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
.
6
,
Decem
b
er
20
25
,
p
p
.
6
0
4
3
~
6
0
5
2
I
SS
N:
2088
-
8
7
0
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijece.
v
15
i
6
.
pp
6
0
4
3
-
6
0
5
2
6043
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m
Neura
l
-
ne
twork
b
a
sed represen
tati
o
n f
ra
mewo
rk
for
adv
ersa
ry
identif
i
ca
tion in
i
nterne
t
of
things
T
ha
nu
j
a
Na
ra
s
im
ha
m
urt
hy
,
G
un
a
v
a
t
hi H
o
s
a
ha
lli Swa
my
D
e
p
a
r
t
me
n
t
o
f
C
o
mp
u
t
e
r
S
c
i
e
n
c
e
a
n
d
En
g
i
n
e
e
r
i
n
g
,
B
a
n
g
a
l
o
r
e
I
n
st
i
t
u
t
e
o
f
T
e
c
h
n
o
l
o
g
y
,
B
e
n
g
a
l
u
r
u
,
a
f
f
i
l
i
a
t
e
t
o
V
i
s
v
e
s
v
a
r
a
y
a
Te
c
h
n
o
l
o
g
i
c
a
l
U
n
i
v
e
r
si
t
y
,
B
e
l
a
g
a
v
i
,
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
Au
g
1
,
2
0
2
4
R
ev
is
ed
Ma
r
1
8
,
2
0
2
5
Acc
ep
ted
Ma
y
2
3
,
2
0
2
5
M
a
c
h
in
e
lea
rn
in
g
is
o
n
e
o
f
th
e
p
o
ten
ti
a
l
so
l
u
ti
o
n
s
t
o
wa
rd
s
o
p
ti
m
izin
g
th
e
se
c
u
rit
y
stre
n
g
th
to
wa
r
d
s
id
e
n
ti
fy
in
g
c
o
m
p
lex
fo
rm
s
o
f
t
h
re
a
ts
in
i
n
tern
e
t
of
th
in
g
s
(Io
T).
Ho
we
v
e
r,
a
re
v
i
e
w
o
f
e
x
isti
n
g
m
a
c
h
i
n
e
lea
rn
in
g
-
b
a
se
d
a
p
p
ro
a
c
h
e
s
sh
o
wc
a
se
s
th
e
ir
su
b
-
o
p
ti
m
a
l
p
e
rfo
rm
a
n
c
e
wh
e
n
e
x
p
o
se
d
to
d
y
n
a
m
ic
f
o
rm
s
o
f
u
n
se
e
n
t
h
re
a
ts
with
o
u
t
a
n
y
a
p
ri
o
ri
in
f
o
rm
a
ti
o
n
d
u
ri
n
g
t
h
e
train
in
g
sta
g
e
.
He
n
c
e
,
th
is
m
a
n
u
sc
rip
t
p
re
se
n
ts
a
n
o
v
e
l
m
a
c
h
in
e
-
lea
rn
in
g
fra
m
e
wo
rk
to
wa
rd
s
p
o
ten
ti
a
l
th
re
a
t
d
e
tec
ti
o
n
c
a
p
a
b
le
o
f
id
e
n
ti
fy
in
g
th
e
u
n
d
e
rl
y
i
n
g
p
a
tt
e
rn
s
o
f
ra
p
id
l
y
e
v
o
lv
i
n
g
th
re
a
ts.
Th
e
p
ro
p
o
se
d
sy
st
e
m
u
se
s
a
n
e
u
ra
l
n
e
two
r
k
-
b
a
se
d
lea
rn
in
g
m
o
d
e
l
e
m
p
h
a
siz
in
g
re
p
re
se
n
tati
o
n
lea
rn
in
g
wh
e
re
a
n
e
x
p
li
c
it
m
a
sk
e
d
in
d
e
x
i
n
g
m
e
c
h
a
n
ism
is
p
re
se
n
ted
f
o
r
h
ig
h
-
lev
e
l
se
c
u
rit
y
a
g
a
in
st
u
n
k
n
o
wn
a
n
d
d
y
n
a
m
ic
a
d
v
e
rsa
ries
.
Th
e
b
e
n
c
h
m
a
rk
e
d
o
u
tco
m
e
o
f
th
e
st
u
d
y
sh
o
ws
to
a
c
c
o
m
p
li
sh
1
1
%
m
a
x
imi
z
e
d
th
re
a
t
d
e
tec
ti
o
n
a
c
c
u
ra
c
y
a
n
d
3
3
%
m
in
imiz
e
d
a
lg
o
rit
h
m
p
ro
c
e
ss
in
g
ti
m
e
.
K
ey
w
o
r
d
s
:
Dy
n
am
ic
t
h
r
ea
ts
I
n
d
ex
in
g
I
n
ter
n
et
of
th
in
g
s
Ma
ch
in
e
lear
n
in
g
Neu
r
al
n
etwo
r
k
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
:
T
h
an
u
ja
Nar
asimh
am
u
r
th
y
C
o
m
p
u
ter
Scien
ce
an
d
E
n
g
i
n
e
er
in
g
,
B
an
g
alo
r
e
I
n
s
titu
te
o
f
T
ec
h
n
o
lo
g
y
af
f
iliate
to
Vis
v
esv
ar
ay
a
T
ec
h
n
o
lo
g
ical
Un
iv
er
s
ity
Vis
h
wesh
war
ap
u
r
a,
B
asav
an
ag
u
d
i,
B
en
g
alu
r
u
,
Kar
n
atak
a
5
6
0
0
0
4
,
I
n
d
ia
E
m
ail:
n
th
an
u
ja@
b
it
-
b
an
g
alo
r
e.
ed
u
.
in
1.
I
NT
RO
D
UCT
I
O
N
I
n
ter
n
et
of
t
h
in
g
s
(
I
o
T
)
co
m
p
r
is
es
v
ar
io
u
s
f
o
r
m
s
o
f
s
o
f
twa
r
e
an
d
h
ar
d
war
e
c
o
n
n
ec
te
d
i
n
a
lar
g
er
co
n
tex
t
to
d
is
s
em
in
ate
an
d
a
n
aly
ze
d
ata
f
r
o
m
v
ar
io
u
s
ap
p
licatio
n
p
er
s
p
ec
tiv
e
s
[
1
]
.
I
r
r
e
s
p
ec
tiv
e
o
f
v
ar
io
u
s
r
elate
d
co
n
ce
r
n
s
,
s
ec
u
r
ity
is
s
u
es
d
o
m
in
ate
th
e
r
esear
c
h
co
m
m
u
n
ities
,
ev
en
if
a
m
o
r
e
ex
te
n
s
iv
e
s
et
o
f
s
ec
u
r
ity
s
o
lu
tio
n
s
is
b
ein
g
in
tr
o
d
u
ce
d
f
o
r
I
o
T
[
2
]
.
Fr
o
m
th
e
p
er
s
p
ec
tiv
e
o
f
I
o
T
s
ec
u
r
ity
,
v
ar
i
o
u
s
ch
allen
g
es
a
r
e
ass
o
ciate
d
with
p
er
f
o
r
m
in
g
f
r
eq
u
e
n
t
s
et
s
o
f
o
p
er
atio
n
s
,
e.
g
.
,
d
ata
en
cr
y
p
tio
n
,
ac
ce
s
s
co
n
tr
o
l,
f
ir
m
war
e
u
p
d
ates,
an
d
d
ev
ice
au
t
h
en
tica
tio
n
[
3
]
.
So
m
e
k
e
y
s
ec
u
r
ity
is
s
u
es in
I
o
T
ar
e
m
ain
ly
ass
o
ciate
d
with
s
ca
lab
ilit
y
,
lim
ited
r
eso
u
r
ce
s
,
d
iv
er
s
e
ec
o
s
y
s
tem
s
,
p
r
iv
ac
y
co
n
ce
r
n
s
,
an
d
p
h
y
s
ical
s
ec
u
r
ity
[
4
]
.
T
h
er
e
is
an
ev
o
lv
in
g
lan
d
s
ca
p
e
o
f
cy
b
er
th
r
ea
ts
in
I
o
T
,
wh
ic
h
ar
e
co
n
s
is
ten
tl
y
b
ein
g
r
esear
ch
ed
[
5
]
.
Ou
t
o
f
v
ar
io
u
s
s
ec
u
r
ity
s
o
lu
tio
n
s
,
ar
tific
ial
in
tellig
e
n
ce
(
AI
)
h
as
ev
e
n
tu
ally
p
lay
e
d
a
r
o
b
u
s
t
s
o
lu
tio
n
to
d
ea
l
with
th
is
co
m
p
lex
s
ec
u
r
ity
p
r
o
b
lem
with
v
a
r
io
u
s
ev
o
lv
in
g
m
ac
h
in
e
lear
n
in
g
al
g
o
r
ith
m
s
[
6
]
–
[
9
]
.
E
x
is
tin
g
s
tu
d
ies
h
av
e
n
o
ted
th
e
ad
o
p
tio
n
o
f
AI
in
an
o
m
aly
d
etec
tio
n
[
1
0
]
,
ad
ap
tiv
e
s
ec
u
r
ity
[
1
1
]
,
a
u
to
m
ated
th
r
ea
t
r
esp
o
n
s
e
[
1
2
]
,
a
n
d
p
r
ed
ictiv
e
ca
p
a
b
ilit
ies
[
1
3
]
.
Ho
wev
er
,
th
e
r
e
ar
e
r
is
in
g
co
n
c
er
n
s
ab
o
u
t
ad
o
p
tin
g
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
in
I
o
T
s
ec
u
r
ity
,
to
o
.
T
h
e
p
r
im
ar
y
lim
itatio
n
is
ass
o
ciate
d
with
th
e
d
ep
e
n
d
en
c
y
o
f
m
ass
iv
e
d
ata
v
o
lu
m
es
f
o
r
m
ac
h
in
e
lear
n
i
n
g
m
o
d
el
s
wh
i
ch
is
a
r
is
in
g
co
n
ce
r
n
t
o
war
d
s
s
en
s
itiv
ity
an
d
d
ata
p
r
iv
ac
y
.
T
h
is
p
r
o
b
lem
is
also
lin
k
ed
to
th
e
in
cr
ea
s
in
g
r
eq
u
ir
em
en
t
f
o
r
tr
ain
i
n
g
d
ata
o
f
h
ig
h
er
q
u
ality
an
d
m
o
d
el
c
o
m
p
le
x
ity
.
Fu
r
th
er
,
th
e
r
e
is
alwa
y
s
th
e
r
is
k
o
f
o
v
er
f
itti
n
g
th
er
e
b
y
r
estrictin
g
th
e
ca
p
ab
ilit
y
o
f
th
e
alg
o
r
ith
m
to
ad
ap
t
f
o
r
f
lu
ctu
atin
g
d
ata.
Ho
wev
er
,
o
n
e
n
o
ta
b
le
p
r
o
b
lem
th
at
is
n
o
r
m
ally
o
v
e
r
s
ee
n
b
y
th
e
m
aj
o
r
ity
o
f
e
x
is
tin
g
s
tu
d
ies
is
th
at
o
n
ce
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
.
6
,
Decem
b
e
r
20
25
:
6
0
4
3
-
6
0
5
2
6044
th
e
d
ataset
is
tr
ain
ed
,
th
e
lea
r
n
in
g
m
o
d
el
c
o
m
p
letely
d
ep
e
n
d
s
o
n
it
f
o
r
its
p
r
e
d
ictiv
e
an
al
y
s
is
.
Un
d
er
s
u
ch
a
s
ce
n
ar
io
,
if
th
er
e
is
an
y
f
o
r
m
o
f
d
e
v
iatio
n
in
th
e
n
etwo
r
k
s
ce
n
ar
io
o
r
if
t
h
er
e
is
an
y
f
o
r
m
o
f
u
n
k
n
o
wn
o
r
d
y
n
am
ic
th
r
ea
ts
,
th
e
tr
ain
ed
m
o
d
el
f
ails
to
s
u
p
p
o
r
t
th
e
h
i
g
h
er
an
d
m
o
r
e
r
eliab
le
th
r
ea
t
d
etec
tio
n
b
y
th
e
ex
is
tin
g
m
ac
h
in
e
lear
n
i
n
g
m
o
d
el.
Hen
ce
,
ex
is
tin
g
s
tu
d
ies
u
s
in
g
m
ac
h
in
e
lear
n
i
n
g
m
ay
s
o
u
n
d
g
o
o
d
,
b
u
t
in
r
ea
lity
,
th
e
ir
ap
p
licab
ilit
y
is
r
estricte
d
to
s
tatic
en
v
ir
o
n
m
en
t i
m
p
lem
en
tatio
n
o
n
ly
.
Hen
ce
,
t
h
e
p
r
im
e
g
ap
o
f
th
e
ex
is
tin
g
s
tu
d
y
is
m
ain
ly
r
el
ated
to
th
e
non
-
u
p
d
atin
g
p
r
o
ce
s
s
f
o
r
th
e
tr
ai
n
ed
m
o
d
el
d
u
r
in
g
th
e
ac
t
u
al
d
ep
lo
y
m
e
n
t
p
r
o
ce
s
s
b
y
ex
is
tin
g
m
ac
h
i
n
e
lear
n
in
g
m
o
d
els.
Fu
r
th
er
,
th
er
e
is
a
n
ee
d
to
d
e
v
elo
p
s
u
ch
a
m
o
d
el
th
at
ca
n
f
u
r
th
e
r
ad
d
r
ess
th
e
o
v
er
f
itti
n
g
a
n
d
r
eliab
ilit
y
p
r
o
b
lem
s
wh
ile
u
s
in
g
m
ac
h
in
e
l
ea
r
n
in
g
in
o
r
d
er
t
o
b
alan
ce
th
e
d
em
an
d
s
o
f
s
u
c
ce
s
s
f
u
l,
u
p
d
ated
,
an
d
r
eliab
l
e
th
r
ea
t
d
etec
tio
n
with
a
co
s
t
-
ef
f
ec
tiv
e
lear
n
in
g
m
o
d
el.
T
h
e
r
elate
d
b
ac
k
g
r
o
u
n
d
wo
r
k
o
f
liter
atu
r
e
s
h
o
ws
th
at
th
er
e
ar
e
v
ar
io
u
s
co
n
tr
ib
u
tio
n
s
o
f
m
ac
h
in
e
lear
n
in
g
-
b
ased
ap
p
r
o
ac
h
es
to
war
d
s
b
o
o
s
tin
g
th
e
s
ec
u
r
ity
f
ea
tu
r
es
in
I
o
T
[
1
4
]
.
T
h
e
r
e
ar
e
v
a
r
io
u
s
s
tu
d
ies
r
ev
iewe
d
co
n
s
id
er
in
g
a
d
o
p
te
d
m
eth
o
d
o
lo
g
ies,
f
ea
tu
r
es,
a
n
d
d
ataset
s
.
T
ab
le
1
h
ig
h
lig
h
t
s
th
e
ch
ar
ac
ter
is
tic
b
ein
g
o
b
tain
ed
f
r
o
m
ex
is
tin
g
s
tu
d
ies
[
1
5
]
–
[
3
0
]
.
Hen
ce
,
it
c
an
b
e
s
ee
n
th
at
XGBo
o
s
t,
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM
)
,
r
an
d
o
m
f
o
r
est
(
RF
)
,
an
d
d
ec
is
io
n
tr
ee
(
DT
)
ar
e
f
r
eq
u
en
tly
ad
o
p
ted
ap
p
r
o
ac
h
e
s
n
o
ted
in
e
x
is
tin
g
ap
p
r
o
ac
h
es.
T
ab
le
1
.
R
ev
iew
o
f
e
x
is
tin
g
s
tu
d
ies
S
t
u
d
i
e
s
M
e
t
h
o
d
s
F
e
a
t
u
r
e
s
D
a
t
a
s
e
t
[
1
5
]
M
u
l
t
i
l
a
y
e
r
p
e
r
c
e
p
t
r
o
n
,
S
V
M
,
R
F
H
e
a
l
t
h
-
r
e
l
a
t
e
d
W
U
S
TL
-
EH
M
C
2
0
2
0
,
I
C
U
-
I
O
M
T
[
1
6
]
DT
1
8
se
n
s
o
r
-
r
e
l
a
t
e
d
f
e
a
t
u
r
e
s
WSN
-
DS
[
2
3
]
C
o
l
l
a
b
o
r
a
t
i
v
e
l
e
a
r
n
i
n
g
Tr
a
f
f
i
c
d
a
t
a
EC
U
-
I
o
H
T
[
2
4
]
X
G
B
o
o
st
,
F
i
s
h
e
r
s
c
o
r
e
m
o
d
e
l
,
g
e
n
e
t
i
c
a
l
g
o
r
i
t
h
m
1
1
5
f
e
a
t
u
r
e
s
f
r
o
m
t
r
a
f
f
i
c
d
a
t
a
N
-
B
a
I
o
T
[
2
5
]
En
se
mb
l
e
4
F
e
a
t
u
r
e
s fr
o
m
n
e
t
w
o
r
k
t
r
a
f
f
i
c
N
S
L
-
K
D
D
[
2
6
]
C
N
N
,
n
a
t
u
r
e
-
i
n
s
p
i
r
e
d
a
l
g
o
r
i
t
h
m
3
4
f
e
a
t
u
r
e
s fr
o
m
n
e
t
w
o
r
k
t
r
a
f
f
i
c
I
o
T
-
M
Q
TT
[
2
7
]
X
G
B
o
o
st
,
R
F
1
1
5
f
e
a
t
u
r
e
s
f
r
o
m
t
r
a
f
f
i
c
d
a
t
a
N
-
B
a
I
o
T
[
2
8
]
[
2
9
]
S
V
M
D
a
t
a
f
e
a
t
u
r
e
s
r
e
l
a
t
e
d
t
o
a
c
t
i
v
i
t
y
,
80
-
t
r
a
f
f
i
c
f
e
a
t
u
r
e
S
y
n
t
h
e
t
i
c
,
C
S
E
-
C
I
C
-
I
D
S
2
0
1
8
[
3
0
]
S
V
M
(
mu
l
t
i
-
c
l
a
ss)
,
D
T
2
0
f
e
a
t
u
r
e
s
f
r
o
m
I
o
T
t
r
a
f
f
i
c
M
u
l
t
i
st
e
p
c
y
b
e
r
-
a
t
t
a
c
k
d
a
t
a
se
t
[
1
7
]
S
V
M
,
K
-
n
e
a
r
e
st
n
e
i
g
h
b
o
r
,
R
F
A
l
a
r
g
e
n
u
mb
e
r
o
f
f
e
a
t
u
r
e
s
M
u
l
t
i
p
l
e
d
a
t
a
se
t
s
[
1
8
]
C
N
N
F
e
a
t
u
r
e
s f
o
r
2
5
c
l
a
ss
e
s
M
a
l
e
V
i
s
[
1
9
]
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
,
e
n
c
r
y
p
t
i
o
n
F
e
a
t
u
r
e
s r
e
l
a
t
e
d
t
o
t
r
a
f
f
i
c
f
l
o
w
S
y
n
t
h
e
t
i
c
d
a
t
a
s
e
t
[
2
0
]
M
u
l
t
i
p
l
e
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
Tr
a
f
f
i
c
-
r
e
l
a
t
e
d
f
e
a
t
u
r
e
I
o
T
-
23
[
2
1
]
M
u
l
t
i
p
l
e
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
45
-
f
e
a
t
u
r
e
s
TO
N
-
I
o
T
[
2
2
]
M
u
l
t
i
p
l
e
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
6
-
f
e
a
t
u
r
e
s
r
e
l
a
t
e
d
t
o
t
r
a
f
f
i
c
f
l
o
w
N
F
U
Q
-
N
I
D
S
-
v2
T
h
e
r
esear
ch
p
r
o
b
lem
s
ass
o
c
iated
with
ex
is
tin
g
s
tu
d
ies
a
r
e
m
an
if
o
ld
.
W
ith
t
h
e
ad
v
an
ce
m
en
t
o
f
v
ar
ied
n
etwo
r
k
in
g
i
n
lar
g
e
s
ca
le
an
d
d
is
tr
ib
u
ted
m
an
n
e
r
,
th
er
e
is
also
an
im
m
en
s
e
i
n
cr
ea
s
e
in
d
if
f
e
r
en
t
th
r
ea
ts
.
Su
ch
th
r
ea
ts
h
av
e
b
ee
n
r
ep
o
r
ted
to
h
a
v
e
h
ig
h
p
o
t
en
tial
an
d
ca
n
b
y
p
ass
an
y
s
ec
u
r
ity
m
o
n
ito
r
i
n
g
s
y
s
tem
in
I
o
T
(
e.
g
.
f
ir
ewa
ll)
.
T
h
e
ex
is
tin
g
s
ec
u
r
ity
f
ir
ewa
ll
s
y
s
tem
ca
n
o
n
ly
r
esis
t
t
h
e
th
r
ea
ts
alr
ea
d
y
co
n
s
id
er
ed
wh
ile
d
e
v
elo
p
in
g
t
h
e
s
ec
u
r
ity
p
atch
f
o
r
t
h
e
f
ir
e
wall.
Hen
ce
,
f
ir
ewa
ll
s
y
s
tem
s
ca
n
n
o
t
id
e
n
tify
an
y
th
r
ea
ts
th
at
ar
e
n
o
t
d
ec
lar
e
d
a
n
d
d
ef
in
ed
well
in
th
eir
s
tan
d
ar
d
d
ata
b
ase
s
y
s
tem
.
Hen
ce
,
t
h
er
e
is
a
n
ee
d
f
o
r
a
n
in
n
o
v
ativ
e
s
y
s
tem
ca
p
ab
le
o
f
id
en
tify
in
g
a
s
ec
u
r
ity
th
r
ea
t
o
f
u
n
k
n
o
wn
f
o
r
m
d
if
f
e
r
en
t
f
r
o
m
th
at
u
s
ed
in
d
ev
elo
p
in
g
a
f
ir
ewa
ll sy
s
tem
.
T
h
er
e
f
o
r
e,
th
e
p
r
o
p
o
s
ed
s
t
u
d
y
ai
m
s
to
d
ev
e
lo
p
a
co
m
p
u
t
a
tio
n
al
m
o
d
e
l
ca
p
ab
le
o
f
h
ar
n
es
s
in
g
th
e
p
o
ten
ti
al
o
f
a
m
ac
h
in
e
-
le
ar
n
in
g
ap
p
r
o
ac
h
to
war
d
s
d
y
n
am
ic
,
u
n
k
n
o
wn
,
an
d
u
n
c
er
t
ain
th
r
ea
t
d
e
te
ct
io
n
in
I
o
T
.
T
h
e
v
a
lu
e
-
ad
d
ed
co
n
t
r
ib
u
t
io
n
s
in
cu
r
r
e
n
t
wo
r
k
ar
e
a
s
f
o
l
lo
ws
:
i
)
a
n
o
v
e
l
th
r
ea
t
d
e
te
ct
io
n
s
ch
em
e
h
as
b
ee
n
p
r
e
s
en
ted
ca
p
ab
l
e
o
f
lea
r
n
in
g
th
e
d
y
n
am
i
c
p
at
te
r
n
s
an
d
co
r
r
el
at
io
n
o
f
p
o
ten
ti
al
v
u
ln
er
ab
i
li
ti
e
s
,
ii
)
a
d
is
t
in
c
t
u
p
d
at
in
g
s
ch
em
e
h
a
s
b
e
en
in
tr
o
d
u
c
ed
th
a
t
c
an
o
p
t
im
iz
e
t
h
e
d
a
ta
q
u
al
it
y
o
f
tr
a
in
in
g
m
o
d
el
to
r
ea
li
ze
th
e
d
y
n
am
i
c
ev
en
t
s
i
n
i
n
co
m
in
g
tr
a
f
f
i
c,
ii
i)
a
s
im
p
li
s
tic
an
d
y
et
in
n
o
v
a
tiv
e
n
e
u
r
a
l
n
e
t
w
o
r
k
b
a
s
ed
m
a
ch
i
n
e
lea
r
n
in
g
m
o
d
el
h
a
s
b
ee
n
p
r
e
s
en
ted
t
h
a
t
c
an
p
er
f
o
r
m
r
ep
r
es
en
t
at
io
n
le
ar
n
in
g
to
ex
tr
a
c
t
m
o
r
e
s
ig
n
if
i
ca
n
t
u
n
d
er
ly
in
g
p
a
t
ter
n
s
o
f
th
r
e
at
s
,
i
v
)
a
u
n
iq
u
e
m
a
s
k
ed
in
d
ex
in
g
m
ec
h
an
i
s
m
h
as
b
ee
n
p
r
e
s
en
ted
to
w
ar
d
s
s
af
eg
u
a
r
d
in
g
th
e
l
ab
e
l
s
o
f
tr
ain
e
d
d
a
ta
f
r
o
m
ad
v
er
s
ar
i
e
s
.
T
h
e
p
ap
er
's
o
r
g
a
n
i
z
at
io
n
i
s
:
Se
ct
io
n
2
p
r
e
s
en
t
s
a
d
is
cu
s
s
io
n
o
f
m
e
th
o
d
s
,
wh
il
e
th
e
r
e
s
u
lt
is
d
is
cu
s
s
ed
in
S
ec
tio
n
3
,
an
d
th
e
co
n
clu
s
i
o
n
i
s
p
r
e
s
en
t
ed
in
Se
ct
io
n
-
4
.
2.
M
E
T
H
O
D
T
h
e
p
r
o
p
o
s
ed
s
y
s
tem
aim
s
to
d
ev
elo
p
a
n
o
v
el
d
etec
tio
n
s
c
h
em
e
o
f
p
o
ten
tial
th
r
ea
ts
b
y
h
ar
n
ess
in
g
m
ac
h
in
e
lear
n
in
g
to
war
d
s
s
ec
u
r
in
g
f
u
tu
r
e
c
o
m
m
u
n
icatio
n
s
y
s
tem
s
i.e
.
I
o
T
.
A
n
I
o
T
i
n
f
r
a
s
tr
u
ctu
r
e
is
f
o
r
m
ed
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
N
eu
r
a
l
-
n
etw
o
r
k
b
a
s
ed
r
ep
r
es
en
ta
tio
n
fr
a
mewo
r
k
fo
r
a
d
ve
r
s
a
r
y
…
(
Th
a
n
u
j
a
N
a
r
a
s
imh
a
mu
r
th
y
)
6045
b
y
co
n
s
id
er
in
g
a
s
p
ec
if
ic
n
u
m
b
er
o
f
I
o
T
d
e
v
ices
th
at
ca
p
tu
r
e
th
e
d
ata
an
d
f
o
r
war
d
it
t
o
a
p
r
o
ce
s
s
in
g
u
n
it
h
o
s
ted
with
in
an
ed
g
e
d
e
v
ice.
T
h
e
p
r
o
ce
s
s
in
g
u
n
it
,
co
n
s
id
er
in
g
a
n
a
d
d
itio
n
al
5
0
+
f
ea
tu
r
e
s
u
s
in
g
a
s
tan
d
ar
d
d
ataset
r
elate
d
to
tr
af
f
ic
f
lo
w
,
f
u
r
th
er
e
x
ec
u
tes
th
e
alg
o
r
ith
m
s
to
war
d
s
s
cr
u
t
in
izin
g
th
r
ea
ts
th
at
ar
e
h
o
s
ted
i
n
a
co
n
n
ec
ted
clo
u
d
en
v
ir
o
n
m
e
n
t
.
An
aly
tical
m
o
d
ellin
g
is
co
n
s
id
er
ed
f
o
r
d
ev
elo
p
in
g
th
e
r
ese
ar
ch
m
eth
o
d
o
lo
g
y
,
wh
er
e
th
e
c
o
m
p
lete
o
p
er
atio
n
is
class
if
ied
in
to
d
u
al
im
p
lem
en
tatio
n
s
tag
e
s
.
T
h
e
f
ir
s
t
im
p
lem
en
tatio
n
is
to
war
d
s
id
en
tify
in
g
t
h
e
p
o
ten
t
ial
th
r
ea
t
,
wh
ile
th
e
s
ec
o
n
d
im
p
lem
en
tatio
n
is
ab
o
u
t
ap
p
ly
in
g
a
n
eu
r
al
n
etwo
r
k
b
ased
m
ac
h
in
e
lear
n
in
g
m
o
d
el
to
p
er
f
o
r
m
u
p
d
atin
g
o
f
in
co
m
in
g
t
h
r
ea
ts
.
Fig
u
r
e
1
h
ig
h
lig
h
ts
th
e
ad
o
p
ted
ar
ch
itectu
r
e
o
f
th
e
p
r
o
p
o
s
ed
m
eth
o
d
o
lo
g
y
.
I
n
d
e
x
e
d
D
a
t
a
U
n
i
n
d
e
x
e
d
D
a
t
a
C
o
n
s
t
r
u
c
t
i
o
n
O
p
e
r
a
t
o
r
R
e
c
o
n
s
t
r
u
c
t
i
o
n
O
p
e
r
a
t
o
r
F
e
a
t
u
r
e
s
F
e
a
t
u
r
e
s
C
o
r
r
e
l
a
t
i
o
n
M
e
a
n
R
e
g
u
l
a
r
F
e
a
t
u
r
e
N
o
r
m
a
l
D
i
s
t
r
i
b
u
t
i
o
n
C
o
s
t
f
u
n
c
t
i
o
n
G
e
n
e
r
a
t
e
M
a
s
k
e
d
I
n
d
e
x
P
e
r
f
o
r
m
U
p
d
a
t
i
n
g
T
h
r
e
a
t
D
e
t
e
c
t
i
o
n
N
e
u
r
a
l
N
e
t
w
o
r
k
Fig
u
r
e
1
.
Ar
c
h
itectu
r
e
o
f
p
r
o
p
o
s
ed
r
esear
ch
m
eth
o
d
o
l
o
g
y
Acc
o
r
d
in
g
t
o
Fig
u
r
e
1
,
th
e
tr
af
f
ic
in
f
o
r
m
atio
n
is
s
p
lit
in
to
in
d
ex
ed
a
n
d
u
n
in
d
e
x
ed
d
ata
wh
ich
is
s
u
b
jecte
d
to
a
n
eu
r
al
n
etwo
r
k
m
o
d
el
with
an
in
clu
s
io
n
o
f
co
n
s
tr
u
ctio
n
a
n
d
r
ec
o
n
s
tr
u
c
tio
n
o
p
er
at
o
r
.
T
h
e
n
eu
r
al
n
etwo
r
k
m
o
d
el
ad
o
p
te
d
to
war
d
s
p
er
f
o
r
m
in
g
lear
n
i
n
g
o
f
tr
af
f
ic
d
ata
r
ep
r
esen
ta
tio
n
wh
er
e
th
e
in
p
u
t
tr
af
f
ic
d
ata
is
in
itially
s
u
b
jec
ted
to
f
o
r
war
d
p
ass
wh
er
e
co
n
s
tr
u
ctio
n
o
p
e
r
ato
r
is
u
s
ed
to
g
en
er
ate
a
laten
t
r
ep
r
esen
tatio
n
wh
ile
r
ec
o
n
s
tr
u
ctio
n
o
p
er
ato
r
is
u
s
ed
to
g
en
e
r
ate
r
ec
o
n
s
tr
u
ctio
n
o
f
o
r
ig
i
n
al
in
p
u
t
u
s
in
g
laten
t
r
ep
r
esen
tatio
n
g
e
n
er
ated
in
p
r
io
r
s
tag
e.
T
h
e
o
p
e
r
atio
n
y
ield
s
f
ea
tu
r
e
th
at
ar
e
s
u
b
jecte
d
to
co
r
r
elatio
n
an
al
y
s
is
f
o
llo
wed
b
y
u
s
in
g
co
s
t
f
u
n
ct
io
n
with
an
o
b
jectiv
e
to
m
ak
e
th
e
r
ec
o
n
s
tr
u
ctio
n
m
atch
in
g
with
th
e
o
r
ig
in
al
in
p
u
t.
T
h
e
c
o
s
t
f
u
n
ctio
n
a
d
o
p
t
ed
in
th
e
p
r
o
p
o
s
ed
s
y
s
tem
is
b
asically
d
esig
n
ed
f
o
r
lear
n
i
n
g
r
ep
r
esen
tatio
n
th
at
is
ca
p
ab
le
o
f
e
x
tr
ac
tin
g
b
o
t
h
s
i
m
ilar
ities
an
d
d
if
f
er
en
ce
s
am
o
n
g
d
ata
p
o
in
ts
.
T
h
is
is
d
o
n
e
w
ith
th
e
o
b
jectiv
e
to
in
ce
n
tiv
es
r
eg
u
lar
tr
af
f
ic
an
d
p
en
alize
m
alicio
u
s
tr
af
f
ic.
T
h
e
n
o
v
elty
o
f
th
is
n
eu
r
al
n
etw
o
r
k
m
o
d
el
is
th
at
it
u
s
es
its
o
p
er
ato
r
an
d
co
s
t
f
u
n
ctio
n
to
r
ed
u
ce
th
e
d
is
tan
ce
b
etwe
en
th
e
p
o
s
itiv
e
tr
af
f
ic
d
ata
(
r
ep
r
esen
tin
g
r
eg
u
lar
tr
a
f
f
ic)
an
d
in
cr
ea
s
in
g
th
e
d
is
tan
ce
b
etwe
en
d
is
s
i
m
il
ar
d
ata
(
r
e
p
r
esen
tin
g
m
alicio
u
s
tr
af
f
ic)
.
T
h
e
o
p
e
r
atio
n
s
o
f
t
h
e
co
s
t
f
u
n
ctio
n
r
esu
lt
in
m
ea
n
r
e
g
u
l
ar
f
ea
tu
r
es
,
wh
ich
,
alo
n
g
with
ex
tr
ac
ted
v
alu
es
o
f
n
o
r
m
al
d
is
tr
ib
u
tio
n
,
ar
e
s
u
b
jecte
d
to
war
d
s
ass
ess
m
en
t
o
f
r
e
g
u
lar
an
d
m
alicio
u
s
tr
af
f
ic
,
th
e
r
eb
y
ac
co
m
p
lis
h
in
g
t
h
e
th
r
ea
t
d
et
ec
tio
n
o
p
er
atio
n
.
Af
ter
th
e
t
h
r
ea
t
d
etec
tio
n
is
p
e
r
f
o
r
m
ed
,
th
e
n
ex
t
task
is
to
in
cr
ea
s
e
th
e
ad
ap
ta
b
ilit
y
o
f
th
e
m
o
d
el
to
p
er
f
o
r
m
d
etec
tio
n
o
v
er
a
s
tr
ea
m
o
f
d
y
n
am
ic
t
r
af
f
ic
in
f
o
r
m
atio
n
.
T
h
e
s
am
e
n
eu
r
al
n
etwo
r
k
m
o
d
el
is
tr
ain
ed
with
a
r
an
d
o
m
weig
h
t
,
r
esu
ltin
g
in
th
e
m
ask
ed
in
d
e
x
wh
ich
r
ep
r
esen
ts
a
s
ec
o
n
d
lay
er
o
f
en
co
d
e
d
v
al
u
e
o
f
lab
elled
d
ata
o
f
tr
af
f
ic.
T
h
e
b
en
ef
icial
asp
ec
t
o
f
th
is
o
p
er
atio
n
is
th
at
th
e
ad
o
p
tio
n
o
f
m
ask
ed
lab
el
o
f
f
e
r
s
f
u
r
th
e
r
d
ata
in
teg
r
ity
as
wel
l
as
d
ata
p
r
iv
ac
y
as
n
o
attac
k
e
r
will
ev
er
b
e
ab
le
to
co
m
p
u
te
t
h
e
ac
tu
al
lab
el
in
f
o
r
m
atio
n
o
f
th
e
tr
af
f
ic
d
ata.
An
o
th
er
b
e
n
ef
icial
asp
ec
t
o
f
t
h
is
m
o
d
u
le
is
th
at
–
th
e
m
o
d
el
ca
n
n
o
w
b
e
ab
le
t
o
ea
s
ily
id
en
tify
a
n
y
f
o
r
m
o
f
s
p
u
r
io
u
s
tr
af
f
ic
if
an
attac
k
e
r
attem
p
ts
to
tam
p
er
with
eith
er
th
e
d
ata
o
r
t
h
e
lab
el
o
f
it.
T
h
e
tam
p
e
r
ed
lab
el
o
f
th
e
d
ata
b
y
th
e
attac
k
er
is
l
ess
lik
ely
to
m
atch
with
th
e
co
m
p
u
ted
v
alu
e
o
f
th
e
m
ask
ed
in
d
e
x
p
r
esen
ted
i
n
th
e
p
r
o
p
o
s
ed
s
y
s
tem
an
d
h
en
ce
s
tr
o
n
g
ly
ac
ts
s
im
ilar
ly
to
a
tr
ap
d
o
o
r
f
u
n
ctio
n
u
s
ed
in
a
c
o
n
v
e
n
tio
n
al
cr
y
p
t
o
g
r
ap
h
y
-
b
ased
ap
p
r
o
ac
h
.
Fu
r
th
er
,
th
e
p
r
o
p
o
s
ed
s
tu
d
y
in
tr
o
d
u
ce
s
u
p
d
atin
g
o
p
er
atio
n
s
in
o
r
d
er
to
in
cr
ea
s
e
th
e
ca
p
ab
ilit
y
o
f
co
v
er
ag
e
o
f
tr
ain
e
d
d
ata
b
y
t
h
e
p
r
o
p
o
s
ed
n
eu
r
al
n
etwo
r
k
m
o
d
el
to
b
e
ab
le
t
o
r
esis
t
p
o
ten
tially
d
y
n
am
ic
f
o
r
m
s
o
f
th
r
ea
ts
.
I
n
ter
esti
n
g
ly
,
th
e
in
ter
n
al
o
p
e
r
atio
n
o
f
th
e
s
ch
em
e
is
q
u
ite
p
r
o
g
r
ess
iv
e
an
d
less
iter
ativ
e
.
E
v
en
tu
ally
,
it
u
s
es
m
ac
h
in
e
l
ea
r
n
in
g
m
eth
o
d
o
lo
g
y
wh
ich
c
o
n
tr
ib
u
tes
to
war
d
s
an
e
f
f
ec
tiv
e
b
alan
ce
b
etwe
en
s
ec
u
r
ity
f
ea
tu
r
es a
n
d
co
m
p
u
tatio
n
al
f
ea
tu
r
es.
An
illu
s
tr
atio
n
o
f
d
esig
n
im
p
lem
en
tatio
n
f
o
ll
o
ws n
ex
t:
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
.
6
,
Decem
b
e
r
20
25
:
6
0
4
3
-
6
0
5
2
6046
2
.
1
.
P
o
t
ent
ia
l
t
hrea
t
identif
ica
t
i
o
n
T
h
is
is
th
e
f
ir
s
t m
o
d
u
le
o
f
im
p
lem
en
tatio
n
th
at
is
r
esp
o
n
s
ib
le
f
o
r
th
r
ea
t i
d
en
tific
atio
n
.
T
h
e
p
r
im
e
task
o
f
th
is
alg
o
r
ith
m
is
b
asically
to
p
er
f
o
r
m
th
e
id
e
n
tific
atio
n
o
f
p
o
ten
tial
th
r
ea
t
s,
f
o
llo
wed
b
y
th
e
g
en
er
atio
n
o
f
a
m
ask
ed
in
d
e
x
.
B
asically
,
th
e
ter
m
m
ask
ed
in
d
ex
is
m
ea
n
t
f
o
r
s
af
eg
u
ar
d
in
g
th
e
lab
el
s
ass
ig
n
ed
f
o
r
th
e
in
co
m
in
g
tr
af
f
ic
d
ata
,
wh
ich
o
th
er
wis
e
co
u
ld
b
e
tam
p
e
r
ed
w
ith
o
r
ea
v
esd
r
o
p
p
ed
b
y
a
n
atta
ck
er
.
T
h
e
s
tep
s
o
f
th
e
p
r
o
p
o
s
ed
alg
o
r
ith
m
to
war
d
s
th
r
ea
t id
en
tific
atio
n
a
r
e
as
Alg
o
r
ith
m
1
:
Alg
o
r
ith
m
1
.
Alg
o
r
ith
m
f
o
r
th
r
ea
t id
en
tific
atio
n
Input
: T
Output
: TI
Start
1. init T
→
(T
r
, T
m
)
2.
(T
r
, T
m
)
→
[(α
r
, β
r
) (α
m
, β
m
)]
3. compute
ρ
=
f
1
(
tr
i
,
tr
j
)
4. λ=
f
2
(
ρ
, δ)
5. λ
1
=
arg
min
(Σλ)
6. TI=
f
3
(
ρ
(
r
,
m
))
End
T
h
e
illu
s
tr
atio
n
o
f
th
e
alg
o
r
it
h
m
is
as
f
o
llo
ws:
T
h
e
alg
o
r
i
th
m
tak
es
t
h
e
in
p
u
t
o
f
T
(
tr
a
f
f
ic
d
ata)
,
wh
ich
,
u
p
o
n
p
r
o
ce
s
s
in
g
,
will
lead
to
th
e
g
e
n
er
atio
n
o
f
TI
(
t
h
r
ea
t
id
en
tific
atio
n
)
.
T
h
e
alg
o
r
i
th
m
in
itializes
th
e
tr
af
f
ic
d
ata
T
co
n
ce
r
n
i
n
g
r
eg
u
lar
tr
af
f
ic
T
r
a
n
d
m
alicio
u
s
tr
af
f
ic
T
m
(
L
in
e
-
1
)
.
Fu
r
th
e
r
,
th
e
tr
af
f
ic
d
ata
is
r
ep
r
esen
ted
in
th
e
f
o
r
m
o
f
in
p
u
t
tr
af
f
ic
v
ec
to
r
(
α
r,
α
m
)
o
f
r
eg
u
lar
an
d
m
alicio
u
s
f
o
r
m
s
a
s
well
as
as
s
o
ciate
d
in
d
ex
es
(
β
r,
β
m
)
o
f
r
eg
u
lar
an
d
m
alicio
u
s
f
o
r
m
s,
r
esp
ec
tiv
ely
(
L
in
e
-
2
)
.
I
t
will
ev
en
tu
ally
m
ea
n
th
at
th
e
tr
af
f
ic
d
ata
T
will
co
n
s
is
t
o
f
in
p
u
t
tr
af
f
ic
v
ec
to
r
α
o
f
s
s
ize
wh
ile
ea
ch
in
d
ex
β
is
in
itialized
wit
h
a
b
in
ar
y
r
an
g
e
o
f
[
0
,
1
]
wh
er
e
0
a
n
d
1
d
ep
ict
r
e
g
u
lar
an
d
m
alicio
u
s
tr
af
f
ic
,
r
e
s
p
ec
tiv
ely
.
T
h
e
p
r
o
p
o
s
ed
alg
o
r
ith
m
co
n
s
id
er
s
th
e
tr
ain
ed
r
ep
r
esen
tatio
n
ass
o
ciate
d
with
ea
ch
in
p
u
t
tr
af
f
ic
d
ata
as
tr
(
∈
T
R
)
,
wh
ile
th
e
p
r
o
p
o
s
ed
n
eu
r
al
n
etwo
r
k
m
o
d
el
u
s
ed
f
o
r
th
r
ea
t d
etec
tio
n
is
r
ep
r
esen
ted
b
y
π
w
,
wh
e
r
e
w
r
ep
r
ese
n
ts
th
e
weig
h
t a
s
s
o
ci
ated
with
th
e
n
eu
r
al
m
o
d
el.
T
h
e
p
r
o
p
o
s
ed
s
y
s
tem
u
s
es
a
u
n
iq
u
e
co
s
t
f
u
n
ctio
n
th
at
tar
g
ets
to
in
cr
ea
s
e
th
e
co
r
r
elatio
n
b
etwe
en
r
eg
u
lar
tr
a
f
f
ic
(
tr
r,
i
,
tr
r,
j
)
an
d
d
ec
r
ea
s
e
th
e
s
am
e
f
o
r
r
ef
er
en
ce
s
am
p
le
tr
r,
i
an
d
m
alicio
u
s
tr
af
f
ic
tr
m,
k.
T
h
e
v
ar
iab
le
tr
r,
i
r
ep
r
esen
ts
th
e
r
e
f
er
en
ce
s
am
p
le
ass
o
ciate
d
with
r
eg
u
lar
tr
a
f
f
ic
in
f
o
r
m
atio
n
w
h
ile
th
e
v
alu
e
o
f
th
e
s
u
b
s
cr
ip
ts
(
i
,
j
)
is
(
1
,
2
,
3
,
.
.
.
.
.
,
I
r
)
,
w
h
er
e
v
a
r
iab
le
I
r
r
ep
r
e
s
en
ts
th
e
in
d
ex
f
o
r
r
e
g
u
lar
tr
af
f
ic.
T
h
e
v
alu
e
o
f
s
u
b
s
cr
ip
t
k
is
(
1
,
2
,
3
,
.
.
.
.
I
m
)
,
wh
er
e
v
a
r
iab
le
I
m
r
e
p
r
esen
ts
t
h
e
in
d
e
x
f
o
r
m
alicio
u
s
tr
af
f
ic.
I
n
th
e
co
n
s
ec
u
tiv
e
p
ar
t
o
f
alg
o
r
ith
m
im
p
lem
en
ta
tio
n
,
th
e
s
ch
em
e
co
m
p
u
tes
c
o
r
r
el
a
tio
n
attr
ib
u
te
ρ
b
y
c
o
n
s
tr
u
ctin
g
a
n
e
x
p
licit
f
u
n
ctio
n
f
1
(
x
)
c
o
n
s
id
er
in
g
in
p
u
t
ar
g
u
m
en
ts
o
f
r
e
g
u
lar
tr
af
f
ic
in
f
o
r
m
atio
n
tr
i
an
d
tr
j
(
L
in
e
-
3
)
.
T
h
e
f
u
n
ctio
n
f
1
(
x
)
,
wh
ich
is
in
ten
d
ed
to
ev
a
lu
ate
th
e
s
im
ilar
ity
b
etwe
en
p
air
s
o
f
tr
af
f
ic
d
ata
p
o
in
ts
,
is
u
s
ed
b
y
th
e
s
y
s
tem
to
ca
lcu
late
th
e
co
r
r
elatio
n
attr
i
b
u
te
ρ
.
I
n
p
ar
ticu
lar
,
f
1
(
x
)
co
m
p
u
tes
th
e
d
o
t
p
r
o
d
u
ct
o
f
t
wo
tr
af
f
ic
s
am
p
les'
f
ea
tu
r
e
v
ec
t
o
r
s
an
d
n
o
r
m
alize
s
th
e
o
u
tco
m
e
u
s
in
g
th
e
E
u
clid
ea
n
d
is
tan
ce
.
T
h
is
p
r
o
ce
s
s
m
ak
es
s
u
r
e
th
at
t
h
e
co
r
r
elatio
n
ac
c
u
r
ately
r
e
p
r
ese
n
ts
h
o
w
s
im
ilar
th
e
s
am
p
les
a
r
e,
wh
ich
is
ess
en
tial
f
o
r
d
if
f
e
r
en
tiatin
g
m
alicio
u
s
tr
af
f
ic
f
r
o
m
leg
itima
te
tr
af
f
ic.
T
h
e
n
eu
r
al
n
etwo
r
k
lear
n
s
to
d
is
tin
g
u
is
h
b
etwe
en
m
alicio
u
s
an
d
leg
itima
te
tr
af
f
ic
b
y
u
s
in
g
th
e
c
o
r
r
elatio
n
s
co
r
e
as
in
p
u
t
to
th
e
co
s
t
f
u
n
ctio
n
λ
.
T
h
e
co
m
p
u
tatio
n
o
f
th
is
ex
p
licit
f
u
n
ctio
n
f
1
(
x
)
is
em
p
ir
ically
ex
p
r
ess
ed
as
(
1
)
:
1
(
)
=
(
.
)
.
[
(
,
)
]
−
1
(
1
)
Af
ter
th
e
ex
p
licit
f
u
n
ctio
n
f
1
(
x
)
co
m
p
u
tatio
n
,
th
e
s
y
s
tem
c
o
m
p
u
tes
th
e
v
al
u
e
o
f
t
h
e
co
r
r
elatio
n
attr
ib
u
te
ρ
,
wh
ich
is
f
u
r
th
er
u
tili
ze
d
to
w
ar
d
s
its
co
n
s
ec
u
tiv
e
s
tep
s
f
o
r
co
m
p
u
tin
g
co
s
t
f
u
n
ctio
n
λ
.
A
clo
s
er
lo
o
k
in
t
o
L
in
e
-
4
s
h
o
wca
s
e
s
th
at
th
e
c
o
m
p
u
tatio
n
o
f
co
s
t
f
u
n
ctio
n
λ
is
ca
r
r
ied
o
u
t
b
y
d
e
v
elo
p
i
n
g
a
n
o
th
er
ex
p
licit
f
u
n
ctio
n
f
2
(
x
)
with
an
in
p
u
t
a
r
g
u
m
en
t
o
f
co
r
r
elatio
n
attr
i
b
u
te
ρ
an
d
ad
j
u
s
tm
en
t
p
ar
a
m
eter
δ
.
T
h
e
s
tep
-
wis
e
co
m
p
u
tatio
n
o
f
c
o
s
t f
u
n
ctio
n
λ
is
s
h
o
wn
as
(
2
)
an
d
(
3
)
:
,
=
.
1
2
(
2
)
=
[
]
−
1
∑
,
(
3
)
A
clo
s
er
lo
o
k
in
t
o
th
e
ex
p
r
ess
io
n
s
(
2
)
a
n
d
(
3
)
s
h
o
w
th
at
t
h
e
p
r
o
p
o
s
ed
s
ch
em
e
i
n
itially
co
m
p
u
tes
ex
p
r
ess
io
n
(
2
)
,
wh
ic
h
is
th
en
u
s
ed
in
ex
p
r
ess
io
n
(
3
)
.
I
n
th
e
in
itial
co
m
p
u
tatio
n
al
s
tep
o
f
ex
p
r
ess
io
n
(
2
)
,
th
e
v
ar
iab
le
A
1
r
e
p
r
esen
ts
an
ex
p
o
n
en
tial
f
o
r
m
o
f
co
r
r
elatio
n
attr
ib
u
te
ρ
f
o
r
(
i
,
j
)
d
iv
id
e
d
b
y
ad
j
u
s
tm
en
t
p
ar
am
eter
δ
.
At
th
e
s
am
e
tim
e,
th
e
v
ar
ia
b
le
A
2
r
ep
r
esen
ts
th
e
s
u
m
m
ati
o
n
o
f
t
h
e
n
ewly
ac
co
m
p
lis
h
ed
v
alu
e
o
f
A
1
an
d
all
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
N
eu
r
a
l
-
n
etw
o
r
k
b
a
s
ed
r
ep
r
es
en
ta
tio
n
fr
a
mewo
r
k
fo
r
a
d
ve
r
s
a
r
y
…
(
Th
a
n
u
j
a
N
a
r
a
s
imh
a
mu
r
th
y
)
6047
ad
d
itiv
e
v
alu
e
s
o
f
A
1
with
th
e
lo
wer
r
an
g
e
o
f
(
i
,
k
)
a
n
d
in
d
e
x
r
an
g
e
o
f
(
I
r
,
I
m
)
.
I
n
th
e
n
ex
t
co
m
p
u
tatio
n
al
s
te
p
o
f
ex
p
r
ess
io
n
(
3
)
,
th
e
v
ar
iab
l
e
I
r
ep
r
esen
ts
I
r
(
I
r
-
1
)
,
wh
er
e
I
r
is
th
e
in
d
ex
f
o
r
r
eg
u
lar
tr
a
f
f
ic
an
d
th
e
s
ec
o
n
d
co
m
p
o
n
en
t
r
ep
r
esen
ts
all
ad
d
itiv
e
v
alu
es
o
f
e
x
p
r
ess
io
n
(
2
)
.
T
h
e
alg
o
r
ith
m
m
in
im
izes
th
e
co
s
t
f
u
n
ctio
n
ass
o
ciate
d
with
b
o
th
co
n
s
tr
u
ctio
n
co
an
d
r
ec
o
n
s
tr
u
ctio
n
re
to
ef
f
icien
tly
lear
n
th
e
tr
af
f
i
c
r
ep
r
esen
tatio
n
to
o
b
tain
th
e
n
ew
v
alu
e
o
f
c
o
s
t
f
u
n
ctio
n
λ
1
(
L
in
e
-
5
)
.
T
h
e
v
ar
ia
b
le
Σλ
r
ep
r
esen
ts
th
e
s
u
m
m
ati
o
n
o
f
λ
co
a
n
d
λ
re
.
I
n
th
e
f
in
al
s
tep
o
f
t
h
e
alg
o
r
ith
m
(
L
in
e
-
6
)
,
a
n
o
th
er
ex
p
lici
t
m
eth
o
d
f
3
(
x
)
is
u
s
ed
f
o
r
h
ig
h
er
p
r
o
b
a
b
ilit
y
ass
es
s
m
en
t
(
HPA)
in
o
r
d
er
to
d
er
iv
e
an
ac
c
u
r
ate
co
r
r
elatio
n
s
co
r
e
ρ
with
th
e
n
o
r
m
al
d
is
tr
ib
u
tio
n
o
f
tr
af
f
i
c
with
r
esp
ec
t to
b
o
th
r
eg
u
la
r
an
d
m
alicio
u
s
.
T
h
is
in
f
o
r
m
atio
n
o
f
th
e
id
e
n
tifie
d
th
r
ea
t is st
o
r
e
d
in
m
atr
ix
T
I
.
2
.
2
.
Upda
t
ing
o
pera
t
io
n f
o
r
inco
m
ing
t
hrea
t
T
h
is
is
th
e
s
ec
o
n
d
p
ar
t
o
f
im
p
l
em
en
tatio
n
wh
ich
is
m
ain
ly
co
n
ce
n
tr
ated
to
war
d
s
p
er
f
o
r
m
in
g
u
p
d
ate
d
o
p
er
atio
n
s
f
o
r
all
in
co
m
i
n
g
th
r
ea
t
-
p
r
o
n
e
tr
af
f
ic.
A
clo
s
er
lo
o
k
in
to
th
e
ad
o
p
tio
n
o
f
m
ac
h
in
e
lear
n
in
g
ap
p
r
o
ac
h
es
to
war
d
s
th
r
ea
t
d
et
ec
tio
n
in
e
x
is
tin
g
s
y
s
tem
s
is
alwa
y
s
f
o
u
n
d
to
d
ep
e
n
d
on
its
p
r
ed
ictiv
e
o
p
er
atio
n
o
n
its
tr
ain
e
d
d
ata.
T
h
e
v
er
y
ass
u
m
p
tio
n
th
at
all
tr
ain
ed
d
a
ta
will
co
n
s
is
t
o
f
co
m
p
lete
p
o
s
s
ib
ilit
ie
s
o
f
tr
af
f
ic
ev
en
ts
u
p
o
n
ex
p
o
s
in
g
th
e
m
i
n
a
d
ep
lo
y
m
en
t
s
ce
n
ar
io
is
i
m
p
r
ac
tical,
esp
ec
ially
r
eg
ar
d
i
n
g
d
y
n
am
ic
th
r
ea
ts
.
Hen
ce
,
th
e
p
r
im
e
o
b
jectiv
e
o
f
th
is
alg
o
r
ith
m
is
to
in
cr
ea
s
e
th
e
ad
ap
ta
b
ilit
y
o
f
th
e
t
h
r
ea
t
d
etec
tio
n
m
o
d
el
to
u
n
d
er
s
tan
d
an
d
r
ea
lize
th
e
d
y
n
am
icity
in
v
o
lv
e
d
in
th
e
tr
af
f
i
c
en
v
ir
o
n
m
e
n
t
b
y
g
en
e
r
atin
g
u
p
d
ated
in
f
o
r
m
atio
n
o
n
in
c
o
m
in
g
th
r
ea
t
s
.
T
h
e
c
o
r
e
co
n
tr
ib
u
tio
n
o
f
th
is
m
o
d
el
is
i
ts
f
r
ee
d
o
m
f
r
o
m
an
y
f
o
r
m
o
f
d
em
an
d
s
o
f
m
a
n
u
al
in
d
ex
in
g
task
s
wh
ile
p
er
f
o
r
m
in
g
tr
ain
in
g
o
p
er
atio
n
s
o
v
e
r
s
tr
ea
m
s
o
f
tr
af
f
ic
in
f
o
r
m
atio
n
.
T
h
e
alg
o
r
ith
m
ic
s
tep
s
ar
e
as sh
o
wn
in
Alg
o
r
ith
m
2
:
Alg
o
r
ith
m
2
.
Alg
o
r
ith
m
f
o
r
u
p
d
atin
g
in
co
m
in
g
th
r
ea
t
Input
:
T
o
,
w
Output
:
up
Start
1.
init
w
2. train
param
3. α, β
α
o
, β
o
4.
For
i=α
i
5. N[(
co
,
re
)(
r
,
m
)]=
f
4
(
param
1
)
6.
For
j=0: (
h
-
1) do
7. res(
co
,
re
)
i+j
=π
w
(α
i+j
)
8.
β
1
=
f
5
(
param
2
)
9.
up
(α, β)=
f
6
[(α, β), (α
i+j
, β
i+j
)]
10.
End
11.
End
12. initiate
training
(
param
2
)
End
T
h
e
ab
o
v
e
-
s
h
o
w
n
alg
o
r
ith
m
t
ak
es
th
e
in
p
u
t
o
f
T
o
(
d
ataset
f
o
r
tr
ain
i
n
g
)
an
d
w
(
weig
h
t
)
,
w
h
ich
,
af
te
r
p
r
o
ce
s
s
in
g
,
y
ield
s
an
o
u
tco
m
e
o
f
up
(
u
p
d
ated
i
n
f
o
r
m
atio
n
o
f
t
h
r
ea
t
)
.
Neu
r
al
n
etwo
r
k
f
r
am
ewo
r
k
π
w
is
in
itialized
co
n
s
id
er
in
g
weig
h
t
s
w
o
f
ar
b
itra
r
y
f
o
r
m
(
L
in
e
-
1
)
is
u
s
ed
f
o
r
tr
ain
in
g
th
e
s
y
s
tem
co
n
s
id
er
in
g
a
p
ilo
t
s
am
p
le
o
f
in
d
e
x
ed
tr
a
f
f
ic
in
f
o
r
m
atio
n
th
at
co
n
s
is
t
s
o
f
in
p
u
t
tr
af
f
ic
v
ec
to
r
α
o
an
d
ass
o
ciate
d
in
d
ex
es
β
o
co
n
s
id
er
ed
f
o
r
p
ilo
t
ep
o
ch
r
o
u
n
d
e
o
(
L
in
e
-
1
an
d
L
in
e
-
2
)
.
T
h
e
v
ar
iab
le
p
a
r
a
m
r
e
p
r
esen
ts
in
p
u
t
tr
af
f
ic
v
ec
to
r
α
o,
ass
o
ciate
d
i
n
d
ex
es
β
o,
a
n
d
ep
o
c
h
e
o
(
L
i
n
e
-
2
)
.
T
h
e
i
n
p
u
t
tr
af
f
ic
v
ec
to
r
α
o
an
d
ass
o
cia
ted
in
d
e
x
es
β
o
ar
e
f
u
r
th
er
ass
ig
n
ed
t
o
th
e
n
ew
m
ap
p
in
g
attr
ib
u
te
o
f
in
p
u
t
tr
a
f
f
i
c
v
ec
to
r
α
an
d
ass
o
ciate
d
i
n
d
e
x
es
β
(
L
in
e
-
3
)
.
T
h
e
f
o
llo
win
g
lin
e
o
f
o
p
er
atio
n
co
n
s
id
er
s
all
th
e
in
p
u
t
tr
af
f
ic
v
ec
to
r
α
i
(
L
in
e
-
4)
,
f
o
llo
wed
b
y
an
ev
alu
atio
n
o
f
th
e
n
o
r
m
al
d
is
tr
ib
u
tio
n
N
(
L
in
e
-
5
)
.
Fo
r
th
is
p
u
r
p
o
s
e,
a
n
ex
p
licit
f
u
n
ctio
n
f
4
(
x
)
is
co
n
s
tr
u
cted
to
ass
ess
th
e
s
u
itab
ilit
y
o
f
n
o
r
m
al
d
is
tr
ib
u
t
io
n
co
n
s
id
er
in
g
p
a
r
a
m
1
r
ep
r
e
s
en
tin
g
p
ilo
t
in
p
u
t
tr
af
f
ic
v
ec
to
r
α
o
,
in
p
u
t
tr
a
f
f
ic
v
ec
to
r
α
,
an
d
n
eu
r
al
n
etwo
r
k
f
r
am
ewo
r
k
π
w
(
L
i
n
e
-
5
)
.
T
h
e
e
v
alu
ated
n
o
r
m
al
d
is
tr
ib
u
tio
n
s
u
itab
ilit
y
s
co
r
e
is
th
en
ass
ig
n
ed
to
m
atr
i
x
N
co
n
s
id
er
in
g
co
n
s
tr
u
ctio
n
o
p
er
at
o
r
co
,
r
ec
o
n
s
tr
u
ctio
n
o
p
e
r
ato
r
re
ass
o
ciate
d
with
r
eg
u
lar
tr
af
f
ic
r
an
d
m
alicio
u
s
tr
af
f
ic
m
(
L
in
e
-
5
)
.
I
n
s
im
p
le
wo
r
d
s
,
th
is
o
p
er
atio
n
(
L
in
e
-
5
)
ex
tr
ac
ts
th
e
n
o
r
m
al
d
is
tr
ib
u
tio
n
o
f
co
n
s
tr
u
ctio
n
o
p
er
ato
r
co
a
n
d
r
ec
o
n
s
tr
u
cto
r
o
p
er
ato
r
re
u
s
in
g
p
il
o
t tr
af
f
ic
in
f
o
r
m
atio
n
T
o
.
Fu
r
th
er
,
th
e
alg
o
r
ith
m
g
en
e
r
at
es
th
e
m
ask
ed
in
d
ex
(
L
in
e
-
5
t
o
L
in
e
-
8
)
f
o
llo
wed
b
y
th
e
g
en
er
atio
n
o
f
u
p
d
ated
in
f
o
r
m
atio
n
(
L
in
e
-
9
t
o
L
in
e
-
1
2
)
.
C
o
n
s
id
er
in
g
th
e
p
h
ases
in
v
o
lv
ed
in
T
r
ain
in
g
r
an
g
in
g
f
r
o
m
0
to
(
h
-
1
)
as
s
h
o
wn
in
L
in
e
-
6
,
t
h
e
alg
o
r
i
th
m
o
b
tain
s
r
esu
lt
s
r
es
co
n
ce
r
n
in
g
co
n
s
tr
u
ctio
n
o
p
er
at
o
r
co
an
d
r
ec
o
n
s
t
r
u
ctio
n
o
p
er
ato
r
r
e
(
L
in
e
-
7
)
.
I
t
ev
en
t
u
ally
in
f
er
s
to
war
d
s
ac
co
m
p
lis
h
in
g
r
esu
lts
co
n
ce
r
n
in
g
i
n
p
u
t
tr
af
f
ic
v
ec
to
r
α
(
L
in
e
-
7)
.
I
n
th
e
co
n
s
ec
u
tiv
e
p
r
o
ce
s
s
,
th
e
alg
o
r
ith
m
allo
ca
te
s
th
e
m
ask
ed
in
d
e
x
to
all
th
e
r
ec
en
tly
g
en
er
ated
in
p
u
t
tr
af
f
ic
α
j
th
at
h
as
alr
ea
d
y
b
ee
n
d
etec
ted
i
n
th
e
f
ir
s
t
alg
o
r
ith
m
u
s
in
g
th
e
Neu
r
al
n
etwo
r
k
f
r
am
ewo
r
k
π
w.
T
h
e
f
u
n
ctio
n
f
5
(
x
)
is
m
ea
n
t
f
o
r
u
n
d
er
ta
k
in
g
a
d
ec
is
io
n
co
n
s
id
er
in
g
n
ewly
g
en
er
ated
r
esu
lt
r
es
in
th
e
p
r
ev
io
u
s
s
tep
co
n
ce
r
n
in
g
p
a
r
a
m
2
(
L
in
e
-
8
)
th
at
ca
n
b
e
f
u
r
th
er
e
m
p
ir
ica
lly
ex
p
r
ess
ed
as
(
4
)
:
(
)
+
→
(
)
,
(
,
)
,
(
,
)
(
4
)
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
.
6
,
Decem
b
e
r
20
25
:
6
0
4
3
-
6
0
5
2
6048
(
)
+
→
(
)
,
(
,
)
,
(
,
)
(
5
)
I
n
th
e
e
x
p
r
ess
io
n
(
2
)
an
d
(
3
)
,
th
e
{
(
)
,
(
,
)
,
(
,
)
}
an
d
{
(
)
,
(
,
)
,
(
,
)
}
to
g
eth
er
c
o
n
s
titu
te
o
f
p
a
r
a
m
2
.
T
h
e
f
u
n
ctio
n
f
5
(
x
)
g
en
e
r
ates
a
v
alu
e
o
f
m
ask
in
g
i
n
d
ex
β
a
s
1
if
it
f
in
d
s
er
r
o
r
-
p
r
o
n
e
in
p
u
t
f
o
llo
we
d
b
y
a
f
lag
g
in
g
aler
t
to
th
e
s
y
s
tem
.
I
t
will
also
m
ea
n
t
h
at
th
e
p
r
ev
i
o
u
s
alg
o
r
ith
m
p
lay
s
th
e
r
o
le
o
f
in
f
e
r
en
ce
to
y
ield
o
u
tc
o
m
e
s
in
th
is
m
o
d
el
wh
ile
p
er
f
o
r
m
in
g
g
en
er
atio
n
f
o
r
m
ask
e
d
in
d
ex
es.
On
ly
th
e
o
u
tco
m
e
o
f
m
ask
ed
in
d
ex
β
i+
j
th
at
is
f
o
u
n
d
to
h
av
e
a
h
ig
h
er
s
co
r
e
is
co
n
s
id
er
ed
f
u
r
t
h
er
,
f
o
l
lo
wed
b
y
r
a
n
d
o
m
l
y
p
er
m
u
tin
g
m
ask
ed
in
d
e
x
β
.
T
h
is
alg
o
r
ith
m
'
s
f
in
al
s
tep
i
s
u
p
d
atin
g
b
o
th
in
p
u
t
tr
af
f
ic
v
ec
t
o
r
α
an
d
ass
o
ciate
d
m
ask
ed
in
d
ex
es
β
(
L
in
e
-
9
)
.
F
o
r
th
is
p
u
r
p
o
s
e,
f
u
n
ctio
n
f
6
(
x
)
h
as
b
ee
n
co
n
s
tr
u
cted
,
wh
ich
is
m
ain
ly
p
er
f
o
r
m
in
g
u
n
io
n
o
p
er
atio
n
s
b
etwe
en
α
,
α
i+
j
,
an
d
β
i+
j
wh
ile
th
e
T
r
ain
in
g
is
ca
r
r
ied
o
u
t
o
n
p
a
r
a
m
2
,
wh
ic
h
co
n
s
is
t
s
o
f
α
,
β
,
π
w
,
an
d
n
ew
ep
o
c
h
e
1
(
L
i
n
e
-
1
2
)
.
I
t
ca
n
b
e
s
ee
n
t
h
at
tr
ain
in
g
d
ataset
T
is
n
o
w
ex
p
an
d
ed
wit
h
th
e
in
clu
s
io
n
o
f
a
n
ew
in
p
u
t
tr
af
f
ic
v
ec
to
r
α
j
a
n
d
its
ass
o
ciate
d
m
ask
ed
i
n
d
ex
β
.
C
o
n
s
id
er
in
g
h
n
u
m
b
er
o
f
in
co
m
in
g
tr
af
f
ic
b
ein
g
n
ewly
id
en
tifie
d
,
n
ew
ep
o
ch
e
1
is
u
s
ed
f
o
r
ad
ju
s
tin
g
th
e
n
eu
r
al
n
etwo
r
k
m
o
d
el
π
w
,
wh
ich
ev
en
tu
ally
m
ea
n
s
th
at
th
e
f
in
al
i
n
d
ex
co
n
s
is
t
s
o
f
b
o
th
g
r
o
u
n
d
-
tr
u
th
in
f
o
r
m
atio
n
in
d
ex
an
d
m
ask
ed
i
n
d
ex
.
T
h
e
f
o
llo
win
g
s
ec
tio
n
d
is
cu
s
s
es a
b
o
u
t a
cc
o
m
p
lis
h
ed
r
esu
lts
.
B
y
d
o
in
g
awa
y
with
m
a
n
u
al
lab
elin
g
a
n
d
o
n
g
o
in
g
r
etr
ai
n
in
g
,
t
h
e
m
ask
ed
in
d
e
x
in
g
m
ec
h
an
is
m
p
r
esen
ts
a
r
ev
o
lu
tio
n
a
r
y
m
eth
o
d
f
o
r
d
y
n
am
ic
th
r
ea
t
d
etec
ti
o
n
.
B
y
em
p
lo
y
in
g
"m
ask
e
d
in
d
ex
es"
to
co
n
ce
al
tr
af
f
ic
lab
els
an
d
th
war
t
lab
el
m
an
ip
u
latio
n
b
y
ad
v
er
s
ar
ies,
i
t
en
ab
les
th
e
s
y
s
tem
to
a
u
to
m
atica
lly
id
en
tify
an
d
ad
ju
s
t
to
ch
an
g
in
g
th
r
ea
ts
in
r
ea
l
-
tim
e.
E
v
en
in
d
y
n
a
m
ic,
lab
el
-
f
r
ee
s
itu
atio
n
s
,
wh
ich
ar
e
ty
p
ical
o
f
I
o
T
n
etwo
r
k
s
,
th
is
ap
p
r
o
ac
h
g
u
ar
a
n
tees
r
eliab
le
th
r
ea
t
d
etec
tio
n
.
B
ec
au
s
e
iter
ativ
e
r
e
-
tr
ain
in
g
o
f
th
e
s
y
s
tem
is
n
o
t
n
ec
ess
ar
y
,
it
also
r
ed
u
ce
s
co
m
p
u
tin
g
o
v
er
h
ea
d
,
in
cr
ea
s
in
g
ef
f
icien
cy
.
B
ec
au
s
e
o
f
th
e
in
cr
ea
s
ed
ac
cu
r
ac
y
,
s
h
o
r
ter
p
r
o
ce
s
s
in
g
tim
es,
an
d
m
o
r
e
t
r
u
s
two
r
th
y
th
r
ea
t
d
etec
t
io
n
th
at
r
esu
lts
,
it
is
esp
ec
ially
u
s
ef
u
l
in
s
ettin
g
s
wh
er
e
d
ata
p
atter
n
s
ar
e
ev
er
-
c
h
an
g
in
g
.
3.
RE
SU
L
T
T
h
e
s
cr
ip
tin
g
o
f
th
e
p
r
o
p
o
s
ed
s
tu
d
y
is
ca
r
r
ied
o
u
t
o
n
a
p
y
th
o
n
en
v
ir
o
n
m
en
t
o
n
a
n
o
r
m
al
6
4
-
b
it
W
in
d
o
ws
en
v
ir
o
n
m
en
t
co
n
s
id
er
in
g
two
s
tan
d
ar
d
b
en
ch
m
a
r
k
ed
d
atasets
im
p
lem
en
ted
o
n
a
s
tan
d
ar
d
6
4
-
b
it
W
in
d
o
ws
s
y
s
tem
with
NVI
DI
A
GeFo
r
ce
GT
X
with
1
6
GB
R
AM
an
d
I
n
tel
C
o
r
e
i5
p
r
o
ce
s
s
o
r
.
T
h
e
en
v
ir
o
n
m
en
t
was
r
etain
ed
th
e
s
am
e
f
o
r
ass
ess
in
g
th
e
p
r
o
p
o
s
ed
ex
is
tin
g
s
y
s
tem
.
T
h
e
f
ir
s
t
d
ataset
is
th
e
NSL
-
KDD
d
ataset
[
3
1
]
,
wh
e
r
e
2
5
class
es
o
f
m
alicio
u
s
f
o
r
m
s
ar
e
co
n
s
id
er
ed
f
o
r
tr
ain
in
g
o
p
er
atio
n
s
en
ca
p
s
u
l
a
tin
g
attac
k
er
s
th
at
attem
p
t
to
ex
p
lo
r
e
v
ar
i
o
u
s
ty
p
es
o
f
attac
k
s
as
f
o
llo
ws:
i)
Attack
er
T
y
p
e
-
1
:
T
h
is
ty
p
e
o
f
attac
k
er
s
ex
p
lo
it
th
e
wea
k
n
ess
o
f
n
etwo
r
k
s
tr
u
ctu
r
e
,
e.
g
.
,
Po
r
t
s
ca
n
n
in
g
,
p
in
g
s
wee
p
in
g
,
n
etwo
r
k
m
ap
p
in
g
,
s
er
v
ice
en
u
m
er
atio
n
,
v
u
ln
er
ab
i
li
ty
s
ca
n
n
in
g
.
ii)
Attack
er
T
y
p
e
-
2
:
T
h
is
ty
p
e
o
f
cy
b
er
-
attac
k
to
g
ain
illeg
itima
te
ac
ce
s
s
to
co
n
tr
o
l
t
h
e
n
etwo
r
k
o
r
m
ac
h
in
e
o
f
th
e
v
ictim
n
o
d
e
r
em
o
tely
,
e.
g
.
,
SQL
in
jectio
n
,
b
u
f
f
er
o
v
er
f
lo
w,
r
em
o
te
co
d
e
ex
ec
u
ti
o
n
.
iii)
Attack
er
T
y
p
e
-
3
:
T
h
is
ty
p
e
o
f
cy
b
er
-
attac
k
is
u
s
ed
f
o
r
g
ain
in
g
ac
ce
s
s
to
r
o
o
t
ad
m
in
is
tr
ativ
e
ac
co
u
n
t
s
illeg
ally
,
e.
g
.
,
Ker
n
el
e
x
p
lo
its
,
m
is
co
n
f
ig
u
r
atio
n
o
f
s
et
g
r
o
u
p
I
D
(
SGI
D)
o
r
s
et
u
s
er
I
D
(
SUI
D)
.
i
v
)
Attack
er
T
y
p
e
-
4
:
T
h
is
attac
k
er
is
m
ea
n
t
to
f
lo
o
d
illeg
itima
te
r
eq
u
est
s
in
o
r
d
er
to
d
is
r
u
p
t
th
e
n
o
r
m
al
s
er
v
ices
f
r
o
m
a
s
er
v
er
,
e.
g
.
,
Vo
lu
m
e
-
b
ased
attac
k
s
,
p
r
o
to
co
l
attac
k
s
,
ap
p
licatio
n
lay
er
attac
k
s
,
d
en
ial
-
of
-
s
er
v
ice
(
Do
S),
an
d
d
is
tr
ib
u
ted
Do
S
(
DDo
S).
T
h
e
s
ec
o
n
d
d
ataset
is
th
e
UNSW
-
NB
1
5
d
ataset
[
3
2
]
,
with
m
o
r
e
th
an
1
0
0
GB
o
f
r
a
w
tr
af
f
ic
in
f
o
r
m
atio
n
ex
tr
ac
ted
f
r
o
m
ar
t
i
f
icially
co
n
s
tr
u
cted
ad
v
er
s
ar
ies
with
9
class
es
o
f
attac
k
with
4
9
f
ea
tu
r
es.
T
h
is
d
ataset
p
r
o
v
id
es
v
u
ln
er
ab
le
en
v
ir
o
n
m
e
n
tal
ass
ess
m
en
t
f
o
r
Do
S,
wo
r
m
s
,
an
d
m
alwa
r
e
.
T
h
er
e
is
a
to
tal
o
f
1
7
5
,
3
4
1
,
an
d
8
2
,
3
3
2
r
ec
o
r
d
s
u
s
ed
f
o
r
tr
ain
in
g
an
d
test
in
g
.
T
h
e
p
r
o
p
o
s
ed
s
ch
em
e
u
s
es
5
lay
e
r
s
with
f
o
llo
win
g
s
izes
i)
f
o
r
f
ir
s
t
d
ataset
(
s
ize
o
f
i
n
p
u
t
lay
er
=
1
2
1
,
s
ize
o
f
co
n
s
tr
u
cto
r
lay
er
=
6
4
,
s
ize
o
f
h
id
d
en
la
y
er
=
3
2
,
s
ize
o
f
r
ec
o
n
s
t
r
u
cto
r
lay
er
=
6
4
,
s
ize
o
f
o
u
tp
u
t
lay
er
=
1
2
1
)
,
an
d
ii)
f
o
r
s
ec
o
n
d
d
ataset
(
s
iz
e
o
f
in
p
u
t
lay
e
r
=
1
9
6
,
s
ize
o
f
c
o
n
s
tr
u
cto
r
la
y
er
=
1
2
8
,
s
ize
o
f
h
i
d
d
en
lay
e
r
=
6
4
,
s
ize
o
f
r
ec
o
n
s
tr
u
ct
o
r
lay
e
r
=
1
2
8
,
s
ize
o
f
o
u
tp
u
t
lay
e
r
=
1
9
6
)
.
T
h
e
ad
ju
s
tm
en
t
p
ar
a
m
e
ter
δ
is
s
et
to
0
.
0
2
with
1
2
8
b
atch
s
ize
,
wh
ile
0
.
0
0
1
is
s
et
f
o
r
t
h
e
r
ate
o
f
lear
n
i
n
g
co
n
s
id
er
in
g
r
ec
tili
n
ea
r
u
n
it
(
R
eL
U)
as
th
e
ac
tiv
atio
n
f
u
n
ctio
n
.
T
h
e
s
ch
em
e
u
s
es g
r
a
d
ien
t d
escen
t o
f
s
to
c
h
asti
c
f
o
r
m
to
p
lay
th
e
r
o
le
o
f
o
p
tim
izat
io
n
in
th
e
p
r
o
p
o
s
ed
m
ac
h
in
e
lear
n
in
g
m
o
d
el.
T
h
e
NSL
-
KDD
d
ataset
co
n
s
is
ts
o
f
n
etwo
r
k
-
b
ased
attac
k
s
,
wh
ile
th
e
UNSW
-
NB
-
15
d
ataset
co
n
s
is
ts
o
f
m
o
r
e
d
iv
er
s
e
attac
k
co
v
er
ag
e
o
n
m
o
r
e
r
ea
lis
tic
s
ce
n
ar
io
s
o
f
I
o
T
.
Ad
o
p
tio
n
o
f
th
ese
d
atasets
f
ac
ili
t
ates
to
s
im
u
lat
io
n
o
f
g
en
er
al
attac
k
s
b
ased
o
n
n
etwo
r
k
v
u
ln
e
r
ab
ilit
ies
in
I
o
T
.
T
h
e
o
u
tco
m
e
is
ass
es
s
ed
f
o
r
v
ar
ied
ac
cu
r
ac
y
p
ar
am
eter
s
(
p
r
ec
is
io
n
,
r
ec
all,
an
d
F1
-
s
co
r
e)
,
o
v
e
r
all
ac
cu
r
ac
y
,
an
d
alg
o
r
ith
m
p
r
o
ce
s
s
in
g
tim
e.
T
h
e
b
e
n
ch
m
ar
k
in
g
is
ca
r
r
ied
o
u
t
b
y
c
o
m
p
ar
in
g
th
e
p
r
o
p
o
s
ed
s
y
s
tem
wit
h
ex
is
tin
g
m
ac
h
in
e
lear
n
in
g
al
g
o
r
ith
m
s
th
at
ar
e
f
r
eq
u
en
tly
r
ep
o
r
ted
in
th
e
liter
at
u
r
e
,
e.
g
.
,
XGBo
o
s
t,
SVM,
R
F,
an
d
DT
as
s
h
o
wn
in
Fig
u
r
e
s
2
to
6
.
T
h
e
o
u
tco
m
e
o
f
th
e
an
aly
s
is
ev
en
tu
ally
s
h
o
ws
th
e
p
r
o
p
o
s
ed
s
ch
em
e
o
f
a
n
eu
r
al
n
etwo
r
k
to
p
er
f
o
r
m
b
etter
p
er
f
o
r
m
a
n
ce
in
co
n
tr
ast
to
ex
is
tin
g
m
ac
h
in
e
l
ea
r
n
in
g
ap
p
r
o
ac
h
es
with
r
esp
e
ct
to
ac
cu
r
ac
y
a
n
d
p
r
o
ce
s
s
in
g
tim
e
-
b
ased
e
v
alu
at
io
n
m
etr
ic.
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
N
eu
r
a
l
-
n
etw
o
r
k
b
a
s
ed
r
ep
r
es
en
ta
tio
n
fr
a
mewo
r
k
fo
r
a
d
ve
r
s
a
r
y
…
(
Th
a
n
u
j
a
N
a
r
a
s
imh
a
mu
r
th
y
)
6049
Fig
u
r
e
2
.
Acc
u
r
ac
y
f
o
r
NSL
-
KDD
d
ataset
Fig
u
r
e
3
.
Acc
u
r
ac
y
f
o
r
UNSW
-
NB
1
5
d
ataset
Fig
u
r
e
4
.
Acc
u
r
ac
y
s
co
r
e
f
o
r
NSL
-
KDD
d
ataset
Fig
u
r
e
5
.
Acc
u
r
ac
y
s
co
r
e
f
o
r
UNSW
-
N
B
1
5
d
ataset
Dis
cu
s
s
io
n
o
f
ac
cu
r
ac
y
o
u
tc
o
m
es
is
illu
s
tr
ated
in
Fig
u
r
e
s
2
to
5
.
T
h
e
q
u
a
n
tific
atio
n
o
f
ac
cu
r
ac
y
o
u
tco
m
es
s
h
o
ws
th
at
th
er
e
is
n
o
s
ig
n
if
ican
t
d
if
f
er
e
n
ce
in
ac
cu
r
ac
y
o
u
tco
m
es
f
o
r
th
e
f
ir
s
t
an
d
s
ec
o
n
d
d
ataset
s
.
T
h
e
p
r
o
p
o
s
ed
s
y
s
tem
is
f
o
u
n
d
to
o
f
f
er
a
p
p
r
o
x
im
ately
1
2
%
an
d
1
0
%
b
etter
ac
c
u
r
ac
y
s
co
r
e
s
r
ec
o
r
d
ed
f
o
r
th
e
f
ir
s
t
an
d
s
ec
o
n
d
d
ataset
s,
r
esp
ec
tiv
ely
,
in
co
n
tr
ast
to
ex
is
tin
g
m
ac
h
i
n
e
lear
n
i
n
g
a
p
p
r
o
ac
h
e
s
.
T
h
e
ju
s
tific
atio
n
is
as
f
o
llo
ws:
Alth
o
u
g
h
t
h
e
XGBo
o
s
t
ap
p
r
o
ac
h
is
k
n
o
wn
f
o
r
its
h
ig
h
er
ac
cu
r
ac
y
p
er
f
o
r
m
an
ce
with
b
etter
h
an
d
lin
g
ca
p
ac
ity
o
f
im
b
alan
ce
d
d
ata,
it
is
th
e
co
m
p
u
tatio
n
ally
in
ten
s
iv
e
m
o
d
el
wh
e
n
e
x
p
o
s
ed
to
a
lar
g
e
r
d
ataset.
I
t
wo
r
k
s
well
f
o
r
s
m
all
d
ataset
s,
b
u
t
i
ts
ac
cu
r
ac
y
s
ta
r
ts
d
ec
lin
in
g
wh
en
ex
p
o
s
ed
to
s
tr
ea
m
s
o
f
d
ata.
A
s
im
ilar
p
r
o
b
lem
is
also
en
co
u
n
ter
ed
f
o
r
t
h
e
SVM
ap
p
r
o
ac
h
,
an
d
its
p
er
f
o
r
m
an
ce
is
p
o
te
n
tially
d
ep
en
d
en
t
o
n
an
ap
p
r
o
p
r
iate
k
er
n
el
f
u
n
ctio
n
s
elec
tio
n
.
Alth
o
u
g
h
th
is
s
elec
tio
n
is
s
o
m
ewh
at
ea
s
ier
f
o
r
s
m
aller
an
d
s
tatic
s
et
s
o
f
d
ata
its
ac
cu
r
ac
y
s
tar
ts
d
eg
r
ad
in
g
wh
e
n
ass
ess
ed
with
in
cr
ea
s
in
g
r
an
d
o
m
s
izes
o
f
in
co
m
in
g
tr
af
f
ic
o
f
d
ata.
T
h
e
p
er
f
o
r
m
a
n
ce
o
f
R
F
is
f
o
u
n
d
to
b
e
q
u
ite
g
o
o
d
in
ac
cu
r
a
cy
f
o
r
th
e
s
ec
o
n
d
d
ataset
,
wh
i
le
th
e
DT
a
p
p
r
o
ac
h
is
f
o
u
n
d
to
b
e
b
etter
f
o
r
b
o
t
h
d
ataset
s
.
Ho
wev
er
,
th
e
DT
ap
p
r
o
ac
h
ca
n
n
o
t
b
e
co
n
s
id
e
r
ed
s
u
itab
le
in
th
is
im
p
lem
en
tatio
n
s
ce
n
ar
i
o
as
it
ca
n
lead
to
a
h
ig
h
er
d
eg
r
ee
o
f
in
s
tab
ilit
y
in
ca
s
e
o
f
s
m
aller
ch
an
g
es
in
d
ata
.
Ho
wev
er
,
t
h
e
p
r
o
p
o
s
ed
s
ch
e
m
e
o
f
f
er
s
b
etter
ac
cu
r
ac
y
m
a
in
ly
b
ec
a
u
s
e
o
f
th
e
s
ec
o
n
d
al
g
o
r
ith
m
to
war
d
s
th
e
u
p
d
atin
g
p
r
o
ce
s
s
,
m
in
im
izin
g
all
th
e
ex
tr
a
tim
e
r
eq
u
ir
e
d
t
o
f
i
n
d
o
p
tim
al
r
esu
lt
s
.
Owin
g
to
f
itm
en
t
test
in
g
with
p
r
o
ce
s
s
ed
f
ea
tu
r
e
s
with
n
o
r
m
al
d
is
tr
ib
u
tio
n
,
t
h
e
d
etec
tio
n
o
u
tco
m
e
is
q
u
ite
r
eliab
le
with
r
esp
ec
t
to
its
o
v
er
all
ac
cu
r
ac
y
s
co
r
e.
T
h
e
ad
o
p
tio
n
o
f
co
r
r
elatio
n
al
s
co
r
e
b
etwe
en
tr
ain
ed
r
e
p
r
esen
tatio
n
o
f
f
e
r
s
a
b
etter
f
o
r
m
o
f
d
ata
r
ep
r
esen
tatio
n
co
n
s
id
er
i
n
g
o
p
tim
al
co
s
t
f
u
n
ctio
n
ass
o
ciate
d
with
th
e
co
n
s
tr
u
ct
o
r
an
d
r
ec
o
n
s
tr
u
ct
o
r
o
p
er
ato
r
.
T
h
is
is
f
u
r
th
er
ju
s
tifie
d
b
y
e
v
alu
atin
g
its
co
m
p
u
tatio
n
al
ef
f
icien
cy
with
r
e
s
p
ec
t
to
alg
o
r
ith
m
p
r
o
ce
s
s
in
g
tim
e
r
esu
lts
ex
h
ib
i
t
ed
in
Fig
u
r
e
6.
Dis
cu
s
s
io
n
o
f
alg
o
r
ith
m
p
r
o
c
ess
in
g
tim
e
i
s
illu
s
tr
ated
in
F
ig
u
r
e
6
.
T
h
e
q
u
an
tifie
d
o
u
tco
m
e
s
h
o
ws
th
at
th
e
p
r
o
p
o
s
ed
s
y
s
tem
o
f
f
er
s
ap
p
r
o
x
im
ately
2
8
%
an
d
3
8
%
o
f
r
ed
u
ce
d
alg
o
r
ith
m
p
r
o
ce
s
s
in
g
tim
e
i
n
co
m
p
ar
is
o
n
t
o
th
e
m
ea
n
o
f
ex
is
tin
g
m
ac
h
in
e
lea
r
n
in
g
alg
o
r
ith
m
s
f
o
r
th
e
f
ir
s
t
an
d
s
ec
o
n
d
d
ataset
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
.
6
,
Decem
b
e
r
20
25
:
6
0
4
3
-
6
0
5
2
6050
r
esp
ec
tiv
ely
.
T
h
e
ju
s
tific
atio
n
b
eh
in
d
th
is
o
u
tc
o
m
e
is
as
f
o
llo
ws:
B
o
th
XG
B
o
o
s
t
an
d
S
VM
ar
e
n
o
ted
with
co
n
s
id
er
ab
ly
h
ig
h
e
r
alg
o
r
ith
m
p
r
o
ce
s
s
in
g
tim
e
,
wh
ich
is
m
ai
n
ly
d
u
e
to
t
h
eir
ex
te
n
s
iv
e
h
y
p
er
p
ar
am
eter
t
u
n
in
g
an
d
s
lo
wer
co
m
p
u
tatio
n
al
p
r
o
ce
s
s
r
esp
ec
tiv
ely
.
Fo
r
b
o
th
d
a
taset
s
,
th
eir
p
er
f
o
r
m
an
ce
is
n
ea
r
ly
th
e
s
am
e.
R
F
alg
o
r
ith
m
g
e
n
er
ates
a
lar
g
e
n
u
m
b
er
o
f
tr
ee
s
th
at
ev
e
n
tu
ally
lead
s
to
in
cr
ea
s
ed
co
m
p
lex
it
y
esp
ec
ially
d
u
r
i
n
g
th
e
tr
ain
in
g
o
p
er
atio
n
.
DT
h
a
s
b
etter
p
er
f
o
r
m
an
ce
in
co
n
tr
ast
to
XG
B
o
o
s
t,
SVM,
an
d
R
F
,
wh
ich
is
m
ain
ly
d
u
e
to
its
in
d
ep
en
d
en
ce
f
r
o
m
n
o
r
m
aliza
tio
n
o
r
f
ea
tu
r
e
s
ca
lin
g
s
tep
;
h
o
wev
er
,
w
h
en
ex
p
o
s
ed
to
u
n
in
d
ex
ed
d
ata,
it
is
n
o
ted
to
b
e
p
r
o
n
e
t
o
o
v
er
f
itti
n
g
,
lead
in
g
to
in
f
er
io
r
g
en
e
r
aliza
tio
n
o
f
u
n
s
ee
n
d
ata.
Ho
wev
er
,
t
h
e
m
eth
o
d
o
l
o
g
y
in
v
o
l
v
ed
in
th
e
p
r
o
p
o
s
ed
s
ch
em
e
ca
lls
f
o
r
u
n
s
u
p
er
v
is
ed
lear
n
in
g
,
wh
ile
t
h
e
id
en
tific
atio
n
o
f
o
u
tlier
s
is
ea
s
ily
d
o
n
e
b
y
esti
m
atin
g
r
ec
o
n
s
tr
u
ctio
n
er
r
o
r
wi
th
o
u
t
m
u
c
h
d
ep
e
n
d
en
c
y
o
n
la
b
elled
d
ata.
Fu
r
th
e
r
,
th
e
m
o
d
el
is
h
ig
h
l
y
ad
a
p
tiv
e
t
o
co
m
p
lex
p
atter
n
s
o
f
n
ew
t
y
p
es
o
f
in
tr
u
s
io
n
b
y
f
i
n
etu
n
in
g
t
h
e
d
ata
,
as
n
o
ted
in
th
e
s
ec
o
n
d
alg
o
r
ith
m
o
f
th
e
p
r
o
p
o
s
ed
s
y
s
tem
.
Hen
ce
,
wit
h
o
u
t
t
h
e
in
cl
u
s
io
n
o
f
a
n
y
s
o
p
h
is
ticated
iter
ativ
e
s
tep
s
,
th
e
alg
o
r
ith
m
ca
n
ca
r
r
y
o
u
t
p
r
o
g
r
ess
iv
e
o
p
er
atio
n
s
to
id
en
tify
ev
en
s
m
aller
ch
an
g
es
in
tr
af
f
ic
with
h
ig
h
er
ac
cu
r
ac
y
.
T
h
is
g
r
o
u
n
d
o
f
f
ac
t
is
attr
ib
u
te
d
to
war
d
s
r
ed
u
ce
d
alg
o
r
ith
m
p
r
o
c
ess
in
g
tim
e
f
o
r
th
e
p
r
o
p
o
s
ed
m
ac
h
in
e
lear
n
in
g
-
b
ased
s
ec
u
r
ity
s
ch
em
e.
Fig
u
r
e
6
.
Pro
ce
s
s
in
g
tim
e
a
n
al
y
s
is
4.
CO
NCLU
SI
O
N
T
h
e
ad
o
p
tio
n
o
f
m
ac
h
in
e
lea
r
n
in
g
alg
o
r
ith
m
s
h
as
b
ee
n
f
r
e
q
u
en
tly
witn
ess
ed
in
ex
is
tin
g
liter
atu
r
e
to
war
d
s
th
r
ea
t
d
etec
tio
n
o
n
v
ar
io
u
s
f
o
r
m
s
o
f
n
etwo
r
k
s
y
s
tem
s
.
Kee
p
in
g
asid
e
th
e
p
o
t
en
tial
ca
pa
b
ilit
y
o
f
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
t
o
war
d
s
ad
d
r
ess
in
g
s
ec
u
r
ity
th
r
ea
ts
,
o
n
e
o
f
th
e
ch
allen
g
in
g
p
r
o
b
lem
s
id
en
tifie
d
is
th
at
–
a
tr
ain
e
d
m
ac
h
in
e
le
ar
n
in
g
m
o
d
el
is
n
o
t
s
u
f
f
icien
t
to
m
itig
ate
d
y
n
am
ic
f
o
r
m
s
o
f
th
r
ea
ts
,
wh
ich
is
a
m
ajo
r
g
a
p
in
th
e
liter
atu
r
e.
T
h
is
g
ap
is
ad
d
r
ess
ed
in
p
r
o
p
o
s
ed
s
y
s
tem
b
y
f
o
llo
win
g
n
o
tab
le
co
n
tr
i
b
u
tio
n
:
i)
p
r
o
p
o
s
ed
s
ch
em
e
im
p
le
m
en
ts
a
h
ig
h
ly
ad
ap
tab
le
an
d
f
lex
ib
le
ar
ch
itectu
r
e
wh
ich
is
ca
p
ab
le
o
f
id
en
tify
in
g
d
y
n
am
ic
th
r
ea
ts
with
f
aster
r
e
s
p
o
n
s
e
an
d
h
i
g
h
er
ac
cu
r
ac
y
,
ii)
n
eu
r
al
n
etwo
r
k
-
b
ased
lear
n
in
g
m
o
d
el
h
as
b
ee
n
d
esig
n
ed
with
a
n
in
clu
s
io
n
o
f
two
ad
d
itio
n
al
la
y
er
s
(
co
n
s
tr
u
ctio
n
o
p
e
r
ato
r
a
n
d
r
ec
o
n
s
tr
u
ct
o
r
o
p
er
ato
r
)
wh
ich
ass
is
ts
to
war
d
s
b
etter
f
o
r
m
o
f
lear
n
in
g
r
e
p
r
esen
tatio
n
o
f
v
a
r
iab
le
an
d
u
n
ce
r
tain
tr
af
f
ic
p
a
tter
n
,
iii)
p
r
o
p
o
s
ed
s
ch
em
e
is
co
m
p
letely
f
r
ee
f
r
o
m
an
y
f
o
r
m
o
f
h
u
m
a
n
in
ter
v
e
n
tio
n
wh
ic
h
ca
n
u
n
d
er
tak
e
its
o
wn
d
ec
is
io
n
th
r
ea
t
d
etec
tio
n
an
d
u
p
d
atin
g
p
r
o
ce
s
s
,
iv
)
an
in
n
o
v
ativ
e
co
n
ce
p
t o
f
m
ask
ed
in
d
ex
h
as b
ee
n
in
tr
o
d
u
ce
d
to
ad
d
r
ess
th
e
task
o
f
lab
ellin
g
s
o
th
at
an
i
n
v
o
lu
n
tar
y
t
h
r
ea
t
d
etec
tio
n
s
ch
e
m
e
ca
n
b
e
d
e
f
in
ed
o
n
d
y
n
am
i
c
en
v
ir
o
n
m
en
t,
an
d
v
)
p
r
o
p
o
s
ed
s
y
s
tem
o
f
f
er
s
ap
p
r
o
x
im
ately
m
ea
n
o
f
1
1
%
in
cr
ea
s
ed
ac
cu
r
ac
y
a
n
d
3
3
%
o
f
r
ed
u
ce
d
p
r
o
ce
s
s
in
g
tim
e
wh
en
co
m
p
ar
ed
with
f
r
eq
u
e
n
tly
u
s
ed
m
ac
h
in
e
le
ar
n
in
g
to
wa
r
d
s
cy
b
e
r
s
ec
u
r
i
ty
.
Fo
r
r
ea
l
-
wo
r
ld
ap
p
licab
ilit
y
,
th
e
m
o
d
el
n
ee
d
s
to
b
e
s
y
n
ch
r
o
n
ized
with
ex
is
tin
g
ed
g
e
d
e
v
ices
f
o
r
p
a
r
ticu
lar
s
er
v
ice
p
r
o
v
i
d
er
s
o
f
I
o
T
-
clo
u
d
s
y
s
tem
s
.
Var
io
u
s
ch
allen
g
es
th
at
co
u
ld
s
u
r
f
ac
e
ar
e
h
an
d
lin
g
an
d
p
r
o
ce
s
s
in
g
m
ass
iv
e
in
co
m
in
g
tr
af
f
ic
f
r
o
m
I
o
T
d
ev
ices
with
r
esp
ec
t
to
d
esig
n
ated
ed
g
e
d
ev
ice
s
.
T
h
e
co
m
p
lete
s
y
s
tem
o
p
er
atio
n
is
ca
r
r
ied
o
u
t
co
n
s
id
er
in
g
f
au
lt
-
to
ler
an
t
p
r
o
ce
s
s
in
g
ca
r
r
ied
o
u
t
with
in
ed
g
e
d
ev
ice
s
an
d
s
er
v
er
s
th
at
ex
ec
u
te
alg
o
r
ith
m
s
in
a
cl
o
u
d
en
v
ir
o
n
m
e
n
t.
T
h
is
d
ep
e
n
d
en
c
y
c
o
u
ld
b
e
o
n
e
o
f
th
e
p
r
im
e
lim
itatio
n
s
alth
o
u
g
h
it
will
n
o
t
af
f
ec
t
alg
o
r
ith
m
ic
p
er
f
o
r
m
an
ce
.
He
n
ce
.
F
u
tu
r
e
wo
r
k
will
b
e
co
n
tin
u
ed
to
war
d
s
im
p
r
o
v
in
g
th
e
s
am
e
m
o
d
el
with
m
o
r
e
p
ar
am
et
r
ic
in
clu
s
io
n
o
f
clo
u
d
attr
ib
u
tes
an
d
e
d
g
e
d
ev
ice
attr
ib
u
tes
ap
ar
t
f
r
o
m
I
o
T
d
ev
ices,
wh
ich
was
th
e
p
r
im
e
f
o
cu
s
o
f
th
is
s
tu
d
y
.
T
h
e
f
u
tu
r
e
w
o
r
k
will
h
av
e
a
n
in
clu
s
io
n
t
o
war
d
s
n
o
v
el
f
o
r
m
u
latio
n
o
f
th
r
ea
t
p
r
ev
en
tio
n
s
tr
ateg
y
co
n
s
id
er
in
g
lar
g
e
I
o
T
en
v
ir
o
n
m
en
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:
2088
-
8
7
0
8
N
eu
r
a
l
-
n
etw
o
r
k
b
a
s
ed
r
ep
r
es
en
ta
tio
n
fr
a
mewo
r
k
fo
r
a
d
ve
r
s
a
r
y
…
(
Th
a
n
u
j
a
N
a
r
a
s
imh
a
mu
r
th
y
)
6051
ACK
NO
WL
E
DG
E
M
E
NT
S
W
e
wis
h
to
co
n
f
ir
m
th
at
n
o
k
n
o
wn
co
n
f
licts
o
f
in
ter
est
ar
e
ass
o
ciate
d
with
th
is
p
u
b
licati
o
n
an
d
al
l
th
e
au
th
o
r
s
h
av
e
co
n
tr
ib
u
ted
e
q
u
ally
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
Au
th
o
r
s
s
tate
n
o
f
u
n
d
in
g
in
v
o
lv
ed
.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
T
h
is
jo
u
r
n
al
u
s
es
th
e
co
n
t
r
ib
u
to
r
r
o
les
tax
o
n
o
m
y
(
C
R
ed
iT)
to
r
ec
o
g
n
ize
in
d
iv
i
d
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
llab
o
r
atio
n
.
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
Th
a
n
u
ja Na
ra
sim
h
a
m
u
rth
y
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
G
u
n
a
v
a
th
i
Ho
sa
h
a
ll
i
S
wa
m
y
✓
✓
✓
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
Au
th
o
r
s
s
tate
n
o
co
n
f
lict o
f
in
t
er
est.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
d
a
t
a
t
h
a
t
s
u
p
p
o
r
t
t
h
e
f
i
n
d
i
n
g
s
o
f
t
h
i
s
s
t
u
d
y
a
r
e
a
v
a
il
a
b
l
e
o
n
r
e
q
u
e
s
t
f
r
o
m
t
h
e
c
o
r
r
e
s
p
o
n
d
in
g
a
u
t
h
o
r
.
RE
F
E
R
E
NC
E
S
[
1
]
A
.
S
.
M
a
s
h
a
l
e
h
e
t
a
l
.
,
“
E
v
a
l
u
a
t
i
o
n
o
f
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
n
d
d
e
e
p
l
e
a
r
n
i
n
g
m
e
t
h
o
d
s
f
o
r
e
a
r
l
y
d
e
t
e
c
t
i
o
n
o
f
i
n
t
e
r
n
e
t
o
f
t
h
i
n
g
s
b
o
t
n
e
t
s
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
E
l
e
c
t
ri
c
a
l
a
n
d
C
o
m
p
u
t
e
r
E
n
g
i
n
e
e
ri
n
g
(
I
J
EC
E)
,
v
o
l
.
1
4
,
n
o
.
4
,
p
.
4
7
3
2
,
A
u
g
.
2
0
2
4
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
c
e
.
v
1
4
i
4
.
p
p
4
7
3
2
-
4
7
4
4
.
[
2
]
T.
N
a
g
a
r
a
j
a
n
d
R
.
K
.
C
h
a
n
n
a
r
a
y
a
p
p
a
,
“
A
n
e
f
f
i
c
i
e
n
t
sec
u
r
i
t
y
f
r
a
mew
o
r
k
f
o
r
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
a
n
d
p
r
e
v
e
n
t
i
o
n
i
n
i
n
t
e
r
n
e
t
-
of
-
t
h
i
n
g
s
u
si
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
,
”
I
n
t
e
r
n
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
r
i
n
g
(
I
J
E
C
E)
,
v
o
l
.
1
4
,
n
o
.
2
,
p
.
2
3
1
3
,
A
p
r
.
2
0
2
4
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
c
e
.
v
1
4
i
2
.
p
p
2
3
1
3
-
2
3
2
1
.
[
3
]
L.
Zh
a
n
g
a
n
d
L.
W
a
n
g
,
“
A
h
y
b
r
i
d
e
n
c
r
y
p
t
i
o
n
a
p
p
r
o
a
c
h
f
o
r
e
f
f
i
c
i
e
n
t
a
n
d
sec
u
r
e
d
a
t
a
t
r
a
n
smis
si
o
n
i
n
I
o
T
d
e
v
i
c
e
s,
”
J
o
u
r
n
a
l
o
f
En
g
i
n
e
e
ri
n
g
a
n
d
A
p
p
l
i
e
d
S
c
i
e
n
c
e
,
v
o
l
.
7
1
,
n
o
.
1
,
p
.
1
3
8
,
D
e
c
.
2
0
2
4
,
d
o
i
:
1
0
.
1
1
8
6
/
s
4
4
1
4
7
-
0
2
4
-
0
0
4
5
9
-
x.
[
4
]
T.
R
a
j
mo
h
a
n
,
P
.
H
.
N
g
u
y
e
n
,
a
n
d
N
.
F
e
r
r
y
,
“
A
d
e
c
a
d
e
o
f
r
e
sea
r
c
h
o
n
p
a
t
t
e
r
n
s
a
n
d
a
r
c
h
i
t
e
c
t
u
r
e
s
f
o
r
I
o
T
se
c
u
r
i
t
y
,
”
C
y
b
e
rse
c
u
r
i
t
y
,
v
o
l
.
5
,
n
o
.
1
,
p
.
2
,
D
e
c
.
2
0
2
2
,
d
o
i
:
1
0
.
1
1
8
6
/
s
4
2
4
0
0
-
0
2
1
-
0
0
1
0
4
-
7.
[
5
]
K
.
K
a
n
d
a
s
a
my
,
S
.
S
r
i
n
i
v
a
s
,
K
.
A
c
h
u
t
h
a
n
,
a
n
d
V
.
P
.
R
a
n
g
a
n
,
“
I
o
T
c
y
b
e
r
r
i
sk
:
a
h
o
l
i
s
t
i
c
a
n
a
l
y
s
i
s
o
f
c
y
b
e
r
r
i
sk
a
ss
e
ssm
e
n
t
f
r
a
mew
o
r
k
s,
r
i
s
k
v
e
c
t
o
r
s,
a
n
d
r
i
s
k
r
a
n
k
i
n
g
p
r
o
c
e
ss,
”
EU
RA
S
I
P
J
o
u
rn
a
l
o
n
I
n
f
o
rm
a
t
i
o
n
S
e
c
u
ri
t
y
,
v
o
l
.
2
0
2
0
,
n
o
.
1
,
p
.
8
,
D
e
c
.
2
0
2
0
,
d
o
i
:
1
0
.
1
1
8
6
/
s
1
3
6
3
5
-
020
-
0
0
1
1
1
-
0.
[
6
]
T.
M
a
z
h
a
r
e
t
a
l
.
,
“
A
n
a
l
y
s
i
s
o
f
I
o
T
sec
u
r
i
t
y
c
h
a
l
l
e
n
g
e
s a
n
d
i
t
s so
l
u
t
i
o
n
s
u
si
n
g
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
,
”
Br
a
i
n
S
c
i
e
n
c
e
s
,
v
o
l
.
1
3
,
n
o
.
4
,
p
.
6
8
3
,
A
p
r
.
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
b
r
a
i
n
s
c
i
1
3
0
4
0
6
8
3
.
[
7
]
S
.
S
e
l
v
a
r
a
j
a
n
e
t
a
l
.
,
“
A
n
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
l
i
g
h
t
w
e
i
g
h
t
b
l
o
c
k
c
h
a
i
n
se
c
u
r
i
t
y
m
o
d
e
l
f
o
r
se
c
u
r
i
t
y
a
n
d
p
r
i
v
a
c
y
i
n
I
I
o
T
sy
s
t
e
ms,
”
J
o
u
rn
a
l
o
f
C
l
o
u
d
C
o
m
p
u
t
i
n
g
,
v
o
l
.
1
2
,
n
o
.
1
,
p
.
3
8
,
M
a
r
.
2
0
2
3
,
d
o
i
:
1
0
.
1
1
8
6
/
s1
3
6
7
7
-
0
2
3
-
0
0
4
1
2
-
y.
[
8
]
M
.
A
.
F
e
r
r
a
g
,
L
.
M
a
g
l
a
r
a
s
,
a
n
d
M
.
B
e
n
b
o
u
z
i
d
,
“
B
l
o
c
k
c
h
a
i
n
a
n
d
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
a
s
e
n
a
b
l
e
r
s
o
f
c
y
b
e
r
s
e
c
u
r
i
t
y
i
n
t
h
e
e
r
a
o
f
I
o
T
a
n
d
I
I
o
T
a
p
p
l
i
c
a
t
i
o
n
s,
”
J
o
u
r
n
a
l
o
f
S
e
n
s
o
r
a
n
d
Ac
t
u
a
t
o
r
N
e
t
w
o
r
k
s
,
v
o
l
.
1
2
,
n
o
.
3
,
p
.
4
0
,
M
a
y
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
j
s
a
n
1
2
0
3
0
0
4
0
.
[
9
]
K
.
C
h
a
k
r
a
b
o
r
t
y
,
D
.
K
a
p
i
l
a
,
S
.
K
u
m
a
r
,
B
h
u
p
a
t
i
,
N
.
S
h
a
i
k
,
a
n
d
A
.
S
i
n
g
h
,
“
I
n
t
e
l
l
i
g
e
n
t
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
b
a
s
e
d
i
n
t
e
r
n
e
t
o
f
t
h
i
n
g
s
(
I
o
T)
r
e
so
u
r
c
e
a
l
l
o
c
a
t
i
o
n
,
”
i
n
RA
i
S
E
-
2
0
2
3
,
D
e
c
.
2
0
2
3
,
p
.
7
3
.
d
o
i
:
1
0
.
3
3
9
0
/
e
n
g
p
r
o
c
2
0
2
3
0
5
9
0
7
3
.
[
1
0
]
T.
La
i
,
F
.
F
a
r
i
d
,
A
.
B
e
l
l
o
,
a
n
d
F
.
S
a
b
r
i
n
a
,
“
En
s
e
m
b
l
e
l
e
a
r
n
i
n
g
b
a
se
d
a
n
o
m
a
l
y
d
e
t
e
c
t
i
o
n
f
o
r
I
o
T
c
y
b
e
r
sec
u
r
i
t
y
v
i
a
B
a
y
e
si
a
n
h
y
p
e
r
p
a
r
a
me
t
e
r
s s
e
n
s
i
t
i
v
i
t
y
a
n
a
l
y
si
s,
”
C
y
b
e
rs
e
c
u
ri
t
y
,
v
o
l
.
7
,
n
o
.
1
,
p
.
4
4
,
J
u
n
.
2
0
2
4
,
d
o
i
:
1
0
.
1
1
8
6
/
s
4
2
4
0
0
-
0
2
4
-
0
0
2
3
8
-
4.
[
1
1
]
C
.
Li
u
e
t
a
l
.
,
“
D
i
sse
c
t
i
n
g
z
e
r
o
t
r
u
st
:
r
e
sea
r
c
h
l
a
n
d
s
c
a
p
e
a
n
d
i
t
s
i
mp
l
e
m
e
n
t
a
t
i
o
n
i
n
I
o
T,
”
C
y
b
e
rs
e
c
u
ri
t
y
,
v
o
l
.
7
,
n
o
.
1
,
p
.
2
0
,
M
a
y
2
0
2
4
,
d
o
i
:
1
0
.
1
1
8
6
/
s4
2
4
0
0
-
0
2
4
-
0
0
2
1
2
-
0.
[
1
2
]
S
.
M
i
s
h
r
a
,
A
.
A
l
b
a
r
a
k
a
t
i
,
a
n
d
S
.
K
.
S
h
a
r
ma,
“
C
y
b
e
r
t
h
r
e
a
t
i
n
t
e
l
l
i
g
e
n
c
e
f
o
r
I
o
T
u
si
n
g
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
,
”
Pr
o
c
e
sses
,
v
o
l
.
1
0
,
n
o
.
1
2
,
p
.
2
6
7
3
,
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
p
r
1
0
1
2
2
6
7
3
.
[
1
3
]
M
.
E.
E.
A
l
a
h
i
e
t
a
l
.
,
“
I
n
t
e
g
r
a
t
i
o
n
o
f
I
o
T
-
e
n
a
b
l
e
d
t
e
c
h
n
o
l
o
g
i
e
s
a
n
d
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
(
A
I
)
f
o
r
smar
t
c
i
t
y
sce
n
a
r
i
o
:
r
e
c
e
n
t
a
d
v
a
n
c
e
me
n
t
s
a
n
d
f
u
t
u
r
e
t
r
e
n
d
s,
”
S
e
n
so
rs
,
v
o
l
.
2
3
,
n
o
.
1
1
,
p
.
5
2
0
6
,
M
a
y
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
s
2
3
1
1
5
2
0
6
.
[
1
4
]
T.
N
a
r
a
s
i
m
h
a
m
u
r
t
h
y
a
n
d
G
.
H
o
sa
h
a
l
l
i
S
w
a
m
y
,
“
I
n
si
g
h
t
s
o
f
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
-
b
a
s
e
d
t
h
r
e
a
t
i
d
e
n
t
i
f
i
c
a
t
i
o
n
sc
h
e
mes
i
n
a
d
v
a
n
c
e
d
n
e
t
w
o
r
k
sy
s
t
e
m
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
E
l
e
c
t
ri
c
a
l
a
n
d
C
o
m
p
u
t
e
r
En
g
i
n
e
e
ri
n
g
(
I
J
EC
E)
,
v
o
l
.
1
4
,
n
o
.
4
,
p
.
4
6
6
4
,
A
u
g
.
2
0
2
4
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
c
e
.
v
1
4
i
4
.
p
p
4
6
6
4
-
4
6
7
4
.
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
.
6
,
Decem
b
e
r
20
25
:
6
0
4
3
-
6
0
5
2
6052
[
1
5
]
D
.
A
l
sal
ma
n
,
“
A
c
o
mp
a
r
a
t
i
v
e
st
u
d
y
o
f
a
n
o
m
a
l
y
d
e
t
e
c
t
i
o
n
t
e
c
h
n
i
q
u
e
s
f
o
r
I
o
T
sec
u
r
i
t
y
u
s
i
n
g
a
d
a
p
t
i
v
e
mac
h
i
n
e
l
e
a
r
n
i
n
g
f
o
r
I
o
T
t
h
r
e
a
t
s
,
”
I
E
EE
Ac
c
e
ss
,
v
o
l
.
1
2
,
p
p
.
1
4
7
1
9
–
1
4
7
3
0
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
4
.
3
3
5
9
0
3
3
.
[
1
6
]
M
.
A
.
E
l
sa
d
i
g
,
“
D
e
t
e
c
t
i
o
n
o
f
d
e
n
i
a
l
-
of
-
serv
i
c
e
a
t
t
a
c
k
i
n
w
i
r
e
l
e
ss
s
e
n
s
o
r
n
e
t
w
o
r
k
s:
a
l
i
g
h
t
w
e
i
g
h
t
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
,
”
I
EEE
A
c
c
e
ss
,
v
o
l
.
1
1
,
p
p
.
8
3
5
3
7
–
8
3
5
5
2
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
3
.
3
3
0
3
1
1
3
.
[
1
7
]
U
.
A
k
r
a
m
e
t
a
l
.
,
“
I
o
TTPS
:
e
n
s
e
m
b
l
e
R
K
S
V
M
m
o
d
e
l
-
b
a
s
e
d
i
n
t
e
r
n
e
t
o
f
t
h
i
n
g
s
t
h
r
e
a
t
p
r
o
t
e
c
t
i
o
n
sy
s
t
e
m
,
”
S
e
n
so
rs
,
v
o
l
.
2
3
,
n
o
.
1
4
,
p
.
6
3
7
9
,
Ju
l
.
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
s
2
3
1
4
6
3
7
9
.
[
1
8
]
S
.
B
e
n
A
t
i
t
a
l
l
a
h
,
M
.
D
r
i
ss,
a
n
d
I
.
A
l
m
o
ma
n
i
,
“
A
n
o
v
e
l
d
e
t
e
c
t
i
o
n
a
n
d
m
u
l
t
i
-
c
l
a
ss
i
f
i
c
a
t
i
o
n
a
p
p
r
o
a
c
h
f
o
r
I
o
T
-
mal
w
a
r
e
u
s
i
n
g
r
a
n
d
o
m
f
o
r
e
st
v
o
t
i
n
g
o
f
f
i
n
e
-
t
u
n
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
s,
”
S
e
n
so
r
s
,
v
o
l
.
2
2
,
n
o
.
1
1
,
p
.
4
3
0
2
,
J
u
n
.
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
s
2
2
1
1
4
3
0
2
.
[
1
9
]
A
.
H
a
mars
h
e
h
,
“
A
n
a
d
a
p
t
i
v
e
sec
u
r
i
t
y
f
r
a
mew
o
r
k
f
o
r
i
n
t
e
r
n
e
t
o
f
t
h
i
n
g
s
n
e
t
w
o
r
k
s
l
e
v
e
r
a
g
i
n
g
S
D
N
a
n
d
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
,
”
A
p
p
l
i
e
d
S
c
i
e
n
c
e
s
,
v
o
l
.
1
4
,
n
o
.
1
1
,
p
.
4
5
3
0
,
M
a
y
2
0
2
4
,
d
o
i
:
1
0
.
3
3
9
0
/
a
p
p
1
4
1
1
4
5
3
0
.
[
2
0
]
A
.
A
l
r
e
f
a
e
i
a
n
d
M
.
I
l
y
a
s
,
“
U
si
n
g
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
mu
l
t
i
c
l
a
ss
c
l
a
ssi
f
i
c
a
t
i
o
n
t
e
c
h
n
i
q
u
e
t
o
d
e
t
e
c
t
I
o
T
a
t
t
a
c
k
s
i
n
r
e
a
l
t
i
me
,
”
S
e
n
s
o
rs
,
v
o
l
.
2
4
,
n
o
.
1
4
,
p
.
4
5
1
6
,
J
u
l
.
2
0
2
4
,
d
o
i
:
1
0
.
3
3
9
0
/
s2
4
1
4
4
5
1
6
.
[
2
1
]
J.
L
i
,
M
.
S
.
O
t
h
ma
n
,
H
.
C
h
e
n
,
a
n
d
L.
M
.
Y
u
su
f
,
“
O
p
t
i
m
i
z
i
n
g
I
o
T
i
n
t
r
u
s
i
o
n
d
e
t
e
c
t
i
o
n
s
y
st
e
m:
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
v
e
r
s
u
s
f
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
i
n
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
,
”
J
o
u
r
n
a
l
o
f
Bi
g
D
a
t
a
,
v
o
l
.
1
1
,
n
o
.
1
,
p
.
3
6
,
F
e
b
.
2
0
2
4
,
d
o
i
:
1
0
.
1
1
8
6
/
s
4
0
5
3
7
-
024
-
0
0
8
9
2
-
y.
[
2
2
]
H
.
A
.
G
o
u
d
a
,
M
.
A
.
A
h
me
d
,
a
n
d
M
.
I
.
R
o
u
s
h
d
y
,
“
O
p
t
i
mi
z
i
n
g
a
n
o
m
a
l
y
-
b
a
se
d
a
t
t
a
c
k
d
e
t
e
c
t
i
o
n
u
si
n
g
c
l
a
ss
i
f
i
c
a
t
i
o
n
mac
h
i
n
e
l
e
a
r
n
i
n
g
,
”
N
e
u
r
a
l
C
o
m
p
u
t
i
n
g
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
3
6
,
n
o
.
6
,
p
p
.
3
2
3
9
–
3
2
5
7
,
F
e
b
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
0
7
/
s0
0
5
2
1
-
0
2
3
-
0
9
3
0
9
-
y.
[
2
3
]
Z.
E.
E
k
o
l
l
e
,
H
.
O
c
h
i
a
i
,
a
n
d
R
.
K
o
h
n
o
,
“
C
o
l
l
a
b
o
:
a
c
o
l
l
a
b
o
r
a
t
i
v
e
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
m
o
d
e
l
a
n
d
i
t
s
a
p
p
l
i
c
a
t
i
o
n
t
o
t
h
e
sec
u
r
i
t
y
o
f
h
e
t
e
r
o
g
e
n
e
o
u
s
me
d
i
c
a
l
d
a
t
a
i
n
a
n
I
o
T
n
e
t
w
o
r
k
,
”
I
EE
E
Ac
c
e
ss
,
v
o
l
.
1
1
,
p
p
.
1
4
2
6
6
3
–
1
4
2
6
7
5
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
3
.
3
3
4
1
8
3
7
.
[
2
4
]
M
.
A
l
q
a
h
t
a
n
i
,
H
.
M
a
t
h
k
o
u
r
,
a
n
d
M
.
M
.
B
e
n
I
smai
l
,
“
I
o
T
B
o
t
n
e
t
a
t
t
a
c
k
d
e
t
e
c
t
i
o
n
b
a
se
d
o
n
o
p
t
i
m
i
z
e
d
e
x
t
r
e
me
g
r
a
d
i
e
n
t
b
o
o
s
t
i
n
g
a
n
d
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
,
”
S
e
n
s
o
rs
,
v
o
l
.
2
0
,
n
o
.
2
1
,
p
.
6
3
3
6
,
N
o
v
.
2
0
2
0
,
d
o
i
:
1
0
.
3
3
9
0
/
s
2
0
2
1
6
3
3
6
.
[
2
5
]
H
.
E
l
-
S
o
f
a
n
y
,
S
.
A
.
El
-
S
e
o
u
d
,
O
.
H
.
K
a
r
a
m
,
a
n
d
B
.
B
o
u
a
l
l
e
g
u
e
,
“
U
si
n
g
mac
h
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
ms
t
o
e
n
h
a
n
c
e
I
o
T
sy
s
t
e
m
sec
u
r
i
t
y
,
”
S
c
i
e
n
t
i
f
i
c
Re
p
o
r
t
s
,
v
o
l
.
1
4
,
n
o
.
1
,
p
.
1
2
0
7
7
,
M
a
y
2
0
2
4
,
d
o
i
:
1
0
.
1
0
3
8
/
s
4
1
5
9
8
-
0
2
4
-
6
2
8
6
1
-
y.
[
2
6
]
M
.
S
a
l
b
e
t
a
l
.
,
“
En
h
a
n
c
i
n
g
i
n
t
e
r
n
e
t
o
f
t
h
i
n
g
s
n
e
t
w
o
r
k
s
e
c
u
r
i
t
y
u
s
i
n
g
h
y
b
r
i
d
C
N
N
a
n
d
X
G
B
o
o
s
t
m
o
d
e
l
t
u
n
e
d
v
i
a
mo
d
i
f
i
e
d
r
e
p
t
i
l
e
sea
r
c
h
a
l
g
o
r
i
t
h
m,
”
A
p
p
l
i
e
d
S
c
i
e
n
c
e
s
,
v
o
l
.
1
3
,
n
o
.
2
3
,
p
.
1
2
6
8
7
,
N
o
v
.
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
a
p
p
1
3
2
3
1
2
6
8
7
.
[
2
7
]
J.
A
l
F
a
y
sa
l
e
t
a
l
.
,
“
XGB
-
R
F
:
a
h
y
b
r
i
d
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
f
o
r
I
o
T
i
n
t
r
u
s
i
o
n
d
e
t
e
c
t
i
o
n
,
”
T
e
l
e
c
o
m
,
v
o
l
.
3
,
n
o
.
1
,
p
p
.
5
2
–
6
9
,
Jan
.
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
t
e
l
e
c
o
m
3
0
1
0
0
0
3
.
[
2
8
]
C
.
I
o
a
n
n
o
u
a
n
d
V
.
V
a
ss
i
l
i
o
u
,
“
N
e
t
w
o
r
k
a
t
t
a
c
k
c
l
a
ssi
f
i
c
a
t
i
o
n
i
n
I
o
T
u
si
n
g
su
p
p
o
r
t
v
e
c
t
o
r
ma
c
h
i
n
e
s,
”
J
o
u
r
n
a
l
o
f
S
e
n
s
o
r
a
n
d
Ac
t
u
a
t
o
r
N
e
t
w
o
rks
,
v
o
l
.
1
0
,
n
o
.
3
,
p
.
5
8
,
A
u
g
.
2
0
2
1
,
d
o
i
:
1
0
.
3
3
9
0
/
j
s
a
n
1
0
0
3
0
0
5
8
.
[
2
9
]
K
.
M
.
A
b
u
a
l
i
,
L.
N
i
ss
i
r
a
t
,
a
n
d
A
.
A
l
-
S
a
maw
i
,
“
A
d
v
a
n
c
i
n
g
n
e
t
w
o
r
k
sec
u
r
i
t
y
w
i
t
h
A
I
:
S
V
M
-
b
a
se
d
d
e
e
p
l
e
a
r
n
i
n
g
f
o
r
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
,
”
S
e
n
so
rs
,
v
o
l
.
2
3
,
n
o
.
2
1
,
p
.
8
9
5
9
,
N
o
v
.
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
s
2
3
2
1
8
9
5
9
.
[
3
0
]
S
.
D
a
l
a
l
e
t
a
l
.
,
“
N
e
x
t
-
g
e
n
e
r
a
t
i
o
n
c
y
b
e
r
a
t
t
a
c
k
p
r
e
d
i
c
t
i
o
n
f
o
r
I
o
T
sy
s
t
e
ms
:
l
e
v
e
r
a
g
i
n
g
mu
l
t
i
-
c
l
a
ss
S
V
M
a
n
d
o
p
t
i
m
i
z
e
d
C
H
A
I
D
d
e
c
i
si
o
n
t
r
e
e
,
”
J
o
u
r
n
a
l
o
f
C
l
o
u
d
C
o
m
p
u
t
i
n
g
,
v
o
l
.
1
2
,
n
o
.
1
,
p
.
1
3
7
,
S
e
p
.
2
0
2
3
,
d
o
i
:
1
0
.
1
1
8
6
/
s1
3
6
7
7
-
0
2
3
-
0
0
5
1
7
-
4.
[
3
1
]
J.
Jan
g
,
Y
.
A
n
,
D
.
K
i
m
,
a
n
d
D
.
C
h
o
i
,
“
F
e
a
t
u
r
e
i
mp
o
r
t
a
n
c
e
-
b
a
se
d
b
a
c
k
d
o
o
r
a
t
t
a
c
k
i
n
N
S
L
-
K
D
D
,
”
E
l
e
c
t
r
o
n
i
c
s
,
v
o
l
.
1
2
,
n
o
.
2
4
,
p
.
4
9
5
3
,
D
e
c
.
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
2
4
4
9
5
3
.
[
3
2
]
S
.
M
o
r
e
,
M
.
I
d
r
i
ssi
,
H
.
M
a
h
m
o
u
d
,
a
n
d
A
.
T.
A
sy
h
a
r
i
,
“
E
n
h
a
n
c
e
d
i
n
t
r
u
s
i
o
n
d
e
t
e
c
t
i
o
n
s
y
st
e
ms
p
e
r
f
o
r
ma
n
c
e
w
i
t
h
U
N
S
W
-
N
B
1
5
d
a
t
a
a
n
a
l
y
si
s
,
”
Al
g
o
ri
t
h
m
s
,
v
o
l
.
1
7
,
n
o
.
2
,
p
.
6
4
,
F
e
b
.
2
0
2
4
,
d
o
i
:
1
0
.
3
3
9
0
/
a
1
7
0
2
0
0
6
4
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Th
a
n
u
j
a
Na
r
a
sim
h
a
m
u
r
th
y
is
As
sista
n
t
P
ro
fe
ss
o
r
at
De
p
a
rt
m
e
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
,
Ba
n
g
a
lo
re
I
n
stit
u
te
o
f
Tec
h
n
o
lo
g
y
,
Be
n
g
a
lu
r
u
.
C
o
m
p
lete
d
G
ra
d
u
a
ti
o
n
a
n
d
P
o
st
-
G
ra
d
u
a
ti
o
n
fr
o
m
Visv
e
sv
a
ra
y
a
Tec
h
n
o
l
o
g
ica
l
Un
iv
e
rsity
in
In
fo
rm
a
ti
o
n
/Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
i
n
e
e
rin
g
S
trea
m
.
P
u
b
li
sh
e
d
p
a
p
e
rs
in
In
tern
a
ti
o
n
a
l
Jo
u
rn
a
ls
a
n
d
Co
n
fe
re
n
c
e
s
in
t
h
e
d
o
m
a
in
m
a
c
h
in
e
lea
rn
i
n
g
a
n
d
i
n
tern
e
t
o
f
th
i
n
g
s
.
Re
se
a
rc
h
fo
c
u
s is
o
n
n
e
tw
o
rk
i
n
g
sy
ste
m
s,
i
n
tern
e
t
o
f
t
h
in
g
s
,
a
n
d
m
a
c
h
in
e
lea
rn
in
g
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
n
th
a
n
u
ja@
b
it
-
b
a
n
g
a
lo
re
.
e
d
u
.
i
n
.
G
u
n
a
v
a
th
i
H
o
sa
h
a
ll
i
S
wa
m
y
is
As
so
c
iate
De
a
n
,
S
k
il
l
De
v
e
lo
p
m
e
n
t
a
n
d
As
sista
n
t
P
ro
fe
ss
o
r
at
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
i
n
e
e
rin
g
,
Ba
n
g
a
lo
re
I
n
stit
u
te
o
f
Tec
h
n
o
l
o
g
y
,
Be
n
g
a
lu
r
u
.
S
h
e
h
a
s
p
u
b
li
sh
e
d
m
a
n
y
re
se
a
rc
h
p
a
p
e
rs
in
in
tern
a
ti
o
n
a
l
j
o
u
r
n
a
ls
a
n
d
i
n
tern
a
ti
o
n
a
l
c
o
n
fe
re
n
c
e
s.
Ac
ti
v
e
ly
i
n
v
o
lv
e
d
i
n
o
rg
a
n
izin
g
e
v
e
n
ts/co
n
fe
re
n
c
e
,
a
lso
o
rg
a
n
ize
d
m
a
n
y
wo
r
k
sh
o
p
s/F
D
P
s.
M
a
j
o
r
a
r
e
a
s
o
f
r
e
s
e
a
r
c
h
i
n
t
e
r
e
s
t
a
r
e
i
m
a
g
e
p
r
o
c
e
s
s
i
n
g
,
p
a
t
t
e
r
n
r
e
c
o
g
n
i
t
i
o
n
,
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
,
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
,
d
a
t
a
s
c
i
e
n
c
e
,
a
n
d
I
o
T
.
S
h
e
c
a
n
b
e
c
o
n
t
a
c
t
e
d
b
y
g
u
n
a
v
a
t
h
i
h
s
@
b
i
t
-
b
a
n
g
a
l
o
r
e
.
e
d
u
.
i
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.