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
.
14
, N
o.
5
,
O
c
to
be
r
2025
, pp.
3493
~
3502
I
S
S
N
:
2252
-
8938
,
D
O
I
:
10.11591/
ij
a
i.
v
14
.i
5
.pp
3493
-
3502
3493
Jou
r
n
al
h
om
e
page
:
ht
tp
:
//
ij
ai
.
ia
e
s
c
or
e
.c
om
O
p
t
i
m
i
z
i
n
g
n
on
l
i
n
e
ar
au
t
or
e
gr
e
ss
i
ve
w
i
t
h
e
xoge
n
ou
s i
n
p
u
t
s
n
e
t
w
or
k
ar
c
h
i
t
e
c
t
u
r
e
f
or
aga
r
w
ood
o
i
l
q
u
al
i
t
y ass
e
ss
m
e
n
t
M
u
h
am
m
ad
I
k
h
s
an
R
o
s
la
n
1
, N
oor
A
id
a S
yak
ir
a A
h
m
ad
S
ab
r
i
1
, N
u
r
A
t
h
ir
ah
S
yaf
iq
ah
N
or
am
li
1
,
N
u
r
la
il
a I
s
m
ai
l
1
, Z
ak
ia
h
M
oh
d
Y
u
s
of
f
1
, A
li
A
b
d
A
lm
i
s
r
e
b
2
,
S
ai
f
u
l
N
iz
am
T
aj
u
d
d
in
3
, M
oh
d
N
as
ir
T
ai
b
1
1
A
dva
nc
e
d S
i
gna
l
P
r
oc
e
s
s
i
ng
R
e
s
e
a
r
c
h I
nt
e
r
e
s
t
G
r
oup, F
a
c
ul
t
y of
E
l
e
c
t
r
i
c
a
l
E
n
gi
ne
e
r
i
ng, U
ni
ve
r
s
i
t
i
T
e
knol
ogi
M
A
R
A
,
S
ha
h A
l
a
m
, M
a
l
a
ys
i
a
2
F
a
c
ul
t
y of
C
om
put
e
r
S
c
i
e
nc
e
a
nd E
ngi
ne
e
r
i
ng, I
nt
e
r
na
t
i
ona
l
U
ni
ve
r
s
i
t
y of
S
a
r
a
j
e
vo, I
l
i
dž
a
, B
os
ni
a
a
nd H
e
r
z
e
govi
na
3
B
i
o A
r
om
a
t
i
c
R
e
s
e
a
r
c
h
C
e
nt
r
e
of
E
xc
e
l
l
e
nc
e
(
B
A
R
C
E
)
, U
ni
ve
r
s
i
t
i
M
a
l
a
y
s
i
a
P
a
ha
ng A
l
-
S
ul
t
a
n A
bdul
l
a
h, G
a
m
ba
ng, M
a
l
a
y
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
O
c
t
21, 2024
R
e
vi
s
e
d
J
un 30, 2025
A
c
c
e
pt
e
d
J
ul
13, 2025
Agarwood
oil
is
highly
valued
in
perfumes,
incense,
and
trad
itional
medicine.
However
,
the
lack
of
standardiz
ed
grading
methods
poses
challenges
for
consist
ent
quality
assessmen
t.
This
study
proposes
a
data
-
driven
classification
approach
using
the
nonlinear
autoregressive
with
exogenous
inputs
(NARX)
model,
implem
ented
in
MATLAB
R2020
a
with
the
Levenberg
-
Marquardt
(LM)
algorithm.
The
dataset,
sourced
fro
m
the
Universiti
Malaysia
Pahang
Al
-
Sultan
Abdullah
und
er
the
Bio
Ar
omatic
Resear
ch
Centre
of
Excelle
nce
(BARCE)
and
Forest
Research
Institut
e
Malaysia
(FRIM),
comprises
chemical
compound
data
used
for
model
training
and
validation.
To
optimize
model
performa
nce,
the
num
ber
of
hidden
neurons
is
systematically
adjusted.
Model
evaluation
uses
performance
metrics
such
as
mean
squared
error
(MSE),
root
mean
s
quared
error
(RMSE),
mean
absolute
error
(MAE),
coefficient
of
determi
nation
(R²
),
epochs,
accuracy,
and
model
validation.
Results
show
that
the
NARX
model
effectively
classifies
a
garwood
oil
into
four
quality
grades
w
hich
is
high,
medium
-
high,
medium
-
low,
and
low.
The
best
performance
is
achieved
with
three
hidden
neurons,
offering
a
balance
between
ac
curacy
and
computational
efficiency.
This
work
demonstrates
the
poten
tial
of
automated
,
standardi
zed
a
garwood
oil
quality
grading.
Future
re
search
should
explore
alternative
training
algorithms
and
larger
datasets
to
further
enhance mod
el robus
tness an
d generali
zabilit
y.
K
e
y
w
o
r
d
s
:
A
ga
r
w
ood oil
qua
li
ty
G
r
a
de
di
s
c
r
im
in
a
ti
on
H
id
de
n ne
ur
ons
L
e
ve
nbe
r
g
-
M
a
r
qua
r
dt
M
a
c
hi
ne
l
e
a
r
ni
ng
N
A
R
X
ne
ur
a
l
ne
twor
k
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
:
N
ur
la
il
a
I
s
m
a
il
A
dva
nc
e
d S
ig
na
l
P
r
oc
e
s
s
in
g
R
e
s
e
a
r
c
h I
nt
e
r
e
s
t
G
r
oup, F
a
c
ul
ty
of
E
le
c
tr
ic
a
l
E
ngi
ne
e
r
in
g
U
ni
ve
r
s
it
i
T
e
knol
ogi
M
A
R
A
S
ha
h A
la
m
, M
a
la
ys
i
a
E
m
a
il
:
nur
la
il
a
0583@
ui
tm
.e
du.my
1.
I
N
T
R
O
D
U
C
T
I
O
N
A
ga
r
w
ood,
s
c
ie
nt
if
ic
a
ll
y
known
a
s
A
qui
la
r
ia
m
al
ac
c
e
n
s
is
L
a
m
,
be
lo
ngs
to
th
e
T
hym
e
la
e
a
c
e
a
e
f
a
m
il
y a
nd i
s
r
e
now
ne
d f
or
i
ts
hi
ghl
y va
lu
a
bl
e
, f
r
a
gr
a
nt
r
e
s
in
. T
hi
s
r
e
s
in
ous
w
ood, f
or
m
e
d i
n t
he
r
oot
s
, s
te
m
s
,
a
nd
br
a
nc
he
s
of
A
qui
la
r
ia
a
nd
G
y
r
in
op
s
tr
e
e
s
,
w
id
e
ly
ut
il
iz
e
d
in
tr
a
di
ti
ona
l
m
e
di
c
in
e
a
nd
r
e
li
gi
ous
pr
a
c
ti
c
e
s
a
c
r
os
s
S
out
he
a
s
t
A
s
ia
a
nd
N
or
th
e
a
s
t
I
ndi
a
.
O
ve
r
th
e
ye
a
r
s
,
th
e
s
e
r
e
gi
ons
ha
ve
be
c
om
e
th
e
pr
im
a
r
y
pr
oduc
e
r
s
of
a
ga
r
w
ood due
t
o i
ts
i
nc
r
e
a
s
in
g gl
oba
l
de
m
a
nd [
1]
–
[
3]
. C
ount
r
ie
s
s
uc
h a
s
C
hi
n
a
, I
ndi
a
, V
ie
tn
a
m
, I
ndone
s
ia
,
M
a
la
ys
ia
,
a
nd T
ha
il
a
nd
r
e
c
ogni
z
e
a
g
a
r
w
ood
f
or
it
s
di
s
ti
nc
ti
ve
a
r
om
a
,
le
a
di
ng
to
it
s
hi
gh
m
a
r
ke
t
va
lu
e
.
I
n
th
e
in
te
r
na
ti
ona
l
tr
a
de
,
pr
e
m
iu
m
-
gr
a
de
a
ga
r
w
ood
is
c
ons
id
e
r
e
d
m
o
r
e
va
lu
a
bl
e
th
a
n
gol
d.
H
ow
e
ve
r
,
th
e
im
m
e
ns
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
.
14
, N
o.
5
,
O
c
to
be
r
2025
:
3493
-
3502
3494
de
m
a
n
d
h
a
s
r
e
s
ul
t
e
d
in
s
e
v
e
r
e
de
pl
e
ti
on
o
f
n
a
tu
r
a
l
A
qui
la
r
i
a
f
or
e
s
t
s
,
a
s
n
e
a
r
ly
a
ll
c
o
unt
r
ie
s
in
v
ol
v
e
d
in
a
g
a
r
w
ood
h
a
r
v
e
s
ti
ng
ha
ve
f
a
c
e
d
de
f
or
e
s
ta
ti
o
n
d
ue
to
it
s
h
i
gh
e
c
on
om
i
c
w
or
t
h
[
2]
,
[
4]
,
[
5]
.
A
uni
qu
e
c
h
a
r
a
c
t
e
r
i
s
t
ic
of
A
q
ui
l
a
r
ia
t
r
e
e
s
i
s
t
ha
t
a
he
a
lt
hy
t
r
e
e
d
oe
s
not
na
tu
r
a
ll
y pr
o
du
c
e
a
g
a
r
w
oo
d. O
nl
y a
ppr
ox
im
a
te
ly
10%
of
th
e
s
e
t
r
e
e
s
d
e
v
e
lo
p
a
g
a
r
w
ood
w
it
hi
n
th
e
h
e
a
r
tw
oo
d,
p
r
i
m
a
r
i
ly
i
n
r
e
s
pon
s
e
t
o
e
xt
e
r
n
a
l
s
tr
e
s
s
or
s
s
u
c
h
a
s
li
gh
tn
i
ng
s
tr
i
ke
s
,
in
s
e
c
t
a
t
ta
c
k
s
,
or
b
a
c
te
r
ia
l
a
n
d
f
un
g
a
l
in
f
e
c
ti
o
ns
.
T
r
a
di
ti
on
a
l
a
g
a
r
w
ood
oi
l
gr
a
d
in
g
r
e
l
ie
s
o
n
s
e
n
s
or
y
e
v
a
lu
a
t
io
n
b
a
s
e
d
o
n
c
o
lo
r
,
vi
s
c
o
s
i
ty
,
o
dor
in
te
ns
it
y
,
a
n
d
p
e
r
s
i
s
t
e
n
c
e
[
6]
–
[
8]
.
H
o
w
e
ve
r
,
t
hi
s
m
e
t
ho
d
i
s
s
u
bj
e
c
ti
v
e
,
c
os
tl
y
,
a
nd
in
c
o
n
s
i
s
t
e
nt
,
a
s
it
d
e
p
e
n
ds
on
th
e
phy
s
ic
a
l
a
nd
e
m
ot
i
on
a
l
c
o
ndi
ti
o
n
of
e
va
lu
a
to
r
,
a
s
w
e
l
l
a
s
e
xt
e
r
n
a
l
e
nv
ir
on
m
e
nt
a
l
f
a
c
to
r
s
.
T
he
li
m
it
a
ti
o
n
s
h
a
v
e
d
e
m
o
ns
tr
a
te
d
a
hi
g
h
d
e
m
a
nd
f
or
a
n
e
f
f
i
c
i
e
nt
a
n
d
s
t
a
n
da
r
di
z
e
d
m
e
th
od
i
n
gr
a
d
in
g a
g
a
r
w
oo
d
oi
l.
T
hi
s
s
tu
dy a
im
s
t
o
a
d
dr
e
s
s
t
h
a
t
g
a
p by
u
ti
l
iz
i
ng a
d
va
nc
e
m
e
n
t
s
in
m
a
c
hi
ne
l
e
a
r
ni
n
g
to
de
ve
lo
p
a
m
od
e
l
f
or
a
g
a
r
w
oo
d
oi
l
qu
a
l
it
y
a
s
s
e
s
s
m
e
nt
[
9]
–
[
11]
.
R
e
c
e
nt
a
dv
a
nc
e
m
e
nt
s
in
a
g
a
r
w
ood
gr
a
di
ng
ha
ve
s
how
n
th
a
t
us
in
g
c
he
m
ic
a
l
c
om
po
s
it
io
n
pr
ovi
de
s
a
m
or
e
a
c
c
ur
a
te
a
nd
r
e
li
a
bl
e
s
ol
ut
io
n
f
or
de
te
r
m
in
in
g
qua
li
ty
.
C
ol
la
bor
a
ti
ons
w
it
h
in
dus
tr
ia
l
s
e
c
to
r
s
ha
ve
id
e
nt
if
ie
d
th
e
ke
y
m
a
r
ke
r
c
om
pounds
th
a
t
in
f
lu
e
nc
e
th
e
s
c
e
nt
of
a
ga
r
w
ood
oi
l,
in
c
lu
di
ng
φ
-
e
ude
s
m
ol
,
α
-
a
ga
r
of
ur
a
n,
β
-
a
ga
r
of
ur
a
n,
a
nd
10
-
e
pi
-
φ
-
e
ude
s
m
ol
[
12]
–
[
14
]
.
R
e
ga
r
dl
e
s
s
,
th
e
r
e
i
s
no
uni
ve
r
s
a
ll
y
a
c
c
e
pt
e
d
c
la
s
s
if
ic
a
ti
on
s
ys
te
m
f
or
gr
a
di
ng
a
ga
r
w
ood
oi
l
ye
t
to
a
ppe
a
r
.
M
ul
ti
pl
e
c
ount
r
ie
s
s
ti
ll
c
ont
in
ue
w
it
h
th
e
ir
ow
n
gr
a
di
ng
m
e
th
od
th
a
t
a
r
e
ba
s
e
d
on
c
li
e
nt
pr
e
f
e
r
e
nc
e
s
a
nd
pe
r
c
e
pt
io
ns
.
T
hi
s
in
di
c
a
te
s
how
de
m
a
ndi
ng
is
th
e
a
ga
r
w
ood
m
a
r
ke
t
f
or
gl
oba
l
s
ta
nda
r
d
in
gr
a
di
ng
pr
oc
e
s
s
,
c
ons
i
de
r
in
g
th
e
e
s
t
a
bl
is
he
d
e
xpl
or
a
ti
on
of
c
he
m
ic
a
l
pr
of
il
in
g
in
a
ga
r
w
ood
[
15]
–
[
17]
.
T
he
a
ga
r
w
ood
in
du
s
tr
y
hol
ds
m
a
jo
r
e
c
onomi
c
va
lu
e
,
w
it
h
pr
ic
e
s
hi
tt
in
g
th
e
r
a
nge
s
f
r
om
R
M
19,999
to
R
M
29,999
pe
r
ki
lo
gr
a
m
de
pe
nds
on
th
e
qua
li
ty
of
th
e
oi
l.
F
r
om
2019
to
2025
,
a
ga
r
w
ood
oi
l
ha
s
gr
ow
n
a
t
a
r
a
te
of
6.46%
a
nd
is
e
xpe
c
te
d
to
r
e
a
c
h
U
S
$201.03
m
il
li
on
in
gl
oba
l
m
a
r
ke
t.
M
a
la
ys
ia
r
a
nks
a
m
ong
th
e
to
p
e
xpor
te
r
s
of
a
ga
r
w
ood
f
or
gl
ob
a
l
di
s
tr
ib
ut
io
n
[
18]
,
[
19]
.
I
n
r
e
s
pons
e
to
th
is
,
m
a
ny
c
ol
la
bor
a
ti
ve
e
f
f
or
ts
ha
ve
e
m
e
r
ge
d,
in
c
lu
di
ng
th
e
pa
r
t
ne
r
s
hi
p
be
twe
e
n
U
ni
ve
r
s
it
i
M
a
la
ys
ia
P
a
ha
ng
Al
-
S
ul
ta
n
A
bdul
la
h,
U
ni
ve
r
s
it
i
T
e
knol
ogi
M
A
R
A
(
U
i
T
M
)
,
a
n
d
in
dus
tr
y
pl
a
ye
r
s
,
w
or
ki
ng
to
de
t
e
r
m
in
e
th
e
qua
li
ty
of
a
ga
r
w
ood oil
by i
de
nt
if
yi
ng ke
y c
he
m
ic
a
l
c
om
pone
nt
s
[
20]
–
[
22]
.
N
e
ur
a
l
ne
twor
ks
ha
v
e
be
e
n
us
e
d
in
c
la
s
s
if
ic
a
ti
on
ta
s
k
s
a
nd
a
r
e
known
f
or
th
e
ir
a
bi
li
ty
to
ha
ndl
e
c
om
pl
e
x
da
ta
s
e
ts
.
I
n
pa
r
ti
c
ul
a
r
,
th
e
nonl
in
e
a
r
a
ut
o
r
e
gr
e
s
s
iv
e
w
it
h
e
xoge
nous
(
N
A
R
X
)
m
ode
l
e
f
f
e
c
ti
ve
ly
c
a
pt
ur
e
s
dyna
m
ic
r
e
la
ti
ons
hi
ps
in
s
e
que
nt
ia
l
da
ta
,
m
a
ki
ng
it
us
e
f
ul
f
or
a
ga
r
w
ood
oi
l
gr
a
di
ng
[
23]
–
[
25
]
.
Q
ua
li
ty
e
va
lu
a
ti
on
ba
s
e
d
on
c
hr
om
a
to
gr
a
phi
c
te
c
hni
que
s
,
c
he
m
ic
a
l
c
om
pos
it
io
n
a
na
ly
s
is
,
a
nd
a
r
om
a
ti
c
a
s
s
e
s
s
m
e
nt
s
uc
h
a
s
ga
s
c
hr
om
a
to
gr
a
phy
-
f
la
m
e
io
ni
z
a
ti
on
de
te
c
ti
on
(
G
C
-
F
I
D
)
a
nd
ga
s
c
hr
om
a
to
gr
a
phy
-
m
a
s
s
s
pe
c
tr
om
e
tr
y
(
G
C
-
M
S
)
known
f
or
hi
gh
p
r
e
c
is
io
n.
H
ow
e
ve
r
,
t
he
s
e
m
e
th
ods
r
e
qui
r
e
hi
gh
c
os
ts
,
s
pe
c
ia
li
z
e
d
e
xpe
r
ti
s
e
to
op
e
r
a
te
th
e
e
qui
pm
e
nt
,
a
nd
s
ti
ll
in
vol
ve
s
ubj
e
c
ti
ve
s
e
n
s
or
y
e
va
lu
a
ti
on
[
26]
,
[
27]
.
M
a
c
hi
ne
le
a
r
ni
ng
of
f
e
r
s
s
c
a
la
bl
e
a
nd
a
ut
om
a
te
d
c
la
s
s
if
ic
a
ti
on
m
e
th
ods
. T
he
r
e
f
or
e
,
th
is
s
tu
dy
le
ve
r
a
ge
s
th
e
c
a
pa
bi
li
ti
e
s
of
m
a
c
hi
ne
le
a
r
ni
ng
by
a
ppl
yi
ng
th
e
N
A
R
X
m
ode
l
u
s
in
g
th
e
L
e
ve
nbe
r
g
-
M
a
r
qua
r
dt
(
L
M
)
a
lg
or
it
hm
to
a
s
s
ig
n
f
our
di
s
ti
nc
t
gr
a
de
s
of
a
ga
r
w
ood
oi
l:
hi
gh,
m
e
di
um
-
hi
gh,
m
e
di
um
-
lo
w
,
a
nd
lo
w
.
A
lt
hough
va
r
io
us
s
tu
di
e
s
ha
ve
e
xpl
or
e
d
m
a
c
hi
ne
le
a
r
ni
ng
f
or
oi
l
c
la
s
s
if
ic
a
ti
on,
r
e
s
e
a
r
c
h
on
th
e
a
ppl
ic
a
ti
on
of
N
A
R
X
f
or
a
g
a
r
w
ood
oi
l
gr
a
di
ng
is
s
ti
ll
li
m
it
e
d.
M
or
e
ove
r
,
none
of
th
e
m
e
m
pha
s
iz
e
th
e
im
pa
c
t
of
va
r
yi
ng
th
e
num
be
r
of
hi
dde
n
ne
ur
ons
on
c
la
s
s
if
ic
a
ti
on
a
c
c
ur
a
c
y.
B
y
s
y
s
te
m
a
ti
c
a
ll
y
m
o
di
f
yi
ng
th
e
hi
dde
n
ne
ur
ons
a
nd
e
va
lu
a
ti
ng
pe
r
f
or
m
a
nc
e
, t
hi
s
s
tu
dy of
f
e
r
s
a
de
e
pe
r
unde
r
s
ta
ndi
ng of
op
ti
m
i
z
in
g N
A
R
X
f
or
c
la
s
s
if
ic
a
ti
on t
a
s
ks
[
28]
, [
29
]
.
D
e
e
p
le
a
r
ni
ng
m
ode
l
s
s
u
c
h
a
s
c
onvolut
io
na
l
ne
ur
a
l
ne
twor
ks
(
C
N
N
s
)
a
nd
r
e
c
ur
r
e
nt
ne
ur
a
l
ne
twor
k
s
(
R
N
N
s
)
ha
ve
im
pr
ove
d
oi
l
c
la
s
s
if
ic
a
ti
on
a
c
c
ur
a
c
y,
pa
r
ti
c
ul
a
r
ly
in
f
e
a
tu
r
e
e
xt
r
a
c
ti
on
[
30
]
,
[
31
]
.
H
e
nc
e
,
th
is
s
tu
dy
a
li
gns
w
it
h t
he
ongoing deve
lo
pm
e
nt
o
f
m
a
c
hi
ne
l
e
a
r
ni
ng by
a
dopt
in
g N
A
R
X
-
ba
s
e
d c
la
s
s
if
ic
a
ti
on t
e
c
hni
que
s
a
nd
f
oc
us
in
g
on
hi
dde
n
ne
ur
on
opt
im
iz
a
ti
on.
M
A
T
L
A
B
R
202
0a
s
of
twa
r
e
is
us
e
d
in
th
is
gr
a
di
ng
pr
oc
e
s
s
to
s
uppor
t
th
e
im
pl
e
m
e
nt
a
ti
on,
s
im
ul
a
ti
on,
a
nd
v
a
li
da
ti
on
of
th
e
a
ga
r
w
ood
oi
l
m
ode
l.
B
y
in
c
or
por
a
ti
ng
a
ll
th
is
in
f
or
m
a
ti
on,
th
is
s
tu
dy
c
ont
r
ib
ut
e
s
to
th
e
d
e
ve
lo
pm
e
nt
of
a
s
ta
nda
r
di
z
e
d
in
te
ll
ig
e
nt
gr
a
di
ng
s
ys
te
m
f
or
hi
gh
-
va
lu
e
oi
ls
l
ik
e
a
ga
r
w
ood oil
.
2.
M
E
T
H
O
D
T
hi
s
s
e
c
t
io
n
is
d
iv
id
e
d
in
to
t
hr
e
e
m
a
i
n
pa
r
ts
:
s
u
b
s
e
c
ti
on
2.
1
is
N
A
R
X
m
od
e
l
de
ve
l
op
m
e
n
t
w
it
h
L
M
a
lg
o
r
i
th
m
,
s
ub
s
e
c
t
io
n
2
.2
is
e
xpe
r
i
m
e
n
ta
l
s
e
t
-
up
,
a
nd
s
u
b
s
e
c
t
io
n
2.3
is
p
e
r
f
o
r
m
a
nc
e
e
va
l
ua
t
io
n.
T
he
f
i
r
s
t
pa
r
t
w
a
lk
t
h
r
ou
gh
t
he
de
ve
l
op
m
e
n
t
o
f
th
e
N
A
R
X
m
ode
l,
hi
g
hl
ig
ht
o
n
how
th
e
L
M
a
lg
o
r
i
th
m
is
a
pp
li
e
d
w
it
h
a
n
o
pe
n
-
lo
o
p
s
t
r
uc
tu
r
e
a
n
d
s
e
le
c
te
d
ne
two
r
k
c
on
f
i
gu
r
a
t
io
ns
.
T
he
s
e
c
ond
pa
r
t
d
is
c
us
s
e
s
th
e
e
xpe
r
i
m
e
n
ta
l
s
e
t
-
up
,
c
ove
r
i
ng
da
ta
s
e
t
p
r
e
pr
oc
e
s
s
i
ng
,
f
e
a
tu
r
e
s
e
le
c
t
io
n,
a
n
d
d
a
ta
pa
r
t
it
io
ni
ng
.
T
he
f
in
a
l
pa
r
t
c
ove
r
s
pe
r
f
o
r
m
a
nc
e
e
va
lu
a
ti
on,
e
x
pl
a
in
in
g
th
e
s
ta
ti
s
ti
c
a
l
a
nd
va
l
id
a
ti
on
a
pp
r
oa
c
he
s
us
e
d
to
m
e
a
s
u
r
e
m
o
de
l
a
c
c
u
r
a
c
y
a
n
d
r
o
bus
t
ne
s
s
.
2.1. NA
R
X
m
od
e
l
d
e
ve
lo
p
m
e
n
t
w
it
h
L
M
al
gor
it
h
m
T
he
a
ga
r
w
ood
oi
l
s
a
m
pl
e
s
u
s
e
d
in
th
is
s
tu
dy
w
e
r
e
obt
a
in
e
d
th
r
ough
a
c
ol
la
bor
a
ti
on
be
twe
e
n
th
e
U
ni
ve
r
s
it
i
M
a
la
ys
ia
P
a
ha
ng
A
l
-
S
ul
ta
n
A
bdul
la
h
unde
r
th
e
B
io
A
r
om
a
ti
c
R
e
s
e
a
r
c
h
C
e
nt
r
e
of
E
xc
e
ll
e
nc
e
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
O
pt
imi
z
in
g nonli
ne
ar
aut
or
e
g
r
e
s
s
iv
e
w
it
h e
x
oge
nous
i
nput
s
ne
t
w
or
k
…
(
M
uhamm
ad I
k
hs
an R
os
la
n
)
3495
(
B
A
R
C
E
)
a
nd
F
or
e
s
t
R
e
s
e
a
r
c
h
I
ns
ti
tu
te
M
a
la
ys
ia
(
F
R
I
M
)
.
T
h
e
s
a
m
pl
e
s
f
r
om
th
e
da
ta
s
e
ts
us
e
d
in
th
is
s
tu
dy
ha
d
be
e
n
a
na
ly
z
e
d
by
pr
e
vi
ou
s
r
e
s
e
a
r
c
he
r
s
th
r
ough
s
ta
ti
s
ti
c
a
l
a
nd
pa
tt
e
r
n
a
na
ly
s
is
m
e
th
ods
[
32]
–
[
34]
.
I
n
th
is
s
tu
dy,
a
N
A
R
X
m
od
e
l
im
pl
e
m
e
nt
e
d
th
e
L
M
a
lg
or
it
hm
in
a
n
ope
n
-
lo
op
s
tr
uc
tu
r
e
.
I
n
th
i
s
a
r
r
a
nge
m
e
nt
,
pr
e
di
c
ti
ons
a
r
e
dr
iv
e
n
by
r
e
a
l
out
put
s
a
nd
in
put
va
r
ia
bl
e
s
,
w
it
h
no
f
e
e
dba
c
k
f
r
om
th
e
pr
e
vi
ous
pr
e
di
c
ti
ons
.
F
ig
ur
e
1
s
how
s
a
th
r
e
e
-
la
ye
r
ne
twor
k
a
r
c
hi
te
c
tu
r
e
c
ons
is
ts
of
in
put
,
hi
dde
n,
a
nd
out
put
la
ye
r
s
to
a
ll
ow
c
om
pl
e
x r
e
la
ti
ons
hi
ps
w
it
hi
n t
he
da
ta
t
o be
c
a
pt
ur
e
d [
35]
.
F
ig
ur
e
1. F
e
e
df
or
w
a
r
d L
M
-
N
A
R
X
ne
twor
k c
onf
ig
ur
a
ti
on
T
he
N
A
R
X
m
ode
l
de
ve
lo
pm
e
nt
f
ol
lo
w
s
a
s
tr
uc
tu
r
e
d
w
or
kf
lo
w
s
ta
r
ti
ng
w
it
h
s
a
m
pl
e
c
ol
le
c
ti
on
a
nd
da
ta
pr
e
pa
r
a
ti
on;
th
e
da
ta
s
e
t
c
on
s
is
ts
of
660
s
a
m
pl
e
s
c
ont
a
in
i
ng
e
le
ve
n
ke
y
c
he
m
ic
a
l
c
om
pounds
id
e
nt
if
ie
d
vi
a
s
ta
ti
s
ti
c
a
l
a
nd
pa
tt
e
r
n
a
na
ly
s
is
.
F
e
a
tu
r
e
s
e
le
c
ti
on
a
nd
pr
e
pr
oc
e
s
s
in
g;
ke
y
c
he
m
ic
a
l
c
om
pounds
w
e
r
e
s
e
le
c
te
d
b
a
s
e
d
on
th
e
ir
r
e
le
va
nc
e
to
a
ga
r
w
ood
oi
l
qu
a
li
ty
.
P
r
e
pr
oc
e
s
s
in
g
s
te
ps
in
c
lu
d
e
s
nor
m
a
li
z
a
ti
on,
r
a
ndomi
z
a
ti
on,
a
nd
da
ta
s
e
t
pa
r
ti
ti
oni
ng
(
70:
15:
15)
w
e
r
e
a
ppl
ie
d
to
e
nha
nc
e
m
ode
l
ge
ne
r
a
li
z
a
ti
on
[
36]
.
N
e
xt
on t
he
ne
twor
k
s
tr
uc
tu
r
e
s
e
le
c
ti
on;
a
t
hr
e
e
-
la
ye
r
f
e
e
df
or
w
a
r
d a
r
c
hi
te
c
tu
r
e
w
a
s
c
hos
e
n, c
om
pr
is
in
g i
nput
,
a
hi
dde
n
(
1
-
10)
,
a
nd
out
put
.
T
he
N
A
R
X
in
(
1
)
de
f
in
e
s
th
e
de
p
e
nde
nc
y
on
pr
e
vi
ous
in
put
s
a
nd
out
put
s
,
w
it
h
in
put
de
la
y (
D
x)
a
nd output
de
la
y (
D
y)
c
a
pt
ur
in
g t
e
m
por
a
l
de
pe
nde
nc
ie
s
[
37]
:
(
)
=
(
(
−
1
)
,
(
−
2
)
,
…
,
(
−
)
,
(
−
1
)
,
(
−
2
)
,
…
,
(
−
)
)
(
1)
T
r
a
in
in
g
pha
s
e
;
th
e
m
ode
l
w
a
s
tr
a
in
e
d
u
s
in
g
th
e
L
M
a
lg
or
it
hm
,
s
e
le
c
t
e
d
f
or
it
s
f
a
s
t
c
onve
r
ge
nc
e
a
nd
s
ta
bi
li
ty
in
ha
ndl
in
g
nonl
in
e
a
r
r
e
gr
e
s
s
io
n
pr
obl
e
m
s
.
A
f
te
r
th
a
t,
th
e
va
li
da
ti
on
a
nd
pe
r
f
or
m
a
nc
e
e
va
lu
a
ti
on
w
he
r
e
th
e
m
ode
l
w
a
s
va
li
da
te
d
us
in
g
m
ul
ti
pl
e
p
e
r
f
or
m
a
nc
e
m
e
tr
ic
s
(
m
e
a
n
s
qua
r
e
d
e
r
r
or
(
M
S
E
)
,
r
oot
m
e
a
n
s
qua
r
e
d
e
r
r
or
(
R
M
S
E
)
,
m
e
a
n
a
bs
ol
ut
e
e
r
r
or
(
M
A
E
)
,
c
oe
f
f
ic
ie
nt
of
de
te
r
m
in
a
ti
on
(
R
²)
,
e
poc
hs
,
a
nd a
c
c
ur
a
c
y)
.
T
he
m
ode
l
w
a
s
a
ls
o a
s
s
e
s
s
e
d
th
r
ough pe
r
f
or
m
a
nc
e
pl
ot
s
, a
nd r
e
gr
e
s
s
io
n pl
ot
s
.
O
pt
im
iz
a
ti
on
a
nd
m
ode
l
tu
ni
ng;
if
pe
r
f
or
m
a
nc
e
th
r
e
s
hol
ds
w
e
r
e
not
m
e
t,
it
e
r
a
ti
ve
a
dj
us
tm
e
nt
s
s
uc
h
a
s
m
odi
f
yi
ng
th
e
num
be
r
of
hi
dde
n
ne
ur
ons
or
tu
ni
ng
tr
a
in
in
g
pa
r
a
m
e
te
r
s
w
e
r
e
im
pl
e
m
e
nt
e
d
to
im
pr
ove
pr
e
di
c
ti
ve
a
c
c
ur
a
c
y.
T
o
im
pr
ove
th
e
a
bi
li
ty
of
m
ode
l
to
le
a
r
n
c
om
pl
e
x
pa
tt
e
r
ns
w
it
hi
n
th
e
d
a
ta
s
e
t,
th
e
in
put
de
la
y
is
a
dj
us
te
d
f
r
om
1:
2
to
1:
4.
T
hi
s
e
nh
a
nc
e
m
e
nt
a
ll
ow
s
t
he
N
A
R
X
m
ode
l
to
c
a
pt
ur
e
lo
nge
r
te
m
por
a
l
de
pe
nde
nc
ie
s
,
pr
ovi
di
ng
a
m
or
e
c
om
pr
e
he
n
s
iv
e
unde
r
s
t
a
ndi
ng
of
pa
s
t
in
put
s
[
3
8
]
,
[
3
9
]
.
B
y
in
c
or
por
a
ti
ng
a
br
oa
de
r
hi
s
to
r
ic
a
l
c
ont
e
xt
,
th
e
m
od
e
l
a
c
hi
e
v
e
s
be
tt
e
r
g
e
ne
r
a
li
z
a
ti
on,
r
e
duc
e
s
th
e
r
is
k
of
ove
r
f
it
ti
ng,
a
nd
ul
ti
m
a
te
ly
e
nha
nc
e
s
pr
e
di
c
ti
ve
a
c
c
ur
a
c
y.
2.2. E
xp
e
r
im
e
n
t
al
s
e
t
-
up
A
s
il
lu
s
tr
a
te
d
in
F
ig
ur
e
2,
th
e
da
ta
s
e
t
c
on
s
is
ts
of
660
s
a
m
pl
e
s
,
e
a
c
h
c
ont
a
in
in
g
e
le
ve
n
k
e
y
c
he
m
ic
a
l
c
om
pounds
:
γ
-
e
ude
s
m
ol
,
β
-
di
hydr
oa
ga
r
of
ur
a
n,
a
ll
o
-
a
r
om
a
de
ndr
e
ne
e
poxi
de
,
α
-
a
ga
r
of
ur
a
n,
a
r
-
c
ur
c
um
e
ne
,
va
le
r
ia
nol
,
α
-
gua
ie
ne
,
10
-
e
pi
-
γ
-
e
ude
s
m
ol
,
di
hydr
oc
ol
lu
m
e
ll
a
r
in
,
γ
-
c
a
di
ne
ne
,
β
-
a
ga
r
of
ur
a
n.
A
ll
c
he
m
ic
a
l
c
om
pounds
e
xhi
bi
t
va
r
yi
ng
a
bunda
nc
e
(
%
)
a
c
r
os
s
s
a
m
pl
e
s
,
pr
ovi
di
ng
va
r
ia
bi
li
ty
f
or
a
na
ly
z
in
g
th
e
r
e
la
ti
ons
hi
p
be
twe
e
n
c
he
m
ic
a
l
c
om
po
s
it
io
n
a
nd
a
ga
r
w
ood
oi
l
qua
li
ty
.
P
r
io
r
to
tr
a
in
in
g,
th
e
da
ta
s
e
t
unde
r
goe
s
pr
e
pr
oc
e
s
s
in
g,
in
c
lu
di
ng
nor
m
a
li
z
a
ti
on,
r
a
ndomi
z
a
ti
on,
a
nd
pa
r
ti
ti
oni
ng
in
to
tr
a
in
in
g,
va
li
da
ti
on,
a
nd
te
s
ti
ng
s
ubs
e
ts
.
T
he
s
e
s
te
ps
he
lp
to
r
e
gul
a
te
th
e
r
a
nge
of
in
put
,
im
pr
ove
da
ta
qua
li
ty
,
a
nd
r
e
duc
e
noi
s
e
s
w
it
hi
n
th
e
da
ta
s
e
t.
T
he
da
ta
s
e
t
i
s
di
vi
de
d
us
in
g
di
vi
d
e
r
a
nd
f
unc
ti
on
of
M
A
T
L
A
B
a
s
f
ol
lo
w
s
:
70%
of
tr
a
in
in
g
(
462
s
a
m
pl
e
s
)
,
15%
of
va
li
da
ti
on
(
99
s
a
m
pl
e
s
)
,
a
nd
15%
of
te
s
ti
ng
(
99
s
a
m
pl
e
s
)
.
T
hi
s
s
pl
it
he
lp
s
to
ba
la
nc
e
tr
a
in
in
g
pr
oc
e
s
s
a
nd
e
ns
ur
e
m
ode
l
to
pe
r
f
or
m
w
e
ll
on
uns
e
e
n
da
ta
.
I
n
a
s
s
e
s
s
in
g
th
e
pe
r
f
or
m
a
nc
e
,
di
f
f
e
r
e
nt
num
be
r
of
ne
ur
ons
u
ti
li
z
e
d
in
hi
dde
n
la
ye
r
c
onf
ig
ur
a
ti
ons
(
1
t
o
10
ne
ur
ons
)
.
M
A
T
L
A
B
R
2020a
is
s
e
le
c
te
d
f
or
im
pl
e
m
e
nt
a
ti
on
due
to
it
s
s
pe
c
ia
li
z
e
d
ne
ur
a
l
ne
twor
k
to
o
lb
ox,
w
hi
c
h
pr
ovi
de
s
bui
lt
-
in
c
a
pa
bi
li
ti
e
s
f
or
dyna
m
ic
s
ys
te
m
m
ode
li
ng
a
nd
ti
m
e
-
s
e
r
ie
s
f
or
e
c
a
s
ti
ng,
m
a
ki
ng
it
a
s
ui
ta
bl
e
c
hoi
c
e
ove
r
a
lt
e
r
na
ti
ve
s
s
uc
h
a
s
T
e
ns
or
F
lo
w
a
nd P
yT
or
c
h [
40]
.
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
.
14
, N
o.
5
,
O
c
to
be
r
2025
:
3493
-
3502
3496
F
ig
ur
e
2. D
e
ta
il
e
d e
xpe
r
im
e
nt
a
l
s
e
tu
p f
or
t
he
L
M
-
N
A
R
X
m
ode
l,
i
ll
us
tr
a
ti
ng da
ta
s
e
t
pr
e
pr
oc
e
s
s
in
g,
de
ve
lo
pm
e
nt
, t
e
s
ti
ng, a
nd
e
va
lu
a
ti
ng w
or
kf
lo
w
2.3. P
e
r
f
or
m
an
c
e
e
val
u
at
io
n
T
he
e
f
f
e
c
ti
v
e
ne
s
s
of
th
e
N
A
R
X
m
od
e
l
w
a
s
a
s
s
e
s
s
e
d
u
s
in
g
m
u
lt
ip
le
p
e
r
f
or
m
a
n
c
e
m
e
tr
i
c
s
,
in
c
lu
di
ng
M
S
E
,
R
M
S
E
,
M
A
E
,
R
²,
a
nd
a
c
c
ur
a
c
y.
T
he
s
e
p
e
r
f
or
m
a
n
c
e
m
e
tr
ic
s
w
e
r
e
c
hos
e
n
f
or
t
he
ir
s
i
gni
f
ic
a
nc
e
in
e
va
lu
a
ti
ng
bot
h
r
e
gr
e
s
s
io
n
a
nd
c
l
a
s
s
if
ic
a
ti
on
ta
s
k
s
. M
S
E
m
e
a
s
ur
e
s
t
he
a
ve
r
a
ge
s
qua
r
e
d
e
r
r
or
but
i
s
s
e
ns
it
iv
e
to
la
r
ge
d
e
vi
a
ti
ons
.
R
M
S
E
,
e
xpr
e
s
s
e
d
in
t
he
s
a
m
e
u
ni
t
a
s
th
e
t
a
r
g
e
t
va
r
ia
bl
e
,
im
pr
ove
s
in
te
r
pr
e
ta
bi
li
ty
but
d
oe
s
not
di
s
t
in
gui
s
h
ov
e
r
-
a
n
d
und
e
r
-
pr
e
di
c
ti
on
s
.
M
S
E
a
nd
R
M
S
E
qua
nt
if
y
pr
e
di
c
ti
on
e
r
r
or
s
,
w
it
h
R
M
S
E
pl
a
c
in
g
gr
e
a
t
e
r
e
m
pha
s
i
s
on
la
r
ge
r
d
is
c
r
e
p
a
nc
i
e
s
.
M
A
E
,
w
hi
c
h
c
a
l
c
ul
a
te
s
a
b
s
ol
ut
e
e
r
r
or
s
,
i
s
l
e
s
s
s
e
ns
it
iv
e
to
out
li
e
r
s
but
doe
s
not
h
e
a
v
il
y
pe
na
li
z
e
l
a
r
ge
d
e
vi
a
t
io
n
s
.
R
²
a
s
s
e
s
s
e
s
m
ode
l
f
it
,
w
it
h
va
l
ue
s
ne
a
r
1
in
di
c
a
ti
n
g
s
tr
on
g
c
or
r
e
la
t
io
n,
t
hough
hi
gh
R
²
doe
s
n
ot
gu
a
r
a
nt
e
e
ge
n
e
r
a
li
z
a
ti
on.
A
c
c
ur
a
c
y
m
e
a
s
ur
e
s
c
or
r
e
c
t
c
la
s
s
if
ic
a
t
io
n
s
but
m
a
y
ov
e
r
lo
ok
c
l
a
s
s
im
b
a
la
nc
e
s
.
A
not
h
e
r
k
e
y
f
a
c
to
r
i
s
th
e
n
um
be
r
of
e
po
c
hs
,
w
hi
c
h
r
e
pr
e
s
e
nt
s
t
he
t
ot
a
l
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
O
pt
imi
z
in
g nonli
ne
ar
aut
or
e
g
r
e
s
s
iv
e
w
it
h e
x
oge
nous
i
nput
s
ne
t
w
or
k
…
(
M
uhamm
ad I
k
hs
an R
os
la
n
)
3497
it
e
r
a
ti
o
ns
th
e
m
od
e
l
und
e
r
go
e
s
d
ur
in
g
tr
a
in
in
g.
S
e
le
c
ti
ng
a
n
opt
im
a
l
e
p
oc
h
c
ount
is
c
r
it
ic
a
l
w
h
e
r
e
a
s
in
s
uf
f
ic
ie
nt
e
po
c
h
s
m
a
y
l
e
a
d t
o
und
e
r
f
it
ti
ng,
w
he
r
e
a
s
e
xc
e
s
s
iv
e
t
r
a
in
in
g
c
a
n
c
a
us
e
ov
e
r
f
it
ti
ng
[
41]
.
T
o
a
s
s
e
s
s
th
e
r
e
li
a
bi
li
ty
of
N
A
R
X
m
ode
l,
va
li
da
ti
on
te
c
hni
que
s
s
uc
h
a
s
pe
r
f
or
m
a
nc
e
a
nd
r
e
gr
e
s
s
io
n
pl
ot
s
w
e
r
e
ut
il
iz
e
d.
T
he
pe
r
f
or
m
a
nc
e
pl
ot
il
lu
s
tr
a
te
s
tr
a
in
in
g,
va
li
da
ti
on,
a
nd
te
s
t
e
r
r
or
s
a
c
r
os
s
e
poc
hs
,
pr
ovi
de
in
f
or
m
a
ti
on
on
th
e
le
a
r
ni
ng
be
ha
vi
or
of
m
ode
l
a
nd
ge
n
e
r
a
li
z
a
ti
on
c
a
pa
bi
li
ti
e
s
.
I
de
a
ll
y,
a
w
e
ll
-
tr
a
in
e
d
m
ode
l
di
s
pl
a
ys
s
te
a
di
ly
de
c
r
e
a
s
in
g
e
r
r
or
s
th
a
t
la
te
r
s
ta
bi
li
z
e
.
I
n
c
ont
r
a
s
t,
a
gr
ow
in
g
ga
p
be
twe
e
n
tr
a
in
in
g
a
nd
va
li
da
ti
on
e
r
r
or
s
m
a
y
in
di
c
a
te
ove
r
f
it
ti
ng
[
42
]
.
T
he
r
e
gr
e
s
s
io
n
pl
ot
hi
ghl
ig
ht
s
th
e
pr
e
di
c
ti
ons
of
m
ode
l
w
it
h
a
c
tu
a
l
va
lu
e
s
;
w
h
e
n
y=
x,
w
he
r
e
a
s
da
ta
poi
nt
s
li
e
c
lo
s
e
to
th
e
li
ne
in
di
c
a
ti
ng
s
tr
ong
a
c
c
ur
a
c
y
of
th
e
m
ode
l.
W
hi
le
pr
e
s
e
nc
e
s
of
de
vi
a
ti
ons
a
nd outl
ie
r
s
m
a
y
s
ig
na
l
th
a
t
th
e
p
a
tt
e
r
ns
of
t
he
m
ode
l
ha
ve
not
f
ul
ly
c
a
pt
ur
e
d.
3.
R
E
S
U
L
T
S
A
N
D
D
I
S
C
U
S
S
I
O
N
T
hi
s
s
e
c
ti
on
di
s
c
us
s
e
s
th
e
out
c
om
e
s
f
r
om
th
e
s
tu
dy
w
it
h
f
ig
ur
e
s
a
nd
ta
bl
e
s
.
T
h
e
s
e
c
ti
on
is
di
vi
de
d
in
to
two
s
ub
s
e
c
ti
ons
.
S
e
c
ti
on
3.1
f
oc
us
e
s
on
N
A
R
X
m
ode
l
p
e
r
f
or
m
a
nc
e
us
in
g
e
v
a
lu
a
ti
on
m
e
tr
ic
s
s
uc
h
a
s
M
S
E
,
R
M
S
E
,
M
A
E
,
R
²,
e
poc
h
s
,
a
nd
a
c
c
ur
a
c
y.
S
e
c
ti
on
3.2
a
d
dr
e
s
s
e
s
th
e
m
od
e
l
va
li
da
ti
on,
in
te
r
pr
e
ti
ng
th
e
pl
ot
s
f
r
om
th
e
pe
r
f
or
m
a
nc
e
a
nd
r
e
gr
e
s
s
io
n
pl
ot
s
.
T
hi
s
e
ns
ur
e
s
th
e
r
e
li
a
bi
li
ty
,
he
lp
s
de
te
c
t
pot
e
nt
ia
l
is
s
ue
s
s
uc
h a
s
ov
e
r
f
it
ti
ng or
unde
r
f
it
ti
ng, a
nd pr
ovi
de
s
i
nf
or
m
a
ti
on f
o
r
ge
ne
r
a
li
z
a
ti
on a
bi
li
ty
.
3.1. M
S
E
, R
M
S
E
, M
A
E
,
R
2
, e
p
oc
h
s
an
d
ac
c
u
r
ac
y
F
or
th
e
e
va
lu
a
ti
on
of
th
e
L
M
-
ba
s
e
d
N
A
R
X
m
ode
l,
th
e
n
um
be
r
of
ne
ur
ons
in
hi
dde
n
la
ye
r
w
a
s
v
a
r
ie
d
f
r
om
1
unt
il
10
a
s
s
how
n
in
T
a
bl
e
1
th
a
t
s
um
m
a
r
i
z
e
s
th
e
r
e
s
ul
t
s
.
T
he
M
S
E
va
lu
e
s
r
a
nge
d
f
r
om
10⁻
²
to
10⁻
³,
in
di
c
a
ti
ng
pr
e
c
is
e
pr
e
di
c
ti
ons
a
nd
c
ons
i
s
te
nt
da
ta
qua
li
ty
.
M
o
s
t
c
onf
ig
ur
a
ti
ons
a
c
hi
e
v
e
d
c
la
s
s
if
ic
a
ti
on
a
c
c
ur
a
c
y
a
bove
80%
,
w
hi
le
R
²
va
lu
e
s
r
e
m
a
in
e
d
a
t
0.99,
r
e
f
le
c
ti
ng
a
s
tr
ong
c
or
r
e
la
ti
on
be
twe
e
n
pr
e
di
c
te
d
out
put
s
a
nd
a
c
tu
a
l
gr
a
de
s
.
T
he
m
o
s
t
opt
im
a
l
c
onf
ig
ur
a
ti
on
a
c
r
os
s
th
e
pe
r
f
or
m
a
nc
e
m
e
a
s
ur
e
is
th
r
e
e
ne
ur
ons
,
a
c
hi
e
vi
ng
a
n
M
S
E
of
2.158×
10⁻
³, R
M
S
E
of
0.046,
M
A
E
of
0.019,
R
²
of
0.99,
7
e
poc
hs
a
nd
99.54%
a
c
c
ur
a
c
y.
T
hi
s
a
ppr
oa
c
h
e
f
f
e
c
ti
ve
ly
c
la
s
s
if
ie
s
a
ga
r
w
ood
oi
l
in
to
f
ou
r
gr
a
de
s
w
hi
le
m
a
in
ta
in
in
g
m
ode
l
s
im
pl
ic
it
y.
B
e
yond
th
r
e
e
ne
ur
on
s
,
a
ddi
ti
ona
l
c
om
pl
e
xi
ty
pr
ovi
de
d
m
in
im
a
l
im
pr
ove
m
e
nt
.
T
he
s
e
r
e
s
ul
ts
c
onf
ir
m
t
he
L
M
-
N
A
R
X
e
f
f
ic
ie
nc
y a
nd r
obus
tn
e
s
s
of
m
ode
l
a
c
r
os
s
a
ll
c
onf
ig
ur
a
ti
ons
.
T
a
bl
e
1.
P
e
r
f
or
m
a
nc
e
m
e
tr
ic
s
of
L
M
-
N
A
R
X
m
ode
l
w
it
h va
r
yi
n
g hi
dde
n ne
ur
ons
(
1
-
10)
H
i
dde
n ne
ur
ons
M
S
E
R
M
S
E
M
A
E
R
2
E
poc
hs
A
c
c
ur
a
c
y (
%
)
1
5.882×10
-
2
0.243
0.151
0.69
17
81.71
2
1.543×10
-
2
0.124
0.048
0.92
12
95.27
*3
2.158×10
-
3
0.046
0.019
0.99
7
99.54
4
1.991×10
-
3
0.045
0.015
0.99
9
99.54
5
1.971×10
-
3
0.044
0.016
0.99
8
99.54
6
1.595×10
-
3
0.040
0.013
0.99
8
99.54
7
1.528×10
-
3
0.039
0.012
0.99
7
99.85
8
1.656
×
10
-
3
0.040
0.013
0.99
10
99.70
9
1.522
×
10
-
3
0.039
0.016
0.99
9
99.70
10
1.875
×
10
-
3
0.043
0.010
0.99
13
99.70
*B
e
s
t
hi
dde
n ne
ur
on i
n L
M
-
N
A
R
X
m
ode
l
3.2. NA
R
X
m
od
e
l
val
id
at
io
n
S
ubs
e
c
ti
on
3.1
e
s
ta
bl
is
h
e
d
th
a
t
th
r
e
e
hi
dde
n
ne
ur
ons
yi
e
ld
e
d
t
he
be
s
t
pe
r
f
or
m
a
nc
e
,
but
th
is
s
e
c
ti
on
e
va
lu
a
te
s
th
e
N
A
R
X
m
ode
l
w
it
h
th
r
e
e
hi
dde
n
ne
ur
ons
to
f
ur
th
e
r
a
na
ly
z
e
it
s
be
ha
vi
or
.
T
he
pe
r
f
or
m
a
nc
e
pl
ot
in
F
ig
ur
e
3
il
lu
s
tr
a
te
s
th
e
le
a
r
ni
ng
pr
oc
e
s
s
,
tr
a
c
ki
ng
tr
a
in
in
g,
v
a
li
da
ti
on,
a
nd
te
s
t
e
r
r
or
s
of
m
ode
l
ov
e
r
e
poc
h
s
.
I
ni
ti
a
ll
y,
a
ll
e
r
r
o
r
s
de
c
li
ne
to
ge
th
e
r
,
r
e
f
le
c
ti
ng
e
f
f
e
c
ti
ve
pa
tt
e
r
n
le
a
r
ni
ng.
T
he
y
e
ve
nt
ua
ll
y
s
ta
bi
li
z
e
,
in
di
c
a
ti
ng
opt
im
a
l
tr
a
in
in
g
w
it
h
no
f
ur
th
e
r
s
ig
ni
f
ic
a
nt
im
p
r
ove
m
e
nt
.
A
s
m
a
ll
ga
p
be
twe
e
n
tr
a
in
in
g
a
nd
va
li
da
ti
on/
te
s
t
e
r
r
or
s
s
ugge
s
ts
good
ge
ne
r
a
li
z
a
ti
on,
w
hi
le
a
la
r
ge
ga
p
w
oul
d
i
ndi
c
a
te
ove
r
f
it
ti
ng.
T
he
pa
r
a
ll
e
l
tr
e
nd
b
e
twe
e
n
va
li
da
ti
on
a
nd
te
s
t
e
r
r
or
s
c
onf
ir
m
s
m
ode
l
c
ons
is
te
nc
y.
S
i
nc
e
a
ll
c
ur
ve
s
c
onve
r
ge
s
m
oot
hl
y
w
it
hout
di
ve
r
ge
nc
e
, t
he
m
ode
l
de
m
ons
tr
a
te
s
e
f
f
e
c
ti
ve
r
e
gul
a
r
iz
a
ti
on a
n
d a
voi
ds
ove
r
f
it
ti
ng.
T
he
r
e
gr
e
s
s
io
n
pl
ot
in
F
ig
ur
e
4
e
va
lu
a
te
s
th
e
c
or
r
e
la
ti
on
be
twe
e
n
pr
e
di
c
te
d
a
nd
a
c
tu
a
l
out
put
s
.
T
he
r
e
gr
e
s
s
io
n
li
ne
s
f
or
tr
a
in
in
g,
va
li
da
ti
on,
a
nd
te
s
t
s
e
ts
c
lo
s
e
ly
a
li
gn
w
it
h
th
e
id
e
a
l
y=
x
li
ne
,
c
onf
ir
m
in
g
a
s
tr
ong
pr
e
di
c
ti
ve
r
e
la
ti
ons
hi
p.
L
ow
r
e
s
id
ua
l
va
lu
e
s
in
th
e
pl
ot
s
in
di
c
a
te
hi
gh
a
c
c
ur
a
c
y
of
th
e
m
ode
l,
th
u
s
by
ha
vi
ng
th
e
s
e
f
our
te
e
n
da
ta
poi
nt
s
of
out
li
e
r
s
m
os
t
l
ik
e
ly
r
e
s
ul
t
f
r
om
th
e
va
r
ia
bi
li
ty
w
it
hi
n
th
e
da
ta
s
e
t
a
nd
di
d
not
s
ig
ni
f
ic
a
nt
ly
a
f
f
e
c
t
th
e
ove
r
a
ll
m
ode
l
pe
r
f
or
m
a
nc
e
.
T
he
f
in
di
ngs
a
s
s
ur
e
s
th
a
t
th
e
pr
opos
e
d
N
A
R
X
m
ode
l
is
c
a
pa
bl
e
i
n pr
oduc
in
g a
c
c
ur
a
te
a
nd c
on
s
is
te
nt
c
la
s
s
if
ic
a
ti
on r
e
s
ul
ts
w
it
h only m
in
or
va
r
ia
nc
e
s
.
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
.
14
, N
o.
5
,
O
c
to
be
r
2025
:
3493
-
3502
3498
F
ig
ur
e
3. T
r
a
in
in
g, va
li
da
ti
on, a
nd t
e
s
t
pe
r
f
or
m
a
nc
e
of
t
he
N
A
R
X
m
ode
l
w
it
h t
hr
e
e
hi
dde
n l
a
ye
r
s
, s
ho
w
in
g
e
r
r
or
c
onve
r
ge
nc
e
a
nd s
ta
bi
li
ty
a
c
r
os
s
e
poc
h
s
F
ig
ur
e
4. R
e
gr
e
s
s
io
n pl
ot
of
N
A
R
X
m
ode
l
on t
hr
e
e
hi
dde
n ne
ur
ons
w
it
h s
m
a
ll
r
e
s
id
ua
ls
a
nd outl
ie
r
s
4.
C
O
N
C
L
U
S
I
O
N
T
hi
s
s
tu
dy
opt
im
iz
e
s
th
e
a
ppl
ic
a
ti
on
of
th
e
N
A
R
X
ne
ur
a
l
n
e
twor
k
in
c
la
s
s
if
yi
ng
a
ga
r
w
ood
oi
l
qua
li
ty
us
in
g
th
e
L
M
a
lg
or
it
hm
.
B
y
m
odi
f
yi
ng
th
e
hi
dde
n
ne
u
r
on,
m
ode
l
pe
r
f
or
m
a
nc
e
w
a
s
m
e
a
s
ur
e
d
us
in
g
m
e
tr
ic
s
,
in
c
lu
di
ng
M
S
E
,
R
M
S
E
,
M
A
E
,
R
²,
num
be
r
of
e
poc
h
s
,
a
nd
a
c
c
ur
a
c
y.
T
he
c
onf
ig
ur
a
ti
on
w
it
h
th
r
e
e
hi
dde
n ne
ur
ons
c
om
e
out
a
s
t
he
m
o
s
t
e
f
f
e
c
ti
ve
, s
tr
ik
e
s
a
b
a
la
nc
e
be
twe
e
n a
c
c
ur
a
c
y a
nd e
f
f
ic
ie
nc
y.
T
he
r
e
s
ul
ts
s
how
e
d
m
in
im
a
l
pr
e
di
c
ti
on
e
r
r
or
,
a
nd
s
tr
ong
c
onve
r
ge
nc
e
be
ha
vi
or
.
T
he
r
e
gr
e
s
s
io
n
a
na
ly
s
is
pr
ove
d
good
ba
la
nc
e
of
pr
e
di
c
te
d
a
nd
a
c
tu
a
l
va
lu
e
s
w
it
h
di
s
tr
ib
ut
e
d
r
e
s
id
ua
l
s
.
T
h
e
s
e
f
in
di
ngs
s
ugge
s
t
th
a
t
th
e
L
M
-
N
A
R
X
m
ode
l
is
s
ui
ta
bl
e
f
or
c
la
s
s
if
yi
ng
a
ga
r
w
ood
oi
l.
H
ow
e
v
e
r
,
th
e
r
e
is
li
m
it
a
ti
ons
to
hi
ghl
ig
ht
in
th
is
s
tu
dy.
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
O
pt
imi
z
in
g nonli
ne
ar
aut
or
e
g
r
e
s
s
iv
e
w
it
h e
x
oge
nous
i
nput
s
ne
t
w
or
k
…
(
M
uhamm
ad I
k
hs
an R
os
la
n
)
3499
W
hi
le
th
e
da
ta
s
e
t
of
660
s
a
m
pl
e
s
pr
ovi
de
s
a
r
obu
s
t
r
e
s
ul
t
,
e
xpa
ndi
ng
th
e
d
a
ta
s
e
t
s
c
oul
d
im
pr
ove
ge
ne
r
a
li
z
a
bi
li
ty
.
F
ur
th
e
r
im
pr
ove
m
e
nt
s
c
oul
d
a
ls
o
be
m
a
d
e
by
f
in
e
-
tu
ni
ng
hype
r
pa
r
a
m
e
te
r
s
,
s
uc
h
a
s
th
e
le
a
r
ni
ng
r
a
te
.
E
xpl
or
in
g
ot
he
r
m
ode
ls
li
ke
L
S
T
M
m
a
y
of
f
e
r
a
dva
nt
a
ge
s
in
ha
ndl
in
g
c
om
pl
e
x
da
ta
s
tr
uc
tu
r
e
s
.
F
or
r
e
a
l
-
w
or
ld
us
e
, t
he
m
ode
l
w
oul
d ne
e
d t
o be
va
li
da
te
d a
ga
in
s
t
s
ta
nda
r
di
z
e
d i
ndus
tr
y da
ta
s
e
t
s
t
o c
onf
ir
m
i
ts
s
c
a
la
bi
li
ty
a
nd
r
e
s
il
ie
nc
e
.
F
ut
ur
e
w
or
k
m
a
y
in
vol
v
e
e
xpe
r
im
e
nt
in
g
w
it
h
a
lt
e
r
na
ti
ve
tr
a
in
in
g
te
c
hni
que
s
to
f
ur
th
e
r
e
nha
nc
e
th
e
a
da
pt
a
bi
li
ty
.
O
ve
r
a
ll
,
th
is
s
tu
dy
unde
r
s
c
or
e
s
th
e
va
lu
e
of
m
a
c
hi
ne
le
a
r
ni
ng
in
pr
oduc
t
qua
li
ty
e
va
lu
a
ti
on a
nd i
ts
pot
e
nt
ia
l
to
i
m
pr
ove
gr
a
di
ng c
ons
is
te
nc
y a
nd c
om
m
e
r
c
ia
l
va
lu
e
.
A
C
K
N
O
WL
E
D
G
M
E
N
T
S
T
he
a
ut
hor
s
w
oul
d
li
ke
to
e
xpr
e
s
s
th
e
ir
s
in
c
e
r
e
gr
a
ti
tu
de
to
th
e
M
in
is
tr
y
of
H
ig
he
r
E
duc
a
ti
on
M
a
la
ys
ia
f
or
f
in
a
nc
ia
l
s
uppor
t
pr
ovi
de
d
th
r
ough
U
ni
ve
r
s
it
i
T
e
knol
ogi
M
A
R
A
,
unde
r
th
e
I
ns
ti
tu
te
of
P
os
tg
r
a
dua
te
S
tu
di
e
s
(
I
P
S
is
)
a
nd
th
e
F
a
c
ul
ty
of
E
le
c
tr
ic
a
l
E
ng
in
e
e
r
in
g
(
F
K
E
)
,
vi
a
th
e
J
our
na
l
S
uppor
t
F
und
(
J
S
F
)
.
S
pe
c
ia
l
th
a
nks
a
r
e
e
xt
e
nde
d
to
th
e
m
e
m
be
r
s
of
th
e
A
dva
nc
e
d
S
ig
na
l
P
r
oc
e
s
s
in
g
R
e
s
e
a
r
c
h
I
nt
e
r
e
s
t
G
r
oup
f
or
th
e
ir
c
ons
tr
uc
ti
ve
f
e
e
dba
c
k
a
nd
c
ont
r
ib
ut
io
ns
.
T
he
a
ut
hor
s
a
ls
o
w
is
h
to
a
c
knowle
dge
th
e
va
lu
a
bl
e
s
uppor
t
of
th
e
F
or
e
s
t
R
e
s
e
a
r
c
h
I
n
s
ti
tu
te
M
a
la
y
s
ia
(
F
R
I
M
)
a
nd
th
e
B
io
A
r
om
a
ti
c
R
e
s
e
a
r
c
h
C
e
nt
r
e
of
E
xc
e
ll
e
nc
e
(
B
A
R
C
E
)
a
t
U
ni
ve
r
s
it
i
M
a
l
a
ys
ia
P
a
h
a
ng
A
l
-
S
ul
ta
n
A
bdul
la
h
(
U
M
P
S
A
)
f
or
th
e
ir
a
s
s
is
t
a
nc
e
in
da
ta
c
ol
le
c
ti
on a
nd pr
e
pa
r
a
ti
on.
F
U
N
D
I
N
G
I
N
F
O
R
M
A
T
I
O
N
T
hi
s
w
or
k
w
a
s
s
uppor
te
d
by
f
undi
ng
f
r
om
th
e
M
in
is
tr
y
of
H
ig
he
r
E
duc
a
ti
on
M
a
la
ys
ia
,
pr
ovi
de
d
th
r
ough
U
ni
ve
r
s
it
i
T
e
knol
ogi
M
A
R
A
(
U
iT
M
)
unde
r
th
e
I
n
s
ti
tu
te
of
P
os
tg
r
a
dua
te
S
tu
di
e
s
(
I
P
S
is
)
a
nd
th
e
F
a
c
ul
ty
of
E
le
c
tr
ic
a
l
E
ngi
ne
e
r
in
g (
F
K
E
)
, t
h
r
ough the
J
our
na
l
S
uppor
t
F
und (
J
S
F
)
.
A
U
T
H
O
R
C
O
N
T
R
I
B
U
T
I
O
N
S
S
T
A
T
E
M
E
N
T
T
hi
s
jo
ur
na
l
us
e
s
th
e
C
ont
r
ib
ut
or
R
ol
e
s
T
a
xonomy
(
C
R
e
di
T
)
to
r
e
c
ogni
z
e
in
di
vi
dua
l
a
ut
hor
c
ont
r
ib
ut
io
ns
, r
e
duc
e
a
ut
hor
s
hi
p di
s
put
e
s
,
a
nd f
a
c
il
it
a
te
c
ol
la
bo
r
a
ti
on.
N
am
e
o
f
A
u
t
h
or
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
M
uha
m
m
a
d I
khs
a
n
R
os
la
n
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
N
oor
A
id
a
S
ya
ki
r
a
A
hm
a
d S
a
br
i
✓
✓
✓
✓
✓
N
ur
A
th
ir
a
h S
ya
f
iq
a
h
N
or
a
m
li
✓
✓
✓
✓
✓
N
ur
la
il
a
I
s
m
a
il
✓
✓
✓
✓
✓
✓
✓
✓
Z
a
ki
a
h M
ohd Yus
of
f
✓
✓
✓
✓
✓
✓
✓
A
li
A
bd A
lm
is
r
e
b
✓
✓
✓
✓
S
a
if
ul
N
iz
a
m
T
a
ju
ddi
n
✓
✓
✓
✓
M
ohd Na
s
ir
T
a
ib
✓
✓
✓
✓
✓
✓
✓
C
:
C
onc
e
pt
ua
l
i
z
a
t
i
on
M
:
M
e
t
hodol
ogy
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
da
t
i
on
Fo
:
Fo
r
m
a
l
a
na
l
ys
i
s
I
:
I
nve
s
t
i
ga
t
i
on
R
:
R
e
s
our
c
e
s
D
:
D
a
t
a
C
ur
a
t
i
on
O
:
W
r
i
t
i
ng
-
O
r
i
gi
na
l
D
r
a
f
t
E
:
W
r
i
t
i
ng
-
R
e
vi
e
w
&
E
di
t
i
ng
Vi
:
Vi
s
ua
l
i
z
a
t
i
on
Su
:
Su
pe
r
vi
s
i
on
P
:
P
r
oj
e
c
t
a
dm
i
ni
s
t
r
a
t
i
on
Fu
:
Fu
ndi
ng a
c
qui
s
i
t
i
on
C
O
N
F
L
I
C
T
O
F
I
N
T
E
R
E
S
T
S
T
A
T
E
M
E
N
T
A
ut
hor
s
s
ta
te
no c
onf
li
c
t
of
i
nt
e
r
e
s
t.
D
A
T
A
A
V
A
I
L
A
B
I
L
I
T
Y
D
a
ta
a
va
il
a
bi
li
ty
is
not
a
ppl
ic
a
bl
e
to
th
is
p
a
pe
r
a
s
no
ne
w
da
ta
w
e
r
e
c
r
e
a
te
d
or
a
n
a
ly
z
e
d
in
th
is
s
tu
dy.
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
.
14
, N
o.
5
,
O
c
to
be
r
2025
:
3493
-
3502
3500
R
E
F
E
R
E
N
C
E
S
[
1]
S
.
M
a
,
Y
.
C
he
n,
T
.
Y
a
n,
J
.
Q
i
n,
a
nd
G
.
L
i
,
“
U
l
t
r
a
s
ound
-
l
a
c
c
a
s
e
pr
e
-
t
r
e
a
t
m
e
nt
e
nha
nc
e
s
a
g
a
r
w
ood
e
s
s
e
nt
i
a
l
oi
l
e
xt
r
a
c
t
i
on
a
nd
bi
oa
c
t
i
vi
t
y,”
I
nt
e
r
nat
i
onal
J
our
nal
of
B
i
ol
ogi
c
al
M
ac
r
om
ol
e
c
ul
e
s
, vol
. 295, M
a
r
. 2025, doi
:
10.1016/
j
.i
j
bi
om
a
c
.2025.139654.
[
2]
G
.
L
i
e
t
al
.
,
“
C
l
oni
ng
a
nd
f
unc
t
i
ona
l
a
na
l
ys
i
s
of
nove
l
t
e
r
pe
ne
s
ynt
ha
s
e
ge
ne
s
f
r
om
A
qui
l
a
r
i
a
s
i
ne
ns
i
s
a
nd
i
ns
i
ght
s
i
nt
o
t
he
bi
os
ynt
he
s
i
s
of
s
e
s
qui
t
e
r
pe
ne
,”
I
ndus
t
r
i
al
C
r
op
s
and P
r
oduc
t
s
, vol
. 217, O
c
t
. 2
024, doi
:
10.1016/
j
.i
ndc
r
op.2024.118835.
[
3]
K
.
X
.
Q
i
n
e
t
a
l
.
,
“
T
w
o
u
np
r
e
c
e
de
n
t
e
d
5
,6,
7,
8
-
t
e
t
r
a
hy
dr
o
-
2
-
(2
-
ph
e
ny
l
e
t
hy
l
)
c
h
r
om
o
ne
s
w
i
t
h
a
nt
i
-
H
e
l
i
c
oba
c
t
e
r
py
l
o
r
i
a
c
t
i
v
i
t
y
f
r
o
m
a
ga
r
w
o
od
o
f
A
qu
i
l
a
r
i
a
s
i
ne
ns
i
s
‘
R
e
ke
1
’
,
”
P
hy
t
oc
h
e
m
i
s
t
r
y
L
e
t
t
e
r
s
,
vo
l
.
66
, p
p.
77
–
80
, A
pr
.
202
5,
do
i
:
1
0.1
01
6/
j
.p
hy
t
o
l
.
202
5.
01.
10
02.
[
4]
X
.
C
he
n,
Y
.
Y
a
ng,
Y
.
L
i
u,
C
.
S
ui
,
a
nd
J
.
W
e
i
,
“
A
n
a
l
ys
i
s
of
f
unga
l
c
om
po
s
i
t
i
on
i
n
di
f
f
e
r
e
nt
l
a
ye
r
s
of
B
a
nt
ou
a
ga
r
w
ood
-
f
or
m
i
ng
t
r
unk of
A
qui
l
a
r
i
a
s
i
ne
ns
i
s
r
e
ve
a
l
i
ng pr
e
s
e
nc
e
of
A
s
p
e
r
gi
l
l
us
-
i
nhi
bi
t
i
ng s
ubs
t
a
n
c
e
s
i
n a
g
a
r
w
ood s
i
t
e
s
,”
C
hi
ne
s
e
H
e
r
bal
M
e
di
c
i
ne
s
,
vol
. 17, no. 2, pp. 315
–
321, A
pr
. 2025, doi
:
10.1016/
j
.c
hm
e
d.2025.02.001.
[
5]
T
.
Y
.
D
u
e
t
al
.
,
“
T
he
f
unga
l
s
t
r
a
i
n
p
r
om
ot
e
s
r
a
pi
d
a
ga
r
w
ood
r
e
s
i
n
pr
oduc
t
i
on
w
i
t
h
m
e
di
c
i
na
l
l
y
a
c
c
e
pt
e
d
a
ga
r
ot
e
t
r
ol
l
e
ve
l
,”
F
ood
B
i
os
c
i
e
nc
e
, vol
. 62, D
e
c
. 2024, doi
:
10.1016/
j
.f
bi
o.2024.105535.
[
6]
S
.
M
a
,
T
.
Y
a
n,
Y
.
C
he
n,
a
nd
G
.
L
i
,
“
C
he
m
i
c
a
l
c
om
pos
i
t
i
on
a
nd
bi
oa
c
t
i
v
i
t
y
va
r
i
a
bi
l
i
t
y
o
f
t
w
o
-
s
t
e
p
e
xt
r
a
c
t
s
de
r
i
ve
d
f
r
om
t
r
a
di
t
i
ona
l
a
nd ‘
Q
i
N
a
n’
a
ga
r
w
ood (
A
qui
l
a
r
i
a
s
pp.)
,”
F
i
t
ot
e
r
api
a
, vol
. 176,
J
ul
.
2024, doi
:
10.1016/
j
.f
i
t
ot
e
.2024.106012.
[
7]
Z
.
W
u
a
nd
L
.
Y
u,
“
C
ha
r
a
c
t
e
r
i
s
t
i
c
qua
l
i
t
y
a
na
l
ys
i
s
f
or
bi
ol
ogi
c
a
l
l
y
i
nduc
e
d
a
ga
r
w
ood
c
ol
um
ns
i
n
A
qui
l
a
r
i
a
s
i
ne
ns
i
s
,”
E
nv
i
r
onm
e
nt
al
R
e
s
e
ar
c
h
, vol
. 235, O
c
t
. 2023, doi
:
10.1016/
j
.e
nvr
e
s
.2023.1166
33.
[
8]
J
.
D
ut
t
a
,
D
.
L
a
hon,
P
.
K
.
B
or
a
,
M
.
B
huya
n,
a
nd
S
.
H
a
l
da
r
,
“
A
hi
gh
pe
r
f
or
m
a
nc
e
t
hi
n
l
a
ye
r
c
hr
om
a
t
ogr
a
phy
(
H
P
T
L
C
)
m
e
t
hod
f
o
r
t
he
qua
l
i
t
y
a
s
s
e
s
s
m
e
nt
of
a
ga
r
w
ood
(
A
qui
l
a
r
i
a
m
a
l
a
c
c
e
ns
i
s
)
oi
l
f
r
om
N
or
t
he
a
s
t
I
ndi
a
,”
N
at
ur
al
P
r
oduc
t
R
e
s
e
ar
c
h
,
vol
.
38,
no.
9,
pp. 1634
–
1638, M
a
y 2024, doi
:
10.1080/
14786419.2022.2162516.
[
9]
Y
.
L
i
n
e
t
al
.
,
“
D
N
A
ba
r
c
odi
ng
i
de
nt
i
f
i
c
a
t
i
on
of
I
U
C
N
r
e
d
l
i
s
t
e
d
t
hr
e
a
t
e
ne
d
s
pe
c
i
e
s
i
n
t
he
ge
nus
A
qui
l
a
r
i
a
(
T
hym
e
l
a
e
a
c
e
a
e
)
us
i
ng
m
a
c
hi
ne
l
e
a
r
ni
ng a
ppr
oa
c
he
s
,”
P
hy
t
oc
he
m
i
s
t
r
y
L
e
t
t
e
r
s
, vol
. 55, pp. 105
–
111, J
un. 2023, doi
:
10.1016/
j
.phyt
ol
.2023.04.007.
[
10]
B
.
Z
ha
ng
e
t
al
.
,
“
T
he
e
s
t
a
bl
i
s
hm
e
nt
of
ha
m
gr
a
de
,
s
e
ns
or
y
s
c
or
e
s
a
nd
ke
y
f
l
a
vor
s
ubs
t
a
nc
e
s
pr
e
di
c
t
i
on
m
ode
l
s
f
or
J
i
nhua
ha
m
vi
a
e
-
nos
e
c
om
bi
ne
d w
i
t
h m
a
c
hi
ne
l
e
a
r
ni
ng,”
F
ood C
he
m
i
s
t
r
y
, vol
. 472, A
pr
. 2025
, doi
:
10.1016/
j
.f
oodc
he
m
.2025.142847.
[
11]
H
.
Y
a
ng,
Y
.
W
a
ng,
J
.
Z
ha
o,
P
.
L
i
,
L
.
L
i
,
a
nd
F
.
W
a
ng,
“
A
m
a
c
hi
ne
l
e
a
r
ni
ng
m
e
t
hod
f
or
j
ui
c
e
hum
a
n
s
e
ns
or
y
he
doni
c
p
r
e
di
c
t
i
on
us
i
ng e
l
e
c
t
r
oni
c
s
e
n
s
or
y f
e
a
t
ur
e
s
,”
C
u
r
r
e
nt
R
e
s
e
a
r
c
h i
n F
ood Sc
i
e
nc
e
, vol
. 7, 2
023, doi
:
10.1016/
j
.c
r
f
s
.2023.100576.
[
12]
K
.
H
i
r
i
m
bur
e
ga
m
a
a
nd
R
.
P
i
yum
a
l
,
“
I
n
vi
t
r
o
pr
oduc
t
i
on
of
i
ndus
t
r
i
a
l
v
a
l
ue
d
bi
oa
c
t
i
ve
s
e
c
onda
r
y
m
e
t
a
bol
i
t
e
s
f
r
om
s
e
l
e
c
t
e
d
m
e
di
c
i
na
l
pl
a
nt
s
of
S
r
i
L
a
nk
a
,”
i
n
B
i
ot
e
c
hnol
ogi
c
al
P
r
oduc
t
i
on
of
B
i
oac
t
i
v
e
P
hy
t
oc
he
m
i
c
al
s
of
M
e
di
c
i
nal
V
al
ue
:
a
C
om
pr
e
he
n
s
i
v
e
T
r
e
at
i
s
e
, E
l
s
e
vi
e
r
, 2024, pp. 67
–
102
,
doi
:
10.1016/
B
978
-
0
-
443
-
21818
-
7.00019
-
8.
[
13]
A
.
B
.
P
e
r
ovi
ć
e
t
al
.
,
“
N
ove
l
hydr
odi
s
t
i
l
l
a
t
i
on
a
nd
s
t
e
a
m
di
s
t
i
l
l
a
t
i
on
m
e
t
hods
of
e
s
s
e
nt
i
a
l
oi
l
r
e
c
ove
r
y
f
r
om
l
a
ve
nde
r
:
A
c
om
pr
e
he
ns
i
ve
r
e
vi
e
w
,”
I
ndus
t
r
i
al
C
r
ops
and P
r
odu
c
t
s
, vol
. 211, M
a
y 2024, d
oi
:
10.1016/
j
.i
ndc
r
op.2024.118244.
[
14]
A
.
F
.
M
.
A
m
i
d
on
, N
.
Z
.
M
a
ha
b
ob,
S
.
M
. H
.
M
.
H
uz
i
r
,
Z
.
M
.
Y
us
o
f
f
,
N
. I
s
m
a
i
l
,
a
nd
M
.
N
. T
a
i
b
,
“
T
he
gr
a
di
ng
of
a
ga
r
w
o
od
o
i
l
qua
l
i
t
y
ba
s
e
d
on m
ul
t
i
c
l
a
s
s
s
u
pp
or
t
v
e
c
t
or
m
a
c
hi
ne
(
M
S
V
M
)
m
od
e
l
,
”
i
n
20
22
I
E
E
E
18
t
h
I
n
t
e
r
n
at
i
on
al
C
ol
l
o
qui
um
on S
i
g
na
l
P
r
oc
e
s
s
i
ng
an
d
A
p
pl
i
c
a
t
i
ons
,
C
S
P
A
20
22
,
M
a
y
2
022
,
pp.
9
1
–
95
,
do
i
:
1
0.1
10
9/
C
S
P
A
55
07
6.2
02
2.
978
202
1.
[
15]
M
.
Z
.
H
.
C
how
dhur
y,
“
E
va
l
ua
t
i
on
of
a
ga
r
w
ood
oi
l
a
ut
he
nt
i
c
i
t
y
w
i
t
h
s
m
a
r
t
phone
-
ba
s
e
d
ha
ndhe
l
d
ne
a
r
-
i
nf
r
a
r
e
d
s
pe
c
t
r
om
e
t
e
r
,”
M
i
c
r
oc
he
m
i
c
al
J
our
nal
, vol
. 209, F
e
b. 2025, doi
:
10.1016/
j
.m
i
c
r
oc
.2025.11279
2.
[
16]
S
. L
i
e
t
al
.
, “
C
he
m
i
c
a
l
c
on
s
t
i
t
ue
nt
s
, bi
ol
ogi
c
a
l
a
c
t
i
vi
t
i
e
s
, a
nd qu
a
l
i
t
y e
va
l
ua
t
i
on
of
a
ga
r
w
ood pr
oduc
e
d f
r
om
t
he
qi
-
na
n ge
r
m
pl
a
s
m
of
A
qui
l
a
r
i
a
s
i
ne
ns
i
s
,”
P
hy
t
oc
he
m
i
s
t
r
y
L
e
t
t
e
r
s
, vol
. 65, pp. 68
–
76, F
e
b. 2025, d
oi
:
10.1016/
j
.phyt
ol
.2024.12.005.
[
17]
Y
.
X
i
e
e
t
al
.
,
“
U
nc
ove
r
i
ng
t
he
s
e
c
r
e
t
s
of
a
ga
r
w
ood
a
r
om
a
a
c
c
or
di
ng
t
o
r
e
gi
ons
a
nd
gr
a
de
s
us
i
ng
a
c
om
pr
e
he
ns
i
ve
a
n
a
l
yt
i
c
a
l
s
t
r
a
t
e
gy,”
C
he
m
i
c
al
C
o
m
m
uni
c
at
i
ons
, vol
. 60, no. 27, pp. 3709
–
3712, 2024, do
i
:
10.1039/
d3c
c
06164g.
[
18]
S
.
N
ga
di
r
a
n
e
t
al
.
,
“
T
he
i
nduc
t
i
on
t
e
c
hni
que
s
of
r
e
s
i
nous
a
g
a
r
w
ood
f
or
m
a
t
i
on:
a
r
e
vi
e
w
,”
B
i
or
e
s
our
c
e
T
e
c
hnol
ogy
R
e
por
t
s
,
vol
. 21, F
e
b. 2023, doi
:
10.1016/
j
.bi
t
e
b.2023.101337.
[
19]
N
.
A
.
A
.
R
a
hm
a
n
e
t
al
.
,
“
I
m
pa
c
t
of
na
t
ur
a
l
l
e
m
ongr
a
s
s
a
nd
a
ga
r
w
ood
e
s
s
e
nt
i
a
l
oi
l
di
f
f
us
i
on
on
i
ndoor
a
i
r
bor
ne
pol
l
ut
a
nt
s
:
a
c
a
s
e
s
t
udy of
of
f
i
c
e
e
nvi
r
onm
e
nt
s
,”
B
ui
l
di
ng and E
nv
i
r
onm
e
nt
, vol
. 270, F
e
b. 2025,
doi
:
10.1016/
j
.bui
l
de
nv.2024.112514.
[
20]
R
. G
ogoi
e
t
al
.
,
“
A
ga
r
w
ood (
A
qui
l
ar
i
a m
al
ac
c
e
n
s
i
s
L
.
)
a
qua
l
i
t
y f
r
a
gr
a
nt
a
nd
m
e
di
c
i
na
l
l
y s
i
gni
f
i
c
a
nt
pl
a
nt
ba
s
e
d e
s
s
e
nt
i
a
l
oi
l
w
i
t
h
pha
r
m
a
c
ol
ogi
c
a
l
pot
e
nt
i
a
l
s
a
nd
g
e
not
oxi
c
i
t
y,”
I
ndus
t
r
i
al
C
r
ops
a
nd
P
r
oduc
t
s
,
vol
.
197,
J
ul
.
2023,
doi
:
10.1016/
j
.i
ndc
r
op.2023.116535.
[
21]
A
.
H
.
N
our
,
R
.
H
.
M
od
a
t
he
r
,
R
.
M
.
Y
unus
,
A
.
A
.
M
.
E
l
nour
,
a
nd
N
.
A
.
I
s
m
a
i
l
,
“
C
ha
r
a
c
t
e
r
i
z
a
t
i
on
of
bi
oa
c
t
i
ve
c
om
pounds
i
n
pa
t
c
houl
i
oi
l
us
i
ng
m
i
c
r
ow
a
ve
-
a
s
s
i
s
t
e
d
a
nd
t
r
a
di
t
i
ona
l
hydr
odi
s
t
i
l
l
a
t
i
on
m
e
t
hods
,”
I
ndus
t
r
i
al
C
r
ops
and
P
r
odu
c
t
s
,
vol
.
208,
F
e
b
.
2024, doi
:
10.1016/
j
.i
ndc
r
op.2023.117901.
[
22]
F
.
Z
.
H
os
s
e
i
ni
e
t
al
.
,
“
A
r
om
a
t
he
r
a
py
w
i
t
h
R
os
a
da
m
a
s
c
e
na
m
i
l
l
.
t
o
r
e
l
i
e
ve
t
he
s
ym
pt
om
s
of
po
s
t
pa
r
t
um
de
pr
e
s
s
i
on
a
nd
s
l
e
e
p
qua
l
i
t
y
i
n
pr
i
m
i
pa
r
ous
w
om
e
n:
a
r
a
ndom
i
s
e
d
c
ont
r
ol
l
e
d
t
r
i
a
l
,”
J
our
nal
of
H
e
r
bal
M
e
di
c
i
ne
,
vol
.
46,
A
ug.
2024,
doi
:
10.1016/
j
.he
r
m
e
d.2024.100912.
[
23]
W
.
M
.
A
s
hr
a
f
a
nd
V
.
D
ua
,
“
P
a
r
t
i
a
l
de
r
i
va
t
i
ve
-
ba
s
e
d
dyna
m
i
c
s
e
ns
i
t
i
vi
t
y
a
na
l
ys
i
s
e
xpr
e
s
s
i
on
f
or
non
-
l
i
ne
a
r
a
ut
o
r
e
g
r
e
s
s
i
ve
w
i
t
h
e
xoge
nous
(
N
A
R
X
)
m
ode
l
–
c
a
s
e
s
t
udi
e
s
on
di
s
t
i
l
l
a
t
i
on
c
ol
um
ns
a
nd
m
ode
l
’
s
i
nt
e
r
pr
e
t
a
t
i
on
i
nve
s
t
i
ga
t
i
on,”
C
he
m
i
c
al
E
ngi
ne
e
r
i
ng
J
our
nal
A
dv
anc
e
s
, vol
. 18,
M
a
y 2024, doi
:
10.1016/
j
.c
e
j
a
.2024.100605.
[
24]
S
.
M
.
A
hm
a
d
a
nd
A
.
T
a
nve
e
r
,
“
N
A
R
X
m
ode
l
i
ng
a
nd
s
i
m
ul
a
t
i
on
o
f
he
a
ve
dyna
m
i
c
s
w
i
t
h
a
ppl
i
c
a
t
i
on
of
r
obus
t
c
ont
r
ol
of
a
n
unde
r
a
c
t
ua
t
e
d unde
r
w
a
t
e
r
ve
hi
c
l
e
,”
O
c
e
an E
ngi
ne
e
r
i
ng
, vol
. 325, M
a
y 2025, d
oi
:
10.1016/
j
.oc
e
a
ne
ng.2025.120790.
[
25]
H
.
V
u
a
nd
D
.
C
ha
ng,
“
O
pt
i
m
i
s
i
ng
c
om
put
a
t
i
ona
l
e
f
f
i
c
i
e
nc
y
i
n
dyn
a
m
i
c
m
ode
l
l
i
ng
of
pr
ot
on
e
xc
ha
nge
m
e
m
br
a
ne
f
ue
l
c
e
l
l
pow
e
r
s
ys
t
e
m
s
u
s
i
ng N
A
R
X
n
e
t
w
or
k,”
E
ne
r
gy
C
onv
e
r
s
i
on and M
anage
m
e
nt
:
X
, vol
. 26, A
pr
. 2025, doi
:
10.1016/
j
.e
c
m
x.2025.100908.
[
26]
L
.
U
.
K
ha
s
a
na
h,
S
.
A
r
i
vi
a
ni
,
E
.
P
ur
w
a
nt
o,
a
nd
D
.
P
r
a
s
e
pt
i
a
ngga
,
“
C
he
m
i
c
a
l
c
om
pos
i
t
i
on
a
nd
c
i
t
r
a
l
c
ont
e
nt
of
e
s
s
e
nt
i
a
l
oi
l
of
l
e
m
ongr
a
s
s
(
C
ym
bopogon c
i
t
r
a
t
us
(
D
C
.)
S
t
a
pf
)
l
e
a
f
w
a
s
t
e
pr
e
pa
r
e
d w
i
t
h va
r
i
ou
s
pr
oduc
t
i
on m
e
t
hods
,”
J
ou
r
nal
of
A
gr
i
c
ul
t
ur
e
and
F
ood R
e
s
e
ar
c
h
, vol
. 19, M
a
r
. 2025, doi
:
10.1016/
j
.j
a
f
r
.2024.101570.
[
27]
J
. L
i
e
t
al
.
, “
C
h
e
m
i
c
a
l
a
na
l
y
s
i
s
of
di
f
f
e
r
e
nt
pa
r
t
s
f
r
om
a
ga
r
w
ood c
ol
um
ns
by a
r
t
i
f
i
c
i
a
l
l
y a
ga
r
w
ood
-
i
nduc
i
ng m
e
t
hod ba
s
e
d on
G
C
–
M
S
a
nd U
P
L
C
-
T
O
F
-
M
S
,”
F
i
t
ot
e
r
api
a
, vol
. 178, O
c
t
. 2024, doi
:
10.1016/
j
.f
i
t
ot
e
.2024.106156.
[
28]
A
.
F
.
M
.
A
m
i
don,
Z
.
M
.
Y
us
of
f
,
N
.
I
s
m
a
i
l
,
A
.
A
.
A
l
m
i
s
r
e
b,
a
nd
M
.
N
.
T
a
i
b,
“
K
N
N
E
uc
l
i
de
a
n
di
s
t
a
nc
e
m
ode
l
pe
r
f
or
m
a
nc
e
on
A
qui
l
ar
i
a
m
al
ac
c
e
ns
i
s
oi
l
qu
a
l
i
t
i
e
s
,”
J
our
nal
of
A
dv
anc
e
d
R
e
s
e
ar
c
h
i
n
A
ppl
i
e
d
Sc
i
e
nc
e
s
and
E
ngi
ne
e
r
i
ng
T
e
c
hnol
ogy
,
vol
.
48,
no. 2, pp. 16
–
28, J
ul
. 2025, doi
:
10.37934/
a
r
a
s
e
t
.48.2.1628.
[
29]
M
.
J
.
A
.
A
.
R
a
j
a
e
t
al
.
,
“
D
e
s
i
gn
of
i
nt
e
l
l
i
ge
nt
B
a
ye
s
i
a
n
r
e
gul
a
r
i
z
e
d
de
e
p
c
a
s
c
a
d
e
d
N
A
R
X
ne
ur
os
t
r
uc
t
ur
e
f
or
pr
e
di
c
t
i
ve
a
na
l
ys
i
s
of
F
i
t
z
H
ugh
-
N
a
gum
o
bi
oe
l
e
c
t
r
i
c
a
l
m
ode
l
i
n
ne
ur
ona
l
c
e
l
l
m
e
m
br
a
ne
,”
B
i
om
e
di
c
al
Si
gnal
P
r
oc
e
s
s
i
ng
and
C
ont
r
ol
,
vol
.
101,
M
a
r
.
2025, doi
:
10.1016/
j
.bs
pc
.2024.107192.
[
30]
S
.
S
.
Z
a
r
e
e
n,
G
.
S
un,
M
.
K
undi
,
S
.
F
.
Q
a
dr
i
,
a
nd
S
.
Q
a
dr
i
,
“
E
nha
nc
i
ng
s
ki
n
c
a
nc
e
r
di
a
gnos
i
s
w
i
t
h
de
e
p
l
e
a
r
ni
ng:
a
hybr
i
d
C
N
N
-
R
N
N
a
ppr
oa
c
h,”
C
om
put
e
r
s
, M
at
e
r
i
al
s
and C
ont
i
nua
, vol
. 79, no. 1, pp. 1497
–
1519, 2024, doi
:
10.32604/
c
m
c
.2024.047418.
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
O
pt
imi
z
in
g nonli
ne
ar
aut
or
e
g
r
e
s
s
iv
e
w
it
h e
x
oge
nous
i
nput
s
ne
t
w
or
k
…
(
M
uhamm
ad I
k
hs
an R
os
la
n
)
3501
[
31]
F
. Z
ha
ng, J
. Y
i
n, N
. W
u,
X
. H
u,
S
. S
un, a
nd
Y
.
W
a
ng, “
A
dua
l
-
pa
t
h m
ode
l
m
e
r
gi
ng C
N
N
a
nd
R
N
N
w
i
t
h a
t
t
e
nt
i
on
m
e
c
ha
ni
s
m
f
or
c
r
op c
l
a
s
s
i
f
i
c
a
t
i
on,”
E
ur
ope
an J
ou
r
nal
of
A
gr
onom
y
, vol
. 159, S
e
p. 2024, doi
:
10.1016/
j
.e
j
a
.2024.127273.
[
32]
Z
.
M
.
Y
us
of
f
a
nd
N
.
I
s
m
a
i
l
,
“
D
a
t
a
s
e
t
s
of
c
he
m
i
c
a
l
c
om
pounds
i
n
t
hr
e
e
di
f
f
e
r
e
nt
s
pe
c
i
e
s
of
a
qui
l
a
r
i
a
us
i
ng
G
C
-
M
S
c
oupl
e
d
w
i
t
h
GC
-
F
I
D
a
na
l
ys
i
s
,”
D
at
a i
n B
r
i
e
f
, vol
. 53, A
pr
. 2024, doi
:
10.1016/
j
.di
b.2024.110209.
[
33]
S
. M
. H
. M
.
H
uz
i
r
, S
. N
. T
a
j
uddi
n, Z
. M
. Y
u
s
of
f
, N
. I
s
m
a
i
l
, A
. A
.
A
l
m
i
s
r
e
b, a
n
d M
. N
. T
a
i
b, “
A
g
a
r
w
ood oi
l
qua
l
i
t
y i
de
nt
i
f
i
c
a
t
i
on
us
i
ng
a
r
t
i
f
i
c
i
a
l
ne
ur
a
l
ne
t
w
or
k
m
ode
l
l
i
ng
f
o
r
f
i
ve
gr
a
de
s
,”
I
nt
e
r
nat
i
onal
J
our
nal
of
E
l
e
c
t
r
i
c
al
and
C
om
put
e
r
E
ngi
ne
e
r
i
ng
,
vol
.
14,
no. 2, pp. 2254
–
2261, A
pr
. 2024, doi
:
10.11591/
i
j
e
c
e
.v14i
2.pp2254
-
2261.
[
34]
S
.
M
.
H
.
M
.
H
uz
i
r
e
t
al
.
,
“
P
r
e
-
p
r
oc
e
s
s
i
ng
t
e
c
hni
que
of
A
qui
l
a
r
i
a
s
pe
c
i
e
s
f
r
o
m
M
a
l
a
ys
i
a
f
or
f
ou
r
di
f
f
e
r
e
nt
qua
l
i
t
i
e
s
,”
B
ul
l
e
t
i
n
o
f
E
l
e
c
t
r
i
c
al
E
ngi
ne
e
r
i
ng and I
nf
or
m
at
i
c
s
, vol
. 13, no. 1, pp. 152
–
159, F
e
b. 2024,
doi
:
10.11591/
e
e
i
.v13i
1.5577.
[
35]
H
.
H
.
A
.
A
boul
E
l
a
,
“
A
nove
l
hybr
i
d
A
N
F
I
S
-
N
A
R
X
a
nd
N
A
R
X
-
ANN
m
ode
l
s
t
o
pr
e
di
c
t
t
he
pr
of
i
t
a
bi
l
i
t
y
of
E
gypt
i
a
n
i
ns
ur
a
nc
e
c
om
pa
ni
e
s
,”
St
at
i
s
t
i
c
s
,
O
pt
i
m
i
z
at
i
on
and
I
nf
or
m
at
i
on
C
om
put
i
ng
,
vol
.
12,
no.
6,
pp.
1934
–
1955,
A
ug.
2024,
doi
:
10.19139/
s
oi
c
-
2310
-
5070
-
2104.
[
36]
W
.
S
hi
,
X
.
W
a
n,
F
.
Z
ha
o,
a
nd
R
.
D
e
ng,
“
A
dua
l
-
m
ode
l
f
r
a
m
e
w
or
k
c
om
bi
ni
n
g
nonl
i
ne
a
r
a
ut
or
e
gr
e
s
s
i
ve
w
i
t
h
e
xog
e
nous
i
nput
s
(
N
A
R
X
)
a
nd
L
S
T
M
ne
t
w
or
ks
f
o
r
e
nha
nc
e
d
da
i
l
y r
unof
f
pr
e
di
c
t
i
on
a
nd
e
r
r
o
r
c
or
r
e
c
t
i
on,”
E
nv
i
r
onm
e
nt
al
M
ode
l
l
i
ng
and
So
f
t
w
ar
e
,
vol
. 192, 2025, doi
:
10.1016/
j
.e
nvs
of
t
.2025.106570.
[3
7
]
Y
. S
ha
o, J
. Z
ha
o, J
. X
u, A
.
F
u, a
nd
M
.
L
i
, “
A
ppl
i
c
a
t
i
on
of
r
a
i
nf
a
l
l
-
r
unof
f
s
i
m
ul
a
t
i
on ba
s
e
d
o
n t
he
N
A
R
X
dyn
a
m
i
c
ne
ur
a
l
n
e
t
w
or
k
m
ode
l
,”
W
at
e
r
, vol
. 14, no. 13, J
un. 2022, doi
:
10.3390/
w
14132082.
[3
8
]
W
.
r
ui
Z
ha
ng
e
t
al
.
,
“
F
or
e
c
a
s
t
i
ng
gr
oundw
a
t
e
r
l
e
ve
l
of
ka
r
s
t
a
qui
f
e
r
i
n
a
l
a
r
ge
m
i
ni
ng
a
r
e
a
us
i
ng
pa
r
t
i
a
l
m
ut
ua
l
i
nf
or
m
a
t
i
on
a
nd
N
A
R
X
hybr
i
d m
ode
l
,”
E
nv
i
r
onm
e
nt
al
R
e
s
e
ar
c
h
, vol
. 213, O
c
t
. 2022, doi
:
10.1
016/
j
.e
nvr
e
s
.2022.113747.
[3
9
]
Y
.
M
i
ky,
M
.
R
.
K
a
l
oop,
M
.
T
.
E
l
na
bw
y,
A
.
B
a
i
k,
a
nd
A
.
A
l
s
houny,
“
A
r
e
c
ur
r
e
nt
-
c
a
s
c
a
de
-
ne
ur
a
l
ne
t
w
or
k
-
nonl
i
ne
a
r
a
ut
or
e
gr
e
s
s
i
ve
ne
t
w
or
k
s
w
i
t
h
e
xog
e
nous
i
np
ut
s
(
N
A
R
X
)
a
ppr
oa
c
h
f
or
l
ong
-
t
e
r
m
t
i
m
e
-
s
e
r
i
e
s
pr
e
di
c
t
i
on
of
w
a
v
e
he
i
ght
b
a
s
e
d
on
w
a
ve
c
ha
r
a
c
t
e
r
i
s
t
i
c
s
m
e
a
s
ur
e
m
e
nt
s
,”
O
c
e
an E
ngi
ne
e
r
i
ng
, vol
. 240, N
ov. 2021, doi
:
10.1016/
j
.oc
e
a
ne
ng.2021.109958.
[
40
]
D
.
M
a
n,
Y
.
F
u,
Y
.
L
i
,
a
nd
M
.
Q
i
a
o,
“
A
n
a
l
ys
i
s
of
t
he
qua
l
i
t
y
a
nd
c
ha
r
a
c
t
e
r
i
s
t
i
c
c
he
m
i
c
a
l
c
om
pone
nt
s
of
a
g
a
r
w
ood
ba
s
e
d
on
t
h
e
dr
i
l
l
i
ng m
e
t
hod f
or
Q
i
-
N
a
n a
ga
r
w
ood,”
I
ndus
t
r
i
al
C
r
ops
and P
r
odu
c
t
s
, vol
. 22
2, D
e
c
. 2024, doi
:
10.1016/
j
.i
ndc
r
op.2024.119841.
[4
1
]
X
.
Z
ha
ng,
Y
.
X
i
a
,
C
.
L
i
,
L
.
Y
a
ng,
a
nd
D
.
P
.
M
a
ndi
c
,
“
A
f
ul
l
s
e
c
ond
-
or
de
r
s
t
a
t
i
s
t
i
c
a
l
a
na
l
ys
i
s
of
s
t
r
i
c
t
l
y
l
i
ne
a
r
a
nd
w
i
de
l
y
l
i
ne
a
r
e
s
t
i
m
a
t
or
s
w
i
t
h M
S
E
a
nd G
a
us
s
i
a
n e
nt
r
opy c
r
i
t
e
r
i
a
,”
Si
gnal
P
r
oc
e
s
s
i
ng
, vol
. 19
2, M
a
r
. 2022, doi
:
10.1016/
j
.s
i
gpr
o.2021.108403.
[
42]
K
.
G
ue
dr
i
,
R
.
Z
a
r
i
n,
B
.
M
.
M
a
khdoum
,
S
.
A
l
j
a
r
boa
,
a
nd
H
.
A
.
E
.
W
.
K
ha
l
i
f
a
,
“
A
dva
nc
e
d
c
om
put
a
t
i
ona
l
m
ode
l
i
ng
of
m
a
l
a
r
i
a
us
i
ng
a
r
t
i
f
i
c
i
a
l
ne
ur
a
l
ne
t
w
or
ks
a
nd
s
t
oc
ha
s
t
i
c
s
ol
v
e
r
s
,”
K
no
w
l
e
dge
-
B
as
e
d
Sy
s
t
e
m
s
,
vol
.
325,
2025,
doi
:
10.1016/
j
.knos
ys
.2025.113976.
B
I
O
G
R
A
P
H
I
E
S
O
F
A
U
T
H
O
R
S
Muhammad
Ikhsa
n
Roslan
earned
his
Master
of
Science
in
Elect
ronic
Systems
Design
Engineering
from
Universiti
Sains
Malaysia
(USM),
Penang,
Malaysia,
in
2022
with
first
-
class
honors.
He
is
currently
a
server
validation
engineer
specializing
in
IP
-
level
validation
at
AMD
Global
Services,
whil
e
also
pursuing
full
-
time
po
stgraduate
studies
at
the
Faculty
of
Electr
ical
Engine
ering,
Univer
siti
Teknol
ogi
MARA
(UiTM)
,
Shah
Alam,
Malaysia.
With
a
strong
passion
for
researc
h
in
engineering,
p
articularly
in
artificial
intelligence
,
he
combines
ac
ademic
excellen
ce
with
practica
l
ex
perienc
e,
showcasing
a
dedicated
commitment
to
advancing
the
field.
He
can
be
contacted
at
email:
muhammadikhsa
nroslan@gmail.com
.
Noor
Aida
Syakira
Ahmad
Sabri
received
her
Bachelor
of
Engin
eering
(Hons)
in
Electronic
Engineering
from
Universiti
Teknologi
MARA
(UiTM),
Shah
Alam,
Malaysia,
in
2022.
Currently,
she
is
pursuing
postgraduate
studies
at
the
Faculty
of
Electrica
l
Engineering,
Universiti
Teknologi
MARA
(UiTM),
Shah
Alam,
Malaysia.
Her
research
interests
focus
on
advance
d
signal
processing
and
machine
learning.
Sh
e
can
be
contacted
at
email:
aidasyaki
raaa01@
gmail.co
m
.
Nur
Athirah
Syafiqah
Noramli
received
her
B.Sc.
(Hons)
in
Co
mputer
Science
from
Universiti
Teknologi
MARA
(UiTM)
Cawangan
Melaka
Kampu
s
Jasin.
She
is
currently
pursuing
her
studies
as
a
postgraduate
student
at
the
Faculty
of
El
ectrical
Engineering,
at
Universiti
Teknologi
MARA
(UiTM)
Shah
Alam,
Selangor,
Malaysia.
Her
research
interests
include
advance
d
signal
processing
,
machine
learning,
and
deep
learning.
Sh
e
can
be
contacted
at email
:
athirah.n
oramli1
@
gmail.co
m
.
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
.
14
, N
o.
5
,
O
c
to
be
r
2025
:
3493
-
3502
3502
Assoc.
Prof.
Ir.
Ts.
Dr.
Nurlaila
Ismail
received
her
Ph
.
D
.
in
Electrical
Engineering
from
Universiti
Teknologi
MARA,
Malaysia.
She
is
curr
ently
a
senior
lecturer
at
Faculty
of
Electr
ical
Engine
ering,
Univer
siti
Teknol
ogi
MARA
Shah
Alam,
Malays
ia.
Her
research
interests
include
advanced
signal
processing
and
artificial
i
ntelligence.
Sh
e
can
be
contacted
at email
:
nurlaila0583@
uitm.edu.my
.
Assoc.
Prof.
Ts.
Dr.
Zakiah
Mohd
Yusoff
received
her
bachel
or’s
degree
in
Electrical
Engineering
and
Ph
.
D
.
in
Electrical
Engineering
from
Univ
ersiti
Teknologi
MARA
Shah
Alam,
in
2009
and
2014,
respectively.
She
is
a
senior
lecturer
who
is
currently
working
at
Faculty
of
Electrical
Engineering,
Universiti
Teknologi
MAR
A
(UiTM)
Shah
Alam
,
Malaysia. In Ma
y 2014,
she joined
Universiti Teknologi
MARA as
a t
eaching staff
. Her major
interests
include
process
control,
system
identifica
tion,
and
essential
oil
extractio
n
systems.
Sh
e can be con
tacted at
email:
zakiah9018
@
uitm.ed
u.my
.
Assoc.
Prof.
Ali
Abd
Almisreb
received
a
master’s
degree
in
Co
mputer
Science
and
doctorate doctor’s degre
e
in Electric
al Enginee
ring/Computer
Engineerin
g from Unive
rsiti
Teknologi
MARA,
Malaysia.
He
is
currently
an
Associate
Profe
ssor
at
the
Faculty
of
Computer
Science
s
and
Enginee
ring,
Directo
r
of
Gradua
te
Council
and
Editor
in
Chief
a
t
International
University
of
Sarajevo.
His
major
interests
include
deep
learning,
machine
learning,
computer
vision
voice
recognition
,
and
quantum
computing.
He
can
be
contacted
at
email:
alimes96
@
yahoo.com
.
Prof.
Dr.
Saiful
Nizam
Tajuddin
received
his
Ph
.
D
.
from
Universiti
Malaysia
Pahang
(UMP)
. He i
s an A
ssocia
te Prof
essor
and
Director
of Bio
A
romatic Res
earch Center of
Excellence
(BARCE)
at
Universiti
Malaysia
Pahang.
He
is
a
director
and
researcher
at
Synbion
Sdn
Bhd,
Kuanta
n,
Pahang
,
Malays
ia.
He
has
been
a
very
ac
tive
resea
rche
r
and
ov
er
the
years
had
author
and/or
co
-
author
many
papers
published
in
refereed
journals
and
conferences.
He can be contacted at email:
saifulnizam@ump.edu.my
.
Prof.
Ir.
Ts.
Dr.
Haji
Mohd
Nasir
Taib
received
the
degre
e
in
Electrical
Engineering
from
the
University
of
Tasmania,
Hobart,
Australia,
the
M.Sc.
degree
in
Control
Engineering
from
Sheffield
University,
UK,
and
the
Ph.D.
degree
in
i
nstrumentation
from
the
University
of
Manchester
Institute
of
Science
and
Technology,
UK.
He
is
currently
an
Honorary
Professor
at
Universiti
Teknologi
MARA
(UiTM),
M
alaysia.
He
Heads
the
Advanced
Signal
Processing
Research
Group
at
the
Faculty
of
Electri
cal
Engineering,
UiTM.
He
has
been
a
very
a
ctive
researcher
and
ove
r
the
years
had
author
and/or
co
-
author
many
papers
published
in
refereed
journals
and
conferences.
He
can
be
contacted
at
email
:
dr.nasir@
uitm.edu.my
.
Evaluation Warning : The document was created with Spire.PDF for Python.