IA
E
S
In
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
A
r
t
i
f
i
c
i
a
l
In
t
e
l
l
i
g
e
n
c
e
(
IJ
-
AI
)
Vo
l
.
8
, N
o
. 1
, M
a
r
c
h
2
0
1
9
, p
p
.
77
~
86
IS
S
N
:
2
2
5
2
-
8938
,
DOI
:
1
0
.
1
1
5
9
1
/
i
j
a
i
.
v
8
.i
1
.p
p
77
-
86
r
77
Jou
r
n
al
h
om
e
page
:
ht
t
p:
/
/
i
ae
s
co
r
e
.c
o
m
/o
n
lin
e
/in
d
e
x
.p
h
p
/I
J
AI
An
i
m
p
r
o
v
e
d
h
y
b
r
i
d
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
m
e
t
h
o
d
f
o
r
h
u
g
e
di
m
e
ns
i
o
na
l
da
t
a
s
e
t
s
F.
R
o
s
i
t
a
K
a
m
a
l
a
1
,
P.
R
a
n
j
i
t
J
e
b
a
T
h
a
n
g
a
i
a
h
2
1
De
p
a
r
t
me
n
t
o
f
C
o
mp
u
t
e
r
S
c
i
e
n
c
e
,
B
h
a
r
a
t
h
i
a
r
Un
i
v
e
r
s
i
t
y
,
I
n
d
i
a
2
De
p
a
r
t
me
n
t
o
f
I
n
f
o
r
ma
t
i
o
n
T
e
c
h
n
o
l
o
g
y
,
Ka
r
u
n
y
a
In
s
t
i
t
u
t
e
o
f
T
e
c
h
n
o
l
o
g
y
a
n
d
S
c
i
e
n
c
e
s
,
In
d
i
a
Ar
t
i
c
l
e
I
n
f
o
AB
S
T
RACT
Ar
t
i
c
l
e
h
i
s
t
o
r
y
:
Re
c
e
i
v
e
d
N
o
v
2
5,
2018
Re
v
i
s
e
d
F
e
b
1
,
2
0
1
9
Ac
c
e
p
t
e
d
Fe
b
2
2
,
2
0
1
9
Hi
g
h
d
i
me
n
s
i
o
n
s
o
f
d
a
t
a
c
a
u
s
e
o
v
e
r
f
i
t
t
i
n
g
i
n
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
mo
d
e
l
s
,
can
le
a
d
to
r
e
d
u
c
tio
n
in
a
c
c
u
r
a
c
y
dur
i
ng
c
l
a
s
s
i
f
i
c
a
t
i
on
of
i
ns
t
a
nc
e
s
.
Va
r
i
a
b
l
e
s
el
ect
i
o
n
i
s
t
h
e
m
o
st
essen
t
i
al
f
u
n
ct
i
o
n
i
n
p
r
ed
i
ct
i
v
e
an
al
y
t
i
cs,
th
a
t
r
e
d
u
c
e
s
th
e
d
im
e
n
s
io
n
a
lity
,
w
ith
o
u
t
lo
s
in
g
a
n
a
p
p
r
o
p
r
ia
te
in
f
o
r
m
a
tio
n
b
y
s
e
le
c
tin
g
a
fe
w
s
i
g
n
i
fi
c
a
n
t
fe
a
t
u
re
s
o
f
m
a
c
h
i
n
e
l
e
a
rn
i
n
g
p
ro
b
l
e
m
s
.
T
h
e
m
a
j
o
r
t
e
c
h
n
i
q
u
e
s
in
v
o
lv
e
d
in
th
is
p
r
o
c
e
s
s
a
r
e
f
ilte
r
a
n
d
w
r
a
p
p
e
r
m
e
th
o
d
o
lo
g
ie
s
.
W
h
ile
f
ilte
r
s
me
a
s
u
r
e
t
h
e
we
i
g
h
t
o
f
f
e
a
t
u
r
e
s
b
a
s
e
d
o
n
t
h
e
a
t
t
r
i
b
u
t
e
we
i
g
h
t
i
n
g
c
r
i
t
e
r
i
o
n
,
t
he
wr
a
p
p
e
r
a
p
p
r
o
a
c
h
c
o
mp
u
t
e
s
t
h
e
c
o
mp
e
t
e
n
c
e
o
f
t
h
e
v
a
r
i
a
b
l
e
s
e
l
e
c
t
i
o
n
al
g
o
r
i
t
h
m
s.
Th
e
w
r
a
p
p
e
r
a
p
p
r
o
a
c
h
is
a
c
h
ie
v
e
d
b
y
th
e
s
e
le
c
tio
n
o
f
f
e
a
tu
r
e
su
b
g
r
o
u
p
s
b
y
p
r
u
n
i
n
g
t
h
e
f
eat
u
r
e
sp
ace
i
n
i
t
s
sear
ch
sp
ace.
Th
e
o
b
j
e
c
t
i
v
e
o
f
th
is
p
a
p
e
r
is
to
c
h
o
o
s
e
th
e
mo
s
t
fa
v
o
u
ra
b
l
e
a
t
t
ri
b
u
t
e
s
u
b
s
e
t
fro
m
t
h
e
n
o
v
e
l
s
e
t
of
f
e
a
t
ur
e
s
,
by
us
i
ng
t
he
c
om
bi
na
t
i
on
m
e
t
hod
t
ha
t
uni
t
e
s
t
he
m
e
r
i
t
s
of
f
i
l
t
e
r
s
an
d
w
r
ap
p
er
s
.
To
a
c
h
i
e
v
e
t
h
i
s
o
b
j
e
c
t
i
v
e
,
a
n
I
m
p
r
o
v
e
d
H
y
b
r
i
d
Fe
a
t
u
r
e
Se
l
e
c
t
i
o
n
(I
H
FS)
me
t
h
o
d
is
pe
r
f
or
m
e
d
t
o
c
r
e
a
t
e
w
e
l
l
-
or
ga
ni
z
e
d
l
e
a
r
ne
r
s
.
Th
e
re
s
u
l
t
s
o
f
t
h
i
s
s
t
u
d
y
s
h
o
w
s
t
h
a
t
t
h
e
I
H
FS
a
l
g
o
r
i
t
h
m
can
b
u
i
l
d
co
m
p
et
en
t
bus
i
ne
s
s
a
ppl
i
c
a
t
i
ons
,
w
hi
c
h
ha
ve
got
a
be
t
t
e
r
pr
e
c
i
s
i
on
t
ha
n
t
ha
t
of
t
he
co
n
st
r
u
ct
ed
wh
i
c
h
i
s
s
t
a
t
e
d
by
t
he
pr
e
vi
ous
hybr
i
d
va
r
i
a
bl
e
s
el
ect
i
o
n
al
g
o
r
i
t
h
m
s.
Ex
p
e
r
i
m
e
n
t
a
t
i
o
n
w
i
t
h
UC
I
(
Un
i
v
e
r
s
i
t
y
o
f
C
a
l
i
f
o
r
n
i
a
,
I
r
v
i
n
e
)
re
p
o
s
i
t
o
ry
d
a
t
a
s
e
t
s
af
f
i
r
m
s
th
a
t
th
is
m
e
th
o
d
h
a
v
e
g
o
t
b
e
tte
r
p
r
e
d
ic
tio
n
pe
r
f
or
m
a
nc
e
,
m
or
e
r
obus
t
t
o
i
nput
noi
s
e
a
nd
out
l
i
e
r
s
,
ba
l
a
nc
e
s
w
e
l
l
w
i
t
h
t
he
av
ai
l
ab
l
e
f
eat
u
r
es,
w
h
en
pe
r
f
or
m
e
d
c
om
pa
r
i
s
on
w
i
t
h
t
he
pr
e
s
e
nt
a
l
gor
i
t
hm
s
in
th
e
lite
r
a
tu
r
e
r
e
v
ie
w
.
Ke
y
wo
r
d
s
:
Fe
a
t
u
r
e
s
e
l
e
c
t
i
o
n
Hy
b
r
i
d
a
p
p
r
o
a
c
h
Ma
c
h
i
n
e
l
e
a
r
n
i
n
g
Ov
e
r
f
i
t
t
i
n
g
Pr
e
d
i
c
t
i
v
e
a
n
a
l
y
t
i
c
s
Va
r
i
a
b
l
e
s
e
l
e
c
t
i
o
n
Co
p
y
r
i
g
h
t
©
2
0
1
9
In
s
t
i
t
u
t
e
o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
r
i
n
g
a
n
d
S
c
i
e
n
c
e
.
Al
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
Co
r
r
e
s
p
o
n
d
i
n
g
Au
t
h
o
r
:
Ro
s
i
t
a
K
a
m
a
l
a
F
,
De
p
a
r
t
m
e
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
Bh
a
r
a
t
h
i
a
r
U
n
i
v
e
r
s
i
t
y
,
Co
i
m
b
a
t
o
r
e
,
T
a
m
i
l
N
a
d
u
,
I
n
d
i
a
.
Em
a
i
l
:
r
o
s
i
t
a
k
a
m
a
l
a
@
g
m
a
i
l
.
c
o
m
1.
IN
T
R
O
D
U
C
T
IO
N
Th
e
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
p
r
o
b
l
e
m
s
u
s
e
t
h
e
t
e
r
m
c
u
r
s
e
o
f
d
i
m
e
n
s
i
o
n
a
l
i
t
y
t
o
r
e
f
e
r
a
n
ex
p
o
n
en
t
i
al
i
n
cr
eas
e
of
m
or
e
num
be
r
of
di
m
e
ns
i
ons
of
f
e
a
t
ur
e
s
i
n
a
m
a
t
he
m
a
t
i
c
a
l
s
pa
c
e
[
1]
.
H
i
gh
di
m
e
ns
i
ona
l
da
t
a
i
s
f
ound
t
o
be
a
m
aj
o
r
p
r
o
b
l
em
i
d
en
t
i
f
i
ed
i
n
s
u
p
er
v
i
s
ed
an
d
u
n
s
u
p
er
v
i
s
ed
l
ear
n
i
n
g
.
H
i
g
h
d
i
m
en
s
i
o
n
al
i
t
y
o
f
t
en
en
t
ai
l
s
h
i
g
h
va
r
i
a
nc
e
,
l
e
a
di
ng
t
o
uns
t
a
bl
e
l
e
a
r
ni
ng
out
c
om
e
s
.
T
o
pr
oduc
e
s
t
a
bl
e
l
e
a
r
ni
ng
i
n
s
t
a
t
i
s
t
i
c
a
l
m
ode
l
s
of
hi
ghe
r
di
m
e
ns
i
ons
,
a
l
a
r
ge
num
be
r
of
s
a
m
pl
e
s
a
r
e
r
e
qui
r
e
d.
L
a
r
ge
r
vol
um
e
s
r
e
s
ul
t
hi
gh
va
r
i
a
nc
e
,
c
a
us
i
ng
uns
t
a
bl
e
le
a
r
n
in
g
o
u
tc
o
m
e
s
.
L
a
r
g
e
r
c
a
lc
u
la
tio
n
is
e
n
f
o
r
c
e
d
f
o
r
d
e
a
lin
g
w
ith
h
ig
h
-
di
m
e
ns
i
ona
l
da
t
a
s
e
t
s
.
N
ow
a
da
ys
,
it
is
b
e
c
o
m
in
g
a
b
i
g
ch
al
l
en
g
e
t
o
d
at
a
s
ci
en
t
i
s
t
s
an
d
b
u
s
i
n
es
s
an
al
y
s
t
s
.
T
h
e
i
n
cr
eas
e
o
f
f
eat
u
r
es
l
ead
s
t
o
va
r
i
ous
pr
obl
e
m
s
l
i
ke
noi
s
e
,
e
r
r
or
a
nd
ove
r
f
i
t
t
i
ng
[
2]
.
I
t
a
l
s
o
l
e
a
ds
t
o
i
nc
r
e
a
s
e
i
n
c
om
put
i
ng
c
os
t
,
s
t
or
i
ng
c
os
t
an
d
m
ak
e
d
at
a
m
i
n
i
n
g
a
ch
al
l
en
g
i
n
g
t
as
k
i
n
v
ar
i
o
u
s
w
a
y
s.
T
h
e
r
e
d
u
c
t
i
o
n
i
n
c
l
a
ssi
f
i
c
a
t
i
o
n
p
e
r
f
o
r
m
a
n
c
e
w
i
t
h
num
be
r
of
f
e
a
t
ur
e
s
i
s
s
how
n
i
n
F
ig
u
r
e
1
.
T
h
e
m
o
s
t
e
f
f
e
c
tiv
e
w
a
y
to
id
e
n
tif
y
r
e
le
v
a
n
t
f
e
a
tu
r
e
s
in
m
a
c
h
in
e
le
a
r
n
in
g
is
f
e
a
tu
r
e
s
e
le
c
tio
n
.
T
o
a
c
h
ie
v
e
m
o
r
e
a
c
c
u
r
a
te
p
r
e
d
ic
tio
n
,
th
e
c
o
n
c
e
p
t
o
f
r
e
le
v
a
n
t
f
e
at
u
r
es
i
s
u
s
ed
i
n
Evaluation Warning : The document was created with Spire.PDF for Python.
r
IS
S
N
:
2252
-
8938
IJ
-
AI
Vo
l
.
8
, N
o
.
1,
M
a
r
c
h
201
9
:
77
–
86
78
mo
r
e
s
t
a
b
l
e
l
e
a
r
n
i
n
g
mo
d
e
l
s
.
T
h
e
s
e
mo
d
e
l
s
a
r
e
e
a
s
y
t
o
u
n
d
e
r
s
t
a
n
d
a
n
d
a
p
p
l
y
.
F
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
(
F
S
)
i
s
a
cr
i
t
i
cal
p
r
o
ced
u
r
e
t
o
i
d
en
t
i
f
y
r
el
at
ed
s
u
b
s
et
s
o
f
f
eat
u
r
es
f
o
r
m
ak
i
ng
a
c
c
ur
a
t
e
pr
e
di
c
t
i
on
i
n
l
a
r
ge
di
m
e
ns
i
ona
l
da
t
a
s
e
t
s
[
3]
.
T
he
m
e
r
i
t
s
o
f
v
a
r
i
a
b
l
e
se
l
e
c
t
i
o
n
a
r
e
m
u
l
t
i
f
o
l
d
a
n
d
a
p
p
l
i
c
a
t
i
o
n
d
e
p
e
n
d
e
n
t
.
Fi
g
u
r
e
1
.
Cl
a
s
s
i
f
i
c
a
t
i
o
n
p
e
r
f
o
r
m
a
n
c
e
V
s
D
i
m
e
n
s
i
o
n
a
l
i
t
y
o
f
f
e
a
t
u
r
e
s
1.
1
.
Ba
c
k
g
r
o
u
n
d
Va
r
i
a
b
l
e
s
e
l
e
c
t
i
o
n
p
o
s
t
u
l
a
t
e
s
o
f
a
l
g
o
r
i
t
h
m
s
a
r
e
b
r
o
a
d
l
y
c
l
a
s
s
i
f
i
e
d
i
n
t
o
t
h
r
e
e
c
a
t
e
g
o
r
i
e
s
t
o
m
e
a
s
u
r
e
re
l
e
v
a
n
c
e
a
n
d
re
d
u
n
d
a
n
c
y
o
f
fe
a
t
u
re
s
.
T
h
e
y
a
re
fi
l
t
e
r,
w
ra
p
p
e
r,
a
n
d
h
y
b
ri
d
m
e
t
h
o
d
s
.
F
i
l
t
e
r
m
e
t
h
o
d
s
a
d
o
p
t
a
me
a
s
u
r
e
o
f
s
t
a
t
i
s
t
i
c
s
t
o
a
l
l
o
c
a
t
e
a
c
o
u
n
t
f
o
r
e
a
c
h
a
n
d
e
v
e
r
y
f
e
a
t
u
r
e
s
l
i
k
e
n
u
me
r
i
c
a
l
o
r
c
o
n
t
i
n
u
o
u
s
,
n
o
mi
n
a
l
o
r
di
s
c
r
e
t
e
a
nd
c
l
a
s
s
l
a
be
l
va
l
ue
s
.
B
a
s
e
d
on
t
he
c
ount
,
t
he
f
e
a
t
ur
e
s
a
r
e
r
a
nke
d
a
nd
e
i
t
he
r
pr
e
f
e
r
r
e
d
t
o
be
ke
pt
or
el
i
m
i
n
at
ed
f
r
o
m
t
h
e
d
at
as
et
.
T
h
i
s
i
s
s
eem
ed
t
o
b
e
v
er
y
s
i
m
p
l
e
an
d
s
cal
e
as
t
h
e
n
u
m
b
er
o
f
s
am
p
l
es
an
d
di
m
e
ns
i
ons
i
nc
r
e
a
s
e
.
Th
e
f
i
l
t
e
r
s
a
r
e
s
e
l
e
c
t
e
d
a
s
t
h
e
m
o
s
t
p
r
o
d
u
c
t
i
v
e
m
e
t
h
o
d
i
n
c
o
mp
a
r
i
s
o
n
w
i
t
h
w
r
a
p
p
e
r
a
n
d
em
b
ed
d
ed
m
et
h
o
d
s
h
av
i
n
g
l
ear
n
i
n
g
i
n
d
ep
en
d
en
ce,
eas
e
o
f
i
m
p
l
em
en
t
at
i
o
n
,
g
o
o
d
g
en
er
al
i
zat
i
o
n
ab
i
l
i
t
y
an
d
be
t
t
e
r
c
om
put
a
t
i
ons
[
3]
.
T
he
l
i
m
i
t
a
t
i
on
of
f
i
l
t
e
r
m
e
t
hods
i
s
t
he
f
e
a
t
ur
e
s
a
r
e
c
a
l
c
ul
a
t
e
d
one
by
one
.
I
t
a
l
s
o
ig
n
o
r
e
s
th
e
a
s
s
o
c
i
at
i
o
n
am
o
n
g
f
eat
u
r
es
an
d
o
v
er
l
o
o
k
s
t
h
e
co
l
l
ab
o
r
at
i
o
n
w
i
t
h
t
h
e
l
ear
n
er
.
Th
e
c
h
o
i
c
e
o
f
a
f
e
a
t
u
r
e
s
u
b
s
e
t
i
s
p
e
r
f
o
r
m
e
d
i
n
w
r
a
p
p
e
r
m
e
t
h
o
d
s
a
s
a
s
e
a
r
c
h
p
r
o
b
l
e
m
[
4
]
.
S
e
a
r
c
h
i
n
g
re
fe
rs
t
o
g
l
o
b
a
l
a
n
d
l
o
c
a
l
s
e
a
rc
h
.
G
l
o
b
a
l
s
e
a
rc
h
s
e
a
rc
h
e
s
d
i
s
t
i
n
c
t
i
v
e
a
re
a
s
i
n
t
h
e
s
e
a
r
ch
s
p
ace,
an
d
s
ear
ch
i
n
g
in
th
e
lo
c
a
l
s
e
a
r
c
h
s
p
a
c
e
is
lo
c
a
l
s
e
a
r
c
h
.
A
b
r
o
a
d
c
la
s
s
if
ic
a
tio
n
o
f
s
u
b
s
e
t
e
x
a
m
in
a
tio
n
a
p
p
r
o
a
c
h
e
s
m
a
y
b
e
sy
st
e
m
a
t
i
c
su
c
h
a
s
a
B
F
S
(
B
e
st
F
i
r
st
S
e
a
r
c
h
)
a
n
d
a
st
o
c
h
a
st
i
c
se
a
r
c
h
,
su
c
h
a
s
r
a
n
d
o
m
h
i
l
l
c
l
i
m
b
i
n
g
a
l
g
o
r
i
t
h
m
,
br
a
nc
h
a
nd
bound,
an
d
ev
o
l
u
t
i
o
n
ar
y
m
et
h
o
d
s
.
T
h
e
k
i
n
d
s
o
f
g
r
eed
y
s
e
ar
ch
s
t
r
at
eg
i
es
ar
e
h
eu
r
i
s
t
i
cs
.
Th
e
y
a
r
e
f
o
r
w
a
r
d
s
t
e
p
w
i
s
e
s
e
l
e
c
t
i
n
g
o
p
t
i
o
n
,
w
h
i
c
h
i
n
c
l
u
d
e
s
v
a
r
i
a
b
l
e
s
g
r
a
d
u
a
l
l
y
i
n
t
o
i
n
c
r
e
a
s
i
n
g
f
e
a
t
u
r
e
su
b
se
t
s
a
n
d
b
a
c
k
w
a
r
d
st
e
p
w
i
se
e
l
i
m
i
n
a
t
i
n
g
o
p
t
i
o
n
b
e
g
i
n
s
f
r
o
m
a
l
l
v
a
r
i
a
b
l
es
a
nd
gr
a
dua
l
l
y
r
e
m
ove
t
he
mi
n
i
mu
m
fa
v
o
u
ra
b
l
e
o
n
e
s
[4
].
A
m
o
n
g
a
l
l
,
g
re
e
d
y
s
e
a
rc
h
m
e
t
h
o
d
s
a
re
m
o
re
a
d
v
a
n
t
a
g
e
o
u
s
c
o
l
l
e
c
t
i
v
e
l
y
a
n
d
st
r
o
n
g
a
g
a
i
n
st
o
v
e
r
f
i
t
t
i
n
g
.
B
u
t
,
t
h
e
w
r
a
p
p
e
r
m
e
t
h
o
d
i
n
t
e
r
a
c
t
s
w
i
t
h
a
c
l
a
ssi
f
i
e
r
.
I
t
u
su
a
l
l
y
e
v
a
l
u
a
t
e
s
t
h
e
f
e
a
t
u
r
e
s
co
n
j
o
i
n
t
l
y
a
nd
c
ons
i
de
r
s
t
he
c
ont
i
nge
nc
y
a
m
ong
t
he
m
t
o
s
e
l
e
c
t
t
he
m
os
t
i
de
a
l
f
e
a
t
ur
e
s
a
ga
i
ns
t
t
he
e
xi
s
t
i
ng
fe
a
t
u
re
s
s
e
t
.
D
i
s
a
d
v
a
n
t
a
g
e
s
o
f
w
ra
p
p
e
rs
e
n
t
a
i
l
m
o
re
e
x
p
e
n
s
e
c
o
m
p
u
t
a
t
i
o
n
a
l
l
y
t
ha
n
t
he
r
e
s
t
of
t
he
m
e
t
hods
.
It
c
o
n
s
u
m
e
s
m
o
re
t
i
m
e
,
m
o
re
v
u
l
n
e
ra
b
l
e
t
o
c
a
u
s
e
o
v
e
rfi
t
tin
g
,
a
n
d
m
o
r
e
le
a
r
n
in
g
d
e
p
e
n
d
e
n
c
y
.
F
o
r
th
is
r
e
a
s
o
n
,
hybr
i
d
m
e
t
hods
a
r
e
a
dopt
e
d
t
o
e
nha
nc
e
t
he
s
e
a
r
c
h
a
l
gor
i
t
hm
.
H
ybr
i
d
s
t
r
a
t
e
gi
e
s
a
r
e
m
or
e
or
l
e
s
s
r
e
l
a
t
e
d
t
o
t
he
wr
a
p
p
e
r
s
t
r
a
t
e
g
i
e
s
.
Hy
b
r
i
d
m
e
t
h
o
d
s
c
o
n
s
i
d
e
r
t
h
e
g
o
o
d
c
h
a
r
a
c
t
e
r
i
s
t
i
c
s
o
f
m
o
r
e
t
h
a
n
o
n
e
t
e
c
h
n
i
q
ue
a
r
e
j
oi
ne
d
to
im
p
r
o
v
e
th
e
s
ig
n
if
ic
a
n
c
e
o
f
th
e
s
e
te
c
h
n
iq
u
e
s
[
5
]
.
T
h
e
y
le
a
r
n
w
h
ic
h
f
e
a
tu
r
e
s
c
o
n
tr
ib
u
te
th
e
b
e
s
t
to
th
e
pr
e
c
is
io
n
of
t
he
m
ode
l
,
w
he
n
t
he
m
ode
l
i
s
c
ons
t
r
uc
t
e
d.
F
e
a
t
ur
e
s
ubs
e
t
s
a
r
e
e
nha
nc
e
d
by
c
e
r
t
a
i
n
goodne
s
s
cr
i
t
er
i
a.
F
eat
u
r
es
ar
e
s
el
ect
ed
dur
i
ng
t
r
a
i
ni
ng,
but
i
t
i
s
done
s
e
pa
r
a
t
e
l
y
i
n
w
r
a
ppe
r
s
.
T
he
t
r
a
i
ni
ng
da
t
a
i
s
us
e
d
in
a
b
e
tte
r
w
a
y
to
e
v
a
lu
a
te
s
u
b
s
e
ts
b
y
n
o
t
r
e
q
u
ir
in
g
a
s
e
p
a
r
a
te
v
a
lid
a
tio
n
se
t
.
T
h
i
s
m
e
t
h
o
d
a
l
so
su
p
p
o
r
t
s
fa
s
t
t
ra
i
n
i
n
g
.
1.
2.
Ob
j
e
c
t
i
v
e
s
In
c
o
n
c
e
p
t
u
a
l
l
e
v
e
l
,
t
h
e
c
o
n
c
e
p
t
le
a
r
n
in
g
ta
s
k
is
d
iv
id
e
d
in
to
tw
o
s
u
b
ta
s
k
s
:
S
e
le
c
tio
n
o
f
f
e
a
tu
r
e
s
a
n
d
de
c
i
s
i
on
a
bout
f
e
a
t
ur
e
c
om
bi
na
t
i
on.
I
n
t
hi
s
obs
e
r
va
t
i
on,
t
he
obj
e
c
t
i
ve
of
t
hi
s
pa
pe
r
i
s
out
l
i
ne
d
a
s
be
l
ow
.
-
To
b
u
i
l
d
u
p
a
p
r
o
p
o
s
e
d
m
e
t
h
o
d
o
l
o
g
y
w
i
t
h
t
h
e
h
y
b
r
i
d
f
r
a
m
e
w
o
r
k
o
f
f
i
l
t
e
r
s
a
n
d
wr
a
p
p
e
r
a
n
d
t
o
p
e
r
f
o
r
m
ex
p
er
i
m
en
t
s
t
o
ev
al
u
at
e
t
h
e
p
er
f
o
r
m
an
ce
o
u
t
co
m
es
f
o
r
co
n
t
i
n
u
o
u
s
,
cat
eg
o
r
i
cal
an
d
h
y
b
r
i
d
d
at
a.
-
To
g
e
t
r
i
d
o
f
f
e
a
t
u
r
e
s
w
h
i
c
h
a
r
e
o
f
i
r
r
e
l
e
v
a
n
t
a
n
d
r
e
d
u
n
d
a
n
t
.
-
To
r
e
d
u
c
e
e
r
r
o
r
a
n
d
t
o
b
o
o
s
t
t
h
e
a
c
c
u
r
a
c
y
o
f
c
l
a
s
s
i
f
i
c
a
t
i
o
n
r
e
s
u
l
t
s
-
To
s
e
l
e
c
t
a
s
u
b
s
et
o
f
o
p
t
i
m
al
f
eat
u
r
es
f
r
o
m
t
h
e
en
t
i
r
e
s
et
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
IS
S
N
:
2252
-
8938
r
An
i
m
p
r
o
v
e
d
h
y
b
r
i
d
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
m
e
t
h
o
d
f
o
r
h
u
g
e
d
i
m
e
n
s
i
o
n
a
l
d
a
t
a
s
e
t
s
(
F.
Ro
s
i
t
a
K
a
m
a
l
a
)
79
1.
3
Li
t
e
r
a
t
u
r
e
R
e
v
i
e
w
Th
e
b
e
n
e
f
i
t
s
o
f
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
m
e
t
h
o
d
s
i
n
c
l
u
d
e
g
o
o
d
i
n
t
e
r
p
r
e
t
a
b
i
l
i
t
y
o
f
m
o
d
e
l
s
,
t
a
k
e
v
e
r
y
s
h
o
r
t
tr
a
in
in
g
c
o
m
p
u
ta
tio
n
tim
e
a
n
d
r
e
d
u
c
e
o
v
e
r
f
ittin
g
b
y
im
p
r
o
v
e
d
g
e
n
e
r
a
liz
at
i
o
n
i
n
cl
a
ssi
f
i
c
a
t
i
o
n
m
o
d
e
l
s.
In
t
h
e
m
o
s
t
re
c
e
n
t
d
e
c
a
d
e
s
,
s
e
v
e
ra
l
h
y
b
ri
d
t
e
c
h
n
i
q
u
e
s
b
a
s
e
d
o
n
P
a
rt
i
c
l
e
S
w
a
rm
O
p
t
i
m
i
z
a
t
i
o
n
(P
S
O
)
h
a
v
e
be
e
n
put
f
or
w
a
r
d,
w
i
t
h
be
t
t
e
r
out
c
om
e
s
.
A
c
om
bi
na
t
i
on
m
e
t
hod
i
nvol
vi
ng
P
S
O
a
nd
A
C
O
(A
n
t
Co
l
o
n
y
O
p
t
i
m
i
z
a
t
i
o
n
)
w
a
s
a
d
o
p
t
e
d
i
n
[
6
]
.
T
h
i
s
h
ybr
i
d
m
e
t
hod
ove
r
c
om
e
s
t
he
de
m
e
r
i
t
s
of
P
S
O
,
by
not
co
n
v
er
t
i
n
g
t
h
e
n
o
m
i
n
al
i
n
t
o
b
i
n
ar
y
an
d
g
et
r
i
d
o
f
t
h
e
p
r
ep
r
o
ces
s
i
n
g
p
h
as
e.
T
h
i
s
al
g
o
r
i
t
h
m
w
as
m
o
d
i
f
i
ed
t
o
PSO
/
A
C
O
1
f
o
r
o
p
t
i
m
i
z
i
n
g
b
o
t
h
t
h
e
c
o
n
t
i
n
u
o
u
s
a
n
d
n
o
m
i
n
a
l
a
t
t
r
i
b
u
t
e
s
a
n
d
PSO
/
A
C
O
1
f
o
r
m
a
n
a
g
i
n
g
co
n
t
i
n
u
ous
da
t
a
.
T
he
s
e
m
e
t
hods
pr
ove
be
t
t
e
r
pe
r
f
or
m
a
nc
e
i
n
ge
ne
r
a
t
i
ng
s
m
a
l
l
a
nd
s
i
m
pl
e
r
ul
e
s
e
t
s
.
N
e
kka
su
g
g
e
st
e
d
a
h
y
b
r
i
d
se
a
r
c
h
m
e
t
h
o
d
b
y
t
h
e
c
o
m
b
i
n
a
t
i
o
n
o
f
H
a
r
m
o
n
y
S
e
a
r
c
h
A
l
g
o
r
i
t
h
m
(
H
S
A
)
a
n
d
S
t
o
c
h
a
st
i
c
Lo
c
a
l
S
e
a
r
c
h
(
S
LS
)
f
o
r
l
e
a
r
n
i
n
g
p
r
o
b
l
e
m
s
i
n
m
a
c
h
i
n
e
l
e
a
r
n
i
ng
[
7]
.
T
hi
s
w
r
a
ppe
r
a
l
gor
i
t
hm
us
e
s
S
uppor
t
V
ect
o
r
M
ach
i
n
e
(
S
V
M
)
cl
as
s
i
f
i
er
.
Th
e
e
x
p
e
r
i
m
e
n
t
a
t
i
o
n
o
u
t
c
o
m
e
s
v
i
n
d
i
c
a
t
e
t
h
a
t
H
S
A
-
SL
S
i
s
g
i
v
i
n
g
b
e
t
t
e
r
re
s
u
l
t
s
t
h
a
n
H
S
A
a
n
d
G
e
n
e
t
i
c
A
l
g
o
ri
t
h
m
(G
A
).
A
b
d
u
l
l
a
h
S
a
e
e
d
G
h
a
re
b
e
t
a
l
.
p
ro
p
o
s
e
d
a
h
y
b
ri
d
m
e
t
h
o
d
o
l
o
g
y
to
im
p
ro
v
e
t
h
e
c
ro
s
s
o
v
e
r
a
n
d
m
u
t
a
t
i
o
n
o
p
e
ra
t
o
rs
o
f
G
A
i
n
c
o
n
s
i
d
e
ra
t
i
o
n
o
f
t
h
e
b
e
n
e
fi
t
s
o
f
fi
l
t
e
r
t
e
c
h
n
i
q
u
e
s
[8
].
In
t
h
e
n
e
x
t
s
t
e
p
,
s
u
b
s
e
t
s
o
f
d
i
ffe
re
n
t
s
i
z
e
s
a
n
d
i
m
p
o
rt
a
n
c
e
w
e
re
d
e
v
e
l
o
p
e
d
u
s
i
n
g
h
y
b
ri
d
a
p
p
ro
a
c
h
e
s
.
Hy
b
r
i
d
a
p
p
r
o
a
c
h
e
s
p
r
o
v
e
d
a
n
e
f
f
e
c
t
i
v
e
i
m
p
r
o
v
e
m
e
n
t
in
te
r
m
s
o
f
p
e
r
f
o
r
m
a
n
c
e
a
n
d
tim
e
.
A
f
e
f
B
e
n
B
r
a
h
im
e
t
al
.
s
u
g
g
es
t
ed
a
f
i
l
t
er
w
r
ap
p
er
h
y
b
r
i
d
m
et
h
o
d
b
y
s
el
ect
i
n
g
a
f
ew
n
u
m
b
er
o
f
f
eat
u
r
es
i
n
t
h
e
f
i
l
t
er
p
h
as
e,
b
as
ed
on
i
ns
t
a
nc
e
l
e
a
r
ni
ng
[
9]
.
I
n
t
he
s
e
c
ond
pha
s
e
,
a
c
oope
r
a
t
i
ve
s
ubs
e
t
s
e
a
r
c
h
w
a
s
us
e
d
a
s
a
w
r
a
ppe
r
a
nd
cl
as
s
i
f
i
cat
i
o
n
al
g
o
r
i
t
h
m
.
E
x
p
er
i
m
en
t
at
i
o
n
w
i
t
h
can
cer
d
at
as
et
s
p
r
o
v
ed
t
h
at
t
h
e
h
y
b
r
i
d
m
et
h
o
d
o
u
t
p
er
f
o
r
m
s
t
h
e
st
a
t
e
-
of
-
th
e
-
ar
t
al
g
o
r
i
t
h
m
s
.
Tw
o
h
y
b
r
i
d
F
S
a
l
g
o
r
i
t
h
m
s
,
w
h
i
c
h
a
r
e
s
i
m
p
l
e
a
n
d
e
f
f
i
c
i
e
n
t
c
o
m
b
i
n
e
t
h
e
w
r
a
p
p
e
r
m
e
t
h
o
d
u
s
i
n
g
a
Bi
n
a
r
y
D
i
f
f
e
r
e
n
t
ia
l
E
v
o
lu
tio
n
(
B
D
E
)
a
lg
o
r
ith
m
b
y
A
p
o
llo
n
i
e
t
a
l
[
1
0
]
.
E
x
p
e
r
im
e
n
ta
tio
n
s
w
ith
m
ic
r
o
a
r
r
a
y
da
t
a
s
e
t
s
l
e
s
s
e
n
t
he
qua
nt
i
t
y
of
t
he
c
hos
e
n
f
e
a
t
ur
e
s
e
f
f
e
c
t
i
ve
l
y,
r
e
s
ul
t
i
ng
a
be
t
t
e
r
a
c
c
ur
a
t
e
c
l
a
s
s
i
f
i
e
r
s
t
ha
n
i
n
t
he
ma
j
o
r
i
t
y
o
c
c
a
s
i
o
n
s
a
n
d
r
o
b
u
s
t
n
e
s
s
.
E
z
g
i
e
t
a
l
.
s
u
g
g
e
st
e
d
a
h
y
b
r
i
d
m
e
t
h
o
d
b
y
c
o
m
b
i
n
i
n
g
a
r
t
i
f
i
c
i
a
l
b
e
e
c
o
l
o
n
y
opt
i
m
i
z
a
t
i
on
m
e
t
hod
w
i
t
h
di
f
f
e
r
e
nt
e
vol
ut
i
on
a
l
gor
i
t
hm
s
f
or
c
l
a
s
s
i
f
i
c
a
t
i
on
ta
s
k
in
f
e
a
tu
r
e
s
e
le
c
tio
n
[
1
1
]
.
Th
i
s
s
t
u
d
y
r
e
v
e
a
l
s
t
h
a
t
t
h
e
h
y
b
r
i
d
m
e
t
h
o
d
e
n
h
a
n
c
e
s
t
h
e
c
l
a
s
s
i
f
i
c
a
t
i
o
n
r
e
s
u
l
t
s
a
n
d
r
u
n
t
i
m
e
.
A
tw
o
s
te
p
s
tr
a
te
g
y
is
w
e
ll
-
lik
e
d
in
f
e
a
tu
r
e
s
e
le
c
tio
n
o
n
la
r
g
e
d
im
e
n
s
io
n
s
o
f
d
a
ta
,
b
y
e
m
p
lo
y
in
g
f
ilte
r
s
to
m
in
im
iz
e
th
e
di
m
e
ns
i
ons
of
f
e
a
t
ur
e
s
.
A
f
ur
t
he
r
t
w
o
-
pha
s
e
i
n
w
hi
c
h
G
a
i
n
R
a
t
i
o
w
a
s
a
dopt
e
d
a
s
a
f
i
l
t
e
r
t
o
c
hoos
e
t
he
be
s
t
sc
o
r
e
d
d
i
m
e
n
si
o
n
s
a
n
d
c
o
m
b
i
n
e
d
w
i
t
h
ba
c
kw
a
r
d
e
l
i
m
i
na
t
i
on
a
l
gor
i
t
hm
pr
i
or
t
o
e
xe
c
ut
i
ng
P
S
O
by
R
os
i
t
a
e
t
al
.
O
u
t
co
m
es
v
i
n
d
i
cat
e
t
h
at
t
h
i
s
m
et
h
o
d
p
r
o
d
u
ced
a
b
et
t
er
cap
ab
i
l
i
t
y
f
o
r
t
h
e
n
u
m
er
i
cal
d
at
as
et
s
b
u
t
n
o
t
f
o
r
nom
i
na
l
a
nd
m
i
c
r
oa
r
r
a
y
da
t
a
s
e
t
s
w
i
t
h
m
or
e
e
xe
c
ut
i
on
t
i
m
e
[
12]
.
H
ui
j
ua
n
L
u
et
al
.
s
u
g
g
es
t
ed
a
h
y
b
r
i
d
i
zed
F
S
al
g
o
r
i
t
h
m
u
s
i
n
g
m
u
t
u
al
i
n
f
o
r
m
at
i
o
n
m
ax
i
m
i
zat
i
o
n
an
d
ad
ap
t
i
v
e
g
en
et
i
c
al
g
o
r
i
t
h
m
co
m
b
i
n
at
i
o
n
f
o
r
g
en
e
d
at
a
to
e
n
h
a
n
c
e
th
e
M
I
M
A
G
A
a
lg
o
r
ith
m
b
y
e
m
p
lo
y
in
g
f
o
u
r
c
la
s
s
if
ie
r
s
.
E
x
p
e
r
im
e
n
ta
tio
n
v
in
d
ic
a
te
s
th
a
t
th
e
accu
r
acy
r
at
es
f
o
r
al
l
d
at
as
et
s
p
r
o
v
ed
t
o
b
e
h
i
g
h
er
t
h
an
8
0
%
,
w
he
n
de
m
ons
t
r
a
t
e
d
t
he
r
obus
t
ne
s
s
of
t
he
a
l
gor
i
t
hm
[1
3
].
M
o
h
a
m
e
d
e
t
a
l
.
a
d
o
p
t
e
d
a
H
y
b
ri
d
B
i
n
a
ry
B
a
t
E
n
h
a
n
c
e
d
P
a
rt
i
c
l
e
S
w
a
rm
O
p
t
i
m
i
z
a
t
i
o
n
Al
g
o
r
i
t
h
m
wi
t
h
b
a
t
a
l
g
o
r
i
t
h
m
a
n
d
e
n
h
a
n
c
e
d
P
S
O
f
o
r
p
e
r
f
o
r
m
a
n
c
e
i
m
p
ro
v
e
m
e
n
t
s
i
n
UC
I
da
t
a
s
e
t
s
[
14]
.
To
e
n
h
a
n
c
e
t
h
e
k
-
Ne
a
r
e
s
t
Ne
i
g
h
b
o
u
r
c
l
a
s
s
i
f
i
e
r
(
k
NN)
,
P
S
O
i
s
u
p
d
a
t
e
d
wi
t
h
th
e
n
o
v
e
l
f
itn
e
s
s
f
u
n
c
tio
n
[
1
5
-
16]
.
Th
e
o
u
t
c
o
m
e
s
h
a
v
e
p
r
o
v
e
d
h
i
g
h
e
r
a
c
c
u
r
a
c
y
o
f
v
e
r
y
s
m
a
l
l
f
e
a
t
u
r
e
s
u
b
s
e
t
s
.
Li
t
e
r
a
t
u
r
e
s
t
u
d
y
r
e
v
e
a
l
s
t
h
a
t
d
i
f
f
e
r
e
n
t
h
y
b
r
i
d
m
e
t
h
o
d
s
h
av
e
b
een
p
r
o
p
o
s
ed
n
o
t
o
n
l
y
t
o
s
o
l
v
e
f
eat
u
r
e
se
l
e
c
t
i
o
n
i
ssu
e
s
b
u
t
a
l
so
t
o
r
e
so
l
v
e
o
p
t
i
m
i
z
a
t
i
o
n
p
r
o
b
l
e
m
s.
C
o
n
se
q
u
e
n
t
l
y
,
t
h
i
s
p
a
p
e
r
p
r
o
p
o
se
s
a
m
e
t
h
o
d
o
l
o
g
y
of
t
he
hybr
i
d
f
r
a
m
e
w
or
k
by
c
om
bi
ni
ng
t
he
m
e
r
i
t
s
of
f
i
l
t
e
r
a
nd
w
r
a
ppe
r
m
e
t
hods
f
or
F
S
pr
obl
e
m
s
i
n
m
a
c
hi
ne
le
a
r
n
in
g
.
I
n
th
is
p
a
p
e
r
,
th
e
f
ilte
r
s
c
h
i
s
q
u
a
r
e
,
F
-
st
a
t
i
st
i
c
,
a
n
d
m
u
t
u
a
l
i
n
f
o
r
m
a
t
i
o
n
,
a
n
d
w
r
a
p
p
e
r
P
S
O
a
r
e
ev
al
u
at
ed
.
T
h
e
p
r
o
p
o
s
ed
m
et
h
o
d
o
l
o
g
y
p
r
o
v
es
b
et
t
er
o
u
t
co
m
es
ag
ai
n
s
t
s
t
at
e
o
f
t
h
e
ar
t
o
f
t
h
e
f
i
el
d
m
en
t
i
o
n
ed
in
th
e
lite
r
a
tu
r
e
s
u
r
v
e
y
u
s
in
g
U
C
I
a
n
d
m
ic
ro
a
rra
y
d
a
t
a
s
e
t
s
.
T
h
e
re
s
u
l
t
s
o
b
t
a
i
n
e
d
p
ro
v
e
t
h
a
t
t
h
e
m
e
t
h
o
d
IH
F
S
is
f
o
u
n
d
to
b
e
th
e
b
e
s
t
r
e
g
a
r
d
in
g
a
c
c
u
r
a
c
y
,
a
n
d
c
o
m
p
u
ta
tio
n
tim
e
w
h
e
n
c
o
m
p
a
r
e
d
w
ith
th
e
v
a
r
ie
ty
o
f
m
e
th
o
d
s
in
th
e
e
x
is
tin
g
lite
r
a
tu
r
e
.
H
o
w
e
v
e
r
,
if
its
a
c
c
u
r
a
c
y
o
f
c
la
s
s
if
ie
r
s
is
im
p
r
o
v
e
d
w
it
h
t
he
r
un
t
i
m
e
,
i
t
i
s
c
ons
i
de
r
e
d
as
t
h
e
t
o
p
b
es
t
an
d
r
es
u
l
t
i
n
g
b
et
t
er
i
m
p
r
o
v
em
en
t
s
f
o
r
b
i
g
g
er
d
i
m
en
s
i
o
n
s
.
Th
e
o
r
g
a
n
i
z
a
t
i
o
n
o
f
t
h
i
s
p
a
p
e
r
i
s
a
s
fo
l
l
o
w
s
.
T
h
e
d
e
t
a
i
l
e
d
re
p
re
s
e
n
t
a
t
i
o
n
o
f
t
h
e
p
ro
p
o
s
e
d
h
y
b
ri
d
fra
m
ew
o
r
k
i
s
o
u
t
l
i
n
ed
i
n
s
ect
i
o
n
2
.
Th
e
e
x
p
e
r
i
m
e
n
t
a
t
i
on
out
c
om
e
s
of
t
he
pr
opos
e
d
m
e
t
hod
a
nd
t
he
c
om
pa
r
a
t
i
ve
s
t
udy
w
i
t
h
t
he
r
e
l
e
va
nt
w
or
ks
ar
e
o
u
t
l
i
n
ed
i
n
s
ect
i
o
n
3
.
I
n
t
h
e
l
as
t
p
ar
t
,
t
h
e
co
n
cl
u
d
i
n
g
o
b
s
er
v
at
i
o
n
s
an
d
t
h
e
i
n
f
o
r
m
at
i
o
n
f
o
r
f
u
t
u
r
e
s
co
p
e
ar
e
pr
e
s
e
nt
e
d
i
n
s
e
c
t
i
on
4.
2.
ME
T
H
O
D
A
N
D
MA
T
E
R
I
A
L
S
2.
1
. M
e
t
h
o
d
s
o
f
Th
e
Pr
o
p
o
s
e
d
W
o
r
k
In
t
h
i
s
p
a
p
e
r,
a
h
y
b
ri
d
m
e
t
h
o
d
i
s
fo
rm
u
l
a
t
e
d
fro
m
C
h
i
s
q
u
a
re
,
F
-
St
a
t
i
s
t
i
c
,
m
u
t
u
a
l
i
n
f
o
r
m
a
t
i
o
n
,
a
n
d
PSO
a
l
g
o
r
i
t
h
m
s
t
o
f
i
n
d
s
o
l
u
t
i
o
n
s
f
o
r
FS
pr
obl
e
m
s
i
n
m
a
c
hi
ne
l
e
a
r
ni
ng
t
a
s
ks
.
Evaluation Warning : The document was created with Spire.PDF for Python.
r
IS
S
N
:
2252
-
8938
IJ
-
AI
Vo
l
.
8
, N
o
.
1,
M
a
r
c
h
201
9
:
77
–
86
80
2.
1.
1.
Ch
i
-
Squa
r
e
T
e
s
t
(
C
H
I
)
CH
I
c
o
m
p
u
t
e
s
,
h
o
w
th
e
o
b
s
e
r
v
e
d
f
r
e
q
u
e
n
c
y
d
a
ta
v
a
lu
e
s
f
its
w
e
ll
w
ith
th
e
e
x
p
e
c
te
d
f
r
e
q
u
e
n
c
y
d
a
ta
va
l
ue
s
of
i
nde
pe
nde
nt
va
r
i
a
bl
e
s
.
S
o,
i
t
i
s
na
m
e
d
a
"
goodne
s
s
of
f
i
t
"
t
e
s
t
[1
7
]
.
T
h
e
c
h
i
s
q
u
a
r
e
a
ttr
ib
u
te
w
e
ig
h
t
ope
r
a
t
or
m
e
a
s
ur
e
s
t
he
a
t
t
r
i
but
e
s
w
e
i
ght
r
e
ga
r
di
ng
t
he
l
a
be
l
a
t
t
r
i
b
ut
e
by
m
e
a
ns
of
t
he
c
hi
s
qua
re
s
t
a
t
i
s
t
i
c
.
If
t
h
e
a
t
t
ri
b
u
t
e
w
e
i
g
h
t
i
s
h
i
g
h
e
r,
t
h
e
a
t
t
ri
b
u
t
e
s
a
re
c
o
n
s
i
d
e
re
d
m
o
re
re
l
e
v
a
n
t
.
N
o
rm
a
l
l
y
,
t
h
e
C
H
I
s
u
m
u
p
t
h
e
sq
u
a
r
e
s
o
f
t
h
e
d
i
sc
r
e
p
a
n
c
i
e
s
b
e
t
w
e
e
n
t
h
e
e
x
p
e
c
t
e
d
o
c
c
u
r
r
e
n
c
e
a
n
d
t
h
e
o
b
se
r
v
e
d
o
c
c
u
r
r
e
n
c
e
,
t
o
t
h
e
e
x
p
e
c
t
e
d
fre
q
u
e
n
c
y
o
f
o
u
t
c
o
m
e
s
[1
8
]
.
T
h
e
C
H
I
te
s
t
is
a
n
o
n
p
a
r
a
m
e
tr
ic
s
to
c
h
a
s
tic
p
r
o
c
e
d
u
r
e
.
T
h
e
r
e
a
r
e
s
o
m
e
ad
v
an
t
ag
es
o
f
n
o
n
p
ar
am
et
r
i
c
t
ech
n
i
q
u
es
.
I
t
i
s
f
ai
r
l
y
eas
y
t
o
co
m
p
u
t
e.
I
t
m
eas
u
r
es
d
at
a
o
n
t
h
e
b
as
i
s
o
f
cl
as
s
i
f
i
cat
i
o
n
.
T
he
C
H
I
te
s
t
can
o
n
l
y
b
e
ap
p
l
i
ed
f
o
r
l
ab
el
s
o
f
cat
eg
o
r
i
cal
v
ar
i
ab
l
es
.
He
n
c
e
,
th
is
s
tu
d
y
h
a
s
ta
k
e
n
C
H
I
to
ach
i
ev
e
b
et
t
er
r
es
u
l
t
s
.
2.
1.
2.
F
-
St
a
t
i
s
t
i
c
F
St
a
t
i
s
t
i
c
i
s
a
t
e
s
t
i
n
s
t
a
t
i
s
t
i
c
s
i
n
t
h
a
t
u
n
d
e
r
t
h
e
n
u
l
l
h
y
p
o
t
h
e
s
i
s
,
t
h
e
t
e
s
t
d
a
t
a
h
a
v
e
F
d
i
s
t
r
i
b
u
t
i
o
n
.
On
e
can
m
eas
ur
e
i
t
,
i
f
t
he
di
m
e
ns
i
on
i
s
num
e
r
i
c
[
1
9
].
H
o
w
e
v
e
r,
t
h
e
c
l
a
s
s
h
a
v
i
n
g
o
n
e
o
f
C
d
i
s
t
i
n
c
t
n
o
m
i
n
a
l
va
l
ue
s
i
s
m
e
nt
i
one
d
be
l
ow
.
≔
2
/
)
(
∈
)
)
/
)
(1
)
W
he
r
e
P
c
-
th
e
s
a
m
p
le
in
d
ic
e
s
p
a
r
titio
n
{
1
,2
,3
,....n
}
,
be
l
ongi
ng
to
th
e
p
a
r
titio
n
in
d
e
x
e
d
b
y
c
,
a
n
d
∶
=
∈
.
Th
i
s
f
o
r
m
u
l
a
c
o
r
r
e
s
p
o
n
d
s
to
th
e
f
r
a
c
tio
n
o
f
th
e
v
a
r
ia
tio
n
a
m
id
c
lu
s
te
r
s
a
n
d
th
e
me
a
n
v
a
r
i
a
n
c
e
i
n
s
i
d
e
t
h
e
c
l
u
s
t
e
r
s
.
L
a
r
g
e
r
r
e
l
e
v
a
n
c
e
i
mp
l
i
e
s
h
i
g
h
v
a
l
u
e
d
.
2.
1.
3.
Mu
t
u
a
l
In
f
o
r
m
a
t
i
o
n
(
M
I)
MI
c
o
m
p
u
t
e
s
t
h
e
f
r
e
q
u
e
n
t
i
n
f
o
r
m
a
t
i
o
n
a
m
i
d
a
n
y
t
w
o
f
e
a
t
u
r
e
s
t
h
a
t
a
r
e
a
r
b
i
t
r
a
r
y
i
n
n
a
t
u
r
e
,
i
f
b
o
t
h
t
h
e
di
m
e
ns
i
on
a
nd
t
he
c
l
a
s
s
ha
v
e
nom
i
na
l
va
l
ue
s
[
20
].
L
a
rg
e
r
v
a
l
u
e
s
s
h
o
w
s
u
p
e
ri
o
r
s
i
g
n
i
fi
c
a
n
c
e
.
A
d
i
m
e
n
s
i
o
n
s
e
t
D
=
{
d
1
,
d
2
,.....d
n
}
of
a
n
e
xa
m
pl
e
s
e
t
of
n
di
m
e
ns
i
ons
,
t
he
di
m
e
ns
i
on
r
e
duc
t
i
on
pr
oc
e
s
s
e
s
t
a
bl
i
s
he
s
a
s
ubs
e
t
P
wi
t
h
k
d
i
m
e
n
s
i
o
n
s
,
wh
e
r
e
k
≤
n
a
n
d
P
⊆
D.
T
h
e
r
e
f
o
r
e
,
P
t
h
e
s
u
b
s
e
t
s
h
o
u
l
d
y
i
e
l
d
e
q
u
a
l
o
r
s
u
p
e
r
i
o
r
a
c
c
u
r
a
c
y
o
f
cl
as
s
i
f
i
er
s
w
h
en
co
m
p
ar
ed
t
o
t
h
e
o
r
i
g
i
n
al
d
i
m
en
s
i
o
n
s
et
.
S
p
eci
f
i
cal
l
y
,
th
e
d
im
e
n
s
io
n
r
e
d
u
c
tio
n
d
e
f
in
e
s
th
e
su
b
se
t
o
f
d
i
m
e
n
si
o
n
s
t
h
a
t
i
m
p
r
o
v
e
s
M
I
w
i
t
h
a
n
out
put
c
l
a
s
s
C
i
s
M
I
(
P
,
C)
[
3
]
.
MI
(
X
,
Y)
=
H(
Y)
–
H(
Y|
X)
(2
)
wh
e
r
e
e
n
t
r
o
p
y
H(
)
,
X
a
n
d
Y
a
r
e
r
a
n
d
o
m
v
a
r
i
a
b
l
e
s
[2
1
]
.
De
f
i
n
i
t
i
o
n
1
.
Di
m
e
n
s
i
o
n
s
i
g
n
i
f
i
c
a
n
c
e
:
Di
m
e
n
s
i
o
n
d
i
is
m
o
r
e
s
ig
n
if
ic
a
n
t
to
th
e
o
u
tp
u
t
c
la
s
s
C
th
e
n
,
di
m
e
ns
i
on
d
j
in
th
e
p
e
r
s
p
e
c
tiv
e
o
f
th
e
c
h
o
s
e
n
s
u
b
s
e
t P
w
h
e
n
,
MI
(
d
i
, P
;C
)
>
M
I
(
d
j
,
P;
C
)
.
(3
)
2.
1.
4.
PS
O
Th
e
w
r
a
p
p
e
r
PSO
i
s
se
l
e
c
t
e
d
o
n
a
c
c
o
u
n
t
o
f
a
n
u
m
b
e
r
o
f
m
e
r
i
t
s
o
u
t
l
i
n
e
d
a
s
b
e
l
o
w
.
-
PSO
i
s
a
n
ev
o
l
u
t
i
o
n
ar
y
m
et
h
o
d
b
as
ed
o
n
p
o
p
u
l
at
i
o
n
.
-
Un
l
i
k
e
m
a
n
y
c
o
n
v
e
n
t
i
o
n
a
l
t
e
c
h
n
i
q
u
e
s
,
i
t
i
s
a
n
a
l
g
o
r
i
t
h
m
wi
t
h
f
e
we
r
d
e
r
i
v
a
t
i
o
n
s
an
d
a
l
o
w
co
m
p
u
t
at
i
o
n
tim
e
.
-
It
i
s
fl
e
x
i
b
l
e
t
o
i
n
t
e
g
ra
t
e
w
i
t
h
a
d
d
i
t
i
o
n
a
l
o
p
t
i
m
i
z
a
t
i
o
n
t
e
c
h
n
i
q
u
e
s
t
o
fo
rm
u
l
a
t
e
h
y
b
ri
d
m
e
t
h
o
d
s
.
In
1
9
9
5
,
Ja
m
e
s
K
e
n
n
e
d
y
a
n
d
R
u
sse
l
l
E
b
e
r
h
a
r
t
d
e
v
e
l
o
p
e
d
P
S
O
,
a
f
t
e
r
b
e
i
n
g
e
n
t
h
u
se
d
b
y
t
h
e
b
i
o
l
o
g
i
st
Fr
a
n
k
H
e
p
p
n
e
r
'
s
s
t
u
d
y
o
f
t
h
e
b
i
r
d
f
l
o
c
k
i
n
g
b
e
h
a
v
i
o
u
r
[
22
].
P
S
O
i
s
a
m
e
t
h
o
d
o
l
o
g
y
t
o
d
i
s
c
o
v
e
r
s
o
l
u
t
i
o
n
s
t
o
pr
obl
e
m
s
a
nd
s
pe
c
i
f
i
e
d
a
s
a
l
oc
us
i
n
a
s
ol
ut
i
on
s
pa
c
e
of
n
di
m
e
ns
i
ons
.
A
c
l
us
t
e
r
of
a
r
bi
t
r
a
r
y
s
pe
c
ks
(s
o
l
u
t
i
o
n
s
)
i
n
i
t
i
a
l
i
z
e
s
P
S
O
s
e
a
rc
h
fo
r
o
p
t
i
m
u
m
v
a
l
u
e
s
,
b
y
mo
d
i
f
y
i
n
g
p
r
o
p
a
g
a
t
i
o
n
s
.
A
l
a
r
g
e
n
u
mb
e
r
o
f
pa
r
t
i
c
l
e
s
a
r
e
c
hos
e
n
i
nt
o
a
c
t
i
on
t
hr
ough
t
hi
s
s
pa
c
e
r
a
ndom
l
y.
T
he
y
e
xa
m
i
ne
t
he
"
f
i
t
ne
s
s
"
of
t
he
s
e
pa
r
t
i
c
l
e
s
an
d
t
h
ei
r
ne
i
ghbour
s
i
n
e
a
c
h
i
t
e
r
a
t
i
on
t
o
"
e
m
ul
a
t
e
"
t
hr
i
vi
ng
ne
i
ghbour
s
by
a
dva
nc
i
ng
t
ow
a
r
ds
t
he
m
[
12
].
Th
e
r
e
a
r
e
d
i
f
f
e
r
e
n
t
m
e
t
h
o
d
s
t
o
g
r
o
u
p
p
a
r
t
i
c
l
e
s
i
n
t
o
c
h
a
l
l
e
n
g
i
n
g
s
e
m
i
-
in
d
e
p
e
n
d
e
n
t
f
lo
c
k
s
o
r
a
s
in
g
le
g
lo
b
a
l
fl
o
c
k
,
i
n
c
l
u
d
i
n
g
a
l
l
t
h
e
p
a
rt
i
c
l
e
s
t
h
a
t
b
e
l
o
n
g
.
T
h
i
s
s
e
em
s
t
o
b
e
v
er
y
ef
f
ect
i
v
e
acr
o
s
s
t
h
e
d
i
f
f
er
en
t
p
r
o
b
l
em
dom
a
i
ns
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
IS
S
N
:
2252
-
8938
r
An
i
m
p
r
o
v
e
d
h
y
b
r
i
d
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
m
e
t
h
o
d
f
o
r
h
u
g
e
d
i
m
e
n
s
i
o
n
a
l
d
a
t
a
s
e
t
s
(
F.
Ro
s
i
t
a
K
a
m
a
l
a
)
81
2.
2
. T
h
e
Pr
o
p
o
s
e
d
IH
F
S
Al
g
o
r
i
t
h
m
.
Th
e
p
o
w
e
r
s
o
f
f
i
l
t
e
r
s
a
n
d
w
r
a
p
p
e
r
s
a
r
e
c
o
m
b
i
n
e
d
i
n
t
h
e
i
m
p
r
o
v
e
d
h
y
b
r
i
d
Fe
a
t
u
r
e
Se
l
e
c
t
i
o
n
al
g
o
r
i
t
h
m
[
5
-
14]
.
A
n
o
v
e
l
f
r
a
m
e
w
o
r
k
,
n
a
m
e
d
th
e
I
m
p
r
o
v
e
d
H
y
b
r
id
Fe
a
t
u
r
e
Se
l
e
c
t
i
o
n
(IH
F
S
)
M
e
t
h
o
d
i
s
de
ve
l
ope
d
.
T
h
e
p
r
o
p
o
s
e
d
I
H
F
S
a
lg
o
r
ith
m
'
s
p
s
e
u
d
o
c
o
d
e
is
s
u
m
m
a
r
iz
e
d
in
F
ig
u
r
e
2
.
Li
n
e
s
1
-
22
a
c
c
ount
f
or
t
he
pha
s
e
-
1
of
I
H
F
S
.
I
n
t
he
f
i
r
s
t
pha
s
e
,
th
e
tw
o
f
ilte
r
m
e
th
o
d
s
a
r
e
s
e
le
c
te
d
a
s
th
e
in
itia
l
p
a
r
titio
n
,
to
e
lim
in
a
te
th
e
mo
s
t
u
n
n
e
e
d
e
d
o
r
e
x
t
r
a
n
e
o
u
s
f
e
a
t
u
r
e
s
.
C
h
i
s
q
u
a
r
e
(
C
H
I
)
,
F
-
St
a
t
i
s
t
i
c
(
FSt
a
t
)
,
a
n
d
M
u
t
u
a
l
I
n
f
o
r
m
a
t
i
o
n
(
M
I
)
a
r
e
th
e
c
e
n
tr
a
l
p
a
r
t
o
f
in
itia
l
s
e
le
c
tio
n
.
T
h
e
s
e
w
e
ig
h
tin
g
f
u
n
c
tio
n
s
a
r
e
s
e
le
c
te
d
,
b
a
s
e
d
o
n
th
e
ty
p
e
s
o
f
d
a
ta
at
t
r
i
b
u
t
es
l
i
k
e
n
u
m
er
i
cal
o
r
co
n
t
i
n
u
o
u
s
,
nom
i
na
l
or
di
s
c
r
e
t
e
a
nd
c
l
a
s
s
l
a
be
l
va
l
ue
s
l
i
ke
F
S
t
a
t
f
or
num
e
r
i
c
di
m
e
ns
i
ons
a
nd
M
I
f
or
nom
i
na
l
di
m
e
ns
i
ons
.
T
he
m
e
t
hods
out
l
i
ne
d
a
bove
,
e
va
l
ua
t
e
t
he
i
m
por
t
a
nc
e
of
t
he
fe
a
t
u
re
s
b
y
e
v
a
l
u
a
t
i
n
g
fo
r
e
v
e
ry
fe
a
t
u
re
o
f
t
h
e
d
a
t
a
s
e
t
,
t
h
e
i
m
p
o
rt
a
n
c
e
o
f
t
h
e
fi
l
t
e
r
m
e
t
hod
w
i
t
h
r
e
f
e
r
e
nc
e
t
o
th
e
c
la
s
s
la
b
e
l.
T
h
e
w
e
ig
h
ts
o
f
f
e
a
tu
r
e
s
w
h
ic
h
c
o
n
te
n
te
d
th
e
p
r
e
c
is
e
c
o
n
d
itio
n
w
ith
r
e
f
e
r
e
n
c
e
to
th
e
w
e
ig
h
ts
o
f
in
p
u
t
f
e
a
tu
r
e
s
a
r
e
s
e
le
c
te
d
f
o
r
th
e
d
a
ta
s
e
ts
.
T
h
e
f
ir
s
t
s
e
le
c
tio
n
f
hi
gh
is
d
o
n
e
b
y
C
H
I
.
Th
e
s
e
c
o
n
d
s
e
l
e
c
t
i
o
n
f
1
or
f
2
is
d
one
by
e
i
t
he
r
F
S
t
a
t
or
M
I
.
F
r
om
a
l
l
e
xi
s
t
i
ng
f
e
a
t
ur
e
s
,
t
he
s
e
f
e
a
t
ur
e
s
a
r
e
m
e
a
s
ur
e
d
a
s
t
he
m
os
t
r
e
l
e
va
nt
la
b
e
l
a
s
s
o
c
ia
te
d
f
e
a
tu
r
e
s
.
T
o
a
r
r
iv
e
a
t
b
e
tte
r
r
e
s
u
lts
,
th
e
f
r
a
m
e
w
o
r
k
is
to
s
u
c
c
e
s
s
f
u
lly
c
o
n
jo
in
th
e
tw
o
v
a
r
ia
b
le
su
b
se
t
s,
w
h
i
c
h
e
l
i
m
i
n
a
t
e
s
t
h
e
f
r
e
q
u
e
n
t
fe
a
t
u
re
s
o
f
b
o
t
h
t
h
e
s
e
t
s
.
T
h
u
s
v
a
ri
a
b
l
e
s
e
l
e
c
t
i
o
n
i
s
a
c
h
i
e
v
e
d
t
o
l
e
sse
n
th
e
d
im
e
n
s
io
n
o
f
fe
a
t
u
re
s
.
Th
e
t
w
o
r
e
s
u
l
t
a
n
t
f
e
a
t
u
r
e
s
u
b
s
e
t
s
o
f
t
h
e
f
i
r
s
t
p
h
a
s
e
h
a
v
e
u
n
d
e
r
g
o
n
e
a
c
o
m
b
i
n
a
t
i
o
n
mo
d
e
l
t
o
r
e
mo
v
e
unr
e
l
a
t
e
d
f
e
a
t
ur
e
s
an
d
o
v
er
f
i
t
t
i
n
g
f
o
r
t
h
e
f
i
n
et
u
n
ed
p
r
ep
ro
c
e
s
s
e
d
fe
a
t
u
re
s
u
b
s
e
t
.
A
m
e
rg
e
pr
oc
e
dur
e
,
f
or
e
xa
m
pl
e
t
he
uni
on
i
s
opt
e
d.
A
s
a
n
out
c
om
e
,
t
he
num
be
r
of
f
e
a
t
ur
e
s
a
r
e
e
l
i
m
i
na
t
e
d
t
o
a
n
ap
p
r
eci
ab
l
e
m
an
n
er
.
T
h
u
s
,
t
h
e
p
r
el
i
m
i
n
ar
y
p
r
ep
r
o
ces
s
i
n
g
s
t
ep
i
s
p
er
f
o
r
m
ed
f
o
r
U
C
I
an
d
m
i
cr
o
ar
r
ay
re
p
o
s
i
t
o
ry
d
a
t
a
s
e
t
s
.
B
y
co
m
b
i
n
i
n
g
t
h
e
r
es
u
l
t
an
t
p
r
ep
r
o
ces
s
ed
f
eat
u
r
e
s
u
b
s
et
s
i
s
n
o
t
an
ex
cel
l
en
t
ch
o
i
ce
as
t
h
e
re
m
o
v
a
l
o
f
re
d
u
n
d
a
n
c
y
w
i
l
l
l
e
a
d
t
o
a
n
i
m
p
ro
v
e
m
e
n
t
o
f
a
c
c
u
ra
c
y
.
In
p
u
t
:
Se
t
o
f
f
e
a
t
u
r
e
s
F
={
f
i
, i =
1
.....n
}
C:
Cl
a
s
s
l
a
b
e
l
s
.
Ou
t
p
u
t
:
S
Se
l
e
c
t
e
d
f
e
a
t
u
r
e
s
1.
S
←
{}
2.
F
or
each
f
i
in
F
3.
W
e
i
ght
(
f
i
)
=
F
i
n
d
C
h
i
(f
i
,C
)
4.
E
nd
f
or
5.
S
or
t
w
e
i
ght
i
n
a
c
c
or
da
nc
e
w
i
t
h
w
e
i
ght
.
6.
f
hi
gh
=
C
h
o
o
s
e
f
e
a
t
u
r
e
s
w
i
t
h
m
o
r
e
w
e
i
g
h
t
.
7.
A
ppe
nd
f
hi
gh
to
S
. (
S
←
S
∪
{f
hi
gh
})
8.
i
f
F
i
s
N
um
e
r
i
c
9.
F
or
e
a
c
h
f
i
in
F
10.
W
e
i
ght
(
f
i
)
=
F
i
n
d
F
S
t
a
t
(f
i
,C
)
u
s
in
g
Eq
n
(
1
)
.
11.
E
nd
f
or
12.
S
or
t
w
e
i
ght
i
n
a
c
c
or
da
nc
e
w
i
t
h
w
e
i
ght
.
13.
f
1
=
C
h
o
o
s
e
f
e
a
t
u
r
e
s
w
i
t
h
m
o
r
e
w
e
i
g
h
t
.
14.
A
ppe
nd
f
1
to
S
. (
S
←
S
∪
{f
1
})
15.
e
l
s
e
i
f
F
i
s
N
om
i
na
l
16.
F
or
e
a
c
h
f
i
in
F
17.
W
e
i
ght
(
f
i
)
=
F
i
n
d
M
I(f
i
, C
)
u
s
in
g
E
q
n
2
.
18.
E
nd
f
or
19.
S
or
t
w
e
i
gh
t in
a
c
c
o
r
d
a
n
c
e
w
ith
w
e
ig
h
t.
20.
f
2
=
C
h
o
o
s
e
f
e
a
t
u
r
e
s
w
i
t
h
m
o
r
e
w
e
i
g
h
t
.
21.
A
ppe
nd
f
2
to
S
. (
S
←
S
∪
{f
2
})
22.
E
nd
i
f
23.
D
i
vi
de
S
i
nt
o
da
t
a
s
e
t
s
f
or
t
r
a
i
ni
ng
a
nd
t
e
s
t
i
ng.
Sw
a
r
m
i
n
i
t
i
a
l
i
z
e
d
.
24.
S
pe
c
i
f
y
t
he
m
a
xi
m
um
i
t
e
r
a
t
i
ons
I
max
25.
F
or
t
he
f
e
a
t
ur
e
s
e
t
S
ge
n
er
at
e
p
ar
t
i
cl
es
P
t
h
ei
r
pos
i
t
i
ons
P
os
(
m
,
n)
a
nd
ve
l
oc
i
t
y
V
e
l
(
m
,
n)
26.
C
om
put
e
t
he
m
a
xi
m
um
i
m
pe
da
nc
e
pe
a
k
ma
x
_
i
mp
e
d
(
m)
c
o
r
r
e
s
p
o
n
d
i
n
g
t
o
m
p
a
r
t
i
c
l
e
s
.
27.
D
e
s
c
r
i
be
t
he
be
s
t
l
oc
a
l
pa
r
t
i
c
l
e
l
_be
s
t
a
nd
f
i
nd
t
he
be
s
t
gl
oba
l
pa
r
t
i
c
l
e
g_be
s
t
.
28.
W
hi
l
e
(
t
≤
I
m
a
x
)
29.
F
or
e
a
c
h
pa
r
t
i
c
l
e
m
=
1
:
P
30.
U
pda
t
e
i
ne
r
t
i
a
w
e
i
ght
ω
,
ve
l
oc
i
t
y,
pos
i
t
i
on.
31.
ω
=
(
ω
m
-
ω
f
)
I
max
-
1/
I
max
//
ω
to
m
a
n
a
g
e
th
e
in
f
lu
e
n
c
e
o
f
th
e
p
r
e
c
e
d
in
g
v
e
lo
c
itie
s
on
t
he
pr
e
s
e
nt
ve
l
oc
i
t
y.
32.
V
e
l
(
m
,
n)
=
ω
(
t
)
V
e
l
(
m
,
n)
+
C
1
r
1
(l
_
b
e
s
t
(m
,
n
)
-
Po
s
(
m
,
n
)
+
C
2
r
2
(g
_
b
e
s
t
)
-
Po
s
(
m
,
n
)
C
1
-
co
g
n
i
zan
ce
f
act
o
r
f
o
r
l
ear
n
i
n
g
,
C
2
-
So
c
i
a
l
f
a
c
t
o
r
f
o
r
le
a
r
n
in
g
, r
1
, r
2
-
Un
i
f
o
r
ml
y
g
e
n
e
r
a
t
e
d
r
a
n
d
o
m
n
u
mb
e
r
s
in
th
e
r
a
n
g
e
[
0
, 1
]
.
33.
P
os
(
m
,
n)
=
P
os
(
m
,
n)
+
V
e
l
(
m
,
n)
34.
W
i
t
hi
n
t
he
l
ow
e
r
a
nd
uppe
r
bounds
l
i
m
i
t
t
he
pos
i
t
i
ons
a
nd
ve
l
oc
i
t
y
35.
E
nd
f
or
36.
U
pda
t
e
l
_be
s
t
(
m
,
n)
a
nd
g_be
s
t
(
m
,
n)
a
c
c
or
di
ngl
y.
37.
t
←
t
+
1
38.
E
nd
w
hi
l
e
39.
R
e
t
ur
n
t
he
be
s
t
pa
r
t
i
c
l
e
a
s
s
ol
ut
i
on.
40.
C
r
os
s
V
a
l
i
da
t
i
on(
C
V
)
f
or
t
he
num
be
r
of
ne
a
r
e
s
t
ne
i
ghbor
s
k
m
i
n
an
d
k
max
41.
T
r
a
i
ni
ng
s
e
t
ha
vi
ng
C
V
w
i
t
h
va
l
ue
s
1,
2,
3,
.
.
.
.
.
V
.
42.
F
or
e
a
c
h
k
∈
[k
min
,
k
max
]
43.
C
om
put
a
t
i
on
of
a
ve
r
a
ge
e
r
r
or
r
a
t
e
V
CV
k
=
∑
e
v
/V
v=
1
44.
e
v
-
er
r
o
r
r
at
e,
O
p
t
i
m
al
k
=
ar
g
{
m
i
n
C
V
k
: k
m
i
n
≤
k
≤
k
max
}
45.
C
l
a
s
s
i
f
i
c
a
t
i
on
r
e
s
ul
t
s
pr
e
di
c
t
i
on
pe
r
f
or
m
a
nc
e
f
or
S
as
fi
n
a
l
s
e
l
e
c
t
e
d
s
u
b
s
e
t
o
f
fe
a
t
u
re
s
.
Fi
g
u
r
e
2
.
T
h
e
p
s
e
u
d
o
c
o
d
e
o
f
t
h
e
p
r
o
p
o
s
e
d
I
H
FS
a
l
g
o
r
i
t
h
m
Li
n
e
s
2
3
-
39
c
or
r
e
s
pond
t
o
a
s
e
c
ond
pha
s
e
of
t
he
P
S
O
w
r
a
ppe
r
a
ppr
oa
c
h
t
o
e
xpl
or
e
f
or
t
he
opt
i
m
a
l
fe
a
t
u
re
s
u
b
s
e
t
i
n
t
h
e
fe
a
t
u
re
s
p
a
c
e
S
.
E
v
e
ry
p
a
rt
i
c
l
e
i
s
n
e
e
d
t
o
b
e
m
o
d
i
fi
e
d
b
y
t
h
e
t
w
o
"
fi
n
e
s
t
"
v
a
l
u
e
s
fo
r
e
a
c
h
Evaluation Warning : The document was created with Spire.PDF for Python.
r
IS
S
N
:
2252
-
8938
IJ
-
AI
Vo
l
.
8
, N
o
.
1,
M
a
r
c
h
201
9
:
77
–
86
82
ite
r
a
tio
n
.
T
h
e
p
a
r
tic
le
s
w
a
r
m
o
p
tim
iz
e
r
tr
a
c
e
s
th
e
s
u
b
s
e
q
u
e
n
t
th
e
be
s
t
r
e
s
ul
t
a
t
t
a
i
ne
d
unt
i
l
now
by
a
f
e
w
el
em
en
t
s
i
n
t
h
e
p
o
p
u
l
at
i
o
n
ar
e
r
ef
er
r
ed
g
b
es
t
,
g
l
o
b
al
l
y
b
es
t
.
W
h
en
ev
er
s
p
eck
s
p
ar
t
i
ci
p
at
e
as
t
o
p
o
l
o
g
i
cal
ne
i
ghbour
s
i
n
t
he
popul
a
t
i
on,
t
he
f
i
ne
s
t
r
e
s
ul
t
i
s
c
a
l
l
e
d
l
be
s
t
,
l
oc
a
l
l
y
be
s
t
.
T
he
opt
i
m
a
l
s
ear
ch
u
s
i
n
g
P
S
O
w
i
t
h
im
p
r
o
v
e
m
e
n
t
in
V
e
lo
c
ity
V
e
l(
m
,
n
)
a
n
d
th
e
s
e
a
r
c
h
s
p
a
c
e
P
o
s
(
m
,
n
)
o
f
e
v
e
r
y
p
a
r
tic
le
is
g
iv
e
n
in
lin
e
s
3
2
-
33.
Fi
n
a
l
l
y
,
L
i
n
e
s
4
0
-
45
c
or
r
e
s
pond
t
o
t
he
c
ons
e
que
nt
i
a
l
f
e
a
t
ur
e
s
e
t
i
s
a
ppl
i
e
d
by
a
dopt
i
ng
10
f
ol
d
C
r
os
s
Va
l
i
d
a
t
i
o
n
(
C
V)
a
n
d
s
t
r
a
t
i
fi
e
d
s
a
m
p
l
i
n
g
t
o
a
d
v
a
n
c
e
t
h
e
c
l
a
s
s
i
fi
e
rs
a
c
c
u
ra
c
y
.
T
h
i
s
i
s
t
h
e
e
x
c
e
l
l
e
n
t
t
u
n
i
n
g
s
t
e
p
to
y
ie
ld
th
e
f
in
e
s
t
f
e
a
tu
r
e
s
e
t.
IH
F
S
i
s
v
e
ry
e
ffi
c
i
e
n
t
i
n
e
l
i
m
i
n
a
t
i
n
g
t
h
e
i
rre
l
e
v
a
n
t
a
n
d
u
s
e
l
e
s
s
fe
a
t
u
re
s
.
Be
c
a
u
s
e
,
t
h
e
m
a
j
o
r
i
t
y
o
f
t
h
e
i
n
s
i
g
n
i
f
i
c
a
n
t
f
e
a
t
u
r
e
s
a
r
e
r
u
l
e
d
o
ut
,
s
ubs
e
que
nt
t
o
t
he
f
i
r
s
t
s
t
e
p
of
t
he
f
i
l
t
e
r
me
t
h
o
d
.
I
t
a
l
s
o
e
l
i
mi
n
a
t
e
s
t
h
e
e
x
p
o
n
e
n
t
i
a
l
c
a
l
c
u
l
a
t
i
o
n
p
r
o
b
l
e
m
o
f
w
r
a
p
p
e
r
a
p
p
r
o
a
c
h
i
n
t
h
e
su
b
se
q
u
e
n
t
st
e
p
.
Th
e
e
x
p
e
r
i
m
e
n
t
a
l
r
e
s
u
l
t
s
r
e
c
o
m
m
e
n
d
t
h
a
t
t
h
e
I
H
F
S
w
o
r
k
s
w
e
l
l
o
n
a
n
e
x
t
e
n
s
i
v
e
r
a
n
g
e
o
f
p
r
o
b
l
e
m
s
.
2.
3
.
k
-
Ne
a
r
e
s
t
Ne
i
g
h
b
o
u
r
Cl
a
s
s
i
f
i
c
a
t
i
o
n
(
k
-
NN)
On
e
o
f
t
h
e
m
o
s
t
c
o
m
m
o
n
n
o
n
p
a
r
a
m
e
t
r
i
c
m
e
t
h
o
d
s
i
s
t
h
e
k
-
NN[
1
5
]
.
T
h
e
o
n
e
a
n
d
o
n
ly
p
a
r
a
m
e
te
r
,
k
is
th
e
q
u
a
n
tity
o
f
n
e
a
r
b
y
n
e
ig
h
b
o
u
r
s
,
c
a
n
b
e
d
e
te
r
m
in
e
d
a
n
d
im
p
le
m
e
n
te
d
e
a
s
ily
.
T
h
e
k
e
y
p
e
r
f
o
r
m
a
n
c
e
o
f
th
e
cl
as
s
i
f
i
cat
i
o
n
pr
oc
e
s
s
i
s
t
he
num
be
r
of
ne
a
r
by
ne
i
ghbour
s
.
k
-
NN
c
a
l
c
u
l
a
t
e
s
t
h
e
m
i
n
i
m
u
m
d
i
s
t
a
n
c
e
c
a
l
l
e
d
Eu
c
l
i
d
e
a
n
d
i
s
t
a
n
c
e
,
i
s
a
n
e
w
e
n
t
i
t
y
o
f
t
e
s
t
s
a
m
p
l
e
s
f
r
o
m
t
h
e
t
r
a
i
n
i
n
g
s
a
m
p
l
e
s
[
16
].
T
o
e
n
h
a
n
c
e
t
h
e
l
e
a
rn
i
n
g
pe
r
f
or
m
a
nc
e
,
t
he
va
r
i
a
bl
e
k
m
us
t
be
m
odi
f
i
e
d
i
n
a
c
c
or
da
nc
e
wi
t
h
t
h
e
v
a
ri
o
u
s
d
a
t
a
s
e
t
s
d
i
s
t
i
n
c
t
i
v
e
n
e
s
s
.
To
c
l
a
s
s
i
f
y
n
s
a
m
p
l
e
s
,
t
h
e
y
a
r
e
s
u
b
d
i
v
i
d
e
d
i
n
t
o
o
n
e
t
e
s
t
i
n
s
t
a
n
c
e
a
n
d
n
-
1
t
r
a
i
ni
ng
i
ns
t
a
nc
e
s
i
n
e
ve
r
y
i
t
e
r
a
t
i
ve
pr
oc
e
s
s
of
e
va
l
ua
t
i
on.
Th
e
s
u
b
s
e
q
u
e
n
t
s
t
e
p
s
a
r
e
r
e
q
u
i
r
e
d
t
o
p
e
r
f
o
r
m
k
-
NN.
-
Se
l
e
c
t
i
n
g
k
:
A
f
t
e
r
s
o
m
e
i
t
er
at
i
o
n
s
,
k
v
al
u
e
i
s
as
cer
t
ai
n
ed
b
y
t
h
e
f
i
n
es
t
r
es
u
l
t
.
-
Ca
l
c
u
l
a
t
i
n
g
D
i
s
p
l
a
c
e
m
e
n
t
:
E
u
c
l
i
d
e
a
n
d
i
s
t
a
n
c
e
m
e
t
h
o
d
i
s
a
d
o
p
t
e
d
.
-
So
r
t
i
n
g
d
i
s
t
a
n
c
e
i
n
a
s
c
e
n
d
i
n
g
o
r
d
e
r
:
T
h
e
m
i
n
i
m
u
m
k
d
i
s
t
a
n
c
e
s
a
r
e
f
o
u
n
d
b
y
t
h
e
d
i
s
t
a
n
c
e
s
o
r
t
e
d
i
n
as
cen
d
i
n
g
o
r
d
er
.
3.
RE
S
UL
T
S
A
ND
ANAL
YS
I
S
To
a
s
s
e
s
s
t
h
e
p
r
o
p
o
s
e
d
m
e
t
h
o
d
o
l
o
g
y
,
a
n
a
n
a
l
y
s
i
s
i
s
p
e
r
f
o
r
m
e
d
p
r
a
g
m
a
t
i
c
a
l
l
y
.
To
a
c
c
o
m
p
l
i
s
h
an
al
y
s
i
s
,
an
o
p
er
at
i
o
n
al
v
er
s
i
o
n
o
f
t
h
e
p
r
o
p
o
s
ed
al
g
o
r
i
t
h
m
w
as
ex
ecu
t
ed
in
J
a
v
a
,
w
ith
th
e
R
a
p
id
M
in
e
r
te
c
h
n
o
lo
g
y
.
I
n
tu
r
n
to
r
e
a
c
h
th
e
b
e
s
t
e
v
a
lu
a
tio
n
o
f
th
e
f
itn
e
s
s
fu
n
c
t
i
o
n
,
a
1
0
fo
l
d
s
t
ra
t
i
fi
e
d
C
V
m
e
t
h
o
d
i
s
fo
l
l
o
w
e
d
i
n
a
l
l
c
l
a
s
s
i
fi
e
rs
.
3.
1.
D
at
as
e
t
s
an
d
Pa
r
a
m
e
t
e
r
S
e
t
t
i
n
g
In
t
h
i
s
s
t
u
d
y
,
ei
g
h
t
da
t
a
s
e
t
s
a
r
e
us
e
d
f
r
om
U
C
I
r
e
pos
i
t
or
y
[
2
3
]
t
o
c
o
n
fi
rm
t
h
e
s
u
c
c
e
s
s
,
a
n
d
e
ffi
c
i
e
n
c
y
of
t
he
a
l
gor
i
t
hm
s
i
s
de
s
c
r
i
be
d
i
n
T
a
bl
e
1.
Th
e
e
x
p
e
r
i
m
e
n
t
s
a
r
e
a
c
c
o
m
p
l
i
s
h
e
d
o
n
d
a
t
a
s
e
t
s
w
i
t
h
m
o
r
e
t
h
a
n
3
0
fe
a
t
u
re
s
.
T
h
e
a
l
g
o
ri
t
h
m
IH
F
S
i
s
a
l
s
o
e
x
p
e
ri
m
e
n
t
e
d
w
i
t
h
s
i
x
w
e
l
l
-
know
n
m
i
c
r
oa
r
r
a
y
da
t
a
s
e
t
s
[
2
4
]
s
u
m
m
a
ri
z
e
d
br
i
e
f
l
y
i
n
T
a
bl
e
2.
Ta
b
l
e
1
.
Ex
p
e
r
i
m
e
n
t
a
l
d
a
t
a
s
e
t
i
n
f
o
r
m
a
t
i
o
n
.
Da
t
a
s
e
t
s
#S
#F
e
a
t
ur
e
s
#C
To
t
a
l
Nu
me
r
i
c
No
mi
n
a
l
Gl
a
s
s
214
9
9
0
6
So
n
a
r
208
60
60
0
2
Ve
h
i
c
l
e
846
18
18
0
4
Io
n
o
s
p
h
e
re
(Io
n
o
)
351
34
34
0
2
Ch
e
s
s
3196
36
0
36
2
Sp
l
i
c
e
3190
61
0
61
3
He
p
a
t
i
t
i
s
155
19
6
13
2
Ly
m
p
h
148
18
3
15
4
#S
(
N
o.
of
S
a
m
pl
e
s
)
#C
(
N
o.
of
Cl
a
s
s
e
s
)
Ta
b
l
e
2
.
H
i
g
h
d
i
m
e
n
s
i
o
n
a
l
mi
c
r
o
a
r
r
a
y
d
a
t
a
s
e
t
s
.
Da
t
a
s
e
t
s
#G
e
ne
s
#S
#C
Ov
a
r
i
a
n
C
a
n
c
e
r
(O
v
a
ri
a
n
)
15154
253
2
ML
L
12582
72
3
Le
u
k
e
m
i
a
2
C
(L
e
u
k
-
2C
)
7129
72
2
Lu
n
g
C
a
n
c
e
r
(L
u
n
g
)
12533
181
2
CN
S
7129
60
2
(L
e
u
k
-
4C
)
7129
72
4
#S
(
N
o.
of
S
a
m
pl
e
s
)
#C
(
N
o.
of
C
l
a
s
s
e
s
)
Th
e
s
e
t
t
i
n
g
o
f
p
a
r
a
m
e
t
e
r
s
f
o
r
P
S
O
i
s
a
s
f
o
l
l
o
w
s
.
P
o
p
u
l
a
t
i
o
n
s
i
z
e
i
s
1
0
0
.
U
p
p
e
r
l
i
m
i
t
n
u
m
b
e
r
o
f
ge
ne
r
a
t
i
ons
i
s
30.
T
he
pa
r
a
m
e
t
e
r
s
l
i
ke
i
ne
r
t
i
a
w
e
i
ght
,
l
oc
a
l
be
s
t
w
e
i
ght
a
nd
gl
oba
l
be
s
t
w
e
i
ght
a
r
e
s
e
t
t
o
1.
0.
Dy
n
a
m
i
c
i
n
e
r
t
i
a
we
i
g
h
t
i
s
t
r
u
e
s
o
t
h
a
t
t
h
e
i
n
e
r
t
i
a
we
i
g
h
t
i
s
e
n
h
a
n
c
e
d
d
u
r
i
n
g
e
x
e
c
u
t
i
o
n
.
T
o
e
v
a
l
u
a
t
e
t
h
e
le
a
r
n
in
g
m
o
d
e
l,
e
x
p
e
r
im
e
n
ts
a
r
e
c
o
n
d
u
c
te
d
on
a
l
l
i
ns
t
a
nc
e
s
w
i
t
h
10
-
fo
l
d
C
V
a
n
d
a
d
o
p
t
e
d
a
k
-
NN
t
e
c
h
n
i
q
u
e
to
a
c
h
ie
v
e
a
b
e
tte
r
p
e
r
f
o
r
m
a
n
c
e
.
I
n
k
N
N
c
la
s
s
if
ie
r
,
m
e
a
s
u
r
e
ty
p
e
is
th
e
'
M
ix
e
d
M
e
a
s
u
r
e
'
a
n
d
m
ix
e
d
m
e
a
s
u
r
e
is
th
e
'
M
ix
e
d
E
u
c
lid
e
a
n
D
is
ta
n
c
e
'
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
IS
S
N
:
2252
-
8938
r
An
i
m
p
r
o
v
e
d
h
y
b
r
i
d
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
m
e
t
h
o
d
f
o
r
h
u
g
e
d
i
m
e
n
s
i
o
n
a
l
d
a
t
a
s
e
t
s
(
F.
Ro
s
i
t
a
K
a
m
a
l
a
)
83
3.
2.
P
e
r
f
or
m
an
c
e
An
a
l
y
s
i
s
Th
e
C
V
m
a
i
n
t
a
i
n
s
a
n
a
ppa
r
e
nt
a
m
ount
of
da
t
a
f
or
t
e
s
t
in
g
an
d
u
s
e
t
h
e
r
es
t
f
o
r
t
r
ai
n
i
n
g
.
T
h
e
n
f
o
l
d
CV
cl
as
s
i
f
i
es
t
h
e
s
am
p
l
e
d
at
a
t
o
n
r
o
u
g
h
l
y
eq
u
al
d
i
v
i
s
i
o
n
s
,
o
n
e
i
s
u
s
ed
f
o
r
t
es
t
i
n
g
,
an
d
t
h
e
r
es
t
f
o
r
t
r
ai
n
i
n
g
.
T
h
u
s
,
th
e
C
V
b
r
in
g
s
a
ll
in
s
ta
n
c
e
s
to
p
a
r
tic
ip
a
te
b
o
th
in
tr
a
in
in
g
a
n
d
te
s
tin
g
m
o
d
e
ls
.
Cl
a
s
s
i
f
i
c
a
t
i
o
n
p
e
r
f
o
r
m
a
n
c
e
i
s
be
s
t
i
nt
e
r
pr
e
t
e
d
by
a
n
a
ppr
opr
ia
te
ly
n
a
m
e
d
to
o
l
c
a
lle
d
th
e
c
o
n
f
u
s
io
n
m
a
tr
ix
[
25
].
T
h
e
c
l
a
s
s
i
fi
e
r
a
c
c
u
ra
c
y
c
a
n
be
vi
e
w
e
d
i
n
one
of
f
our
pos
s
i
bl
e
w
a
ys
.
T
he
y
a
r
e
T
P
(
T
r
ue
_P
os
i
t
i
ve
)
,
F
P
(
F
a
l
s
e
_P
os
i
t
i
ve
)
,
F
N
(F
a
l
s
e
_
N
e
g
a
t
i
v
e
)
a
n
d
T
N
(T
ru
e
_
N
e
g
a
t
i
v
e
).
Th
e
a
c
c
u
r
a
c
y
o
f
t
h
e
p
r
o
p
o
s
e
d
I
H
F
S
a
l
g
o
r
i
t
h
m
i
s
c
o
mp
a
r
e
d
w
i
t
h
th
e
tr
a
d
itio
n
a
l
m
et
h
o
d
s
o
f
f
eat
u
r
e
s
el
ect
i
o
n
i
n
F
ig
u
r
e
3
.
Th
e
r
e
s
u
l
t
s
o
f
f
i
t
n
e
s
s
f
u
n
c
t
i
o
n
s
P
r
e
c
i
s
i
o
n
(
P
r
)
,
a
n
d
Re
c
a
l
l
(
Rc
)
o
f
t
h
e
I
H
F
S
w
i
t
h
d
i
f
f
e
r
e
n
t
c
o
m
b
i
n
a
t
i
o
n
s
a
r
e
s
h
o
w
n
i
n
F
ig
u
r
e
4
.
Sp
e
c
i
f
i
c
i
t
y
i
s
T
N
/
(
T
N
+FP)
.
Pr
e
c
i
s
i
o
n
i
s
T
P
/
(
T
P
+
FP
).
R
e
c
a
l
l
i
s
a
l
s
o
g
i
v
e
n
b
y
T
P
/
(T
P
+
F
N
).
A
c
c
u
ra
c
y
=
[T
P
+
T
N
]/
[T
P
+
F
P
+
T
N
+
FN
]
.
Th
e
re
s
u
l
t
of
t
hi
s
st
u
d
y
in
F
ig
u
r
e
3
de
pi
c
t
s
th
e
m
o
s
t
s
ig
n
if
ic
a
n
t
o
u
tc
o
m
e
s
of
t
he
pr
opos
e
d
a
l
gor
i
t
hm
.
In
al
m
o
s
t
al
l
d
at
as
et
s
t
h
e
accu
r
acy
of
t
he
pr
opos
e
d
st
u
d
y
sh
o
w
s
a
n
e
n
or
m
ous
i
m
pr
ove
m
e
nt
t
ha
n
t
he
t
r
a
di
t
i
ona
l
fi
l
t
e
r
m
e
t
h
o
d
s
C
H
I,
FSt
a
t
,
M
I
,
a
n
d
w
r
a
p
p
e
r
PSO
.
De
p
e
nd
on
t
he
out
c
om
e
s
obt
a
i
ne
d
i
n
F
ig
u
r
e
4
,
th
e
pr
os
pe
c
t
i
ve
m
e
t
hod
IH
F
S
is
f
o
u
n
d
to
b
e
th
e
b
e
s
t
in
te
r
m
s
o
f
p
r
e
c
is
io
n
a
n
d
r
e
c
a
ll
in
c
o
m
p
a
r
is
o
n
w
ith
th
e
di
f
f
e
r
e
nt
t
r
a
di
tio
n
a
l
me
t
h
o
d
s
in
th
e
lite
r
a
tu
r
e
lik
e
CH
I
,
FSt
a
t
,
M
I
,
a
n
d
w
r
a
p
p
e
r
PSO
an
d
y
i
el
d
s
a
m
as
s
i
v
e
im
p
r
o
v
e
m
e
n
t.
Fi
g
u
r
e
3
. A
c
c
u
r
a
c
y
C
o
m
p
a
r
is
o
n
o
f
th
e
f
r
a
m
e
w
o
r
k
I
H
F
S
w
ith
f
ilte
r
s
a
n
d
w
r
a
p
p
e
r
s
.
Fi
g
u
r
e
4
. C
o
m
p
a
r
a
tiv
e
s
tu
d
y
o
f
P
r
e
c
is
io
n
, a
n
d
R
e
c
a
ll o
f
th
e
I
H
F
S
w
ith
tr
a
d
itio
n
a
l e
x
is
tin
g
m
e
th
o
d
s
.
3.
3.
An
a
l
y
s
i
s
o
f
Pe
r
f
o
r
m
a
n
c
e
M
e
a
s
u
r
e
A
s
K
a
p
p
a
Co
h
e
n
s
K
a
p
p
a
s
t
a
t
i
s
t
i
c
i
s
s
t
a
t
i
s
t
i
c
a
l
l
y
s
t
r
o
n
g
[
2
6
].
It
i
s
c
o
n
s
i
d
e
re
d
a
s
o
n
e
o
f
t
h
e
e
v
a
l
u
a
t
i
o
n
m
e
a
s
u
re
s
in
c
la
s
s
if
ic
a
tio
n
p
e
r
f
o
r
m
a
n
c
e
.
I
t
is
s
u
p
p
o
r
tiv
e
a
n
d
a
c
c
e
p
ta
b
le
to
te
s
t,
a
s
a
n
e
v
a
lu
a
tio
n
m
e
a
s
u
r
e
o
f
th
e
v
a
r
ia
b
le
se
l
e
c
t
i
o
n
m
e
t
h
o
d
s.
T
h
u
s,
t
h
e
K
a
p
p
a
c
o
e
f
f
i
c
i
e
n
t
i
s
b
e
n
e
f
i
c
i
a
l
,
an
d
i
t
u
s
es
a
s
m
al
l
er
n
u
m
b
er
o
f
at
t
r
i
b
u
t
es
.
Th
e
p
r
o
p
o
s
e
d
I
H
F
S
p
e
r
f
e
c
t
l
y
a
g
r
e
e
s
w
i
t
h
t
h
e
a
b
o
ve
-
me
n
t
i
o
n
e
d
K
a
p
p
a
i
n
t
er
p
r
et
at
i
o
n
i
s
r
ep
r
es
en
t
ed
in
T
a
b
le
3.
Ka
p
p
a
a
s
a
n
e
f
f
i
c
i
e
n
c
y
m
e
a
s
u
r
e
,
e
n
h
a
n
c
e
s
t
h
e
s
t
r
e
n
g
t
h
o
f
t
h
e
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
s
t
r
a
t
e
g
i
e
s
.
T
h
i
s
i
s
be
c
a
us
e
K
a
ppa
puni
s
he
s
r
a
ndom
ne
s
s
a
nd
not
pe
r
f
or
m
a
nc
e
.
T
he
noi
s
y
f
e
a
t
ur
e
s
t
ha
t
do
not
a
f
f
e
c
t
t
he
co
r
r
ect
le
a
r
n
in
g
r
a
te
,
b
u
t
im
p
a
c
t
K
a
p
p
a
.
T
h
e
r
e
f
o
r
e
,
Ka
p
p
a
i
s
t
h
e
m
o
s
t
a
p
p
r
o
p
r
i
a
t
e
m
e
a
s
u
r
e
,
wh
i
c
h
s
e
l
e
c
t
s
t
h
e
s
u
b
s
e
t
s
0
50
100
150
P
r
e
c
i
s
i
o
n
(
%)
Si
g
ni
f
i
c
a
nc
e
m
e
a
s
ur
e
P
r
e
c
i
s
i
o
n
IH
F
S
Fi
l
t
e
r
(
F
-
S
t
at
/
M
I
)
Fi
l
t
e
r
(
C
H
I
)
PS
O
0
50
100
150
Gl
a
s
s
So
na
r
Ve
h
i
c
l
e
Io
n
o
Ch
e
s
s
Spl
i
c
e
He
p
a
t
i
t
i
s
Ly
m
p
h
Ov
a
r
i
a
n
ML
L
Le
u
k
-
2C
Lu
n
g
CNS
Le
u
k
-
4C
R
e
c
a
l
l
(
%)
Si
g
ni
f
i
c
a
nc
e
m
e
a
s
ur
e
R
e
c
a
l
l
IH
F
S
Fi
l
t
e
r
(
F
-
S
t
at
/
M
I
)
Fi
l
t
e
r
(
C
H
I
)
PS
O
Evaluation Warning : The document was created with Spire.PDF for Python.
r
IS
S
N
:
2252
-
8938
IJ
-
AI
Vo
l
.
8
, N
o
.
1,
M
a
r
c
h
201
9
:
77
–
86
84
of
s
m
a
l
l
e
r
s
i
z
e
a
nd
doe
s
not
i
nc
l
ude
t
he
noi
s
y
f
e
a
t
ur
e
s
.
A
K
a
ppa
of
a
l
i
m
i
t
f
r
om
0.
81
t
o
0.
99
e
nt
a
i
l
s
a
l
m
os
t
a
n
id
e
a
l a
g
r
e
e
m
e
n
t. I
d
e
a
l a
g
r
eem
en
t
w
o
u
l
d
eq
u
at
e
t
o
a
K
ap
p
a
o
f
one
.
3.
4.
MA
E
(
Me
a
n
A
b
s
o
l
u
t
e
E
r
r
o
r
)
a
n
d
R
MS
E
(
R
o
o
t
Me
a
n
S
q
u
a
r
e
d
E
r
r
o
r
)
a
n
a
l
y
s
i
s
Th
e
t
w
o
e
r
r
o
r
m
e
a
s
u
r
e
s
M
A
E
a
n
d
R
M
S
E
e
s
t
i
m
a
t
e
t
h
e
a
c
c
u
r
a
c
y
p
r
e
d
i
c
t
i
o
n
[2
7
]
.
M
A
E
is
d
e
f
in
e
d
a
s
th
e
m
e
a
n
o
f
th
e
d
if
f
e
r
e
n
c
e
o
f
tw
o
p
o
in
ts
lik
e
a
ct
u
al
an
d
f
i
t
t
ed
p
o
i
n
t
s
.
I
t
can
t
ak
e
v
al
u
es
f
r
o
m
zer
o
t
o
i
n
f
i
n
i
t
y
.
Th
e
i
d
e
a
l
f
i
t
i
s
p
r
e
v
a
i
l
e
d
i
f
M
A
E
i
s
z
e
r
o
.
R
M
S
E
i
s
a
m
e
a
s
u
r
e
o
f
e
r
r
o
r
i
n
a
b
s
o
l
u
t
e
v
a
l
u
e
.
I
t
c
a
l
c
u
l
a
t
e
s
t
h
e
d
if
f
e
r
e
n
c
e
s
q
u
a
r
e
o
f
p
o
s
itiv
e
a
n
d
n
e
g
a
tiv
e
d
e
v
ia
tio
n
s
to
c
a
n
c
e
l
o
n
e
a
n
o
th
e
r
o
u
t.
T
he
l
e
a
s
t
va
l
ue
of
R
M
S
E
im
p
lie
s
a
m
o
d
e
l o
f
g
o
o
d
s
ig
n
if
ic
a
n
c
e
.
Ta
b
l
e
4
d
e
p
i
c
t
s
t
h
e
M
A
E
a
n
d
R
M
S
E
o
u
t
c
o
m
e
s
.
3.
5.
Co
m
p
a
r
a
t
i
v
e
An
a
l
y
s
i
s
o
f
T
h
e
P
r
o
p
o
s
e
d
W
o
r
k
T
o
St
a
t
e
-
o
f
-
Th
e
-
Ar
t
Fe
a
t
u
r
e
S
e
l
e
c
t
i
o
n
M
e
t
h
o
d
s
In
s
u
m
m
a
ry
,
t
h
e
o
u
t
c
o
m
e
s
i
n
T
a
b
l
e
5
a
n
d
T
a
b
l
e
6
w
e
l
l
v
a
l
i
da
t
e
t
he
e
f
f
i
c
i
e
nc
y
of
I
H
F
S
i
n
t
e
r
m
s
of
accu
r
acy
an
d
C
P
U
t
i
m
e
r
es
p
ect
i
v
el
y
f
o
r
U
C
I
an
d
l
ar
g
e
-
sc
a
l
e
m
i
c
r
o
a
r
r
a
y
d
a
t
a
se
t
s.
T
h
e
d
a
t
a
se
t
s'
a
c
c
u
r
a
t
e
n
e
ss
ach
i
ev
ed
b
y
t
h
i
s
f
r
am
ew
o
r
k
i
s
co
m
p
ar
ed
i
n
T
ab
l
e
5
w
i
t
h
t
h
e
r
es
u
l
t
s
d
i
r
ect
l
y
t
ak
en
f
r
o
m
each
o
f
t
h
e
al
g
o
r
i
t
h
m
s
of
t
he
m
os
t
r
e
l
e
va
nt
w
or
ks
i
n t
he
r
e
c
e
nt
l
i
t
e
r
a
t
ur
e
.
I
n gl
a
s
s
,
s
ona
r
,
ve
hi
c
l
e
,
a
nd i
onos
phe
r
e
da
t
a
s
e
t
s
(n
u
m
e
ri
c
a
l
),
t
h
e
p
ro
p
o
s
e
d
IH
F
S
o
u
t
p
e
rfo
r
ms
t
h
e
r
e
l
e
v
a
n
t
s
t
u
d
y
o
f
[
2
8
,
2
1
,
3
3
,
12]
by
obt
a
i
ni
ng
t
he
m
os
t
co
m
p
et
i
t
i
v
e
r
es
u
l
t
s
i
n
t
er
m
s
o
f
accu
r
acy
an
d
C
P
U
tim
e
.
I
n
c
h
e
s
s
d
a
ta
s
e
t
b
o
th
th
e
a
c
c
u
r
a
c
y
a
n
d
r
u
n
tim
e
a
r
e
not
s
upe
r
i
or
ove
r
t
he
r
e
s
ul
t
s
obt
a
i
ne
d
i
n
[
21]
.
E
ve
n
t
hough
t
he
F
S
S
M
C
m
e
t
hods
i
n
[
31]
gi
vi
ng
s
i
gni
f
i
c
a
nt
co
m
p
u
t
at
i
o
n
t
i
m
e
f
o
r
s
p
l
i
ce
d
at
as
et
,
t
h
e
p
r
o
p
o
s
ed
I
H
F
S
o
u
t
p
er
f
o
r
m
s
t
h
e
accu
r
acy
o
f
t
h
e
F
S
S
M
C
me
t
h
o
d
s
wi
t
h
C
4
.
5
,
a
n
d
NB
c
l
a
s
s
i
f
i
e
r
s
.
T
h
e
a
c
c
u
r
a
c
y
o
f
I
HF
S
a
l
s
o
o
u
t
p
e
r
f
o
r
m
s
t
h
e
e
n
s
e
m
b
l
e
c
l
a
s
s
i
f
i
e
r
R
C
R
F
m
e
t
h
o
d
us
e
d
i
n
[
34]
f
or
s
pl
i
c
e
da
t
a
s
e
t
.
Ta
b
l
e
3
. K
a
p
p
a
C
o
e
f
f
ic
ie
n
t o
f
I
H
F
S
.
Da
t
a
s
e
t
s
Ka
p
p
a
Gl
a
s
s
0.
985
So
n
a
r
0.
954
Ve
h
i
c
l
e
0.
965
Io
n
o
1.
0
Ch
e
s
s
0.
924
Sp
l
i
c
e
0.
952
He
p
a
t
i
t
i
s
0.
961
Ly
m
p
h
1.
0
ML
L
1.
00
CN
S
0.
75
Ov
a
r
i
a
n
1.
0
Le
u
k
-
2C
1.
0
Lu
n
g
1.
0
Le
u
k
-
4C
1.
0
Ta
b
l
e
4
. M
A
E
a
n
d
R
M
S
E
in
m
ik
r
o
Da
t
a
s
e
t
s
MA
E
RM
S
E
Gl
a
s
s
0.
006
0.
108
So
n
a
r
0.
056
0.
154
Ve
h
i
c
l
e
0.
032
0.
224
Io
n
o
0.
000
0.
000
Ch
e
s
s
0.
063
0.
83
Sp
l
i
c
e
0.
044
0.
112
He
p
a
t
i
t
i
s
0.
000
0.
000
Ly
m
p
h
0.
000
0.
000
ML
L
0.
000
0.
000
CN
S
0.
500
0.
289
Ov
a
r
i
a
n
0.
000
0.
000
Le
u
k
-
2C
0.
000
0.
000
Lu
n
g
0.
000
0.
000
Le
u
k
-
4C
0.
000
0.
000
Ev
e
n
t
h
o
u
g
h
t
h
e
m
e
t
h
o
d
s
u
s
e
d
i
n
[
2
8
-
29]
c
oul
d
gi
ve
be
t
t
e
r
accu
r
acy
f
o
r
h
ep
at
i
t
i
s
d
at
as
et
,
t
h
er
e
i
s
n
o
si
g
n
i
f
i
c
a
n
c
e
i
n
c
o
m
p
u
t
a
t
i
o
n
t
i
m
e
.
B
u
t
t
h
e
p
r
o
p
o
se
d
I
H
F
S
g
i
v
e
s
si
g
n
i
f
i
c
a
n
t
r
e
su
l
t
s
i
n
t
e
r
m
s
o
f
b
o
t
h
a
c
c
u
r
a
c
y
an
d
r
u
n
t
i
m
e
f
o
r
h
ep
at
i
t
i
s
d
at
as
et
.
I
n
l
y
m
p
h
,
L
eu
k
em
i
a
2
C
an
d
M
L
L
d
at
as
et
s
ev
en
t
h
o
u
g
h
t
h
e
ex
i
s
t
i
n
g
me
t
h
ods
i
n
[
29]
c
oul
d
gi
ve
s
i
m
i
l
a
r
r
e
s
ul
t
s
f
or
a
c
c
ur
a
c
y
but
c
oul
d
not
gi
ve
be
t
t
e
r
r
e
s
ul
t
s
i
n
t
e
r
m
s
of
C
P
U
t
i
m
e
an
d
t
h
e
p
r
o
p
o
s
ed
I
H
F
S
i
s
s
u
p
er
i
o
r
i
n
t
er
m
s
o
f
r
u
n
t
i
m
e
f
o
r
t
h
es
e
d
at
as
et
s
.
I
n
O
v
ar
i
an
can
cer
d
at
as
et
ev
en
th
o
u
g
h
th
e
e
x
is
tin
g
m
e
th
o
d
s
u
s
e
d
in
[
3
0
]
a
n
d
[
10]
c
oul
d
gi
ve
c
om
pe
t
i
t
i
ve
r
e
s
ul
t
s
f
or
a
c
c
ur
a
c
y
but
c
oul
d
not
gi
ve
be
t
t
e
r
r
e
s
ul
t
s
i
n
t
e
r
m
s
of
C
P
U
t
i
m
e
.
I
n
C
N
S
da
t
a
s
e
t
t
he
a
c
c
ur
a
c
y
i
s
not
s
i
gni
f
i
c
a
nt
t
ha
n
e
xi
s
t
i
ng
m
e
t
hods
of
[
32]
but
t
he
I
H
F
S
m
e
t
hod
s
how
s
s
i
gni
f
i
c
a
nt
i
m
pr
ove
m
e
nt
i
n
c
om
put
i
ng
t
i
m
e
ove
r
th
e
m
e
th
o
d
s
in
[
2
9
]
.
Th
o
u
g
h
t
h
e
m
e
t
h
o
d
s
u
s
e
d
i
n
[
5
]
c
o
u
l
d
g
i
v
e
b
e
t
t
e
r
r
u
n
t
i
m
e
f
o
r
l
u
n
g
d
a
t
a
s
e
t
,
t
h
e
r
e
i
s
n
o
s
i
g
n
i
f
i
c
a
n
c
e
i
n
accu
r
acy
.
B
u
t
t
h
e
p
r
o
p
o
s
ed
I
H
F
S
g
i
v
es
s
i
g
n
i
f
i
can
t
r
es
u
l
t
s
f
o
r
l
u
n
g
d
at
as
et
i
n
t
er
m
s
o
f
b
o
t
h
accu
r
acy
an
d
r
u
n
tim
e
.
I
n
L
e
u
k
e
m
ia
4
C
mi
c
r
o
a
r
r
a
y
d
a
t
a
s
e
t
s
t
h
e
p
r
o
p
o
s
e
d
I
H
F
S
o
u
t
p
e
r
f
o
r
ms
t
h
e
r
e
l
e
v
a
n
t
s
t
u
d
y
i
n
[
2
9
]
b
y
obt
a
i
ni
ng
t
he
be
s
t
c
om
pe
t
i
t
i
ve
r
e
s
ul
t
s
i
n
t
e
r
m
s
of
a
c
c
ur
a
c
y
a
nd
C
P
U
t
i
m
e
.
As
s
h
o
wn
fro
m
T
ab
l
e
5
t
h
at
,
o
u
t
o
f
1
4
d
at
as
et
s
t
h
e
p
r
o
p
o
s
ed
m
et
h
o
d
s
h
o
w
s
s
i
g
n
i
f
i
can
t
l
y
b
et
t
er
o
n
tw
e
lv
e
d
at
as
et
s
an
d
w
o
r
s
e
o
n
t
w
o
d
at
as
et
s
i
n
t
er
m
s
o
f
accu
r
acy
.
I
n
t
er
m
s
o
f
co
m
p
u
t
i
n
g
t
i
m
e,
t
h
e
p
r
o
p
o
s
ed
me
t
h
o
d
h
a
s
t
e
n
s
i
g
n
i
f
i
c
a
n
t
l
y
b
e
t
t
e
r
r
e
s
u
l
t
s
,
t
w
o
w
o
r
s
e
r
e
s
u
l
t
s
(c
a
t
e
g
o
ri
c
a
l
)
a
n
d
n
o
s
i
g
n
i
fi
c
a
n
t
d
i
ffe
re
n
c
e
o
n
tw
o
r
e
s
u
lts
(l
u
n
g
a
n
d
H
e
p
a
t
i
t
i
s
).
T
h
e
s
u
c
c
e
s
s
of
I
H
F
S
c
oul
d
be
a
t
t
r
i
but
a
bl
e
t
o
i
t
s
a
bi
l
i
t
y
t
o
i
de
nt
i
f
y
s
t
r
ongl
y
re
l
e
v
a
n
t
fe
a
t
u
re
s
.
T
h
e
s
e
fe
a
t
u
re
s
i
n
c
re
a
s
e
t
h
e
l
i
k
e
l
i
h
o
o
d
t
o
a
s
c
e
rt
a
i
n
a
n
o
p
t
i
m
a
l
fe
a
t
u
re
s
u
b
s
e
t
.
E
v
e
n
t
h
o
u
g
h
th
e
a
c
c
u
r
a
c
ie
s
o
f
th
e
p
e
r
f
o
r
m
a
n
c
e
o
f
th
e
c
la
s
s
if
ie
r
f
o
r
m
ic
r
o
a
r
r
a
y
d
a
ta
s
e
ts
a
r
e
ve
r
y c
om
pe
t
i
t
i
ve
i
n T
a
bl
e
5,
t
he
st
r
e
n
g
t
h
o
f
t
h
e
I
H
F
S
a
l
g
o
r
i
t
h
m
c
a
n
b
e
v
i
n
d
i
c
a
t
e
d
i
n
T
a
b
l
e
6
b
y
t
h
e
r
a
p
i
d
c
o
m
p
u
t
a
t
i
o
n
t
i
m
e
o
f
t
h
e
m
i
c
r
o
a
r
r
a
y
da
t
a
s
e
t
s
t
ha
n
t
he
s
t
a
t
e
-
of
-
th
e
-
ar
t
w
o
r
k
s
i
n
t
h
e
r
ecen
t
s
t
u
d
y
,
p
r
o
v
es
t
h
e
i
m
p
r
o
v
ed
p
er
f
o
r
m
an
ce
o
f
t
h
e
fra
m
e
w
o
rk
IH
F
S
. O
v
e
r
a
ll, th
e
I
H
F
S
h
a
d
th
e
b
e
s
t p
e
r
f
o
r
m
a
n
c
e
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
IS
S
N
:
2252
-
8938
r
An
i
m
p
r
o
v
e
d
h
y
b
r
i
d
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
m
e
t
h
o
d
f
o
r
h
u
g
e
d
i
m
e
n
s
i
o
n
a
l
d
a
t
a
s
e
t
s
(
F.
Ro
s
i
t
a
K
a
m
a
l
a
)
85
Ta
b
l
e
5
.
Co
m
p
a
r
a
t
i
v
e
a
n
a
l
y
s
i
s
o
f
t
h
e
f
r
a
m
e
w
o
r
k
I
H
F
S
t
o
t
h
e
m
o
s
t
a
p
p
r
o
p
r
i
a
t
e
w
o
r
k
s
.
Da
t
a
s
e
t
s
IH
F
S
%
Re
s
u
l
t
s
o
b
t
a
i
n
e
d
f
r
o
m
t
h
e
l
i
t
e
r
a
t
u
r
e
(
Th
e
m
o
s
t
e
x
c
e
l
l
e
n
t
r
e
s
u
l
t
s
a
r
e
s
h
o
w
n
i
n
b
o
l
d
)
Me
t
h
o
d
o
l
o
g
y
&
A
c
c
u
r
a
c
y
Me
t
h
o
d
o
l
o
g
y
&
A
c
c
u
r
a
c
y
Me
t
h
o
d
o
l
o
g
y
&
A
c
c
u
r
a
c
y
Gl
a
s
s
98.
84
[
28
]
MA
+
S
V
M
opt
i
m
i
z
e
d
85.
98
[3
3
]IE
M
85.
56
[
28
]
MA
+
S
V
M
de
f
a
ul
t
76.
63
So
n
a
r
97.
62
[
28
]
MA
+
S
V
M
opt
i
m
i
z
e
d
97.
11
[
12
]H
F
S
M
97.
13
[
21
]P
S
O
P
G
3
87.
3
Ve
h
i
c
l
e
94.
99
[
21
]P
S
O
In
i
2
87.
99
[
21
]P
S
O
In
i
3
87.
8
[
28
]
MA
+
S
V
M
opt
i
m
i
z
e
d
83.
33
Io
n
o
100
[
28
]
MA
+
S
V
M
opt
i
m
i
z
e
d
96.
74
[
21
]P
S
O
P
G
2
95.
24
[
21
]P
S
O
In
i
2
94.
29
Ch
e
s
s
96.
64
[
21
]P
S
O
E
-
DT
99.
44
[3
3
]B
P
N
N
99.
28
[
21
]P
S
O
M
I
-
KNN
95.
62
Sp
l
i
c
e
98.
75
[
34]
RCRF
96.
48
[
31
]F
S
S
M
C
+
C
4
.
5
88.
4
[
31
]F
S
S
M
C
+
N
B
95.
6
He
p
a
t
i
t
i
s
100
[
28
]
MA
+
S
V
M
opt
i
m
i
z
e
d
100
[
29
]S
V
E
G
A
-
ANN
99.
81
[
29
]S
V
E
G
A
-
KNN
88.
45
Ly
m
p
h
100
[
29
]S
V
E
G
A
-
ANN
100
[
29
]
S
V
E
G
A
-
SV
M
87.
32
[
29
]S
V
E
G
A
-
KNN
91.
56
ML
L
100
[2
9
]
SV
EG
A
-
KNN
100
[
29
]S
V
E
G
A
-
NB
100
[
29
]S
V
E
G
A
-
SV
M
98.
61
CN
S
91.
67
[
32
]M
F
+
G
A
+
T
S
99.
33
[
29]
SV
EG
A
-
SV
M
93.
35
[
29
]S
V
E
G
A
-
ANN
95
Ov
a
r
i
a
n
100
[
10
]
BD
E
-
SV
M
Ra
n
k
100
[
30
]
IW
S
S
3
(1
N
N
)
100
[3
0
]
IW
S
S
3
(
3
NN)
99.
2
Le
u
k
-
2C
100
[
29
]S
V
E
G
A
-
NB
100
[
32
]M
F
+
G
A
+
T
S
99.
50
[
29
]S
V
E
G
A
-
SV
M
97.
2
Lu
n
g
100
[
32
]M
F
+
G
A
+
T
S
99.
17
[
10
]B
D
E
-
SV
M
98.
7
[
10
]B
D
E
-
KNN
Ra
n
k
98.
7
Le
u
k
-
4C
100
[
29
]
S
V
E
G
A
-
NB
97.
22
[
29
]S
V
E
G
A
-
SV
M
98.
86
[
29
]S
V
E
G
A
-
ANN
98.
61
Ta
b
l
e
6
.
Co
m
p
a
r
i
s
o
n
s
o
n
c
o
m
p
u
t
a
t
i
o
n
a
l
t
i
m
e
(
i
n
S
e
c
o
n
d
s
)
t
o
t
h
e
m
o
s
t
r
e
l
e
v
a
n
t
w
o
r
k
s
.
Da
t
a
s
e
t
s
IH
F
S
(S
e
c
)
Re
s
u
l
t
s
o
b
t
a
i
n
e
d
f
r
o
m
t
h
e
l
i
t
e
r
a
t
u
r
e
(
i
n
S
e
c
o
n
d
s
)
T
h
e
b
e
s
t
r
e
s
u
l
t
s
a
r
e
s
h
o
w
n
i
n
b
o
l
d
.
Me
t
h
o
d
o
l
o
g
y
&
C
P
U
t
i
m
e
Me
t
h
o
d
o
l
o
g
y
&
C
P
U
tim
e
Me
t
h
o
d
o
l
o
g
y
&
C
P
U
t
i
m
e
Gl
a
s
s
4
[2
8
]M
A
+
S
V
M
de
f
a
ul
t
80.
91
[
28
]S
L
S
+
S
V
M
61.
14
[
28
]G
A
+
S
V
M
70.
79
So
n
a
r
7
[2
8
]M
A
+
S
V
M
opt
i
m
i
z
e
d
126.
12
[
21
]P
S
O
In
i
1
18
[
21
]P
S
O
P
G
2
25.
2
Ve
h
i
c
l
e
17
[2
8
]M
A
+
S
V
M
opt
i
m
i
z
e
d
153.
26
[
12
]H
F
S
M
46.
98
[
21
]S
P
E
A
2
331.
8
Io
n
o
9
[
21
]P
S
O
In
i
P
G
42.
6
[
21
]P
S
O
P
G
2
52.
8
[
21
]N
S
G
A
IF
63.
6
Ch
e
s
s
132
[2
8
]M
A
+
S
V
M
opt
i
m
i
z
e
d
176.
3
[
21
]N
S
G
A
IIE
17.
78
[
21
]P
S
O
E
(
α
e
=0
.
5
)
256.
17
Sp
l
i
c
e
131
[
31
]F
S
S
M
C
25.
7
[
31
]R
e
l
i
e
f
164
[
31
]R
e
l
i
e
f
-
RS
18.
4
He
p
a
t
i
t
i
s
1
[2
8
]M
A
+
S
V
M
opt
i
m
i
z
e
d
22.
79
[
29
]S
V
E
G
A
-
ANN
2
.
4
6
[
29
]S
V
E
G
A
-
K
NN
.5
9
Ly
m
p
h
1
[
28
]S
L
S
+
S
V
M
53.
31
[
28
]G
A
+
S
V
M
98.
90
[2
8
]M
A
+
S
V
M
de
f
a
ul
t
122.
31
ML
L
68
[
29
]S
V
E
G
A
-
SV
M
240.
4
[
29
]S
V
E
G
A
-
ANN
288.
3
[
29
]S
V
E
G
A
-
J4
8
390.
1
CN
S
15
[
29
]S
V
E
G
A
-
SV
M
67.
1
[
29
]S
V
E
G
A
-
KNN
71.
16
[
29
]S
V
E
G
A
-
NB
80.
2
Ov
a
r
i
a
n
70
[
30
]
IW
S
S
2
(1
N
N
)
426.
4
[
30
]S
F
S
(1
N
N
)
678.
5
[3
0
]IW
S
S
2
(3
N
N
)
754.
6
Le
u
k
-
2C
10
[
29
]S
V
E
G
A
-
SV
M
112.
5
[
29
]S
V
E
G
A
-
NB
150
[
29
]S
V
E
G
A
-
J4
8
312.
6
Lu
n
g
59
[
5
]L
F
S
1.
26
[
5
]F
C
B
F
1.
68
[
5
]H
C
10.
92
Le
u
k
-
4C
8
[
29
]S
V
E
G
A
-
SV
M
234.
3
[
29
]S
V
E
G
A
-
NB
272
[
29
]S
V
E
G
A
-
J4
8
399.
7
4.
CO
NCL
US
I
O
N
Th
e
I
H
F
S
i
s
e
x
p
l
o
r
e
d
i
n
v
i
e
w
o
f
t
h
e
ad
v
an
t
ag
es
of
bot
h
f
e
a
t
ur
e
s
e
l
e
c
t
i
on
m
e
t
hods
l
i
ke
f
i
l
t
e
r
s
a
nd
wr
a
p
p
e
r
s
r
e
s
p
e
c
t
i
v
e
l
y
.
A
d
r
a
s
t
i
c
e
x
p
e
r
i
m
e
n
t
a
l
s
t
u
d
y
wa
s
c
o
n
d
u
c
t
e
d
wi
t
h
d
a
t
a
s
e
t
s
o
f
UC
I
a
n
d
m
i
c
r
o
a
r
r
a
y
re
p
o
s
i
t
o
ry
w
i
t
h
m
o
re
fe
a
t
u
re
s
.
Th
e
p
e
r
f
o
r
m
e
d
o
u
t
c
o
m
e
s
a
r
e
c
o
m
p
a
r
e
d
t
o
me
a
s
u
r
e
th
e
e
f
f
e
c
tiv
e
n
es
s
o
f
t
h
e
pr
opos
e
d
I
H
F
S
a
l
gor
i
t
hm
.
T
he
c
om
pe
t
e
nc
e
of
t
he
I
H
F
S
de
t
e
r
m
i
ne
s
t
he
be
s
t
pos
s
i
bl
e
f
e
a
t
ur
e
s
ubs
e
t
s
w
i
t
h
ut
m
os
t
e
f
f
i
c
i
e
nc
y,
i
n
c
om
pa
r
i
s
on
w
i
t
h
ot
he
r
di
ve
r
s
e
up
-
to
-
da
t
e
F
S
m
e
t
hodol
ogi
e
s
of
t
he
pr
ove
n
r
e
s
e
a
r
c
h
fi
n
d
i
n
g
s
w
i
t
h
s
i
m
i
l
a
r
d
a
t
a
s
e
t
s
.
In
v
i
e
w
of
t
hi
s
s
t
udy,
t
he
I
H
F
S
ha
s
i
m
pr
ove
d
t
he
c
l
a
s
s
i
f
i
e
r
a
c
c
ur
a
c
y
a
nd
co
m
p
u
t
at
i
o
n
al
t
i
m
e.
T
h
e
s
t
u
d
y
f
i
n
d
i
n
g
s
ar
e
v
er
y
n
o
t
ew
o
r
t
h
y
an
d
h
av
e
o
b
t
ai
n
ed
a
h
i
g
h
l
y
co
m
p
et
i
t
i
v
e
me
t
h
o
d
o
l
o
g
y
i
n
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
p
r
o
b
l
e
ms
,
i
mp
l
y
i
n
g
a
b
e
t
t
e
r
p
e
r
f
o
r
ma
n
c
e
.
F
o
r
f
u
t
u
r
e
s
t
u
d
i
e
s
,
th
is
f
r
a
m
e
w
o
r
k
can
al
s
o
b
e
pr
ogr
e
s
s
e
d
to
d
if
f
e
r
e
n
t
ty
p
e
s
o
f
d
im
e
n
s
io
n
s
u
b
s
e
ts
o
f
im
a
g
e
s
,
te
x
t
a
n
d
m
e
d
ic
a
l
d
a
ta
s
e
ts
.
I
H
F
S
c
a
n
al
s
o
b
e
cu
s
t
o
m
i
zed
t
o
m
ak
e
h
y
b
r
i
d
i
zat
i
o
n
w
i
t
h
o
t
h
er
P
S
O
t
ech
n
i
q
u
es
.
RE
F
E
RE
NCE
S
[1
]
Ri
c
h
a
r
d
E
r
n
e
s
t
Be
l
l
m
a
n
,
"
D
y
n
a
m
i
c
P
r
o
g
r
a
m
m
i
n
g
,
"
Co
ur
i
e
r
D
ove
r
P
ubl
i
c
a
t
i
ons
,
2003.
[2
]
B.
S
.
E
v
e
r
i
t
t
,
a
n
d
A
.
S
k
r
o
n
d
a
l
,
"
Ca
m
b
r
i
d
g
e
D
i
c
t
i
o
n
a
r
y
o
f
S
t
a
t
i
s
t
i
c
s
,
"
Ca
m
b
r
i
d
g
e
U
n
i
v
e
r
s
i
t
y
P
r
e
s
s
,
2010.
[3
]
A.
Gu
y
o
n
,
a
n
d
E
l
i
s
s
e
e
f
f
,
"
An
i
n
t
r
o
d
u
c
t
i
o
n
t
o
v
a
r
i
a
b
l
e
a
n
d
f
e
a
t
u
r
e
s
s
e
l
e
c
t
i
o
n
,
"
J.
o
f
M
a
ch
.
L
ea
rn
.
R
es
.,
v
o
l.3
,
pp.
1157
–
1182,
2003.
[4
]
R.
K
o
h
a
v
i
,
a
n
d
G
H
.
J
o
h
n
,
"
W
r
a
p
p
e
r
s
f
o
r
f
e
a
t
u
r
e
s
u
b
s
e
t
s
e
l
e
c
t
i
o
n
,
"
Ar
t
i
f
.
I
n
t
e
l
l
.
,
vol
.
97(
1
-
2)
,
pp.
273
–
324,
1997.
[5
]
Pa
b
l
o
B
e
r
m
e
j
o
,
J
o
s
e
A
G
a
m
e
z
,
a
n
d
J
o
s
e
M
P
u
e
r
ta
,
"
A
G
R
A
S
P
a
lg
o
r
ith
m
f
o
r
f
a
s
t
h
y
b
r
id
(
f
ilte
r
-
wr
a
p
p
e
r
)
f
e
a
t
u
r
e
su
b
set
sel
ect
i
o
n
i
n
h
i
g
h
-
di
m
e
ns
i
ona
l
da
t
a
s
e
t
s
,
"
Pa
t
t
e
r
n
Re
c
o
g
n
.
Le
t
t
.
,
vol
.
32,
pp.
701
–
711,
2011.
[6
]
N.
Ho
l
d
e
n
,
a
n
d
A.
F
r
e
i
t
a
s
,
"
A
Hy
b
r
i
d
P
S
O/
AC
O
a
l
g
o
r
i
t
h
m
f
o
r
d
i
s
co
v
er
i
n
g
cl
assi
f
i
cat
i
o
n
r
u
l
es
i
n
d
at
a
m
i
n
i
n
g
,
"
Hi
n
d
a
w
i
P
u
b
l
i
s
h
i
n
g
C
o
r
p
o
r
a
t
i
o
n
J
o
u
r
n
a
l
o
f
A
r
t
i
f
i
c
i
a
l
E
v
o
l
u
t
i
o
n
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
V
o
l
.
2
0
0
8
,
A
r
tic
le
I
D
3
1
6
1
4
5
,
1
1
pa
ge
s
,
2008.
[7
]
M.
N
e
k
k
a
a
a
n
d
D
.
B
o
u
g
h
a
c
i
,
"
H
y
b
r
i
d
h
a
r
m
o
y
s
e
a
r
c
h
c
o
m
b
i
n
e
d
w
i
t
h
s
t
o
c
h
a
s
t
i
c
l
o
c
a
l
s
e
a
rc
h
fo
r
fe
a
t
u
re
s
e
l
e
c
t
i
o
n
,
"
Ne
u
r
a
l
P
r
o
c
e
s
s
i
n
g
L
e
t
t
e
r
s
, p
p
.1
-
22,
2015.
[8
]
Ab
d
u
l
l
a
h
S
a
e
e
d
Gh
a
r
e
b
,
Az
u
r
a
l
i
z
a
Ab
u
B
a
k
a
r
,
Ab
d
u
l
R
a
z
a
k
Ha
md
a
n
,
"
Hy
b
r
i
d
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
b
a
s
e
d
o
n
e
n
h
an
ced
ge
ne
t
i
c
a
l
gor
i
t
hm
f
or
t
e
xt
c
a
t
e
gor
i
z
a
t
i
on
,"
Ex
p
e
r
t
S
y
s
t
e
m
s
w
i
t
h
Ap
p
l
i
c
a
t
i
o
n
s
,
vol
. 4
9
, p
p
. 3
1
-
47,
2016.
Evaluation Warning : The document was created with Spire.PDF for Python.
r
IS
S
N
:
2252
-
8938
IJ
-
AI
Vo
l
.
8
, N
o
.
1,
M
a
r
c
h
201
9
:
77
–
86
86
[9
]
Af
e
f
B
e
n
B
r
a
h
i
m,
M
o
h
a
me
d
L
i
ma
m
.
"
A
h
y
b
r
id
f
e
a
tu
r
e
s
e
le
c
tio
n
m
e
th
o
d
b
a
s
e
d
o
n
in
s
ta
n
c
e
le
a
r
n
in
g
a
n
d
co
o
p
er
at
i
v
e
su
b
set
sear
ch
,
"
Pa
t
t
e
r
n
Re
c
o
g
n
i
t
i
o
n
Le
t
t
e
r
s
,
vol
. 6
9
, p
p
. 2
8
-
34,
2016.
[1
0
]
Ap
o
l
l
o
n
i
,
Gu
i
l
l
e
r
mo
L
e
g
u
i
z
a
mó
n
,
a
n
d
E
n
r
i
q
u
e
Al
b
a
,
"
T
wo
h
y
b
r
i
d
wr
a
p
p
e
r
-
fi
l
t
e
r
fe
a
t
u
re
s
e
l
e
c
t
i
o
n
a
l
g
o
ri
t
h
m
s
ap
p
l
i
ed
t
o
h
i
g
h
-
di
m
e
ns
i
ona
l
m
i
c
r
oa
r
r
a
y
e
xpe
r
i
m
e
nt
s
,
"
Ap
p
l
.
S
o
f
t
.
C
o
m
p
u
t
.
,
vol
.
38,
pp.
922
–
932,
2016
.
[1
1
]
Ez
g
i
Zo
r
a
r
p
a
c
ı
,
a
n
d
Se
l
m
a
A
y
s
e
O
z
e
l
,
"
A
h
y
b
r
i
d
a
p
p
r
o
a
c
h
o
f
d
i
f
f
e
r
e
n
t
i
a
l
e
v
o
l
u
t
i
o
n
a
n
d
a
r
t
i
f
i
c
i
a
l
b
e
e
c
o
l
o
n
y
f
o
r
fe
a
t
u
re
s
e
l
e
c
t
i
o
n
,"
Ex
p
e
r
t
S
y
s
t
e
m
s
W
i
t
h
A
ppl
i
c
at
i
ons
,
vol
.
62
, p
p
. 9
1
-
103
,
2016.
[1
2
]
F.
R
o
s
i
t
a
K
a
m
a
l
a
,
a
n
d
D
r
.
R
a
n
j
i
t
J
e
b
a
Th
a
n
g
a
i
a
h
P,
"
A
p
r
o
p
o
s
e
d
t
w
o
p
h
a
s
e
h
y
b
r
i
d
f
e
a
t
u
re
s
e
l
e
c
t
i
o
n
m
e
t
h
o
d
u
s
i
n
g
ba
c
kw
a
r
d
E
l
i
m
i
na
t
i
on
a
nd
P
S
O
,
"
In
t
.
J
.
o
f
A
p
p
l
.
E
n
g
.
R
e
s
,
vol
.
11(
1)
,
pp.
77
-
83,
2016.
[1
3
]
Hu
i
j
u
a
n
L
u
,
J
u
n
y
i
n
g
C
h
e
n
,
Ke
Ya
n
,
Qu
n
J
i
n
,
a
n
d
Z
h
i
g
a
n
g
Ga
o
,
"
A
h
y
b
r
i
d
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
a
l
g
o
r
i
t
h
m
f
o
r
g
e
n
e
ex
p
r
essi
o
n
d
at
a
cl
assi
f
i
cat
i
o
n
,
"
N
eu
r
o
co
m
p
u
t
i
n
g
,
v
o
l
.
2
5
6
,
p
p
.
5
6
-
62,
2017.
[1
4
]
Mo
h
a
m
e
d
A
.
T
a
w
h
i
d
,
a
n
d
K
e
v
i
n
B
.
D
s
o
u
z
a
,
"
H
y
b
r
i
d
B
i
n
a
r
y
B
a
t
E
n
h
a
n
c
e
d
P
a
r
t
i
c
l
e
S
w
a
r
m
O
p
t
i
m
i
z
a
t
i
o
n
Al
g
o
r
i
t
h
m
f
o
r
s
o
l
v
i
n
g
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
p
r
o
b
l
e
ms
"
Ap
p
l
i
e
d
C
o
m
p
u
t
i
n
g
a
n
d
I
n
f
o
r
m
a
t
i
c
s
(A
rt
i
c
l
e
i
n
P
re
s
s
)
O
p
e
n
Ac
c
e
s
s
,
Ap
r
i
l
2
0
1
8
.
[1
5
]
De
b
o
j
i
t
B
o
r
o
,
Dh
r
u
b
a
K.
Bh
a
t
t
a
c
h
a
r
y
y
a
,
“
Pa
r
t
i
c
l
e
Sw
a
r
m
O
p
t
i
m
i
z
a
t
i
o
n
b
a
s
e
d
K
N
N
f
o
r
i
m
p
r
o
v
i
n
g
K
N
N
a
n
d
en
sem
b
l
e
cl
assi
f
i
cat
i
o
n
p
er
f
o
r
m
an
ce,
"
In
t
e
r
n
a
t
i
o
n
a
l
j
o
u
r
n
a
l
o
f
i
n
n
o
v
a
t
i
v
e
c
o
m
p
u
t
i
n
g
a
n
d
a
p
p
l
i
c
a
t
i
o
n
s
(IJ
IC
A
)
,
v
o
l.
6(
3/
4)
,
pp.
145
-
162,
2015.
[1
6
]
M.
J
a
b
b
a
r
"
A
P
r
e
d
i
c
t
i
o
n
o
f
h
e
a
r
t
d
i
s
e
a
s
e
u
s
i
n
g
k
n
e
a
r
e
s
t
n
e
i
g
h
b
o
r
a
n
d
p
a
r
t
i
c
l
e
s
w
a
r
m
o
p
t
i
m
i
s
a
t
i
o
n
,
"
Bi
o
m
e
d
i
c
a
l
Re
s
e
a
r
c
h
, v
o
l. 2
8
(
9
)
, p
p
. 4
1
5
4
-
4158,
2017.
[1
7
]
P.
E.
G
r
e
e
n
w
o
o
d
,
a
n
d
M
.
S.
N
i
k
u
l
i
n
,
"
A
g
u
i
d
e
t
o
c
h
i
-
sq
u
ar
ed
t
est
i
n
g
,
"
Jo
h
n
W
i
l
ey
&
S
o
n
s,
N
ew
Y
o
r
k
,
1
9
9
6
.
[1
8
]
H.
O.
L
a
n
c
a
s
t
e
r
,
"
T
h
e
c
h
i
-
sq
u
ar
ed
d
i
st
r
i
b
u
t
i
o
n
,
"
Jo
h
n
W
i
l
ey
&
S
o
n
s,
N
ew
Y
o
r
k
,
1
9
6
9
.
[1
9
]
Pa
u
l
J
.
La
v
r
a
k
a
s
, "
E
n
c
y
c
lo
p
e
d
ia
o
f
S
u
r
v
e
y
R
e
s
e
a
r
c
h
M
e
th
o
d
s
,"
S
A
G
E
P
u
b
lic
a
tio
n
s
, 2
0
0
8
.
[2
0
]
Ch
r
i
s
t
o
p
h
e
r
D
.
Ma
n
n
i
n
g
,
P
r
a
b
h
a
k
a
r
R
a
g
h
a
v
a
n
,
H
i
n
r
i
c
h
S
c
h
u
t
z
e
,
"
A
n
I
n
t
r
o
d
u
c
t
i
o
n
t
o
I
n
f
o
r
m
a
t
i
o
n
Re
t
r
i
e
v
a
l
,
"
Ca
m
b
r
i
d
g
e
U
n
i
v
e
r
s
i
t
y
P
r
e
s
s
, 2
0
0
8
.
[2
1
]
Bi
n
g
X
u
e
,
"
P
a
r
t
i
c
l
e
S
w
a
r
m
O
p
t
im
is
a
tio
n
f
o
r
F
e
a
tu
r
e
S
e
le
c
tio
n
in
C
la
s
s
if
ic
a
tio
n
,"
Ph
.
D
Th
e
s
i
s
,
Vi
c
t
o
r
i
a
U
n
i
v
e
r
s
i
t
y
of
W
e
l
l
i
ngt
on
, W
e
llin
g
to
n
, 2
0
1
4
.
[2
2
]
J.
K
en
n
ed
y
,
an
d
R
.
C
.
E
b
er
h
ar
t
,
"P
ar
t
i
cl
e
sw
ar
m
o
p
t
i
m
i
zat
i
o
n
,
"
i
n
Pr
o
c
.
o
f
I
EEE
I
n
t
.
C
o
n
f
.
o
n
N
e
u
r
a
l
N
e
t
w
o
r
k
s
,
Pe
r
t
h
,
2
7
N
o
v
e
m
b
e
r
-
1
D
e
c
e
m
be
r
1
9
9
5
,
1
9
9
5
,
p
p
.
1
9
4
2
–
1948.
[2
3
]
M.
L
i
c
h
m
a
n
,
"
U
C
I
Ma
c
h
i
n
e
L
e
a
r
n
i
n
g
R
e
p
o
s
i
t
o
r
y
"
I
r
v
i
n
e
,
C
A
:
U
n
i
v
e
r
s
i
t
y
o
f
C
a
l
i
f
o
r
n
i
a
,
S
c
h
o
o
l
o
f
I
n
f
o
r
m
a
t
i
o
n
a
n
d
Co
m
p
u
t
e
r
S
c
i
e
n
c
e
.
h
t
t
p
:
/
/
a
r
c
h
i
v
e
.
i
c
s
.
u
c
i
.
e
d
u
/
m
l
.
L
a
s
t
v
i
s
i
t
J
u
l
y
2
0
1
8
.
[2
4
]
Zh
u
z
x
.
M
i
c
r
o
a
r
r
a
y
D
a
t
a
s
e
t
s
i
n
We
k
a
A
R
F
F
f
o
r
m
a
t
,
h
ttp
://
c
s
s
e
.s
z
u
.e
d
u
.c
n
/
s
ta
f
f
/
z
h
u
z
x
/
D
a
ta
s
e
ts
.h
tm
l
(
8
J
u
ly
2018,
da
t
e
l
a
s
t
a
c
c
e
s
s
e
d)
.
[2
5
]
V.
S
t
e
h
ma
n
,
a
n
d
S
t
e
p
h
e
n
,
"
S
e
l
e
c
t
i
n
g
a
n
d
i
n
t
e
r
p
r
e
t
i
n
g
me
a
s
u
r
e
s
o
f
t
h
e
ma
t
i
c
c
l
a
s
s
i
f
i
c
a
t
i
on
a
c
c
ur
a
c
y,
"
Re
m
o
t
e
Se
ns
i
ng
of
E
nv
i
r
onm
e
nt
, v
o
l. 6
2
(
1
)
, p
p
. 7
7
–
89,
1997.
[2
6
]
An
t
h
o
n
y
Vi
e
r
a
J
.
,
a
n
d
J
o
a
n
n
e
Ga
r
r
e
t
t
M
.
,
"
Un
d
e
r
s
t
a
n
d
i
n
g
I
n
t
e
r
o
b
s
e
r
v
e
r
Ag
r
e
e
me
n
t
:
T
h
e
Ka
p
p
a
S
t
a
t
i
s
t
i
c
,
"
Re
s
.
Se
r
i
e
s
.
F
am
i
l
y
M
e
di
c
i
ne
,
vol
.
37(
5)
,
pp.
360
-
363,
2005.
[2
7
]
Wi
l
l
m
o
t
t
,
C
o
r
t
J
,
M
a
ts
u
u
r
a
,
K
e
n
ji,
"
A
d
v
a
n
ta
g
e
s
o
f
th
e
m
e
a
n
a
b
s
o
lu
te
e
r
r
o
r
(
M
A
E
)
o
v
e
r
th
e
r
o
o
t
m
e
a
n
s
q
u
a
r
e
e
r
r
o
r
(R
M
S
E
)
i
n
a
s
s
e
s
s
i
n
g
a
v
e
ra
g
e
m
o
d
e
l
p
e
rfo
rm
a
n
c
e
,
"
Cl
i
m
a
t
e
R
e
s
e
a
r
c
h
, v
o
l 3
0
, p
p
7
9
–
82,
2005.
[2
8
]
Me
s
s
a
o
u
d
a
N
e
k
k
a
a
,
a
n
d
D
a
l
i
l
a
B
o
u
g
h
a
c
i
,
"
A
m
e
m
e
t
i
c
a
l
g
o
r
i
t
h
m
w
i
t
h
s
u
p
por
t
ve
c
t
or
m
a
c
hi
ne
f
or
f
e
a
t
ur
e
s
e
l
e
c
t
i
on
an
d
cl
assi
f
i
cat
i
o
n
,
"
Me
m
e
t
i
c
C
o
m
p
u
t
.
,
vol
.
7,
pp.
59
–
73,
2015.
[2
9
]
S.
Sa
s
i
k
a
l
a
,
S.
A
p
p
a
v
u
a
l
i
a
s
B
a
l
a
m
u
r
u
g
a
n
,
a
n
d
S.
G
e
e
t
h
a
,
"
A
n
o
v
e
l
a
d
a
p
t
i
v
e
f
e
a
t
u
r
e
s
e
l
e
c
t
o
r
f
o
r
s
u
p
e
r
v
i
s
e
d
cl
assi
f
i
cat
i
o
n
,
"
In
f
o
r
m
a
t
i
o
n
P
r
o
c
e
s
s
i
n
g
L
e
t
t
e
rs
, v
o
l. 1
1
7
, p
p
. 2
5
–
34,
2017.
[3
0
]
Ai
g
u
o
W
a
n
g
,
Ni
n
g
An
,
Gu
i
l
i
n
C
h
e
n
,
L
i
a
n
L
i
,
a
n
d
Gi
l
Al
t
e
r
o
v
i
t
z
,
"
Ac
c
e
l
e
r
a
t
i
n
g
wr
a
p
p
e
r
-
ba
s
e
d
f
e
a
t
ur
e
s
e
l
e
c
t
i
on
wi
t
h
K
-
ne
a
r
e
s
t
-
ne
i
ghbour
,"
K
n
o
w
l.
-
Ba
s
e
d
S
y
s
t
.
, v
o
l. 8
3
, p
p
. 8
1
–
91,
2015.
[3
1
]
Yu
e
Hu
a
n
g
,
P
a
u
l
J
.
M
c
C
u
l
l
a
g
h
,
a
n
d
No
r
ma
n
D
B
l
a
c
k
,
"
An
o
p
t
i
mi
z
a
t
i
o
n
o
f
R
e
l
i
e
f
F
f
o
r
c
l
a
s
s
i
f
i
c
a
t
i
o
n
i
n
l
a
r
g
e
da
t
a
s
e
t
s
,
"
Da
t
a
a
n
d
K
n
o
w
l
.
E
n
g
.
,
vol
.
68,
pp.
1348
–
1356,
2009.
[3
2
]
Ed
m
u
n
d
o
B
o
n
i
l
l
a
-
Hu
e
r
t
a
,
Al
b
e
r
t
o
He
r
n
a
n
d
e
z
-
Mo
n
t
i
e
l
,
R
o
b
e
r
t
o
Mo
r
a
l
e
s
-
Ca
p
o
r
a
l
,
a
n
d
M
a
r
c
o
A
r
j
o
n
a
-
Lo
p
e
z
,
"H
y
b
r
i
d
F
r
am
ew
o
r
k
U
si
n
g
M
u
l
t
ip
le
-
Fi
l
t
e
r
s
a
n
d
a
n
Em
b
e
d
d
e
d
A
p
p
r
o
a
c
h
f
o
r
a
n
Ef
f
i
c
i
e
n
t
Se
l
e
c
t
i
o
n
a
n
d
Cl
a
s
s
i
f
i
c
a
t
i
o
n
o
f
M
i
c
r
o
a
r
r
a
y
D
a
t
a
,
"
IE
E
E
/
A
C
M
T
r
a
n
s
a
c
t
i
o
n
s
o
n
C
o
m
p
u
t
a
t
i
o
n
a
l
B
i
o
l
o
g
y
A
n
d
B
i
o
i
n
f
o
r
m
a
t
i
c
s
Ja
n
u
a
ry/
F
eb
ru
a
ry
, v
o
l.1
3
(
1
)
, p
p
. 1
2
-
26,
2016
.
[3
3
]
Ku
n
g
-
Jen
g
W
an
g
,
A
n
g
el
i
a
M
el
an
i
A
d
r
i
an
,
Ku
n
-
Hu
a
n
g
C
h
e
n
,
Ku
n
g
-
Mi
n
Wa
n
g
,
"
A
n
i
m
p
r
o
v
e
d
e
l
e
c
t
r
o
m
a
g
n
e
t
i
s
m
-
lik
e
m
e
c
h
a
n
is
m
a
lg
o
r
ith
m
a
n
d
its
a
p
p
lic
a
tio
n
to
th
e
p
r
e
d
ic
tio
n
o
f
d
ia
b
e
te
s
m
e
llitu
s
,"
Jo
u
rn
a
l
o
f
B
i
o
m
ed
i
ca
l
In
f
o
r
m
a
t
i
c
s
, v
o
l. 5
4
, p
p
. 2
2
0
–
229,
2015.
[3
4
]
Jo
aq
u
i
n
A
b
el
l
an
,
C
ar
l
o
s
J.
M
an
t
as,
Jav
i
er
G
.
C
ast
el
l
an
o
,
S
er
af
i
n
M
o
r
al
-
Ga
r
c
i
a
,
"
I
n
c
r
e
a
s
i
n
g
d
i
v
e
r
s
i
t
y
i
n
r
a
n
d
o
m
fo
re
s
t
l
e
a
rn
i
n
g
a
l
g
o
ri
t
h
m
v
i
a
i
m
p
re
c
i
s
e
p
ro
b
a
b
i
l
i
t
i
e
s
"
,
Ex
p
e
r
t
S
y
s
t
e
m
s
W
i
t
h
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l.
9
7
,
p
p
.
2
2
8
–
243,
2018.
Evaluation Warning : The document was created with Spire.PDF for Python.