TELK
O
M
N
I
K
A
Te
l
e
c
ommu
n
i
c
at
i
on
,
C
omp
u
ti
n
g
,
El
e
c
t
r
on
i
c
s
an
d
C
on
tr
o
l
V
ol
.
18
,
N
o
.
2
,
A
pri
l
2020
,
p
p.
8
1
5
~
8
2
1
IS
S
N
:
1693
-
6930
,
a
c
c
r
e
di
t
e
d
F
i
rs
t
G
r
a
de
by
K
e
m
e
n
ri
s
t
e
kdi
k
t
i
,
D
e
c
r
e
e
N
o:
21
/
E
/
K
P
T
/
2
018
D
O
I:
10
.
129
28/
T
E
L
K
O
M
N
IK
A
.
v1
8i
2
.
1478
5
815
Jou
r
n
al
h
o
m
e
pa
ge
:
ht
t
p:
/
/
j
our
nal
.
uad
.
a
c
.
i
d/
i
nd
e
x
.
php
/
T
E
L
K
O
MNIKA
C
o
mp
a
r
i
n
g
r
a
n
d
o
m
f
o
r
e
st
a
n
d
su
p
p
o
r
t
v
e
c
t
o
r
m
a
c
h
i
n
e
s
f
o
r
b
r
e
a
st
c
a
n
c
e
r
c
l
a
ssi
f
i
c
a
t
i
o
n
C
h
e
l
vi
an
A
r
o
e
f
,
Y
u
d
a
R
i
van
,
Zu
h
e
r
m
an
R
u
s
tam
D
e
pa
r
t
m
e
n
t
of
M
a
t
he
m
a
t
i
c
s
,
U
ni
ve
r
s
i
t
y
o
f
I
nd
one
s
i
a
,
I
ndon
e
s
i
a
A
r
ti
c
l
e
I
n
fo
A
BS
TR
A
C
T
Ar
t
i
c
l
e
h
i
s
t
or
y
:
Re
c
e
i
v
e
d
A
ug
31
,
201
9
Re
vi
s
e
d
J
a
n
15
,
202
0
A
c
c
e
pt
e
d
F
e
b
14
,
202
0
T
he
r
e
a
r
e
m
or
e
t
h
a
n
10
0
t
y
pe
s
o
f
c
a
nc
e
r
a
r
ou
nd
t
he
w
or
l
d
w
i
t
h
d
i
f
f
e
r
e
n
t
s
ym
pt
om
s
a
n
d
d
i
f
f
i
c
ul
t
y
i
n
p
r
e
d
i
c
t
i
ng
i
t
s
a
pp
e
a
r
a
n
c
e
i
n
a
p
e
r
s
on
due
t
o
i
t
s
r
a
nd
om
a
nd
s
udd
e
n
a
t
t
a
c
k
m
e
t
hod
.
H
ow
e
v
e
r
,
t
he
a
ppe
a
r
a
nc
e
of
c
a
nc
e
r
i
s
ge
ne
r
a
l
l
y
m
a
r
k
e
d
by
t
he
g
r
ow
t
h
o
f
s
o
m
e
a
bno
r
m
a
l
c
e
l
l
.
S
o
m
e
one
m
i
ght
be
di
a
g
nos
e
d
e
a
r
l
y
a
nd
qu
i
c
k
l
y
t
r
e
a
t
e
d
,
b
ut
t
he
c
a
nc
e
r
ou
s
c
e
l
l
m
os
t
t
i
m
e
s
hi
d
e
s
i
n
t
he
bo
dy
of
i
t
s
v
i
c
t
i
m
a
nd
r
e
a
pp
e
a
r
,
on
l
y
t
o
k
i
l
l
i
t
s
s
uf
f
e
r
e
r
.
O
ne
o
f
t
he
m
o
s
t
c
om
m
on
c
a
nc
e
r
s
i
s
b
r
e
a
s
t
c
a
nc
e
r
.
A
c
c
or
d
i
ng
t
o
M
i
n
i
s
t
r
y
o
f
H
e
a
l
t
h
,
i
n
20
18
,
br
e
a
s
t
c
a
n
c
e
r
a
t
t
a
c
ke
d
42
o
ut
o
f
e
ve
r
y
10
0.
000
pe
opl
e
i
n
I
ndo
n
e
s
i
a
w
i
t
h
a
ppr
oxi
m
a
t
e
l
y
1
7
d
e
a
t
hs
.
I
n
a
dd
i
t
i
on
,
t
h
e
M
i
ni
s
t
r
y
r
e
c
o
r
de
d
a
y
e
a
r
l
y
i
nc
r
e
a
s
e
i
n
c
a
nc
e
r
pa
t
i
e
n
t
s
.
T
he
r
e
f
o
r
e
,
t
h
e
r
e
i
s
a
de
q
ua
t
e
ne
e
d
t
o
be
a
bl
e
t
o
de
t
e
r
m
i
ne
t
hos
e
a
f
f
e
c
t
e
d
by
t
hi
s
di
s
e
a
s
e
.
T
hi
s
s
t
udy
a
ppl
i
e
d
t
h
e
B
or
u
t
a
f
e
a
t
u
r
e
s
e
l
e
c
t
i
on
t
o
de
t
e
r
m
i
n
e
t
he
m
os
t
i
m
p
or
t
a
n
t
f
e
a
t
ur
e
s
i
n
m
a
k
i
ng
a
m
a
c
hi
ne
l
e
a
r
ni
ng
m
od
e
l
.
F
ur
t
he
r
m
or
e
,
t
h
e
R
a
ndo
m
F
or
e
s
t
(
R
F
)
a
n
d
S
u
ppo
r
t
V
e
c
t
o
r
M
a
c
hi
n
e
s
(
S
V
M
)
w
e
r
e
t
he
m
a
c
h
i
ne
l
e
a
r
n
i
ng
m
ode
l
us
e
d
,
w
i
t
h
h
i
gh
e
s
t
a
c
c
ur
a
c
i
e
s
o
f
90
%
a
nd
9
5%
r
e
s
p
e
c
t
i
v
e
l
y
.
F
r
o
m
t
h
e
r
e
s
u
l
t
s
o
bt
a
i
n
e
d
,
t
he
S
V
M
i
s
a
b
e
t
t
e
r
m
ode
l
t
ha
n
r
a
nd
o
m
f
or
e
s
t
i
n
t
e
r
m
s
o
f
a
c
c
ur
a
c
y
.
Ke
y
w
or
d
s
:
Bre
a
s
t
c
a
n
c
e
r
Ra
ndo
m
fore
s
t
S
upport
ve
c
t
o
r
m
a
c
hi
ne
s
T
hi
s
i
s
an
op
e
n
ac
c
e
s
s
ar
t
i
c
l
e
u
nde
r
t
he
C
C
B
Y
-
SA
l
i
c
e
n
s
e
.
Cor
r
e
s
pon
di
n
g
Au
t
h
or
:
Z
uhe
rm
a
n
Rus
t
a
m
,
D
e
pa
r
t
m
e
nt
of
M
a
t
he
m
a
t
i
c
s
,
U
ni
ve
rs
i
t
y
of
Indon
e
s
i
a
,
J
l
.
M
a
rgond
a
Ra
y
a
,
P
ond
ok
Ci
n
a
,
K
e
c
a
m
a
t
a
n
B
e
j
i
,
K
ot
a
D
e
po
k,
J
a
w
a
B
a
ra
t
16424
,
Indon
e
s
i
a
.
E
m
a
i
l
:
rus
t
a
m
@
ui
.
a
c
.
i
d
1.
I
N
TR
O
D
U
C
TI
O
N
Ca
nc
e
r
i
s
one
of
t
he
d
e
a
dl
i
e
s
t
di
s
e
a
s
e
s
i
n
t
h
e
w
or
l
d.
A
c
c
ordi
n
g
t
o
W
orl
d
H
e
a
l
t
h
O
rg
a
ni
z
a
t
i
on
(W
H
O
)
,
i
n
2018
,
i
t
i
s
t
he
s
e
c
ond
l
e
a
di
ng
c
a
us
e
o
f
de
a
t
h
gl
ob
a
l
l
y
a
n
d
r
e
s
pons
i
b
l
e
f
or
a
p
proxi
m
a
t
e
l
y
9
.
6
b
i
l
l
i
on
d
e
a
t
hs
i
n
2018
.
T
h
e
re
a
r
e
ove
r
1
00
di
f
fe
r
e
nt
t
ype
o
f
c
a
nc
e
r
t
h
a
t
a
ff
e
c
t
h
um
a
n
.
H
ow
e
v
e
r,
t
hi
s
s
t
ud
y,
a
i
m
s
t
o
a
n
a
l
yz
e
t
he
br
e
a
s
t
c
a
n
c
e
r
,
a
di
s
e
a
s
e
i
n
w
h
i
c
h
c
e
l
l
s
grow
out
of
c
ont
r
ol
t
o
for
m
a
t
u
m
or
w
hi
c
h
t
e
n
ds
t
o
a
ff
e
c
t
a
not
he
r
pa
rt
of
t
he
b
ody.
T
h
e
re
a
re
t
hre
e
c
o
m
m
on
pa
r
t
s
of
bre
a
s
t
w
ho
s
e
c
e
l
l
s
ha
s
t
he
a
bi
l
i
t
y
t
o
t
u
rn
i
n
t
o
c
a
nc
e
r
na
m
e
l
y
l
obul
e
s
,
duc
t
s
,
a
nd
t
h
e
c
onn
e
c
t
i
v
e
t
i
s
s
ue
.
L
obu
l
e
s
a
re
gl
a
nds
w
hi
c
h
produ
c
e
s
m
i
l
k
w
hi
l
e
t
he
du
c
t
s
a
re
t
h
i
n
t
ube
s
t
ha
t
c
a
r
ry
m
i
l
k
a
w
a
y
fro
m
t
h
e
l
ob
ul
e
.
T
he
c
onne
c
t
i
ve
t
i
s
s
ue
c
ons
i
s
t
s
of
f
i
brous
a
n
d
fa
t
t
y
t
i
s
s
ue
s
w
hi
c
h
hol
ds
t
he
bre
a
s
t
a
n
d
g
i
v
e
s
i
t
s
h
a
pe
a
nd
s
i
z
e
.
H
ow
e
ve
r
,
i
n
m
os
t
c
a
s
e
s
,
i
t
be
g
i
ns
i
n
t
he
l
obul
e
s
o
r
d
uc
t
s
.
T
he
e
xa
c
t
c
a
us
e
s
of
bre
a
s
t
c
a
nc
e
r
a
r
e
s
t
i
l
l
not
know
n
,
but
e
xp
e
rt
s
a
r
e
o
f
t
he
op
i
ni
o
n
t
ha
t
a
n
i
n
t
e
r
a
c
t
i
o
n
be
t
w
e
e
n
g
e
n
e
s
w
i
t
h
l
i
f
e
s
t
y
l
e
,
e
n
vi
ron
m
e
nt
,
a
n
d
hor
m
o
ne
,
t
e
n
ds
t
o
pro
voke
a
bno
rm
a
l
c
e
l
l
grow
t
h
.
T
he
r
e
a
r
e
s
e
v
e
r
a
l
f
a
c
t
ors
t
h
a
t
i
nc
r
e
a
s
e
t
h
e
ri
s
k
of
ge
t
t
i
n
g
bre
a
s
t
c
a
nc
e
r
s
uc
h
a
s
a
g
e
.
A
c
c
o
rdi
ng
t
o
r
e
s
e
a
rc
h
,
m
os
t
c
a
s
e
s
p
e
opl
e
a
r
e
di
a
gnos
e
d
a
f
t
e
r
t
he
a
ge
of
50
.
M
e
n
s
t
i
l
l
h
a
ve
a
r
i
s
k
of
ge
t
t
i
ng
br
e
a
s
t
c
a
n
c
e
r
e
v
e
n
t
hou
gh
i
t
i
s
a
l
o
t
l
ow
e
r
t
ha
n
w
o
m
e
n.
S
o
m
e
one
w
h
o
ha
d
e
a
rl
y
m
e
n
s
t
rua
l
pe
r
i
ods
b
e
for
e
t
h
e
a
g
e
of
12
a
n
d
s
t
a
rt
i
ng
m
e
no
pa
us
e
a
f
t
e
r
55
,
s
t
a
nd
a
h
i
ghe
r
r
i
s
k
of
b
e
i
ng
a
ff
e
c
t
e
d
.
Ra
d
i
a
t
i
on
t
he
r
a
py
i
s
a
l
s
o
a
n
ot
h
e
r
fa
c
t
o
r
w
h
i
c
h
m
a
ke
s
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
1693
-
6930
T
E
L
K
O
M
N
IK
A
T
e
l
e
c
om
m
un
Co
m
put
E
l
Con
t
rol
,
V
ol
.
18
,
N
o.
2
,
A
pri
l
2
020:
8
1
5
-
8
2
1
816
c
e
l
l
grow
s
a
bnor
m
a
l
l
y
.
F
urt
he
r
m
or
e
,
w
o
m
e
n
ha
v
e
a
hi
gh
e
r
r
i
s
k
of
g
e
t
t
i
ng
br
e
a
s
t
c
a
nc
e
r
a
s
s
um
i
n
g
he
r
fi
rs
t
re
l
a
t
i
ve
(m
o
t
he
r,
da
u
ght
e
r,
or
s
i
s
t
e
r
)
w
a
s
di
a
gnos
e
d
w
i
t
h
i
t
,
w
hi
c
h
i
n
m
os
t
c
a
s
e
s
i
s
unc
ha
ng
e
a
bl
e
.
T
he
r
e
a
re
a
l
s
o
s
om
e
f
a
c
t
ors
for
i
ns
t
a
nc
e
ov
e
rw
e
i
gh
t
w
o
m
e
n
a
ft
e
r
m
e
n
opa
us
e
s
t
a
nd
a
h
i
ghe
r
ri
s
k
t
h
a
n
t
hos
e
w
i
t
h
no
rm
a
l
b
ody
w
e
i
gh
t
.
Ca
r
e
s
hou
l
d
t
he
r
e
for
e
b
e
t
a
ke
n
by
t
hos
e
w
i
t
h
i
nc
re
a
s
e
d
h
orm
o
ne
s
a
ft
e
r
m
e
no
pa
us
e
,
a
s
i
t
r
a
i
s
e
s
t
h
e
ri
s
k
of
g
e
t
t
i
ng
a
ffe
c
t
e
d
by
b
re
a
s
t
c
a
n
c
e
r.
W
h
e
n
s
o
m
e
o
ne
ha
s
a
l
l
t
h
e
a
bov
e
m
e
n
t
i
o
ne
d
fa
c
t
o
rs
,
doe
s
n’t
m
e
a
n
t
he
y
a
re
s
uff
e
re
rs
,
a
nd
vi
c
e
v
e
rs
a
.
S
ym
pt
o
m
s
of
bre
a
s
t
c
a
nc
e
r
d
i
ff
e
r
fro
m
pe
rs
ons
.
H
ow
e
ve
r
,
s
om
e
c
o
m
m
o
n
s
ym
pt
o
m
s
i
nc
l
ude
s
ki
n
c
ha
n
ge
s
,
s
uc
h
a
s
s
w
e
l
l
i
ng,
re
d
ne
s
s
,
vi
s
i
b
l
e
di
ff
e
r
e
nc
e
s
i
n
o
ne
or
bot
h
bre
a
s
t
s
,
a
p
pe
a
ra
n
c
e
of
a
l
u
m
p
w
hi
c
h
doe
s
n’t
go
a
w
a
y
a
f
t
e
r
s
om
e
pe
ri
od
,
f
e
e
l
i
n
g
of
pa
i
n
or
bur
ni
ng
s
e
ns
a
t
i
on
a
rou
nd
t
he
b
re
a
s
t
a
re
a
e
ve
n
w
i
t
h
no
pre
s
s
ur
e
,
a
c
h
a
ng
e
i
n
t
he
ni
pp
l
e
,
i
t
c
h
e
s
,
e
t
c
.
O
n
c
e
yo
u
c
o
m
e
a
c
r
os
s
a
ny
of
t
he
s
e
s
ym
p
t
o
m
s
,
c
ons
ul
t
a
doc
t
or
i
m
m
e
di
a
t
e
l
y
.
T
he
t
re
a
t
m
e
nt
for
bre
a
s
t
c
a
n
c
e
r
i
s
di
ff
e
re
nt
a
nd
d
e
pe
n
de
n
t
o
n
t
he
t
yp
e
,
t
he
t
um
or
s
i
z
e
,
a
nd
how
fa
r
i
t
ha
s
s
pre
a
d
i
n
t
h
e
body
(s
t
a
ge
of
t
he
c
a
nc
e
r)
.
T
h
e
m
os
t
c
o
m
m
o
n
t
re
a
t
m
e
n
t
m
e
t
ho
d
i
s
s
urg
e
ry
,
w
h
i
c
h
i
s
us
e
d
t
o
r
e
m
ove
t
h
e
t
u
m
or
a
nd
t
i
s
s
ue
s
k
now
n
a
s
l
um
pe
c
t
o
m
y
or
t
he
w
ho
l
e
br
e
a
s
t
c
a
l
l
e
d
m
a
s
t
e
c
t
o
m
y.
In
a
d
di
t
i
on
,
onc
e
t
he
c
a
nc
e
r
ha
s
a
l
re
a
dy
s
pr
e
a
d
i
n
t
he
bod
y,
t
h
e
c
o
m
m
on
t
re
a
t
m
e
n
t
i
s
r
a
di
a
t
i
on
t
h
e
r
a
py,
t
he
i
nt
e
nt
i
on
i
s
t
o
ki
l
l
i
t
s
c
e
l
l
s
us
i
ng
hi
gh
e
n
e
rgy
w
a
v
e
s
.
T
h
e
ot
h
e
r
w
a
y
t
o
ki
l
l
c
a
nc
e
r
c
e
l
l
s
i
s
Che
m
o
t
he
r
a
py
,
w
hi
c
h
i
s
t
he
us
e
of
dru
gs
,
how
e
v
e
r
,
t
hi
s
t
r
e
a
t
m
e
nt
a
l
s
o
h
a
s
i
t
s
s
i
de
e
ffe
c
t
s
uc
h
a
s
h
a
i
r
l
os
s
,
e
a
rl
y
m
e
nop
a
us
e
,
a
nd
f
a
t
i
gue
.
T
he
us
e
of
m
e
di
c
i
n
e
t
o
pre
ve
n
t
hor
m
on
e
s
,
e
s
pe
c
i
a
l
l
y
e
s
t
roge
n,
a
l
s
o
w
o
rks
a
s
a
t
re
a
t
m
e
nt
.
But
s
a
d
l
y,
c
urr
e
nt
l
y
t
he
r
e
i
s
no
c
ure
for
c
a
n
c
e
r
c
om
p
l
e
t
e
l
y.
T
he
r
e
for
e
,
t
h
e
s
oone
r
t
h
e
be
t
t
e
r
t
o
k
now
s
o
m
e
one
i
s
s
uffe
r
i
ng
c
a
nc
e
r
or
no
t
,
s
o
i
t
c
a
n
be
t
r
e
a
t
e
d
e
a
r
l
y.
M
a
ny
m
a
c
hi
n
e
l
e
a
r
ni
ng
m
e
t
ho
ds
ha
v
e
b
e
e
n
a
pp
l
i
e
d
for
b
re
a
s
t
c
a
n
c
e
r
c
l
a
s
s
i
f
i
c
a
t
i
on
,
s
uc
h
a
s
S
uppor
t
V
e
c
t
or
M
a
c
h
i
ne
s
[1]
a
nd
N
e
t
w
ork
-
b
a
s
e
d
[2]
.
H
ow
e
ve
r
,
t
hi
s
re
s
e
a
rc
h
c
om
p
a
r
e
s
bo
t
h
i
n
t
e
r
m
s
of
a
c
c
u
ra
c
y.
S
V
M
i
s
a
l
r
e
a
d
y
a
w
i
de
l
y
know
n
m
e
t
hod
us
e
d
for
c
l
a
s
s
i
fi
c
a
t
i
on
s
uc
h
a
c
ut
e
s
i
nus
i
t
i
s
[3
],
f
a
c
e
i
d
e
n
t
i
f
i
c
a
t
i
on
[4]
,
pre
di
c
t
i
ng
b
a
nk
fa
i
l
ur
e
[
5],
Int
r
us
i
on
D
e
t
e
c
t
i
on
S
ys
t
e
m
[6,
7,
8]
,
Cl
a
s
s
i
fi
c
a
t
i
on
of
S
c
hi
z
ophr
e
ni
a
[9
],
D
e
t
e
c
t
i
on
of
T
r
a
ffi
c
Inc
i
de
nt
[10]
,
a
nd
F
a
c
e
Re
c
ogn
i
t
i
on
[11
,
12].
S
om
e
p
re
v
i
ous
s
t
udi
e
s
u
t
i
l
i
z
e
d
r
a
ndo
m
f
o
re
s
t
for
ge
n
e
s
e
l
e
c
t
i
on
a
nd
c
l
a
s
s
i
fi
c
a
t
i
on
[13]
,
c
l
a
s
s
i
f
i
c
a
t
i
on
of
a
ndro
i
d
m
a
l
w
a
r
e
[14]
,
pr
e
di
c
t
ba
n
k
fa
i
l
ure
[
15],
pre
d
i
c
t
pros
t
a
t
e
c
a
nc
e
r
[16
],
a
nd
os
t
e
oa
rt
hr
i
t
i
s
c
l
a
s
s
i
fi
c
a
t
i
on
[17]
.
T
hi
s
re
s
e
a
r
c
h
i
s
e
xpe
c
t
e
d
t
o
he
l
p
t
he
he
a
l
t
h
s
e
c
t
or
t
o
c
l
a
s
s
i
f
y
br
e
a
s
t
c
a
n
c
e
r
s
uff
e
re
rs
.
2.
R
ES
EA
R
C
H
M
ET
H
O
D
2.
1
.
D
at
a
T
he
d
a
t
a
i
n
t
h
i
s
s
t
udy
w
a
s
t
a
ke
n
f
rom
U
CI
m
a
c
hi
ne
l
e
a
r
ni
ng
r
e
pos
i
t
ory
[18]
.
T
h
e
da
t
a
c
ons
i
s
t
s
of
n
i
ne
fe
a
t
ure
s
,
a
s
f
ol
l
ow
s
:
−
A
ge
(Y
e
a
rs
)
−
BM
I
(kg
/
m
²
)
−
G
l
uc
os
e
(
m
g/
dL
)
−
Ins
ul
i
n
(µ
U
/
m
L
)
−
HOMA
−
L
e
p
t
i
n
(ng
/
m
L
)
−
A
di
pon
e
c
t
i
n
(µ
g
/
m
L
)
−
Re
s
i
s
t
i
n
(
ng/
m
L
)
−
M
CP
1
(
pg/
m
L
)
T
he
r
e
a
re
t
w
o
c
l
a
s
s
i
f
i
c
a
t
i
on
c
l
a
s
s
,
w
i
t
h
116
obs
e
rv
a
t
i
ons
c
ons
i
s
t
i
ng
of
52
h
e
a
l
t
hy
(
1)
a
n
d
64
pa
t
i
e
n
t
s
(2)
.
T
a
b
l
e
1
.
D
a
t
a
of
br
e
a
s
t
c
a
n
c
e
r
fro
m
U
CI
m
a
c
h
i
ne
l
e
a
r
ni
ng
r
e
pos
i
t
ory
A
g
e
BM
I
G
l
u
c
o
s
e
In
s
u
l
i
n
H
O
M
A
L
e
p
t
i
n
A
d
i
p
o
n
e
c
t
i
n
Re
s
i
s
t
i
n
M
CP
1
Cl
a
s
s
76
2
7
.
1
110
2
6
.
2
7
.
1
1
2
1
.
7
8
4
.
9
3
8
.
4
9
4
5
.
8
4
1
77
2
5
.
9
85
4
.
5
8
0
.
9
6
1
3
.
7
4
9
.
7
5
1
1
.
7
7
4
8
8
.
8
2
1
45
2
1
.
3
102
1
3
.
8
5
2
3
.
4
8
7
.
6
4
2
1
.
0
5
2
3
.
0
3
5
5
2
.
4
4
2
45
2
0
.
8
3
74
4
.
5
6
0
.
8
3
7
.
7
5
8
.
2
3
2
8
.
0
3
3
8
2
.
9
5
2
2.
2
.
S
u
p
e
r
vi
s
e
d
l
e
a
r
n
i
n
g
S
upe
rvi
s
e
d
l
e
a
r
ni
ng
i
s
a
m
e
t
hod
t
ha
t
prov
i
d
e
s
d
i
s
c
r
e
t
e
pr
e
di
c
t
i
on,
c
a
l
l
e
d
c
l
a
s
s
i
fi
c
a
t
i
on
.
It
s
p
l
i
t
t
h
e
da
t
a
i
nt
o
t
ra
i
ni
ng
,
w
hi
c
h
i
s
us
e
d
t
o
pr
e
di
c
t
m
ode
l
t
o
ob
t
a
i
n
t
h
e
be
s
t
p
a
ra
m
e
t
e
r
a
nd
t
he
t
e
s
t
d
a
t
a
w
he
r
e
t
h
e
obt
a
i
n
e
d
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
K
O
M
N
IK
A
T
e
l
e
c
om
m
un
Co
m
put
E
l
Con
t
rol
Com
par
i
ng
r
andom
f
or
e
s
t
and
s
uppor
t
v
e
c
t
or
m
a
c
hi
ne
s
…
(
Che
v
i
an
A
r
o
e
f
)
817
re
s
ul
t
s
a
re
a
p
pl
i
e
d
.
S
u
pe
rv
i
s
e
d
l
e
a
r
ni
n
g
k
e
e
ps
up
da
t
i
ng
i
t
s
e
l
f
t
o
m
a
ke
t
h
e
be
s
t
m
od
e
l
pos
s
i
b
l
e
,
a
nd
by
us
i
ng
i
t
,
a
n
e
w
d
a
t
a
ha
s
t
h
e
a
bi
l
i
t
y
of
b
e
i
ng
i
npu
t
t
e
d
a
nd
c
l
a
s
s
i
f
i
e
d
.
2.
3
.
D
e
c
i
s
i
on
tr
e
e
D
e
c
i
s
i
on
t
r
e
e
i
s
a
m
od
e
l
d
i
a
g
ra
m
t
h
a
t
c
ons
i
s
t
s
o
f
nod
e
a
n
d
e
d
ge
.
T
h
e
re
a
re
s
e
ve
r
a
l
t
yp
e
s
of
nod
e
s
s
uc
h
a
s
,
r
oot
,
p
a
r
e
nt
(i
n
t
e
r
na
l
),
a
n
d
c
hi
l
d
no
de
s
(l
e
a
f)
.
R
oot
nod
e
i
s
t
h
e
b
e
gi
nni
n
g
of
a
nod
e
t
ha
t
m
a
k
e
s
a
not
he
r
bra
nc
h,
know
n
a
s
t
h
e
pa
r
e
nt
no
de
.
T
hi
s
m
a
ke
s
a
no
t
he
r
br
a
n
c
h
k
now
n
a
s
c
hi
l
d
nod
e
,
w
hi
c
h
c
ons
i
s
t
s
of
r
i
ght
a
nd
l
e
ft
nod
e
s
.
F
ur
t
he
r
m
or
e
,
w
he
n
t
he
c
h
i
l
d
n
ode
doe
s
n’
t
ha
ve
a
n
y
br
a
nc
h
,
i
t
’s
c
a
l
l
e
d
t
e
rm
i
na
l
nod
e
.
F
i
gur
e
1
s
how
s
a
s
i
m
p
l
e
d
e
c
i
s
i
on
t
r
e
e
c
ons
i
s
t
s
of
root
nod
e
,
i
nt
e
rna
l
n
ode
a
nd
l
e
a
f
no
de
.
F
i
gure
1
.
D
i
a
gr
a
m
of
d
e
c
i
s
i
on
t
r
e
e
c
o
ns
i
s
t
s
of
r
oot
nod
e
,
i
nt
e
rna
l
nod
e
,
a
nd
l
e
a
f
nod
e
D
e
c
i
s
i
on
t
r
e
e
i
s
one
of
t
he
e
x
a
m
p
l
e
s
of
m
a
c
hi
n
e
l
e
a
r
ni
ng
m
o
de
l
,
w
hi
c
h
i
s
e
a
s
y
t
o
un
de
rs
t
a
nd
b
e
c
a
us
e
i
t
i
s
v
i
s
ua
l
i
z
e
d
.
It
fol
l
ow
s
t
h
e
rul
e
of
B
ool
e
a
n
a
l
ge
br
a
.
T
r
e
e
i
s
m
a
d
e
of
b
i
na
r
y
re
c
urs
i
v
e
pro
c
e
s
s
of
t
h
e
w
hol
e
da
t
a
s
o
t
h
e
va
r
i
a
b
l
e
f
rom
e
a
c
h
d
a
t
a
w
i
l
l
be
h
o
m
og
e
nou
s
[19]
.
M
a
k
i
ng
a
de
c
i
s
i
on
t
r
e
e
m
od
e
l
i
nvol
ve
t
he
s
e
pro
c
e
s
s
e
s
[19
]:
−
S
pl
i
t
t
h
e
pa
r
e
nt
no
de
i
n
t
o
c
h
i
l
d
nod
e
w
i
t
h
goo
dne
s
s
of
s
p
l
i
t
c
ri
t
e
ri
on
−
S
t
a
t
e
t
he
t
e
r
m
i
n
a
l
nod
e
w
h
i
c
h
i
s
t
he
l
a
s
t
node
of
t
h
e
de
c
i
s
i
on
t
hre
e
−
Cl
a
s
s
a
s
s
i
gn
m
e
n
t
A
l
gori
t
hm
for
m
a
ki
ng
t
h
e
d
e
c
i
s
i
on
t
re
e
m
od
e
l
i
s
us
i
n
g
di
vi
de
a
nd
c
onq
ue
r
re
c
urs
i
v
e
l
y
[20]
.
T
he
a
l
gor
i
t
h
m
fol
l
ow
s
t
he
s
t
e
ps
b
e
l
ow
:
−
Choos
e
a
fe
a
t
u
re
t
o
b
e
n
a
m
e
d
a
s
root
no
de
a
nd
m
a
k
e
bra
n
c
h
for
a
l
l
pos
s
i
b
l
e
f
e
a
t
ure
.
−
D
i
vi
d
e
t
he
t
r
a
i
ni
ng
s
e
t
w
i
t
h
on
e
s
e
t
for
a
bra
nc
h
.
−
Re
c
u
rs
i
ve
l
y
re
p
e
a
t
t
h
e
pr
oc
e
s
s
for
e
v
e
ry
br
a
nc
h
us
i
ng
i
t
s
da
t
a
.
−
S
t
op
w
h
e
n
al
l
t
h
e
d
at
a
h
as
sam
e
c
l
ass.
2.
4
.
B
or
u
ta
f
e
atu
r
e
s
e
l
e
c
t
i
on
Borut
a
f
e
a
t
ur
e
s
e
l
e
c
t
i
on
i
s
bu
i
l
t
a
round
t
he
r
a
ndo
m
f
ore
s
t
c
l
a
s
s
i
fi
c
a
t
i
on
a
l
gor
i
t
h
m
,
w
h
i
c
h
i
s
c
a
rr
i
e
d
out
w
i
t
hout
t
un
i
ng
of
p
a
r
a
m
e
t
e
rs
a
nd
num
e
ri
c
a
l
e
s
t
i
m
a
t
e
of
t
h
e
i
m
por
t
a
n
t
f
e
a
t
ure
.
R
a
ndo
m
for
e
s
t
i
s
a
c
l
a
s
s
i
fi
c
a
t
i
on
m
e
t
hod
w
hi
c
h
i
s
p
e
rfo
rm
e
d
by
vot
i
ng
o
f
m
ul
t
i
pl
e
unb
i
a
s
e
d
de
c
i
s
i
on
t
re
e
s
bui
l
t
f
rom
s
a
m
pl
e
s
of
t
he
t
r
a
i
ni
ng
s
e
t
[21]
.
T
h
e
i
m
port
a
n
c
e
of
t
he
fe
a
t
ure
i
s
ob
t
a
i
n
e
d
fro
m
t
he
l
os
s
of
a
c
c
ur
a
c
y
of
c
l
a
s
s
i
fi
c
a
t
i
on
.
In
a
ddi
t
i
o
n,
t
he
de
c
i
s
i
on
of
a
t
re
e
i
s
n’t
i
nfl
ue
n
c
e
d
b
y
a
n
ot
h
e
r
i
n
t
he
for
e
s
t
.
T
h
e
Boru
t
a
F
e
a
t
ur
e
S
e
l
e
c
t
i
on
a
l
gori
t
h
m
[2
2]:
−
Bui
l
d
e
xt
e
nde
d
s
ys
t
e
m
i
nfor
m
a
t
i
on
by
a
ddi
n
g
c
opi
e
s
o
f
a
l
l
f
e
a
t
ur
e
r
a
ndo
m
l
y
p
e
r
m
ut
e
d
a
c
ros
s
ob
j
e
c
t
s
.
−
S
huffl
e
t
he
s
ys
t
e
m
t
o
m
i
ni
m
i
z
e
t
he
c
or
re
l
a
t
i
ons
w
i
t
h
t
he
re
s
p
ons
e
−
P
e
rform
r
a
ndo
m
fore
s
t
t
o
o
bt
a
i
n
t
h
e
Z
s
c
or
e
s
c
om
put
e
d.
−
D
e
t
e
r
m
i
ne
t
h
e
m
a
xi
m
u
m
Z
s
c
o
re
f
or
t
he
e
xt
e
nd
e
d
f
e
a
t
ur
e
a
nd
a
s
s
i
gn
t
h
e
f
e
a
t
ure
a
s
s
u
m
i
ng
i
t
ha
s
a
b
e
t
t
e
r
s
c
ore
t
h
a
n
t
h
e
e
xt
e
nd
e
d.
F
ur
t
he
r
m
or
e
,
run
t
w
o
-
s
i
d
e
d
t
e
s
t
of
e
qu
a
l
i
t
y
us
i
ng
t
he
Z
s
c
ore
for
t
h
e
e
x
t
e
n
de
d
fe
a
t
ure
of
e
a
c
h
a
t
t
ri
bu
t
e
w
i
t
h
unde
t
e
r
m
i
ne
d
i
m
por
t
a
nc
e
.
−
L
a
b
e
l
t
h
e
f
e
a
t
ure
w
i
t
h
l
ow
e
r
Z
s
c
ore
for
t
he
e
x
t
e
nde
d
‘un
i
m
p
ort
a
n
t
’
f
e
a
t
ur
e
a
nd
r
e
m
ove
i
t
fro
m
t
he
s
ys
t
e
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
1693
-
6930
T
E
L
K
O
M
N
IK
A
T
e
l
e
c
om
m
un
Co
m
put
E
l
Con
t
rol
,
V
ol
.
18
,
N
o.
2
,
A
pri
l
2
020:
8
1
5
-
8
2
1
818
−
L
a
b
e
l
t
h
e
fe
a
t
ur
e
w
i
t
h
h
i
ghe
r
Z
s
c
ore
for
t
he
e
x
t
e
n
de
d
‘i
m
por
t
a
n
t
’
f
e
a
t
ur
e
.
−
Re
m
o
ve
a
l
l
c
o
pi
e
d
f
e
a
t
ur
e
a
nd
re
p
e
a
t
t
h
e
pr
oc
e
dur
e
t
i
l
l
non
e
i
s
re
m
ov
e
d
2.
5
.
R
an
d
om
fo
r
e
s
t
Ra
ndo
m
fore
s
t
w
a
s
f
i
rs
t
i
n
t
rodu
c
e
d
by
H
o
i
n
199
5
t
o
s
p
l
i
t
n
o
de
s
.
It
i
s
t
h
e
e
ns
e
m
b
l
e
of
m
a
ny
de
c
i
s
i
on
t
re
e
s
us
i
ng
boo
t
s
t
r
a
ppi
ng
a
nd
r
a
ndo
m
fe
a
t
ur
e
s
e
l
e
c
t
i
on
.
R
a
ndo
m
f
or
e
s
t
i
s
s
ui
t
a
bl
e
for
t
h
i
s
s
t
ud
y
b
e
c
a
us
e
i
t
p
e
rfor
m
s
w
e
l
l
on
l
a
rge
d
a
t
a
s
e
t
s
.
F
i
g
ure
2
s
how
s
a
di
a
gra
m
o
f
r
a
ndo
m
fo
re
s
t
w
hi
c
h
b
ui
l
t
of
m
a
n
y
de
c
i
s
i
on
t
r
e
e
.
F
i
gure
2
.
D
i
a
gr
a
m
of
r
a
ndo
m
fo
re
s
t
w
hi
c
h
c
o
ns
i
s
t
s
of
m
a
ny
de
c
i
s
i
on
t
r
e
e
s
Ra
ndo
m
f
ore
s
t
i
s
a
c
l
a
s
s
i
fi
e
r
c
ons
i
s
t
i
ng
of
c
l
a
s
s
i
f
i
c
a
t
i
on
t
re
e
w
he
r
e
i
s
a
v
e
c
t
or
w
i
t
h
i
nd
e
pe
nde
nt
l
y
a
n
d
i
d
e
n
t
i
c
a
l
l
y
d
i
s
t
r
i
but
e
d
w
i
t
h
e
a
c
h
t
re
e
v
ot
e
a
t
t
h
e
i
npu
t
.
T
h
e
a
c
c
ura
c
y
f
rom
de
c
i
s
i
on
t
r
e
e
i
s
m
or
e
s
t
a
b
l
e
a
nd
a
c
c
ur
a
t
e
.
H
ow
e
v
e
r
,
t
he
ra
nd
om
f
ore
s
t
p
e
rfor
m
s
b
e
t
t
e
r
a
c
c
ura
c
y
w
h
i
c
h
m
a
k
e
s
t
he
c
orr
e
l
a
t
i
on
of
t
re
e
s
i
gn
i
fi
c
a
n
t
.
By
for
m
i
ng
a
l
ot
of
t
re
e
i
nt
o
ra
ndo
m
fore
s
t
t
h
e
ri
s
k
o
f
ge
t
t
i
ng
ov
e
r
fi
t
t
i
ng
i
s
re
du
c
e
d,
w
i
t
h
t
h
e
e
rror
a
nd
c
onv
e
rg
e
i
n
t
o
s
o
m
e
va
l
u
e
g
e
ne
ra
l
i
z
e
d
.
G
i
v
e
n
e
ns
e
m
bl
e
c
l
a
s
s
i
fi
e
r
w
i
t
h
ra
nd
om
t
r
a
i
n
i
ng
d
a
t
a
ob
t
a
i
ne
d
fro
m
v
e
c
t
or
X
a
nd
Y
,
t
he
func
t
i
on
of
t
h
e
m
a
rgi
n
i
s
w
ri
t
t
e
n
a
s
:
(1)
W
he
r
e
i
s
t
he
i
ndi
c
a
t
or
fu
nc
t
i
on
a
nd
t
he
a
ve
r
a
g
e
,
w
i
t
h
t
he
r
e
s
ul
t
o
f
c
l
a
s
s
i
f
i
c
a
t
i
on,
w
h
e
r
e
Y
i
s
t
he
c
l
a
s
s
pre
di
c
t
i
on
a
nd
i
s
t
h
e
re
s
u
l
t
of
c
l
a
s
s
i
fi
c
a
t
i
on
w
i
t
h
j
.
M
a
rg
i
n
i
s
us
e
d
t
o
de
t
e
r
m
i
n
e
t
he
a
v
e
r
a
ge
va
l
u
e
of
vo
t
e
s
X
,
Y
.
A
gre
a
t
e
r
m
a
rgi
n
v
a
l
u
e
m
e
a
ns
a
m
o
re
a
c
c
ur
a
t
e
c
l
a
s
s
i
fi
c
a
t
i
on.
(2)
de
n
ot
e
s
t
he
ge
n
e
r
a
l
i
z
a
t
i
o
n
e
rror
,
w
hi
l
e
i
ndi
c
a
t
e
t
h
a
t
t
he
pro
ba
bi
l
i
t
y
i
s
m
or
e
t
ha
n
X
,
Y
di
m
e
ns
i
on
[23]
.
2.
6
.
S
u
p
p
or
t
v
e
c
to
r
ma
c
h
i
n
e
s
S
upport
V
e
c
t
or
M
a
c
h
i
ne
s
i
s
on
e
of
t
h
e
s
up
e
rv
i
s
e
d
l
e
a
r
ni
ng
m
e
t
hods
w
i
d
e
l
y
i
n
t
rod
uc
e
d
by
V
a
pn
i
k
i
n
t
he
l
a
t
e
9
0s
.
D
uri
ng
i
t
s
e
a
r
l
y
da
ys
,
i
t
w
a
s
us
e
d
on
l
y
for
c
l
a
s
s
i
fi
c
a
t
i
on,
how
e
ve
r,
i
t
h
a
s
d
e
v
e
l
op
e
d
,
a
nd
c
a
pa
bl
e
of
s
ol
v
i
ng
re
gr
e
s
s
i
on
prob
l
e
m
s
.
S
V
M
t
ry
t
o
s
ol
v
e
t
he
c
l
a
s
s
i
f
i
c
a
t
i
o
n
pr
obl
e
m
by
form
i
ng
a
hyp
e
rpl
a
ne
w
hi
c
h
m
a
x
i
m
i
z
e
s
t
h
e
m
a
rg
i
n
by
di
v
i
di
ng
t
h
e
d
a
t
a
i
nt
o
c
l
a
s
s
e
s
.
T
h
e
ne
a
r
e
s
t
d
i
s
t
a
nc
e
fro
m
t
he
h
ype
r
pl
a
ne
t
o
t
he
poi
n
t
of
e
a
c
h
c
l
a
s
s
i
s
know
n
a
s
m
a
rgi
n.
F
i
gu
re
3
i
s
a
n
i
l
l
us
t
r
a
t
i
on
o
f
S
V
M
m
ode
l
.
G
i
ve
n
a
da
t
a
s
e
t
S
V
M
t
r
y
t
o
s
ol
v
e
t
he
fol
l
ow
i
ng
e
qu
a
t
i
on:
(3)
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
K
O
M
N
IK
A
T
e
l
e
c
om
m
un
Co
m
put
E
l
Con
t
rol
Com
par
i
ng
r
andom
f
or
e
s
t
and
s
uppor
t
v
e
c
t
or
m
a
c
hi
ne
s
…
(
Che
v
i
an
A
r
o
e
f
)
819
F
or
s
om
e
e
rror
c
a
s
e
s
,
p
a
r
a
m
e
t
e
r
a
nd
s
l
a
c
k
va
r
i
a
b
l
e
w
a
s
a
dd
e
d
t
o
t
he
e
qu
a
t
i
on
:
(4)
K
e
rne
l
f
unc
t
i
o
n
i
s
us
e
d
for
s
om
e
p
robl
e
m
w
h
i
c
h
c
a
n’
t
b
e
s
o
l
ve
d
us
i
ng
l
i
ne
a
r
hype
rpl
a
ne
.
It
s
fun
c
t
i
on
i
s
de
fi
n
e
d
a
s
:
(5)
T
he
s
e
a
re
t
h
e
m
os
t
c
om
m
on
k
e
rne
l
func
t
i
o
n
[2
4,
25]
:
-
L
i
n
e
a
r
:
-
Ra
di
a
l
B
a
s
i
s
F
un
c
t
i
on
:
-
P
ol
yno
m
i
a
l
:
F
i
gure
3
.
S
V
M
s
o
l
v
e
c
l
a
s
s
i
fi
c
a
t
i
on
pro
bl
e
m
by
pe
rf
orm
i
ng
a
hype
rp
l
a
n
e
w
hi
c
h
m
a
xi
m
i
z
e
t
he
m
a
rg
i
n
o
f
t
he
d
a
t
a
2.
7
.
M
od
e
l
p
e
r
f
or
m
an
c
e
va
l
i
d
at
i
on
T
hi
s
s
t
udy
a
ppl
i
e
d
H
ol
d
-
O
ut
V
a
l
i
d
a
t
i
on
t
o
v
a
l
i
da
t
e
t
h
e
p
e
rfor
m
a
n
c
e
of
t
he
m
o
de
l
.
It
fun
c
t
i
ons
by
s
p
l
i
t
t
i
ng
t
h
e
da
t
a
i
nt
o
t
ra
i
ni
ng
a
n
d
t
e
s
t
i
ng
d
a
t
a
w
i
t
h
t
h
e
m
ode
l
bu
i
l
t
f
rom
t
he
t
ra
i
ni
ng
da
t
a
a
nd
t
e
s
t
e
d
w
i
t
h
t
he
t
e
s
t
i
ng
da
t
a
.
A
di
f
fe
r
e
nt
pe
r
c
e
nt
a
ge
of
t
ra
i
ni
ng
d
a
t
a
w
a
s
a
ppl
i
e
d
,
t
o
ove
r
c
o
m
e
t
h
e
w
e
a
kn
e
s
s
of
H
o
l
d
-
O
u
t
V
a
l
i
d
a
t
i
on
,
w
hi
c
h
w
a
s
p
e
rfor
m
e
d
ni
ne
t
i
m
e
s
w
i
t
h
di
ffe
re
nt
p
e
r
c
e
n
t
a
g
e
of
t
he
t
ra
i
n
i
ng
d
a
t
a
ut
i
l
i
z
e
d.
T
he
pe
rfor
m
a
n
c
e
of
t
h
e
m
od
e
l
i
s
ob
t
a
i
ne
d
fro
m
t
h
e
a
c
c
ur
a
c
y
w
i
t
h
t
h
e
f
orm
u
l
a
w
ri
t
t
e
n
a
s
be
l
ow
:
(6)
F
or
T
P
,
T
N
,
F
P
,
F
N
r
e
pr
e
s
e
nt
s
T
r
ue
P
os
i
t
i
ve
,
T
ru
e
N
e
g
a
t
i
ve
,
F
a
l
s
e
P
os
i
t
i
v
e
,
a
nd
F
a
l
s
e
N
e
ga
t
i
v
e
.
3.
R
ES
U
LTS
A
N
D
A
N
A
LY
S
I
S
T
hi
s
s
t
udy
us
e
d
s
of
t
w
a
r
e
Rs
t
udi
o
v
e
rs
i
o
n
1
.
1
.
463
for
bo
t
h
r
a
n
dom
f
ore
s
t
a
nd
s
uppor
t
ve
c
t
or
m
a
c
hi
n
e
s
.
3.
1
.
R
e
s
u
l
t
of
b
o
r
u
ta
fe
atu
r
e
s
e
l
e
c
ti
on
T
he
re
s
u
l
t
o
f
Bor
ut
a
F
e
a
t
ur
e
S
e
l
e
c
t
i
on
de
t
e
r
m
i
n
e
s
w
h
e
t
h
e
r
t
he
f
e
a
t
ure
i
s
i
m
p
ort
a
nt
or
no
t
a
s
s
how
n
i
n
T
a
bl
e
2
.
A
c
c
ord
i
ng
t
o
t
h
e
r
e
s
ul
t
s
h
ow
n
on
t
he
t
a
b
l
e
,
t
he
i
m
port
a
nt
f
e
a
t
ure
s
a
re
A
ge
,
BM
I
,
G
l
u
c
os
e
,
H
O
M
A
,
a
nd
R
e
s
i
s
t
i
n.
H
ow
e
v
e
r
,
t
he
fe
a
t
ur
e
s
w
hi
c
h
a
r
e
l
a
be
l
e
d
i
m
port
a
n
t
t
o
m
a
k
e
t
h
e
m
a
c
h
i
ne
l
e
a
rn
i
ng
m
od
e
l
a
re
r
a
ndom
f
or
e
s
t
a
n
d
s
up
port
v
e
c
t
or
m
a
c
h
i
ne
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
1693
-
6930
T
E
L
K
O
M
N
IK
A
T
e
l
e
c
om
m
un
Co
m
put
E
l
Con
t
rol
,
V
ol
.
18
,
N
o.
2
,
A
pri
l
2
020:
8
1
5
-
8
2
1
820
T
a
b
l
e
2
.
R
e
s
ul
t
o
f
bo
rut
a
f
e
a
t
ur
e
s
e
l
e
c
t
i
on
F
e
a
t
u
re
Re
s
u
l
t
A
g
e
Im
p
o
rt
a
n
t
BM
I
Im
p
o
rt
a
n
t
G
l
u
c
o
s
e
Im
p
o
rt
a
n
t
In
s
u
l
i
n
U
n
i
m
p
o
rt
a
n
t
H
O
M
A
Im
p
o
rt
a
n
t
L
e
p
t
i
n
U
n
i
m
p
o
rt
a
n
t
A
d
i
p
o
n
e
c
t
i
n
U
n
i
m
p
o
rt
a
n
t
Re
s
i
s
t
i
n
Im
p
o
rt
a
n
t
M
CP
.
1
U
n
i
m
p
o
rt
a
n
t
3.
2
.
B
r
e
as
t
c
an
c
e
r
c
l
as
s
i
fi
c
at
i
on
u
s
i
n
g
r
an
d
om
f
or
e
s
t
A
c
c
ord
i
ng
t
o
t
h
e
r
e
s
ul
t
s
how
n
on
t
he
F
i
gur
e
4
,
t
h
e
m
ode
l
p
e
rform
e
d
b
e
s
t
w
i
t
h
80%
of
t
r
a
i
n
i
ng
da
t
a
,
re
s
ul
t
i
ng
i
n
a
n
a
c
c
ur
a
c
y
of
9
0.
9
0%.
Conv
e
rs
e
l
y
,
t
he
w
o
rs
t
a
c
c
u
ra
c
y
w
a
s
re
c
ord
e
d
a
t
74.
7
5%
w
i
t
h
10%
t
ra
i
ni
ng
d
a
t
a
.
F
i
gure
4
.
R
e
s
ul
t
s
of
a
c
c
ura
c
y
of
bre
a
s
t
c
a
nc
e
r
c
l
a
s
s
i
fi
c
a
t
i
o
n
p
e
rfor
m
e
d
by
ra
ndo
m
fo
re
s
t
3.
3
.
B
r
e
as
t
c
an
c
e
r
c
l
as
s
i
fi
c
at
i
on
u
s
i
n
g
s
u
p
p
or
t
ve
c
to
r
m
a
c
h
i
n
e
s
A
c
c
ord
i
ng
t
o
t
he
re
s
u
l
t
s
how
n
on
t
h
e
F
i
gur
e
5
,
t
h
e
m
od
e
l
p
e
rform
e
d
be
s
t
us
i
ng
80%
of
t
h
e
t
ra
i
ni
ng
da
t
a
w
h
i
c
h
r
e
s
ul
t
e
d
i
n
a
n
a
c
c
ura
c
y
of
95.
45%
.
Conv
e
rs
e
l
y,
t
h
e
w
ors
t
a
c
c
ura
c
y
i
s
re
c
ord
e
d
a
t
72
.
81%
w
i
t
h
10%
t
ra
i
ni
ng
d
a
t
a
.
F
i
gure
5
.
R
e
s
ul
t
s
of
a
c
c
ura
c
y
of
bre
a
s
t
c
a
nc
e
r
c
l
a
s
s
i
fi
c
a
t
i
o
n
p
e
rfor
m
e
d
by
s
uppor
t
ve
c
t
or
m
a
c
hi
n
e
s
w
i
t
h
RBF
ke
rne
l
,
p
a
ra
m
e
t
e
r
a
n
d
4.
C
O
N
C
LU
S
I
O
N
T
hi
s
s
t
u
dy
us
e
d
Ra
n
dom
F
or
e
s
t
(
RF
)
a
nd
S
uppor
t
V
e
c
t
or
M
a
c
h
i
ne
s
(S
V
M
)
a
s
t
h
e
m
a
c
h
i
n
e
l
e
a
rn
i
ng
m
e
t
hods
t
o
c
l
a
s
s
i
fy
br
e
a
s
t
c
a
n
c
e
r
.
F
urt
h
e
r
m
ore
,
t
h
e
H
o
l
d
-
O
ut
V
a
l
i
da
t
i
on
w
a
s
us
e
d
t
o
v
a
l
i
da
t
e
a
nd
e
va
l
u
a
t
e
t
he
p
e
rfor
m
a
nc
e
of
t
he
m
od
e
l
,
fro
m
t
he
s
i
m
u
l
a
t
i
o
n
for
S
V
M
w
i
t
h
R
a
di
a
l
Ba
s
i
s
F
u
nc
t
i
on
(
RBF
)
k
e
rne
l
,
w
i
t
h
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
K
O
M
N
IK
A
T
e
l
e
c
om
m
un
Co
m
put
E
l
Con
t
rol
Com
par
i
ng
r
andom
f
or
e
s
t
and
s
uppor
t
v
e
c
t
or
m
a
c
hi
ne
s
…
(
Che
v
i
an
A
r
o
e
f
)
821
a
nd
a
s
t
h
e
be
s
t
p
a
ra
m
e
t
e
r
of
t
h
e
m
od
e
l
.
A
c
c
ord
i
ng
t
o
t
he
e
xpe
r
i
m
e
n
t
re
s
ul
t
,
RF
s
c
ore
d
t
he
b
e
s
t
a
c
c
ur
a
c
y
a
t
90
.
90%
us
i
ng
80%
t
ra
i
ni
ng
d
a
t
a
,
w
h
i
l
e
S
V
M
ha
d
b
e
t
t
e
r
a
c
c
ura
c
y
a
t
9
5.
4
5%
us
i
ng
80
%.
T
he
s
e
r
e
s
ul
t
s
s
how
t
h
a
t
t
h
e
pe
rfo
rm
a
nc
e
of
S
V
M
i
s
be
t
t
e
r
t
ha
n
RF
i
n
t
e
rm
s
of
a
c
c
ur
a
c
y
.
A
C
K
N
O
WL
ED
G
E
M
EN
TS
T
hi
s
re
s
e
a
r
c
h
s
uppor
t
e
d
f
i
na
nc
i
a
l
l
y
by
t
he
M
i
n
i
s
t
ry
of
R
e
s
e
a
rc
h
a
nd
H
i
ghe
r
E
duc
a
t
i
on
R
e
pub
l
i
c
of
In
done
s
i
a
(K
E
M
E
N
RIS
T
E
K
D
IK
T
I)
w
i
t
h
a
P
T
U
P
T
202
0
r
e
s
e
a
rc
h
gra
n
t
s
c
h
e
m
e
,
ID
nu
m
b
e
r
1621/
U
N
2
.
R3
.
1/
H
K
P
.
05.
00/
2
019
R
EF
ER
EN
C
ES
[
1]
M
.
F
.
A
k
a
y,
“
S
uppo
r
t
v
e
c
t
or
m
a
c
hi
ne
s
c
o
m
b
i
n
e
d
w
i
t
h
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
f
or
b
r
e
a
s
t
c
a
nc
e
r
di
a
gn
os
i
s
,
”
E
x
p
e
r
t
s
y
s
t
e
m
s
w
i
t
h
a
ppl
i
c
at
i
on
s
,
vo
l
.
36
,
n
o.
2,
pp.
3
240
-
324
7
,
20
09.
[
2]
H
-
Y
.
C
h
ua
n
g,
E
.
L
e
e
,
Y
-
T
.
L
i
u
,
D
.
L
e
e
,
a
nd
T
.
I
de
ke
r
,
“
N
e
t
w
or
k
-
ba
s
e
d
c
l
a
s
s
i
f
i
c
a
t
i
on
o
f
b
r
e
a
s
t
c
a
nc
e
r
m
e
t
a
s
t
a
s
i
s
,
”
M
o
l
.
Sy
s
t
.
B
i
ol
.
,
v
ol
.
3
,
no
.
140
,
pp
.
1
-
10,
2
007
.
[
3]
A
r
f
i
a
n
i
,
Z
.
R
u
s
t
a
m
,
J
.
P
a
nde
l
a
ki
,
a
nd
A
.
S
i
a
ha
a
n
,
“
K
e
r
n
e
l
s
p
he
r
i
c
a
l
K
-
m
e
a
ns
a
nd
s
uppo
r
t
ve
c
t
or
m
a
c
h
i
n
e
f
or
a
c
ut
e
s
i
nu
s
i
t
i
s
c
l
a
s
s
i
f
i
c
a
t
i
on
,”
I
O
P
C
on
f
e
r
e
nc
e
Se
r
i
e
s
:
M
a
t
e
r
i
a
l
s
Sc
i
e
nc
e
and
E
ng
i
ne
e
r
i
ng
,
vol
.
546
,
pp
.
1
-
9
,
201
9.
[
4]
T
z
ot
s
os
,
A
nge
l
os
&
A
r
g
i
a
l
a
s
,
D
e
m
e
t
r
e
,
“
S
up
por
t
V
e
c
t
or
M
a
c
hi
n
e
C
l
a
s
s
i
f
i
c
a
t
i
on
f
o
r
O
b
j
e
c
t
-
B
a
s
e
d
I
m
a
g
e
,”
O
b
j
e
c
t
-
B
a
s
e
d
I
m
ag
e
A
n
al
y
s
i
s
,
pp
.
663
-
677
,
J
a
nua
r
y
20
08
.
[
5]
Z
.
R
us
t
a
m
,
F
.
N
a
d
hi
f
a
a
nd
M
.
A
c
a
r
,
“
C
o
m
pa
r
i
s
on
of
S
V
M
a
nd
F
S
V
M
f
o
r
p
r
e
d
i
t
c
i
ng
b
a
nk
f
a
i
l
ur
e
s
u
s
i
ng
c
h
i
-
s
qu
a
r
e
f
e
a
t
u
r
e
s
e
l
e
c
t
i
on
,
”
I
O
P
C
o
nf
e
r
e
nc
e
Se
r
i
e
s
:
J
ou
r
na
l
of
P
h
y
s
i
c
s
,
vo
l
.
1108
,
no.
1
,
pp
.
1
-
7
,
201
8.
[
6]
J
.
M
a
h
a
r
a
ni
a
nd
Z
.
R
us
t
a
m
,
“
T
h
e
a
pp
l
i
c
a
t
i
on
o
f
m
ul
t
i
-
c
l
a
s
s
s
uppo
r
t
ve
c
t
o
r
m
a
c
hi
n
e
s
o
n
i
n
t
r
us
i
on
d
e
t
e
c
t
i
o
n
s
y
s
t
e
m
w
i
t
h
t
h
e
f
e
a
t
ur
e
s
e
l
e
c
t
i
on
u
s
i
ng
i
nf
or
m
a
t
i
on
g
a
i
n
,
”
1s
t
A
nnu
al
I
n
t
e
r
nat
i
o
nal
C
onf
e
r
e
nc
e
on
M
a
t
he
m
a
t
i
c
s
,
S
c
i
e
nc
e
,
and
E
duc
at
i
on
(
I
C
o
M
SE
)
,
v
ol
.
218
,
pp
.
1
-
2
,
2017
.
[
7]
Z
.
R
us
t
a
m
a
nd
N.
O
l
i
e
v
r
a
,
“
C
o
m
p
a
r
i
s
o
n
o
f
f
u
z
z
y
r
obus
t
ke
r
ne
l
C
-
m
e
a
ns
a
n
d
s
upp
or
t
ve
c
t
o
r
m
a
c
h
i
ne
f
o
r
i
n
t
r
us
i
on
de
t
e
c
t
i
o
n
s
y
s
t
e
m
s
u
s
i
n
g
m
od
i
f
i
e
d
ke
r
ne
l
ne
a
r
e
s
t
ne
i
gh
bor
f
e
a
t
ur
e
s
e
l
e
c
t
i
on
,
”
A
I
P
C
on
f
e
r
e
nc
e
P
r
oc
e
e
d
i
ngs
:
3r
d
I
n
t
e
r
na
t
i
ona
l
Sy
m
po
s
s
i
um
on
C
ur
r
e
n
t
P
r
o
gr
e
s
s
i
n
M
a
t
he
m
at
i
c
s
and
S
c
i
e
nc
e
s
(
I
SC
P
M
S)
,
pp
.
0202
15
-
1
-
6,
20
17
.
[
8]
Z
.
R
us
t
a
m
a
n
d
N
.
P
.
A
.
A
.
A
r
i
a
nt
a
r
i
.
C
o
m
p
a
r
i
s
o
n
B
e
t
w
e
e
n
S
u
ppo
r
t
V
e
c
t
o
r
M
a
c
h
i
ne
a
nd
F
u
z
z
y
K
e
r
ne
l
C
-
M
e
a
ns
a
s
C
l
a
s
s
i
f
i
e
r
f
o
r
I
nt
r
us
i
on
D
e
t
e
c
t
i
on
S
y
s
t
e
m
us
i
ng
C
hi
-
S
q
ua
r
e
F
e
a
t
u
r
e
S
e
l
e
c
t
i
o
n,
A
I
P
C
onf
e
r
e
nc
e
P
r
o
c
e
e
di
ngs
:
3
r
d
I
n
t
e
r
na
t
i
ona
l
Sy
m
po
s
s
i
um
on
C
ur
r
e
n
t
P
r
o
gr
e
s
s
i
n
M
a
t
he
m
at
i
c
s
and
S
c
i
e
nc
e
s
(
I
SC
P
M
S)
,
vo
l
.
2
023
,
2018
.
[
9]
T
.
V
.
R
a
m
p
i
s
e
l
a
a
n
d
Z
.
R
us
t
a
m
,
“
C
l
a
s
s
i
f
i
c
a
t
i
o
n
o
f
s
c
hi
z
oph
r
e
n
i
a
d
a
t
a
u
s
i
n
g
s
upp
or
t
ve
c
t
o
r
m
a
c
h
i
ne
(
S
V
M
)
,
”
I
O
P
C
onf
e
r
e
n
c
e
Se
r
i
e
s
:
J
ou
r
na
l
of
P
hy
s
i
c
s
,
vo
l
.
11
08
,
p
p.
1
-
7,
2
018
.
[
10]
J
.
X
i
a
o
,
“
S
V
M
a
n
d
K
N
N
e
ns
e
m
bl
e
l
e
a
r
ni
ng
f
or
t
r
a
f
f
i
c
i
n
c
i
de
n
t
de
t
e
c
t
i
on
,
”
P
h
y
s
i
c
a
A
:
S
t
a
t
i
s
t
i
c
a
l
M
e
c
han
i
c
s
and
i
t
s
A
ppl
i
c
a
t
i
ons
,
E
l
s
e
v
i
e
r
,
v
ol
.
517
,
p
p.
29
-
35,
2
019
.
[
11]
Z
.
R
us
t
a
m
a
nd
A
.
A
.
R
uv
i
t
a
,
“
A
pp
l
i
c
a
t
i
on
s
uppo
r
t
ve
c
t
o
r
m
a
c
hi
n
e
on
f
a
c
e
r
e
c
o
gni
t
i
on
f
or
ge
n
de
r
c
l
a
s
s
i
f
i
c
a
t
i
on
.
C
onf
e
r
e
n
c
e
Se
r
i
e
s
:
J
ou
r
na
l
of
P
hy
s
i
c
s
,
vo
l
.
11
08,
p
p.
1
-
5
,
2
018
.
[
12]
R
us
t
a
m
,
Z
u
he
r
m
a
n
&
F
a
r
a
di
n
a
,
R
i
d
ha
n
i
.
(
20
18)
.
F
a
c
e
R
e
c
o
gni
t
i
on
t
o
I
d
e
nt
i
f
y
L
oo
k
-
A
l
i
k
e
F
a
c
e
s
us
i
ng
S
upp
or
t
V
e
c
t
o
r
M
a
c
h
i
ne
.
J
o
ur
n
a
l
o
f
P
hys
i
c
s
:
C
on
f
e
r
e
n
c
e
S
e
r
i
e
s
.
1108
.
012
0
71.
10
.
10
88/
1742
-
6
596
/
110
8/
1
/
0
1207
1.
[
13]
R
.
D
i
a
z
-
U
r
i
a
r
t
e
a
nd
S
.
A
.
de
A
nd
r
e
s
,
“
G
e
ne
s
e
l
e
c
t
i
on
a
nd
c
l
a
s
s
i
f
i
c
a
t
i
on
o
f
m
i
c
r
oa
r
r
a
y
da
t
a
us
i
n
g
r
a
n
do
m
f
o
r
e
s
t
,”
B
M
C
B
i
o
i
n
f
or
m
at
i
c
s
,
vo
l
.
7
,
no
.
3,
pp
.
1
-
13
,
2006
.
[
14]
M
.
A
l
a
m
a
n
d
S
.
T
.
V
uo
ng
,
“
R
a
ndo
m
f
o
r
e
s
t
c
l
a
s
s
i
f
i
c
a
t
i
o
n
f
or
d
e
t
e
c
t
i
ng
a
n
dr
o
i
d
m
a
l
w
a
r
e
,
”
I
E
E
E
I
nt
.
C
onf
.
on
G
r
e
e
n
C
om
p
.
and
C
om
m
.
and
I
E
E
E
I
n
t
e
r
ne
t
o
f
T
h
i
ng
s
a
nd
I
E
E
E
C
y
be
r
,
P
hy
s
i
c
a
l
and
So
c
.
C
om
p
.
,
p
p.
66
3
-
669
,
2013
.
[
15]
Z
.
R
us
t
a
m
a
nd
G
.
S
a
r
a
g
i
h,
“
P
r
e
d
i
c
t
i
n
g
B
a
nk
F
i
n
a
nc
i
a
l
F
a
i
l
ur
e
s
u
s
i
ng
R
a
n
d
om
F
or
e
s
t
,
”
I
n
t
e
r
na
t
i
ona
l
W
or
k
s
ho
p
on
B
i
g
D
at
a
a
nd
I
n
f
o
r
m
at
i
on
S
e
c
u
r
i
t
y
(
I
W
B
I
S)
,
pp.
81
-
8
6,
20
1
8.
[
16]
M
.
H
u
l
j
a
na
h,
Z
.
R
us
t
a
m
,
S
.
U
t
a
m
a
,
a
n
d
T
.
S
i
s
w
a
nt
i
n
i
ng
,
“
F
e
a
t
u
r
e
s
e
l
e
c
t
i
on
us
i
ng
r
a
nd
om
f
or
e
s
t
c
l
a
s
s
i
f
i
e
r
f
or
pr
e
d
i
c
t
i
ng
pr
os
t
a
t
e
c
a
nc
e
r
,
”
I
O
P
C
on
f
e
r
e
nc
e
Se
r
i
e
s
:
M
at
e
r
i
a
l
s
Sc
i
e
nc
e
and
E
n
gi
n
e
e
r
i
n
g
,
vo
l
.
5
46
,
p
p.
1
-
8,
2019
.
[
17]
U
.
A
p
r
i
l
l
i
a
ni
a
n
d
Z
.
R
us
t
a
m
,
“
O
s
t
e
oa
r
t
hr
i
t
i
s
d
i
s
e
a
s
e
pr
e
di
c
t
i
on
ba
s
e
d
o
n
r
a
n
do
m
f
o
r
e
s
t
,
”
I
n
t
e
r
na
t
i
o
na
l
C
o
nf
e
r
e
nc
e
on
A
d
v
an
c
e
d
C
om
p
ut
e
r
Sc
i
e
n
c
e
and
I
nf
or
m
a
t
i
on
Sy
s
t
e
m
s
(
I
C
A
C
SI
S
)
,
pp
.
237
-
24
0,
20
18.
[
18]
U
C
I
M
a
c
h
i
ne
L
e
a
r
n
i
ng
R
e
po
s
i
t
o
r
y
,
A
c
c
e
s
s
e
d
10
A
ugus
t
201
9,
A
v
a
i
l
a
b
l
e
a
t
h
t
t
p
s
:
/
/
a
r
c
h
i
ve
.
i
c
s
.
u
c
i
.
e
d
u/
m
l
/
da
t
a
s
e
t
s
/
B
r
e
a
s
t
+
C
a
nc
e
r
+
C
oi
m
br
a
.
[
19]
B
.
S
.
E
ve
r
i
t
t
,
“
C
l
a
s
s
i
f
i
c
a
t
i
on
a
n
d
r
e
g
r
e
s
s
i
on
t
r
e
e
s
,”
J
oh
n
W
i
l
e
y
&
S
ons
,
I
nc
.
,
2
005
.
[
20]
C
.
B
i
s
h
op
,
“
P
a
t
t
e
r
n
re
c
ogn
i
t
i
on
a
nd
m
a
c
h
i
ne
l
e
a
r
n
i
ng
,
”
S
p
r
i
nge
r
,
N
e
w
Y
o
r
k,
2
006
.
[
21]
M
.
B
.
K
u
r
s
a
a
nd
W
.
R
.
R
u
dn
i
c
k
i
,
“
F
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
w
i
t
h
B
or
ut
a
p
a
c
ka
ge
,”
J
o
ur
n
al
o
f
S
t
at
i
s
t
i
c
a
l
So
f
t
w
a
r
e
,
vo
l
.
36,
no.
11
,
pp
.
1
-
1
3,
20
10.
[
22]
M
.
B
.
K
ur
s
a
,
A.
J
a
nkow
s
k
i
,
a
nd
W
.
R
udn
i
c
ki
,
“
B
o
r
u
t
a
-
a
s
y
s
t
e
m
f
o
r
f
e
a
t
u
r
e
s
e
l
e
c
t
i
on
,”
F
und
am
.
I
nf
or
m
.
,
v
ol
.
101
,
no.
4
,
pp.
2
71
-
285
,
2010
.
[
23]
L
.
B
r
e
i
m
a
n
,
“
R
a
n
do
m
f
o
r
e
s
t
,
”
M
ac
h
i
ne
L
e
ar
ni
n
g
,
vo
l
.
4
5,
pp
.
5
-
32
,
200
1.
[
24]
N
.
C
r
i
s
t
i
a
n
i
n
i
a
nd
J
.
S
.
T
a
yl
o
r
,
“
A
n
i
n
t
r
o
duc
t
i
on
t
o
s
uppo
r
t
v
e
c
t
o
r
m
a
c
h
i
ne
s
a
nd
o
t
he
r
ke
r
ne
l
-
b
a
s
e
d
l
e
a
r
n
i
ng
m
e
t
ho
ds
,
”
C
am
br
i
dg
e
U
ni
v
e
r
s
i
t
y
P
r
e
s
s
,
20
00.
[
25]
A
.
B
e
n
-
H
u
r
a
nd
J
.
W
e
s
t
on
,
“
A
u
s
e
r
’
s
gu
i
de
t
o
s
uppo
r
t
ve
c
t
or
m
a
c
h
i
ne
s
,
”
M
e
t
hod
s
i
n
M
o
l
e
c
u
l
ar
B
i
ol
o
gy
(
N
.
J
.
C
l
i
f
t
on
)
,
vo
l
.
60
9,
pp
.
223
-
2
3
9,
20
10
.
Evaluation Warning : The document was created with Spire.PDF for Python.