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
.
3
,
J
u
n
e
2021
,
p
p
.
2631
~
2
6
3
9
I
SS
N:
2
0
8
8
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v
1
1
i
3
.
pp
2
6
3
1
-
2
6
3
9
2631
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m
Ex
pert
ca
ncer
m
o
del using
supe
rv
ised a
lg
o
rith
m
s w
ith
a
LASSO
selec
tion
a
ppro
a
ch
P
ro
na
b G
ho
s
h
1
,
A
s
if
K
a
ri
m
2
,
Sy
eda
T
a
njila
At
ik
3
,
Sa
i
m
a
Af
rin
4
,
M
o
hd
.
Sa
if
uzza
m
a
n
5
1,
3,
4,
5
De
p
a
rtm
e
n
t
o
f
Co
m
p
u
ter S
c
ien
c
e
a
n
d
E
n
g
in
e
e
rin
g
,
Da
ff
o
d
il
In
tern
a
ti
o
n
a
l
Un
iv
e
rsity
,
Dh
a
k
a
,
Ba
n
g
lad
e
sh
2
Co
ll
e
g
e
o
f
En
g
in
e
e
rin
g
,
IT
a
n
d
En
v
iro
n
m
e
n
t,
Ch
a
rles
Da
rw
i
n
Un
iv
e
rsit
y
,
Ca
su
a
rin
a
,
NT
,
A
u
stra
li
a
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
u
l 5
,
2
0
2
0
R
ev
i
s
ed
Sep
1
1
,
2
0
2
0
A
cc
ep
ted
Oct
1
3
,
2
0
2
0
On
e
o
f
th
e
m
o
st
c
rit
ica
l
issu
e
s
o
f
th
e
m
o
rtalit
y
ra
t
e
in
th
e
m
e
d
ica
l
f
ield
in
c
u
rre
n
t
ti
m
e
s
is
b
re
a
st
c
a
n
c
e
r.
No
w
a
d
a
y
s,
a
larg
e
n
u
m
b
e
r
o
f
m
e
n
a
n
d
w
o
m
e
n
a
re
f
a
c
in
g
c
a
n
c
e
r
-
re
late
d
d
e
a
th
s
d
u
e
to
t
h
e
lac
k
o
f
e
a
rl
y
d
iag
n
o
s
is
s
y
ste
m
s
a
n
d
p
r
o
p
e
r
trea
tm
e
n
t
p
e
r
y
e
a
r.
T
o
tac
k
l
e
th
e
issu
e
,
v
a
rio
u
s
d
a
ta
m
in
in
g
a
p
p
ro
a
c
h
e
s
h
a
v
e
b
e
e
n
a
n
a
ly
z
e
d
to
b
u
il
d
a
n
e
ff
e
c
ti
v
e
m
o
d
e
l
th
a
t
h
e
lp
s
t
o
id
e
n
ti
f
y
th
e
d
iff
e
r
e
n
t
sta
g
e
s
o
f
d
e
a
d
ly
c
a
n
c
e
rs.
T
h
e
stu
d
y
su
c
c
e
ss
f
u
ll
y
p
ro
p
o
se
s
a
n
e
a
rl
y
c
a
n
c
e
r
d
ise
a
se
m
o
d
e
l
b
a
se
d
o
n
f
iv
e
d
iff
e
re
n
t
s
u
p
e
rv
ise
d
a
lg
o
rit
h
m
s
su
c
h
a
s
lo
g
isti
c
re
g
re
ss
io
n
(h
e
n
c
e
f
o
rth
L
R),
d
e
c
i
sio
n
tree
(h
e
n
c
e
f
o
rth
DT
),
ra
n
d
o
m
f
o
re
st
(h
e
n
c
e
f
o
rth
RF
),
S
u
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
(h
e
n
c
e
f
o
rth
S
V
M
),
a
n
d
K
-
n
e
a
r
e
st
n
e
ig
h
b
o
r
(h
e
n
c
e
f
o
rth
KN
N).
A
f
t
e
r
a
n
a
p
p
ro
p
riate
p
re
p
ro
c
e
ss
in
g
o
f
t
h
e
d
a
tas
e
t,
lea
st
a
b
so
l
u
te
sh
r
i
n
k
a
g
e
a
n
d
se
lec
ti
o
n
o
p
e
ra
t
o
r
(L
A
S
S
O)
w
a
s
u
se
d
f
o
r
f
e
a
tu
re
se
lec
ti
o
n
(F
S
)
u
sin
g
a
1
0
-
f
o
ld
c
ro
ss
-
v
a
li
d
a
ti
o
n
(CV
)
a
p
p
r
o
a
c
h
.
Em
p
lo
y
in
g
LA
S
S
O
w
it
h
1
0
-
f
o
ld
c
ro
ss
-
v
a
li
d
a
ti
o
n
h
a
s
b
e
e
n
a
n
o
v
e
l
ste
p
s
in
tr
o
d
u
c
e
d
i
n
th
is
re
se
a
rc
h
.
Af
ter
wa
rd
s,
d
if
fe
re
n
t
p
e
rf
o
r
m
a
n
c
e
e
v
a
lu
a
ti
o
n
m
e
tri
c
s
we
re
m
e
a
su
re
d
to
sh
o
w
a
c
c
u
ra
te
p
re
d
ictio
n
s
b
a
se
d
o
n
t
h
e
p
r
o
p
o
se
d
a
lg
o
rit
h
m
s.
T
h
e
re
su
lt
in
d
i
c
a
ted
to
p
a
c
c
u
ra
c
y
wa
s
re
c
e
iv
e
d
f
ro
m
RF
c
las
si
f
ier,
a
p
p
ro
x
im
a
tel
y
9
9
.
4
1
%
w
it
h
th
e
in
teg
ra
ti
o
n
o
f
LA
S
S
O.
F
in
a
ll
y
,
a
c
o
m
p
re
h
e
n
siv
e
c
o
m
p
a
riso
n
wa
s
c
a
rried
o
u
t
on
W
isc
o
n
si
n
b
re
a
st
c
a
n
c
e
r
(
d
iag
n
o
stic)
d
a
tas
e
t
(W
BCD)
to
g
e
th
e
r
w
it
h
so
m
e
c
u
rre
n
t
w
o
rk
s c
o
n
tain
i
n
g
a
ll
f
e
a
tu
re
s.
K
ey
w
o
r
d
s
:
B
r
ea
s
t c
an
ce
r
Dec
is
io
n
tr
ee
SVM
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
:
P
r
o
n
ab
Gh
o
s
h
Dep
ar
t
m
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
an
d
E
n
g
i
n
ee
r
in
g
Daf
f
o
d
il I
n
ter
n
atio
n
al
U
n
iv
er
s
it
y
Dh
a
k
a,
B
an
g
lad
es
h
E
m
ail:
p
r
o
n
ab
1
7
1
2
@
g
m
a
il.c
o
m
1.
I
NT
RO
D
UCT
I
O
N
T
h
e
s
ec
o
n
d
-
h
i
g
h
est
d
ea
d
l
y
d
i
s
ea
s
e
o
v
er
th
e
w
o
r
ld
an
d
a
s
i
g
n
i
f
ica
n
t
r
ea
s
o
n
f
o
r
w
o
m
e
n
'
s
d
ea
th
s
in
th
is
co
n
te
m
p
o
r
ar
y
t
i
m
e
is
b
r
ea
s
t
ca
n
ce
r
.
I
t
cr
ea
tes
a
s
ig
n
i
f
i
ca
n
t
ch
a
llen
g
e
to
w
o
m
en
’
s
h
ea
lth
i
n
t
h
e
w
o
r
ld
to
d
ay
.
As
p
er
th
e
s
tati
s
tic
s
o
f
th
e
w
o
r
ld
h
ea
lt
h
o
r
g
a
n
izatio
n
(
W
HO)
,
2
.
1
m
i
llio
n
w
o
m
en
a
r
e
g
ettin
g
a
f
f
ec
ted
d
u
e
to
b
r
ea
s
t
ca
n
ce
r
an
n
u
all
y
.
T
h
e
r
ate
o
f
d
ea
th
is
ap
p
r
o
x
i
m
atel
y
1
5
%
a
m
o
n
g
all
w
o
m
en
[
1
]
,
a
r
ep
o
r
te
d
n
u
m
b
er
o
f
6
2
7
,
0
0
0
f
e
m
ales
d
ied
d
u
e
to
ca
n
ce
r
-
r
elate
d
is
s
u
es
in
2
0
1
8
.
I
t
h
as
b
ee
n
p
r
ed
icted
th
at
1
2
7
.
5
ar
e
d
iag
n
o
s
ed
,
w
h
er
ea
s
2
0
.
6
f
e
m
ales
h
av
e
d
ied
p
er
1
0
0
,
0
0
0
ea
ch
y
ea
r
[
2
]
.
A
s
u
r
v
e
y
w
as
ta
k
en
f
r
o
m
Glo
b
o
ca
n
w
h
ic
h
en
s
u
r
es
t
h
at
8
7
,
0
9
0
f
em
ales
h
av
e
d
ied
in
th
e
y
ea
r
o
f
2
0
1
8
[
3
]
b
u
t
5
8
%
o
f
d
ea
t
h
s
w
er
e
n
o
ticed
in
d
ev
elo
p
ed
co
u
n
tr
ies
as
p
er
th
e
r
ep
o
r
t
o
f
2
0
0
8
[
4
]
.
C
o
n
s
id
er
in
g
t
h
e
n
u
m
b
er
o
f
d
ea
th
r
ec
o
r
d
s
,
I
n
d
ia
h
as
ac
h
iev
ed
t
h
e
f
ir
s
t r
an
k
w
h
er
ea
s
T
h
ailan
d
h
as t
h
e
f
o
u
r
t
h
m
o
s
t
,
w
it
h
5
,
9
0
2
d
ea
th
s
in
th
e
s
a
m
e
y
ea
r
[
3
]
.
T
h
e
k
e
y
co
n
ce
p
t
b
eh
in
d
t
h
e
p
r
o
p
o
s
ed
r
esear
ch
is
to
d
ev
elo
p
a
f
r
a
m
e
w
o
r
k
f
o
r
b
r
ea
s
t
ca
n
ce
r
d
iag
n
o
s
i
s
th
at
is
co
m
p
letel
y
b
a
s
ed
o
n
m
ac
h
in
e
lear
n
i
n
g
.
T
h
e
s
t
u
d
y
ai
m
s
to
ad
d
r
ess
d
if
f
er
en
t a
lg
o
r
it
h
m
s
s
u
c
h
as
L
R
,
DT
,
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
.
3
,
J
u
n
e
2
0
2
1
:
2
6
3
1
-
2639
2632
R
F,
S
VM
,
an
d
KNN
f
o
r
id
en
ti
f
icatio
n
t
h
e
a
f
f
ec
ted
p
eo
p
l
e
o
f
b
r
ea
s
t
ca
n
ce
r
.
T
h
e
m
o
s
t
r
eliab
le
s
elec
tio
n
tech
n
iq
u
e
s
u
ch
as
L
A
SS
O
i
s
also
u
s
ed
to
d
et
er
m
i
n
e
t
h
e
m
o
s
t
r
ele
v
an
t
a
n
d
s
tr
o
n
g
l
y
a
s
s
o
ci
ated
attr
ib
u
tes
t
h
at
s
h
o
w
co
n
s
id
er
ab
le
in
f
l
u
e
n
ce
o
n
th
e
p
r
ed
icted
f
ea
tu
r
e.
T
h
i
s
an
al
y
s
is
is
ca
r
r
ied
o
u
t
u
s
in
g
1
0
-
f
o
ld
C
V
f
o
r
m
ak
in
g
it
m
o
r
e
r
eliab
le.
Dif
f
e
r
en
t
p
er
f
o
r
m
a
n
ce
ass
e
s
s
m
e
n
t
m
etr
ics
s
u
c
h
as
co
n
f
u
s
io
n
m
at
r
ix
,
ac
cu
r
ac
y
(
A
cc
)
,
p
r
ec
is
io
n
(
P
s
co
r
e)
,
r
ec
a
ll
(
R
s
co
r
e)
,
s
p
ec
if
ic
it
y
(
Sp
e)
,
n
eg
a
tiv
e
p
r
ed
ictiv
e
v
al
u
e
(
NP
V)
,
f
alse
d
is
co
v
er
y
r
ate
(
FDR
)
,
f
alse
n
e
g
ati
v
e
r
ate
(
F
NR
)
,
an
d
f
al
s
e
p
o
s
iti
v
e
r
ate
(
F
P
R
)
h
av
e
b
ee
n
u
s
ed
to
p
r
o
p
er
l
y
e
v
alu
a
te
cla
s
s
i
f
ier
p
er
f
o
r
m
a
n
ce
.
B
esid
es,
th
e
p
r
o
ce
d
u
r
e
o
f
d
ata
p
r
e
p
r
o
ce
s
s
in
g
i
s
ap
p
lied
to
th
e
d
ataset
o
f
b
r
e
ast
ca
n
ce
r
.
T
h
e
k
e
y
o
b
j
ec
tiv
es o
f
t
h
e
p
r
esen
t r
esea
r
ch
ar
e:
A
ll
f
ea
t
u
r
es
h
av
e
b
ee
n
p
r
ep
o
s
s
ess
ed
w
it
h
t
h
e
h
elp
o
f
s
tan
d
a
r
d
s
ca
ler
tech
n
iq
u
e
to
k
ee
p
th
e
v
alu
e
s
in
t
h
e
r
an
g
e
o
f
[
0
,
1
]
.
T
h
e
ev
alu
atio
n
p
r
o
ce
s
s
es
o
f
v
ar
io
u
s
m
o
d
el
s
h
a
v
e
b
ee
n
ex
p
er
im
e
n
ted
w
it
h
th
e
s
ep
ar
atio
n
o
f
8
0
:
2
0
b
y
L
A
S
SO
w
it
h
1
0
-
f
o
ld
cr
o
s
s
-
v
al
id
atio
n
.
T
h
e
s
tu
d
y
ca
r
r
ied
o
u
t
a
co
m
p
r
eh
en
s
iv
e
co
m
p
ar
is
o
n
o
f
th
e
p
er
f
o
r
m
a
n
ce
o
f
L
A
S
SO
s
elec
ted
f
ea
tu
r
es
a
n
d
cu
r
r
en
t
ex
is
ti
n
g
w
o
r
k
s
to
id
en
ti
f
y
t
h
e
a
f
f
ec
ted
ca
n
ce
r
p
atien
ts
,
w
h
ic
h
h
i
g
h
li
g
h
t
s
th
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
p
r
o
p
o
s
ed
in
telli
g
e
n
t
s
y
s
te
m
.
W
e
h
av
e
u
s
ed
th
e
d
ef
a
u
lt
s
ett
in
g
s
a
v
ailab
le
at
s
ci
k
it
-
lear
n
f
o
r
L
A
S
SO,
an
d
h
av
e
n
o
t
o
p
tim
ized
an
y
s
p
ec
i
f
ic
p
ar
a
m
eter
f
o
r
p
er
f
o
r
m
a
n
c
e
tu
n
i
n
g
.
T
h
e
d
ef
au
lt
s
et
tin
g
p
r
o
v
id
ed
g
o
o
d
en
o
u
g
h
r
es
u
lt
s
.
Sev
er
al
m
ac
h
i
n
e
lear
n
i
n
g
ap
p
r
o
ac
h
es
h
av
e
b
ee
n
e
v
alu
ated
to
p
r
e
d
ict
an
ac
cu
r
ate
o
u
tco
m
e
o
n
th
e
d
ataset
o
f
b
r
ea
s
t
ca
n
ce
r
.
So
m
e
o
f
th
e
m
ar
e
ex
p
lai
n
ed
to
s
h
o
w
t
h
e
r
esear
ch
er
s
'
f
i
n
d
in
g
s
.
L
atc
h
o
u
m
i
et
a
l.
[
5
]
ex
p
lain
ed
a
w
ei
g
h
ted
p
ar
ticle
s
w
ar
m
o
p
ti
m
izatio
n
tec
h
n
iq
u
e
u
s
in
g
a
s
m
o
o
th
s
u
p
p
o
r
t
v
ec
to
r
m
et
h
o
d
to
g
iv
e
th
is
r
esear
ch
n
o
v
elt
y
.
E
ar
n
i
n
g
a
lo
w
er
r
o
r
r
ate
w
a
s
t
h
e
m
ai
n
tar
g
et
o
f
t
h
is
r
esear
ch
t
ec
h
n
i
q
u
e
.
B
e
s
i
d
e
s
,
i
t
s
u
c
c
e
s
s
f
u
l
l
y
g
e
n
e
r
a
t
e
s
9
8
.
4
2
%
a
c
c
u
r
a
c
y
u
s
i
n
g
t
h
i
s
a
l
g
o
r
i
t
h
m
.
D
i
s
t
i
n
c
t
i
v
e
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
c
a
l
c
u
l
a
t
i
o
n
s
h
a
v
e
b
ee
n
co
n
te
m
p
lated
a
n
d
u
s
ed
to
an
ti
cip
ate
th
e
ea
r
l
y
d
etec
tio
n
o
f
b
r
ea
s
t
ca
n
ce
r
in
t
h
e
i
n
v
esti
g
ati
o
n
o
f
J
.
R
o
h
it
et
a
l.
[
6
]
t
h
ey
also
w
o
r
k
ed
o
n
b
r
ea
s
t
ca
n
ce
r
d
ataset
b
y
u
s
in
g
d
if
f
e
r
en
t
p
r
ed
ictiv
e
m
o
d
els
an
d
id
en
ti
f
ied
an
o
p
ti
m
al
s
o
lu
tio
n
co
n
s
id
er
i
n
g
v
ar
io
u
s
s
tag
e
s
o
f
ca
n
ce
r
.
T
h
e
m
o
d
els
w
er
e
ass
e
s
s
ed
s
ep
ar
atel
y
o
n
th
e
b
asis
o
f
t
h
eir
d
ep
lo
y
m
en
t
s
tr
ate
g
y
.
A
lico
v
ic
et
a
l.
u
s
ed
a
g
en
e
tic
alg
o
r
it
h
m
[
7
]
b
ased
o
n
FS
an
d
m
u
lt
ip
le
class
i
f
icatio
n
s
to
m
ak
e
t
h
eir
r
esear
ch
m
o
r
e
s
p
e
cif
ic.
T
h
at
r
esear
c
h
h
elp
s
to
f
i
n
d
o
u
t
t
h
e
I
n
d
iv
id
u
al
ac
c
u
r
ac
y
a
n
d
d
iv
er
s
it
y
w
it
h
v
er
y
s
en
s
iti
v
e
ac
c
u
r
ac
y
.
T
h
e
o
b
s
er
v
atio
n
h
as
b
ee
n
d
ep
lo
y
e
d
f
o
r
id
en
ti
f
y
i
n
g
n
o
cla
s
s
o
r
d
if
f
er
e
n
t
cla
s
s
es.
A
m
u
lti
-
la
y
er
ed
alg
o
r
ith
m
h
as
b
ee
n
g
en
er
ated
b
y
K.
A
r
u
tch
elv
an
et
a
l.
[
8
]
th
r
o
u
g
h
t
h
e
co
m
b
i
n
atio
n
o
f
DT
tech
n
iq
u
es.
T
h
is
alg
o
r
it
h
m
h
e
lp
s
to
d
iag
n
o
s
e
th
e
p
r
ed
ictio
n
o
f
ca
n
ce
r
r
is
k
an
d
o
th
er
cr
itical
d
is
ea
s
es.
Un
d
er
th
e
s
u
p
er
v
is
io
n
o
f
d
ata
m
in
i
n
g
ap
p
r
o
ac
h
es,
it e
asil
y
d
etec
ts
w
h
et
h
er
a
p
atien
t h
a
s
ca
n
ce
r
o
r
n
o
t.
T
h
e
r
esear
ch
er
s
co
llected
t
h
e
clin
ical
d
ataset
to
e
v
al
u
ate
t
h
e
g
r
a
m
m
atica
l
p
r
o
b
le
m
u
s
i
n
g
m
ac
h
in
e
lear
n
in
g
to
o
ls
.
Ku
m
ar
et
a
l.
s
h
o
w
ed
[
9
]
an
id
ea
o
n
th
e
b
r
e
ast
ca
n
ce
r
d
ataset
w
h
ic
h
h
elp
ed
to
e
r
ad
icate
th
e
ea
r
l
y
d
ea
th
r
is
k
b
ec
au
s
e
o
f
t
h
eir
co
m
p
r
eh
e
n
s
i
v
e
r
esear
c
h
.
Dif
f
er
en
t
d
ata
m
i
n
i
n
g
to
o
ls
w
er
e
d
ev
elo
p
ed
to
m
ak
e
a
p
r
ed
ictio
n
s
y
s
te
m
f
o
r
B
r
ea
s
t
ca
n
ce
r
b
y
A
.
F.
M.
Ag
ar
ap
[
1
0
]
.
Six
ap
p
r
o
ac
h
es
w
er
e
p
er
f
o
r
m
ed
o
n
t
h
is
s
y
s
te
m
to
ac
h
ie
v
e
s
p
ec
i
f
icit
y
a
n
d
r
ec
all
r
esu
lts
.
T
h
e
r
ep
o
r
ted
ac
cu
r
ac
y
ex
ce
ed
s
9
0
%
in
t
h
eir
p
r
o
p
o
s
ed
s
y
s
te
m
.
I
n
th
e
r
esea
r
ch
w
o
r
k
o
f
Na
u
ck
et
a
l.
[
1
1
]
9
5
.
5
7
%
o
f
ac
cu
r
ac
y
h
as
b
ee
n
s
h
o
w
n
w
i
th
t
h
e
f
u
zz
y
cl
u
s
ter
in
g
tech
n
iq
u
e.
P
.
G
u
p
ta
et
a
l.
e
x
p
lain
ed
a
tec
h
n
iq
u
e
o
n
t
h
e
ca
n
ce
r
d
atase
t
o
f
U
C
I
I
r
v
i
n
e
m
ac
h
i
n
e
lear
n
in
g
r
ep
o
s
ito
r
y
co
m
b
i
n
i
n
g
th
r
ee
al
g
o
r
ith
m
s
[
1
2
]
(
C
A
R
T
,
R
F,
a
n
d
KNN)
to
s
h
o
w
t
h
e
p
r
ed
icted
p
er
f
o
r
m
a
n
ce
.
KNN
m
o
d
el,
h
o
w
e
v
er
,
p
r
o
v
id
es
9
7
%
ac
cu
r
ac
y
a
m
o
n
g
all
o
f
t
h
e
m
.
Y.
K
h
o
u
r
d
i
f
i
et
a
l.
[
1
3
]
an
al
y
ze
d
t
h
e
ea
r
l
y
p
r
ed
ictio
n
r
ate
o
f
b
r
ea
s
t
ca
n
ce
r
d
ep
en
d
in
g
o
n
v
ar
io
u
s
cla
s
s
if
ier
s
,
s
u
c
h
as
NB
,
R
F,
SV
M,
an
d
KNN.
I
n
th
e
ir
s
tu
d
y
,
t
h
e
b
est
r
e
s
u
l
t
w
as
o
b
tain
ed
f
r
o
m
S
VM
ar
o
u
n
d
9
7
.
9
%.
C
.
Sh
r
a
v
y
a
et
a
l.
[
1
4
]
s
u
g
g
ested
a
d
ia
g
n
o
s
is
ap
p
r
o
ac
h
o
n
th
e
b
asis
o
f
ca
n
c
er
d
ataset
co
m
b
in
in
g
th
r
ee
cla
s
s
i
f
ier
s
s
u
c
h
as
L
R
,
SVM,
an
d
KNN
to
ev
alu
ate
th
e
p
er
f
o
r
m
a
n
ce
.
T
h
e
y
a
ls
o
i
llu
s
tr
ated
d
if
f
er
e
n
t
p
er
f
o
r
m
a
n
ce
in
d
ices
in
ter
m
s
o
f
ac
cu
r
a
c
y
,
p
r
ec
is
io
n
,
an
d
s
en
s
iti
v
it
y
.
T
h
e
h
i
g
h
est
p
r
ed
ictab
le
ac
cu
r
ac
y
o
f
9
2
.
0
7
%
w
as
g
en
er
ated
b
y
SVM
m
o
d
el.
N.
M.
A
li
et
a
l.
[
15
]
s
u
g
g
e
s
ted
a
m
o
d
el
th
at
w
as
i
m
p
le
m
en
ted
b
y
SVM
a
n
d
L
G
alg
o
r
ith
m
s
b
as
ed
o
n
B
o
r
u
ta
an
d
L
A
SS
O
F
S
tech
n
iq
u
es.
I
n
t
h
eir
ex
p
er
i
m
e
n
t,
t
h
e
b
est
p
er
f
o
r
m
an
ce
w
a
s
n
o
ticed
o
n
L
A
S
SO
b
y
L
G
(
9
8
.
6
1
%)
alg
o
r
ith
m
.
T
h
r
ee
s
tr
ateg
ies
f
o
r
th
e
id
en
tif
icat
io
n
o
f
ca
n
ce
r
s
ta
g
es
wer
e
test
ed
o
v
er
th
e
b
r
ea
s
t
ca
n
ce
r
d
ataset
b
y
V.
C
h
a
u
r
asia
et
a
l.
w
it
h
t
h
e
co
m
p
letio
n
o
f
th
e
p
r
e
-
p
r
o
ce
s
s
in
g
s
tep
,
an
a
v
er
ag
e
r
es
u
lt
w
as
ac
h
iev
ed
a
m
o
n
g
all
o
f
th
e
m
th
r
o
u
g
h
a
s
i
m
p
le
lo
g
i
s
tic
m
eth
o
d
[
16
]
,
b
u
t it
w
as a
b
o
u
t
to
7
4
.
4
7
%.
I
n
th
e
s
tu
d
y
o
f
[
1
2
-
1
4
]
,
th
e
y
e
v
alu
a
ted
th
e
o
v
er
all
r
es
u
lt
s
w
i
th
s
o
m
e
clas
s
if
ier
s
w
it
h
o
u
t
u
s
i
n
g
a
n
y
FS
ap
p
r
o
ac
h
.
Am
o
n
g
all
o
f
th
e
m
,
th
e
b
est
r
esu
lt
w
a
s
o
b
tain
ed
f
r
o
m
SVM
in
[
1
3
]
w
h
ic
h
w
as
9
7
.
9
%.
A
n
o
th
er
s
tu
d
y
[
17
]
illu
s
tr
ated
t
w
o
F
S
t
ec
h
n
iq
u
es
i
n
clu
d
i
n
g
L
ASSO
a
n
d
B
o
r
u
ta
an
d
g
o
t
9
8
.
6
1
%
b
y
L
R
.
Ho
w
e
v
er
,
o
u
r
s
y
s
te
m
h
as o
b
tai
n
ed
9
9
.
4
1
% b
y
R
F
u
s
in
g
t
h
e
L
ASS
O
FS
ap
p
r
o
ac
h
w
h
ich
o
u
tp
er
f
o
r
m
s
o
t
h
er
p
r
ev
io
u
s
s
t
u
d
ies.
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:
2
0
8
8
-
8708
E
xp
ert ca
n
ce
r
mo
d
el
u
s
in
g
s
u
p
ervis
ed
a
lg
o
r
ith
ms w
ith
a
LAS
S
O
s
elec
tio
n
a
p
p
r
o
a
ch
(
P
r
o
n
a
b
Gh
o
s
h
)
2633
2.
O
UR
P
RO
P
O
SE
D
AL
G
O
R
I
T
H
M
S
T
h
e
lear
n
in
g
tec
h
n
iq
u
e
o
f
a
m
ac
h
in
e
is
a
s
o
lu
t
io
n
t
h
at
s
u
p
p
o
r
ts
a
d
is
tin
ct
e
s
ti
m
atio
n
.
I
t
d
iv
id
es
t
h
e
d
ata
in
to
tr
ain
in
g
to
p
r
ed
ict
t
h
e
b
est
p
ar
am
e
ter
o
f
th
e
m
o
d
el
an
d
g
ath
er
ed
o
u
tco
m
e
s
ar
e
ap
p
lied
t
o
th
e
test
d
ata.
T
h
e
lear
n
in
g
a
l
g
o
r
i
t
h
m
[
1
8
]
k
e
e
p
s
o
n
u
p
g
r
a
d
i
n
g
i
t
s
e
l
f
o
p
t
i
m
u
m
p
r
e
d
i
c
t
i
o
n
a
n
d
i
n
t
e
r
p
r
e
t
a
t
i
o
n
o
f
n
ew
d
ata.
2
.
1
.
L
o
g
is
t
ic
re
g
re
s
s
io
n
L
o
g
i
s
tic
R
e
g
r
es
s
io
n
,
w
h
ich
is
also
ca
lled
a
s
tatis
tical
a
p
p
r
o
ac
h
,
h
elp
s
to
class
if
y
a
class
i
f
ied
v
ar
iab
le
[
1
9
]
.
B
y
est
i
m
at
in
g
th
e
p
r
o
b
ab
ilit
y
o
f
a
p
ar
ticu
lar
class
,
L
R
g
e
n
er
ates
a
m
o
d
el
th
at
d
i
f
f
er
en
tiate
s
b
et
w
ee
n
s
a
m
p
les.
Hav
in
g
t
w
o
d
if
f
er
e
n
t o
u
tco
m
e
s
,
1
r
ep
r
esen
ts
tr
u
e
an
d
0
r
ep
r
ese
n
ts
f
al
s
e.
Fig
u
r
e
1
s
h
o
w
s
th
e
w
o
r
k
i
n
g
ap
p
r
o
ac
h
[
20
]
.
Fig
u
r
e
1
.
An
alg
o
r
it
h
m
o
f
lo
g
i
s
tic
r
eg
r
es
s
io
n
T
h
r
ee
d
if
f
er
en
t
en
titi
es
h
a
v
e
b
ee
n
s
elec
ted
as
i
n
p
u
t
s
u
c
h
as
d
ata
s
et,
b
ase
lear
n
er
,
an
d
t
h
e
n
u
m
b
er
o
f
b
ase
lear
n
er
s
.
Af
ter
th
e
i
n
p
u
t,
b
ase
lear
n
er
is
s
elec
ted
f
o
r
t
r
ain
in
g
an
d
it
co
n
ti
n
u
es
till
it
r
ea
ch
es
th
e
u
p
p
er
li
m
it.
A
s
a
r
es
u
lt,
t
h
e
e
x
p
ec
te
d
o
u
tp
u
t
is
g
ai
n
ed
t
h
r
o
u
g
h
th
e
H(
x
)
v
ar
iab
le
w
h
er
e
b
ase
le
an
er
is
d
iv
id
ed
b
y
n
u
m
b
er
o
f
b
ase
lear
n
er
s
(
B
).
2
.
2
.
Dec
is
io
n
t
re
e
Dec
is
io
n
T
r
ee
u
s
es
h
ier
ar
ch
ic
al
tr
ee
ap
p
r
o
ac
h
es
in
w
h
er
e
ea
ch
n
o
d
e
illu
s
tr
ates
a
f
ea
tu
r
e,
a
b
r
an
ch
illu
s
tr
ates
t
h
e
r
u
le
o
f
d
ec
is
io
n
,
an
d
lea
v
es
s
h
o
w
a
n
o
u
tco
m
e
[
21
]
.
I
t
ca
n
b
e
ap
p
lied
f
o
r
class
i
f
icatio
n
a
n
d
r
eg
r
ess
io
n
tr
ee
s
p
r
o
b
lem
.
T
h
e
d
ataset
h
as
b
ee
n
d
i
v
id
ed
in
to
tw
o
s
u
b
g
r
o
u
p
s
to
s
h
o
w
t
h
e
ill
u
s
tr
atio
n
p
r
o
ce
s
s
o
f
C
A
R
T
.
An
o
v
er
all
id
ea
is
ill
u
s
tr
ated
in
th
e
eq
u
atio
n
s
[
2
2
].
Gain
o
f
i
n
f
o
r
m
atio
n
(
IG
)
,
I (N
,
P) =
−
(
+
l
og2
+
+
+
l
og2
+
)
(
1
)
Valu
e
o
f
E
n
tr
o
p
y
,
E
(
A
)
=
∑
(
+
+
)
∗
G
a
in
of
in
for
ma
tio
n
=
1
(
2
)
T
h
er
e
ar
e
tw
o
d
i
f
f
er
e
n
t
p
r
o
b
ab
ilit
ies
(
P
an
d
N)
,
th
at
h
av
e
b
ee
n
s
u
cc
e
s
s
f
u
ll
y
u
t
ilized
t
o
p
r
o
d
u
ce
in
f
o
r
m
atio
n
g
ai
n
(
I
G)
.
T
h
e
s
u
m
m
atio
n
s
o
f
p
r
o
b
ab
ilis
tic
o
u
t
co
m
e
s
ar
e
ca
lcu
lated
to
g
et
t
h
e
I
G
v
al
u
e
i
n
(
1
)
.
T
h
e
v
alu
e
o
f
e
n
tr
o
p
y
(
E
(
A
)
)
is
ca
lcu
la
ted
in
(
2
)
b
ased
o
n
I
G
v
alu
e.
I
G
v
al
u
e
is
m
u
ltip
lied
b
y
t
h
e
p
r
o
b
ab
ilis
tic
o
u
tco
m
es a
n
d
it is
co
n
t
in
u
ed
u
n
til to
g
et
n
v
a
lu
e.
2
.
3
.
Ra
nd
o
m
f
o
re
s
t
R
an
d
o
m
f
o
r
est
h
a
s
b
ee
n
d
ep
lo
y
ed
as
an
en
s
e
m
b
le
m
o
d
el
th
a
t
is
a
co
m
m
o
n
to
o
l
f
o
r
class
i
f
i
ca
tio
n
an
d
r
eg
r
ess
io
n
.
T
o
im
p
r
o
v
e
ac
c
u
r
ac
y
alo
n
g
w
i
th
o
v
er
f
it
tin
g
p
r
o
b
lem
s
,
it
co
m
b
i
n
e
s
m
u
lt
ip
le
DT
[
2
3
]
in
a
s
in
g
le
u
n
i
t.
E
ac
h
tr
ee
is
n
o
r
m
all
y
b
u
ilt
to
ac
h
ie
v
e
a
s
a
m
p
l
e
o
f
n
e
w
tr
ai
n
in
g
d
ata
w
i
th
a
n
av
er
ag
ed
v
alu
e
o
f
f
i
n
al
p
r
ed
ictio
n
.
2
.
4
.
Su
pp
o
rt
v
ec
t
o
r
m
a
chine
Su
p
p
o
r
t
Vec
to
r
Ma
ch
in
e
is
u
s
ed
as
a
tr
ain
i
n
g
m
et
h
o
d
to
s
tu
d
y
cla
s
s
i
f
icatio
n
an
d
r
eg
r
es
s
io
n
r
u
le
s
f
r
o
m
a
lar
g
e
n
u
m
b
er
o
f
d
ata.
An
o
p
ti
m
a
l
h
y
p
er
p
lan
e
[
2
4
]
is
b
ein
g
p
r
o
d
u
ce
d
to
ca
teg
o
r
ize
test
d
ata
b
y
SV
M
p
r
o
v
id
ed
a
s
et
o
f
lab
eled
tr
ain
in
g
i
s
av
a
ilab
le.
SVM
i
s
ex
t
en
s
i
v
el
y
u
s
ed
to
id
en
ti
f
y
th
e
s
tag
e
o
f
ca
n
ce
r
s
i
n
h
is
to
p
at
h
o
lo
g
y
i
m
a
g
es.
A
t
n
-
l
ev
els,
a
h
y
p
er
p
lan
e
is
d
escr
ib
e
d
in
(
3
)
[2
5
]
.
0
+
1
1
+
2
2
+
.
.
.
+
=
0
(
3
)
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
.
3
,
J
u
n
e
2
0
2
1
:
2
6
3
1
-
2639
2634
w
h
er
e
0
,
1
,
2
…
r
ep
r
esen
ts
h
y
p
o
t
h
etica
l
v
alu
e
s
an
d
Xn
s
h
o
w
s
d
ata
p
o
in
ts
in
t
h
e
s
a
m
p
le
s
p
ac
e
o
f
n
d
i
m
en
s
io
n
.
T
h
e
in
it
ial
in
te
n
ti
o
n
f
o
r
d
ev
elo
p
in
g
SVM
w
as
t
o
s
o
lv
e
2
-
clas
s
class
if
ica
tio
n
p
r
o
b
lem
,
later
o
n
,
it
w
a
s
t
u
n
ed
f
o
r
m
u
lt
i
-
cla
s
s
p
r
o
b
le
m
s
.
T
h
e
al
g
o
r
ith
m
ta
k
es
a
1
-
vs
-
r
es
t
ap
p
r
o
ac
h
w
h
er
e
i
t
at
te
m
p
ts
to
s
ep
ar
ate
a
s
in
g
le
cla
s
s
f
r
o
m
all
o
t
h
er
clas
s
es.
A
t
t
h
e
ti
m
e
o
f
te
s
ti
n
g
,
t
h
e
class
lab
el
o
f
z
o
f
a
cla
s
s
p
att
er
n
y
i
s
d
eter
m
i
n
ed
as
:
=
{
,
(
)
+
>
0
0
,
(
)
+
<
0
(
4
)
w
h
er
e
dn
(
y
)
=
{
dn
}
=
1
,
di
(
y)
i
s
th
e
d
is
ta
n
ce
f
r
o
m
y
to
th
e
SVM
h
y
p
er
p
lan
e
co
r
r
esp
o
n
d
in
g
to
class
i,
an
d
t
l
is
th
e
cla
s
s
i
f
icatio
n
th
r
e
s
h
o
ld
.
2
.
5
.
K
-
Nea
re
s
t
neig
hb
o
r
K
NN
is
t
h
e
m
o
s
t
co
m
m
o
n
l
y
u
s
ed
alg
o
r
ith
m
s
in
m
ac
h
in
e
lea
r
n
in
g
b
ec
au
s
e
o
f
its
f
le
x
ib
ilit
y
.
B
esid
es,
th
e
lear
n
i
n
g
s
ta
g
e
is
n
o
t
n
ec
es
s
ar
y
li
k
e
o
th
er
alg
o
r
it
h
m
s
[
2
6
]
.
KNN
is
ca
lled
a
class
i
f
ied
an
d
a
lazy
al
g
o
r
ith
m
in
d
ata
m
in
in
g
.
E
u
clid
ea
n
d
is
t
an
ce
is
s
h
o
w
n
in
t
h
e
eq
u
atio
n
is
:
[
2
7
]
T
w
o
d
ata
p
o
in
ts
ar
e
tak
en
f
o
r
E
u
clid
ea
n
d
is
tan
ce
.
(
,
)
=
√
∑
(
−
)
2
=
1
(
5
)
T
h
e
co
o
r
d
in
ate
o
f
y
p
o
i
n
t
is
s
u
b
tr
ac
ted
f
r
o
m
t
h
e
co
o
r
d
in
ato
r
o
f
x
p
o
in
t
u
n
d
er
th
e
r
o
o
t
s
q
u
ar
e
v
al
u
e.
T
h
e
o
b
tain
ab
le
r
esu
lt
is
ca
lle
d
E
u
clid
ea
n
d
i
s
tan
ce
li
k
e
D(
x
,
y
)
.
T
h
e
f
o
llo
w
i
n
g
d
ata
s
e
t
i
s
ca
te
g
o
r
ized
u
s
i
n
g
tr
ain
i
n
g
d
ata
.
3.
RE
S
E
ARCH
M
E
T
H
O
DO
L
O
G
Y
R
esear
ch
m
eth
o
d
o
lo
g
y
h
elp
s
to
o
b
tain
a
lo
g
ical
k
n
o
w
led
g
e
o
f
t
h
e
r
e
s
ea
r
ch
w
o
r
k
.
R
e
s
ea
r
ch
s
u
b
j
ec
t
an
d
in
s
tr
u
m
e
n
tatio
n
h
a
v
e
b
ee
n
ex
p
lai
n
ed
to
aid
in
t
h
e
e
s
tab
lis
h
m
e
n
t
o
f
a
clea
r
co
n
ce
p
t.
Si
n
ce
d
ata
is
t
h
e
m
o
s
t
s
ig
n
i
f
ica
n
t p
ar
t o
f
m
ac
h
i
n
e
lea
r
n
in
g
w
o
r
k
,
a
co
n
ce
p
tu
al
d
esc
r
ip
tio
n
h
as b
ee
n
ad
d
ed
to
th
e
d
ata
co
llectio
n
.
3
.
1
.
Da
t
a
s
et
co
llect
io
n
I
n
o
u
r
p
r
o
p
o
s
ed
s
y
s
te
m
,
W
is
co
n
s
i
n
b
r
ea
s
t
ca
n
ce
r
(d
iag
n
o
s
ti
c)
d
ataset
(
W
B
DC
)
[
2
8
]
g
ath
er
ed
f
r
o
m
UC
I
m
ac
h
i
n
e
lear
n
i
n
g
r
ep
o
s
it
o
r
y
h
as
b
ee
n
e
v
al
u
at
ed
to
p
r
ed
ict
th
e
ac
cu
r
ac
y
r
ate.
Mo
s
t
o
f
t
h
e
attr
ib
u
te
s
w
er
e
in
n
u
m
er
ic
v
al
u
es,
e
x
ce
p
t
d
ia
g
n
o
s
is
,
w
h
ic
h
h
as
b
ee
n
i
n
ca
te
g
o
r
ical
v
al
u
e.
As
a
r
es
u
lt,
w
e
h
av
e
co
n
v
er
ted
th
i
s
ca
teg
o
r
ical
v
a
lu
e
in
to
n
u
m
er
i
c
v
al
u
e
f
o
r
m
a
k
i
n
g
a
p
r
ed
icti
o
n
.
I
n
5
6
9
in
s
ta
n
ce
s
w
i
th
3
2
att
r
ib
u
tes,
3
5
7
ar
e
b
en
ig
n
clas
s
(
B
)
,
an
d
2
1
2
ar
e
m
ali
g
n
a
n
t (
M)
clas
s
.
T
w
o
clas
s
es o
f
ca
n
ce
r
d
is
ea
s
e
ar
e
s
h
o
wn
in
T
ab
le
1
.
T
ab
le
1
.
A
ttrib
u
te,
an
d
t
y
p
e
o
f
v
alu
e
s
S
e
r
i
a
l
N
u
mb
e
r
s
A
t
t
r
i
b
u
t
e
s o
f
B
r
e
a
st
C
a
n
c
e
r
S
e
r
i
a
l
N
u
mb
e
r
s
A
t
t
r
i
b
u
t
e
s o
f
B
r
e
a
st
C
a
n
c
e
r
S
e
r
i
a
l
N
u
mb
e
r
s
A
t
t
r
i
b
u
t
e
s o
f
B
r
e
a
st
C
a
n
c
e
r
1
I
D
n
u
mb
e
r
12
f
r
a
c
t
a
l
_
d
i
me
n
si
o
n
_
me
a
n
23
r
a
d
i
u
s
_
w
o
r
st
2
D
i
a
g
n
o
si
s
13
r
a
d
i
u
s
_
se
24
t
e
x
t
u
r
e
_
w
o
r
st
3
r
a
d
i
u
s
_
me
a
n
14
t
e
x
t
u
r
e
_
se
25
p
e
r
i
me
t
e
r
_
w
o
r
st
4
t
e
x
t
u
r
e
_
me
a
n
15
p
e
r
i
me
t
e
r
_
se
26
a
r
e
a
_
w
o
r
st
5
p
e
r
i
me
t
e
r
_
me
a
n
16
a
r
e
a
_
se
27
smo
o
t
h
n
e
ss
_
w
o
r
st
6
a
r
e
a
_
me
a
n
17
smo
o
t
h
n
e
ss
_
se
28
c
o
mp
a
c
t
n
e
ss_
w
o
r
st
7
smo
o
t
h
n
e
s_
me
a
n
18
c
o
mp
a
c
t
n
e
ss_
se
29
c
o
n
c
a
v
i
t
y
_
w
o
r
st
8
c
o
mp
a
c
t
n
e
ss_
me
a
n
19
c
o
n
c
a
v
i
t
y
_
se
30
c
o
n
c
a
v
e
p
o
i
n
t
s
_
w
o
r
st
9
c
o
n
c
a
v
i
t
y
_
me
a
n
20
c
o
n
c
a
v
e
p
o
i
n
t
s
_
se
31
sy
mm
e
t
r
y
_
w
o
r
st
10
c
o
n
c
a
v
e
_
p
o
i
n
t
s
_
me
a
n
21
sy
mm
e
t
r
y
_
se
32
f
r
a
c
t
a
l
_
d
i
me
n
si
o
n
_
w
o
r
st
11
sy
mm
e
t
r
y
_
me
a
n
22
f
r
a
c
t
a
l
d
i
me
n
si
o
n
_
se
3
.
2
.
Da
t
a
prepro
ce
s
s
ing
T
h
e
ap
p
lied
d
ataset
is
p
ick
e
d
f
r
o
m
th
e
U
C
I
m
ac
h
i
n
e
lea
r
n
in
g
r
ep
o
s
ito
r
y
to
d
etec
t
th
e
s
tag
e
s
o
f
ca
n
ce
r
d
is
ea
s
e
a
n
d
s
tan
d
ar
d
s
ca
ler
[
2
9
]
a
p
p
r
o
ac
h
h
as
b
ee
n
ad
d
r
ess
ed
to
k
ee
p
th
e
m
i
n
t
h
e
r
an
g
e
o
f
[
0
,
1
]
.
Af
ter
w
ar
d
,
w
e
h
av
e
to
co
n
v
er
t
o
n
e
ca
te
g
o
r
ical
f
ea
t
u
r
e
s
u
ch
as
‘
Dia
g
n
o
s
is
’
in
to
n
u
m
b
er
s
b
y
u
s
i
n
g
lab
el
en
co
d
in
g
[
30
]
tech
n
iq
u
e.
Fo
r
ex
a
m
p
le,
w
e
lab
el
t
h
e
‘
B
en
ig
n
’
an
d
‘
Ma
li
g
n
a
n
t
’
as 0
an
d
1
r
esp
ec
tiv
el
y
.
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:
2
0
8
8
-
8708
E
xp
ert ca
n
ce
r
mo
d
el
u
s
in
g
s
u
p
ervis
ed
a
lg
o
r
ith
ms w
ith
a
LAS
S
O
s
elec
tio
n
a
p
p
r
o
a
ch
(
P
r
o
n
a
b
Gh
o
s
h
)
2635
3
.
3
.
An e
x
pect
ed
o
utc
o
m
e
o
n a
s
e
lect
ed
f
ea
t
ure
s
elec
t
io
n a
lg
o
r
it
h
m
Featu
r
e
s
elec
tio
n
[
31
]
is
a
te
ch
n
iq
u
e
t
h
at
h
elp
s
to
s
elec
t
t
h
e
ap
p
r
o
p
r
iate
f
ea
tu
r
es
f
o
r
g
ettin
g
t
h
e
h
ig
h
e
s
t
o
u
tco
m
es
b
ased
o
n
t
h
e
g
at
h
er
ed
d
ataset.
B
ef
o
r
e
p
er
f
o
r
m
i
n
g
a
d
ata
e
x
p
er
i
m
e
n
t,
t
h
e
s
elec
tio
n
p
r
o
ce
s
s
o
f
t
h
e
f
u
n
ctio
n
m
u
s
t
e
x
a
m
i
n
e
th
e
d
ata
s
et.
T
h
e
s
elec
tio
n
o
f
f
ea
tu
r
es
i
n
t
h
is
f
r
a
m
e
w
o
r
k
i
s
o
n
l
y
u
s
ed
f
o
r
i
m
p
r
o
v
i
n
g
m
o
d
el
e
f
f
icie
n
c
y
,
a
n
d
also
h
elp
s
to
r
ed
u
ce
e
x
ec
u
t
io
n
ti
m
e.
W
e
u
s
ed
o
n
e
o
f
th
e
e
m
b
ed
d
ed
m
eth
o
d
s
s
u
c
h
as
t
h
e
L
A
S
SO
s
tr
ateg
y
.
Usi
n
g
a
r
an
d
o
m
l
y
g
e
n
er
ated
s
u
b
s
et
o
f
k
e
y
w
o
r
d
s
f
r
o
m
t
h
e
co
r
r
esp
o
n
d
in
g
s
u
b
-
r
eg
io
n
,
th
e
e
f
f
icien
c
y
o
f
t
h
is
f
u
n
ctio
n
ca
n
b
e
i
m
p
r
o
v
ed
b
y
r
ep
ea
tin
g
th
e
ab
o
v
e
p
r
o
ce
d
u
r
e.
I
t
is
ca
lled
th
e
r
an
d
o
m
ized
L
ASSO
f
u
n
ctio
n
th
at
w
a
s
i
n
tr
o
d
u
ce
d
W
an
g
[
3
2
]
.
I
n
ad
d
itio
n
,
L
A
SS
O
is
c
o
n
s
id
er
ed
th
e
m
o
s
t
s
ig
n
i
f
ica
n
t f
ea
tu
r
e
co
n
tai
n
ed
i
n
q
i
w
h
ic
h
s
y
m
b
o
lize
s
th
e
v
ec
to
r
o
f
th
e
s
i
m
ilar
s
u
b
-
r
e
g
io
n
k
e
y
s
i
n
Fi
g
u
r
e
2
.
3
.
3
.
1
.
L
ea
s
t
a
bs
o
lute
s
hrin
k
a
g
e
a
nd
s
elec
t
io
n o
pera
t
o
r
f
e
a
t
ure
s
elec
t
io
n a
lg
o
rit
h
m
(
L
A
SS
O
)
T
h
e
L
A
S
SO
f
u
n
ctio
n
alit
y
o
f
th
e
o
p
er
ato
r
is
d
ep
en
d
en
t
u
p
o
n
u
p
d
atin
g
t
h
e
ab
s
o
lu
te
v
a
l
u
e
o
f
th
e
f
u
n
ctio
n
co
ef
f
icien
t.
Var
io
u
s
co
ef
f
icie
n
t
r
an
k
s
o
f
t
h
e
ch
ar
a
cter
is
tics
ar
e
ze
r
o
,
an
d
th
o
s
e
ch
ar
ac
ter
is
tic
s
w
it
h
n
eg
at
iv
e
co
e
f
f
icie
n
ts
ar
e
r
e
m
o
v
ed
f
r
o
m
t
h
e
s
u
b
s
e
t
o
f
f
ea
t
u
r
es.
T
h
e
L
ASSO
p
er
f
o
r
m
s
w
ell
w
i
th
lo
w
co
ef
f
icie
n
c
y
f
u
n
ct
io
n
v
al
u
es.
A
s
u
b
s
et
o
f
d
esire
d
f
u
n
ctio
n
s
in
cl
u
d
in
g
ir
r
elev
a
n
t
f
ea
t
u
r
es
m
a
y
b
e
s
elec
ted
i
n
th
e
L
A
SS
O
ap
p
r
o
ac
h
[
3
3
].
T
h
e
L
ASSO
s
elec
ts
clo
s
el
y
r
el
ated
ch
ar
ac
ter
is
tics
t
h
at
ar
e
to
b
e
v
ie
w
ed
as
tr
u
e,
an
d
th
e
r
est
as
f
al
s
e.
Af
ter
co
m
p
leti
n
g
t
h
e
ev
al
u
atio
n
p
r
o
ce
s
s
o
f
L
ASS
O,
W
e
g
et
f
o
u
r
es
s
en
tia
l
f
ea
t
u
r
es
th
a
t
h
av
e
b
ee
n
clea
r
l
y
s
h
o
w
n
i
n
Fi
g
u
r
e
3
.
Ov
er
all,
te
x
tu
r
e_
w
o
r
s
t
co
n
tain
s
th
e
h
i
g
h
e
s
t sco
r
e
(
0
.
0
1
7
4
8
)
.
Fig
u
r
e
2
.
T
h
e
w
o
r
k
in
g
p
r
o
ce
d
u
r
e
o
f
L
ASSO a
l
g
o
r
it
h
m
Fig
u
r
e
3
.
T
h
e
g
r
ap
h
ical
v
ie
w
o
f
s
elec
ti
v
e
f
ea
tu
r
es b
y
L
ASS
O
alg
o
r
ith
m
3
.
4
.
G
ra
ph
ica
l r
epre
s
ent
a
t
io
n o
f
pro
po
s
ed
m
o
del
T
h
e
ca
n
ce
r
d
ataset
h
as
b
ee
n
co
llected
th
r
o
u
g
h
an
o
n
li
n
e
r
ep
o
s
ito
r
y
to
d
etec
t
th
e
d
iag
n
o
s
e
r
ate
o
f
ca
n
ce
r
.
Sin
ce
t
h
e
co
llected
d
ataset
o
f
ca
n
ce
r
h
a
s
n
o
m
is
s
i
n
g
v
al
u
e,
it
is
d
ir
ec
tl
y
t
r
an
s
m
itted
to
th
e
p
r
ep
r
o
ce
s
s
in
g
tec
h
n
iq
u
e.
I
n
t
h
is
tech
n
iq
u
e,
1
0
-
f
o
ld
C
V
a
p
p
r
o
ac
h
is
ta
k
en
f
o
r
a
n
e
x
p
er
im
en
t
to
d
ea
l
w
i
t
h
o
v
er
f
itti
n
g
an
d
u
n
d
er
f
i
tti
n
g
i
s
s
u
e
s
.
Af
ter
s
u
cc
es
s
f
u
l
l
y
ap
p
l
y
in
g
f
i
v
e
al
g
o
r
ith
m
s
to
t
h
e
g
i
v
en
d
ataset,
th
e
m
o
s
t
s
u
itab
le
o
u
tco
m
e
is
r
ec
ei
v
ed
f
r
o
m
R
F
-
b
ased
o
n
v
alid
ati
o
n
d
ataset
a
m
o
n
g
all
alg
o
r
i
th
m
s
.
T
h
e
o
v
er
a
ll
d
escr
ip
tio
n
p
r
o
ce
s
s
u
s
i
n
g
p
s
eu
d
o
co
d
e
h
as b
ee
n
s
h
o
w
n
as a
g
r
ap
h
ical
f
o
r
m
at
i
n
Fi
g
u
r
e
4
.
Step 1: Function Pre
-
processing ()
Step 2:
import dataset
Step 3:
Convert dataset to csv format
Step 4:
Label encoder for categorical values
Step 5: End Pre
-
p
rocessing ()
Step 6: Function SpecificFeaturesSelection ()
Step 7:
Go to Step 1
Step 8: Choose selection model
Step 9: Apply LASSO selection algorithm on preprocessed dataset
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
.
3
,
J
u
n
e
2
0
2
1
:
2
6
3
1
-
2639
2636
Step 10: End SpecificFeature
sSelection ()
Step 11: Function TrainTestSplit ()
Step 12:
Go to Step 7
Step 13: Choose 10
-
fold cross
-
validation method
Step 14: End TrainTestSplit ()
Step 15: Function BestClassifiersSelectionApproach ()
Step 16:
Choose 5 algorithms (LR, DT, RF, SVM, and KNN)
Step 17: For i = 0: 5
Step 18: Go to Step 12
Step 19:
Predict class of data
Step 20:
Evaluate result
Step 21: Comparison among overall outcomes
Step 22: Recommend the best model over the dataset by LASSO technique
Step 23: End BestCLassifiersSelectionApproach ()
3
.
5
.
Va
lid
a
t
io
n t
ec
hn
iqu
e
o
f
cla
s
s
if
iers
I
n
k
-
f
o
ld
cr
o
s
s
-
v
alid
atio
n
,
t
h
e
co
llected
d
ata
w
as
b
r
o
k
e
n
d
o
w
n
i
n
to
k
eq
u
al
p
ar
ts
w
h
er
e
k
-
1
ca
teg
o
r
ies
h
a
v
e
b
ee
n
ch
o
s
en
to
tr
ain
t
h
e
m
o
d
els
a
n
d
o
t
h
er
p
ar
ts
ar
e
e
v
al
u
ated
i
n
e
ac
h
p
h
a
s
e
to
tes
t
p
er
f
o
r
m
a
n
ce
.
T
h
e
p
r
o
ce
s
s
o
f
v
alid
atio
n
is
iter
ated
f
o
r
k
-
ti
m
es.
T
h
e
class
i
f
ier
's
e
f
f
icien
c
y
is
ca
lc
u
lated
b
y
th
e
r
esu
lt
s
o
f
k
.
Var
io
u
s
v
al
u
es
o
f
k
ar
e
c
h
o
s
en
f
o
r
C
V.
W
e'v
e
o
n
l
y
u
s
ed
k
=
1
0
i
n
o
u
r
e
x
p
er
i
m
e
n
t
b
ec
au
s
e
t
h
eir
o
u
tp
u
t
i
s
g
o
o
d
[
3
4
]
.
I
n
th
e
1
0
-
f
o
ld
C
r
o
s
s
-
v
alid
atio
n
s
y
s
t
e
m
,
d
ata
is
8
0
%
allo
ca
ted
f
o
r
tr
ain
in
g
a
n
d
th
e
r
e
m
ai
n
in
g
f
o
r
ex
p
er
i
m
en
tal
p
u
r
p
o
s
es.
Fo
r
ea
ch
f
o
ld
,
th
e
p
r
o
ce
s
s
w
a
s
ca
r
r
ied
o
v
er
f
o
r
1
0
ti
m
es;
e
v
er
y
d
ata
p
o
in
ts
in
t
h
e
tes
tin
g
a
n
d
tr
ain
in
g
s
ets
w
er
e
r
an
d
o
m
l
y
d
is
tr
ib
u
ted
o
v
er
th
e
en
tire
d
ataset
b
ef
o
r
e
th
e
s
elec
tio
n
tr
ain
i
n
g
an
d
te
s
ti
n
g
o
f
n
e
w
s
ets
f
o
r
ea
ch
it
er
atio
n
.
Af
ter
t
h
e
co
m
p
let
io
n
o
f
1
0
-
f
o
ld
p
er
io
d
,
av
er
ag
es
o
f
al
l
p
er
f
o
r
m
a
n
ce
m
etr
ics ar
e
ca
lcu
lated
.
Fig
u
r
e
4
.
Sk
eleto
n
o
f
th
e
p
r
o
p
o
s
ed
ca
n
ce
r
d
etec
tio
n
m
o
d
el
3
.
6
.
P
er
f
o
r
m
a
nce
m
ea
s
ure
i
nd
ice
s
T
h
e
p
er
f
o
r
m
an
ce
a
n
d
co
r
r
ec
tn
es
s
h
a
v
e
b
ee
n
s
u
r
v
e
y
ed
d
u
e
to
d
o
m
ea
s
u
r
in
g
s
o
m
e
p
er
f
o
r
m
an
ce
in
d
ices
to
r
ed
u
ce
th
e
d
ea
th
r
i
s
k
[
3
5
]
u
s
i
n
g
m
ac
h
i
n
e
lear
n
i
n
g
m
et
h
o
d
s
.
T
h
is
f
o
r
m
u
la
h
as
b
ee
n
ap
p
lied
to
f
i
n
d
A
cc
,
P
_
s
co
r
e,
R
_
s
co
r
e,
an
d
Sp
e.
[
3
6
].
4.
E
XP
E
R
I
M
E
NT
A
L
RE
SUL
T
S AN
D
D
I
SC
USS
I
O
N
Fiv
e
al
g
o
r
ith
m
s
ar
e
u
s
ed
to
p
r
ed
ict
b
r
ea
s
t
ca
n
ce
r
o
u
tco
m
es
o
n
s
elec
ted
f
ea
tu
r
e
s
(
p
er
i
m
eter
_
m
ea
n
r
ad
iu
s
_
w
o
r
s
t,
tex
tu
r
e_
w
o
r
s
t,
p
er
im
e
ter
_
w
o
r
s
t)
b
y
t
h
e
L
A
S
SO
F
S
al
g
o
r
ith
m
t
h
at
h
elp
s
to
p
r
ed
ict
th
e
b
est
o
u
tco
m
e
o
v
er
t
h
e
e
x
tr
ac
ted
co
lu
m
n
s
.
T
o
ev
alu
ate
v
ar
io
u
s
m
et
h
o
d
s
,
s
o
m
e
p
er
f
o
r
m
a
n
c
e
m
etr
ics
s
u
c
h
a
s
C
o
n
f
u
s
io
n
Ma
tr
ix
,
A
cc
,
FP
R
,
FN
R
,
Sp
e,
an
d
P
_
Sco
r
e
.
h
av
e
b
ee
n
u
s
ed
to
ev
al
u
ate
v
ar
io
u
s
m
o
d
els
o
f
alg
o
r
ith
m
s
.
Am
o
n
g
t
h
e
m
,
R
F
h
as p
r
o
v
id
ed
th
e
h
i
g
h
est ac
c
u
r
ac
y
f
o
r
b
o
th
tr
ain
i
n
g
an
d
test
i
n
g
d
ata.
4
.
1
.
E
x
peri
m
e
nta
l c
o
ns
equence
a
m
o
ng
s
ev
er
a
l a
lg
o
rit
h
m
s
L
R
w
h
ich
is
ca
l
led
s
tati
s
tical
m
ac
h
in
e
lear
n
i
n
g
tech
n
iq
u
e,
o
n
th
e
o
t
h
er
h
an
d
,
SV
M
ca
n
b
e
u
s
ed
i
n
b
o
th
class
i
f
icatio
n
a
n
d
r
eg
r
es
s
io
n
.
W
e
h
av
e
o
b
tain
ed
f
r
o
m
b
o
th
9
8
.
2
4
%
ac
c,
9
7
%
R
_
s
c
o
r
e,
9
8
%
s
co
r
e
f
o
r
b
o
th
Sp
e
an
d
P_
s
co
r
e,
0
%
is
ac
h
iev
ed
b
y
FP
R
an
d
FD
R
.
DT
w
o
r
k
s
li
k
e
a
b
in
ar
y
tr
ee
w
h
er
e
ev
er
y
d
ata
p
o
in
t
d
en
o
tes
an
a
ttrib
u
te,
an
d
K
NN
is
co
n
s
id
er
ed
a
n
o
n
-
p
ar
a
m
etr
ic
al
g
o
r
ith
m
a
lo
n
g
w
it
h
clas
s
i
f
icatio
n
a
n
d
r
eg
r
ess
io
n
u
s
a
g
e.
W
e
h
a
v
e
ac
h
iev
ed
a
s
i
m
ilar
r
es
u
lt,
w
h
er
e
ac
c
is
f
o
u
n
d
a
t
9
6
.
7
5
%
.
O
t
h
e
r
p
e
r
f
o
r
m
a
n
c
e
i
n
d
i
c
e
s
s
u
c
h
a
s
N
P
V
,
R
_
s
c
o
r
e
,
P
_
s
c
o
r
e
,
a
n
d
F
N
R
a
r
e
s
h
o
w
n
9
7
.
4
1
%
,
9
8
%
,
9
8
%
,
a
n
d
2
%
r
e
s
p
e
c
t
i
v
e
l
y
f
o
r
d
is
ea
s
e
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:
2
0
8
8
-
8708
E
xp
ert ca
n
ce
r
mo
d
el
u
s
in
g
s
u
p
ervis
ed
a
lg
o
r
ith
ms w
ith
a
LAS
S
O
s
elec
tio
n
a
p
p
r
o
a
ch
(
P
r
o
n
a
b
Gh
o
s
h
)
2637
p
r
ed
ictio
n
.
R
F
h
as
g
o
t
t
h
e
h
ig
h
est
Acc
(
9
7
.
4
1
%)
w
it
h
t
h
e
lo
w
e
s
t
er
r
o
r
r
a
t
e
[
17
]
c
o
m
p
a
r
a
t
i
v
e
l
y
t
h
a
n
o
t
h
e
r
s
.
T
h
i
s
i
s
t
h
e
h
i
g
h
e
s
t
p
r
e
d
i
c
t
a
b
l
e
s
c
o
r
e
r
e
g
a
r
d
i
n
g
b
r
e
a
s
t
c
a
n
c
e
r
d
a
t
a
s
e
t
.
A
s
h
o
r
t
e
x
p
l
a
n
a
t
i
o
n
i
s
a
d
d
e
d
i
n
T
ab
le
2
.
T
ab
le
2
.
Ma
ch
in
e
lear
n
i
n
g
cla
s
s
i
f
ier
s
p
er
f
o
r
m
a
n
ce
o
n
10
-
f
o
l
d
cr
o
s
s
-
v
al
id
atio
n
tech
n
iq
u
e
D
i
me
n
si
o
n
L
o
g
i
st
i
c
R
e
g
r
e
ssi
o
n
D
e
c
i
si
o
n
T
r
e
e
R
a
n
d
o
m F
o
r
e
st
S
u
p
p
o
r
t
V
e
c
t
o
r
M
a
c
h
i
n
e
K
-
N
e
a
r
e
st
N
e
i
g
h
b
o
r
C
o
n
f
u
s
i
o
n
mat
r
i
x
[
6
1
.
6
%
0
%
]
[
3
%
3
5
.
4
%
]
[
6
1
.
7
%
0
%
]
[
2
%
3
6
.
3
%
]
[
6
1
.
7
%
0
%
]
[
1
%
3
7
.
3
%
]
[
6
1
.
6
%
0
%
]
[
3
%
3
5
.
4
%
]
[
6
1
.
7
%
0
%
]
[
2
%
3
6
.
3
%
]
A
c
c
9
8
.
2
4
%
9
8
.
8
8
%
9
9
.
4
1
%
9
8
.
2
4
%
9
8
.
8
8
%
P
_
sco
r
e
9
8
%
9
8
%
9
9
%
9
8
%
9
8
%
R
_
s
c
o
r
e
9
7
%
9
8
%
9
9
%
9
7
%
9
8
%
S
p
e
9
8
%
9
9
%
9
9
%
9
8
%
9
9
%
N
P
V
9
8
%
9
7
.
4
1
%
9
8
.
4
1
%
9
8
%
9
7
.
4
1
%
F
D
R
0%
0%
0%
0%
0%
F
P
R
0%
0%
0%
0%
0%
F
N
R
2
.
9
%
2%
0
.
9
%
2
.
9
%
2%
4
.
2
.
Co
m
pa
ring
w
it
h e
x
is
t
ing
s
y
s
t
e
m
s
a
cc
ura
cy
T
ab
le
3
p
r
o
v
id
es
a
co
m
p
ar
ativ
e
v
ie
w
i
n
ter
m
s
o
f
A
cc
u
r
ac
y
ac
h
iev
ed
b
y
v
ar
io
u
s
al
g
o
r
it
h
m
s
w
i
th
ex
is
t
in
g
s
y
s
te
m
s
.
W
e
ca
n
ea
s
il
y
s
ee
f
r
o
m
t
h
e
tab
le
th
at
t
h
e
p
r
ed
ictio
n
r
ate
o
f
o
u
r
p
r
o
p
o
s
ed
s
y
s
te
m
i
s
v
er
y
h
i
g
h
r
ath
er
th
a
n
p
r
ev
io
u
s
w
o
r
k
s
e
x
p
lain
ed
r
e
f
er
r
in
g
to
[
1
2
-
15
]
.
I
n
th
e
tab
le,
b
o
th
L
R
a
n
d
S
VM
s
h
o
w
a
s
i
m
ilar
ac
cu
r
ac
y
o
f
9
8
.
2
4
%
f
r
o
m
o
u
r
s
y
s
te
m
.
L
R
p
r
o
v
id
es
9
2
.
1
0
%
f
o
r
[
1
4
]
,
7
3
.
6
1
%
an
d
9
8
.
6
1
%
f
o
r
[
3
5
]
th
at
is
n
o
ticed
f
r
o
m
B
o
r
tu
a
a
n
d
L
A
S
SO
FS
tech
n
iq
u
e
s
,
w
h
er
ea
s
S
VM
is
f
o
u
n
d
at
ab
o
u
t
9
7
.
9
%
[
1
3
]
,
9
2
.
7
8
%
[
1
4
]
,
6
9
.
4
4
%
an
d
5
9
.
7
2
%
f
o
r
[
3
5
]
th
at
is
tak
e
n
f
r
o
m
B
o
r
tu
a
a
n
d
L
ASSO
F
S
tec
h
n
iq
u
es
r
esp
e
ctiv
el
y
.
C
o
n
ce
r
n
in
g
R
F,
o
u
r
1
0
-
f
o
ld
C
V
tec
h
n
iq
u
e
h
as
ac
h
ie
v
ed
a
b
etter
p
r
ed
ictio
n
r
ate
o
f
9
9
.
4
1
%,
w
h
ich
is
t
h
e
h
i
g
h
e
s
t
o
u
tp
u
t
o
f
o
u
r
m
o
d
el,
co
m
p
ar
ed
to
r
ec
e
n
t
w
o
r
k
s
o
f
[
1
2
]
(
9
6
.
4
7
%).
R
eg
ar
d
in
g
DT
an
d
KNN,
o
u
r
s
y
s
te
m
o
u
tp
er
f
o
r
m
s
th
an
al
l
g
i
v
en
e
x
is
tin
g
tec
h
n
i
q
u
es.
T
h
e
p
r
ed
icted
a
cc
u
r
ac
y
is
r
ec
eiv
ed
9
8
.
8
8
%,
o
n
th
e
co
n
tr
ar
y
,
9
2
.
3
5
%
is
f
o
u
n
d
f
o
r
DT
in
[
1
2
]
an
d
th
e
o
u
tco
m
es
o
f
KNN
i
s
o
b
tain
ed
9
7
%
[
1
2
]
,
9
6
.
1
%
[
1
3
]
an
d
9
2
.
2
3
%
[
1
4
]
.
Fin
all
y
,
w
e
ca
n
ea
s
il
y
s
a
y
t
h
at
i
n
d
iv
i
d
u
als
w
it
h
t
h
ese
f
ea
t
u
r
es
co
n
t
ain
h
ig
h
r
is
k
s
o
f
b
ein
g
af
f
ec
t
ed
b
y
b
r
ea
s
t
ca
n
ce
r
w
h
ic
h
h
a
s
b
ee
n
b
r
ief
l
y
d
escr
ib
ed
in
th
e
o
v
er
all
co
n
tex
t.
T
ab
le
3
.
C
o
m
p
ar
is
o
n
o
f
th
e
p
r
o
p
o
s
ed
s
y
s
te
m
w
i
th
e
x
i
s
ti
n
g
s
y
s
te
m
s
M
o
d
e
l
s
A
c
c
u
r
a
c
y
o
f
1
0
-
f
o
l
d
c
r
o
ss
v
a
l
i
d
a
t
i
o
n
me
t
h
o
d
Ex
i
st
i
n
g
S
y
st
e
ms [
1
2
]
Ex
i
st
i
n
g
S
y
st
e
ms [
1
3
]
Ex
i
st
i
n
g
S
y
st
e
ms [
1
4
]
Ex
i
st
i
n
g
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
[
3
6
]
LR
9
8
.
2
4
%
-
-
9
2
.
1
0
%
B
o
r
u
t
a
(
7
3
.
6
1
%),
L
A
S
S
O
(
9
8
.
6
1
%)
DT
9
8
.
8
8
%
9
2
.
3
5
%
-
-
-
RF
9
9
.
4
1
%
9
6
.
4
7
%
9
6
%
-
-
S
V
M
9
8
.
2
4
%
-
9
7
.
9
%
9
2
.
7
8
%
B
o
r
u
t
a
(
6
9
.
4
4
%),
L
A
S
S
O
(
5
9
.
7
2
%)
K
N
N
9
8
.
8
8
%
9
7
%
9
6
.
1
%
9
2
.
2
3
%
-
5.
CO
NCLU
SI
O
N
AND
F
U
T
U
RE
WO
RK
W
e
h
av
e
s
t
u
d
ied
t
h
e
u
s
e
o
f
d
if
f
er
e
n
t
M
L
to
o
ls
to
p
r
ed
ict
th
e
ea
r
l
y
d
iag
n
o
s
i
s
r
ate.
All
o
f
t
h
e
s
e
tech
n
iq
u
es
t
h
r
o
u
g
h
t
h
e
L
ASS
O
f
ea
tu
r
e
s
elec
t
io
n
h
a
v
e
b
ee
n
ev
al
u
ated
f
o
r
g
etti
n
g
a
m
o
r
e
o
p
tim
i
s
tic
r
esu
lt.
Af
ter
t
h
e
e
f
f
ec
ti
v
e
f
ea
tu
r
e
s
el
ec
tio
n
s
tep
s
,
a
r
ath
er
p
r
o
m
i
s
i
n
g
o
u
tco
m
e
h
as
b
ee
n
o
b
tain
ed
f
r
o
m
R
F
a
lg
o
r
it
h
m
w
it
h
9
9
.
4
1
%
ac
cu
r
ac
y
i
n
co
m
p
ar
is
o
n
to
o
th
er
all
tec
h
n
iq
u
e
s
.
T
o
ad
d
r
ess
p
r
o
b
a
b
le
o
v
er
f
it
tin
g
is
s
u
e
s
,
w
e
ar
e
alr
ea
d
y
tr
y
in
g
to
co
llect
a
la
r
g
e
n
u
m
b
er
o
f
d
atasets
f
o
r
ca
lcu
lati
n
g
th
e
p
er
f
o
r
m
a
n
ce
ev
en
m
o
r
e
p
r
ec
is
el
y
.
Af
ter
all,
o
u
r
p
r
o
p
o
s
ed
tech
n
iq
u
e
h
a
s
b
ee
n
s
u
cc
ee
d
ed
in
g
en
er
atin
g
m
o
r
e
s
ec
u
r
e
an
d
e
f
f
icie
n
t
r
es
u
lt
s
w
it
h
v
er
y
lo
w
er
r
o
r
r
ates.
I
n
f
u
t
u
r
e,
w
e
w
il
l
d
ev
elo
p
a
n
o
n
li
n
e
an
d
r
o
id
ap
p
to
s
h
o
w
t
h
e
r
el
ev
an
t
s
y
m
p
to
m
s
o
f
b
r
ea
s
t c
an
ce
r
at
th
e
ea
r
lies
t a
s
a
to
o
l f
o
r
ea
r
ly
d
etec
tio
n
o
f
s
u
ch
t
y
p
e
o
f
ca
n
ce
r
.
ACK
NO
WL
E
D
G
E
M
E
NT
S
T
h
e
au
th
o
r
s
ar
e
g
r
atef
u
l to
Da
f
f
o
d
il
I
n
t
e
r
n
a
t
i
o
n
a
l
U
n
i
v
e
r
s
i
t
y
,
D
h
a
k
a
,
B
a
n
g
l
a
d
e
s
h
f
o
r
a
l
l
t
h
e
s
u
p
p
o
r
t.
RE
F
E
R
E
NC
E
S
[1
]
“
W
o
rld
He
a
lt
h
Org
a
n
iza
ti
o
n
(W
HO
),
”
2
0
2
0
.
[
On
l
in
e
]
.
Av
a
il
a
b
le:
h
tt
p
s:/
/www
.
w
h
o
.
in
t/
c
a
n
c
e
r/p
re
v
e
n
ti
o
n
/d
iag
n
o
siss
c
re
e
n
in
g
/b
re
a
stc
a
n
c
e
r/en
/.
[2
]
“
W
o
m
e
n
Bre
a
st Can
c
e
r,
”
2
0
2
0
.
[On
li
n
e
]
.
A
v
a
il
a
b
le:
h
tt
p
s://
se
e
r.
c
a
n
c
e
r.
g
o
v
/statfa
c
ts/h
tm
l/
b
r
e
a
st.h
t
m
l.
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
.
3
,
J
u
n
e
2
0
2
1
:
2
6
3
1
-
2639
2638
[3
]
W
o
rld
He
a
lt
h
Org
a
n
iza
ti
o
n
In
ter
n
a
ti
o
n
a
l
Ag
e
n
c
y
f
o
r
Re
se
a
r
c
h
o
n
Ca
n
c
e
r,
“
G
lo
b
a
l
Ca
n
c
e
r
Ob
se
r
v
a
to
r
y
,
”
2
0
2
0
.
[
On
li
n
e
]
.
A
v
a
il
a
b
le:
h
tt
p
:/
/g
c
o
.
i
a
r
c
.
f
r/.
[4
]
“
Bre
a
st
Ca
n
c
e
r
D
a
tas
e
t
S
u
rv
e
y
b
y
WHO,”
[
On
li
n
e
]
.
A
v
a
il
a
b
le:
w
h
o
.
in
t/
c
a
n
c
e
r/d
e
tec
ti
o
n
/
b
re
a
stc
a
n
c
e
r
/en
/i
n
d
e
x
1
.
h
tm
l
[5
]
L
.
L
a
tch
o
u
m
i,
e
t
a
l.
,
“
A
b
n
o
rm
a
li
ty
De
te
c
ti
o
n
Us
in
g
W
e
i
g
h
e
d
P
a
r
ti
c
le
S
wa
r
m
Op
ti
m
iza
ti
o
n
a
n
d
S
m
o
o
th
S
u
p
p
o
r
t
V
e
c
to
r
M
a
c
h
in
e
,
”
Bi
o
me
d
ica
l
Re
se
a
rc
h
,
v
o
l.
2
8
,
n
o
.
1
1
,
p
p
.
4
7
4
9
-
4
7
5
1
,
2
0
1
7
.
[6
]
J.
Ro
h
it
,
e
t
a
l.
,
"
S
tag
e
-
sp
e
c
i
f
ic
P
re
d
ictiv
e
M
o
d
e
ls
f
o
r
Bre
a
st
Ca
n
c
e
r
S
u
rv
iv
a
b
il
it
y
,
"
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
M
e
d
ica
l
In
f
o
rm
a
ti
c
s
,
v
o
l
.
9
7
,
p
p
.
3
0
4
-
3
1
1
,
2
0
1
7
.
[7
]
E.
A
li
č
k
o
v
ić
a
n
d
A
.
S
u
b
a
si,
"
Bre
a
st
Ca
n
c
e
r
Dia
g
n
o
sis
Us
in
g
GA
F
e
a
tu
re
S
e
lec
ti
o
n
a
n
d
Ro
tati
o
n
F
o
re
st,"
Ne
u
ra
l
Co
mp
u
t
in
g
a
n
d
Ap
p
li
c
a
t
io
n
s
,
v
o
l.
2
8
,
p
p
.
7
5
3
-
7
6
3
,
2
0
1
7
.
[8
]
K.
A
ru
tch
e
lv
a
n
a
n
d
R.
P
e
riy
a
sa
m
y
,
"
C
a
n
c
e
r
P
re
d
ictio
n
S
y
ste
m
Us
in
g
Da
ta
m
in
in
g
T
e
c
h
n
iq
u
e
s
,
"
In
ter
n
a
ti
o
n
a
l
Res
e
a
rc
h
J
o
u
rn
a
l
o
f
E
n
g
i
n
e
e
rin
g
a
n
d
T
e
c
h
n
o
l
o
g
y
(
IRJET
)
,
v
o
l.
2
,
n
o
.
8
,
p
p
.
1
1
7
9
-
1
1
8
3
,
2
0
1
5
.
[9
]
M
.
Ku
m
a
r,
S
.
S
.
T
o
m
a
r
a
n
d
B.
G
a
u
r,
"
M
in
in
g
Ba
se
d
Op
ti
m
iz
a
ti
o
n
f
o
r
Bre
a
st
C
a
n
c
e
r
A
n
a
l
y
si
s:
A
Re
v
ie
w
,
"
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
C
o
mp
u
ter
Ap
p
l
ica
ti
o
n
s
,
v
o
l.
1
1
9
,
n
o
.
1
3
,
p
p
.
1
-
6
,
2
0
1
5
.
[1
0
]
A
.
F
.
M
.
A
g
a
r
a
p
,
“
On
Bre
a
st
Ca
n
c
e
r
De
te
c
ti
o
n
:
a
n
A
p
p
li
c
a
ti
o
n
o
f
M
a
c
h
in
e
L
e
a
rn
in
g
A
lg
o
rit
h
m
s
o
n
th
e
W
isc
o
n
sin
Dia
g
n
o
stic
Da
tas
e
t,
”
ICM
L
S
C
'
1
8
:
Pro
c
e
e
d
in
g
s o
f
t
h
e
2
n
d
In
ter
n
a
t
io
n
a
l
Co
n
fer
e
n
c
e
o
n
M
a
c
h
in
e
L
e
a
rn
in
g
a
n
d
S
o
f
t
Co
mp
u
t
in
g
,
p
p
.
5
-
9
,
2
0
1
8
.
[1
1
]
Na
u
c
k
,
e
t
a
l.
,
"
Ob
tain
in
g
In
terp
r
e
tab
le
F
u
z
z
y
Clas
si
f
ica
ti
o
n
Ru
les
f
ro
m
M
e
d
ica
l
Da
t
a
,
"
Arti
fi
c
ia
l
In
telli
g
e
n
c
e
i
n
M
e
d
icin
e
,
v
o
l.
1
6
,
n
o
.
2
,
p
p
.
1
4
9
-
1
6
9
,
2
0
1
5
.
[1
2
]
P
.
G
u
p
ta
a
n
d
L
.
S
h
a
li
n
i
,
“
A
n
a
l
y
s
is
o
f
M
a
c
h
in
e
L
e
a
rn
in
g
T
e
c
h
n
iq
u
e
s
f
o
r
Bre
a
st
C
a
n
c
e
r
P
re
d
ictio
n
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
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.
7
,
no
.
5
,
p
p
.
2
3
8
9
1
-
2
3
8
9
,
2
0
1
8
.
[1
3
]
Y.
Kh
o
u
rd
if
i,
e
t
a
l.
,
"
A
p
p
ly
i
n
g
Be
st
M
a
c
h
in
e
L
e
a
rn
in
g
Alg
o
rit
h
m
s
f
o
r
Bre
a
st
Ca
n
c
e
r
P
re
d
ictio
n
a
n
d
Clas
sif
ic
a
ti
o
n
,
"
2
0
1
8
I
n
ter
n
a
t
i
o
n
a
l
C
o
n
fer
e
n
c
e
o
n
El
e
c
tro
n
ics
,
Co
n
tro
l,
Op
ti
miz
a
ti
o
n
a
n
d
Co
mp
u
ter
S
c
ien
c
e
(
ICECOCS
)
,
Ke
n
it
ra
,
2
0
1
8
,
p
p
.
1
-
5.
[1
4
]
C.
S
h
ra
v
y
a
,
K.
P
ra
v
a
li
k
a
,
a
n
d
S
.
S
u
b
h
a
n
i,
“
P
re
d
icti
o
n
o
f
Bre
a
st
Ca
n
c
e
r
Us
in
g
S
u
p
e
rv
ise
d
M
a
c
h
in
e
L
e
a
rn
in
g
T
e
c
h
n
iq
u
e
s,”
In
t
e
rn
a
ti
o
n
a
l
J
o
u
r
n
a
l
o
f
In
n
o
v
a
t
ive
T
e
c
h
n
o
lo
g
y
a
n
d
Exp
lo
rin
g
En
g
in
e
e
rin
g
(
IJ
IT
EE
)
,
v
o
l.
8
,
no
.
6
,
p
p
.
1
1
0
6
-
1
1
1
0
,
2
0
1
9
.
[1
5
]
N.
M
.
A
li
,
N.
A
.
A
.
A
z
i
z
a
n
d
R
.
Be
sa
r,
"
Co
m
p
a
riso
n
o
f
M
icro
a
r
ra
y
Bre
a
st
Ca
n
c
e
r
Clas
si
f
ic
a
ti
o
n
Us
in
g
S
u
p
p
o
r
t
V
e
c
to
r
M
a
c
h
i
n
e
a
n
d
L
o
g
isti
c
Re
g
r
e
ss
io
n
w
it
h
LA
S
S
O
a
n
d
BOR
UTA
F
e
a
tu
re
S
e
le
c
ti
o
n
,
"
In
d
o
n
e
sia
n
J
o
u
rn
a
l
o
f
El
e
c
trica
l
En
g
in
e
e
rin
g
a
n
d
Co
m
p
u
ter
S
c
ien
c
e
(
IJ
EE
CS
)
,
v
o
l.
2
0
,
n
o
.
2
,
p
p
.
7
1
2
-
7
1
9
,
2
0
2
0
.
[1
6
]
V
.
C
h
a
u
ra
sia
,
e
t
a
l.
,
“
Da
ta
M
in
i
n
g
Tec
h
n
iq
u
e
s:
t
o
P
re
d
ict
a
n
d
Re
s
o
lv
e
Bre
a
st
Ca
n
c
e
r
S
u
rv
iv
a
b
il
it
y
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
M
o
b
il
e
C
o
mp
u
t
in
g
IJ
CS
M
C
,
v
o
l.
3
,
n
o
.
1
,
p
p
.
1
0
-
2
2
,
2
0
1
7
.
[1
7
]
S
.
Yu
a
n
h
a
n
g
,
e
t
a
l.
,
“
O
n
Ex
ten
d
e
d
L
o
n
g
S
h
o
rt
-
term
M
e
m
o
r
y
a
n
d
De
p
e
n
d
e
n
t
Bi
d
irec
ti
o
n
a
l
R
e
c
u
rre
n
t
Ne
u
ra
l
Ne
tw
o
rk
,
”
N
e
u
ro
c
o
mp
u
t
in
g
,
v
o
l.
3
5
6
,
p
p
.
1
5
1
-
1
6
1
,
2
0
1
9
.
[1
8
]
F
.
M
.
J.
M
.
S
h
a
m
ra
t,
M
.
A
s
a
d
u
z
z
a
m
a
n
,
P
.
G
h
o
sh
,
D.
S
u
lt
a
n
a
n
d
Z.
T
a
sn
i
m
,
“
A
Web
Ba
se
d
A
p
p
li
c
a
ti
o
n
f
o
r
Ag
ricu
lt
u
re
:
“
S
m
a
rt
F
a
r
m
in
g
S
y
s
tem
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
E
m
e
rg
in
g
T
re
n
d
s
in
En
g
in
e
e
rin
g
R
e
se
a
rc
h
,
v
o
l.
8
,
n
o
.
6
,
p
p
.
2
3
0
9
-
2
3
2
0
,
2
0
2
0
.
[1
9
]
S
.
Bh
a
ra
ti
,
M
.
A
.
Ra
h
m
a
n
a
n
d
P
.
P
o
d
d
e
r,
"
Bre
a
st
Ca
n
c
e
r
P
re
d
icti
o
n
A
p
p
ly
in
g
Diffe
re
n
t
Clas
si
f
ica
ti
o
n
A
lg
o
rit
h
m
w
it
h
Co
m
p
a
r
a
ti
v
e
A
n
a
l
y
sis
u
sin
g
W
EK
A
,
"
2
0
1
8
4
t
h
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
El
e
c
trica
l
E
n
g
i
n
e
e
rin
g
a
n
d
In
fo
rm
a
t
io
n
&
Co
mm
u
n
ica
ti
o
n
T
e
c
h
n
o
l
o
g
y
(
iCE
Ei
CT
),
D
h
a
k
a
,
B
a
n
g
l
a
d
e
sh
,
2
0
1
8
,
p
p
.
5
8
1
-
5
8
4
.
[2
0
]
F
.
M
u
ra
to
re
,
e
t
a
l.
,
“
Co
rre
latio
n
s b
e
tw
e
e
n
Histo
p
a
th
o
lo
g
ica
l
F
i
n
d
i
n
g
s
a
n
d
Cli
n
ica
l
M
a
n
if
e
sta
ti
o
n
s
in
Bio
p
sy
-
p
ro
v
e
n
G
ian
t
Ce
ll
A
rteriti
s
,
”
J
o
u
rn
a
l
o
f
Au
to
imm
u
n
it
y
,
v
o
l.
6
9
,
p
p
.
9
4
‐
1
0
1
,
2
0
1
6
.
[2
1
]
P
.
G
h
o
sh
,
M
.
Z
.
Ha
sa
n
,
O.A
.
D
h
o
re
,
A
.
A
.
M
o
h
a
m
m
a
d
a
n
d
M
.
I.
Ja
b
i
u
ll
a
h
,
“
On
t
h
e
A
p
p
li
c
a
ti
o
n
o
f
M
a
c
h
in
e
L
e
a
rn
in
g
to
P
re
d
ictin
g
Ca
n
c
e
r
Ou
tco
m
e
,”
Pro
c
e
e
d
in
g
s
o
f
th
e
I
n
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
El
e
c
tro
n
ics
a
n
d
ICT
–
2
0
1
8
,
o
rg
a
n
ize
d
b
y
B
a
n
g
la
d
e
sh
E
lec
tro
n
ics
S
o
c
iety
(
BE
S
),
2
0
1
8
.
[2
2
]
F
.
M
.
J.
M
.
S
h
a
m
ra
t,
Z.
Tas
n
im,
P
.
G
h
o
sh
,
A
.
M
a
ju
m
d
e
r
a
n
d
M
d
.
Z
.
Ha
sa
n
,
“
P
e
rso
n
a
li
z
a
ti
o
n
o
f
Jo
b
Circu
lar
A
d
v
e
rti
s
e
m
e
n
t
to
Ca
n
d
id
a
tes
Us
i
n
g
De
c
isio
n
T
re
e
Clas
sif
ica
ti
o
n
A
l
g
o
rit
h
m
,”
2
0
2
0
I
EE
E
I
n
ter
n
a
ti
o
n
a
l
C
o
n
fer
e
n
c
e
fo
r In
n
o
v
a
ti
o
n
i
n
T
e
c
h
n
o
l
o
g
y
,
2
0
2
0
.
[2
3
]
S
.
Ra
g
h
a
v
e
n
d
ra
,
e
t
a
l.
,
“
P
e
rf
o
r
m
a
n
c
e
Ev
a
lu
a
ti
o
n
o
f
Ra
n
d
o
m
F
o
re
st w
it
h
F
e
a
tu
re
S
e
le
c
ti
o
n
M
e
th
o
d
s i
n
P
re
d
ictio
n
o
f
Dia
b
e
tes
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
r
n
a
l
o
f
El
e
c
trica
l
a
n
d
C
o
mp
u
ter
En
g
in
e
e
rin
g
(
IJ
EC
E)
,
v
o
l.
1
0
,
n
o
.
1
,
p
p
.
3
5
3
-
3
5
9
,
2
0
2
0
.
[2
4
]
M
.
S
riv
e
n
k
a
tes
h
,
“
P
re
d
icti
o
n
o
f
Bre
a
st
Ca
n
c
e
r
Dise
a
se
Us
in
g
M
a
c
h
in
e
L
e
a
rn
in
g
A
lg
o
rit
h
m
s,”
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
In
n
o
v
a
ti
v
e
T
e
c
h
n
o
lo
g
y
a
n
d
Exp
lo
ri
n
g
E
n
g
in
e
e
rin
g
(
IJ
IT
EE
),
v
o
l.
9
,
n
o
.
4
,
p
p
.
2
8
6
8
-
2
8
7
8
,
2
0
2
0
.
[2
5
]
M
.
M
.
Is
lam
,
H.
Iq
b
a
l,
M
.
R.
Ha
q
u
e
a
n
d
M
.
K.
Ha
sa
n
,
“
P
re
d
ictio
n
o
f
Bre
a
st Can
c
e
r
Us
in
g
S
u
p
p
o
rt
V
e
c
to
r
M
a
c
h
in
e
a
n
d
K
-
Ne
a
re
st
Ne
ig
h
b
o
rs,”
2
0
1
7
IEE
E
Reg
i
o
n
1
0
H
u
ma
n
it
a
ri
a
n
T
e
c
h
n
o
l
o
g
y
Co
n
fer
e
n
c
e
(
R1
0
-
HTC
)
,
Dh
a
k
a
,
2
0
1
7
,
p
p
.
2
2
6
-
2
2
9
.
[2
6
]
S
.
A
.
M
e
d
jah
e
d
,
A
.
Be
n
y
e
tt
o
u
a
n
d
T
.
A
.
S
a
a
d
i,
“
Bre
a
st
Ca
n
c
e
r
Dia
g
n
o
sis
b
y
Us
in
g
K
-
N
e
a
re
st
Ne
ig
h
b
o
r
w
it
h
Diff
e
r
e
n
t
Dista
n
c
e
s
a
n
d
Clas
sif
i
c
a
ti
o
n
Ru
les
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Co
mp
u
ter
Ap
p
li
c
a
ti
o
n
s
,
v
o
l.
6
2
,
n
o
.
1
,
p
p
.
1
-
5
,
2
0
1
3
.
[2
7
]
P
.
G
h
o
sh
,
M
.
Z.
Ha
sa
n
a
n
d
M
.
I.
Ja
b
iu
ll
a
h
,
“
A
Co
m
p
a
ra
ti
v
e
S
tu
d
y
o
f
M
a
c
h
in
e
L
e
a
rn
in
g
A
p
p
ro
a
c
h
e
s
o
n
Da
tas
e
t
to
P
re
d
ictin
g
Ca
n
c
e
r
Ou
tc
o
m
e
,
”
Ba
n
g
l
a
d
e
sh
El
e
c
tro
n
ic
S
o
c
iety
,
v
o
l
.
1
8
,
p
p
.
0
1
-
0
5
,
p
p
.
8
1
-
8
6
,
2
0
1
8
.
[2
8
]
“
Bre
a
st
Ca
n
c
e
r
W
is
c
o
n
sin
(Di
a
g
n
o
stic)
D
a
ta
S
e
t,
”
[
On
li
n
e
].
A
v
a
il
a
b
le:
h
tt
p
s:/
/arc
h
iv
e
.
ics
.
u
c
i.
e
d
u
/m
l/
d
a
tas
e
ts/Brea
st
+
Ca
n
c
e
r
+
W
i
sc
o
n
sin
+
(Dia
g
n
o
stic).
[2
9
]
“
S
tan
d
a
rd
S
c
a
ler
T
e
c
h
n
iq
u
e
,
”
[o
n
l
in
e
]
.
A
v
a
il
a
b
le:
h
tt
p
s://
sc
ik
it
-
lea
rn
.
o
rg
/stab
le/m
o
d
u
les
/g
e
n
e
ra
ted
/sk
lea
rn
.
p
re
p
r
o
c
e
ss
in
g
.
S
tan
d
a
r
d
S
c
a
ler.h
tm
l
.
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:
2
0
8
8
-
8708
E
xp
ert ca
n
ce
r
mo
d
el
u
s
in
g
s
u
p
ervis
ed
a
lg
o
r
ith
ms w
ith
a
LAS
S
O
s
elec
tio
n
a
p
p
r
o
a
ch
(
P
r
o
n
a
b
Gh
o
s
h
)
2639
[3
0
]
“
En
c
o
d
i
n
g
Ca
teg
o
rica
l
Da
ta
to
P
r
e
p
a
re
D
a
tas
e
t,
”
2
0
2
0
.
[
On
l
in
e
]
.
A
v
a
il
a
b
le:
h
tt
p
s:/
/t
o
w
a
rd
sd
a
tas
c
ien
c
e
.
c
o
m
/s
m
a
rter
-
w
a
y
s
-
to
-
e
n
c
o
d
e
-
c
a
teg
o
rica
ld
a
ta
-
f
o
r
-
m
a
c
h
in
e
-
lea
rn
in
g
-
p
a
rt
-
1
-
of
-
3
-
6
d
c
a
2
f
7
1
b
1
5
9
.
[3
1
]
Y.
S
u
,
e
t
a
l.
,
"
Eff
i
c
ien
t
Tex
t
Cl
a
ss
if
i
c
a
ti
o
n
Us
in
g
T
r
e
e
-
stru
c
tu
re
d
M
u
lt
i
-
li
n
e
a
r
P
rin
c
i
p
a
l
Co
m
p
o
n
e
n
t
A
n
a
l
y
sis,"
2
0
1
8
2
4
t
h
I
n
ter
n
a
t
io
n
a
l
Co
n
fer
e
n
c
e
o
n
P
a
tt
e
rn
Rec
o
g
n
i
ti
o
n
(
ICPR
),
Be
ij
in
g
,
2
0
1
8
,
p
p
.
5
8
5
-
5
9
0
.
[3
2
]
C.
Zh
o
u
,
a
n
d
A
.
W
ie
se
r,
“
J
a
c
c
a
r
d
A
n
a
l
y
sis
a
n
d
LA
S
S
O
-
Ba
s
e
d
F
e
a
tu
re
S
e
lec
ti
o
n
f
o
r
L
o
c
a
ti
o
n
F
in
g
e
rp
rin
ti
n
g
w
it
h
L
i
m
it
e
d
Co
m
p
u
tatio
n
a
l
C
o
m
p
lex
it
y
,
”
LB
S
2
0
1
8
:
1
4
t
h
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
L
o
c
a
ti
o
n
B
a
se
d
S
e
rv
ice
s
,
p
p
.
7
1
-
8
7
,
2
0
1
8
.
[3
3
]
H.
Dh
a
h
ri,
e
t
a
l.
,
“
A
u
to
m
a
ted
Bre
a
st
Ca
n
c
e
r
Dia
g
n
o
sis
Ba
se
d
o
n
M
a
c
h
in
e
L
e
a
rn
in
g
A
l
g
o
rit
h
m
s
,
”
J
o
u
rn
a
l
o
f
h
e
a
lt
h
c
a
re
e
n
g
i
n
e
e
rin
g
,
v
o
l.
2
0
1
9
,
2
0
1
9
.
[3
4
]
A
.
U.
Ha
q
,
e
t
a
l.
,
“
A
H
y
b
ri
d
In
telli
g
e
n
t
S
y
ste
m
F
ra
m
e
w
o
rk
f
o
r
t
h
e
P
re
d
ict
io
n
o
f
He
a
rt
Dise
a
se
Us
in
g
M
a
c
h
in
e
L
e
a
rn
in
g
A
lg
o
rit
h
m
s,”
M
o
b
il
e
In
f
o
rm
a
ti
o
n
S
y
ste
ms
,
v
o
l.
2
0
1
8
,
2
0
1
8
.
[3
5
]
G
.
S
a
ra
n
y
a
,
e
t
a
l.
,
“
A
Co
m
p
re
h
e
n
siv
e
S
tu
d
y
o
n
Dise
a
se
Ris
k
P
re
d
ictio
n
s
i
n
M
a
c
h
i
n
e
L
e
a
rn
in
g
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
trica
l
a
n
d
C
o
mp
u
t
e
r E
n
g
i
n
e
e
rin
g
(
IJ
ECE
)
,
v
o
l.
1
0
,
n
o
.
4
,
p
p
.
4
2
1
7
-
4
2
2
5
,
2
0
2
0
.
[3
6
]
F
.
M
.
J.
M
.
S
h
a
m
ra
t,
P
.
G
h
o
sh
,
M
.
H.
S
a
d
e
k
,
A
.
Ka
z
i,
a
n
d
S
.
S
h
u
lt
a
n
a
,
“
Im
p
le
m
e
n
tatio
n
o
f
M
a
c
h
in
e
L
e
a
rn
in
g
A
l
g
o
rit
h
m
s
to
De
t
e
c
t
th
e
P
ro
g
n
o
sis
Ra
te
o
f
Kid
n
e
y
Dise
a
se
,
”
2
0
2
0
IEE
E
In
ter
n
a
t
io
n
a
l
Co
n
fer
e
n
c
e
fo
r
In
n
o
v
a
t
io
n
in
T
e
c
h
n
o
l
o
g
y
,
2
0
2
0
.
B
I
O
G
RAP
H
I
E
S
O
F
AUTH
O
RS
Pro
n
a
b
G
h
o
s
h
a
c
q
u
ired
h
is
Ba
c
h
e
lo
r
o
f
S
c
ien
c
e
d
e
g
re
e
in
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
i
n
e
e
rin
g
(CS
E)
De
p
a
rtm
e
n
t
f
ro
m
Da
ff
o
d
il
In
tern
a
ti
o
n
a
l
U
n
iv
e
rsity
in
th
e
y
e
a
r
o
f
2
0
1
9
.
He
h
a
s
b
e
e
n
in
v
o
lv
e
d
e
x
ten
siv
e
l
y
in
c
o
ll
a
b
o
r
a
ti
v
e
re
se
a
rc
h
a
c
ti
v
it
ies
,
p
a
rti
c
u
l
a
rl
y
in
th
e
f
ield
s
o
f
m
a
c
h
in
e
lea
rn
in
g
,
c
lo
u
d
c
o
m
p
u
ti
n
g
,
a
n
d
Io
T
.
In
a
d
d
it
io
n
,
h
e
h
a
s
w
rit
ten
a
n
d
p
u
b
li
sh
e
d
d
if
fe
re
n
t
a
rti
c
les
in
c
o
n
f
e
re
n
c
e
s a
n
d
in
tern
a
ti
o
n
a
l
j
o
u
r
n
a
ls.
As
if
K
a
r
i
m
is
a
P
h
.
D.
re
se
a
rc
h
e
r
a
t
Ch
a
rles
D
a
r
w
in
Un
iv
e
rsit
y
a
n
d
li
v
e
s
in
Da
rw
in
'
s
p
o
rt
c
it
y
.
His
re
se
a
r
c
h
a
re
a
in
c
lu
d
e
s
Cry
p
to
g
ra
p
h
ic
Co
m
m
u
n
ica
ti
o
n
a
n
d
A
rti
f
icia
l
In
telli
g
e
n
c
e
.
H
e
is
p
re
se
n
tl
y
a
i
m
in
g
to
b
u
il
d
a
r
o
b
u
st
a
n
d
so
p
h
isti
c
a
ted
e
m
a
il
f
il
terin
g
f
ra
m
e
w
o
r
k
u
sin
g
M
a
c
h
i
n
e
L
e
a
rn
in
g
m
o
d
e
ls.
A
si
f
h
a
s
a
b
ro
a
d
b
a
c
k
g
ro
u
n
d
in
t
h
e
IT
in
d
u
stry
,
e
sp
e
c
iall
y
i
n
so
f
t
w
a
r
e
e
n
g
in
e
e
rin
g
.
S
y
e
d
a
Ta
n
jil
a
Ati
k
h
a
s
c
o
m
p
le
ted
h
e
r
b
o
t
h
Ba
c
h
e
lo
r
a
n
d
M
a
st
e
rs
in
IT
,
b
o
t
h
f
ro
m
th
e
IIT
,
Ja
h
a
n
g
irn
a
g
a
r
Un
iv
e
rsit
y
in
2
0
1
6
a
n
d
2
0
1
8
,
re
sp
e
c
ti
v
e
ly
.
S
h
e
h
a
s
b
e
e
n
re
c
e
n
tl
y
s
e
rv
in
g
a
s
a
lec
tu
re
r
a
t
Da
ff
o
d
il
In
tern
a
ti
o
n
a
l
Un
iv
e
rsit
y
sin
c
e
A
p
ril
2
0
1
7
.
S
h
e
w
o
rk
s
o
n
e
d
g
e
c
o
m
p
u
ti
n
g
,
d
a
ta m
in
in
g
,
Io
T
,
a
n
d
m
a
c
h
in
e
le
a
rn
in
g
.
S
a
i
m
a
Af
r
in
c
o
m
p
lete
d
h
e
r
Ba
c
h
e
lo
r
o
f
S
c
ien
c
e
d
e
g
re
e
in
th
e
f
ield
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
in
e
e
rin
g
f
ro
m
th
e
r
e
n
o
w
n
e
d
Da
ff
o
d
il
In
tern
a
ti
o
n
a
l
Un
iv
e
rsity
in
th
e
y
e
a
r
o
f
2
0
1
9
.
A
t
p
re
se
n
t
sh
e
is
a
lec
tu
re
r
o
f
Da
ff
o
d
il
In
t
e
rn
a
ti
o
n
a
l
U
n
iv
e
rsity
in
CS
E
d
e
p
a
rtm
e
n
t.
He
r
re
se
a
rc
h
a
re
a
c
o
n
tain
s
M
a
c
h
in
e
L
e
a
rn
in
g
a
n
d
I
o
T
.
M
o
h
d
.
S
a
ifu
z
z
a
m
a
n
h
a
s
o
b
tain
e
d
h
is
M
.
S
c
.
i
n
CS
f
ro
m
Ja
h
a
n
g
ir
n
a
g
a
r
Un
iv
e
rsit
y
a
n
d
B.
S
c
.
in
CS
E
f
ro
m
Da
ff
o
d
il
In
tern
a
ti
o
n
a
l
Un
iv
e
rsit
y
.
He
c
u
rre
n
tl
y
w
o
rk
s
a
s
a
l
e
c
tu
re
r
in
th
e
CS
E
De
p
a
rtme
n
t
a
t
Da
ff
o
d
il
In
te
rn
a
ti
o
n
a
l
Un
iv
e
rsity
in
Ba
n
g
lad
e
sh
sin
c
e
S
e
p
tem
b
e
r
2
0
1
7
.
He
p
re
f
e
r
s
Io
T
w
o
rk
a
n
d
Da
ta
M
in
in
g
w
o
rk
.
He
is
a
lso
a
n
a
c
ti
v
e
m
e
m
b
e
r
o
f
DIU
-
N
L
P
a
n
d
M
a
c
h
in
e
L
e
a
rn
in
g
Re
se
a
rc
h
L
a
b
Evaluation Warning : The document was created with Spire.PDF for Python.