I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
p
ute
r
E
ng
in
ee
ring
(
I
J
E
CE
)
Vo
l.
10
,
No
.
5
,
Octo
b
er
2
0
2
0
,
p
p
.
5
4
4
5
~
5
4
5
2
I
SS
N:
2
0
8
8
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v
1
0
i
5
.
pp
5
4
4
5
-
5
4
5
2
5445
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m/in
d
ex
.
p
h
p
/I
JE
C
E
Co
m
pa
riso
n
s
t
ud
y
of
m
a
c
hine
lea
r
ning
c
la
ss
ifiers
to
dete
ct
a
no
m
a
li
es
Nis
ha
P
.
Sh
et
t
y
1
,
J
a
y
a
s
hree
Sh
et
t
y
2
,
Ro
hil
Na
rula
3
,
K
us
h
a
g
ra
T
a
nd
o
na
4
1,
2,
4
De
p
a
rtm
e
n
t
o
f
In
f
o
rm
a
ti
o
n
a
n
d
Co
m
m
u
n
ica
ti
o
n
T
e
c
h
n
o
lo
g
y
,
M
a
n
ip
a
l
I
n
stit
u
te
o
f
T
e
c
h
n
o
lo
g
y
,
M
a
n
ip
a
l
A
c
a
d
e
m
y
o
f
Hi
g
h
e
r
Ed
u
c
a
ti
o
n
,
I
n
d
ia
3
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter S
c
ien
c
e
a
n
d
E
n
g
in
e
e
rin
g
,
M
a
n
ip
a
l
I
n
stit
u
te o
f
T
e
c
h
n
o
lo
g
y
,
M
a
n
ip
a
l
A
c
a
d
e
m
y
o
f
Hi
g
h
e
r
Ed
u
c
a
ti
o
n
,
I
n
d
ia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
No
v
2
9
,
2
0
1
9
R
ev
i
s
ed
A
p
r
3
,
2
0
2
0
A
cc
ep
ted
A
p
r
1
3
,
2
0
2
0
In
th
is
e
ra
o
f
In
tern
e
t
e
n
su
rin
g
th
e
c
o
n
f
id
e
n
ti
a
li
ty
,
a
u
th
e
n
ti
c
a
ti
o
n
a
n
d
in
teg
rit
y
o
f
a
n
y
re
so
u
rc
e
e
x
c
h
a
n
g
e
d
o
v
e
r
th
e
n
e
t
is
th
e
i
m
p
e
ra
ti
v
e
.
P
re
se
n
c
e
o
f
in
tru
sio
n
p
re
v
e
n
ti
o
n
tec
h
n
iq
u
e
s
li
k
e
stro
n
g
p
a
ss
w
o
rd
,
f
ir
e
w
a
ll
s
e
tc.
a
re
n
o
t
su
ff
icie
n
t
to
m
o
n
it
o
r
s
u
c
h
v
o
l
u
m
in
o
u
s
n
e
tw
o
rk
traff
ic
a
s
th
e
y
c
a
n
b
e
b
re
a
c
h
e
d
e
a
sil
y
.
Ex
isti
n
g
sig
n
a
tu
re
b
a
se
d
d
e
tec
ti
o
n
tec
h
n
i
q
u
e
s
li
k
e
a
n
ti
v
iru
s
o
n
ly
o
f
fe
rs
p
ro
tec
ti
o
n
a
g
a
in
st
k
n
o
w
n
a
tt
a
c
k
s
w
h
o
se
sig
n
a
tu
re
s
a
re
sto
re
d
i
n
th
e
d
a
tab
a
se
.
T
h
u
s,
th
e
n
e
e
d
f
o
r
re
a
l
-
ti
m
e
d
e
tec
ti
o
n
o
f
a
b
e
r
ra
ti
o
n
s is
o
b
se
rv
e
d
.
Ex
isti
n
g
sig
n
a
tu
re
b
a
se
d
d
e
tec
ti
o
n
tec
h
n
i
q
u
e
s
li
k
e
a
n
ti
v
ir
u
s
o
n
ly
o
ff
e
rs
p
ro
tec
ti
o
n
a
g
a
in
st
k
n
o
w
n
a
tt
a
c
k
s
w
h
o
se
sig
n
a
tu
re
s
a
re
sto
re
d
i
n
th
e
d
a
tab
a
se
.
M
a
c
h
in
e
lea
rn
i
n
g
c
las
sif
ier
s
a
re
i
m
p
le
m
e
n
ted
h
e
re
to
lea
rn
h
o
w
th
e
v
a
lu
e
s
o
f
v
a
rio
u
s
f
ield
s
li
k
e
so
u
rc
e
b
y
tes
,
d
e
stin
a
ti
o
n
b
y
tes
e
t
c
.
in
a
n
e
two
rk
p
a
c
k
e
t
d
e
c
id
e
s
if
th
e
p
a
c
k
e
t
is
c
o
m
p
ro
m
is
e
d
o
r
n
o
t
.
F
i
n
a
ll
y
th
e
a
c
c
u
r
a
c
y
o
f
th
e
ir
d
e
tec
ti
o
n
is
c
o
m
p
a
re
d
to
c
h
o
o
se
th
e
b
e
st
su
it
e
d
c
las
sif
ier
f
o
r
th
is
p
u
rp
o
se
.
T
h
e
o
u
tco
m
e
th
u
s
p
ro
d
u
c
e
d
m
a
y
b
e
u
se
f
u
l
to
o
ff
e
r
re
a
l
ti
m
e
d
e
te
c
ti
o
n
w
h
il
e
e
x
c
h
a
n
g
in
g
se
n
siti
v
e
in
f
o
r
m
a
ti
o
n
su
c
h
a
s c
re
d
it
c
a
rd
d
e
tails.
K
ey
w
o
r
d
s
:
B
ag
g
i
n
g
I
DS
Net
w
o
r
k
attac
k
s
Neu
r
al
n
et
w
o
r
k
R
an
d
o
m
f
o
r
est
SVM
Co
p
y
rig
h
t
©
2
0
2
0
In
stit
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
All
rig
h
ts re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
J
ay
a
s
h
r
ee
S
h
ett
y
,
Dep
ar
t
m
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
an
d
E
n
g
i
n
ee
r
in
g
,
Ma
n
ip
al
I
n
s
tit
u
te
o
f
T
ec
h
n
o
lo
g
y
,
Ma
n
ip
al
A
ca
d
e
m
y
o
f
Hi
g
h
er
E
d
u
ca
tio
n
,
Ma
n
ip
al
-
5
7
6
1
0
4
I
n
d
ia
.
E
m
ail:
j
a
y
as
h
r
ee
.
s
s
h
ett
y
@
m
a
n
ip
al.
ed
u
1.
I
NT
RO
D
UCT
I
O
N
I
n
cr
ea
s
in
g
r
ate
o
f
c
y
b
er
cr
i
m
es
is
a
g
r
av
e
co
n
ce
r
n
n
o
w
a
d
ay
s
.
O
w
in
g
to
th
e
i
n
cr
ea
s
e
d
u
s
ag
e
o
f
I
n
ter
n
et
in
all
zo
n
e
s
o
f
lif
e
p
r
iv
ac
y
a
n
d
s
ec
u
r
it
y
h
as
b
ec
o
m
e
th
e
n
ee
d
o
f
th
e
h
o
u
r
.
An
y
m
a
n
ip
u
lat
io
n
d
o
n
e
to
r
eso
u
r
ce
b
y
a
n
u
n
a
u
t
h
o
r
ized
en
tit
y
w
it
h
th
e
i
n
te
n
s
io
n
o
f
c
au
s
i
n
g
h
ar
m
is
ter
m
ed
as
in
tr
u
s
io
n
.
an
i
n
tr
u
s
io
n
d
etec
tio
n
s
y
s
te
m
(
I
DS)
i
s
a
d
ef
en
s
e
s
y
s
te
m
w
h
ich
s
cr
ee
n
s
th
e
ac
ti
v
itie
s
i
n
a
co
m
p
u
ter
s
y
s
te
m
o
r
a
n
e
t
w
o
r
k
au
to
m
at
icall
y
to
d
etec
t b
r
ea
ch
es a
n
d
s
u
b
s
eq
u
e
n
tl
y
n
o
ti
f
ies t
h
e
u
s
er
ab
o
u
t a
n
y
v
io
lat
io
n
s
[
1
]
.
T
h
er
e
ar
e
m
ai
n
l
y
f
o
u
r
ca
ta
g
o
r
ies
o
f
at
tack
s
[
2
]
.
I
n
Do
S
A
tt
ac
k
,
attac
k
er
s
p
r
ev
e
n
t
o
th
er
u
s
er
s
to
u
s
e
a
leg
i
ti
m
ate
s
er
v
ice
f
o
r
a
p
er
io
d
o
f
ti
m
e
b
y
p
r
ev
e
n
ti
n
g
ac
c
ess
to
o
th
er
s
.
B
an
k
s
w
eb
s
ites
,
VT
U
s
ites
etc.
ar
e
p
r
o
n
e
to
th
is
k
in
d
o
f
attac
k
s
.
I
n
r
e
m
o
te
to
u
s
er
(
R
2
L
)
,
th
e
th
r
ea
t
ca
u
s
ed
b
y
a
s
ec
lu
d
ed
p
er
s
o
n
to
g
ain
co
n
tr
o
l
o
f
a
tar
g
et
r
eso
u
r
ce
.
So
cial
E
n
g
in
ee
r
i
n
g
is
o
n
e
s
u
c
h
attac
k
.
I
n
u
s
er
to
r
o
o
t
(
U2
R
)
,
p
er
s
o
n
w
it
h
lo
ca
l
p
r
iv
ileg
e
s
ab
u
s
es
t
h
e
s
y
s
te
m
’
s
v
u
l
n
er
ab
ilit
ies
to
g
et
s
u
p
er
u
s
er
r
ig
h
t
s
.
B
u
f
f
er
o
v
er
f
lo
w
er
r
o
r
s
an
d
er
r
o
r
s
ca
u
s
ed
b
y
ir
r
eg
u
lar
itie
s
in
e
n
v
ir
o
n
m
e
n
t
al
ass
u
m
p
tio
n
s
ar
e
s
o
m
e
co
m
m
o
n
ex
a
m
p
les.
I
n
P
r
o
b
in
g
,
A
ttac
k
er
ex
a
m
i
n
e
s
th
e
s
y
s
te
m
to
f
in
d
all
it
s
liab
ili
ties
.
B
y
u
s
in
g
t
h
ese
v
u
l
n
er
ab
il
ities
t
h
e
s
y
s
te
m
is
ab
u
s
ed
.
T
w
o
co
m
m
o
n
l
y
u
s
ed
I
DS
b
ased
o
n
th
e
lo
c
atio
n
ar
e
[
3
]
n
et
w
o
r
k
b
ased
i
n
tr
u
s
io
n
d
etec
t
io
n
s
y
s
te
m
(
NI
DS)
(
T
r
af
f
ic
f
lo
w
i
n
g
in
t
h
e
n
et
w
o
r
k
i
s
ex
a
m
i
n
ed
)
an
d
h
o
s
t
b
ased
in
tr
u
s
io
n
d
etec
t
io
n
s
y
s
te
m
(
HI
D
S
)
(
T
r
af
f
ic
o
r
i
g
in
ated
f
r
o
m
o
r
is
d
esti
n
ed
to
a
p
ar
tic
u
lar
h
o
s
t
is
s
cr
u
ti
n
ized
).
B
ased
o
n
d
etec
ti
o
n
tec
h
n
iq
u
es
I
D
S
ca
n
b
e
ca
te
g
o
r
ized
as
[
4
]
Mi
s
u
s
e
d
etec
tio
n
a
n
d
An
o
m
al
y
d
etec
tio
n
.
I
n
m
is
u
s
e
d
etec
tio
n
s
ig
n
at
u
r
es
o
f
a
ll
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
10
,
No
.
5
,
Octo
b
e
r
2
0
2
0
:
5
4
4
5
-
5452
5446
k
n
o
w
n
attac
k
s
ar
e
d
o
cu
m
e
n
te
d
.
Sig
n
a
tu
r
es
o
f
e
v
er
y
n
e
w
p
a
ck
et
e
n
co
u
n
ter
ed
ar
e
co
m
p
ar
e
d
w
it
h
t
h
e
d
atab
ase
to
ch
ec
k
w
h
et
h
er
it
is
a
n
attac
k
o
r
n
o
t.
Alth
o
u
g
h
t
h
is
tec
h
n
i
q
u
e
p
r
o
v
id
es
a
h
i
g
h
d
etec
tio
n
r
ate
it
is
v
er
y
ti
m
e
co
n
s
u
m
i
n
g
an
d
o
n
l
y
is
e
f
f
ec
ti
v
e
f
o
r
k
n
o
w
n
at
tack
s
o
n
l
y
.
I
n
An
o
m
al
y
d
etec
tio
n
,
a
n
y
v
ar
iat
io
n
f
r
o
m
t
h
e
n
o
r
m
al
ex
p
ec
ted
b
eh
av
io
r
is
f
la
g
g
ed
as
attac
k
s
w
i
th
o
u
t
a
n
y
p
r
io
r
m
aster
y
o
n
att
ac
k
s
.
A
h
i
g
h
er
f
al
s
e
alar
m
r
ate
i
s
o
b
tain
ed
b
y
t
h
is
m
et
h
o
d
.
P
r
o
b
lem
De
f
in
itio
n
.
I
n
cr
ea
s
e
in
I
n
ter
n
et
cr
i
m
e
s
n
o
w
ad
a
y
s
ex
e
m
p
li
f
ies
th
e
n
ee
d
f
o
r
a
co
m
p
eten
t
in
tr
u
s
io
n
d
etec
tio
n
s
y
s
te
m
.
E
v
er
y
s
ec
to
r
in
t
h
e
s
o
ciet
y
is
co
m
p
u
ter
ized
,
th
u
s
a
lar
g
e
v
o
lu
m
e
o
f
i
m
p
o
r
t
an
t
in
f
o
r
m
atio
n
s
u
c
h
as
p
er
s
o
n
al
p
r
o
f
iles
an
d
cr
ed
it
ca
r
d
in
f
o
r
m
atio
n
ar
e
en
ter
ed
,
ed
ited
an
d
tr
an
s
f
er
r
ed
ac
r
o
s
s
th
e
n
et
w
o
r
k
d
ail
y
.
T
h
is
s
h
i
f
t
f
r
o
m
ce
n
tr
alize
d
co
m
p
u
tin
g
t
o
n
et
w
o
r
k
ed
en
v
ir
o
n
m
e
n
t
h
as
in
v
o
k
ed
a
n
ee
d
to
i
m
p
r
o
v
e
th
e
s
ec
u
r
it
y
o
f
t
h
e
n
et
w
o
r
k
s
.
F
a
u
lt
y
p
ac
k
et
f
ilter
i
n
g
tec
h
n
o
lo
g
y
o
f
f
ir
e
w
a
lls
,
g
en
er
alit
y
p
r
o
b
le
m
o
f
an
ti
v
ir
u
s
,
h
u
g
e
co
s
t
an
d
p
er
f
o
r
m
an
ce
b
o
ttlen
ec
k
o
f
ap
p
licat
io
n
g
a
te
w
a
y
w
h
ich
s
lo
w
s
d
o
w
n
t
h
e
n
et
w
o
r
k
etc
d
o
es
n
o
t
allo
w
t
h
e
m
to
ev
ad
e
all
attac
k
er
s
a
n
d
ar
e
n
o
t
co
m
p
letel
y
ef
f
ic
ien
t.
Ma
c
h
i
n
e
lear
n
in
g
cla
s
s
i
f
ier
s
ar
e
i
m
p
le
m
en
ted
h
er
e
to
lear
n
h
o
w
t
h
e
v
al
u
es
o
f
v
ar
io
u
s
f
ield
s
lik
e
s
o
u
r
ce
b
y
te
s
,
d
es
tin
atio
n
b
y
te
s
etc.
in
a
n
et
w
o
r
k
p
ac
k
et
d
ec
id
es
if
t
h
e
p
ac
k
e
t
is
co
m
p
r
o
m
is
e
d
o
r
n
o
t
.
T
h
is
r
esear
ch
p
o
in
t
s
o
u
t
t
h
e
n
ee
d
f
o
r
a
p
r
o
f
icien
t
in
tr
u
s
io
n
d
etec
tio
n
s
y
s
te
m
(
I
DS)
w
h
ic
h
e
x
p
o
s
es
m
alicio
u
s
p
ac
k
e
ts
e
f
f
ec
t
iv
el
y
ev
e
n
i
f
a
b
r
o
ad
r
an
g
e
o
f
i
n
tr
u
s
io
n
s
ar
e
en
co
u
n
ter
ed
an
d
ca
n
n
o
t b
e
ta
m
p
er
ed
.
Sig
n
i
f
ica
n
ce
o
f
P
r
o
p
o
s
ed
R
esear
ch
.
T
h
e
f
ir
s
t
r
ea
s
o
n
f
o
r
ch
o
o
s
in
g
t
h
e
r
esear
ch
i
s
t
h
at
I
n
ter
n
et
i
s
a
p
ar
t o
f
ev
er
y
d
a
y
r
o
u
ti
n
e
n
o
w
ad
a
y
s
f
o
r
m
o
s
t o
f
t
h
e
p
eo
p
le
en
co
m
p
a
s
s
i
n
g
all
a
s
p
ec
ts
f
r
o
m
o
n
lin
e
s
h
o
p
p
in
g
to
s
o
cial
m
ed
ia.
Hen
ce
en
s
u
r
i
n
g
th
at
o
n
l
y
s
a
n
ctio
n
ed
p
eo
p
le
s
h
o
u
ld
h
a
v
e
ac
ce
s
s
to
p
r
i
v
ate
in
f
o
r
m
atio
n
w
h
ile
p
r
eser
v
in
g
its
i
n
te
g
r
it
y
i
s
q
u
i
te
n
ec
ess
ar
y
.
Seco
n
d
l
y
,
Sig
n
a
tu
r
e
b
ased
m
et
h
o
d
s
d
esp
ite
o
f
h
a
v
i
n
g
lo
w
f
alse
p
o
s
itiv
e
r
ates
is
in
e
f
f
ec
ti
v
e
i
n
p
r
o
v
id
in
g
d
e
f
en
s
e
a
g
ai
n
s
t
u
n
k
n
o
w
n
a
ttack
s
.
Sta
tis
tica
l a
n
o
m
al
y
b
ased
d
etec
tio
n
ex
p
lo
r
es
o
n
d
is
cr
ep
an
c
y
o
f
tr
af
f
ic
ch
ar
ac
ter
i
s
tics
f
r
o
m
n
o
r
m
al
i
n
ter
m
s
o
f
v
o
lu
m
e.
I
t
f
a
ils
w
h
en
attac
k
er
is
cr
af
t
y
e
n
o
u
g
h
to
k
ee
p
th
e
in
c
o
n
g
r
u
it
y
b
elo
w
ce
r
tai
n
le
v
el
s
.
Fin
a
ll
y
,
m
ac
h
i
n
e
lear
n
in
g
al
g
o
r
ith
m
s
ar
e
ch
o
s
en
as
t
h
e
y
h
a
v
e
p
r
o
v
e
n
to
b
e
a
n
e
f
f
ec
ti
v
e
s
o
l
u
tio
n
i
n
id
e
n
ti
f
y
in
g
ab
n
o
r
m
alit
ies
i
m
m
ed
iat
el
y
w
i
th
o
u
t
b
ei
n
g
s
u
s
ce
p
tib
le
to
an
y
s
o
r
t o
f
m
a
n
i
p
u
latio
n
s
f
r
o
m
at
tack
er
s
.
2.
L
I
T
E
R
AT
U
RE
R
E
VI
E
W
U.
C
a
v
u
s
o
g
l
u
[
5
]
e
m
p
lo
y
ed
v
ar
io
u
s
m
ac
h
in
e
lear
n
in
g
a
lg
o
r
ith
m
s
to
e
v
alu
a
te
w
h
ic
h
cla
s
s
if
ier
g
a
v
e
b
etter
d
etec
tio
n
f
o
r
ea
ch
at
tac
k
t
y
p
e.
Data
p
r
ep
r
o
ce
s
s
in
g
a
n
d
n
e
w
f
ea
t
u
r
e
r
ed
u
ctio
n
m
eth
o
d
s
C
f
s
S
u
b
s
etE
v
al
an
d
W
r
ap
p
er
Su
b
s
etE
v
al
w
er
e
u
s
ed
.
T
h
e
m
et
h
o
d
ca
n
b
e
f
u
r
t
h
er
ex
ten
d
ed
to
f
i
n
d
o
n
e
o
p
ti
m
u
m
c
lass
if
ier
w
h
ic
h
g
iv
e
s
t
h
e
o
p
ti
m
u
m
d
etec
tio
n
f
o
r
all
ca
teg
o
r
ies o
f
attac
k
s
.
Kan
g
et
al
[
6
]
h
av
e
d
em
o
n
s
tr
ated
in
tr
u
s
io
n
d
etec
tio
n
at
th
e
clu
s
ter
h
ea
d
b
y
e
m
p
lo
y
in
g
S
NORT
an
d
MY
SQ
L
d
ata
b
ases
.
C
lu
s
ter
h
ea
d
r
ec
eiv
es
a
g
g
r
eg
ated
i
n
f
o
r
m
at
io
n
f
r
o
m
e
n
tire
n
et
w
o
r
k
m
a
k
i
n
g
d
etec
tio
n
q
u
ick
er
.
T
h
e
p
r
ese
n
ted
r
esea
r
ch
is
m
o
s
t
s
u
itab
le
f
o
r
o
r
g
a
n
izatio
n
h
av
i
n
g
lar
g
e
a
m
o
u
n
t
o
f
d
ata.
Ho
w
e
v
er
th
e
tec
h
n
iq
u
e
is
i
m
p
le
m
e
n
ted
o
n
l
y
o
n
s
tatic
n
et
w
o
r
k
a
n
d
S
NORT
alth
o
u
g
h
o
f
f
er
s
g
o
o
d
d
etec
tio
n
f
o
r
k
n
o
wn
attac
k
s
f
ail
s
f
o
r
an
o
m
al
y
d
etec
tio
n
.
B
ay
k
ar
a
a
n
d
Das
[
7
]
in
co
r
p
o
r
ated
a
h
o
n
e
y
p
o
t
b
a
s
ed
ap
p
r
o
ac
h
f
o
r
r
ea
l
ti
m
e
in
tr
u
s
io
n
d
etec
tio
n
.
T
h
e
p
r
o
p
o
s
ed
s
y
s
te
m
r
ed
u
ce
s
f
alse
p
o
s
itiv
e
le
v
el
an
d
p
r
o
v
id
es
p
r
o
tectio
n
ag
ain
s
t
attac
k
s
s
u
c
h
as
ze
r
o
d
ay
attac
k
.
Ho
w
e
v
er
th
is
ap
p
r
o
ac
h
is
co
s
tl
y
i
n
ter
m
s
o
f
co
n
f
i
g
u
r
atio
n
,
in
s
tallatio
n
a
n
d
m
a
n
a
g
e
m
en
t
o
f
h
o
n
e
y
p
o
ts
w
h
e
n
co
m
p
ar
ed
to
m
ac
h
i
n
e
l
ea
r
n
in
g
clas
s
i
f
ier
s
.
I
f
r
e
g
u
lar
l
y
th
e
at
tack
s
ig
n
at
u
r
es
co
llect
ed
f
r
o
m
lo
g
f
ile
o
f
h
o
n
e
y
p
o
t
s
ar
e
n
o
t u
p
d
ated
in
t
h
e
d
atab
ase
th
e
n
th
e
d
etec
tio
n
r
ate
s
u
f
f
er
s
.
Z
h
ao
et
al.
[
8
]
u
s
ed
P
r
in
cip
al
C
o
m
p
o
n
e
n
t
An
al
y
s
i
s
to
r
ed
u
ce
th
e
d
i
m
e
n
s
io
n
s
f
o
r
lar
g
e
d
ataset
to
m
ak
e
it
s
u
itab
le
I
OT
d
ev
ice
s
.
A
cc
u
r
ac
y
o
f
So
f
t
m
a
x
a
n
d
KNN
C
la
s
s
i
f
ier
s
is
co
m
p
ar
ed
,
w
h
er
e
s
o
f
t
m
a
x
r
eg
r
ess
io
n
s
h
o
w
s
b
etter
ti
m
e
p
er
f
o
r
m
a
n
ce
.
Un
s
u
p
er
v
is
ed
lear
n
in
g
alg
o
r
ith
m
s
ca
n
b
e
u
s
ed
s
o
th
at
m
a
n
y
b
r
o
ad
e
r
r
an
g
e
o
f
a
ttack
s
ca
n
b
e
d
is
co
v
er
ed
.
A
l
s
o
s
i
n
ce
t
h
e
al
g
o
r
ith
m
is
to
b
e
d
ep
lo
y
ed
o
n
I
OT
m
e
m
o
r
y
s
av
i
n
g
tech
n
iq
u
e
s
s
h
o
u
ld
b
e
ap
p
lied
.
Sin
g
h
et
al.
[
9
]
p
r
o
p
o
s
ed
a
f
o
u
r
tier
ar
c
h
itect
u
r
e
h
a
v
i
n
g
d
ata
p
r
ep
r
o
ce
s
s
in
g
i
n
f
ir
s
t
t
i
er
,
f
ea
tu
r
e
ex
tr
ac
tio
n
i
n
s
ec
o
n
d
tier
,
clas
s
if
icatio
n
in
t
h
ir
d
tier
an
d
u
s
er
in
ter
f
ac
e
i
n
f
o
u
r
t
h
tier
.
Gen
er
alize
d
d
is
cr
i
m
i
n
an
t
an
al
y
s
is
w
as
u
s
ed
f
o
r
ex
tr
ac
ti
n
g
f
ea
t
u
r
e
s
f
r
o
m
KDD
C
u
p
9
9
d
ata
s
et.
C
4
.
5
o
f
f
er
ed
b
etter
d
etec
tio
n
f
o
r
n
o
r
m
al
an
d
p
r
o
b
e
class
es,
iSVM
d
etec
ted
n
o
r
m
al
an
d
Do
S
attac
k
s
an
d
h
y
b
r
id
C
4
.
5
-
i
SVM
p
er
ce
iv
ed
U2
R
an
d
R
2
L
attac
k
s
.
Alt
h
o
u
g
h
th
e
in
d
i
v
i
d
u
al
clas
s
i
f
ier
s
o
f
f
er
ed
g
o
o
d
ac
cu
r
ac
y
th
er
e
i
s
a
r
o
o
m
f
o
r
i
m
p
r
o
v
e
m
e
n
t
i
n
d
etec
tio
n
o
f
U2
R
an
d
R
2
L
att
ac
k
s
.
Ho
q
u
e
et
al.
[
1
0
]
u
s
ed
g
en
eti
c
alg
o
r
ith
m
to
d
etec
t
v
ar
io
u
s
t
y
p
es
o
f
at
tack
s
.
Fit
n
es
s
o
f
c
h
r
o
m
o
s
o
m
e
w
a
s
r
ea
lized
u
s
i
n
g
s
ta
n
d
ar
d
d
ev
iatio
n
m
et
h
o
d
w
h
ic
h
ca
n
b
e
m
ad
e
b
etter
b
y
u
s
i
n
g
h
e
u
r
is
tic
ap
p
r
o
ac
h
es.
L
i
n
et
a
l.
[
1
1
]
d
ev
elo
p
e
d
an
ap
p
r
o
ac
h
w
h
ic
h
co
m
b
in
e
s
lo
g
f
ile
an
al
y
s
i
s
tech
n
o
lo
g
y
a
n
d
B
P
n
eu
r
al
n
et
w
o
r
k
tech
n
o
lo
g
y
.
E
v
e
n
th
o
u
g
h
th
i
s
tech
n
iq
u
e
d
etec
ted
b
o
th
m
i
s
u
s
e
an
d
a
n
o
m
a
l
y
d
ata,
lo
g
f
iles
u
s
ed
ar
e
m
o
n
ito
r
ed
b
y
d
ae
m
o
n
s
m
a
k
i
n
g
it
le
s
s
tr
u
s
t
w
o
r
t
h
y
.
L
e
u
a
n
d
L
in
[
1
2
]
em
p
lo
y
ed
C
h
i
-
Sq
u
ar
e
m
et
h
o
d
to
d
etec
t
v
ar
iatio
n
i
n
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:
2
0
8
8
-
8708
C
o
mp
a
r
is
o
n
s
tu
d
y
o
f
ma
ch
in
e
lea
r
n
in
g
cla
s
s
ifier
s
to
d
etec
t a
n
o
ma
lies
(
N
i
s
h
a
P
.
S
h
etty)
5447
p
ac
k
et
s
tatis
tics
w
h
ich
h
ap
p
en
s
u
s
u
a
ll
y
i
n
ca
s
e
o
f
attac
k
s
.
On
t
h
e
co
n
tr
ar
y
to
clea
r
l
y
estab
lis
h
n
o
r
m
al
d
is
tr
ib
u
tio
n
h
u
g
e
a
m
o
u
n
ts
o
f
d
ata
m
u
s
t b
e
f
o
r
k
ed
t
h
r
o
u
g
h
w
h
ich
is
ti
m
e
co
n
s
u
m
i
n
g
.
Seo
[
1
3
]
im
p
le
m
e
n
ted
M
u
ltip
le
Su
p
p
o
r
t
Vec
to
r
Ma
ch
in
e
s
i
n
w
h
ich
e
v
er
y
h
y
p
er
p
la
n
e
is
tr
ain
ed
to
d
etec
t
s
p
ec
if
ic
attac
k
,
th
er
eb
y
d
ec
r
ea
s
in
g
t
h
e
f
alse
p
o
s
it
iv
e
r
ate.
B
u
t,
s
i
n
ce
M
SVM
h
as
b
ig
g
er
m
ar
g
in
th
a
n
class
ical
SVM,
s
o
m
eti
m
es
e
v
en
th
e
n
o
r
m
a
l
p
ac
k
ets
ar
e
class
i
f
ied
as
atta
ck
p
ac
k
ets.
M
u
k
k
a
m
ala
et
al.
[
1
4
]
co
m
p
ar
ed
th
e
ac
c
u
r
ac
y
o
f
SV
M
an
d
n
e
u
r
al
n
e
t
w
o
r
k
o
n
D
A
R
P
A
d
atase
t
.
SVM
w
a
s
o
b
s
er
v
ed
to
b
e
p
er
f
o
r
m
i
n
g
b
etter
th
a
n
NN
f
o
r
t
h
e
s
elec
te
d
1
3
f
ea
tu
r
es.
Ho
w
ev
er
SVM
w
a
s
li
m
ited
o
n
l
y
i
n
m
a
k
in
g
b
i
n
ar
y
c
lass
if
ica
tio
n
s
an
d
th
e
m
eth
o
d
co
u
ld
b
e
ex
te
n
d
ed
to
d
etec
t m
o
r
e
v
ar
ia
n
ts
o
f
attac
k
s
.
3.
P
RO
P
O
SE
D
M
E
T
H
O
D
O
L
O
G
Y
Fig
u
r
e
1
h
ig
h
li
g
h
ts
t
h
e
m
eth
o
d
o
lo
g
y
f
o
llo
w
ed
in
t
h
e
p
ap
er
w
h
er
e
ea
ch
f
ield
is
d
escr
ib
ed
b
elo
w
.
Fig
u
r
e
1
.
Ov
er
all
m
et
h
o
d
o
lo
g
y
a
.
Data
s
et
T
h
e
KDD9
9
d
ataset
[
1
5
]
em
b
o
d
ies
4
1
attr
ib
u
tes
an
d
th
e
‘
clas
s
’
a
ttrib
u
tes
[
1
6
]
w
h
ic
h
s
p
ec
i
f
ie
s
w
h
et
h
er
a
g
i
v
en
ca
s
e
i
s
a
n
o
r
m
al
o
r
an
attac
k
as s
h
o
w
n
i
n
F
ig
u
r
e
2
.
b
.
P
r
e
-
p
r
o
ce
s
s
in
g
No
is
y
,
r
ed
u
n
d
a
n
t,
in
co
m
p
le
te
an
d
d
ata
h
av
i
n
g
d
i
f
f
er
en
t
d
ata
t
y
p
es
is
o
b
s
er
v
ed
.
W
ith
o
u
t
s
tan
d
ar
d
izatio
n
t
h
e
p
r
o
ce
s
s
o
f
class
i
f
icatio
n
w
ill
b
e
h
a
m
p
er
e
d
.
Var
io
u
s
R
p
r
ep
r
o
ce
s
s
in
g
p
a
ck
ag
e
s
ar
e
ap
p
lied
to
eli
m
i
n
ate
m
is
s
in
g
r
ec
o
r
d
s
h
av
in
g
i
n
co
m
p
lete
d
ata
an
d
to
g
et
d
ata
in
u
n
i
f
o
r
m
f
o
r
m
.
c
.
P
r
in
cip
al
co
m
p
o
n
e
n
t a
n
al
y
s
is
Mo
s
t
o
f
th
e
f
ea
tu
r
e
s
i
n
t
h
e
N
SL
-
KD
D
d
ataset
lar
g
el
y
d
o
n
o
t
ac
co
u
n
t
f
o
r
m
o
s
t
o
f
t
h
e
v
a
r
ian
ce
i
n
th
e
r
es
u
lt
s
.
T
h
er
ef
o
r
e,
a
m
e
th
o
d
ca
lled
P
C
A
[
1
7
,
1
8
]
is
u
s
e
d
to
g
et
a
m
o
r
e
co
n
ci
s
e
d
atas
et
w
it
h
les
s
f
ea
t
u
r
es
th
at
ac
co
u
n
t
f
o
r
m
o
s
t
o
f
t
h
e
v
ar
ian
ce
i
n
th
e
d
ata.
T
h
e
P
C
A
m
eth
o
d
,
d
ev
elo
p
ed
b
y
Kar
l
P
er
s
o
n
in
1
9
0
1
,
u
s
e
s
a
n
o
r
th
o
g
o
n
al
tr
a
n
s
f
o
r
m
atio
n
a
n
d
co
n
v
er
ts
t
h
e
p
o
s
s
i
b
ly
co
r
r
elate
d
d
ata
to
li
n
ea
r
ly
u
n
co
r
r
ela
ted
d
ata
s
o
r
ted
in
ter
m
s
o
f
v
ar
y
in
g
d
eg
r
ee
s
o
f
co
n
tr
ib
u
tio
n
o
f
v
ar
ian
ce
to
th
e
f
i
n
al
r
esu
lt,
s
u
ch
t
h
at,
th
e
f
ir
s
t
co
m
p
o
n
e
n
t
e
x
p
lain
s
m
o
r
e
v
a
r
ian
ce
th
a
n
th
e
n
e
x
t
an
d
s
o
o
n
.
T
h
e
v
ar
ian
ce
s
ex
p
lai
n
ed
ar
e
ca
lcu
lated
b
y
s
q
u
ar
in
g
t
h
e
E
ig
e
n
v
alu
e
s
.
I
n
th
is
w
a
y
,
th
e
f
ir
s
t
k
co
m
p
o
n
en
t
s
ca
n
b
e
s
elec
ted
i
n
s
u
ch
a
w
a
y
t
h
at
t
h
ese
k
co
m
p
o
n
e
n
t
s
e
x
p
lain
m
o
s
t
o
f
t
h
e
v
ar
ian
ce
i
n
t
h
e
d
ata.
I
n
t
h
is
w
a
y
,
o
n
l
y
k
f
ea
t
u
r
es
ar
e
o
b
tain
ed
as
a
r
esu
lt,
w
it
h
o
u
t
m
u
c
h
c
h
an
g
e
i
n
t
h
e
v
ar
ian
ce
e
x
p
lai
n
ed
.
T
h
is
m
e
t
h
o
d
o
f
r
ed
u
ci
n
g
th
e
d
i
m
e
n
s
io
n
s
o
f
t
h
e
d
ata
h
elp
s
d
ata
v
is
u
al
izatio
n
an
d
also
m
ed
iates
s
o
m
e
o
f
th
e
h
ig
h
v
ar
i
an
ce
p
r
o
b
lem
s
o
cc
u
r
r
i
n
g
d
u
e
to
ex
ce
s
s
f
ea
t
u
r
es
h
av
i
n
g
litt
le
o
r
n
o
co
n
tr
ib
u
tio
n
to
th
e
r
esu
lt
s
.
d
.
C
ateg
o
r
ize
t
h
e
p
ac
k
ets
1)
Neu
r
al
n
et
w
o
r
k
:
P
r
e
-
p
r
o
ce
s
s
e
d
d
ata
is
d
i
v
id
ed
in
to
3
s
et
s
-
tr
ain
i
n
g
,
v
alid
atio
n
a
n
d
tes
t
s
e
ts
i
n
6
0
:2
0
:2
0
r
atio
.
Mo
d
el
is
tr
ain
ed
u
s
i
n
g
th
e
ab
o
v
e
m
eth
o
d
[
1
9
]
f
o
r
d
if
f
er
e
n
t
v
alu
e
s
o
f
h
id
d
en
la
y
er
d
ata
f
r
o
m
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
10
,
No
.
5
,
Octo
b
e
r
2
0
2
0
:
5
4
4
5
-
5452
5448
th
e
tr
ain
in
g
s
e
t
an
d
ac
cu
r
ac
y
w
as
te
s
ted
s
u
b
s
eq
u
e
n
tl
y
u
s
in
g
t
h
e
v
al
id
atio
n
s
et.
C
las
s
if
ica
tio
n
er
r
o
r
E=
M
-
Y
i
s
ca
lc
u
lated
u
s
i
n
g
v
alid
atio
n
s
et
w
h
er
e
M
i
s
th
e
e
x
p
ec
ted
o
u
tp
u
t
v
ec
to
r
tak
en
f
r
o
m
th
e
v
alid
atio
n
s
et
an
d
Y
is
t
h
e
co
m
p
u
ted
o
u
tp
u
t
r
es
u
l
ti
n
g
f
r
o
m
th
e
clas
s
i
f
icatio
n
(
Y=
W
*
X)
h
av
i
n
g
w
ei
g
h
t
W
an
d
in
p
u
t
X.
W
h
en
th
e
er
r
o
r
o
b
s
er
v
ed
is
lo
w
th
e
tr
ain
in
g
p
h
ase
en
d
s
.
T
h
e
en
ti
r
e
p
r
o
ce
s
s
is
r
ep
ea
ted
k
ti
m
es
(
k
f
o
ld
cr
o
s
s
v
alid
atio
n
)
[
2
0
]
f
o
r
d
if
f
er
en
t
r
an
d
o
m
l
y
s
elec
ted
d
ata
s
a
m
p
les
to
f
in
d
th
e
m
o
s
t
o
p
ti
m
u
m
v
al
u
e
f
o
r
h
id
d
en
la
y
er
e
n
s
u
r
in
g
t
h
at
th
e
m
o
d
el
w
o
n
t
o
v
er
f
it
t
h
e
n
et
w
o
r
k
.
Mo
d
el
s
o
b
s
er
v
ed
w
it
h
t
h
e
h
i
g
h
e
s
t
v
ali
d
atio
n
ac
cu
r
ac
y
ar
e
tak
e
n
a
n
d
test
ed
w
it
h
t
h
e
test
s
et.
Ag
ai
n
k
-
f
o
ld
cr
o
s
s
v
alid
atio
n
i
s
ap
p
lied
to
f
in
d
th
e
o
p
tim
u
m
v
al
u
e.
2)
B
ag
g
i
n
g
:
W
h
en
th
e
m
o
d
el
is
b
ag
g
ed
[
2
1
-
23]
,
s
ev
er
al
r
esam
p
les
o
f
th
e
d
ata
ar
e
tak
e
n
i
n
iter
atio
n
a
n
d
th
e
m
o
d
el
is
tr
ain
ed
o
n
th
es
e
s
a
m
p
les.
T
h
en
th
e
p
r
ed
ictio
n
s
ar
e
av
er
ag
ed
o
v
er
th
e
s
a
m
p
les
.
T
h
is
m
et
h
o
d
is
p
ar
ticu
lar
l
y
u
s
ef
u
l
w
h
e
n
t
h
e
m
o
d
el
h
as
a
lo
w
v
ar
ian
ce
a
s
it
h
elp
s
in
cr
ea
s
e
th
e
v
ar
ian
c
e
o
f
th
e
m
o
d
el
an
d
h
a
v
in
g
li
ttle
e
f
f
ec
t
o
n
t
h
e
b
ias.
T
h
is
is
d
o
n
e
w
it
h
t
h
e
r
an
d
o
m
f
o
r
ests
a
s
w
ell
a
s
w
it
h
t
h
e
lin
ea
r
SVM
m
o
d
el
to
co
m
p
ar
e
r
esu
lts
.
C
r
o
s
s
v
a
lid
atio
n
is
i
n
cl
u
d
ed
in
o
r
d
er
to
f
in
d
th
e
m
o
s
t o
p
ti
m
u
m
v
alu
e
[
2
4
-
26]
.
3)
A
d
d
itio
n
al
l
y
,
s
i
m
p
le
d
ec
is
io
n
tr
ee
s
w
it
h
d
i
f
f
er
e
n
t
co
m
p
lex
i
t
y
p
ar
a
m
eter
s
(
cp
)
(
cr
o
s
s
v
al
id
atio
n
)
ar
e
also
u
s
ed
to
co
m
p
ar
e
r
esu
lt
s
w
it
h
t
h
e
ab
o
v
e
m
o
d
els.
e
.
A
cc
u
r
ac
y
ca
lc
u
latio
n
an
d
c
o
m
p
ar
is
o
n
:
T
h
e
m
o
s
t o
p
ti
m
u
m
cla
s
s
i
f
ier
i
s
s
elec
ted
in
t
h
i
s
s
tep
.
Fig
u
r
e
2
.
Field
s
in
d
ata
s
et
4.
RE
SU
L
T
S
A
ND
AN
AL
Y
SI
S
T
ab
le
1
to
T
a
b
le
1
1
s
h
o
w
t
h
e
r
esu
lt
s
o
b
tain
ed
f
o
r
th
e
cla
s
s
i
f
i
er
s
u
s
ed
.
a.
Neu
r
al
n
et
w
o
r
k
T
h
e
T
ab
le
1
s
h
o
w
h
id
d
en
la
y
er
=
2
an
d
6
w
as
c
h
o
s
e
n
a
s
t
h
e
p
ar
a
m
eter
f
o
r
h
a
v
i
n
g
th
e
h
i
g
h
e
s
t
cr
o
s
s
v
alid
atio
n
ac
c
u
r
ac
y
an
d
t
h
e
m
o
d
el
w
as
test
ed
a
g
ain
,
w
it
h
t
h
e
test
s
e
t.
T
h
e
T
ab
le
2
s
h
o
w
h
id
d
en
la
y
er
=
2
is
f
o
u
n
d
to
b
e
id
ea
l f
o
r
th
is
d
atas
et.
Fin
al
A
cc
u
r
ac
y
(
tes
t set)
: 9
6
.
2
6
% u
s
in
g
h
id
d
en
la
y
er
=
2
.
T
ab
le
1
.
Neu
r
al
n
et
w
o
r
k
r
esu
l
t
s
(
c
r
o
s
s
v
alid
atio
n
)
H
i
d
d
e
n
L
a
y
e
r
s
A
c
c
u
r
a
c
y
(
C
r
o
ss V
a
l
i
d
a
t
i
o
n
)
1
9
0
.
6
%
2
9
7
.
6
7
%
3
9
6
.
3
4
%
4
9
7
.
0
1
%
5
9
7
.
3
4
%
6
9
7
.
6
7
%
7
9
7
.
0
0
%
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:
2
0
8
8
-
8708
C
o
mp
a
r
is
o
n
s
tu
d
y
o
f
ma
ch
in
e
lea
r
n
in
g
cla
s
s
ifier
s
to
d
etec
t a
n
o
ma
lies
(
N
i
s
h
a
P
.
S
h
etty)
5449
T
ab
le
2
.
Neu
r
al
n
et
w
o
r
k
r
esu
l
t
s
(
t
est
d
ata
s
et)
H
i
d
d
e
n
L
a
y
e
r
s
T
e
st
A
c
c
u
r
a
c
y
2
9
6
.
2
6
%
6
9
5
.
9
7
%
b.
P
r
in
cip
le
co
m
p
o
n
e
n
t a
n
al
y
s
is
3
8
f
ea
tu
r
es
ar
e
ta
k
en
a
f
ter
r
e
m
o
v
i
n
g
t
h
e
f
ac
to
r
s
.
First
6
c
o
m
p
o
n
en
t
s
ar
e
s
elec
ted
,
s
i
n
ce
th
e
y
h
a
v
e
th
e
h
i
g
h
est
v
ar
ian
ce
.
T
ab
le
3
.
P
C
A
I
S
t
a
n
d
a
r
d
d
e
v
i
a
t
i
o
n
P
r
o
p
o
r
t
i
o
n
o
f
V
a
r
i
a
n
c
e
C
u
m
u
l
a
t
i
v
e
P
r
o
p
o
r
t
i
o
n
P
C
1
1
.
5
4
3
e
+
0
4
7
.
7
8
3
e
-
01
7
.
7
8
3
e
-
01
P
C
2
8
0
2
5
.
9
5
3
3
0
.
2
1
0
6
0
.
9
8
8
8
P
C
3
1
.
8
3
3
e
+
0
3
1
.
0
9
8
e
-
02
9
.
9
9
8
e
-
01
P
C
4
1
8
0
.
0
3
3
1
3
0
.
0
0
0
1
1
0
.
9
9
9
9
2
P
C
5
1
2
1
.
5
6
6
2
8
0
.
0
0
0
0
5
0
.
9
9
9
9
7
P
C
6
9
4
.
5
1
1
5
1
0
.
0
0
0
0
3
1
.
0
0
0
0
0
P
C
7
2
6
.
0
1
0
.
0
0
1
.
0
0
P
C
8
3
.
9
7
3
0
.
0
0
0
1
.
0
0
0
P
C
9
0
.
8
0
2
7
0
.
0
0
0
0
1
.
0
0
0
0
P
C
1
0
0
.
5
0
9
6
0
.
0
0
0
0
1
.
0
0
0
0
P
C
1
1
0
.
4
5
2
4
0
.
0
0
0
0
1
.
0
0
0
0
P
C
1
2
0
.
4
3
3
5
0
.
0
0
0
0
1
.
0
0
0
0
T
ab
le
4
.
P
C
A
I
I
S
t
a
n
d
a
r
d
d
e
v
i
a
t
i
o
n
P
r
o
p
o
r
t
i
o
n
o
f
V
a
r
i
a
n
c
e
C
u
m
u
l
a
t
i
v
e
P
r
o
p
o
r
t
i
o
n
P
C
1
3
P
C
2
6
0
.
0
0
0
0
1
.
0
0
0
0
P
C
1
4
0
.
3
9
5
2
0
.
0
0
0
0
1
.
0
0
0
0
P
C
1
5
0
.
3
5
3
0
.
0
0
0
0
1
.
0
0
0
0
P
C
1
6
0
.
3
1
1
0
.
0
0
0
0
1
.
0
0
0
0
P
C
1
7
0
.
2
6
3
4
0
.
0
0
0
0
1
.
0
0
0
0
P
C
1
8
0
.
2
2
7
1
0
.
0
0
0
0
1
.
0
0
0
0
P
C
1
9
0
.
2
2
5
7
0
.
0
0
0
0
1
.
0
0
0
0
P
C
2
0
0
.
1
6
1
1
0
.
0
0
0
0
1
.
0
0
0
0
P
C
2
1
0
.
1
4
7
1
0
.
0
0
0
0
1
.
0
0
0
0
P
C
2
2
0
.
1
3
7
5
0
.
0
0
0
0
1
.
0
0
0
0
P
C
2
3
0
.
1
3
5
4
0
.
0
0
0
0
1
.
0
0
0
0
P
C
2
4
0
.
1
1
1
9
0
.
0
0
0
0
1
.
0
0
0
0
P
C
2
5
0
.
0
9
6
6
7
0
.
0
0
0
0
1
.
0
0
0
0
P
C
2
6
0
.
0
8
9
3
0
.
0
0
0
0
1
.
0
0
0
0
T
ab
le
5
.
P
C
A
I
I
I
S
t
a
n
d
a
r
d
d
e
v
i
a
t
i
o
n
P
r
o
p
o
r
t
i
o
n
o
f
V
a
r
i
a
n
c
e
C
u
m
u
l
a
t
i
v
e
P
r
o
p
o
r
t
i
o
n
P
C
2
7
0
.
0
8
6
9
6
0
.
0
0
0
0
1
.
0
0
0
0
P
C
2
8
0
.
0
7
9
4
5
0
.
0
0
0
0
1
.
0
0
0
0
P
C
2
9
0
.
0
6
6
8
8
0
.
0
0
0
0
1
.
0
0
0
0
P
C
3
0
0
.
0
5
3
7
7
0
.
0
0
0
0
1
.
0
0
0
0
P
C
3
1
0
.
0
4
7
6
9
0
.
0
0
0
0
1
.
0
0
0
0
P
C
3
2
0
.
0
3
9
8
8
0
.
0
0
0
0
1
.
0
0
0
0
P
C
3
3
0
.
0
3
7
1
0
.
0
0
0
0
1
.
0
0
0
0
P
C
3
4
0
.
0
2
9
0
8
0
.
0
0
0
0
1
.
0
0
0
0
P
C
3
5
0
.
0
2
9
0
8
0
.
0
0
0
0
1
.
0
0
0
0
P
C
3
6
0
.
0
1
8
3
2
0
.
0
0
0
0
1
.
0
0
0
0
P
C
3
7
1
.
5
0
6
e
-
12
0
.
0
0
0
e
+
0
0
1
.
0
0
0
e
+
0
0
P
C
3
8
1
.
5
0
6
e
-
12
0
.
0
0
0
e
+
0
0
1
.
0
0
0
e
+
0
0
c.
Fo
r
d
ec
is
io
n
tr
ee
w
it
h
P
C
A
T
h
e
f
i
n
al
v
al
u
e
u
s
ed
f
o
r
th
e
m
o
d
el
cp
=
0
.
0
5
8
3
0
1
6
5
.
A
cc
u
r
a
c
y
a
g
ai
n
s
t
th
e
test
d
ata
s
e
t
w
a
s
o
b
tain
ed
to
b
e
0
.
6
9
2
3
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
10
,
No
.
5
,
Octo
b
e
r
2
0
2
0
:
5
4
4
5
-
5452
5450
T
ab
le
6
.
Dec
is
io
n
tr
ee
F
o
r
C
r
o
ss V
a
l
i
d
a
t
i
o
n
D
a
t
a
S
e
t
cp
A
c
c
u
r
a
c
y
K
a
p
p
a
0
.
0
5
8
3
0
1
6
5
0
.
6
8
5
1
4
0
4
0
.
4
9
8
9
2
5
3
0
.
0
7
2
2
4
3
3
5
0
.
5
8
2
8
3
9
1
0
.
2
8
4
5
2
5
1
0
.
0
7
4
7
7
8
2
0
0
.
5
7
3
0
8
0
4
0
.
2
5
6
1
0
8
9
d.
B
ag
g
i
n
g
w
i
th
r
a
n
d
o
m
f
o
r
est
with
P
C
A
n
b
ag
g
=
3
0
is
s
elec
ted
.
T
h
e
f
in
al
ac
cu
r
ac
y
: 0
.
9
7
1
7
.
T
ab
le
7
.
R
an
d
o
m
f
o
r
est
F
o
r
C
r
o
ss V
a
l
i
d
a
t
i
o
n
D
a
t
a
S
e
t
n
b
a
g
g
A
c
c
u
r
a
c
y
10
0
.
9
4
9
7
15
0
.
9
5
6
0
20
0
.
9
5
6
0
25
0
.
9
6
2
3
30
0
.
9
7
1
7
e.
B
ag
g
i
n
g
w
i
th
SVM
w
it
h
P
C
A
T
h
e
f
in
al
v
al
u
e
u
s
ed
f
o
r
th
e
m
o
d
el
is
C
=
1
a
n
d
n
b
a
g
g
=
2
5
.
T
h
e
f
in
a
l
ac
cu
r
ac
y
f
o
r
tes
t
d
ata
=
0
.
8
6
4
1
9
2
2
f
o
r
C
=
1
an
d
n
b
ag
g
=
25.
T
ab
le
8
.
SVM
I
F
o
r
C
r
o
ss V
a
l
i
d
a
t
i
o
n
D
a
t
a
S
e
t
n
b
a
g
g
=
2
0
c
A
c
c
u
r
a
c
y
1
0
.
8
3
8
1
5
3
8
10
0
.
8
3
7
5
3
8
5
1
0
0
0
.
8
3
6
9
2
3
1
1
0
0
0
0
.
8
3
3
8
4
6
2
T
ab
le
9
.
SVM
I
I
F
o
r
C
r
o
ss V
a
l
i
d
a
t
i
o
n
D
a
t
a
S
e
t
n
b
a
g
g
=
1
0
c
A
c
c
u
r
a
c
y
1
0
.
8
3
8
7
6
9
2
10
0
.
8
3
8
2
5
3
4
1
0
0
0
.
8
3
5
7
9
3
4
1
0
0
0
0
.
8
3
0
2
5
8
3
T
ab
le
1
0
.
SVM
I
I
I
F
o
r
C
r
o
ss V
a
l
i
d
a
t
i
o
n
D
a
t
a
S
e
t
n
b
a
g
g
=
2
5
c
A
c
c
u
r
a
c
y
1
.
0
0
.
8
6
4
2
9
7
8
1
0
.
0
0
.
8
6
5
8
3
4
1
1
0
0
.
0
0
.
8
6
5
0
0
9
0
1
0
0
0
.
0
0
.
8
6
3
0
6
2
6
f.
Fin
al
ac
c
u
r
ac
y
T
ab
le
1
1
.
Fin
al
a
cc
u
r
ac
y
c
o
m
p
ar
is
o
n
C
l
a
ssi
f
i
e
r
A
c
c
u
r
a
c
y
(
i
n
p
e
r
c
e
n
t
)
N
u
e
r
a
l
N
e
t
w
o
r
k
(
H
i
d
d
e
n
L
a
y
e
r
=
2
)
9
6
.
2
6
R
a
n
d
o
m F
o
r
e
st
(
n
b
a
g
g
=
3
0
)
9
7
.
1
7
S
V
M
(
n
b
a
g
=
2
5
a
n
d
c
=
1
)
8
6
.
4
1
D
e
c
i
si
o
n
T
r
e
e
(
C
p
=
0
.
0
5
8
3
0
1
6
5
)
6
9
.
2
3
g.
Sig
n
i
f
ica
n
ce
o
f
r
esear
ch
1)
R
esear
ch
o
n
i
n
tr
u
s
io
n
d
ete
ctio
n
s
y
s
te
m
m
ak
e
s
s
ig
n
i
f
i
ca
n
t
co
n
tr
ib
u
tio
n
s
to
t
h
e
s
o
ciet
y
.
B
u
s
in
e
s
s
o
r
g
an
izatio
n
s
,
b
an
k
in
g
s
ec
to
r
s
,
d
ef
en
s
e
s
y
s
te
m
etc.
ca
n
d
ep
lo
y
s
u
ch
tec
h
n
iq
u
e
s
to
s
af
e
g
u
ar
d
th
eir
v
ital d
ata.
2)
Seco
n
d
l
y
,
t
h
is
r
esear
ch
m
a
k
es
a
s
ig
n
i
f
ica
n
t
co
n
tr
ib
u
tio
n
to
th
e
b
o
d
y
o
f
k
n
o
w
led
g
e
b
y
estab
li
s
h
in
g
m
et
h
o
d
s
w
h
ich
o
n
lear
n
i
n
g
i
d
ea
l
s
y
s
te
m
b
eh
a
v
io
r
f
r
o
m
p
ast
d
ata
d
is
co
v
er
s
t
h
e
p
atter
n
s
an
d
d
e
v
iatio
n
s
au
to
m
at
icall
y
,
w
h
ic
h
is
o
t
h
er
w
i
s
e
d
if
f
ic
u
lt to
d
etec
t.
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:
2
0
8
8
-
8708
C
o
mp
a
r
is
o
n
s
tu
d
y
o
f
ma
ch
in
e
lea
r
n
in
g
cla
s
s
ifier
s
to
d
etec
t a
n
o
ma
lies
(
N
i
s
h
a
P
.
S
h
etty)
5451
3)
Fin
all
y
,
co
m
p
ar
in
g
a
n
d
an
al
y
zin
g
t
h
e
ac
cu
r
ac
y
o
f
v
ar
io
u
s
clas
s
i
f
ier
s
a
n
d
f
i
n
d
in
g
th
e
m
o
s
t
s
u
itab
le
class
i
f
ier
(
s
u
b
j
ec
tiv
e
to
th
e
e
n
v
ir
o
n
m
e
n
t
i
n
w
h
i
c
h
t
h
e
s
y
s
te
m
i
s
d
ep
lo
y
ed
,
t
h
e
co
s
t
a
n
d
co
m
p
u
tatio
n
p
r
ec
in
cts
a
n
d
t
h
e
s
ec
u
r
it
y
lev
el
n
ec
e
s
s
ar
y
)
co
n
tr
ib
u
tes
co
n
s
id
er
ab
ly
to
t
h
e
t
h
eo
r
y
b
u
ild
in
g
f
o
r
u
p
g
r
ad
ed
s
y
s
te
m
d
es
ig
n
.
5.
CO
NCLU
SI
O
N
W
ith
t
h
e
p
r
o
f
u
s
io
n
in
t
h
e
u
s
a
g
e
o
f
I
n
ter
n
et
f
o
r
ap
p
licatio
n
s
s
u
c
h
a
s
e
-
co
m
m
er
ce
w
eb
s
ites
,
o
n
lin
e
b
an
k
i
n
g
etc.
p
r
o
tectio
n
o
f
cr
u
cial
in
f
o
r
m
atio
n
tr
av
ell
i
n
g
o
v
er
t
h
e
n
et
w
o
r
k
o
r
r
esid
in
g
in
h
o
s
t
m
ac
h
i
n
es
b
ec
o
m
e
s
cr
u
cial.
E
f
f
ec
ti
v
en
e
s
s
o
f
a
n
y
d
etec
tio
n
tec
h
n
iq
u
e
d
ep
en
d
s
o
n
t
h
e
t
y
p
e
a
n
d
b
eh
av
io
r
o
f
t
h
e
d
ata
i
n
th
e
s
y
s
te
m
,
t
h
e
e
n
v
ir
o
n
m
e
n
t
in
w
h
ich
th
e
s
y
s
te
m
is
d
e
p
lo
y
ed
,
th
e
t
y
p
e
o
f
a
n
o
m
a
li
es
an
d
attac
k
s
t
h
at
th
e
s
y
s
te
m
e
n
co
u
n
ter
s
,
t
h
e
co
s
t
an
d
co
m
p
u
tat
io
n
l
i
m
itat
io
n
s
a
s
s
i
g
n
ed
f
o
r
t
h
e
p
ar
tic
u
lar
o
p
er
atio
n
a
n
d
th
e
s
ec
u
r
it
y
le
v
el
r
eq
u
ir
ed
.
Fi
r
e
w
all
s
ac
t
as
a
f
e
n
ce
ar
o
u
n
d
th
e
o
r
g
an
izatio
n
’
s
n
e
t
w
o
r
k
b
u
t
d
o
n
o
t
p
r
o
v
id
e
p
r
o
tectio
n
f
r
o
m
i
n
s
id
er
attac
k
s
.
User
a
u
t
h
en
t
icatio
n
m
et
h
o
d
s
ar
e
co
s
tlier
i
n
ter
m
s
o
f
eq
u
i
p
m
e
n
t
an
d
f
ail
s
i
f
th
e
s
ec
r
et
k
e
y
w
h
ic
h
au
t
h
e
n
tic
ates th
e
p
er
s
o
n
i
s
leak
ed
.
T
h
u
s
,
th
er
e
is
a
m
a
m
m
o
th
n
ee
d
f
o
r
a
d
etec
tio
n
s
y
s
te
m
w
h
ic
h
ca
n
ca
te
g
o
r
ize
an
y
p
ac
k
et
ac
c
u
r
atel
y
as
n
o
r
m
al
o
r
in
tr
u
s
i
v
e
i
n
r
ea
l
t
i
m
e
w
it
h
o
u
t
h
a
v
in
g
to
r
el
y
o
n
an
y
d
atab
ase
an
d
b
ein
g
m
ed
d
led
b
y
an
y
at
tack
er
.
H
y
b
r
id
m
e
th
o
d
s
en
co
m
p
as
s
i
n
g
a
co
m
b
i
n
atio
n
o
f
s
i
g
n
at
u
r
e
b
ased
an
d
an
o
m
al
y
b
ased
d
e
tectio
n
ca
n
b
e
i
m
p
le
m
e
n
ted
in
f
u
t
u
r
e
to
o
f
f
er
r
ea
l
ti
m
e
d
etec
tio
n
w
it
h
g
o
o
d
d
etec
tio
n
r
ate.
RE
F
E
R
E
NC
E
S
[1
]
N.
P
.
S
h
e
tt
y
,
“
A
S
u
rv
e
y
o
n
A
re
a
s
in
Da
ta
M
in
in
g
f
o
r
In
tru
sio
n
De
tec
ti
o
n
,”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Da
ta
M
in
i
n
g
a
n
d
Kn
o
wled
g
e
En
g
in
e
e
rin
g
, v
o
l
.
7
,
n
o
.
1
,
p
p
.
1
-
3
,
2
0
1
5
.
[2
]
N.
P
.
S
h
e
tt
y
,
“
Us
in
g
c
lu
ste
rin
g
t
o
c
a
p
t
u
re
a
tt
a
c
k
e
rs
,”
in
2
0
1
6
I
n
t
e
rn
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
I
n
v
e
n
ti
v
e
Co
mp
u
t
a
ti
o
n
T
e
c
h
n
o
l
o
g
ies
(
ICICT
)
,
p
p
.
1
-
5
,
2
0
1
6
.
[3
]
A
.
S
h
e
n
f
ield
,
e
t
a
l.
,
“
In
telli
g
e
n
t
i
n
tru
si
o
n
d
e
tec
ti
o
n
sy
ste
m
s
u
sin
g
a
rti
f
icia
l
n
e
u
ra
l
n
e
t
w
o
rk
s
,”
ICT
Exp
re
ss
,
v
o
l.
4
,
n
o
.
2
,
p
p
.
9
5
-
99
,
2
0
1
8
.
[4
]
O.
De
p
re
n
,
e
t
a
l.
,
“
A
n
in
tell
ig
e
n
t
in
tru
si
o
n
d
e
tec
ti
o
n
sy
ste
m
(
I
DS
)
f
o
r
a
n
o
m
a
l
y
a
n
d
m
isu
se
d
e
tec
ti
o
n
i
n
c
o
m
p
u
ter
n
e
tw
o
rk
s
,”
Exp
e
rt S
y
st
e
ms
wit
h
A
p
p
l
ic
a
ti
o
n
s
,
v
o
l
.
2
9
,
n
o
.
4
,
p
p
.
7
1
3
-
7
2
2
,
2
0
0
5
.
[5
]
U
.
Ca
v
u
so
g
lu
,
“
A
n
e
w
h
y
b
rid
a
p
p
ro
a
c
h
f
o
r
in
tru
si
o
n
d
e
tec
ti
o
n
u
sin
g
m
a
c
h
in
e
lea
rn
in
g
m
e
t
h
o
d
s
,”
Ap
p
li
e
d
In
telli
g
e
n
c
e
,
v
o
l.
4
9
,
n
o
.
2
,
2
0
1
9
.
[6
]
S.
K
.
Ka
n
g
,
e
t
a
l.
,
“
A
n
im
p
le
m
e
n
tatio
n
o
f
h
iera
rc
h
ica
l
i
n
tru
si
o
n
d
e
tec
ti
o
n
sy
ste
m
s
u
sin
g
sn
o
r
t
a
n
d
f
e
d
e
ra
ted
d
a
tab
a
se
s
,
”
in
2
0
1
8
1
7
t
h
IEE
E
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
On
T
ru
st,
S
e
c
u
rity
A
n
d
Priva
c
y
In
Co
mp
u
t
in
g
A
n
d
Co
mm
u
n
ica
ti
o
n
s/1
2
th
IEE
E
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
On
Bi
g
Da
ta
S
c
ie
n
c
e
An
d
En
g
in
e
e
rin
g
(
T
ru
st
-
Co
m/Bi
g
D
a
ta
S
E)
,
p
p
.
1
5
2
1
-
1
5
2
5
,
2
0
1
8
.
[7
]
M
.
Ba
y
k
a
ra
a
n
d
R.
Da
s
,
“
A
n
o
v
e
l
h
o
n
e
y
p
o
t
b
a
se
d
se
c
u
rit
y
a
p
p
r
o
a
c
h
f
o
r
re
a
l
-
ti
m
e
in
tru
sio
n
d
e
tec
ti
o
n
a
n
d
p
re
v
e
n
ti
o
n
sy
ste
m
s
,”
J
o
u
rn
a
l
o
f
I
n
fo
rm
a
t
io
n
S
e
c
u
rity
a
n
d
A
p
p
l
ica
ti
o
n
s
,
v
o
l
.
4
1
,
p
p
.
1
0
3
-
1
1
6
,
2
0
1
8
.
[8
]
S
.
Zh
a
o
,
e
t
a
l.
,
“
A
d
ime
n
sio
n
re
d
u
c
ti
o
n
m
o
d
e
l
a
n
d
c
las
si
f
ier
f
o
r
a
n
o
m
a
l
y
-
b
a
s
e
d
in
tru
sio
n
d
e
tec
ti
o
n
in
in
tern
e
t
o
f
th
in
g
s
,
”
i
n
2
0
1
7
IE
EE
1
5
t
h
I
n
t
l
Co
n
f
o
n
De
p
e
n
d
a
b
le,
Au
t
o
n
o
mic
a
n
d
S
e
c
u
re
C
o
mp
u
ti
n
g
,
1
5
t
h
In
t
l
Co
n
f
o
n
Per
v
a
siv
e
In
telli
g
e
n
c
e
a
n
d
C
o
mp
u
ti
n
g
,
3
r
d
In
tl
C
o
n
f
o
n
Bi
g
Da
t
a
In
telli
g
e
n
c
e
a
n
d
Co
m
p
u
t
in
g
a
n
d
Cy
b
e
r
S
c
ien
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
C
o
n
g
re
ss
(
DAS
C/
Pi
Co
m/Da
t
a
Co
m/Cy
b
e
rS
c
i
T
e
c
h
)
,
p
p
.
8
3
6
-
8
4
3
,
2
0
1
7
.
[9
]
S
.
S
i
n
g
h
a
n
d
S
.
S
i
lak
a
ri
,
“
A
n
e
n
s
e
m
b
le
a
p
p
ro
a
c
h
f
o
r
c
y
b
e
r
-
a
tt
a
c
k
d
e
tec
t
io
n
sy
ste
m
:
A
g
e
n
e
ric
f
ra
m
e
w
o
rk
,”
in
2
0
1
3
1
4
t
h
ACIS
In
ter
n
a
ti
o
n
a
l
Co
n
f
e
r
e
n
c
e
o
n
S
o
ft
wa
re
En
g
in
e
e
rin
g
,
Arti
fi
c
i
a
l
In
telli
g
e
n
c
e
,
Ne
two
rk
in
g
a
n
d
Pa
ra
ll
e
l/
Distri
b
u
te
d
Co
m
p
u
t
in
g
,
p
p
.
7
9
-
84
,
2
0
1
3
.
[1
0
]
M
.
S
.
Ho
q
u
e
,
e
t
a
l.
,
“
A
n
im
p
le
m
e
n
tatio
n
o
f
in
tru
sio
n
d
e
tec
ti
o
n
s
y
ste
m
u
sin
g
g
e
n
e
ti
c
a
lg
o
rit
h
m
,”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Ne
two
rk
S
e
c
u
rity
a
n
d
I
ts A
p
p
li
c
a
ti
o
n
s
,
v
o
l.
4
,
n
o
.
2
,
p
p
.
1
0
9
-
1
2
0
,
2
0
1
2
.
[1
1
]
Y.
L
in
,
e
t
a
l.
,
“
T
h
e
d
e
sig
n
a
n
d
im
p
le
m
e
n
tatio
n
o
f
h
o
st
-
b
a
se
d
in
tru
si
o
n
d
e
tec
ti
o
n
sy
ste
m
,
”
in
2
0
1
0
T
h
ird
In
ter
n
a
t
io
n
a
l
S
y
mp
o
si
u
m o
n
I
n
tel
li
g
e
n
t
I
n
f
o
rm
a
ti
o
n
T
e
c
h
n
o
lo
g
y
a
n
d
S
e
c
u
rity I
n
fo
rm
a
ti
c
s
,
p
p
.
5
9
5
-
5
9
8
,
2
0
1
0
.
[1
2
]
F.
Y.
L
e
u
a
n
d
I
.
L
.
L
in
,
“
A
DO
S
/
DD
OS
a
tt
a
c
k
d
e
tec
ti
o
n
sy
ste
m
u
sin
g
c
h
i
-
sq
u
a
re
sta
ti
stic
a
p
p
r
o
a
c
h
,”
S
y
ste
mic
s,
Cy
b
e
rn
e
ti
c
s a
n
d
In
f
o
rm
a
ti
c
s
,
v
o
l.
8
,
n
o
.
2
,
p
p
.
4
1
-
5
1
,
2
0
1
0
.
[1
3
]
J.
S
e
o
,
“
A
n
a
tt
a
c
k
c
las
si
f
ic
a
ti
o
n
m
e
c
h
a
n
is
m
b
a
se
d
o
n
m
u
lt
ip
l
e
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
s
,
”
in
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
Co
m
p
u
t
a
ti
o
n
a
l
S
c
i
e
n
c
e
a
n
d
Its
A
p
p
li
c
a
ti
o
n
s
,
p
p
.
9
4
-
103
,
2
0
0
7
.
[1
4
]
S
.
M
u
k
k
a
m
a
la,
e
t
a
l.
,
“
In
tru
sio
n
d
e
tec
ti
o
n
u
si
n
g
n
e
u
ra
l
n
e
tw
o
rk
s
a
n
d
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
s
,
”
in
Pro
c
e
e
d
in
g
s
o
f
th
e
2
0
0
2
In
ter
n
a
ti
o
n
a
l
J
o
in
t
Co
n
f
e
re
n
c
e
o
n
Ne
u
ra
l
Ne
two
rk
s.
IJ
CNN’0
2
,
v
o
l.
2
,
p
p
.
1
7
0
2
-
1
7
0
7
,
2
0
0
2
.
[1
5
]
“
KD
D Cu
p
1
9
9
9
Da
ta
,
”
1
9
9
9
.
[
O
n
li
n
e
]
,
A
v
a
il
a
b
le
:
h
tt
p
:/
/k
d
d
.
ics
.
u
c
i.
e
d
u
/d
a
ta
b
a
se
s/k
d
d
c
u
p
9
9
/k
d
d
c
u
p
9
9
.
h
tm
l
.
[1
6
]
M
.
T
a
v
a
ll
a
e
e
,
e
t
a
l.
,
“
A
d
e
taile
d
a
n
a
l
y
sis
o
f
th
e
KD
D
CUP
9
9
d
a
ta
se
t
,”
i
n
Pro
c
e
e
d
i
n
g
s
o
f
th
e
2
0
0
9
I
EE
E
S
y
mp
o
si
u
m
o
n
Co
m
p
u
t
a
ti
o
n
a
l
I
n
t
e
ll
ig
e
n
c
e
in
S
e
c
u
rity a
n
d
D
e
fen
se
A
p
p
li
c
a
ti
o
n
s (
CIS
DA’0
9
)
,
p
p
.
53
-
58
,
2
0
0
9
.
[1
7
]
M
.
Bre
m
s,
“
A
On
e
-
S
to
p
S
h
o
p
f
o
r
P
ri
n
c
ip
a
l
Co
m
p
o
n
e
n
t
An
a
ly
sis
,
”
M
e
d
iu
m
,
2
0
1
7
.
[
On
li
n
e
]
,
A
v
a
il
a
b
le
:
h
tt
p
s:/
/t
o
w
a
rd
sd
a
tas
c
ien
c
e
.
c
o
m
/a
-
one
-
sto
p
-
sh
o
p
-
f
o
r
-
p
ri
n
c
ip
a
l
-
c
o
m
p
o
n
e
n
t
-
a
n
a
ly
sis
-
5
5
8
2
f
b
7
e
0
a
9
c
.
[1
8
]
S
.
M
.
A
lma
n
so
b
a
n
d
S
.
S
.
L
o
m
te,
“
A
d
d
re
ss
in
g
c
h
a
ll
e
n
g
e
s
f
o
r
in
tru
sio
n
d
e
tec
ti
o
n
sy
ste
m
u
sin
g
n
a
iv
e
Ba
y
e
s
a
n
d
P
CA
a
lg
o
rit
h
m
,
”
2
0
1
7
2
n
d
I
n
ter
n
a
ti
o
n
a
l
C
o
n
fer
e
n
c
e
fo
r
C
o
n
v
e
rg
e
n
c
e
in
T
e
c
h
n
o
lo
g
y
(
I2
CT
)
,
M
u
m
b
a
i,
p
p
.
5
6
5
-
568
,
2
0
1
7
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
10
,
No
.
5
,
Octo
b
e
r
2
0
2
0
:
5
4
4
5
-
5452
5452
[1
9
]
G
.
Ka
r
a
tas
a
n
d
O.
K.
S
a
h
in
g
o
z
,
“
Ne
u
ra
l
n
e
tw
o
rk
b
a
se
d
in
tru
sio
n
d
e
tec
ti
o
n
sy
ste
m
s
w
it
h
d
iff
e
r
e
n
t
train
in
g
f
u
n
c
ti
o
n
s,
”
2
0
1
8
6
th
In
ter
n
a
ti
o
n
a
l
S
y
mp
o
si
u
m o
n
Di
g
it
a
l
F
o
re
n
sic
a
n
d
S
e
c
u
rity (
IS
DFS
)
,
A
n
taly
a
,
p
p
.
1
-
6
,
2
0
1
8
.
[2
0
]
J.
Bro
w
n
lee
,
“
A
G
e
n
tl
e
In
tro
d
u
c
ti
o
n
t
o
k
-
f
o
ld
Cro
ss
-
V
a
li
d
a
ti
o
n
,
”
M
a
c
h
in
e
L
e
a
r
n
i
n
g
M
a
ste
ry
,
2
0
1
9
.
[
O
n
li
n
e
],
Av
a
il
a
b
le
:
h
tt
p
s://
m
a
c
h
in
e
lea
rn
in
g
m
a
ste
r
y
.
c
o
m
/k
-
f
o
ld
-
c
ro
ss
-
v
a
li
d
a
ti
o
n
/
.
[2
1
]
Ud
a
c
it
y
,
“
Bo
o
tstrap
a
g
g
re
g
a
ti
n
g
b
a
g
g
in
g
,
”
[
o
n
li
n
e
]
.
A
v
a
il
a
b
le
:
h
tt
p
s://
ww
w
.
y
o
u
tu
b
e
.
c
o
m
/wa
tch
?
v
=
2
M
g
8
QD
0
F
1
d
Q
[2
2
]
M
.
Zare
a
p
o
o
r
a
n
d
P
.
S
h
a
m
so
lmo
a
li
,
“
A
p
p
li
c
a
ti
o
n
o
f
Cre
d
it
Ca
rd
F
ra
u
d
De
tec
ti
o
n
:
Ba
se
d
o
n
Ba
g
g
in
g
En
se
m
b
le
Clas
sif
ier,
”
Pro
c
e
d
ia
C
o
mp
u
ter
S
c
ien
c
e
,
v
ol
.
4
8
,
pp.
6
7
9
-
6
8
5
,
2
0
1
5
.
[2
3
]
M
.
H
.
D
.
M
.
Ri
b
e
iro
a
n
d
L
.
d
o
s
S
.
Co
e
lh
o
,
“
En
se
m
b
le
a
p
p
ro
a
c
h
b
a
se
d
o
n
b
a
g
g
in
g
,
b
o
o
sti
n
g
a
n
d
sta
c
k
in
g
f
o
r
sh
o
rt
-
term
p
re
d
ictio
n
i
n
a
g
rib
u
si
n
e
ss
ti
m
e
se
ries
,
”
Ap
p
li
e
d
S
o
ft
Co
mp
u
ti
n
g
,
v
ol
.
86,
p
p
.
1
0
5
8
3
7
,
2
0
2
0
.
[2
4
]
M
.
Be
lo
u
c
h
,
e
t
a
l.
,
“
P
e
rf
o
rm
a
n
c
e
e
v
a
lu
a
ti
o
n
o
f
in
tru
si
o
n
d
e
tec
ti
o
n
b
a
se
d
o
n
m
a
c
h
in
e
lea
rn
in
g
u
sin
g
A
p
a
c
h
e
S
p
a
rk
,
”
Pro
c
e
d
ia
C
o
mp
u
ter
S
c
ien
c
e
,
v
ol
.
1
2
7
,
p
p
.
1
-
6,
2
0
1
8
.
[2
5
]
F.
A.
Kh
a
n
a
n
d
A.
G
u
m
a
e
i
,
“
A
Co
m
p
a
ra
ti
v
e
S
tu
d
y
o
f
M
a
c
h
in
e
L
e
a
rn
in
g
Clas
si
f
iers
f
o
r
Ne
t
w
o
rk
In
tru
sio
n
De
tec
ti
o
n
,
”
In
ter
n
a
ti
o
n
a
l
C
n
fer
e
n
c
e
o
n
Arti
f
icia
l
I
n
telli
g
e
n
c
e
a
n
d
S
e
c
u
rity
,
p
p
.
7
5
-
86
,
2
0
1
9
.
[2
6
]
S
.
Zw
a
n
e
,
e
t
a
l.
,
“
P
e
rf
o
r
m
a
n
c
e
A
n
a
l
y
sis
o
f
M
a
c
h
in
e
L
e
a
rn
in
g
Clas
sif
i
e
rs
f
o
r
In
tru
sio
n
De
tec
ti
o
n
,
”
2
0
1
8
In
ter
n
a
t
io
n
a
l
Co
n
fer
e
n
c
e
o
n
In
telli
g
e
n
t
a
n
d
I
n
n
o
v
a
ti
v
e
Co
mp
u
ti
n
g
A
p
p
l
ica
ti
o
n
s
(
ICONIC)
,
P
l
a
in
e
M
a
g
n
ien
,
p
p
.
1
-
5
,
2
0
1
8
.
B
I
O
G
RAP
H
I
E
S
O
F
AUTH
O
RS
Nish
a
P.
S
h
e
tty
h
a
s
p
u
b
li
sh
e
d
in
th
e
a
re
a
s
n
e
tw
o
rk
se
c
u
rit
y
a
n
d
m
a
c
h
in
e
lea
rn
in
g
.
C
u
rre
n
tl
y
sh
e
is w
o
rk
in
g
in
th
e
a
re
a
o
f
so
c
ial
n
e
tw
o
rk
se
c
u
rit
y
.
J
a
y
a
sh
r
e
e
is
c
u
rre
n
tl
y
w
o
rk
in
g
in
m
a
c
h
in
e
lea
rn
in
g
d
o
m
a
in
s.
S
h
e
is
in
tere
ste
d
in
e
x
p
lo
ri
n
g
th
e
d
o
m
a
in
o
f
m
a
c
h
in
e
lea
rn
in
g
a
n
d
a
n
a
ly
ti
c
s in
m
e
d
ica
l
re
a
l
m
.
Ro
h
il
Na
r
u
l
a
is i
n
tere
ste
d
in
a
re
a
s li
k
e
m
a
c
h
in
e
lea
rn
in
g
a
n
d
a
n
a
ly
ti
c
s.
K
u
sh
a
g
r
a
Ta
n
d
o
n
is i
n
tere
ste
d
i
n
a
re
a
s li
k
e
m
a
c
h
in
e
lea
rn
in
g
a
n
d
a
n
a
ly
ti
c
s.
Evaluation Warning : The document was created with Spire.PDF for Python.