I
A
E
S
I
n
t
e
r
n
at
io
n
al
Jou
r
n
al
of
A
r
t
if
ic
ia
l
I
n
t
e
ll
ig
e
n
c
e
(
I
J
-
AI
)
V
ol
.
13
, N
o.
4
,
D
e
c
e
m
be
r
2024
, pp.
4054
~
4061
I
S
S
N
:
2252
-
8938
,
D
O
I
:
10.11591/
ij
a
i.
v
13
.i
4
.pp
4054
-
4061
4054
Jou
r
n
al
h
om
e
page
:
ht
tp
:
//
ij
ai
.
ia
e
s
c
or
e
.c
om
E
n
h
a
n
c
i
n
g i
n
t
e
r
p
r
e
t
ab
i
l
i
t
y i
n
r
an
d
o
m
f
or
e
st
:
L
e
ve
r
agi
n
g
i
n
T
r
e
e
s f
or
ass
oc
i
at
i
on
r
u
l
e
e
x
t
r
ac
t
i
on
i
n
si
g
h
t
s
F
at
m
a H
il
al
i
M
oh
’
d
,
K
h
ai
r
il
A
n
w
ar
N
ot
od
ip
u
t
r
o, Y
e
n
n
i
A
n
gr
ai
n
i
D
e
pa
r
t
m
e
nt
of
S
t
a
t
i
s
t
i
c
s
,
F
a
c
ul
t
y of
M
a
t
he
m
a
t
i
c
s
a
nd
N
a
t
ur
a
l
S
c
i
e
nc
e
s
, I
P
B
U
n
i
ve
r
s
i
t
y, B
ogor
, I
ndone
s
i
a
A
r
t
ic
le
I
n
f
o
A
B
S
T
R
A
C
T
A
r
ti
c
le
h
is
to
r
y
:
R
e
c
e
iv
e
d
D
e
c
29, 2023
R
e
vi
s
e
d
F
e
b 29, 2024
A
c
c
e
pt
e
d
M
a
r
21, 2024
The
random
forest
model
is
a
powerful
supervised
learner,
recognized
for
its
abilit
y
to
learn
the
pattern
within
data
with
superior
predicti
ve
acc
uracy.
However,
it
is
a
black
box
model
because
it
lacks
interpretability
.
Thi
s
study
addressed
the
interpret
able
challenge
by
employi
ng
the
inTree
fram
ework.
The
rules
were
extracted
from
each
decision
tree
in
a
random
forest
model,
and
the
association
rules
were
determined
through
measured
matrix
s
upport
and
confidence
to
reveal
th
e
frequent
variable
interaction
s
for
pre
dicting
unemployment.
This
approach
provided
insight
into
the
relatio
nships
between
specific
variables
and
unemployment
outcomes.
The
dev
eloped
method
used
data
set
from
the
integrated
labor
force
survey
(ILFS)
2020/2021
in
Zanzibar.
Zanzibar’s
unemployment
rate
consi
stently
increase
d
across
surveys
conducted
in
2006,
2014,
and
2020/2021.
Results
have
shown
that
the
rules
that
most
predict
unemployment
for
individuals
are
female
and
lack
of
health
insurance
and
secondary
education
level,
female
and
youth
age
group
and
lack
of
health
insurance
and
sec
ondary
education
level
with
a
high
confidence
level.
This
study
provides
pr
actical
insights
for
Zanzibar
’s
governmen
t
to
develop
effec
tive
interve
ntions,
programs,
and
policies.
Improving
the
interpretability
of
the
random
forest
model enha
nces dec
ision
-
making to add
ress unemplo
yment challe
nges.
K
e
y
w
o
r
d
s
:
A
s
s
oc
ia
ti
on r
ul
e
E
xt
r
a
c
ti
on r
ul
e
In
T
r
e
e
s
R
a
ndom f
or
e
s
t
U
ne
m
pl
oym
e
nt
This is an
open
acce
ss artic
le unde
r the
CC BY
-
SA
license.
C
or
r
e
s
pon
di
n
g A
u
th
or
:
K
ha
ir
il
A
nw
a
r
N
ot
odi
put
r
o
D
e
pa
r
tm
e
nt
of
S
ta
ti
s
ti
c
s
, F
a
c
ul
ty
of
M
a
th
e
m
a
ti
c
s
a
nd N
a
tu
r
a
l
S
c
ie
nc
e
S
c
ie
nc
e
s
, I
P
B
U
ni
ve
r
s
it
y
B
ogor
, I
ndone
s
ia
E
m
a
il
:
kha
ir
il
@
a
pps
.i
pb.a
c
.i
d
1.
I
N
T
R
O
D
U
C
T
I
O
N
E
ns
e
m
bl
e
tr
e
e
s
[
1]
‒[
4]
a
r
e
ve
r
y
pow
e
r
f
ul
s
upe
r
vi
s
e
d
le
a
r
ne
r
s
,
r
e
c
ogni
z
e
d
f
or
th
e
ir
hi
ghl
y
e
xc
e
pt
io
na
l
pr
e
di
c
ti
ve
pr
e
c
is
io
n
due
to
c
a
pt
ur
in
g
in
f
or
m
a
ti
on
in
da
ta
.
T
he
r
a
ndom
f
or
e
s
t
m
ode
l
is
th
e
m
os
t
us
e
d
e
ns
e
m
bl
e
tr
e
e
due
to
th
e
ir
a
dj
us
ta
bl
e
pa
r
a
m
e
te
r
s
,
e
a
s
e
of
us
e
,
a
nd
c
a
pa
c
it
y
to
m
a
na
ge
li
m
it
e
d
s
a
m
pl
e
s
iz
e
a
nd
hi
gh
-
di
m
e
ns
io
na
l
f
e
a
tu
r
e
s
pa
c
e
s
[
5]
,
[
6]
.
D
ue
to
th
e
ir
hi
gh
pr
e
di
c
ti
ve
pe
r
f
o
r
m
a
nc
e
,
th
e
y
a
r
e
c
ons
id
e
r
e
d
a
m
u
s
t
-
tr
y
m
e
th
od
f
or
s
ol
vi
ng
r
e
a
l
pr
obl
e
m
s
.
T
he
y
ha
ve
a
c
hi
e
ve
d
out
s
t
a
ndi
ng
s
c
or
e
s
in
va
r
io
us
da
ta
m
in
in
g c
om
pe
ti
ti
ons
, i
nc
lu
di
ng pr
e
di
c
ti
ng t
he
e
m
pl
oym
e
nt
s
ta
tu
s
of
j
ob
a
ppl
ic
a
nt
s
[
7]
, [
8]
.
H
ow
e
ve
r
, r
a
ndom f
or
e
s
t
is
t
he
m
os
t
c
om
pe
ti
ti
ve
m
ode
l;
i
t
ha
s
be
e
n c
r
it
ic
iz
e
d f
or
be
in
g a
‘
bl
a
c
k box
’
m
ode
l
due
to
it
s
c
om
pl
e
xi
ty
[
9]
‒
[
11]
.
B
e
c
a
us
e
r
a
ndom
f
or
e
s
ts
c
om
bi
ne
m
ul
ti
pl
e
de
c
is
io
n
tr
e
e
s
w
it
h
di
f
f
e
r
e
nt
r
ul
e
s
,
it
be
c
om
e
s
di
f
f
ic
ul
t
to
in
te
r
p
r
e
t
or
unde
r
s
ta
nd
th
e
r
e
la
ti
o
ns
hi
p
be
twe
e
n
pr
e
di
c
to
r
s
a
nd
pr
e
di
c
ti
ons
[
12]
.
T
he
m
ode
l’
s
a
c
c
ur
a
c
y
gr
ow
s
w
it
h
r
ul
e
c
om
pl
e
xi
ty
,
a
lt
hou
gh
th
is
c
om
pl
e
xi
ty
hi
nde
r
s
in
te
r
pr
e
ta
bi
li
ty
.
I
nt
e
r
pr
e
ta
bi
li
ty
is
e
s
s
e
nt
ia
l
f
or
le
a
r
ni
ng
f
r
om
th
e
m
ode
l
to
unde
r
s
ta
nd
th
e
r
e
a
s
on
s
be
hi
nd
de
c
is
io
ns
[
13]
,
[
14]
.
A
n
in
te
r
pr
e
ta
bl
e
m
ode
l
a
ll
ow
s
a
qua
li
ta
ti
ve
unde
r
s
ta
ndi
ng
of
th
e
r
e
la
ti
ons
hi
p
be
twe
e
n
pr
e
di
c
to
r
a
nd
out
c
om
e
va
r
ia
bl
e
s
[
15]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
E
nhanc
in
g i
nt
e
r
pr
e
ta
bi
li
ty
i
n r
andom
fo
r
e
s
t:
L
e
v
e
r
agi
ng i
nT
r
e
e
s
f
or
as
s
oc
ia
ti
on
…
(
F
at
m
a H
il
al
i
M
oh’
d
)
4055
L
ik
e
de
c
is
io
n
tr
e
e
s
,
tr
e
e
-
ba
s
e
d
m
ode
ls
of
f
e
r
in
te
r
pr
e
ta
bi
li
ty
by
tr
a
ns
la
ti
ng
th
e
ir
tr
e
e
s
in
to
r
ul
e
s
,
m
a
ki
ng
th
e
m
e
a
s
ie
r
to
unde
r
s
ta
nd
a
nd
a
ppl
y
[
16]
.
D
e
c
i
s
io
n
r
ul
e
s
ta
k
e
th
e
f
or
m
of
if
(
c
ondi
ti
on)
-
th
e
n(
pr
e
di
c
ti
on)
s
ta
te
m
e
nt
s
[
17]
,
[
18]
.
T
he
c
ondi
ti
on
out
li
n
e
s
s
pe
c
if
ic
in
put
va
r
ia
bl
e
va
lu
e
s
,
w
hi
le
th
e
pr
e
di
c
ti
on
in
di
c
a
te
s
th
e
e
xpe
c
te
d
va
lu
e
of
th
e
out
c
om
e
va
r
ia
bl
e
w
he
n
a
n
obs
e
r
va
ti
on
m
e
e
ts
th
e
gi
ve
n
r
e
qui
r
e
m
e
nt
in
th
a
t
c
ondi
ti
on.
T
he
s
e
r
ul
e
s
op
e
r
a
te
on
a
c
onj
unc
ti
ve
ba
s
i
s
,
r
e
qui
r
in
g
f
ul
f
il
lm
e
nt
of
e
ve
r
y
c
ondi
ti
on;
if
a
n
obs
e
r
va
ti
on
f
a
il
s
to
m
e
e
t
th
e
c
ondi
ti
on,
th
e
r
ul
e
is
not
us
e
d
f
or
th
a
t
obs
e
r
va
ti
on.
R
a
ndom
f
or
e
s
ts
, a
n e
ns
e
m
bl
e
of
de
c
is
io
n t
r
e
e
s
, d
e
m
ons
tr
a
te
va
r
io
us
r
ul
e
pa
tt
e
r
ns
.
A
s
s
oc
ia
ti
on
r
ul
e
s
,
in
tr
oduc
e
d
by
A
gr
a
w
a
l
e
t
al
.
[
19]
,
e
xt
r
a
c
t
c
or
r
e
la
ti
ons
,
f
r
e
que
nt
pa
tt
e
r
ns
,
a
nd
a
s
s
oc
ia
ti
on
s
a
m
ong
s
e
ts
of
it
e
m
s
in
da
ta
ba
s
e
s
.
A
s
s
oc
ia
ti
o
n
r
ul
e
s
a
r
e
“
if
-
th
e
n”
s
ta
te
m
e
nt
s
id
e
nt
if
yi
ng
c
om
bi
na
ti
ons
of
it
e
m
s
f
r
e
que
nt
ly
oc
c
ur
r
in
g
to
ge
th
e
r
in
la
r
ge
da
ta
s
e
ts
.
I
n
th
e
c
ont
e
xt
of
th
i
s
r
e
s
e
a
r
c
h,
th
e
it
e
m
s
e
t
r
e
f
e
r
s
to
th
e
de
c
is
io
n
r
ul
e
s
w
it
hi
n
th
e
r
a
ndom
f
or
e
s
t
m
o
de
l
[
20]
.
P
r
opos
e
d
a
n
in
T
r
e
e
s
f
r
a
m
e
w
or
k
to
in
te
r
pr
e
t
r
a
ndom
f
or
e
s
ts
by
e
xt
r
a
c
ti
ng
r
ul
e
s
,
m
e
a
s
ur
in
g
r
ul
e
s
,
e
xt
r
a
c
ti
ng
f
r
e
que
nt
va
r
ia
bl
e
in
te
r
a
c
ti
on
s
,
a
nd
ha
ndl
in
g
th
e
r
ul
e
s
pr
oduc
e
d
by
th
e
de
c
is
io
n
tr
e
e
s
in
th
e
f
o
r
e
s
t.
T
he
in
T
r
e
e
s
f
r
a
m
e
w
or
k
ha
s
be
e
n
ut
il
iz
e
d
in
va
r
io
us
s
tu
di
e
s
f
or
in
te
r
pr
e
ti
ng
tr
e
e
e
ns
e
m
bl
e
s
,
in
c
lu
di
ng
a
ppl
ic
a
ti
ons
in
br
e
a
s
t
c
a
nc
e
r
di
a
gnos
is
[
21]
a
nd
th
e
pove
r
ty
s
ta
tu
s
of
hous
e
hol
ds
[
12]
.
U
ne
m
pl
oym
e
nt
is
a
s
e
ve
r
e
pr
obl
e
m
th
a
t
ha
s
a
n
im
pa
c
t
on
va
r
io
us
a
s
pe
c
t
s
,
s
uc
h
a
s
f
in
a
nc
e
s
,
m
e
nt
a
l
he
a
lt
h,
a
nd
in
c
r
e
a
s
e
d
s
ui
c
id
e
r
a
te
s
[
22]
.
W
e
ob
s
e
r
ve
d
a
not
a
bl
e
in
c
r
e
a
s
e
in
Z
a
nz
ib
a
r
’
s
une
m
pl
oym
e
nt
r
a
te
in
I
L
F
S
s
ur
ve
ys
c
onduc
te
d
in
2006,
2014,
a
nd
2020/2021
a
t
5.5
%
,
14.3%
,
a
nd
19.6%
,
r
e
s
pe
c
ti
ve
ly
[
23]
.
G
iv
e
n
th
is
s
it
ua
ti
on,
pr
e
di
c
ti
on
is
ne
e
de
d
to
id
e
nt
i
f
y
s
ig
ni
f
ic
a
nt
va
r
ia
bl
e
s
th
a
t
c
ont
r
ib
ut
e
to
une
m
pl
oym
e
nt
to
he
l
p
c
r
e
a
te
be
tt
e
r
jo
b
oppor
tu
ni
ti
e
s
.
T
hi
s
s
tu
dy
is
th
e
f
ir
s
t
in
Z
a
nz
ib
a
r
to
us
e
a
r
a
ndom
f
or
e
s
t
m
ode
l
to
pr
e
di
c
t
une
m
pl
oym
e
nt
s
ta
tu
s
.
I
t
is
a
ls
o
th
e
in
it
ia
l
a
ppl
ic
a
ti
on
of
th
e
in
T
r
e
e
s
f
r
a
m
e
w
or
k
in
th
e
f
ie
ld
of
une
m
pl
oym
e
nt
pr
e
di
c
ti
on, e
nha
nc
in
g i
nt
e
r
pr
e
ta
bi
li
ty
a
nd pr
ovi
di
ng a
uni
que
c
ont
r
ib
ut
io
n t
o t
he
l
it
e
r
a
tu
r
e
.
T
hi
s
s
tu
dy
ha
s
ut
il
iz
e
d
th
e
in
te
gr
a
te
d
la
bor
f
or
c
e
s
ur
ve
y
(
I
L
F
S
)
da
ta
f
r
om
Z
a
nz
ib
a
r
in
2020/
20
21
to
bui
ld
th
e
r
a
ndom
f
or
e
s
t
m
ode
l
f
or
pr
e
di
c
ti
ng
une
m
pl
oym
e
nt
s
ta
tu
s
a
nd
us
e
d
th
e
in
T
r
e
e
s
(
in
te
r
pr
e
ta
bl
e
tr
e
e
s
)
f
r
a
m
e
w
or
k
to
e
xt
r
a
c
t
a
nd
a
s
s
e
s
s
r
ul
e
s
f
r
om
r
a
ndom
f
or
e
s
t
m
ode
ls
,
a
im
in
g
to
unc
ove
r
f
r
e
que
nt
va
r
ia
bl
e
in
te
r
a
c
ti
ons
th
a
t
c
ont
r
ib
ut
e
to
une
m
pl
oym
e
nt
.
T
hi
s
doc
um
e
nt
is
or
ga
ni
z
e
d
a
s
f
ol
lo
w
s
:
s
e
c
ti
on
2
de
s
c
r
ib
e
s
th
e
r
e
s
e
a
r
c
h
m
e
th
od
f
or
th
e
ut
il
iz
e
d
d
a
ta
,
th
e
v
a
r
ia
bl
e
s
,
a
nd
th
e
r
a
ndom
f
or
e
s
t
m
ode
l
a
nd
in
T
r
e
e
f
r
a
m
e
w
or
k
a
ppr
oa
c
he
s
.
I
n
s
e
c
ti
on
3,
r
e
s
ul
ts
a
nd
di
s
c
us
s
io
n
a
r
e
pr
e
s
e
nt
e
d.
F
in
a
ll
y,
th
e
c
onc
lu
s
io
n
a
nd
s
ugge
s
ti
on
of
th
is
pa
pe
r
a
r
e
br
ie
f
ly
out
li
ne
d i
n
s
e
c
ti
ons
4 a
nd 5, r
e
s
pe
c
ti
ve
ly
.
2.
M
E
T
H
O
D
2.1.
D
at
a an
d
var
ia
b
le
s
T
he
ut
il
iz
e
d
da
ta
f
or
th
is
in
ve
s
ti
ga
ti
on
is
s
e
c
ond
a
r
y
da
ta
f
r
om
th
e
I
L
F
S
2020/2021
c
onduc
te
d
by
th
e
of
f
ic
e
of
th
e
C
hi
e
f
G
ove
r
nm
e
nt
S
ta
ti
s
ti
c
ia
n
Z
a
nz
ib
a
r
(
O
C
G
S
)
.
T
he
da
ta
s
e
t
ha
s
9608
obs
e
r
va
ti
on
s
,
in
c
lu
di
ng
in
di
vi
dua
ls
f
r
om
th
e
s
ur
ve
ye
d
hous
e
hol
ds
w
it
hi
n
th
e
w
or
k
in
g
a
ge
r
a
nge
of
15
to
64
ye
a
r
s
.
T
he
out
c
om
e
va
r
ia
bl
e
is
bi
na
r
y:
une
m
pl
oym
e
nt
s
ta
tu
s
(
e
m
pl
oye
d=
0,
un
e
m
p
lo
ye
d=
1)
.
T
he
te
n
pr
e
di
c
to
r
va
r
ia
bl
e
s
u
s
e
d
to
pr
e
di
c
t
une
m
pl
oym
e
nt
s
ta
tu
s
a
r
e
di
s
pl
a
y
e
d i
n T
a
bl
e
1.
T
a
bl
e
1
.
D
a
ta
v
a
r
ia
bl
e
s
V
a
r
i
a
bl
e
s
D
e
s
c
r
i
pt
i
on of
va
r
i
a
bl
e
s
c
a
t
e
gor
i
e
s
s
e
x
S
e
x of
r
e
s
ponde
nt
.
M
a
l
e
=1 F
e
m
a
l
e
=2
a
ge
A
ge
of
r
e
s
ponde
nt
Y
out
h=1 A
dul
t
=2
i
ns
u
H
e
a
l
t
h i
ns
ur
a
nc
e
Y
e
s
=1 N
o=2
di
s
a
b
D
i
s
a
bi
l
i
t
y
Y
e
s
=1 N
o=2
m
a
r
i
t
M
a
r
i
t
a
l
s
t
a
t
us
S
i
ngl
e
=1
M
a
r
r
i
e
d=2
W
i
dow
e
d=3
c
i
t
i
z
C
i
t
i
z
e
ns
hi
p
Y
e
s
=1 N
o=2
r
e
a
d
L
a
ngua
ge
pr
of
i
c
i
e
nc
y
K
i
s
w
a
hi
l
i
=1
E
ngl
i
s
h &
K
i
s
w
a
hi
l
i
=2
C
a
nnot
=3
e
dul
e
v
E
duc
a
t
i
on l
e
ve
l
N
e
ve
r
a
t
t
e
nde
d=1
P
r
i
m
a
r
y=2
S
e
c
onda
r
y=3
V
oc
a
t
i
ona
l
t
r
a
i
ni
ng=4
T
e
r
t
i
a
r
y/
non
-
un
i
ve
r
s
i
t
y
U
ni
ve
r
s
i
t
y=6
t
r
a
i
ni
T
r
a
i
n a
t
t
e
nde
d
N
one
=1 Y
e
s
=2
ur
br
ur
R
e
s
i
de
nc
e
R
ur
a
l
=1 U
r
ba
n=2
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
13
, N
o.
4
,
D
e
c
e
m
be
r
2024
:
4054
-
4061
4056
2.2.
M
od
e
l
A
r
a
ndom
f
o
r
e
s
t
m
ode
l
w
a
s
f
it
te
d
by
e
m
pl
oyi
ng
op
ti
m
a
l
pa
r
a
m
e
te
r
s
obt
a
in
e
d
f
r
om
a
g
r
id
s
e
a
r
c
h
a
ppr
oa
c
h
f
or
pr
e
di
c
ti
ng
une
m
pl
oy
m
e
nt
s
ta
tu
s
.
T
hi
s
m
ode
l
w
a
s
tr
a
in
e
d
w
it
h
100
de
c
is
io
n
tr
e
e
s
a
nd
4
pr
e
di
c
to
r
s
f
or
s
pl
it
ti
ng
a
t
e
a
c
h node
, w
it
h 10 mi
ni
m
um
s
iz
e
s
of
t
e
r
m
in
a
l
node
s
a
nd 200 maximum
numbe
r
s
o
f
te
r
m
in
a
l
node
s
.
F
or
e
a
c
h
de
c
is
io
n
tr
e
e
,
r
a
ndoml
y
s
e
le
c
t
a
s
ub
s
e
t
of
th
e
da
ta
by
boot
s
tr
a
p
m
e
th
od
a
nd
s
e
le
c
t
4
s
e
ts
of
pr
e
di
c
to
r
s
f
or
e
a
c
h
tr
e
e
in
th
e
e
ns
e
m
bl
e
.
A
f
te
r
th
e
m
ode
l
w
a
s
tr
a
in
e
d,
th
e
m
ode
l
us
e
d
out
-
of
-
ba
g
(
O
O
B
)
da
ta
,
w
hi
c
h
m
e
a
ns
da
ta
not
us
e
d
f
or
tr
a
in
in
g,
to
e
s
ti
m
a
te
m
ode
l
pe
r
f
or
m
a
nc
e
.
T
he
ove
r
a
ll
pr
e
di
c
ti
o
n
f
or
e
a
c
h
obs
e
r
va
ti
on
is
obt
a
in
e
d
by
ta
ki
ng
a
m
a
jo
r
it
y
vot
e
f
r
om
in
di
vi
dua
l
tr
e
e
pr
e
di
c
ti
ons
[
24]
.
T
he
m
ode
l
w
a
s
e
va
lu
a
t
e
d on te
s
ti
ng da
ta
u
s
in
g a
c
c
ur
a
c
y,
s
e
n
s
it
iv
it
y, s
pe
c
if
ic
it
y, a
nd a
r
e
a
unde
r
t
he
c
ur
ve
(
A
U
C
)
.
H
ow
e
ve
r
,
r
a
ndom
f
or
e
s
t
is
di
f
f
ic
ul
t
to
in
te
r
pr
e
t
be
c
a
us
e
it
’
s
li
ke
a
bl
a
c
k
box.
W
e
s
ugge
s
te
d
us
in
g
in
T
r
e
e
s
f
or
a
s
s
o
c
ia
ti
on
r
ul
e
e
xt
r
a
c
ti
on
to
m
a
ke
it
m
or
e
in
te
r
pr
e
ta
bl
e
.
T
hi
s
a
ppr
oa
c
h
pr
ovi
de
s
in
s
ig
ht
s
in
to
th
e
m
ode
l
’
s
de
c
is
io
ns
a
nd i
m
pr
ove
s
i
ts
i
nt
e
r
pr
e
ta
bi
li
ty
[
20]
.
2.3.
R
u
le
e
xt
r
ac
t
io
n
f
r
om
d
e
c
is
io
n
t
r
e
e
s
D
e
c
is
io
n
r
ul
e
s
,
a
s
de
s
c
r
ib
e
d
by
[
17]
,
a
r
e
in
va
lu
a
bl
e
f
o
r
s
im
pl
if
yi
ng
de
c
is
io
n
tr
e
e
s
a
nd
im
pr
ov
in
g
th
e
ir
in
te
r
pr
e
ta
bi
li
ty
.
A
de
c
is
io
n
tr
e
e
c
a
n
be
tr
a
ns
f
or
m
e
d
f
r
om
a
tr
e
e
f
or
m
in
to
a
de
c
is
io
n
r
ul
e
by
ta
ki
ng
th
e
f
or
m
“
if
[
c
ondi
ti
on]
th
e
n
[
pr
e
di
c
ti
on
]
.
”
F
ig
ur
e
1
is
a
n
e
xa
m
pl
e
of
e
xt
r
a
c
ti
ng
d
e
c
is
io
n
r
ul
e
s
f
r
om
a
d
e
c
is
io
n
tr
e
e
.
T
he
num
be
r
of
r
ul
e
s
f
or
m
e
d
in
r
a
ndom
f
o
r
e
s
ts
va
r
ie
s
ba
s
e
d
on
th
e
num
be
r
of
f
in
a
l
node
s
in
e
a
c
h
de
c
is
io
n t
r
e
e
.
A
c
c
or
di
ng t
o F
ig
ur
e
1, t
he
r
e
s
ul
t
of
t
he
r
ul
e
s
f
or
m
e
d i
s
a
s
f
ol
lo
w
s
:
r
ul
e
1=
{
(
[
s
a
la
r
y
≥
$100]
)
⇒
(
[
A
c
c
e
pt
j
ob]
)
}
r
ul
e
2=
{
(
[
s
a
la
r
y
<
$100]
)
&
(
[
G
e
nde
r
=
M
a
le
]
)
⇒
(
[
R
e
je
c
t
jo
b]
)
}
r
ul
e
3=
{
(
[
s
a
la
r
y
<
$100]
)
&
(
[
G
e
nde
r
=
F
e
m
a
le
]
)
⇒
(
[
A
c
c
e
pt
j
ob]
)
}
F
ig
ur
e
1
.
D
e
c
is
io
n t
r
e
e
s
tr
uc
tu
r
e
2.4.
T
h
e
in
t
r
e
e
s
f
r
am
e
w
or
k
T
he
in
T
r
e
e
s
f
r
a
m
e
w
or
k
w
a
s
pr
opos
e
d
by
[
20]
to
in
te
r
pr
e
t
r
a
ndom
f
or
e
s
ts
by
e
xt
r
a
c
ti
ng
r
ul
e
s
,
m
e
a
s
ur
in
g
r
ul
e
s
,
e
xt
r
a
c
ti
ng
f
r
e
que
nt
v
a
r
ia
bl
e
in
te
r
a
c
ti
ons
,
a
nd
ha
ndl
in
g
th
e
r
ul
e
s
f
or
m
e
d
by
th
e
de
c
i
s
io
n
tr
e
e
s
in
th
e
f
or
e
s
t.
T
h
e
r
ul
e
qua
li
ty
i
s
m
e
a
s
ur
e
d
by
th
e
le
ngt
h
of
th
e
num
be
r
of
it
e
m
s
in
e
a
c
h
c
ondi
ti
on,
f
r
e
que
nc
y
is
th
e
popula
r
it
y
of
th
e
r
ul
e
ba
s
e
d
on
m
a
tc
hi
ng
d
a
ta
r
ow
s
,
a
nd
e
r
r
or
is
th
e
pr
opor
ti
on
of
r
ow
s
w
it
h
m
is
m
a
tc
he
d
r
e
s
pon
s
e
v
a
r
ia
bl
e
va
lu
e
s
.
M
a
th
e
m
a
ti
c
a
ll
y,
th
e
f
r
e
que
nc
y
a
nd
e
r
r
or
va
lu
e
s
a
r
e
e
xpr
e
s
s
e
d
a
s
f
ol
lo
w
s
[
12]
:
=
ℎ
ℎ
ℎ
ℎ
ℎ
ℎ
(
1)
=
ℎ
ℎ
ℎ
ℎ
ℎ
ℎ
(
2)
2.5.
A
s
s
oc
ia
t
io
n
r
u
le
F
r
om
[
25]
‒
[
27]
,
a
s
s
oc
ia
ti
on
r
ul
e
s
in
vol
ve
unc
ove
r
in
g
f
r
e
que
nt
a
s
s
oc
ia
ti
on
s
or
r
e
la
ti
ons
hi
ps
be
twe
e
n
it
e
m
s
or
va
r
ia
bl
e
s
in
a
da
ta
s
e
t.
A
s
s
oc
ia
ti
on
r
ul
e
s
ta
ke
th
e
f
or
m
“
I
f
a
nt
e
c
e
de
nt
,
th
e
n c
ons
e
que
nt
”
,
a
lo
ng
w
it
h
a
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
E
nhanc
in
g i
nt
e
r
pr
e
ta
bi
li
ty
i
n r
andom
fo
r
e
s
t:
L
e
v
e
r
agi
ng i
nT
r
e
e
s
f
or
as
s
oc
ia
ti
on
…
(
F
at
m
a H
il
al
i
M
oh’
d
)
4057
m
e
a
s
ur
e
of
t
he
s
uppor
t
a
nd c
onf
id
e
nc
e
a
s
s
o
c
ia
te
d w
it
h t
he
r
ul
e
.
S
uppor
t
r
e
pr
e
s
e
nt
s
t
he
pe
r
c
e
nt
a
ge
of
da
ta
ba
s
e
c
ondi
ti
ons
th
a
t
s
a
ti
s
f
y
th
e
gi
ve
n
r
ul
e
.
C
onf
id
e
nc
e
m
e
a
s
ur
e
s
th
e
l
e
ve
l
of
c
e
r
ta
in
ty
in
th
e
id
e
nt
if
ie
d
a
s
s
oc
ia
ti
on.
M
a
th
e
m
a
ti
c
a
ll
y, t
he
s
uppor
t
a
nd c
onf
id
e
nc
e
f
or
m
ul
a
a
r
e
e
xpr
e
s
s
e
d a
s
f
ol
lo
w
s
[
27]
:
(
⇒
)
=
ℎ
(
3)
(
⇒
)
=
ℎ
(
4)
w
he
r
e
r
e
pr
e
s
e
nt
c
ondi
ti
on a
nd pr
e
di
c
ti
on i
n t
he
r
ul
e
, r
e
s
pe
c
ti
ve
ly
.
2.6.
D
at
a
an
al
ys
is
p
r
oc
e
d
u
r
e
T
he
da
ta
a
na
ly
s
is
pr
oc
e
dur
e
us
e
d
in
th
is
s
tu
dy
is
de
m
ons
tr
a
te
d
in
F
ig
ur
e
2
.
A
s
s
how
n
in
F
ig
ur
e
2,
da
ta
w
a
s
im
por
te
d,
a
nd
th
e
n
d
a
ta
e
xpl
or
a
ti
on
w
a
s
don
e
.
A
f
te
r
da
ta
w
a
s
s
pl
it
in
to
80%
f
or
tr
a
in
in
g
a
nd
20%
f
or
te
s
ti
ng
da
ta
,
th
e
tr
a
in
in
g
da
ta
w
a
s
ba
la
nc
e
d
us
in
g
th
e
s
ynt
he
ti
c
m
in
or
it
y
ove
r
-
s
a
m
pl
in
g
te
c
hni
que
(
S
M
O
T
E
)
;
th
e
n
a
r
a
ndom
f
or
e
s
t
m
ode
l
w
a
s
f
it
te
d.
A
f
te
r
th
e
m
ode
l
w
a
s
f
it
te
d,
r
ul
e
s
w
e
r
e
e
xt
r
a
c
te
d
f
r
om
r
a
ndom
f
or
e
s
ts
,
a
nd
th
e
m
os
t
f
r
e
que
nt
r
ul
e
s
w
e
r
e
us
e
d
to
i
de
nt
if
y
f
r
e
que
nt
va
r
ia
bl
e
s
.
T
he
c
ons
tr
uc
ti
on
r
a
ndom
f
or
e
s
t
m
ode
l
w
a
s
f
a
c
il
it
a
te
d
us
in
g
th
e
r
e
gul
a
r
iz
e
d
r
a
ndom
f
or
e
s
t
(
RRF
)
pa
c
ka
ge
,
a
nd
th
e
in
T
r
e
e
s
pa
c
ka
ge
[
28]
w
a
s
e
m
pl
oye
d t
o
e
xt
r
a
c
t
th
e
r
ul
e
in
t
he
tr
e
e
s
a
nd
c
om
put
e
th
e
m
e
a
s
ur
e
of
t
he
r
ul
e
i
nt
e
r
e
s
t.
F
ig
ur
e
2
.
F
lo
w
c
ha
r
t
of
t
he
pr
oc
e
dur
e
of
da
ta
a
na
ly
s
is
3.
R
E
S
U
L
T
S
A
N
D
D
I
S
C
U
S
S
I
O
N
3.1.
D
at
a
e
xp
lo
r
at
io
n
T
he
r
e
a
r
e
no
in
s
ta
nc
e
s
of
m
is
s
in
g
da
ta
in
a
ny
of
th
e
pr
e
di
c
to
r
v
a
r
ia
bl
e
s
or
th
e
out
c
om
e
v
a
r
ia
bl
e
. T
he
a
na
ly
s
is
of
th
e
out
c
om
e
va
r
ia
bl
e
‘
une
m
pl
oym
e
nt
s
ta
tu
s
’
s
how
e
d
a
di
s
ti
nc
t
c
l
a
s
s
di
s
tr
ib
ut
io
n,
w
he
r
e
‘
e
m
pl
oye
d’
a
c
c
ount
s
f
or
81.5
%
a
nd
‘
une
m
pl
oye
d’
f
or
18.5
%
of
th
e
da
ta
s
e
t,
a
s
s
how
n
in
T
a
bl
e
2.
T
he
s
e
obs
e
r
va
ti
ons
i
ndi
c
a
te
a
s
ig
ni
f
ic
a
nt
da
ta
i
m
ba
la
nc
e
.
T
a
bl
e
2
.
C
la
s
s
di
s
tr
ib
ut
io
n of
a
n une
m
pl
oym
e
nt
s
ta
tu
s
U
ne
m
pl
oym
e
nt
s
t
a
t
us
P
r
opor
t
i
on
E
m
pl
oye
d
0.815
U
ne
m
pl
oye
d
0.185
U
nba
la
nc
e
d
da
ta
c
a
n
ha
r
m
pr
e
di
c
ti
on
a
c
c
ur
a
c
y
be
c
a
us
e
m
o
de
ls
a
r
e
ty
pi
c
a
ll
y
bi
a
s
e
d
to
w
a
r
d
th
e
dom
in
a
nt
c
la
s
s
i
n t
he
da
ta
s
e
t.
T
o m
it
ig
a
te
t
hi
s
i
s
s
ue
, w
e
e
m
pl
oy
e
d a
t
e
c
hni
que
c
a
ll
e
d S
M
O
T
E
i
n t
r
a
in
in
g da
ta
to
m
a
ke
da
ta
ba
la
n
c
e
.
S
M
O
T
E
is
de
s
ig
ne
d
to
a
ddr
e
s
s
im
ba
la
n
c
e
d
da
ta
by
c
r
e
a
ti
ng
a
r
ti
f
ic
ia
l
s
a
m
pl
e
s
f
or
th
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
13
, N
o.
4
,
D
e
c
e
m
be
r
2024
:
4054
-
4061
4058
m
in
or
it
y
c
la
s
s
,
th
us
e
nh
a
nc
in
g
th
e
m
ode
l’
s
c
a
pa
c
it
y
to
pr
e
di
c
t
bot
h
c
la
s
s
e
s
w
it
hi
n
our
‘
une
m
pl
oym
e
nt
s
ta
tu
s
’
va
r
ia
bl
e
m
or
e
a
c
c
ur
a
te
ly
[
27]
‒[
30]
.
3.2.
R
an
d
om
f
or
e
s
t
m
od
e
l
T
he
r
a
ndom
f
or
e
s
t
m
ode
l
w
a
s
bui
lt
us
in
g
th
e
opt
im
a
l
p
a
r
a
m
e
te
r
f
r
om
th
e
gr
id
s
e
a
r
c
h
a
ppr
oa
c
h.
T
h
e
m
ode
l
w
a
s
tr
a
in
e
d
w
it
h
100
de
c
is
io
n
tr
e
e
s
a
nd
4
pr
e
di
c
to
r
s
,
w
it
h
10
m
in
im
um
s
iz
e
s
of
te
r
m
in
a
l
node
s
a
nd
200
m
a
xi
m
um
num
be
r
s
of
te
r
m
in
a
l
node
s
.
S
ubs
e
que
nt
ly
,
a
m
ode
l
w
a
s
e
va
lu
a
te
d
u
s
in
g
te
s
t
da
ta
,
pr
e
s
e
nt
in
g
th
e
pe
r
f
or
m
a
nc
e
out
c
om
e
s
i
n T
a
bl
e
3.
T
a
bl
e
3
.
M
od
e
l
pe
r
f
or
m
a
nc
e
on t
e
s
t
da
ta
P
e
r
f
or
m
a
nc
e
m
e
t
r
i
c
s
V
a
l
ue
(%)
A
c
c
ur
a
c
y
66.51
S
e
ns
i
t
i
vi
t
y
74.72
S
pe
c
i
f
i
c
i
t
y
64.65
AUC
69.68
T
he
m
ode
l
de
s
ig
ne
d
f
or
pr
e
di
c
ti
ng
une
m
pl
oym
e
nt
s
ta
tu
s
a
c
hi
e
ve
d
a
n
ove
r
a
ll
a
c
c
ur
a
c
y
of
66.51%
.
W
it
h
a
s
e
ns
it
iv
it
y
of
74.72%
,
it
de
m
ons
tr
a
te
s
th
e
a
bi
li
ty
to
pr
e
d
ic
t
th
e
une
m
pl
oye
d
c
la
s
s
,
w
it
h
a
s
pe
c
if
ic
it
y
of
64.65%
,
hi
ghl
ig
ht
in
g
it
s
a
c
c
ur
a
c
y
in
pr
e
di
c
ti
ng
th
e
e
m
pl
oy
e
d
c
la
s
s
.
T
he
A
U
C
is
69.68%
,
in
di
c
a
ti
ng
a
m
ode
r
a
te
l
e
ve
l
of
a
c
c
ur
a
c
y i
n di
s
ti
ngui
s
hi
ng be
tw
e
e
n une
m
pl
oy
e
d a
nd e
m
pl
oye
d i
ndi
vi
dua
ls
.
3.3.
R
u
le
e
xt
r
ac
t
io
n
an
d
m
e
as
u
r
e
m
e
n
t
T
hi
s
r
e
s
e
a
r
c
h
obt
a
in
e
d
544
uni
que
r
ul
e
s
f
r
om
5
,
531
to
ta
l
r
ul
e
s
e
xt
r
a
c
te
d
f
r
om
th
e
f
ir
s
t
100
tr
e
e
s
.
A
m
ong
544
uni
que
r
ul
e
s
,
325
r
ul
e
s
pr
e
di
c
t
e
m
pl
oye
d
w
h
il
e
219
r
ul
e
s
pr
e
di
c
t
un
e
m
pl
oye
d.
T
a
bl
e
4
de
m
ons
tr
a
te
s
uni
que
r
ul
e
s
b
a
s
e
d on le
ngt
h
e
xt
r
a
c
te
d f
r
om
a
r
a
ndom f
or
e
s
t.
T
hi
s
s
tu
dy
f
oc
u
s
e
d
on
va
r
ia
bl
e
in
te
r
a
c
ti
ons
w
it
h
a
m
in
im
um
l
e
ngt
h
of
2
(
le
ngt
h=
2)
,
f
ol
lo
w
in
g
th
e
a
ppr
oa
c
h
us
e
d
in
pove
r
ty
s
ta
tu
s
pr
e
di
c
ti
on
li
te
r
a
tu
r
e
[
12]
,
hi
ghl
ig
ht
in
g
th
e
s
ig
ni
f
ic
a
nc
e
of
c
ons
id
e
r
in
g
in
te
r
a
c
ti
ons
be
twe
e
n
two
va
r
ia
bl
e
s
to
und
e
r
s
ta
nd
be
tt
e
r
th
e
d
yna
m
ic
s
in
f
lu
e
nc
in
g
th
e
pr
e
di
c
te
d
out
c
om
e
s
.
R
e
f
e
r
r
in
g t
o T
a
bl
e
4, t
he
r
e
a
r
e
203 unique r
ul
e
s
de
r
iv
e
d f
r
om
t
h
e
r
ul
e
s
w
it
h
2
≤
l
e
n
g
t
h
≤
4
th
a
t
c
ont
r
ib
ut
e
t
o
pr
e
di
c
ti
ng
une
m
pl
oym
e
nt
a
m
ong
in
di
vi
dua
ls
.
T
a
bl
e
5
s
how
s
th
e
s
ta
ti
s
ti
c
a
l
de
s
c
r
ip
ti
on
of
f
our
r
ul
e
m
e
a
s
ur
e
s
ge
ne
r
a
te
d
by
th
e
in
T
r
e
e
s
f
r
a
m
e
w
or
k:
f
r
e
que
nc
y,
e
r
r
or
,
s
uppor
t,
a
nd
c
onf
id
e
nc
e
of
203
uni
que
r
ul
e
s
f
or
m
e
d
in
pr
e
di
c
ti
ng une
m
pl
oym
e
nt
.
T
a
bl
e
5
di
s
pl
a
ys
th
e
s
ta
ti
s
ti
c
a
l
s
um
m
a
r
y
of
th
e
r
ul
e
’
s
m
e
a
s
ur
e
m
e
nt
in
vol
vi
ng
f
r
e
que
nc
y
a
nd
e
r
r
o
r
obt
a
in
e
d
f
r
om
th
e
da
ta
s
e
t.
A
ddi
ti
ona
ll
y,
s
uppor
t
a
nd
c
onf
id
e
nc
e
a
r
e
u
s
e
d
f
or
th
e
a
s
s
oc
ia
ti
on
r
ul
e
to
de
t
e
r
m
in
e
f
r
e
que
nt
va
r
ia
bl
e
s
f
or
pr
e
di
c
ti
ng
a
n
out
c
om
e
va
r
ia
bl
e
(
une
m
p
lo
ym
e
nt
s
ta
tu
s
)
.
T
he
f
r
e
que
nc
y
v
a
lu
e
s
r
a
nge
be
twe
e
n
0.2%
a
nd
89.6%
,
a
ve
r
a
gi
ng
25.3%
.
H
ow
e
ve
r
,
e
r
r
or
va
lu
e
s
va
r
y
f
r
om
21.3%
to
50.0%
,
a
ve
r
a
gi
ng
38.4%
a
c
r
os
s
th
e
s
a
m
e
da
ta
s
e
t
r
ow
s
.
F
or
r
ul
e
s
pr
e
di
c
ti
ng
une
m
pl
oym
e
nt
s
ta
tu
s
,
th
e
s
uppor
t
va
lu
e
s
s
pa
n
f
r
om
0.010
to
0.101,
a
ve
r
a
gi
ng
a
t
0.024
a
nd
w
it
h
a
m
e
di
a
n
of
0.017
.
C
onf
id
e
nc
e
va
lu
e
s
r
a
nge
be
tw
e
e
n
0.500
a
nd
1.00, with an a
ve
r
a
ge
of
0.707 a
nd a
m
e
di
a
n of
0.695.
T
a
bl
e
4
.
U
ni
que
r
ul
e
c
ount
s
by l
e
ngt
h f
or
r
a
ndom f
or
e
s
t
m
ode
l
pr
e
di
c
ti
ons
L
e
ngt
h
E
m
pl
oye
d
U
ne
m
pl
oye
d
T
ot
a
l
uni
que
r
ul
e
s
1
24
16
40
2
130
82
212
3
129
96
225
4
39
25
64
5
3
0
3
T
ot
a
l
325
219
544
T
a
bl
e
5
.
D
e
s
c
r
ip
ti
ve
s
ta
ti
s
ti
c
s
of
r
ul
e
m
e
a
s
ur
e
s
f
or
pr
e
di
c
ti
ng u
ne
m
pl
oym
e
nt
(
R
ul
e
s
w
it
h 2≤L
e
ngt
h≤4)
M
e
a
s
ur
e
s
F
r
e
que
nc
y
E
r
r
or
S
uppor
t
C
onf
i
de
nc
e
M
i
n
0.002
0.213
0.010
0.500
1s
t
Q
u
0.132
0.321
0.012
0.610
M
e
di
a
n
0.209
0.389
0.017
0.695
M
e
a
n
0.253
0.384
0.024
0.707
3r
d Q
u
0.344
0.458
0.030
0.792
M
a
x
0.896
0.500
0.101
1.000
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
E
nhanc
in
g i
nt
e
r
pr
e
ta
bi
li
ty
i
n r
andom
fo
r
e
s
t:
L
e
v
e
r
agi
ng i
nT
r
e
e
s
f
or
as
s
oc
ia
ti
on
…
(
F
at
m
a H
il
al
i
M
oh’
d
)
4059
3.4.
T
h
e
m
os
t
var
ia
b
le
f
r
e
q
u
e
n
t
in
t
e
r
ac
t
io
n
s
T
he
a
na
ly
s
is
f
oc
u
s
e
d
on
f
in
di
ng
f
r
e
que
nt
va
r
ia
bl
e
in
te
r
a
c
ti
ons
th
a
t
s
ig
ni
f
ic
a
nt
ly
c
ont
r
ib
ut
e
to
pr
e
di
c
ti
ng
une
m
pl
oym
e
nt
.
T
a
bl
e
6
di
s
pl
a
ys
th
e
to
p
te
n
f
r
e
que
nt
va
r
ia
bl
e
in
te
r
a
c
ti
ons
,
out
li
ni
ng
th
e
ir
r
e
s
pe
c
ti
ve
m
e
tr
ic
s
w
it
hi
n
th
e
in
T
r
e
e
s
f
r
a
m
e
w
or
k.
L
e
ngt
h
(
L
e
n)
s
how
s
th
e
num
be
r
of
it
e
m
s
or
va
r
ia
bl
e
s
in
e
a
c
h
c
ondi
ti
on
or
r
ul
e
.
F
r
e
que
nc
y
(
F
r
e
q)
a
nd
e
r
r
or
(
E
r
r
)
a
r
e
m
e
a
s
ur
e
m
e
nt
s
th
a
t
d
e
s
c
r
ib
e
th
e
r
ul
e
ba
s
e
d
on
th
e
da
ta
s
e
t.
S
uppor
t
a
nd c
onf
id
e
nc
e
a
r
e
m
e
a
s
ur
e
m
e
nt
s
t
h
a
t
de
s
c
r
ib
e
t
he
r
ul
e
ba
s
e
d on pr
e
di
c
ti
on.
T
a
bl
e
6
.
T
op
te
n i
nt
e
r
a
c
ti
ons
of
va
r
ia
bl
e
s
pr
e
di
c
ti
ng une
m
pl
oy
m
e
nt
s
or
te
d by highes
t
s
uppor
t,
w
it
h
c
onf
id
e
nc
e
≥ 88.0%
L
e
n
C
ondi
t
i
on
F
r
e
q
E
r
r
L
a
be
l
s
uppor
t
c
onf
i
de
pr
e
d
4
S
e
x=2 & a
ge
=1 & i
ns
u=2 & ur
br
ur
=2
0.195
0.223
1
0.029
0.891
1
2
E
dul
e
v=(
1,2,3)
&
ur
b
r
ur
=2
0.475
0.408
1
0.018
0.882
1
2
S
e
x=2 & e
dul
e
v=2
0.124
0.415
1
0.018
0.882
1
4
S
e
x=2 & a
ge
=1& e
dul
e
v=3& ur
br
ur
=1
0.13
0
0.264
1
0.017
0.933
1
3
S
e
x=2 & e
dul
e
v=3 & t
r
a
i
ni
=1
0.257
0.278
1
0.016
0.964
1
4
S
e
x=2 & a
ge
=1 & e
dul
e
v = (
2,3)
&
ur
br
ur
=1
0.164
0.283
1
0.016
0.9
00
1
4
S
e
x=2 & a
ge
=1& e
dul
e
v=3& t
r
a
i
ni
=1
0.194
0.213
1
0.015
0.962
1
3
S
e
x=2 & i
ns
u=2 & e
dul
e
v=3
0.366
0.282
1
0.015
1
.000
1
3
S
e
x=2 & a
ge
=1& e
dul
e
v=2
0.06
0
0.321
1
0.015
0.929
1
4
S
e
x=2 & a
ge
=1& i
ns
u=2 & e
dul
e
v=3
0.278
0.227
1
0.014
1
.000
1
B
a
s
e
d
on
T
a
bl
e
6,
th
e
f
r
e
q
ue
nc
y
a
n
d
e
r
r
or
of
th
e
f
i
r
s
t
c
on
di
t
io
n (
S
e
x
=
2
&
a
ge
=
1
&
in
s
u
=
2 &
ur
br
ur
=
2)
a
r
e
0.
19
5
a
nd
0.
22
3,
r
e
s
p
e
c
ti
ve
ly
.
T
hi
s
in
d
ic
a
t
e
s
t
ha
t
t
he
c
on
di
ti
on’
s
p
op
ul
a
r
it
y
i
n
t
he
d
a
ta
s
e
t
i
s
1
9.
5%
,
but
th
e
e
r
r
or
i
s
22
.3%
,
i
m
pl
yi
n
g
t
h
a
t
th
e
c
o
ndi
ti
o
n
di
d
no
t
a
li
gn
w
i
th
th
e
da
ta
s
e
t
’
s
ou
tc
om
e
c
l
a
s
s
v
a
r
i
a
bl
e
(
un
e
m
p
lo
y
e
d
=
1)
.
C
onv
e
r
s
e
ly
,
t
he
s
up
por
t
a
nd
c
onf
i
d
e
n
c
e
a
r
e
0
.0
29
a
n
d
0.8
91
,
r
e
s
p
e
c
ti
v
e
l
y,
pr
e
di
c
ti
ng
une
m
pl
oy
m
e
nt
.
T
hi
s
m
e
a
ns
th
a
t
a
2.
9%
pr
op
or
ti
on
of
r
ul
e
s
is
f
or
m
e
d,
s
a
ti
s
f
yi
n
g
th
e
r
u
le
f
or
pr
e
d
ic
ti
n
g
une
m
pl
oy
m
e
nt
w
i
th
a
c
o
nf
id
e
n
c
e
of
89
.1%
.
I
n
ot
he
r
w
or
d
s
,
w
h
e
n
th
i
s
c
o
ndi
ti
o
n
o
c
c
ur
s
,
th
e
r
e
i
s
a
n
8
9.
1%
pr
ob
a
b
il
it
y
of
pr
e
di
c
t
in
g
un
e
m
p
lo
y
m
e
nt
.
T
a
bl
e
7
s
ho
w
s
th
e
c
ha
r
a
c
t
e
r
i
s
ti
c
s
of
u
ne
m
pl
oy
e
d
in
di
v
id
ua
l
s
i
n
Z
a
n
z
ib
a
r
b
a
s
e
d
o
n
a
c
on
f
id
e
n
c
e
v
a
l
ue
of
88
.0%
a
bo
ve
a
nd
s
or
t
e
d
by
m
o
s
t
e
x
c
e
l
le
nt
s
upp
or
t
a
nd
l
o
w
e
s
t
e
r
r
or
.
I
n
T
a
bl
e
7,
th
e
c
onf
id
e
nc
e
va
lu
e
is
th
e
pr
oba
bi
li
ty
of
a
n
in
di
vi
dua
l
be
in
g
c
la
s
s
if
ie
d
a
s
un
e
m
pl
oye
d
ba
s
e
d
on
th
e
in
te
r
a
c
ti
on
va
r
ia
bl
e
in
a
r
ul
e
f
or
m
e
d.
A
m
ong
10
r
ul
e
s
f
or
m
e
d,
th
e
f
our
dom
in
a
nt
va
r
ia
bl
e
s
a
r
e
be
in
g
f
e
m
a
le
,
a
ge
gr
oup
yout
h,
h
a
vi
ng
s
e
c
onda
r
y
e
duc
a
ti
on,
a
nd
ha
vi
ng
pr
im
a
r
y
e
duc
a
ti
on.
T
he
s
e
r
ul
e
s
r
e
ve
a
l
ke
y c
ha
r
a
c
te
r
is
ti
c
s
t
ha
t
s
tr
ongl
y i
ndi
c
a
te
une
m
pl
oym
e
nt
a
m
ong indi
vi
dua
ls
.
−
F
e
m
a
le
s
w
it
h
s
e
c
onda
r
y
e
du
c
a
ti
on
a
nd
l
a
c
ki
ng
he
a
lt
h
in
s
ur
a
nc
e
s
ho
w
c
a
s
e
a
100.0%
pr
oba
bi
li
ty
of
une
m
pl
oym
e
nt
.
−
Y
out
hf
ul
f
e
m
a
le
s
la
c
ki
ng
he
a
lt
h
in
s
ur
a
nc
e
a
nd
s
e
c
ond
a
r
y
e
duc
a
ti
on
di
s
pl
a
y
a
100%
pr
oba
bi
li
ty
of
une
m
pl
oym
e
nt
.
−
F
e
m
a
le
s
w
it
h s
e
c
onda
r
y e
duc
a
ti
on a
nd no tr
a
in
in
g a
tt
e
nda
n
c
e
h
a
ve
a
96.4%
l
ik
e
li
hood of
une
m
pl
oym
e
nt
.
−
Y
out
hf
ul
f
e
m
a
le
s
w
it
h
s
e
c
onda
r
y
e
duc
a
ti
on
a
nd
no
tr
a
in
in
g
a
tt
e
nda
nc
e
di
s
pl
a
y
a
96.2%
pr
oba
bi
li
ty
of
une
m
pl
oym
e
nt
.
−
Y
out
hf
ul
f
e
m
a
le
s
w
it
h
s
e
c
ond
a
r
y
e
duc
a
ti
on
a
nd
r
e
s
id
in
g
in
r
ur
a
l
a
r
e
a
s
di
s
pl
a
y
a
93.3%
pr
oba
bi
li
ty
of
une
m
pl
oym
e
nt
.
−
Y
out
hf
ul
f
e
m
a
le
s
w
it
h pr
im
a
r
y e
duc
a
ti
on ha
ve
a
92.9%
pr
oba
bi
li
ty
of
une
m
pl
oym
e
nt
.
−
Y
out
hf
ul
f
e
m
a
le
s
w
it
h
pr
im
a
r
y
o
r
s
e
c
onda
r
y
e
duc
a
ti
on
a
nd
r
e
s
id
in
g
in
r
ur
a
l
a
r
e
a
s
ha
ve
a
90.0%
pr
oba
bi
li
ty
of
une
m
pl
oym
e
nt
.
T
a
bl
e
7
.
C
h
a
r
a
c
te
r
is
ti
c
s
of
une
m
pl
oye
d i
ndi
vi
dua
ls
b
a
s
e
d on
c
o
nf
id
e
nc
e
va
lu
e
s
≥ 88
.0%
R
ul
e
s
e
x
a
ge
H
e
a
l
t
h i
ns
ur
a
nc
e
E
duc
a
t
i
on l
e
ve
l
A
t
t
e
nde
d t
r
a
i
ni
ng
R
e
s
i
de
nc
e
C
onf
i
de
nc
e
(%)
1
f
e
m
a
l
e
-
no
s
e
c
onda
r
y
-
-
100.0
2
f
e
m
a
l
e
yout
h
no
s
e
c
onda
r
y
-
-
100.0
3
f
e
m
a
l
e
-
-
s
e
c
onda
r
y
no
-
96.4
4
f
e
m
a
l
e
yout
h
-
s
e
c
onda
r
y
no
-
96.2
5
f
e
m
a
l
e
yout
h
-
s
e
c
onda
r
y
-
r
ur
a
l
93.3
6
f
e
m
a
l
e
yout
h
-
pr
i
m
a
r
y
-
-
92.9
7
f
e
m
a
l
e
yout
h
-
P
r
i
m
a
r
y/
s
e
c
onda
r
y
-
r
ur
a
l
90.0
8
f
e
m
a
l
e
yout
h
no
-
-
ur
ba
n
89.1
9
-
-
-
N
e
ve
r
a
t
t
e
nde
d/
pr
i
m
a
r
y/
s
e
c
onda
r
y
-
ur
ba
n
88.2
10
f
e
m
a
l
e
-
-
pr
i
m
a
r
y
-
-
88.2
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
13
, N
o.
4
,
D
e
c
e
m
be
r
2024
:
4054
-
4061
4060
4.
C
O
N
C
L
U
S
I
O
N
T
hi
s
s
tu
dy
s
uc
c
e
s
s
f
ul
ly
a
ppl
ie
d
th
e
in
te
r
pr
e
ta
bl
e
r
a
ndom
f
or
e
s
t
m
ode
l
to
pr
e
di
c
t
une
m
pl
oym
e
nt
a
t
th
e
in
di
vi
dua
l
le
ve
ls
in
Z
a
nz
ib
a
r
.
T
he
in
te
r
pr
e
ta
bl
e
m
ode
l
ge
ne
r
a
te
d
us
in
g
th
e
in
T
r
e
e
s
f
r
a
m
e
w
or
k
r
e
ve
a
le
d
in
s
ig
ht
f
ul
pa
tt
e
r
ns
a
nd
f
r
e
que
nt
va
r
ia
bl
e
in
te
r
a
c
ti
ons
th
a
t
c
ont
r
ib
ut
e
to
une
m
pl
oym
e
nt
.
T
he
ke
y
f
in
di
ngs
th
a
t
c
ont
r
ib
ut
e
to
une
m
pl
oym
e
nt
a
r
e
f
e
m
a
le
s
e
x
a
nd
ha
vi
ng
no
he
a
lt
h
in
s
ur
a
nc
e
a
nd
he
r
e
duc
a
ti
on
le
ve
l
is
s
e
c
onda
r
y,
f
e
m
a
le
a
nd
he
r
a
ge
gr
oup
is
yout
h
a
nd
ha
vi
ng
no
he
a
lt
h
in
s
ur
a
nc
e
a
nd
he
r
e
duc
a
ti
on
le
ve
l
is
s
e
c
onda
r
y,
bot
h
w
it
h
c
onf
id
e
nc
e
of
100%
.
T
he
s
e
f
in
di
ngs
pr
e
s
e
nt
th
e
oppor
tu
ni
ty
to
ta
il
or
in
te
r
ve
nt
io
ns
a
nd
f
or
m
ul
a
te
ta
r
ge
te
d
pol
ic
ie
s
th
a
t
a
ddr
e
s
s
th
e
s
e
va
r
ia
bl
e
s
,
pot
e
nt
ia
ll
y
le
a
di
ng
to
m
or
e
e
f
f
e
c
ti
ve
e
m
pl
oym
e
nt
s
tr
a
te
gi
e
s
.
T
he
s
e
r
e
s
ul
ts
pr
ovi
de
a
be
tt
e
r
unde
r
s
ta
ndi
ng of
t
he
u
ne
m
pl
oym
e
nt
dyna
m
ic
i
n Z
a
nz
ib
a
r
.
5.
S
U
G
G
E
S
T
I
O
N
I
n
th
e
f
ut
ur
e
,
th
is
r
e
s
e
a
r
c
h
c
a
n
be
e
xt
e
nde
d
to
de
ve
lo
p
th
e
i
nt
e
r
pr
e
ta
bi
li
ty
of
th
e
double
r
a
ndom
f
or
e
s
t.
T
he
double
r
a
ndom
f
o
r
e
s
t
c
a
n
s
uc
c
e
s
s
f
ul
ly
m
ode
l
th
e
unde
r
f
it
ti
ng
da
ta
a
nd
ha
s
be
e
n
pr
e
li
m
in
a
r
y
in
ve
s
ti
ga
te
d by
pr
e
vi
ous
s
tu
di
e
s
,
but
ne
e
d
s
im
pr
ove
m
e
nt
.
A
C
K
N
O
WL
E
D
G
E
M
E
N
T
S
T
he
a
ut
hor
w
oul
d
li
ke
to
e
xt
e
nd
he
a
r
t
f
e
lt
gr
a
ti
tu
de
to
th
os
e
w
ho
ha
ve
s
uppor
te
d
th
is
r
e
s
e
a
r
c
h.
S
pe
c
ia
l
th
a
nks
to
th
e
D
e
p
a
r
tm
e
nt
of
S
ta
ti
s
ti
c
s
,
F
a
c
ul
ty
of
M
a
th
e
m
a
ti
c
s
a
nd
N
a
tu
r
a
l
S
c
ie
nc
e
s
,
I
P
B
U
ni
v
e
r
s
it
y,
a
nd
M
in
is
tr
y
of
E
duc
a
ti
on
of
I
ndone
s
ia
f
or
th
e
ir
in
va
lu
a
bl
e
s
p
ons
or
s
hi
p
of
m
y
s
tu
di
e
s
in
I
ndone
s
ia
th
r
ough
th
e
P
a
r
tn
e
r
s
hi
p
f
or
D
e
ve
lo
pi
ng
C
ount
r
y
(
K
N
B
)
s
c
hol
a
r
s
hi
p
pr
ogr
a
m
,
of
f
e
r
e
d
by
th
e
I
ndone
s
ia
n
gove
r
nm
e
nt
,
a
s
w
e
ll
a
s
O
f
f
ic
e
of
C
hi
e
f
G
ove
r
nm
e
nt
S
ta
ti
s
ti
c
ia
n
(
O
C
G
S
)
of
Z
a
nz
ib
a
r
f
o
r
pr
ov
is
io
n
of
da
ta
s
e
t
w
e
r
e
in
s
tr
um
e
nt
a
l
in
t
hi
s
r
e
s
e
a
r
c
h.
R
E
F
E
R
E
N
C
E
S
[
1]
B
.
S
i
s
w
oyo,
Z
.
A
.
A
ba
s
,
A
.
N
.
C
.
P
e
e
,
R
.
K
om
a
l
a
s
a
r
i
,
a
nd
N
.
S
ur
ya
na
,
“
E
ns
e
m
bl
e
m
a
c
hi
ne
l
e
a
r
ni
ng
a
l
gor
i
t
hm
opt
i
m
i
z
a
t
i
on
o
f
ba
nkr
upt
c
y
pr
e
di
c
t
i
on
of
ba
nk,”
I
A
E
S
I
nt
e
r
nat
i
onal
J
our
nal
of
A
r
t
i
f
i
c
i
al
I
nt
e
l
l
i
ge
nc
e
(
I
J
-
A
I
)
,
vol
.
11,
no.
2,
pp.
679
–
686,
J
un.
2022, doi
:
10.11591/
i
j
a
i
.v11.i
2.pp679
-
686.
[
2]
A
.
B
.
G
um
e
l
a
r
,
A
.
Y
oga
t
a
m
a
,
D
.
P
.
A
di
,
F
.
F
r
i
s
m
a
nda
,
a
nd
I
.
S
ugi
a
r
t
o,
“
F
or
w
a
r
d
f
e
a
t
ur
e
s
e
l
e
c
t
i
on
f
or
t
oxi
c
s
p
e
e
c
h
c
l
a
s
s
i
f
i
c
a
t
i
on
us
i
ng
s
uppor
t
ve
c
t
or
m
a
c
hi
ne
a
nd
r
a
ndom
f
or
e
s
t
,”
I
A
E
S
I
nt
e
r
nat
i
onal
J
our
nal
of
A
r
t
i
f
i
c
i
al
I
nt
e
l
l
i
ge
nc
e
(
I
J
-
A
I
)
,
vol
.
11,
no.
2,
pp.
717
–
726, J
un. 2022, doi
:
10.11591/
i
j
a
i
.v11.i
2.pp717
-
726.
[
3]
L
. B
r
e
i
m
a
n, “
R
a
ndom
f
or
e
s
t
s
,”
M
ac
hi
ne
L
e
ar
ni
ng
, vol
. 45, no. 1, pp. 5
–
32, 200
1, doi
:
10.1023/
A
:
1010933404324.
[
4]
A
.
T
he
be
l
t
,
J
.
K
r
onqvi
s
t
,
M
.
M
i
s
t
r
y,
R
.
M
.
L
e
e
,
N
.
S
.
-
M
e
r
x,
a
nd
R
.
M
i
s
e
ne
r
,
“
E
N
T
M
O
O
T
:
A
f
r
a
m
e
w
or
k
f
o
r
opt
i
m
i
z
a
t
i
on
ove
r
e
ns
e
m
bl
e
t
r
e
e
m
ode
l
s
,”
C
om
put
e
r
s
and C
he
m
i
c
al
E
ngi
ne
e
r
i
ng
, vol
. 151, 2021,
doi
:
10.1016/
j
.c
om
pc
he
m
e
ng.2021.107343.
[
5]
G
. B
i
a
u a
nd E
. S
c
or
ne
t
, “
A
r
a
ndom
f
or
e
s
t
gui
de
d t
our
,”
T
e
s
t
, vol
. 25, no. 2, pp. 197
–
227, 2016, doi
:
10.1007/
s
11749
-
016
-
0481
-
7.
[
6]
F
.
F
a
r
ooq
e
t
al
.
,
“
A
c
om
pa
r
a
t
i
ve
s
t
udy
o
f
r
a
ndom
f
or
e
s
t
a
nd
ge
ne
t
i
c
e
ngi
ne
e
r
i
ng
pr
ogr
a
m
m
i
ng
f
or
t
he
pr
e
di
c
t
i
on
of
c
om
pr
e
s
s
i
ve
s
t
r
e
ngt
h of
hi
gh s
t
r
e
ngt
h c
onc
r
e
t
e
(
H
S
C
)
,”
A
ppl
i
e
d Sc
i
e
nc
e
s
, vol
. 10, no. 20, O
c
t
. 2020, doi
:
10.3390/
a
pp10207330.
[
7]
O
.
A
w
uj
ool
a
,
P
.
O
.
O
di
on,
M
.
E
.
I
r
he
bhude
,
a
nd
H
.
A
m
i
nu,
“
P
e
r
f
or
m
a
nc
e
e
v
a
l
ua
t
i
on
of
m
a
c
hi
ne
l
e
a
r
ni
ng
pr
e
di
c
t
i
ve
a
na
l
yt
i
c
a
l
m
ode
l
f
or
de
t
e
r
m
i
ni
ng
t
he
j
ob
a
ppl
i
c
a
nt
s
e
m
pl
oym
e
nt
s
t
a
t
u
s
’
,”
M
al
ay
s
i
an
J
ou
r
nal
of
A
ppl
i
e
d
Sc
i
e
nc
e
s
,
vol
.
6,
no.
1,
pp.
67
–
79
,
A
pr
. 2021, doi
:
10.37231/
m
yj
a
s
.2021.6.1.276.
[
8]
M
.
G
.
C
e
l
bi
ş
,
“
S
oc
i
a
l
ne
t
w
or
ks
,
f
e
m
a
l
e
une
m
pl
oym
e
nt
,
a
nd
t
he
u
r
ba
n
-
r
ur
a
l
di
vi
de
i
n
T
u
r
ke
y:
E
vi
de
nc
e
f
r
om
t
r
e
e
-
ba
s
e
d
m
a
c
hi
ne
l
e
a
r
ni
ng a
l
gor
i
t
hm
s
,”
Sos
y
oe
k
onom
i
, vol
. 29, no. 50, pp. 73
–
93, O
c
t
. 2021, doi
:
10.17233/
s
os
yoe
konom
i
.2021.04.04.
[
9]
C
.
B
é
na
r
d,
G
.
B
i
a
u,
S
.
D.
V
e
i
ga
,
a
nd
E
.
S
c
or
ne
t
,
“
I
nt
e
r
pr
e
t
a
bl
e
r
a
ndom
f
or
e
s
t
s
vi
a
r
ul
e
e
xt
r
a
c
t
i
o
n,”
A
r
x
i
v
-
St
at
i
s
t
i
c
s
,
vol
.
130,
pp.
937
–
945, A
pr
. 2020,
doi
:
10.48550/
a
r
X
i
v.2004.14841.
[
10]
A
.
J
.
S
a
ge
,
Y
.
L
i
u,
a
nd
J
.
S
a
t
o,
“
F
r
om
bl
a
c
k
box
t
o
s
hi
ni
ng
s
pot
l
i
ght
:
U
s
i
ng
r
a
ndom
f
or
e
s
t
pr
e
di
c
t
i
on
i
nt
e
r
va
l
s
t
o
i
l
l
um
i
na
t
e
t
he
i
m
pa
c
t
of
a
s
s
um
pt
i
ons
i
n
l
i
ne
a
r
r
e
gr
e
s
s
i
on,”
T
he
A
m
e
r
i
c
an
St
at
i
s
t
i
c
i
an
,
vol
.
76,
no.
4,
pp.
414
–
429,
O
c
t
.
2022,
doi
:
10.1080/
00031305.2022.2107568.
[
11]
S
.
W
e
nc
k,
M
.
C
r
e
ydt
,
J
.
H
a
ns
e
n,
F
.
G
ä
r
be
r
,
M
.
F
i
s
c
he
r
,
a
nd
S
.
S
e
i
f
e
r
t
,
“
O
pe
ni
ng
t
he
r
a
ndom
f
or
e
s
t
bl
a
c
k
box
of
t
he
m
e
t
a
bol
om
e
by t
he
a
ppl
i
c
a
t
i
on of
s
ur
r
oga
t
e
m
i
ni
m
a
l
de
pt
h,”
M
e
t
abol
i
t
e
s
, vol
. 12, no. 1, D
e
c
. 2021, doi
:
10.3390/
m
e
t
a
bo12010005.
[
12]
H
.
I
l
m
a
,
K
.
A
.
N
ot
odi
put
r
o,
a
nd
B
.
S
a
r
t
ono,
“
A
s
s
oc
i
a
t
i
on
r
ul
e
s
i
n
r
a
ndom
f
or
e
s
t
f
or
t
he
m
os
t
’
,”
B
A
R
E
K
E
N
G
:
J
ur
nal
I
l
m
u
M
at
e
m
at
i
k
a dan T
e
r
apan
, vol
. 17, no. 1, pp. 0185
–
0196, A
pr
. 2023, doi
:
10.30598/
ba
r
e
ke
ngvol
17i
s
s
1pp0185
-
0196.
[
13]
T
i
m
M
i
l
l
e
r
,
“
E
xpl
a
na
t
i
on
i
n
a
r
t
i
f
i
c
i
a
l
i
nt
e
l
l
i
ge
nc
e
:
I
ns
i
ght
s
f
r
om
t
he
s
oc
i
a
l
s
c
i
e
nc
e
s
,”
A
r
t
i
f
i
c
i
al
I
nt
e
l
l
i
ge
nc
e
,
vol
.
267,
pp.
1
–
38,
2019,
doi
:
10.1016/
j
.a
r
t
i
nt
.2018.07.007.
[
14]
A
.
V
e
l
l
i
do,
“
T
he
i
m
por
t
a
nc
e
of
i
nt
e
r
p
r
e
t
a
bi
l
i
t
y
a
nd
vi
s
ua
l
i
z
a
t
i
on
i
n
m
a
c
hi
ne
l
e
a
r
ni
ng
f
or
a
ppl
i
c
a
t
i
ons
i
n
m
e
di
c
i
ne
a
nd
he
a
l
t
h
c
a
r
e
,”
N
e
ur
al
C
om
put
i
ng and A
ppl
i
c
at
i
ons
, vol
. 32, no. 24, pp. 18069
–
18083, 2020, doi
:
10.1007/
s
00521
-
019
-
04051
-
w.
[
15]
G
.
J
a
m
e
s
,
D
.
W
i
t
t
e
n,
T
.
H
a
s
t
i
e
,
a
nd
R
.
T
i
bs
hi
r
a
ni
,
“
A
n
i
nt
r
oduc
t
i
on
t
o
s
t
a
t
i
s
t
i
c
a
l
l
e
a
r
n
i
ng,
”
i
n
Spr
i
nge
r
T
e
x
t
s
i
n
St
at
i
s
t
i
c
s
,
vol
.
103, no. 10
,
N
e
w
Y
or
k:
S
pr
i
nge
r
, 2013. doi
:
10.1007/
978
-
1
-
4614
-
7138
-
7.
[
16]
T
.
E
l
om
a
a
,
“
I
n
D
e
f
e
ns
e
of
C
4.5:
not
e
s
o
n
l
e
a
r
ni
ng
one
-
l
e
ve
l
de
c
i
s
i
on
t
r
e
e
s
,”
i
n
M
ac
hi
ne
L
e
ar
ni
ng
P
r
oc
e
e
di
ngs
1994
,
E
l
s
e
vi
e
r
,
1994, pp. 62
–
69. doi
:
10.1016/
B
978
-
1
-
55860
-
335
-
6.50016
-
7.
[
17]
M
.
F
okke
m
a
,
N
.
S
m
i
t
s
,
H
.
K
e
l
de
r
m
a
n,
a
nd
B
.
W
.
J
.
H
.
P
e
nni
nx,
“
C
onn
e
c
t
i
n
g
c
l
i
ni
c
a
l
a
nd
a
c
t
ua
r
i
a
l
pr
e
di
c
t
i
on
w
i
t
h
r
ul
e
-
ba
s
e
d
m
e
t
hods
,”
P
s
y
c
hol
ogi
c
al
A
s
s
e
s
s
m
e
nt
, vol
. 27, no. 2, pp. 636
–
644, 2015, doi
:
10.1037/
pa
s
0000072.
[
18]
O
.
S
a
gi
a
nd
L
.
R
oka
c
h,
“
A
ppr
oxi
m
a
t
i
ng
X
G
B
oo
s
t
w
i
t
h
a
n
i
nt
e
r
pr
e
t
a
bl
e
de
c
i
s
i
on
t
r
e
e
,”
I
nf
or
m
at
i
on
Sc
i
e
nc
e
s
,
vol
.
572,
pp.
522
–
542, S
e
p. 2021, doi
:
10.1016/
j
.i
ns
.2021.05.055.
[
19]
R
.
A
gr
a
w
a
l
,
T
.
I
m
i
e
l
i
ńs
ki
,
a
nd
A
.
S
w
a
m
i
,
“
M
i
ni
ng
a
s
s
oc
i
a
t
i
on
r
ul
e
s
be
t
w
e
e
n
s
e
t
s
of
i
t
e
m
s
i
n
l
a
r
ge
da
t
a
ba
s
e
s
,
”
A
C
M
SI
G
M
O
D
R
e
c
or
d
, vol
. 22, no. 2, pp. 207
–
216, J
un. 1993, doi
:
10.1145/
170036.170072.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
E
nhanc
in
g i
nt
e
r
pr
e
ta
bi
li
ty
i
n r
andom
fo
r
e
s
t:
L
e
v
e
r
agi
ng i
nT
r
e
e
s
f
or
as
s
oc
ia
ti
on
…
(
F
at
m
a H
il
al
i
M
oh’
d
)
4061
[
20]
H
.
D
e
ng,
“
I
nt
e
r
pr
e
t
i
ng
t
r
e
e
e
ns
e
m
bl
e
s
w
i
t
h
i
nT
r
e
e
s
,”
I
nt
e
r
nat
i
onal
J
ou
r
nal
of
D
at
a
Sc
i
e
nc
e
and
A
nal
y
t
i
c
s
,
vol
.
7,
no.
4,
pp.
277
–
287, J
un. 2019, doi
:
10.1007/
s
41060
-
018
-
0144
-
8.
[
21]
S
.
W
a
ng,
Y
.
W
a
ng,
D
.
W
a
ng,
Y
.
Y
i
n,
Y
.
W
a
ng,
a
nd
Y
.
J
i
n,
“
A
n
i
m
pr
ove
d
r
a
ndom
f
or
e
s
t
-
ba
s
e
d
r
ul
e
e
xt
r
a
c
t
i
on
m
e
t
hod
f
or
br
e
a
s
t
c
a
nc
e
r
di
a
gnos
i
s
,
”
A
ppl
i
e
d Sof
t
C
om
put
i
ng
, vol
. 86, J
a
n. 2020, doi
:
10.1016/
j
.a
s
oc
.2019.105941.
[
22]
F
.
B
i
a
nc
hi
,
G
.
B
i
a
nc
hi
,
a
nd
D
.
S
ong,
“
T
he
l
ong
-
t
e
r
m
i
m
pa
c
t
of
t
he
C
O
V
I
D
-
19
une
m
pl
oym
e
nt
s
hoc
k
on
l
i
f
e
e
xpe
c
t
a
nc
y
a
nd
m
or
t
a
l
i
t
y r
a
t
e
s
,”
J
our
nal
of
E
c
onom
i
c
D
y
nam
i
c
s
and
C
ont
r
ol
, vol
. 146, J
a
n. 2023, doi
:
10.1016/
j
.j
e
dc
.2022.104581.
[
23]
N
B
S
, “
I
nt
e
gr
a
t
e
d l
a
bour
f
or
c
e
s
ur
ve
y 2020
-
2021,”
T
anz
ani
a N
at
i
onal
B
ur
e
au of
St
at
i
s
t
i
c
s
, D
odom
a
, T
a
nz
a
ni
a
:
N
B
S
,
2021.
[
24]
H
.
V
.
T
.
M
a
i
,
T
.
A
.
N
guye
n,
H
.
B
.
L
y,
a
nd
V
.
Q
.
T
r
a
n,
“
P
r
e
di
c
t
i
on
c
om
pr
e
s
s
i
ve
s
t
r
e
ngt
h
of
c
onc
r
e
t
e
c
ont
a
i
ni
ng
G
G
B
F
S
us
i
ng
r
a
ndom
f
or
e
s
t
m
ode
l
,”
A
dv
anc
e
s
i
n C
i
v
i
l
E
ngi
ne
e
r
i
ng
, vol
. 2021, 2021, doi
:
10.1155/
2021/
6671448.
[
25]
N
. F
. I
dr
i
s
, M
. A
. I
s
m
a
i
l
, M
. S
. M
oh
a
m
a
d, S
. K
a
s
i
m
, Z
. Z
a
ka
r
i
a
, a
nd T
. S
ut
i
kno, “
B
r
e
a
s
t
c
a
nc
e
r
di
s
e
a
s
e
c
l
a
s
s
i
f
i
c
a
t
i
on u
s
i
ng f
uz
z
y
-
I
D
3 a
l
gor
i
t
hm
ba
s
e
d on a
s
s
o
c
i
a
t
i
on f
unc
t
i
on,”
I
A
E
S I
nt
e
r
nat
i
onal
J
our
nal
of
A
r
t
i
f
i
c
i
al
I
nt
e
l
l
i
ge
nc
e
(
I
J
-
A
I
)
, vol
. 11, no. 2, pp. 448
–
461, J
un. 2022, doi
:
10.11591/
i
j
a
i
.v11.i
2.pp448
-
461.
[
26]
T
. D
a
ni
e
l
, C
. L
a
r
os
e
, a
nd
D
. L
a
r
os
e
,
D
at
a m
i
ni
ng and pr
e
di
c
t
i
v
e
anal
y
t
i
c
s
, N
e
w
Y
or
k:
J
ohn W
i
l
e
y &
S
ons
, 2015.
[
27]
M
.
J
.
H
ua
ng,
H
.
S
.
S
ung,
T
.
J
.
H
s
i
e
h,
M
.
C
.
W
u,
a
nd
S
.
H
.
C
hung,
“
A
ppl
yi
ng
da
t
a
-
m
i
ni
ng
t
e
c
hni
que
s
f
or
di
s
c
ove
r
i
ng
a
s
s
oc
i
a
t
i
o
n
r
ul
e
s
,”
Sof
t
C
om
put
i
ng
, vol
. 24, no. 11, pp. 8069
–
8075, 2020, doi
:
10.1007/
s
00500
-
019
-
04163
-
4.
[
28]
H
.
D
e
ng,
X
.
G
ua
n,
a
nd
V
.
K
hot
i
l
ovi
c
h,
“
i
nT
r
e
e
s
:
i
nt
e
r
p
r
e
t
t
r
e
e
e
ns
e
m
bl
e
s
,”
C
R
A
N
R
P
r
oj
e
c
t
,
2014
,
doi
:
10.32614/
C
R
A
N
.pa
c
ka
g
e
.i
nT
r
e
e
s
.
[
29]
D
.
E
l
r
e
e
dy
a
nd
A
.
F
.
A
t
i
ya
,
“
A
c
om
pr
e
he
ns
i
ve
a
na
l
ys
i
s
of
s
ynt
he
t
i
c
m
i
nor
i
t
y
ove
r
s
a
m
pl
i
ng
t
e
c
hni
que
(
S
M
O
T
E
)
f
or
ha
ndl
i
ng
c
l
a
s
s
i
m
ba
l
a
nc
e
,”
I
nf
or
m
at
i
on Sc
i
e
nc
e
s
, vol
. 505, pp. 32
–
64, D
e
c
. 2019, doi
:
10
.1016/
j
.i
ns
.2019.07.070.
[
30]
G
. A
. P
r
a
di
pt
a
, R
.
W
a
r
doyo, A
.
M
us
dhol
i
f
a
h, I
. N
.
H
. S
a
nj
a
ya
,
a
nd
M
. I
s
m
a
i
l
,
“
S
M
O
T
E
f
or
ha
ndl
i
ng i
m
ba
l
a
nc
e
d da
t
a
pr
obl
e
m
:
a
r
e
vi
e
w
,”
i
n
2021
Si
x
t
h
I
nt
e
r
nat
i
onal
C
onf
e
r
e
n
c
e
on
I
nf
or
m
at
i
c
s
and
C
om
put
i
ng
(
I
C
I
C
)
,
I
E
E
E
,
N
ov.
2021,
pp.
1
–
8.
doi
:
10.1109/
I
C
I
C
54025.2021.9632912.
B
I
O
G
R
A
P
H
I
E
S
O
F
A
U
T
H
O
R
S
Fatma
Hilali
Moh’d
is
a
statistician
at
the
Office
of
Chief
Government
of
Statistic
ians
in
Zanzibar
-
Tanzania.
She
completed
her
Bachelor
of
Science
in
Statistics
in
2016
from
the
University
of
Dodoma,
Tanzania.
She
is
pursuing
a
Master
of
Science
in
Statistic
s
and
Data
Scienc
e
at
IPB
Univer
sity,
Bogor,
Indone
sia.
Si
nce
2018,
she
has
been
actively
involv
ed
in
various
roles
at
O
CGS
Statis
tics
Zanzibar,
contri
buting
as
a
staff
member
across
multiple
national
surveys
and
censuses.
Her
contributions
enc
ompass
a
range
of
areas,
including
employment
and
earnings
surveys,
household
budget
surve
ys,
integr
ated
labor
force
surveys,
population
censuses,
and
industrial
production
censuses.
S
he
can
be
contacted
at
email:
fatmahmohd@
apps.ipb.ac.i
d
.
Khairil
Anwar
Notodiputro
e
arned
B
.
S
.
and
Master’s
degrees
from
IPB
University
and
Ph.D
.
in
Statistics
from
Macquarie
University,
Aust
ralia.
His
thesis
title
is
“Modified
fisher
scoring
algorithms
for
image
reconstruction
from
projection
”.
He
has
been
invited
as
a
“Visiting
professor
at
Prince
of
Songkla
University
a
nd
Kasetsar
t
University,
Thailand”.
He
is
a
Professor
of
Statistics
and
Data
Science
study
Pro
gram
at
IPB
University.
He
is
a
member
of
the
International
Statistical
Institute
and
a
f
ormer
Chairman
of
the
Indonesia
Statist
ical
Associatio
n.
He
is
actively
doing
competiti
ve
r
esearch
and
supervisin
g
undergraduate
and
postgraduate
students.
His
research
interests
inclu
de
mixed
models,
small
area estim
ation,
time s
eries analy
sis, an
d statistical machine learning.
During the last ten years,
he
has
taught
the
following
subjects
:
time
series
analysis
and
for
ecasting,
advanced
data
analysis
,
statis
tical
analysis
,
statis
tical
analysis
,
statis
tical
method,
and
data
science
.
He
can
be
contacted
at email
:
khairil@
apps.ipb.ac.id
.
Yenni
Angraini
is
academic
journey
led
her
to
excel
in
statis
tic
s,
earning
her
Bachelo
r’s,
Master’
s,
and
Doctora
te
degree
s
in
Statistics
from
IP
B
Universi
ty.
Currentl
y
serving
as
a
dedicated
faculty
member
.
She
is
a
respected
lecture
r
at
the
Department
of
Statistic
s,
Faculty
of
Mathem
atics
and
Natur
al
Scienc
es,
IPB
Univer
s
ity.
Her
commitme
nt
to
education
extends
beyond
the
classroom
,
as
evidenced
by
her
active
involv
ement
in
professional
organizations.
She
is
a
proud
m
ember
of
the
Ikat
an
Statistics
Indonesia
(Indonesian
Statist
ician
Associatio
n)
and
the
Forum
Pendidik
an
Tin
ggi
Statist
ika
Indonesia
(Indonesian
Higher
Education
Statist
ics
Forum).
For
inquiries
and
collaboratio
n
opportuni
ties
.
She
can be cont
acted at em
ail:
y_angraini@
apps.ipb.ac.id.
Evaluation Warning : The document was created with Spire.PDF for Python.