I
AE
S
I
n
t
e
r
n
at
ion
al
Jou
r
n
al
of
Ar
t
if
icial
I
n
t
e
ll
ig
e
n
c
e
(
I
J
-
AI
)
Vol.
14,
No.
4:
Augus
t
2025
,
pp.
2889
~
2898
I
S
S
N:
2252
-
8938
,
DO
I
:
10
.
11591/i
jai
.
v14.
i4
.
pp
28
89
-
2898
2889
Jou
r
n
al
h
omepage
:
ht
tp:
//
ij
ai
.
iaes
c
or
e
.
c
om
A
p
p
li
c
a
t
io
n
of
s
e
lf
-
o
r
gan
iz
in
g
m
ap
f
o
r
m
o
d
e
li
n
g t
h
e
A
q
u
il
ari
a
mal
ac
c
e
n
si
s
oi
l
u
si
n
g c
h
e
m
ic
a
l
c
om
p
o
u
n
d
M
oh
am
m
ad
Ar
i
f
F
ah
m
i
Che
Has
s
an
,
Z
ak
ia
h
M
oh
d
Yus
of
f
,
Nurlail
a
I
s
m
ail
,
M
oh
d
Nas
ir
T
aib
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
Ar
t
icle
I
n
f
o
AB
S
T
RA
CT
A
r
ti
c
le
h
is
tor
y
:
R
e
c
e
ived
Apr
8,
2024
R
e
vis
e
d
M
a
r
25,
2025
Ac
c
e
pted
J
un
8,
2025
A
g
ar
w
o
o
d
o
i
l
,
k
n
o
w
n
as
‘
b
l
ac
k
g
o
l
d
’
o
r
t
h
e
‘
w
o
o
d
o
f
G
o
d
,
’
i
s
a
g
l
o
b
a
l
l
y
p
ri
ze
d
es
s
en
t
i
a
l
o
i
l
d
eri
v
ed
n
a
t
u
ra
l
l
y
fro
m
t
h
e
A
q
u
i
l
a
r
i
a
t
ree.
D
e
s
p
i
t
e
i
t
s
s
i
g
n
i
f
i
ca
n
ce,
t
h
e
cu
rren
t
n
o
n
-
s
t
an
d
ard
i
zed
g
ra
d
i
n
g
s
y
s
t
e
m
v
ari
es
w
o
r
l
d
w
i
d
e,
rel
y
i
n
g
o
n
s
u
b
j
ec
t
i
v
e
as
s
es
s
men
t
s
.
T
h
i
s
p
a
p
er
ad
d
re
s
s
es
t
h
e
n
eed
f
o
r
a
co
n
s
i
s
t
e
n
t
c
l
as
s
i
f
i
cat
i
o
n
mo
d
el
b
y
p
res
e
n
t
i
n
g
an
o
v
erv
i
e
w
o
f
A
q
u
i
l
a
r
i
a
m
a
l
a
cce
n
s
i
s
o
i
l
q
u
a
l
i
t
y
u
s
i
n
g
t
h
e
s
e
l
f
-
o
r
g
an
i
zi
n
g
ma
p
(SO
M)
al
g
o
r
i
t
h
m.
D
eri
v
ed
fro
m
t
h
e
T
h
y
mel
aeacea
e
fami
l
y
,
A
q
u
i
l
a
r
i
a
m
a
l
a
ccen
s
i
s
i
s
a
p
ri
mar
y
s
o
u
rce
o
f
ag
arw
o
o
d
t
ree
s
i
n
t
h
e
Mal
a
y
A
rch
i
p
e
l
ag
o
.
A
g
arw
o
o
d
o
i
l
ex
t
rac
t
i
o
n
i
n
v
o
l
v
e
s
t
rad
i
t
i
o
n
al
me
t
h
o
d
s
l
i
k
e
s
o
l
v
e
n
t
e
x
t
rac
t
i
o
n
a
n
d
h
y
d
r
o
-
d
i
s
t
i
l
l
a
t
i
o
n
,
y
i
e
l
d
i
n
g
a
co
m
p
l
e
x
mi
x
t
u
re
o
f
c
h
ro
m
o
n
e
d
eri
v
at
i
v
e
s
,
o
x
y
g
e
n
at
e
d
s
es
q
u
i
t
er
p
en
e
s
,
an
d
s
e
s
q
u
i
t
erp
e
n
e
h
y
d
ro
car
b
o
n
s
.
T
h
i
s
s
t
u
d
y
cat
eg
o
ri
ze
s
ag
arw
o
o
d
o
i
l
i
n
t
o
h
i
g
h
an
d
l
o
w
g
ra
d
es
b
a
s
ed
o
n
ch
emi
ca
l
co
mp
o
u
n
d
s
,
u
t
i
l
i
zi
n
g
t
h
e
SO
M
a
l
g
o
ri
t
h
m
w
i
t
h
i
n
p
u
t
s
o
f
t
h
ree
s
p
eci
f
i
c
co
m
p
o
u
n
d
s
:
β
-
ag
aro
f
u
ran
,
α
-
ag
ar
o
fu
ra
n
,
an
d
1
0
-
ep
i
-
φ
-
e
u
d
e
s
mo
l
.
Fi
n
d
i
n
g
s
d
em
o
n
s
t
ra
t
e
t
h
e
eff
i
cacy
o
f
S
O
M
-
b
a
s
ed
q
u
al
i
t
y
g
ra
d
i
n
g
i
n
d
i
s
t
i
n
g
u
i
s
h
i
n
g
a
g
arw
o
o
d
o
i
l
gr
ad
e
s
,
o
fferi
n
g
a
s
i
g
n
i
fi
can
t
co
n
t
r
i
b
u
t
i
o
n
t
o
t
h
e
fi
el
d
.
T
h
e
n
o
n
-
s
t
an
d
ard
i
ze
d
g
rad
i
n
g
s
y
s
t
em's
i
n
effi
c
i
en
c
y
an
d
s
u
b
j
ect
i
v
i
t
y
u
n
d
ers
c
o
r
e
t
h
e
n
eces
s
i
t
y
fo
r
a
s
t
a
n
d
ar
d
i
ze
d
mo
d
el
,
mak
i
n
g
t
h
i
s
res
earc
h
cru
c
i
al
f
o
r
t
h
e
ag
arw
o
o
d
i
n
d
u
s
t
r
y
's
ad
v
a
n
cemen
t
.
K
e
y
w
o
r
d
s
:
Aga
r
wood
oil
A
quil
ar
ia
s
pe
c
ies
Gr
a
ding
c
las
s
if
ica
ti
on
S
e
lf
-
or
ga
nizing
map
S
ys
tem
identif
ica
ti
on
Th
i
s
i
s
a
n
o
p
en
a
c
ces
s
a
r
t
i
c
l
e
u
n
d
e
r
t
h
e
CC
B
Y
-
SA
l
i
ce
n
s
e.
C
or
r
e
s
pon
din
g
A
u
th
or
:
Z
a
kiah
M
ohd
Yus
of
f
F
a
c
ult
y
of
E
lec
tr
ica
l
E
nginee
r
ing
,
Unive
r
s
it
i
T
e
kn
ologi
M
AR
A
40450
S
ha
h
Ala
m,
S
e
langor
,
M
a
lays
ia
E
mail:
z
a
kiah9018@uitm
.
e
du.
my
1.
I
NT
RODU
C
T
I
ON
Aga
r
wood
oil
,
a
ls
o
known
a
s
"
g
a
ha
r
u"
oi
l
in
M
a
l
a
ys
ia
a
nd
I
ndone
s
ia,
is
e
xtr
a
c
ted
f
r
o
m
a
ga
r
wood
tr
e
e
s
of
the
ge
nus
A
quil
ar
ia
m
alacc
e
ns
is
a
nd
the
T
hymela
e
a
c
e
a
e
f
a
mi
ly
[
1]
–
[
4]
.
T
he
f
or
mation
of
matur
e
d
a
ga
r
wood
r
e
s
ult
s
f
r
om
va
r
ious
f
a
c
tor
s
,
including
a
nim
a
l
g
r
a
z
ing,
ins
e
c
t
a
tt
a
c
ks
,
mi
c
r
obial
invas
i
o
ns
,
a
nd
li
ghtni
ng
s
tr
ikes
[
2]
,
[
5]
.
C
ons
e
que
ntl
y,
a
ga
r
woo
d
is
a
c
knowle
dge
d
a
s
r
e
s
in
-
im
pr
e
gna
ted
he
a
r
twood,
with
e
ve
r
y
pa
r
t
o
f
the
plant
s
e
r
ving
a
pu
r
pos
e
,
includin
g
the
t
r
e
e
tr
unks
,
br
a
nc
he
s
,
a
nd
a
ga
r
wood
s
tems
.
T
he
s
tem
c
a
n
unde
r
go
pr
oc
e
s
s
ing
to
yield
e
s
s
e
nti
a
l
oil
,
c
omm
only
r
e
f
e
r
r
e
d
to
a
s
a
ga
r
wood
oil
[
6]
.
I
n
c
ont
e
mpor
a
r
y
ti
mes
,
a
ga
r
wood
oil
holds
high
r
e
ga
r
d
f
or
it
s
a
p
pli
c
a
ti
ons
in
pe
r
f
umer
y,
a
s
a
s
ymbol
of
luxu
r
y,
medic
inal
us
e
s
,
a
nd
r
e
li
gious
r
i
tuals
,
lea
ding
to
a
s
tea
dy
incr
e
a
s
e
in
de
mand
[
1]
,
[
5]
,
[
7]
.
S
ome
ye
a
r
s
a
go,
a
ga
r
wood
oil
gr
a
ding
r
e
li
e
d
on
t
r
a
dit
ional
methods
ba
s
e
d
on
f
a
c
tor
s
s
uc
h
a
s
c
olor
a
nd
odo
r
[
8]
.
How
e
ve
r
,
us
ing
human
s
e
ns
or
y
pa
ne
ls
,
pa
r
ti
c
ular
ly
the
s
e
ns
e
of
s
mell,
f
or
gr
a
ding
a
ga
r
wood
oil
wa
s
c
ons
ider
e
d
inef
f
icie
nt
[
7]
.
T
his
a
ppr
oa
c
h
pr
e
s
e
nted
mor
e
dr
a
wba
c
ks
than
a
dva
nt
a
ge
s
,
including
a
high
leve
l
of
s
ubjec
ti
vit
y
a
nd
c
a
us
ing
f
a
ti
gue
a
mong
the
s
e
ns
or
y
pa
ne
l
due
to
the
r
e
pe
ti
ti
ve
a
nd
c
onti
nuous
na
tur
e
o
f
the
method
[
9
]
.
Ove
r
the
c
our
s
e
o
f
tec
hnologi
c
a
l
a
dva
nc
e
ments
,
the
g
r
a
ding
o
f
a
ga
r
wood
oil
ha
s
e
volved,
incor
por
a
ti
ng
moder
n
tec
h
niques
a
li
gne
d
with
c
ur
r
e
nt
de
ve
lopm
e
nts
.
I
ntelli
ge
nt
methods
,
s
uc
h
a
s
Z
-
s
c
or
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
Ar
ti
f
I
ntell
,
Vol.
14,
No.
4:
Augus
t
2025
:
288
9
-
2898
2890
a
na
lys
is
,
a
r
ti
f
icia
l
ne
ur
a
l
ne
twor
ks
(
AN
N)
,
mul
ti
laye
r
pe
r
c
e
ptr
on
(
M
L
P
)
,
s
uppor
t
ve
c
tor
mac
hine
(
S
VM
)
,
k
-
ne
a
r
e
s
t
ne
ighbor
s
,
a
nd
li
ne
a
r
r
e
gr
e
s
s
ion
,
ha
ve
be
e
n
pr
opos
e
d
f
or
gr
a
din
g
[
10]
–
[
15
]
.
T
he
s
e
c
ontempor
a
r
y
gr
a
ding
a
ppr
oa
c
he
s
r
e
ly
on
the
c
he
mi
c
a
l
pr
ope
r
ti
e
s
of
a
ga
r
wood
oil
,
a
im
ing
to
e
nha
nc
e
the
a
c
c
ur
a
c
y
a
nd
r
e
li
a
bil
it
y
of
the
g
r
a
ding
s
ys
tem
[
13]
,
[
14]
,
[
16]
.
Va
r
ious
e
xtr
a
c
ti
on
tec
hniques
a
r
e
e
mpl
oye
d
to
o
btain
a
ga
r
wood
oil
,
including
s
upe
r
c
r
it
ica
l
f
lui
d
e
xtr
a
c
ti
on,
s
olvent
e
xtr
a
c
ti
on
,
hyd
r
o
-
dis
ti
ll
a
ti
on,
a
nd
other
s
.
P
r
e
-
tr
e
a
tm
e
nt
of
a
ga
r
wood
s
a
mpl
e
s
,
i
nvolvi
ng
c
he
mi
c
a
l
tr
e
a
tm
e
nt,
s
oa
king
i
n
wa
ter
,
a
nd
s
oni
c
a
ti
on,
is
de
e
med
ne
c
e
s
s
a
r
y
be
f
or
e
e
xtr
a
c
ti
on
[
4]
,
[
17]
.
S
ome
s
tudi
e
s
uti
li
z
e
ga
s
c
hr
omatogr
a
phy
-
mas
s
s
pe
c
tr
omete
r
(
GC
-
M
S
)
[
12]
,
[
18
]
,
[
19]
a
nd
s
ol
id
-
pha
s
e
mi
c
r
oe
xtr
a
c
ti
o
n
(
S
P
M
E
)
[
18]
,
[
20]
f
or
f
ur
the
r
a
na
l
ys
is
of
the
e
xtr
a
c
ted
oil
.
S
e
ve
r
a
l
r
e
s
e
a
r
c
h
s
tudi
e
s
i
nc
or
por
a
te
s
tati
s
ti
c
a
l
a
na
lys
is
tec
hniques
,
s
u
c
h
a
s
the
z
-
s
c
or
e
,
to
de
ter
mi
ne
the
qua
li
ty
gr
a
de
s
of
a
ga
r
wood
oil
.
Addi
ti
ona
ll
y,
mac
hine
lea
r
ning
a
lgor
it
hms
,
including
AN
N,
s
uppor
t
ve
c
tor
c
las
s
if
ier
s
,
a
nd
r
a
ndom
f
or
e
s
ts
,
a
r
e
e
mpl
oy
e
d
to
va
li
da
te
thes
e
gr
a
de
s
[
21]
–
[
24
]
.
T
he
z
-
s
c
or
e
method
r
e
li
e
s
on
de
tec
ti
ng
va
r
iations
in
the
a
bunda
nc
e
pa
tt
e
r
ns
of
indi
vidual
c
ompounds
,
a
nd
in
the
c
a
s
e
of
a
ga
r
wood
oil
qua
li
ty
g
r
a
ding,
s
e
ve
n
s
pe
c
if
ic
c
ompounds
:
β
-
a
ga
r
of
ur
a
n,
α
-
a
ga
r
of
ur
a
n,
10
-
e
pi
-
ɤ
-
e
ud
e
s
mol
,
ɤ
-
e
ude
s
mol
,
longi
f
olol
,
he
xa
de
c
a
nol,
a
nd
e
ude
s
mol
s
igni
f
ica
nt
ly
inf
luenc
e
the
a
s
s
e
s
s
ment
[
25]
,
[
26
]
.
T
he
s
e
c
o
mpounds
play
a
pivot
a
l
r
ole
in
de
ter
mi
ning
the
qua
li
ty
of
a
ga
r
wood
oil
.
T
he
β
-
a
ga
r
of
ur
a
n,
α
-
a
ga
r
of
ur
a
n,
10
-
e
pi
-
ɤ
-
e
ude
s
mol
,
a
nd
ɤ
-
e
ude
s
mol
a
r
e
ins
tr
umenta
l
in
g
r
a
din
g
high
-
qua
li
ty
a
ga
r
wood
oil
,
while
longi
f
olol
,
he
xa
de
c
a
nol,
a
nd
e
ude
s
mol
c
ontr
ibut
e
to
the
gr
a
ding
of
low
-
qua
li
ty
a
ga
r
wood
oil
[
25]
.
I
n
s
pe
c
if
ic
a
ga
r
w
ood
oil
s
a
mpl
e
s
,
J
B
D
a
nd
M
A2
we
r
e
identif
ied
a
s
high
qua
li
ty,
whe
r
e
a
s
C
KE
,
HD
,
a
nd
R
5
we
r
e
c
las
s
if
ied
a
s
low
qua
li
ty
[
21]
.
W
he
n
us
ing
the
z
-
s
c
or
e
me
thod
f
o
r
gr
a
ding
a
ga
r
wood
oil
qua
li
ty,
both
AN
N
a
nd
r
a
ndo
m
f
o
r
e
s
t
a
lgor
it
hms
ha
ve
p
r
ove
n
e
f
f
e
c
ti
ve
in
a
c
c
ur
a
tely
c
a
tegor
izing
a
ga
r
wood
oil
a
s
e
it
he
r
high
or
low
qu
a
li
ty,
with
mi
nim
a
l
pr
e
diction
e
r
r
or
[
24]
,
[
27]
.
T
he
s
e
lf
-
or
ga
nizing
map
(
S
OM
)
,
a
n
ANN
e
mpl
o
ying
a
c
lus
ter
ing
a
lgor
it
hm
f
o
r
high
-
dim
e
ns
ional
vis
ua
li
z
a
ti
on,
is
a
ls
o
known
a
s
the
Kohone
n
ne
twor
k,
a
c
onc
e
pt
int
r
oduc
e
d
by
T
e
uvo
Kohone
n
in
1
981.
T
he
a
dva
ntage
s
of
us
ing
S
OM
include
the
f
oll
owing
[
2
7]
–
[
29]
:
‒
Dimens
ional
r
e
duc
ti
on
:
S
OM
f
a
c
il
it
a
tes
the
r
e
du
c
ti
on
of
dim
e
ns
ions
,
s
im
pli
f
ying
the
int
e
r
pr
e
tatio
n
of
c
lus
ter
ing
outcome
s
.
B
y
tr
a
ns
f
or
mi
ng
a
high
-
dim
e
ns
ional
input
s
pa
c
e
int
o
a
lowe
r
-
dim
e
ns
ional
o
utput
s
pa
c
e
,
it
r
e
tains
the
o
r
igi
na
l
topo
logi
c
a
l
r
e
lations
hi
ps
.
‒
S
uit
a
bil
it
y
f
or
c
ompl
e
x
da
ta
:
S
OM
is
a
ppli
c
a
ble
in
s
c
e
na
r
ios
whe
r
e
a
c
ompr
e
he
ns
ive
unde
r
s
tanding
o
f
the
input
da
ta's
c
ha
r
a
c
ter
is
ti
c
s
is
a
b
s
e
nt.
I
t
e
xc
e
ls
a
t
identif
ying
pa
tt
e
r
ns
a
nd
r
e
lations
hips
e
ve
n
whe
n
the
da
ta
is
not
thor
oughly
unde
r
s
tood.
‒
E
a
s
e
of
us
e
:
the
a
lgor
it
hm
is
unc
ompl
ica
ted
a
nd
e
a
s
y
to
c
omput
e
,
e
nha
nc
ing
it
s
pr
a
c
ti
c
a
li
ty
a
nd
us
a
bil
it
y
a
c
r
os
s
diver
s
e
a
ppli
c
a
ti
ons
.
S
OM
,
a
ls
o
r
e
f
e
r
r
e
d
to
a
s
the
Kohone
n
ne
twor
k
,
p
r
ove
s
to
be
a
va
luable
too
l
f
or
both
vis
ua
li
z
ing
a
nd
c
lus
ter
ing
high
-
dim
e
ns
ional
da
ta.
I
ts
ve
r
s
a
ti
li
ty,
s
im
pli
c
it
y,
a
nd
br
oa
d
a
ppli
c
a
bil
it
y
make
it
a
n
e
xc
e
ll
e
nt
c
hoice
f
or
va
r
ious
da
ta
a
na
lys
is
tas
ks
.
F
igur
e
1
il
lus
tr
a
tes
the
ne
ur
a
l
ne
twor
k
s
tr
uc
tur
e
of
the
S
OM
,
c
ompr
is
ing
two
laye
r
s
:
the
inpu
t
laye
r
a
nd
the
output
laye
r
,
of
ten
c
a
ll
e
d
the
c
ompetit
ion
laye
r
.
T
he
numbe
r
of
ne
ur
ons
in
the
inpu
t
laye
r
is
de
ter
mi
ne
d
by
the
qua
nti
ty
of
ve
c
tor
s
in
the
inpu
t
ne
twor
k.
T
he
s
e
input
laye
r
ne
ur
ons
e
s
tablis
h
c
onne
c
ti
ons
with
ne
ur
ons
in
the
ou
tput
laye
r
thr
ough
we
ight
s
,
de
noted
a
s
W
[
30
]
.
E
a
c
h
ne
u
r
on
in
the
output
laye
r
c
a
n
be
c
onc
e
ptualize
d
a
s
r
e
pr
e
s
e
nti
ng
a
c
las
s
or
c
lus
te
r
that
c
ha
r
a
c
ter
ize
s
the
input
s
[
20]
.
T
he
or
ga
ni
z
a
ti
on
of
ne
ur
ons
in
the
output
laye
r
f
o
r
ms
a
two
-
dim
e
ns
ional
gr
id
or
node
matr
ix
,
f
a
c
il
it
a
ti
ng
the
vis
ua
li
z
a
ti
on
a
nd
or
ga
niza
ti
on
of
the
c
lus
ter
ing
pr
oc
e
s
s
withi
n
the
S
OM
.
T
his
c
onf
igur
a
ti
on
e
na
bles
the
S
OM
to
c
a
ptur
e
a
nd
r
e
pr
e
s
e
nt
int
r
ica
te
pa
tt
e
r
ns
a
nd
r
e
lations
hips
pr
e
s
e
nt
in
the
input
da
ta
.
F
igur
e
1.
S
OM
a
r
c
hit
e
c
tur
e
[
20]
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
Ar
ti
f
I
ntell
I
S
S
N:
2252
-
8938
A
ppli
c
ati
on
of
s
e
lf
-
or
ganiz
ing
map
for
mode
li
ng
th
e
A
quil
ar
ia
…
(
M
ohamm
ad
A
r
if
F
ahmi
C
he
Has
s
an
)
2891
A
S
OM
f
unc
ti
ons
a
s
a
c
ompetit
ive
ne
ur
a
l
ne
twor
k
,
f
oll
owing
the
pr
inciples
of
c
ompeti
ti
ve
lea
r
n
ing.
I
n
the
output
laye
r
,
a
ls
o
known
a
s
the
c
ompetit
ion
laye
r
,
ne
ur
ons
e
nga
ge
in
c
ompetit
ion
to
be
s
e
lec
ted
a
s
the
'winne
r
,
'
de
ter
mi
ne
d
by
their
pr
oxi
mi
ty
to
inpu
t
d
a
ta
ve
c
tor
s
.
Onc
e
the
winning
ne
ur
on
a
nd
it
s
ne
i
ghbor
ing
ne
ur
ons
a
r
e
identif
ied,
the
we
ight
ve
c
tor
s
a
s
s
oc
iate
d
with
them
unde
r
go
modi
f
ica
ti
on.
T
his
a
djus
tm
e
nt
pr
oc
e
s
s
is
s
tr
a
tegic
a
ll
y
de
s
igned
to
e
nha
nc
e
the
r
e
s
pons
ivene
s
s
of
the
winning
ne
ur
on
a
nd
it
s
ne
ighbor
s
to
s
im
il
a
r
input
pa
tt
e
r
ns
,
f
os
ter
ing
a
s
e
lf
-
or
ga
nizing
mec
ha
nis
m.
T
he
s
e
f
unda
menta
l
s
teps
e
nc
a
ps
ulate
Kohone
n's
S
OM
a
ppr
oa
c
h
[
28]
,
[
31
]
.
M
ode
li
ng
the
c
ompl
e
x
r
e
lations
hips
be
twe
e
n
the
s
igni
f
ica
nt
c
ompounds
a
nd
o
il
qua
li
ty
is
e
s
s
e
nti
a
l
f
or
unlocking
the
f
ull
po
tential
of
A
qu
il
ar
ia
mal
ac
c
e
ns
is
in
va
r
ious
a
ppli
c
a
ti
ons
.
S
OM
ha
ve
e
mer
ge
d
a
s
a
powe
r
f
ul
tool
f
o
r
modeling
c
ompl
e
x
da
tas
e
ts
a
nd
identif
ying
pa
tt
e
r
ns
in
mul
ti
dim
e
ns
ional
da
ta
[
6]
.
S
OM
's
a
bil
it
y
to
c
lus
ter
a
nd
map
high
-
dim
e
ns
ional
da
ta
make
s
it
a
p
r
omi
s
ing
a
ppr
oa
c
h
f
or
unde
r
s
tan
ding
the
int
r
ica
te
r
e
lations
hips
withi
n
the
s
e
s
quit
e
r
pe
noi
d
c
ompos
it
ion
of
A
quil
ar
ia
malac
c
e
ns
is
oil
.
I
t
e
mpl
oys
uns
upe
r
vis
e
d
lea
r
ning,
a
ll
owing
it
to
r
e
ve
a
l
hidd
e
n
pa
tt
e
r
ns
a
nd
s
tr
uc
tu
r
e
s
withi
n
s
e
s
quit
e
r
pe
noid
da
tas
e
ts
without
the
ne
e
d
f
or
p
r
e
de
f
ined
c
a
tegor
ies
.
T
his
c
ha
r
a
c
ter
is
ti
c
is
pa
r
ti
c
ular
ly
a
dva
ntage
ous
in
e
xplor
ing
the
diver
s
e
c
he
mi
c
a
l
c
ompos
it
ion
of
A
quil
ar
ia
malac
c
e
ns
is
oil
[
32]
,
[
33]
.
S
OM
pr
ovides
topol
ogica
l
mapping,
pr
e
s
e
r
ving
the
s
pa
ti
a
l
r
e
lations
hips
be
twe
e
n
dif
f
e
r
e
nt
s
e
s
quit
e
r
pe
noid
c
ompounds
.
T
his
f
e
a
tur
e
is
c
r
uc
ial
f
or
unde
r
s
tanding
how
s
ubtl
e
va
r
iations
in
the
c
he
mi
c
a
l
s
tr
uc
tur
e
im
pa
c
t
oil
qua
li
ty
[
6]
,
[
33
]
.
Unlike
c
onve
nti
ona
l
tec
hniques
that
ma
y
r
e
ly
on
s
ubjec
ti
v
e
a
s
s
e
s
s
ment
s
,
S
OM
pr
ovides
a
n
objec
ti
ve
a
nd
da
t
a
-
dr
iven
a
ppr
oa
c
h
to
modeling
s
e
s
quit
e
r
pe
noids
,
of
f
e
r
ing
a
mor
e
nua
nc
e
d
unde
r
s
tanding
of
their
r
ole
in
oil
qua
li
ty.
S
OM
,
a
type
of
ANN
,
ha
s
pr
ove
n
e
f
f
e
c
ti
ve
in
va
r
i
ous
f
ields
,
including
c
he
mi
nf
o
r
matics
a
nd
c
he
mi
c
a
l
pa
tt
e
r
n
r
e
c
ognit
ion.
I
n
the
c
ontext
o
f
na
tur
a
l
pr
oduc
ts
,
S
OM
ha
s
be
e
n
s
uc
c
e
s
s
f
ull
y
a
ppli
e
d
to
model
c
ompl
e
x
c
he
mi
c
a
l
da
ta,
pr
ovidi
ng
ins
ight
s
int
o
c
ompound
r
e
lations
hips
a
nd
c
las
s
if
ica
ti
ons
[
32]
,
[
33]
.
How
e
ve
r
,
it
s
a
ppli
c
a
ti
on
in
gr
a
ding
the
qua
li
ty
of
A
quil
ar
ia
m
alacc
e
ns
is
oil
,
pa
r
ti
c
ular
ly
c
ons
ider
ing
s
e
s
quit
e
r
pe
noid
pr
of
il
e
s
,
r
e
mains
a
n
unde
r
e
xplo
r
e
d
a
r
e
a
.
2.
T
HE
ORI
T
I
CA
L
WORK
T
he
outcome
s
de
r
ived
f
r
om
S
OM
lea
r
ning
of
f
e
r
va
luable
ins
ight
s
int
o
the
r
e
lations
hips
a
mong
ne
ighbor
ing
ne
ur
ons
,
known
a
s
S
OM
ne
ighbo
r
dis
tanc
e
s
,
a
nd
the
dis
tr
ibut
ion
of
we
ight
va
lues
,
vis
u
a
li
z
e
d
a
s
S
OM
we
ight
plane
s
.
T
yp
ica
ll
y
,
thes
e
r
e
s
ult
s
a
r
e
pr
e
s
e
nted
us
ing
c
olor
maps
[
32]
,
[
34]
.
I
n
S
OM
ne
ighbor
dis
tanc
e
s
,
he
xa
gons
a
nd
r
e
d
li
ne
s
de
pict
ne
ur
ons
a
nd
their
c
onne
c
ti
ons
,
r
e
s
pe
c
ti
ve
ly.
T
he
da
r
kne
s
s
of
the
c
olor
r
e
f
lec
ts
the
de
gr
e
e
o
f
d
is
tanc
e
,
with
da
r
k
e
r
s
ha
de
s
indi
c
a
ti
ng
gr
e
a
ter
dis
tanc
e
s
a
nd
li
ghte
r
s
h
a
de
s
indi
c
a
ti
ng
s
maller
one
s
[
34]
.
S
OM
we
ight
plane
s
vis
ua
ll
y
de
mon
s
tr
a
te
the
li
nk
be
twe
e
n
c
olor
a
nd
the
we
ight
of
the
output
ne
ur
on,
a
s
de
picte
d
in
F
igur
e
2.
I
n
thi
s
r
e
pr
e
s
e
n
tation,
li
ghter
a
nd
da
r
ke
r
c
olo
r
s
c
or
r
e
s
pond
to
la
r
ge
r
a
nd
s
maller
we
ight
s
,
r
e
s
pe
c
ti
ve
ly.
W
he
n
the
c
onn
e
c
ti
o
n
pa
tt
e
r
ns
of
two
input
s
e
xhibi
t
a
high
de
gr
e
e
of
s
i
mi
lar
it
y,
mea
ning
the
s
ha
pe
a
nd
c
olo
r
o
f
ne
u
r
ons
a
r
e
the
s
a
me
f
or
both
input
s
a
nd
it
s
ugge
s
ts
a
s
tr
ong
c
o
r
r
e
lation
be
twe
e
n
thos
e
input
s
[
34]
.
T
he
s
il
houe
tt
e
ind
e
x
(
S
I
)
f
unc
ti
ons
a
s
a
va
luable
tool
f
or
c
lus
ter
a
s
s
e
s
s
ment,
a
idi
ng
in
the
identif
ica
ti
on
o
f
objec
ts
that
a
r
e
a
p
pr
opr
iate
ly
plac
e
d
withi
n
their
a
s
s
igned
c
lus
ter
s
a
nd
thos
e
that
mi
ght
f
a
ll
in
be
twe
e
n
c
lus
ter
s
.
I
n
F
igu
r
e
3
[
35]
two
c
lus
ter
s
a
r
e
labe
led
a
s
A
a
nd
C
.
F
igur
e
2
.
S
OM
we
ight
p
lane
s
of
input
[
33
]
F
igur
e
3
.
C
omput
ing
s
il
houe
tt
e
index
T
o
c
omput
e
the
S
I
,
be
gin
by
s
e
lec
ti
ng
a
nd
labe
li
ng
a
n
objec
t
in
c
lus
ter
A
a
s
"
i,
"
then
c
a
lcula
te
a
(
i)
,
r
e
pr
e
s
e
nti
ng
the
a
ve
r
a
ge
dis
s
im
il
a
r
it
y
of
"
i"
to
oth
e
r
objec
ts
withi
n
A.
T
his
pr
ovides
ins
ight
int
o
the
a
ve
r
a
ge
length
of
c
onne
c
ti
ons
withi
n
c
lus
ter
A.
S
ubs
e
que
n
tl
y,
c
a
lcula
te
c
(
i
,
C
)
,
de
noti
ng
the
a
ve
r
a
ge
dis
s
im
i
lar
it
y
o
f
"
i"
to
a
ll
objec
ts
in
c
lus
ter
C
,
indi
c
a
ti
ng
the
a
ve
r
a
ge
length
of
c
onne
c
ti
ons
f
r
om
"
i"
(
in
c
lus
ter
A
)
to
c
lus
ter
C
.
P
r
oc
e
e
d
to
c
a
lcula
te
a
ll
c
(
i
,
C
)
va
lues
.
F
inally,
ide
nti
f
y
the
s
malles
t
number
a
mong
thes
e
va
lues
[
34
]
,
[
35]
.
I
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
Ar
ti
f
I
ntell
,
Vol.
14,
No.
4:
Augus
t
2025
:
288
9
-
2898
2892
e
s
s
e
nc
e
,
the
SI
e
va
luate
s
c
lus
ter
ing
qua
li
ty
by
c
o
mpar
ing
how
we
ll
a
n
objec
t
with
in
c
lus
ter
A
is
i
nter
na
ll
y
c
onne
c
ted
(
a
(
i)
)
ve
r
s
us
it
s
c
onne
c
ti
ons
to
objec
ts
in
other
c
lus
ter
s
(
c
(
i
,
C
)
)
,
with
a
lowe
r
va
lue
in
dica
ti
ng
be
tt
e
r
c
lus
ter
ing.
T
he
SI
r
a
nge
s
f
r
om
-
1
to
1,
o
f
f
e
r
ing
c
r
uc
ial
in
f
or
mation
a
bout
c
lus
ter
ing
qua
li
ty.
An
S
I
va
lue
c
los
e
to
1
s
igni
f
ies
that
the
objec
t
"
i"
is
s
igni
f
ica
ntl
y
c
lo
s
e
r
to
other
ob
jec
ts
withi
n
the
s
a
me
c
lus
ter
than
to
thos
e
in
the
ne
a
r
e
s
t
ne
ighbor
ing
c
lus
ter
,
indi
c
a
ti
ng
a
r
ob
us
t
a
nd
we
ll
-
de
f
ined
c
lus
ter
.
An
S
I
c
los
e
to
0
s
ugge
s
ts
unc
e
r
tainty
or
a
mbi
guit
y
in
the
c
lus
ter
ing
of
the
f
oc
a
l
objec
t,
im
plyi
ng
it
may
not
di
s
ti
nc
tl
y
be
lon
g
to
a
ny
s
pe
c
if
ic
c
lus
ter
.
An
S
I
va
lue
ne
a
r
-
1
ind
ica
tes
m
is
c
lus
ter
ing,
whe
r
e
the
objec
t
s
e
e
ms
mor
e
a
li
gne
d
with
a
dif
f
e
r
e
nt
c
lus
ter
than
it
s
a
s
s
igned
one
.
T
o
o
f
f
e
r
a
pr
a
c
ti
c
a
l
int
e
r
p
r
e
tation:
An
S
I
f
a
ll
ing
be
twe
e
n
0
.
71
a
nd
1.
00
is
de
e
med
a
n
"
e
xc
e
ll
e
nt
s
pli
t
,
"
indi
c
a
ti
ng
a
r
obus
t
a
nd
c
lea
r
ly
de
f
ined
c
lus
ter
s
e
pa
r
a
ti
on.
An
S
I
r
a
nging
f
r
om
0.
51
to
0
.
70
is
l
a
be
led
a
"
r
e
a
s
ona
ble
s
pli
t
,
"
s
igni
f
ying
a
r
e
a
s
ona
bly
we
ll
-
s
e
pa
r
a
ted
c
lus
ter
.
An
S
I
be
twe
e
n
0.
26
a
nd
0.
50
is
c
a
tegor
ize
d
a
s
a
"
we
a
k
s
pli
t
,
"
s
u
gge
s
ti
ng
a
les
s
dis
ti
nc
t
c
lus
ter
s
e
pa
r
a
ti
on.
An
S
I
be
low
0
.
5
is
ter
med
a
"
ba
d
s
pli
t
,
"
indi
c
a
ti
ng
a
poor
s
e
pa
r
a
ti
on
of
c
lus
ter
s
[
35
]
.
Addit
ionally,
c
a
lcula
ti
ng
the
a
ve
r
a
ge
S
I
va
lues
ove
r
a
c
l
us
ter
c
a
n
pr
ovide
a
n
a
s
s
e
s
s
ment
of
the
ove
r
a
ll
qua
li
ty
or
"
g
oodne
s
s
"
of
that
c
lus
ter
.
T
he
a
dva
ntage
s
of
S
I
a
r
e
outl
ined
a
s
f
oll
ows
:
‒
S
I
va
li
da
tes
c
lus
ter
ing
a
t
the
point
leve
l,
pr
ovidi
ng
the
f
ines
t
gr
a
nular
it
y
.
‒
S
I
is
indepe
nde
nt
o
f
a
ny
s
pe
c
if
ic
a
lgor
it
hm
.
‒
I
t
r
e
li
e
s
s
olely
on
pa
ir
wis
e
s
im
il
a
r
it
ies
-
dis
s
im
il
a
r
it
ies
a
nd
the
membe
r
s
hip
matr
ix
a
s
input
.
‒
S
I
is
a
ppli
e
d
f
or
e
va
luating
the
c
lus
ter
ing
qua
li
ty
of
a
s
e
pa
r
a
ti
on,
f
ulf
il
l
ing
the
objec
ti
ve
o
f
c
lus
ter
i
ng
by
a
s
s
e
s
s
ing
both
c
los
e
ne
s
s
a
nd
s
e
pa
r
a
ti
on.
3.
M
E
T
HO
D
OL
OG
Y
3.
1.
S
a
m
p
le
ac
q
u
is
it
ion
T
he
a
ga
r
wood
o
il
s
a
mpl
e
s
uti
li
z
e
d
in
thi
s
s
tudy
a
r
e
e
xc
lus
ively
de
r
ived
f
r
om
A
quil
ar
ia
s
pe
c
ies
,
so
ur
c
e
d
f
r
om
the
F
or
e
s
t
R
e
s
e
a
r
c
h
I
ns
ti
tut
e
M
a
lays
ia
(
F
R
I
M
)
a
nd
Unive
r
s
it
i
M
a
lays
ia
P
a
ha
ng
(
UM
P
)
.
A
tot
a
l
of
660
s
a
mpl
e
s
,
c
ompr
is
ing
22
pr
im
a
r
y
s
a
mpl
e
s
na
med
a
s
C
KE
,
C
M
,
E
O2,
E
O3
,
E
O4
,
HD
,
HG
,
J
B
D,
KB
,
L
A,
L
G,
M
,
M
A,
M
A1,
M
A2,
M
N,
M
NS,
M
P
E
,
M
S
,
R
5,
R
G,
a
nd
T
,
we
r
e
e
mpl
oye
d
in
thi
s
r
e
s
e
a
r
c
h.
E
a
c
h
s
a
mpl
e
c
ons
is
t
e
d
of
103
c
he
mi
c
a
l
c
ompounds
,
whic
h
we
r
e
meticulous
ly
e
xtr
a
c
ted
a
nd
a
na
lyze
d
us
ing
GC
-
M
S
.
T
he
GC
-
M
S
a
ppa
r
a
tus
wa
s
c
onf
igur
e
d
wi
th
the
f
ol
lowing
s
e
tt
ings
:
‒
T
he
ini
ti
a
l
tempe
r
a
tur
e
o
f
the
a
ppa
r
a
tus
wa
s
s
e
t
a
t
60
º
C
f
o
r
10
mi
nutes
.
‒
T
he
tempe
r
a
tur
e
gr
a
dua
ll
y
incr
e
a
s
e
d,
r
e
a
c
hing
230
º
C
with
a
n
incr
e
ment
of
3
º
C
pe
r
m
inut
e
.
‒
T
he
f
low
r
a
te
o
f
the
he
li
u
m
ga
s
c
a
r
r
ie
r
wa
s
maintai
ne
d
a
t
1
ml
pe
r
mi
nute.
‒
T
he
tempe
r
a
tur
e
of
the
ion
s
our
c
e
wa
s
s
e
t
a
t
280
º
C
.
I
de
nti
f
i
c
a
ti
on
of
s
igni
f
ica
nt
c
he
mi
c
a
l
c
ompounds
wa
s
a
c
c
ompl
is
h
e
d
by
matc
hing
them
to
the
mas
s
s
pe
c
tr
a
l
li
br
a
r
y
(
HPC
H2205.
L
;
W
il
e
y7Nis
t05a.
L
;
NI
S
T
05a
.
L
)
,
a
ided
by
a
c
he
mi
s
t.
3.
2.
De
s
ign
at
ion
of
agar
wood
oil
gr
ad
e
s
I
n
th
is
s
e
c
ti
on,
we
e
mpl
oye
d
the
S
OM
c
lus
ter
ing
tec
hnique
f
or
c
a
tegor
izing
a
ga
r
wood
oil
g
r
a
de
s
.
T
he
input
da
ta
f
or
tr
a
ini
ng
a
nd
tes
ti
ng
pu
r
pos
e
s
wa
s
de
r
ived
f
r
om
pr
incipa
l
c
omponent
a
na
lys
is
(
P
C
A)
a
nd
P
e
a
r
s
on’
s
c
or
r
e
lation.
I
nit
ially,
the
da
ta
unde
r
w
e
nt
pe
r
-
r
ow
r
a
ndomi
z
a
ti
on,
f
oll
owe
d
by
d
ivi
s
ion
int
o
a
n
80:20
r
a
ti
o
f
or
the
t
r
a
ini
ng
a
nd
tes
ti
ng
da
tas
e
ts
.
S
ubs
e
que
ntl
y,
e
a
c
h
da
tas
e
t
unde
r
we
nt
tr
a
ns
pos
it
ion.
P
r
ior
to
a
pplyi
ng
the
S
OM
a
lgor
it
hm
,
a
thor
ough
a
s
s
e
s
s
ment
e
ns
ur
e
d
the
inclus
ion
of
a
ll
e
s
s
e
nti
a
l
s
a
mpl
e
s
in
both
da
tas
e
ts
.
I
f
thi
s
c
ondit
ion
wa
s
met,
the
pa
r
a
mete
r
s
of
the
S
OM
,
including
di
mens
ion,
topol
ogy,
a
nd
dis
tanc
e
f
unc
ti
on,
we
r
e
s
e
t.
I
n
c
a
s
e
s
whe
r
e
the
c
r
it
e
r
ia
we
r
e
not
s
a
ti
s
f
ied,
the
r
a
ndomi
z
a
ti
on,
divi
s
ion,
a
nd
tr
a
ns
pos
e
pr
oc
e
s
s
e
s
we
r
e
it
e
r
a
ted.
T
he
S
OM
tr
a
ini
ng
a
nd
tes
ti
ng
pr
oc
e
dur
e
s
we
r
e
then
e
xe
c
uted.
F
oll
owing
the
t
r
a
ini
ng
a
nd
tes
ti
ng
pha
s
e
s
,
s
il
ho
ue
tt
e
va
lues
of
the
c
lus
ter
s
we
r
e
c
omput
e
d
a
nd
s
c
r
uti
nize
d
f
or
ne
ga
ti
ve
va
lues
in
both
da
tas
e
ts
.
I
f
ne
ga
ti
ve
va
lues
we
r
e
de
tec
ted,
the
e
nti
r
e
c
lus
ter
ing
pr
oc
e
s
s
wa
s
r
e
c
a
lcula
ted.
Only
c
lus
ter
s
e
xhi
bit
ing
pos
it
ive
s
il
houe
tt
e
va
lues
we
r
e
a
c
knowle
dge
d.
Upon
a
c
knowle
dgment,
the
S
OM
va
li
da
ti
on
pr
oc
e
dur
e
wa
s
ini
ti
a
ted.
T
he
c
lus
ter
ing
r
ules
a
ppli
e
d
in
t
he
S
OM
a
lgor
it
hm
a
r
e
outl
ined
a
s
f
ol
lows
:
a.
I
nput:
c
he
mi
c
a
l
c
ompounds
b.
Output:
number
o
f
ne
ur
ons
that
r
e
p
r
e
s
e
nt
the
gr
a
de
s
of
a
ga
r
wood
oil
c.
Dimens
ion:
‒
1
by
2
gr
id
f
o
r
2
g
r
a
de
s
(
e
a
c
h
ne
ur
on
r
e
pr
e
s
e
n
ts
a
c
lus
ter
r
e
pr
e
s
e
nti
ng
e
it
he
r
high
or
low
gr
a
de
of
a
ga
r
wood
oil
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
Ar
ti
f
I
ntell
I
S
S
N:
2252
-
8938
A
ppli
c
ati
on
of
s
e
lf
-
or
ganiz
ing
map
for
mode
li
ng
th
e
A
quil
ar
ia
…
(
M
ohamm
ad
A
r
if
F
ahmi
C
he
Has
s
an
)
2893
‒
1
by
3
gr
id
f
or
3
g
r
a
de
s
(
e
a
c
h
ne
ur
on
r
e
pr
e
s
e
nts
a
c
lus
ter
r
e
pr
e
s
e
nti
ng
e
it
he
r
high
,
medium
,
o
r
low
gr
a
de
of
a
ga
r
wood
oil
)
‒
2
by
2
g
r
id
f
o
r
4
gr
a
de
s
(
e
a
c
h
ne
ur
on
r
e
p
r
e
s
e
nts
a
c
lus
ter
r
e
pr
e
s
e
nti
ng
e
it
he
r
high
,
medium
-
high,
medium
-
low,
or
low
g
r
a
de
of
a
ga
r
wood
oil
)
d.
T
opology
f
unc
ti
on
:
He
xtop
(
he
xa
gona
l
pa
tt
e
r
n)
e.
Dis
tanc
e
f
unc
ti
on:
E
uc
li
de
a
n
dis
tanc
e
s
Out
of
103
c
he
mi
c
a
l
c
ompounds
,
only
thr
e
e
we
r
e
s
e
lec
ted
a
s
input
va
r
iable
s
f
or
the
S
OM
:
β
-
a
ga
r
of
ur
a
n,
α
-
a
ga
r
of
ur
a
n,
a
nd
10
-
e
pi
-
ɤ
-
e
ude
s
mo
l.
T
he
r
e
s
ult
ing
outpu
t
wa
s
indi
c
a
ti
ve
of
the
a
ga
r
wood
oil
gr
a
de
,
c
a
tegor
ize
d
a
s
e
it
he
r
h
igh
o
r
low
.
T
he
S
OM
c
lus
ter
ing
pr
oc
e
s
s
a
dhe
r
e
d
to
the
f
oll
owing
r
ules
:
‒
T
he
r
a
ti
o
of
tr
a
ini
ng
a
nd
tes
ti
ng:
80
to
20
‒
T
wo
ne
ur
ons
:
e
a
c
h
ne
ur
on
r
e
pr
e
s
e
nts
a
c
lus
ter
that
is
e
it
he
r
high
or
low
qua
li
ty
‒
Dimens
ion:
1
by
2
gr
id
‒
T
opology
f
unc
ti
on
:
he
xtop
(
he
xa
gona
l
pa
tt
e
r
n)
‒
Dis
tanc
e
f
unc
ti
on:
E
uc
li
de
a
n
dis
tanc
e
s
‒
C
ove
r
S
teps
:
100
‒
I
nit
Ne
ighbor
:
1
T
he
s
e
s
pe
c
if
ied
r
ules
we
r
e
incor
por
a
ted
int
o
the
S
OM
a
lgor
it
hm
f
or
im
pleme
ntation.
T
he
ini
ti
a
l
s
tep
invol
ve
d
r
a
ndomi
z
ing
the
da
ta
on
a
pe
r
-
r
ow
ba
s
is
,
f
oll
owe
d
by
their
div
is
ion
int
o
tr
a
ini
ng
a
nd
tes
ti
ng
da
tas
e
ts
with
a
n
80:20
r
a
ti
o.
S
ubs
e
que
ntl
y,
e
a
c
h
da
tas
e
t
unde
r
we
nt
tr
a
ns
pos
it
ion.
P
r
ior
to
the
S
OM
c
omput
a
ti
on,
the
tes
ti
ng
da
tas
e
t
unde
r
we
nt
s
c
r
uti
ny
to
e
ns
ur
e
the
inc
lus
ion
of
a
ll
pr
im
a
r
y
s
a
mpl
e
s
.
F
oll
owing
thi
s
,
ke
y
S
OM
p
r
ope
r
ti
e
s
,
including
di
mens
ion,
c
ove
r
S
teps
,
ini
tNe
ighbor
,
topol
ogy,
a
nd
dis
tanc
e
f
unc
ti
on,
we
r
e
c
onf
igur
e
d.
T
he
da
tas
e
ts
f
or
tr
a
ini
ng
,
tes
ti
ng,
a
nd
va
li
da
ti
on
we
r
e
then
c
omput
e
d
s
e
que
nti
a
ll
y.
S
ubs
e
que
ntl
y,
s
il
houe
tt
e
va
lues
f
o
r
e
a
c
h
c
lus
ter
we
r
e
c
omput
e
d
a
nd
s
c
r
uti
nize
d
f
or
ne
ga
ti
ve
va
lues
in
both
tr
a
ini
ng
a
nd
tes
ti
ng
da
tas
e
t
s
.
Only
c
lus
ter
s
e
xhibi
ti
ng
pos
it
ive
s
il
houe
tt
e
va
l
ue
s
we
r
e
c
ons
ider
e
d
va
li
d.
T
he
pr
ogr
a
m
c
onc
luded
upon
the
f
ulf
i
ll
ment
o
f
thi
s
c
ondit
ion
,
a
s
il
lus
tr
a
ted
in
Algo
r
it
hm
1
.
Algor
it
hm
1
.
S
OM
a
lgor
it
hm
f
or
c
lus
ter
ing
Input: data T, training data Tr, testing data Ts, validation data Tv, main samples M
Output: predicted cluster of training data Dtr, predicted cluster of testing data Dts
,
predicted cluster of validation data Dtv
1
load
T,
Tv
2
while
silhouette
values
of
Dtr
≤
0 or
Dts
≤
0 do
3
While
Ts
⊅
M
do
4
randomize
T
5
split
T to
Tr and
Ts
with
80 to
20
ratio
6
Tr’
←
Tr;
Ts’ ←
Ts
7
end
while
8
set
SOM
parameters:
dimensions,
topologyFc
n,
distanceFcn
9
start
training
10
start
testing
11
return
Dtr,
Dts
12
calcul
ate
and
plot
silhouette
values
of
Dtr,
Dts
13
calcul
ate
average
silhouette
values
of
Dtr,
Dts
14
end
while
15
start
validation
16
return
Dtv
4.
RE
S
UL
T
S
AN
D
DI
S
CU
S
S
I
ON
T
he
outcome
s
e
nc
ompas
s
ing
we
ight
dis
tanc
e
s
be
twe
e
n
ne
ur
ons
,
c
ompound
we
ight
s
to
ne
u
r
ons
,
s
il
houe
tt
e
va
lues
f
o
r
both
tr
a
ini
ng
a
nd
tes
ti
ng,
a
nd
the
a
s
s
ignm
e
nt
of
ne
ur
ons
to
a
ga
r
wood
oil
s
a
mpl
e
s
will
be
pr
e
s
e
nted
a
nd
dis
c
u
s
s
e
d
withi
n
thi
s
s
ub
s
e
c
ti
on.
F
igur
e
4
il
lus
tr
a
tes
the
we
ight
dis
tanc
e
be
twe
e
n
ne
ur
o
ns
,
with
e
a
c
h
ne
ur
on
de
picte
d
a
s
a
blue
he
xa
gon.
Ne
ur
on
1
is
s
it
ua
ted
a
t
the
bott
om,
while
ne
ur
on
2
is
pos
i
ti
one
d
a
t
the
top.
T
he
c
olor
a
ti
on
of
the
r
e
gion
be
twe
e
n
the
ne
ur
ons
s
e
r
ve
s
a
s
a
n
indi
c
a
tor
of
their
dis
tanc
e
,
wi
th
da
r
ke
r
hue
s
s
igni
f
ying
g
r
e
a
ter
dis
tanc
e
s
a
nd
vice
ve
r
s
a
.
I
n
F
igur
e
4
,
the
r
e
gion
is
c
olor
e
d
r
e
d,
indi
c
a
ti
ng
a
moder
a
te
dis
tanc
e
be
twe
e
n
the
ne
u
r
ons
.
T
his
vis
ua
l
r
e
pr
e
s
e
ntation
a
ids
in
the
int
e
r
pr
e
tation
o
f
the
ne
u
r
a
l
r
e
la
ti
ons
hips
in
the
c
ontex
t
o
f
the
s
tudy.
F
igur
e
5
dis
plays
the
c
ompound
we
ight
s
a
s
s
igned
to
ne
ur
ons
,
with
he
xa
gons
r
e
pr
e
s
e
nti
ng
ne
ur
ons
labe
led
a
s
N
e
ur
ons
1
a
nd
2
a
t
the
bott
om
a
nd
top,
r
e
s
pe
c
ti
ve
ly.
T
he
c
olor
s
of
thes
e
ne
ur
ons
s
igni
f
y
the
r
e
s
pe
c
ti
ve
c
ompound
we
ight
s
,
whe
r
e
li
ghte
r
a
n
d
da
r
ke
r
s
ha
de
s
de
note
lar
ge
r
a
nd
s
maller
c
ont
r
ibut
ions
.
Nota
bly,
the
c
ompounds
β
-
a
ga
r
of
ur
a
n,
α
-
a
ga
r
of
ur
a
n,
a
nd
10
-
e
pi
-
ɤ
-
e
ude
s
mol
de
mons
tr
a
ted
a
mor
e
s
ubs
tantial
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
Ar
ti
f
I
ntell
,
Vol.
14,
No.
4:
Augus
t
2025
:
288
9
-
2898
2894
c
ontr
ibut
ion
to
ne
ur
on
1
c
ompar
e
d
to
ne
u
r
on
2.
L
it
e
r
a
tur
e
s
our
c
e
s
ha
ve
c
ons
i
s
tently
identi
f
ied
thes
e
c
ompounds
a
s
s
igni
f
ica
nt
c
ontr
ibut
or
s
to
high
-
qua
li
ty
oil
.
C
ons
e
que
ntl
y,
ne
ur
on
1
is
indi
c
a
ti
ve
of
a
high
-
gr
a
de
c
lus
ter
,
while
ne
ur
on
2
r
e
pr
e
s
e
nts
a
low
-
gr
a
de
c
lus
ter
.
T
he
c
onne
c
ti
on
pa
tt
e
r
n
of
thes
e
c
o
mpounds
r
e
maine
d
c
ons
is
tent
a
c
r
os
s
a
ll
inpu
ts
,
with
ne
ur
on
1
be
ing
ye
ll
ow
a
nd
ne
ur
on
2
be
ing
blac
k,
highl
i
ghti
ng
a
s
tr
ong
c
or
r
e
lation
be
twe
e
n
the
c
ompounds
.
F
igur
e
4.
W
e
ight
dis
tanc
e
be
twe
e
n
ne
ur
ons
F
igur
e
5.
W
e
ight
o
f
c
ompounds
to
ne
ur
ons
1
(
high
gr
a
de
)
a
nd
2
(
low
gr
a
de
)
F
igur
e
6
pr
e
s
e
nts
s
il
houe
tt
e
plot
s
f
or
e
a
c
h
ne
ur
on,
the
tr
a
ini
ng
plo
t
s
hown
in
F
igur
e
6
(
a
)
a
nd
the
tes
ti
ng
plo
t
s
hown
in
F
igur
e
6(
b
)
.
T
he
da
tas
e
t
c
ompr
is
e
d
528
t
r
a
ini
ng
s
a
mpl
e
s
a
nd
132
tes
t
s
a
mpl
e
s
,
a
ll
of
whic
h
e
xhibi
ted
pos
it
ive
s
il
houe
tt
e
va
lues
in
both
t
r
a
ini
ng
a
nd
tes
ti
ng
pha
s
e
s
.
S
pe
c
if
ica
ll
y,
in
the
tr
a
ini
ng
s
e
t,
the
a
ve
r
a
ge
s
il
houe
tt
e
va
lues
f
or
ne
ur
on
1
(
r
e
pr
e
s
e
nti
ng
the
h
igh
-
gr
a
de
c
lus
ter
)
a
nd
ne
ur
on
2
(
r
e
pr
e
s
e
nti
ng
the
low
-
gr
a
de
c
lus
ter
)
we
r
e
0.
82
a
n
d
0.
67
,
r
e
s
pe
c
ti
ve
ly.
Du
r
ing
tes
ti
ng,
the
a
ve
r
a
ge
s
il
houe
tt
e
va
lues
f
or
ne
ur
on
1
a
nd
ne
u
r
on
2
we
r
e
0
.
79
a
nd
0.
58,
r
e
s
pe
c
ti
ve
ly.
T
his
obs
e
r
va
ti
on
s
igni
f
ies
that
the
high
-
gr
a
de
c
lus
ter
(
ne
ur
on
1)
c
ons
is
tently
yielde
d
s
up
e
r
ior
a
ve
r
a
ge
s
il
houe
tt
e
va
lues
c
ompar
e
d
to
the
l
ow
-
gr
a
de
c
lus
ter
(
ne
ur
on
2
)
in
both
the
tr
a
ini
ng
a
nd
tes
ti
ng
da
tas
e
ts
.
T
he
s
il
houe
tt
e
plot
s
s
ugge
s
t
that
the
s
a
m
ples
a
li
gn
we
ll
withi
n
their
de
s
ignate
d
c
lus
ter
s
,
dis
ti
n
guis
hing
be
twe
e
n
low
a
nd
high
gr
a
de
s
,
but
e
xhibi
t
poor
e
r
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
Ar
ti
f
I
ntell
I
S
S
N:
2252
-
8938
A
ppli
c
ati
on
of
s
e
lf
-
or
ganiz
ing
map
for
mode
li
ng
th
e
A
quil
ar
ia
…
(
M
ohamm
ad
A
r
if
F
ahmi
C
he
Has
s
an
)
2895
a
li
gnment
with
their
ne
ighbor
ing
c
lus
ter
s
.
T
his
a
li
gnment
pa
tt
e
r
n
pr
ovides
va
luable
ins
ight
s
int
o
the
e
f
f
ica
c
y
of
the
c
lus
ter
ing
pr
oc
e
s
s
a
nd
the
dis
ti
nc
t
s
e
pa
r
a
ti
on
of
low
a
nd
high
-
gr
a
de
c
lus
ter
s
.
(
a
)
(
b)
F
igur
e
6.
S
il
houe
tt
e
plot
f
or
(
a
)
tr
a
ini
ng
da
tas
e
t
a
nd
(
b)
tes
ti
ng
da
tas
e
t
T
he
a
ll
oc
a
ti
on
of
p
r
im
a
r
y
s
a
mpl
e
s
to
dis
ti
nc
t
a
ga
r
wood
oil
gr
a
de
s
is
meticulous
ly
de
tailed
in
T
a
ble
1,
p
r
ovidi
ng
a
c
ompr
e
he
ns
ive
ove
r
view
of
the
tr
a
ini
ng
a
nd
tes
ti
ng
pha
s
e
s
.
Nota
bly,
the
h
i
gh
gr
a
de
c
a
tegor
y
e
nc
ompas
s
e
s
J
B
D,
KB
,
L
A,
M
A,
M
A1
,
M
A2,
M
NS
,
M
P
E
,
R
G,
a
nd
T
,
c
oll
e
c
ti
ve
ly
f
o
r
mi
ng
a
c
ohe
s
ive
unit
withi
n
the
high
-
gr
a
de
c
lus
ter
a
s
s
oc
iate
d
with
ne
ur
on
1
.
C
onve
r
s
e
ly,
the
low
g
r
a
de
c
lus
ter
,
r
e
pr
e
s
e
nted
by
ne
ur
ons
2,
c
ompr
is
e
s
C
KE
,
C
M
,
E
O2,
E
O3
,
E
O4,
HD
,
L
G,
M
,
M
N,
M
S
,
a
nd
R
5.
T
he
numer
ic
va
lues
f
e
a
tur
e
d
in
the
table
de
note
the
qua
nti
ty
of
s
a
mpl
e
s
withi
n
e
a
c
h
r
e
s
pe
c
ti
ve
c
a
tegor
y.
T
his
s
tr
a
tegic
a
s
s
ignm
e
nt
of
p
r
im
a
r
y
s
a
mpl
e
s
un
de
r
s
c
or
e
s
the
pr
e
c
is
ion
of
ou
r
a
ppr
oa
c
h,
a
li
gning
with
the
ne
ur
a
l
ne
twor
k's
a
bil
it
y
to
dis
c
e
r
n
a
nd
c
las
s
if
y
a
ga
r
wood
oil
gr
a
de
s
.
T
he
e
xpli
c
i
t
de
taili
n
g
of
the
s
a
mpl
e
dis
tr
ibut
ion
a
mong
ne
ur
ons
e
nha
nc
e
s
the
tr
a
ns
pa
r
e
nc
y
a
nd
r
e
pli
c
a
bil
it
y
of
our
methodolo
gy.
T
he
r
e
s
ult
s
manif
e
s
t
a
c
lea
r
de
mar
c
a
ti
on
be
twe
e
n
high
a
nd
low
-
gr
a
de
c
lus
ter
s
,
s
e
tt
ing
the
s
tage
f
or
a
r
obus
t
e
va
luation
of
the
pr
opos
e
d
c
las
s
if
ica
ti
on
model.
T
a
ble
1
.
T
r
a
in
ing
a
nd
tes
ti
ng
da
ta
S
a
mpl
e
s
T
r
a
in
in
g
T
e
s
ti
ng
N
e
ur
on1 (
hi
gh gr
a
de
)
N
e
ur
on2 (
lo
w
gr
a
de
)
N
e
ur
on1 (
hi
gh gr
a
de
)
N
e
ur
on2 (
lo
w
gr
a
de
)
C
K
E
0
22
0
8
CM
0
24
0
6
E
O
2
0
23
0
7
E
O
3
0
21
0
9
E
O
4
0
24
0
6
HD
0
22
0
8
HG
27
0
3
0
J
B
D
25
0
5
0
KB
27
0
3
0
LA
23
0
7
0
LG
0
25
0
5
M
0
24
0
6
MA
26
0
4
0
M
A
1
24
0
6
0
M
A
2
23
0
7
0
MN
0
27
0
3
M
N
S
25
0
5
0
M
P
E
26
0
4
0
MS
0
21
0
9
R5
0
25
0
5
RG
21
0
9
0
T
23
0
7
0
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
Ar
ti
f
I
ntell
,
Vol.
14,
No.
4:
Augus
t
2025
:
288
9
-
2898
2896
5.
CONC
L
USI
ON
T
he
a
na
lys
is
de
mons
tr
a
tes
that
a
mi
nim
a
l
ye
t
t
a
r
ge
ted
s
e
lec
ti
on
of
thr
e
e
c
he
mi
c
a
l
c
ompounds
:
β
-
a
ga
r
of
ur
a
n,
α
-
a
ga
r
of
ur
a
n,
a
nd
10
-
e
pi
-
π
-
e
ud
e
s
m
ol
c
a
n
e
f
f
e
c
ti
ve
ly
c
las
s
if
y
a
ga
r
wood
oil
int
o
hig
h
or
low
gr
a
de
s
.
T
h
is
f
indi
ng
r
e
inf
o
r
c
e
s
the
r
e
li
a
bil
it
y
a
nd
dis
c
r
im
inative
powe
r
of
thes
e
s
pe
c
if
ic
mar
ke
r
s
in
a
s
s
e
s
s
ing
a
ga
r
wood
oil
qua
li
ty,
making
them
va
luabl
e
i
ndica
tor
s
f
or
both
r
e
s
e
a
r
c
h
a
nd
indus
tr
y
a
ppli
c
a
ti
ons
.
Additi
ona
ll
y
,
the
a
ppli
c
a
ti
on
of
S
OM
de
mons
tr
a
tes
notable
pr
of
icie
nc
y
in
c
lus
ter
ing
a
ga
r
wood
oil
int
o
high
a
nd
low
gr
a
de
s
,
a
s
e
videnc
e
d
by
a
ve
r
a
ge
s
il
houe
tt
e
va
lues
r
a
nging
f
r
om
0
.
58
to
0
.
82.
T
his
n
ot
only
r
e
inf
or
c
e
s
the
e
f
f
ica
c
y
of
the
c
hos
e
n
c
he
mi
c
a
l
c
ompounds
b
ut
a
ls
o
highl
igh
ts
the
a
bil
it
y
of
S
OM
to
p
r
ovide
a
r
e
li
a
ble
a
nd
a
c
c
ur
a
te
c
las
s
if
ica
ti
on
of
a
ga
r
wood
oil
qua
li
ty.
I
n
c
onc
lus
ion,
the
pr
opos
e
d
method
of
qua
li
ty
gr
a
ding
f
or
a
ga
r
wood
oil
,
r
e
lyi
ng
on
the
c
he
mi
c
a
l
c
ompo
unds
β
-
a
ga
r
of
ur
a
n,
α
-
a
ga
r
of
ur
a
n,
a
nd
10
-
e
pi
-
ɤ
-
e
ude
s
mol
thr
ough
S
OM
,
ha
s
be
e
n
va
li
da
ted
a
s
a
n
e
f
f
e
c
ti
ve
a
nd
de
pe
nda
ble
a
ppr
oa
c
h.
As
a
dir
e
c
ti
on
f
or
f
utur
e
r
e
s
e
a
r
c
h,
e
xtending
the
qua
li
ty
c
las
s
if
ica
ti
on
int
o
high
,
m
e
dium
,
a
nd
low
gr
a
d
e
s
c
ould
o
f
f
e
r
a
mor
e
nua
n
c
e
d
a
nd
r
e
f
ined
unde
r
s
tanding
of
a
ga
r
wood
oil
va
r
iati
ons
.
T
his
e
xpa
ns
ion
would
c
ontr
ibut
e
f
ur
the
r
to
the
a
dva
nc
e
ment
of
a
ga
r
wood
indus
tr
y
s
tanda
r
ds
a
nd
de
e
pe
n
our
ins
ight
in
to
the
nua
nc
e
d
gr
a
da
ti
ons
wi
thi
n
thi
s
va
luable
e
s
s
e
nti
a
l
oil
.
AC
KNOWL
E
DGM
E
N
T
S
T
he
a
uthor
s
would
li
ke
to
e
xpr
e
s
s
their
s
ince
r
e
gr
a
ti
tude
to
a
ll
pa
r
ti
e
s
invol
ve
d
a
nd
to
Unive
r
s
it
i
T
e
knologi
M
AR
A
(
UiT
M
)
f
or
their
c
onti
nuous
s
uppor
t
a
nd
c
ont
r
ibut
ions
thr
oughout
the
c
ou
r
s
e
of
th
is
wor
k.
F
UN
DI
NG
I
NF
ORM
AT
I
ON
T
he
a
uthor
s
wis
h
to
e
xtend
their
a
ppr
e
c
iatio
n
to
the
F
a
c
ult
y
of
E
lec
tr
ica
l
E
nginee
r
ing
a
t
UiT
M
S
ha
h
Ala
m,
S
e
langor
f
o
r
their
c
onti
nuous
f
inanc
ial
s
uppor
t
dur
ing
thi
s
r
e
s
e
a
r
c
h
unde
r
F
R
GS
gr
a
nt
(
600
-
R
M
C
/F
R
GS
5/3
(
154/2023)
.
AU
T
HO
R
CONT
RI
B
U
T
I
ONS
S
T
AT
E
M
E
N
T
Nam
e
of
Au
t
h
or
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
M
oha
mm
a
d
Ar
if
F
a
hmi
C
he
Ha
s
s
a
n
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
Z
a
kiah
M
ohd
Yus
of
f
✓
✓
✓
✓
✓
✓
✓
✓
✓
Nur
laila
I
s
mail
✓
✓
✓
✓
✓
✓
✓
✓
M
ohd
Na
s
ir
T
a
ib
✓
✓
✓
✓
✓
✓
✓
✓
✓
C
:
C
onc
e
pt
ua
li
z
a
ti
on
M
:
M
e
th
odol
ogy
So
:
So
f
twa
r
e
Va
:
Va
li
da
ti
on
Fo
:
Fo
r
ma
l
a
na
ly
s
is
I
:
I
nve
s
ti
ga
ti
on
R
:
R
e
s
our
c
e
s
D
:
D
a
ta
C
ur
a
ti
on
O
:
W
r
it
in
g
-
O
r
ig
in
a
l
D
r
a
f
t
E
:
W
r
it
in
g
-
R
e
vi
e
w
&
E
di
ti
ng
Vi
:
Vi
s
ua
li
z
a
ti
on
Su
:
Su
pe
r
vi
s
io
n
P
:
P
r
oj
e
c
t
a
dmi
ni
s
tr
a
ti
on
Fu
:
Fu
ndi
ng a
c
qui
s
it
io
n
CONF
L
I
CT
OF
I
NT
E
RE
S
T
S
T
AT
E
M
E
N
T
Author
s
s
tate
no
c
onf
li
c
t
of
int
e
r
e
s
t.
I
NF
ORM
E
D
CONSE
NT
W
e
ha
ve
obtaine
d
inf
or
med
c
ons
e
nt
f
r
om
a
ll
ind
ivi
dua
ls
include
d
in
thi
s
s
tudy.
E
T
HI
CA
L
AP
P
ROVA
L
Not
a
ppli
c
a
ble.
DA
T
A
AV
AI
L
A
B
I
L
I
T
Y
Da
ta
a
va
il
a
bil
it
y
is
n
ot
a
pp
li
c
a
b
le
to
th
is
pa
pe
r
a
s
n
o
ne
w
d
a
ta
we
r
e
c
r
e
a
te
d
or
a
na
lyze
d
in
thi
s
s
tu
dy
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
Ar
ti
f
I
ntell
I
S
S
N:
2252
-
8938
A
ppli
c
ati
on
of
s
e
lf
-
or
ganiz
ing
map
for
mode
li
ng
th
e
A
quil
ar
ia
…
(
M
ohamm
ad
A
r
if
F
ahmi
C
he
Has
s
an
)
2897
RE
F
E
RE
NC
E
S
[
1]
P
.
S
a
r
ma
h
e
t
al
.
,
“
A
n
in
s
ig
ht
in
to
th
e
im
munom
odul
a
to
r
y
pot
e
nt
ia
l
of
w
ood
oi
l
of
A
qui
la
r
ia
ma
la
c
c
e
ns
is
L
a
m.
w
it
h
a
n
e
mpha
s
i
s
in
to
r
e
la
te
d
phyt
ome
di
c
in
e
,
bi
oma
r
ke
r
s
,
pha
r
ma
c
ol
ogy,
a
nd
to
xi
c
it
y,”
Sout
h
A
fr
ic
an
J
our
nal
of
B
ot
any
,
vo
l.
151,
pp.
695
–
7
12,
2022, doi:
10.1016/j
.s
a
jb
.2022.10.020.
[
2]
N
.
Z
.
M
a
ha
bob,
Z
.
M
.
Y
us
of
f
,
A
.
F
.
M
.
A
mi
don,
N
.
I
s
ma
il
,
a
n
d
M
.
N
.
T
a
ib
,
“
A
nove
l
a
ppl
ic
a
ti
on
of
a
r
ti
f
ic
ia
l
ne
ur
a
l
ne
twor
k
f
o
r
c
la
s
s
if
yi
ng
a
ga
r
w
ood
e
s
s
e
nt
ia
l
oi
l
qua
li
ty
,
”
I
nt
e
r
nat
io
nal
J
o
ur
nal
of
E
le
c
tr
ic
al
and
C
o
m
put
e
r
E
ngi
ne
e
r
in
g
,
vol
.
12,
no.
6,
pp. 6645
–
6652, 2022, doi:
10.11591/i
je
c
e
.v12i6.pp6645
-
6652.
[
3]
R
.
G
og
oi
e
t
al
.
,
“
A
g
a
r
w
o
od
(
A
qu
il
a
r
i
a
m
a
l
a
c
c
e
n
s
i
s
L
.)
a
qu
a
li
ty
f
r
a
g
r
a
nt
a
nd
m
e
d
i
c
in
a
ll
y
s
i
gn
if
i
c
a
nt
p
la
nt
b
a
s
e
d
e
s
s
e
n
ti
a
l
o
il
w
it
h
p
h
a
r
m
a
c
o
lo
gi
c
a
l
po
t
e
nt
i
a
l
s
a
nd
g
e
n
ot
ox
ic
it
y,
”
I
n
du
s
t
r
ia
l
C
r
o
p
s
an
d
P
r
o
du
c
t
s
,
vol
.
19
7,
2
02
3,
d
oi
:
10
.1
01
6/
j.
i
nd
c
r
op
.2
02
3.
11
65
35
.
[
4]
P
.
W
u
e
t
al
.
,
“
E
xt
r
a
c
ti
on
pr
o
c
e
s
s
,
c
h
e
mi
c
a
l
pr
of
il
e
,
a
nd
bi
ol
ogi
c
a
l
a
c
ti
v
it
y
of
a
r
om
a
ti
c
oi
l
f
r
om
a
g
a
r
w
o
od
le
a
f
(
A
qui
l
a
r
i
a
s
in
e
n
s
i
s
)
by s
upe
r
c
r
it
i
c
a
l
c
a
r
bon
di
ox
id
e
e
xt
r
a
c
t
io
n,
”
J
ou
r
nal
of
C
O
2
U
ti
li
z
at
io
n
, v
ol
.
77,
2023
, doi
:
10
.101
6/
j.
j
c
ou
.202
3.10
2615
.
[
5]
N
.
Z
ha
ng
e
t
al
.
,
“
E
f
f
e
c
ts
of
v
a
r
io
us
a
r
ti
f
ic
ia
l
a
ga
r
w
ood
-
in
duc
ti
on
te
c
hni
que
s
on
th
e
me
ta
bol
ome
of
A
qui
la
r
ia
s
in
e
n
s
is
,”
B
M
C
P
la
nt
B
io
lo
gy
, vol
. 21, no. 1, 2021, doi
:
10.1186
/s
12870
-
021
-
0
3378
-
8.
[
6]
S
.
P
.
M
una
s
in
ghe
,
S
.
S
oma
r
a
tn
e
,
S
.
R
.
W
e
e
r
a
koon,
a
nd
C
.
R
a
na
s
in
ghe
,
“
E
c
ol
ogi
c
a
l
or
ig
in
of
th
e
a
ppe
a
r
a
nc
e
of
s
e
s
qui
te
r
pe
ne
s
in
G
yr
in
ops
w
a
ll
a
G
a
e
tn
e
r
by
w
ood
a
na
to
mi
c
a
l
a
nd
c
he
mi
c
a
l
a
na
ly
s
is
,”
J
ou
r
nal
of
th
e
I
ndi
an
A
c
ade
m
y
of
W
ood
Sc
ie
nc
e
,
vol
. 18, no. 2,
pp. 97
–
105, 2021, doi:
10.1007/s
13196
-
021
-
002
85
-
1.
[
7]
F
.
A
.
A
.
K
a
di
r
,
K
.
A
.
A
z
iz
a
n,
a
nd
R
.
O
th
ma
n,
“
D
a
ta
s
e
ts
o
f
e
s
s
e
nt
ia
l
oi
ls
f
r
om
na
tu
r
a
ll
y
f
or
me
d
a
nd
s
ynt
he
ti
c
a
ll
y
in
du
c
e
d
A
qui
la
r
ia
ma
la
c
c
e
ns
i
s
a
ga
r
w
oods
,
”
D
at
a i
n B
r
ie
f
, vol
. 28, 2020, doi:
10.1016/j
.d
ib
.2019.104987.
[
8]
H
.
E
.
S
ha
ia
n,
C
.
M
.
A
iz
a
l,
a
nd
S
.
N
.
T
a
ju
ddi
n,
“
C
he
mi
c
a
l
c
o
mpos
it
io
n
of
a
ga
r
w
ood
e
s
s
e
nt
ia
l
oi
l
(
A
qui
la
r
ia
ma
la
c
c
e
ns
i
s
)
u
pon
e
xpos
ur
e
to
w
a
r
ds
he
a
t
c
ondi
ti
on,”
M
al
ay
s
ia
n
J
our
nal
of
C
he
m
is
tr
y
,
vol
.
23,
no.
2,
pp.
199
–
204,
2021,
doi
:
10.55373/m
jc
he
m.v23i
2.1013
.
[
9]
M
.
H
.
A
bdul
la
h,
S
.
A
.
C
.
G
ha
ni
,
a
nd
S
.
R
.
M
oha
me
d,
“
E
xpe
r
im
e
nt
a
nd
nume
r
ic
a
l
s
tu
di
e
s
on
he
a
t
lo
s
s
in
a
hydr
o
-
di
s
ti
ll
a
ti
on
w
it
h
va
r
io
us
i
ns
ul
a
ti
ons
,”
M
A
T
E
C
W
e
b of
C
onf
e
r
e
nc
e
s
, vol
. 225, 20
18, doi:
10.1051/m
a
te
c
c
onf
/2
01822502014.
[
10]
S
.
M
.
H
.
M
.
H
uz
ir
e
t
al
.
,
“
S
te
pw
is
e
r
e
gr
e
s
s
io
n
of
a
ga
r
w
ood
oi
l
s
ig
ni
f
ic
a
nt
c
he
mi
c
a
l
c
ompounds
in
to
f
our
qua
li
ty
di
f
f
e
r
e
nt
ia
ti
on,”
I
ndone
s
ia
n
J
our
nal
of
E
le
c
t
r
ic
al
E
ngi
ne
e
r
in
g
and
C
o
m
put
e
r
Sc
ie
n
c
e
,
vol
.
29,
no.
2,
pp.
735
–
741,
2023,
doi
:
10.11591/i
je
e
c
s
.v29.i2.pp735
-
741.
[1
1]
J
.
D
ut
ta
,
D
.
L
a
hon,
P
.
K
.
B
or
a
,
M
.
B
huya
n,
a
nd
S
.
H
a
ld
a
r
,
“
A
hi
gh
pe
r
f
or
ma
nc
e
th
in
la
ye
r
c
hr
oma
to
gr
a
phy
(
H
P
T
L
C
)
me
th
od
f
or
th
e
qua
li
ty
a
s
s
e
s
s
m
e
nt
of
a
ga
r
w
ood
(
A
qui
la
r
ia
ma
la
c
c
e
ns
is
)
oi
l
f
r
om
N
or
th
e
a
s
t
I
ndi
a
,”
N
at
ur
al
P
r
oduc
t
R
e
s
e
ar
c
h
,
vol
.
38,
n
o.
9,
pp
. 1634
–
1638, 2024, doi:
10.1080/14786419.
2022.2162516.
[
12]
S
.
L
ia
s
,
N
.
A
.
M
.
A
li
,
M
.
J
a
mi
l,
M
.
H
.
Z
a
in
a
l,
A
.
M
.
J
a
li
l,
a
nd
N
.
I
s
ma
il
,
“
C
la
s
s
if
ic
a
ti
on
of
a
ga
r
w
ood
oi
ls
us
in
g
K
-
N
N
K
-
f
ol
d,”
Se
ns
or
L
e
tt
e
r
s
, vol
. 12, no. 6
–
7, pp. 1038
–
1040, 2014, doi:
10.1
166/
s
l.
2014.3187.
[
13]
M
. M
oha
ma
d
e
t
al
.
, “
kN
N
:
c
la
s
s
if
ic
a
ti
on of
a
ga
r
w
ood type
s
i
n oi
l
a
nd w
oode
n us
in
g e
-
nos
e
,”
i
n
T
he
6t
h I
nt
e
r
nat
io
nal
C
onf
e
r
e
nc
e
on E
le
c
tr
ic
al
, C
ont
r
ol
and C
om
put
e
r
E
ngi
ne
e
r
in
g
, 2022, pp. 57
5
–
586. doi:
10.1007/978
-
981
-
16
-
8690
-
0_52.
[
14]
Y.
C
he
n
e
t
al
.
,
“
S
tr
uc
tu
r
e
a
nd
c
he
mi
c
a
l
c
ha
r
a
c
te
r
is
ti
c
s
of
ly
uq
in
a
n
a
ga
r
w
ood
f
r
om
H
ui
dong,”
C
hi
ne
s
e
J
our
nal
of
W
ood
Sc
ie
nc
e
and T
e
c
hnol
ogy
, vol
. 36, no. 1, pp. 49
–
56, 2022, doi:
10.12326/j
.2096
-
9694.2021059.
[
15]
N
.
H
.
M
.
H
us
s
a
in
,
A
.
A
nua
r
,
S
.
M
a
s
r
om,
T
.
M
oh
d,
N
.
A
.
A
hma
d,
a
nd
H
.
B
yr
d,
“
M
a
c
hi
ne
le
a
r
ni
ng
in
r
e
ve
r
s
e
mi
gr
a
ti
on
c
la
s
s
if
ic
a
ti
on,”
J
our
nal
of
A
dv
anc
e
d
R
e
s
e
a
r
c
h
in
A
ppl
ie
d
Sc
ie
nc
e
s
and
E
ngi
ne
e
r
in
g
T
e
c
hnol
ogy
,
vol
.
38,
no.
2,
pp.
45
–
55,
2
024,
doi
:
10.37934/ar
a
s
e
t.
38.2.4555.
[
16]
D
.
T
.
A
hm
a
e
d,
A
.
M
.
M
a
s
a
a
d,
a
nd
S
.
N
.
T
a
ju
ddi
n
,
“
I
d
e
nt
if
i
c
a
t
io
n
of
a
ga
r
w
ood
(
A
qui
l
a
r
ia
m
a
la
c
c
e
n
s
i
s
)
c
h
ip
s
in
c
e
n
s
e
s
mo
ke
a
n
d
he
a
d
s
p
a
c
e
v
ol
a
t
il
e
c
om
poun
ds
by
gc
-
M
S
.E
I
.Q
.T
O
F
, S
P
M
E
,”
M
al
ay
s
ia
n
J
o
ur
nal
of
C
he
m
is
tr
y
, v
ol
.
20,
no.
2, p
p. 1
54
–
1
62,
2018
.
[
17]
P
.
M
oha
n
a
nd
K
.
P
a
ti
l,
“
D
e
e
p
le
a
r
ni
ng
ba
s
e
d
w
e
ig
ht
e
d
S
O
M
t
o
f
or
e
c
a
s
t
w
e
a
th
e
r
a
nd
c
r
op
p
r
e
di
c
ti
on
f
o
r
a
gr
ic
ul
tu
r
e
a
ppl
ic
a
ti
on,”
I
nt
e
r
nat
io
nal
J
our
nal
of
I
nt
e
ll
ig
e
nt
E
ngi
ne
e
r
in
g
and
Sy
s
te
m
s
,
vol
.
11,
no.
4,
pp.
167
–
176,
A
ug.
2018,
doi
:
10.22266/i
ji
e
s
2018.0831.17.
[
18]
C
.
Y
.
S
h
u
,
H
.
L
.
L
a
i
,
P
.
Y
.
L
i
n
,
K
.
C
.
C
.
C
h
o
u
,
a
n
d
M
.
S
h
e
l
o
m
i
,
“
P
r
e
l
i
m
i
n
a
r
y
s
t
u
d
i
e
s
o
n
t
h
e
e
f
f
e
c
t
s
o
f
T
a
i
w
a
n
e
s
e
r
i
t
u
a
l
s
m
o
k
e
o
n
A
e
d
e
s
a
e
g
y
p
t
i
(
L
i
n
n
a
e
u
s
,
1
7
6
2
)
(
D
i
p
t
e
r
a
:
C
u
l
i
c
i
d
a
e
)
,
”
P
a
n
-
P
a
c
i
f
i
c
E
n
t
o
m
o
l
o
g
i
s
t
,
v
o
l
.
9
7
,
n
o
.
2
,
p
p
.
6
7
–
74, 2021, doi:
10.3956/
2021
-
97.2.67.
[
19]
H
.
H
.
W
a
ng,
Y
.
C
.
W
a
ng,
B
.
L
.
W
e
e
,
a
nd
S
.
W
.
S
im
,
“
N
ove
l
f
e
a
tu
r
e
e
xt
r
a
c
ti
on
f
or
oi
l
pa
lm
bunc
h
e
s
c
la
s
s
if
ic
a
ti
on,”
J
ou
r
n
al
of
A
dv
anc
e
d
R
e
s
e
ar
c
h
in
A
ppl
ie
d
Sc
ie
nc
e
s
and
E
ngi
ne
e
r
i
ng
T
e
c
hnol
ogy
,
vol
.
34,
no.
1,
pp.
350
–
360,
2023,
doi
:
10.37934/ar
a
s
e
t.
34.1.350360.
[
20]
A
.
N
.
M
.
R
a
ml
i,
S
.
Y
u
s
of
,
P
.
B
huya
r
,
A
.
W
.
A
mi
na
n,
S
.
N
.
T
a
ju
ddi
n,
a
nd
H
.
A
.
H
a
mi
d,
“
P
r
oduc
ti
on
of
vol
a
ti
le
c
ompounds
b
y
a
va
r
ie
ty
of
f
ungi
in
a
r
ti
f
ic
ia
ll
y
in
oc
ul
a
te
d
a
nd
n
a
tu
r
a
ll
y
in
f
e
c
te
d
A
qui
la
r
ia
ma
la
c
c
e
ns
i
s
,”
C
u
r
r
e
n
t
M
ic
r
obi
ol
ogy
,
vol
.
79,
n
o.
5,
2022, doi:
10.1007/s
00284
-
022
-
02840
-
6.
[
21]
N
.
I
s
ma
il
,
N
.
A
.
M
.
A
li
,
M
.
J
a
mi
l,
M
.
H
.
F
.
R
a
hi
ma
n,
M
.
N
.
T
a
ib
,
a
nd
S
.
N
.
T
a
ju
ddi
n,
“
M
a
jo
r
vol
a
ti
le
c
h
e
mi
c
a
l
c
ompound
s
of
a
ga
r
w
ood
oi
ls
f
r
om
M
a
la
ys
ia
ba
s
e
d
on
z
-
s
c
or
e
te
c
hni
que
,”
C
h
e
m
is
tr
y
of
N
at
ur
al
C
om
pounds
,
vol
.
51,
no.
4,
pp.
776
–
779,
2015,
doi
:
10.1007/s
10600
-
015
-
1409
-
2.
[
22]
S
.
L
ia
s
,
A
.
M
.
J
a
li
l,
M
.
J
a
mi
l,
M
.
S
.
Y
.
T
ol
ma
na
n,
N
.
I
s
ma
il
,
a
nd
M
.
N
.
T
a
ib
,
“
T
h
e
a
ppl
ic
a
ti
on
of
e
le
c
tr
oni
c
nos
e
c
oupl
e
d
w
it
h
80:
20 k
-
ne
a
r
e
s
t
ne
ig
hbor
s
c
la
s
s
if
ic
a
ti
on t
e
c
hni
que
f
or
a
ga
r
w
oo
d oi
l
qua
li
ty
in
de
x e
s
ta
bl
is
hme
nt
,”
M
al
ay
s
ia
n J
our
nal
of
A
nal
y
t
ic
al
Sc
ie
nc
e
s
, vol
. 26, no. 6, pp. 1240
–
1248, 2022.
[
23]
R
.
K
a
lr
a
a
nd
N
.
K
a
us
hi
k, “
A
r
e
vi
e
w
of
c
he
mi
s
tr
y,
qu
a
li
ty
a
nd a
na
ly
s
is
of
in
f
e
c
te
d
a
g
a
r
w
ood
tr
e
e
(
A
q
ui
la
r
ia
s
p.)
,”
P
hy
to
c
he
m
i
s
tr
y
R
e
v
ie
w
s
, vol
. 16, no. 5, pp. 1045
–
1079, 2017, doi:
10.1007/s
11
101
-
017
-
9518
-
0.
[
24]
F
.
A
.
A
di
ta
ma
,
L
.
Z
u
lf
ik
r
i,
L
.
M
a
r
di
a
na
,
T
.
M
ul
ya
ni
ngs
ih
,
N
.
Q
oma
r
iy
a
h,
a
nd
R
.
W
ir
a
w
a
n,
“
E
le
c
tr
oni
c
nos
e
s
e
ns
or
de
ve
lo
pme
nt
us
in
g
A
N
N
b
a
c
kpr
opa
ga
ti
on
f
or
L
ombok
a
ga
r
w
ood
c
la
s
s
if
ic
a
ti
on,”
R
e
s
e
a
r
c
h
in
A
gr
ic
ul
tu
r
al
E
ngi
ne
e
r
in
g
,
vol
.
66,
no.
3,
pp.
97
–
103, 2020, doi:
10.17221/26/
2020
-
R
A
E
.
[
25]
E
.
H
a
s
li
na
A
bd
L
a
ti
b,
A
.
K
.
M
.
M
oye
e
nul
H
uq,
S
.
Z
a
r
e
e
n,
a
nd
S
.
N
iz
a
m
T
a
ju
ddi
n,
“
G
a
s
c
hr
oma
to
gr
a
phy
ma
s
s
s
pe
c
tr
om
e
tr
y
c
oupl
e
w
it
h
qua
dr
upol
e
ti
me
-
of
-
f
li
ght
(
G
C
-
Q
T
O
F
M
S
)
a
s
a
po
w
e
r
f
ul
to
ol
f
or
pr
of
i
li
ng
of
oxyge
na
te
d
s
e
s
qui
te
r
pe
ne
s
in
a
g
a
r
w
ood
oi
l,
”
A
r
abi
an J
our
nal
of
C
he
m
is
t
r
y
, vol
. 16, no. 9, 2023, doi
:
1
0.1016/j
.a
r
a
bj
c
.2023.105025.
[
26]
Z
.
W
u
a
nd
L
.
Y
u,
“
C
ha
r
a
c
te
r
is
ti
c
qua
li
ty
a
na
ly
s
i
s
f
or
bi
ol
ogi
c
a
ll
y
in
duc
e
d
a
ga
r
w
ood
c
ol
umns
in
A
qui
la
r
ia
s
in
e
n
s
is
,”
E
nv
ir
onm
e
nt
al
R
e
s
e
ar
c
h
, vol
. 235, 2023, doi:
10.1016/j
.e
nvr
e
s
.
2023.116633.
[
27]
N
. Z
. M
a
ha
bob
e
t
al
.
,
“
A
s
tu
dy on
A
N
N
pe
r
f
o
r
ma
nc
e
t
ow
a
r
ds
t
hr
e
e
s
ig
ni
f
ic
a
nt
c
ompound
s
of
h
ig
h qua
li
ty
a
ga
r
w
ood oil
,”
i
n
2
022
I
E
E
E
18t
h
I
nt
e
r
nat
io
nal
C
ol
lo
qui
um
on
Si
gnal
P
r
oc
e
s
s
i
ng
&
A
pp
li
c
at
io
ns
(
C
SP
A
)
,
I
E
E
E
,
2022,
pp.
116
–
120.
doi
:
10.1109/C
S
P
A
55076.2022.9782017.
[
28]
S
.
R
e
je
b,
C
.
D
uve
a
u,
a
nd
T
.
R
e
b
a
f
ka
,
“
S
e
lf
-
or
ga
ni
z
in
g
ma
ps
f
or
e
xpl
or
a
ti
on
of
pa
r
ti
a
ll
y
obs
e
r
ve
d
da
ta
a
nd
im
put
a
ti
on
of
mi
s
s
in
g
va
lu
e
s
,”
C
h
e
m
om
e
tr
i
c
s
and I
nt
e
ll
ig
e
nt
L
abor
at
or
y
S
y
s
te
m
s
, vol
. 231, 2022, doi:
10.1016/j
.c
he
mol
a
b.2022.104653.
[
29]
K
.
G
oe
l
a
nd
W
.
T
a
bi
b,
“
I
nc
r
e
me
nt
a
l
mul
ti
moda
l
s
ur
f
a
c
e
ma
p
pi
ng
vi
a
s
e
lf
-
or
ga
ni
z
in
g
ga
us
s
ia
n
mi
xt
ur
e
mode
ls
,”
I
E
E
E
R
ob
ot
ic
s
and A
ut
om
at
io
n L
e
tt
e
r
s
, vol
. 8, no. 12, pp. 8358
–
8365, 2023, doi:
10.1109/L
R
A
.2023.3327670.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
Ar
ti
f
I
ntell
,
Vol.
14,
No.
4:
Augus
t
2025
:
288
9
-
2898
2898
[
30]
C
.
L
i
,
C
.
X
ia
o
,
M
.
L
i,
L
.
X
u
,
a
nd
N
.
H
e
,
“
T
h
e
qu
a
l
it
y
a
nd
q
u
a
n
ti
ty
o
f
S
O
M
d
e
te
r
mi
ne
s
th
e
mi
n
e
r
a
li
z
a
ti
on
of
r
e
c
e
nt
l
y
a
d
d
e
d
la
bi
le
C
a
n
d
pr
i
mi
ng
of
n
a
t
iv
e
S
O
M
i
n
gr
a
z
e
d
gr
a
s
s
l
a
n
d
s
,
”
G
e
o
de
r
m
a
,
vo
l.
43
2,
A
pr
.
2
02
3,
d
oi
:
10
.1
01
6/
j
.g
e
od
e
r
m
a
.2
02
3.
11
63
85
.
[
31]
A
.
H
.
E
ls
he
ik
h
e
t
al
.
,
“
A
n
a
r
ti
f
ic
ia
l
ne
ur
a
l
ne
twor
k
b
a
s
e
d
a
p
pr
oa
c
h
f
or
pr
e
di
c
ti
on
th
e
th
e
r
ma
l
c
onduc
ti
vi
ty
of
na
nof
lu
id
s
,”
SN
A
ppl
ie
d Sc
ie
nc
e
s
, vol
. 2, no. 2, F
e
b. 2020, doi:
10.1007/s
42452
-
019
-
1610
-
1.
[
32]
J
.
P
ol
a
ns
ki
,
“
U
ns
up
e
r
vi
s
e
d
le
a
r
ni
ng
in
dr
ug
de
s
ig
n
f
r
om
s
e
lf
-
or
ga
ni
z
a
ti
on
to
de
e
p
c
he
mi
s
tr
y,”
I
nt
e
r
nat
io
nal
J
ou
r
nal
of
M
ol
e
c
u
la
r
Sc
ie
nc
e
s
, vol
. 23, no. 5, 2022, doi
:
10.3390/i
jm
s
23052797.
[
33]
S
.
L
i
c
e
n,
A
.
A
s
t
e
l,
a
n
d
S
.
T
s
a
k
ov
s
ki
,
“
S
e
lf
-
o
r
g
a
n
iz
in
g
m
a
p
a
lg
or
it
h
m
f
or
a
s
s
e
s
s
in
g
s
p
a
ti
a
l
a
n
d
t
e
m
por
a
l
pa
tt
e
r
n
s
of
p
ol
l
ut
a
nt
s
i
n
e
n
vi
r
on
me
nt
a
l
c
o
mp
a
r
t
m
e
nt
s
:
A
r
e
vi
e
w
,
”
S
c
i
e
n
c
e
of
t
h
e
T
o
ta
l
E
nv
i
r
on
m
e
nt
,
vo
l.
8
78
,
20
23
,
do
i:
10
.1
01
6/
j.
s
c
i
to
te
nv
.2
02
3.
16
30
84
.
[
34]
K
. S
a
xe
na
, S
. J
a
lo
r
e
e
, R
. S
. T
ha
kur
, a
nd S
. K
a
ml
e
y, “
S
e
lf
or
ga
ni
z
in
g ma
p (
S
O
M
)
ba
s
e
d mode
ll
in
g t
e
c
hni
que
f
or
s
tu
de
nt
a
c
a
d
e
mi
c
pe
r
f
or
ma
nc
e
pr
e
di
c
ti
on,”
I
nt
e
r
nat
io
nal
J
our
nal
on F
ut
ur
e
R
e
v
ol
ut
io
n i
n C
om
put
e
r
Sc
ie
nc
e
& C
om
m
uni
c
at
io
n E
ngi
ne
e
r
in
g
, vol
. 3,
no. 9, pp. 115
–
120, 2017.
[
35]
O
.
M
ur
a
d
a
nd
M
.
M
a
lk
a
w
i,
“
A
n
opt
im
iz
a
ti
on
me
th
od
u
s
in
g
c
lu
s
te
r
in
g
te
c
hni
que
f
or
th
e
huma
n
e
mot
io
ns
de
te
c
ti
on
a
r
ti
f
ic
ia
l
ne
ur
o
-
f
uz
z
y
lo
gi
c
s
ys
te
m,”
I
nt
e
r
nat
io
nal
J
our
nal
of
C
om
put
e
r
s
&
T
e
c
hnol
ogy
,
vol
.
15,
no.
9,
pp.
7090
–
709
6,
2016,
doi
:
10.24297/i
jc
t.
v15i
9.695.
B
I
OG
RA
P
HI
E
S
OF
AU
T
HO
RS
M
o
ha
m
m
a
d
A
r
i
f
F
a
hm
i
C
he
H
a
s
s
a
n
w
as
b
o
rn
i
n
Mal
ay
s
i
a,
o
n
May
1
9
9
6
.
H
e
recei
v
e
d
h
i
s
Bach
e
l
o
r
o
f
Ch
em
i
cal
E
n
g
i
n
ee
ri
n
g
T
ec
h
n
o
l
o
g
y
(H
o
n
s
)
(Bi
o
t
ec
h
n
o
l
o
g
y
In
d
u
s
t
r
y
)
fro
m
U
n
i
v
er
s
i
t
i
Mal
a
y
s
i
a
Perl
i
s
.
H
e
i
s
cu
rre
n
t
l
y
en
g
i
n
e
er
at
V
i
s
i
o
n
In
d
u
s
t
ri
e
s
an
d
at
t
h
e
s
ame
t
i
me
as
fu
l
l
-
t
i
me
p
o
s
t
g
rad
u
at
e
s
t
u
d
en
t
s
at
Sch
o
o
l
o
f
E
l
ect
r
i
cal
E
n
g
i
n
eer
i
n
g
,
Co
l
l
e
g
e
o
f
E
n
g
i
n
eeri
n
g
,
U
n
i
v
er
s
i
t
i
T
ek
n
o
l
o
g
i
MA
R
A
,
U
i
T
M
Sh
ah
A
l
am,
Mal
ay
s
i
a.
H
e
can
b
e
co
n
t
ac
t
e
d
at
emai
l
:
A
r
i
ffah
m
i
5
3
3
@
g
ma
i
l
.
co
m.
A
s
s
o
c.
P
ro
f.
Ts
.
D
r.
Za
k
i
a
h
M
o
hd
Y
u
s
o
ff
i
s
a
s
en
i
o
r
l
ec
t
u
rer
w
h
o
i
s
cu
rre
n
t
l
y
w
o
r
k
i
n
g
at
U
i
T
M
Pas
i
r
G
u
d
a
n
g
.
Sh
e
recei
v
ed
t
h
e
B.
E
n
g
.
i
n
E
l
ec
t
ri
ca
l
E
n
g
i
n
eeri
n
g
an
d
Ph
.
D
.
i
n
E
l
e
ct
r
i
cal
E
n
g
i
n
eer
i
n
g
fro
m
U
i
T
M
Sh
ah
A
l
am,
i
n
2
0
0
9
an
d
2
0
1
4
,
res
p
ec
t
i
v
el
y
.
In
May
2
0
1
4
,
s
h
e
j
o
i
n
e
d
U
i
T
M
Pas
i
r
G
u
d
an
g
as
a
t
eac
h
i
n
g
s
t
af
f.
H
er
maj
o
r
i
n
t
ere
s
t
s
i
n
c
l
u
d
e
p
r
o
ces
s
co
n
t
ro
l
,
s
y
s
t
em
i
d
e
n
t
i
fi
ca
t
i
o
n
,
an
d
es
s
e
n
t
i
al
o
i
l
ex
t
ract
i
o
n
s
y
s
t
em.
Sh
e
can
b
e
co
n
t
ac
t
ed
a
t
emai
l
:
zak
i
a
h
9
0
1
8
@
u
i
t
m.
ed
u
.
my
.
A
s
s
o
c.
P
ro
f.
I
r.
Ts
.
D
r.
N
url
a
i
l
a
I
s
m
a
i
l
recei
v
ed
h
er
Ph
.
D
.
i
n
E
l
ect
r
i
cal
E
n
g
i
n
eeri
n
g
fr
o
m
U
n
i
v
ers
i
t
i
T
e
k
n
o
l
o
g
i
M
A
RA
,
Mal
a
y
s
i
a.
Sh
e
i
s
c
u
rren
t
l
y
a
s
e
n
i
o
r
l
ec
t
u
r
er
a
t
Sch
o
o
l
o
f
E
l
ect
r
i
cal
E
n
g
i
n
eer
i
n
g
,
Co
l
l
e
g
e
o
f
E
n
g
i
n
e
eri
n
g
,
U
n
i
v
er
s
i
t
i
T
e
k
n
o
l
o
g
i
MA
R
A
,
Mal
ay
s
i
a.
H
er
res
earch
i
n
t
ere
s
t
s
i
n
c
l
u
d
e
ad
v
an
ce
d
s
i
g
n
a
l
p
ro
ce
s
s
i
n
g
an
d
art
i
fi
c
i
al
i
n
t
e
l
l
i
g
e
n
ce.
Sh
e
can
b
e
co
n
t
ac
t
ed
a
t
emai
l
:
n
u
rl
a
i
l
a
0
5
8
3
@
u
i
t
m.
e
d
u
.
my
.
P
ro
f.
I
r.
Ts
.
D
r.
M
o
hd
N
a
s
i
r
T
a
i
b
recei
v
e
d
h
i
s
Ph
.
D
.
fro
m
U
MIST
,
U
K
.
H
e
i
s
a
Sen
i
o
r
Pro
fe
s
s
o
r
a
t
U
n
i
v
ers
i
t
i
T
ek
n
o
l
o
g
i
M
A
RA
(
U
i
T
M).
H
e
h
ead
s
t
h
e
A
d
v
an
ce
d
S
i
g
n
a
l
Pro
ces
s
i
n
g
Res
earch
G
ro
u
p
at
t
h
e
Sch
o
o
l
o
f
E
l
ec
t
ri
cal
E
n
g
i
n
eeri
n
g
,
Co
l
l
eg
e
o
f
E
n
g
i
n
eeri
n
g
,
U
i
T
M.
H
e
h
as
b
een
a
v
ery
act
i
v
e
res
earch
er
an
d
o
v
er
t
h
e
y
ears
h
a
d
au
t
h
o
r
an
d
/
o
r
co
-
au
t
h
o
r
man
y
p
a
p
ers
p
u
b
l
i
s
h
ed
i
n
refereed
j
o
u
rn
a
l
s
an
d
co
n
fer
en
ces
.
H
e
can
b
e
co
n
t
ac
t
ed
at
ema
i
l
:
d
r.
n
as
i
r@
u
i
t
m.
ed
u
.
my
.
Evaluation Warning : The document was created with Spire.PDF for Python.