
TP
o
p
1
0
0
0
= H
ig
h P
o
sitive
AN
D Co
ntLand
= L
o
w Pos
itive)
OR (
TP
o
p
1
0
0
0
= H
ig
h P
o
sitive
AN
D Co
ntLand
=
H
i
g
h P
o
sitive)
TH
EN
T
o
ta
l of Ar
m
exp
o
rt = H
ig
h
ELSE Tota
l of Ar
m
exp
o
rt = Low
3.2.3.
T
otal
of
arms
im
po
r
t
The deci
sio
n
tr
ee f
or
patte
r
n o
f
the
arm
s ex
port
var
ia
ble is
pr
ese
nted
in Fi
gure
6.
p
e
c
1
0
0
0
c
o
a
l
t
o
n
E
n
t
r
o
p
y
=
0
.
1
0
5
2
2
D
u
r
a
b
i
l
i
t
y
E
n
t
r
o
p
y
=
0
.
1
5
1
6
6
C
o
n
t
l
a
n
d
E
n
t
r
o
p
y
=
0
.
0
9
4
9
8
M
i
l
e
x
p
G
D
P
E
n
t
r
o
p
y
=
0
.
3
2
0
7
L
o
w
P
o
s
i
t
i
v
e
H
i
g
h
P
o
s
i
t
v
e
M
i
d
d
l
e
L
o
w
P
o
s
i
t
i
v
e
H
i
g
h
P
o
s
i
t
v
e
M
i
d
d
l
e
L
o
w
(
+
0
,
-
9
9
)
L
o
w
(
+
1
9
,
-
1
8
3
)
H
i
g
h
(
+
2
6
,
-
2
4
)
L
o
w
P
o
s
i
t
i
v
e
H
i
g
h
P
o
s
i
t
v
e
N
o
n
e
n
L
o
w
(
+
1
,
-
3
6
)
H
i
g
h
(
+
7
8
,
-
6
4
)
H
i
g
h
(
+
9
4
,
-
2
4
)
L
o
w
P
o
s
i
t
i
v
e
H
i
g
h
P
o
s
i
t
v
e
M
i
d
d
l
e
L
o
w
(
+
0
,
-
3
1
)
H
i
g
h
(
+
1
1
7
,
-
3
7
)
H
i
g
h
(
+
2
3
,
-
5
)
Figure
6
.
Decis
ion
t
ree fo
r
ar
m
s i
m
po
rt g
oal
The deci
sio
n r
ules fo
r
Fi
gure
6
a
re
descr
i
bed as
s
how
n
in
.
IF
(
Pec10
0
0
Co
a
lt
o
n = M
id
d
le AND
Dura
b
ility = H
ig
h
Pos
itive)
OR
(P
ec10
0
0
Co
a
l
to
n
=
Low
Pos
itiv
e
AND
Co
ntla
nd
=
Low
Pos
itive)
OR
(P
ec10
0
0
Co
a
lto
n
=
Low
Pos
itive
AND
Co
ntla
nd
=
H
ig
h
P
o
sitive)
OR
(
Pec
1
0
0
0
C
o
a
lto
n
AN
D
Milexp
GDP
=
Low
Pos
itive)
OR
(P
ec10
0
0
Co
a
lto
n
=
Low
Pos
itive
AND
Milexp
GD
P
=
H
ig
h
Pos
itive)
TH
EN
Arm
im
p
o
r
t = H
ig
h
ELSE A
rm
im
p
o
rt
= L
o
w
4.
RESU
LT
S
AND
DI
SCUS
S
ION
Ba
sed
on
th
e
hy
po
the
sis
in
th
e
introd
uction
sect
ion
,
t
her
e
are
thre
e
ob
j
ec
ti
ves
in
the
st
udy
nam
el
y:
(1)
de
fen
s
e
bu
dg
et
,
(2)
arm
s
trade
ex
port,
a
nd
(3)
arm
s
trade
im
po
rt.
Each
goal
has
different
decisi
on
tree
of
m
achine lear
nin
g o
utcom
e. T
he follo
wing is
a d
et
ai
le
d disc
us
sio
n of eac
h objecti
ve
.
4.1.
Def
e
nse
b
ud
get
Ba
sed
on
the
r
esult,
def
e
ns
e
budget
is
m
os
t
sign
ific
antly
de
te
rm
ined
by
sta
te
’s
direct
sea
con
ti
guit
y
(contsea)
.
It
s
ugge
sts
that
the
m
or
e
sta
te
ha
s
direct
sea
c
onti
gu
it
y;
the
st
at
e
will
sp
en
d
m
or
e
fo
r
it
s
de
fen
se
budget
.
Anothe
r
stu
dy
al
so
su
ggest
s
sim
ila
r
pre
po
sit
io
n
that
co
ntig
uity
has
posit
ive
eff
ect
or
pos
it
ively
associat
ed
with
m
i
li
ta
ry
sp
end
i
ng
[
29
]
.
Re
gardin
g
the
dir
ect
sea
con
ti
gu
it
y,
it
has
thre
e
der
i
vative
va
riables
wh
ic
h
are:
(
1)
GDP
pe
r
Ca
pita
wh
ic
h
occ
ur
for
gro
up
of
sta
te
with
no
dir
ect
sea
con
ti
gu
it
y,
(2
)
sta
te
fragil
it
y
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2502
-
4752
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci,
Vo
l.
23
, N
o.
3
,
Se
ptem
ber
20
21
:
16
54
-
16
62
1660
ind
e
x
wh
ic
h
oc
cur
f
or
gro
up
of
sta
te
with
low
num
ber
of
di
rect
sea
c
ont
igu
it
y,
a
nd
(
3)
sta
te
’s
dura
bili
ty
wh
ic
h
occ
ur
f
or
gr
oup
of
sta
te
with
hig
h
nu
m
ber
of
dir
ect
sea.
Acco
r
ding
to
the
res
ult,
stud
y
on
de
fens
e
budget
relat
ed
to
sea
co
ntig
uity
is
i
m
po
rtant
for
the
st
udy
on
m
arit
i
m
e
power
a
nd/o
r
m
arit
i
m
e
secur
it
y.
A
sta
te
who
per
c
ei
ves
m
arit
i
m
e
as
strat
egic
dri
ver
s
a
nd
f
or
ce
s
will
seek
m
or
e
po
wer
f
or
it
s
m
ariti
m
e
def
ens
e
po
st
ur
e
.
A
st
udy
by
U
nited
Stat
es
d
e
par
tm
ent
of
The
Na
vy
al
so
m
ade
cl
assifi
cat
ion
about
sta
te
s
by
it
s
m
arit
i
m
e p
owe
r
[
30]
.
To
c
oncl
ude,
a
de
fense
budget
ca
n
be
e
xpla
ined
by
ide
nti
fyi
ng
sta
te
’
s
sp
at
ia
l
c
ondi
ti
on
-
in
this
c
on
te
xt
is
dire
ct
sea
co
ntig
ui
ty
-
whi
c
h
is
po
sit
ively
associat
ed
wit
h
GDP
pe
r
Ca
pita,
St
at
e’s
F
rag
il
it
y
ind
e
x
,
and
sta
te
’s
dur
abili
ty
.
GD
P
pe
r
Ca
pita
will
sign
ific
a
ntly
determ
ine
def
ense
budg
et
am
on
g
sta
te
with
no
direct
sea
co
ntig
uity
or
a
la
ndlo
cke
d
c
ountry.
Stat
e’s
fr
a
gili
ty
will
sign
ific
a
ntly
determ
ine
de
f
ense
budget
a
m
on
g
sta
te
with
few
direct
sea
cont
igu
it
ie
s.
Stat
e’s
dura
bili
ty
will
sign
ific
antly
determ
ine
def
ense
budget
a
m
ong
sta
te
w
it
h
m
any d
irect
sea
co
ntigu
it
ie
s.
4.2.
Arms
e
xpor
t
Ba
sed
on
the
resu
lt
of
m
achine
le
arn
in
g
de
ci
sion
tree
m
od
el
outc
om
e,
siz
e
of
the
sta
te
wh
ic
h
is
translat
ed
to
it
s
total
popu
la
t
ion
has
a
signi
ficant
relat
ion
with
it
s
arm
s
trade
ex
port.
big
siz
e
po
pu
l
at
ion
country te
nds to
ex
port m
or
e in the
arm
s tra
de.
T
h
is analy
sis resu
lt
w
as c
orres
ponded w
it
h
cl
assic
al
th
eor
y o
n
econom
ic
and
m
anu
fact
ur
e
d
pro
du
ct
io
n
t
ha
t
there
is
a
ge
ne
ral
posit
ive
re
la
ti
on
sh
i
p
between
po
pula
ti
on
siz
e
and d
e
ns
it
y wit
h
e
xport
p
e
rform
ance of th
e
c
ountry
[31],
[32]
.
Re
gardin
g
th
e
decisi
on
tr
ee
m
od
el
of
a
rm
s
ex
port,
t
otal
of
po
pu
la
ti
on
var
ia
ble
has
one
der
i
vate
var
ia
ble
that
is
direct
la
nd
co
ntigu
it
y
.
It
sug
gests
that
if
the
popu
la
ti
on
of
the
sta
te
is
b
ig
and
it
has
m
uch
direct
co
ntig
uity
by
la
nd
,
the
sta
te
te
nd
to
get
higher
am
ount
of
arm
s
t
rad
e
e
xport
an
d
vice
ve
rsa.
Existi
ng
li
te
ratur
es
al
so
hav
e
ex
plaine
d
ho
w
la
nd
c
onti
gu
it
y
was
a
sso
ci
at
ed
with
no
t
on
ly
inter
na
ti
on
al
co
nf
li
ct
s
but
al
so
m
i
li
ta
ry
e
xp
e
ndit
ur
es
[
33]
,
[
34
]
.
Our
a
naly
sis
resu
lt
s
uggests
new
i
nsi
gh
t
that
sta
te
’s
num
ber
of
direct
la
nd
c
onti
gu
it
y
was
po
sit
ively
associat
ed
with
the
am
ount
of
arm
s
trade
ex
port.
T
o
c
on
cl
ud
e
,
arm
s
expo
rt
ca
n
be
ex
plained
by
identify
ing
s
ta
te
’s
total
of
popula
ti
on
w
hi
ch
is
po
sit
ivel
y
associat
ed
w
it
h
sta
te
’s
nu
m
ber
of
direct la
nd c
on
ti
gu
it
y.
4.3.
Arms
i
m
po
r
t
In
te
rm
of
arm
s
trade
im
po
rt
,
our
m
achine
le
arn
i
ng
decisi
on
tree
m
od
el
ou
tc
om
e
su
gg
e
sts
pr
im
ary
energy
c
on
s
um
pt
ion
si
gn
i
ficantl
y
determ
i
ne
the
am
ou
nt
of
sta
te
’s
arm
s
trade
im
po
rt.
As
the
r
oo
t
va
riable,
pr
im
ary
ener
gy
con
s
um
ption
has
th
ree
der
i
vative
var
ia
ble
s
wh
ic
h
are:
(
1)
sta
te
durab
i
li
ty
wh
ic
h
occ
ur
for
gro
up
of
sta
te
with
lo
w
pr
im
ary
ene
rg
y
c
on
su
m
ption
,
(2)
direct
la
nd
c
onti
gu
it
y
occur
f
or
gro
up
of
sta
te
with
m
idd
le
pri
m
ar
y
energy
co
nsu
m
pt
ion
,
a
nd
(
3)
m
i
li
ta
ry
exp
e
nd
it
ure
by
pe
rc
ent
of
GDP
oc
cur
f
or
gro
up
of
sta
te
with
high
pri
m
ary
ener
gy
con
s
um
ption
.
A
stud
y
by
V
incenz
o
Bo
ve
on
the
relat
io
ns
hi
p
betwee
n
ener
gy
consum
ption
a
nd
arm
s
trade
sta
te
d
that
e
nergy
dep
e
nde
nce
le
ads
to
a
hig
he
r
volum
e
of
arm
s
transf
e
r
s
betwee
n
c
ount
ries
[
35]
.
T
he
s
tud
y
fou
nd
tha
t
a
co
untry
with
a
rece
nt
discov
e
ry
of
ne
w
oi
l
fiel
ds
will
increas
e
it
s
i
m
po
rt
of
weapo
ns
f
r
om
oil
-
de
pe
nd
e
nt
econom
ie
s
by
56%
[
35
]
.
I
n
de
eper
disc
us
si
on,
oil
m
igh
t
play
an
even
la
r
ge
r
r
ol
e
in
influ
e
ncing
ec
onom
ic
and
poli
ti
cal
decisi
on
s
[
35]
.
To
co
nclu
de,
arm
s
i
m
po
rt
c
an
be
exp
la
ine
d
by
identify
in
g
sta
t
e’s
co
nsum
pti
on
on
pri
m
ary
energy
w
hich
is
determ
ined
by
Stat
e’s
D
urabil
it
y,
nu
m
ber
of
di
re
ct
land
c
onti
gu
it
y
,
and
am
ount
o
f
m
i
li
ta
ry exp
en
ditu
re
by percent
of GDP
5.
CONCL
US
I
O
N
T
h
i
s
r
e
s
e
a
r
c
h
p
r
e
s
e
n
t
s
r
o
b
u
s
t
e
v
i
d
e
n
c
e
t
h
a
t
‘
d
e
c
i
s
i
o
n
t
r
e
e
’
o
f
m
a
c
hi
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
hm
c
a
n
c
l
a
s
s
i
f
y
a
n
d
p
r
e
d
i
c
t
d
e
f
e
n
s
e
b
u
d
g
e
t
,
a
r
m
s
t
r
a
d
e
e
x
p
o
r
t
,
a
n
d
a
r
m
s
t
r
a
de
i
m
p
o
r
t
i
n
S
o
u
t
h
e
a
s
t
A
s
i
a
c
o
u
n
t
r
i
e
s
.
E
a
c
h
v
a
r
i
a
b
l
e
h
a
s
i
t
s
o
w
n
r
o
o
t
v
a
r
i
a
b
l
e
s
o
r
i
t
s
m
o
s
t
s
i
g
n
i
f
i
c
a
n
t
f
a
c
t
o
r
/
v
a
r
i
a
b
l
e
t
o
d
e
t
e
r
m
i
ne
t
h
o
s
e
d
e
p
e
n
de
n
t
v
a
r
i
a
b
l
e
s
.
D
e
f
e
n
s
e
b
u
d
g
e
t
w
a
s
d
e
t
e
r
m
i
n
e
d
b
y
s
t
a
t
e
’
s
d
i
r
e
c
t
s
e
a
c
o
n
t
i
g
u
i
t
y
.
A
r
m
s
e
x
p
o
r
t
w
a
s
d
e
t
e
r
m
i
n
e
d
b
y
s
t
a
t
e
’
s
t
ot
a
l
p
o
p
u
l
a
t
i
o
n
.
A
r
m
s
i
m
p
o
r
t
w
a
s
d
e
t
e
r
m
i
n
e
d
b
y
p
r
i
m
a
r
y
e
n
e
r
g
y
c
o
n
s
um
p
t
i
o
n
.
T
o
s
u
m
u
p
,
a
r
m
s
d
y
n
a
m
i
c
i
n
t
h
e
S
o
u
t
h
e
a
s
t
A
s
i
a
c
o
u
n
t
r
i
e
s
-
a
n
d
s
om
e
s
t
r
a
t
e
gi
c
c
o
u
n
t
r
i
e
s
p
a
r
t
n
e
r
s
a
s
w
e
l
l
-
w
a
s
h
i
g
h
l
y
a
s
s
o
c
i
a
t
e
d
w
i
t
h
i
t
s
d
i
r
e
c
t
s
e
a
c
o
n
t
i
g
u
i
t
y
,
t
o
t
a
l
p
o
p
u
l
a
t
i
o
n
,
a
n
d
p
r
i
m
a
r
y
e
n
e
r
g
y
c
o
n
s
um
pt
i
o
n
.
I
n
s
p
i
t
e
o
f
t
h
a
t
,
t
hi
s
r
e
s
e
a
r
c
h
p
r
e
s
e
n
t
s
t
h
r
e
e
t
h
e
o
r
e
t
i
c
a
l
i
m
pl
i
c
a
t
i
o
n
s
.
F
i
r
s
t
,
a
s
t
a
t
e
w
h
o
p
e
r
c
e
i
v
e
s
m
a
r
i
t
i
m
e
a
s
s
t
r
a
t
e
g
i
c
d
r
i
v
e
r
s
a
n
d
f
o
r
c
e
s
w
i
l
l
s
e
e
k
m
o
r
e
p
o
w
e
r
f
o
r
i
t
s
m
a
r
i
t
i
m
e
d
e
f
e
ns
e
p
o
s
t
u
r
e
w
h
i
c
h
i
s
t
r
a
n
s
l
a
t
e
d
t
o
d
e
f
e
n
s
e
b
u
d
g
e
t
.
T
h
a
t
i
s
w
h
y
s
e
a
c
o
n
t
i
gu
i
t
y
a
c
c
o
r
d
i
n
g
t
o
o
u
r
a
n
a
l
y
s
i
s
d
e
t
e
r
m
i
n
e
d
e
f
e
n
s
e
b
u
d
g
e
t
s
i
g
n
i
f
i
c
a
n
t
l
y
.
S
e
c
o
n
d
,
b
i
g
s
i
z
e
c
o
u
n
t
r
i
e
s
t
e
n
d
t
o
b
e
a
n
a
r
m
s
e
x
p
o
r
t
e
r
c
o
u
n
t
r
y
.
Y
e
t
,
i
t
i
s
n
o
t
j
u
s
t
a
b
o
u
t
t
h
e
s
i
z
e
o
f
t
h
e
c
o
u
n
t
r
y
,
b
e
c
a
u
s
e
a
c
c
o
r
d
i
n
g
t
o
t
h
e
r
e
s
e
a
r
c
h
a
n
a
l
y
s
i
s
,
i
t
m
u
s
t
b
e
a
c
o
u
n
t
r
y
w
i
t
h
m
a
n
y
d
i
r
e
c
t
l
a
n
d
c
o
n
t
i
g
u
i
t
i
e
s
.
T
hi
r
d
,
r
e
l
a
t
i
o
n
s
h
i
p
b
e
t
w
e
e
n
e
n
e
r
g
y
c
o
n
s
um
p
t
i
o
n
a
n
d
a
r
m
s
t
r
a
d
e
s
t
a
t
e
d
t
h
a
t
e
n
e
r
g
y
d
e
p
e
n
d
e
n
c
e
l
e
a
d
s
t
o
a
h
i
g
h
e
r
v
o
l
um
e
o
f
a
r
m
s
t
r
a
n
s
f
e
r
s
b
e
t
w
e
e
n
c
o
u
n
t
r
i
e
s
.
A
m
o
ng
t
h
e
s
t
a
t
e
s
w
i
t
h
h
i
g
h
p
r
i
m
a
r
y
e
n
e
r
g
y
c
o
n
s
u
m
pt
i
o
n
,
a
h
i
g
h
l
e
v
e
l
o
f
m
i
l
i
t
a
r
y
e
x
p
e
n
d
i
t
u
r
e
b
y
p
e
r
c
e
n
t
o
f
G
D
P
w
i
l
l
l
e
a
d
t
o
a
h
i
g
h
l
e
v
e
l
o
f
a
r
m
s
t
r
a
d
e
i
m
po
r
t
a
n
d
v
i
c
e
v
e
r
s
a
.
I
n
t
e
r
e
s
t
i
n
g
l
y
,
t
h
e
c
om
bi
n
a
t
i
o
n
o
f
t
h
e
s
e
t
h
r
e
e
f
i
n
d
i
n
g
s
c
a
n b
e
t
o
o
l
s
/
s
u
p
p
o
r
t
s
f
o
r
Evaluation Warning : The document was created with Spire.PDF for Python.
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci
IS
S
N:
25
02
-
4752
Mac
hin
e le
ar
ni
ng d
e
plo
y
me
nt
for
ar
ms
d
y
na
mics
pa
tt
ern
…
(
Zu
l I
ndr
a
)
1661
a
c
a
d
e
m
i
c
s
a
n
d
d
e
c
i
s
i
o
n
/
p
o
l
i
c
y
m
a
k
e
r
s
t
o
p
e
r
c
e
i
v
e
a
n
d
c
a
l
c
u
l
a
t
e
t
h
e
p
o
w
e
r
p
r
o
j
e
c
t
i
o
n
s
a
n
d
r
e
g
i
o
n
a
l
p
o
l
i
t
i
c
a
l
l
a
n
d
s
c
a
p
e
i
n
o
r
d
e
r
t
o
f
o
r
m
u
l
a
t
e
c
e
r
t
a
i
n
s
t
r
a
t
e
g
i
c
p
o
l
i
c
i
e
s
a
nd
f
o
r
e
i
g
n
p
o
l
i
c
y
.
H
o
w
e
v
e
r
,
f
u
r
t
h
e
r
r
e
s
e
a
r
c
h
n
e
e
d
s
t
o
b
e
d
o
n
e
i
n
o
r
d
e
r
t
o
d
e
e
p
e
n
a
n
d
e
n
r
i
c
h
t
h
i
s
r
e
s
e
a
r
c
h
t
o
p
i
c
,
s
u
c
h
a
s
b
y
a
p
p
l
y
i
n
g
r
a
n
d
o
m
f
o
r
e
s
t
a
n
d
u
n
s
e
r
v
i
s
e
d
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
hm
s
.
T
h
e
r
a
n
d
o
m
f
o
r
e
s
t
a
l
g
o
r
i
t
hm
i
s
a
d
e
c
i
s
i
o
n
t
r
e
e
d
e
v
e
l
o
pm
e
n
t
a
l
g
o
r
i
t
hm
h
e
n
c
e
i
t
s
i
m
pl
e
m
e
nt
a
t
i
o
n
i
s
e
x
p
e
c
t
e
d
t
o
c
o
v
e
r
t
h
e
l
i
m
i
t
a
t
i
o
n
o
f
t
h
e
d
e
c
i
s
i
o
n
t
r
e
e
a
l
g
o
r
i
t
hm
u
s
e
d
i
n
t
h
i
s
s
t
u
d
y
,
w
h
i
c
h
i
s
t
h
a
t
i
t
c
a
n
o
n
l
y
c
l
a
s
s
i
f
y
t
w
o
c
l
a
s
s
e
s
.
I
n
t
e
r
m
s
o
f
i
m
p
l
e
m
e
n
t
i
n
g
u
n
s
u
p
e
r
v
i
s
e
d
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
hm
s
,
c
l
u
s
t
e
r
o
f
c
o
u
n
t
r
i
e
s
c
a
n
b
e
f
o
r
m
e
d
b
a
s
e
d
o
n
t
h
e
i
r
w
e
a
p
o
n
t
r
a
d
i
n
g
p
a
t
t
e
r
n
s
b
y
u
s
i
n
g
a
n
u
n
s
u
p
e
r
v
i
s
e
d
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
hm
.
T
h
e
s
e
c
o
u
n
t
r
y
c
l
u
s
t
e
r
s
c
a
n
b
e
a
n
i
m
p
o
r
t
a
n
t
i
n
f
o
r
m
a
t
i
o
n
t
o
c
o
n
d
u
c
t
a
n
a
l
y
s
i
s
f
o
r
t
h
e
p
o
l
a
r
i
z
a
t
i
o
n
o
f
t
h
e
s
e
c
o
u
n
t
r
i
e
s
i
n
t
h
e
i
r
p
o
l
i
t
i
c
a
l
a
n
d
d
e
f
e
n
s
e
p
o
l
i
c
i
e
s
.
REFERE
NCE
S
[1]
J.
McCar
th
y
and
E.
Feig
enba
um
,
“
In
Mem
oria
m
Arthur
Sam
uel
:
Pionee
r
in
Ma
ch
ine
L
ea
rning
,
”
A
I
Magazine
,
vol
.
11,
no
.
3
.
p
.
11
,
1990,
doi
:
10
.
16
09/a
imag.
v11
i3.840.
[2]
A.
L
.
Sam
uel,
"
Som
e
Studie
s
in
Mac
hin
e
Le
arn
i
ng
Us
ing
the
Ga
m
e
of
Che
cke
rs,
"
in
I
BM
Journa
l
of
Re
sear
ch
an
d
Dev
el
opmen
t
,
vo
l.
3
,
no
.
3
,
pp
.
21
0
-
229,
Jul
y
1959
,
doi
:
10
.
1147/rd
.
33.
0210
.
[3]
J
.
H.
Friedman,
“
Data
Mining
a
nd
Stat
isti
cs:
W
hat
’s
the
Conne
ct
ion
?
,
”
Comput
ing
scie
n
ce
and
stati
stic
s
,
vol.
2
9,
no.
1
,
pp
.
3
-
9
,
1
998.
[4]
R.
M.
Nea
l
,
"
Pa
t
te
rn
R
ec
ogni
ti
on
and
Ma
chi
n
e Learning
,
"
New Yo
rk:
Springer
,
vo
l
.
49
,
no
.
3
,
2007
.
[5]
D.
Mali
n
ia
k
,
S.
Pete
rson,
R
.
Po
wers,
and
M.
J.
Ti
ern
e
y
,
“
TRIP
2014
W
orld
Surve
y
Repor
t,”
W
i
ll
ia
m
sburg,
2014
.
[6]
P.
E.
Shea
and
P.
Poast,
“
W
ar
and
Defa
ult,”
Jo
urnal
of
Confl
ict
Re
soluti
on
,
vol
.
62,
no.
9,
pp.
1876
-
1904,
2018,
doi:
10
.
1177/00
22002717707239.
[7]
M.
Digiuse
ppe
a
nd
P.
Poast,
“
Arm
s
ver
sus
Dem
o
cra
t
ic
All
ie
s,
”
B
riti
sh
Journal
of
Pol
itical
S
cienc
e
,
vol
.
48,
no
.
4,
pp.
981
-
1003
,
2
018.
[8]
C.
Kaout
za
nis
,
P.
Poast,
and
J.
Urpel
ai
n
en,
“
No
t
letti
ng
‘ba
d
ap
ple
s’’
spoil
the
bunch:
Dem
ocr
a
ti
z
at
ion
and
str
i
ct
int
ern
at
ion
a
l
org
ani
z
at
ion
a
ccess
ion
rule
s,’”
The
Re
v
ie
w
of
Inte
rn
ati
onal
Or
ganizati
ons
,
vol.
11
,
no.
4,
pp.
399
-
41
8,
2016,
doi
:
10
.
10
07/s11558
-
015
-
9237
-
5.
[9]
P.
Poast,
M.
J.
Bom
m
ari
to,
and
D.
M.
Katz,
“
The
Elec
troni
c
W
orld
Tre
aty
In
dex:
Coll
ecting
the
Population
of
Inte
rna
ti
ona
l
Agree
m
ent
s
in
the
20th
Cent
ur
y
,
”
SSRN
El
ec
troni
c
J
ournal
,
pp.
1
-
29,
2015,
doi:
10.
2139/ssrn.26
52760.
[10]
D.
W
ie
ns,
P.
P
oast,
and
W
.
R.
Cla
rk
,
“
The
Polit
i
ca
l
Resourc
e
Curse:
An
E
m
piri
ca
l
Re
-
evalua
ti
on
,
”
Po
li
tic
al
Re
search
Quar
ter
ly
,
vo
l. 67
,
no.
4,
pp
.
783
-
794
,
2014.
[11]
P.
Poast,
“
Cent
ral
Banks
at
W
ar,
”
Inte
rnat
i
onal
Or
ganizati
on
,
vol.
69
,
no.
1,
pp.
63
-
95,
2014,
doi:
10.
1017/S0020818314000265.
[12]
P.
Poast
and
J
.
Urpel
ai
n
en,
“
Fit
and
Feasible:
W
h
y
Dem
o
cra
t
iz
ing
States
Form
,
not
Joi
n,
Int
ern
a
ti
ona
l
Organi
zations,”
Inte
rnational
Stu
die
s Quar
te
rly
,
v
ol.
57
,
no
.
4
,
pp
.
831
-
841,
2013
,
doi:
10
.
1111/i
sq
u.
12031.
[13]
W
.
R.
Cla
rk
,
S.
N.
Golder
,
and
P.
Poast
,
“
Moneta
r
y
insti
tut
ions
and
the
po
li
t
ic
a
l
survival
of
dem
ocr
atic
l
ea
d
ers,
”
Inte
rnational
Stu
die
s Quar
te
rly
,
v
ol.
57
,
no
.
3
,
pp
.
556
-
567,
2013
,
doi:
10
.
1111/i
sq
u.
12013.
[14]
J.
Era
se
y
,
“
Cau
sal
Infe
r
ence
w
it
h
Obs
erv
a
ti
on
al
Da
ta
Introdu
ct
ion
to
Ma
tc
hi
ng,
”
i
n
Ana
lyti
cs,
Po
li
c
y,
and
Gove
rnance
,
Ne
w Have
n:
Yale
U
nive
rsit
y
Press
,
2015
,
pp
.
1
-
37
,
doi: 10.
1177
/15
36867X0800700403.
[15]
W
.
D.
Berr
y
et
al.
,
“
Te
sting
for
Inte
r
ac
t
ion
in
Bina
r
y
Log
it
an
d
Probit
Models
:
Is
a
Produc
t
T
erm
Essentia
l
?
,
”
Ame
rican Journal
of
Politi
cal Science
,
vol. 54, n
o.
1
,
pp
.
248
-
26
6,
2018
,
doi
:
10
.
1111/j
.
1540
-
590
7.
2009.
00429
.
x.
[16]
M.
Som
vanshi,
P.
Chava
n
,
S.
Tam
bade
and
S.
V.
Shinde,
"A
r
eview
of
m
ac
hin
e
l
ea
rning
techniqu
es
using
de
ci
sio
n
tre
e
and
support
vec
tor
m
ac
hin
e,
"
2016
Inte
rna
ti
onal
Confe
re
nce
on
Computing
Comm
unic
ati
on
Control
and
automati
on
(
ICCUBEA
)
,
2016,
p
p.
1
-
7
,
doi
:
10
.
1
109/ICCUBEA.2016
.
7860040
.
[17]
S.
Gava
nkar
and
S.
Sawarka
r,
"D
ec
ision
Tr
ee:
Revi
ew
of
Te
ch
nique
s
for
Miss
ing
Value
s
a
t
Training,
Te
sting
a
nd
Com
pat
ibi
lit
y
,
"
2015
3rd
Inte
rnational
Conf
ere
nce
on
Art
if
i
ci
al
Inte
l
li
gen
ce,
M
odel
li
ng
and
Si
mulati
on
(
AIM
S)
,
2015,
pp
.
122
-
1
26,
doi
:
10
.
1109
/AIMS
.
2015.
29
.
[18]
G.
G.
Moisen,
"
Cla
ss
ifi
c
at
ion
a
nd
Regre
ss
ion
Tre
es
,"
in
Jørge
nsen,
Sve
n
Erik
;
Fat
h,
Brian
D.
(
Edi
tor
-
in
-
Chie
f
)
.
Enc
y
cl
oped
ia
o
f Ec
olog
y
,
CRC p
ress,
vol
.
1
,
pp
.
582
-
588,
2008
.
[19]
W
.
Loh,
“
Cla
ss
ifi
c
at
ion
and
r
e
gre
ss
ion
tre
es,
”
Wil
ey
in
te
rdis
ci
pli
nary
re
vi
e
ws:
data
mining
and
knowl
ed
ge
discov
ery
,
vo
l. 1
,
no
.
1
,
pp
.
14
-
2
3,
2011
,
doi
:
10
.
1002/widm.
8.
[20]
Z.
Indra
,
A.
Setiawan,
A.
B.
Saputra
,
and
Y.
Jus
m
an,
“
Ar
m
s
Dy
namics
Cla
ss
ifi
c
at
ion
S
y
st
em
for
South
Ea
st
As
ia
Regi
on
b
y
Us
ing
K
-
NN
Algorit
h
m
,
”
Journal
of P
hysic
s: Conf
ere
n
ce
S
erie
s
,
vol
.
1
811,
no
.
1
,
p
.
12
095,
2021
.
[21]
I.
V.
Polit
y
,
“
Pol
ity
IV P
roj
ect:
P
oli
tical Re
g
ime
Chara
c
te
rist
ic
s
a
nd
Tra
nsi
ti
ons,
1
800
-
2013
"
Vien
na,
2012
.
[22]
M.
G.
Marshal
l
and
B.
R.
Col
e,
“
Global
Report
on
Confli
ct,
Governa
nc
e
and
Sta
t
e
Fragil
i
t
y
2008,
”
Forei
gn
Po
li
c
y
Bul
letin
,
vol
.
8
,
no.
1
,
3
-
21
,
200
8
.
[23]
Pw
C,
“
Global
Defe
nse
Perspe
cti
ves:
Mapping
Priorit
i
zation
and
Pos
ture
in
a
Ch
al
l
engi
ng
W
orld
,
”
London,
U.K.
,
2015.
[24]
K.
M.
Kade
r
a
a
nd
G.
L
.
Soroki
n,
“
Mea
suring
n
at
ion
al
power
,
”
Inte
rnational
Int
erac
ti
ons
,
vol.
3
0,
no.
3,
pp
.
211
-
230,
2004
,
doi
:
1
0.
1080/0305062
049049
2097.
[25]
D.
J.
Singer,
“
Rec
onstruc
ti
ng
the
Corre
l
at
es
o
f
W
ar
Data
set
on
Mate
rial
Ca
pabi
litie
s
of
Sta
te
s,
1816
-
1985,
”
Inte
rnational
Int
erac
ti
ons
,
vol
.
1
4,
pp
.
115
-
132
,
1987,
doi
:
10
.
10
80/03050628808434695.
[26]
D.
J.
Singer
,
S.
Bremer,
and
J.
S
tuc
ke
y
,
“
Capa
b
ilit
y
Distribu
ti
on,
Unce
rtaint
y
,
and
Major
Pow
er
W
ar,
1820
-
1965
,
”
in
P
eac
e
,
War,
a
nd
Numbers
,
B.
Russ
et
,
Ed
.
B
ev
erly
Hi
ll
s: SAG
E
Publi
ca
t
ions,
1
972.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2502
-
4752
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci,
Vo
l.
23
, N
o.
3
,
Se
ptem
ber
20
21
:
16
54
-
16
62
1662
[27]
S.
M.
W
al
t
,
“
Alli
an
ce
Form
at
io
n
and
th
e
Balan
ce
of
W
orl
d
Or
der
,
”
in
The
Per
il
s
of
Anarchy:
Conte
mpor
ary
Re
ali
sm
and
Int
e
rnational
S
ec
urit
y
,
M
.
Brown,
S. L
y
nn
-
Jones
,
and
S.
Mil
le
r
,
Eds.
Cambridge:
MIT
Press
,
1995.
[28]
D.
M.
Stinnet,
J.
Ti
r,
P.
Schaf
er,
P.
F.
Diehl,
a
n
d
C.
Gochm
an,
“
The
Corre
l
at
es
of
W
ar
Projec
t
Dire
ct
Con
ti
gu
i
t
y
Data
,
Version
3,
”
Confl
ict
Manage
ment
a
nd
Pe
a
ce
Scie
nce
,
vol
.
19
,
no.
2,
pp.
58
-
66,
2002
,
doi
:
10.
1177/073889
420201900203.
[29]
T.
Böhm
el
t
an
d
V.
Bove,
“
Forec
asti
ng
m
ilita
r
y
exp
endi
tur
e,
”
R
ese
arch
a
nd
Pol
it
i
cs
,
vo
l.
1,
no.
1,
p
.
205316801453590,
2014
,
doi
:
10
.
1177/20531680
14535909.
[30]
U.
S.
Marine
Corps,
U.S.
Navy
,
and
U.S.
Co
ast
Guard,
“
Nava
l
Opera
t
ions
Conce
pt
2010:
Im
ple
m
ent
ing
the
Marit
ime
Stra
teg
y
,
”
Uni
te
d
St
ates.
Depa
r
tment
of
the
Nav
y
;
U
nit
ed
St
at
es
.
Ma
rine
Cor
ps;
Uni
t
ed
States.
Co
ast
Guard,
W
ashing
ton
D.C.,
2010
.
[31]
T.
K.
Morrison,
“
The
Eff
e
ct
s
of
Populat
ion
Size
and
Popul
at
ion
Densi
t
y
o
n
the
Manuf
act
ure
d
Expor
ts
of
Deve
lopi
ng
Cou
ntri
es,
”
Sou
the
rn
Ec
onomi
c Journal
,
vo
l
.
43,
no.
3,
p
.
1368
,
Jan
.
1977,
doi
:
10
.
23
07/1057796.
[32]
E.
Nuroglu,
“
The
Im
pac
t
of
P
opula
ti
on
on
Bi
la
t
era
l
Tr
ade
Fl
ows
in
the
ca
se
of
OIC,”
in
2
nd
Inte
rnational
Confe
renc
e
on
I
slamic
E
conomi
c
Int
egrati
on
,
20
10.
[33]
A.
Q.
Flores,
“
Alli
an
ce
s
as
cont
i
guity
in
spa
ti
a
l
m
odel
s
of
m
il
it
a
r
y
expe
ndi
ture
s,
”
Confl
i
ct
Mana
geme
nt
and
P
ea
c
e
Sci
en
ce
,
vol
.
28
,
no.
4,
pp.
402
-
41
8,
Sep
.
2011
,
d
oi:
10
.
1177/073
8894211413064.
[34]
P.
D.
Senese
,
“T
err
i
tor
y
,
contig
uity
,
and
intern
at
ion
al
conf
lict
:
Assessing
a
ne
w
joi
nt
expl
anation,”
Ame
rican
Journal
of
Politi
cal
S
ci
en
ce
,
vo
l.
49,
no
.
4
,
pp
.
76
9
-
779,
2005
,
doi
:
10.
2307
/36476
96.
[35]
V.
Bove
,
“
How
t
he
arms
tr
ade i
s
used
to
se
cur
e
a
c
ce
ss
to
o
il,”
The
Conve
rs
ati
on
,
2
018.
BIOGR
AP
HI
ES OF
A
UTH
ORS
Z
ul
Indra
is
a
cur
ren
tly
as
L
ec
tur
er
at
Univ
ersit
as
Abdurra
b
,
Indone
sia.
H
e
obta
ine
d
his
Master
in
Com
pute
r
and
Infor
m
at
ion
Sci
enc
e
s
from
Univer
siti
Te
kno
logi
P
ET
RON
AS
.
Presently
pu
rsuing
h
is
Ph.D.
in
Com
pute
r
Scie
nce
.
His
rese
ar
c
h
int
ere
st
in
cl
u
des
software
deve
lopment
,
d
a
ta
m
ini
ng
,
and
m
ac
hin
e
l
ea
rn
ing.
A
z
hari
Setia
w
a
n
is
a
cur
r
ent
l
y
as
Lectu
rer
at
Univer
sit
as
Abdurrab,
Indon
esia.
He
obta
in
ed
his
Master
in
I
nte
rna
ti
ona
l
Re
l
at
ions
from
Univer
sit
y
of
Indo
nesia
.
His
rese
a
rch
intere
st
inc
lud
es
Inte
rn
at
ion
al
Relati
o
ns,
AS
EAN
S
ec
uri
t
y
Com
m
unity
,
stra
te
gi
c
studie
s,
a
nd
int
ern
at
ion
al
s
ecurit
y
.
Yess
i
Jus
ma
n
is
a
cur
ren
t
l
y
as
L
ec
tur
er
at
Unive
rsita
s
Muham
m
d
i
y
ah
Yog
y
aka
rt
a
,
Indone
sia
.
She
obta
ine
d
h
er
PhD
fro
m
U
nive
rsiti
Mal
a
y
a
.
Her
rese
arc
h
int
er
est
inc
lud
es
int
el
l
ige
nt
s
y
stem,
d
ata
pro
ce
ss
ing,
and
m
a
c
hine
le
arn
ing.
Arisman
Adna
n
is
a
cur
ren
tly
as
Le
ct
ur
er
at
Univer
sit
as
Ria
u,
Indone
sia.
He
o
bta
in
ed
a
PhD
in
Stat
ist
ic
s,
Uni
ver
sit
y
of
Newc
astl
e
UK
.
He
is
working
on
stat
i
stic
al
m
odel
li
ng
,
ordina
l
d
ata,
and
sta
ti
sti
ca
l
co
m
puti
ng.
Evaluation Warning : The document was created with Spire.PDF for Python.