I
nd
o
ne
s
ia
n J
o
urna
l o
f
E
lect
rica
l En
g
ineering
a
nd
Co
m
pu
t
er
Science
Vo
l.
36
,
No
.
2
,
N
o
v
em
b
e
r
20
24
,
p
p
.
1
28
4
~1
29
3
I
SS
N:
2
502
-
4
7
52
,
DOI
: 1
0
.
1
1
5
9
1
/ijee
cs
.v
36.
i
2
.
p
p
1
28
4
-
1
29
3
1284
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ee
cs
.
ia
esco
r
e.
co
m
Enha
ncing
net
wo
rk securi
ty using
unsu
perv
ised lea
r
ning
a
ppro
a
ch t
o
com
ba
t
zero
-
da
y
at
tack
Ra
j
a
k
u
m
a
r
P
er
um
a
l
1
,
T
a
m
il
a
ra
s
i K
a
rupp
ia
h
2
,
Uppi
l
ira
j
a
P
a
nn
ee
rselv
a
m
1
,
Venka
t
esa
n Anna
m
a
la
i
3
,
P
ra
bu
K
a
liy
a
perum
a
l
1
1
S
c
h
o
o
l
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
r
i
n
g
,
G
a
l
g
o
t
i
a
s U
n
i
v
e
r
si
t
y
,
G
r
e
a
t
e
r
N
o
i
d
a
,
I
n
d
i
a
2
D
e
p
a
r
t
me
n
t
o
f
I
n
f
o
r
mat
i
o
n
Te
c
h
n
o
l
o
g
y
,
P
a
n
i
mal
a
r
E
n
g
i
n
e
e
r
i
n
g
C
o
l
l
e
g
e
,
V
a
r
a
d
a
r
a
j
a
p
u
r
a
m
,
I
n
d
i
a
3
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
i
c
a
l
,
E
l
e
c
t
r
o
n
i
c
s
a
n
d
C
o
mm
u
n
i
c
a
t
i
o
n
E
n
g
i
n
e
e
r
i
n
g
,
G
a
l
g
o
t
i
a
s
U
n
i
v
e
r
si
t
y
,
G
r
e
a
t
e
r
N
o
i
d
a
,
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
Ap
r
11
,
2
0
2
4
R
ev
is
ed
Au
g
6
,
2
0
2
4
Acc
ep
ted
Au
g
15
,
2
0
2
4
M
a
c
h
in
e
lea
rn
i
n
g
(M
L)
a
n
d
a
d
v
a
n
c
e
d
n
e
u
ra
l
n
e
tw
o
rk
m
e
th
o
d
o
l
o
g
ies
l
ik
e
d
e
e
p
lea
rn
i
n
g
(DL)
tec
h
n
i
q
u
e
s
h
a
v
e
b
e
e
n
i
n
c
re
a
sin
g
ly
u
ti
li
z
e
d
in
d
e
v
e
lo
p
i
n
g
in
tru
si
o
n
d
e
tec
ti
o
n
sy
ste
m
s
(IDS)
.
Ho
we
v
e
r,
t
h
e
g
r
o
win
g
q
u
a
n
ti
ty
a
n
d
d
iv
e
rsit
y
o
f
c
y
b
e
r
-
a
tt
a
c
k
s
p
o
se
a
sig
n
ifi
c
a
n
t
c
h
a
ll
e
n
g
e
f
o
r
IDS
so
lu
ti
o
n
s
re
li
a
n
t
o
n
h
isto
rica
l
a
tt
a
c
k
sig
n
a
t
u
re
s.
Th
is
h
i
g
h
li
g
h
ts
t
h
e
in
d
u
str
y
'
s
n
e
e
d
f
o
r
re
sili
e
n
t
IDSs
th
a
t
c
a
n
i
d
e
n
ti
f
y
z
e
ro
-
d
a
y
a
tt
a
c
k
s.
C
u
rre
n
t
stu
d
ies
fo
c
u
sin
g
o
n
o
u
tl
ier
-
b
a
se
d
z
e
ro
-
d
a
y
d
e
tec
ti
o
n
a
re
h
in
d
e
re
d
b
y
e
lev
a
ted
fa
lse
-
n
e
g
a
ti
v
e
ra
tes
,
t
h
e
re
b
y
c
o
n
stra
in
i
n
g
th
e
i
r
p
ra
c
ti
c
a
l
e
ffica
c
y
.
Th
is
p
a
p
e
r
su
g
g
e
sts
u
ti
li
z
i
n
g
a
n
a
u
t
o
e
n
c
o
d
e
r
(AE)
a
p
p
ro
a
c
h
f
o
r
z
e
ro
-
da
y
a
tt
a
c
k
d
e
tec
ti
o
n
,
a
imi
n
g
to
a
c
h
iev
e
h
ig
h
re
c
a
ll
wh
il
e
m
in
imiz
in
g
fa
lse
n
e
g
a
ti
v
e
s.
E
v
a
lu
a
ti
o
n
is
c
o
n
d
u
c
ted
u
sin
g
we
ll
-
e
sta
b
li
s
h
e
d
IDS
d
a
tas
e
ts,
CICIDS
2
0
1
7
a
n
d
CS
ECICIDS
2
0
1
8
.
Th
e
m
o
d
e
l'
s
e
ffica
c
y
is
d
e
m
o
n
stra
ted
b
y
c
o
n
tr
a
stin
g
it
s
p
e
rfo
rm
a
n
c
e
with
t
h
a
t
o
f
a
one
-
c
l
a
ss
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
i
n
e
(OCS
VM).
Th
e
re
se
a
rc
h
u
n
d
e
rsc
o
re
s
th
e
OCSVM
'
s
c
a
p
a
b
il
it
y
i
n
d
isti
n
g
u
ish
in
g
z
e
ro
-
d
a
y
a
tt
a
c
k
s
fro
m
n
o
rm
a
l
b
e
h
a
v
i
o
r.
Lev
e
ra
g
i
n
g
th
e
e
n
c
o
d
i
n
g
-
d
e
c
o
d
i
n
g
c
a
p
a
b
il
it
ies
o
f
AE
s,
t
h
e
p
ro
p
o
se
d
m
o
d
e
l
e
x
h
i
b
it
s
p
r
o
m
isin
g
re
su
lt
s
in
d
e
tec
ti
n
g
c
o
m
p
le
x
z
e
ro
-
d
a
y
a
tt
a
c
k
s,
a
c
h
iev
in
g
a
c
c
u
ra
c
ies
ra
n
g
in
g
fro
m
9
3
%
to
9
9
%
a
c
ro
ss
d
a
tas
e
ts.
F
in
a
ll
y
,
th
e
p
a
p
e
r
d
isc
u
ss
e
s
th
e
b
a
lan
c
e
b
e
twe
e
n
re
c
a
ll
a
n
d
fa
ll
o
u
t,
o
ffe
rin
g
v
a
lu
a
b
le i
n
sig
h
ts i
n
to
m
o
d
e
l
p
e
rfo
rm
a
n
c
e
.
K
ey
w
o
r
d
s
:
Au
to
en
co
d
e
r
C
lo
u
d
s
ec
u
r
ity
Dee
p
lear
n
in
g
I
n
tr
u
s
io
n
d
etec
tio
n
s
y
s
tem
Su
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
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
:
R
ajak
u
m
ar
Per
u
m
al
Sch
o
o
l o
f
C
o
m
p
u
ter
Scien
ce
a
n
d
E
n
g
in
ee
in
g
,
Galg
o
tias
Un
iv
er
s
ity
Gr
ea
ter
No
id
a
,
Uttar
Pre
d
esh
,
2
0
1
3
1
0
,
I
n
d
ia
E
m
ail: r
ajk
u
m
ar
.
jcet@
g
m
ail.
c
o
m
1.
I
NT
RO
D
UCT
I
O
N
C
r
itical
to
ad
d
r
ess
in
g
th
e
e
x
p
o
n
e
n
tial
in
cr
ea
s
e
in
cy
b
er
-
attac
k
s
[
1
]
,
[
2
]
is
th
e
d
ep
l
o
y
m
en
t
o
f
in
tr
u
s
io
n
d
etec
tio
n
s
y
s
tem
s
(
I
DS)
ca
p
ab
le
o
f
id
en
tif
y
in
g
ze
r
o
-
d
a
y
cy
b
e
r
-
attac
k
s
.
Ma
ch
in
e
lear
n
in
g
(
ML
)
m
eth
o
d
o
l
o
g
ies
h
av
e
b
ee
n
wi
d
ely
em
p
lo
y
ed
i
n
d
ev
el
o
p
in
g
r
esil
ien
t
I
DS
[
3
]
,
[
4
]
s
o
lu
tio
n
s
.
As
d
ep
icted
i
n
Fig
u
r
e
1
,
I
DS
s
tan
d
s
as
a
v
ital
co
m
p
o
n
en
t
o
f
m
o
d
e
r
n
cy
b
er
s
ec
u
r
ity
[
5
]
,
f
o
r
tify
in
g
n
etwo
r
k
s
ag
ain
s
t
escalatin
g
d
ig
ital
r
is
k
s
.
Ho
wev
er
,
c
o
n
t
em
p
o
r
ar
y
I
DS
s
y
s
tem
s
ex
ce
l
in
d
etec
tin
g
k
n
o
w
n
attac
k
s
b
u
t
s
tr
u
g
g
le
with
id
en
tify
in
g
n
o
v
el
ze
r
o
-
d
a
y
att
ac
k
s
[
6
]
d
u
e
to
r
elian
ce
o
n
p
r
e
d
eter
m
in
ed
p
atter
n
s
a
n
d
s
ig
n
a
tu
r
es.
Ad
d
itio
n
ally
,
elev
ated
f
alse
-
p
o
s
itiv
e
r
ates
lim
it
th
eir
ef
f
ec
tiv
en
ess
an
d
p
r
ac
ticality
in
r
ea
l
-
wo
r
ld
d
ep
lo
y
m
en
ts
.
C
o
n
s
eq
u
en
tly
,
a
n
o
ta
b
le
p
o
r
ti
o
n
o
f
ze
r
o
-
d
a
y
attac
k
s
e
v
ad
e
d
etec
tio
n
,
e
x
ac
er
b
atin
g
th
eir
r
ep
er
cu
s
s
io
n
s
s
u
ch
as
d
en
ial
o
f
s
er
v
ice
an
d
th
ef
t
o
f
cu
s
to
m
er
in
f
o
r
m
atio
n
.
Acc
o
r
d
in
g
to
Ali
et
a
l.
[
7
]
,
a
ze
r
o
-
d
a
y
attac
k
is
a
tr
af
f
ic
p
atter
n
th
at
is
o
f
in
ter
est
an
d
o
f
ten
lack
s
m
atch
in
g
p
atter
n
s
in
m
alwa
r
e
o
r
n
etwo
r
k
attac
k
d
et
ec
tio
n
elem
en
ts
.
Sak
th
im
u
r
u
g
a
n
et
a
l.
[
8
]
d
el
v
e
in
to
th
e
r
ea
l
-
wo
r
l
d
r
am
if
icatio
n
s
o
f
ze
r
o
-
d
ay
attac
k
s
.
T
h
eir
r
esear
ch
is
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
E
n
h
a
n
cin
g
n
etw
o
r
k
s
ec
u
r
ity
u
s
in
g
u
n
s
u
p
ervis
ed
lea
r
n
in
g
a
p
p
r
o
a
ch
to
…
(
R
a
ja
k
u
ma
r
P
eru
ma
l
)
1285
ce
n
ter
ed
o
n
e
x
am
in
in
g
th
e
i
m
p
ac
t
an
d
p
r
e
v
alen
ce
o
f
th
e
s
e
attac
k
s
.
T
h
e
au
th
o
r
s
em
p
h
asize
th
at
ze
r
o
-
d
a
y
attac
k
s
ar
e
co
n
s
id
er
a
b
ly
m
o
r
e
wid
esp
r
ea
d
t
h
an
a
n
ticip
ated
.
T
h
eir
an
al
y
s
is
o
f
1
8
attac
k
s
r
ev
ea
ls
th
at
1
1
wer
e
p
r
ev
io
u
s
ly
u
n
id
en
tifie
d
.
T
h
ei
r
d
is
co
v
er
ies
in
d
icate
d
t
h
at
ze
r
o
-
d
a
y
attac
k
s
m
ay
en
d
u
r
e
o
v
er
a
s
u
b
s
tan
tial
p
er
io
d
,
with
an
a
v
er
ag
e
o
f
1
0
m
o
n
th
s
,
b
ef
o
r
e
d
etec
tio
n
.
T
h
is
p
r
o
lo
n
g
ed
p
r
esen
ce
allo
ws
th
em
to
co
m
p
r
o
m
is
e
s
y
s
tem
s
o
v
er
th
e
s
aid
p
er
io
d
.
Mo
r
eo
v
er
,
Z
o
p
p
i
et
a
l.
[
9
]
cite
a
s
tati
s
tical
in
v
esti
g
atio
n
r
ev
ea
lin
g
th
at
6
2
%
o
f
attac
k
s
ar
e
d
etec
ted
o
n
ly
af
ter
s
y
s
tem
s
h
av
e
b
ee
n
s
u
cc
ess
f
u
l
ly
co
m
p
r
o
m
is
ed
.
Fu
r
th
er
m
o
r
e,
th
e
co
u
n
t
o
f
ze
r
o
-
d
ay
attac
k
s
in
2
0
2
3
s
u
r
p
ass
es th
at
o
f
th
e
p
r
ec
ed
in
g
th
r
ee
y
ea
r
s
[
1
0
]
.
On
e
p
r
im
ar
y
r
esear
c
h
av
en
u
e
f
o
r
d
etec
tin
g
ze
r
o
-
d
a
y
attac
k
s
r
ev
o
lv
es
ar
o
u
n
d
id
e
n
tify
in
g
o
u
tlier
s
,
wh
ich
ar
e
in
s
tan
ce
s
th
at
d
ev
iate
f
r
o
m
n
o
r
m
al
n
etwo
r
k
tr
af
f
ic.
Ho
wev
e
r
,
t
h
e
m
ai
n
li
m
itatio
n
o
f
cu
r
r
en
t
d
etec
tio
n
tech
n
iq
u
es
b
ased
o
n
o
u
tlier
s
lies
in
t
h
eir
r
elativ
el
y
lo
w
ac
c
u
r
ac
y
r
ates,
s
tem
m
i
n
g
f
r
o
m
b
o
th
h
ig
h
f
alse
p
o
s
itiv
e
r
ate
(
FP
R
)
an
d
f
alse
n
eg
ativ
e
o
u
tco
m
es.
As
m
en
tio
n
ed
ea
r
lier
,
th
e
elev
ated
f
alse
-
n
eg
ativ
e
r
ates
(
FNR
)
m
ak
e
th
e
s
etu
p
v
u
l
n
er
ab
le
to
attac
k
s
[
1
1
]
,
wh
er
ea
s
th
e
elev
ated
FP
R
u
n
n
ec
ess
ar
ily
b
u
r
d
en
cy
b
er
s
ec
u
r
ity
o
p
er
atio
n
ce
n
te
r
s
,
with
o
n
ly
3
2
%
o
f
in
v
esti
g
ated
in
tr
u
s
io
n
s
p
r
o
v
in
g
to
b
e
g
en
u
in
e
.
E
s
s
a
an
d
B
h
ay
a
[
1
2
]
u
n
d
er
s
co
r
e
th
e
li
m
itatio
n
s
p
o
s
ed
b
y
f
alse
n
eg
ativ
es
in
I
DS
d
e
v
elo
p
m
e
n
t,
n
o
tab
ly
th
eir
ad
v
er
s
e
im
p
ac
t o
n
I
DS e
f
f
ec
tiv
en
ess
.
Fig
u
r
e
1
.
C
o
m
p
o
n
en
ts
o
f
in
tr
u
s
io
n
d
etec
tio
n
s
y
s
tem
Hair
ab
et
a
l.
[
1
3
]
in
tr
o
d
u
ce
d
a
s
y
s
tem
tailo
r
ed
to
id
e
n
tify
p
r
e
v
io
u
s
ly
u
n
s
ee
n
attac
k
s
with
in
in
ter
co
n
n
ec
ted
d
ev
ice
n
etwo
r
k
s
.
T
h
eir
ap
p
r
o
ac
h
r
elies
o
n
a
d
is
tr
ib
u
ted
d
etec
tio
n
m
ec
h
a
n
is
m
.
T
h
ey
u
tili
ze
d
ar
tific
ial
n
eu
r
al
n
etwo
r
k
s
f
o
r
s
u
p
er
v
is
ed
a
n
o
m
aly
-
b
ased
ML
,
attain
in
g
a
d
etec
tio
n
r
ate
o
f
9
4
%
o
n
th
e
NSL
-
KDD
d
ataset.
Ho
wev
er
,
th
e
s
tu
d
y
'
s
l
im
itatio
n
is
th
e
r
elati
v
ely
s
m
all
d
ataset
u
s
ed
f
o
r
s
u
p
er
v
is
ed
lear
n
in
g
.
Su
n
et
a
l.
[
1
4
]
i
n
tr
o
d
u
ce
d
a
B
ay
esian
p
r
o
b
ab
ilis
tic
ap
p
r
o
a
ch
with
th
e
o
b
jectiv
e
o
f
d
etec
tin
g
ze
r
o
-
d
a
y
attac
k
p
ath
way
s
.
T
h
ey
r
ep
r
esen
ted
at
tack
s
with
in
a
s
tr
u
ctu
r
e
r
esem
b
lin
g
a
g
r
ap
h
an
d
p
r
esen
ted
a
p
r
o
to
ty
p
e
aim
ed
at
id
en
tify
in
g
th
ese
attac
k
s
.
Du
o
n
g
an
d
Hai
[
1
5
]
p
r
o
p
o
s
ed
a
s
em
i
-
s
u
p
er
v
is
ed
ML
ap
p
r
o
ac
h
f
o
r
ze
r
o
-
d
ay
attac
k
d
etec
tio
n
u
tili
zin
g
th
e
NSL
-
KDD
d
ataset.
T
h
ey
u
tili
ze
d
a
K
-
m
ea
n
s
clu
s
ter
in
g
m
o
d
el
to
r
em
o
v
e
o
u
tlier
s
d
u
r
in
g
tr
ain
i
n
g
,
aim
in
g
f
o
r
a
t
r
ain
in
g
d
ataset
with
o
n
ly
b
e
n
ig
n
n
etwo
r
k
tr
af
f
ic
.
Desp
ite
f
e
atu
r
e
d
is
cr
ep
a
n
cy
,
th
eir
m
eth
o
d
ac
h
iev
e
d
a
9
1
% d
etec
tio
n
ac
cu
r
ac
y
.
T
h
is
p
ap
er
ad
v
o
ca
tes
lev
er
ag
in
g
u
n
s
u
p
er
v
is
ed
lear
n
in
g
ca
p
ab
ilit
ies
f
o
r
o
u
tlier
d
e
tectio
n
to
ef
f
ec
tiv
ely
id
en
tif
y
p
r
ev
i
o
u
s
ly
u
n
k
n
o
wn
attac
k
s
with
a
h
i
g
h
r
ate
o
f
ac
c
u
r
ac
y
.
T
h
e
g
o
a
l
en
tails
cr
af
tin
g
an
in
tr
u
s
io
n
d
etec
tio
n
m
o
d
el
t
h
a
t
is
ef
f
icien
t
at
id
en
tify
in
g
n
o
v
el
in
tr
u
s
io
n
s
an
d
ze
r
o
-
d
ay
attac
k
s
with
a
h
ig
h
ac
cu
r
ac
y
,
t
h
er
eb
y
en
s
u
r
i
n
g
a
l
o
w
f
allo
u
t
r
ate.
T
h
u
s
,
p
o
s
s
ess
in
g
a
s
tr
o
n
g
a
b
ilit
y
to
d
etec
t
ze
r
o
-
d
ay
attac
k
s
ca
n
allev
iate
th
e
ch
allen
g
es a
n
d
co
n
ce
r
n
s
ty
p
ically
lin
k
ed
with
n
o
v
el
attac
k
s
.
T
h
is
wo
r
k
o
f
f
er
s
th
r
ee
s
ig
n
if
ic
an
t c
o
n
tr
ib
u
tio
n
s
:
a)
I
n
tr
o
d
u
cin
g
a
n
d
d
e
p
lo
y
in
g
an
in
n
o
v
ativ
e
a
n
d
ef
f
icien
t
au
to
e
n
co
d
er
(
AE
)
f
r
a
m
ewo
r
k
f
o
r
z
er
o
-
d
a
y
attac
k
d
etec
tio
n
in
in
tr
u
s
io
n
d
etec
tio
n
s
y
s
tem
.
b)
C
r
ea
tin
g
an
an
o
m
aly
d
etec
tio
n
m
o
d
el
u
tili
zin
g
one
-
class
s
u
p
p
o
r
t v
ec
to
r
m
ac
h
i
n
e
(
OC
SVM
)
.
c)
Ass
es
s
in
g
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
OC
SVM
m
o
d
el,
u
tili
ze
d
as
a
f
o
u
n
d
atio
n
al
o
u
tlier
-
b
ased
d
etec
to
r
,
co
m
p
ar
ed
with
th
e
s
u
g
g
ested
AE
f
r
am
ewo
r
k
,
wh
ile
a
ls
o
co
m
p
a
r
in
g
th
e
p
r
o
p
o
s
ed
m
o
d
el
wit
h
b
en
ch
m
ar
k
ed
alg
o
r
ith
m
s
.
2.
T
H
E
CO
M
P
RE
H
E
NS
I
VE
T
H
E
O
RE
T
I
CA
L
B
ASI
S
2
.
1
.
Aut
o
enco
der
T
h
e
m
o
d
el
p
r
esen
ted
in
th
is
m
an
u
s
cr
ip
t
p
r
im
a
r
ily
lev
er
ag
e
s
th
e
f
ea
tu
r
es
o
f
AE
.
T
h
e
o
b
je
ctiv
e
is
f
o
r
th
e
AE
to
o
p
e
r
ate
as
a
lig
h
t
weig
h
t
an
o
m
aly
d
etec
to
r
[
1
6
]
,
th
u
s
f
ac
ilit
atin
g
ze
r
o
-
d
ay
att
ac
k
d
etec
tio
n
.
A
Es
wer
e
in
itially
in
tr
o
d
u
ce
d
b
y
R
u
m
elh
ar
t
et
a
l.
[
1
7
]
t
o
ad
d
r
ess
b
ac
k
p
r
o
p
ag
atio
n
ch
allen
g
es
in
an
u
n
s
u
p
er
v
is
ed
co
n
tex
t
b
y
e
m
p
lo
y
i
n
g
t
h
e
in
p
u
t
as
th
e
tar
g
et.
A
E
s
ar
e
class
if
ied
as
s
elf
-
s
u
p
er
v
is
ed
m
o
d
els
b
ec
au
s
e
t
h
eir
in
p
u
t
an
d
o
u
tp
u
t
ar
e
ess
en
tially
th
e
s
am
e.
B
r
u
n
n
er
et
a
l.
[
1
8
]
ex
p
l
ain
ed
th
at
an
AE
is
a
n
eu
r
al
n
etwo
r
k
v
ar
ia
n
t
th
at,
th
r
o
u
g
h
tr
ain
in
g
,
lear
n
s
to
p
r
o
d
u
ce
o
u
tp
u
t r
esem
b
lin
g
its
in
p
u
t.
Fig
u
r
e
2
d
ep
icts
th
e
b
asic s
tr
u
ctu
r
e
o
f
an
AE
.
T
h
e
s
etu
p
o
f
a
n
AE
an
d
its
laten
t
lay
er
co
u
n
t
v
ar
y
d
e
p
en
d
in
g
o
n
th
e
d
o
m
ai
n
an
d
u
s
ag
e
co
n
tex
t.
T
h
e
re
-
co
n
s
tr
u
ctio
n
er
r
o
r
is
co
m
p
u
ted
u
s
in
g
a
f
u
n
ctio
n
th
at
m
ea
s
u
r
es
th
e
d
is
p
ar
ity
b
etwe
en
th
e
in
p
u
t
an
d
th
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
36
,
No
.
2
,
No
v
em
b
er
20
24
:
1
28
4
-
1
29
3
1286
o
u
tp
u
t,
wh
er
e
th
e
o
u
tp
u
t
is
th
e
r
ec
o
n
s
tr
u
cted
in
p
u
t
[
1
9
]
.
M
ea
n
s
q
u
ar
e
er
r
o
r
(
MSE
)
an
d
m
ea
n
ab
s
o
lu
te
er
r
o
r
(
MA
E
)
ar
e
ty
p
ical
f
u
n
ctio
n
s
em
p
lo
y
e
d
to
co
m
p
u
te
th
e
r
ec
o
n
s
tr
u
ctio
n
er
r
o
r
[
2
0
]
.
B
y
tr
ain
in
g
th
e
A
E
to
m
in
im
ize
r
ec
o
n
s
tr
u
ctio
n
e
r
r
o
r
,
it
en
ab
les
p
r
ec
is
e
d
if
f
e
r
en
tia
tio
n
b
etwe
en
n
o
r
m
al
an
d
m
al
icio
u
s
tr
af
f
ic.
T
h
e
ar
ch
itectu
r
e
d
ep
icted
i
n
Fig
u
r
e
2
s
h
o
wca
s
es
th
e
ef
f
ec
tiv
en
ess
o
f
th
e
AE
m
o
d
el
in
ex
tr
ac
tin
g
cr
u
cia
l
in
f
o
r
m
atio
n
,
lead
in
g
to
r
o
b
u
s
t
class
if
icatio
n
r
esu
lts
[
2
1
]
.
Fig
u
r
e
2
.
Ar
c
h
itectu
r
e
o
f
an
AE
2
.
2
.
O
ne
-
cl
a
s
s
s
up
po
rt
v
ec
t
o
r
m
a
chine
S
u
p
p
o
r
t
v
e
c
t
o
r
m
a
c
h
i
n
e
(
S
V
M
)
i
s
r
e
n
o
w
n
e
d
a
s
a
d
e
e
p
l
y
r
o
o
t
e
d
a
n
d
r
o
b
u
s
t
s
u
p
e
r
v
i
s
e
d
ML
m
e
t
h
o
d
[
2
2
]
.
W
ith
th
e
p
r
o
v
id
ed
tr
ain
in
g
d
ata,
an
SVM
i
s
tr
ain
ed
to
d
e
f
in
e
a
h
y
p
er
p
la
n
e
in
a
m
u
lti
-
d
im
en
s
io
n
al
s
p
ac
e,
aim
in
g
to
ef
f
ec
tiv
ely
d
if
f
er
en
t
iate
b
etwe
en
th
e
class
es
[
2
3
]
.
I
n
th
e
ca
s
e
o
f
two
d
im
en
s
io
n
s
,
th
is
h
y
p
er
p
la
n
e
is
r
ep
r
esen
ted
as
a
lin
e,
wh
ile
i
n
th
r
ee
o
r
h
ig
h
er
d
im
en
s
io
n
s
,
it
m
an
if
ests
as
a
p
lan
e
[
2
4
]
.
W
h
en
d
ata
e
x
h
ib
its
non
-
lin
ea
r
s
ep
ar
a
b
ilit
y
,
a
k
er
n
el
is
u
tili
ze
d
to
tr
an
s
f
o
r
m
t
h
e
in
p
u
t
f
ea
tu
r
es
i
n
to
a
h
ig
h
er
-
d
im
en
s
io
n
al
s
p
ac
e,
en
ab
lin
g
b
etter
s
ep
a
r
atio
n
[
2
4
]
.
I
n
th
is
tr
an
s
f
o
r
m
ed
s
p
ac
e,
a
h
y
p
er
p
lan
e
ca
n
ac
c
u
r
ately
d
is
tin
g
u
is
h
b
etwe
en
th
e
class
es [
2
3
]
.
T
h
e
OC
SVM
d
if
f
er
s
f
r
o
m
its
s
u
p
er
v
is
ed
co
u
n
ter
p
ar
t
b
y
f
u
n
ctio
n
in
g
as
an
u
n
s
u
p
er
v
is
ed
v
ar
iatio
n
.
I
t
is
ch
ar
ac
ter
ized
as
a
s
y
s
tem
ad
ep
t
at
id
en
tify
in
g
"n
o
v
elty
"
[
2
2
]
.
T
h
e
g
o
al
o
f
th
e
OC
SVM
is
to
d
ef
in
e
a
h
y
p
er
p
lan
e
[
2
5
]
,
illu
s
tr
ated
in
Fig
u
r
e
3
,
th
at
ac
ts
as
a
b
o
u
n
d
ar
y
.
T
h
is
b
o
u
n
d
ar
y
ef
f
ec
tiv
e
ly
en
co
m
p
ass
es
all
tr
ain
in
g
d
ata
p
o
i
n
ts
wh
ile
ex
clu
d
in
g
a
n
y
o
th
er
s
.
T
h
e
o
u
tc
o
m
e
o
f
tr
ain
in
g
an
OC
SVM
is
o
b
s
er
v
ed
as
a
b
o
u
n
d
ar
y
with
a
s
p
h
er
ical
s
h
ap
e
[
2
2
]
.
Giv
e
n
its
r
ep
u
tat
io
n
as
a
h
ig
h
l
y
estab
lis
h
ed
o
u
tlier
-
b
ased
M
L
tech
n
iq
u
es,
th
e
OC
SVM
s
er
v
es
as
an
o
p
tim
al
b
en
ch
m
ar
k
f
o
r
ass
ess
in
g
th
e
ef
f
ec
ti
v
en
ess
o
f
a
d
ee
p
n
eu
r
al
n
etwo
r
k
(
DNN)
-
b
ased
AE
.
Fig
u
r
e
3
.
Vis
u
aliza
tio
n
o
f
th
e
OC
SVM
h
y
p
er
p
lan
e
3.
RE
S
E
ARCH
M
E
T
H
O
D
T
h
e
p
r
o
p
o
s
ed
d
etec
tio
n
m
o
d
el
en
co
m
p
ass
es
s
ev
er
al
s
tag
es.
I
t
b
eg
in
s
with
p
r
ep
r
o
ce
s
s
in
g
tr
af
f
ic
d
ata,
f
o
llo
wed
b
y
c
o
n
d
u
ctin
g
d
i
m
e
n
s
io
n
ality
r
ed
u
ctio
n
,
tr
ain
in
g
th
e
d
etec
tio
n
m
o
d
el,
id
en
tify
i
n
g
an
o
m
alies,
an
d
ev
alu
atin
g
its
p
er
f
o
r
m
a
n
ce
.
T
h
e
ar
ch
itectu
r
e
illu
s
tr
ated
in
Fig
u
r
e
4
elu
cid
ates
th
e
d
esig
n
o
f
th
e
attac
k
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
E
n
h
a
n
cin
g
n
etw
o
r
k
s
ec
u
r
ity
u
s
in
g
u
n
s
u
p
ervis
ed
lea
r
n
in
g
a
p
p
r
o
a
ch
to
…
(
R
a
ja
k
u
ma
r
P
eru
ma
l
)
1287
d
etec
tio
n
m
eth
o
d
.
I
n
itially
,
tr
a
f
f
ic
d
ata
u
n
d
er
g
o
n
o
r
m
aliza
tio
n
an
d
s
tan
d
ar
d
izatio
n
to
p
r
ep
a
r
e
n
o
r
m
al
n
etwo
r
k
tr
af
f
ic
d
atasets
.
Su
b
s
eq
u
en
tly
,
im
p
lem
en
tin
g
u
n
s
u
p
er
v
is
ed
p
r
e
-
clu
s
ter
in
g
with
AE
p
r
o
d
u
ce
s
d
is
tin
ct
s
u
b
s
ets
with
in
th
e
i
n
itial
laten
t
lay
e
r
,
h
ig
h
lig
h
tin
g
u
n
i
q
u
e
p
atter
n
s
wh
ile
r
e
d
u
cin
g
d
im
en
s
io
n
s
.
Du
r
in
g
th
e
tr
ai
n
in
g
p
h
ase,
th
ese
s
u
b
s
ets
ar
e
in
d
iv
id
u
ally
p
r
o
ce
s
s
ed
th
r
o
u
g
h
t
h
e
s
am
e
AE
in
laten
t
s
p
ac
es
2
t
o
4
,
a
n
d
OC
SVM.
T
h
e
"e
n
co
d
in
g
-
d
ec
o
d
in
g
"
p
r
o
ce
s
s
in
AE
ev
alu
ates
th
e
a
v
er
a
g
e
r
ec
o
n
s
tr
u
ctio
n
e
r
r
o
r
,
d
eter
m
in
in
g
th
e
d
etec
tio
n
th
r
esh
o
ld
.
I
n
th
e
a
n
o
m
aly
d
e
tectio
n
p
h
ase,
th
e
tr
ain
ed
m
o
d
el
u
tili
ze
s
test
d
ata
to
g
en
er
ate
r
ec
o
n
s
tr
u
cted
o
u
tp
u
ts
,
with
th
e
r
ec
o
n
s
tr
u
ctio
n
er
r
o
r
ca
lcu
lated
u
s
in
g
t
h
e
m
ea
n
s
q
u
a
r
ed
er
r
o
r
.
I
f
t
h
e
ca
lcu
lated
er
r
o
r
s
u
r
p
ass
es
th
e
p
r
ed
ef
in
ed
th
r
e
s
h
o
ld
,
th
e
tr
af
f
ic
is
id
en
tifie
d
as
ze
r
o
-
d
ay
attac
k
;
o
th
er
wis
e,
it
is
class
if
ied
as
n
o
r
m
al
[
1
8
]
.
OC
SVM
o
p
er
a
tes
with
in
a
h
i
g
h
-
d
im
en
s
io
n
al
k
er
n
el
s
p
ac
e
b
y
t
r
an
s
f
o
r
m
in
g
s
am
p
les
f
r
o
m
r
ed
u
ce
d
d
im
en
s
io
n
s
u
s
in
g
a
m
ap
p
in
g
f
u
n
ctio
n
.
Fig
u
r
e
4
.
Ar
c
h
itectu
r
e
o
f
p
r
o
p
o
s
ed
au
to
en
c
o
d
er
b
ased
I
DS
3
.
1
.
P
re
pro
ce
s
s
ing
da
t
a
a
nd
re
du
cing
dim
ens
io
na
lity
T
h
is
s
tu
d
y
m
ak
es
u
s
e
o
f
b
e
n
ch
m
ar
k
d
atasets
f
o
r
in
t
r
u
s
io
n
d
etec
tio
n
,
s
u
ch
as
C
I
C
I
DS2
0
1
7
an
d
C
SECI
C
I
DS2
0
1
8
.
Stan
d
ar
d
s
ca
lin
g
tech
n
iq
u
es
ar
e
em
p
l
o
y
ed
to
s
tan
d
ar
d
ize
n
u
m
e
r
ical
d
ata,
lin
ea
r
ly
tr
an
s
f
o
r
m
in
g
it.
T
h
e
n
o
r
m
alize
d
d
ata
is
s
u
b
s
eq
u
e
n
tly
tr
an
s
f
o
r
m
ed
to
a
r
a
n
g
e
with
a
m
ea
n
(
μ
)
o
f
0
a
n
d
a
s
tan
d
ar
d
d
e
v
iatio
n
(
σ
)
o
f
1
th
r
o
u
g
h
Z
-
s
co
r
e
n
o
r
m
aliza
tio
n
.
T
h
e
A
E
,
a
n
e
u
r
al
n
etwo
r
k
d
es
ig
n
ed
f
o
r
u
n
s
u
p
er
v
is
ed
lear
n
in
g
an
d
d
im
e
n
s
io
n
ality
r
e
d
u
ctio
n
,
h
a
n
d
les
in
p
u
t
n
o
r
m
al
tr
af
f
ic
d
ata
v
ia
its
en
co
d
er
n
etwo
r
k
.
E
v
e
r
y
lay
er
em
p
lo
y
weig
h
ts
an
d
b
iases
to
m
o
d
if
y
th
e
d
ata
in
to
a
r
ep
r
esen
tatio
n
o
f
lo
wer
d
im
en
s
io
n
ality
,
u
ltima
tely
r
e
s
u
ltin
g
in
th
e
laten
t
lay
er
g
en
er
atin
g
an
en
c
o
d
e
d
r
ep
r
esen
tatio
n
.
T
h
e
c
o
m
p
lete
n
etwo
r
k
,
in
clu
d
in
g
b
o
th
th
e
en
co
d
er
an
d
d
ec
o
d
er
,
ex
p
er
ien
ce
s
en
d
-
to
-
en
d
tr
ain
in
g
th
r
o
u
g
h
th
e
u
tili
za
tio
n
o
f
b
ac
k
p
r
o
p
ag
atio
n
an
d
g
r
ad
ien
t
d
escen
t
alg
o
r
ith
m
s
.
T
h
r
o
u
g
h
th
is
u
n
s
u
p
er
v
is
ed
lear
n
in
g
p
r
o
ce
s
s
,
a
co
m
p
r
ess
ed
r
ep
r
esen
ta
tio
n
o
f
th
e
i
n
p
u
t
d
ata
is
o
b
tain
ed
,
ef
f
ec
ti
v
ely
ca
p
tu
r
in
g
its
f
u
n
d
am
en
tal
f
ea
t
u
r
es.
Hav
in
g
b
e
en
t
r
ain
ed
o
n
u
n
lab
eled
d
ata,
th
e
A
E
'
s
en
co
d
ed
r
ep
r
esen
tatio
n
ac
ts
as
a
lo
wer
-
d
im
en
s
io
n
al
a
n
d
p
r
e
-
clu
s
ter
ed
v
ar
ian
t
o
f
th
e
o
r
ig
in
al
n
o
r
m
al
d
ata,
th
e
r
eb
y
en
ab
lin
g
ef
f
icie
n
t
an
aly
s
is
an
d
an
o
m
al
y
d
etec
tio
n
.
3
.
2
.
Det
ec
t
ing
zero
-
da
y
a
t
t
a
ck
s
wit
h
AE
T
h
e
A
E
,
a
n
eu
r
al
n
etwo
r
k
tr
ain
ed
with
o
u
t
s
u
p
e
r
v
is
io
n
,
e
x
tr
ac
ts
h
ier
ar
ch
ical
f
ea
tu
r
es
to
en
h
an
ce
class
if
icatio
n
.
B
y
lev
er
ag
in
g
th
e
laten
t
s
p
ac
e
1
with
in
t
h
e
A
E
f
o
r
d
im
e
n
s
io
n
ality
r
e
d
u
ctio
n
,
t
h
is
s
tu
d
y
co
n
s
tr
u
cts
an
A
E
m
o
d
el
t
o
ef
f
ec
tiv
ely
ca
p
tu
r
e
d
ata
f
ea
t
u
r
es
ac
r
o
s
s
clu
s
ter
in
g
s
u
b
s
ets.
T
r
ain
in
g
th
e
A
E
r
ed
u
ce
s
th
e
r
ec
o
n
s
tr
u
ctio
n
er
r
o
r
,
e
n
ab
lin
g
p
r
ec
is
e
d
is
cr
im
in
atio
n
b
etwe
en
n
o
r
m
al
a
n
d
m
a
licio
u
s
tr
af
f
ic.
T
h
e
ar
ch
itectu
r
e
d
e
p
icted
in
Fig
u
r
e
4
s
h
o
wca
s
es
th
e
ef
f
ec
tiv
en
e
s
s
o
f
th
e
A
E
m
o
d
el
in
e
x
tr
ac
tin
g
v
ital
in
f
o
r
m
atio
n
to
ac
h
iev
e
r
o
b
u
s
t c
lass
if
icatio
n
r
esu
lts
.
Du
r
in
g
th
e
tr
ain
in
g
p
h
ase,
th
e
A
E
m
o
d
el
p
r
o
ce
s
s
es
b
atch
es
o
f
n
o
r
m
al
tr
af
f
ic
d
ata
with
th
e
o
b
jectiv
e
o
f
m
in
im
izin
g
th
e
r
ec
o
n
s
tr
u
cti
o
n
er
r
o
r
.
T
h
e
d
etec
tio
n
th
r
esh
o
ld
d
y
n
a
m
ically
ad
ju
s
ts
b
ased
o
n
th
e
cu
m
u
lativ
e
lo
s
s
in
cu
r
r
ed
d
u
r
in
g
th
e
m
o
d
el's
tr
ain
in
g
p
r
o
ce
s
s
.
T
h
e
p
r
o
ce
s
s
iter
ates
f
o
r
a
d
esig
n
ated
n
u
m
b
e
r
o
f
ep
o
c
h
s
,
y
ield
in
g
th
e
tr
ai
n
ed
A
E
m
o
d
el
an
d
d
etec
tio
n
th
r
esh
o
ld
as
o
u
tp
u
ts
.
Du
r
in
g
th
e
d
etec
tio
n
p
h
ase,
th
e
tr
ain
ed
A
E
m
o
d
el
is
ass
e
s
s
ed
u
s
in
g
th
e
t
est
d
ataset
to
d
etec
t
an
o
m
alies.
T
h
e
r
ec
o
n
s
tr
u
ctio
n
er
r
o
r
f
o
r
ea
ch
d
ata
p
o
in
t
is
d
eter
m
in
ed
b
y
co
m
p
ar
i
n
g
th
e
o
r
ig
in
al
an
d
r
ec
o
n
s
tr
u
cted
d
at
a.
Data
p
o
in
ts
ex
ce
ed
in
g
th
e
d
etec
tio
n
th
r
esh
o
ld
in
r
ec
o
n
s
tr
u
ctio
n
er
r
o
r
ar
e
cl
ass
if
ied
as
ze
r
o
-
d
ay
attac
k
;
o
th
er
wis
e,
th
ey
ar
e
lab
eled
n
o
r
m
al.
T
h
e
o
u
tp
u
t
o
b
tain
ed
in
d
icate
s
th
e
class
if
icatio
n
o
f
ea
ch
d
ata
p
o
i
n
t
in
th
e
test
d
ataset
as
eith
er
n
o
r
m
al
o
r
ze
r
o
-
d
ay
attac
k
.
T
h
is
test
in
g
p
r
o
ce
s
s
is
cr
u
cial
f
o
r
ev
alu
atin
g
th
e
ef
f
ec
tiv
e
n
ess
o
f
th
e
tr
ain
ed
d
etec
tio
n
m
o
d
el
o
n
n
ew
d
ata,
ass
is
tin
g
in
th
e
id
en
tific
atio
n
o
f
p
o
ten
tial a
n
o
m
alies.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
36
,
No
.
2
,
No
v
em
b
er
20
24
:
1
28
4
-
1
29
3
1288
3
.
3
.
Det
ec
t
ing
zero
-
da
y
a
t
t
a
ck
s
wit
h O
CSVM
OC
SVM
s
tan
d
s
o
u
t
as
a
u
n
iq
u
e
ap
p
r
o
ac
h
in
o
n
e
-
class
class
if
icatio
n
,
f
o
cu
s
in
g
s
o
lely
o
n
a
s
in
g
u
lar
d
ata
class
[
2
2
]
,
u
n
lik
e
b
in
a
r
y
o
r
m
u
lti
-
class
m
eth
o
d
s
.
Af
ter
tr
ain
in
g
,
th
e
m
o
d
el
d
eter
m
i
n
es
wh
eth
er
a
n
ew
s
am
p
le
b
elo
n
g
s
to
th
e
d
esig
n
ated
tar
g
et
class
.
OC
SVM
o
p
er
ates
in
a
h
ig
h
-
d
im
e
n
s
io
n
al
k
er
n
el
s
p
ac
e
b
y
tr
an
s
f
o
r
m
in
g
s
am
p
les
u
s
in
g
a
m
ap
p
in
g
f
u
n
ctio
n
.
I
t
f
o
r
m
u
lat
es
a
h
y
p
er
p
la
n
e
to
s
ep
ar
ate
d
a
ta
f
r
o
m
th
e
o
r
ig
in
,
ad
d
r
ess
in
g
a
q
u
a
d
r
atic
p
r
o
b
le
m
with
a
weig
h
t
v
ec
to
r
a
n
d
m
ar
g
in
.
T
h
e
h
y
p
er
p
ar
a
m
eter
‘
v’
co
n
tr
o
ls
th
e
tr
ad
e
-
o
f
f
,
lim
itin
g
th
e
p
r
o
p
o
r
tio
n
o
f
an
o
m
alies.
T
h
e
in
clu
s
io
n
o
f
n
o
n
-
n
eg
ativ
e
s
lack
v
a
r
iab
les
allo
ws
f
o
r
s
o
f
t
m
ar
g
in
s
,
ad
ju
s
tin
g
s
en
s
itiv
ity
to
o
u
tlier
s
.
T
h
e
Gau
s
s
ian
k
er
n
el
f
u
n
ctio
n
e
f
f
ec
tiv
ely
r
e
p
r
es
en
ts
d
ata,
with
th
e
h
y
p
er
p
ar
am
eter
‘
γ
’
s
h
a
p
in
g
t
h
e
d
ec
is
io
n
b
o
u
n
d
a
r
y
.
Z
er
o
-
d
a
y
attac
k
s
ar
e
id
e
n
tifie
d
wh
en
n
ew
s
am
p
les
f
all
o
n
th
e
in
co
r
r
ec
t sid
e
o
f
th
e
h
y
p
e
r
p
lan
e,
r
esu
ltin
g
i
n
n
eg
ativ
e
v
a
lu
es in
th
e
d
ec
is
io
n
f
u
n
ctio
n
.
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
e
s
u
g
g
ested
m
eth
o
d
s
ar
e
e
v
alu
ated
u
s
in
g
b
e
n
ch
m
ar
k
in
t
r
u
s
io
n
d
etec
tio
n
d
atasets
,
with
a
s
p
ec
if
i
c
f
o
cu
s
o
n
th
e
to
p
f
o
u
r
attac
k
s
as
ze
r
o
-
d
a
y
o
cc
u
r
r
en
ce
s
wit
h
in
b
o
t
h
th
e
C
I
C
I
DS2
0
1
7
an
d
C
SECICIDS2
0
1
8
d
atasets
.
T
h
e
ev
alu
atio
n
in
v
o
lv
es
co
m
p
a
r
in
g
t
h
e
p
r
o
p
o
s
ed
m
o
d
el
ag
ain
s
t
s
tan
d
ar
d
an
o
m
aly
d
etec
tio
n
alg
o
r
ith
m
s
OC
SVM
o
n
u
n
-
cl
u
s
ter
ed
d
ata,
f
o
llo
wed
b
y
an
ass
ess
m
en
t
o
f
th
eir
p
er
f
o
r
m
an
ce
.
T
h
e
T
ab
le
1
p
r
esen
ts
th
e
d
ataset
u
tili
ze
d
an
d
th
e
q
u
an
tity
o
f
s
elec
ted
s
am
p
les f
o
r
th
is
r
esear
ch
.
T
ab
le
1
.
E
x
p
er
im
en
tal
s
am
p
le
s
C
I
C
I
D
S
2
0
1
7
C
S
EC
I
C
I
D
S
2
0
1
8
Tr
a
i
n
i
n
g
cl
a
ss
B
e
n
i
g
n
B
e
n
i
g
n
Te
st
i
n
g
c
l
a
ss
D
o
S
-
D
D
o
S
-
B
r
u
t
e
F
o
r
c
e
-
B
o
t
n
e
t
-
B
e
n
i
g
n
F
e
a
t
u
r
e
s
83
80
B
e
n
i
g
n
2
,
3
5
9
,
0
8
7
2
,
3
7
4
,
8
7
1
A
t
t
a
c
k
2
2
4
,
8
9
3
2
3
9
,
8
4
2
4
.
1
.
P
er
f
o
r
m
a
nce
m
e
t
rics
E
v
alu
atin
g
t
h
e
ef
f
icien
cy
o
f
an
in
tr
u
s
io
n
d
etec
tio
n
m
o
d
el
u
s
in
g
a
c
o
n
f
u
s
io
n
m
atr
i
x
[
2
6
]
in
v
o
lv
es
ass
es
s
in
g
ac
cu
r
ac
y
,
r
ec
all,
an
d
s
p
ec
if
icity
.
R
ec
all,
as
d
ep
ict
ed
in
(
1
)
,
r
ep
r
esen
ts
th
e
ac
cu
r
ate
id
en
tific
atio
n
o
f
p
o
s
itiv
e
in
s
tan
ce
s
.
Sp
ec
if
icity
,
o
u
tlin
e
d
in
(
2
)
,
g
au
g
es
th
e
s
y
s
tem
's
ca
p
ab
ilit
y
to
co
r
r
ec
tly
r
ec
o
g
n
ize
tr
u
e
n
eg
ativ
e
in
s
tan
ce
s
,
s
er
v
in
g
as a
co
m
p
r
e
h
en
s
iv
e
ev
alu
atio
n
m
etr
ic.
Acc
u
r
ac
y
,
as
in
d
icate
d
in
(
3
)
,
m
ea
s
u
r
es
th
e
p
r
ec
is
io
n
o
f
class
if
icatio
n
s
.
=
+
(
1
)
=
+
(2
)
=
+
+
(
3)
T
h
e
s
u
cc
ess
f
u
l
p
er
f
o
r
m
an
ce
o
f
th
e
m
o
d
el
h
in
g
es
o
n
cr
itical
f
ac
to
r
s
,
in
clu
d
in
g
th
e
s
u
itab
le
ar
ch
itectu
r
e
an
d
h
y
p
er
p
ar
am
e
ter
co
n
f
ig
u
r
atio
n
s
f
o
r
th
e
A
E
d
etec
tio
n
m
o
d
el.
E
x
te
n
s
iv
e
ex
p
er
im
en
tatio
n
was
u
n
d
er
tak
e
n
to
id
en
tify
th
e
m
o
s
t
ef
f
ec
tiv
e
h
y
p
er
p
ar
am
eter
co
n
f
ig
u
r
atio
n
s
.
T
h
e
A
E
p
r
o
p
o
s
ed
in
th
e
m
eth
o
d
co
m
p
r
is
es
an
in
p
u
t
lay
e
r
,
f
i
v
e
laten
t
s
p
ac
e,
an
d
an
o
u
t
p
u
t
la
y
er
.
Ad
d
itio
n
ally
,
t
h
e
A
E
in
te
g
r
ates
th
r
ee
h
i
d
d
en
lay
er
s
,
with
u
n
it
s
izes
cu
s
to
m
ized
ac
co
r
d
in
g
t
o
th
e
lo
s
s
f
u
n
ctio
n
.
T
h
is
ap
p
r
o
ac
h
g
u
ar
an
tees
an
o
p
tim
al
ad
ap
tatio
n
to
t
h
e
d
iv
e
r
s
e
f
ea
tu
r
e
d
im
en
s
io
n
s
with
in
th
e
t
r
ain
i
n
g
d
ata.
4
.
2
.
T
ra
ini
ng
a
nd
t
esting
t
he
det
ec
t
io
n m
o
del
T
h
is
ap
p
r
o
ac
h
en
tails
d
iv
id
in
g
th
e
d
ataset
in
to
tr
ain
in
g
(
7
0
%)
an
d
test
in
g
(
3
0
%)
s
u
b
s
ets
co
m
p
r
is
in
g
b
en
ig
n
d
ata.
All
attac
k
d
ata
is
u
tili
ze
d
f
o
r
ev
alu
atio
n
p
u
r
p
o
s
es.
T
h
e
m
o
d
el
u
n
d
er
g
o
es
tr
ain
in
g
with
ea
r
ly
s
to
p
p
in
g
cr
iter
ia
to
ascer
tain
th
e
id
ea
l
n
u
m
b
er
o
f
ep
o
c
h
s
.
R
ec
tifie
d
lin
ea
r
u
n
it
(
R
eL
U
)
is
em
p
lo
y
ed
as
th
e
ac
tiv
atio
n
f
u
n
ctio
n
,
wh
ile
Ad
a
m
o
p
tim
izatio
n
with
a
lear
n
in
g
r
ate
s
et
at
0
.
0
0
0
1
is
u
tili
ze
d
.
Dete
ctio
n
ac
cu
r
ac
y
is
ev
alu
ated
b
y
em
p
lo
y
in
g
M
SE
as
th
e
lo
s
s
f
u
n
ctio
n
.
Fig
u
r
e
5
illu
s
tr
ates
th
e
test
in
g
ac
cu
r
ac
y
o
f
th
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
ac
r
o
s
s
b
o
th
d
atasets
,
em
p
h
asizin
g
th
e
ef
f
icac
y
o
f
th
e
m
o
d
el.
Dete
ctio
n
m
o
d
els
co
n
s
tr
u
cted
u
s
in
g
b
o
th
No
r
m
al
an
d
Attack
d
ata
f
r
o
m
th
e
C
I
C
I
DS2
0
1
7
an
d
C
SECICIDS2
0
1
8
d
atasets
d
em
o
n
s
tr
ate
s
ig
n
if
ican
t
p
er
f
o
r
m
an
ce
ac
r
o
s
s
all
d
atasets
.
T
h
e
ex
p
e
r
im
en
tal
r
esu
lts
o
f
th
e
AE
m
o
d
el
,
av
er
a
g
ed
ac
r
o
s
s
ten
ex
p
er
im
e
n
ts
,
d
em
o
n
s
tr
ate
s
ig
n
if
ican
t
p
er
f
o
r
m
an
ce
ac
r
o
s
s
b
o
th
d
atasets
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
E
n
h
a
n
cin
g
n
etw
o
r
k
s
ec
u
r
ity
u
s
in
g
u
n
s
u
p
ervis
ed
lea
r
n
in
g
a
p
p
r
o
a
ch
to
…
(
R
a
ja
k
u
ma
r
P
eru
ma
l
)
1289
Fig
u
r
e
5
.
Pro
p
o
s
ed
AE
lo
s
s
f
u
n
ctio
n
4
.
3
.
Ass
ess
ing
t
he
AE
o
n t
he
CICI
D
S
2
0
1
7
a
nd
CSE
CIC
I
DS2
0
1
8
da
t
a
s
et
s
As
s
tated
ea
r
lier
,
th
e
A
E
is
tr
a
in
ed
u
s
in
g
7
0
%
o
f
th
e
b
e
n
ig
n
in
s
tan
ce
s
.
T
h
e
o
p
tim
ized
ar
ch
i
tectu
r
e
o
f
th
e
AE
f
o
r
th
e
C
I
C
I
DS2
0
1
7
an
d
C
SECICIDS2
0
1
8
d
atas
ets
co
n
s
is
ts
o
f
a
n
eu
r
al
n
etw
o
r
k
co
m
p
r
is
in
g
1
9
n
eu
r
o
n
s
in
laten
t sp
ac
e
1
an
d
5
,
alo
n
g
with
th
r
ee
h
id
d
en
la
y
er
s
co
n
s
is
tin
g
o
f
1
4
,
1
0
,
an
d
1
4
n
eu
r
o
n
s
.
T
h
e
i
d
ea
l
b
atch
s
ize
f
o
r
o
p
tim
al
p
er
f
o
r
m
an
ce
is
1
,
0
2
4
.
Alo
n
g
with
o
th
er
o
p
tim
ized
p
ar
am
eter
s
s
u
c
h
as
a
m
ea
n
s
q
u
ar
e
er
r
o
r
lo
s
s
(
MSeL
)
f
u
n
ctio
n
,
L
2
r
eg
u
lar
izatio
n
s
et
at
0
.
0
0
0
1
a
n
d
tr
ai
n
in
g
d
u
r
atio
n
s
o
f
1
1
4
ep
o
ch
s
f
o
r
C
I
C
I
DS2
0
1
7
an
d
1
2
6
e
p
o
ch
s
f
o
r
C
SECICIDS2
0
1
8
d
atasets
,
r
esp
ec
tiv
ely
.
T
ab
le
2
p
r
o
v
id
es
a
s
u
m
m
a
r
y
o
f
th
e
AE
ac
c
u
r
ac
y
f
o
r
all
class
e
s
in
th
e
C
I
C
I
DS
2
0
1
7
a
n
d
C
SECI
C
I
DS2
0
1
8
d
atasets
.
A
cc
u
r
ac
y
d
if
f
e
r
s
f
o
r
b
en
ig
n
an
d
attac
k
class
es.
Fo
r
b
en
i
g
n
,
it
r
ef
lects
s
p
ec
if
icity
,
in
d
icatin
g
in
s
tan
ce
s
co
r
r
ec
tly
class
if
ied
as
n
o
n
-
ze
r
o
-
d
ay
.
Fo
r
attac
k
s
,
it
r
ep
r
esen
ts
r
ec
all,
ca
p
tu
r
in
g
co
r
r
ec
tly
id
en
tifie
d
in
s
tan
ce
s
.
T
ab
le
2
.
A
E
ac
c
u
r
ac
y
f
o
r
all
c
lass
es in
th
e
C
I
C
I
DS2
0
1
7
an
d
C
SECI
C
I
DS2
0
1
8
d
atasets
with
d
if
f
er
e
n
t
th
r
esh
o
ld
v
al
u
es
C
I
C
I
D
S
2
0
1
7
C
S
EC
I
C
I
D
S
2
0
1
8
Th
r
e
s
h
o
l
d
/
a
t
t
a
c
k
0
.
1
5
0
.
1
0
.
0
5
0
.
1
5
0
.
1
0
.
0
5
B
e
n
i
g
n
0
.
9
6
5
9
0
.
9
5
2
9
0
.
9
1
4
7
0
.
9
8
8
9
0
.
9
5
9
8
0
.
9
2
4
8
D
o
S
0
.
9
8
4
2
0
.
9
8
5
8
0
.
9
8
8
7
0
.
9
8
7
7
0
.
9
8
7
8
0
.
9
8
9
8
DDOS
0
.
9
8
2
3
0
.
9
8
9
8
0
.
9
8
9
9
0
.
9
8
8
2
0
.
9
8
8
1
0
.
9
8
9
1
B
r
u
t
e
F
o
r
c
e
0
.
9
3
0
1
0
.
9
6
4
8
0
.
9
8
4
7
0
.
9
4
1
2
0
.
9
7
8
4
0
.
9
8
7
8
B
o
t
n
e
t
0
.
9
3
4
8
0
.
9
7
8
8
0
.
9
8
9
7
0
.
9
4
8
3
0
.
9
8
2
8
0
.
9
8
9
7
Fro
m
T
ab
le
2
,
th
e
ac
c
u
r
ac
y
f
o
r
b
en
i
g
n
i
n
s
tan
ce
s
is
9
6
.
5
9
%,
9
5
.
2
9
%,
an
d
9
1
.
4
7
%
with
th
r
esh
o
ld
s
o
f
0
.
1
5
,
0
.
1
,
an
d
0
.
0
5
,
r
esp
ec
tiv
e
ly
.
Dec
r
ea
s
in
g
th
e
th
r
esh
o
ld
g
en
er
ally
in
cr
ea
s
es
ac
cu
r
ac
y
a
cr
o
s
s
all
class
es
in
b
o
th
d
atasets
.
C
SECI
C
I
DS2
0
1
8
an
d
C
I
C
I
DS2
0
1
7
c
o
n
s
is
ten
tly
s
h
o
ws
h
ig
h
er
ac
cu
r
ac
y
at
all
th
r
esh
o
ld
s
.
T
h
e
AE
ef
f
ec
tiv
ely
id
e
n
tifie
s
b
en
i
g
n
in
s
tan
ce
s
,
with
ac
cu
r
ac
y
r
an
g
in
g
f
r
o
m
9
1
.
4
7
%
to
9
6
.
5
9
%
in
C
I
C
I
DS2
0
1
7
an
d
9
2
.
4
8
%
to
9
8
.
8
9
%
in
C
SECICID
S2
0
1
8
.
A
d
d
itio
n
ally
,
attac
k
s
s
u
ch
as
d
en
ial
-
of
-
s
er
v
ice
(
Do
S
)
,
d
is
tr
ib
u
ted
Do
S
(
DD
o
S
)
,
B
r
u
t
e
Fo
r
ce
,
an
d
B
o
tn
et
ar
e
ac
c
u
r
ately
d
etec
ted
,
with
ac
cu
r
ac
y
co
n
s
is
ten
tly
ab
o
v
e
9
0
%
f
o
r
m
o
s
t
th
r
esh
o
ld
s
an
d
d
atasets
.
L
o
wer
th
r
esh
o
ld
s
(
0
.
0
5
)
ten
d
to
y
ield
h
ig
h
e
r
ac
cu
r
ac
y
f
o
r
m
o
s
t
class
es,
in
d
icatin
g
a
p
r
e
f
er
en
ce
f
o
r
h
i
g
h
er
s
en
s
itiv
ity
at
th
e
ex
p
e
n
s
e
o
f
s
p
ec
if
icity
.
Ov
er
all,
th
e
AE
ex
h
ib
its
r
o
b
u
s
t
p
er
f
o
r
m
an
ce
in
d
etec
tin
g
b
o
t
h
b
en
ig
n
a
n
d
attac
k
in
s
tan
ce
s
ac
r
o
s
s
v
ar
io
u
s
th
r
esh
o
ld
s
an
d
d
atasets
,
with
g
en
er
ally
h
i
g
h
er
ac
c
u
r
ac
y
o
b
s
er
v
ed
in
t
h
e
b
o
t
h
C
I
C
I
DS2
0
1
7
an
d
C
SECICIDS2
0
1
8
d
ataset.
4
.
4
.
Ass
ess
ing
t
he
O
CSV
M
o
n t
he
CICI
D
S
2
0
1
7
a
nd
CSE
CIC
I
DS2
0
1
8
da
t
a
s
et
s
T
h
e
OC
SVM
i
s
tr
ain
ed
ex
clu
s
iv
ely
with
n
o
n
-
m
alicio
u
s
o
cc
u
r
an
ce
s
.
T
o
f
ac
ilit
ate
th
is
tr
ain
in
g
,
a
s
p
ec
if
ic
'
ν
'
v
alu
e
was
d
esig
n
a
ted
f
o
r
th
e
OC
SVM.
T
h
e
'
ν
'
p
ar
am
eter
r
an
g
es
b
etwe
en
0
a
n
d
1
,
d
e
f
in
in
g
b
o
th
th
e
m
in
im
u
m
an
d
m
ax
im
u
m
th
r
esh
o
ld
s
f
o
r
th
e
n
u
m
b
er
o
f
e
x
am
p
les
id
en
tifie
d
as
s
u
p
p
o
r
t
v
ec
to
r
s
,
in
clu
d
in
g
th
o
s
e
th
at
lie
o
n
th
e
wr
o
n
g
s
id
e
o
f
th
e
h
y
p
er
p
lan
e
.
T
h
e
d
ef
au
lt
v
alu
e
f
o
r
'
ν
'
is
s
et
to
0
.
5
,
im
p
ly
i
n
g
th
at
it
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
36
,
No
.
2
,
No
v
em
b
er
20
24
:
1
28
4
-
1
29
3
1290
in
co
r
p
o
r
ates 5
0
% o
f
th
e
tr
ain
in
g
s
am
p
les with
in
th
e
h
y
p
er
p
l
an
e.
Ho
wev
er
,
f
o
r
th
e
p
u
r
p
o
s
e
o
f
th
is
ex
p
er
im
en
t,
s
ev
er
al
'
ν
'
v
alu
es
wer
e
s
e
lec
ted
(
0
.
2
,
0
.
1
5
,
0
.
1
)
.
T
h
ese
v
alu
es
wer
e
ch
o
s
en
to
ev
alu
ate
an
d
g
au
g
e
th
e
p
er
f
o
r
m
an
ce
o
f
t
h
e
AE
.
Sev
en
ty
p
er
ce
n
t
o
f
th
e
n
o
n
-
attac
k
s
a
m
p
les
ar
e
u
tili
ze
d
to
tr
ain
th
e
OC
SVM
m
o
d
el.
I
n
co
n
tr
ast
to
th
e
AE
m
o
d
el,
t
h
e
ev
alu
atio
n
o
f
a
n
OC
SVM
tr
ain
ed
m
o
d
el
r
esu
lts
in
a
b
i
n
ar
y
v
alu
e
0
an
d
1
.
T
h
e
o
u
tp
u
t
in
d
icate
s
wh
eth
er
an
o
cc
u
r
an
ce
is
class
if
ied
w
it
h
in
th
e
SVM
'
s
f
it
ted
clas
s
.
C
o
n
s
eq
u
en
tly
,
th
e
ass
es
s
m
en
t o
f
ea
ch
attac
k
r
elies o
n
th
e
n
u
m
b
er
o
f
in
s
tan
ce
s
p
r
ed
icted
with
a
'
0
'
o
u
tp
u
t f
r
o
m
th
e
OC
SVM.
Fig
u
r
e
6
p
r
esen
ts
a
s
u
m
m
ar
y
o
f
th
e
r
esu
lts
o
b
tain
ed
f
r
o
m
th
e
OC
SVM.
Up
o
n
ex
a
m
in
in
g
th
e
OC
SVM
r
e
s
u
lts
,
two
o
b
s
er
v
atio
n
s
b
ec
o
m
e
ap
p
a
r
en
t.
Firstl
y
,
alter
in
g
th
e
'
ν
'
v
alu
e
d
o
es
n
o
t
n
o
tab
ly
im
p
ac
t
d
etec
tio
n
ac
cu
r
ac
y
.
Seco
n
d
ly
,
th
e
class
es
th
at
ex
h
ib
it
h
ig
h
a
cc
u
r
ac
y
in
th
e
A
E
r
esu
lts
in
T
ab
le
2
ar
e
s
im
ilar
ly
id
en
tifie
d
b
y
th
e
OC
SVM.
N
ev
er
th
eless
,
it
d
o
es
n
o
t
s
u
cc
ee
d
in
id
en
tify
in
g
two
ad
d
itio
n
al
ca
teg
o
r
ies:
th
o
s
e
ex
h
ib
itin
g
e
n
h
an
ce
d
ac
cu
r
ac
y
at
lo
wer
th
r
esh
o
ld
s
,
an
d
th
o
s
e
d
em
o
n
s
tr
atin
g
p
o
o
r
d
etec
ti
o
n
ac
cu
r
ac
y
.
T
h
is
ar
is
es
f
r
o
m
th
e
co
n
s
tr
ain
ts
o
f
th
e
OC
SV
M
alg
o
r
ith
m
,
wh
ich
aim
s
to
estab
li
s
h
a
s
p
h
er
ical
h
y
p
er
p
lan
e
to
d
is
tin
g
u
is
h
th
e
b
en
ig
n
class
f
r
o
m
o
th
er
s
.
No
n
et
h
eless
,
class
es
f
allin
g
with
in
th
is
h
y
p
er
p
la
n
e
ar
e
co
n
s
is
ten
tly
class
if
ied
as
b
en
ig
n
.
T
h
is
c
an
b
e
f
u
r
th
er
v
is
u
alize
d
in
Fig
u
r
e
7
.
OC
SVM
is
p
ar
ticu
lar
ly
ef
f
ec
tiv
e
in
id
en
tify
in
g
r
ec
o
g
n
izab
le
ze
r
o
-
d
ay
attac
k
s
.
Yet,
AE
ex
ce
l
in
h
an
d
lin
g
in
tr
icate
ze
r
o
-
d
a
y
attac
k
s
,
b
o
asti
n
g
n
o
tab
ly
s
u
p
e
r
io
r
p
er
f
o
r
m
an
ce
r
an
k
in
g
s
.
Fig
u
r
e
6
.
OC
SVM
ac
cu
r
ac
y
e
v
alau
tio
n
o
n
C
I
C
I
DS2
0
1
7
Fig
u
r
e
7
.
OC
SVM
ac
cu
r
ac
y
e
v
alau
tio
n
o
n
C
SECI
C
I
DS2
0
1
8
4
.
5
.
P
er
f
o
r
m
a
nce
co
m
pa
riso
n
Fig
u
r
e
8
illu
s
tr
ates
AE
an
d
OC
SVM
a
lg
o
r
ith
m
p
er
f
o
r
m
a
n
ce
ac
r
o
s
s
v
ar
ied
h
y
p
er
p
ar
a
m
eter
s
o
n
C
I
C
I
DS2
0
1
7
an
d
C
SECIC
I
DS2
0
1
8
d
atasets
.
AE
's
d
etec
tio
n
ac
cu
r
ac
y
im
p
r
o
v
es
with
d
ec
r
ea
s
in
g
h
y
p
er
p
ar
am
eter
s
,
co
n
s
is
ten
tly
o
u
tp
er
f
o
r
m
in
g
OC
SVM.
A
E
ac
h
iev
es
h
ig
h
est
ac
cu
r
ac
y
o
n
C
I
C
I
DS2
0
1
7
(
9
8
.
8
8
%)
an
d
C
SECICIDS2
0
1
8
(
9
8
.
9
1
%)
with
a
h
y
p
er
p
ar
a
m
eter
o
f
0
.
0
5
.
OC
SVM's
ac
cu
r
ac
y
also
im
p
r
o
v
es
with
r
ed
u
ce
d
h
y
p
er
p
a
r
am
eter
s
b
u
t
r
em
ain
s
s
lig
h
tly
in
f
er
io
r
to
AE
,
n
o
tab
ly
o
n
C
I
C
I
DS2
0
1
7
.
I
ts
h
ig
h
est
ac
cu
r
ac
ies
ar
e
9
0
.
7
4
%
o
n
C
I
C
I
DS2
0
1
7
an
d
9
0
.
7
6
%
o
n
C
SECICIDS2
0
1
8
with
a
h
y
p
er
p
ar
am
eter
o
f
0
.
1
.
Ov
er
all,
AE
ex
h
ib
its
s
u
p
er
i
o
r
p
er
f
o
r
m
an
ce
,
with
d
ec
r
ea
s
i
n
g
h
y
p
er
p
a
r
am
eter
s
co
r
r
elatin
g
with
en
h
an
ce
d
ac
cu
r
ac
y
,
s
u
g
g
esti
n
g
C
SECIC
I
DS2
0
1
8
'
s
p
o
ten
tial f
o
r
ze
r
o
-
d
ay
d
etec
tio
n
.
T
h
e
r
a
d
ar
c
h
ar
t
i
n
Fig
u
r
e
9
co
m
p
ar
es
th
e
ac
cu
r
ac
y
o
f
v
ar
io
u
s
an
o
m
aly
d
etec
tio
n
alg
o
r
ith
m
s
f
o
r
ze
r
o
-
d
ay
attac
k
d
etec
tio
n
.
A
E
lead
s
with
9
8
.
9
0
%
ac
cu
r
ac
y
,
f
o
ll
o
wed
clo
s
ely
b
y
OC
SVM
at
9
0
.
7
5
%.
K
-
n
ea
r
est
n
eig
h
b
o
r
s
(
KNN)
[
2
6
]
tr
ails
b
eh
in
d
at
8
2
.
1
7
%,
i
n
d
icatin
g
wea
k
er
p
er
f
o
r
m
a
n
ce
co
m
p
ar
ed
t
o
A
E
an
d
OC
SVM.
Den
s
ity
-
b
ased
s
p
atial
clu
s
ter
in
g
o
f
ap
p
licatio
n
s
with
n
o
is
e
(
DB
SC
AN)
[
2
7
]
ac
h
iev
es
8
6
.
5
4
%
ac
cu
r
ac
y
,
f
allin
g
b
etwe
en
A
E
an
d
OC
SVM.
Ov
er
all,
A
E
d
em
o
n
s
tr
ates
s
u
p
er
io
r
ac
c
u
r
ac
y
a
m
o
n
g
th
e
ev
alu
ated
alg
o
r
ith
m
s
,
with
OC
SVM,
DB
S
C
AN,
an
d
KNN
f
o
llo
win
g
s
u
it.
T
h
e
co
m
b
i
n
ed
r
esu
lts
an
d
d
i
s
cu
s
s
io
n
u
n
d
er
s
co
r
e
th
e
s
tr
o
n
g
p
er
f
o
r
m
an
ce
o
f
o
u
r
d
etec
tio
n
m
o
d
el,
with
th
e
A
E
co
n
s
is
ten
tly
ac
h
iev
in
g
h
i
g
h
ac
c
u
r
ac
y
o
n
th
e
C
I
C
I
DS2
0
1
7
an
d
C
SECICIDS
2
0
1
8
d
atasets
.
Key
f
in
d
in
g
s
in
d
icate
th
at
o
u
r
m
o
d
el
ef
f
ec
tiv
ely
d
etec
ts
b
o
th
b
en
ig
n
an
d
attac
k
in
s
tan
ce
s
,
m
ain
tain
in
g
ac
cu
r
ac
ies
co
n
s
is
ten
tly
ab
o
v
e
9
0
% a
cr
o
s
s
m
o
s
t th
r
esh
o
ld
s
.
Fig
u
r
es 5
an
d
6
v
is
u
ally
s
u
p
p
o
r
t th
ese
r
esu
lts
,
p
r
o
v
id
in
g
clea
r
ev
id
en
ce
o
f
th
e
m
o
d
el'
s
ef
f
ec
tiv
en
ess
.
Ou
r
m
o
d
el
d
e
m
o
n
s
tr
ates
s
u
p
e
r
io
r
ac
cu
r
ac
y
co
m
p
ar
ed
to
p
r
ev
io
u
s
s
tu
d
ies,
esp
ec
ially
in
id
en
tify
in
g
ze
r
o
-
d
a
y
attac
k
s
.
Alth
o
u
g
h
t
h
e
A
E
o
u
tp
er
f
o
r
m
s
th
e
OC
SVM,
o
u
r
s
tu
d
y
also
h
ig
h
lig
h
ts
s
o
m
e
lim
itatio
n
s
,
s
u
c
h
as
th
e
OC
SVM
's
d
if
f
icu
lty
in
d
etec
tin
g
s
p
ec
if
ic
attac
k
ca
teg
o
r
ies.
Su
r
p
r
is
in
g
ly
,
lo
wer
th
r
e
s
h
o
ld
s
led
to
h
i
g
h
er
ac
cu
r
ac
y
,
in
d
icatin
g
a
tr
a
d
e
-
o
f
f
b
etwe
en
s
en
s
itiv
ity
an
d
s
p
ec
i
f
icity
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
E
n
h
a
n
cin
g
n
etw
o
r
k
s
ec
u
r
ity
u
s
in
g
u
n
s
u
p
ervis
ed
lea
r
n
in
g
a
p
p
r
o
a
ch
to
…
(
R
a
ja
k
u
ma
r
P
eru
ma
l
)
1291
T
h
e
m
ain
o
b
jectiv
e
o
f
th
is
s
tu
d
y
was
to
cr
ea
te
an
ef
f
ec
tiv
e
an
o
m
aly
d
etec
tio
n
m
o
d
el.
Ou
r
f
in
d
in
g
s
h
ig
h
lig
h
t
th
e
A
E
'
s
cr
u
cial
r
o
le
in
im
p
r
o
v
in
g
d
etec
tio
n
ac
cu
r
ac
y
,
wh
ich
h
as
im
p
o
r
tan
t
im
p
licatio
n
s
f
o
r
f
u
tu
r
e
cy
b
er
s
ec
u
r
ity
ap
p
licatio
n
s
.
Fu
tu
r
e
r
esear
ch
s
h
o
u
l
d
f
o
c
u
s
o
n
f
u
r
t
h
er
o
p
tim
izin
g
h
y
p
er
p
a
r
am
eter
s
an
d
ad
d
r
ess
in
g
th
e
id
e
n
tifie
d
lim
itatio
n
s
,
esp
ec
ially
f
o
r
m
o
r
e
co
m
p
lex
attac
k
s
ce
n
ar
io
s
.
Fig
u
r
e
8
.
Per
f
o
r
m
an
c
e
co
m
p
ar
is
o
n
o
f
AE
with
OC
SVM
Fig
u
r
e
9
.
Per
f
o
r
m
an
c
e
co
m
p
ar
is
o
n
with
B
en
ch
m
ar
k
alg
o
r
ith
m
s
5.
CO
NCLU
SI
O
N
T
h
e
m
an
u
s
cr
ip
t
in
tr
o
d
u
ce
s
a
n
in
n
o
v
ativ
e
ap
p
r
o
ac
h
to
d
ete
ct
ze
r
o
-
d
a
y
c
y
b
er
-
attac
k
s
u
s
in
g
o
u
tlier
-
b
ased
tech
n
iq
u
es,
aim
in
g
to
a
d
d
r
ess
lim
itatio
n
s
o
f
ex
is
tin
g
I
DS
.
T
h
e
p
r
o
p
o
s
ed
m
eth
o
d
r
ev
o
lv
es
ar
o
u
n
d
an
in
tellig
en
t
I
DS
m
o
d
el
em
p
lo
y
in
g
AE
ar
ch
itectu
r
e,
ca
p
italizin
g
o
n
its
en
co
d
in
g
-
d
ec
o
d
i
n
g
c
ap
ab
ilit
ies.
Acr
o
s
s
th
e
C
I
C
I
DS2
0
1
7
an
d
C
SECICIDS2
0
1
8
d
atasets
,
th
e
AE
m
o
d
el
e
x
h
ib
its
r
o
b
u
s
t
d
etec
tio
n
ac
c
u
r
ac
y
,
r
a
n
g
in
g
f
r
o
m
9
8
.
4
7
%
to
9
8
.
9
9
%.
C
o
m
p
ar
ativ
e
an
aly
s
is
ag
ain
s
t
OC
SVM
u
n
d
er
s
co
r
es
th
e
s
u
p
er
io
r
ac
cu
r
ac
y
o
f
th
e
p
r
o
p
o
s
ed
m
o
d
el,
ac
c
o
m
p
a
n
ied
b
y
lo
w
f
alse
-
p
o
s
itiv
e
r
ates.
Fu
tu
r
e
en
d
ea
v
o
r
s
e
n
tail
ass
ess
in
g
th
ese
m
o
d
els
with
s
p
ec
ialized
n
etwo
r
k
I
DS
d
atasets
,
s
u
ch
as
th
o
s
e
p
er
tai
n
in
g
t
o
I
o
T
a
n
d
c
r
itical
I
n
f
r
astru
ctu
r
e
n
etwo
r
k
s
.
T
h
is
iter
ativ
e
p
r
o
ce
s
s
aim
s
to
ad
ap
t
an
d
r
ef
in
e
t
h
e
ap
p
r
o
ac
h
,
f
o
s
ter
in
g
d
ee
p
er
in
s
ig
h
ts
in
to
ze
r
o
-
d
ay
attac
k
d
etec
tio
n
an
d
f
ac
ilit
atin
g
ex
p
l
o
r
atio
n
o
f
ad
d
itio
n
al
ML
tech
n
iq
u
es to
en
h
an
ce
ef
f
icac
y
.
RE
F
E
R
E
NC
E
S
[
1
]
M
.
S
a
r
h
a
n
,
S
.
L
a
y
e
g
h
y
,
M
.
G
a
l
l
a
g
h
e
r
,
a
n
d
M
.
P
o
r
t
ma
n
n
,
“
F
r
o
m
z
e
r
o
-
sh
o
t
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
t
o
z
e
r
o
-
d
a
y
a
t
t
a
c
k
d
e
t
e
c
t
i
o
n
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
I
n
f
o
rm
a
t
i
o
n
S
e
c
u
r
i
t
y
,
v
o
l
.
2
2
,
n
o
.
4
,
p
p
.
9
4
7
–
9
5
9
,
A
u
g
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
0
7
/
s1
0
2
0
7
-
023
-
0
0
6
7
6
-
0.
[
2
]
M
.
P
a
r
i
c
h
e
r
l
a
,
M
.
R
i
t
o
n
g
a
,
S
.
R
.
S
h
i
n
d
e
,
S
.
M
.
C
h
a
u
d
h
a
r
i
,
R
.
Li
n
u
r
,
a
n
d
A
.
R
a
g
h
u
v
a
n
s
h
i
,
“
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
f
o
r
a
c
c
u
r
a
t
e
c
l
a
s
si
f
i
c
a
t
i
o
n
a
n
d
d
e
t
e
c
t
i
o
n
o
f
i
n
t
r
u
si
o
n
s
i
n
c
o
m
p
u
t
e
r
n
e
t
w
o
r
k
,
”
B
u
l
l
e
t
i
n
o
f
E
l
e
c
t
r
i
c
a
l
En
g
i
n
e
e
ri
n
g
a
n
d
I
n
f
o
rm
a
t
i
c
s
,
v
o
l
.
1
2
,
n
o
.
4
,
p
p
.
2
3
4
0
–
2
3
4
7
,
A
u
g
.
2
0
2
3
,
d
o
i
:
1
0
.
1
1
5
9
1
/
e
e
i
.
v
1
2
i
4
.
4
7
0
8
.
[
3
]
I
.
M
b
o
n
a
a
n
d
J.
H
.
P
.
E
l
o
f
f
,
“
D
e
t
e
c
t
i
n
g
z
e
r
o
-
d
a
y
i
n
t
r
u
s
i
o
n
a
t
t
a
c
k
s
u
si
n
g
semi
-
s
u
p
e
r
v
i
se
d
mac
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
e
s
,
”
I
EE
E
Ac
c
e
ss
,
v
o
l
.
1
0
,
p
p
.
6
9
8
2
2
–
6
9
8
3
8
,
2
0
2
2
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
2
.
3
1
8
7
1
1
6
.
[
4
]
P
.
C
h
a
n
d
r
e
,
P
.
M
a
h
a
l
l
e
,
a
n
d
G
.
S
h
i
n
d
e
,
“
I
n
t
r
u
si
o
n
p
r
e
v
e
n
t
i
o
n
s
y
s
t
e
m
u
s
i
n
g
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
f
o
r
w
i
r
e
l
e
ss
s
e
n
so
r
n
e
t
w
o
r
k
,
”
I
A
ES
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
Ar
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
1
1
,
n
o
.
2
,
p
p
.
5
0
4
–
5
1
5
,
Ju
n
.
2
0
2
2
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
a
i
.
v
1
1
.
i
2
.
p
p
5
0
4
-
5
1
5
.
[
5
]
A
.
S
.
A
l
f
o
u
d
i
e
t
a
l
.
,
“
H
y
p
e
r
c
l
u
st
e
r
i
n
g
m
o
d
e
l
f
o
r
d
y
n
a
mi
c
n
e
t
w
o
r
k
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
,
”
I
ET
C
o
m
m
u
n
i
c
a
t
i
o
n
s
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
4
9
/
c
m
u
2
.
1
2
5
2
3
.
[
6
]
B
.
M
.
S
e
r
i
n
e
l
l
i
,
A
.
C
o
l
l
e
n
,
a
n
d
N
.
A
.
N
i
j
d
a
m,
“
O
n
t
h
e
a
n
a
l
y
si
s
o
f
o
p
e
n
s
o
u
r
c
e
d
a
t
a
s
e
t
s
:
V
a
l
i
d
a
t
i
n
g
I
D
S
i
mp
l
e
me
n
t
a
t
i
o
n
f
o
r
w
e
l
l
-
k
n
o
w
n
a
n
d
z
e
r
o
d
a
y
a
t
t
a
c
k
d
e
t
e
c
t
i
o
n
,
”
i
n
Pro
c
e
d
i
a
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
E
l
se
v
i
e
r
B
.
V
.
,
2
0
2
1
,
p
p
.
1
9
2
–
1
9
9
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
r
o
c
s
.
2
0
2
1
.
0
7
.
0
2
4
.
[
7
]
S
.
A
l
i
,
S
.
U
.
R
e
h
ma
n
,
A
.
I
mr
a
n
,
G
.
A
d
e
e
m,
Z.
I
q
b
a
l
,
a
n
d
K
.
I
l
K
i
m
,
“
C
o
m
p
a
r
a
t
i
v
e
e
v
a
l
u
a
t
i
o
n
o
f
AI
-
b
a
sed
t
e
c
h
n
i
q
u
e
s
f
o
r
z
e
r
o
-
d
a
y
a
t
t
a
c
k
s
d
e
t
e
c
t
i
o
n
,
”
E
l
e
c
t
r
o
n
i
c
s
(
S
w
i
t
ze
rl
a
n
d
)
,
v
o
l
.
1
1
,
n
o
.
2
3
,
D
e
c
.
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
e
l
e
c
t
r
o
n
i
c
s
1
1
2
3
3
9
3
4
.
[
8
]
S
.
S
a
k
t
h
i
mu
r
u
g
a
n
,
A
.
S
a
n
j
a
y
K
u
m
a
a
r
,
V
.
V
i
g
n
e
s
h
,
a
n
d
P
.
S
a
n
t
h
i
,
“
A
ssess
men
t
o
f
z
e
r
o
-
d
a
y
v
u
l
n
e
r
a
b
i
l
i
t
y
u
si
n
g
mac
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
,
”
EAI
En
d
o
rse
d
T
r
a
n
sa
c
t
i
o
n
s o
n
I
n
t
e
r
n
e
t
o
f
T
h
i
n
g
s
,
v
o
l
.
1
0
,
2
0
2
4
,
d
o
i
:
1
0
.
4
1
0
8
/
e
e
t
i
o
t
.
4
9
7
8
.
[
9
]
T.
Zo
p
p
i
,
A
.
C
e
c
c
a
r
e
l
l
i
,
a
n
d
A
.
B
o
n
d
a
v
a
l
l
i
,
“
U
n
s
u
p
e
r
v
i
se
d
a
l
g
o
r
i
t
h
ms
t
o
d
e
t
e
c
t
z
e
r
o
-
d
a
y
a
t
t
a
c
k
s:
s
t
r
a
t
e
g
y
a
n
d
a
p
p
l
i
c
a
t
i
o
n
,”
I
EE
E
Ac
c
e
ss
,
v
o
l
.
9
,
p
p
.
9
0
6
0
3
–
9
0
6
1
5
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
1
.
3
0
9
0
9
5
7
.
[
1
0
]
M
.
K
a
n
t
e
,
V
.
S
h
a
r
m
a
,
a
n
d
K
.
G
u
p
t
a
,
“
M
i
t
i
g
a
t
i
n
g
r
a
n
s
o
mw
a
r
e
a
t
t
a
c
k
s
t
h
r
o
u
g
h
c
y
b
e
r
t
h
r
e
a
t
i
n
t
e
l
l
i
g
e
n
c
e
a
n
d
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
,
”
I
n
d
o
n
e
si
a
n
J
o
u
r
n
a
l
o
f
El
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
3
,
n
o
.
3
,
p
p
.
1
9
5
8
–
1
9
6
5
,
M
a
r
.
2
0
2
4
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
e
c
s.
v
3
3
.
i
3
.
p
p
1
9
5
8
-
1
9
6
5
.
[
1
1
]
Y
.
F
a
n
g
a
n
d
G
.
M
o
g
o
s,
“
D
e
t
e
c
t
i
n
g
a
t
t
a
c
k
s
o
n
e
-
ma
i
l
,
”
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
E
n
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
3
,
n
o
.
3
,
p
p
.
1
5
7
6
–
1
5
8
8
,
M
a
r
.
2
0
2
4
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
e
c
s.
v
3
3
.
i
3
.
p
p
1
5
7
6
-
1
5
8
8
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
36
,
No
.
2
,
No
v
em
b
er
20
24
:
1
28
4
-
1
29
3
1292
[
1
2
]
H
.
A
.
A
.
Essa
a
n
d
W
.
S
.
B
h
a
y
a
,
“
E
n
semb
l
e
l
e
a
r
n
i
n
g
c
l
a
ss
i
f
i
e
r
s
h
y
b
r
i
d
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
f
o
r
e
n
h
a
n
c
i
n
g
p
e
r
f
o
r
m
a
n
c
e
o
f
i
n
t
r
u
s
i
o
n
d
e
t
e
c
t
i
o
n
sy
s
t
e
m
,
”
B
u
l
l
e
t
i
n
o
f
El
e
c
t
ri
c
a
l
E
n
g
i
n
e
e
r
i
n
g
a
n
d
I
n
f
o
rm
a
t
i
c
s
,
v
o
l
.
1
3
,
n
o
.
1
,
p
p
.
6
6
5
–
6
7
6
,
F
e
b
.
2
0
2
4
,
d
o
i
:
1
0
.
1
1
5
9
1
/
e
e
i
.
v
1
3
i
1
.
5
8
4
4
.
[
1
3
]
B
.
I
.
H
a
i
r
a
b
,
H
.
K
.
A
sl
a
n
,
M
.
S
.
El
s
a
y
e
d
,
A
.
D
.
J
u
r
c
u
t
,
a
n
d
M
.
A
.
A
z
e
r
,
“
A
n
o
m
a
l
y
d
e
t
e
c
t
i
o
n
o
f
z
e
r
o
-
d
a
y
a
t
t
a
c
k
s
b
a
se
d
o
n
C
N
N
a
n
d
r
e
g
u
l
a
r
i
z
a
t
i
o
n
t
e
c
h
n
i
q
u
e
s
,
”
E
l
e
c
t
r
o
n
i
c
s
(
S
w
i
t
zer
l
a
n
d
)
,
v
o
l
.
1
2
,
n
o
.
3
,
F
e
b
.
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
e
l
e
c
t
r
o
n
i
c
s
1
2
0
3
0
5
7
3
.
[
1
4
]
X
.
S
u
n
,
J.
D
a
i
,
P
.
L
i
u
,
A
.
S
i
n
g
h
a
l
,
a
n
d
J
.
Y
e
n
,
“
U
s
i
n
g
B
a
y
e
si
a
n
n
e
t
w
o
r
k
s
f
o
r
p
r
o
b
a
b
i
l
i
s
t
i
c
i
d
e
n
t
i
f
i
c
a
t
i
o
n
o
f
z
e
r
o
-
d
a
y
a
t
t
a
c
k
p
a
t
h
s,
”
I
EEE
T
r
a
n
s
a
c
t
i
o
n
s
o
n
I
n
f
o
rm
a
t
i
o
n
F
o
r
e
n
s
i
c
s
a
n
d
S
e
c
u
r
i
t
y
,
v
o
l
.
1
3
,
n
o
.
1
0
,
p
p
.
2
5
0
6
–
2
5
2
1
,
2
0
1
8
,
d
o
i
:
1
0
.
1
1
0
9
/
TI
F
S
.
2
0
1
8
.
2
8
2
1
0
9
5
.
[
1
5
]
N
.
H
.
D
u
o
n
g
a
n
d
H
.
D
.
H
a
i
,
“
A
sem
i
-
su
p
e
r
v
i
se
d
m
o
d
e
l
f
o
r
n
e
t
w
o
r
k
t
r
a
f
f
i
c
a
n
o
ma
l
y
d
e
t
e
c
t
i
o
n
,
”
i
n
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
Ad
v
a
n
c
e
d
C
o
m
m
u
n
i
c
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
,
I
C
AC
T
,
2
0
1
5
,
p
p
.
7
0
–
7
5
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
A
C
T.
2
0
1
5
.
7
2
2
4
7
5
9
.
[
1
6
]
Y
.
R
e
n
,
K
.
F
e
n
g
,
F
.
H
u
,
L
.
C
h
e
n
,
a
n
d
Y
.
C
h
e
n
,
“
A
L
i
g
h
t
w
e
i
g
h
t
U
n
s
u
p
e
r
v
i
s
e
d
I
n
t
r
u
si
o
n
D
e
t
e
c
t
i
o
n
M
o
d
e
l
B
a
s
e
d
o
n
V
a
r
i
a
t
i
o
n
a
l
A
u
t
o
-
En
c
o
d
e
r
,
”
S
e
n
s
o
rs (B
a
se
l
)
,
v
o
l
.
2
3
,
n
o
.
2
0
,
O
c
t
.
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
s
2
3
2
0
8
4
0
7
.
[
1
7
]
D
.
E
.
R
u
me
l
h
a
r
t
,
G
.
E
.
H
i
n
t
o
n
,
a
n
d
R
.
J.
W
i
l
l
i
a
ms
,
“
L
e
a
r
n
i
n
g
i
n
t
e
r
n
a
l
r
e
p
r
e
se
n
t
a
t
i
o
n
s
b
y
e
r
r
o
r
p
r
o
p
a
g
a
t
i
o
n
,
”
Bi
o
m
e
t
r
i
k
a
,
1
9
8
6
,
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s:
/
/
a
p
i
.
s
e
ma
n
t
i
c
s
c
h
o
l
a
r
.
o
r
g
/
C
o
r
p
u
sI
D
:
6
2
2
4
5
7
4
2
[
1
8
]
C
.
B
r
u
n
n
e
r
,
A
.
K
o
,
a
n
d
S
.
F
o
d
o
r
,
“
A
n
a
u
t
o
e
n
c
o
d
e
r
-
e
n
h
a
n
c
e
d
s
t
a
c
k
i
n
g
n
e
u
r
a
l
n
e
t
w
o
r
k
mo
d
e
l
f
o
r
i
n
c
r
e
a
si
n
g
t
h
e
p
e
r
f
o
r
m
a
n
c
e
o
f
i
n
t
r
u
s
i
o
n
d
e
t
e
c
t
i
o
n
,
”
J
o
u
r
n
a
l
o
f
Ar
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
S
o
f
t
C
o
m
p
u
t
i
n
g
Re
se
a
rc
h
,
v
o
l
.
1
2
,
n
o
.
2
,
p
p
.
1
4
9
–
1
6
3
,
A
p
r
.
2
0
2
1
,
d
o
i
:
1
0
.
2
4
7
8
/
j
a
i
s
c
r
-
2
0
2
2
-
0
0
1
0
.
[
1
9
]
W
.
Zh
a
n
g
a
n
d
Y
.
Z
h
a
n
g
,
“
I
n
t
r
u
s
i
o
n
d
e
t
e
c
t
i
o
n
m
o
d
e
l
f
o
r
i
n
d
u
s
t
r
i
a
l
i
n
t
e
r
n
e
t
o
f
t
h
i
n
g
s
b
a
se
d
o
n
i
mp
r
o
v
e
d
a
u
t
o
e
n
c
o
d
e
r
,
”
C
o
m
p
u
t
a
t
i
o
n
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
N
e
u
r
o
sc
i
e
n
c
e
,
v
o
l
.
2
0
2
2
,
2
0
2
2
,
d
o
i
:
1
0
.
1
1
5
5
/
2
0
2
2
/
1
4
0
6
2
1
4
.
[
2
0
]
M
.
Le
o
n
,
T
.
M
a
r
k
o
v
i
c
,
a
n
d
S
.
P
u
n
n
e
k
k
a
t
,
“
F
e
a
t
u
r
e
e
n
c
o
d
i
n
g
w
i
t
h
a
u
t
o
e
n
c
o
d
e
r
a
n
d
d
i
f
f
e
r
e
n
t
i
a
l
e
v
o
l
u
t
i
o
n
f
o
r
n
e
t
w
o
r
k
i
n
t
r
u
s
i
o
n
d
e
t
e
c
t
i
o
n
u
si
n
g
mac
h
i
n
e
l
e
a
r
n
i
n
g
,
”
i
n
G
EC
C
O
2
0
2
2
C
o
m
p
a
n
i
o
n
-
Pro
c
e
e
d
i
n
g
s
o
f
t
h
e
2
0
2
2
G
e
n
e
t
i
c
a
n
d
Ev
o
l
u
t
i
o
n
a
r
y
C
o
m
p
u
t
a
t
i
o
n
C
o
n
f
e
re
n
c
e
,
A
ss
o
c
i
a
t
i
o
n
f
o
r
C
o
m
p
u
t
i
n
g
M
a
c
h
i
n
e
r
y
,
I
n
c
,
Ju
l
.
2
0
2
2
,
p
p
.
2
1
5
2
–
2
1
5
9
.
d
o
i
:
1
0
.
1
1
4
5
/
3
5
2
0
3
0
4
.
3
5
3
4
0
0
9
.
[
2
1
]
C
.
W
a
n
g
,
H
.
L
i
u
,
C
.
Li
,
Y
.
S
u
n
,
W
.
W
a
n
g
,
a
n
d
B
.
W
a
n
g
,
“
R
o
b
u
s
t
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
f
o
r
i
n
d
u
st
r
i
a
l
c
o
n
t
r
o
l
s
y
st
e
ms
u
s
i
n
g
i
mp
r
o
v
e
d
a
u
t
o
e
n
c
o
d
e
r
a
n
d
B
a
y
e
s
i
a
n
G
a
u
ss
i
a
n
mi
x
t
u
r
e
m
o
d
e
l
,”
Ma
t
h
e
m
a
t
i
c
s
,
v
o
l
.
1
1
,
n
o
.
9
,
M
a
y
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
ma
t
h
1
1
0
9
2
0
4
8
.
[
2
2
]
A
.
B
i
n
b
u
s
a
y
y
i
s
a
n
d
T
.
V
a
i
y
a
p
u
r
i
,
“
U
n
su
p
e
r
v
i
s
e
d
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
c
o
mb
i
n
i
n
g
c
o
n
v
o
l
u
t
i
o
n
a
l
a
u
t
o
e
n
c
o
d
e
r
a
n
d
o
n
e
-
c
l
a
ss SV
M
,
”
Ap
p
l
i
e
d
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
5
1
,
n
o
.
1
0
,
p
p
.
7
0
9
4
–
7
1
0
8
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
0
4
8
9
-
0
2
1
-
0
2
2
0
5
-
9.
[
2
3
]
C
.
W
a
n
g
,
Y
.
S
u
n
,
S
.
Lv
,
C
.
W
a
n
g
,
H
.
Li
u
,
a
n
d
B
.
W
a
n
g
,
“
I
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
s
y
st
e
m
b
a
s
e
d
o
n
o
n
e
-
c
l
a
ss
s
u
p
p
o
r
t
v
e
c
t
o
r
mac
h
i
n
e
a
n
d
G
a
u
ss
i
a
n
m
i
x
t
u
r
e
m
o
d
e
l
,”
El
e
c
t
r
o
n
i
c
s
(
S
w
i
t
zer
l
a
n
d
)
,
v
o
l
.
1
2
,
n
o
.
4
,
F
e
b
.
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
e
l
e
c
t
r
o
n
i
c
s
1
2
0
4
0
9
3
0
.
[
2
4
]
M
.
A
.
A
l
ma
i
a
h
e
t
a
l
.
,
“
P
e
r
f
o
r
ma
n
c
e
i
n
v
e
st
i
g
a
t
i
o
n
o
f
p
r
i
n
c
i
p
a
l
c
o
m
p
o
n
e
n
t
a
n
a
l
y
si
s
f
o
r
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
s
y
st
e
m
u
s
i
n
g
d
i
f
f
e
r
e
n
t
su
p
p
o
r
t
v
e
c
t
o
r
m
a
c
h
i
n
e
k
e
r
n
e
l
s
,
”
E
l
e
c
t
ro
n
i
c
s (
S
w
i
t
z
e
rl
a
n
d
)
,
v
o
l
.
1
1
,
n
o
.
2
1
,
N
o
v
.
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
e
l
e
c
t
r
o
n
i
c
s1
1
2
1
3
5
7
1
.
[
2
5
]
S
.
S
o
k
k
a
l
i
n
g
a
m
a
n
d
R
.
R
a
ma
k
r
i
s
h
n
a
n
,
“
A
n
i
n
t
e
l
l
i
g
e
n
t
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
sy
s
t
e
m
f
o
r
d
i
st
r
i
b
u
t
e
d
d
e
n
i
a
l
o
f
s
e
r
v
i
c
e
a
t
t
a
c
k
s:
A
su
p
p
o
r
t
v
e
c
t
o
r
m
a
c
h
i
n
e
w
i
t
h
h
y
b
r
i
d
o
p
t
i
m
i
z
a
t
i
o
n
a
l
g
o
r
i
t
h
m
b
a
se
d
a
p
p
r
o
a
c
h
,
”
C
o
n
c
u
rr
e
n
c
y
a
n
d
C
o
m
p
u
t
a
t
i
o
n
,
v
o
l
.
3
4
,
n
o
.
2
7
,
p
.
e
7
3
3
4
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
2
/
c
p
e
.
7
3
3
4
.
[
2
6
]
Z.
L
i
u
,
B
.
X
u
,
B
.
C
h
e
n
g
,
X
.
H
u
,
a
n
d
M
.
D
a
r
b
a
n
d
i
,
“
I
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
s
y
st
e
ms
i
n
t
h
e
c
l
o
u
d
c
o
m
p
u
t
i
n
g
:
A
c
o
mp
r
e
h
e
n
si
v
e
a
n
d
d
e
e
p
l
i
t
e
r
a
t
u
r
e
r
e
v
i
e
w
,
”
C
o
n
c
u
rre
n
c
y
a
n
d
C
o
m
p
u
t
a
t
i
o
n
,
v
o
l
.
3
4
,
n
o
.
4
,
p
.
e
6
6
4
6
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
2
/
c
p
e
.
6
6
4
6
.
[
2
7
]
H
a
r
i
n
t
a
k
a
a
n
d
C
.
W
i
j
a
y
a
,
“
A
u
t
o
ma
t
i
c
p
o
i
n
t
c
l
o
u
d
se
g
me
n
t
a
t
i
o
n
u
s
i
n
g
R
A
N
S
A
C
a
n
d
D
B
S
C
A
N
a
l
g
o
r
i
t
h
m
f
o
r
i
n
d
o
o
r
m
o
d
e
l
,
”
T
ELKO
MN
I
K
A
(
T
e
l
e
c
o
m
m
u
n
i
c
a
t
i
o
n
C
o
m
p
u
t
i
n
g
E
l
e
c
t
r
o
n
i
c
s
a
n
d
C
o
n
t
ro
l
)
,
v
o
l
.
2
1
,
n
o
.
6
,
p
p
.
1
3
1
7
–
1
3
2
5
,
2
0
2
3
,
d
o
i
:
1
0
.
1
2
9
2
8
/
TEL
K
O
M
N
I
K
A
.
V
2
1
I
6
.
2
5
2
9
9
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Ra
ja
k
u
m
a
r
Per
u
m
a
l
As
sista
n
t
P
r
o
fe
ss
o
r
i
n
S
c
h
o
o
l
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
i
n
e
e
rin
g
a
t
G
a
lg
o
ti
a
s
Un
i
v
e
rsity
,
h
o
l
d
s
2
2
y
e
a
rs
o
f
tea
c
h
in
g
e
x
p
e
rien
c
e
a
n
d
is
p
u
rs
u
in
g
a
Ph
.
D
.
in
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
i
n
e
e
rin
g
a
t
S
h
ri
Ve
n
k
a
tes
h
wa
ra
Un
iv
e
rsit
y
.
Wi
th
a
n
M
.
E
CS
E
fro
m
An
n
a
Un
i
v
e
rsity
,
h
e
h
a
s
p
u
b
li
sh
e
d
4
p
a
ten
ts
a
n
d
6
re
se
a
rc
h
p
a
p
e
rs
,
sp
e
c
ializin
g
i
n
n
e
two
rk
s,
c
l
o
u
d
c
o
m
p
u
ti
n
g
,
s
o
ft
wa
re
e
n
g
in
e
e
rin
g
,
a
n
d
m
a
c
h
i
n
e
l
e
a
rn
in
g
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
ra
jk
u
m
a
r.
jce
t@g
m
a
il
.
c
o
m
.
Dr
.
Ta
m
il
a
r
a
si
K
a
r
u
p
p
i
a
h
As
so
c
iate
P
ro
fe
ss
o
r
in
De
p
a
rtme
n
t
o
f
In
f
o
rm
a
ti
o
n
Tec
h
n
o
l
o
g
y
a
t
P
a
n
ima
lar
En
g
in
e
e
rin
g
C
o
ll
e
g
e
,
a
c
c
u
m
u
lati
n
g
2
4
y
e
a
rs
o
f
tea
c
h
in
g
e
x
p
e
rien
c
e
.
S
h
e
e
a
rn
e
d
h
e
r
P
h
.
D
.
re
c
o
r
d
wit
h
7
p
a
ten
ts,
5
b
o
o
k
c
h
a
p
ters
,
a
n
d
1
8
re
se
a
rc
h
p
a
p
e
rs
p
u
b
li
sh
e
d
in
e
ste
e
m
e
d
i
n
tern
a
ti
o
n
a
l
jo
u
rn
a
ls
a
n
d
c
o
n
fe
re
n
c
e
s.
He
r
e
x
p
e
rti
se
sp
a
n
s
c
y
b
e
r
se
c
u
rit
y
,
n
e
two
rk
s,
c
l
o
u
d
c
o
m
p
u
ti
n
g
,
a
n
d
m
a
c
h
in
e
lea
rn
i
ng.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
th
a
m
izh
a
n
n
a
@g
m
a
il
.
c
o
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
E
n
h
a
n
cin
g
n
etw
o
r
k
s
ec
u
r
ity
u
s
in
g
u
n
s
u
p
ervis
ed
lea
r
n
in
g
a
p
p
r
o
a
ch
to
…
(
R
a
ja
k
u
ma
r
P
eru
ma
l
)
1293
Upp
il
ira
j
a
P
a
n
n
e
e
r
se
lv
a
m
As
sista
n
t
P
ro
fe
ss
o
r
i
n
S
c
h
o
o
l
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
in
e
e
ri
n
g
a
t
G
a
lg
o
ti
a
s
Un
i
v
e
rsity
,
h
o
l
d
s
1
1
y
e
a
rs
o
f
tea
c
h
in
g
e
x
p
e
rien
c
e
.
Wi
t
h
a
n
M
.
E
CS
E
fro
m
An
n
a
Un
i
v
e
rsity
,
h
e
h
a
s
p
u
b
li
sh
e
d
2
p
a
ten
ts
a
n
d
5
re
se
a
rc
h
p
a
p
e
rs,
sp
e
c
ializin
g
i
n
c
lo
u
d
c
o
m
p
u
ti
n
g
,
c
o
m
p
u
ter
n
e
tw
o
rk
s,
so
ftwa
re
e
n
g
i
n
e
e
rin
g
,
a
n
d
m
a
c
h
in
e
lea
rn
in
g
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
u
p
p
i
li
ra
ja@
g
m
a
il
.
c
o
m
.
Ve
n
k
a
te
sa
n
A
n
n
a
m
a
la
i
As
sista
n
t
P
ro
fe
ss
o
r
in
De
p
a
rtme
n
t
o
f
El
e
c
tri
c
a
l,
El
e
c
tro
n
ic
s
a
n
d
Co
m
m
u
n
ica
ti
o
n
En
g
i
n
e
e
rin
g
a
t
G
a
lg
o
ti
a
s
Un
i
v
e
rsity
,
h
o
l
d
s
1
4
y
e
a
rs
o
f
tea
c
h
in
g
e
x
p
e
rien
c
e
.
Wi
th
a
n
M
.
E
Co
m
m
.
S
y
ste
m
s
fro
m
An
n
a
U
n
iv
e
rsit
y
,
h
e
h
a
s
p
u
b
li
sh
e
d
3
p
a
ten
ts
a
n
d
4
re
se
a
rc
h
p
a
p
e
rs,
sp
e
c
ializin
g
i
n
c
o
m
p
u
ter
n
e
t
wo
rk
s,
tele
c
o
m
m
u
n
ica
ti
o
n
s,
c
y
b
e
rse
c
u
rit
y
i
n
c
o
m
m
u
n
ica
ti
o
n
sy
ste
m
s
a
n
d
m
a
c
h
in
e
lea
rn
in
g
.
H
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
v
e
n
k
a
tes
a
n
@g
a
l
g
o
ti
a
su
n
i
v
e
rsity
.
e
d
u
.
i
n
.
Pra
b
u
K
a
li
y
a
p
e
r
u
m
a
l
As
sista
n
t
P
r
o
fe
ss
o
r
i
n
S
c
h
o
o
l
o
f
C
o
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
i
n
e
e
rin
g
a
t
G
a
lg
o
ti
a
s
Un
iv
e
rs
it
y
,
h
a
s
1
6
y
e
a
rs
o
f
tea
c
h
in
g
e
x
p
e
rien
c
e
.
Cu
rre
n
tl
y
p
u
rsu
i
n
g
a
P
h
.
D,
h
e
h
o
ld
s
a
n
M
.
Tec
h
i
n
CS
E
fro
m
S
RM
Un
i
v
e
rsity
a
n
d
M
B
A
fro
m
An
n
a
U
n
iv
e
rsit
y
.
He
h
a
s
p
u
b
li
s
h
e
d
4
p
a
ten
ts
a
n
d
9
re
se
a
rc
h
p
a
p
e
rs
in
in
tern
a
ti
o
n
a
l
j
o
u
rn
a
ls
a
n
d
c
o
n
fe
re
n
c
e
s.
His
e
x
p
e
rti
se
in
c
lu
d
e
s
c
y
b
e
r
se
c
u
rit
y
,
n
e
two
rk
s,
c
l
o
u
d
c
o
m
p
u
ti
n
g
,
a
n
d
m
a
c
h
in
e
lea
rn
in
g
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
m
e
g
a
.
p
ra
b
u
@g
m
a
il
.
c
o
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.