Int
ern
at
i
onal
Journ
al of Ele
ctrical
an
d
Co
mput
er
En
gin
eeri
ng
(IJ
E
C
E)
Vo
l.
9
, No
.
6
,
Decem
ber
201
9
, p
p.
5537~
5544
IS
S
N: 20
88
-
8708
,
DOI: 10
.11
591/
ijece
.
v9
i
6
.
pp5537
-
55
44
5537
Journ
al h
om
e
page
:
http:
//
ia
es
core
.c
om/
journa
ls
/i
ndex.
ph
p/IJECE
Estimati
on of fin
es amou
nt in sy
ariah
crimi
nal offences us
ing
adapti
ve neuro
-
f
uzzy infe
re
n
ce sy
stem (A
NFIS)
en
hanced
with an
alytic
hie
rarchy
process
(AHP)
Ah
m
ad
Fitri
Ma
z
lam,
Wan
Nu
r
al
Ja
w
ah
i
r Hj W
an
Yu
s
so
f
,
Rabiei
M
am
at
School
of
In
for
m
at
ic
s a
nd
app
lied
Math
ematics,
Univer
sit
i
Mal
a
y
sia
T
ere
ngg
anu
,
Mal
a
y
s
ia
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
Feb
15
, 201
9
Re
vised
A
pr
18
, 2
01
9
Accepte
d
J
un
1
0
, 201
9
All
s
y
ariah
cri
m
ina
l
ca
ses
,
espe
c
ia
lly
in
khal
w
at
offe
nce
hav
e
th
e
ir
c
ase
-
fa
ct,
and
th
e
judg
es
t
y
pi
call
y
look
for
ward
to
al
l
th
e
f
ac
ts
whi
ch
wer
e
ta
bul
at
ed
b
y
the
prosec
utors.
A
var
ie
t
y
of
cri
t
eri
a
is
conside
r
e
d
b
y
the
judg
e
t
o
det
ermine
the
fin
es
amoun
t
t
ha
t
should
b
e
impos
ed
on
an
ac
cuse
d
who
pl
ea
ds
guilt
y
.
In
Te
ren
gg
anu,
t
her
e
were
te
n
(1
0)
judge
s,
and
th
e
judgments
wer
e
m
ade
b
y
the
indi
v
idual
d
ec
ision
upon
th
e
trial
to
d
ecide
the
c
ase
.
Each
judge
has
a
stake,
principl
es
and
disti
nc
tive
criter
i
a
in
de
t
ermining
fin
es
amount
on
an
a
cc
used
who
ple
ads
gu
il
t
y
a
nd
convi
c
te
d
.
T
his
rese
ar
ch
pap
er
pre
sen
ts
an
Adapti
v
e
Ne
uro
-
fuz
z
y
Inf
erence
S
y
st
em
(AN
FIS
)
te
chni
que
combining
with
Anal
y
t
ic
Hier
arc
h
y
Pro
c
ess
(AH
P)
for
esti
m
at
ing
f
ine
s
amount
in
S
y
ariah
(kh
al
wa
t)
cr
iminal
.
Dat
a
sets
were
col
l
ect
ed
under
the
supervision
of
reg
istra
r
and
s
yar
ie
judge
in
th
e
Depa
r
tm
ent
o
f
S
y
ar
ia
h
Judi
ciar
y
State
o
f
Te
ren
gg
anu,
Ma
lay
s
ia.
The
resul
ts
show
ed
tha
t
AN
FIS
+A
HP
co
uld
esti
m
at
e
fine
s e
ff
icientl
y
tha
n
the t
rad
it
io
nal
m
et
hod
with
a
ver
y
m
ini
m
al
e
rror.
Ke
yw
or
d
s
:
Ad
a
ptive
n
eu
r
o
f
uzzy
i
nf
e
re
nce
s
yst
e
m
(
AN
F
I
S)
Am
ou
nt
of
f
in
es
A
naly
ti
c
h
ie
ra
r
chy
p
r
ocess
(AHP
)
Syari
ah
c
rim
in
al
Copyright
©
201
9
Instit
ut
e
o
f Ad
vanc
ed
Engi
n
ee
r
ing
and
S
cienc
e
.
Al
l
rights re
serv
ed
.
Corres
pond
in
g
Aut
h
or
:
Wan N
ural
Ja
wah
i
r Hj
W
a
n Yu
s
sof,
School
of In
for
m
at
ic
s an
d Appli
ed
Ma
them
at
ic
s,
Un
i
ver
sit
i M
al
ay
sia
Tereng
ga
nu,
21030 K
uala
N
eru
s
, Te
reng
ga
nu D
a
r
ul I
m
an,
Mal
ay
sia
.
Em
a
il
: wan
nur
wy@
um
t.edu
.
m
y
1.
INTROD
U
CTION
Decisi
on
m
aki
ng
is
the
proc
ess
of
helpi
ng
so
m
eon
e
so
l
ve
s
the
pro
ble
m
by
assessi
ng
al
te
rn
at
ive
reso
l
ution.
It
is
a
m
and
at
ory
act
ivit
y
in
real
l
ife
wh
ic
h
in
volves
se
ver
al
ste
ps
incl
ud
i
ng
i
de
ntifyi
ng
t
he
r
esults
and
gat
her
i
ng
inf
or
m
at
ion
.
T
ypic
al
ly
,
m
any
unce
rtai
nties
and
inacc
ur
at
e
data
a
nd
crit
e
ria
are
us
e
d
t
o
m
ak
e
decisi
ons.
S
om
et
i
m
es,
it
se
e
m
s
natur
al
to
decide,
bu
t
wh
e
n
it
com
e
s
to
m
ulti
-
crit
eria
to
be
co
nsi
der
e
d,
the d
eci
si
on
-
m
akin
g process
beco
m
es a fuz
zy
task.
Since
de
ci
sion
-
m
aking
can
be
reg
a
rded
as
hum
an
need
s
,
th
ere
are
te
c
hn
i
ques
us
ed
by
pe
op
le
to
deal
with
the
pro
bl
e
m
s.
Trad
it
io
na
ll
y,
people
al
ways
hold
m
eet
ing
s
betwee
n
ex
per
ts,
vote
for
m
ajo
rity
de
ci
sion
s
and
j
um
p
to
a
con
cl
usi
on
to
en
d
the
pro
cess.
This
pro
cess
is
so
m
ew
hat
su
bject
ive
.
Ther
e
fore
,
m
od
e
r
n
m
et
ho
ds
of
de
ci
sion
m
aking
hav
e
been
i
nt
rodu
ce
d
ra
pidl
y
ov
er
se
ver
a
l
decad
es.
In
1992,
[
1]
disc
us
se
d
sever
al
m
et
hods o
f
m
ulti
-
crit
eria decisi
on m
akin
g
a
nd they
updated
the
art
ic
le
in
2008
[2].
A
da
ptive
Neuro
-
F
uzzy
I
nf
e
ren
ce
Syst
e
m
(ANF
IS
)
m
et
hod
i
s
one
of
t
he
m
os
t
widely
use
d
appr
oach
es
t
o
handle
the
m
ulti
-
crit
eria
decis
ion
-
m
aking
pr
ocess.
A
NF
I
S
is
a
fu
zzy
infe
r
ence
syst
em
that
has
been
us
e
d
in
nu
m
erous
ar
eas
of
resea
rc
h
s
uch
a
s
dia
gnos
is
[3
-
7],
m
od
el
li
ng
[8
]
and
predict
io
n
[
9]
.
In
dia
gnos
is
a
pp
li
cat
io
n,
A
N
FI
S
was
em
plo
ye
d
f
or
ris
k
diag
nosis
ris
k
in
de
ngue
patie
nts
[
3].
The
us
e
of
the
A
NFIS
f
or
the
diag
no
si
s
of
m
al
aria
has
been
perf
or
m
ed
by
[
4]
to
pro
vid
e
be
tt
er
decisi
on
than
the
tradit
ion
al
diag
nosis
of
m
et
hods
cha
ract
erized
by
er
otic
gu
ess
w
ork
a
n
d
patie
nt
obs
erv
at
io
ns
by
doct
ors.
This
w
ork
ach
ie
ves
a
ver
y
cl
os
e
res
ult
to
the
exp
ect
at
io
n
of
th
e
resea
rch
e
rs
with
a
ver
y
m
ini
m
al
error.
Anothe
r
dia
gnos
is
w
orks
us
ing
ANFI
S
was
pr
ese
nted
i
n
[
5
-
7],
as
a
base
for
hype
rtensi
on
diag
nosis
and
f
or
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
9
, N
o.
6
,
Dece
m
ber
2
01
9
:
5537
-
5544
5538
hear
t
diseas
e
diag
nosis,
res
pe
ct
ively
.
Be
sides
diag
nosis,
ANFIS
m
od
el
was
us
e
d
to
f
or
ecast
the
return
on
stock
pri
ce
in
de
x
of
the
Istan
bu
l
Stoc
k
E
xc
hange
(ISE)
[
8]
and
predict
t
he
fu
t
ur
e
act
io
ns
of
the
e
xc
ha
ng
e
rate [
9].
Me
anwhil
e,
A
naly
ti
c
Hierar
chy
Process
(
AHP)
is
a
m
et
hod
to
fi
nd
a
weig
ht
betw
een
facto
rs.
AHP
m
easur
es
al
l
the
factor
s
i
nvolv
e
d
i
n
the
decisi
on
-
m
aking
pro
cess
thr
ough
pair
wise
com
par
is
on.
The
c
om
par
iso
n
is
co
nducte
d
us
in
g
a
scal
e
of
abs
olu
te
j
ud
gem
ent
that
repr
esents
ho
w
vit
al
on
e
factor
t
o
on
e
ano
t
her
an
d
it
s
al
so
been
us
e
d
by
resea
rch
e
rs
in
m
any
fiel
ds
a
nd
resea
rc
hes.
[
10
]
s
how
s
an
im
ple
m
entat
ion
of
A
HP
in
ide
ntifyi
ng
t
he
fa
ct
or
s
as
so
ci
at
e
d
wit
h
obesi
ty
.
The
res
ult
indi
cat
ed
that
the
facto
r
of
Se
de
ntary
Lifest
yl
e
(S
E
D)
co
ntribute
s
the
highest
weig
ht
com
pa
red
to
Me
dic
al
Psychia
tric
Illness
(ME
D)
an
d
Gen
et
ic
s
(GEN
).
Ba
sed
on
t
he
stren
gth
of
ANFIS
an
d
A
HP
repor
te
d
i
n
the
li
te
ratu
r
e,
this
pa
pe
r
pr
ese
nts
how
ANFIS
+
A
HP
cou
l
d
help
sya
rie
ju
dg
e
to
est
i
m
at
e
fines
am
ount
to
the
kha
lwat
offe
nces
base
d
on
fi
ve
crit
eri
a
befor
e
tria
l
without
interf
erin
g
of
influ
e
nce
jud
ic
ia
l
decisi
on.
T
hese
crit
eri
a
are
cho
s
en
be
cause
the
re
ar
e
on
ly
five
crit
eria
th
at
hav
e
been
no
te
d
in
the
c
ase
fact
ta
bu
la
te
d
by
the
e
nfor
cem
ent
officer
f
or
prosec
ut
ion.
Data
that
has
been
c
ol
le
ct
ed
is
a
co
m
bin
at
ion
of
diff
e
re
nt
j
udge
s
base
d
on
five
crit
eria
.
In
c
urren
t
pract
ic
e,
so
m
e
j
udges
e
stim
at
e
a
m
ea
gr
e
am
ount
of
fines
wh
il
e
s
om
e
j
udges
put
a
ve
ry
high
a
m
ou
nt
of
fines
f
or
the
sam
e
crit
e
ria.
I
n
the
pr
evio
us
stu
dy
[11],
we
on
ly
i
m
ple
m
ented
AN
F
IS
for
the
sam
e
case
stud
y.
Ther
e
f
or
e,
this
pap
e
r
ai
m
s to
im
pr
ov
e
pr
e
vious
wor
k by inc
orp
or
at
in
g ANFIS wit
h A
HP
.
The
rem
ai
nd
er
of
t
his
pa
per
i
s
orga
nized
as
fo
ll
ows:
Sect
io
n
2
disc
us
ses
on
t
he
ba
sic
de
finiti
on
f
or
ANFIS
as
a
m
et
hod
us
ed
f
or
est
i
m
ating
f
i
ne
s
am
ou
t.
Sect
ion
3
def
i
nes
a
si
m
ple
ste
p
f
or
AHP.
A
f
ram
ewor
k
for
est
i
m
at
ion
fines
am
ou
nt
us
in
g
A
NFI
S+A
HP
is
pr
esented
i
n
Se
ct
ion
4.
Ex
pe
rim
ental
resu
lts
and
discuss
i
on are
ta
bu
la
te
d i
n Se
ct
ion
5
a
nd f
i
na
ll
y, Sect
ion
6 wil
l con
cl
ude t
he pape
r.
2.
METHO
DO
L
OGY
In
this
pa
per
,
75
cases
we
re
trai
ne
d,
a
nd
30
cases
we
re
te
ste
d
with
A
NF
I
S
s
olu
ti
on.
The
n
A
HP
so
luti
on
is
use
d
to
determ
ine
the
wei
gh
t
age
an
d
finall
y
the
AN
F
IS+
AHP
so
l
utio
n
is
us
e
d
to
gen
e
rate
a n
e
w result.
2.1.
AN
F
IS
s
olu
tio
n
This
stu
dy
us
e
d
dataset
s
fro
m
the
Dep
art
m
ent
of
Syari
ah
Judic
ia
ry
Stat
e
of
Tere
ngga
nu
wh
ic
h
is
an
insti
tuti
on
that
has
Appeal
Court,
High
C
ourt
an
d
S
ub
Ordina
ry
Cour
t.
Ob
se
r
vation,
li
te
ratur
e
surv
ey
and
intervie
w
we
re
us
e
d
to
gath
e
r
in
form
at
ion
about
the
accu
sed
per
s
on
who
pleade
d
guil
ty
and
fines
a
m
ou
nt
charge
d
to
t
ha
t
per
s
on.
I
n
t
hi
s
stud
y,
75
da
ta
set
s
cases
s
entence
d
by
fi
nes
wer
e
c
ollec
te
d.
Ta
ble
1
li
sts
the
at
trib
utes
of
K
halw
at
data
set
wh
il
e
Ta
bl
e
2
li
sts
the
va
lue
of
se
x,
m
a
rita
l
sta
tus,
loc
at
ion
t
ype
of
a
rr
est
and
the
ti
m
e
of
ar
rest.
The
figure
of
a
ge
and
the
fines
a
m
ou
nt
is
us
ed
as
the
valu
e
fo
r
ag
e
and
fine
s
at
tribu
te
s
acco
rd
i
ng
ly
.
Ta
ble
3
sho
ws
the
ra
w
dataset
that
had
bee
n
colle
ct
ed
an
d
the
se
t
of
data
that
ha
s
be
e
n
conve
rted
i
nto
value
is
sh
own
in
Ta
ble
4.
Table
1.
Attrib
utes
of
khal
wat
off
e
nces
datas
et
Ab
b
reviatio
n
Decsripti
o
n
Rep
resentatio
n
of
Fu
zzy
V
ariables
ag
e
Ag
e in y
e
ars
1
sex
Sex
(
m
a
le; f
e
m
a
le)
2
m
a
rstatu
s
Mar
ital
statu
s (div
o
rced,
m
a
rr
i
ed
,
b
achelo
r)
3
lo
catio
n
Locatio
n
of
cri
m
e
(ho
tel,
resid
en
ce,
clo
sed
ar
ea,
op
en
ar
ea)
4
ti
m
e
Ti
m
e
of
ar
rest
Day
(
7
.00
a
m
–
7
.
0
0
p
m
)
Nig
h
t (
7
.01
p
m
–
1
2
.00
a
m
)
Ear
l
y
Morn
in
g
(
1
2
.01
a
m
–
6
.59
a
m
)
5
f
in
es
Fin
es a
m
o
u
n
t
Table
2.
Value
for
s
ex
, m
arit
a
l st
at
us
, lo
c
at
io
n
a
nd tim
e
v
alu
e
sex
m
a
rstatu
s
lo
catio
n
ti
m
e
1
Fe
m
ale
Sin
g
le
Op
en
Ar
ea
Day
2
Male
Mar
ried
Clo
sed
Ar
ea
Nig
h
t
3
Div
o
rced
Res
id
en
ce
Ear
l
y
Morn
in
g
4
Ho
tel
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N: 20
88
-
8708
Esti
ma
ti
on
of f
ines
amo
un
t i
n syaria
h
c
rimin
al o
ff
en
c
es
us
i
ng ad
ap
ti
ve
ne
ur
o
-
fuz
z
y
.
..
(
A
hma
d
Fit
ri M
azl
am
)
5539
Table
3.
Ra
w d
at
aset
f
or
kh
al
wat offe
nces
Cas
e No
ag
e
sex
m
ar
statu
s
lo
catio
n
ti
m
e
1
1
0
0
3
-
143
-
0
0
1
2
-
2
0
1
5
17
f
e
m
a
le
b
achelo
r
resid
en
ce
early
m
o
rnin
g
1
1
0
0
8
-
143
-
0
0
0
1
-
2
0
1
5
28
m
a
le
m
a
r
ried
h
o
tel
n
ig
h
t
1
1
0
0
3
-
143
-
0
0
2
4
-
2
0
1
5
59
m
a
le
d
iv
o
rced
resid
en
ce
early
m
o
rnin
g
1
1
0
0
5
-
143
-
0
1
0
7
-
2
0
1
5
37
f
e
m
a
le
d
iv
o
rced
resid
en
ce
early
m
o
rnin
g
…
…
…
…
…
…
1
1
0
0
9
-
143
-
0
0
2
8
-
2
0
1
5
28
f
e
m
a
le
d
iv
o
rced
clo
sed
ar
ea
early
m
o
rnin
g
Table
4.
C
onve
rted dat
a set ac
cordin
g
t
o valu
e d
e
fine
d
in
Ta
ble 3
Cas
e No
ag
e
sex
m
a
r
statu
s
lo
catio
n
ti
m
e
1
1
0
0
3
-
143
-
0
0
1
2
-
2
0
1
5
17
1
1
3
3
1
1
0
0
8
-
143
-
0
0
0
1
-
2
0
1
5
28
2
2
4
2
1
1
0
0
3
-
143
-
0
0
2
4
-
2
0
1
5
59
2
3
3
3
1
1
0
0
5
-
143
-
0
1
0
7
-
2
0
1
5
37
1
3
3
3
…
…
…
…
…
…
1
1
0
0
9
-
143
-
0
0
2
8
-
2
0
1
5
28
1
3
2
3
Ba
sed
on
the
data
colle
ct
ed
and
AN
F
IS
pa
ram
et
ers
need
,
ru
le
s
are
ge
ne
rated
us
i
ng
nu
m
erical
data
introd
uced
by
[1
2
]
.
These
ru
l
es
will
be
us
e
d
in
the
la
ye
rs
in
AN
F
IS.
Table
5
is
a
set
of
data
that
ha
s
been
so
rte
d by age
.
2.1.1.
Genera
tin
g
ru
le
s using num
eri
cal dat
a
In g
e
ner
at
in
g
t
he rules,
ei
ght
ste
ps
hav
e
to
be
conside
red
:
Step
1
:
Id
e
ntify t
he n
um
ber
o
f data
c
ollec
te
d,
=
75
Step
2
:
Id
e
ntify t
he n
um
ber
o
f
at
tri
bute
s,
= 6
(ag
e
, se
x,
m
arstat
us
, l
ocati
on, tim
e, f
ines)
Step
3
:
Ca
lc
ulate
the total n
um
ber
of
ru
le
s,
=
=
75
6
=
12
Step
4
:
Data are
sorted
acco
rd
i
ng to
a
ge
as
sho
wn in
Table
4.
Step
5
:
The
m
axi
m
u
m
v
al
ue
is
ide
ntif
ie
d
f
or each
va
riable
Step
6
:
The fuzzy
num
ber
s
in Ta
ble
4 f
or
m
s the an
te
ceden
ts
and c
onse
qu
e
nt
par
ts
of the
ru
le
.
Step
7
:
Rule 1
is f
ram
e
d
as
If
(
1
is 1
7) or
(
2
is fem
al
e) o
r (
3
is bac
helo
r) or
(
4
is h
otel)
or (
5
is early
m
or
ning)
then
is
2.7.
Step
8
:
To
fr
am
e
the
nex
t
r
ule
,
the
nex
t
6
data
are
ta
ken
an
d
the
4
th
and
5
th
ste
ps
are
rep
eat
ed
a
nd
gen
e
rated
rule
1
a
re show
n
in
Table
5.
Table
5.
First f
ram
e o
f
data
to
g
e
ner
at
e
ru
le
1
Cas
e No
ag
e
sex
m
ar
statu
s
lo
catio
n
ti
m
e
f
in
es
1
1
0
0
9
-
143
-
0
0
1
6
-
2
0
1
5
15
1
1
3
3
2
.0
1
1
0
0
4
-
143
-
0
1
2
0
-
2
0
1
4
15
1
1
3
3
2
.6
1
1
0
0
3
-
143
-
0
0
1
2
-
2
0
1
5
17
1
1
3
3
2
.3
1
1
0
0
6
-
143
-
0
0
7
2
-
2
0
1
5
17
1
1
4
2
2
.5
1
1
0
0
6
-
143
-
0
0
7
3
-
2
0
1
5
17
1
1
4
2
2
.5
1
1
0
0
5
-
143
-
0
1
4
1
-
2
0
1
5
17
1
1
4
2
2
.7
Maxi
m
u
m
valu
es f
o
r
f
u
zzy
v
ariables
Ru
le 1
17
1
1
4
3
2
.7
2.1.2.
Predi
cting
out
put
vari
abl
e u
sing
li
ne
ar
eq
ua
ti
on
Ba
sed on 1
2 rul
es f
ram
ed
earl
ie
r,
ou
t
pu
t
var
i
able,
f
or
eac
h ru
le
e
quat
ion i
s the
n
pre
dicte
d
f
or li
ne
a
r
equ
at
io
n
a
s fol
low:
=
1
1
+
2
2
+
3
3
+
4
4
+
5
5
+
(1)
w
hile
par
am
et
e
r
is p
red
ic
te
d
us
in
g
t
he
li
nea
r
e
qu
at
io
n usi
ng the
for
m
ula intr
oduce
d by [
1
2
]
as
foll
ow:
=
(
)
(
)
+
(
)
(
)
(
)
(
.
)
(2)
wh
e
re
j
=
1,2,3
,4
,
5
a
nd
i
=
1,2
,3
…
11,
12.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
9
, N
o.
6
,
Dece
m
ber
2
01
9
:
5537
-
5544
5540
The
pre
dicte
d
fu
zzy
val
ues
a
nd
ge
ner
at
e
d
r
ules
are
us
ed
i
n
A
NFIS
m
et
ho
d
t
o
est
i
m
at
e
fines
am
ou
nt
are
pr
ese
nted
in
Ta
ble
6.
For
exam
ple:
Rule
1:
If
(
1
is
young)
or
(
2
is
fem
al
e)
or
(
3
is
bach
el
or)
or
(
4
is
hote
l)
or
(
5
is
earlym
or
n
in
g)
then
is
high,
1
=
1
1
1
+
1
2
2
+
1
3
3
+
1
4
4
+
1
5
5
+
1
.
=
(
)
(
)
+
(
)
(
)
(
)
(
.
)
=
2
.
4
17
+
2
.
4
16
(
2
.
4
)
(
5
)
=
0
.
02
(3)
Table
6.
Pr
e
dic
te
d
f
uzzy
value
s for
f
or Rule
1
ag
e
1
sex
2
m
a
r
statu
s
3
lo
catio
n
4
ti
m
e
5
(
)
2
.4
(
)
17
1
1
4
3
(
)
16
1
1
4
3
n
o
.
o
f
attr
ib
u
tes
5
(
)
(
)
+
(
)
(
)
(
)
(
.
)
1
1
1
2
1
3
1
4
1
5
0
.02
0
.40
0
.40
0
.11
0
.15
F
or
al
l
r
ules
a
r
e
cal
culat
ed
usi
ng
li
near
eq
ua
ti
on
w
it
h
the val
ue
f
or
eac
h
r
ule
f
ram
ed.
F
or
e
xam
ple,
1
is cal
culat
ed
as
foll
ow
s:
1
=
1
−
1
1
1
+
1
2
2
+
1
3
3
+
1
4
4
+
1
5
5
1
=
2
.
7
−
(
(
0
.
02
)
(
17
)
+
(
0
.
40
)
(
1
)
+
(
0
.
40
)
(
1
)
+
(
0
.
11
)
(
4
)
+
(
0
.
15
)
(
3
)
)
(4)
1
=
0
.
67
The
val
ues for
Rule
1
to
Rule
12 a
re
pr
ese
nt
ed
in
Ta
ble 7.
Table
7.
Pr
e
dic
te
d
f
uzzy
value
for
,
1
2
3
4
5
Ru
le 1
0
.02
0
.4
0
.4
0
.11
0
.15
0
.67
Ru
le
2
0
.02
0
.23
0
.4
0
.13
0
.15
0
.51
Ru
le 3
0
.02
0
.21
0
.4
0
.11
0
.14
0
.72
Ru
le 4
0
.02
0
.27
0
.22
0
.13
0
.17
0
.33
Ru
le 5
0
.02
0
.25
0
.2
0
.1
0
.15
0
.59
Ru
le 6
0
.02
0
.2
0
.4
0
.11
0
.14
0
.82
Ru
le 7
0
.01
0
.22
0
.17
0
.11
0
.15
0
.88
Ru
le 8
0
.01
0
.22
0
.15
0
.11
0
.17
0
.85
Ru
le 9
0
.01
0
.23
0
.17
0
.11
0
.14
0
.84
Ru
le 10
0
.01
0
.25
0
.14
0
.13
0
.15
0
.89
Ru
le 11
0
.01
0
.22
0
.15
0
.11
0
.14
0
.68
Ru
le 12
0
.01
0
.22
0
.14
0
.11
0
.13
0
.72
2.2.
A
HP
so
lu
tio
n
In
the
stu
dy,
fi
ve
crit
eria
invo
lved
to
est
im
a
t
e
the
fines
am
o
un
t.
Howe
ver,
j
ud
ges
al
so
ha
ve
dif
fer
e
nt
thoughts
reg
a
r
ding
w
hic
h
do
m
ai
n
factor
c
ontrib
ute
to
t
he
order.
T
his
si
tuati
on
pro
ved
in
[
11
]
wh
e
re
so
m
e
cases
that
hav
e
the
sa
m
e
values
sh
ow
di
ff
e
re
nt
fines
am
ou
nt
con
cer
ning
a d
iffe
re
nt
j
ud
ge
.
The
di
ff
e
ren
c
e
will
aff
ect
the
trai
ni
ng
d
at
a
and
t
he
final
res
ult
at
on
ce.
A
sur
ve
y
had
bee
n
c
onduct
ed
to
find
t
he
weig
ht
of
eac
h
crit
erion
usi
ng
A
HP
t
o
e
ns
ur
e
the
c
onsist
en
cy
of
the
res
ults.
Tw
o
se
ct
io
ns
wer
e
sur
ve
ye
d
w
hich
is
f
irst
to
identify
the
e
xperie
nce
of
j
ud
ges
in
ha
nd
li
ng
the
khal
wat
ca
ses
an
d
the
s
econd
one
is
to
cal
culat
e
the
dom
ai
n
factor o
f
eac
h crit
erion de
fine
d by the
judges
.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N: 20
88
-
8708
Esti
ma
ti
on
of f
ines
amo
un
t i
n syaria
h
c
rimin
al o
ff
en
c
es
us
i
ng ad
ap
ti
ve
ne
ur
o
-
fuz
z
y
.
..
(
A
hma
d
Fit
ri M
azl
am
)
5541
2.2.1.
Findi
ng w
ei
ghtage
ju
d
ge
e
xper
ie
nce
Ba
sed on t
he s
urvey c
ollec
te
d,
h
e
re a
re the
e
xp
e
rience
of ea
ch jud
ge
as
pre
sent in
Tab
le
8.
Table
8.
J
udge
ye
ars
of
e
xperi
ence
Ju
d
g
e Na
m
e
Exp
erience (
y
ea
rs)
Ju
d
g
e A
15
Ju
d
g
e B
20
Ju
d
g
e C
4
Ju
d
g
e D
2
Step
1: S
um
all th
e ye
ars
of e
xperie
nce,
= 20 +
15 +
4 + 2
=
41
(5)
Step
2: Find t
he
experie
nce
w
ei
gh
ta
ge
b
y
di
vid
in
g
eac
h ex
per
ie
nce
by
Y
Ex
per
ie
nc
e
We
igh
ta
ge
,
.
n
=
(6)
Fo
r
e
xam
ple, Ju
dge
A Ex
per
i
ence
Weig
htag
e,
.
A
=
.
A
=
15
41
= 0
.
37
(7)
The wei
ghta
ge
of all
ju
dg
e
experie
nce a
re ta
bu
la
te
d i
n Tabl
e 9
.
Table
9.
J
udge
exp
e
rience
w
ei
gh
ta
ge
Ju
d
g
e Na
m
e
W
eig
h
tag
e
Ju
d
g
e A
0
.37
Ju
d
g
e B
0
.49
Ju
d
g
e C
0
.10
Ju
d
g
e D
0
.05
2.2.2.
Findi
ng w
ei
ghtage cri
te
ri
a
Step
1: Com
par
e the
f
act
or
s.
10 p
ai
r
-
wise c
om
par
isons
we
re
qu
est
io
ne
d
t
o four j
udge
s i
n
the
stu
dy are
a.
a.
Ag
e
co
m
par
e
d wit
h
Se
x
b.
Ag
e
co
m
par
e
d wit
h
Ma
rita
l St
at
us
c.
Ag
e
co
m
par
e
d wit
h
Tim
e o
f Ar
rest
d.
Ag
e
co
m
par
e
d wit
h
L
ocati
on
of Arrest
e.
Sex
c
om
par
ed
with Marit
al
St
at
us
f.
Sex
c
om
par
ed
with Tim
e o
f Ar
rest
g.
Sex
c
om
par
ed
with L
ocati
on
of Arrest
h.
Ma
rita
l Sta
tus
com
par
ed wit
h Ti
m
e o
f
A
rr
es
t
i.
Ma
rita
l Sta
tus
com
par
ed wit
h Locat
io
n of A
rr
est
j.
Ti
m
e o
f A
r
rest
co
m
par
ed
w
it
h
L
ocati
on
of
Arrest
Step 2
: The d
e
gr
ee o
f
scal
e by
the f
our
judg
es w
as colle
ct
ed
an
d
the r
es
ult i
s co
m
plete
d
in the
m
at
rix
fo
r
eac
h
j
ud
ge.
Table
10 s
hows
an exa
m
ple o
f
pair
-
w
ise
co
m
par
iso
n by
j
ud
ge A.
Table
10. P
ai
r
-
wise c
om
par
ison res
ult by
ju
dg
e
A
Facto
r
Ag
e
Sex
Mar
ital
Statu
s
Ti
m
e
of
Arr
est
Locatio
n
of
Arr
est
Ag
e
1
.00
8
.00
0
.13
0
.14
0
.17
Sex
0
.13
1
.00
0
.17
0
.17
0
.17
Mar
ital
Status
8
.00
6
.00
1
.00
0
.33
6
.00
Ti
m
e
of
Ar
rest
7
.00
6
.00
3
.00
1
.00
7
.00
Locatio
n
of
Ar
rest
6
.00
6
.00
0
.17
0
.14
1
.00
Total
2
2
.13
2
7
.00
4
.46
1
.79
1
4
.33
Step
3:
Norm
a
li
zed
the
pair
-
wise
com
par
is
on
for
eac
h
ju
dg
e
.
As
sho
wn
in
Table
11,
t
he
norm
al
iz
ed
values
are
pr
ese
nted
a
s prio
rity
v
ect
or
of
c
rite
ria
by
al
l t
he
ju
dg
es
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
9
, N
o.
6
,
Dece
m
ber
2
01
9
:
5537
-
5544
5542
Table
11. Prio
r
it
y vector of cr
it
eria by all
judges
Ju
d
g
e A
Ju
d
g
e B
Ju
d
g
e C
Ju
d
g
e D
Priority
Vector
or
W
eig
h
t
Priority
Vector
or
W
eig
h
t
Priority
Vector
or
W
eig
h
t
Priority
Vector
or
W
eig
h
t
Ag
e
0
.09
0
.26
0
.08
0
.07
Sex
0
.04
0
.03
0
.03
0
.03
Mar
ital
Status
0
.28
0
.31
0
.56
0
.54
Ti
m
e
of
Ar
rest
0
.45
0
.17
0
.20
0
.21
Locatio
n
of
Ar
rest
0
.14
0
.23
0
.13
0
.15
2.2.3.
Findi
ng w
ei
ghtage cri
te
ri
a
i
n consider
at
io
n wi
th
ju
d
ge
e
xp
eri
ence
Step
1: Multi
pl
y ea
ch fact
or we
igh
ta
ge
w
it
h j
udge
e
xperie
nc
e w
ei
ghta
ge
.
Step
2: T
otal t
he
resu
lt
for
ea
ch fact
or to
get
the
final
weig
htage. T
h
e r
esu
lt
s ar
e s
how
n
i
n
Fi
gure
1.
Figure
1
.
Resul
t of final c
rite
ri
a w
ei
ghta
ge
2.3.
AN
F
IS e
nh
ance
d wi
th
A
HP
s
olu
tion
Ba
sed
on
(
4
)
,
ou
t
pu
t
va
riable
,
f
or
eac
h
r
ule
eq
uatio
n
is
t
hen
pr
e
dicte
d
again
f
or
li
nea
r
e
quat
ion
enh
a
nce
d wit
h AHP
weig
htag
e res
ult as in
(
5
)
a
nd
(
6
)
. T
he r
esults are
the
n show
n
in
Ta
ble 1
2
.
=
1
1
1
+
2
2
2
+
3
3
3
+
4
4
4
+
5
5
5
+
(8)
1
=
1
−
1
1
1
+
2
2
2
+
3
3
3
+
4
4
4
+
5
5
5
1
=
2
.
7
−
(
(
0
.
02
)
(
0
.
194
)
(
17
)
+
(
0
.
40
)
(
0
.
029
)
(
1
)
+
(
0
.
40
)
(
0
.
338
)
(
1
)
+
(
0
.
11
)
(
0
.
196
)
(
4
)
+
(
0
.
15
)
(
0
.
252
)
(
3
)
)
1
=
2
.
29
(9)
Table
1
2
.
P
red
i
ct
ed
f
uzzy
valu
e f
or
,
(
)
1
2
3
4
5
(
)
Ru
le 1
0
.02
0
.4
0
.4
0
.11
0
.15
0
.67
2
.29
Ru
le 2
0
.02
0
.23
0
.4
0
.13
0
.15
0
.51
2
.27
Ru
le 3
0
.02
0
.21
0
.4
0
.11
0
.14
0
.72
2
.38
Ru
le 4
0
.02
0
.27
0
.22
0
.13
0
.17
0
.33
2
.45
Ru
le 5
0
.02
0
.25
0
.2
0
.1
0
.15
0
.59
2
.50
Ru
le 6
0
.02
0
.2
0
.4
0
.11
0
.14
0
.82
2
.56
Ru
le 7
0
.01
0
.22
0
.17
0
.11
0
.15
0
.88
2
.56
Ru
le 8
0
.01
0
.22
0
.15
0
.11
0
.17
0
.85
2
.56
Ru
le 9
0
.01
0
.23
0
.17
0
.11
0
.14
0
.84
2
.56
Ru
le 10
0
.01
0
.25
0
.14
0
.13
0
.15
0
.89
2
.59
Ru
le 11
0
.01
0
.22
0
.15
0
.11
0
.14
0
.68
2
.37
Ru
le 12
0
.01
0
.22
0
.14
0
.11
0
.13
0
.72
2
.55
3.
RESU
LT
S
A
ND
DI
SCUS
S
ION
S
In
this
pa
pe
r,
75
cases
wer
e
trai
ned
,
a
nd
30
cases
wer
e
te
ste
d
with
ANFIS
a
nd
A
NF
I
S+A
HP.
The
cases
in
Table
1
3
a
re
com
par
ed
beca
us
e
of
the
sam
e
value
for
se
x,
m
arit
al
sta
t
us
,
tim
e
of
arre
st
and
locat
ion
of
a
rrest
crit
eria.
Howev
e
r,
the
val
ue
of
age
is
di
f
fer
e
nt.
T
he
res
ult
sho
ws
that
the
hum
an
judgem
ent
is
incon
sist
ent
wh
e
re
the
fine
s
a
m
ou
nt
for
a
25
-
ye
a
r
-
old
m
an
is
lower
tha
n
a
23
-
ye
ar
-
ol
d
m
an
with
the
sa
m
e
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N: 20
88
-
8708
Esti
ma
ti
on
of f
ines
amo
un
t i
n syaria
h
c
rimin
al o
ff
en
c
es
us
i
ng ad
ap
ti
ve
ne
ur
o
-
fuz
z
y
.
..
(
A
hma
d
Fit
ri M
azl
am
)
5543
value
f
or
t
he
ot
her
crit
eria.
T
he
age
diff
e
rence
is
on
ly
two
ye
ars,
an
d
it
is
qu
it
e
co
nfusing.
By
us
in
g
A
NFI
S
,
the
ga
p
of
fine
s
am
ou
nt
is
s
m
al
le
r
wh
e
re
ANFIS
+
A
HP
sh
ows
a
ver
y
consi
ste
nt
re
sul
t
fo
r
the
ra
ng
e
of
a
ge
unde
r 35 y
ears
old
.
Table
1
3
. Res
ul
t com
par
ison
betwee
n hu
m
an jud
gem
ent, A
N
FI
S
and
A
NF
I
S+A
HP
Cas
e No
Ag
e
Hu
m
an
Ju
d
g
e
m
en
t
ANFIS
ANFI
S+AH
P
1
1
0
0
3
-
143
-
0
0
0
5
-
2
0
1
6
23
2
.9
2
.7
2
.8
1
1
0
0
4
-
143
-
0
0
8
0
-
2
0
1
6
24
2
.9
2
.8
2
.8
1
1
0
0
5
-
143
-
0
0
2
9
-
2
0
1
6
25
2
.7
2
.7
2
.8
1
1
0
0
3
-
143
-
0
0
3
4
-
2
0
1
5
31
2
.9
2
.8
2
.8
1
1
0
0
5
-
143
-
0
0
4
7
-
2
0
1
6
32
2
.7
2
.8
2
.8
ANFIS
+
A
HP
has
processe
d
the
eq
uatio
n
a
nd
present
th
e
resu
lt
in
it
s
r
ang
e
.
A
s
an
e
xam
ple,
fo
r
the
sam
e
value
of
oth
e
r
crit
eria
(se
x=f
em
al
e,
m
ari
ta
l
st
at
us
=si
ngle
,
lo
cat
ion
of
a
rr
es
t=
op
e
n
area
,
ti
m
e
of
arr
est
=
d
ay
)
a
nd only
age
is
diff
e
re
nt, here
Table
14
is t
he
r
a
ng
e
for fine
s
am
ou
nt.
Table
1
4
. F
i
ne
s am
ou
nt r
a
nge
b
ase
d o
n
ra
ng
e of a
ge
Ran
g
e of
Age
Fin
es A
m
o
u
n
t us
in
g
ANFIS+
AHP
0
-
16
2
.6
17
-
51
2
.7
52
-
85
2
.8
8
6
and
ab
o
v
e
2
.9
4.
CONCL
US
I
O
N
This
pa
per
pr
opos
e
d
A
NF
I
S+A
HP
t
o
es
tim
a
te
fines
a
m
ou
nt
base
d
on
pr
e
vi
ou
s
ju
dg
m
ents.
75
dataset
s
we
re
us
ed
f
or
tra
ining,
a
nd
30
dataset
s
we
re
us
e
d
f
or
te
sti
ng.
The
est
im
a
t
ion
c
onside
re
d
five
inputs
an
d
one
sing
le
outp
ut
base
d
on
case
fact
from
the
Syari
ah
crim
inal
file
s
in
the
stud
ie
d
de
par
t
m
ent.
The
resu
lt
of
t
he
propose
d
m
et
hod
has
pro
v
en
that
A
NF
I
S
+AHP
is
a
n
ef
f
ic
ie
nt
way
to
e
stim
at
e
the
fines
an
d
helps
the
judge
m
ake a
pr
el
im
inary
j
ud
gm
ent
at o
nce
.
ACKN
OWLE
DGE
MENTS
The
a
utho
rs
w
ou
l
d
li
ke
to
tha
nk
the
De
pa
rtm
ent
of
Syari
a
h
J
udic
ia
ry
Sta
te
of
Te
reng
ga
nu
es
pecial
ly
Y.A
T
n
H
j
Ka
m
al
ru
azm
i
bin
Ism
ail,
the
Syari
ah
Highco
ur
t
Ju
dge
f
or
the
ir
act
ive
co
op
e
rati
on
a
nd
the
data
pro
vid
e
d.
REFERE
NCE
S
[1]
S.
D.
Jam
es,
et
al.
,
“
Multi
pl
e
Crit
eria
Dec
ision
Making,
Mult
ia
tt
r
ibute
Util
i
t
y
The
or
y
:
Th
e
Next
Te
n
Ye
ars,
”
Manage
ment
S
cienc
e
,
v
o
l. 38, pp
.
645
-
654
,
1992
.
[2]
W
.
J
y
rki
,
et
a
l.
,
“
Multi
ple
Cr
it
er
ia
De
ci
sion
Mak
ing,
Mul
ti
a
tt
ribu
te
Uti
li
t
y
Th
eor
y
:
Re
ce
nt
Acc
o
m
pli
shm
ent
s
and
W
hat
Lies Ahead
,”
Journal
Man
ageme
nt
S
cienc
e
,
v
ol
.
54
,
pp
.
133
6
-
1349
,
2008
.
[3]
T.
Faisal,
e
t
al.
,
“
Adapti
ve
Neur
o
-
Fuzz
y
Inf
ere
n
ce
S
y
st
em
for
di
agnosis
risk
in
dengue
patien
ts
,”
Ex
pert
Syst
ems
wit
h
App
licati
on
s
,
vol.
39,
pp.
44
83
-
4495
,
2012
.
[4]
R.
Appiah,
“
Impl
ementation
of
ada
pt
ive
n
eur
o
fuz
z
y
inf
ere
n
ce
s
y
stem
for
m
alari
a
d
ia
gnosis
,”
A
ca
se
s
tud
y
a
t
Kw
esimintsim
P
ol
y
cl
in
ic
,
PhD
di
ss
.
,
2016.
[5]
R.
Nohria
and
P.
S.
Mann,
“
Diagnosis
of
H
y
per
t
ension
Us
ing
Adapti
ve
Neuro
-
Fuzz
y
In
fer
ence
S
y
st
em,
”
Inte
rnational
Jo
urnal
of
Comput
er
Scienc
e
and
Technol
og
y
,
v
ol
.
6,
pp
.
36
-
40
,
20
15
.
[6]
M.
A.
Abus
har
iah,
et
al
.
,
“
Auto
m
at
ic
he
art
dise
ase
dia
gnosis
s
ystem
base
d
on
ar
ti
ficia
l
neur
al
n
e
twork
(AN
N)
and
ada
pt
ive
neu
ro
-
f
uzzy
inf
er
en
ce
s
y
stems
(AN
FIS
)
appr
oac
h
es
,”
J
ournal
of
softwa
re
engi
ne
ering
a
nd
appli
ca
ti
ons
,
vol.
7
,
p
p.
1055
,
2014
.
[7]
N.
Ziasabounc
hi
and
I.
As
ker
za
d
e,
“
AN
FIS
base
d
cl
assifi
cation
m
od
el
for
hear
t
disea
se
pr
edi
c
ti
o
n
,”
In
te
rnationa
l
Journal
of
Elec
t
rical
&
Compute
r Sc
ie
n
ce
s I
JE
C
S
-
IJE
NS
,
vol
.
14
,
pp
.
7
-
12
,
2014
.
[8]
D
.
Bohra
and
S.
Bhat
ia,
“
Portfol
io
ret
urn
m
odell
ing
using
AN
FI
S
,”
Inte
rnat
iona
l
Journal
of
Eng
ine
ering
,
vol.
1
,
pp.
1
-
4
,
2012
.
[9]
M.
A.
Bo
y
a
ci
og
l
u
and
D
.
Avc
i,
“
An
ada
pt
ive
netw
ork
-
base
d
fuz
z
y
infe
r
ence
s
y
s
tem
(AN
FI
S)
for
t
he
pre
d
ic
t
ion
of
stock
m
ark
et
re
turn:
th
e
c
ase
of
the
Ist
anbul
stock
ex
cha
ng
e
,”
Ex
p
ert
S
yste
ms
wit
h
App
licati
ons
,
vol.
37
,
pp.
7908
-
7912
,
2
010
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
9
, N
o.
6
,
Dece
m
ber
2
01
9
:
5537
-
5544
5544
[10]
L.
Abdull
ah
an
d
F.
N.
Azm
an,
“
W
ei
ghts
of
obesity
fa
ct
ors
u
s
ing
anal
y
tic
hi
er
arc
h
y
pro
ce
ss
,”
Int.
J.
Re
s.
Rev.
App.
S
ci
,
vol
.
7
,
pp.
57
-
83
,
2011
.
[11]
A.
F.
Maz
la
m
,
et
al
.
,
“
Esti
m
at
i
on
of
Fines
Am
ount
in
S
y
ar
ia
h
Criminal
Offen
c
es
Us
ing
Adaptive
Neuro
-
Fuzz
y
Infe
ren
c
e
S
y
ste
m
(A
NF
IS)
,”
Journal
of
Tele
co
mm
unic
ati
on,
E
l
ec
troni
c
and
Computer
Engi
nee
r
ing
(
JTEC
)
,
vol.
9,
pp.
153
-
156
,
20
17
.
[12]
A.
S.
Kum
ar,
“
Gene
rating
r
ule
s
for
adva
n
ce
d
fu
zzy
reso
lut
ion
m
e
cha
nis
m
to
dia
gnosis
hea
r
t
dis
ea
se
,”
Inte
rnational
Jo
urnal
of
Comput
er
Applications
,
vol.
77
,
2013
.
BIOGR
AP
H
I
ES
OF
A
UTH
ORS
Ah
mad
Fi
tri
Maz
lam
is
an
I
CT
Offic
er
in
I
nstit
ut
Pendidi
k
an
Guru
Kam
pu
s
Dato’
Raz
ali
Ism
ai
l
and
a
Pos
t
-
Gradua
te
St
udent
a
t
School
of
Inform
at
i
cs
and
Appli
ed
Mathe
m
at
i
cs,
Univer
siti
Mal
a
y
sia
T
ere
ngg
anu
.
He
obta
in
ed
Bac
he
lor
Degre
e
in
Scie
nc
e
(Com
pute
r)
from
Univer
siti
Te
kn
ologi
Ma
l
a
y
sia
in
2005.
His
re
sea
rch
es
are
in
fields
of
De
cis
ion
Support
S
y
stem,
Art
ifi
c
i
al
In
te
l
li
gen
ce
a
nd
Artificial
Ne
ura
l
Ne
twork.
R
ec
en
tly
,
s
y
st
em’s
appl
i
ca
t
ion
on
s
y
ar
ia
h
judi
c
i
ar
y
dep
art
m
ent
h
as
bee
n
tackle
d
and
his
ar
ti
c
le
h
as
bee
n
pub
li
she
d
in
JTEC
in
20
17.
Besid
es,
he
is
al
so
invo
l
ved
in
NG
Os
,
student
associati
ons,
and
m
an
ag
ing
non
-
profi
t
founda
ti
on
.
Wan
Nu
ral
Jaw
ahir
Hj
Wan
Y
us
sof
rec
ei
ved
h
er
B.
IT
in
Softw
are
Engi
n
ee
r
ing
and
M.Sc.
in
Artifi
c
ia
l
Intelligence
from
Kolej
Univ
ersiti
Sains
dan
Teknolo
gi
Mal
a
y
s
ia
.
In
2014
,
she
obta
ine
d
her
Ph.D.
from
Univer
siti
Mal
a
y
s
ia
Te
ren
gg
anu.
She
is
cur
ren
tly
a
se
nior
le
c
ture
r
at
School
of
I
nform
at
ic
s
and
Applie
d
Ma
th
emati
cs,
Univer
siti
Mal
a
y
s
ia
Te
ren
gg
anu.
Her
rese
arc
h
intere
sts a
r
e in
2D/
3D i
m
age
ana
l
y
s
is a
nd
und
erwa
t
e
r
vide
o
proc
essing.
Rab
ie
i
Mam
at
recei
v
ed
his
M.Sc.
d
egr
ee
i
n
High
Perform
anc
e
S
y
st
em
fr
om
Univer
si
t
y
Coll
ege
of
Science
and
Techno
log
y
Mal
a
y
s
ia
(
KU
STEM)
in
2
004
and
get
ti
ng
his
P
h.
D.
i
n
Inform
at
ion
T
echnolog
y
from
th
e
Univer
si
ti
Tun
Hus
sein
Onn
(
UTHM
)
in
2014
.
Curre
n
tly
h
e
is
a
le
c
ture
r
i
n
School
of
Inform
at
ic
s
and
Applie
d
Math
emati
cs,
Univ
er
siti
Malay
si
a
Te
ren
gg
anu
(U
MT).
His
teac
hi
ng
topi
cs
in
cl
ud
es
W
eb
and
Mobile
Appl
ic
a
ti
on
Deve
lopment
and
Program
m
ing.
His
r
ese
ar
ch
i
nte
rests
inc
lud
e S
oft
Com
puti
ng
and
Dat
a
Min
ing
.
Evaluation Warning : The document was created with Spire.PDF for Python.