Internati
o
nal
Journal of Ele
c
trical
and Computer
Engineering
(IJE
CE)
V
o
l.
5, N
o
. 1
,
Febr
u
a
r
y
201
5,
pp
. 84
~91
I
S
SN
: 208
8-8
7
0
8
84
Jo
urn
a
l
h
o
me
pa
ge
: h
ttp
://iaesjo
u
r
na
l.com/
o
n
lin
e/ind
e
x.ph
p
/
IJECE
IQ Clas
s
i
fication via Brai
nwave
Features: Review on Artificial
Intelligence Techniques
Aisyah H
a
rtin
i
Jahidin, Mohd Nasi
r T
a
i
b
, N
o
ori
t
aw
ati
Md T
a
hi
r,
Me
ga
t
S
y
ahirul Amin Me
ga
t Ali
Facult
y of Ele
c
tr
ica
l
Eng
i
ne
ering
,
Universiti Tekn
ologi
M
A
RA
, M
a
la
ys
i
a
Article Info
A
B
STRAC
T
Article histo
r
y:
Received Oct 29, 2014
Rev
i
sed
D
ec 20
, 20
14
Accepte
d Ja
n
6, 2015
Intelligen
ce stud
y
is on
e of key
s
tone to
d
i
stinguish individual dif
f
erences in
cognitiv
e ps
y
c
h
o
log
y
. Convent
i
onal ps
y
c
hom
etr
i
c tests are l
i
m
ited in term
s
of as
s
e
s
s
m
ent ti
m
e
, and
exis
ten
c
e of bi
as
nes
s
is
s
u
es
. Apart
from
that
, th
ere
is
still
lack
in kno
wledge
to c
l
assi
f
y
IQ b
a
sed on
EEG signals
an
d inte
llig
ent
signal processin
g
(ISP) techniq
u
e.
IS
P
purpos
e is
to extra
c
t
as
m
u
ch
information as possible
from
signal a
nd no
ise data using
lear
ning and
/
or
other s
m
art
te
ch
niques
.
Ther
efor
e,
as
a
firs
t
at
te
m
p
t in c
l
as
s
i
f
y
in
g IQ fe
ature
via sci
e
ntif
ic ap
proach,
it is im
portant to
iden
ti
f
y
a
rel
e
vant
te
c
hnique with
prom
inent parad
i
gm
that is suitable for
this area
of applica
tion.
Thus, this
arti
cle r
e
vi
ews
s
e
veral IS
P
app
r
oaches
to prov
ide cons
ol
idat
ed
s
ource of
information.
This in particular
f
o
cuse
s on prominent par
a
digm that suitable
for pattern
clas
s
i
fica
tion in biom
edic
al ar
ea
. The
review le
ads
to s
e
lec
tion of
ANN
since it has been widely implemen
ted for pattern cl
assification in
biomedical engineering
.
Keyword:
A
NN
EEG
Expert system
Fuzzy logic
IQ
Copyright ©
201
5 Institut
e
o
f
Ad
vanced
Engin
eer
ing and S
c
i
e
nce.
All rights re
se
rve
d
.
Co
rresp
ond
i
ng
Autho
r
:
Ai
sy
ah Hart
i
n
i
Jahi
di
n,
Facu
lty of Electri
cal Engineering,
Un
i
v
ersiti Tekn
o
l
o
g
i
M
A
RA,
4
045
0 Sh
ah A
l
a
m
, Selan
gor
,
Malaysia.
Em
ail: aisyah23@gm
ail.co
m
1.
INTRODUCTION
Co
gn
itiv
e ab
il
ity is a
su
b
-
d
i
v
i
sion
of hu
man
po
te
n
tial wh
ich
refers to
in
d
i
v
i
d
u
a
l
’
s ch
aracteristic
ap
pro
ach
in
i
n
fo
rm
atio
n
p
r
o
c
essin
g
. Th
is
h
a
s b
een
well-estab
lish
e
d
with
in
th
e do
m
a
in
o
f
h
u
m
an
in
tellig
en
ce
and is strictly related to intelligence
quotient
(IQ).
To
date, IQ is assessa
ble using conventi
onal m
e
thods s
u
c
h
as Stan
fo
rd
-Bi
n
et In
tellig
en
ce Scales [1
], Rav
e
n’s Prog
ressiv
e Matrices [2
], and
W
e
ch
sler In
tellig
en
ce Scales
[3]
,
whi
c
h can t
h
e
n
be
qua
nt
i
f
i
e
d f
o
r e
v
a
l
uat
i
on
of m
e
nt
al
per
f
o
r
m
a
nce.
A d
r
aw
b
ack t
o
t
h
ese t
y
pe
o
f
assessm
en
t are th
at it is
relativ
ely in
sen
s
itiv
e to ind
i
v
i
d
u
al u
n
d
e
rstand
i
n
g in an
swerin
g th
e p
s
ych
o
metric
tests. Moreove
r
, there would be bias
ness issue that is unique to each of
the assess
ment batteries [2, 4]
.
C
e
rt
ai
nl
y
,
EE
G
has
becom
e
i
n
creasi
ngl
y
i
m
port
a
nt
as i
t
can
reco
r
d
v
a
st
am
ount
s
o
f
com
p
l
e
x
ne
ur
o
n
al
activ
ity fro
m
th
e hu
m
a
n
b
r
ain
.
Th
u
s
, th
e q
u
a
litativ
e in
fo
rm
atio
n
can
b
e
ov
erco
me with
qu
an
titativ
e
measurem
ent provid
ed
b
y
th
e E
E
G
.
In
ge
neral,
EE
G can be cate
g
orized i
n
to
prim
ar
y an
d
seco
nd
ar
y sign
als. A
p
r
im
ar
y EEG
can
b
e
o
b
s
erv
e
d
and
in
terp
reted
d
i
rectly d
u
r
ing
th
e EEG record
i
n
g
.
Th
ese si
g
n
a
ls h
a
v
e
b
e
en
utilized
ex
ten
s
iv
ely t
o
assist clin
ician
s
in
d
i
ag
no
sing
acu
te p
a
ed
iatric en
ceph
a
l
opath
y [
5
], anaes
t
h
esi
a
[
6
], st
ro
ke [
7
]
,
schi
zo
phre
n
i
a
[8]
,
and dem
e
nt
i
a
s or Al
zheim
e
r [9]
.
U
n
der
sev
e
r
e
cas
e
s
, it is also used to ascertai
n
brain deat
h [10].
Meanwhile, the seconda
ry E
E
G is used
for m
o
re sophist
icated applica
t
i
ons
.
Thi
s
h
o
w
eve
r
, w
oul
d req
u
i
r
e
com
p
l
e
x dat
a
pr
ocessi
ng a
p
p
r
oac
h
es
fo
r si
g
n
al
m
a
ni
pul
at
i
o
n
.
T
h
e sec
o
n
d
ary
EE
G
ha
v
e
bee
n
i
m
pl
em
ent
e
d i
n
pers
on rec
o
gni
tion [11], and
brai
n–c
om
puter interface [
1
2
]
, as well as
i
nve
stigation on
ne
urophysiol
ogical
co
rrelates with p
s
ych
oph
ysio
l
o
g
y
,
wh
ich
norm
a
l
l
y
in
cl
u
d
e
s e
m
o
tio
n
recog
n
ition
stud
ies [1
3]
an
d
its effects
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
IQ Cl
a
ssifica
tio
n via
Bra
i
n
w
a
ve Fea
t
u
r
es:
Review o
n
Arti
ficia
l In
tellig
ence Techn
i
qu
es
(Aisya
h
H
J
)
85
fro
m
ex
tern
al
sti
m
u
l
u
s
[1
4]
. Oth
e
r app
licatio
n
o
f
EEG can
also
b
e
rel
a
ted
to
cog
n
iti
v
e
ab
ilities such
as
in
tellig
en
ce [15
]
and
learn
i
ng style [16
]
.
The
features e
x
tracted from
raw EE
G sign
als withou
t lo
si
ng
its
o
r
i
g
in
al co
n
t
en
t are cru
c
ial in
ord
e
r
to
d
i
stingu
ish
an
d classify
b
r
ain
activ
ity effectiv
ely.
Th
is
wou
l
d en
ab
le
co
rrelatio
n of
p
a
ram
e
ters with
t
h
e
neu
r
opsy
c
hol
o
g
i
cal
fu
nct
i
o
n
i
ng o
f
t
h
e
br
ai
n.
W
i
t
h
t
h
e
ai
d of v
a
ri
o
u
s si
g
n
al
pr
o
cessi
ng a
p
p
r
o
aches
,
charact
e
r
i
s
at
i
ons o
f
brai
nw
ave feat
ures i
n
t
h
e
past
ha
ve t
a
ke
n n
u
m
e
ro
us a
p
p
r
oac
h
es
un
der a
br
o
a
d
pers
pect
i
v
e
of
EEG st
u
d
i
e
s [
1
7-
1
9
]
.
Suc
h
va
l
u
abl
e
i
n
fo
rm
ati
on
obt
ai
ne
d t
h
ro
u
gh i
n
n
ovat
i
ve si
g
n
al
p
r
oce
ssi
n
g
is co
mm
o
n
l
y in
corpo
r
ated
with
th
e
u
s
e
o
f
in
tellig
en
t
classifiers an
d h
e
n
ce, resu
lt in
the co
n
c
ep
tio
n
of
intelligent signal proces
si
ng (IS
P) technique. T
h
e approach re
fers
t
o
the im
ple
m
entation of m
ode
l-free
t
echni
q
u
es
f
o
r
feat
u
r
e ext
r
a
c
t
i
on a
nd m
odel
l
i
ng
p
u
r
p
o
s
es [2
0]
. T
h
es
e wo
ul
d
f
u
rt
h
e
r co
nt
ri
but
e
t
o
t
h
e
enha
ncem
ent of knowledge and lead to
a wi
d
e
r rang
e m
u
lt
id
iscip
lin
ary ap
p
lication
s
. Certain
l
y, th
is p
r
esen
ts
an e
x
cel
l
e
nt
op
po
rt
u
n
i
t
y
t
o
a
d
vance
t
h
e
re
search
in
relatin
g EEG
with
IQ via
ISP.
Cu
rren
tly, th
ere is a
v
a
riet
y o
f
artificial in
tellig
en
ce
(AI) tech
iqu
e
s [21-24
] th
at
cou
l
d
b
e
incorporate
d
for IQ classifi
c
a
tion via the
brainwave
features. In order
to j
u
stify th
e sel
ectio
n
of a p
a
rticu
l
ar
m
e
t
hod, t
h
e
pa
per at
t
e
m
p
t
s
t
o
pro
v
i
d
e a gen
e
ral
ove
rvi
e
w and a
ppl
i
cat
i
o
ns o
f
est
a
bl
i
s
h
e
d t
echni
qu
es,
whi
c
h
are ex
pert
sy
st
em
, genet
i
c
al
g
o
ri
t
h
m
,
fuzzy
l
ogi
c a
nd a
r
t
i
f
i
c
i
a
l
neural
net
w
or
k (
A
NN
) [
2
1
-
2
4
]
.
The
di
sc
u
ssi
on
t
h
en
f
o
cuse
s
o
n
sel
ect
i
o
n
of
AN
N as
s
u
i
t
a
b
l
e ISP
m
e
t
hod
fo
r cl
assi
fi
cat
i
o
n
o
f
IQ
i
n
dex
base
d
on
EE
G.
2.
ESTABLISHED INTELLIGENT SIGNAL
PROCESSING APPROACHES
Th
rou
gho
u
t
the years, p
a
ttern
reco
gn
ition
an
d
classi
ficatio
n
h
a
s
b
een
mad
e
p
o
s
sib
l
e v
i
a ISP t
h
at
sp
ecifically focu
ses
on
an
al
yzin
g
an
d m
o
d
e
llin
g of com
p
lex
d
a
ta. Th
e aim
o
f
ISP is to
ex
tract
as m
u
ch
i
n
f
o
rm
at
i
on as
p
o
ssi
bl
e
fr
om
si
gnal
a
n
d n
o
i
se dat
a
usi
n
g
de
di
cat
ed l
e
a
r
ni
ng t
e
c
hni
q
u
e
s [
25]
.
ISP
d
i
ffers
fu
n
d
am
ent
a
l
l
y
fr
om
t
h
e cl
assical
appr
oac
h
o
f
st
at
i
s
t
i
cal
si
gnal
pr
ocessi
ng
i
n
t
h
at
t
h
e i
n
p
u
t
-
out
put
be
ha
vi
o
u
r
of a c
o
m
p
lex
syste
m
is
m
odelled
using intelligent
or
m
odel-free tec
hni
que
s,
rathe
r
tha
n
relying on t
h
e
lim
itations of
m
a
them
atical
m
odels [20].
To achie
ve
such goals, num
e
rous m
e
thods
can be utilised; each
havi
ng i
t
s
o
w
n u
n
i
q
ue ad
va
nt
ages a
nd
dr
awbac
k
s
.
The
r
efo
r
e, i
t
i
s
impo
rt
ant
t
o
i
d
e
n
t
i
f
y
a part
i
c
u
l
ar ISP
t
echni
q
u
e t
h
at
coul
d m
eet the speci
fi
cat
i
ons re
q
u
i
r
ed
f
o
r m
odel
l
i
ng
t
h
e rel
a
t
i
ons
hi
p bet
w
een
IQ
an
d
brai
nwa
v
e feat
ures
.
Th
ere are
wid
e
v
a
rieties of
ISP techn
i
qu
es cu
rren
tly
av
ailab
l
e.
Howev
e
r,
th
is p
a
p
e
r
will on
ly fo
cu
s
o
n
tech
n
i
q
u
e
s
m
o
st co
mm
o
n
l
y fo
und
in
t
h
e
literatu
re. Th
ese co
m
p
rise o
f
ex
p
e
rt system
s
,
g
e
n
e
tic alg
o
rith
m
s
,
fu
zzy log
i
c and
ANN.
It is no
ted
that th
e meth
od
s
o
r
igi
n
at
e from
hum
an-related phe
n
omena and an
offs
pri
ng
of b
r
oade
r di
s
c
i
p
l
i
n
e kn
o
w
n
as AI. I
n
ge
neral
,
t
h
e t
ech
ni
q
u
es ex
hi
bi
t
sim
i
l
a
r chara
c
t
e
ri
st
i
c
s of si
m
p
le
co
m
p
u
t
atio
n
a
l
stag
es, and
o
f
t
e
n
co
m
p
le
m
e
n
t
ed
b
y
rep
e
titiv
e learn
i
n
g
cycle [2
6
]
.
It
has bee
n
im
pl
em
ent
e
d i
n
di
verse areas
, su
ch as sci
e
nce [
27]
, en
gi
neeri
n
g [2
7,
28]
, a
g
ri
cul
t
u
re [
2
7
,
29
-
33]
, m
e
di
cal
[2
7,
3
4
-
41]
,
bi
om
edi
cal
[3
4,
4
2
,
43]
, c
o
m
put
er sci
e
nce
[2
7]
, a
n
d
fi
na
nci
a
l
[
4
4
,
45]
. Th
e
appl
i
cat
i
o
n pa
radi
gm
s
i
n
cl
ude
c
o
nt
rol
,
de
si
gn
, di
ag
no
si
s,
i
n
st
ruct
i
o
n,
i
n
t
e
r
p
ret
a
t
i
o
n, m
oni
t
o
ri
ng
, pl
anni
n
g
,
p
a
ttern
classi
ficatio
n
,
p
r
escri
p
tio
n, p
r
ed
iction
,
selectio
n
,
sch
e
du
ling
,
m
a
in
ten
a
n
ce and
targ
etin
g,
o
p
tim
is
atio
n
,
identification, clustering
a
n
d feat
ure
ext
r
actuin
[27-32, 34
,
35, 37-39, 44-78].
Ta
ble 1 s
u
mmarises
each ISP
ap
pro
ach
with th
e
resp
ectiv
e ap
p
lication
p
a
rad
i
g
m
s.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
IJEC
E V
o
l
.
5, No
. 1, Feb
r
uar
y
20
1
5
:
8
4
– 91
86
Tabl
e 1. Sum
m
a
ry
o
f
ISP Tec
hni
que
an
d A
p
pl
i
cat
i
on
P
a
ra
d
i
gm
s
T
echniques Application
Par
a
dig
m
s
Ar
eas
E
xper
t
Sy
stem
contr
o
l [27,
57]
,
d
e
sign [2
7]
,
pr
escr
iption [
27]
,
diagnosis
[27,
28,
31,
34-
37,
56,
57]
,
sor
ting [29]
,
identification [30], instru
ction, interpretation,
m
onitor
i
ng [38,
57]
,
selection [44]
,
m
a
naging [
45]
,
planning [59]
,
classificati
on [3
2]
,
pr
ediction [6
3]
che
m
istry
[27], ge
ology
[27], space technology
[2
7],
electr
i
c r
a
ilway [28]
,
egg gr
ading [29
]
,
plant
pr
otection in pepp
er
[30]
,
fish disease diagnosis [3
1]
,
pollen gr
ains ident
i
ficaition
[3
2]
,
com
puter
networ
ks
43]
,
stock exchang
e
[44]
,
finance
m
a
nagem
e
nt [45]
,
health [56]
,
industr
ial [57]
,
dr
ug
m
e
tabolis
m
[63]
.
Gener
tic Algor
ithm
optim
ization
[68-
7
1
,
73,
74]
,
identification [78]
,
scheduling
[65,
66
]
,
featur
e
extr
actio
n [67]
,
contr
o
l
[68]
,
epilepsy
stage identification,
industr
ial,
power
syste
m
,
environ
m
e
n
t, electro
m
a
gnetic
s related to
antenna,
m
e
dical phy
sics,
contr
o
l theor
y
,
and
econo
m
i
cs [67,
70, 74,
79-
81]
.
Fuzzy
L
ogic
Contr
o
l [7
5,
76,
82
-
84]
,
decision
m
a
k
i
ng [6
0]
,
m
onitor
i
ng[39]
,
diagnosis [
40]
,
identification [8
5]
,
patter
n
r
ecognition/classi
fication [40,
42,
76]
r
obotic [84]
,
indus
tr
ial (auto
m
otive)
[82]
,
power
[83]
,
geoscience [86]
,
instr
u
m
e
ntation [87]
,
stock tr
ading
[60], ty
phoid fever
[40],
m
u
ltifunctio
nal prosthesis
contr
o
l [42]
.
ANN
Contr
o
l [2
4,
64]
,
function appr
o
x
im
ation,
pr
ediction [33,
77]
,
patter
n
classification [8
8-
92]
,
f
o
recasting [
9
3], cl
ustering/categorisation,
diagnosis [94]
,
fuel consum
ption in wheat pr
oduction [33]
,
sleep
scor
ing [43]
,
r
i
sk in dengue
patients [
77]
,
post-
dialy
s
is
blood ur
ea concent
r
ation [41]
,
r
obotics and
m
achine
em
bodim
e
nts,
adaptive contr
o
l of co
m
p
lex sy
ste
m
s,
power
[95]
,
m
a
nufactur
ing,
tr
anspor
tation [88]
,
electr
i
c nose sensor
s [96]
,
envir
o
n
m
ental sy
ste
m
[97]
,
ener
gy
s
y
stem
s
[33]
,
epilepsy
[92]
.
Ex
pert
sy
st
em
i
s
k
n
o
w
l
e
d
g
e
-
base
d al
g
o
r
i
t
h
m
t
h
at
em
ul
at
es t
h
e
be
ha
vi
o
u
r
of
h
u
m
a
n ex
pert
s
i
n
t
e
rm
s
o
f
tho
ugh
t an
d reason
ing
p
r
ocess [98
]
. Th
e
ex
p
e
rt syste
m
s
can
d
e
sign
ed
as a p
r
ob
lem
-
so
lv
ing
ab
ility
m
o
d
e
l,
w
h
ich
invo
lv
es kn
ow
ledg
e,
r
eason
ing
,
conclu
sion
an
d
ex
p
l
an
atio
ns si
milar to
hu
m
a
n
ex
p
e
rt in ord
e
r to
anal
y
s
e and s
o
l
v
e com
p
l
e
x pr
obl
em
s [27]
. T
h
e fi
rst
ex
pert
sy
st
em
was devel
o
ped i
n
t
h
e
m
i
d-1
9
6
0
s [
2
1]
, but
its
ap
p
licatio
n p
r
o
liferated
in th
e
198
0s [27
,
9
9
]
.
Th
e
techniq
u
e
is su
itab
l
e fo
r cl
o
s
ed
-syste
m
ap
p
licatio
n
s
for
wh
ich
inp
u
t
s are literal an
d
precise, lead
ing
to
lo
g
i
cal
ou
tpu
t
s [9
8
]
. Th
roug
hou
t th
e years, its i
m
p
l
e
m
en
tatio
n
i
s
m
o
st
ly
i
n
t
e
nde
d
f
o
r
di
ag
nosi
s
p
u
r
p
oses
[2
7]
.
Ex
pert
sy
st
em
s have
a p
r
o
f
o
u
n
d
ap
pl
i
cat
i
on i
n
h
eal
t
h
d
i
agn
o
stic syste
m
s [34
,
36
,
38
,
39
,
59
],
wh
ich
in
terpret m
e
d
i
cal test resu
lts [2
7
]
.
Co
nv
ersely, gen
e
tic alg
o
r
ithm is a
so
lu
tion
search
ing
tech
n
i
q
u
e
, wh
ich
is roo
t
ed
in th
e id
eas o
f
ev
o
l
u
tio
n
pr
o
c
ess and
n
a
t
u
r
a
l p
o
p
u
l
ation
gen
e
tics [
7
4
]
.
Th
e techn
i
qu
e th
at w
a
s concep
tu
alized
b
y
John
Ho
lland
in
t
h
e 1
960
s,
h
a
s
d
r
i
v
en
t
h
e in
terest in
h
e
uris
tic search
al
g
o
rithm
s
with
foun
datio
n
s
in
n
a
tural an
d
phy
si
cal
p
r
oce
sses [7
9]
. It
s m
o
st
pr
om
i
n
ent
i
m
pl
em
ent
a
t
i
on i
s
i
n
opt
i
m
i
zati
on [
6
8-
71
, 7
3
,
74]
. C
o
nse
que
nt
l
y
,
g
e
n
e
tic algor
ith
m
is
m
o
st su
ccessfu
l
in
so
lv
ing
p
r
o
b
l
em
s th
at ar
e
r
e
lated
to
ch
ar
acter
izatio
n
i
n
physical
scien
ces [7
4
]
.
A
n
o
t
h
e
r
ISP tech
n
i
q
u
e
, th
e fuzzy lo
g
i
c w
a
s fo
und
ed
i
n
1965
b
y
Lo
f
ti Zadeh
[2
3
]
. Th
e co
n
c
ep
t
was ad
o
p
t
e
d f
r
om
hum
an t
h
i
nki
n
g
a
nd m
u
ch resem
b
l
e
s t
h
e nat
u
ral
l
a
n
gua
ge c
o
m
p
ared t
o
t
h
e t
r
a
d
i
t
i
onal
logical syste
m
s [75].
W
ithin the ne
xt two decades
, fuz
z
y
logic has b
een widely implem
ented to solve
pr
o
b
l
e
m
s
from
deci
si
on
-m
aki
ng t
h
e
o
ry
. The
t
echni
q
u
e i
s
m
o
st
successfu
l
l
y
appl
i
e
d i
n
cont
rol
p
r
o
b
l
e
m
s
[64,
75
, 7
6
,
8
2
-
8
4]
. The sy
st
em
howe
v
e
r
, i
s
n
o
t
capabl
e
of l
e
a
r
ni
n
g
[
1
00]
.
He
nce, i
t
s
i
m
pl
em
ent
a
t
i
on i
n
t
h
e area
o
f
p
a
ttern
recog
n
ition
is still l
i
m
i
t
ed
. To
m
i
n
i
m
i
ze su
ch
d
r
awb
a
ck
,
fu
zzy l
o
g
i
c
will n
eed
to
b
e
in
tegrated
with
ot
he
r I
SP t
e
c
h
ni
q
u
es
[4
2,
8
7
]
.
Mean
wh
ile, ANN is a
n
o
n
-
li
n
ear artificial i
n
tellig
en
ce
app
r
o
a
ch
th
at is in
sp
ired
b
y
th
e work
i
n
g
o
f
biological neurons in the bra
i
n [101]
. T
h
e technique that was introduce
d
in the 1940s
ha
s recently seen a
sh
arp
in
crease in
its i
m
p
l
e
m
e
n
tatio
n
[24
]
. ANN has b
e
en
an
altern
ativ
e to th
e trad
itio
n
a
l
statist
i
cal
m
o
d
e
llin
g
tech
n
i
qu
es in
variou
s scien
tifi
c
d
i
scip
lin
es [10
2
]
. Its m
a
in
ad
v
a
n
t
ag
e lies in
its ab
ility
to
learn
and
g
e
n
e
ralize
so
lu
tion
s
fo
r co
m
p
lex
p
r
o
b
l
em
s
[1
01
]. Hence, th
e
m
e
th
od
i
s
part
i
c
ul
arl
y
useful
f
o
r s
o
l
v
i
n
g a p
r
o
b
e
m
for
whi
c
h l
a
r
g
e a
m
ount
o
f
d
a
t
a
i
s
i
n
v
o
l
v
e
d
,
b
u
t
wi
t
h
un
k
n
o
w
n i
n
t
e
r-
rel
a
t
i
o
n
s
hi
p
[
1
0
2
]
.
It
can a
p
p
r
oxi
m
a
t
e
t
h
e
no
n
-
l
i
n
ear rel
a
t
i
ons
hi
p bet
w
een t
h
e i
n
p
u
t
vari
a
b
l
e
s and t
h
e
o
u
t
p
ut
of a so
p
h
i
s
t
i
cat
ed sy
st
em
[1
03]
.
Im
pl
em
ent
a
ti
on of t
h
e
AN
N can be m
o
st
l
y
fo
u
nd i
n
bi
om
edi
cal
appl
i
cat
i
ons [
4
6, 4
7
, 5
2
]
.
M
o
re
ove
r,
i
t
was
also disc
ove
re
d that ANN
has
been
very
success
f
ul
wh
en i
n
tegrated with innovati
ve signal proc
essing
ap
pro
ach fo
r pattern
recogn
itio
n purpo
ses [88
-
91
,
10
4
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
IQ Cl
a
ssifica
tio
n via
Bra
i
n
w
a
ve Fea
t
u
r
es:
Review o
n
Arti
ficia
l In
tellig
ence Techn
i
qu
es
(Aisya
h
H
J
)
87
3.
CO
MP
AR
AT
IVE AN
ALY
S
IS
Table
2 s
u
mmarizes on t
h
e s
e
lection criteri
as fo
r t
h
e c
o
mparative
analy
s
is. The
selection c
r
iteria
in
clu
d
e
cap
a
b
ilities to
g
e
n
e
ral
i
ze so
lu
tio
n
for co
m
p
lex
p
r
ob
lem
s
, to
self-learn, an
alyze an
d
m
o
d
e
l no
n-lin
ear
rel
a
t
i
ons
hi
ps
, a
s
wel
l
as i
t
s
pri
m
ary
pu
rp
ose
of
i
m
pl
em
ent
a
ti
on.
Tabl
e 2. Sel
ect
i
on
C
r
i
t
e
ri
as
Criterias
ES
GA
FL
ANN
1
Generalize
Solution for Co
m
p
lex
Pr
oblem
s
Yes Yes
No
Yes
2 Self-learning
Capability
No Yes No
Yes
3
Analy
z
e and M
odel Non-
linear
Relationships
No No No
Yes
4
Pur
pose of Im
ple
m
entation
Diagnosis
Opti
m
i
zation
Decision-m
a
king
Pattern
recognition
Up to th
is
p
o
i
nt, it h
a
s b
e
en
i
d
en
tified
t
h
at
ANN is th
e m
o
st su
itab
l
e
app
r
oach
t
o
b
e
im
p
l
e
m
en
ted
for
IQ m
odelling
via the EEG.
This is attribut
ed
by its
pre
v
i
ous successes
as a ro
bust m
odelling techni
que
in
b
i
o
m
ed
ical ap
p
licatio
n
s
,
p
a
rticu
l
arly for
p
a
ttern
reco
g
n
itio
n and
classifi
catio
n
.
Thu
s
, t
h
e fo
llowing
section
will furth
e
r elab
orate
o
n
th
e i
m
p
o
r
tan
t
aspects o
f
ANN
wh
i
c
h
in
cl
u
d
e
its
m
o
st p
o
p
u
l
ar arch
itecture.
3.
1. Ar
ti
fi
ci
al
Neur
al
Ne
tw
o
r
k
ANN was p
i
on
eered
b
y
McCu
llo
ch
and
Pitts in
th
e 19
40
s.
Later, t
h
e p
e
rcep
t
r
on co
nv
erg
e
n
ce
th
eorem
h
a
s b
een
in
tro
d
u
c
ed b
y
Ro
senb
latt in
th
e 1
960
s [1
01
]. Desp
ite th
is, th
e th
eo
ry
was still h
a
v
i
n
g
its
li
mitatio
n
s
, w
h
ich
r
e
su
lted
in
slo
w
d
o
wn
of
th
e r
e
sear
ch
ar
ea. H
o
w
e
v
e
r, the en
thu
s
iasm
r
e
su
rg
ed
in
1982
w
ith
th
e in
tro
d
u
c
tion
of b
a
ck-propag
a
tio
n
le
arn
i
ng
algo
rith
m
b
y
W
e
rbo
s
fo
r t
h
e
m
u
ltila
yer p
e
rcep
tron
n
e
t
w
ork. In
19
8
6
[2
4]
, i
t
was f
u
rt
he
r p
o
pul
a
r
i
zed by
R
u
m
e
l
h
art
.
Ev
er since, the use
of ANN ha
s seen a steady growt
h
wi
t
h
a
ppl
i
cat
i
o
ns s
p
a
n
i
n
g ac
r
o
ss a
wi
de
ran
g
e
of
p
r
o
b
l
e
m
dom
ai
ns as
pre
v
i
o
usl
y
m
e
nt
i
on i
n
Ta
bl
e 1
.
Th
e m
u
ltilayer
p
e
rcep
t
r
on
is curren
tly th
e
m
o
st estab
lish
e
d
sup
e
rv
ised
n
e
ural
n
e
two
r
k
m
o
d
e
l for
p
r
actical ap
p
l
i
catio
n
s
i
n
so
l
v
ing
d
i
v
e
rse
an
d co
m
p
le
x
p
r
ob
lem
s
[9
1
]
. As an
i
n
tellig
en
t tech
n
i
q
u
e, th
e
m
u
l
tilayer p
e
rcep
tro
n
h
a
s
b
een wi
d
e
ly u
s
ed
for
o
p
t
i
m
isatio
n
,
m
o
d
e
llin
g
,
pred
ictio
n
an
d fun
c
tio
n
app
r
oxi
m
a
t
i
on pu
rp
oses
[
1
0
5
]
.
H
o
we
ve
r,
i
t
has al
so be
en suc
cessf
ul
l
y
appl
i
e
d t
o
a vari
et
y
of
p
a
t
t
e
r
n
recogn
itio
n
and
classification
p
r
ob
lem
s
[5
4,
5
5
]
.
Such app
licatio
n
s
in
clu
d
e
d
i
sease reco
gn
itio
n [77
]
,
phy
si
ol
o
g
i
cal
anal
y
s
i
s
and m
odel
i
ng
[
46]
, ca
ncer
det
ect
i
on
and cl
assi
fi
cat
i
on
[
47]
, m
odel
l
i
ng o
f
heart
di
seas
e
reco
g
n
i
t
i
on
[
1
0
6
]
,
di
ag
nosi
s
o
f
c
o
r
ona
ry
art
e
ry
di
sease
[
49]
,
an
d
ot
he
r rel
a
t
e
d st
udi
es
[
5
2
,
89]
.
3.
2. I
n
te
gr
ati
o
n o
f
Arti
fi
ci
al
Neur
al
Ne
tw
o
r
k a
nd E
E
G
f
o
r I
Q
Cl
assi
fi
cati
on
Im
pl
em
ent
a
ti
on
of
A
N
N
i
n
bi
om
edi
cal
appl
i
cat
i
o
n
has
bee
n
o
b
ser
v
e
d
i
n
si
g
n
al
c
o
m
p
ressi
o
n
,
enha
ncem
ent
and i
n
t
e
rp
ret
a
t
i
on
[8
9]
. The
p
r
o
p
o
sed
resear
ch on IQ classification vi
a brainwa
v
e features fall
with
in th
e
do
main
o
f
sign
al interp
retation
,
wh
ereb
y th
e
p
a
ttern of EEG sub
-
b
a
n
d
f
eat
u
r
es will b
e
recogn
ized
t
h
r
o
u
g
h
a l
ear
n
i
ng
pr
ocess
an
d l
a
t
e
r cl
assi
fi
e
d
i
n
t
o
di
scret
e
IQ l
e
vel
s
[
1
07]
. A
n
unc
om
pro
m
i
s
i
ng ad
va
nt
age
o
f
ANN also
lies
in
th
e ab
ility to
cro
ss-correlate d
a
ta co
rrectly fro
m
u
n
kno
wn relatio
n
s
h
i
p
s
.
ANN
has cert
a
inly established itself as the
m
o
st
successfully
m
odelling techni
que
for biom
edical
appl
i
cat
i
o
ns [
5
2,
1
08]
,
pa
rt
i
c
ul
arl
y
i
n
t
h
e a
r
ea
of
pat
t
e
r
n
reco
g
n
i
t
i
on
[5
4]
. O
v
e
r
t
h
e
y
ears, se
ve
ral
t
y
pes
o
f
AN
N ha
ve
bee
n
de
vel
o
pe
d, e
ach wi
t
h
uni
qu
e pr
ope
rt
i
e
s t
h
at
m
a
ke t
h
em
m
o
re sui
t
a
bl
e fo
r cert
a
i
n
t
a
s
k
o
v
er
t
h
e ot
he
rs.
The
net
w
or
k arc
h
i
t
ect
ure va
ri
es i
n
t
e
rm
s of st
ru
ct
ure, act
i
v
at
i
o
n f
unct
i
on a
n
d
l
earni
n
g
al
g
o
r
i
t
h
m
.
In ge
ne
ral, the
ANN ca
n be im
plem
ented in su
per
v
ised
a
nd
uns
u
p
er
vise
d learnin
g
m
odes [
1
0
9
]
.
U
n
der the
form
er settin
g
,
th
e n
e
two
r
k
will h
a
v
e
to
recog
n
i
ze th
e
p
a
ttern
o
n
l
y fro
m
t
h
e inp
u
t
variables. Co
nv
ersely, the
later will al
lo
w th
e n
e
two
r
k to
learn
b
y
reco
gn
izing
th
e relatio
n
s
h
i
p
between
th
e inp
u
t
v
a
riab
les
an
d
th
e
o
u
t
p
u
t
.
Hen
c
e in
ou
r
p
r
op
osed
work, th
e n
e
two
r
k
will b
e
im
p
l
e
m
en
ted
in
sup
e
rv
i
s
ed
learn
i
ng
settin
g
s
whe
r
e
b
y the E
E
G feat
ures
wi
ll be assigne
d
as the inputs a
n
d
t
h
e d
i
stin
ct
IQ lev
e
ls as the actu
a
l ou
tpu
t
. Th
is
wo
ul
d
al
l
o
w
t
h
e net
w
o
r
k
t
o
b
e
t
r
ai
ne
d
by
co
nfi
r
m
i
ng i
t
s
pe
rf
orm
a
nce t
o
t
h
e
pr
o
v
i
d
e
d
ou
t
put
.
4.
CO
NCL
USI
O
N
Am
ong
vari
ous ISP a
p
proac
h
es, t
h
e ANN is percei
ved as a signific
ant technique
for
pattern
recogn
itio
n
and
classificatio
n th
ro
ugh
its
mo
d
e
lling
cap
a
bilit
ies. Its i
m
p
l
e
m
en
tatio
n
h
a
s ex
tend
ed
t
o
a wide
ran
g
e
o
f
bi
om
edi
cal
ap
pl
i
cat
i
ons
. T
h
us,
A
N
N
i
s
c
o
nsi
d
e
r
e
d
as
t
h
e
m
o
st
sui
t
a
bl
e m
e
t
hod
t
o
m
odel
t
h
e
I
Q
fr
om
brai
nwa
v
e feat
ures
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
IJEC
E V
o
l
.
5, No
. 1, Feb
r
uar
y
20
1
5
:
8
4
– 91
88
ACKNOWLE
DGE
M
ENTS
Aut
h
ors e
x
tend their appreci
ation to t
h
e Ministry
o
f
E
duc
at
i
on, M
a
l
a
y
s
i
a
an
d U
n
i
v
e
r
si
t
i
Tekn
ol
o
g
i
M
A
R
A
f
o
r t
h
e fi
nanci
a
l
su
pp
o
r
t
t
h
ro
u
gh
t
h
e Fu
ndam
e
nt
al
R
e
search Gra
n
t
Schem
e
(60
0
-
R
M
I/
FR
GS 5/
3
(
7
2
/
201
2)
) and MyPh
D sch
o
l
ar
sh
i
p
.
REFERE
NC
ES
[1]
S
.
B.I.
S
cal
es
, "H
is
tor
y
of
the
S
t
a
n
ford-Binet
Int
e
l
ligen
ce S
c
al
es
:
Content
and P
s
ychom
etrics
,"
200
3.
[2]
J. Raven
,
"
T
he
Raven
'
s progres
sive matrices: change
and stab
il
it
y
ov
er cultur
e
and
t
i
m
e
,"
Cogn
itiv
e psy
c
holog
y
,
vol. 41
, pp
. 1-48
, 2000
.
[3]
D. W
echsler, "W
echsler adult
intel
ligen
ce sca
l
e–Fourth Editio
n (W
AIS–IV),"
San Antonio, TX: NCS Pearso
n,
2008.
[4]
D.F. Lohman
, et
al.
, "Identif
y
i
ng Academically
Gi
fted Eng
lis
h-Languag
e
Learners Using Nonverbal
Tests:
A
Comparison of the Raven,
NNAT,
and CogAT,
"
Gifted Chi
l
d Quarterly,
vo
l. 52
, pp. 275-296,
October 1, 200
8
2008.
[5]
E. Hussain an
d D. Nordli,
"EEG Patte
rn
s
in Acut
e P
e
diatri
c En
ceph
a
lopathi
e
s
,
"
Jou
r
nal of clinica
l
neurophysiology: officia
l
publica
tion of th
e American
Electroen
cephalographic So
ciety,
vol. 30
, pp
. 539-544
, 2013
.
[6]
V.
V.
Moca
, et
al.
, "EEG under
anesthesia--f
eatu
r
e
extr
action with
TESPAR,"
Computer Methods and Programs i
n
Biomedicine,
vo
l. 95
, pp
. 191-20
2, Sep
2009.
[7]
M.J. van Put
t
en
and D.L
.
Tav
y
,
"Continuous quantitative E
E
G m
onitoring
in h
e
m
i
spheric stroke p
a
ti
ents using th
e
brain s
y
mmetr
y
index,"
Stroke,
v
o
l. 35
, pp
. 2489-
92, Nov 2004
.
[8]
M. Koukkou
, et al.
, "An EEG approach to the neurodevelopmen
tal h
y
po
thesis of schizophrenia stud
y
i
ng
schizophren
ics, normal
cont
ro
ls and
ado
l
escents,"
Journal of Psychiatric Research,
vol. 34
, pp
. 57
-73, 2000
.
[9]
C.E. J
acks
on a
nd P
.
J
.
S
n
y
d
e
r
,
"Elec
t
roen
ceph
a
lograph
y
and event-r
e
la
ted
po
tenti
a
ls
as bio
m
arkers of mild
cognitiv
e
im
pair
m
e
nt and
m
ild A
l
zhe
i
m
e
r’s disea
s
e,"
Alzhe
i
me
r's &
D
e
me
ntia,
vol. 4
,
pp
. S137-S143, 2008
.
[10]
D.
M.
Murray
, et al.
, "Early
EEG findings in hy
poxi
c-isch
emic enceph
a
l
opath
y predict ou
tcomes at 2
y
ears,"
Pediatrics,
vo
l.
124, pp
. e459
-67
,
Sep
2009.
[11]
P.
Ca
mpisi
, et al.
,
"EEG for
Auto
m
a
tic Person
Re
cognition
,
"
Com
puter,
vo
l. 45, p
p
. 87-89
, 2012
.
[12]
S. Sun, "Extreme energ
y
differ
e
nce fo
r featur
e extraction of EEG signals,"
Expert Systems with
Applica
tions,
vo
l.
37, pp
. 4350-43
57, 2010
.
[13]
N. Sharm
a
and
T. Ged
e
on,
"Object
ive m
easur
es,
sensor
s and co
m
putational
t
ech
niques for str
e
ss recogn
ition
and
clas
s
i
fi
cat
ion: a s
u
rve
y
,"
Computer
Methods and Progr
ams in Bio
m
edicin
e,
vol. 1
08, pp
. 1287-30
1, Dec 2012
.
[14]
T. B
a
umgartner
, et al.
, "From emotion perception to
emotio
n expe
r
i
ence:
Emotions evoked
b
y
pictures
an
d
c
l
a
ssica
l music,
"
International Jo
urnal
of Psycho
physiology,
vol.
60, pp
. 34-43
, 2
006.
[15]
A.H. Jahidin
, et al.
, "Brainwave
sub-band power ratio ch
aracte
r
i
st
ics in intel
lig
enc
e
assessm
ent," in
IEEE Control
and System Gra
duate
Research
Colloquium (
I
CSGRC)
, 2012, pp
. 318-321.
[16]
M.
S.
A.
Me
ga
t Ali
, et
a
l
.
, "Learning Sty
l
e Cla
ssification via EEG
Subband Spectral
Centro
id Frequ
e
ncy
Features,"
International Jo
urnal of
Electrical
and Computer Engin
eering
(
I
JECE)
,
vol. 4
,
pp
. 931-938, 2014.
[17]
S.
Sa
ne
i a
n
d J.
A
.
Cha
m
be
rs,
EEG signal process
i
ng
: Wiley
-Inter
s
cience, 2008.
[18]
S
.
Xie and S
.
Kris
hnan, "W
avele
t
-bas
ed s
p
ars
e
function
a
l
line
a
r m
odel with applica
tions to EEGs seizure dete
ct
io
n
and ep
ileps
y
diagnosis,"
Medical
&
Biological En
gineering
&
Co
mputing,
vo
l. 51, pp. 49-60, 2013
/02/01 2013
.
[19]
G. Chen, "Auto
m
atic
EEG seizure detection us
i
ng dual-
tree co
mplex wavelet-
Fourier features
,"
Expert System
s
with App
l
i
c
ation
s
,
vol. 41, pp. 23
91-2394, 2014
.
[20]
S.
S.
Hay
k
in a
n
d
B.
Kosko,
In
telligent signa
l processing
: Wiley
-
I
E
EE Press, 2001.
[21]
L. Shu-Hsien, "Expert s
y
stem
methodologies and
applications
—a
decade rev
i
ew from 1995 to 2004,"
Expert
Systems with
Ap
plications,
vo
l. 2
8
, pp
. 93-103
, 2
005.
[22]
J.
F.
Fre
n
ze
l
,
"Gene
t
ic
al
gori
t
h
ms,"
Poten
tials
, IE
E
E
,
vo
l. 12, pp. 2
1
-24, 1993
.
[23]
L.A
.
Zad
e
h,
"F
uzz
y
s
e
ts
,"
In
formation and
contro
l,
vo
l. 8, pp. 338
-353, 1965
.
[24]
A.
K.
Jain
, et
al.
,
"Artifici
a
l
neura
l
networks: A
tut
o
rial
,"
IE
EE
com
puter
,
vo
l. 29, p
p
. 31-44
, 1996
.
[25]
S. Ha
y
k
in and
B. Kosko, "Spec
i
al I
ssue On Int
e
llig
ent Sign
al P
r
ocessing,"
Proceedings of
the I
EEE,
vol. 86, p
p
.
2119-2120, 199
8.
[26]
K.
T.
Yang,
"Artificial Ne
ural Networks (ANNs):
A
New P
a
radigm
for Therm
a
l S
c
ien
ce and
Engi
neering
,
"
Journa
l
of Heat Transfer,
vol. 130
, p
.
093
001, 2008
.
[27]
J
.
Durkin, "Res
earch revi
ew: Applic
ation
of exp
e
rt s
y
s
t
em
s
in the s
c
ienc
es
,"
Ohio Journal of Science,
vol. 90
, pp.
171-179, 1990
.
[28]
Z. Liu and
Z. Han,
"Fault
diagnosis
of electr
ic railway
traction substa
tion w
ith model-b
a
sed
relation
guidin
g
algorithm,"
Exp
e
rt Systems wi
th
Applica
tions,
vo
l. 41
, pp
. 1730-1
741, 2014
.
[29]
V.
Patel
, et
al.
,
"Development and evaluation of
an
expert s
y
stem for egg sortin
g,"
Computers a
nd Electronics in
Agric
u
lture,
vo
l. 20, pp. 97-116,
1998.
[30]
L
.
Gonz
al
ez
-Di
a
z
, et al.
, "Exp
ert s
y
stem for integrated p
l
ant p
r
otection in pepp
er
(Capsicum ann
uun L.)
,
"
Exp
e
rt
Systems with
Ap
plications,
vo
l. 3
6
, pp
. 8975-897
9, 2009
.
[31]
D.
Li
, et al
.
, "Fish-Expert: a web-based expert s
y
s
t
em for fish dise
ase diagnosis,"
Expert Systems
with Application
s
,
vol. 23
, pp
. 311-
320, 2002
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
IQ Cl
a
ssifica
tio
n via
Bra
i
n
w
a
ve Fea
t
u
r
es:
Review o
n
Arti
ficia
l In
tellig
ence Techn
i
qu
es
(Aisya
h
H
J
)
89
[32]
Y.
Kay
a
, et al.
, "An expert class
i
fication s
y
s
t
em of polle
n of Onopordum using
a rough set appr
oach,"
R
evi
ew o
f
Palaeobotan
y an
d Palynology,
vo
l. 189
, pp
. 50-56
, 2013
.
[33]
M. Safa
and S.
Sam
a
rasinghe,
"
M
odelling fu
el
consum
ption in
wheat produ
c
tio
n using art
i
fi
cial
neural networks,"
Energy,
vol. 49
,
pp. 337-343
, 20
13.
[34]
J. Nahar
, et
a
l
.
,
"Com
putationa
l
int
e
llig
enc
e
for
m
i
croarr
a
y
da
t
a
and
biom
edi
c
al
im
age
anal
ys
is for th
e
earl
y
diagnosis of br
east cancer,"
Exp
e
rt Syst
ems with
Applica
tions,
vo
l. 39
, pp
. 12371-
12377, 2012
.
[35]
G.
Sardana
, et a
l
.
, "
E
merging biomarkers for the diagnos
is and p
r
ognosis of prostate can
cer
,"
Cli
n
ical chemistr
y
(
B
altimore, Md
.)
,
vol. 54
, pp
. 195
1-60, 2008
.
[36]
F. Bascif
tci an
d A. Eldem,
"
U
sing reduced
rule b
a
se with
Expert
S
y
stem for the diagnos
is of disease in
h
y
per
t
ension
,"
Medica
l
&
Biological Engin
eerin
g
&
Computing,
vol. 51
, pp
. 1287
-93, Dec 2013
.
[37]
P. Baptist
a
, et
al.
, "Gold nano
particles for
th
e dev
e
lopm
ent
of clin
ical d
i
ag
nosis methods,"
Analytical and
Bioanalytical Ch
emistry,
vo
l. 391
, pp
. 943-50
, Jun
2008.
[38]
E. S
e
to
, et
al.
,
"Developing he
alth
care ru
le-ba
s
ed expert s
y
s
t
e
m
s
:
cas
e s
t
ud
y
of a heart f
a
ilur
e
tel
e
m
onitorin
g
sy
s
t
e
m
,
"
Int J M
e
d Inform,
vol. 8
1
, pp
. 556-65
, A
ug 2012.
[39]
M. Mansoor B
a
ig
, et
al.
, "F
uzz
y
logi
c bas
e
d
anaes
thes
i
a
m
onitori
ng s
y
stems for the detection of absolute
h
y
povolaemia,"
Computers in Biology and
Med
i
cine,
vol. 43
, pp
.
683-92, Jul 2013
.
[40]
O.
W.
Samuel
, et
al.
, "A web
b
a
sed decision su
pport s
y
stem driven
b
y
fu
zzy
lo
gic for
the diag
nosis of ty
phoid
fever,"
Expert S
y
stems with
Applications,
vol. 40
,
pp. 4164-4171
,
2013.
[41]
A.T. Azar, "Fast neural network learni
ng
algorithms for medical
applications,"
Neural Computing an
d
Applica
tions,
vo
l. 23
, pp
. 1019-1
034, 2012
.
[42]
A.B. Ajibo
y
e
and R.F. Weir, "
A
heuristic fuzzy
log
i
c appro
a
ch to EMG pattern recognit
i
on f
o
r m
u
ltifunction
a
l
prosthesis contr
o
l,"
Neural Syst
ems and Rehabilitat
i
on Engi
ne
e
r
ing, IEEE Transactions on,
vol. 13, pp. 280-291,
2005.
[43]
M. Ronzhin
a
, et
al
.
, "Sleep
scoring using ar
tif
icial neur
al network
s
,"
Sleep M
e
d Rev,
vo
l. 16, pp. 25
1-63, Jun 2012
.
[44]
P. Xidonas
, et al.
, "On the select
i
on of equit
y
securities: An ex
p
e
r
t
s
y
stem
s m
e
thodolog
y
and an application on th
e
Athens Stock
Ex
change,"
Expert
Systems with
Ap
plications,
vo
l. 3
6
, pp
. 11966-11
980, 2009
.
[45]
W.
Shi
u
e
, et
a
l
.
, "A frame kno
wledge s
y
s
t
em f
o
r ma
naging f
i
n
a
ncial d
ecision
knowledge,"
Expe
rt Sy
ste
m
s with
Applica
tions,
vo
l. 35
, pp
. 1068-1
079, 2008
.
[46]
M.
Duta
, et al.
, "Neural networ
k analy
s
is of the mastoi
d EEG for the assessment of vigilan
ce,"
Inte
rnat
i
o
nal
Journal of Human-Computer Interaction
,
vol. 17
, pp
. 171-195
, 2
004/06/01 2004
.
[47]
J.
Kha
n
, et al.
,
"Classification and diagnostic pr
ed
iction of cancers using gene e
xpression profiling and artificial
neural networks,"
Natur
e
m
e
d
i
ci
ne,
vo
l. 7, pp. 67
3-679, 2001
.
[48]
J. Kuruvilla
an
d K. Gunav
a
thi, "Lung
cancer
classi
fication us
ing neur
al netw
orks for CT images,"
Computer
Methods and
Programs in Biom
edicine,
vol. 113
, pp
. 202-9
,
Jan
2014.
[49]
R. Alizad
ehs
a
ni
, et al.
, "A data
mining approach
for diagnos
is of
coronar
y
arter
y
disease,"
Computer Methods and
Programs in Bio
m
edicin
e,
vol. 1
11, pp
. 52-61
, Ju
l 2013.
[50]
C. Ze
cchin
, et
al.
, "Jump neur
al network for on
line short-
time prediction of
blood glucose f
r
om continuous
monitoring sensors and meal inf
o
rmation,"
Computer Methods a
nd Programs in Biomedicine,
vol. 113, pp
. 144-5
2
,
Jan 2014.
[51]
A. Kupusinac
, et al.
, "Predictin
g bod
y
fat percentag
e
based on
ge
nder,
age an
d BMI by
using
artificial neur
al
networks,"
Com
puter Methods a
nd Programs in
Biomedicine,
vo
l. 113
, pp
. 610-6
19, Feb
2014.
[52]
P
.
J
.
Lis
boa
, "A review of ev
iden
ce of
he
al
th bene
fit from
artif
ici
a
l
neural n
e
tworks in m
e
dica
l int
e
r
v
ention
,
"
Neura
l
networks,
vo
l. 1
5
, pp
. 11-39
, 20
02.
[53]
E. P
o
let
t
i
, et
al.
, "A modular fr
amework for the
automatic classification of ch
romoso
mes in
Q-band images,"
Computer Metho
d
s and Programs in Biomedicine,
vol. 105
, pp
. 12
0-130, 2012
.
[54]
J.
L.
Ama
r
al
, e
t
al
.
, "An improved
method of
ear
ly diagnosis of
sm
oking-induced respirator
y
chang
e
s using machine
learn
i
ng algorith
m
s,"
Computer Methods and
Progr
ams in Biom
edicine,
vol. 112
, pp
. 441-54
, Dec 2013.
[55]
S.
İ
çer
, et al.
, "Com
parison of
m
u
ltilay
e
r perceptron train
i
ng al
gorithm
s
for por
tal venous dopp
ler signals in th
e
cirrhosis disease,"
Expert S
y
stem
s with
Applicatio
ns,
vol. 31, pp. 4
06-413, 2006
.
[56]
Y.
Chen
, et al.
, "Constructing a nutrition di
a
gnosis e
xpe
rt sy
stem,
"
Expert Systems with Applications,
vol. 39, p
p
.
2132-2156, 201
2.
[57]
V.
Uraikul
, et
a
l
.
, "Artif
ici
a
l in
te
lligen
ce for m
o
n
itoring and sup
e
rvisor
y
con
t
rol o
f
process s
y
stem
s,"
Engineering
Applica
tions o
f
Artifi
cial
Int
e
ll
ig
ence
,
vo
l. 20, pp
. 115-131
, 2007
.
[58]
E. Ném
e
th
, et al.
, "Predic
tio
n-based diagno
si
s and loss prevent
i
on using qualit
ativ
e m
u
lti-sc
ale m
odels
,"
Information Sciences,
vo
l. 177, p
p
. 1916-1930
, 2
007.
[59]
C.L. Hsu
, et al.
, "A secur
e
IR
B s
y
stem
for
as
sisting the
dev
e
lopm
ent of
inte
l
ligent
m
e
dic
a
l
devic
e
s,"
Expert
Systems with
Ap
plications,
vo
l. 3
9
, pp
. 12512-12
521, 2012
.
[60]
L. D
y
mova
, et
al.
, "A stock tr
ading exper
t
s
y
stem based on th
e rule-bas
e evi
d
enti
al re
as
oning us
ing Level 2
Quotes,"
Expert
Systems with
Ap
plications,
vo
l. 3
9
, pp
. 7150-715
7, 2012
.
[61]
J.S. Cope
, et al.
,
"Revers
e
engin
eering exp
e
rt vi
s
u
al obs
ervation
s
: F
r
om fixation
s
to the learning
of spatial fil
t
ers
with a neur
al-gas algorithm,"
Expe
rt Sy
ste
m
s wi
th Applications,
v
o
l. 40
, pp
. 6707-
6712, 2013
.
[62]
C. Chen
, et al.
, "An integrated
architecture
fo
r fault diagnosis and failure
pr
ognosis of complex engin
eer
in
g
s
y
ste
m
s,
"
Expert Systems with
Ap
plications,
vo
l. 3
9
, pp
. 9031-904
0, 2012
.
[63]
A. Long, "Drug m
e
tabolism
in
silico - the kno
wledge-based
ex
pert s
y
stem
appr
oach. Histori
c
al perspect
ives and
current strategies,"
Drug Discov
Today Technol,
vol. 10
, pp
.
e147
-53, Spring
2013
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
IJEC
E V
o
l
.
5, No
. 1, Feb
r
uar
y
20
1
5
:
8
4
– 91
90
[64]
D.A. Linkens and H.O. Ny
ong
es
a, "Le
a
rning
s
y
s
t
em
s
in intel
ligent
control
:
a
n
apprais
a
l of f
u
zz
y, neu
r
al an
d
genetic algo
rith
m control applications,"
IE
E Proceed
ings-Control Theory and Applicat
ions,
vol. 143, pp. 367-38
6,
1996.
[65]
E.S
.
Hou
, et a
l
.
, "A genetic
alg
o
rithm
for m
u
ltiprocessor scheduling,"
Parallel and Distribut
ed Systems, IEEE
Transactions on,
vol. 5, pp. 113-
120, 1994
.
[66]
J.C. Chen
, et al.
,
"Flex
i
ble job shop
scheduling
with parallel mach
in
es using Genetic Algorith
m
and Grouping
Genetic Algor
ith
m,"
Expert
Syst
e
m
s with App
lica
t
ions,
vol. 39
, pp
. 10016-10021, 2
012.
[67]
M. Lee and
E.K. Kang, "Ter
mite
tunneling
feature extr
act
ion using
genetic algor
ith
m,"
Soft Computing,
vol. 16
, pp
.
419-425, 2011
.
[68]
K.
S.
Tang
, et al.
, "Genetic algorithms and their applications,"
Sig
nal Processing Magazine, IEEE,
vol. 13, pp. 22-
37, 1996
.
[69]
D. W
h
itl
e
y
,
"A genet
i
c
algor
ith
m
tutoria
l
,"
S
t
atistics and
computing,
vo
l. 4, pp. 6
5
-85, 1994
.
[70]
U. Kesgin, "Ge
n
eti
c
a
l
gorithm
and a
r
tif
ici
a
l
neural
ne
twork
for engin
e
op
timisation of
eff
i
ciency
and N
O
x
e
m
ission,
"
Fue
l
,
vol. 83
, pp
. 885-
895, 2004
.
[71]
T. Ba
ck
, et
al.
, "Evolution
a
r
y
co
mputation:
comment
s on the his
t
or
y
and curr
ent
state,"
Evo
l
ution
a
ry Computatio
n,
IEEE Transactio
ns on
vol. 1, pp.
3-17, 1997
.
[72]
C.F.M. To
ledo
, et
al.
, "Global o
p
timization usin
g a g
e
netic
algo
rith
m with hierarchically
s
t
ructu
r
ed population
,
"
Journal of Computational
and Ap
plied
Mathematics,
vol. 261
, pp
.
341-351, 2014
.
[73]
M
.
S
r
inivas
and
L.M
.
P
a
tna
i
k,
"Geneti
c
algori
t
h
m
s
:
a s
u
rve
y
,"
C
o
mputer,
vol. 27
, pp
. 17-26
, 199
4.
[74]
D.
A.
Coley
,
"Ge
n
e
t
ic
al
gori
t
h
ms,"
Contemporary Physics,
vol. 37
, pp. 145-154, 19
96.
[75]
C.C. L
e
e
,
"F
uz
z
y
logi
c in
con
t
rol s
y
s
t
em
s
:
fuzz
y log
i
c
contr
o
ller
.
I,"
S
y
stems, Man and Cybernetics, I
E
EE
Transactions on,
vol. 20, pp. 419
-435, 1990
.
[76]
P. Guillem
i
n, "F
uzz
y
log
i
c ap
pli
e
d to m
o
tor cont
rol,"
Industry Ap
plications,
IEEE Transactions on,
vol. 32
, pp
. 51-
56, 1996
.
[77]
T. F
a
isal
, et al
.
,
"A noninvasive intelligent
approach for pr
edicting
the risk in dengue patients,"
Ex
pe
rt Sy
ste
m
s with
Applica
tions,
vo
l. 37
, pp
. 2175-2
181, 2010
.
[78]
A. Sotelo
, et
al.
,
"Identifi
cat
ion
of epil
eps
y
s
t
ag
es
from
ECoG
using genetic pro
g
ramming
clas
s
i
fiers
,
" Com
puter
s
in Biolog
y
and
Medicin
e
, vol. 4
3
, pp
. 1713-23
,
Nov 2013.
[79]
R. L
eard
i
,
"Gene
tic
algor
ithm
s
in
chem
istr
y,"
Journal of Chromatography A,
vol. 1
158, pp
. 226-33
, Jul 27 2007
.
[80]
E.S. Moghaddam, "Design of a
Printed
Quadrifilar-Helical Anten
n
a on a Dielec
tric C
y
linder b
y
M
eans of a Genetic
Algorithm [Antenna Applicatio
ns Corner]
,
"
Antennas and Propagation Magazine, I
EEE,
vol.
53, pp. 262-268
,
2011.
[81]
R.O. R
y
an
and S. Jinjun, "Active
vibr
ation
contro
l using genetic
algorithm-ba
sed system
id
entif
ica
t
ion and posit
ive
position f
eedback,"
Smart Materials and Structures,
vol. 21
, p
.
05
5002, 2012
.
[82]
N.J. Schouten
, et
al.
, "Fuzzy
logic contro
l for pa
rallel h
y
brid
vehicles,"
Control Systems Technology, IEEE
Transactions on,
vol. 10, pp. 460
-468, 2002
.
[83]
C. Cecati
, et
a
l
.
, "A Multil
ev
el Inver
t
er
for
Photovoltaic
S
y
stem
s With Fuzzy
Log
i
c Co
ntrol,"
Industria
l
Electronics, IEEE Transactions
on,
vol. 57
, pp
. 4
115-4125, 2010
.
[84]
P.M. Aubin
, et al.
, "A Robotic Cadav
e
ric Gait
Simulator With Fuzzy
Log
i
c
Ve
rtical Ground
Reaction
Force
Control,"
Robotics, IEEE Transactions on
,
vo
l. 2
8
, pp
. 246-255
,
2012.
[85]
F. Cismondi
, et
al.
,
"Reducing
u
nnecessar
y
lab t
e
sting in
the IC
U with arti
fic
i
al
inte
llig
enc
e
,"
In
t
J
Me
d
I
n
f
o
rm,
vol. 82
, pp
. 345-
58, May
2013.
[86]
J.A. Giannin
i
and C. Kilgus,
"A
fuzzy
log
i
c technique for
correcting
climatological ion
o
spheric
models,"
Geoscience and
Remote Sens
ing
,
IEEE Transactions on,
vol. 35
,
pp. 470-474
, 19
97.
[87]
F.
Russo,
"Fuz
z
y
s
y
s
t
e
m
s in instru
mentation: fu
zzy
signal processing,"
Instrumentation and M
e
asurement, IEEE
Transactions on,
vol. 45, pp. 683
-689, 1996
.
[88]
R. Kozma
, et
a
l
.
,
"Advances in neural
n
e
twor
ks
research
: an
intr
oduction
,
"
Neur
al Networks,
vol. 22, pp
. 489-90
,
Jul-Aug 2009.
[89]
S. Kara
and M. Okandan, "Atria
l f
i
bril
lat
i
on c
l
assifica
tion wi
th
artificial neur
al networks,"
Pattern Recognition,
vol. 40
, pp
. 2967
-2973, 2007
.
[90]
R.P. Lippmann,
"Patte
rn
classification using n
e
u
r
al n
e
tworks,"
C
o
mmunications Magazine,
I
E
EE,
vol. 27
, pp
. 47
-
50, 1989
.
[91]
G.P
.
Zhang
,
"N
eural n
e
tworks
f
o
r clas
s
i
fi
ca
tion:
a s
u
rve
y
,"
Systems, Man, and
Cybernetic
s,
Pa
rt C: Applicatio
n
s
and Reviews
,
I
E
EE Transactions
on,
vol. 30, pp.
451-462, 2000
.
[92]
E. S
ezer
, et al
.
, "Emplo
y
m
ent an
d Comparison of
Different Artific
ial Neural Networks for Epileps
y
Diagnosis from
EEG Signals,"
Journal of M
e
dical Systems,
vol.
36, pp
. 347-362
, 2012/02/01
201
2.
[93]
K. Methapr
a
y
o
o
n
, et al., "An in
tegration of AN
N wind
power e
s
tim
ation in
to u
n
it com
m
itm
ent
considering
the
forecast
i
ng un
ce
rtaint
y
,
" Industr
y
App
lic
ations, I
EEE
Trans
act
ion
s
on, vol
. 43
, pp
. 1441-1448, 200
7.
[94]
E.
D.
Übey
l
ı˙
an
d
İ
. Güler, "Feature extraction from Doppler ultras
ound signals for automate
d diagnostic s
y
stems,"
Computers in Biology and
Med
i
cine,
vol. 35
, pp
.
735-764, 2005
.
[95]
D. Singhal
and
K.S. Swarup, "E
lec
t
ric
i
t
y
pr
ic
e
f
o
recast
i
ng using
art
i
fic
i
al
neur
al
networks,"
In
ternational Journ
a
l
of Electrical Po
wer
&
Energy Systems,
vol. 33
,
pp. 550-555
, 20
11.
[96]
B.
G.
Kermani
, et al.
, "Performan
ce of th
e Levenb
erg–Marquardt n
e
ur
al n
e
twork tr
aining method
in
electronic nose
applications,"
Sensors and Actuators B: Chemical,
vol. 110, pp. 13
-22, 2005
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJECE
ISS
N
:
2088-8708
IQ Cl
a
ssifica
tio
n via
Bra
i
n
w
a
ve Fea
t
u
r
es:
Review o
n
Arti
ficia
l In
tellig
ence Techn
i
qu
es
(Aisya
h
H
J
)
91
[97]
B. He
, et
a
l
.
, "
E
stim
ating m
o
n
t
hl
y tot
a
l n
itrog
en conc
entr
atio
n
in streams b
y
using arti
ficial neural network
,
"
Journal of En
vironm
ental Manag
ement,
vol. 92
, p
p
. 172-177
, 201
1.
[98]
M.
Ya
hia
, et
al
.
,
"Rough neural expert s
y
stems,"
Ex
pe
rt Sy
ste
m
s
with Applic
ations,
vol. 18, pp. 87
-99, 2000
.
[99]
F. Hay
e
s-Roth
, "
K
nowledge-B
as
ed Exp
e
rt
S
y
s
t
e
m
s
,
"
Computer,
vol. 17
, pp
. 263-
273, 1984
.
[100]
D. Chakrbort
y
a
nd N. R. P
a
l, "I
ntegra
ted f
eatur
e ana
l
y
s
is
and f
u
zz
y rul
e
-bas
ed
s
y
s
t
em
iden
tifi
c
ation
in a n
e
uro
-
fuzzy
p
a
radigm,"
Systems, Man,
and Cybernetics, Part B:
C
y
bern
etics, IEEE Transactions on,
vo
l.
31, pp. 391-40
0
,
2001.
[101]
S
.
H
a
ykin
,
Neur
al networks: a
comprehensive
fo
undation
: Pr
enti
ce Ha
ll
PTR,
19
94.
[102]
E. Grossi and M. Buscema, "Intr
oduc
tion to ar
tificial neural n
e
tw
orks,"
European Journal of Gast
roenterology an
d
Hepatology,
vo
l. 19, pp. 1046-10
54, 2007
.
[103]
M.
N.
Ta
ib
, e
t
al.
, "
E
xtend
i
ng th
e response r
a
nge of
an
optical fibre pH sensor u
s
i
ng an
artif
icial neural network,"
Analyt
ica
Chimi
c
a A
c
ta
,
vol. 330
, pp
. 31-40
, 199
6.
[104]
N.
Chen
, et al.
, "Edge Dete
ction
Based on Biom
im
etic Patt
ern Re
cognition
,
"
TELKOMNIKA Indonesian Journal of
E
l
ec
t
r
i
c
al
E
n
gi
ne
e
r
i
n
g
,
vol. 12
,
pp. 6965-6968
,
2014.
[105]
Y.
Sun
, et
al
.
, "
A
pplica
tion of
a
r
tifi
c
ia
l neur
al n
e
tworks in th
e
d
e
sign of con
t
rolled rel
ease drug
deliv
er
y
s
y
stems,"
Advanc
ed
Dr
ug Deliv
er
y Rev
i
ew
s
,
vol. 55, pp. 12
01-1215, 2003
.
[106]
A.E. Zad
e
h
, et
al.
, "Clas
s
i
fi
ca
tion of
the
ele
c
t
rocard
i
ogram
s
i
gnals
us
ing s
u
pervis
ed
clas
s
i
fi
ers
and
effic
i
en
t
featur
es
,"
Comp
uter Methods an
d Pr
ograms in Biomedicine,
vol.
99, pp
. 179-194
, 2010.
[107]
A.
H.
Jahidin
, e
t
al.
, "Classification of in
tellig
en
ce quotient via brainw
ave su
b-ba
nd power r
a
tio f
eatu
r
es an
d
artif
icial neural
network,"
Comp
uter Methods an
d Pr
ograms in Biomedicine,
vol.
114, pp
. 50-59
,
2014.
[108]
F.
Ama
t
o
, et al.
, "Artificial neural networ
ks in me
dical diagnosis,"
Journal of Applied Biomed
icin
e,
vol. 11, pp. 47-
58, 2013
.
[109]
A.
K.
Jain
, et
al
.
, "Statistical p
a
ttern
r
ecognition
:
A review,"
Pa
ttern Anal
ysis and Machine In
t
e
llig
ence, I
E
EE
Transactions on,
vol. 22, pp. 4-3
7
, 2000
.
BIOGRAP
HI
ES OF
AUTH
ORS
Ais
y
ah Har
tini Jahidin ob
tain
ed the B.Eng (Tel
ecommunication) and M.Eng.
S
c
(El
ectr
i
ca
l) from
Universiti Mal
a
ya
, Mal
a
y
s
ia
. She is current
l
y
a postgraduat
e
research
er at
t
h
e Facult
y of
Electrical Engin
eering
,
Univ
ersiti Teknologi MARA, Malay
s
ia.
Her main r
e
search
inter
e
sts
includ
e hum
an intell
igen
ce, cog
n
itive
abil
it
y
,
E
E
G and non-linear m
odelling of
brain behav
i
or
via
inte
llig
ent
si
gnal pro
cessing
t
echniqu
e.
Mohd Nasir Taib obtained the
B.Eng (Electr
i
cal)
from the Uni
v
ersity
of Tasmania, Australia,
M.Sc. (Control
S
y
stems) from
University
of She
ffield,
and Ph.D. (Control
& I
n
strumentation
)
from
Universit
y
of Manch
e
ster
Institute of
Science
and T
echno
log
y
, Uni
t
ed Ki
ngdom
. He is
current
l
y
a Profe
ssor and the Dea
n
of the Facu
lt
y
of Ele
c
tri
c
a
l
Eng
i
neer
ing, Univ
er
siti Tekno
logi
MARA, Malay
s
ia. He is leading
an active r
e
search group and supe
rvising a pool
of research
ers
in advan
ced sig
n
al processing with application
s
in control s
y
s
t
ems and process, biomedical
engineering, and
nonlin
ear
s
y
stems.
Nooritawati Md
Tahir received
the B
.
Eng (
E
l
ectronics) from th
e Universiti
Tek
nologi MARA,
Malay
s
ia, M.Sc. (Micro
electr
o
n
ics & Teleco
mmunications) fro
m University
of Liv
e
rpool,
United Kingdo
m, and Ph.D. in Electrical E
ngineer
ing (Pattern Recognition
& Artificial
Intell
igen
ce) fro
m
Universiti Kebangsaan Mal
a
ysia
, Mala
ysi
a
.
She is currentl
y
an Associate
Professor at the Faculty
of Electrical
Engine
er
ing and the Dir
ector of R
e
sear
ch Innovation
Business Unit, Universiti T
e
kn
ologi MARA, Mala
y
s
ia
. Her r
e
search
inter
e
sts include
im
age
processing,
pat
t
e
r
n recogn
ition
,
c
o
m
puter vision
a
nd art
i
fic
i
a
l
in
tel
ligen
ce.
Megat S
y
ahirul
Am
in Megat Ali rec
e
ived
th
e B.Eng (Biom
e
dic
a
l) from
Universiti Ma
la
ya
,
Malay
s
ia, and
M.Sc. (Biomedical
Engin
eering
)
from University of Surrey
,
United Kingdom. He
is currentl
y
a senior lecturer at
the Facult
y
of
Elect
rical Engi
neering
,
Univer
siti Teknolog
i
MARA, Mala
y
s
ia. His resear
c
h
interests inc
l
ude EEG and
intel
ligen
t m
odelling of bra
i
n
behavior
with
ap
plic
ation
to
expe
rienti
al
le
arning
theor
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.