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.
8
,
No
.
5
,
Octo
b
e
r
2
0
1
8
,
p
p
.
3
9
6
6
~3
9
7
5
I
SS
N:
2
0
8
8
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v8
i
5
.
p
p
3
9
6
6
-
397
5
3966
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
co
r
e
.
co
m/
jo
u
r
n
a
ls
/in
d
ex
.
p
h
p
/
I
JE
C
E
A Co
m
pa
ra
tive A
na
ly
sis
on the
Ev
a
lua
tion o
f
Cla
ss
ificatio
n
Alg
o
rith
m
s
in
the
P
redi
ction o
f
Dia
betes
Ra
t
na
P
a
t
il
1
,
S
ha
ra
v
a
ri
T
a
m
a
ne
2
1
De
p
a
rt
m
e
n
t
o
f
Co
m
p
u
ter E
n
g
in
e
e
rin
g
,
Ba
b
a
sa
h
e
b
Am
b
e
d
k
a
r
M
a
r
a
th
w
a
d
a
Un
iv
e
rsit
y
,
In
d
ia
2
De
p
a
rt
m
e
n
t
o
f
Co
m
p
u
ter E
n
g
in
e
e
rin
g
,
JN
EC,
A
u
ra
n
g
a
b
a
d
,
In
d
i
a
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
a
n
9
,
2
0
1
8
R
ev
i
s
ed
Mar
15
,
2
0
1
8
A
cc
ep
ted
J
u
l
4
,
2
0
1
8
Da
ta
m
in
in
g
tec
h
n
i
q
u
e
s
a
re
a
p
p
li
e
d
i
n
m
a
n
y
a
p
p
li
c
a
ti
o
n
s
a
s
a
sta
n
d
a
rd
p
ro
c
e
d
u
re
f
o
r
a
n
a
ly
z
in
g
th
e
larg
e
v
o
lu
m
e
o
f
a
v
a
il
a
b
le
d
a
ta,
e
x
trac
t
in
g
u
se
f
u
l
in
f
o
rm
a
ti
o
n
a
n
d
k
n
o
w
led
g
e
to
su
p
p
o
rt
th
e
m
a
jo
r
d
e
c
isio
n
-
m
a
k
in
g
p
ro
c
e
ss
e
s.
Dia
b
e
tes
m
e
ll
it
u
s
is
a
c
o
n
ti
n
u
i
n
g
,
g
e
n
e
ra
l,
d
e
a
d
l
y
s
y
n
d
ro
m
e
o
c
c
u
rrin
g
a
ll
a
ro
u
n
d
t
h
e
w
o
rld
.
It
is
c
h
a
ra
c
teriz
e
d
b
y
h
y
p
e
rg
l
y
c
e
m
ia
o
c
c
u
rrin
g
d
u
e
t
o
a
b
n
o
rm
a
li
ti
e
s
in
i
n
su
li
n
se
c
re
ti
o
n
w
h
ich
w
o
u
ld
i
n
tu
r
n
re
su
lt
in
irr
e
g
u
lar
rise
o
f
g
lu
c
o
se
lev
e
l.
In
re
c
e
n
t
y
e
a
rs,
th
e
im
p
a
c
t
o
f
Dia
b
e
tes
m
e
ll
it
u
s
h
a
s
in
c
re
a
se
d
to
a
g
re
a
t
e
x
ten
t
e
sp
e
c
ial
ly
in
d
e
v
e
lo
p
in
g
c
o
u
n
tri
e
s
li
k
e
In
d
ia.
T
h
is
is
m
a
in
ly
d
u
e
to
th
e
irreg
u
lariti
e
s in
th
e
f
o
o
d
h
a
b
it
s
a
n
d
li
f
e
st
y
le.
Th
u
s,
e
a
rly
d
iag
n
o
sis
a
n
d
c
las
sif
ica
ti
o
n
o
f
th
i
s
d
e
a
d
ly
d
ise
a
se
h
a
s
b
e
c
o
m
e
a
n
a
c
ti
v
e
a
re
a
o
f
re
se
a
rc
h
in
th
e
las
t
d
e
c
a
d
e
.
Nu
m
e
ro
u
s
c
lu
ste
rin
g
a
n
d
c
las
sif
ica
ti
o
n
s
tec
h
n
iq
u
e
s
a
re
a
v
a
il
a
b
le
in
th
e
li
tera
tu
re
to
v
isu
a
li
z
e
te
m
p
o
ra
l
d
a
ta
to
id
e
n
ti
fy
tren
d
s
f
o
r
c
o
n
tro
ll
in
g
d
iab
e
tes
m
e
ll
it
u
s.
T
h
is
w
o
rk
p
re
se
n
ts
a
n
e
x
p
e
rim
e
n
tal
stu
d
y
o
f
se
v
e
ra
l
a
lg
o
rit
h
m
s
w
h
ich
c
las
si
f
ies
Dia
b
e
tes
M
e
l
li
tu
s
d
a
ta
e
ffe
c
ti
v
e
l
y
.
T
h
e
e
x
isti
n
g
a
lg
o
rit
h
m
s
a
re
a
n
a
l
y
z
e
d
th
o
ro
u
g
h
ly
to
id
e
n
ti
fy
th
e
ir
a
d
v
a
n
tag
e
s
a
n
d
li
m
it
a
ti
o
n
s.
T
h
e
p
e
rf
o
r
m
a
n
c
e
a
ss
e
ss
m
e
n
t
o
f
th
e
e
x
isti
n
g
a
lg
o
rit
h
m
s is
c
a
rried
o
u
t
t
o
d
e
term
in
e
th
e
b
e
st ap
p
r
o
a
c
h
.
K
ey
w
o
r
d
:
Data
m
i
n
i
n
g
Diab
etes
m
el
lit
u
s
C
las
s
i
f
icatio
n
Ma
ch
i
n
e
l
ea
r
n
i
n
g
R
OC
Co
p
y
rig
h
t
©
2
0
1
8
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
atn
a
P
atil
,
Dep
ar
t
m
en
t o
f
C
o
m
p
u
ter
E
n
g
i
n
ee
r
in
g
,
B
ab
asah
eb
Am
b
ed
k
ar
Ma
r
ath
w
ad
a
U
n
iv
er
s
it
y
,
I
n
d
ia
.
E
m
ail:
r
at
n
a.
n
it
in
.
p
atil
@
g
m
a
il
.
co
m
1.
I
NT
RO
D
UCT
I
O
N
Dia
b
et
es
m
ellitu
s
is
a
g
r
o
u
p
o
f
m
etab
o
l
ic
d
is
ea
s
es
in
w
h
ich
a
p
e
r
s
o
n
ex
p
er
ien
c
es
h
ig
h
b
l
o
o
d
g
lu
c
o
s
e
lev
els
ei
th
er
b
e
ca
u
s
e
th
e
b
o
d
y
p
r
o
d
u
ce
s
in
a
d
e
q
u
at
e
in
s
u
lin
o
r
th
e
b
o
d
y
ce
lls
d
o
n
o
t
r
esp
o
n
d
p
r
o
p
e
r
ly
to
th
e
in
s
u
lin
p
r
o
d
u
ce
d
b
y
th
e
b
o
d
y
.
P
a
tien
ts
w
ith
d
iab
etes
o
f
ten
e
x
p
e
r
ien
c
e
f
r
e
q
u
en
t
u
r
in
ati
o
n
(
p
o
ly
u
r
ia)
,
in
c
r
e
ase
d
th
ir
s
t
(
p
o
ly
d
i
p
s
ia
)
an
d
in
cr
ea
s
e
d
h
u
n
g
er
(
p
o
ly
p
h
ag
i
a)
[
1
]
,
[
2
]
.
T
h
e
3
T
y
p
es
o
f
D
ia
b
e
tes:
a.
T
y
p
e
1
Dia
b
e
tes
I
n
th
is
ty
p
e
o
f
d
ia
b
etes
,
th
e
b
o
d
y
d
o
es
n
o
t
p
r
o
d
u
c
e
en
o
u
g
h
i
n
s
u
lin
.
T
h
is
ty
p
e
p
f
d
i
ab
et
es
is
also
r
ef
e
r
r
ed
t
o
as
in
s
u
lin
-
d
e
p
en
d
en
t
d
i
ab
et
es,
j
u
v
en
il
e
d
ia
b
e
tes
o
r
ea
r
ly
-
o
n
s
et
d
ia
b
e
tes
.
T
y
p
e
1
d
ia
b
et
es
u
s
u
ally
d
ev
el
o
p
s
b
ef
o
r
e
a
p
e
r
s
o
n
is
4
0
-
y
ea
r
s
-
o
l
d
i.e
.
,
in
ea
r
ly
ad
u
lth
o
o
d
o
r
teen
ag
e.
Pati
en
ts
w
ith
t
y
p
e
1
d
ia
b
e
tes w
ill n
ee
d
t
o
tak
e
in
s
u
lin
in
j
ec
ti
o
n
s
f
o
r
th
e
r
est
o
f
th
e
ir
lif
e.
T
h
ey
m
u
s
t
also
en
s
u
r
e
p
r
o
p
e
r
b
l
o
o
d
-
g
l
u
co
s
e
l
ev
els
b
y
ca
r
r
y
in
g
o
u
t
r
eg
u
la
r
b
l
o
o
d
t
est
s
an
d
f
o
ll
o
w
in
g
a
s
p
e
ci
al
d
i
et
.
b.
T
y
p
e
2
Dia
b
e
tes
I
n
T
y
p
e
2
D
ia
b
e
tes
,
th
e
b
o
d
y
d
o
es
n
o
t
p
r
o
d
u
c
e
en
o
u
g
h
in
s
u
lin
o
r
th
e
ce
lls
in
th
e
b
o
d
y
d
is
p
l
ay
in
s
u
lin
r
esis
tan
c
e.
So
m
e
p
e
o
p
le
m
a
y
b
e
a
b
l
e
t
o
c
o
n
tr
o
l
th
e
ir
ty
p
e
2
d
i
ab
etes
s
y
m
p
to
m
s
b
y
lo
s
in
g
w
eig
h
t,
f
o
ll
o
w
in
g
a
h
ea
lth
y
d
iet
,
d
o
in
g
p
l
en
ty
o
f
ex
er
cise
,
an
d
m
o
n
ito
r
in
g
th
eir
b
l
o
o
d
g
lu
c
o
s
e
lev
els.
H
o
w
ev
er
,
ty
p
e
2
d
ia
b
etes
is
ty
p
ical
ly
a
p
r
o
g
r
es
s
iv
e
d
is
e
ase
–
it
g
r
a
d
u
a
lly
g
ets
w
o
r
s
e
–
an
d
th
e
p
a
tien
t
w
ill
p
r
o
b
a
b
ly
en
d
u
p
h
av
in
g
to
tak
e
in
s
u
lin
,
u
s
u
al
ly
in
tab
let
f
o
r
m
.
B
ein
g
o
v
er
w
eig
h
t,
p
h
y
s
ically
in
ac
tiv
e
an
d
ea
tin
g
th
e
w
r
o
n
g
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
A
C
o
mp
a
r
a
tive
A
n
a
lysi
s
o
n
th
e
E
va
lu
a
tio
n
o
f
C
la
s
s
ifica
tio
n
A
lg
o
r
ith
ms in
th
e
P
r
ed
ictio
n
.
..
(
R
a
tn
a
P
a
til)
3967
f
o
o
d
s
all
c
o
n
t
r
i
b
u
te
t
o
o
u
r
r
is
k
o
f
d
ev
el
o
p
in
g
ty
p
e
2
d
ia
b
etes
.
T
h
e
r
is
k
o
f
d
ev
el
o
p
in
g
T
y
p
e
2
d
i
ab
et
es
als
o
in
cr
ea
s
es
w
ith
ag
e
[
3
]
,
[
4
]
.
c.
Gesta
ti
o
n
al
Di
ab
etes
T
h
is
ty
p
e
af
f
ec
ts
f
em
ales
d
u
r
i
n
g
p
r
eg
n
an
cy
.
So
m
e
w
o
m
en
h
av
e
v
er
y
h
ig
h
lev
els
o
f
g
lu
co
s
e
in
th
ei
r
b
l
o
o
d
,
an
d
th
ei
r
b
o
d
i
es
a
r
e
u
n
ab
le
t
o
p
r
o
d
u
ce
en
o
u
g
h
in
s
u
lin
to
tr
an
s
p
o
r
t
a
ll
o
f
th
e
g
lu
c
o
s
e
in
to
th
ei
r
c
ells
,
r
esu
lt
in
g
in
p
r
o
g
r
ess
iv
e
ly
r
is
in
g
lev
els
o
f
g
lu
c
o
s
e
.
T
h
e
m
ajo
r
ity
o
f
g
estati
o
n
al
d
i
ab
etes
p
a
ti
en
ts
ca
n
c
o
n
t
r
o
l
th
eir
d
ia
b
e
tes
w
ith
ex
er
c
is
e
an
d
d
i
et
.
B
e
tw
ee
n
1
0
%
t
o
2
0
%
o
f
th
em
w
ill
n
ee
d
t
o
t
ak
e
s
o
m
e
k
in
d
o
f
b
l
o
o
d
-
g
lu
co
s
e
-
c
o
n
t
r
o
llin
g
m
ed
ic
ati
o
n
s
.
Un
d
iag
n
o
s
e
d
o
r
u
n
co
n
t
r
o
l
led
g
es
tat
io
n
a
l
d
ia
b
et
es
ca
n
r
aise
th
e
r
is
k
o
f
co
m
p
lic
ati
o
n
s
d
u
r
in
g
ch
i
ld
b
i
r
t
h
.
2.
P
RO
CE
SS
WO
RK
F
L
O
W
Fig
u
r
e
1
s
h
o
w
s
t
h
e
p
r
o
ce
s
s
o
f
co
n
ce
p
tu
al
f
r
a
m
e
w
o
r
k
.
Fig
u
r
e
1
.
T
h
e
p
r
o
ce
s
s
o
f
co
n
c
ep
tu
al
f
r
a
m
e
w
o
r
k
3.
M
O
DE
L
CO
NS
T
RUC
T
I
O
N
Mo
d
el
C
o
n
s
tr
u
ct
io
n
w
ill
ta
k
e
p
lace
u
s
in
g
L
o
g
i
s
tic
R
e
g
r
es
s
i
o
n
,
K
Nea
r
est
Neig
h
b
o
r
s
(
KNN)
,
SVM,
Gr
ad
ien
t
B
o
o
s
t,
Dec
is
io
n
tr
ee
,
ML
P
,
R
an
d
o
m
Fo
r
est
an
d
G
au
s
s
ian
Naïv
e
B
a
y
es
a
n
d
th
ei
r
p
er
f
o
r
m
a
n
ce
w
il
l
b
e
ev
alu
ated
[
5
]
,
[
6
]
.
3
.
1
.
L
o
g
is
t
ic
r
eg
re
s
s
io
n
L
o
g
i
s
tic
r
eg
r
es
s
io
n
i
s
b
asical
l
y
a
lin
ea
r
m
o
d
el
f
o
r
class
i
f
i
ca
tio
n
r
ath
er
th
a
n
r
eg
r
ess
io
n
.
I
t
is
also
k
n
o
w
n
a
s
t
h
e
lo
g
it
r
eg
r
es
s
io
n
,
m
a
x
i
m
u
m
-
e
n
tr
o
p
y
cla
s
s
i
f
ica
tio
n
(
Ma
x
E
n
t)
o
r
th
e
lo
g
-
li
n
e
ar
clas
s
i
f
ier
.
I
n
t
h
i
s
m
o
d
el,
w
e
u
s
e
lo
g
is
tic
r
eg
r
es
s
io
n
to
m
o
d
el
p
r
o
b
ab
ilis
ticall
y
d
escr
ib
ed
o
u
tco
m
e
s
o
f
a
s
in
g
l
e
tr
ial.
I
t
is
a
b
asic
m
o
d
el
w
h
ich
d
escr
ib
es
d
ich
o
to
m
o
u
s
o
u
tp
u
t
v
ar
iab
les
a
n
d
ca
n
b
e
e
x
te
n
d
ed
f
o
r
d
i
s
ea
s
e
cla
s
s
i
f
icatio
n
p
r
ed
ictio
n
[
7
]
,
[
8
]
.
Su
p
p
o
s
e
th
er
e
ar
e
N
in
p
u
t
v
ar
iab
les
w
h
er
e
th
eir
v
al
u
es
ar
e
in
d
icate
d
b
y
m
1
,
m
2
,
m
3
,
…,
m
N.
L
et
u
s
ass
u
m
e
th
a
t
th
e
P
p
r
o
b
ab
ilit
y
o
f
t
h
at
an
ev
e
n
t
w
il
l
o
cc
u
r
an
d
1
-
P
b
e
a
p
r
o
b
ab
ilit
y
th
at
ev
e
n
t
w
ill
n
o
t
o
cc
u
r
.
L
o
g
i
s
tic
r
eg
r
es
s
io
n
m
o
d
el
is
g
i
v
en
b
y
(
)
(
)
(
1
)
3
.
2
.
K
NN
K
n
ea
r
est
n
ei
g
h
b
o
r
s
is
a
s
i
m
p
le
alg
o
r
ith
m
t
h
at
s
to
r
es
all
av
ailab
le
ca
s
e
s
a
n
d
clas
s
if
ies
n
e
w
ca
s
es
b
ased
o
n
a
s
i
m
ilar
i
t
y
m
ea
s
u
r
e
(
e.
g
.
,
d
is
tan
ce
f
u
n
ctio
n
s
)
.
C
as
e
is
cla
s
s
i
f
ied
b
y
a
m
aj
o
r
ity
v
o
te
o
f
its
n
ei
g
h
b
o
r
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.
8
,
No
.
5
,
Octo
b
er
2
0
1
8
:
3
9
6
6
–
3
9
7
5
3968
w
it
h
t
h
e
ca
s
e
b
ein
g
ass
ig
n
ed
to
th
ec
lass
m
o
s
t
co
m
m
o
n
a
m
o
n
g
s
t
i
ts
K
n
ea
r
es
t
n
ei
g
h
b
o
r
s
m
ea
s
u
r
ed
b
y
a
d
is
tan
ce
f
u
n
ctio
n
[
9
]
.
(
)
√
(
)
(
)
(
)
(
2
)
I
f
K
=
1
,
th
e
n
t
h
e
ca
s
e
is
s
i
m
p
ly
ass
ig
n
ed
to
th
e
c
lass
o
f
it
s
n
ea
r
est
n
ei
g
h
b
o
r
.
Si
m
ilar
it
y
is
d
ef
i
n
e
d
ac
co
r
d
in
g
to
a
d
is
ta
n
ce
m
etr
i
c
b
et
w
ee
n
t
w
o
d
ata
p
o
in
t
s
.
A
p
o
p
u
lar
ch
o
ice
is
t
h
e
E
u
clid
ea
n
d
is
ta
n
ce
.
Mo
r
e
f
o
r
m
all
y
,
g
i
v
en
a
p
o
s
iti
v
e
i
n
t
eg
er
K,
a
n
u
n
s
ee
n
o
b
s
er
v
at
i
o
n
x
an
d
a
s
i
m
ilar
i
t
y
m
etr
ic
d
,
KNN
cla
s
s
i
f
ier
p
er
f
o
r
m
s
t
h
e
f
o
llo
w
i
n
g
t
w
o
s
t
ep
s
:
a.
I
t
r
u
n
s
t
h
r
o
u
g
h
t
h
e
w
h
o
le
d
ata
s
et
co
m
p
u
t
in
g
d
b
et
w
ee
n
x
a
n
d
ea
ch
tr
ain
in
g
o
b
s
er
v
atio
n
.
W
e’
ll
ca
ll
th
e
K
p
o
in
ts
in
t
h
e
tr
ai
n
in
g
d
ata
th
at
ar
e
clo
s
est
to
x
th
e
s
e
t
A.
No
te
th
at
K
is
u
s
u
a
ll
y
o
d
d
to
p
r
ev
en
t
tie
s
itu
a
tio
n
s
.
b.
I
t
th
e
n
e
s
ti
m
ates
t
h
e
co
n
d
itio
n
al
p
r
o
b
ab
il
it
y
f
o
r
ea
c
h
cla
s
s
,
th
at
is
,
t
h
e
f
r
ac
tio
n
o
f
p
o
in
ts
in
A
w
it
h
t
h
a
t
g
iv
e
n
c
lass
lab
el.
(
No
te
I(
x
)
is
th
e
i
n
d
ica
to
r
f
u
n
ctio
n
w
h
ich
e
v
alu
a
tes to
1
w
h
e
n
t
h
e
ar
g
u
m
e
n
t
x
i
s
tr
u
e
an
d
0
o
th
er
w
i
s
e)
(
|
)
∑
(
(
)
)
(
3
)
Fin
all
y
,
o
u
r
i
n
p
u
t
x
g
et
s
ass
i
g
n
ed
to
th
e
class
w
i
th
t
h
e
lar
g
es
t p
r
o
b
a
b
ilit
y
.
3
.
3
.
SVM
A
Su
p
p
o
r
t
Vec
to
r
Ma
ch
i
n
e
(
SVM)
is
a
d
i
s
cr
i
m
in
at
iv
e
c
la
s
s
i
f
ier
f
o
r
m
all
y
d
ef
in
ed
b
y
a
s
ep
ar
atin
g
h
y
p
er
p
lan
e
[
1
0
]
.
I
n
t
h
e
li
n
ea
r
class
if
ier
m
o
d
el,
w
e
as
s
u
m
e
d
th
at
tr
ain
i
n
g
e
x
a
m
p
les
p
lo
tt
ed
in
s
p
ac
e.
T
h
ese
d
ata
p
o
in
ts
ar
e
e
x
p
ec
ted
to
b
e
s
ep
ar
ated
b
y
an
ap
p
ar
en
t
g
ap
.
I
t
p
r
ed
icts
a
s
tr
aig
h
t
h
y
p
er
p
lan
e
d
iv
id
i
n
g
2
class
es.
T
h
e
p
r
i
m
ar
y
f
o
cu
s
wh
ile
d
r
a
w
i
n
g
t
h
e
h
y
p
er
p
lan
e
i
s
o
n
m
a
x
i
m
izi
n
g
t
h
e
d
i
s
ta
n
ce
f
r
o
m
h
y
p
er
p
la
n
e
to
th
e
n
ea
r
est
d
ata
p
o
in
t
o
f
eit
h
e
r
class
.
T
h
e
d
r
a
w
n
h
y
p
er
p
la
n
e
ca
lled
as
a
m
a
x
i
m
u
m
-
m
ar
g
i
n
h
y
p
er
p
lan
e
[
1
1
]
.
T
h
e
class
if
ica
t
io
n
p
r
o
ce
s
s
o
f
S
VM
class
i
f
ier
.
Fi
g
u
r
e
2
s
h
o
w
s
th
e
SVM
h
y
p
er
p
lan
es
.
Fig
u
r
e
2
.
SVM
h
y
p
er
p
lan
e
s
⃗
⃗
⃗
⃗
(
4
)
W
h
er
e
‖
⃗
⃗
‖
is
n
o
r
m
al
v
ec
to
r
to
th
e
h
y
p
er
p
la
n
e,
d
en
o
tes
class
es
an
d
d
en
o
tes
f
ea
t
u
r
es.
T
h
e
Dis
t
an
c
e
b
et
w
ee
n
t
w
o
h
y
p
er
p
lan
e
s
is
‖
⃗
⃗
‖
,
to
m
a
x
i
m
ize
th
is
d
is
ta
n
ce
d
en
o
m
in
ato
r
v
al
u
e
s
h
o
u
ld
b
e
m
in
i
m
ized
i.e
,
‖
⃗
⃗
‖
s
h
o
u
ld
b
e
m
in
i
m
ized
.
Fo
r
p
r
o
p
er
class
if
icatio
n
,
w
e
ca
n
b
u
ild
a
co
m
b
i
n
ed
eq
u
atio
n
:
‖
⃗
⃗
‖
(
⃗
⃗
)
(
5
)
3
.
4
.
G
ra
dient
b
o
o
s
t
B
o
o
s
tin
g
r
e
f
er
s
to
a
f
a
m
i
l
y
o
f
alg
o
r
it
h
m
s
t
h
at
ar
e
ab
le
to
c
o
n
v
er
t
w
ea
k
lear
n
er
s
to
s
tr
o
n
g
lear
n
er
s
.
T
h
e
m
ai
n
p
r
in
cip
le
o
f
b
o
o
s
ti
n
g
i
s
to
f
it
a
s
eq
u
e
n
ce
o
f
w
ea
k
lear
n
er
s
−
m
o
d
els
t
h
at
ar
e
o
n
l
y
s
lig
h
tl
y
b
etter
th
a
n
r
an
d
o
m
g
u
e
s
s
i
n
g
,
s
u
c
h
as
s
m
all
d
ec
is
io
n
tr
ee
s
−
to
w
eig
h
te
d
v
er
s
io
n
s
o
f
t
h
e
d
ata.
Mo
r
e
w
eig
h
t
is
g
i
v
e
n
to
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
A
C
o
mp
a
r
a
tive
A
n
a
lysi
s
o
n
th
e
E
va
lu
a
tio
n
o
f
C
la
s
s
ifica
tio
n
A
lg
o
r
ith
ms in
th
e
P
r
ed
ictio
n
.
..
(
R
a
tn
a
P
a
til)
3969
ex
a
m
p
le
s
t
h
at
w
er
e
m
is
cla
s
s
i
f
ied
b
y
ea
r
lier
r
o
u
n
d
s
.
T
h
e
p
r
ed
ictio
n
s
ar
e
th
e
n
co
m
b
i
n
ed
t
h
r
o
u
g
h
a
w
ei
g
h
ted
m
aj
o
r
ity
v
o
te
(
class
i
f
icatio
n
)
o
r
a
w
ei
g
h
ted
s
u
m
(
r
e
g
r
ess
io
n
)
to
p
r
o
d
u
ce
th
e
f
i
n
al
p
r
ed
ictio
n
.
Gr
ad
ien
t
T
r
ee
B
o
o
s
tin
g
s
a
g
en
er
al
izatio
n
o
f
b
o
o
s
tin
g
t
o
ar
b
itra
r
y
d
if
f
e
r
en
tiab
le
lo
s
s
f
u
n
ctio
n
s
.
I
t
ca
n
b
e
u
s
ed
f
o
r
b
o
th
r
eg
r
ess
io
n
an
d
clas
s
if
icatio
n
p
r
o
b
lem
s
.
Gr
ad
ie
n
t B
o
o
s
tin
g
b
u
ild
s
t
h
e
m
o
d
el
i
n
a
s
eq
u
e
n
tial
w
a
y
.
(
)
(
)
(
)
(
6
)
A
t e
ac
h
s
ta
g
e
t
h
e
d
ec
is
io
n
tr
ee
h
m
(
x
)
is
c
h
o
s
en
to
m
i
n
i
m
ize
a
lo
s
s
f
u
n
c
tio
n
L
g
i
v
e
n
th
e
c
u
r
r
en
t
m
o
d
el
F
m
-
1
(
x
)
:
(
)
(
)
∑
(
(
)
(
)
)
(
7
)
T
h
e
alg
o
r
ith
m
s
f
o
r
r
eg
r
es
s
io
n
an
d
class
i
f
icat
io
n
d
if
f
er
in
t
h
e
t
y
p
e
o
f
lo
s
s
f
u
n
ctio
n
u
s
ed
.
3
.
5
.
Dec
is
io
n t
ree
Dec
is
io
n
tr
ee
i
s
a
s
i
m
p
le,
d
eter
m
i
n
is
tic
d
ata
s
tr
u
ct
u
r
e
f
o
r
m
o
d
ell
in
g
d
ec
is
io
n
r
u
le
s
f
o
r
a
s
p
ec
if
i
c
class
i
f
icatio
n
p
r
o
b
le
m
.
A
t
ea
ch
n
o
d
e,
o
n
e
f
ea
t
u
r
e
is
s
ele
cted
to
m
ak
e
s
ep
ar
atin
g
d
ec
i
s
io
n
.
W
e
ca
n
s
to
p
s
p
litt
i
n
g
o
n
ce
th
e
lea
f
n
o
d
e
h
as
o
p
tim
a
ll
y
les
s
d
ata
p
o
in
ts
.
Su
ch
lea
f
n
o
d
e
th
en
g
iv
e
s
u
s
i
n
s
ig
h
t
in
to
th
e
f
i
n
al
r
esu
lt
(
P
r
o
b
ab
ilit
ies
f
o
r
d
if
f
er
en
t
class
e
s
in
ca
s
e
o
f
class
i
f
ic
atio
n
)
.
T
h
e
m
o
s
t
d
ec
is
i
v
e
f
ac
t
o
r
f
o
r
th
e
ef
f
icie
n
c
y
o
f
a
d
ec
is
io
n
tr
ee
is
t
h
e
e
f
f
i
cie
n
c
y
o
f
it
s
s
p
litt
in
g
p
r
o
ce
s
s
as
s
h
o
w
n
i
n
Fi
g
u
r
e
3.
W
e
s
p
lit
at
ea
ch
n
o
d
e
i
n
s
u
c
h
a
w
a
y
t
h
at
t
h
e
r
esu
lt
in
g
p
u
r
it
y
is
m
ax
i
m
u
m
.
W
ell,
p
u
r
it
y
j
u
s
t
r
ef
er
s
to
h
o
w
w
ell
w
e
ca
n
s
e
g
r
eg
ate
t
h
e
class
e
s
an
d
in
cr
ea
s
e
o
u
r
k
n
o
w
led
g
e
b
y
th
e
s
p
lit p
er
f
o
r
m
ed
[
1
2
]
.
Fig
u
r
e
3
.
Dec
is
io
n
t
r
ee
3
.
6
.
M
L
P
T
h
e
Mu
ltil
a
y
er
P
er
ce
p
tio
n
(
M
L
P
)
is
p
er
h
ap
s
th
e
m
o
s
t p
o
p
u
l
ar
n
et
w
o
r
k
ar
ch
itec
tu
r
e
i
n
u
s
e
to
d
ay
b
o
t
h
f
o
r
clas
s
if
icatio
n
an
d
r
e
g
r
ess
i
o
n
.
M
L
P
s
ar
e
f
ee
d
f
o
r
w
ar
d
n
eu
r
al
n
et
w
o
r
k
s
w
h
ich
ar
e
t
y
p
icall
y
co
m
p
o
s
ed
o
f
s
ev
er
al
la
y
er
s
o
f
n
o
d
es
w
it
h
u
n
id
ir
ec
tio
n
al
co
n
n
ec
tio
n
s
,
o
f
ten
tr
ain
ed
b
y
b
ac
k
p
r
o
p
ag
atio
n
[
1
3
]
,
[
1
4
]
.
T
h
e
lear
n
in
g
p
r
o
ce
s
s
o
f
ML
P
n
et
wo
r
k
is
b
ased
o
n
th
e
d
ata
s
am
p
l
es
co
m
p
o
s
ed
o
f
th
e
N
-
d
i
m
e
n
s
i
o
n
al
in
p
u
t
v
ec
to
r
x
an
d
th
e
M
-
d
i
m
e
n
s
io
n
al
d
esire
d
o
u
tp
u
t
v
ec
to
r
d
,
ca
lled
d
esti
n
atio
n
.
B
y
p
r
o
ce
s
s
in
g
t
h
e
i
n
p
u
t
v
ec
to
r
x
,
t
h
e
M
L
P
p
r
o
d
u
ce
s
th
e
o
u
tp
u
t
s
ig
n
al
v
ec
to
r
y
(
x
,
w
)
w
h
er
e
w
is
t
h
e
v
ec
t
o
r
o
f
ad
ap
ted
w
eig
h
t
s
.
T
h
e
er
r
o
r
s
ig
n
al
p
r
o
d
u
ce
d
ac
tu
ates
a
co
n
tr
o
l
m
ec
h
a
n
is
m
o
f
th
e
lear
n
i
n
g
a
lg
o
r
it
h
m
.
T
h
e
co
r
r
ec
tiv
e
ad
j
u
s
t
m
e
n
ts
ar
e
d
e
s
ig
n
ed
to
m
a
k
e
t
h
e
o
u
tp
u
t
s
ig
n
al
y
k
(
k
=
1
,
2
,
…
,
M
)
to
th
e
d
esire
d
r
esp
o
n
s
e
d
k
i
n
a
s
tep
b
y
s
tep
m
an
n
er
.
I
f
a
m
u
ltil
a
y
er
p
er
ce
p
tr
o
n
h
as
a
li
n
ea
r
ac
ti
v
atio
n
f
u
n
ctio
n
i
n
all
n
eu
r
o
n
s
,
t
h
at
i
s
,
a
l
in
e
ar
f
u
n
ctio
n
t
h
at
m
ap
s
t
h
e
w
ei
g
h
ted
in
p
u
ts
to
t
h
e
o
u
tp
u
t
o
f
ea
c
h
n
eu
r
o
n
,
th
e
n
l
in
ea
r
alg
eb
r
a
s
h
o
w
s
t
h
at
a
n
y
n
u
m
b
er
o
f
la
y
er
s
ca
n
b
e
r
ed
u
ce
d
to
a
t
w
o
-
la
y
er
in
p
u
t
-
o
u
tp
u
t
m
o
d
el.
I
n
ML
P
s
s
o
m
e
n
e
u
r
o
n
s
u
s
e
a
n
o
n
li
n
ea
r
ac
tiv
atio
n
f
u
n
ctio
n
t
h
at
w
as
d
ev
elo
p
ed
to
m
o
d
el
th
e
f
r
eq
u
e
n
c
y
o
f
ac
tio
n
p
o
ten
t
ials
,
o
r
f
ir
in
g
,
o
f
b
io
lo
g
ical
n
e
u
r
o
n
s
[
1
5
]
.
T
h
e
t
w
o
co
m
m
o
n
ac
tiv
atio
n
f
u
n
ctio
n
s
ar
e
b
o
th
s
ig
m
o
id
s
,
an
d
ar
e
d
escr
ib
ed
b
y
(
)
(
)
(
)
(
)
(
8
)
T
h
e
f
ir
s
t
is
a
h
y
p
er
b
o
lic
ta
n
g
en
t
t
h
at
r
an
g
es
f
r
o
m
-
1
to
1
,
w
h
ile
th
e
o
th
er
is
th
e
lo
g
i
s
tic
f
u
n
ctio
n
,
w
h
ic
h
i
s
s
i
m
ilar
in
s
h
a
p
e
b
u
t
r
an
g
e
s
f
r
o
m
0
to
1
.
Her
e
y
i
is
th
e
o
u
tp
u
t
o
f
t
h
e
i
th
n
o
d
e
(
n
eu
r
o
n
)
an
d
v
i
i
s
t
h
e
w
ei
g
h
ted
s
u
m
o
f
t
h
e
i
n
p
u
t
co
n
n
ec
t
io
n
s
.
T
h
e
lear
n
in
g
al
g
o
r
ith
m
o
f
M
L
P
is
b
ased
o
n
t
h
e
m
i
n
i
m
izatio
n
o
f
t
h
e
er
r
o
r
f
u
n
ctio
n
d
e
f
i
n
ed
o
n
th
e
l
ea
r
n
in
g
s
et
(
x
i
,
d
i
)
f
o
r
i
=1
,
2
,
…
,
N
u
s
in
g
t
h
e
E
u
clid
ea
n
n
o
r
m
:
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.
8
,
No
.
5
,
Octo
b
er
2
0
1
8
:
3
9
6
6
–
3
9
7
5
3970
(
)
∑
‖
(
)
‖
(
9
)
T
h
e
m
i
n
i
m
izat
io
n
o
f
t
h
i
s
er
r
o
r
lead
s
to
th
e
o
p
ti
m
al
v
al
u
es
o
f
w
ei
g
h
ts
.
T
h
e
m
o
s
t
ef
f
ec
ti
v
e
m
et
h
o
d
s
o
f
m
i
n
i
m
izatio
n
ar
e
th
e
g
r
ad
ie
n
t
alg
o
r
it
h
m
s
,
f
r
o
m
w
h
ich
t
h
e
m
o
s
t
e
f
f
ec
t
iv
e
is
t
h
e
L
e
v
en
b
er
g
–
Ma
r
q
u
ar
d
alg
o
r
ith
m
f
o
r
m
ed
i
u
m
s
ize
n
et
w
o
r
k
s
a
n
d
co
n
j
u
g
ate
g
r
ad
ien
t
f
o
r
lar
g
e
s
ize
n
e
t
w
o
r
k
s
.
Fi
g
u
r
e
4
s
h
o
w
s
th
e
M
L
P
s
tr
u
ct
u
r
e
.
Fig
u
r
e
4
.
ML
P
s
tr
u
ct
u
r
e
3
.
7
.
Ra
nd
o
m
f
o
re
s
t
R
an
d
o
m
f
o
r
est
is
j
u
s
t
a
n
i
m
p
r
o
v
e
m
e
n
t
o
v
er
th
e
to
p
o
f
th
e
d
ec
is
io
n
tr
ee
alg
o
r
ith
m
.
T
h
e
co
r
e
id
ea
b
eh
in
d
R
a
n
d
o
m
Fo
r
est is
to
g
e
n
er
ate
m
u
lt
ip
le
s
m
all
d
ec
is
io
n
tr
ee
s
f
r
o
m
r
a
n
d
o
m
s
u
b
s
et
s
o
f
th
e
d
ata
(
h
e
n
ce
t
h
e
n
a
m
e
“R
an
d
o
m
Fo
r
est”)
.
E
ac
h
o
f
th
e
d
ec
is
io
n
tr
ee
g
iv
e
s
a
b
iased
class
if
ier
(
as
it
o
n
l
y
co
n
s
id
er
s
a
s
u
b
s
et
o
f
th
e
d
ata)
.
T
h
ey
ea
c
h
ca
p
tu
r
e
d
if
f
er
en
t tr
e
n
d
s
in
t
h
e
d
ata
as s
h
o
w
n
in
F
ig
u
r
e
5
.
Fig
u
r
e
5
.
R
an
d
o
m
f
o
r
est
T
h
is
en
s
e
m
b
le
o
f
tr
ee
s
is
li
k
e
a
tea
m
o
f
e
x
p
er
ts
ea
c
h
w
it
h
a
litt
le
k
n
o
w
led
g
e
o
v
er
th
e
o
v
er
all
s
u
b
j
ec
t
b
u
t
th
o
r
o
u
g
h
i
n
t
h
eir
ar
ea
o
f
ex
p
er
tis
e.
No
w
,
i
n
ca
s
e
o
f
class
i
f
icatio
n
t
h
e
m
aj
o
r
ity
v
o
te
is
co
n
s
id
er
ed
to
class
i
f
y
a
cla
s
s
.
I
n
an
a
lo
g
y
w
it
h
e
x
p
er
ts
,
it
i
s
li
k
e
a
s
k
i
n
g
t
h
e
s
a
m
e
m
u
ltip
le
c
h
o
ice
q
u
esti
o
n
to
ea
ch
e
x
p
er
t a
n
d
tak
i
n
g
t
h
e
an
s
w
er
as
th
e
o
n
e
t
h
at
m
o
s
t
n
o
.
o
f
ex
p
er
ts
v
o
te
as
co
r
r
ec
t.
I
n
ca
s
e
o
f
R
eg
r
es
s
i
o
n
,
w
e
ca
n
u
s
e
th
e
av
g
.
o
f
al
l
tr
ee
s
as
o
u
r
p
r
ed
ictio
n
.
I
n
ad
d
itio
n
to
th
is
,
w
e
ca
n
also
w
e
ig
h
t
s
o
m
e
m
o
r
e
d
ec
is
iv
e
tr
ee
s
h
ig
h
r
elativ
e
to
o
th
er
s
b
y
test
in
g
o
n
th
e
v
alid
atio
n
d
ata
[
1
6
]
.
M
aj
o
r
ity
v
o
te
is
ta
k
en
f
r
o
m
t
h
e
ex
p
er
ts
(
tr
ee
s
)
f
o
r
class
i
f
icatio
n
.
3
.
8
.
G
a
us
s
ia
n n
a
ïv
e
b
a
y
es
I
n
Gau
s
s
ia
n
Nai
v
e
B
a
y
es,
c
o
n
tin
u
o
u
s
v
al
u
es
a
s
s
o
ciate
d
w
it
h
ea
c
h
f
ea
t
u
r
e
ar
e
ass
u
m
ed
t
o
b
e
d
is
tr
ib
u
ted
ac
co
r
d
in
g
to
a
Gau
s
s
ia
n
d
is
tr
ib
u
tio
n
[
1
7
]
.
A
Ga
u
s
s
ian
d
is
tr
ib
u
tio
n
i
s
also
ca
lled
No
r
m
a
l
d
is
tr
ib
u
tio
n
.
W
h
e
n
p
lo
tted
,
it
g
iv
e
s
a
b
ell
s
h
ap
ed
cu
r
v
e
wh
ich
i
s
s
y
m
m
e
tr
ic
ab
o
u
t
th
e
m
ea
n
o
f
t
h
e
f
ea
t
u
r
e
v
alu
e
s
as s
h
o
w
n
in
F
ig
u
r
e
6
.
T
h
e
lik
elih
o
o
d
o
f
th
e
f
ea
t
u
r
es
is
ass
u
m
ed
to
b
e
Gau
s
s
ia
n
,
h
e
n
ce
,
co
n
d
itio
n
al
p
r
o
b
ab
ilit
y
is
g
iv
e
n
b
y
:
(
|
)
√
(
(
)
)
(
1
0
)
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
A
C
o
mp
a
r
a
tive
A
n
a
lysi
s
o
n
th
e
E
va
lu
a
tio
n
o
f
C
la
s
s
ifica
tio
n
A
lg
o
r
ith
ms in
th
e
P
r
ed
ictio
n
.
..
(
R
a
tn
a
P
a
til)
3971
Fig
u
r
e
6
.
Gau
s
s
ian
c
u
r
v
e
4.
P
E
RF
O
RM
ANCE E
VA
L
U
AT
I
O
N
CR
I
T
E
R
I
A
F
O
R
M
O
DE
L
T
o
an
aly
ze
a
n
d
co
m
p
ar
e
t
h
e
p
er
f
o
r
m
a
n
ce
o
f
t
h
e
d
ata
m
in
i
n
g
m
et
h
o
d
s
p
r
ese
n
ted
in
o
u
r
s
tu
d
y
,
w
e
ap
p
ly
v
ar
io
u
s
s
tatis
tics
s
u
c
h
as
MA
E
,
R
M
SE,
NR
M
SE
an
d
C
o
n
f
u
s
io
n
Ma
tr
ix
co
m
p
u
ted
as
f
o
llo
w
s
[
1
8
]
-
[
2
0
]
.
a.
Me
an
ab
s
o
lu
te
e
r
r
o
r
(
MA
E
)
MA
E
m
ea
s
u
r
es
th
e
a
v
er
ag
e
m
ag
n
it
u
d
e
o
f
th
e
er
r
o
r
s
in
a
s
et
o
f
p
r
ed
ictio
n
s
,
w
it
h
o
u
t
c
o
n
s
id
er
in
g
th
eir
d
ir
ec
tio
n
.
I
t’
s
th
e
av
er
a
g
e
o
v
e
r
th
e
test
s
a
m
p
le
o
f
t
h
e
ab
s
o
lu
te
d
if
f
er
en
ce
s
b
et
w
ee
n
p
r
ed
ictio
n
an
d
ac
tu
al
o
b
s
er
v
atio
n
w
h
er
e
all
in
d
i
v
id
u
al
d
if
f
er
e
n
ce
s
h
av
e
eq
u
a
l
w
eig
h
t.
∑
|
̂
|
(
1
1
)
b.
R
o
o
t
m
ea
n
s
q
u
ar
e
e
r
r
o
r
(
R
MSE
)
R
MSE
is
a
q
u
ad
r
a
tic
s
c
o
r
in
g
r
u
le
th
at
als
o
m
ea
s
u
r
es
th
e
av
e
r
ag
e
m
ag
n
itu
d
e
o
f
th
e
e
r
r
o
r
.
I
t’
s
th
e
s
q
u
ar
e
r
o
o
t
o
f
th
e
av
er
ag
e
o
f
s
q
u
a
r
e
d
d
if
f
e
r
en
c
es
b
e
tw
ee
n
p
r
e
d
ic
ti
o
n
an
d
ac
tu
a
l
o
b
s
e
r
v
ati
o
n
.
√
∑
(
̂
)
(
1
2
)
c.
C
o
n
f
u
s
io
n
m
atr
i
x
T
h
e
in
f
o
r
m
atio
n
ab
o
u
t
ac
t
u
al
an
d
p
r
ed
icted
clas
s
if
icatio
n
s
y
s
te
m
is
h
o
ld
b
y
t
h
e
C
o
n
f
u
s
io
n
m
atr
i
x
.
I
t
d
em
o
n
s
tr
ate
s
th
e
ac
cu
r
ac
y
o
f
th
e
s
o
lu
tio
n
to
a
class
if
icat
io
n
p
r
o
b
lem
.
T
ab
le
1
s
h
o
w
s
th
e
c
o
n
f
u
s
io
n
m
atr
i
x
f
o
r
a
t
w
o
cla
s
s
clas
s
i
f
ier
.
T
h
e
en
tr
ies
in
t
h
e
co
n
f
u
s
io
n
m
atr
ix
h
a
v
e
t
h
e
f
o
llo
w
i
n
g
m
ea
n
i
n
g
i
n
t
h
e
co
n
tex
t o
f
o
u
r
s
t
u
d
y
.
T
p
is
th
e
n
u
m
b
er
o
f
co
r
r
ec
t
p
r
ed
ictio
n
s
t
h
at
a
n
i
n
s
ta
n
ce
i
s
p
o
s
iti
v
e.
F
n
i
s
t
h
e
n
u
m
b
er
o
f
in
co
r
r
ec
t
p
r
ed
ictio
n
s
t
h
at
a
n
in
s
tan
ce
is
n
eg
a
tiv
e.
F
p
is
th
e
n
u
m
b
er
o
f
i
n
co
r
r
ec
t
p
r
ed
ictio
n
s
t
h
at
a
n
in
s
ta
n
ce
is
p
o
s
it
iv
e
a
n
d
i
s
th
e
n
u
m
b
er
o
f
co
r
r
ec
t p
r
ed
ictio
n
s
th
at
an
i
n
s
tan
ce
i
s
n
e
g
ati
v
e.
T
ab
le
1
.
T
h
e
C
o
n
f
u
s
io
n
M
atr
i
x
f
o
r
a
t
w
o
cla
s
s
C
las
s
i
f
ier
P
r
e
d
i
c
t
e
d
P
o
si
t
i
v
e
N
e
g
a
t
i
v
e
A
c
t
u
a
l
P
o
si
t
i
v
e
N
e
g
a
t
i
v
e
d.
P
r
ec
is
io
n
P
r
ec
is
io
n
lo
o
k
s
at
th
e
r
atio
o
f
co
r
r
ec
t p
o
s
itiv
e
o
b
s
er
v
atio
n
s
.
T
h
e
f
o
r
m
u
la
i
s
,
(
1
3
)
e.
R
ec
all
/
tr
u
e
p
o
s
itiv
e
r
ate
/
s
en
s
it
iv
it
y
R
ec
all
is
al
s
o
k
n
o
w
n
as
s
en
s
iti
v
it
y
o
r
tr
u
e
p
o
s
iti
v
e
r
ate.
I
t’
s
t
h
e
r
atio
o
f
co
r
r
ec
tly
p
r
ed
icted
p
o
s
itiv
e
ev
e
n
t
s
.
(
1
4
)
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.
8
,
No
.
5
,
Octo
b
er
2
0
1
8
:
3
9
6
6
–
3
9
7
5
3972
f.
A
cc
u
r
ac
y
T
h
e
p
r
o
p
o
r
tio
n
o
f
th
e
to
tal
n
u
m
b
er
o
f
p
r
ed
ictio
n
s
th
at
w
er
e
co
r
r
ec
t
is
k
n
o
w
n
to
b
e
as
A
c
cu
r
ac
y
(
AC
)
.
I
t
s
h
o
w
s
o
v
er
all
ef
f
ec
ti
v
e
n
ess
o
f
class
i
f
ier
.
I
t is d
eter
m
i
n
ed
u
s
i
n
g
t
h
e
eq
u
atio
n
:
(
1
5
)
g.
R
OC
A
r
ec
ei
v
er
o
p
er
atin
g
c
h
ar
ac
ter
is
tics
(
R
O
C
)
g
r
ap
h
is
a
m
et
h
o
d
f
o
r
co
n
ce
p
tu
al
i
ze
,
o
r
g
an
izin
g
a
n
d
s
elec
ti
n
g
class
i
f
ier
s
o
n
t
h
e
b
asis
o
f
t
h
ei
r
p
er
f
o
r
m
a
n
ce
[
2
1
]
,
[
2
2
]
.
R
OC
g
r
ap
h
s
ar
e
b
i
-
d
i
m
e
n
s
io
n
al
g
r
ap
h
s
w
h
er
e
o
n
th
e
Y
ax
i
s
t
p
r
ate
is
p
lo
tted
an
d
o
n
th
e
X
ax
is
f
p
r
ate
is
p
lo
tt
ed
.
A
R
OC
g
r
ap
h
d
escr
ib
e
r
elativ
e
tr
ad
e
-
o
f
f
s
b
et
w
ee
n
b
en
e
f
its
(
tr
u
e
p
o
s
it
iv
es)
an
d
co
s
ts
(
f
al
s
e
p
o
s
iti
v
es)
[
2
3
]
.
5.
E
XP
E
R
I
M
E
NT
A
L
RE
SUL
T
S AN
D
O
B
S
E
RVA
T
I
O
NS
I
n
E
x
p
er
i
m
e
n
tal
s
tu
d
ie
s
th
e
d
ataset
h
a
v
e
b
ee
n
p
ar
titi
o
n
ed
b
et
w
ee
n
7
0
–
3
0
%
(
5
3
8
–
2
3
0
)
f
o
r
tr
ain
in
g
an
d
test
in
g
p
u
r
p
o
s
e.
T
ab
le
2
s
h
o
w
s
L
o
g
i
s
tic
R
eg
r
e
s
s
io
n
b
ein
g
t
h
e
s
i
m
p
le
s
t
cla
s
s
i
f
ier
h
av
e
p
er
f
o
r
m
ed
w
e
ll
w
it
h
a
n
ac
cu
r
ac
y
o
f
7
9
.
5
4
%,
w
h
ile
h
a
v
i
n
g
r
elativ
e
ab
s
o
l
u
te
er
r
o
r
2
1
.
6
5
%.
Am
o
n
g
t
h
e
ap
p
lied
alg
o
r
ith
m
s
L
o
g
i
s
tic
R
eg
r
es
s
io
n
h
as
h
i
g
h
er
ac
cu
r
ac
y
w
h
ic
h
i
s
q
u
ite
well
an
d
h
a
v
in
g
t
h
e
lo
w
est
R
MSE
v
al
u
e
4
6
.
5
2
%.
T
ab
le
2
s
h
o
w
s
co
m
p
ar
ativ
e
a
n
al
y
s
i
s
o
f
al
g
o
r
ith
m
i
n
ter
m
s
o
f
Me
an
A
b
s
o
l
u
te
E
r
r
o
r
,
R
o
o
t
Me
an
Sq
u
ar
e
E
r
r
o
r
an
d
A
cc
u
r
ac
y
s
co
r
e
[
4
]
.
R
OC
is
p
lo
tted
f
o
r
all
th
e
alg
o
r
it
h
m
s
.
Mo
r
e
th
e
ar
ea
co
v
er
ed
b
etter
is
th
e
clas
s
i
f
ier
.
T
h
ese
m
ea
s
u
r
e
m
en
t
s
ar
e
ta
k
en
b
y
u
s
in
g
Sp
y
d
er
to
o
l
o
n
P
im
a
I
n
d
ian
Diab
etes
Data
s
et
tak
e
n
f
r
o
m
U
C
I
r
ep
o
s
ito
r
y
.
T
h
e
r
es
u
lts
ar
e
s
h
o
w
n
i
n
T
ab
le
2
.
T
h
e
r
esu
lt
s
m
a
y
b
e
i
m
p
r
o
v
ed
b
y
ap
p
l
y
i
n
g
lar
g
e
s
ize
u
p
d
ated
d
ata
s
ets
o
f
r
ea
lis
tic
co
n
te
x
t.
Ho
w
e
v
er
w
e
n
ee
d
to
ap
p
ly
o
th
er
m
ac
h
in
e
lear
n
in
g
al
g
o
r
ith
m
s
u
s
i
n
g
r
ea
l
d
ata
s
et
b
ef
o
r
e
g
en
er
alizi
n
g
th
e
r
es
u
lt
s
.
T
ab
le
2.
Su
m
m
ar
y
o
f
P
r
ed
ictio
n
f
o
r
d
if
f
er
en
t
A
l
g
o
r
ith
m
s
A
l
g
o
r
i
t
h
m
M
A
E
R
M
S
E
A
c
c
u
r
a
c
y
S
c
o
r
e
L
o
g
i
st
i
c
R
e
g
r
e
ssi
o
n
0
.
2
1
6
5
0
.
4
6
5
2
0
.
7
9
5
4
K
N
N
e
i
g
h
b
o
r
s
0
.
2
5
1
1
0
.
5
0
1
1
0
.
7
4
8
9
L
i
n
e
a
r
S
V
M
0
.
3
2
0
3
0
.
5
6
6
0
0
.
6
7
9
7
G
r
a
d
i
e
n
t
B
o
o
st
i
n
g
0
.
2
0
7
8
0
.
4
5
5
8
0
.
7
9
2
2
D
e
c
i
si
o
n
t
r
e
e
0
.
2
6
8
4
0
.
5
1
8
1
0
.
7
3
1
6
M
L
P
0
.
3
5
9
3
0
.
5
9
9
4
0
.
6
4
0
7
R
a
n
d
o
m F
o
r
e
st
0
.
2
3
8
1
0
.
4
8
8
0
0
.
7
6
1
9
G
a
u
ssi
a
n
N
a
ï
v
e
B
a
y
e
s
0
.
2
3
8
1
0
.
4
8
8
0
0
.
7
6
Fig
u
r
e
7
s
h
o
w
s
t
h
e
co
m
p
ar
ati
v
e
an
al
y
s
is
i
n
ter
m
s
o
f
ac
cu
r
a
c
y
.
Fig
u
r
e
7
.
C
o
m
p
ar
ativ
e
An
al
y
s
is
in
ter
m
s
o
f
Acc
u
r
ac
y
T
ab
le
3
s
h
o
w
s
C
o
m
p
ar
is
o
n
o
f
A
l
g
o
r
ith
m
s
f
o
r
tr
ain
i
n
g
t
i
m
e,
T
r
ain
in
g
an
d
Sco
r
e.
L
o
g
is
tic
R
eg
r
es
s
io
n
g
iv
e
s
t
h
e
b
est
te
s
ti
n
g
s
co
r
e
o
f
7
7
%.
Neu
r
al
Net
C
las
s
i
f
ier
t
ak
es
th
e
lo
n
g
est
ti
m
e
to
tr
ain
th
e
d
atase
t.
R
ec
a
ll,
P
r
ec
is
io
n
,
A
cc
u
r
ac
y
ca
lc
u
late
d
u
s
in
g
co
n
f
u
s
io
n
m
atr
i
x
an
d
t
h
e
co
m
p
ar
is
o
n
i
s
d
o
n
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
A
C
o
mp
a
r
a
tive
A
n
a
lysi
s
o
n
th
e
E
va
lu
a
tio
n
o
f
C
la
s
s
ifica
tio
n
A
lg
o
r
ith
ms in
th
e
P
r
ed
ictio
n
.
..
(
R
a
tn
a
P
a
til)
3973
T
ab
le
3.
C
o
m
p
ar
is
o
n
o
f
A
lg
o
r
ith
m
s
f
o
r
tr
ain
i
n
g
ti
m
e
a
n
d
Sc
o
r
e
C
l
a
ssi
f
i
e
r
T
r
a
i
n
_
S
c
o
r
e
T
e
st
_
S
c
o
r
e
T
r
a
i
n
i
n
g
_
t
i
me
N
a
ï
v
e
B
a
y
e
s
0
.
7
6
7
2
0
.
7
6
1
9
0
.
0
0
4
1
L
o
g
i
st
i
c
R
e
g
r
e
ssi
o
n
0
.
7
6
7
2
0
.
7
8
3
6
0
.
0
1
9
0
R
a
n
d
o
m F
o
r
e
st
0
.
9
9
6
3
0
.
7
7
0
6
0
.
1
1
4
6
K
N
e
a
r
e
st
N
e
i
g
h
b
o
r
s
0
.
7
8
9
6
0
.
7
4
8
9
0
.
0
0
3
0
G
r
a
d
i
e
n
t
B
o
o
st
i
n
g
0
.
9
3
3
0
0
.
7
8
3
6
0
.
3
4
1
4
D
e
c
i
si
o
n
t
r
e
e
1
.
0
0
0
0
0
.
7
4
0
3
0
.
0
0
7
9
L
i
n
e
a
r
S
V
M
1
.
0
0
0
0
0
.
6
7
9
7
0
.
1
7
7
7
N
e
u
r
a
l
N
e
t
0
.
7
5
2
3
0
.
7
1
4
3
0
.
9
1
7
7
Fig
u
r
e
8
s
h
o
w
s
t
h
e
co
m
p
ar
ati
v
e
an
al
y
s
is
i
n
ter
m
s
o
f
s
co
r
e
an
d
tr
ain
i
n
g
t
i
m
e
.
Fig
u
r
e
8
.
C
o
m
p
ar
ativ
e
An
al
y
s
is
in
ter
m
s
o
f
Sco
r
e
an
d
tr
ain
i
n
g
ti
m
e
T
ab
le
4
s
h
o
w
s
t
h
e
r
es
u
lt
s
f
o
r
P
I
MA
o
n
alg
o
r
it
h
m
s
.
T
ab
le
4.
R
esu
lts
f
o
r
P
I
MA
o
n
alg
o
r
ith
m
s
C
l
a
ssi
f
i
e
r
P
r
e
c
i
si
o
n
R
e
c
a
l
l
F1
–
M
e
a
s
u
r
e
R
O
C
N
a
ï
v
e
B
a
y
e
s
0
.
7
2
9
9
0
.
6
9
6
2
0
.
7
0
7
0
0
.
7
0
L
o
g
i
st
i
c
R
e
g
r
e
ssi
o
n
0
.
7
6
2
2
0
.
7
1
5
7
0
.
7
2
9
8
0
.
7
5
R
a
n
d
o
m F
o
r
e
st
0
.
7
2
8
8
0
.
6
9
9
8
0
.
7
0
9
6
0
.
7
0
K
N
e
a
r
e
st
N
e
i
g
h
b
o
r
s
0
.
7
1
1
0
0
.
6
9
0
3
0
.
6
9
7
8
0
.
6
9
G
r
a
d
i
e
n
t
B
o
o
st
i
n
g
C
l
a
ss
i
f
i
e
r
0
.
7
7
3
6
0
.
7
4
7
1
0
.
7
5
4
0
0
.
7
5
D
e
c
i
si
o
n
T
r
e
e
0
.
6
9
6
0
0
.
7
0
6
1
0
.
7
0
0
.
7
1
L
i
n
e
a
r
S
V
M
0
.
3
3
9
8
0
.
5
0
0
.
4
0
4
6
0
.
5
0
N
e
u
r
a
l
N
e
t
0
.
6
1
2
3
0
.
6
2
4
9
0
.
6
1
2
8
0
.
6
2
Fig
u
r
e
9
s
h
o
w
s
t
h
e
c
o
m
p
ar
ati
v
e
an
al
y
s
is
o
f
al
g
o
r
ith
m
s
in
te
r
m
s
o
f
R
OC
.
Fig
u
r
e
9
.
C
o
m
p
ar
ativ
e
a
n
al
y
s
i
s
o
f
alg
o
r
it
h
m
s
i
n
ter
m
s
o
f
R
O
C
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.
8
,
No
.
5
,
Octo
b
er
2
0
1
8
:
3
9
6
6
–
3
9
7
5
3974
Fig
u
r
e
1
0
s
h
o
w
s
t
h
e
c
o
m
p
ar
at
iv
e
an
al
y
s
is
o
f
al
g
o
r
ith
m
s
in
t
er
m
s
o
f
r
ec
all,
p
r
ec
is
io
n
,
ac
c
u
r
ac
y
,
R
O
C
.
Fig
u
r
e
1
0
.
C
o
m
p
ar
ativ
e
an
a
l
y
s
is
o
f
al
g
o
r
ith
m
s
in
ter
m
s
o
f
R
ec
all,
P
r
ec
is
io
n
,
A
cc
u
r
ac
y
,
R
O
C
6.
CO
NC
L
U
SI
O
N
I
n
th
i
s
p
ap
er
,
w
e
h
av
e
i
n
s
p
ec
t
ed
th
e
ex
ec
u
t
io
n
o
f
ei
g
h
t
m
ac
h
in
e
lear
n
i
n
g
al
g
o
r
ith
m
s
n
a
m
e
l
y
L
o
g
is
tic
R
eg
r
es
s
io
n
,
K
Nea
r
est
Nei
g
h
b
o
r
s
(
KNN)
,
SVM,
Gr
a
d
ien
t
B
o
o
s
t,
Dec
is
io
n
tr
ee
,
ML
P
,
R
an
d
o
m
Fo
r
est
a
n
d
Gau
s
s
ia
n
Naïv
e
to
p
r
ed
ict
th
e
p
o
p
u
latio
n
w
h
o
ar
e
m
o
s
t
li
k
el
y
to
d
ev
elo
p
d
iab
etes
o
n
P
im
a
I
n
d
ia
n
d
iab
etes
d
ata.
T
h
e
p
er
f
o
r
m
an
ce
m
ea
s
u
r
e
m
en
t
is
co
m
p
ar
ed
in
ter
m
s
o
f
M
A
E
,
R
M
SE,
R
O
C
,
T
est
A
cc
u
r
ac
y
,
P
r
ec
is
io
n
an
d
R
ec
all
o
b
tain
ed
f
r
o
m
th
e
test
s
et.
Her
e
th
e
s
t
u
d
ies
co
n
cl
u
d
e
th
at
L
o
g
is
tic
R
e
g
r
ess
io
n
an
d
Gr
ad
ien
t
B
o
o
s
t
class
i
f
ier
s
ac
h
ie
v
e
h
i
g
h
er
tes
t
ac
cu
r
ac
y
o
f
7
9
%
t
h
an
o
th
er
c
lass
i
f
ier
s
.
Fu
r
t
h
er
,
w
e
p
lan
to
r
ec
r
ea
te
o
u
r
s
t
u
d
y
o
f
C
la
s
s
i
f
icat
io
n
m
o
d
el
s
b
y
in
tr
o
d
u
ci
n
g
th
e
in
te
lli
g
en
t
m
ac
h
in
e
lear
n
i
n
g
alg
o
r
it
h
m
s
ap
p
lied
to
a
lar
g
e
co
llectio
n
o
f
r
ea
l
li
f
e
d
ata
s
e
t.
U
s
i
n
g
Ga
u
s
s
ian
F
u
zz
y
d
ec
is
io
n
tr
ee
al
g
o
r
ith
m
f
o
r
t
h
e
d
iag
n
o
s
i
s
ac
c
u
r
ac
y
o
b
tain
ed
w
as
7
5
%
[
2
4
]
.
Desig
n
o
f
a
Diab
etic
Dia
g
n
o
s
is
S
y
s
te
m
Usi
n
g
R
o
u
g
h
Set
s
ac
c
u
r
ac
y
o
b
tai
n
ed
w
a
s
7
6
%
[
2
5
]
.
T
h
e
r
esu
lt
s
o
b
tai
n
e
d
b
y
o
u
r
e
x
p
er
i
m
e
n
tal
al
g
o
r
ith
m
s
ca
n
b
e
f
u
r
t
h
er
i
m
p
r
o
v
ed
b
y
ap
p
l
y
i
n
g
o
u
tl
ier
d
etec
tio
n
b
ef
o
r
e
class
i
f
icatio
n
.
T
h
is
s
tu
d
y
ca
n
b
e
u
s
ed
to
s
ele
ct
b
est cla
s
s
i
f
ier
f
o
r
p
r
ed
ictin
g
d
iab
etes.
RE
F
E
R
E
NC
E
S
[1
]
K.
S
e
lv
a
k
u
b
e
ra
n
,
e
t
a
l.
,
“
A
n
Ef
f
i
c
ien
t
F
e
a
tu
re
S
e
lec
ti
o
n
M
e
th
o
d
f
o
r
Clas
sif
ica
ti
o
n
i
n
He
a
lt
h
Ca
re
S
y
st
e
m
s
Us
in
g
M
a
c
h
in
e
L
e
a
rn
in
g
T
e
c
h
n
iq
u
e
s”
,
IEE
E
,
pp.
8
6
1
0
-
8
6
1
5
,
2
0
1
1
.
[2
]
M.
S
e
e
ra
,
e
t
a
l
.
,
“
A
H
y
b
rid
I
n
te
ll
ig
e
n
t
S
y
ste
m
f
o
r
M
e
d
ica
l
Da
ta
C
las
sif
ic
a
ti
o
n
”
,
Exp
e
rt
El
se
v
ier
:
S
y
ste
ms
wit
h
Ap
p
li
c
a
ti
o
n
s
,
v
o
l
.
4
1
,
p
p
.
2
2
3
9
-
2249
,
2
0
1
4
.
[3
]
T
.
K
a
rth
ik
e
y
a
n
,
e
t
a
l.
,
“
A
n
In
telli
g
e
n
t
T
y
p
e
-
II
Dia
b
e
tes
M
e
ll
it
u
s
Dia
g
n
o
sis
A
p
p
ro
a
c
h
u
sin
g
Im
p
ro
v
e
d
F
P
-
g
ro
w
th
w
it
h
H
y
b
rid
Clas
sif
ier
Ba
se
d
A
r
m
Re
s
e
a
rc
h
”
,
J
o
u
rn
a
l
o
f
Ap
p
li
e
d
S
c
ien
c
e
s,
E
n
g
i
n
e
e
rin
g
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
1
1
,
n
o
.
5
,
p
p
.
5
4
9
-
5
5
8
,
2
0
1
5
.
[4
]
D
.
K.
Ka
ru
m
a
n
c
h
i,
e
t
a
l.
,
“
Early
d
iag
n
o
sis
o
f
Dia
b
e
tes
m
e
ll
it
u
s
th
r
o
u
g
h
t
h
e
e
y
e
”
,
2
n
d
I
n
ter
n
a
t
io
n
a
l
Co
n
fer
e
n
c
e
o
n
En
d
o
c
rin
o
lo
g
y
,
2
0
1
4
.
[5
]
M
.
B.
W
a
n
k
h
a
d
e
a
n
d
A
.
A
.
G
u
rjar
,
“
A
n
a
l
y
sis
o
f
Dise
a
se
u
sin
g
Re
ti
n
a
l
Bl
o
o
d
V
e
ss
e
ls
De
t
e
c
ti
o
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
ol
.
5
,
n
o
.
12
,
pp.
1
9
6
4
4
-
1
9
6
4
7
,
2
0
1
6
.
[6
]
S
.
B
.
Ch
o
i
,
e
t
a
l
.
,
“
S
c
re
e
n
i
n
g
f
o
r
P
re
d
iab
e
tes
Us
in
g
M
a
c
h
in
e
L
e
a
rn
in
g
M
o
d
e
ls”
,
Hi
n
d
a
wi
Pu
b
li
sh
in
g
Co
rp
o
r
a
ti
o
n
Co
m
p
u
t
a
ti
o
n
a
l
a
n
d
M
a
th
e
ma
ti
c
a
l
M
e
th
o
d
s in
M
e
d
ici
n
e
,
2
0
1
4
.
[7
]
M
.
S
.
Kle
in
a
n
d
J
.
S
h
e
a
re
r
,
“
M
e
tab
o
l
o
m
ics
a
n
d
T
y
p
e
2
Dia
b
e
tes
:
T
ra
n
sla
ti
n
g
Ba
sic
Re
se
a
r
c
h
i
n
to
Cli
n
ica
l,
A
p
p
li
c
a
ti
o
n
”
,
Hi
n
d
a
wi
P
u
b
li
s
h
in
g
Co
r
p
o
ra
t
io
n
J
o
u
rn
a
l
o
f
Dia
b
e
tes
Res
e
a
r
c
h
,
2
0
1
5
.
[8
]
M
.
Ko
th
a
i
n
a
y
a
k
i
a
n
d
P
.
T
h
a
n
g
a
ra
j
,
“
Clu
ste
rin
g
a
n
d
Clas
sify
in
g
Dia
b
e
ti
c
Da
ta
S
e
ts
Us
in
g
K
-
M
e
a
n
s
A
lg
o
rit
h
m
”
.
A
rti
c
le c
a
n
b
e
a
c
c
e
ss
e
d
o
n
li
n
e
a
t
h
tt
p
:
//
ww
w
.
p
u
b
li
s
h
in
g
in
d
ia
.
[9
]
M.
N
.
De
v
i,
e
t
a
l
.,
“
A
n
A
m
a
lg
a
m
K
NN
to
P
re
d
ict
D
ia
b
e
tes
m
e
ll
it
u
s
”
,
IEE
E
,
2
0
1
3
.
[1
0
]
N
.
H.
Ba
ra
k
a
t,
e
t
a
l.
,
“
In
tell
i
g
ib
le
S
u
p
p
o
rt
V
e
c
to
r
M
a
c
h
in
e
s
f
o
r
Dia
g
n
o
sis
o
f
Dia
b
e
tes
M
e
ll
it
u
s”
,
I
E
E
E
T
ra
n
s
a
c
ti
o
n
s
o
n
In
fo
rm
a
t
io
n
T
e
c
h
n
o
l
o
g
y
i
n
Bi
o
me
d
icin
e
,
vo
l
.
14
,
n
o
.
4
,
2
0
1
0
.
[1
1
]
T
.
S
a
n
th
a
n
a
m
a
n
d
M.
S
.
P
a
d
m
a
v
a
th
i
,
“
A
p
p
li
c
a
ti
o
n
o
f
K
-
M
e
a
n
s
a
n
d
G
e
n
e
ti
c
A
lg
o
rit
h
m
s
f
o
r
Di
m
e
n
sio
n
Re
d
u
c
ti
o
n
b
y
In
teg
ra
ti
n
g
S
V
M
f
o
r
Dia
b
e
tes
Dia
g
n
o
sis”
,
Pr
o
c
e
d
ia
C
o
mp
u
ter
S
c
ien
c
e
,
v
o
l.
47
,
p
p
.
76
-
83
,
2
0
1
5
.
[1
2
]
A
.
G
.
Ka
re
g
o
w
d
a
,
e
t
a
l.
,
“
Ru
le
b
a
se
d
c
las
si
f
ica
ti
o
n
f
o
r
d
iab
e
ti
c
p
a
ti
e
n
ts
u
si
n
g
c
a
sc
a
d
e
d
K
-
m
e
a
n
s
a
n
d
d
e
c
isio
n
tree
C4
.
5
”
,
I
n
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
C
o
mp
u
ter
Ap
p
li
c
a
t
io
n
s
,
v
o
l
.
4
5
,
n
o
.
1
2
,
p
p
.
0
9
7
5
-
8
8
8
7
,
2
0
1
2
.
[1
3
]
H.
T
e
m
u
rtas
,
e
t
a
l.
,
“
A
C
o
m
p
a
ra
ti
v
e
S
tu
d
y
o
n
Dia
b
e
tes
d
ise
a
se
D
i
a
g
n
o
sis
u
sin
g
N
e
u
ra
l
Ne
tw
o
rk
s
”
,
El
se
v
ier
:
Exp
e
rt
S
y
ste
ms
wit
h
A
p
p
li
c
a
ti
o
n
s
,
v
o
l.
3
6
,
2
0
0
9
.
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
A
C
o
mp
a
r
a
tive
A
n
a
lysi
s
o
n
th
e
E
va
lu
a
tio
n
o
f
C
la
s
s
ifica
tio
n
A
lg
o
r
ith
ms in
th
e
P
r
ed
ictio
n
.
..
(
R
a
tn
a
P
a
til)
3975
[1
4
]
K.
S
rin
iv
a
s,
e
t
a
l.
,
“
H
y
b
rid
A
p
p
ro
a
c
h
f
o
r
P
re
d
ictio
n
o
f
Ca
rd
io
v
a
sc
u
lar
Dise
a
s
e
Us
in
g
Cl
a
ss
As
so
c
iatio
n
Ru
les
a
n
d
M
L
P
”
,
I
n
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
E
lec
trica
l
a
n
d
C
o
mp
u
ter
En
g
in
e
e
ri
n
g
,
v
ol
.
6
,
n
o
.
4
,
p
p
.
1
8
0
0
-
1
8
1
0
,
2
0
1
6
.
[1
5
]
R.
Ka
la,
e
t
a
l
.
,
“
Dia
g
n
o
sis
o
f
Bre
a
st
c
a
n
c
e
r
b
y
M
o
d
u
l
a
r
E
v
o
lu
ti
o
n
a
ry
Ne
u
ra
l
N
e
tw
o
rk
s
”
,
In
d
e
rs
c
ien
c
e
:
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
B
io
me
d
i
c
a
l
En
g
in
e
e
rin
g
a
n
d
T
e
c
h
n
o
l
o
g
y
(
IJ
BE
T
)
,
v
ol
.
7
,
n
o
.
2
,
p
p
.
1
9
4
-
2
1
1
,
2
0
1
1
.
[1
6
]
C.
Hs
ieh
,
e
t
a
l.
,
“
No
v
e
l
S
o
lu
ti
o
n
s
f
o
r
a
n
o
ld
d
ise
a
se
:
Dia
g
n
o
sis
o
f
A
c
u
te A
p
p
e
n
d
iciti
s
w
it
h
ra
n
d
o
m
f
o
re
st,
S
u
p
p
o
rt
V
e
c
to
r
M
a
c
h
in
e
s,
a
n
d
A
rti
f
icia
l
Ne
u
ra
l
Ne
tw
o
rk
s
”
,
S
u
r
g
e
ry
,
v
o
l
.
1
4
9
,
n
o
.
1
,
p
p
.
8
7
-
9
3
,
2
0
1
1
.
[1
7
]
M
d
.
M
.
M
o
tt
a
li
b
,
e
t
a
l.
,
“
De
tec
ti
o
n
o
f
th
e
On
se
t
o
f
Dia
b
e
tes
M
e
ll
i
tu
s
b
y
Ba
y
e
sia
n
Clas
sif
ier
B
a
se
d
M
e
d
ica
l
Ex
p
e
rt
S
y
st
e
m
”
,
T
ra
n
sa
c
ti
o
n
o
n
M
a
c
h
in
e
L
e
a
rn
in
g
a
n
d
Art
if
icia
l
In
tell
ig
e
n
c
e
,
2
0
1
6
.
[1
8
]
P
.
Y
a
so
d
h
a
a
n
d
M
.
Ka
n
n
a
n
,
“
A
n
a
ly
sis
o
f
a
p
o
p
u
lat
io
n
o
f
d
iab
e
ti
c
p
a
ti
e
n
t’
s
d
a
tab
a
se
s
in
W
e
k
a
to
o
l”
,
Pr
o
c
e
e
d
in
g
s
o
f
th
e
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
S
c
ien
ti
fi
c
&
En
g
in
e
e
rin
g
Res
e
a
r
c
h
,
v
o
l
.
2
,
n
o
.
5
,
2
0
1
1
.
[1
9
]
H
.
M
a
h
a
jan
,
e
t
a
l
.
,
“
He
a
lt
h
I
n
terv
e
n
ti
o
n
Im
p
a
c
t
A
ss
e
ss
m
e
n
t
o
n
G
l
y
c
e
m
ic
S
tatu
s
o
f
Dia
b
e
ti
c
P
a
ti
e
n
ts
”
,
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Dia
b
e
tes
Res
e
a
r
c
h
,
p
p
.
73
-
80
,
2
0
1
2
.
[2
0
]
M
.
A
b
d
a
r
,
e
t
a
l
.
,
“
C
o
m
p
a
rin
g
P
e
r
f
o
r
m
a
n
c
e
o
f
Da
ta
M
in
in
g
A
lg
o
rit
h
m
s
in
P
re
d
icti
o
n
He
a
rt
Dise
a
se
s
”
,
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
,
v
ol
.
5
,
n
o
.
6
,
p
p
.
1
5
6
9
-
1
5
7
6
,
2
0
1
5
.
[2
1
]
R
.
M
.
Ra
h
m
a
n
a
n
d
F
.
A
f
o
z
,
“
Co
m
p
a
riso
n
o
f
V
a
rio
u
s
Clas
sif
ica
ti
o
n
T
e
c
h
n
iq
u
e
s
u
sin
g
d
if
f
e
r
e
n
t
Da
ta
M
in
in
g
T
o
o
ls
f
o
r
Dia
b
e
tes
Dia
g
n
o
sis
”,
J
o
u
rn
a
l
o
f
S
o
ft
w
a
r
e
En
g
in
e
e
rin
g
a
n
d
Ap
p
l
ica
ti
o
n
s
,
v
o
l.
6,
p
p
.
85
-
97
,
2
0
1
3
.
[2
2
]
V
.
P
e
ll
a
k
u
ri
,
e
t
a
l
.
,
“
P
e
rf
o
rm
a
n
c
e
A
n
a
l
y
sis
a
n
d
Op
ti
m
iza
ti
o
n
o
f
S
u
p
e
rv
ise
d
L
e
a
rn
in
g
T
e
c
h
n
iq
u
e
s
f
o
r
M
e
d
ica
l
Dia
g
n
o
sis
Us
in
g
Op
e
n
S
o
u
rc
e
T
o
o
ls”
,
I
n
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
C
o
mp
u
ter
S
c
ien
c
e
a
n
d
In
f
o
rm
a
ti
o
n
T
e
c
h
n
o
l
o
g
ies
,
v
ol
.
6
,
n
o
.
1
,
p
p
.
3
8
0
-
3
8
3
,
2
0
1
5
.
[2
3
]
R
.
R.
Ra
o
a
n
d
K
.
M
a
k
k
it
h
a
y
a
,
“
L
e
a
rn
in
g
f
ro
m
a
Clas
s
I
m
b
a
lan
c
e
d
P
u
b
li
c
He
a
lt
h
Da
tas
e
t:
A
Co
st
-
b
a
se
d
Co
m
p
a
riso
n
o
f
Clas
sif
ier
P
e
rf
o
rm
a
n
c
e
”
,
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
Co
mp
u
ter
En
g
i
n
e
e
rin
g
,
v
ol
.
7
,
n
o
.
4
,
p
p
.
2
2
1
5
-
2
2
2
2
,
2
0
1
7
.
[2
4
]
K
.
V.
S.
R.
P
.
V
a
rm
a
,
e
t
a
l.
,
“
A
Co
m
p
u
tatio
n
a
l
I
n
telli
g
e
n
c
e
A
p
p
ro
a
c
h
f
o
r
a
b
e
tt
e
r
Dia
g
n
o
sis
o
f
Di
a
b
e
ti
c
P
a
ti
e
n
ts”
,
Co
mp
u
ter
s
a
n
d
El
e
c
trica
l
E
n
g
in
e
e
rin
g
,
v
o
l
.
40
,
p
p
.
1
7
5
8
-
1
7
6
5
,
2
0
1
4
.
[2
5
]
M
.
A
n
o
u
n
c
ia
S
.
,
e
t
a
l.
,
“
De
sig
n
o
f
a
Dia
b
e
ti
c
Dia
g
n
o
sis
S
y
ste
m
u
sin
g
Ro
u
g
h
S
e
ts”
,
Cy
b
e
rn
e
ti
c
s
a
n
d
In
fo
rm
a
t
io
n
T
e
c
h
n
o
l
o
g
ies
,
v
ol
.
13
,
n
o
.
3
,
2
0
1
3
.
Evaluation Warning : The document was created with Spire.PDF for Python.