I
A
E
S
I
n
t
e
r
n
at
io
n
al
Jou
r
n
al
of
A
r
t
if
ic
ia
l
I
n
t
e
ll
ig
e
n
c
e
(
I
J
-
AI
)
V
ol
.
14
, N
o.
4
,
A
ugus
t
20
25
, pp.
2839
~
2848
I
S
S
N
:
2252
-
8938
,
D
O
I
:
10.11591/
ij
a
i.
v
14
.i
4
.pp
2839
-
2848
2839
Jou
r
n
al
h
om
e
page
:
ht
tp
:
//
ij
ai
.
ia
e
s
c
or
e
.c
om
Im
p
r
ovi
n
g
f
i
r
e
w
al
l
p
e
r
f
or
m
an
c
e
u
si
n
g h
yb
r
i
d
of
op
t
i
m
i
z
at
i
on
al
gor
i
t
h
m
s an
d
d
e
c
i
si
on
t
r
e
e
s c
l
ass
i
f
i
e
r
M
os
le
h
M
. A
b
u
al
h
aj
1
, A
h
m
ad
A
d
e
l
A
b
u
-
S
h
ar
e
h
a
2
, S
u
m
aya
N
ab
il
Al
-
K
h
at
i
b
1
,
A
d
e
e
b
M
. A
ls
aai
d
ah
1
,
M
oh
am
m
e
d
A
n
b
ar
3
1
D
e
pa
r
t
m
e
nt
of
N
e
t
w
or
ks
a
nd C
ybe
r
s
e
c
ur
i
t
y,
F
a
c
ul
t
y of
I
nf
or
m
a
t
i
on
T
e
c
hnol
og
y
,
Al
-
A
hl
i
yya
A
m
m
a
n U
ni
ve
r
s
i
t
y, A
m
m
a
n, J
or
da
n
2
D
e
pa
r
t
m
e
nt
of
D
a
t
a
S
c
i
e
nc
e
a
nd A
r
t
i
f
i
c
i
a
l
I
nt
e
l
l
i
ge
nc
e
,
F
a
c
ul
t
y of
I
nf
or
m
a
t
i
on
T
e
c
hnol
ogy
,
Al
-
A
hl
i
yya
A
m
m
a
n U
ni
ve
r
s
i
t
y,
A
m
m
a
n, J
or
da
n
3
C
ybe
r
s
e
c
ur
i
t
y
R
e
s
e
a
r
c
h
C
e
nt
e
r
,
U
ni
ve
r
s
i
t
i
S
a
i
ns
M
a
l
a
y
s
i
a
, P
e
na
ng, M
a
l
a
ys
i
a
A
r
t
ic
le
I
n
f
o
A
B
S
T
R
A
C
T
A
r
ti
c
le
h
is
to
r
y
:
R
e
c
e
iv
e
d
S
e
p
24
,
2024
R
e
vi
s
e
d
M
a
r
16
,
2025
A
c
c
e
pt
e
d
J
un
8
,
2025
One
of
the
primary
concerns
of
governments,
corporations,
and
even
individual
users
is
their
level
of
online
protection.
This
is
because
a
large
number
of
attacks
target
their
primary
assets.
A
firewall
is
a
critical
to
ol
that
almost
every
organization
uses
to
protect
its
assets.
However,
fir
ewalls
become less reliable when
they deal with
large amounts
of data. One
method
for
reducing
the
amount
of
data
and
enhancing
firewall
performa
nce
is
feature
selection.
The
main
aim
of
this
study
is
to
enhance
the
fir
ewall'
s
performance
by
proposing
a
new
feature
selection
m
ethod.
The
pr
oposed
feature
selection
method
combines
the
strengths
of
Harris
Hawks
optimization
(HHO)
and
whale
optimization
algorithm
(WOA).
Exper
iments
were
performed
utilizing
the
NSL
-
KDD
dataset
to
measure
the
effectiveness
of
the
proposed
method.
The
experiments
employed
the
decision
trees
(DTs)
as
a
machine
classifier.
The
experimental
results
show
that
the
ac
hieved
accuracy
is
98.46%
when
using
HHO/WOA
for
feature
selectio
n
and DT
for
classif
ication,
outperfor
ming
the
HHO
and
WOA
when
used
separat
ely
for
feature
selection.
The
study'
s
findings
offer
insight
ful
informati
on
for
researchers
and
practition
ers
looking
to
improve
firewall
effectivene
ss
and
efficiency i
n defendin
g intern
et connect
ions agai
nst chang
ing th
reats
.
K
e
y
w
o
r
d
s
:
D
e
c
is
io
n t
r
e
e
s
F
e
a
tu
r
e
s
e
le
c
ti
on
F
ir
e
w
a
ll
H
a
r
r
is
h
a
w
ks
a
lg
or
it
hm
W
ha
le
opt
im
iz
a
ti
on a
lg
or
it
hm
This is an
open
acce
ss artic
le unde
r the
CC BY
-
SA
license.
C
or
r
e
s
pon
di
n
g A
u
th
or
:
M
os
le
h M
.
A
bua
lh
a
j
D
e
pa
r
tm
e
nt
of
N
e
twor
ks
a
nd C
ybe
r
s
e
c
ur
it
y,
F
a
c
ul
ty
of
I
nf
or
m
a
ti
on
T
e
c
hnol
ogy
Al
-
A
hl
iy
ya
A
m
m
a
n
U
ni
ve
r
s
it
y
A
m
m
a
n
19111, J
or
da
n
E
m
a
il
:
m
.a
bua
lh
a
j@a
m
m
a
nu.e
du.j
o
1.
I
N
T
R
O
D
U
C
T
I
O
N
C
ybe
r
a
tt
a
c
ks
a
r
e
de
li
be
r
a
t
e
a
tt
e
m
pt
s
to
ha
c
k
or
ta
ke
a
dva
nt
a
g
e
of
c
om
put
e
r
s
ys
te
m
s
,
ne
twor
ks
,
or
ot
he
r
te
c
hnol
ogy.
T
he
num
be
r
of
c
ybe
r
a
tt
a
c
k
s
is
a
pr
obl
e
m
t
ha
t
is
c
ont
in
ua
ll
y
c
h
a
ngi
ng
a
nd
e
xpa
ndi
ng
a
s
m
or
e
c
om
pa
ni
e
s
,
or
ga
ni
z
a
ti
ons
,
a
nd
pe
opl
e
r
e
ly
on
di
gi
ta
l
te
c
hnol
ogi
e
s
to
s
to
r
e
a
nd
s
e
nd
s
e
ns
it
iv
e
in
f
or
m
a
ti
on.
C
ybe
r
a
tt
a
c
ks
s
pr
e
a
d
f
r
om
one
ne
twor
k
or
s
y
s
te
m
to
a
not
he
r
[
1]
–
[
3]
.
S
e
ve
r
a
l
r
e
por
ts
in
di
c
a
te
th
a
t,
dur
in
g
th
e
pa
s
t
f
e
w
ye
a
r
s
,
th
e
num
be
r
of
c
ybe
r
a
tt
a
c
ks
h
a
s
be
e
n
c
ont
in
uous
ly
r
is
in
g.
F
or
in
s
ta
nc
e
,
th
e
num
be
r
of
phi
s
hi
ng
w
e
bs
it
e
s
in
c
r
e
a
s
e
d
by
350%
in
2020
[
4]
,
a
nd
th
e
num
be
r
of
r
a
ns
om
w
a
r
e
a
tt
a
c
ks
in
c
r
e
a
s
e
d by 400%
[
5]
.
A
f
ir
e
w
a
ll
is
a
to
ol
w
id
e
ly
us
e
d
by
or
ga
ni
z
a
ti
ons
to
pr
ot
e
c
t
th
e
ir
a
s
s
e
ts
.
F
ir
e
w
a
ll
s
a
na
ly
z
e
n
e
twor
k
da
ta
in
r
e
a
l
-
ti
m
e
,
c
ont
r
a
s
ti
ng
it
w
it
h
known
pa
tt
e
r
ns
of
m
a
li
c
io
us
a
c
ti
vi
ty
a
nd
a
ppl
yi
ng
a
lg
or
it
hm
s
to
f
in
d
pot
e
nt
ia
l
th
r
e
a
ts
.
F
ig
ur
e
1
c
la
r
if
ie
s
th
e
r
o
le
of
f
ir
e
w
a
ll
s
.
I
t
is
c
r
uc
ia
l
to
r
e
m
e
m
be
r
th
a
t
th
e
y
a
r
e
not
f
ool
pr
oo
f
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
14
, N
o.
4
,
A
ugus
t
20
25
:
2839
-
2848
2840
a
nd
c
a
n
be
d
e
f
e
a
te
d
by
c
unni
ng
a
tt
a
c
ke
r
s
[
6]
–
[
8]
.
T
he
r
e
f
or
e
,
bui
ld
in
g
s
ys
te
m
s
w
it
h
a
dva
nc
e
d
a
lg
or
it
hm
s
is
c
r
uc
ia
l
to
pr
ovi
di
ng c
om
pr
e
he
ns
iv
e
pr
ot
e
c
ti
on a
ga
in
s
t
c
ybe
r
a
tt
a
c
ks
.
F
ig
ur
e
1.
F
ir
e
w
a
ll
f
unc
ti
on
M
ode
r
n
f
ir
e
w
a
ll
s
us
e
m
a
c
hi
ne
le
a
r
ni
ng
(
M
L
)
te
c
hni
que
s
to
s
t
op
th
e
ne
w
ty
pe
s
of
c
ybe
r
a
tt
a
c
k
s
.
B
y
in
c
or
por
a
ti
ng
M
L
in
to
f
ir
e
w
a
ll
s
,
s
e
c
ur
it
y
i
s
s
ue
s
c
a
n
be
di
s
c
o
ve
r
e
d
a
nd
a
voi
de
d
w
it
h
in
c
r
e
di
bl
e
s
pe
e
d
a
nd
a
c
c
ur
a
c
y.
H
ow
e
ve
r
,
in
a
c
c
ur
a
te
da
ta
c
la
s
s
if
ic
a
ti
on
is
a
c
om
m
o
n
pr
obl
e
m
th
a
t
f
ir
e
w
a
ll
-
ba
s
e
d
M
L
f
r
e
que
nt
ly
f
a
c
e
s
.
I
na
c
c
ur
a
t
e
da
ta
c
l
a
s
s
if
ic
a
ti
on
h
a
ppe
ns
w
he
n
a
f
ir
e
w
a
ll
i
nc
or
r
e
c
tl
y
la
be
ls
a
r
e
gul
a
r
n
e
twor
k
a
c
ti
vi
ty
a
s
m
a
li
c
io
us
, w
a
s
ti
ng r
e
s
our
c
e
s
a
nd c
a
us
in
g unwa
nt
e
d a
la
r
m
s
[
9]
–
[
11]
.
F
e
a
tu
r
e
s
e
le
c
ti
on
is
a
w
id
e
ly
ut
il
iz
e
d
te
c
hni
que
in
f
ir
e
w
a
ll
-
ba
s
e
d
M
L
to
r
e
duc
e
th
e
in
a
c
c
ur
a
c
y
of
da
ta
c
la
s
s
if
ic
a
ti
on.
S
e
le
c
ti
ng
th
e
f
e
a
tu
r
e
s
or
v
a
r
ia
bl
e
s
m
os
t
li
ke
ly
to
di
s
ti
ngui
s
h
be
twe
e
n
m
a
li
c
io
u
s
a
nd
le
gi
ti
m
a
te
da
ta
is
e
s
s
e
nt
ia
l.
T
h
e
M
L
a
lg
or
it
hm
s
c
a
n
id
e
nt
if
y
ne
twor
k
da
ta
m
or
e
a
c
c
ur
a
te
ly
by
e
m
pha
s
iz
in
g
th
e
m
os
t
va
lu
a
bl
e
f
e
a
tu
r
e
s
[
12]
,
[
13]
.
F
e
a
tu
r
e
s
e
le
c
ti
on
c
a
n
be
im
pl
e
m
e
nt
e
d
us
in
g
f
il
te
r
-
ba
s
e
d
a
nd
w
r
a
ppe
r
-
ba
s
e
d
te
c
hni
qu
e
s
.
T
he
f
il
te
r
-
ba
s
e
d
te
c
hni
que
s
s
e
le
c
t
f
e
a
tu
r
e
s
f
e
a
tu
r
e
-
by
-
f
e
a
tu
r
e
,
w
hi
c
h
e
nf
or
c
e
s
in
de
pe
nde
nc
ie
s
be
twe
e
n
f
e
a
tu
r
e
s
.
W
r
a
ppe
r
-
ba
s
e
d
t
e
c
hni
que
s
s
e
le
c
t
f
e
a
tu
r
e
s
c
ol
la
bor
a
ti
ve
ly
,
ye
t
th
e
y
r
e
qui
r
e
va
s
t
a
m
ount
s
of
ti
m
e
a
nd
r
e
s
our
c
e
s
,
a
s
a
ll
pos
s
ib
le
f
e
a
tu
r
e
c
om
bi
na
ti
ons
s
houl
d
be
te
s
te
d
to
pr
oduc
e
th
e
out
put
s
e
t
[
14]
–
[
16]
.
A
c
c
or
di
ngl
y,
m
e
ta
he
ur
is
ti
c
a
lg
or
it
hm
s
e
a
s
e
th
e
c
om
put
a
ti
ona
l
r
e
qui
r
e
m
e
nt
s
of
w
r
a
ppe
r
-
ba
s
e
d
f
e
a
tu
r
e
s
e
le
c
ti
on.
T
hi
s
pa
pe
r
w
il
l
us
e
th
e
H
a
r
r
is
H
a
w
ks
opt
im
iz
a
ti
on
(
H
H
O
)
a
nd
w
ha
le
opt
im
iz
a
ti
on a
lg
or
it
hm
(
W
O
A
)
m
e
ta
he
ur
is
ti
c
a
lg
or
it
hm
s
t
o s
e
le
c
t
f
e
a
tu
r
e
s
f
or
f
ir
e
w
a
ll
-
ba
s
e
d M
L
.
N
um
e
r
ous
w
or
ks
ha
ve
b
e
e
n
pr
opos
e
d
to
m
it
ig
a
te
e
m
e
r
gi
ng
c
ybe
r
a
tt
a
c
ks
.
T
he
a
ut
hor
s
in
[
17]
s
ugge
s
te
d
th
e
double
-
la
ye
r
e
d
hybr
id
a
ppr
oa
c
h
(
D
L
H
A
)
to
ha
ndl
e
th
e
is
s
u
e
of
th
e
la
r
ge
di
f
f
e
r
e
nc
e
in
th
e
pa
tt
e
r
ns
of
a
tt
a
c
ks
w
he
n
us
in
g
ne
twor
k
in
tr
us
io
n
de
te
c
ti
on
s
ys
te
m
(
N
I
D
S
)
.
T
he
f
ir
s
t
la
ye
r
of
D
L
H
A
us
e
s
a
na
iv
e
B
a
ye
s
(
N
B
)
c
la
s
s
if
ie
r
to
de
te
c
t
de
ni
a
l
of
s
e
r
vi
c
e
(
D
oS
)
a
nd
pr
obe
a
tt
a
c
ks
.
T
he
s
e
c
ond
la
ye
r
of
D
L
H
A
us
e
s
a
s
uppor
t
ve
c
to
r
m
a
c
hi
ne
(
S
V
M
)
c
la
s
s
if
ie
r
to
de
te
c
t
r
e
m
ot
e
to
lo
c
a
l
(
R
2L
)
a
nd
us
e
r
to
r
oo
t
(
U
2R
)
a
tt
a
c
ks
.
T
he
D
L
H
A
a
ppr
oa
c
h
c
om
bi
ne
s
th
e
out
put
s
of
bot
h
l
a
ye
r
s
(
N
B
a
nd
S
V
M
la
ye
r
s
)
to
c
a
te
gor
iz
e
th
e
ne
twor
k
tr
a
f
f
ic
a
s
nor
m
a
l
or
a
nom
a
lo
us
,
w
hi
c
h
e
nha
nc
e
s
a
c
c
ur
a
c
y
a
nd
le
s
s
e
n
s
th
e
f
a
ls
e
-
pos
it
iv
e
r
a
te
.
T
he
s
ugge
s
te
d
a
ppr
oa
c
h
a
c
hi
e
ve
s
a
n
a
c
c
ur
a
c
y
of
88.97%
a
nd
a
f
a
l
s
e
-
pos
it
iv
e
r
a
te
of
0.12%
on
th
e
w
id
e
ly
us
e
d
N
S
L
-
K
D
D
da
ta
s
e
t.
A
c
c
or
di
ng
to
M
ugha
id
e
t
al
.
[
18]
,
th
e
de
te
c
ti
on
m
ode
l
u
s
in
g
M
L
te
c
hni
que
s
ha
s
be
e
n
pr
opo
s
e
d
by
s
pl
it
ti
ng
th
e
da
ta
s
e
t
f
or
th
e
d
e
te
c
ti
on
m
ode
l
tr
a
in
in
g
a
nd
r
e
s
ul
ts
va
li
da
ti
on.
A
l
s
o,
th
is
w
or
k
a
im
s
to
c
a
pt
ur
e
in
he
r
e
nt
c
ha
r
a
c
te
r
is
ti
c
s
f
r
om
e
m
a
il
te
xt
a
lo
ng
w
it
h
ot
he
r
f
e
a
tu
r
e
s
.
T
he
s
e
f
e
a
tu
r
e
s
a
r
e
c
la
s
s
ifi
e
d
a
s
phi
s
hi
ng
or
non
-
phi
s
hi
ng i
nvol
vi
ng t
hr
e
e
di
f
f
e
r
e
nt
da
ta
s
e
ts
. T
he
e
va
lu
a
ti
on ha
d be
e
n c
onduc
te
d b
a
s
e
d on thr
e
e
s
upe
r
vi
s
e
d
da
ta
s
e
ts
,
th
e
n
m
a
de
a
c
om
pa
r
is
on
be
twe
e
n
th
e
s
e
c
la
s
s
ifi
e
r
s
. T
h
e
m
a
in
f
in
di
ng
of
th
i
s
w
or
k
i
s
th
e
hi
gh
le
v
e
l
of
a
c
c
ur
a
c
y
w
h
e
n
u
s
in
g
phi
s
hi
ng
e
m
a
il
de
t
e
c
ti
on. T
he
not
ic
e
a
bl
e
r
e
s
ul
ts
c
ol
le
c
te
d
f
r
om
th
e
c
om
pa
r
is
on
be
twe
e
n
a
lg
or
it
hm
s
a
r
e
ba
s
e
d
on
th
e
m
ul
ti
-
f
e
a
tu
r
e
of
(
50)
,
w
hi
c
h
in
tu
r
n
obt
a
in
s
th
e
hi
ghe
s
t
a
c
c
ur
a
c
y. H
ow
e
ve
r
,
w
hi
le
us
in
g
f
e
w
e
r
f
e
a
tu
r
e
s
th
a
n
20,
th
e
a
c
c
ur
a
c
y
r
e
gi
s
te
r
e
d
a
n
a
c
c
e
pt
a
bl
e
va
lu
e
,
but
th
is
s
ta
tu
s
is
not
e
f
f
e
c
ti
ve
e
nough
to
de
te
c
t
phi
s
hi
ng
e
m
a
il
s
.
T
he
ove
r
a
ll
f
in
di
ng
of
th
is
w
or
k
is
th
a
t
th
e
be
s
t
M
L
a
lg
or
it
hm
a
c
c
ur
a
c
ie
s
a
r
e
0.88%
, 0.97%
,
a
nd 100%
c
ons
e
c
ut
iv
e
ly
f
or
s
tr
e
ngt
he
ni
ng t
he
d
e
c
is
io
n t
r
e
e
(
D
T
)
on t
he
a
ppl
ie
d d
a
ta
s
e
t
s
.
L
iu
e
t
al
.
[
19
]
pr
opos
e
a
nove
l
a
ppr
oa
c
h
f
or
de
te
c
ti
ng
ne
twor
k
in
tr
us
io
ns
in
im
ba
la
nc
e
d
ne
twor
k
tr
a
f
f
ic
da
ta
,
w
he
r
e
th
e
num
be
r
of
nor
m
a
l
ne
twor
k
tr
a
f
f
ic
in
s
ta
nc
e
s
s
ig
ni
f
ic
a
nt
ly
out
w
e
ig
hs
th
e
num
be
r
of
in
tr
us
io
n
in
s
ta
nc
e
s
.
T
h
e
s
ugge
s
te
d
di
f
f
ic
ul
t
s
e
t
s
a
m
pl
in
g
te
c
hni
que
(
D
S
S
T
E
)
a
ppr
oa
c
h
us
e
s
bot
h
M
L
a
nd
de
e
p
le
a
r
ni
ng
to
ha
ndl
e
th
is
is
s
ue
.
T
he
D
S
S
T
E
te
c
hni
que
le
s
s
e
ns
th
e
im
ba
la
nc
e
of
th
e
or
ig
in
a
l
tr
a
in
in
g
s
e
t
a
nd
pr
ovi
de
s
ta
r
ge
te
d
da
ta
a
ugm
e
nt
a
ti
on
f
or
th
e
unde
r
r
e
pr
e
s
e
nt
e
d
c
la
s
s
th
a
t
ne
e
ds
to
le
a
r
n.
T
he
r
e
f
or
e
,
th
e
D
S
S
T
E
te
c
hni
que
e
na
bl
e
s
th
e
c
l
a
s
s
if
ie
r
to
pe
r
f
or
m
be
tt
e
r
dur
in
g
c
la
s
s
if
ic
a
ti
on
a
nd
be
tt
e
r
le
a
r
n
th
e
E
x
te
r
n
a
l
N
e
tw
o
r
k
I
n
te
r
n
a
l
N
e
tw
o
r
k
D
r
o
p
I
n
c
o
m
in
g
Tr
a
f
f
ic
B
e
n
i
g
n
T
r
a
f
f
i
c
M
a
l
i
c
i
o
u
s
T
r
a
f
f
i
c
F
i
r
ew
a
l
l
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
I
m
pr
ov
in
g f
ir
e
w
al
l
pe
r
fo
r
m
anc
e
us
in
g hy
b
r
id
of
opt
imi
z
at
io
n al
gor
it
hm
s
and
…
(
M
os
le
h M
. A
bual
haj
)
2841
di
s
ti
nc
ti
ons
dur
in
g t
he
t
r
a
in
in
g s
ta
ge
. T
he
t
e
s
t
f
in
di
ngs
s
how
t
h
a
t
th
e
s
ugge
s
te
d
D
S
S
T
E
t
e
c
hni
que
a
c
hi
e
v
e
s
a
n
a
c
c
ur
a
c
y of
82.84%
, pr
e
c
is
io
n 84.64%
, a
nd r
e
c
a
ll
82.78%
, i
n m
ul
ti
c
la
s
s
c
la
s
s
if
ic
a
ti
on.
2.
M
E
T
H
O
D
2.1.
N
S
L
-
K
D
D
d
at
as
e
t
N
S
L
-
K
D
D
is
c
ons
id
e
r
e
d
a
s
uf
f
ic
ie
nt
da
t
a
s
e
t
th
a
t
he
lp
s
s
e
c
ur
it
y
r
e
s
e
a
r
c
he
r
s
in
in
v
e
s
ti
ga
ti
ng
num
e
r
ous
f
ir
e
w
a
ll
s
. I
t
is
pos
s
ib
le
t
o s
uc
c
e
s
s
f
ul
ly
c
onduc
t
th
e
e
x
pe
r
im
e
nt
s
a
nd a
na
ly
z
e
t
he
out
c
om
e
s
us
in
g t
he
N
S
L
-
K
D
D
da
ta
s
e
t
s
in
c
e
it
ha
s
a
s
uf
f
ic
ie
nt
num
be
r
of
r
e
c
or
ds
[
20]
.
T
he
N
S
L
-
K
D
D
da
ta
s
e
t
c
ont
a
in
s
148
,
517
s
a
m
pl
e
s
a
nd
41
f
e
a
tu
r
e
s
in
c
lu
di
ng
th
e
l
a
be
l
c
ol
um
n.
T
he
r
e
a
r
e
38
ty
pe
s
of
a
tt
a
c
k
s
in
th
e
N
S
L
-
K
D
D
da
ta
s
e
t
gr
ope
d i
nt
o f
our
m
a
in
t
ype
s
:
‒
D
oS
a
tt
a
c
k:
T
he
D
oS
a
tt
a
c
k
a
im
s
to
m
a
ke
a
ne
twor
k
or
s
ys
te
m
una
va
il
a
bl
e
by
ove
r
w
he
lm
in
g
it
w
it
h
tr
a
f
f
ic
or
r
e
que
s
ts
.
‒
P
r
obe
a
tt
a
c
k:
T
he
p
r
obe
a
tt
a
c
k
in
vol
ve
s
th
e
a
tt
a
c
ke
r
a
tt
e
m
pt
in
g
to
ga
th
e
r
in
f
or
m
a
ti
on
a
bout
th
e
ta
r
ge
t
ne
twor
k or
s
ys
te
m
.
‒
R
oot
U
2R
a
tt
a
c
k:
t
he
U
2R
a
tt
a
c
k
in
vol
ve
s
a
n
una
ut
hor
iz
e
d
us
e
r
ga
in
in
g
e
le
va
te
d
pr
iv
il
e
ge
s
on
a
ta
r
ge
t
s
ys
te
m
.
‒
R
2L
a
tt
a
c
ks
:
T
h
e
R
2L
a
tt
a
c
ks
in
vol
ve
s
a
n
a
tt
a
c
ke
r
ga
in
in
g
a
c
c
e
s
s
to
a
s
ys
te
m
th
r
ough
a
r
e
m
ot
e
c
onne
c
ti
on, s
uc
h a
s
e
xpl
oi
ti
ng a
vul
ne
r
a
bi
li
ty
i
n a
s
e
r
vi
c
e
or
a
p
pl
ic
a
ti
on.
B
e
s
id
e
s
,
th
e
N
S
L
-
K
D
D
da
ta
s
e
t
c
ont
a
in
s
a
"
n
or
m
a
l,
”
ty
pe
w
hi
c
h
r
e
pr
e
s
e
nt
s
r
e
gul
a
r
ne
twor
k
tr
a
f
f
ic
[
20]
.
T
he
num
be
r
of
r
e
c
or
ds
i
n t
he
N
S
L
-
K
D
D
da
ta
s
e
t
is
br
oke
n down by
a
tt
a
c
k t
ype
i
n T
a
bl
e
1.
T
a
bl
e
1.
N
um
be
r
of
r
e
c
or
ds
f
or
e
a
c
h a
tt
a
c
k
A
t
t
a
c
k
t
ype
N
um
be
r
of
r
e
c
or
ds
D
oS
53
,
387
P
r
obe
14
,
077
U
2R
119
R
2L
3
,
880
N
or
m
a
l
77
,
055
2.2.
F
e
at
u
r
e
s
e
le
c
t
io
n
u
s
in
g
H
ar
r
is
H
aw
k
s
op
t
im
iz
at
io
n
an
d
w
h
al
e
op
t
im
iz
at
io
n
al
gor
it
h
m
S
e
le
c
ti
ng
th
e
m
os
t
pe
r
ti
ne
nt
f
e
a
tu
r
e
s
or
v
a
r
ia
bl
e
s
f
r
om
a
da
ta
s
e
t
to
in
c
lu
de
in
a
m
ode
l
i
s
known
a
s
f
e
a
tu
r
e
s
e
le
c
ti
on
a
nd
is
a
c
r
it
ic
a
l
s
te
p
in
a
n
M
L
m
ode
l.
D
ue
to
th
e
ir
c
a
pa
c
it
y
to
s
c
a
n
th
e
w
hol
e
f
e
a
tu
r
e
s
pa
c
e
a
nd
id
e
nt
if
y
th
e
id
e
a
l
s
ubs
e
t
of
f
e
a
tu
r
e
s
,
opt
im
iz
a
ti
on
a
lg
or
it
hm
s
a
r
e
f
r
e
que
nt
ly
ut
il
iz
e
d
in
f
e
a
tu
r
e
s
e
le
c
ti
on.
A
s
m
e
nt
io
ne
d
e
a
r
li
e
r
,
th
e
H
H
O
a
nd
W
O
A
opt
im
iz
a
ti
on
a
lg
or
it
hm
s
w
il
l
be
u
s
e
d
to
s
e
le
c
t
th
e
a
tt
a
c
k
f
e
a
tu
r
e
s
th
a
t
th
e
f
ir
e
w
a
ll
c
a
n
us
e
to
de
te
c
t
th
e
a
tt
a
c
k
s
.
T
h
e
H
H
O
a
nd
W
O
A
ha
ve
be
e
n
w
id
e
ly
te
s
te
d
in
c
ybe
r
s
e
c
ur
it
y
a
nd
pr
ove
n
r
obus
t
a
nd
e
f
f
ic
ie
nt
.
I
n
a
ddi
ti
on,
c
om
bi
ni
ng
th
e
s
e
two
a
lg
or
it
hm
s
c
a
n
pot
e
nt
ia
ll
y
le
ve
r
a
g
e
th
e
ir
r
e
s
pe
c
ti
ve
s
tr
e
ngt
hs
,
pr
ovi
di
ng
a
m
or
e
r
obus
t
a
nd
e
f
f
e
c
ti
ve
op
ti
m
iz
a
ti
on
s
tr
a
te
gy
to
s
e
le
c
t
th
e
m
o
s
t
r
e
le
va
nt
f
e
a
tu
r
e
s
to
de
te
c
t
th
e
a
tt
a
c
ks
.
F
ur
th
e
r
m
or
e
,
W
O
A
is
r
e
now
ne
d
f
or
it
s
r
obus
t
gl
oba
l
e
xpl
o
r
a
ti
on
s
ki
ll
s
,
e
na
bl
in
g
it
to
qui
c
kl
y
na
vi
ga
te
th
e
s
e
a
r
c
h
s
pa
c
e
a
nd
a
voi
d
b
e
in
g
tr
a
ppe
d
in
lo
c
a
l
opt
im
a
.
H
H
O
de
m
ons
tr
a
te
s
e
f
f
e
c
ti
ve
e
xpl
or
a
ti
on
by
ut
il
iz
in
g
m
a
ny
s
ta
ge
s
of
hunt
in
g
be
h
a
vi
or
,
in
c
lu
di
ng
e
xpl
or
a
ti
on,
in
te
r
c
e
pt
io
n,
a
nd
a
tt
a
c
k
[
21]
,
[
22]
.
T
he
pr
opos
e
d
f
e
a
tu
r
e
s
e
le
c
ti
on
in
th
i
s
w
or
k
pr
opos
e
s
c
om
bi
ni
ng
th
e
f
e
a
tu
r
e
s
s
e
le
c
te
d
by
th
e
H
H
O
a
nd
W
O
A
opt
im
iz
a
ti
on
a
lg
or
it
hm
s
in
to
one
s
ubs
e
t
of
f
e
a
tu
r
e
s
.
T
he
H
H
O
a
lg
or
it
hm
id
e
nt
if
ie
d
a
s
ubs
e
t
of
13
f
e
a
tu
r
e
s
,
w
hi
le
th
e
W
O
A
a
lg
or
it
hm
id
e
nt
if
ie
d
a
s
ubs
e
t
of
16
f
e
a
tu
r
e
s
.
T
he
uni
on
of
th
e
s
e
two
s
ubs
e
ts
c
r
e
a
te
s
a
f
in
a
l
s
ubs
e
t
of
25
f
e
a
tu
r
e
s
.
F
ig
ur
e
2
s
how
s
th
e
pr
opo
s
e
d
f
e
a
tu
r
e
s
e
le
c
ti
on
s
te
ps
.
T
a
bl
e
2
s
how
s
th
e
c
r
e
a
te
d s
ubs
e
t
of
f
e
a
tu
r
e
s
by e
a
c
h m
e
th
od.
2.3.
D
e
c
is
io
n
t
r
e
e
c
la
s
s
i
f
ie
r
I
n
th
is
w
or
k,
th
e
D
T
c
la
s
s
if
ie
r
c
a
te
gor
iz
e
s
ne
twor
k
tr
a
f
f
ic
a
s
be
ni
gn
or
a
tt
a
c
k
tr
a
f
f
ic
.
B
a
s
e
d
on
th
e
f
e
a
tu
r
e
s
of
th
e
in
put
da
ta
,
D
T
c
la
s
s
if
ie
r
c
ons
tr
uc
ts
a
m
ode
l
of
de
c
is
io
ns
a
nd
pot
e
nt
ia
l
out
c
om
e
s
th
a
t
r
e
s
e
m
bl
e
s
a
tr
e
e
.
E
a
c
h
in
te
r
na
l
node
of
th
e
tr
e
e
r
e
pr
e
s
e
nt
s
a
d
e
c
is
io
n
ba
s
e
d
on
a
s
pe
c
if
ic
f
e
a
tu
r
e
,
a
nd
e
a
c
h
le
a
f
node
r
e
pr
e
s
e
nt
s
a
c
la
s
s
la
be
l
or
a
de
c
is
io
n
out
c
om
e
.
T
h
e
c
ons
tr
uc
ti
on
of
a
DT
s
ta
r
ts
w
it
h
th
e
e
nt
ir
e
da
ta
s
e
t,
a
nd
a
t
e
a
c
h
s
te
p,
th
e
a
lg
or
it
hm
s
e
l
e
c
ts
th
e
f
e
a
tu
r
e
th
a
t
pr
ovi
de
s
th
e
m
os
t
in
f
or
m
a
ti
on
a
bout
th
e
c
la
s
s
la
be
ls
. T
he
a
lg
or
it
hm
s
pl
it
s
th
e
da
ta
s
e
t
ba
s
e
d
on
th
e
s
e
le
c
t
e
d
f
e
a
tu
r
e
a
nd
it
s
pos
s
ib
le
va
lu
e
s
a
nd
c
r
e
a
t
e
s
a
ne
w
node
f
or
e
a
c
h
s
pl
it
.
T
he
pr
oc
e
s
s
is
r
e
p
e
a
te
d
r
e
c
ur
s
iv
e
ly
unt
il
a
s
to
ppi
ng
c
r
it
e
r
io
n
is
m
e
t,
s
uc
h
a
s
a
m
a
xi
m
um
de
pt
h
of
th
e
tr
e
e
or
a
m
in
im
um
nu
m
be
r
of
in
s
ta
nc
e
s
pe
r
le
a
f
.
F
ig
ur
e
3
c
la
r
if
ie
s
th
e
D
T
te
c
hni
que
.
T
he
f
unc
ti
on
th
a
t
w
il
l
be
us
e
d
w
it
h
D
T
in
th
e
p
r
opos
e
d
s
ys
te
m
to
m
e
a
s
ur
e
th
e
qua
li
ty
o
f
a
s
pl
it
is
"
G
in
i
im
pur
it
y"
. T
he
G
in
i
i
m
pur
it
y
is
de
f
in
e
d a
s
t
he
pr
oba
bi
li
ty
o
f
m
i
s
c
la
s
s
if
yi
ng a
r
a
ndoml
y c
hos
e
n e
le
m
e
nt
i
n t
he
s
e
t
if
i
t
w
e
r
e
r
a
ndoml
y
la
be
le
d a
c
c
or
di
ng t
o t
he
di
s
tr
ib
ut
io
n o
f
l
a
be
ls
i
n t
he
s
ubs
e
t
[
23]
, [
24]
.
A
s
i
n
(
1
)
is
us
e
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
14
, N
o.
4
,
A
ugus
t
20
25
:
2839
-
2848
2842
f
or
c
a
lc
ul
a
ti
ng
G
in
i
im
pur
it
y.
W
he
r
e
J
is
th
e
num
be
r
of
c
la
s
s
e
s
,
a
nd
p(
i)
is
th
e
pr
opor
t
io
n
of
th
e
s
a
m
pl
e
s
th
a
t
be
lo
ng t
o c
la
s
s
i
.
G
in
i
I
m
pur
it
y =
1
−
(
∑
(
i
=
1
to
J
)
p
(
i
)
2
)
(
1)
F
ig
ur
e
2
.
F
e
a
tu
r
e
s
e
le
c
ti
on pr
oc
e
s
s
T
a
bl
e
2
.
S
e
le
c
te
d f
e
a
tu
r
e
by dif
f
e
r
e
nt
m
e
th
ods
O
pt
i
m
i
z
e
r
S
e
l
e
c
t
e
d f
e
a
t
ur
e
s
(
f
e
a
t
ur
e
#)
W
O
A
S
e
r
vi
c
e
, F
l
a
g, s
r
c
_byt
e
s
,
num
_f
a
i
l
e
d_l
ogi
ns
, num
_r
oot
, num
_a
c
c
e
s
s
_f
i
l
e
s
, num
_out
bound_c
m
ds
, i
s
_hos
t
_l
ogi
n,
i
s
_gue
s
t
_l
ogi
n, s
r
v_c
ount
, s
e
r
r
or
_r
a
t
e
, s
r
v_s
e
r
r
or
_r
a
t
e
, s
a
m
e
_s
r
v_r
a
t
e
HHO
pr
ot
oc
ol
_t
ype
, F
l
a
g, s
r
c
_byt
e
s
, ds
t
_byt
e
s
, ur
ge
nt
, hot
, num
_a
c
c
e
s
s
_f
i
l
e
s
, C
ount
,
di
f
f
_s
r
v_r
a
t
e
H
H
O
&
W
O
A
pr
ot
oc
ol
_t
ype
, s
e
r
vi
c
e
, F
l
a
, s
r
c
_byt
e
s
, d
s
t
_byt
e
s
, ur
ge
nt
, hot
, num
_f
a
i
l
e
d_l
ogi
ns
, num
_r
oot
, num
_a
c
c
e
s
s
_f
i
l
e
s
,
num
_out
bound_c
m
ds
, i
s
_hos
t
_l
ogi
n, i
s
_gue
s
t
_l
ogi
n, C
ount
, s
r
v_c
ount
, s
e
r
r
or
_r
a
t
e
, s
r
v_s
e
r
r
or
_r
a
t
e
, s
a
m
e
_s
r
v_r
a
t
e
,
di
f
f
_s
r
v_r
a
t
e
, s
r
v_di
f
f
_hos
t
_r
a
t
e
, ds
t
_hos
t
_c
ount
, ds
t
_hos
t
_s
r
v_c
ount
, ds
t
_hos
t
_di
f
f
_s
r
v_r
a
t
e
,
ds
t
_hos
t
_s
a
m
e
_
s
r
c
_por
t
_r
a
t
e
, ds
t
_hos
t
_r
e
r
r
or
_r
a
t
e
F
ig
ur
e
3
.
D
T
t
e
c
hni
que
s
c
h
e
m
e
2.4.
A
t
t
ac
k
d
e
t
e
c
t
io
n
T
hi
s
s
e
c
ti
on
di
s
c
u
s
s
e
s
th
e
s
te
ps
in
vol
v
e
d
in
f
ir
e
w
a
ll
-
ba
s
e
d
M
L
de
te
c
ti
on.
F
ig
ur
e
4
s
how
s
th
e
a
tt
a
c
k
de
te
c
ti
on
s
te
ps
.
F
ir
s
t,
a
ll
th
e
non
-
num
e
r
ic
da
ta
in
th
e
N
S
L
-
K
D
D
da
ta
s
e
t
ha
s
be
e
n
tr
a
n
s
f
or
m
e
d
in
to
num
be
r
s
,
F
e
a
tu
r
e
S
e
l
e
c
ti
o
n
S
t
a
rt
H
H
O
O
p
t
i
mi
z
er
:
13
F
ea
t
u
res
W
O
A
O
p
t
i
mi
z
e
r
:
16
F
ea
t
u
res
M
u
t
u
a
l
o
f
H
H
O
&
W
O
A
:
25
F
ea
t
u
res
E
n
d
R
ed
u
c
ed
NS
L
-
K
D
D
D
a
t
a
s
e
t
(
25
F
e
a
t
u
re
s
)
NS
L
-
K
D
D
D
a
t
a
s
et
(
40
F
e
a
t
u
re
s
)
s
R
o
o
t
N
o
d
e
De
c
i
s
i
o
n
N
o
d
e
De
c
i
s
i
o
n
N
o
d
e
De
c
i
s
i
o
n
N
o
d
e
L
e
a
f
N
o
d
e
L
e
a
f
N
o
d
e
L
e
a
f
N
o
d
e
L
e
a
f
N
o
d
e
L
e
a
f
N
o
d
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
I
m
pr
ov
in
g f
ir
e
w
al
l
pe
r
fo
r
m
anc
e
us
in
g hy
b
r
id
of
opt
imi
z
at
io
n al
gor
it
hm
s
and
…
(
M
os
le
h M
. A
bual
haj
)
2843
us
in
g
th
e
l
a
be
l
-
e
nc
odi
ng
m
e
th
od,
to
e
ns
ur
e
th
a
t
im
pl
e
m
e
nt
in
g
th
e
D
T
c
la
s
s
if
ie
r
w
il
l
be
e
r
r
or
-
f
r
e
e
[
20]
,
[
25
]
.
N
e
xt
,
th
e
e
nt
ir
e
N
S
L
-
K
D
D
da
ta
s
e
t
w
a
s
nor
m
a
li
z
e
d
us
in
g
th
e
m
in
-
m
a
x
s
c
a
le
r
m
e
th
od
to
e
ns
ur
e
a
ll
da
ta
poi
nt
s
f
a
ll
w
it
hi
n
th
e
s
a
m
e
r
a
nge
.
N
or
m
a
li
z
a
ti
on
pr
e
ve
nt
s
da
ta
w
it
h
la
r
ge
r
va
lu
e
s
f
r
om
dom
in
a
ti
ng
th
e
da
ta
w
it
h
s
m
a
ll
va
lu
e
s
dur
in
g
th
e
D
T
c
la
s
s
if
ie
r
im
pl
e
m
e
nt
a
ti
on
[
20]
,
[
2
5]
.
T
he
f
in
a
l
s
te
p
in
pr
e
pa
r
in
g
th
e
da
ta
is
to
s
e
le
c
t
th
e
f
e
a
tu
r
e
s
w
it
h
th
e
hi
ghe
s
t
im
pa
c
t
on
de
te
c
ti
ng
a
tt
a
c
ks
.
T
h
e
r
e
f
or
e
,
onl
y
th
e
c
r
it
ic
a
l
f
e
a
tu
r
e
s
th
a
t
pr
ovi
de
us
e
f
ul
in
f
or
m
a
ti
on
a
r
e
us
e
d
f
or
a
tt
a
c
k
de
te
c
ti
on,
im
pr
ovi
ng
th
e
D
T
c
la
s
s
if
ie
r
'
s
a
c
c
ur
a
c
y.
T
he
pr
opos
e
d f
e
a
tu
r
e
s
s
e
le
c
ti
on me
th
od i
s
di
s
c
u
s
s
e
d i
n
s
e
c
ti
on 2.2. Af
te
r
pr
e
pa
r
in
g t
he
da
ta
, t
he
c
la
s
s
if
ic
a
ti
on
pr
oc
e
s
s
s
ta
r
ts
us
in
g
th
e
D
T
c
l
a
s
s
if
ie
r
.
T
he
D
T
c
la
s
s
if
ie
r
ha
s
be
e
n
tr
a
in
e
d
a
nd
te
s
te
d
to
m
e
a
s
ur
e
it
s
pe
r
f
or
m
a
nc
e
i
n a
tt
a
c
k de
te
c
ti
on.
F
ig
ur
e
4
.
A
tt
a
c
k de
te
c
ti
on mode
l
T
he
D
T
c
la
s
s
if
ie
r
w
a
s
im
pl
e
m
e
nt
e
d
us
in
g
th
e
K
-
f
ol
d
c
r
os
s
-
va
li
da
ti
on
m
e
th
od.
T
he
K
-
f
ol
d
m
e
th
od
di
vi
de
s
th
e
a
va
il
a
bl
e
da
ta
in
to
K
e
qua
l
-
s
iz
e
d
f
ol
ds
or
s
ubs
e
ts
,
u
s
e
s
K
-
1
f
ol
ds
f
or
m
ode
l
tr
a
in
in
g,
a
nd
us
e
s
th
e
la
s
t
f
ol
d
f
or
m
ode
l
te
s
ti
ng.
E
a
c
h
of
th
e
K
f
ol
ds
i
s
ut
il
iz
e
d
a
s
v
a
li
da
ti
on
da
ta
e
xa
c
tl
y
on
c
e
dur
in
g
th
e
K
ti
m
e
s
th
is
pr
oc
e
s
s
is
c
ondu
c
te
d.
I
n
or
de
r
to
pr
ovi
de
a
n
ove
r
a
ll
e
s
ti
m
a
te
of
th
e
m
ode
l'
s
pe
r
f
or
m
a
nc
e
,
th
e
r
e
s
ul
ts
a
r
e
a
ve
r
a
ge
d
ove
r
th
e
K
it
e
r
a
ti
ons
.
K
-
f
ol
d
c
r
os
s
-
va
li
da
ti
on
ha
s
th
e
be
ne
f
it
of
a
ll
ow
in
g
f
or
a
m
or
e
pr
e
c
is
e
e
s
ti
m
a
ti
on
of
th
e
m
ode
l'
s
pe
r
f
or
m
a
nc
e
a
nd
c
a
n
a
id
in
a
voi
di
ng
ove
r
f
it
ti
ng
[
20]
,
[
25]
.
T
he
pe
r
f
o
r
m
a
nc
e
of
th
e
D
T
c
la
s
s
if
ie
r
s
ha
s
b
e
e
n e
va
lu
a
te
d us
in
g
a
c
c
ur
a
c
y, r
e
c
a
ll
, pr
e
c
i
s
io
n, a
nd F
1
-
s
c
or
e
.
3.
R
E
S
U
L
T
S
A
N
D
D
I
S
C
U
S
S
I
O
N
T
he
r
e
s
ul
ts
of
th
e
pr
opos
e
d
f
ir
e
w
a
ll
m
ode
l
a
r
e
c
om
put
e
d
ba
s
e
d
on
th
e
e
le
m
e
nt
s
of
th
e
c
onf
us
io
n
m
a
tr
ix
:
tr
ue
pos
it
iv
e
(
T
P
o)
,
tr
ue
ne
ga
ti
ve
(
T
N
e
)
,
f
a
l
s
e
po
s
it
iv
e
(
F
P
o)
,
a
nd
f
a
ls
e
n
e
ga
ti
ve
(
F
N
e
)
.
S
e
ve
r
a
l
m
e
tr
ic
s
a
r
e
c
a
lc
ul
a
te
d
ba
s
e
d
on
th
e
s
e
e
le
m
e
nt
s
,
in
c
lu
di
ng
a
c
c
ur
a
c
y,
pr
e
c
is
io
n,
r
e
c
a
ll
,
a
nd
F
1
-
s
c
or
e
.
A
c
c
ur
a
c
y
e
va
lu
a
te
s
ove
r
a
ll
c
or
r
e
c
tn
e
s
s
but
m
a
y
be
m
i
s
le
a
di
ng
w
it
h
im
ba
la
nc
e
d
da
t
a
.
T
h
e
a
c
c
ur
a
c
y
of
th
e
pr
opos
e
d
f
ir
e
w
a
ll
m
ode
l
is
c
a
lc
ul
a
te
d
us
in
g
(
2
)
.
P
r
e
c
is
io
n
m
in
im
iz
e
s
f
a
ls
e
pos
it
iv
e
s
,
m
a
ki
ng
it
id
e
a
l
f
or
a
ppl
ic
a
ti
ons
w
he
r
e
f
a
ls
e
a
la
r
m
s
a
r
e
c
os
tl
y.
T
h
e
p
r
e
c
is
io
n
of
th
e
pr
opos
e
d
f
ir
e
w
a
ll
m
ode
l
is
c
a
lc
ul
a
te
d
us
in
g
(
3
)
.
R
e
c
a
ll
r
e
duc
e
s
f
a
ls
e
ne
ga
ti
ve
s
,
e
n
s
ur
in
g
im
por
ta
nt
in
s
ta
nc
e
s
a
r
e
not
m
is
s
e
d.
T
he
r
e
c
a
ll
of
th
e
pr
opos
e
d
f
ir
e
w
a
ll
m
ode
l
is
c
a
lc
ul
a
te
d
u
s
in
g
(
4
)
.
F
1
-
s
c
or
e
ba
la
n
c
e
s
pr
e
c
is
io
n
a
nd
r
e
c
a
ll
,
m
a
ki
ng
it
s
ui
ta
bl
e
f
or
im
ba
la
nc
e
d
da
ta
s
e
ts
.
T
he
F
1
-
s
c
or
e
of
th
e
pr
opos
e
d
f
ir
e
w
a
ll
m
ode
l
is
c
a
lc
u
la
te
d
us
in
g
(
5
)
[
20]
,
[
25]
.
T
he
s
e
f
our
m
e
tr
ic
s
ha
ve
be
e
n
c
a
lc
ul
a
t
e
d
f
or
D
T
w
it
h
H
H
O
(
D
T
-
H
H
O
)
m
e
th
od,
D
T
w
it
h
W
O
A
(
D
T
-
W
O
A
)
m
e
th
od,
a
nd
D
T
w
it
h H
H
O
/W
O
A
(
D
T
-
H
H
O
/W
O
A
)
m
e
th
od t
ha
t
is
u
s
e
d w
it
h t
h
e
pr
opos
e
d f
ir
e
w
a
ll
m
ode
l.
=
(
+
)
(
+
+
+
)
(
2)
=
(
+
)
(
3)
D
a
t
a
P
re
p
ro
c
e
s
s
i
n
g
S
t
a
r
t
Nor
m
a
l
i
z
a
t
i
on
M
i
n
-
m
a
x
S
c
a
l
e
r
F
e
a
t
u
r
e
S
e
l
e
c
t
i
o
n
H
H
O
&
W
O
A
D
a
t
a
T
r
a
n
s
f
o
r
m
a
t
i
o
n
L
a
b
e
l
E
n
c
o
d
e
r
R
e
d
u
c
e
d
N
S
L
-
K
D
D
D
a
t
a
s
e
t
(
25
F
e
a
t
u
r
e
s
)
NS
L
-
K
D
D
D
a
t
a
s
e
t
(
40
F
e
a
t
u
r
e
s
)
A
t
t
a
c
k
D
e
t
e
c
t
i
o
n
C
l
a
s
s
i
f
i
c
a
t
i
on
D
T
C
l
a
s
s
i
f
i
e
r
P
e
r
f
or
m
a
n
c
e
E
v
a
l
u
a
t
i
on
K
-
F
o
l
d
Cr
o
s
s
-
V
a
l
i
d
a
t
i
o
n
R
e
s
u
l
t
s
A
c
c
u
ra
c
y
,
R
e
c
a
l
l
,
Pr
e
c
i
s
i
o
n
,
M
C
C
,
a
n
d
F
1
-
s
c
o
re
E
n
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
14
, N
o.
4
,
A
ugus
t
20
25
:
2839
-
2848
2844
=
(
+
)
(
4)
1
−
=
+
(
5)
F
ig
ur
e
5
pr
e
s
e
nt
s
th
e
a
c
c
ur
a
c
y
a
c
hi
e
ve
d
by
th
e
pr
opo
s
e
d
f
ir
e
w
a
ll
m
ode
l.
T
he
D
T
-
H
H
O
m
e
th
od
ha
s
a
n a
c
c
ur
a
c
y of
97.59%
, t
he
D
T
-
W
O
A
m
e
th
od h
a
s
a
n a
c
c
ur
a
c
y
of
97.5%
, a
nd t
he
D
T
-
H
H
O
/W
O
A
m
e
th
od ha
s
a
n
a
c
c
ur
a
c
y
of
98.46%
.
T
he
a
c
c
ur
a
c
y
a
c
hi
e
ve
d
by
th
e
D
T
-
H
H
O
/W
O
A
m
e
th
od
out
pe
r
f
or
m
e
d
th
e
a
c
c
ur
a
c
y
a
c
hi
e
ve
d
by
th
e
D
T
-
H
H
O
m
e
th
od
a
nd
by
th
e
D
T
-
W
O
A
m
e
th
od
by
0.87%
a
nd
0.96%
,
r
e
s
pe
c
ti
ve
ly
.
T
he
r
e
f
or
e
, t
he
pr
opos
e
d D
T
-
H
H
O
/W
O
A
m
e
th
od i
m
pr
ove
s
t
he
f
ir
e
w
a
ll
'
s
de
te
c
ti
on a
tt
a
c
k a
c
c
ur
a
c
y.
F
ig
ur
e
5
. A
c
c
ur
a
c
y of
th
e
pr
opos
e
d f
ir
e
w
a
ll
m
ode
l
F
ig
ur
e
6
p
r
e
s
e
nt
s
th
e
r
e
c
a
ll
a
c
hi
e
ve
d
by
th
e
pr
opos
e
d
f
i
r
e
w
a
ll
m
ode
l.
T
he
D
T
-
H
H
O
m
e
th
od
ha
s
a
r
e
c
a
ll
of
97.59%
, t
he
D
T
-
W
O
A
m
e
th
od ha
s
a
r
e
c
a
ll
of
97.5
%
,
a
nd t
he
D
T
-
H
H
O
/W
O
A
m
e
th
od ha
s
a
r
e
c
a
ll
of
98.46%
.
T
he
r
e
c
a
ll
a
c
hi
e
ve
d
by
th
e
D
T
-
H
H
O
/W
O
A
m
e
th
od
out
pe
r
f
or
m
e
d
th
e
r
e
c
a
ll
a
c
hi
e
ve
d
by
th
e
DT
-
H
H
O
m
e
th
od
a
nd
by
th
e
D
T
-
W
O
A
m
e
th
od
by
0.87%
a
nd
0.96%
,
r
e
s
pe
c
ti
ve
ly
.
T
he
r
e
f
or
e
,
th
e
pr
opos
e
d
DT
-
H
H
O
/W
O
A
m
e
th
od i
m
pr
ove
s
t
he
f
ir
e
w
a
ll
'
s
de
t
e
c
ti
on a
tt
a
c
k r
e
c
a
ll
.
F
ig
ur
e
6
.
R
e
c
a
ll
of
th
e
pr
opos
e
d f
ir
e
w
a
ll
m
ode
l
9
7
.
5
9
%
9
7
.
5
0
%
9
8
.
4
6
%
9
7
.
0
0
%
9
7
.
2
0
%
9
7
.
4
0
%
9
7
.
6
0
%
9
7
.
8
0
%
9
8
.
0
0
%
9
8
.
2
0
%
9
8
.
4
0
%
9
8
.
6
0
%
A
c
c
u
ra
c
y
(
%
)
M
e
t
h
o
d
A
c
c
u
r
a
c
y
H
H
O
W
O
A
H
H
O
&
W
O
A
9
7
.
5
9
%
9
7
.
5
0
%
9
8
.
4
6
%
9
7
.
0
0
%
9
7
.
2
0
%
9
7
.
4
0
%
9
7
.
6
0
%
9
7
.
8
0
%
9
8
.
0
0
%
9
8
.
2
0
%
9
8
.
4
0
%
9
8
.
6
0
%
R
e
c
a
l
l
(
%
)
M
e
t
h
o
d
Re
c
a
l
l
H
H
O
W
O
A
H
H
O
&W
O
A
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
I
m
pr
ov
in
g f
ir
e
w
al
l
pe
r
fo
r
m
anc
e
us
in
g hy
b
r
id
of
opt
imi
z
at
io
n al
gor
it
hm
s
and
…
(
M
os
le
h M
. A
bual
haj
)
2845
F
ig
ur
e
7
pr
e
s
e
nt
s
th
e
pr
e
c
is
io
n
a
c
hi
e
ve
d
by
th
e
pr
opos
e
d
f
ir
e
w
a
ll
m
ode
l.
T
h
e
D
T
-
H
H
O
m
e
th
od
ha
s
a
pr
e
c
is
io
n of
97.59%
, t
he
D
T
-
W
O
A
m
e
th
od ha
s
a
pr
e
c
is
io
n of
97.5%
, a
nd t
he
D
T
-
H
H
O
/W
O
A
m
e
th
od ha
s
a
pr
e
c
is
io
n
of
98.46%
.
T
h
e
pr
e
c
i
s
io
n
a
c
hi
e
v
e
d
by
th
e
D
T
-
H
H
O
/W
O
A
m
e
th
od
out
pe
r
f
or
m
e
d
th
e
pr
e
c
is
io
n
a
c
hi
e
ve
d
by
th
e
D
T
-
H
H
O
m
e
th
od
a
nd
by
th
e
D
T
-
W
O
A
m
e
th
od
by
0.87%
a
nd
0.96%
,
r
e
s
pe
c
ti
ve
ly
.
T
he
r
e
f
or
e
, t
he
pr
opos
e
d D
T
-
H
H
O
/W
O
A
m
e
th
od i
m
pr
ove
s
t
he
f
ir
e
w
a
ll
'
s
de
te
c
ti
on a
tt
a
c
k pr
e
c
is
io
n.
F
ig
ur
e
7. P
r
e
c
is
io
n of
t
he
pr
opos
e
d f
ir
e
w
a
ll
m
ode
l
F
ig
ur
e
8
pr
e
s
e
nt
s
th
e
F
1
-
s
c
or
e
a
c
hi
e
ve
d
by
th
e
pr
opo
s
e
d
f
ir
e
w
a
ll
m
ode
l.
T
he
D
T
-
H
H
O
m
e
th
od
ha
s
a
n
F1
-
s
c
or
e
of
97.59%
,
t
he
D
T
-
W
O
A
m
e
th
od ha
s
a
n
F1
-
s
c
or
e
of
97.5%
, a
nd t
he
D
T
-
H
H
O
/W
O
A
m
e
th
od ha
s
a
n
F1
-
s
c
or
e
of
98.46%
.
T
he
F1
-
s
c
or
e
a
c
hi
e
ve
d
by
th
e
D
T
-
H
H
O
/W
O
A
m
e
th
od
out
pe
r
f
or
m
e
d
th
e
F1
-
s
c
or
e
a
c
hi
e
ve
d
by
th
e
D
T
-
H
H
O
m
e
th
od
a
nd
by
th
e
D
T
-
W
O
A
m
e
th
od
by
0.87%
a
nd
0.96%
,
r
e
s
pe
c
ti
ve
ly
.
T
he
r
e
f
or
e
, t
he
pr
opos
e
d D
T
-
H
H
O
/W
O
A
m
e
th
od i
m
pr
ove
s
t
he
f
ir
e
w
a
ll
'
s
de
te
c
ti
on a
tt
a
c
k a
n
F1
-
s
c
or
e
.
F
ig
ur
e
8. F
1
-
s
c
or
e
of
t
he
pr
opos
e
d f
ir
e
w
a
ll
m
ode
l
I
n
s
um
m
a
r
y,
th
e
s
upe
r
io
r
pe
r
f
o
r
m
a
nc
e
of
th
e
pr
opos
e
d
m
ode
l
s
te
m
s
f
r
om
th
e
c
om
bi
ne
d
s
tr
e
ngt
hs
of
H
H
O
a
nd
W
O
A
in
f
e
a
tu
r
e
s
e
le
c
ti
on.
H
H
O
e
nha
nc
e
s
e
xpl
or
a
ti
on,
w
hi
le
W
O
A
r
e
f
in
e
s
lo
c
a
l
e
xpl
oi
ta
ti
on,
r
e
s
ul
ti
ng
in
a
n
opt
i
m
iz
e
d
f
e
a
tu
r
e
s
ubs
e
t.
T
hi
s
im
pr
ove
s
th
e
DT
c
la
s
s
if
ie
r
’
s
a
c
c
ur
a
c
y,
r
e
duc
in
g
ir
r
e
le
va
nt
f
e
a
tu
r
e
s
a
nd
e
nha
nc
in
g
a
tt
a
c
k
de
te
c
ti
on.
T
h
e
a
c
hi
e
ve
d
98.46
%
a
c
c
ur
a
c
y
c
onf
ir
m
s
th
e
e
f
f
e
c
ti
ve
ne
s
s
of
th
is
9
7
.
5
9
%
9
7
.
5
0
%
9
8
.
4
6
%
9
7
.
0
0
%
9
7
.
2
0
%
9
7
.
4
0
%
9
7
.
6
0
%
9
7
.
8
0
%
9
8
.
0
0
%
9
8
.
2
0
%
9
8
.
4
0
%
9
8
.
6
0
%
P
re
c
i
s
i
o
n
(
%
)
M
e
t
h
o
d
P
r
e
c
i
s
i
o
n
H
H
O
W
O
A
H
H
O
&W
O
A
9
7
.
5
9
%
9
7
.
5
0
%
9
8
.
4
6
%
9
7
.
0
0
%
9
7
.
2
0
%
9
7
.
4
0
%
9
7
.
6
0
%
9
7
.
8
0
%
9
8
.
0
0
%
9
8
.
2
0
%
9
8
.
4
0
%
9
8
.
6
0
%
F1
-
Sc
o
re
(
%
)
M
e
t
h
o
d
F1
-
S
c
o
r
e
H
H
O
W
O
A
H
H
O
&W
O
A
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
14
, N
o.
4
,
A
ugus
t
20
25
:
2839
-
2848
2846
a
ppr
oa
c
h
ove
r
us
in
g
H
H
O
or
W
O
A
s
e
pa
r
a
te
ly
.
A
ddi
ti
ona
ll
y,
t
he
m
e
th
od
r
e
duc
e
s
c
om
put
a
ti
ona
l
c
om
pl
e
xi
ty
,
e
na
bl
in
g
f
a
s
te
r
pr
oc
e
s
s
in
g
w
hi
le
m
a
in
ta
in
in
g
hi
gh
de
t
e
c
ti
on
a
c
c
ur
a
c
y.
T
he
s
e
r
e
s
ul
ts
de
m
on
s
tr
a
te
th
e
pr
a
c
ti
c
a
l
be
ne
f
it
s
of
th
e
pr
opos
e
d a
ppr
oa
c
h i
n i
m
pr
ovi
ng f
ir
e
w
a
ll
e
f
f
ic
ie
nc
y a
ga
in
s
t
e
vol
vi
ng c
ybe
r
t
hr
e
a
t
s
.
4.
C
O
N
C
L
U
S
I
O
N
A
f
ir
e
w
a
ll
is
on
e
of
th
e
ke
y
c
om
pone
nt
s
th
a
t
pr
ot
e
c
t
s
th
e
in
t
e
r
na
l
ne
twor
k
f
r
om
i
nt
e
r
ne
t
a
tt
a
c
k
s
.
T
r
a
di
ti
ona
l
f
ir
e
w
a
ll
s
do
not
c
ope
w
it
h
r
e
c
e
nt
a
tt
a
c
ks
th
a
t
us
e
s
ophi
s
ti
c
a
te
d
te
c
hni
que
s
.
I
n
th
is
pa
pe
r
,
w
e
ha
ve
pr
opos
e
d
a
f
ir
e
w
a
ll
th
a
t
us
e
s
M
L
m
e
th
ods
to
s
to
p
th
e
s
e
s
ophi
s
ti
c
a
te
d
a
tt
a
c
k
te
c
hni
que
s
.
T
he
pr
opos
e
d
f
ir
e
w
a
ll
e
m
pl
oys
th
e
H
H
O
a
nd
W
O
A
a
lg
or
it
hm
s
f
or
f
e
a
tu
r
e
s
e
le
c
ti
on.
T
he
m
a
in
pur
pos
e
of
H
H
O
a
nd
W
O
A
is
t
o s
e
le
c
t
onl
y t
he
ke
y f
e
a
tu
r
e
s
of
t
he
t
r
a
f
f
ic
t
ha
t
c
a
n i
de
nt
if
y
t
he
a
tt
a
c
ks
. T
he
H
H
O
ha
s
s
e
le
c
te
d 13 f
e
a
tu
r
e
s
,
w
hi
le
th
e
W
O
A
ha
s
s
e
le
c
te
d
16
f
e
a
tu
r
e
s
f
r
om
40
f
e
a
tu
r
e
s
.
T
he
c
om
m
on
f
e
a
tu
r
e
s
be
twe
e
n
th
e
two
a
lg
or
it
hm
s
a
r
e
25. C
om
bi
ni
ng t
he
f
e
a
tu
r
e
s
f
r
om
t
he
t
w
o
a
lg
or
it
hm
s
ha
s
e
nha
nc
e
d t
he
f
ir
e
w
a
ll
pe
r
f
or
m
a
nc
e
. F
or
e
xa
m
pl
e
,
th
e
a
c
c
ur
a
c
y
a
c
hi
e
ve
d
w
he
n
u
s
in
g
H
H
O
is
97.59%
,
a
nd
W
O
A
is
97.5%
,
w
hi
le
w
he
n
us
in
g
th
e
c
om
m
on
f
e
a
tu
r
e
s
of
th
e
two
a
lg
or
it
hm
s
,
th
e
a
c
c
ur
a
c
y
r
e
a
c
h
e
d
98.46%
.
T
he
a
r
c
hi
v
e
d
r
e
s
ul
t
pr
ove
d
th
a
t
th
e
pr
opos
e
d
ML
-
ba
s
e
d f
ir
e
w
a
ll
i
s
a
pr
om
is
in
g s
ol
ut
io
n t
o m
it
ig
a
te
t
he
a
tt
a
c
k
s
on t
he
i
nt
e
r
na
l
ne
twor
k
.
F
U
N
D
I
N
G
I
N
F
O
R
M
A
T
I
O
N
A
ut
hor
s
s
ta
te
no f
undi
ng i
nvol
ve
d.
A
U
T
H
O
R
C
O
N
T
R
I
B
U
T
I
O
N
S
S
T
A
T
E
M
E
N
T
T
hi
s
jo
ur
na
l
us
e
s
th
e
C
ont
r
ib
ut
or
R
ol
e
s
T
a
xonomy
(
C
R
e
di
T
)
to
r
e
c
ogni
z
e
in
di
vi
dua
l
a
ut
hor
c
ont
r
ib
ut
io
ns
, r
e
duc
e
a
ut
hor
s
hi
p di
s
put
e
s
,
a
nd f
a
c
il
it
a
te
c
ol
la
bo
r
a
ti
on.
N
am
e
o
f
A
u
t
h
or
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
M
os
le
h M
.
A
bua
lh
a
j
✓
✓
✓
✓
✓
✓
A
hm
a
d A
de
l
A
bu
-
S
ha
r
e
ha
✓
✓
✓
✓
S
um
a
ya
N
a
bi
l
A
l
-
K
ha
ti
b
✓
✓
✓
✓
✓
✓
✓
A
de
e
b M
. A
ls
a
a
id
a
h
✓
✓
✓
✓
✓
✓
M
oha
m
m
e
d A
nba
r
✓
✓
✓
✓
✓
C
:
C
onc
e
pt
ua
l
i
z
a
t
i
on
M
:
M
e
t
hodol
ogy
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
da
t
i
on
Fo
:
Fo
r
m
a
l
a
na
l
ys
i
s
I
:
I
nve
s
t
i
ga
t
i
on
R
:
R
e
s
our
c
e
s
D
:
D
a
t
a
C
ur
a
t
i
on
O
:
W
r
i
t
i
ng
-
O
r
i
gi
na
l
D
r
a
f
t
E
:
W
r
i
t
i
ng
-
R
e
vi
e
w
&
E
di
t
i
ng
Vi
:
Vi
s
ua
l
i
z
a
t
i
on
Su
:
Su
pe
r
vi
s
i
on
P
:
P
r
oj
e
c
t
a
dm
i
ni
s
t
r
a
t
i
on
Fu
:
Fu
ndi
ng a
c
qui
s
i
t
i
on
C
O
N
F
L
I
C
T
O
F
I
N
T
E
R
E
S
T
S
T
A
T
E
M
E
N
T
A
ut
hor
s
s
ta
te
no c
onf
li
c
t
of
i
nt
e
r
e
s
t.
D
A
T
A
A
V
A
I
L
A
B
I
L
I
T
Y
T
he
da
ta
th
a
t
s
uppor
t
th
e
f
in
di
ngs
of
th
is
s
tu
dy
a
r
e
ope
nl
y
a
va
il
a
bl
e
in
[
C
a
na
di
a
n
I
ns
ti
tu
te
f
or
C
ybe
r
s
e
c
ur
it
y]
a
t
ht
tp
s
:/
/ww
w
.unb.c
a
/c
ic
/d
a
ta
s
e
t
s
/i
ds
.ht
m
l
[
doi
:
10.1016/j
.c
os
e
.2011.12.012]
, r
e
f
e
r
e
nc
e
[
20]
.
R
E
F
E
R
E
N
C
E
S
[
1]
M
.
C
ui
,
J
.
W
a
ng,
a
nd
B
.
C
he
n,
“
F
l
e
xi
bl
e
m
a
c
hi
ne
l
e
a
r
ni
ng
-
ba
s
e
d
c
yb
e
r
a
t
t
a
c
k
de
t
e
c
t
i
on
us
i
ng
s
pa
t
i
ot
e
m
por
a
l
pa
t
t
e
r
ns
f
or
di
s
t
r
i
but
i
on s
ys
t
e
m
s
,”
I
E
E
E
T
r
ans
ac
t
i
ons
on Sm
ar
t
G
r
i
d
, vol
. 11, no. 2, pp. 1805
–
1808, 2020, doi
:
10.1109/
T
S
G
.2020.2965797.
[
2]
M
.
M
.
A
bua
l
ha
j
,
A
.
A
.
A
bu
-
S
ha
r
e
ha
,
Q
.
Y
.
S
ha
m
bour
,
A
.
A
l
s
a
a
i
da
h,
S
.
N
.
A
l
-
K
ha
t
i
b,
a
nd
M
.
A
nba
r
,
“
C
us
t
om
i
z
e
d
K
-
ne
a
r
e
s
t
ne
i
ghbor
s
’
a
l
gor
i
t
hm
f
or
m
a
l
w
a
r
e
de
t
e
c
t
i
on,”
I
nt
e
r
nat
i
onal
J
our
nal
of
D
at
a
and
N
e
t
w
or
k
Sc
i
e
nc
e
,
vol
.
8,
no.
1,
pp.
431
–
438,
2024, doi
:
10.5267/
j
.i
j
dns
.2023.9.012.
[
3]
A
.
O
.
A
l
uko,
R
.
M
us
um
puka
,
a
nd
D
.
G
.
D
or
r
e
l
l
,
“
C
ybe
r
a
t
t
a
c
k
-
r
e
s
i
l
i
e
nt
s
e
c
onda
r
y
f
r
e
que
nc
y
c
ont
r
ol
s
c
he
m
e
f
or
s
t
a
nd
-
a
l
one
m
i
c
r
ogr
i
ds
,”
I
E
E
E
T
r
ans
ac
t
i
ons
on
I
ndus
t
r
i
al
E
l
e
c
t
r
oni
c
s
,
vol
.
70,
no.
2,
pp.
1622
–
1634,
F
e
b.
2023,
doi
:
10.1109/
T
I
E
.2022.3159965.
[
4]
D
.
-
J
.
L
i
u
,
G
.
-
G
.
G
e
ng
,
X
.
-
B
.
J
i
n,
a
nd
W
.
W
a
ng
,
“
A
n
e
f
f
i
c
i
e
nt
m
ul
t
i
s
t
a
ge
phi
s
hi
ng
w
e
bs
i
t
e
de
t
e
c
t
i
on
m
ode
l
ba
s
e
d
on
t
he
C
A
S
E
f
e
a
t
ur
e
f
r
a
m
e
w
or
k:
A
i
m
i
ng
a
t
t
he
r
e
a
l
w
e
b
e
nvi
r
on
m
e
nt
,
”
C
om
put
e
r
s
&
Se
c
ur
i
t
y
.
vol
.
110,
pp.
1
-
11,
N
ov.
2021,
do
i
:
10.1016/
j
.c
os
e
.2021.102421
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
I
m
pr
ov
in
g f
ir
e
w
al
l
pe
r
fo
r
m
anc
e
us
in
g hy
b
r
id
of
opt
imi
z
at
io
n al
gor
it
hm
s
and
…
(
M
os
le
h M
. A
bual
haj
)
2847
[
5]
D
.
H
.
K
a
s
s
,
“
F
B
I
:
c
ovi
d
-
19
c
ybe
r
a
t
t
a
c
ks
s
pi
ke
400%
i
n
pa
nde
m
i
c
,
”
M
SSP
A
l
e
r
t
.
A
c
c
e
s
s
e
d:
S
e
p.
14,
2024.
[
O
nl
i
ne
]
.
A
va
i
l
a
bl
e
:
ht
t
ps
:
/
/
w
w
w
.m
s
s
pa
l
e
r
t
.c
om
/
c
ybe
r
s
e
c
ur
i
t
y
-
ne
w
s
/
f
bi
-
c
ovi
d
-
19
-
c
ybe
r
a
t
t
a
c
ks
-
s
pi
ke
-
400
-
in
-
pa
nde
m
i
c
/
[
6]
H
. G
u
e
t
al
.
, “
D
I
A
V
A
:
a
t
r
a
f
f
i
c
-
ba
s
e
d f
r
a
m
e
w
or
k f
or
de
t
e
c
t
i
on of
S
Q
L
i
nj
e
c
t
i
o
n a
t
t
a
c
ks
a
nd vul
n
e
r
a
bi
l
i
t
y a
na
l
ys
i
s
of
l
e
a
k
e
d da
t
a
,
”
I
E
E
E
T
r
ans
ac
t
i
ons
on R
e
l
i
abi
l
i
t
y
, vol
. 69, no. 1, pp. 188
–
202, M
a
r
. 2020, doi
:
10.1109/
T
R
.2019.2925415.
[
7]
H
. J
m
a
l
, F
. B
e
n
H
m
i
da
, N
.
B
a
s
t
a
,
M
. I
kr
a
m
, M
.
A
. K
a
a
f
a
r
, a
nd A
.
W
a
l
ke
r
,
“
S
P
G
N
N
-
A
P
I
:
a
t
r
a
ns
f
e
r
a
bl
e
gr
a
ph ne
ur
a
l
ne
t
w
or
k f
or
a
t
t
a
c
k
pa
t
hs
i
de
nt
i
f
i
c
a
t
i
on
a
nd
a
ut
onom
ous
m
i
t
i
ga
t
i
on,”
I
E
E
E
T
r
ans
ac
t
i
ons
on
I
nf
or
m
at
i
on
F
or
e
ns
i
c
s
and
Se
c
ur
i
t
y
,
vol
.
19,
pp. 1601
–
1613, 2024, doi
:
10.1109/
T
I
F
S
.2023.3338965.
[
8]
F
.
C
he
n,
B
.
B
r
uha
de
s
hw
a
r
,
a
nd
A
.
X
.
L
i
u,
“
C
r
os
s
-
dom
a
i
n
pr
i
va
c
y
-
pr
e
s
e
r
vi
ng
c
oope
r
a
t
i
ve
f
i
r
e
w
a
l
l
opt
i
m
i
z
a
t
i
on,”
I
E
E
E
/
A
C
M
T
r
ans
ac
t
i
ons
on N
e
t
w
or
k
i
ng
, vol
. 21, no. 3, pp. 857
–
868, J
un. 2013, doi
:
10.1109/
T
N
E
T
.2012.2217985.
[
9]
P
.
Z
hou,
H
.
Z
ha
ng,
a
nd
W
.
L
i
a
ng,
“
R
e
s
e
a
r
c
h
on
hybr
i
d
i
nt
r
us
i
on
de
t
e
c
t
i
on
ba
s
e
d
on
i
m
pr
ove
d
ha
r
r
i
s
ha
w
k
opt
i
m
i
z
a
t
i
o
n
a
l
gor
i
t
hm
,”
C
onne
c
t
i
on Sc
i
e
nc
e
, vol
. 35, no. 1, D
e
c
. 2023, doi
:
10.1080/
095400
91.2023.2195595.
[
10]
G
.
Y
e
dukonda
l
u,
G
.
H
.
B
i
ndu,
J
.
P
a
va
n,
G
.
V
e
nka
t
e
s
h,
a
nd
A
.
S
a
i
T
e
j
a
,
“
I
nt
r
us
i
on
de
t
e
c
t
i
on
s
ys
t
e
m
f
r
a
m
e
w
or
k
us
i
ng
m
a
c
hi
ne
l
e
a
r
ni
ng,”
i
n
2021
T
hi
r
d
I
nt
e
r
nat
i
onal
C
onf
e
r
e
nc
e
on
I
nv
e
nt
i
v
e
R
e
s
e
ar
c
h
i
n
C
om
put
i
ng
A
ppl
i
c
at
i
ons
(
I
C
I
R
C
A
)
,
S
e
p.
2021,
pp. 1224
–
1230
,
doi
:
10.1109/
I
C
I
R
C
A
51532.2021.9544717.
[
11]
J
.
A
.
A
br
a
ha
m
a
nd
V
.
R
.
B
i
ndu,
“
I
nt
r
us
i
on
de
t
e
c
t
i
on
a
nd
pr
e
ve
nt
i
on
i
n
ne
t
w
or
ks
us
i
ng
m
a
c
hi
ne
l
e
a
r
ni
ng
a
nd
d
e
e
p
l
e
a
r
ni
ng
a
ppr
oa
c
he
s
:
a
r
e
vi
e
w
,
”
i
n
2021 I
nt
e
r
nat
i
onal
C
onf
e
r
e
n
c
e
on A
dv
anc
e
m
e
nt
s
i
n
E
l
e
c
t
r
i
c
al
, E
l
e
c
t
r
oni
c
s
,
C
om
m
uni
c
at
i
on, C
om
put
i
n
g
and A
ut
om
at
i
on (
I
C
A
E
C
A
)
, O
c
t
. 2021, pp. 1
–
4
,
doi
:
10.1109/
I
C
A
E
C
A
52838.2021.9675595.
[
12]
A
.
G
upt
a
e
t
al
.
,
“
O
n
t
he
ut
i
l
i
t
y
of
pow
e
r
s
pe
c
t
r
a
l
t
e
c
hni
que
s
w
i
t
h
f
e
a
t
ur
e
s
e
l
e
c
t
i
on
t
e
c
hni
que
s
f
or
e
f
f
e
c
t
i
ve
m
e
nt
a
l
t
a
s
k
c
l
a
s
s
i
f
i
c
a
t
i
on i
n noni
nva
s
i
ve
B
C
I
,”
I
E
E
E
T
r
ans
ac
t
i
ons
on S
y
s
t
e
m
s
, M
an, and C
y
be
r
ne
t
i
c
s
:
Sy
s
t
e
m
s
, vol
. 51, no. 5, pp. 3080
–
3092
,
M
a
y 2021, doi
:
10.1109/
T
S
M
C
.2019.2917599.
[
13]
D
.
P
a
l
a
c
i
os
,
I
.
D
e
-
la
-
B
a
nde
r
a
,
A
.
G
om
e
z
-
A
ndr
a
de
s
,
L
.
F
l
or
e
s
,
a
nd
R
.
B
a
r
c
o,
“
A
ut
om
a
t
i
c
f
e
a
t
ur
e
s
e
l
e
c
t
i
on
t
e
c
hni
que
f
or
ne
xt
ge
ne
r
a
t
i
on
s
e
l
f
-
or
ga
ni
z
i
ng
ne
t
w
or
ks
,”
I
E
E
E
C
om
m
uni
c
at
i
ons
L
e
t
t
e
r
s
,
vol
.
22,
no.
6,
pp.
1272
–
1275,
J
un.
2018,
doi
:
10.1109/
L
C
O
M
M
.2018.2825392.
[
14]
M
.
B
a
ne
r
j
e
e
a
nd
N
.
R
.
P
a
l
,
“
U
ns
upe
r
vi
s
e
d
f
e
a
t
ur
e
s
e
l
e
c
t
i
on
w
i
t
h
c
ont
r
ol
l
e
d
r
e
dunda
nc
y
(
U
F
e
S
C
oR
)
,”
I
E
E
E
T
r
ans
ac
t
i
ons
on
K
now
l
e
dge
and D
at
a E
ngi
ne
e
r
i
ng
, vol
. 27, no. 12, pp. 3390
–
3403, D
e
c
. 2015,
doi
:
10.1109/
T
K
D
E
.2015.2455509.
[
15]
M
. M
. S
a
kr
, M
. A
.
T
a
w
f
e
e
q, a
nd A
. B
. E
l
-
S
i
s
i
, “
F
i
l
t
e
r
ve
r
s
us
w
r
a
ppe
r
f
e
a
t
ur
e
s
e
l
e
c
t
i
on f
or
ne
t
w
or
k i
nt
r
us
i
on de
t
e
c
t
i
on s
ys
t
e
m
,”
i
n
2019
N
i
nt
h
I
nt
e
r
nat
i
onal
C
onf
e
r
e
n
c
e
on
I
nt
e
l
l
i
ge
nt
C
om
put
i
ng
and
I
nf
or
m
at
i
o
n
Sy
s
t
e
m
s
(
I
C
I
C
I
S)
,
D
e
c
.
2019,
pp.
209
–
214
,
doi
:
10.1109/
I
C
I
C
I
S
46948.2019.9014797.
[
16]
S
.
M
.
K
a
s
ongo
a
nd
Y
.
S
un,
“
P
e
r
f
or
m
a
nc
e
a
na
l
ys
i
s
of
i
nt
r
us
i
on
de
t
e
c
t
i
on
s
ys
t
e
m
s
us
i
ng
a
f
e
a
t
ur
e
s
e
l
e
c
t
i
on
m
e
t
hod
on
t
he
U
N
S
W
-
N
B
15 da
t
a
s
e
t
,
”
J
our
nal
of
B
i
g D
at
a
, vol
. 7, no. 1, D
e
c
. 2020, doi
:
10.1186/
s
40
537
-
020
-
00379
-
6.
[
17]
T
.
W
i
s
a
nw
a
ni
c
ht
ha
n
a
nd
M
.
T
ha
m
m
a
w
i
c
ha
i
,
“
A
doubl
e
-
l
a
ye
r
e
d
hybr
i
d
a
ppr
oa
c
h
f
or
ne
t
w
or
k
i
nt
r
us
i
on
de
t
e
c
t
i
on
s
ys
t
e
m
us
i
ng
c
om
bi
ne
d na
i
ve
B
a
ye
s
a
nd S
V
M
,”
I
E
E
E
A
c
c
e
s
s
, vol
. 9, pp. 138432
–
138450, 2
021, doi
:
10.1109/
A
C
C
E
S
S
.2021.3118573.
[
18]
A
.
M
ugha
i
d,
S
.
A
l
Z
u’
bi
,
A
.
H
na
i
f
,
S
.
T
a
a
m
ne
h,
A
.
A
l
na
j
j
a
r
,
a
nd
E
.
A
.
E
l
s
oud
,
“
A
n
i
nt
e
l
l
i
ge
nt
c
ybe
r
s
e
c
ur
i
t
y
phi
s
hi
ng
de
t
e
c
t
i
o
n
s
ys
t
e
m
us
i
ng
de
e
p
l
e
a
r
ni
ng
t
e
c
hni
que
s
,”
C
l
us
t
e
r
C
om
put
i
ng
,
vol
.
25,
no.
6,
pp.
3819
–
3828,
2022,
doi
:
10.1007/
s
10586
-
022
-
03604
-
4.
[
19]
L
.
L
i
u,
P
.
W
a
ng,
J
.
L
i
n,
a
nd
L
.
L
i
u,
“
I
nt
r
us
i
on
de
t
e
c
t
i
on
of
i
m
ba
l
a
nc
e
d
ne
t
w
or
k
t
r
a
f
f
i
c
ba
s
e
d
on
m
a
c
hi
ne
l
e
a
r
ni
ng
a
nd
de
e
p
l
e
a
r
ni
ng,”
I
E
E
E
A
c
c
e
s
s
, vol
. 9, pp. 7550
–
7563, 2021, doi
:
10.1109/
A
C
C
E
S
S
.2020.3048198.
[
20]
A
.
S
hi
r
a
va
ni
,
M
.
H
.
S
a
dr
e
ddi
ni
,
a
nd
H
.
N
.
N
a
hook,
“
N
e
t
w
or
k
i
nt
r
us
i
on
de
t
e
c
t
i
on
us
i
ng
da
t
a
di
m
e
ns
i
ons
r
e
duc
t
i
on
t
e
c
hni
que
s
,
”
J
our
nal
of
B
i
g D
at
a
, vol
. 10, no. 1, M
a
r
. 2023,
doi
:
10.1186/
s
40537
-
023
-
00697
-
5.
[
21]
M
.
M
.
A
bua
l
ha
j
,
S
.
N
.
A
l
-
K
ha
t
i
b,
A
.
A
l
-
A
l
l
a
w
e
e
,
A
.
M
unt
he
r
,
a
nd
M
.
A
nba
r
,
“
E
nha
nc
i
ng
ne
t
w
or
k
i
nt
r
us
i
on
de
t
e
c
t
i
on
s
ys
t
e
m
s
t
hr
ough
di
m
e
ns
i
ona
l
i
t
y
r
e
duc
t
i
on,”
R
e
c
e
nt
A
dv
anc
e
s
on
Sof
t
C
om
put
i
ng
and
D
at
a
M
i
ni
ng
,
pp.
244
–
253,
2024,
doi
:
10.1007/
978
-
3
-
031
-
66965
-
1_24.
[
22]
M
.
M
.
A
bua
l
ha
j
,
A
.
A
.
A
bu
-
S
ha
r
e
ha
,
A
.
A
l
-
A
l
l
a
w
e
e
,
A
.
M
unt
he
r
,
a
nd
M
.
A
nba
r
,
“
P
e
r
f
or
m
a
nc
e
e
va
l
ua
t
i
on
of
w
ha
l
e
a
nd
ha
r
r
i
s
ha
w
ks
opt
i
m
i
z
a
t
i
on
a
l
gor
i
t
hm
s
w
i
t
h
i
nt
r
us
i
on
pr
e
ve
nt
i
on
s
y
s
t
e
m
s
,
”
R
e
c
e
nt
A
dv
anc
e
s
on
Sof
t
C
om
put
i
ng
and
D
at
a
M
i
ni
ng
,
pp. 254
–
265, 2024, doi
:
10.1007/
978
-
3
-
031
-
66965
-
1_25.
[
23]
M
.
M
.
A
bua
l
ha
j
,
A
.
S
.
A
l
-
S
ha
m
a
yl
e
h,
A
.
M
unt
he
r
,
S
.
N
.
A
l
kha
t
i
b,
M
.
O
.
H
i
a
r
i
,
a
nd
M
.
A
nba
r
,
“
E
nha
nc
i
ng
s
pyw
a
r
e
de
t
e
c
t
i
on
b
y
ut
i
l
i
z
i
ng
de
c
i
s
i
on
t
r
e
e
s
w
i
t
h
hype
r
pa
r
a
m
e
t
e
r
opt
i
m
i
z
a
t
i
on,”
B
ul
l
e
t
i
n
of
E
l
e
c
t
r
i
c
al
E
ngi
ne
e
r
i
ng
and
I
nf
or
m
at
i
c
s
,
vol
.
13,
no.
5,
pp. 3653
–
3662, 2024, doi
:
10.11591/
e
e
i
.v13i
5.7939.
[
24]
X.
-
Y
.
S
hi
h,
Y
.
C
hi
u,
a
nd
H
.
-
E
.
W
u,
“
D
e
s
i
gn
a
nd
i
m
pl
e
m
e
nt
a
t
i
on
of
de
c
i
s
i
on
-
t
r
e
e
(
D
T
)
onl
i
ne
t
r
a
i
ni
ng
ha
r
dw
a
r
e
us
i
ng
di
vi
de
r
-
f
r
e
e
G
I
c
a
l
c
ul
a
t
i
on
a
nd
s
pe
e
di
ng
-
up
doubl
e
-
r
oot
c
l
a
s
s
i
f
i
e
r
,”
I
E
E
E
T
r
ans
ac
t
i
ons
on
C
i
r
c
ui
t
s
and
Sy
s
t
e
m
s
I
:
R
e
gul
ar
P
ape
r
s
,
vol
. 70, no. 2, pp. 759
–
771, F
e
b. 2023, doi
:
10.1109/
T
C
S
I
.2022.3222515.
[
25]
M
.
M
.
A
bua
l
ha
j
,
A
.
A
.
A
bu
-
S
ha
r
e
ha
,
M
.
O
.
H
i
a
r
i
,
Y
.
A
l
r
a
ba
na
h,
M
.
A
l
-
Z
youd,
a
nd
M
.
A
.
A
l
s
ha
r
a
i
a
h,
“
A
pa
r
a
di
gm
f
or
D
oS
a
t
t
a
c
k
di
s
c
l
o
s
ur
e
us
i
ng
m
a
c
hi
ne
l
e
a
r
ni
ng
t
e
c
hni
que
s
,”
I
nt
e
r
nat
i
onal
J
our
nal
of
A
dv
anc
e
d
C
om
put
e
r
S
c
i
e
nc
e
and
A
ppl
i
c
at
i
ons
,
vol
. 13, no. 3, 2022, doi
:
10.14569/
I
J
A
C
S
A
.2022.0130325.
B
I
O
G
R
A
P
H
I
E
S
O
F
A
U
T
H
O
R
S
Mosleh
M.
Abualhaj
is
a
senior
lecturer
at
Al
-
Ahliyya
Amman
University.
He
received
his
first
degree
in
Computer
Science
from
Philadelp
hia
Uni
versity,
Jordan,
in
2004,
master
degree
in
Computer
Information
System
from
the
Arab
Ac
ademy
for
Banking
an
d
Financi
al
Scienc
es,
Jordan
in
2007,
and
Ph.D.
in
Multimed
ia
N
etwor
ks
Protoco
ls
from
Universiti
Sains
Malaysia
in
2011.
His
research
area
of
interest
inc
ludes
VoIP,
c
ongestion
c
ontrol,
and
c
ybersecurity
data
mining
and
o
ptimization.
He
can
be
contacted
at
email:
m.abualhaj@ammanu.edu.jo
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
14
, N
o.
4
,
A
ugus
t
20
25
:
2839
-
2848
2848
Ahmad
Adel
Abu
-
Shareha
received
his
first
degree
in
Computer
Science
fro
m
Al
Al
-
Bayt
University,
Jordan,
2004,
m
aster
degree
from
Universiti
Sains
Malaysia
(USM),
Malaysia,
in
2006,
and
Ph.D
.
degree
from
USM,
Malaysia,
in
2012.
His
research
focuses
on
data
mining,
artificial
intelligent
,
and
multimedia
security
.
He
inve
stigated
many
machine
learning
algorithms
and
employed
artificia
l
intelligen
ce
in
variety
of
fields,
such
as
network,
medical
information
process,
knowledge
construction
and
extraction
.
He
can
be
contacted
at
email:
a.abushareha@
ammanu.ed
u.jo
.
Sumaya
Nabil
Al
-
Khatib
is
a
senior
lecturer
in
Al
-
Ahliyya
Amm
an
University.
She
received
his
first
degree
in
Computer
Science
from
Baghdad
University,
Iraq,
in
June
1994
and
master
degree
in
Computer
Information
System
from
t
he
Arab
Academy
for
Banking
and
Financia
l
Science
s,
Jordan
in
Februar
y.
Her
resea
rch
area
of
interes
t
includes
VoIP,
multimedia
networking,
and
congestion
control
.
She
can
be
contacted
at
email:
sumayakh@ammanu.edu.jo
.
Adeeb
M.
Alsaaidah
received
the
bachelor’s
degree
in
Compute
r
Engineering
from
the
Faculty
of
Engineering,
A
l
-
Balqa
Applied
University,
the
master’s
degree
in
Networking
and
Computer
Security
from
NYIT
University,
and
the
P
h.D.
degree
in
Computer
Network
from
Universiti
Sains
Islam
Malaysia
,
Malaysia.
He
is
currently
an
Assistant
Profes
sor
in
Networ
k
and
Cybers
ecur
ity
depar
tment
at
Al
-
Ahliyya
Amman
University.
His
research
interests
include
network
performance,
multim
edia
netwo
rks,
network
quality
of
service
(QoS),
the
IoT,
network
modeling
and
simulation,
netwo
rk
security,
and
c
loud
security.
He can be contacted at email:
a.alsaaidah
@
ammanu.ed
u.jo.
Mohammed
Anbar
received
the
B.Sc.
degree
in
Softwa
re
Eng
ineer
ing
from
Al
-
Azhar
University,
Palestine,
in
2008,
the
M.Sc.
degree
in
Infor
mation
Technology
from
Universiti
Utara
Malaysia,
in
2009,
and
the
Ph.D.
degree
in
Advanc
ed
Internet
Security
a
nd
Monitoring
from
Universiti
Sains
Malaysia,
in
2013.
He
is
currently
a
senior
lecturer
with
the
National
Advanced
IPv6
Centre
(NAv6),
Universiti
Sains
Malaysi
a
.
His
current
research
interests
include
malware
detection,
intrusion
detection
systems
(ID
Ss),
intrusion
prevention
systems
(IPSs),
network
monitoring,
the
internet
of
thing
s
(IoT),
soft
ware
-
defined
networking
(SDN)
security,
cloud
computing
security,
and
IPv6
security.
He
can
be
contacted
at
email:
anbar@
usm.my
.
Evaluation Warning : The document was created with Spire.PDF for Python.