I
AE
S In
t
er
na
t
io
na
l J
o
urna
l o
f
Art
if
icia
l In
t
ellig
ence
(
I
J
-
AI
)
Vo
l.
15
,
No
.
1
,
Feb
r
u
ar
y
2
0
2
6
,
p
p
.
469
~
480
I
SS
N:
2
2
5
2
-
8
9
3
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijai.v
15
.i
1
.
p
p
4
6
9
-
4
8
0
469
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
Cla
ss
ifying
menta
l worklo
a
d of esp
o
rts play
ers using
ma
chine
lea
rning
Ais
y
Al
F
a
w
wa
z
1
,
O
s
ma
lin
a
Nur
Ra
h
ma
1
,
2
,
S
a
y
y
id
ul I
s
t
i
g
hfa
r
I
t
t
a
qil
l
a
h
1,
3
,
An
g
e
lin
e
S
ha
ne
K
urni
a
w
a
n
1
,
Rev
it
a
No
v
ia
nti
P
utr
i
1
,
Rich
a
Va
ry
a
n
1
,
Aura
Adin
da
3
,
K
hu
s
n
ul Ai
n
1,
2
,
Rif
a
i C
ha
i
4
1
B
i
o
me
d
i
c
a
l
E
n
g
i
n
e
e
r
i
n
g
S
t
u
d
y
P
r
o
g
r
a
m
,
F
a
c
u
l
t
y
o
f
S
c
i
e
n
c
e
a
n
d
Te
c
h
n
o
l
o
g
y
,
U
n
i
v
e
r
si
t
a
s
A
i
r
l
a
n
g
g
a
,
S
u
r
a
b
a
y
a
,
I
n
d
o
n
e
si
a
2
B
i
o
me
d
i
c
a
l
E
n
g
i
n
e
e
r
i
n
g
I
n
n
o
v
a
t
i
o
n
R
e
se
a
r
c
h
G
r
o
u
p
,
F
a
c
u
l
t
y
o
f
S
c
i
e
n
c
e
a
n
d
Te
c
h
n
o
l
o
g
y
,
U
n
i
v
e
r
si
t
a
s
A
i
r
l
a
n
g
g
a
,
S
u
r
a
b
a
y
a
,
I
n
d
o
n
e
s
i
a
3
M
a
s
t
e
r
o
f
B
i
o
me
d
i
c
a
l
E
n
g
i
n
e
e
r
i
n
g
,
F
a
c
u
l
t
y
o
f
S
c
i
e
n
c
e
a
n
d
Te
c
h
n
o
l
o
g
y
,
U
n
i
v
e
r
si
t
a
s Air
l
a
n
g
g
a
,
S
u
r
a
b
a
y
a
,
I
n
d
o
n
e
si
a
4
D
e
p
a
r
t
me
n
t
o
f
E
n
g
i
n
e
e
r
i
n
g
T
e
c
h
n
o
l
o
g
i
e
s,
S
c
h
o
o
l
o
f
S
c
i
e
n
c
e
,
C
o
m
p
u
t
i
n
g
a
n
d
E
n
g
i
n
e
e
r
i
n
g
T
e
c
h
n
o
l
o
g
i
e
s,
S
w
i
n
b
u
r
n
e
U
n
i
v
e
r
si
t
y
o
f
T
e
c
h
n
o
l
o
g
y
,
M
e
l
b
o
u
r
n
e
,
A
u
s
t
r
a
l
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
u
n
3
,
2024
R
ev
is
ed
Au
g
6
,
2
0
2
5
Acc
ep
ted
Oct
18
,
2
0
2
5
El
e
c
tro
d
e
rm
a
l
a
c
ti
v
it
y
(EDA)
p
e
a
k
c
o
u
n
ts,
d
e
ri
v
e
d
fro
m
b
o
t
h
to
n
ic
a
n
d
p
h
a
sic
c
o
m
p
o
n
e
n
ts,
a
re
wid
e
ly
u
se
d
a
s
p
h
y
sio
lo
g
ica
l
p
r
o
x
ies
f
o
r
m
e
n
tal
wo
rk
l
o
a
d
in
c
o
g
n
it
i
v
e
ly
d
e
m
a
n
d
in
g
tas
k
s,
su
c
h
a
s
e
sp
o
rts.
Ho
w
e
v
e
r,
t
h
e
ir
sp
e
c
ifi
c
it
y
re
m
a
in
s
u
n
c
e
rtain
,
p
a
r
ti
c
u
larly
g
i
v
e
n
p
o
ten
ti
a
l
c
o
n
f
o
u
n
d
in
g
e
ffe
c
t
o
f
ti
m
e
-
on
-
tas
k
.
Th
is
st
u
d
y
a
n
a
ly
z
e
s
9
2
c
o
m
p
e
ti
ti
v
e
g
a
m
e
p
lay
se
ss
io
n
s
fro
m
a
m
u
lt
imo
d
a
l
e
sp
o
r
ts
d
a
tas
e
t
u
si
n
g
th
re
e
d
e
c
o
m
p
o
siti
o
n
tec
h
n
iq
u
e
s:
c
o
n
v
e
x
d
e
c
o
m
p
o
siti
o
n
(
c
v
x
EDA
)
,
sp
a
rs
e
d
e
c
o
n
v
o
l
u
ti
o
n
(s
p
a
rse
EDA
)
,
a
n
d
ti
m
e
-
v
a
r
y
in
g
s
y
m
p
a
th
e
ti
c
a
c
ti
v
i
ty
(
T
VSy
m
p
)
.
F
ro
m
e
a
c
h
m
e
th
o
d
,
p
h
a
sic
,
a
n
d
to
n
ic
p
e
a
k
c
o
u
n
ts
(TP
C)
,
a
s
we
ll
a
s
th
e
ir
n
o
rm
a
li
z
e
d
ra
tes
,
we
re
e
x
trac
ted
.
We
e
x
a
m
in
e
d
t
h
e
ir
re
lati
o
n
s
h
i
p
wit
h
se
lf
-
re
p
o
rted
w
o
rk
l
o
a
d
th
r
o
u
g
h
c
o
rre
latio
n
a
n
a
ly
se
s,
p
a
rti
a
l
c
o
rre
latio
n
s
c
o
n
tro
ll
in
g
f
o
r
se
ss
io
n
d
u
r
a
ti
o
n
,
a
n
d
li
n
e
a
r
m
ix
e
d
-
e
ffe
c
ts
m
o
d
e
l
s
(L
M
M
s)
.
Wh
il
e
b
o
th
p
e
a
k
ty
p
e
s
e
x
h
ib
it
e
d
stro
n
g
p
o
siti
v
e
c
o
rre
latio
n
s
with
g
a
m
e
p
lay
d
u
ra
ti
o
n
(r=
0
.
9
1
5
f
o
r
p
h
a
sic
a
n
d
r=
0
.
8
5
6
f
o
r
to
n
ic),
t
h
e
ir
a
ss
o
c
iati
o
n
wi
th
p
e
rc
e
iv
e
d
wo
rk
lo
a
d
v
a
n
i
sh
e
d
o
n
c
e
ti
m
e
wa
s
a
c
c
o
u
n
ted
fo
r.
Ac
ro
s
s
m
e
th
o
d
s,
TVS
y
m
p
y
ield
e
d
th
e
h
ig
h
e
st
d
isc
rimin
a
ti
v
e
v
a
li
d
it
y
wit
h
a
n
a
re
a
u
n
d
e
r
c
u
rv
e
(
AUC
)
o
f
0
.
8
8
0
in
c
las
sify
in
g
h
ig
h
v
e
rsu
s
lo
w
wo
r
k
lo
a
d
.
M
a
c
h
in
e
lea
rn
i
n
g
(M
L)
c
las
sifiers
train
e
d
so
lel
y
o
n
EDA
-
b
a
se
d
fe
a
t
u
re
s
u
n
d
e
r
a
lea
v
e
-
one
-
su
b
jec
t
-
o
u
t
(LOS
O)
sc
h
e
m
e
o
u
tp
e
rf
o
rm
e
d
m
u
l
ti
m
o
d
a
l
m
o
d
e
ls
t
h
a
t
i
n
c
o
r
p
o
ra
ted
h
e
a
rt
ra
te
v
a
riab
il
it
y
(HRV
)
.
Th
e
se
re
su
lt
s
u
n
d
e
rsc
o
re
n
e
e
d
to
d
ise
n
tan
g
le
tem
p
o
ra
l
stru
c
tu
re
fro
m
c
o
g
n
it
i
v
e
sig
n
a
l
s
wh
e
n
i
n
terp
re
ti
n
g
EDA
a
n
d
c
a
ll
in
t
o
q
u
e
stio
n
t
h
e
a
ss
u
m
p
ti
o
n
t
h
a
t
EDA
p
e
a
k
c
o
u
n
ts
a
l
o
n
e
re
li
a
b
ly
e
n
c
o
d
e
m
e
n
tal
wo
rk
l
o
a
d
a
c
ro
ss
in
d
i
v
id
u
a
ls.
K
ey
w
o
r
d
s
:
E
DA
d
ec
o
m
p
o
s
itio
n
E
lectr
o
d
er
m
al
ac
tiv
ity
Me
n
tal
wo
r
k
lo
ad
Ph
y
s
io
lo
g
ical
s
ig
n
al
an
aly
s
is
T
em
p
o
r
al
c
o
n
f
o
u
n
d
s
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
:
Osma
lin
a
Nu
r
R
ah
m
a
B
io
m
ed
ical
E
n
g
in
ee
r
in
g
Stu
d
y
Pro
g
r
am
,
Facu
lty
o
f
Scien
ce
an
d
T
ec
h
n
o
lo
g
y
,
Un
i
v
er
s
itas
Air
lan
g
g
a
Su
r
ab
ay
a,
I
n
d
o
n
esia
E
m
ail: o
s
m
alin
a.
n
.
r
ah
m
a@
f
s
t.u
n
air
.
ac
.
id
1.
I
NT
RO
D
UCT
I
O
N
Ov
er
th
e
p
ast
d
ec
ad
e,
t
h
e
ex
p
o
n
e
n
tial
g
r
o
wth
o
f
co
m
p
etitiv
e
elec
tr
o
n
ic
s
p
o
r
ts
(
esp
o
r
ts
)
h
as
s
ig
n
if
ican
tly
r
esh
ap
ed
th
e
la
n
d
s
ca
p
e
o
f
d
ig
ital
in
ter
ac
tio
n
an
d
h
u
m
an
p
er
f
o
r
m
an
ce
[
1
]
–
[
3
]
.
E
s
p
o
r
ts
is
n
o
lo
n
g
er
a
ca
s
u
al
p
asti
m
e.
I
t
n
o
w
d
em
an
d
s
s
u
s
tain
ed
co
g
n
itiv
e
f
o
cu
s
,
em
o
tio
n
al
r
eg
u
latio
n
,
an
d
r
ap
id
d
ec
is
io
n
-
m
ak
in
g
u
n
d
e
r
p
r
ess
u
r
e
[
4
]
,
[
5
]
.
T
h
ese
c
o
g
n
itiv
e
a
n
d
af
f
ec
tiv
e
d
em
a
n
d
s
in
c
r
ea
s
in
g
ly
r
esem
b
le
t
h
o
s
e
en
co
u
n
ter
e
d
in
h
ig
h
-
s
tak
es d
o
m
ain
s
s
u
ch
as a
v
iatio
n
,
m
ilit
ar
y
o
p
er
atio
n
s
,
an
d
em
er
g
e
n
cy
r
esp
o
n
s
e
[
6
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
1
,
Feb
r
u
ar
y
20
26
:
469
-
4
8
0
470
I
n
r
esp
o
n
s
e
to
th
is
co
n
v
er
g
e
n
ce
,
r
esear
ch
in
af
f
ec
tiv
e
co
m
p
u
tin
g
h
as
s
h
if
ted
its
f
o
c
u
s
to
war
d
p
h
y
s
io
lo
g
ical
s
ig
n
als
as
a
m
ea
n
s
to
m
o
n
ito
r
m
e
n
tal
s
tates
i
n
r
ea
l
-
tim
e
[
7
]
–
[
9
]
.
E
lectr
o
d
e
r
m
al
ac
tiv
ity
(
E
DA)
h
as e
m
er
g
ed
as o
n
e
o
f
th
e
m
o
s
t stu
d
ied
s
ig
n
als d
u
e
to
its
s
e
n
s
itiv
ity
to
s
y
m
p
ath
etic
ar
o
u
s
a
l a
n
d
its
p
o
ten
tial to
tr
ac
k
s
u
b
tle
ch
an
g
es in
co
g
n
iti
v
e
ef
f
o
r
t
d
u
r
in
g
n
atu
r
alis
tic
task
s
[
1
0
]
.
I
n
a
p
r
ev
io
u
s
s
tu
d
y
i
n
v
o
lv
in
g
9
6
p
lay
er
s
ac
r
o
s
s
2
1
esp
o
r
ts
m
atch
es,
we
ev
alu
ated
th
e
p
r
e
d
ictiv
e
p
o
te
n
tial
o
f
f
ea
tu
r
es
d
er
iv
e
d
f
r
o
m
E
DA
an
d
h
ea
r
t
r
ate
v
ar
iab
ilit
y
(
HR
V)
in
esti
m
atin
g
p
er
ce
iv
ed
wo
r
k
lo
ad
u
s
in
g
m
ac
h
in
e
lear
n
in
g
(
ML
)
m
o
d
e
ls
[
1
1
]
.
Am
o
n
g
th
e
ex
tr
ac
ted
f
ea
tu
r
es,
to
n
ic
p
ea
k
co
u
n
t
(
T
PC
)
an
d
p
h
asic
p
ea
k
co
u
n
ts
(
PP
C
)
wer
e
o
b
tain
ed
th
r
o
u
g
h
co
n
v
ex
d
ec
o
m
p
o
s
itio
n
(
cv
x
E
DA)
a
n
d
co
n
s
is
ten
tly
o
u
tp
er
f
o
r
m
ed
o
t
h
er
in
d
icato
r
s
,
y
iel
d
in
g
ar
ea
u
n
d
er
c
u
r
v
e
(
AUC)
s
co
r
es
ab
o
v
e
0
.
8
8
.
T
h
ese
p
ea
k
-
b
ased
f
ea
tu
r
es
p
r
o
v
id
ed
a
c
o
m
p
ac
t
an
d
in
ter
p
r
etab
le
s
u
m
m
ar
y
o
f
au
to
n
o
m
ic
ac
tiv
ity
,
wh
ich
m
ad
e
th
em
p
r
o
m
is
in
g
f
o
r
a
p
p
licatio
n
s
in
ad
a
p
tiv
e
s
y
s
tem
s
an
d
r
ea
l
-
tim
e
d
e
cisi
o
n
s
u
p
p
o
r
t.
Ho
wev
er
,
t
h
e
v
alid
ity
o
f
E
DA
p
ea
k
s
as
in
d
icato
r
s
o
f
co
g
n
iti
v
e
d
em
an
d
r
em
ain
s
u
n
r
eso
lv
e
d
.
L
o
n
g
e
r
g
am
ep
lay
s
ess
io
n
s
n
atu
r
ally
ac
cu
m
u
late
m
o
r
e
E
DA
p
ea
k
s
,
r
aisi
n
g
th
e
p
o
s
s
ib
ilit
y
th
at
p
ea
k
-
b
ased
m
o
d
els m
ay
co
n
f
late
task
d
u
r
atio
n
with
m
e
n
tal
ef
f
o
r
t.
W
ith
o
u
t p
r
o
p
e
r
ly
d
is
en
tan
g
lin
g
tem
p
o
r
al
co
n
f
o
u
n
d
s
,
class
if
ier
s
m
ay
o
v
er
f
it to
tim
e
-
on
-
task
e
f
f
ec
ts
an
d
m
is
in
ter
p
r
et
a
r
o
u
s
al
d
r
i
v
e
n
b
y
e
x
p
o
s
u
r
e
as wo
r
k
lo
a
d
-
in
d
u
ce
d
.
T
h
e
p
r
esen
t
s
tu
d
y
a
d
d
r
ess
es
th
is
am
b
ig
u
ity
t
h
r
o
u
g
h
a
co
m
p
r
eh
en
s
iv
e
tem
p
o
r
al
an
aly
s
is
o
f
E
DA
s
ig
n
als
r
ec
o
r
d
ed
d
u
r
i
n
g
9
2
co
m
p
etitiv
e
g
am
ep
lay
s
ess
io
n
s
.
Usi
n
g
th
r
ee
d
ec
o
m
p
o
s
itio
n
te
ch
n
iq
u
es:
cv
x
E
DA,
s
p
ar
s
e
d
ec
o
n
v
o
lu
tio
n
(
s
p
a
r
s
eE
DA
)
,
an
d
tim
e
-
v
ar
y
in
g
s
y
m
p
ath
etic
ac
tiv
ity
(
T
VSy
m
p
)
,
we
is
o
late
to
n
ic
an
d
p
h
asic
co
m
p
o
n
e
n
ts
an
d
q
u
a
n
tify
th
eir
p
ea
k
co
u
n
ts
an
d
r
ate
s
.
W
e
th
en
ap
p
ly
s
tatis
tical
c
o
n
tr
o
ls
,
m
u
ltil
ev
el
m
o
d
elin
g
,
a
n
d
ML
with
leav
e
-
one
-
s
u
b
ject
-
o
u
t
(
L
OSO)
v
alid
atio
n
to
d
eter
m
in
e
wh
eth
e
r
th
ese
f
ea
tu
r
es
r
ef
lect
g
en
u
in
e
c
o
g
n
itiv
e
w
o
r
k
lo
a
d
o
r
ar
e
b
etter
ex
p
lain
ed
b
y
s
ess
io
n
len
g
th
.
B
ey
o
n
d
in
f
er
en
tial
an
aly
s
is
,
we
ass
e
s
s
th
e
g
en
er
aliza
b
ilit
y
o
f
p
ea
k
-
d
er
iv
ed
f
ea
t
u
r
es
u
s
in
g
ML
class
if
ier
s
tr
ain
ed
u
n
d
er
s
u
b
ject
-
in
d
ep
en
d
en
t
v
alid
atio
n
.
W
e
co
m
p
ar
e
u
n
im
o
d
al
E
DA
m
o
d
els
ag
ain
s
t
m
u
ltimo
d
al
v
ar
ia
n
ts
th
at
in
clu
d
e
HR
V,
ev
alu
atin
g
wh
eth
er
a
d
d
in
g
HR
V
im
p
r
o
v
es
cr
o
s
s
-
s
u
b
ject
p
r
ed
ictio
n
o
r
in
tr
o
d
u
ce
s
v
ar
ian
ce
th
at
h
in
d
e
r
s
p
er
f
o
r
m
a
n
ce
.
B
y
d
is
tin
g
u
is
h
in
g
b
etwe
en
tim
e
-
d
r
i
v
en
an
d
wo
r
k
lo
ad
-
s
p
ec
if
ic
p
h
y
s
io
lo
g
ical
s
ig
n
als,
th
is
s
tu
d
y
co
n
tr
i
b
u
tes
to
th
e
d
ev
el
o
p
m
en
t
o
f
af
f
ec
tiv
e
co
m
p
u
tin
g
s
y
s
tem
s
th
at
ar
e
b
o
th
in
ter
p
r
etab
le
an
d
r
eliab
le
in
d
y
n
am
ic,
r
ea
l
-
wo
r
l
d
en
v
i
r
o
n
m
e
n
ts
s
u
ch
as e
s
p
o
r
ts
.
2.
M
E
T
H
O
D
2
.
1
.
Da
t
a
s
et
a
nd
pa
rt
icipa
nt
T
h
e
s
tu
d
y
u
tili
ze
d
th
e
p
u
b
licly
av
ailab
le
esp
o
r
ts
s
en
s
o
r
s
d
ataset
[
1
2
]
,
wh
ich
co
n
tain
s
m
u
ltimo
d
al
p
h
y
s
io
lo
g
ical
r
ec
o
r
d
in
g
s
f
r
o
m
2
2
co
m
p
etitiv
e
leag
u
e
o
f
leg
en
d
s
(
L
OL
)
m
atch
es
in
v
o
lv
in
g
two
team
s
o
f
f
iv
e
p
lay
er
s
ea
ch
,
to
talin
g
1
1
0
in
d
iv
id
u
al
s
ess
io
n
in
s
tan
ce
s
.
All
p
ar
ticip
an
ts
wer
e
m
ale,
ag
ed
1
8
to
3
5
y
ea
r
s
,
an
d
wer
e
ca
teg
o
r
ized
in
to
two
co
h
o
r
ts
b
ased
o
n
th
eir
lev
el
o
f
ex
p
er
tis
e.
Pro
f
ess
io
n
al
p
lay
er
s
r
ep
o
r
ted
5
,
0
0
0
to
1
0
,
0
0
0
h
o
u
r
s
o
f
g
am
e
p
lay
,
wh
ile
am
ateu
r
s
h
ad
b
etwe
en
4
0
0
an
d
1
,
2
0
0
h
o
u
r
s
.
E
ac
h
s
ess
io
n
in
clu
d
ed
s
y
n
c
h
r
o
n
ized
r
ec
o
r
d
in
g
s
o
f
E
D
A,
h
ea
r
t
r
ate
(
HR
)
,
a
n
d
p
o
s
t
-
m
atch
s
elf
-
r
ep
o
r
ted
wo
r
k
lo
a
d
,
co
l
lecte
d
v
ia
wr
is
t
-
m
o
u
n
ted
b
io
s
en
s
o
r
s
in
n
atu
r
alis
tic,
to
u
r
n
a
m
en
t
-
s
ty
le
en
v
ir
o
n
m
en
ts
.
R
ich
m
etad
ata
ac
co
m
p
an
ies
ea
ch
s
ess
io
n
,
in
c
lu
d
in
g
m
atch
d
u
r
atio
n
,
p
lay
er
r
o
le,
ca
len
d
ar
d
ay
o
f
g
am
e
p
lay
,
an
d
th
e
o
r
d
er
o
f
g
am
ep
la
y
with
in
t
h
e
d
a
y
.
Ma
tch
es
wer
e
d
is
tr
ib
u
ted
ac
r
o
s
s
f
o
u
r
c
o
n
s
ec
u
tiv
e
d
ay
s
,
with
ea
c
h
d
ay
co
m
p
r
is
in
g
m
u
ltip
le
s
ess
io
n
s
in
d
ex
e
d
b
o
th
g
lo
b
ally
an
d
lo
ca
lly
(
e.
g
.
,
m
atch
3
o
n
d
ay
1
an
d
m
atch
6
o
n
d
ay
2
)
,
e
n
ab
li
n
g
p
r
ec
is
e
tem
p
o
r
al
alig
n
m
en
t
.
A
q
u
ality
co
n
tr
o
l
p
r
o
to
co
l
was
ap
p
lied
to
ex
clu
d
e
s
ess
io
n
s
with
m
is
s
in
g
o
r
co
r
r
u
p
ted
E
DA,
in
s
u
f
f
icien
t
HR
d
ata,
o
r
in
c
o
m
p
lete
s
elf
-
r
ep
o
r
ts
.
T
h
is
r
esu
lted
in
a
clea
n
an
aly
tic
co
h
o
r
t
s
u
itab
le
f
o
r
d
o
wn
s
tr
ea
m
s
tatis
tical
an
d
ML
an
aly
s
es.
Po
s
t
-
m
atch
wo
r
k
lo
ad
r
atin
g
s
,
co
llected
o
n
a
f
iv
e
-
p
o
in
t
L
ik
er
t
s
ca
le,
wer
e
b
in
ar
ized
in
to
h
ig
h
(
4
-
5
)
an
d
lo
w
(
1
-
3
)
wo
r
k
lo
ad
lab
els,
wh
ich
s
er
v
ed
as
th
e
p
r
im
ar
y
tar
g
et
v
ar
iab
le
.
T
h
e
d
ataset'
s
tem
p
o
r
al
s
tr
u
ctu
r
e
allo
ws
f
o
r
an
al
y
s
is
o
f
f
atig
u
e
ef
f
ec
ts
,
cir
ca
d
ia
n
v
ar
iatio
n
,
an
d
n
ested
m
o
d
elin
g
ac
r
o
s
s
p
lay
er
s
an
d
s
ess
io
n
s
.
2
.
2
.
E
lect
ro
derma
l sig
na
l dec
o
m
po
s
it
io
n
R
aw
E
DA
s
ig
n
als
wer
e
f
ir
s
t
d
en
o
is
ed
u
s
in
g
Dau
b
ec
h
ies
-
4
wav
elet
tr
an
s
f
o
r
m
s
to
atten
u
ate
m
o
tio
n
ar
tifa
cts
an
d
s
u
p
p
r
ess
h
i
g
h
-
f
r
eq
u
e
n
cy
n
o
is
e.
W
e
t
h
en
d
ec
o
m
p
o
s
ed
th
e
s
ig
n
als
in
t
o
to
n
ic
an
d
p
h
asic
co
m
p
o
n
en
ts
u
s
in
g
th
r
ee
wid
ely
r
ec
o
g
n
ized
m
eth
o
d
s
:
cv
x
E
DA,
s
p
ar
s
eE
DA,
an
d
T
VS
y
m
p
.
T
h
e
cv
x
E
DA
m
o
d
el,
in
tr
o
d
u
ce
d
b
y
Gr
ec
o
et
a
l
.
[
1
3
]
ass
u
m
es
th
at
th
e
o
b
s
er
v
ed
s
k
in
c
o
n
d
u
ctan
ce
s
ig
n
al
y(
t)
ca
n
b
e
r
ep
r
esen
ted
as (
1
)
.
(
)
=
(
)
+
(
ℎ
∗
)
(
)
+
(
)
(
1
)
W
h
er
e
r
(
t)
is
th
e
s
lo
wly
v
ar
y
in
g
to
n
ic
co
m
p
o
n
en
t,
(
)
≥
0
is
a
s
p
ar
s
e
p
h
asic
d
r
iv
er
,
h
(
t
)
is
th
e
ca
n
o
n
ical
im
p
u
ls
e
r
esp
o
n
s
e
o
f
th
e
s
u
d
o
m
o
to
r
n
e
r
v
e
ac
tiv
ity
,
∗
d
en
o
te
s
co
n
v
o
lu
tio
n
,
an
d
ε
(
t)
is
Gau
s
s
ian
n
o
is
e.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
C
la
s
s
i
fyin
g
men
ta
l wo
r
klo
a
d
o
f e
s
p
o
r
ts
p
la
ye
r
s
u
s
in
g
ma
ch
in
e
lea
r
n
in
g
(
A
is
y
A
l F
a
w
w
a
z
)
471
Sp
ar
s
eE
DA,
p
r
o
p
o
s
ed
b
y
Gal
leg
o
et
a
l
.
[
1
4
]
,
also
m
o
d
els
th
e
E
DA
s
ig
n
al
as
th
e
s
u
m
o
f
to
n
ic
an
d
p
h
asic
co
m
p
o
n
en
ts
b
u
t
f
o
c
u
s
es
o
n
r
ec
o
v
e
r
in
g
p
(
t)
th
r
o
u
g
h
n
o
n
-
n
eg
ativ
e
least
ab
s
o
lu
te
s
h
r
in
k
ag
e
an
d
s
elec
tio
n
o
p
er
ato
r
(
L
ASSO
)
-
b
ased
s
p
ar
s
eE
DA
.
T
h
e
m
o
d
el
i
s
f
o
r
m
u
lated
as (
2
)
.
≥
0
‖
−
ℎ
∗
‖
2
+
‖
‖
1
(
2
)
W
ith
th
e
to
n
ic
co
m
p
o
n
e
n
t
es
tim
ated
u
s
in
g
ad
ap
tiv
e
p
o
ly
n
o
m
ial
s
m
o
o
th
i
n
g
.
T
h
is
a
p
p
r
o
ac
h
is
lig
h
tweig
h
t,
ef
f
icien
t,
an
d
h
ig
h
ly
s
u
itab
le
f
o
r
em
b
e
d
d
ed
o
r
wea
r
ab
le
a
p
p
l
icatio
n
s
wh
er
e
in
ter
p
r
etab
ilit
y
an
d
co
m
p
u
tatio
n
al
s
p
ee
d
ar
e
cr
itical.
I
n
co
n
tr
ast,
T
VSy
m
p
,
p
r
o
p
o
s
ed
b
y
Qu
in
ter
o
et
a
l
.
[
1
5
]
,
a
v
o
id
s
ex
p
licit
d
ec
o
n
v
o
lu
tio
n
.
I
t
in
s
tead
an
aly
ze
s
th
e
E
DA
s
ig
n
al
i
n
th
e
tim
e
-
f
r
eq
u
en
cy
d
o
m
ain
u
s
in
g
v
ar
iab
le
-
f
r
eq
u
en
cy
co
m
p
lex
d
em
o
d
u
latio
n
.
T
h
e
s
y
m
p
ath
etic
en
er
g
y
E
(
t)
is
ex
tr
ac
ted
f
r
o
m
th
e
s
ig
n
al’
s
tim
e
-
v
ar
y
in
g
s
p
ec
tr
al
d
en
s
ity
S
(
t,f
)
em
p
h
asizin
g
th
e
f
r
eq
u
e
n
cy
b
a
n
d
ass
o
ciate
d
with
s
y
m
p
ath
etic
ac
tiv
atio
n
as in
(
3
)
.
(
)
=
∫
(
,
)
2
1
(
3
)
Ou
r
s
elec
tio
n
o
f
th
ese
th
r
ee
m
eth
o
d
s
was
in
f
o
r
m
e
d
b
y
a
b
en
ch
m
a
r
k
co
m
p
ar
is
o
n
co
n
d
u
cted
b
y
Vee
r
an
k
i
et
a
l
.
[
1
6
]
,
wh
ich
d
em
o
n
s
tr
ated
s
u
b
s
tan
tial
v
ar
iab
ilit
y
in
d
ec
o
m
p
o
s
itio
n
p
er
f
o
r
m
an
ce
ac
r
o
s
s
af
f
ec
tiv
e
co
m
p
u
tin
g
task
s
.
Giv
en
th
at
esp
o
r
ts
’
s
ettin
g
s
in
v
o
lv
e
co
n
tin
u
o
u
s
co
g
n
itiv
e
e
n
g
a
g
em
en
t,
ex
ter
n
ally
p
ac
ed
task
d
em
an
d
s
,
an
d
m
in
i
m
al
g
r
o
s
s
m
o
to
r
m
o
v
e
m
en
t,
it
r
em
ain
s
a
n
o
p
en
q
u
esti
o
n
wh
ich
d
ec
o
m
p
o
s
itio
n
ap
p
r
o
ac
h
y
ield
s
t
h
e
m
o
s
t
b
eh
av
i
o
r
ally
c
o
n
g
r
u
en
t
f
e
atu
r
es
u
n
d
e
r
r
ea
l
-
tim
e
wo
r
k
lo
ad
.
Fr
o
m
ea
c
h
d
ec
o
m
p
o
s
itio
n
o
u
tp
u
t,
we
e
x
tr
ac
ted
T
PC
an
d
PP
C
u
s
in
g
a
ze
r
o
-
cr
o
s
s
in
g
-
b
ased
lo
ca
l
m
ax
im
a
d
etec
to
r
im
p
lem
en
ted
v
ia
s
cip
y
.
s
ig
n
al.
f
in
d
_
p
ea
k
s
[
1
7
]
.
T
h
ese
f
ea
tu
r
e
s
wer
e
co
m
p
u
ted
ac
r
o
s
s
en
tire
g
am
ep
lay
s
ess
io
n
s
an
d
with
in
f
ix
ed
-
le
n
g
th
tem
p
o
r
al
s
eg
m
en
ts
to
s
u
p
p
o
r
t
with
in
-
s
ess
io
n
m
o
d
elin
g
.
T
o
d
eter
m
in
e
th
e
m
o
s
t
s
u
itab
le
m
eth
o
d
f
o
r
w
o
r
k
lo
a
d
m
o
d
elin
g
,
we
e
v
alu
ated
t
h
e
s
tatis
tical
co
r
r
esp
o
n
d
en
ce
b
etwe
en
ex
tr
ac
ted
f
ea
tu
r
es
an
d
p
lay
er
s
’
s
elf
-
r
e
p
o
r
ted
m
e
n
tal
wo
r
k
l
o
ad
.
Sp
ea
r
m
an
co
r
r
elatio
n
c
o
ef
f
icie
n
ts
wer
e
co
m
p
u
te
d
b
etwe
en
T
PC
,
PP
C
,
an
d
th
eir
r
esp
ec
tiv
e
n
o
r
m
alize
d
r
ates a
g
ain
s
t b
in
ar
ized
wo
r
k
lo
ad
lab
el
s
.
2
.
3
.
T
em
po
r
a
l seg
m
ent
a
t
io
n a
nd
pea
k
ra
t
e
m
o
delin
g
T
o
ex
am
in
e
wh
et
h
er
E
DA
p
e
ak
co
u
n
ts
ex
h
i
b
it
in
tr
in
s
ic
tem
p
o
r
al
ac
cu
m
u
latio
n
,
ea
c
h
s
ess
io
n
-
lev
el
E
DA
s
ig
n
al
was
s
eg
m
en
ted
in
to
n
o
n
-
o
v
e
r
lap
p
in
g
win
d
o
w
s
o
f
eq
u
al
d
u
r
atio
n
.
W
ith
in
ea
ch
s
eg
m
en
t
s
,
we
co
m
p
u
ted
th
e
p
h
asic p
ea
k
r
ate
(
)
,
an
d
t
o
n
ic
p
ea
k
r
ate
(
)
,
d
ef
in
ed
as (
4
)
.
(
)
=
(
)
Δ
,
(
)
=
(
)
Δ
(
4
)
W
h
er
e
(
)
an
d
(
)
d
en
o
te
th
e
n
u
m
b
er
o
f
d
etec
ted
p
h
asic
an
d
to
n
ic
p
ea
k
s
with
in
s
eg
m
en
t
s
,
an
d
Δ
t
r
ep
r
esen
ts
th
e
f
ix
ed
s
eg
m
en
t d
u
r
atio
n
in
m
in
u
tes.
T
o
ass
ess
th
e
p
r
esen
ce
o
f
tim
e
-
d
ep
en
d
en
t
ac
cu
m
u
latio
n
,
we
ca
lcu
lated
th
e
Pear
s
o
n
co
r
r
elatio
n
co
ef
f
icien
t
b
etwe
en
th
e
s
eg
m
e
n
t
in
d
ex
an
d
co
r
r
esp
o
n
d
in
g
p
e
ak
r
ates
ac
r
o
s
s
all
s
eg
m
en
ts
f
o
r
ea
ch
p
ar
ticip
an
t.
T
h
is
q
u
an
tifie
s
th
e
lin
ea
r
as
s
o
ciatio
n
b
etwe
en
tim
e
elap
s
ed
an
d
p
h
y
s
io
lo
g
ical
r
ea
ctiv
i
ty
.
A
s
ig
n
if
ican
tly
p
o
s
itiv
e
co
r
r
elatio
n
wo
u
ld
in
d
icate
a
s
y
s
tem
atic
in
cr
ea
s
e
in
E
DA
p
ea
k
s
o
v
er
tim
e,
in
d
ep
e
n
d
en
t
o
f
s
u
b
jectiv
e
wo
r
k
lo
ad
lev
els.
W
e
ad
d
itio
n
ally
co
m
p
u
ted
Sp
ea
r
m
an
r
a
n
k
-
o
r
d
er
co
r
r
elatio
n
s
.
T
h
is
d
u
al
-
m
etr
ic
ap
p
r
o
ac
h
p
r
o
v
id
es
a
m
o
r
e
c
o
m
p
r
eh
en
s
i
v
e
ass
ess
m
en
t
o
f
tem
p
o
r
al
tr
e
n
d
s
,
d
is
tin
g
u
is
h
in
g
b
etwe
en
li
n
ea
r
an
d
m
o
n
o
to
n
ic
ac
cu
m
u
latio
n
p
atter
n
s
.
2
.
4
.
St
a
t
is
t
ica
l a
na
ly
s
is
a
nd
t
em
po
ra
l dis
a
m
big
ua
t
io
n
T
o
d
is
e
n
ta
n
g
l
e
c
o
g
n
iti
v
e
w
o
r
k
lo
a
d
e
f
f
ec
ts
f
r
o
m
ti
m
e
-
on
-
t
as
k
c
o
n
f
o
u
n
d
s
,
we
e
m
p
l
o
y
e
d
a
m
u
lt
i
-
t
ie
r
e
d
in
f
e
r
e
n
ti
al
f
r
am
ew
o
r
k
t
h
at
i
n
c
o
r
p
o
r
at
ed
b
i
v
a
r
i
ate
c
o
r
r
el
ati
o
n
,
p
a
r
t
ial
c
o
r
r
el
ati
o
n
,
li
n
ea
r
m
i
x
e
d
-
e
f
f
ec
ts
m
o
d
el
s
(
L
M
Ms)
,
a
n
d
r
esa
m
p
li
n
g
-
b
as
ed
v
ali
d
a
ti
o
n
.
W
e
f
i
r
s
t
c
o
m
p
u
te
d
Sp
ea
r
m
a
n
co
r
r
el
ati
o
n
s
b
etwe
e
n
t
o
t
al
s
ess
i
o
n
d
u
r
ati
o
n
a
n
d
b
o
th
p
h
as
ic
an
d
T
PC
.
T
h
is
a
n
a
ly
s
is
ass
ess
e
d
w
h
et
h
er
lo
n
g
e
r
g
a
m
e
p
l
a
y
s
ess
i
o
n
s
n
at
u
r
al
ly
g
en
e
r
at
e
m
o
r
e
E
DA
p
ea
k
s
r
eg
ar
d
l
ess
o
f
c
o
g
n
iti
v
e
d
e
m
a
n
d
,
s
e
r
v
in
g
as
b
as
eli
n
e
i
n
d
ic
at
o
r
o
f
t
em
p
o
r
al
ac
c
u
m
u
lat
io
n
.
T
o
is
o
late
th
e
s
p
ec
if
ic
ass
o
cia
tio
n
b
etwe
en
E
DA
d
y
n
a
m
ics
an
d
p
e
r
ce
iv
ed
wo
r
k
lo
a
d
,
in
d
e
p
en
d
en
t
o
f
s
ess
io
n
len
g
th
,
we
th
en
co
m
p
u
ted
p
ar
tial
c
o
r
r
elatio
n
s
b
etwe
en
n
o
r
m
alize
d
p
ea
k
r
ates
(
p
ea
k
s
p
er
m
in
u
te)
an
d
b
in
ar
y
wo
r
k
lo
a
d
lab
els,
co
n
tr
o
llin
g
f
o
r
s
ess
io
n
d
u
r
atio
n
.
T
h
i
s
s
tep
en
ab
led
u
s
to
ass
ess
wh
eth
er
E
DA
-
d
er
i
v
ed
f
ea
tu
r
es
r
etain
ed
e
x
p
lan
at
o
r
y
p
o
wer
af
ter
ac
c
o
u
n
tin
g
f
o
r
tem
p
o
r
al
ex
p
o
s
u
r
e.
Nex
t
,
we
im
p
lem
en
ted
h
ier
ar
ch
ical
L
MM
s
with
p
h
asic
an
d
to
n
ic
p
ea
k
r
ates
as
d
ep
en
d
en
t
v
ar
iab
les.
W
o
r
k
lo
a
d
was
en
ter
ed
as
a
f
ix
ed
ef
f
ec
t,
an
d
r
an
d
o
m
in
ter
ce
p
ts
wer
e
s
p
ec
if
ied
f
o
r
p
lay
er
id
e
n
tity
an
d
m
atch
d
a
y
to
ac
c
o
u
n
t
f
o
r
in
ter
in
d
iv
id
u
al
an
d
s
ess
io
n
-
lev
el
v
ar
iab
ilit
y
.
T
h
e
b
ase
m
o
d
el
to
o
k
th
e
f
o
r
m
(
5
)
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
1
,
Feb
r
u
ar
y
20
26
:
469
-
4
8
0
472
=
β
0
+
β
1
×
+
(
)
+
(
)
+
ϵ
(
5
)
W
h
er
e
is
th
e
p
ea
k
r
ate
f
o
r
s
ess
io
n
,
is
th
e
wo
r
k
lo
ad
c
o
n
d
iti
o
n
(
0
=lo
w,
1
=h
ig
h
)
,
an
d
(
)
,
(
)
ar
e
r
an
d
o
m
ef
f
ec
ts
.
T
h
is
m
o
d
e
l w
as f
itted
s
ep
ar
ately
f
o
r
p
h
asic a
n
d
to
n
ic
c
o
m
p
o
n
en
ts
.
T
o
in
v
esti
g
ate
p
o
ten
tial
in
te
r
a
ctio
n
ef
f
ec
ts
b
etwe
en
tim
e
a
n
d
wo
r
k
lo
ad
,
we
ex
te
n
d
ed
th
e
m
o
d
el
to
in
clu
d
e
s
eg
m
en
t in
d
ex
a
n
d
its
in
ter
ac
tio
n
with
wo
r
k
lo
ad
(
6
)
.
=
β
0
+
β
1
×
+
β
2
×
+
β
3
×
(
×
)
+
(
)
+
ε
(
6
)
W
h
er
e
is
th
e
s
eg
m
en
t
in
d
e
x
.
A
s
ig
n
if
ican
t
in
ter
ac
tio
n
ter
m
(
3
)
wo
u
l
d
in
d
icate
th
at
tem
p
o
r
a
l
ac
cu
m
u
latio
n
p
atter
n
s
d
if
f
er
b
y
wo
r
k
lo
a
d
co
n
d
itio
n
.
Fin
ally
,
to
ass
ess
m
o
d
el
r
o
b
u
s
tn
ess
an
d
g
en
er
aliza
b
ilit
y
,
w
e
co
n
d
u
cte
d
a
co
m
b
in
ato
r
ial
r
esam
p
lin
g
an
aly
s
is
.
Sp
ec
if
ically
,
we
g
en
er
ated
1
,
0
0
0
s
tr
atif
ied
b
o
o
ts
tr
ap
s
u
b
s
ets
o
f
p
lay
er
s
a
n
d
s
ess
io
n
s
,
r
ep
ea
tin
g
th
e
f
u
ll
co
r
r
elatio
n
an
d
m
ix
ed
-
m
o
d
el
an
aly
s
es
f
o
r
ea
ch
.
T
h
es
e
y
ield
ed
d
is
tr
ib
u
tio
n
s
o
f
p
a
r
am
eter
esti
m
ates
(
e.
g
.
,
1
)
f
r
o
m
wh
ic
h
we
d
er
iv
ed
co
n
f
id
e
n
ce
in
ter
v
als
an
d
s
tab
ilit
y
in
d
ices.
Su
ch
r
esam
p
lin
g
allo
ws
u
s
to
esti
m
ate
p
ar
am
eter
v
ar
iab
ilit
y
u
n
d
e
r
d
if
f
e
r
en
t
s
am
p
lin
g
co
n
f
ig
u
r
atio
n
s
an
d
ass
ess
th
e
r
eliab
ilit
y
o
f
o
u
r
in
f
er
en
ce
s
ac
r
o
s
s
s
u
b
s
ets.
2
.
5
.
M
a
chine le
a
rning
cla
s
s
i
f
ica
t
io
n f
o
r
m
ent
a
l lo
a
d
wit
h L
O
SO
T
o
co
m
p
lem
e
n
t
o
u
r
in
f
er
e
n
t
ial
an
aly
s
es
an
d
ev
alu
ate
th
e
p
r
ed
ictiv
e
v
alid
ity
o
f
E
D
A
-
d
er
iv
ed
f
ea
tu
r
es,
we
im
p
lem
en
te
d
a
s
u
p
er
v
is
ed
ML
f
r
am
ewo
r
k
u
s
in
g
L
OSO
cr
o
s
s
-
v
alid
atio
n
ap
p
r
o
ac
h
[
1
8
]
.
T
h
is
ev
alu
atio
n
p
r
o
to
c
o
l
was
s
elec
ted
to
en
s
u
r
e
s
tr
ict
g
e
n
er
aliza
tio
n
to
u
n
s
ee
n
in
d
iv
i
d
u
als,
w
h
ich
is
p
ar
ticu
lar
ly
cr
itical
in
p
h
y
s
io
lo
g
ical
c
o
m
p
u
tin
g
,
w
h
er
e
in
te
r
-
s
u
b
ject
v
ar
iab
ilit
y
ca
n
o
b
s
cu
r
e
m
o
d
el
p
e
r
f
o
r
m
a
n
ce
.
I
n
ea
c
h
L
OSO
iter
atio
n
,
d
ata
f
r
o
m
a
s
in
g
le
p
ar
ticip
a
n
t
was
h
eld
o
u
t
as
th
e
test
s
et,
wh
ile
m
o
d
els
wer
e
tr
ain
ed
o
n
d
ata
f
r
o
m
all
r
em
ain
in
g
p
ar
ticip
a
n
ts
.
T
h
is
p
r
o
ce
s
s
was
r
ep
ea
ted
u
n
til
ea
ch
p
ar
ticip
an
t
h
ad
s
er
v
ed
as
th
e
test
s
et
ex
ac
tly
o
n
ce
,
th
er
eb
y
en
s
u
r
in
g
co
m
p
r
e
h
en
s
iv
e
an
d
u
n
b
iased
v
alid
atio
n
ac
r
o
s
s
th
e
wh
o
le
co
h
o
r
t.
F
e
a
t
u
r
e
v
e
c
t
o
r
s
we
r
e
c
o
n
s
t
r
u
ct
e
d
u
s
i
n
g
b
o
t
h
p
h
y
s
i
o
l
o
g
i
c
a
l
a
n
d
c
o
n
t
e
x
t
u
a
l
a
t
t
r
i
b
u
t
e
s
.
T
h
ese
i
n
c
l
u
d
e
d
t
h
e
P
PC
a
n
d
T
P
C
a
n
d
t
h
e
i
r
n
o
r
m
a
l
i
z
e
d
r
a
t
es
,
c
o
m
p
u
t
e
d
o
v
e
r
b
o
t
h
t
h
e
w
h
o
l
e
s
e
s
s
i
o
n
a
n
d
t
e
m
p
o
r
a
l
l
y
s
e
g
m
e
n
t
e
d
i
n
t
e
r
v
a
l
s
.
T
e
m
p
o
r
a
l
s
l
o
p
e
c
o
e
f
f
i
c
i
e
n
ts
we
r
e
c
al
c
u
l
a
te
d
f
r
o
m
p
e
a
k
r
a
t
e
t
r
e
n
d
s
o
v
e
r
t
i
m
e
.
W
e
a
l
s
o
i
n
c
l
u
d
e
d
H
R
V
f
e
a
t
u
r
e
s
,
s
u
c
h
as
s
t
a
n
d
a
r
d
d
e
v
i
a
t
i
o
n
o
f
n
o
r
m
a
l
-
to
-
n
o
r
m
a
l
i
n
te
r
v
a
l
s
(
S
DN
N
)
,
r
o
o
t
m
e
a
n
s
q
u
a
r
e
o
f
s
u
c
c
ess
i
v
e
d
i
f
f
e
r
e
n
c
e
s
(
R
MS
S
D
)
,
c
o
e
f
f
i
c
ie
n
t
o
f
v
a
r
i
a
t
i
o
n
(
C
V
)
,
a
n
d
S
h
a
n
n
o
n
e
n
t
r
o
p
y
.
S
e
s
s
i
o
n
m
e
t
a
d
a
t
a
,
s
u
c
h
a
s
m
at
c
h
d
u
r
a
t
i
o
n
,
g
a
m
e
p
l
a
y
o
r
d
e
r
w
i
t
h
in
t
h
e
d
a
y
,
a
n
d
p
l
a
y
e
r
r
o
l
e
,
w
e
r
e
a
p
p
e
n
d
e
d
t
o
c
a
p
t
u
r
e
s
it
u
a
t
i
o
n
a
l
f
a
c
t
o
r
s
.
All
f
ea
tu
r
es
wer
e
s
tan
d
ar
d
ize
d
u
s
in
g
z
-
s
co
r
e
n
o
r
m
aliza
tio
n
,
co
m
p
u
te
d
ex
clu
s
iv
ely
o
n
th
e
tr
ain
in
g
f
o
ld
s
to
p
r
ev
e
n
t
in
f
o
r
m
atio
n
leak
ag
e
in
to
th
e
test
s
et.
T
o
m
o
d
el
th
e
d
ata,
we
tr
ain
ed
f
o
u
r
class
if
ier
s
,
r
ep
r
esen
tin
g
b
o
th
lin
ea
r
an
d
n
o
n
lin
ea
r
p
ar
a
d
ig
m
s
:
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
s
(
SVM)
with
a
r
ad
ial
b
asis
f
u
n
ctio
n
k
er
n
el
(
R
B
F),
a
m
u
l
tilay
er
p
er
ce
p
tr
o
n
(
ML
P)
,
n
e
u
r
al
n
etwo
r
k
(
NN)
,
d
ec
is
io
n
t
r
ee
s
(
DT
)
,
ex
tr
em
e
g
r
ad
ien
t
b
o
o
s
tin
g
(
XGBo
o
s
t)
,
an
d
lo
g
is
tic
r
eg
r
ess
io
n
(
L
R
)
as
a
b
aselin
e.
Hy
p
er
p
ar
am
eter
s
f
o
r
ea
ch
class
if
ier
wer
e
o
p
tim
ized
v
ia
n
ested
g
r
i
d
s
ea
r
ch
with
in
ea
ch
tr
ain
in
g
f
o
ld
.
W
e
co
m
p
u
ted
m
u
ltip
le
cl
ass
if
icatio
n
m
etr
ics
,
in
clu
d
in
g
ac
cu
r
ac
y
,
AUC,
F1
-
s
co
r
e,
an
d
lo
g
ar
ith
m
ic
lo
s
s
.
Acc
u
r
ac
y
was
d
ef
in
ed
as
th
e
p
r
o
p
o
r
tio
n
o
f
co
r
r
ec
tly
class
if
ied
in
s
tan
ce
s
o
v
er
th
e
to
tal
n
u
m
b
er
o
f
p
r
ed
ictio
n
s
.
AUC
was
esti
m
ated
b
y
n
u
m
er
ically
in
teg
r
atin
g
th
e
r
ec
eiv
er
o
p
er
at
in
g
ch
ar
ac
ter
is
tic
(
R
OC
)
cu
r
v
e
u
s
in
g
th
e
tr
ap
ez
o
id
al
r
u
le,
ca
p
tu
r
in
g
t
h
e
m
o
d
el
’
s
ab
ilit
y
to
r
a
n
k
p
o
s
itiv
e
in
s
tan
ce
s
h
ig
h
er
th
an
n
eg
ativ
e
o
n
es
ir
r
esp
ec
tiv
e
o
f
class
if
icatio
n
th
r
esh
o
l
d
.
T
o
p
en
alize
o
v
er
c
o
n
f
i
d
en
t m
is
class
if
icatio
n
s
,
we
also
r
ep
o
r
ted
l
o
g
lo
s
s
,
d
ef
in
e
d
as (
7
)
.
L
o
g
L
o
s
s
=
−
1
∑
[
l
og
(
̂
)
+
(
1
−
)
l
og
(
1
−
̂
)
]
=
1
(
7
)
W
h
er
e
∈
{
0
,
1
}
is
th
e
tr
u
e
lab
el
a
n
d
̂
∈
(
0
,
1
)
is
th
e
p
r
ed
icted
p
r
o
b
a
b
ilit
y
o
f
th
e
p
o
s
itiv
e
class
.
T
o
m
itig
ate
class
im
b
alan
ce
,
we
ap
p
lied
class
we
ig
h
tin
g
in
v
er
s
ely
p
r
o
p
o
r
tio
n
al
to
class
f
r
eq
u
en
c
y
,
an
d
u
s
ed
s
tr
atif
ied
s
am
p
lin
g
wh
er
e
a
p
p
licab
le.
T
o
ass
ess
in
ter
-
s
u
b
ject
r
o
b
u
s
tn
ess
,
we
tr
ac
k
ed
p
ar
ticip
a
n
t
-
l
ev
el
co
n
f
u
s
io
n
m
atr
ices
an
d
co
m
p
u
ted
th
e
av
er
a
g
e
R
OC
cu
r
v
e
ac
r
o
s
s
L
OSO
iter
atio
n
s
.
All
m
o
d
el
in
g
p
r
o
ce
d
u
r
es
wer
e
im
p
lem
e
n
ted
in
Py
th
o
n
3
.
9
u
s
in
g
th
e
Scik
it
-
lear
n
an
d
XGBo
o
s
t
lib
r
ar
ies.
T
h
ey
wer
e
ex
ec
u
ted
in
a
clo
u
d
-
b
ased
en
v
ir
o
n
m
en
t
(
Go
o
g
le
C
o
lab
)
t
o
en
s
u
r
e
r
ep
r
o
d
u
cib
ilit
y
.
3.
RE
SU
L
T
S
3
.
1
.
Da
t
a
s
et
des
cr
iptiv
es a
nd
qu
a
lity
co
ntr
o
l
T
h
e
f
in
al
d
ataset
in
clu
d
ed
9
2
p
lay
er
s
ess
io
n
s
co
llected
f
r
o
m
2
2
L
OL
m
atch
es,
ea
ch
in
v
o
lv
in
g
f
iv
e
in
d
iv
id
u
al
p
la
y
er
s
with
co
m
p
lete
r
ec
o
r
d
in
g
s
o
f
E
DA
an
d
HR
.
Ses
s
io
n
s
wer
e
r
etain
ed
af
ter
ap
p
ly
in
g
b
asic
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
C
la
s
s
i
fyin
g
men
ta
l wo
r
klo
a
d
o
f e
s
p
o
r
ts
p
la
ye
r
s
u
s
in
g
ma
ch
in
e
lea
r
n
in
g
(
A
is
y
A
l F
a
w
w
a
z
)
473
q
u
ality
ch
ec
k
s
to
en
s
u
r
e
d
at
a
co
m
p
leten
ess
an
d
s
ig
n
al
u
s
ab
ilit
y
.
An
d
th
e
av
er
ag
e
m
atch
d
u
r
atio
n
was
1
,
3
5
7
.
5
s
ec
o
n
d
s
(
ap
p
r
o
x
im
atel
y
2
2
.
6
m
in
u
tes).
3
.
2
.
Dec
o
m
po
s
it
io
n m
e
t
ho
d c
o
m
pa
riso
n
F
i
g
u
r
e
1
s
h
o
w
s
a
m
o
n
g
t
h
e
m
e
t
h
o
d
s
,
T
V
S
y
m
p
c
o
n
s
i
s
t
e
n
t
l
y
o
u
t
p
e
r
f
o
r
m
e
d
b
o
t
h
c
v
x
E
D
A
a
n
d
s
p
a
r
s
E
D
A
a
c
r
o
s
s
a
l
l
e
v
a
l
u
a
t
i
o
n
a
x
e
s
.
I
t
y
i
e
l
d
e
d
t
h
e
s
t
r
o
n
g
e
s
t
m
o
n
o
t
o
n
i
c
a
s
s
o
c
i
a
t
i
o
n
b
e
t
w
e
e
n
P
P
C
a
n
d
m
e
n
t
a
l
l
o
a
d
r
a
t
i
n
g
s
(
S
p
e
a
r
m
a
n
'
s
ρ
=
0
.
6
4
0
)
,
w
i
t
h
a
c
o
m
p
a
r
a
b
l
e
l
i
n
e
a
r
r
e
l
a
t
i
o
n
s
h
i
p
,
a
s
s
h
o
w
n
b
y
P
e
a
r
s
o
n
'
s
r
=
0
.
6
0
2
.
A
s
i
l
l
u
s
t
r
a
t
e
d
i
n
F
i
g
u
r
e
1
(
a
)
,
T
V
S
y
m
p
d
e
m
o
n
s
t
r
a
t
e
s
s
u
p
e
r
i
o
r
p
h
a
s
i
c
r
e
s
p
o
n
s
e
d
e
t
e
c
t
i
o
n
,
c
a
p
t
u
r
i
n
g
d
i
s
t
i
n
c
t
p
e
a
k
s
d
u
r
i
n
g
h
i
g
h
-
a
r
o
u
s
a
l
p
e
r
i
o
d
s
t
h
a
t
a
r
e
l
e
s
s
p
r
o
n
o
u
n
c
e
d
i
n
c
v
x
E
D
A
a
n
d
s
p
a
r
s
E
D
A
d
e
c
o
m
p
o
s
i
t
i
o
n
s
.
N
o
t
a
b
l
y
,
t
h
i
s
s
u
g
g
e
s
t
s
t
h
a
t
t
h
e
e
n
v
e
l
o
p
e
-
e
x
t
r
a
c
t
e
d
r
e
s
p
o
n
s
e
f
r
o
m
t
h
e
H
i
l
b
e
r
t
-
t
r
a
n
s
f
o
r
m
e
d
b
a
n
d
p
a
s
s
-
f
i
l
t
e
r
e
d
E
D
A
c
a
p
t
u
r
e
s
h
i
g
h
-
f
r
e
q
u
e
n
c
y
E
D
A
d
y
n
a
m
i
c
s
t
h
a
t
s
c
a
l
e
w
i
t
h
t
r
a
n
s
i
e
n
t
c
o
g
n
i
t
i
v
e
a
r
o
u
s
a
l
,
a
n
e
x
p
e
c
t
e
d
h
a
l
l
m
a
r
k
o
f
p
h
a
s
i
c
s
y
m
p
a
t
h
e
t
i
c
a
c
t
i
v
a
t
i
o
n
d
u
r
i
n
g
h
i
g
h
-
i
n
t
e
n
s
i
t
y
g
a
m
e
p
l
a
y
s
e
g
m
e
n
t
s
.
T
h
e
o
t
h
e
r
m
e
t
h
o
d
s
,
w
h
i
l
e
s
t
i
l
l
p
o
s
i
t
i
v
e
l
y
c
o
r
r
e
l
a
t
e
d
w
i
t
h
w
o
r
k
l
o
a
d
,
s
h
o
w
e
d
a
t
t
e
n
u
a
t
e
d
s
e
n
s
i
t
i
v
i
t
y
:
c
v
x
E
D
A
p
r
o
d
u
c
e
d
a
m
o
d
e
r
a
t
e
c
o
r
r
e
l
a
t
i
o
n
(
ρ
=
0
.
5
6
2
)
,
w
h
e
r
e
a
s
s
p
a
r
s
E
D
A
'
s
s
i
g
n
a
l
-
l
o
a
d
a
l
i
g
n
m
e
n
t
w
a
s
s
u
b
s
t
a
n
t
i
a
l
l
y
w
e
a
k
e
r
(
ρ
=
0
.
2
2
8
)
.
I
n
ter
m
s
o
f
class
if
icatio
n
b
et
wee
n
h
ig
h
-
a
n
d
lo
w
-
w
o
r
k
lo
a
d
s
tates,
T
VSy
m
p
ag
ain
ac
h
iev
ed
th
e
h
ig
h
est
ar
ea
u
n
d
er
th
e
R
OC
cu
r
v
e
(
AUC=0
.
8
8
0
)
,
r
ef
le
ctin
g
ex
ce
llen
t
d
is
cr
im
in
ativ
e
p
o
wer
.
T
h
is
was
s
u
p
p
o
r
ted
b
y
a
h
ig
h
ly
s
ig
n
if
i
ca
n
t
d
if
f
er
en
ce
i
n
PP
C
b
etwe
en
wo
r
k
lo
a
d
class
es
(
p
<1
e
-
7
,
Ma
n
n
–
W
h
itn
ey
U)
an
d
a
lar
g
e
m
ag
n
itu
d
e
ef
f
ec
t
s
ize
(
C
o
h
en
's
d
=
-
1
.
2
7
)
.
T
h
es
e
f
in
d
in
g
s
in
d
icate
th
at
T
VSy
m
p
-
d
er
i
v
ed
p
h
asic
f
ea
tu
r
es
ar
e
n
o
t
o
n
ly
s
tatis
tic
ally
r
o
b
u
s
t
b
u
t
also
p
r
ac
ticall
y
m
ea
n
i
n
g
f
u
l
in
d
is
tin
g
u
is
h
in
g
co
g
n
itiv
e
d
e
m
an
d
s
tates
u
n
d
er
g
am
in
g
co
n
d
itio
n
s
.
cv
x
E
DA,
wh
ile
s
lig
h
tly
less
s
en
s
itiv
e,
s
till
d
em
o
n
s
tr
ated
s
tr
o
n
g
p
er
f
o
r
m
a
n
ce
(
AUC=0
.
8
0
7
,
d
=1
.
0
8
)
,
u
n
d
e
r
s
co
r
in
g
its
co
n
tin
u
ed
r
elev
a
n
ce
in
s
em
i
-
co
n
tr
o
lled
en
v
ir
o
n
m
en
ts
.
I
n
co
n
tr
ast,
s
p
ar
s
E
DA
u
n
d
er
p
er
f
o
r
m
ed
r
e
lativ
e
to
b
o
th
m
eth
o
d
s
(
AUC=0
.
6
1
3
,
d
=−
0
.
4
7
)
,
lik
ely
d
u
e
to
its
r
elian
ce
o
n
s
p
ar
s
ity
-
d
r
iv
en
ass
u
m
p
tio
n
s
,
wh
ich
m
ay
b
e
less
s
tab
le
in
n
o
is
y
,
r
ea
l
-
wo
r
ld
d
atasets
.
W
h
ile
p
h
asic
co
m
p
o
n
en
ts
s
h
o
wed
co
n
s
is
ten
t
tr
en
d
s
,
to
n
ic
c
o
m
p
o
n
en
ts
r
ev
ea
led
m
o
r
e
h
et
er
o
g
en
e
o
u
s
b
eh
av
io
r
ac
r
o
s
s
m
eth
o
d
s
.
Fi
g
u
r
e
1
(
b
)
r
e
v
ea
ls
th
at
to
n
ic
d
ec
o
m
p
o
s
itio
n
s
ex
h
ib
it
g
r
ea
t
er
v
ar
ia
b
ilit
y
ac
r
o
s
s
m
eth
o
d
s
,
with
cv
x
E
DA
s
h
o
win
g
s
m
o
o
th
er
b
aselin
e
tr
en
d
s
co
m
p
ar
ed
to
th
e
m
o
r
e
o
s
cillato
r
y
p
atter
n
s
o
f
T
VSy
m
p
an
d
s
p
ar
s
E
DA.
I
n
te
r
esti
n
g
ly
,
cv
x
E
DA
ex
h
i
b
ited
th
e
m
o
s
t
p
o
ten
t
to
n
ic
–
l
o
ad
r
el
atio
n
s
h
ip
(
ρ
=0
.
4
6
4
,
d=
-
0
.
7
4
)
,
s
u
g
g
esti
n
g
th
at
its
s
lo
w
-
v
ar
y
in
g
b
aselin
e
m
o
d
elin
g
ef
f
ec
tiv
el
y
ca
p
tu
r
es
r
esid
u
a
l
au
to
n
o
m
ic
s
h
if
ts
,
wh
ich
m
ay
b
e
lin
k
ed
t
o
s
u
s
tain
ed
ef
f
o
r
t
o
r
f
atig
u
e.
T
VSy
m
p
,
d
esp
ite
its
p
h
asic
s
tr
en
g
th
,
p
r
o
d
u
ce
d
wea
k
e
r
to
n
ic
ass
o
ciatio
n
s
(
ρ
=0
.
3
7
7
,
d
=
-
0
.
4
7
)
,
p
o
s
s
ib
ly
r
ef
lectin
g
its
lim
ited
ab
ilit
y
to
is
o
late
lo
w
-
f
r
eq
u
e
n
cy
b
aselin
e
tr
en
d
s
.
s
p
ar
s
E
DA,
ag
ain
,
lag
g
ed
b
eh
in
d
b
o
th
c
o
r
r
elatio
n
s
a
n
d
ef
f
ec
t
s
izes,
wh
ich
wer
e
n
ea
r
th
e
n
o
is
e
f
lo
o
r
.
T
ak
en
to
g
eth
e
r
,
th
ese
f
in
d
i
n
g
s
h
ig
h
lig
h
t
th
at
T
VSy
m
p
o
f
f
er
s
th
e
m
o
s
t
r
o
b
u
s
t
an
d
d
is
cr
im
in
ativ
e
d
ec
o
m
p
o
s
itio
n
f
o
r
r
ea
l
-
tim
e
m
en
tal
wo
r
k
lo
ad
m
o
n
ito
r
in
g
in
esp
o
r
ts
,
p
a
r
ticu
lar
ly
wh
e
n
th
e
g
o
al
is
to
tr
ac
k
f
ast,
tr
an
s
ien
t
s
h
if
ts
in
ar
o
u
s
al.
C
Vx
E
DA
r
em
ain
s
a
s
o
lid
alter
n
ativ
e,
esp
ec
ially
wh
en
t
o
n
ic
m
o
d
u
latio
n
o
r
in
ter
p
r
etab
ilit
y
is
d
esire
d
.
(
a)
(
b
)
Fig
u
r
e
1
.
C
o
m
p
a
r
is
o
n
o
f
E
DA
s
ig
n
al
d
ec
o
m
p
o
s
itio
n
m
eth
o
d
s
: (
a)
p
h
asic a
n
d
(
b
)
t
o
n
ic
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
1
,
Feb
r
u
ar
y
20
26
:
469
-
4
8
0
474
3
.
3
.
T
em
po
r
a
l seg
m
ent
a
t
io
n a
nd
t
re
nd
a
na
ly
s
is
Usi
n
g
th
e
T
VSy
m
p
d
ec
o
m
p
o
s
itio
n
m
eth
o
d
,
we
a
n
aly
ze
d
h
o
w
E
DA
ch
an
g
es
th
r
o
u
g
h
o
u
t
g
am
ep
lay
an
d
its
r
elatio
n
s
h
ip
to
p
er
c
eiv
ed
m
en
tal
wo
r
k
l
o
ad
.
As
d
em
o
n
s
tr
ated
in
Fig
u
r
e
2
,
b
o
th
p
h
asic
an
d
T
PC
s
h
o
wed
s
tr
o
n
g
p
o
s
itiv
e
co
r
r
elatio
n
s
with
g
am
ep
lay
d
u
r
atio
n
(
r
=0
.
9
1
5
an
d
r
=0
.
8
5
6
,
r
esp
ec
tiv
e
ly
;
b
o
th
ρ
<
0
.
0
0
1
)
,
in
d
icatin
g
th
at
s
y
m
p
ath
etic
a
r
o
u
s
al
ac
cu
m
u
lates
s
tead
ily
d
u
r
in
g
p
r
o
l
o
n
g
e
d
m
atch
es.
T
h
e
lin
ea
r
tr
en
d
lin
es
clea
r
ly
illu
s
tr
ate
th
is
tim
e
-
d
ep
en
d
en
t
ac
cu
m
u
latio
n
,
with
p
h
asic
p
ea
k
s
s
h
o
win
g
a
s
teep
er
g
r
ad
ien
t
co
m
p
ar
ed
to
to
n
ic
p
ea
k
s
.
T
h
e
i
n
cr
ea
s
in
g
p
h
asic
ac
tiv
ity
lik
el
y
r
e
f
lects
r
ep
ea
ted
m
o
m
en
tar
y
r
es
p
o
n
s
es
to
i
n
-
g
am
e
s
tim
u
li,
wh
ile
th
e
to
n
ic
co
m
p
o
n
en
t
m
ay
r
ep
r
esen
t
s
lo
wer
ad
ap
tatio
n
s
s
u
ch
as
s
u
s
ta
in
ed
en
g
ag
em
e
n
t,
s
tr
es
s
b
u
ild
u
p
,
o
r
p
h
y
s
io
lo
g
ical
f
atig
u
e.
T
h
e
p
lo
ts
in
Fig
u
r
e
2
r
e
v
ea
l
co
n
s
id
er
ab
le
in
ter
-
s
ess
io
n
v
ar
iab
ilit
y
ar
o
u
n
d
th
e
r
eg
r
ess
io
n
lin
es,
p
ar
ticu
lar
ly
ev
id
en
t
in
th
e
t
o
n
ic
co
m
p
o
n
en
t,
wh
er
e
d
ata
p
o
in
ts
s
h
o
w
g
r
ea
ter
d
is
p
er
s
io
n
at
lo
n
g
er
d
u
r
atio
n
s
.
T
h
is
h
eter
o
g
en
eity
s
u
g
g
ests
th
at
wh
ile
t
h
e
o
v
er
all
tem
p
o
r
al
tr
e
n
d
is
r
o
b
u
s
t,
in
d
i
v
id
u
al
m
atch
es
ex
h
ib
it
u
n
i
q
u
e
p
h
y
s
io
lo
g
ical
s
ig
n
atu
r
es
th
at
g
am
e
d
y
n
am
ics,
p
lay
e
r
s
tr
ateg
ies,
o
r
co
n
tex
tu
al
f
ac
to
r
s
m
ay
in
f
lu
e
n
ce
.
Fig
u
r
e
2
.
PPC
an
d
T
PC
as a
f
u
n
ctio
n
o
f
m
atch
d
u
r
atio
n
.
B
o
th
s
h
o
w
s
tr
o
n
g
p
o
s
itiv
e
co
r
r
elatio
n
s
(
p
h
asic: r
=0
.
9
1
5
,
to
n
ic:
r
=
0
.
8
5
6
; p
<
0
.
0
0
1
)
Ho
wev
er
,
th
e
r
elatio
n
s
h
ip
b
et
wee
n
E
DA
p
ea
k
r
ates
an
d
s
u
b
jectiv
e
wo
r
k
lo
ad
was
wea
k
.
P
h
asic
p
ea
k
r
ate
was
o
n
ly
m
a
r
g
in
ally
c
o
r
r
elate
d
with
s
elf
-
r
ep
o
r
ted
m
en
tal
lo
ad
(
r
=
0
.
0
6
9
,
ρ
=0
.
5
1
5
)
,
a
n
d
to
n
ic
p
ea
k
r
ate
was
ef
f
ec
tiv
ely
u
n
co
r
r
elate
d
(
r
=
-
0
.
0
2
3
,
ρ
=0
.
8
2
7
)
.
T
h
is
s
u
g
g
ests
th
at
p
h
y
s
io
lo
g
ical
ar
o
u
s
al
d
o
es
n
o
t
d
ir
ec
tly
alig
n
with
co
ar
s
e
m
atch
-
lev
el
wo
r
k
lo
ad
lab
els.
Sev
er
al
f
ac
to
r
s
m
ay
co
n
tr
ib
u
te,
in
clu
d
in
g
th
e
lim
ited
g
r
an
u
lar
ity
o
f
m
en
tal
lo
a
d
an
n
o
tatio
n
s
,
in
ter
-
in
d
iv
id
u
al
v
a
r
iab
ilit
y
in
au
to
n
o
m
ic
r
ea
ctiv
it
y
,
an
d
v
a
r
iatio
n
s
in
g
am
e
p
ac
in
g
o
r
p
lay
er
r
o
les.
3
.
4
.
Co
rr
el
a
t
io
n a
nd
pa
rt
ia
l
co
rr
ela
t
io
n a
na
ly
s
is
B
iv
ar
iate
Sp
ea
r
m
an
co
r
r
elatio
n
s
r
ev
ea
led
s
tr
o
n
g
p
o
s
itiv
e
ass
o
ciatio
n
s
b
etwe
en
to
tal
s
ess
io
n
d
u
r
atio
n
an
d
E
DA
p
ea
k
co
u
n
ts
f
o
r
b
o
t
h
p
h
asic
(
r
=0
.
8
3
5
,
ρ
<0
.
0
0
1
)
a
n
d
to
n
ic
co
m
p
o
n
en
ts
(
r
=
0
.
6
3
0
,
ρ
<0
.
0
0
1
)
.
T
h
ese
f
in
d
in
g
s
s
u
g
g
est
th
at
as
th
e
d
u
r
atio
n
o
f
g
am
e
p
lay
in
cr
ea
s
es,
th
e
cu
m
u
lativ
e
n
u
m
b
er
o
f
E
D
A
ev
en
ts
also
ten
d
s
to
r
is
e,
p
o
s
s
ib
ly
r
e
f
lectin
g
s
u
s
tain
ed
en
g
a
g
em
en
t,
ac
cu
m
u
latio
n
o
f
ar
o
u
s
al,
o
r
f
atig
u
e
b
u
ild
u
p
o
v
e
r
tim
e
.
Ho
wev
er
,
wh
e
n
c
o
n
tr
o
llin
g
f
o
r
s
ess
io
n
d
u
r
atio
n
,
th
e
ass
o
ciatio
n
b
etwe
en
n
o
r
m
alize
d
p
ea
k
r
ates
a
n
d
s
u
b
jectiv
e
wo
r
k
lo
ad
b
ec
am
e
m
u
ch
wea
k
er
.
Par
tial
co
r
r
elatio
n
s
b
etwe
en
m
en
tal
lo
ad
lab
el
an
d
p
h
asic
r
ate
(r=
-
0
.
0
9
2
,
ρ
=0
.
3
8
5
)
an
d
b
et
wee
n
wo
r
k
lo
ad
an
d
to
n
ic
r
ate
(
r
=
-
0
.
1
8
5
,
ρ
=0
.
0
7
8
)
wer
e
n
o
t
s
tatis
tically
s
ig
n
if
ican
t.
T
h
is
s
u
g
g
ests
th
at
in
cr
ea
s
es
in
E
DA
p
ea
k
d
en
s
it
y
o
v
er
tim
e
m
ay
b
e
m
o
r
e
attr
i
b
u
tab
le
to
tim
e
-
on
-
task
th
an
to
p
er
ce
iv
ed
co
g
n
itiv
e
d
em
an
d
.
T
h
e
lack
o
f
ass
o
ciatio
n
m
ay
also
r
ef
lect
lim
itatio
n
s
in
th
e
g
r
an
u
lar
ity
o
f
g
lo
b
al
wo
r
k
lo
ad
lab
els,
wh
ich
d
o
n
o
t c
ap
t
u
r
e
i
n
tr
a
-
m
atch
c
o
g
n
itiv
e
v
ar
iab
ilit
y
.
3
.
5
.
M
ix
ed
-
ef
f
ec
t
s
m
o
delin
g
a
nd
bo
o
t
s
t
ra
pp
ed
v
a
lid
a
t
io
n
T
o
f
u
r
th
e
r
test
th
e
p
r
ed
ictiv
e
r
o
le
o
f
wo
r
k
lo
ad
o
n
E
DA
d
y
n
am
ics,
L
MM
s
wer
e
f
itted
u
s
in
g
p
h
asic
an
d
to
n
ic
r
ate
p
e
r
m
in
u
te
as
d
ep
en
d
e
n
t
v
ar
iab
les,
with
wo
r
k
lo
ad
as
a
f
ix
ed
ef
f
ec
t
an
d
p
lay
er
as
a
r
an
d
o
m
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
C
la
s
s
i
fyin
g
men
ta
l wo
r
klo
a
d
o
f e
s
p
o
r
ts
p
la
ye
r
s
u
s
in
g
ma
ch
in
e
lea
r
n
in
g
(
A
is
y
A
l F
a
w
w
a
z
)
475
in
ter
ce
p
t.
T
h
e
m
o
d
el
f
o
r
p
h
asi
c
ac
tiv
ity
p
r
o
d
u
ce
d
a
n
o
n
-
s
ig
n
if
ican
t
co
ef
f
icien
t (
β=0
.
0
3
6
,
ρ
=0
.
9
2
9
)
,
in
d
icatin
g
n
o
m
ea
n
in
g
f
u
l
r
elatio
n
s
h
i
p
b
e
twee
n
wo
r
k
lo
a
d
class
an
d
p
h
asic
ac
tiv
atio
n
r
ate.
Me
a
n
wh
i
le,
th
e
to
n
ic
m
o
d
el
y
ield
ed
a
m
ar
g
in
ally
n
o
n
-
s
i
g
n
if
ican
t
ad
v
er
s
e
ef
f
ec
t
(
β=
-
0
.
2
5
3
,
ρ
=0
.
0
9
3
)
,
h
in
tin
g
at
a
p
o
ten
tial
in
v
er
s
e
ass
o
ciatio
n
b
etwe
en
h
ig
h
er
w
o
r
k
lo
ad
an
d
to
n
ic
ac
tiv
ity
.
Ho
wev
er
,
th
is
d
id
n
o
t
m
ee
t
co
n
v
en
tio
n
al
th
r
esh
o
ld
s
f
o
r
s
tatis
tical
s
ig
n
if
ican
ce
.
T
o
a
s
s
e
s
s
t
h
e
r
o
b
u
s
t
n
e
s
s
o
f
th
e
s
e
f
i
n
d
i
n
g
s
,
b
o
o
t
s
t
r
ap
r
e
s
a
m
p
l
i
n
g
w
a
s
u
s
e
d
to
e
s
t
i
m
a
t
e
th
e
s
t
a
b
i
l
i
ty
o
f
th
e
p
h
a
s
i
c
m
o
d
e
l
’
s
wo
r
k
l
o
a
d
co
e
f
f
i
c
i
en
t
.
Ac
r
o
s
s
1
,
0
0
0
it
e
r
a
t
i
o
n
s
,
th
e
m
e
an
b
o
o
t
s
t
r
a
p
p
e
d
β
1
w
a
s
0
.
0
3
9
7
,
w
i
t
h
a
9
5
%
co
n
f
i
d
en
c
e
in
t
e
r
v
a
l
r
an
g
in
g
f
r
o
m
-
0
.
8
7
0
5
t
o
0
.
8
2
1
6
.
T
h
i
s
in
t
e
r
v
a
l
co
n
s
i
s
te
n
t
l
y
s
p
a
n
n
e
d
z
e
r
o
,
f
u
r
th
e
r
s
u
p
p
o
r
t
i
n
g
th
e
i
n
t
er
p
r
e
t
a
t
i
o
n
t
h
a
t
an
y
wo
r
k
l
o
a
d
e
f
f
e
c
t
o
n
E
D
A
p
e
a
k
r
a
t
e
s
i
s
w
ea
k
an
d
u
n
s
t
a
b
l
e
i
n
t
h
i
s
d
a
t
a
s
e
t.
3
.
6
.
M
a
t
ch
o
rder
,
r
o
le,
a
nd
da
y
-
lev
el
ef
f
ec
t
s
o
n E
DA
dy
na
m
ics
De
s
p
ite
ex
p
ec
ta
tio
n
s
th
at
co
n
tex
tu
a
l
g
am
ep
lay
v
ar
ia
b
le
s
m
ig
h
t
s
h
ap
e
E
D
A,
s
ta
ti
s
ti
ca
l
m
o
d
e
lin
g
r
ev
ea
le
d
o
n
ly
m
o
d
e
s
t
an
d
i
n
co
n
s
i
s
ten
t
e
f
f
e
ct
s
.
A
s
s
h
o
wn
in
F
ig
u
r
e
3
,
p
h
a
s
ic
p
e
ak
r
at
es
f
lu
c
tu
a
ted
ac
r
o
s
s
m
at
ch
o
r
d
er
a
n
d
t
o
u
r
n
am
en
t
d
ay
s
,
wi
th
a
s
l
ig
h
t
in
cr
ea
s
e
o
b
s
er
v
ed
o
n
d
ay
1
r
el
at
iv
e
to
d
ay
0
(
β=0
.
1
4
7
,
p
=
0
.
0
3
6
)
.
Ho
w
ev
er
,
t
h
i
s
p
a
tt
er
n
w
as
n
o
t
r
ep
li
ca
t
ed
o
n
d
a
y
2
,
an
d
m
a
tch
o
r
d
er
i
t
s
e
lf
d
id
n
o
t
y
ie
ld
s
i
g
n
if
ic
an
t
e
f
f
e
ct
s
(
p
=0
.
5
1
0
)
.
T
o
n
ic
p
ea
k
r
a
te
s
f
o
l
lo
w
ed
s
i
m
i
lar
ly
n
o
n
-
s
ig
n
if
ic
an
t
tr
en
d
s
(
p
=0
.
6
0
1
)
,
r
ein
f
o
r
c
in
g
t
h
e
in
t
er
p
r
e
ta
t
io
n
th
a
t
s
y
m
p
a
th
et
ic
a
ct
iv
at
io
n
i
s
n
o
t
co
n
s
i
s
ten
tly
m
o
d
u
l
ated
b
y
g
am
ep
l
ay
s
eq
u
en
c
e
o
r
d
a
ily
p
r
o
g
r
e
s
s
io
n
.
Fig
u
r
e
3
.
I
n
f
lu
e
n
ce
o
f
to
u
r
n
a
m
en
t d
ay
,
m
atch
o
r
d
er
,
an
d
p
l
ay
er
r
o
le
o
n
p
h
asic a
ctiv
atio
n
an
d
wo
r
k
lo
ad
p
er
ce
p
tio
n
On
th
e
r
ig
h
t
p
an
el
o
f
Fig
u
r
e
3
,
we
v
is
u
alize
d
m
e
n
tal
lo
ad
d
is
tr
ib
u
tio
n
b
y
p
lay
e
r
r
o
le.
W
h
ile
s
o
m
e
r
o
les,
s
u
ch
as su
p
p
o
r
t a
n
d
attac
k
d
am
ag
e
ca
r
r
y
(
ADC),
ex
h
ib
ited
h
ig
h
er
f
r
eq
u
e
n
cies o
f
lo
w
-
lo
ad
s
ess
io
n
s
,
an
d
ju
n
g
le
an
d
m
id
r
o
les
h
ad
s
lig
h
tly
m
o
r
e
h
ig
h
-
l
o
ad
s
ess
io
n
s
,
t
h
ese
d
if
f
er
en
ce
s
wer
e
n
o
t
s
tatis
tically
s
ig
n
if
ican
t.
No
n
eth
eless
,
th
e
r
o
le
-
b
ased
tr
en
d
s
s
u
g
g
est
th
at
task
c
o
m
p
le
x
ity
o
r
r
ea
l
-
tim
e
d
ec
is
io
n
-
m
a
k
in
g
d
em
an
d
s
m
ay
s
u
b
tly
in
f
lu
e
n
ce
p
er
ce
iv
ed
wo
r
k
lo
ad
.
T
ak
e
n
to
g
eth
er
,
th
e
r
e
s
u
lts
s
u
g
g
est
th
at
b
r
o
ad
co
n
te
x
tu
al
f
ea
tu
r
es,
s
u
ch
as
d
ay
,
m
atch
o
r
d
er
,
o
r
p
r
ed
e
f
in
ed
r
o
les,
o
f
f
er
lim
ited
ex
p
l
an
ato
r
y
p
o
wer
f
o
r
s
y
m
p
ath
etic
ar
o
u
s
al
p
atter
n
s
.
T
h
is
u
n
d
er
s
co
r
es
th
e
im
p
o
r
ta
n
ce
o
f
ac
co
u
n
tin
g
f
o
r
with
in
-
m
atch
d
y
n
am
ics
o
r
in
d
iv
id
u
a
l
d
if
f
er
en
ce
s
wh
en
m
o
d
elin
g
E
DA
r
esp
o
n
s
es in
c
o
m
p
etitiv
e
s
ettin
g
s
.
3
.
6
.
M
a
chine le
a
rning
cla
s
s
i
f
ica
t
io
n
T
o
co
m
p
ar
e
t
h
e
ef
f
ec
ts
o
f
e
v
alu
atio
n
s
tr
ateg
y
an
d
f
ea
tu
r
e
s
elec
tio
n
o
n
m
o
d
el
p
er
f
o
r
m
an
ce
,
th
e
r
esu
lts
f
r
o
m
b
o
th
th
e
p
r
e
v
io
u
s
s
tr
atif
ied
k
-
f
o
ld
s
etu
p
an
d
t
h
e
cu
r
r
en
t
L
OSO
v
alid
atio
n
ar
e
s
u
m
m
ar
ize
d
in
T
ab
le
1
.
Un
d
e
r
s
tr
atif
ied
k
-
f
o
l
d
,
wh
er
e
with
in
-
s
u
b
ject
s
am
p
l
es
co
u
ld
a
p
p
ea
r
in
b
o
th
tr
ain
in
g
an
d
test
in
g
f
o
ld
s
,
class
if
icatio
n
m
etr
ics
wer
e
n
o
tab
ly
h
ig
h
er
ac
r
o
s
s
all
m
o
d
els.
SVM
an
d
LR
b
o
th
ac
h
iev
ed
8
1
.
9
7
%
ac
cu
r
ac
y
with
an
AUC
o
f
0
.
8
8
2
,
s
u
g
g
e
s
tin
g
s
tr
o
n
g
with
in
-
s
u
b
ject
p
r
ed
ictio
n
ca
p
ab
ilit
ies.
Ho
wev
e
r
,
th
ese
v
alu
es
m
a
y
o
v
er
esti
m
ate
g
en
er
aliza
b
ilit
y
d
u
e
to
p
o
ten
tial d
ata
leak
a
g
e.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
1
,
Feb
r
u
ar
y
20
26
:
469
-
4
8
0
476
T
ab
le
1
.
C
lass
if
ier
p
er
f
o
r
m
a
n
c
e
co
m
p
ar
is
o
n
b
etwe
en
s
tr
atif
ied
k
-
f
o
ld
an
d
L
OSO
u
s
in
g
E
D
A
an
d
HR
V
M
o
d
e
l
P
r
e
v
i
o
u
s s
t
u
d
y
[
1
1
]
(
ED
A
+
H
R
V
,
s
t
r
a
t
i
f
i
e
d
k
-
f
o
l
d
)
C
u
r
r
e
n
t
st
u
d
y
(
ED
A
o
n
l
y
,
LO
S
O
)
C
u
r
r
e
n
t
st
u
d
y
(
ED
A
+
H
R
V
,
LO
S
O
)
LR
A
c
c
u
r
a
c
y
:
8
1
.
9
7
A
U
C
:
0
.
8
8
2
F1
-
sc
o
r
e
:
-
A
c
c
u
r
a
c
y
:
7
6
.
8
8
A
U
C
:
0
.
7
8
8
F1
-
sc
o
r
e
:
0
.
6
1
7
A
c
c
u
r
a
c
y
:
6
9
.
6
4
A
U
C
:
0
.
7
6
9
F1
-
sc
o
r
e
:
0
.
4
7
3
S
V
M
(
R
B
F
)
A
c
c
u
r
a
c
y
:
8
1
.
9
7
A
U
C
:
0
.
8
8
2
F1
-
sc
o
r
e
:
-
A
c
c
u
r
a
c
y
:
7
5
.
3
3
A
U
C
:
0
.
7
8
1
F1
-
sc
o
r
e
:
0
.
6
1
7
A
c
c
u
r
a
c
y
:
6
3
.
2
7
A
U
C
:
0
.
7
2
2
F1
-
sc
o
r
e
:
0
.
3
5
9
M
LP
A
c
c
u
r
a
c
y
:
8
2
.
0
3
A
U
C
:
0
.
8
4
0
F1
-
sc
o
r
e
:
-
A
c
c
u
r
a
c
y
:
7
0
.
7
1
A
U
C
:
0
.
6
6
4
F1
-
sc
o
r
e
:
0
.
4
3
7
A
c
c
u
r
a
c
y
:
6
5
.
5
1
A
U
C
:
0
.
5
5
9
F1
-
sc
o
r
e
:
0
.
2
8
3
X
G
B
o
o
st
A
c
c
u
r
a
c
y
:
7
9
.
6
3
A
U
C
:
0
.
8
7
4
F1
-
sc
o
r
e
:
-
A
c
c
u
r
a
c
y
:
6
9
.
4
4
A
U
C
:
0
.
7
6
7
F1
-
sc
o
r
e
:
0
.
4
8
2
A
c
c
u
r
a
c
y
:
6
7
.
3
0
%
A
U
C
:
0
.
6
6
9
F1
-
sc
o
r
e
:
0
.
4
6
2
I
n
co
n
tr
ast,
th
e
L
OSO
f
r
am
e
wo
r
k
,
wh
ich
en
s
u
r
es
s
u
b
ject
-
l
ev
el
s
ep
ar
atio
n
d
u
r
i
n
g
test
in
g
,
p
r
o
d
u
ce
d
m
o
r
e
co
n
s
er
v
ativ
e
b
u
t
ar
g
u
a
b
l
y
m
o
r
e
r
ea
lis
tic
m
etr
ics.
T
h
e
b
est
m
o
d
el
u
n
d
er
th
ese
co
n
d
it
io
n
s
was
L
R
u
s
in
g
o
n
ly
E
DA
p
ea
k
f
ea
tu
r
es,
wh
ich
ac
h
iev
ed
7
6
.
8
8
%
ac
c
u
r
ac
y
an
d
an
AUC
o
f
0
.
7
8
8
.
No
tab
ly
,
wh
en
HR
V
f
ea
tu
r
es
wer
e
in
cl
u
d
ed
alo
n
g
s
id
e
E
DA,
p
e
r
f
o
r
m
an
ce
d
ec
lin
ed
ac
r
o
s
s
all
class
if
ier
s
.
Fo
r
ex
am
p
le,
t
h
e
s
am
e
L
R
m
o
d
el
ac
h
iev
e
d
an
ac
c
u
r
ac
y
o
f
6
9
.
6
4
%
an
d
F1
-
s
co
r
e
o
f
0
.
4
7
3
.
Similar
d
eg
r
a
d
atio
n
s
wer
e
o
b
s
er
v
ed
in
SVM,
ML
P,
an
d
XGBo
o
s
t,
u
n
d
er
s
co
r
in
g
th
e
p
o
s
s
ib
ilit
y
th
at
HR
V
s
ig
n
als,
wh
ile
p
h
y
s
io
lo
g
ically
m
ea
n
in
g
f
u
l,
m
ay
in
tr
o
d
u
ce
s
u
b
ject
-
s
p
ec
if
ic
n
o
is
e
th
at
h
in
d
e
r
s
cr
o
s
s
-
in
d
iv
i
d
u
al
g
en
e
r
aliza
tio
n
.
4.
DIS
CU
SS
I
O
N
T
h
is
s
tu
d
y
ex
a
m
in
ed
w
h
eth
er
E
DA
p
ea
k
c
o
u
n
ts
ar
e
m
o
r
e
r
e
f
lectiv
e
o
f
m
en
tal
wo
r
k
lo
ad
o
r
p
r
im
ar
il
y
d
r
iv
en
b
y
task
d
u
r
atio
n
.
Acr
o
s
s
9
2
g
am
ep
lay
s
ess
io
n
s
,
we
f
o
u
n
d
th
at
b
o
th
PPC
an
d
T
PC
wer
e
s
ig
n
if
ican
tly
co
r
r
elate
d
with
s
ess
io
n
len
g
th
.
Ho
wev
er
,
o
n
ce
d
u
r
atio
n
was
s
tatis
tical
ly
co
n
tr
o
lled
,
th
e
as
s
o
ciatio
n
b
etwe
en
E
DA
p
ea
k
r
ates a
n
d
s
elf
-
r
ep
o
r
ted
wo
r
k
lo
ad
wea
k
en
e
d
s
u
b
s
tan
tially
.
T
h
is
p
atter
n
s
u
g
g
ests
t
h
at
tim
e
-
on
-
task
is
a
d
o
m
in
an
t
d
r
iv
e
r
o
f
E
DA
p
ea
k
ac
cu
m
u
latio
n
,
an
d
th
at
E
DA
m
etr
ics
alo
n
e
m
ay
o
v
er
esti
m
ate
co
g
n
itiv
e
d
em
an
d
if
tem
p
o
r
al
ef
f
ec
ts
ar
e
n
o
t ta
k
en
in
to
ac
co
u
n
t.
T
h
e
r
esu
lts
alig
n
with
p
r
i
o
r
f
i
n
d
in
g
s
th
at
lin
k
s
u
s
tain
ed
s
y
m
p
ath
etic
ac
tiv
atio
n
with
p
r
o
l
o
n
g
ed
task
en
g
ag
em
e
n
t
[
1
9
]
,
[
2
0
]
.
Ho
wev
er
,
u
n
lik
e
s
tu
d
ies
th
at
in
ter
p
r
et
E
DA
in
cr
ea
s
es
as
a
d
ir
ec
t
r
ef
lectio
n
o
f
co
g
n
itiv
e
ef
f
o
r
t
[
2
1
]
,
[
2
2
]
,
o
u
r
ap
p
r
o
ac
h
ex
p
licitly
s
ep
ar
ated
d
u
r
atio
n
ef
f
ec
ts
th
r
o
u
g
h
p
ar
ti
al
co
r
r
elatio
n
s
an
d
m
ix
ed
-
ef
f
ec
ts
m
o
d
elin
g
.
T
h
e
an
aly
s
is
s
h
o
wed
th
at
p
h
asic
an
d
to
n
ic
r
ates
d
id
n
o
t
s
tr
o
n
g
l
y
p
r
e
d
ict
wo
r
k
l
o
ad
lab
els,
in
d
icatin
g
th
at
tr
an
s
ien
t
E
DA
f
lu
ctu
atio
n
s
ar
e
m
o
r
e
te
m
p
o
r
ally
s
tr
u
ctu
r
ed
th
a
n
co
g
n
itiv
ely
s
p
ec
if
ic.
W
e
f
u
r
th
er
in
v
esti
g
ated
co
n
te
x
tu
al
v
ar
iab
les,
in
clu
d
in
g
g
a
m
ep
lay
o
r
d
e
r
,
to
u
r
n
am
e
n
t
d
ay
,
an
d
p
lay
e
r
r
o
le.
T
h
e
r
eg
r
ess
io
n
m
o
d
els
r
ev
ea
led
n
o
m
ea
n
i
n
g
f
u
l
im
p
ac
t
o
f
m
atch
s
eq
u
en
ce
o
r
d
ay
p
r
o
g
r
ess
io
n
o
n
E
DA
p
ea
k
r
ates.
Alth
o
u
g
h
r
o
le
-
b
ased
co
m
p
ar
is
o
n
s
s
h
o
wed
s
lig
h
t
v
ar
iatio
n
s
in
wo
r
k
lo
a
d
lab
e
ls
,
th
e
d
if
f
er
en
ce
s
wer
e
n
o
t
s
tatis
t
ically
r
o
b
u
s
t.
T
h
is
s
u
g
g
ests
th
at
in
ter
-
s
ess
io
n
s
ch
ed
u
lin
g
o
r
g
a
m
ep
lay
r
es
p
o
n
s
ib
ilit
ies
h
av
e
a
lim
ited
in
f
lu
en
ce
o
n
p
h
y
s
io
lo
g
ical
wo
r
k
lo
ad
m
ar
k
e
r
s
,
an
d
th
at
with
in
-
m
atch
d
y
n
a
m
ics
m
ay
p
la
y
a
m
o
r
e
s
ig
n
if
ican
t r
o
le.
T
o
ass
ess
p
r
ed
ictiv
e
u
tili
ty
,
we
ap
p
lied
f
o
u
r
ML
class
if
ier
s
with
L
OSO
cr
o
s
s
-
v
alid
at
io
n
.
W
h
en
tr
ain
ed
u
s
in
g
o
n
l
y
PPC
an
d
T
PC
,
L
R
ac
h
iev
ed
th
e
b
est
r
esu
lts
,
with
7
6
.
9
%
ac
cu
r
ac
y
an
d
an
AUC
o
f
0
.
7
8
8
.
SVM
s
h
o
wed
s
im
ilar
p
er
f
o
r
m
an
ce
,
wh
ile
MLP
an
d
XGBo
o
s
t
m
o
d
els
u
n
d
er
p
er
f
o
r
m
ed
.
Ad
d
in
g
HR
V
f
ea
tu
r
e
s
d
id
n
o
t
im
p
r
o
v
e
class
if
icatio
n
p
er
f
o
r
m
an
ce
.
I
n
s
tead
,
it
led
t
o
co
n
s
is
ten
t
d
ec
lin
es
i
n
F1
-
s
c
o
r
es
an
d
in
cr
ea
s
ed
u
n
ce
r
tain
ty
ac
r
o
s
s
all
m
o
d
els.
T
h
is
f
in
d
in
g
c
o
n
tr
asts
with
o
u
r
p
r
ev
io
u
s
s
tu
d
y
[
1
1
]
,
[
2
3
]
,
w
h
ich
u
s
ed
s
tr
atif
ied
k
-
f
o
ld
v
alid
atio
n
an
d
in
clu
d
ed
all
E
DA
an
d
HR
V
f
ea
tu
r
es.
I
n
th
at
s
etu
p
,
SVM
a
ch
iev
ed
an
ac
cu
r
ac
y
o
f
8
1
.
9
7
%
an
d
an
AUC
o
f
0
.
8
8
2
.
T
h
e
d
is
cr
ep
an
cy
lik
ely
s
tem
s
f
r
o
m
th
e
v
alid
atio
n
s
tr
a
teg
y
.
Stra
tifie
d
k
-
f
o
ld
allo
ws
d
ata
f
r
o
m
th
e
s
am
e
s
u
b
ject
to
ap
p
ea
r
in
b
o
th
tr
ai
n
in
g
an
d
test
in
g
f
o
ld
s
,
p
o
te
n
t
ially
in
f
latin
g
m
o
d
el
p
er
f
o
r
m
an
ce
[
2
3
]
,
[
2
4
]
.
I
n
co
n
tr
ast,
th
e
L
OSO
p
r
o
t
o
co
l
e
n
f
o
r
ce
s
s
u
b
ject
-
lev
el
i
n
d
ep
e
n
d
en
ce
an
d
b
etter
r
ef
lects
r
ea
l
-
w
o
r
ld
g
e
n
er
aliza
tio
n
s
ce
n
ar
io
s
wh
er
e
ca
lib
r
atio
n
d
a
ta
m
ay
b
e
u
n
av
ailab
le.
T
h
e
d
r
o
p
in
p
er
f
o
r
m
an
ce
a
f
ter
in
clu
d
in
g
HR
V
f
ea
tu
r
es
s
u
g
g
ests
th
at
th
ese
m
etr
ics,
alth
o
u
g
h
p
h
y
s
io
lo
g
ically
v
alid
,
m
ay
en
co
d
e
in
d
iv
id
u
al
-
s
p
ec
if
ic
tr
aits
th
at
r
ed
u
ce
cr
o
s
s
-
s
u
b
ject
g
en
e
r
aliza
b
ilit
y
.
HR
V
is
in
f
lu
en
ce
d
b
y
v
a
r
io
u
s
in
d
iv
id
u
al
-
s
p
ec
if
ic
f
ac
to
r
s
u
ch
as
g
e
n
d
er
,
ag
e,
an
d
cu
r
r
en
t
p
h
y
s
ic
al
co
n
d
itio
n
,
wh
ic
h
ca
n
lead
to
v
ar
iab
ilit
y
in
d
if
f
er
en
t
in
d
iv
id
u
als
[
2
5
]
,
[
2
6
]
.
E
DA
-
d
er
iv
ed
f
ea
tu
r
es,
esp
ec
ially
T
PC
an
d
PP
C
,
ap
p
ea
r
m
o
r
e
co
n
s
is
ten
t
an
d
s
u
b
ject
-
in
v
ar
ia
n
t,
m
ak
in
g
th
em
b
etter
s
u
ited
f
o
r
wo
r
k
lo
a
d
m
o
d
elin
g
in
d
y
n
am
ic
an
d
h
eter
o
g
en
e
o
u
s
p
o
p
u
latio
n
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
C
la
s
s
i
fyin
g
men
ta
l wo
r
klo
a
d
o
f e
s
p
o
r
ts
p
la
ye
r
s
u
s
in
g
ma
ch
in
e
lea
r
n
in
g
(
A
is
y
A
l F
a
w
w
a
z
)
477
T
h
ese
f
in
d
in
g
s
h
ig
h
lig
h
t
th
e
im
p
o
r
tan
ce
o
f
b
o
th
f
ea
tu
r
e
s
elec
tio
n
an
d
v
alid
atio
n
s
t
r
ateg
y
in
p
h
y
s
io
lo
g
ical
co
m
p
u
tin
g
.
C
o
m
p
ac
t
E
DA
-
b
ased
in
d
icato
r
s
ca
n
p
r
o
v
id
e
r
eliab
le
in
s
ig
h
ts
in
to
co
g
n
itiv
e
s
tate
with
o
u
t th
e
co
m
p
lex
ity
o
r
v
ar
i
ab
ilit
y
in
tr
o
d
u
ce
d
b
y
m
u
ltimo
d
al
s
ig
n
als.
Fu
tu
r
e
wo
r
k
s
h
o
u
l
d
ex
p
lo
r
e
s
eg
m
en
t
-
lev
el
wo
r
k
lo
ad
la
b
elin
g
,
in
c
o
r
p
o
r
ate
b
eh
av
i
o
r
al
an
d
c
o
n
t
ex
tu
al
d
ata,
an
d
e
v
alu
ate
ad
ap
tiv
e
m
o
d
els
th
at
p
er
s
o
n
alize
p
r
e
d
ictio
n
s
with
o
u
t r
eq
u
ir
in
g
ex
p
licit c
alib
r
atio
n
.
5.
CO
NCLU
SI
O
N
T
h
is
s
tu
d
y
d
em
o
n
s
tr
ates
th
at
E
DA
p
ea
k
co
u
n
ts
,
p
ar
ticu
la
r
ly
p
h
asic
r
esp
o
n
s
es,
a
r
e
m
o
r
e
s
tr
o
n
g
l
y
in
f
lu
en
ce
d
b
y
task
d
u
r
atio
n
t
h
an
b
y
m
en
tal
wo
r
k
l
o
ad
.
W
h
ile
r
aw
p
ea
k
ac
cu
m
u
latio
n
alig
n
s
clo
s
ely
with
g
am
ep
lay
len
g
th
,
th
eir
n
o
r
m
alize
d
r
ates
s
h
o
w
wea
k
an
d
in
co
n
s
is
ten
t
as
s
o
ciatio
n
s
with
s
elf
-
r
ep
o
r
ted
co
g
n
itiv
e
d
em
an
d
.
E
v
e
n
af
ter
co
n
tr
o
llin
g
f
o
r
d
u
r
atio
n
,
b
o
t
h
s
tatis
tica
l
an
d
ML
an
aly
s
es
r
ev
ea
l
th
at
E
DA
p
ea
k
s
h
av
e
lim
ited
p
r
ed
ictiv
e
v
alu
e
f
o
r
w
o
r
k
lo
ad
class
if
icatio
n
ac
r
o
s
s
in
d
iv
id
u
als.
T
h
ese
f
in
d
in
g
s
s
u
g
g
est
th
at
E
DA
p
ea
k
co
u
n
ts
ar
e
p
r
e
d
o
m
in
a
n
tly
tim
e
-
d
ep
en
d
en
t
a
n
d
s
h
o
u
ld
b
e
in
ter
p
r
eted
ca
u
tio
u
s
ly
as
s
tan
d
a
lo
n
e
in
d
icato
r
s
o
f
m
en
tal
wo
r
k
l
o
ad
.
Fo
r
r
o
b
u
s
t
a
n
d
g
e
n
er
aliza
b
le
m
o
d
elin
g
,
f
u
tu
r
e
s
y
s
tem
s
m
u
s
t
ac
co
u
n
t
f
o
r
tem
p
o
r
al
s
tr
u
ctu
r
e
an
d
p
r
i
o
r
itize
v
alid
atio
n
p
r
o
t
o
co
ls
th
at
r
ef
lect
r
ea
l
-
wo
r
ld
v
ar
iab
ilit
y
.
ACK
NO
WL
E
DG
E
M
E
NT
S
T
h
e
au
th
o
r
s
th
an
k
t
h
e
Facu
lty
o
f
Scien
ce
an
d
T
ec
h
n
o
lo
g
y
,
U
n
iv
er
s
itas
Air
lan
g
g
a
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
T
h
e
au
th
o
r
s
th
an
k
th
e
Facu
lt
y
o
f
Scien
ce
an
d
T
ec
h
n
o
lo
g
y
,
Un
iv
er
s
itas
Air
lan
g
g
a,
f
o
r
f
u
n
d
in
g
th
is
r
esear
ch
u
n
d
er
t
h
e
Air
lan
g
g
a
R
esear
ch
Fu
n
d
(
I
n
te
r
n
at
io
n
al
R
esear
ch
Netwo
r
k
)
with
g
r
an
t
Nu
m
b
er
1
6
6
9
/UN3
.
L
PP
M/PT
.
0
1
.
0
3
/2
0
2
3
.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
T
h
is
jo
u
r
n
al
u
s
es
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
iv
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
llab
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
Ais
y
Al
Fawwa
z
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
Osma
lin
a
Nu
r
R
ah
m
a
✓
✓
✓
✓
✓
✓
Say
y
id
u
l I
s
tig
h
f
a
r
I
ttaq
illah
✓
✓
✓
✓
An
g
elin
e
Sh
an
e
Ku
rn
iaw
a
n
✓
✓
✓
R
ev
ita
No
v
ian
ti Pu
tr
i
✓
✓
✓
R
ich
a
Var
y
an
✓
✓
✓
Au
r
a
Ad
in
d
a
✓
✓
✓
Kh
u
s
n
u
l A
in
✓
✓
✓
✓
✓
✓
R
if
ai
C
h
ai
✓
✓
✓
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
s
i
s
I
:
I
n
v
e
s
t
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
s
i
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
Au
th
o
r
s
s
tate
n
o
co
n
f
lict o
f
in
t
er
est.
E
T
H
I
CAL AP
P
RO
V
AL
T
h
is
s
tu
d
y
u
tili
ze
d
s
ec
o
n
d
ar
y
d
ata
f
r
o
m
th
e
p
u
b
licly
a
v
ailab
le
d
ataset.
E
th
ical
ap
p
r
o
v
al
an
d
in
f
o
r
m
e
d
co
n
s
en
t
wer
e
o
b
tain
ed
b
y
t
h
e
o
r
i
g
in
al
au
th
o
r
s
d
u
r
in
g
d
ata
c
o
llectio
n
.
As
th
is
r
esear
ch
in
v
o
lv
ed
o
n
l
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
1
,
Feb
r
u
ar
y
20
26
:
469
-
4
8
0
478
s
ec
o
n
d
ar
y
an
al
y
s
is
o
f
p
u
b
lic
ly
ac
ce
s
s
ib
le
d
ata
an
d
d
id
n
o
t
in
clu
d
e
an
y
d
ir
ec
t
in
ter
ac
tio
n
with
h
u
m
an
p
ar
ticip
an
ts
,
ad
d
itio
n
al
et
h
ical
ap
p
r
o
v
al
was n
o
t r
e
q
u
ir
ed
.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
eSp
o
r
t
s
en
s
o
r
s
d
ataset
is
p
r
o
v
id
e
d
in
csv
.
f
o
r
m
a
t
an
d
o
p
en
ly
a
v
ailab
le
in
Gith
u
b
at
h
ttp
s
://g
ith
u
b
.
co
m
/s
m
er
d
o
v
/e
Sp
o
r
ts
_
Sen
s
o
r
s
_
Data
s
et.
RE
F
E
R
E
NC
E
S
[
1
]
B
.
W
a
t
so
n
e
t
a
l
.
,
“
Es
p
o
r
t
s
a
n
d
h
i
g
h
p
e
r
f
o
r
ma
n
c
e
H
C
I
,
”
i
n
Ex
t
e
n
d
e
d
A
b
s
t
ra
c
t
s
o
f
t
h
e
2
0
2
1
C
H
I
C
o
n
f
e
re
n
c
e
o
n
H
u
m
a
n
F
a
c
t
o
rs
i
n
C
o
m
p
u
t
i
n
g
S
y
s
t
e
m
s
,
N
e
w
Y
o
r
k
,
U
n
i
t
e
d
S
t
a
t
e
s
:
A
C
M
,
M
a
y
2
0
2
1
,
p
p
.
1
–
5
,
d
o
i
:
1
0
.
1
1
4
5
/
3
4
1
1
7
6
3
.
3
4
4
1
3
1
3
.
[
2
]
I
.
V.
H
i
l
v
o
o
r
d
e
,
“
E
d
i
t
o
r
i
a
l
:
e
s
p
o
r
t
s
a
n
d
d
i
g
i
t
a
l
i
z
a
t
i
o
n
o
f
sp
o
r
t
s
,
”
Fr
o
n
t
i
e
rs
i
n
S
p
o
rt
s
a
n
d
A
c
t
i
v
e
L
i
v
i
n
g
,
v
o
l
.
4
,
S
e
p
.
2
0
2
2
,
d
o
i
:
1
0
.
3
3
8
9
/
f
s
p
o
r
.
2
0
2
2
.
1
0
4
0
4
6
8
.
[
3
]
D
.
E
k
d
a
h
l
,
I
.
V.
H
i
l
v
o
o
r
d
e
,
Z.
A
.
R
u
c
i
ń
s
k
a
,
a
n
d
S
.
R
a
v
n
,
“
E
d
i
t
o
r
i
a
l
:
w
h
a
t
i
s
e
s
p
o
r
t
s
p
e
r
f
o
r
m
a
n
c
e
?
,
”
Fr
o
n
t
i
e
rs
i
n
S
p
o
rt
s
a
n
d
Ac
t
i
v
e
L
i
v
i
n
g
,
v
o
l
.
6
,
D
e
c
.
2
0
2
4
,
d
o
i
:
1
0
.
3
3
8
9
/
f
sp
o
r
.
2
0
2
4
.
1
5
3
8
6
8
6
.
[
4
]
B
.
T.
S
h
a
r
p
e
e
t
a
l
.
,
“
R
e
a
p
p
r
a
i
sa
l
a
n
d
mi
n
d
s
e
t
i
n
t
e
r
v
e
n
t
i
o
n
s
o
n
p
r
e
ss
u
r
i
s
e
d
e
s
p
o
r
t
p
e
r
f
o
r
ma
n
c
e
,
”
Ap
p
l
i
e
d
Ps
y
c
h
o
l
o
g
y
,
v
o
l
.
7
3
,
n
o
.
4
,
p
p
.
2
1
7
8
–
2
1
9
9
,
O
c
t
.
2
0
2
4
,
d
o
i
:
1
0
.
1
1
1
1
/
a
p
p
s.
1
2
5
4
4
.
[
5
]
O
.
Le
i
s,
B
.
T
.
S
h
a
r
p
e
,
V
.
P
e
l
i
k
a
n
,
J.
F
r
i
t
s
c
h
,
A
.
R
.
N
i
c
h
o
l
l
s
,
a
n
d
D
.
P
o
u
l
u
s,
“
S
t
r
e
ss
o
r
s
a
n
d
c
o
p
i
n
g
st
r
a
t
e
g
i
e
s
i
n
e
s
p
o
r
t
s:
a
sy
st
e
ma
t
i
c
r
e
v
i
e
w
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
Re
v
i
e
w
o
f
S
p
o
r
t
a
n
d
E
x
e
rci
s
e
Ps
y
c
h
o
l
o
g
y
,
p
p
.
1
–
3
1
,
A
u
g
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
8
0
/
1
7
5
0
9
8
4
X
.
2
0
2
4
.
2
3
8
6
5
2
8
.
[
6
]
C
.
R
e
a
l
e
e
t
a
l
.
,
“
D
e
c
i
si
o
n
-
m
a
k
i
n
g
d
u
r
i
n
g
h
i
g
h
-
r
i
s
k
e
v
e
n
t
s:
a
sy
s
t
e
mat
i
c
l
i
t
e
r
a
t
u
r
e
r
e
v
i
e
w
,
”
J
o
u
r
n
a
l
o
f
C
o
g
n
i
t
i
v
e
En
g
i
n
e
e
ri
n
g
a
n
d
D
e
c
i
s
i
o
n
Ma
k
i
n
g
,
v
o
l
.
1
7
,
n
o
.
2
,
p
p
.
1
8
8
–
2
1
2
,
Ju
n
.
2
0
2
3
,
d
o
i
:
1
0
.
1
1
7
7
/
1
5
5
5
3
4
3
4
2
2
1
1
4
7
4
1
5
.
[
7
]
I
.
T.
P
a
v
l
i
d
i
s,
T
.
C
h
a
s
p
a
r
i
,
a
n
d
D
.
M
c
D
u
f
f
,
“
E
d
i
t
o
r
i
a
l
:
s
p
e
c
i
a
l
i
ssu
e
o
n
u
n
o
b
t
r
u
s
i
v
e
p
h
y
si
o
l
o
g
i
c
a
l
me
a
s
u
r
e
men
t
m
e
t
h
o
d
s
f
o
r
a
f
f
e
c
t
i
v
e
a
p
p
l
i
c
a
t
i
o
n
s,
”
I
EEE
T
r
a
n
sa
c
t
i
o
n
s
o
n
Af
f
e
c
t
i
v
e
C
o
m
p
u
t
i
n
g
,
v
o
l
.
1
4
,
n
o
.
4
,
p
p
.
2
5
6
4
–
2
5
6
6
,
O
c
t
.
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
TA
F
F
C
.
2
0
2
3
.
3
2
8
6
7
6
9
.
[
8
]
P
.
J.
B
o
t
a
,
C
.
W
a
n
g
,
A
.
L
.
N
.
F
r
e
d
,
a
n
d
H
.
P
.
D
.
S
i
l
v
a
,
“
A
r
e
v
i
e
w
,
c
u
r
r
e
n
t
c
h
a
l
l
e
n
g
e
s
,
a
n
d
f
u
t
u
r
e
p
o
ssi
b
i
l
i
t
i
e
s
o
n
e
m
o
t
i
o
n
r
e
c
o
g
n
i
t
i
o
n
u
s
i
n
g
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
a
n
d
p
h
y
s
i
o
l
o
g
i
c
a
l
si
g
n
a
l
s,”
I
EEE
A
c
c
e
ss
,
v
o
l
.
7
,
p
p
.
1
4
0
9
9
0
–
1
4
1
0
2
0
,
2
0
1
9
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
1
9
.
2
9
4
4
0
0
1
.
[
9
]
C
.
F
i
l
i
p
p
i
n
i
e
t
a
l
.
,
“
A
u
t
o
ma
t
e
d
a
f
f
e
c
t
i
v
e
c
o
mp
u
t
i
n
g
b
a
se
d
o
n
b
i
o
-
si
g
n
a
l
s
a
n
a
l
y
s
i
s
a
n
d
d
e
e
p
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
,
”
S
e
n
so
rs
,
v
o
l
.
2
2
,
n
o
.
5
,
F
e
b
.
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
s2
2
0
5
1
7
8
9
.
[
1
0
]
G
.
G
e
r
ša
k
,
“
E
l
e
c
t
r
o
d
e
r
m
a
l
a
c
t
i
v
i
t
y
-
a
b
e
g
i
n
n
e
r
’
s
g
u
i
d
e
,
”
E
l
e
k
t
r
o
t
e
h
n
i
sk
i
V
e
st
n
i
k
,
v
o
l
.
8
7
,
n
o
.
4
,
p
p
.
1
7
5
–
1
8
2
,
2
0
2
0
.
[
1
1
]
A
.
A
l
F
a
w
w
a
z
,
O
.
N
.
R
a
h
ma
,
K
.
A
i
n
,
S
.
I
.
I
t
t
a
q
i
l
l
a
h
,
a
n
d
R
.
C
h
a
i
,
“
M
e
a
s
u
r
e
men
t
o
f
me
n
t
a
l
w
o
r
k
l
o
a
d
u
si
n
g
h
e
a
r
t
r
a
t
e
v
a
r
i
a
b
i
l
i
t
y
a
n
d
e
l
e
c
t
r
o
d
e
r
ma
l
a
c
t
i
v
i
t
y
,
”
I
EE
E
A
c
c
e
ss
,
v
o
l
.
1
2
,
p
p
.
1
9
7
5
8
9
–
1
9
7
6
0
1
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
4
.
3
5
2
1
6
4
9
.
[
1
2
]
A
.
S
merd
o
v
,
B
.
Z
h
o
u
,
P
.
L
u
k
o
w
i
c
z
,
a
n
d
A
.
S
o
mo
v
,
“
C
o
l
l
e
c
t
i
o
n
a
n
d
v
a
l
i
d
a
t
i
o
n
o
f
p
s
y
c
h
o
p
h
y
si
o
l
o
g
i
c
a
l
d
a
t
a
f
r
o
m p
r
o
f
e
ss
i
o
n
a
l
a
n
d
a
mat
e
u
r
p
l
a
y
e
r
s
:
a
mu
l
t
i
m
o
d
a
l
e
s
p
o
r
t
s
d
a
t
a
s
e
t
,
”
a
rX
i
v
:
2
0
1
1
.
0
0
9
5
8
,
A
u
g
.
2
0
2
1
.
[
1
3
]
A
.
G
r
e
c
o
,
G
.
V
a
l
e
n
z
a
,
A
.
La
n
a
t
a
,
E
.
S
c
i
l
i
n
g
o
,
a
n
d
L.
C
i
t
i
,
“
c
v
x
ED
A
:
a
c
o
n
v
e
x
o
p
t
i
m
i
z
a
t
i
o
n
a
p
p
r
o
a
c
h
t
o
e
l
e
c
t
r
o
d
e
r
ma
l
a
c
t
i
v
i
t
y
p
r
o
c
e
ss
i
n
g
,
”
I
EE
E
T
r
a
n
s
a
c
t
i
o
n
s
o
n
Bi
o
m
e
d
i
c
a
l
E
n
g
i
n
e
e
ri
n
g
,
v
o
l
.
6
3
,
n
o
.
4
,
p
p
.
1
–
1
,
2
0
1
6
,
d
o
i
:
1
0
.
1
1
0
9
/
TB
M
E
.
2
0
1
5
.
2
4
7
4
1
3
1
.
[
1
4
]
F
.
H
.
-
G
a
l
l
e
g
o
,
D
.
Lu
e
n
g
o
,
a
n
d
A
.
A
.
-
R
o
d
r
i
g
u
e
z
,
“
F
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
o
f
g
a
l
v
a
n
i
c
s
k
i
n
r
e
sp
o
n
ses
b
y
n
o
n
n
e
g
a
t
i
v
e
s
p
a
r
se
d
e
c
o
n
v
o
l
u
t
i
o
n
,
”
I
EEE
J
o
u
r
n
a
l
o
f
Bi
o
m
e
d
i
c
a
l
a
n
d
H
e
a
l
t
h
I
n
f
o
rm
a
t
i
c
s
,
v
o
l
.
2
2
,
n
o
.
5
,
p
p
.
1
3
8
5
–
1
3
9
4
,
S
e
p
.
2
0
1
8
,
d
o
i
:
1
0
.
1
1
0
9
/
J
B
H
I
.
2
0
1
7
.
2
7
8
0
2
5
2
.
[
1
5
]
H
.
F
.
P
.
-
Q
u
i
n
t
e
r
o
,
J
.
P
.
F
l
o
r
i
a
n
,
Á
.
D
.
O
.
-
C
a
ñ
ó
n
,
a
n
d
K
.
H
.
C
h
o
n
,
“
H
i
g
h
l
y
sen
s
i
t
i
v
e
i
n
d
e
x
o
f
s
y
m
p
a
t
h
e
t
i
c
a
c
t
i
v
i
t
y
b
a
se
d
o
n
t
i
me
-
f
r
e
q
u
e
n
c
y
sp
e
c
t
r
a
l
a
n
a
l
y
s
i
s
o
f
e
l
e
c
t
r
o
d
e
r
ma
l
a
c
t
i
v
i
t
y
,
”
Am
e
ri
c
a
n
J
o
u
r
n
a
l
o
f
P
h
y
s
i
o
l
o
g
y
-
Re
g
u
l
a
t
o
ry,
I
n
t
e
g
ra
t
i
v
e
a
n
d
C
o
m
p
a
ra
t
i
v
e
Ph
y
s
i
o
l
o
g
y
,
v
o
l
.
3
1
1
,
n
o
.
3
,
p
p
.
R
5
8
2
–
R
5
9
1
,
S
e
p
.
2
0
1
6
,
d
o
i
:
1
0
.
1
1
5
2
/
a
j
p
r
e
g
u
.
0
0
1
8
0
.
2
0
1
6
.
[
1
6
]
Y
.
R
.
V
e
e
r
a
n
k
i
,
N
.
G
a
n
a
p
a
t
h
y
,
R
.
S
w
a
mi
n
a
t
h
a
n
,
a
n
d
H
.
F
.
P
.
-
Q
u
i
n
t
e
r
o
,
“
C
o
m
p
a
r
i
s
o
n
o
f
e
l
e
c
t
r
o
d
e
r
m
a
l
a
c
t
i
v
i
t
y
s
i
g
n
a
l
d
e
c
o
m
p
o
si
t
i
o
n
t
e
c
h
n
i
q
u
e
s
f
o
r
e
m
o
t
i
o
n
r
e
c
o
g
n
i
t
i
o
n
,
”
I
E
EE
Ac
c
e
ss
,
v
o
l
.
1
2
,
p
p
.
1
9
9
5
2
–
1
9
9
6
6
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
4
.
3
3
6
1
8
3
2
.
[
1
7
]
P
.
V
i
r
t
a
n
e
n
e
t
a
l
.
,
“
S
c
i
P
y
1
.
0
:
f
u
n
d
a
me
n
t
a
l
a
l
g
o
r
i
t
h
ms
f
o
r
s
c
i
e
n
t
i
f
i
c
c
o
mp
u
t
i
n
g
i
n
P
y
t
h
o
n
,
”
N
a
t
u
r
e
Me
t
h
o
d
s
,
v
o
l
.
1
7
,
n
o
.
3
,
p
p
.
2
6
1
–
2
7
2
,
M
a
r
.
2
0
2
0
,
d
o
i
:
1
0
.
1
0
3
8
/
s4
1
5
9
2
-
019
-
0
6
8
6
-
2.
[
1
8
]
S
.
K
u
n
j
a
n
e
t
a
l
.
,
“
T
h
e
n
e
c
e
ssi
t
y
o
f
l
e
a
v
e
o
n
e
s
u
b
j
e
c
t
o
u
t
(
LO
S
O
)
c
r
o
ss
v
a
l
i
d
a
t
i
o
n
f
o
r
EEG
d
i
se
a
se
d
i
a
g
n
o
s
i
s,
”
i
n
Br
a
i
n
I
n
f
o
rm
a
t
i
c
s:
1
4
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
ren
c
e
,
BI
2
0
2
1
,
V
i
r
t
u
a
l
e
v
e
n
t
:
S
p
r
i
n
g
e
r
,
C
h
a
m
,
2
0
2
1
,
p
p
.
5
5
8
–
5
6
7
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
0
3
0
-
8
6
9
9
3
-
9
_
5
0
.
[
1
9
]
H
.
F
.
P
.
-
Q
u
i
n
t
e
r
o
,
J.
P
.
F
l
o
r
i
a
n
,
A
.
D
.
O
.
-
C
a
ñ
ó
n
,
a
n
d
K
.
H
.
C
h
o
n
,
“
El
e
c
t
r
o
d
e
r
ma
l
a
c
t
i
v
i
t
y
i
s
se
n
si
t
i
v
e
t
o
c
o
g
n
i
t
i
v
e
st
r
e
ss
u
n
d
e
r
w
a
t
e
r
,
”
Fr
o
n
t
i
e
rs
i
n
P
h
y
s
i
o
l
o
g
y
,
v
o
l
.
8
,
J
a
n
.
2
0
1
8
,
d
o
i
:
1
0
.
3
3
8
9
/
f
p
h
y
s.
2
0
1
7
.
0
1
1
2
8
.
[
2
0
]
M
.
D
i
a
r
r
a
,
J.
Th
e
u
r
e
l
,
a
n
d
B
.
P
a
t
y
,
“
S
y
s
t
e
ma
t
i
c
r
e
v
i
e
w
o
f
n
e
u
r
o
p
h
y
si
o
l
o
g
i
c
a
l
a
ssess
me
n
t
t
e
c
h
n
i
q
u
e
s
a
n
d
m
e
t
r
i
c
s
f
o
r
me
n
t
a
l
w
o
r
k
l
o
a
d
e
v
a
l
u
a
t
i
o
n
i
n
r
e
a
l
-
w
o
r
l
d
s
e
t
t
i
n
g
s,”
F
ro
n
t
i
e
rs
i
n
N
e
u
ro
e
rg
o
n
o
m
i
c
s
,
v
o
l
.
6
,
p
p
.
1
-
1
9
,
A
p
r
.
2
0
2
5
,
d
o
i
:
1
0
.
3
3
8
9
/
f
n
r
g
o
.
2
0
2
5
.
1
5
8
4
7
3
6
.
[
2
1
]
E.
Lu
t
i
n
,
R
.
H
a
s
h
i
mo
t
o
,
W
.
D
.
R
a
e
d
t
,
a
n
d
C
.
V
.
H
o
o
f
,
“
F
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
f
o
r
st
r
e
ss
d
e
t
e
c
t
i
o
n
i
n
e
l
e
c
t
r
o
d
e
r
ma
l
a
c
t
i
v
i
t
y
,
”
i
n
1
4
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
i
n
t
C
o
n
f
e
re
n
c
e
o
n
Bi
o
m
e
d
i
c
a
l
En
g
i
n
e
e
ri
n
g
S
y
s
t
e
m
s
a
n
d
T
e
c
h
n
o
l
o
g
i
e
s
,
R
o
ma
,
I
t
a
l
y
:
S
C
I
TEPRE
S
S
-
S
c
i
e
n
c
e
a
n
d
Te
c
h
n
o
l
o
g
y
P
u
b
l
i
c
a
t
i
o
n
s
,
2
0
2
1
,
p
p
.
1
7
7
–
1
8
5
,
d
o
i
:
1
0
.
5
2
2
0
/
0
0
1
0
2
4
4
6
0
1
7
7
0
1
8
5
.
[
2
2
]
A
.
B
o
f
f
e
t
,
L.
M
.
A
r
sa
c
,
V
.
I
b
a
n
e
z
,
F
.
S
a
u
v
e
t
,
a
n
d
V
.
D
.
-
A
r
sac
,
“
D
e
t
e
c
t
i
o
n
o
f
c
o
g
n
i
t
i
v
e
l
o
a
d
m
o
d
u
l
a
t
i
o
n
b
y
ED
A
a
n
d
H
R
V
,
”
S
e
n
so
rs
,
v
o
l
.
2
5
,
n
o
.
8
,
A
p
r
.
2
0
2
5
,
d
o
i
:
1
0
.
3
3
9
0
/
s2
5
0
8
2
3
4
3
.
[
2
3
]
M
.
R
o
se
n
b
l
a
t
t
,
L.
T
e
j
a
v
i
b
u
l
y
a
,
R
.
Ji
a
n
g
,
S
.
N
o
b
l
e
,
a
n
d
D
.
S
c
h
e
i
n
o
s
t
,
“
D
a
t
a
l
e
a
k
a
g
e
i
n
f
l
a
t
e
s
p
r
e
d
i
c
t
i
o
n
p
e
r
f
o
r
m
a
n
c
e
i
n
c
o
n
n
e
c
t
o
m
e
-
b
a
se
d
mac
h
i
n
e
l
e
a
r
n
i
n
g
mo
d
e
l
s
,
”
N
a
t
u
r
e
C
o
m
m
u
n
i
c
a
t
i
o
n
s
,
v
o
l
.
1
5
,
n
o
.
1
,
p
p
.
1
-
1
5
,
F
e
b
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
2
4
-
4
6
1
5
0
-
w.
[
2
4
]
R
.
P
.
F
r
a
g
a
,
Z
.
K
a
n
g
,
a
n
d
C
.
M
.
A
x
t
h
e
l
m,
“
Ef
f
e
c
t
o
f
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
c
r
o
s
s
-
v
a
l
i
d
a
t
i
o
n
a
l
g
o
r
i
t
h
ms
c
o
n
s
i
d
e
r
i
n
g
h
u
ma
n
p
a
r
t
i
c
i
p
a
n
t
s
a
n
d
t
i
m
e
-
seri
e
s:
a
p
p
l
i
c
a
t
i
o
n
o
n
b
i
o
me
t
r
i
c
d
a
t
a
o
b
t
a
i
n
e
d
f
r
o
m
a
v
i
r
t
u
a
l
r
e
a
l
i
t
y
e
x
p
e
r
i
me
n
t
,
”
Pro
c
e
e
d
i
n
g
s
o
f
t
h
e
H
u
m
a
n
F
a
c
t
o
rs
a
n
d
Erg
o
n
o
m
i
c
s
S
o
c
i
e
t
y
A
n
n
u
a
l
M
e
e
t
i
n
g
,
v
o
l
.
6
7
,
n
o
.
1
,
p
p
.
2
1
6
2
–
2
1
6
7
,
S
e
p
.
2
0
2
3
,
d
o
i
:
1
0
.
1
1
7
7
/
2
1
6
9
5
0
6
7
2
3
1
1
9
2
2
5
8
.
[
2
5
]
E.
O
r
t
e
g
a
a
n
d
C
.
J
.
K
.
W
a
n
g
,
“
P
r
e
-
p
e
r
f
o
r
man
c
e
p
h
y
si
o
l
o
g
i
c
a
l
st
a
t
e
:
h
e
a
r
t
r
a
t
e
v
a
r
i
a
b
i
l
i
t
y
a
s a
p
r
e
d
i
c
t
o
r
o
f
s
h
o
o
t
i
n
g
p
e
r
f
o
r
ma
n
c
e
,
”
Ap
p
l
i
e
d
Ps
y
c
h
o
p
h
y
s
i
o
l
o
g
y
a
n
d
Bi
o
f
e
e
d
b
a
c
k
,
v
o
l
.
4
3
,
n
o
.
1
,
p
p
.
7
5
–
8
5
,
M
a
r
.
2
0
1
8
,
d
o
i
:
1
0
.
1
0
0
7
/
s1
0
4
8
4
-
0
1
7
-
9
3
8
6
-
9.
Evaluation Warning : The document was created with Spire.PDF for Python.