I
nd
o
ne
s
ia
n J
o
urna
l o
f
E
lect
rica
l En
g
ineering
a
nd
Co
m
pu
t
er
Science
Vo
l.
3
8
,
No
.
2
,
Ma
y
20
2
5
,
p
p
.
1
149
~
1
1
6
1
I
SS
N:
2
502
-
4
7
52
,
DOI
: 1
0
.
1
1
5
9
1
/ijee
cs
.v
3
8
.
i
2
.
pp
1
1
4
9
-
1
1
6
1
1149
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ee
cs
.
ia
esco
r
e.
co
m
Enha
ncing
uncoll
a
teralized loa
n ris
k as
ses
sm
ent
a
ccu
ra
cy
throug
h f
ea
t
ure s
elect
io
n and
adv
a
nced ma
chin
e lea
rning
techniqu
es
Sh
a
hrul Niza
m
Sa
la
hu
di
n
,
Yo
s
za
Da
s
ril
,
Yo
s
y
Arisa
nd
y
F
a
c
u
l
t
y
o
f
T
e
c
h
n
o
l
o
g
y
M
a
n
a
g
e
m
e
n
t
a
n
d
B
u
si
n
e
ss,
U
n
i
v
e
r
s
i
t
i
T
u
n
H
u
ss
e
i
n
O
n
n
M
a
l
a
y
s
i
a
,
J
o
h
o
r
,
M
a
l
a
y
s
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Ju
n
1
,
2
0
2
4
R
ev
is
ed
No
v
6
,
2
0
2
4
Acc
ep
ted
No
v
11
,
2
0
2
4
Ac
c
u
ra
c
y
in
e
v
a
l
u
a
ti
n
g
t
h
e
ris
k
o
f
c
re
d
it
a
p
p
li
c
a
ti
o
n
s
is
c
ru
c
ial
f
o
r
len
d
e
rs,
p
a
rti
c
u
larly
wh
e
n
d
e
a
li
n
g
with
u
n
se
c
u
re
d
lo
a
n
s.
Ac
c
u
ra
c
y
c
a
n
b
e
e
n
h
a
n
c
e
d
b
y
se
lec
ti
n
g
su
i
tab
le
fe
a
tu
re
s
f
o
r
a
m
a
c
h
in
e
lea
rn
in
g
m
o
d
e
l.
To
b
e
tt
e
r
id
e
n
ti
f
y
h
ig
h
-
risk
b
o
rro
we
rs,
th
is
stu
d
y
a
p
p
li
e
s
a
n
e
lab
o
ra
te
fe
a
tu
re
se
lec
ti
o
n
tec
h
n
iq
u
e
.
Th
is
stu
d
y
u
se
s
th
e
li
g
h
t
g
ra
d
ien
t
b
o
o
stin
g
m
a
c
h
in
e
(LG
BM
)
Clas
sifier
m
o
d
e
l
with
b
o
o
stin
g
ty
p
e
g
ra
d
ien
t
b
o
o
stin
g
d
e
c
isio
n
tree
(
G
BDT)
a
lg
o
rit
h
m
a
n
d
n
_
e
stima
to
r
v
a
lu
e
1
0
0
fo
r
fe
a
tu
re
se
lec
ti
o
n
p
r
o
c
e
ss
.
Th
is wo
r
k
u
se
s
a
d
v
a
n
c
e
d
m
a
c
h
in
e
lea
rn
in
g
tec
h
n
i
q
u
e
s
n
a
m
e
ly
sta
c
k
i
n
g
t
o
imp
ro
v
e
a
c
c
u
ra
c
y
m
o
d
e
l
p
e
rfo
rm
.
Th
e
d
a
tas
e
t
c
o
n
sists
o
f
3
0
7
,
5
0
6
a
p
p
li
c
a
n
ts
fro
m
Eu
ro
p
e
a
n
len
d
e
rs
wh
o
h
a
v
e
a
p
p
li
e
d
f
o
r
l
o
a
n
s
in
S
o
u
t
h
e
a
st
As
ia.
Eac
h
a
p
p
li
c
a
n
t
is
d
e
sc
rib
e
d
b
y
1
2
6
d
iffere
n
t
fe
a
tu
re
s
.
Us
in
g
G
DBT
a
lg
o
rit
h
m
G
BDT,
3
0
b
e
st
fe
a
tu
re
s
we
re
se
lec
ted
b
a
se
d
o
n
th
e
ir
m
a
x
imu
m
a
c
c
u
ra
c
y
c
o
m
p
a
re
d
to
a
n
o
th
e
r
fe
a
tu
re
.
By
e
m
p
lo
y
in
g
a
sta
c
k
in
g
tec
h
n
iq
u
e
th
a
t
c
o
m
b
in
e
s
t
h
e
LG
BM
,
g
ra
d
ien
t
b
o
o
sti
n
g
(G
B)
,
a
n
d
ra
n
d
o
m
f
o
re
st
(RF
)
m
o
d
e
ls,
a
n
d
u
ti
li
z
in
g
l
o
g
isti
c
r
e
g
re
ss
io
n
(LR)
a
s
th
e
fi
n
a
l
e
sti
m
a
to
r,
a
n
a
c
c
u
ra
c
y
o
f
0
.
9
9
6
3
7
wa
s
re
a
c
h
e
d
.
T
h
is
st
u
d
y
d
e
m
o
n
stra
tes
a
n
imp
ro
v
e
d
t
h
e
a
c
c
u
ra
c
y
c
o
m
p
a
re
d
to
p
re
v
io
u
s
re
se
a
rc
h
.
Th
is
d
isc
o
v
e
ry
in
d
i
c
a
tes
th
a
t
u
ti
li
z
i
n
g
fe
a
tu
r
e
se
lec
ti
o
n
a
n
d
sta
c
k
in
g
m
e
th
o
d
c
a
n
p
r
o
v
i
d
e
o
n
e
o
f
th
e
m
o
st
p
re
c
ise
c
h
o
ice
s
fo
r
m
o
d
e
ll
i
n
g
t
h
e
b
i
n
a
ry
c
las
s
c
las
sifica
ti
o
n
a
m
o
n
g
th
e
c
u
rre
n
t
m
o
d
e
ls
.
K
ey
w
o
r
d
s
:
Acc
u
r
ac
y
Featu
r
e
s
elec
tio
n
Ma
ch
in
e
lear
n
in
g
R
is
k
as
s
ess
m
en
t
Stack
in
g
Un
co
llater
alize
d
l
o
an
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
:
Yo
s
y
Ar
is
an
d
y
Facu
lty
o
f
T
ec
h
n
o
lo
g
y
Ma
n
a
g
em
en
t a
n
d
B
u
s
in
ess
,
Un
iv
er
s
iti
T
u
n
Hu
s
s
ein
On
n
Ma
lay
s
ia
8
6
4
0
0
J
o
h
o
r
,
Ma
lay
s
ia
E
m
ail:
g
p
2
0
0
0
0
2
@
s
tu
d
en
t.u
th
m
.
ed
u
.
m
y
1.
I
NT
RO
D
UCT
I
O
N
E
ac
h
y
ea
r
,
a
co
n
s
id
er
a
b
le
p
er
c
en
tag
e
o
f
b
o
r
r
o
wer
s
with
u
n
s
ec
u
r
ed
lo
an
ar
e
d
e
f
au
lt
[
1
]
.
E
m
p
h
asizin
g
th
e
ess
en
tial
r
eq
u
ir
em
e
n
t
f
o
r
p
r
ec
is
e
d
ata
an
d
d
ep
e
n
d
ab
l
e
ea
r
ly
d
etec
tio
n
m
o
d
els
to
ac
c
u
r
ately
ev
al
u
ate
d
ef
au
lt
r
is
k
[
2
]
,
m
ac
h
in
e
lear
n
in
g
tech
n
iq
u
es
ar
e
p
r
o
g
r
ess
iv
ely
em
p
lo
y
e
d
f
o
r
th
is
ass
es
s
m
en
t
to
en
s
u
r
e
th
e
q
u
ality
o
f
tar
g
ets
in
th
e
d
ataset
[
3
]
,
[
4
]
.
T
h
e
in
teg
r
atio
n
o
f
m
ac
h
in
e
lear
n
in
g
an
d
e
n
h
an
ci
n
g
th
e
alg
o
r
ith
m
h
as
p
r
o
v
id
e
d
a
m
o
r
e
n
u
an
ce
d
a
p
p
r
o
ac
h
[
5
]
to
g
et
s
o
lu
tio
n
s
t
o
g
lo
b
al
o
p
tim
izatio
n
m
o
d
ell
in
g
p
r
o
b
lem
s
th
an
tr
ad
itio
n
al
r
is
k
ev
alu
atio
n
m
o
d
el
[
6
]
.
Ho
wev
e
r
,
s
o
m
e
o
u
ts
tan
d
in
g
co
n
ce
r
n
s
co
n
tin
u
e
to
ex
is
t,
esp
ec
ially
co
n
ce
r
n
in
g
f
ea
t
u
r
e
s
elec
tio
n
.
Desp
ite
th
e
co
m
p
letio
n
o
f
f
e
atu
r
e
im
p
o
r
tan
ce
a
n
aly
s
is
,
th
er
e
is
p
o
ten
tial
f
o
r
en
h
an
ce
m
e
n
t in
d
eter
m
in
in
g
wh
ich
f
ea
tu
r
es m
o
s
t sig
n
if
ican
tly
co
n
tr
ib
u
te
t
o
lo
an
d
ef
au
lt.
Featu
r
e
s
elec
tio
n
in
m
ac
h
i
n
e
lear
n
in
g
is
th
e
p
r
o
ce
s
s
o
f
s
elec
tin
g
th
e
o
p
tim
al
f
ea
t
u
r
es
f
o
r
a
class
if
icatio
n
p
r
o
b
lem
in
o
r
d
e
r
to
in
cr
ea
s
e
th
e
class
if
icatio
n
’
s
ac
cu
r
ac
y
[
7
]
.
R
ec
u
r
s
iv
e
f
ea
t
u
r
e
elim
in
atio
n
is
a
tech
n
iq
u
e
f
o
r
l
o
wer
in
g
t
h
e
ef
f
ec
t
o
f
n
o
is
y
d
ata
a
n
d
in
c
r
ea
s
in
g
co
m
p
u
tatio
n
al
p
er
f
o
r
m
an
c
e
[
8
]
.
T
h
e
ac
c
u
r
a
c
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
8
,
No
.
2
,
May
20
2
5
:
1
1
4
9
-
1
1
6
1
1150
o
f
th
e
m
o
d
el
is
in
f
lu
en
ce
d
b
y
th
e
n
u
m
b
er
o
f
f
ea
tu
r
es
th
at
ar
e
ex
am
in
ed
.
W
h
en
a
g
r
ea
ter
n
u
m
b
er
o
f
o
p
tim
ally
s
elec
ted
f
ea
tu
r
es
ar
e
e
x
am
in
e
d
,
th
e
le
v
el
o
f
ac
cu
r
ac
y
will
i
n
cr
ea
s
e
[
9
]
.
Path
an
et
a
l.
[
1
0
]
,
f
ea
tu
r
e
s
elec
tio
n
tech
n
iq
u
es,
s
u
ch
as
th
o
s
e
u
s
ed
in
g
r
ad
ien
t b
o
o
s
tin
g
d
ec
is
io
n
tr
ee
s
(
GB
DT
s
)
,
is
u
s
ef
u
l
in
id
en
tify
in
g
i
m
p
o
r
ta
n
t
f
ea
tu
r
es
b
y
ta
k
in
g
in
t
o
ac
co
u
n
t
th
eir
co
n
tr
ib
u
tio
n
to
th
e
m
o
d
el
’
s
p
er
f
o
r
m
a
n
ce
an
d
d
e
alin
g
with
co
m
p
lex
r
elatio
n
s
h
ip
s
in
th
e
d
ata,
an
d
ac
h
iev
e
h
ig
h
er
p
r
ed
ictiv
e
ac
c
u
r
ac
y
an
d
av
o
id
b
ein
g
in
f
lu
en
ce
d
b
y
ir
r
elev
an
t
o
r
n
o
is
y
f
ea
tu
r
es
[
1
1
]
.
Mo
r
e
o
v
e
r
S
ev
er
al
p
r
ev
io
u
s
s
tu
d
ies
h
a
v
e
f
o
u
n
d
th
at
u
s
in
g
f
ea
tu
r
e
s
elec
tio
n
tech
n
iq
u
es,
p
ar
ticu
la
r
ly
GB
DT
,
ca
n
im
p
r
o
v
e
m
o
d
el
p
e
r
f
o
r
m
an
ce
,
r
ed
u
ce
c
o
m
p
u
tatio
n
al
co
m
p
le
x
ity
,
an
d
im
p
r
o
v
e
g
en
er
aliza
tio
n
to
n
ew,
p
r
ev
io
u
s
ly
u
n
s
ee
n
d
ata
[
1
2
]
–
[
1
4
]
.
I
n
co
n
tr
ast
to
p
r
ev
io
u
s
r
esear
ch
th
at
f
o
c
u
s
ed
o
n
g
en
er
ic
GB
DT
ap
p
licatio
n
s
,
th
is
s
tu
d
y
ex
p
lo
r
es
th
e
f
ea
tu
r
es
o
f
cr
ed
it
d
ata
to
id
en
tify
th
e
m
o
s
t
p
r
ed
ictiv
e
f
ac
to
r
s
o
f
b
o
r
r
o
wer
d
ef
a
u
lts
.
T
h
e
d
ataset
we
u
tili
ze
is
p
ar
ticu
lar
l
y
co
m
p
ellin
g
,
h
a
v
in
g
b
ee
n
ex
p
lo
r
e
d
in
m
u
ltip
le
s
tu
d
ies
th
at
d
em
o
n
s
tr
ate
its
r
elev
a
n
ce
a
n
d
r
o
b
u
s
tn
ess
ac
r
o
s
s
v
ar
i
o
u
s
r
ese
ar
ch
co
n
tex
ts
in
clu
d
in
g
C
h
en
et
a
l
.
[
1
5
]
u
s
ed
1
0
4
f
ea
tu
r
es
in
h
is
s
tu
d
y
as
th
e
r
es
u
lt
o
f
tr
ee
f
ea
tu
r
es
s
elec
tio
n
an
d
u
s
in
g
th
e
i
n
teg
r
ated
m
eth
o
d
th
at
co
m
b
i
n
es
th
e
d
ee
p
lear
n
in
g
f
r
am
ew
o
r
k
Dee
p
GB
M
with
C
atN
N
h
an
d
lin
g
s
p
ar
s
e
ca
teg
o
r
ical
d
ata
an
d
G
B
DT
2
NN
h
an
d
lin
g
d
en
s
e
n
u
m
er
ical
d
ata
,
th
u
s
o
b
t
ain
in
g
th
e
b
est
a
r
ea
u
n
d
er
th
e
cu
r
v
e
(
AUC
)
v
alu
e
o
f
0
.
7
5
5
8
3
2
.
T
ia
n
et
a
l
.
[
1
6
]
u
tili
ze
d
t
h
e
P
ea
r
s
o
n
c
o
r
r
elatio
n
co
ef
f
icie
n
t
as
a
m
eth
o
d
o
f
f
ea
tu
r
e
s
elec
tio
n
was
u
s
ed
in
th
e
s
elec
ted
f
ea
tu
r
e
s
o
th
at
ar
o
u
n
d
8
0
f
ea
tu
r
es
wer
e
p
r
o
d
u
ce
d
wh
ich
p
r
o
d
u
ce
d
t
h
e
b
est
ac
cu
r
ac
y
o
f
9
0
.
9
9
%
u
s
in
g
th
e
GB
DT
m
o
d
el
.
Featu
r
e
en
g
i
n
ee
r
in
g
a
n
d
co
m
p
a
r
in
g
f
ea
t
u
r
es
ac
r
o
s
s
all
m
o
d
els
b
y
Ma
h
m
u
d
i
et
a
l
.
[
1
7
]
ex
tr
ac
te
d
4
0
f
ea
tu
r
es
an
d
f
o
u
n
d
t
h
e
b
est
ac
cu
r
ac
y
o
f
9
8
.
4
7
%
u
s
in
g
e
x
tr
em
e
g
r
a
d
ien
t
b
o
o
s
tin
g
(
X
g
b
o
o
s
t
)
.
XGBo
o
s
t
ex
h
ib
its
th
e
s
u
p
er
io
r
ef
f
icac
y
o
f
th
e
XGB
clas
s
if
ier
,
d
em
o
n
s
tr
atin
g
a
s
ig
n
if
ican
t
ca
p
ac
ity
to
f
o
r
ec
ast
cr
ed
itwo
r
th
in
ess
with
co
n
s
id
er
ab
le
p
r
ec
is
io
n
[
1
8
]
.
As
an
ad
d
e
d
b
e
n
ef
it,
t
h
is
p
ap
e
r
u
t
ilized
XGBo
o
s
t
f
o
r
m
o
d
ellin
g
b
u
t a
l
s
o
f
o
cu
s
o
n
p
i
ck
in
g
th
e
m
o
s
t o
p
tim
al
f
ea
tu
r
e
s
,
th
at
h
av
e
a
s
ig
n
if
ican
t im
p
ac
t o
n
ac
h
iev
in
g
th
e
h
ig
h
est
ac
cu
r
ac
y
v
alu
e
b
y
u
tili
ze
d
b
o
o
s
tin
g
f
ea
tu
r
e
s
elec
tio
n
tech
n
iq
u
e
ca
lled
GB
DT
em
b
ed
d
ed
m
eth
o
d
an
d
p
u
r
p
o
s
e
th
e
s
tack
in
g
ap
p
r
o
ac
h
f
o
r
m
o
d
el
ev
al
u
atio
n
.
2.
M
E
T
HOD
Fig
u
r
e
1
d
em
o
n
s
tr
ates
th
e
d
ataset
p
r
o
ce
s
s
in
g
s
tep
s
in
th
is
r
esear
ch
.
I
n
ex
p
lo
r
at
o
r
y
d
at
a
an
aly
s
is
,
r
elev
an
t
d
ata
is
co
llected
an
d
v
is
u
alize
d
.
T
h
e
d
ata
s
o
u
r
ce
s
an
d
d
o
m
ain
k
n
o
wled
g
e
elem
en
ts
u
s
ed
to
en
h
an
ce
th
e
d
ataset
s
h
o
u
ld
b
e
co
n
s
id
er
ed
alo
n
g
with
th
e
c
o
r
r
elatio
n
,
im
p
ac
ts
,
an
d
in
ter
a
ctio
n
s
b
etw
ee
n
v
ar
iab
les.
T
h
e
s
ec
o
n
d
p
h
ase
in
m
ac
h
in
e
lear
n
in
g
is
d
ata
p
r
ep
r
o
ce
s
s
in
g
,
wh
ich
clea
n
s
an
d
o
r
g
a
n
izes
r
aw
d
ata
f
o
r
m
o
d
el
cr
ea
tio
n
an
d
tr
ain
in
g
.
Data
p
r
ep
r
o
ce
s
s
in
g
im
p
r
o
v
es
d
ata
q
u
ality
to
en
ab
le
u
s
ef
u
l
in
s
ig
h
ts
.
I
n
p
r
ep
r
o
ce
s
s
in
g
,
u
n
b
alan
ce
d
d
ata
is
h
a
n
d
led
with
s
y
n
th
etic
m
in
o
r
ity
o
v
er
-
s
a
m
p
lin
g
tech
n
iq
u
e
(
SMOT
E
)
.
T
h
e
n
ex
t
s
tep
af
ter
d
ataset
clea
n
in
g
is
id
en
tify
in
g
im
p
o
r
ta
n
t
f
ea
tu
r
es
f
o
r
tar
g
et
p
r
ed
ictio
n
.
GB
DT
s
ar
e
an
ef
f
e
ctiv
e
m
ac
h
in
e
lear
n
in
g
a
p
p
r
o
ac
h
f
o
r
f
ea
tu
r
e
im
p
o
r
tan
ce
d
eter
m
in
atio
n
.
T
h
is
s
tu
d
y
s
et
tr
ain
in
g
s
ets
at
8
0
%
an
d
test
s
ets
at
2
0
%.
I
n
tr
ain
in
g
,
th
e
tr
ain
in
g
an
d
v
alid
atio
n
s
ets ar
e
m
er
g
ed
to
cr
ea
te
a
m
o
d
el.
Fin
ally
,
e
v
alu
atio
n
m
etr
ics
ar
e
u
tili
ze
d
to
ev
alu
ate
th
e
r
is
k
ass
ess
m
en
t m
o
d
el.
Fig
u
r
e
1
.
R
esear
ch
f
lo
wc
h
ar
t
2
.
1
.
Da
t
a
c
o
llect
io
n
T
h
e
d
ataset
p
r
o
v
id
es
a
co
m
p
r
eh
en
s
iv
e
d
escr
ip
tio
n
o
f
ea
ch
ap
p
lican
t,
co
n
s
is
tin
g
o
f
1
2
6
f
ea
tu
r
es
o
r
co
lu
m
n
s
,
e
n
co
m
p
ass
in
g
a
to
ta
l
o
f
3
0
7
,
5
0
6
ap
p
licatio
n
s
.
T
h
e
d
ataset
is
a
co
m
p
ilatio
n
o
f
cu
s
to
m
er
d
ata
lo
an
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
E
n
h
a
n
ci
n
g
u
n
co
lla
tera
liz
ed
lo
a
n
r
is
k
a
s
s
ess
men
t a
cc
u
r
a
cy
th
r
o
u
g
h
fe
a
tu
r
e
… (
S
h
a
h
r
u
l Ni
z
a
m
S
a
la
h
u
d
i
n)
1151
f
r
o
m
Kaz
a
k
h
s
tan
,
R
u
s
s
ia,
Vietn
am
,
C
h
in
a,
I
n
d
o
n
esia,
a
n
d
t
h
e
Ph
ilip
p
in
es
[
1
9
]
.
T
h
is
th
o
r
o
u
g
h
s
et
o
f
f
ea
tu
r
es
en
co
m
p
ass
es
an
ex
ten
s
iv
e
a
r
r
ay
o
f
ap
p
licatio
n
in
f
o
r
m
atio
n
,
in
clu
d
in
g
d
em
o
g
r
a
p
h
ic
d
ata,
f
in
an
cial
co
n
d
itio
n
,
an
d
p
r
io
r
lo
a
n
h
is
to
r
y
.
A
d
ataset
th
at
i
s
co
m
p
letely
let
s
u
s
d
o
co
m
p
r
eh
en
s
iv
e
an
al
y
s
es
an
d
d
is
co
v
er
s
ig
n
if
ican
tly
r
eg
ar
d
in
g
t
h
e
is
s
u
es th
at
af
f
ec
t lo
an
a
p
p
licatio
n
s
an
d
r
esu
lts
.
2
.
2
.
Da
t
a
p
re
pro
ce
s
s
ing
Data
p
r
ep
ar
atio
n
in
v
o
l
v
es
n
u
m
er
o
u
s
ess
en
tial
p
r
o
ce
d
u
r
es.
T
h
ese
en
co
m
p
ass
im
p
o
r
tin
g
t
h
e
r
eq
u
is
ite
lib
r
ar
ies,
r
ec
tify
in
g
m
is
s
in
g
v
alu
es,
en
co
d
in
g
ca
teg
o
r
ical
v
ar
iab
les,
r
em
o
v
e
o
u
tlier
s
,
s
p
litt
in
g
d
ataset
,
an
d
ex
ec
u
tin
g
f
ea
tu
r
e
s
ca
lin
g
[
2
0
]
.
Fu
r
th
e
r
m
o
r
e
,
elim
in
atin
g
o
u
tlier
s
an
d
u
s
in
g
f
ea
t
u
r
e
s
ca
lin
g
en
h
a
n
ce
th
e
d
ataset
’
s
b
alan
ce
an
d
r
e
p
r
esen
tativ
en
ess
,
h
en
ce
im
p
r
o
v
in
g
m
o
d
el
p
er
f
o
r
m
a
n
ce
an
d
g
en
e
r
aliza
tio
n
.
2
.
2
.
1
.
I
m
po
rt
ing
lib
ra
ries a
nd
da
t
a
s
et
I
m
p
o
r
tin
g
d
ata
in
to
th
e
Py
th
o
n
en
v
ir
o
n
m
en
t
c
o
n
s
titu
tes
th
e
in
itial
p
h
ase
o
f
d
ata
an
aly
s
is
.
T
h
e
im
p
o
r
t
f
o
r
m
at
f
o
r
co
m
m
a
s
ep
a
r
ated
v
alu
es
(
C
SV
)
f
iles
,
wh
ich
s
tan
d
s
f
o
r
co
m
m
a
-
s
ep
a
r
ated
v
alu
es
.
T
h
is
is
th
e
f
o
r
m
at
em
p
lo
y
ed
b
y
p
an
d
as
to
im
p
o
r
t
lo
ca
l
d
atasets
in
to
Py
th
o
n
f
o
r
p
r
ep
r
o
ce
s
s
in
g
in
th
is
r
ese
ar
ch
.
Su
b
s
eq
u
en
tly
,
th
e
lib
r
ar
ies
u
tili
ze
d
f
o
r
p
r
et
r
ea
tm
en
t
an
d
ad
d
itio
n
al
d
ata
p
r
o
ce
s
s
in
g
wer
e
im
p
o
r
ted
.
Ma
ch
in
e
lear
n
in
g
p
r
o
jects
in
v
ar
iab
ly
u
tili
ze
th
e
Nu
m
Py
lib
r
ar
y
f
o
r
th
e
m
an
ag
em
e
n
t
o
f
v
ec
to
r
s
an
d
m
atr
ices.
N
u
m
Py
en
co
m
p
ass
es
f
u
n
d
am
e
n
tal
ar
r
a
y
d
ata
ty
p
es
an
d
o
p
er
atio
n
s
,
i
n
clu
d
in
g
in
d
e
x
in
g
,
s
o
r
tin
g
,
r
esh
ap
in
g
,
an
d
e
lem
en
tal
f
u
n
ctio
n
s
.
SciPy
en
co
m
p
ass
es
all
n
u
m
er
ical
co
d
e.
T
h
e
wid
ely
u
tili
ze
d
Pan
d
a
’
s
lib
r
ar
y
is
r
en
o
wn
e
d
f
o
r
its
ef
f
icac
y
in
m
an
ag
in
g
tim
e
s
er
ies
an
d
tab
u
lar
d
ata
s
tr
u
ctu
r
es.
Su
b
s
eq
u
e
n
tly
,
t
h
er
e
is
Ma
tp
lo
tlib
.
T
h
e
p
y
p
lo
t
lib
r
ar
y
is
a
p
o
wer
f
u
l
d
ata
v
is
u
aliza
tio
n
an
d
g
r
ap
h
ical
ch
ar
tin
g
p
ac
k
a
g
e
cr
ea
ted
f
o
r
Py
th
o
n
a
n
d
Nu
m
Py
,
ca
p
a
b
le
o
f
o
p
er
atin
g
o
n
m
u
ltip
le
p
latf
o
r
m
s
[
2
1
]
.
2
.
2
.
2
.
F
ind
ing
m
is
s
ing
v
a
lue a
nd
ha
nd
li
ng
T
h
is
s
tu
d
y
u
s
ed
two
d
if
f
er
en
t
m
eth
o
d
s
to
h
an
d
le
m
is
s
in
g
v
alu
es,
wh
ich
ar
e
c
r
itical
to
m
ain
tain
in
g
d
ata
in
teg
r
ity
.
T
h
ese
m
eth
o
d
s
in
clu
d
e
th
e
av
er
ag
in
g
tech
n
iq
u
e
to
im
p
u
te
m
is
s
in
g
n
u
m
er
ic
v
alu
es
an
d
th
e
u
s
e
o
f
s
u
b
s
titu
te
v
alu
es
to
f
ill
i
n
m
is
s
in
g
ca
teg
o
r
ical
v
alu
es.
B
y
im
p
lem
e
n
tin
g
th
ese
tec
h
n
iq
u
es,
th
is
s
tu
d
y
en
s
u
r
ed
th
at
th
e
d
ata
s
et
r
e
m
ain
ed
as
co
m
p
lete
as
p
o
s
s
i
b
le,
th
er
eb
y
m
in
im
izin
g
th
e
im
p
ac
t
o
f
m
is
s
in
g
in
f
o
r
m
atio
n
o
n
th
e
an
aly
s
is
r
e
s
u
lts
[
2
2
]
.
2
.
2
.
3
.
L
a
bel
en
co
din
g
Du
r
in
g
th
is
s
tag
e,
ca
teg
o
r
y
d
ata
is
co
n
v
er
ted
o
r
e
n
co
d
e
d
in
to
n
u
m
er
ical
v
alu
es.
So
m
e
ty
p
es
o
f
m
ac
h
in
e
lear
n
in
g
,
lik
e
Dee
p
lear
n
in
g
,
n
ee
d
n
u
m
e
r
ical
d
ata
to
wo
r
k
.
I
t
n
ee
d
s
to
b
e
tu
r
n
e
d
in
to
n
u
m
b
e
r
s
b
ef
o
r
e
ca
teg
o
r
y
d
ata
ca
n
b
e
u
s
ed
to
f
it
an
d
test
a
m
o
d
el.
Du
m
m
y
v
ar
iab
les
o
r
L
ab
el
E
n
co
d
in
g
ca
n
b
e
u
s
ed
to
s
o
lv
e
th
at
p
r
o
b
lem
.
I
n
t
h
is
s
tu
d
y
,
th
e
L
ab
elE
n
c
o
d
er
m
eth
o
d
was
u
s
ed
to
t
u
r
n
ca
teg
o
r
ical
d
ata
i
n
to
n
u
m
b
er
v
alu
es.
Ad
d
itio
n
ally
,
th
e
Stan
d
ar
d
Sca
ler
m
eth
o
d
is
u
s
ed
i
n
th
e
p
r
e
p
r
o
ce
s
s
in
g
s
tep
o
f
th
is
s
tu
d
y
.
T
h
is
im
p
r
o
v
es
th
e
p
er
f
o
r
m
an
ce
,
in
te
r
p
r
etab
ilit
y
,
an
d
r
esil
ien
ce
o
f
m
ac
h
in
e
lear
n
in
g
m
o
d
els tr
ain
ed
o
n
th
e
d
at
aset
[
2
3
]
.
2
.
2
.
4
.
Rem
o
v
e
o
utlier
T
h
e
p
r
esen
t
s
tu
d
y
a
d
d
r
ess
ed
o
u
tlier
s
u
tili
zin
g
th
e
in
ter
-
q
u
ar
tile
r
an
g
e
(
I
QR
)
Sco
r
e.
T
h
is
wo
r
k
s
s
im
ilar
to
a
b
o
x
p
lo
t
an
d
z
-
s
co
r
e
in
th
e
s
en
s
e
th
at
a
th
r
esh
o
ld
I
QR
v
alu
e
is
d
ef
in
ed
.
I
QR
is
th
e
f
ir
s
t
q
u
ar
tile
s
u
b
tr
ac
ted
f
r
o
m
th
e
th
ir
d
q
u
ar
t
ile.
An
y
p
o
in
t
b
elo
w
th
e
th
r
esh
o
ld
I
QR
is
r
em
o
v
e
d
[
2
4
]
.
T
h
is
m
eth
o
d
id
en
tifie
s
an
d
r
e
m
o
v
es
d
ata
p
o
in
ts
th
at
f
all
s
ig
n
if
ican
tly
o
u
ts
id
e
th
e
n
o
r
m
al
r
an
g
e,
h
elp
in
g
to
im
p
r
o
v
e
th
e
q
u
ality
o
f
th
e
d
ataset.
B
y
m
an
ag
in
g
o
u
tlier
s
,
th
e
s
tu
d
y
en
s
u
r
es
a
m
o
r
e
r
eliab
le
an
al
y
s
is
,
m
in
im
izin
g
th
e
i
n
f
lu
en
ce
o
f
ex
tr
em
e
v
alu
es o
n
m
o
d
el
ac
cu
r
ac
y
.
2.
2
.
5
B
a
la
ncing
d
a
t
a
T
h
e
d
ataset
u
s
ed
f
o
r
m
o
d
el
tr
ain
in
g
co
n
tain
s
im
b
alan
ce
d
class
es.
T
h
is
lead
s
to
h
ig
h
v
ar
iatio
n
in
p
er
f
o
r
m
an
ce
r
esu
lts
,
esp
ec
ially
in
ac
cu
r
ac
y
[
2
5
]
an
d
s
p
ec
if
icity
r
ate
[
2
6
]
.
Hen
ce
,
to
tack
le
s
u
ch
a
s
itu
atio
n
,
we
ap
p
lied
SMOT
E
o
v
er
s
am
p
lin
g
ap
p
r
o
ac
h
to
p
r
o
d
u
cin
g
r
eliab
le
an
d
ac
cu
r
ate
r
esu
lts
b
y
r
ed
u
cin
g
th
e
ef
f
ec
t
s
o
f
a
b
iased
class
.
I
n
ter
m
s
o
f
ac
cu
r
ac
y
,
Alam
s
y
ah
et.
al
s
u
g
g
ests
th
at
o
v
er
s
am
p
lin
g
is
p
r
ef
er
ab
le
to
u
n
d
er
s
am
p
lin
g
[
2
7
]
.
Un
d
er
s
a
m
p
lin
g
r
u
n
s
th
e
r
is
k
o
f
elim
in
atin
g
s
o
m
e
p
o
r
tio
n
s
o
f
th
e
d
ataset
th
at
in
clu
d
e
cr
u
cial
in
f
o
r
m
atio
n
,
p
er
h
a
p
s
r
esu
ltin
g
in
m
o
d
el
o
v
er
f
itti
n
g
.
C
o
n
v
er
s
ely
,
th
e
m
o
s
t
ef
f
e
ctiv
e
o
v
er
s
am
p
lin
g
ap
p
r
o
ac
h
es
ar
e
th
o
s
e
t
h
at
g
e
n
er
ate
n
ew
d
ata
f
o
r
th
e
m
in
o
r
ity
class
in
s
tead
o
f
ess
en
tially
d
u
p
licatin
g
ex
is
tin
g
d
ata
[
2
8
]
.
2
.
2
.
6
Sp
litt
ing
d
a
t
a
s
et
Fo
llo
win
g
d
ata
co
llectio
n
an
d
p
r
ep
ar
atio
n
,
th
e
d
ataset
is
s
p
lit
in
to
two
s
et
s
,
in
clu
d
in
g
tr
ain
in
g
an
d
test
in
g
.
T
h
e
tr
ain
in
g
s
et
s
er
v
es
as
th
e
f
o
u
n
d
atio
n
f
o
r
tr
ain
in
g
th
e
m
ac
h
in
e
lear
n
in
g
m
o
d
el,
wh
er
ea
s
th
e
test
in
g
s
et
is
r
eq
u
ir
ed
to
ev
alu
ate
th
e
m
o
d
el
’
s
p
er
f
o
r
m
an
ce
.
T
o
a
ch
iev
e
a
f
air
ass
ess
m
en
t,
a
r
an
d
o
m
d
ata
s
p
lit
is
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
8
,
No
.
2
,
May
20
2
5
:
1
1
4
9
-
1
1
6
1
1152
p
er
f
o
r
m
ed
.
T
h
e
c
o
m
p
ar
is
o
n
o
f
th
e
p
er
ce
n
tag
e
o
f
d
atasets
f
o
r
tr
ain
in
g
an
d
v
alid
atio
n
8
0
%
an
d
2
0
%
test
in
g
.
I
n
th
is
s
tu
d
y
,
th
e
T
r
ain
i
n
g
s
ets s
ettin
g
b
y
8
0
% a
n
d
test
s
et
b
y
2
0
% b
ased
o
n
R
o
s
eb
r
o
ck
[
2
9
]
.
2.
3
.
F
e
a
t
ure
s
elec
t
io
n
T
h
is
s
tu
d
y
em
p
lo
y
s
a
b
o
o
s
tin
g
tech
n
iq
u
e
u
tili
zin
g
th
e
em
b
ed
d
in
g
m
eth
o
d
to
d
eter
m
in
e
t
h
e
o
p
tim
al
n
u
m
b
er
o
f
f
ea
tu
r
es
f
o
r
m
o
d
elin
g
p
u
r
p
o
s
es.
T
h
e
lig
h
t
g
r
ad
ien
t
b
o
o
s
tin
g
m
ac
h
in
e
(
L
GB
M
)
C
la
s
s
if
ier
is
co
n
f
ig
u
r
ed
with
a
b
o
o
s
tin
g
ty
p
e
o
f
GB
DT
an
d
a
n
_
esti
m
at
o
r
v
alu
e
o
f
1
0
0
,
as
d
etailed
in
T
ab
le
1
.
T
r
ain
a
m
o
d
el
with
o
n
e
f
ea
tu
r
e
an
d
c
o
m
p
ar
e
its
p
er
f
o
r
m
an
ce
ag
ain
s
t
th
e
m
o
d
el
with
all
f
ea
tu
r
es
to
d
eter
m
in
e
f
ea
tu
r
e
r
elev
an
ce
.
W
h
ile
T
ab
le
2
o
u
tli
n
e
th
e
GB
DT
f
ea
tu
r
e
im
p
o
r
tan
t p
r
o
ce
s
s
ac
cu
r
ac
y
c
o
m
p
ar
is
o
n
.
T
ab
le
1
.
Hy
p
er
p
ar
a
m
eter
s
o
f
f
ea
tu
r
e
s
elec
tio
n
tech
n
iq
u
es
M
e
t
h
o
d
Te
c
h
n
i
q
u
e
H
y
p
e
r
p
a
r
a
me
t
e
r
Emb
e
d
d
e
d
m
e
t
h
o
d
b
o
o
s
t
i
n
g
LG
B
M
C
l
a
ssi
f
i
e
r
,
o
b
j
e
c
t
i
v
e
=
b
i
n
a
r
y
,
b
o
o
s
t
i
n
g
_
t
y
p
e
=
g
b
d
t
,
n
_
e
st
i
ma
t
o
r
s=1
0
0
T
ab
le
2
.
Featu
r
e
im
p
o
r
tan
ce
s
elec
tio
n
n
_
f
e
a
t
u
r
e
s
p
r
o
p
_
f
e
a
r
u
r
e
s
mea
n
_
a
c
c
u
r
a
c
y
mea
n
_
r
o
c
_
a
u
c
mea
n
_
f
i
t
_
t
i
me
55
1
.
0
0
0
0
0
0
0
.
9
1
9
2
1
1
0
.
7
1
0
9
0
8
6
.
9
4
4
8
3
8
54
0
.
9
8
1
8
1
8
0
.
9
1
9
2
2
4
0
.
7
1
0
3
3
2
9
.
4
8
6
7
8
7
53
0
.
9
6
3
6
3
6
0
.
9
1
9
2
1
8
0
.
7
1
0
4
5
9
1
3
.
0
7
7
0
0
4
52
0
.
9
4
5
4
5
5
0
.
9
1
9
2
1
8
0
.
7
1
0
4
6
0
8
.
6
4
6
9
5
3
51
0
.
9
2
7
2
7
3
0
.
9
1
9
2
1
8
0
.
7
1
0
4
6
0
6
.
8
9
0
5
2
2
50
0
.
9
0
9
0
9
1
0
.
9
1
9
2
3
1
0
.
7
1
0
7
6
1
1
0
.
1
4
5
0
2
1
49
0
.
8
9
0
9
0
9
0
.
9
1
9
2
1
8
0
.
7
1
0
6
8
9
1
0
.
4
9
0
0
2
0
48
0
.
8
7
2
7
2
7
0
.
9
1
9
2
2
8
0
.
7
1
0
7
1
8
1
0
.
0
6
9
5
5
9
47
0
.
8
5
4
5
4
5
0
.
9
1
9
2
2
4
0
.
7
1
0
7
0
8
7
.
0
8
0
6
6
1
46
0
.
8
3
6
3
6
4
0
.
9
1
9
2
2
1
0
.
7
1
0
7
4
2
6
.
7
7
9
1
4
0
45
0
.
8
1
8
1
8
2
0
.
9
1
9
2
6
3
0
.
7
1
0
9
6
9
9
.
2
8
4
7
3
4
44
0
.
8
0
0
0
0
0
0
.
9
1
9
2
2
4
0
.
7
0
9
5
9
7
8
.
4
9
4
2
1
9
43
0
.
7
8
1
8
1
8
0
.
9
1
9
2
3
1
0
.
7
1
0
6
9
2
6
.
4
9
1
2
6
5
42
0
.
7
6
3
6
3
6
0
.
9
1
9
2
4
4
0
.
7
0
9
7
8
7
8
.
2
3
4
1
6
1
41
0
.
7
4
5
4
5
5
0
.
9
1
9
2
2
4
0
.
7
1
0
0
3
5
9
.
7
3
4
1
4
2
40
0
.
7
2
7
2
7
3
0
.
9
1
9
2
5
0
0
.
7
1
0
1
0
4
6
.
8
9
3
9
7
9
39
0
.
7
0
9
0
9
1
0
.
9
1
9
2
2
1
0
.
7
1
0
0
9
8
6
.
3
3
7
1
6
7
38
0
.
6
9
0
9
0
9
0
.
9
1
9
2
6
3
0
.
7
1
0
8
0
8
9
.
1
6
9
5
7
8
37
0
.
6
7
2
7
2
7
0
.
9
1
9
2
0
5
0
.
7
0
9
8
1
7
6
.
8
1
3
4
1
8
36
0
.
6
5
4
5
4
5
0
.
9
1
9
2
6
3
0
.
7
0
9
8
2
4
6
.
1
9
6
5
2
2
35
0
.
6
3
6
3
6
4
0
.
9
1
9
2
6
3
0
.
7
0
9
2
6
5
9
.
4
4
9
1
9
8
34
0
.
6
1
8
1
8
2
0
.
9
1
9
2
6
7
0
.
7
0
9
9
5
7
7
.
0
7
4
7
4
6
33
0
.
6
0
0
0
0
0
0
.
9
1
9
2
3
7
0
.
7
0
8
9
5
2
5
.
9
5
2
1
2
7
32
0
.
5
8
1
8
1
8
0
.
9
1
9
2
3
1
0
.
7
1
0
4
1
5
8
.
9
7
5
7
1
4
31
0
.
5
6
3
6
3
6
0
.
9
1
9
2
0
8
0
.
7
0
9
9
8
1
6
.
1
8
1
9
7
5
30
0
.
5
4
5
4
5
5
0
.
9
1
9
2
8
9
0
.
7
0
9
0
4
8
6
.
0
8
9
9
8
1
29
0
.
5
2
7
2
7
3
0
.
9
1
9
2
7
0
0
.
7
0
9
7
1
2
8
.
7
3
0
7
6
5
28
0
.
5
0
9
0
9
1
0
.
9
1
9
2
7
6
0
.
7
0
8
6
3
9
5
.
7
8
3
9
3
5
27
0
.
4
0
9
0
9
1
0
.
9
1
9
2
3
4
0
.
7
0
9
8
8
5
6
.
4
2
5
9
1
0
S
o
u
r
c
e
:
d
a
t
a
p
r
o
c
e
ssi
n
g
GB
DT
is
th
e
m
o
s
t
p
o
p
u
lar
s
tan
d
ar
d
ap
p
r
o
ac
h
f
o
r
tr
ai
n
in
g
DT
-
b
ased
m
o
d
els
[
3
0
]
.
T
h
e
alg
o
r
ith
m
ex
ec
u
tes
tr
ain
in
g
b
y
iter
ativ
e
ly
s
tar
tin
g
with
a
b
ase
m
o
d
e
l,
wh
er
e
th
e
n
e
x
t
m
o
d
el
is
g
o
in
g
to
u
tili
ze
th
e
m
is
tak
e
o
b
tain
ed
in
th
e
p
r
ev
i
o
u
s
p
r
o
ce
s
s
[
3
1
]
.
T
h
is
is
d
if
f
er
en
t
f
r
o
m
th
e
r
an
d
o
m
f
o
r
est
(
R
F)
m
o
d
el,
wh
ich
u
tili
ze
s
m
an
y
DT
m
o
d
els
in
d
ep
en
d
en
tly
[
3
2
]
.
On
e
c
h
ar
ac
t
er
is
tic
o
f
GB
DT
is
th
at
in
th
e
last
iter
atio
n
s
,
th
e
m
o
d
el
ten
d
s
to
o
v
er
-
s
p
ec
ializ
e,
f
o
cu
s
in
g
o
n
ly
o
n
a
f
ew
f
ea
t
u
r
es
th
at
h
av
e
a
h
ig
h
c
o
r
r
elati
o
n
with
th
e
tr
ain
in
g
o
u
tco
m
e
[
3
3
]
.
Af
ter
u
tili
zin
g
L
GB
M
with
b
o
o
s
tin
g
ty
p
e
GB
DT
,
th
e
cr
iter
ia
f
o
r
m
ax
im
u
m
ac
cu
r
ac
y
co
m
p
ar
ed
to
o
th
er
co
llectio
n
s
,
o
n
ly
3
1
o
f
t
h
e
1
2
6
ch
ar
ac
ter
is
tics
,
in
clu
d
in
g
th
e
ta
r
g
et,
wer
e
p
i
ck
ed
.
T
h
e
f
ea
t
u
r
es
m
en
tio
n
ed
a
b
o
v
e
ar
e
r
ec
o
g
n
iz
ed
in
th
e
T
a
b
le
3
.
2.
4
.
M
o
dellin
g
Su
p
er
v
is
ed
m
ac
h
i
n
e
lear
n
i
n
g
an
d
s
tack
in
g
ar
e
m
o
d
eled
b
y
s
ev
er
al
m
ac
h
in
e
lear
n
in
g
m
o
d
e
ls
T
ab
le
4
,
in
clu
d
in
g
RF
,
lo
g
is
tic
r
eg
r
ess
io
n
(
L
R
)
,
g
r
ad
i
e
n
t
b
o
o
s
tin
g
(
GB
)
,
an
d
L
GB
M
,
C
at
B
o
o
s
t
(
C
B
)
,
an
d
XGBo
o
s
t.
Stack
in
g
ap
p
r
o
ac
h
o
u
t
p
er
f
o
r
m
s
th
e
o
th
er
tech
n
iq
u
es
in
ter
m
s
o
f
y
ield
s
h
ig
h
p
e
r
f
o
r
m
an
ce
,
n
o
t
o
n
ly
in
te
r
m
s
o
f
class
if
icatio
n
ac
cu
r
ac
y
[
3
4
]
,
o
u
tp
er
f
o
r
m
s
tr
ad
itio
n
al
cr
e
d
it sco
r
in
g
m
o
d
els in
ter
m
s
o
f
ac
c
u
r
ac
y
an
d
ef
f
icien
c
y
[
3
5
]
,
p
r
ed
ictio
n
ac
cu
r
ac
y
b
u
t a
ls
o
in
p
r
ec
is
io
n
an
d
r
ec
all
[
3
6
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
E
n
h
a
n
ci
n
g
u
n
co
lla
tera
liz
ed
lo
a
n
r
is
k
a
s
s
ess
men
t a
cc
u
r
a
cy
th
r
o
u
g
h
fe
a
tu
r
e
… (
S
h
a
h
r
u
l Ni
z
a
m
S
a
la
h
u
d
i
n)
1153
T
ab
le
3
.
Featu
r
e
im
p
o
r
tan
ce
with
d
escr
ip
tio
n
an
d
weig
h
t
F
e
a
t
u
r
e
i
m
p
o
r
t
a
n
c
e
F
e
a
t
u
r
e
d
e
scri
p
t
i
o
n
W
e
i
g
h
t
i
mp
o
r
t
a
n
c
e
_
n
o
r
ma
l
i
z
e
d
C
R
ED
I
T
TE
R
M
Th
e
l
e
n
g
t
h
o
f
t
h
e
p
a
y
me
n
t
i
n
m
o
n
t
h
s
(
si
n
c
e
t
h
e
a
n
n
u
i
t
y
i
s
t
h
e
m
o
n
t
h
l
y
a
mo
u
n
t
d
u
e
)
5
5
0
0
.
1
8
3
3
3
3
D
A
Y
S
B
I
R
TH
C
l
i
e
n
t
’
s
a
g
e
i
n
d
a
y
s
a
t
t
h
e
t
i
me
o
f
a
p
p
l
i
c
a
t
i
o
n
1
7
4
0
.
0
5
8
0
0
0
D
A
Y
S
I
D
P
U
B
LI
S
H
H
o
w
ma
n
y
d
a
y
s
b
e
f
o
r
e
t
h
e
a
p
p
l
i
c
a
t
i
o
n
d
i
d
c
l
i
e
n
t
c
h
a
n
g
e
t
h
e
i
d
e
n
t
i
t
y
d
o
c
u
m
e
n
t
w
i
t
h
w
h
i
c
h
h
e
a
p
p
l
i
e
d
f
o
r
t
h
e
l
o
a
n
1
5
4
0
.
0
5
1
3
3
3
A
M
T
G
O
O
D
S
P
R
I
C
E
F
o
r
c
o
n
su
m
e
r
l
o
a
n
s
i
t
i
s
t
h
e
p
r
i
c
e
o
f
t
h
e
g
o
o
d
s f
o
r
w
h
i
c
h
t
h
e
l
o
a
n
i
s
g
i
v
e
n
(
c
o
n
s
u
mer
l
o
a
n
)
1
5
0
0
.
0
5
0
0
0
0
D
A
Y
S
LA
S
T
P
H
O
N
E
C
H
A
N
G
E
H
o
w
ma
n
y
d
a
y
s
b
e
f
o
r
e
a
p
p
l
i
c
a
t
i
o
n
d
i
d
c
l
i
e
n
t
c
h
a
n
g
e
p
h
o
n
e
1
4
5
0
.
0
4
8
3
3
3
D
A
Y
S
R
EG
I
S
TR
A
TI
O
N
H
o
w
ma
n
y
d
a
y
s
b
e
f
o
r
e
t
h
e
a
p
p
l
i
c
a
t
i
o
n
d
i
d
c
l
i
e
n
t
c
h
a
n
g
e
h
i
s
r
e
g
i
s
t
r
a
t
i
o
n
1
3
0
0
.
0
4
3
3
3
3
A
M
T
A
N
N
U
I
TY
Lo
a
n
a
n
n
u
i
t
y
1
2
0
0
.
0
4
0
0
0
0
R
EG
I
O
N
P
O
P
U
LA
T
I
O
N
R
EA
TI
V
E
N
o
r
mal
i
z
e
d
p
o
p
u
l
a
t
i
o
n
o
f
r
e
g
i
o
n
w
h
e
r
e
c
l
i
e
n
t
l
i
v
e
s
(
h
i
g
h
e
r
n
u
m
b
e
r
me
a
n
s
t
h
e
c
l
i
e
n
t
l
i
v
e
s
i
n
mo
r
e
p
o
p
u
l
a
t
e
d
r
e
g
i
o
n
)
1
1
8
0
.
0
3
9
3
3
3
A
M
T
C
R
ED
I
T
C
r
e
d
i
t
a
m
o
u
n
t
o
f
t
h
e
l
o
a
n
1
1
4
0
.
0
3
8
0
0
0
D
A
Y
S
EM
P
LO
Y
ED
H
o
w
ma
n
y
d
a
y
s
b
e
f
o
r
e
t
h
e
a
p
p
l
i
c
a
t
i
o
n
t
h
e
p
e
r
so
n
st
a
r
t
e
d
c
u
r
r
e
n
t
e
m
p
l
o
y
me
n
t
92
0
.
0
3
0
6
6
7
D
A
Y
S
EM
P
LO
Y
ED
P
ER
C
EN
T
Th
e
p
e
r
c
e
n
t
a
g
e
o
f
t
h
e
d
a
y
s
e
mp
l
o
y
e
d
r
e
l
a
t
i
v
e
t
o
t
h
e
c
l
i
e
n
t
’
s
a
g
e
92
0
.
0
3
0
6
6
7
C
R
ED
I
T
I
N
C
O
M
E
P
ER
C
EN
T
Th
e
p
e
r
c
e
n
t
a
g
e
o
f
t
h
e
c
r
e
d
i
t
a
m
o
u
n
t
r
e
l
a
t
i
v
e
t
o
a
c
l
i
e
n
t
’
s
i
n
c
o
m
e
89
0
.
0
2
9
6
6
7
A
M
T
R
EQ
C
R
ED
I
T
B
U
R
EA
U
Y
EA
R
N
u
mb
e
r
o
f
e
n
q
u
i
r
i
e
s
t
o
C
r
e
d
i
t
B
u
r
e
a
u
a
b
o
u
t
t
h
e
c
l
i
e
n
t
o
n
e
d
a
y
y
e
a
r
(
e
x
c
l
u
d
i
n
g
l
a
st
3
mo
n
t
h
s
b
e
f
o
r
e
a
p
p
l
i
c
a
t
i
o
n
)
70
0
.
0
2
3
3
3
3
A
N
N
U
I
TY
I
N
C
O
M
E
P
ER
C
EN
T
t
h
e
p
e
r
c
e
n
t
a
g
e
o
f
t
h
e
l
o
a
n
a
n
n
u
i
t
y
r
e
l
a
t
i
v
e
t
o
a
c
l
i
e
n
t
’
s
i
n
c
o
m
e
70
0
.
0
2
3
3
3
3
A
M
T
I
N
C
O
M
E
TO
TA
L
I
n
c
o
me
o
f
t
h
e
c
l
i
e
n
t
w
i
t
h
s
t
a
n
d
a
r
d
b
a
s
e
c
u
r
r
e
n
c
y
b
y
H
o
mec
r
e
d
i
t
(
Tu
a
n
a
n
h
k
k
,
2
0
1
9
)
63
0
.
0
2
1
0
0
0
O
C
C
U
P
A
TI
O
N
TY
P
E
W
h
a
t
k
i
n
d
o
f
o
c
c
u
p
a
t
i
o
n
d
o
e
s
t
h
e
c
l
i
e
n
t
h
a
v
e
62
0
.
0
2
0
6
6
7
O
W
N
C
A
R
A
G
E
A
g
e
o
f
c
l
i
e
n
t
’
s
c
a
r
56
0
.
0
1
8
6
6
7
N
A
M
E
F
A
M
I
LY
S
TA
TU
S
F
a
mi
l
y
s
t
a
t
u
s
o
f
t
h
e
c
l
i
e
n
t
54
0
.
0
1
8
0
0
0
R
EG
I
O
N
R
A
TI
N
G
C
LI
EN
T
W
C
I
TY
O
u
r
r
a
t
i
n
g
o
f
t
h
e
r
e
g
i
o
n
w
h
e
r
e
c
l
i
e
n
t
l
i
v
e
s wi
t
h
t
a
k
i
n
g
c
i
t
y
i
n
t
o
a
c
c
o
u
n
t
(
1
,
2
,
3
)
51
0
.
0
1
7
0
0
0
O
R
G
A
N
I
ZA
T
I
O
N
TY
P
E
Ty
p
e
o
f
o
r
g
a
n
i
z
a
t
i
o
n
w
h
e
r
e
c
l
i
e
n
t
w
o
r
k
s
50
0
.
0
1
6
6
6
7
H
O
U
R
A
P
P
R
P
R
O
C
ESS
S
TA
R
T
Ty
p
e
o
f
o
r
g
a
n
i
z
a
t
i
o
n
w
h
e
r
e
c
l
i
e
n
t
w
o
r
k
s
48
0
.
0
1
6
0
0
0
N
A
M
E
C
O
N
TR
A
C
T
TY
P
E
I
d
e
n
t
i
f
i
c
a
t
i
o
n
i
f
l
o
a
n
i
s
c
a
s
h
o
r
r
e
v
o
l
v
i
n
g
48
0
.
0
1
6
0
0
0
C
O
D
E
G
EN
D
ER
G
e
n
d
e
r
o
f
t
h
e
c
l
i
e
n
t
44
0
.
0
1
4
6
6
7
A
M
T
R
EQ
C
R
ED
I
T
B
U
R
EA
U
Q
R
T
N
u
mb
e
r
o
f
e
n
q
u
i
r
i
e
s
t
o
C
r
e
d
i
t
B
u
r
e
a
u
a
b
o
u
t
t
h
e
c
l
i
e
n
t
3
mo
n
t
h
b
e
f
o
r
e
a
p
p
l
i
c
a
t
i
o
n
(
e
x
c
l
u
d
i
n
g
o
n
e
mo
n
t
h
b
e
f
o
r
e
a
p
p
l
i
c
a
t
i
o
n
)
38
0
.
0
1
2
6
6
7
D
EF 3
0
C
N
T
S
O
C
I
A
L
C
I
R
C
LE
H
o
w
ma
n
y
o
b
s
e
r
v
a
t
i
o
n
s
o
f
c
l
i
e
n
t
’
s s
o
c
i
a
l
s
u
r
r
o
u
n
d
i
n
g
s
d
e
f
a
u
l
t
e
d
o
n
3
0
D
P
D
(
d
a
y
s
p
a
st
d
u
e
)
35
0
.
0
1
1
6
6
7
N
A
M
E
ED
U
C
A
TI
O
N
TY
P
E
Le
v
e
l
o
f
h
i
g
h
e
s
t
e
d
u
c
a
t
i
o
n
t
h
e
c
l
i
e
n
t
a
c
h
i
e
v
e
d
32
0
.
0
1
0
6
6
7
F
LA
G
W
O
R
K
P
H
O
N
E
D
i
d
c
l
i
e
n
t
p
r
o
v
i
d
e
h
o
me
p
h
o
n
e
(
1
=
Y
ES,
0
=
N
O
)
31
0
.
0
1
0
3
3
3
O
B
S
3
0
C
N
T
S
O
C
I
A
L
C
I
R
C
LE
H
o
w
ma
n
y
o
b
s
e
r
v
a
t
i
o
n
s
o
f
c
l
i
e
n
t
’
s s
o
c
i
a
l
s
u
r
r
o
u
n
d
i
n
g
s w
i
t
h
o
b
s
e
r
v
a
b
l
e
3
0
D
P
D
(
d
a
y
s
p
a
s
t
d
u
e
)
d
e
f
a
u
l
t
27
0
.
0
0
9
0
0
0
EM
E
R
G
EN
C
Y
S
TA
T
E
M
O
D
E
N
o
r
mal
i
z
e
d
i
n
f
o
r
m
a
t
i
o
n
a
b
o
u
t
b
u
i
l
d
i
n
g
w
h
e
r
e
t
h
e
c
l
i
e
n
t
l
i
v
e
s
23
0
.
0
0
7
6
6
7
F
LA
G
O
W
N
C
A
R
M
e
a
n
s
l
o
a
n
a
p
p
l
i
c
a
n
t
h
a
v
e
o
w
n
s
a
c
a
r
(
1
=
Y
ES
,
0
=
N
O
)
23
0
.
0
0
7
6
6
7
TA
R
G
ET
Ta
r
g
e
t
v
a
r
i
a
b
l
e
(
1
-
C
l
i
e
n
t
w
i
t
h
p
a
y
m
e
n
t
d
i
f
f
i
c
u
l
t
i
e
s
:
h
e
/
s
h
e
h
a
d
l
a
t
e
p
a
y
me
n
t
m
o
r
e
t
h
a
n
X
d
a
y
s
o
n
a
t
l
e
a
st
o
n
e
o
f
t
h
e
f
i
r
st
Y
i
n
s
t
a
l
l
m
e
n
t
s
o
f
t
h
e
l
o
a
n
,
0
-
a
l
l
o
t
h
e
r
c
a
s
e
s)
-
-
S
o
u
r
c
e
:
d
a
t
a
p
r
o
c
e
ssi
n
g
T
ab
le
4
.
Hy
p
er
p
ar
a
m
eter
s
o
f
a
lg
o
r
ith
m
s
Ty
p
e
A
l
g
o
r
i
t
h
ms
H
y
p
e
r
p
a
r
a
me
t
e
r
S
u
p
e
r
v
i
se
d
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
RF
R
a
n
d
o
m
f
o
r
e
s
t
c
l
a
ss
i
f
i
e
r
,
n
_
e
st
i
ma
t
o
r
s=8
0
0
LR
Lo
g
i
s
t
i
c
r
e
g
r
e
ssi
o
n
GB
G
r
a
d
i
e
n
t
b
o
o
s
t
i
n
g
c
l
a
ss
i
f
i
e
r
LG
B
M
l
g
b
.
LG
B
M
c
l
a
ss
i
f
i
e
r
CB
Cat
b
o
o
s
t
c
l
a
ssi
f
i
e
r
,
i
t
e
r
a
t
i
o
n
s
=
5
,
l
e
a
r
n
i
n
g
_
r
a
t
e
=
0
.
1
,
l
o
ss
_
f
u
n
c
t
i
o
n
=
C
r
o
ssE
n
t
r
o
p
y
X
G
B
o
o
st
XGB
c
l
a
ssi
f
i
e
r
S
t
a
c
k
i
n
g
G
B
,
R
F
,
a
n
d
LG
B
M
w
i
t
h
LR
a
s fi
n
a
l
e
st
i
ma
t
o
r
S
t
a
c
k
i
n
g
c
l
a
ss
i
f
i
e
r
,
e
s
t
i
mat
o
r
s
=
s
t
a
c
,
f
i
n
a
l
_
e
s
t
i
m
a
t
o
r
=
l
r
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
8
,
No
.
2
,
May
20
2
5
:
1
1
4
9
-
1
1
6
1
1154
T
h
is
s
tu
d
y
th
e
s
tak
in
g
m
eth
o
d
was
u
s
ed
b
y
co
m
b
in
in
g
GB
,
R
F,
an
d
L
GB
M
with
LR
as
f
in
al
esti
m
ato
r
.
T
h
e
r
esear
ch
em
p
l
o
y
ed
LR
as
th
e
f
in
al
esti
m
at
o
r
d
u
e
t
o
its
d
em
o
n
s
tr
ated
r
el
iab
ilit
y
in
s
elec
tin
g
m
an
ag
ea
b
le
s
u
b
s
ets
o
f
in
d
icat
o
r
s
[
3
7
]
an
d
Of
f
er
s
p
r
o
b
ab
ilit
ies
f
o
r
p
o
ten
tial
r
esu
lts
,
wh
ich
ca
n
b
e
v
alu
ab
le
f
o
r
m
ak
in
g
in
f
o
r
m
e
d
d
ec
is
io
n
s
[
3
8
]
.
2.
5
.
P
er
f
o
r
m
a
nce
ev
a
lua
t
io
n m
et
rics
I
n
o
r
d
er
to
ev
alu
ate
th
e
ef
f
ica
cy
o
f
th
e
p
r
o
p
o
s
ed
m
eth
o
d
o
lo
g
y
,
m
a
n
y
p
e
r
f
o
r
m
an
ce
m
et
r
ics
h
av
e
b
ee
n
u
tili
ze
d
,
in
clu
d
in
g
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
a
n
d
F1
-
s
co
r
e
an
d
it
c
o
n
s
is
ts
o
f
a
m
atr
ix
o
f
f
o
u
r
d
if
f
e
r
en
t
co
m
b
in
atio
n
s
t
r
u
e
n
e
g
ativ
e
(
T
N)
,
f
alse
n
eg
ativ
e
(
FN)
,
f
alse
p
o
s
itiv
e
(
FP
)
,
an
d
t
r
u
e
p
o
s
itiv
e
(TP
)
o
f
p
r
e
d
icted
an
d
ac
tu
al
v
al
u
es
o
n
Fig
u
r
e
2
C
o
n
f
u
s
io
n
m
atr
ix
.
=
+
+
+
+
×
100
(
1
)
=
+
(
2)
=
+
(
3)
1
−
=
2
×
+
(
4)
Fig
u
r
e
2
.
T
h
e
c
o
m
p
o
n
e
n
ts
o
f
2
×
2
co
n
f
u
s
io
n
m
at
r
ix
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
I
n
th
is
s
ec
tio
n
,
th
e
s
u
p
er
v
is
ed
lear
n
in
g
alg
o
r
ith
m
s
im
p
lem
e
n
ted
in
clu
d
e
L
GB
M,
C
B
,
LR
,
RF
,
GB
,
an
d
XGBo
o
s
t
.
Ad
d
itio
n
ally
,
an
en
s
em
b
le
lear
n
in
g
alg
o
r
ith
m
,
s
tack
in
g
,
is
u
tili
ze
d
to
co
m
b
in
e
th
e
s
tr
en
g
th
s
o
f
m
u
ltip
le
m
o
d
els.
E
ac
h
o
f
th
es
e
alg
o
r
ith
m
s
is
s
elec
ted
f
o
r
its
ef
f
ec
tiv
en
ess
in
h
an
d
lin
g
d
if
f
er
en
t
d
ata
p
atter
n
s
an
d
im
p
r
o
v
in
g
m
o
d
el
p
er
f
o
r
m
an
ce
.
3
.
1
.
L
ig
ht
g
r
a
dient
bo
o
s
t
ing
m
a
chine
Acc
o
r
d
in
g
to
th
e
L
GB
M
a
lg
o
r
ith
m
co
n
f
u
s
io
n
m
atr
ix
in
Fig
u
r
e
3
,
th
er
e
wer
e
1
1
1
2
7
9
in
s
tan
ce
s
wh
er
e
th
e
m
o
d
el
ac
cu
r
ately
p
r
ed
icte
d
th
e
p
o
s
itiv
e
class
,
also
k
n
o
wn
as
T
P.
A
to
tal
o
f
0
o
cc
u
r
r
en
ce
s
wer
e
r
e
co
r
d
e
d
in
wh
ich
th
e
m
o
d
el
ac
cu
r
ately
p
r
ed
icted
th
e
n
e
g
ativ
e
class
,
also
k
n
o
wn
as
T
N.
T
h
e
n
u
m
b
er
o
f
o
cc
u
r
r
en
ce
s
in
wh
ich
th
e
m
o
d
el
p
r
o
d
u
ce
d
f
al
s
e
p
r
ed
ictio
n
s
b
y
ca
teg
o
r
izi
n
g
th
em
as
p
o
s
itiv
e
(
T
y
p
e
I
er
r
o
r
)
,
g
e
n
er
ally
k
n
o
wn
as
FP
,
is
0
.
T
h
e
ca
p
t
u
r
ed
n
u
m
b
er
o
f
i
n
s
tan
ce
s
in
wh
ic
h
t
h
e
m
o
d
el
g
e
n
er
ated
in
ac
cu
r
ate
p
r
ed
ictio
n
s
f
o
r
th
e
n
eg
ativ
e
class
,
k
n
o
wn
as
T
y
p
e
I
I
er
r
o
r
o
r
FN,
is
1
7
9
4
.
T
h
e
ac
cu
r
ac
y
v
alu
e
is
0
.
9
8
4
1
3
,
p
r
ec
is
io
n
v
alu
e
is
r
ep
o
r
ted
as 1
.
0
,
t
h
e
r
ec
all
v
alu
e
as 0
.
9
8
4
1
3
,
a
n
d
th
e
F1
-
s
co
r
e
v
alu
e
as 0
.
9
9
2
0
0
.
3
.
2
.
Ca
t
B
o
o
s
t
T
h
e
n
u
m
b
er
o
f
in
s
tan
ce
s
wh
en
th
e
m
o
d
el
co
r
r
ec
tly
p
r
e
d
i
cted
th
e
p
o
s
itiv
e
class
,
k
n
o
w
n
as
T
P,
is
1
0
0
7
8
9
,
as
s
h
o
w
n
in
th
e
co
n
f
u
s
io
n
m
atr
i
x
C
B
Fig
u
r
e
4
.
T
h
er
e
wer
e
n
o
in
s
tan
ce
s
in
wh
ich
th
e
m
o
d
el
s
u
cc
ess
f
u
lly
p
r
ed
icted
th
e
n
eg
ativ
e
class
,
also
r
ef
er
r
ed
to
as T
N.
T
h
e
q
u
a
n
tity
o
f
in
s
tan
ce
s
in
wh
ich
th
e
m
o
d
e
l
m
ad
e
an
in
ac
c
u
r
ate
p
r
e
d
ictio
n
o
f
th
e
p
o
s
itiv
e
class
(
T
y
p
e
I
er
r
o
r
)
r
ef
e
r
r
ed
to
as
FP
,
is
0
.
T
h
e
n
u
m
b
e
r
o
f
in
s
tan
ce
s
in
wh
ich
th
e
m
o
d
el
p
r
o
d
u
ce
d
in
ac
cu
r
ate
p
r
ed
icti
o
n
s
f
o
r
th
e
n
eg
ativ
e
class
,
class
if
ied
as
T
y
p
e
I
I
er
r
o
r
s
o
r
FN,
is
1
2
2
8
4
.
T
h
e
a
cc
u
r
ac
y
n
u
m
b
er
is
0
.
8
9
1
3
6
,
wh
ile
th
e
p
r
ec
is
io
n
1
.
0
,
r
ec
all
0
.
8
9
1
3
6
,
an
d
F1
-
s
co
r
e
v
alu
es a
r
e
0
.
9
4
2
5
6
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
E
n
h
a
n
ci
n
g
u
n
co
lla
tera
liz
ed
lo
a
n
r
is
k
a
s
s
ess
men
t a
cc
u
r
a
cy
th
r
o
u
g
h
fe
a
tu
r
e
… (
S
h
a
h
r
u
l Ni
z
a
m
S
a
la
h
u
d
i
n)
1155
Fig
u
r
e
3
.
C
o
n
f
u
s
io
n
m
atr
i
x
o
f
L
GB
M
Fig
u
r
e
4
.
C
o
n
f
u
s
io
n
m
atr
i
x
o
f
CB
3.
3
.
L
o
g
is
t
ic
re
g
re
s
s
io
n
Fig
u
r
e
5
d
em
o
n
s
tr
ates
th
at
th
e
co
n
f
u
s
io
n
m
etr
ic
f
o
r
LR
m
o
d
el
’
s
ac
cu
r
ac
y
in
p
r
ed
ictin
g
o
u
tco
m
es,
ca
teg
o
r
ized
b
y
class
,
an
d
in
c
lu
d
es
th
e
co
u
n
ts
o
f
b
o
t
h
co
r
r
ec
t
an
d
in
c
o
r
r
ec
t
p
r
ed
ictio
n
s
.
Acc
o
r
d
in
g
to
th
e
co
n
f
u
s
io
n
m
atr
ix
L
R
,
th
e
co
u
n
t
o
f
o
cc
u
r
r
e
n
ce
s
in
wh
ich
th
e
m
o
d
el
ac
cu
r
ately
p
r
ed
icted
th
e
p
o
s
itiv
e
class
,
also
k
n
o
w
n
as T
P,
is
6
1
6
0
1
.
T
h
er
e
was a
to
tal
o
f
0
ca
s
es in
wh
ich
th
e
m
o
d
el
ac
cu
r
ately
p
r
ed
icted
th
e
n
eg
ati
v
e
class
,
o
f
ten
k
n
o
w
n
as
T
N.
T
h
e
n
u
m
b
er
o
f
ca
s
es
in
wh
ich
th
e
m
o
d
el
wr
o
n
g
ly
p
r
ed
icte
d
th
e
p
o
s
itiv
e
class
(
T
y
p
e
I
er
r
o
r
)
,
also
k
n
o
wn
as
FP
,
is
0
.
T
h
e
co
u
n
t
o
f
o
cc
u
r
r
en
ce
s
in
w
h
ich
th
e
m
o
d
el
m
ad
e
in
ac
c
u
r
ate
p
r
ed
ictio
n
s
f
o
r
th
e
n
e
g
ativ
e
class
,
o
f
ten
k
n
o
wn
as
T
y
p
e
I
I
er
r
o
r
o
r
FN,
is
5
1
4
7
2
.
T
h
e
a
cc
u
r
ac
y
v
alu
e
is
0
.
5
4
4
7
9
,
t
h
e
p
r
ec
is
io
n
v
alu
e
is
1
.
0
,
th
e
r
ec
all
v
alu
e
is
0
.
5
4
4
7
9
,
an
d
t
h
e
F1
-
s
co
r
e
v
alu
e
is
0
.
7
0
5
3
3
.
Fig
u
r
e
5
.
C
o
n
f
u
s
io
n
m
atr
i
x
o
f
LR
3.
4
.
R
a
nd
o
m
f
o
re
s
t
B
ased
o
n
th
e
co
n
f
u
s
io
n
m
at
r
ix
p
r
esen
ted
in
Fig
u
r
e
6
,
t
h
e
R
F
alg
o
r
ith
m
p
r
o
p
e
r
ly
p
r
e
d
icted
th
e
p
o
s
itiv
e
class
,
r
ef
er
r
e
d
to
as
T
P,
in
a
to
tal
o
f
1
1
0
3
5
4
o
cc
u
r
r
en
ce
s
.
A
cu
m
u
lativ
e
co
u
n
t
o
f
0
i
n
s
tan
ce
s
was
o
b
s
er
v
ed
in
wh
ich
th
e
m
o
d
el
s
u
cc
ess
f
u
lly
p
r
ed
icted
th
e
n
eg
ativ
e
class
,
co
m
m
o
n
ly
r
ef
e
r
r
ed
to
as
T
N.
T
h
e
q
u
an
tity
o
f
in
s
tan
ce
s
in
wh
ic
h
th
e
m
o
d
el
m
ad
e
an
in
ac
cu
r
ate
p
r
ed
ictio
n
o
f
th
e
p
o
s
itiv
e
class
(
T
y
p
e
I
er
r
o
r
)
,
wid
ely
k
n
o
wn
as
FP
,
is
0
.
T
h
e
f
r
eq
u
e
n
cy
o
f
in
s
tan
ce
s
in
wh
i
ch
th
e
m
o
d
el
p
r
o
d
u
ce
d
in
ac
cu
r
ate
p
r
ed
ictio
n
s
f
o
r
th
e
n
eg
ativ
e
class
,
r
ef
er
r
ed
to
as T
y
p
e
I
I
e
r
r
o
r
o
r
FN,
am
o
u
n
ts
to
2
7
1
9
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
8
,
No
.
2
,
May
20
2
5
:
1
1
4
9
-
1
1
6
1
1156
T
h
e
Acc
u
r
ac
y
m
etr
ic
is
r
ep
o
r
ted
as
0
.
9
7
5
9
5
,
in
d
icatin
g
th
e
p
r
o
p
o
r
tio
n
o
f
c
o
r
r
ec
tly
class
if
ied
in
s
tan
ce
s
in
th
e
d
ataset.
T
h
e
Pre
cisi
o
n
m
etr
ic
is
r
ep
o
r
ted
as
1
.
0
,
r
ep
r
esen
tin
g
t
h
e
p
r
o
p
o
r
tio
n
o
f
co
r
r
ec
tly
p
r
ed
icted
p
o
s
itiv
e
in
s
t
an
ce
s
o
u
t
o
f
all
in
s
tan
ce
s
p
r
ed
icte
d
as
p
o
s
itiv
e.
T
h
e
R
ec
all
m
etr
ic
is
r
ep
o
r
ted
as
0
.
9
7
5
9
5
,
in
d
icatin
g
th
e
p
r
o
p
o
r
tio
n
o
f
co
r
r
ec
tly
p
r
ed
icted
p
o
s
itiv
e
in
s
tan
ce
s
o
u
t
o
f
a
ll
ac
tu
al
p
o
s
itiv
e
in
s
tan
ce
s
.
L
astl
y
,
th
e
F1
-
s
co
r
e
m
etr
ic
is
r
ep
o
r
ted
as
0
.
9
8
7
8
3
,
wh
ich
is
th
e
h
ar
m
o
n
ic
m
ea
n
o
f
Pre
cisi
o
n
an
d
R
ec
all,
p
r
o
v
id
in
g
an
o
v
er
all
m
ea
s
u
r
e
o
f
th
e
m
o
d
el
’
s
p
e
r
f
o
r
m
an
ce
.
Fig
u
r
e
6
.
C
o
n
f
u
s
io
n
m
atr
i
x
o
f
R
F
3.
5
.
G
ra
dient
bo
o
s
t
ing
Acc
o
r
d
in
g
to
th
e
co
n
f
u
s
io
n
m
atr
ix
d
e
p
icted
in
Fig
u
r
e
7
,
th
e
GB
m
eth
o
d
ac
cu
r
ately
id
en
tifie
d
in
s
tan
ce
s
b
elo
n
g
in
g
to
th
e
p
o
s
itiv
e
class
,
d
en
o
ted
as
T
P,
in
1
0
6
1
3
8
i
n
s
tan
ce
s
.
A
to
tal
o
f
0
o
cc
u
r
r
en
ce
s
wer
e
r
ec
o
r
d
e
d
in
wh
ich
th
e
m
o
d
el
ac
cu
r
ately
p
r
ed
icted
th
e
n
e
g
ativ
e
class
,
also
k
n
o
wn
as
T
N.
T
h
e
n
u
m
b
er
o
f
o
cc
u
r
r
e
n
ce
s
in
wh
ich
th
e
m
o
d
el
p
r
o
d
u
ce
d
an
in
co
r
r
ec
t
f
o
r
ec
ast
o
f
th
e
p
o
s
itiv
e
class
(
T
y
p
e
I
er
r
o
r
)
,
o
f
te
n
k
n
o
wn
as
FP
,
is
ze
r
o
.
T
h
e
o
cc
u
r
r
en
ce
r
ate
o
f
er
r
o
n
eo
u
s
p
r
ed
ictio
n
s
f
o
r
th
e
n
eg
ativ
e
cla
s
s
,
o
f
ten
k
n
o
wn
a
s
T
y
p
e
I
I
m
is
tak
e
o
r
FN,
is
6
9
3
5
.
T
h
e
a
cc
u
r
ac
y
m
ea
s
u
r
e
is
p
r
o
v
id
ed
as
0
.
9
3
8
6
7
,
wh
ic
h
s
ig
n
if
ies
th
e
p
r
o
p
o
r
tio
n
o
f
i
n
s
tan
ce
s
in
th
e
d
ataset
th
at
h
av
e
b
ee
n
p
r
o
p
e
r
ly
ca
teg
o
r
ized
.
T
h
e
p
r
ec
is
io
n
m
ea
s
u
r
e
is
p
r
o
v
id
ed
as
1
.
0
,
in
d
icatin
g
th
e
r
atio
o
f
ac
cu
r
ately
p
r
e
d
icted
p
o
s
itiv
e
in
s
tan
ce
s
to
th
e
to
tal
n
u
m
b
er
o
f
in
s
tan
ce
s
p
r
o
jecte
d
as
p
o
s
itiv
e.
T
h
e
r
ec
all
m
ea
s
u
r
e
is
r
ep
o
r
ted
as
0
.
9
3
8
6
7
,
d
en
o
tin
g
th
e
r
atio
o
f
ac
cu
r
ately
an
ticip
ated
p
o
s
itiv
e
in
s
tan
ce
s
to
th
e
to
tal
n
u
m
b
er
o
f
ac
tu
al
p
o
s
itiv
e
in
s
tan
ce
s
.
Fin
ally
,
th
e
F1
-
s
c
o
r
e
m
etr
ic
is
s
h
o
wn
as
0
.
9
6
8
3
6
,
r
e
p
r
esen
tin
g
th
e
h
ar
m
o
n
ic
m
ea
n
o
f
p
r
ec
is
io
n
an
d
r
ec
all.
T
h
is
m
etr
ic
s
er
v
es
as
a
co
m
p
r
eh
en
s
iv
e
ass
ess
m
en
t
o
f
t
h
e
m
o
d
el
’
s
p
er
f
o
r
m
an
ce
.
Fig
u
r
e
7
.
C
o
n
f
u
s
io
n
m
atr
i
x
o
f
GB
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
E
n
h
a
n
ci
n
g
u
n
co
lla
tera
liz
ed
lo
a
n
r
is
k
a
s
s
ess
men
t a
cc
u
r
a
cy
th
r
o
u
g
h
fe
a
tu
r
e
… (
S
h
a
h
r
u
l Ni
z
a
m
S
a
la
h
u
d
i
n)
1157
3.
6
.
XG
B
o
o
s
t
T
h
e
n
u
m
b
er
o
f
i
n
s
tan
ce
s
wh
er
e
th
e
m
o
d
el
co
r
r
ec
tly
p
r
ed
ict
ed
th
e
p
o
s
itiv
e
class
,
co
m
m
o
n
ly
k
n
o
w
n
as
T
P,
was
1
1
1
3
9
8
,
as
s
h
o
wn
in
th
e
o
r
ig
in
al
XGBo
o
s
t
a
lg
o
r
ith
m
co
n
f
u
s
io
n
m
atr
ix
Fig
u
r
e
8
.
A
cu
m
u
lativ
e
co
u
n
t
o
f
0
in
s
tan
ce
s
was
o
b
s
er
v
ed
in
w
h
ich
t
h
e
m
o
d
el
s
u
cc
es
s
f
u
lly
p
r
ed
icted
th
e
n
eg
ati
v
e
class
,
co
m
m
o
n
ly
r
ef
er
r
ed
to
as
T
N.
T
h
e
q
u
a
n
tity
o
f
in
s
tan
ce
s
in
wh
ich
th
e
m
o
d
el
m
a
d
e
i
n
ac
cu
r
ate
p
r
e
d
ictio
n
s
b
y
class
if
y
in
g
th
em
as
p
o
s
itiv
e
(
T
y
p
e
I
er
r
o
r
)
,
co
m
m
o
n
ly
r
ef
er
r
ed
to
as
FP
,
is
0
.
T
h
e
q
u
an
tity
o
f
in
s
tan
ce
s
in
wh
ich
t
h
e
m
o
d
el
p
r
o
d
u
ce
d
er
r
o
n
eo
u
s
f
o
r
ec
asts
f
o
r
n
eg
ativ
e
class
,
co
m
m
o
n
ly
r
e
f
er
r
ed
t
o
as
T
y
p
e
I
I
er
r
o
r
o
r
FN,
is
r
ec
o
r
d
e
d
as
1
6
7
5
.
T
h
e
p
r
ec
is
i
o
n
v
alu
e
is
r
ep
o
r
ted
as
1
.
0
,
th
e
R
ec
all
v
alu
e
as
0
.
9
8
5
1
9
,
an
d
th
e
F1
-
s
co
r
e
v
alu
e
as 0
.
9
9
2
5
4
.
T
h
e
h
i
g
h
est
co
u
n
t o
f
ac
cu
r
ac
y
co
m
p
a
r
ed
to
th
e
L
R
,
XGBo
o
s
t
,
R
F,
GB
an
d
C
B
alg
o
r
ith
m
s
,
XGBo
o
s
t
h
as
a
s
ig
n
if
ican
t
im
p
ac
t
o
n
th
e
ac
h
iev
ed
ac
cu
r
ac
y
0
.
9
8
5
1
9
,
wh
ich
is
th
e
b
est
ac
cu
r
ac
y
wh
en
co
m
p
ar
ed
to
th
e
o
th
er
f
iv
e
alg
o
r
ith
m
s
.
T
h
e
ca
lcu
latio
n
r
esu
lts
f
o
r
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
a
n
d
F1
-
s
co
r
e
ar
e
p
r
esen
ted
in
T
ab
le
5
.
Fig
u
r
e
8
.
C
o
n
f
u
s
io
n
m
atr
i
x
o
f
XGBo
o
s
t
T
ab
le
5
.
Su
m
m
a
r
y
o
f
co
n
f
u
s
io
n
m
etr
ics
A
l
g
o
r
i
t
h
m
A
c
c
u
r
a
c
y
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F1
-
sc
o
r
e
LG
B
M
0
.
9
8
4
1
3
1
.
0
0
.
9
8
4
1
3
0
.
9
9
2
0
0
CB
0
.
8
9
1
3
6
1
.
0
0
.
8
9
1
3
6
0
.
9
4
2
5
6
LR
0
.
5
4
4
7
9
1
.
0
0
.
5
4
4
7
9
0
.
7
0
5
3
3
RF
0
.
9
7
5
9
5
1
.
0
0
.
9
7
5
9
5
0
.
9
8
7
8
3
GB
0
.
9
3
7
0
7
1
.
0
0
.
9
3
7
0
7
0
.
9
6
7
5
1
X
G
B
o
o
st
0
.
9
8
5
1
9
1
.
0
0
.
9
8
5
1
9
0
.
9
9
2
5
4
Acc
o
r
d
in
g
t
o
th
e
d
ata
s
h
o
wn
in
T
ab
le
1
.
T
h
e
XGBo
o
s
t
alg
o
r
ith
m
d
em
o
n
s
tr
ates
th
e
h
ig
h
est
lev
el
o
f
ac
cu
r
ac
y
,
with
a
n
o
tab
le
ac
c
u
r
ac
y
r
ate
o
f
9
8
.
5
2
%.
XGBo
o
s
t
is
a
GB
alg
o
r
ith
m
th
at
u
s
es
DT
as
wea
k
lear
n
er
s
to
cr
ea
te
a
s
tr
o
n
g
lear
n
er
,
wh
ich
ca
n
h
an
d
le
b
o
th
n
u
m
er
ica
l
an
d
ca
teg
o
r
ical
f
ea
tu
r
es
ef
f
ec
tiv
ely
,
m
ak
in
g
it
s
u
itab
le
f
o
r
a
wid
e
r
an
g
e
o
f
d
atasets
.
T
h
is
i
s
p
ar
ticu
lar
ly
u
s
ef
u
l
f
o
r
th
e
h
o
m
e
c
r
ed
it
d
ataset,
wh
ich
co
n
tain
s
a
m
ix
o
f
n
u
m
er
ical
a
n
d
ca
teg
o
r
i
ca
l f
ea
tu
r
es with
lar
g
e
d
ataset.
On
e
o
f
th
e
n
o
tab
le
s
tr
en
g
th
s
o
f
XGBo
o
s
t
is
its
ef
f
icac
y
in
m
an
ag
in
g
im
b
alan
ce
d
d
atasets
.
T
h
e
h
o
m
e
cr
ed
it
d
ataset
ex
h
ib
its
an
im
b
alan
ce
,
ch
ar
ac
ter
ized
b
y
a
r
el
ativ
ely
lo
w
p
r
o
p
o
r
tio
n
o
f
d
ef
au
lter
s
.
T
h
is
s
tu
d
y
em
p
lo
y
ed
s
tr
ateg
ies
s
u
ch
as
SMOT
E
to
ad
d
r
ess
class
im
b
alan
ce
in
th
e
d
ataset
an
d
en
h
an
ce
th
e
ef
f
icac
y
o
f
GB
alg
o
r
ith
m
s
,
s
p
ec
if
ic
ally
XGBo
o
s
t.
T
h
e
p
r
esen
t
s
tu
d
y
ex
clu
s
iv
ely
em
p
lo
y
s
th
e
SMOT
E
tech
n
iq
u
e
as
a
m
ea
n
s
to
ad
d
r
ess
th
e
is
s
u
e
o
f
i
m
b
alan
ce
d
d
atasets
,
with
o
u
t
u
s
in
g
th
e
ADASYN
(
ad
ap
tiv
e
s
y
n
th
etic
)
a
p
p
r
o
ac
h
u
tili
ze
d
b
y
Ma
h
m
u
d
i
et
a
l.
[
1
7
]
.
Nev
e
r
th
eless
,
th
e
r
esear
ch
m
o
d
el
d
e
m
o
n
s
tr
ates
im
p
r
o
v
ed
ac
cu
r
ac
y
in
its
o
u
tco
m
es th
an
Ma
h
m
u
d
i
et
a
l
.
[
1
7
]
f
in
d
in
g
u
s
in
g
XGBo
o
s
t
.
Mo
r
eo
v
er
,
XGBo
o
s
t
is
r
en
o
w
n
ed
f
o
r
its
n
o
ta
b
le
ef
f
icien
c
y
an
d
s
ca
lab
ilit
y
,
en
a
b
lin
g
it
to
ef
f
ec
tiv
ely
m
an
ag
e
ex
ten
s
iv
e
d
atasets
in
clu
d
in
g
a
s
u
b
s
tan
tial
n
u
m
b
e
r
o
f
f
ea
tu
r
es.
T
h
e
h
o
m
e
cr
ed
i
t
d
atas
et,
with
3
0
f
ea
tu
r
es
an
d
3
0
7
,
5
0
6
ca
p
tu
r
es,
o
f
f
e
r
s
s
ig
n
if
ican
t
a
d
v
an
ta
g
es.
I
n
co
n
clu
s
io
n
,
th
e
n
o
tab
le
ac
c
u
r
ac
y
o
f
XGBo
o
s
t
in
b
in
ar
y
class
class
if
icatio
n
u
tili
zin
g
th
e
d
ataset
m
ay
b
e
a
s
cr
ib
ed
to
its
ca
p
ac
ity
t
o
m
a
n
ag
e
m
ix
e
d
f
ea
t
u
r
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
8
,
No
.
2
,
May
20
2
5
:
1
1
4
9
-
1
1
6
1
1158
ty
p
es,
its
ef
f
icac
y
in
ad
d
r
ess
in
g
im
b
alan
ce
d
d
atasets
,
an
d
its
ef
f
icien
cy
an
d
s
ca
lab
ilit
y
in
m
an
ag
in
g
ex
ten
s
iv
e
d
atasets
with
n
u
m
e
r
o
u
s
f
ea
tu
r
es.
So
,
XGBo
o
s
t
is
a
h
i
g
h
ly
r
ec
o
m
m
en
d
e
d
m
eth
o
d
f
o
r
b
in
ar
y
class
class
if
icatio
n
in
th
e
co
n
tex
t o
f
ass
es
s
in
g
d
ef
au
lt r
is
k
,
p
ar
ticu
lar
ly
wh
en
a
p
p
lied
to
h
o
m
e
c
r
ed
it d
atasets
.
3.
7
.
St
a
ck
ing
L
R
was
u
s
ed
as
th
e
f
in
al
esti
m
ato
r
in
th
is
r
esear
ch
b
ec
au
s
e
its
r
eliab
ilit
y
is
m
o
s
t
ev
id
en
t
in
its
ca
p
ac
ity
to
s
elec
t
m
an
a
g
ea
b
le
s
u
b
s
ets
o
f
in
d
icato
r
s
[
3
7
]
an
d
Pro
v
id
es
p
r
o
b
a
b
ilit
ies
f
o
r
o
u
tco
m
es,
wh
ich
ca
n
b
e
u
s
ef
u
l
f
o
r
d
ec
is
io
n
-
m
ak
i
n
g
[
3
8
]
.
L
R
is
a
s
tatis
tical
m
o
d
elin
g
tech
n
iq
u
e
u
s
ed
f
o
r
p
r
ed
ictiv
e
an
aly
s
is
.
T
h
is
m
eth
o
d
o
lo
g
y
is
em
p
lo
y
ed
to
el
u
cid
ate
th
e
ass
o
ciatio
n
b
etwe
en
d
is
cr
ete
b
in
a
r
y
v
a
r
iab
les.
L
R
ex
h
ib
its
s
en
s
itiv
ity
to
th
e
p
r
esen
ce
o
f
m
u
ltiv
ar
iate
co
llin
ea
r
ity
am
o
n
g
th
e
in
d
ep
en
d
en
t
v
ar
ia
b
le
s
with
in
th
e
m
o
d
el,
wh
er
ein
th
e
in
f
lu
e
n
ce
o
f
a
s
in
g
le
v
ar
iab
le
ca
n
s
ig
n
if
ican
tly
i
m
p
ac
t
th
e
o
th
er
v
a
r
iab
les.
W
h
en
co
n
f
r
o
n
ted
with
a
m
u
ltit
u
d
e
o
f
f
ac
to
r
s
,
th
e
r
esu
ltin
g
p
er
f
o
r
m
a
n
ce
m
a
y
n
o
t
m
ee
t
e
x
p
ec
tatio
n
s
.
T
h
e
r
e
m
ar
k
el
u
cid
ates
th
e
r
atio
n
ale
f
o
r
th
e
v
er
y
m
o
d
est
ac
cu
r
ac
y
o
u
tco
m
e
o
f
LR
,
wh
i
ch
r
ec
o
r
d
ed
a
v
alu
e
o
f
0
.
5
4
4
7
9
,
in
th
e
c
o
n
tex
t
o
f
b
in
ar
y
class
if
icatio
n
in
s
id
e
th
is
r
esear
ch
.
T
h
is
p
er
f
o
r
m
a
n
c
e
was
co
m
p
ar
ativ
e
in
co
m
p
a
r
is
o
n
to
th
e
r
esu
lts
ac
h
iev
ed
b
y
alter
n
ativ
e
m
eth
o
d
o
lo
g
ies em
p
lo
y
ed
in
th
is
in
v
e
s
tig
atio
n
.
T
h
e
o
u
tco
m
es
o
f
e
m
p
lo
y
i
n
g
t
h
e
S
tack
in
g
tech
n
iq
u
e,
wh
ich
in
teg
r
ates
L
GB
M,
R
F,
an
d
GB
m
o
d
els,
with
L
R
s
er
v
in
g
as
th
e
u
ltim
ate
esti
m
ato
r
,
en
co
m
p
ass
th
e
f
o
llo
win
g
m
etr
ics:
ac
cu
r
ac
y
:
0
.
9
9
6
3
7
,
p
r
ec
is
io
n
:
1
.
0
,
r
ec
all:
0
.
9
9
6
3
7
,
an
d
F1
-
s
co
r
e
:
0
.
9
9
8
1
8
d
is
p
lay
in
T
ab
l
e
6
.
T
h
e
o
u
tp
u
ts
o
f
th
e
co
n
f
u
s
io
n
m
atr
ix
r
esu
ltin
g
f
r
o
m
th
e
in
teg
r
atio
n
o
f
L
GB
M
,
R
F,
an
d
GB
m
o
d
els in
th
e
s
tack
in
g
m
o
d
el
ar
e
d
ep
icted
in
Fig
u
r
e
9
.
T
ab
le
6
.
C
o
n
f
u
s
io
n
m
atr
ix
s
tack
in
g
M
e
t
h
o
d
e
A
c
c
u
r
a
c
y
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F1
-
sc
o
r
e
S
t
a
c
k
i
n
g
0
.
9
9
6
3
7
1
.
0
0
.
9
9
6
3
7
0
.
9
9
8
1
8
Fig
u
r
e
9
.
C
o
n
f
u
s
io
n
m
atr
i
x
f
o
r
s
tack
in
g
co
m
b
in
es GB,
R
F
,
an
d
L
GB
M
T
h
e
s
tack
in
g
m
eth
o
d
’
s
o
u
ts
tan
d
in
g
ac
cu
r
ac
y
d
em
o
n
s
tr
ates
its
p
r
o
m
is
e
as
r
eliab
le
in
s
tr
u
m
en
ts
f
o
r
ass
es
s
in
g
u
n
co
llater
alize
d
lo
a
n
r
is
k
.
T
h
is
h
as
im
p
o
r
tan
t
r
a
m
if
icatio
n
s
f
o
r
f
in
an
cial
in
s
ti
tu
tio
n
s
lo
o
k
in
g
to
im
p
r
o
v
e
th
eir
m
o
d
el
ac
cu
r
ac
y
.
Fu
r
th
er
m
o
r
e,
th
e
u
s
e
o
f
GB
DT
with
n
_
esti
m
ato
r
=1
0
0
to
s
elec
t
f
ea
tu
r
e
im
p
o
r
tan
ce
in
class
if
icatio
n
af
f
ec
ts
ac
cu
r
ac
y
.
T
ab
le
7
p
r
esen
ts
a
co
m
p
ar
is
o
n
o
f
th
e
ac
c
u
r
ac
y
ac
h
iev
ed
u
s
in
g
GB
DT
with
n
_
esti
m
ato
r
=1
0
0
ag
ain
s
t
p
r
ev
io
u
s
s
tu
d
ies.
As
d
em
o
n
s
tr
ated
in
T
ab
le
7
,
o
u
r
m
o
d
el
h
a
s
th
e
h
ig
h
est
ac
cu
r
ac
y
o
f
9
9
.
6
4
%,
wh
i
ch
is
m
u
ch
h
ig
h
er
th
an
p
r
io
r
r
esear
ch
’
ac
cu
r
ac
ies
r
an
g
i
n
g
f
r
o
m
7
5
%
to
9
8
%.
T
h
is
im
p
r
o
v
em
en
t
ca
n
b
e
d
u
e
to
GB
DT
’
s
s
tr
in
g
en
t f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
,
wh
ic
h
en
s
u
r
es th
at
th
e
m
o
d
el
co
n
tain
s
o
n
ly
th
e
m
o
s
t u
s
ef
u
l f
ea
tu
r
es.
T
h
is
s
tu
d
y
e
m
p
lo
y
e
d
s
tr
ateg
i
es
s
u
ch
as
SMOT
E
to
ad
d
r
e
s
s
class
im
b
alan
ce
in
th
e
d
a
taset
an
d
en
h
an
ce
th
e
e
f
f
icac
y
o
f
GB
alg
o
r
ith
m
s
,
s
p
ec
if
ically
XGBo
o
s
t.
T
h
e
p
r
esen
t
s
tu
d
y
ex
clu
s
iv
ely
em
p
lo
y
s
th
e
SMOT
E
tech
n
iq
u
e
with
o
u
t
u
s
in
g
th
e
ADASYN
(
Ad
ap
tiv
e
Sy
n
th
etic)
ap
p
r
o
ac
h
u
tili
ze
d
b
y
Ma
h
m
u
d
i
[
1
7
]
.
Nev
er
th
eless
,
th
e
r
esear
ch
m
o
d
el
d
em
o
n
s
tr
ates
im
p
r
o
v
e
d
ac
cu
r
ac
y
in
its
o
u
tco
m
es.
LR
w
as
u
s
ed
as
th
e
f
in
al
esti
m
ato
r
in
s
tack
in
g
r
esear
c
h
m
eth
o
d
e
b
ec
a
u
s
e
its
r
eliab
ilit
y
is
m
o
s
t
ev
id
en
t
in
its
ca
p
ac
ity
to
s
elec
t
Evaluation Warning : The document was created with Spire.PDF for Python.