I
AE
S In
t
er
na
t
io
na
l J
o
urna
l o
f
Art
if
icia
l In
t
ellig
ence
(
I
J
-
AI
)
Vo
l.
1
5
,
No
.
5
,
Octo
b
er
2
0
2
6
,
p
p
.
4
6
2
4
~
4
6
3
4
I
SS
N:
2
2
5
2
-
8
9
3
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijai.v
15
.i
5
.
p
p
4
6
2
4
-
4
6
3
4
4624
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
Deep r
einforcem
e
nt
lea
rning
-
ba
sed credit
risk
predic
tion mo
del
for multi
-
a
tt
ri
but
e data
Al
-
K
ho
wa
rizm
i
1
,
F
er
ry
F
a
c
hriza
l
2
,
Ro
m
i F
a
dil
la
h Ra
hm
a
t
3
,
Arif
Ridh
o
L
ub
is
2
,
H
a
riy
a
di
4
1
D
e
p
a
r
t
me
n
t
o
f
I
n
f
o
r
mat
i
o
n
Te
c
h
n
o
l
o
g
y
,
F
a
c
u
l
t
y
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
I
n
f
o
r
mat
i
o
n
T
e
c
h
n
o
l
o
g
y
,
U
n
i
v
e
r
si
t
a
s
M
u
h
a
mm
a
d
i
y
a
h
S
u
ma
t
e
r
a
U
t
a
r
a
,
M
e
d
a
n
,
I
n
d
o
n
e
si
a
2
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
En
g
i
n
e
e
r
i
n
g
a
n
d
I
n
f
o
r
ma
t
i
c
s,
P
o
l
i
t
e
k
n
i
k
N
e
g
e
r
i
M
e
d
a
n
,
M
e
d
a
n
,
I
n
d
o
n
e
si
a
3
D
e
p
a
r
t
me
n
t
o
f
I
n
f
o
r
mat
i
o
n
Te
c
h
n
o
l
o
g
y
,
F
a
c
u
l
t
y
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
I
n
f
o
r
mat
i
o
n
T
e
c
h
n
o
l
o
g
y
,
U
n
i
v
e
r
si
t
a
s
S
u
ma
t
e
r
a
U
t
a
r
a
,
M
e
d
a
n
,
I
n
d
o
n
e
s
i
a
4
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
i
c
a
l
En
g
i
n
e
e
r
i
n
g
,
F
a
c
u
l
t
y
o
f
E
n
g
i
n
e
e
r
i
n
g
,
U
n
i
v
e
r
si
t
a
s
M
u
h
a
mm
a
d
i
y
a
h
S
u
ma
t
e
r
a
B
a
r
a
t
,
P
a
d
a
n
g
,
I
n
d
o
n
e
s
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Ma
y
1
4
,
2
0
2
5
R
ev
is
ed
J
u
l 1
8
,
2
0
2
6
Acc
ep
ted
Au
g
1
2
,
2
0
2
6
Th
is
p
a
p
e
r
a
ims
t
o
d
e
v
e
lo
p
a
m
u
l
ti
-
a
tt
rib
u
te
d
a
ta
-
b
a
se
d
c
re
d
it
risk
p
re
d
ictio
n
m
o
d
e
l
u
si
n
g
t
h
e
d
e
e
p
re
i
n
fo
rc
e
m
e
n
t
lea
rn
in
g
(DRL)
a
p
p
ro
a
c
h
.
In
t
h
e
c
o
n
tex
t
o
f
c
re
d
it
w
o
rth
i
n
e
ss
e
v
a
lu
a
ti
o
n
,
a
m
o
d
e
l
is
n
e
e
d
e
d
t
h
a
t
is
n
o
t
o
n
l
y
a
c
c
u
ra
te
b
u
t
a
lso
a
d
a
p
ti
v
e
to
c
o
m
p
lex
d
a
ta
d
y
n
a
m
ics
a
n
d
ris
k
p
a
tt
e
rn
s.
T
h
is
st
u
d
y
u
se
s
a
d
a
tas
e
t
o
f
5
,
0
0
0
e
n
tri
e
s
th
a
t
h
a
v
e
g
o
n
e
t
h
ro
u
g
h
a
p
re
p
ro
c
e
ss
in
g
p
ro
c
e
ss
,
wit
h
fe
a
tu
re
s
su
c
h
a
s
c
r
e
d
it
sc
o
re
,
i
n
c
o
m
e
,
a
g
e
,
l
o
a
n
a
m
o
u
n
t,
a
n
d
e
m
p
lo
y
m
e
n
t
h
ist
o
ry
.
I
n
it
ial
c
o
rre
latio
n
a
n
a
ly
sis
t
h
ro
u
g
h
h
e
a
tma
p
v
isu
a
li
z
a
ti
o
n
sh
o
ws
th
a
t
c
re
d
it
sc
o
re
h
a
s
a
stro
n
g
n
e
g
a
ti
v
e
re
latio
n
sh
ip
with
c
re
d
it
risk
,
w
h
il
e
o
t
h
e
r
fe
a
tu
re
s
sh
o
w
we
a
k
to
m
o
d
e
ra
te
c
o
rre
lat
io
n
s.
T
h
e
e
v
a
lu
a
ti
o
n
o
f
m
o
d
e
l
p
e
rfo
rm
a
n
c
e
is
sh
o
wn
th
r
o
u
g
h
th
e
a
re
a
u
n
d
e
r
th
e
c
u
rv
e
(AU
C)
v
a
lu
e
o
n
th
e
re
c
e
iv
e
r
o
p
e
ra
ti
n
g
c
h
a
ra
c
teristic
(ROC)
c
u
rv
e
o
f
0
.
8
8
,
wh
ich
i
n
d
ica
tes
e
x
c
e
ll
e
n
t
d
isc
rimin
a
ti
o
n
a
b
il
it
y
.
Th
e
fi
n
a
l
e
v
a
l
u
a
ti
o
n
re
su
l
ts
sh
o
w
th
a
t
t
h
e
DRL
m
o
d
e
l
h
a
s
a
n
a
c
c
u
ra
c
y
o
f
9
1
%
,
p
re
c
isio
n
o
f
8
9
%
,
re
c
a
ll
o
f
9
0
%
,
a
n
d
F
1
-
sc
o
re
o
f
8
9
.
5
%
,
i
n
d
ica
ti
n
g
t
h
a
t
th
e
m
o
d
e
l
n
o
t
o
n
ly
h
a
s
h
ig
h
a
c
c
u
ra
c
y
b
u
t
is
a
lso
b
a
lan
c
e
d
i
n
it
s
se
n
siti
v
it
y
a
n
d
sp
e
c
ifi
c
it
y
.
Th
u
s,
t
h
e
a
p
p
li
c
a
ti
o
n
o
f
DRL
h
a
s
p
ro
v
e
n
e
ffe
c
ti
v
e
in
p
re
d
icti
n
g
c
re
d
it
risk
a
d
a
p
ti
v
e
l
y
a
n
d
a
c
c
u
ra
tely
,
a
n
d
c
a
n
b
e
imp
lem
e
n
ted
a
s
a
n
in
telli
g
e
n
t
so
l
u
t
io
n
i
n
a
n
a
u
to
m
a
ted
c
re
d
it
d
e
c
isio
n
-
m
a
k
i
n
g
sy
ste
m
b
a
se
d
o
n
m
u
lt
i
-
a
tt
rib
u
te d
a
ta.
K
ey
w
o
r
d
s
:
Ad
ap
tiv
e
m
o
d
el
C
r
ed
it r
is
k
Dee
p
lear
n
in
g
Fin
an
cial
r
is
k
Mu
lti
-
attr
ib
u
te
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Al
-
Kh
o
war
izm
i
Dep
ar
tm
en
t o
f
I
n
f
o
r
m
atio
n
T
e
ch
n
o
lo
g
y
,
Facu
lty
o
f
C
o
m
p
u
te
r
Scien
ce
an
d
I
n
f
o
r
m
atio
n
T
ec
h
n
o
lo
g
y
Un
iv
er
s
itas
Mu
h
am
m
ad
iy
ah
Su
m
ater
a
Utar
a
Kap
ten
Mu
k
h
tar
B
asri
St.
No
.
3
,
Me
d
an
,
No
r
th
Su
m
at
r
a
2
0
2
3
8
,
I
n
d
o
n
esia
E
m
ail:
alk
h
o
war
izm
i@
u
m
s
u
.
a
c.
id
1.
I
NT
RO
D
UCT
I
O
N
T
h
e
d
ev
elo
p
m
en
t
o
f
in
f
o
r
m
a
tio
n
an
d
co
m
m
u
n
icatio
n
tech
n
o
lo
g
y
i
n
th
e
last
d
ec
ad
e
h
as
b
r
o
u
g
h
t
ab
o
u
t
m
ajo
r
tr
a
n
s
f
o
r
m
atio
n
s
in
v
ar
io
u
s
s
ec
to
r
s
o
f
life
,
in
c
lu
d
in
g
th
e
f
in
a
n
cial
s
ec
to
r
.
Dig
italizatio
n
in
th
e
f
in
an
cial
s
ec
to
r
h
as
d
r
iv
e
n
t
h
e
b
ir
th
o
f
v
ar
i
o
u
s
in
n
o
v
ati
o
n
s
,
r
an
g
in
g
f
r
o
m
d
i
g
ital
b
a
n
k
in
g
an
d
f
in
an
cial
tech
n
o
lo
g
y
to
b
lo
c
k
ch
ain
-
b
ased
p
a
y
m
en
t
s
y
s
tem
s
[
1
]
,
[
2
]
.
On
e
im
p
o
r
tan
t
asp
ec
t
o
f
f
in
a
n
cial
m
an
ag
e
m
en
t,
b
o
th
in
tr
ad
itio
n
al
f
in
an
cial
in
s
titu
tio
n
s
an
d
f
in
tech
,
is
r
is
k
m
an
ag
em
en
t,
esp
ec
ially
cr
e
d
it
r
is
k
[
3
]
.
C
r
ed
it
r
is
k
is
th
e
r
is
k
th
at
ar
is
es
f
r
o
m
th
e
f
ailu
r
e
o
f
t
h
e
b
o
r
r
o
wer
t
o
f
u
lf
ill
d
eb
t
p
ay
m
en
t
o
b
lig
ati
o
n
s
ac
co
r
d
in
g
to
t
h
e
ag
r
ee
m
en
t.
T
h
is
r
is
k
is
o
n
e
o
f
th
e
m
ain
ca
u
s
es
o
f
lo
s
s
es
in
th
e
b
an
k
in
g
s
ec
to
r
an
d
i
s
a
m
ajo
r
f
ac
to
r
in
f
in
an
cial
s
y
s
tem
ic
f
ailu
r
e
if
n
o
t
m
an
ag
ed
p
r
o
p
e
r
ly
[
4
]
.
T
r
ad
itio
n
ally
,
f
in
an
cial
in
s
titu
t
io
n
s
u
s
e
s
tati
s
tical
ap
p
r
o
ac
h
es
s
u
ch
as
lo
g
is
tic
r
eg
r
ess
io
n
o
r
cr
e
d
it
s
co
r
in
g
to
ass
ess
an
in
d
iv
id
u
al'
s
cr
ed
itwo
r
th
in
ess
[
5
]
,
[
6
]
.
T
h
u
s
,
with
th
e
in
c
r
ea
s
in
g
co
m
p
lex
ity
o
f
d
ata
an
d
in
d
iv
id
u
al
f
i
n
an
cial
b
eh
av
io
r
,
t
h
ese
m
eth
o
d
s
ar
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
Dee
p
r
ein
fo
r
ce
men
t le
a
r
n
in
g
-
b
a
s
ed
cred
it ri
s
k
p
r
ed
ictio
n
mo
d
el
fo
r
mu
lti
-
a
ttr
ib
u
te
d
a
ta
(
Al
-
K
h
o
w
a
r
iz
mi
)
4625
b
ec
o
m
in
g
less
ef
f
ec
tiv
e.
B
o
r
r
o
wer
d
ata
is
n
o
lo
n
g
er
lim
ite
d
to
f
o
r
m
al
f
in
an
cial
d
ata,
b
u
t
h
as
ex
p
an
d
e
d
to
m
u
lti
-
attr
ib
u
te
d
ata
s
u
ch
as
s
o
cial
m
ed
ia
h
is
to
r
y
,
elec
tr
o
n
ic
tr
an
s
ac
tio
n
d
ata,
g
e
o
g
r
a
p
h
ic
l
o
ca
tio
n
,
an
d
d
ig
ital
d
ev
ice
u
s
ag
e
h
a
b
its
[
7
]
.
T
h
is
c
r
ea
tes
ch
allen
g
es
an
d
o
p
p
o
r
tu
n
ities
in
cr
ea
tin
g
m
o
r
e
ac
c
u
r
at
e
an
d
ad
ap
tiv
e
r
is
k
p
r
ed
ictio
n
m
o
d
els
[
8
]
.
T
h
e
u
s
e
o
f
ar
tific
ial
in
tellig
en
ce
(
AI
)
tech
n
o
l
o
g
y
,
esp
ec
ially
m
ac
h
in
e
lear
n
in
g
an
d
d
ee
p
le
ar
n
in
g
,
is
a
m
ajo
r
b
r
ea
k
th
r
o
u
g
h
in
th
is
f
ield
.
AI
m
o
d
els
h
av
e
th
e
ab
ilit
y
to
lear
n
co
m
p
lex
p
atter
n
s
f
r
o
m
b
ig
d
ata
th
at
ca
n
n
o
t
b
e
ca
p
tu
r
ed
b
y
co
n
v
e
n
tio
n
al
s
tatis
tical
m
o
d
els.
I
n
th
e
co
n
tex
t
o
f
cr
ed
it
r
is
k
p
r
e
d
ictio
n
,
AI
allo
ws
f
in
an
cial
in
s
titu
tio
n
s
to
an
aly
ze
m
u
ltip
le
attr
ib
u
tes
at
o
n
ce
an
d
p
r
o
d
u
ce
m
o
r
e
p
er
s
o
n
alize
d
,
f
aster
,
an
d
m
o
r
e
r
elev
an
t
r
is
k
p
r
e
d
ictio
n
s
.
A
p
r
ed
ictiv
e
a
p
p
r
o
ac
h
alo
n
e
is
n
o
t
en
o
u
g
h
[
9
]
,
[
1
0
]
.
I
n
th
e
m
id
s
t
o
f
ch
an
g
in
g
m
ar
k
et
an
d
d
e
b
to
r
b
eh
a
v
io
r
d
y
n
am
ics,
s
tatic
p
r
ed
ictiv
e
m
o
d
els
b
ec
o
m
e
less
ef
f
ec
tiv
e
in
t
h
e
lo
n
g
ter
m
.
I
n
th
is
co
n
tex
t,
th
e
d
ee
p
r
ei
n
f
o
r
ce
m
e
n
t
lear
n
in
g
(
DR
L
)
m
eth
o
d
b
e
co
m
es
v
er
y
r
elev
an
t
[
1
1
]
.
DR
L
is
a
co
m
b
in
atio
n
o
f
r
ei
n
f
o
r
ce
m
en
t
lear
n
in
g
an
d
d
ee
p
lear
n
in
g
.
I
n
DR
L
,
th
e
ag
en
t
lear
n
s
th
r
o
u
g
h
d
ir
ec
t
i
n
ter
ac
tio
n
with
th
e
en
v
ir
o
n
m
en
t,
wh
e
r
e
it
tak
es
a
ctio
n
s
,
r
ec
eiv
es
f
ee
d
b
ac
k
i
n
th
e
f
o
r
m
o
f
r
ewa
r
d
s
,
a
n
d
u
p
d
ate
s
its
s
tr
ateg
y
b
ased
o
n
th
at
ex
p
er
ien
ce
[
1
2
]
.
I
n
t
h
e
ca
s
e
o
f
cr
ed
it
r
is
k
,
t
h
e
ag
en
t
ca
n
lear
n
th
e
p
atter
n
o
f
cr
e
d
it
d
ec
is
io
n
-
m
ak
in
g
o
v
er
tim
e,
im
p
r
o
v
i
n
g
its
r
is
k
ass
es
s
m
en
t
p
o
licy
b
ased
o
n
th
e
ac
tu
al
o
u
tco
m
e
o
f
cr
ed
i
t
p
ay
m
en
ts
b
y
t
h
e
b
o
r
r
o
we
r
[
1
3
]
,
[
1
4
]
.
T
h
en
t
h
e
in
teg
r
atio
n
o
f
m
u
lti
-
attr
ib
u
te
d
ata
b
ec
o
m
es
a
s
tr
en
g
t
h
in
t
h
e
DR
L
-
b
ased
r
is
k
p
r
ed
ictio
n
s
y
s
tem
[
1
5
]
.
M
u
lti
-
attr
ib
u
te
d
ata
co
v
er
s
v
ar
io
u
s
d
im
en
s
io
n
s
th
at
ar
e
r
elev
a
n
t
to
th
e
ch
ar
ac
ter
is
tics
an
d
b
eh
av
i
o
r
o
f
b
o
r
r
o
we
r
s
,
s
u
ch
as
ag
e,
in
co
m
e,
ty
p
e
o
f
em
p
lo
y
m
e
n
t,
m
ar
ital
s
tatu
s
,
l
o
an
p
u
r
p
o
s
e,
lo
an
d
u
r
atio
n
,
d
e
b
t
-
to
-
in
c
o
m
e
r
atio
,
an
d
p
r
e
v
io
u
s
p
a
y
m
en
t
tr
ac
k
r
ec
o
r
d
[
1
6
]
.
T
h
is
co
m
b
in
atio
n
o
f
d
ata,
if
m
an
ag
e
d
p
r
o
p
er
l
y
,
can
p
r
o
v
id
e
a
h
o
lis
tic
an
d
m
o
r
e
co
n
tex
tu
al
p
ictu
r
e
o
f
cr
ed
it r
is
k
[
1
7
]
,
[
1
8
]
.
E
f
f
ec
tiv
e
cr
e
d
it
r
is
k
m
an
ag
e
m
en
t
is
an
ess
en
tial
f
o
u
n
d
at
io
n
f
o
r
f
in
an
cial
s
y
s
tem
s
tab
ilit
y
[
1
9
]
.
Am
id
s
t
th
e
in
cr
ea
s
in
g
co
m
p
lex
ity
o
f
co
n
s
u
m
er
b
eh
a
v
i
o
r
an
d
th
e
ev
e
r
-
in
cr
ea
s
in
g
v
o
lu
m
e
o
f
d
ata,
co
n
v
en
tio
n
al
ap
p
r
o
ac
h
es
ar
e
in
cr
ea
s
in
g
ly
i
n
ad
eq
u
ate
in
ad
d
r
ess
in
g
th
e
ch
alle
n
g
es
o
f
r
is
k
p
r
ed
ictio
n
.
T
h
e
u
s
e
o
f
d
ee
p
lear
n
in
g
tech
n
o
lo
g
y
,
esp
ec
ially
DR
L
b
ased
o
n
m
u
lti
-
attr
ib
u
te
d
ata,
p
r
esen
ts
an
ad
ap
tiv
e,
ef
f
icien
t,
an
d
in
tellig
en
t
s
o
lu
tio
n
in
m
o
d
elin
g
cr
e
d
it
r
is
k
[
2
0
]
,
[
2
1
]
.
Alth
o
u
g
h
t
h
er
e
ar
e
s
till
ch
al
len
g
es
in
ter
m
s
o
f
im
p
lem
en
tatio
n
an
d
in
ter
p
r
eta
tio
n
,
th
e
p
o
ten
tial
b
en
ef
its
o
f
f
er
ed
ar
e
en
o
r
m
o
u
s
.
W
ith
th
e
s
u
p
p
o
r
t
o
f
ad
eq
u
ate
d
ata
in
f
r
astru
ctu
r
e
an
d
p
r
o
g
r
ess
iv
e
r
eg
u
lato
r
y
p
o
licies,
DR
L
-
b
ased
p
r
ed
ictio
n
s
y
s
tem
s
ca
n
b
e
a
s
tr
ateg
i
c
in
n
o
v
atio
n
to
in
cr
ea
s
e
f
in
a
n
cial
in
clu
s
io
n
,
s
tr
en
g
th
en
r
i
s
k
m
an
ag
em
e
n
t,
a
n
d
d
r
iv
e
s
u
s
tain
ab
le
d
ig
ital
tr
an
s
f
o
r
m
atio
n
in
th
e
f
i
n
an
cial
s
ec
to
r
.
2.
L
I
T
E
R
AT
U
RE
R
E
VI
E
W
2
.
1
.
F
ina
ncia
l r
is
k
Fin
an
cial
r
is
k
is
th
e
p
o
s
s
ib
ilit
y
o
f
lo
s
s
r
elate
d
to
f
in
an
cial
ac
tiv
ities
,
s
u
ch
as
in
v
estme
n
t,
len
d
in
g
,
tr
ad
in
g
,
o
r
o
th
er
b
u
s
in
ess
ac
tiv
ities
.
T
h
is
r
is
k
ar
is
es
b
ec
au
s
e
o
f
u
n
ce
r
tain
ty
in
t
h
e
ex
p
ec
te
d
r
esu
lts
,
b
o
th
f
r
o
m
in
ter
n
al
an
d
ex
ter
n
al
f
ac
to
r
s
.
I
n
th
e
w
o
r
ld
o
f
f
in
an
ce
,
r
is
k
is
an
u
n
av
o
id
ab
le
p
ar
t,
s
o
it
m
u
s
t
b
e
id
en
tifie
d
,
an
aly
ze
d
,
an
d
m
a
n
ag
ed
p
r
o
p
e
r
ly
[
2
2
]
.
Fin
an
cial
r
is
k
ca
n
b
e
d
iv
id
ed
in
to
s
ev
er
al
m
ain
ty
p
es.
First,
m
ar
k
et
r
is
k
,
wh
ich
is
th
e
r
is
k
th
at
ar
is
es
d
u
e
to
p
r
ice
f
lu
ctu
atio
n
s
in
th
e
m
ar
k
et
s
u
ch
as
s
to
ck
s
,
b
o
n
d
s
,
o
r
co
m
m
o
d
ities
.
Seco
n
d
,
c
r
ed
it
r
is
k
,
wh
ich
is
th
e
p
o
s
s
ib
ilit
y
th
at
th
e
d
eb
to
r
f
ails
to
f
u
lf
il
l
th
eir
o
b
lig
atio
n
s
.
T
h
ir
d
,
liq
u
id
ity
r
is
k
,
wh
ich
o
c
cu
r
s
wh
en
a
co
m
p
an
y
o
r
in
d
iv
id
u
al
is
u
n
a
b
le
to
f
u
lf
ill
f
in
a
n
c
ial
o
b
lig
atio
n
s
d
u
e
to
d
if
f
icu
lty
co
n
v
er
tin
g
ass
ets
in
to
ca
s
h
.
Fo
u
r
th
,
o
p
er
atio
n
al
r
is
k
,
wh
ich
co
m
es
f
r
o
m
in
ter
n
al
co
m
p
an
y
f
ailu
r
es su
ch
as h
u
m
a
n
,
s
y
s
tem
,
o
r
p
r
o
ce
s
s
er
r
o
r
s
[
2
3
]
.
2
.
2
.
M
ulti
-
a
t
t
ribute
T
h
e
u
s
e
o
f
d
if
f
e
r
en
t
f
ea
tu
r
es
o
r
ch
ar
ac
te
r
is
tics
o
f
b
o
r
r
o
w
er
d
ata
to
b
u
ild
a
c
r
ed
it
r
is
k
p
r
e
d
ictio
n
m
o
d
el.
T
h
ese
attr
ib
u
tes
ca
n
in
clu
d
e
d
em
o
g
r
ap
h
ic
d
ata
(
s
u
ch
as
ag
e,
em
p
lo
y
m
en
t,
an
d
ed
u
ca
tio
n
)
,
f
i
n
an
cial
d
ata
(
s
u
ch
as
in
co
m
e
,
to
tal
d
e
b
t,
an
d
in
s
tallm
en
t
r
atio
)
,
a
n
d
cr
ed
it
h
is
to
r
y
(
s
u
ch
as
p
ay
m
e
n
t
h
is
to
r
y
an
d
cr
ed
it
ca
r
d
u
s
ag
e)
.
E
ac
h
attr
ib
u
te
c
ar
r
ies
u
n
iq
u
e
in
f
o
r
m
atio
n
th
a
t
ca
n
h
elp
in
e
v
alu
atin
g
a
b
o
r
r
o
wer
'
s
ab
ilit
y
an
d
willin
g
n
ess
to
m
ee
t
th
eir
f
i
n
a
n
cial
o
b
lig
atio
n
s
[
2
4
]
.
B
y
co
n
s
id
er
in
g
m
u
ltip
le
attr
ib
u
tes
s
im
u
ltan
eo
u
s
ly
,
th
e
p
r
ed
ictio
n
m
o
d
el
b
ec
o
m
es
m
o
r
e
ac
cu
r
ate
an
d
is
ab
le
to
ca
p
tu
r
e
th
e
co
m
p
lex
ity
o
f
th
e
r
e
latio
n
s
h
ip
s
b
etwe
en
f
ac
to
r
s
th
at
in
f
lu
en
ce
cr
ed
it
r
i
s
k
.
I
n
th
e
DR
L
-
b
ased
ap
p
r
o
a
ch
,
th
is
m
u
lti
-
attr
ib
u
te
d
ata
is
u
s
ed
to
g
u
id
e
th
e
lear
n
in
g
ag
en
t in
m
a
k
in
g
th
e
b
est d
ec
is
io
n
,
n
am
ely
p
r
ed
ictin
g
wh
eth
er
a
b
o
r
r
o
wer
is
h
ig
h
o
r
lo
w
r
is
k
.
T
h
e
u
s
e
o
f
m
u
lti
-
attr
ib
u
tes
allo
ws
th
e
m
o
d
el
to
m
ak
e
m
o
r
e
co
n
tex
tu
al
an
d
ad
ap
tiv
e
d
ec
is
io
n
s
to
co
m
p
lex
p
atter
n
s
in
h
is
to
r
ical
d
ata
[
2
5
]
.
2
.
3
.
Dee
p r
einf
o
rc
e
m
ent
lea
rning
DR
L
is
a
b
r
an
ch
o
f
m
ac
h
in
e
l
ea
r
n
in
g
th
at
co
m
b
in
es th
e
p
r
in
cip
les o
f
r
ein
f
o
r
ce
m
e
n
t le
ar
n
i
n
g
with
th
e
p
o
wer
o
f
d
ee
p
lear
n
i
n
g
[
2
6
]
,
[
2
7
]
.
DR
L
allo
ws
an
ag
en
t p
r
o
g
r
am
o
r
m
o
d
el
to
lear
n
t
o
m
ak
e
o
p
tim
al
d
ec
is
io
n
s
in
an
en
v
ir
o
n
m
en
t
b
y
tr
ial
an
d
er
r
o
r
a
n
d
b
y
r
ec
eiv
in
g
f
ee
d
b
ac
k
in
th
e
f
o
r
m
o
f
r
ewa
r
d
s
o
r
p
u
n
is
h
m
en
ts
.
I
n
tr
ad
itio
n
al
r
ein
f
o
r
ce
m
en
t
lear
n
in
g
,
ag
en
ts
lear
n
f
r
o
m
in
te
r
ac
tio
n
s
with
th
e
en
v
ir
o
n
m
en
t
b
as
ed
o
n
th
e
p
r
in
cip
le
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
5
,
No
.
5
,
Octo
b
er
2
0
2
6
:
4
6
2
4
-
4
6
3
4
4626
o
f
r
ein
f
o
r
ce
m
en
t,
b
u
t
o
f
ten
h
av
e
lim
itatio
n
s
in
p
r
o
ce
s
s
in
g
co
m
p
lex
o
r
h
ig
h
-
d
im
en
s
io
n
al
d
ata
[
2
8
]
.
T
h
is
is
wh
er
e
d
ee
p
lear
n
i
n
g
c
o
m
es
i
n
:
d
ee
p
n
eu
r
al
n
etwo
r
k
s
ar
e
u
s
ed
to
a
p
p
r
o
x
im
ate
im
p
o
r
ta
n
t
f
u
n
ctio
n
s
,
s
u
ch
as
v
alu
e
f
u
n
ctio
n
s
,
p
o
licy
f
u
n
ct
io
n
s
,
o
r
en
v
ir
o
n
m
e
n
tal
m
o
d
e
l
f
u
n
ctio
n
s
.
T
h
is
allo
ws
DR
L
to
h
a
n
d
le
v
er
y
co
m
p
lex
in
p
u
ts
s
u
ch
as im
ag
e
s
,
s
o
u
n
d
s
,
o
r
la
r
g
e
m
u
lti
-
attr
ib
u
te
d
ata.
2
.
4
.
P
re
v
io
us
re
s
ea
rc
h
T
h
e
r
esear
ch
th
at
h
as
b
ee
n
co
n
d
u
cte
d
is
as
f
o
llo
ws.
Ac
co
r
d
in
g
t
o
Kiziele
wicz
et
a
l.
[
2
9
]
,
th
e
in
tellig
en
t
ch
ar
ac
ter
is
tic
o
b
je
cts
m
eth
o
d
(
I
NC
OM
E
)
is
a
n
ew
m
eth
o
d
in
m
u
lti
-
c
r
iter
ia
d
ec
is
io
n
an
aly
s
is
(
MCDA)
b
ased
o
n
d
ata
k
n
o
wled
g
e.
Dif
f
er
e
n
t
f
r
o
m
t
r
ad
i
tio
n
al
MCDA
th
at
r
elies
o
n
s
u
b
jectiv
e
in
p
u
t,
I
NC
OM
E
u
s
es
d
ata
an
aly
s
is
t
o
id
en
tify
im
p
o
r
tan
t c
h
ar
ac
ter
i
s
tics
o
f
o
b
jects a
n
d
b
u
ild
d
ec
is
io
n
s
au
to
m
atica
lly
.
I
n
th
is
way
,
th
e
d
ec
is
io
n
-
m
ak
in
g
p
r
o
ce
s
s
b
ec
o
m
es
m
o
r
e
o
b
jectiv
e,
co
n
s
is
ten
t,
an
d
ac
cu
r
a
te.
T
h
is
m
eth
o
d
is
ef
f
ec
tiv
e
in
co
m
p
lex
p
r
o
b
lem
s
with
lar
g
e
o
r
in
c
o
m
p
lete
d
at
a,
s
u
ch
as
in
r
is
k
a
n
aly
s
is
an
d
f
in
an
cial
p
lan
n
in
g
.
W
ei
et
a
l.
[
3
0
]
d
e
v
elo
p
ed
t
h
e
s
p
ar
r
o
w
–
o
p
p
o
s
itio
n
–
lear
n
i
n
g
s
alp
s
war
m
alg
o
r
ith
m
(
S
OL
SS
A
)
alg
o
r
ith
m
(
a
co
m
b
in
atio
n
o
f
s
p
ar
r
o
w
s
ea
r
ch
,
s
alp
s
war
m
,
o
p
p
o
s
itio
n
-
b
ased
lear
n
in
g
(
OB
L
)
,
a
n
d
later
al
m
u
tatio
n
s
tr
ateg
y
)
to
o
p
tim
ize
th
e
ca
teg
o
r
ical
b
o
o
s
tin
g
(
C
atB
o
o
s
t)
m
o
d
el
i
n
ca
r
d
io
v
ascu
lar
d
is
ea
s
e
(
C
V
D)
r
is
k
ass
es
s
m
en
t.
C
o
m
p
ar
ed
with
s
i
x
o
th
er
m
ac
h
in
e
lear
n
i
n
g
m
o
d
els
an
d
f
o
u
r
o
p
tim
izatio
n
alg
o
r
ith
m
s
,
SOLSSA
-
C
atB
o
o
s
t
s
h
o
wed
th
e
b
est
p
r
ed
ictio
n
r
esu
lts
with
F1
-
s
co
r
es
o
f
9
0
%
a
n
d
8
1
.
5
1
%
o
n
t
wo
C
VD
d
atasets
,
h
elp
in
g
to
im
p
r
o
v
e
th
e
ac
cu
r
a
cy
o
f
p
atien
t r
is
k
ass
es
s
m
en
t.
B
ar
ly
b
ay
ev
et
a
l.
[
3
1
]
in
tr
o
d
u
ce
s
a
n
ew
m
o
d
el
f
o
r
f
lex
ib
le
an
d
co
m
p
r
e
h
en
s
iv
e
i
n
f
o
r
m
atio
n
s
ec
u
r
ity
r
is
k
ass
ess
m
en
t.
T
h
is
m
o
d
el
in
teg
r
ate
s
s
tan
d
ar
d
s
,
ex
p
er
t
k
n
o
wled
g
e,
m
ac
h
in
e
lear
n
in
g
,
an
d
o
n
to
lo
g
y
m
o
d
elin
g
to
co
m
p
r
eh
e
n
s
iv
ely
u
n
d
e
r
s
tan
d
an
d
m
an
ag
e
r
is
k
s
.
Me
th
o
d
s
s
u
ch
as
k
-
m
ea
n
s
cl
u
s
ter
an
aly
s
is
,
m
ac
h
in
e
lear
n
in
g
-
b
ased
p
r
o
g
r
am
m
in
g
,
u
s
e
o
f
h
ea
t
m
a
p
s
f
o
r
v
is
u
aliza
tio
n
,
an
d
d
ev
el
o
p
m
en
t
o
f
o
n
to
l
o
g
ies
f
o
r
r
is
k
c
lass
if
icatio
n
im
p
r
o
v
e
th
e
ef
f
ec
tiv
en
ess
o
f
r
is
k
id
en
tific
atio
n
an
d
m
an
a
g
em
en
t.
T
h
is
in
n
o
v
atio
n
o
f
f
er
s
a
m
o
r
e
r
o
b
u
s
t
an
d
a
d
ap
tiv
e
a
p
p
r
o
ac
h
to
ad
d
r
ess
in
g
th
e
ev
er
-
ev
o
lv
in
g
c
h
allen
g
es o
f
in
f
o
r
m
atio
n
s
ec
u
r
ity
.
3.
M
E
T
H
O
D
T
h
is
p
ap
er
will
u
s
e
m
u
lti
-
attr
ib
u
te
d
ata
th
at
i
n
clu
d
es
b
o
r
r
o
wer
d
e
m
o
g
r
ap
h
ic
attr
ib
u
te
s
,
i
n
co
m
e
attr
ib
u
tes,
cr
ed
it
h
is
to
r
y
attr
ib
u
tes,
cr
ed
it
s
co
r
e
attr
ib
u
tes,
y
ea
r
s
o
f
s
er
v
ice
attr
ib
u
tes
,
an
d
r
is
k
attr
ib
u
tes.
T
h
e
d
ata
u
s
ed
is
a
co
llectio
n
o
f
h
is
to
r
ical
cr
ed
it
d
ata
th
at
in
cl
u
d
e
s
v
ar
io
u
s
b
o
r
r
o
wer
c
h
ar
ac
ter
is
tics
f
r
o
m
f
in
an
cial
in
s
titu
tio
n
s
.
T
h
is
s
tu
d
y
aim
s
to
b
u
ild
a
m
o
d
el
th
at
ca
n
p
r
e
d
ict
cr
ed
it
r
is
k
b
ased
o
n
m
u
lti
-
a
ttrib
u
te
d
ata.
I
n
th
is
s
tu
d
y
th
er
e
is
a
m
eth
o
d
o
lo
g
y
co
n
s
is
tin
g
o
f
d
ata
p
r
o
ce
s
s
in
g
s
tag
es,
f
ea
tu
r
e
v
ec
to
r
r
ep
r
ese
n
tatio
n
,
ap
p
licatio
n
o
f
th
e
DR
L
alg
o
r
ith
m
,
an
d
m
o
d
el
ev
alu
atio
n
,
wh
ich
ar
e
ex
p
l
ain
ed
in
Fig
u
r
e
1
.
D
a
ta
M
u
l
ti
-
A
t
r
i
b
u
t
D
a
ta
P
r
o
c
e
s
s
i
n
g
A
c
c
u
ra
c
y
,
P
re
c
i
s
i
o
n
,
Re
c
a
l
l
,
F
1
-
s
c
o
r
e
D
e
e
p
R
e
i
n
f
o
r
c
e
m
e
n
t
L
e
a
r
n
i
n
g
S
t
a
t
e
(S
)
A
c
t
i
o
n
(
A
)
RM
S
E
E
v
a
l
u
a
t
i
o
n
M
o
d
e
l
F
e
a
t
u
r
e
V
e
c
t
o
r
R
e
p
r
e
s
e
n
t
a
t
i
o
n
C
l
e
a
n
i
n
g
D
a
t
a
T
r
a
n
s
f
o
r
m
a
t
i
o
n
D
a
t
a
E
x
p
l
o
r
a
t
o
r
y
D
a
t
a
A
n
a
l
y
s
i
s
Ca
t
e
g
o
r
i
c
a
l
A
t
t
ri
b
u
t
e
E
n
c
o
d
i
n
g
N
o
rm
a
l
i
z
a
t
i
o
n
a
n
d
S
t
a
n
d
a
rd
i
z
a
t
i
o
n
F
e
a
t
u
re
S
e
l
e
c
t
i
o
n
V
e
c
t
o
ri
z
a
t
i
o
n
a
n
d
M
e
r
g
i
n
g
Re
w
a
rd
(R)
V
a
l
u
e
F
u
n
c
t
i
o
n
D
a
t
a
s
e
t
L
e
n
d
i
n
g
D
e
m
o
g
ra
p
h
i
c
s
J
o
b
a
n
d
E
c
o
n
o
m
i
c
a
t
t
ri
b
u
t
e
s
F
i
n
a
n
c
i
a
l
A
t
t
ri
b
u
t
e
s
Cre
d
i
t
H
i
s
t
o
ry
A
t
t
ri
b
u
t
e
s
Fig
u
r
e
1
.
R
esear
ch
ar
c
h
itectu
r
e
f
o
r
cr
e
d
it r
is
k
p
r
e
d
ictio
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
Dee
p
r
ein
fo
r
ce
men
t le
a
r
n
in
g
-
b
a
s
ed
cred
it ri
s
k
p
r
ed
ictio
n
mo
d
el
fo
r
mu
lti
-
a
ttr
ib
u
te
d
a
ta
(
Al
-
K
h
o
w
a
r
iz
mi
)
4627
Fig
u
r
e
1
ex
p
lain
s
wh
at
will
b
e
p
r
o
ce
s
s
ed
in
b
u
ild
in
g
a
m
o
d
el
th
at
ca
n
p
r
ed
ict
cr
ed
it
r
is
k
b
ased
o
n
m
u
lti
-
attr
ib
u
te
d
ata.
T
h
e
f
o
llo
win
g
is
an
ex
p
lan
atio
n
o
f
th
e
r
esear
ch
ar
ch
itectu
r
e:
i)
T
h
is
s
tu
d
y
m
o
d
els
cr
ed
it
r
is
k
p
r
ed
ictio
n
with
a
m
u
lti
-
attr
i
b
u
te
d
ata
-
b
ased
ap
p
r
o
ac
h
u
s
i
n
g
DR
L
.
T
h
e
d
ataset
u
s
ed
in
th
is
s
tu
d
y
c
o
n
s
is
ts
o
f
s
ev
er
al
attr
ib
u
te
c
ateg
o
r
ies,
s
u
ch
as
b
o
r
r
o
wer
d
em
o
g
r
ap
h
ic
attr
ib
u
tes,
i
n
co
m
e
attr
ib
u
tes,
cr
ed
it
h
is
to
r
y
attr
ib
u
tes,
cr
ed
i
t
s
co
r
e
attr
ib
u
tes,
y
ea
r
s
o
f
w
o
r
k
attr
ib
u
tes
,
an
d
r
is
k
attr
ib
u
tes.
ii)
T
h
e
d
ata
p
r
o
ce
s
s
in
g
b
eg
i
n
s
with
th
e
d
ata
p
r
o
ce
s
s
in
g
s
tag
e,
wh
ich
in
clu
d
es
d
ata
clea
n
i
n
g
to
r
e
m
o
v
e
d
u
p
licates
an
d
co
r
r
ec
tin
g
m
is
s
in
g
v
alu
es,
d
ata
tr
a
n
s
f
o
r
m
ati
o
n
to
ch
a
n
g
e
th
e
f
o
r
m
at
an
d
co
m
b
in
e
d
at
a
f
r
o
m
v
ar
i
o
u
s
s
o
u
r
ce
s
,
an
d
ex
p
lo
r
ato
r
y
d
ata
an
aly
s
is
(
E
DA
)
to
u
n
d
er
s
tan
d
th
e
d
is
tr
ib
u
tio
n
p
atter
n
s
an
d
r
elatio
n
s
h
ip
s
b
etwe
en
attr
ib
u
tes
in
th
e
d
ataset.
T
h
is
s
tag
e
i
s
v
er
y
im
p
o
r
ta
n
t
to
en
s
u
r
e
th
at
th
e
d
ata
u
s
ed
is
o
f
h
ig
h
q
u
ality
an
d
r
ep
r
esen
tativ
e.
iii)
Nex
t,
d
ata
r
ep
r
esen
tatio
n
is
ca
r
r
ied
o
u
t in
t
o
f
ea
tu
r
e
v
ec
to
r
r
e
p
r
esen
tatio
n
.
T
h
is
s
tag
e
in
clu
d
es c
ateg
o
r
ical
attr
ib
u
te
en
co
d
in
g
to
c
h
an
g
e
n
o
n
-
n
u
m
er
ic
attr
ib
u
tes
in
to
n
u
m
b
er
s
,
n
o
r
m
aliza
tio
n
an
d
s
tan
d
ar
d
izatio
n
t
o
alig
n
th
e
d
ata
s
ca
le,
f
ea
tu
r
e
s
e
lectio
n
to
s
elec
t
th
e
m
o
s
t
r
ele
v
an
t
attr
ib
u
tes
to
c
r
ed
it
r
is
k
p
r
ed
ictio
n
,
an
d
v
ec
to
r
izatio
n
a
n
d
m
er
g
i
n
g
t
o
co
m
b
i
n
e
all
f
ea
t
u
r
es
in
to
o
n
e
in
p
u
t
f
o
r
m
th
at
ca
n
b
e
p
r
o
ce
s
s
ed
b
y
th
e
m
ac
h
in
e
lear
n
in
g
m
o
d
el.
iv
)
At
th
e
m
o
d
ellin
g
s
tag
e,
th
e
DR
L
ap
p
r
o
ac
h
is
u
s
ed
,
wh
ic
h
d
ef
i
n
es
s
tate
(
S)
as
a
r
e
p
r
e
s
en
tatio
n
o
f
th
e
cu
r
r
en
t
d
ata
c
o
n
d
itio
n
,
ac
tio
n
(
A)
as
a
p
r
e
d
ictio
n
o
f
ac
tio
n
o
n
cr
e
d
it
r
is
k
,
r
ewa
r
d
(
R
)
as
f
e
ed
b
ac
k
o
n
th
e
q
u
ality
o
f
p
r
ed
ictio
n
,
an
d
a
v
alu
e
f
u
n
ctio
n
to
esti
m
ate
th
e
lo
n
g
-
ter
m
v
alu
e
o
f
th
e
d
ec
is
io
n
tak
e
n
.
T
h
is
m
o
d
el
lear
n
s
ad
a
p
tiv
ely
f
r
o
m
co
n
tin
u
o
u
s
in
ter
ac
tio
n
with
d
a
ta.
v)
Fin
ally
,
th
e
m
o
d
el
is
ev
alu
ated
u
s
in
g
v
ar
io
u
s
p
er
f
o
r
m
an
ce
m
etr
ics,
in
clu
d
in
g
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
an
d
F1
-
s
co
r
e
to
m
ea
s
u
r
e
class
if
icatio
n
p
er
f
o
r
m
an
ce
,
an
d
r
o
o
t
m
ea
n
s
q
u
ar
e
er
r
o
r
(
R
MSE
)
t
o
m
ea
s
u
r
e
th
e
lev
el
o
f
p
r
ed
ictio
n
e
r
r
o
r
.
T
h
is
ev
alu
atio
n
aim
s
to
en
s
u
r
e
th
a
t
th
e
r
esu
ltin
g
m
o
d
el
ca
n
b
e
u
s
ed
to
p
r
ed
ict
cr
ed
it r
is
k
ac
cu
r
ately
.
I
n
th
is
s
tu
d
y
,
th
er
e
will
b
e
a
d
ataset
th
at
will
b
e
u
s
ed
to
b
u
il
d
a
m
o
d
el
,
wh
ic
h
ca
n
b
e
s
ee
n
in
T
ab
le
1
.
T
ab
le
1
s
h
o
ws
th
at
th
is
d
ata
is
a
co
llectio
n
o
f
cu
s
to
m
er
p
r
o
f
iles
u
s
ed
f
o
r
cr
e
d
it
s
co
r
i
n
g
,
wh
e
r
e
f
i
n
an
cial
in
s
titu
tio
n
s
ev
alu
ate
an
i
n
d
i
v
id
u
al'
s
elig
ib
ilit
y
f
o
r
a
lo
a
n
b
ased
o
n
t
h
eir
f
in
an
cial
h
is
to
r
y
an
d
p
er
s
o
n
al
s
tab
ilit
y
.
Var
iab
les
s
u
ch
as
ag
e,
in
co
m
e,
an
d
len
g
th
o
f
em
p
lo
y
m
e
n
t
p
r
o
v
i
d
e
a
g
en
er
al
o
v
er
v
iew
o
f
an
ap
p
lican
t'
s
ec
o
n
o
m
ic
ca
p
ac
ity
an
d
s
tab
ilit
y
,
wh
ile
th
e
lo
a
n
a
m
o
u
n
t
s
er
v
es
as
th
e
b
asis
f
o
r
m
ea
s
u
r
in
g
th
e
d
eb
t
b
u
r
d
e
n
,
th
e
y
will
in
cu
r
.
T
h
e
m
o
s
t
im
p
o
r
tan
t f
ac
to
r
in
th
is
d
at
a
is
th
e
cr
ed
it
s
co
r
e,
wh
ic
h
s
er
v
es
as
an
in
d
icato
r
o
f
a
cu
s
to
m
er
'
s
p
ast
b
eh
a
v
io
r
in
m
ee
tin
g
th
ei
r
f
i
n
an
cial
o
b
lig
atio
n
s
.
T
h
r
o
u
g
h
th
e
r
is
k
c
o
lu
m
n
,
w
h
ich
h
as
a
v
alu
e
o
f
ze
r
o
o
r
o
n
e,
t
h
is
d
ata
s
h
o
ws
a
p
atter
n
wh
e
r
e
cu
s
to
m
er
s
with
h
ig
h
c
r
ed
it
s
co
r
es
an
d
lo
n
g
ten
u
r
e
ar
e
g
en
er
ally
co
n
s
id
er
e
d
s
af
e,
wh
i
le
th
o
s
e
with
p
o
o
r
c
r
ed
it
h
is
to
r
ies
o
r
j
o
b
i
n
s
tab
ilit
y
ar
e
ca
teg
o
r
ized
as
h
ig
h
-
r
is
k
cu
s
to
m
er
s
with
a
h
ig
h
p
r
o
b
a
b
i
lity
o
f
d
ef
a
u
lt.
Ov
er
all,
th
is
d
a
taset
o
f
5
,
0
0
0
en
t
r
ies
s
er
v
es
as
th
e
f
o
u
n
d
atio
n
f
o
r
b
u
ild
in
g
a
n
au
to
m
ate
d
p
r
e
d
ictiv
e
m
o
d
el
th
at
ca
n
h
elp
b
an
k
s
m
in
im
ize
lo
s
s
es b
y
s
cr
ee
n
in
g
p
o
ten
tial a
p
p
lican
ts
m
o
r
e
ac
cu
r
ately
an
d
o
b
jectiv
e
ly
.
T
ab
le
1
.
C
r
ed
it r
is
k
p
r
ed
ictio
n
d
ataset
No
A
g
e
I
n
c
o
me
Lo
a
n
a
m
o
u
n
t
C
r
e
d
i
t
sc
o
r
e
Y
e
a
r
s
e
m
p
l
o
y
e
d
R
i
s
k
1
56
1
2
5
1
8
6
1
9
5
4
6
6
6
7
4
0
2
69
5
4
6
7
4
2
0
1
2
9
5
1
9
2
1
3
46
5
5
8
5
4
2
5
9
1
6
3
6
0
0
4
32
6
6
2
7
1
1
2
3
0
3
5
2
2
2
1
5
60
9
3
6
8
8
3
1
5
6
1
6
5
9
5
0
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
1
1
1
1
25
5
8
5
1
8
7
1
8
3
6
2
8
2
0
1
1
1
2
38
1
3
1
0
7
6
2
7
5
7
2
7
6
2
7
0
1
1
1
3
56
3
0
2
6
7
3
5
0
8
0
6
8
1
0
1
1
1
1
4
36
1
1
0
8
2
5
2
1
6
9
5
4
6
8
7
1
1
1
1
5
40
2
1
0
6
2
8
3
1
4
5
1
6
3
1
1
1
1
6
28
1
3
6
6
6
4
2
0
6
2
3
6
7
7
5
0
1
1
1
7
28
7
0
4
4
8
3
1
3
9
4
1
8
7
1
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
4
9
9
8
41
1
1
5
4
6
2
4
8
4
5
6
4
8
9
2
1
4
9
9
9
53
2
6
2
2
9
3
2
2
9
0
6
6
9
0
1
5
0
0
0
57
7
6
7
9
3
1
9
0
8
8
4
4
8
6
1
T
h
en
in
th
is
r
esear
ch
th
er
e
wil
l b
e
a
m
ath
em
atica
l f
o
r
m
u
latio
n
o
f
DR
L
m
o
d
el
as in
(
1
)
an
d
(
2
)
[
3
2
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
5
,
No
.
5
,
Octo
b
er
2
0
2
6
:
4
6
2
4
-
4
6
3
4
4628
(
)
=
[
∑
∞
=
0
(
,
)
|
0
=
]
(
1
)
(
,
)
=
[
∑
∞
=
0
(
,
)
|
0
=
,
0
=
]
(
2
)
I
n
(
1
)
a
n
d
(
2
)
th
er
e
is
a
s
y
m
b
o
l
S
wh
ich
is
a
s
tate
s
p
ac
e
th
at
r
ep
r
esen
ts
th
e
d
ata
o
f
cr
ed
i
t
ap
p
lican
ts
wh
ich
co
n
s
is
ts
o
f
m
u
lti
-
attr
ib
u
tes s
u
ch
as in
co
m
e,
cr
e
d
it h
is
to
r
y
,
a
g
e,
em
p
lo
y
m
en
t statu
s
,
th
en
s
y
m
b
o
l A
is
an
ac
tio
n
s
p
ac
e,
f
o
r
ex
am
p
le
th
e
d
ec
is
io
n
to
ap
p
r
o
v
e
o
r
r
ejec
t
a
cr
ed
it
ap
p
licatio
n
.
Sy
m
b
o
l
R
(
,
)
w
h
ich
is
a
r
ewa
r
d
f
u
n
ctio
n
,
wh
ich
r
ef
lects
th
e
r
esu
lts
o
f
d
ec
is
io
n
m
a
k
in
g
r
e
g
ar
d
in
g
a
cr
e
d
it
ap
p
licatio
n
a
n
d
s
y
m
b
o
l
T
(
,
)
wh
ich
is
th
e
p
r
o
b
ab
ilit
y
o
f
tr
a
n
s
itio
n
f
r
o
m
s
tate
s
to
s
tate
s
′.
Af
ter
d
o
in
g
ac
tio
n
,
th
en
th
e
r
e
is
(
2
)
[
3
3
]
.
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
At
th
is
s
tag
e,
th
e
r
esu
lts
r
elate
d
to
cr
ed
it r
is
k
p
r
ed
ictio
n
m
o
d
elin
g
b
ased
o
n
m
u
lti
-
attr
ib
u
te
d
ata
u
s
in
g
th
e
DR
L
m
eth
o
d
will
b
e
d
is
p
lay
ed
.
I
n
th
is
s
ec
tio
n
,
th
e
d
ata
u
s
ed
co
n
s
is
ts
o
f
5
,
0
0
0
d
ata
th
at
h
av
e
g
o
n
e
th
r
o
u
g
h
th
e
p
r
ep
r
o
ce
s
s
in
g
s
tag
e
to
en
s
u
r
e
d
ata
q
u
ality
an
d
co
n
s
is
ten
cy
.
T
h
en
m
o
d
elin
g
is
ca
r
r
ied
o
u
t
u
s
in
g
th
e
DR
L
ap
p
r
o
ac
h
to
b
u
il
d
a
p
r
ed
ictiv
e
ag
en
t
th
at
is
ab
le
to
lear
n
r
is
k
p
atter
n
s
b
ased
o
n
th
e
av
ailab
le
attr
ib
u
tes.
T
h
e
p
u
r
p
o
s
e
o
f
th
is
s
tu
d
y
is
to
d
e
v
elo
p
a
m
o
d
el
th
at
is
ab
le
to
p
e
r
f
o
r
m
task
s
in
t
er
m
s
o
f
id
e
n
tify
in
g
cr
ed
it
r
is
k
,
s
o
th
at
it
ca
n
im
p
r
o
v
e
p
r
ed
ictio
n
ac
cu
r
ac
y
an
d
a
s
s
is
t
in
m
o
r
e
ef
f
ec
tiv
e
cr
ed
it
d
ec
is
io
n
m
ak
in
g
.
I
n
th
is
ca
s
e,
th
er
e
will b
e
E
DA
an
d
th
e
r
esu
lts
o
f
t
h
e
f
o
llo
win
g
m
o
d
els
.
4
.
1
.
H
ea
t
m
a
p
T
h
is
s
tu
d
y
will
u
s
e
th
e
tech
n
i
q
u
e
to
u
n
d
er
s
tan
d
th
e
r
elatio
n
s
h
ip
b
etwe
en
f
ea
tu
r
es
in
t
h
e
c
r
ed
it
r
is
k
d
ataset,
a
co
r
r
elatio
n
an
aly
s
is
is
ca
r
r
ied
o
u
t
wh
ich
is
v
is
u
ali
ze
d
in
th
e
f
o
r
m
o
f
a
h
ea
tm
ap
.
T
h
is
h
ea
tm
ap
h
elp
s
to
d
escr
ib
e
h
o
w
s
tr
o
n
g
th
e
l
in
ea
r
r
elatio
n
s
h
ip
is
b
etwe
en
o
n
e
f
ea
tu
r
e
a
n
d
an
o
th
er
,
in
clu
d
in
g
th
e
tar
g
et
v
ar
iab
le,
n
am
el
y
cr
ed
it
r
is
k
.
I
n
Fig
u
r
e
2
th
e
r
e
ar
e
th
e
h
ea
tm
a
p
r
esu
lts
f
r
o
m
th
e
d
ata
as f
o
llo
ws.
Fig
u
r
e
2
.
C
o
r
r
elatio
n
h
ea
tm
a
p
f
r
o
m
t
h
e
d
ata
I
n
Fig
u
r
e
2
,
it
is
ex
p
lain
ed
th
at
th
e
co
r
r
elatio
n
h
ea
tm
ap
s
h
o
ws
th
e
r
elatio
n
s
h
ip
b
etwe
en
f
ea
tu
r
es
in
cr
ed
it
r
is
k
d
ata.
C
o
r
r
elatio
n
is
m
ea
s
u
r
ed
f
r
o
m
-
1
to
1
,
wh
er
e
v
alu
es
a
p
p
r
o
ac
h
in
g
1
in
d
icate
a
s
tr
o
n
g
p
o
s
itiv
e
r
elatio
n
s
h
ip
,
wh
ile
v
al
u
es
ap
p
r
o
ac
h
in
g
-
1
i
n
d
icate
a
s
tr
o
n
g
n
eg
ativ
e
r
elatio
n
s
h
ip
.
B
ased
o
n
th
e
v
is
u
aliza
tio
n
r
esu
lts
,
th
e
cr
ed
it_
s
co
r
e
f
ea
tu
r
e
h
as
a
s
tr
o
n
g
n
eg
ativ
e
c
o
r
r
e
latio
n
to
th
e
r
is
k
v
ar
iab
le
with
a
v
alu
e
o
f
-
0
.
7
9
.
T
h
is
in
d
icate
s
th
at
th
e
lo
wer
a
p
er
s
o
n
'
s
cr
ed
it
s
co
r
e,
th
e
h
ig
h
er
th
eir
cr
e
d
it
r
is
k
.
Me
an
wh
ile,
th
e
in
co
m
e
f
ea
tu
r
e
also
s
h
o
ws
a
n
eg
ativ
e
co
r
r
elatio
n
t
o
r
is
k
,
alth
o
u
g
h
wea
k
er
,
with
a
v
alu
e
o
f
-
0
.
1
4
,
wh
ich
m
ea
n
s
th
at
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
Dee
p
r
ein
fo
r
ce
men
t le
a
r
n
in
g
-
b
a
s
ed
cred
it ri
s
k
p
r
ed
ictio
n
mo
d
el
fo
r
mu
lti
-
a
ttr
ib
u
te
d
a
ta
(
Al
-
K
h
o
w
a
r
iz
mi
)
4629
h
ig
h
er
i
n
co
m
e
te
n
d
s
to
b
e
s
lig
h
tly
r
elate
d
t
o
lo
wer
r
is
k
.
Oth
er
f
ea
tu
r
es
s
u
ch
as
ag
e,
lo
an
_
am
o
u
n
t,
an
d
y
ea
r
s
_
em
p
lo
y
ed
s
h
o
w
a
v
er
y
wea
k
co
r
r
elatio
n
to
r
is
k
,
i
n
d
icatin
g
th
at
th
eir
in
f
lu
e
n
c
e
o
n
cr
ed
it
r
is
k
is
r
elativ
ely
s
m
all.
4
.
2
.
Cre
dit
da
t
a
v
is
ua
liza
t
io
n
I
n
th
is
s
ec
tio
n
,
th
e
r
e
will
b
e
a
cr
ed
it
v
is
u
aliza
tio
n
.
T
h
e
im
ag
e
ab
o
v
e
is
th
e
r
esu
lt
o
f
a
v
is
u
aliza
tio
n
o
f
m
u
lti
-
attr
ib
u
te
d
ata
o
f
c
r
ed
it
a
p
p
lican
ts
wh
o
s
e
d
im
en
s
io
n
s
h
av
e
b
ee
n
r
e
d
u
ce
d
u
s
in
g
th
e
p
r
in
cip
al
co
m
p
o
n
en
t
an
aly
s
is
(
PC
A)
tech
n
iq
u
e.
T
h
is
v
is
u
aliza
tio
n
aim
s
to
p
r
o
v
id
e
an
o
v
er
v
iew
o
f
th
e
d
is
tr
ib
u
tio
n
o
f
cr
e
d
it
r
is
k
d
ata
in
two
m
ai
n
d
im
e
n
s
io
n
s
(
PC
A
c
o
m
p
o
n
en
ts
1
an
d
2
)
,
wh
ich
r
ep
r
esen
t
a
lin
ea
r
c
o
m
b
in
atio
n
o
f
v
a
r
io
u
s
f
ea
tu
r
es
s
u
ch
as
in
c
o
m
e,
cr
e
d
it
s
co
r
e,
p
ay
m
e
n
t
h
is
to
r
y
,
ag
e
,
an
d
t
o
tal
ac
tiv
e
lo
a
n
s
.
E
ac
h
p
o
in
t
o
n
th
e
g
r
ap
h
r
ep
r
esen
ts
o
n
e
in
d
i
v
id
u
al,
wh
er
e
th
e
g
r
ee
n
co
lo
r
in
d
icate
s
ap
p
lican
ts
with
lo
w
c
r
ed
it
r
is
k
,
wh
ile
th
e
r
ed
co
lo
r
in
d
icate
s
ap
p
lican
ts
with
h
ig
h
r
is
k
.
T
h
e
v
is
u
aliza
tio
n
o
f
th
e
c
r
ed
it d
ata
is
s
h
o
wn
in
Fig
u
r
e
3
.
Fig
u
r
e
3
s
h
o
ws
th
at
ev
e
n
th
o
u
g
h
th
er
e
is
o
v
er
lap
b
etwe
e
n
th
e
h
ig
h
a
n
d
lo
w
r
is
k
ca
te
g
o
r
ies,
th
e
d
is
tr
ib
u
tio
n
p
atter
n
o
f
ea
ch
g
r
o
u
p
s
till
s
h
o
ws
a
ten
d
en
cy
th
at
ca
n
b
e
u
tili
ze
d
b
y
th
e
p
r
e
d
ictiv
e
m
o
d
el.
I
n
th
e
co
n
tex
t
o
f
th
is
r
esear
ch
,
th
is
v
is
u
aliza
tio
n
is
th
e
b
a
s
is
f
o
r
th
e
ap
p
licatio
n
o
f
th
e
DR
L
alg
o
r
ith
m
wh
ich
aim
s
to
lear
n
th
e
o
p
tim
al
d
ec
is
io
n
-
m
ak
in
g
p
o
licy
in
p
r
ed
ictin
g
cr
ed
it
r
is
k
b
ased
o
n
c
o
m
p
lex
an
d
d
y
n
am
ic
d
ata
.
B
y
m
o
d
elin
g
th
e
r
is
k
e
v
alu
at
io
n
p
r
o
ce
s
s
as
a
Ma
r
k
o
v
d
ec
is
io
n
p
r
o
ce
s
s
(
MD
P)
p
r
o
b
le
m
,
DR
L
is
ab
le
t
o
u
tili
ze
m
u
lti
-
attr
ib
u
te
v
ec
to
r
r
ep
r
esen
tatio
n
as
a
s
tate,
an
d
p
r
o
d
u
ce
p
r
ed
ictiv
e
ac
tio
n
s
in
th
e
f
o
r
m
o
f
cr
ed
it
r
is
k
class
if
icatio
n
.
Fig
u
r
e
3
.
PC
A
-
b
ased
v
is
u
aliza
tio
n
o
f
c
r
ed
it r
is
k
d
ata
4
.
3
.
Dee
p r
einf
o
rc
e
m
ent
lea
rning
m
o
del r
esu
lt
T
h
e
ap
p
licatio
n
o
f
DR
L
in
cr
ed
it
r
is
k
p
r
ed
ictio
n
m
o
d
ellin
g
b
ased
o
n
m
u
lti
-
attr
ib
u
te
d
ata
is
ca
r
r
ied
o
u
t
b
y
d
esig
n
in
g
a
c
r
ed
itwo
r
th
in
ess
ev
alu
atio
n
s
y
s
tem
as
an
ad
ap
tiv
e
d
ec
is
io
n
-
m
ak
i
n
g
p
r
o
ce
s
s
b
y
an
in
tellig
en
t a
g
en
t
[
2
1
]
.
E
ac
h
cr
ed
it a
p
p
lican
t is r
ep
r
esen
ted
a
s
a
s
tate
co
n
s
i
s
tin
g
o
f
v
ar
io
u
s
im
p
o
r
tan
t a
ttrib
u
tes,
s
u
ch
as
in
co
m
e
,
lo
a
n
am
o
u
n
t,
cr
ed
it
s
co
r
e
,
an
d
y
ea
r
o
f
c
o
m
p
letio
n
.
T
h
is
m
o
d
el
th
en
d
ec
i
d
es
th
e
b
est
ac
tio
n
,
n
am
ely
to
ap
p
r
o
v
e,
r
ejec
t,
o
r
r
ev
iew
b
ased
o
n
th
e
h
ig
h
est
v
alu
e
o
f
th
e
ac
tio
n
f
u
n
ctio
n
(
,
;
)
lear
n
ed
th
r
o
u
g
h
a
n
eu
r
al
n
etwo
r
k
.
R
e
war
d
s
ar
e
g
iv
en
b
ased
o
n
th
e
ac
cu
r
ac
y
o
f
c
r
ed
it
r
is
k
p
r
ed
ict
io
n
;
d
ec
is
io
n
s
th
at
ar
e
in
ac
c
o
r
d
an
ce
with
p
a
y
m
en
t
r
ea
lizatio
n
will
r
esu
lt
in
p
o
s
itiv
e
r
ewa
r
d
s
,
wh
ile
wr
o
n
g
d
ec
is
io
n
s
will
b
e
s
u
b
ject
to
p
en
alties.
T
h
e
im
p
lem
en
tatio
n
o
f
th
e
d
ee
p
Q
-
n
etwo
r
k
alg
o
r
ith
m
allo
ws
to
ex
p
lo
r
e
v
ar
io
u
s
p
ar
am
eter
co
m
b
in
atio
n
s
a
n
d
lea
r
n
f
r
o
m
h
is
to
r
ical
ex
p
er
ien
ce
in
t
h
e
d
a
taset
[
3
4
]
.
D
u
r
in
g
th
e
tr
ai
n
in
g
p
r
o
ce
s
s
,
th
e
DR
L
m
o
d
el
g
r
a
d
u
ally
im
p
r
o
v
es
i
ts
p
o
licy
th
r
o
u
g
h
ex
p
lo
r
atio
n
-
ex
p
lo
itatio
n
m
ec
h
an
is
m
s
as
well
as
tar
g
et
n
etwo
r
k
-
b
ased
u
p
d
ates
an
d
ex
p
er
ien
ce
r
e
p
lay
,
wh
ich
ac
ce
ler
ates
co
n
v
er
g
en
ce
s
tab
ilit
y
.
On
e
o
f
th
e
r
ea
l
ad
v
an
tag
es
o
f
th
is
a
p
p
r
o
ac
h
is
its
ab
ilit
y
to
a
d
ap
t
to
d
y
n
am
i
c
an
d
co
m
p
le
x
d
ata,
s
u
ch
as
c
h
an
g
es
in
b
o
r
r
o
wer
b
eh
av
io
r
tr
en
d
s
o
r
ec
o
n
o
m
ic
co
n
d
itio
n
s
.
I
n
th
e
ap
p
licatio
n
o
f
t
h
e
m
o
d
el,
th
er
e
is
a
r
ec
eiv
er
o
p
er
atin
g
ch
ar
ac
ter
is
tic
(
R
OC
)
v
alu
e
wh
ich
ca
n
b
e
s
ee
n
i
n
Fig
u
r
e
4
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
5
,
No
.
5
,
Octo
b
er
2
0
2
6
:
4
6
2
4
-
4
6
3
4
4630
Fig
u
r
e
4
.
R
OC
cu
r
v
e
f
o
r
cr
e
d
i
t r
is
k
class
if
icatio
n
Fig
u
r
e
4
s
h
o
ws
th
e
R
OC
cu
r
v
e
u
s
ed
to
ev
alu
ate
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
class
if
icatio
n
m
o
d
el
in
cr
e
d
it
r
is
k
p
r
ed
ictio
n
.
T
h
is
cu
r
v
e
ill
u
s
tr
ates
th
e
r
elatio
n
s
h
ip
b
etwe
en
th
e
tr
u
e
p
o
s
itiv
e
r
ate
(
T
PR
)
o
r
s
en
s
itiv
ity
with
th
e
f
alse
p
o
s
itiv
e
r
ate
(
FP
R
)
at
v
ar
io
u
s
d
ec
is
io
n
t
h
r
esh
o
ld
v
alu
es.
I
n
th
is
co
n
tex
t,
T
PR
s
h
o
ws
th
e
m
o
d
el'
s
ab
ilit
y
to
co
r
r
ec
tly
id
en
tif
y
h
i
g
h
-
r
is
k
b
o
r
r
o
wer
s
,
wh
ile
FP
R
s
h
o
ws
th
e
er
r
o
r
r
ate
in
class
if
y
in
g
b
o
r
r
o
wer
s
wh
o
ar
e
ac
tu
ally
wo
r
th
y
o
f
b
ein
g
at
r
is
k
.
T
h
e
R
OC
cu
r
v
e
s
h
o
w
n
s
h
o
ws
th
at
th
e
m
o
d
el
h
as
v
e
r
y
g
o
o
d
p
e
r
f
o
r
m
an
ce
,
with
an
ar
ea
u
n
d
er
th
e
cu
r
v
e
(
AUC)
v
alu
e
o
f
0
.
8
8
.
T
h
e
AUC v
alu
e
s
h
o
ws th
at
th
e
m
o
d
el
is
ab
le
to
d
is
tin
g
u
is
h
b
etwe
en
h
ig
h
-
an
d
lo
w
-
r
is
k
c
r
ed
it
ap
p
lican
ts
with
a
h
ig
h
l
ev
el
o
f
ac
cu
r
ac
y
.
T
h
e
clo
s
er
to
1
,
th
e
h
ig
h
er
t
h
e
m
o
d
el'
s
d
is
cr
im
in
atio
n
ab
ilit
y
,
an
d
in
th
is
ca
s
e,
an
AUC
o
f
0
.
8
8
s
h
o
ws
th
at
th
e
m
o
d
el
is
n
o
t
o
n
ly
ac
c
u
r
ate
i
n
tr
ain
in
g
,
b
u
t
also
h
as
g
o
o
d
g
e
n
er
aliza
tio
n
ab
ilit
y
to
n
ew
d
at
a.
T
h
en
t
h
er
e
is
th
e
p
er
f
o
r
m
an
ce
f
o
r
t
h
e
ac
cu
r
ac
y
o
f
th
e
DR
L
m
o
d
el
s
h
o
wn
in
Fi
g
u
r
e
5
.
Fig
u
r
e
5
.
Acc
u
r
ac
y
o
f
DR
L
m
o
d
el
T
h
e
g
r
ap
h
in
Fig
u
r
e
5
s
h
o
ws
th
e
in
cr
ea
s
in
g
tr
en
d
o
f
th
e
ac
cu
r
ac
y
o
f
th
e
DR
L
m
o
d
el
in
p
r
ed
ictin
g
cr
ed
it
r
is
k
d
u
r
in
g
1
0
0
tr
ain
in
g
ep
o
ch
s
.
T
h
e
b
l
u
e
c
u
r
v
e
s
h
o
ws
th
e
ac
cu
r
ac
y
o
f
th
e
tr
ai
n
in
g
d
ata,
th
e
g
r
ee
n
cu
r
v
e
s
h
o
ws
th
e
v
alid
atio
n
a
cc
u
r
ac
y
,
a
n
d
th
e
r
ed
cu
r
v
e
s
h
o
ws
th
e
ac
cu
r
ac
y
o
n
th
e
tes
tin
g
d
ata.
All
th
r
ee
cu
r
v
es
s
h
o
w
co
n
s
is
ten
t
im
p
r
o
v
em
en
t
o
v
e
r
tim
e,
in
d
icatin
g
th
at
th
e
m
o
d
el
n
o
t
o
n
l
y
lear
n
s
well
b
u
t
is
also
ab
le
to
g
en
er
alize
to
n
ew
d
ata
with
o
u
t
o
v
er
f
itti
n
g
.
W
ith
a
test
ac
c
u
r
ac
y
ap
p
r
o
ac
h
i
n
g
0
.
9
1
0
at
th
e
en
d
o
f
th
e
e
p
o
ch
,
th
e
p
er
f
o
r
m
an
ce
o
f
th
is
DR
L
p
r
o
v
es
its
r
eliab
ilit
y
in
h
an
d
lin
g
co
m
p
lex
an
d
m
u
lti
-
attr
ib
u
te
cr
ed
it
r
is
k
d
ata.
T
h
en
th
er
e
will b
e
an
ev
al
u
atio
n
g
r
a
p
h
s
h
o
win
g
th
e
v
alu
es in
Fig
u
r
e
6
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
Dee
p
r
ein
fo
r
ce
men
t le
a
r
n
in
g
-
b
a
s
ed
cred
it ri
s
k
p
r
ed
ictio
n
mo
d
el
fo
r
mu
lti
-
a
ttr
ib
u
te
d
a
ta
(
Al
-
K
h
o
w
a
r
iz
mi
)
4631
I
n
Fig
u
r
e
6
,
th
e
r
e
will
b
e
a
b
ar
g
r
ap
h
th
at
is
th
e
r
esu
lt
o
f
th
e
DR
L
m
o
d
el
p
er
f
o
r
m
a
n
ce
ev
alu
atio
n
in
p
r
ed
ictin
g
cr
e
d
it
r
is
k
b
ased
o
n
m
u
lti
-
attr
ib
u
te
d
ata.
T
h
e
f
o
u
r
m
ain
m
etr
ics
d
is
p
lay
ed
in
clu
d
e
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
an
d
F1
-
s
co
r
e.
B
ased
o
n
th
e
v
is
u
aliza
tio
n
,
th
e
DR
L
m
o
d
el
s
h
o
ws
v
er
y
g
o
o
d
p
er
f
o
r
m
an
ce
,
with
an
ac
cu
r
ac
y
o
f
9
1
%,
in
d
icatin
g
th
at
m
o
s
t
o
f
th
e
p
r
ed
i
ctio
n
s
m
ad
e
b
y
th
e
m
o
d
el
ar
e
in
ac
co
r
d
an
ce
with
th
e
ac
tu
al
v
alu
e
o
f
th
e
a
p
p
lic
an
t'
s
cr
ed
it
r
is
k
.
I
n
ad
d
itio
n
,
a
p
r
ec
is
io
n
o
f
8
9
%
in
d
icate
s
th
at
th
e
m
o
d
el
h
as
a
h
ig
h
ab
ilit
y
to
av
o
i
d
er
r
o
r
s
in
class
if
y
in
g
ap
p
lican
ts
wh
o
ar
e
ac
tu
ally
elig
ib
le
as
r
is
k
y
,
th
er
eb
y
r
ed
u
cin
g
th
e
ch
an
ce
o
f
ty
p
e
I
er
r
o
r
s
.
Me
an
wh
ile,
a
r
ec
all
o
f
9
0
%
r
ef
lects
th
e
m
o
d
el'
s
ab
ilit
y
to
d
etec
t
alm
o
s
t
all
ap
p
lican
ts
wh
o
ar
e
tr
u
l
y
h
i
g
h
r
is
k
,
t
h
u
s
m
in
im
izin
g
t
y
p
e
I
I
er
r
o
r
s
.
T
h
e
h
ig
h
F1
-
s
co
r
e
v
al
u
e,
wh
ic
h
is
8
9
.
5
%,
in
d
icate
s
th
at
th
e
m
o
d
el
is
n
o
t
o
n
ly
ac
cu
r
ate
b
u
t
also
b
alan
ce
d
b
etwe
en
ac
cu
r
ac
y
an
d
s
en
s
itiv
ity
i
n
its
class
if
icatio
n
.
T
h
is
g
r
ap
h
s
h
o
ws
th
at
th
e
f
o
u
r
m
etr
ics
h
av
e
h
ig
h
an
d
r
el
ativ
ely
b
alan
ce
d
v
alu
es,
in
d
ic
atin
g
th
at
th
e
DR
L
m
o
d
el
d
o
es
n
o
t
e
x
p
er
ien
ce
o
v
er
f
itti
n
g
an
d
is
ab
le
to
g
en
e
r
al
ize
well
to
n
ew
d
ata.
T
h
e
d
if
f
e
r
en
t
co
lo
r
s
o
n
ea
c
h
b
ar
m
ak
e
it
ea
s
y
to
id
en
tify
ea
ch
m
etr
ic,
an
d
th
e
n
u
m
b
er
lab
els
ab
o
v
e
ea
ch
b
ar
p
r
o
v
id
e
clea
r
in
f
o
r
m
atio
n
ab
o
u
t
its
q
u
a
n
titativ
e
v
alu
e.
Ov
er
all,
th
is
g
r
ap
h
r
ein
f
o
r
ce
s
th
e
f
in
d
in
g
t
h
at
th
e
a
p
p
licatio
n
o
f
DR
L
in
cr
e
d
it
r
is
k
p
r
ed
ictio
n
s
y
s
tem
s
n
o
t
o
n
ly
im
p
r
o
v
es
ac
cu
r
ac
y
,
b
u
t
also
p
r
o
v
id
es
co
n
s
is
ten
t
an
d
r
eliab
le
r
esu
lts
in
au
to
m
ated
an
d
ad
a
p
tiv
e
cr
ed
it
d
ec
is
io
n
-
m
ak
in
g
.
Fig
u
r
e
6
.
Per
f
o
r
m
an
c
e
ev
alu
at
io
n
o
f
DR
L
m
o
d
el
4
.
4
.
Dis
cus
s
io
n
T
h
is
s
tu
d
y
d
is
cu
s
s
es
th
e
ap
p
licatio
n
o
f
th
e
DR
L
m
eth
o
d
to
b
u
ild
a
cr
ed
it r
is
k
p
r
ed
ictio
n
m
o
d
el
b
ased
o
n
co
m
p
lex
a
n
d
d
y
n
am
ic
m
u
lti
-
attr
ib
u
te
d
ata.
Usi
n
g
a
d
at
aset
o
f
5
,
0
0
0
e
n
tr
ies
th
at
h
av
e
g
o
n
e
th
r
o
u
g
h
th
e
p
r
ep
r
o
ce
s
s
in
g
s
tag
e,
th
e
m
o
d
e
l
is
d
ev
elo
p
e
d
to
b
e
a
b
le
to
le
ar
n
f
r
o
m
p
atter
n
s
th
at
a
p
p
ea
r
in
cr
ed
it
a
p
p
lican
t
d
ata.
T
h
e
m
ain
o
b
jectiv
e
o
f
th
is
ap
p
r
o
ac
h
is
to
d
ev
elo
p
a
n
in
tellig
en
t
p
r
ed
ictiv
e
ag
en
t
th
a
t
is
ab
le
to
p
r
o
v
id
e
m
o
r
e
ac
c
u
r
ate
an
d
a
d
ap
tiv
e
r
is
k
class
if
icatio
n
,
an
d
s
u
p
p
o
r
t
m
o
r
e
ap
p
r
o
p
r
iate
cr
ed
it
d
ec
is
io
n
m
ak
in
g
.
T
h
e
DR
L
m
o
d
el
in
th
is
s
tu
d
y
is
i
m
p
lem
en
ted
u
s
in
g
th
e
d
ee
p
Q
-
n
etwo
r
k
ap
p
r
o
ac
h
,
wh
ich
f
o
r
m
s
an
ev
alu
ativ
e
ag
en
t
th
at
lear
n
s
th
e
v
alu
e
o
f
th
e
ac
tio
n
f
u
n
ctio
n
(
,
;
)
.
I
n
th
is
m
o
d
el,
it
will
d
ec
id
e
wh
eth
e
r
t
h
e
cr
ed
it
ap
p
licatio
n
will
b
e
a
p
p
r
o
v
ed
,
r
ejec
ted
,
o
r
r
ev
iewe
d
.
R
ewa
r
d
s
ar
e
g
i
v
en
b
ased
o
n
th
e
ac
cu
r
ac
y
o
f
th
e
d
ec
is
io
n
o
n
th
e
ac
tu
al
d
ata.
T
h
is
m
o
d
el
also
ap
p
lies
ex
p
er
ien
ce
r
ep
la
y
an
d
ta
r
g
et
n
etwo
r
k
tech
n
iq
u
es
to
ac
ce
ler
ate
an
d
s
tab
ilize
th
e
tr
ain
in
g
p
r
o
ce
s
s
.
On
e
o
f
th
e
s
ig
n
if
ica
n
t
ad
v
a
n
tag
es
o
f
t
h
is
m
eth
o
d
is
its
ab
ilit
y
to
ad
a
p
t
to
ch
an
g
es in
d
ata
t
r
en
d
s
,
f
o
r
ex
a
m
p
le
d
u
e
t
o
ch
an
g
es in
ec
o
n
o
m
ic
co
n
d
itio
n
s
o
r
cu
s
to
m
er
b
e
h
av
io
r
p
atter
n
s
.
I
n
th
is
s
tu
d
y
,
to
m
ea
s
u
r
e
m
o
d
el
p
er
f
o
r
m
an
ce
,
a
R
OC
cu
r
v
e
is
d
is
p
lay
ed
wh
ich
d
escr
ib
es
t
h
e
b
alan
ce
b
etwe
en
T
PR
an
d
FP
R
.
T
h
e
AUC
v
alu
e
o
f
0
.
8
8
in
d
icate
s
th
at
th
e
m
o
d
el
h
as
ex
ce
llen
t
d
is
cr
im
in
atio
n
ab
ilit
y
in
d
is
tin
g
u
is
h
in
g
h
ig
h
an
d
lo
w
r
is
k
ap
p
lican
ts
.
T
h
is
in
d
ica
tes
th
at
DR
L
n
o
t
o
n
ly
e
x
ce
ls
in
tr
ain
i
n
g
b
u
t
ca
n
also
p
er
f
o
r
m
s
tr
o
n
g
g
en
er
aliz
atio
n
to
n
ew
d
ata.
T
h
en
th
e
m
o
d
el
ac
cu
r
ac
y
p
er
f
o
r
m
a
n
ce
g
r
ap
h
f
o
r
1
0
0
ep
o
c
h
s
s
h
o
ws
th
at
th
e
ac
cu
r
ac
y
c
u
r
v
es
f
o
r
tr
ain
i
n
g
,
v
alid
atio
n
,
an
d
test
in
g
d
ata
i
n
cr
ea
s
e
co
n
s
is
ten
tly
.
T
h
e
test
in
g
ac
cu
r
ac
y
a
p
p
r
o
ac
h
ed
0
.
9
1
0
at
th
e
en
d
o
f
tr
ain
i
n
g
,
i
n
d
icatin
g
th
at
th
e
m
o
d
el
d
id
n
o
t
ex
p
er
ie
n
ce
o
v
e
r
f
itti
n
g
an
d
was a
b
le
to
lear
n
well
f
r
o
m
h
i
s
to
r
ical
d
ata.
I
n
th
is
s
tu
d
y
,
th
er
e
will
b
e
m
ain
ev
alu
atio
n
m
etr
ics
with
a
cc
u
r
ac
y
(
9
1
%),
p
r
ec
is
io
n
(
8
9
%),
r
ec
all
(
9
0
%),
an
d
F1
-
s
co
r
e
(
8
9
.
5
%)
wh
ich
s
h
o
w
th
at
th
e
DR
L
m
o
d
el
is
n
o
t
o
n
ly
ac
cu
r
ate
b
u
t
also
b
alan
ce
d
an
d
r
eliab
le.
Hig
h
p
r
ec
is
io
n
i
n
d
ica
tes
lo
w
f
alse
p
o
s
itiv
es,
wh
ile
h
ig
h
r
ec
all
in
d
icate
s
th
at
th
e
m
o
d
el
d
o
es
n
o
t
m
is
s
m
an
y
h
ig
h
-
r
is
k
ca
s
es.
F1
-
s
co
r
es
th
at
ar
e
clo
s
e
to
th
e
p
r
ec
is
i
o
n
an
d
r
ec
all
v
alu
es
s
tr
en
g
th
e
n
th
at
th
e
m
o
d
el
is
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
5
,
No
.
5
,
Octo
b
er
2
0
2
6
:
4
6
2
4
-
4
6
3
4
4632
ab
le
to
m
ain
tain
a
b
alan
ce
b
et
wee
n
th
e
two
.
Vis
u
ally
,
th
e
g
r
ap
h
clar
if
ies
th
e
d
o
m
i
n
an
ce
o
f
DR
L
p
er
f
o
r
m
an
ce
o
v
er
th
e
b
aselin
e,
s
u
p
p
o
r
tin
g
th
e
claim
th
at
th
is
m
eth
o
d
is
e
f
f
ec
tiv
e
f
o
r
r
ea
l
im
p
le
m
en
tatio
n
in
a
n
au
to
m
ated
cr
ed
it r
is
k
ev
alu
atio
n
s
y
s
tem
.
5.
CO
NCLU
SI
O
N
T
h
is
s
tu
d
y
p
r
o
v
es
th
at
th
e
a
p
p
licatio
n
o
f
t
h
e
DR
L
m
eth
o
d
,
e
s
p
ec
ially
th
e
d
ee
p
Q
-
n
etwo
r
k
ap
p
r
o
ac
h
,
is
v
er
y
ef
f
ec
tiv
e
i
n
b
u
ild
i
n
g
a
cr
ed
it
r
is
k
p
r
e
d
ictio
n
m
o
d
e
l
b
ased
o
n
c
o
m
p
lex
a
n
d
d
y
n
am
ic
m
u
lti
-
attr
ib
u
te
d
ata.
B
y
u
tili
zin
g
5
,
0
0
0
p
r
ev
i
o
u
s
ly
p
r
o
ce
s
s
ed
d
ata,
th
e
m
o
d
el
is
ab
le
to
lear
n
f
r
o
m
h
is
t
o
r
ical
p
atter
n
s
an
d
p
r
o
d
u
ce
ac
cu
r
ate,
ad
ap
tiv
e
p
r
ed
ictiv
e
d
ec
is
io
n
s
,
an
d
s
u
p
p
o
r
t
m
o
r
e
p
r
ec
is
e
cr
ed
it
d
ec
is
io
n
m
ak
in
g
an
d
th
is
s
tu
d
y
h
as
m
o
d
el
p
er
f
o
r
m
a
n
c
e
s
h
o
wn
t
h
r
o
u
g
h
an
AUC
v
alu
e
o
f
0
.
8
8
an
d
a
test
ac
cu
r
ac
y
o
f
u
p
t
o
9
1
%
in
d
icatin
g
h
ig
h
g
en
er
aliza
tio
n
ab
ilit
y
an
d
m
in
im
al
r
is
k
o
f
o
v
er
f
itti
n
g
.
Fu
r
t
h
er
ev
alu
atio
n
th
r
o
u
g
h
p
r
ec
is
io
n
m
etr
ics
o
f
8
9
%,
r
ec
all
o
f
9
0
%,
an
d
F1
-
s
co
r
e
o
f
8
9
.
5
%
s
h
o
ws
th
e
b
ala
n
ce
an
d
r
eliab
ilit
y
o
f
p
r
ed
ictio
n
s
.
T
h
e
m
ain
ad
v
a
n
tag
e
o
f
DR
L
lies
i
n
its
ab
ilit
y
to
ad
ap
t to
ch
an
g
e
s
in
d
ata
tr
en
d
s
.
ACK
NO
WL
E
DG
M
E
N
T
S
T
h
e
au
th
o
r
s
wo
u
ld
lik
e
to
e
x
p
r
ess
o
u
r
s
in
ce
r
e
th
an
k
s
to
Pr
o
f
.
Dr
.
Ak
r
im
,
M.
Pd
.
,
r
ec
to
r
o
f
Un
iv
er
s
itas
Mu
h
am
m
ad
iy
a
h
Su
m
ater
a
Uta
r
a,
Me
d
an
,
I
n
d
o
n
esia,
f
o
r
s
u
p
p
o
r
tin
g
th
is
r
esear
ch
p
r
o
ce
s
s
in
ter
m
s
o
f
f
u
n
d
in
g
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
T
h
is
r
esear
ch
was
f
u
n
d
e
d
b
y
Pro
f
.
Dr
.
Ak
r
im
,
M.
Pd
.
,
r
ec
to
r
o
f
Un
iv
e
r
s
itas
Mu
h
a
m
m
ad
iy
ah
Su
m
ater
a
Utar
a,
Me
d
an
,
I
n
d
o
n
esia.
T
h
r
o
u
g
h
s
cien
tific
p
u
b
li
ca
tio
n
co
s
t a
s
s
is
tan
ce
f
r
o
m
L
P2
M
UM
SU.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
T
h
is
jo
u
r
n
al
u
s
es
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT
)
to
r
ec
o
g
n
ize
in
d
iv
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
llab
o
r
atio
n
.
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
Al
-
Kh
o
war
izm
i
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
Fer
r
y
Fach
r
izal
✓
✓
✓
✓
✓
✓
✓
R
o
m
i Fad
illah
R
ah
m
at
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
Ar
if
R
id
h
o
L
u
b
is
✓
✓
✓
✓
✓
✓
✓
Har
iy
ad
i
✓
✓
✓
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
Au
th
o
r
s
s
tate
n
o
co
n
f
lict o
f
in
t
er
est.
DATA AV
AI
L
AB
I
L
I
T
Y
Data
was
o
b
tain
ed
d
ir
ec
tly
f
r
o
m
I
n
d
o
n
esian
b
a
n
k
s
,
b
u
t
it
is
n
o
t
p
er
m
itted
t
o
m
e
n
tio
n
th
e
n
am
es
o
f
b
an
k
s
s
p
ec
if
ically
.
RE
F
E
R
E
NC
E
S
[
1
]
J.
P
.
N
o
r
i
e
g
a
,
L
.
A
.
R
i
v
e
r
a
,
a
n
d
J.
A
.
H
e
r
r
e
r
a
,
“
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
f
o
r
c
r
e
d
i
t
r
i
s
k
p
r
e
d
i
c
t
i
o
n
:
a
s
y
st
e
ma
t
i
c
l
i
t
e
r
a
t
u
r
e
r
e
v
i
e
w
,
”
D
a
t
a
,
v
o
l
.
8
,
n
o
.
1
1
,
N
o
v
.
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
d
a
t
a
8
1
1
0
1
6
9
.
[
2
]
W
.
Z
h
a
n
g
,
S
.
Y
a
n
,
J.
L
i
,
R
.
P
e
n
g
,
a
n
d
X
.
Ti
a
n
,
“
D
e
e
p
r
e
i
n
f
o
r
c
e
m
e
n
t
l
e
a
r
n
i
n
g
i
m
b
a
l
a
n
c
e
d
c
r
e
d
i
t
r
i
s
k
o
f
S
M
E
s
i
n
s
u
p
p
l
y
c
h
a
i
n
f
i
n
a
n
c
e
,
”
A
n
n
a
l
s
o
f
O
p
e
r
a
t
i
o
n
s
Re
se
a
rch
,
p
p
.
1
–
3
1
,
M
a
r
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
0
4
7
9
-
0
2
4
-
0
5
9
2
1
-
w.
[
3
]
S
.
B
h
a
t
o
r
e
,
L
.
M
o
h
a
n
,
a
n
d
Y
.
R
.
R
e
d
d
y
,
“
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s
f
o
r
c
r
e
d
i
t
r
i
s
k
e
v
a
l
u
a
t
i
o
n
:
a
sy
s
t
e
m
a
t
i
c
l
i
t
e
r
a
t
u
r
e
r
e
v
i
e
w
,
”
J
o
u
rn
a
l
o
f
Ba
n
k
i
n
g
a
n
d
F
i
n
a
n
c
i
a
l
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
4
,
n
o
.
1
,
p
p
.
1
1
1
–
1
3
8
,
A
p
r
.
2
0
2
0
,
d
o
i
:
1
0
.
1
0
0
7
/
s
4
2
7
8
6
-
0
2
0
-
0
0
0
2
0
-
3.
[
4
]
S
.
P
a
u
l
,
A
.
G
u
p
t
a
,
A
.
K
.
K
a
r
,
a
n
d
V
.
S
i
n
g
h
,
“
A
n
a
u
t
o
ma
t
i
c
d
e
e
p
r
e
i
n
f
o
r
c
e
men
t
l
e
a
r
n
i
n
g
b
a
s
e
d
c
r
e
d
i
t
s
c
o
r
i
n
g
m
o
d
e
l
u
s
i
n
g
d
e
e
p
-
Q
n
e
t
w
o
r
k
f
o
r
c
l
a
ss
i
f
i
c
a
t
i
o
n
o
f
c
u
st
o
m
e
r
c
r
e
d
i
t
r
e
q
u
e
s
t
s,
”
i
n
2
0
2
3
I
EEE
I
n
t
e
rn
a
t
i
o
n
a
l
S
y
m
p
o
si
u
m
o
n
T
e
c
h
n
o
l
o
g
y
a
n
d
S
o
c
i
e
t
y
(
I
S
T
AS
)
,
S
e
p
.
2
0
2
3
,
p
p
.
1
–
8
,
d
o
i
:
1
0
.
1
1
0
9
/
I
S
TA
S
5
7
9
3
0
.
2
0
2
3
.
1
0
3
0
6
1
1
1
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
Dee
p
r
ein
fo
r
ce
men
t le
a
r
n
in
g
-
b
a
s
ed
cred
it ri
s
k
p
r
ed
ictio
n
mo
d
el
fo
r
mu
lti
-
a
ttr
ib
u
te
d
a
ta
(
Al
-
K
h
o
w
a
r
iz
mi
)
4633
[
5
]
P
.
A
d
d
o
,
D
.
G
u
e
g
a
n
,
a
n
d
B
.
H
a
ssa
n
i
,
“
C
r
e
d
i
t
r
i
sk
a
n
a
l
y
si
s
u
s
i
n
g
ma
c
h
i
n
e
a
n
d
d
e
e
p
l
e
a
r
n
i
n
g
m
o
d
e
l
s,”
R
i
sk
s
,
v
o
l
.
6
,
n
o
.
2
,
A
p
r
.
2
0
1
8
,
d
o
i
:
1
0
.
3
3
9
0
/
r
i
s
k
s6
0
2
0
0
3
8
.
[
6
]
K
.
S
i
v
a
ma
y
i
l
v
e
l
a
n
,
E.
R
a
j
a
s
e
k
a
r
,
S
.
V
a
i
r
a
v
a
s
u
n
d
a
r
a
m,
S
.
B
a
l
a
c
h
a
n
d
r
a
n
,
a
n
d
V
.
S
u
r
e
s
h
,
“
F
l
e
x
i
b
l
e
r
e
c
o
mm
e
n
d
a
t
i
o
n
f
o
r
o
p
t
i
mi
z
i
n
g
t
h
e
d
e
b
t
c
o
l
l
e
c
t
i
o
n
p
r
o
c
e
ss
b
a
se
d
o
n
c
u
st
o
mer
r
i
s
k
u
si
n
g
d
e
e
p
r
e
i
n
f
o
r
c
e
me
n
t
l
e
a
r
n
i
n
g
,
”
Ex
p
e
r
t
S
y
st
e
m
s
w
i
t
h
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
2
5
6
,
D
e
c
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
s
w
a
.
2
0
2
4
.
1
2
4
9
5
1
.
[
7
]
R
.
C
.
T
o
r
r
e
s,
G
.
M
.
O
l
a
i
v
a
r
,
a
n
d
S
.
I
.
B
r
i
t
a
n
i
c
o
,
“
C
o
n
s
u
mers
’
f
i
n
a
n
c
i
a
l
h
a
b
i
t
s
o
n
ser
v
e
r
-
b
a
se
d
e
l
e
c
t
r
o
n
i
c
mo
n
e
y
a
s
i
t
a
f
f
e
c
t
s
t
h
e
i
r
f
i
n
a
n
c
i
a
l
b
e
h
a
v
i
o
r
:
mo
d
e
r
a
t
e
d
b
y
mo
n
t
h
l
y
t
r
a
n
sa
c
t
i
o
n
s,
”
i
n
9
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
B
u
si
n
e
ss
a
n
d
I
n
f
o
rm
a
t
i
o
n
Ma
n
a
g
e
m
e
n
t
,
S
e
p
.
2
0
2
5
,
p
p
.
5
2
–
55
,
d
o
i
:
1
0
.
1
1
4
5
/
3
7
8
5
1
7
1
.
3
7
8
5
1
9
2
.
[
8
]
V
.
S
i
n
g
h
,
S
.
-
S
.
C
h
e
n
,
M
.
S
i
n
g
h
a
n
i
a
,
B
.
N
a
n
a
v
a
t
i
,
A
.
K
u
mar
K
a
r
,
a
n
d
A
.
G
u
p
t
a
,
“
H
o
w
a
r
e
r
e
i
n
f
o
r
c
e
m
e
n
t
l
e
a
r
n
i
n
g
a
n
d
d
e
e
p
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
ms
u
se
d
f
o
r
b
i
g
d
a
t
a
-
b
a
s
e
d
d
e
c
i
si
o
n
ma
k
i
n
g
i
n
f
i
n
a
n
c
i
a
l
i
n
d
u
s
t
r
i
e
s
–
a
r
e
v
i
e
w
a
n
d
r
e
s
e
a
r
c
h
a
g
e
n
d
a
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
I
n
f
o
rm
a
t
i
o
n
Ma
n
a
g
e
m
e
n
t
D
a
t
a
I
n
si
g
h
t
s
,
v
o
l
.
2
,
n
o
.
2
,
N
o
v
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
j
i
m
e
i
.
2
0
2
2
.
1
0
0
0
9
4
.
[
9
]
P
.
Z.
L
a
p
p
a
s
a
n
d
A
.
N
.
Y
a
n
n
a
c
o
p
o
u
l
o
s,
“
A
mac
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
c
o
mb
i
n
i
n
g
e
x
p
e
r
t
k
n
o
w
l
e
d
g
e
w
i
t
h
g
e
n
e
t
i
c
a
l
g
o
r
i
t
h
ms
i
n
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
f
o
r
c
r
e
d
i
t
r
i
s
k
a
ssess
men
t
,
”
A
p
p
l
i
e
d
S
o
f
t
C
o
m
p
u
t
i
n
g
,
v
o
l
.
1
0
7
,
A
u
g
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
so
c
.
2
0
2
1
.
1
0
7
3
9
1
.
[
1
0
]
V
.
C
h
a
n
g
,
Q
.
A
.
X
u
,
S
.
H
.
A
k
i
n
l
o
y
e
,
V
.
B
e
n
so
n
,
a
n
d
K
.
H
a
l
l
,
“
P
r
e
d
i
c
t
i
o
n
o
f
b
a
n
k
c
r
e
d
i
t
w
o
r
t
h
i
n
e
ss
t
h
r
o
u
g
h
c
r
e
d
i
t
r
i
sk
a
n
a
l
y
s
i
s:
a
n
e
x
p
l
a
i
n
a
b
l
e
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
st
u
d
y
,
”
A
n
n
a
l
s
o
f
O
p
e
ra
t
i
o
n
s
R
e
se
a
r
c
h
,
v
o
l
.
3
5
4
,
n
o
.
1
,
p
p
.
2
4
7
–
2
7
1
,
N
o
v
.
2
0
2
5
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
0
4
7
9
-
024
-
0
6
1
3
4
-
x.
[
1
1
]
S
.
L
i
n
,
D
.
S
o
n
g
,
B
.
C
a
o
,
X
.
G
u
,
a
n
d
J.
Li
,
“
C
r
e
d
i
t
r
i
s
k
a
ss
e
ssm
e
n
t
o
f
a
u
t
o
m
o
b
i
l
e
l
o
a
n
s
u
si
n
g
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
-
b
a
se
d
S
h
a
p
l
e
y
a
d
d
i
t
i
v
e
e
x
p
l
a
n
a
t
i
o
n
s
a
p
p
r
o
a
c
h
,
”
En
g
i
n
e
e
r
i
n
g
A
p
p
l
i
c
a
t
i
o
n
s
o
f
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
1
4
7
,
M
a
y
2
0
2
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
n
g
a
p
p
a
i
.
2
0
2
5
.
1
1
0
2
3
6
.
[
1
2
]
X
.
W
a
n
g
e
t
a
l
.
,
“
D
e
e
p
r
e
i
n
f
o
r
c
e
me
n
t
l
e
a
r
n
i
n
g
:
a
s
u
r
v
e
y
,
”
I
EE
E
T
ra
n
s
a
c
t
i
o
n
s
o
n
N
e
u
r
a
l
N
e
t
w
o
rk
s
a
n
d
L
e
a
r
n
i
n
g
S
y
s
t
e
m
s
,
v
o
l
.
3
5
,
n
o
.
4
,
p
p
.
5
0
6
4
–
5
0
7
8
,
A
p
r
.
2
0
2
4
,
d
o
i
:
1
0
.
1
1
0
9
/
TN
N
LS
.
2
0
2
2
.
3
2
0
7
3
4
6
.
[
1
3
]
D
.
M
h
l
a
n
g
a
,
“
F
i
n
a
n
c
i
a
l
i
n
c
l
u
s
i
o
n
i
n
e
merg
i
n
g
e
c
o
n
o
mi
e
s:
t
h
e
a
p
p
l
i
c
a
t
i
o
n
o
f
mac
h
i
n
e
l
e
a
r
n
i
n
g
a
n
d
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
i
n
c
r
e
d
i
t
r
i
sk
a
ssessm
e
n
t
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
F
i
n
a
n
c
i
a
l
S
t
u
d
i
e
s
,
v
o
l
.
9
,
n
o
.
3
,
J
u
l
.
2
0
2
1
,
d
o
i
:
1
0
.
3
3
9
0
/
i
j
f
s
9
0
3
0
0
3
9
.
[
1
4
]
I
.
G
a
smi,
S
.
N
e
j
i
,
N
.
M
a
n
s
o
u
r
i
,
a
n
d
M
.
S
o
u
i
,
“
B
a
n
k
c
r
e
d
i
t
r
i
sk
p
r
e
d
i
c
t
i
o
n
u
si
n
g
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
mo
d
e
l
,
”
N
e
u
r
a
l
C
o
m
p
u
t
i
n
g
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
3
7
,
n
o
.
1
6
,
p
p
.
1
0
3
3
3
–
1
0
3
5
0
,
J
u
n
.
2
0
2
5
,
d
o
i
:
1
0
.
1
0
0
7
/
s
0
0
5
2
1
-
025
-
1
1
0
4
4
-
5.
[
1
5
]
A
.
D
u
a
n
d
A
.
G
h
a
v
i
d
e
l
,
“
P
a
r
a
m
e
t
e
r
i
z
e
d
d
e
e
p
r
e
i
n
f
o
r
c
e
me
n
t
l
e
a
r
n
i
n
g
-
e
n
a
b
l
e
d
m
a
i
n
t
e
n
a
n
c
e
d
e
c
i
si
o
n
-
su
p
p
o
r
t
a
n
d
l
i
f
e
-
c
y
c
l
e
r
i
s
k
a
ssessm
e
n
t
f
o
r
h
i
g
h
w
a
y
b
r
i
d
g
e
p
o
r
t
f
o
l
i
o
s
,
”
S
t
r
u
c
t
u
ra
l
S
a
f
e
t
y
,
v
o
l
.
9
7
,
J
u
l
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
s
t
r
u
s
a
f
e
.
2
0
2
2
.
1
0
2
2
2
1
.
[
1
6
]
D
.
Y
a
n
g
a
n
d
B
.
X
i
a
o
,
“
F
e
a
t
u
r
e
-
e
n
h
a
n
c
e
d
e
n
se
mb
l
e
m
o
d
e
l
i
n
g
w
i
t
h
v
o
t
i
n
g
o
p
t
i
m
i
z
a
t
i
o
n
f
o
r
c
r
e
d
i
t
r
i
s
k
a
ssess
me
n
t
,
”
I
EEE
A
c
c
e
ss
,
v
o
l
.
1
2
,
p
p
.
1
1
5
1
2
4
–
1
1
5
1
3
6
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
4
.
3
4
4
5
4
9
9
.
[
1
7
]
C
.
Y
u
,
Y
.
J
i
n
,
Q
.
X
i
n
g
,
Y
.
Z
h
a
n
g
,
S
.
G
u
o
,
a
n
d
S
.
M
e
n
g
,
“
A
d
v
a
n
c
e
d
u
ser
c
r
e
d
i
t
r
i
s
k
p
r
e
d
i
c
t
i
o
n
m
o
d
e
l
u
si
n
g
Li
g
h
t
G
B
M
,
X
G
B
o
o
s
t
a
n
d
T
a
b
n
e
t
w
i
t
h
S
M
O
T
EEN
N
,
”
i
n
2
0
2
4
I
EEE
6
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
Po
w
e
r
,
I
n
t
e
l
l
i
g
e
n
t
C
o
m
p
u
t
i
n
g
a
n
d
S
y
st
e
m
s
(
I
C
PI
C
S
)
,
J
u
l
.
2
0
2
4
,
p
p
.
8
7
6
–
883
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
P
I
C
S
6
2
0
5
3
.
2
0
2
4
.
1
0
7
9
6
2
4
7
.
[
1
8
]
M
.
S
c
h
mi
t
t
,
“
D
e
e
p
l
e
a
r
n
i
n
g
v
s.
g
r
a
d
i
e
n
t
b
o
o
st
i
n
g
:
b
e
n
c
h
m
a
r
k
i
n
g
s
t
a
t
e
-
of
-
t
h
e
-
a
r
t
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
ms
f
o
r
c
r
e
d
i
t
sc
o
r
i
n
g
,
”
2
0
2
2
,
a
rX
i
v
:
2
2
0
5
.
1
0
5
3
5
.
[
1
9
]
B
.
V
.
S
a
mo
r
o
d
o
v
,
G
.
M
.
A
z
a
r
e
n
k
o
v
a
,
O
.
G
.
G
o
l
o
v
k
o
,
O
.
Y
u
.
M
i
r
o
sh
n
i
k
,
a
n
d
M
.
V
.
B
a
b
e
n
k
o
,
“
C
r
e
d
i
t
r
i
s
k
m
a
n
a
g
e
me
n
t
i
n
t
h
e
b
a
n
k
’
s
f
i
n
a
n
c
i
a
l
s
t
a
b
i
l
i
t
y
sy
s
t
e
m,”
F
i
n
a
n
c
i
a
l
a
n
d
c
r
e
d
i
t
a
c
t
i
v
i
t
y
p
r
o
b
l
e
m
s
o
f
t
h
e
o
r
y
a
n
d
p
ra
c
t
i
c
e
,
v
o
l
.
4
,
n
o
.
3
1
,
p
p
.
3
0
1
–
3
1
0
,
Jan
.
2
0
2
1
,
d
o
i
:
1
0
.
1
8
3
7
1
/
f
c
a
p
t
p
.
v
4
i
3
1
.
1
9
0
9
2
0
.
[
2
0
]
C
.
H
u
a
n
d
M
.
X
u
,
“
A
d
a
p
t
i
v
e
e
x
p
l
o
r
a
t
i
o
n
s
t
r
a
t
e
g
y
w
i
t
h
m
u
l
t
i
-
a
t
t
r
i
b
u
t
e
d
e
c
i
s
i
o
n
-
ma
k
i
n
g
f
o
r
r
e
i
n
f
o
r
c
e
me
n
t
l
e
a
r
n
i
n
g
,
”
I
EE
E
Ac
c
e
ss
,
v
o
l
.
8
,
p
p
.
3
2
3
5
3
–
3
2
3
6
4
,
2
0
2
0
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
0
.
2
9
7
3
1
6
9
.
[
2
1
]
F
.
Li
a
o
,
“
D
y
n
a
m
i
c
c
r
e
d
i
t
r
a
t
i
n
g
e
v
a
l
u
a
t
i
o
n
a
l
g
o
r
i
t
h
m
f
o
r
c
u
s
t
o
m
e
r
s
u
s
i
n
g
r
e
i
n
f
o
r
c
e
m
e
n
t
l
e
a
r
n
i
n
g
a
n
d
m
u
l
t
i
-
f
e
a
t
u
r
e
w
e
i
g
h
t
a
n
a
l
y
si
s
,
”
i
n
S
e
c
o
n
d
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
Bi
g
D
a
t
a
,
C
o
m
p
u
t
a
t
i
o
n
a
l
I
n
t
e
l
l
i
g
e
n
c
e
,
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
(
BD
C
I
A
2
0
2
4
)
,
M
a
r
.
2
0
2
5
,
p
p
.
1
0
6
9
–
1
0
7
5
,
d
o
i
:
1
0
.
1
1
1
7
/
1
2
.
3
0
5
9
6
5
6
.
[
2
2
]
A
.
B
e
l
h
a
d
i
,
S
.
S
.
K
a
m
b
l
e
,
V
.
M
a
n
i
,
I
.
B
e
n
k
h
a
t
i
,
a
n
d
F
.
E.
T
o
u
r
i
k
i
,
“
A
n
e
n
s
e
mb
l
e
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
f
o
r
f
o
r
e
c
a
st
i
n
g
c
r
e
d
i
t
r
i
sk
o
f
a
g
r
i
c
u
l
t
u
r
a
l
S
M
Es’
i
n
v
e
s
t
me
n
t
s
i
n
a
g
r
i
c
u
l
t
u
r
e
4
.
0
t
h
r
o
u
g
h
s
u
p
p
l
y
c
h
a
i
n
f
i
n
a
n
c
e
,
”
A
n
n
a
l
s
o
f
O
p
e
r
a
t
i
o
n
s R
e
sea
r
c
h
,
v
o
l
.
3
4
5
,
n
o
.
2
–
3
,
p
p
.
7
7
9
–
8
0
7
,
F
e
b
.
2
0
2
5
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
0
4
7
9
-
0
2
1
-
0
4
3
6
6
-
9.
[
2
3
]
M.
-
J.
A
.
-
G
a
r
z
ó
n
,
J.
A
r
r
o
y
o
,
M
.
-
J
.
S
.
-
V
a
r
g
a
s,
a
n
d
A
.
C
a
p
a
r
r
i
n
i
,
“
P
r
o
f
i
t
-
se
n
s
i
t
i
v
e
mac
h
i
n
e
l
e
a
r
n
i
n
g
c
l
a
ssi
f
i
c
a
t
i
o
n
w
i
t
h
e
x
p
l
a
n
a
t
i
o
n
s
i
n
c
r
e
d
i
t
r
i
s
k
:
t
h
e
c
a
se
o
f
sma
l
l
b
u
si
n
e
sses
i
n
p
e
e
r
-
to
-
p
e
e
r
l
e
n
d
i
n
g
,
”
E
l
e
c
t
r
o
n
i
c
C
o
m
m
e
rc
e
Re
se
a
rc
h
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
6
7
,
S
e
p
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
l
e
r
a
p
.
2
0
2
4
.
1
0
1
4
2
8
.
[
2
4
]
S
.
S
a
t
h
e
e
s
h
k
u
m
a
r
,
M
.
D
a
k
s
h
a
n
a
,
K
.
G
u
n
a
l
a
n
,
P
.
A
n
a
n
d
a
n
,
R
.
S
a
v
e
e
t
h
a
,
a
n
d
M
.
N
i
t
h
y
a
,
“
Le
v
e
r
a
g
i
n
g
mac
h
i
n
e
l
e
a
r
n
i
n
g
a
n
d
f
o
r
e
c
a
st
i
n
g
t
e
c
h
n
i
q
u
e
s
t
o
e
n
h
a
n
c
e
c
r
e
d
i
t
r
i
s
k
a
n
a
l
y
si
s
a
n
d
p
r
e
d
i
c
t
i
o
n
,
”
i
n
2
0
2
4
2
n
d
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
S
e
l
f
S
u
st
a
i
n
a
b
l
e
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
S
y
s
t
e
m
s (I
C
S
S
A
S
)
,
O
c
t
.
2
0
2
4
,
p
p
.
7
8
1
–
7
8
6
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
S
S
A
S
6
4
0
0
1
.
2
0
2
4
.
1
0
7
6
0
7
4
6
.
[
2
5
]
X
.
Z
h
a
n
g
a
n
d
L
.
Y
u
,
“
C
o
n
s
u
m
e
r
c
r
e
d
i
t
r
i
sk
a
ssessm
e
n
t
:
a
r
e
v
i
e
w
f
r
o
m
t
h
e
s
t
a
t
e
-
of
-
t
h
e
-
a
r
t
c
l
a
ss
i
f
i
c
a
t
i
o
n
a
l
g
o
r
i
t
h
ms,
d
a
t
a
t
r
a
i
t
s
,
a
n
d
l
e
a
r
n
i
n
g
m
e
t
h
o
d
s,
”
E
x
p
e
rt
S
y
st
e
m
s w
i
t
h
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
2
3
7
,
M
a
r
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
sw
a
.
2
0
2
3
.
1
2
1
4
8
4
.
[
2
6
]
Z.
J.
K
.
A
b
a
d
i
,
N
.
M
a
n
s
o
u
r
i
,
a
n
d
M
.
M
.
J
a
v
i
d
i
,
“
D
e
e
p
r
e
i
n
f
o
r
c
e
m
e
n
t
l
e
a
r
n
i
n
g
-
b
a
se
d
s
c
h
e
d
u
l
i
n
g
i
n
d
i
s
t
r
i
b
u
t
e
d
s
y
st
e
ms:
a
c
r
i
t
i
c
a
l
r
e
v
i
e
w
,
”
K
n
o
w
l
e
d
g
e
a
n
d
I
n
f
o
rm
a
t
i
o
n
S
y
s
t
e
m
s
,
v
o
l
.
6
6
,
n
o
.
1
0
,
p
p
.
5
7
0
9
–
5
7
8
2
,
O
c
t
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
0
1
1
5
-
0
2
4
-
0
2
1
6
7
-
7.
[
2
7
]
X
.
Ta
n
g
e
t
a
l
.
,
“
R
e
i
n
f
o
r
c
e
me
n
t
l
e
a
r
n
i
n
g
-
b
a
s
e
d
e
n
e
r
g
y
m
a
n
a
g
e
me
n
t
f
o
r
h
y
b
r
i
d
p
o
w
e
r
sy
s
t
e
ms
:
s
t
a
t
e
-
of
-
t
h
e
-
a
r
t
s
u
r
v
e
y
,
r
e
v
i
e
w
,
a
n
d
p
e
r
s
p
e
c
t
i
v
e
s
,
”
C
h
i
n
e
s
e
J
o
u
r
n
a
l
o
f
Me
c
h
a
n
i
c
a
l
E
n
g
i
n
e
e
r
i
n
g
,
v
o
l
.
3
7
,
n
o
.
1
,
M
a
y
2
0
2
4
,
d
o
i
:
1
0
.
1
1
8
6
/
s
1
0
0
3
3
-
0
2
4
-
0
1
0
2
6
-
4.
[
2
8
]
I
.
Š
t
i
mac
,
A
.
V
i
d
o
v
i
ć
,
T.
M
i
h
e
t
e
c
,
a
n
d
M
.
D
r
l
j
a
č
a
,
“
O
p
t
i
m
i
z
a
t
i
o
n
o
f
a
i
r
p
o
r
t
c
a
p
a
c
i
t
y
e
f
f
i
c
i
e
n
c
y
b
y
se
l
e
c
t
i
n
g
o
p
t
i
mal
a
i
r
c
r
a
f
t
a
n
d
a
i
r
l
i
n
e
b
u
s
i
n
e
ss
mo
d
e
l
,
”
S
u
s
t
a
i
n
a
b
i
l
i
t
y
,
v
o
l
.
1
2
,
n
o
.
1
0
,
M
a
y
2
0
2
0
,
d
o
i
:
1
0
.
3
3
9
0
/
s
u
1
2
1
0
3
9
8
8
.
[
2
9
]
B
.
K
i
z
i
e
l
e
w
i
c
z
,
A
.
S
h
e
k
h
o
v
t
s
o
v
,
J.
W
i
ę
c
k
o
w
sk
i
,
J.
W
ą
t
r
ó
b
s
k
i
,
a
n
d
W
.
S
a
ł
a
b
u
n
,
“
I
n
t
e
l
l
i
g
e
n
t
c
h
a
r
a
c
t
e
r
i
s
t
i
c
o
b
j
e
c
t
s
me
t
h
o
d
(
I
N
C
O
M
E)
:
a
d
a
t
a
k
n
o
w
l
e
d
g
e
-
b
a
s
e
d
mu
l
t
i
-
c
r
i
t
e
r
i
a
d
e
c
i
s
i
o
n
a
n
a
l
y
s
i
s,
”
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
Re
v
i
e
w
,
v
o
l
.
5
7
,
n
o
.
1
0
,
A
u
g
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
0
4
6
2
-
024
-
1
0
8
9
2
-
2.
[
3
0
]
X
.
W
e
i
,
C
.
R
a
o
,
X
.
X
i
a
o
,
L
.
C
h
e
n
,
a
n
d
M
.
G
o
h
,
“
R
i
sk
a
ssessm
e
n
t
o
f
c
a
r
d
i
o
v
a
s
c
u
l
a
r
d
i
se
a
se
b
a
s
e
d
o
n
S
O
LSS
A
-
C
a
t
B
o
o
st
m
o
d
e
l
,
”
Ex
p
e
rt
S
y
st
e
m
s w
i
t
h
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
2
1
9
,
J
u
n
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
sw
a
.
2
0
2
3
.
1
1
9
6
4
8
.
[
3
1
]
A
.
B
a
r
l
y
b
a
y
e
v
,
A
.
S
h
a
r
i
p
b
a
y
,
G
.
S
h
a
k
h
me
t
o
v
a
,
a
n
d
A
.
Z
h
u
m
a
d
i
l
l
a
y
e
v
a
,
“
D
e
v
e
l
o
p
me
n
t
o
f
a
f
l
e
x
i
b
l
e
i
n
f
o
r
m
a
t
i
o
n
s
e
c
u
r
i
t
y
r
i
sk
mo
d
e
l
u
s
i
n
g
mac
h
i
n
e
l
e
a
r
n
i
n
g
me
t
h
o
d
s a
n
d
o
n
t
o
l
o
g
i
e
s,”
A
p
p
l
i
e
d
S
c
i
e
n
c
e
s
,
v
o
l
.
1
4
,
n
o
.
2
1
,
O
c
t
.
2
0
2
4
,
d
o
i
:
1
0
.
3
3
9
0
/
a
p
p
1
4
2
1
9
8
5
8
.
[
3
2
]
S
.
B
o
u
h
o
u
l
a
,
M
.
A
v
g
e
r
i
s
,
A
.
Le
i
v
a
d
e
a
s,
a
n
d
I
.
La
m
b
a
d
a
r
i
s,
“
D
R
L
-
b
a
se
d
t
r
a
j
e
c
t
o
r
y
p
l
a
n
n
i
n
g
a
n
d
s
e
n
so
r
t
a
sk
sc
h
e
d
u
l
i
n
g
f
o
r
e
d
g
e
r
o
b
o
t
i
c
s,”
i
n
2
0
2
4
I
E
E
E
1
0
t
h
W
o
rl
d
Fo
r
u
m
o
n
I
n
t
e
rn
e
t
o
f
T
h
i
n
g
s
(
WF
-
I
o
T
)
,
N
o
v
.
2
0
2
4
,
p
p
.
3
5
2
–
3
5
7
,
d
o
i
:
1
0
.
1
1
0
9
/
W
F
-
I
o
T
6
2
0
7
8
.
2
0
2
4
.
1
0
8
1
1
3
9
4
.
[
3
3
]
M
.
P
.
N
o
o
n
a
n
,
B
.
K
.
H
.
C
h
a
u
,
M
.
F
.
S
.
R
u
s
h
w
o
r
t
h
,
a
n
d
L.
K
.
F
e
l
l
o
w
s,
“
C
o
n
t
r
a
st
i
n
g
e
f
f
e
c
t
s
o
f
me
d
i
a
l
a
n
d
l
a
t
e
r
a
l
o
r
b
i
t
o
f
r
o
n
t
a
l
c
o
r
t
e
x
l
e
si
o
n
s
o
n
c
r
e
d
i
t
a
ssi
g
n
me
n
t
a
n
d
d
e
c
i
si
o
n
-
m
a
k
i
n
g
i
n
h
u
m
a
n
s
,
”
T
h
e
J
o
u
rn
a
l
o
f
N
e
u
r
o
s
c
i
e
n
c
e
,
v
o
l
.
3
7
,
n
o
.
2
9
,
p
p
.
7
0
2
3
–
7
0
3
5
,
Ju
l
.
2
0
1
7
,
d
o
i
:
1
0
.
1
5
2
3
/
JN
EU
R
O
S
C
I
.
0
6
9
2
-
1
7
.
2
0
1
7
.
[
3
4
]
S
.
M
a
i
t
r
a
,
“
C
o
v
a
r
i
a
n
c
e
ma
t
r
i
x
a
d
a
p
t
a
t
i
o
n
,
e
v
o
l
u
t
i
o
n
a
r
y
g
a
m
e
t
h
e
o
r
y
a
n
d
d
e
e
p
Q
-
n
e
t
w
o
r
k
s
i
n
t
e
g
r
a
t
e
d
e
v
o
l
u
t
i
o
n
a
r
y
a
l
g
o
r
i
t
h
m,
”
J
o
u
rn
a
l
o
f
M
o
d
e
l
l
i
n
g
i
n
Ma
n
a
g
e
m
e
n
t
,
p
p
.
1
–
3
4
,
M
a
y
2
0
2
6
,
d
o
i
:
1
0
.
1
1
0
8
/
J
M
2
-
03
-
2
0
2
5
-
0
0
9
2
.
Evaluation Warning : The document was created with Spire.PDF for Python.