I
nte
rna
t
io
na
l
J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
p
ute
r
E
ng
in
ee
ring
(
I
J
E
CE
)
Vo
l.
10
,
No
.
1
,
Feb
r
u
ar
y
2020
,
p
p
.
353
~
359
I
SS
N:
2088
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v
10
i
1
.
pp
3
5
3
-
359
353
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m/in
d
ex
.
p
h
p
/I
JE
C
E
Perf
o
r
m
a
nce
e
v
a
lua
tion
o
f
ra
ndo
m f
o
rest
w
ith
f
ea
tu
re
s
elec
tion
m
e
thods
in
p
redic
tion o
f
d
ia
be
tes
Ra
g
ha
v
endra
S
1
, S
a
nto
s
h K
u
m
a
r
J
2
1
De
p
a
rt
m
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
,
CHRIST
De
e
m
e
d
T
o
Be
Un
iv
e
rsit
y
,
In
d
ia
2
De
p
a
rt
m
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
,
KSS
E
M
,
I
n
d
i
a
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
an
10
,
2
0
1
9
R
ev
i
s
ed
A
u
g
7
,
20
19
A
cc
ep
ted
A
u
g
29
,
2
0
19
Da
ta
m
in
in
g
is
n
o
th
i
n
g
b
u
t
t
h
e
p
r
o
c
e
ss
o
f
v
ie
w
in
g
d
a
ta
in
d
if
f
e
re
n
t
a
n
g
le
a
n
d
c
o
m
p
il
in
g
it
in
t
o
a
p
p
r
o
p
riate
i
n
f
o
rm
a
ti
o
n
.
Re
c
e
n
t
im
p
ro
v
e
m
e
n
ts
in
th
e
a
re
a
o
f
d
a
ta
m
in
in
g
a
n
d
m
a
c
h
in
e
lea
rn
in
g
h
a
v
e
e
m
p
o
w
e
re
d
th
e
r
e
se
a
rc
h
in
b
io
m
e
d
ica
l
f
ield
to
i
m
p
ro
v
e
th
e
c
o
n
d
it
i
o
n
o
f
g
e
n
e
ra
l
h
e
a
lt
h
c
a
re
.
S
in
c
e
th
e
w
ro
n
g
c
las
sif
i
c
a
ti
o
n
m
a
y
le
a
d
to
p
o
o
r
p
re
d
ictio
n
,
th
e
re
is
a
n
e
e
d
to
p
e
rf
o
r
m
th
e
b
e
tt
e
r
c
las
sif
ic
a
ti
o
n
w
h
ich
f
u
rth
e
r
im
p
ro
v
e
s
th
e
p
re
d
i
c
ti
o
n
ra
te
o
f
th
e
m
e
d
ica
l
d
a
tas
e
ts.
W
h
e
n
m
e
d
ica
l
d
a
ta
m
in
in
g
is
a
p
p
li
e
d
o
n
t
h
e
m
e
d
ica
l
d
a
tas
e
ts
th
e
im
p
o
rtan
t
a
n
d
d
if
f
i
c
u
lt
c
h
a
ll
e
n
g
e
s
a
re
th
e
c
las
sif
ic
a
ti
o
n
a
n
d
p
re
d
ictio
n
.
In
th
is
p
ro
p
o
se
d
w
o
rk
w
e
e
v
a
lu
a
te
th
e
P
IM
A
In
d
ian
Dia
b
tes
d
a
ta
se
t
o
f
UCI
r
e
p
o
sito
ry
u
sin
g
m
a
c
h
in
e
lea
rn
in
g
a
lg
o
rit
h
m
li
k
e
Ra
n
d
o
m
F
o
re
st
a
lo
n
g
w
it
h
f
e
a
tu
re
se
l
e
c
t
io
n
m
e
th
o
d
s
su
c
h
a
s
f
o
rw
a
rd
se
lec
ti
o
n
a
n
d
b
a
c
k
w
a
rd
e
li
m
in
a
ti
o
n
b
a
se
d
o
n
e
n
tro
p
y
e
v
a
lu
a
ti
o
n
m
e
th
o
d
u
si
n
g
p
e
rc
e
n
tag
e
s
p
li
t
a
s tes
t
o
p
ti
o
n
.
T
h
e
e
x
p
e
ri
m
e
n
t
wa
s
c
o
n
d
u
c
te
d
u
sin
g
R
stu
d
i
o
p
latf
o
r
m
a
n
d
we
a
c
h
iev
e
d
c
las
si
f
ica
ti
o
n
a
c
c
u
ra
c
y
o
f
8
4
.
1
%
.
F
ro
m
re
su
lt
s
w
e
c
a
n
sa
y
th
a
t
Ra
n
d
o
m
F
o
re
st
p
re
d
icts
d
iab
e
tes
b
e
tt
e
r
th
a
n
o
th
e
r
tec
h
n
i
q
u
e
s
w
it
h
les
s
n
u
m
b
e
r
o
f
a
tt
rib
u
tes
so
th
a
t
o
n
e
c
a
n
a
v
o
id
lea
st
i
m
p
o
rtan
t
tes
t
f
o
r
id
e
n
ti
f
y
in
g
d
iab
e
tes
.
K
ey
w
o
r
d
s
:
C
las
s
i
f
icatio
n
ac
c
u
r
ac
y
Data
m
i
n
i
n
g
Featu
r
e
s
elec
t
io
n
m
eth
o
d
P
er
ce
n
tag
e
s
p
lit
R
an
d
o
m
f
o
r
est
Co
p
y
rig
h
t
©
2
0
2
0
In
stit
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
Al
l
rig
h
ts
re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
R
ag
h
a
v
en
d
r
a
S
,
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
,
C
HR
I
ST
Dee
m
ed
T
o
B
e
Un
iv
er
s
it
y
,
Kan
m
an
i
k
e,
K
u
m
b
al
g
o
d
u
,
M
y
s
o
r
e
R
o
ad
,
B
an
g
alo
r
e
-
5
6
0
0
7
4
.
C
o
n
tact:
+9
1
9
7
4
0
8
5
7
5
0
1
E
m
ail:
r
ag
h
a
v
.
tr
g
@
g
m
a
il.c
o
m
1.
I
NT
RO
D
UCT
I
O
N
R
ec
en
t
i
m
p
r
o
v
e
m
en
t
s
in
t
h
e
ar
ea
o
f
d
ata
m
i
n
in
g
an
d
m
ac
h
i
n
e
lear
n
in
g
h
a
v
e
e
m
p
o
w
er
ed
th
e
r
esear
ch
in
b
io
m
ed
ical
f
ield
to
i
m
p
r
o
v
e
th
e
co
n
d
it
io
n
o
f
g
e
n
er
al
h
ea
lth
ca
r
e.
I
n
m
a
n
y
p
ar
ts
o
f
t
h
e
w
o
r
ld
th
e
ten
d
e
n
c
y
f
o
r
m
ai
n
tain
in
g
lo
n
g
-
last
i
n
g
r
ec
o
r
d
s
co
n
s
is
ti
n
g
o
f
m
ed
ical
d
ata
is
b
ec
o
m
i
n
g
a
n
ac
ce
p
ted
p
r
ac
tice.
I
n
ad
d
itio
n
to
th
is
,
t
h
e
n
e
w
er
m
ed
ical
eq
u
ip
m
e
n
t
’
s
a
n
d
th
e
tec
h
n
iq
u
es
u
s
ed
in
d
ia
g
n
o
s
is
,
p
r
o
d
u
ce
s
co
m
p
o
s
i
te
an
d
h
u
g
e
d
ata.
T
h
er
ef
o
r
e,
to
h
an
d
le
th
ese
ill
-
s
tr
u
ct
u
r
ed
b
io
m
ed
ical
d
ata,
in
tell
ig
e
n
t
a
lg
o
r
it
h
m
s
f
o
r
d
ata
m
i
n
i
n
g
an
d
m
ac
h
in
e
lear
n
i
n
g
ar
e
r
eq
u
ir
ed
in
o
r
d
er
to
tak
e
lo
g
ical
r
ea
s
o
n
i
n
g
f
r
o
m
t
h
e
s
av
ed
r
a
w
d
ata,
wh
ich
is
co
n
s
id
er
ed
as
m
ed
ical
d
ata
m
i
n
i
n
g
.
W
ith
i
n
th
e
m
ed
ical
d
ata,
th
e
m
ed
ical
d
ata
m
i
n
i
n
g
s
ea
r
ch
e
s
f
o
r
p
atter
n
s
a
n
d
r
elatio
n
s
h
ip
s
w
h
ic
h
ca
n
p
r
o
v
id
e
u
s
ef
u
l
in
f
o
r
m
a
tio
n
f
o
r
ap
p
r
o
p
r
iate
m
ed
ical
d
iag
n
o
s
is
[
1
]
.
Data
m
i
n
in
g
tech
n
iq
u
es
ar
e
ap
p
lied
to
d
i
f
f
er
en
t
m
ed
ical
d
o
m
ai
n
s
(
h
ea
lt
h
ca
r
e
d
atab
ases
o
r
m
ed
ical
d
atasets
)
to
i
m
p
r
o
v
e
th
e
m
ed
ical
d
ia
g
n
o
s
is
.
T
o
ch
ec
k
f
o
r
an
y
in
v
is
ib
le
p
atter
n
s
i
n
s
id
e
th
e
m
ed
ical
d
atasets
,
m
ed
ical
d
ata
m
in
i
n
g
i
s
s
tr
o
n
g
l
y
r
ec
o
m
m
e
n
d
ed
.
I
n
m
ed
ical
d
ata
m
i
n
i
n
g
,
t
h
e
ac
t
u
al
tas
k
s
(
c
h
allen
g
es)
ar
e
t
h
e
clas
s
i
f
icatio
n
a
n
d
p
r
ed
ictio
n
o
f
m
ed
ical
d
atase
ts
.
On
e
o
f
th
e
t
ec
h
n
iq
u
es
t
h
at
i
s
u
s
ed
f
o
r
t
h
e
class
i
f
icatio
n
a
n
d
p
r
ed
ictio
n
i
s
r
an
d
o
m
f
o
r
est
[
2
]
.
R
an
d
o
m
f
o
r
est
is
a
n
e
n
s
e
m
b
le
lear
n
i
n
g
m
et
h
o
d
f
o
r
r
eg
r
e
s
s
io
n
,
clas
s
i
f
icatio
n
,
a
n
d
o
th
er
j
o
b
s
th
a
t
f
u
n
ctio
n
s
b
y
m
ak
in
g
an
a
s
s
e
m
b
l
y
o
f
d
ec
is
io
n
tr
ee
s
at
tr
ain
i
n
g
ti
m
e
an
d
g
en
er
ati
n
g
th
e
cla
s
s
t
h
at
is
t
h
e
class
if
icatio
n
o
r
r
eg
r
ess
io
n
o
f
th
e
d
i
s
tin
c
t tr
ee
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.
10
,
No
.
1
,
Feb
r
u
ar
y
2
0
2
0
:
353
-
359
354
R
an
d
o
m
f
o
r
est i
s
a
v
er
s
at
ile
al
g
o
r
ith
m
u
s
ed
f
o
r
clas
s
i
f
icatio
n
s
u
ited
f
o
r
th
e
an
a
l
y
s
is
o
f
lar
g
e
d
atasets
.
R
an
d
o
m
f
o
r
est
is
p
o
p
u
lar
b
e
ca
u
s
e
r
a
n
d
o
m
f
o
r
est
cla
s
s
i
f
ic
atio
n
m
o
d
el
s
h
a
v
e
h
i
g
h
p
r
ed
ictio
n
ac
cu
r
ac
y
a
n
d
p
r
o
v
id
es
in
f
o
r
m
atio
n
o
n
i
m
p
o
r
tan
t
v
ar
iab
les
f
o
r
class
if
ica
tio
n
.
R
a
n
d
o
m
f
o
r
est
p
r
o
v
id
es
t
wo
i
m
p
o
r
tan
t
a
s
p
ec
ts
f
o
r
d
ata
m
i
n
in
g
i.e
.
h
ig
h
p
r
ed
ict
io
n
ac
cu
r
ac
y
a
n
d
in
f
o
r
m
atio
n
r
elate
d
to
i
m
p
o
r
ta
n
ce
o
f
attr
ib
u
te
i
n
class
i
f
icatio
n
[
3
]
.
On
e
o
f
t
h
e
s
i
g
n
if
ican
t
s
tep
s
in
a
n
y
d
ata
m
i
n
i
n
g
r
esear
c
h
b
ased
o
n
clas
s
if
icatio
n
an
d
p
r
ed
ictio
n
m
o
d
el
s
is
th
e
f
ea
tu
r
e
s
elec
ti
o
n
.
Featu
r
e
s
elec
tio
n
is
v
er
y
m
u
ch
i
m
p
o
r
tan
t
b
ec
au
s
e
w
h
e
n
w
e
co
n
s
tr
u
c
t
a
m
ed
ical
d
ata
m
i
n
i
n
g
m
o
d
el,
th
e
m
ed
ical
d
ataset
m
a
y
g
en
er
all
y
co
n
s
is
ts
o
f
f
u
r
t
h
e
r
in
f
o
r
m
atio
n
th
a
n
th
e
ac
tu
al
i
n
f
o
r
m
atio
n
n
ee
d
e
d
to
c
o
n
s
tr
u
ct
th
e
m
o
d
el.
I
f
w
e
p
r
eser
v
e
th
e
attr
ib
u
te
c
o
lu
m
n
s
t
h
at
ar
e
n
o
t
ac
tu
all
y
n
ee
d
ed
,
th
en
i
t
lead
s
to
w
asta
g
e
o
f
m
e
m
o
r
y
a
n
d
m
o
r
e
C
P
U
tim
e
is
n
ee
d
ed
f
o
r
th
e
tr
ain
i
n
g
p
r
o
ce
s
s
an
d
th
e
q
u
alit
y
o
f
t
h
e
ex
p
l
o
r
ed
p
atter
n
m
a
y
b
e
d
eter
io
r
ated
b
y
t
h
e
s
e
ad
d
itio
n
al
att
r
ib
u
tes
b
ec
au
s
e
o
f
th
e
f
o
llo
w
i
n
g
r
ea
s
o
n
:
a.
I
t
is
d
if
f
ic
u
lt
to
d
is
co
v
er
m
ea
n
in
g
f
u
l
p
atter
n
f
r
o
m
d
ata
b
ec
au
s
e
s
o
m
e
attr
ib
u
tes
m
a
y
b
e
r
e
d
u
n
d
an
t
,
an
d
n
o
is
y
.
b.
Fo
r
id
en
tify
i
n
g
ex
ce
lle
n
t p
atter
n
,
th
e
m
aj
o
r
it
y
o
f
d
ata
m
i
n
i
n
g
alg
o
r
ith
m
s
n
ee
d
lar
g
er
tr
ai
n
i
n
g
d
ataset
b
u
tth
e
d
ata
u
s
ed
f
o
r
tr
ain
i
n
g
i
s
ex
tr
em
el
y
s
m
all
i
n
f
e
w
d
ata
m
i
n
in
g
ap
p
licatio
n
s
.
Featu
r
e
s
elec
tio
n
as
s
is
t
in
s
o
l
v
in
g
t
h
ese
p
r
o
b
le
m
s
b
y
h
a
v
i
n
g
to
o
litt
le
d
ata
o
f
h
ig
h
v
al
u
e
r
ath
er
th
a
n
to
o
m
u
c
h
d
ata
o
f
litt
le
v
al
u
e.
Featu
r
e
s
elec
t
io
n
h
as
ad
v
a
n
tag
e
s
i
n
th
e
c
lass
if
icatio
n
o
f
d
ata
an
d
t
h
er
e
is
r
ed
u
ctio
n
in
co
m
p
u
tatio
n
al
co
m
p
lex
i
t
y
d
u
e
to
r
e
d
u
ctio
n
in
d
i
m
e
n
s
io
n
[
4
]
.
I
n
th
e
p
r
o
p
o
s
ed
r
esear
ch
w
o
r
k
w
e
ap
p
l
y
d
ata
m
i
n
i
n
g
tec
h
n
iq
u
e
li
k
e
r
a
n
d
o
m
f
o
r
est
with
f
ea
t
u
r
e
s
elec
tio
n
m
eth
o
d
s
o
n
d
iab
etes
d
ataset
an
d
h
av
e
id
en
ti
f
ied
th
e
k
e
y
attr
ib
u
tes
f
o
r
in
cr
ea
s
i
n
g
th
e
clas
s
if
icatio
n
ac
cu
r
ac
y
.
2.
L
I
T
E
R
AT
U
RE
R
E
VI
E
W
Fro
m
th
e
e
x
is
ti
n
g
liter
at
u
r
e
w
e
f
o
u
n
d
m
an
y
d
if
f
er
e
n
t
m
et
h
o
d
s
ar
e
ap
p
lied
o
n
P
I
MA
I
n
d
ian
d
iab
etes
d
ataset.
T
h
e
m
et
h
o
d
s
p
r
o
p
o
s
ed
b
y
d
i
f
f
er
e
n
t
r
esear
c
h
er
s
an
d
t
h
e
cla
s
s
i
f
icatio
n
ac
c
u
r
ac
y
ac
h
ie
v
ed
ar
e
ex
p
lain
ed
b
elo
w
:
T
o
ex
h
ib
it
th
e
ef
f
icie
n
c
y
o
f
th
e
h
y
b
r
id
class
if
ier
b
ased
o
n
e
v
o
lu
t
io
n
ar
y
co
m
p
u
tatio
n
o
n
d
iab
etes
d
ataset,
a
m
eth
o
d
b
ased
o
n
h
y
b
r
id
cla
s
s
i
f
ier
alo
n
g
w
it
h
k
-
n
ea
r
est
n
ei
g
h
b
o
r
w
as
p
r
o
p
o
s
ed
.
B
ased
o
n
th
e
cla
s
s
i
f
icat
io
n
ac
c
u
r
ac
y
,
it
w
as
clea
r
t
h
at
o
n
o
v
er
5
0
r
u
n
s
th
e
h
y
b
r
id
clas
s
i
f
ier
ac
h
ie
v
ed
g
o
o
d
ac
c
u
r
ac
y
of
8
0
% [
5
]
.
I
n
s
tead
o
f
u
s
i
n
g
a
tr
ad
itio
n
al
n
eu
r
o
n
w
h
ic
h
p
r
o
d
u
ce
s
o
u
t
p
u
t
f
o
r
a
g
i
v
e
n
in
p
u
t
i
n
ea
ch
iter
atio
n
,
a
s
p
ik
i
n
g
n
eu
r
o
n
w
h
ic
h
g
et
s
a
ctiv
ated
af
ter
ea
c
h
T
m
s
w
i
th
an
in
p
u
t
is
d
esi
g
n
ed
.
T
h
e
o
u
t
p
u
t
ca
n
b
e
ch
an
g
ed
in
to
a
p
ar
ticu
lar
f
ir
in
g
r
ate
f
u
r
th
er
m
o
r
e
it
ca
n
p
er
f
o
r
m
th
e
d
ata
class
i
f
icat
io
n
d
ep
en
d
i
n
g
o
n
a
f
ir
i
n
g
r
ate
cr
ea
ted
f
r
o
m
in
p
u
t
s
i
g
n
a
l.
Fo
r
a
s
et
o
f
ca
s
es
b
elo
n
g
in
g
to
o
n
e
a
m
o
n
g
k
clas
s
es,
ev
er
y
i
n
p
u
t
is
co
n
n
ec
ted
to
in
p
u
t
cu
r
r
e
n
t
a
n
d
t
h
e
s
p
ik
i
n
g
n
eu
r
o
n
g
ets
ex
ci
ted
af
ter
T
m
s
,
at
l
ast
th
e
f
ir
in
g
a
m
o
u
n
t
i
s
ca
lcu
lated
f
o
r
ea
c
h
ca
s
e.
W
eig
h
t
s
f
o
r
th
e
s
p
i
k
in
g
n
eu
r
o
n
ar
e
o
p
ti
m
ized
u
s
in
g
a
g
r
av
ita
tio
n
al
s
ea
r
c
h
alg
o
r
it
h
m
.
T
h
e
ca
p
ab
ilit
y
o
f
th
e
p
r
o
j
ec
ted
m
et
h
o
d
is
co
m
p
ar
ed
w
it
h
t
h
e
id
e
n
tical
s
p
ik
in
g
n
eu
r
o
n
i
m
p
le
m
e
n
ted
w
it
h
p
ar
ticle
s
w
ar
m
opt
im
izatio
n
(
P
SO)
,
cu
ck
o
o
s
ea
r
ch
alg
o
r
it
h
m
a
n
d
d
if
f
er
e
n
tial
ev
o
l
u
tio
n
s
.
T
h
e
m
o
d
el
is
i
m
p
le
m
en
ted
o
n
d
iab
etes d
ataset
an
d
th
e
g
r
av
it
atio
n
al
s
ea
r
ch
al
g
o
r
ith
m
ac
h
ie
v
ed
g
o
o
d
ac
cu
r
ac
y
o
f
7
6
.
6
1
% [
6
]
.
Fo
r
o
p
ti
m
izi
n
g
t
h
e
p
ar
a
m
e
ter
f
o
r
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
i
n
e
(
S
VM
)
,
a
n
ad
j
u
s
ted
b
at
al
g
o
r
ith
m
(
A
B
A
)
is
p
r
o
p
o
s
ed
.
T
h
e
ex
p
er
i
m
e
n
ts
ar
e
co
n
d
u
cted
o
n
t
h
e
d
iab
ete
s
d
ataset.
T
h
e
ex
p
er
i
m
en
tal
r
e
s
u
lt
w
as
co
m
p
ar
ed
w
it
h
th
e
Gr
id
-
SVM
an
d
o
t
h
er
ap
p
r
o
ac
h
es.
B
ased
o
n
th
e
r
es
u
lt,
A
B
A
-
S
VM
is
co
n
s
id
er
ed
as
a
b
etter
clas
s
i
f
ier
th
an
Gr
id
-
SVM
an
d
co
m
p
a
r
ed
to
o
th
er
ap
p
r
o
ac
h
es
lik
e
P
SO
-
SVM,
th
e
A
B
A
-
SV
M
ac
h
iev
ed
b
etter
class
i
f
icatio
n
ac
cu
r
ac
y
o
f
7
7
.
3
4
% [
7
]
.
A
m
o
d
el
is
p
r
o
p
o
s
ed
to
h
an
d
l
e
th
e
p
r
o
b
le
m
s
th
at
ca
n
ap
p
ea
r
w
h
e
n
lear
n
i
n
g
f
r
o
m
v
er
y
s
m
a
ll d
ata
t
h
a
t
ar
e
alr
ea
d
y
clas
s
i
f
ied
.
T
h
e
m
o
d
el
d
ep
en
d
s
o
n
L
o
g
ical
An
al
y
s
i
s
o
f
Data
(
L
A
D)
an
d
is
p
r
o
v
id
ed
w
it
h
ad
d
itio
n
al
in
f
o
r
m
atio
n
o
b
tain
ed
f
r
o
m
t
h
e
co
n
s
id
er
atio
n
o
f
d
ata
s
tatis
ticall
y
.
So
th
e
n
e
w
p
r
o
p
o
s
ed
m
o
d
el
is
ca
lled
SLA
D.
T
h
e
p
er
f
o
r
m
a
n
ce
o
f
SLA
D
is
co
m
p
ar
ed
w
it
h
L
AD,
SVM
an
d
lab
el
p
r
o
p
ag
atio
n
al
g
o
r
ith
m
.
T
h
e
ex
p
er
im
e
n
t
w
as
co
n
d
u
cte
d
o
n
d
iab
etes
d
ataset.
Fro
m
th
e
r
esu
lts
o
b
tain
ed
,
it
w
as
f
o
u
n
d
th
at
f
o
r
b
o
th
5
%
tr
ain
i
n
g
a
n
d
1
0
% tr
ain
in
g
S
L
A
D
ac
h
ie
v
ed
b
etter
ac
cu
r
ac
y
o
f
7
2
.
8
7
% c
o
m
p
ar
ed
to
th
e
o
th
er
m
et
h
o
d
s
[
8
]
.
A
m
e
th
o
d
to
u
s
e
s
eq
u
en
t
ial
v
ar
iatio
n
al
i
n
f
er
en
ce
an
d
k
al
m
an
f
ilter
i
n
g
o
n
d
iab
etes
d
atase
t
to
p
r
ed
ict
th
e
clas
s
i
f
icatio
n
ac
c
u
r
ac
y
is
p
r
o
p
o
s
ed
[
9
]
.
Fro
m
t
h
e
o
u
t
p
u
t
o
f
t
h
e
m
e
th
o
d
,
it
w
as
cl
ea
r
th
at
s
eq
u
e
n
tia
l
v
ar
iatio
n
al
i
n
f
er
en
ce
ac
h
iev
ed
b
etter
ac
cu
r
ac
y
o
f
8
0
% c
o
m
p
ar
ed
to
7
6
% a
ch
iev
ed
b
y
k
al
m
an
f
i
lter
in
g
.
B
y
co
m
b
i
n
i
n
g
t
h
e
ad
v
a
n
ta
g
e
s
o
f
g
r
ap
h
an
d
co
m
b
i
n
ato
r
ial
m
eth
o
d
,
a
clu
s
ter
i
n
g
en
s
e
m
b
le
m
et
h
o
d
w
a
s
d
ev
elo
p
ed
u
s
in
g
De
m
p
s
ter
-
S
h
ef
er
ev
id
en
ce
th
eo
r
e
m
[
1
0
]
.
T
h
e
m
o
d
el
w
as
i
m
p
le
m
en
ted
o
n
d
iab
etes
d
ataset
an
d
f
r
o
m
t
h
e
e
x
p
er
im
en
tal
r
es
u
lt
s
it
w
as
id
en
t
if
ied
th
at
th
e
p
r
o
p
o
s
ed
th
eo
r
em
ac
h
ie
v
ed
b
etter
ac
cu
r
ac
y
o
f
6
9
.
2
7
% c
o
m
p
ar
ed
to
o
th
er
m
e
th
o
d
s
.
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
P
erfo
r
ma
n
ce
ev
a
lu
a
tio
n
o
f ra
n
d
o
m
fo
r
est w
ith
fea
tu
r
e
s
elec
tio
n
m
eth
o
d
s
in
p
r
ed
ictio
n
...
(
R
a
g
h
a
ve
n
d
r
a
S
)
355
A
m
et
h
o
d
s
i
m
ilar
to
p
r
in
cip
al
co
m
p
o
n
en
t
a
n
al
y
s
i
s
w
a
s
u
s
e
d
to
s
elec
t
th
e
i
m
p
o
r
tan
t
at
tr
ib
u
tes
w
a
s
d
ev
elo
p
ed
[
1
1
]
.
T
h
ese
attr
ib
u
tes
ar
e
g
iv
e
n
as
a
n
in
p
u
t
to
th
e
f
ee
d
f
o
r
w
ar
d
ar
tif
icial
n
eu
r
al
n
et
w
o
r
k
.
T
h
e
r
esu
l
t
ac
h
iev
ed
b
y
t
h
e
m
et
h
o
d
is
m
ea
s
u
r
ed
u
p
w
it
h
o
th
er
m
eth
o
d
s
o
f
th
e
f
ea
t
u
r
e
s
elec
tio
n
l
ik
e
T
ar
r
’
s
,
R
U
C
K’
s
,
p
r
in
cip
al
co
m
p
o
n
e
n
t
a
n
al
y
s
i
s
an
d
t
-
test
.
T
h
e
n
e
w
m
o
d
el
w
a
s
ap
p
lied
o
n
t
h
e
d
iab
etes
d
ataset.
T
esti
n
g
i
s
d
o
n
e
u
s
i
n
g
2
0
%
o
f
d
ata
a
n
d
r
e
m
ai
n
in
g
8
0
%
is
u
s
ed
f
o
r
tr
ain
i
n
g
.
T
h
e
p
r
o
p
o
s
ed
m
et
h
o
d
ac
h
ie
v
e
d
g
o
o
d
ac
cu
r
ac
y
o
f
7
5
.
2
2
% w
it
h
les
s
n
u
m
b
er
o
f
at
tr
ib
u
tes.
A
s
elec
t
iv
e
b
a
y
es
ian
cla
s
s
i
f
ier
is
p
r
o
p
o
s
ed
an
d
is
i
m
p
le
m
e
n
t
ed
o
n
d
iab
etes
d
ata
s
et
u
s
in
g
5
-
f
o
ld
cr
o
s
s
v
alid
atio
n
s
a
m
p
le
[
1
2
]
.
T
h
e
au
g
m
e
n
ted
b
a
y
esia
n
clas
s
i
f
ier
is
also
i
m
p
le
m
e
n
ted
o
n
t
h
e
s
a
m
e
d
ataset.
T
h
e
r
esu
l
t
o
f
s
elec
ti
v
e
b
a
y
es
ian
cla
s
s
i
f
ie
r
is
co
m
p
ar
ed
w
it
h
n
aï
v
e
b
a
y
es
a
n
d
au
g
m
e
n
ted
b
a
y
esi
an
clas
s
i
f
ier
.
Fro
m
th
e
r
es
u
lt
it
w
as
clea
r
t
h
at
s
elec
ti
v
e
b
a
y
esia
n
cla
s
s
i
f
i
er
g
iv
es
b
etter
ac
c
u
r
ac
y
t
h
a
n
t
h
e
n
aï
v
e
b
a
y
es
a
n
d
au
g
m
e
n
ted
b
a
y
esia
n
clas
s
i
f
ier
.
I
n
ad
d
itio
n
to
th
is
,
th
e
s
elec
ti
v
e
b
a
y
esia
n
clas
s
i
f
ier
ac
h
ie
v
es
b
etter
ac
cu
r
ac
y
o
f
7
9
.
9
4
% th
r
o
u
g
h
les
s
er
a
m
o
u
n
t
s
o
f
attr
ib
u
te
s
th
u
s
b
y
r
ed
u
cin
g
th
e
s
ize
o
f
t
h
e
d
ataset.
Fo
r
in
d
u
cti
v
e
co
n
ce
p
t
lear
n
in
g
an
E
v
o
lu
tio
n
ar
y
C
o
n
ce
p
t
L
ea
r
n
er
(
E
C
L
)
w
a
s
d
ev
elo
p
ed
an
d
th
r
e
e
d
if
f
er
e
n
t
s
elec
tio
n
m
ec
h
a
n
is
m
s
o
f
E
C
L
:
U
S
(
US
s
elec
t
io
n
o
p
er
atio
n
)
,
w
ei
g
h
ted
US
(
W
US
)
an
d
ex
p
o
n
e
n
tial
l
y
w
ei
g
h
ted
US
(
E
W
US)
w
er
e
i
m
p
le
m
e
n
ted
o
n
d
iab
etes
d
ataset
[
1
3
]
.
Fro
m
t
h
e
r
e
s
u
l
t
it
w
a
s
f
o
u
n
d
t
h
at
th
e
av
er
a
g
e
ac
cu
r
ac
y
ac
h
ie
v
ed
b
y
E
W
US
w
a
s
7
7
% a
n
d
b
etter
th
an
co
m
p
ar
ed
to
US a
n
d
W
US.
A
m
o
d
el
th
a
t
m
a
k
es
u
s
e
o
f
g
en
etic
alg
o
r
it
h
m
to
s
elec
t
i
m
p
o
r
tan
t
f
ea
tu
r
es
is
d
ev
elo
p
ed
in
p
ar
allel
w
it
h
m
ap
r
ed
u
ce
f
r
a
m
e
w
o
r
k
[
1
4
]
.
T
h
e
s
elec
te
d
f
ea
tu
r
es
ar
e
p
r
o
d
u
ce
d
to
k
-
Nea
r
est
Neig
h
b
o
r
class
if
ier
.
T
h
e
ex
p
er
i
m
e
n
t
i
s
ca
r
r
ied
o
u
t
o
n
d
iab
etes
d
ataset.
T
h
e
ac
cu
r
ac
y
o
f
f
it
n
es
s
i
s
ca
lc
u
late
d
u
s
i
n
g
k
-
Nea
r
est
Neig
h
b
o
r
.
Fro
m
t
h
e
r
esu
lt it
was seen
t
h
at
p
ar
allel
g
e
n
etic
al
g
o
r
ith
m
p
r
o
d
u
ce
s
b
etter
ac
cu
r
a
c
y
o
f
8
0
.
5
1
%.
A
p
o
w
er
f
u
l
m
et
h
o
d
is
p
r
o
p
o
s
ed
f
o
r
lo
w
d
i
m
en
s
io
n
a
l
clas
s
if
ica
tio
n
a
n
d
esti
m
atio
n
o
f
r
eg
r
ess
io
n
p
r
o
b
lem
s
[
1
5
]
.
C
las
s
i
f
icatio
n
d
if
f
ic
u
lt
y
m
a
y
b
e
co
n
s
id
er
ed
as
a
d
if
f
ic
u
lt
y
o
f
ap
p
r
o
x
i
m
ati
n
g
th
e
tr
ain
i
n
g
s
et.
A
m
u
lti
r
eso
lu
tio
n
f
r
a
m
e
w
o
r
k
i
s
b
u
i
lt
b
ased
o
n
ap
p
r
o
x
i
m
atio
n
s
a
n
d
o
r
g
a
n
ized
i
n
t
h
e
f
o
r
m
o
f
a
tr
ee
.
T
h
is
s
u
p
p
o
r
ts
f
o
r
ef
f
icie
n
t tr
ain
in
g
.
T
h
e
m
o
d
el
is
e
x
p
er
i
m
e
n
ted
o
n
d
iab
etes d
ataset
an
d
ac
h
ie
v
es
a
g
o
o
d
ac
cu
r
ac
y
.
An
ar
tif
ic
ial
i
m
m
u
n
e
r
ec
o
g
n
it
io
n
s
y
s
te
m
w
h
ic
h
ca
n
n
o
tice
th
e
ex
i
s
ten
ce
o
r
n
o
n
e
x
i
s
ten
ce
o
f
d
is
ea
s
e
is
d
ev
elo
p
ed
.
T
h
e
d
iab
etes
d
ataset
is
r
u
n
o
n
t
h
e
m
ac
h
i
n
e
o
n
an
a
v
er
ag
e
o
f
3
r
u
n
s
u
s
in
g
1
0
-
f
o
ld
cr
o
s
s
v
alid
atio
n
s
a
m
p
le.
T
h
e
ca
p
ab
ilit
y
o
f
th
e
m
o
d
el
is
co
m
p
ar
e
d
w
i
th
t
h
e
s
u
p
p
le
m
e
n
tar
y
m
e
th
o
d
s
li
k
e
I
n
cNE
t,
L
o
g
d
i
s
c
an
d
Dip
o
l9
2
.
T
h
e
ac
cu
r
ac
y
o
b
tain
ed
b
y
th
e
p
r
o
p
o
s
ed
s
y
s
te
m
w
a
s
7
4
.
1
%
an
d
w
as
b
etter
t
h
a
n
th
e
o
th
er
s
[
1
6
]
.
A
g
r
o
w
i
n
g
-
p
r
u
n
i
n
g
s
p
ik
in
g
n
e
u
r
o
n
n
et
w
o
r
k
co
n
s
i
s
ti
n
g
o
f
2
s
tag
e
lear
n
in
g
a
lg
o
r
it
h
m
is
d
ev
elo
p
ed
f
o
r
h
an
d
li
n
g
th
e
p
r
o
b
lem
s
o
f
p
atter
n
class
i
f
icatio
n
.
T
h
e
p
r
o
p
o
s
ed
n
et
w
o
r
k
is
co
n
s
is
ted
o
f
th
r
ee
lay
er
s
an
d
t
w
o
s
tag
e
s
o
f
lear
n
i
n
g
al
g
o
r
ith
m
an
d
ex
p
er
i
m
en
ted
o
n
d
iab
ete
s
d
ataset
[
1
7
]
.
T
h
e
o
u
tco
m
e
s
ar
e
ev
al
u
ated
w
it
h
b
atch
an
d
o
n
li
n
e
s
p
ik
in
g
n
e
u
r
o
n
.
Fro
m
t
h
e
r
esu
lt,
it
w
as
i
d
en
tifie
d
th
at
p
r
o
p
o
s
ed
g
r
o
w
i
n
g
-
p
r
u
n
i
n
g
s
p
ik
i
n
g
n
eu
r
al
n
et
w
o
r
k
a
c
h
iev
ed
b
ette
r
ac
cu
r
ac
y
o
f
7
1
.
1
%.
Data
m
i
n
i
n
g
m
et
h
o
d
s
lik
e
lo
g
is
t
ic
r
eg
r
ess
io
n
an
d
ar
ti
f
icia
l
n
eu
r
al
n
et
w
o
r
k
s
w
it
h
f
ea
tu
r
e
s
elec
tio
n
m
et
h
o
d
s
lik
e
f
o
r
w
ar
d
s
elec
t
io
n
an
d
b
ac
k
w
ar
d
eli
m
i
n
ati
o
n
ar
e
ap
p
lied
o
n
d
iab
etes
d
ataset
b
ased
o
n
th
e
en
tr
o
p
y
e
v
al
u
atio
n
m
et
h
o
d
[
1
8
]
.
T
h
e
ex
p
er
im
e
n
t
w
a
s
co
n
d
u
cted
u
s
in
g
W
E
K
A
.
Fro
m
th
e
r
es
u
lt
i
t
w
as
id
en
ti
f
ied
th
at
t
h
e
n
e
u
r
al
n
e
t
w
o
r
k
w
i
th
b
ac
k
w
ar
d
eli
m
i
n
at
io
n
u
s
in
g
p
er
ce
n
ta
g
e
s
p
lit
ac
h
iev
ed
a
n
ac
cu
r
ac
y
o
f
7
8
.
9
0
%.
Data
m
i
n
i
n
g
tec
h
n
iq
u
es
lik
e
l
o
g
is
tic
r
e
g
r
ess
io
n
a
n
d
ar
tific
ia
l
n
e
u
r
al
n
e
t
w
o
r
k
ar
e
ap
p
lied
o
n
d
iab
etes
d
ataset
w
it
h
f
ea
tu
r
e
s
elec
tio
n
m
et
h
o
d
s
li
k
e
f
o
r
w
ar
d
s
elec
ti
o
n
an
d
b
ac
k
w
ar
d
eli
m
i
n
atio
n
b
ased
o
n
th
e
m
ea
n
v
alu
e
o
f
t
h
e
attr
ib
u
te
s
[
1
9
]
.
Fro
m
th
e
e
x
p
er
i
m
e
n
t
r
es
u
lt
i
t
was
f
o
u
n
d
t
h
at
an
ac
c
u
r
ac
y
o
f
8
0
.
4
6
%
is
ac
h
iev
ed
b
y
lo
g
is
t
ic
r
eg
r
ess
io
n
w
h
en
co
m
p
ar
ed
to
n
e
u
r
al
n
et
w
o
r
k
.
Usi
n
g
th
e
t
h
r
esh
o
ld
v
al
u
e
o
f
ea
ch
attr
ib
u
te
a
n
e
x
p
er
i
m
e
n
t
w
a
s
co
n
d
u
cted
o
n
d
iab
etes
d
ataset
a
n
d
th
e
p
er
f
o
r
m
a
n
ce
s
o
f
t
h
e
d
ata
m
i
n
i
n
g
m
et
h
o
d
s
li
k
e
lo
g
i
s
ti
c
r
eg
r
ess
io
n
an
d
ar
ti
f
icial
n
e
u
r
al
n
et
w
o
r
k
s
ar
e
ev
al
u
a
ted
w
it
h
f
ea
tu
r
e
s
elec
ti
o
n
m
e
th
o
d
s
.
Fro
m
t
h
e
r
es
u
lt
it
w
as
id
en
ti
f
ied
t
h
at
lo
g
i
s
ti
c
r
eg
r
ess
io
n
u
s
i
n
g
b
ac
k
w
ar
d
eli
m
i
n
atio
n
ac
h
ie
v
e
d
an
o
f
ac
cu
r
ac
y
o
f
8
2
.
8
1
% w
h
en
co
m
p
ar
ed
to
n
e
u
r
al
n
et
w
o
r
k
[
2
0
]
.
Usi
n
g
n
eu
r
al
n
et
w
o
r
k
w
it
h
b
ac
k
p
r
o
p
ag
atio
n
an
d
d
if
f
er
en
t
d
at
a
m
i
n
i
n
g
tec
h
n
iq
u
e
s
lik
e
J
4
8
,
n
aïv
e
b
ay
e
s
a
n
d
SVM
ar
e
ap
p
lied
o
n
d
iab
etes
d
ataset
to
p
r
ed
ict
th
e
p
r
ese
n
ce
o
r
ab
s
en
ce
o
f
d
iab
etes
in
a
p
er
s
o
n
.
A
5
-
f
o
ld
cr
o
s
s
v
al
id
atio
n
s
a
m
p
le
is
u
s
ed
to
i
m
p
r
o
v
e
th
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
m
o
d
el.
B
a
s
ed
o
n
th
e
e
x
p
er
i
m
e
n
tal
r
esu
l
t
co
n
d
u
cted
an
ac
cu
r
ac
y
o
f
8
3
.
1
1
% w
a
s
ac
h
ie
v
ed
b
y
b
ac
k
p
r
o
p
ag
atio
n
alg
o
r
it
h
m
[
2
1
]
.
I
n
m
ed
ical
f
ield
to
e
x
p
lo
it
t
h
e
p
atie
n
ts
i
n
f
o
r
m
atio
n
,
c
las
s
if
ica
tio
n
s
y
s
te
m
s
ar
e
w
id
el
y
u
s
ed
o
n
d
iab
etes
d
ataset.
T
h
e
n
aïv
e
b
ay
es
i
s
ap
p
lied
f
o
r
class
if
icatio
n
an
d
f
o
r
attr
i
b
u
te
s
elec
tio
n
g
en
etic
al
g
o
r
ith
m
is
u
s
ed
[
2
2
]
.
Fr
o
m
th
e
e
x
p
er
i
m
e
n
tal
r
esu
lts
an
ac
cu
r
ac
y
o
f
7
8
.
6
9
%
is
ac
h
iev
ed
.
P
o
p
u
lar
tech
n
iq
u
es
li
k
e
d
ee
p
n
eu
r
al
n
e
t
w
o
r
k
s
an
d
SVM
ar
e
u
s
ed
to
id
en
tify
t
h
e
p
r
esen
ce
o
r
ab
s
en
ce
o
f
d
iab
etes
b
ased
o
n
th
e
ac
cu
r
ac
y
o
f
cr
o
s
s
v
al
id
atio
n
s
a
m
p
le
o
n
d
ia
b
etes d
ataset.
An
ac
cu
r
ac
y
o
f
7
7
.
8
6
% w
as a
c
h
ie
v
ed
f
r
o
m
t
h
e
s
aid
m
et
h
o
d
[
2
3
]
.
Usi
n
g
f
ea
t
u
r
e
s
elec
tio
n
f
o
r
cl
ass
i
f
icatio
n
o
f
th
e
d
ata
h
a
s
a
lar
g
e
n
u
m
b
er
o
f
b
en
e
f
its
:
d
ec
r
ea
s
e
i
n
co
m
p
u
tatio
n
al
d
if
f
ic
u
lt
y
d
u
e
to
d
ec
r
ea
s
e
in
d
i
m
en
s
io
n
a
lit
y
[
2
4
]
an
d
r
ed
u
ctio
n
in
n
o
is
e
to
e
n
h
a
n
c
e
th
e
clas
s
i
f
icatio
n
ac
cu
r
ac
y
[
2
5
]
.
Fro
m
t
h
e
liter
at
u
r
e
s
u
r
v
e
y
w
e
ca
n
id
e
n
ti
f
y
th
at
m
a
n
y
d
if
f
er
en
t
m
et
h
o
d
s
ar
e
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.
10
,
No
.
1
,
Feb
r
u
ar
y
2
0
2
0
:
353
-
359
356
ap
p
lied
o
n
th
e
d
iab
etes
d
atas
et.
So
m
e
m
et
h
o
d
s
u
s
in
g
f
u
ll
s
et
o
f
at
tr
ib
u
tes
an
d
s
o
m
e
u
s
e
s
th
e
s
u
b
s
ets
o
f
th
e
attr
ib
u
tes.
T
h
e
cla
s
s
i
f
icati
o
n
ac
cu
r
ac
y
ac
h
iev
ed
i
s
n
o
t
s
atis
f
ac
to
r
y
a
n
d
it
ca
n
b
e
f
u
r
t
h
er
i
m
p
r
o
v
ed
.
I
n
t
h
i
s
r
esear
ch
w
o
r
k
w
e
tr
y
to
i
m
p
r
o
v
e
th
e
ac
cu
r
ac
y
b
y
u
s
i
n
g
th
e
d
ata
m
i
n
i
n
g
tec
h
n
iq
u
e
lik
e
r
an
d
o
m
f
o
r
est
w
it
h
f
ea
t
u
r
e
s
elec
tio
n
m
eth
o
d
s
lik
e
f
o
r
w
ar
d
s
elec
tio
n
an
d
b
ac
k
w
ar
d
eli
m
i
n
atio
n
u
s
in
g
p
er
ce
n
tag
e
s
p
li
t
as
test
o
p
tio
n
.
I
n
th
e
n
e
x
t sect
io
n
w
e
s
tu
d
y
th
e
p
r
o
p
o
s
ed
f
r
a
m
e
w
o
r
k
o
f
th
e
r
esear
ch
ca
r
r
ied
o
u
t.
3.
P
RO
P
O
SE
D
F
RAM
E
WO
RK
T
h
e
p
r
o
p
o
s
ed
f
r
am
e
w
o
r
k
f
o
r
ev
alu
a
tin
g
t
h
e
P
I
MA
I
n
d
ian
Diab
etes
d
ataset
w
it
h
f
ea
tu
r
e
s
elec
tio
n
m
et
h
o
d
s
u
s
in
g
R
i
s
s
h
o
w
n
i
n
Fig
u
r
e
1
.
T
h
e
p
r
o
ce
s
s
o
f
ev
al
u
atio
n
is
as
f
o
llo
w
s
:
a.
T
h
e
f
ir
s
t
s
tep
is
t
h
e
s
elec
t
io
n
o
f
th
e
d
iab
etes d
atase
t.
b.
Fo
r
an
y
m
is
s
in
g
v
alu
e
s
in
t
h
e
d
ataset,
p
r
e
-
p
r
o
ce
s
s
in
g
i
s
d
o
n
e.
Sin
ce
th
e
d
ataset
co
n
s
id
er
ed
h
av
e
n
o
m
is
s
i
n
g
v
al
u
es,
s
o
n
o
p
r
e
-
p
r
o
ce
s
s
i
n
g
i
s
r
eq
u
ir
ed
.
T
h
e
d
ataset
is
tak
e
n
in
i
ts
o
r
ig
i
n
al
f
o
r
m
.
c.
W
e
f
in
d
th
e
e
n
tr
o
p
y
v
al
u
e
o
f
e
ac
h
attr
ib
u
te
o
f
t
h
e
d
ataset
u
s
i
n
g
(
1
)
I
nfo
(
D)
=
∑
(
)
(
1
)
w
h
er
e
D
is
t
h
e
attr
ib
u
te,
i
is
th
e
attr
ib
u
te
i
n
d
ex
,
p
i
is
th
e
p
r
o
b
ab
ilit
y
th
a
t
a
n
at
tr
ib
u
te
in
D
b
elo
n
g
s
t
o
a
class
an
d
m
is
t
h
e
to
tal
co
u
n
t
o
f
attr
ib
u
tes.
d.
No
t
all
th
e
f
ea
t
u
r
es i
n
th
e
d
ata
s
et
ar
e
i
m
p
o
r
tan
t i
n
p
r
ed
ictio
n
.
B
ased
o
n
th
e
en
tr
o
p
y
v
al
u
e
o
f
ea
ch
a
ttrib
u
te
,
ap
p
ly
f
ea
t
u
r
e
s
elec
tio
n
m
et
h
o
d
s
lik
e
f
o
r
w
ar
d
s
elec
tio
n
a
n
d
b
ac
k
w
ar
d
eli
m
i
n
atio
n
to
o
b
tain
t
h
e
d
if
f
er
e
n
t
s
u
b
s
et
s
o
f
f
ea
tu
r
e
s
.
e.
Fo
r
ea
ch
s
u
b
s
et
o
f
f
ea
t
u
r
e
we
ev
al
u
ate
t
h
e
p
er
f
o
r
m
an
ce
o
f
r
an
d
o
m
f
o
r
est
u
s
i
n
g
p
er
ce
n
t
ag
e
s
p
lit
as
tes
t
o
p
tio
n
.
f.
Fin
all
y
t
h
e
s
u
b
s
et
o
f
f
ea
t
u
r
es
w
h
ic
h
ac
h
ie
v
es b
etter
ac
cu
r
ac
y
ar
e
co
n
s
id
er
s
as
v
er
y
i
m
p
o
r
t
an
t a
ttrib
u
te
s
.
Fig
u
r
e
1
.
Fra
m
e
w
o
r
k
f
o
r
th
e
p
r
o
p
o
s
ed
w
o
r
k
b
ased
o
n
r
an
d
o
m
f
o
r
est
u
s
in
g
p
er
ce
n
ta
g
e
s
p
lit
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
I
n
th
i
s
r
esear
ch
w
o
r
k
w
e
ev
a
lu
ate
th
e
p
er
f
o
r
m
an
ce
o
f
t
h
e
d
ata
m
i
n
i
n
g
m
et
h
o
d
lik
e
r
an
d
o
m
f
o
r
est
w
it
h
f
ea
t
u
r
e
s
elec
tio
n
m
e
th
o
d
s
u
s
i
n
g
p
er
ce
n
ta
g
e
s
p
lit
as
test
o
p
tio
n
o
n
d
iab
etes
d
ataset
u
s
i
n
g
R
.
T
h
e
d
ia
b
etes
d
ataset
co
n
s
id
er
ed
in
th
e
r
ese
ar
ch
w
o
r
k
co
n
s
is
ts
o
f
7
6
8
in
s
tan
ce
s
a
n
d
9
attr
ib
u
tes.
T
ab
le
1
g
iv
es
t
h
e
lis
t
o
f
attr
ib
u
te
in
d
ataset
an
d
th
e
ir
m
ea
n
in
g
.
T
ab
le
2
an
d
T
a
b
le
3
g
iv
es
t
h
e
d
if
f
er
en
t
s
u
b
s
ets
o
f
attr
ib
u
tes
o
b
tain
ed
af
ter
ap
p
l
y
in
g
t
h
e
f
ea
t
u
r
e
s
e
le
ctio
n
m
et
h
o
d
s
an
d
t
h
e
b
est ac
c
u
r
ac
y
ac
h
iev
ed
b
y
r
an
d
o
m
f
o
r
est.
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
P
erfo
r
ma
n
ce
ev
a
lu
a
tio
n
o
f ra
n
d
o
m
fo
r
est w
ith
fea
tu
r
e
s
elec
tio
n
m
eth
o
d
s
in
p
r
ed
ictio
n
...
(
R
a
g
h
a
ve
n
d
r
a
S
)
357
T
ab
le
1
.
Diab
etes d
ataset
attr
ib
u
tes an
d
m
ea
n
in
g
A
t
t
r
i
b
u
t
e
M
e
a
n
i
n
g
p
r
e
g
N
u
mb
e
r
o
f
t
i
me
s p
r
e
g
n
a
n
t
p
l
a
s
P
l
a
sma
g
l
u
c
o
se
c
o
n
c
e
n
t
r
a
t
i
o
n
a
2
h
o
u
r
s i
n
a
n
o
r
a
l
g
l
u
c
o
se
t
o
l
e
r
a
n
c
e
t
e
st
p
r
e
s
D
i
a
st
o
l
i
c
b
l
o
o
d
p
r
e
ssu
r
e
(
mm
H
g
)
sk
i
n
T
r
i
c
e
p
s sk
i
n
f
o
l
d
t
h
i
c
k
n
e
ss (m
m)
i
n
s
u
2
-
H
o
u
r
se
r
u
m
i
n
s
u
l
i
n
(
m
u
U
/
ml
)
mass
B
o
d
y
mass
i
n
d
e
x
(
w
e
i
g
h
t
i
n
k
g
/
(
h
e
i
g
h
t
i
n
m)
^
2
)
p
e
d
i
D
i
a
b
e
t
e
s
p
e
d
i
g
r
e
e
f
u
n
c
t
i
o
n
a
g
e
A
g
e
(
y
e
a
r
s)
c
l
a
ss
C
l
a
ss v
a
r
i
a
b
l
e
(
0
o
r
1
)
T
ab
le
2
.
Dif
f
er
en
t
s
u
b
s
et
s
o
b
tain
ed
an
d
th
e
b
est ac
c
u
r
ac
y
ac
h
iev
ed
a
f
ter
ap
p
ly
i
n
g
f
o
r
w
ar
d
s
elec
tio
n
b
a
s
ed
o
n
en
tr
o
p
y
v
al
u
e
S
u
b
se
t
N
o
.
S
u
b
se
t
o
f
A
t
t
r
i
b
u
t
e
s
N
o
.
o
f
A
t
t
r
i
b
u
t
e
s
C
l
a
ssi
f
i
c
a
t
i
o
n
A
c
c
u
r
a
c
y
1
p
r
e
s,
c
l
a
ss
2
5
7
.
4
%
2
p
r
e
s,
p
e
d
i
,
c
l
a
ss
3
5
8
.
5
%
3
p
r
e
g
,
p
r
e
s,
p
e
d
i
,
c
l
a
ss
4
6
2
%
4
p
r
e
g
,
p
r
e
s,
s
k
i
n
,
p
e
d
i
,
c
l
a
ss
5
6
3
.
6
%
5
p
r
e
g
,
p
r
e
s,
s
k
i
n
,
i
n
su
,
p
e
d
i
,
c
l
a
ss
6
6
8
%
6
p
r
e
g
,
p
r
e
s,
s
k
i
n
,
i
n
su
,
p
e
d
i
,
a
g
e
,
c
l
a
ss
7
7
3
.
2
%
7
p
r
e
g
,
p
r
e
s,
s
k
i
n
,
i
n
su
,
mass
,
p
e
d
i
,
a
g
e
,
c
l
a
ss
8
7
5
.
5
%
8
F
u
l
l
s
e
t
o
f
a
t
t
r
i
b
u
t
e
s
9
8
3
.
8
%
T
ab
le
3
.
Dif
f
er
en
t
s
u
b
s
et
s
o
b
tain
ed
an
d
th
e
b
est ac
c
u
r
ac
y
ac
h
iev
ed
a
f
ter
ap
p
ly
i
n
g
b
ac
k
w
ar
d
eli
m
in
atio
n
b
ased
o
n
en
tr
o
p
y
v
al
u
e
S
u
b
se
t
N
o
.
S
u
b
se
t
o
f
A
t
t
r
i
b
u
t
e
s
N
o
.
o
f
A
t
t
r
i
b
u
t
e
s
C
l
a
ssi
f
i
c
a
t
i
o
n
A
c
c
u
r
a
c
y
1
p
r
e
g
,
p
l
a
s,
sk
i
n
,
i
n
s
u
,
mass,
p
e
d
i
,
a
g
e
,
c
l
a
ss
8
8
4
.
1
%
2
p
r
e
g
,
p
l
a
s,
sk
i
n
,
i
n
s
u
,
mass,
a
g
e
,
c
l
a
ss
7
8
3
.
1
%
3
p
l
a
s,
s
k
i
n
,
i
n
s
u
,
m
a
ss,
a
g
e
,
c
l
a
ss
6
8
1
.
4
%
4
p
l
a
s,
i
n
s
u
,
mass
,
a
g
e
,
c
l
a
ss
5
8
1
.
7
%
5
p
l
a
s,
m
a
ss,
a
g
e
,
c
l
a
ss
4
8
0
.
8
%
6
p
l
a
s,
m
a
ss,
c
l
a
ss
3
7
4
.
6
%
7
p
l
a
s,
c
l
a
ss
2
7
1
.
3
%
Fro
m
t
h
e
cla
s
s
i
f
icat
io
n
ac
c
u
r
ac
y
o
b
tain
ed
f
o
r
d
i
f
f
er
e
n
t
s
et
o
f
at
tr
ib
u
tes
as
s
h
o
w
n
i
n
T
ab
le
2
an
d
T
ab
le
3
,
w
e
id
en
tifie
d
th
at:
a.
T
h
e
ac
cu
r
ac
y
ac
h
ie
v
ed
f
o
r
f
u
ll
s
et
o
f
attr
ib
u
te
s
is
8
3
.
8
%.
b.
T
h
e
b
est
class
i
f
icatio
n
ac
h
ie
v
ed
is
b
y
Su
b
s
e
t
No
.
1
o
f
T
ab
l
e
3
is
8
4
.
1
%,
w
h
ic
h
is
b
etter
t
h
an
t
h
e
ac
c
u
r
ac
y
ac
h
iev
ed
f
o
r
f
u
ll
s
et
o
f
attr
ib
u
tes
an
d
o
th
er
m
eth
o
d
s
ap
p
lied
o
n
d
iab
etes
d
ataset.
T
h
e
co
m
p
ar
is
o
n
b
et
w
ee
n
th
e
ac
cu
r
ac
y
ac
h
ie
v
ed
b
y
th
e
p
r
o
p
o
s
ed
m
et
h
o
d
an
d
s
o
m
e
o
f
th
e
ex
i
s
ti
n
g
m
et
h
o
d
s
is
s
h
o
wn
in
Fig
u
r
e
2
.
Fig
u
r
e
2
.
C
o
m
p
ar
is
o
n
o
f
cla
s
s
if
icatio
n
ac
cu
r
ac
y
o
f
th
e
e
x
i
s
ti
n
g
m
et
h
o
d
w
it
h
t
h
e
p
r
o
p
o
s
ed
m
et
h
o
d
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.
10
,
No
.
1
,
Feb
r
u
ar
y
2
0
2
0
:
353
-
359
358
5.
CO
NCLU
SI
O
N
I
n
th
is
r
e
s
ea
r
ch
w
o
r
k
t
h
e
d
ata
m
i
n
in
g
m
et
h
o
d
li
k
e
r
an
d
o
m
f
o
r
est
w
ith
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
s
li
k
e
f
o
r
w
ar
d
s
elec
tio
n
a
n
d
b
ac
k
war
d
eli
m
in
a
tio
n
i
s
ap
p
lied
o
n
di
ab
etes
d
ataset
u
s
i
n
g
p
er
ce
n
tag
e
s
p
lit
a
s
test
o
p
tio
n
.
T
h
e
ex
p
er
im
e
n
t
is
i
m
p
le
m
e
n
ted
in
R
s
t
u
d
io
u
s
i
n
g
R
p
r
o
g
r
am
m
i
n
g
lan
g
u
ag
e
s
.
Fro
m
th
e
e
x
p
er
i
m
e
n
tal
r
esu
lt
s
w
e
ca
n
o
b
s
er
v
e
t
h
e
f
o
ll
o
w
i
n
g
:
a.
T
h
e
class
if
ica
tio
n
ac
c
u
r
ac
y
ac
h
iev
ed
f
o
r
f
u
ll set o
f
at
tr
ib
u
tes
is
8
3
.
8
%.
b.
T
h
e
class
if
icat
io
n
ac
cu
r
ac
y
a
ch
iev
ed
b
y
t
h
e
p
r
o
p
o
s
ed
m
et
h
o
d
is
8
4
.
1
%
w
it
h
7
attr
ib
u
tes
n
a
m
el
y
p
r
eg
,
p
las,
s
k
i
n
,
in
s
u
,
m
as
s
,
p
ed
i,
an
d
ag
e.
c.
T
h
e
ac
cu
r
ac
y
ac
h
iev
ed
is
b
etter
th
an
an
y
o
f
th
e
e
x
is
ti
n
g
m
et
h
o
d
s
o
n
d
iab
etes
d
ataset
an
d
th
e
ac
cu
r
a
c
y
ac
h
iev
ed
b
y
u
s
i
n
g
f
u
ll
s
et
o
f
at
tr
ib
u
tes
as
s
h
o
w
n
in
F
ig
u
r
e
2
.
Fu
r
t
h
er
r
esear
ch
er
s
ca
n
co
m
p
ar
e
th
e
C
las
s
i
f
icatio
n
A
cc
u
r
ac
y
w
it
h
o
th
er
tech
n
iq
u
e
s
lik
e
S
VM
,
NN,
an
d
NN
w
it
h
K
f
o
ld
,
d
ee
p
lear
n
in
g
a
n
d
m
a
y
ac
h
ie
v
e
b
etter
C
lass
i
f
icatio
n
A
cc
u
r
ac
y
.
ACK
NO
WL
E
D
G
E
M
E
NT
S
I
w
o
u
ld
li
k
e
e
x
p
r
ess
m
y
d
ee
p
g
r
atit
u
d
e
to
t
h
e
Ho
D
a
n
d
Staf
f
o
f
C
o
m
p
u
ter
Scie
n
ce
a
n
d
E
n
g
i
n
ee
r
i
n
g
d
ep
ar
tm
e
n
t o
f
C
HR
I
ST
Dee
m
ed
to
b
e
Un
iv
er
s
it
y
,
B
an
g
alo
r
e
f
o
r
s
u
p
p
o
r
tin
g
m
e
i
n
d
o
in
g
t
h
i
s
r
esear
ch
w
o
r
k
.
RE
F
E
R
E
NC
E
S
[1
]
S
.
K.
W
a
sa
n
,
e
t
a
l
.
,
“
T
h
e
Im
p
a
c
t
o
f
Da
ta
M
in
in
g
T
e
c
h
n
iq
u
e
s
o
n
M
e
d
ica
l
Dia
g
n
o
stics
,
”
Da
ta
S
c
ien
c
e
J
o
u
rn
a
l
,
v
o
l.
5
,
p
p
.
1
1
9
-
1
2
6
,
2
0
0
6
.
[2
]
T
.
K.
Ho
,
“
T
h
e
Ra
n
d
o
m
S
u
b
sp
a
c
e
M
e
th
o
d
f
o
r
Co
n
stru
c
ti
n
g
De
c
isio
n
F
o
re
sts,”
IEE
E
T
ra
n
s
a
c
t
io
n
o
n
Pa
t
ter
n
An
a
lys
is
a
n
d
M
a
c
h
i
n
e
In
tell
ig
e
n
c
e
,
v
o
l.
2
0
,
p
p
.
8
3
2
-
8
4
4
,
1
9
9
8
.
[3
]
W
.
G
.
T
o
u
w
,
e
t
a
l
.
,
“
Da
ta
M
in
i
n
g
in
th
e
L
if
e
S
c
ien
c
e
s
w
it
h
Ra
n
d
o
m
F
o
re
st:
a
W
a
lk
in
th
e
p
a
rk
o
r
lo
st
i
n
Ju
n
g
le?
”
Briefin
g
s i
n
Bi
o
i
n
fo
rm
a
ti
c
s
,
v
o
l.
1
4
,
p
p
.
3
1
5
-
3
2
6
,
2
0
1
2
.
[4
]
R.
A
b
ra
h
a
m
,
e
t
a
l
.
,
“
Eff
e
c
ti
v
e
Disc
re
ti
z
a
ti
o
n
a
n
d
Hy
b
rid
F
e
a
tu
r
e
S
e
lec
ti
o
n
Us
in
g
Na
ïv
e
Ba
y
e
si
a
n
Clas
sif
ier
f
o
r
M
e
d
ica
l
Da
ta M
in
i
n
g
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
C
o
mp
u
ta
t
io
n
a
l
I
n
t
e
ll
ig
e
n
c
e
Res
e
a
rc
h
,
v
o
l.
5
,
p
p
.
1
1
6
-
1
2
9
,
2
0
0
9
.
[5
]
M
.
L
.
Ra
y
m
e
r,
e
t
a
l
.
,
“
Kn
o
w
led
g
e
D
isc
o
v
e
r
y
in
M
e
d
ica
l
a
n
d
Bi
o
l
o
g
ica
l
Da
tas
e
ts
Us
in
g
a
H
y
b
rid
B
a
y
e
s
Clas
si
f
ier/
Ev
o
lu
ti
o
n
a
ry
A
lg
o
rit
h
m
,
”
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
S
y
ste
ms
,
M
a
n
a
n
d
Cy
b
e
rn
a
ti
c
s
,
v
o
l
.
3
3
,
p
p
.
8
0
2
-
8
1
3
,
2
0
0
3
.
[6
]
M
.
B.
Do
w
latsh
a
h
i
a
n
d
M
.
Re
z
a
e
ian
,
“
T
ra
in
in
g
S
p
ik
in
g
Ne
u
ro
n
s
w
it
h
G
ra
v
it
a
ti
o
n
a
l
S
e
a
rc
h
A
l
g
o
rit
h
m
f
o
r
Da
ta
Clas
sif
ic
a
ti
o
n
,
”
IEE
E
1
st
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
S
wa
r
m
In
telli
g
e
n
c
e
a
n
d
Evo
lu
ti
o
n
a
ry
Co
mp
u
ta
ti
o
n
,
p
p
.
5
3
-
5
8
,
2
0
1
6
.
[7
]
E.
T
u
b
a
,
e
t
a
l
.
,
“
A
d
ju
ste
d
Ba
t
A
l
g
o
rit
h
m
f
o
r
T
u
n
in
g
o
f
S
u
p
p
o
rt
Ve
c
to
r
M
a
c
h
in
e
P
a
ra
m
e
ters
,
”
IEE
E
Co
n
g
re
ss
on
Evo
lu
ti
o
n
a
ry
Co
mp
u
t
a
ti
o
n
,
p
p
.
2
2
2
5
-
2
2
3
2
,
2
0
1
6
.
[8
]
R.
Bru
n
i
a
n
d
G
.
Bian
c
h
i,
“
Eff
e
c
ti
v
e
Cla
ss
i
f
ica
ti
o
n
Us
in
g
a
S
m
a
ll
T
r
a
in
in
g
S
e
t
Ba
se
d
o
n
Disc
re
ti
z
a
ti
o
n
a
n
d
S
tatisti
c
a
l
A
n
a
l
y
sis,”
IEE
E
T
ra
n
s
a
c
ti
o
n
s o
n
Kn
o
wled
g
e
a
n
d
D
a
ta
En
g
i
n
e
e
rin
g
,
v
o
l.
2
7
,
p
p
.
2
3
4
9
-
2
3
6
1
,
2
0
1
5
.
[9
]
P
.
S
y
k
a
c
e
k
a
n
d
S
.
R
o
b
e
rts,
“
Ad
a
p
ti
v
e
Clas
sif
ica
ti
o
n
b
y
V
a
riat
io
n
a
l
Ka
lm
a
n
F
il
teri
n
g
,
”
Ad
v
a
n
c
e
s
in
Ne
u
ra
l
In
fo
rm
a
t
io
n
Pro
c
e
ss
in
g
S
y
ste
ms
(
NIPS
)
,
p
p
.
7
3
7
-
7
4
4
,
2
0
0
2
.
[1
0
]
F
.
J.
L
i,
e
t
a
l
.
,
“
M
u
lt
ig
ra
n
u
lati
o
n
In
f
o
rm
a
ti
o
n
F
u
si
o
n
:
A
De
m
p
ste
r
-
S
h
a
f
e
r
Ev
id
e
n
c
e
T
h
e
o
r
y
B
a
se
d
Clu
ste
rin
g
En
se
m
b
le
M
e
th
o
d
,
”
Pr
o
c
e
e
d
in
g
s
o
f
IEE
E
I
n
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
M
a
c
h
i
n
e
L
e
a
rn
i
n
g
a
n
d
Cy
b
e
rn
e
ti
c
s
(
ICM
L
C)
,
v
o
l.
1
,
p
p
.
58
-
6
3
,
2
0
1
5
.
[1
1
]
S
.
J.
P
e
ra
n
to
n
is
a
n
d
V
.
V
i
rv
il
is,
“
In
p
u
t
F
e
a
tu
re
Ex
trac
ti
o
n
f
o
r
M
u
lt
il
a
y
e
re
d
P
e
rc
e
p
tro
n
s
Us
in
g
S
u
p
e
rv
i
se
d
P
ri
n
c
ip
a
l
Co
m
p
o
n
e
n
t
A
n
a
ly
sis,”
Ne
u
ra
l
Pro
c
e
ss
in
g
L
e
tt
e
rs
,
v
o
l.
1
0
,
p
p
.
2
4
3
-
2
5
2
,
1
9
9
9
.
[1
2
]
C.
A
.
Ra
tan
a
m
a
h
a
tan
a
a
n
d
D.
Gu
n
o
p
u
lo
s,
“
F
e
a
tu
re
S
e
lec
ti
o
n
f
o
r
th
e
Na
ïv
e
Ba
y
e
sia
n
Clas
sif
ier
Us
in
g
De
c
isio
n
T
re
e
s,”
Ap
p
li
e
d
Arti
fi
c
i
a
l
I
n
telli
g
e
n
c
e
(
AA
I)
,
v
o
l.
1
7
,
p
p
.
4
7
5
-
4
8
7
,
2
0
0
3
.
[1
3
]
F
.
Div
in
a
a
n
d
E.
M
a
rc
h
io
ri
,
“
Kn
o
w
led
g
e
-
Ba
se
d
Ev
o
lu
ti
o
n
a
ry
S
e
a
rc
h
f
o
r
In
d
u
c
ti
v
e
Co
n
c
e
p
t
L
e
a
rn
i
n
g
,
”
Kn
o
wled
g
e
In
c
o
rp
o
ra
ti
o
n
in
Ev
o
lu
ti
o
n
a
ry
Co
mp
u
ta
ti
o
n
,
S
p
ri
n
g
e
r
,
v
o
l
.
1
6
7
,
p
p
.
2
3
7
-
2
5
3
,
2
0
0
5
.
[1
4
]
G
.
T
.
Hild
a
a
n
d
R
.
R.
Ra
jala
x
m
i,
“
Eff
e
c
ti
v
e
F
e
a
tu
r
e
S
e
lec
ti
o
n
f
o
r
S
u
p
e
rv
ise
d
L
e
a
rn
in
g
Us
in
g
G
e
n
e
ti
c
A
lg
o
rit
h
m
,
”
IEE
E
2
nd
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
E
lec
tro
n
ics
a
n
d
Co
mm
u
n
ica
t
io
n
S
y
ste
ms
(
ICECS
2
0
1
5
)
,
p
p
.
9
0
9
-
9
1
4
,
2
0
1
5
.
[1
5
]
I.
Blay
v
a
s
a
n
d
R.
Kim
m
e
l,
“
M
a
c
h
in
e
L
e
a
rn
in
g
v
ia
M
u
lt
ires
o
l
u
ti
o
n
A
p
p
r
o
x
im
a
ti
o
n
s,”
IEI
CE
T
ra
n
sa
c
ti
o
n
o
n
In
fo
rm
a
t
io
n
S
y
ste
m,
v
ol
.
E8
6
-
D,
p
p
.
1
1
7
2
-
1
1
8
0
,
2
0
0
3
.
[1
6
]
A
.
W
a
t
k
in
s,
e
t
a
l
.
,
“
A
rti
f
icia
l
I
m
m
u
n
e
R
e
c
o
g
n
it
io
n
S
y
ste
m
(
A
I
RS
):
a
n
I
m
m
u
n
e
In
sp
ired
S
u
p
e
r
v
ise
d
L
e
a
rn
in
g
A
l
g
o
rit
h
m
,
”
Ge
n
e
ti
c
Pro
g
ra
mm
in
g
a
n
d
Evo
lva
b
le M
a
c
h
i
n
e
s
,
v
o
l.
5
,
p
p
.
2
9
1
-
3
1
7
,
2
0
0
4
.
[1
7
]
S
.
Do
ra
,
e
t
a
l
.
,
“
A
Tw
o
S
tag
e
L
e
a
rn
in
g
A
lg
o
rit
h
m
f
o
r
a
G
ro
w
i
n
g
-
P
r
u
n
i
n
g
S
p
ik
i
n
g
Ne
u
ra
l
Ne
tw
o
rk
f
o
r
P
a
tt
e
rn
Clas
sif
ic
a
ti
o
n
P
r
o
b
lem
s,”
In
ter
n
a
ti
o
n
a
l
J
o
in
t
C
o
n
fer
e
n
c
e
o
n
Ne
u
r
a
l
Ne
two
rk
s (
IJ
CNN)
,
p
p
.
1
-
7
,
2
0
1
5
.
[1
8
]
S.
Ra
g
h
a
v
e
n
d
ra
a
n
d
M
.
In
d
iram
m
a
,
“
P
e
rf
o
rm
a
n
c
e
Ev
a
lu
a
ti
o
n
o
f
L
o
g
isti
c
Re
g
re
ss
io
n
a
n
d
A
rti
f
icia
l
Ne
u
ra
l
Ne
tw
o
rk
M
o
d
e
l
w
it
h
F
e
a
tu
re
S
e
lec
ti
o
n
M
e
th
o
d
s
Us
in
g
Cro
ss
V
a
li
d
a
ti
o
n
S
a
m
p
le
a
n
d
P
e
rc
e
n
tag
e
S
p
li
t
o
n
M
e
d
ica
l
Da
tas
e
ts,
”
In
ter
n
a
t
io
n
a
l
Co
n
fer
e
n
c
e
o
n
Em
e
rg
in
g
Res
e
a
rc
h
i
n
Co
m
p
u
t
in
g
,
In
fo
rm
a
t
io
n
,
C
o
mm
u
n
ica
ti
o
n
a
n
d
A
p
p
li
c
a
ti
o
n
s
,
v
o
l.
2
,
2
0
1
4
.
[1
9
]
S.
Ra
g
h
a
v
e
n
d
ra
a
n
d
M
.
In
d
iram
m
a
,
“
Clas
si
f
ic
a
ti
o
n
a
n
d
P
re
d
icti
o
n
M
o
d
e
l
u
sin
g
Hy
b
rid
T
e
c
h
n
iq
u
e
f
o
r
M
e
d
ica
l
Da
tas
e
ts,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
r
n
a
l
o
f
Co
m
p
u
ter
A
p
p
l
ica
ti
o
n
s
,
v
o
l.
1
2
7
,
p
p
.
2
0
-
1
5
,
2
0
1
5
.
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
P
erfo
r
ma
n
ce
ev
a
lu
a
tio
n
o
f ra
n
d
o
m
fo
r
est w
ith
fea
tu
r
e
s
elec
tio
n
m
eth
o
d
s
in
p
r
ed
ictio
n
...
(
R
a
g
h
a
ve
n
d
r
a
S
)
359
[2
0
]
S.
Ra
g
h
a
v
e
n
d
ra
a
n
d
M
.
In
d
iram
m
a
,
“
H
y
b
rid
Da
t
a
M
in
in
g
M
o
d
e
l
f
o
r
th
e
Cl
a
ss
i
f
ica
ti
o
n
a
n
d
P
re
d
ictio
n
o
f
M
e
d
ica
l
Da
tas
e
ts,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
r
n
a
l
o
f
Kn
o
wled
g
e
En
g
i
n
e
e
rin
g
a
n
d
S
o
ft
Da
ta
P
a
ra
d
ig
ms
,
v
o
l.
5
,
p
p
.
2
6
2
-
2
8
4
,
2
0
1
7
.
[2
1
]
F
.
G
.
W
o
ld
e
m
ich
a
e
l
a
n
d
S
.
M
e
n
a
ria,
“
P
re
d
ictio
n
o
f
Dia
b
e
tes
Us
i
n
g
Da
t
a
M
in
i
n
g
T
e
c
h
n
iq
u
e
s,”
2
nd
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
T
re
n
d
s In
El
e
c
tro
n
ics
a
n
d
I
n
fo
rm
a
ti
c
s
,
p
p
.
4
1
4
-
4
1
8
,
2
0
1
8
.
[2
2
]
D.
K.
Ch
o
u
b
e
y
,
e
t
a
l
.
,
“
Clas
sif
ic
a
ti
o
n
o
f
P
im
a
In
d
ian
Dia
b
e
tes
Da
tas
e
t
Us
i
n
g
Na
ï
v
e
Ba
y
e
s
w
it
h
G
e
n
e
ti
c
A
l
g
o
rit
h
m
a
s an
A
tt
rib
u
te S
e
lec
ti
o
n
,
”
Co
mm
u
n
ica
ti
o
n
a
n
d
Co
mp
u
ti
n
g
S
y
ste
ms
,
T
a
y
l
o
r
&
Fra
n
c
is
Gr
o
u
p
,
p
p
.
4
5
1
-
4
5
5
,
2
0
1
7
.
[2
3
]
S
.
W
e
i,
e
t
a
l
.
,
“
A
Co
m
p
re
h
e
n
siv
e
Ex
p
lo
ra
ti
o
n
to
th
e
M
a
c
h
i
n
e
L
e
a
r
n
in
g
T
e
c
h
n
iq
u
e
f
o
r
Dia
b
e
tes
Da
tas
e
t,
”
IEE
E
4
th
W
o
rld
Fo
r
u
m o
n
I
n
ter
n
e
t
o
f
T
h
in
g
s
,
p
p
.
2
9
1
-
2
9
5
,
2
0
1
8
.
[2
4
]
R.
A
b
ra
h
a
m
,
e
t
a
l
.,
“
Eff
e
c
ti
v
e
Disc
re
ti
z
a
ti
o
n
a
n
d
Hy
b
rid
F
e
a
tu
r
e
S
e
lec
ti
o
n
Us
in
g
Na
ïv
e
Ba
y
e
si
a
n
Clas
sif
ier
f
o
r
M
e
d
ica
l
Da
ta M
in
i
n
g
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
C
o
mp
u
ta
t
io
n
a
l
I
n
t
e
ll
ig
e
n
c
e
Res
e
a
rc
h
,
v
o
l.
5
,
p
p
.
1
1
6
-
1
2
9
,
2
0
0
9
.
[2
5
]
Q.
Ch
e
n
g
,
e
t
a
l
.,
“
L
o
g
isti
c
Re
g
re
s
sio
n
f
o
r
F
e
a
tu
re
S
e
lec
ti
o
n
a
n
d
S
o
f
t
Clas
sif
i
c
a
ti
o
n
o
f
Re
m
o
te S
e
n
si
n
g
Da
ta,”
IEE
E
Ge
o
sc
ien
c
e
a
n
d
Rem
o
te S
e
n
si
n
g
L
e
tt
e
rs
,
v
o
l.
3
,
p
p
.
4
9
1
-
4
9
4
,
2
0
0
6
.
B
I
O
G
RAP
H
I
E
S
O
F
AUTH
O
RS
Dr
.
Ra
g
h
a
v
e
n
d
r
a
S
is
c
u
rre
n
tl
y
w
o
rk
in
g
a
s
A
ss
o
c
iate
P
ro
f
e
ss
o
r
in
th
e
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
a
t
CHRIST
De
e
m
e
d
to
b
e
Un
iv
e
rsit
y
,
Ba
n
g
a
lo
re
.
He
c
o
m
p
lete
d
h
is
P
h
.
D.
d
e
g
re
e
in
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
in
e
e
rin
g
f
ro
m
V
T
U,
B
e
lg
a
u
m
,
In
d
ia
in
2
0
1
7
a
n
d
h
a
s
1
4
y
e
a
rs
o
f
tea
c
h
in
g
e
x
p
e
rien
c
e
.
His
i
n
tere
sts
in
c
lu
d
e
Da
ta
M
i
n
in
g
a
n
d
Bi
g
d
a
ta.
S
a
n
to
sh
K
u
m
a
r
J
is
c
u
rre
n
tl
y
w
o
rk
in
g
a
s
A
ss
o
c
iate
P
ro
f
e
ss
o
r
in
th
e
De
p
a
rtme
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
a
t
K
.
S
.
S
c
h
o
o
l
o
f
E
n
g
in
e
e
rin
g
a
n
d
M
a
n
a
g
e
m
e
n
t
,
B
a
n
g
a
l
o
r
e
.
He
i
s
p
u
rsu
i
n
g
P
h
.
D.
i
n
V
T
U,
Be
lg
a
u
m
,
In
d
ia.
He
h
a
s
1
0
y
e
a
r
s
o
f
t
e
a
c
h
in
g
a
n
d
3
y
e
a
rs
o
f
in
d
u
stry
e
x
p
e
rien
c
e
.
H
e
is
sp
e
c
ialize
d
in
Big
d
a
ta
stre
a
m
in
g
a
n
a
l
y
sis.
His
re
se
ra
c
h
to
p
ics
in
c
lu
d
e
sBig
d
a
ta w
it
h
m
a
c
h
in
e
lea
rn
in
g
.
Evaluation Warning : The document was created with Spire.PDF for Python.