I
A
E
S
I
n
t
e
r
n
at
io
n
al
Jou
r
n
al
of
A
r
t
if
ic
ia
l
I
n
t
e
ll
ig
e
n
c
e
(
I
J
-
AI
)
V
ol
. 14, No. 5, O
c
to
be
r
2025
, pp.
3847
~
3857
I
S
S
N
:
2252
-
8938
,
D
O
I
:
10.11591/
ij
a
i.
v
14
.i
5
.pp
3847
-
3857
3847
Jou
r
n
al
h
om
e
page
:
ht
tp
:
//
ij
ai
.
ia
e
s
c
or
e
.c
om
R
e
c
om
m
e
n
d
at
i
on
sys
t
e
m
f
or
f
oot
b
al
l
p
l
aye
r
r
e
c
r
u
i
t
m
e
n
t
u
si
n
g
k
-
n
e
ar
e
st
n
e
i
gh
b
or
M
au
k
ar
1
, R
od
ia
h
2
1
M
a
s
t
e
r
of
S
ys
t
e
m
I
nf
or
m
a
t
i
on M
a
na
ge
m
e
nt
,
U
ni
ve
r
s
i
t
a
s
G
una
da
r
m
a
,
D
e
pok
, I
ndone
s
i
a
2
D
e
pa
r
t
m
e
nt
of
I
nf
or
m
a
t
i
c
s
,
F
a
c
ul
t
y of
I
ndus
t
r
y
T
e
c
hnol
ogy,
U
ni
ve
r
s
i
t
a
s
G
una
da
r
m
a
,
D
e
pok
, I
ndone
s
i
a
A
r
t
ic
le
I
n
f
o
A
B
S
T
R
A
C
T
A
r
ti
c
le
h
is
to
r
y
:
R
e
c
e
iv
e
d
A
ug 25, 2024
R
e
vi
s
e
d
J
un 23, 2025
A
c
c
e
pt
e
d
J
ul
10, 2025
In
modern
professional
football,
achieving
a
competitive
edge
depen
ds
not
only
on
on
-
field
performance
but
also
on
effective
off
-
field
stra
tegies,
particularly
in
player
recruitment.
This
study
proposes
a
machine
lea
rning
-
based
recommendation
system
to
support
talent
identification
and
o
ptimal
player
placement
using
statistical
performance
data.
The
model
anal
yzes
a
wide
range
of
features,
including
shots,
expected
goals,
expected
assists,
pass
types,
offensive
contributions,
and
defensive
actions
across
fi
eld
zones.
The
dataset
undergoes
preprocessing
steps
such
as
normalization
(
per
90
minutes)
and
dimensionality
reduction.
A
key
innovation
of
this
rese
arch
is
the
use
of
principal
component
analysis
(PCA)
to
reduce
f
eature
dimensionality,
minimizing
redundancy
while
retaining
es
sential
information,
which
improves
model
efficie
ncy
and
scalability.
The
r
efined
data
is
then
processed
using
the
k
-
nearest
neighbors
(KNN)
algorith
m
with
cosine
similarity,
allowing
the
system
to
identify
players
with
similar
perfo
rmance
profiles
based
on
directional
similarity
in
a
high
-
dime
nsional
space.
This
combination
enhances
recommendation
accuracy
by
focusing
on
performance
structure
rather
than
raw
values.
The
resulting
system
pr
ovides
actionabl
e insig
hts in
to play
er
suitab
ility
and
potent
ial, offeri
ng clubs
a data
-
driven
tool
for
informed
scouting
and
recruitment
decisions.
The
ap
proach
demonstrates
the
effectiveness
of
combining
PCA
and
KNN
in
opti
mizing
football p
layer recommendat
ion syst
ems.
K
e
y
w
o
r
d
s
:
C
os
in
e
s
im
il
a
r
it
y
F
e
a
tu
r
e
d
F
oot
ba
ll
pl
a
ye
r
P
r
e
di
c
ti
on
R
e
c
om
m
e
nda
ti
on s
y
s
te
m
This is an
open
acce
ss artic
le unde
r the
CC BY
-
SA
license.
C
or
r
e
s
pon
di
n
g A
u
th
or
:
R
odi
a
h
D
e
pa
r
tm
e
nt
of
I
nf
or
m
a
ti
c
s
, F
a
c
ul
ty
of
I
ndus
tr
y T
e
c
hnol
ogy,
U
ni
ve
r
s
it
a
s
G
una
da
r
m
a
M
a
r
gonda
R
a
ya
100, P
ondok C
in
a
,
D
e
pok,
W
e
s
t
J
a
v
a
, I
ndone
s
i
a
E
m
a
il
:
r
odi
a
h@
s
ta
f
f
.
guna
da
r
m
a
.a
c
.i
d
1.
I
N
T
R
O
D
U
C
T
I
O
N
T
he
a
p
pl
ic
a
t
io
n
o
f
da
ta
s
c
ie
nc
e
a
nd
m
a
c
hi
n
e
le
a
r
ni
ng
in
s
po
r
t
s
,
pa
r
t
ic
u
la
r
l
y
p
r
o
f
e
s
s
io
na
l
f
o
ot
ba
l
l,
ha
s
g
r
ow
n
s
i
gn
if
ic
a
nt
ly
in
r
e
c
e
nt
ye
a
r
s
[
1
]
.
T
he
s
e
t
e
c
h
nol
og
i
e
s
a
r
e
in
c
r
e
a
s
in
gl
y
us
e
d
n
ot
on
ly
to
e
nha
nc
e
te
a
m
pe
r
f
o
r
m
a
nc
e
o
n
th
e
f
i
e
l
d
bu
t
a
ls
o
to
s
upp
o
r
t
s
tr
a
te
gi
c
d
e
c
is
i
ons
of
f
th
e
f
ie
l
d
[
2
]
,
s
uc
h
a
s
op
ti
m
iz
in
g
pl
a
y
e
r
r
e
c
r
u
it
m
e
n
t
a
nd
p
la
c
e
m
e
n
t
[
3]
.
A
s
c
om
pe
t
it
io
n
in
te
ns
i
f
ie
s
a
c
r
os
s
t
op
l
e
a
g
ue
s
,
c
l
ubs
s
e
e
k
to
ga
in
a
c
om
p
e
t
it
iv
e
e
d
ge
by
i
nt
e
g
r
a
t
in
g
d
a
ta
-
d
r
iv
e
n
a
pp
r
oa
c
he
s
i
n
to
s
c
ou
ti
ng
p
r
oc
e
s
s
e
s
.
F
oo
tb
a
ll
c
l
ubs
m
us
t
c
ons
t
a
n
tl
y
e
va
l
ua
te
a
nd
r
e
pl
a
c
e
pl
a
ye
r
s
due
t
o
tr
a
ns
f
e
r
s
,
in
j
ur
ie
s
,
o
r
pe
r
f
or
m
a
nc
e
is
s
ue
s
[
4
]
.
T
r
a
di
ti
ona
l
s
c
out
in
g
m
e
th
ods
,
w
h
il
e
va
lu
a
bl
e
,
a
r
e
of
te
n
s
ub
je
c
ti
v
e
a
n
d
c
o
s
tl
y
.
M
a
c
hi
ne
le
a
r
ni
ng
of
f
e
r
s
a
m
o
r
e
s
c
a
la
bl
e
a
nd
ob
je
c
ti
ve
s
o
lu
t
io
n,
c
a
pa
b
le
o
f
e
va
lu
a
ti
ng
va
s
t
da
ta
s
e
ts
t
o
id
e
nt
if
y
pl
a
ye
r
s
w
h
os
e
s
ta
ti
s
ti
c
a
l
p
r
o
f
i
le
s
a
li
gn
w
i
th
te
a
m
ne
e
ds
[
5
]
.
A
dva
nc
e
d
f
oot
ba
ll
s
ta
t
is
t
ic
s
s
uc
h
a
s
e
xpe
c
te
d
goa
ls
(
x
G
)
,
e
xpe
c
te
d
a
s
s
is
ts
(
x
A
)
,
t
a
ke
-
o
ns
,
a
nd
de
f
e
ns
i
ve
a
c
t
io
ns
[
6
]
,
a
r
e
now
c
o
m
m
on
ly
us
e
d
t
o
a
s
s
e
s
s
p
la
y
e
r
pe
r
f
or
m
a
nc
e
[
7
]
.
P
r
e
vi
ous
r
e
s
e
a
r
c
h
[
8]
,
[
9
]
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
. 14, No. 5, O
c
to
be
r
2025
:
3847
-
3857
3848
ha
s
e
x
pl
or
e
d
p
r
e
di
c
ti
ve
m
od
e
l
in
g
f
or
pl
a
ye
r
t
r
a
ns
f
e
r
s
,
pe
r
f
o
r
m
a
nc
e
e
va
lu
a
ti
on
,
a
n
d
r
ol
e
c
l
a
s
s
i
f
ic
a
t
io
n
.
F
o
r
e
xa
m
pl
e
,
s
om
e
s
t
ud
ie
s
de
ve
lo
pe
d
p
r
e
di
c
ti
ve
m
o
de
ls
ba
s
e
d
on
t
r
a
ns
f
e
r
da
t
a
[
1
0]
,
w
h
il
e
o
th
e
r
s
a
pp
li
e
d
c
lu
s
te
r
in
g
a
nd
c
la
s
s
if
ic
a
ti
on
m
e
th
o
ds
to
id
e
nt
i
f
y
pl
a
ye
r
r
o
le
s
or
a
na
ly
z
e
g
a
m
e
s
t
r
a
t
e
g
ie
s
[
1
1]
.
H
ow
e
ve
r
,
m
a
ny
o
f
t
he
s
e
a
pp
r
o
a
c
he
s
e
i
th
e
r
o
ve
r
lo
ok
d
im
e
ns
i
ona
li
ty
i
s
s
u
e
s
[
12
]
in
h
ig
h
-
di
m
e
ns
io
na
l
da
ta
s
e
ts
or
f
a
i
l
to
in
t
e
g
r
a
t
e
s
i
m
i
la
r
i
ty
-
b
a
s
e
d
m
o
de
ls
e
f
f
e
c
t
iv
e
ly
i
nt
o
p
la
ye
r
r
e
c
o
m
m
e
nda
ti
on s
ys
te
m
s
.
S
e
ve
r
a
l
s
tu
d
ie
s
r
e
la
t
e
d
to
r
e
c
om
m
e
nda
ti
on
s
ys
te
m
f
o
r
f
o
ot
ba
l
l
pl
a
ye
r
s
b
a
s
e
d
on
th
e
i
r
pe
r
f
or
m
a
n
c
e
a
nd
t
a
le
n
t
us
i
ng
da
ta
s
c
ie
nc
e
a
nd
m
a
c
h
in
e
l
e
a
r
ni
ng
c
o
nc
e
p
ts
h
a
ve
be
e
n
c
a
r
r
ie
d
ou
t
by
p
r
e
v
io
us
r
e
s
e
a
r
c
he
r
s
.
D
in
s
da
le
a
nd
G
a
l
la
g
he
r
[
1
3
]
c
r
e
a
te
d
a
m
ode
l
to
p
r
e
d
ic
t
th
e
p
e
r
f
o
r
m
a
nc
e
o
f
a
f
o
ot
b
a
l
l
pl
a
ye
r
a
f
t
e
r
m
ov
in
g
f
r
om
one
c
l
ub
to
a
no
th
e
r
a
nd
w
it
h
a
m
o
de
l
th
a
t
to
o
k
i
np
ut
f
r
om
13
f
e
a
tu
r
e
s
o
f
s
ta
ti
s
ti
c
a
l
da
ta
on
th
e
pe
r
f
o
r
m
a
nc
e
o
f
a
f
oo
tb
a
ll
pl
a
ye
r
o
bt
a
in
e
d
f
r
om
26,
00
0
s
a
m
pl
e
s
o
f
tr
a
ns
f
e
r
a
n
d
no
n
-
t
r
a
ns
f
e
r
da
ta
f
r
om
32
do
m
e
s
t
ic
le
a
gue
s
f
r
om
2
017
.
M
ode
ls
in
r
e
s
e
a
r
c
h
[
13
]
c
o
m
pa
r
e
d
to
a
s
a
m
pl
e
o
f
2
,
65
9
hi
s
to
r
ic
a
l
t
r
a
ns
f
e
r
da
ta
a
nd
8
,
67
7
h
is
t
or
ic
a
l
no
n
-
t
r
a
ns
f
e
r
da
ta
a
n
d
c
o
m
pa
r
e
d
t
o
th
e
pr
e
di
c
t
io
ns
of
a
m
od
e
l
t
ha
t
ha
s
be
e
n
c
r
e
a
te
d
w
he
r
e
pl
a
ye
r
s
a
r
e
a
s
s
u
m
e
d
t
o
c
on
ti
nue
to
p
r
od
uc
e
t
he
s
a
m
e
p
e
r
f
o
r
m
a
nc
e
f
r
om
b
e
f
or
e
a
nd
a
f
te
r
t
he
c
lu
b
t
r
a
ns
f
e
r
.
B
unk
e
r
a
n
d
T
h
a
b
ta
h
[
14
]
e
xp
lo
r
e
th
e
us
e
o
f
m
a
c
hi
n
e
le
a
r
n
in
g
i
n
pr
e
di
c
t
in
g
o
ut
c
om
e
s
i
n
th
e
in
c
r
e
a
s
in
g
ly
im
po
r
ta
nt
p
r
o
f
e
s
s
i
ona
l
s
po
r
ts
w
o
r
ld
o
f
s
p
o
r
ts
te
a
m
m
a
na
ge
m
e
n
t
a
s
w
e
l
l
a
s
i
n
th
e
ga
m
b
li
ng
in
d
us
t
r
y
.
R
e
s
e
a
r
c
he
r
s
a
r
g
ue
th
a
t
m
a
c
h
in
e
le
a
r
ni
ng
is
a
p
r
o
m
is
in
g
m
e
th
od
to
a
c
h
ie
v
e
h
ig
h
-
a
c
c
u
r
a
c
y
pr
e
di
c
ti
ons
i
n
th
is
f
ie
l
d,
a
n
d
r
e
s
e
a
r
c
he
r
s
pr
op
os
e
a
ne
w
f
r
a
m
e
w
or
k
c
a
l
le
d
s
p
or
ts
r
e
s
ul
t
p
r
e
d
ic
t
io
n
c
r
os
s
‑
i
nd
us
t
r
y
s
ta
nda
r
d
p
r
oc
e
s
s
f
or
da
ta
m
i
ni
ng
(
S
R
P
-
C
R
I
S
P
-
DM
)
to
pr
e
di
c
t
s
po
r
ts
ou
tc
om
e
s
,
t
hi
s
f
r
a
m
e
w
o
r
k
in
c
lu
de
s
s
e
v
e
r
a
l
s
ta
ge
s
s
uc
h
a
s
da
ta
c
ol
le
c
ti
on
a
nd
pr
e
p
r
oc
e
s
s
in
g
,
f
e
a
tu
r
e
s
e
le
c
ti
o
n,
m
o
de
l
c
r
e
a
ti
on
a
nd
m
ode
l
e
v
a
l
ua
t
io
n
a
nd
de
p
lo
ym
e
nt
.
C
ha
va
n
[
1
5
]
tr
ie
s
t
o
f
in
d
a
s
o
lu
ti
on
to
t
he
p
r
obl
e
m
to
f
i
nd
th
e
c
lo
s
e
s
t
m
a
tc
h
o
f
th
e
p
la
ye
r
t
o
be
r
e
pl
a
c
e
d
us
in
g
m
a
c
hi
ne
l
e
a
r
n
in
g
a
l
go
r
i
th
m
s
.
P
l
a
ye
r
s
w
il
l
be
c
la
s
s
i
f
ie
d
ba
s
e
d
o
n
r
a
t
in
gs
,
in
th
is
s
tu
dy
s
ix
m
a
c
h
in
e
le
a
r
ni
ng
a
lg
or
it
hm
s
w
e
r
e
us
e
d,
na
m
e
ly
s
upp
or
t
ve
c
to
r
m
a
c
h
in
e
(
S
V
M
)
,
l
in
e
a
r
d
is
c
r
i
m
i
na
n
t
a
na
ly
s
is
(
L
D
A
)
,
n
a
ïv
e
B
a
ye
s
,
d
e
c
is
io
n
t
r
e
e
,
X
G
B
o
os
t
,
a
n
d
k
-
ne
a
r
e
s
t
ne
ig
h
bo
r
(
K
N
N
)
,
t
he
n
a
c
om
pa
r
is
on
w
a
s
m
a
d
e
be
twe
e
n
t
he
a
l
go
r
it
h
m
s
a
nd
i
t
w
a
s
f
oun
d
t
ha
t
L
D
A
a
n
d
S
V
M
ha
d
t
he
be
s
t
a
c
c
u
r
a
c
y
w
i
th
8
3.7
7
%
a
nd
80
.3
1%
w
hi
le
th
e
K
N
N
a
l
g
or
it
hm
p
r
o
duc
e
d
r
e
s
u
l
ts
th
a
t
ha
d
t
he
c
lo
s
e
s
t
m
a
tc
h
to
th
e
p
r
e
di
c
te
d
p
la
ye
r
.
L
i
e
t
al
.
[
16
]
c
ha
r
a
c
te
r
iz
e
s
th
e
t
ype
of
p
la
y
of
f
o
ot
b
a
l
le
r
s
i
n
th
e
C
h
in
e
s
e
f
o
ot
ba
l
l
s
up
e
r
le
a
g
ue
(
C
S
L
)
le
a
gue
o
bt
a
in
e
d
f
r
o
m
96
0
m
a
tc
h
da
ta
f
r
o
m
20
16
-
2
0
19.
T
he
f
i
r
s
t
p
la
y
e
r
w
il
l
be
c
l
us
te
r
e
d
in
t
o
8
pos
i
ti
ons
th
e
n
a
one
-
pl
a
ye
r
ve
c
to
r
w
il
l
be
c
r
e
a
te
d
f
o
r
e
a
c
h
p
la
ye
r
i
n
e
a
c
h
m
a
tc
h
ba
s
e
d
on
p
la
y
e
r
ve
c
t
o
r
s
us
in
g
n
on
ne
ga
ti
ve
m
a
t
r
ix
f
a
c
to
r
i
z
a
t
io
n
(
N
M
F
)
.
A
s
a
r
e
s
ul
t,
1
8
t
ype
s
of
p
la
ye
r
s
w
e
r
e
f
ou
nd
to
p
la
y
in
th
e
C
S
L
a
nd
in
ge
ne
r
a
l
t
he
t
ype
o
f
p
la
y
in
g
f
o
r
w
a
r
d
a
nd
m
i
df
ie
ld
e
r
is
d
ir
e
c
t
ly
p
r
opo
r
ti
o
na
l
to
t
he
t
r
e
nd
o
f
th
e
e
vol
ut
io
n
of
f
o
ot
ba
l
l
pe
r
f
o
r
m
a
nc
e
,
w
hi
le
th
e
ty
p
e
of
p
la
yi
n
g
de
f
e
n
de
r
s
m
us
t
be
r
e
c
ons
id
e
r
e
d,
t
he
ty
pe
o
f
m
ul
ti
f
unc
ti
ona
l
p
la
y
is
a
ls
o
f
o
und
a
m
on
g
C
S
L
p
la
ye
r
s
.
Y
e
a
n
e
t
a
l.
[
17
]
f
o
un
d
th
a
t
m
a
c
h
in
e
le
a
r
ni
ng
a
lg
o
r
it
hm
s
c
a
n
a
ls
o
be
a
pp
li
e
d
to
s
e
ve
r
a
l
c
la
s
s
if
ic
a
ti
on
p
r
o
bl
e
m
s
in
c
lu
di
ng
c
li
ni
c
a
l
s
t
ud
ie
s
,
o
ne
o
f
w
hi
c
h
is
in
t
h
e
a
na
l
ys
is
o
f
t
he
e
m
o
ti
ons
o
f
s
t
r
o
ke
pa
t
ie
n
ts
.
T
he
K
N
N
a
l
go
r
i
th
m
r
e
l
ie
s
o
n
m
e
t
r
ic
di
s
ta
nc
e
to
c
a
lc
u
la
te
th
e
ne
a
r
e
s
t
c
la
s
s
f
o
r
c
la
s
s
i
f
ic
a
ti
on
.
T
he
pu
r
p
os
e
o
f
th
is
s
tu
dy
w
a
s
t
o
c
om
pa
r
e
t
he
pe
r
f
o
r
m
a
nc
e
o
f
s
e
ve
r
a
l
di
f
f
e
r
e
nt
di
s
ta
n
c
e
m
e
t
r
i
c
s
t
o
be
a
p
pl
ie
d
to
th
e
c
la
s
s
i
f
i
c
a
t
io
n
of
e
m
ot
io
n
a
l
e
le
c
t
r
oe
nc
e
p
ha
l
og
r
a
m
(
E
E
G
)
be
tw
e
e
n
s
tr
oke
pa
t
ie
n
ts
a
n
d
o
r
d
in
a
r
y
pe
o
pl
e
.
T
he
r
e
s
ul
t
is
th
a
t
t
he
c
i
ty
bl
oc
k
di
s
ta
nc
e
m
e
tr
ic
h
a
s
th
e
be
s
t
pe
r
f
o
r
m
a
nc
e
a
m
on
g
ot
he
r
s
.
L
i
e
t
al
.
[
1
6]
w
a
s
f
ou
nd
t
ha
t
1
8
ty
pe
s
o
f
f
oo
tb
a
ll
e
r
s
p
la
ye
d
in
t
h
e
C
S
L
,
t
h
e
ty
p
e
of
p
la
y
in
g
a
tt
a
c
ki
ng
p
la
ye
r
s
a
n
d
m
i
df
ie
ld
e
r
s
is
di
r
e
c
tl
y
p
r
op
or
ti
o
na
l
to
t
h
e
tr
e
nd
o
f
t
he
e
vo
lu
ti
on
o
f
pl
a
yi
ng
s
ty
l
e
,
w
hi
le
t
he
p
la
y
in
g
s
ty
le
o
f
d
e
f
e
nde
r
s
m
us
t
be
r
e
c
o
ns
id
e
r
e
d
.
A
w
id
e
va
r
ie
ty
o
f
d
is
t
a
nc
e
m
e
t
r
ic
s
w
e
r
e
tr
ie
d
i
n
t
he
s
tu
d
y
[
1
7]
a
n
d
i
t
w
a
s
f
ou
nd
th
a
t
th
e
d
if
f
e
r
e
n
c
e
in
t
he
m
e
t
r
i
c
s
us
e
d
h
a
d
a
n
e
f
f
e
c
t
o
n
t
he
pe
r
f
o
r
m
a
nc
e
o
f
th
e
m
ode
l.
T
hi
s
s
tu
d
y
a
dd
r
e
s
s
e
s
t
he
s
e
li
m
i
ta
ti
ons
by
p
r
o
pos
in
g
a
p
r
in
c
ip
a
l
c
om
pon
e
nt
a
na
l
ys
is
(
P
C
A
)
e
nha
nc
e
d
K
N
N
r
e
c
o
m
m
e
nda
ti
on
s
ys
te
m
us
in
g
c
os
in
e
s
i
m
i
la
r
i
ty
to
r
e
c
o
m
m
e
nd
f
oo
tb
a
ll
p
la
ye
r
s
ba
s
e
d
o
n
pe
r
f
o
r
m
a
nc
e
s
im
il
a
r
it
y.
P
C
A
is
e
m
p
lo
y
e
d
t
o
r
e
duc
e
f
e
a
tu
r
e
d
im
e
ns
i
ona
li
ty
w
hi
le
p
r
e
s
e
r
v
in
g
c
r
it
ic
a
l
s
ta
t
is
t
ic
a
l
i
nf
o
r
m
a
ti
on,
t
hus
im
pr
ov
in
g
m
ode
l
p
e
r
f
o
r
m
a
nc
e
a
nd
in
te
r
p
r
e
ta
bi
li
ty
.
B
y
t
r
a
ns
f
or
m
i
ng
hi
gh
-
di
m
e
ns
i
ona
l
pl
a
ye
r
pe
r
f
or
m
a
n
c
e
d
a
ta
i
nt
o
a
l
ow
e
r
-
di
m
e
ns
io
na
l
s
pa
c
e
,
P
C
A
m
in
im
iz
e
s
r
e
dun
da
n
c
y
a
nd
hi
g
hl
ig
ht
s
t
he
m
os
t
in
f
l
ue
n
ti
a
l
f
e
a
t
ur
e
s
.
T
he
s
e
opt
i
m
iz
e
d
f
e
a
t
ur
e
s
a
r
e
th
e
n
ut
il
iz
e
d
in
a
K
N
N
m
ode
l
w
it
h
c
os
in
e
s
i
m
i
la
r
i
ty
,
w
h
ic
h
c
a
lc
u
la
te
s
t
he
a
n
gu
la
r
s
i
m
i
la
r
it
y
be
twe
e
n
pl
a
ye
r
ve
c
to
r
s
,
m
a
ki
ng
i
t
p
a
r
ti
c
ul
a
r
ly
e
f
f
e
c
t
iv
e
i
n
i
de
n
ti
f
yi
n
g
p
la
ye
r
s
w
i
th
s
t
r
uc
tu
r
a
ll
y
s
i
m
i
la
r
p
la
y
s
ty
le
s
,
in
de
pe
nde
nt
o
f
r
a
w
m
a
g
ni
tu
de
.
T
hi
s
m
e
t
hod
ol
og
ic
a
l
c
o
m
b
in
a
ti
on
e
n
ha
nc
e
s
b
ot
h t
he
pr
e
c
is
io
n a
nd
s
c
a
la
b
il
it
y
of
t
he
r
e
c
om
m
e
n
da
t
io
n p
r
oc
e
s
s
.
T
he
da
ta
s
e
t
us
e
d
in
t
hi
s
s
tu
d
y
is
s
ou
r
c
e
d
f
r
om
f
o
ot
ba
l
l
-
r
e
f
e
r
e
n
c
e
(
F
B
r
e
f
)
[
18
]
a
nd
c
ons
is
ts
of
pl
a
y
e
r
s
ta
t
is
t
ic
s
f
r
om
th
e
to
p
f
i
ve
E
u
r
o
pe
a
n
f
o
ot
ba
l
l
le
a
gu
e
s
du
r
in
g
t
he
20
22
–
20
23
s
e
a
s
on
[
1
9]
.
T
hi
r
te
e
n
pe
r
f
o
r
m
a
nc
e
f
e
a
tu
r
e
s
a
r
e
s
e
le
c
t
e
d
f
or
o
ut
f
i
e
l
d
pl
a
ye
r
s
,
in
c
lu
d
in
g
s
h
ot
s
,
x
G
,
xA
,
c
r
os
s
e
s
,
t
ot
a
l
pa
s
s
e
s
,
a
n
d
va
r
io
us
ty
pe
s
o
f
de
f
e
ns
iv
e
a
n
d
of
f
e
ns
i
ve
c
on
tr
ib
ut
io
ns
.
P
C
A
r
e
duc
e
s
t
he
s
e
f
e
a
tu
r
e
s
w
h
il
e
m
a
i
nt
a
in
in
g
da
ta
in
t
e
g
r
i
ty
.
T
he
m
o
de
l
t
he
n
id
e
nt
i
f
ie
s
th
e
m
os
t
s
i
m
i
la
r
p
la
ye
r
s
ba
s
e
d
on
c
os
i
ne
s
im
il
a
r
it
y
s
c
o
r
e
s
,
o
f
f
e
r
i
ng
a
pr
a
c
ti
c
a
l
t
oo
l
f
o
r
s
c
ou
ts
a
nd a
na
ly
s
ts
t
o
m
a
ke
i
nf
or
m
e
d de
c
is
i
ons
a
bo
ut
po
te
n
ti
a
l
r
e
c
r
ui
ts
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
R
e
c
om
m
e
ndat
io
n s
y
s
t
e
m
f
or
f
oot
bal
l
pl
ay
e
r
r
e
c
r
ui
tme
nt
us
in
g k
-
ne
ar
e
s
t
ne
ig
hbor
(
M
auk
a
r
)
3849
2.
M
E
T
H
O
D
T
hi
s
r
e
s
e
a
r
c
h c
o
ns
i
s
t
s
of
s
e
ve
r
a
l
s
t
a
ge
s
of
t
h
e
pr
oc
e
s
s
s
ta
r
ti
ng f
r
om
da
ta
c
ol
l
e
c
ti
o
n t
o t
he
c
r
e
a
t
io
n of
a
r
e
c
om
m
e
nda
ti
on
s
ys
te
m
b
a
s
e
d on t
he
s
im
il
a
r
it
y of
m
a
t
c
h hi
s
to
r
i
c
a
l
d
a
ta
a
s
c
a
n b
e
s
e
e
n i
n F
ig
ur
e
1. A
t
th
e
f
in
a
l
s
ta
g
e
,
th
e
da
t
a
s
e
t
th
a
t
i
s
r
e
a
dy
w
il
l
be
gi
v
e
n
in
to
t
he
m
od
e
l
to
be
c
r
e
a
te
d
,
na
m
e
ly
th
e
pl
a
y
e
r
r
e
c
o
m
m
e
n
da
ti
o
n
s
ys
te
m
.
T
h
e
m
od
e
l
w
il
l
tr
y t
o pr
e
di
c
t
t
he
di
s
ta
nc
e
be
t
w
e
e
n da
t
a
poi
nt
s
us
i
ng t
h
e
KNN
a
lg
or
it
hm
a
n
d t
he
c
o
s
in
e
s
im
il
a
r
it
y
m
e
tr
ic
.
T
h
e
d
a
ta
s
e
t
us
e
d i
s
d
a
ta
th
a
t
ha
s
b
e
e
n
di
m
e
n
s
i
ona
li
ty
r
e
du
c
ti
on
be
f
or
e
.
T
h
e
r
e
s
e
a
r
c
h
b
e
g
a
n
w
it
h
t
he
pr
oc
e
s
s
of
c
ol
le
c
ti
ng
d
a
t
a
f
r
om
a
pl
a
ye
r
s
t
a
ti
s
t
ic
s
c
ol
l
e
c
ti
on
s
it
e
c
a
l
le
d
F
B
r
e
f
[
18
]
.
T
h
e
d
a
t
a
c
ol
l
e
c
t
e
d
i
nc
lu
d
e
s
v
a
r
i
ou
s
f
e
a
t
ur
e
s
s
uc
h
a
s
xG
,
xA
,
a
n
d
s
e
ve
r
a
l
ot
h
e
r
f
e
a
tu
r
e
s
.
T
h
is
da
ta
c
ol
le
c
ti
on
pr
o
c
e
s
s
i
s
c
a
r
r
ie
d
o
ut
by
s
c
r
a
pi
ng
da
ta
on
pl
a
y
e
r
s
t
a
ti
s
t
ic
s
a
v
a
i
la
bl
e
o
n
t
he
s
it
e
.
A
f
t
e
r
t
he
d
a
t
a
i
s
s
u
c
c
e
s
s
f
u
ll
y
c
ol
le
c
t
e
d
,
th
e
ne
xt
s
te
p
i
s
to
c
a
r
r
y
out
d
a
t
a
pr
e
pr
oc
e
s
s
in
g
w
h
e
r
e
t
hi
s
s
ta
ge
w
il
l
c
on
s
is
t
of
s
e
v
e
r
a
l
s
t
a
g
e
s
,
t
h
e
f
ir
s
t
of
w
hi
c
h
b
e
gi
ns
w
i
th
th
e
d
a
t
a
e
xp
lo
r
a
ti
on
s
t
a
g
e
,
na
m
e
l
y
t
he
e
xpl
or
a
ti
o
n
a
nd
a
na
l
y
s
i
s
of
e
a
c
h
f
e
a
tu
r
e
in
t
he
d
a
t
a
s
e
t
in
or
d
e
r
t
o
ge
t
a
c
c
ur
a
te
i
nf
or
m
a
t
io
n
on
t
he
b
e
s
t
s
te
p
s
to
b
e
ta
ke
n
ne
xt
.
A
f
t
e
r
th
e
d
a
ta
e
xp
lo
r
a
ti
on
pr
oc
e
s
s
i
s
c
o
m
pl
e
t
e
,
t
he
n
e
x
t
s
te
p i
s
t
o
do
d
a
t
a
c
le
a
n
i
ng.
T
h
is
p
r
o
c
e
s
s
in
vol
ve
s
r
e
m
o
vi
ng
d
a
t
a
t
h
a
t
i
s
ir
r
e
l
e
v
a
n
t
t
o
t
he
pur
po
s
e
of
th
e
s
t
ud
y
or
in
a
c
c
ur
a
te
d
a
t
a
,
a
s
w
e
l
l
a
s
h
a
n
dl
i
ng
nul
l
d
a
t
a
o
r
m
i
s
s
i
ng
da
ta
.
I
n
a
dd
it
i
on
,
d
a
t
a
n
or
m
a
li
z
a
ti
o
n
w
il
l
a
l
s
o
b
e
c
a
r
r
ie
d
o
ut
,
w
h
ic
h
i
s
t
h
e
pr
oc
e
s
s
of
c
h
a
n
gi
n
g
t
he
v
a
lu
e
s
in
th
e
d
a
t
a
s
e
t
s
o
th
a
t
t
he
d
a
t
a
i
s
o
n
t
he
s
a
m
e
s
c
a
l
e
.
A
f
te
r
d
a
t
a
c
l
e
a
ni
ng,
t
he
d
a
t
a
tr
a
n
s
f
or
m
a
ti
on
pr
oc
e
s
s
w
il
l
b
e
c
a
r
r
i
e
d
o
ut
.
T
h
is
pr
o
c
e
s
s
in
v
ol
ve
s
c
onv
e
r
t
in
g
t
he
d
a
t
a
in
t
o
a
f
or
m
a
t
th
a
t
i
s
m
or
e
s
u
it
a
bl
e
f
or
m
o
de
l
in
g
l
a
t
e
r
,
s
u
c
h
a
s
c
on
ve
r
ti
ng
c
a
t
e
gor
ic
a
l
d
a
t
a
i
nt
o
n
um
e
r
i
c
a
l
d
a
t
a
,
a
f
te
r
d
a
t
a
tr
a
n
s
f
o
r
m
a
t
io
n
, f
e
a
tu
r
e
s
e
l
e
c
ti
o
n
w
il
l
b
e
c
a
r
r
i
e
d
ou
t.
T
hi
s
pr
oc
e
s
s
in
vol
ve
s
s
e
le
c
ti
ng
th
e
m
os
t
r
e
le
va
nt
a
nd
in
f
or
m
a
ti
ve
f
e
a
tu
r
e
s
to
us
e
in
th
e
m
ode
l.
T
he
s
e
f
e
a
tu
r
e
s
a
r
e
c
ho
s
e
n
b
a
s
e
d
on
th
e
ir
im
por
ta
nc
e
in
pr
e
di
c
ti
ng
ta
r
ge
t
va
r
ia
bl
e
s
,
r
e
duc
in
g
th
e
c
ha
nc
e
of
unde
r
f
it
ti
ng
a
s
w
e
ll
a
s
ove
r
f
it
ti
ng
[
20
]
but
s
ti
ll
r
e
pr
e
s
e
nt
s
th
e
da
ta
s
e
t
a
s
a
w
hol
e
.
A
f
te
r
f
e
a
tu
r
e
s
e
le
c
ti
on,
m
ode
l
c
r
e
a
ti
on
a
nd
tr
a
in
in
g
a
r
e
c
a
r
r
ie
d
out
.
T
hi
s
m
ode
l
w
il
l
be
us
e
d
to
c
r
e
a
te
a
r
e
c
om
m
e
nd
a
ti
on
s
ys
te
m
w
he
r
e
us
e
r
s
c
a
n
e
nt
e
r
in
put
in
th
e
f
or
m
of
pl
a
ye
r
s
w
ho
w
il
l
be
lo
oki
ng
f
or
th
e
m
os
t
s
im
il
a
r
pl
a
ye
r
a
c
c
or
di
ng
to
th
e
f
e
a
tu
r
e
or
s
ta
ti
s
ti
c
s
th
a
t
r
e
pr
e
s
e
nt
th
e
pl
a
ye
r
.
T
he
m
ode
l
w
il
l
th
e
n
be
te
s
te
d
to
s
e
e
how
a
c
c
ur
a
te
th
e
pr
e
di
c
ti
ons
a
r
e
.
O
nc
e
th
e
m
ode
l
ha
s
be
e
n
c
r
e
a
te
d
a
nd
tr
a
in
e
d
a
nd
th
e
r
e
s
ul
ts
a
r
e
obt
a
in
e
d,
th
e
m
ode
l
w
il
l
be
de
pl
oye
d us
in
g t
he
s
tr
e
a
m
li
t
f
r
a
m
e
w
or
k
[
21]
.
F
ig
ur
e
1. G
e
ne
r
a
l
c
ha
r
t
of
f
oot
ba
ll
pl
a
ye
r
r
e
c
om
m
e
nda
ti
on s
ys
te
m
r
e
s
e
a
r
c
h m
e
th
ods
2
.1.
D
at
a c
ol
le
c
t
io
n
an
d
p
r
e
p
ar
at
io
n
T
he
de
ve
lo
pm
e
nt
of
th
is
r
e
c
om
m
e
nda
ti
on
s
ys
te
m
be
gi
ns
w
it
h
th
e
c
ol
le
c
ti
on
a
nd
pr
e
pa
r
a
ti
on
of
r
e
le
va
nt
da
ta
.
H
ig
h
-
qua
li
ty
a
nd
w
e
ll
-
s
tr
uc
tu
r
e
d
da
ta
i
s
e
s
s
e
nt
ia
l
f
or
bui
ld
in
g
a
n
a
c
c
ur
a
te
a
nd
r
e
li
a
bl
e
m
a
c
hi
n
e
le
a
r
ni
ng
m
ode
l,
pa
r
ti
c
ul
a
r
ly
in
th
e
c
ont
e
xt
of
f
oot
ba
ll
pl
a
ye
r
pe
r
f
or
m
a
nc
e
a
na
ly
s
is
.
T
hi
s
s
ta
ge
in
vol
ve
s
s
e
le
c
ti
ng a
c
r
e
di
bl
e
da
t
a
s
our
c
e
, i
de
nt
if
yi
ng ke
y pe
r
f
or
m
a
nc
e
f
e
a
tu
r
e
s
, a
nd pe
r
f
or
m
in
g da
ta
c
le
a
ni
ng t
o e
ns
ur
e
c
ons
is
te
nc
y
a
nd c
om
pl
e
te
ne
s
s
be
f
or
e
pr
oc
e
e
di
ng t
o t
he
m
ode
li
n
g pr
oc
e
s
s
.
2
.
1
.1.
D
at
a s
ou
r
c
e
an
d
s
e
le
c
t
io
n
T
hi
s
s
tu
dy
u
s
e
s
pl
a
ye
r
pe
r
f
or
m
a
nc
e
da
ta
s
our
c
e
d
f
r
om
F
B
r
e
f
[
18]
,
a
r
e
put
a
bl
e
f
oot
ba
ll
s
ta
ti
s
ti
c
s
pl
a
tf
or
m
th
a
t
a
ggr
e
ga
te
s
a
dva
nc
e
d
pl
a
ye
r
m
e
tr
ic
s
a
c
r
os
s
m
a
jo
r
f
oot
ba
ll
le
a
gue
s
.
F
B
r
e
f
is
w
id
e
ly
a
dopt
e
d
i
n
pr
of
e
s
s
io
na
l
s
por
ts
a
na
ly
ti
c
s
due
to
it
s
da
ta
a
c
c
ur
a
c
y,
br
e
a
dt
h,
a
nd
a
c
c
e
s
s
ib
il
it
y.
F
or
th
e
pur
pos
e
of
th
is
r
e
s
e
a
r
c
h,
w
e
f
oc
us
on
s
ta
ti
s
ti
c
a
l
da
ta
f
r
om
th
e
to
p
f
iv
e
E
ur
ope
a
n
le
a
gue
s
-
P
r
e
m
ie
r
le
a
gue
,
L
a
li
ga
,
B
unde
s
li
ga
,
S
e
r
ie
A
, a
nd
L
ig
ue
1,
s
p
e
c
if
ic
a
ll
y
f
r
om
th
e
2022
-
2023
s
e
a
s
on.
T
h
e
da
ta
s
e
t
in
c
lu
de
s
bot
h
out
f
ie
ld
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
. 14, No. 5, O
c
to
be
r
2025
:
3847
-
3857
3850
pl
a
ye
r
s
a
nd
go
a
lk
e
e
pe
r
s
,
a
nd
is
c
ur
a
te
d
to
e
ns
ur
e
c
ons
i
s
te
nc
y
i
n
th
e
m
e
tr
ic
s
a
na
ly
z
e
d
a
c
r
os
s
a
ll
le
a
gue
s
.
T
he
s
e
le
c
ti
on
of
f
e
a
tu
r
e
s
is
ba
s
e
d
on
th
e
ir
r
e
le
v
a
nc
e
in
m
e
a
s
ur
in
g
a
pl
a
ye
r
’
s
ove
r
a
ll
c
ont
r
ib
ut
io
n
a
nd
pos
it
io
na
l
e
f
f
e
c
ti
ve
ne
s
s
, c
ov
e
r
in
g of
f
e
ns
iv
e
, de
f
e
ns
iv
e
, a
nd t
r
a
n
s
it
io
na
l
a
s
pe
c
ts
of
t
he
ga
m
e
.
2
.
1
.
2
.
F
e
at
u
r
e
s
e
le
c
t
io
n
T
hi
r
te
e
n
ke
y
pe
r
f
or
m
a
nc
e
m
e
tr
ic
s
w
e
r
e
s
e
le
c
t
e
d
f
or
out
f
ie
ld
pl
a
ye
r
s
,
in
c
lu
di
ng:
s
hot
s
,
xG
,
xA
,
c
r
os
s
e
s
,
to
ta
l
pa
s
s
e
s
,
s
hor
t
p
a
s
s
e
s
(
<
32
m)
,
lo
ng
pa
s
s
e
s
(
≥3
2
m
)
,
pa
s
s
e
s
in
a
tt
a
c
ki
ng
th
ir
d,
pe
na
lt
y
a
r
e
a
e
nt
r
ie
s
,
ta
ke
-
ons
,
de
f
e
n
s
iv
e
a
c
ti
on
s
in
ow
n
th
ir
d,
de
f
e
ns
iv
e
a
c
t
io
ns
in
m
id
dl
e
th
ir
d,
a
nd
de
f
e
ns
iv
e
a
c
ti
ons
in
oppos
it
io
n
th
ir
d
.
F
or
goa
lk
e
e
pe
r
s
,
f
our
pos
it
io
n
-
s
pe
c
if
ic
m
e
tr
ic
s
w
e
r
e
c
on
s
id
e
r
e
d,
w
it
h
P
C
A
la
te
r
r
e
du
c
in
g
th
is
to
th
r
e
e
c
om
pone
nt
s
.
T
he
s
e
le
c
t
e
d
f
e
a
tu
r
e
s
e
n
s
ur
e
a
c
o
m
pr
e
he
ns
iv
e
r
e
pr
e
s
e
nt
a
ti
on
of
e
a
c
h
pl
a
y
e
r
'
s
in
-
ga
m
e
be
ha
vi
or
a
nd r
ol
e
.
2
.
1
.
3
.
D
at
a c
le
an
in
g an
d
i
n
t
e
gr
at
io
n
F
ol
lo
w
in
g
c
ol
le
c
ti
on,
th
e
da
ta
s
e
ts
w
e
r
e
m
e
r
ge
d
a
nd
c
le
a
n
e
d.
P
la
ye
r
s
w
it
h
in
c
om
pl
e
te
or
m
is
s
in
g
va
lu
e
s
in
th
e
s
e
le
c
te
d
f
e
a
tu
r
e
s
w
e
r
e
e
xc
lu
de
d
to
pr
e
s
e
r
ve
da
ta
in
te
gr
it
y
du
r
in
g
m
ode
li
ng.
T
he
f
in
a
l
da
ta
s
e
t
w
a
s
th
e
n
s
e
gm
e
nt
e
d
by
pos
it
io
n
(
out
f
ie
ld
vs
.
goa
lk
e
e
pe
r
)
to
f
a
c
il
it
a
te
f
e
a
tu
r
e
-
s
pe
c
if
ic
di
m
e
ns
io
na
li
ty
r
e
duc
ti
on a
nd mode
li
ng.
2
.2.
D
at
a e
xp
lo
r
at
io
n
A
f
te
r
c
om
pl
e
ti
ng
th
e
da
t
a
c
ol
le
c
ti
on
a
n
d
c
le
a
ni
ng
s
t
a
ge
s
,
a
n
e
xpl
or
a
to
r
y
d
a
ta
a
n
a
ly
s
i
s
(
E
D
A
)
w
a
s
c
ondu
c
te
d
to
ga
in
a
de
e
pe
r
u
nde
r
s
ta
ndi
ng
of
th
e
d
a
ta
s
e
t
a
n
d
to
gui
de
s
ub
s
e
que
nt
pr
e
pr
oc
e
s
s
in
g
de
c
i
s
io
n
s
. T
hi
s
s
te
p
is
e
s
s
e
nt
i
a
l
to
id
e
nt
if
y
s
tr
uc
t
ur
a
l
i
s
s
ue
s
,
a
s
s
e
s
s
t
he
c
om
pl
e
te
n
e
s
s
of
th
e
d
a
ta
[
2
2]
,
a
nd
pr
e
pa
r
e
it
f
or
di
m
e
n
s
io
n
a
li
ty
r
e
du
c
ti
on
a
nd
m
od
e
li
ng
[
23]
.
E
xpl
or
a
ti
on
w
a
s
c
ondu
c
te
d
u
s
in
g
th
e
P
a
nd
a
s
li
br
a
r
y
in
P
yt
hon
[
24]
,
w
hi
c
h
of
f
e
r
s
p
ow
e
r
f
ul
da
t
a
h
a
ndl
in
g
c
a
pa
b
il
it
ie
s
.
T
h
e
da
ta
s
e
t
w
a
s
f
ir
s
t
in
s
pe
c
te
d
u
s
in
g
t
he
.
s
h
a
pe
a
tt
r
ib
ut
e
to
und
e
r
s
t
a
nd
i
ts
s
tr
u
c
tu
r
e
.
F
or
o
ut
f
ie
ld
pl
a
ye
r
s
,
t
he
da
t
a
s
e
t
c
ons
is
t
e
d
of
2,82
7
r
ow
s
a
nd
151
c
ol
u
m
ns
,
r
e
pr
e
s
e
nt
in
g
in
di
vi
dua
l
pl
a
ye
r
s
a
nd
th
e
ir
pe
r
f
or
m
a
nc
e
f
e
a
tu
r
e
s
,
r
e
s
pe
c
ti
ve
ly
.
A
ke
y
f
oc
us
of
th
e
e
xpl
or
a
ti
on
w
a
s
th
e
ha
ndl
in
g
of
m
is
s
in
g
va
lu
e
s
,
w
hi
c
h
c
a
n
s
ig
ni
f
ic
a
nt
ly
im
pa
c
t
m
ode
l
a
c
c
ur
a
c
y.
M
is
s
in
g
va
lu
e
s
(
nul
l)
i
n
th
is
c
ont
e
xt
ty
pi
c
a
ll
y
a
r
is
e
due
to
pl
a
ye
r
s
not
r
e
c
or
di
ng
a
v
a
lu
e
in
a
pa
r
ti
c
ul
a
r
s
ta
ti
s
ti
c
a
l
c
a
te
gor
y
of
te
n
be
c
a
us
e
th
e
y
di
d
not
e
nga
ge
in
th
a
t
ty
pe
of
pl
a
y
dur
in
g
th
e
s
e
a
s
on.
T
o
qua
nt
if
y
th
is
,
th
e
is
nul
l(
)
f
unc
ti
on
w
a
s
us
e
d
in
c
onj
unc
ti
on
w
it
h
s
um
(
)
,
r
e
ve
a
li
ng
a
to
ta
l
of
4,585
m
i
s
s
in
g
e
nt
r
ie
s
a
c
r
os
s
va
r
io
us
c
ol
um
ns
.
R
a
th
e
r
th
a
n
im
put
in
g
pot
e
nt
ia
ll
y
bi
a
s
e
d
v
a
lu
e
s
,
r
ow
s
w
it
h
m
is
s
in
g
da
t
a
in
c
r
it
ic
a
l
f
e
a
tu
r
e
s
w
e
r
e
e
xc
lu
d
e
d
to
pr
e
s
e
r
ve
th
e
s
ta
ti
s
ti
c
a
l
in
te
gr
it
y
of
th
e
da
ta
s
e
t.
I
n
a
ddi
ti
on
to
m
is
s
in
g
d
a
ta
,
th
e
pr
e
s
e
nc
e
of
dupl
ic
a
te
e
nt
r
ie
s
w
a
s
a
l
s
o
in
ve
s
ti
ga
te
d.
D
upl
ic
a
te
r
e
c
or
ds
in
th
is
s
tu
dy
w
e
r
e
pr
im
a
r
il
y
du
e
to
pl
a
ye
r
s
tr
a
ns
f
e
r
r
in
g
be
twe
e
n
te
a
m
s
w
it
hi
n
th
e
s
a
m
e
s
e
a
s
on, whic
h r
e
s
ul
te
d i
n m
ul
ti
pl
e
e
nt
r
ie
s
unde
r
t
he
s
a
m
e
pl
a
ye
r
’
s
na
m
e
. T
hi
s
w
a
s
ve
r
if
ie
d us
in
g t
he
dupl
ic
a
te
d(
)
f
unc
ti
on
[
25]
on
th
e
'
pl
a
ye
r
'
c
ol
um
n,
w
he
r
e
70
dupl
ic
a
te
d
r
e
c
or
ds
w
e
r
e
id
e
nt
if
ie
d.
I
ns
te
a
d
of
e
li
m
in
a
ti
ng
dupl
ic
a
te
s
bl
in
dl
y,
dom
a
in
-
s
pe
c
if
ic
c
ons
id
e
r
a
ti
ons
w
e
r
e
a
ppl
ie
d:
th
e
m
os
t
c
om
pl
e
te
r
e
c
or
d
o
r
th
e
la
te
s
t
c
lu
b da
ta
f
or
t
he
s
e
a
s
on w
a
s
r
e
ta
in
e
d t
o e
ns
ur
e
r
e
le
v
a
nc
e
t
o r
e
c
r
ui
tm
e
nt
a
na
ly
s
is
.
2
.
3
.
D
at
a p
r
e
p
r
oc
e
s
s
in
g
A
t
th
is
s
ta
ge
,
th
e
m
a
jo
r
it
y
of
th
e
in
f
or
m
a
ti
on
on
th
e
da
ta
s
e
t
ha
s
be
e
n
known
th
a
nks
to
s
e
ve
r
a
l
s
ta
ge
s
th
a
t
ha
ve
be
e
n
c
a
r
r
ie
d
out
be
f
or
e
,
s
o
th
a
t
th
e
m
os
t
e
f
f
ic
ie
nt
s
ta
ge
c
a
n
be
c
a
r
r
ie
d
out
to
c
ont
in
ue
th
e
r
e
s
e
a
r
c
h.
I
n t
hi
s
pr
e
pr
oc
e
s
s
in
g s
ta
g
e
, i
t
is
di
vi
de
d i
nt
o s
e
ve
r
a
l
s
m
a
ll
e
r
s
ta
ge
s
, na
m
e
ly
:
i)
D
a
ta
c
le
a
ni
ng
s
ta
ge
f
or
r
e
dunda
nt
a
nd
va
lu
a
bl
e
da
ta
nul
l
om
it
te
d
f
r
om
da
ta
s
e
t
to
e
ns
ur
e
m
a
xi
m
um
m
ode
l
pe
r
f
or
m
a
nc
e
[
26]
.
O
n
th
e
c
ode
s
ni
ppe
t
be
lo
w
s
om
e
f
e
a
tu
r
e
d
da
ta
s
uc
h
a
s
nom
in
a
l
da
ta
s
uc
h
a
s
w
hi
c
h t
e
a
m
t
he
pl
a
ye
r
pl
a
ye
d i
n, t
he
a
ge
of
t
he
pl
a
ye
r
,
a
nd t
he
na
ti
ona
li
ty
of
t
he
pl
a
ye
r
a
r
e
o
m
it
te
d f
r
om
th
e
da
ta
s
e
t
s
o
th
a
t
th
e
onl
y
da
ta
le
f
t
is
qua
nt
it
a
ti
ve
da
t
a
th
a
t
w
il
l
be
us
e
d
a
s
da
ta
tr
a
in
in
g
f
or
th
e
m
ode
l
th
a
t
w
il
l
be
m
a
de
l
a
te
r
.
ii)
I
n
a
ddi
ti
on
to
th
e
d
a
ta
s
e
t
f
e
a
tu
r
e
,
a
dupl
ic
a
te
da
ta
c
a
n
a
ls
o
b
e
s
a
id
to
be
dupl
ic
a
te
,
a
c
c
or
di
ng
to
th
e
e
xpl
or
a
ti
on
c
a
r
r
ie
d
out
in
th
e
pr
e
vi
ous
s
ta
ge
,
it
is
known
th
a
t
t
he
r
e
a
r
e
70
dupl
ic
a
te
da
ta
in
th
e
'
pl
a
ye
r
'
c
ol
um
n
to
ove
r
c
om
e
th
i
s
,
e
a
c
h
pl
a
ye
r
w
il
l
be
gi
ve
n
a
uni
qu
e
i
d
a
nd
e
a
c
h
da
t
a
ow
ne
d
by
th
e
pl
a
ye
r
w
il
l
be
c
om
bi
ne
d i
nt
o one
t
o e
a
c
h unique
i
d.
iii)
A
t
th
e
da
ta
s
ta
ge
e
xpl
or
a
ti
on
pr
e
vi
ous
ly
,
it
w
a
s
f
ound
th
a
t
th
e
r
e
w
e
r
e
s
ti
ll
4
,
585
a
m
ount
s
of
da
t
a
th
a
t
w
e
r
e
nul
l
th
e
n
it
is
ne
c
e
s
s
a
r
y
to
ta
ke
s
te
ps
to
e
li
m
in
a
te
th
e
s
e
da
ta
,
by
us
in
g
f
unc
ti
on
ot
he
r
pr
ovi
de
d
li
br
a
r
y
P
a
nda
s
a
r
e
f
il
ln
a
(
)
[
27]
va
lu
a
bl
e
da
ta
nul
l
r
e
pl
a
c
e
d w
it
h
a
va
lu
e
of
z
e
r
o or
0 a
s
c
a
n be
s
e
e
n i
n t
he
c
ode
s
ni
ppe
t.
D
a
ta
th
a
t
is
nul
l
,
t
hi
s
i
s
r
e
pl
a
c
e
d
w
it
h
0
s
o
th
a
t
t
he
da
ta
r
e
m
a
in
s
c
on
s
is
te
nt
,
a
nd
th
e
s
ha
p
e
of
t
he
da
ta
i
s
l
e
s
s
s
ke
w
e
d t
o one
s
id
e
.
2
.
4
.
D
at
a t
r
an
s
f
or
m
at
io
n
O
nc
e
th
e
da
ta
is
c
le
a
n
of
r
e
dunda
nt
da
ta
a
nd
th
e
va
lu
e
of
nul
l
ne
xt
is
to
c
ha
nge
th
e
w
hol
e
da
ta
s
e
t
in
to
a
f
or
m
t
ha
t
s
uppor
ts
m
a
xi
m
um
m
ode
l
pe
r
f
o
r
m
a
nc
e
. I
n a
f
o
ot
ba
ll
s
ta
ti
s
ti
c
, a
pl
a
ye
r
w
ho ha
s
m
or
e
m
in
ut
e
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
R
e
c
om
m
e
ndat
io
n s
y
s
t
e
m
f
or
f
oot
bal
l
pl
ay
e
r
r
e
c
r
ui
tme
nt
us
in
g k
-
ne
ar
e
s
t
ne
ig
hbor
(
M
auk
a
r
)
3851
pl
a
ye
d t
e
nds
t
o ha
ve
a
hi
ghe
r
s
ta
ti
s
ti
c
a
l
va
lu
e
t
ha
n a
pl
a
ye
r
w
ho
ha
s
l
e
s
s
m
in
ut
e
s
of
pl
a
y da
ta
[
28]
. T
he
r
e
f
or
e
,
th
is
c
a
n
be
ove
r
c
om
e
to
nor
m
a
li
z
e
s
ta
ti
s
ti
c
a
l
da
ta
in
to
a
f
or
m
pe
r
90
m
in
ut
e
s
[
29]
na
m
e
ly
nor
m
a
li
z
in
g
th
e
da
ta
s
o t
ha
t
a
ll
pl
a
ye
r
da
ta
i
s
c
ons
id
e
r
e
d
e
qua
l
e
ve
n t
hough the
y
ha
ve
di
f
f
e
r
e
nt
m
in
ut
e
s
of
pl
a
y us
in
g
th
e
(
1)
.
90
=
×
90
(
1)
2.5. F
e
at
u
r
e
s
e
le
c
t
io
n
of
f
oot
b
al
l
p
la
ye
r
s
N
ot
a
ll
f
e
a
tu
r
e
d
a
t
da
ta
s
e
t
w
il
l
be
pa
r
t
of
th
e
tr
a
in
in
g
d
a
ta
a
ga
in
s
t
th
e
m
ode
l,
to
r
e
duc
e
th
e
unde
r
f
it
ti
ng
a
nd
ove
r
f
it
ti
ng
th
e
n
it
is
ne
c
e
s
s
a
r
y
to
s
e
le
c
t
th
e
da
ta
f
e
a
tu
r
e
or
f
e
a
tu
r
e
s
e
le
c
ti
on
[
30]
.
T
he
pl
a
ye
r
s
ta
ti
s
ti
c
a
l
da
ta
f
e
a
tu
r
e
is
s
e
le
c
te
d
th
e
be
s
t
to
r
e
pr
e
s
e
nt
e
a
c
h
pl
a
ye
r
,
w
hi
c
h
is
s
e
le
c
te
d
a
s
m
a
ny
a
s
13
f
e
a
tu
r
e
d
da
ta
f
or
e
a
c
h pl
a
ye
r
out
f
ie
ld
a
s
f
ol
lo
w
s
:
i)
S
hot
s
:
th
e
num
be
r
of
s
hot
s
t
he
pl
a
ye
r
ha
s
m
a
d
e
.
ii)
xG
:
th
e
pr
oba
bi
li
ty
o
f
a
pl
a
ye
r
s
c
or
in
g a
goa
l
in
e
a
c
h ki
c
k t
a
ke
n
(
on a
s
c
a
le
of
0
-
1)
.
iii)
xA
:
th
e
pr
oba
bi
li
ty
of
a
pl
a
ye
r
s
c
or
in
g
a
pa
s
s
th
a
t
w
il
l
be
c
onve
r
te
d
in
to
a
goa
l
by
a
te
a
m
m
a
te
(
on
a
s
c
a
le
of
0
-
1)
.
iv
)
C
r
os
s
e
s
:
th
e
num
be
r
of
t
im
e
s
a
pl
a
ye
r
m
a
ke
s
c
r
os
s
e
s
.
v)
T
ot
a
l
pa
s
s
e
s
:
th
e
to
ta
l
p
a
s
s
e
s
m
a
de
by t
h
e
pl
a
ye
r
.
vi
)
T
ot
a
l
s
hor
t
pa
s
s
e
s
(
<
32
m
)
:
a
s
hor
t
pa
s
s
or
pa
s
s
t
ha
t
m
ove
s
s
hor
te
r
t
ha
n 32 me
te
r
s
by a
pl
a
ye
r
.
vi
i)
T
ot
a
l
lo
ng
pa
s
s
e
s
(
≥
32
m
)
:
th
e
num
be
r
of
lo
ng
pa
s
s
e
s
or
pa
s
s
e
s
th
a
t
m
ove
m
or
e
th
a
n
32
m
e
te
r
s
by
a
pl
a
ye
r
.
vi
ii
)
P
a
s
s
e
s
i
n
a
tt
a
c
ki
ng t
hi
r
ds
:
th
e
numbe
r
of
pa
s
s
e
s
a
pl
a
y
e
r
m
a
ke
s
in
1/
3 of
t
he
f
ie
ld
i
n t
he
oppone
nt
'
s
a
r
e
a
.
ix
)
P
e
na
lt
y
a
r
e
a
e
nt
r
ie
s
:
th
e
num
be
r
of
pa
s
s
e
s
a
pl
a
y
e
r
m
a
ke
s
i
nt
o t
he
box.
x)
T
a
ke
-
ons
:
th
e
num
be
r
of
a
tt
e
m
pt
s
t
o pa
s
s
a
pl
a
ye
r
by dr
ib
bl
in
g.
xi
)
D
e
f
e
ns
iv
e
a
c
ti
ons
in
ow
n
th
ir
d:
th
e
num
be
r
of
de
f
e
ns
iv
e
a
c
ti
ons
pe
r
f
or
m
e
d
by
pl
a
ye
r
s
in
1/
3
of
th
e
i
r
ow
n f
ie
ld
.
xi
i)
D
e
f
e
ns
iv
e
a
c
ti
ons
in
m
id
dl
e
th
ir
d:
th
e
num
be
r
o
f
de
f
e
ns
iv
e
a
c
ti
ons
m
a
de
by
pl
a
ye
r
s
in
th
e
m
id
dl
e
o
f
th
e
f
ie
ld
.
xi
ii
)
D
e
f
e
ns
iv
e
a
c
ti
ons
in
oppo
s
it
io
n
th
ir
ds
:
th
e
num
be
r
of
de
f
e
ns
iv
e
a
c
ti
ons
a
pl
a
ye
r
pe
r
f
or
m
s
on
1/
3
of
th
e
f
ie
ld
i
n t
he
oppone
nt
'
s
a
r
e
a
.
I
n
th
is
s
tu
dy,
th
e
s
e
le
c
te
d
pl
a
ye
r
s
a
r
e
a
ls
o
pl
a
ye
r
s
w
ho
ha
v
e
90
m
in
ut
e
s
or
f
ul
l
pl
a
yi
ng
num
be
r
s
dur
in
g
a
m
a
tc
h
a
t
le
a
s
t
3
ti
m
e
s
to
r
e
duc
e
th
e
num
be
r
of
pl
a
ye
r
s
w
ho
ha
ve
m
in
im
a
l
da
t
a
s
a
m
pl
e
s
us
in
g
th
e
f
ol
lo
w
in
g P
s
e
udoc
ode
:
o
u
t
_
d
f
=
g
r
a
n
d
[
g
r
a
n
d
[
'
9
0
s
'
]
>
=
3
]
g
k
_
d
f
=
g
k
_
g
r
a
n
d
[
g
k
_
g
r
a
n
d
[
'
9
0
s
'
]
>
=
3
]
F
ur
th
e
r
m
or
e
,
it
w
i
ll
be
c
a
r
r
ie
d
out
di
m
e
ns
io
na
li
ty
r
e
duc
ti
on
u
s
i
ng
th
e
P
C
A
pr
ovi
de
d
by
L
ib
r
a
r
y
s
kl
e
a
r
n
[
31]
,
a
t
th
is
s
ta
ge
th
e
da
ta
s
e
t
w
il
l
a
ga
in
be
s
e
le
c
te
d
f
or
th
e
da
ta
f
e
a
tu
r
e
s
th
a
t
be
s
t
r
e
pr
e
s
e
nt
th
e
e
nt
ir
e
in
f
or
m
a
ti
o
n
f
r
om
t
he
da
ta
us
in
g t
he
f
ol
lo
w
in
g ps
e
udoc
ode
:
p
c
a
=
d
e
c
o
m
p
o
s
i
t
i
o
n
.
P
C
A
(
)
p
c
a
.
n
_
c
o
m
p
o
n
e
n
t
s
=
1
3
p
c
a
_
d
a
t
a
=
p
c
a
.
f
i
t
_
t
r
a
n
s
f
o
r
m
(
o
u
t
_
d
a
t
a
)
A
t
th
e
pr
e
pr
oc
e
s
s
in
g
c
on
s
is
ts
of
f
e
a
tu
r
e
s
e
l
e
c
ti
on
us
in
g
P
C
A
w
he
r
e
th
e
da
ta
di
m
e
ns
io
n
is
r
e
duc
e
d
but
s
ti
ll
r
e
ta
in
s
th
e
m
a
jo
r
it
y
of
th
e
in
f
or
m
a
ti
on
[
32]
da
ta
,
th
e
r
e
by
r
e
duc
in
g
th
e
pos
s
ib
il
it
y
of
unde
r
f
it
t
in
g
a
nd
ove
r
f
it
ti
ng
.
D
im
e
ns
io
n
r
e
duc
ti
on
us
in
g
P
C
A
is
c
a
r
r
ie
d
out
on
two
da
ta
,
na
m
e
ly
pl
a
y
e
r
da
ta
w
it
h
pos
it
io
n
out
f
ie
ld
a
nd a
ls
o pl
a
ye
r
s
w
it
h t
he
goa
lk
e
e
pe
r
pos
it
io
n.
2.6. Cr
e
at
io
n
of
r
e
c
o
m
m
e
n
d
at
io
n
s
ys
t
e
m
A
t
th
is
s
ta
g
e
,
th
e
d
a
ta
s
e
t
i
s
r
e
a
dy
to
be
f
e
d
in
to
th
e
m
ode
l
to
b
e
c
r
e
a
te
d,
th
e
m
ode
l
w
il
l
tr
y
to
pr
e
di
c
t
th
e
di
s
ta
nc
e
be
twe
e
n
da
ta
poi
nt
s
us
in
g
th
e
KNN
a
lg
or
it
hm
a
nd
th
e
c
os
in
e
s
im
il
a
r
it
y
m
e
tr
ic
.
T
he
da
ta
s
e
t
us
e
d
is
da
ta
th
a
t
h
a
s
b
e
e
n
di
m
e
n
s
io
na
li
ty
r
e
duc
ti
on
be
f
or
e
.
L
ib
r
a
r
y
S
c
iP
y
pr
ovi
de
s
a
f
unc
ti
on
na
m
e
d
di
s
ta
nc
e
w
hi
c
h
w
il
l
w
or
k
by
pe
r
f
or
m
in
g
c
om
put
a
ti
ons
be
twe
e
n
two
or
m
or
e
da
ta
poi
nt
s
in
N
-
di
m
e
ns
io
na
l
s
pa
c
e
[
33]
.
I
n
th
e
im
pl
e
m
e
nt
a
ti
on,
e
a
c
h
pl
a
y
e
r
is
a
s
s
ig
ne
d
a
uni
que
id
e
nt
i
f
ie
r
,
w
hi
c
h
is
th
e
n
us
e
d
to
s
y
s
te
m
a
ti
c
a
ll
y
p
a
ir
a
nd
c
om
pa
r
e
pl
a
ye
r
s
in
a
lo
op
e
d
pr
oc
e
s
s
.
T
he
c
om
pa
r
is
on
is
c
onduc
te
d
it
e
r
a
ti
ve
ly
u
s
in
g
a
di
s
ta
nc
e
f
unc
ti
on
th
a
t
c
a
lc
ul
a
te
s
s
im
il
a
r
it
y
s
c
or
e
s
be
twe
e
n
a
ll
pos
s
ib
le
pl
a
ye
r
pa
i
r
s
.
F
ur
th
e
r
m
or
e
,
th
e
r
e
s
ul
t
da
ta
is
no
r
m
a
li
z
e
d
on
a
s
c
a
le
of
0
-
100
to
obt
a
in
da
ta
th
a
t
is
e
xc
lu
s
iv
e
a
c
r
os
s
a
ll
c
om
pone
nt
s
da
ta
s
e
t
.
A
ll
r
e
s
ul
ts
f
r
om
th
e
m
ode
l
a
r
e
th
e
n
f
e
d
in
to
th
e
f
or
m
of
pi
c
kl
e
or
a
f
or
m
of
s
to
r
a
ge
pr
ovi
de
d
by
P
yt
hon
s
o
th
a
t
th
e
da
ta
f
r
om
m
ode
l
tr
a
in
in
g c
a
n be
s
to
r
e
d a
nd r
e
us
e
d l
a
te
r
[
34]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
. 14, No. 5, O
c
to
be
r
2025
:
3847
-
3857
3852
3.
R
E
S
U
L
T
S
A
N
D
D
I
S
C
U
S
S
I
O
N
T
o
r
e
duc
e
di
m
e
ns
io
na
li
ty
a
nd
im
pr
ove
m
ode
l
e
f
f
ic
ie
nc
y,
P
C
A
w
a
s
a
ppl
ie
d
s
e
pa
r
a
te
ly
to
out
f
ie
ld
pl
a
ye
r
s
a
nd
goa
lk
e
e
p
e
r
s
.
T
he
go
a
l
w
a
s
to
r
e
ta
in
th
e
m
a
jo
r
it
y
of
va
r
ia
nc
e
(
≥95%
)
w
hi
le
m
in
im
iz
in
g
th
e
num
be
r
of
f
e
a
tu
r
e
s
to
a
voi
d
ove
r
f
it
ti
ng
a
nd
unde
r
f
it
t
in
g
dur
i
ng
K
N
N
-
ba
s
e
d
s
im
il
a
r
it
y
m
a
tc
hi
ng.
F
ig
ur
e
2
s
how
s
th
e
e
xpl
a
in
e
d
va
r
ia
nc
e
r
a
ti
o
f
or
out
f
ie
ld
p
la
ye
r
da
ta
.
O
r
ig
in
a
ll
y,
12
f
e
a
tu
r
e
s
w
e
r
e
us
e
d.
A
f
te
r
pe
r
f
or
m
in
g
P
C
A
,
it
w
a
s
obs
e
r
ve
d
th
a
t
10
c
om
pone
nt
s
r
e
ta
in
e
d
98.7%
of
th
e
to
ta
l
va
r
ia
nc
e
,
m
a
ki
ng
th
e
m
s
uf
f
ic
ie
nt
f
or
r
e
pr
e
s
e
nt
in
g
th
e
da
ta
w
it
hout
s
ig
ni
f
ic
a
nt
in
f
or
m
a
ti
on
lo
s
s
.
T
hi
s
s
e
le
c
ti
on
w
a
s
m
a
de
u
s
in
g
th
e
c
um
ul
a
ti
ve
e
xpl
a
in
e
d va
r
ia
nc
e
t
hr
e
s
hol
d, a
s
ta
nd
a
r
d pr
a
c
ti
c
e
i
n
P
C
A
-
ba
s
e
d m
ode
li
ng.
S
im
il
a
r
ly
,
goa
lk
e
e
pe
r
da
ta
be
ga
n
w
i
th
4
f
e
a
tu
r
e
s
.
B
a
s
e
d
on
th
e
r
e
s
ul
ts
of
th
e
P
C
A
,
onl
y
th
r
e
e
pr
in
c
ip
a
l
c
om
pone
nt
s
w
e
r
e
r
e
qui
r
e
d
to
r
e
ta
in
100%
of
th
e
va
r
ia
nc
e
in
th
e
da
ta
s
e
t.
T
hi
s
de
m
ons
tr
a
te
s
P
C
A
'
s
e
f
f
e
c
ti
ve
ne
s
s
in
ha
ndl
in
g
bot
h
la
r
ge
a
nd
c
om
pa
c
t
f
e
a
tu
r
e
s
e
ts
,
e
s
pe
c
ia
ll
y
f
or
hi
ghl
y s
pe
c
ia
li
z
e
d
pos
it
io
ns
s
u
c
h
a
s
goa
lk
e
e
pe
r
s
.
T
he
di
m
e
n
s
io
na
li
ty
r
e
duc
ti
on
pr
oc
e
s
s
s
ig
ni
f
ic
a
nt
ly
r
e
duc
e
d
noi
s
e
in
th
e
da
ta
a
nd
im
pr
ove
d
th
e
c
om
put
a
ti
ona
l
pe
r
f
or
m
a
nc
e
of
th
e
m
ode
l
w
hi
le
pr
e
s
e
r
vi
ng
th
e
unde
r
ly
in
g
pl
a
ye
r
be
ha
vi
or
pa
tt
e
r
ns
.
H
ow
e
ve
r
,
w
hi
le
P
C
A
e
f
f
e
c
ti
ve
ly
r
e
duc
e
s
f
e
a
tu
r
e
s
pa
c
e
a
nd
m
a
in
ta
in
s
hi
gh
in
f
or
m
a
ti
on
r
e
te
nt
io
n,
it
is
e
s
s
e
nt
ia
l
to
c
r
it
ic
a
ll
y e
xa
m
in
e
i
ts
l
im
it
a
ti
ons
. S
ugge
s
te
d a
lt
e
r
na
t
iv
e
s
f
or
f
ut
ur
e
w
or
k c
a
n be
s
e
e
n i
n T
a
bl
e
1.
F
ig
ur
e
2. P
C
A
f
or
out
f
ie
ld
pos
it
io
n pl
a
ye
r
da
ta
T
a
bl
e
1. S
ugge
s
t
e
d a
lt
e
r
na
ti
ve
s
f
or
f
ut
ur
e
w
or
k
T
e
c
hni
que
S
t
r
e
ngt
hs
L
i
m
i
t
a
t
i
ons
P
C
A
F
a
s
t
, r
e
t
a
i
ns
m
a
xi
m
um
va
r
i
a
nc
e
, w
i
de
l
y us
e
d
L
i
ne
a
r
, l
e
s
s
i
nt
e
r
pr
e
t
a
bl
e
t
-
di
s
t
r
i
but
e
d
s
t
oc
ha
s
t
i
c
ne
i
ghbor
e
m
be
ddi
ng
C
a
pt
ur
e
s
nonl
i
ne
a
r
r
e
l
a
t
i
ons
hi
ps
w
e
l
l
C
om
put
a
t
i
ona
l
l
y e
xpe
ns
i
ve
, poor
f
or
ne
w
da
t
a
U
ni
f
or
m
m
a
ni
f
ol
d
a
pp
r
oxi
m
a
t
i
on
a
nd pr
oj
e
c
t
i
on
P
r
e
s
e
r
ve
s
bot
h l
oc
a
l
a
nd
gl
oba
l
s
t
r
uc
t
ur
e
M
a
y r
e
qui
r
e
f
i
ne
-
t
uni
ng, l
e
s
s
i
nt
e
r
pr
e
t
a
bl
e
A
ut
oe
nc
ode
r
s
L
e
a
r
ns
de
e
p, nonl
i
ne
a
r
f
e
a
t
ur
e
s
R
e
qui
r
e
s
l
a
r
ge
r
da
t
a
s
e
t
s
a
nd t
r
a
i
ni
ng t
i
m
e
3.1.
I
m
p
ac
t
on
r
e
c
o
m
m
e
n
d
at
io
n
r
e
s
u
lt
s
T
he
r
e
s
ul
ts
of
th
e
r
e
duc
ti
on
by
s
e
le
c
ti
ng
f
e
a
tu
r
e
s
us
in
g
P
C
A
s
uc
c
e
e
d
e
d
in
r
e
c
om
m
e
ndi
ng
pl
a
ye
r
s
w
it
h
pa
tt
e
r
ns
th
a
t
c
lo
s
e
ly
m
a
tc
h
ba
s
e
d
on
f
oot
ba
ll
-
s
pe
c
if
ic
be
h
a
vi
or
.
F
or
e
xa
m
pl
e
,
a
s
s
how
n
in
F
ig
ur
e
3,
th
e
r
e
c
om
m
e
nda
ti
on s
ys
te
m
i
de
nt
if
ie
d R
iy
a
d M
a
hr
e
z
a
s
t
he
m
os
t
s
i
m
il
a
r
pl
a
ye
r
t
o L
io
ne
l
M
e
s
s
i,
w
it
h a
s
im
il
a
r
it
y
s
c
or
e
of
91.85%
.
T
hi
s
r
e
s
ul
t
il
lu
s
tr
a
te
s
th
e
s
tr
e
ngt
h
of
P
C
A
in
c
a
pt
ur
in
g
la
te
nt
be
ha
vi
or
a
l
p
a
tt
e
r
ns
a
c
r
o
s
s
hi
gh
-
di
m
e
ns
io
na
l
f
oot
ba
ll
pe
r
f
or
m
a
nc
e
da
ta
.
B
y
f
il
te
r
in
g
ou
t
l
e
s
s
r
e
le
va
nt
f
e
a
tu
r
e
s
a
nd
r
e
ta
in
in
g
th
os
e
th
a
t
c
ont
r
ib
ut
e
m
os
t
to
th
e
va
r
ia
nc
e
,
P
C
A
e
na
bl
e
s
th
e
r
e
c
om
m
e
nda
ti
on
s
ys
te
m
to
m
a
tc
h
pl
a
ye
r
s
w
it
h
r
e
m
a
r
ka
bl
e
pr
e
c
is
io
n.
T
he
s
im
il
a
r
it
y
s
c
or
e
of
91.85
%
be
twe
e
n
L
io
ne
l
M
e
s
s
i
a
nd
R
iy
a
d
M
a
hr
e
z
in
F
ig
ur
e
3,
r
e
f
le
c
ts
how
P
C
A
pr
e
s
e
r
ve
s
nua
n
c
e
d pl
a
yi
ng s
ty
le
s
w
hi
le
e
nha
n
c
in
g c
om
put
a
ti
ona
l
e
f
f
ic
ie
nc
y.
F
ig
ur
e
3. E
xa
m
pl
e
of
r
e
c
om
m
e
nda
ti
on s
ys
te
m
a
f
te
r
f
e
a
tu
r
e
s
e
le
c
ti
on us
in
g P
C
A
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
R
e
c
om
m
e
ndat
io
n s
y
s
t
e
m
f
or
f
oot
bal
l
pl
ay
e
r
r
e
c
r
ui
tme
nt
us
in
g k
-
ne
ar
e
s
t
ne
ig
hbor
(
M
auk
a
r
)
3853
3.
2
.
R
e
s
u
lt
s
of
s
h
ot
s
p
os
it
io
n
d
is
t
r
ib
u
t
io
n
r
e
p
r
e
s
e
n
t
at
io
n
an
al
ys
is
on
t
h
e
f
oot
b
al
l
f
ie
ld
F
or
e
xa
m
pl
e
, F
ig
ur
e
4
s
how
s
a
r
e
pr
e
s
e
nt
a
ti
on of
s
hot
s
pos
it
io
n di
s
tr
ib
ut
io
n on the
f
oo
tb
a
ll
f
ie
ld
[
35]
.
P
la
ye
r
s
w
it
h t
he
goa
lk
e
e
pe
r
pos
it
io
n, t
he
f
e
a
tu
r
e
w
il
l
be
s
e
pa
r
a
t
e
d i
nt
o t
hr
e
e
, na
m
e
ly
:
i)
A
ve
r
a
ge
pa
s
s
l
e
ngt
h:
t
he
num
be
r
of
goa
lk
e
e
pe
r
s
pa
s
s
di
s
ta
nc
e
s
(
in
m
e
te
r
s
)
.
ii)
A
ve
r
a
ge
goa
l
ki
c
k l
e
ngt
h:
t
he
to
ta
l
di
s
ta
nc
e
of
a
goa
lk
e
e
pe
r
(
in
m
e
te
r
s
)
.
iii)
C
r
os
s
e
s
s
to
ppe
d:
t
he
num
be
r
of
c
r
os
s
e
s
t
ha
t
th
e
goa
lk
e
e
pe
r
ha
s
s
to
ppe
d.
T
he
r
e
s
ul
ts
of
th
e
f
in
di
ngs
s
how
th
a
t
ba
s
e
d
on
a
ll
s
ta
ti
s
ti
c
s
,
th
e
pl
a
ye
r
s
w
ho
a
r
e
r
e
pr
e
s
e
nt
e
d
a
c
c
or
di
ng
to
th
e
pos
it
io
n
a
nd
r
ol
e
pl
a
y
e
d
in
a
te
a
m
a
r
e
s
e
le
c
te
d
a
c
c
ur
a
te
ly
a
nd
c
a
n
be
c
a
te
gor
iz
e
d
in
to
f
our
c
a
te
gor
ie
s
th
a
t
de
s
c
r
ib
e
th
e
a
bi
li
ti
e
s
of
out
f
ie
ld
pl
a
ye
r
s
th
a
t
c
a
n
be
s
e
e
n
a
s
in
T
a
bl
e
2.
O
ne
of
th
e
pr
of
il
e
s
of
c
ha
m
pi
ons
le
a
gue
f
oot
ba
ll
pl
a
y
e
r
s
A
le
x
a
nde
r
I
s
a
k
f
r
om
B
or
u
s
s
ia
D
or
tm
und
c
a
n
b
e
s
e
e
n
in
F
ig
ur
e
5
w
h
e
n
vi
s
it
in
g
A
th
li
ti
kos
P
odos
f
e
r
ik
os
O
m
il
os
E
ll
in
on
L
e
f
kos
ia
s
(
A
P
O
E
L
)
N
ic
os
ia
i
n
O
c
to
be
r
2017
[
36]
. A
s
c
a
n be
s
e
e
n
in
one
e
xa
m
pl
e
of
a
E
ur
ope
a
n
le
a
gue
pr
of
il
e
,
S
w
e
de
n s
tr
ik
e
r
A
le
xa
nde
r
I
s
a
k
ha
s
40 c
a
ps
w
it
h
two
goa
ls
in
th
e
P
r
e
m
ie
r
le
a
gue
f
r
om
hi
s
f
ir
s
t
th
r
e
e
ga
m
e
s
.
T
he
pl
a
ye
r
s
ti
ll
s
c
or
e
s
10
goa
ls
in
th
e
le
a
gu
e
(
e
ig
ht
non
-
pe
na
lt
y
goa
ls
)
.
B
a
s
e
d
on
th
e
s
ta
ti
s
ti
c
s
in
F
ig
ur
e
5,
out
of
a
to
ta
l
of
52
s
hot
s
(
32
r
ig
ht
f
oot
a
nd
9
le
f
t
f
oot
)
,
he
a
d i
s
11 with a
n a
c
c
um
ul
a
ti
on of
xG
of
6.7 a
nd x
G
pe
r
s
hot
of
0.14 a
nd puts
t
he
pl
a
ye
r
a
s
a
c
e
nt
e
r
f
or
w
a
r
d.
P
C
A
a
ls
o
pr
ove
s
b
e
ne
f
ic
ia
l
in
a
na
ly
z
in
g
di
s
tr
ib
ut
io
n
-
ba
s
e
d
f
e
a
tu
r
e
s
s
u
c
h
a
s
s
hot
s
a
nd
pos
it
io
na
l
be
ha
vi
or
s
.
A
s
de
m
ons
tr
a
te
d
in
F
ig
ur
e
5
,
th
e
pr
of
il
e
of
A
le
xa
nde
r
I
s
a
k,
P
C
A
a
bs
tr
a
c
ts
c
om
pl
e
x
f
e
a
tu
r
e
in
te
r
a
c
ti
ons
(
e
.g.,
f
oot
pr
e
f
e
r
e
nc
e
,
xG
e
f
f
ic
ie
nc
y,
a
nd
h
e
a
di
ng
a
bi
li
ty
)
in
to
c
onc
is
e
c
om
pone
nt
s
.
T
he
s
e
c
om
pone
nt
s
r
e
ta
in
th
e
e
s
s
e
nt
ia
l
va
r
ia
nc
e
r
e
qui
r
e
d
to
di
f
f
e
r
e
n
ti
a
te
pl
a
ye
r
ty
pe
s
a
nd
r
ol
e
s
.
T
hi
s
a
b
s
tr
a
c
ti
on
e
na
bl
e
s
th
e
r
e
c
om
m
e
nda
ti
on
e
ngi
ne
to
m
a
tc
h
pl
a
ye
r
s
not
on
ly
on
a
bs
ol
ut
e
s
hoot
in
g
m
e
tr
ic
s
but
a
ls
o
on
nua
nc
e
d,
c
a
te
gor
y
-
le
ve
l
be
ha
vi
or
pa
tt
e
r
n
s
.
T
he
pr
e
s
e
r
va
ti
on
of
pl
a
ye
r
c
ha
r
a
c
te
r
is
ti
c
s
a
c
r
os
s
f
e
a
tu
r
e
c
a
te
gor
ie
s
,
s
uc
h
a
s
s
hoot
in
g
a
bi
li
ty
a
nd
s
ur
vi
va
bi
li
ty
,
f
ur
th
e
r
il
lu
s
tr
a
te
s
P
C
A
'
s
s
tr
e
ngt
h
in
m
a
in
ta
in
in
g
f
oot
ba
ll
-
s
pe
c
if
ic
c
ont
e
xt
w
it
hi
n a
r
e
duc
e
d di
m
e
ns
io
na
l
s
p
a
c
e
.
F
ig
ur
e
4. S
hot
pos
it
io
n di
s
tr
ib
ut
io
n on the
f
oot
ba
ll
f
ie
ld
[
35]
T
a
bl
e
2. C
a
te
gor
y
di
vi
s
io
n of
f
e
a
tu
r
e
da
ta
It
C
a
t
e
gor
y
F
e
a
t
ur
e
d
1
S
hoot
i
ng a
bi
l
i
t
y
S
hot
s
, xG
2.
B
a
i
t
a
bi
l
i
t
y
xA
, c
r
os
s
e
s
, t
ot
a
l
pa
s
s
e
s
,
t
ot
a
l
s
hor
t
pa
s
s
e
s
, t
ot
a
l
l
ong pa
s
s
e
s
,
pa
s
s
e
s
i
n a
t
t
a
c
ki
ng t
hi
r
ds
, pe
na
l
t
y a
r
e
a
e
nt
r
i
e
s
3
B
a
l
l
-
c
a
r
r
yi
ng a
bi
l
i
t
y
T
a
ke
-
ons
4
S
ur
vi
va
bi
l
i
t
y
D
e
f
e
ns
i
ve
a
c
t
i
ons
i
n ow
n t
hi
r
d, de
f
e
ns
i
ve
a
c
t
i
ons
i
n m
i
ddl
e
t
hi
r
d,
de
f
e
ns
i
ve
a
c
t
i
ons
i
n oppo
s
i
t
i
on t
hi
r
ds
F
ig
ur
e
5. T
he
pr
of
il
e
of
one
of
t
he
c
ha
m
pi
ons
l
e
a
gue
f
oot
ba
ll
pl
a
ye
r
s
[
36]
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
. 14, No. 5, O
c
to
be
r
2025
:
3847
-
3857
3854
3.
3
.
P
r
e
d
ic
t
io
n
s
ys
t
e
m
r
e
s
u
lt
s
b
as
e
d
on
p
la
ye
r
t
al
e
n
t
A
f
te
r
th
e
de
ve
lo
pm
e
nt
a
nd
e
v
a
lu
a
ti
on
of
th
e
r
e
c
om
m
e
nda
t
io
n
m
ode
l,
th
e
f
in
a
l
s
te
p
in
vol
ve
d
de
pl
oyi
ng
th
e
tr
a
in
e
d
m
ode
l
to
a
n
in
te
r
a
c
ti
ve
a
nd
us
e
r
-
f
r
ie
ndl
y
pl
a
tf
or
m
.
T
hi
s
w
a
s
a
c
c
om
pl
is
h
e
d
us
in
g
S
tr
e
a
m
li
t,
a
P
yt
hon
-
ba
s
e
d
f
r
a
m
e
w
or
k
id
e
a
l
f
or
bui
ld
in
g
w
e
b
a
ppl
ic
a
ti
ons
f
or
da
ta
s
c
i
e
nc
e
pr
oj
e
c
ts
.
T
he
de
pl
oym
e
nt
w
a
s
c
a
r
r
ie
d
out
w
it
hi
n
a
G
oogl
e
C
ol
a
b
not
e
book,
in
te
gr
a
ti
ng
th
e
tr
a
in
e
d
m
ode
l,
pl
a
ye
r
da
ta
s
e
t,
a
nd i
nt
e
r
f
a
c
e
l
ogi
c
.
B
e
f
or
e
de
pl
oym
e
nt
,
di
m
e
ns
io
na
li
ty
r
e
duc
ti
on
us
in
g
P
C
A
w
a
s
a
ppl
ie
d
to
th
e
da
ta
s
e
t.
T
hi
s
s
te
p
w
a
s
e
s
s
e
nt
ia
l
f
or
im
pr
ovi
ng
c
om
put
a
ti
ona
l
e
f
f
ic
ie
nc
y
a
nd
e
li
m
in
a
ti
ng
noi
s
e
f
r
om
ir
r
e
le
va
nt
f
e
a
tu
r
e
s
.
T
he
num
be
r
of
r
e
ta
in
e
d
pr
in
c
ip
a
l
c
om
pone
nt
s
w
a
s
de
te
r
m
in
e
d
b
a
s
e
d
on
th
e
c
um
ul
a
ti
ve
e
xpl
a
in
e
d
va
r
ia
nc
e
c
r
it
e
r
io
n,
e
ns
ur
in
g
th
a
t
a
t
le
a
s
t
95%
of
th
e
da
ta
s
e
t’
s
va
r
ia
nc
e
w
a
s
pr
e
s
e
r
ve
d.
F
or
out
f
ie
ld
pl
a
ye
r
s
,
th
is
r
e
s
ul
te
d
in
th
e
r
e
te
nt
io
n
of
11
pr
in
c
ip
a
l
c
om
pone
nt
s
,
w
hi
le
f
or
goa
lk
e
e
pe
r
s
,
3
c
om
pone
nt
s
w
e
r
e
s
uf
f
ic
ie
nt
due
to
th
e
ir
m
or
e
s
pe
c
ia
li
z
e
d
a
nd
f
e
w
e
r
pe
r
f
or
m
a
nc
e
m
e
tr
ic
s
.
T
hi
s
s
e
le
c
ti
on
ba
la
nc
e
s
di
m
e
n
s
io
na
li
ty
r
e
duc
ti
on
w
it
h
in
f
or
m
a
ti
on
p
r
e
s
e
r
va
ti
on,
opt
im
iz
in
g
th
e
pe
r
f
or
m
a
nc
e
o
f
th
e
K
N
N
a
lg
or
it
hm
w
hi
le
m
in
im
iz
in
g
ove
r
f
it
ti
n
g
a
nd
c
om
put
a
ti
ona
l
ove
r
he
a
d.
T
he
d
e
pl
oye
d
s
y
s
te
m
e
na
bl
e
s
u
s
e
r
s
,
s
u
c
h
a
s
a
na
ly
s
t
s
,
s
c
out
s
,
or
c
oa
c
he
s
to
id
e
nt
if
y
pl
a
ye
r
s
w
it
h
s
im
il
a
r
s
ta
ti
s
ti
c
a
l
pr
of
il
e
s
ba
s
e
d
on
hi
s
to
r
ic
a
l
m
a
tc
h
da
ta
.
T
he
ke
y
f
e
a
tu
r
e
s
of
th
e
pl
a
tf
or
m
i
nc
lu
de
:
‒
P
la
ye
r
ty
pe
s
e
l
e
c
ti
on
:
us
e
r
s
c
a
n
s
p
e
c
if
y
w
he
th
e
r
th
e
que
r
y
ta
r
ge
ts
a
n
out
f
ie
ld
pl
a
y
e
r
or
a
go
a
lk
e
e
pe
r
,
e
na
bl
in
g pos
it
io
n
-
s
pe
c
if
ic
r
e
c
om
m
e
nda
ti
ons
.
‒
S
e
a
r
c
h
f
unc
ti
ona
li
ty
:
a
te
xt
in
put
a
ll
ow
s
us
e
r
s
to
s
e
a
r
c
h
f
or
a
pl
a
ye
r
by
na
m
e
.
U
pon
s
e
le
c
ti
on,
th
e
pl
a
ye
r
’
s
c
lu
b a
nd othe
r
ba
s
ic
i
nf
or
m
a
ti
on a
r
e
di
s
pl
a
ye
d.
‒
C
us
to
m
iz
a
bl
e
r
e
s
ul
t
c
ount
:
by
de
f
a
ul
t
,
th
e
s
ys
te
m
r
e
tu
r
ns
f
iv
e
s
im
il
a
r
pl
a
ye
r
s
,
but
us
e
r
s
c
a
n
m
odi
f
y
th
is
to
di
s
pl
a
y be
twe
e
n t
hr
e
e
a
nd t
e
n r
e
c
om
m
e
nda
ti
ons
.
‒
L
e
a
gue
f
il
te
r
:
a
dr
opdown
m
e
nu
a
ll
ow
s
f
il
te
r
in
g
r
e
s
ul
ts
by
s
pe
c
if
ic
le
a
gue
s
(
e
.g.,
P
r
e
m
ie
r
le
a
gue
a
nd
L
a
li
ga
)
, w
it
h t
he
de
f
a
ul
t
s
e
t
to
i
nc
lu
de
a
ll
l
e
a
gue
s
.
‒
P
os
it
io
n
m
a
tc
hi
ng
f
il
te
r
:
us
e
r
s
c
a
n
c
hoo
s
e
w
he
th
e
r
to
di
s
pl
a
y
onl
y
pl
a
ye
r
s
f
r
om
th
e
s
a
m
e
pos
it
io
n
or
f
r
om
a
ny pos
it
io
n. B
y de
f
a
ul
t,
a
ll
pos
it
io
ns
a
r
e
i
nc
lu
de
d.
‒
A
ge
f
il
te
r
:
a
s
li
de
r
pr
ovi
de
s
f
il
te
r
in
g
ba
s
e
d
on
pl
a
ye
r
a
ge
,
r
a
ngi
ng
f
r
om
15
to
45
ye
a
r
s
,
w
it
h
a
de
f
a
ul
t
r
a
nge
of
15 t
o 41 ye
a
r
s
t
o m
a
tc
h c
om
m
on pr
of
e
s
s
io
na
l
c
a
r
e
e
r
s
pa
ns
.
T
he
out
put
i
s
pr
e
s
e
nt
e
d i
n a
s
or
ta
bl
e
t
a
bl
e
t
ha
t
di
s
pl
a
ys
:
th
e
na
m
e
of
e
a
c
h r
e
c
om
m
e
nde
d pl
a
ye
r
,
th
e
ir
s
im
il
a
r
it
y
pe
r
c
e
nt
a
ge
(
ba
s
e
d
on
c
os
in
e
s
im
il
a
r
it
y)
,
pos
it
io
n,
le
a
gue
,
a
ge
,
a
nd
to
ta
l
num
be
r
of
m
a
tc
he
s
pl
a
ye
d
(
e
xpr
e
s
s
e
d
in
90
-
m
in
ut
e
e
qui
va
le
nt
s
)
.
A
n
e
xa
m
pl
e
of
th
e
s
y
s
te
m
’
s
in
te
r
f
a
c
e
a
nd
r
e
c
om
m
e
nda
ti
on
r
e
s
ul
t
s
is
pr
e
s
e
nt
e
d
in
F
ig
ur
e
6,
w
h
e
r
e
us
e
r
s
c
a
n
in
te
r
a
c
ti
ve
ly
e
xpl
or
e
a
n
d
e
va
lu
a
te
pl
a
ye
r
s
w
ho
m
o
s
t
c
lo
s
e
ly
r
e
s
e
m
bl
e
th
e
s
ta
ti
s
ti
c
a
l
pr
of
il
e
of
a
s
e
le
c
te
d
in
di
vi
dua
l.
T
he
s
y
s
te
m
'
s
de
s
ig
n
e
m
pha
s
iz
e
s
tr
a
ns
pa
r
e
nc
y
a
nd
f
le
xi
bi
li
ty
,
m
a
ki
ng i
t
a
n e
f
f
e
c
ti
ve
de
c
is
io
n
-
s
uppor
t
to
ol
i
n t
he
c
ont
e
xt
of
t
a
le
nt
i
de
nt
if
ic
a
ti
on a
nd r
e
c
r
ui
tm
e
nt
.
F
ig
ur
e
6. P
la
ye
r
r
e
c
om
m
e
nda
ti
on s
ys
te
m
3.
4
.
C
om
p
ar
is
on
w
it
h
p
r
e
vi
ou
s
s
t
u
d
ie
s
an
d
p
e
r
f
or
m
an
c
e
e
v
al
u
at
io
n
T
a
bl
e
3
pr
ovi
de
s
a
s
um
m
a
r
y
of
r
e
le
va
nt
pr
io
r
r
e
s
e
a
r
c
h
in
th
e
dom
a
in
of
f
oot
ba
ll
pl
a
ye
r
pr
e
di
c
ti
on
a
nd
r
e
c
om
m
e
nda
ti
on
s
ys
te
m
s
,
hi
ghl
ig
ht
in
g
th
e
ir
m
e
th
ods
,
li
m
it
a
ti
ons
,
a
nd
th
e
c
om
pa
r
a
ti
ve
s
tr
e
ngt
h
s
of
th
e
c
ur
r
e
nt
s
tu
dy.
W
hi
le
pr
e
vi
ous
s
tu
di
e
s
h
a
ve
a
ddr
e
s
s
e
d
pl
a
y
e
r
c
la
s
s
if
ic
a
ti
on
a
nd
pe
r
f
or
m
a
nc
e
pr
e
di
c
ti
on,
th
e
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
R
e
c
om
m
e
ndat
io
n s
y
s
t
e
m
f
or
f
oot
bal
l
pl
ay
e
r
r
e
c
r
ui
tme
nt
us
in
g k
-
ne
ar
e
s
t
ne
ig
hbor
(
M
auk
a
r
)
3855
of
te
n
ove
r
lo
ok
two
c
r
it
ic
a
l
a
s
p
e
c
ts
:
e
f
f
e
c
ti
ve
di
m
e
ns
io
na
li
ty
r
e
duc
ti
on
a
nd
th
e
us
e
of
a
ppr
opr
ia
te
s
im
il
a
r
it
y
m
e
tr
ic
s
.
M
a
ny
e
xi
s
ti
ng
m
ode
l
s
s
tr
uggl
e
w
it
h
hi
gh
-
di
m
e
ns
i
ona
l
da
ta
,
w
hi
c
h
c
a
n
hi
nde
r
s
c
a
la
bi
li
ty
a
nd
a
c
c
ur
a
c
y.
I
n
c
ont
r
a
s
t,
th
is
s
tu
dy
e
m
pl
oys
P
C
A
to
r
e
ta
in
onl
y
th
e
m
os
t
r
e
le
va
nt
f
e
a
tu
r
e
s
,
pr
e
s
e
r
vi
ng
up
to
98.7%
of
th
e
or
ig
in
a
l
in
f
or
m
a
ti
on
w
hi
le
r
e
duc
in
g
m
ode
l
c
om
pl
e
xi
ty
.
A
ddi
ti
ona
ll
y,
th
e
a
dopt
io
n
of
c
os
in
e
s
im
il
a
r
it
y
pr
ovi
de
s
a
m
or
e
r
obus
t
m
e
a
s
ur
e
of
s
im
il
a
r
it
y
by
c
a
p
tu
r
in
g
th
e
di
r
e
c
ti
ona
l
pa
tt
e
r
n
of
pe
r
f
or
m
a
nc
e
,
r
a
th
e
r
th
a
n
r
e
ly
in
g
s
ol
e
ly
on
m
a
gni
tu
de
-
ba
s
e
d
di
f
f
e
r
e
nc
e
s
,
th
e
r
e
by
of
f
e
r
in
g
a
m
or
e
m
e
a
ni
ngf
ul
c
om
pa
r
is
on
be
twe
e
n pl
a
ye
r
s
w
it
h va
r
yi
ng pla
y i
nt
e
n
s
it
ie
s
but
s
im
il
a
r
s
ty
le
s
.
T
a
bl
e
3. C
om
pa
r
is
on of
c
ur
r
e
nt
s
tu
dy w
it
h pr
e
vi
ous
r
e
s
e
a
r
c
h
S
t
udy
M
e
t
hod us
e
d
F
e
a
t
ur
e
s
e
l
e
c
t
i
on
D
i
s
t
a
nc
e
m
e
t
r
i
c
M
a
i
n l
i
m
i
t
a
t
i
ons
P
e
r
f
om
a
nc
e
hi
ghl
i
ght
C
om
pa
r
i
s
on
a
dva
nt
a
ge
[
13]
R
e
gr
e
s
s
i
on
m
ode
l
f
or
t
r
a
ns
f
e
r
pr
e
di
c
t
i
on
M
a
nua
l
s
e
l
e
c
t
i
on
(
13 f
e
a
t
ur
e
s
)
N
/
A
F
oc
us
e
d onl
y on
pos
t
-
t
r
a
ns
f
e
r
pe
r
f
or
m
a
nc
e
P
r
e
di
c
t
i
ve
a
c
c
ur
a
c
y
f
or
t
r
a
ns
f
e
r
s
L
i
m
i
t
e
d t
o t
r
a
ns
f
e
r
out
c
om
e
s
, not
t
a
l
e
nt
m
a
t
c
hi
ng
[
14]
S
R
P
-
C
R
I
S
P
-
D
M
f
r
a
m
e
w
or
k
F
i
l
t
e
r
-
ba
s
e
d
N
ot
s
pe
c
i
f
i
e
d
G
e
ne
r
a
l
pr
e
di
c
t
i
on
f
r
a
m
e
w
or
k
S
t
r
uc
t
ur
e
d
a
ppr
oa
c
h
D
oe
s
not
pr
ovi
de
s
i
m
i
l
a
r
i
t
y
-
ba
s
e
d
r
e
c
om
m
e
nda
t
i
on
[
15]
M
ul
t
i
pl
e
M
L
a
l
gor
i
t
hm
s
(
S
V
M
, L
D
A
,
a
nd
KNN)
N
ot
e
m
pha
s
i
z
e
d
E
uc
l
i
de
a
n
di
s
t
a
nc
e
I
nc
ons
i
s
t
e
nt
f
e
a
t
ur
e
pr
e
pr
oc
e
s
s
i
ng
L
D
A
a
c
c
ur
a
c
y:
83.77%
N
o di
m
e
ns
i
ona
l
i
t
y
r
e
duc
t
i
on;
l
ow
e
r
i
nt
e
r
pr
e
t
a
bi
l
i
t
y
[
16]
N
M
F
+
c
l
us
t
e
r
i
ng
M
a
nua
l
, pos
i
t
i
on
-
ba
s
e
d
N
/
A
F
oc
us
e
d on C
S
L
pl
a
ye
r
s
onl
y
F
ound 18 pl
a
ye
r
t
ype
s
C
ont
e
xt
-
l
i
m
i
t
e
d
a
nd uns
ui
t
e
d f
or
s
i
m
i
l
a
r
i
t
y r
a
nki
ng
[
17]
K
N
N
(
w
i
t
h
m
ul
t
i
pl
e
di
s
t
a
nc
e
m
e
t
r
i
c
s
)
N
one
C
i
t
y
bl
oc
k
F
oc
us
e
d on
c
l
i
ni
c
a
l
E
E
G
da
t
a
C
i
t
y
bl
oc
k
pe
r
f
or
m
e
d be
s
t
I
r
r
e
l
e
va
nt
dom
a
i
n;
i
ns
i
ght
s
not
di
r
e
c
t
l
y
t
r
a
ns
f
e
r
a
bl
e
C
ur
r
e
nt
S
t
udy
P
C
A
+ K
N
N
w
i
t
h
c
os
i
ne
s
i
m
i
l
a
r
i
t
y
P
C
A
-
ba
s
e
d
di
m
e
ns
i
ona
l
i
t
y
r
e
duc
t
i
on
(
O
ut
f
i
e
l
d:
13→10,
G
K
:
4→3)
C
os
i
ne
s
i
m
i
l
a
r
i
t
y
O
nl
y l
i
ne
a
r
t
r
a
ns
f
or
m
a
t
i
on
c
ons
i
de
r
e
d
(
P
C
A
)
M
a
i
nt
a
i
ne
d ≥
98%
va
r
i
a
nc
e
, M
a
hr
e
z
91.85%
s
i
m
i
l
a
r
i
t
y
t
o M
e
s
s
i
S
upe
r
i
or
ba
l
a
nc
e
of
di
m
e
ns
i
ona
l
i
t
y
r
e
duc
t
i
on,
pe
r
f
or
m
a
nc
e
, a
nd
i
nt
e
r
pr
e
t
a
bi
l
i
t
y
4.
C
O
N
C
L
U
S
I
O
N
T
hi
s
s
tu
dy
pr
opos
e
s
a
s
tr
uc
tu
r
e
d
a
nd
s
c
a
la
bl
e
r
e
c
om
m
e
nda
ti
on mode
l
f
or
f
oot
ba
ll
pl
a
ye
r
r
e
c
r
ui
tm
e
nt
by
in
te
gr
a
ti
ng
P
C
A
f
or
di
m
e
ns
io
na
li
ty
r
e
duc
ti
on
w
it
h
K
N
N
a
nd
c
os
in
e
s
im
il
a
r
it
y
f
or
pe
r
f
o
r
m
a
nc
e
-
ba
s
e
d
c
om
pa
r
is
on.
T
he
P
C
A
e
f
f
e
c
ti
ve
ly
r
e
duc
e
s
f
e
a
tu
r
e
c
om
pl
e
xi
ty
w
hi
le
pr
e
s
e
r
vi
ng
up
to
98.7%
of
th
e
o
r
ig
in
a
l
da
ta
va
r
ia
nc
e
,
e
n
s
ur
in
g
th
a
t
e
s
s
e
nt
ia
l
pe
r
f
or
m
a
nc
e
c
ha
r
a
c
te
r
is
ti
c
s
a
r
e
r
e
ta
in
e
d.
C
os
in
e
s
im
il
a
r
it
y
f
ur
th
e
r
e
nha
nc
e
s
th
e
m
ode
l
by
c
a
pt
ur
in
g
th
e
di
r
e
c
ti
ona
l
a
li
gnm
e
nt
of
pl
a
ye
r
pe
r
f
or
m
a
nc
e
pa
tt
e
r
ns
,
e
na
bl
in
g
m
e
a
ni
ngf
ul
c
om
pa
r
is
ons
b
e
twe
e
n
in
di
vi
dua
ls
w
it
h
di
f
f
e
r
e
nt
pl
a
y
in
te
ns
it
ie
s
but
s
im
il
a
r
s
ty
le
s
.
K
e
y
pr
e
pr
oc
e
s
s
in
g s
te
ps
, s
u
c
h a
s
pe
r
-
90
-
m
in
ut
e
nor
m
a
li
z
a
ti
on, c
ons
is
te
nt
pl
a
ye
r
i
de
nt
if
ic
a
ti
on a
c
r
os
s
s
e
a
s
ons
, a
nd
pos
it
io
n
-
ba
s
e
d
c
a
te
gor
iz
a
ti
on,
s
uppor
t
f
a
ir
a
nd
r
obus
t
c
om
p
a
r
is
ons
w
it
hi
n
a
dyna
m
ic
d
a
ta
s
e
t.
T
he
m
ode
l
de
m
ons
tr
a
te
d
it
s
e
f
f
e
c
ti
ve
ne
s
s
by
id
e
nt
if
yi
ng
R
iy
a
d
M
a
hr
e
z
a
s
t
he
m
os
t
s
im
il
a
r
pl
a
ye
r
to
L
io
ne
l
M
e
s
s
i
w
it
h a
91.85%
s
im
il
a
r
it
y
s
c
or
e
,
hi
ghl
ig
ht
in
g
it
s
pr
a
c
ti
c
a
l
r
e
le
va
nc
e
f
or
s
c
out
in
g
a
nd
ta
le
nt
id
e
nt
if
ic
a
ti
on.
T
hi
s
w
or
k
c
ont
r
ib
ut
e
s
to
th
e
a
dva
nc
e
m
e
nt
of
s
por
ts
a
na
ly
ti
c
s
by
of
f
e
r
in
g
a
n
in
te
r
pr
e
ta
bl
e
,
da
ta
-
dr
iv
e
n,
a
nd
a
ppl
ic
a
ti
on
r
e
a
dy s
ol
ut
io
n f
or
m
ode
r
n f
oot
ba
ll
r
e
c
r
ui
tm
e
nt
s
tr
a
te
gi
e
s
.
A
C
K
N
O
WL
E
D
G
M
E
N
T
S
T
he
a
ut
hor
s
w
oul
d
li
ke
to
e
xpr
e
s
s
our
a
ppr
e
c
ia
ti
on
to
th
e
R
e
s
e
a
r
c
h
I
ns
ti
tu
ti
ons
of
U
ni
ve
r
s
it
a
s
G
una
da
r
m
a
f
or
t
he
i
r
c
ont
in
ue
d f
a
c
il
it
a
ti
on a
nd s
uppor
t
th
r
ough
out
t
he
r
e
s
e
a
r
c
h a
c
ti
vi
ti
e
s
.
F
U
N
D
I
N
G
I
N
F
O
R
M
A
T
I
O
N
A
ut
hor
s
s
ta
te
no f
undi
ng i
nvol
ve
d.
A
U
T
H
O
R
C
O
N
T
R
I
B
U
T
I
O
N
S
S
T
A
T
E
M
E
N
T
T
hi
s
jo
ur
na
l
us
e
s
th
e
C
ont
r
ib
ut
or
R
ol
e
s
T
a
xonomy
(
C
R
e
di
T
)
to
r
e
c
ogni
z
e
in
di
vi
dua
l
a
ut
hor
c
ont
r
ib
ut
io
ns
, r
e
duc
e
a
ut
hor
s
hi
p di
s
put
e
s
,
a
nd f
a
c
il
it
a
te
c
ol
la
bo
r
a
ti
on.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
. 14, No. 5, O
c
to
be
r
2025
:
3847
-
3857
3856
N
am
e
o
f
A
u
t
h
or
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
M
a
uka
r
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
R
odi
a
h
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
C
:
C
onc
e
pt
ua
l
i
z
a
t
i
on
M
:
M
e
t
hodol
ogy
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
da
t
i
on
Fo
:
Fo
r
m
a
l
a
na
l
ys
i
s
I
:
I
nve
s
t
i
ga
t
i
on
R
:
R
e
s
our
c
e
s
D
:
D
a
t
a
C
ur
a
t
i
on
O
:
W
r
i
t
i
ng
-
O
r
i
gi
na
l
D
r
a
f
t
E
:
W
r
i
t
i
ng
-
R
e
vi
e
w
&
E
di
t
i
ng
Vi
:
Vi
s
ua
l
i
z
a
t
i
on
Su
:
Su
pe
r
vi
s
i
on
P
:
P
r
oj
e
c
t
a
dm
i
ni
s
t
r
a
t
i
on
Fu
:
Fu
ndi
ng a
c
qui
s
i
t
i
on
C
O
N
F
L
I
C
T
O
F
I
N
T
E
R
E
S
T
S
T
A
T
E
M
E
N
T
A
ut
hor
s
s
ta
te
no c
onf
li
c
t
of
i
nt
e
r
e
s
t.
I
N
F
O
R
M
E
D
C
O
N
S
E
N
T
T
hi
s
s
tu
dy did not
i
nvol
ve
huma
n pa
r
ti
c
ip
a
nt
s
;
th
e
r
e
f
or
e
, i
nf
or
m
e
d c
ons
e
nt
w
a
s
not
r
e
qui
r
e
d.
E
T
H
I
C
A
L
A
P
P
R
O
V
A
L
T
hi
s
s
tu
dy
ut
il
iz
e
d
pe
s
t
da
ta
s
e
t
a
nd
di
d
not
in
vol
ve
hum
a
n
or
ve
r
te
br
a
te
a
ni
m
a
l
s
ubj
e
c
t
s
.
T
he
r
e
f
or
e
,
e
th
ic
a
l
a
ppr
ova
l
w
a
s
not
r
e
qui
r
e
d.
D
A
T
A
A
V
A
I
L
A
B
I
L
I
T
Y
T
hi
s
s
tu
dy
ut
il
iz
e
s
pl
a
ye
r
pe
r
f
or
m
a
nc
e
da
ta
obt
a
in
e
d
f
r
om
F
B
r
e
f
,
a
w
e
ll
-
e
s
ta
bl
is
he
d
a
nd
r
e
put
a
bl
e
f
oot
ba
ll
s
ta
ti
s
ti
c
s
pl
a
tf
or
m
th
a
t
c
om
pi
le
s
c
om
pr
e
he
ns
iv
e
a
nd
a
dva
nc
e
d
pl
a
ye
r
m
e
tr
ic
s
a
c
r
os
s
m
a
jo
r
pr
of
e
s
s
io
na
l
le
a
gue
s
.
T
he
da
ta
s
e
t
e
m
pl
oye
d i
n t
hi
s
r
e
s
e
a
r
c
h i
s
p
ubl
ic
ly
a
c
c
e
s
s
ib
le
at
ht
tp
s
:/
/f
br
e
f
.c
om
, t
he
r
e
by
a
ll
ow
in
g
f
or
pot
e
nt
ia
l
r
e
pl
ic
a
ti
on
of
th
e
da
ta
c
ol
le
c
ti
on
p
r
oc
e
s
s
by
ot
he
r
r
e
s
e
a
r
c
he
r
s
.
D
e
ta
il
e
d
de
s
c
r
ip
ti
ons
of
th
e
da
ta
pr
e
pr
oc
e
s
s
in
g
pr
oc
e
dur
e
s
,
f
e
a
tu
r
e
s
e
l
e
c
ti
on
c
r
it
e
r
ia
,
a
nd
va
r
ia
bl
e
tr
a
ns
f
or
m
a
ti
ons
a
ppl
ie
d
in
th
is
s
tu
dy a
r
e
a
va
il
a
bl
e
upon r
e
que
s
t
f
r
om
t
he
c
or
r
e
s
ponding a
ut
hor
.
R
E
F
E
R
E
N
C
E
S
[
1]
L
. L
ol
l
i
e
t
al
.
, “
D
a
t
a
a
n
a
l
yt
i
c
s
i
n t
h
e
f
oot
ba
l
l
i
ndus
t
r
y:
a
s
ur
ve
y i
nve
s
t
i
ga
t
i
ng o
pe
r
a
t
i
ona
l
f
r
a
m
e
w
or
ks
a
nd pr
a
c
t
i
c
e
s
i
n pr
of
e
s
s
i
ona
l
c
l
ubs
a
nd
na
t
i
ona
l
f
e
de
r
a
t
i
ons
f
r
om
a
r
ound
t
he
w
or
l
d,”
Sc
i
e
nc
e
and
M
e
di
c
i
ne
i
n
F
oot
bal
l
,
vol
.
9,
no.
2,
pp.
189
–
198,
2025,
doi
:
10.1080/
24733938.2024.2341837.
[
2]
Z
.
B
a
i
a
nd
X
.
B
a
i
,
“
S
por
t
s
bi
g
da
t
a
:
m
a
na
g
e
m
e
nt
,
a
na
l
y
s
i
s
,
a
ppl
i
c
a
t
i
ons
,
a
nd
c
ha
l
l
e
nge
s
,”
C
om
pl
e
x
i
t
y
,
vol
.
2021,
2021,
doi
:
10.1155/
2021/
6676297.
[
3]
N
.
C
hi
nt
ha
m
u
a
nd
M
.
K
a
r
ukur
i
,
“
D
a
t
a
s
c
i
e
nc
e
a
nd
a
ppl
i
c
a
t
i
ons
,”
J
ou
r
nal
of
D
at
a
Sc
i
e
nc
e
and
I
nt
e
l
l
i
ge
nt
Sy
s
t
e
m
s
,
vol
.
1,
no.
2
,
pp. 83
–
91, 2023, doi
:
10.47852/
bonvi
e
w
j
ds
i
s
3202837.
[
4]
W
. B
ul
l
a
nd M
. F
a
ur
e
, “
A
g
e
nt
s
i
n t
he
s
por
t
i
ng f
i
e
l
d:
a
l
a
w
a
nd e
c
onom
i
c
s
pe
r
s
p
e
c
t
i
ve
,”
I
nt
e
r
nat
i
onal
Spor
t
s
L
a
w
J
our
nal
, vol
. 22
,
no. 1, pp. 17
–
32, 2022, doi
:
10.1007/
s
40318
-
021
-
00195
-
x.
[
5]
J
.
H
.
H
e
w
i
t
t
a
nd
O
.
K
a
r
a
kuş
,
“
A
m
a
c
hi
ne
l
e
a
r
ni
ng
a
ppr
oa
c
h
f
or
pl
a
ye
r
a
nd
pos
i
t
i
on
a
dj
us
t
e
d
e
xpe
c
t
e
d
goa
l
s
i
n
f
oot
ba
l
l
(
s
oc
c
e
r
)
,”
F
r
ank
l
i
n O
pe
n
, vol
. 4, 2023, doi
:
10.1016/
j
.f
r
a
ope
.2023.100034.
[
6]
V
.
C
.
P
a
nt
z
a
l
i
s
a
nd
C
.
T
j
or
t
j
i
s
,
“
S
por
t
s
a
na
l
yt
i
c
s
f
or
f
oot
ba
l
l
l
e
a
gue
t
a
bl
e
a
nd
pl
a
ye
r
pe
r
f
or
m
a
nc
e
pr
e
di
c
t
i
on,”
i
n
2020
11t
h
I
nt
e
r
nat
i
onal
C
onf
e
r
e
n
c
e
on
I
nf
or
m
at
i
on,
I
nt
e
l
l
i
ge
nc
e
,
Sy
s
t
e
m
s
and
A
ppl
i
c
at
i
ons
,
2020,
pp.
1
–
8
,
doi
:
10.1109/
I
I
S
A
50023.2020.9284352.
[
7]
T
.
G
.
R
um
s
e
y,
“
A
s
t
a
t
i
s
t
i
c
a
l
l
ook
i
nt
o
how
c
om
m
on
s
oc
c
e
r
m
e
t
r
i
c
s
i
nf
l
ue
nc
e
e
xpe
c
t
e
d
goa
l
m
e
a
s
ur
e
s
i
n
t
he
pr
of
e
s
s
i
ona
l
ga
m
e
,”
B
.S. T
he
s
i
s
, D
e
pa
r
t
m
e
nt
of
M
a
t
he
m
a
t
i
c
a
l
S
c
i
e
nc
e
s
, B
ut
l
e
r
U
ni
ve
r
s
i
t
y, I
ndi
a
na
p
ol
i
s
, U
ni
t
e
d S
t
a
t
e
s
, 2024.
[
8]
J
.
M
e
a
d,
A
.
O
’
H
a
r
e
,
a
nd
P
.
M
c
M
e
ne
m
y,
“
E
xpe
c
t
e
d
goa
l
s
i
n
f
oot
ba
l
l
:
I
m
pr
ov
i
ng
m
ode
l
pe
r
f
or
m
a
nc
e
a
nd
de
m
ons
t
r
a
t
i
ng
va
l
ue
,”
P
L
oS O
N
E
, vol
. 18, no. 4 A
pr
i
l
, 2023, doi
:
10.1371
/
j
our
na
l
.pone
.0282295.
[
9]
M
.
R
oc
c
e
t
t
i
,
F
.
B
e
r
ve
gl
i
e
r
i
,
a
nd
G
.
C
a
ppi
e
l
l
o,
“
F
oot
ba
l
l
da
t
a
a
na
l
ys
i
s
:
t
he
p
r
e
di
c
t
i
ve
pow
e
r
of
e
xpe
c
t
e
d
goa
l
s
(
xG
)
,”
i
n
25t
h
I
nt
e
r
nat
i
onal
C
onf
e
r
e
nc
e
on I
nt
e
l
l
i
ge
nt
G
am
e
s
and Si
m
ul
at
i
on, G
A
M
E
-
O
N
202
4
, 2024, pp. 20
–
24.
[
10]
G
.
H
a
dda
d
a
nd
D
.
O
’
C
onnor
,
“
D
e
ve
l
opi
ng
pl
a
ye
r
s
f
or
a
t
hl
e
t
e
l
e
a
de
r
s
hi
p
g
r
oups
i
n
pr
of
e
s
s
i
ona
l
f
oot
ba
l
l
t
e
a
m
s
:
Q
ua
l
i
t
a
t
i
ve
i
ns
i
ght
s
f
r
om
he
a
d c
oa
c
he
s
a
nd a
t
hl
e
t
e
l
e
a
de
r
s
,”
P
L
oS O
N
E
, vol
. 17, 2022, doi
:
10.1371/
j
our
na
l
.pone
.0271093.
[
11]
S
e
c
r
e
t
a
r
y
of
S
t
a
t
e
f
or
C
ul
t
u
r
e
M
e
di
a
a
nd
S
por
t
,
“
A
s
us
t
a
i
na
bl
e
f
ut
ur
e
-
r
e
f
o
r
m
i
ng
c
l
ub
f
oot
ba
l
l
gove
r
na
nc
e
,”
U
ni
t
e
d
K
i
ngdom
G
ov
e
r
nm
e
nt
.
2023.
[
O
nl
i
ne
]
.
A
va
i
l
a
bl
e
:
ht
t
p
s
:
/
/
w
w
w
.gov.uk/
gove
r
nm
e
nt
/
publ
i
c
a
t
i
ons
/
a
-
s
us
t
a
i
na
bl
e
-
f
ut
ur
e
-
r
e
f
or
m
i
ng
-
c
l
ub
-
f
oot
ba
l
l
-
gove
r
na
nc
e
/
a
-
s
us
t
a
i
na
bl
e
-
f
ut
ur
e
-
r
e
f
o
r
m
i
ng
-
c
l
ub
-
f
oot
ba
l
l
-
gove
r
na
nc
e
[
12]
M
.
M
us
a
i
gw
a
,
“
T
he
r
ol
e
of
l
e
a
de
r
s
hi
p
i
n
m
a
na
gi
ng
c
ha
nge
,”
I
nt
e
r
nat
i
onal
R
e
v
i
e
w
of
M
anage
m
e
nt
and
M
ar
k
e
t
i
ng
,
vol
.
13,
no.
6
,
pp. 1
–
9, 2023, doi
:
10.32479/
i
r
m
m
.13526.
[
13]
D
.
D
i
ns
da
l
e
a
nd
J
.
G
a
l
l
a
ghe
r
,
“
T
r
a
ns
f
e
r
po
r
t
a
l
:
a
c
c
ur
a
t
e
l
y
f
o
r
e
c
a
s
t
i
ng
t
he
i
m
pa
c
t
of
a
pl
a
ye
r
t
r
a
ns
f
e
r
i
n
s
oc
c
e
r
,”
Sc
i
Spac
e
,
pp. 1
–
25, 2020.
[
14]
R
.
P
.
B
unk
e
r
a
nd
F
.
T
ha
bt
a
h,
“
A
m
a
c
hi
ne
l
e
a
r
ni
ng
f
r
a
m
e
w
or
k
f
or
s
por
t
r
e
s
ul
t
pr
e
di
c
t
i
on,”
A
ppl
i
e
d
C
om
put
i
ng
and
I
nf
or
m
at
i
c
s
,
vol
. 15, no. 1, pp. 27
–
33, 2019, doi
:
10.1016/
j
.a
c
i
.2017.09.005.
Evaluation Warning : The document was created with Spire.PDF for Python.