I
nd
o
ne
s
ia
n J
o
urna
l o
f
E
lect
rica
l En
g
ineering
a
nd
Co
m
pu
t
er
Science
Vo
l.
3
8
,
No
.
2
,
Ma
y
20
2
5
,
p
p
.
1
137
~
1
1
4
8
I
SS
N:
2
502
-
4
7
52
,
DOI
: 1
0
.
1
1
5
9
1
/ijee
cs
.v
3
8
.
i
2
.
pp
1
1
3
7
-
1
1
4
8
1137
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ee
cs
.
ia
esco
r
e.
co
m
Bo
o
sting
stro
ke p
rediction
with
ens
emble lea
rning
on
imba
la
nced healt
hca
re data
O
utm
a
ne
L
a
ba
y
bi
1
,
M
o
ha
med B
enna
n
i Ta
j
1
,
K
ha
lid
E
l
F
a
hs
s
i
2
,
Sa
id
E
l G
a
ro
ua
ni
2
,
M
o
ha
m
ed
L
a
m
rini
1
,
M
o
ha
m
ed
E
l F
a
r
1
1
La
b
o
r
a
t
o
r
y
LPA
I
S
,
D
e
p
a
r
t
m
e
n
t
o
f
C
o
mp
u
t
e
r
S
c
i
e
n
c
e
,
F
a
c
u
l
t
y
o
f
S
c
i
e
n
c
e
D
h
a
r
E
l
M
a
h
r
a
z
,
U
n
i
v
e
r
s
i
t
y
S
i
d
i
M
o
h
a
me
d
B
e
n
A
b
d
e
l
l
a
h
,
F
e
z
,
M
o
r
o
c
c
o
2
La
b
o
r
a
t
o
r
y
LI
S
A
C
,
D
e
p
a
r
t
m
e
n
t
o
f
C
o
mp
u
t
e
r
S
c
i
e
n
c
e
,
F
a
c
u
l
t
y
o
f
S
c
i
e
n
c
e
D
h
a
r
E
l
M
a
h
r
a
z
,
U
n
i
v
e
r
s
i
t
y
S
i
d
i
M
o
h
a
me
d
B
e
n
A
b
d
e
l
l
a
h
,
F
e
z
,
M
o
r
o
c
c
o
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Mar
25
,
2
0
2
4
R
ev
is
ed
No
v
6
,
2
0
2
4
Acc
ep
ted
No
v
11
,
2
0
2
4
De
tec
ti
n
g
stro
k
e
s
a
t
t
h
e
e
a
rly
d
a
y
is
c
ru
c
ial
fo
r
p
re
v
e
n
ti
n
g
h
e
a
lt
h
i
ss
u
e
s
a
n
d
p
o
ten
ti
a
ll
y
sa
v
in
g
li
v
e
s.
P
re
d
icti
n
g
stro
k
e
s
a
c
c
u
ra
tely
c
a
n
b
e
c
h
a
ll
e
n
g
in
g
,
e
sp
e
c
ially
wh
e
n
wo
r
k
in
g
wit
h
u
n
b
a
lan
c
e
d
h
e
a
lt
h
c
a
re
d
a
tas
e
ts.
In
t
h
is
a
rti
c
le,
we
su
g
g
e
st a t
h
o
r
o
u
g
h
m
e
th
o
d
c
o
m
b
in
in
g
m
a
c
h
in
e
lea
rn
i
n
g
(M
L)
a
lg
o
rit
h
m
s
a
n
d
e
n
se
m
b
le
lea
rn
in
g
tec
h
n
iq
u
e
s
to
imp
ro
v
e
th
e
a
c
c
u
ra
c
y
o
f
p
re
d
ictin
g
stro
k
e
s.
Ou
r
a
p
p
r
o
a
c
h
in
c
l
u
d
e
s
u
sin
g
p
re
p
ro
c
e
ss
in
g
m
e
th
o
d
s
f
o
r
tac
k
li
n
g
imb
a
lan
c
e
d
d
a
ta,
fe
a
tu
re
e
n
g
i
n
e
e
rin
g
fo
r
e
x
trac
ti
n
g
k
e
y
i
n
fo
rm
a
ti
o
n
,
a
n
d
u
ti
li
z
i
n
g
d
iffere
n
t
ML
a
lg
o
r
it
h
m
s su
c
h
a
s
ra
n
d
o
m
fo
re
sts
(R
F
)
,
d
e
c
isio
n
tree
s
(DT)
,
a
n
d
g
ra
d
ien
t
b
o
o
sti
n
g
(G
B
o
o
st)
c
las
sifiers
.
T
h
ro
u
g
h
t
h
e
u
ti
l
iza
ti
o
n
o
f
e
n
se
m
b
le
lea
rn
in
g
,
we
a
m
a
lg
a
m
a
te
th
e
a
d
v
a
n
ta
g
e
s
o
f
v
a
ri
o
u
s
m
o
d
e
ls
i
n
o
rd
e
r
t
o
g
e
n
e
ra
te
stro
n
g
e
r
a
n
d
m
o
re
re
li
a
b
le
p
re
d
icti
o
n
s.
By
c
o
n
d
u
c
ti
n
g
th
o
r
o
u
g
h
tes
ts
a
n
d
a
ss
e
ss
m
e
n
ts
o
n
a
v
a
riet
y
o
f
d
a
tas
e
ts,
we
d
e
m
o
n
stra
te
th
e
e
ffica
c
y
o
f
o
u
r
a
p
p
r
o
a
c
h
i
n
a
d
d
r
e
ss
in
g
t
h
e
imb
a
lan
c
e
d
str
o
k
e
d
a
t
a
se
ts
a
n
d
g
re
a
tl
y
e
n
h
a
n
c
e
s
p
re
d
ictio
n
a
c
c
u
ra
c
y
.
We
c
o
n
d
u
c
ted
c
o
m
p
re
h
e
n
si
v
e
tes
ti
n
g
a
n
d
v
a
li
d
a
ti
o
n
to
e
n
s
u
re
t
h
e
re
li
a
b
il
it
y
a
n
d
a
p
p
li
c
a
b
il
i
ty
o
f
o
u
r
m
e
th
o
d
,
imp
ro
v
i
n
g
th
e
a
c
c
u
ra
c
y
o
f
str
o
k
e
p
re
d
ictio
n
a
n
d
su
p
p
o
r
ti
n
g
h
e
a
lt
h
c
a
re
p
lan
n
in
g
a
n
d
re
so
u
rc
e
a
ll
o
c
a
ti
o
n
stra
teg
ies
.
K
ey
w
o
r
d
s
:
Ar
tific
ial
in
tellig
en
t
E
n
s
em
b
le
l
ea
r
n
in
g
I
m
b
alan
ce
d
d
ataset
R
an
d
o
m
f
o
r
est
Str
o
k
e
p
r
ed
ictio
n
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Mo
h
am
ed
B
en
n
a
n
i
T
aj
L
ab
o
r
ato
r
y
L
PAI
S,
Dep
ar
tm
e
n
t
o
f
C
o
m
p
u
ter
Scien
ce
,
Facu
l
ty
o
f
Scien
ce
Dh
a
r
E
l M
ah
r
az
Un
iv
er
s
ity
Sid
i M
o
h
am
ed
B
en
Ab
d
ellah
Fez,
Mo
r
o
cc
o
E
m
ail:
b
en
n
an
i.ta
j@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
Acc
o
r
d
in
g
th
e
wo
r
ld
h
ea
lth
o
r
g
an
izatio
n
(
W
HO)
,
ab
o
u
t
1
5
m
illi
o
n
p
e
o
p
le
h
a
v
e
s
tr
o
k
es
[
1
]
e
v
er
y
y
ea
r
all
o
v
er
th
e
wo
r
ld
[
2
]
,
[
3
]
.
W
HO
d
ef
in
es
s
tr
o
k
e
as
a
b
r
ain
-
r
elate
d
illn
ess
th
at
lead
s
to
th
e
d
y
s
f
u
n
ctio
n
o
f
th
e
b
r
ain
.
T
h
e
r
e
ar
e
two
ty
p
es
o
f
s
tr
o
k
es,
h
em
o
r
r
h
ag
ic
s
tr
o
k
e
(
wh
en
a
b
l
o
o
d
v
ess
el
b
r
ea
k
s
an
d
ca
u
s
es
b
leed
in
g
in
t
h
e
b
r
ain
)
an
d
is
ch
em
ic
s
tr
o
k
e
(
wh
e
n
a
b
lo
o
d
v
ess
el
g
ets
b
lo
ck
ed
)
[
4
]
,
[
5
]
.
T
h
is
d
y
s
f
u
n
ctio
n
is
m
ain
ly
a
r
esu
lt
o
f
v
ess
el
p
r
o
b
lem
s
,
an
d
it
las
ts
f
o
r
lo
n
g
er
th
an
2
4
h
o
u
r
s
.
I
t
’
s
im
p
o
r
tan
t
to
k
n
o
w
th
e
ty
p
e
o
f
s
tr
o
k
e
b
ec
au
s
e
tr
ea
tm
en
ts
d
e
p
en
d
o
n
it.
Dete
ctin
g
s
tr
o
k
e
ea
r
ly
is
cr
u
cial
f
o
r
b
etter
tr
ea
tm
e
n
t r
esu
lts
[
6
]
.
It
’
s
a
cr
itical
m
ed
ical
c
o
n
d
itio
n
,
th
at
r
eq
u
ir
es
ac
c
u
r
ate
an
d
t
im
ely
p
r
e
d
ictio
n
to
f
ac
ilit
ate
p
r
ev
en
tiv
e
m
ea
s
u
r
es
an
d
im
p
r
o
v
e
p
atien
t
o
u
tco
m
es.
T
r
ad
itio
n
al
m
o
d
els
o
f
ten
f
ac
e
ch
allen
g
es
in
h
an
d
lin
g
co
m
p
lex
p
atter
n
s
an
d
r
elatio
n
s
h
ip
s
with
in
h
ea
lth
ca
r
e
d
ata
s
ets
[
7
]
.
T
o
ad
d
r
ess
th
ese
ch
allen
g
es,
we
ex
p
lo
r
e
a
s
cien
tific
ap
p
r
o
ac
h
th
at
le
v
er
ag
es
th
e
s
y
n
er
g
y
o
f
en
s
em
b
le
lear
n
in
g
,
h
y
p
er
p
a
r
am
eter
tu
n
in
g
an
d
m
ac
h
in
e
lear
n
i
n
g
(
ML
)
alg
o
r
ith
m
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
8
,
No
.
2
,
May
20
2
5
:
1
1
3
7
-
1
1
4
8
1138
ML
,
it
’
s
a
p
a
r
t
o
f
ar
tific
ial
in
t
ellig
en
ce
,
h
as
r
e
v
o
lu
tio
n
ized
h
ea
lth
ca
r
e
b
y
p
r
o
v
i
d
in
g
t
o
o
ls
to
an
aly
ze
v
ast
d
atasets
,
d
etec
t
p
atter
n
s
,
an
d
m
a
k
e
p
r
ed
ictio
n
s
.
I
n
t
h
e
co
n
tex
t
o
f
s
tr
o
k
e,
ML
al
g
o
r
ith
m
s
o
f
f
er
th
e
p
o
ten
tial to
r
ef
i
n
e
r
is
k
p
r
e
d
ictio
n
m
o
d
els,
co
n
tr
ib
u
tin
g
to
ea
r
ly
d
iag
n
o
s
is
an
d
p
r
ev
e
n
tiv
e
s
tr
ateg
ies.
An
in
cr
ea
s
in
g
n
u
m
b
e
r
o
f
r
esear
ch
es
h
av
e
in
v
esti
g
ated
th
e
ap
p
licatio
n
o
f
ML
m
o
d
els
in
s
tr
o
k
e
p
r
ed
ictio
n
in
t
h
e
last
d
ec
ad
e.
Ma
h
esh
war
i
et
a
l
.
[
8
]
p
r
o
v
i
d
e
s
a
s
tu
d
y
o
f
v
ar
io
u
s
r
is
k
f
ac
to
r
s
to
u
n
d
er
s
tan
d
t
h
e
p
r
o
b
a
b
ilit
y
o
f
s
tr
o
k
e.
I
t
u
s
ed
a
r
eg
r
ess
io
n
-
b
ased
ap
p
r
o
ac
h
t
o
id
en
tify
th
e
r
elatio
n
b
etwe
e
n
a
f
ac
to
r
an
d
its
co
r
r
esp
o
n
d
in
g
im
p
ac
t o
n
s
tr
o
k
e.
E
x
p
lo
r
in
g
a
Kag
g
le
d
ataset,
Sailas
y
a
an
d
Ku
m
ar
i
[
9
]
d
elv
ed
in
to
s
tr
o
k
e
p
r
ed
ictio
n
u
s
in
g
v
ar
io
u
s
ML
alg
o
r
ith
m
s
,
in
clu
d
in
g
l
o
g
is
tic
r
eg
r
ess
io
n
(
L
R
)
,
k
-
n
ea
r
est
n
eig
h
b
o
u
r
(
KNN)
,
r
an
d
o
m
f
o
r
est
(
R
F)
,
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM)
,
N
aïv
e
B
ay
es
(
N
B
)
,
an
d
d
ec
is
io
n
tr
ee
(
DT
)
alg
o
r
ith
m
s
.
T
o
ad
d
r
e
s
s
im
b
alan
ce
d
d
ata,
an
u
n
d
e
r
s
am
p
lin
g
m
eth
o
d
wa
s
em
p
lo
y
ed
.
T
h
e
f
in
d
in
g
s
r
ev
ea
led
th
at
NB
ex
h
ib
ited
th
e
h
ig
h
est
p
er
f
o
r
m
a
n
ce
,
b
o
asti
n
g
an
o
v
er
all
ac
cu
r
ac
y
o
f
8
2
%.
I
n
co
m
p
ar
is
o
n
,
KNN
an
d
SVM
b
o
th
ac
h
iev
e
d
a
n
8
0
%
ac
cu
r
ac
y
,
wh
il
e
LR
y
ield
ed
a
s
lig
h
tly
lo
wer
ac
cu
r
ac
y
o
f
7
8
%.
Nwo
s
u
et
a
l
.
[
1
0
]
h
ar
n
ess
ed
e
lectr
o
n
ic
h
ea
lth
r
ec
o
r
d
s
a
n
d
u
tili
ze
d
a
d
ataset
p
r
o
v
i
d
ed
b
y
Mc
Kin
s
ey
an
d
c
o
m
p
a
n
y
,
en
c
o
m
p
ass
in
g
1
1
d
is
tin
ct
attr
ib
u
tes
s
u
ch
as
b
o
d
y
m
ass
in
d
ex
,
h
ea
r
t
d
is
ea
s
e,
m
ar
ital
s
tatu
s
,
ag
e,
av
er
ag
e
b
lo
o
d
g
lu
c
o
s
e,
an
d
s
m
o
k
in
g
s
tatu
s
.
W
ith
in
th
is
d
ataset,
5
4
8
p
atien
ts
h
ad
ex
p
e
r
ien
ce
d
a
s
tr
o
k
e,
wh
ile
2
8
5
2
4
p
atien
ts
h
ad
n
o
t
en
c
o
u
n
ter
ed
an
y
p
r
io
r
s
tr
o
k
es.
D
u
e
to
th
e
d
ataset
’
s
im
b
alan
c
e,
1
0
0
0
d
o
wn
s
izin
g
ex
p
er
im
en
ts
wer
e
co
n
d
u
cted
t
o
m
itig
ate
s
am
p
lin
g
b
ias.
Su
b
s
eq
u
en
tly
,
7
0
%
o
f
th
e
d
ataset
was
allo
ca
ted
f
o
r
tr
ain
in
g
,
with
th
e
r
em
ai
n
in
g
3
0
%
r
eser
v
ed
f
o
r
test
in
g
p
u
r
p
o
s
es.
Acr
o
s
s
th
e
1
0
0
0
d
o
wn
s
izin
g
ex
p
e
r
im
en
ts
,
th
e
n
eu
r
al
n
etwo
r
k
m
o
d
el
d
em
o
n
s
tr
ated
s
u
p
er
io
r
p
er
f
o
r
m
a
n
c
e,
ac
h
iev
in
g
th
e
h
ig
h
est
ac
cu
r
ac
y
at
7
5
.
0
2
%.
Fo
llo
win
g
clo
s
ely
,
th
e
RF
m
o
d
el
attain
ed
an
ac
cu
r
ac
y
o
f
7
4
.
5
3
%,
an
d
th
e
DT
m
o
d
el
e
x
h
ib
ited
an
ac
cu
r
ac
y
o
f
7
4
.
3
1
%.
I
n
th
e
s
tu
d
y
r
e
f
er
en
ce
d
as
[
1
1
]
,
th
e
r
esear
ch
e
r
s
o
p
ted
f
o
r
in
tr
icate
alg
o
r
ith
m
s
lik
e
ADABo
o
s
t
an
d
XGB,
ac
h
iev
in
g
o
u
tco
m
es
co
m
p
ar
ab
le
t
o
o
u
r
s
.
Ho
wev
e
r
,
o
u
r
s
tu
d
y
ac
h
iev
e
d
im
p
r
ess
iv
e
r
esu
lts
u
s
in
g
s
im
p
ler
alg
o
r
ith
m
s
,
a
m
o
r
e
p
r
ef
e
r
ab
le
an
d
ef
f
icien
t
a
p
p
r
o
ac
h
.
I
n
a
s
tu
d
y
b
y
Sailas
y
a
an
d
Ku
m
a
r
i
[
9
]
,
s
im
ilar
to
th
e
fi
n
d
in
g
s
in
r
e
f
er
en
ce
[
1
2
]
,
t
h
e
Kag
g
le
d
ataset
was
u
tili
ze
d
a
lo
n
g
with
v
ar
io
u
s
alg
o
r
ith
m
s
i
n
clu
d
in
g
DT
,
NB
,
SVM
,
RF
,
K
NN
,
an
d
LR
.
T
h
eir
r
esu
lts
d
em
o
n
s
tr
ated
th
at
DT
o
u
tp
er
f
o
r
m
ed
t
h
e
o
th
er
al
g
o
r
ith
m
s
,
ac
h
iev
i
n
g
th
e
h
ig
h
est p
er
f
o
r
m
a
n
ce
,
f
o
llo
wed
b
y
KNN
with
an
ac
c
u
r
ac
y
o
f
9
6
.
3
%.
T
h
e
p
r
o
b
lem
is
th
at
th
e
r
is
in
g
in
cid
en
ce
o
f
s
tr
o
k
es
em
p
h
asizes
th
e
n
ee
d
f
o
r
e
f
f
ec
tiv
e
p
r
ed
ictio
n
m
o
d
els
th
at
ac
cu
r
ately
id
e
n
tify
in
d
iv
i
d
u
als
a
t
r
is
k
.
W
h
ile
p
r
ev
io
u
s
s
tu
d
ies
h
av
e
tack
l
ed
th
is
is
s
u
e
u
s
in
g
class
ical
m
eth
o
d
s
,
th
ese
ap
p
r
o
ac
h
es h
av
e
n
o
t p
r
o
d
u
ce
d
s
atis
f
ac
to
r
y
r
esu
lts
.
Ou
r
co
n
tr
ib
u
tio
n
aim
s
to
in
v
esti
g
ate
h
o
w
ML
a
n
d
en
s
em
b
le
lear
n
in
g
tech
n
iq
u
es
ca
n
b
e
u
s
ed
to
p
r
ed
ict
s
tr
o
k
es.
B
y
ex
am
in
i
n
g
th
e
co
n
tr
ib
u
tio
n
s
o
f
ea
c
h
m
eth
o
d
o
lo
g
y
an
d
th
eir
s
y
n
er
g
ies,
th
e
r
esear
ch
s
ee
k
s
to
p
r
o
v
id
e
a
co
m
p
r
eh
e
n
s
iv
e
u
n
d
er
s
tan
d
in
g
o
f
h
o
w
ad
v
a
n
ce
d
co
m
p
u
tatio
n
al
tech
n
iq
u
es
ca
n
b
e
h
ar
n
ess
ed
to
en
h
an
ce
ac
c
u
r
ac
y
,
i
n
ter
p
r
eta
b
ilit
y
,
an
d
clin
ical
r
elev
a
n
ce
in
s
tr
o
k
e
p
r
ed
ictio
n
m
o
d
els.
T
h
e
r
est
o
f
th
is
ar
ticle
is
s
tr
u
ctu
r
ed
in
to
s
ev
e
r
al
s
ec
tio
n
s
,
in
clu
d
in
g
o
n
e
th
at
d
escr
ib
es
o
u
r
m
eth
o
d
.
An
o
th
er
s
ec
tio
n
is
d
ed
icate
d
to
r
esu
lts
an
d
d
is
cu
s
s
io
n
,
p
r
esen
tin
g
r
esear
ch
co
n
cl
u
s
io
n
s
alo
n
g
with
co
m
p
ar
is
o
n
s
to
o
th
er
s
im
ilar
t
ec
h
n
iq
u
es
.
Fin
ally
,
a
s
ec
tio
n
s
u
m
m
ar
izes
th
e
f
in
d
in
g
s
an
d
s
u
g
g
ests
d
ir
ec
tio
n
s
f
o
r
f
u
tu
r
e
r
esear
c
h
.
2.
M
E
T
H
O
D
I
n
th
e
s
ec
tio
n
s
th
at
f
o
llo
w,
a
d
etailed
d
escr
ip
tio
n
o
f
th
e
m
eth
o
d
s
th
at
was
u
s
ed
f
o
r
th
is
wo
r
k
is
p
r
o
v
id
e
d
.
I
n
s
ec
tio
n
2
.
1
,
th
e
d
etails
o
f
th
e
d
ataset
th
at
was
u
s
ed
ar
e
ex
p
lain
e
d
.
Sectio
n
2
.
2
o
u
tlin
es
th
e
d
ata
p
r
ep
r
o
c
ess
in
g
tech
n
iq
u
e.
I
n
s
ec
tio
n
2
.
3
,
th
e
p
r
o
p
o
s
ed
m
et
h
o
d
is
p
r
esen
ted
,
wh
ile
in
s
ec
tio
n
2
.
4
,
th
e
ML
alg
o
r
ith
m
s
u
s
ed
f
o
r
s
tr
o
k
e
p
r
e
d
ictio
n
ar
e
ex
p
lain
ed
i
n
g
r
ea
te
r
d
etail.
2
.
1
.
Descript
io
n o
f
da
t
a
s
et
T
h
e
d
ataset
u
s
ed
in
o
u
r
s
tu
d
y
is
ca
lled
‘
s
tr
o
k
e
p
r
e
d
ictio
n
d
a
taset’,
it
co
n
tain
s
im
p
o
r
ta
n
t
in
f
o
r
m
atio
n
f
r
o
m
m
ed
ical
r
ec
o
r
d
s
,
lik
e
wh
eth
er
a
p
atien
t
h
as
h
y
p
e
r
t
en
s
io
n
,
h
ea
r
t
d
is
ea
s
e,
v
ar
io
u
s
p
h
y
s
io
lo
g
ical
an
d
en
v
ir
o
n
m
en
tal
d
etails.
T
h
e
d
at
aset is o
r
g
an
ized
in
to
r
o
ws an
d
co
lu
m
n
s
,
with
ea
ch
r
o
w
r
ep
r
esen
tin
g
a
d
if
f
er
en
t
p
atien
t.
T
h
e
d
ataset
h
as
5
1
1
0
r
o
ws,
a
n
d
ea
ch
r
o
w
is
in
f
o
a
b
o
u
t
o
n
e
p
er
s
o
n
.
T
h
er
e
ar
e
1
2
co
lu
m
n
s
.
T
en
tell u
s
th
in
g
s
ab
o
u
t
th
e
p
eo
p
le,
lik
e
h
ea
lth
co
n
d
itio
n
s
.
On
e
co
lu
m
n
h
as
an
I
D,
an
d
an
o
th
er
s
ay
s
if
th
e
p
er
s
o
n
h
ad
a
s
tr
o
k
e
(
1
)
o
r
n
o
t
(
0
)
.
T
h
e
d
ataset
is
n
o
t
b
alan
ce
d
;
4
8
6
1
p
eo
p
le
ar
e
n
o
r
m
al,
an
d
2
4
9
h
ad
a
s
tr
o
k
e.
T
h
is
im
b
alan
ce
m
ig
h
t
af
f
ec
t
o
u
r
m
o
d
els,
s
o
we
’
r
e
g
o
in
g
to
f
ix
it
d
u
r
i
n
g
tr
ain
i
n
g
to
m
ak
e
th
in
g
s
m
o
r
e
e
v
en
.
Yo
u
c
an
g
et
th
is
d
ataset
o
n
Kag
g
le
u
s
in
g
th
is
lin
k
:
s
tr
o
k
e
p
r
ed
ictio
n
d
a
taset
.
We
’
r
e
s
tu
d
y
in
g
th
is
d
ataset
to
b
u
ild
g
o
o
d
m
o
d
els
f
o
r
p
r
ed
ictin
g
s
tr
o
k
es,
k
ee
p
in
g
in
m
i
n
d
th
e
u
n
e
v
en
n
u
m
b
er
o
f
n
o
r
m
al
an
d
s
tr
o
k
e
ca
s
es.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
B
o
o
s
tin
g
s
tr
o
ke
p
r
ed
ictio
n
w
ith
en
s
emb
le
lea
r
n
in
g
o
n
imb
a
l
a
n
ce
d
h
ea
lth
ca
r
e
d
a
ta
(
Ou
tma
n
e
La
b
a
yb
i
)
1139
2
.
2
.
Da
t
a
prepro
ce
s
s
ing
Han
d
le
m
is
s
in
g
v
alu
es
:
t
h
e
w
ay
we
f
ill
in
m
is
s
in
g
d
ata
d
e
p
en
d
s
o
n
wh
at
k
in
d
o
f
d
ata
it
i
s
an
d
wh
at
th
e
d
ataset
is
lik
e.
T
h
e
B
MI
co
lu
m
n
h
as
2
0
1
an
d
th
e
s
m
o
k
i
n
g
s
tatu
s
h
as
1544
m
is
s
in
g
v
alu
es.
T
o
h
an
d
le
th
ese
m
is
s
in
g
v
alu
es,
we
h
av
e
a
m
o
r
e
o
p
tio
n
s
.
Fo
r
u
s
,
b
ec
au
s
e
B
MI
an
d
s
m
o
k
in
g
s
tatu
s
ar
e
im
p
o
r
tan
t
f
ac
t
o
r
s
an
d
we
’
r
e
m
is
s
in
g
q
u
ite
a
f
ew
v
alu
es
f
o
r
t
h
em
,
it m
ak
es
s
en
s
e
t
o
f
ill
in
t
h
o
s
e
m
is
s
in
g
v
alu
es.
Fo
r
th
ese
we
u
s
e
th
e
KNN
alg
o
r
ith
m
to
im
p
u
te
B
MI
m
is
s
in
g
v
alu
es,
f
o
r
ea
ch
m
is
s
in
g
v
alu
e,
f
in
d
its
KNN
b
ased
o
n
o
th
er
f
ea
tu
r
es
a
n
d
f
o
r
th
e
s
m
o
k
i
n
g
s
tatu
s
we
u
tili
ze
th
e
RF
alg
o
r
ith
m
to
im
p
u
te
m
is
s
in
g
v
alu
es.
T
r
ain
a
RF
m
o
d
el
o
n
th
e
s
u
b
s
et
o
f
d
ata
with
co
m
p
lete
in
f
o
r
m
atio
n
,
an
d
p
r
e
d
ict
th
e
m
i
s
s
in
g
v
alu
es b
ased
o
n
o
th
er
f
e
atu
r
es.
E
n
co
d
e
ca
teg
o
r
ical
v
ar
ia
b
les
:
c
o
n
v
er
tin
g
ca
teg
o
r
ical
v
a
r
iab
les
to
n
u
m
er
ical
f
o
r
m
at
u
s
in
g
a
m
et
h
o
d
ca
lled
o
n
e
-
h
o
t
en
co
d
in
g
is
a
c
o
m
m
o
n
p
r
ep
r
o
ce
s
s
in
g
s
tep
to
cr
ea
te
b
in
a
r
y
co
lu
m
n
s
(
0
o
r
1
)
f
o
r
ea
c
h
ca
teg
o
r
y
,
ef
f
ec
tiv
ely
tr
an
s
f
o
r
m
in
g
it in
to
a
s
et
o
f
n
u
m
er
ical
f
ea
tu
r
es.
I
m
b
alan
ce
d
d
ataset
h
a
n
d
lin
g
:
h
an
d
lin
g
im
b
alan
ce
d
d
atasets
is
cr
u
cial
i
n
ML
,
as
m
o
d
els
tr
ain
ed
o
n
s
u
ch
d
atasets
m
ig
h
t
h
a
v
e
a
b
ias
to
war
d
s
th
e
m
ajo
r
ity
class
[
1
3
]
.
I
n
t
h
e
co
n
tex
t
o
f
a
s
tr
o
k
e
d
ataset,
wh
er
e
s
tr
o
k
es
ar
e
lik
ely
to
b
e
a
m
in
o
r
ity
class
,
ad
d
r
ess
in
g
t
h
e
im
b
a
lan
ce
F
ig
u
r
e
1
is
im
p
o
r
tan
t
f
o
r
c
r
e
a
t
i
n
g
a
r
e
l
i
a
b
l
e
a
n
d
e
f
f
e
c
t
i
v
e
m
o
d
e
l
.
T
h
e
r
e
a
r
e
s
e
v
e
r
a
l
m
e
t
h
o
d
s
t
o
c
o
m
p
e
n
s
a
t
e
f
o
r
a
n
i
m
b
a
l
a
n
c
e
o
f
c
l
a
s
s
e
s
i
n
a
d
a
t
a
s
e
t
.
D
e
p
e
n
d
i
n
g
o
n
t
h
e
a
m
o
u
n
t
o
f
d
a
t
a
a
v
a
i
l
a
b
l
e
,
w
e
w
i
l
l
t
h
e
n
c
h
o
o
s
e
o
n
e
o
r
o
t
h
e
r
o
f
t
h
e
f
o
l
l
o
w
i
n
g
m
e
t
h
o
d
s
.
Data
s
ca
lin
g
:
i
n
s
tr
o
k
e
p
r
e
d
ictio
n
m
o
d
els,
th
e
d
ataset
m
ay
co
n
tain
n
u
m
er
ical
f
ea
tu
r
es
s
u
ch
as
ag
e
,
b
lo
o
d
p
r
ess
u
r
e,
av
g
_
g
lu
c
o
s
e_
lev
el,
an
d
B
MI
F
ig
u
r
e
2
.
T
h
ese
f
ea
tu
r
es
ca
n
h
av
e
v
astl
y
d
if
f
er
en
t
s
ca
les
an
d
u
n
its
,
wh
ich
ca
n
a
f
f
ec
t
t
h
e
p
e
r
f
o
r
m
an
ce
o
f
ML
alg
o
r
ith
m
s
[
1
4
]
.
I
n
th
is
s
tu
d
y
,
we
u
s
e
th
e
m
in
-
m
ax
s
ca
lin
g
()
.
T
h
is
tech
n
iq
u
e
s
ca
les th
e
f
ea
tu
r
es to
a
f
ix
ed
r
an
g
e
,
ty
p
ically
b
etwe
en
0
an
d
1
F
ig
u
r
e
3
.
Fig
u
r
e
1
.
Un
d
er
s
am
p
lin
g
an
d
o
v
er
s
am
p
lin
g
Fig
u
r
e
2.
Or
ig
i
n
al
n
u
m
er
ical
f
ea
tu
r
es
Fig
u
r
e
3
.
D
ataset
s
ca
led
af
ter
u
s
in
g
Min
Ma
x
Scaler
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
8
,
No
.
2
,
May
20
2
5
:
1
1
3
7
-
1
1
4
8
1140
=
(
−
)
(
−
)
(
1
)
I
t
is
ca
lcu
lated
u
s
in
g
th
e
eq
4
wh
er
e
X
r
ep
r
esen
ts
th
e
o
r
ig
in
al
v
al
u
e
o
f
t
h
e
f
ea
t
u
r
e,
X
m
in
is
th
e
s
m
allest
v
alu
e
o
f
th
e
f
ea
tu
r
e,
an
d
Xm
ax
is
th
e
lar
g
est
v
alu
e
o
f
th
e
f
e
atu
r
e.
T
h
is
p
r
ep
r
o
ce
s
s
in
g
s
tep
en
h
an
ce
s
th
e
p
er
f
o
r
m
an
ce
a
n
d
in
te
r
p
r
et
ab
ilit
y
o
f
th
e
m
o
d
els,
u
ltima
te
ly
lead
in
g
to
b
etter
h
ea
lth
ca
r
e
o
u
tco
m
es.
2
.
3
.
P
r
o
po
s
ed
m
et
ho
d
At
th
e
b
eg
i
n
n
in
g
o
f
o
u
r
e
x
p
er
i
m
en
t,
p
r
esen
ted
in
F
ig
u
r
e
4
w
e
ca
r
ef
u
lly
p
r
e
p
ar
ed
th
e
d
ataset
to
m
ak
e
s
u
r
e
it
was
g
o
o
d
t
o
u
s
e.
Af
ter
we
im
p
u
te
an
d
f
ix
e
d
an
y
m
is
s
in
g
in
f
o
r
m
atio
n
,
d
ec
o
d
e
th
e
ca
teg
o
r
ies
v
ar
iab
les
to
n
u
m
b
er
s
u
s
in
g
o
n
e
h
o
t e
n
co
d
in
g
,
a
n
d
m
a
d
e
s
u
r
e
all
th
e
d
ata
was in
th
e
s
am
e
r
an
g
e.
Nex
t
s
tep
,
we
s
p
lit
o
u
r
d
ataset
in
to
tr
ain
in
g
d
ataset
wh
ich
h
ad
8
0
%
o
f
t
h
e
d
ata,
an
d
test
in
g
d
ataset
wh
ich
h
a
d
2
0
%.
Do
in
g
th
is
h
elp
ed
u
s
s
ee
h
o
w
well
th
e
m
o
d
els
wo
r
k
ed
.
T
h
e
n
,
we
d
ea
lt
with
th
e
p
r
o
b
lem
o
f
th
er
e
b
ein
g
m
o
r
e
o
f
o
n
e
ty
p
e
o
f
d
ata
th
an
th
e
o
th
er
in
th
e
tr
ain
in
g
s
et.
W
e
u
s
ed
a
tech
n
iq
u
e
ca
lled
SMOT
E
to
m
ak
e
m
o
r
e
o
f
t
h
e
m
in
o
r
ity
cla
s
s
(
ca
s
es wh
er
e
p
eo
p
le
h
ad
s
tr
o
k
es).
T
h
is
was to
h
elp
t
h
e
m
o
d
els lea
r
n
b
etter
.
On
ce
we
h
ad
a
b
alan
ce
d
tr
ai
n
in
g
s
et,
we
tau
g
h
t
th
r
ee
d
if
f
er
en
t
m
o
d
els
RF
,
XGBo
o
s
t,
an
d
SVM
u
s
in
g
th
at
d
ata.
E
ac
h
m
o
d
el
l
ea
r
n
ed
t
h
e
p
atter
n
s
in
t
h
e
d
ata
.
T
o
m
ak
e
th
e
p
r
ed
ictio
n
s
e
v
e
n
b
etter
,
we
u
s
ed
a
m
eth
o
d
ca
lled
m
o
d
el
s
tack
in
g
.
T
h
is
m
ea
n
s
we
p
u
t
t
h
e
th
r
ee
m
o
d
els
to
g
eth
er
an
d
let
th
em
lear
n
f
r
o
m
ea
ch
o
th
er
.
T
h
is
h
elp
ed
m
a
k
e
th
e
p
r
ed
ictio
n
s
m
o
r
e
ac
c
u
r
ate
b
y
u
s
in
g
th
e
s
tr
en
g
th
s
o
f
ea
c
h
m
o
d
el.
Af
ter
s
tack
in
g
th
e
m
o
d
els,
we
co
m
b
in
e
d
th
eir
p
r
ed
ictio
n
s
u
s
in
g
a
v
o
tin
g
s
y
s
tem
.
T
h
is
let
u
s
u
s
e
all
th
e
m
o
d
els
to
g
eth
er
to
m
ak
e
p
r
ed
ictio
n
s
.
B
y
d
o
in
g
t
h
is
,
we
co
u
ld
m
ak
e
b
etter
p
r
ed
ictio
n
s
o
v
er
all.
Fin
ally
,
we
test
ed
h
o
w
well
o
u
r
co
m
b
in
ed
m
o
d
el
wo
r
k
ed
u
s
in
g
th
e
test
s
et.
W
e
ex
am
in
ed
its
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
an
d
F1
-
s
co
r
e
t
o
s
ee
h
o
w
well
i
t p
r
ed
icted
th
e
li
k
elih
o
o
d
o
f
s
o
m
eo
n
e
h
a
v
in
g
a
s
tr
o
k
e.
Fig
u
r
e
4
.
I
ll
u
s
tr
ates th
e
wo
r
k
f
l
o
w
f
o
r
p
r
ed
ictin
g
s
tr
o
k
es u
s
in
g
th
e
p
r
o
v
id
e
d
d
ataset
2
.
4
M
a
chine
lea
rning
m
o
dels
2
.
4
.
1
.
Ra
nd
o
m
f
o
re
s
t
RF
i
s
a
s
tr
o
n
g
co
m
p
u
ter
m
et
h
o
d
th
at
co
m
b
in
es
m
an
y
DT
an
d
co
m
b
in
es
th
eir
g
u
ess
es
to
m
ak
e
b
etter
p
r
ed
ictio
n
s
o
v
er
all.
I
t
’
s
r
ea
lly
g
o
o
d
at
d
ea
lin
g
with
lo
ts
o
f
in
f
o
r
m
atio
n
an
d
co
m
p
licated
co
n
n
ec
tio
n
s
b
etwe
en
d
if
f
er
en
t
p
a
r
ts
o
f
th
e
d
ata,
wh
ich
m
ak
es
it
g
r
ea
t
f
o
r
p
r
e
d
ictin
g
s
tr
o
k
es.
Ou
r
r
esu
lts
co
r
r
o
b
o
r
ate
p
r
ev
io
u
s
f
in
d
in
g
s
d
em
o
n
s
tr
atin
g
th
e
ef
f
icac
y
o
f
RF
in
h
ea
lth
ca
r
e
ap
p
l
ica
tio
n
s
[
1
5
]
.
B
y
h
ar
n
ess
in
g
th
e
en
s
em
b
le
n
atu
r
e
o
f
RF
,
we
en
h
an
ce
t
h
e
r
o
b
u
s
tn
ess
an
d
g
en
er
aliza
tio
n
ca
p
a
b
ilit
y
o
f
o
u
r
p
r
ed
ictiv
e
m
o
d
el
,
o
f
f
er
i
n
g
v
al
u
ab
le
in
s
ig
h
ts
in
to
s
tr
o
k
e
r
is
k
ass
ess
m
en
t.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
B
o
o
s
tin
g
s
tr
o
ke
p
r
ed
ictio
n
w
ith
en
s
emb
le
lea
r
n
in
g
o
n
imb
a
l
a
n
ce
d
h
ea
lth
ca
r
e
d
a
ta
(
Ou
tma
n
e
La
b
a
yb
i
)
1141
2
.
4
.
2
.
Su
pp
o
rt
v
ec
t
o
r
m
a
chi
nes
SVM
[
1
6
]
ar
e
p
o
wer
f
u
l
s
u
p
er
v
is
ed
lear
n
in
g
alg
o
r
ith
m
s
t
h
at
f
in
d
th
e
o
p
tim
al
h
y
p
er
p
lan
e
t
o
s
ep
ar
ate
class
es
in
th
e
f
ea
tu
r
e
s
p
ac
e.
SVMs
m
ax
im
ize
th
e
m
ar
g
in
b
etwe
en
th
e
class
es
an
d
ca
n
h
an
d
le
n
o
n
-
lin
ea
r
d
ec
is
io
n
b
o
u
n
d
ar
ies
u
s
in
g
k
e
r
n
el
f
u
n
ctio
n
s
.
SVM
[
1
7
]
ar
e
g
o
o
d
at
d
iv
id
i
n
g
d
ata
in
to
t
wo
g
r
o
u
p
s
an
d
ar
e
f
am
o
u
s
f
o
r
b
ein
g
ab
le
to
wo
r
k
well
ev
en
with
n
ew
d
ata
t
h
ey
h
av
en
’
t seen
b
ef
o
r
e.
2
.
4
.
3
.
E
ns
em
ble
l
ea
rning
Ma
n
y
s
tu
d
ies
in
h
ea
lth
ca
r
e,
a
s
s
h
o
wn
in
r
ef
er
en
c
es
[
1
8
]
,
[
1
9
]
,
h
a
v
e
u
s
ed
en
s
em
b
le
lear
n
i
n
g
.
T
h
e
s
e
s
tu
d
ies
f
o
cu
s
m
ain
ly
o
n
u
s
in
g
th
e
s
am
e
ty
p
e
o
f
m
ac
h
in
e
-
lear
n
in
g
m
eth
o
d
s
as
t
h
eir
b
as
ic
to
o
ls
,
wh
ich
ar
e
o
f
ten
ca
lled
wea
k
lear
n
e
r
s
.
T
h
r
ee
p
o
p
u
lar
m
eth
o
d
s
,
b
o
o
ts
tr
ap
ag
g
r
eg
atin
g
(
b
ag
g
in
g
)
,
s
tac
k
in
g
,
an
d
b
o
o
s
tin
g
,
co
m
b
in
e
th
ese
wea
k
lea
r
n
er
s
.
B
ag
g
in
g
wo
r
k
s
b
y
t
r
ain
in
g
m
an
y
co
p
ies
o
f
th
e
s
am
e
b
asic
l
ea
r
n
in
g
m
eth
o
d
o
n
d
if
f
e
r
en
t
p
ar
ts
o
f
th
e
tr
ain
in
g
d
ata
F
ig
u
r
e
5
,
ch
o
s
e
n
r
an
d
o
m
l
y
b
u
t
with
r
ep
lace
m
en
t.
W
h
en
p
r
ed
ictin
g
s
tr
o
k
e
s
,
b
ag
g
in
g
co
u
ld
b
e
u
s
ed
with
DT
o
r
RF
.
I
n
th
is
m
eth
o
d
,
ea
ch
tr
ee
in
th
e
g
r
o
u
p
is
tr
ain
ed
o
n
a
d
if
f
e
r
en
t
r
a
n
d
o
m
s
am
p
le
o
f
th
e
d
ataset.
B
ag
g
in
g
h
elp
s
m
ak
e
p
r
ed
ictio
n
s
m
o
r
e
r
eliab
le
an
d
ac
cu
r
ate
b
y
co
m
b
in
in
g
th
e
p
r
e
d
ictio
n
s
o
f
m
u
ltip
le
m
o
d
els
[
2
0
]
.
̂
f
b
a
ggin
g
(
)
=
1
∑
(
)
=
1
(
2
)
−
̂
f
b
a
ggin
g
(
)
is
th
e
en
s
em
b
le
p
r
ed
ictio
n
.
−
th
e
n
u
m
b
er
o
f
b
asic le
ar
n
er
s
.
−
f
b
(
x
)
r
ef
er
s
to
t
h
e
p
r
e
d
ictio
n
m
ad
e
b
y
th
e
b
-
t
h
b
asic le
ar
n
er
.
Fig
u
r
e
1
.
B
ag
g
in
g
en
s
em
b
le
T
h
e
d
iag
r
am
in
th
e
F
ig
u
r
e
4
illu
s
tr
ates
h
o
w
th
e
B
ag
g
in
g
en
s
em
b
le
m
eth
o
d
wo
r
k
s
.
I
n
B
ag
g
in
g
,
m
u
ltip
le
p
r
o
ce
s
s
es
h
ap
p
en
s
im
u
ltan
eo
u
s
ly
.
T
h
e
p
r
im
ar
y
g
o
al
o
f
B
ag
g
in
g
is
to
d
ec
r
ea
s
e
v
ar
iab
ilit
y
in
th
e
p
r
ed
ictio
n
s
m
a
d
e
b
y
t
h
e
en
s
e
m
b
le.
B
o
o
s
tin
g
is
a
m
eth
o
d
wh
er
e
wea
k
lear
n
er
s
a
r
e
tr
ai
n
ed
o
n
e
af
ter
a
n
o
th
e
r
in
a
s
er
ies
o
f
s
t
ep
s
,
wh
er
e
ea
ch
n
ew
m
o
d
el
f
o
c
u
s
es
o
n
th
e
in
s
tan
ce
s
th
at
wer
e
m
i
s
class
if
ied
b
y
th
e
p
r
ev
io
u
s
m
o
d
els
F
ig
u
r
e
6
.
Alg
o
r
ith
m
s
lik
e
ad
ap
tiv
e
b
o
o
s
tin
g
(
Ad
aBo
o
s
t
)
an
d
g
r
a
d
ien
t
b
o
o
s
tin
g
m
ac
h
i
n
es
(
GB
M)
ar
e
co
m
m
o
n
ly
u
s
ed
b
o
o
s
tin
g
m
eth
o
d
s
.
I
n
th
e
co
n
tex
t
o
f
s
tr
o
k
e
p
r
ed
ictio
n
,
b
o
o
s
tin
g
alg
o
r
ith
m
s
co
u
ld
b
e
ap
p
l
ied
to
DT
o
r
o
th
e
r
wea
k
lear
n
er
s
to
iter
ativ
ely
im
p
r
o
v
e
t
h
e
p
r
ed
ictio
n
ac
cu
r
a
cy
b
y
em
p
h
asizi
n
g
d
if
f
icu
lt
-
to
-
class
if
y
in
s
tan
ce
s
r
elate
d
to
s
tr
o
k
e
r
is
k
f
ac
to
r
s
[
2
1
]
.
̂
b
oostin
g
(
)
=
1
1
(
)
+
2
2
(
)
+
⋯
+
(
)
(
3
)
−
̂
b
oostin
g
(
)
is
th
e
en
s
em
b
le
p
r
ed
ictio
n
.
−
i
is
th
e
weig
h
t a
s
s
ig
n
ed
to
th
e
-
th
b
ase
lear
n
er
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
8
,
No
.
2
,
May
20
2
5
:
1
1
3
7
-
1
1
4
8
1142
Fig
u
r
e
6
.
I
ll
u
s
tr
ates th
e
b
o
o
s
tin
g
en
s
em
b
le
m
et
h
o
d
T
h
e
m
ain
g
o
al
o
f
b
o
o
s
tin
g
is
to
d
ec
r
ea
s
e
th
e
m
is
tak
e
s
m
ad
e
in
th
e
en
s
em
b
le
d
e
cisi
o
n
.
So
,
th
e
class
if
ier
s
p
ick
ed
f
o
r
th
e
g
r
o
u
p
u
s
u
ally
s
h
o
u
ld
h
av
e
le
s
s
ch
an
ce
o
f
b
ein
g
wr
o
n
g
b
u
t
m
ay
b
e
s
im
p
ler
,
with
f
ewe
r
th
in
g
s
to
lear
n
.
Stack
in
g
also
k
n
o
wn
as
s
tack
e
d
g
en
e
r
aliza
tio
n
,
m
ix
es
th
e
g
u
ess
es
f
r
o
m
m
an
y
d
if
f
er
e
n
t
b
as
ic
m
o
d
els
u
s
in
g
a
s
p
ec
ial
lear
n
er
.
T
h
ese
b
asic
m
o
d
els
ca
n
b
e
v
a
r
io
u
s
k
in
d
s
o
f
co
m
p
u
ter
m
eth
o
d
s
tr
ain
ed
o
n
th
e
s
am
e
d
ata.
Fo
r
ex
am
p
le,
f
o
r
p
r
e
d
ictin
g
s
tr
o
k
es,
s
tack
in
g
m
ig
h
t
u
s
e
d
if
f
er
en
t
m
o
d
els
lik
e
DT
,
L
R
,
an
d
SVM
o
n
th
e
s
tr
o
k
e
d
ata.
T
h
en
,
a
n
o
th
er
s
p
e
cial
lear
n
er
(
lik
e
LR
o
r
an
o
th
er
co
m
p
u
t
er
m
eth
o
d
)
is
tr
ain
ed
o
n
th
e
g
u
ess
es
o
f
th
ese
b
asic m
o
d
els to
m
ak
e
th
e
f
in
al
g
u
ess
[
2
2
]
.
̂
f
s
ta
c
k
in
g
(
)
=
(
∑
=
1
(
)
)
(
4
)
−
̂
f
s
ta
c
k
in
g
(
)
is
th
e
en
s
em
b
le
p
r
ed
ictio
n
.
−
is
th
e
m
eta
-
lear
n
er
.
−
is
th
e
weig
h
t a
s
s
ig
n
ed
to
th
e
-
th
b
ase
lear
n
er
.
−
(
)
is
th
e
p
r
ed
ictio
n
f
r
o
m
t
h
e
-
th
b
ase
lear
n
er
.
I
n
th
e
d
iag
r
am
ab
o
v
e
Fig
u
r
e
7
,
we
s
ee
o
n
e
lev
el
o
f
s
tack
in
g
.
Ho
wev
er
,
t
h
er
e
ar
e
also
m
o
r
e
co
m
p
lex
s
tack
in
g
m
eth
o
d
s
with
m
u
ltip
le
lay
er
s
o
f
class
if
ier
s
ad
d
ed
in
b
etwe
en
.
Fig
u
r
e
7
.
I
ll
u
s
tr
ates h
o
w
th
e
s
tack
in
g
en
s
em
b
le
m
eth
o
d
wo
r
k
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
B
o
o
s
tin
g
s
tr
o
ke
p
r
ed
ictio
n
w
ith
en
s
emb
le
lea
r
n
in
g
o
n
imb
a
l
a
n
ce
d
h
ea
lth
ca
r
e
d
a
ta
(
Ou
tma
n
e
La
b
a
yb
i
)
1143
2
.
5
.
P
er
f
o
r
m
a
nce
ev
a
lua
t
io
n m
et
rics
T
o
s
ee
h
o
w
well
a
class
if
icatio
n
alg
o
r
ith
m
is
d
o
i
n
g
,
we
u
s
e
d
if
f
er
en
t
m
eth
o
d
s
.
On
e
o
f
th
e
m
is
ca
lled
th
e
co
n
f
u
s
io
n
m
atr
ix
s
h
o
wn
in
F
ig
u
r
e
8
.
I
t
’
s
is
a
tab
le
th
at
s
h
o
ws
h
o
w
well
a
s
u
p
er
v
is
ed
lear
n
in
g
alg
o
r
ith
m
is
d
o
in
g
.
E
ac
h
r
o
w
r
ep
r
esen
ts
th
e
ac
tu
al
in
s
tan
ce
s
o
f
a
class
,
an
d
ea
c
h
co
l
u
m
n
r
ep
r
esen
t
th
e
p
r
ed
icted
i
n
s
tan
ce
s
o
f
a
class
.
Fro
m
th
is
tab
le,
we
ca
n
ca
lcu
late
all
th
e
m
etr
ics to
ev
alu
ate
th
e
p
e
r
f
o
r
m
an
ce
o
f
t
h
e
alg
o
r
ith
m
[
2
3
]
.
Fig
u
r
e
8
.
C
o
n
f
u
s
io
n
m
atr
i
x
Acc
u
r
ac
y
: th
e
p
r
o
p
o
r
tio
n
o
f
c
o
r
r
ec
t p
r
e
d
ictio
n
s
m
ad
e
b
y
th
e
class
if
ier
.
=
(
+
)
(
+
+
+
)
(
5
)
R
ec
all
: k
n
o
wn
as
th
e
tr
u
e
n
e
g
ativ
e
r
ate,
is
ca
lcu
lated
b
y
u
s
in
g
(
6
)
.
=
(
+
)
(
6
)
Pre
cisi
o
n
in
d
icate
s
th
e
p
r
o
p
o
r
t
io
n
o
f
p
o
s
itiv
e
p
r
ed
ictio
n
s
th
at
ar
e
ac
tu
ally
co
r
r
ec
t.
I
t
m
ea
s
u
r
es
h
o
w
ac
cu
r
ately
th
e
class
if
ier
id
en
tifie
s
p
o
s
itiv
e
ca
s
es.
=
(
+
)
(
7
)
T
h
e
F1
-
s
co
r
e
as
s
h
o
wn
in
(
8
)
,
it
is
ca
lcu
lated
as
th
e
tr
u
e
p
o
s
itiv
e
d
iv
id
ed
b
y
th
e
s
u
m
o
f
tr
u
e
p
o
s
itiv
e
an
d
o
n
e
-
h
alf
o
f
t
h
e
s
u
m
o
f
f
alse p
o
s
itiv
e
an
d
f
alse n
e
g
ativ
e.
F1
−
Score
=
(
+
1
2
(
+
)
)
(
8
)
R
ec
eiv
er
o
p
er
atin
g
ch
ar
ac
ter
i
s
tic
(
R
OC
)
cu
r
v
es
ar
e
a
R
O
C
cu
r
v
e,
is
a
g
r
ap
h
ical
p
lo
t
wh
ich
illu
s
tr
ates
th
e
p
er
f
o
r
m
an
ce
o
f
a
b
i
n
ar
y
class
if
icatio
n
alg
o
r
ith
m
as a
f
u
n
ctio
n
o
f
tr
u
e
p
o
s
itiv
e
r
ate
a
n
d
f
als
e
p
o
s
itiv
e
r
ate.
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
is
s
tu
d
y
in
tr
o
d
u
ce
d
a
n
e
w
m
eth
o
d
to
ev
alu
ate
th
e
e
f
f
ec
tiv
en
ess
o
f
f
o
u
r
ML
class
if
icatio
n
alg
o
r
ith
m
s
,
alo
n
g
with
o
n
e
h
y
b
r
id
m
o
d
el,
in
p
r
e
d
ictin
g
s
tr
o
k
e
.
W
e
ass
e
s
s
ed
th
e
p
er
f
o
r
m
a
n
ce
o
f
ea
ch
m
o
d
el
b
ased
o
n
f
iv
e
k
ey
c
r
iter
ia:
s
p
ec
if
icity
,
r
ec
all,
p
r
ec
is
io
n
,
F1
-
s
co
r
e,
an
d
ar
ea
u
n
d
er
th
e
cu
r
v
e
(
AUC).
T
h
e
p
er
f
o
r
m
an
ce
ev
al
u
atio
n
cr
iter
i
a
f
o
r
th
e
d
if
f
er
e
n
t
ML
a
lg
o
r
ith
m
s
ar
e
p
r
esen
ted
in
T
ab
le
1
.
T
ab
le
1
.
Mo
d
el
co
m
p
ar
is
o
n
f
o
r
m
u
ltip
le
m
etr
ics
M
o
d
e
l
F1
-
sc
o
r
e
A
c
c
u
r
a
c
y
R
e
c
a
l
l
P
r
e
c
i
s
i
o
n
R
O
C
A
U
C
RF
9
8
.
9
%
9
8
.
9
0
%
1
0
0
.
0
0
%
9
7
.
8
%
9
8
.
8
%
X
G
B
o
o
st
9
6
.
9
0
%
9
6
.
8
0
%
1
0
0
.
0
0
%
9
4
.
0
0
%
9
6
.
7
0
%
S
V
M
9
0
.
4
0
%
8
9
.
7
0
%
9
4
.
8
8
6
.
4
0
%
8
9
.
6
0
%
LR
7
7
.
3
0
%
7
6
.
2
0
%
7
9
.
4
0
%
7
5
.
3
0
%
7
6
.
1
0
%
O
u
r
p
r
o
p
o
s
e
d
h
y
b
r
i
d
m
o
d
e
l
u
s
i
n
g
M
L,
S
t
a
k
i
n
g
,
a
n
d
v
o
t
i
n
g
9
9
.
7
0
%
9
9
.
7
0
%
1
0
0
%
9
9
.
5
0
%
9
9
.
7
0
%
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
8
,
No
.
2
,
May
20
2
5
:
1
1
3
7
-
1
1
4
8
1144
T
h
e
R
F
m
o
d
el
ac
h
iev
ed
an
F1
-
s
co
r
e
a
n
d
ac
c
u
r
ac
y
o
f
9
8
.
9
0
%,
with
a
R
OC
AUC
o
f
9
8
.
8
%,
in
d
icatin
g
e
x
ce
llen
t
p
e
r
f
o
r
m
a
n
ce
.
XGBo
o
s
t
f
o
llo
we
d
clo
s
e
ly
with
a
n
F1
-
s
co
r
e
o
f
9
6
.
9
0
%
an
d
ac
c
u
r
ac
y
o
f
9
6
.
8
0
%.
T
h
e
SVM
m
o
d
el
s
h
o
wed
s
o
lid
r
ec
all
b
u
t
lo
wer
p
r
e
cisi
o
n
,
with
a
R
OC
AUC
o
f
8
9
.
6
0
%.
LR
h
ad
an
F1
-
s
co
r
e
o
f
7
7
.
3
0
%,
d
em
o
n
s
tr
atin
g
r
ea
s
o
n
ab
le
p
er
f
o
r
m
an
ce
b
u
t
less
ef
f
ec
tiv
en
ess
c
o
m
p
ar
ed
to
th
e
m
o
r
e
co
m
p
lex
m
o
d
els.
T
h
e
p
r
o
p
o
s
ed
h
y
b
r
id
m
o
d
el,
u
tili
zin
g
M
L
,
s
tack
in
g
,
an
d
v
o
tin
g
tech
n
i
q
u
es,
o
u
tp
e
r
f
o
r
m
ed
all
o
th
er
s
,
ac
h
iev
in
g
an
F1
-
s
c
o
r
e
an
d
ac
cu
r
ac
y
o
f
9
9
.
7
0
%,
alo
n
g
with
a
R
OC
AUC
o
f
9
9
.
7
0
%,
in
d
icatin
g
s
u
p
er
io
r
p
e
r
f
o
r
m
an
ce
.
To
v
is
u
alize
th
ese
r
esu
lts
,
we
cr
ea
te
a
b
ar
p
lo
t
f
o
r
co
m
p
ar
in
g
th
e
F1
-
s
co
r
e,
ac
cu
r
ac
y
,
r
ec
all,
p
r
ec
is
io
n
,
an
d
R
OC
AUC
f
o
r
ea
ch
m
o
d
el
,
th
is
b
ar
p
lo
t in
F
i
g
u
r
e
9
r
e
p
r
esen
ts
th
e
p
er
f
o
r
m
a
n
ce
m
etr
ics o
f
ea
ch
m
o
d
el,
en
a
b
lin
g
a
clea
r
co
m
p
ar
is
o
n
o
f
th
ei
r
s
tr
en
g
th
s
an
d
wea
k
n
ess
es
,
i
n
Fig
u
r
e
1
0
,
we
p
r
esen
ted
th
e
R
OC
cu
r
v
e
co
m
p
ar
in
g
th
e
p
e
r
f
o
r
m
a
n
ce
o
f
d
if
f
er
e
n
t
m
o
d
els.
It
’
s
clea
r
f
r
o
m
th
e
p
lo
t
th
at
th
e
p
r
o
p
o
s
ed
h
y
b
r
id
m
o
d
e
l
o
u
tp
er
f
o
r
m
s
th
e
o
th
er
m
o
d
els
ac
r
o
s
s
all
m
etr
ics,
in
d
icatin
g
its
s
u
p
er
io
r
ity
in
class
if
icatio
n
task
s
.
Fig
u
r
e
9
.
C
o
m
p
a
r
is
o
n
o
f
m
o
d
el
p
er
f
o
r
m
an
ce
m
etr
ics
Fig
u
r
e
10
.
R
OC
c
u
r
v
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
B
o
o
s
tin
g
s
tr
o
ke
p
r
ed
ictio
n
w
ith
en
s
emb
le
lea
r
n
in
g
o
n
imb
a
l
a
n
ce
d
h
ea
lth
ca
r
e
d
a
ta
(
Ou
tma
n
e
La
b
a
yb
i
)
1145
I
n
o
u
r
s
tu
d
y
,
Fig
u
r
e
1
1
a
n
d
T
a
b
le
2
d
is
cu
s
s
es
an
d
co
m
p
a
r
es
o
u
r
p
r
o
p
o
s
ed
m
o
d
el
with
s
ev
e
r
al
r
elate
d
s
tu
d
ies
in
th
e
d
o
m
ain
o
f
s
tr
o
k
e
p
r
ed
ictio
n
.
Firstl
y
,
Alam
o
u
d
i
an
d
A
b
d
allah
[
5
]
u
tili
ze
d
LR
,
DT
,
R
F,
KNN,
SVM,
an
d
NB
class
if
ier
s
,
ac
h
iev
in
g
a
n
ac
cu
r
ac
y
o
f
8
2
%
o
n
a
s
tr
o
k
e
p
r
e
d
ictio
n
d
ataset.
Similar
ly
,
J
av
ale
a
n
d
De
s
ai
[
1
9
]
em
p
lo
y
ed
R
F,
KNN,
an
d
L
R
,
ac
h
iev
in
g
a
h
ig
h
e
r
ac
cu
r
ac
y
o
f
9
3
.
3
2
%.
B
r
eim
an
et
a
l
.
[
2
0
]
u
tili
ze
d
L
R
,
DT
,
v
o
tin
g
,
a
n
d
R
F,
at
tain
in
g
an
im
p
r
ess
iv
e
ac
c
u
r
a
cy
o
f
9
6
%.
Sch
ap
ir
e
et
a
l
.
[
2
1
]
ex
p
lo
r
ed
R
F,
XGBo
o
s
t,
an
d
L
ig
h
tGB
M
,
ac
h
iev
in
g
a
h
i
g
h
ac
cu
r
ac
y
o
f
9
6
.
3
4
%.
Ad
d
itio
n
ally
,
D
r
u
ck
er
et
a
l
.
[
2
2
]
ex
p
er
im
en
ted
with
L
R
,
SVM,
ar
tific
ial
n
eu
r
al
n
etwo
r
k
s
(
ANN)
,
XGBo
o
s
t,
an
d
R
F,
o
b
tain
in
g
a
n
o
tab
le
ac
cu
r
ac
y
o
f
9
7
%.
I
n
c
o
m
p
a
r
is
o
n
to
th
ese
s
tu
d
ies,
o
u
r
f
ir
s
t
m
o
d
el,
wh
ich
u
s
ed
th
e
b
ase
ML
tech
n
iq
u
es
lik
e
SVM,
XGBo
o
s
t,
an
d
R
F,
ac
h
iev
ed
an
ac
cu
r
ac
y
o
f
9
4
.
9
0
%.
No
tab
ly
,
th
is
ac
cu
r
ac
y
s
u
r
p
ass
es th
at
o
f
Alam
o
u
d
i
an
d
Ab
d
alla
[
5
]
,
J
av
ale
an
d
D
esai
[
1
9
]
,
alth
o
u
g
h
s
lig
h
tly
lo
wer
th
an
th
at
ac
h
iev
ed
b
y
B
r
e
im
an
[
2
0
]
,
Sch
ap
i
r
e
[
2
1
]
,
a
n
d
Dr
u
ck
er
et
a
l
.
[
2
2
]
,
[2
3
].
T
ab
le
2
.
co
m
p
ar
is
o
n
s
tr
o
k
e
p
r
ed
ictio
n
with
r
elate
d
s
tu
d
ies
Ref
M
e
t
h
o
d
o
l
o
g
y
A
c
c
u
r
a
c
y
D
a
t
a
s
e
t
S
a
i
l
a
sy
a
a
n
d
K
u
mar
i
[
9
]
LR
,
S
V
M
,
D
T,
K
N
N
,
R
F
,
a
n
d
NB
w
a
s t
h
e
b
e
s
t
8
2
%
S
t
r
o
k
e
p
r
e
d
i
c
t
i
o
n
d
a
t
a
s
e
t
B
a
d
r
i
y
a
h
et
al
.
[
2
4
]
R
F
,
K
N
N
,
LR
,
R
F
w
h
i
c
h
i
s t
h
e
b
e
st
9
3
.
3
2
%
S
t
r
o
k
e
p
r
e
d
i
c
t
i
o
n
d
a
t
a
s
e
t
Ta
z
i
n
,
e
t
a
l
.
[
2
5
]
D
T,
L
R
,
V
o
t
i
n
g
,
a
n
d
R
F
w
a
s
t
h
e
b
e
s
t
.
9
6
%
S
t
r
o
k
e
p
r
e
d
i
c
t
i
o
n
d
a
t
a
s
e
t
A
l
r
u
i
l
y
e
t
a
l
.
[
2
6
]
R
F
,
L
i
g
h
t
G
B
M
,
a
n
d
X
G
B
o
o
s
t
9
6
.
3
4
%
S
t
r
o
k
e
p
r
e
d
i
c
t
i
o
n
d
a
t
a
s
e
t
A
l
h
a
k
a
mi
e
t
a
l
.
[
2
7
]
LR
,
S
V
M
,
A
N
N
,
X
G
B
o
o
s
t
,
a
n
d
R
F
,
w
h
i
c
h
w
a
s
t
h
e
b
e
s
t
9
7
%
S
t
r
o
k
e
p
r
e
d
i
c
t
i
o
n
d
a
t
a
s
e
t
p
r
o
p
o
se
d
m
o
d
e
l
u
s
i
n
g
b
a
s
e
M
L
S
V
M
,
X
G
B
o
o
st
,
R
F
w
i
c
h
w
a
s
t
h
e
b
e
s
t
9
4
.
9
0
%
S
t
r
o
k
e
p
r
e
d
i
c
t
i
o
n
d
a
t
a
s
e
t
P
r
o
p
o
se
d
m
o
d
e
l
u
si
n
g
h
y
b
r
i
d
M
L
,
e
n
se
mb
l
e
l
e
a
r
n
i
n
g
R
F
,
S
V
M
,
X
G
B
o
o
st
,
s
t
a
k
i
n
g
,
a
n
d
v
o
t
i
n
g
9
9
.
7
4
%
S
t
r
o
k
e
p
r
e
d
i
c
t
i
o
n
d
a
t
a
s
e
t
Fig
u
r
e
11
.
C
o
m
p
ar
is
o
n
o
f
s
tr
o
k
e
p
r
ed
ictio
n
s
tu
d
ies
Ho
wev
er
,
th
e
im
p
o
r
tan
t
co
n
tr
ib
u
tio
n
o
f
o
u
r
s
tu
d
y
lies
in
th
e
d
ev
elo
p
m
en
t
o
f
a
n
o
v
el
h
y
b
r
id
ML
ap
p
r
o
ac
h
,
u
tili
zin
g
en
s
em
b
le
l
ea
r
n
in
g
tec
h
n
iq
u
es
s
u
c
h
as
s
tack
in
g
an
d
v
o
tin
g
.
T
h
is
h
y
b
r
id
m
o
d
el,
c
o
m
b
in
i
n
g
R
F,
SVM,
XG
B
o
o
s
t,
s
tack
in
g
,
an
d
v
o
tin
g
,
r
em
a
r
k
ab
l
y
ac
h
i
ev
ed
an
ac
cu
r
ac
y
o
f
9
9
.
7
4
%,
o
u
tp
er
f
o
r
m
in
g
all
p
r
ev
io
u
s
s
tu
d
ies.
T
h
e
in
teg
r
atio
n
o
f
e
n
s
em
b
le
lear
n
in
g
tec
h
n
iq
u
es h
elp
ed
u
s
m
ak
e
th
e
m
o
s
t o
f
th
e
s
tr
en
g
th
s
o
f
ea
ch
m
o
d
el
wh
ile
co
m
p
en
s
at
in
g
f
o
r
t
h
eir
lim
itatio
n
s
,
r
esu
ltin
g
in
a
h
ig
h
l
y
ac
cu
r
ate
p
r
ed
ictiv
e
m
o
d
el
f
o
r
s
tr
o
k
e
o
cc
u
r
r
en
ce
.
Ou
r
p
r
o
p
o
s
ed
h
y
b
r
i
d
m
o
d
el,
wh
ich
co
m
b
i
n
es
ML
tech
n
iq
u
es
s
u
ch
as
s
tack
in
g
an
d
v
o
tin
g
,
r
ep
r
esen
ts
a
s
ig
n
if
ican
t
ad
v
an
ce
m
en
t
in
s
tr
o
k
e
p
r
ed
ictio
n
co
m
p
a
r
ed
t
o
p
r
ev
io
u
s
s
tu
d
ies.
B
y
u
tili
zin
g
th
e
s
tr
en
g
th
s
o
f
m
u
ltip
le
b
ase
m
o
d
els
o
f
ML
a
n
d
o
u
r
ap
p
r
o
ac
h
,
we
ac
h
iev
ed
a
r
em
a
r
k
ab
le
ac
cu
r
ac
y
o
f
9
9
.
7
4
%,
s
u
r
p
ass
in
g
all
p
r
ev
io
u
s
s
tu
d
ies in
s
tr
o
k
e
p
r
e
d
ictio
n
ac
cu
r
ac
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
8
,
No
.
2
,
May
20
2
5
:
1
1
3
7
-
1
1
4
8
1146
4.
C
O
NCLU
SI
O
N
I
n
th
is
s
tu
d
y
,
we
’
v
e
ex
p
lo
r
e
d
h
o
w
ML
an
d
en
s
em
b
le
lear
n
in
g
alg
o
r
ith
m
s
ca
n
b
e
co
m
b
in
ed
f
o
r
p
r
ed
ictin
g
s
tr
o
k
es,
r
esu
ltin
g
i
n
th
e
d
ev
elo
p
m
en
t
o
f
a
n
ew
h
y
b
r
id
m
o
d
el.
Usi
n
g
a
lar
g
e
s
tr
o
k
e
d
ataset,
we
’
v
e
s
h
o
wn
th
at
o
u
r
ap
p
r
o
ac
h
is
ef
f
ec
tiv
e
in
id
en
tify
in
g
in
d
iv
i
d
u
als
at
r
is
k
o
f
s
tr
o
k
e
ac
cu
r
ately
b
y
co
m
b
in
in
g
d
if
f
er
en
t
ML
m
o
d
els
an
d
lev
er
ag
in
g
th
eir
u
n
iq
u
e
s
tr
en
g
th
s
.
Ho
wev
er
,
o
u
r
h
y
b
r
i
d
m
o
d
el
a
ch
iev
es
ex
ce
p
tio
n
al
p
r
ed
ictiv
e
ac
c
u
r
ac
y
,
s
u
r
p
ass
in
g
p
r
e
v
io
u
s
b
en
ch
m
ar
k
s
i
n
s
tr
o
k
e
p
r
ed
ictio
n
.
Ad
d
itio
n
ally
,
e
n
s
em
b
le
lear
n
in
g
is
ess
en
tial
in
o
u
r
h
y
b
r
id
m
o
d
e
l
.
Me
th
o
d
s
lik
e
s
tack
in
g
an
d
v
o
tin
g
h
elp
u
s
m
er
g
e
i
n
s
ig
h
ts
f
r
o
m
d
if
f
er
en
t
m
o
d
els,
wh
ich
h
elp
s
o
v
er
co
m
e
th
e
lim
itatio
n
s
o
f
in
d
iv
i
d
u
al
alg
o
r
ith
m
s
an
d
en
h
a
n
ce
s
o
v
er
all
p
r
ed
ictiv
e
ac
cu
r
ac
y
.
B
y
u
s
in
g
e
n
s
em
b
le
lear
n
in
g
,
we
m
ax
i
m
ize
th
e
p
o
ten
tial
o
f
o
u
r
p
r
e
d
ictiv
e
m
o
d
el,
o
f
f
er
i
n
g
s
tr
o
n
g
an
d
d
ep
e
n
d
ab
le
s
tr
o
k
e
r
is
k
ass
es
s
m
en
ts
f
o
r
clin
ical
d
ec
is
io
n
-
m
ak
in
g
.
Fin
ally
,
o
u
r
s
tu
d
y
p
r
esen
ts
an
in
n
o
v
ativ
e
way
t
o
p
r
ed
ict
s
tr
o
k
es
b
y
co
m
b
in
in
g
ML
a
n
d
e
n
s
em
b
le
lear
n
in
g
tech
n
iq
u
es.
W
e
f
o
u
n
d
th
at
o
u
r
h
y
b
r
id
m
o
d
el
p
er
f
o
r
m
s
b
etter
t
h
an
p
r
ev
io
u
s
m
et
h
o
d
s
,
o
f
f
er
in
g
p
r
ec
is
e
r
is
k
ass
ess
m
en
ts
.
E
n
s
em
b
le
lear
n
in
g
g
r
ea
tly
im
p
r
o
v
es
th
e
m
o
d
el
’
s
p
er
f
o
r
m
an
ce
,
s
h
o
win
g
h
o
w
im
p
o
r
tan
t
it
is
to
m
er
g
e
d
if
f
er
en
t
alg
o
r
ith
m
s
f
o
r
b
etter
ac
cu
r
ac
y
.
I
n
f
u
tu
r
e
wo
r
k
,
we
p
lan
to
im
p
lem
en
t
ass
o
ciatio
n
r
u
les
to
f
u
r
th
er
ju
s
tify
o
u
r
r
esu
lts
an
d
e
x
p
lo
r
e
d
ee
p
lear
n
in
g
m
o
d
els
f
o
r
im
ag
e
an
aly
s
is
to
p
r
ed
ict
s
tr
o
k
es
with
en
h
an
ce
d
p
er
f
o
r
m
an
ce
.
T
h
is
co
u
ld
s
ig
n
if
ican
tly
ad
v
an
ce
th
e
u
s
e
o
f
p
r
ed
icti
v
e
an
aly
tics
in
clin
ical
s
etti
n
g
s
,
en
a
b
lin
g
h
ea
lth
ca
r
e
p
r
o
f
ess
io
n
als
to
m
ak
e
in
f
o
r
m
e
d
d
ec
is
io
n
s
b
ased
o
n
r
o
b
u
s
t d
ata
ACK
NO
WL
E
DG
E
M
E
NT
S
T
h
er
e
ar
e
n
o
s
ig
n
if
ican
t
f
in
d
i
n
g
s
to
r
ep
o
r
t.
W
e
ar
e
s
h
ar
in
g
th
is
ar
ticle
s
o
lely
f
o
r
th
e
b
en
ef
it
o
f
th
e
s
cien
tific
co
m
m
u
n
ity
.
RE
F
E
R
E
NC
E
S
[
1
]
V
.
L.
F
e
i
g
i
n
e
t
a
l
.
,
“
G
l
o
b
a
l
,
r
e
g
i
o
n
a
l
,
a
n
d
n
a
t
i
o
n
a
l
b
u
r
d
e
n
o
f
st
r
o
k
e
a
n
d
i
t
s
r
i
sk
f
a
c
t
o
r
s,
1
9
9
0
-
2
0
1
9
:
a
s
y
st
e
ma
t
i
c
a
n
a
l
y
s
i
s
f
o
r
t
h
e
g
l
o
b
a
l
b
u
r
d
e
n
o
f
d
i
s
e
a
s
e
st
u
d
y
2
0
1
9
,
”
T
h
e
L
a
n
c
e
t
N
e
u
r
o
l
o
g
y
,
v
o
l
.
2
0
,
n
o
.
1
0
,
p
p
.
7
9
5
–
8
2
0
,
O
c
t
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
S
1
4
7
4
-
4
4
2
2
(
2
1
)
0
0
2
5
2
-
0.
[
2
]
V
.
R
a
j
i
n
i
k
a
n
t
h
a
n
d
S
.
C
.
S
a
t
a
p
a
t
h
y
,
“
S
e
g
m
e
n
t
a
t
i
o
n
o
f
i
sc
h
e
m
i
c
st
r
o
k
e
l
e
si
o
n
i
n
b
r
a
i
n
M
R
I
b
a
se
d
o
n
s
o
c
i
a
l
g
r
o
u
p
o
p
t
i
m
i
z
a
t
i
o
n
a
n
d
f
u
z
z
y
-
t
s
a
l
l
i
s
e
n
t
r
o
p
y
,
”
Ar
a
b
i
a
n
J
o
u
rn
a
l
f
o
r
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
r
i
n
g
,
v
o
l
.
4
3
,
n
o
.
8
,
p
p
.
4
3
6
5
–
4
3
7
8
,
A
u
g
.
2
0
1
8
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
3
369
-
017
-
3
0
5
3
-
6.
[
3
]
S
.
V
.
S
h
e
t
t
y
,
H
.
S
a
r
o
j
a
d
e
v
i
,
S
.
A
n
k
a
l
a
k
i
,
C
.
D
e
d
e
e
p
y
a
,
S
.
S
h
r
e
e
r
a
k
s
h
a
,
a
n
d
N
.
G
a
n
a
v
i
,
“
E
a
r
l
y
d
e
t
e
c
t
i
o
n
o
f
s
t
r
o
k
e
u
s
i
n
g
M
R
I
i
ma
g
e
s
-
A
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
,
”
i
n
AI
P
C
o
n
f
e
r
e
n
c
e
Pr
o
c
e
e
d
i
n
g
s
,
2
0
2
4
,
v
o
l
.
3
1
2
2
,
n
o
.
1
,
p
.
0
8
0
0
2
0
,
d
o
i
:
1
0
.
1
0
6
3
/
5
.
0
2
1
6
0
7
6
.
[
4
]
M
.
R
u
d
i
a
n
s
y
a
h
,
T
.
A
.
S
a
r
d
j
o
n
o
,
a
n
d
R
.
M
a
r
d
i
y
a
n
t
o
,
“
S
e
g
m
e
n
t
a
t
i
o
n
o
f
t
h
e
i
n
t
r
a
c
e
r
e
b
r
a
l
h
e
m
o
r
r
h
a
g
i
c
st
r
o
k
e
s
(
B
l
e
e
d
s)
f
r
o
m
b
r
a
i
n
C
T
i
ma
g
e
b
a
se
d
o
n
G
V
F
s
n
a
k
e
,
”
i
n
Pro
c
e
e
d
i
n
g
-
2
0
1
8
I
n
t
e
r
n
a
t
i
o
n
a
l
S
e
m
i
n
a
r
o
n
I
n
t
e
l
l
i
g
e
n
t
T
e
c
h
n
o
l
o
g
y
a
n
d
I
t
s
A
p
p
l
i
c
a
t
i
o
n
,
I
S
I
T
I
A
2
0
1
8
,
A
u
g
.
2
0
1
8
,
p
p
.
4
6
5
–
4
7
0
.
d
o
i
:
1
0
.
1
1
0
9
/
I
S
I
T
I
A
.
2
0
1
8
.
8
7
1
1
1
5
5
.
[
5
]
A
.
A
l
a
m
o
u
d
i
a
n
d
Y
.
A
b
d
a
l
l
a
h
,
“
C
h
a
r
a
c
t
e
r
i
z
a
t
i
o
n
o
f
b
r
a
i
n
st
r
o
k
e
u
si
n
g
i
m
a
g
e
a
n
d
si
g
n
a
l
p
r
o
c
e
ssi
n
g
t
e
c
h
n
i
q
u
e
s
,
”
i
n
Bi
o
m
e
d
i
c
a
l
S
i
g
n
a
l
a
n
d
I
m
a
g
e
Pr
o
c
e
ssi
n
g
,
I
n
t
e
c
h
O
p
e
n
,
2
0
2
1
.
d
o
i
:
1
0
.
5
7
7
2
/
i
n
t
e
c
h
o
p
e
n
.
9
6
2
8
8
.
[
6
]
W
.
S
.
A
l
a
z
a
w
e
e
,
Z
.
H
.
N
a
j
i
,
a
n
d
W
.
T.
A
l
i
,
“
A
n
a
l
y
z
i
n
g
a
n
d
d
e
t
e
c
t
i
n
g
h
e
m
o
r
r
h
a
g
i
c
a
n
d
i
s
c
h
e
mi
c
s
t
r
o
k
e
-
b
a
s
e
d
o
n
b
i
t
p
l
a
n
e
sl
i
c
i
n
g
a
n
d
e
d
g
e
d
e
t
e
c
t
i
o
n
a
l
g
o
r
i
t
h
m
s,”
I
n
d
o
n
e
s
i
a
n
J
o
u
r
n
a
l
o
f
E
l
e
c
t
ri
c
a
l
En
g
i
n
e
e
ri
n
g
a
n
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
(
I
J
E
EC
S
)
,
v
o
l
.
2
5
,
n
o
.
2
,
p
p
.
1
0
0
3
–
1
0
1
0
,
F
e
b
.
2
0
2
2
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
e
c
s.
v
2
5
.
i
2
.
p
p
1
0
0
3
-
1
0
1
0
.
[
7
]
N
.
S
.
I
.
M
.
R
a
f
e
i
,
R
.
H
a
ss
a
n
,
R
.
D
.
R
.
S
a
e
d
u
d
i
n
,
A
.
F
.
M
.
R
a
f
f
e
i
,
Z.
Z
a
k
a
r
i
a
,
a
n
d
S
.
K
a
s
i
m,
“
C
o
m
p
a
r
i
so
n
o
f
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
t
e
c
h
n
i
q
u
e
s
i
n
c
l
a
ss
i
f
y
i
n
g
st
r
o
k
e
d
o
c
u
me
n
t
s
,
”
I
n
d
o
n
e
si
a
n
J
o
u
rn
a
l
o
f
El
e
c
t
ri
c
a
l
En
g
i
n
e
e
r
i
n
g
a
n
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
(
I
J
EEC
S
)
,
v
o
l
.
1
4
,
n
o
.
3
,
p
p
.
1
2
4
4
–
1
2
5
0
,
J
u
n
.
2
0
1
9
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
e
c
s
.
v
1
4
.
i
3
.
p
p
1
2
4
4
-
1
2
5
0
.
[
8
]
H
.
M
a
h
e
s
h
w
a
r
i
,
D
.
Y
a
d
a
v
,
a
n
d
U
.
C
h
a
n
d
r
a
,
“
B
r
a
i
n
st
r
o
k
e
p
r
e
d
i
c
t
i
o
n
u
s
i
n
g
t
h
e
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
,
”
i
n
C
o
m
m
u
n
i
c
a
t
i
o
n
s
i
n
C
o
m
p
u
t
e
r
a
n
d
I
n
f
o
rm
a
t
i
o
n
S
c
i
e
n
c
e
,
v
o
l
.
1
7
4
2
C
C
I
S
,
2
0
2
2
,
p
p
.
1
–
1
1
.
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
0
3
1
-
2
3
6
4
7
-
1
_
1
.
[
9
]
G
.
S
a
i
l
a
s
y
a
a
n
d
G
.
L
.
A
.
K
u
m
a
r
i
,
“
A
n
a
l
y
z
i
n
g
t
h
e
p
e
r
f
o
r
ma
n
c
e
o
f
s
t
r
o
k
e
p
r
e
d
i
c
t
i
o
n
u
s
i
n
g
M
L
c
l
a
ss
i
f
i
c
a
t
i
o
n
a
l
g
o
r
i
t
h
ms
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
Ad
v
a
n
c
e
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
1
2
,
n
o
.
6
,
p
p
.
5
3
9
–
5
4
5
,
2
0
2
1
,
d
o
i
:
1
0
.
1
4
5
6
9
/
I
JA
C
S
A
.
2
0
2
1
.
0
1
2
0
6
6
2
.
[
1
0
]
C
.
S
.
N
w
o
s
u
,
S
.
D
e
v
,
P
.
B
h
a
r
d
w
a
j
,
B
.
V
e
e
r
a
v
a
l
l
i
,
a
n
d
D
.
J
o
h
n
,
“
P
r
e
d
i
c
t
i
n
g
st
r
o
k
e
f
r
o
m e
l
e
c
t
r
o
n
i
c
h
e
a
l
t
h
r
e
c
o
r
d
s
,
”
i
n
Pro
c
e
e
d
i
n
g
s
o
f
t
h
e
An
n
u
a
l
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
f
t
h
e
I
EEE
E
n
g
i
n
e
e
r
i
n
g
i
n
M
e
d
i
c
i
n
e
a
n
d
B
i
o
l
o
g
y
S
o
c
i
e
t
y
,
E
M
B
S
,
J
u
l
.
2
0
1
9
,
p
p
.
5
7
0
4
–
5
7
0
7
.
d
o
i
:
1
0
.
1
1
0
9
/
E
M
B
C
.
2
0
1
9
.
8
8
5
7
2
3
4
.
[
11
]
M
.
U
.
Emo
n
,
M
.
S
.
K
e
y
a
,
T.
I
.
M
e
g
h
l
a
,
M
.
M
.
R
a
h
ma
n
,
M
.
S
.
A
l
M
a
m
u
n
,
a
n
d
M
.
S
.
K
a
i
s
e
r
,
“
P
e
r
f
o
r
m
a
n
c
e
a
n
a
l
y
si
s
o
f
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
e
s
i
n
s
t
r
o
k
e
p
r
e
d
i
c
t
i
o
n
,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
4
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
El
e
c
t
ro
n
i
c
s,
C
o
m
m
u
n
i
c
a
t
i
o
n
a
n
d
Ae
ro
s
p
a
c
e
T
e
c
h
n
o
l
o
g
y
,
I
C
E
C
A
2
0
2
0
,
N
o
v
.
2
0
2
0
,
p
p
.
1
4
6
4
–
1
4
6
9
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
EC
A
4
9
3
1
3
.
2
0
2
0
.
9
2
9
7
5
2
5
.
[
1
2
]
E.
D
r
i
t
sas
a
n
d
M
.
Tr
i
g
k
a
,
“
S
t
r
o
k
e
r
i
s
k
p
r
e
d
i
c
t
i
o
n
w
i
t
h
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s,
”
S
e
n
s
o
rs
,
v
o
l
.
2
2
,
n
o
.
1
3
,
p
.
4
6
7
0
,
J
u
n
.
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
s
2
2
1
3
4
6
7
0
.
[
1
3
]
A.
K
.
H
a
m
o
u
d
e
t
a
l
.
,
“
A
p
r
e
d
i
c
t
i
o
n
mo
d
e
l
b
a
se
d
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
ms
w
i
t
h
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
a
p
p
r
o
a
c
h
e
s
o
v
e
r
i
m
b
a
l
a
n
c
e
d
d
a
t
a
se
t
,
”
I
n
d
o
n
e
s
i
a
n
J
o
u
rn
a
l
o
f
E
l
e
c
t
ri
c
a
l
E
n
g
i
n
e
e
r
i
n
g
a
n
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
(
I
J
E
EC
S
)
,
v
o
l
.
2
8
,
n
o
.
2
,
p
p
.
1
1
0
5
–
1
1
1
6
,
N
o
v
.
2
0
2
2
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
e
c
s
.
v
2
8
.
i
2
.
p
p
1
1
0
5
-
1
1
1
6
.
[
1
4
]
T.
A
.
A
sse
g
i
e
,
V
.
El
a
n
a
n
g
a
i
,
J.
S
.
P
a
u
l
r
a
j
,
M
.
V
e
l
mu
r
u
g
a
n
,
a
n
d
D
.
F
.
D
e
v
e
sa
n
,
“
E
v
a
l
u
a
t
i
o
n
o
f
f
e
a
t
u
r
e
s
c
a
l
i
n
g
f
o
r
i
m
p
r
o
v
i
n
g
t
h
e
p
e
r
f
o
r
m
a
n
c
e
o
f
s
u
p
e
r
v
i
se
d
l
e
a
r
n
i
n
g
me
t
h
o
d
s,
”
B
u
l
l
e
t
i
n
o
f
El
e
c
t
ri
c
a
l
E
n
g
i
n
e
e
r
i
n
g
a
n
d
I
n
f
o
rm
a
t
i
c
s
(
BEEI)
,
v
o
l
.
1
2
,
n
o
.
3
,
p
p
.
1
8
3
3
–
1
8
3
8
,
Ju
n
.
2
0
2
3
,
d
o
i
:
1
0
.
1
1
5
9
1
/
e
e
i
.
v
1
2
i
3
.
5
1
7
0
.
[
1
5
]
A
.
R
.
C
h
o
w
d
h
u
r
y
,
T.
C
h
a
t
t
e
r
j
e
e
,
a
n
d
S
.
B
a
n
e
r
j
e
e
,
“
A
r
a
n
d
o
m
f
o
r
e
st
c
l
a
ssi
f
i
e
r
-
b
a
s
e
d
a
p
p
r
o
a
c
h
i
n
t
h
e
d
e
t
e
c
t
i
o
n
o
f
a
b
n
o
r
m
a
l
i
t
i
e
s
i
n
t
h
e
r
e
t
i
n
a
,
”
M
e
d
i
c
a
l
a
n
d
Bi
o
l
o
g
i
c
a
l
E
n
g
i
n
e
e
r
i
n
g
a
n
d
C
o
m
p
u
t
i
n
g
,
v
o
l
.
5
7
,
n
o
.
1
,
p
p
.
1
9
3
–
2
0
3
,
Ja
n
.
2
0
1
9
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
5
1
7
-
018
-
1
8
7
8
-
0.
Evaluation Warning : The document was created with Spire.PDF for Python.