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.
11
,
No
.
4
,
A
u
g
u
s
t
2021
,
p
p
.
3
3
9
3
~
3
4
0
2
I
SS
N:
2
0
8
8
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v
1
1
i
4
.
pp
3
3
9
3
-
3
4
0
2
3393
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m
Co
m
pa
ra
tive
a
na
ly
sis
of
R
elief
F
-
S
VM
and C
FS
-
S
V
M
f
o
r
m
icro
a
rray
d
a
ta
c
la
ss
ificatio
n
M
o
cha
m
a
d Ag
us
t
a
Na
o
f
a
l H
a
k
i
m
,
A
diw
ij
a
y
a
,
Widi
Astut
i
De
p
a
rt
m
e
n
t
o
f
In
f
o
rm
a
ti
c
En
g
in
e
e
rin
g
,
T
e
lk
o
m
Un
iv
e
rsit
y
,
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
Ju
l
10
,
2
0
20
R
ev
i
s
ed
J
an
4
,
202
1
A
cc
ep
ted
J
an
1
3
,
2
0
2
1
Ca
n
c
e
r
is
o
n
e
o
f
th
e
m
a
in
c
a
u
se
s
o
f
d
e
a
th
in
t
h
e
w
o
rld
w
h
e
re
th
e
W
o
rld
He
a
lt
h
Org
a
n
iza
ti
o
n
(W
HO
)
re
c
o
g
n
ize
d
c
a
n
c
e
r
a
s
a
m
o
n
g
th
e
to
p
c
a
u
se
s
o
f
d
e
a
th
i
n
2
0
1
8
.
T
h
u
s
,
d
e
tec
ti
n
g
c
a
n
c
e
r
s
y
m
p
to
m
s
is
p
a
ra
m
o
u
n
t
i
n
o
r
d
e
r
t
o
c
u
re
a
n
d
s
u
b
se
q
u
e
n
tl
y
re
d
u
c
e
th
e
c
a
su
a
lt
ies
d
u
e
t
o
c
a
n
c
e
r
d
i
se
a
se
.
M
a
n
y
stu
d
ies
h
a
v
e
b
e
e
n
d
e
v
e
lo
p
e
d
d
a
t
a
m
in
in
g
a
p
p
r
o
a
c
h
e
s
to
d
e
tec
t
sy
m
p
to
m
s
o
f
c
a
n
c
e
r
th
ro
u
g
h
a
c
la
ss
i
fy
in
g
h
u
m
a
n
g
e
n
e
d
a
ta
e
x
p
re
ss
io
n
.
On
e
p
o
p
u
lar
a
p
p
ro
a
c
h
is
u
sin
g
m
icro
a
rra
y
d
a
ta
b
a
se
d
o
n
DN
A
.
Ho
w
e
v
e
r,
DN
A
m
icro
a
rra
y
d
a
ta
h
a
s
m
a
n
y
d
i
m
e
n
sio
n
s
th
a
t
c
a
n
h
a
v
e
a
d
e
tri
m
e
n
tal
e
ff
e
c
t
o
n
th
e
a
c
c
u
ra
c
y
o
f
c
las
sif
ic
a
ti
o
n
.
T
h
e
re
f
o
re
,
b
e
f
o
re
p
e
rf
o
r
m
in
g
c
las
si
f
ica
t
io
n
,
a
f
e
a
tu
re
se
lec
ti
o
n
tec
h
n
iq
u
e
m
u
st
b
e
u
se
d
to
e
li
m
in
a
te
fe
a
tu
re
s
t
h
a
t
d
o
n
o
t
h
a
v
e
i
m
p
o
rtan
t
i
n
f
o
rm
a
ti
o
n
to
su
p
p
o
rt
t
h
e
c
las
sif
ic
a
ti
o
n
p
r
o
c
e
ss
.
T
h
e
fe
a
tu
re
se
lec
ti
o
n
tec
h
n
i
q
u
e
s
u
se
d
w
e
re
Re
li
e
f
F
a
n
d
c
o
rre
latio
n
-
b
a
se
d
f
e
a
tu
re
se
lec
ti
o
n
(CF
S
)
a
n
d
a
c
las
s
if
ic
a
t
io
n
tec
h
n
iq
u
e
u
se
d
in
th
is
stu
d
y
is
su
p
p
o
r
t
v
e
c
to
r
m
a
c
h
in
e
(S
V
M
).
S
e
v
e
ra
l
tes
ti
n
g
sc
h
e
m
e
s
we
re
a
p
p
li
e
d
i
n
t
h
is
a
n
a
ly
sis
to
c
o
m
p
a
re
th
e
p
e
rf
o
r
m
a
n
c
e
o
f
Re
li
e
f
F
a
n
d
CF
S
w
it
h
S
V
M
.
It
s
h
o
w
e
d
th
a
t
th
e
Re
li
e
f
F
o
u
tp
e
rf
o
rm
e
d
c
o
m
p
a
re
d
w
it
h
CF
S
a
s
m
icro
a
rra
y
d
a
ta
c
las
si
f
ica
ti
o
n
a
p
p
r
o
a
c
h
.
K
ey
w
o
r
d
s
:
C
las
s
i
f
icatio
n
C
o
r
r
elatio
n
-
b
ased
f
ea
tu
r
e
s
elec
tio
n
Featu
r
e
s
elec
t
io
n
Mic
r
o
ar
r
ay
d
ata
R
elie
f
F
Su
p
p
o
r
t v
ec
to
r
m
ac
h
i
n
e
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
ch
a
m
ad
Ag
u
s
ta
Nao
f
al
Ha
k
i
m
Dep
ar
t
m
en
t
I
n
f
o
r
m
atic
E
n
g
i
n
e
er
in
g
T
elk
o
m
U
n
i
v
er
s
it
y
T
elek
o
m
u
n
ik
a
s
i
Stre
et
,
B
an
d
u
n
g
T
o
w
n
s
h
ip
,
W
est J
a
w
a
C
o
u
n
t
y
40257
,
I
n
d
o
n
esia
E
m
ail:
n
ao
f
al
h
a
k
i
m
@
s
t
u
d
en
ts
.
telk
o
m
u
n
iv
er
s
it
y
.
ac
.
id
1.
I
NT
RO
D
UCT
I
O
N
A
cc
o
r
d
in
g
to
t
h
e
W
o
r
ld
Hea
lt
h
Or
g
an
izat
io
n
(
W
HO)
,
ca
n
ce
r
is
o
n
e
o
f
t
h
e
w
o
r
ld
'
s
lead
in
g
ca
u
s
e
s
o
f
d
ea
th
w
h
er
e
it
is
r
a
n
k
ed
s
ec
o
n
d
o
f
t
h
e
to
p
ten
ca
u
s
e
s
o
f
d
e
ath
g
lo
b
all
y
,
w
it
h
es
ti
m
ates
r
ea
ch
in
g
9
.
6
m
i
llio
n
d
ea
th
s
in
2
0
1
8
[
1
]
.
Ma
n
y
s
tu
d
ies
h
av
e
d
ev
elo
p
ed
w
a
y
s
to
d
etec
t
ca
n
ce
r
ca
u
s
alit
y
,
a
n
d
o
n
e
o
f
th
e
p
o
p
u
lar
ap
p
r
o
ac
h
es
is
t
h
r
o
u
g
h
g
e
n
etic
ex
p
r
ess
io
n
.
T
h
e
h
u
m
an
g
e
n
e
d
ata
h
as
b
ee
n
e
x
p
r
ess
ed
i
n
t
h
e
f
o
r
m
o
f
a
s
to
r
ag
e
m
ed
iu
m
ca
lled
DN
A
m
icr
o
ar
r
a
y
d
ata
[
2
]
.
Us
u
all
y
,
t
h
e
cla
s
s
if
icatio
n
p
r
o
ce
s
s
m
u
s
t
f
ir
s
t
b
e
u
s
ed
o
n
t
h
e
DN
A
m
icr
o
ar
r
a
y
d
ata
to
d
etec
t
th
e
c
an
ce
r
ca
u
s
a
lit
y
.
Ho
w
e
v
er
,
DN
A
m
icr
o
ar
r
a
y
d
ata
h
a
s
m
a
n
y
d
i
m
e
n
s
io
n
s
t
h
at
ca
n
ad
v
er
s
el
y
a
f
f
ec
t
cla
s
s
i
f
icat
io
n
p
er
f
o
r
m
an
ce
[
1
,
2
]
.
On
e
p
o
s
s
ib
le
ap
p
r
o
ac
h
to
o
v
er
co
m
e
t
h
is
p
r
o
b
le
m
i
s
b
y
s
elec
ti
n
g
f
ea
tu
r
es
t
h
at
ar
e
co
n
s
id
er
ed
m
o
r
e
r
elev
an
t
in
p
r
o
d
u
cin
g
o
p
ti
m
al
class
i
f
icat
io
n
p
er
f
o
r
m
an
ce
[
3
]
.
T
h
er
ef
o
r
e,
it
is
n
ec
es
s
ar
y
to
a
p
p
ly
a
f
ea
t
u
r
e
s
elec
tio
n
tech
n
i
q
u
e
p
r
io
r
to
co
n
d
u
ctin
g
th
e
cl
ass
i
f
icatio
n
p
r
o
ce
s
s
to
ac
h
ie
v
e
m
o
r
e
o
p
ti
m
al
ca
n
c
er
d
etec
tio
n
.
Ho
w
ev
er
,
n
o
t
al
l
f
ea
t
u
r
e
s
e
lectio
n
tech
n
iq
u
es
a
n
d
i
m
p
le
m
e
n
tatio
n
s
ch
e
m
es
ca
n
o
p
ti
m
ize
class
i
f
icatio
n
p
r
o
ce
s
s
;
m
an
y
ex
p
er
i
m
en
ts
ar
e
n
ee
d
ed
to
c
o
m
p
ar
e
f
ea
tu
r
e
s
elec
tio
n
tech
n
iq
u
es
a
n
d
th
e
ap
p
licatio
n
o
f
th
e
cla
s
s
i
f
icat
io
n
s
c
h
e
m
e
to
ac
h
iev
e
th
e
m
o
s
t
o
p
ti
m
al
r
esu
lts
.
T
h
e
t
w
o
p
o
p
u
lar
ap
p
r
o
ac
h
es a
r
e
R
elief
F a
n
d
co
r
r
elatio
n
-
b
ased
f
ea
tu
r
e
s
elec
tio
n
f
o
r
f
ea
tu
r
e
s
e
lectio
n
p
r
o
ce
s
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
11
,
No
.
4
,
A
u
g
u
s
t 2
0
2
1
:
3
3
9
3
-
3402
3394
T
h
is
p
ap
er
w
o
u
ld
li
k
e
to
u
s
e
s
t
w
o
ap
p
r
o
ac
h
es,
i.e
.
,
R
elie
f
F
an
d
C
o
r
r
elatio
n
-
b
ased
f
ea
tu
r
e
s
elec
tio
n
(
C
FS
)
to
f
ilter
f
ea
t
u
r
e
s
elec
ti
o
n
.
T
h
ese
t
w
o
tec
h
n
iq
u
e
s
w
e
r
e
ch
o
s
en
b
ec
au
s
e
t
h
e
y
ca
n
i
n
d
ep
en
d
en
tl
y
a
s
s
e
s
s
th
e
q
u
a
lit
y
o
f
g
o
o
d
f
ea
t
u
r
es,
an
d
h
a
v
e
a
s
i
m
p
ler
an
d
f
as
ter
co
m
p
u
tat
io
n
ti
m
e
co
m
p
ar
ed
to
o
th
er
tech
n
iq
u
e
s
[
4
,
5
]
;
th
er
ef
o
r
e,
th
e
f
il
ter
tech
n
iq
u
e
is
f
o
u
n
d
ap
p
r
o
p
r
iate
to
u
s
e
f
o
r
m
icr
o
ar
r
ay
d
ata
w
i
th
m
a
n
y
d
i
m
en
s
io
n
s
.
T
h
e
R
elie
f
F
ap
p
r
o
ac
h
is
b
ased
o
n
r
an
k
in
g
a
p
p
r
o
ac
h
,
w
h
ile
t
h
e
C
FS
is
b
ased
o
n
a
s
u
b
s
et
ap
p
r
o
ac
h
.
R
elief
F
g
i
v
e
s
a
r
a
n
k
i
n
g
v
al
u
e
to
ea
ch
f
ea
tu
r
e
ag
ain
s
t
i
ts
cla
s
s
a
ttrib
u
te
s
;
t
h
e
f
ea
t
u
r
es
w
i
th
th
e
h
ig
h
e
s
t
w
ei
g
h
t
w
il
l
p
o
s
iti
v
el
y
i
m
p
ac
t
t
h
e
cla
s
s
i
f
icatio
n
p
r
o
ce
s
s
.
Me
a
n
w
h
ile,
t
h
e
C
FS
h
elp
s
ass
e
s
s
w
h
et
h
er
a
s
u
b
s
et
o
f
f
ea
t
u
r
es
u
s
es
m
er
it_
s
ca
lcu
latio
n
s
b
ased
o
n
t
h
e
co
r
r
elatio
n
b
et
w
ee
n
f
ea
t
u
r
es
a
n
d
class
es,
as
w
el
l
as
th
e
co
r
r
elatio
n
b
et
w
ee
n
f
ea
t
u
r
es
w
it
h
o
t
h
er
f
ea
t
u
r
es;
t
h
e
g
r
ea
ter
th
e
m
er
it_
s
v
al
u
e
o
f
a
s
u
b
s
et,
t
h
e
b
etter
it
s
i
m
p
ac
t
o
n
t
h
e
cla
s
s
i
f
icatio
n
p
r
o
ce
s
s
[
6
]
.
T
h
e
s
u
p
p
o
r
t
v
ec
t
o
r
m
ac
h
i
n
e
(
SVM)
cla
s
s
i
f
icat
io
n
tech
n
iq
u
e
w
as
ch
o
s
en
b
ec
au
s
e
it
ca
n
p
r
o
d
u
ce
b
etter
ac
cu
r
ac
y
w
it
h
m
icr
o
a
r
r
ay
d
ata
co
m
p
ar
ed
to
s
ev
er
al
o
th
er
class
if
ica
tio
n
tech
n
iq
u
es [
7
-
9
]
.
T
h
e
d
ataset
u
s
ed
i
n
t
h
i
s
p
a
p
er
w
as
ta
k
en
f
r
o
m
t
h
e
Ke
n
t
R
id
g
e
b
io
-
m
ed
ical
d
atase
t
r
ep
o
s
ito
r
y
(
h
ttp
://l
eo
.
u
g
r
.
es/el
v
ir
a/DB
C
R
ep
o
s
i
to
r
y
/)
co
m
p
r
is
i
n
g
co
lo
n
tu
m
o
u
r
,
o
v
ar
ian
,
b
r
ea
s
t
ca
n
ce
r
,
lu
n
g
ca
n
ce
r
,
p
r
o
s
tate,
th
e
ce
n
tr
al
n
er
v
o
u
s
s
y
s
te
m
,
an
d
M
L
L
leu
k
ae
m
ia
d
ata.
B
y
co
n
d
u
cti
n
g
t
h
i
s
r
esear
ch
,
th
e
b
est
i
m
p
le
m
en
ta
tio
n
an
d
p
er
f
o
r
m
an
ce
o
f
R
elie
f
F
an
d
C
F
S
tech
n
iq
u
es
f
o
r
i
m
p
r
o
v
in
g
th
e
clas
s
i
f
i
ca
tio
n
p
er
f
o
r
m
a
n
ce
o
f
t
h
e
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
i
n
e
(
SVM)
cla
s
s
if
icatio
n
tec
h
n
iq
u
e
o
n
m
icr
o
ar
r
ay
d
ata
ca
n
b
e
id
en
ti
f
ied
.
I
t
w
as
s
h
o
w
n
t
h
at
t
h
e
R
elie
f
F
o
u
tp
er
f
o
r
m
ed
co
m
p
ar
ed
w
it
h
C
FS
f
o
r
m
icr
o
ar
r
ay
d
ata
cla
s
s
i
f
icat
io
n
tech
n
iq
u
e.
T
h
e
r
est
o
f
th
e
p
ap
er
is
o
r
g
an
i
ze
d
as
f
o
llo
w
s
:
Sectio
n
2
,
d
is
cu
s
s
es
r
esear
ch
m
et
h
o
d
o
lo
g
y
o
n
h
o
w
t
h
e
s
tep
o
f
r
esear
ch
.
T
h
e
r
es
u
lts
a
n
d
d
is
cu
s
s
io
n
ar
e
p
r
esen
ted
i
n
s
ec
tio
n
3
to
s
u
p
p
o
r
t
th
e
a
n
a
l
y
s
i
s
o
f
t
h
e
r
es
u
lt
s
.
Sectio
n
4
co
n
clu
d
e
s
w
it
h
a
s
u
m
m
ar
y
o
f
t
h
e
co
m
p
ar
is
o
n
b
etw
ee
n
t
w
o
tech
n
iq
u
e
s
.
2.
RE
S
E
ARCH
M
E
T
H
O
D
I
n
th
e
r
esear
ch
d
ev
e
lo
p
m
en
t
p
h
ase,
t
w
o
f
ea
t
u
r
e
s
elec
tio
n
tec
h
n
iq
u
es
w
er
e
co
m
p
ar
ed
,
n
a
m
e
l
y
R
elie
f
F
an
d
co
r
r
elatio
n
-
b
ased
f
ea
t
u
r
e
s
elec
tio
n
(
C
FS
)
.
T
h
ese
tech
n
i
q
u
es
ar
e
s
i
m
p
le
a
n
d
w
id
el
y
u
s
ed
.
T
h
is
s
tu
d
y
ai
m
s
to
ascer
tain
t
h
e
p
er
f
o
r
m
a
n
ce
o
f
b
o
th
tec
h
n
iq
u
es
i
n
t
h
e
cla
s
s
if
ica
tio
n
p
r
o
ce
s
s
u
s
i
n
g
b
ased
o
n
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SV
M)
.
B
ased
o
n
Fig
u
r
e
1
,
th
e
s
y
s
te
m
in
c
lu
d
es
t
h
e
n
o
r
m
aliza
tio
n
o
f
d
ata
in
to
v
alu
e
s
b
et
w
ee
n
[
0
,
1
]
[
1
0
]
.
T
h
e
d
ata
w
as
d
iv
id
ed
i
n
to
tr
a
in
i
n
g
d
ata
an
d
te
s
ti
n
g
d
ata,
r
esp
ec
tiv
el
y
.
T
h
en
,
t
h
e
f
ea
t
u
r
e
s
elec
tio
n
p
r
o
ce
s
s
,
n
a
m
e
l
y
R
elie
f
F
a
n
d
co
r
r
elatio
n
-
b
a
s
ed
f
ea
t
u
r
e
s
elec
tio
n
(
C
F
S)
b
e
ca
r
r
ied
o
u
t
o
n
th
e
tr
ai
n
i
n
g
d
ata,
p
r
o
d
u
cin
g
to
p
-
r
an
k
f
ea
t
u
r
es
t
h
at
w
er
e
th
en
u
s
ed
as
n
e
w
tr
ain
in
g
d
ata
to
f
o
r
m
th
e
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
i
n
e
(
SVM
)
class
i
f
ier
m
o
d
el
[
1
1
]
.
T
h
e
f
i
n
a
l
s
tep
in
v
o
lv
ed
test
i
n
g
t
h
e
s
y
s
t
e
m
b
y
i
n
p
u
t
tin
g
th
e
tes
tin
g
d
ata
to
d
eter
m
i
n
e
th
e
a
cc
u
r
ac
y
o
f
t
h
e
s
y
s
te
m
.
Fig
u
r
e
1
.
Gen
er
al
s
y
s
te
m
f
lo
w
ch
ar
t
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
C
o
mp
a
r
a
tive
a
n
a
lysi
s
o
f
R
eliefF
-
S
V
M a
n
d
C
F
S
-
S
V
M fo
r
micro
a
r
r
a
y
...
(
Mo
ch
a
ma
d
A
g
u
s
ta
N
a
o
fa
l H
a
kim)
3395
2
.
1
.
T
he
da
t
a
s
et
s
T
h
e
d
ataset
u
s
ed
i
n
t
h
i
s
p
ap
er
is
m
icr
o
ar
r
a
y
d
ata
co
m
p
r
i
s
i
n
g
s
e
v
en
DN
A
m
icr
o
ar
r
ay
s
.
T
h
e
d
ataset
w
a
s
o
b
tain
ed
f
r
o
m
th
e
Ken
t
-
R
id
g
e
b
io
m
ed
ical
d
ata
r
ep
o
s
it
o
r
y
,
as
r
ep
o
r
ted
b
y
L
i
[
1
0
]
.
E
ac
h
m
icr
o
ar
r
ay
d
ata
w
a
s
g
r
o
u
p
ed
in
to
tr
ai
n
in
g
d
ata
an
d
test
i
n
g
d
ata
ac
co
r
d
in
g
to
t
h
e
r
atio
lis
ted
in
T
ab
le
1
.
T
h
is
d
iv
is
io
n
w
as d
o
n
e
b
y
d
ir
ec
tl
y
d
i
v
id
in
g
t
h
e
d
ata
r
an
d
o
m
l
y
,
alo
n
g
w
it
h
t
h
e
r
es
u
l
ts
o
f
t
h
e
d
is
tr
ib
u
tio
n
o
f
tr
ai
n
in
g
an
d
te
s
ti
n
g
d
ata
,
as
s
h
o
w
n
i
n
T
ab
le
1
.
T
ab
le
1
.
Dis
tr
ib
u
tio
n
o
f
d
ata
i
n
to
test
i
n
g
d
ata
an
d
tr
ain
i
n
g
d
ata
D
a
t
a
A
t
t
r
i
b
u
t
e
T
r
a
i
n
i
n
g
T
e
st
i
n
g
N
u
mb
e
r
o
f
C
l
a
ss
B
r
e
a
st
c
a
n
c
e
r
2
4
4
8
1
78
19
2
C
o
l
o
n
t
u
mo
r
2
0
0
0
43
19
2
O
v
a
r
i
a
n
c
a
n
c
e
r
1
5
1
5
4
1
7
7
76
2
L
u
n
g
c
a
n
c
e
r
1
2
5
3
3
1
4
9
32
2
P
r
o
st
a
t
e
c
a
n
c
e
r
1
2
6
0
0
1
0
2
34
2
M
L
L
l
e
u
k
e
mi
a
1
2
5
8
3
57
15
3
C
e
n
t
r
a
l
N
e
r
v
o
u
s Sy
st
e
m
7
1
2
9
42
18
2
No
r
m
a
lizatio
n
is
u
s
e
f
u
l
f
o
r
co
n
v
er
ti
n
g
attr
ib
u
te
v
al
u
es
in
to
a
r
an
g
e
b
et
w
ee
n
0
an
d
1
s
o
t
h
at
th
e
d
ata
w
il
l b
e
ea
s
ier
to
p
r
o
ce
s
s
[
1
2
]
.
T
h
e
n
o
r
m
al
izatio
n
w
a
s
d
o
n
e
u
s
in
g
(
1
)
[
1
0
]
.
′
=
−
−
(
1
)
E
q
u
atio
n
(
1
)
w
as
u
s
ed
to
ca
lcu
late
th
e
n
o
r
m
aliza
t
io
n
v
al
u
e,
w
h
er
e
′
is
th
e
r
esu
lt
o
f
n
o
r
m
a
lizatio
n
,
is
o
r
ig
i
n
al
v
a
lu
e
o
f
th
e
at
tr
ib
u
te,
is
th
e
s
m
alle
s
t
v
al
u
e
o
f
a
ll
th
e
d
ata
i
n
o
n
e
a
ttrib
u
te,
an
d
is
th
e
lar
g
est
v
alu
e
o
f
all
t
h
e
d
ata
in
o
n
e
attr
ib
u
te.
2
.
2
.
T
he
Relief
F
t
ec
hn
i
qu
e
T
h
e
in
itial
p
r
o
ce
s
s
o
f
s
elec
ti
n
g
th
e
R
elie
f
F
f
ea
tu
r
e
r
eq
u
ir
es
th
e
s
ep
ar
atio
n
o
f
tr
ai
n
in
g
d
ata
b
ased
o
n
class
a
n
d
t
h
e
d
eter
m
i
n
atio
n
o
f
th
e
n
u
m
b
er
o
f
p
o
s
s
ib
le
i
n
s
ta
n
ce
s
b
ased
o
n
it
s
cla
s
s
[
1
3
,
1
4
]
.
T
h
e
tr
ain
in
g
d
ata
w
a
s
s
ep
ar
ated
ac
co
r
d
in
g
to
it
s
class
to
f
ac
ilit
a
te
th
e
s
ea
r
ch
f
o
r
n
ea
r
-
h
it
an
d
n
ea
r
-
mi
s
s
f
o
r
ea
ch
r
an
d
o
m
l
y
s
elec
ted
in
s
ta
n
ce
,
w
h
er
e
n
ea
r
-
h
it
an
d
n
ea
r
-
m
is
s
s
ea
r
ch
w
er
e
ca
lcu
lated
b
ased
o
n
th
e
lev
el
o
f
s
i
m
ilar
it
y
o
f
d
ata
u
s
i
n
g
th
e
E
u
clid
ia
n
d
is
ta
n
ce
f
o
r
m
u
la
o
f
(
2
)
[
1
0
]
,
an
d
co
n
s
i
d
er
in
g
t
h
at
t
h
e
tr
ain
in
g
d
ata
w
a
s
n
u
m
er
ical
an
d
co
n
tin
u
o
u
s
.
(
,
)
=
√
∑
(
−
)
2
(
2
)
E
q
u
atio
n
(
2
)
w
as
u
s
ed
to
ca
lc
u
late
d
i
s
tan
ce
v
al
u
e,
w
h
er
e
an
d
ar
e
in
s
ta
n
ce
s
,
an
d
is
th
e
a
ttrib
u
te
v
alu
e.
T
h
e
n
ea
r
-
h
it
a
n
d
n
ea
r
-
m
is
s
v
al
u
es
d
eter
m
in
ed
w
er
e
th
en
u
s
ed
to
ca
lcu
late
t
h
e
w
e
ig
h
t
o
f
all
f
ea
t
u
r
e
s
b
ased
o
n
th
e
f
o
r
m
u
la
o
f
t
h
e
R
elief
F tec
h
n
iq
u
e
g
i
v
en
b
y
(
3
)
.
[
]
=
[
]
−
(
,
,
)
+
∑
[
(
)
∗
(
,
,
(
)
)
]
≠
(
)
(
3
)
w
h
er
e
is
th
e
w
ei
g
h
t
o
f
ea
c
h
f
ea
t
u
r
e,
is
t
h
e
f
ea
tu
r
e
t
h
at
i
s
b
ein
g
ca
lc
u
lated
,
is
th
e
s
ele
cte
d
r
an
d
o
m
in
s
ta
n
ce
,
is
t
h
e
n
ea
r
-
h
it
i
n
s
tan
ce
v
alu
e,
is
t
h
e
v
alu
e
o
f
n
ea
r
-
m
is
s
i
n
s
ta
n
ce
,
(
)
is
th
e
p
r
io
r
p
r
o
b
a
b
ilit
y
o
f
clas
s
,
an
d
is
t
h
e
n
u
m
b
er
o
f
r
ep
etitio
n
s
o
f
th
e
r
an
d
o
m
in
s
tan
ce
r
etr
ie
v
al
[
1
5
,
1
6
]
.
T
h
e
f
u
n
c
tio
n
(
,
1
,
2
)
w
as
u
s
ed
to
ca
lcu
late
t
h
e
v
al
u
e
o
f
u
s
i
n
g
th
e
d
if
f
er
en
ce
i
n
v
alu
e
b
et
w
ee
n
1
an
d
2
in
s
tan
ce
s
.
T
h
is
ca
lcu
latio
n
is
d
ef
i
n
ed
b
y
(
4
)
.
(
,
1
,
2
)
=
(
,
1
)
−
(
,
2
)
m
ax
(
)
−
m
i
n
(
)
(
4
)
w
h
er
e
is
th
e
f
ea
tu
r
e
t
h
at
i
s
b
ein
g
ca
lc
u
lated
,
1
is
th
e
s
ele
cted
r
an
d
o
m
i
n
s
ta
n
ce
,
an
d
2
is
th
e
n
ea
r
-
h
it
in
s
ta
n
ce
s
o
r
n
ea
r
-
m
is
s
in
s
ta
n
ce
s
[
1
7
]
.
T
h
e
ca
lcu
latio
n
o
f
p
o
s
s
ib
le
o
cc
u
r
r
en
ce
s
o
f
i
n
s
tan
ce
s
u
s
in
g
p
r
io
r
p
r
o
b
a
b
i
lit
y
ca
n
b
e
d
o
n
e
u
s
in
g
(
5
)
.
(
)
=
1
−
(
5
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
11
,
No
.
4
,
A
u
g
u
s
t 2
0
2
1
:
3
3
9
3
-
3402
3396
w
h
er
e
is
a
c
lass
t
h
at
w
i
ll c
alc
u
late
t
h
e
p
r
o
b
ab
ilit
y
o
f
o
cc
u
r
r
en
ce
w
h
ic
h
co
n
s
is
t
s
o
f
a
clas
s
p
r
o
b
ab
ilit
y
a
n
d
an
in
s
tan
ce
p
r
o
b
ab
ilit
y
.
T
h
e
f
o
llo
w
i
n
g
s
ec
tio
n
p
r
o
v
id
es
a
f
l
o
w
c
h
ar
t
d
escr
ib
in
g
t
h
e
s
y
s
te
m
an
d
p
r
o
ce
s
s
o
f
th
e
R
el
ief
F f
ea
t
u
r
e
s
elec
tio
n
te
ch
n
iq
u
e
as s
h
o
w
n
in
F
ig
u
r
e
2
.
Fig
u
r
e
2
.
R
elief
F f
ea
t
u
r
e
s
elec
tio
n
f
lo
w
c
h
ar
t
2
.
2
.
1
.
T
he
Relief
F
a
lg
o
rit
h
m
T
h
e
alg
o
r
ith
m
o
f
R
elie
f
F tec
h
n
iq
u
e
i
s
ad
o
p
ted
f
r
o
m
[
1
8
]
,
i
t is b
ased
o
n
s
ev
er
al
s
tep
s
as s
h
o
w
n
:
The Input
:
for each training instance a vector of
attribute values and the class value
Output
:
T
he vector w of estimations of the qualities of attributes.
1.
set weight
of features
[
]
:
=
0
.
0
2.
for
:
=
1
to m do
3.
select an instance
randomly;
4.
find k
-
nearest hits
ℎ
;
5.
for each
≠
(
)
do
6.
from class C find k nearest misses
(
)
;
7.
for
≔
1
to a do
8.
Calculate:
[
]
=
[
]
−
∑
[
(
,
,
ℎ
)
(
,
)
=
1
+
∑
1
−
(
)
∑
(
,
,
ℎ
)
]
=
1
≠
(
)
(
,
)
9. end.
2
.
3
.
T
he
CF
S t
ec
hn
iqu
e
T
h
e
C
FS
r
u
n
s
b
y
ca
lc
u
lati
n
g
c
o
m
p
o
n
en
t
s
b
ased
o
n
a
h
eu
r
i
s
ti
cs
v
al
u
e
ca
lled
t
h
e
v
alu
e,
w
h
i
ch
r
ep
r
esen
ts
th
e
q
u
alit
y
o
f
ea
c
h
f
ea
t
u
r
e
co
m
b
in
atio
n
o
r
f
ea
tu
r
e
s
u
b
s
et.
is
ca
lcu
lated
u
s
i
n
g
(
6
)
:
=
̅
̅
̅
̅
̅
√
+
(
−
1
)
̅
(
6
)
w
h
er
e
is
s
u
b
s
et
f
ea
tu
r
e
v
al
u
e,
is
n
u
m
b
er
o
f
f
ea
t
u
r
es,
̅
̅
̅
̅
is
a
v
e
r
ag
e
v
a
lu
e
o
f
cla
s
s
m
in
u
s
th
e
f
ea
t
u
r
e
co
r
r
elatio
n
,
an
d
̅
is
av
er
ag
e
v
al
u
e
o
f
f
ea
tu
r
e
m
i
n
u
s
th
e
f
ea
t
u
r
e
in
ter
co
r
r
elatio
n
[
6
]
.
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
C
o
mp
a
r
a
tive
a
n
a
lysi
s
o
f
R
eliefF
-
S
V
M a
n
d
C
F
S
-
S
V
M fo
r
micro
a
r
r
a
y
...
(
Mo
ch
a
ma
d
A
g
u
s
ta
N
a
o
fa
l H
a
kim)
3397
T
h
e
co
m
p
o
n
en
ts
o
f
in
clu
d
e
th
e
co
r
r
elatio
n
v
al
u
e
o
f
a
f
ea
tu
r
e
w
ith
o
t
h
er
f
ea
t
u
r
es
an
d
t
h
e
co
r
r
elatio
n
b
et
w
ee
n
f
ea
t
u
r
es
an
d
clas
s
es
o
w
n
ed
.
I
n
t
h
i
s
c
ase,
a
v
alu
e
s
ea
r
ch
w
as
d
o
n
e
u
s
i
n
g
t
h
e
f
o
r
w
ar
d
s
elec
tio
n
al
g
o
r
ith
m
,
s
tar
tin
g
w
it
h
a
b
lan
k
f
ea
t
u
r
e
s
u
b
s
et
u
n
til
th
e
b
est
co
m
b
in
a
ti
o
n
ac
co
r
d
in
g
to
t
h
e
th
r
es
h
o
ld
w
a
s
f
o
u
n
d
[
1
9
,
2
0
]
.
T
h
en
,
th
e
v
al
u
e
o
f
ea
ch
s
u
b
s
et
w
ill
b
e
ca
lcu
lated
,
a
n
d
s
u
b
s
et
w
it
h
t
h
e
b
ig
est
v
alu
e
w
i
ll b
e
s
elec
ted
.
Fig
u
r
e
3
s
h
o
w
s
t
h
e
C
FS
f
lo
w
c
h
ar
t.
Fig
u
r
e
3
.
C
FS
f
lo
w
ch
ar
t
Usi
n
g
R
elie
f
F
an
d
C
F
S,
th
e
f
e
atu
r
es
o
f
t
h
e
m
icr
o
ar
r
a
y
d
ata
w
er
e
o
b
tain
ed
.
T
h
ese
f
ea
t
u
r
es
w
er
e
th
e
n
u
s
ed
as
p
ar
a
m
eter
s
f
o
r
d
ev
el
o
p
in
g
t
h
e
cla
s
s
i
f
icat
io
n
m
o
d
el.
T
h
e
d
ev
elo
p
m
en
t
o
f
th
e
SVM
clas
s
i
f
icatio
n
m
o
d
el
i
n
cl
u
d
es
d
ev
elo
p
in
g
th
e
b
est
h
y
p
er
p
la
n
e
m
o
d
el
to
s
ep
ar
ate
th
e
d
ata
ac
co
r
d
in
g
to
ea
ch
class
,
b
ased
o
n
s
u
p
p
o
r
tin
g
p
o
in
t
s
o
r
s
u
p
p
o
r
t
v
ec
to
r
s
th
at
ar
e
at
t
h
e
cla
s
s
s
ep
ar
atio
n
li
m
it
[
2
1
,
2
2
]
.
T
h
e
s
elec
tio
n
o
f
p
o
in
t
s
a
s
s
u
p
p
o
r
t
v
ec
to
r
s
w
a
s
i
n
f
l
u
e
n
ce
d
b
y
th
e
s
h
ap
e
an
d
c
h
ar
ac
ter
o
r
co
n
d
itio
n
o
f
th
e
f
ea
t
u
r
es
o
f
t
h
e
d
ata.
B
y
g
etti
n
g
th
e
b
es
t
f
ea
t
u
r
es,
t
h
e
m
ar
g
in
o
n
t
h
e
s
u
p
p
o
r
t
v
ec
to
r
ca
n
b
e
m
a
x
i
m
ized
[
2
3
-
2
5
]
.
E
q
u
ati
o
n
(
7
)
p
r
esen
t
s
t
h
e
f
u
n
ctio
n
th
a
t
m
u
s
t
b
e
o
p
tim
ized
f
r
o
m
th
e
h
y
p
er
p
la
n
e
m
ar
g
in
w
h
er
e
w
is
t
h
e
u
n
it
v
ec
to
r
f
o
u
n
d
i
n
th
e
h
y
p
er
p
lan
e.
m
i
n
w
⃗
⃗
⃗
τ
(
)
=
1
2
|
|
⃗
⃗
|
|
2
(
7
)
Ho
w
e
v
er
,
b
ef
o
r
e
u
s
in
g
th
e
d
ata
to
d
ev
elo
p
th
e
m
o
d
el,
th
e
d
ataset
w
a
s
f
ir
s
t
d
iv
id
ed
in
to
test
in
g
d
ata
an
d
tr
ain
in
g
d
ata.
B
ec
au
s
e
th
e
test
in
g
d
ata
co
n
s
is
t
s
o
f
2
an
d
3
class
es,
th
e
t
y
p
e
o
f
k
er
n
el
to
b
e
u
s
ed
f
o
r
SVM
w
a
s
f
ir
s
t
te
s
ted
ag
ai
n
s
t
al
l
d
a
ta.
T
h
e
k
er
n
el
t
h
at
s
h
o
w
ed
g
o
o
d
co
m
p
atib
ilit
y
an
d
g
o
o
d
ac
cu
r
ac
y
f
o
r
all
d
ata
w
a
s
th
e
n
u
s
ed
as o
n
e
o
f
t
h
e
p
ar
a
m
eter
s
f
o
r
SVM.
2
.
3
.
1
.
T
he
CSF
a
lg
o
rit
h
m
T
h
e
alg
o
r
ith
m
o
f
C
SF
tech
n
i
q
u
e
u
s
i
n
g
f
o
r
w
ar
d
s
elec
tio
n
b
eg
in
s
w
it
h
th
e
ze
r
o
v
alu
e
o
f
s
u
b
s
et
th
e
n
ad
d
a
f
ea
tu
r
e
o
n
e
b
y
o
n
e
an
d
ca
lcu
late
th
e
v
alu
e
of
ea
ch
f
ea
t
u
r
es
co
m
b
in
atio
n
g
r
ee
d
i
ly
[
1
9
]
.
I
t
is
b
ased
o
n
s
ev
er
al
s
tep
s
a
s
s
h
o
w
n
:
The Input
:
F
or each training instance a vector of attribute values and the class value
.
Output
:
maximum subset
value.
1.
set subset S := [ ], threshold := determined by writer
2. while threshold != 0
3.
for each attribute
4.
̅
̅
̅
̅
∶
=
average attribut relation with class
5.
̅
∶
=
average attribut relation with another attribut
6.
∶
=
count number of subset
7.
=
̅
̅
̅
̅
̅
√
+
(
−
1
)
̅
8.
if maximum
value has finded then
9.
reintialize threshold number to determined value
10.
else threshold minus 1
11. end.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
11
,
No
.
4
,
A
u
g
u
s
t 2
0
2
1
:
3
3
9
3
-
3402
3398
3.
RE
SU
L
T
S
A
ND
D
I
SCU
SS
I
O
N
I
n
r
esu
lt
s
an
d
d
is
cu
s
s
io
n
s
s
ec
t
io
n
,
w
i
ll
d
is
cu
s
s
ab
o
u
t
r
es
u
lts
o
f
r
esear
ch
.
T
h
e
r
esu
lt
ar
e
d
i
v
id
ed
in
t
o
th
r
ee
s
ce
n
ar
io
s
a
n
d
p
r
esen
ted
i
n
a
f
i
g
u
r
e
an
d
tab
le.
3
.
1
.
Scena
rio
1
I
n
th
is
te
s
tin
g
p
r
o
ce
s
s
,
a
co
m
p
ar
is
o
n
w
a
s
m
ad
e
b
et
w
ee
n
th
e
k
er
n
el
s
to
b
e
u
s
ed
in
th
e
SVM
class
if
ier
ac
co
r
d
in
g
to
th
e
d
ata
d
is
tr
ib
u
tio
n
a
n
d
to
p
r
o
v
id
e
th
e
m
o
s
t
o
p
ti
m
al
ac
cu
r
ac
y
r
es
u
lt
s
.
T
h
e
k
er
n
els
ch
o
s
e
n
in
cl
u
d
ed
th
e
R
B
F,
P
o
ly
n
o
m
ial
,
an
d
L
in
ea
r
k
er
n
els
w
it
h
p
ar
am
eter
s
(
d
)
d
eg
r
ee
s
w
it
h
a
v
al
u
e
o
f
3
an
d
(
C
)
w
it
h
a
v
al
u
e
o
f
1
,
w
h
i
le
t
h
e
te
s
ti
n
g
d
ata
co
n
s
i
s
t
ed
o
f
7
ca
n
ce
r
d
ata.
T
ab
le
2
s
h
o
w
s
th
e
r
es
u
l
ts
o
f
t
h
e
3
t
y
p
es
o
f
k
er
n
el
e
x
p
er
i
m
e
n
ts
f
o
r
SVM
u
s
in
g
m
icr
o
ar
r
ay
d
ata.
T
ab
le
2
.
T
h
e
r
esu
lts
o
f
te
s
ti
n
g
d
ata
u
s
i
n
g
t
h
e
t
h
r
ee
SVM
k
er
n
els
K
e
r
n
e
l
A
c
c
u
r
a
c
y
(
%)
C
o
l
o
n
t
u
mo
r
L
u
n
g
c
a
n
c
e
r
P
r
o
st
a
t
e
M
L
L
l
e
u
k
e
mi
a
B
r
e
a
st
c
a
n
c
e
r
O
v
a
r
i
a
n
C
e
n
t
r
a
l
n
e
r
v
o
u
s
sy
st
e
m
A
v
e
r
a
g
e
RBF
7
8
,
9
5
50
50
80
3
6
,
8
4
8
8
,
1
6
7
2
,
2
2
6
5
,
1
7
P
o
l
y
n
o
mi
a
l
6
8
,
4
2
50
7
3
,
5
3
7
3
,
3
3
3
6
,
8
4
6
8
,
4
2
6
1
,
1
1
6
1
,
6
7
L
i
n
e
a
r
7
3
,
6
8
8
4
,
3
6
9
4
,
1
2
8
6
,
6
7
8
4
,
2
1
1
0
0
6
1
,
1
1
8
3
,
4
5
I
n
test
i
n
g
t
h
e
s
y
s
te
m
w
it
h
th
e
f
ir
s
t
te
s
ti
n
g
s
ce
n
ar
io
,
th
e
ac
c
u
r
ac
y
,
a
s
s
h
o
w
n
in
T
ab
le
2
,
in
d
icate
s
t
h
at
th
e
m
o
s
t
s
u
itab
le
S
VM
k
er
n
e
l
f
o
r
all
test
i
n
g
d
ata
w
as
t
h
e
l
in
ea
r
k
er
n
el,
w
h
ich
r
et
u
r
n
ed
t
h
e
h
i
g
h
est
a
v
er
ag
e
ac
cu
r
ac
y
o
f
8
3
.
4
5
%,
w
h
ile
t
h
e
R
B
F
k
er
n
el
p
r
o
d
u
ce
d
an
av
er
ag
e
ac
cu
r
ac
y
o
f
6
5
.
1
7
%
an
d
th
e
P
o
ly
n
o
m
ial
k
er
n
el
r
ea
c
h
ed
6
1
.
6
7
%
ac
cu
r
ac
y
o
n
l
y
.
A
lt
h
o
u
g
h
th
e
lin
ea
r
k
er
n
el
p
r
o
d
u
ce
d
g
o
o
d
av
er
ag
e
ac
cu
r
ac
y
,
i
n
tes
ts
co
n
d
u
cted
o
n
s
o
m
e
d
ata
s
u
c
h
as
th
e
ce
n
tr
al
n
er
v
o
u
s
s
y
s
te
m
an
d
co
lo
n
t
u
m
o
r
d
ata,
it
s
ti
ll
p
r
o
d
u
c
ed
a
lo
w
er
ac
cu
r
ac
y
co
m
p
ar
ed
to
th
e
R
B
F
k
er
n
e
l.
Fro
m
t
h
e
r
es
u
lts
,
i
t
ca
n
b
e
co
n
clu
d
ed
t
h
at
m
o
s
t
o
f
th
e
d
is
tr
ib
u
tio
n
o
f
d
ata
u
s
ed
in
th
i
s
tes
t
f
o
llo
w
ed
a
lin
ea
r
o
r
lin
ea
r
s
ep
ar
ab
le
d
ata
d
is
tr
ib
u
tio
n
,
e
x
ce
p
t
f
o
r
s
o
m
e
d
ata
s
u
ch
a
s
t
h
e
ce
n
tr
al
n
er
v
o
u
s
s
y
s
te
m
a
n
d
co
lo
n
tu
m
o
r
d
ata.
3
.
2
.
Scena
rio
2
T
h
e
s
ec
o
n
d
test
i
n
g
s
ce
n
ar
io
p
r
o
d
u
ce
d
v
alu
es
i
n
t
h
e
f
o
r
m
o
f
an
o
p
ti
m
al
n
u
m
b
er
o
f
f
e
atu
r
es,
t
h
e
h
ig
h
e
s
t
ac
cu
r
ac
y
,
an
d
th
e
a
v
er
ag
e
ac
cu
r
ac
y
f
o
r
ea
ch
f
ea
t
u
r
e
s
u
b
s
et
s
elec
tio
n
,
as
o
u
tli
n
ed
in
T
ab
le
3
.
A
th
r
es
h
o
ld
p
ar
am
eter
w
it
h
a
v
a
lu
e
o
f
3
0
0
w
as
i
n
s
er
ted
i
n
th
i
s
p
ap
er
.
A
th
r
es
h
o
ld
is
u
s
ed
as
a
d
eli
m
iter
o
f
t
h
e
n
u
m
b
er
o
f
f
ea
t
u
r
es
t
h
at
w
ill
b
e
test
ed
as
a
f
ea
tu
r
e
s
u
b
s
et.
Fiv
e
f
ea
tu
r
es
w
er
e
ta
k
en
i
n
a
s
u
b
s
et
u
n
til
t
h
e
th
r
es
h
o
ld
w
as
d
eter
m
i
n
ed
an
d
th
e
n
u
m
b
er
w
a
s
s
ea
r
ch
ed
.
T
h
e
f
ea
t
u
r
e
s
u
b
s
et
s
h
o
u
ld
p
r
o
v
id
e
th
e
m
o
s
t
o
p
ti
m
al
ac
cu
r
ac
y
.
B
esid
es
t
h
e
p
ar
a
m
e
ter
in
th
e
f
o
r
m
o
f
a
t
h
r
es
h
o
ld
,
th
e
n
u
m
b
er
o
f
Nea
r
es
t
I
n
s
ta
n
ce
to
b
e
u
s
ed
w
as
also
test
ed
i
n
t
h
is
s
ce
n
ar
io
,
wh
er
e
th
e
n
ea
r
est
in
s
ta
n
ce
v
a
lu
e
w
a
s
c
h
o
s
en
r
a
n
d
o
m
l
y
f
r
o
m
1
,
2
,
o
r
3
.
T
h
e
C
F
S
th
r
es
h
o
ld
p
ar
a
m
eter
v
alu
e
s
o
f
3
an
d
5
w
er
e
i
n
s
er
ted
i
n
th
i
s
p
ap
er
.
T
h
e
C
FS
t
h
r
es
h
o
ld
p
ar
am
eter
w
a
s
u
s
ed
a
s
a
li
m
it
to
allo
w
t
h
e
f
o
r
w
ar
d
s
el
ec
tio
n
alg
o
r
ith
m
to
w
id
en
its
s
ea
r
ch
if
t
h
e
n
e
w
r
esu
lts
w
er
e
less
er
th
a
n
th
e
r
esu
lt
s
o
f
p
r
ev
io
u
s
.
Oth
e
r
w
i
s
e,
an
o
th
er
s
u
b
s
et
w
ill
b
e
g
iv
e
n
th
e
ch
a
n
ce
to
ad
d
m
e
m
b
er
s
in
th
e
f
o
r
m
o
f
n
e
w
f
ea
tu
r
e
s
.
T
ab
le
3
.
T
est r
esu
lts
o
f
SV
M+
R
elie
f
F a
n
d
S
VM
+
C
FS
D
a
t
a
S
V
M
+
R
e
l
i
e
f
F
S
V
M
+
C
F
S
A
c
c
u
r
a
c
y
(
%)
A
t
t
r
i
b
u
t
e
A
c
c
u
r
a
c
y
(
%)
A
t
t
r
i
b
u
t
e
C
o
l
o
n
t
u
mo
r
1
0
0
15
8
9
,
4
7
13
L
u
n
g
c
a
n
c
e
r
9
6
,
8
8
10
8
4
,
3
8
48
P
r
o
st
a
t
e
c
a
n
c
e
r
9
4
,
1
2
1
7
0
8
2
,
3
5
46
O
v
a
r
i
a
n
c
a
n
c
e
r
1
0
0
1
3
0
9
8
,
6
8
50
B
r
e
a
st
c
a
n
c
e
r
8
4
,
2
1
65
6
3
,
1
6
51
M
L
L
l
e
u
k
e
mi
a
1
0
0
5
8
6
,
6
7
19
C
e
n
t
r
a
l
n
e
r
v
o
u
s sy
st
e
m
8
8
,
8
9
1
2
5
8
3
,
3
3
12
B
ased
o
n
T
ab
le
3
,
w
h
e
n
te
s
ti
n
g
t
h
e
S
VM
clas
s
i
f
icatio
n
tec
h
n
iq
u
e
u
s
in
g
t
h
e
R
elie
f
F
s
elec
ti
o
n
f
ea
t
u
r
e,
th
e
co
lo
n
t
u
m
o
r
d
ata
w
as
c
las
s
if
ied
w
it
h
a
n
ac
c
u
r
ac
y
o
f
1
0
0
%
w
it
h
1
5
f
ea
tu
r
es
a
n
d
3
n
e
ar
est
i
n
s
ta
n
ce
s
.
T
h
e
lu
n
g
ca
n
ce
r
d
ata
w
as
cla
s
s
i
f
ie
d
w
it
h
9
6
.
8
8
%
ac
cu
r
ac
y
w
i
th
1
0
f
ea
tu
r
es
an
d
3
n
ea
r
est
i
n
s
t
an
ce
s
.
T
h
e
o
v
ar
ian
ca
n
ce
r
d
ata
w
as
clas
s
i
f
ied
w
it
h
1
0
0
%
ac
cu
r
ac
y
i
n
al
l
n
u
m
b
er
o
f
i
n
s
ta
n
ce
s
,
w
it
h
t
h
e
lea
s
t
f
ea
t
u
r
es
b
ei
n
g
1
3
0
f
ea
t
u
r
es.
T
h
e
p
r
o
s
tate
ca
n
ce
r
d
ata
w
as
c
lass
if
ied
w
i
th
a
m
a
x
i
m
u
m
ac
c
u
r
ac
y
o
f
9
4
.
1
2
%
in
al
l
i
n
s
ta
n
ce
s
a
n
d
1
7
0
as th
e
m
i
n
i
m
u
m
n
u
m
b
er
o
f
f
ea
tu
r
e
s
.
B
r
ea
s
t
ca
n
ce
r
d
ata
w
a
s
clas
s
i
f
ied
w
it
h
8
4
.
2
1
% a
cc
u
r
ac
y
w
i
th
at
least
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
C
o
mp
a
r
a
tive
a
n
a
lysi
s
o
f
R
eliefF
-
S
V
M a
n
d
C
F
S
-
S
V
M fo
r
micro
a
r
r
a
y
...
(
Mo
ch
a
ma
d
A
g
u
s
ta
N
a
o
fa
l H
a
kim)
3399
6
5
f
ea
tu
r
es
an
d
a
to
tal
o
f
3
n
ea
r
est
i
n
s
ta
n
ce
s
.
T
h
e
ML
L
l
eu
k
e
m
ia
d
ata
w
it
h
3
class
e
s
w
a
s
class
if
ied
w
it
h
1
0
0
%
ac
cu
r
ac
y
at
all
n
ea
r
es
t
in
s
tan
ce
s
,
w
it
h
5
f
ea
tu
r
e
s
m
in
i
m
u
m
a
t
t
h
e
n
ea
r
e
s
t
i
n
s
ta
n
c
es
o
f
1
a
n
d
2
.
T
h
e
ce
n
tr
al
n
er
v
o
u
s
s
y
s
te
m
d
ata
was
clas
s
i
f
ied
w
it
h
t
h
e
h
i
g
h
e
s
t
ac
cu
r
ac
y
o
f
8
8
.
8
9
%
at
th
e
n
e
ar
est
in
s
ta
n
ce
s
o
f
2
an
d
3
an
d
a
m
in
i
m
u
m
n
u
m
b
er
o
f
f
ea
t
u
r
es o
f
1
2
5
at
in
s
ta
n
ce
2
.
Fro
m
t
h
e
tes
tin
g
s
ce
n
ar
io
s
co
n
d
u
cted
,
it c
an
b
e
s
aid
t
h
at
u
s
i
n
g
t
h
e
R
e
lief
F
f
ea
t
u
r
e
s
elec
tio
n
o
n
all
t
h
e
d
ata
s
u
cc
e
s
s
f
u
ll
y
p
r
o
v
id
ed
m
o
r
e
o
p
tim
al
r
esu
lts
,
co
m
p
ar
e
d
w
it
h
s
ce
n
ar
io
s
w
it
h
o
u
t
u
s
i
n
g
R
el
ief
F.
T
h
is
i
s
ev
id
en
ce
d
b
y
th
e
i
n
cr
ea
s
ed
a
cc
u
r
ac
y
w
it
h
f
ea
tu
r
e
r
ed
u
ct
io
n
b
u
t
f
o
r
s
o
m
e
d
ata
in
p
ar
ticu
lar
s
u
c
h
as
co
lo
n
tu
m
o
r
,
l
u
n
g
ca
n
ce
r
,
M
L
L
leu
k
e
m
ia,
an
d
th
e
ce
n
tr
al
n
er
v
o
u
s
s
y
s
te
m
d
ata,
w
h
ic
h
s
a
w
i
n
cr
ea
s
ed
class
i
f
icatio
n
ac
cu
r
ac
y
i
n
all
test
s
ce
n
ar
io
s
.
T
h
e
n
u
m
b
er
o
f
f
ea
t
u
r
es
w
as
n
o
t
t
h
e
o
n
l
y
p
ar
a
m
eter
t
h
at
af
f
ec
ted
cla
s
s
i
f
icatio
n
ac
c
u
r
ac
y
.
I
f
t
h
e
n
ea
r
est
in
s
tan
ce
p
ar
a
m
eter
f
o
r
s
o
m
e
d
ata
w
er
e
to
b
e
o
b
s
er
v
ed
s
u
c
h
a
s
co
lo
n
tu
m
o
r
,
p
r
o
s
t
ate
ca
n
ce
r
,
o
v
ar
ian
ca
n
ce
r
,
lu
n
g
ca
n
ce
r
,
M
L
L
le
u
k
e
m
ia,
an
d
th
e
ce
n
tr
al
n
er
v
o
u
s
s
y
s
te
m
d
ata,
an
i
n
cr
ea
s
e
in
th
e
n
u
m
b
er
o
f
t
h
e
n
ea
r
est
i
n
s
ta
n
c
e
s
r
es
u
lted
in
an
ac
cu
r
ac
y
t
h
at
w
a
s
n
o
t
lo
w
er
t
h
a
n
th
e
ac
cu
r
ac
y
w
it
h
t
h
e
s
m
aller
n
ea
r
est
in
s
ta
n
ce
,
alt
h
o
u
g
h
t
h
e
b
r
ea
s
t
c
an
ce
r
d
ata
at
th
e
s
ec
o
n
d
n
ea
r
est
i
n
s
ta
n
ce
h
ad
a
lo
w
er
ac
c
u
r
a
c
y
t
h
an
t
h
e
n
ea
r
est
in
s
ta
n
ce
o
f
1
.
Ho
w
ev
er
,
i
n
t
h
e
th
ir
d
i
n
s
ta
n
ce
,
t
h
e
ac
cu
r
ac
y
ag
ain
in
cr
ea
s
ed
an
d
w
a
s
g
r
ea
t
er
th
an
th
e
s
ec
o
n
d
in
s
ta
n
ce
.
T
h
is
is
li
k
el
y
b
ec
au
s
e
th
er
e
i
s
a
n
ea
r
es
t i
n
s
ta
n
ce
th
at
ca
lcu
la
tes
th
e
w
ei
g
h
t
o
f
th
e
f
ea
t
u
r
e
t
h
at
m
ad
e
it
less
o
p
ti
m
al,
s
o
th
e
to
p
f
ea
t
u
r
es th
at
h
ad
w
e
ig
h
t
w
er
e
th
e
m
o
s
t lik
e
l
y
to
co
n
tai
n
n
o
is
e.
3
.
3
.
Scena
rio
3
Fo
r
th
e
th
ir
d
te
s
ti
n
g
s
ce
n
ar
io
s
h
o
w
n
in
T
ab
le
4
,
an
in
cr
ea
s
e
in
ac
cu
r
ac
y
w
as
o
b
s
er
v
ed
in
al
m
o
s
t
al
l
m
icr
o
ar
r
a
y
d
ata
a
f
ter
t
h
e
s
ele
ctio
n
f
ea
t
u
r
e
tec
h
n
iq
u
e
w
a
s
a
p
p
lied
.
T
h
e
co
lo
n
t
u
m
o
r
d
ata
w
a
s
cla
s
s
i
f
ied
w
i
th
1
0
0
%
an
d
8
9
.
4
7
%
ac
cu
r
ac
y
u
s
i
n
g
t
h
e
R
elie
f
F
tec
h
n
iq
u
e
a
n
d
th
e
C
FS
tech
n
iq
u
e,
r
e
s
p
ec
tiv
el
y
.
T
h
e
C
e
n
tr
al
Ner
v
o
u
s
S
y
s
te
m
d
ata
w
as
i
n
it
iall
y
clas
s
i
f
ied
w
ith
a
n
ac
cu
r
a
c
y
o
f
6
1
.
1
1
%,
w
h
ic
h
in
cr
ea
s
e
d
to
8
8
.
8
9
%
u
s
i
n
g
R
elie
f
F
a
n
d
8
3
.
3
3
%
u
s
i
n
g
C
F
S.
B
esid
es
t
h
at,
t
h
e
cla
s
s
i
f
icat
io
n
ac
cu
r
ac
y
o
f
l
u
n
g
ca
n
ce
r
,
p
r
o
s
tate
ca
n
ce
r
,
a
n
d
ML
L
le
u
k
e
m
ia
d
ata
also
i
n
cr
ea
s
ed
u
s
i
n
g
t
h
e
R
e
lie
f
F
tec
h
n
iq
u
e,
b
u
t
t
h
e
C
FS
tech
n
iq
u
e
p
r
o
d
u
ce
d
th
e
s
a
m
e
ac
cu
r
ac
y
f
o
r
s
o
m
e
d
ata
-
t
h
e
s
a
m
e
as
t
h
at
o
f
test
in
g
w
it
h
o
u
t
f
ea
tu
r
e
s
elec
ti
o
n
i.e
.
8
4
.
3
7
%
f
o
r
lu
n
g
ca
n
ce
r
d
ata,
an
d
8
6
.
6
7
%
f
o
r
ML
L
le
u
k
e
m
ia
d
ata.
Dec
r
ea
s
ed
ac
cu
r
ac
y
w
as
o
b
s
er
v
ed
i
n
s
o
m
e
d
atasets
s
u
c
h
as
f
r
o
m
8
4
.
2
1
%
to
6
3
.
1
6
%
f
o
r
b
r
ea
s
t
ca
n
ce
r
d
ata
;
f
r
o
m
1
0
0
%
to
9
8
.
6
8
%
f
o
r
o
v
ar
ia
n
ca
n
ce
r
d
ata;
a
n
d
f
r
o
m
9
4
.
1
2
%
to
8
2
.
3
6
% f
o
r
p
r
o
s
tat
e
ca
n
ce
r
d
at
a.
T
ab
le
4
.
R
ec
ap
itu
latio
n
o
f
t
h
e
r
esu
lt
s
o
f
SVM,
R
elie
f
F
-
SVM
an
d
C
FS
–
SV
M
test
i
n
g
D
a
t
a
A
c
c
u
r
a
c
y
(
%)
S
V
M
S
V
M
+
R
e
l
i
e
f
F
S
V
M
+
C
F
S
C
o
l
o
n
t
u
mo
r
7
3
,
6
8
1
0
0
8
9
,
4
7
L
u
n
g
c
a
n
c
e
r
8
4
,
3
8
9
6
,
8
8
8
4
,
3
8
P
r
o
st
a
t
e
c
a
n
c
e
r
9
4
,
1
2
9
4
,
1
2
8
2
,
3
5
O
v
a
r
i
a
n
c
a
n
c
e
r
1
0
0
1
0
0
9
8
,
6
8
B
r
e
a
st
c
a
n
c
e
r
8
4
,
2
1
8
4
,
2
1
6
3
,
1
6
M
L
L
l
e
u
k
e
mi
a
8
6
,
6
7
1
0
0
8
6
,
6
7
C
e
n
t
r
a
l
n
e
r
v
o
u
s sy
st
e
m
6
1
,
1
1
8
8
,
8
9
8
3
,
3
3
Fro
m
T
ab
le
4
an
Fig
u
r
e
4
,
th
e
SVM
clas
s
i
f
icatio
n
tech
n
iq
u
e
u
s
i
n
g
C
F
S
f
o
r
co
lo
n
tu
m
o
r
d
ata
p
r
o
d
u
ce
d
an
ac
cu
r
ac
y
o
f
8
9
.
4
7
%
w
it
h
1
6
f
ea
t
u
r
es
an
d
a
th
r
esh
o
ld
o
f
3
.
f
o
r
th
e
lu
n
g
ca
n
ce
r
d
ata
,
class
if
icatio
n
ac
cu
r
ac
y
r
ea
ch
ed
8
4
.
3
7
5
%
w
i
th
4
5
f
ea
t
u
r
es
at
a
t
h
r
es
h
o
ld
v
alu
e
o
f
3
;
o
v
ar
ia
n
ca
n
ce
r
d
ata
ac
h
iev
ed
9
8
.
6
8
%
ac
cu
r
ac
y
f
o
r
all
n
u
m
b
er
s
o
f
i
n
s
ta
n
ce
s
,
w
it
h
t
h
e
lea
s
t
n
u
m
b
er
o
f
f
ea
tu
r
e
s
(
3
0
)
:
p
r
o
s
tate
ca
n
ce
r
d
ata
y
ield
ed
a
m
ax
i
m
u
m
ac
c
u
r
ac
y
o
f
u
p
to
8
2
.
3
5
%
at
a
th
r
es
h
o
ld
o
f
5
w
it
h
4
6
f
ea
tu
r
es
;
b
r
ea
s
t
ca
n
ce
r
d
ata
o
b
tain
ed
6
3
.
1
6
%
ac
cu
r
ac
y
w
it
h
5
1
f
ea
tu
r
e
s
at
a
th
r
esh
o
ld
v
al
u
e
o
f
5
;
M
L
L
leu
k
e
m
ia
d
ata
w
it
h
3
class
es
o
b
tain
ed
an
ac
cu
r
ac
y
o
f
8
6
.
6
7
%
at
a
th
r
esh
o
ld
o
f
5
w
it
h
1
9
f
ea
t
u
r
es;
a
n
d
t
h
e
ce
n
t
r
al
n
er
v
o
u
s
s
y
s
te
m
d
ata
ac
h
ie
v
ed
an
ac
c
u
r
ac
y
o
f
7
2
.
2
2
% w
it
h
1
2
f
ea
t
u
r
es,
f
o
r
all
th
r
es
h
o
ld
s
.
Fro
m
t
h
e
te
s
ti
n
g
s
ce
n
ar
io
s
c
o
n
d
u
cted
,
u
s
i
n
g
C
F
S
f
ea
t
u
r
e
s
elec
tio
n
f
o
r
s
o
m
e
d
ata
s
u
c
h
as
M
L
L
leu
k
e
m
ia
an
d
l
u
n
g
ca
n
ce
r
p
r
o
d
u
ce
d
m
o
r
e
o
p
ti
m
al
r
es
u
lts
co
m
p
ar
ed
to
o
n
l
y
SV
M.
M
ea
n
w
h
ile,
t
h
e
co
lo
n
tu
m
o
r
an
d
ce
n
tr
al
n
er
v
o
u
s
S
y
s
te
m
d
ata
also
p
r
o
d
u
ce
d
m
o
r
e
o
p
ti
m
al
r
esu
l
ts
an
d
e
x
p
er
ien
ce
d
in
cr
ea
s
ed
ac
cu
r
ac
y
co
m
p
ar
ed
to
th
e
SVM
s
ce
n
ar
io
s
o
n
l
y
,
as
s
h
o
w
n
b
y
t
h
e
f
e
w
er
n
u
m
b
er
o
f
f
ea
t
u
r
e
s
u
s
ed
,
b
u
t
ac
cu
r
ac
y
w
a
s
s
t
ill
m
ai
n
tai
n
ed
an
d
i
m
p
r
o
v
ed
.
Ho
w
e
v
er
,
in
s
o
m
e
d
ata
s
u
c
h
as
b
r
ea
s
t c
an
ce
r
,
p
r
o
s
tate
ca
n
ce
r
,
an
d
o
v
ar
ia
n
ca
n
ce
r
d
ata,
in
th
e
te
s
ti
n
g
s
ce
n
ar
io
s
,
th
e
ac
c
u
r
ac
y
o
f
ea
c
h
ex
p
er
i
m
e
n
t
d
ec
r
ea
s
ed
.
T
h
is
is
lik
el
y
b
ec
au
s
e
t
h
e
se
ar
ch
p
r
o
ce
s
s
u
s
ed
a
Gr
ee
d
y
f
o
r
w
ar
d
s
elec
tio
n
tech
n
iq
u
e
th
at
w
a
s
n
o
t
o
p
ti
m
al,
as
ca
n
b
e
p
r
o
v
en
f
r
o
m
t
h
e
r
esu
lts
o
f
t
h
e
n
u
m
b
e
r
o
f
f
ea
tu
r
e
s
o
b
tain
ed
b
y
th
e
R
elie
f
F
tec
h
n
iq
u
e
f
o
r
t
h
e
th
r
e
e
d
ata
ab
o
v
e,
w
h
ic
h
w
a
s
a
lar
g
er
n
u
m
b
er
t
h
an
th
e
o
th
er
d
ata,
w
h
ile
th
e
m
a
x
i
m
u
m
f
ea
t
u
r
e
s
u
b
s
et
f
o
u
n
d
b
y
C
FS
w
as
lo
w
er
i
n
co
m
p
ar
is
o
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
11
,
No
.
4
,
A
u
g
u
s
t 2
0
2
1
:
3
3
9
3
-
3402
3400
Fig
u
r
e
4
.
C
o
m
p
ar
is
o
n
r
es
u
lt
s
o
f
SVM,
SVM
–
R
e
lie
f
F,
an
d
S
VM
–
C
FS
te
s
ti
n
g
As
f
o
r
th
e
e
f
f
ec
t
o
f
t
h
r
es
h
o
ld
o
n
th
e
ac
cu
r
ac
y
p
r
o
d
u
ce
d
,
alm
o
s
t
all
d
ata
w
it
h
m
o
r
e
t
h
r
esh
o
ld
s
h
ad
an
ac
cu
r
ac
y
th
at
w
a
s
eq
u
a
l
o
r
b
e
tter
th
a
n
s
m
aller
t
h
r
es
h
o
ld
v
al
u
es.
Hen
ce
,
it
ca
n
b
e
co
n
cl
u
d
ed
th
at
a
t
h
r
es
h
o
ld
v
alu
e
o
f
5
s
till
allo
w
ed
f
o
r
i
n
c
r
ea
s
ed
ac
cu
r
ac
y
ev
e
n
w
it
h
an
ad
d
itio
n
al
n
u
m
b
er
o
f
f
ea
tu
r
e
s
.
Ho
w
e
v
er
,
th
e
lu
n
g
ca
n
ce
r
d
ata
th
r
es
h
o
ld
o
f
3
h
ad
b
etter
ac
cu
r
ac
y
t
h
a
n
a
th
r
e
s
h
o
ld
o
f
5
,
lik
el
y
d
u
e
to
t
h
e
p
r
es
en
ce
o
f
n
o
is
e
w
h
e
n
th
e
th
r
e
s
h
o
ld
w
as r
ed
u
ce
d
.
B
esid
es
th
at,
th
e
e
f
f
ec
t
o
f
ac
c
u
r
ac
y
g
e
n
er
ated
f
r
o
m
th
e
R
eli
ef
F
-
SVM
test
in
g
s
ch
e
m
e
ca
n
b
e
s
aid
to
b
e
s
tab
le
ag
ai
n
s
t
t
h
e
r
atio
o
f
v
ar
io
u
s
d
ata
-
s
o
lv
in
g
r
atio
s
.
T
h
is
is
ev
id
e
n
t
f
o
r
th
e
d
ata
w
it
h
a
r
atio
o
f
7
0
%
s
u
c
h
as
th
e
o
v
ar
ian
ca
n
ce
r
,
co
lo
n
t
u
m
o
r
,
an
d
t
h
e
ce
n
tr
al
n
er
v
o
u
s
s
y
s
te
m
tr
ain
in
g
d
ata,
w
h
ich
p
r
o
d
u
ce
d
i
n
cr
ea
s
ed
ac
cu
r
ac
y
.
Fo
r
lu
n
g
ca
n
ce
r
an
d
ML
L
le
u
k
e
m
ia
d
ata,
w
i
th
m
o
r
e
tr
ai
n
in
g
d
ata
ac
co
r
d
in
g
to
a
r
atio
o
f
8
0
%
to
8
2
%,
an
in
cr
ea
s
e
in
ac
c
u
r
ac
y
co
u
ld
also
b
e
ac
h
ie
v
ed
.
I
n
ad
d
itio
n
,
alt
h
o
u
g
h
th
er
e
w
a
s
n
o
i
n
cr
ea
s
e
i
n
ac
cu
r
ac
y
f
o
r
th
e
p
r
o
s
tate
ca
n
ce
r
an
d
b
r
ea
s
t
ca
n
ce
r
d
ata
u
s
in
g
a
tr
ain
i
n
g
d
ata
r
atio
o
f
7
5
%
-
8
0
%,
th
e
n
u
m
b
er
o
f
f
ea
tu
r
e
s
w
er
e
s
ti
ll
s
u
b
s
ta
n
tiall
y
r
ed
u
ce
d
an
d
th
e
s
a
m
e
ac
c
u
r
ac
y
w
it
h
test
in
g
w
as
o
b
s
er
v
ed
w
it
h
o
u
t
u
s
i
n
g
t
h
e
f
ea
tu
r
e
s
elec
tio
n
tech
n
iq
u
e.
T
h
i
s
r
es
u
lt
w
as
al
s
o
o
b
s
er
v
ed
i
n
o
th
er
test
i
n
g
d
ata
u
s
i
n
g
th
e
SVM
-
C
FS
s
c
h
e
m
e,
w
it
h
d
ata
b
ased
o
n
a
r
atio
o
f
7
0
%
i
n
t
h
e
co
lo
n
t
u
m
o
r
an
d
t
h
e
ce
n
tr
al
n
er
v
o
u
s
s
y
s
te
m
tr
ain
in
g
d
ata.
B
o
th
o
f
w
h
ic
h
w
a
s
ab
le
to
p
r
o
v
id
e
in
cr
ea
s
ed
ac
cu
r
ac
y
.
Me
a
n
w
h
ile,
f
o
r
ML
L
leu
k
e
m
ia
d
ata
an
d
L
u
n
g
ca
n
ce
r
d
ata
w
i
th
a
r
atio
o
f
8
0
%
-
82%
,
alt
h
o
u
g
h
t
h
er
e
w
a
s
n
o
in
cr
ea
s
e
i
n
ac
c
u
r
ac
y
,
t
h
e
n
u
m
b
er
o
f
f
ea
t
u
r
es
m
a
n
ag
ed
to
b
e
r
ed
u
ce
d
s
u
b
s
ta
n
tial
l
y
.
Ho
w
ev
er
,
f
o
r
th
e
p
r
o
s
tate
ca
n
ce
r
,
o
v
ar
ian
ca
n
ce
r
,
an
d
b
r
ea
s
t
ca
n
ce
r
d
ata
w
it
h
d
if
f
er
e
n
t
d
ata
r
atio
s
,
all
ex
p
er
ien
ce
d
a
d
ec
r
ea
s
e
in
ac
cu
r
ac
y
,
b
u
t
t
h
e
r
atio
w
a
s
p
r
o
b
ab
ly
n
o
t
th
e
m
ain
ca
u
s
e
o
f
d
ec
r
ea
s
ed
ac
cu
r
ac
y
;
a
n
d
o
th
er
f
ac
to
r
s
s
u
ch
as t
h
r
esh
o
ld
s
.
i
n
t
h
e
s
ch
e
m
e
co
u
ld
also
ca
u
s
e
t
h
is
r
es
u
lt.
4.
CO
NCLU
SI
O
N
AND
F
U
T
U
RE
WO
RK
B
ased
o
n
t
h
e
r
es
u
lts
o
f
th
e
t
esti
n
g
s
ce
n
ar
io
co
n
d
u
cted
i
n
th
is
s
t
u
d
y
,
it
ca
n
b
e
co
n
clu
d
ed
th
at
t
h
e
R
elie
f
F
a
n
d
co
r
r
elatio
n
f
ea
t
u
r
e
s
elec
tio
n
(
C
FS
)
tec
h
n
iq
u
e
s
o
n
m
icr
o
ar
r
a
y
d
ata
cla
s
s
i
f
icat
io
n
u
s
i
n
g
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SM
V)
ca
n
g
en
er
all
y
p
r
o
v
id
e
m
o
r
e
o
p
ti
m
al
r
esu
lt
s
co
m
p
ar
ed
to
th
e
class
i
f
icatio
n
p
r
o
ce
s
s
w
it
h
o
u
t
u
s
i
n
g
f
ea
tu
r
e
s
elec
tio
n
.
Ho
w
e
v
er
,
s
o
m
e
test
s
u
n
d
e
r
th
e
C
SF
-
SVM
s
ce
n
ar
io
s
p
r
o
d
u
ce
d
r
esu
lt
s
w
it
h
d
ec
r
ea
s
ed
ac
cu
r
ac
y
co
m
p
ar
ed
to
SVM
w
i
th
o
u
t f
ea
tu
r
e
s
elec
t
io
n
.
I
n
SVM
test
i
n
g
,
w
it
h
o
u
t
u
s
in
g
f
ea
tu
r
e
s
elec
tio
n
,
b
y
te
s
ti
n
g
s
e
v
er
al
t
y
p
e
s
o
f
k
er
n
els
to
d
eter
m
in
e
o
p
tim
a
l
class
i
f
icatio
n
p
er
f
o
r
m
an
ce
,
th
e
b
est
ac
cu
r
ac
y
w
as
a
ch
iev
ed
u
s
i
n
g
t
h
e
L
i
n
ea
r
k
er
n
el,
w
h
ich
r
etu
r
n
ed
an
av
er
ag
e
ac
c
u
r
ac
y
o
f
8
3
.
4
5
%.
I
n
co
n
tr
ast,
th
e
R
B
F
k
er
n
el
p
r
o
d
u
ce
d
an
av
er
ag
e
ac
cu
r
ac
y
o
f
6
5
.
1
7
%
an
d
th
e
P
o
ly
n
o
m
ial
k
er
n
el
p
r
o
d
u
ce
d
6
1
.
6
7
% a
cc
u
r
ac
y
.
T
h
e
f
in
a
l
ac
c
u
r
ac
y
o
b
tai
n
ed
f
r
o
m
th
e
th
r
ee
test
i
n
g
s
ce
n
ar
i
o
s
f
o
r
all
d
ata
u
n
d
er
th
e
SV
M
s
ce
n
ar
i
o
w
it
h
o
u
t
u
s
in
g
t
h
e
f
ea
t
u
r
e
s
ele
ctio
n
tech
n
iq
u
e
w
a
s
an
a
v
er
ag
e
ac
cu
r
ac
y
o
f
8
3
.
4
5
%.
Me
an
w
h
ile,
R
el
ief
F
-
SVM
p
r
o
d
u
ce
d
an
av
er
ag
e
ac
c
u
r
ac
y
o
f
9
4
.
8
7
%,
an
d
C
FS
-
S
VM
p
r
o
d
u
ce
d
8
4
%
ac
cu
r
ac
y
.
Fro
m
t
h
ese
r
es
u
lt
s
,
it
ca
n
b
e
co
n
clu
d
ed
th
at
t
h
e
test
in
g
s
ch
e
m
e
i
n
v
o
l
v
i
n
g
R
elie
f
F
a
s
t
h
e
f
ea
t
u
r
e
s
elec
tio
n
tec
h
n
iq
u
e
w
it
h
S
VM
h
ad
t
h
e
b
est cla
s
s
i
f
icatio
n
ac
cu
r
ac
y
.
So
m
e
s
u
g
g
e
s
tio
n
s
r
elate
d
to
th
is
r
esear
c
h
i
n
clu
d
e
i
m
p
r
o
v
i
n
g
t
h
e
C
FS
-
SV
M
test
i
n
g
s
c
h
e
m
e
u
s
i
n
g
alg
o
r
ith
m
s
f
o
r
s
ea
r
c
h
in
g
s
u
c
h
as
Fo
r
w
ar
d
Select
io
n
a
n
d
w
it
h
a
C
FS
t
h
r
es
h
o
ld
w
ith
5
as
t
h
e
m
a
x
i
m
u
m
.
T
h
i
s
w
o
u
ld
s
t
ill
r
et
u
r
n
a
h
i
g
h
p
r
o
b
ab
ilit
y
o
f
i
n
cr
ea
s
ed
ac
c
u
r
ac
y
,
s
o
it
is
b
etter
f
o
r
f
u
t
u
r
e
s
t
u
d
i
es
to
u
s
e
a
g
r
ea
ter
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
C
o
mp
a
r
a
tive
a
n
a
lysi
s
o
f
R
eliefF
-
S
V
M a
n
d
C
F
S
-
S
V
M fo
r
micro
a
r
r
a
y
...
(
Mo
ch
a
ma
d
A
g
u
s
ta
N
a
o
fa
l H
a
kim)
3401
v
alu
e
f
o
r
d
ata
w
it
h
lar
g
e
d
i
m
e
n
s
io
n
s
s
u
ch
as
m
icr
o
ar
r
ay
d
ata.
T
h
en
,
to
p
r
o
v
id
e
m
o
r
e
v
ali
d
test
r
e
s
u
l
ts
f
o
r
al
l
d
ata
u
s
ed
,
cr
o
s
s
-
v
alid
atio
n
ev
alu
atio
n
co
u
ld
b
e
d
o
n
e
to
p
r
o
v
id
e
m
o
r
e
v
alid
e
v
alu
a
tio
n
r
esu
lt
s
b
ec
au
s
e
t
h
i
s
tech
n
iq
u
e
tr
ea
ts
all
d
ata
i
n
cl
u
d
in
g
te
s
ti
n
g
d
ata
an
d
tr
ain
i
n
g
d
ata.
RE
F
E
R
E
NC
E
S
[1
]
W
.
Yip
,
S.
B.
Am
in
,
a
n
d
C
.
Li
,
“
A
S
u
rv
e
y
o
f
Clas
sifi
c
a
ti
o
n
T
e
c
h
n
iq
u
e
s
f
o
r
M
icro
a
rra
y
,
”
Ha
n
d
b
o
o
k
o
f
S
t
a
ti
stica
l
Bi
o
in
fo
rm
a
t
ics
,
v
o
l.
3
,
p
p
.
1
9
3
–
2
2
3
,
2
0
1
1
.
[2
]
G
.
J.
G
o
rd
o
n
e
t
al
.,
“
T
ra
n
sla
ti
o
n
o
f
m
icro
a
rra
y
d
a
ta
in
to
c
li
n
ic
a
ll
y
re
le
v
a
n
t
c
a
n
c
e
r
d
iag
n
o
stic
tes
ts
u
sin
g
g
e
n
e
e
x
p
re
ss
io
n
ra
ti
o
s
i
n
l
u
n
g
c
a
n
c
e
r
a
n
d
m
e
so
th
e
li
o
m
a
,
”
Am
e
ric
a
n
Asso
c
i
a
ti
o
n
f
o
r
Ca
n
c
e
r
Res
e
a
rc
h
,
v
o
l.
6
2
,
p
p
.
4
9
6
3
–
4
9
6
7
,
S
e
p
.
2
0
0
2
.
[3
]
S
.
Ch
o
rm
u
n
g
e
a
n
d
S
.
Je
n
a
,
“
Co
rre
latio
n
b
a
se
d
F
e
a
tu
re
S
e
lec
ti
o
n
w
it
h
Clu
ste
rin
g
f
o
r
Hig
h
Di
m
e
n
sio
n
a
l
Da
ta,”
J
o
u
rn
a
l
o
f
El
e
c
trica
l
S
y
ste
ms
a
n
d
In
fo
rm
a
ti
o
n
T
e
c
h
n
o
l
o
g
y
,
v
o
l.
5
,
p
p
.
5
4
2
–
5
4
9
,
De
c
.
2
0
1
8
.
[4
]
A
.
A
.
Ya
h
y
a
,
“
F
e
a
tu
re
S
e
lec
ti
o
n
f
o
r
Hig
h
Dim
e
n
sio
n
a
l
Da
ta:
A
n
Ev
o
lu
ti
o
n
a
ry
F
il
ter
A
p
p
ro
a
c
h
,
”
J
o
u
rn
a
l
o
f
Co
mp
u
ter
S
c
ien
c
e
,
v
o
l.
7
,
p
p
.
8
0
0
–
8
2
0
,
M
a
y
2
0
1
1
.
[5
]
M
.
Ch
e
rri
n
g
to
n
,
F
.
T
h
a
b
tah
,
J.
L
u
,
a
n
d
Q.
X
u
,
“
F
e
a
tu
re
S
e
lec
ti
o
n
:
F
il
ter
M
e
th
o
d
s
P
e
rf
o
rm
a
n
c
e
Ch
a
ll
e
n
g
e
s,”
In
ter
n
a
t
io
n
a
l
C
o
n
fer
e
n
c
e
o
n
C
o
mp
u
ter
a
n
d
In
f
o
rm
a
ti
o
n
S
c
ien
c
e
s
,
M
a
y
2
0
1
9
,
p
p
.
1
–
4.
[6
]
R.
J.
P
a
lm
a
-
M
e
n
d
o
z
a
,
L
.
d
e
M
a
rc
o
s,
D.
Ro
d
íi
g
u
e
z
,
a
n
d
A
.
A
l
o
n
so
-
Be
tan
z
o
s,
“
Distrib
u
ted
C
o
r
re
latio
n
-
Ba
se
d
F
e
a
tu
re
S
e
lec
ti
o
n
in
S
p
a
rk
,
”
a
rXi
v
:
1
9
0
1
.
1
1
2
8
6
,
p
p
.
1
–
2
5
,
Ja
n
.
2
0
1
9
.
[7
]
V
.
B
o
ló
n
-
Ca
n
e
d
o
,
N.
S
á
n
c
h
e
z
-
M
a
ro
ñ
o
,
a
n
d
A
.
A
lo
n
so
-
Be
tan
z
o
s,
“
Distrib
u
ted
f
e
a
tu
re
se
lec
ti
o
n
:
A
n
a
p
p
li
c
a
ti
o
n
t
o
m
icro
a
rra
y
d
a
ta cla
ss
i
f
ica
ti
o
n
,
”
Ap
p
li
e
d
S
o
ft
C
o
mp
u
ti
n
g
,
v
o
l.
3
0
,
p
p
.
1
3
6
–
1
5
0
,
M
a
y
2
0
1
5
.
[8
]
M
.
G
h
o
sh
,
S
.
A
d
h
ik
a
r
y
,
K.
K.
G
h
o
sh
,
A
.
S
a
rd
a
r,
S
.
Be
g
u
m
,
a
n
d
R.
S
a
rk
a
r,
“
G
e
n
e
ti
c
a
l
g
o
rit
h
m
b
a
se
d
c
a
n
c
e
ro
u
s
g
e
n
e
id
e
n
ti
f
ica
ti
o
n
f
ro
m
m
icro
a
rr
a
y
d
a
ta
u
s
in
g
e
n
se
m
b
le
o
f
f
il
ter
m
e
th
o
d
s,”
M
e
d
ica
l
&
Bi
o
lo
g
ica
l
En
g
i
n
e
e
rin
g
&
Co
mp
u
t
in
g
,
v
o
l
.
5
7
,
p
p
.
1
5
9
–
1
7
6
,
F
e
b
.
2
0
1
8
.
[9
]
V
.
B
o
ló
n
-
Ca
n
e
d
o
,
N.
S
á
n
c
h
e
z
-
M
a
ro
ñ
o
,
a
n
d
A
.
A
lo
n
so
-
Be
tan
z
o
s,
“
A
n
e
n
se
m
b
le
o
f
f
il
ters
a
n
d
c
las
sif
iers
f
o
r
m
icro
a
rra
y
d
a
ta cla
ss
i
f
ica
ti
o
n
,
”
Pa
tt
e
rn
Rec
o
g
n
it
io
n
,
v
o
l
.
4
5
,
p
p
.
5
3
1
–
5
3
9
,
Ja
n
.
2
0
1
2
.
[1
0
]
H
.
Ay
d
a
d
e
n
ta
a
n
d
A
d
iw
ij
a
y
a
,
“
O
n
t
h
e
c
las
sif
ica
ti
o
n
tec
h
n
i
q
u
e
s
in
d
a
ta
m
in
in
g
f
o
r
m
icro
a
rra
y
d
a
ta
c
las
si
f
ica
ti
o
n
,
”
J
o
u
rn
a
l
o
f
P
h
y
sic
s: Co
n
fer
e
n
c
e
S
e
rie
s
,
v
o
l.
9
7
1
,
p
p
.
1
–
1
0
,
2
0
1
8
.
[1
1
]
N.
S
á
n
c
h
e
z
-
M
a
ro
ñ
o
,
O.
F
o
n
ten
la
-
Ro
m
e
ro
,
a
n
d
B.
P
é
re
z
-
S
á
n
c
h
e
z
,
“
Clas
si
f
ica
ti
o
n
o
f
M
icro
a
rra
y
D
a
t
a
,
”
M
e
th
o
d
s
in
mo
lec
u
la
r
b
io
l
o
g
y
,
v
o
l.
1
9
8
6
,
p
p
.
1
8
5
–
2
0
5
,
Ja
n
.
2
0
1
9
.
[1
2
]
O
.
M
a
im
o
n
a
n
d
L
.
Ro
k
a
c
h
,
“
Da
ta
M
in
in
g
a
n
d
Kn
o
w
led
g
e
Dis
c
o
v
e
r
y
Ha
n
d
b
o
o
k
,
2
n
d
e
d
,
”
J
o
u
rn
a
l
o
f
El
e
c
trica
l
S
y
ste
ms
a
n
d
I
n
f
o
rm
a
ti
o
n
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
2
,
p
p
.
1
0
1
–
1
0
5
,
Ja
n
.
2
0
1
0
.
[1
3
]
R
.
J.
Urb
a
n
o
w
icz
,
R
.
S
.
Olso
n
,
P
.
S
c
h
m
it
t,
M
.
M
e
e
k
e
r
,
a
n
d
J
.
H
.
M
o
o
re
,
“
Be
n
c
h
m
a
rk
in
g
re
li
e
f
-
b
a
s
e
d
f
e
a
tu
re
se
lec
ti
o
n
tec
h
n
i
q
u
e
s
f
o
r
b
i
o
in
f
o
r
m
a
ti
c
s
d
a
ta
m
in
in
g
,
”
J
o
u
rn
a
l
o
f
Bi
o
me
d
ica
l
In
f
o
rm
a
ti
c
s
,
v
o
l
.
8
5
,
p
p
.
1
6
8
–
1
8
8
,
S
e
p
.
2
0
1
8
.
[1
4
]
W
.
M
e
g
c
h
e
len
b
rin
k
,
E.
M
a
rc
h
i
o
ri
,
a
n
d
P
.
L
u
c
a
s,
“
Re
li
e
f
-
b
a
se
d
f
e
a
tu
re
se
lec
ti
o
n
in
b
i
o
in
f
o
r
m
a
ics
:
d
e
tec
ti
n
g
f
u
n
c
ti
o
n
a
l
sp
e
c
ifi
c
it
y
r
e
sid
u
e
s
f
ro
m
m
u
lt
ip
le
se
q
u
e
n
c
e
a
li
g
n
m
e
n
ts,
”
M
a
ste
r
th
e
sis
in
in
fo
r
ma
ti
o
n
sc
ien
c
e
,
p
p
.
2
9
–
4
0
,
J
u
l
.
2
0
1
0
.
[1
5
]
B.
T
a
n
g
a
n
d
L
.
Zh
a
n
g
,
“
M
u
lt
i
-
c
las
s
S
e
m
i
-
su
p
e
rv
ise
d
L
o
g
isti
c
I
-
REL
IEF
F
e
a
tu
re
S
e
lec
ti
o
n
Ba
se
d
o
n
Ne
a
re
st
Ne
ig
h
b
o
r,
”
P
a
c
if
ic
-
Asia
Co
n
fer
e
n
c
e
o
n
Kn
o
wled
g
e
Disc
o
v
e
ry
a
n
d
D
a
ta
M
in
i
n
g
,
v
o
l.
1
1
4
4
0
,
p
p
.
2
8
1
–
2
9
2
,
M
a
r
.
2
0
1
9
.
[1
6
]
Y.
Zh
a
n
g
,
X
.
Re
n
,
a
n
d
J.
Zh
a
n
g
,
“
In
tru
sio
n
d
e
tec
ti
o
n
m
e
th
o
d
b
a
se
d
o
n
in
f
o
rm
a
ti
o
n
g
a
in
a
n
d
Re
li
e
f
F
f
e
a
tu
re
se
lec
ti
o
n
,
”
In
ter
n
a
ti
o
n
a
l
J
o
i
n
t
C
o
n
fer
e
n
c
e
o
n
Ne
u
r
a
l
Ne
two
rk
s (
IJ
CNN)
,
Ju
l.
2
0
1
9
,
p
p
.
1
–
5.
[1
7
]
N
.
S
p
o
laô
r
,
E.
A
.
Ch
e
r
m
a
n
,
M
.
C.
M
o
n
a
rd
,
a
n
d
H.
D.
L
e
e
,
“
Re
li
e
f
F
f
o
r
M
u
lt
i
-
lab
e
l
F
e
a
tu
re
S
e
lec
ti
o
n
,
”
Bra
zili
a
n
Co
n
fer
e
n
c
e
o
n
In
tell
ig
e
n
t
S
y
ste
ms
,
Oc
t.
2
0
1
3
,
p
p
.
6
–
11
.
[1
8
]
R.
P
.
L
.
Du
rg
a
b
a
i
a
n
d
Y.
R.
B
h
u
sh
a
n
,
“
F
e
a
tu
re
S
e
lec
ti
o
n
u
si
n
g
Re
li
e
f
F
A
lg
o
rit
h
m
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Ad
v
a
n
c
e
d
Res
e
a
rc
h
in
Co
mp
u
ter
a
n
d
Co
mm
u
n
ica
t
io
n
En
g
i
n
e
e
rin
g
,
p
p
.
8
2
1
5
–
8
2
1
8
,
Oc
t
.
2
0
1
4
.
[1
9
]
M
.
A
.
Ha
ll
,
“
Co
rre
latio
n
-
b
a
se
d
F
e
a
tu
re
S
e
lec
ti
o
n
f
o
r
M
a
c
h
in
e
L
e
a
rn
in
g
,
”
M
a
ste
r
th
e
sis
in
in
f
o
r
m
a
ti
o
n
sc
ien
c
e
,
v
o
l.
1
9
,
p
p
.
5
1
–
7
4
,
Ju
n
.
2
0
0
0
.
[2
0
]
G
.
S
o
sa
-
Ca
b
re
ra
,
M
.
G
a
r
c
ía
-
T
o
rr
e
s,
S
.
G
ó
m
e
z
-
G
u
e
rre
ro
,
C
.
E.
S
c
h
a
e
re
r,
a
n
d
F
.
Div
in
a
,
“
A
m
u
lt
iv
a
ri
a
te
a
p
p
ro
a
c
h
t
o
th
e
sy
m
m
e
tri
c
a
l
u
n
c
e
rtain
ty
m
e
a
su
re
:
A
p
p
li
c
a
ti
o
n
to
f
e
a
tu
re
se
le
c
ti
o
n
p
ro
b
lem
,
”
In
fo
rm
a
ti
o
n
S
c
ie
n
c
e
s
,
v
o
l.
4
9
4
,
p
p
.
1
–
2
0
,
A
u
g
.
2
0
1
9
.
[2
1
]
H
.
A
lsh
a
m
l
a
n
,
G
.
B
a
d
r,
a
n
d
Y.
A
lo
h
a
li
,
“
M
icro
a
rra
y
Ge
n
e
S
e
l
e
c
ti
o
n
a
n
d
Ca
n
c
e
r
Clas
sif
ic
a
ti
o
n
M
e
th
o
d
Us
in
g
A
rti
f
icia
l
Be
e
Co
lo
n
y
a
n
d
S
V
M
A
l
g
o
rit
h
m
s
(
A
BC
-
S
V
M
),
”
Pro
c
e
e
d
in
g
s
o
f
t
h
e
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
Da
t
a
En
g
i
n
e
e
rin
g
2
0
1
5
,
v
o
l.
5
2
0
,
A
u
g
.
2
0
1
9
,
p
p
.
5
7
5
–
5
84
.
[2
2
]
V
.
N.
V
a
p
n
ik
,
“
S
tatisti
c
a
l
L
e
a
rn
in
g
T
h
e
o
ry
,
”
A
W
il
e
y
-
I
n
ter
sc
ien
c
e
p
u
b
li
c
a
t
io
n
J
o
h
n
W
il
e
y
&
S
o
n
s
,
I
n
c
,
p
p
.
40
1
–
41
0
,
S
e
p
.
1
9
9
8
.
[2
3
]
M
.
P
.
S
.
Bro
w
n
e
t
a
l
.
,
“
Kn
o
w
led
g
e
-
b
a
se
d
a
n
a
ly
sis
o
f
m
icro
a
rra
y
g
e
n
e
e
x
p
re
ss
io
n
d
a
ta
b
y
u
sin
g
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
s,
”
Pro
c
e
e
d
in
g
s
o
f
t
h
e
Na
ti
o
n
a
l
Ac
a
d
e
my
o
f
S
c
ien
c
e
s o
f
th
e
Un
it
e
d
S
t
a
tes
o
f
Ame
ric
a
,
2
0
0
0
,
p
p
.
2
6
2
–
2
67
.
[2
4
]
R
.
A
.
M
u
sh
e
e
r
,
C
.
K.
V
e
rm
a
,
a
n
d
N.
S
riv
a
sta
v
a
,
“
No
v
e
l
m
a
c
h
in
e
lea
rn
in
g
a
p
p
r
o
a
c
h
f
o
r
c
las
sif
i
c
a
ti
o
n
o
f
h
ig
h
-
d
im
e
n
sio
n
a
l
m
icro
a
rra
y
d
a
ta
,”
S
o
ft
Co
mp
u
ti
n
g
,
v
o
l.
2
3
,
p
p
.
1
3
4
0
9
–
1
3
4
2
1
,
M
a
r
.
2
0
1
9
.
[2
5
]
Y
.
P
e
n
g
,
W
.
Li
,
a
n
d
Y.
L
iu
,
“
A
Hy
b
rid
A
p
p
ro
a
c
h
f
o
r
Bio
m
a
rk
e
r
Di
sc
o
v
e
r
y
f
ro
m
M
icro
a
rra
y
G
e
n
e
Ex
p
re
ss
io
n
Da
ta
f
o
r
Ca
n
c
e
r
Clas
si
f
ica
ti
o
n
,
”
Ca
n
c
e
r In
fo
rm
a
ti
c
s
,
v
o
l.
2
,
p
p
.
3
0
1
–
3
1
1
,
F
e
b
.
2
0
1
9
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
11
,
No
.
4
,
A
u
g
u
s
t 2
0
2
1
:
3
3
9
3
-
3402
3402
B
I
O
G
RAP
H
I
E
S
O
F
AUTH
O
RS
M
o
c
h
a
m
a
d
Ag
u
st
a
N
a
o
f
a
l
H
a
k
i
m
wa
s
b
o
rn
in
Ng
a
n
ju
k
,
In
d
o
e
n
sia
,
in
2
0
1
5
re
c
e
iv
e
d
th
e
Ba
c
h
e
lo
r
d
e
g
re
e
o
f
In
f
o
r
m
a
ti
c
s
En
g
in
e
rin
g
f
ro
m
T
e
lk
o
m
Un
iv
e
r
sity
B
a
n
d
u
n
g
.
I
n
d
o
e
n
sia
.
His
re
se
a
rc
h
in
tere
st i
n
c
lu
d
e
s
o
n
Da
ta
M
in
i
n
g
,
M
a
c
h
i
n
e
L
e
a
rn
in
g
,
a
n
d
M
icro
a
rra
y
D
a
ta
.
Ad
iw
ija
y
a
is
a
p
ro
f
e
ss
o
r
o
f
m
a
t
h
e
m
a
ti
c
s
a
t
S
c
h
o
o
l
o
f
Co
m
p
u
ti
n
g
,
T
e
l
k
o
m
Un
iv
e
r
sit
y
.
He
is
in
tere
ste
d
in
t
h
e
re
se
a
rc
h
a
re
a
o
f
g
ra
p
h
th
e
o
ry
a
n
d
it
s
a
p
p
li
c
a
ti
o
n
s,
d
a
ta
sc
ien
c
e
,
a
n
d
in
f
o
rm
a
ti
o
n
sc
ien
c
e
.
He
jo
in
e
d
T
e
lk
o
m
Un
iv
e
rsity
sin
c
e
2
0
0
0
a
n
d
h
a
s b
e
c
o
m
e
p
ro
f
e
ss
o
r
sin
c
e
2
0
1
6
.
W
i
d
i
As
tu
ti
o
b
tain
e
d
b
a
c
h
e
lo
r
d
e
g
re
e
a
n
d
m
a
ste
r
d
e
g
re
e
o
f
In
f
o
r
m
a
ti
c
s
a
t
T
e
lk
o
m
Un
iv
e
r
sit
y
.
He
r
re
se
a
rc
h
in
tere
sts in
c
lu
d
e
d
a
ta m
in
in
g
,
b
ig
d
a
ta,
m
a
c
h
in
e
lea
rn
in
g
a
n
d
a
rti
f
icia
l
in
telli
g
e
n
c
e
.
Evaluation Warning : The document was created with Spire.PDF for Python.