I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
p
ute
r
E
ng
in
ee
ring
(
I
J
E
CE
)
Vo
l.
10
,
No
.
1
,
Feb
r
u
ar
y
2020
,
p
p
.
1
0
7
9
~
1
0
8
4
I
SS
N:
2088
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v
10
i
1
.
pp
1
0
7
9
-
1
0
8
4
1079
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m/in
d
ex
.
p
h
p
/I
JE
C
E
A new
m
o
del f
o
r
i
ris da
ta set
clas
sif
ica
tion ba
sed o
n l
inea
r
supp
o
rt
v
ector
ma
chine
pa
ra
m
eter
's o
pti
m
i
z
a
tion
Z
a
hra
a
F
a
i
z
H
us
s
a
in
1
,
H
ind
Ra
a
d Ib
ra
hee
m
2
,
M
o
ha
mm
a
d Alsa
j
ri
3
,
Ah
m
e
d H
us
s
e
in
Ali
4
,
M
o
hd
Arf
ia
n Is
m
a
il
5
,
S
ha
hr
ee
n K
a
s
i
m
6
,
T
o
le
Su
t
i
kn
o
7
1,
2,
3,
4
C
o
m
p
u
ter S
c
ien
c
e
De
p
a
rtm
e
n
t
,
A
L
S
a
la
m
Un
iv
e
rsit
y
Co
ll
e
g
e
,
Ira
q
3,
4
De
p
a
rtm
e
n
t
o
f
Co
m
p
u
ter S
c
ien
c
e
,
Co
ll
e
g
e
o
f
Ed
u
c
a
ti
o
n
,
A
l
-
Ira
q
ia Un
iv
e
rsit
y
,
Ira
q
5
F
a
c
u
lt
y
o
f
Co
m
p
u
ter S
y
ste
m
s &
S
o
f
tw
a
r
e
En
g
in
e
e
rin
g
,
Un
iv
e
rsiti
M
a
la
y
sia
P
a
h
a
n
g
,
M
a
lay
sia
6
F
a
c
u
lt
y
o
f
Co
m
p
u
ter
S
c
ien
c
e
&
In
f
o
rm
a
ti
o
n
T
e
c
h
n
o
l
o
g
y
,
Un
iv
e
rsi
ti
T
u
n
Hu
ss
e
in
On
n
M
a
lay
sia
,
M
a
la
y
sia
7
De
p
a
rtme
n
t
o
f
El
e
c
tri
c
a
l
a
n
d
Co
m
p
u
ter E
n
g
in
e
e
rin
g
,
Un
iv
e
rsitas
A
h
m
a
d
Da
h
lan
,
In
d
o
n
e
sia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
A
p
r
9
,
2
0
1
9
R
ev
i
s
ed
Sep
2
9
,
2
0
1
9
A
cc
ep
ted
Oct
1
0
,
2
0
1
9
Da
ta
m
in
in
g
is
k
n
o
w
n
a
s
th
e
p
ro
c
e
ss
o
f
d
e
te
c
ti
o
n
c
o
n
c
e
rn
i
n
g
p
a
tt
e
rn
s
f
ro
m
e
ss
e
n
ti
a
l
a
m
o
u
n
ts
o
f
d
a
ta.
A
s
a
p
ro
c
e
ss
o
f
k
n
o
w
led
g
e
d
isc
o
v
e
r
y
.
Clas
sif
ic
a
ti
o
n
is
a
d
a
ta
a
n
a
ly
sis
th
a
t
e
x
trac
ts
a
m
o
d
e
l
w
h
ich
d
e
sc
rib
e
s
a
n
im
p
o
rtan
t
d
a
ta
c
las
se
s.
On
e
o
f
th
e
o
u
tstan
d
i
n
g
c
las
si
f
ica
ti
o
n
s
m
e
th
o
d
s
in
d
a
ta
m
in
in
g
is
su
p
p
o
r
t
v
e
c
to
r
m
a
c
h
in
e
c
las
sif
ica
ti
o
n
(S
V
M
)
.
It
is
c
a
p
a
b
le
o
f
e
n
v
isa
g
in
g
re
su
lt
s
a
n
d
m
o
stly
e
ff
e
c
ti
v
e
th
a
n
o
t
h
e
r
c
las
sif
ica
ti
o
n
m
e
th
o
d
s.
T
h
e
S
V
M
is
a
o
n
e
tec
h
n
i
q
u
e
o
f
m
a
c
h
in
e
lea
rn
in
g
tec
h
n
i
q
u
e
s
th
a
t
is
w
e
ll
k
n
o
w
n
tec
h
n
i
q
u
e
,
lea
rn
in
g
w
it
h
su
p
e
rv
ise
d
a
n
d
h
a
v
e
b
e
e
n
a
p
p
li
e
d
p
e
rf
e
c
tl
y
to
a
v
a
r
y
p
ro
b
lem
s
o
f
:
r
e
g
re
ss
io
n
,
c
las
sif
ic
a
ti
o
n
,
a
n
d
c
lu
ste
ri
n
g
in
d
iv
e
rse
d
o
m
a
in
s
su
c
h
a
s
g
e
n
e
e
x
p
re
ss
io
n
,
w
e
b
tex
t
m
in
in
g
.
In
t
h
is
stu
d
y
,
w
e
p
ro
p
o
se
d
a
n
e
w
ly
m
o
d
e
f
o
r
c
las
sify
in
g
iri
s
d
a
ta
se
t
u
sin
g
S
VM
c
las
sif
ie
r
a
n
d
g
e
n
e
ti
c
a
lg
o
rit
h
m
to
o
p
ti
m
iz
e
c
a
n
d
g
a
m
m
a
p
a
ra
m
e
ter
s
o
f
li
n
e
a
r
S
V
M
,
in
a
d
d
it
i
o
n
p
ri
n
c
ip
le
c
o
m
p
o
n
e
n
ts
a
n
a
ly
sis
(P
CA
)
a
lg
o
rit
h
m
wa
s
u
se
f
o
r
f
e
a
tu
re
s red
u
c
ti
o
n
.
K
ey
w
o
r
d
s
:
C
las
s
i
f
icatio
n
Data
m
i
n
i
n
g
Gen
etic
al
g
o
r
ith
m
I
r
is
d
ataset
P
ar
am
eter
o
p
ti
m
izat
io
n
SVM
Co
p
y
rig
h
t
©
2
0
2
0
In
stit
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
Al
l
rig
h
ts
re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Mo
h
a
m
m
ad
A
ls
aj
r
i
,
C
o
m
p
u
ter
Scien
ce
Dep
ar
t
m
en
t
,
AL
Sala
m
U
n
i
v
er
s
it
y
C
o
lle
g
e
,
B
A
GHD
AD
-
S
id
i
y
a
h
: N
ea
r
A
l W
alee
d
C
e
n
ter
,
I
r
aq
.
E
m
ail:
m
o
h
a
m
m
ad
.
cs8
8
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
C
las
s
i
f
icatio
n
is
a
m
a
n
n
er
o
f
d
ata
an
al
y
s
i
s
w
h
ic
h
u
s
ed
to
el
icit
a
class
if
ier
to
class
if
y
i
m
p
o
r
tan
t
d
ata
class
es.
T
h
ese
cla
s
s
i
f
ier
s
ca
n
ex
p
ec
t
ca
te
g
o
r
ical
d
ata
(
d
etac
h
ed
,
u
n
o
r
d
er
ed
)
class
lab
el
[
1
]
.
Als
o
cla
s
s
i
f
icatio
n
is
an
i
m
p
o
r
tan
t
f
ield
in
t
h
e
d
ata
m
in
in
g
an
d
m
ac
h
in
e
lear
n
in
g
b
y
co
n
cl
u
d
in
g
o
b
s
cu
r
e
clas
s
es
o
f
s
a
m
p
les
u
s
i
n
g
th
e
lear
n
in
g
o
f
r
en
o
w
n
ed
cla
s
s
es
o
f
s
a
m
p
les
[
2
-
3
]
.
As
an
ex
a
m
p
le,
r
ati
n
g
b
an
k
lo
an
ap
p
licatio
n
ca
n
b
e
class
i
f
ied
as sa
f
et
y
o
r
r
is
k
y
af
t
er
a
co
n
s
tr
u
ctio
n
o
f
clas
s
i
f
icati
o
n
m
o
d
el.
T
h
is
an
al
y
s
i
s
s
u
p
p
lied
u
s
w
it
h
a
b
etter
co
m
p
r
e
h
en
s
io
n
o
f
t
h
e
d
ata
at
b
ig
m
a
g
n
itu
d
e.
Ma
n
y
clas
s
i
f
ic
atio
n
ap
p
r
o
ac
h
es
h
av
e
b
ee
n
s
u
g
g
e
s
ted
in
m
ac
h
i
n
e
lear
n
in
g
,
p
atter
n
r
ec
o
g
n
itio
n
,
an
d
s
tati
s
tics
.
C
las
s
i
f
icatio
n
ca
n
b
e
ac
h
iev
ed
i
n
a
p
r
o
ce
s
s
o
f
t
w
o
-
s
tep
s
.
T
h
e
co
n
s
tr
u
ctio
n
o
f
a
clas
s
i
f
ic
atio
n
b
ased
o
n
p
r
ec
ed
in
g
d
ata
is
ac
h
ie
v
ed
in
t
h
e
f
ir
s
t
s
tag
e.
I
n
th
e
s
ec
o
n
d
s
ta
g
e,
s
p
ec
if
ie
s
i
f
th
e
ac
cu
r
ac
y
o
f
a
m
o
d
el
is
ad
m
i
s
s
ib
le,
an
d
if
s
o
,
w
e
u
t
ilize
th
e
m
o
d
el
to
cla
s
s
i
f
y
f
r
es
h
d
ata
[
4
]
.
Su
p
p
o
r
t
v
ec
to
r
m
ac
h
i
n
e
(
SV
M)
C
lass
i
f
ier
is
a
r
en
o
w
n
ed
class
i
f
icatio
n
m
et
h
o
d
e
m
p
lo
y
ed
f
o
r
p
r
ed
ictin
g
th
e
r
esu
l
ts
o
f
d
atasets
[
5
]
.
T
h
e
p
r
o
p
o
s
ed
m
o
d
el
w
a
s
ass
e
s
s
ed
o
n
an
I
R
I
S
d
ataset
g
ai
n
ed
f
r
o
m
th
e
U
C
I
Ma
ch
i
n
e
L
e
ar
n
i
n
g
Data
b
ase
[
6
]
.
T
h
e
cr
ea
tio
n
o
f
SVM
m
o
d
el
w
it
h
h
i
g
h
p
r
ed
ictio
n
ac
c
u
r
a
c
y
a
n
d
co
n
s
is
te
n
c
y
i
s
b
ased
o
n
s
ee
k
i
n
g
th
e
id
ea
l
p
ar
a
m
eter
s
o
n
SVM
,
s
i
n
ce
it
p
la
y
s
a
n
e
s
s
e
n
tial
r
o
le.
W
ea
k
n
e
s
s
c
lass
if
ica
tio
n
p
er
f
o
r
m
a
n
ce
r
esu
lt
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2088
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
10
,
No
.
1
,
Feb
r
u
ar
y
2
0
2
0
:
10
79
-
1
0
8
4
1080
f
r
o
m
i
n
d
ec
en
t
p
ar
a
m
eter
s
e
tt
in
g
s
,
w
h
ile
th
e
p
er
f
ec
t
ca
teg
o
r
izatio
n
ac
cu
r
ac
y
o
f
SVM
s
te
m
s
f
r
o
m
s
ee
k
i
n
g
o
p
tim
a
l p
ar
a
m
eter
s
.
a.
T
h
e
au
th
o
r
s
s
u
b
m
i
tted
a
n
e
w
m
an
n
er
w
h
ich
o
p
ti
m
ize
SV
M
's
p
ar
a
m
eter
s
e
f
f
ec
t
iv
el
y
a
n
d
r
ed
u
ce
th
e
ti
m
e
o
f
o
p
ti
m
izatio
n
a
n
d
ca
lcu
la
tio
n
s
co
s
t
u
s
i
n
g
to
w
n
este
d
r
ea
l
v
alu
ed
g
e
n
e
tic
alg
o
r
ith
m
(
NR
G
A
)
.
T
h
e
NR
G
A
co
m
p
ar
ed
to
th
e
co
n
v
e
n
tio
n
al
o
p
ti
m
izat
io
n
m
e
ch
an
i
s
m
s
w
h
ic
h
o
p
er
ate
o
n
s
e
ek
in
g
t
h
e
w
h
o
le
p
ar
am
eter
s
to
g
et
h
er
[
7
]
.
b.
A
n
o
tatio
n
w
as
s
u
b
m
itted
in
[
8
]
f
o
r
d
eter
m
i
n
i
n
g
SVM
p
ar
a
m
eter
s
d
ep
en
d
in
g
o
n
m
in
d
s
f
r
o
m
d
esi
g
n
o
f
ex
p
er
ien
ce
s
,
w
h
ich
i
n
itia
ted
w
it
h
an
e
x
tr
e
m
el
y
r
o
u
g
h
g
r
i
d
co
m
p
r
is
i
n
g
t
h
e
co
m
p
lete
s
ea
r
ch
r
an
g
e
an
d
r
ep
etitiv
e
r
e
v
is
ed
b
o
th
t
h
e
g
r
i
d
r
eso
lu
tio
n
an
d
s
ea
r
ch
b
o
ar
d
er
s
,
s
a
f
eg
u
ar
d
in
g
t
h
e
n
u
m
b
er
o
f
f
o
r
m
s
at
ea
c
h
p
h
ase
al
m
o
s
t c
o
n
s
tan
t
s
.
c.
Gen
etic
a
lg
o
r
it
h
m
(
G
A
)
i
s
tr
e
n
d
to
b
e
co
m
p
letel
y
p
r
ett
y
at
f
i
n
d
in
g
i
n
g
e
n
er
al
p
er
f
ec
t
u
n
i
v
er
s
al
s
o
lu
t
io
n
s
.
GA
h
a
s
b
ee
n
v
ast
l
y
ad
o
p
ted
f
o
r
p
ar
a
m
eter
s
ett
in
g
.
I
n
[
9
]
a
m
a
n
n
er
b
ased
o
n
G
A
was
s
u
g
g
es
ted
to
s
i
m
u
lta
n
eo
u
s
l
y
o
p
ti
m
ize
SV
M
'
S
p
ar
a
m
eter
s
a
n
d
attr
ib
u
te
s
u
b
s
et.
I
n
[
1
0
]
GA
is
f
u
s
ed
w
it
h
a
s
y
m
p
to
tic
attitu
d
es
o
f
SVM
w
h
ic
h
t
h
e
n
g
u
id
e
s
t
h
e
s
ea
r
c
h
to
t
h
e
r
i
g
h
t
l
in
e
o
f
p
er
f
ec
t
g
e
n
er
aliza
tio
n
e
r
r
o
r
in
th
e
s
u
p
er
p
ar
am
eter
s
p
ac
e.
d.
T
h
is
s
t
u
d
y
[
9
]
d
ev
elo
p
s
a
n
o
v
el
m
a
n
n
er
ter
m
ed
P
SO+
SVM.
P
SO
b
ased
ap
p
r
o
ac
h
f
o
r
p
ar
a
m
eter
d
eter
m
in
i
n
g
an
d
f
ea
t
u
r
e
s
ele
ctio
n
,
an
d
th
e
n
a
co
m
p
ar
is
o
n
is
co
n
d
u
cted
o
f
g
ain
ed
r
e
s
u
lt
s
w
it
h
o
th
er
ap
p
r
o
ac
h
es.
T
h
e
SVM+
P
SO g
ain
ed
a
b
etter
ac
cu
r
ac
y
o
f
clas
s
if
ica
tio
n
t
h
an
o
t
h
er
test
s
.
2.
CL
AS
SI
F
I
E
R
S
C
las
s
i
f
icatio
n
is
i
m
p
er
ati
v
e
f
o
r
d
ata
m
i
n
i
n
g
.
T
h
e
lear
n
i
n
g
alg
o
r
ith
m
[
1
1
]
estab
li
s
h
e
s
a
c
lass
i
f
ier
i
n
a
g
iv
e
n
s
et
o
f
m
ea
s
u
r
e
m
en
t,
f
o
r
in
s
tan
ce
,
a
s
et
o
f
c
h
ar
ac
ter
i
s
tic
d
ata
(
x
1
,
x
2
,
….
,
x
n
)
,
w
h
er
e
x
i
d
en
o
tes
f
ea
tu
r
e
d
ata
Xi.
T
h
e
p
u
r
p
o
s
e
o
f
cla
s
s
if
icatio
n
i
s
to
i
n
itiate
th
e
ac
tu
alit
y
o
f
g
r
o
u
p
s
w
h
e
n
g
iv
e
n
a
s
et
o
f
o
b
s
er
v
atio
n
(
u
n
s
u
p
er
v
i
s
ed
lear
n
in
g
)
o
r
w
h
er
e
v
ar
io
u
s
ca
teg
o
r
ies
p
r
ev
ail
a
n
d
t
h
e
t
ar
g
et
i
s
cla
s
s
if
ied
in
to
o
n
e
o
f
th
e
p
r
ev
io
u
s
ca
te
g
o
r
ies
(
s
u
p
e
r
v
is
ed
lear
n
i
n
g
)
[
1
2
]
.
Su
p
er
v
is
ed
lear
n
in
g
h
a
s
b
ee
n
e
m
p
lo
y
ed
in
th
is
s
t
u
d
y
as
th
e
clas
s
i
f
icatio
n
m
et
h
o
d
.
2
.
1
.
SVM
I
n
t
h
is
p
ar
t,
w
e
f
o
cu
s
S
VM
,
a
m
an
n
er
u
s
i
n
g
f
o
r
a
cla
s
s
i
f
icatio
n
t
h
e
l
in
ea
r
a
n
d
n
o
n
li
n
ea
r
d
ata.
T
h
e
SVM
alg
o
r
it
h
m
o
p
er
ates
as
f
o
llo
w
s
:
t
h
e
n
o
n
li
n
ea
r
m
ap
p
in
g
i
s
u
s
ed
to
co
n
v
er
t
t
h
e
tr
ain
i
n
g
d
ata
in
to
a
h
i
g
h
er
d
i
s
tan
ce
,
u
n
d
er
th
e
f
r
es
h
d
is
tan
ce
;
it
i
n
v
esti
g
at
es
f
o
r
t
h
e
l
in
ea
r
p
er
f
ec
t
s
e
g
r
eg
atin
g
h
y
p
er
p
la
n
e
(
i.e
.
,
a
“
d
ec
i
s
io
n
b
o
u
n
d
ar
y
”
s
eg
r
eg
ati
n
g
t
h
e
tu
p
le
s
o
f
o
n
e
c
lass
f
r
o
m
a
n
o
th
er
)
.
W
ith
a
co
n
v
e
n
ie
n
t
n
o
n
lin
ea
r
m
ap
p
in
g
to
an
ad
eq
u
atel
y
elev
ated
d
is
ta
n
ce
,
th
e
d
ata
o
f
t
w
o
cla
s
s
e
s
ca
n
b
e
al
wa
y
s
s
e
g
r
eg
a
ted
b
y
a
h
y
p
er
p
la
n
e.
T
h
e
SVM
f
i
n
d
s
th
i
s
h
y
p
er
p
lan
e
u
s
in
g
s
u
p
p
o
r
t
v
ec
to
r
s
(
“
e
s
s
e
n
tial”
tr
ain
i
n
g
tu
p
les)
a
n
d
ed
g
e
s
(
d
ef
i
n
ed
b
y
th
e
s
u
p
p
o
r
t v
ec
to
r
s
)
[
1
3
,
1
4
]
.
2
.
2
.
G
enet
ic
a
lg
o
rit
h
m
(
G
A)
Gen
etic
al
g
o
r
ith
m
s
(
G
A
)
o
p
er
ate
w
i
th
a
co
llec
tio
n
o
f
n
o
m
in
ee
s
o
lu
tio
n
s
n
a
m
ed
a
p
o
p
u
latio
n
.
Dep
en
d
in
g
o
n
t
h
e
Dar
w
in
ian
p
r
in
cip
le
o
f
„
e
x
is
te
n
ce
o
f
t
h
e
f
itte
s
t‟
,
t
h
e
G
A
ea
r
n
s
t
h
e
p
er
f
ec
t
s
o
lu
tio
n
a
f
ter
s
eq
u
en
ce
s
o
f
r
ed
u
p
licate
ca
lcu
latio
n
s
.
G
A
p
r
o
d
u
cts
co
n
s
ec
u
ti
v
e
p
o
p
u
latio
n
s
o
f
alter
n
at
e
s
o
lu
tio
n
s
th
at
i
s
r
ep
r
esen
tativ
e
b
y
a
ch
r
o
m
o
s
o
m
e,
i.e
.
a
s
o
lu
tio
n
to
th
e
p
r
o
b
lem
,
till
ac
ce
p
tab
le
r
esu
lts
ar
e
ea
r
n
ed
.
GA
a
g
en
er
al
ad
ap
tiv
e
o
p
ti
m
izatio
n
s
ea
r
ch
m
e
th
o
d
o
lo
g
y
b
ased
o
n
a
d
ir
ec
t
an
alo
g
y
to
Dar
w
i
n
ia
n
n
at
u
r
al
s
elec
tio
n
an
d
g
e
n
etic
s
i
n
b
io
lo
g
ical
s
y
s
t
e
m
s
is
a
p
r
o
m
i
s
i
n
g
a
lter
n
at
iv
e
to
co
n
v
e
n
tio
n
a
l
h
e
u
r
is
tic
m
et
h
o
d
s
.
I
n
t
h
is
s
t
u
d
y
,
w
e
e
s
s
e
n
tiall
y
u
tili
ze
G
A
to
r
ef
i
n
e
th
e
p
ar
a
m
eter
s
(
C
a
n
d
γ
)
o
f
th
e
SVM
m
o
d
el
f
o
r
ir
is
d
at
aset [
1
5
,
1
6
]
.
GA
as
a
w
r
ap
p
er
m
et
h
o
d
co
m
b
in
ed
w
it
h
P
C
A
a
s
f
ilter
m
et
h
o
d
an
d
test
ed
u
s
i
n
g
SV
M
to
clas
s
if
icatio
n
lea
v
es
[
1
6
]
.
T
h
e
r
esu
lts
s
h
o
w
ed
t
h
at
G
A
co
m
b
i
n
ed
w
it
h
SVM
g
i
v
e
n
c
o
m
p
u
ti
n
g
ti
m
e
e
f
f
ec
ti
v
el
y
an
d
i
m
p
r
o
v
e
ac
c
u
r
ac
y
.
GA
also
u
s
ed
to
s
elec
t
i
m
p
o
r
tan
t
f
ea
t
u
r
es
an
d
in
s
tan
ce
s
t
h
en
tes
ted
u
s
i
n
g
SV
M
an
d
k
-
n
ea
r
e
s
t
n
ei
g
h
b
o
r
s
(
KNN)
[
1
7
-
1
9
]
.
Gain
R
atio
(
f
ilter
)
co
m
b
in
ed
w
it
h
s
eq
u
e
n
ti
al
f
o
r
w
ar
d
s
elec
t
io
n
(
SF
S)
w
r
ap
p
er
p
r
o
p
o
s
ed
to
d
ea
l
w
it
h
th
r
ee
d
atasets
;
ir
is
,
b
r
ea
s
t,
an
d
d
e
r
m
ato
lo
g
y
[
2
0
,
2
1
]
.
A
v
ar
io
u
s
f
ea
tu
r
e
s
ele
ctio
n
m
et
h
o
d
s
also
co
m
p
ar
ed
,
th
e
y
w
er
e
in
f
o
r
m
atio
n
g
ai
n
,
g
ai
n
r
atio
(
GR
)
,
s
y
m
m
etr
ical
u
n
ce
r
tai
n
t
y
(
SU)
,
C
h
i
s
q
u
ar
e
(
C
S
)
,
r
elief
,
an
d
co
r
r
elatio
n
b
ased
f
ea
tu
r
e
s
elec
tio
n
(
C
FS
)
[
1
9
]
.
T
h
e
r
esu
lt
s
h
o
w
ed
th
at
C
F
S
w
a
s
th
e
m
o
s
t
s
tab
le
w
it
h
t
h
e
h
i
g
h
est ac
c
u
r
ac
y
f
o
r
h
an
d
li
n
g
d
ata
w
i
th
t
w
o
cla
s
s
es
.
3.
M
E
T
H
O
D
As
m
en
tio
n
ed
b
ef
o
r
e
S
VM
class
i
f
ier
w
as
b
u
ilt
to
clas
s
if
y
ir
i
s
d
ataset
i
n
to
d
if
f
er
e
n
t
cla
s
s
e
s
.
T
h
e
u
s
in
g
o
f
G
A
i
s
to
o
p
ti
m
ize
SV
M
'
s
p
ar
a
m
eter
s
(
c,
g
a
m
m
a)
,
in
o
r
d
er
to
o
b
tai
n
h
i
g
h
er
an
d
b
est
ac
cu
r
ac
y
[
2
2
]
.
T
h
e
ir
is
d
ataset
h
as
f
o
u
r
attr
ib
u
tes,
p
r
in
ci
p
le
co
m
p
o
n
e
n
ts
a
n
al
y
s
i
s
(
P
C
A
)
al
g
o
r
ith
m
w
as
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8708
A
n
ew mo
d
el
fo
r
ir
is
d
a
ta
s
et
cla
s
s
ifica
tio
n
b
a
s
ed
o
n
lin
e
a
r
s
u
p
p
o
r
t v
ec
to
r
ma
ch
in
e
…
(
Za
h
r
a
a
F
a
iz
Hu
s
s
a
in
)
1081
ap
p
lied
to
r
ed
u
ce
th
ese
f
ea
t
u
r
es
(
f
ea
t
u
r
e
r
ed
u
ctio
n
)
,
a
n
d
t
h
en
o
n
l
y
t
h
r
ee
f
ea
t
u
r
es
w
er
e
ch
o
o
s
es.
W
h
er
ea
s
p
r
in
cip
al
co
m
p
o
n
e
n
t
a
n
al
y
s
is
(
P
C
A
)
is
a
m
at
h
e
m
a
tical
e
x
ec
u
t
io
n
t
h
at
co
n
v
er
t
s
a
n
u
m
b
er
o
f
(
p
o
s
s
ib
l
y
)
co
r
r
elate
d
v
ar
iab
les
in
to
a
(
s
m
aller
)
n
u
m
b
er
o
f
u
n
co
r
r
elate
d
v
ar
iab
les
n
a
m
ed
p
r
in
cip
al
c
o
m
p
o
n
en
t
s
.
P
C
A
i
s
a
d
i
m
e
n
s
io
n
-
d
ec
r
ea
s
i
n
g
i
n
s
tr
u
m
en
t
t
h
at
ca
n
b
e
u
tili
ze
d
to
d
r
o
o
p
in
g
a
g
r
ea
t
s
et
o
f
i
n
co
n
s
ta
n
t
to
a
lit
tle
s
et
t
h
at
s
ta
y
i
n
v
o
lv
es
m
o
s
t o
f
th
e
i
n
f
o
r
m
atio
n
i
n
t
h
e
b
ig
s
e
t [
1
2
,
2
]
.
T
h
e
p
r
esen
ted
tec
h
n
iq
u
e
i
n
th
is
s
t
u
d
y
u
s
ed
t
h
e
I
R
I
S
d
a
taset
a
cq
u
ir
ed
f
r
o
m
t
h
e
U
C
I
Ma
ch
i
n
e
L
ea
r
n
i
n
g
R
ep
o
s
ito
r
y
.
T
h
e
d
at
aset
i
s
i
n
a
m
u
lt
iv
ar
iate
g
r
o
u
p
as
i
t
p
r
o
v
id
es
th
e
s
tat
is
tic
o
n
th
e
I
r
is
p
la
n
t
t
y
p
e
b
ased
o
n
f
o
u
r
ch
ar
ac
ter
is
tics
w
h
ic
h
in
cl
u
d
e
w
id
t
h
,
w
id
th
a
n
d
p
etal
-
len
g
t
h
,
s
ep
al
-
le
n
g
th
,
a
n
d
v
alu
e
s
as
p
r
esen
te
d
in
Fi
g
u
r
e
1
.
T
h
e
d
ataset
i
s
co
m
p
o
s
ed
o
f
t
h
r
ee
g
r
o
u
p
s
w
it
h
5
0
ca
s
e
s
ea
c
h
an
d
a
to
tal
o
f
5
0
ca
s
e
s
.
T
h
e
d
ataset
w
er
e
f
ir
s
t
p
r
o
ce
s
s
ed
b
y
r
e
m
o
v
i
n
g
m
is
s
i
n
g
d
at
a
v
alu
e
s
.
T
h
e
ty
p
e
o
f
I
r
is
p
lan
t
is
t
h
e
f
o
r
ec
asted
ch
ar
ac
ter
is
tic
i
n
th
is
d
ataset
[
5
]
.
Fig
u
r
e
1
.
I
R
I
S
d
ataset
Step
-
by
-
s
tep
o
f
n
e
w
m
o
d
el
in
th
i
s
r
esear
ch
f
o
r
ir
is
d
ata
s
et
class
i
f
icatio
n
b
ased
o
n
lin
ea
r
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
p
ar
a
m
e
ter
's o
p
ti
m
izat
io
n
is
:
Step
-
1
: T
h
e
I
r
is
d
ataset
in
C
S
V
is
co
m
p
u
ted
as t
h
e
i
n
p
u
t.
Step
-
2
:
Di
v
id
e
t
h
e
d
ata
i
n
to
t
est
an
d
tr
ain
i
n
g
d
ataset
s
.
I
n
t
h
is
s
t
u
d
y
,
t
h
e
d
atase
t
w
as
p
ar
titi
o
n
ed
in
to
7
0
%
tr
ain
i
n
g
a
n
d
3
0
% testi
n
g
.
Step
-
3
: D
is
tin
g
u
is
h
t
h
e
tr
ain
i
n
g
d
ataset
b
ased
o
n
th
e
cla
s
s
v
a
lu
es,
t
h
at
is
,
1
,
2
an
d
3
.
Step
-
4
:
Dete
r
m
i
n
e
th
e
s
tan
d
a
r
d
d
ev
iatio
n
an
d
m
ea
n
v
a
lu
e
s
f
o
r
th
e
in
d
iv
id
u
al
d
ata
ca
s
e
b
ased
o
n
th
e
class
v
alu
e
s
.
Step
-
5
: C
h
o
o
s
e
th
e
S
VM
(
C
a
n
d
γ
)
p
ar
am
eter
s
as i
n
p
u
t to
g
e
n
etic
al
g
o
r
ith
m
o
p
ti
m
izat
io
n
.
Step
-
6
:
A
p
p
l
y
t
h
e
o
p
ti
m
al
v
al
u
e
o
f
t
h
e
(
C
an
d
γ
)
p
ar
a
m
eter
s
as a
n
i
n
it
ial
v
al
u
e
to
t
h
e
p
r
o
ce
s
s
o
f
cla
s
s
i
f
icatio
n
u
s
i
n
g
SVM.
Step
-
7
: U
tili
ze
t
h
e
m
o
d
el
a
n
d
g
en
er
ate
p
r
ed
ictio
n
s
.
Step
-
8
:
Dete
r
m
i
n
e
th
e
p
r
ed
i
ctio
n
ac
cu
r
ac
y
t
h
r
o
u
g
h
t
h
e
co
m
p
ar
is
o
n
o
f
t
h
e
cla
s
s
d
at
a
o
f
test
d
ata
s
et.
T
h
is
ac
cu
r
ac
y
is
e
v
al
u
ated
d
ep
en
d
in
g
o
n
t
h
e
r
atio
b
et
w
ee
n
0
to
1
0
0
%.
4.
RE
SU
L
T
S AN
D
CO
RR
E
L
A
T
I
O
NS
T
h
e
s
u
g
g
ested
m
o
d
el
p
r
esen
t
ed
in
Sectio
n
4
w
as
p
er
f
o
r
m
ed
o
n
th
e
I
r
is
d
ataset
w
ith
a
n
d
w
i
th
o
u
t
Step
-
5
.
I
n
ea
ch
r
u
n
,
t
h
e
o
b
tain
ed
r
esu
lt
s
w
er
e
ev
al
u
ated
b
ased
o
n
th
e
ac
cu
r
ac
y
o
f
t
h
e
SVM
cla
s
s
i
f
ier
.
T
h
e
o
b
tain
ed
r
esu
lts
s
h
o
w
ed
t
h
at
th
e
ac
c
u
r
ac
y
o
f
t
h
e
SVM
in
cr
ea
s
ed
to
9
8
.
7
u
s
in
g
Step
-
5
an
d
ab
o
u
t
9
5
.
3
%
w
it
h
o
u
t Step
-
5
.
A
ll t
h
e
r
es
u
lts
,
w
it
h
t
h
e
o
p
ti
m
izatio
n
,
ar
e
p
r
e
s
en
ted
i
n
Fi
g
u
r
es 2
,
3
,
4
,
5
,
6
a
n
d
7
,
r
esp
ec
tiv
el
y
.
T
h
e
r
esu
lts
o
f
p
r
o
p
o
s
ed
m
et
h
o
d
s
h
o
w
t
h
e
p
o
w
er
f
u
l o
f
u
s
i
n
g
g
en
et
ic
al
g
o
r
ith
m
to
o
p
ti
m
ize
th
e
(
C
an
d
g
a
m
m
a)
p
ar
am
eter
o
f
SVM
clas
s
i
f
ier
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2088
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
10
,
No
.
1
,
Feb
r
u
ar
y
2
0
2
0
:
10
79
-
1
0
8
4
1082
Fig
u
r
e
2
.
T
h
e
s
ca
tter
p
lo
t w
it
h
o
u
t g
e
n
etic
Fig
u
r
e
3
.
T
h
e
R
OC
cu
r
v
e
w
i
th
o
u
t g
e
n
etic
Fig
u
r
e
4
.
T
h
e
co
n
f
u
s
io
n
m
a
tr
i
x
w
it
h
o
u
t
g
en
et
ic
Fig
u
r
e
5
.
T
h
e
s
ca
tter
p
lo
t
w
it
h
g
en
etic
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8708
A
n
ew mo
d
el
fo
r
ir
is
d
a
ta
s
et
cla
s
s
ifica
tio
n
b
a
s
ed
o
n
lin
e
a
r
s
u
p
p
o
r
t v
ec
to
r
ma
ch
in
e
…
(
Za
h
r
a
a
F
a
iz
Hu
s
s
a
in
)
1083
Fig
u
r
e
6
.
T
h
e
R
OC
cu
r
v
e
w
i
th
g
en
etic
Fig
u
r
e
7
.
T
h
e
co
n
f
u
s
io
n
m
a
tr
i
x
w
it
h
g
e
n
etic
5.
CO
NCLU
SI
O
NS A
ND
RE
C
O
M
M
E
NDATI
O
N
T
h
is
p
ap
er
h
av
e
p
r
o
p
o
s
ed
a
n
e
w
l
y
m
o
d
e
f
o
r
cla
s
s
i
f
y
i
n
g
i
r
is
d
ata
s
e
t
u
s
i
n
g
S
VM
cla
s
s
if
ier
an
d
g
en
et
ic
alg
o
r
it
h
m
,
in
ad
d
itio
n
P
C
A
al
g
o
r
ith
m
w
a
s
u
s
e
f
o
r
f
ea
tu
r
e
s
r
ed
u
ctio
n
.
T
h
is
p
r
o
p
o
s
ed
m
o
d
e
is
to
o
p
tim
ize
c
a
n
d
g
a
m
m
a
p
ar
a
m
eter
s
o
f
li
n
ea
r
SVM.
A
s
s
h
o
wn
ab
o
v
e
th
e
r
es
u
lts
o
b
tain
ed
f
r
o
m
ap
p
lied
G
A
o
n
ir
is
d
ataset
i
s
9
8
.
7
an
d
w
ith
o
u
t
G
A
i
s
9
7
.
7
8
.
GA
w
as
u
s
ed
to
o
p
ti
m
ize
SV
M
'
s
p
ar
a
m
eter
s
(
c,
g
a
m
m
a)
,
in
o
r
d
er
to
p
r
o
m
o
tio
n
an
ef
f
ic
ac
io
u
s
SV
M
m
o
d
el
w
it
h
h
i
g
h
ac
cu
r
ac
y
a
n
d
s
tab
ilit
y
,
t
h
e
o
p
tim
al
p
ar
a
m
eter
s
ee
k
o
n
SVM
p
la
y
s
a
f
a
tef
u
l
r
o
le.
I
n
ad
v
is
ab
le
p
ar
a
m
eter
s
etti
n
g
s
r
es
u
lt
in
i
n
f
er
io
r
class
i
f
ic
atio
n
p
er
f
o
r
m
an
ce
.
Fo
r
th
e
f
u
tu
r
e
w
o
r
k
,
t
h
is
s
tu
d
y
ca
n
b
e
ex
te
n
d
in
to
t
w
o
p
ar
t;
f
ir
s
tl
y
b
y
i
m
p
r
o
v
in
g
t
h
e
p
er
f
o
r
m
an
ce
o
f
G
A
s
u
c
h
as
h
y
b
r
id
G
A
w
it
h
o
th
e
r
m
et
h
o
d
as
w
o
r
k
s
d
o
n
e
b
y
[
2
2
-
2
4
]
,
an
d
s
ec
o
n
d
l
y
b
y
ap
p
l
y
f
ea
t
u
r
e
s
elec
tio
n
m
et
h
o
d
in
SVM
f
o
r
o
p
ti
m
al
p
ar
a
m
eter
s
ettin
g
as p
r
o
p
o
s
ed
in
[
2
5
]
.
RE
F
E
R
E
NC
E
S
[1
]
Z.
L
n
lan
,
e
t
a
l,
"
Us
in
g
Ge
n
e
ti
c
A
l
g
o
rit
h
m
to
Op
ti
m
i
z
e
P
a
ra
m
e
ters
o
f
S
u
p
p
o
rt
V
e
c
to
r
M
a
c
h
in
e
a
n
d
Its
A
p
p
li
c
a
ti
o
n
in
M
a
teria
l
F
a
ti
g
u
e
L
ife
P
re
d
ictio
n
,
"
S
c
h
o
o
l
o
f
M
e
c
h
a
n
ica
l
E
n
g
i
n
e
e
rin
g
,
S
h
a
n
g
h
a
i
Un
ive
rs
it
y
o
f
En
g
i
n
e
e
rin
g
S
c
ien
c
e
,
S
h
a
n
g
h
a
i,
C
h
in
a
.
,
Ad
v
a
n
c
e
s in
N
a
tu
ra
l
S
c
ien
c
e
,
v
o
l.
8
(
1
)
,
2
0
1
5
[2
]
X
.
Z.
L
i
a
n
d
J
M
.
Ko
n
g
,
"
A
p
p
li
c
a
ti
o
n
o
f
GA
–
S
V
M
m
e
th
o
d
w
it
h
p
a
ra
m
e
ter
o
p
ti
m
iza
ti
o
n
f
o
rlan
d
slid
e
d
e
v
e
lo
p
m
e
n
t
p
re
d
ictio
n
,
"
Na
t
.
Ha
z
a
rd
s E
a
rth
S
y
st.
S
c
i.
,
v
o
l
.
1
4
,
p
p
.
5
2
5
–
5
3
3
,
2
0
1
4
.
[3
]
M
a
o
,
K.
Z.
,
"
F
e
a
tu
re
su
b
se
t
se
lec
ti
o
n
f
o
r
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
s
th
r
o
u
g
h
d
isc
rim
in
a
ti
v
e
f
u
n
c
ti
o
n
p
r
u
n
i
n
g
a
n
a
ly
sis,"
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
S
y
ste
ms
,
M
a
n
,
a
n
d
Cy
b
e
rn
e
ti
c
s
,
v
o
l.
3
4
(
1
),
p
p
.
6
0
-
6
7
,
2
0
0
4
.
[4
]
A
b
b
a
s
F
.
H.
A
lh
a
ra
n
,
Ha
y
d
e
r
K.
F
a
tl
a
w
i,
Na
b
e
e
l
S
a
li
h
A
li
,
“
A
c
lu
ste
r
-
b
a
se
d
f
e
a
tu
re
se
lec
ti
o
n
m
e
th
o
d
f
o
r
im
a
g
e
tex
tu
re
c
las
si
f
ica
ti
o
n
,”
I
n
d
o
n
esi
an
J
o
u
r
n
al
o
f
E
lectr
ical
E
n
g
i
n
ee
r
in
g
a
n
d
C
o
m
p
u
ter
Scie
n
ce
,
V
o
l
1
4
,
No
3
:
p
p
1
4
3
3
-
1
4
4
2
,
J
u
n
e
2
0
1
9
.
[5
]
M
a
ry
a
m
,
N.
Ak
h
m
a
d
S
e
ti
a
w
a
n
,
a
n
d
O.
W
a
h
y
u
n
g
g
o
ro
.
,
"
A
H
y
b
rid
F
e
a
tu
re
S
e
lec
ti
o
n
M
e
th
o
d
Us
in
g
M
u
lt
icla
ss
S
VM
f
o
r
Dia
g
n
o
sis
o
f
Er
y
th
e
m
a
t
o
-
S
q
u
a
m
o
u
s
Dise
a
se
,
"
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
M
a
th
e
ma
ti
c
s
W
o
rld
C
o
n
g
re
ss
o
n
,
2
0
1
7
.
[6
]
L
.
Tala
v
e
ra
.
,
"
A
n
e
v
a
lu
a
ti
o
n
o
f
f
il
ter
a
n
d
w
r
a
p
p
e
r
m
e
th
o
d
s
fo
r
f
e
a
tu
re
se
le
c
ti
o
n
in
c
a
teg
o
ri
c
a
l
c
lu
ste
rin
g
,"
Ad
v
.
In
tell.
Da
ta
An
a
l.
VI
, p
p
.
7
4
2
,
2
0
0
5
.
[7
]
P
.
L
iao
,
X.
Z
h
a
n
g
,
a
n
d
K.
L
i.
,
"
P
a
ra
m
e
ter
Op
ti
m
iz
a
ti
o
n
f
o
r
S
u
p
p
o
rt
V
e
c
to
r
M
a
c
h
i
n
e
Ba
se
d
o
n
Ne
ste
d
G
e
n
e
ti
c
A
l
g
o
rit
h
m
s
,"
In
fo
rm
a
ti
o
n
E
n
g
i
n
e
e
rin
g
S
c
h
o
o
l,
N
a
n
c
h
a
n
g
Un
ive
rs
i
ty,
Na
n
c
h
a
n
g
,
C
h
in
a
,
J
o
u
rn
a
l
o
f
Au
to
m
a
ti
o
n
a
n
d
Co
n
tro
l
E
n
g
i
n
e
e
rin
g
,
2
0
1
6
.
[8
]
C.
S
tae
li
n
.
,
Pa
ra
me
ter
se
lec
ti
o
n
fo
r
su
p
p
o
rt
v
e
c
to
r
ma
c
h
i
n
e
s
,
T
e
c
h
n
ica
l
Re
p
o
rt
H
P
L
-
2
0
0
2
-
3
5
4
(R.
1
)
,
H
P
L
a
b
o
ra
to
ries
Isra
e
l,
2
0
0
3
.
[9
]
C.
L
.
Hu
a
n
g
a
n
d
C.
J.
W
a
n
g
,
"
A
GA
-
b
a
s
e
d
fe
a
tu
re
se
le
c
ti
o
n
a
n
d
p
a
ra
m
e
ter
s
o
p
ti
m
iza
t
io
n
f
o
r
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
,
"
Exp
e
rt S
y
ste
ms
wit
h
A
p
p
li
c
a
ti
o
n
s
,
v
o
l
.
3
1
(
2
)
,
p
p
.
2
3
1
-
2
4
0
,
2
0
0
6
.
[1
0
]
C.
H.
W
u
,
G
.
H.
T
z
e
n
g
,
Y.
J.
G
o
o
,
a
n
d
W
.
C
.
F
a
n
g
.
,
"
A
re
a
l
-
v
a
lu
e
d
g
e
n
e
ti
c
a
lg
o
rit
h
m
to
o
p
ti
m
iz
e
th
e
p
a
ra
m
e
ters
o
f
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
fo
r
p
re
d
ictin
g
b
a
n
k
ru
p
tcy
,
"
Ex
p
e
rt
S
y
ste
ms
wit
h
A
p
p
li
c
a
ti
o
n
s
,
v
o
l.
3
2
(
2
)
,
p
p
.
3
9
7
-
4
0
8
,
2
0
0
7
.
[1
1
]
S
.
W
.
L
in
,
K.
C.
Yin
g
,
S
.
C.
C
h
e
n
,
a
n
d
Z.
J.
L
e
e
.
,
"
P
a
rti
c
le
sw
a
r
m
o
p
ti
m
iz
a
ti
o
n
f
o
r
p
a
ra
m
e
ter
d
e
t
e
r
m
in
a
ti
o
n
a
n
d
f
e
a
tu
re
se
lec
t
io
n
o
f
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
s,"
Exp
e
rt S
y
ste
ms
wit
h
Ap
p
li
c
a
ti
o
n
s
,
v
o
l
.
3
5
,
p
p
.
1
8
1
7
-
1
8
2
4
,
2
0
0
8
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2088
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
10
,
No
.
1
,
Feb
r
u
ar
y
2
0
2
0
:
10
79
-
1
0
8
4
1084
[1
2
]
M
in
,
J.
H.
a
n
d
L
e
e
,
Y.
C.
"
B
a
n
k
ru
p
tcy
p
re
d
ictio
n
u
si
n
g
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
w
it
h
o
p
ti
m
a
l
c
h
o
ice
o
f
k
e
rn
e
l
f
u
n
c
ti
o
n
p
a
ra
m
e
t
e
rs
,"
Exp
e
rt S
y
st.
Ap
p
l.
,
v
o
l.
2
8
,
p
p
.
6
0
3
-
6
1
4
,
2
0
0
5
.
[1
3
]
G
u
,
J.
Y.,
L
iu
,
J.
F
,
a
n
d
Ch
e
n
,
M
.
,
"
A
m
o
d
if
ied
re
g
re
ss
io
n
p
re
d
ictio
n
a
lg
o
rit
h
m
o
f
lar
g
e
sa
m
p
le
d
a
ta
b
a
se
d
o
n
S
V
M
,"
Co
mp
u
ter
En
g
in
e
e
rin
g
,
v
o
l.
4
0
(
1
),
p
p
.
1
6
1
-
1
6
6
,
2
0
1
4
.
[1
4
]
M
.
Z.
A
.
A
h
m
e
d
Hu
ss
e
in
A
li
,
"
A
S
u
rv
e
y
o
n
V
e
rti
c
a
l
a
n
d
Ho
riz
o
n
tal
S
c
a
li
n
g
P
latf
o
rm
s
f
o
r
Big
Da
ta
A
n
a
l
y
ti
c
s,"
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
I
n
teg
ra
ted
En
g
i
n
e
e
rin
g
,
2
0
1
8
.
[1
5
]
M
.
Z.
A
.
A
h
m
e
d
Hu
ss
e
in
A
li
,
"
A
n
Eff
icie
n
t
M
o
d
e
l
f
o
r
Da
ta
Clas
sif
ica
ti
o
n
Ba
se
d
o
n
S
VM
G
rid
P
a
ra
m
e
ter
Op
ti
m
iza
ti
o
n
a
n
d
P
S
O F
e
a
tu
re
W
e
ig
h
t
S
e
lec
ti
o
n
,
"
I
n
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
In
te
g
ra
ted
En
g
in
e
e
rin
g
,
2
0
1
8
.
[1
6
]
C.
T
sa
i,
W
.
Eb
e
rle,
a
n
d
C.
Ch
u
., "
Kn
o
w
led
g
e
-
B
a
se
d
S
y
ste
m
s
G
e
n
e
ti
c
a
lg
o
rit
h
m
s
in
f
e
a
tu
re
a
n
d
in
st
a
n
c
e
se
lec
ti
o
n
,
"
Kn
o
wled
g
e
-
Ba
se
d
S
y
st
.
,
v
o
l.
3
9
,
p
p
.
2
4
0
-
2
4
7
,
2
0
1
3
.
[1
7
]
M
.
Ka
ra
b
a
tak
a
n
d
M
.
C.
In
c
e
,
"
A
n
e
w
f
e
a
tu
re
s
e
lec
ti
o
n
m
e
th
o
d
b
a
se
d
o
n
a
ss
o
c
iatio
n
r
u
les
f
o
r
d
iag
n
o
sis
o
f
e
r
y
th
e
m
a
to
-
sq
u
a
m
o
u
s d
ise
a
se
s,"
Exp
e
rt S
y
st.
A
p
p
l
.
,
v
o
l.
3
6
(
10
),
p
p
.
1
2
5
0
0
-
1
2
5
0
5
,
2
0
0
9
.
[1
8
]
D.
Zh
a
n
g
,
e
t
a
l
,
"
A
G
e
n
e
ti
c
A
lg
o
rit
h
m
Ba
se
d
S
u
p
p
o
rt
V
e
c
to
r
M
a
c
h
in
e
M
o
d
e
l
f
o
r
Blo
o
d
-
Bra
i
n
Ba
rr
ier
P
e
n
e
tratio
n
P
re
d
ictio
n
,
"
Bi
o
M
e
d
Res
e
a
rc
h
In
t
e
rn
a
ti
o
n
a
l
,
2
0
1
5
.
[1
9
]
Ka
si
m
S
.
,
Ha
s
sa
n
R.
,
M
o
h
d
N.
S
.
,
Ra
m
lan
R.
,
M
a
h
d
in
H,
a
n
d
F
u
d
z
e
e
M. F.
M.,
"
A
Co
m
p
a
ra
ti
v
e
S
tu
d
y
o
f
Di
ff
e
r
e
n
t
T
e
m
p
late
M
a
tch
in
g
T
e
c
h
n
iq
u
e
s
f
o
r
Tw
in
Iris
Re
c
o
g
n
it
io
n
,
"
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
n
Ad
v
a
n
c
e
d
S
c
ien
c
e
,
En
g
i
n
e
e
rin
g
a
n
d
I
n
fo
rm
a
ti
o
n
T
e
c
h
n
o
lo
g
y
,
v
o
l.
7
(
4
-
2
),
p
p
.
1
5
8
1
-
1
5
8
8
,
2
0
1
7
.
[2
0
]
Ka
si
m
S
.
,
Ha
ss
a
n
R.
,
Zain
i
N.
H.,
S
y
i
f
a
a
‟
A
h
m
a
d
A
.
,
Ra
m
li
A
.
A
,
a
n
d
S
a
e
d
u
d
in
R.
R
.
,
"
A
S
tu
d
y
o
n
F
a
c
ial
Ex
p
re
ss
io
n
Re
c
o
g
n
it
i
o
n
Us
i
n
g
L
o
c
a
l
Bin
a
ry
P
a
tt
e
rn
,
"
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
n
A
d
v
a
n
c
e
d
S
c
ien
c
e
,
E
n
g
i
n
e
e
rin
g
a
n
d
In
fo
rm
a
t
io
n
T
e
c
h
n
o
l
o
g
y
,
v
o
l.
7
(5
)
,
p
p
.
1
6
2
1
-
1
6
2
6
,
2
0
1
7
.
[2
1
]
Zi
n
N.A
.
M
.
,
A
s
m
u
n
i
H.,
Ha
m
e
d
H.N.A
.
,
Oth
m
a
n
R.
M
.
,
Ka
sim
S.
,
Ha
ss
a
n
R.
,
Zak
a
ria
Z
,
a
n
d
Ro
sla
n
R.
,
"
Co
n
tac
t
len
s
c
las
si
f
ica
ti
o
n
b
y
u
sin
g
se
g
m
e
n
ted
len
s
b
o
u
n
d
a
ry
f
e
a
tu
re
s
,"
In
d
o
n
e
si
a
n
J
o
u
rn
a
l
o
f
El
e
c
trica
l
E
n
g
i
n
e
e
rin
g
a
n
d
Co
mp
u
ter
S
c
ien
c
e
(
IJ
EE
CS
)
,
v
o
l.
1
1
(
3
),
p
p
.
1
1
2
9
-
1
1
3
5
,
2
0
1
9
.
[2
2
]
Is
m
a
il
M
.
A
.
,
Ra
z
a
k
K.
A
.
B.
,
M
o
o
rth
y
K.,
M
e
z
h
u
y
e
v
V
.
,
Ka
sim
S
,
a
n
d
Ib
ra
h
im
A
.
O.,
"
Ne
w
to
n
Co
m
p
e
ti
ti
v
e
G
e
n
e
ti
c
A
l
g
o
rit
h
m
M
e
th
o
d
f
o
r
Op
ti
m
i
z
a
ti
o
n
th
e
P
ro
d
u
c
ti
o
n
o
f
Bio
c
h
e
m
ic
a
l
S
y
ste
m
s
,"
Ad
v
a
n
c
e
d
S
c
ien
c
e
L
e
tt
e
rs
,
v
o
l.
2
4
(1
0
),
p
p
.
7
4
8
1
-
7
4
8
5
,
2
0
1
8
.
[2
3
]
Is
m
a
il
M
.
A
.
,
M
e
z
h
u
y
e
v
V
.
,
De
ris
S
.
,
M
o
h
a
m
a
d
,
M
.
S
.
,
Ka
sim
S
,
a
n
d
S
a
e
d
u
d
i
n
R.
R.
,
"
M
u
l
ti
-
o
b
jec
ti
v
e
Op
ti
m
iza
ti
o
n
o
f
Bio
c
h
e
m
i
c
a
l
S
y
ste
m
P
ro
d
u
c
t
io
n
Us
in
g
a
n
Im
p
ro
v
e
Ne
w
to
n
Co
m
p
e
ti
ti
v
e
Di
ff
e
r
e
n
ti
a
l
Ev
o
lu
ti
o
n
M
e
th
o
d
,"
In
ter
n
a
t
io
n
a
l
J
o
u
r
n
a
l
o
n
A
d
v
a
n
c
e
d
S
c
ien
c
e
,
En
g
i
n
e
e
rin
g
a
n
d
I
n
f
o
rm
a
ti
o
n
T
e
c
h
n
o
lo
g
y
,
v
o
l.
7
(4
-
2
),
p
p
.
1
5
3
5
-
1
5
4
2
,
2
0
1
7
.
[2
4
]
Is
m
a
il
M
.
A
.
,
M
e
z
h
u
y
e
v
V
.
M
o
o
rth
y
K.,
Ka
si
m
S
,
a
n
d
Ib
ra
h
im
A
.
O.,
"
Op
ti
m
isa
ti
o
n
o
f
Bio
c
h
e
m
ic
a
l
S
y
ste
m
s
P
r
o
d
u
c
ti
o
n
u
sin
g
Hy
b
rid
o
f
Ne
w
to
n
m
e
th
o
d
,
Dif
fe
re
n
ti
a
l
Ev
o
lu
ti
o
n
A
lg
o
rit
h
m
a
n
d
Co
o
p
e
ra
ti
v
e
Co
e
v
o
lu
ti
o
n
A
l
g
o
rit
h
m
,"
In
d
o
n
e
si
a
n
J
o
u
rn
a
l
o
f
El
e
c
trica
l
E
n
g
i
n
e
e
rin
g
a
n
d
C
o
mp
u
ter
S
c
ien
c
e
,
v
o
l.
8
(1
),
p
p
.
2
7
-
3
5
,
2
0
1
7
.
[2
5
]
Ib
ra
h
im
A
.
,
Hu
ss
ien
W.
,
Ya
g
o
o
p
A
,
a
n
d
Ism
a
il
M
.
,
"
F
e
a
tu
re
S
e
lec
ti
o
n
a
n
d
Ra
d
ial
Ba
sis
F
u
n
c
ti
o
n
Ne
tw
o
rk
f
o
r
P
a
rk
in
so
n
Dise
a
se
Clas
sifi
c
a
ti
o
n
,"
Ku
rd
ista
n
J
o
u
r
n
a
l
o
f
Ap
p
li
e
d
R
e
se
a
rc
h
,
v
o
l.
2
(
3
),
p
p
.
1
6
7
-
1
7
1
,
2
0
1
7
.
Evaluation Warning : The document was created with Spire.PDF for Python.