I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
pu
t
er
E
ng
ineering
(
I
J
E
CE
)
Vo
l.
15
,
No
.
2
,
A
p
r
il
20
25
,
p
p
.
2
4
9
1
~
2
4
9
8
I
SS
N:
2088
-
8
7
0
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijece.
v
15
i
2
.
pp
2
4
9
1
-
2
4
9
8
2491
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m
Intrusio
n de
tect
io
n bas
ed on g
ener
a
tive a
dv
ersa
ria
l net
wo
rk
with
ra
ndo
m
f
o
re
st f
o
r cloud
ne
two
rks
G
na
na
m
J
eba
R
o
s
lin
e
1
,
P
us
hp
a
Ra
ni
2
1
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
a
n
d
S
o
f
t
w
a
r
e
En
g
i
n
e
e
r
i
n
g
,
M
o
t
h
e
r
Th
e
r
e
sa
W
o
m
e
n
’
s U
n
i
v
e
r
s
i
t
y
,
K
o
d
a
i
k
a
n
a
l
,
I
n
d
i
a
2
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
M
o
t
h
e
r
Th
e
r
e
sa
W
o
me
n
’
s Un
i
v
e
r
si
t
y
,
K
o
d
a
i
k
a
n
a
l
,
I
n
d
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Ma
y
2
8
,
2
0
2
4
R
ev
is
ed
Oct
2
1
,
2
0
2
4
Acc
ep
ted
No
v
2
0
,
2
0
2
4
Th
e
d
e
v
e
lo
p
m
e
n
t
o
f
c
lo
u
d
c
o
m
p
u
ti
n
g
e
n
a
b
les
i
n
d
i
v
id
u
a
ls
a
n
d
o
rg
a
n
iza
ti
o
n
s
to
a
c
c
e
ss
a
wid
e
ra
n
g
e
o
f
o
n
li
n
e
p
r
o
g
ra
m
s
a
n
d
se
rv
ice
s.
Be
c
a
u
se
o
f
it
s
n
a
tu
re
,
n
u
m
e
ro
u
s
u
se
rs
c
a
n
a
c
c
e
ss
a
n
d
d
istri
b
u
te
c
lo
u
d
i
n
fra
str
u
c
tu
re
.
I
n
c
lo
u
d
c
o
m
p
u
ti
n
g
se
v
e
ra
l
se
c
u
rit
y
th
re
a
ts
c
h
a
n
g
e
th
e
d
a
ta
a
n
d
o
p
e
ra
ti
o
n
s.
A
n
e
two
rk
'
s
a
b
il
it
y
t
o
d
e
tec
t
m
a
li
c
io
u
s
a
c
ti
v
it
y
a
n
d
p
o
ss
ib
le
th
re
a
ts
is
g
re
a
tl
y
a
id
e
d
b
y
i
n
tru
si
o
n
d
e
tec
ti
o
n
.
T
o
so
lv
e
th
e
se
iss
u
e
s,
i
n
tru
si
o
n
d
e
tec
ti
o
n
b
a
se
d
o
n
g
e
n
e
ra
ti
v
e
a
d
v
e
rsa
rial
n
e
two
r
k
wit
h
ra
n
d
o
m
fo
re
st
(G
AN
-
RF
)
fo
r
c
lo
u
d
n
e
two
rk
s
is
i
n
tr
o
d
u
c
e
d
.
T
h
e
fu
n
c
ti
o
n
o
f
t
h
e
g
e
n
e
ra
ti
v
e
a
d
v
e
rsa
rial
n
e
two
rk
s
(G
AN
s)
b
a
se
d
n
e
two
rk
a
b
n
o
rm
a
l
it
y
re
c
o
g
n
it
i
o
n
sy
ste
m
is
e
v
a
lu
a
t
e
d
.
It
u
se
s
th
e
CICIDS2
0
1
8
d
a
tas
e
t
to
d
e
tec
t
in
tru
sio
n
.
G
AN
is
u
ti
li
z
e
d
to
imp
ro
v
e
n
e
two
rk
a
n
o
m
a
ly
d
e
tec
ti
o
n
in
c
o
n
ju
n
c
ti
o
n
wit
h
a
n
e
n
se
m
b
le
ra
n
d
o
m
fo
re
s
t
(RF
)
c
las
sifier.
Th
e
G
A
N
-
RF
m
o
d
e
l
a
c
h
iev
e
d
9
5
.
0
1
%
o
f
a
c
c
u
ra
c
y
fo
r
in
tru
si
o
n
d
e
tec
ti
o
n
a
n
d
o
b
tain
b
e
tt
e
r
re
c
a
ll
a
n
d
F
1
-
sc
o
re
.
Ex
ten
si
v
e
a
ss
e
ss
m
e
n
ts
a
n
d
v
a
lu
a
ti
o
n
s il
lu
str
a
te
th
e
e
fficie
n
c
y
o
f
th
e
G
AN
-
RF
a
p
p
r
o
a
c
h
in
a
c
c
u
ra
tely
i
d
e
n
ti
f
y
in
g
n
e
two
rk
issu
e
s.
K
ey
w
o
r
d
s
:
C
I
C
I
DS2
0
1
8
d
ataset
Gen
er
ativ
e
ad
v
e
r
s
ar
ial
n
etwo
r
k
s
I
n
tr
u
s
io
n
d
etec
tio
n
Netwo
r
k
an
o
m
al
y
R
an
d
o
m
f
o
r
est
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Gn
an
am
J
eb
a
R
o
s
lin
e
Dep
ar
tm
en
t o
f
C
o
m
p
u
ter
a
n
d
So
f
twar
e
E
n
g
in
ee
r
in
g
,
M
o
th
er
T
h
er
esa
W
o
m
en
’
s
Un
iv
er
s
ity
Ko
d
aik
an
al,
T
am
il Na
d
u
,
I
n
d
i
a
E
m
ail: je
b
ar
o
s
elin
e8
7
1
@
g
m
ai
l.c
o
m
1.
I
NT
RO
D
UCT
I
O
N
T
h
e
in
ter
n
et
s
er
v
ice
s
ec
to
r
is
a
d
y
n
am
ic
p
a
r
ad
ig
m
f
o
r
lar
g
e
-
s
ca
le
in
f
r
astru
ctu
r
e,
e
n
co
m
p
ass
in
g
f
ield
s
lik
e
clo
u
d
co
m
p
u
tin
g
[
1
]
.
T
h
e
clo
u
d
c
o
m
p
u
tin
g
s
er
v
ice
is
a
less
ex
p
en
s
iv
e.
T
h
e
way
th
ese
clo
u
d
p
r
o
v
id
e
r
s
h
an
d
le
u
s
er
s
'
d
ata
r
aises
s
ec
u
r
ity
an
d
p
r
iv
ac
y
c
o
n
ce
r
n
s
,
n
o
twith
s
tan
d
in
g
th
e
im
p
ac
t
an
d
ef
f
ec
tiv
e
s
er
v
ices
th
ese
ap
p
s
h
a
v
e
p
r
o
v
id
ed
[
2
]
,
[
3
]
.
As
clo
u
d
co
m
p
u
tin
g
s
er
v
i
ce
s
b
ec
o
m
e
m
o
r
e
wid
ely
av
ail
ab
le,
m
an
y
b
an
k
s
,
g
o
v
er
n
m
en
ts
,
an
d
en
ter
p
r
is
es
h
av
e
em
b
r
ac
ed
th
em
.
Stro
n
g
s
ec
u
r
ity
m
ea
s
u
r
es
ar
e
n
ec
e
s
s
ar
y
b
ec
au
s
e
th
is
tr
an
s
f
o
r
m
atio
n
ex
p
o
s
ed
th
ese
s
y
s
tem
s
to
v
a
r
io
u
s
i
n
tr
u
s
io
n
s
f
r
o
m
h
ac
k
er
s
an
d
o
th
e
r
in
tr
u
d
er
s
.
C
u
r
r
e
n
t
in
tr
u
s
io
n
d
etec
tio
n
s
y
s
tem
s
(
I
DS)
u
s
e
an
o
m
aly
o
r
s
ig
n
at
u
r
e
d
etec
tio
n
as
th
eir
o
p
er
atin
g
m
ec
h
an
is
m
[
4
]
.
A
n
ad
ap
tiv
e
s
ec
u
r
ity
m
ec
h
an
is
m
i
s
n
ec
ess
ar
y
f
o
r
a
s
ec
u
r
e
cl
o
u
d
d
ep
lo
y
m
e
n
t to
f
o
s
ter
a
h
ig
h
d
e
g
r
ee
o
f
u
s
er
tr
u
s
t.
T
h
e
co
n
v
en
tio
n
al
m
ac
h
in
e
le
ar
n
in
g
(
ML
)
tech
n
iq
u
es
[
5
]
,
i
n
clu
d
in
g
s
u
p
er
v
is
ed
n
etwo
r
k
I
DS,
h
av
e
d
em
o
n
s
tr
ated
r
esp
ec
tab
le
r
esu
lts
in
id
en
tify
in
g
m
alev
o
len
t
p
ay
lo
ad
s
th
at
h
av
e
b
ee
n
ass
ig
n
ed
a
g
r
o
u
n
d
tr
u
t
h
lab
el.
I
n
o
r
d
er
to
en
lar
g
e
r
ec
o
g
n
itio
n
ac
c
u
r
ac
y
estab
lis
h
ed
o
n
in
tr
u
s
io
n
d
etec
tio
n
ap
p
r
o
ac
h
es,
a
v
ar
iety
o
f
g
en
er
ativ
e
m
o
d
els
h
av
e
b
ee
n
d
ev
elo
p
ed
[
6
]
,
[
7
]
.
T
h
ese
m
o
d
els
ca
n
p
r
o
d
u
ce
r
eliab
le
d
a
ta
s
ets.
I
n
o
r
d
er
t
o
o
v
er
co
m
e
th
e
is
s
u
e
o
f
im
b
ala
n
ce
d
d
atasets
,
g
en
e
r
ativ
e
ad
v
e
r
s
ar
ial
n
etwo
r
k
s
(
GANs)
ca
n
p
r
o
v
id
e
a
v
a
r
iety
o
f
s
y
n
th
etic
d
ata
t
o
s
u
p
p
lem
en
t
t
h
e
s
ca
n
t
am
o
u
n
t
o
f
r
ea
l
-
wo
r
ld
in
tr
u
s
io
n
d
ata
[
8
]
.
B
y
id
e
n
tify
in
g
m
in
u
te
p
atter
n
s
an
d
ab
n
o
r
m
alities
in
n
etwo
r
k
d
ata,
GANs
ca
n
h
elp
th
e
I
D
S
to
id
en
tify
n
ew
ty
p
es
o
f
at
tack
s
.
T
h
e
r
a
n
d
o
m
f
o
r
est
(
R
F)
is
co
m
p
o
s
ed
o
f
s
ev
er
al
d
ec
is
io
n
tr
ee
s
,
ea
c
h
o
f
wh
ich
will
g
r
o
w
to
f
u
ll
m
at
u
r
ity
,
d
o
n
o
t
n
ee
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
2
,
Ap
r
il
20
25
:
2
4
9
1
-
2
4
9
8
2492
p
r
u
n
in
g
,
p
r
o
v
id
e
r
esu
lts
th
at
ar
e
m
o
r
e
ac
cu
r
ate
th
e
m
o
r
e
tr
e
es
it
h
as,
an
d
av
o
id
o
v
er
f
itti
n
g
.
T
h
e
R
F
m
eth
o
d
,
wh
ich
h
as th
e
b
e
n
ef
it o
f
a
u
to
m
ated
f
ea
tu
r
e
s
elec
tio
n
a
m
o
n
g
o
th
er
th
in
g
s
,
will p
er
f
o
r
m
th
e
o
v
er
all
esti
m
ate.
T
h
e
s
tr
u
ctu
r
e
o
f
th
is
ar
ticle
as
f
o
llo
ws.
T
h
e
r
elev
a
n
t
wo
r
k
s
b
ased
o
n
in
tr
u
s
io
n
d
etec
tio
n
m
o
d
els
ar
e
s
u
m
m
ar
ized
in
s
ec
tio
n
2
.
T
h
e
p
r
o
p
o
s
ed
m
o
d
el
th
at
u
s
es
an
R
F
cla
s
s
if
ier
an
d
GAN
is
ex
p
lain
ed
in
s
ec
tio
n
3
.
T
h
e
r
esu
lts
,
ev
alu
atio
n
,
an
d
d
ataset
ar
e
co
v
er
ed
in
s
ec
tio
n
4
.
Fin
ally
,
s
ec
tio
n
5
p
r
esen
ts
co
n
clu
s
io
n
with
f
u
tu
r
e
wo
r
k
.
2.
RE
L
AT
E
D
WO
RK
T
h
e
u
s
er
m
ay
b
en
ef
it
f
r
o
m
a
m
u
ltit
u
d
e
o
f
s
er
v
ices
o
f
f
er
ed
b
y
clo
u
d
co
m
p
u
tin
g
,
in
clu
d
in
g
in
f
r
astru
ctu
r
e,
s
to
r
ag
e
ca
p
ac
it
y
an
d
a
p
p
licatio
n
s
.
A
clo
u
d
u
s
er
m
o
s
tly
u
s
es
th
e
in
ter
n
et
to
ac
ce
s
s
an
d
m
o
d
if
y
h
ar
d
war
e
an
d
s
o
f
twar
e
to
s
u
it
th
eir
r
eq
u
ir
em
en
ts
.
Alth
o
u
g
h
th
er
e
ar
e
n
u
m
er
o
u
s
ad
v
an
ta
g
es
to
u
s
in
g
clo
u
d
co
m
p
u
tin
g
,
th
er
e
ar
e
also
d
r
a
wb
ac
k
s
an
d
d
if
f
ic
u
lties
.
C
lo
u
d
co
m
p
u
tin
g
p
r
esen
ts
a
n
u
m
b
e
r
o
f
is
s
u
es,
in
clu
d
in
g
lo
ad
b
alan
cin
g
,
p
r
iv
ac
y
,
s
ec
u
r
ity
,
an
d
p
er
f
o
r
m
a
n
ce
m
a
n
ag
e
m
en
t.
T
h
e
m
o
s
t
s
ig
n
if
ica
n
t
is
s
u
e
am
o
n
g
th
em
is
s
ec
u
r
ity
s
in
ce
u
s
er
d
ata
an
d
ap
p
s
ar
e
lo
ca
ted
o
n
clo
u
d
in
f
r
a
s
tr
u
ctu
r
e.
Ad
d
itio
n
ally
,
it
g
u
a
r
d
s
ag
ain
s
t
s
o
f
twar
e
q
u
er
y
lan
g
u
ag
e
i
n
jectio
n
,
c
r
o
s
s
-
s
ite
s
cr
ip
tin
g
,
d
ata
m
a
n
i
p
u
latio
n
,
s
o
f
twar
e
v
u
ln
er
a
b
ilit
ies,
an
d
f
lo
o
d
in
g
attac
k
s
.
An
ex
ten
s
iv
e
r
an
g
e
o
f
d
ee
p
le
ar
n
in
g
(
DL
)
an
d
m
ac
h
in
e
lear
n
in
g
(
ML
)
alg
o
r
it
h
m
s
h
av
e
r
e
ce
n
tly
b
ee
n
im
p
lem
en
ted
in
to
in
tr
u
s
io
n
d
etec
tio
n
s
y
s
tem
s
as
a
r
esu
l
t
o
f
th
e
q
u
ick
g
r
o
wth
o
f
ar
tific
ial
in
tellig
en
ce
tech
n
o
lo
g
y
.
A
h
y
b
r
id
d
ee
p
l
ea
r
n
in
g
d
etec
tio
n
ap
p
r
o
ac
h
with
an
ar
ea
u
n
d
er
th
e
cu
r
v
e
(
AUC)
o
f
0
.
9
7
is
d
em
o
n
s
tr
ated
in
s
tu
d
y
[
9
]
.
A
n
en
h
an
ce
d
r
estricte
d
B
o
ltzm
an
n
m
ac
h
in
e
was
u
s
ed
to
ex
t
r
ac
t
an
d
r
ed
u
ce
t
h
e
d
ata
ch
ar
ac
ter
is
tics
b
ef
o
r
e
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
es
(
SV
M)
[
1
0
]
,
[
1
1
]
wer
e
u
s
ed
f
o
r
class
if
icatio
n
.
An
au
to
en
co
d
er
-
b
ased
f
r
a
m
ewo
r
k
f
o
r
n
etwo
r
k
i
n
tr
u
s
io
n
d
etec
ti
o
n
s
y
s
tem
(
NI
DS)
is
ex
p
lain
e
d
in
s
tu
d
y
[
1
2
]
.
Fo
r
im
p
r
o
v
e
d
class
if
icatio
n
,
th
e
f
r
am
ewo
r
k
c
o
m
b
in
e
d
th
e
au
t
o
en
co
d
e
r
an
d
u
n
s
u
p
er
v
is
ed
clu
s
ter
in
g
m
o
d
u
le'
s
co
o
p
er
ativ
e
tr
ain
i
n
g
o
f
th
e
r
ec
o
n
s
tr
u
ctio
n
lo
s
s
an
d
class
if
ica
tio
n
lo
s
s
.
I
n
o
r
d
er
to
b
o
o
s
t
th
e
ef
f
ec
tiv
en
ess
an
d
g
en
er
aliza
tio
n
o
f
class
if
ier
s
R
am
ap
r
ab
a
et
a
l.
[
1
3
]
p
r
o
p
o
s
ed
a
GAN
-
b
ased
I
DS.
T
h
e
s
t
r
ateg
y
cr
ea
ted
f
alse
lab
el
s
am
p
les
co
n
tin
u
ally
u
s
i
n
g
a
g
en
e
r
ativ
e
ap
p
r
o
ac
h
to
p
r
o
v
id
e
t
h
e
class
if
ier
s
en
h
a
n
ce
th
eir
d
etec
tio
n
ab
ilit
y
,
an
d
also
em
p
lo
y
ed
ad
v
er
s
ar
ial
tr
ain
in
g
to
e
n
h
an
ce
t
h
e
class
if
ier
s
[
1
4
]
.
A
d
is
tr
ib
u
ted
GAN
-
b
ased
I
DS th
at
ca
n
id
en
tify
in
ter
n
et
o
f
th
in
g
s
(
I
o
T
)
i
n
tr
u
s
io
n
with
litt
le
n
ee
d
o
n
a
ce
n
tr
al
d
e
v
ice
[
1
5
]
.
I
n
o
r
d
er
t
o
d
etec
t
in
ter
n
al
an
d
ex
ter
n
al
d
an
g
er
s
,
ea
ch
in
ter
n
et
o
f
th
in
g
s
d
ev
ice
(
I
o
T
D)
h
as
th
e
ab
ilit
y
to
an
aly
ze
b
o
t
h
its
o
wn
d
ata
an
d
th
at
o
f
its
s
u
r
r
o
u
n
d
in
g
I
o
T
Ds.
A
f
r
esh
i
n
v
esti
g
atio
n
o
n
d
ee
p
lear
n
in
g
a
p
p
licatio
n
is
s
u
g
g
ested
in
[
1
6
]
.
T
h
ey
c
o
n
tr
asted
f
o
u
r
p
o
p
u
la
r
d
ee
p
lear
n
i
n
g
tech
n
iq
u
es
with
co
n
v
en
tio
n
al
m
ac
h
i
n
e
lear
n
in
g
tech
n
iq
u
es.
Usi
n
g
th
e
NSL
-
KDD
d
ataset
,
Ng
u
y
en
et
a
l.
[
1
7
]
p
r
esen
ted
a
d
ee
p
lear
n
in
g
-
b
ased
d
etec
tio
n
alg
o
r
ith
m
f
o
r
n
etwo
r
k
I
DS.
T
o
i
m
p
r
o
v
e
th
e
d
etec
tio
n
r
ate
o
f
ass
au
lts
o
n
m
o
b
il
e
clo
u
d
co
m
p
u
tin
g
e
n
v
ir
o
n
m
en
t
s
Kh
an
et
a
l.
[
1
8
]
s
u
g
g
ested
an
en
s
em
b
le
m
o
d
el
in
wh
ich
f
ea
tu
r
e
s
elec
tio
n
is
d
o
n
e
u
s
in
g
r
estricte
d
B
o
ltz
m
an
n
m
ac
h
in
e
(
R
B
M)
an
d
d
im
en
s
io
n
r
ed
u
ctio
n
is
d
o
n
e
ap
p
ly
i
n
g
p
r
in
cip
al
co
m
p
o
n
en
t
an
aly
s
is
.
A
co
s
t
-
s
en
s
itiv
e
d
ee
p
n
eu
r
al
n
etwo
r
k
wh
ich
ca
n
r
ep
ea
te
d
ly
d
is
co
v
er
r
eliab
le
ch
ar
ac
ter
is
tic
d
eleg
ac
ies
is
ex
p
lain
ed
in
[
1
9
]
.
T
h
e
r
elatio
n
s
h
ip
s
b
etwe
en
th
e
p
h
y
s
ical
an
d
cy
b
e
r
d
o
m
ain
s
to
d
ev
elo
p
a
co
n
d
itio
n
al
GAN
b
a
s
ed
m
o
d
el
f
o
r
o
b
s
er
v
in
g
c
r
itical
s
ec
u
r
ity
n
ee
d
s
[
2
0
]
.
A
co
m
b
in
atio
n
o
f
an
en
h
an
ce
d
au
to
e
n
co
d
e
r
k
n
o
wn
as
im
p
r
o
v
ed
c
o
n
d
itio
n
al
v
ar
iatio
n
al
a
u
to
en
co
d
er
(
I
C
VAE
)
an
d
a
n
in
tr
u
s
io
n
d
et
ec
tio
n
m
o
d
el
is
in
tr
o
d
u
ce
d
in
[
2
1
]
;
r
ea
ch
ed
ac
cu
r
ac
y
o
f
8
5
.
9
7
%
an
d
7
5
.
4
3
%
o
n
th
e
NSLK
DD
an
d
UNSW
N
B
1
5
d
atasets
,
r
esp
ec
tiv
ely
.
Usi
n
g
th
e
KDDT
est+
an
d
UNSW
N
B
1
5
d
atasets
,
co
r
r
esp
o
n
d
in
g
ly
,
T
ian
et
a
l.
[
2
2
]
d
ev
elo
p
ed
a
n
I
DS
b
ased
o
n
GAN
with
ac
c
u
r
ac
y
o
f
8
4
.
4
5
%
an
d
8
2
.
5
3
%
r
esp
ec
tiv
ely
.
Usi
n
g
th
e
UNSW
NB
1
5
d
ataset,
p
r
esen
ted
an
I
DS
estab
lis
h
ed
o
n
en
h
a
n
ce
d
d
ee
p
b
elief
n
etwo
r
k
(
DB
N)
th
at
ac
h
iev
ed
ac
cu
r
ac
y
o
f
8
6
.
4
9
%
o
n
UNSW
NB
1
5
d
ataset.
A
two
-
s
tag
e
class
if
i
er
en
s
em
b
le
f
o
r
an
in
tellig
en
t
an
o
m
aly
-
b
ased
I
D
S
is
d
escr
ib
ed
in
s
tu
d
y
[
2
3
]
.
T
wo
-
s
tag
e
en
s
em
b
le
in
tr
u
s
io
n
d
etec
tio
n
s
y
s
tem
(
T
SE
-
I
DS)
h
as
d
em
o
n
s
tr
ated
9
1
.
2
7
%,
7
2
.
5
2
%,
an
d
8
5
.
7
9
%,
an
d
class
if
icatio
n
ac
cu
r
ac
y
o
n
UNSW
-
NB
1
5
,
KDDT
est
-
2
1
an
d
KDDT
est+d
atasets
.
B
ay
esian
d
ec
is
io
n
m
o
d
el
b
ase
d
r
eliab
le
r
o
u
te
f
o
r
m
atio
n
m
o
d
el
d
etec
ts
th
e
u
n
r
eliab
le
n
o
d
e
d
etec
tio
n
.
Activ
e
an
d
p
ass
iv
e
attac
k
r
ec
o
g
n
itio
n
m
eth
o
d
s
r
ec
o
g
n
ize
u
n
r
eliab
le
n
o
d
e.
R
em
ain
in
g
en
er
g
y
,
n
o
d
e
d
e
g
r
ee
,
an
d
p
ac
k
et
tr
a
n
s
m
is
s
io
n
r
ate
p
ar
am
eter
s
to
m
o
n
ito
r
th
eir
n
o
d
e
p
o
s
s
ib
ilit
ies
f
o
r
r
ec
o
g
n
izin
g
th
e
p
ass
iv
e
u
n
r
eliab
le
n
o
d
es
[
2
4
]
.
Netwo
r
k
in
tr
u
s
io
n
d
etec
tio
n
s
y
s
tem
b
y
ap
p
l
y
in
g
e
n
s
em
b
le
m
o
d
el
t
o
co
r
r
ec
t
th
e
er
r
o
r
s
u
n
til
n
o
f
u
r
th
e
r
im
p
r
o
v
em
e
n
t
s
[
2
5
]
.
K
-
m
ea
n
s
clu
s
ter
in
g
i
m
p
r
o
v
es
r
eso
u
r
ce
allo
ca
tio
n
ef
f
icien
cy
an
d
p
a
v
es
th
e
way
f
o
r
p
r
ec
is
e
au
to
-
s
ca
lin
g
[
2
6
]
.
Den
ial
o
f
s
er
v
ice
(
D
o
S)
attac
k
d
etec
tio
n
an
d
h
ill
clim
b
in
g
(
DDHC)
b
ased
o
p
tim
al
f
o
r
war
d
er
s
elec
tio
n
m
ec
h
an
is
m
to
r
ec
o
g
n
i
ze
d
en
ial
o
f
s
er
v
ice
attac
k
s
.
Fu
zz
y
lear
n
in
g
is
p
r
o
p
o
s
ed
to
Do
S
th
r
ea
ts
.
T
h
e
n
o
d
e
b
a
n
d
wid
th
,
c
o
n
n
ec
t
iv
ity
,
p
ac
k
et
r
ec
eiv
ed
r
ate
,
u
tili
ze
d
en
er
g
y
an
d
r
esp
o
n
s
e
tim
e
p
ar
am
eter
s
to
n
o
tice
th
e
n
o
d
e
a
b
n
o
r
m
ality
.
T
h
is
ab
n
o
r
m
ality
c
o
n
f
ir
m
s
th
e
n
o
d
e'
s
f
u
tu
r
e
s
tate
an
d
o
b
s
er
v
es
th
e
Do
S
attac
k
er
.
A
f
u
zz
y
lear
n
in
g
to
d
is
tin
g
u
is
h
Do
S
attac
k
s
th
at
r
aises
attac
k
d
etec
tio
n
ac
cu
r
ac
y
[
2
7
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
I
n
tr
u
s
io
n
d
etec
tio
n
b
a
s
ed
o
n
g
en
era
tive
a
d
ve
r
s
a
r
ia
l n
etw
o
r
k
w
ith
…
(
Gn
a
n
a
m
Je
b
a
R
o
s
li
n
e
)
2493
3.
P
RO
P
O
SE
D
M
E
T
H
O
D
I
DS
is
a
s
ig
n
if
ican
t
s
ec
u
r
ity
s
o
lu
tio
n
f
o
r
id
en
tif
y
in
g
attac
k
s
.
T
h
e
co
n
v
en
tio
n
al
ML
alg
o
r
it
h
m
s
f
ailed
to
s
atis
f
y
th
e
n
ec
ess
ity
f
o
r
cy
b
er
s
ec
u
r
ity
.
An
ess
en
tial id
ea
b
eh
in
d
th
e
I
DS is to
r
ec
o
g
n
ize
d
ec
eitf
u
l a
ctio
n
s
to
p
r
o
tecte
d
u
s
er
d
ata
as
well
a
s
clo
u
d
s
er
v
ices.
T
h
e
GAN
-
R
F
m
ec
h
an
is
m
p
r
o
p
o
s
es
a
d
i
s
tin
ctiv
e
m
eth
o
d
to
ap
p
ly
GANs
to
d
ev
el
o
p
s
ec
u
r
i
ty
in
clo
u
d
n
etwo
r
k
s
.
T
h
e
e
x
p
l
o
it
o
f
GANs
to
o
f
f
er
ar
tific
ial
d
ata
th
at
s
im
u
lates
ty
p
ical
n
etwo
r
k
ac
tio
n
to
e
n
h
a
n
ce
th
e
e
f
f
ec
tiv
en
ess
o
f
I
DS.
T
h
r
o
u
g
h
tr
ai
n
in
g
with
b
o
t
h
att
ac
k
an
d
u
s
u
al
d
ata,
th
e
GAN
im
p
r
o
v
es th
e
s
y
s
tem
'
s
ab
ili
ty
to
d
is
tin
g
u
is
h
b
etwe
e
n
m
alicio
u
s
an
d
s
af
e
n
etwo
r
k
ac
tiv
ity
.
3
.
1
.
G
ener
a
t
iv
e
a
dv
er
s
a
ria
l net
w
o
rk
An
u
n
s
u
p
er
v
is
ed
d
ee
p
lear
n
in
g
n
etwo
r
k
ca
lled
t
h
e
g
en
e
r
ativ
e
ad
v
er
s
ar
ial
n
etwo
r
k
[
2
8
]
,
[
2
9
]
d
o
es
n
o
t
n
ee
d
lab
ellin
g
o
f
th
e
tr
ain
in
g
d
ataset
o
r
its
s
tr
u
ctu
r
e.
Usi
n
g
th
e
r
ea
l
d
ata
f
r
o
m
th
e
tr
ain
i
n
g
d
ataset
as
its
in
p
u
t,
th
e
G
jo
b
is
to
p
r
o
d
u
ce
f
alse
d
ata
wh
ich
is
eq
u
iv
alen
t
to
th
e
r
ea
l
d
ata
b
y
a
d
d
in
g
n
o
is
e
d
ata
an
d
e
x
tr
ac
tin
g
laten
t
ch
ar
ac
ter
is
tics
f
r
o
m
th
e
r
ea
l
d
ata.
T
h
e
d
is
cr
im
in
ato
r
(
D)
an
d
g
en
e
r
ato
r
(
G)
a
r
e
th
e
t
wo
co
m
p
o
n
en
ts
o
f
a
GAN
as d
em
o
n
s
tr
ated
in
Fig
u
r
e
1
.
I
n
ess
en
ce
,
th
e
D
r
ep
r
esen
ts
a
d
ee
p
n
eu
r
al
n
etwo
r
k
class
if
ier
th
at
in
p
u
ts
b
o
th
ac
tu
al
an
d
f
ak
e
d
ata
p
r
o
d
u
ce
d
b
y
th
e
G
b
e
f
o
r
e
p
r
o
d
u
cin
g
its
ju
d
g
m
e
n
tal
r
esu
lt.
T
h
e
D
an
d
th
e
G
will
r
e
ce
iv
e
in
d
e
p
en
d
e
n
t
in
s
tr
u
ctio
n
d
u
r
in
g
th
is
p
r
o
ce
s
s
.
T
h
e
lo
s
s
o
p
er
atio
n
o
f
GAN
is
g
iv
en
in
(
1
)
.
(
,
)
=
(
)
[
(
)
]
+
(
)
[
(
1
−
(
(
)
)
)
]
(
1
)
wh
er
e
,
d
en
o
tes
th
e
in
p
u
t
s
am
p
le;
d
ep
icts
th
e
r
an
d
o
m
n
o
is
e;
p
(
x
)
r
ep
r
esen
ts
th
e
d
is
tr
ib
u
tio
n
o
f
;
p
′
(
z
)
r
ep
r
esen
ts
th
e
d
is
tr
ib
u
tio
n
o
f
;
G
(
z
)
an
d
D
(
x
)
d
escr
ib
es
th
e
o
u
tp
u
ts
o
f
G
an
d
D
r
esp
ec
tiv
ely
.
W
h
i
le
th
e
D
ac
cu
r
ate
r
ate
is
h
ig
h
,
it
m
u
s
t
b
e
ad
ju
s
ted
,
an
d
th
e
G
s
etti
n
g
s
m
u
s
t
b
e
ad
ju
s
ted
to
p
r
o
d
u
ce
m
o
r
e
r
ea
lis
tic
-
lo
o
k
in
g
p
h
o
n
ey
d
ata.
W
h
e
n
t
h
e
d
is
cr
im
in
ato
r
'
s
er
r
o
r
r
ate
i
s
lar
g
e,
th
e
G
is
to
b
e
r
ep
air
ed
,
an
d
p
a
r
am
eter
tu
n
in
g
is
d
o
n
e
b
y
th
e
D
to
im
p
r
o
v
e
its
d
is
cr
im
in
atin
g
p
er
f
o
r
m
an
ce
.
Fig
u
r
e
1
.
Stru
ctu
r
e
o
f
GAN
3
.
2
.
Ra
nd
o
m
f
o
re
s
t
A
tr
ad
itio
n
al
ML
m
o
d
el
ca
lle
d
r
a
n
d
o
m
f
o
r
est
[
3
0
]
is
f
r
eq
u
e
n
tly
em
p
l
o
y
ed
to
ad
d
r
ess
ca
teg
o
r
izatio
n
is
s
u
es.
Sev
er
al
d
ec
is
io
n
tr
ee
(
DT
)
m
o
d
els m
ak
e
u
p
a
R
F st
r
u
ctu
r
e
[
3
1
]
.
A
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
c
r
ea
tes th
e
d
iv
id
in
g
s
tan
d
a
r
d
s
o
f
th
e
p
r
esen
t
n
o
d
e
in
th
e
DT
m
o
d
el,
an
d
s
u
ch
m
eth
o
d
iter
ativ
ely
cr
ea
tes
n
o
d
es
d
escen
d
in
g
to
p
r
o
d
u
ce
a
s
tr
u
ctu
r
e
s
im
ilar
to
a
tr
ee
.
I
n
f
o
r
m
atio
n
en
tr
o
p
y
is
a
p
o
p
u
l
ar
f
ea
tu
r
e
s
elec
tio
n
ap
p
r
o
ac
h
am
o
n
g
th
e
m
an
y
th
at
ar
e
av
ailab
le.
A
r
an
d
o
m
v
ar
iab
le'
s
u
n
ce
r
tain
ty
is
r
ep
r
esen
ted
b
y
its
in
f
o
r
m
atio
n
en
tr
o
p
y
,
wh
e
r
e
a
h
ig
h
er
e
n
tr
o
p
y
n
u
m
b
er
i
n
d
icate
s
a
g
r
ea
ter
am
o
u
n
t
o
f
in
f
o
r
m
atio
n
in
th
e
v
ar
iab
le.
T
h
e
p
r
ed
ictio
n
(
P)
f
o
r
R
F a
lg
o
r
ith
m
is
g
iv
en
b
y
(
2
)
.
(
)
=
1
∑
=
1
(
2
)
wh
er
e,
is
th
e
p
r
o
b
ab
ilit
y
f
o
r
i
th
n
o
d
e
an
d
n
r
ep
r
esen
ts
th
e
n
u
m
b
er
o
f
d
ata
lab
els.
R
F
ap
p
r
o
ac
h
is
im
p
lem
en
ted
in
Py
th
o
n
ML
f
r
am
ewo
r
k
s
lik
e
s
cik
it
-
lear
n
,
d
esp
ite
r
eq
u
ir
in
g
a
lo
t
o
f
p
ar
a
m
eter
s
an
d
in
tr
icat
e
in
ter
ac
tio
n
s
.
T
h
e
attr
ib
u
te
with
th
e
g
r
ea
test
v
alu
e
is
id
en
tifi
ed
b
y
t
h
e
p
r
esen
t
n
o
d
e
b
y
co
m
p
u
tin
g
t
h
e
en
tr
o
p
y
o
f
th
e
attr
ib
u
tes
in
th
e
p
r
esen
t
attr
ib
u
te
s
et.
E
v
er
y
DT
is
b
u
ilt
u
s
in
g
a
co
m
p
ar
ab
le
p
r
o
ce
d
u
r
e,
an
d
u
ltima
tely
R
F
m
o
d
el
is
f
o
r
m
ed
b
y
th
e
co
m
b
in
atio
n
o
f
s
ev
er
al
DT
s
.
E
ac
h
d
ec
is
io
n
tr
ee
in
th
e
class
if
icatio
n
p
r
o
b
lem
in
d
icate
s
th
e
class
p
r
o
b
ab
ilit
y
o
f
th
e
in
p
u
t
s
am
p
le;
th
e
class
i
f
icatio
n
o
u
tc
o
m
e
is
d
eter
m
in
e
d
b
y
t
h
e
R
F
m
o
d
e
l
b
y
s
elec
tin
g
th
e
DT
with
t
h
e
h
ig
h
est
p
r
o
b
ab
ilit
y
.
T
h
e
f
lo
w
ch
ar
t
f
o
r
GAN
-
R
F
m
ec
h
an
is
m
is
d
em
o
n
s
tr
ated
in
Fig
u
r
e
2
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
2
,
Ap
r
il
20
25
:
2
4
9
1
-
2
4
9
8
2494
Fig
u
r
e
2
.
Flo
wch
ar
t
f
o
r
GAN
-
R
F in
tr
u
s
io
n
d
etec
tio
n
m
o
d
el
4.
RE
SU
L
T
S AN
A
L
YS
I
S AN
D
DIS
CU
SS
I
O
N
T
h
e
p
r
o
p
o
s
ed
m
ec
h
an
is
m
is
ex
ec
u
ted
b
y
ap
p
ly
in
g
Py
th
o
n
with
its
s
u
itab
le
lib
r
ar
ies.
C
I
C
I
DS2
0
1
8
d
ataset
is
em
p
lo
y
ed
in
th
is
w
o
r
k
wh
ic
h
is
th
e
m
o
s
t
r
ec
en
t,
lar
g
est,
an
d
m
o
s
t
im
p
o
r
tan
t
in
tr
u
s
io
n
d
etec
tio
n
d
ataset
av
ailab
le
f
o
r
f
r
ee
[
3
2
]
.
B
o
th
b
en
ig
n
an
d
m
alicio
u
s
co
m
m
u
n
icatio
n
s
ca
n
b
e
f
o
u
n
d
in
C
SV
f
iles
.
T
en
f
iles
in
all,
to
talin
g
6
.
4
1
GB
,
ar
e
in
clu
d
ed
i
n
th
e
c
o
llectio
n
[
3
3
]
.
T
h
e
r
e
ar
e
1
6
,
2
3
3
,
0
0
2
n
u
m
b
e
r
o
f
in
s
tan
ce
s
in
th
e
C
I
C
I
DS2
0
1
8
d
ataset.
All
th
ese
d
ataset
s
ar
e
u
tili
ze
d
in
th
is
wo
r
k
f
o
r
ass
ess
m
en
t.
T
h
e
d
ataset
in
clu
d
es
8
3
d
ata
attr
ib
u
tes,
in
clu
d
in
g
p
ac
k
et
co
u
n
t,
d
u
r
atio
n
,
b
y
tes,
in
ad
d
itio
n
to
a
s
tr
ea
m
o
f
p
ac
k
ets.
E
ac
h
d
ataset
s
am
p
le
co
n
clu
d
es
wit
h
a
lab
el
d
esig
n
atin
g
wh
eth
er
n
etwo
r
k
tr
af
f
ic
f
alls
in
to
th
e
b
en
ig
n
o
r
attac
k
ca
teg
o
r
y
.
T
h
is
s
ec
tio
n
m
ea
s
u
r
es
th
e
p
ar
am
eter
s
,
f
o
r
ex
am
p
le,
p
r
ec
is
io
n
(
PR
)
,
r
ec
all
(
R
E
)
,
F1
-
m
ea
s
u
r
e,
an
d
ac
cu
r
ac
y
(
AC
C
)
,
ar
e
u
tili
ze
d
t
o
ev
alu
ate
th
e
f
u
n
ctio
n
o
f
th
e
GAN
-
R
F
f
o
r
id
en
tif
y
in
g
in
tr
u
s
io
n
th
r
o
u
g
h
o
u
t
t
h
e
tr
ials
.
T
h
e
ca
lcu
latio
n
eq
u
atio
n
f
o
r
esti
m
atio
n
p
ar
am
eter
s
is
s
p
ec
if
ied
b
elo
w.
=
+
+
+
+
(
3
)
=
+
(
4
)
=
+
(
5
)
1
−
=
2
×
×
+
(
6
)
wh
er
e
d
en
o
tes
th
e
f
alse
n
eg
ativ
es,
in
d
icate
s
th
e
tr
u
e
n
eg
ativ
es,
r
ep
r
esen
ts
th
e
f
alse
p
o
s
itiv
es
an
d
d
ep
icts
th
e
tr
u
e
p
o
s
itiv
es.
T
ab
le
1
g
iv
es
p
er
f
o
r
m
an
ce
an
aly
s
is
o
f
p
r
o
p
o
s
ed
m
o
d
el
co
m
p
ar
ed
with
ex
is
tin
g
in
tr
u
s
i
o
n
d
etec
tio
n
m
eth
o
d
s
.
T
ab
le
1
.
Gan
-
R
F m
ec
h
an
is
m
p
er
ce
n
tag
e
o
f
p
r
ec
is
io
n
r
ec
all,
ac
cu
r
ac
y
a
n
d
F1
-
s
co
r
e
A
t
t
a
c
k
s
A
c
c
u
r
a
c
y
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F1
-
sc
o
r
e
B
e
n
i
g
n
9
5
.
7
5
%
9
4
.
2
4
%
9
5
.
3
2
%
9
4
.
7
8
%
B
r
u
t
e
f
o
r
c
e
9
4
.
7
1
%
9
3
.
1
1
%
9
2
.
1
7
%
9
2
.
6
4
%
D
o
S
9
5
%
9
3
.
8
7
%
9
4
.
5
4
%
9
4
.
2
0
%
W
e
b
9
5
.
2
4
%
9
4
.
5
1
%
9
5
.
2
2
%
9
4
.
8
6
%
I
n
f
i
l
t
r
a
t
i
o
n
9
3
.
7
8
%
9
3
.
1
4
%
9
2
.
9
1
%
9
3
.
0
2
%
B
o
t
n
e
t
9
4
.
8
0
%
9
4
%
9
3
.
8
7
%
9
3
.
9
3
%
D
D
o
S
9
5
.
7
7
%
9
4
.
2
8
%
9
3
.
6
9
%
9
3
.
9
8
%
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
I
n
tr
u
s
io
n
d
etec
tio
n
b
a
s
ed
o
n
g
en
era
tive
a
d
ve
r
s
a
r
ia
l n
etw
o
r
k
w
ith
…
(
Gn
a
n
a
m
Je
b
a
R
o
s
li
n
e
)
2495
I
n
th
is
wo
r
k
,
s
y
n
th
etic
m
in
o
r
i
ty
o
v
er
s
am
p
lin
g
tech
n
o
lo
g
y
(
SMOT
E
)
an
aly
s
is
is
n
o
t
em
p
l
o
y
ed
as
it
m
ay
ca
u
s
e
r
e
p
licatin
g
th
e
e
x
is
tin
g
d
ata
wh
ich
r
esu
lts
in
p
o
o
r
p
er
f
o
r
m
an
ce
.
T
ab
le
1
estab
lis
h
es
th
e
in
tr
o
d
u
ce
d
GAN
-
R
F
m
ec
h
an
is
m
p
er
ce
n
tag
e
o
f
p
r
ec
is
io
n
,
r
ec
all,
a
cc
u
r
ac
y
an
d
F1
-
s
co
r
e.
GAN
-
R
F
m
ec
h
an
is
m
ac
h
iev
e
d
av
er
ag
e
ac
cu
r
ac
y
r
ate
is
9
5
.
7
5
%
an
d
9
4
.
1
7
%
f
o
r
b
en
ig
n
a
s
well
as
o
th
er
attac
k
s
.
T
h
e
p
r
o
p
o
s
ed
GAN
-
R
F
m
ec
h
an
is
m
d
etec
ts
s
ev
er
al
ty
p
es o
f
attac
k
ac
cu
r
ac
y
p
er
ce
n
ta
g
e,
wh
ich
is
illu
s
tr
ated
in
Fig
u
r
e
3
.
Fro
m
Fig
u
r
e
3
,
c
o
m
p
a
r
ed
to
7
ty
p
es
o
f
attac
k
s
th
e
GA
N
-
R
F
m
ec
h
an
is
m
d
etec
tio
n
ac
cu
r
ac
y
o
f
b
en
ig
n
,
web
an
d
DDo
S
h
a
v
e
ab
o
v
e
9
5
%.
T
h
e
in
f
iltra
tio
n
h
as
b
elo
w
9
4
%.
Fig
u
r
e
4
ex
p
lain
s
GAN
-
R
F
m
ec
h
an
is
m
p
r
ec
is
io
n
p
er
ce
n
ta
g
e
co
m
p
a
r
ed
to
s
ev
er
al
ty
p
es o
f
attac
k
s
.
Fro
m
Fig
u
r
e
4
,
co
m
p
ar
ed
t
o
7
t
y
p
es
o
f
attac
k
s
th
e
GAN
-
R
F
m
ec
h
an
is
m
d
etec
ts
th
e
web
attac
k
p
r
e
cisi
o
n
p
er
ce
n
tag
e
is
h
ig
h
c
o
m
p
ar
ed
to
th
e
o
t
h
er
attac
k
s
.
T
h
e
b
r
u
te
f
o
r
ce
an
d
in
f
iltra
tio
n
h
av
e
b
elo
w
9
3
.
5
%.
Fig
u
r
e
5
ex
p
lain
s
GAN
-
R
F
m
ec
h
an
is
m
r
ec
all
p
er
ce
n
tag
e
c
o
m
p
ar
e
d
to
s
ev
er
al
ty
p
es o
f
attac
k
s
.
Fro
m
Fig
u
r
e
5
,
t
h
e
GAN
-
R
F
m
ec
h
an
is
m
r
ec
all
p
e
r
ce
n
tag
e
f
o
r
all
ty
p
e
attac
k
h
av
e
g
r
ea
ter
th
an
9
2
%.
I
n
ad
d
itio
n
,
t
h
e
b
e
n
ig
n
an
d
web
ty
p
e
o
f
attac
k
r
ec
all
p
er
ce
n
tag
e
is
ab
o
v
e
c
o
m
p
ar
e
d
t
o
th
e
o
th
er
ty
p
es
o
f
attac
k
s
.
Fig
u
r
e
6
ex
p
lain
s
GA
N
-
R
F
m
ec
h
an
is
m
F1
-
s
co
r
e
f
o
r
s
ev
er
al
ty
p
es
o
f
attac
k
s
.
C
o
m
p
ar
ed
to
all
ty
p
es
o
f
attac
k
s
,
th
e
GAN
-
R
F
m
ec
h
an
is
m
F1
-
s
co
r
e
f
o
r
web
a
n
d
b
e
n
ig
n
h
av
e
ab
o
v
e
9
4
%
th
an
o
th
er
ty
p
es
o
f
attac
k
s
.
Sp
ec
if
ically
,
b
o
th
th
e
m
in
o
r
ity
class
's
an
d
th
e
r
eg
u
lar
class
'
s
p
er
f
o
r
m
an
ce
o
u
g
h
t
to
b
e
en
h
an
ce
d
if
th
e
n
u
m
b
er
o
f
m
in
o
r
ity
class
es
-
lik
e
B
o
t,
I
n
f
iltra
tio
n
,
a
n
d
B
r
u
tef
o
r
ce
-
is
o
v
er
s
am
p
led
.
Fig
u
r
e
3
.
GAN
-
R
F m
ec
h
an
is
m
ac
cu
r
ac
y
v
er
s
u
s
attac
k
ty
p
e
s
Fig
u
r
e
4
.
GAN
-
R
F m
ec
h
an
is
m
p
r
ec
is
io
n
v
e
r
s
u
s
attac
k
ty
p
e
s
Fig
u
r
e
5
.
GAN
-
R
F m
ec
h
an
is
m
r
ec
all
v
er
s
u
s
attac
k
ty
p
es
92.
50%
93.
00%
93.
50%
94.
00%
94.
50%
95.
00%
95.
50%
96.
00%
B
e
ni
g
n
B
r
ut
e
f
o
r
c
e
D
o
S
W
e
b
Inf
i
l
t
r
at
i
o
n
B
o
t
ne
t
D
D
o
S
A
ccur
acy
(%
)
A
t
t
a
c
k
s
92.
00%
92.
50%
93.
00%
93.
50%
94.
00%
94.
50%
95.
00%
B
e
ni
g
n
B
r
ut
e
f
o
r
c
e
D
o
S
W
e
b
Inf
i
l
t
r
at
i
o
n
B
o
t
ne
t
D
D
o
S
P
re
ci
si
on
(%
)
A
t
t
a
c
k
s
90.
00%
91.
00%
92.
00%
93.
00%
94.
00%
95.
00%
96.
00%
B
e
ni
g
n
B
r
ut
f
o
r
c
e
D
o
S
W
e
b
Inf
i
l
t
r
at
i
o
n
B
o
t
ne
t
D
D
o
S
R
e
cal
l
(%)
A
t
t
a
c
k
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
2
,
Ap
r
il
20
25
:
2
4
9
1
-
2
4
9
8
2496
Fig
u
r
e
6
.
GAN
-
R
F m
ec
h
an
is
m
F1
s
co
r
e
v
er
s
u
s
attac
k
ty
p
es
B
ec
au
s
e
th
e
f
ea
tu
r
es
o
f
th
e
m
in
o
r
ity
class
d
if
f
er
g
r
ea
tly
f
r
o
m
th
o
s
e
o
f
o
th
er
class
es,
lear
n
in
g
m
o
r
e
ab
o
u
t
t
h
em
a
p
p
ea
r
s
t
o
h
av
e
b
o
o
s
ted
th
e
p
er
f
o
r
m
a
n
ce
o
f
t
h
e
r
eg
u
lar
class
.
T
h
e
f
o
llo
win
g
ar
e
th
e
ca
u
s
es
o
f
d
ec
r
ea
s
ed
Pre
cisi
o
n
f
o
r
I
n
f
il
tr
atio
n
.
Dep
e
n
d
in
g
o
n
t
h
e
q
u
an
tity
o
f
d
ata,
p
r
ec
is
io
n
a
n
d
r
ec
all
f
r
eq
u
en
tly
co
n
tr
ad
ict
o
n
e
an
o
th
er
;
m
o
r
e
p
r
ec
is
io
n
ty
p
ically
r
esu
lts
in
l
o
wer
r
ec
all
an
d
th
e
o
th
er
way
ar
o
u
n
d
.
T
h
e
GAN
-
R
F
m
ec
h
an
is
m
co
m
p
ar
ed
wit
h
I
C
VAE
,
GAN,
DB
N
an
d
T
SE
-
I
DS
in
T
ab
le
2
.
T
h
e
o
v
er
all
ac
cu
r
ac
y
f
o
r
th
e
p
r
o
p
o
s
ed
GAN
-
R
F
is
ac
h
iev
e
d
9
5
.
0
1
%
wh
ich
is
s
u
p
er
io
r
t
o
o
th
er
a
p
p
r
o
ac
h
es.
T
h
e
ac
cu
r
ac
y
o
f
GAN
-
R
F,
I
C
VAE
,
GAN,
DB
N
an
d
T
SE
-
I
DS m
ec
h
an
is
m
s
ar
e
s
p
ec
if
ie
d
in
Fig
u
r
e
7
.
C
o
m
p
ar
ed
to
all
o
th
er
m
ec
h
an
is
m
s
,
th
e
I
C
VAE
m
ec
h
an
is
m
ac
cu
r
ac
y
p
er
ce
n
ta
g
e
is
8
5
.
9
7
%,
GAN
m
ec
h
an
is
m
is
8
4
.
4
5
%,
D
B
N
ac
cu
r
ac
y
v
alu
e
is
8
6
.
4
9
,
T
SE
-
I
DS
m
ec
h
an
is
m
a
cc
u
r
ac
y
r
ate
is
9
1
.
4
7
%
an
d
p
r
o
p
o
s
ed
GAN
-
R
F
is
r
ea
ch
ed
9
5
.
0
1
%
th
at
is
h
ig
h
er
th
an
o
th
er
m
ec
h
an
is
m
s
.
T
ab
le
2
.
Gan
-
R
F m
ec
h
an
is
m
co
m
p
ar
ed
with
I
C
VAE
,
GAN,
DB
N
an
d
T
SE
-
I
DS m
ec
h
an
is
m
s
M
o
d
e
l
s
D
a
t
a
s
e
t
A
c
c
u
r
a
c
y
(
%)
I
C
V
A
E
N
S
LK
D
D
8
5
.
9
7
GAN
K
D
D
Te
s
t
+
8
4
.
4
5
D
B
N
U
N
S
W
N
B
1
5
d
a
t
a
se
t
8
6
.
4
9
TSE
-
I
D
S
U
N
S
W
-
N
B
1
5
9
1
.
2
7
GAN
-
RF
C
I
C
-
I
D
S
2
0
1
8
9
5
.
0
1
Fig
u
r
e
7
.
Acc
u
r
ac
y
o
f
GAN
-
R
F,
I
C
VAE
,
GAN,
DB
N
an
d
T
SE
-
I
DS m
ec
h
an
is
m
s
5.
CO
NCLU
SI
O
N
T
h
is
wo
r
k
p
r
o
p
o
s
ed
th
e
in
te
g
r
atio
n
o
f
GAN
an
d
R
F
class
if
ier
f
o
r
in
tr
u
s
io
n
d
etec
tio
n
in
clo
u
d
in
f
r
astru
ctu
r
e.
W
h
en
it
co
m
e
s
to
in
tr
u
s
io
n
d
etec
tio
n
,
R
F
ca
n
b
e
u
s
ed
o
n
t
h
e
ch
ar
ac
ter
is
tics
th
at
th
e
GAN
ex
tr
ac
ted
in
o
r
d
e
r
to
d
eter
m
in
e
wh
ich
f
ea
tu
r
es
ar
e
m
o
s
t
p
er
tin
en
t
an
d
d
is
cr
im
in
ativ
e
in
ter
m
s
o
f
id
en
tify
in
g
n
o
r
m
al
f
r
o
m
an
o
m
alo
u
s
ac
tiv
ity
.
T
h
e
GAN
was
tr
ain
ed
o
n
th
e
d
ata
co
u
n
ts
o
f
th
e
r
ar
e
class
es
Hea
r
tb
leed
,
I
n
f
iltra
tio
n
,
an
d
B
o
t,
t
h
en
o
v
er
s
am
p
led
1
0
,
0
0
0
d
ata
p
o
in
ts
to
e
v
alu
ate
class
if
icatio
n
p
er
f
o
r
m
an
ce
.
T
h
e
test
f
in
d
in
g
s
d
em
o
n
s
tr
ate
th
at
R
F
class
if
icatio
n
p
er
f
o
r
m
an
ce
,
f
o
llo
win
g
GAN
r
esam
p
lin
g
,
o
u
tp
er
f
o
r
m
ed
s
in
g
le
R
F
class
if
icatio
n
with
o
u
t
r
esa
m
p
lin
g
.
R
esu
lts
s
p
ec
if
ically
i
n
d
icate
d
th
at
m
in
o
r
ity
class
es
p
er
f
o
r
m
ed
b
etter
in
class
if
icatio
n
th
an
r
eg
u
la
r
cla
s
s
es
d
id
.
B
ec
au
s
e
th
e
ch
ar
ac
t
er
is
tics
o
f
th
e
m
in
o
r
ity
class
d
if
f
er
g
r
ea
tly
f
r
o
m
th
o
s
e
o
f
o
th
er
class
es,
lear
n
in
g
m
o
r
e
ab
o
u
t
t
h
em
a
p
p
ea
r
s
to
h
av
e
e
n
h
an
ce
d
p
e
r
f
o
r
m
an
ce
i
n
class
if
y
in
g
n
o
r
m
al
class
es.
T
h
e
p
r
o
p
o
s
ed
GAN
-
R
F
m
o
d
el
ac
h
iev
ed
9
5
.
0
1
%
o
f
class
if
icatio
n
ac
cu
r
ac
y
wh
ich
o
u
tp
er
f
o
r
m
e
d
91.
50%
92.
00%
92.
50%
93.
00%
93.
50%
94.
00%
94.
50%
95.
00%
95.
50%
B
e
ni
g
n
B
r
ut
f
o
r
c
e
D
o
S
W
e
b
Inf
i
l
t
r
at
i
o
n
B
o
t
ne
t
D
D
o
S
F1
S
co
re
(
%
)
A
t
t
a
c
k
s
75
80
85
90
95
100
IC
VA
E
G
A
N
D
B
N
T
S
E
-
IDS
G
A
N
-
R
F
M
e
c
h
a
n
i
sm
s
A
c
c
u
r
a
c
y
(
%
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
I
n
tr
u
s
io
n
d
etec
tio
n
b
a
s
ed
o
n
g
en
era
tive
a
d
ve
r
s
a
r
ia
l n
etw
o
r
k
w
ith
…
(
Gn
a
n
a
m
Je
b
a
R
o
s
li
n
e
)
2497
v
ar
io
u
s
attac
k
s
wh
ic
h
is
h
i
g
h
lig
h
ted
in
r
esu
lt
s
ec
tio
n
.
T
h
e
ea
v
esd
r
o
p
p
e
r
lis
ten
in
g
th
e
d
etails
o
f
d
ata
f
o
r
war
d
in
g
an
d
r
ec
eiv
in
g
in
t
h
e
n
etwo
r
k
.
I
n
f
u
tu
r
e,
p
r
o
tects
th
e
d
ata
f
r
o
m
ea
v
esd
r
o
p
p
er
a
ttack
er
in
th
e
clo
u
d
in
f
r
astru
ctu
r
e.
RE
F
E
R
E
NC
E
S
[
1
]
Z.
Ta
r
i
,
“
S
e
c
u
r
i
t
y
a
n
d
p
r
i
v
a
c
y
i
n
c
l
o
u
d
c
o
mp
u
t
i
n
g
,
”
I
EEE
C
l
o
u
d
C
o
m
p
u
t
i
n
g
,
v
o
l
.
1
,
n
o
.
1
,
p
p
.
5
4
–
5
7
,
M
a
y
2
0
1
4
,
d
o
i
:
1
0
.
1
1
0
9
/
M
C
C
.
2
0
1
4
.
2
0
.
[
2
]
M
.
B
.
M
o
l
l
a
h
,
M
.
A
.
K
.
A
z
a
d
,
a
n
d
A
.
V
a
si
l
a
k
o
s
,
“
S
e
c
u
r
i
t
y
a
n
d
p
r
i
v
a
c
y
c
h
a
l
l
e
n
g
e
s
i
n
mo
b
i
l
e
c
l
o
u
d
c
o
mp
u
t
i
n
g
:
S
u
r
v
e
y
a
n
d
w
a
y
a
h
e
a
d
,
”
J
o
u
r
n
a
l
o
f
N
e
t
w
o
rk
a
n
d
C
o
m
p
u
t
e
r
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
8
4
,
p
p
.
3
8
–
5
4
,
A
p
r
.
2
0
1
7
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
n
c
a
.
2
0
1
7
.
0
2
.
0
0
1
.
[
3
]
Y
.
S
.
A
b
d
u
l
s
a
l
a
m
a
n
d
M
.
H
e
d
a
b
o
u
,
“
S
e
c
u
r
i
t
y
a
n
d
p
r
i
v
a
c
y
i
n
c
l
o
u
d
c
o
m
p
u
t
i
n
g
:
t
e
c
h
n
i
c
a
l
r
e
v
i
e
w
,
”
F
u
t
u
r
e
I
n
t
e
rn
e
t
,
v
o
l
.
1
4
,
n
o
.
1
,
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
f
i
1
4
0
1
0
0
1
1
.
[
4
]
L
.
V
u
,
Q
.
U
.
N
g
u
y
e
n
,
D
.
N
.
N
g
u
y
e
n
,
D
.
T
.
H
o
a
n
g
,
a
n
d
E
.
D
u
t
k
i
e
w
i
c
z
,
“
D
e
e
p
G
e
n
e
r
a
t
i
v
e
L
e
a
r
n
i
n
g
M
o
d
e
l
s
f
o
r
C
l
o
u
d
I
n
t
r
u
s
i
o
n
D
e
t
e
c
t
i
o
n
S
y
s
t
e
m
s
,
”
I
E
E
E
T
r
a
n
s
a
c
t
i
o
n
s
o
n
C
y
b
e
r
n
e
t
i
c
s
,
v
o
l
.
5
3
,
n
o
.
1
,
p
p
.
5
6
5
–
5
7
7
,
J
a
n
.
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
T
C
Y
B
.
2
0
2
2
.
3
1
6
3
8
1
1
.
[
5
]
R
.
K
.
V
a
n
a
k
a
mam
i
d
i
,
N
.
A
b
i
r
a
mi
,
C
.
S
a
si
K
u
m
a
r
,
L
.
R
a
ma
l
i
n
g
a
m,
S
.
P
r
i
y
a
n
k
a
,
a
n
d
S
.
M
u
r
u
g
a
n
,
“
I
o
T
se
c
u
r
i
t
y
b
a
se
d
o
n
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
,
”
i
n
2
0
2
3
2
n
d
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
S
m
a
rt
T
e
c
h
n
o
l
o
g
i
e
s
f
o
r
S
m
a
r
t
N
a
t
i
o
n
,
S
m
a
r
t
T
e
c
h
C
o
n
2
0
2
3
,
A
u
g
.
2
0
2
3
,
p
p
.
6
8
3
–
6
8
7
,
d
o
i
:
1
0
.
1
1
0
9
/
S
mart
Te
c
h
C
o
n
5
7
5
2
6
.
2
0
2
3
.
1
0
3
9
1
7
2
7
.
[
6
]
D
.
P
.
K
i
n
g
ma
a
n
d
M
.
W
e
l
l
i
n
g
,
“
A
u
t
o
-
e
n
c
o
d
i
n
g
v
a
r
i
a
t
i
o
n
a
l
b
a
y
e
s
,
”
a
rX
i
v
p
r
e
p
r
i
n
t
a
rX
i
v
:
1
3
1
2
.
6
1
1
4
,
2
0
1
4
.
[
7
]
M
.
H
.
S
h
a
h
r
i
a
r
,
N
.
I
.
H
a
q
u
e
,
M
.
A
.
R
a
h
m
a
n
,
a
n
d
M
.
A
l
o
n
s
o
,
“
G
-
I
D
S
:
G
e
n
e
r
a
t
i
v
e
a
d
v
e
r
sar
i
a
l
n
e
t
w
o
r
k
s
a
ssi
st
e
d
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
sy
st
e
m,
”
i
n
2
0
2
0
I
EEE
4
4
t
h
A
n
n
u
a
l
C
o
m
p
u
t
e
rs
,
S
o
f
t
w
a
r
e
,
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
C
o
n
f
e
re
n
c
e
(
C
O
M
PS
A
C
)
,
J
u
l
.
2
0
2
0
,
p
p
.
3
7
6
–
3
8
5
,
d
o
i
:
1
0
.
1
1
0
9
/
C
O
M
P
S
A
C
4
8
6
8
8
.
2
0
2
0
.
0
-
2
1
8
.
[
8
]
W
.
X
u
,
J
.
Ja
n
g
-
J
a
c
c
a
r
d
,
T.
L
i
u
,
F
.
S
a
b
r
i
n
a
,
a
n
d
J
.
K
w
a
k
,
“
I
mp
r
o
v
e
d
b
i
d
i
r
e
c
t
i
o
n
a
l
G
A
N
-
b
a
s
e
d
a
p
p
r
o
a
c
h
f
o
r
n
e
t
w
o
r
k
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
u
si
n
g
o
n
e
-
c
l
a
ss
c
l
a
ss
i
f
i
e
r
,
”
C
o
m
p
u
t
e
rs
,
v
o
l
.
1
1
,
n
o
.
6
,
p
.
8
5
,
M
a
y
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
c
o
m
p
u
t
e
r
s
1
1
0
6
0
0
8
5
.
[
9
]
S
.
G
a
r
g
,
K
.
K
a
u
r
,
N
.
K
u
mar,
a
n
d
J.
J.
P
.
C
.
R
o
d
r
i
g
u
e
s,
“
H
y
b
r
i
d
d
e
e
p
-
l
e
a
r
n
i
n
g
-
b
a
s
e
d
a
n
o
mal
y
d
e
t
e
c
t
i
o
n
sc
h
e
me
f
o
r
su
s
p
i
c
i
o
u
s
f
l
o
w
d
e
t
e
c
t
i
o
n
i
n
S
D
N
:
A
s
o
c
i
a
l
m
u
l
t
i
me
d
i
a
p
e
r
s
p
e
c
t
i
v
e
,
”
I
EEE
T
r
a
n
sa
c
t
i
o
n
s
o
n
M
u
l
t
i
m
e
d
i
a
,
v
o
l
.
2
1
,
n
o
.
3
,
p
p
.
5
6
6
–
5
7
8
,
2
0
1
9
,
d
o
i
:
1
0
.
1
1
0
9
/
TM
M
.
2
0
1
9
.
2
8
9
3
5
4
9
.
[
1
0
]
S
.
K
.
S
e
k
a
r
e
t
a
l
.
,
“
R
a
n
d
o
m
f
o
r
e
s
t
a
l
g
o
r
i
t
h
m
w
i
t
h
h
i
l
l
c
l
i
mb
i
n
g
a
l
g
o
r
i
t
h
m
t
o
i
m
p
r
o
v
e
i
n
t
r
u
s
i
o
n
d
e
t
e
c
t
i
o
n
a
t
e
n
d
p
o
i
n
t
a
n
d
n
e
t
w
o
r
k
,
”
I
n
d
o
n
e
si
a
n
J
o
u
r
n
a
l
o
f
E
l
e
c
t
ri
c
a
l
E
n
g
i
n
e
e
r
i
n
g
a
n
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
3
7
,
n
o
.
1
,
p
p
.
1
3
4
–
1
4
2
,
2
0
2
5
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
e
c
s.
v
3
7
.
i
1
.
p
p
1
3
4
-
1
4
2
.
[
1
1
]
B
.
M
e
e
n
a
k
s
h
i
,
B
.
G
o
p
i
,
L.
R
a
ma
l
i
n
g
a
m
,
A
.
V
a
n
a
t
h
i
,
S
.
S
a
n
g
e
e
t
h
a
,
a
n
d
S
.
M
u
r
u
g
a
n
,
“
W
i
r
e
l
e
ss
s
e
n
s
o
r
n
e
t
w
o
r
k
s
f
o
r
d
i
sas
t
e
r
man
a
g
e
me
n
t
a
n
d
e
mer
g
e
n
c
y
r
e
s
p
o
n
s
e
u
s
i
n
g
S
V
M
c
l
a
ss
i
f
i
e
r
,
”
i
n
2
0
2
3
2
n
d
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
S
m
a
rt
T
e
c
h
n
o
l
o
g
i
e
s
f
o
r
S
m
a
r
t
N
a
t
i
o
n
,
S
m
a
r
t
T
e
c
h
C
o
n
2
0
2
3
,
A
u
g
.
2
0
2
3
,
p
p
.
6
4
7
–
6
5
1
,
d
o
i
:
1
0
.
1
1
0
9
/
S
martT
e
c
h
C
o
n
5
7
5
2
6
.
2
0
2
3
.
1
0
3
9
1
4
3
5
.
[
1
2
]
G
.
T
h
a
h
n
i
y
a
t
h
e
t
a
l
.
,
“
C
l
o
u
d
b
a
s
e
d
p
r
e
d
i
c
t
i
o
n
o
f
e
p
i
l
e
p
t
i
c
s
e
i
z
u
r
e
s
u
s
i
n
g
r
e
a
l
-
t
i
me
e
l
e
c
t
r
o
e
n
c
e
p
h
a
l
o
g
r
a
ms
a
n
a
l
y
s
i
s,”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
t
r
i
c
a
l
a
n
d
C
o
m
p
u
t
e
r
E
n
g
i
n
e
e
r
i
n
g
,
v
o
l
.
1
4
,
n
o
.
5
,
p
p
.
6
0
4
7
–
6
0
5
6
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
c
e
.
v
1
4
i
5
.
p
p
6
0
4
7
-
6
0
5
6
.
[
1
3
]
P
.
S
.
R
a
m
a
p
r
a
b
a
e
t
a
l
.
,
“
I
mp
l
e
m
e
n
t
i
n
g
c
l
o
u
d
c
o
m
p
u
t
i
n
g
i
n
d
r
u
g
d
i
sc
o
v
e
r
y
a
n
d
t
e
l
e
me
d
i
c
i
n
e
f
o
r
q
u
a
n
t
i
t
a
t
i
v
e
st
r
u
c
t
u
r
e
-
a
c
t
i
v
i
t
y
r
e
l
a
t
i
o
n
s
h
i
p
a
n
a
l
y
si
s
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
t
r
i
c
a
l
a
n
d
C
o
m
p
u
t
e
r
E
n
g
i
n
e
e
r
i
n
g
,
v
o
l
.
1
5
,
n
o
.
1
,
p
p
.
1
1
3
2
–
1
1
4
1
,
2
0
2
5
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
c
e
.
v
1
5
i
1
.
p
p
1
1
3
2
-
1
1
4
1
.
[
1
4
]
A
.
F
e
r
d
o
w
si
a
n
d
W
.
S
a
a
d
,
“
G
e
n
e
r
a
t
i
v
e
a
d
v
e
r
s
a
r
i
a
l
n
e
t
w
o
r
k
s
f
o
r
d
i
st
r
i
b
u
t
e
d
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
i
n
t
h
e
i
n
t
e
r
n
e
t
o
f
t
h
i
n
g
s,
”
i
n
2
0
1
9
I
EEE
G
l
o
b
a
l
C
o
m
m
u
n
i
c
a
t
i
o
n
s
C
o
n
f
e
r
e
n
c
e
(
G
L
O
BE
C
O
M)
,
D
e
c
.
2
0
1
9
,
p
p
.
1
–
6
,
d
o
i
:
1
0
.
1
1
0
9
/
G
LO
B
E
C
O
M
3
8
4
3
7
.
2
0
1
9
.
9
0
1
4
1
0
2
.
[
1
5
]
R
.
Zh
a
o
,
R
.
Y
a
n
,
Z.
C
h
e
n
,
K
.
M
a
o
,
P
.
W
a
n
g
,
a
n
d
R
.
X
.
G
a
o
,
“
D
e
e
p
l
e
a
r
n
i
n
g
a
n
d
i
t
s
a
p
p
l
i
c
a
t
i
o
n
s
t
o
mac
h
i
n
e
h
e
a
l
t
h
mo
n
i
t
o
r
i
n
g
,
”
Me
c
h
a
n
i
c
a
l
S
y
st
e
m
s
a
n
d
S
i
g
n
a
l
Pr
o
c
e
ssi
n
g
,
v
o
l
.
1
1
5
,
p
p
.
2
1
3
–
2
3
7
,
Ja
n
.
2
0
1
9
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
y
mss
p
.
2
0
1
8
.
0
5
.
0
5
0
.
[
1
6
]
A
.
Jav
a
i
d
,
Q
.
N
i
y
a
z
,
W
.
S
u
n
,
a
n
d
M
.
A
l
a
m,
“
A
d
e
e
p
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
f
o
r
n
e
t
w
o
r
k
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
sy
st
e
m,”
i
n
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
9
t
h
EAI
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
B
i
o
-
i
n
s
p
i
re
d
I
n
f
o
rm
a
t
i
o
n
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
s
T
e
c
h
n
o
l
o
g
i
e
s
(
f
o
rm
e
r
l
y
BI
O
N
ETI
C
S
)
,
2
0
1
6
,
p
p
.
2
1
–
2
6
,
d
o
i
:
1
0
.
4
1
0
8
/
e
a
i
.
3
-
12
-
2
0
1
5
.
2
2
6
2
5
1
6
.
[
1
7
]
K
.
K
.
N
g
u
y
e
n
,
D
.
T.
H
o
a
n
g
,
D
.
N
i
y
a
t
o
,
P
.
W
a
n
g
,
D
.
N
g
u
y
e
n
,
a
n
d
E
.
D
u
t
k
i
e
w
i
c
z
,
“
C
y
b
e
r
a
t
t
a
c
k
d
e
t
e
c
t
i
o
n
i
n
m
o
b
i
l
e
c
l
o
u
d
c
o
m
p
u
t
i
n
g
:
A
d
e
e
p
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
,
”
i
n
I
EEE
W
i
re
l
e
ss
C
o
m
m
u
n
i
c
a
t
i
o
n
s
a
n
d
N
e
t
w
o
rk
i
n
g
C
o
n
f
e
r
e
n
c
e
,
W
C
N
C
,
A
p
r
.
2
0
1
8
,
v
o
l
.
2
0
1
8
-
A
p
r
i
l
,
p
p
.
1
–
6
,
d
o
i
:
1
0
.
1
1
0
9
/
W
C
N
C
.
2
0
1
8
.
8
3
7
6
9
7
3
.
[
1
8
]
S
.
H
.
K
h
a
n
,
M
.
H
a
y
a
t
,
M
.
B
e
n
n
a
m
o
u
n
,
F
.
A
.
S
o
h
e
l
,
a
n
d
R
.
T
o
g
n
e
r
i
,
“
C
o
st
-
sen
s
i
t
i
v
e
l
e
a
r
n
i
n
g
o
f
d
e
e
p
f
e
a
t
u
r
e
r
e
p
r
e
se
n
t
a
t
i
o
n
s
f
r
o
m
i
mb
a
l
a
n
c
e
d
d
a
t
a
,
”
I
E
EE
T
r
a
n
sa
c
t
i
o
n
s
o
n
N
e
u
r
a
l
N
e
t
w
o
r
k
s
a
n
d
L
e
a
rn
i
n
g
S
y
st
e
m
s
,
v
o
l
.
2
9
,
n
o
.
8
,
p
p
.
3
5
7
3
–
3
5
8
7
,
A
u
g
.
2
0
1
8
,
d
o
i
:
1
0
.
1
1
0
9
/
TN
N
LS.
2
0
1
7
.
2
7
3
2
4
8
2
.
[
1
9
]
S
.
R
.
C
h
h
e
t
r
i
,
A
.
B
.
L
o
p
e
z
,
J.
W
a
n
,
a
n
d
M
.
A
.
A
l
F
a
r
u
q
u
e
,
“
G
A
N
-
S
e
c
:
g
e
n
e
r
a
t
i
v
e
a
d
v
e
r
sari
a
l
n
e
t
w
o
r
k
mo
d
e
l
i
n
g
f
o
r
t
h
e
se
c
u
r
i
t
y
a
n
a
l
y
si
s
o
f
c
y
b
e
r
-
p
h
y
si
c
a
l
p
r
o
d
u
c
t
i
o
n
sy
s
t
e
ms,”
i
n
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
2
0
1
9
D
e
si
g
n
,
A
u
t
o
m
a
t
i
o
n
a
n
d
T
e
s
t
i
n
Eu
ro
p
e
C
o
n
f
e
r
e
n
c
e
a
n
d
Ex
h
i
b
i
t
i
o
n
,
D
AT
E
2
0
1
9
,
2
0
1
9
,
p
p
.
7
7
0
–
7
7
5
,
d
o
i
:
1
0
.
2
3
9
1
9
/
D
A
TE
.
2
0
1
9
.
8
7
1
5
2
8
3
.
[
2
0
]
Y
.
Y
a
n
g
,
K
.
Z
h
e
n
g
,
C
.
W
u
,
a
n
d
Y
.
Y
a
n
g
,
“
I
mp
r
o
v
i
n
g
t
h
e
c
l
a
ssi
f
i
c
a
t
i
o
n
e
f
f
e
c
t
i
v
e
n
e
ss
o
f
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
b
y
u
s
i
n
g
i
mp
r
o
v
e
d
c
o
n
d
i
t
i
o
n
a
l
v
a
r
i
a
t
i
o
n
a
l
a
u
t
o
e
n
c
o
d
e
r
a
n
d
d
e
e
p
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
S
e
n
s
o
rs
(
S
w
i
t
z
e
rl
a
n
d
)
,
v
o
l
.
1
9
,
n
o
.
1
1
,
p
.
2
5
2
8
,
J
u
n
.
2
0
1
9
,
d
o
i
:
1
0
.
3
3
9
0
/
s
1
9
1
1
2
5
2
8
.
[
2
1
]
S
.
H
u
a
n
g
a
n
d
K
.
L
e
i
,
“
I
G
A
N
-
I
D
S
:
A
n
i
m
b
a
l
a
n
c
e
d
g
e
n
e
r
a
t
i
v
e
a
d
v
e
r
sari
a
l
n
e
t
w
o
r
k
t
o
w
a
r
d
s
i
n
t
r
u
s
i
o
n
d
e
t
e
c
t
i
o
n
sy
st
e
m
i
n
a
d
-
h
o
c
n
e
t
w
o
r
k
s,”
A
d
H
o
c
N
e
t
w
o
rks
,
v
o
l
.
1
0
5
,
A
u
g
.
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
d
h
o
c
.
2
0
2
0
.
1
0
2
1
7
7
.
[
2
2
]
Q
.
T
i
a
n
,
D
.
H
a
n
,
K
.
C
.
L
i
,
X
.
L
i
u
,
L.
D
u
a
n
,
a
n
d
A
.
C
a
s
t
i
g
l
i
o
n
e
,
“
A
n
i
n
t
r
u
s
i
o
n
d
e
t
e
c
t
i
o
n
a
p
p
r
o
a
c
h
b
a
se
d
o
n
i
m
p
r
o
v
e
d
d
e
e
p
b
e
l
i
e
f
n
e
t
w
o
r
k
,
”
A
p
p
l
i
e
d
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
5
0
,
n
o
.
1
0
,
p
p
.
3
1
6
2
–
3
1
7
8
,
M
a
y
2
0
2
0
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
0
4
8
9
-
020
-
0
1
6
9
4
-
4.
[
2
3
]
B
.
A
.
T
a
ma
,
M
.
C
o
m
u
z
z
i
,
a
n
d
K
.
H
.
R
h
e
e
,
“
TSE
-
I
D
S
:
A
Tw
o
-
S
t
a
g
e
C
l
a
ss
i
f
i
e
r
E
n
se
mb
l
e
f
o
r
I
n
t
e
l
l
i
g
e
n
t
A
n
o
m
a
l
y
-
B
a
s
e
d
I
n
t
r
u
si
o
n
D
e
t
e
c
t
i
o
n
S
y
st
e
m,
”
I
EE
E
A
c
c
e
ss
,
v
o
l
.
7
,
p
p
.
9
4
4
9
7
–
9
4
5
0
7
,
2
0
1
9
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
1
9
.
2
9
2
8
0
4
8
.
[
2
4
]
C
.
S
.
R
a
n
g
a
n
a
t
h
a
n
,
R
.
R
a
ma
n
,
K
.
K
.
S
u
t
a
r
i
a
,
R
.
A
V
a
r
ma
,
a
n
d
S
.
M
u
r
u
g
a
n
,
“
N
e
t
w
o
r
k
s
e
c
u
r
i
t
y
i
n
c
y
b
e
r
sp
a
c
e
u
s
i
n
g
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s
,
”
i
n
7
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
E
l
e
c
t
r
o
n
i
c
s
,
C
o
m
m
u
n
i
c
a
t
i
o
n
a
n
d
Ae
r
o
s
p
a
c
e
T
e
c
h
n
o
l
o
g
y
,
I
C
E
C
A
2
0
2
3
-
Pro
c
e
e
d
i
n
g
s
,
N
o
v
.
2
0
2
3
,
p
p
.
1
7
5
5
–
1
7
5
9
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
E
C
A
5
8
5
2
9
.
2
0
2
3
.
1
0
3
9
4
9
6
2
.
[
2
5
]
M
.
J.
K
u
mar
,
S
.
M
i
s
h
r
a
,
E.
G
.
R
e
d
d
y
,
M
.
R
a
j
m
o
h
a
n
,
S
.
M
u
r
u
g
a
n
,
a
n
d
N
.
A
.
V
i
g
n
e
s
h
,
“
B
a
y
e
s
i
a
n
d
e
c
i
s
i
o
n
mo
d
e
l
b
a
s
e
d
r
e
l
i
a
b
l
e
r
o
u
t
e
f
o
r
m
a
t
i
o
n
i
n
i
n
t
e
r
n
e
t
o
f
t
h
i
n
g
s,”
I
n
d
o
n
e
si
a
n
J
o
u
r
n
a
l
o
f
E
l
e
c
t
r
i
c
a
l
En
g
i
n
e
e
ri
n
g
a
n
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
3
4
,
n
o
.
3
,
p
p
.
1
6
6
5
–
1
6
7
3
,
Ju
n
.
2
0
2
4
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
e
c
s.
v
3
4
.
i
3
.
p
p
1
6
6
5
-
1
6
7
3
.
[
2
6
]
M
.
A
mr
u
e
t
a
l
.
,
“
N
e
t
w
o
r
k
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
sy
st
e
m
b
y
a
p
p
l
y
i
n
g
e
n
se
m
b
l
e
mo
d
e
l
f
o
r
smar
t
h
o
me
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
t
r
i
c
a
l
a
n
d
C
o
m
p
u
t
e
r
En
g
i
n
e
e
r
i
n
g
,
v
o
l
.
1
4
,
n
o
.
3
,
p
p
.
3
4
8
5
–
3
4
9
4
,
J
u
n
.
2
0
2
4
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
c
e
.
v
1
4
i
3
.
p
p
3
4
8
5
-
3
4
9
4
.
[
2
7
]
A
.
R
.
R
a
t
h
i
n
a
m,
B
.
S
.
V
a
t
h
a
n
i
,
A
.
K
o
ma
t
h
i
,
J
.
L
e
n
i
n
,
B
.
B
h
a
r
a
t
h
i
,
a
n
d
S
.
M
.
U
r
u
g
a
n
,
“
A
d
v
a
n
c
e
s
a
n
d
p
r
e
d
i
c
t
i
o
n
s
i
n
p
r
e
d
i
c
t
i
v
e
a
u
t
o
-
sc
a
l
i
n
g
a
n
d
ma
i
n
t
e
n
a
n
c
e
a
l
g
o
r
i
t
h
ms
f
o
r
c
l
o
u
d
c
o
m
p
u
t
i
n
g
,
”
2
n
d
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
A
u
t
o
m
a
t
i
o
n
,
C
o
m
p
u
t
i
n
g
a
n
d
Re
n
e
w
a
b
l
e
S
y
st
e
m
s,
I
C
A
C
RS
2
0
2
3
-
Pro
c
e
e
d
i
n
g
s
,
p
p
.
3
9
5
–
4
0
0
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
A
C
R
S
5
8
5
7
9
.
2
0
2
3
.
1
0
4
0
4
1
8
6
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
2
,
Ap
r
il
20
25
:
2
4
9
1
-
2
4
9
8
2498
[
2
8
]
P
.
R
a
d
h
a
k
r
i
sh
n
a
n
e
t
a
l
.
,
“
D
o
S
a
t
t
a
c
k
d
e
t
e
c
t
i
o
n
a
n
d
h
i
l
l
c
l
i
m
b
i
n
g
b
a
s
e
d
o
p
t
i
m
a
l
f
o
r
w
a
r
d
e
r
se
l
e
c
t
i
o
n
,
”
I
n
d
o
n
e
si
a
n
J
o
u
r
n
a
l
o
f
El
e
c
t
r
i
c
a
l
En
g
i
n
e
e
r
i
n
g
a
n
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
3
6
,
n
o
.
2
,
p
p
.
8
8
2
–
8
9
1
,
N
o
v
.
2
0
2
4
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
e
c
s
.
v
3
6
.
i
2
.
p
p
8
8
2
-
8
9
1
.
[
2
9
]
I
.
G
o
o
d
f
e
l
l
o
w
e
t
a
l
.
,
“
G
e
n
e
r
a
t
i
v
e
a
d
v
e
r
sari
a
l
n
e
t
w
o
r
k
s
,
”
C
o
m
m
u
n
i
c
a
t
i
o
n
s
o
f
t
h
e
AC
M
,
v
o
l
.
6
3
,
n
o
.
1
1
,
p
p
.
1
3
9
–
1
4
4
,
O
c
t
.
2
0
2
0
,
d
o
i
:
1
0
.
1
1
4
5
/
3
4
2
2
6
2
2
.
[
3
0
]
N
.
M
o
h
a
n
k
u
m
a
r
e
t
a
l
.
,
“
A
d
v
a
n
c
i
n
g
c
h
r
o
n
i
c
p
a
i
n
r
e
l
i
e
f
c
l
o
u
d
-
b
a
se
d
r
e
m
o
t
e
ma
n
a
g
e
m
e
n
t
w
i
t
h
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
i
n
h
e
a
l
t
h
c
a
r
e
,
”
I
n
d
o
n
e
si
a
n
J
o
u
rn
a
l
o
f
E
l
e
c
t
r
i
c
a
l
En
g
i
n
e
e
ri
n
g
a
n
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
3
7
,
n
o
.
2
,
p
p
.
1
0
4
2
–
1
0
5
2
,
2
0
2
5
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
e
c
s.
v
3
7
.
i
2
.
p
p
1
0
4
2
-
1
0
5
2
.
[
3
1
]
A
.
J.
M
y
l
e
s,
R
.
N
.
F
e
u
d
a
l
e
,
Y
.
Li
u
,
N
.
A
.
W
o
o
d
y
,
a
n
d
S
.
D
.
B
r
o
w
n
,
“
A
n
i
n
t
r
o
d
u
c
t
i
o
n
t
o
d
e
c
i
si
o
n
t
r
e
e
mo
d
e
l
i
n
g
,
”
J
o
u
rn
a
l
o
f
C
h
e
m
o
m
e
t
ri
c
s
,
v
o
l
.
1
8
,
n
o
.
6
,
p
p
.
2
7
5
–
2
8
5
,
J
u
n
.
2
0
0
4
,
d
o
i
:
1
0
.
1
0
0
2
/
c
e
m.
8
7
3
.
[
3
2
]
H
.
A
t
t
o
u
,
A
.
G
u
e
z
z
a
z
,
S
.
B
e
n
k
i
r
a
n
e
,
M
.
A
z
r
o
u
r
,
a
n
d
Y
.
F
a
r
h
a
o
u
i
,
“
C
l
o
u
d
-
b
a
s
e
d
i
n
t
r
u
s
i
o
n
d
e
t
e
c
t
i
o
n
a
p
p
r
o
a
c
h
u
s
i
n
g
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s,
”
B
i
g
D
a
t
a
M
i
n
i
n
g
a
n
d
An
a
l
y
t
i
c
s
,
v
o
l
.
6
,
n
o
.
3
,
p
p
.
3
1
1
–
3
2
0
,
S
e
p
.
2
0
2
3
,
d
o
i
:
1
0
.
2
6
5
9
9
/
B
D
M
A
.
2
0
2
2
.
9
0
2
0
0
3
8
.
[
3
3
]
M
.
R
.
S
u
d
h
a
e
t
a
l
.
,
“
P
r
e
d
i
c
t
i
v
e
m
o
d
e
l
i
n
g
f
o
r
h
e
a
l
t
h
c
a
r
e
w
o
r
k
e
r
w
e
l
l
-
b
e
i
n
g
w
i
t
h
c
l
o
u
d
c
o
m
p
u
t
i
n
g
a
n
d
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
f
o
r
st
r
e
ss
man
a
g
e
me
n
t
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
El
e
c
t
r
i
c
a
l
a
n
d
C
o
m
p
u
t
e
r
E
n
g
i
n
e
e
ri
n
g
,
v
o
l
.
1
5
,
n
o
.
1
,
p
p
.
1
2
1
8
–
1
2
2
8
,
2
0
2
5
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
c
e
.
v
1
5
i
1
.
p
p
1
2
1
8
-
1
2
2
8
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
G
n
a
n
a
m
J
e
b
a
Ro
slin
e
c
o
m
p
lete
d
M
a
ste
r
o
f
Co
m
p
u
ter
Ap
p
li
c
a
ti
o
n
s
fro
m
Un
iv
e
rsity
o
f
M
a
d
ra
s
a
n
d
P
u
rsu
i
n
g
P
h
.
D.
a
s
p
a
rt
ti
m
e
c
a
n
d
id
a
te
i
n
M
o
t
h
e
r
Tere
sa
W
o
m
e
n
’s
Un
iv
e
rsity
,
K
o
d
a
ik
a
n
a
l,
Tam
il
N
a
d
u
,
In
d
ia
u
n
d
e
r
t
h
e
g
u
i
d
e
sh
ip
o
f
Dr.
P
u
s
h
p
a
Ra
n
i.
Cu
rre
n
tl
y
wo
rk
i
n
g
a
s
a
Lec
tu
re
r
i
n
Un
i
v
e
rsity
o
f
Tec
h
n
o
lo
g
y
a
n
d
A
p
p
li
e
d
S
c
ien
c
e
s,
M
u
sc
a
t
.
Are
a
o
f
re
se
a
rc
h
in
c
lu
d
e
s
n
e
two
rk
se
c
u
ri
ty
a
n
d
a
rti
ficia
l
in
tell
ig
e
n
c
e
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
jeb
a
ro
slin
e
ies
@g
m
a
il
.
c
o
m
.
Pu
shp
a
Ra
n
i
c
o
m
p
lete
d
M
a
ste
r
o
f
Co
m
p
u
ter
A
p
p
li
c
a
ti
o
n
s
fro
m
Bh
a
ra
t
h
iar
Un
iv
e
rsity
,
In
d
ia
a
n
d
P
h
.
D
.
fr
o
m
M
a
d
u
ra
i
Ka
m
a
ra
j
Un
iv
e
rsity
,
I
n
d
ia.
Cu
rre
n
tl
y
wo
r
k
in
g
s
a
s
a
Dire
c
to
r,
De
p
a
rtme
n
t
o
f
Co
m
p
u
t
e
r
S
c
ien
c
e
,
M
o
th
e
r
Tere
sa
Wo
m
e
n
’s
Un
iv
e
rsity
,
Ko
d
a
i
k
a
n
a
l
,
Tam
il
Na
d
u
.
I
n
d
ia.
Re
se
a
rc
h
g
u
i
d
e
wh
o
h
a
s
p
u
b
li
sh
e
d
1
0
0
Jo
u
rn
a
ls,
1
0
0
c
o
n
fe
re
n
c
e
p
ro
c
e
e
d
in
g
s,
4
b
o
o
k
s,
5
p
ro
jec
t
s
a
n
d
m
a
n
y
fu
n
d
e
d
p
r
o
jec
ts.
E
x
p
e
rti
se
a
n
d
re
se
a
rc
h
a
re
a
in
c
lu
d
e
b
io
m
e
tri
c
s,
a
d
a
p
ti
v
e
lea
rn
in
g
sy
ste
m
,
in
fo
rm
a
ti
o
n
re
tri
e
v
a
l,
ima
g
e
p
ro
c
e
ss
in
g
,
c
lo
u
d
c
o
m
p
u
ti
n
g
,
n
e
two
r
k
se
c
u
rit
y
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
d
r
p
u
s
h
p
a
.
m
twu
@g
m
a
il
.
c
o
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.