T
E
L
K
O
M
NIKA
T
elec
o
mm
un
ica
t
io
n Co
m
pu
t
i
ng
E
lect
ro
nics
a
nd
Co
ntr
o
l
Vo
l.
24
,
No
.
3
,
J
u
n
e
20
26
,
p
p
.
891
~
9
03
I
SS
N:
1
6
9
3
-
6
9
3
0
,
DOI
: 1
0
.
1
2
9
2
8
/
T
E
L
KOM
NI
K
A
.
v
24
i
3
.
27584
891
J
o
ur
na
l
ho
m
ep
a
g
e
:
h
ttp
:
//jo
u
r
n
a
l.u
a
d
.
a
c.
id
/in
d
ex
.
p
h
p
/TELK
OM
N
I
K
A
Cla
ss
ificatio
n
o
f
p
re
m
a
ture
ca
rdia
c
co
ntraction
s ba
se
d
o
n
RFECV
and
ense
m
ble
learning
E
ls
a
Sa
ri
H
a
y
un
a
h Nurdi
niy
a
h,
A
’
is
y
a
Nur
Aulia
Yus
uf,
No
r
m
a
A
m
a
lia
,
Wi
dh
ia
t
m
o
ko
H
er
ry
P
urno
m
o
,
Aziz
a
h Na
j
da
H
a
f
izha
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
i
c
a
l
En
g
i
n
e
e
r
i
n
g
,
F
a
c
u
l
t
y
o
f
En
g
i
n
e
e
r
i
n
g
,
U
n
i
v
e
r
si
t
as
J
e
n
d
e
r
a
l
S
o
e
d
i
r
man
,
P
u
r
b
a
l
i
n
g
g
a
,
I
n
d
o
n
e
si
a
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Oct
2
,
2
0
2
5
R
ev
i
s
ed
Feb
5
,
2
0
2
6
A
cc
ep
ted
Ma
r
2
9
,
2
0
2
6
P
re
m
a
tu
re
c
a
rd
iac
c
o
n
trac
ti
o
n
s,
in
c
lu
d
in
g
p
re
m
a
tu
re
a
tri
a
l
c
o
n
trac
ti
o
n
s
(P
A
Cs)
a
n
d
p
re
m
a
tu
re
v
e
n
tri
c
u
lar
c
o
n
trac
ti
o
n
s
(
P
V
Cs)
,
a
re
c
o
m
m
o
n
a
rrh
y
th
m
ias
th
a
t
m
a
y
in
c
re
a
se
th
e
risk
o
f
c
a
rd
io
v
a
sc
u
lar
c
o
m
p
li
c
a
ti
o
n
s
w
h
e
n
th
e
y
o
c
c
u
r
f
re
q
u
e
n
tl
y
.
Ac
c
u
ra
te
c
las
si
f
ica
ti
o
n
o
f
th
e
se
e
v
e
n
ts
f
ro
m
e
lec
tro
c
a
rd
io
g
ra
m
(EC
G
)
sig
n
a
ls r
e
m
a
in
s
c
h
a
ll
e
n
g
in
g
d
u
e
to
n
o
ise
a
n
d
sig
n
a
l
v
a
riab
il
it
y
.
T
h
is
stu
d
y
p
ro
p
o
se
s
a
m
a
c
h
in
e
lea
rn
in
g
–
b
a
se
d
c
las
sif
ic
a
ti
o
n
f
ra
m
e
w
o
rk
th
a
t
c
o
m
b
in
e
s
re
c
u
rsiv
e
fe
a
tu
re
e
li
m
in
a
ti
o
n
w
it
h
c
ro
ss
-
v
a
li
d
a
t
io
n
f
o
r
f
e
a
tu
re
se
lec
ti
o
n
a
n
d
a
n
e
n
se
m
b
le
lea
rn
in
g
s
trate
g
y
to
im
p
ro
v
e
c
las
si
f
ica
ti
o
n
ro
b
u
st
n
e
ss
.
T
h
e
a
p
p
r
o
a
c
h
w
a
s
e
v
a
lu
a
ted
u
sin
g
th
e
M
a
ss
a
c
h
u
se
tt
s
In
stit
u
te
o
f
T
e
c
h
n
o
l
o
g
y
–
Be
th
Isra
e
l
Ho
s
p
it
a
l
(
M
IT
-
BIH
)
A
rrh
y
th
m
ia
d
a
tab
a
se
a
n
d
a
c
h
iev
e
d
h
ig
h
c
las
sif
ica
ti
o
n
p
e
rf
o
rm
a
n
c
e
,
w
it
h
a
n
a
c
c
u
ra
c
y
o
f
9
5
.
3
4
%
,
F1
-
sc
o
re
o
f
9
2
.
1
1
%
,
a
n
d
b
a
lan
c
e
d
p
re
c
isio
n
a
n
d
re
c
a
ll
f
o
r
P
V
C
a
n
d
P
A
C
.
In
a
d
d
it
i
o
n
,
SH
a
p
le
y
A
d
d
it
iv
e
e
x
P
lan
a
ti
o
n
s (S
HA
P
)
we
re
e
m
p
lo
y
e
d
to
id
e
n
ti
f
y
th
e
m
o
s
t
in
f
lu
e
n
ti
a
l
f
e
a
tu
re
s,
e
n
h
a
n
c
in
g
m
o
d
e
l
in
terp
re
tab
il
it
y
.
T
h
e
r
e
su
lt
s
d
e
m
o
n
stra
te
th
a
t
th
e
p
ro
p
o
se
d
fra
m
e
w
o
rk
p
ro
v
id
e
s
a
re
li
a
b
le
a
n
d
in
terp
re
t
a
b
le
so
lu
t
io
n
f
o
r
d
isti
n
g
u
ish
i
n
g
p
re
m
a
tu
re
c
a
rd
iac
c
o
n
trac
ti
o
n
s,
h
ig
h
li
g
h
ti
n
g
it
s p
o
ten
t
ial
a
p
p
li
c
a
ti
o
n
i
n
c
li
n
ica
l
d
e
c
i
sio
n
su
p
p
o
rt
sy
ste
m
s.
K
ey
w
o
r
d
s
:
E
n
s
e
m
b
le
lear
n
i
n
g
M
ac
h
i
n
e
lear
n
i
n
g
P
r
em
at
u
r
e
atr
ial
co
n
tr
ac
tio
n
s
P
r
em
at
u
r
e
ca
r
d
iac
co
n
tr
ac
tio
n
s
P
r
em
at
u
r
e
v
e
n
tr
icu
lar
co
n
tr
ac
tio
n
s
r
ec
u
r
s
iv
e
f
ea
t
u
r
e
eli
m
in
at
io
n
w
it
h
cr
o
s
s
-
v
alid
atio
n
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
E
ls
a
Sar
i H
a
y
u
n
ah
N
u
r
d
in
i
y
a
h
Dep
ar
t
m
en
t o
f
E
lectr
ical
E
n
g
i
n
ee
r
in
g
,
Fac
u
lt
y
o
f
E
n
g
i
n
ee
r
in
g
,
Un
i
v
er
s
i
t
as
J
en
d
er
al
So
ed
ir
m
an
R
a
y
a
Ma
y
j
e
n
S
u
n
g
k
o
n
o
Stre
et
No
.
KM
5
,
Du
s
u
n
2
,
B
later
,
Pu
r
b
alin
g
g
a,
J
a
w
a
T
en
g
ah
,
I
n
d
o
n
esia
53371
E
m
ail:
el
s
a.
n
u
r
d
in
i
y
ah
@
u
n
s
o
ed
.
ac
.
id
1.
I
NT
RO
D
UCT
I
O
N
P
r
em
at
u
r
e
co
n
tr
ac
tio
n
s
ar
e
ab
n
o
r
m
al
h
ea
r
tb
ea
ts
th
a
t
o
r
ig
i
n
a
te
o
u
ts
id
e
th
e
s
i
n
o
atr
ial
n
o
d
e.
Gen
er
all
y
,
th
ese
i
m
p
u
ls
es
o
r
ig
in
ate
i
n
t
h
e
atr
ia
o
r
v
en
tr
icles.
I
f
t
h
e
i
m
p
u
ls
e
o
r
ig
i
n
ates
i
n
t
h
e
atr
ial
r
eg
io
n
,
it
i
s
ca
lled
a
p
r
em
at
u
r
e
atr
ial
co
n
tr
ac
tio
n
(
P
A
C
)
,
an
d
i
f
t
h
e
i
m
p
u
ls
e
o
r
i
g
in
a
tes
i
n
t
h
e
v
e
n
tr
ic
u
lar
r
eg
io
n
,
it
is
c
alled
a
p
r
em
at
u
r
e
v
en
tr
ic
u
lar
co
n
tr
ac
ti
o
n
(
P
VC
)
.
P
A
C
s
ar
e
g
e
n
er
all
y
co
n
s
id
er
ed
b
en
ig
n
a
n
d
o
f
te
n
r
e
q
u
ir
e
n
o
tr
ea
t
m
en
t,
o
cc
u
r
r
in
g
ac
r
o
s
s
all
ag
e
g
r
o
u
p
s
w
it
h
i
n
cr
ea
s
i
n
g
p
r
ev
ale
n
ce
in
o
ld
er
p
o
p
u
latio
n
s
[
1
]
.
Ho
w
e
v
er
,
ex
ce
s
s
i
v
e
P
AC
s
h
av
e
b
ee
n
ass
o
ciate
d
w
ith
p
ath
o
lo
g
ical
ch
a
n
g
es
in
th
e
le
f
t
atr
iu
m
an
d
in
cr
ea
s
ed
r
is
k
s
o
f
atr
ial
f
ib
r
illatio
n
,
s
tr
o
k
e,
m
o
r
talit
y
,
atr
ial
ca
r
d
io
m
y
o
p
ath
y
,
an
d
k
id
n
e
y
d
is
o
r
d
er
s
,
in
d
icatin
g
b
r
o
ad
er
s
y
s
te
m
ic
co
m
p
licatio
n
s
in
v
o
l
v
e
m
en
t
ev
e
n
i
n
p
atie
n
ts
w
i
th
o
u
t
a
p
r
io
r
h
is
to
r
y
o
f
a
tr
ial
f
ib
r
illatio
n
[
2
]
–
[
5
]
.
Sim
i
lar
l
y
,
P
VC
i
s
als
o
co
n
s
id
er
ed
a
g
e
n
er
all
y
h
ar
m
le
s
s
ca
r
d
iac
ab
n
o
r
m
alit
y
[
6
]
.
I
n
ce
r
tain
co
n
d
itio
n
s
,
f
r
eq
u
e
n
t
P
VC
s
ar
e
ass
o
ciate
d
w
it
h
a
n
i
n
cr
ea
s
ed
r
is
k
o
f
atr
ial
f
ib
r
illatio
n
a
n
d
is
c
h
ae
m
ic
s
tr
o
k
e
a
n
d
m
a
y
co
n
tr
ib
u
te
t
o
ar
r
h
y
th
m
ia
s
a
n
d
ca
r
d
io
m
y
o
p
at
h
y
[
7
]
–
[
9
]
.
No
ta
b
ly
,
P
VC
s
ar
e
o
n
e
o
f
th
e
m
o
s
t
co
m
m
o
n
f
o
r
m
s
o
f
ab
n
o
r
m
al
h
ea
r
tb
ea
t
an
d
h
av
e
s
er
io
u
s
co
n
s
eq
u
e
n
ce
s
,
esp
ec
ia
ll
y
i
n
th
e
eld
er
l
y
p
o
p
u
latio
n
[
1
0
]
.
Gen
er
a
ll
y
,
h
ea
r
t
r
h
y
t
h
m
ab
n
o
r
m
al
ities
,
s
u
c
h
as
p
r
e
m
at
u
r
e
co
n
tr
ac
tio
n
s
,
ar
e
d
etec
ted
u
s
i
n
g
a
n
elec
tr
o
ca
r
d
io
g
r
am
(
E
C
G)
.
Un
f
o
r
tu
n
atel
y
,
m
an
u
al
i
n
ter
p
r
etati
o
n
o
f
E
C
G
s
i
g
n
als
r
eq
u
ir
e
s
h
ig
h
ex
p
er
ti
s
e,
is
ti
m
e
-
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
3
,
J
u
n
e
20
26
:
8
9
1
-
9
03
892
co
n
s
u
m
i
n
g
,
an
d
is
p
r
o
n
e
to
er
r
o
r
s
d
u
e
to
n
o
is
e
an
d
s
u
b
j
ec
ti
v
it
y
.
R
ec
e
n
t
ad
v
an
ce
s
i
n
m
ac
h
in
e
lear
n
i
n
g
h
a
v
e
en
ab
led
au
to
m
ated
ar
r
h
y
t
h
m
ia
class
if
icatio
n
s
h
o
w
i
n
g
p
r
o
m
is
in
g
r
es
u
lt
s
.
Var
io
u
s
m
ac
h
i
n
e
l
ea
r
n
in
g
alg
o
r
it
h
m
s
h
av
e
b
ee
n
w
id
el
y
u
s
ed
in
h
ea
r
tb
ea
t
class
if
ica
tio
n
,
in
cl
u
d
i
n
g
K
-
n
ea
r
es
t
n
ei
g
h
b
o
r
s
(
KNN)
[
1
1
]
–
[
1
4
]
,
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
i
n
e
(
SVM
)
[
1
5
]
,
[
1
6
]
,
l
ig
h
t
g
r
ad
ien
t
b
o
o
s
tin
g
m
ac
h
in
e
(
L
ig
h
tG
B
M
)
[
1
7
]
,
[
1
8
]
.
T
h
e
u
s
e
o
f
t
h
ese
alg
o
r
ith
m
s
s
h
o
w
s
f
a
ir
l
y
g
o
o
d
p
er
f
o
r
m
an
ce
,
b
u
t
an
ap
p
r
o
ac
h
th
at
co
m
b
in
e
s
s
ev
er
al
m
et
h
o
d
s
k
n
o
w
n
a
s
en
s
e
m
b
le
lear
n
in
g
(
E
L
)
o
f
te
n
p
r
o
v
id
es
m
o
r
e
o
p
ti
m
al
p
er
f
o
r
m
an
ce
[
1
9
]
–
[
2
2
]
.
T
h
is
is
d
u
e
to
t
h
e
ab
ilit
y
o
f
E
L
to
co
m
b
in
e
th
e
s
tr
e
n
g
t
h
s
o
f
ea
ch
alg
o
r
it
h
m
,
r
esu
l
tin
g
i
n
a
m
o
r
e
co
m
p
r
e
h
en
s
i
v
e
an
d
r
o
b
u
s
t
class
if
ica
ti
o
n
m
o
d
el
co
m
p
ar
ed
to
u
s
i
n
g
a
s
in
g
le
m
o
d
el.
R
esear
ch
o
n
p
r
e
m
at
u
r
e
ca
r
d
iac
co
n
tr
ac
tio
n
cla
s
s
i
f
icatio
n
r
e
m
ai
n
s
li
m
i
ted
,
w
ith
m
o
s
t
ex
i
s
t
in
g
s
t
u
d
ies
f
o
cu
s
in
g
p
r
i
m
ar
il
y
o
n
P
VC
.
Stu
d
ies
t
h
at
ad
d
r
ess
b
o
th
P
AC
a
n
d
P
VC
o
f
te
n
e
x
h
ib
it
s
ig
n
i
f
ican
t
p
er
f
o
r
m
a
n
ce
i
m
b
alan
ce
b
et
w
ee
n
t
h
e
t
w
o
cl
ass
es
(
P
A
C
a
n
d
P
VC
)
[
2
3
]
–
[
2
7
]
.
Fu
r
th
er
m
o
r
e,
m
an
y
s
tu
d
ie
s
h
av
e
n
o
t
ap
p
lied
an
in
ter
-
p
atie
n
t
s
ce
n
ar
io
,
w
h
ich
i
n
v
o
l
v
es
test
in
g
th
e
m
o
d
el
o
n
d
ata
f
r
o
m
p
atien
t
s
n
o
t
u
s
ed
in
th
e
tr
ain
in
g
s
tag
e
[
2
2
]
,
[
2
8
]
;
th
u
s
,
th
e
m
o
d
el
’
s
ab
ilit
y
to
g
en
er
alize
to
n
e
w
p
atien
t
s
r
e
m
ain
s
w
ea
k
.
I
n
a
cli
n
ical
co
n
te
x
t,
in
ter
-
p
atien
t
g
e
n
er
alis
atio
n
is
cr
u
ci
al
d
u
e
to
p
h
y
s
io
lo
g
ical
v
ar
iati
o
n
s
b
et
w
ee
n
in
d
i
v
id
u
al
s
.
T
h
i
s
co
n
d
itio
n
in
d
icate
s
th
at,
d
esp
ite
n
u
m
er
o
u
s
f
ea
t
u
r
e
s
b
ein
g
co
n
s
id
er
ed
,
th
er
e
is
n
o
r
o
b
u
s
t
f
ea
t
u
r
e
s
elec
tio
n
s
tep
t
o
d
eter
m
i
n
e
w
h
ich
f
ea
t
u
r
es
ar
e
tr
u
l
y
i
n
f
o
r
m
ati
v
e
i
n
d
is
ti
n
g
u
is
h
in
g
b
et
w
ee
n
P
A
C
,
P
VC
,
an
d
n
o
r
m
a
l
clas
s
es,
a
s
w
ell
as
to
m
ai
n
tai
n
p
er
f
o
r
m
a
n
ce
w
h
en
t
h
e
m
o
d
el
i
s
ap
p
lied
to
u
n
s
ee
n
d
ata.
T
h
er
ef
o
r
e,
th
is
s
t
u
d
y
u
s
es
r
ec
u
r
s
i
v
e
f
ea
t
u
r
e
eli
m
in
a
tio
n
w
it
h
cr
o
s
s
-
v
alid
atio
n
(
R
FE
C
V
)
as
a
f
ea
t
u
r
e
s
elec
tio
n
m
et
h
o
d
in
th
e
p
r
e
m
atu
r
e
ca
r
d
iac
co
n
tr
ac
to
n
cla
s
s
if
icatio
n
m
o
d
el.
R
FE
C
V
eli
m
in
ates
th
e
f
ea
t
u
r
es
g
r
ad
u
all
y
d
u
r
i
n
g
cr
o
s
s
-
v
alid
at
io
n
at
ea
ch
s
tep
,
allo
w
i
n
g
f
o
r
th
e
id
en
ti
f
icatio
n
o
f
th
e
m
o
s
t
in
f
o
r
m
ati
v
e
f
ea
t
u
r
e
s
u
b
s
et
w
h
ile
a
s
s
e
s
s
i
n
g
m
o
d
el
p
er
f
o
r
m
a
n
ce
b
et
w
ee
n
f
o
ld
s
.
S
ev
er
al
s
t
u
d
ies h
av
e
d
e
m
o
n
s
tr
ated
t
h
e
ef
f
ec
ti
v
en
e
s
s
o
f
f
ea
t
u
r
e
s
elec
tio
n
m
et
h
o
d
s
in
d
is
ea
s
e
cla
s
s
i
f
icatio
n
,
p
ar
ticu
lar
l
y
t
h
r
o
u
g
h
t
h
e
ap
p
licatio
n
o
f
R
FECV.
Fo
r
ex
a
m
p
le,
in
th
is
r
esear
c
h
R
FE
C
V
s
u
cc
e
s
s
f
u
ll
y
s
elec
ted
2
5
o
u
t
o
f
ap
p
r
o
x
i
m
ate
l
y
1
2
8
E
C
G
-
p
h
o
n
o
ca
r
d
io
g
r
a
m
(
P
C
G
)
f
ea
tu
r
es,
th
er
eb
y
i
m
p
r
o
v
in
g
m
o
d
el
ac
cu
r
ac
y
[
2
9
]
.
An
o
th
er
s
t
u
d
y
also
d
e
m
o
n
s
tr
ated
th
at
th
e
u
s
e
o
f
R
FEC
V
o
n
h
ea
r
t
d
is
ea
s
e
d
ata
s
ets
ca
n
o
p
ti
m
ize
R
an
d
o
m
Fo
r
est
p
er
f
o
r
m
an
ce
t
h
r
o
u
g
h
m
o
r
e
r
elev
an
t
f
ea
t
u
r
e
s
elec
tio
n
[
3
0
]
.
Si
m
ilar
f
in
d
i
n
g
s
w
er
e
r
ep
o
r
ted
in
t
h
i
s
s
t
u
d
y
w
h
er
e
th
is
m
et
h
o
d
p
r
o
d
u
ce
d
h
ig
h
F1
-
s
co
r
e
a
n
d
r
ec
all
v
alu
e
s
af
ter
th
e
f
ea
tu
r
e
s
elec
tio
n
p
r
o
ce
s
s
[
3
1
]
.
B
ased
o
n
th
i
s
ev
id
e
n
ce
,
th
e
ap
p
licati
o
n
o
f
R
FE
C
V
i
n
th
e
class
i
f
icatio
n
o
f
p
r
e
m
at
u
r
e
h
e
ar
t
co
n
tr
ac
tio
n
s
is
e
x
p
ec
ted
to
en
h
a
n
ce
P
AC
clas
s
if
icatio
n
p
er
f
o
r
m
an
ce
,
r
ed
u
ce
o
v
er
f
itti
n
g
,
an
d
i
m
p
r
o
v
e
i
n
ter
-
p
atien
t g
e
n
er
aliza
ti
o
n
.
T
h
is
s
tu
d
y
p
r
o
p
o
s
es
a
class
if
i
ca
tio
n
f
r
a
m
e
w
o
r
k
f
o
r
p
r
em
at
u
r
e
h
ea
r
t
co
n
tr
ac
tio
n
s
in
to
n
o
r
m
al,
P
A
C
,
an
d
P
VC
clas
s
es
b
y
ap
p
l
y
in
g
R
FECV
f
o
r
f
ea
tu
r
e
s
e
lecti
o
n
an
d
i
n
teg
r
ati
n
g
KNN
w
it
h
L
ig
h
tGB
M
i
n
a
n
en
s
e
m
b
le
m
o
d
el.
T
h
is
co
m
b
in
atio
n
ai
m
s
to
p
r
o
d
u
ce
a
m
o
d
el
t
h
at
is
b
o
th
h
ig
h
l
y
ac
cu
r
ate
a
n
d
co
m
p
u
tatio
n
a
ll
y
ef
f
icien
t,
w
h
ile
al
s
o
b
ein
g
ea
s
y
to
u
n
d
er
s
ta
n
d
.
T
h
e
s
tu
d
y
u
tili
ze
s
s
h
ap
le
y
ad
d
itiv
e
e
x
p
l
an
atio
n
s
(
SH
A
P
)
to
q
u
an
ti
f
y
t
h
e
co
n
tr
ib
u
t
io
n
o
f
ea
ch
f
ea
tu
r
e
to
t
h
e
d
ec
is
io
n
.
T
h
is
m
ak
e
s
t
h
e
r
es
u
lts
m
o
r
e
tr
an
s
p
ar
en
t
an
d
cli
n
icall
y
m
ea
n
in
g
f
u
l.
T
h
is
ap
p
r
o
ac
h
s
h
o
u
ld
i
m
p
r
o
v
e
th
e
clas
s
i
f
icatio
n
o
f
m
in
o
r
it
y
cla
s
s
e
s
,
s
u
ch
as
P
A
C
a
n
d
P
VC
,
a
n
d
en
h
a
n
ce
t
h
e
m
o
d
el
’
s
ab
ilit
y
to
g
en
er
alize
ac
r
o
s
s
p
atie
n
ts
.
2.
M
E
T
H
O
D
Fig
u
r
e
1
s
h
o
w
s
th
e
s
tag
e
s
o
f
class
i
f
y
i
n
g
p
r
e
m
atu
r
e
co
n
tr
a
ctio
n
s
u
s
ed
in
th
is
s
t
u
d
y
.
T
h
ese
s
tag
e
s
in
cl
u
d
e
p
r
ep
r
o
ce
s
s
in
g
,
f
ea
t
u
r
e
ex
tr
ac
tio
n
,
s
elec
tio
n
o
f
clas
s
i
f
icatio
n
f
ea
t
u
r
es,
an
d
ev
al
u
ati
o
n
o
f
class
i
f
icatio
n
r
esu
lt
s
u
s
i
n
g
p
ar
a
m
eter
s
r
ec
al
l,
p
r
ec
is
io
n
,
an
d
F1
-
s
co
r
e.
A
ll
o
f
t
h
ese
s
ta
g
es
w
er
e
s
y
s
te
m
a
ticall
y
d
esi
g
n
ed
to
en
s
u
r
e
th
e
ac
c
u
r
ac
y
o
f
E
C
G
s
i
g
n
al
d
etec
tio
n
in
id
e
n
ti
f
y
in
g
t
y
p
e
s
o
f
p
r
e
m
at
u
r
e
co
n
tr
ac
tio
n
s
.
2
.
1
.
Da
t
a
T
h
e
d
ata
u
s
ed
in
th
is
s
t
u
d
y
ar
e
o
n
e
-
d
i
m
e
n
s
io
n
al
E
C
G
s
i
g
n
al
d
ata
o
b
tain
ed
f
r
o
m
th
e
MI
T
-
B
I
H
A
r
r
h
y
t
h
m
ia
d
atab
ase
[
3
2
]
.
T
h
e
E
C
G
s
ig
n
al
s
w
er
e
d
ig
itis
ed
at
a
s
am
p
li
n
g
f
r
eq
u
e
n
c
y
o
f
3
6
0
Hz
w
it
h
1
1
-
b
it
r
eso
lu
tio
n
i
n
t
h
e
r
an
g
e
o
f
1
0
m
V.
T
h
er
e
w
a
s
a
to
tal
o
f
4
8
E
C
G
r
ec
o
r
d
in
g
s
,
ea
ch
co
n
s
is
ti
n
g
o
f
t
w
o
c
h
an
n
el
s
:
th
e
u
p
p
er
ML
I
I
ch
a
n
n
e
l
an
d
t
h
e
lo
w
er
V1
,
V4
,
V5
,
an
d
V6
ch
a
n
n
els,
w
h
ich
v
ar
ied
f
o
r
ea
c
h
r
ec
o
r
d
in
g
.
I
n
th
i
s
s
t
u
d
y
,
d
ata
f
r
o
m
o
n
l
y
o
n
e
ch
a
n
n
el,
ML
I
I
,
w
er
e
u
s
ed
b
ec
a
u
s
e
t
h
e
elec
tr
ical
ac
tiv
i
t
y
r
ec
o
r
d
ed
b
y
ML
I
I
co
r
r
esp
o
n
d
s
to
th
e
d
ir
ec
tio
n
o
f
ca
r
d
iac
i
m
p
u
ls
e
co
n
d
u
ctio
n
,
a
n
d
th
e
Q
R
S
co
m
p
l
ex
s
h
ap
e
is
b
es
t
v
i
s
u
al
ized
in
t
h
e
ML
I
I
ch
a
n
n
e
l
[
3
3
]
.
I
n
ad
d
itio
n
,
w
ea
r
ab
l
e
s
tu
d
ies
d
e
m
o
n
s
tr
ate
th
at
P
AC
/
P
VC
d
etec
tio
n
ca
n
b
e
p
er
f
o
r
m
ed
in
a
s
in
g
le
-
lead
f
o
r
m
at
(
e.
g
.
,
a
1
5
-
s
ec
o
n
d
w
ea
r
ab
le
E
C
G)
an
d
th
at
a
s
in
g
le
l
ea
d
p
r
o
v
id
es
s
u
f
f
icie
n
t
s
i
g
n
a
l
to
d
etec
t
p
r
em
at
u
r
e
b
ea
ts
[
3
4
]
.
I
n
th
is
s
tu
d
y
,
n
o
t
all
E
C
G
r
ec
o
r
d
in
g
s
w
er
e
u
s
ed
,
o
n
l
y
r
ec
o
r
d
in
g
s
n
u
m
b
er
ed
1
0
0
,
1
0
1
,
1
0
3
,
1
0
5
,
1
0
6
,
1
0
8
,
1
1
2
,
1
1
6
,
1
1
8
,
1
1
9
,
1
2
4
,
2
0
0
,
2
0
1
,
2
0
2
,
2
0
3
,
2
0
5
,
2
0
7
,
2
0
8
,
2
0
9
,
2
1
3
,
2
1
4
,
2
1
5
,
2
1
9
,
2
2
0
,
2
2
1
,
2
2
2
,
2
2
3
,
228,
an
d
2
3
2
w
er
e
u
s
ed
.
T
h
e
d
eter
m
in
at
io
n
o
f
t
h
ese
r
ec
o
r
d
in
g
s
w
as
b
ased
o
n
s
ev
er
al
cr
it
er
ia,
in
clu
d
i
n
g
th
a
t
r
ec
o
r
d
in
g
u
s
ed
f
o
r
tr
ain
in
g
d
ata
w
o
u
ld
n
o
t b
e
u
s
ed
f
o
r
test
d
a
ta
an
d
th
at
r
ec
o
r
d
in
g
s
ca
m
e
f
r
o
m
t
h
e
u
p
p
er
ML
I
I
ch
an
n
el
(
i
n
ter
-
p
atien
t)
.
T
h
e
s
e
lectio
n
a
n
d
d
iv
i
s
io
n
o
f
r
ec
o
r
d
in
g
s
u
s
ed
f
o
r
tr
ain
i
n
g
d
ata
a
n
d
test
d
ata
f
o
llo
w
ed
th
e
s
a
m
e
s
elec
t
io
n
an
d
d
iv
is
io
n
p
r
o
ce
s
s
as th
o
s
e
ca
r
r
ied
o
u
t
b
y
Sh
i
et
a
l.
[
3
3
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
C
la
s
s
i
fica
tio
n
o
f p
r
ema
tu
r
e
ca
r
d
ia
c
co
n
tr
a
ctio
n
s
b
a
s
ed
o
n
R
F
E
C
V
…
(
E
ls
a
S
a
r
i H
a
yu
n
a
h
N
u
r
d
in
iya
h
)
893
Fig
u
r
e
1
.
P
r
o
p
o
s
ed
m
eth
o
d
2
.
2
.
P
re
pro
ce
s
s
ing
T
h
e
E
C
G
s
ig
n
a
ls
o
b
t
a
in
e
d
f
r
o
m
th
e
d
a
t
a
b
as
e
w
e
r
e
c
l
e
an
e
d
u
s
i
n
g
a
m
e
d
i
an
f
il
t
e
r
w
ith
a
w
i
d
th
o
f
2
0
0
m
s
an
d
6
0
0
m
s
to
eli
m
in
ate
b
aseli
n
e
w
an
d
er
[
2
6
]
.
Me
an
w
h
ile,
h
i
g
h
-
f
r
eq
u
e
n
c
y
p
o
w
er
li
n
e
n
o
i
s
e
w
a
s
r
e
m
o
v
ed
u
s
i
n
g
a
B
u
tter
w
o
r
t
h
lo
w
-
p
as
s
f
ilter
(
L
P
F)
w
it
h
a
c
u
t
-
o
f
f
f
r
eq
u
e
n
c
y
o
f
4
0
Hz
[
2
6
]
.
Nex
t,
P
QR
ST
s
ig
n
a
l
s
e
g
m
en
tatio
n
w
a
s
p
er
f
o
r
m
ed
b
y
tak
i
n
g
1
3
0
s
a
m
p
les b
e
f
o
r
e
an
d
1
7
0
s
am
p
l
es a
f
ter
t
h
e
R
p
o
s
itio
n
[
3
5
]
.
2
.
3
.
F
e
a
t
ure
ex
t
ra
ct
io
ns
I
n
th
is
s
tu
d
y
,
v
a
r
i
o
u
s
ty
p
es
o
f
f
e
at
u
r
es
f
r
o
m
s
ev
e
r
al
d
o
m
ai
n
s
w
e
r
e
u
s
e
d
,
in
c
lu
d
in
g
t
em
p
o
r
al
,
s
p
e
c
t
r
al
,
s
t
a
ti
s
t
i
c
al
,
a
n
d
R
-
R
p
e
ak
in
t
e
r
v
a
l
m
o
r
p
h
o
l
o
g
i
c
a
l
f
e
atu
r
e
s
.
T
em
p
o
r
a
l
,
s
p
e
c
t
r
a
l
,
an
d
s
ta
t
is
t
ic
a
l
f
ea
t
u
r
e
ex
t
r
ac
t
i
o
n
w
as
p
e
r
f
o
r
m
e
d
u
s
in
g
th
e
t
im
e
s
e
r
i
es
f
e
a
t
u
r
e
ex
t
r
a
ct
i
o
n
li
b
r
a
r
y
(
T
S
F
E
L
)
av
a
il
a
b
l
e
in
P
y
th
o
n
,
w
h
ich
p
r
o
v
i
d
es
a
v
a
r
i
e
ty
o
f
f
e
atu
r
e
s
f
o
r
an
a
ly
s
in
g
t
im
e
s
i
g
n
a
ls
[
3
6
]
,
[
3
7
]
.
Me
an
w
h
i
le
,
f
o
r
t
h
e
R
-
R
in
t
e
r
v
a
l m
o
r
p
h
o
l
o
g
y
d
o
m
a
in
,
th
is
s
tu
d
y
em
p
l
o
y
e
d
f
o
u
r
m
ain
f
e
at
u
r
e
s
:
R
Po
s
it
i
o
n
,
Pr
e
_
R
,
Po
s
t
_
R
,
an
d
L
o
c
a
l
_
R
[
2
6
]
,
[
3
8
]
.
Pr
e
_
R
i
s
d
e
f
in
e
d
a
s
th
e
in
t
e
r
v
a
l
b
e
tw
e
en
th
e
o
b
s
e
r
v
e
d
R
an
d
t
h
e
p
r
e
v
i
o
u
s
R
.
M
e
an
w
h
il
e
,
P
o
s
t_
R
i
s
th
e
in
t
e
r
v
al
b
e
tw
e
en
th
e
o
b
s
e
r
v
e
d
R
a
n
d
th
e
s
u
b
s
e
q
u
en
t
R
,
an
d
L
o
c
a
l
_
R
is
d
e
f
in
e
d
as
th
e
av
e
r
ag
e
o
f
1
0
R
-
R
in
t
e
r
v
al
s
a
r
o
u
n
d
th
e
o
b
s
e
r
v
ed
R
.
2
.
4
.
F
e
a
t
ure
s
ele
ct
io
ns
F
e
a
tu
r
e
s
el
e
c
ti
o
n
is
p
e
r
f
o
r
m
e
d
t
o
r
e
d
u
c
e
c
o
m
p
u
t
a
ti
o
n
a
l
l
o
a
d
.
D
u
r
in
g
t
h
i
s
s
t
ag
e
,
f
e
a
tu
r
e
s
c
al
i
n
g
t
ak
es
p
l
a
c
e
,
f
o
l
l
o
w
e
d
b
y
t
h
e
r
em
o
v
al
o
f
f
ea
tu
r
e
s
w
it
h
l
o
w
v
a
r
ia
n
ce
.
T
h
e
p
r
o
c
e
s
s
c
o
n
t
in
u
es
w
ith
Pe
a
r
s
o
n
c
o
r
r
e
la
t
i
o
n
a
n
a
ly
s
is
a
n
d
c
o
n
c
lu
d
e
s
w
ith
f
e
at
u
r
e
r
an
k
in
g
u
s
in
g
R
F
E
C
V
.
T
h
e
f
e
a
tu
r
es
o
b
t
a
in
e
d
a
r
e
f
i
r
s
t
s
c
ale
d
t
o
n
o
r
m
a
li
s
e
th
e
v
a
lu
es
,
e
n
s
u
r
in
g
t
h
ey
f
a
ll
w
i
th
in
th
e
s
am
e
r
an
g
e
.
T
h
i
s
i
s
d
o
n
e
u
s
in
g
th
e
s
t
an
d
a
r
d
z
-
s
c
o
r
e
,
w
h
ic
h
r
e
l
i
es
o
n
th
e
m
e
an
(
)
a
n
d
s
t
an
d
a
r
d
d
e
v
i
at
i
o
n
(
)
o
f
t
h
e
s
am
p
le
.
A
f
t
e
r
s
c
al
in
g
,
f
ea
t
u
r
es
w
i
th
l
o
w
v
a
r
i
an
ce
a
r
e
r
e
m
o
v
e
d
u
s
in
g
t
h
e
Py
th
o
n
l
i
b
r
a
r
y
V
a
r
ia
n
c
e
T
h
r
e
s
h
o
l
d
(
)
,
e
l
im
in
at
in
g
t
h
o
s
e
b
e
l
o
w
a
s
et
th
r
e
s
h
o
l
d
.
T
h
e
r
em
a
in
i
n
g
f
e
a
tu
r
es
a
r
e
t
h
en
c
h
e
ck
e
d
f
o
r
c
o
r
r
e
l
a
t
i
o
n
;
in
P
y
th
o
n
,
t
h
e
Pe
a
r
s
o
n
c
o
r
r
e
l
a
ti
o
n
c
al
c
u
l
a
t
es
th
e
c
o
r
r
e
l
a
ti
o
n
v
a
lu
es
.
A
n
y
f
e
a
tu
r
es
w
i
th
a
c
o
r
r
e
l
a
t
i
o
n
a
b
o
v
e
0
.
7
a
r
e
r
em
o
v
e
d
,
a
s
0
.
7
a
l
r
e
a
d
y
i
n
d
i
c
a
t
es
a
v
e
r
y
h
ig
h
c
o
r
r
e
l
a
ti
o
n
b
e
t
w
e
en
tw
o
f
e
a
tu
r
es
[
3
9
]
.
H
ig
h
ly
c
o
r
r
e
la
t
e
d
f
ea
tu
r
e
s
c
an
n
o
t
a
c
cu
r
a
t
ely
r
e
p
r
es
en
t
th
e
d
i
s
t
i
n
c
t
c
h
a
r
a
c
t
e
r
i
s
t
i
cs
o
f
th
e
t
h
r
ee
c
l
a
s
s
e
s
,
p
o
t
en
t
ia
l
ly
c
o
m
p
r
o
m
i
s
in
g
c
l
as
s
if
i
c
at
i
o
n
a
cc
u
r
a
cy
.
F
in
a
lly
,
R
F
E
C
V
i
s
em
p
l
o
y
e
d
t
o
r
a
n
k
an
d
s
e
le
c
t
th
e
m
o
s
t
p
r
e
d
i
c
t
iv
e
f
e
a
tu
r
es
u
s
in
g
r
an
d
o
m
f
o
r
e
s
t
a
s
an
el
im
i
n
at
o
r
w
i
th
5
-
f
o
l
d
c
r
o
s
s
-
v
a
l
i
d
ati
o
n
.
St
a
r
t
in
g
f
r
o
m
a
ll
r
em
a
in
in
g
f
e
a
tu
r
es
,
R
F
E
C
V
i
t
e
r
a
ti
v
e
ly
r
em
o
v
es
th
e
le
a
s
t
im
p
o
r
t
a
n
t
f
e
at
u
r
e
b
a
s
e
d
o
n
im
p
o
r
ta
n
c
e
s
c
o
r
e
s
o
b
t
ain
e
d
f
r
o
m
th
e
r
a
n
d
o
m
f
o
r
es
t
m
o
d
e
l
t
r
a
in
e
d
o
n
th
e
f
u
l
l
t
r
a
in
in
g
s
e
t
.
E
a
c
h
r
e
s
u
lt
in
g
f
e
at
u
r
e
s
u
b
s
e
t
is
e
v
a
lu
a
t
e
d
u
s
in
g
5
-
f
o
l
d
c
r
o
s
s
-
v
a
l
i
d
a
t
i
o
n
,
w
h
e
r
e
t
h
e
s
am
e
s
u
b
s
e
t
i
s
t
e
s
t
e
d
a
c
r
o
s
s
al
l
f
o
l
d
s
an
d
th
e
v
a
li
d
a
t
i
o
n
p
e
r
f
o
r
m
an
c
e
i
s
a
v
e
r
a
g
e
d
.
T
h
e
f
e
a
tu
r
e
s
u
b
s
e
t
a
ch
i
ev
in
g
th
e
h
ig
h
es
t
a
v
e
r
a
g
e
c
r
o
s
s
-
v
a
li
d
a
t
i
o
n
p
e
r
f
o
r
m
an
c
e
is
s
e
l
e
ct
e
d
as
o
p
t
im
a
l
[
4
0
]
–
[
42]
.
B
y
s
e
l
e
ct
in
g
f
e
a
tu
r
es
b
as
e
d
o
n
a
v
e
r
ag
e
d
p
e
r
f
o
r
m
an
ce
a
c
r
o
s
s
f
o
l
d
s
r
a
th
e
r
t
h
a
n
a
s
in
g
l
e
d
a
t
a
s
p
l
i
t
,
R
F
E
C
V
im
p
l
ic
i
tly
p
r
o
m
o
te
s
f
e
a
tu
r
e
s
ta
b
i
l
ity
,
im
p
r
o
v
e
s
g
en
e
r
a
l
i
za
b
i
l
ity
,
a
n
d
r
e
d
u
c
es
th
e
r
is
k
o
f
o
v
e
r
f
it
t
in
g
.
2
.
5
.
Cla
s
s
if
ica
t
io
n
I
n
th
e
c
lass
if
ica
ti
o
n
p
r
o
ce
s
s
,
t
h
e
i
m
b
a
lan
ce
i
n
t
h
e
n
u
m
b
er
o
f
an
n
o
tatio
n
s
b
et
w
ee
n
t
h
e
m
a
j
o
r
ity
an
d
m
i
n
o
r
it
y
cla
s
s
es
in
t
h
e
tr
ai
n
i
n
g
d
ata
o
f
te
n
b
ec
o
m
es
a
s
er
io
u
s
p
r
o
b
lem
b
ec
au
s
e
t
h
e
m
o
d
el
ten
d
s
to
b
e
b
iased
to
w
ar
d
s
th
e
m
aj
o
r
ity
clas
s
an
d
ig
n
o
r
e
th
e
m
i
n
o
r
it
y
clas
s
.
Si
n
ce
th
e
a
m
o
u
n
t
o
f
d
ata
f
o
r
th
e
n
o
r
m
al,
P
VC
,
an
d
P
A
C
clas
s
es
i
n
th
e
tr
ai
n
in
g
d
ata
is
u
n
b
ala
n
ce
d
,
th
e
s
y
n
t
h
eti
c
m
in
o
r
it
y
o
v
er
-
s
a
m
p
li
n
g
tech
n
iq
u
e
(
SMOT
E
)
is
u
s
ed
as
a
d
ata
b
ala
n
ci
n
g
m
et
h
o
d
in
t
h
e
tr
ai
n
in
g
d
ata.
SMO
T
E
w
o
r
k
s
b
y
g
e
n
er
ati
n
g
s
y
n
t
h
etic
s
a
m
p
le
s
i
n
t
h
e
m
i
n
o
r
i
t
y
clas
s
,
n
o
t
s
i
m
p
l
y
d
u
p
licatin
g
d
ata,
b
u
t
b
y
cr
ea
tin
g
n
e
w
d
ata
t
h
r
o
u
g
h
i
n
ter
p
o
lati
o
n
b
et
w
ee
n
ex
i
s
ti
n
g
m
i
n
o
r
it
y
s
a
m
p
les
a
n
d
th
eir
n
e
ig
h
b
o
u
r
s
in
f
ea
tu
r
e
s
p
ac
e.
T
h
u
s
,
th
e
cla
s
s
d
is
tr
ib
u
tio
n
b
ec
o
m
es
m
o
r
e
b
alan
ce
d
,
allo
w
i
n
g
th
e
m
o
d
el
to
lear
n
p
atter
n
s
f
r
o
m
t
h
e
m
i
n
o
r
it
y
cla
s
s
m
o
r
e
ef
f
ec
ti
v
el
y
,
w
h
ich
i
m
p
r
o
v
es
clas
s
if
icatio
n
p
er
f
o
r
m
a
n
ce
,
p
ar
tic
u
lar
l
y
o
n
m
etr
ics
s
u
ch
as
r
ec
all
an
d
F1
-
s
co
r
e.
A
f
ter
d
ata
b
alan
ci
n
g
,
tr
ai
n
in
g
a
n
d
te
s
ti
n
g
ar
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
3
,
J
u
n
e
20
26
:
8
9
1
-
9
03
894
p
er
f
o
r
m
ed
u
s
i
n
g
t
h
e
s
elec
ted
m
ac
h
in
e
lear
n
in
g
m
o
d
els.
H
y
p
er
p
ar
am
eter
tu
n
in
g
f
o
r
KNN,
L
ig
h
tGB
M,
an
d
th
e
en
s
e
m
b
le
clas
s
i
f
ier
is
co
n
d
u
cted
u
s
i
n
g
Gr
id
Sear
c
h
C
V
w
i
th
s
tr
ati
f
ied
5
-
f
o
ld
cr
o
s
s
-
v
ali
d
atio
n
to
ev
alu
a
te
co
m
b
i
n
atio
n
s
o
f
t
h
eir
r
esp
ec
ti
v
e
p
ar
a
m
eter
s
.
2
.
5
.
1
.
K
NN
KNN
is
o
n
e
o
f
t
h
e
s
i
m
p
lest
s
u
p
er
v
is
ed
m
ac
h
in
e
lear
n
in
g
al
g
o
r
ith
m
s
.
T
h
is
m
ea
n
s
t
h
at
KN
N
r
eq
u
ir
es
lab
els
as
o
u
tp
u
t
to
r
ec
o
g
n
is
e
p
atter
n
s
b
et
w
ee
n
i
n
p
u
t
a
n
d
o
u
tp
u
t.
KNN
is
a
n
o
n
-
p
ar
a
m
etr
ic
a
l
g
o
r
ith
m
t
h
at
m
a
k
es
n
o
ass
u
m
p
tio
n
s
ab
o
u
t
t
h
e
d
ata.
T
h
e
w
o
r
k
i
n
g
p
r
in
cip
le
o
f
K
NN
is
to
s
elec
t
a
n
u
m
b
er
o
f
K
-
n
ea
r
e
s
t
d
ata
p
o
in
t
s
as
th
e
b
asis
f
o
r
d
eter
m
i
n
i
n
g
th
e
class
.
K
NN
w
il
l
ca
lc
u
late
t
h
e
d
is
tan
ce
f
r
o
m
th
e
o
b
s
er
v
ed
d
ata
to
all
d
ata
p
o
in
t
s
in
t
h
e
d
ataset.
Ne
x
t,
th
e
K
-
n
ea
r
est
d
ata
p
o
in
ts
is
s
elec
ted
,
an
d
th
e
n
u
m
b
er
o
f
ea
ch
cla
s
s
i
s
ca
lcu
lated
f
r
o
m
th
e
s
e
K
-
n
ea
r
est
d
ata
p
o
in
ts
.
T
h
e
d
ata
b
ein
g
o
b
s
er
v
ed
w
il
l
b
e
class
i
f
ied
in
to
th
e
clas
s
w
it
h
t
h
e
m
o
s
t
n
u
m
er
o
u
s
K
-
n
ea
r
est
d
ata
p
o
in
t
s
ar
o
u
n
d
i
t.
So
m
e
i
m
p
o
r
tan
t
p
ar
a
m
ete
r
s
in
KNN
i
n
cl
u
d
e
th
e
n
u
m
b
er
o
f
n
ei
g
h
b
o
u
r
s
(
_
ℎ
)
an
d
th
e
w
ei
g
h
ts
o
f
ea
ch
n
ei
g
h
b
o
u
r
d
ata
p
o
in
t
th
a
t
ar
e
tak
en
i
n
to
co
n
s
id
er
at
io
n
.
H
y
p
er
p
ar
am
e
ter
tu
n
i
n
g
is
p
er
f
o
r
m
ed
o
n
b
o
th
p
ar
am
eter
s
,
w
h
er
e
th
e
v
ar
iatio
n
in
_
ℎ
is
=
4
−
10
.
As f
o
r
th
e
w
ei
g
h
t,
it is
“
d
is
tan
ce
”
an
d
“u
n
i
f
o
r
m
”.
2
.
5
.
2
.
L
ig
ht
G
B
M
L
ig
h
tG
B
M
i
s
an
al
g
o
r
i
th
m
f
r
o
m
t
h
e
t
r
ee
-
b
a
s
e
d
l
e
a
r
n
in
g
g
r
o
u
p
th
at
u
ti
l
i
z
es
th
e
g
r
a
d
ie
n
t
b
o
o
s
ti
n
g
t
e
c
h
n
i
q
u
e
.
I
t
w
o
r
k
s
s
im
i
l
a
r
ly
to
a
d
e
c
i
s
i
o
n
t
r
e
e
,
b
u
t
th
e
d
if
f
er
e
n
c
e
li
e
s
i
n
h
o
w
th
e
m
o
d
e
l
b
u
i
l
d
s
d
e
c
is
i
o
n
t
r
e
es
i
t
e
r
a
ti
v
e
ly
.
I
n
L
ig
h
t
G
B
M
,
e
a
ch
n
ew
t
r
e
e
i
s
b
u
il
t
t
o
im
p
r
o
v
e
o
n
th
e
e
r
r
o
r
s
o
f
th
e
p
r
ev
i
o
u
s
t
r
ee
b
y
m
in
im
i
z
in
g
t
h
e
l
o
s
s
f
u
n
c
t
i
o
n
u
s
in
g
th
e
g
r
a
d
i
en
t
d
e
s
c
en
t
m
e
th
o
d
.
U
n
l
ik
e
o
th
e
r
a
l
g
o
r
i
th
m
s
,
L
ig
h
t
G
B
M
u
s
e
s
a
l
e
a
f
-
w
is
e
g
r
o
w
th
a
p
p
r
o
a
c
h
,
w
h
e
r
e
b
y
t
r
e
es
a
r
e
d
ev
e
l
o
p
e
d
at
le
av
e
s
th
a
t
h
av
e
th
e
g
r
e
a
te
s
t
p
o
t
en
t
ia
l
f
o
r
l
o
s
s
r
e
d
u
ct
i
o
n
.
T
h
is
a
p
p
r
o
a
ch
m
ak
es
L
ig
h
t
G
B
M
m
o
r
e
e
f
f
i
c
i
en
t
an
d
c
a
p
a
b
l
e
o
f
p
r
o
d
u
ci
n
g
h
ig
h
e
r
a
c
cu
r
a
cy
w
i
th
f
ew
e
r
it
e
r
at
i
o
n
s
,
e
s
p
e
ci
a
lly
o
n
l
a
r
g
e
d
at
a
s
et
s
.
T
o
d
e
t
e
r
m
in
e
th
e
p
r
e
d
i
c
t
i
o
n
r
e
s
u
lt
s
o
n
n
ew
d
a
ta
,
L
ig
h
tG
B
M
w
il
l
s
u
m
t
h
e
c
o
n
t
r
i
b
u
t
i
o
n
s
o
f
a
l
l
th
e
t
r
e
e
s
th
a
t
h
av
e
b
e
e
n
b
u
i
l
t
.
T
h
e
f
in
al
r
e
s
u
l
t
i
s
a
c
o
m
b
in
a
ti
o
n
o
f
a
l
l
th
es
e
t
r
e
es
,
en
a
b
l
in
g
t
h
e
m
o
d
e
l
t
o
p
r
o
v
i
d
e
m
o
r
e
s
t
a
b
l
e
an
d
a
c
cu
r
a
t
e
p
r
e
d
i
ct
i
o
n
s
.
T
o
o
b
t
a
i
n
r
o
b
u
s
t
r
e
s
u
lt
s
,
h
y
p
e
r
p
a
r
am
et
e
r
tu
n
in
g
is
p
e
r
f
o
r
m
e
d
f
o
r
s
e
v
e
r
al
im
p
o
r
t
an
t
p
a
r
am
et
e
r
s
,
in
clu
d
in
g
t
h
e
n
u
m
b
e
r
o
f
e
s
t
im
a
t
o
r
s
[
1
0
0
,
2
0
0
]
,
m
ax
im
u
m
t
r
ee
d
e
p
t
h
[
3
,
5
,
7
]
,
l
e
a
r
n
in
g
r
at
e
[
0
.
0
0
1
,
0
.
0
1
,
0
.
1
]
,
m
ax
im
u
m
n
u
m
b
e
r
o
f
l
ea
v
e
s
in
ea
ch
t
r
e
e
[
3
1
,
6
3
]
,
a
n
d
s
u
b
s
am
p
l
e
[
0
.
6
,
0
.
8
,
1
.
0
]
.
2
.
5
.
3
.
E
ns
e
m
b
le
le
a
rning
E
n
s
e
m
b
le
lear
n
i
n
g
i
s
a
co
m
b
in
atio
n
o
f
s
ev
er
al
m
ac
h
in
e
le
ar
n
in
g
alg
o
r
it
h
m
s
t
h
at
w
o
r
k
t
o
g
eth
er
to
p
er
f
o
r
m
a
s
p
ec
i
f
ic
f
u
n
ct
io
n
,
s
u
ch
as
clas
s
if
icatio
n
.
T
h
e
ad
v
a
n
tag
e
o
f
en
s
e
m
b
le
lear
n
i
n
g
is
th
at
it
ca
n
p
r
o
d
u
ce
a
g
en
er
al
an
d
b
ala
n
ce
d
clas
s
if
ic
atio
n
m
et
h
o
d
,
en
s
u
r
in
g
th
at
t
h
e
cla
s
s
i
f
icatio
n
r
es
u
lt
s
i
n
ea
ch
cla
s
s
ar
e
n
o
t
to
o
d
is
p
ar
ate.
T
h
e
en
s
e
m
b
le
lear
n
i
n
g
ap
p
r
o
ac
h
e
m
p
lo
y
ed
in
t
h
is
s
tu
d
y
u
til
izes
a
v
o
ti
n
g
clas
s
i
f
ie
r
s
tr
ateg
y
w
it
h
s
o
f
t
v
o
tin
g
.
Vo
ti
n
g
cla
s
s
i
f
ier
s
u
til
i
s
e
th
e
ad
v
a
n
ta
g
es
o
f
ea
ch
m
ac
h
in
e
lear
n
i
n
g
m
et
h
o
d
u
s
ed
to
p
r
ed
ict
o
r
class
if
y
a
class
.
Fi
g
u
r
e
2
i
llu
s
tr
ates
t
h
e
o
p
er
atio
n
o
f
th
e
s
o
f
t
v
o
tin
g
class
i
f
ier
.
I
n
s
o
f
t
v
o
tin
g
,
ea
c
h
m
ac
h
i
n
e
lear
n
in
g
m
et
h
o
d
d
o
es
n
o
t
d
ir
ec
tly
ass
i
g
n
a
class
to
th
e
class
if
ied
d
ata
b
u
t
p
r
o
v
id
es
d
if
f
er
en
t
p
r
ed
ictio
n
p
r
o
b
a
b
ilit
ies
(
)
f
o
r
ea
ch
c
lass
if
ied
class
.
I
f
th
e
r
e
ar
e
n
cl
a
s
s
e
s
,
ea
ch
m
ac
h
i
n
e
lear
n
in
g
m
et
h
o
d
w
ill
p
r
o
v
id
e
f
o
r
ea
ch
c
lass
.
I
n
ad
d
itio
n
to
th
e
p
r
ed
ictio
n
p
r
o
b
ab
ilit
y
,
a
n
o
th
er
i
m
p
o
r
tan
t
p
ar
am
eter
is
th
e
w
eig
h
t
(
)
.
Ma
ch
in
e
lear
n
in
g
u
s
er
s
ca
n
d
eter
m
i
n
e
th
e
w
e
ig
h
t
t
h
e
y
w
a
n
t
to
ass
ig
n
;
h
o
w
e
v
er
,
if
i
t
is
n
o
t
s
p
ec
if
ied
,
th
e
s
y
s
te
m
as
s
ig
n
s
t
h
e
s
a
m
e
w
ei
g
h
t
to
ea
ch
m
ac
h
in
e
lear
n
i
n
g
m
eth
o
d
.
A
f
ter
ea
ch
m
ac
h
i
n
e
lear
n
in
g
m
et
h
o
d
p
er
f
o
r
m
s
t
h
e
cla
s
s
i
f
i
ca
tio
n
p
r
o
ce
s
s
,
t
h
e
s
u
m
o
f
th
e
w
ei
g
h
t
an
d
th
e
p
r
ed
ictio
n
p
r
o
b
a
b
ilit
y
o
f
ea
ch
m
ac
h
in
e
lear
n
in
g
m
et
h
o
d
w
i
ll
b
e
ca
lc
u
lated
.
T
h
e
o
b
s
er
v
ed
d
ata
w
ill
b
e
clas
s
if
ied
ac
co
r
d
in
g
to
th
e
class
w
i
th
th
e
h
i
g
h
est v
alu
e.
Si
m
ilar
to
o
th
er
m
et
h
o
d
s
,
o
r
th
e
w
ei
g
h
ts
f
o
r
b
o
th
m
ac
h
in
e
lear
n
i
n
g
m
o
d
els
w
ill
b
e
v
ar
ied
as
f
o
llo
w
s
:
[
0
.
9
;0
.
1
]
,
[
0
.
8
;0
.
2
]
,
[
0
.
7
;0
.
3
]
,
[
0
.
6
;0
.
4
]
,
[
0
.
5
;0
.
5
]
.
T
h
e
f
ir
s
t
n
u
m
b
er
in
d
icate
s
t
h
e
w
ei
g
h
t
f
o
r
m
ac
h
in
e
lear
n
i
n
g
1
,
a
n
d
th
e
s
ec
o
n
d
n
u
m
b
e
r
in
d
icate
s
t
h
e
w
ei
g
h
t f
o
r
m
ac
h
i
n
e
lear
n
i
n
g
2
.
C
las
s
i
f
icatio
n
p
er
f
o
r
m
a
n
ce
i
s
ev
alu
a
ted
u
s
i
n
g
p
r
ec
is
io
n
,
r
ec
all,
an
d
F1
-
s
co
r
e,
w
h
ic
h
ar
e
p
ar
ticu
lar
l
y
in
f
o
r
m
ati
v
e
f
o
r
i
m
b
ala
n
ce
d
d
atasets
.
T
h
ese
m
etr
ics
ar
e
d
er
i
v
ed
f
r
o
m
th
e
co
n
f
u
s
io
n
m
atr
i
x
,
w
h
ich
co
m
p
r
i
s
es
tr
u
e
p
o
s
iti
v
es
(
T
P
)
,
f
alse
p
o
s
itiv
es
(
FP
)
,
tr
u
e
n
eg
at
iv
e
s
(
T
N)
,
an
d
f
al
s
e
n
eg
ati
v
e
s
(
FN)
.
I
n
ad
d
itio
n
,
Mc
Ne
m
ar
’
s
test
is
e
m
p
lo
y
ed
to
s
tati
s
ticall
y
ass
es
s
th
e
s
i
g
n
i
f
ica
n
ce
o
f
p
er
f
o
r
m
a
n
ce
d
if
f
er
e
n
ce
s
b
et
w
ee
n
class
i
f
ier
s
b
ased
o
n
p
air
ed
p
r
ed
ictio
n
s
o
n
th
e
s
a
m
e
test
s
et.
=
+
×
100
(
1
)
=
+
×
100
(
2
)
1
=
2
×
(
×
)
+
×
100
(
3
)
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
C
la
s
s
i
fica
tio
n
o
f p
r
ema
tu
r
e
ca
r
d
ia
c
co
n
tr
a
ctio
n
s
b
a
s
ed
o
n
R
F
E
C
V
…
(
E
ls
a
S
a
r
i H
a
yu
n
a
h
N
u
r
d
in
iya
h
)
895
2
.
6
.
SH
AP
(
f
ea
t
ures int
er
pret
a
t
io
n)
SH
A
P
is
a
m
et
h
o
d
f
o
r
in
ter
p
r
etin
g
m
ac
h
i
n
e
lear
n
i
n
g
m
o
d
el
s
b
y
f
air
l
y
m
ea
s
u
r
in
g
t
h
e
co
n
t
r
ib
u
tio
n
o
f
ea
ch
f
ea
tu
r
e
to
a
p
r
ed
ictio
n
u
s
i
n
g
S
h
ap
le
y
v
al
u
es
f
r
o
m
co
o
p
er
ativ
e
g
a
m
e
th
eo
r
y
[
4
3
]
,
[
4
4
]
.
B
y
ca
lcu
la
tin
g
h
o
w
p
r
ed
ictio
n
s
ch
a
n
g
e
w
h
en
f
ea
tu
r
es
ar
e
ad
d
ed
o
r
r
e
m
o
v
ed
,
SH
A
P
en
s
u
r
es
in
ter
p
r
etab
ilit
y
b
ased
o
n
lo
ca
l
ac
cu
r
ac
y
,
co
n
s
is
te
n
c
y
,
an
d
t
h
e
p
r
esen
ce
o
f
m
i
s
s
i
n
g
v
al
u
es.
I
n
th
i
s
s
tu
d
y
,
SH
A
P
w
as
u
s
ed
to
ex
p
lain
th
e
b
es
t
en
s
e
m
b
le
m
o
d
el
’
s
E
C
G
-
b
ased
h
ea
r
t c
o
n
tr
ac
tio
n
clas
s
i
f
icatio
n
,
p
r
o
v
id
in
g
tr
a
n
s
p
ar
en
c
y
an
d
clin
ical
r
ele
v
an
ce
.
T
h
e
SHA
P
r
esu
lt
s
ar
e
v
is
u
a
li
ze
d
in
t
w
o
w
a
y
s
:
a
s
u
m
m
ar
y
p
lo
t
an
d
class
-
s
p
ec
i
f
ic
p
lo
ts
.
I
n
class
-
s
p
ec
if
ic
p
lo
ts
,
th
e
x
-
a
x
is
s
h
o
w
s
t
h
e
S
H
A
P
v
al
u
e,
r
ep
r
esen
t
in
g
t
h
e
d
ir
ec
tio
n
a
n
d
m
ag
n
it
u
d
e
o
f
ea
ch
f
ea
t
u
r
e
’
s
in
f
lu
e
n
ce
o
n
th
e
p
r
o
b
ab
ilit
y
o
f
p
r
ed
ictin
g
a
g
iv
e
n
clas
s
.
Do
ts
r
ep
r
esen
t
s
a
m
p
les,
w
it
h
co
lo
r
in
d
icati
n
g
o
r
ig
i
n
al
fe
at
u
r
e
v
alu
e
s
(
r
ed
f
o
r
h
ig
h
,
b
lu
e
f
o
r
lo
w
)
.
R
ed
p
o
in
ts
o
n
t
h
e
r
ig
h
t
i
n
d
icate
th
at
a
h
ig
h
f
ea
t
u
r
e
v
al
u
e
in
cr
ea
s
e
s
th
e
li
k
eli
h
o
o
d
o
f
t
h
at
clas
s
,
wh
ile
b
l
u
e
p
o
in
t
s
o
n
t
h
e
r
i
g
h
t
o
r
r
ed
p
o
in
ts
o
n
th
e
le
f
t
s
h
o
w
th
e
o
p
p
o
s
ite.
T
h
e
s
u
m
m
ar
y
p
lo
t
r
an
k
s
f
ea
t
u
r
e
g
l
o
b
all
y
b
y
th
e
ir
av
er
ag
e
ab
s
o
lu
te
SH
A
P
v
al
u
e,
h
i
g
h
li
g
h
tin
g
t
h
e
m
o
s
t
i
n
f
l
u
e
n
tia
l
f
ea
t
u
r
es in
t
h
e
m
o
d
el.
Fig
u
r
e
2
.
Flo
w
c
h
ar
t so
f
t
v
o
ti
n
g
clas
s
if
ier
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
I
n
th
i
s
s
ec
tio
n
,
it i
s
ex
p
lai
n
ed
th
e
r
esu
l
ts
o
f
r
esear
c
h
an
d
at
th
e
s
a
m
e
ti
m
e
is
g
i
v
e
n
t
h
e
co
m
p
r
eh
en
s
iv
e
d
is
cu
s
s
io
n
.
3
.
1
.
T
he
cla
s
s
if
ica
t
io
n r
esu
lt
s
o
f
K
NN
T
h
e
s
elec
tio
n
o
f
h
y
p
er
p
ar
a
m
et
er
=
4
w
i
th
d
is
ta
n
ce
w
ei
g
h
tin
g
m
e
an
s
t
h
e
KNN
m
o
d
el
d
ec
id
es
b
ased
o
n
th
e
f
o
u
r
n
ea
r
est d
ata
p
o
in
ts
.
Data
clo
s
er
in
d
is
tan
ce
co
n
tr
ib
u
tes
m
o
r
e
to
th
e
d
ec
is
io
n
.
T
h
is
m
ak
e
s
th
e
m
o
d
el
s
en
s
iti
v
e
to
t
h
e
lo
ca
l
s
tr
u
ct
u
r
e
o
f
th
e
d
ata.
Fo
r
ex
a
m
p
le,
if
th
e
r
e
is
a
‘
P
AC
’
c
lass
ex
a
m
p
le
i
n
an
ar
ea
w
h
er
e
m
o
s
t
n
ea
r
b
y
da
ta
ar
e
“
P
VC
”
o
r
‘
N
’
,
th
e
m
o
d
el
w
ill
lean
to
w
ar
d
s
th
e
d
o
m
i
n
a
n
t
class
.
Dis
ta
n
ce
weig
h
tin
g
r
ed
u
ce
s
th
e
‘
v
o
te
tie
’
ef
f
ec
t
co
m
p
ar
ed
to
u
n
i
f
o
r
m
w
ei
g
h
ti
n
g
.
Ho
w
e
v
er
,
it
in
cr
ea
s
es
t
h
e
i
m
p
ac
t
o
f
ea
c
h
n
ea
r
est
d
ata
p
o
in
t,
w
h
ic
h
co
u
ld
b
e
an
o
u
tlier
o
r
n
ea
r
l
y
id
en
tical
b
et
w
ee
n
cla
s
s
e
s
(
class
o
v
er
lap
)
.
T
h
is
co
m
b
in
atio
n
o
f
=
4
an
d
d
is
tan
ce
w
ei
g
h
ts
r
es
u
lted
in
h
i
g
h
o
v
er
all
ac
cu
r
ac
y
.
T
h
e
tr
ad
e
-
o
f
f
:
t
h
e
m
in
o
r
P
AC
clas
s
s
u
f
f
e
r
ed
f
r
o
m
lo
w
r
ec
all.
A
lt
h
o
u
g
h
t
h
e
d
etec
tio
n
o
f
P
AC
w
h
e
n
p
r
ed
icted
w
a
s
q
u
ite
ac
cu
r
ate,
m
a
n
y
P
AC
d
ata
p
o
in
ts
w
er
e
n
o
t
r
ec
o
g
n
i
s
ed
as P
A
C
.
F
ig
u
r
e
3
s
h
o
w
t
h
e
c
l
as
s
if
i
c
at
i
o
n
r
es
u
l
ts
o
f
K
NN
.
T
h
e
c
l
ass
if
i
ca
t
i
o
n
r
e
s
u
lt
s
u
s
in
g
KN
N
s
h
o
w
ed
e
x
c
el
l
en
t
p
e
r
f
o
r
m
an
c
e
in
th
e
n
o
r
m
al
c
l
as
s
,
w
ith
a
r
e
c
a
l
l
o
f
9
7
.
4
5
%
a
n
d
a
p
r
e
c
i
s
i
o
n
o
f
9
8
.
5
%
,
i
n
d
i
c
at
i
n
g
th
at
n
e
a
r
ly
al
l
n
o
r
m
a
l
s
am
p
l
e
s
w
e
r
e
a
c
cu
r
at
e
ly
d
et
e
c
te
d
w
i
th
a
v
e
r
y
l
o
w
e
r
r
o
r
r
a
t
e
.
I
n
t
h
e
PV
C
c
l
ass
,
t
h
e
m
o
d
e
l
w
as
a
l
s
o
q
u
i
te
r
e
l
i
a
b
l
e
w
it
h
a
r
e
c
a
ll
o
f
9
3
.
39
%
an
d
p
r
e
ci
s
i
o
n
o
f
8
8
.
47
%
,
a
lt
h
o
u
g
h
th
e
r
e
w
e
r
e
s
t
i
ll
a
n
u
m
b
e
r
o
f
f
a
ls
e
p
o
s
i
ti
v
e
s
,
m
a
in
ly
f
r
o
m
th
e
P
A
C
c
la
s
s
.
Me
a
n
w
h
il
e
,
th
e
P
A
C
c
l
as
s
h
a
d
r
el
a
t
iv
e
ly
l
o
w
e
r
p
e
r
f
o
r
m
an
ce
,
as
i
n
d
i
ca
t
e
d
b
y
a
r
e
ca
ll
o
f
8
1
.
1
%
an
d
a
p
r
e
c
is
i
o
n
o
f
8
7
.
26
%
,
w
h
e
r
e
m
o
s
t
o
f
th
e
o
r
ig
i
n
al
P
A
C
s
w
e
r
e
c
l
ass
if
i
e
d
a
s
PV
C
.
T
h
i
s
in
d
i
ca
t
es
f
e
at
u
r
e
o
v
e
r
l
a
p
b
e
t
w
e
en
PA
C
a
n
d
PV
C
,
as
w
e
ll
a
s
l
im
it
at
i
o
n
s
in
th
e
r
e
p
r
es
en
t
a
ti
o
n
o
f
P
A
C
d
a
t
a
.
Ov
e
r
a
ll
,
th
e
m
o
d
e
l
’
s
a
cc
u
r
ac
y
r
e
a
ch
e
d
9
4
.
1
8
%
,
w
ith
th
e
p
r
im
a
r
y
c
h
al
l
en
g
e
b
e
in
g
t
o
im
p
r
o
v
e
th
e
s
en
s
i
tiv
i
ty
o
f
P
A
C
d
e
t
e
ct
i
o
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
3
,
J
u
n
e
20
26
:
8
9
1
-
9
03
896
3
.
2
.
T
he
cla
s
s
if
ica
t
io
n r
esu
lt
s
o
f
L
ig
htG
B
M
Af
ter
th
e
t
u
n
in
g
p
r
o
ce
s
s
,
t
h
e
L
ig
h
t
GB
M
m
o
d
el
w
as
s
e
lecte
d
w
it
h
th
e
f
o
llo
w
i
n
g
p
ar
a
m
eter
s
:
_
=
200
,
_
=
0
.
1
,
_
ℎ
=
5
,
_
=
31
,
an
d
p
ar
tial
s
a
m
p
le
an
d
f
ea
t
u
r
e
s
etti
n
g
s
,
s
p
ec
if
ical
l
y
s
u
b
s
a
m
p
le
=
0
.
8
.
T
h
is
co
m
b
i
n
ati
o
n
b
alan
ce
s
m
o
d
e
l
co
m
p
le
x
it
y
an
d
g
en
er
ali
s
atio
n
ab
ilit
y
:
m
ax
_
d
ep
th
=
5
li
m
its
t
h
e
m
a
x
i
m
u
m
d
ep
th
o
f
th
e
tr
ee
s
o
th
at
it
is
n
o
t
to
o
d
ee
p
,
th
er
eb
y
r
ed
u
ci
n
g
th
e
r
i
s
k
o
f
o
v
er
f
itti
n
g
;
_
=
31
p
r
o
v
id
es
an
a
d
e
q
u
ate
n
u
m
b
er
o
f
leav
es
s
o
th
at
th
e
tr
ee
is
s
till
f
lex
ib
l
e
en
o
u
g
h
to
ca
p
tu
r
e
d
ata
v
ar
iatio
n
s
i
n
e
ac
h
b
o
o
s
tin
g
iter
atio
n
,
b
u
t
is
n
o
t
to
o
co
m
p
licated
.
_
=
0
.
1
s
ets
a
m
o
d
er
ate
lear
n
in
g
s
p
ee
d
,
allo
w
i
n
g
m
a
n
y
tr
ee
s
(
e
n
s
e
m
b
les)
to
co
n
tr
ib
u
te
g
r
ad
u
a
ll
y
,
w
h
il
e
_
=
200
p
r
o
v
id
es
en
o
u
g
h
b
o
o
s
tin
g
r
o
u
n
d
s
f
o
r
th
e
m
o
d
el
to
lear
n
co
m
p
lex
p
atter
n
s
.
T
h
e
s
u
b
s
a
m
p
l
e
s
ettin
g
o
f
0
.
8
(
s
elec
tin
g
8
0
%
o
f
d
ata
r
o
w
s
p
er
iter
atio
n
)
in
tr
o
d
u
ce
s
an
ele
m
en
t
o
f
r
an
d
o
m
n
es
s
th
at
h
elp
s
r
ed
u
ce
o
v
er
f
itti
n
g
an
d
i
m
p
r
o
v
e
r
o
b
u
s
t
n
e
s
s
.
T
h
e
r
an
d
o
m
s
tate
is
s
e
t
s
o
t
h
at
t
h
e
r
esu
lt
s
ca
n
b
e
r
ep
r
o
d
u
ce
d
.
Ov
er
all,
t
h
is
t
u
n
i
n
g
ap
p
ea
r
s
to
h
av
e
s
u
cc
es
s
f
u
ll
y
cr
ea
ted
a
g
o
o
d
t
r
ad
e
-
o
f
f
b
etw
ee
n
b
ias
an
d
v
ar
ian
ce
,
en
ab
lin
g
th
e
m
o
d
el
to
g
en
er
alize
s
tr
o
n
g
l
y
w
h
ile
s
till
ca
p
tu
r
in
g
i
m
p
o
r
tan
t p
atter
n
s
w
it
h
o
u
t e
x
ce
s
s
i
v
e
o
v
er
f
itti
n
g
.
I
n
th
is
s
tu
d
y
Fi
g
u
r
e
4
s
h
o
w
t
h
e
cla
s
s
i
f
icat
io
n
r
es
u
lt
s
o
f
L
ig
h
tGB
M,
t
h
e
clas
s
i
f
ica
tio
n
m
o
d
el
d
em
o
n
s
tr
ated
a
v
er
y
h
ig
h
ca
p
ab
ilit
y
in
r
ec
o
g
n
izi
n
g
cla
s
s
N,
w
it
h
a
p
r
ec
is
io
n
o
f
ar
o
u
n
d
9
8
.
1
5
%
an
d
a
r
ec
all
o
f
n
ea
r
l
y
9
9
.
8
2
%,
r
esu
lti
n
g
in
a
class
N
F1
-
s
co
r
e
o
f
ap
p
r
o
x
i
m
atel
y
9
8
.
9
8
%.
T
h
is
m
ea
n
s
t
h
at
t
h
e
m
o
d
el
m
ad
e
al
m
o
s
t
n
o
m
is
ta
k
es
i
n
p
r
ed
i
ctin
g
s
a
m
p
le
s
f
r
o
m
clas
s
N,
w
h
er
e
o
u
t
o
f
2
,
2
3
7
d
ata
p
o
in
ts
,
o
n
l
y
4
w
er
e
m
is
c
lass
if
ied
.
Fo
r
th
e
P
VC
class
,
th
e
p
er
f
o
r
m
a
n
ce
w
a
s
also
q
u
ite
g
o
o
d
: P
VC
r
ec
all
w
a
s
ar
o
u
n
d
9
5
.
3
8
%,
w
h
ile
p
r
ec
is
io
n
w
as
in
th
e
r
an
g
e
o
f
8
8
.
2
2
%,
r
esu
ltin
g
i
n
a
n
F1
-
s
co
r
e
o
f
9
1
.
6
6
%.
T
h
is
in
d
icate
s
th
a
t
m
o
s
t
P
VC
s
a
m
p
le
s
w
er
e
s
u
cc
e
s
s
f
u
ll
y
r
ec
o
g
n
ized
,
alth
o
u
g
h
t
h
er
e
w
er
e
s
t
ill
p
r
ed
ictio
n
er
r
o
r
s
f
r
o
m
o
th
er
cla
s
s
e
s
,
esp
ec
iall
y
P
AC
,
w
h
ic
h
w
a
s
also
m
i
s
clas
s
i
f
ie
d
as
P
VC
.
T
h
e
P
AC
cla
s
s
,
d
esp
ite
ac
h
ie
v
i
n
g
a
h
ig
h
p
r
e
cisi
o
n
o
f
9
5
.
1
9
%,
ex
p
er
ien
ce
d
a
s
ig
n
i
f
ica
n
t
d
ec
r
ea
s
e
in
r
ec
all
o
f
ap
p
r
o
x
i
m
atel
y
7
1
.
4
9
%,
r
esu
ltin
g
in
a
P
A
C
F1
-
s
co
r
e
o
f
o
n
ly
8
1
.
6
5
%.
T
h
is
r
ef
lect
s
th
at
alt
h
o
u
g
h
m
a
n
y
d
ata
w
er
e
s
u
cc
e
s
s
f
u
ll
y
cla
s
s
i
f
ied
as
P
AC
,
2
7
%
o
f
P
A
C
s
a
m
p
les
w
er
e
n
o
t
id
en
ti
f
ied
as
P
AC
an
d
w
er
e
class
i
f
ied
in
to
th
e
P
VC
clas
s
.
T
h
e
d
if
f
er
en
ce
in
m
etr
ics
b
et
w
ee
n
clas
s
es
clea
r
l
y
s
h
o
w
s
t
h
at
th
e
m
o
d
el
’
s
p
er
f
o
r
m
an
ce
i
s
m
o
r
e
co
n
s
is
ten
t
a
n
d
r
eliab
le
f
o
r
class
es w
ith
h
ig
h
er
p
r
ev
alen
ce
,
s
u
c
h
as
N
an
d
P
VC
.
I
n
co
n
tr
ast
,
f
o
r
m
i
n
o
r
it
y
cla
s
s
es,
s
u
c
h
as
P
A
C
,
d
e
s
p
ite
h
i
g
h
p
r
ec
is
io
n
,
s
e
n
s
i
tiv
it
y
r
e
m
ai
n
s
r
elat
iv
el
y
lo
w
i
n
th
i
s
d
ataset
a
n
d
ev
al
u
at
io
n
s
etti
n
g
.
3
.
3
.
T
he
cla
s
s
if
ica
t
io
n r
esu
lt
s
o
f
ens
e
m
ble le
a
rni
ng
I
n
th
is
e
n
s
e
m
b
le
lear
n
in
g
ap
p
r
o
ac
h
,
KNN
a
n
d
L
ig
h
tGB
M
w
e
r
e
u
tili
ze
d
w
i
th
p
ar
a
m
eter
s
d
er
iv
ed
f
r
o
m
p
r
ev
io
u
s
h
y
p
er
p
ar
a
m
eter
tu
n
i
n
g
.
Me
an
w
h
i
le,
th
e
s
o
f
t
v
o
tin
g
w
e
ig
h
ts
t
h
at
y
ield
ed
th
e
o
p
ti
m
u
m
r
es
u
lt
s
w
er
e
[
0
.
6
,
0
.
4
]
f
o
r
ea
ch
m
ac
h
in
e
le
ar
n
in
g
m
o
d
el.
T
h
e
class
if
icati
o
n
r
esu
lts
i
n
Fig
u
r
e
5
o
b
tain
ed
in
d
icate
th
at
th
e
m
o
d
el
e
x
h
i
b
its
v
er
y
h
i
g
h
p
er
f
o
r
m
a
n
ce
f
o
r
th
e
N
clas
s
,
w
it
h
a
p
r
ec
is
io
n
o
f
ap
p
r
o
x
i
m
atel
y
9
8
.
7
5
%
an
d
a
r
ec
all
o
f
ap
p
r
o
x
im
ate
l
y
9
9
.
0
6
%,
r
es
u
lti
n
g
i
n
an
F1
-
s
co
r
e
o
f
ap
p
r
o
x
i
m
atel
y
9
8
.
9
0
%.
T
h
ese
f
ig
u
r
es
in
d
icate
th
at
t
h
e
m
o
d
el
al
m
o
s
t
al
w
a
y
s
s
u
cc
ee
d
s
in
co
r
r
ec
tl
y
r
ec
o
g
n
is
i
n
g
N
s
a
m
p
les
a
n
d
m
a
k
es
v
er
y
f
e
w
f
al
s
e
p
o
s
itiv
es
a
n
d
f
a
ls
e
n
eg
at
iv
e
s
in
t
h
is
cla
s
s
.
Fo
r
th
e
P
VC
class
,
th
e
p
er
f
o
r
m
a
n
ce
is
also
s
tr
o
n
g
.
T
h
e
p
r
ec
is
io
n
o
f
P
VC
p
r
ed
ictio
n
s
r
ea
ch
es
9
0
.
5
9
%,
in
d
icatin
g
th
at
m
o
s
t
P
VC
p
r
ed
ictio
n
s
ar
e
co
r
r
ec
t,
alth
o
u
g
h
th
er
e
ar
e
a
n
u
m
b
er
o
f
in
co
r
r
ec
t
p
r
ed
ictio
n
s
f
r
o
m
o
th
er
class
es.
T
h
e
P
VC
r
ec
all
r
ea
ch
es
9
4
.
4
3
%,
m
ea
n
i
n
g
th
at
m
o
s
t
P
VC
s
am
p
le
s
ar
e
s
u
cc
e
s
s
f
u
ll
y
id
en
ti
f
ied
.
T
h
e
F
1
-
s
co
r
e
f
o
r
th
e
P
VC
c
lass
r
ea
ch
ed
9
2
.
4
7
%,
in
d
icatin
g
a
g
o
o
d
b
alan
ce
b
et
w
ee
n
p
r
ec
is
io
n
an
d
r
ec
all
f
o
r
th
e
P
VC
cl
as
s
,
alt
h
o
u
g
h
n
o
t
as
h
i
g
h
as
t
h
at
o
f
t
h
e
N
clas
s
.
Me
an
w
h
ile,
t
h
e
P
AC
cla
s
s
s
h
o
w
ed
s
l
ig
h
tl
y
d
i
f
f
er
e
n
t
r
es
u
lts
.
T
h
e
P
AC
p
r
ec
is
io
n
w
as
q
u
ite
h
i
g
h
,
r
ea
c
h
in
g
8
9
.
9
1
%,
in
d
icatin
g
t
h
at
w
h
e
n
th
e
m
o
d
el
p
r
ed
icted
P
A
C
,
t
h
e
p
r
ed
ictio
n
w
a
s
r
elati
v
el
y
r
elia
b
le.
Ho
w
e
v
er
,
th
e
P
A
C
r
ec
all
is
lo
w
er
,
at
ar
o
u
n
d
8
0
.
5
2
%,
w
h
ich
m
ea
n
s
th
at
a
p
r
o
p
o
r
ti
o
n
o
f
P
AC
d
ata
is
n
o
t
s
u
cc
es
s
f
u
ll
y
d
etec
ted
as
P
A
C
(
ap
p
r
o
x
im
a
tel
y
1
9
-
2
0
%
o
f
P
A
C
s
a
m
p
le
s
ar
e
p
r
e
d
icted
as
o
th
er
class
es,
esp
ec
iall
y
P
VC
)
.
T
h
e
P
A
C
F1
-
s
co
r
e
is
ap
p
r
o
x
i
m
atel
y
8
4
.
96
%,
in
d
icatin
g
t
h
at
alt
h
o
u
g
h
t
h
e
P
A
C
p
r
ed
ictio
n
i
s
q
u
ite
p
r
ec
is
e,
th
e
o
v
er
all
p
er
f
o
r
m
a
n
ce
o
f
th
e
P
A
C
cla
s
s
is
s
ti
ll
a
f
f
ec
ted
b
y
th
e
r
elat
iv
el
y
lo
w
r
ec
all
v
alu
e.
T
h
e
d
i
f
f
er
e
n
ce
s
i
n
t
h
ese
m
etr
ic
s
co
n
s
is
te
n
tl
y
s
h
o
w
t
h
at
cla
s
s
e
s
w
it
h
h
i
g
h
er
p
r
ev
ale
n
ce
(
p
ar
ticu
lar
l
y
t
h
e
N
cla
s
s
)
ac
h
iev
e
t
h
e
m
o
s
t
o
p
ti
m
a
l
p
er
f
o
r
m
a
n
ce
,
w
h
ile
m
i
n
o
r
it
y
cla
s
s
e
s
,
s
u
c
h
as
P
A
C
,
d
esp
ite
o
b
tain
in
g
g
o
o
d
p
r
ec
is
io
n
,
s
till
ex
p
er
ien
ce
a
g
ap
in
s
en
s
iti
v
it
y
(
r
ec
all
)
.
I
n
th
e
co
n
tex
t
o
f
ap
p
licatio
n
s
w
h
er
e
er
r
o
r
s
in
d
e
tectin
g
th
e
P
AC
class
ca
r
r
y
s
ig
n
i
f
i
ca
n
t
w
ei
g
h
t,
th
e
lo
w
er
r
ec
all
o
f
P
A
C
b
ec
o
m
e
s
a
cr
itical
p
o
in
t
in
in
ter
p
r
eti
n
g
p
er
f
o
r
m
a
n
ce
.
Fo
r
s
cie
n
ti
f
ic
r
e
p
o
r
ts
,
th
e
m
ar
g
in
b
et
w
ee
n
p
r
ec
is
io
n
a
n
d
r
ec
all
f
o
r
ea
ch
clas
s
i
s
i
m
p
o
r
ta
n
t
to
n
o
te,
as
it
i
n
d
icate
s
th
e
m
o
d
el
’
s
b
i
as
to
w
ar
d
s
th
e
d
o
m
i
n
an
t
clas
s
in
t
h
e
d
atase
t,
an
d
s
h
o
w
s
t
h
at
a
n
o
v
er
all
ev
al
u
atio
n
u
s
in
g
o
n
l
y
ac
cu
r
ac
y
o
r
g
lo
b
al
m
e
tr
ics
m
a
y
m
as
k
th
e
r
elativ
e
l
y
w
ea
k
p
er
f
o
r
m
a
n
ce
o
n
s
m
al
ler
class
e
s
[
4
5
]
.
I
n
th
is
s
tu
d
y
,
s
o
f
t
v
o
ti
n
g
w
eig
h
t
s
w
er
e
v
ar
ied
u
s
in
g
d
is
cr
ete
in
cr
e
m
en
t
s
to
p
r
o
v
id
e
a
tr
an
s
p
ar
en
t
an
d
co
m
p
u
tatio
n
all
y
ef
f
icie
n
t
co
m
p
ar
i
s
o
n
.
W
h
i
le
th
i
s
ap
p
r
o
ac
h
w
as
s
u
f
f
icie
n
t
f
o
r
id
en
ti
f
y
i
n
g
an
ef
f
ec
ti
v
e
en
s
e
m
b
le
co
n
f
i
g
u
r
atio
n
,
m
o
r
e
ad
v
an
ce
d
o
p
ti
m
izatio
n
m
et
h
o
d
s
s
u
ch
as
B
a
y
e
s
ian
o
p
tim
izatio
n
o
r
ev
o
lu
t
io
n
ar
y
alg
o
r
ith
m
s
co
u
ld
b
e
ex
p
lo
r
ed
in
f
u
t
u
r
e
w
o
r
k
to
p
o
ten
tiall
y
f
u
r
th
er
i
m
p
r
o
v
e
en
s
e
m
b
le
p
er
f
o
r
m
a
n
ce
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
C
la
s
s
i
fica
tio
n
o
f p
r
ema
tu
r
e
ca
r
d
ia
c
co
n
tr
a
ctio
n
s
b
a
s
ed
o
n
R
F
E
C
V
…
(
E
ls
a
S
a
r
i H
a
yu
n
a
h
N
u
r
d
in
iya
h
)
897
Fig
u
r
e
3
.
C
o
n
f
u
s
io
n
m
atr
ix
o
f
KNN
Fig
u
r
e
4
.
C
o
n
f
u
s
io
n
m
atr
ix
o
f
L
i
g
h
tGB
M
Fig
u
r
e
5
.
C
o
n
f
u
s
io
n
m
atr
ix
o
f
en
s
e
m
b
le
lear
n
i
n
g
3
.
4
.
Dis
cus
s
io
n
T
ab
le
1
s
h
o
w
s
t
h
e
cla
s
s
i
f
icati
o
n
r
esu
l
ts
o
f
t
h
e
KNN,
L
i
g
h
t
GB
M,
an
d
en
s
e
m
b
le
lear
n
i
n
g
alg
o
r
ith
m
s
.
I
n
ev
al
u
ati
n
g
p
er
f
o
r
m
a
n
ce
b
e
t
w
ee
n
s
in
g
le
m
et
h
o
d
s
a
n
d
en
s
e
m
b
le
m
et
h
o
d
s
,
L
ig
h
tGB
M
s
h
o
w
ed
s
u
f
f
icien
t
p
er
f
o
r
m
a
n
ce
a
n
d
p
r
ed
ictio
n
s
t
ab
ilit
y
co
m
p
ar
ed
to
K
NN.
L
i
g
h
tGB
M,
a
tr
ee
-
b
ased
g
r
ad
ien
t b
o
o
s
tin
g
a
lg
o
r
it
h
m
,
ca
p
tu
r
es
co
m
p
lex
f
ea
tu
r
e
in
ter
ac
tio
n
s
th
r
o
u
g
h
a
leaf
-
w
is
e
g
r
o
w
t
h
s
tr
ateg
y
th
a
t
iter
ativ
el
y
s
elec
ts
th
e
lea
f
w
i
t
h
th
e
m
a
x
i
m
u
m
lo
s
s
r
ed
u
ctio
n
,
o
f
ten
lead
in
g
to
d
ee
p
er
tr
ee
s
an
d
b
etter
p
r
e
d
ictiv
e
p
er
f
o
r
m
an
ce
[
4
6
]
,
[
4
7
]
.
C
o
n
v
er
s
el
y
,
KNN
h
a
s
an
ad
v
a
n
tag
e
i
n
r
ec
all
o
r
s
e
n
s
it
iv
i
t
y
in
class
i
f
y
in
g
P
A
C
.
T
h
i
s
is
b
ec
au
s
e
KNN
is
a
m
et
h
o
d
th
at
r
elies
o
n
t
h
e
n
ea
r
e
s
t
n
ei
g
h
b
o
u
r
in
g
clas
s
in
t
h
e
f
ea
tu
r
e
s
p
ac
e,
s
o
w
h
e
n
P
AC
s
a
m
p
le
s
a
r
e
in
a
lo
ca
l
clu
s
ter
th
at
h
as
s
i
m
ilar
,
KN
N
w
il
l
b
e
b
et
ter
ab
le
to
class
i
f
y
t
h
e
P
A
C
class
co
r
r
ec
tl
y
,
ev
e
n
t
h
o
u
g
h
t
h
e
n
u
m
b
er
o
f
P
AC
s
is
s
m
aller
th
a
n
o
t
h
er
clas
s
es.
Me
an
w
h
ile,
L
i
g
h
tGB
M
is
v
er
y
e
f
f
ec
tiv
e
in
g
lo
b
al
er
r
o
r
o
p
ti
m
is
a
tio
n
,
m
a
k
in
g
it
less
r
esp
o
n
s
i
v
e
to
r
ar
e
lo
ca
l
p
at
ter
n
s
(
w
h
ic
h
ap
p
ea
r
in
m
i
n
o
r
cl
ass
es)
,
p
ar
tic
u
lar
l
y
w
h
e
n
s
u
c
h
p
atter
n
s
co
n
tr
ib
u
te
m
i
n
i
m
all
y
to
o
v
er
all
lo
s
s
r
ed
u
ctio
n
d
u
r
in
g
b
o
o
s
tin
g
.
On
th
e
o
th
er
h
a
n
d
,
th
e
KNN
–
L
ig
h
tGB
M
en
s
e
m
b
le
s
u
cc
es
s
f
u
ll
y
co
m
b
in
e
s
th
e
ad
v
a
n
ta
g
es
o
f
b
o
th
m
et
h
o
d
s
an
d
u
tili
s
es
L
i
g
h
t
GB
M
’
s
s
tr
en
g
th
i
n
h
a
n
d
li
n
g
n
o
n
-
lin
ea
r
f
ea
t
u
r
es
an
d
d
ataset
co
m
p
lex
i
t
y
,
as
w
ell
as
KNN
’
s
ab
ilit
y
to
p
r
o
v
id
e
s
en
s
i
tiv
it
y
to
lo
ca
l
v
ar
iab
ilit
y
,
s
o
th
at
th
is
en
s
e
m
b
le
s
h
o
w
s
an
i
m
p
r
o
v
e
m
e
n
t
in
m
e
tr
ics
s
u
c
h
as
r
ec
all
a
n
d
F1
-
s
co
r
e
i
n
m
i
n
o
r
class
es
co
m
p
ar
ed
to
th
e
t
w
o
s
i
n
g
le
m
eth
o
d
s
.
Mo
r
e
g
en
er
all
y
,
e
n
s
e
m
b
le
ap
p
r
o
ac
h
es
ten
d
to
ac
h
ie
v
e
b
etter
r
esu
lt
s
i
n
i
m
b
ala
n
ce
d
d
ata
[
4
8
]
–
[
5
1
]
.
T
h
is
tr
en
d
is
r
e
f
lec
ted
in
o
u
r
f
in
d
i
n
g
s
,
w
h
er
e
th
e
e
n
s
e
m
b
le
m
o
d
el
s
h
o
w
s
en
h
a
n
ce
d
d
etec
tio
n
p
er
f
o
r
m
an
ce
e
s
p
ec
ia
ll
y
f
o
r
m
in
o
r
it
y
class
es
s
u
c
h
as
P
VC
an
d
P
AC
.
T
h
er
ef
o
r
e,
f
r
o
m
a
s
cien
ti
f
ic
p
o
in
t
o
f
v
ie
w
,
t
h
e
K
NN
–
L
i
g
h
tGB
M
en
s
e
m
b
le
m
e
th
o
d
w
a
s
c
h
o
s
en
as
th
e
p
r
ef
er
r
ed
m
et
h
o
d
b
ec
au
s
e
th
e
r
esear
ch
o
b
j
ec
tiv
e
em
p
h
a
s
is
ed
th
e
b
est
p
er
f
o
r
m
a
n
ce
in
th
e
P
VC
an
d
P
A
C
class
es.
T
o
s
u
b
s
tan
tia
te
th
e
o
b
s
er
v
ed
p
er
f
o
r
m
a
n
ce
d
if
f
er
en
ce
s
,
Mc
Ne
m
ar
’
s
test
w
as
ap
p
lied
t
o
p
air
ed
p
r
ed
ictio
n
s
g
en
er
ated
f
r
o
m
t
h
e
s
a
m
e
te
s
t
s
et.
T
h
e
r
esu
lts
in
d
icate
th
at
th
e
e
n
s
e
m
b
le
m
o
d
el
s
i
g
n
i
f
ica
n
tl
y
o
u
tp
er
f
o
r
m
s
KNN
(
=
7
.
66
×
10⁻
⁸
)
,
w
h
ile
n
o
s
tatis
ticall
y
s
ig
n
i
f
ica
n
t
d
if
f
er
en
ce
is
o
b
s
er
v
ed
b
etw
ee
n
th
e
en
s
e
m
b
le
m
o
d
el
a
n
d
L
i
g
h
tGB
M
(
=
0
.
1359
)
,
s
u
g
g
esti
n
g
co
m
p
ar
ab
le
p
r
ed
ictiv
e
p
er
f
o
r
m
an
ce
.
I
m
p
o
r
tan
tl
y
,
th
e
ab
s
e
n
ce
o
f
s
ta
tis
tica
l
s
i
g
n
if
ica
n
ce
r
elati
v
e
to
L
i
g
h
tGB
M
in
d
icate
s
p
er
f
o
r
m
an
ce
eq
u
iv
ale
n
ce
r
ath
er
t
h
an
in
f
er
io
r
it
y
.
T
h
er
ef
o
r
e,
th
e
en
s
e
m
b
le
ap
p
r
o
ac
h
r
em
ain
s
p
r
ef
er
ab
le
d
u
e
t
o
its
h
ig
h
er
F1
-
s
co
r
e
f
o
r
th
e
P
VC
an
d
P
A
C
cla
s
s
e
s
,
w
h
ich
ar
e
ce
n
tr
al
to
th
e
o
b
j
ec
tiv
es
o
f
t
h
is
s
t
u
d
y
,
as
w
ell
a
s
it
s
i
n
cr
ea
s
ed
r
o
b
u
s
tn
es
s
ac
h
ie
v
ed
th
r
o
u
g
h
t
h
e
i
n
teg
r
atio
n
o
f
h
ete
r
o
g
en
eo
u
s
class
if
ier
s
[
5
2
]
.
B
ased
o
n
th
e
co
m
p
ar
is
o
n
r
e
s
u
lts
i
n
T
ab
le
2
,
th
is
s
tu
d
y
s
h
o
w
s
b
etter
p
er
f
o
r
m
a
n
ce
co
m
p
ar
e
d
to
m
o
s
t
p
r
ev
io
u
s
s
t
u
d
ies,
esp
ec
iall
y
i
n
ter
m
s
o
f
th
e
av
er
ag
e
F1
-
s
co
r
e,
w
h
ic
h
r
ea
ch
ed
8
8
.
7
2
%,
h
ig
h
er
th
an
Ma
teo
an
d
T
alav
er
a
(
7
1
.
4
2
%)
[
2
3
]
,
Yan
g
et
a
l.
(
8
2
.
4
0
%)
[
2
7
]
,
Gu
s
ev
(
8
5
.
0
1
%)
[
2
4
]
,
Z
ak
ar
ia
et
a
l.
(
8
4
.
9
2
%)
[
2
6
]
,
an
d
Den
g
et
a
l.
(
av
er
ag
e
n
o
t r
ec
o
r
d
ed
)
[
2
5
]
.
Fo
r
th
e
P
VC
class
,
t
h
is
s
tu
d
y
r
ec
o
r
d
ed
a
p
r
ec
is
io
n
o
f
9
0
.
5
9
%,
a
r
ec
all
o
f
9
4
.
4
3
%,
an
d
an
F1
-
s
co
r
e
o
f
9
2
.
4
7
%.
T
h
ese
v
al
u
es
ar
e
r
ela
tiv
el
y
b
alan
ce
d
b
et
w
ee
n
p
r
ec
is
io
n
an
d
r
ec
all,
a
n
d
ar
e
s
u
p
er
io
r
to
Z
ak
ar
ia
et
a
l.
(
7
6
%)
[
2
6
]
an
d
Den
g
et
a
l.
(
7
2
.
9
6
%
)
[
2
5
]
.
I
n
f
ac
t,
its
p
er
f
o
r
m
an
ce
is
co
m
p
ar
ab
le
to
th
at
o
f
Ma
teo
a
n
d
T
alav
er
a
(
9
2
.
3
0
%)
[
2
3
]
an
d
Yan
g
et
a
l.
(
9
2
.
6
0
%)
[
2
7
]
,
b
u
t
w
it
h
an
ad
v
a
n
tag
e
in
t
h
e
b
alan
ce
b
et
w
ee
n
p
r
ec
is
io
n
a
n
d
r
ec
all.
T
h
is
d
e
m
o
n
s
tr
ates
t
h
at
th
e
m
et
h
o
d
e
m
p
lo
y
ed
is
ca
p
ab
le
o
f
r
ed
u
ci
n
g
P
V
C
class
i
f
icatio
n
er
r
o
r
s
w
i
th
o
u
t
co
m
p
r
o
m
is
i
n
g
s
e
n
s
iti
v
it
y
.
Fo
r
th
e
P
A
C
clas
s
,
th
is
s
tu
d
y
p
r
o
d
u
ce
d
a
p
r
ec
is
io
n
o
f
8
9
.
9
1
%,
a
r
ec
all
o
f
8
0
.
5
2
%,
a
n
d
an
F1
-
s
co
r
e
o
f
8
4
.
96
%.
C
o
m
p
ar
ed
to
Ma
teo
et
a
l.
,
w
h
o
a
ch
iev
ed
a
n
F1
-
s
co
r
e
o
f
o
n
l
y
5
0
.
5
8
%,
th
e
i
m
p
r
o
v
e
m
en
t
is
v
er
y
s
i
g
n
i
f
ican
t.
Ho
wev
e
r
,
w
h
e
n
co
m
p
ar
ed
to
Z
ak
ar
ia
et
a
l.
(
F1
-
s
co
r
e
o
f
9
3
.
8
3
%)
[
2
6
]
o
r
Den
g
et
a
l.
(
r
ec
all
o
f
8
9
.
4
5
%)
[
2
5
]
,
th
is
s
t
u
d
y
s
ti
ll
lag
s
s
li
g
h
t
l
y
i
n
s
en
s
it
iv
it
y
,
esp
ec
iall
y
i
n
r
ec
o
g
n
is
i
n
g
P
AC
,
w
h
ic
h
is
r
el
ativ
el
y
f
e
w
er
in
n
u
m
b
er
.
Ne
v
e
r
th
eles
s
,
th
e
p
r
ec
is
io
n
o
f
P
A
C
i
n
t
h
is
s
tu
d
y
is
m
u
c
h
h
ig
h
er
t
h
an
i
n
m
o
s
t
o
th
er
s
t
u
d
ies,
in
d
icat
i
n
g
a
lo
w
er
p
r
ed
ictio
n
er
r
o
r
r
ate
w
h
en
t
h
e
m
o
d
el
class
i
f
ies
P
A
C
.
I
n
ad
d
itio
n
,
th
e
class
if
icatio
n
o
f
P
VC
an
d
P
A
C
i
n
t
h
is
s
tu
d
y
g
e
n
er
all
y
y
ield
s
g
o
o
d
an
d
r
elativ
el
y
b
alan
ce
d
r
esu
lts
,
co
m
p
ar
ed
to
o
th
er
s
tu
d
ie
s
th
at
o
f
ten
s
h
o
w
s
i
g
n
i
f
ican
t d
i
f
f
er
en
ce
s
i
n
p
er
f
o
r
m
a
n
ce
b
et
w
ee
n
P
VC
an
d
P
A
C
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
3
,
J
u
n
e
20
26
:
8
9
1
-
9
03
898
T
h
ese
r
esu
lt
s
i
n
d
icate
t
h
at
t
h
e
f
ea
t
u
r
e
s
elec
tio
n
p
r
o
ce
s
s
e
f
f
ec
tiv
el
y
r
ed
u
ce
d
t
h
e
d
i
m
en
s
io
n
a
lit
y
o
f
t
h
e
f
ea
t
u
r
e
s
p
ac
e,
en
ab
lin
g
t
h
e
m
o
d
el
to
lear
n
m
o
r
e
ef
f
icie
n
tl
y
w
h
ile
f
o
c
u
s
i
n
g
o
n
th
e
m
o
s
t
i
n
f
o
r
m
ati
v
e
v
ar
iab
les.
I
n
itiall
y
,
P
ea
r
s
o
n
co
r
r
elatio
n
f
ilter
in
g
w
a
s
ap
p
lied
w
it
h
a
th
r
es
h
o
ld
o
f
0
.
7
t
o
r
em
o
v
e
s
tr
o
n
g
l
y
r
ed
u
n
d
an
t
f
ea
tu
r
e
s
,
s
i
m
p
li
f
y
i
n
g
t
h
e
f
ea
tu
r
e
s
et
b
e
f
o
r
e
th
e
m
ai
n
s
e
lectio
n
p
r
o
ce
s
s
.
Su
b
s
eq
u
e
n
tl
y
,
R
FE
C
V
w
a
s
p
er
f
o
r
m
ed
,
u
s
i
n
g
cr
o
s
s
-
v
alid
atio
n
at
ea
c
h
iter
at
io
n
to
s
y
s
te
m
atica
ll
y
e
v
alu
a
t
e
f
ea
t
u
r
e
i
m
p
o
r
tan
ce
a
n
d
r
etain
o
n
l
y
t
h
o
s
e
t
h
at
co
n
tr
ib
u
ted
s
ig
n
i
f
ica
n
tl
y
to
class
i
f
icatio
n
p
er
f
o
r
m
a
n
ce
.
T
h
is
s
tag
ed
eli
m
i
n
atio
n
ap
p
r
o
ac
h
n
o
t
o
n
l
y
p
r
ev
en
te
d
o
v
er
f
itti
n
g
ca
u
s
ed
b
y
an
ex
ce
s
s
iv
e
n
u
m
b
er
o
f
f
ea
tu
r
es b
u
t a
ls
o
im
p
r
o
v
ed
m
o
d
el
g
en
er
aliza
ti
o
n
an
d
p
r
o
d
u
ce
d
a
s
i
m
p
ler
y
et
m
ea
n
in
g
f
u
l
r
ep
r
esen
tatio
n
o
f
t
h
e
d
ata
f
o
r
clin
ical
an
al
y
s
i
s
[
5
3
]
.
R
FECV
s
e
lecte
d
s
ev
er
al
i
m
p
o
r
ta
n
t
f
ea
t
u
r
es,
g
r
o
u
p
ed
as
f
o
llo
w
s
:
m
o
r
p
h
o
lo
g
ical
f
ea
t
u
r
es
(
P
r
e_
R
,
L
o
ca
l_
R
,
P
o
s
t_
R
)
;
s
tatis
tic
al
f
ea
tu
r
es
s
u
c
h
a
s
m
ed
ian
a
n
d
k
u
r
to
s
i
s
;
s
p
ec
tr
al
f
ea
t
u
r
es
s
u
c
h
as
m
ed
ian
f
r
eq
u
en
c
y
,
s
p
ec
tr
al
d
is
ta
n
ce
,
a
n
d
s
p
ec
tr
al
.
W
h
ile
th
e
P
ea
r
s
o
n
co
r
r
elatio
n
th
r
es
h
o
l
d
o
f
0
.
7
p
r
o
v
ed
ef
f
ec
ti
v
e,
alter
n
ati
v
e
t
h
r
es
h
o
ld
s
o
r
o
th
er
m
u
ltico
lli
n
ea
r
it
y
d
iag
n
o
s
t
ics co
u
ld
b
e
ex
p
lo
r
ed
in
f
u
t
u
r
e
w
o
r
k
to
f
u
r
th
er
r
ef
i
n
e
f
ea
tu
r
e
q
u
al
it
y
.
T
ab
le
1
.
T
h
e
class
if
icat
io
n
r
es
u
lts
o
f
all
alg
o
r
it
h
m
s
C
l
a
ss
P
r
e
c
i
si
o
n
R
e
c
a
l
l
F1
-
S
c
o
r
e
A
c
c
u
r
a
c
y
K
N
N
N
9
8
.
5
0
%
9
7
.
4
5
%
9
7
.
9
7
%
9
4
.
1
8
%
P
V
C
8
8
.
4
7
%
9
3
.
3
9
%
9
0
.
86
%
P
A
C
8
7
.
2
6
%
8
1
.
1
2
%
8
4
.
0
8
%
L
i
g
h
t
G
B
M
N
9
8
.
1
5
%
9
9
.
8
2
%
9
8
.
9
8
%
9
4
.
8
6
%
P
V
C
8
8
.
2
2
%
9
5
.
3
8
%
9
1
.
6
6
%
P
A
C
9
5
.
1
9
%
7
1
.
4
9
%
8
1
.
6
5
%
En
se
mb
l
e
l
e
a
r
n
i
n
g
N
9
8
.
75
%
9
9
.
06
%
9
8
.
9
0
%
9
5
.
3
4
%
P
V
C
9
0
.
5
9
%
9
4
.
4
3
%
9
2
.
4
7
%
P
A
C
8
9
.
9
1
%
8
0
.
5
2
%
8
4
.
96
%
T
ab
le
2
.
C
o
m
p
ar
is
o
n
w
i
th
o
t
h
er
r
esear
ch
O
t
h
e
r
r
e
se
a
r
c
h
P
V
C
P
A
C
A
v
e
r
a
g
e
o
f
F1
-
s
c
o
r
e
P
r
e
c
i
si
o
n
R
e
c
a
l
l
F1
-
sco
r
e
P
r
e
c
i
si
o
n
R
e
c
a
l
l
F1
-
sco
r
e
M
a
t
e
o
a
n
d
T
a
l
a
v
e
r
a
[
2
3
]
8
9
.
4
0
%
9
5
.
3
8
%
9
2
.
3
0
%
5
1
.
9
5
%
4
9
.
2
6
%
5
0
.
5
8
%
7
1
.
4
2
%
Y
a
n
g
e
t
a
l
.
[
2
7
]
9
2
.
6
0
%
7
2
.
2
0
%
8
2
.
4
0
%
G
u
se
v
[
2
4
]
8
5
.
0
1
%
Z
a
k
a
r
i
a
e
t
a
l
.
[
2
6
]
7
6
%
7
6
%
7
6
%
9
0
%
9
8
%
9
3
.
8
3
%
8
4
.
9
2
%
D
e
n
g
e
t
a
l
.
[
2
5
]
7
2
.
9
6
%
8
9
.
4
5
%
T
h
i
s st
u
d
y
9
0
.
5
9
%
9
4
.
4
3
%
9
2
.
4
7
%
8
9
.
9
1
%
8
0
.
5
2
%
8
4
.
96
%
8
8
.
7
2
%
3
.
5
.
SH
AP
-
f
ea
t
ures
inte
rpre
t
a
t
io
n
T
h
e
r
esu
lts
o
f
i
n
ter
p
r
etin
g
th
e
Vo
ti
n
g
C
la
s
s
i
f
ier
(
KN
N+
L
i
g
h
tGB
M)
m
o
d
el
u
s
i
n
g
SH
A
P
Ker
n
elE
x
p
lai
n
er
s
h
o
w
t
h
at
th
e
R
in
ter
v
al,
s
p
ec
tr
al,
an
d
s
tati
s
tical
f
ea
t
u
r
es
p
la
y
a
cr
u
cial
r
o
le
in
d
is
tin
g
u
is
h
i
n
g
b
et
w
ee
n
t
h
e
n
o
r
m
al
,
P
A
C
,
a
n
d
P
VC
cla
s
s
es.
T
h
e
m
ai
n
i
n
f
lu
e
n
tial
f
ea
tu
r
es
ar
e
P
r
e_
R
,
Me
d
ian
f
r
eq
u
e
n
c
y
,
Sp
ec
tr
al
d
is
tan
ce
,
L
o
ca
l_
R
,
m
ed
ian
,
a
n
d
k
u
r
to
s
i
s
.
T
h
e
S
HA
P
Ker
n
e
l
m
eth
o
d
o
f
f
er
s
h
ig
h
f
le
x
ib
ilit
y
i
n
ex
p
lain
i
n
g
en
s
e
m
b
le
m
o
d
els;
h
o
w
ev
er
,
th
e
r
esu
l
ts
ca
n
b
e
u
n
s
tab
le
d
u
e
to
its
r
elian
ce
o
n
s
to
ch
ast
ic
s
a
m
p
li
n
g
,
w
h
ic
h
m
a
y
y
ield
d
if
f
er
e
n
t
attr
ib
u
tio
n
s
ac
r
o
s
s
r
u
n
s
.
R
ec
en
t
s
t
u
d
ies
h
a
v
e
s
h
o
w
n
t
h
at
Ker
n
el
SH
A
P
’
s
in
s
tab
ilit
y
s
te
m
s
f
r
o
m
i
ts
r
an
d
o
m
n
eig
h
b
o
r
s
elec
tio
n
p
r
o
ce
d
u
r
e
[
5
4
]
.
I
n
f
u
t
u
r
e
w
o
r
k
,
r
ep
ea
ted
SH
A
P
an
al
y
s
es
o
r
co
r
r
o
b
o
r
atio
n
u
s
i
n
g
alter
n
a
tiv
e
ex
p
lai
n
ab
ilit
y
m
eth
o
d
s
s
u
ch
as
L
I
ME
an
d
p
er
m
u
tat
io
n
f
ea
t
u
r
e
i
m
p
o
r
tan
ce
co
u
ld
b
e
em
p
lo
y
ed
to
f
u
r
t
h
er
ass
es
s
th
e
r
o
b
u
s
t
n
es
s
o
f
f
ea
tu
r
e
attr
ib
u
tio
n
s
[
5
5
]
,
[
5
6
]
.
Fig
u
r
es
6
to
8
s
h
o
w
th
e
l
is
t
o
f
to
p
f
ea
t
u
r
es
f
o
r
ea
ch
cla
s
s
.
I
n
th
e
n
o
r
m
al
c
lass
,
h
i
g
h
P
r
e_
R
v
al
u
es
al
m
o
s
t
al
w
a
y
s
i
n
cr
ea
s
e
th
e
p
r
o
b
ab
ilit
y
o
f
a
n
o
r
m
a
l
p
r
ed
ictio
n
.
T
h
is
ef
f
ec
t
is
r
ei
n
f
o
r
ce
d
b
y
s
tati
s
tical
f
ea
tu
r
es
s
u
c
h
as
m
ed
ian
a
n
d
k
u
r
to
s
is
t
h
e
h
i
g
h
er
th
e
s
e
v
al
u
es,
t
h
e
g
r
ea
ter
th
e
s
u
p
p
o
r
t
f
o
r
n
o
r
m
a
l
c
lass
i
f
icatio
n
.
I
n
t
h
e
P
A
C
cla
s
s
,
P
r
e_
R
,
Me
d
ian
f
r
eq
u
en
c
y
,
an
d
Sp
ec
tr
al
d
is
ta
n
ce
p
lay
a
m
aj
o
r
r
o
le.
L
o
w
v
al
u
es
o
f
P
r
e_
R
,
Me
d
ian
f
r
eq
u
en
c
y
,
Sp
ec
tr
al
d
i
s
tan
ce
,
an
d
L
o
ca
l_
R
g
e
n
er
all
y
i
n
cr
ea
s
e
th
e
P
AC
p
r
ed
ictio
n
,
as
s
h
o
w
n
b
y
b
l
u
e
d
o
ts
o
n
th
e
r
ig
h
t
s
id
e
(
p
o
s
itiv
e
SH
A
P
v
alu
e
s
)
.
T
h
e
r
ed
d
o
ts
o
n
th
e
r
i
g
h
t
al
s
o
in
d
icate
t
h
at,
u
n
d
er
ce
r
tain
co
n
d
itio
n
s
,
h
ig
h
v
al
u
es
o
f
t
h
ese
f
ea
t
u
r
es
ca
n
s
t
ill
i
n
cr
ea
s
e
t
h
e
p
r
o
b
ab
ilit
y
o
f
P
AC
,
s
u
g
g
e
s
ti
n
g
f
ea
t
u
r
e
in
ter
ac
tio
n
s
.
T
h
u
s
,
P
A
C
ca
n
b
e
tr
i
g
g
er
ed
b
y
lo
w
-
v
alu
e
f
ea
tu
r
es
o
r
f
r
o
m
a
co
m
b
in
atio
n
o
f
h
i
g
h
f
ea
t
u
r
e
v
alu
e
s
w
it
h
o
t
h
er
f
ac
to
r
s
.
I
n
th
e
P
VC
clas
s
,
a
s
i
m
i
lar
p
atter
n
o
cc
u
r
s
:
P
r
e_
R
r
em
ai
n
s
d
o
m
i
n
an
t
;
Sp
ec
tr
al
d
is
tan
ce
a
n
d
Me
d
ian
f
r
eq
u
e
n
c
y
f
o
llo
w
a
s
k
e
y
f
ea
t
u
r
es
in
d
ete
ctin
g
p
r
em
a
tu
r
e
v
e
n
tr
ic
u
lar
b
ea
ts
.
T
h
e
b
lu
e
an
d
r
ed
d
o
ts
f
o
r
P
r
e_
R
o
n
th
e
r
ig
h
t
s
id
e
in
d
icate
t
h
at
b
o
th
h
i
g
h
a
n
d
lo
w
v
al
u
es
ca
n
in
cr
ea
s
e
P
VC
p
r
ed
ictio
n
,
d
ep
en
d
in
g
o
n
th
e
co
n
te
x
t
o
f
o
t
h
er
f
ea
t
u
r
es.
Fo
r
Sp
ec
tr
al
d
i
s
tan
ce
an
d
Me
d
ia
n
f
r
eq
u
en
c
y
,
h
i
g
h
v
alu
e
s
co
n
s
i
s
ten
tl
y
in
cr
ea
s
e
P
VC
p
r
o
b
ab
ilit
y
.
Fo
r
A
b
s
o
l
u
te
en
er
g
y
,
lo
w
v
al
u
es d
r
iv
e
th
e
p
r
ed
ictio
n
to
w
ar
d
th
e
P
VC
c
lass
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
C
la
s
s
i
fica
tio
n
o
f p
r
ema
tu
r
e
ca
r
d
ia
c
co
n
tr
a
ctio
n
s
b
a
s
ed
o
n
R
F
E
C
V
…
(
E
ls
a
S
a
r
i H
a
yu
n
a
h
N
u
r
d
in
iya
h
)
899
Fig
u
r
e
9
s
h
o
w
s
th
e
li
s
t
o
f
to
p
f
ea
t
u
r
es
b
ased
o
n
SHA
P
v
al
u
es.
T
h
e
to
p
f
iv
e
f
ea
t
u
r
es
P
r
e_
R
,
Me
d
ian
f
r
eq
u
en
c
y
,
Sp
ec
tr
al
d
is
tan
ce
,
L
o
ca
l_
R
,
an
d
A
b
s
o
lu
te
en
er
g
y
w
er
e
s
elec
ted
b
y
t
h
e
m
o
d
el.
T
h
ey
ar
e
p
h
y
s
io
lo
g
icall
y
li
n
k
ed
to
p
r
em
at
u
r
e
b
ea
ts
(
b
o
th
P
A
C
an
d
P
VC
)
an
d
n
o
r
m
al
r
h
y
t
h
m
.
P
r
e_
R
r
ep
r
esen
ts
th
e
in
ter
v
a
l
o
r
ch
ar
ac
ter
is
tic
s
b
ef
o
r
e
th
e
R
w
a
v
e.
P
r
e
m
atu
r
e
b
ea
ts
t
y
p
icall
y
a
f
f
ec
t
th
e
R
-
to
-
R
i
n
ter
v
a
l
o
r
th
e
w
a
v
e
f
o
r
m
ar
o
u
n
d
R
,
m
a
k
i
n
g
f
ea
tu
r
es
th
a
t
m
ea
s
u
r
e
th
e
i
n
ter
v
al
b
ef
o
r
e
o
r
af
ter
R
h
ig
h
l
y
i
n
f
o
r
m
ati
v
e
[
3
4
]
.
T
h
e
Med
ian
f
r
eq
u
e
n
c
y
an
d
Sp
ec
tr
a
l
d
is
tan
ce
f
ea
t
u
r
es
d
escr
ib
e
ch
an
g
e
s
i
n
f
r
eq
u
e
n
c
y
co
n
te
n
t
w
h
en
th
e
s
i
g
n
al
s
h
o
w
s
ab
n
o
r
m
al
f
l
u
ctu
a
tio
n
s
,
s
u
ch
as
p
r
em
a
tu
r
e
b
ea
t
s
,
n
o
i
s
e,
o
r
R
-
R
i
n
ter
v
al
v
ar
iatio
n
s
.
L
o
ca
l_
R
lik
el
y
ca
p
t
u
r
es
lo
ca
l
ch
an
g
es
i
n
th
e
s
i
g
n
al
ar
o
u
n
d
th
e
R
w
a
v
e.
T
h
ese
in
clu
d
e
ab
n
o
r
m
al
R
-
R
in
ter
v
al
ch
a
n
g
e
s
o
r
u
n
u
s
u
al
s
h
ap
e
s
b
ef
o
r
e
o
r
af
ter
R
,
w
h
ic
h
s
i
g
n
al
p
r
em
at
u
r
e
b
ea
ts
.
T
h
e
A
b
s
o
lu
t
e
en
er
g
y
f
ea
tu
r
e
r
ef
lects
t
h
e
to
tal
‘
s
tr
en
g
th
’
o
f
th
e
s
ig
n
al.
P
r
e
m
at
u
r
e
b
ea
ts
o
f
te
n
ca
u
s
e
e
n
er
g
y
s
u
r
g
e
s
o
r
a
m
p
l
itu
d
e
f
lu
ct
u
atio
n
s
,
t
h
er
eb
y
i
n
c
r
ea
s
in
g
th
e
o
v
er
all
s
ig
n
al
e
n
er
g
y
r
ela
tiv
e
to
n
o
r
m
al
s
ig
n
als
[
5
7
]
,
[
5
8
]
.
I
m
p
o
r
tan
tl
y
,
t
h
ese
f
ea
tu
r
e
-
le
v
el
in
s
i
g
h
t
s
ca
n
as
s
is
t
cli
n
ician
s
b
y
h
i
g
h
l
ig
h
ti
n
g
w
h
ic
h
s
i
g
n
a
l
ch
ar
ac
ter
is
tic
s
ar
e
m
o
s
t
in
f
o
r
m
ati
v
e
f
o
r
d
is
tin
g
u
i
s
h
in
g
p
r
em
at
u
r
e
b
ea
ts
,
p
o
ten
tiall
y
s
u
p
p
o
r
tin
g
m
o
r
e
f
o
cu
s
ed
E
C
G
i
n
ter
p
r
etatio
n
a
n
d
m
o
n
i
t
o
r
in
g
s
tr
ateg
ie
s
.
Fro
m
a
s
y
s
te
m
d
ev
elo
p
m
en
t
p
er
s
p
ec
ti
v
e,
t
h
e
id
en
t
if
ica
tio
n
o
f
th
ese
k
e
y
f
ea
tu
r
es
m
a
y
al
s
o
g
u
id
e
f
u
tu
r
e
d
ata
co
llectio
n
an
d
m
o
d
el
r
ef
i
n
e
m
en
t
b
y
p
r
io
r
itizin
g
p
h
y
s
io
lo
g
icall
y
r
elev
an
t
s
i
g
n
al
ch
ar
ac
ter
is
tic
s
.
Nev
er
th
eles
s
,
th
e
id
en
ti
f
i
ed
f
ea
tu
r
es
o
r
ig
i
n
ate
f
r
o
m
th
e
s
p
ec
if
ic
m
o
d
el
an
d
d
ataset
u
s
ed
in
t
h
i
s
s
t
u
d
y
an
d
d
o
n
o
t
c
o
n
s
tit
u
te
cli
n
ical
c
o
n
f
ir
m
atio
n
.
B
r
o
ad
er
clin
ical
d
ata,
in
ter
-
p
atien
t
ev
alu
a
tio
n
,
a
n
d
ex
ter
n
al
v
alid
a
tio
n
ar
e
r
eq
u
ir
ed
b
ef
o
r
e
th
e
s
e
i
n
s
i
g
h
ts
ca
n
b
e
r
eliab
l
y
ap
p
lied
in
m
ed
ical
p
r
ac
tice.
Fig
u
r
e
6
.
SHA
P
f
o
r
n
o
r
m
al
Fig
u
r
e
7
.
SHA
P
f
o
r
P
AC
Fig
u
r
e
8
.
SHA
P
f
o
r
P
VC
Fig
u
r
e
9
.
SHA
P
s
u
m
m
ar
y
p
lo
t
-
to
p
f
ea
t
u
r
es
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
3
,
J
u
n
e
20
26
:
8
9
1
-
9
03
900
4.
CO
NCLU
SI
O
N
T
h
is
s
t
u
d
y
p
r
o
p
o
s
es
a
r
o
b
u
s
t
f
r
a
m
e
w
o
r
k
f
o
r
class
if
y
i
n
g
p
r
e
m
atu
r
e
ca
r
d
iac
co
n
tr
ac
tio
n
s
u
s
i
n
g
R
FECV
-
b
ased
f
ea
tu
r
e
s
e
lectio
n
a
n
d
a
n
en
s
e
m
b
le
o
f
KNN
a
n
d
L
ig
h
tGB
M.
T
h
e
m
o
d
el
ac
h
ie
v
es
h
ig
h
a
n
d
b
alan
ce
d
p
er
f
o
r
m
a
n
ce
ac
r
o
s
s
n
o
r
m
al,
P
VC
,
an
d
P
A
C
cla
s
s
e
s
,
in
cl
u
d
in
g
u
n
d
er
in
ter
-
p
atie
n
t
v
a
lid
atio
n
,
d
e
m
o
n
s
tr
atin
g
it
s
r
eliab
ilit
y
an
d
co
m
p
etit
iv
e
n
es
s
w
ith
e
x
is
t
in
g
ap
p
r
o
ac
h
es.
SHA
P
-
b
a
s
ed
in
ter
p
r
etab
ilit
y
a
n
al
y
s
i
s
co
n
f
ir
m
s
th
a
t
f
ea
t
u
r
es
s
u
c
h
as
P
r
e_
R
,
m
ed
ian
f
r
eq
u
e
n
c
y
,
s
p
ec
tr
al
d
is
tan
ce
,
L
o
ca
l_
R
,
an
d
ab
s
o
lu
te
en
er
g
y
ar
e
th
e
m
o
s
t
in
f
o
r
m
ati
v
e,
s
u
p
p
o
r
tin
g
p
h
y
s
io
lo
g
ical
r
elev
a
n
ce
an
d
clin
ical
i
n
ter
p
r
etab
ilit
y
.
B
ey
o
n
d
p
er
f
o
r
m
an
c
e
i
m
p
r
o
v
e
m
en
t
s
,
th
is
f
r
a
m
e
w
o
r
k
p
r
o
v
id
es
a
f
o
u
n
d
atio
n
f
o
r
d
ev
elo
p
in
g
r
eliab
le
a
n
d
ex
p
l
ain
ab
le
ar
r
h
y
th
m
ia
d
etec
tio
n
s
y
s
te
m
s
th
at
ca
n
s
u
p
p
o
r
t
w
ea
r
ab
le
m
o
n
ito
r
i
n
g
an
d
clin
ical
d
ec
is
io
n
-
s
u
p
p
o
r
t
ap
p
li
ca
tio
n
s
.
Fu
tu
r
e
w
o
r
k
w
il
l
f
o
cu
s
o
n
v
alid
atio
n
u
s
i
n
g
lar
g
er
an
d
m
o
r
e
d
iv
er
s
e
p
atien
t
p
o
p
u
lat
io
n
s
,
in
te
g
r
atio
n
w
it
h
d
ee
p
lear
n
in
g
ap
p
r
o
ac
h
es,
an
d
ev
alu
atio
n
o
f
r
ea
l
-
ti
m
e
i
m
p
le
m
e
n
tatio
n
f
o
r
p
r
ac
tical
clin
ical
d
ep
lo
y
m
e
n
t.
ACK
NO
WL
E
D
G
M
E
NT
S
T
h
is
r
esear
ch
w
as
s
u
p
p
o
r
ted
b
y
L
P
P
M
Un
iv
er
s
itas
J
en
d
e
r
al
So
ed
ir
m
a
n
t
h
r
o
u
g
h
UN
S
OE
D
B
L
U
R
esear
ch
Gr
a
n
t 2
0
2
5
.
F
UNDIN
G
I
NF
O
RM
AT
I
O
N
T
h
e
au
th
o
r
s
w
o
u
ld
li
k
e
to
th
an
k
L
P
P
M
Un
iv
er
s
itas
J
en
d
e
r
al
So
ed
ir
m
a
n
f
o
r
f
u
n
d
in
g
th
i
s
r
esear
ch
th
r
o
u
g
h
t
h
e
UN
S
OE
D
B
L
U
R
esear
ch
Gr
an
t 2
0
2
5
.
AUTHO
R
CO
NT
RIB
UT
I
O
NS ST
A
T
E
M
E
NT
T
h
is
j
o
u
r
n
al
u
s
e
s
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT
)
to
r
ec
o
g
n
ize
in
d
i
v
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
t
h
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
lla
b
o
r
atio
n
.
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
E
ls
a
Sar
i H
a
y
u
n
ah
Nu
r
d
in
i
y
a
h
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
A
’
i
s
y
a
N
u
r
A
u
l
ia
Yu
s
u
f
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
No
r
m
a
Am
al
ia
✓
✓
✓
✓
✓
✓
W
id
h
iat
m
o
k
o
Her
r
y
P
u
r
n
o
m
o
✓
✓
✓
✓
✓
A
ziza
h
Naj
d
a
Haf
iz
h
a
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
si
s
I
:
I
n
v
e
st
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
si
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
T
h
er
e
is
n
o
co
n
f
lict o
f
in
ter
e
s
t
in
th
i
s
r
esear
ch
.
I
NF
O
RM
E
D
CO
NSE
N
T
W
e
h
av
e
o
b
tain
ed
in
f
o
r
m
ed
c
o
n
s
en
t f
r
o
m
al
l in
d
i
v
id
u
al
s
in
c
lu
d
ed
in
th
is
s
t
u
d
y
.
E
T
H
I
CAL AP
P
RO
V
AL
T
h
e
s
tu
d
y
u
tili
ze
d
o
p
en
ac
ce
s
s
d
atab
ase
(
h
ttp
s
://p
h
y
s
io
n
et.
o
r
g
/co
n
te
n
t/
m
itd
b
/1
.
0
.
0
/)
an
d
th
er
ef
o
r
e
d
id
n
o
t r
eq
u
ir
e
eth
ical
ap
p
r
o
v
al,
as it c
o
m
p
lied
w
it
h
all
r
elev
a
n
t
n
atio
n
al
r
e
g
u
latio
n
s
an
d
in
s
tit
u
tio
n
al
p
o
licie
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.