I
n
t
e
r
n
at
ion
al
Jou
r
n
a
l
of
I
n
f
o
r
m
at
ics
an
d
Com
m
u
n
icat
ion
T
e
c
h
n
ol
ogy
(
I
J
-
I
CT
)
Vo
l
.
1
5
,
N
o
.
1
,
M
a
r
c
h
20
2
6
,
pp.
1
8
9
~
1
9
7
I
S
S
N:
2252
-
8776
,
DO
I
:
10
.
11591/
i
ji
c
t
.
v
1
5
i
1
.
pp
1
8
9
-
19
7
189
Jou
r
n
al
h
o
m
e
page
:
ht
tp:
//
ij
ict
.
iaes
c
or
e
.
c
om
A
n
i
n
t
e
gr
at
io
n
c
lu
st
e
r
in
g an
d
m
u
lti
-
t
ar
ge
t
c
la
ssi
f
ic
a
t
io
n
ap
p
r
oac
h
t
o e
xp
lo
r
e
e
m
p
lo
y
ab
il
ity
an
d
c
ar
e
e
r
l
in
e
ar
it
y
Nadzl
a
And
r
it
a
I
n
t
an
Gh
ayat
r
ie
,
De
vi
F
it
r
ian
ah
D
e
pa
r
tm
e
nt
of
C
o
mput
e
r
S
c
i
e
n
c
e
,
B
I
N
U
S
G
r
a
dua
te
P
r
o
gr
a
m
-
M
a
s
te
r
of
C
o
mput
e
r
S
c
i
e
n
c
e
, B
in
a
N
us
a
nt
a
r
a
U
ni
v
e
r
s
it
y
,
J
a
ka
r
ta
, I
ndo
ne
s
ia
Ar
t
ic
l
e
I
n
f
o
AB
S
T
RA
CT
A
r
ti
c
le
h
is
tor
y
:
R
e
c
e
i
ve
d
F
e
b
23,
2025
R
e
vi
s
e
d
Oc
t
17,
2025
A
c
c
e
pt
e
d
N
o
v
5,
2025
T
h
i
s
s
t
u
d
y
a
n
al
y
z
e
s
j
o
b
p
l
aceme
n
t
w
a
i
t
i
n
g
t
i
me
s
an
d
j
o
b
l
i
n
e
ari
t
y
a
m
o
n
g
fem
al
e
s
c
i
en
ce
,
t
ec
h
n
o
l
o
g
y
,
en
g
i
n
ee
r
i
n
g
,
a
n
d
mat
h
em
at
i
c
s
(
ST
E
M
)
g
rad
u
at
e
s
u
s
i
n
g
cl
u
s
t
e
r
i
n
g
a
n
d
mu
l
t
i
-
t
arg
e
t
cl
as
s
i
fi
c
at
i
o
n
(MT
C)
mo
d
el
s
.
T
h
e
K
-
me
an
s
l
e
as
t
t
ri
mme
d
s
q
u
are
(
L
T
S
)
al
g
o
r
i
t
h
m
,
k
n
o
w
n
fo
r
i
t
s
ro
b
u
s
t
n
e
s
s
ag
ai
n
s
t
o
u
t
l
i
e
rs
,
w
as
em
p
l
o
y
ed
f
o
r
c
l
u
s
t
e
ri
n
g
.
W
i
t
h
k
=
2
an
d
a
t
ri
mm
i
n
g
p
e
r
cen
t
ag
e
o
f
3
0
%
,
t
h
e
m
o
d
e
l
a
c
h
i
e
v
e
d
a
s
i
l
h
o
u
e
t
t
e
s
c
o
r
e
o
f
7
7
%
,
re
s
u
l
t
i
n
g
i
n
t
w
o
d
i
s
t
i
n
c
t
cl
u
s
t
e
rs
:
i
d
e
a
l
a
n
d
n
o
n
-
i
d
e
a
l
.
T
o
e
n
h
an
ce
t
h
e
d
at
as
e
t
fo
r
c
l
as
s
i
fi
c
at
i
o
n
,
s
y
n
t
h
e
t
i
c
d
at
a
w
as
g
en
e
rat
ed
u
s
i
n
g
t
h
e
a
d
ap
t
i
v
e
s
y
n
t
h
e
t
i
c
(
A
D
A
SY
N
)
-
g
au
s
s
i
an
me
t
h
o
d
.
Pri
n
ci
p
al
c
o
m
p
o
n
e
n
t
an
al
y
s
i
s
(PCA
)
w
as
u
s
e
d
f
o
r
v
i
s
u
al
i
zat
i
o
n
p
u
rp
o
s
e
s
,
al
o
n
g
w
i
t
h
o
v
e
r
l
ap
p
i
n
g
h
i
s
t
o
g
ra
m
s
,
t
o
i
l
l
u
s
t
rat
e
t
h
at
t
h
e
s
y
n
t
h
e
t
i
c
d
at
a
d
i
s
t
ri
b
u
t
i
o
n
cl
o
s
e
l
y
r
e
s
em
b
l
e
d
t
h
e
o
r
i
g
i
n
a
l
.
Fo
r
cl
as
s
i
fi
c
at
i
o
n
,
a
ran
d
o
m
fo
r
e
s
t
(RF)
m
o
d
e
l
w
as
u
s
ed
t
o
p
re
d
i
c
t
b
o
t
h
j
o
b
s
w
a
i
t
i
n
g
t
i
me
a
n
d
j
o
b
l
i
n
e
ari
t
y
.
H
y
p
e
rp
arame
t
e
r
t
u
n
i
n
g
p
ro
d
u
ce
d
an
o
p
t
i
m
a
l
mo
d
e
l
w
i
t
h
a
c
l
as
s
i
fi
c
at
i
o
n
a
ccu
ra
cy
o
f
9
2
%
.
Cr
o
s
s
-
v
al
i
d
at
i
o
n
(CV
)
c
o
n
fi
r
med
t
h
e
m
o
d
e
l
’
s
r
o
b
u
s
t
n
e
s
s
,
w
i
t
h
F1
-
m
i
c
r
o
a
n
d
F1
-
m
a
c
ro
s
co
r
e
s
o
f
9
4
%
an
d
9
3
%
,
r
e
s
p
ec
t
i
v
e
l
y
.
R
e
s
u
l
t
s
s
h
o
w
t
h
at
al
t
h
o
u
g
h
w
o
me
n
i
n
ST
E
M
ar
e
u
n
d
e
rr
e
p
r
e
s
en
t
ed
,
7
3
%
o
f
t
h
e
f
em
al
e
a
l
u
mn
i
a
n
al
y
z
e
d
b
e
l
o
n
g
ed
t
o
t
h
e
s
h
o
rt
j
o
b
w
ai
t
i
n
g
g
ro
u
p
.
Fu
rt
h
e
r
m
o
r
e,
a
s
t
ro
n
g
n
eg
at
i
v
e
co
rr
el
at
i
o
n
b
e
t
w
ee
n
G
PA
an
d
j
o
b
w
a
i
t
i
n
g
t
i
me
s
u
g
g
e
s
t
s
t
h
at
h
i
g
h
e
r
-
G
PA
g
rad
u
at
e
s
t
en
d
t
o
s
ecu
r
e
em
p
l
o
y
men
t
mo
r
e
q
u
i
ck
l
y
.
K
e
y
w
o
r
d
s
:
Da
t
a
a
ugm
e
n
t
a
t
i
o
n
Ge
n
de
r
ga
p
K
-
m
e
a
n
s
l
e
a
s
t
t
r
i
m
m
e
d
s
qua
r
e
M
u
l
t
i
-
t
a
r
ge
t
r
a
n
do
m
f
o
r
e
s
t
W
o
m
e
n
i
n
s
c
i
e
n
c
e
,
t
e
c
hn
o
l
o
g
y
,
e
n
g
i
ne
e
r
i
ng,
a
n
d
m
a
t
h
e
m
a
t
i
c
s
Th
i
s
i
s
a
n
o
p
en
a
c
ces
s
a
r
t
i
c
l
e
u
n
d
e
r
t
h
e
CC
B
Y
-
SA
l
i
cen
s
e.
C
or
r
e
s
pon
din
g
A
u
th
or
:
Na
dz
l
a
An
dr
i
t
a
I
n
t
a
n
G
h
a
y
a
t
r
i
e
C
o
m
put
e
r
S
c
i
e
n
c
e
De
pa
r
t
m
e
n
t
,
B
I
NU
S
Gr
a
dua
t
e
P
r
o
gr
a
m
,
M
a
s
t
e
r
o
f
C
o
m
put
e
r
S
c
i
e
n
c
e
B
i
n
a
Nu
s
a
n
t
a
r
a
Uni
v
e
r
s
i
t
y
11480,
J
a
ka
r
t
a
,
I
n
do
n
e
s
i
a
E
m
a
i
l
:
n
a
dz
l
a
.
g
h
a
y
a
t
r
i
e
@bi
nus
.
a
c
.
i
d
1.
I
NT
RODU
C
T
I
ON
T
h
e
un
de
r
r
e
pr
e
s
e
n
t
a
t
i
o
n
o
f
wo
m
e
n
i
n
s
c
i
e
nc
e
,
t
e
c
hn
o
l
o
g
y
,
e
n
g
i
ne
e
r
i
ng,
a
n
d
m
a
t
he
m
a
t
i
c
s
(
S
T
E
M
)
r
e
m
a
i
ns
a
pr
e
s
s
i
ng
c
o
n
c
e
r
n
i
n
g
l
o
b
a
l
ge
n
de
r
d
i
s
pa
r
i
t
y
.
A
c
c
o
r
d
i
n
g
to
[
1]
,
t
h
e
l
a
b
o
r
f
o
r
c
e
pa
r
t
i
c
i
p
a
t
i
o
n
ga
p
a
m
o
n
g
i
nd
i
v
i
dua
l
s
a
g
e
d
15
a
n
d
a
b
o
v
e
r
e
a
c
h
e
s
27.
1%
,
wi
t
h
wo
m
e
n
c
o
m
pr
i
s
i
ng
46.
8%
a
n
d
m
e
n
73.
9%
.
W
o
m
e
n
a
l
s
o
h
o
l
d
o
nl
y
26.
2%
o
f
pa
r
l
i
a
m
e
n
t
a
r
y
s
e
a
t
s
c
o
m
pa
r
e
d
to
73
.
8%
f
o
r
m
e
n
,
hi
g
hli
g
h
t
i
n
g
t
he
ur
ge
n
t
n
e
e
d
t
o
a
ddr
e
s
s
t
h
e
pe
r
s
i
s
t
e
n
t
gl
o
ba
l
im
ba
l
a
n
c
e
i
n
wor
kf
o
r
c
e
pa
r
t
i
c
i
pa
t
i
o
n
.
I
n
S
T
E
M
f
i
e
l
ds
s
p
e
c
i
f
i
c
a
l
ly
[
2]
,
f
e
m
a
l
e
r
e
pr
e
s
e
n
t
a
t
i
o
n
i
nc
r
e
a
s
e
d
o
nl
y
s
li
g
h
t
l
y
-
f
r
o
m
26.
1%
i
n
2016
to
28
.
6%
i
n
2024
-
whil
e
wo
m
e
n
’
s
pa
r
t
i
c
i
pa
t
i
o
n
i
n
n
o
n
-
S
T
E
M
s
e
c
tor
s
h
a
s
n
e
a
r
e
d
50%
.
T
hi
s
c
o
n
t
r
a
s
t
un
de
r
s
c
o
r
e
s
t
h
e
i
m
po
r
t
a
n
c
e
of
f
ur
t
h
e
r
r
e
s
e
a
r
c
h
i
n
t
o
f
e
m
a
l
e
S
T
E
M
g
r
a
dua
t
e
s
’
e
m
p
l
o
ym
e
n
t
o
u
t
c
o
m
e
s
,
pa
r
t
i
c
u
l
a
r
ly
j
o
b
wa
i
t
i
n
g
t
i
m
e
a
n
d
j
o
b
l
i
ne
a
r
i
t
y
(
t
h
e
r
e
l
e
v
a
nc
e
o
f
t
h
e
i
r
wo
r
k
to
t
h
e
i
r
f
i
e
l
d
o
f
s
t
ud
y
)
.
M
a
c
hi
ne
l
e
a
r
ni
ng
(
M
L
)
[
3]
,
o
f
f
e
r
s
po
we
r
f
u
l
t
oo
l
s
f
o
r
c
l
us
t
e
r
i
n
g
a
n
d
c
l
a
s
s
if
i
c
a
t
i
o
n
,
b
o
l
s
t
e
r
e
d
by
da
t
a
a
ug
m
e
n
t
a
t
i
o
n
t
e
c
hni
que
s
to
t
a
c
kl
e
c
l
a
s
s
im
ba
l
a
nc
e
[
4]
,
[
5]
.
A
s
de
p
i
c
t
e
d
i
n
F
i
gur
e
1
,
f
e
m
a
l
e
a
l
u
m
n
i
a
r
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8776
I
n
t
J
I
n
f
&
C
o
m
m
u
n
T
e
c
hn
o
l
,
Vo
l
.
1
5
,
N
o.
1
,
M
a
r
c
h
20
2
6
:
1
8
9
-
1
9
7
190
s
i
g
nif
i
c
a
n
t
l
y
o
ut
n
u
m
b
e
r
e
d
by
m
a
l
e
s
.
Ho
we
v
e
r
,
t
h
e
r
e
i
s
h
o
pe
.
T
o
c
o
un
t
e
r
t
hi
s
i
m
ba
l
a
nc
e
,
o
v
e
r
s
a
m
p
li
ng
m
e
t
h
o
ds
s
uc
h
a
s
a
da
pt
i
v
e
s
y
n
t
h
e
t
i
c
(
A
D
A
S
YN
)
[
6]
a
n
d
S
M
OT
E
[
7
]
a
r
e
c
o
m
m
o
n
ly
e
m
p
l
o
y
e
d.
AD
A
S
Y
N
ge
n
e
r
a
t
e
s
s
y
n
t
h
e
t
i
c
da
t
a
i
n
s
pa
r
s
e
mi
n
o
r
i
t
y
r
e
g
i
o
ns
t
h
r
o
ugh
l
i
ne
a
r
i
n
t
e
r
po
l
a
t
i
o
n
,
whil
e
i
t
s
Ga
us
s
i
a
n
-
e
nh
a
nc
e
d
v
a
r
i
a
n
t
(
A
DA
S
YN
-
Ga
u
s
s
i
a
n
)
i
n
t
r
o
duc
e
s
gr
e
a
t
e
r
d
i
ve
r
s
i
t
y
a
n
d
r
e
s
i
s
t
a
n
c
e
t
o
o
u
t
l
i
e
r
s
.
T
hi
s
s
t
ud
y
ha
r
ne
s
s
e
s
t
h
e
pot
e
n
t
i
a
l
o
f
A
D
A
S
Y
N
-
Ga
us
s
i
a
n
t
o
b
a
l
a
n
c
e
t
h
e
m
in
o
r
i
t
y
(
F
e
m
a
l
e
a
l
u
m
n
i
)
c
l
a
s
s
,
o
f
f
e
r
i
ng
a
pr
o
m
i
s
i
ng
pa
t
h
to
a
ddr
e
s
s
ge
n
de
r
d
i
s
p
a
r
i
t
i
e
s
i
n
S
T
E
M
.
F
i
gur
e
1.
C
o
m
pa
r
i
s
o
n
o
f
w
o
m
e
n
i
n
S
T
E
M
a
n
d
n
on
-
S
T
E
M
d
i
e
l
d
s
B
e
y
o
n
d
c
l
us
t
e
r
i
n
g
j
o
b
wa
i
t
i
ng
t
i
m
e
s
,
t
hi
s
r
e
s
e
a
r
c
h
a
l
s
o
e
x
a
m
i
ne
s
j
o
b
l
i
ne
a
r
i
t
y
,
whi
c
h
r
e
f
e
r
s
to
t
h
e
de
gr
e
e
o
f
pr
e
d
i
c
t
a
bil
i
t
y
i
n
j
o
b
wa
i
t
i
ng
t
i
m
e
s
,
us
i
ng
b
ot
h
un
s
up
e
r
vi
s
e
d
a
n
d
s
upe
r
vi
s
e
d
a
ppr
o
a
c
h
e
s
.
C
l
us
t
e
r
i
n
g
i
s
u
s
e
d
to
gr
o
up
j
o
b
wa
i
t
i
ng
dur
a
t
i
o
ns
,
whil
e
c
las
s
if
i
c
a
t
i
o
n
pr
e
d
i
c
t
s
b
o
t
h
j
o
b
s
wa
i
t
i
ng
c
a
t
e
go
r
y
a
n
d
j
o
b
li
ne
a
r
i
t
y
[
8]
.
P
r
i
o
r
s
t
udi
e
s
ha
v
e
e
x
p
l
o
r
e
d
s
im
il
a
r
t
h
e
m
e
s
.
F
o
r
i
ns
t
a
n
c
e
,
[
9]
a
pp
l
i
e
d
f
uz
z
y
c
-
m
e
a
n
s
(
F
C
M
)
to
c
l
u
s
t
e
r
j
o
b
wa
i
t
i
n
g
t
i
m
e
s
i
n
t
o
“
f
a
s
t
,
”
“
m
o
de
r
a
t
e
,
”
a
n
d
“
s
l
o
w”
c
a
t
e
go
r
i
e
s
,
f
o
l
l
o
we
d
by
C
4.
5
de
c
i
s
i
o
n
t
r
e
e
c
l
a
s
s
if
i
c
a
t
i
o
n
,
a
c
hi
e
vi
ng
o
nly
86%
a
c
c
ur
a
c
y
w
i
t
h
o
ut
k
-
v
a
l
ue
o
pt
i
m
i
z
a
t
i
o
n
.
An
o
t
h
e
r
s
t
udy
[
10]
e
m
p
l
o
y
e
d
f
uz
z
y
c
l
us
t
e
r
i
n
g
to
i
de
n
t
i
f
y
e
m
p
l
o
ym
e
n
t
f
a
c
t
o
r
s
,
e
m
p
h
a
s
i
z
i
ng
j
o
b
s
t
a
bil
i
t
y
(
40.
7%
)
a
n
d
e
m
p
l
o
ym
e
n
t
r
a
t
e
(
34.
4%
)
a
s
ke
y
i
nf
l
u
e
n
c
e
s
.
A
pr
e
m
il
i
na
r
y
v
e
r
s
i
o
n
o
f
t
hi
s
r
e
s
e
a
r
c
h
[
11]
,
pr
o
p
o
s
e
d
a
c
o
m
bi
ne
d
c
l
us
t
e
r
i
ng
a
n
d
m
u
l
t
i
-
t
a
r
ge
t
c
l
a
s
s
if
i
c
a
t
i
o
n
(
M
T
C
)
f
r
a
m
e
wo
r
k,
a
c
hi
e
vi
ng
77%
a
c
c
ur
a
c
y
a
n
d
a
s
i
l
h
o
ue
t
t
e
s
c
o
r
e
o
f
0.
61.
B
u
i
l
d
i
ng
o
n
t
h
a
t
,
t
h
e
pr
e
s
e
n
t
r
e
s
e
a
r
c
h
e
nh
a
n
c
e
s
pe
r
f
o
r
m
a
n
c
e
t
h
r
o
ugh
o
p
t
i
m
i
z
e
d
m
o
de
l
s
e
l
e
c
t
i
o
n
,
us
i
n
g
K
-
m
e
a
ns
L
T
S
f
o
r
i
t
s
r
o
b
us
t
n
e
s
s
t
o
o
u
t
l
i
e
r
s
-
v
a
li
da
t
e
d
a
c
r
o
s
s
t
e
n
b
e
n
c
hm
a
r
k
da
t
a
s
e
t
s
[
12]
.
T
hi
s
s
t
udy
pr
e
d
i
c
t
s
t
wo
t
a
r
ge
t
v
a
r
i
a
bl
e
s
s
i
m
u
l
t
a
n
e
o
us
ly
,
r
e
qu
i
r
i
n
g
a
M
T
C
a
ppr
o
a
c
h
[
13]
,
[
14
]
.
I
n
[
15]
,
P
r
i
o
r
r
e
s
e
a
r
c
h
s
uppo
r
t
s
r
a
n
do
m
f
o
r
e
s
t
(
R
F
)
e
f
f
e
c
t
i
v
e
n
e
s
s
i
n
pr
e
d
i
c
t
i
n
g
gr
a
dua
t
e
e
m
p
l
o
y
a
bil
i
t
y
.
S
i
m
il
a
r
ly
,
[
16]
us
e
d
ge
n
e
t
i
c
a
l
go
r
i
t
hm
s
(
G
A
s
)
to
pr
e
d
i
c
t
c
a
r
e
e
r
s
uc
c
e
s
s
(
87.
61%
f
i
t
)
a
n
d
[
17]
a
pp
l
ied
Na
ï
v
e
B
a
y
e
s
w
i
t
h
90%
a
c
c
ur
a
c
y
o
n
li
mi
t
e
d
da
t
a
.
C
o
m
p
a
r
a
t
i
v
e
e
v
a
l
ua
t
i
o
n
s
[
18]
,
h
a
v
e
f
ur
t
h
e
r
s
h
o
wn
RF
s
upe
r
i
o
r
pe
r
f
o
r
m
a
n
c
e
,
r
o
b
us
t
n
e
s
s
to
n
o
i
s
e
,
a
n
d
r
e
s
i
s
t
a
nc
e
to
o
v
e
r
f
i
t
t
i
n
g,
m
a
k
i
ng
i
t
a
r
e
l
i
a
bl
e
c
h
o
i
c
e
f
o
r
t
h
i
s
s
t
udy
’
s
o
u
t
l
i
e
r
-
h
e
a
vy
da
t
a
s
e
t
.
T
hi
s
pa
pe
r
pr
e
s
e
n
t
s
a
c
o
m
pr
e
he
n
s
i
ve
i
n
t
e
gr
a
t
e
d
m
o
de
l
i
ng
f
r
a
m
e
wo
r
k
t
h
a
t
c
o
m
bi
ne
s
A
D
A
S
YN
-
Ga
us
s
i
a
n
a
ug
m
e
n
t
a
t
i
o
n
,
K
-
m
e
a
n
s
L
T
S
c
l
u
s
t
e
r
i
ng,
a
n
d
m
u
l
t
i
-
t
a
r
ge
t
r
a
n
do
m
f
o
r
e
s
t
(
M
T
R
F
)
c
l
a
s
s
if
i
c
a
t
i
o
n
.
Unli
ke
e
a
r
l
i
e
r
wo
r
ks
t
h
a
t
f
o
c
us
e
d
o
n
s
i
n
g
l
e
-
t
a
r
ge
t
pr
e
di
c
t
i
o
n
o
r
m
o
de
l
s
l
a
c
k
i
ng
o
u
t
l
i
e
r
r
e
s
i
li
e
n
c
e
,
thi
s
s
t
ud
y
j
o
i
n
t
ly
m
o
de
l
s
j
o
b
wa
i
t
i
ng
t
i
m
e
a
n
d
j
o
b
li
ne
a
r
i
t
y
i
n
a
unif
i
e
d
p
i
pe
li
ne
.
T
o
o
ur
kn
o
wl
e
dge
,
n
o
pr
e
vi
o
us
r
e
s
e
a
r
c
h
h
a
s
e
x
a
mi
n
e
d
t
h
e
s
e
dua
l
e
m
p
l
o
ym
e
n
t
o
u
t
c
o
m
e
s
f
o
r
f
e
m
a
l
e
S
T
E
M
gr
a
dua
t
e
s
us
i
n
g
s
uc
h
a
n
i
n
t
e
gr
a
t
e
d
a
ppr
o
a
c
h
.
2.
RE
S
E
AR
CH
M
E
T
HO
D
T
hi
s
s
t
udy
c
o
n
s
i
s
t
s
o
f
t
h
r
e
e
m
a
i
n
s
t
a
ge
s
:
d
a
t
a
p
r
e
pa
r
a
t
i
o
n
,
K
-
m
e
a
ns
L
T
S
c
l
u
s
t
e
r
i
n
g,
a
n
d
t
h
e
M
T
R
F
c
l
a
s
s
if
i
c
a
t
i
o
n
pr
o
c
e
s
s
,
a
s
i
ll
us
t
r
a
t
e
d
i
n
F
i
gur
e
2
.
I
n
t
h
e
da
t
a
p
r
e
pa
r
a
t
i
o
n
ph
a
s
e
,
t
h
e
pr
e
-
pr
o
c
e
s
s
i
n
g
s
t
e
p
e
n
s
ur
e
s
t
h
e
da
t
a
s
e
t
i
s
f
r
e
e
f
r
o
m
m
i
s
s
i
ng
v
a
l
ue
s
a
n
d
dup
li
c
a
t
e
r
e
c
o
r
ds
.
On
c
e
t
h
e
da
t
a
i
s
c
l
e
a
n
,
da
t
a
a
ug
m
e
n
t
a
t
i
o
n
i
s
c
o
n
duc
t
e
d
to
a
ddr
e
s
s
t
h
e
s
i
g
nif
i
c
a
n
t
i
m
ba
l
a
nc
e
be
t
we
e
n
m
a
l
e
a
n
d
f
e
m
a
l
e
a
l
u
m
n
i
r
e
c
o
r
ds
.
T
hi
s
s
t
ud
y
a
pp
l
i
e
s
t
h
e
A
D
A
S
YN
-
g
a
u
s
s
i
a
n
t
e
c
hni
que
t
o
ge
n
e
r
a
t
e
s
y
n
t
h
e
t
i
c
da
t
a
f
o
r
t
h
e
m
i
n
o
r
i
t
y
c
l
a
s
s
(
f
e
m
a
l
e
a
l
u
mni
)
.
Af
t
e
r
a
ug
m
e
n
t
a
t
i
o
n
,
t
h
e
da
t
a
s
e
l
e
c
t
i
o
n
pr
o
c
e
s
s
i
s
pe
r
f
o
r
m
e
d
to
i
s
o
l
a
t
e
o
n
l
y
f
e
m
a
l
e
a
l
u
m
n
i
r
e
c
o
r
ds
f
o
r
f
ur
t
h
e
r
m
o
de
li
ng.
Ne
x
t
,
t
h
e
K
-
m
e
a
n
s
L
T
S
c
l
us
t
e
r
i
n
g
pr
o
c
e
s
s
i
s
a
pp
li
e
d
to
g
r
o
up
t
h
e
j
o
b
wa
i
t
i
ng
t
i
m
e
s
o
f
t
h
e
s
e
l
e
c
t
e
d
a
l
u
m
n
i
.
T
hi
s
i
nv
o
l
ve
s
hy
pe
r
pa
r
a
m
e
t
e
r
t
uni
n
g
t
o
d
e
t
e
r
m
i
ne
t
h
e
o
p
t
i
m
a
l
n
u
m
be
r
o
f
c
l
us
t
e
r
s
(
K
)
a
n
d
t
h
e
b
e
s
t
t
r
i
m
mi
ng
pe
r
c
e
n
t
a
ge
.
T
h
e
f
i
na
l
s
t
a
ge
i
s
t
h
e
c
l
a
s
s
if
ica
t
i
o
n
p
r
o
c
e
s
s
,
wh
e
r
e
t
h
e
gr
o
upe
d
da
t
a
i
s
us
e
d
a
s
in
put
f
o
r
a
M
T
R
F
m
o
de
l
t
o
pr
e
di
c
t
b
ot
h
j
o
b
s
wa
i
t
i
ng
t
i
m
e
a
n
d
j
o
b
li
ne
a
r
i
t
y
s
im
u
l
t
a
n
e
o
us
ly
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t
J
I
n
f
&
C
o
m
m
u
n
T
e
c
hn
o
l
I
S
S
N:
2252
-
8776
A
n
int
e
gr
ati
on
c
lus
ter
ing
and
multi
-
tar
ge
t
c
las
s
if
ic
ati
on
appr
oac
h
…
(
N
adz
la
A
ndr
it
a
I
ntan
Ghay
atr
i
e
)
191
F
i
gur
e
2.
R
e
s
e
a
r
c
h
f
r
a
m
e
wo
r
k
2.
1.
Dat
a
c
ol
l
e
c
t
ion
T
h
e
da
t
a
us
e
d
i
n
t
hi
s
s
t
ud
y
wa
s
s
o
ur
c
e
d
f
r
o
m
s
e
ve
r
a
l
pr
i
v
a
t
e
uni
ve
r
s
i
t
i
e
s
i
n
I
n
do
ne
s
i
a
.
T
wo
t
y
pe
s
o
f
da
t
a
s
e
t
s
we
r
e
c
o
l
l
e
c
t
e
d:
a
l
u
m
n
i
a
c
a
de
mi
c
r
e
c
o
r
ds
a
n
d
po
s
t
-
gr
a
dua
t
i
o
n
e
m
p
l
o
y
m
e
n
t
da
t
a
.
T
h
e
da
t
a
c
o
l
l
e
c
t
i
o
n
pr
o
c
e
s
s
i
nv
o
l
v
e
d
s
e
n
d
i
ng
f
o
r
m
a
l
r
e
que
s
t
s
to
e
a
c
h
uni
v
e
r
s
i
t
y
,
a
c
c
o
m
p
a
ni
e
d
by
a
l
e
t
t
e
r
o
f
a
gr
e
e
m
e
nt
s
t
a
t
i
ng
t
h
a
t
t
h
e
da
t
a
w
o
ul
d
b
e
us
e
d
s
t
r
i
c
t
l
y
f
o
r
r
e
s
e
a
r
c
h
pur
po
s
e
s
,
w
o
ul
d
r
e
m
a
i
n
c
o
nf
i
de
n
t
i
a
l
,
a
n
d
wo
ul
d
not
r
e
v
e
a
l
t
h
e
i
de
n
t
i
t
y
o
f
a
ny
pa
r
t
i
c
i
pa
t
i
n
g
i
ns
t
i
t
ut
i
o
n
s
.
Upo
n
a
ppr
o
v
a
l
a
n
d
t
h
e
s
i
g
ni
ng
o
f
a
m
e
m
o
r
a
n
du
m
o
f
un
de
r
s
t
a
n
d
i
ng
(
M
o
U)
,
t
h
e
uni
ve
r
s
i
t
i
e
s
pr
o
vi
de
d
t
h
e
r
e
que
s
t
e
d
da
t
a
s
e
t
s
,
whi
c
h
we
r
e
t
h
e
n
pr
o
c
e
s
s
e
d
f
o
r
r
e
s
e
a
r
c
h
a
n
a
ly
s
i
s
.
B
a
s
e
d
o
n
t
h
e
c
o
l
l
e
c
t
e
d
da
t
a
,
i
t
wa
s
f
o
un
d
t
h
a
t
f
e
m
a
l
e
a
l
u
m
n
i
m
a
de
up
o
nly
15%
o
f
t
h
e
tot
a
l
r
e
c
o
r
ds
,
c
o
nf
i
r
m
i
ng
t
h
e
c
l
a
s
s
i
m
ba
l
a
n
c
e
a
ddr
e
s
s
e
d
t
h
r
o
ugh
a
ugm
e
n
t
a
t
i
o
n
.
2.
2.
Dat
a
p
r
e
p
ar
at
ion
p
r
oc
e
s
s
2.
2.
1.
Dat
a
p
r
e
-
p
r
oc
e
s
s
in
g
Da
t
a
pr
e
-
p
r
o
c
e
s
s
i
ng
wa
s
c
o
n
duc
t
e
d
to
e
n
s
ur
e
da
t
a
qua
l
i
t
y
a
n
d
e
l
im
i
na
t
e
a
n
o
m
a
li
e
s
[
19]
.
T
hi
s
s
t
a
ge
i
nv
o
l
ve
d
s
e
v
e
r
a
l
s
t
e
ps
,
i
n
c
l
ud
i
ng
c
h
e
c
k
i
ng
f
o
r
n
u
ll
va
l
ue
s
,
r
e
m
o
vi
ng
dup
l
i
c
a
t
e
r
e
c
or
ds
,
a
n
d
i
d
e
n
t
i
f
yi
ng
o
u
t
l
i
e
r
s
,
a
l
l
o
f
w
hi
c
h
a
r
e
e
s
s
e
n
t
i
a
l
t
o
a
v
o
i
d
s
u
b
o
pti
m
a
l
m
o
de
li
ng
o
u
t
c
o
m
e
s
.
Ou
t
l
i
e
r
de
t
e
c
t
i
o
n
wa
s
pe
r
f
o
r
m
e
d
us
i
n
g
t
h
e
i
n
t
e
r
qua
r
t
i
l
e
r
a
n
g
e
(
I
QR
)
m
e
t
h
o
d
a
n
d
vi
s
u
a
l
i
z
e
d
t
h
r
o
ugh
b
o
x
p
l
o
t
s
ge
n
e
r
a
t
e
d
wi
t
h
t
h
e
m
a
t
p
l
o
t
l
ib
li
b
r
a
r
y
[
20]
.
2.
2.
2.
Dat
a
au
gm
e
n
t
at
ion
A
D
A
S
YN
i
s
a
da
t
a
a
ug
m
e
n
t
a
t
i
o
n
t
e
c
hni
que
t
h
a
t
ge
n
e
r
a
t
e
s
s
y
n
t
h
e
t
i
c
s
a
m
p
l
e
s
t
h
r
o
ugh
l
i
ne
a
r
i
n
t
e
r
po
l
a
t
i
o
n
b
e
t
we
e
n
m
i
n
o
r
i
t
y
c
l
a
s
s
i
n
s
t
a
nc
e
s
a
n
d
t
h
e
i
r
n
e
a
r
e
s
t
n
e
i
g
hb
o
r
s
[
6
]
.
H
o
we
v
e
r
,
t
hi
s
l
i
ne
a
r
a
ppr
o
a
c
h
m
a
y
be
i
n
a
de
qu
a
t
e
wh
e
n
t
h
e
m
i
n
o
r
i
t
y
c
l
a
s
s
e
xhi
b
i
t
s
a
c
o
m
p
l
e
x
da
t
a
di
s
t
r
i
b
ut
i
o
n
[
21]
,
p
ot
e
n
t
i
a
ll
y
l
e
a
d
i
n
g
to
s
y
n
t
he
t
i
c
da
t
a
t
h
a
t
l
a
c
ks
d
i
ve
r
s
i
t
y
o
r
f
a
il
s
t
o
r
e
f
lec
t
t
h
e
un
de
r
lyi
ng
d
i
s
t
r
i
b
ut
i
o
n
a
c
c
ur
a
t
e
l
y
.
T
o
a
d
dr
e
s
s
t
hi
s
li
mi
t
a
t
i
o
n
,
A
D
A
S
YN
-
Ga
u
s
s
i
a
n
,
a
m
o
d
i
f
i
e
d
v
e
r
s
io
n
o
f
A
D
A
S
YN
,
i
n
c
o
r
por
a
t
e
s
t
h
e
Ga
us
s
i
a
n
d
i
s
t
r
i
b
ut
i
o
n
to
ge
n
e
r
a
t
e
s
y
n
t
he
t
i
c
da
t
a
.
T
hi
s
s
t
ud
y
a
do
p
t
s
t
h
e
m
u
l
t
i
v
a
r
i
a
t
e
Ga
us
s
i
a
n
d
i
s
t
r
i
b
ut
i
o
n
,
s
u
i
t
a
bl
e
f
o
r
t
w
o
or
m
o
r
e
da
t
a
di
m
e
n
s
i
o
ns
,
a
s
de
f
i
ne
d
i
n
(
1)
.
(
;
,
Σ
)
=
1
(
2
)
2
|
Σ
|
1
2
e
xp
(
−
1
2
(
−
)
Σ
−
1
(
−
)
(
1)
T
h
e
pa
r
a
m
e
t
e
r
s
k
a
n
d
β
a
r
e
c
r
i
t
i
c
a
l
i
n
t
h
e
a
ug
m
e
n
t
a
t
i
o
n
t
e
c
h
ni
que
.
T
h
e
pa
r
a
m
e
t
e
r
k
de
t
e
r
m
i
n
e
s
h
o
w
f
a
r
a
n
d
s
im
il
a
r
t
h
e
ge
n
e
r
a
t
e
d
s
y
n
t
h
e
t
i
c
s
a
m
p
les
a
r
e
.
I
n
c
or
r
e
c
t
k
v
a
l
ue
s
c
a
n
l
e
a
d
to
o
v
e
r
f
it
t
i
n
g,
a
s
a
ug
m
e
n
t
a
t
i
o
n
c
a
n
b
e
a
r
e
gu
l
a
r
i
z
e
r
c
o
n
t
r
o
l
li
ng
m
o
de
l
c
o
m
p
l
e
xi
t
y
[
22]
.
T
h
e
pa
r
a
m
e
t
e
r
β
a
dds
v
a
r
i
a
t
i
o
n
t
h
r
o
ugh
c
o
n
t
r
o
l
l
e
d
n
o
i
s
e
w
hil
e
pr
e
s
e
r
vi
ng
t
h
e
o
ve
r
a
l
l
r
e
l
e
v
a
n
c
e
o
f
t
h
e
da
t
a
[
23]
.
S
e
l
e
c
t
i
n
g
t
h
e
a
ppr
o
pr
i
a
t
e
β
va
l
ue
e
ns
ur
e
s
t
h
a
t
t
h
e
a
ug
m
e
n
t
e
d
da
t
a
i
s
r
e
l
e
v
a
n
t
a
nd
d
o
e
s
n
o
t
de
vi
a
t
e
s
i
g
nif
i
c
a
n
t
l
y
f
r
o
m
t
h
e
or
i
g
in
a
l
da
t
a
.
A
D
A
S
YN
-
g
a
u
s
s
i
a
n
a
ut
o
m
a
t
i
c
a
ll
y
t
un
e
s
t
h
e
s
e
pa
r
a
m
e
t
e
r
s
,
m
a
k
i
ng
a
pp
l
yi
ng
t
hi
s
t
e
c
h
ni
que
a
n
d
ge
n
e
r
a
t
i
n
g
s
y
n
t
he
t
i
c
da
t
a
s
t
r
a
i
g
h
t
f
o
r
wa
r
d.
2.
2.
3.
Dat
a
s
e
l
e
c
t
ion
T
h
e
da
t
a
s
e
l
e
c
t
i
o
n
pr
o
c
e
s
s
wa
s
c
o
n
duc
t
e
d
to
i
s
o
l
a
t
e
a
n
d
r
e
t
a
i
n
o
nl
y
t
h
e
r
e
c
o
r
ds
o
f
f
e
m
a
l
e
a
l
u
m
n
i
a
f
t
e
r
t
h
e
a
ug
m
e
n
t
a
t
i
o
n
s
t
e
p.
Gi
v
e
n
t
h
a
t
f
e
m
a
l
e
a
l
u
m
n
i
c
o
n
s
t
i
t
ut
e
d
o
nl
y
15%
o
f
t
h
e
i
ni
t
i
a
l
da
t
a
s
e
t
,
da
t
a
a
ug
m
e
n
t
a
t
i
o
n
wa
s
e
s
s
e
n
t
i
a
l
t
o
a
ddr
e
s
s
t
hi
s
s
igni
f
i
c
a
n
t
i
m
ba
l
a
nc
e
pr
i
o
r
to
m
o
de
l
i
ng.
F
o
l
l
o
w
i
n
g
t
h
e
a
ug
m
e
n
t
a
t
i
o
n
pr
o
c
e
s
s
,
whi
c
h
c
o
n
s
i
de
r
e
d
t
h
e
or
i
g
i
na
l
d
i
s
t
r
i
b
ut
i
o
n
o
f
f
e
m
a
l
e
a
l
u
m
n
i
d
a
t
a
,
o
nl
y
t
h
e
r
e
l
e
va
n
t
f
e
m
a
l
e
a
l
u
m
ni
r
e
c
o
r
ds
we
r
e
f
il
t
e
r
e
d
a
n
d
s
e
l
e
c
t
e
d
f
o
r
us
e
i
n
t
h
e
s
u
b
s
e
qu
e
n
t
m
o
de
l
i
ng
s
t
a
ge
s
.
2.
3.
K
-
m
e
an
s
LTS
K
-
m
e
a
n
s
l
e
a
s
t
t
r
i
m
m
e
d
s
qua
r
e
(
L
T
S
)
[
12]
i
s
a
m
o
d
i
f
i
c
a
t
i
o
n
o
f
t
h
e
K
-
m
e
a
n
s
a
l
go
r
i
t
hm
t
h
a
t
t
r
i
m
s
o
u
t
l
i
e
r
s
a
f
t
e
r
c
l
u
s
t
e
r
s
a
r
e
f
o
r
m
e
d.
C
l
u
s
t
e
r
i
n
g
i
s
e
m
p
l
o
y
e
d
to
g
r
o
up
a
l
u
m
n
i
b
a
s
e
d
o
n
s
i
mi
l
a
r
i
t
i
e
s
i
n
t
h
e
i
r
c
h
a
r
a
c
t
e
r
i
s
t
i
c
w
i
t
hi
n
t
h
e
da
t
a
s
e
t
.
T
hi
s
a
ppr
o
a
c
h
he
l
p
s
un
c
o
v
e
r
t
h
e
n
a
t
ur
a
l
s
t
r
uc
t
ur
e
o
f
t
h
e
da
t
a
a
n
d
pr
o
vi
de
s
i
n
i
t
i
a
l
i
ns
i
g
h
t
s
i
n
t
o
t
h
e
n
u
m
b
e
r
a
n
d
na
t
ur
e
o
f
t
h
e
r
e
s
u
l
t
i
n
g
gr
o
ups
.
T
h
e
s
e
c
l
u
s
t
e
r
s
s
e
r
v
e
a
s
t
h
e
b
a
s
i
s
f
o
r
f
ur
t
h
e
r
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8776
I
n
t
J
I
n
f
&
C
o
m
m
u
n
T
e
c
hn
o
l
,
Vo
l
.
1
5
,
N
o.
1
,
M
a
r
c
h
20
2
6
:
1
8
9
-
1
9
7
192
a
n
a
ly
s
i
s
,
whi
c
h
i
s
t
h
e
n
f
o
l
l
o
we
d
by
c
l
a
s
s
i
f
i
c
a
t
i
o
n
s
t
a
ge
s
.
C
l
u
s
t
e
r
i
n
g
i
s
pe
r
f
o
r
m
e
d
pr
i
o
r
to
c
l
a
s
s
if
i
c
a
t
i
o
n
to
b
e
t
t
e
r
un
de
r
s
t
a
n
d
t
h
e
di
s
t
r
i
b
ut
i
o
n
a
n
d
s
t
r
uc
t
ur
e
o
f
t
h
e
da
t
a
,
e
s
pe
c
i
a
ll
y
w
h
e
n
l
a
b
e
l
s
s
uc
h
a
s
e
mp
l
o
y
m
e
n
t
pa
tt
e
r
n
s
or
b
e
h
a
vi
o
r
a
r
e
n
ot
e
x
p
l
i
c
i
t
y
a
va
i
l
a
b
l
e
.
I
n
t
hi
s
m
o
de
l
,
t
h
e
o
p
t
i
m
a
l
n
u
m
be
r
o
f
c
l
us
t
e
r
s
(
)
a
nd
t
h
e
t
r
i
m
mi
ng
pr
o
p
o
r
t
i
o
n
a
r
e
de
t
e
r
m
i
ne
d
t
h
r
o
ugh
t
uni
ng
f
u
n
c
t
i
o
ns
pr
o
vi
de
d
w
i
t
hi
n
t
h
e
a
l
go
r
i
t
hm
’
s
c
l
a
s
s
.
Al
t
o
ugh
c
e
n
t
r
o
i
d
i
ni
t
i
a
li
z
a
t
i
o
n
o
f
K
-
m
e
a
n
s
c
a
n
i
nf
l
ue
n
c
e
t
h
e
f
i
na
l
c
l
us
t
e
r
i
n
g
due
to
i
t
s
s
us
c
e
pt
i
bil
i
t
y
to
l
o
c
a
l
o
pt
i
m
a
[
24]
,
t
hi
s
i
s
s
u
e
i
s
mi
t
i
g
a
t
e
d
i
n
K
-
m
e
a
ns
L
T
S
.
I
n
t
hi
s
m
e
t
h
o
d,
c
l
u
s
t
e
r
i
n
g
i
s
p
e
r
f
o
r
m
e
d
us
i
ng
m
u
l
t
i
p
l
e
r
a
n
do
m
i
ni
t
i
a
li
z
a
t
i
o
ns
,
a
n
d
t
h
e
b
e
s
t
r
e
s
u
l
t
i
s
s
e
l
e
c
t
e
d
b
a
s
e
d
o
n
t
h
e
l
o
we
s
t
tot
a
l
w
i
t
hin
-
c
l
us
t
e
r
v
a
r
i
a
n
c
e
.
F
ur
t
h
e
r
m
o
r
e
,
t
h
e
tr
i
mm
i
ng
pr
o
c
e
s
s
r
e
m
o
v
e
s
a
s
pe
c
i
f
i
e
d
pr
o
p
o
r
t
i
o
n
o
f
da
t
a
p
o
i
n
t
s
w
i
t
h
t
h
e
hi
g
h
e
s
t
r
e
s
i
dua
l
s
a
f
t
e
r
c
l
us
t
e
r
i
n
g,
whi
c
h
he
l
ps
r
e
duc
e
t
h
e
i
nf
l
ue
n
c
e
o
f
poo
r
l
y
i
n
i
t
i
a
li
z
e
d
c
e
n
t
r
o
i
d
s
a
n
d
o
ut
l
i
e
r
s
.
T
h
e
s
e
t
w
o
m
e
c
h
a
ni
s
m
s
(
r
e
pe
a
t
e
d
i
ni
t
i
a
li
z
a
t
i
o
n
a
n
d
t
r
i
mm
i
ng)
c
o
n
t
r
i
b
ut
e
to
a
m
o
r
e
r
o
b
us
t
a
n
d
s
t
a
bl
e
c
l
u
s
t
e
r
i
n
g
o
u
t
c
o
m
e
.
2.
3.
1.
F
in
d
K
op
t
im
al
an
d
b
e
s
t
p
e
r
c
e
n
t
age
T
hi
s
pr
o
c
e
s
s
i
s
a
pa
r
a
m
e
t
e
r
t
uni
n
g
s
t
e
p
r
e
qu
i
r
e
d
f
o
r
t
h
e
m
o
d
i
f
i
e
d
K
-
m
e
a
n
s
m
o
de
l
.
Us
i
ng
t
h
e
t
r
i
m
mi
ng
c
o
n
c
e
pt
a
s
i
n
[
25]
,
K
-
m
e
a
ns
L
T
S
r
e
q
u
i
r
e
s
a
n
o
p
t
i
m
a
l
o
ut
l
i
e
r
t
r
i
mm
i
ng
pe
r
c
e
n
t
a
ge
.
T
uni
ng
i
s
pe
r
f
o
r
m
e
d
by
a
s
s
e
s
s
i
ng
t
h
e
s
il
h
o
ue
t
t
e
s
c
o
r
e
,
a
n
d
t
h
e
pa
r
a
m
e
t
e
r
c
o
m
bi
na
t
i
o
n
t
h
a
t
y
i
e
l
ds
t
h
e
b
e
s
t
s
i
l
h
o
ue
t
t
e
s
c
o
r
e
i
s
s
e
l
e
c
t
e
d
f
o
r
m
o
de
l
i
ng.
2.
3.
2.
Cl
u
s
t
e
r
in
g
p
r
oc
e
s
s
K
-
m
e
a
n
s
L
T
S
l
o
o
ks
l
i
ke
r
o
b
us
t
t
r
i
mm
e
d
K
-
m
e
a
ns
(
R
T
K
M
)
[
25]
ut
i
l
i
z
i
ng
t
h
e
c
o
n
c
e
pt
o
f
t
r
i
mm
i
ng
o
u
t
l
i
e
r
s
b
a
s
e
d
o
n
L
T
S
.
I
n
K
-
m
e
a
n
s
L
T
S
[
12]
,
t
h
e
pr
o
c
e
s
s
i
nv
o
l
ve
s
s
o
r
t
i
n
g
t
h
e
c
l
us
t
e
r
e
d
da
t
a
s
e
t
by
i
t
s
d
i
s
t
a
nc
e
to
t
h
e
c
e
n
t
r
o
i
d
a
n
d
t
r
i
mm
i
ng
t
h
e
f
a
r
t
h
e
s
t
da
t
a
p
o
i
n
t
s
(
o
u
t
l
i
e
r
s
)
b
a
s
e
d
o
n
a
s
pe
c
i
f
i
c
pe
r
c
e
n
t
a
ge
.
Da
ta
f
a
l
li
ng
w
i
t
hi
n
t
h
e
f
a
r
t
h
e
s
t
n
%
f
r
o
m
t
h
e
c
e
n
t
r
o
i
d
a
r
e
t
r
i
m
m
e
d.
T
hi
s
a
l
go
r
i
t
hm
m
a
n
a
g
e
s
c
o
m
p
l
e
xi
t
y
we
ll
,
wi
t
h
m
a
in
s
t
e
ps
s
uc
h
a
s
c
e
n
t
r
o
i
d
i
n
i
t
i
a
l
i
z
a
t
i
o
n
a
n
d
c
l
u
s
t
e
r
I
D
a
s
s
i
g
nm
e
n
t
(
(
×
)
)
,
c
a
l
c
u
l
a
t
i
n
g
da
t
a
di
s
t
a
n
c
e
s
to
t
h
e
c
e
n
t
r
o
i
d
(
(
×
×
)
)
,
s
or
t
i
n
g
t
h
e
d
i
s
t
a
nc
e
s
f
o
r
o
u
t
l
i
e
r
t
r
i
mm
i
ng
(
(
∗
log
(
)
)
)
,
a
n
d
r
e
a
s
s
i
g
ni
ng
c
l
u
s
t
e
r
s
(
(
×
)
)
.
Gi
ve
n
t
hi
s
c
o
m
p
l
e
xi
t
y
,
K
-
m
e
a
ns
L
T
S
i
s
s
u
i
t
a
bl
e
f
o
r
m
e
d
i
u
m
-
s
i
z
e
d
da
t
a
s
e
t
s
,
b
a
l
a
nc
i
ng
e
f
f
i
c
i
e
nc
y
a
n
d
c
o
m
put
a
t
i
o
n
.
2.
3.
3.
Cl
u
s
t
e
r
in
g
e
va
l
u
at
ion
C
l
u
s
t
e
r
i
n
g
r
e
s
u
l
t
s
a
r
e
e
v
a
l
ua
t
e
d
us
i
n
g
t
h
e
s
il
h
o
ue
t
te
s
c
o
r
e
,
s
ui
t
a
bl
e
f
o
r
da
t
a
s
e
t
s
l
a
c
k
i
ng
a
t
r
a
i
ni
ng
s
e
t
f
o
r
m
o
de
l
e
v
a
l
u
a
t
i
o
n
[
26]
.
T
h
e
s
i
l
h
o
ue
t
t
e
s
c
o
r
e
m
e
a
s
ur
e
s
t
h
e
c
l
us
t
e
r
i
n
g
qua
l
i
t
y
by
c
o
m
pa
r
i
ng
a
n
e
l
e
m
e
n
t
’
s
a
v
e
r
a
ge
d
i
s
t
a
n
c
e
to
m
e
m
be
r
s
o
f
i
t
s
c
l
us
t
e
r
(
)
a
n
d
t
h
e
n
e
a
r
e
s
t
ot
h
e
r
c
l
us
t
e
r
(
)
.
T
h
e
s
c
o
r
e
r
a
n
ge
s
f
r
o
m
-
1
to
1
,
wi
t
h
hi
g
h
po
s
i
t
i
v
e
v
a
l
ue
s
i
n
d
i
c
a
t
i
n
g
w
e
l
l
-
c
l
us
t
e
r
e
d
e
l
e
m
e
n
t
s
,
whil
e
n
e
g
a
t
i
ve
v
a
l
ue
s
s
ugge
s
t
e
l
e
m
e
n
t
s
mi
g
h
t
b
e
i
n
c
o
r
r
e
c
t
l
y
c
l
us
t
e
r
e
d
[
27]
.
A
m
a
xim
a
l
s
il
h
o
ue
t
t
e
s
c
o
r
e
(
(
)
=
1
)
i
s
a
c
hi
e
v
e
d
w
h
e
n
e
l
e
m
e
n
t
s
a
r
e
c
l
o
s
e
to
t
h
e
i
r
c
l
us
t
e
r
a
n
d
f
a
r
f
r
o
m
o
t
h
e
r
s
.
2.
4.
M
T
RF
2.
4.
1.
Hyp
e
r
p
ar
a
m
e
t
e
r
t
u
n
in
g
H
y
pe
r
pa
r
a
m
e
t
e
r
t
uni
n
g
i
s
c
r
uc
i
a
l
i
n
M
L
m
o
de
li
ng
,
e
s
pe
c
i
a
ll
y
f
o
r
t
r
e
e
-
b
a
s
e
d
m
o
de
l
s
w
i
t
h
n
u
m
e
r
o
us
pa
r
a
m
e
t
e
r
s
[
28]
.
I
t
r
e
f
e
r
s
to
b
u
i
l
d
i
ng
a
n
o
pt
i
m
a
l
m
o
de
l
by
c
o
nf
i
gur
i
n
g
hy
pe
r
pa
r
a
m
e
t
e
r
s
t
h
r
o
ugh
a
s
e
a
r
c
h
s
t
r
a
t
e
g
y
[
29
]
.
T
hi
s
s
t
ud
y
us
e
s
Gr
i
d
s
e
a
r
c
h
f
o
r
hy
pe
r
pa
r
a
m
e
t
e
r
t
uni
n
g
i
n
t
h
e
M
T
R
F
m
o
de
l
.
Gr
i
d
s
e
a
r
c
h
i
s
a
de
c
i
s
i
o
n
-
t
h
e
o
r
e
t
i
c
a
l
a
ppr
o
a
c
h
i
nv
o
l
vi
ng
e
xh
a
u
s
t
i
v
e
s
e
a
r
c
h
o
v
e
r
a
s
e
t
r
a
n
ge
o
f
hy
p
e
r
pa
r
a
m
e
t
e
r
v
a
l
ue
s
[
30]
.
Gr
i
d
s
e
a
r
c
h
u
s
e
s
c
r
o
s
s
-
va
l
i
d
a
t
i
o
n
(
C
V)
f
o
r
e
a
c
h
c
o
m
b
i
na
t
i
o
n
o
f
pa
r
a
m
e
t
e
r
s
.
T
h
e
da
t
a
i
s
s
p
li
t
i
n
t
o
10
pa
r
t
s
,
‘
n_s
pli
ts
’
a
n
d
f
o
r
e
a
c
h
pa
r
t
,
o
n
e
s
e
c
t
i
o
n
i
s
u
s
e
d
a
s
t
e
s
t
da
t
a
,
whi
l
e
t
h
e
r
e
s
t
i
s
t
r
a
i
ni
ng
da
t
a
.
Af
ter
t
e
s
t
i
n
g
a
l
l
pa
r
a
m
e
t
e
r
c
o
m
bi
na
t
i
o
n
s
,
Gr
i
d
s
e
a
r
c
h
s
e
l
e
c
t
s
t
h
e
b
e
s
t
c
o
m
bi
na
t
i
o
n
b
a
s
e
d
o
n
t
h
e
e
v
a
l
ua
t
i
o
n
s
c
o
r
e
‘
s
c
o
r
i
n
g
’
.
T
hi
s
s
t
udy
us
e
s
F
1
-
mi
c
r
o
a
n
d
F1
-
m
a
c
r
o
s
c
o
r
e
s
a
s
t
h
e
e
v
a
l
u
a
t
i
o
n
m
e
t
r
i
c
s
f
o
r
m
u
l
t
i
-
t
a
r
ge
t
m
o
de
l
i
ng
[
31]
.
T
hi
s
s
t
udy
us
e
s
f
o
ur
t
h
a
t
a
r
e
s
i
g
ni
f
i
c
a
n
t
f
o
r
m
o
de
l
pe
r
f
o
r
m
a
n
c
e
.
‘
n_e
s
ti
mator
s
’
r
e
f
e
r
s
to
t
h
e
n
u
m
b
e
r
o
f
tr
e
e
s
,
a
i
mi
ng
to
b
o
o
s
t
m
o
de
l
pe
r
f
o
r
m
a
n
c
e
by
r
e
duc
i
ng
va
r
i
a
n
c
e
.
‘
max
_de
pth
’
c
o
n
tr
o
l
s
t
h
e
m
a
xim
u
m
de
pt
h
o
f
e
a
c
h
de
c
i
s
i
o
n
t
r
e
e
.
De
e
pe
r
t
r
e
e
s
a
r
e
m
o
r
e
pr
o
n
e
to
o
v
e
r
f
i
t
t
i
n
g
[
32]
,
s
o
t
hi
s
s
t
ud
y
l
im
i
t
s
t
h
e
de
pt
h
to
10,
20,
a
n
d
30.
‘
min_s
a
mple
_s
pli
t
’
s
e
t
s
t
h
e
m
i
n
i
m
u
m
n
u
m
be
r
o
f
s
a
m
p
l
e
s
r
e
qu
i
r
e
d
to
s
pl
i
t
a
n
o
de
,
a
n
d
‘
min_s
ample
_leaf
’
[
33]
c
o
n
t
r
o
l
s
t
h
e
mi
n
im
u
m
n
u
m
be
r
o
f
s
a
m
p
l
e
s
r
e
qu
i
r
e
d
a
t
e
a
c
h
l
e
a
f
n
o
de
.
T
h
e
v
a
l
u
e
s
[
1
]
,
[
2
]
,
[
4]
b
a
l
a
n
c
e
c
a
pt
ur
i
n
g
gr
a
n
u
l
a
r
pa
tt
e
r
n
s
a
n
d
pr
e
v
e
n
t
i
n
g
o
v
e
r
f
i
t
t
i
n
g.
2.
4.
2.
M
T
C
p
r
oc
e
s
s
T
h
e
c
l
a
s
s
i
f
i
c
a
t
i
o
n
m
o
de
l
i
s
a
c
r
uc
i
a
l
s
t
e
p
i
n
t
hi
s
s
tud
y
,
wh
e
r
e
ML
a
l
go
r
i
t
hm
s
a
r
e
a
pp
l
i
e
d
to
p
r
e
di
c
t
a
n
d
c
a
t
e
go
r
i
z
e
a
l
u
m
da
t
a
b
a
s
e
d
o
n
t
h
e
i
r
wa
i
t
i
n
g
ti
m
e
f
o
r
a
j
o
b
a
n
d
t
h
e
l
i
ne
a
r
i
t
y
o
f
t
h
e
i
r
c
a
r
e
e
r
.
T
h
i
s
s
t
ud
y
us
e
s
t
h
e
M
T
R
F
s
i
nc
e
t
w
o
t
a
r
ge
t
v
a
r
i
a
b
l
e
s
n
e
e
d
t
o
b
e
c
l
a
s
s
i
f
i
e
d.
A
c
c
o
r
d
i
n
g
to
[
34]
,
M
T
C
w
or
ks
a
s
f
o
l
l
o
ws
,
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t
J
I
n
f
&
C
o
m
m
u
n
T
e
c
hn
o
l
I
S
S
N:
2252
-
8776
A
n
int
e
gr
ati
on
c
lus
ter
ing
and
multi
-
tar
ge
t
c
las
s
if
ic
ati
on
appr
oac
h
…
(
N
adz
la
A
ndr
it
a
I
ntan
Ghay
atr
i
e
)
193
T
h
e
t
r
a
i
ni
ng
da
t
a
s
e
t
c
o
n
s
i
s
t
s
o
f
t
r
i
p
l
e
t
s
(
,
,
)
,
wh
e
r
e
∈
de
s
c
r
i
be
s
t
h
e
r
e
l
a
t
i
o
ns
hi
p
b
e
t
we
e
n
i
ns
t
a
nc
e
a
n
d
t
a
r
ge
t
.
Dur
i
n
g
t
r
a
i
ni
ng,
di
f
f
e
r
e
n
t
i
n
s
t
a
n
c
e
s
a
n
d
di
f
f
e
r
e
n
t
t
a
r
ge
t
s
,
wh
e
r
e
n
a
n
d
m
a
r
e
f
i
n
i
t
e
n
u
m
be
r
s
.
T
h
e
s
c
o
r
e
s
f
r
o
m
t
h
e
tr
a
i
ni
ng
da
t
a
a
r
e
a
r
r
a
n
ge
d
i
n
a
n
×
m
a
t
r
i
x
,
o
f
t
e
n
i
nc
o
m
p
l
e
t
e
due
to
m
i
s
s
i
n
g
v
a
l
ue
s
T
h
e
s
c
o
r
e
s
e
t
i
s
o
n
e
-
d
i
m
e
ns
i
o
n
a
l
a
n
d
c
o
n
s
i
s
t
s
o
f
no
m
i
na
l
,
o
r
di
na
l
,
o
r
f
un
da
m
e
n
t
a
l
v
a
l
ue
s
.
T
hi
s
pr
o
c
e
s
s
a
i
m
s
t
o
pr
e
di
c
t
s
c
o
r
e
s
f
o
r
e
a
c
h
i
n
s
t
a
nc
e
-
t
a
r
ge
t
pa
i
r
(
,
)
i
n
t
h
e
s
e
t
×
.
2.
4.
3.
Cl
as
s
if
i
c
at
ion
e
val
u
at
ion
E
v
a
l
ua
t
i
n
g
m
u
l
t
i
-
t
a
r
ge
t
m
o
de
l
s
c
a
nn
ot
r
e
l
y
o
n
s
t
a
n
da
r
d
c
l
a
s
s
if
i
c
a
t
i
o
n
m
e
t
r
i
c
s
li
ke
a
c
c
ur
a
c
y
,
pr
e
c
i
s
i
o
n
,
r
e
c
a
ll
,
o
r
F
1
-
s
c
o
r
e
.
I
n
M
T
C
,
t
h
e
s
e
m
e
t
r
i
c
s
m
u
s
t
b
e
s
li
g
h
t
l
y
m
o
d
if
i
e
d,
r
e
s
u
l
t
i
n
g
i
n
mi
c
r
o
a
v
e
r
a
g
i
ng
[
35]
a
n
d
m
a
c
r
o
a
v
e
r
a
g
i
n
g
[
36]
.
A
s
e
x
p
l
a
i
ne
d
[
11]
,
m
i
c
r
o
-
a
v
e
r
a
g
i
n
g
s
u
m
s
t
h
e
v
a
l
ue
s
i
n
t
h
e
c
o
nf
us
i
o
n
m
a
t
r
i
x
to
pr
o
duc
e
a
c
c
ur
a
c
y
,
pr
e
c
i
s
i
o
n
,
r
e
c
a
ll
,
a
n
d
F1
-
s
c
o
r
e
.
M
a
c
r
o
a
v
e
r
a
g
i
n
g
t
a
ke
s
t
h
e
a
v
e
r
a
ge
o
f
e
a
c
h
v
a
lue
i
n
t
h
e
c
o
nf
us
i
o
n
m
a
t
r
i
x
.
M
i
c
r
o
a
v
e
r
a
g
i
ng
i
s
us
e
f
u
l
w
h
e
n
o
v
e
r
a
l
l
pe
r
f
o
r
m
a
n
c
e
i
s
pr
i
o
r
i
t
i
z
e
d,
w
hi
l
e
m
a
c
r
o
a
v
e
r
a
g
i
ng
i
s
pr
e
f
e
r
r
e
d
wh
e
n
e
a
c
h
t
a
r
ge
t
v
a
r
i
a
bl
e
i
s
e
qua
l
ly
im
po
r
t
a
n
t
.
2.
5.
K
-
f
ol
d
c
r
os
s
val
id
at
ion
CV
i
s
a
n
im
po
r
t
a
n
t
ML
a
n
d
s
t
a
t
i
s
t
i
c
a
l
m
o
de
li
ng
m
e
t
h
o
d
t
h
a
t
s
e
l
e
c
t
s
m
o
de
l
s
,
a
s
s
e
s
s
e
s
pe
r
f
o
r
m
a
n
c
e
,
a
n
d
t
un
e
s
hy
p
e
r
pa
r
a
m
e
t
e
r
s
[
37]
.
T
hi
s
m
e
t
h
o
d
i
nv
o
l
ve
s
s
p
li
t
t
i
n
g
t
h
e
da
t
a
i
n
t
o
tr
a
i
ni
ng
a
n
d
t
e
s
t
i
n
g
s
e
t
s
,
tr
a
i
ni
n
g
t
h
e
m
o
de
l
o
n
t
h
e
t
r
a
i
ni
ng
s
e
t
,
a
n
d
e
va
l
u
a
t
i
n
g
i
t
s
p
e
r
f
o
r
m
a
n
c
e
w
i
t
h
t
h
e
t
e
s
t
i
n
g
s
e
t
.
T
hi
s
s
t
ud
y
u
s
e
s
10
-
f
o
l
d
CV
,
whi
c
h
m
e
a
ns
t
h
e
r
e
a
r
e
10
-
f
o
l
d
s
w
i
t
h
d
if
f
e
r
e
n
t
t
r
a
i
ni
ng
a
n
d
t
e
s
t
i
n
g
s
e
t
s
i
n
e
a
c
h
f
o
l
d.
T
he
s
a
m
p
l
e
i
nd
e
x
a
ll
o
w
s
da
t
a
i
n
de
x
t
o
b
e
pa
r
t
o
f
t
h
e
t
r
a
i
n
s
e
t
i
n
f
o
l
d
1
a
n
d
p
a
r
t
o
f
t
h
e
t
e
s
t
s
e
t
i
n
f
o
l
d
2.
I
n
t
hi
s
pr
o
c
e
s
s
,
t
h
e
m
e
t
r
i
c
s
us
e
d
to
a
s
s
e
s
s
t
h
e
pe
r
f
o
r
m
a
n
c
e
o
f
e
a
c
h
f
o
l
d
a
r
e
F
1
-
m
i
c
r
o
a
n
d
F
1
-
m
a
c
r
o
s
c
or
e
s
.
3.
RE
S
UL
T
S
AN
D
DI
S
CU
S
S
I
ON
3.
1.
Dat
a
a
u
gm
e
n
t
at
ion
As
m
e
n
t
i
o
n
e
d
i
n
s
e
c
t
i
o
n
3,
A
DA
S
YN
-
ga
u
s
s
i
a
n
us
e
s
ga
us
s
i
a
n
d
i
s
t
r
i
b
ut
i
o
n
t
o
ge
n
e
r
a
t
e
s
y
n
t
he
t
i
c
da
t
a
.
T
h
e
pa
r
a
m
e
t
e
r
s
k
a
n
d
b
e
t
a
us
e
d
i
n
t
hi
s
s
t
a
ge
a
r
e
k
=
4
a
n
d
β
=
0.
5,
wi
t
h
t
h
e
m
o
s
t
o
p
t
i
m
a
l
F1
-
s
c
o
r
e
b
e
i
ng
0.
8405.
Unl
i
ke
S
M
OT
E
[
38]
,
A
DA
S
YN
-
g
a
us
s
i
a
n
f
o
c
us
e
s
o
n
ge
n
e
r
a
t
i
n
g
s
y
n
t
h
e
t
i
c
da
t
a
b
a
s
e
d
o
n
i
t
s
de
n
s
i
t
y
.
T
h
e
r
e
f
o
r
e
,
t
h
e
n
u
m
b
e
r
o
f
m
i
n
o
r
i
t
y
s
y
n
t
h
e
t
i
c
da
t
a
i
s
n
ot
e
qua
l
to
t
h
e
m
a
j
o
r
i
t
y
da
t
a
.
I
n
t
hi
s
s
t
ud
y
,
b
a
l
a
n
c
i
ng
t
h
e
da
t
a
b
e
t
we
e
n
m
a
l
e
s
a
n
d
f
e
m
a
l
e
s
i
s
n
o
t
a
s
i
m
p
o
r
t
a
n
t
b
e
c
a
us
e
o
nl
y
f
e
m
a
l
e
da
t
a
i
s
us
e
d
f
o
r
m
o
de
l
i
ng,
a
s
m
e
n
t
i
o
n
e
d
i
n
s
e
c
t
i
o
n
3
.
F
i
gur
e
3
s
h
o
ws
t
h
e
o
v
e
r
l
a
pp
i
ng
hi
s
t
o
gr
a
m
vi
s
ua
li
z
a
t
i
o
n
f
o
r
t
h
e
f
e
a
t
ur
e
s
‘
gpa
’
a
n
d
‘
w
ait
ing_t
ime
’
.
I
n
t
h
e
vi
s
ua
li
z
a
t
i
o
n
o
f
t
h
e
‘
gpa
’
f
e
a
t
ur
e
,
t
h
e
s
y
n
t
h
e
t
i
c
da
t
a
ge
ne
r
a
t
e
d
h
a
s
a
hi
g
h
e
r
de
ns
i
t
y
n
e
a
r
a
GPA
o
f
4.
0.
B
ot
h
di
s
t
r
i
b
ut
i
o
ns
h
a
v
e
s
im
il
a
r
s
ha
pe
s
,
s
h
o
w
i
n
g
t
h
a
t
t
h
e
s
y
n
t
h
e
t
i
c
da
t
a
f
o
l
l
o
w
s
t
h
e
o
r
i
g
i
na
l
d
i
s
t
r
i
b
ut
i
o
n
.
F
o
r
t
h
e
‘
w
ait
ing_t
ime
’
f
e
a
t
ur
e
,
m
o
s
t
of
t
h
e
or
i
g
i
na
l
d
a
t
a
i
s
c
o
n
c
e
n
t
r
a
t
e
d
a
r
o
un
d
a
wa
i
t
i
n
g
t
i
m
e
o
f
0
–
5
m
o
n
t
hs
.
B
ot
h
di
s
t
r
i
b
ut
i
o
n
s
(
o
r
i
g
i
na
l
a
n
d
s
y
n
t
h
e
t
i
c
da
t
a
)
t
a
pe
r
o
f
f
s
im
il
a
r
ly
a
s
t
h
e
wa
i
t
i
n
g
t
i
m
e
i
n
c
r
e
a
s
e
s
b
e
y
o
n
d
20
m
o
n
t
h
s
.
F
i
gur
e
3
.
Ov
e
r
l
a
pp
i
ng
hi
s
t
o
gr
a
m
f
o
r
‘
gpa
’
f
e
a
t
ur
e
(
l
e
f
t
)
a
n
d
‘
wa
i
t
i
ng_t
i
m
e
’
f
e
a
t
ur
e
(
r
i
g
h
t
)
3.
2.
Cl
u
s
t
e
r
in
g
p
r
oc
e
s
s
On
e
o
f
t
h
e
b
e
n
e
f
i
t
s
o
f
us
i
ng
K
-
m
e
a
n
s
L
T
S
i
n
c
l
u
s
t
e
r
i
n
g
m
o
de
li
ng
i
s
t
h
a
t
K
-
m
e
a
n
s
L
T
S
i
s
a
r
o
b
us
t
m
o
de
l
a
g
a
i
ns
t
o
u
t
l
i
e
r
s
.
T
h
e
pr
e
s
e
n
c
e
o
f
o
u
t
l
i
e
r
s
i
n
t
h
e
da
t
a
i
s
s
u
bs
t
a
n
t
i
a
l
,
e
s
pe
c
i
a
ll
y
i
n
t
h
e
‘
GP
A
,
’
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8776
I
n
t
J
I
n
f
&
C
o
m
m
u
n
T
e
c
hn
o
l
,
Vo
l
.
1
5
,
N
o.
1
,
M
a
r
c
h
20
2
6
:
1
8
9
-
1
9
7
194
‘
length_s
tudy
,
’
a
n
d
‘
w
ait
ing_t
ime
’
f
e
a
t
ur
e
s
.
Us
i
n
g
k
=
2
a
n
d
a
tr
i
mm
i
ng
pe
r
c
e
n
t
a
ge
o
f
30%
,
a
s
de
s
c
r
i
b
e
d
i
n
s
e
c
t
i
o
n
3,
t
h
e
n
u
m
be
r
o
f
o
ut
l
i
e
r
s
i
s
s
i
g
ni
f
i
c
a
n
t
l
y
r
e
duc
e
d.
Ho
we
v
e
r
,
a
c
o
n
s
i
de
r
a
bl
e
n
u
m
be
r
o
f
o
ut
l
i
e
r
s
r
e
m
a
in
i
n
t
h
e
‘
length_s
tudy
’
whi
c
h
i
s
e
x
pe
c
t
e
d,
a
s
t
h
e
a
v
e
r
a
ge
s
t
ud
y
dur
a
t
i
o
n
f
o
r
f
e
m
a
l
e
a
l
u
m
ni
i
n
t
hi
s
da
t
a
s
e
t
i
s
4
y
e
a
r
s
,
a
s
t
a
n
da
r
d
t
i
m
e
f
r
a
m
e
f
o
r
un
de
r
gr
a
dua
t
e
s
tud
i
e
s
i
n
I
n
do
n
e
s
i
a
n
uni
ve
r
s
i
t
i
e
s
.
F
i
gur
e
4
di
s
p
l
a
y
s
t
h
e
vi
s
u
a
l
i
z
a
t
i
o
n
o
f
t
h
e
pl
o
t
r
e
s
u
l
t
s
f
o
r
e
a
c
h
f
o
r
m
e
d
c
l
u
s
t
e
r
.
I
n
t
h
e
2D
pl
o
t,
c
l
us
t
e
r
s
0
a
n
d
1
a
r
e
c
l
e
a
r
ly
we
ll
-
s
e
pa
r
a
t
e
d.
S
i
mi
l
a
r
l
y
,
t
h
e
3D
pl
o
t
s
h
o
ws
t
w
o
di
s
t
i
nc
t
c
l
us
t
e
r
s
po
s
i
t
i
o
n
e
d
c
l
o
s
e
ly
b
ut
s
t
i
ll
s
e
pa
r
a
bl
e
.
W
i
t
h
=
2
a
n
d
a
30%
t
r
i
mm
i
ng
r
a
t
e
,
t
h
e
K
-
m
e
a
n
s
L
T
S
m
o
de
l
a
c
hi
e
ve
s
a
s
il
h
o
ue
t
t
e
s
c
o
r
e
o
f
0.
77,
i
n
d
i
c
a
t
i
n
g
s
t
r
o
n
g
c
l
us
t
e
r
i
n
g
pe
r
f
o
r
m
a
n
c
e
.
Al
t
h
o
ugh
f
ur
t
h
e
r
o
p
t
i
m
i
z
a
t
i
o
n
i
s
p
o
s
s
i
bl
e
,
t
h
i
s
s
c
o
r
e
de
m
o
ns
t
r
a
t
e
s
s
u
f
f
i
c
i
e
n
t
s
e
pa
r
a
t
i
o
n
b
e
t
we
e
n
t
h
e
two
c
l
us
t
e
r
s
.
T
h
e
c
l
u
s
t
e
r
s
f
o
r
m
e
d
c
a
n
b
e
i
n
t
e
r
pr
e
t
e
d
a
s
f
o
l
l
o
ws
,
C
l
u
s
t
e
r
0
(
544
m
e
m
be
r
s
)
:
a
gr
o
up
o
f
f
e
m
a
l
e
a
l
u
m
n
i
w
i
t
h
l
o
n
g
j
o
b
wa
i
t
i
ng
t
i
m
e
s
,
l
o
we
r
GPA
s
t
h
a
n
C
l
u
s
t
e
r
1
(
a
v
e
r
a
ge
GPA
3.
24)
,
a
n
d
a
s
t
udy
dur
a
t
i
o
n
s
li
g
h
t
l
y
l
o
n
ge
r
a
t
a
n
a
v
e
r
a
ge
o
f
4.
13
y
e
a
r
s
.
C
l
u
s
t
e
r
1
(
1
,
483
m
e
m
be
r
s
)
:
a
gr
o
up
o
f
f
e
m
a
l
e
a
l
u
m
n
i
w
i
t
h
s
h
o
r
t
j
o
b
wa
i
t
i
n
g
t
i
m
e
s
,
hi
g
h
e
r
GP
As
(
a
v
e
r
a
ge
GP
A
3.
6)
,
a
n
d
a
s
t
udy
dur
a
t
i
o
n
o
f
4.
11
y
e
a
r
s
.
F
o
r
c
o
m
pa
r
i
s
o
n
,
a
ppl
yi
ng
s
t
a
n
da
r
d
K
-
m
e
a
n
s
c
l
u
s
t
e
r
i
n
g
w
i
t
h
t
h
e
s
a
m
e
n
u
m
be
r
o
f
c
l
u
s
t
e
r
s
(
=
2
)
r
e
s
u
l
t
s
i
n
a
l
o
we
r
s
i
l
h
o
ue
t
t
e
s
c
o
r
e
o
f
0.
63,
i
n
d
i
c
a
t
i
n
g
l
e
s
s
d
i
s
t
i
n
c
t
s
e
pa
r
a
t
i
o
n
.
T
hi
s
c
o
m
pa
r
i
s
o
n
e
m
p
ha
s
i
z
e
s
t
h
e
e
f
f
e
c
t
i
v
e
n
e
s
s
o
f
K
-
m
e
a
n
s
L
T
S
i
n
ha
n
d
l
i
ng
o
ut
l
i
e
r
s
a
n
d
i
m
pr
o
vi
ng
c
l
us
t
e
r
i
n
g
qua
l
i
t
y
w
i
t
hi
n
t
hi
s
da
t
a
s
e
t.
F
i
gur
e
4
.
Vi
s
ua
l
i
z
a
t
i
o
n
o
f
t
h
e
f
o
r
m
e
d
c
l
us
t
e
r
s
p
l
o
t
,
(
l
e
f
t
)
p
l
o
t
wi
t
h
2
d
i
m
e
ns
i
o
ns
,
a
n
d
(
r
i
g
h
t
)
pl
o
t
wi
t
h
3
di
m
e
ns
i
o
n
s
3.
3.
M
T
C
p
r
oc
e
s
s
T
h
e
b
e
s
t
m
o
de
l
a
r
c
hi
t
e
c
t
ur
e
f
r
o
m
t
h
e
hy
p
e
r
pa
r
a
m
e
t
e
r
t
uni
n
g
pe
r
f
o
r
m
e
d
‘
max
_de
pth
’
=
20,
‘
n_e
s
ti
mator
s
’
=
300,
‘
min_s
ample
s
_leaf
’
=
1
,
a
n
d
‘
min_s
ample
_s
pli
t
’
=
2.
F
r
o
m
t
hi
s
a
r
c
hi
t
e
c
t
ur
e
,
t
h
e
b
e
s
t
s
c
o
r
e
a
c
hi
e
v
e
d
i
s
92%
or
0
.
9188.
T
a
bl
e
1
s
h
o
ws
t
h
e
r
e
s
u
l
t
s
o
f
e
a
c
h
m
e
t
r
i
c
us
e
d,
a
n
d
t
h
e
m
i
c
r
o
a
v
e
r
a
ge
pr
o
vi
de
s
a
hi
g
h
e
r
r
e
s
u
l
t
t
h
a
n
t
h
e
m
a
c
r
o
a
v
e
r
a
ge
.
T
a
bl
e
1.
M
T
R
F
m
o
de
l
li
ng
r
e
s
u
l
t
s
o
f
hy
p
e
r
pa
r
a
m
e
t
e
r
t
uni
n
g
M
e
tr
i
c
s
M
ic
r
o
a
v
e
r
a
ge
M
a
c
r
o
a
ve
r
a
g
e
P
r
e
c
is
i
o
n
0.930612
0.912371
R
e
c
a
ll
0.913828
0.894089
F1
-
s
c
o
r
e
0.922144
0.903203
T
h
e
a
b
o
ve
t
a
bl
e
c
o
m
bi
ne
s
t
h
e
r
e
s
u
l
t
s
f
o
r
b
oth
t
a
r
ge
t
v
a
r
i
a
bl
e
s
,
l
i
ne
a
r
i
t
y
a
n
d
wa
i
t
i
ng
t
i
m
e
.
F
r
o
m
c
a
l
c
u
l
a
t
i
o
n
o
f
c
o
n
f
u
s
i
o
n
m
a
t
r
i
x
us
i
n
g
t
h
e
a
c
c
ur
a
c
y
f
o
r
m
u
l
a
(
(
+
)
/
(
+
=
+
+
)
)
)
,
t
h
e
a
c
c
ur
a
c
y
f
o
r
t
h
e
‘
li
ne
ar
it
y
’
t
a
r
ge
t
f
e
a
t
ur
e
i
s
81
%
o
r
0.
809
,
a
n
d
f
o
r
t
h
e
‘
w
ait
ing_t
ime
’
f
e
a
t
ur
e
,
i
t
i
s
100%
o
r
1.
0
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t
J
I
n
f
&
C
o
m
m
u
n
T
e
c
hn
o
l
I
S
S
N:
2252
-
8776
A
n
int
e
gr
ati
on
c
lus
ter
ing
and
multi
-
tar
ge
t
c
las
s
if
ic
ati
on
appr
oac
h
…
(
N
adz
la
A
ndr
it
a
I
ntan
Ghay
atr
i
e
)
195
3.
4.
10
-
f
ol
d
c
r
os
s
val
id
at
ion
F
r
o
m
t
h
e
pr
e
vi
o
u
s
hy
pe
r
pa
r
a
m
e
t
e
r
t
uni
n
g
r
e
s
u
l
t
s
,
t
h
e
10
-
f
o
l
d
v
a
li
da
t
i
o
n
pr
o
c
e
s
s
us
e
d
t
h
e
be
s
t_es
ti
mator
_
a
n
d
wa
s
s
a
ve
d
a
s
t
h
e
b
e
s
t
M
T
R
F
m
o
de
l
f
o
r
t
hi
s
s
t
ud
y
.
T
h
e
da
t
a
i
s
us
e
d
f
o
r
CV
e
v
a
l
ua
t
i
o
n
a
f
t
e
r
a
ug
m
e
n
t
a
t
i
o
n
a
n
d
c
l
us
t
e
r
i
n
g
pr
o
c
e
s
s
e
s
.
He
n
c
e
,
t
h
e
r
e
qu
i
r
e
d
f
e
a
t
ur
e
s
a
r
e
a
l
r
e
a
d
y
a
v
a
il
a
bl
e
.
T
h
e
da
t
a
wi
l
l
b
e
s
p
l
i
t
i
n
t
o
10
-
f
o
l
d
s
w
i
t
h
e
qua
l
pr
o
por
t
i
o
n
s
o
f
t
r
a
i
n
a
n
d
t
e
s
t
i
n
e
a
c
h
f
o
l
d.
T
a
bl
e
2
s
h
o
ws
t
h
e
e
va
l
ua
t
i
o
n
r
e
s
u
l
t
s
o
f
F1
-
s
c
o
r
e
s
,
b
o
t
h
m
i
c
r
o
a
n
d
m
a
c
r
o
,
f
o
r
e
a
c
h
f
o
l
d
ge
n
e
r
a
t
e
d
dur
i
n
g
t
h
e
CV
pr
o
c
e
s
s
.
T
h
e
b
e
s
t
v
a
l
u
e
s
we
r
e
a
c
hi
e
v
e
d
i
n
t
h
e
s
e
v
e
n
t
h
f
o
l
d,
w
i
t
h
a
m
i
c
r
o
s
c
o
r
e
o
f
0.
939
a
n
d
a
m
a
c
r
o
s
c
or
e
o
f
0.
929
.
T
h
e
c
on
s
i
s
t
e
n
c
y
o
f
s
c
o
r
e
s
b
e
t
we
e
n
f
o
l
ds
i
s
c
r
i
t
i
c
a
l
i
n
a
s
s
e
s
s
i
ng
m
o
de
l
s
t
a
bil
i
t
y
[
37]
.
T
h
e
v
a
l
ue
s
ge
ne
r
a
t
e
d
i
n
T
a
bl
e
2
s
h
o
w
c
o
n
s
i
s
t
e
n
c
y
,
i
n
d
i
c
a
t
i
n
g
t
h
a
t
t
h
e
M
T
R
F
m
o
de
l
i
s
s
t
a
bl
e
a
n
d
pr
o
vi
de
s
s
a
t
i
s
f
y
i
ng
e
v
a
l
ua
t
i
o
n
r
e
s
u
l
t
s
.
T
a
bl
e
2.
F
1
-
m
i
c
r
o
a
n
d
F
1
-
m
a
c
r
o
r
e
s
u
l
t
s
o
f
e
a
c
h
f
o
l
d
n
-
f
o
ld
M
ic
r
o
a
v
e
r
a
ge
M
a
c
r
o
a
ve
r
a
g
e
1
-
f
o
ld
0.91017964
0.8902439
2
-
f
o
ld
0.93279022
0.92105263
3
-
f
o
ld
0.91170431
0.8919598
4
-
f
o
ld
0.91836735
0.90654206
5
-
f
o
ld
0.91053678
0.88341969
6
-
f
o
ld
0.92337165
0.9
7
-
f
o
ld
0.93962264
0.92920354
8
-
f
o
ld
0.92116183
0.9
9
-
f
o
ld
0.91358025
0.890625
10
-
f
o
ld
0.93256262
0.91935484
4.
CONC
L
USI
ON
T
hi
s
s
t
ud
y
i
nv
e
s
t
i
ga
t
e
d
t
h
e
m
o
de
l
i
ng
o
f
c
l
us
t
e
r
i
ng
a
n
d
M
T
C
to
a
n
a
l
y
z
e
j
o
b
wa
i
t
i
n
g
t
i
m
e
a
n
d
j
o
b
li
ne
a
r
i
t
y
a
m
o
n
g
f
e
m
a
l
e
S
T
E
M
a
l
u
m
s
i
n
I
n
do
ne
s
i
a
.
Us
i
n
g
K
-
m
e
a
n
s
L
T
S
f
o
r
c
l
us
t
e
r
i
n
g,
we
c
o
n
duc
t
e
d
hy
pe
r
pa
r
a
m
e
t
e
r
t
uni
n
g
a
n
d
f
o
u
n
d
t
h
e
o
p
t
i
m
a
l
n
u
m
be
r
o
f
c
l
us
t
e
r
s
k
=
2
w
i
t
h
a
t
r
i
mm
i
ng
pe
r
c
e
n
t
a
ge
o
f
30%
,
yi
e
l
d
i
ng
a
s
il
h
o
ue
t
t
e
s
c
or
e
o
f
0.
77.
B
a
s
e
d
o
n
t
h
e
s
e
r
e
s
u
l
t
s
,
we
i
de
n
t
i
f
i
e
d
t
wo
di
s
t
i
n
c
t
a
l
u
m
gr
o
ups
:
t
h
o
s
e
wi
t
h
f
a
s
t
a
n
d
l
o
n
g
j
o
b
wa
i
t
i
ng
t
i
m
e
s
.
Not
a
bl
y
,
73%
o
f
f
e
m
a
l
e
a
l
u
m
s
b
e
l
o
n
ge
d
t
o
t
h
e
f
a
s
t
j
o
b
wa
i
t
i
n
g
gr
o
up.
T
h
e
r
e
s
u
l
t
i
n
g
c
l
us
t
e
r
s
we
r
e
f
ur
t
h
e
r
us
e
d
i
n
a
M
T
R
F
c
l
a
s
s
i
f
i
c
a
t
i
o
n
m
o
de
l
t
o
p
r
e
di
c
t
b
ot
h
j
o
b
wa
i
t
in
g
t
i
m
e
c
a
t
e
g
o
r
y
a
n
d
j
o
b
l
i
ne
a
r
i
t
y
.
T
h
e
b
e
s
t
pe
r
f
o
r
m
a
n
c
e
wa
s
a
c
hi
e
v
e
d
us
i
ng
a
t
un
e
d
RF
c
l
a
s
s
if
i
e
r
w
i
t
h
pa
r
a
m
e
t
e
r
s
m
a
x
_de
pt
h
=
20,
n
_e
s
t
i
m
a
t
o
r
s
=
300,
m
i
n_
s
a
m
p
les
_
l
e
a
f
=
1,
a
n
d
m
i
n_
s
a
m
p
l
e
s
_s
p
li
t
=
2.
E
v
a
l
ua
t
i
o
n
us
i
n
g
10
-
f
o
l
d
CV
r
e
ve
a
l
e
d
s
t
r
o
n
g
a
n
d
c
o
n
s
i
s
t
e
n
t
r
e
s
u
l
t
s
,
wi
t
h
F
1
-
m
i
c
r
o
r
e
a
c
hi
n
g
94%
a
n
d
F
1
-
m
a
c
r
o
r
e
a
c
h
i
ng
93
%
i
n
t
h
e
b
e
s
t
f
o
l
d.
T
h
e
c
l
a
s
s
i
f
i
c
a
t
i
o
n
o
f
j
o
b
wa
i
t
i
ng
t
i
m
e
yi
e
l
de
d
e
x
c
e
pt
i
o
n
a
ll
y
hi
g
h
pe
r
f
o
r
m
a
n
c
e
,
a
c
hi
e
vi
n
g
100%
a
c
c
ur
a
c
y
,
pr
e
c
i
s
i
o
n
,
r
e
c
a
l
l
,
a
n
d
F1
-
s
c
o
r
e
.
I
n
c
o
n
t
r
a
s
t,
t
h
e
j
o
b
l
i
ne
a
r
i
t
y
c
l
a
s
s
if
i
c
a
t
i
o
n
a
c
hi
e
v
e
d
l
o
we
r
a
c
c
ur
a
c
y
(
81%
)
due
to
t
h
e
c
o
m
p
l
e
xi
t
y
o
f
t
h
a
t
t
a
r
ge
t
a
n
d
t
h
e
l
i
mi
t
e
d
f
e
a
t
ur
e
s
a
v
a
il
a
bl
e
.
C
o
m
pa
r
e
d
t
o
pr
e
vi
o
us
wo
r
ks
t
h
a
t
a
ppl
i
e
d
f
uz
z
y
l
o
g
i
c
t
o
c
l
a
s
s
if
y
j
o
b
o
ut
c
o
m
e
s
a
n
d
t
h
a
t
us
e
d
GAs
f
o
r
o
p
t
i
mi
z
a
t
i
o
n
,
o
ur
a
ppr
o
a
c
h
de
m
o
ns
t
r
a
t
e
d
hi
g
h
e
r
i
n
t
e
r
pr
e
t
a
bi
li
t
y
a
n
d
i
m
pr
o
v
e
d
c
l
a
s
s
if
i
c
a
t
i
o
n
pe
r
f
o
r
m
a
n
c
e
t
h
r
o
ugh
a
c
o
m
bi
na
t
i
o
n
o
f
o
u
t
l
i
e
r
-
r
o
b
us
t
c
l
us
t
e
r
i
n
g
(
K
-
m
e
a
ns
L
T
S
)
a
n
d
m
u
l
t
i
-
t
a
r
ge
t
m
o
de
l
i
ng.
A
dd
it
i
o
n
a
ll
y
,
whil
e
m
a
ny
pr
i
o
r
s
t
udi
e
s
t
r
e
a
t
e
d
c
l
a
s
s
i
f
i
c
a
t
i
o
n
t
a
s
ks
i
n
d
e
pe
n
de
n
t
l
y
,
t
hi
s
wo
r
k
hi
g
hl
i
g
h
t
s
t
h
e
va
l
ue
o
f
m
o
de
li
ng
m
u
l
t
i
p
l
e
i
n
t
e
r
r
e
l
a
t
e
d
o
u
t
c
o
m
e
s
s
im
u
l
t
a
ne
o
us
l
y
us
i
ng
M
T
R
F
.
T
hi
s
s
t
udy
,
h
o
we
ve
r
,
i
s
n
o
t
wi
t
h
o
ut
li
mi
t
a
t
i
o
ns
.
F
i
r
s
t
,
t
h
e
f
e
a
t
ur
e
s
us
e
d
to
p
r
e
di
c
t
j
o
b
l
i
ne
a
r
i
t
y
m
a
y
n
o
t
f
u
l
ly
c
a
pt
ur
e
t
h
e
r
e
a
l
-
wo
r
l
d
f
a
c
t
o
r
s
i
n
f
l
u
e
n
c
i
ng
j
o
b
r
e
l
e
va
n
c
e
,
s
uc
h
a
s
j
o
b
t
y
pe
,
i
ndus
t
r
y
a
l
i
g
nm
e
n
t
,
or
pe
r
s
o
n
a
l
c
a
r
e
e
r
pr
e
f
e
r
e
n
c
e
s
.
T
h
e
da
t
a
s
e
t
wa
s
a
l
s
o
l
i
mi
t
e
d
to
f
e
m
a
l
e
gr
a
dua
t
e
s
f
r
o
m
a
s
m
a
l
l
s
a
m
p
l
e
o
f
pr
i
v
a
t
e
uni
ve
r
s
i
t
i
e
s
,
whi
c
h
m
a
y
a
f
f
e
c
t
t
h
e
ge
n
e
r
a
li
z
a
bil
i
t
y
o
f
t
h
e
f
i
nd
i
ngs
.
F
ut
ur
e
s
t
udi
e
s
s
h
o
u
l
d
c
o
n
s
i
de
r
e
x
pa
n
d
i
ng
t
h
e
f
e
a
t
ur
e
s
e
t
wi
t
h
r
i
c
h
e
r
a
c
a
de
mi
c
a
n
d
e
m
p
l
o
ym
e
n
t
hi
s
t
o
r
y
da
t
a
,
i
n
c
l
ud
i
ng
qua
li
t
a
t
i
v
e
i
nf
o
r
m
a
t
i
o
n
f
r
o
m
s
ur
v
e
y
s
o
r
i
n
t
e
r
vi
e
ws
.
A
dd
i
t
i
o
n
a
ll
y
,
a
pp
l
yi
ng
a
lt
e
r
n
a
t
i
v
e
m
u
l
t
i
-
t
a
r
ge
t
l
e
a
r
n
e
r
s
s
uc
h
a
s
de
e
p
n
e
ur
a
l
n
e
t
wor
ks
o
r
e
n
s
e
m
b
l
e
m
e
t
h
o
ds
m
a
y
im
pr
o
v
e
ge
n
e
r
a
li
z
a
bil
i
t
y
,
e
s
pe
c
i
a
ll
y
f
o
r
m
o
r
e
c
o
m
p
l
e
x
t
a
r
ge
t
s
l
i
ke
j
o
b
li
ne
a
r
i
t
y
.
I
n
s
u
mm
a
r
y
,
t
hi
s
r
e
s
e
a
r
c
h
de
m
o
ns
t
r
a
t
e
s
t
h
a
t
c
om
b
i
n
i
ng
c
l
u
s
t
e
r
i
n
g
(
K
-
m
e
a
n
s
L
T
S
)
a
n
d
M
T
R
F
m
o
de
li
ng
c
a
n
e
f
f
e
c
t
i
v
e
ly
u
n
c
o
v
e
r
pa
tt
e
r
n
s
i
n
gr
a
dua
t
e
e
m
p
l
o
ym
e
n
t
o
u
t
c
o
m
e
s
,
pa
r
t
i
c
u
l
a
r
ly
i
n
j
o
b
w
a
i
t
i
n
g
t
i
m
e
.
De
s
p
i
t
e
t
h
e
u
n
de
r
r
e
pr
e
s
e
n
t
a
t
i
o
n
o
f
wo
m
e
n
i
n
S
T
E
M
,
t
h
e
da
t
a
r
e
v
e
a
l
t
h
a
t
a
s
i
g
nif
i
c
a
n
t
p
o
r
t
i
o
n
o
f
f
e
m
a
l
e
a
l
u
m
ni
s
uc
c
e
s
s
f
u
l
ly
e
n
t
e
r
t
h
e
w
o
r
kf
o
r
c
e
qui
c
k
ly
a
n
d
s
h
o
w
s
tr
o
n
g
a
c
a
de
mi
c
pe
r
f
o
r
m
a
n
c
e
.
T
h
e
s
e
f
i
nd
i
n
g
s
s
uppo
r
t
t
h
e
n
o
t
i
o
n
t
h
a
t
f
e
m
a
l
e
S
T
E
M
gr
a
dua
t
e
s
p
o
s
s
e
s
s
c
o
m
pe
t
i
t
i
v
e
c
a
pa
bi
li
t
i
e
s
a
n
d
r
e
s
il
i
e
n
c
e
i
n
t
h
e
j
o
b
m
a
r
ke
t
.
AC
K
NOWL
E
DGM
E
NT
S
T
hi
s
r
e
s
e
a
r
c
h
wa
s
c
o
n
duc
t
e
d
un
de
r
t
h
e
a
us
p
i
c
e
s
of
B
i
na
Nu
s
a
n
t
a
r
a
Uni
v
e
r
s
i
t
y
.
I
s
i
n
c
e
r
e
ly
t
h
a
n
k
t
h
e
f
a
c
u
l
t
y
,
s
t
a
f
f
,
a
n
d
a
l
l
c
o
n
t
r
i
b
ut
o
r
s
f
o
r
t
h
e
i
r
v
a
l
ua
ble
s
uppo
r
t
a
n
d
a
s
s
i
s
t
a
n
c
e
.
T
hi
s
pa
pe
r
a
l
s
o
u
t
i
li
z
e
d
A
I
too
l
s
to
a
s
s
i
s
t
wi
t
h
t
r
a
n
s
l
a
t
i
o
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8776
I
n
t
J
I
n
f
&
C
o
m
m
u
n
T
e
c
hn
o
l
,
Vo
l
.
1
5
,
N
o.
1
,
M
a
r
c
h
20
2
6
:
1
8
9
-
1
9
7
196
F
UN
DI
NG
I
N
F
ORM
AT
I
ON
T
hi
s
r
e
s
e
a
r
c
h
r
e
c
e
i
ve
d
n
o
gr
a
n
t
f
r
o
m
a
ny
f
u
n
d
i
ng
a
ge
nc
y
i
n
t
h
e
pu
bli
c
,
c
o
m
m
e
r
c
i
a
l
,
o
r
n
ot
-
f
o
r
-
pr
o
f
i
t
s
e
c
t
or
s
.
AU
T
HO
R
CONT
RI
B
UT
I
ONS
S
T
AT
E
M
E
NT
T
hi
s
j
o
ur
n
a
l
u
s
e
s
t
h
e
C
o
n
t
r
i
b
ut
or
R
o
l
e
s
T
a
x
o
n
o
m
y
(
C
R
e
d
i
T
)
to
r
e
c
o
gni
z
e
i
n
d
i
v
i
dua
l
a
ut
h
o
r
c
o
n
t
r
i
b
ut
i
o
ns
,
r
e
duc
e
a
ut
h
or
s
hi
p
d
i
s
put
e
s
,
a
n
d
f
a
c
il
i
t
a
t
e
c
o
l
l
a
b
o
r
a
t
i
o
n
.
Nam
e
of
Aut
h
or
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
Na
dz
l
a
An
dr
i
t
a
I
n
t
a
n
Gh
a
y
a
t
r
i
e
De
vi
F
i
t
r
i
a
n
a
h
C
:
C
o
n
c
e
pt
ua
li
z
a
ti
o
n
M
:
M
e
th
o
d
o
l
o
g
y
So
:
So
f
twa
r
e
Va
:
Va
li
da
ti
o
n
Fo
:
Fo
r
ma
l
a
na
l
y
s
is
I
:
I
nve
s
ti
ga
ti
o
n
R
:
R
e
s
o
u
r
c
e
s
D
:
D
a
ta
C
ur
a
ti
o
n
O
:
W
r
it
in
g
-
O
r
ig
in
a
l
D
r
a
f
t
E
:
W
r
it
in
g
-
R
e
v
i
e
w
&
E
di
ti
ng
Vi
:
Vi
s
ua
li
z
a
ti
o
n
Su
:
Su
pe
r
v
is
io
n
P
:
P
r
o
j
e
c
t
a
dmi
ni
s
tr
a
ti
o
n
Fu
:
Fu
ndi
ng a
c
qui
s
it
i
o
n
CONF
L
I
CT
OF
I
NT
E
RE
S
T
S
T
AT
E
M
E
NT
T
h
e
a
ut
h
o
r
s
de
c
l
a
r
e
n
o
c
o
nf
li
c
t
o
f
i
n
t
e
r
e
s
t
.
DA
T
A
AV
AI
L
AB
I
L
I
T
Y
T
h
e
da
t
a
us
e
d
i
n
t
hi
s
s
t
ud
y
a
r
e
n
ot
publ
i
c
ly
a
v
a
il
a
bl
e
due
to
s
t
r
i
c
t
c
o
nf
i
d
e
n
t
i
a
l
i
t
y
a
gr
e
e
m
e
n
t
s
wi
t
h
m
u
l
t
i
p
l
e
pa
r
t
n
e
r
uni
ve
r
s
i
t
i
e
s
.
S
h
a
r
i
n
g
o
f
t
hi
s
da
t
a
i
s
r
e
s
t
r
i
c
t
e
d
a
s
pe
r
t
h
e
s
i
g
n
e
d
a
gr
e
e
m
e
n
t
s
a
n
d
i
ns
t
i
t
ut
i
o
n
a
l
po
l
i
c
i
e
s
.
RE
F
E
R
E
NC
E
S
[
1]
U
N
D
P
, “
G
e
nd
e
r
i
n
e
qua
li
t
y
i
nd
e
x
,”
pp. 156
–
159, 2013, d
o
i:
10.
18356/c
791776a
-
e
n.
[
2]
W
.
E
.
F
o
r
um,
“
G
l
o
ba
l
g
e
nde
r
ga
p
r
e
p
o
r
t,
”
W
E
F
,
2023,
[
O
nl
in
e
]
.
A
v
a
i
la
bl
e
:
ht
tp
s
:/
/ww
w
3.w
e
f
o
r
um.
or
g/
d
o
c
s
/W
E
F
_
G
G
G
R
_2023.pd
f
.
[
3]
P
.
P
.
L
ia
ng,
A
.
Z
a
de
h,
a
nd
L
.
P
.
M
or
e
n
c
y
,
“
F
o
unda
ti
o
ns
a
nd
t
r
e
nds
in
mul
ti
mo
da
l
ma
c
hi
ne
l
e
a
r
ni
ng:
pr
in
c
ip
l
e
s
,
c
ha
ll
e
nge
s
,
a
nd
o
p
e
n qu
e
s
ti
o
ns
,”
A
C
M
C
om
put
in
g Sur
v
e
y
s
, v
o
l.
56, n
o
. 10, 202
4, do
i:
10.1145/3656580.
[
4]
K
. M
a
ha
r
a
na
, S
. M
o
nda
l,
a
nd
B
. N
e
ma
de
,
“
A
r
e
v
ie
w
:
D
a
ta
pr
e
-
pr
oc
e
s
s
in
g a
nd da
ta
a
ugme
nt
a
ti
o
n
t
e
c
hni
qu
e
s
,”
G
lo
bal
T
r
ans
it
i
ons
P
r
oc
e
e
di
ngs
, vo
l.
3, n
o
. 1, pp. 91
–
99, 2022,
d
o
i:
10.1016/
j.
gl
tp
.
2022.04.020.
[
5]
C
.
K
ho
s
la
a
nd
B
.
S
.
S
a
in
i,
“
E
nha
nc
in
g
p
e
r
f
or
ma
nc
e
of
d
e
e
p
le
a
r
ni
ng
m
o
de
ls
w
it
h
di
f
f
e
r
e
nt
da
ta
a
ugme
nt
a
ti
o
n
t
e
c
hni
q
ue
s
:
a
s
ur
v
e
y
,”
P
r
oc
e
e
di
ngs
of
I
nt
e
r
nat
io
nal
C
on
f
e
r
e
nc
e
on
I
nt
e
ll
ig
e
nt
E
ngi
ne
e
r
in
g
and
M
anage
m
e
nt
,
I
C
I
E
M
2020
,
pp.
79
–
85,
2020,
do
i:
10.1109/
I
C
I
E
M
48762.2020.9160048.
[
6]
H
.
H
e
,
Y
.
B
a
i,
E
.
A
.
G
a
r
c
ia
,
a
nd
S
.
L
i,
“
A
D
A
S
Y
N
:
a
da
pt
i
ve
s
y
nt
h
e
ti
c
s
a
mpl
in
g
a
pp
r
o
a
c
h
f
o
r
im
ba
la
nc
e
d
le
a
r
n
in
g,”
P
r
oc
e
e
di
ngs
of
t
he
I
nt
e
r
nat
io
nal
J
oi
nt
C
on
f
e
r
e
nc
e
on N
e
u
r
al
N
e
tw
or
k
s
, pp.
132
2
–
1328, 2008, do
i:
10.1109/
I
J
C
N
N
.2008.4633969.
[
7]
N
.
V
.
C
ha
w
l
a
,
K
.
W
.
B
o
w
y
e
r
,
L
.
O
.
H
a
ll
,
a
nd
W
.
P
.
K
e
g
e
lm
e
y
e
r
,
“
S
M
O
T
E
:
s
y
n
th
e
t
ic
mi
n
or
it
y
ove
r
-
s
a
mpl
in
g
te
c
hni
q
ue
,”
J
our
nal
of
A
r
ti
f
ic
ia
l
I
nt
e
ll
ig
e
nc
e
R
e
s
e
ar
c
h
, v
o
l.
16, pp. 321
–
35
7, 2002, do
i:
10.1613/j
a
i
r
.953.
[
8]
L
.
R
o
ka
c
h
a
nd
O
.
M
a
im
o
n,
“
C
lu
s
te
r
in
g
me
th
o
ds
,”
D
at
a
M
in
in
g
and
K
now
le
dge
D
is
c
ov
e
r
y
H
andboo
k
,
pp.
321
–
352,
20
06,
do
i:
10.1007/0
-
387
-
25465
-
x
_15.
[
9]
D
.
F
it
r
ia
na
h
a
nd
Y
.
H
a
r
w
a
ni
,
“
A
lu
mni
a
bs
or
pt
i
o
n
p
r
e
di
c
ti
o
n
f
r
a
m
e
w
o
r
k
ba
s
e
d
o
n
ma
c
hi
n
e
l
e
a
r
ni
ng
(
in
I
ndone
s
ia
n
)
,”
E
ngi
ne
e
r
in
g,
M
A
th
e
m
at
ic
s
and
C
om
put
e
r
S
c
ie
nc
e
(
E
M
A
C
S)
J
our
nal
,
vo
l.
2,
no
.
1,
pp.
21
–
28,
2020,
do
i:
10.21512/e
ma
c
s
j
o
ur
na
l.
v
2i
1.6251.
[
10]
Y
.
J
in
g
a
nd
J
.
W
a
ng,
“
I
nf
lu
e
n
c
in
g
f
a
c
t
o
r
s
a
na
l
y
s
is
of
c
o
ll
e
g
e
e
mpl
oy
m
e
nt
ba
s
e
d
o
n
f
uz
z
y
c
lu
s
te
r
in
g,”
Ad
v
anc
e
s
in
T
r
ans
di
s
c
ip
li
nar
y
E
ngi
ne
e
r
in
g
, v
o
l.
47, pp. 1206
–
1213, 2024, d
o
i:
10.3233/A
T
D
E
231305.
[
11]
N
.
A
.
I
.
G
ha
y
a
t
r
i
e
,
W
.
P
.
S
a
r
i,
D
.
F
it
r
ia
na
h,
a
nd
E
.
C
he
w
,
“
A
na
ly
z
in
g
S
T
E
M
w
o
m
e
n
’
s
e
mpl
oy
a
bi
l
it
y
:
a
ma
c
hi
n
e
l
e
a
r
ni
ng
f
r
a
me
w
o
r
k
w
it
h
c
lu
s
te
r
in
g
a
nd
mul
ti
-
ta
r
g
e
t
c
la
s
s
if
i
c
a
ti
o
n,”
I
C
oC
SE
T
I
2025
-
I
nt
e
r
nat
io
nal
C
onf
e
r
e
nc
e
on
C
om
put
e
r
Sc
ie
n
c
e
s
,
E
ngi
ne
e
r
in
g, and T
e
c
hnol
ogy
I
nnov
at
io
n, P
r
oc
e
e
di
ng
, pp. 750
–
755, 2025, do
i:
10.1109/
I
C
o
C
S
E
T
I
63724.2025.11019987.
[
12]
T
.
E
s
t
e
ll
a
,
N
.
A
nd
r
it
a
I
nt
a
n
G
ha
y
a
tr
i
e
,
a
nd
A
.
W
ib
o
w
o
,
“
O
ut
l
ie
r
ha
ndl
in
g
in
c
lu
s
t
e
r
in
g:
a
c
o
mpa
r
a
ti
ve
e
x
pe
r
im
e
nt
of
K
-
m
e
a
ns
,
r
o
bus
t
tr
im
me
d
K
-
m
e
a
ns
,
a
nd
K
-
m
e
a
ns
le
a
s
t
tr
im
me
d
s
qua
r
e
d,
”
I
nt
e
r
nat
io
nal
J
our
nal
o
f
C
om
put
in
g
and
D
ig
it
al
S
y
s
te
m
s
,
v
o
l.
16,
no
. 1, pp. 1029
–
1039, 2024, d
o
i:
10.12785/i
jc
ds
/1
60175.
[
13]
F
.
K
.
N
a
ka
no
,
K
.
P
li
a
k
o
s
,
a
nd
C
.
V
e
ns
,
“
D
e
e
p
tr
e
e
-
e
ns
e
mbl
e
s
f
o
r
mul
ti
-
o
ut
put
p
r
e
di
c
ti
o
n,”
P
at
te
r
n
R
e
c
ogni
ti
on
,
vol
.
121,
2
021,
do
i:
10.1016/j
.pa
t
c
o
g.2021.108211.
[
14]
D
.
X
u,
Y
.
S
hi
,
I
.
W
.
T
s
a
ng,
Y
.
S
.
O
ng,
C
.
G
o
ng,
a
nd
X
.
S
he
n,
“
S
ur
ve
y
o
n
mul
ti
-
o
ut
put
l
e
a
r
ni
ng,”
I
E
E
E
T
r
ans
ac
ti
ons
on
N
e
ur
al
N
e
tw
or
k
s
and L
e
ar
ni
ng Sy
s
te
m
s
, v
ol
. 31, n
o
. 7, pp. 2409
–
2429,
2020, do
i:
10.1109/
T
N
N
L
S
.2019.2945133.
[
15]
Y
.
Y
us
of
,
F
.
B
a
ha
r
o
m,
N
.
J
a
ma
lu
di
n,
a
nd
S
.
B
a
dr
o
ddi
n,
“
B
in
a
r
y
-
ba
s
e
d
c
la
s
s
if
i
c
a
ti
o
n
m
o
d
e
ls
f
o
r
M
a
la
y
s
ia
n
gr
a
dua
te
s
e
mpl
oy
m
e
nt
a
na
l
y
s
is
,”
pp. 501
–
513, 2025, d
o
i:
10.1007/978
-
3
-
031
-
91485
-
0_39.
[
16]
R
.
P
ic
o
-
S
a
lt
o
s
,
J
.
G
a
r
z
á
s
,
A
.
R
e
dc
h
uk
,
P
.
E
s
c
a
nd
ón
-
P
a
nc
h
a
n
a
,
a
nd
F
.
M
o
r
a
nt
e
-
C
a
r
ba
ll
o
,
“
R
o
le
o
f
a
l
um
ni
p
r
o
gr
a
m
i
n
t
he
pr
e
d
ic
ti
o
n
o
f
c
a
r
e
e
r
s
uc
c
e
s
s
i
n
a
n e
c
u
a
do
r
i
a
n
pu
bl
ic
U
ni
v
e
r
s
it
y
,”
A
p
pl
i
e
d
Sc
ie
nc
e
s
(
Sw
it
z
e
r
l
an
d)
,
v
o
l.
12
,
no
.
19
,
20
22
,
do
i:
1
0.
33
90/
a
pp
12
19
989
2
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t
J
I
n
f
&
C
o
m
m
u
n
T
e
c
hn
o
l
I
S
S
N:
2252
-
8776
A
n
int
e
gr
ati
on
c
lus
ter
ing
and
multi
-
tar
ge
t
c
las
s
if
ic
ati
on
appr
oac
h
…
(
N
adz
la
A
ndr
it
a
I
ntan
Ghay
atr
i
e
)
197
[
17]
R
.
A
ma
li
a
a
nd
A
.
W
ib
o
w
o
,
“
P
r
e
di
c
ti
o
n
of
th
e
w
a
it
in
g
ti
me
p
e
r
i
o
d
f
o
r
ge
t
ti
ng
a
j
o
b
us
in
g
th
e
na
i
ve
B
a
y
e
s
a
lg
or
it
hm,”
I
nt
e
r
nat
io
nal
R
e
s
e
ar
c
h J
ou
r
nal
o
f
A
dv
anc
e
d E
ngi
ne
e
r
in
g and S
c
ie
nc
e
, v
o
l.
5, n
o
. 2, pp. 225
–
229, 2020.
[
18]
M
.
B
a
s
ga
lu
pp,
R
.
C
e
r
r
i,
L
.
S
c
h
ie
tg
a
t,
I
.
T
r
ig
u
e
r
o
,
a
nd
C
.
V
e
ns
,
“
B
e
y
o
nd
gl
o
ba
l
a
nd
l
o
c
a
l
mul
ti
-
ta
r
g
e
t
le
a
r
n
in
g,”
I
nf
or
m
at
io
n
Sc
ie
nc
e
s
, v
o
l.
579, pp. 508
–
524, 2021, d
o
i:
10.1016/j
.i
ns
.2021.
08
.022.
[
19]
O
.
W
.
S
a
mue
l,
G
.
M
.
A
s
o
gb
o
n,
A
.
K
.
S
a
nga
ia
h,
a
nd
G
.
L
i,
“
C
o
mput
a
ti
o
na
l
in
t
e
ll
ig
e
n
c
e
e
na
bl
in
g
th
e
d
e
v
e
l
o
pme
n
t
of
e
f
f
i
c
i
e
nt
c
li
ni
c
a
l
d
e
c
is
i
o
n
s
upp
o
r
t
s
y
s
t
e
ms
:
C
a
s
e
s
tu
d
y
of
he
a
r
t
f
a
il
ur
e
,”
C
om
put
at
io
nal
I
nt
e
ll
ig
e
nc
e
f
or
M
ul
ti
m
e
di
a
B
ig
D
at
a
on
th
e
C
lo
ud
w
it
h E
ngi
ne
e
r
in
g A
ppl
ic
at
io
ns
, pp. 123
–
133, 2018, d
o
i:
10.101
6/
B
978
-
0
-
12
-
813314
-
9.00005
-
0.
[
20]
J
. H
unt
e
r
, D
. D
a
le
, E
.
F
ir
in
g, a
nd M
. D
r
oe
tt
b
oo
m, “
ma
tp
l
o
tl
ib
.p
y
pl
o
t.
b
ox
pl
o
t,
”
m
at
pl
ot
li
b
.
[
21]
X
.
W
a
ng,
J
.
X
u,
T
.
Z
e
ng,
a
nd
L
.
J
in
g,
“
L
oc
a
l
di
s
tr
ib
ut
i
o
n
-
ba
s
e
d
a
da
pt
i
ve
mi
n
o
r
it
y
ove
r
s
a
mpl
in
g
f
o
r
im
ba
la
nc
e
d
da
ta
c
la
s
s
if
i
c
a
ti
o
n,”
N
e
ur
oc
om
put
in
g
, vo
l.
422, pp. 200
–
213, 2021,
do
i:
10.1016/j
.n
e
u
c
o
m.2020.05.030.
[
22]
T
.
D
a
o
,
A
.
G
u,
A
.
J
.
R
a
tn
e
r
,
V
.
S
mi
th
,
C
.
D
e
S
a
,
a
nd
C
.
R
é
,
“
A
ke
r
n
e
l
th
e
o
r
y
of
m
o
d
e
r
n
da
ta
a
ugm
e
nt
a
ti
o
n,”
36t
h
I
nt
e
r
nat
io
nal
C
onf
e
r
e
nc
e
on M
ac
hi
ne
L
e
ar
ni
ng, I
C
M
L
2019
, vo
l.
2019
-
J
une
,
pp. 2755
–
2775, 2019.
[
23]
Z
.
D
ua
n,
C
.
W
a
ng,
a
nd
W
.
Z
ho
ng,
“
S
S
G
C
L
:
S
im
pl
e
s
o
c
ia
l
r
e
c
o
mm
e
nda
ti
o
n
w
it
h
gr
a
ph
c
o
nt
r
a
s
ti
ve
le
a
r
ni
ng,”
M
at
he
m
at
ic
s
,
vo
l.
12, n
o
. 7, 2024, d
o
i
:
10.3390/
ma
th
12071107.
[
24]
A
.
A
.
A
bdul
na
s
s
a
r
a
nd
L
.
R
.
N
a
ir
,
“
P
e
r
f
o
r
ma
n
c
e
a
na
l
y
s
is
of
K
m
e
a
ns
w
it
h
m
o
di
f
ie
d
in
it
ia
l
c
e
nt
r
o
id
s
e
l
e
c
ti
o
n
a
lg
o
r
it
h
ms
a
nd
de
ve
l
o
pe
d
K
m
e
a
ns
9+
mo
d
e
l,
”
M
e
as
ur
e
m
e
nt
:
Se
ns
or
s
, v
o
l.
25,
2023, do
i:
10.1016/j
.
me
a
s
e
n.2023.100666.
[
25]
O
.
D
o
r
a
bi
a
la
,
J
.
N
.
K
ut
z
,
a
nd
A
.
Y
.
A
r
a
v
ki
n,
“
R
o
bus
t
tr
im
me
d
K
-
m
e
a
ns
,”
P
at
te
r
n
R
e
c
ogni
ti
on
L
e
tt
e
r
s
,
v
o
l.
161,
pp.
9
–
16,
20
22,
do
i:
10.1016/j
.pa
t
r
e
c
.2022.07.007.
[
26]
M
.
S
hut
a
y
w
i
a
nd
N
.
N
.
K
a
c
h
o
u
ie
,
“
S
il
h
o
u
e
tt
e
a
na
l
y
s
is
f
or
pe
r
f
or
ma
nc
e
e
v
a
lu
a
ti
o
n
in
ma
c
h
in
e
l
e
a
r
ni
ng
w
it
h
a
ppl
ic
a
ti
o
n
s
to
c
lu
s
te
r
in
g,”
E
nt
r
opy
, vo
l.
23, n
o
. 6, 2021, d
oi
:
10.3390/
e
23060
759.
[
27]
K
.
R
.
S
ha
ha
pur
e
a
nd
C
.
N
ic
h
o
la
s
,
“
C
lu
s
te
r
qua
li
t
y
a
na
l
y
s
is
us
i
ng
s
il
ho
u
e
tt
e
s
c
o
r
e
,”
in
P
r
oc
e
e
di
ngs
-
2020
I
E
E
E
7t
h
I
n
te
r
nat
io
nal
C
onf
e
r
e
nc
e
on Data Sc
ie
nc
e
and A
dv
an
c
e
d A
nal
y
ti
c
s
, D
SA
A
2020
, 2020, pp. 747
–
748, do
i:
10.1109/DS
A
A
49011.2020.00096.
[
28]
F
. H
ut
t
e
r
,
L
.
K
o
tt
h
of
f
, a
nd J
. V
a
ns
c
ho
r
e
n,
A
u
to
m
at
e
d m
ac
hi
ne
l
e
ar
ni
n
g
. C
ha
m:
S
pr
in
ge
r
I
nt
e
r
na
ti
o
na
l
P
ub
li
s
hi
ng, 2019.
[
29]
L
.
Y
a
ng a
nd A
. S
ha
mi
, “
O
n h
y
p
e
r
pa
r
a
me
t
e
r
o
pt
im
i
z
a
ti
o
n
of
ma
c
hi
n
e
l
e
a
r
ni
ng a
lg
o
r
it
h
ms
:
T
h
e
o
r
y
a
nd pr
a
c
ti
c
e
,”
N
e
ur
oc
om
put
in
g
,
vo
l.
415, pp. 295
–
316, 2020, d
o
i:
10.1016/
j.
ne
u
c
o
m.2020.07.06
1.
[
30]
F
.
A
bba
s
e
t
al
.
,
“
O
pt
im
iz
in
g
ma
c
hi
n
e
l
e
a
r
ni
ng
a
lg
o
r
it
hms
f
or
la
nds
li
de
s
us
c
e
pt
ib
i
li
t
y
ma
ppi
ng
a
lo
ng
th
e
ka
r
a
k
o
r
a
m
hi
gh
w
a
y
,
gi
lg
it
ba
lt
is
ta
n,
pa
ki
s
ta
n:
a
c
o
mpa
r
a
ti
v
e
s
tu
d
y
of
ba
s
e
li
ne
,
ba
ye
s
ia
n,
a
nd
me
ta
he
ur
is
ti
c
h
y
p
e
r
pa
r
a
me
t
e
r
o
pt
im
i
z
a
ti
o
n
t
e
c
hn
iq
u
e
s
,”
Se
ns
or
s
, v
o
l.
23, n
o
. 15, 2023, d
o
i:
10.3390/s
23156843.
[
31]
S
.
W
u,
Y
.
C
he
n,
Z
.
L
i,
J
.
L
i,
F
.
Z
ha
o
,
a
nd
X
.
S
u,
“
T
o
w
a
r
ds
mul
ti
-
la
be
l
c
la
s
s
if
ic
a
ti
o
n:
n
e
x
t
s
t
e
p
of
ma
c
hi
n
e
l
e
a
r
ni
ng
f
or
mi
c
r
o
bi
ome
r
e
s
e
a
r
c
h,”
C
om
put
at
io
nal
and
St
r
uc
tu
r
al
B
io
te
c
hnol
ogy
J
our
nal
,
v
ol
.
19,
pp
.
2742
–
2749,
2021,
do
i:
10.1016/j
.c
s
bj
.2021.04.054.
[
32]
B
.
P
.
K
oy
a
,
S
.
A
n
e
ja
,
R
.
G
upt
a
,
a
nd
C
.
V
a
le
o
,
“
C
o
mpa
r
a
ti
ve
a
na
l
y
s
is
of
d
i
f
f
e
r
e
nt
ma
c
hi
ne
l
e
a
r
ni
ng
a
lg
o
r
i
th
ms
to
pr
e
di
c
t
me
c
ha
ni
c
a
l
p
r
o
pe
r
ti
e
s
of
c
o
nc
r
e
te
,”
M
e
c
hani
c
s
of
A
dv
anc
e
d
M
at
e
r
ia
ls
and
St
r
uc
tu
r
e
s
,
v
ol
.
29,
n
o
.
25,
pp.
4032
–
4043,
2
022,
do
i:
10.1080/15376494.
2021.1917021.
[
33]
K
.
H
of
f
ma
n,
J
.
Y
.
S
ung,
a
nd
A
.
Z
a
z
z
e
r
a
,
“
M
ul
ti
-
o
ut
put
r
a
ndo
m
f
or
e
s
t
r
e
gr
e
s
s
io
n
t
o
e
mul
a
t
e
th
e
e
a
r
li
e
s
t
s
ta
ge
s
of
pl
a
ne
t
f
or
ma
ti
o
n,”
2021
I
E
E
E
Sy
s
te
m
s
and
I
nf
or
m
at
io
n
E
ngi
n
e
e
r
in
g
D
e
s
ig
n
Sy
m
pos
iu
m
,
SI
E
D
S
2021
,
2021,
do
i:
10.1109/S
I
E
D
S
52267.2021.9483749.
[
34]
W
.
W
a
e
ge
ma
n,
K
.
D
e
mb
c
z
y
ńs
ki
,
a
nd
E
.
H
ül
l
e
r
m
e
i
e
r
,
“
M
ul
ti
-
ta
r
ge
t
pr
e
di
c
ti
o
n:
a
uni
f
y
in
g
v
i
e
w
o
n
pr
o
b
le
ms
a
nd
me
th
ods
,”
D
at
a M
in
in
g and K
now
le
dge
D
is
c
ov
e
r
y
, v
o
l.
33, n
o
. 2,
pp. 293
–
324, 2019, do
i:
10.1007/s
10618
-
018
-
0595
-
5.
[
35]
M
.
G
r
a
ndi
ni
,
E
.
B
a
gl
i,
a
nd
G
.
V
is
a
ni
,
“
M
e
t
r
i
c
s
f
o
r
mul
ti
-
c
la
s
s
c
la
s
s
i
f
ic
a
ti
o
n
:
a
n
ove
r
v
i
e
w
,”
2020,
[
O
nl
in
e
]
.
A
v
a
il
a
bl
e
:
ht
tp
:/
/a
r
x
i
v
.
o
r
g/
a
bs
/2
008.05756.
[
36]
K
.
T
a
ka
ha
s
hi
,
K
.
Y
a
ma
mo
t
o
,
A
.
K
u
c
hi
ba
,
a
n
d
T
.
K
oy
a
ma
,
“
C
o
n
f
id
e
n
c
e
in
t
e
r
v
a
l
f
o
r
mi
c
r
o
-
a
ve
r
a
g
e
d
F
1
a
nd
ma
c
r
o
-
a
ve
r
a
ge
d
F
1
s
c
or
e
s
,”
A
ppl
ie
d I
n
te
ll
ig
e
nc
e
, vo
l.
52, n
o
. 5, pp. 4961
–
4972, 20
22, do
i:
10.1007/s
10489
-
021
-
02635
-
5.
[
37]
L
.
A
.
Y
a
te
s
,
Z
.
A
a
nda
hl
,
S
.
A
.
R
ic
ha
r
ds
,
a
nd
B
.
W
.
B
r
oo
k,
“
C
r
o
s
s
v
a
li
d
a
ti
o
n
f
o
r
m
o
d
e
l
s
e
le
c
ti
o
n
:
a
r
e
v
i
e
w
w
it
h
e
x
a
mpl
e
s
f
r
o
m
e
c
o
l
o
g
y
,”
E
c
ol
ogi
c
al
M
onogr
aphs
, vo
l.
93, n
o
. 1, 2023, d
o
i:
10
.1002/e
c
m.1557.
[
38]
J
.
B
r
a
ndt
a
nd
E
.
L
a
n
z
é
n,
“
A
c
o
mpa
r
a
ti
ve
r
e
v
i
e
w
of
S
M
O
T
E
a
nd
A
D
A
S
Y
N
in
im
ba
la
nc
e
d
da
ta
c
la
s
s
if
i
c
a
ti
o
n,”
2021
,
p.
42,
2
020,
[
O
nl
in
e
]
.
A
v
a
il
a
bl
e
:
ht
tp
s
:/
/ww
w
.di
v
a
-
p
o
r
ta
l.
or
g/
s
ma
s
h/
r
e
c
o
r
d.
js
f
?
pi
d=
di
v
a
2:
1519153.
B
I
OG
RA
P
HI
E
S
OF
AU
T
HO
RS
N
a
dzl
a
A
n
dri
ta
I
n
ta
n
G
h
a
y
a
tri
e
rece
i
v
e
d
h
i
s
fi
rs
t
d
eg
ree
fr
o
m
B
i
n
a
N
u
s
an
t
ara
U
n
i
v
e
rs
i
t
y
,
Co
m
p
u
t
e
r
S
c
i
en
ce
,
J
ak
art
a,
i
n
2
0
2
4
.
Sh
e
h
as
al
s
o
m
as
t
e
r
’
s
d
e
g
r
ee
fro
m
Bi
n
a
N
u
s
an
t
ara
U
n
i
v
e
rs
i
t
y
,
J
ak
art
a,
I
n
d
o
n
e
s
i
a,
i
n
2
0
2
5
.
H
e
r
r
e
s
e
ar
c
h
i
n
t
e
r
e
s
t
s
ce
n
t
e
r
ar
o
u
n
d
ad
d
r
e
s
s
i
n
g
g
e
n
d
e
r
g
ap
i
s
s
u
e
s
t
h
ro
u
g
h
i
n
n
o
v
at
i
v
e
ap
p
l
i
c
at
i
o
n
s
o
f
d
at
a
s
c
i
en
ce
,
ML
,
an
d
AI
.
Sh
e
c
an
b
e
co
n
t
a
c
t
ed
at
em
ai
l
:
n
ad
z
l
a.
g
h
a
y
at
ri
e@
b
i
n
u
s
.
ac
.
i
d
.
D
ev
i
F
i
tr
i
a
n
a
h
i
s
a
l
ec
t
u
r
e
r
an
d
r
e
s
e
ar
ch
e
r
at
t
h
e
Mas
t
e
r
o
f
Co
m
p
u
t
e
r
S
ci
e
n
ce
D
e
p
art
men
t
,
Bi
n
a
N
u
s
an
t
ara
U
n
i
v
e
rs
i
t
y
.
S
h
e
e
ar
n
e
d
h
e
r
B.
Sc.
i
n
C
o
m
p
u
t
e
r
S
ci
e
n
ce
(2
0
0
0
)
fro
m
Bi
n
a
N
u
s
an
t
ara
U
n
i
v
e
rs
i
t
y
,
a
n
d
h
e
r
M.
I
T
(2
0
0
8
)
an
d
Ph
.
D
.
i
n
Co
m
p
u
t
e
r
S
c
i
en
ce
(2
0
1
5
)
fro
m
U
n
i
v
e
rs
i
t
as
In
d
o
n
e
s
i
a
.
In
2
0
1
4
,
s
h
e
j
o
i
n
ed
t
h
e
PR
I
PG
IS
L
ab
,
M
i
c
h
i
g
a
n
St
at
e
U
n
i
v
e
rs
i
t
y
,
U
SA
,
t
h
ro
u
g
h
a
s
an
d
w
i
ch
p
ro
g
ra
m,
an
d
i
s
cu
rr
e
n
t
l
y
a
f
e
l
l
o
w
r
e
s
e
ar
c
h
e
r
at
E
u
r
e
k
a
R
o
b
o
t
i
c
s
L
ab
,
Card
i
ff
M
e
t
ro
p
o
l
i
t
an
U
n
i
v
e
rs
i
t
y
,
U
K
.
H
e
r
r
e
s
e
ar
c
h
i
n
t
ere
s
t
s
i
n
c
l
u
d
e
d
at
a
mi
n
i
n
g
,
ML
,
AI
,
a
n
d
ap
p
l
i
e
d
r
em
o
t
e
s
en
s
i
n
g
.
S
h
e
c
a
n
b
e
c
o
n
t
ac
t
ed
at
em
ai
l
:
d
e
v
i
.
f
i
t
ri
an
a
h
@
b
i
n
u
s
.
a
c
.
i
d
.
Evaluation Warning : The document was created with Spire.PDF for Python.