I
n
t
e
r
n
at
i
on
al
Jou
r
n
al
of
E
l
e
c
t
r
i
c
al
an
d
C
o
m
p
u
t
er E
n
g
i
n
eeri
n
g
(
I
J
E
C
E
)
V
o
l.
11
,
N
o.
6
,
D
ecem
b
er
202
1
,
pp.
55
49
~
5
557
I
S
S
N
:
2088
-
8708
,
D
O
I
:
10.
11
591/
i
j
ece
.
v1
1
i
6
.
pp
55
49
-
555
7
5549
Jou
r
n
al
h
om
e
p
age
:
h
ttp
:
//ije
c
e
.
ia
e
s
c
o
r
e
.
c
o
m
Co
m
pa
ny
ba
n
k
r
u
pt
cy
predi
ct
io
n
f
r
a
m
ew
o
rk
ba
s
ed o
n t
he
m
o
s
t
inf
luen
t
ia
l f
ea
t
u
r
es
us
ing
XG
B
o
o
s
t
a
nd s
t
a
c
k
ing
e
ns
e
m
bl
e
lea
rning
M
uc
h
A
z
i
z
M
us
l
i
m
1
,
Y
os
z
a D
as
r
i
l
2
1
F
a
c
ul
t
y
of
T
e
c
hnol
og
y
M
a
na
ge
m
e
nt
a
nd B
us
i
ne
s
s
,
U
ni
v
e
r
s
i
t
i
T
u
n H
us
s
e
i
n O
nn
M
a
l
a
y
s
i
a
,
M
a
l
a
y
s
i
a
1
D
ep
ar
t
m
en
t
o
f
C
o
m
p
u
t
er
S
ci
en
ce,
U
n
i
v
er
s
i
t
as
N
eg
er
i
S
e
m
ar
an
g
,
I
n
d
o
n
es
i
a
2
F
a
c
ul
t
y
of
T
e
c
hnol
og
y
M
a
na
ge
m
e
nt
a
nd B
us
i
ne
s
s
,
U
ni
v
e
r
s
i
t
i
T
u
n H
us
s
e
i
n O
nn
M
a
l
a
y
s
i
a
,
B
a
t
u
P
a
ha
t
,
M
a
l
a
y
s
i
a
A
r
t
ic
l
e
In
f
o
AB
S
T
RAC
T
A
r
tic
le
h
is
to
r
y
:
R
ecei
v
ed
N
ov
16,
2020
Re
v
i
se
d
Ap
r
8
,
20
2
1
A
ccep
t
ed
M
a
y
11
, 2
0
2
1
C
om
pa
n
y
ba
nk
r
upt
c
y
i
s
of
t
e
n a
v
e
r
y
bi
g
pr
o
bl
e
m
f
or
c
om
pa
ni
e
s
.
T
he
i
m
pa
c
t
of
ba
nk
r
upt
c
y
c
a
n c
a
us
e
l
os
s
e
s
t
o e
l
e
m
e
nt
s
of
t
he
c
o
m
pa
n
y
s
uc
h
a
s
o
w
ne
r
s
,
i
nv
e
s
t
or
s
,
e
m
pl
oy
e
e
s
,
a
nd c
o
ns
u
m
e
r
s
.
O
ne
w
a
y
t
o pr
e
v
e
nt
ba
nk
r
upt
c
y
i
s
t
o
pr
e
di
c
t
t
he
pos
s
i
bi
l
i
t
y
of
ba
nk
r
upt
c
y
ba
s
e
d on t
he
c
om
pa
ny
's
f
i
n
a
nc
i
a
l
da
t
a
.
T
he
r
e
f
or
e
,
t
hi
s
s
t
u
dy
a
i
m
s
t
o f
i
nd t
he
be
s
t
pr
e
di
c
t
i
v
e
m
ode
l
or
m
e
t
hod t
o
pr
e
di
c
t
c
om
pa
n
y
ba
nk
r
upt
c
y
us
i
ng
t
h
e d
at
as
et
f
r
o
m
P
o
l
i
s
h
co
m
p
an
i
es
b
an
k
r
u
p
t
cy
.
T
h
e p
r
ed
i
ct
i
o
n
an
al
y
s
i
s
p
r
o
ces
s
u
s
es
t
h
e b
es
t
f
eat
u
r
e s
el
ect
i
o
n
an
d
en
s
em
b
l
e l
ear
n
i
n
g
.
T
h
e b
es
t
f
e
at
u
r
e s
el
ect
i
o
n
i
s
s
el
ect
ed
u
s
i
n
g
f
e
at
u
r
e
i
m
por
t
a
nc
e
t
o X
G
B
oos
t
w
i
t
h a
w
e
i
g
ht
v
a
l
ue
f
i
l
t
e
r
o
f
10.
T
he
e
ns
e
m
bl
e
l
e
a
r
ni
ng
m
e
t
hod us
e
d i
s
s
t
a
c
k
i
ng.
S
t
a
c
k
i
ng
i
s
c
o
m
pos
e
d of
t
he
b
a
s
e
m
ode
l
an
d
m
et
a l
ear
n
er
.
T
h
e b
as
e
m
o
d
el
co
n
s
i
s
t
s
o
f
K
-
n
ear
es
t
n
ei
g
h
b
o
r
,
d
eci
s
i
o
n
t
r
ee
,
s
uppor
t
v
e
c
t
or
m
a
c
hi
ne
s
(
SV
M
)
,
a
nd
r
a
n
dom
f
or
e
s
t
,
w
h
ile
th
e
m
e
ta
l
ear
n
er
u
s
ed
i
s
L
i
g
h
t
G
B
M
.
T
h
e s
t
ack
i
n
g
m
o
d
el
accu
r
ac
y
r
es
u
l
t
s
ca
n
o
u
t
p
er
f
o
r
m
t
h
e b
as
e
m
o
d
el
accu
r
ac
y
w
i
t
h
an
accu
r
ac
y
r
at
e o
f
9
7
%
.
Ke
y
wo
rd
s
:
B
a
nc
kr
up
t
c
y p
r
e
d
i
c
t
i
o
n
E
n
s
e
m
b
l
e l
ear
n
i
n
g
F
eat
u
r
e i
m
p
o
r
t
a
n
ce
S
t
ack
i
n
g
XGB
o
o
s
t
T
hi
s
i
s
an
ope
n ac
c
e
s
s
ar
t
i
c
l
e
u
nd
e
r
t
he
CC B
Y
-
SA
l
i
cen
s
e.
Co
rre
sp
o
n
d
i
n
g
Au
t
h
o
r
:
M
u
ch
A
z
i
z M
u
s
l
i
m
Fa
c
ul
t
y o
f
T
e
c
hno
l
o
g
y M
a
na
g
e
m
e
n
t
U
ni
ve
r
s
i
t
i
T
un H
u
s
s
e
i
n O
n
n
M
a
l
a
y
s
i
a
B
at
u
P
ah
at
,
J
o
h
o
r
,
M
al
ay
s
i
a
E
m
a
il:
a2
1
2
m
u
s
l
i
m
@
m
ai
l
.
u
n
n
es
.
ac.
i
d
1.
I
NT
RO
D
UCT
I
O
N
P
r
e
di
c
t
i
n
g
c
o
m
pa
ny
ba
nk
r
u
pt
c
y
i
s
on
e
o
f
t
h
e
m
os
t
i
m
por
t
a
nt
pa
r
t
s
of
m
a
n
a
g
e
m
e
n
t
s
c
i
e
n
c
e
pr
obl
e
m
s
.
T
h
e m
ai
n
p
u
r
p
o
s
e o
f
t
h
i
s
p
r
ed
i
ct
i
o
n
i
s
t
o
cat
e
g
o
r
i
ze co
m
p
an
i
es
t
h
at
ar
e s
a
f
e a
n
d
u
n
s
a
f
e
o
r
b
an
k
r
u
p
t
[
1
]
.
I
n
a
ddi
t
i
on
,
t
h
e
w
r
ong
de
c
i
s
i
on
-
m
a
ki
ng
i
n
f
in
a
n
c
ia
l i
n
s
t
itu
tio
n
s
t
h
a
t a
r
e
in
f
i
n
a
n
c
ia
l d
if
f
ic
u
lt
y
o
r
d
is
tr
e
s
s
i
s
ex
p
er
i
en
ced
b
y
m
an
y
s
o
ci
al
co
s
t
s
s
u
c
h
as
o
w
n
er
s
o
r
s
h
ar
eh
o
l
d
er
s
,
m
a
n
a
g
er
s
,
g
o
v
e
r
n
m
e
n
t
a
n
d
o
t
h
er
s
.
T
h
er
ef
o
r
e,
t
h
e p
r
e
d
i
ct
i
o
n
o
f
co
m
p
an
y
b
a
n
k
r
u
p
t
c
y
h
as
b
eco
m
e a s
p
eci
al
co
n
cer
n
am
o
n
g
i
n
d
u
s
tr
ia
l
p
r
act
i
t
i
o
n
er
s
as
w
el
l
as
acad
e
m
i
cs
o
r
r
es
ear
ch
er
s
[
2
]
-
[
5
].
N
o
w
ad
a
y
s
,
m
ac
h
i
n
e l
ear
n
i
n
g
t
ech
n
i
q
u
es
[
6
]
an
d
ar
t
i
f
i
ci
al
i
n
t
el
l
i
g
e
n
ce [
7
]
co
m
p
u
t
a
t
i
o
n
h
av
e
b
een
w
i
de
l
y
u
s
e
d b
y
r
e
s
e
a
r
c
h
e
r
s
t
o s
ol
v
e
ba
n
k
r
u
pt
c
y
pr
e
di
c
t
i
on
pr
obl
e
m
s
s
u
c
h
a
s
s
u
ppor
t
v
e
c
t
or
m
ac
h
i
n
es
(
S
V
M)
[8
]
-
[
16]
,
de
c
i
s
i
on
t
r
e
e
s
[
17
]
-
[
2
3
],
a
rt
i
f
i
c
i
a
l
n
e
u
ra
l
n
e
t
w
o
rk
s
(A
N
N
) [2
4
]
-
[
3
1
]
a
n
d
d
is
c
u
s
s
io
n
w
it
h
s
y
s
te
m
a
tic
lite
r
a
tu
r
e
r
e
v
ie
w
te
c
h
n
iq
u
e
[
3
2
]
-
[
37
]
.
M
e
a
n
w
hi
l
e
,
i
m
p
r
o
v
e
m
e
n
t
i
n
m
a
c
hi
ne
l
e
a
r
ni
ng t
e
c
hni
q
ue
s
t
hr
o
u
gh
v
ar
i
o
u
s
s
t
r
at
e
g
i
es
h
a
s
al
s
o
b
ee
n
car
r
i
ed
o
u
t
s
u
ch
as
b
o
o
s
t
i
n
g
i
m
p
r
o
v
e
m
e
n
t
b
as
ed
o
n
f
eat
u
r
e s
el
ect
i
o
n
k
n
o
w
n
a
s F
S
-
B
o
o
s
t
i
n
g
i
s
p
r
o
v
en
t
o
h
av
e
g
o
o
d
p
er
f
o
r
m
a
n
ce as
a l
ea
r
n
er
an
d
h
as
h
i
g
h
er
acc
u
r
ac
y
an
d
d
i
v
er
s
i
t
y
b
as
ed
on
t
w
o s
e
l
e
c
t
e
d c
o
m
pa
ny
ba
n
k
r
u
pt
c
y
da
t
a
s
e
t
s
[
38]
.
T
h
e
c
om
bi
n
a
t
i
o
n
of
S
V
M
a
n
d
A
N
N
in
te
g
r
a
te
d
w
it
h
d
r
o
po
ut
,
a
ut
o
-
en
co
d
er
p
r
o
v
ed
t
o
p
r
o
d
u
ce b
et
t
er
accu
r
ac
y
t
h
an
l
o
g
i
s
t
i
c r
e
g
r
es
s
i
o
n
,
g
en
et
i
c al
g
o
r
i
t
h
m
a
n
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
20
88
-
8708
In
t
J
E
l
e
c
&
C
o
m
p
E
n
g
,
V
o
l.
11
, N
o
.
6
,
D
ecem
b
er
2
0
2
1
:
5
549
-
55
57
5550
i
n
du
c
t
i
v
e
l
e
a
r
ni
ng
[
39]
.
A
hy
br
i
d a
p
pr
oa
c
h
ba
s
e
d
on
s
y
nt
he
t
i
c
m
i
n
or
i
t
y
ov
e
r
-
s
a
m
p
l
i
n
g t
e
c
hni
q
ue
k
no
w
n a
s
t
h
e S
M
O
T
E
t
ech
n
i
q
u
e
w
i
t
h
t
he
e
n
s
e
m
b
l
e
l
e
a
r
ni
ng
m
e
t
ho
d
,
i
.
e
.
B
o
o
s
t
i
ng,
B
a
g
gi
ng,
N
a
i
ve
B
a
y
e
s
,
A
N
N
,
R
a
n
do
m
f
or
e
s
t
,
R
ot
a
t
i
on f
or
e
s
t
a
n
d
di
v
e
r
s
e
e
n
s
e
m
bl
e
c
r
e
a
t
i
on
b
y
oppos
i
t
i
on
a
l
r
e
l
a
be
l
i
ng
of
m
e
a
n
i
n
gf
u
l
t
r
a
i
ni
ng
e
xa
m
p
l
e
s
(
D
E
C
O
R
A
T
E
)
ar
e
p
r
o
v
en
t
o
ef
f
i
c
i
e
n
t
l
y
i
m
p
r
o
v
e
p
er
f
o
r
m
an
ce
p
ar
a
m
et
er
s
s
u
c
h
as
accu
r
ac
y
,
A
U
C
,
er
r
o
r
t
y
p
es
1
an
d
2
,
G
-
m
e
a
n
t
h
r
oug
h
t
h
e
c
ol
l
e
c
t
e
d da
t
a
s
e
t
of
S
pa
n
i
s
h
c
om
pa
n
i
e
s
[
40]
.
T
h
e
i
n
t
e
g
r
a
t
i
on a
ppr
oa
c
h
of
S
V
M
pr
opor
t
i
on
s
,
boos
t
i
n
g a
n
d b
a
g
g
i
ng
i
n a
n e
ns
e
m
bl
e
s
t
r
a
t
e
gy
c
a
l
l
e
d B
a
gg
e
d
-
pS
V
M
a
n
d B
oos
t
e
d
-
p
S
V
M
w
h
i
ch
i
s
b
as
ed
o
n
a l
ear
n
i
n
g
p
e
r
s
p
ect
i
v
e
w
i
t
h
l
ab
el
p
r
o
p
o
r
t
i
o
n
s
w
h
er
e
u
n
l
ab
el
ed
l
ear
n
i
n
g
d
at
a ar
e p
r
o
v
i
d
ed
w
i
t
h
d
i
f
f
er
en
t
b
ag
s
an
d
o
n
l
y
g
i
v
e
n
a b
ag
b
as
ed
o
n
t
h
e p
r
o
p
o
r
t
i
o
n
o
f
i
n
s
t
an
ce
s
l
ev
el
w
i
t
h
p
ar
t
i
cu
l
ar
cl
as
s
e
s
.
T
h
i
s
ap
p
r
o
ach
i
s
p
r
o
p
o
s
ed
t
o
o
v
er
co
m
e a
l
ar
g
e
n
u
m
b
er
o
f
i
ns
t
a
nc
e
-
l
ev
el
l
ab
el
ed
le
a
r
n
in
g
d
a
ta
[
41]
.
T
h
e
hy
br
i
d of
S
M
O
T
E
-
ed
i
t
ed
n
ear
es
t
n
ei
g
h
b
o
r
(
SM
O
T
E
-
E
NN)
as
o
v
er
-
s
a
m
p
lin
g
t
e
c
h
ni
qu
e
a
n
d
C
B
oos
t
a
l
g
or
i
t
hm
a
s
c
os
t
-
s
e
n
s
i
t
i
v
e
l
e
a
r
n
i
ng
or
pr
e
di
c
t
i
v
e
m
ode
l
.
T
h
i
s
hy
br
i
d
pr
odu
c
e
s
t
h
e
be
s
t
p
er
f
o
r
m
a
n
ce o
f
e
x
i
s
t
i
n
g
l
ear
n
i
n
g
t
ech
n
i
q
u
es
[
4
2
]
.
R
ed
u
c
i
n
g
t
h
e
u
n
b
al
a
n
ced
cl
as
s
o
f
b
an
k
r
u
p
t
c
y
d
at
a s
et
s
us
i
n
g o
ve
r
-
s
a
m
p
l
i
n
g
o
r
S
M
O
T
E
t
ech
n
i
q
u
e
s
t
h
en
A
N
N
as
a p
r
ed
i
ct
i
v
e
m
o
d
el
.
T
h
i
s
co
n
cep
t
r
es
u
l
t
ed
i
n
s
i
g
n
i
f
i
ca
n
t
p
er
f
o
r
m
a
n
ce t
h
a
n
t
h
e
A
N
N
a
n
d
w
eak
l
ear
n
er
s
t
r
ai
n
ed
i
n
t
h
e
A
U
C
s
ect
i
o
n
[
4
3
]
.
B
o
r
d
e
r
lin
e
s
ynt
he
t
i
c
m
i
no
r
i
t
y
o
ve
r
-
s
a
m
p
l
i
ng t
e
c
h
ni
q
ue
(
B
S
M
)
a
nd
s
t
ac
k
ed
au
t
o
-
e
n
co
d
er
(
S
A
E
)
ba
s
e
d on
t
h
e
S
of
t
-
ma
x
c
l
a
s
s
i
f
i
e
r
a
r
e
pr
opos
e
d
t
o
s
ol
v
e
t
h
e
un
ba
l
a
n
c
e
d
c
l
a
s
s
i
f
i
c
a
t
i
on
of
c
o
m
pa
ny
ba
nk
r
u
pt
c
y
pr
e
di
c
t
i
on
pr
obl
e
m
s
.
T
h
i
s
c
om
bi
n
a
t
i
on
a
ppr
oa
c
h
i
s
c
on
s
i
d
er
ed
m
o
r
e ef
f
i
ci
e
n
t
t
h
a
n
t
h
e co
m
b
i
n
at
i
o
n
o
f
B
S
M
w
i
t
h
m
ac
h
i
n
e l
ear
n
i
n
g
t
e
c
hn
i
q
ue
s
a
nd
m
a
c
hi
ne
l
e
a
r
ni
ng t
e
c
h
ni
q
ue
s
w
i
t
ho
ut
o
ve
r
-
s
a
m
p
l
i
ng
[
44]
.
A
t
t
h
e
s
a
m
e t
i
m
e,
t
h
e p
r
o
ces
s
o
f
r
u
n
n
i
n
g
t
h
e co
m
p
a
n
y
’
s
b
u
s
i
n
es
s
p
r
o
d
u
ces
f
i
n
an
ci
a
l
d
at
a t
h
at
ca
n
b
e
u
s
e
d t
o pr
e
di
c
t
ba
n
k
r
u
pt
c
y
[
45]
.
T
h
e
l
a
t
e
s
t
di
s
c
us
s
i
o
n
r
e
ga
r
di
n
g
ba
nk
r
u
pt
c
y
pr
e
di
c
t
i
on f
oc
us
e
s
on
f
e
a
t
u
r
e
s
el
ect
i
o
n
[
3
3
]
.
C
o
m
p
an
y
f
i
n
a
n
ci
al
d
at
a s
u
c
h
as
s
al
es
,
p
r
o
f
i
t
an
d
as
s
et
d
at
a af
f
ect
t
h
e a
n
al
y
s
i
s
p
r
o
ces
s
o
f
b
a
nkr
up
t
c
y p
r
e
d
i
c
t
i
o
ns
.
T
he
r
e
s
ul
t
i
n
g c
o
m
p
a
n
y
f
i
na
nc
i
a
l
d
at
a h
as
m
a
n
y
f
eat
u
r
e
s
s
o
t
h
at
t
h
e b
es
t
f
eat
u
r
e
an
al
y
s
i
s
p
r
o
ces
s
i
s
n
eed
ed
t
o
i
m
p
r
o
v
e t
h
e q
u
al
i
t
y
o
f
p
r
ed
i
ct
i
o
n
s
.
T
w
o
t
y
p
es
f
eat
u
r
e s
el
ect
i
o
n
b
as
ed
o
n
f
i
l
t
er
a
n
d
w
r
a
ppe
r
w
i
t
h
t
w
o t
y
pe
s
c
l
a
s
s
i
f
i
c
a
t
i
o
n
t
e
c
h
n
i
q
u
e
s
ba
s
e
d
on
ba
gg
i
ng
a
n
d boos
t
i
ng
e
n
s
e
m
bl
e
c
l
a
s
s
i
f
ie
r
to
m
ode
l
pr
e
di
c
t
i
v
e
[
46]
.
So
n
e
t a
l.
[
47]
u
s
e
d S
k
e
w
ne
s
s
r
e
du
c
t
i
on
f
or
da
t
a
n
or
m
a
l
i
z
a
t
i
o
n
a
n
d
X
B
oos
t
a
l
g
or
i
t
hm
to
s
e
le
c
t
f
e
a
tu
r
e
s
i
m
p
o
r
ta
n
t to
s
e
r
v
e
a
s
a
ttr
ib
u
te
s
o
f
b
a
n
k
r
u
p
t
c
y
p
r
e
d
ic
tio
n
s
.
T
h
e
r
e
s
u
lt o
f
S
o
n
e
t a
l.
’
s
m
e
t
h
od
can
i
m
p
r
o
v
e p
r
ed
i
ct
i
o
n
s
w
i
t
h
an
acc
u
r
ac
y
o
f
1
7
%
o
f t
h
e
A
U
C
l
ev
el
.
N
o
b
r
e
[
48]
u
s
e
d t
h
e
X
G
B
oos
t
a
l
g
or
i
t
hm
t
o
f
eat
u
r
e s
el
ect
i
o
n
co
m
b
i
n
e
d
w
i
t
h
p
r
i
nc
i
p
a
l
c
o
m
p
o
ne
nt
a
na
l
ys
i
s
(P
C
A
) a
n
d
d
is
c
r
e
te
w
a
v
e
le
t tr
a
n
s
f
o
r
m
(
D
W
T
)
t
o
an
al
y
ze b
an
k
r
u
p
t
c
y
p
r
ed
i
ct
i
o
n
s
.
T
h
e r
es
u
l
t
s
o
f
t
h
e an
al
y
s
i
s
s
h
o
w
t
h
a
t
t
h
e
m
et
h
o
d
u
s
ed
h
as
a r
et
u
r
n
v
a
l
u
e
o
f
49.
26
%
B
as
ed
o
n
p
r
ev
i
o
u
s
r
es
ear
ch
,
i
n
cr
eas
i
n
g
acc
u
r
ac
y
i
s
t
h
e
m
ai
n
f
o
c
u
s
i
n
p
r
ed
i
ct
i
v
e s
t
u
d
i
es
o
f
co
r
p
o
r
at
e
b
an
k
r
u
p
t
c
y
.
C
o
m
b
i
n
ed
ap
p
r
o
ach
es
o
r
i
m
p
r
o
v
ed
m
et
h
o
d
s
ar
e s
t
i
l
l
v
er
y
m
u
c
h
n
eed
ed
t
o
ach
i
ev
e b
et
t
er
accu
r
ac
y
.
T
h
er
ef
o
r
e,
T
h
i
s
s
t
u
d
y
u
s
es
a
f
eat
u
r
e
an
al
y
s
i
s
ap
p
r
o
ach
t
o
s
el
ect
t
h
e
b
e
s
t
f
eat
u
r
es
,
an
d
co
m
b
i
n
e
s
s
ev
er
al
m
ach
i
n
e l
ear
n
i
n
g
al
g
o
r
i
t
h
m
s
(
s
t
ac
k
i
n
g
e
n
s
e
m
b
l
e)
t
o
i
m
p
r
o
v
e accu
r
ac
y
.
X
G
B
o
o
s
t
f
eat
u
r
e i
m
p
o
r
t
an
ce
i
s
u
s
ed
t
o
s
el
ect
h
i
g
h
l
y
i
n
f
l
u
en
t
i
al
f
eat
u
r
e
s
b
as
ed
o
n
t
h
e
w
ei
g
h
t
v
al
u
e o
f
each
f
eat
u
r
e d
u
r
i
n
g
t
h
e p
r
ed
i
ct
i
o
n
an
al
y
s
i
s
p
r
o
ces
s
[
4
9
]
.
I
n
ad
d
i
t
i
o
n
t
o
s
el
ect
i
n
g
t
h
e b
es
t
f
ea
t
u
r
es
,
t
h
i
s
s
t
u
d
y
al
s
o
co
m
b
i
n
e
s
m
ac
h
i
n
e l
ear
n
i
n
g
m
e
t
h
ods
c
ons
i
s
t
i
ng
o
f
K
-
ne
a
r
e
s
t
ne
i
g
hb
o
r
,
d
e
c
is
io
n
tr
e
e
,
S
V
M
an
d
ra
n
d
o
m
f
o
re
s
t
i
n
t
h
i
s
c
as
e cal
l
e
d
en
s
em
b
l
e
l
e
a
r
ni
n
g
w
i
t
h t
he
s
t
a
c
ki
ng
m
e
t
ho
d
[
3
3
]
.
T
he
p
ur
p
o
s
e
o
f
t
hi
s
s
t
ud
y
w
a
s
t
o
f
i
nd
t
he
hi
ghe
s
t
a
c
c
ur
a
c
y b
y
s
el
ect
i
n
g
t
h
e b
es
t
s
el
ect
i
o
n
f
eat
u
r
e an
d
co
m
b
i
n
i
n
g
s
e
v
er
al
m
ach
i
n
e l
ear
n
i
n
g
m
et
h
o
d
s
u
s
i
n
g
a s
t
ac
k
i
n
g
en
s
e
m
b
l
e.
2.
TH
EO
R
ET
I
C
A
L B
A
C
K
G
R
O
U
N
D
2
.1
.
B
o
o
s
t
i
n
g
t
ree m
et
h
o
d
B
oos
t
i
n
g
i
s
a
s
u
pe
r
i
or
m
e
t
h
o
d
i
n
c
o
m
bi
ni
ng
s
e
v
e
r
a
l
ba
s
i
c
c
l
a
s
s
i
f
i
c
a
t
i
ons
t
o
pr
odu
c
e
a
n
a
l
g
or
i
t
hm
th
a
t is
s
u
p
e
r
io
r
in
a
c
h
ie
v
in
g
a
c
c
u
r
a
c
y
t
h
a
n
o
th
e
r
c
la
s
s
i
f
ic
a
ti
o
n
a
lg
o
r
ith
m
s
.
B
o
o
s
tin
g
is
a
n
a
d
d
itiv
e
e
n
s
e
m
b
le
m
e
t
h
od t
h
a
t
w
or
k
s
b
y
a
ddi
ng
n
e
w
m
o
d
el
s
t
o
r
ed
u
ce er
r
o
r
s
m
ad
e b
y
o
l
d
er
o
r
ex
i
s
t
i
n
g
m
o
d
el
s
.
S
eq
u
en
t
i
al
l
y
,
t
h
e
m
ode
l
s
a
r
e
a
dde
d i
n s
u
c
h
a
w
a
y
t
h
a
t
n
o pos
s
i
bl
e
i
m
pr
ove
m
e
nt
oc
c
u
r
s
.
B
oos
t
e
d
m
ode
l
s
c
a
n pr
odu
c
e
g
ood
accu
r
ac
y
e
v
en
t
h
o
u
g
h
t
h
e b
as
i
c cl
as
s
i
f
i
cat
i
o
n
h
a
s
o
n
l
y
s
l
i
g
h
t
l
y
b
et
t
er
accu
r
ac
y
t
h
an
r
a
n
d
o
m
cl
as
s
i
f
i
cat
i
o
n
,
s
o
t
h
at
t
h
e b
as
i
c cl
as
s
i
f
i
cat
i
o
n
i
s
co
n
s
i
d
er
ed
a
w
eak
l
ear
n
er
[
5
0
]
.
I
n
a s
u
p
er
v
i
s
ed
l
ear
n
i
n
g
s
et
t
i
n
g
,
L
et
d
at
a
-
se
t
D
=
{
(
,
)
:
∈
ℝ
,
∈
ℝ
}
ar
r
an
g
ed
o
f
n
d
at
a
w
i
t
h
m
f
ea
t
u
r
es
a
n
d
n
l
ab
el
s
,
a b
o
o
s
t
i
n
g
t
r
ee
m
ode
l
u
se
s
K
a
d
d
itiv
e
f
u
n
c
tio
n
s
(
)
t
o pr
e
di
c
t
t
h
e
ou
t
pu
t
.
=
(
)
=
∑
=
(
)
w
h
er
e
(
)
=
W
(
)
.
C
le
a
r
ly
,
:
ℝ
⟶
i
n
d
i
cat
es
t
h
e s
t
r
u
ct
u
r
e o
f
each
t
r
ee t
h
at
m
a
ps
a
s
a
m
pl
e
t
o t
h
e
c
or
r
e
s
p
on
di
ng
i
n
de
x
of
l
e
a
f
a
n
d
W
∈
i
a a
w
ei
g
h
t
o
f
l
ea
f
w
i
t
h T
l
e
a
ve
s
.
In
o
rd
e
r t
o
l
e
a
r
n
t
he
f
unc
t
i
o
n
s
e
t
,
w
e
m
i
ni
m
i
z
e
t
he
f
unc
t
i
o
n
o
f
l
o
s
s
(
)
=
∑
=
1
(
,
)
+
∑
=
1
(
)
wh
e
r
e
(
)
=
+
∥
∥
2
is
a
te
r
m
o
f
r
e
g
u
la
r
iz
a
tio
n
t
h
a
t
p
e
n
a
l
iz
e
s
m
o
d
e
l
c
o
m
p
le
x
it
y
.
T
h
e
f
u
n
c
tio
n
o
f
lo
s
s
L
(
g)
c
o
nt
a
i
ns
K
-
f
u
n
c
tio
n
a
s
p
a
r
a
m
e
te
r
s
s
o
it
is
s
o
h
a
r
d
to
o
p
ti
m
iz
e
d
ir
e
c
tl
y
.
I
n
s
te
a
d
,
w
e
o
p
ti
m
iz
e
th
e
a
d
d
itiv
e
l
y
m
o
d
e
l.
G
iv
e
n
b
e
ℎ
s
a
m
p
le
p
r
e
d
ic
tio
n
a
t
ℎ
it
e
r
a
tio
n
.
W
e
w
ill a
d
d
to
m
in
i
m
iz
e
.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t
J
E
l
e
c
&
C
o
m
p
E
n
g
I
S
S
N
:
2088
-
8708
C
om
pany
ba
nk
r
upt
c
y
pr
e
di
c
t
i
on f
r
am
e
w
or
k
bas
e
d on t
he
m
os
t
i
nf
l
ue
nt
i
al
…
(
M
u
c
h
A
z
i
z
M
u
sl
i
m
)
5551
(
)
=
∑
=
1
(
,
−
1
+
(
)
)
+
(
)
W
h
i
ch
m
ea
n
s
t
h
at
w
e g
r
ee
d
i
l
y
ad
d
t
h
e
t
h
at
m
o
s
t
i
m
p
r
o
v
e o
u
r
m
o
d
el
f
o
r
each
i
t
er
at
i
o
n
.
W
e u
s
e
a
ppr
ox
i
m
a
t
i
on
o
f
s
e
c
o
n
d
-
o
r
d
e
r
th
a
t
u
s
e
s
a
g
r
a
d
ie
n
t o
n
th
i
s
in
te
r
m
e
d
ia
te
f
u
n
c
tio
n
o
f
lo
s
s
(
)
.
T
hi
s
i
s
t
he
r
e
a
s
o
n
w
e
n
a
m
e
it g
r
a
d
ie
n
t b
o
o
s
tin
g
a
l
g
o
r
ith
m
a
s
s
ho
w
n
i
n a
lg
o
r
it
h
m
1
in
F
ig
u
r
e
1
.
T
h
e
X
g
boos
t
[
49]
i
s
a
n
ope
n
-
s
o
ur
c
e
l
i
b
r
a
r
y
o
f
s
o
f
t
w
a
r
e
t
ha
t
gi
ve
s
f
r
a
m
e
w
o
r
k o
f
gr
a
d
i
e
nt
b
o
o
s
t
i
n
g f
o
r
C
+
+
,
J
a
va
,
P
y
t
ho
n,
m
a
t
h
-
la
b
an
d
R
.
I
t
u
s
es
a
g
r
ad
i
en
t
-
b
o
o
s
tin
g
a
lg
o
r
it
h
m
t
h
a
t r
e
s
u
lt
s
in
a
p
r
e
d
ic
tio
n
m
o
d
e
l in
t
h
e
f
o
r
m
o
f
an
en
s
e
m
b
l
e o
f
w
ea
k
p
r
ed
i
ct
i
o
n
m
o
d
el
s
,
w
h
i
c
h
ar
e d
eci
s
i
o
n
t
r
ees
,
t
y
p
i
cal
l
y
.
Fi
g
u
r
e 1
.
G
r
ad
i
en
t
boos
t
i
n
g
a
l
g
or
i
t
hm
2
.2
.
St
a
c
k
i
ng
e
ns
e
m
b
l
e
m
o
de
l
i
ng
T
h
e s
t
ack
i
n
g
en
s
e
m
b
l
e i
n
t
r
o
d
u
ced
b
y
W
o
l
p
er
t
[
5
1
]
t
h
en
f
o
r
m
al
i
zed
b
y
B
r
ei
m
e
n
[
5
2
]
an
d
th
e
o
r
e
tic
a
ll
y
v
al
i
d
at
ed
b
y
V
a
n
d
er
L
aa
n
e
t a
l.
[
53]
i
s
on
e
of
t
h
e
l
e
a
r
n
i
n
g
a
l
g
or
i
t
hm
s
kn
o
w
n
a
s
a
s
u
pe
r
i
or
l
ear
n
i
n
g
f
r
a
m
e
w
o
r
k
b
as
ed
o
n
g
en
er
al
i
z
i
n
g
l
o
s
s
e
s
.
D
u
e t
o
i
t
s
s
u
p
er
i
o
r
p
er
f
o
r
m
a
n
ce co
m
p
ar
ed
t
o
o
t
h
er
l
ear
n
i
n
g
a
l
g
or
i
t
hm
s
,
S
t
a
c
k
i
n
g
e
n
s
e
m
bl
e
h
a
s
m
a
ny
a
ppl
i
c
a
t
i
ons
f
or
pr
e
di
c
t
i
n
g
c
o
m
pa
ny
ba
n
k
r
u
pt
c
y
.
A
s
d
es
cr
i
b
ed
i
n
a
l
g
or
i
t
hm
2 i
n
F
i
gu
r
e
2.
T
h
e
r
e
f
or
e
,
t
o i
m
pr
ov
e
t
h
e
pr
e
di
c
t
i
on
a
c
c
u
r
a
c
y
,
t
h
e
s
t
a
c
k
i
ng
e
ns
e
m
bl
e
i
s
pr
opos
e
d i
n
t
h
i
s
s
t
u
d
y
t
o be
c
om
bi
n
e
d
w
i
t
h
t
h
e
X
G
boos
t
a
l
g
or
i
t
hm
.
F
i
gu
r
e
2.
S
t
a
c
ki
ng e
ns
e
m
b
l
e
m
o
d
e
l
i
n
g
a
lg
o
r
it
h
m
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
20
88
-
8708
In
t
J
E
l
e
c
&
C
o
m
p
E
n
g
,
V
o
l.
11
, N
o
.
6
,
D
ecem
b
er
2
0
2
1
:
5
549
-
55
57
5552
3.
R
ES
EA
R
C
H
M
ETH
O
D
T
h
e r
es
ear
ch
m
et
h
o
d
o
f
b
an
k
r
u
p
t
c
y
p
r
ed
i
ct
i
o
n
a
n
al
y
s
i
s
u
s
es
s
ev
er
al
s
t
ag
e
s
,
i
.
e.
d
at
a co
l
l
ect
i
o
n
,
p
r
e
-
p
r
o
ces
s
i
n
g
d
at
a,
f
eat
u
r
e
i
m
por
t
a
n
c
e
a
n
d
m
ode
l
i
ng
.
G
e
n
e
r
a
l
l
y
,
t
he
r
e
s
e
a
r
c
h
f
r
a
m
e
w
or
k
c
a
n
be
s
h
o
w
n i
n
F
i
gur
e
3.
T
h
e d
at
a
-
s
e
t in
t
h
is
s
tu
d
y
w
a
s
ta
k
e
n
p
u
b
lic
l
y
f
r
o
m
K
a
g
g
l
e.
T
h
e d
at
a
-
s
e
t is
h
is
to
r
ic
a
l d
a
ta
o
n
b
a
n
k
r
u
p
tc
y
f
r
o
m
P
o
l
i
s
h
co
m
p
a
n
i
es
a
n
d
h
as
a
r
an
g
e
o
f
y
ear
s
t
h
at
ar
e l
i
s
t
ed
i
n
t
h
e d
at
a
-
s
e
t
s
t
a
r
t
i
ng
f
r
o
m
2000 t
o 2012 [
54]
.
T
h
e
d
a
ta
-
s
e
t
i
s
c
o
m
pos
e
d of
65 f
e
a
t
u
r
e
s
r
e
l
a
t
e
d t
o t
h
e
c
o
m
pa
ny
’
s
bu
s
i
n
e
s
s
c
on
t
i
nu
i
t
y
p
r
o
ces
s
.
T
h
e t
o
t
al
d
at
a r
o
w
s
i
n t
he
d
a
t
a
-
s
e
t
a
r
e
42,
627 r
o
w
s
.
T
h
e
t
a
r
g
e
t
da
t
a
-
s
e
t
f
e
a
t
u
r
e
is
in
th
e
"
c
la
s
s
"
c
o
lu
m
n
w
it
h
d
e
ta
ile
d
c
o
n
te
n
ts
,
n
a
m
e
l
y
0
an
d
1
.
V
ar
i
ab
l
e d
at
a
0
m
ea
n
s
t
h
at
i
t
i
s
n
o
t
b
an
k
r
u
p
t
an
d
v
i
ce
v
er
s
a
i
n
d
at
a
v
ar
i
ab
l
e
1
i
n
d
i
cat
es
b
a
nkr
up
t
c
y
.
D
at
a p
r
e
-
pr
oc
e
s
s
i
n
g
m
e
a
n
s
n
or
m
a
l
i
z
i
n
g
da
t
a
s
e
t
s
t
h
a
t
do n
ot
s
u
ppor
t
t
h
e
a
n
a
l
y
s
i
s
pr
oc
e
s
s
[
47]
,
[
5
5
]
.
D
at
a t
h
at
d
o
n
o
t
s
u
p
p
o
r
t
t
h
e an
al
y
s
i
s
p
r
o
ces
s
ar
e r
e
p
et
i
t
i
v
e d
at
a,
b
l
an
k
d
at
a an
d
ab
n
o
r
m
al
d
at
a.
F
eat
u
r
es
t
h
at
ar
e n
o
t
r
el
at
ed
t
o
t
h
e
an
al
y
s
i
s
p
r
o
ces
s
i
n
t
h
e d
at
a s
et
w
i
l
l
b
e n
o
r
m
al
i
zed
[
5
6
]
,
[
5
7
]
.
T
h
e d
at
a
-
p
r
ep
r
o
ces
s
i
n
g
m
et
h
o
d
i
n
t
h
i
s
s
t
u
d
y
i
s
d
at
a s
cal
i
n
g
.
D
a
t
a
s
cal
i
n
g
i
s
a
m
et
h
o
d
o
f
s
i
m
p
l
i
f
y
i
n
g
t
h
e
r
an
g
e o
f
n
u
m
er
i
c d
at
a
v
al
u
es
i
n
a d
at
a
-
s
et
t
h
at
h
a
s
t
h
e s
a
m
e
v
al
u
e
[
5
8
]
.
D
at
a s
cal
i
n
g
cr
eat
es
a
b
al
an
ced
r
an
g
e o
f
n
u
m
er
i
c d
at
a.
I
m
p
o
r
t
an
ce
f
eat
u
r
es
ar
e s
el
ect
ed
b
as
ed
o
n
t
h
e
cal
cu
l
at
i
o
n
o
f
t
h
e X
G
B
o
o
s
t
a
l
g
o
r
i
t
h
m
[
4
8
]
.
T
h
e
m
et
h
o
d
o
f
d
et
er
m
i
n
i
n
g
t
h
e
v
a
l
u
e o
f
t
h
e
f
eat
u
r
e
w
ei
g
h
t
i
s
ca
l
cu
l
at
ed
b
as
ed
o
n
t
h
e e
f
f
ect
o
f
t
h
e
f
eat
u
r
e o
n
t
h
e
r
es
u
l
t
s
o
f
p
r
ed
i
ct
i
v
e
an
al
y
s
i
s
.
T
h
e
f
in
a
l
r
e
s
u
lt
o
f
d
e
te
r
m
in
i
n
g
t
h
e
b
e
s
t
f
e
a
t
u
r
e
s
is
a
p
p
lie
d
to
th
e
d
a
ta
-
s
e
t
to
i
m
p
r
o
v
e t
h
e r
es
u
l
t
s
o
f
p
r
ed
i
ct
i
o
n
accu
r
ac
y
.
F
i
g
u
r
e 3
.
T
h
e r
es
ear
ch
f
r
a
m
e
w
o
r
k
T
h
e m
o
d
el
i
n
g
p
r
o
ces
s
u
s
es
s
t
ack
i
n
g
e
n
s
e
m
b
l
e l
ear
n
i
n
g
,
w
h
i
ch
i
s
t
h
e p
r
o
ces
s
o
f
co
m
b
i
n
i
n
g
s
ev
er
al
m
a
c
h
i
ne
l
e
a
r
ni
ng
a
l
go
r
i
t
h
m
s
s
uc
h
a
s
K
-
n
ear
es
t
n
ei
g
h
b
o
r
,
d
eci
s
i
o
n
t
r
ee,
g
r
ad
i
en
t
b
o
o
s
t
i
n
g
t
r
ee
an
d
r
an
d
o
m
fo
r
e
s
t
[
5
9
]
.
E
ns
e
m
b
l
e
s
t
a
c
ki
n
g i
s
o
ne
o
f
t
he
e
ns
e
m
b
l
e
l
e
a
r
n
i
ng
m
e
t
ho
d
s
a
nd
c
a
n u
s
e
he
t
e
r
o
ge
ne
o
u
s
m
a
c
hi
ne
le
a
r
n
in
g
m
e
t
h
o
d
s
.
S
ta
c
k
in
g
e
n
s
e
m
b
le
le
ar
n
i
n
g
u
s
es
m
et
a
-
l
e
a
r
n
i
ng
a
l
g
or
i
t
hm
s
t
o
f
i
n
d t
he
be
s
t
r
e
s
u
l
t
s
f
or
c
om
bi
n
i
n
g pr
e
di
c
t
i
on
s
f
r
o
m
t
w
o or
m
or
e
ba
s
i
c
m
a
c
h
i
n
e
l
e
a
r
n
i
ng
a
l
g
or
i
t
hm
s
.
T
h
e
s
t
a
c
k
i
n
g e
ns
e
m
bl
e
h
a
s
t
h
e
ad
v
an
t
a
g
e o
f
b
ei
n
g
ab
l
e t
o
t
ak
e ad
v
a
n
t
a
g
e o
f
t
h
e
w
o
r
k
p
r
o
ces
s
es
o
f
s
e
v
er
al
m
ac
h
i
n
e
l
ea
r
ni
n
g a
l
go
r
i
t
h
m
m
o
d
e
l
s
t
h
a
t
f
u
n
c
tio
n
w
e
ll i
n
c
la
s
s
i
f
ic
a
tio
n
o
r
r
e
g
r
e
s
s
io
n
ta
s
k
s
a
n
d
m
a
k
e
p
r
e
d
ic
tio
n
s
b
e
t
te
r
th
a
n
t
h
e
w
o
r
k
p
r
o
ces
s
o
f
o
n
e
m
ac
h
i
n
e l
ear
n
i
n
g
m
o
d
el
i
n
e
n
s
e
m
b
l
e l
ear
n
i
n
g
.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t
J
E
l
e
c
&
C
o
m
p
E
n
g
I
S
S
N
:
2088
-
8708
C
om
pany
ba
nk
r
upt
c
y
pr
e
di
c
t
i
on f
r
am
e
w
or
k
bas
e
d on t
he
m
os
t
i
nf
l
ue
nt
i
al
…
(
M
u
c
h
A
z
i
z
M
u
sl
i
m
)
5553
4.
RE
S
U
L
T
S
AND D
I
S
CU
S
S
I
O
N
T
h
e
an
al
y
s
i
s
p
r
o
ces
s
u
s
es
t
h
e
g
o
o
g
l
e co
l
l
ab
t
o
o
l
w
i
t
h t
he
p
y
t
ho
n
p
r
o
gr
a
m
m
i
ng
l
a
n
g
ua
g
e
a
nd
t
h
e
he
l
p
o
f
t
he
s
c
i
ki
t
-
l
e
a
r
n
,
pa
n
da
s
,
num
p
y
l
i
br
a
r
i
e
s
a
n
d ot
h
e
r
s
u
ppor
t
i
n
g l
i
br
a
r
i
e
s
.
T
h
e
da
t
a
-
s
et
u
s
ed
co
m
e
s
f
r
o
m
K
a
g
g
le
w
ith
a
d
e
ta
ile
d
d
a
ta
-
s
e
t
c
o
n
s
is
tin
g
o
f
5
C
S
V
fi
le
s
w
h
ic
h
a
r
e
c
o
m
b
in
e
d
in
to
o
n
e
to
f
a
c
ilita
te
th
e
p
r
ed
i
ct
i
o
n
an
al
y
s
i
s
p
r
o
ces
s
.
T
he
p
r
e
-
p
r
o
ces
s
i
n
g
d
at
a s
t
ag
e i
s
s
cal
i
n
g
t
h
e d
at
a o
n
t
h
e
d
at
a
-
s
e
t
u
s
i
n
g t
he
s
t
an
d
ar
d
s
cal
er
py
t
h
on
l
i
b
r
ar
y
.
T
h
e d
at
a s
cal
i
n
g
p
r
o
ces
s
w
a
s
ap
p
l
i
ed
t
o
each
o
f
t
h
e n
u
m
e
r
i
cal
d
at
a co
n
t
ai
n
ed
in
th
e
d
a
ta
-
s
e
t.
D
a
ta
tr
a
n
s
f
o
r
m
a
tio
n
i
s
on
l
y
pe
r
f
or
m
e
d
on
f
e
a
t
u
r
e
s
u
s
e
d
f
or
t
h
e
pr
e
di
c
t
i
on
pr
oc
e
s
s
.
T
h
i
s
i
s
b
ecau
s
e t
h
e t
ar
g
et
f
eat
u
r
e
da
t
a
a
r
e
bi
n
a
r
y
,
n
a
m
e
l
y
0 a
n
d 1,
s
o t
h
e
r
e
i
s
n
o n
e
e
d f
or
t
r
a
n
s
f
or
m
a
t
i
on
t
h
r
oug
h
da
t
a
s
c
a
l
i
n
g.
T
he
r
e
s
ul
t
s
o
f
t
he
s
cal
i
n
g
d
at
a ar
e t
h
e
n
a
n
al
y
zed
at
t
h
e f
eat
u
r
e i
m
p
o
r
t
a
nc
e
s
t
a
ge
.
4
.1
.
F
ea
t
u
re
i
m
po
r
t
a
nc
e
T
h
e f
eat
u
r
e i
m
p
o
r
t
a
n
ce s
t
a
g
e
i
s
t
h
e p
r
o
ces
s
o
f
s
el
ect
i
n
g
t
h
e
b
es
t
f
eat
u
r
es
f
r
o
m
t
h
e r
es
ear
ch
d
at
as
et
.
T
h
e p
r
o
ces
s
o
f
d
et
er
m
i
n
i
n
g
t
h
e b
es
t
f
eat
u
r
e
s
u
s
es
a
n
al
g
o
r
i
t
h
m
o
f
f
eat
u
r
e i
m
p
o
r
t
an
ce
f
r
o
m
t
h
e X
G
B
o
o
s
t
m
ach
i
n
e l
e
a
r
n
i
ng
m
e
t
h
od.
T
h
e
i
m
por
t
a
n
t
f
eat
u
r
es
ar
e
s
el
e
ct
ed
b
as
ed
o
n
t
h
e
w
ei
g
h
t
v
al
u
e o
f
eac
h
f
eat
u
r
e
g
en
er
at
ed
d
u
r
i
n
g
t
h
e p
r
ed
i
ct
i
o
n
an
al
y
s
i
s
p
r
o
ces
s
.
t
h
e b
es
t
f
eat
u
r
e i
s
s
el
ect
ed
b
as
ed
o
n
t
h
e
f
eat
u
r
e
w
ei
g
h
t
t
h
a
t
i
s
m
o
r
e t
h
a
n
1
0
.
D
et
ai
l
s
o
f
t
h
e
b
es
t
f
eat
u
r
e s
e
l
ect
i
o
n r
e
s
ul
t
s
a
r
e
s
ho
w
n i
n T
a
b
l
e
1
.
T
a
b
le
1.
F
itu
r
i
m
p
o
r
t
an
ce
F
ea
t
u
r
e
W
ei
g
h
t
A
ttr
ib
u
te
2
7
1
1
1
A
ttr
ib
u
te
3
4
8
1
A
ttr
ib
u
te
5
5
5
A
ttr
ib
u
te
4
6
4
1
A
ttr
ib
u
te
2
1
3
6
A
ttr
ib
u
te
3
5
2
5
A
ttr
ib
u
te
6
2
5
A
ttr
ib
u
te
5
8
2
3
A
ttr
ib
u
te
2
4
2
3
A
ttr
ib
u
te
5
6
2
0
A
ttr
ib
u
te
1
3
2
0
A
ttr
ib
u
te
4
1
1
7
A
ttr
ib
u
te
3
9
1
6
A
ttr
ib
u
te
2
2
1
6
A
ttr
ib
u
te
2
9
1
5
A
ttr
ib
u
te
4
7
1
4
A
ttr
ib
u
te
3
8
1
4
A
ttr
ib
u
te
2
6
1
3
A
ttr
ib
u
te
4
4
1
0
A
ttr
ib
u
te
3
0
1
0
T
h
e m
o
d
el
i
n
g
s
t
ag
e i
s
i
n
t
h
e
f
o
r
m
o
f
a
n
o
r
m
al
i
zed
b
an
k
r
u
p
t
cy
p
r
ed
i
ct
i
o
n
a
n
al
y
s
i
s
p
r
o
ces
s
t
h
r
ou
gh
t
h
e d
at
a p
r
e
-
p
r
o
ces
s
i
n
g
s
t
ag
e.
A
t
t
h
i
s
s
t
a
g
e t
h
e d
at
as
e
t
i
s
a
n
al
y
zed
u
s
i
n
g
v
ar
i
o
u
s
m
ach
i
n
e l
ear
n
i
n
g
m
e
t
h
o
d
s
.
T
h
e
p
r
e
d
ic
tio
n
a
n
a
l
y
s
is
p
r
o
c
e
s
s
b
e
g
in
s
b
y
d
iv
id
in
g
th
e
d
a
ta
in
to
tr
a
in
i
n
g
d
a
ta
a
n
d
te
s
t d
a
ta
w
it
h
a
7
5
:2
5
r
a
tio
.
T
h
e d
at
a s
h
ar
i
n
g
p
r
o
ces
s
w
a
s
s
t
r
at
i
f
i
ed
an
d
r
ep
eat
ed
.
S
t
r
at
i
f
i
ed
i
s
a d
at
a s
h
ar
i
n
g
m
et
h
o
d
b
as
ed
o
n
t
h
e
w
ei
g
h
t
r
at
i
o
r
at
i
o
o
f
t
h
e
f
eat
u
r
es
f
o
r
w
h
i
c
h
t
h
e
s
el
ect
ed
cat
e
g
o
r
y
.
I
n
t
h
i
s
cas
e
t
h
e
cat
eg
o
r
y
f
ea
t
u
r
e
s
el
ec
t
ed
i
s
t
h
e
t
ar
g
et
f
eat
u
r
e.
R
ep
eat
ed
i
s
a m
e
t
h
o
d
i
n
w
h
i
ch
t
h
e d
at
a s
h
ar
i
n
g
p
r
o
ces
s
i
s
r
ep
eat
ed
acco
r
d
i
n
g
t
o
t
h
e
p
ar
am
et
er
s
.
T
h
e l
o
o
p
i
n
g
p
r
o
ces
s
i
s
ad
d
ed
w
i
t
h
d
at
a
s
h
u
f
f
l
e,
r
es
u
l
t
i
n
g
i
n
d
i
f
f
er
en
t
d
at
a
f
o
r
each
i
t
er
at
i
o
n
.
Cr
o
s
s
-
v
a
l
id
a
tio
n
is
i
n
c
l
u
d
e
d
in
th
is
p
r
o
c
e
s
s
to
a
v
o
id
o
v
e
r
f
it a
n
d
u
n
d
e
r
f
it to
m
a
x
i
m
iz
e
th
e
q
u
a
lit
y
o
f
p
r
e
d
ic
t
i
ve
an
al
y
s
i
s
.
O
v
er
f
i
t
i
s
a
m
o
d
el
t
h
at
i
s
h
i
g
h
l
y
d
ep
en
d
e
n
t
o
n
t
h
e d
at
as
et
an
d
h
as
a h
i
g
h
er
r
o
r
v
al
u
e o
n
t
h
e t
e
s
t
i
n
g
d
at
a.
U
n
d
er
f
i
t
i
s
a
m
o
d
el
t
h
at
can
n
o
t
f
u
l
l
y
u
n
d
er
s
t
an
d
t
h
e d
at
as
et
b
ei
n
g
an
al
y
zed
.
4
.2
.
St
a
c
k
i
ng
T
he
m
a
c
hi
ne
l
e
a
r
ni
n
g
m
e
t
ho
d
us
e
d
i
n
t
he
m
o
d
e
l
i
ng
s
t
a
ge
o
f
t
hi
s
s
t
ud
y
i
s
s
t
a
c
ki
n
g
e
ns
e
m
b
l
e
l
e
a
r
n
i
ng
.
S
t
a
c
k
i
ng
m
e
a
ns
s
t
a
c
k
i
ng
,
w
hi
c
h
m
e
a
ns
pi
l
i
ng
u
p t
h
e
w
or
k
pr
oc
e
s
s
of
m
a
c
h
i
n
e
l
e
a
r
n
i
ng
m
e
t
h
ods
t
o
p
r
o
d
u
ce b
et
t
er
p
r
ed
i
ct
i
v
e r
es
u
l
t
s
.
M
ac
h
i
n
e l
ear
n
i
n
g
m
et
h
o
d
s
t
h
at
can
b
e u
s
ed
i
n
s
t
ack
i
n
g
ca
n
b
e s
el
ec
te
d
he
t
e
r
o
ge
ne
o
us
l
y.
T
he
t
yp
e
o
f
s
t
a
c
ki
n
g e
ns
e
m
b
l
e
m
e
t
ho
d
u
s
e
d
i
n t
hi
s
s
t
ud
y i
s
t
he
c
l
a
s
s
i
f
i
c
a
t
i
o
n o
f
b
a
n
kr
up
t
c
y
p
r
e
d
ic
tio
n
s
.
T
he
s
t
a
c
ke
d
m
a
c
hi
ne
l
e
a
r
ni
ng a
l
go
r
i
t
h
m
s
i
n t
hi
s
s
t
ud
y a
r
e
K
-
n
ear
es
t
n
ei
g
h
b
o
r
,
d
eci
s
i
o
n
t
r
ee,
g
r
a
di
e
n
t
boos
t
i
ng
t
r
e
e
a
n
d r
a
n
do
m
f
or
e
st
i
n
t
h
i
s
cas
e cal
l
e
d
t
h
e b
as
e
m
o
d
el
.
T
h
e b
as
e m
o
d
el
ca
n
co
n
s
i
s
t
o
f
m
an
y
al
g
o
r
i
t
h
m
s
,
b
u
t
t
h
e
m
o
r
e al
g
o
r
i
t
h
m
s
ar
e
u
s
ed
,
t
h
e
m
o
r
e r
es
o
u
r
ces
an
d
t
i
m
e i
t
u
s
es
.
A
l
g
o
r
i
t
h
m
s
i
n
t
h
e
ba
s
e
m
ode
l
a
r
e
n
ot
l
i
m
i
t
e
d t
o j
u
s
t
on
e
m
ode
l
,
t
h
e
y
c
a
n
a
l
s
o be
u
s
e
d f
r
o
m
m
a
ny
v
a
r
ia
tio
n
s
o
f
th
e
m
o
d
e
l
acco
r
d
i
n
g
t
o
r
es
ear
ch
n
eed
s
.
T
h
i
s
r
es
ear
c
h
p
r
o
ces
s
u
s
es
t
h
e c
l
as
s
i
f
i
cat
i
o
n
m
et
h
o
d
s
o
t
h
at
t
h
e al
g
o
r
i
t
h
m
u
s
ed
i
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
20
88
-
8708
In
t
J
E
l
e
c
&
C
o
m
p
E
n
g
,
V
o
l.
11
, N
o
.
6
,
D
ecem
b
er
2
0
2
1
:
5
549
-
55
57
5554
a t
y
p
e o
f
cl
as
s
i
f
i
cat
i
o
n
.
T
h
e r
e
s
u
l
t
o
f
t
h
e b
a
s
e
m
o
d
el
b
u
i
l
d
u
p
i
s
cal
c
u
l
at
ed
b
y
t
h
e
m
et
a l
ear
n
er
.
M
et
a l
ear
n
er
i
s
a m
ac
h
i
n
e l
ear
n
i
n
g
a
l
g
o
r
i
t
h
m
t
h
at
i
s
u
s
ed
t
o
an
al
y
ze
an
d
co
m
b
i
n
e
t
h
e r
es
u
l
t
s
o
f
each
b
as
e
m
o
d
el
i
n
o
r
d
er
t
o
o
b
t
ai
n
a b
et
t
er
p
r
ed
i
ct
i
o
n
r
at
e f
r
o
m
t
h
e b
as
e
m
o
d
el
.
T
h
e
m
et
a l
ear
n
er
u
s
ed
i
n
t
h
i
s
s
t
u
d
y
i
s
L
i
g
h
t
G
B
M
.
T
h
e
f
i
n
a
l r
e
s
u
lt
o
f
th
e
s
ta
c
k
i
n
g
m
o
d
e
l is
a
p
r
e
d
ic
tio
n
g
e
n
er
at
e
d
b
y
t
h
e
m
et
a
l
ear
n
er
.
T
h
e
a
ccu
r
ac
y
d
et
ai
l
s
ar
e
s
ho
w
n
i
n
F
i
g
ur
e
4
.
I
n
F
i
g
u
r
e
4
,
i
t
i
s
s
h
o
w
n
t
h
at
t
h
e d
i
f
f
er
e
n
ce b
et
w
ee
n
t
h
e
m
o
d
el
s
v
ar
i
e
s
.
T
h
e l
o
w
e
s
t
l
e
v
el
o
f
accu
r
ac
y
i
s
o
b
t
ai
n
ed
i
n
t
h
e
d
eci
s
i
o
n
t
r
ee al
g
o
r
i
t
h
m
w
i
t
h on
l
y
94.
8
%
.
T
he
hi
ghe
s
t
l
e
v
e
l
o
f
a
c
c
ur
a
c
y i
s
obt
a
i
n
e
d by
t
h
e
s
t
a
c
k
i
n
g
m
ode
l
a
l
g
or
i
t
hm
97
%
.
F
i
gu
r
e
4.
D
e
ta
il
accu
r
ac
y
5.
CO
NCL
U
S
I
O
N
I
n t
hi
s
s
t
ud
y,
a
ne
w
m
e
t
ho
d
h
a
s
b
e
e
n us
e
d
t
o
a
na
l
yz
e
b
a
nkr
up
t
c
y
p
r
e
d
i
c
t
i
o
ns
us
i
ng t
he
b
e
s
t
f
e
a
t
ur
e
s
el
ect
i
o
n
an
d
e
n
s
e
m
b
l
e l
ear
n
i
n
g
.
T
h
e p
r
o
ces
s
o
f
s
el
ect
i
n
g
t
h
e b
es
t
f
eat
u
r
e
s
u
s
es
X
G
B
o
o
s
t
'
s
i
m
p
o
r
t
a
n
t
f
eat
u
r
es
an
d
t
h
e s
t
ac
k
i
n
g
m
et
h
o
d
.
T
h
e
b
as
e
m
o
d
el
u
s
ed
i
s
t
h
e K
-
n
ear
es
t
n
ei
g
h
b
o
r
,
d
eci
s
i
o
n
t
r
ee,
g
r
ad
i
en
t
b
o
o
s
t
i
n
g
t
r
ee
an
d
r
an
d
o
m
f
o
r
es
t
.
T
h
e
m
et
a
l
ear
n
er
u
s
ed
i
s
L
i
g
h
t
G
B
M
.
T
h
e s
t
ac
k
i
n
g
m
o
d
el
acc
u
r
ac
y
r
e
s
u
l
t
s
c
a
n ou
t
pe
r
f
or
m
t
h
e b
as
e
m
o
d
el
accu
r
ac
y
w
i
t
h
an
accu
r
ac
y
r
at
e o
f
9
7
%
.
ACK
NO
W
L
E
D
G
E
M
E
NT
S
T
h
e
a
u
th
o
r
s
w
o
u
ld
lik
e
to
e
x
p
r
e
s
s
th
e
ir
g
r
a
tit
u
d
e
a
n
d
a
p
p
r
e
c
ia
tio
n
to
th
e
U
n
i
v
e
r
s
iti T
u
n
H
u
s
s
e
i
n
O
n
n M
a
l
a
ys
i
a
(
U
T
H
M
)
t
hr
o
ugh t
he
r
e
s
e
a
r
c
h
gr
a
nt
T
I
E
R
1
(
H
7
7
7
)
.
R
EF
ER
EN
C
ES
[
1]
R
.
L
.
C
ons
t
a
nd a
nd R
.
Y
a
z
di
pou
r
,
"
F
i
r
m
f
a
i
l
ur
e
pr
e
di
c
t
i
o
n
m
ode
l
s
:
A
c
r
i
t
i
que
a
nd a
r
e
v
i
e
w
of
r
e
c
e
nt
de
v
e
l
opm
e
nt
s
,
"
A
dv
anc
e
s
i
n E
nt
r
e
pr
e
ne
ur
i
al
F
i
n
anc
e
,
pp.
1
85
-
2
04,
2
01
1
,
do
i
:
10
.
10
07
/
97
8
-
1
-
44
19
-
752
7
-
0_
10
.
[
2]
H
.
Li
,
D
.
A
ndi
na
,
an
d
J
.
Su
n
,
"
M
u
ltip
le
p
r
o
p
o
r
tio
n
c
a
s
e
-
ba
s
i
ng
dr
i
v
e
n C
B
R
E
a
nd i
t
s
a
pp
l
i
c
a
t
i
on
i
n t
he
e
v
a
l
ua
t
i
on of
p
o
s
s
ib
le
f
a
ilu
r
e
o
f
f
ir
m
s
,
"
I
nt
e
r
nat
i
on
al
J
our
nal
of
Sy
s
t
e
m
s
Sc
i
e
nc
e
, v
o
l
. 4
4
,
no
.
8,
pp
.
14
09
-
1
425
,
20
12
,
doi
:
10.
10
80/
00
20
77
21
.
2
01
2.
6
59
68
6
.
[
3]
D.
L
.
Ol
so
n
,
D
.
D
el
en
,
an
d
Y
.
M
en
g
,
"
C
om
pa
r
a
t
i
v
e
a
na
l
y
s
i
s
of
da
t
a
m
i
ni
ng
m
e
t
hods
f
or
ba
nk
r
u
pt
c
y
pr
e
di
c
t
i
on,
"
D
e
c
i
s
i
on
S
up
por
t
Sy
s
t
e
m
s
,
v
o
l
. 5
2
,
n
o.
2,
pp
.
4
64
-
47
3,
20
12
,
d
oi
:
1
0.
1
01
6/
j
.
ds
s
.
2
01
1.
1
0.
00
7
.
[
4]
L
. Z
h
o
u
,
K
.
K
eu
n
g
L
ai
,
a
nd
J
.
Ye
n
,
"
B
a
nk
r
upt
c
y
pr
e
di
c
t
i
o
n
us
i
ng
S
V
M
m
ode
l
s
w
i
t
h a
ne
w
a
ppr
oa
c
h t
o
c
om
bi
ne
f
e
at
u
r
es
s
el
ect
i
o
n
an
d
p
ar
am
et
er
o
p
t
i
m
i
zat
i
o
n
,
"
I
nt
e
r
nat
i
on
al
J
our
na
l
of
Sy
s
t
e
m
s
Sc
i
e
nc
e
, v
o
l
. 4
5
,
n
o.
3,
pp.
24
1
-
2
5
3,
20
14
,
DOI
:
10
.
10
80/
0
02
07
72
1
.
2
01
2.
72
02
93
.
[
5]
L
.
L
i
a
nd J
.
S
un,
"
G
a
us
s
i
a
n c
a
s
e
-
ba
s
e
d r
e
a
s
oni
ng
f
or
bus
i
ne
s
s
f
a
i
l
ur
e
pr
e
di
c
t
i
o
n w
i
t
h e
m
pi
r
i
c
a
l
da
t
a
i
n C
hi
na
,
"
I
nf
or
m
at
i
o
n Sc
i
e
nc
e
s
, v
o
l
. 3
,
no
.
1
-
2
,
pp
.
89
-
1
08
,
20
09
,
d
o
i:
10.
10
16/
j
.
i
ns
.
20
08
.
0
9.
0
03
.
[
6]
B
.
P
r
as
et
i
y
o,
A
l
a
m
s
y
a
h a
nd M
.
A
.
M
u
s
lim
,
"
A
na
l
y
s
i
s
of
b
ui
l
di
ng
e
ne
r
gy
e
f
f
i
c
i
e
nc
y
da
t
a
s
e
t
us
i
ng
na
i
v
e
ba
y
e
s
cl
as
s
i
f
i
cat
i
o
n
cl
as
s
i
f
i
er
,
"
J
our
nal
of
P
hy
s
i
c
s
:
C
onf
e
r
e
nc
e
Se
r
i
e
s
,
v
ol
.
1
79,
n
o
. 3
,
20
19
,
A
r
t
.
no.
0
32
01
6,
do
i
:
10.
10
88/
17
42
-
65
96
/
1
32
1/
3/
03
20
16
.
[
7]
A
.
N
ur
z
a
hput
r
a
,
M.
A
.
M
us
l
i
m
a
nd B
.
P
a
r
s
e
t
i
y
o,
"
O
p
tim
iz
a
tio
n
o
f
C
4
.
5
a
lg
o
r
ith
m
u
s
in
g
m
e
ta
-
le
a
r
n
in
g
in
di
a
g
nos
i
ng
of
c
hr
oni
c
k
i
d
ne
y
di
s
e
a
s
e
s
,
"
J
our
nal
of
P
hy
s
i
c
s
:
C
onf
e
r
e
nc
e
Se
r
i
e
s
,
v
ol
.
1
79,
n
o.
3
,
20
19
,
A
r
t
.
no.
032
02
2,
do
i
:
10.
10
88/
17
42
-
65
96/
13
21/
3/
0
32
02
2
.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t
J
E
l
e
c
&
C
o
m
p
E
n
g
I
S
S
N
:
2088
-
8708
C
om
pany
ba
nk
r
upt
c
y
pr
e
di
c
t
i
on f
r
am
e
w
or
k
bas
e
d on t
he
m
os
t
i
nf
l
ue
nt
i
al
…
(
M
u
c
h
A
z
i
z
M
u
sl
i
m
)
5555
[
8]
Y.
Di
n
g
,
X
.
S
ong
,
a
nd
Y
.
Z
en
,
"
F
o
r
ecas
t
i
n
g
f
i
na
nc
i
a
l
c
ondi
t
i
o
n of
c
hi
ne
s
e
l
i
s
t
e
d c
om
pa
ni
e
s
ba
s
e
d on s
up
por
t
v
e
c
t
or
m
ach
i
n
e,
"
Ex
p
e
r
t S
y
s
te
m
s
w
ith
A
p
p
lic
a
tio
n
s
, v
o
l
. 3
4
,
n
o.
4,
p
p.
30
81
-
30
89,
20
08
,
d
o
i
:
1
0.
1
01
6/
j
.
e
s
w
a
.
2007.
06.
03
7
.
[
9]
B
.
E
.
E
dor
g
a
n,
"
P
r
e
d
i
c
t
i
o
n of
b
a
nk
r
upt
c
y
us
i
ng
s
upp
or
t
v
e
c
t
or
m
a
c
hi
ne
s
:
A
n a
ppl
i
c
a
t
i
on
t
o
ba
n
k
ba
nk
r
upt
c
y
,
"
J
our
n
al
of
St
at
i
s
t
i
c
al
C
om
p
ut
at
i
on &
Si
m
ul
at
i
o
n
,
v
ol
.
83,
n
o
. 8
,
pp.
15
43
-
15
5
5,
20
13
,
d
o
i
:
10.
10
80/
00
94
96
55
.
2
01
2.
6
66
55
0
.
[
1
0]
K
. S
. S
h
i
n
,
T
.
S
.
L
ee,
a
nd
H
.
-
J.
Ki
m
,
"
A
n a
ppl
i
c
a
t
i
on
of
s
uppor
t
v
e
c
t
or
m
a
c
hi
ne
s
i
n ba
nk
r
upt
c
y
pr
e
di
c
t
i
on m
ode
l
,
"
J
our
n
al
o
f
F
i
na
nc
i
al
R
e
s
e
ar
c
h,
v
ol
.
28,
n
o
. 1
,
pp.
1
27
-
13
5,
20
06
,
d
o
i:
10.
10
16/
j
.
e
s
w
a
.
2004.
08
.
0
09
.
[
1
1]
F
.
B
ar
b
o
za
,
H
.
K
i
mu
r
a
,
a
nd
E
.
A
ltm
a
n
,
"
M
a
c
hi
ne
l
e
a
r
ni
ng
m
ode
l
s
a
nd ba
nk
r
upt
c
y
p
r
e
d
ic
tio
n
,
"
E
x
p
e
rt
S
y
st
e
ms w
i
t
h
A
ppl
i
c
at
i
ons
,
vo
l
.
83
,
p
p.
40
5
-
4
17
,
20
17
,
d
o
i:
10.
10
16/
j
.
e
s
w
a
.
2017
.
04.
00
6
.
[
1
2]
R
. G
e
n
g
,
I.
Bo
s
e
b
,
a
nd
X
.
C
h
en
,
"
P
r
e
d
ic
ti
o
n
o
f
f
in
a
n
c
ia
l d
is
tr
e
s
s
:
A
n
e
m
p
ir
ic
a
l s
tu
d
y
o
f
lis
te
d
C
h
in
e
s
e
c
o
m
p
a
n
ie
s
us
i
ng
da
t
a
m
i
ni
ng
,
"
E
ur
ope
a
n
J
our
n
al
of
O
pe
r
a
t
i
o
nal
R
e
s
e
ar
c
h
,
v
ol
.
2
41,
n
o.
1,
p
p.
2
36
-
24
7
,
20
15
,
d
o
i:
10.
10
16/
j
.
e
j
or
.
20
14
.
0
8.
0
16
.
[
1
3]
S
.
L
es
s
m
an
n
,
B
.
B
aes
en
s
,
H.
-
V
. S
e
o
w
,
L
.
C
.
T
h
o
m
as
,
"
B
en
ch
m
ar
k
i
n
g
s
t
at
e
-
of
-
th
e
-
a
r
t c
la
s
s
i
f
ic
a
tio
n
a
lg
o
r
it
h
m
s
f
o
r
cr
ed
i
t
s
co
r
i
n
g
:
A
n
u
p
d
at
e o
f
r
es
ear
ch
,
"
E
u
r
o
pe
an
J
our
nal
o
f
O
pe
r
at
i
o
nal
R
e
s
e
ar
c
h
,
v
ol
.
24
7,
no
.
1,
pp
.
1
24
-
13
6,
201
5
,
d
o
i:
10.
10
16/
j
.
e
j
or
.
20
15
.
0
5
.
03
0
.
[
1
4]
C
.
Lu
o
,
D
es
h
en
g
W
u
,
a
nd
D
e
xi
a
n
g
W
u
, "
A
de
e
p l
e
a
r
ni
ng
a
ppr
oa
c
h f
or
c
r
e
di
t
s
c
or
i
ng
us
i
ng
c
r
e
di
t
de
f
a
ul
t
s
w
a
ps
,
"
En
g
i
n
e
e
r
in
g
Ap
p
lie
d
Ar
tific
ia
l
I
n
t
e
lli
g
en
ce
,
vo
l
.
6
5,
pp
.
4
65
-
47
0,
20
1
7
,
d
o
i
:
1
0.
1
01
6/
j
.
e
ng
a
ppa
i
.
2
01
6.
1
2.
0
02
.
[
1
5]
M
.
S
t
a
nk
ov
a
a
nd D
.
H
a
m
pe
l
,
"
B
a
nk
r
upt
c
y
pr
e
di
c
t
i
on of
e
ng
i
ne
e
r
i
ng
c
o
m
pa
ni
e
s
i
n t
he
E
U
us
i
n
g
c
l
a
s
s
i
f
i
c
a
t
i
on
m
e
t
hods
,
"
A
c
t
a U
ni
v
e
r
s
i
t
at
i
s
A
gr
i
c
ul
t
ur
ae
E
t
S
i
l
v
i
c
ul
t
ur
ae
M
e
nde
l
i
an
ae
B
r
une
ns
is
, v
o
l
. 6
6
,
n
o
. 5
,
pp.
1
34
7
-
13
55,
201
8
,
d
oi
:
10.
11
11
8/
a
c
t
a
un
20
18
6
605
13
47
.
[
1
6]
S
.
K
.
S
hr
i
v
a
s
t
a
v
a
nd P
.
J
.
R
a
m
udu,
"
B
a
nk
r
upt
c
y
pr
e
di
c
t
i
on a
nd s
t
r
e
s
s
qua
nt
i
f
i
c
a
t
i
on us
i
ng
s
up
por
t
v
e
c
t
or
m
a
c
hi
ne
:
ev
i
d
en
ce f
r
o
m
I
n
d
i
an
B
an
k
s
,
"
R
i
s
ks
, v
o
l
.
8,
n
o.
2,
pp.
1
-
22,
2
02
0
,
d
o
i:
10.
33
90/
r
i
s
k
s
802
00
52
.
[
1
7]
A.
Ge
p
p
,
K
.
K
u
ma
r
,
a
nd
S
.
B
h
at
t
ach
ar
y
a
, "
B
us
i
ne
s
s
f
a
i
l
ur
e
pr
e
di
c
t
i
on us
i
ng
de
c
i
s
i
on t
r
e
e
s
,
"
j
our
n
al
of
f
or
e
c
as
t
i
n
g
,
vo
l
.
29
,
no.
6,
pp
.
5
36
-
55
5
,
20
10
,
doi
:
10
.
10
02
/
f
o
r
.
11
53
.
[
1
8]
D
.
D
el
en
,
C
.
K
u
zey
,
a
nd
A
.
U
ya
r
, "
M
eas
u
r
i
n
g
f
i
r
m
p
er
f
o
r
m
an
ce u
s
i
n
g
f
i
n
an
ci
al
r
at
i
o
s
:
A
d
eci
s
i
o
n
t
r
ee ap
p
r
o
ach
,
"
E
x
pe
r
t
Sy
s
t
e
m
s
w
i
t
h A
ppl
i
c
at
i
ons
,
vo
l
.
40
,
n
o.
10
,
p
p.
39
70
-
3
98
3,
2
013
,
d
o
i
:
1
0.
1
01
6/
j
.
e
s
w
a
.
201
3.
0
1.
012
.
[
1
9]
S
.
Y
.
K
i
m
an
d
A
.
U
p
n
ej
a,
"
P
r
ed
i
ct
i
n
g
r
es
t
au
r
an
t
f
i
n
an
ci
al
d
i
s
t
r
es
s
u
s
i
n
g
d
eci
s
i
o
n
t
r
ee an
d
a
da
bo
os
t
e
d de
c
i
s
i
o
n t
r
e
e
m
ode
l
s
,
"
E
c
onom
i
c
M
ode
l
l
i
ng
,
v
ol
.
36,
p
p.
35
4
-
36
2
,
20
14
,
doi
:
1
0
.
10
16/
j
.
e
c
onm
od.
20
13.
10
.
0
05
.
[
2
0]
N
.
O
cal
,
M
.
K
.
E
r
can
,
a
nd
E
.
K
a
d
io
g
lu
, "
P
r
e
d
ic
ti
n
g
f
in
a
n
c
ia
l
f
a
ilu
r
e
u
s
in
g
d
e
c
is
io
n
tr
e
e
a
lg
o
r
ith
m
s
:
A
n
e
m
p
ir
ic
a
l
t
e
s
t
on
t
he
m
a
nuf
a
c
t
ur
i
ng
i
nd
us
t
r
y
a
t
B
or
s
a
I
s
t
a
nbul
,
"
I
nt
e
r
n
at
i
o
nal
J
our
n
al
of
E
c
on
om
i
c
s
an
d F
i
na
nc
e
,
v
o
l
. 7
,
no.
7,
p
p.
18
9
-
20
6,
20
15
,
DOI
:
10
.
55
39/
i
j
e
f
.
v
7n7p
18
9
.
[
2
1]
A
.
G
e
pp a
nd K
.
K
u
m
a
r
,
"
P
r
e
d
ic
tin
g
F
in
a
n
c
ia
l D
is
tr
e
s
s
: A
C
o
m
p
a
r
is
o
n
o
f
S
u
r
v
iv
a
l A
n
a
l
y
s
is
a
n
d
D
e
c
is
io
n
T
r
e
e
T
e
c
hni
que
s
,
"
P
r
oc
e
di
a C
om
p
ut
e
r
Sc
i
e
nc
e
,
vo
l
.
5
4,
pp
.
3
96
-
40
4,
201
5
,
d
o
i:
10.
10
16/
j
.
pr
oc
s
.
2
01
5.
0
6.
046
.
[
2
2]
H
.
S
he
r
m
a
a
nd S
.
K
u
m
a
r
,
"
A
S
ur
v
e
y
on D
e
c
i
s
i
on T
r
e
e
A
l
g
or
i
t
hm
s
of
C
l
a
s
s
i
f
ic
a
tio
n
in
D
a
ta
M
in
in
g
,
"
I
nt
e
r
nat
i
on
a
l
Jo
u
r
n
a
l
S
ci
en
ce R
es
ea
r
ch
,
v
o
l
.
5
,
no.
4,
p
p.
20
94
-
20
97,
2
01
6.
[
2
3]
S
. H
. S
y
e
d N
or
,
S
.
I
s
ma
i
l
,
a
nd
B
.
W
.
Y
ap
Y
ap
,
"
P
e
r
s
o
na
l
ba
nk
r
u
pt
c
y
pr
e
di
c
t
i
on us
i
ng
de
c
i
s
i
o
n t
r
e
e
m
ode
l
,
"
J
our
na
l
of
E
c
on
om
i
c
s
F
i
na
nc
e
and
A
d
m
i
n
i
s
t
r
a
t
i
ve S
ci
en
ce
, v
o
l
.
4
,
n
o.
7,
pp.
1
57
-
17
0,
20
20
,
do
i
:
10.
11
08/
J
E
F
A
S
-
08
-
20
18
-
007
6
.
[
2
4]
R
.
L
.
W
i
l
s
on a
nd
R
.
S
ha
r
da
,
"
B
a
nk
r
upt
c
y
pr
e
di
c
t
i
on
us
i
ng
ne
ur
a
l
ne
t
w
or
k
s
,
"
D
e
c
i
s
i
on S
up
por
t
S
y
s
t
e
m
,
vo
l
.
1
1,
no.
5,
p
p.
54
5
-
55
7,
19
94
,
d
o
i
:
1
0.
101
6/
01
67
-
92
36(
94)
90
02
4
-
8
.
[
2
5]
A
.
F
.
A
t
i
y
a
,
"
B
a
nk
r
upt
c
y
pr
e
di
c
t
i
on f
or
c
r
e
di
t
r
i
s
k
us
i
ng
ne
ur
a
l
ne
t
w
or
k
s
:
A
s
ur
v
e
y
a
nd ne
w
r
e
s
ul
t
s
,
"
I
E
E
E
T
r
ans
ac
t
i
ons
on
N
e
ur
al
N
e
t
w
or
k
s
,
v
ol
.
12,
n
o
. 4
,
pp.
9
29
-
93
5,
20
01
,
d
o
i:
10.
11
09/
72
.
9
35
10
1
.
[
2
6]
M
.
A
n
an
d
ar
aj
an
,
P
.
Le
e
,
a
nd
A
.
A
n
an
d
ar
aj
an
,
"
B
a
nk
r
upt
c
y
pr
e
di
c
t
i
on
of
f
i
na
nc
i
a
l
l
y
s
t
r
e
s
s
e
d
f
i
r
m
s
:
a
n e
x
a
m
i
na
t
i
on
o
f
t
h
e p
r
ed
i
ct
i
v
e accu
r
ac
y
o
f
ar
t
i
f
i
ci
al
n
eu
r
al
n
et
w
o
r
k
s
,
"
I
nt
e
r
nat
i
o
nal
J
our
n
al
I
nt
e
l
l
e
ge
n
t
Sy
s
t
e
m
A
c
c
ount
,
v
o
l
.
1
0
,
no.
2,
p
p.
69
-
8
1,
20
01
,
do
i
:
10
.
100
2
/
i
s
a
f
.
1
99
.
[
2
7]
M
.
J.
Ki
m
a
n
d
D.
K.
K
a
ng
,
"
E
ns
e
m
bl
e
w
i
t
h ne
ur
a
l
ne
t
w
or
k
s
f
or
ba
nk
r
u
pt
c
y
pr
e
di
c
t
i
on,
"
E
xp
er
t
S
ys
t
em
s
w
i
t
h
A
ppl
i
c
at
i
on
s
,
vo
l
.
37
,
n
o.
4,
pp
.
33
7
3
-
33
79
,
20
10
,
d
oi
:
10.
10
16/
j
.
e
s
w
a
.
2009.
10.
01
2
.
[
2
8]
A.
A.
K
a
s
g
a
r
i
,
M
.
D
i
v
s
al
ar
,
M
.
R
.
J
a
vi
d
,
a
nd
S
.
J
.
E
b
r
ah
i
m
i
an
,
"
P
r
e
di
c
t
i
o
n of
ba
nk
r
upt
c
y
I
r
a
ni
a
n c
or
por
a
t
i
ons
t
hr
o
ug
h a
r
t
i
f
i
c
i
a
l
ne
ur
a
l
ne
t
w
or
k a
nd pr
o
bi
t
-
b
as
ed
an
al
y
s
es
,
"
N
e
ur
al
C
om
p
ut
i
ng A
ppl
i
c
at
i
on
, v
o
l
. 2
3
,
no.
3
-
4
,
pp.
92
7
-
9
3
6,
20
13
,
d
oi
:
10.
10
07/
s
005
21
-
01
2
-
10
17
-
z
.
[
2
9]
Z
. D
o
ng
e
t a
l.
"
A
n
ef
f
e
ct
i
v
e
co
m
p
u
t
at
i
o
n
al
m
o
d
el
f
o
r
b
an
k
r
u
p
t
cy
p
r
ed
i
ct
i
o
n
u
s
i
n
g
k
er
n
el
ex
t
r
e
m
e l
ear
n
i
n
g
m
ach
i
n
e
a
ppr
oa
c
h,
"
C
om
put
at
i
on
al
E
c
on
o
m
i
c
s
, v
o
l
. 4
9
,
no.
2,
pp
.
3
25
-
3
41,
201
7
,
d
oi
:
10.
10
07/
s
10
61
4
-
01
6
-
9
562
-
7
.
[
3
0]
G
.
P
.
N
a
i
d
u a
nd
K
.
G
ov
i
nda
,
"
B
a
nk
r
upt
c
y
P
r
e
di
c
t
i
on U
s
i
ng
N
e
ur
a
l
N
e
t
w
or
ks
,
"
2018 2n
d
I
n
t
e
r
nat
i
o
nal
C
onf
e
r
e
nc
e
on I
nv
e
nt
i
v
e
Sy
s
t
e
m
s
an
d C
ont
r
ol
(
I
C
I
SC
)
,
20
18
,
pp
.
2
48
-
25
1
,
d
o
i:
10.
11
09/
I
C
I
S
C
.
2
01
8.
83
99
07
2
.
[
3
1]
T
.
H
os
a
ka
,
"
B
a
nk
r
upt
c
y
pr
e
di
c
t
i
on us
i
ng
i
m
a
g
e
d
f
i
na
nc
i
a
l
r
a
t
i
os
a
nd c
onv
ol
ut
i
ona
l
ne
ur
a
l
ne
t
w
or
k
s
,
"
E
xp
er
t
S
y
s
te
m
s
w
ith
Ap
p
lic
a
tio
n
s
,
v
ol
.
1
17,
pp
.
2
87
-
29
9,
20
19
,
d
o
i
:
1
0.
1
0
16/
j
.
e
s
w
a
.
2018.
09.
03
9
.
[
3
2]
D
.
Li
a
n
g
,
C.
-
C
. L
u
, C
.
-
F
on
g T
s
a
i
,
G
.
-
An
S
hi
ha
,
"
F
i
n
a
n
ci
al
r
at
i
o
s
an
d
co
r
p
o
r
at
e g
o
v
er
n
an
c
e i
n
d
i
cat
o
r
s
i
n
ba
nk
r
upt
c
y
pr
e
di
c
t
i
o
n:
A
c
om
pr
e
he
ns
i
v
e
s
t
udy
,
"
E
ur
ope
an
J
o
ur
nal
of
O
pe
r
at
i
o
nal
R
e
s
e
ar
c
h
,
vo
l
.
25
2,
n
o.
2,
pp.
56
1
-
572
,
2
01
6.
,
d
o
i:
10
.
1
01
6/
j
.
e
j
or
.
20
16.
01
.
0
12
.
[
3
3]
H
.
A.
Al
a
k
a
e
t a
l.
,
"
S
y
s
t
e
m
a
t
i
c
r
e
v
i
e
w
of
ba
nk
r
upt
c
y
pr
e
di
c
t
i
on m
ode
l
s
:
T
ow
a
r
ds
a
f
r
a
m
e
w
or
k
f
o
r
t
ool
s
e
l
e
c
t
i
on,
"
Ex
p
e
r
t S
y
s
te
m
s
W
ith
Ap
p
lic
a
tio
n
s
,
v
ol
.
94,
p
p.
16
4
-
18
4
,
20
18
,
doi
:
10.
10
16/
j
.
e
s
w
a
.
2017.
10
.
0
40
.
[
3
4]
Y
.
S
hi
a
nd
X
.
L
i
,
"
A
n ove
r
v
i
e
w
o
f
ba
nk
r
upt
c
y
pr
e
di
c
t
i
on m
ode
l
s
f
or
c
or
por
a
t
e
f
i
r
m
s
:
A
s
y
s
t
e
m
a
t
i
c
l
i
t
e
r
a
t
ur
e
r
e
v
i
e
w,
"
I
nt
angi
bl
e
C
a
pi
t
al
,
v
ol
.
15,
n
o.
2,
pp
.
1
14
-
12
7,
20
19
,
d
oi
:
10.
39
26/
i
c
.
1
354
.
[
3
5]
Y.
Qu
,
P
.
Q
u
a
n,
M
.
Le
i
,
a
nd
Y
.
Sh
i
,
"
R
e
v
i
e
w
of
ba
nk
r
upt
c
y
pr
e
di
c
t
i
on
us
i
ng
m
a
c
hi
ne
l
e
a
r
ni
ng
a
nd de
e
p l
e
a
r
ni
ng
t
ech
n
i
q
u
es
,
"
P
r
oc
e
di
a
C
om
p
ut
e
r
Sc
i
e
nc
e
,
v
ol
.
16
2,
pp.
8
95
-
89
9,
2
019
,
d
o
i
:
1
0.
1
01
6/
j
.
pr
oc
s
.
20
19.
12
.
06
5
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
20
88
-
8708
In
t
J
E
l
e
c
&
C
o
m
p
E
n
g
,
V
o
l.
11
, N
o
.
6
,
D
ecem
b
er
2
0
2
1
:
5
549
-
55
57
5556
[
3
6]
S
.
S
.
D
ev
i
an
d
Y
.
R
ad
h
i
k
a,
"
A
s
u
r
v
e
y
o
n
m
ach
i
n
e l
ear
n
i
n
g
an
d
s
t
at
i
s
t
i
cal
t
ech
ni
q
ue
s
i
n B
a
nk
r
u
pt
c
y
P
r
e
di
c
t
i
on
,
"
I
nt
e
r
nat
i
o
nal
J
o
ur
n
al
of
M
ac
hi
ne
L
e
ar
ni
n
g a
nd C
om
p
ut
i
n
g
,
v
o
l
. 8
,
n
o
. 2
,
pp
.
1
33
-
1
39,
20
18
,
d
oi
:
10.
18
17
8/
i
j
m
l
c
.
201
8.
8.
2
.
6
76
.
[
3
7]
P
.
G
ni
p a
nd
P
.
D
r
ot
a
r
,
"
E
ns
e
m
bl
e
m
e
t
hods
f
or
s
t
r
ong
l
y
i
m
ba
l
a
n
c
e
d da
t
a
:
ba
nk
r
up
t
c
y
pr
e
di
c
t
i
on
,
"
20
19 I
E
E
E
1
7t
h
I
nt
e
r
nat
i
o
nal
Sy
m
p
os
i
um
on
I
nt
e
l
l
i
ge
nt
Sy
s
t
e
m
s
an
d I
n
f
or
m
at
i
c
s
(
SI
SY
)
,
20
19
,
pp
.
1
55
-
16
0,
d
o
i:
10.
11
09/
S
I
S
Y
4
75
53
.
2
01
9.
9
11
15
57
.
[
3
8]
G
. Wa
n
g
,
J
.
M
a
,
an
d
S
.
Y
an
g
,
"
A
n i
m
pr
ov
e
d b
oos
t
i
ng
ba
s
e
d
on
f
e
a
t
ur
e
s
e
l
e
c
t
i
on f
or
c
or
por
a
t
e
ba
nk
r
upt
c
y
p
r
e
d
ic
tio
n
,
"
Ex
p
e
r
t S
y
s
te
m
s
W
ith
Ap
p
lic
a
t
io
n
s
, v
o
l
. 4
1
,
n
o
. 5
,
pp.
2
35
3
-
23
61,
2
0
14
,
d
o
i:
10.
10
16/
j
.
e
s
w
a
.
2013.
09
.
0
33
.
[
3
9]
N
.
W
a
ng
,
"
B
a
nk
r
upt
c
y
pr
e
di
c
t
i
on us
i
ng
m
a
c
hi
ne
l
e
a
r
ni
ng
,
"
J
our
nal
of
M
at
he
m
at
i
c
al
F
i
n
anc
e
,
v
o
l
. 7
,
no.
4
,
pp.
90
8
-
9
1
8,
20
17
,
d
oi
:
10.
42
36/
j
m
f
.
2017.
74
04
9
.
[
4
0]
H
.
F
ar
i
s
,
W
.
M
an
as
eer
,
M
.
S
a
a
de
h,
A
.
M
o
r
a,
P
.
A
.
C
a
s
tillo
,
I
.
A
lja
r
a
h
,
"
I
m
pr
ov
i
ng
f
i
na
nc
i
a
l
ba
nk
r
upt
c
y
pr
e
di
c
t
i
o
n
i
n a
hi
g
hl
y
i
m
ba
l
a
nc
e
d c
l
a
s
s
di
s
t
r
i
but
i
on
us
i
ng
ov
e
r
s
a
m
pl
i
ng
a
nd e
ns
e
m
bl
e
l
e
a
r
ni
ng
:
a
c
a
s
e
f
r
om
t
he
S
pa
ni
s
h
m
ar
k
et
,
"
P
ro
g
re
ss i
n
Ar
tif
ic
ia
l
I
n
t
el
l
i
g
en
ce
,
v
o
l
. 9
, n
o
.
1
,
p
p
. 3
1
-
53
,
20
19.
[
4
1]
Z
. C
h
e
n
,
W
.
C
he
n,
a
nd
Y
.
Sh
i
,
"
E
ns
e
m
bl
e
l
e
a
r
ni
ng
w
i
t
h l
a
be
l
pr
o
por
t
i
ons
f
or
ba
nk
r
upt
c
y
pr
e
di
c
t
i
on,
"
E
xp
er
t
S
y
st
e
ms
W
ith
Ap
p
lic
a
ti
o
n
s
,
v
ol
.
1
46,
20
20
, A
r
t
. n
o
.
11
31
55
,
d
o
i:
10
.
10
16/
j
.
e
s
w
a
.
2019.
11
31
55
.
[
4
2]
T.
Le
M
in
h
,
T
.
Vo
,
B
.
V
o
, M
i
Y
.
L
ee,
a
nd
S
.
W
oo
k B
a
i
k
,
"
A
hy
br
i
d a
ppr
oa
c
h us
i
ng
ov
e
r
s
a
m
pl
i
n
g
t
e
c
hni
que
a
nd
co
s
t
-
s
e
ns
i
t
i
v
e
l
e
a
r
ni
ng
f
or
ba
nk
r
upt
c
y
pr
e
di
c
t
i
on,
"
C
o
m
p
l
exi
t
y,
v
ol
.
20
19,
no
.
2,
pp.
1
-
1
2,
20
19
,
A
r
t
.
no.
84
60
93
4
,
doi
:
1
0.
1
15
5/
2
01
9/
84
60
93
4
.
[
4
3]
S
.
V
el
l
em
ch
et
i
a
nd P
.
S
i
ng
h "
C
l
a
s
s
i
m
ba
l
a
nc
e
de
e
p l
e
a
r
ni
ng
f
or
ba
nk
r
upt
c
y
pr
e
di
c
t
i
on,
"
20
20 F
i
r
s
t
I
nt
e
r
nat
i
on
al
C
onf
e
r
e
nc
e
on P
ow
e
r
,
C
o
nt
r
ol
an
d C
om
put
i
n
g T
e
c
hn
ol
ogi
e
s
(
I
C
P
C
2T
)
,
20
20
,
pp.
421
-
42
5,
doi
:
10.
11
09/
I
C
P
C
2T
48
08
2.
20
20.
90
7
146
0
.
[
4
4]
S
.
S
m
i
t
i
a
nd M
.
S
o
ui
,
"
B
a
nk
r
u
pt
c
y
pr
e
di
c
t
i
on us
i
ng
de
e
p l
e
a
r
ni
ng
a
ppr
oa
c
h ba
s
e
d
on B
or
de
r
l
i
ne
S
M
O
T
E
,
"
I
nf
or
ma
t
i
o
n
S
y
st
e
ms F
ro
n
t
i
e
rs
,
v
ol
.
22,
n
o.
5,
pp.
1
06
7
-
10
83
,
2
02
0.
[
4
5]
T
.
H.
Ka
n
g
a
,
S.
D.
Ja
m
e
s,
a
nd
F
.
F
ab
i
an
,
"
R
e
a
l
opt
i
ons
a
nd s
t
r
a
t
e
g
i
c
ba
nk
r
upt
c
y
,
"
J
our
nal
o
f
B
us
i
ne
s
s
R
e
s
e
ar
c
h,
v
ol
.
1
17,
p
p.
15
2
-
16
2
,
20
20
,
doi
:
10.
10
16/
j
.
j
b
us
r
e
s
.
2
02
0.
0
5.
05
7
.
[
4
6]
W
. C
.
L
i
n
,
Y
.‐
H
.
L
u
,
a
nd
C.
‐F
.
Ts
a
i
,
"
F
eat
u
r
e s
el
e
ct
i
o
n
i
n
s
i
n
g
l
e an
d
en
s
e
m
b
l
e l
ear
n
i
n
g
-
ba
s
e
d ba
nk
r
upt
c
y
pr
e
di
c
t
i
o
n m
ode
l
s
,
"
E
x
p
e
rt
S
y
st
e
ms,
vo
l
.
36
,
n
o.
1,
pp
.
1
-
8
,
2
01
8,
doi
:
1
0.
1
11
1/
e
x
s
y
.
123
35
.
[
4
7]
H.
S
o
n
,
C
. H
y
u
n
, D
. P
h
a
n
,
H.
J.
Hwa
n
g
,
"
D
a
t
a
a
na
l
y
t
i
c
a
ppr
oa
c
h f
or
ba
nk
r
upt
c
y
pr
e
di
c
t
i
on,
"
E
xp
er
t
S
ys
t
em
s
W
ith
A
ppl
i
c
at
i
ons
,
vo
l
.
13
8
,
20
19
,
A
r
t
. n
o
.
11
28
16
,
doi
:
1
0.
10
16/
j
.
e
s
w
a
.
201
9.
07.
03
3
.
[
4
8]
J
.
N
obr
e
a
nd R
.
F
.
N
e
v
e
s
,
"
C
o
m
bi
ni
ng
P
r
i
nc
i
pa
l
C
om
pone
nt
A
na
l
y
s
i
s
,
D
i
s
c
r
e
t
e
W
a
v
e
l
e
t
T
r
a
ns
f
or
m
a
nd X
G
B
oos
t
to
tr
a
d
e
in
th
e
f
in
a
n
c
ia
l
m
a
r
k
e
ts
,
"
Ex
p
e
r
t S
y
s
te
m
s
W
ith
A
p
p
lic
a
tio
n
s
,
v
o
l.
1
2
5
,
p
p
.
1
8
1
-
19
4
,
20
19
,
d
o
i:
10.
10
16/
j
.
e
s
w
a
.
2019.
01
.
0
83
.
[
4
9]
T
.
C
he
n a
n
d C
.
G
ue
s
t
r
i
n,
"
X
G
B
oos
t
:
A
s
cal
ab
l
e t
r
ee b
o
o
s
t
i
n
g
s
y
s
t
e
m
,"
P
r
oc
e
e
di
ngs
of
t
he
22
nd
A
C
M
SI
G
K
D
D
I
nt
e
r
nat
i
o
nal
C
onf
e
r
e
nc
e
on
K
now
l
e
d
ge
D
i
s
c
ov
e
r
y
and
D
at
a M
i
ni
ng
,
2
01
6,
p
p.
785
-
7
94
,
d
o
i:
10.
11
45/
29
39
67
2.
29
39
78
5
.
[
5
0]
Y
.
F
r
e
und
,
a
nd
R
.
E
.
S
ch
ap
i
r
e
,
"
A
s
hor
t
i
nt
r
od
uc
t
i
o
n t
o bo
os
t
i
n
g
,
"
J
our
nal
of
Ja
p
a
n
es
e S
o
ci
et
y F
o
r
A
r
t
i
f
i
ci
a
l
I
n
te
llig
e
n
c
e
,
v
ol
.
14,
n
o
. 5
,
p
p
.
771
-
7
90,
1
999
.
[
5
1]
D.
H
.
W
o
l
p
er
t
,
"
S
t
ack
ed
g
en
er
al
i
zat
i
o
n
,
"
N
eu
r
a
l
N
et
w
o
r
ks
,
v
o
l
. 5
,
no.
2,
p
p.
24
1
-
2
59,
1
99
2,
doi
:
1
0.
1
01
6/
S
08
93
-
608
0(
05)
80
02
3
-
1
.
[
5
2]
L
.
B
r
ei
m
an
,
"
S
t
ack
ed
r
eg
r
es
s
i
o
n
s
,
"
M
ac
hi
ne
L
e
ar
ni
ng
,
vo
l
.
24
,
no.
1,
p
p.
49
-
6
4,
19
96
.
[
5
3]
M.
J
.
V
an
d
er
L
aan
,
E
.
C
.
P
o
l
l
e
y
, A
. E
. H
u
b
b
a
r
d
,
"
S
u
p
er
l
ear
n
er
,
"
S
ta
tis
tic
a
l
Ap
p
lic
a
t
io
n
s
in
G
e
n
e
tic
s
a
n
d
M
ol
e
c
ul
ar
,
v
o
l
. 6
,
n
o
. 1
,
20
07
,
A
r
t
.
n
o.
25,
d
oi
:
10.
22
02
/
1
54
4
-
61
15.
1
30
9
.
[
5
4]
M
. Z
i
e
b
a
,
S
.
K
.
T
o
m
cz
ak
,
J
.
M.
T
o
m
c
zak
,
"
E
ns
e
m
bl
e
boos
t
e
d t
r
e
e
s
w
i
t
h s
y
nt
he
t
i
c
f
e
a
t
ur
e
s
g
e
ne
r
a
t
i
on i
n
a
ppl
i
c
a
t
i
o
n t
o
ba
nk
r
up
t
c
y
pr
e
di
c
t
i
on,
"
Ex
p
e
r
t
S
y
s
te
m
s
W
ith
A
p
p
lic
a
tio
n
s
,
v
ol
.
58,
p
p.
93
-
10
1,
20
16
,
d
o
i:
10.
10
16/
j
.
e
s
w
a
.
2016.
04
.
0
01
.
[
5
5]
T
.
M
u
s
ta
q
im
,
K
.
U
m
a
m
a
nd M
.
A
.
M
u
s
lim
,
"
T
w
i
t
t
e
r
t
e
x
t
m
i
ni
ng
f
or
s
e
nt
i
m
e
nt
a
na
l
y
s
i
s
on g
ov
e
r
n
m
e
nt
’
s
r
e
s
pons
e
to
f
o
r
e
s
t f
ir
e
s
w
ith
v
a
d
e
r
le
x
ic
o
n
p
o
la
r
ity
d
e
te
c
tio
n
a
n
d
k
-
ne
a
r
e
s
t
ne
i
g
hbor
a
l
g
or
i
t
hm
,
"
J
our
n
al
o
f
P
hy
s
i
c
s
:
C
o
n
f
er
en
ce S
er
i
es
, v
o
l
.
15
67
,
p
p
. 8
-
1
5,
20
20
,
A
r
t
.
no.
03
20
24
,
do
i
:
10.
10
88/
17
42
-
65
96
/
1
56
7/
3/
03
20
24
.
[
5
6]
I.
A
s
h
a
ri
,
M
.
A
.
M
u
s
lim
,
a
nd
A
.
A
l
a
m
s
y
ah
,
"
C
o
m
pa
r
i
s
on
P
e
r
f
or
m
a
nc
e
of
G
e
ne
t
i
c
A
l
g
or
i
t
hm
a
nd A
nt
C
ol
ony
O
p
tim
iz
a
tio
n
in
C
o
u
r
s
e
S
c
h
e
d
u
li
n
g
O
p
tim
iz
in
g
,
"
S
c
ie
n
tific
J
o
u
r
n
a
l I
n
fo
r
m
a
t
ic
s
,
v
o
l
. 3
,
n
o.
2,
p
p.
149
-
15
8
,
20
16
,
doi
:
10
.
15
29
4
/
s
j
i
.
v3
i
2.
79
11
.
[
5
7]
A.
R
.
S
a
f
i
t
r
i
a
nd M
.
A
.
M
us
l
i
m
"
I
m
pr
ov
e
d A
c
c
ur
a
c
y
of
N
a
i
ve
B
a
y
e
s
C
l
a
s
s
i
f
i
e
r
f
or
D
e
t
e
r
m
i
na
t
i
on
of
C
us
t
om
e
r
C
hur
n U
s
e
s
S
M
O
T
E
a
nd
G
e
ne
t
i
c
A
l
g
or
i
t
hm
s
,
"
J
our
nal
Sof
t
C
om
p
ut
i
n
g E
x
pl
or
at
i
on
, v
o
l
. 1
,
no.
1,
p
p.
70
-
75
,
20
20
,
doi
:
10
.
5
24
65/
j
o
s
c
e
x.
v1
i
1
.
5
.
[
5
8]
Y
.
G
uo,
"
C
r
e
di
t
r
i
s
k
a
s
s
e
s
s
m
e
nt
of
P
2
P
l
e
ndi
ng
pl
a
t
f
or
m
t
ow
a
r
ds
bi
g
da
t
a
ba
s
e
d on B
P
ne
ur
a
l
ne
t
w
or
k
,
"
J
our
nal
of
V
i
s
ual
C
om
m
u
ni
c
at
i
on
a
nd I
m
a
ge
R
ep
r
es
en
t
a
t
i
o
n
,
v
o
l
. 7
1
,
2020
,
A
r
t
.
no.
102730,
do
i
:
10
.
1016/
j
.
j
vc
i
r
.
201
9.
102730
.
[
5
9]
D
.
Li
a
n
g
,
C
.
-
F
.
Ts
a
i
,
H
.
-
Y
.
(Ri
c
h
a
rd
) L
u
,
L
.
-
S
.
C
ha
ng
, "
C
o
m
bi
ni
n
g
c
or
por
a
t
e
g
ov
e
r
na
nc
e
i
ndi
c
a
t
or
s
w
i
t
h s
t
a
c
k
i
ng
e
ns
e
m
bl
e
s
f
or
f
i
na
nc
i
a
l
di
s
t
r
e
s
s
pr
e
di
c
t
i
on
,
"
J
our
n
al
of
B
us
i
ne
s
s
R
e
s
e
ar
c
h
,
v
ol
.
1
20,
p
p.
13
7
-
14
6
,
20
20
,
d
o
i
:
10.
10
16/
j
.
j
b
us
r
e
s
.
2
02
0.
0
7.
05
2
.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t
J
E
l
e
c
&
C
o
m
p
E
n
g
I
S
S
N
:
2088
-
8708
C
om
pany
ba
nk
r
upt
c
y
pr
e
di
c
t
i
on f
r
am
e
w
or
k
bas
e
d on t
he
m
os
t
i
nf
l
ue
nt
i
al
…
(
M
u
c
h
A
z
i
z
M
u
sl
i
m
)
5557
B
I
O
G
RAP
H
I
ES
O
F
AUT
H
O
RS
M
uc
h A
z
i
z
M
us
l
i
m
P
h
D
can
d
i
d
at
e i
n
t
h
e f
acu
l
t
y
o
f
m
an
ag
e
m
en
t
t
ech
n
o
l
o
g
y
at
U
n
i
v
er
s
i
t
i
T
u
n
H
us
s
e
i
n O
nn
M
a
l
a
y
s
i
a
(
U
T
H
M
)
.
T
he
s
c
ope
of
r
e
s
e
a
r
c
h he
i
s
c
ur
r
e
nt
l
y
w
or
k
i
ng
on i
s
i
n t
he
f
i
e
l
ds
of
C
o
m
put
e
r
S
c
i
e
nc
e
,
D
a
t
a
M
i
ni
ng
a
nd N
e
t
w
or
k
i
ng
.
be
s
i
de
s
th
a
t h
e
is
a
ls
o
a
le
c
tu
r
e
r
in
th
e
co
m
p
u
t
er
s
ci
en
ce d
ep
ar
t
m
en
t
o
f
t
h
e U
n
i
v
er
s
i
t
as
N
eg
er
i
S
e
m
ar
an
g
(
U
N
N
E
S
)
.
Em
a
il:
a
212m
us
l
i
m
@
m
a
i
l
.
unne
s
.
a
c
.
i
d
Y
os
z
a D
as
r
i
l
r
ecei
v
ed
P
h
D
an
d
M
as
t
er
D
eg
r
ee d
e
g
r
ee i
n
A
p
p
l
i
ed
M
at
h
em
at
i
cs
f
r
o
m
U
n
i
v
er
i
t
i
P
u
t
r
a M
al
ay
s
i
a i
n
2
0
0
3
an
d
1
9
9
9
r
es
p
ect
i
v
el
y
an
d
B
ach
el
o
r
D
eg
r
ee i
n
M
at
h
e
m
at
i
cs
f
r
o
m
U
ni
v
e
r
s
i
t
a
s
R
i
a
u,
I
nd
one
s
i
a
i
n
199
4.
F
r
om
19
99 t
o
20
06,
he
w
a
s
a
l
e
c
t
ur
e
r
i
n t
he
U
n
i
v
e
r
s
i
t
i
Ma
l
a
y
s
i
a
T
e
r
e
ngg
a
nu (
U
M
T
)
.
I
n
t
he
m
i
ddl
e
of
20
06 i
nt
o N
ov
2
01
9,
he
s
e
r
v
e
a
s
a
S
e
ni
or
L
e
c
t
ur
e
r
a
t
t
he
D
e
pa
r
t
m
e
nt
of
E
l
e
c
t
r
oni
c
s
a
nd C
om
put
e
r
E
ng
i
ne
e
r
i
ng
T
e
c
hnol
og
y
,
U
ni
v
e
r
s
i
t
i
T
e
k
ni
k
a
l
M
al
a
y
s
i
a M
el
a
k
a (
U
T
eM
)
.
C
u
r
r
en
t
l
y
,
h
e i
s
a L
ect
u
r
er
at
F
acu
l
t
y
o
f
T
ech
n
o
l
o
g
y
M
an
a
g
e
m
en
t
an
d
B
u
si
n
e
ss,
Un
i
v
e
r
si
t
i
T
u
n
Hu
sse
i
n
On
n
M
a
l
a
y
si
a
(
U
T
HM
)
.
Hi
s r
e
se
a
r
c
h
i
n
t
e
r
e
st
s a
r
e
i
n
O
p
t
i
m
i
zat
i
o
n
,
E
n
g
i
n
eer
i
n
g
M
at
h
e
m
at
i
cs
an
d
C
o
m
p
u
t
at
i
o
n
al
M
at
h
em
at
i
cs
.
Em
a
il:
y
os
z
a
@
ut
e
m
.
e
du.
m
y
Evaluation Warning : The document was created with Spire.PDF for Python.