I
AE
S In
t
er
na
t
io
na
l J
o
urna
l o
f
Art
if
icia
l In
t
ellig
ence
(
I
J
-
AI
)
Vo
l.
14
,
No
.
2
,
A
p
r
il
2
0
2
5
,
p
p
.
1
2
3
2
~
1
2
4
1
I
SS
N:
2
2
5
2
-
8
9
3
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijai.v
14
.i
2
.
p
p
1
2
3
2
-
1
2
4
1
1232
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
Bo
o
sting
i
ndu
stria
l int
er
net
o
f
thi
ng
s
intrusio
n d
etec
ti
o
n:
lev
erag
ing
ma
chine learning
and
f
ea
ture
sele
ction
te
chniques
L
a
hcen
I
do
ug
lid
,
Sa
id
T
k
a
t
e
k
,
K
ha
lid
E
lf
a
y
q
F
a
c
u
l
t
y
o
f
S
c
i
e
n
c
e
s,
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
s R
e
se
a
r
c
h
La
b
o
r
a
t
o
r
y
,
I
b
n
T
o
f
a
i
l
U
n
i
v
e
r
s
i
t
y
,
K
e
n
i
t
r
a
,
M
o
r
o
c
c
o
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Au
g
1
4
,
2
0
2
4
R
ev
is
ed
Oct
3
1
,
2
0
2
4
Acc
ep
ted
No
v
1
4
,
2
0
2
4
Th
e
ra
p
id
i
n
teg
ra
ti
o
n
o
f
i
n
d
u
stria
l
in
tern
e
t
o
f
t
h
in
g
s
(IIo
T)
tec
h
n
o
l
o
g
ies
in
t
o
In
d
u
stry
4
.
0
h
a
s
re
v
o
lu
t
io
n
ize
d
i
n
d
u
strial
e
fficie
n
c
y
a
n
d
a
u
to
m
a
ti
o
n
,
b
u
t
it
h
a
s
a
lso
e
x
p
o
se
d
c
rit
ica
l
v
u
l
n
e
ra
b
il
it
i
e
s
to
c
y
b
e
r
th
re
a
ts.
T
h
is
p
a
p
e
r
d
e
l
v
e
s
in
t
o
a
c
o
m
p
re
h
e
n
siv
e
e
v
a
lu
a
ti
o
n
o
f
m
a
c
h
in
e
lea
rn
in
g
(
M
L)
c
las
sifiers
f
o
r
d
e
tec
ti
n
g
a
n
o
m
a
li
e
s
in
IIo
T
e
n
v
ir
o
n
m
e
n
ts.
By
stra
teg
ica
ll
y
a
p
p
ly
i
n
g
fe
a
tu
r
e
se
lec
ti
o
n
tec
h
n
iq
u
e
s,
we
d
e
m
o
n
stra
te
sig
n
ifi
c
a
n
t
e
n
h
a
n
c
e
m
e
n
ts
in
b
o
t
h
th
e
a
c
c
u
ra
c
y
a
n
d
e
fficie
n
c
y
o
f
t
h
e
se
c
las
sifiers
.
Ou
r
fin
d
in
g
s
re
v
e
a
l
th
a
t
fe
a
tu
re
se
lec
ti
o
n
n
o
t
o
n
l
y
b
o
o
sts
d
e
tec
ti
o
n
ra
tes
b
u
t
a
lso
m
in
imiz
e
s
c
o
m
p
u
tatio
n
a
l
d
e
m
a
n
d
s,
m
a
k
in
g
it
a
c
o
r
n
e
rsto
n
e
fo
r
d
e
v
e
lo
p
i
n
g
re
sili
e
n
t
in
tr
u
sio
n
d
e
tec
ti
o
n
sy
ste
m
s
(IDS)
tailo
re
d
f
o
r
In
d
u
stry
4
.
0
.
T
h
e
in
sig
h
ts
g
a
rn
e
re
d
fro
m
t
h
is
stu
d
y
p
a
v
e
th
e
wa
y
fo
r
d
e
p
l
o
y
i
n
g
m
o
re
ro
b
u
s
t
se
c
u
rit
y
fra
m
e
wo
rk
s,
sa
fe
g
u
a
rd
in
g
th
e
in
teg
rit
y
a
n
d
re
li
a
b
i
li
ty
o
f
IIo
T
in
fra
stru
c
tu
re
s in
m
o
d
e
rn
in
d
u
strial
se
tt
in
g
s.
K
ey
w
o
r
d
s
:
An
o
m
aly
d
etec
tio
n
Featu
r
e
s
elec
tio
n
I
n
d
u
s
t
r
i
a
l
i
n
t
e
r
n
e
t
o
f
t
h
i
n
g
s
s
e
c
u
r
i
t
y
I
n
d
u
s
tr
y
4
.
0
I
n
tr
u
s
io
n
d
etec
tio
n
Ma
ch
in
e
lear
n
in
g
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
:
L
ah
ce
n
I
d
o
u
g
lid
Facu
lty
o
f
Scien
ce
s
,
C
o
m
p
u
te
r
Scien
ce
s
R
esear
ch
L
ab
o
r
ato
r
y
,
I
b
n
T
o
f
ail
Un
i
v
er
s
ity
Ken
itra
,
Mo
r
o
cc
o
E
m
ail:
lah
ce
n
.
id
o
u
g
lid
@
u
it.a
c.
m
a
1.
I
NT
RO
D
UCT
I
O
N
I
n
d
u
s
tr
y
4
.
0
,
k
n
o
wn
as
th
e
f
o
u
r
th
i
n
d
u
s
tr
ial
r
e
v
o
lu
tio
n
,
r
e
p
r
esen
ts
a
m
ajo
r
s
h
if
t
in
m
an
u
f
ac
tu
r
in
g
p
r
o
ce
s
s
es
b
y
in
teg
r
atin
g
c
y
b
e
r
-
p
h
y
s
ical
s
y
s
tem
s
,
au
to
m
atio
n
,
an
d
s
m
ar
t
tech
n
o
lo
g
ies.
A
k
ey
co
m
p
o
n
e
n
t
o
f
I
n
d
u
s
tr
y
4
.
0
is
th
e
i
n
d
u
s
tr
ial
i
n
ter
n
et
o
f
th
in
g
s
(
I
I
o
T
)
,
wh
ic
h
co
n
n
ec
ts
m
ac
h
in
es,
d
ev
ices,
an
d
s
y
s
tem
s
with
in
in
d
u
s
tr
ial
en
v
ir
o
n
m
e
n
ts
th
r
o
u
g
h
ad
v
an
ce
d
co
m
m
u
n
icatio
n
n
etwo
r
k
s
.
T
h
is
co
n
n
ec
tiv
ity
e
n
ab
les
r
ea
l
-
tim
e
d
ata
co
llectio
n
,
an
aly
s
is
,
an
d
d
ec
is
io
n
-
m
ak
in
g
,
s
ig
n
if
ica
n
tly
im
p
r
o
v
in
g
ef
f
icien
c
y
,
p
r
o
d
u
ctiv
i
ty
,
an
d
f
le
x
ib
ilit
y
i
n
m
an
u
f
ac
tu
r
in
g
p
r
o
ce
s
s
es
[
1
]
.
T
h
e
I
I
o
T
p
lay
s
a
p
iv
o
tal
r
o
le
i
n
en
h
an
cin
g
o
p
er
atio
n
al
ca
p
a
b
il
ities
b
y
f
ac
ilit
atin
g
s
ea
m
less
in
f
o
r
m
atio
n
ex
ch
an
g
e
b
etwe
en
m
ac
h
i
n
es a
n
d
s
y
s
tem
s
.
I
ts
ad
o
p
tio
n
allo
w
s
in
d
u
s
tr
ies to
o
p
tim
ize
m
an
u
f
ac
tu
r
in
g
p
r
o
ce
s
s
es,
p
r
ed
ict
m
ain
ten
a
n
ce
n
ee
d
s
,
a
n
d
d
ev
elo
p
s
m
ar
t
f
ac
to
r
ies
th
at
o
p
er
ate
au
to
n
o
m
o
u
s
ly
an
d
a
d
a
p
tiv
ely
.
T
h
is
h
as
led
to
wid
esp
r
ea
d
ad
o
p
tio
n
o
f
I
I
o
T
tech
n
o
lo
g
ies g
lo
b
ally
,
g
iv
in
g
in
d
u
s
tr
ies a
co
m
p
etitiv
e
ed
g
e
[
2
]
.
Ho
wev
er
,
th
e
r
ap
id
im
p
lem
en
tatio
n
o
f
I
I
o
T
in
tr
o
d
u
ce
s
cr
itical
cy
b
er
s
ec
u
r
ity
ch
allen
g
es.
T
h
e
in
ter
co
n
n
ec
ted
n
at
u
r
e
o
f
th
ese
s
y
s
tem
s
m
ak
es
th
em
v
u
l
n
er
a
b
le
to
c
y
b
er
th
r
ea
ts
,
wh
ich
ca
n
lead
to
o
p
er
atio
n
al
d
is
r
u
p
tio
n
s
,
f
in
a
n
cial
lo
s
s
es,
a
n
d
co
m
p
r
o
m
is
ed
s
en
s
itiv
e
d
at
a.
E
n
s
u
r
in
g
t
h
e
s
ec
u
r
ity
o
f
I
I
o
T
en
v
ir
o
n
m
en
ts
h
as
th
u
s
b
ec
o
m
e
a
to
p
p
r
io
r
ity
f
o
r
b
o
th
in
d
u
s
tr
y
lea
d
er
s
an
d
r
esea
r
ch
er
s
as
th
e
r
elian
ce
o
n
I
n
d
u
s
t
r
y
4
.
0
tech
n
o
lo
g
ies
g
r
o
ws
[
3
]
.
As
th
e
I
I
o
T
b
ec
o
m
e
s
in
cr
ea
s
in
g
ly
in
teg
r
ated
in
to
I
n
d
u
s
tr
y
4
.
0
,
its
cy
b
er
s
ec
u
r
ity
b
ec
o
m
es
ev
en
m
o
r
e
ess
en
tial.
I
I
o
T
s
y
s
tem
s
,
wh
ic
h
co
n
n
ec
t
a
v
ast
ar
r
ay
o
f
d
e
v
ices,
s
en
s
o
r
s
,
an
d
m
ac
h
in
er
y
,
ar
e
cr
u
cial
f
o
r
th
e
ef
f
icien
cy
o
f
m
o
d
er
n
in
d
u
s
tr
ial
o
p
er
atio
n
s
.
H
o
wev
er
,
th
eir
in
ter
co
n
n
ec
ted
n
ess
in
tr
o
d
u
ce
s
s
ig
n
if
ican
t
v
u
ln
er
ab
ilit
ies
th
at
ca
n
b
e
ex
p
lo
ited
b
y
cy
b
er
cr
im
i
n
als.
T
h
e
d
is
r
u
p
tio
n
o
f
I
I
o
T
n
etwo
r
k
s
th
r
o
u
g
h
cy
b
er
attac
k
s
ca
n
r
esu
lt in
p
r
o
d
u
ctio
n
d
o
wn
t
im
es,
f
in
an
cial
lo
s
s
es,
an
d
th
r
e
ats to
h
u
m
an
s
af
ety
[
4
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
B
o
o
s
tin
g
I
I
o
T in
tr
u
s
io
n
d
etec
t
io
n
:
leve
r
a
g
in
g
ma
ch
in
e
lea
r
n
in
g
a
n
d
fea
tu
r
e
s
elec
tio
n
…
(
L
a
h
ce
n
I
d
o
u
g
lid
)
1233
T
h
e
ch
ar
ac
ter
is
tics
o
f
I
I
o
T
en
v
ir
o
n
m
en
ts
,
s
u
ch
as
th
eir
s
ca
le,
h
eter
o
g
en
eity
,
an
d
r
ea
l
-
tim
e
o
p
er
atio
n
s
,
m
ak
e
th
em
s
u
s
ce
p
tib
le
to
a
r
a
n
g
e
o
f
cy
b
e
r
th
r
ea
ts
,
in
clu
d
in
g
d
is
tr
ib
u
ted
d
en
ial
o
f
s
er
v
ice
(
DDo
S)
attac
k
s
,
d
ata
b
r
ea
ch
es,
an
d
m
a
n
ip
u
latio
n
o
f
cr
itical
p
r
o
ce
s
s
es.
Ad
d
itio
n
ally
,
th
e
u
s
e
o
f
leg
ac
y
s
y
s
tem
s
w
ith
lim
ited
s
ec
u
r
ity
f
ea
tu
r
es
f
u
r
t
h
er
in
c
r
ea
s
es
th
e
r
is
k
s
,
h
ig
h
lig
h
tin
g
th
e
n
ee
d
f
o
r
r
o
b
u
s
t
s
ec
u
r
ity
m
ea
s
u
r
es
tailo
r
ed
s
p
ec
if
ically
f
o
r
I
I
o
T
en
v
ir
o
n
m
en
ts
[
5
]
.
Secu
r
in
g
th
ese
en
v
ir
o
n
m
en
ts
r
eq
u
ir
e
s
a
m
u
lti
-
lay
er
ed
ap
p
r
o
ac
h
.
T
h
is
in
clu
d
es
n
o
t
o
n
ly
tr
ad
itio
n
al
I
T
s
ec
u
r
ity
p
r
ac
tic
es
b
u
t
also
s
p
ec
ialized
m
ea
s
u
r
es
d
esig
n
ed
f
o
r
in
d
u
s
tr
ial
c
o
n
tr
o
l
s
y
s
tem
s
(
I
C
S),
s
u
ch
as
in
tr
u
s
io
n
d
etec
tio
n
s
y
s
tem
s
(
I
DS)
,
en
cr
y
p
tio
n
tec
h
n
iq
u
es,
an
d
r
ea
l
-
tim
e
m
o
n
it
o
r
in
g
to
d
etec
t
an
d
m
itig
ate
an
o
m
alies
s
wif
tly
.
Giv
en
th
e
p
o
ten
tial
co
n
s
eq
u
en
c
es
o
f
cy
b
e
r
attac
k
s
,
cy
b
e
r
s
ec
u
r
ity
h
as
b
ec
o
m
e
a
cr
itical
co
n
ce
r
n
f
o
r
in
d
u
s
tr
ies ad
o
p
tin
g
I
n
d
u
s
tr
y
4
.
0
tech
n
o
lo
g
ies
[
6
]
.
T
h
e
p
r
im
a
r
y
o
b
jectiv
e
o
f
th
is
s
tu
d
y
is
to
en
h
an
ce
t
h
e
s
ec
u
r
it
y
o
f
I
I
o
T
en
v
ir
o
n
m
en
ts
b
y
ev
a
lu
atin
g
th
e
ef
f
ec
tiv
en
ess
o
f
m
ac
h
in
e
lear
n
in
g
(
ML
)
class
if
ier
s
f
o
r
an
o
m
a
ly
d
etec
tio
n
.
As
I
I
o
T
tec
h
n
o
l
o
g
ies
ar
e
in
cr
ea
s
in
g
ly
in
teg
r
ated
in
t
o
in
d
u
s
tr
ial
p
r
o
c
ess
es,
th
e
n
ee
d
f
o
r
r
o
b
u
s
t
a
n
d
ef
f
icien
t
I
DS
is
p
ar
a
m
o
u
n
t.
T
h
is
r
esear
ch
f
o
cu
s
es
o
n
lev
er
ag
in
g
f
ea
tu
r
e
s
elec
tio
n
tech
n
iq
u
es
to
im
p
r
o
v
e
th
e
ac
c
u
r
ac
y
an
d
c
o
m
p
u
tatio
n
al
ef
f
ici
en
cy
o
f
class
if
ier
s
,
p
r
o
v
id
i
n
g
a
s
ca
lab
le
s
o
lu
tio
n
f
o
r
r
ea
l
-
tim
e
th
r
ea
t
d
etec
tio
n
i
n
I
I
o
T
en
v
i
r
o
n
m
e
n
ts
[
7
]
.
B
y
ad
d
r
ess
in
g
ch
allen
g
es
s
u
ch
as
h
ig
h
-
d
im
en
s
io
n
al
d
ata
an
d
c
o
m
p
lex
in
d
u
s
tr
ial
n
etwo
r
k
s
,
th
is
s
tu
d
y
c
o
n
tr
ib
u
tes
to
t
h
e
d
ev
el
o
p
m
en
t
o
f
r
esil
ien
t secu
r
ity
f
r
am
ewo
r
k
s
f
o
r
I
I
o
T
s
y
s
tem
s
.
T
h
e
f
in
d
in
g
s
ar
e
ex
p
ec
ted
to
ass
is
t in
d
esig
n
in
g
ad
v
a
n
ce
d
I
DS
th
at
m
itig
ate
cy
b
er
attac
k
s
,
en
s
u
r
in
g
th
e
co
n
tin
u
ity
an
d
s
af
ety
o
f
cr
itical
in
d
u
s
tr
ial
o
p
er
atio
n
s
[
8
]
.
ML
h
a
s
p
r
o
v
e
n
t
o
b
e
a
p
o
we
r
f
u
l
t
o
o
l
i
n
t
h
e
d
e
v
e
l
o
p
m
e
n
t
o
f
I
D
S
,
es
p
e
c
ia
l
l
y
w
i
t
h
i
n
t
h
e
c
o
n
te
x
t
o
f
I
I
o
T
a
n
d
I
n
d
u
s
t
r
y
4
.
0
.
U
n
l
i
k
e
t
r
a
d
i
ti
o
n
a
l
r
u
l
e
-
b
as
e
d
I
DS
,
w
h
i
c
h
d
e
p
e
n
d
o
n
p
r
e
d
e
f
i
n
e
d
s
i
g
n
a
t
u
r
es
o
f
k
n
o
w
n
t
h
r
e
a
ts
,
ML
-
b
a
s
e
d
I
D
S
c
a
n
i
d
e
n
t
i
f
y
b
o
t
h
k
n
o
w
n
a
n
d
u
n
k
n
o
w
n
t
h
r
ea
t
s
b
y
l
e
a
r
n
i
n
g
p
at
t
e
r
n
s
f
r
o
m
h
i
s
t
o
r
i
c
al
d
at
a
.
T
h
is
c
a
p
a
b
i
l
i
t
y
is
c
r
u
c
i
al
f
o
r
d
e
t
e
c
t
i
n
g
n
o
v
e
l
a
t
t
a
c
k
s
a
n
d
a
n
o
m
a
l
i
es
th
a
t
m
a
y
c
o
m
p
r
o
m
i
s
e
t
h
e
s
ec
u
r
it
y
o
f
I
I
o
T
n
e
t
w
o
r
k
s
[
1
]
.
S
e
v
e
r
a
l
ML
a
l
g
o
r
it
h
m
s
,
s
u
c
h
a
s
l
o
g
is
ti
c
r
e
g
r
es
s
i
o
n
,
d
e
ci
s
i
o
n
t
r
e
e
s
,
a
n
d
r
a
n
d
o
m
f
o
r
e
s
ts
,
h
a
v
e
b
e
e
n
w
i
d
e
l
y
a
d
o
p
t
e
d
i
n
I
D
S
d
u
e
t
o
t
h
e
i
r
a
b
i
l
i
t
y
t
o
h
a
n
d
l
e
c
o
m
p
l
e
x
d
a
t
a
s
e
t
s
a
n
d
a
c
h
i
e
v
e
h
i
g
h
d
e
t
e
c
t
i
o
n
a
c
c
u
r
a
c
y
.
F
e
a
t
u
r
e
s
e
l
e
ct
i
o
n
t
ec
h
n
i
q
u
e
s
f
u
r
t
h
e
r
e
n
h
a
n
c
e
t
h
e
s
e
a
l
g
o
r
i
t
h
m
s
b
y
r
e
d
u
c
i
n
g
d
a
t
a
d
i
m
e
n
s
i
o
n
a
l
ity
,
t
h
u
s
i
m
p
r
o
v
i
n
g
c
o
m
p
u
t
a
t
i
o
n
a
l
e
f
f
i
c
i
e
n
c
y
wi
t
h
o
u
t
s
a
c
r
i
f
i
ci
n
g
p
e
r
f
o
r
m
a
n
c
e
.
I
n
r
e
s
o
u
r
c
e
-
c
o
n
s
t
r
ai
n
e
d
e
n
v
i
r
o
n
m
e
n
t
s
t
y
p
i
c
a
l
o
f
I
I
o
T
,
ML
o
f
f
e
r
s
s
c
a
l
a
b
l
e
a
n
d
a
d
a
p
ti
v
e
s
o
l
u
ti
o
n
f
o
r
s
a
f
e
g
u
a
r
d
i
n
g
i
n
d
u
s
t
r
i
a
l
s
y
s
t
e
m
s
f
r
o
m
w
i
d
e
r
a
n
g
e
o
f
c
y
b
e
r
t
h
r
e
a
ts
[
2
]
.
2.
RE
L
AT
E
D
WO
RK
T
h
i
s
s
e
c
t
i
o
n
a
n
a
l
y
z
e
s
a
n
d
c
o
n
s
o
l
i
d
a
t
e
s
k
e
y
f
i
n
d
i
n
g
s
a
n
d
m
e
t
h
o
d
o
l
o
g
i
e
s
f
r
o
m
e
x
i
s
t
i
n
g
r
e
s
e
a
r
c
h
o
n
I
D
S
i
n
t
h
e
I
I
o
T
c
o
n
t
e
x
t
.
T
h
e
r
i
s
e
o
f
I
I
o
T
t
e
c
h
n
o
l
o
g
i
e
s
i
n
I
n
d
u
s
t
r
y
4
.
0
h
a
s
s
p
u
r
r
e
d
r
e
s
e
a
r
c
h
i
n
t
o
s
e
c
u
r
i
n
g
t
h
e
s
e
e
n
v
i
r
o
n
m
e
n
t
s
,
w
i
t
h
a
f
o
c
u
s
o
n
a
p
p
l
y
i
n
g
ML
a
n
d
d
e
e
p
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s
t
o
d
e
t
e
c
t
a
n
o
m
a
l
i
e
s
a
n
d
c
y
b
e
r
t
h
r
e
a
t
s
i
n
I
I
o
T
n
e
t
w
o
r
k
s
.
‒
Ma
ch
in
e
lear
n
in
g
-
b
ased
I
DS:
T
h
e
u
s
e
o
f
ML
alg
o
r
ith
m
s
in
I
DS
h
as
g
ain
ed
s
ig
n
if
ica
n
t
atten
tio
n
d
u
e
to
t
h
eir
ab
ilit
y
to
d
etec
t
b
o
th
k
n
o
w
n
a
n
d
u
n
k
n
o
w
n
th
r
ea
ts
.
Mlik
i
et
a
l
.
[
9
]
c
o
n
d
u
cted
a
co
m
p
r
eh
e
n
s
iv
e
s
u
r
v
ey
o
f
ML
tech
n
iq
u
es
ap
p
lied
to
I
I
o
T
s
ec
u
r
ity
,
h
ig
h
lig
h
tin
g
th
e
s
tr
en
g
th
s
an
d
lim
itatio
n
s
o
f
v
ar
io
u
s
alg
o
r
ith
m
s
s
u
ch
as
s
u
p
p
o
r
t
v
ec
t
o
r
m
ac
h
in
es,
d
ec
is
io
n
tr
ee
s
,
an
d
n
eu
r
al
n
etwo
r
k
s
.
T
h
e
s
tu
d
y
u
n
d
e
r
s
co
r
e
s
th
e
im
p
o
r
tan
ce
o
f
f
ea
tu
r
e
s
elec
tio
n
in
im
p
r
o
v
i
n
g
th
e
ef
f
icien
cy
o
f
I
DS
.
‒
Dee
p
lear
n
in
g
a
p
p
r
o
ac
h
es
:
S
o
lim
an
et
a
l
.
[
1
0
]
p
r
o
p
o
s
e
a
d
ee
p
lear
n
in
g
-
b
ased
I
DS
f
o
r
s
ec
u
r
in
g
I
I
o
T
n
etwo
r
k
s
,
ad
d
r
ess
in
g
ch
allen
g
es
lik
e
h
ig
h
f
ea
tu
r
e
d
im
en
s
io
n
s
an
d
im
b
alan
ce
d
d
atasets
.
T
h
eir
m
o
d
el
u
tili
ze
s
s
in
g
u
lar
v
al
u
e
d
ec
o
m
p
o
s
itio
n
(
SVD)
an
d
s
y
n
t
h
etic
m
in
o
r
ity
o
v
er
-
s
am
p
lin
g
tech
n
iq
u
e
(
SMOT
E
)
to
en
h
a
n
ce
d
etec
tio
n
ac
c
u
r
ac
y
an
d
r
e
d
u
c
e
er
r
o
r
r
ates,
ac
h
iev
i
n
g
u
p
to
9
9
.
9
9
%
ac
cu
r
ac
y
i
n
b
i
n
ar
y
cl
ass
if
icatio
n
an
d
9
9
.
9
8
% in
m
u
lti
-
class
class
if
ic
atio
n
o
n
t
h
e
T
o
N_
I
o
T
d
ataset
.
‒
Hy
b
r
id
I
DS
m
o
d
els:
Hy
b
r
id
m
o
d
els
co
m
b
i
n
in
g
m
u
ltip
le
ML
tech
n
iq
u
es
h
av
e
b
ee
n
d
e
v
el
o
p
ed
to
en
h
a
n
ce
th
e
r
o
b
u
s
tn
ess
o
f
I
DS
in
I
I
o
T
en
v
ir
o
n
m
en
ts
.
Gu
ez
za
z
et
a
l
.
[
1
1
]
p
r
o
p
o
s
e
a
lig
h
tweig
h
t
h
y
b
r
id
I
DS
f
r
am
ewo
r
k
th
at
in
teg
r
ates
K
-
n
ea
r
est
n
eig
h
b
o
r
(
K
-
NN)
an
d
p
r
in
cip
al
co
m
p
o
n
en
t
an
aly
s
is
(
PC
A)
f
o
r
ed
g
e
-
b
ased
I
I
o
T
s
ec
u
r
ity
.
T
h
is
ap
p
r
o
ac
h
lev
er
a
g
es
th
e
s
tr
en
g
th
s
o
f
b
o
th
K
-
NN
f
o
r
h
ig
h
d
etec
tio
n
ac
cu
r
ac
y
an
d
PC
A
f
o
r
ef
f
ec
tiv
e
f
ea
tu
r
e
en
g
i
n
ee
r
in
g
,
ac
h
iev
in
g
n
o
ta
b
le
r
esu
lts
with
9
9
.
1
0
%
ac
cu
r
ac
y
a
n
d
9
8
.
4
%
d
etec
tio
n
r
ate
o
n
t
h
e
n
etwo
r
k
s
ec
u
r
ity
lab
-
k
n
o
wled
g
e
d
is
co
v
er
y
an
d
d
ata
m
in
in
g
(
NSL
-
KDD)
d
ata
s
et,
an
d
9
8
.
2
%
ac
cu
r
ac
y
a
n
d
9
7
.
6
%
d
etec
tio
n
r
ate
o
n
th
e
B
o
t
-
I
o
T
d
ataset
.
An
o
th
er
p
ap
e
r
,
B
ak
r
o
et
a
l
.
[
1
2
]
p
r
o
p
o
s
es
a
clo
u
d
-
b
ased
I
DS
th
at
in
teg
r
ate
s
b
io
-
in
s
p
ir
ed
f
ea
tu
r
e
s
elec
tio
n
alg
o
r
ith
m
s
,
n
am
ely
g
r
ass
h
o
p
p
er
o
p
tim
izatio
n
alg
o
r
ith
m
(
GOA)
a
n
d
g
en
etic
alg
o
r
ith
m
(
GA)
,
alo
n
g
with
a
r
an
d
o
m
f
o
r
est
class
if
ier
.
T
h
is
h
y
b
r
id
f
r
am
ewo
r
k
ad
d
r
ess
es
ch
allen
g
es
s
u
ch
as
im
b
alan
ce
d
d
ata
a
n
d
h
ig
h
d
i
m
en
s
io
n
ality
b
y
e
m
p
lo
y
i
n
g
a
d
ap
tiv
e
s
y
n
th
etic
s
am
p
lin
g
(
ADASYN)
an
d
r
an
d
o
m
u
n
d
er
-
s
am
p
lin
g
(
R
US)
.
E
v
alu
ated
o
n
d
atasets
lik
e
UN
SW
-
N
B
1
5
,
C
I
C
-
DDo
S2
0
1
9
,
an
d
C
I
C
B
ell
D
NS
E
XF
2
0
2
1
,
th
e
m
o
d
el
a
ch
iev
ed
ac
cu
r
ac
ies
o
f
9
8
%,
9
9
%,
an
d
9
2
%,
r
esp
e
ctiv
ely
,
d
em
o
n
s
tr
atin
g
s
u
p
er
io
r
m
u
lti
-
class
class
if
icat
io
n
p
er
f
o
r
m
an
ce
a
n
d
ef
f
icien
cy
co
m
p
ar
e
d
to
o
th
er
class
if
ier
s
,
in
clu
d
in
g
s
u
p
p
o
r
t v
ec
to
r
m
ac
h
i
n
e
(
SVM)
,
d
ee
p
n
eu
r
al
n
etwo
r
k
(
DNN)
,
an
d
XGBo
o
s
t
.
‒
A
n
o
m
a
l
y
d
e
t
e
c
t
i
o
n
:
A
w
o
t
u
n
d
e
e
t
a
l
.
[
1
3
]
i
n
t
r
o
d
u
c
e
d
e
n
s
e
m
b
l
e
t
r
e
e
-
b
a
s
e
d
m
o
d
e
l
f
o
r
i
n
t
r
u
s
io
n
d
e
t
e
c
t
i
o
n
i
n
I
I
o
T
n
e
t
w
o
r
k
s
,
e
m
p
l
o
y
i
n
g
c
l
ass
i
f
i
e
r
s
li
k
e
XGB
o
o
s
t
,
B
a
g
g
i
n
g
,
e
x
t
r
a
t
r
e
es
(
E
T
)
,
r
a
n
d
o
m
f
o
r
e
s
t
,
a
n
d
A
d
aB
o
o
s
t.
B
y
u
s
i
n
g
t
h
e
c
h
i
-
s
q
u
a
r
e
s
t
a
ti
s
ti
c
a
l
m
e
t
h
o
d
f
o
r
f
e
a
t
u
r
e
s
e
l
e
ct
io
n
,
t
h
e
i
r
m
o
d
e
l
a
c
h
i
e
v
e
d
h
i
g
h
p
e
r
f
o
r
m
a
n
c
e
i
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
4
,
No
.
2
,
Ap
r
il 2
0
2
5
:
1
2
3
2
-
1
2
4
1
1234
a
c
c
u
r
a
c
y
,
r
e
c
a
l
l
,
p
r
e
c
is
i
o
n
,
a
n
d
F
1
-
s
c
o
r
e
.
A
m
o
n
g
t
h
e
c
l
a
s
s
i
f
i
e
r
s
,
t
h
e
X
GB
o
o
s
t
e
n
s
e
m
b
l
e
e
x
c
e
ll
e
d
i
n
d
e
t
e
c
t
i
n
g
a
n
d
c
l
a
s
s
i
f
y
i
n
g
I
I
o
T
a
t
t
ac
k
s
,
o
f
f
e
r
i
n
g
a
s
i
g
n
i
f
i
ca
n
t
e
n
h
a
n
c
e
m
en
t
t
o
I
D
S
i
n
c
o
m
p
le
x
I
I
o
T
e
n
v
i
r
o
n
m
e
n
t
s
.
‒
Featu
r
e
en
g
i
n
ee
r
in
g
an
d
s
elec
tio
n
:
R
ajash
ek
ar
an
et
a
l
.
[
1
4
]
in
tr
o
d
u
ce
d
th
e
r
ec
u
r
s
iv
e
f
ea
t
u
r
e
elim
in
atio
n
(
RF
E)
-
lo
n
g
s
h
o
r
t
-
ter
m
m
em
o
r
y
(
L
STM
)
-
I
DS
m
o
d
el,
c
o
m
b
i
n
in
g
R
FE
with
L
STM
n
etwo
r
k
s
f
o
r
im
p
r
o
v
ed
f
ea
tu
r
e
s
elec
tio
n
a
n
d
d
y
n
a
m
ic
th
r
ea
t
d
etec
tio
n
in
cl
o
u
d
s
ec
u
r
i
ty
.
T
h
eir
m
o
d
el
ac
h
iev
e
d
9
1
.
5
0
%
an
d
9
2
.
2
1
%
ac
cu
r
ac
y
o
n
NSL
-
KDD
an
d
B
o
T
-
I
o
T
d
atasets
,
r
esp
ec
tiv
ely
,
an
d
s
h
o
wed
p
r
ec
is
io
n
o
f
4
7
.
5
4
% a
n
d
r
e
ca
ll o
f
8
2
.
3
1
%,
h
ig
h
lig
h
tin
g
its
ef
f
ec
t
iv
en
ess
in
h
an
d
lin
g
co
m
p
lex
i
n
tr
u
s
io
n
s
ce
n
ar
io
s
in
I
I
o
T
en
v
i
r
o
n
m
en
ts
.
‒
I
DS
f
o
r
r
ea
l
-
tim
e
ap
p
licati
o
n
s
:
E
f
f
icien
t
r
ea
l
-
tim
e
d
et
ec
tio
n
is
cr
u
cial
f
o
r
I
I
o
T
en
v
ir
o
n
m
e
n
ts
.
Alo
s
aim
i
an
d
Alm
u
tair
i
[
1
5
]
p
r
o
p
o
s
ed
a
n
o
v
el
ap
p
r
o
ac
h
f
o
r
lig
h
tweig
h
t
I
DS
b
y
o
p
t
im
izin
g
f
ea
tu
r
e
s
elec
tio
n
.
T
h
ey
ev
al
u
ated
v
ar
i
o
u
s
ML
alg
o
r
ith
m
s
o
n
th
e
B
o
T
-
I
o
T
2
0
1
8
d
ataset,
f
o
cu
s
in
g
o
n
id
e
n
tify
in
g
th
e
m
o
s
t
ef
f
ec
tiv
e
f
ea
tu
r
e
p
air
s
to
d
ev
elo
p
en
e
r
g
y
-
ef
f
icien
t
I
DS
.
T
h
eir
ap
p
r
o
ac
h
d
e
m
o
n
s
tr
ate
d
th
at
s
elec
tin
g
o
p
tim
al
f
ea
tu
r
e
p
air
s
ca
n
s
ig
n
if
ican
tly
en
h
an
ce
d
etec
tio
n
ac
cu
r
ac
y
wh
ile
m
ain
tain
i
n
g
s
y
s
t
em
ef
f
icien
cy
,
ac
h
iev
in
g
o
v
er
9
0
% a
cc
u
r
ac
y
with
lig
h
tweig
h
t m
o
d
els
.
‒
Nex
t
-
g
en
s
ec
u
r
ity
:
I
d
o
u
g
lid
et
a
l
.
[
1
6
]
d
is
cu
s
s
ed
in
teg
r
atin
g
I
DS
with
ML
tech
n
iq
u
es
to
e
n
h
an
ce
s
ec
u
r
ity
in
I
I
o
T
en
v
ir
o
n
m
e
n
ts
.
T
h
eir
p
ap
er
p
r
o
v
id
es
in
s
ig
h
ts
in
to
th
e
n
ex
t
-
g
en
e
r
atio
n
s
ec
u
r
it
y
m
ea
s
u
r
es
f
o
r
I
n
d
u
s
tr
y
4
.
0
,
f
o
cu
s
in
g
o
n
r
esil
ien
ce
an
d
a
d
v
an
ce
d
th
r
ea
t
d
ete
ctio
n
.
‒
No
v
el
an
o
m
aly
d
etec
tio
n
m
o
d
el
: I
d
o
u
g
lid
et
a
l
.
[
1
7
]
p
r
o
p
o
s
e
d
a
n
o
v
el
a
n
o
m
aly
d
etec
tio
n
m
o
d
el
u
s
in
g
ML
tech
n
iq
u
es
tailo
r
ed
f
o
r
I
I
o
T
en
v
ir
o
n
m
e
n
ts
.
T
h
eir
s
tu
d
y
h
ig
h
lig
h
ts
th
e
ef
f
ec
tiv
en
ess
o
f
ad
v
a
n
ce
d
alg
o
r
ith
m
s
in
im
p
r
o
v
in
g
d
etec
tio
n
ac
cu
r
a
cy
an
d
a
d
d
r
ess
in
g
s
p
ec
if
ic
ch
a
llen
g
es in
I
I
o
T
s
ec
u
r
ity
.
‒
Sec
u
r
it
y
c
h
al
le
n
g
es
i
n
I
I
o
T
:
A
v
d
i
b
asi
c
et
a
l
.
[
3
]
ad
d
r
ess
t
h
e
c
h
al
le
n
g
es
o
f
d
et
ec
ti
n
g
cy
b
e
r
-
at
tac
k
s
i
n
I
o
T
/I
I
o
T
en
v
i
r
o
n
m
e
n
ts
,
s
p
e
ci
f
ic
all
y
f
o
c
u
s
i
n
g
o
n
M
o
d
b
u
s
p
r
o
to
c
o
l
-
b
as
ed
s
y
s
t
em
s
.
T
h
e
y
p
r
o
p
o
s
e
a
n
o
v
el
d
ee
p
le
ar
n
i
n
g
ar
ch
ite
ct
u
r
e
t
h
a
t i
m
p
r
o
v
es
u
p
o
n
t
r
a
d
it
io
n
al
m
et
h
o
d
s
b
y
en
h
an
cin
g
b
o
t
h
b
i
n
a
r
y
a
n
d
m
u
l
ti
-
cl
a
s
s
class
i
f
i
ca
t
io
n
o
f
at
ta
ck
s
.
T
h
ei
r
ex
p
e
r
i
m
e
n
ts
d
e
m
o
n
s
tr
ate
t
h
at
t
h
e
p
r
o
p
o
s
e
d
ar
c
h
it
ec
t
u
r
e
c
o
n
s
is
te
n
tl
y
o
u
t
p
er
f
o
r
m
s
e
x
is
ti
n
g
m
o
d
els,
o
f
f
e
r
i
n
g
e
f
f
ec
t
iv
e
d
e
te
cti
o
n
a
n
d
cl
ass
i
f
ic
ati
o
n
o
f
cy
b
e
r
-
att
ac
k
s
o
n
I
I
o
T
d
e
v
ic
es
.
‒
H
y
p
er
p
ar
a
m
eter
-
o
p
tim
izatio
n
:
I
n
th
eir
s
tu
d
y
,
C
h
im
p
h
lee
an
d
C
h
im
p
h
lee
[
1
8
]
ex
p
lo
r
e
th
e
u
s
e
o
f
h
y
p
er
p
ar
am
eter
-
o
p
tim
ized
XGBo
o
s
t
f
o
r
in
t
r
u
s
io
n
d
etec
t
io
n
o
n
th
e
C
SE
-
C
I
C
-
I
DS2
0
1
8
d
ataset.
B
y
f
in
e
-
tu
n
in
g
p
ar
a
m
eter
s
s
u
ch
as
lear
n
in
g
r
ate,
m
ax
d
e
p
th
,
an
d
g
am
m
a,
th
eir
a
p
p
r
o
ac
h
s
ig
n
if
i
ca
n
tly
im
p
r
o
v
es
m
o
d
el
p
er
f
o
r
m
a
n
ce
.
T
h
e
o
p
ti
m
ized
XGBo
o
s
t
alg
o
r
ith
m
o
u
tp
er
f
o
r
m
s
o
th
er
t
r
ad
itio
n
al
ML
tech
n
iq
u
es,
ac
h
iev
in
g
a
n
im
p
r
ess
iv
e
r
ec
ei
v
er
o
p
er
atin
g
ch
ar
ac
ter
is
tic
(
R
OC
)
s
co
r
e
o
f
0
.
9
9
9
9
2
6
a
n
d
h
i
g
h
ac
cu
r
ac
y
f
o
r
d
etec
tin
g
n
etwo
r
k
in
tr
u
s
io
n
s
.
T
h
eir
wo
r
k
d
em
o
n
s
tr
ates
th
e
i
m
p
o
r
tan
ce
o
f
h
y
p
er
p
ar
am
eter
o
p
tim
i
za
tio
n
in
ad
ap
tin
g
ML
m
o
d
els
to
n
etwo
r
k
s
ec
u
r
ity
task
s
,
h
ig
h
lig
h
tin
g
XGBo
o
s
t
's
ca
p
ac
ity
f
o
r
h
ig
h
ac
cu
r
ac
y
wh
ile
m
ain
tain
in
g
lo
w
f
alse p
o
s
itiv
es
.
3.
M
E
T
H
O
D
T
h
is
s
e
c
ti
o
n
o
u
tlin
es
th
e
s
tep
-
by
-
s
tep
ap
p
r
o
ac
h
ad
o
p
ted
in
th
e
s
tu
d
y
,
in
clu
d
in
g
d
ata
p
r
ep
ar
atio
n
,
p
r
ep
r
o
ce
s
s
in
g
,
f
ea
tu
r
e
en
g
in
e
er
in
g
,
m
o
d
el
d
ev
elo
p
m
e
n
t,
a
n
d
ev
alu
atio
n
.
T
h
e
m
eth
o
d
o
l
o
g
y
is
d
esig
n
ed
to
en
s
u
r
e
r
ep
r
o
d
u
ci
b
ilit
y
b
y
p
r
o
v
i
d
i
n
g
d
etailed
d
escr
ip
tio
n
s
o
f
th
e
tech
n
iq
u
es a
n
d
to
o
ls
u
s
ed
.
3
.
1
.
Da
t
a
prepa
ra
t
i
o
n
3
.
1
.
1
.
UNSW
-
NB
1
5
-
v2
da
t
a
s
et
des
cr
iptio
n
T
h
e
UNSW
-
N
B
1
5
d
ataset,
d
ev
elo
p
ed
b
y
UNSW
C
an
b
er
r
a'
s
C
y
b
er
R
an
g
e
L
ab
,
is
a
k
ey
r
eso
u
r
ce
f
o
r
n
etwo
r
k
i
n
tr
u
s
io
n
d
etec
tio
n
r
e
s
ea
r
ch
.
I
t
i
n
clu
d
es
2
,
5
4
0
,
0
4
4
i
n
s
tan
ce
s
with
5
3
f
ea
tu
r
es,
c
o
v
er
in
g
a
b
r
o
ad
r
an
g
e
o
f
n
etwo
r
k
t
r
af
f
ic
attr
ib
u
tes,
s
u
ch
as
f
lo
w
ch
ar
ac
ter
i
s
tics
,
b
asi
c
an
d
co
n
te
n
t
-
r
elate
d
in
f
o
r
m
ati
o
n
,
an
d
tim
e
-
b
ased
m
etr
ics.
T
h
e
d
ataset
also
f
ea
tu
r
es
g
en
er
ated
m
etr
ics
lik
e
co
n
n
ec
tio
n
c
o
u
n
ts
.
I
t
p
r
o
v
i
d
e
s
clea
r
d
is
tin
ctio
n
s
b
etwe
en
n
o
r
m
al
tr
af
f
ic
a
n
d
v
ar
io
u
s
ty
p
es
o
f
attac
k
s
,
in
clu
d
in
g
d
en
ial
o
f
s
er
v
ice
(
Do
S
)
,
ex
p
lo
its
,
an
d
r
ec
o
n
n
aiss
an
ce
,
m
ak
i
n
g
it
ess
en
tial
f
o
r
ev
alu
atin
g
I
DS.
W
ith
a
s
ize
o
f
7
0
0
MB,
it
is
id
ea
l
f
o
r
ML
an
d
d
ee
p
lear
n
in
g
ap
p
licatio
n
s
in
n
etw
o
r
k
s
ec
u
r
ity
[
1
9
]
,
[
2
0
]
.
W
ith
th
e
d
ataset
s
elec
ted
,
th
e
f
o
llo
win
g
p
r
ep
r
o
ce
s
s
in
g
s
tep
s
wer
e
ca
r
r
ied
o
u
t to
en
s
u
r
e
d
ata
r
ea
d
in
ess
f
o
r
ML
an
aly
s
is
.
3
.
2
.
Da
t
a
p
re
pro
ce
s
s
ing
3
.
2
.
1
.
H
a
nd
le
m
is
s
ing
v
a
lues
Han
d
lin
g
m
is
s
in
g
v
alu
es
is
e
s
s
en
tial
to
av
o
id
b
iased
r
esu
lts
an
d
en
s
u
r
e
ac
cu
r
ate
m
o
d
el
p
r
ed
ictio
n
s
.
T
h
is
p
r
o
ce
s
s
in
v
o
lv
es
tech
n
iq
u
es
s
u
ch
as
im
p
u
tatio
n
,
wh
er
e
m
is
s
in
g
d
ata
is
r
ep
lace
d
with
s
tatis
t
ical
m
ea
s
u
r
es
lik
e
th
e
m
ea
n
,
m
ed
ian
,
o
r
m
o
d
e.
T
h
ese
m
et
h
o
d
s
p
r
eser
v
e
th
e
d
ataset's
co
n
s
is
ten
cy
an
d
in
teg
r
ity
,
en
a
b
lin
g
m
o
r
e
r
eliab
le
an
aly
s
is
an
d
s
u
b
s
eq
u
e
n
t m
o
d
elin
g
[
2
1
]
.
3
.
2
.
2
.
Co
lum
n
d
ro
pp
ing
T
o
r
ed
u
ce
n
o
is
e
an
d
co
m
p
u
ta
tio
n
al
lo
ad
,
ir
r
ele
v
an
t
o
r
r
ed
u
n
d
an
t
co
lu
m
n
s
wer
e
r
em
o
v
ed
f
r
o
m
th
e
d
ataset.
T
h
is
p
r
e
p
r
o
ce
s
s
in
g
s
tep
is
cr
u
cial
f
o
r
o
p
tim
izin
g
th
e
d
ataset
b
y
r
etain
in
g
o
n
l
y
th
e
m
o
s
t
r
elev
a
n
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
B
o
o
s
tin
g
I
I
o
T in
tr
u
s
io
n
d
etec
t
io
n
:
leve
r
a
g
in
g
ma
ch
in
e
lea
r
n
in
g
a
n
d
fea
tu
r
e
s
elec
tio
n
…
(
L
a
h
ce
n
I
d
o
u
g
lid
)
1235
f
ea
tu
r
es.
C
o
n
s
eq
u
en
tly
,
th
e
m
a
ch
in
e
lear
n
in
g
m
o
d
els
b
en
e
f
it
f
r
o
m
im
p
r
o
v
e
d
ef
f
icie
n
cy
a
n
d
p
r
ed
ictio
n
ac
cu
r
ac
y
as th
ey
ar
e
tr
ain
ed
o
n
a
s
tr
ea
m
lin
ed
d
ataset.
3
.
2
.
3
.
O
ne
-
ho
t
enco
din
g
C
ateg
o
r
ical
v
ar
iab
les
in
th
e
d
ataset
wer
e
co
n
v
er
ted
in
t
o
b
in
ar
y
r
ep
r
esen
tatio
n
s
u
s
i
n
g
o
n
e
-
h
o
t
en
co
d
in
g
.
T
h
is
tech
n
iq
u
e
,
im
p
lem
en
ted
with
th
e
p
an
d
as.g
e
t_
d
u
m
m
ies()
f
u
n
ctio
n
in
Py
th
o
n
,
tr
a
n
s
f
o
r
m
s
ea
c
h
ca
teg
o
r
ical
attr
ib
u
te
in
to
s
ep
a
r
ate
b
in
ar
y
(
0
o
r
1
)
c
o
lu
m
n
s
.
B
y
d
o
in
g
s
o
,
m
ac
h
in
e
lear
n
i
n
g
alg
o
r
ith
m
s
ca
n
in
ter
p
r
et
an
d
p
r
o
ce
s
s
th
ese
f
ea
tu
r
es e
f
f
ec
tiv
ely
,
e
n
h
an
cin
g
m
o
d
el
co
m
p
atib
ilit
y
an
d
p
er
f
o
r
m
an
ce
[
2
2
]
.
3
.
2
.
4
.
F
ea
t
ure
s
ca
lin
g
Featu
r
e
s
ca
lin
g
was a
p
p
lied
to
en
s
u
r
e
th
at
all
f
ea
tu
r
es c
o
n
tr
i
b
u
te
eq
u
ally
to
th
e
m
o
d
el’
s
p
e
r
f
o
r
m
a
n
ce
,
p
ar
ticu
lar
ly
f
o
r
alg
o
r
ith
m
s
th
at
r
ely
o
n
d
is
tan
ce
m
etr
ics.
T
h
is
was
d
o
n
e
u
s
in
g
th
e
Mi
n
Ma
x
Scaler
f
r
o
m
s
cik
it
-
lear
n
,
wh
ich
n
o
r
m
alize
d
th
e
f
ea
tu
r
es
to
a
r
a
n
g
e
b
etwe
en
0
a
n
d
1
[
2
3
]
.
Fo
llo
win
g
d
ata
p
r
ep
r
o
ce
s
s
in
g
,
f
ea
tu
r
e
en
g
in
ee
r
in
g
tech
n
iq
u
es we
r
e
ap
p
lied
to
f
u
r
t
h
er
r
e
f
in
e
th
e
d
ataset
an
d
en
h
an
ce
m
o
d
el
p
er
f
o
r
m
an
ce
.
3
.
3
.
F
e
a
t
ure
e
ng
ineering
3
.
3
.
1
.
F
ea
t
ure
s
elec
t
io
n
Featu
r
e
s
elec
tio
n
is
cr
itical
to
i
m
p
r
o
v
i
n
g
t
h
e
p
e
r
f
o
r
m
an
ce
a
n
d
ef
f
icien
c
y
o
f
ML
m
o
d
els.
I
n
th
is
s
tu
d
y
,
two
f
ea
tu
r
e
s
elec
tio
n
tech
n
iq
u
es we
r
e
em
p
lo
y
ed
,
‒
R
F
E
:
R
F
E
w
a
s
u
s
e
d
t
o
i
t
e
r
a
t
i
v
e
l
y
e
l
i
m
i
n
a
t
e
t
h
e
l
e
a
s
t
i
m
p
o
r
t
a
n
t
f
e
a
t
u
r
e
s
,
r
e
d
u
c
i
n
g
t
h
e
d
i
m
e
n
s
i
o
n
a
l
i
t
y
o
f
t
h
e
d
a
t
a
w
h
i
l
e
r
e
t
a
i
n
i
n
g
p
r
e
d
i
c
t
i
v
e
p
o
w
e
r
.
T
h
i
s
p
r
o
c
e
s
s
h
e
l
p
s
t
o
f
o
c
u
s
t
h
e
m
o
d
e
l
o
n
t
h
e
m
o
s
t
i
m
p
a
c
t
f
u
l
v
a
r
i
a
b
l
e
s
[
2
4
]
.
‒
PC
A:
PC
A
was
u
s
ed
to
tr
an
s
f
o
r
m
th
e
d
ataset
in
to
a
lo
wer
-
d
im
en
s
io
n
al
s
p
ac
e
b
y
s
elec
tin
g
th
e
p
r
in
cip
al
co
m
p
o
n
en
ts
th
at
ex
p
lain
th
e
m
o
s
t
v
ar
ian
ce
.
T
h
is
m
eth
o
d
h
elp
ed
i
n
r
ed
u
cin
g
o
v
er
f
itti
n
g
an
d
im
p
r
o
v
in
g
co
m
p
u
tatio
n
al
e
f
f
icien
cy
,
esp
e
cially
f
o
r
h
i
g
h
-
d
im
e
n
s
io
n
al
d
a
ta
[
2
4
]
.
3
.
3
.
2
.
Dim
ens
io
na
lity
r
educt
io
n
Dim
en
s
io
n
ality
r
ed
u
ctio
n
wa
s
p
er
f
o
r
m
ed
u
s
in
g
PC
A
to
r
ed
u
ce
th
e
n
u
m
b
er
o
f
f
ea
tu
r
es
wh
ile
p
r
eser
v
in
g
th
e
u
n
d
er
l
y
in
g
s
tr
u
ctu
r
e
o
f
th
e
d
ata.
B
y
co
m
p
r
es
s
in
g
th
e
f
ea
tu
r
e
s
p
ac
e,
PC
A
e
n
s
u
r
ed
th
at
th
e
ML
m
o
d
els
r
em
ai
n
ed
r
o
b
u
s
t
an
d
e
asier
to
in
ter
p
r
et.
T
h
is
also
h
e
lp
ed
m
itig
ate
is
s
u
es
o
f
o
v
er
f
itti
n
g
,
as
t
h
e
m
o
d
el
wo
u
ld
n
o
t
b
e
o
v
e
r
wh
elm
ed
b
y
ir
r
elev
an
t
f
ea
tu
r
es
[
2
5
]
.
Af
ter
r
ef
in
in
g
th
e
d
ataset
th
r
o
u
g
h
f
ea
tu
r
e
en
g
in
ee
r
i
n
g
,
th
e
ML
m
o
d
els we
r
e
d
ev
elo
p
e
d
an
d
tr
ai
n
ed
.
3
.
4
.
M
o
del
d
ev
elo
pm
ent
3
.
4
.
1
.
Da
t
a
s
pli
t
t
ing
T
h
e
d
ataset
was
s
p
lit
in
to
tr
ain
in
g
an
d
test
in
g
s
u
b
s
ets
in
an
8
0
:2
0
r
atio
u
s
in
g
th
e
tr
ain
_
t
est_
s
p
lit()
f
u
n
ctio
n
f
r
o
m
th
e
s
cik
it
-
lear
n
l
ib
r
ar
y
.
A
r
a
n
d
o
m
_
s
tate
o
f
4
2
was
s
et
to
en
s
u
r
e
th
e
r
ep
r
o
d
u
cib
ilit
y
o
f
th
e
r
esu
lts
.
T
h
e
tr
ain
in
g
s
et
was
u
s
ed
to
tr
ain
th
e
ML
m
o
d
els,
wh
ile
t
h
e
test
in
g
s
et
was
r
eser
v
ed
f
o
r
ev
alu
atin
g
th
eir
g
en
er
aliza
tio
n
a
b
ilit
y
.
3
.
4
.
2
.
Cro
s
s
-
v
a
lid
a
t
io
n
C
r
o
s
s
-
v
alid
atio
n
was
em
p
lo
y
ed
to
f
in
e
-
tu
n
e
t
h
e
h
y
p
er
p
a
r
am
eter
s
o
f
th
e
m
o
d
els
an
d
a
s
s
es
s
th
eir
p
er
f
o
r
m
an
ce
.
A
5
-
f
o
ld
cr
o
s
s
-
v
alid
atio
n
tech
n
iq
u
e
was
u
s
ed
,
en
s
u
r
in
g
th
at
ea
ch
f
o
l
d
s
er
v
ed
as
a
v
alid
atio
n
s
et
o
n
ce
,
wh
ile
th
e
r
em
ain
in
g
f
o
ld
s
wer
e
u
s
ed
f
o
r
tr
ain
i
n
g
.
T
h
is
p
r
o
ce
s
s
h
elp
s
m
itig
ate
o
v
er
f
itti
n
g
an
d
en
s
u
r
es
th
at
th
e
m
o
d
el'
s
p
er
f
o
r
m
a
n
ce
is
ev
alu
ated
o
n
m
u
ltip
le
s
u
b
s
ets o
f
th
e
d
ata
[
2
6
]
.
3
.
4
.
3
.
T
ra
ini
ng
a
nd
v
a
lid
a
t
io
n
T
h
e
ML
class
if
ier
s
u
s
ed
in
th
is
s
tu
d
y
we
r
e
tr
ain
e
d
o
n
t
h
e
p
r
ep
ar
ed
d
ata,
an
d
v
alid
atio
n
was
p
er
f
o
r
m
e
d
to
ass
ess
m
o
d
el
p
er
f
o
r
m
an
c
e
an
d
tu
n
e
h
y
p
er
p
ar
a
m
eter
s
.
R
an
d
o
m
f
o
r
est
,
XGBo
o
s
t,
Ad
aBo
o
s
t,
g
r
ad
ien
t
b
o
o
s
tin
g
,
a
n
d
m
u
lti
-
lay
e
r
p
er
c
ep
tr
o
n
(
ML
P)
we
r
e
th
e
p
r
i
m
ar
y
m
o
d
els
test
ed
,
with
ea
ch
co
n
f
ig
u
r
ed
t
o
o
p
tim
ize
d
etec
tio
n
ac
cu
r
ac
y
an
d
co
m
p
u
tatio
n
al
ef
f
icien
cy
[
2
7
]
.
T
h
e
f
o
llo
win
g
ML
m
o
d
els
wer
e
d
e
v
elo
p
ed
a
n
d
t
r
ain
ed
in
th
is
s
tu
d
y
:
r
an
d
o
m
f
o
r
est
,
XGBo
o
s
t,
Ad
aBo
o
s
t,
g
r
ad
ien
t b
o
o
s
tin
g
,
an
d
ML
P.
E
ac
h
m
o
d
el
was
f
in
e
-
tu
n
e
d
u
s
in
g
g
r
id
s
ea
r
ch
to
o
p
tim
ize
h
y
p
er
p
ar
am
eter
s
.
Fo
r
e
x
am
p
le,
t
h
e
r
an
d
o
m
f
o
r
est
m
o
d
el
was
tr
ain
ed
with
1
0
0
esti
m
ato
r
s
an
d
a
m
a
x
im
u
m
d
ep
th
o
f
1
0
,
wh
ile
th
e
XGBo
o
s
t
m
o
d
el
was
o
p
tim
ized
with
a
lear
n
in
g
r
ate
o
f
0
.
1
an
d
2
0
0
b
o
o
s
tin
g
r
o
u
n
d
s
.
T
h
e
m
o
d
els
wer
e
tr
ain
ed
u
s
in
g
th
e
s
cik
it
-
lear
n
an
d
XGBo
o
s
t
lib
r
ar
ies
in
Py
th
o
n
,
lev
er
a
g
in
g
t
h
e
Kag
g
le
v
i
r
tu
al
m
ac
h
in
e
(
VM
)
e
n
v
ir
o
n
m
e
n
t,
wh
ich
in
clu
d
ed
a
m
u
lti
-
co
r
e
I
n
tel
Xeo
n
p
r
o
ce
s
s
o
r
,
1
3
GB
o
f
R
AM
,
an
d
ac
ce
s
s
to
an
NVI
DI
A
T
esla P1
0
0
GP
U
[
2
6
]
.
T
h
e
m
ath
em
atica
l
f
o
r
m
u
latio
n
s
o
f
th
e
m
o
d
els ar
e
as f
o
llo
ws
:
−
R
an
d
o
m
f
o
r
est:
r
an
d
o
m
f
o
r
est
is
an
en
s
em
b
le
lear
n
in
g
m
eth
o
d
th
at
co
n
s
tr
u
cts
m
u
ltip
le
d
ec
is
io
n
tr
ee
s
.
E
ac
h
tr
ee
p
r
o
v
id
es
a
p
r
ed
ictio
n
,
a
n
d
th
e
f
in
al
o
u
tp
u
t
is
th
e
m
ajo
r
ity
v
o
te
(
class
if
icatio
n
)
o
r
th
e
av
er
a
g
e
(
r
eg
r
ess
io
n
)
o
f
th
e
p
r
ed
ictio
n
s
.
T
h
e
m
ath
em
atica
l f
o
r
m
u
latio
n
f
o
r
r
an
d
o
m
f
o
r
est
is
:
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
4
,
No
.
2
,
Ap
r
il 2
0
2
5
:
1
2
3
2
-
1
2
4
1
1236
(
)
=
1
∑
ℎ
(
)
=
1
(
1
)
W
h
er
e
f
(
x
)
i
s
t
h
e
f
i
n
a
l
p
r
e
d
i
c
t
i
o
n
,
N
i
s
t
h
e
n
u
m
b
e
r
o
f
d
e
c
i
s
i
o
n
t
r
e
e
s
,
a
n
d
ℎ
(
)
i
s
t
h
e
p
r
ed
i
c
t
i
o
n
f
r
o
m
t
h
e
-
t
h
t
r
e
e
.
−
XGBo
o
s
t
:
XG
B
o
o
s
t
is
a
g
r
ad
ien
t
b
o
o
s
tin
g
alg
o
r
ith
m
t
h
at
ad
d
s
n
ew
tr
ee
s
to
co
r
r
ec
t
th
e
r
esid
u
al
er
r
o
r
s
o
f
p
r
ev
io
u
s
tr
ee
s
.
T
h
e
o
b
jectiv
e
f
u
n
ctio
n
is
o
p
tim
ized
iter
ativ
ely
.
T
h
e
o
b
jectiv
e
f
u
n
ctio
n
f
o
r
XGBo
o
s
t is
:
(
)
=
∑
(
,
̂
(
−
1
)
+
(
)
)
+
(
)
=
0
(
2
)
W
h
er
e
L(
θ
)
is
t
h
e
o
b
jectiv
e
f
u
n
ctio
n
;
y
an
d
y
̂
ar
e
th
e
ac
t
u
al
an
d
p
r
ed
icted
v
alu
es,
r
esp
ec
tiv
ely
;
(
x
)
is
th
e
n
ew
tr
ee
ad
d
ed
at
iter
atio
n
; a
n
d
Ω
(
f
)
is
th
e
r
eg
u
lar
izatio
n
ter
m
.
−
Ad
aBo
o
s
t:
Ad
aBo
o
s
t
i
s
an
en
s
em
b
le
m
eth
o
d
th
at
co
m
b
in
es
wea
k
class
if
ier
s
to
f
o
r
m
a
s
tr
o
n
g
class
if
ier
.
T
h
e
alg
o
r
ith
m
ass
ig
n
s
h
ig
h
e
r
weig
h
ts
to
m
is
clas
s
if
ied
in
s
t
an
ce
s
,
th
er
eb
y
im
p
r
o
v
i
n
g
f
u
t
u
r
e
p
r
ed
ictio
n
s
.
T
h
e
f
in
al
p
r
ed
ictio
n
is
g
iv
e
n
b
y
:
f
(
x
)
=
s
ign
∑
α
t
h
t
(
x
)
T
t
=
1
(
3
)
W
h
er
e
ℎ
(
)
is
th
e
wea
k
class
if
ier
at
iter
atio
n
;
α
is
th
e
weig
h
t
ass
ig
n
ed
to
th
e
wea
k
class
if
ier
;
an
d
f
(
x
)
is
th
e
f
in
al
p
r
e
d
ictio
n
.
−
Gr
ad
ien
t
b
o
o
s
tin
g
:
Gr
a
d
ien
t
b
o
o
s
tin
g
wo
r
k
s
b
y
s
eq
u
en
tially
b
u
ild
in
g
m
o
d
els
th
at
m
in
im
i
ze
th
e
r
esid
u
als
(
er
r
o
r
s
)
o
f
p
r
ev
io
u
s
m
o
d
els.
T
h
e
g
en
er
al
l
o
s
s
f
u
n
ctio
n
f
o
r
b
i
n
ar
y
class
if
icatio
n
is
:
L
(
y
i
,
y
̂
)
=
∑
[
y
i
l
og
(
y
̂
i
)
+
(
1
−
y
̂
i
)
l
og
(
1
−
y
̂
i
)
]
N
i
=
0
(
4
)
W
h
er
e
y
an
d
y
̂
ar
e
th
e
ac
tu
al
an
d
p
r
ed
icted
lab
els
; a
n
d
L
(
y
,
y
̂
)
is
th
e
lo
s
s
f
u
n
ctio
n
th
at
is
m
in
im
ized
.
−
MLP
:
ML
P
i
s
a
ty
p
e
o
f
ar
tific
ial
n
eu
r
al
n
etwo
r
k
with
m
u
ltip
le
lay
er
s
o
f
n
eu
r
o
n
s
.
E
ac
h
n
eu
r
o
n
co
m
p
u
tes
a
weig
h
ted
s
u
m
o
f
th
e
in
p
u
ts
an
d
p
ass
es
it
th
r
o
u
g
h
a
n
ac
tiv
atio
n
f
u
n
ctio
n
.
T
h
e
ML
P
m
o
d
el
i
s
r
ep
r
esen
ted
as
:
f
(
x
)
=
σ
(
W
2
⋅
σ
(
W
1
⋅
x
+
b
1
)
+
b
2
)
(
5
)
W
h
er
e
x
is
th
e
in
p
u
t
v
ec
to
r
;
1
an
d
2
ar
e
th
e
weig
h
t
m
atr
ices
f
o
r
th
e
h
id
d
en
an
d
o
u
tp
u
t
lay
er
s
,
r
esp
ec
tiv
ely
;
b
1
an
d
2
ar
e
th
e
b
ias ter
m
s
;
σ
is
th
e
ac
tiv
atio
n
f
u
n
ctio
n
(
e.
g
.
,
r
ec
tifie
d
lin
ea
r
u
n
it (
R
eL
U
)
o
r
s
ig
m
o
id
)
; a
n
d
f
(
x
)
is
th
e
o
u
t
p
u
t o
f
th
e
ML
P.
On
ce
tr
ain
ed
,
th
e
m
o
d
els we
r
e
ev
alu
ated
b
ased
o
n
th
ei
r
p
er
f
o
r
m
an
ce
ac
r
o
s
s
s
ev
er
al
k
ey
m
etr
ics.
3
.
5
.
M
o
del e
v
a
lua
t
i
o
n
3
.
5
.
1
.
E
v
a
lua
t
io
n
m
e
t
rics
T
h
e
m
o
d
els
wer
e
ev
alu
ated
u
s
i
n
g
s
ev
er
al
m
etr
ics:
ac
c
u
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
F1
s
co
r
e,
a
n
d
a
r
ea
u
n
d
er
th
e
cu
r
v
e
(
AUC
)
-
R
OC
.
Acc
u
r
ac
y
m
ea
s
u
r
es
th
e
p
r
o
p
o
r
tio
n
o
f
co
r
r
ec
t
p
r
ed
ictio
n
s
,
wh
ile
p
r
ec
is
io
n
an
d
r
ec
all
p
r
o
v
id
e
in
s
ig
h
ts
in
t
o
th
e
m
o
d
el'
s
h
an
d
lin
g
o
f
p
o
s
itiv
e
cla
s
s
p
r
ed
ictio
n
s
[
2
8
]
.
T
h
e
F1
s
co
r
e
b
alan
ce
s
t
h
e
tr
ad
e
-
o
f
f
b
etwe
en
p
r
ec
is
io
n
a
n
d
r
ec
all,
an
d
AUC
-
R
OC
as
s
ess
es
th
e
m
o
d
el’
s
ab
ilit
y
to
d
is
tin
g
u
is
h
b
etwe
en
class
es a
t v
ar
io
u
s
th
r
esh
o
ld
s
ettin
g
s
[
2
9
]
.
3
.
5
.
2
.
Respo
ns
e
d
ec
i
s
io
n
T
h
e
m
o
d
els
wer
e
u
s
ed
to
clas
s
if
y
n
etwo
r
k
tr
a
f
f
ic
as
eith
er
n
o
r
m
al
o
r
an
o
m
alo
u
s
.
A
t
h
r
esh
o
ld
-
b
ase
d
d
ec
is
io
n
m
ec
h
an
is
m
was
ap
p
li
ed
,
wh
er
e
a
n
o
m
alies
wer
e
f
lag
g
ed
b
ased
o
n
th
e
m
o
d
el'
s
o
u
tp
u
t.
T
h
e
r
esu
lts
wer
e
an
aly
ze
d
t
o
m
i
n
im
ize
f
alse p
o
s
itiv
es
an
d
f
alse n
eg
ativ
es,
e
n
s
u
r
in
g
th
at
le
g
itima
te
tr
af
f
ic
wa
s
n
o
t
m
is
class
if
ied
wh
ile
ac
tu
al
th
r
ea
ts
wer
e
ac
cu
r
ately
d
etec
ted
.
Fig
u
r
e
1
illu
s
tr
ates
th
e
m
eth
o
d
o
lo
g
y
wo
r
k
f
lo
w
,
p
r
o
v
i
d
in
g
a
v
is
u
al
r
ep
r
esen
tatio
n
o
f
th
e
s
tep
-
by
-
s
tep
p
r
o
ce
s
s
.
3
.
6
.
E
x
perim
ent
a
l set
up
All
ex
p
er
im
en
ts
wer
e
co
n
d
u
cted
in
a
Kag
g
le
VM
e
n
v
ir
o
n
m
en
t,
wh
ich
p
r
o
v
id
ed
a
c
lo
u
d
-
b
ased
in
f
r
astru
ctu
r
e
with
s
u
f
f
icien
t
c
o
m
p
u
tin
g
p
o
wer
.
T
h
e
s
etu
p
in
clu
d
ed
a
m
u
lti
-
co
r
e
I
n
tel
Xeo
n
p
r
o
ce
s
s
o
r
,
1
3
GB
o
f
R
AM
,
an
d
ac
ce
s
s
to
an
N
VI
DI
A
T
esla
P1
0
0
GPU
f
o
r
ac
ce
ler
at
ed
p
r
o
ce
s
s
in
g
.
T
h
e
P
y
th
o
n
p
r
o
g
r
am
m
in
g
lan
g
u
ag
e
was u
s
ed
,
with
k
ey
li
b
r
ar
ies in
clu
d
in
g
s
cik
it
-
lear
n
,
XGBo
o
s
t,
an
d
p
an
d
as.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
B
o
o
s
tin
g
I
I
o
T in
tr
u
s
io
n
d
etec
t
io
n
:
leve
r
a
g
in
g
ma
ch
in
e
lea
r
n
in
g
a
n
d
fea
tu
r
e
s
elec
tio
n
…
(
L
a
h
ce
n
I
d
o
u
g
lid
)
1237
Fig
u
r
e
1
.
Me
th
o
d
o
lo
g
y
wo
r
k
f
l
o
w
o
v
er
v
iew
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
4
.
1
.
Resul
t
s
T
h
e
p
er
f
o
r
m
an
ce
o
f
f
iv
e
ML
m
o
d
els
—
r
an
d
o
m
f
o
r
est
,
XGBo
o
s
t,
Ad
aBo
o
s
t,
g
r
ad
ien
t
b
o
o
s
tin
g
,
an
d
MLP
-
was
ev
alu
ated
f
o
r
in
tr
u
s
io
n
d
etec
tio
n
i
n
I
I
o
T
en
v
ir
o
n
m
e
n
ts
.
T
h
e
m
o
d
els
wer
e
ass
ess
ed
b
ased
o
n
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
F1
s
co
r
e
,
an
d
AUC
-
R
O
C
.
T
h
e
r
esu
lts
d
em
o
n
s
tr
ate
th
at
R
an
d
o
m
Fo
r
est
an
d
XGBo
o
s
t
wer
e
th
e
to
p
p
e
r
f
o
r
m
er
s
,
wh
ile
ML
P lag
g
ed
b
e
h
in
d
o
th
er
m
o
d
els.
4
.
1
.
1
.
Co
m
pa
riso
n o
f
mo
del per
f
o
rm
a
nce
T
ab
les
1
a
n
d
2
s
u
m
m
ar
ize
t
h
e
m
o
d
el
ev
alu
atio
n
m
et
r
ics,
s
h
o
win
g
th
e
d
etailed
p
er
f
o
r
m
a
n
ce
o
f
ea
ch
class
if
ier
b
ef
o
r
e
an
d
af
ter
f
ea
tu
r
e
s
elec
tio
n
.
R
an
d
o
m
f
o
r
est
co
n
s
is
ten
tly
o
u
tp
e
r
f
o
r
m
e
d
o
th
e
r
class
if
ier
s
,
ac
h
iev
in
g
th
e
h
ig
h
est
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
an
d
F1
s
c
o
r
es,
with
s
ig
n
if
ican
t
im
p
r
o
v
e
m
en
ts
af
ter
f
ea
tu
r
e
s
elec
tio
n
(
ac
cu
r
ac
y
in
cr
ea
s
ed
f
r
o
m
9
4
.
5
%
to
9
6
.
8
%,
an
d
F1
s
co
r
e
f
r
o
m
9
4
.
7
%
to
9
6
.
7
%).
XGBo
o
s
t
a
ls
o
p
er
f
o
r
m
ed
ex
ce
p
tio
n
ally
well,
with
ac
cu
r
ac
y
r
is
in
g
f
r
o
m
9
2
.
2
%
to
9
5
.
1
%
an
d
F1
s
co
r
e
f
r
o
m
9
1
.
9
%
to
9
5
.
2
%
af
ter
f
ea
tu
r
e
s
elec
tio
n
.
Ad
aBo
o
s
t
an
d
g
r
ad
ien
t
b
o
o
s
tin
g
s
h
o
wed
s
o
lid
p
er
f
o
r
m
an
ce
,
im
p
r
o
v
in
g
in
ac
cu
r
ac
y
an
d
p
r
ec
is
io
n
af
ter
f
ea
tu
r
e
s
elec
tio
n
,
b
u
t
s
lig
h
tly
tr
ailin
g
b
e
h
in
d
r
a
n
d
o
m
f
o
r
est
an
d
XGB
o
o
s
t.
ML
P,
wh
ile
im
p
r
o
v
in
g
f
r
o
m
8
6
.
7
%
to
9
1
.
5
%
in
ac
cu
r
ac
y
p
o
s
t
-
f
ea
tu
r
e
s
elec
tio
n
,
s
till
lag
g
ed
b
eh
in
d
t
h
e
o
th
e
r
m
o
d
els
in
o
v
er
all
p
er
f
o
r
m
a
n
ce
.
T
ab
le
1.
Per
f
o
r
m
an
ce
m
etr
ics o
f
class
if
ier
s
b
ef
o
r
e
an
d
af
ter
f
ea
tu
r
e
s
elec
tio
n
C
l
a
s
si
f
i
e
r
A
c
c
u
r
a
c
y
b
e
f
o
r
e
(
%)
A
c
c
u
r
a
c
y
a
f
t
e
r
(
%)
P
r
e
c
i
s
i
o
n
b
e
f
o
r
e
(
%)
P
r
e
c
i
s
i
o
n
a
f
t
e
r
(
%)
R
e
c
a
l
l
b
e
f
o
r
e
(
%)
R
e
c
a
l
l
a
f
t
e
r
(
%)
F
1
s
c
o
r
e
b
e
f
o
r
e
(
%)
F
1
s
c
o
r
e
a
f
t
e
r
(
%)
R
a
n
d
o
m
f
o
r
e
s
t
9
4
.
5
9
6
.
8
9
5
.
8
9
7
.
3
9
3
.
6
9
6
.
2
9
4
.
7
9
6
.
7
X
G
B
o
o
st
9
2
.
2
9
5
.
1
9
3
.
5
9
6
.
1
9
0
.
4
9
4
.
3
9
1
.
9
9
5
.
2
A
d
a
B
o
o
st
8
9
.
3
9
2
.
4
9
0
.
2
94
88
9
1
.
8
8
9
.
1
9
2
.
9
G
r
a
d
i
e
n
t
b
o
o
s
t
i
n
g
9
0
.
5
9
3
.
6
9
1
.
7
9
4
.
8
8
9
.
1
9
2
.
3
9
0
.
4
9
3
.
5
M
LP
8
6
.
7
9
1
.
5
8
8
.
9
9
3
.
4
8
5
.
6
9
0
.
8
8
7
.
2
9
2
.
1
T
ab
le
2
.
Hy
p
o
th
etica
l A
UC
-
R
OC
tab
le
f
o
r
class
if
ier
s
C
l
a
s
si
f
i
e
r
AUC
-
R
O
C
b
e
f
o
r
e
AUC
-
R
O
C
a
f
t
e
r
R
a
n
d
o
m f
o
r
e
s
t
0
.
9
6
0
.
9
7
X
G
B
o
o
st
0
.
9
4
0
.
9
7
A
d
a
B
o
o
st
0
.
8
9
0
.
9
3
G
r
a
d
i
e
n
t
b
o
o
s
t
i
n
g
0
.
9
1
0
.
9
2
M
LP
0
.
8
7
0
.
9
1
T
h
is
d
etailed
an
al
y
s
is
s
h
o
ws
th
at
r
an
d
o
m
f
o
r
est
a
n
d
XGBo
o
s
t
ar
e
th
e
m
o
s
t
ef
f
ec
tiv
e
class
if
ier
s
in
th
is
s
tu
d
y
,
b
o
th
s
ig
n
i
f
ican
tly
b
en
ef
itin
g
f
r
o
m
f
ea
t
u
r
e
s
elec
tio
n
.
A
d
aBo
o
s
t a
n
d
g
r
ad
ie
n
t b
o
o
s
tin
g
also
p
er
f
o
r
m
well,
b
u
t
to
a
s
lig
h
tly
less
er
ex
ten
t,
wh
ile
ML
P,
d
esp
ite
im
p
r
o
v
e
m
en
ts
,
r
em
ain
s
th
e
least
ef
f
ec
t
iv
e
o
f
th
e
class
if
ier
s
test
ed
.
T
h
e
b
ar
ch
ar
t
b
el
o
w
p
r
o
v
id
es
a
d
etailed
co
m
p
a
r
is
o
n
o
f
class
if
ier
p
er
f
o
r
m
a
n
ce
ac
r
o
s
s
k
ey
m
etr
ics
b
ef
o
r
e
an
d
af
ter
f
ea
tu
r
e
s
elec
tio
n
.
As
s
h
o
wn
in
Fig
u
r
e
2
,
a
d
etailed
co
m
p
ar
is
o
n
o
f
class
if
ier
p
er
f
o
r
m
an
ce
b
ef
o
r
e
an
d
af
ter
f
ea
tu
r
e
s
elec
tio
n
u
n
d
er
s
c
o
r
es
th
e
s
ig
n
if
ican
t
im
p
ac
t
o
f
f
ea
tu
r
e
s
elec
tio
n
o
n
im
p
r
o
v
in
g
m
o
d
el
p
er
f
o
r
m
an
ce
.
T
h
e
r
esu
lts
clea
r
ly
in
d
icate
th
at
r
an
d
o
m
f
o
r
est
a
n
d
XGBo
o
s
t
co
n
s
is
ten
tly
o
u
tp
e
r
f
o
r
m
o
th
er
class
if
ier
s
,
ac
h
iev
in
g
th
e
h
ig
h
est
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
an
d
F1
s
c
o
r
es
in
b
o
th
p
r
e
-
a
n
d
p
o
s
t
-
f
ea
t
u
r
e
s
elec
tio
n
s
tag
es.
T
h
ese
f
in
d
in
g
s
h
i
g
h
lig
h
t t
h
eir
r
o
b
u
s
t su
itab
ilit
y
f
o
r
in
tr
u
s
io
n
d
etec
tio
n
in
I
I
o
T
en
v
ir
o
n
m
en
t
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
4
,
No
.
2
,
Ap
r
il 2
0
2
5
:
1
2
3
2
-
1
2
4
1
1238
Featu
r
e
s
elec
tio
n
h
a
d
a
p
r
o
n
o
u
n
ce
d
ef
f
ec
t o
n
all
class
if
ier
s
.
Fo
r
in
s
tan
ce
,
ML
P d
em
o
n
s
tr
ated
n
o
tab
le
im
p
r
o
v
em
e
n
ts
in
all
m
etr
ics,
p
ar
ticu
lar
ly
in
ac
cu
r
ac
y
a
n
d
F1
s
co
r
e,
af
ter
ap
p
ly
in
g
f
ea
tu
r
e
s
e
lectio
n
tech
n
iq
u
es.
E
v
en
th
e
to
p
-
p
er
f
o
r
m
in
g
class
if
ier
s
,
s
u
ch
as
r
an
d
o
m
f
o
r
est
a
n
d
XGBo
o
s
t,
ex
h
ib
ited
o
b
s
er
v
ab
le
en
h
a
n
ce
m
en
ts
in
p
r
ec
is
io
n
an
d
r
ec
all,
f
u
r
th
er
r
ef
in
in
g
th
eir
ef
f
ec
ti
v
en
ess
in
in
tr
u
s
io
n
d
etec
tio
n
.
Fig
u
r
e
2
also
r
ev
ea
ls
in
ter
es
tin
g
tr
en
d
s
an
d
an
o
m
alies.
W
h
ile
cla
s
s
if
ier
s
s
u
ch
as
Ad
aBo
o
s
t
an
d
g
r
ad
ien
t
b
o
o
s
tin
g
d
is
p
lay
e
d
s
o
lid
im
p
r
o
v
em
e
n
ts
,
th
eir
p
er
f
o
r
m
an
ce
g
ain
s
we
r
e
less
p
r
o
n
o
u
n
ce
d
th
an
th
o
s
e
o
f
r
an
d
o
m
f
o
r
est
an
d
XGBo
o
s
t,
s
u
g
g
esti
n
g
t
h
at
th
ese
m
o
d
el
s
m
ay
b
e
less
s
en
s
itiv
e
to
th
e
ad
v
a
n
tag
es
p
r
o
v
id
ed
b
y
f
ea
tu
r
e
s
elec
tio
n
.
I
n
ca
s
es
wh
er
e
s
lig
h
t
d
ec
r
ea
s
es
in
ce
r
tain
m
etr
ics
wer
e
o
b
s
er
v
e
d
,
th
ese
an
o
m
alies
m
er
i
t
f
u
r
th
er
in
v
esti
g
atio
n
to
u
n
co
v
er
u
n
d
er
ly
in
g
ca
u
s
es.
T
h
e
v
is
u
al
co
m
p
ar
is
o
n
p
r
o
v
id
ed
in
Fig
u
r
e
2
co
m
p
r
eh
e
n
s
iv
ely
illu
s
tr
ates
th
e
im
p
ac
t
o
f
f
ea
tu
r
e
s
elec
tio
n
o
n
m
etr
ics
s
u
ch
as
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
an
d
F1
s
co
r
e.
T
h
is
g
r
ap
h
ical
r
ep
r
e
s
en
tatio
n
en
ab
les
s
tak
eh
o
ld
e
r
s
to
ev
alu
ate
a
n
d
s
elec
t
th
e
m
o
s
t
ef
f
ec
tiv
e
m
o
d
els
f
o
r
I
I
o
T
s
ec
u
r
ity
ap
p
licatio
n
s
.
Fu
r
th
er
m
o
r
e
,
ex
te
n
d
in
g
th
is
a
n
aly
s
is
to
in
clu
d
e
AUC
-
R
OC
cu
r
v
es
co
u
l
d
en
h
an
ce
th
e
v
alid
atio
n
o
f
class
if
ier
p
er
f
o
r
m
an
ce
.
Fig
u
r
e
2
.
C
o
m
p
a
r
is
o
n
o
f
class
if
ier
p
er
f
o
r
m
an
ce
b
ef
o
r
e
an
d
a
f
ter
f
ea
tu
r
e
s
elec
tio
n
T
h
e
T
ab
le
2
p
r
o
v
id
es
a
d
etail
ed
co
m
p
ar
is
o
n
o
f
th
e
AUC
-
R
OC
v
alu
es
f
o
r
ea
ch
class
if
ier
b
ef
o
r
e
an
d
af
ter
f
ea
tu
r
e
s
elec
tio
n
.
R
an
d
o
m
f
o
r
est
an
d
XGBo
o
s
t
em
er
g
ed
as
th
e
s
tr
o
n
g
est
p
er
f
o
r
m
e
r
s
,
with
th
eir
AUC
-
R
OC
v
alu
es
im
p
r
o
v
in
g
f
r
o
m
0
.
9
6
to
0
.
9
8
an
d
0
.
9
4
to
0
.
9
7
,
r
e
s
p
ec
tiv
ely
,
f
o
llo
win
g
f
ea
t
u
r
e
s
elec
tio
n
.
Ad
aBo
o
s
t
an
d
g
r
a
d
ien
t
b
o
o
s
tin
g
also
s
h
o
wed
n
o
tab
le
e
n
h
an
ce
m
en
ts
,
th
o
u
g
h
th
eir
p
e
r
f
o
r
m
an
ce
r
em
ain
ed
s
lig
h
tly
lo
we
r
th
an
th
e
to
p
class
if
ier
s
.
ML
P,
wh
ich
in
itially
h
ad
th
e
lo
west
AUC
-
R
O
C
v
alu
es,
s
aw
a
s
ig
n
i
f
ican
t
in
c
r
ea
s
e
af
ter
f
ea
tu
r
e
s
elec
tio
n
,
h
ig
h
lig
h
tin
g
th
e
p
o
s
itiv
e
im
p
ac
t o
f
th
is
p
r
o
ce
s
s
o
n
m
o
d
el
p
er
f
o
r
m
an
ce
.
T
h
e
lin
e
ch
ar
t b
elo
w
p
r
o
v
id
es a
d
etailed
co
m
p
ar
is
o
n
o
f
th
e
AUC
-
R
OC
v
alu
es f
o
r
ea
ch
class
if
ier
b
ef
o
r
e
a
n
d
af
ter
f
ea
tu
r
e
s
elec
tio
n
.
Fig
u
r
e
3
illu
s
tr
ates
th
e
co
m
p
ar
is
o
n
o
f
AUC
-
R
OC
v
alu
es
f
o
r
v
ar
io
u
s
class
if
ier
s
b
ef
o
r
e
an
d
af
ter
f
ea
tu
r
e
s
elec
tio
n
,
p
r
o
v
id
in
g
a
clea
r
v
i
s
u
al
r
ep
r
esen
tatio
n
o
f
th
e
p
er
f
o
r
m
an
ce
im
p
r
o
v
em
en
ts
.
T
h
e
l
in
e
ch
ar
t
h
ig
h
lig
h
ts
th
e
ef
f
ec
tiv
en
ess
o
f
f
ea
tu
r
e
s
elec
tio
n
in
e
n
h
an
cin
g
m
o
d
el
p
er
f
o
r
m
an
ce
ac
r
o
s
s
all
clas
s
if
ier
s
.
No
tab
ly
,
r
an
d
o
m
f
o
r
est
an
d
XGBo
o
s
t
s
h
o
wed
th
e
m
o
s
t
s
ig
n
if
ican
t
g
ain
s
,
as
ev
id
en
ce
d
b
y
th
e
s
h
ar
p
u
p
war
d
tr
en
d
i
n
th
e
"AUC
-
R
OC
a
f
ter
"
lin
e.
T
h
is
r
esu
lt
r
ea
f
f
ir
m
s
th
e
r
o
b
u
s
tn
ess
o
f
th
ese
m
o
d
els
in
d
etec
tin
g
in
tr
u
s
io
n
s
in
I
I
o
T
en
v
ir
o
n
m
en
ts
.
T
h
e
ch
ar
t
also
r
ev
ea
ls
th
at
w
h
ile
Ad
aBo
o
s
t
an
d
g
r
ad
ien
t
b
o
o
s
tin
g
d
em
o
n
s
tr
ated
im
p
r
o
v
em
en
ts
in
AUC
-
R
O
C
v
alu
es,
th
eir
p
er
f
o
r
m
an
ce
g
ain
s
wer
e
m
o
r
e
m
o
d
est
co
m
p
a
r
ed
to
th
e
t
o
p
-
p
er
f
o
r
m
in
g
m
o
d
els.
ML
P
ex
h
ib
ited
a
n
o
ticea
b
le
im
p
r
o
v
e
m
en
t,
wh
ich
,
th
o
u
g
h
s
till
tr
ailin
g
b
eh
in
d
o
th
er
m
o
d
els,
u
n
d
er
s
co
r
es
th
e
v
alu
e
o
f
f
ea
t
u
r
e
s
elec
tio
n
in
en
h
a
n
cin
g
ev
en
less
co
m
p
et
itiv
e
class
if
ier
s
.
Ov
er
all,
th
e
tr
en
d
s
d
ep
icted
in
Fig
u
r
e
3
em
p
h
asize
th
e
cr
itical
r
o
le
o
f
f
ea
t
u
r
e
s
elec
tio
n
in
r
ef
in
in
g
class
if
ier
p
er
f
o
r
m
an
ce
.
B
y
im
p
r
o
v
i
n
g
th
e
AUC
-
R
O
C
v
alu
es
ac
r
o
s
s
m
u
ltip
le
m
o
d
els,
f
ea
tu
r
e
s
elec
tio
n
p
r
o
v
es
to
b
e
an
ess
en
tial
p
r
e
p
r
o
ce
s
s
in
g
s
tep
f
o
r
ac
h
iev
in
g
h
ig
h
er
d
etec
tio
n
ac
c
u
r
ac
y
an
d
r
eliab
ilit
y
in
I
I
o
T
s
ec
u
r
ity
ap
p
licatio
n
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
B
o
o
s
tin
g
I
I
o
T in
tr
u
s
io
n
d
etec
t
io
n
:
leve
r
a
g
in
g
ma
ch
in
e
lea
r
n
in
g
a
n
d
fea
tu
r
e
s
elec
tio
n
…
(
L
a
h
ce
n
I
d
o
u
g
lid
)
1239
Fig
u
r
e
3.
C
o
m
p
a
r
is
o
n
o
f
AUC
-
R
OC
v
alu
es b
ef
o
r
e
an
d
a
f
ter
f
ea
tu
r
e
s
elec
tio
n
4
.
1
.
2
.
Co
m
pa
riso
n wit
h r
ela
t
ed
s
t
ud
ie
s
T
h
e
r
e
s
u
l
t
s
o
f
t
h
i
s
s
t
u
d
y
a
r
e
c
o
n
s
i
s
t
e
n
t
w
i
t
h
e
x
i
s
t
i
n
g
l
i
t
e
r
a
t
u
r
e
,
p
a
r
t
i
c
u
l
a
r
l
y
i
n
h
i
g
h
l
i
g
h
t
i
n
g
t
h
e
s
u
p
e
r
i
o
r
p
e
r
f
o
r
m
a
n
c
e
o
f
t
r
e
e
-
b
a
s
e
d
m
o
d
e
l
s
f
o
r
i
n
t
r
u
s
i
o
n
d
e
t
e
c
t
i
o
n
.
F
o
r
e
x
a
m
p
l
e
,
i
n
[
9
]
,
r
a
n
d
o
m
f
o
r
e
s
t
w
a
s
s
h
o
w
n
t
o
a
c
h
i
e
v
e
h
i
g
h
a
c
c
u
r
a
c
y
i
n
a
s
i
m
i
l
a
r
I
I
o
T
e
n
v
i
r
o
n
m
e
n
t
,
w
i
t
h
c
o
m
p
a
r
a
b
l
e
i
m
p
r
o
v
e
m
e
n
t
s
w
h
e
n
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
t
e
c
h
n
i
q
u
e
s
w
e
r
e
a
p
p
l
i
e
d
.
A
d
d
i
t
i
o
n
a
l
l
y
,
X
G
B
o
o
s
t
h
a
s
c
o
n
s
i
s
t
e
n
t
l
y
b
e
e
n
h
i
g
h
l
i
g
h
t
e
d
a
s
a
t
o
p
p
e
r
f
o
r
m
e
r
i
n
a
n
o
m
a
l
y
d
e
t
e
c
t
i
o
n
s
t
u
d
i
e
s
,
s
u
c
h
a
s
[
1
3
]
,
w
h
e
r
e
i
t
s
e
n
s
e
m
b
l
e
l
e
a
r
n
i
n
g
m
e
t
h
o
d
a
n
d
b
o
o
s
t
i
n
g
t
e
c
h
n
i
q
u
e
h
e
l
p
e
d
r
e
d
u
c
e
f
a
l
s
e
p
o
s
i
t
i
v
e
s
.
Ah
m
ed
et
a
l.
[
3
0
]
in
v
esti
g
ates
a
h
o
n
ey
p
o
t
-
d
r
iv
en
ap
p
r
o
ac
h
to
s
ec
u
r
in
g
s
m
ar
t
cities
b
y
i
n
teg
r
atin
g
h
o
n
ey
p
o
t
d
ata
with
m
ac
h
i
n
e
le
ar
n
in
g
f
o
r
I
o
T
attac
k
d
etec
tio
n
.
Usi
n
g
h
ig
h
-
i
n
ter
ac
tio
n
h
o
n
ey
p
o
ts
an
d
r
ea
l
-
w
o
r
ld
d
atasets
,
th
e
s
tu
d
y
em
p
lo
y
s
alg
o
r
ith
m
s
lik
e
d
ec
is
io
n
tr
ee
,
n
aïv
e
B
ay
es,
K
-
NN,
s
eq
u
en
tial
n
eu
r
al
n
etwo
r
k
(
SNN)
,
an
d
L
STM
to
class
if
y
attac
k
ty
p
es
s
u
ch
as
p
o
r
t
s
ca
n
n
in
g
,
b
r
u
te
f
o
r
ce
,
an
d
Mir
ai
b
o
t
n
et.
Ach
iev
in
g
h
i
g
h
ac
cu
r
ac
y
ac
r
o
s
s
m
etr
ics,
t
h
is
r
esear
ch
h
ig
h
lig
h
ts
th
e
p
r
ac
tica
l
ap
p
licatio
n
o
f
h
o
n
ey
p
o
ts
f
o
r
ea
r
ly
d
etec
tio
n
a
n
d
m
itig
atio
n
o
f
I
o
T
c
y
b
er
th
r
e
ats,
em
p
h
asizin
g
s
ca
lab
ilit
y
an
d
ef
f
ec
tiv
en
ess
in
s
m
ar
t
city
ec
o
s
y
s
tem
s
.
Al
-
Halb
o
o
s
i
et
a
l.
[
3
1
]
e
x
p
lo
r
e
d
f
ed
er
ate
d
lear
n
in
g
m
o
d
els
with
in
ce
p
tio
n
tim
e
an
d
m
u
lti
-
h
ea
d
atten
tio
n
m
ec
h
an
is
m
s
,
ac
h
iev
in
g
g
lo
b
al
ac
cu
r
ac
ies
o
f
9
3
.
9
1
%
an
d
9
3
.
4
9
%
u
n
d
er
I
I
D
an
d
n
o
n
-
I
I
D
d
is
tr
ib
u
tio
n
s
,
r
esp
ec
tiv
ely
.
T
h
o
u
g
h
th
is
s
tu
d
y
d
o
es
n
o
t
u
s
e
f
ed
e
r
ated
lear
n
in
g
,
its
f
o
cu
s
o
n
r
ea
l
-
tim
e
ap
p
li
ca
b
ilit
y
alig
n
s
with
th
e
u
s
e
o
f
d
i
m
en
s
io
n
ality
r
e
d
u
ctio
n
to
o
p
tim
ize
m
o
d
el
p
er
f
o
r
m
an
ce
f
o
r
I
I
o
T
en
v
i
r
o
n
m
e
n
ts
.
C
o
m
p
ar
ed
to
p
r
ev
io
u
s
r
esear
c
h
[
1
3
]
,
wh
ich
also
u
s
ed
tr
ee
-
b
ased
m
o
d
els
f
o
r
I
I
o
T
in
tr
u
s
io
n
d
etec
tio
n
,
th
is
s
tu
d
y
ac
h
iev
ed
s
im
ilar
p
er
f
o
r
m
an
ce
m
etr
ics,
p
ar
ticu
lar
l
y
in
p
r
ec
is
io
n
a
n
d
r
ec
all.
Ho
wev
er
,
th
e
in
clu
s
io
n
o
f
PC
A
in
th
i
s
s
tu
d
y
as
a
d
im
e
n
s
io
n
ality
r
ed
u
ctio
n
tec
h
n
iq
u
e
p
r
o
v
id
e
d
an
ad
d
itio
n
al
b
o
o
s
t
in
co
m
p
u
tatio
n
a
l
ef
f
icien
cy
an
d
ac
cu
r
ac
y
,
m
a
k
in
g
th
is
ap
p
r
o
ac
h
m
o
r
e
s
u
itab
l
e
f
o
r
r
ea
l
-
tim
e
I
I
o
T
ap
p
licatio
n
s
.
W
h
ile
th
e
ML
P
m
o
d
el
s
h
o
wed
s
o
m
e
im
p
r
o
v
e
m
en
t
af
ter
f
ea
t
u
r
e
s
elec
tio
n
,
it
s
till
lag
g
ed
b
eh
in
d
o
th
er
m
o
d
e
ls
.
Similar
tr
en
d
s
wer
e
o
b
s
er
v
e
d
in
[
1
0
]
,
wh
er
e
ML
P
m
o
d
els,
d
esp
ite
th
eir
f
lex
ib
ilit
y
,
s
tr
u
g
g
led
with
th
e
h
i
g
h
-
d
im
en
s
io
n
ality
a
n
d
co
m
p
lex
ity
o
f
I
I
o
T
d
atasets
.
4
.
2
.
Dis
cus
s
io
n
4
.
2
.
1
.
K
ey
f
ind
ing
s
T
h
e
an
aly
s
is
h
ig
h
lig
h
ts
th
e
ef
f
ec
tiv
en
ess
o
f
r
an
d
o
m
f
o
r
est
an
d
XGBo
o
s
t
a
s
th
e
m
o
s
t
r
o
b
u
s
t
class
if
ier
s
f
o
r
I
I
o
T
in
tr
u
s
io
n
d
etec
tio
n
,
p
ar
ticu
lar
ly
af
ter
ap
p
ly
i
n
g
f
ea
tu
r
e
s
elec
tio
n
tech
n
iq
u
es.
Featu
r
e
s
elec
tio
n
,
s
p
ec
if
ically
PC
A
an
d
R
FE,
s
i
g
n
if
ican
tly
im
p
r
o
v
ed
t
h
e
p
er
f
o
r
m
an
ce
o
f
all
m
o
d
els
b
y
f
o
c
u
s
in
g
o
n
th
e
m
o
s
t
im
p
o
r
tan
t
f
ea
t
u
r
es
an
d
r
ed
u
ci
n
g
d
im
e
n
s
io
n
ality
.
T
h
e
r
esu
lts
also
d
em
o
n
s
tr
ate
th
at
Ad
a
B
o
o
s
t
an
d
g
r
ad
ie
n
t
b
o
o
s
tin
g
a
r
e
s
tr
o
n
g
alter
n
ativ
es,
alth
o
u
g
h
th
ey
d
id
n
o
t
o
u
tp
er
f
o
r
m
r
an
d
o
m
f
o
r
est
an
d
XG
B
o
o
s
t.
ML
P,
wh
ile
im
p
r
o
v
in
g
af
ter
f
ea
tu
r
e
s
elec
t
io
n
,
r
em
ain
s
less
co
m
p
etitiv
e
i
n
h
an
d
lin
g
th
e
co
m
p
le
x
ity
o
f
I
I
o
T
d
atasets
,
a
f
in
d
in
g
t
h
at
is
co
n
s
is
ten
t w
ith
o
th
er
r
esear
ch
s
tu
d
ies.
4
.
2
.
2
.
L
im
it
a
t
io
ns
Desp
ite
th
e
p
r
o
m
is
in
g
r
esu
lts
,
th
is
s
tu
d
y
h
as
s
ev
er
al
l
im
itati
o
n
s
.
First,
th
e
f
in
d
in
g
s
ar
e
b
ased
o
n
th
e
UNSW
-
N
B
1
5
d
ataset,
wh
ich
,
wh
ile
wid
ely
u
s
ed
,
d
o
es
n
o
t
r
ep
r
esen
t
all
p
o
s
s
ib
le
I
I
o
T
a
ttack
ty
p
es.
Fu
tu
r
e
r
esear
ch
s
h
o
u
l
d
ex
p
lo
r
e
m
o
r
e
d
iv
er
s
e
d
atasets
,
s
u
ch
as
T
o
N
_
I
o
T
o
r
B
o
T
-
I
o
T
,
to
v
alid
ate
t
h
e
g
en
e
r
aliza
b
ilit
y
o
f
th
ese
r
esu
lts
.
Seco
n
d
,
th
e
e
x
p
er
im
en
ts
wer
e
co
n
d
u
cted
u
s
in
g
th
e
Kag
g
le
VM
,
wh
ich
m
a
y
n
o
t f
u
lly
r
e
p
licate
th
e
r
eso
u
r
ce
c
o
n
s
tr
ain
ts
o
f
r
ea
l
I
I
o
T
e
n
v
ir
o
n
m
en
ts
.
T
esti
n
g
i
n
lo
w
-
laten
cy
,
r
eso
u
r
ce
-
c
o
n
s
tr
a
in
ed
s
ettin
g
s
co
u
ld
y
ield
d
if
f
er
en
t
p
er
f
o
r
m
an
ce
o
u
tco
m
es.
T
h
ir
d
,
th
e
s
tu
d
y
p
r
i
m
ar
ily
f
o
cu
s
ed
o
n
class
ical
ML
m
o
d
els,
leav
in
g
r
o
o
m
f
o
r
f
u
r
th
er
ex
p
lo
r
atio
n
o
f
d
ee
p
lear
n
in
g
tech
n
i
q
u
es
lik
e
C
NNs
an
d
L
STM
s
,
w
h
ich
m
a
y
en
h
an
ce
an
o
m
aly
d
etec
tio
n
,
p
ar
ticu
lar
ly
in
lar
g
e
r
d
atasets
.
L
ast
ly
,
wh
ile
P
C
A
r
ed
u
ce
d
co
m
p
u
tatio
n
al
co
s
ts
,
f
u
r
th
er
o
p
tim
izatio
n
is
n
ee
d
ed
to
en
s
u
r
e
th
a
t th
ese
m
o
d
els ca
n
f
u
n
ctio
n
e
f
f
ec
tiv
el
y
in
r
ea
l
-
tim
e
I
I
o
T
en
v
i
r
o
n
m
e
n
ts
.
Fu
tu
r
e
r
esear
ch
s
h
o
u
ld
f
o
c
u
s
o
n
d
ev
elo
p
in
g
li
g
h
tweig
h
t,
ad
a
p
tiv
e
m
o
d
els s
u
ited
f
o
r
r
ea
l
-
tim
e
ap
p
licatio
n
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
4
,
No
.
2
,
Ap
r
il 2
0
2
5
:
1
2
3
2
-
1
2
4
1
1240
4
.
2
.
3
.
F
uture
re
s
ea
rc
h direc
t
io
ns
Fu
tu
r
e
r
esear
ch
co
u
ld
e
x
p
an
d
o
n
s
ev
er
al
ar
ea
s
b
ased
o
n
th
is
s
tu
d
y
'
s
f
in
d
in
g
s
.
First,
in
te
g
r
atin
g
d
ee
p
lear
n
in
g
m
o
d
els
lik
e
C
NNs
an
d
L
STM
with
e
n
s
em
b
le
lear
n
in
g
m
et
h
o
d
s
c
o
u
ld
en
h
a
n
ce
in
t
r
u
s
io
n
d
etec
tio
n
in
I
I
o
T
e
n
v
ir
o
n
m
en
ts
.
T
h
ese
m
o
d
els
ar
e
b
etter
s
u
ited
f
o
r
ca
p
tu
r
in
g
tem
p
o
r
al
an
d
s
p
atial
d
ata
r
elatio
n
s
h
ip
s
,
lead
in
g
to
im
p
r
o
v
e
d
an
o
m
aly
d
etec
tio
n
.
Seco
n
d
,
o
p
tim
izin
g
ML
m
o
d
els
f
o
r
r
ea
l
-
tim
e
d
ep
l
o
y
m
en
t
is
ess
en
tial
f
o
r
I
I
o
T
,
wh
e
r
e
lo
w
laten
cy
a
n
d
lim
ited
r
eso
u
r
ce
s
ar
e
cr
itic
al.
Dev
elo
p
in
g
lig
h
tweig
h
t
m
o
d
els
t
h
at
m
ain
tain
h
ig
h
ac
cu
r
ac
y
wh
ile
m
in
im
izin
g
co
m
p
u
tatio
n
al
d
em
an
d
s
s
h
o
u
ld
b
e
a
k
ey
f
o
cu
s
.
L
astl
y
,
ex
p
an
d
in
g
r
esear
ch
to
in
clu
d
e
a
d
d
itio
n
al
d
atasets
,
s
u
ch
as
T
o
N
_
I
o
T
an
d
B
o
T
-
I
o
T
,
will
p
r
o
v
id
e
a
b
r
o
ad
er
u
n
d
er
s
tan
d
in
g
o
f
m
o
d
el
p
er
f
o
r
m
an
ce
ac
r
o
s
s
d
if
f
er
en
t
I
I
o
T
en
v
ir
o
n
m
e
n
ts
an
d
attac
k
t
y
p
es.
5.
CO
NCLU
SI
O
N
T
h
e
r
esu
lts
o
f
th
is
s
tu
d
y
d
e
m
o
n
s
tr
ate
th
e
ef
f
ec
tiv
en
ess
o
f
ML
alg
o
r
ith
m
s
,
p
ar
ticu
lar
ly
tr
ee
-
b
ase
d
m
o
d
els
s
u
ch
as
r
an
d
o
m
f
o
r
est
an
d
XGBo
o
s
t,
f
o
r
in
tr
u
s
io
n
d
etec
tio
n
in
I
I
o
T
en
v
ir
o
n
m
e
n
ts
.
R
an
d
o
m
f
o
r
est
ac
h
iev
ed
th
e
h
ig
h
est
ac
c
u
r
ac
y
,
b
e
n
ef
itin
g
s
ig
n
if
ican
tly
f
r
o
m
th
e
f
ea
tu
r
e
s
elec
tio
n
tec
h
n
iq
u
es
e
m
p
lo
y
e
d
.
XGBo
o
s
t
al
s
o
p
er
f
o
r
m
e
d
well,
o
f
f
er
in
g
a
b
ala
n
ce
b
etwe
en
ac
cu
r
ac
y
an
d
c
o
m
p
u
tatio
n
al
ef
f
icien
cy
.
T
h
e
in
clu
s
io
n
o
f
PC
A
an
d
R
F
E
co
n
tr
ib
u
ted
to
th
e
r
ed
u
ctio
n
o
f
d
im
en
s
io
n
ality
,
im
p
r
o
v
in
g
b
o
th
ac
cu
r
ac
y
an
d
co
m
p
u
tatio
n
al
p
e
r
f
o
r
m
an
ce
.
Ad
aBo
o
s
t
an
d
g
r
ad
ien
t
b
o
o
s
tin
g
m
o
d
els
s
h
o
wed
co
m
p
et
itiv
e
p
er
f
o
r
m
an
ce
,
alth
o
u
g
h
th
ey
d
id
n
o
t
o
u
tp
e
r
f
o
r
m
r
an
d
o
m
f
o
r
est
an
d
XG
B
o
o
s
t.
T
h
e
ML
P
im
p
r
o
v
e
d
s
lig
h
tly
af
ter
f
ea
tu
r
e
s
elec
tio
n
b
u
t
r
em
ain
ed
less
co
m
p
etitiv
e
d
u
e
t
o
its
ch
allen
g
es
with
h
a
n
d
lin
g
h
ig
h
-
d
im
e
n
s
io
n
al
d
ata
in
I
I
o
T
en
v
ir
o
n
m
en
ts
.
T
h
ese
f
in
d
i
n
g
s
p
r
o
v
id
e
v
alu
ab
le
in
s
ig
h
ts
in
to
th
e
d
ev
elo
p
m
en
t
o
f
m
o
r
e
ef
f
icien
t
an
d
s
ca
lab
le
I
DS
f
o
r
I
I
o
T
a
p
p
licatio
n
s
.
Fu
tu
r
e
r
esear
ch
co
u
ld
f
u
r
th
er
en
h
an
ce
th
ese
m
o
d
els
b
y
i
n
teg
r
atin
g
d
ee
p
lear
n
i
n
g
tech
n
iq
u
es,
s
u
ch
as
C
NN
o
r
L
STM
,
o
r
ex
p
lo
r
in
g
f
ed
er
ated
lear
n
in
g
a
p
p
r
o
ac
h
es
to
im
p
r
o
v
e
r
ea
l
-
tim
e
p
er
f
o
r
m
an
ce
an
d
ad
a
p
tab
ilit
y
i
n
r
eso
u
r
ce
-
co
n
s
tr
ain
ed
en
v
ir
o
n
m
en
ts
.
RE
F
E
R
E
NC
E
S
[
1
]
M
.
S
o
o
r
i
,
B
.
A
r
e
z
o
o
,
a
n
d
R
.
D
a
s
t
r
e
s
,
“
I
n
t
e
r
n
e
t
o
f
t
h
i
n
g
s
f
o
r
sm
a
r
t
f
a
c
t
o
r
i
e
s
i
n
i
n
d
u
s
t
r
y
4
.
0
,
a
r
e
v
i
e
w
,
”
I
n
t
e
r
n
e
t
o
f
T
h
i
n
g
s
a
n
d
C
y
b
e
r
-
Ph
y
s
i
c
a
l
S
y
st
e
m
s
,
v
o
l
.
3
,
p
p
.
1
9
2
–
2
0
4
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
o
t
c
p
s.
2
0
2
3
.
0
4
.
0
0
6
.
[
2
]
A
.
A
l
t
h
a
b
a
t
a
h
,
M
.
Y
a
q
o
t
,
B
.
M
e
n
e
z
e
s
,
a
n
d
L.
K
e
r
b
a
c
h
e
,
“
Tr
a
n
sf
o
r
mat
i
v
e
p
r
o
c
u
r
e
me
n
t
t
r
e
n
d
s:
i
n
t
e
g
r
a
t
i
n
g
i
n
d
u
s
t
r
y
4
.
0
t
e
c
h
n
o
l
o
g
i
e
s
f
o
r
e
n
h
a
n
c
e
d
p
r
o
c
u
r
e
me
n
t
p
r
o
c
e
sses,
”
L
o
g
i
s
t
i
c
s
,
v
o
l
.
7
,
n
o
.
3
,
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
l
o
g
i
s
t
i
c
s
7
0
3
0
0
6
3
.
[
3
]
E.
A
v
d
i
b
a
s
i
c
,
A
.
S
.
T
o
k
sa
n
o
v
n
a
,
a
n
d
B
.
D
u
r
a
k
o
v
i
c
,
“
C
y
b
e
r
s
e
c
u
r
i
t
y
c
h
a
l
l
e
n
g
e
s i
n
I
n
d
u
s
t
r
y
4
.
0
:
A
st
a
t
e
o
f
t
h
e
a
r
t
r
e
v
i
e
w
,
”
D
e
f
e
n
s
e
a
n
d
S
e
c
u
ri
t
y
S
t
u
d
i
e
s
,
v
o
l
.
3
,
p
p
.
3
2
–
4
9
,
2
0
2
2
,
d
o
i
:
1
0
.
3
7
8
6
8
/
d
ss
.
v
3
.
i
d
1
8
8
.
[
4
]
F
.
M
e
n
e
g
h
e
l
l
o
,
M
.
C
a
l
o
r
e
,
D
.
Z
u
c
c
h
e
t
t
o
,
M
.
P
o
l
e
s
e
,
a
n
d
A
.
Z
a
n
e
l
l
a
,
“
I
o
T
:
I
n
t
e
r
n
e
t
o
f
t
h
r
e
a
t
s
?
a
s
u
r
v
e
y
o
f
p
r
a
c
t
i
c
a
l
s
e
c
u
r
i
t
y
v
u
l
n
e
r
a
b
i
l
i
t
i
e
s
i
n
r
e
a
l
I
o
T
d
e
v
i
c
e
s
,
”
I
E
E
E
I
n
t
e
r
n
e
t
o
f
T
h
i
n
g
s
J
o
u
r
n
a
l
,
v
o
l
.
6
,
n
o
.
5
,
p
p
.
8
1
8
2
–
8
2
0
1
,
2
0
1
9
,
d
o
i
:
1
0
.
1
1
0
9
/
J
I
O
T
.
2
0
1
9
.
2
9
3
5
1
8
9
.
[
5
]
R
.
H
o
o
d
a
,
A
.
K
u
m
a
r
,
S
h
i
v
a
n
i
,
S
u
d
h
i
r
,
P
o
o
j
a
,
a
n
d
P
a
r
t
i
b
h
a
Y
a
d
a
v
,
“
I
n
d
u
s
t
r
i
a
l
i
n
t
e
r
n
e
t
o
f
t
h
i
n
g
s:
a
n
a
n
a
l
y
s
i
s
o
f
e
mer
g
e
n
c
e
,
c
o
m
p
o
n
e
n
t
a
n
d
c
h
a
l
l
e
n
g
e
s
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
n
R
e
c
e
n
t
a
n
d
I
n
n
o
v
a
t
i
o
n
T
re
n
d
s
i
n
C
o
m
p
u
t
i
n
g
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
,
v
o
l
.
1
1
,
n
o
.
1
0
,
p
p
.
2
0
1
0
–
2
0
1
7
,
2
0
2
3
,
d
o
i
:
1
0
.
1
7
7
6
2
/
i
j
r
i
t
c
c
.
v
1
1
i
1
0
.
8
8
8
4
.
[
6
]
M
.
S
o
o
r
i
,
B
.
A
r
e
z
o
o
,
a
n
d
R
.
D
a
st
r
e
s
,
“
V
i
r
t
u
a
l
ma
n
u
f
a
c
t
u
r
i
n
g
i
n
I
n
d
u
s
t
r
y
4
.
0
:
A
r
e
v
i
e
w
,
”
D
a
t
a
S
c
i
e
n
c
e
a
n
d
M
a
n
a
g
e
m
e
n
t
,
v
o
l
.
7
,
n
o
.
1
,
p
p
.
4
7
–
6
3
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
d
sm
.
2
0
2
3
.
1
0
.
0
0
6
.
[
7
]
O
.
F
.
A
w
a
d
,
L.
R
.
H
a
z
i
m
,
A
.
A
.
Jasi
m
,
a
n
d
O
.
A
t
a
,
“
En
h
a
n
c
i
n
g
I
I
o
T
se
c
u
r
i
t
y
w
i
t
h
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
f
o
r
i
n
t
r
u
s
i
o
n
d
e
t
e
c
t
i
o
n
,
”
M
a
l
a
y
s
i
a
n
J
o
u
r
n
a
l
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
3
7
,
n
o
.
2
,
p
p
.
1
3
9
–
1
5
3
,
2
0
2
4
,
d
o
i
:
1
0
.
2
2
4
5
2
/
mj
c
s
.
v
o
l
3
7
n
o
2
.
3
.
[
8
]
B
.
A
l
o
t
a
i
b
i
,
“
A
s
u
r
v
e
y
o
n
i
n
d
u
s
t
r
i
a
l
i
n
t
e
r
n
e
t
o
f
t
h
i
n
g
s
s
e
c
u
r
i
t
y
:
r
e
q
u
i
r
e
me
n
t
s
,
a
t
t
a
c
k
s
,
A
I
-
b
a
se
d
s
o
l
u
t
i
o
n
s,
a
n
d
e
d
g
e
c
o
m
p
u
t
i
n
g
o
p
p
o
r
t
u
n
i
t
i
e
s,
”
S
e
n
s
o
rs
,
v
o
l
.
2
3
,
n
o
.
1
7
,
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
s
2
3
1
7
7
4
7
0
.
[
9
]
H
.
M
l
i
k
i
,
A
.
K
a
c
e
a
m
,
a
n
d
L
.
C
h
a
a
r
i
,
“
A
c
o
m
p
r
e
h
e
n
si
v
e
s
u
r
v
e
y
o
n
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
-
b
a
se
d
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
f
o
r
I
o
T
n
e
t
w
o
r
k
s,”
I
C
S
T
T
r
a
n
sa
c
t
i
o
n
s
o
n
S
e
c
u
r
i
t
y
a
n
d
S
a
f
e
t
y
,
v
o
l
.
8
,
n
o
.
2
9
,
2
0
2
1
,
d
o
i
:
1
0
.
4
1
0
8
/
e
a
i
.
6
-
10
-
2
0
2
1
.
1
7
1
2
4
6
.
[
1
0
]
S
.
S
o
l
i
ma
n
,
W
.
O
u
d
a
h
,
a
n
d
A
.
A
l
j
u
h
a
n
i
,
“
D
e
e
p
l
e
a
r
n
i
n
g
-
b
a
s
e
d
i
n
t
r
u
s
i
o
n
d
e
t
e
c
t
i
o
n
a
p
p
r
o
a
c
h
f
o
r
se
c
u
r
i
n
g
i
n
d
u
st
r
i
a
l
i
n
t
e
r
n
e
t
o
f
t
h
i
n
g
s,
”
A
l
e
x
a
n
d
ri
a
E
n
g
i
n
e
e
ri
n
g
J
o
u
r
n
a
l
,
v
o
l
.
8
1
,
p
p
.
3
7
1
–
3
8
3
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
e
j
.
2
0
2
3
.
0
9
.
0
2
3
.
[
1
1
]
A
.
G
u
e
z
z
a
z
,
M
.
A
z
r
o
u
r
,
S
.
B
e
n
k
i
r
a
n
e
,
M
.
M
o
h
y
-
E
d
d
i
n
e
,
H
.
A
t
t
o
u
,
a
n
d
M
.
D
o
u
i
b
a
,
“
A
l
i
g
h
t
w
e
i
g
h
t
h
y
b
r
i
d
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
f
r
a
mew
o
r
k
u
si
n
g
mac
h
i
n
e
l
e
a
r
n
i
n
g
f
o
r
e
d
g
e
-
b
a
se
d
I
I
o
T
se
c
u
r
i
t
y
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
A
ra
b
J
o
u
r
n
a
l
o
f
I
n
f
o
rm
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
1
9
,
n
o
.
5
,
p
p
.
8
2
2
–
8
3
0
,
2
0
2
2
,
d
o
i
:
1
0
.
3
4
0
2
8
/
i
a
j
i
t
/
1
9
/
5
/
1
4
.
[
1
2
]
M
.
B
a
k
r
o
e
t
a
l
.
,
“
B
u
i
l
d
i
n
g
a
C
l
o
u
d
-
I
D
S
b
y
h
y
b
r
i
d
b
i
o
-
i
n
sp
i
r
e
d
f
e
a
t
u
r
e
se
l
e
c
t
i
o
n
a
l
g
o
r
i
t
h
ms
a
l
o
n
g
w
i
t
h
r
a
n
d
o
m
f
o
r
e
st
mo
d
e
l
,
”
I
E
E
E
Ac
c
e
ss
,
v
o
l
.
1
2
,
p
p
.
8
8
4
6
–
8
8
7
4
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
4
.
3
3
5
3
0
5
5
.
[
1
3
]
J.
B
.
A
w
o
t
u
n
d
e
e
t
a
l
.
,
“
A
n
e
n
sem
b
l
e
t
r
e
e
-
b
a
se
d
mo
d
e
l
f
o
r
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
i
n
i
n
d
u
st
r
i
a
l
i
n
t
e
r
n
e
t
o
f
t
h
i
n
g
s
n
e
t
w
o
r
k
s,
”
Ap
p
l
i
e
d
S
c
i
e
n
c
e
s
,
v
o
l
.
1
3
,
n
o
.
4
,
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
a
p
p
1
3
0
4
2
4
7
9
.
[
1
4
]
K
.
R
a
j
a
sh
e
k
a
r
a
n
,
R
.
K
a
z
m
i
,
a
n
d
R
.
J
a
i
n
,
“
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
-
e
n
h
a
n
c
e
d
I
D
S
:
R
F
E
-
LS
TM
-
b
a
s
e
d
m
o
d
e
l
f
o
r
c
l
o
u
d
s
e
c
u
r
i
t
y
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
C
o
m
p
u
t
e
r
T
r
e
n
d
s
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
7
2
,
n
o
.
4
,
p
p
.
1
–
1
4
,
2
0
2
4
,
d
o
i
:
1
0
.
1
4
4
4
5
/
2
2
3
1
2
8
0
3
/
i
j
c
t
t
-
v
7
2
i
4
p
1
0
1
.
[
1
5
]
S
.
A
l
o
s
a
i
m
i
a
n
d
S
.
M
.
A
l
m
u
t
a
i
r
i
,
“
A
n
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
sy
s
t
e
m
u
s
i
n
g
B
o
T
-
I
o
T,
”
Ap
p
l
i
e
d
S
c
i
e
n
c
e
s
,
v
o
l
.
1
3
,
n
o
.
9
,
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
a
p
p
1
3
0
9
5
4
2
7
.
[
1
6
]
L.
I
d
o
u
g
l
i
d
,
S
.
T
k
a
t
e
k
,
K
.
El
f
a
y
q
,
a
n
d
A
.
G
u
e
z
z
a
z
,
“
N
e
x
t
-
g
e
n
s
e
c
u
r
i
t
y
i
n
I
I
o
T:
i
n
t
e
g
r
a
t
i
n
g
i
n
t
r
u
s
i
o
n
d
e
t
e
c
t
i
o
n
s
y
st
e
ms
w
i
t
h
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
f
o
r
i
n
d
u
s
t
r
y
4
.
0
r
e
si
l
i
e
n
c
e
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
rn
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
r
i
n
g
,
v
o
l
.
1
4
,
n
o
.
3
,
p
p
.
3
5
1
2
–
3
5
2
1
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
c
e
.
v
1
4
i
3
.
p
p
3
5
1
2
-
3
5
2
1
.
[
1
7
]
L.
I
d
o
u
g
l
i
d
,
S
.
Tk
a
t
e
k
,
K
.
E
l
f
a
y
q
,
a
n
d
A
.
G
u
e
z
z
a
z
,
“
A
n
o
v
e
l
a
n
o
m
a
l
y
d
e
t
e
c
t
i
o
n
m
o
d
e
l
f
o
r
t
h
e
i
n
d
u
s
t
r
i
a
l
i
n
t
e
r
n
e
t
o
f
t
h
i
n
g
s
u
s
i
n
g
mac
h
i
n
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s
,
”
R
a
d
i
o
e
l
e
c
t
r
o
n
i
c
a
n
d
C
o
m
p
u
t
e
r
S
y
s
t
e
m
s
,
v
o
l
.
2
0
2
4
,
n
o
.
1
,
p
p
.
1
4
3
–
1
5
1
,
2
0
2
4
,
d
o
i
:
1
0
.
3
2
6
2
0
/
R
E
K
S
.
2
0
2
4
.
1
.
1
2
.
[
1
8
]
W
.
C
h
i
mp
h
l
e
e
a
n
d
S
.
C
h
i
m
p
h
l
e
e
,
“
H
y
p
e
r
p
a
r
a
me
t
e
r
s
o
p
t
i
mi
z
a
t
i
o
n
X
G
B
o
o
s
t
f
o
r
n
e
t
w
o
r
k
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
u
s
i
n
g
C
S
E
-
C
I
C
-
I
D
S
2
0
1
8
d
a
t
a
se
t
,
”
I
AE
S
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
1
3
,
n
o
.
1
,
p
p
.
8
1
7
–
8
2
6
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
a
i
.
v
1
3
.
i
1
.
p
p
8
1
7
-
8
2
6
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
B
o
o
s
tin
g
I
I
o
T in
tr
u
s
io
n
d
etec
t
io
n
:
leve
r
a
g
in
g
ma
ch
in
e
lea
r
n
in
g
a
n
d
fea
tu
r
e
s
elec
tio
n
…
(
L
a
h
ce
n
I
d
o
u
g
lid
)
1241
[
1
9
]
J.
B
r
o
w
n
l
e
e
,
“
H
o
w
t
o
c
h
o
o
s
e
a
f
e
a
t
u
r
e
se
l
e
c
t
i
o
n
me
t
h
o
d
f
o
r
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
,
”
Ma
c
h
i
n
e
L
e
a
r
n
i
n
g
M
a
st
e
r
y
,
v
o
l
.
1
0
,
n
o
.
1
–
7
,
h
t
t
p
s
:
/
/
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
mas
t
e
r
y
.
c
o
m/
f
e
a
t
u
r
e
-
se
l
e
c
t
i
o
n
-
w
i
t
h
-
r
e
a
l
-
a
n
d
-
c
a
t
e
g
o
r
i
c
a
l
-
d
a
t
a
/
[
2
0
]
N
.
M
o
u
s
t
a
f
a
a
n
d
J
.
S
l
a
y
,
“
U
N
S
W
-
N
B
1
5
:
A
c
o
m
p
r
e
h
e
n
si
v
e
d
a
t
a
s
e
t
f
o
r
n
e
t
w
o
r
k
i
n
t
r
u
s
i
o
n
d
e
t
e
c
t
i
o
n
sy
st
e
ms
(
U
N
S
W
-
N
B
1
5
n
e
t
w
o
r
k
d
a
t
a
se
t
)
,
”
2
0
1
5
M
i
l
i
t
a
ry
C
o
m
m
u
n
i
c
a
t
i
o
n
s
a
n
d
I
n
f
o
rm
a
t
i
o
n
S
y
s
t
e
m
s
C
o
n
f
e
re
n
c
e
,
M
i
l
C
I
S
2
0
1
5
,
2
0
1
5
,
d
o
i
:
1
0
.
1
1
0
9
/
M
i
l
C
I
S
.
2
0
1
5
.
7
3
4
8
9
4
2
.
[
2
1
]
A
.
E
.
K
a
r
r
a
r
,
“
T
h
e
e
f
f
e
c
t
o
f
u
s
i
n
g
d
a
t
a
p
r
e
-
p
r
o
c
e
ss
i
n
g
b
y
i
mp
u
t
a
t
i
o
n
s
i
n
h
a
n
d
l
i
n
g
mi
ss
i
n
g
v
a
l
u
e
s,”
I
n
d
o
n
e
s
i
a
n
J
o
u
rn
a
l
o
f
El
e
c
t
ri
c
a
l
En
g
i
n
e
e
ri
n
g
a
n
d
I
n
f
o
rm
a
t
i
c
s
,
v
o
l
.
1
0
,
n
o
.
2
,
p
p
.
3
7
5
–
3
8
4
,
2
0
2
2
,
d
o
i
:
1
0
.
5
2
5
4
9
/
i
j
e
e
i
.
v
1
0
i
2
.
3
7
3
0
.
[
2
2
]
A
.
S
e
r
a
j
e
t
a
l
.
,
“
C
r
o
ss
-
v
a
l
i
d
a
t
i
o
n
,
”
H
a
n
d
b
o
o
k
o
f
H
y
d
r
o
I
n
f
o
rm
a
t
i
c
s:
V
o
l
u
m
e
I
:
C
l
a
ssi
c
S
o
f
t
-
C
o
m
p
u
t
i
n
g
T
e
c
h
n
i
q
u
e
s
,
p
p
.
8
9
–
1
0
5
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
B
9
7
8
-
0
-
12
-
8
2
1
2
8
5
-
1
.
0
0
0
2
1
-
X.
[
2
3
]
Y
.
D
i
n
g
,
Q
.
P
e
n
g
,
Z.
S
o
n
g
,
a
n
d
H
.
C
h
e
n
,
“
V
a
r
i
a
b
l
e
se
l
e
c
t
i
o
n
a
n
d
r
e
g
u
l
a
r
i
z
a
t
i
o
n
v
i
a
a
r
b
i
t
r
a
r
y
r
e
c
t
a
n
g
l
e
-
r
a
n
g
e
g
e
n
e
r
a
l
i
z
e
d
e
l
a
s
t
i
c
n
e
t
,
”
S
t
a
t
i
s
t
i
c
s
a
n
d
C
o
m
p
u
t
i
n
g
,
v
o
l
.
3
3
,
n
o
.
3
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
2
2
2
-
0
2
3
-
1
0
2
4
0
-
4.
[
2
4
]
H
.
H
.
H
t
u
n
,
M
.
B
i
e
h
l
,
a
n
d
N
.
P
e
t
k
o
v
,
“
S
u
r
v
e
y
o
f
f
e
a
t
u
r
e
sel
e
c
t
i
o
n
a
n
d
e
x
t
r
a
c
t
i
o
n
t
e
c
h
n
i
q
u
e
s
f
o
r
s
t
o
c
k
mar
k
e
t
p
r
e
d
i
c
t
i
o
n
,
”
F
i
n
a
n
c
i
a
l
I
n
n
o
v
a
t
i
o
n
,
v
o
l
.
9
,
n
o
.
1
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
8
6
/
s
4
0
8
5
4
-
0
2
2
-
0
0
4
4
1
-
7.
[
2
5
]
M
.
G
r
e
e
n
a
c
r
e
,
P
.
J.
F
.
G
r
o
e
n
e
n
,
T.
H
a
st
i
e
,
A
.
I
.
D
’
En
z
a
,
A
.
M
a
r
k
o
s,
a
n
d
E
.
T
u
z
h
i
l
i
n
a
,
“
P
r
i
n
c
i
p
a
l
c
o
mp
o
n
e
n
t
a
n
a
l
y
s
i
s,”
N
a
t
u
r
e
Re
v
i
e
w
s M
e
t
h
o
d
s
Pri
m
e
rs
,
v
o
l
.
2
,
n
o
.
1
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
3
8
/
s
4
3
5
8
6
-
0
2
2
-
0
0
1
8
4
-
w.
[
2
6
]
I
.
Tsa
mar
d
i
n
o
s
,
A
.
R
a
k
h
sh
a
n
i
,
a
n
d
V
.
La
g
a
n
i
,
“
P
e
r
f
o
r
m
a
n
c
e
-
e
st
i
mat
i
o
n
p
r
o
p
e
r
t
i
e
s
o
f
c
r
o
s
s
-
v
a
l
i
d
a
t
i
o
n
-
b
a
se
d
p
r
o
t
o
c
o
l
s
w
i
t
h
si
mu
l
t
a
n
e
o
u
s
h
y
p
e
r
-
p
a
r
a
me
t
e
r
o
p
t
i
m
i
z
a
t
i
o
n
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
n
Ar
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
T
o
o
l
s
,
v
o
l
.
2
4
,
n
o
.
5
,
2
0
1
5
,
d
o
i
:
1
0
.
1
1
4
2
/
S
0
2
1
8
2
1
3
0
1
5
4
0
0
2
3
0
.
[
2
7
]
O
.
A
.
V
.
Li
l
i
e
n
f
e
l
d
,
“
I
n
t
r
o
d
u
c
i
n
g
mac
h
i
n
e
l
e
a
r
n
i
n
g
:
S
c
i
e
n
c
e
a
n
d
t
e
c
h
n
o
l
o
g
y
,
”
M
a
c
h
i
n
e
L
e
a
rn
i
n
g
:
S
c
i
e
n
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
1
,
n
o
.
1
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
8
8
/
2
6
3
2
-
2
1
5
3
/
a
b
6
d
5
d
.
[
2
8
]
J.
A
.
I
l
e
m
o
b
a
y
o
e
t
a
l
.
,
“
H
y
p
e
r
p
a
r
a
me
t
e
r
t
u
n
i
n
g
i
n
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
:
a
c
o
m
p
r
e
h
e
n
si
v
e
r
e
v
i
e
w
,
”
J
o
u
r
n
a
l
o
f
E
n
g
i
n
e
e
ri
n
g
R
e
se
a
rc
h
a
n
d
Re
p
o
r
t
s
,
v
o
l
.
2
6
,
n
o
.
6
,
p
p
.
3
8
8
–
3
9
5
,
2
0
2
4
,
d
o
i
:
1
0
.
9
7
3
4
/
j
e
r
r
/
2
0
2
4
/
v
2
6
i
6
1
1
8
8
.
[
2
9
]
Ž.
V
u
j
o
v
i
ć
,
“
C
l
a
ssi
f
i
c
a
t
i
o
n
m
o
d
e
l
e
v
a
l
u
a
t
i
o
n
m
e
t
r
i
c
s,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
Ad
v
a
n
c
e
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
1
2
,
n
o
.
6
,
p
p
.
5
9
9
–
6
0
6
,
2
0
2
1
,
d
o
i
:
1
0
.
1
4
5
6
9
/
I
JA
C
S
A
.
2
0
2
1
.
0
1
2
0
6
7
0
.
[
3
0
]
Y
.
A
h
m
e
d
,
K
.
B
e
y
i
o
k
u
,
a
n
d
M
.
Y
o
u
s
e
f
i
,
“
S
e
c
u
r
i
n
g
smar
t
c
i
t
i
e
s
t
h
r
o
u
g
h
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
:
A
h
o
n
e
y
p
o
t
‐
d
r
i
v
e
n
a
p
p
r
o
a
c
h
t
o
a
t
t
a
c
k
d
e
t
e
c
t
i
o
n
i
n
i
n
t
e
r
n
e
t
o
f
t
h
i
n
g
s
e
c
o
sy
st
e
ms,”
I
ET
S
m
a
r
t
C
i
t
i
e
s
,
v
o
l
.
6
,
n
o
.
3
,
p
p
.
1
8
0
–
1
9
8
,
S
e
p
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
4
9
/
smc2
.
1
2
0
8
4
.
[
3
1
]
I
.
T.
A
l
-
H
a
l
b
o
o
si
,
B
.
M
.
El
b
a
g
o
u
r
y
,
S
.
E
l
-
R
e
g
a
i
l
y
,
a
n
d
E.
S
.
M
.
E
l
-
H
o
r
b
a
t
y
,
“
F
e
d
e
r
a
t
e
d
i
n
c
e
p
t
i
o
n
-
m
u
l
t
i
-
h
e
a
d
a
t
t
e
n
t
i
o
n
mo
d
e
l
s
f
o
r
c
y
b
e
r
-
a
t
t
a
c
k
s
d
e
t
e
c
t
i
o
n
,
”
I
A
ES
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
1
3
,
n
o
.
4
,
p
p
.
4
7
7
8
–
4
7
9
4
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
a
i
.
v
1
3
.
i
4
.
p
p
4
7
7
8
-
4
7
9
4
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
La
h
c
e
n
Id
o
u
g
li
d
is
a
P
h
.
D
.
re
se
a
rc
h
e
r
a
t
I
b
n
To
fa
il
U
n
iv
e
rsit
y
'
s
F
a
c
u
lt
y
o
f
S
c
ien
c
e
s
in
Ke
n
it
ra
,
wo
r
k
s
with
i
n
th
e
Co
m
p
u
ter
S
c
ien
c
e
Re
se
a
rc
h
Lab
o
ra
to
ry
(LaRIT).
His
re
se
a
rc
h
e
n
c
o
m
p
a
ss
e
s
c
o
m
p
u
ter
n
e
two
rk
s,
so
ftwa
re
e
n
g
in
e
e
rin
g
,
a
rti
ficia
l
i
n
telli
g
e
n
c
e
,
a
n
d
se
c
u
rit
y
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
lah
c
e
n
.
id
o
u
g
li
d
@u
it
.
a
c
.
m
a
.
S
a
id
Tk
a
te
k
is
P
ro
fe
ss
o
r
in
c
o
m
p
u
ter
sc
ien
c
e
a
t
Ib
n
To
fa
il
Un
iv
e
rsity
in
Ke
n
i
tra,
h
e
is
a
m
e
m
b
e
r
o
f
th
e
Co
m
p
u
ter
S
c
ien
c
e
Re
se
a
rc
h
Lab
o
ra
to
ry
(L
a
RI
).
His
p
rima
ry
re
se
a
r
c
h
a
c
ro
ss
v
a
rio
u
s
field
s
i
n
c
lu
d
e
s
b
i
g
d
a
ta,
a
rti
ficia
l
i
n
telli
g
e
n
c
e
(AI),
a
n
d
th
e
ir
a
p
p
li
c
a
ti
o
n
s.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
sa
id
.
tk
a
tek
@u
it
.
a
c
.
m
a
.
K
h
a
li
d
Elfa
y
q
a
P
h
.
D
.
re
se
a
rc
h
e
r
a
t
I
b
n
To
fa
i
l
Un
iv
e
rsity
'
s F
a
c
u
lt
y
o
f
S
c
ien
c
e
s
i
n
Ke
n
it
ra
,
wo
rk
s
wit
h
in
th
e
Co
m
p
u
ter
S
c
ien
c
e
Re
se
a
rc
h
Lab
o
ra
t
o
ry
(LaRIT).
His
re
se
a
rc
h
e
n
c
o
m
p
a
ss
e
s
so
ftwa
re
e
n
g
i
n
e
e
rin
g
,
c
o
m
p
u
ter
n
e
two
r
k
s,
a
rti
ficia
l
i
n
telli
g
e
n
c
e
,
a
n
d
a
u
d
io
v
isu
a
l
tec
h
n
o
l
o
g
ies
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
k
h
a
li
d
.
e
lfay
q
@u
it
.
a
c
.
m
a
.
Evaluation Warning : The document was created with Spire.PDF for Python.