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.
14
,
No
.
6
,
Dec
em
b
er
20
24
,
p
p
.
6
6
4
2
~
6
6
5
5
I
SS
N:
2088
-
8
7
0
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijece.
v
14
i
6
.
pp
6
6
4
2
-
6
6
5
5
6642
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m
A comp
a
ra
tive st
udy
of ma
chine
le
a
rning
t
o
o
ls for d
etec
ti
ng
Tro
jan ho
rse inf
e
ctions
in
clo
ud
co
mputing
enviro
n
ments
H
a
s
a
n K
a
na
k
er
1
,
M
o
nther
T
a
ra
wneh
2
,
Na
der
Abdel K
a
rim
3
,
Adeeb
Als
a
a
ida
h
4
,
M
a
he
r
Abuh
a
m
deh
5
,
O
s
a
m
a
Q
t
a
is
h
6
,
E
s
s
a
m
Alhr
o
o
b
1
,
Z
a
id Alha
lho
uli
7
1
D
e
p
a
r
t
me
n
t
o
f
C
y
b
e
r
S
e
c
u
r
i
t
y
,
F
a
c
u
l
t
y
o
f
I
n
f
o
r
m
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
,
I
sr
a
U
n
i
v
e
r
s
i
t
y
,
A
mm
a
n
,
J
o
r
d
a
n
2
D
e
p
a
r
t
me
n
t
o
f
I
n
f
o
r
mat
i
o
n
Te
c
h
n
o
l
o
g
y
,
C
o
l
l
e
g
e
o
f
I
n
f
o
r
mat
i
o
n
Te
c
h
n
o
l
o
g
y
,
Ta
f
i
l
a
Te
c
h
n
i
c
a
l
U
n
i
v
e
r
si
t
y
,
A
l
t
a
f
i
l
a
,
J
o
r
d
a
n
3
D
e
p
a
r
t
me
n
t
o
f
I
n
t
e
l
l
i
g
e
n
t
S
y
st
e
ms,
F
a
c
u
l
t
y
o
f
A
r
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
,
A
l
-
B
a
l
q
a
A
p
p
l
i
e
d
U
n
i
v
e
r
si
t
y
,
A
l
-
S
a
l
t
,
J
o
r
d
a
n
4
D
e
p
a
r
t
me
n
t
o
f
N
e
t
w
o
r
k
s
a
n
d
C
y
b
e
r
S
e
c
u
r
i
t
y
,
F
a
c
u
l
t
y
o
f
I
n
f
o
r
mat
i
o
n
Te
c
h
n
o
l
o
g
y
,
A
l
-
A
h
l
i
y
y
a
A
mm
a
n
U
n
i
v
e
r
si
t
y
,
A
mm
a
n
,
J
o
r
d
a
n
5
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
I
n
f
o
r
m
a
t
i
o
n
S
y
st
e
m,
F
a
c
u
l
t
y
o
f
I
n
f
o
r
m
a
t
i
o
n
Te
c
h
n
o
l
o
g
y
,
I
sr
a
U
n
i
v
e
r
si
t
y
,
A
mm
a
n
,
Jo
r
d
a
n
6
D
e
p
a
r
t
me
n
t
o
f
S
o
f
t
w
a
r
e
En
g
i
n
e
e
r
i
n
g
,
F
a
c
u
l
t
y
o
f
I
n
f
o
r
ma
t
i
o
n
Te
c
h
n
o
l
o
g
y
,
I
sr
a
U
n
i
v
e
r
s
i
t
y
,
A
mm
a
n
,
Jo
r
d
a
n
7
D
e
p
a
r
t
me
n
t
o
f
D
a
t
a
S
c
i
e
n
c
e
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
,
F
a
c
u
l
t
y
o
f
I
n
f
o
r
ma
t
i
o
n
Te
c
h
n
o
l
o
g
y
,
I
sr
a
U
n
i
v
e
r
s
i
t
y
,
A
m
man
,
J
o
r
d
a
n
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Ma
y
2
,
2
0
2
4
R
ev
is
ed
J
u
l 2
0
,
2
0
2
4
Acc
ep
ted
Au
g
6
,
2
0
2
4
Clo
u
d
c
o
m
p
u
ti
n
g
o
ffe
rs
se
v
e
ra
l
a
d
v
a
n
ta
g
e
s,
i
n
c
lu
d
in
g
c
o
st
sa
v
in
g
s
a
n
d
e
a
sy
a
c
c
e
s
s
to
re
so
u
rc
e
s,
it
is
a
lso
c
o
u
ld
b
e
v
u
l
n
e
ra
b
le
t
o
se
rio
u
s
se
c
u
ri
ty
a
tt
a
c
k
s
su
c
h
a
s
c
lo
u
d
Tr
o
jan
h
o
rse
in
fe
c
t
io
n
a
tt
a
c
k
s.
T
o
a
d
d
re
ss
th
is
issu
e
,
m
a
c
h
in
e
lea
rn
in
g
is
a
p
ro
m
isin
g
a
p
p
ro
a
c
h
fo
r
d
e
tec
ti
n
g
th
e
se
t
h
re
a
ts.
Th
u
s
,
d
iffere
n
t
m
a
c
h
in
e
lea
rn
in
g
t
o
o
ls
a
n
d
m
o
d
e
ls
h
a
v
e
b
e
e
n
e
m
p
lo
y
e
d
t
o
d
e
te
c
t
Tro
jan
h
o
rse
in
fe
c
ti
o
n
su
c
h
a
s
Wek
a
a
n
d
P
y
th
o
n
C
o
lab
.
Th
is
st
u
d
y
a
ims
to
c
o
m
p
a
re
th
e
p
e
rf
o
rm
a
n
c
e
o
f
We
k
a
a
n
d
P
y
th
o
n
Co
lab
,
a
s
p
o
p
u
lar
t
o
o
ls
fo
r
b
u
il
d
i
n
g
m
a
c
h
in
e
lea
rn
in
g
m
o
d
e
ls.
Th
is
stu
d
y
e
v
a
l
u
a
tes
th
e
re
c
a
ll
,
a
c
c
u
ra
c
y
,
a
n
d
F1
-
sc
o
re
o
f
m
a
c
h
in
e
lea
rn
i
n
g
m
o
d
e
ls
b
u
il
t
with
Wek
a
a
n
d
P
y
th
o
n
Co
lab
a
n
d
c
o
m
p
a
re
s
th
e
ir
c
o
m
p
u
tatio
n
a
l
re
s
o
u
rc
e
s
re
q
u
ired
e
m
p
lo
y
in
g
se
v
e
ra
l
m
a
c
h
in
e
lea
rn
in
g
a
lg
o
rit
h
m
s.
Th
e
d
a
tas
e
t
c
o
ll
e
c
ted
a
n
d
a
n
a
ly
z
e
d
u
si
n
g
d
y
n
a
m
ic
a
n
a
ly
sis
o
f
Tro
jan
h
o
rse
in
fe
c
ti
o
n
in
c
o
n
tro
l
lab
e
n
v
iro
n
m
e
n
t.
T
h
e
fin
d
in
g
s
o
f
th
is
stu
d
y
c
a
n
h
e
l
p
d
e
term
in
e
th
e
d
e
c
isio
n
a
b
o
u
t
wh
ich
to
o
l
to
u
se
t
o
d
e
tec
t
Tro
jan
h
o
rse
in
fe
c
ti
o
n
s
a
n
d
p
r
o
v
id
e
in
si
g
h
ts
i
n
t
o
th
e
stre
n
g
t
h
s
a
n
d
li
m
it
a
ti
o
n
s o
f
Wek
a
a
n
d
P
y
th
o
n
Co
lab
fo
r
b
u
il
d
in
g
m
a
c
h
in
e
-
lea
rn
in
g
m
o
d
e
ls
in
g
e
n
e
ra
l.
K
ey
w
o
r
d
s
:
C
lo
u
d
co
m
p
u
tin
g
C
o
n
tr
o
l la
b
C
y
b
er
s
ec
u
r
ity
Ma
ch
in
e
lear
n
in
g
Py
th
o
n
T
r
o
jan
h
o
r
s
e
in
f
ec
tio
n
W
ek
a
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
:
Hasan
Kan
ak
er
Dep
ar
tm
en
t o
f
C
y
b
er
Secu
r
ity
,
Facu
lty
o
f
I
n
f
o
r
m
atio
n
T
ec
h
n
o
lo
g
y
,
I
s
r
a
Un
iv
e
r
s
ity
Am
m
an
,
1
1
6
2
2
,
J
o
r
d
a
n
E
m
ail: h
asan
.
k
an
ak
e
r
@
iu
.
ed
u
.
jo
1.
I
NT
RO
D
UCT
I
O
N
C
lo
u
d
co
m
p
u
tin
g
,
an
in
n
o
v
a
tiv
e
m
eth
o
d
o
f
in
f
o
r
m
atio
n
tech
n
o
lo
g
y
,
u
s
es
th
e
in
ter
n
et
an
d
r
em
o
t
e
s
er
v
er
s
to
d
eliv
er
a
co
m
m
o
n
s
et
o
f
co
m
p
u
tin
g
r
eso
u
r
ce
s
an
d
ap
p
licatio
n
s
to
m
ee
t
cu
s
to
m
e
r
r
eq
u
ir
em
e
n
ts
[
1
]
.
T
h
e
ad
v
en
t
o
f
clo
u
d
co
m
p
u
tin
g
tech
n
o
lo
g
ies
in
th
e
f
ir
s
t
d
ec
ad
e
o
f
th
is
ce
n
tu
r
y
u
s
h
er
ed
in
a
n
ew
er
a
in
th
e
d
ev
elo
p
m
e
n
t
o
f
in
f
o
r
m
atio
n
t
ec
h
n
o
lo
g
y
(
I
T
)
in
f
r
astru
ctu
r
e,
as
u
s
er
s
ca
n
n
o
w
o
b
tain
s
o
f
t
war
e
an
d
co
m
p
u
tin
g
p
o
wer
o
v
e
r
th
e
I
n
ter
n
et
o
r
n
etwo
r
k
s
.
T
h
is
h
as
also
led
to
th
e
em
er
g
en
ce
o
f
n
ew
m
o
d
els
f
o
r
h
o
s
tin
g
an
d
d
is
tr
ib
u
tin
g
o
n
lin
e
s
er
v
ices
[
2
]
,
[
3
]
.
C
lo
u
d
co
m
p
u
tin
g
allo
ws
u
s
er
s
to
ea
s
ily
ac
ce
s
s
s
er
v
ices,
f
ac
ilit
atin
g
s
ea
m
less
ac
ce
s
s
to
d
ata
an
d
p
r
o
g
r
a
m
ex
ec
u
tio
n
o
n
a
lar
g
e
n
u
m
b
er
o
f
co
n
n
ec
ted
c
o
m
p
u
ter
s
.
Mo
r
eo
v
e
r
,
it
r
em
o
v
es
th
e
r
e
q
u
ir
em
en
t f
o
r
u
s
er
s
to
in
s
tall
s
o
f
twar
e
o
n
th
eir
p
er
s
o
n
al
co
m
p
u
ter
s
an
d
in
s
tead
p
r
o
v
id
es
ac
ce
s
s
to
r
eso
u
r
ce
s
an
d
ap
p
licatio
n
s
v
ia
an
in
ter
n
et
c
o
n
n
e
ctio
n
at
a
n
y
tim
e
an
d
f
r
o
m
an
y
lo
ca
tio
n
[
2
]
,
[
4
]
.
C
lo
u
d
co
m
p
u
tin
g
h
as
b
ee
n
wid
ely
ad
o
p
ted
b
y
p
u
b
lic
-
s
ec
to
r
b
u
s
in
ess
es
in
m
an
y
co
u
n
tr
ies,
in
clu
d
in
g
Au
s
tr
alia,
th
e
Un
ited
Kin
g
d
o
m
(
UK)
,
th
e
Un
ited
State
(
U
S),
an
d
v
ar
i
o
u
s
E
u
r
o
p
ea
n
c
o
u
n
tr
ies
[
5
]
–
[
7
]
.
T
h
e
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
A
co
mp
a
r
a
tive
s
tu
d
y
o
f m
a
ch
i
n
e
lea
r
n
in
g
to
o
ls
fo
r
d
etec
tin
g
Tr
o
ja
n
h
o
r
s
e
…
(
Ha
s
a
n
K
a
n
a
ke
r
)
6643
g
o
v
er
n
m
en
t
clo
u
d
(G
-
C
lo
u
d
)
in
f
r
astru
ctu
r
e
was
in
tr
o
d
u
ce
d
b
y
th
e
UK
g
o
v
e
r
n
m
en
t
i
n
2
0
1
0
,
s
av
in
g
a
n
esti
m
ated
£
3
.
2
b
illi
o
n
[
5
]
,
[
8
]
,
[
9
]
.
T
h
e
cl
o
u
d
co
m
p
u
tin
g
m
a
ll
was
also
in
tr
o
d
u
ce
d
b
y
t
h
e
US
g
o
v
er
n
m
en
t
in
2009
[
8
]
.
C
lo
u
d
co
m
p
u
tin
g
o
f
f
er
s
n
u
m
er
o
u
s
b
en
ef
its
s
u
ch
as
s
h
ar
ed
ass
et
p
o
o
lin
g
,
tr
e
m
en
d
o
u
s
v
er
s
atility
,
co
s
t
s
av
in
g
s
,
f
lex
ib
ilit
y
,
p
ay
-
as
-
you
-
g
o
p
r
icin
g
,
s
elf
-
p
r
o
v
is
io
n
in
g
o
f
r
eso
u
r
ce
s
,
a
n
d
m
u
lti
-
ten
an
cy
[
1
0
]
–
[
1
2
]
.
Ho
wev
er
,
th
e
r
e
ar
e
s
till
s
ec
u
r
ity
co
n
ce
r
n
s
an
d
r
is
k
s
ass
o
ciate
d
with
clo
u
d
c
o
m
p
u
tin
g
[
1
3
]
,
p
ar
ticu
lar
ly
th
e
r
is
k
o
f
s
ec
u
r
ity
attac
k
s
.
As a
r
esu
lt,
b
o
th
s
er
v
ice
p
r
o
v
i
d
er
s
a
n
d
clien
ts
ar
e
co
n
ce
r
n
ed
ab
o
u
t th
ese
attac
k
s
[
1
4
]
.
Var
io
u
s
f
o
r
m
s
o
f
attac
k
s
p
o
s
e
a
th
r
ea
t
to
clo
u
d
co
m
p
u
ti
n
g
,
in
clu
d
i
n
g
p
h
is
h
in
g
,
au
t
h
en
ticatio
n
,
s
er
v
ice
d
en
ial,
m
an
-
in
-
th
e
-
m
id
d
le,
a
n
d
m
alwa
r
e
in
s
er
tio
n
,
am
o
n
g
o
th
e
r
s
[
2
]
,
[
1
5
]
–
[
1
8
]
.
Am
o
n
g
t
h
em
,
m
alwa
r
e
attac
k
s
ar
e
a
s
ig
n
if
ican
t r
is
k
to
th
e
clo
u
d
co
m
p
u
tin
g
ec
o
s
y
s
tem
.
On
e
s
u
ch
attac
k
is
th
e
"
clo
u
d
T
r
o
jan
in
jectio
n
attac
k
,
"
wh
ic
h
in
v
o
l
v
es
in
jectin
g
a
m
alicio
u
s
p
r
o
g
r
am
in
to
clo
u
d
s
er
v
ices
to
ca
u
s
e
h
ar
m
.
T
h
ese
p
r
o
g
r
a
m
s
ca
n
b
e
d
is
g
u
is
ed
as
n
o
r
m
al
co
m
m
an
d
s
an
d
ex
ec
u
ted
as
s
u
ch
[
1
9
]
.
C
lo
u
d
T
r
o
j
an
in
jectio
n
attac
k
s
ca
n
in
tr
o
d
u
ce
m
alicio
u
s
s
er
v
i
ce
s
,
v
ir
tu
al
m
ac
h
in
es,
an
d
ap
p
licatio
n
s
in
to
clo
u
d
s
y
s
tem
s
,
wh
ich
ca
n
af
f
ec
t
clo
u
d
f
u
n
ctio
n
ality
b
y
i
n
ter
f
e
r
in
g
with
o
r
alter
in
g
it
[
2
0
]
,
[
2
1
]
.
Attack
er
s
u
p
l
o
ad
a
m
alicio
u
s
v
er
s
io
n
o
f
ap
p
licatio
n
,
v
ir
tu
al
m
ac
h
in
e
o
r
s
er
v
ice
to
a
clo
u
d
s
y
s
tem
s
o
th
at
it
will
th
in
k
it
is
a
r
ea
l
in
s
tan
ce
o
f
th
at
ap
p
licatio
n
v
ir
tu
al
m
ac
h
i
n
e
o
r
s
er
v
ice.
W
h
en
a
n
o
r
m
al
u
s
er
r
eq
u
ests
an
in
s
tan
ce
o
f
th
e
m
alicio
u
s
s
er
v
ice
to
r
u
n
,
th
e
m
alicio
u
s
co
d
e
is
e
x
ec
u
ted
[
2
0
]
.
T
r
o
ja
n
h
o
r
s
es
ar
e
ch
allen
g
in
g
to
d
etec
t
u
s
in
g
s
ig
n
atu
r
e
-
b
ased
tech
n
o
lo
g
ies,
wh
ich
is
t
h
e
m
o
s
t
co
m
m
o
n
a
n
ti
-
v
ir
u
s
s
y
s
te
m
d
etec
tio
n
m
eth
o
d
[
2
2
]
.
Fu
r
th
er
m
o
r
e,
o
n
ly
well
-
k
n
o
wn
s
ig
n
at
u
r
es
ca
n
b
e
u
s
ed
with
s
ig
n
atu
r
e
-
b
ased
an
ti
-
v
ir
u
s
to
ac
h
ie
v
e
h
ig
h
ac
cu
r
ac
y
.
T
h
is
ty
p
e
o
f
d
etec
tio
n
h
as
a
d
r
awb
ac
k
in
th
at
it
f
r
e
q
u
en
tly
m
is
s
es
f
r
esh
attac
k
s
wh
e
n
m
alwa
r
e
co
m
p
letely
ch
a
n
g
es
its
s
ig
n
atu
r
e.
W
ith
th
e
in
cr
ea
s
in
g
n
u
m
b
er
o
f
c
y
b
er
-
attac
k
s
,
th
e
n
ee
d
f
o
r
ef
f
ec
tiv
e
to
o
ls
to
d
etec
t
m
a
lwar
e
an
d
T
r
o
jan
h
o
r
s
e
in
f
ec
tio
n
s
is
m
o
r
e
im
p
o
r
tan
t
th
an
e
v
er
.
T
o
p
r
o
tect
an
d
r
ed
u
ce
d
am
ag
e
ca
u
s
ed
b
y
T
r
o
ja
n
h
o
r
s
e
in
f
ec
tio
n
,
m
ac
h
in
e
lear
n
in
g
to
o
ls
h
av
e
b
ee
n
p
r
o
v
e
n
as
an
e
f
f
ec
tiv
e
tech
n
iq
u
e
u
s
ed
t
o
d
ete
ct
s
u
ch
th
r
ea
ts
b
y
an
aly
zin
g
p
atter
n
s
in
th
e
d
at
a
an
d
id
en
tif
y
in
g
p
o
s
s
ib
le
d
an
g
er
o
u
s
b
eh
av
io
r
.
W
ek
a
an
d
Py
th
o
n
C
o
lab
ar
e
k
n
o
wn
to
o
ls
to
u
s
e
an
d
im
p
le
m
en
t
m
ac
h
in
e
lear
n
in
g
m
o
d
el
s
.
W
ek
a
i
s
a
J
av
a
-
b
ased
p
r
o
g
r
am
th
at
p
r
o
v
id
es
a
g
r
ap
h
ical
u
s
er
in
te
r
f
ac
e
ex
p
lo
ited
r
ea
d
y
-
to
-
u
s
e
m
ac
h
in
e
lea
r
n
in
g
m
o
d
els,
wh
ile
Py
th
o
n
C
o
lab
is
an
o
n
lin
e
J
u
p
y
ter
n
o
teb
o
o
k
en
v
ir
o
n
m
e
n
t
th
at
allo
ws
u
s
er
s
to
u
tili
ze
to
im
p
lem
en
t
an
d
r
u
n
Py
th
o
n
m
ac
h
in
e
lear
n
in
g
lib
r
ar
ies
in
th
e
clo
u
d
.
Alth
o
u
g
h
,
it
is
p
o
s
s
ib
le
to
cr
ea
te
an
d
u
tili
ze
m
ac
h
in
e
lear
n
in
g
m
o
d
el
s
to
id
en
tify
T
r
o
jan
h
o
r
s
e
in
f
ec
tio
n
s
u
s
in
g
eith
e
r
W
ek
a
o
r
Py
th
o
n
C
o
lab
,
th
e
r
e
is
lim
ited
r
esear
ch
es
co
m
p
ar
in
g
th
e
p
er
f
o
r
m
a
n
ce
o
f
th
ese
two
to
o
ls
.
T
h
e
g
o
al
o
f
th
is
s
tu
d
y
is
to
co
m
p
ar
e
th
e
p
er
f
o
r
m
a
n
ce
o
f
W
ek
a
an
d
Py
th
o
n
C
o
lab
in
d
etec
tin
g
T
r
o
ja
n
h
o
r
s
e
in
f
ec
tio
n
s
.
W
e
will
ev
alu
ate
th
e
r
ec
all,
ac
cu
r
ac
y
,
a
n
d
F1
-
s
co
r
e
o
f
m
ac
h
i
n
e
lear
n
in
g
m
o
d
els
b
u
ilt
with
W
ek
a
an
d
im
p
lan
ted
u
s
in
g
Py
th
o
n
C
o
lab
.
W
e
will
also
co
m
p
ar
e
th
e
c
o
m
p
u
tatio
n
al
r
e
s
o
u
r
ce
s
r
eq
u
ir
ed
to
b
u
ild
a
n
d
t
r
ain
t
h
e
m
o
d
els,
in
clu
d
in
g
p
r
o
ce
s
s
in
g
tim
e
a
n
d
m
em
o
r
y
u
s
ag
e.
T
h
e
r
esu
lts
o
f
th
is
s
tu
d
y
h
elp
t
o
d
eter
m
in
e
t
h
e
d
ec
is
io
n
ab
o
u
t
wh
ich
to
o
l
to
u
s
e
to
d
etec
t
T
r
o
jan
h
o
r
s
e
in
f
ec
tio
n
s
.
Fu
r
t
h
er
m
o
r
e
,
th
is
s
tu
d
y
p
r
o
v
id
es
i
n
s
ig
h
ts
in
to
th
e
s
tr
e
n
g
th
s
an
d
lim
itatio
n
s
o
f
W
ek
a
an
d
Py
th
o
n
C
o
lab
f
o
r
b
u
ild
i
n
g
m
ac
h
i
n
e
-
lear
n
in
g
m
o
d
els in
T
r
o
ja
n
h
o
r
s
e
d
etec
tio
n
.
2.
B
ACK
G
RO
UND
AN
D
RE
L
AT
E
D
WO
RK
S
T
r
o
jan
h
o
r
s
es
ar
e
a
t
y
p
e
o
f
m
alicio
u
s
s
o
f
twar
e
th
at
p
o
s
e
as
leg
itima
te
o
r
h
elp
f
u
l
p
r
o
g
r
a
m
s
to
g
ai
n
ac
ce
s
s
to
u
s
er
s
'
co
m
p
u
ter
s
.
C
y
b
er
cr
im
in
als
f
r
eq
u
en
tly
u
s
e
s
o
cial
en
g
in
ee
r
in
g
tactics
to
tr
ick
u
s
er
s
in
to
u
n
in
ten
tio
n
ally
in
s
tallin
g
th
e
m
[
2
3
]
.
O
n
ce
in
s
talled
,
T
r
o
jan
h
o
r
s
es
h
a
v
e
th
e
ab
ilit
y
to
r
em
o
tely
m
an
i
p
u
late
th
e
v
ictim
'
s
d
ev
ice,
s
teal
s
en
s
i
tiv
e
d
ata,
watc
h
u
s
er
ac
tiv
ities
,
an
d
co
r
r
u
p
t,
d
estro
y
,
o
r
m
o
d
i
f
y
s
y
s
tem
f
iles
[
2
4
]
,
[
2
5
]
.
Un
lik
e
v
ir
u
s
es
an
d
wo
r
m
s
,
T
r
o
jan
s
r
eq
u
ir
e
u
s
er
in
ter
ac
tio
n
to
p
r
o
p
ag
ate
.
T
h
ey
ar
e
co
n
s
id
er
ed
am
o
n
g
th
e
m
o
s
t d
an
g
e
r
o
u
s
f
o
r
m
s
o
f
m
alwa
r
e
d
u
e
to
t
h
eir
p
r
e
v
alen
ce
[
2
4
]
–
[
2
6
]
.
T
r
o
jan
h
o
r
s
es
ca
n
co
m
e
in
two
ty
p
es:
g
en
er
al
an
d
r
em
o
t
e
-
ac
ce
s
s
T
r
o
jan
s
[
2
4
]
.
Gen
er
al
T
r
o
jan
s
p
er
f
o
r
m
a
v
ar
iety
o
f
m
alicio
u
s
ac
tio
n
s
,
in
clu
d
in
g
jeo
p
ar
d
izin
g
th
e
d
ata
in
teg
r
ity
o
f
v
ictim
m
ac
h
in
es,
d
ir
ec
tin
g
u
s
er
s
'
wo
r
k
s
tatio
n
s
to
s
p
ec
if
ic
web
s
ites
v
ia
s
y
s
tem
f
iles
,
an
d
r
u
n
n
i
n
g
a
d
d
itio
n
al
h
ar
m
f
u
l
p
r
o
g
r
am
s
[
2
6
]
.
T
h
ey
h
av
e
th
e
a
b
ilit
y
to
tr
ac
k
u
s
er
b
eh
av
io
r
a
n
d
p
r
o
v
i
d
e
th
e
i
n
f
o
r
m
atio
n
to
th
e
attac
k
er
.
R
em
o
t
e
-
ac
ce
s
s
T
r
o
jan
s
,
o
n
th
e
o
th
er
h
a
n
d
,
h
av
e
a
s
p
ec
ial
ca
p
ac
ity
t
h
at
allo
ws
h
ac
k
e
r
s
t
o
r
em
o
tely
m
a
n
ip
u
late
th
e
tar
g
et
m
ac
h
in
e
ac
r
o
s
s
th
e
i
n
ter
n
et
o
r
a
lo
ca
l
ar
ea
n
etwo
r
k
(
L
AN)
.
T
h
is
ty
p
e
o
f
T
r
o
jan
is
p
ar
ticu
lar
ly
d
an
g
er
o
u
s
as
it
ca
n
b
e
u
s
ed
to
s
teal
co
n
f
id
en
tial
in
f
o
r
m
atio
n
f
r
o
m
th
e
v
ictim
'
s
p
er
s
o
n
al
co
m
p
u
ter
(
PC
)
an
d
ca
r
r
y
o
u
t
o
th
er
n
ef
a
r
io
u
s
ac
tiv
ities
.
Ma
c
h
i
n
e
lea
r
n
i
n
g
m
e
th
o
d
s
h
av
e
b
ec
o
m
e
i
n
c
r
e
asi
n
g
ly
p
o
p
u
la
r
f
o
r
m
alw
ar
e
d
ete
cti
o
n
i
n
cl
o
u
d
en
v
i
r
o
n
m
e
n
ts
,
as
w
ell
as
in
v
ar
i
o
u
s
o
t
h
e
r
ar
ea
s
[
2
7
]
–
[
3
2
]
.
Su
p
p
o
r
t
v
ec
t
o
r
m
ac
h
i
n
es
(
SV
M)
,
d
e
cisi
o
n
t
r
e
es
(
DT
)
,
r
a
n
d
o
m
f
o
r
est
(
R
F
)
,
a
n
d
n
ai
v
e
B
a
y
es
(
NB
)
ar
e
s
o
m
e
o
f
th
e
m
ac
h
i
n
e
l
ea
r
n
in
g
t
ec
h
n
i
q
u
es
t
h
a
t
h
a
v
e
b
ee
n
u
s
e
d
to
s
u
c
ce
s
s
f
u
ll
y
f
i
n
d
m
al
w
ar
e
i
n
p
r
a
cti
ca
l
ap
p
l
ica
ti
o
n
s
[
3
0
]
,
[
3
3
]
,
[
3
4
]
.
T
h
e
s
iz
e
o
f
th
e
t
r
ai
n
i
n
g
d
atas
et
a
n
d
th
e
n
u
m
b
e
r
o
f
f
ea
t
u
r
es
t
h
at
c
a
n
b
e
r
e
tr
ie
v
e
d
f
r
o
m
it
,
h
o
w
ev
er
,
d
et
e
r
m
i
n
e
h
o
w
ac
cu
r
ate
t
h
es
e
alg
o
r
it
h
m
s
ar
e.
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.
14
,
No
.
6
,
Dec
em
b
e
r
20
24
:
6
6
4
2
-
6
6
5
5
6644
I
n
a
r
ec
en
t
s
tu
d
y
[
2
3
]
,
an
a
p
p
r
o
ac
h
b
ased
o
n
co
n
v
o
lu
ti
o
n
al
n
eu
r
al
n
etwo
r
k
s
(
C
NN)
f
o
r
v
ir
u
s
d
etec
tio
n
in
clo
u
d
p
latf
o
r
m
s
h
as
b
ee
n
s
u
g
g
ested
.
Fo
r
id
en
t
if
y
in
g
m
alwa
r
e,
th
e
au
th
o
r
s
u
s
ed
a
C
NN
m
o
d
el
with
two
d
im
e
n
s
io
n
s
(
2
D)
a
n
d
a
C
NN
m
o
d
el
with
th
r
ee
d
i
m
en
s
io
n
s
(
3
D)
.
Als
o
,
t
h
ey
co
n
d
u
cted
ex
p
e
r
im
en
ts
b
y
in
s
tallin
g
v
ar
io
u
s
in
f
ec
tio
n
s
o
n
s
im
u
lated
PC
s
an
d
ac
h
iev
ed
7
9
%
ac
cu
r
ac
y
with
th
e
2
D
C
NN
m
o
d
el
an
d
9
0
%
ac
cu
r
ac
y
with
th
e
3
D
C
NN
m
o
d
el.
Ho
wev
er
,
th
is
s
tu
d
y
d
id
n
o
t
c
o
m
p
ar
e
C
NN
to
c
o
n
v
en
tio
n
al
m
ac
h
in
e
lear
n
in
g
tech
n
iq
u
es; it o
n
ly
f
o
cu
s
ed
o
n
C
NN
[
2
9
]
.
T
h
e
ef
f
icac
y
o
f
v
ar
io
u
s
m
ac
h
in
e
lear
n
in
g
m
et
h
o
d
s
,
s
u
ch
as
DT
an
d
SVM,
was
p
r
o
v
e
d
u
s
in
g
a
s
y
s
tem
b
ased
o
n
m
ac
h
in
e
le
ar
n
in
g
f
o
r
m
alwa
r
e
d
etec
tio
n
[
3
0
]
.
T
h
e
r
esear
c
h
er
s
em
p
lo
y
ed
th
e
C
u
ck
o
o
San
d
b
o
x
to
test
d
if
f
e
r
en
t
m
al
war
e
k
in
d
s
in
a
s
im
u
lated
en
v
ir
o
n
m
en
t,
p
r
o
d
u
cin
g
a
n
an
aly
s
is
r
ep
o
r
t
d
ep
en
d
in
g
o
n
h
o
w
th
e
m
alwa
r
e
s
am
p
les
ac
t
in
th
e
s
ettin
g
.
T
h
e
C
u
ck
o
o
San
d
b
o
x
h
as
b
ee
n
u
tili
ze
d
b
y
n
u
m
e
r
o
u
s
r
esear
ch
er
s
to
ex
am
in
e
m
alw
ar
e
[
3
0
]
,
[
3
5
]
–
[
3
8
]
e
x
am
in
ed
th
e
v
iab
ilit
y
o
f
u
s
in
g
C
NN,
R
F,
an
d
k
-
n
ea
r
est
n
eig
h
b
o
r
s
(
KNN)
m
o
d
els
an
d
u
s
ed
f
ea
tu
r
es
f
r
o
m
ap
p
licatio
n
p
r
o
g
r
am
m
in
g
in
ter
f
ac
e
(
API
)
ca
lls
to
id
en
tif
y
m
alwa
r
e.
Ad
d
itio
n
ally
,
W
ats
o
n
et
a
l.
[
3
1
]
u
tili
ze
d
th
e
r
an
d
o
m
f
o
r
est
class
if
ier
to
k
ee
p
tr
ac
k
o
f
a
v
ir
tu
a
l
m
ac
h
in
e'
s
p
r
o
ce
s
s
ac
tiv
ity
.
W
ek
a
h
as
b
ee
n
u
s
ed
in
s
tu
d
i
es
o
n
m
alwa
r
e
d
etec
tio
n
b
y
n
u
m
er
o
u
s
s
ec
u
r
ity
r
esear
ch
er
s
,
in
clu
d
in
g
[
2
4
]
,
[
2
6
]
,
[
3
9
]
,
[
4
0
]
.
On
th
e
o
th
er
h
an
d
,
s
ev
er
al
r
esear
c
h
er
s
,
in
clu
d
in
g
[
4
0
]
–
[
4
3
]
h
av
e
u
tili
ze
d
Py
th
o
n
in
th
eir
s
tu
d
ies.
Ku
m
ar
et
a
l.
[
4
4
]
r
e
p
o
r
ted
ly
d
em
o
n
s
tr
ated
a
m
ac
h
in
e
lear
n
in
g
-
b
ased
ap
p
r
o
ac
h
to
ac
cu
r
ately
an
d
ef
f
icien
tly
d
ete
r
m
in
e
w
h
eth
er
a
s
am
p
le
is
m
alwa
r
e
o
r
b
en
ig
n
.
W
ek
a
was
u
tili
ze
d
b
y
th
e
a
u
th
o
r
s
to
im
p
lem
e
n
t
s
ix
class
if
icatio
n
alg
o
r
ith
m
s
.
As a
r
esu
lt o
f
ap
p
ly
in
g
1
0
-
f
o
ld
cr
o
s
s
-
v
alid
atio
n
,
th
e
r
an
d
o
m
f
o
r
est cla
s
s
if
ier
wa
s
ab
le
to
ac
h
iev
e
a
n
ac
cu
r
ac
y
o
f
9
8
.
4
%.
J
4
5
,
lo
g
is
tic
m
o
d
el
tr
ee
(
L
MT
)
,
n
ai
v
e
B
ay
es
(
NB
)
,
r
an
d
o
m
f
o
r
est
(
R
F),
m
u
ltil
ay
er
p
er
ce
p
tr
o
n
(
ML
P)
alg
o
r
ith
m
,
r
an
d
o
m
tr
ee
(
R
T
)
,
r
ed
u
ce
d
er
r
o
r
p
r
u
n
i
n
g
tr
ee
(
R
E
PTr
ee
)
,
B
ag
g
in
g
,
Ad
aBo
o
s
t,
KStar
,
s
im
p
le
lo
g
is
tic,
lazy
k
-
n
ea
r
est
n
eig
h
b
o
r
(
I
B
K)
,
lo
ca
lly
weig
h
ted
lear
n
in
g
(
L
W
L
)
,
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM)
,
an
d
r
ad
ial
b
asis
f
u
n
cti
o
n
(
R
B
F)
n
etwo
r
k
a
r
e
am
o
n
g
th
e
f
if
teen
d
if
f
er
e
n
t
m
ac
h
in
e
l
ea
r
n
in
g
al
g
o
r
ith
m
s
u
s
ed
,
th
is
r
ep
o
r
t
[
2
6
]
p
r
o
v
id
e
d
a
co
m
p
ar
ativ
e
an
aly
s
is
o
f
m
a
lwar
e
id
en
tific
atio
n
.
T
h
e
ex
p
e
r
im
en
t
was
ca
r
r
ied
o
u
t
in
th
e
W
ek
a
en
v
i
r
o
n
m
e
n
t
u
s
in
g
t
h
e
class
if
icatio
n
o
f
m
alwa
r
e
with
p
o
r
tab
le
ex
ec
u
tab
le
(
PE)
h
ea
d
er
s
(
C
laM
P)
d
ataset.
T
h
e
ac
cu
r
ac
y
r
ate
is
s
til
l
p
o
o
r
ev
en
if
th
e
RF
m
eth
o
d
p
er
f
o
r
m
s
b
et
ter
th
an
th
e
o
th
er
tech
n
iq
u
es.
I
n
ac
co
r
d
a
n
ce
with
a
s
tu
d
y
[
2
4
]
,
th
is
s
tu
d
y
ex
a
m
in
ed
eig
h
t
m
ac
h
in
e
lear
n
in
g
class
if
ier
s
to
d
etec
t
T
r
o
jan
h
o
r
s
es
in
clo
u
d
en
v
ir
o
n
m
en
ts
.
I
n
v
esti
g
atio
n
s
e
x
a
m
in
in
g
t
h
e
ac
cu
r
ac
y
o
f
th
e
clo
u
d
T
r
o
jan
h
o
r
s
e
d
etec
tio
n
r
ate
h
av
e
b
ee
n
co
n
d
u
cted
u
s
in
g
th
e
d
ata
m
in
in
g
to
o
l
W
ek
a.
T
h
e
m
o
s
t
ef
f
ec
tiv
e
class
if
ier
s
f
o
r
id
en
tify
in
g
T
r
o
jan
h
o
r
s
es in
a
clo
u
d
-
b
ased
s
ettin
g
h
av
e
b
ee
n
s
h
o
wn
to
b
e
th
e
s
eq
u
en
tial m
i
n
im
al
o
p
tim
izatio
n
(
SMO)
an
d
m
u
ltil
ay
er
p
er
ce
p
tr
o
n
b
ased
o
n
th
e
s
tu
d
ies
th
at
h
av
e
b
ee
n
d
o
n
e.
W
ith
a
9
5
.
8
6
%
ac
cu
r
ac
y
r
ate,
SMO
an
d
m
u
ltil
ay
er
p
e
r
c
ep
tr
o
n
h
av
e
t
h
e
g
r
ea
test
r
ate.
Acc
o
r
d
in
g
to
th
eir
r
esear
ch
p
a
p
er
,
Kan
a
k
er
et
a
l.
[
2
4
]
p
r
o
p
o
s
ed
a
h
y
b
r
id
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
th
at
m
er
g
ed
KNN
an
d
NB
f
o
r
d
etec
tin
g
T
r
o
ja
n
h
o
r
s
es,
an
d
th
ey
ex
ec
u
te
d
th
eir
wo
r
k
u
s
in
g
Py
th
o
n
in
th
e
C
o
lab
en
v
ir
o
n
m
en
t,
wh
er
e
ce
r
tain
m
ac
h
in
e
-
lear
n
in
g
tec
h
n
iq
u
es
ar
e
test
ed
with
9
9
.
5
%
ac
cu
r
ac
y
.
T
h
e
em
p
ir
ical
r
esu
lts
o
f
th
is
s
tu
d
y
d
em
o
n
s
tr
ate
th
at
th
e
h
y
b
r
i
d
alg
o
r
ith
m
is
th
e
m
o
s
t
ef
f
e
ctiv
e
alg
o
r
ith
m
f
o
r
d
etec
tin
g
a
T
r
o
jan
h
o
r
s
e.
Seth
i
et
a
l.
[
4
5
]
d
ev
elo
p
a
m
ac
h
in
e
lear
n
in
g
-
b
ased
m
al
war
e
an
aly
s
is
f
r
am
ewo
r
k
f
o
r
r
ap
id
an
d
ac
cu
r
ate
m
alwa
r
e
d
etec
tio
n
.
A
Py
th
o
n
p
ac
k
ag
e
was
d
ev
el
o
p
ed
f
o
r
f
ea
tu
r
e
s
elec
tio
n
,
f
e
atu
r
e
ex
tr
ac
tio
n
,
an
d
th
e
b
u
ild
in
g
o
f
tr
ain
in
g
an
d
test
in
g
d
atasets
.
Fo
r
th
e
d
etec
tio
n
an
d
class
if
icatio
n
o
f
th
e
p
r
o
v
id
ed
d
ataset,
th
e
y
m
ad
e
u
s
e
o
f
n
u
m
er
o
u
s
m
ac
h
in
e
lear
n
in
g
tech
n
i
q
u
es
o
f
f
er
ed
b
y
th
e
Py
th
o
n
Scik
it
-
le
ar
n
f
r
am
ewo
r
k
.
A
d
ec
is
io
n
tr
ee
with
an
ac
c
u
r
ac
y
o
f
9
9
.
3
7
% h
as a
h
ig
h
d
etec
tio
n
r
ate,
ac
co
r
d
in
g
to
ex
p
er
im
en
tal
r
esu
lts
.
I
n
a
p
u
b
licatio
n
[
4
6
]
,
a
u
n
iq
u
e
m
ac
h
in
e
lear
n
i
n
g
a
p
p
r
o
ac
h
is
s
u
g
g
ested
an
d
im
p
lem
en
ted
in
Py
th
o
n
an
d
MA
T
L
AB
to
r
ec
o
g
n
ize
t
h
e
Mir
ai
m
alwa
r
e.
T
h
is
s
tu
d
y
co
m
p
ar
es
th
e
p
er
f
o
r
m
an
ce
o
f
ar
tific
ial
n
eu
r
al
n
etwo
r
k
(
ANN)
an
d
R
F
m
o
d
e
ls
u
s
in
g
a
d
ataset
cr
ea
ted
b
y
co
m
b
in
in
g
th
e
b
en
ig
n
an
d
m
a
licio
u
s
d
atasets
f
o
r
s
ev
en
in
ter
n
et
o
f
th
in
g
s
(
I
o
T
)
d
ev
ices.
B
ec
au
s
e
o
f
th
e
g
r
ea
te
s
t
p
er
f
o
r
m
an
ce
,
wh
ich
h
ad
an
ac
cu
r
ac
y
o
f
9
2
.
8
%,
th
e
r
esu
lts
ar
e
d
ee
m
ed
t
o
b
e
r
e
liab
le
an
d
ac
cu
r
ate.
T
h
e
f
ir
s
t
s
tep
in
th
e
m
alw
ar
e
d
etec
tio
n
p
r
o
ce
d
u
r
e
is
m
alwa
r
e
an
aly
s
is
.
Ma
lwar
e
ca
n
n
o
t
b
e
d
is
co
v
er
ed
u
n
til
it
is
o
b
s
er
v
ed
an
d
its
b
eh
a
v
io
r
is
u
n
d
er
s
to
o
d
.
As
a
r
esu
lt,
it
is
ea
s
y
to
ad
d
s
ec
u
r
ity
m
ea
s
u
r
es
to
m
alwa
r
e
d
etec
to
r
s
.
T
h
er
e
ar
e
two
ca
teg
o
r
ies
o
f
m
al
war
e
an
aly
s
is
tech
n
iq
u
es:
s
tatic
an
d
d
y
n
am
ic,
d
ep
en
d
i
n
g
o
n
th
e
am
o
u
n
t
o
f
ti
m
e
an
d
tech
n
o
lo
g
y
r
e
q
u
ir
ed
to
co
m
p
lete
th
e
in
v
esti
g
atio
n
.
W
ith
o
u
t
ac
tu
ally
r
u
n
n
in
g
th
e
m
,
s
tatic
an
aly
s
is
ex
am
in
es
m
alwa
r
e
ex
ec
u
tab
le
f
iles
in
a
co
n
tr
o
lled
en
v
ir
o
n
m
en
t
[
4
7
]
–
[
4
9
]
.
Static
d
ata
is
ex
tr
ac
ted
f
r
o
m
t
h
e
c
o
d
e
an
d
u
s
ed
to
d
eter
m
in
e
w
h
eth
er
th
e
s
o
f
twar
e
co
n
tain
s
m
alicio
u
s
co
d
e
[
4
7
]
.
T
h
e
ex
ec
u
tab
le
f
ile
co
n
tain
s
a
lo
t
o
f
s
tatic
f
ea
tu
r
es,
l
ik
e
m
em
o
r
y
co
m
p
ac
tn
ess
.
T
o
d
o
s
tatic
an
aly
s
is
,
a
v
ar
iety
o
f
to
o
ls
ca
n
b
e
em
p
lo
y
e
d
,
i
n
clu
d
in
g
d
ec
o
m
p
iler
s
,
d
is
ass
em
b
ler
s
,
s
o
u
r
ce
co
d
e
an
aly
ze
r
s
,
an
d
d
eb
u
g
g
e
r
s
[
4
7
]
.
Fu
r
th
er
m
o
r
e,
o
n
ly
k
n
o
wn
m
alwa
r
e
s
ig
n
atu
r
es
ca
n
b
e
r
e
liab
ly
d
etec
ted
v
i
a
s
tatic
an
aly
s
is
.
B
ec
au
s
e
o
f
th
i
s
,
it
co
u
l
d
r
a
r
ely
b
e
u
n
ab
le
to
ass
es
s
m
alwa
r
e
s
ig
n
atu
r
es
t
h
a
t
ar
e
u
n
k
n
o
wn
an
d
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
A
co
mp
a
r
a
tive
s
tu
d
y
o
f m
a
ch
i
n
e
lea
r
n
in
g
to
o
ls
fo
r
d
etec
tin
g
Tr
o
ja
n
h
o
r
s
e
…
(
Ha
s
a
n
K
a
n
a
ke
r
)
6645
n
o
t
in
its
d
atab
ase.
R
eg
u
lar
u
p
d
atin
g
a
n
d
p
r
o
d
u
ctio
n
b
y
h
u
m
an
s
with
s
p
ec
if
ic
k
n
o
wled
g
e
a
r
e
r
e
q
u
ir
ed
f
o
r
th
e
s
ig
n
atu
r
es
u
s
ed
in
s
tan
d
ar
d
s
tatic
an
aly
s
is
[
4
]
,
[
7
]
,
[
2
4
]
,
[
3
9
]
.
I
t
is
n
o
t
em
p
lo
y
e
d
in
th
is
s
tu
d
y
d
u
e
to
th
e
r
ea
s
o
n
s
m
en
tio
n
ed
ab
o
v
e.
Dy
n
am
ic
an
aly
s
is
ev
alu
ates
m
alwa
r
e
b
eh
av
io
r
in
a
d
y
n
am
ic
ally
co
n
tr
o
lled
e
n
v
ir
o
n
m
en
t.
I
n
ad
d
itio
n
to
s
tar
tin
g
i
n
p
r
iv
ileg
ed
m
o
d
e,
th
e
v
ir
u
s
m
o
d
if
ies
th
e
r
e
g
is
tr
y
as
it
r
u
n
s
.
W
h
en
th
e
m
alwa
r
e
s
witch
es
to
p
r
iv
ileg
e
m
o
d
e,
t
h
e
o
p
er
atin
g
s
y
s
tem
will
b
e
co
m
p
letely
u
n
d
er
its
co
n
tr
o
l.
All
r
eso
u
r
ce
s
ar
e
co
m
p
letely
u
n
d
e
r
th
e
co
n
tr
o
l
o
f
d
y
n
am
ic
an
aly
tic
s
o
f
twar
e.
I
t
is
ab
le
to
f
u
n
ctio
n
in
a
s
ec
u
r
e
en
v
ir
o
n
m
e
n
t
as
a
r
esu
lt.
T
h
e
p
r
o
g
r
a
m
ca
n
o
p
er
ate
in
d
eb
u
g
m
o
d
e
an
d
alter
c
o
m
p
u
ter
r
eg
is
tr
y
k
ey
s
in
a
co
n
tr
o
lle
d
en
v
ir
o
n
m
en
t
T
h
e
d
y
n
am
ic
en
v
i
r
o
n
m
e
n
t
r
etu
r
n
s
to
th
e
in
itial
s
n
ap
s
h
o
t
th
at
wa
s
tak
en
at
th
e
s
tar
t
o
f
en
v
ir
o
n
m
en
t
d
ev
elo
p
m
en
t
af
ter
ex
ec
u
tin
g
a
n
d
a
n
aly
zin
g
a
m
alwa
r
e
s
am
p
le.
B
ef
o
r
e
l
o
o
k
in
g
at
an
o
th
e
r
m
alwa
r
e
s
am
p
le,
th
is
m
ak
es
s
u
r
e
th
e
ar
ea
is
clea
n
.
B
y
s
tr
en
g
th
en
in
g
th
r
ea
t
id
e
n
tific
atio
n
an
d
p
r
ed
ictio
n
,
a
u
to
m
atin
g
an
d
s
tr
ea
m
lin
in
g
t
h
r
ea
t
r
esp
o
n
s
e,
an
d
f
o
r
tify
in
g
an
d
ad
ap
tin
g
s
ec
u
r
ity
m
ea
s
u
r
es,
th
e
lite
r
atu
r
e
in
th
e
f
ield
o
f
m
ac
h
in
e
lear
n
in
g
an
d
cy
b
er
s
ec
u
r
ity
ad
v
an
ce
s
o
u
r
u
n
d
er
s
tan
d
in
g
an
d
ca
p
ac
ities
.
I
t
ad
v
an
ce
s
th
e
d
is
cip
lin
e
b
y
in
tr
o
d
u
cin
g
n
o
v
el
ap
p
r
o
ac
h
es
an
d
o
f
f
er
i
n
g
u
s
ef
u
l
ca
s
e
s
tu
d
ies
th
at
illu
s
tr
ate
th
eir
ef
f
ec
tiv
en
ess
in
th
e
r
ea
l
w
o
r
ld
.
T
h
is
h
elp
s
to
b
etter
p
r
o
tect
d
ig
ital in
f
r
astru
c
tu
r
es b
y
p
r
o
v
id
i
n
g
b
o
th
th
e
o
r
e
tical
u
n
d
er
s
tan
d
in
g
an
d
u
s
ef
u
l
to
o
ls
.
3.
RE
S
E
ARCH
M
E
T
H
O
D
I
n
o
r
d
e
r
t
o
c
o
m
p
a
r
e
t
h
e
e
f
f
e
c
t
i
v
e
n
e
s
s
o
f
W
e
k
a
a
n
d
P
y
t
h
o
n
C
o
l
a
b
i
n
d
e
t
e
c
t
i
n
g
T
r
o
j
a
n
h
o
r
s
e
i
n
f
e
c
t
i
o
n
s
,
w
e
wi
l
l
u
s
e
a
d
a
t
as
e
t
c
o
n
s
is
ti
n
g
o
f
k
n
o
w
n
m
a
l
w
a
r
e
a
n
d
b
e
n
i
g
n
f
i
l
es
.
W
e
u
ti
l
iz
e
d
s
e
v
e
r
a
l
t
o
o
ls
,
i
n
cl
u
d
i
n
g
S
a
n
d
b
o
x
,
N
E
W
T
P
r
o
,
P
r
o
m
i
s
cD
e
t
e
c
t
,
P
r
o
c
ess
E
x
p
l
o
r
e
r
,
P
o
r
tM
o
n
,
a
n
d
W
i
r
e
s
h
a
r
k
.
D
y
n
a
m
i
c
a
n
a
l
y
s
is
w
a
s
u
s
e
d
i
n
t
h
i
s
s
t
u
d
y
s
i
n
ce
i
t
i
s
m
o
r
e
e
f
f
e
ct
i
v
e
,
p
o
w
e
r
f
u
l
,
a
n
d
r
e
l
ia
b
l
e
.
T
h
e
a
n
a
l
y
s
is
w
as
c
a
r
r
i
e
d
o
u
t
i
n
f
o
u
r
s
t
a
g
e
s
:
−
Data
c
o
llectio
n
:
W
e
g
at
h
er
ed
b
en
ig
n
an
d
T
r
o
jan
h
o
r
s
e
s
am
p
les
f
r
o
m
V
ir
u
s
S
h
a
r
e.
co
m
a
n
d
Vir
u
s
T
o
tal
[
4
]
,
[
2
4
]
,
[
3
0
]
,
[
3
9
]
,
[
5
0
]
,
[
5
1
]
.
On
e
o
f
th
e
lar
g
est
p
u
b
licly
av
aila
b
le
v
ir
u
s
r
ep
o
s
ito
r
ies
o
n
th
e
in
ter
n
et
is
ca
lled
Vir
u
s
Sh
ar
e,
wh
er
ea
s
Vir
u
s
T
o
tal
s
er
v
es
as
a
r
ep
o
s
ito
r
y
f
o
r
m
alwa
r
e
s
am
p
les.
W
e
o
b
tain
ed
3
,
0
0
0
ex
ec
u
tab
le
T
r
o
jan
h
o
r
s
es
f
r
o
m
s
am
p
les
m
ad
e
av
ailab
le
o
n
Vir
u
s
T
o
tal
an
d
Vir
u
s
Sh
ar
e,
wh
ich
wer
e
d
etec
ted
b
y
p
o
wer
f
u
l
an
tiv
ir
u
s
s
y
s
tem
s
s
u
ch
as
Av
ast,
F
-
Sec
u
r
e,
Kasp
er
s
k
y
,
C
o
m
o
d
o
,
Av
i
r
a,
B
itd
ef
en
d
er
,
an
d
o
th
e
r
s
.
Ma
ch
in
e
lear
n
in
g
t
ec
h
n
iq
u
es will b
e
ap
p
lied
o
n
th
is
d
ataset
f
o
r
f
u
r
th
er
m
alwa
r
e
r
esear
ch
.
−
Data
a
n
aly
s
is
:
A
co
n
tr
o
lled
lab
en
v
ir
o
n
m
en
t
is
n
ec
ess
ar
y
f
o
r
T
r
o
ja
n
h
o
r
s
e
an
aly
s
is
.
T
o
estab
lis
h
a
co
m
p
letely
co
n
tr
o
lled
,
s
eg
r
e
g
ated
en
v
ir
o
n
m
en
t
a
n
d
p
r
ev
en
t
th
e
s
p
r
ea
d
o
f
T
r
o
jan
h
o
r
s
e
s
,
th
e
p
h
y
s
ical
n
etwo
r
k
lin
k
m
u
s
t
b
e
s
ev
er
ed
.
Ho
wev
er
,
with
o
u
t
a
n
etwo
r
k
co
n
n
ec
tio
n
it
is
im
p
o
s
s
ib
le
to
d
o
d
y
n
am
ic
an
aly
s
is
o
n
T
r
o
jan
h
o
r
s
e
s
am
p
les.
A
v
ir
tu
al
clo
u
d
e
n
v
ir
o
n
m
en
t
is
cr
ea
ted
u
s
in
g
v
ir
tu
aliza
tio
n
tech
n
o
lo
g
ies,
an
d
is
u
s
ed
to
lin
k
a
d
is
s
im
ilar
s
er
v
er
(
m
o
n
ito
r
in
g
h
o
s
t
an
d
th
e
attac
k
er
)
to
t
h
e
clo
u
d
.
T
h
is
co
n
tr
o
lled
lab
ar
ch
itectu
r
e
is
co
m
p
ar
ab
le
to
th
at
u
s
ed
by
[
2
]
,
[
2
4
]
,
[
2
5
]
,
[
3
0
]
−
Featu
r
e
e
x
tr
ac
tio
n
:
T
h
is
s
tag
e
in
v
o
lv
es
d
eter
m
in
in
g
th
e
s
ig
n
if
ican
ce
o
f
th
e
d
ataset'
s
cu
r
r
en
t
f
ea
tu
r
es.
I
t
r
etain
s
th
e
cr
u
cial
co
m
p
o
n
e
n
ts
wh
ile
elim
in
atin
g
th
e
s
u
p
er
f
lu
o
u
s
o
n
es
[
5
2
]
.
Featu
r
e
e
x
tr
ac
ti
o
n
ca
n
b
e
u
s
ed
to
ac
h
iev
e
h
i
g
h
er
ac
c
u
r
ac
y
r
at
es.
−
T
esti
n
g
f
r
am
ewo
r
k
:
I
n
th
is
s
tag
e,
we
p
er
f
o
r
m
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
o
n
th
e
d
ataset
u
s
in
g
W
ek
a
an
d
Py
th
o
n
in
d
e
p
en
d
en
tly
af
ter
f
e
atu
r
e
ex
tr
ac
tio
n
.
4.
DATA AN
AL
Y
SI
S
Fig
u
r
e
1
d
em
o
n
s
tr
ates
a
c
o
n
tr
o
lled
la
b
s
etu
p
[
2
]
,
[
2
4
]
wh
er
e
a
r
ea
l
d
ev
ice
is
u
s
ed
to
h
o
s
t
a
v
ir
tu
alize
d
v
er
s
io
n
o
f
th
e
clo
u
d
,
is
o
lated
f
r
o
m
th
e
in
ter
n
et.
T
h
e
attac
k
er
in
itiates
th
e
attac
k
f
r
o
m
o
u
ts
id
e
t
h
e
clo
u
d
s
itu
atio
n
,
wh
e
r
ea
s
th
e
s
er
v
er
in
th
e
clo
u
d
h
as
m
o
n
i
to
r
in
g
to
o
ls
in
s
talled
to
tr
ac
k
ac
tiv
ities
s
u
ch
a
s
p
r
o
ce
s
s
m
o
n
ito
r
i
n
g
,
f
ile
m
o
n
ito
r
in
g
,
n
etwo
r
k
m
o
n
ito
r
i
n
g
,
an
d
r
eg
is
tr
y
m
o
n
ito
r
in
g
.
D
y
n
am
ic
an
aly
s
is
an
d
b
eh
av
io
r
m
o
n
ito
r
in
g
th
r
o
u
g
h
th
e
C
u
ck
o
o
San
d
b
o
x
wer
e
u
s
ed
to
ass
ess
th
e
d
etec
tio
n
o
f
T
r
o
jan
h
o
r
s
es
in
th
e
co
n
tr
o
lled
lab
en
v
ir
o
n
m
en
t
.
A
s
im
ilar
ap
p
r
o
ac
h
h
as
b
ee
n
ad
o
p
ted
in
[
2
]
,
[
2
4
]
,
[
3
9
]
,
[
4
5
]
.
T
h
e
ac
c
u
r
ac
y
o
f
d
etec
tio
n
was
ca
lcu
lated
s
ep
ar
ately
u
s
in
g
two
d
if
f
er
e
n
t
e
n
v
ir
o
n
m
en
ts
:
Py
th
o
n
C
o
lab
an
d
th
e
W
ek
a
d
ata
m
in
in
g
to
o
l.
T
h
e
d
y
n
am
ic
a
n
d
C
u
ck
o
o
San
d
b
o
x
an
aly
s
is
h
av
e
b
ee
n
p
er
f
o
r
m
ed
o
n
all
f
iles
.
E
ac
h
f
ile
i
s
ex
ec
u
ted
an
d
its
r
u
n
tim
e
o
p
er
atio
n
s
ar
e
ex
am
in
e
d
to
d
eter
m
i
n
e
th
e
b
e
h
av
io
r
o
f
th
e
v
ir
u
s
in
a
n
ewly
in
s
talled
o
p
er
atin
g
s
y
s
tem
.
T
h
e
T
r
o
jan
h
o
r
s
e
s
a
m
p
les
ar
e
test
ed
in
a
co
n
tr
o
ll
ed
en
v
ir
o
n
m
en
t
m
o
d
e
f
o
r
d
y
n
am
ic
an
aly
s
is
.
On
ce
th
e
test
in
g
is
c
o
m
p
lete,
th
e
Dee
p
f
r
ee
ze
co
m
p
u
ter
p
r
o
g
r
am
is
u
s
ed
to
r
esto
r
e
th
e
co
n
tr
o
lled
en
v
ir
o
n
m
e
n
t
to
its
o
r
ig
in
al
u
n
in
f
ec
ted
s
tate.
Du
r
i
n
g
th
e
d
y
n
am
ic
an
al
y
s
is
s
tag
e,
th
e
b
e
h
av
io
r
o
f
ea
ch
in
jecte
d
T
r
o
ja
n
s
am
p
le
is
o
b
s
er
v
ed
o
n
th
e
h
o
s
t
m
ac
h
in
e.
T
h
e
m
o
n
ito
r
in
g
to
o
ls
u
s
e
d
in
th
is
s
tag
e
in
clu
d
e
lis
ten
in
g
p
o
r
ts
,
r
eg
is
tr
y
,
n
etwo
r
k
,
r
a
n
d
o
m
-
ac
ce
s
s
m
em
o
r
y
(
R
AM
)
,
f
iles
,
tr
an
s
m
i
s
s
io
n
co
n
tr
o
l
p
r
o
to
co
l
(
T
C
P)
,
p
r
o
ce
s
s
es,
an
d
d
y
n
am
ic
-
lin
k
lib
r
a
r
ies
(
DL
L
s
)
.
An
y
s
u
s
p
icio
u
s
b
e
h
av
io
r
d
et
ec
ted
b
y
th
ese
to
o
ls
,
th
e
s
am
p
le
is
id
en
tifie
d
a
n
d
lab
eled
as T
r
o
jan
.
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.
14
,
No
.
6
,
Dec
em
b
e
r
20
24
:
6
6
4
2
-
6
6
5
5
6646
Fig
u
r
e
1
.
C
o
n
tr
o
lled
lab
o
r
ato
r
y
s
etu
p
[
2
]
,
[
2
4
]
4
.
1
.
F
e
a
t
ure
e
x
t
r
a
ct
io
n
T
h
e
p
r
o
ce
s
s
o
f
ev
alu
atin
g
an
e
x
is
tin
g
f
ea
tu
r
e'
s
r
elev
an
ce
in
a
d
ataset
is
ca
lled
f
ea
tu
r
e
e
x
tr
a
ctio
n
.
T
h
e
s
alien
t
ch
ar
ac
ter
is
tics
ar
e
k
e
p
t
wh
ile
th
e
ir
r
elev
an
t
o
n
es
ar
e
r
e
m
o
v
e
d
[
2
4
]
.
Selectin
g
f
ea
tu
r
es
aid
s
in
im
p
r
o
v
in
g
ac
c
u
r
ac
y
r
ates.
Fo
r
th
is
s
tu
d
y
,
we
h
av
e
co
llected
3
0
0
0
T
r
o
jan
h
o
r
s
e
s
am
p
les
f
r
o
m
Vir
u
s
T
o
tal
an
d
Vir
u
s
S
h
ar
e.
T
h
e
p
r
o
ce
s
s
o
f
f
ea
tu
r
e
ex
tr
ac
tio
n
em
p
lo
y
e
d
th
e
C
u
ck
o
o
San
d
b
o
x
an
aly
s
is
r
ep
o
r
t
a
n
d
d
y
n
a
m
ic
an
aly
s
is
.
T
h
r
o
u
g
h
an
aly
s
is
ca
r
r
ied
o
u
t
in
a
c
o
n
tr
o
lled
lab
e
n
v
ir
o
n
m
en
t,
th
e
q
u
alities
th
at
wer
e
r
ec
o
g
n
ize
d
as
p
r
o
p
er
ties
o
f
ea
ch
T
r
o
ja
n
h
o
r
s
e
wer
e
id
en
tifie
d
,
an
d
a
r
elev
a
n
t
d
ataset
was
estab
lis
h
ed
.
T
h
i
s
p
h
ase
is
f
in
is
h
ed
with
th
e
cr
ea
tio
n
o
f
an
ex
t
en
s
iv
e
o
u
tp
u
t
an
aly
s
is
r
ep
o
r
t.
T
h
is
r
ep
o
r
t
co
n
tain
s
d
etails
o
n
th
e
cr
ea
tio
n
,
m
o
d
if
icatio
n
,
d
eletio
n
,
an
d
a
cc
ess
to
f
ile
s
an
d
r
eg
is
tr
y
k
ey
s
as
well
as
DL
L
s
,
R
AM
,
an
d
n
etwo
r
k
in
g
p
r
o
to
co
ls
.
Nin
e
f
ea
tu
r
es
m
ak
e
u
p
th
e
o
u
tp
u
t
f
ile,
wh
ich
h
as
b
ee
n
tr
an
s
lated
to
th
e
s
u
ita
b
le
f
ile
f
o
r
m
at
ar
f
f
.
T
h
e
o
u
tp
u
t
d
ata
is
en
co
d
e
d
i
n
th
is
f
ile
f
o
r
m
at
s
o
th
at
it
m
ay
b
e
u
s
ed
as
in
p
u
t
d
ata
in
th
e
m
ac
h
in
e
lear
n
i
n
g
s
im
u
latio
n
en
v
ir
o
n
m
en
ts
o
f
P
y
th
o
n
C
o
lab
an
d
W
E
KA.
T
h
e
tr
o
jan
h
o
r
s
e
d
ataset
was
s
u
cc
ess
f
u
lly
class
if
ied
in
th
is
s
tu
d
y
u
s
in
g
a
1
0
-
f
o
l
d
cr
o
s
s
-
v
alid
atio
n
alo
n
g
with
a
n
u
m
b
er
o
f
class
if
ier
s
,
in
clu
d
in
g
NB
,
I
B
K,
RF
,
J
4
8
,
an
d
r
eg
r
ess
io
n
.
4
.
2
.
T
esting
f
ra
m
ewo
r
k
T
h
e
f
r
am
ewo
r
k
d
e
p
icted
in
Fig
u
r
e
2
is
d
e
v
elo
p
ed
u
s
in
g
Py
t
h
o
n
in
th
e
C
o
lab
en
v
ir
o
n
m
en
t
an
d
W
ek
a
m
ac
h
in
e
lear
n
in
g
tech
n
iq
u
e,
wh
er
e
ce
r
tain
m
ac
h
in
e
-
lear
n
i
n
g
alg
o
r
ith
m
s
a
r
e
test
ed
a
n
d
im
p
lem
en
ted
.
T
h
e
T
r
o
jan
h
o
r
s
e
d
ataset
is
class
i
f
ied
u
s
in
g
class
if
ier
s
in
clu
d
in
g
1
0
-
f
o
ld
cr
o
s
s
-
v
alid
atio
n
,
J
4
8
,
RF
,
MLP
,
I
B
K,
r
eg
r
ess
io
n
,
NB
,
an
d
o
th
er
s
.
Fig
u
r
e
2
.
Fra
m
ewo
r
k
f
o
r
test
in
g
tr
o
jan
attac
k
d
etec
tio
n
5.
E
XP
E
R
I
M
E
N
T
A
L
RE
SUL
T
S AN
D
D
I
SC
USS
I
O
N
I
n
th
is
ex
p
er
im
en
t,
a
v
a
r
iety
o
f
m
ac
h
in
e
-
lear
n
in
g
tech
n
i
q
u
es
wer
e
ap
p
lied
u
s
in
g
b
o
t
h
th
e
W
ek
a
to
o
l
an
d
Py
th
o
n
C
o
lab
.
Py
th
o
n
is
a
wid
ely
u
s
ed
p
r
o
g
r
am
m
in
g
lan
g
u
ag
e
in
d
ata
s
cien
ce
,
s
cien
tific
co
m
p
u
tin
g
,
an
d
m
ac
h
in
e
lear
n
i
n
g
,
a
n
d
h
as
a
n
ex
ten
s
iv
e
lib
r
ar
y
o
f
to
o
ls
f
o
r
th
ese
f
ield
s
[
3
8
]
.
W
ek
a,
o
n
t
h
e
o
th
er
h
an
d
,
is
a
J
av
a
-
b
ased
to
o
l
d
esig
n
e
d
s
p
e
cif
ically
f
o
r
d
ata
m
in
in
g
an
d
in
clu
d
es
a
r
an
g
e
o
f
m
ac
h
in
e
l
ea
r
n
in
g
tech
n
iq
u
es
[
5
3
]
.
R
esear
ch
er
s
h
av
e
u
s
ed
b
o
th
W
ek
a
an
d
Py
th
o
n
in
clu
s
ter
in
g
,
class
if
icatio
n
,
an
d
d
et
ec
tio
n
s
tu
d
ies
[
2
4
]
,
[
2
6
]
,
[
3
9
]
–
[
4
3
]
.
Fo
r
th
is
ex
p
er
im
en
tatio
n
,
a
1
0
-
f
o
ld
c
r
o
s
s
-
v
alid
atio
n
d
atase
t
was
g
en
er
ated
u
s
in
g
b
o
th
W
ek
a
an
d
Py
th
o
n
.
An
ap
p
r
o
ac
h
th
at
is
f
r
eq
u
en
tly
u
s
ed
to
ev
al
u
ate
th
e
er
r
o
r
r
ate
o
f
a
lear
n
in
g
s
tr
ateg
y
o
n
a
p
ar
ticu
lar
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
A
co
mp
a
r
a
tive
s
tu
d
y
o
f m
a
ch
i
n
e
lea
r
n
in
g
to
o
ls
fo
r
d
etec
tin
g
Tr
o
ja
n
h
o
r
s
e
…
(
Ha
s
a
n
K
a
n
a
ke
r
)
6647
d
ataset
is
th
e
1
0
-
f
o
ld
c
r
o
s
s
v
alid
atio
n
m
eth
o
d
[
5
4
]
,
[
5
5
]
.
T
h
e
d
ataset
was
d
iv
id
ed
in
to
te
n
p
ar
ts
,
o
r
"
f
o
ld
s
,
"
an
d
ea
ch
f
o
ld
was
u
tili
ze
d
n
in
e
tim
es
f
o
r
tr
ain
in
g
an
d
o
n
ce
f
o
r
test
in
g
.
T
h
e
u
s
ag
e
o
f
a
1
0
-
f
o
ld
cr
o
s
s
-
v
alid
atio
n
a
p
p
r
o
ac
h
h
as
two
m
ain
ad
v
an
tag
es:
B
ec
au
s
e
it
u
s
es
d
ata
as
p
o
s
s
ib
le
f
o
r
b
o
t
h
th
e
tr
ain
in
g
an
d
test
in
g
p
h
ases
,
it is
m
o
r
e
ac
cu
r
ate
[
2
4
]
.
Var
io
u
s
estab
lis
h
ed
p
er
f
o
r
m
a
n
ce
m
etr
ics,
in
clu
d
in
g
ac
cu
r
a
cy
r
ate,
F
-
m
ea
s
u
r
e,
p
r
ec
is
io
n
,
an
d
r
ec
all,
h
av
e
b
ee
n
em
p
lo
y
e
d
to
ass
es
s
th
e
class
if
ier
s
'
ef
f
icac
y
.
T
a
b
le
1
s
h
o
ws
th
e
p
e
r
f
o
r
m
an
ce
m
etr
ics
o
f
v
ar
io
u
s
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
ap
p
lied
in
th
e
ex
p
e
r
im
en
t
u
s
in
g
b
o
th
W
E
KA
an
d
Py
th
o
n
.
T
h
e
alg
o
r
ith
m
s
in
clu
d
e
J
4
8
,
I
B
K,
NB
,
r
eg
r
ess
io
n
,
RF
,
SMO
an
d
MLP
.
Fro
m
s
u
b
s
ec
tio
n
s
5
.
1
to
5
.
4
,
e
x
p
lain
s
th
e
ex
p
er
im
en
tal
r
esu
lts
in
d
etail.
T
ab
le
1
.
T
r
o
jan
h
o
r
s
e
d
etec
tio
n
r
esu
lts
u
s
in
g
W
ek
a
an
d
Py
th
o
n
v
ar
i
o
u
s
m
ac
h
in
e
lear
n
in
g
m
eth
o
d
s
W
EK
A
P
y
t
h
o
n
A
l
g
o
r
i
t
h
m
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F
-
M
e
a
s
u
r
e
A
c
c
u
r
a
c
y
(
%)
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F
-
M
e
a
s
u
r
e
A
c
c
u
r
a
c
y
(
%)
N
a
ï
v
e
B
a
y
e
s
0
.
9
5
7
0
.
9
5
7
0
.
9
5
3
9
5
.
6
1
0
.
8
2
3
0
.
9
0
3
8
4
.
5
R
a
n
d
o
m
f
o
r
e
s
t
0
.
9
5
7
0
.
9
5
7
0
.
9
5
3
9
5
.
6
0
.
9
5
8
0
.
9
7
7
0
.
9
6
8
9
4
.
3
I
B
K
0
.
9
5
7
0
.
9
5
7
0
.
9
5
3
9
5
.
6
0
.
9
6
2
1
0
.
9
8
1
9
6
.
6
R
e
g
r
e
ssi
o
n
0
.
9
5
7
0
.
9
5
7
0
.
9
5
3
9
5
.
6
0
.
8
7
6
1
0
.
9
3
4
8
7
.
6
J4
8
0
.
9
5
7
0
.
9
5
7
0
.
9
5
3
9
5
.
6
0
.
9
5
3
0
.
9
7
7
0
.
9
7
4
9
5
.
4
M
u
l
t
i
l
a
y
e
r
p
e
r
c
e
p
t
r
o
n
0
.
9
5
9
0
.
9
5
9
0
.
9
5
5
9
5
.
8
0
.
8
7
6
1
0
.
9
3
4
8
7
.
6
S
M
O
0
.
9
5
9
0
.
9
5
9
0
.
9
5
5
9
5
.
8
0
.
8
7
6
1
0
.
9
3
4
8
7
.
6
5
.
1
.
P
re
cisi
o
n r
a
t
e
re
s
ults in
Wek
a
a
nd
P
y
t
ho
n
Pre
cisi
o
n
p
r
o
v
id
es
an
ex
p
la
n
atio
n
f
o
r
th
e
p
er
ce
n
tag
e
o
f
g
en
u
in
e
p
o
s
itiv
e
class
if
icatio
n
s
in
all
p
o
s
itiv
e
f
in
d
in
g
s
.
I
t
is
th
e
q
u
an
tity
o
f
s
am
p
les
th
at
ar
e
co
r
r
ec
tly
r
ec
o
g
n
ized
a
n
d
d
o
n
o
t
r
ep
r
esen
t
f
alse
p
o
s
itiv
es.
E
q
u
atio
n
(
1
)
s
tates
th
at
p
r
ec
is
io
n
is
ca
lcu
lated
u
s
in
g
th
e
T
P
an
d
FP
r
ates.
Gr
ea
ter
p
r
ec
is
io
n
is
in
d
icate
d
b
y
a
h
ig
h
er
T
P
.
Pr
e
c
isio
n
=
TP
(
TP
+
FP
)
(
1
)
T
h
e
p
r
ec
is
io
n
f
in
d
in
g
s
f
o
r
ea
ch
o
f
th
e
class
if
ier
s
u
s
ed
in
t
h
is
ex
p
er
im
en
t
ar
e
d
is
p
lay
ed
in
T
ab
le
1
an
d
Fig
u
r
e
3
.
Fo
r
W
E
KA,
all
th
e
alg
o
r
ith
m
s
h
av
e
th
e
s
am
e
p
r
ec
is
io
n
r
ate
v
alu
e
o
f
0
.
9
5
7
%
ex
ce
p
t
f
o
r
SMO
an
d
Mu
ltil
ay
er
Per
ce
p
tr
o
n
,
wh
ich
h
a
v
e
a
s
lig
h
tly
h
ig
h
er
p
r
ec
is
io
n
r
ate
v
alu
e
o
f
0
.
9
5
9
%.
Ho
wev
e
r
,
f
o
r
Py
th
o
n
,
th
e
n
aïv
e
B
ay
es a
lg
o
r
i
th
m
h
as th
e
h
ig
h
est p
r
ec
is
io
n
r
ate
v
alu
e
at
1
0
0
%.
Fig
u
r
e
3
.
Pre
cisi
o
n
r
ate
o
f
n
u
m
er
o
u
s
class
if
icatio
n
alg
o
r
ith
m
s
5
.
2
.
Rec
a
ll r
a
t
e
re
s
ults in W
ek
a
a
nd
P
y
t
ho
n
R
ec
all
i
s
a
s
tatis
t
ic
th
at
q
u
an
tifie
s
th
e
f
r
eq
u
en
cy
with
wh
i
ch
a
m
ac
h
in
e
lear
n
in
g
m
o
d
el
p
r
o
p
er
ly
s
elec
ts
p
o
s
itiv
e
s
am
p
les
(
tr
u
e
p
o
s
itiv
es)
f
r
o
m
all
o
f
th
e
r
e
al
p
o
s
itiv
e
s
am
p
les
in
th
e
d
a
taset.
On
th
e
o
th
er
h
an
d
s
,
tr
u
ly
p
o
s
itiv
e
(
T
P)
s
a
m
p
les
ar
e
t
h
o
s
e
f
r
o
m
p
r
e
d
icted
clo
u
d
s
tr
o
ja
n
h
o
r
s
e
s
am
p
l
es
th
at
ar
e
co
r
r
ec
tly
ca
teg
o
r
ized
an
d
tag
g
ed
as d
an
g
er
o
u
s
.
C
alcu
latin
g
r
ec
all
is
d
o
n
e
u
s
in
g
(
2
)
.
0.95
7
0.95
7
0.95
7
0.95
7
0.95
7
0.95
9
0.95
9
1
0.95
8
0.962
0.87
6
0.95
3
0.87
6
0.87
6
0.8
0.8
5
0.9
0.9
5
1
1.0
5
N
aï
v
e
B
ay
e
s
R
an
do
m
Fo
r
e
s
t
IB
K
R
e
g
r
e
ss
i
o
n
J
48
Mul
t
i
l
ay
e
r
P
e
r
c
e
p
t
r
o
n
S
M
O
P
e
r
c
e
n
t
a
g
e
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
A
l
g
o
r
i
t
h
ms
P
rec
ision
W
E
K
A
P
r
e
c
i
si
o
n
P
Y
T
H
O
N
P
r
e
c
i
si
o
n
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.
14
,
No
.
6
,
Dec
em
b
e
r
20
24
:
6
6
4
2
-
6
6
5
5
6648
=
=
(
+
)
(
2
)
Fro
m
th
e
ex
p
er
im
en
tal
r
esu
lts
in
T
ab
le
1
an
d
Fig
u
r
e
4
,
it
is
clea
r
ly
s
ee
n
th
at
f
o
r
W
E
KA,
all
th
e
alg
o
r
ith
m
s
h
av
e
th
e
s
am
e
r
ec
all
r
ate
v
alu
e
o
f
0
.
9
5
7
%,
e
x
ce
p
t
f
o
r
MLP
an
d
SMO,
wh
ich
h
a
v
e
a
s
lig
h
t
ly
h
ig
h
er
r
ec
all
r
ate
v
alu
e
o
f
0
.
9
5
9
%.
Fo
r
Py
th
o
n
,
I
B
K,
r
eg
r
ess
io
n
,
SMO
an
d
th
e
m
u
ltil
ay
er
p
er
ce
p
tr
o
n
alg
o
r
ith
m
h
av
e
th
e
h
ig
h
est r
ec
all
r
ate
v
alu
e
at
1
0
0
%.
Fig
u
r
e
4
.
R
ec
all
r
ate
o
f
n
u
m
e
r
o
u
s
class
if
icatio
n
alg
o
r
ith
m
s
5
.
3
.
F
-
M
ea
s
ure
ra
t
e
re
s
ults i
n We
k
a
a
nd
P
y
t
ho
n
T
h
e
F
-
Me
asu
r
e
is
a
s
y
s
tem
p
er
f
o
r
m
a
n
ce
m
etr
ic
th
at
a
g
g
r
e
g
ates
r
ec
all
an
d
p
r
ec
is
io
n
in
t
o
a
s
in
g
le
n
u
m
b
er
.
T
h
e
ca
lc
u
latio
n
o
f
th
e
F
-
m
ea
s
u
r
e
is
s
h
o
wn
in
(
3
)
.
F
−
M
e
a
s
ure
=
2
x
Recal
l
x
p
r
ec
i
s
i
o
n
Recal
l
+
p
r
ec
i
s
i
o
n
(
3
)
T
h
e
ex
p
er
im
en
tal
r
esu
lts
in
T
ab
le
1
an
d
Fig
u
r
e
5
s
h
o
w
th
at
f
o
r
W
E
KA,
all
th
e
alg
o
r
ith
m
s
h
av
e
th
e
s
am
e
F
-
m
ea
s
u
r
e
r
ate
v
alu
e
o
f
0
.
9
5
3
%,
ex
ce
p
t f
o
r
MLP
an
d
SMO,
wh
ich
h
av
e
s
lig
h
tly
h
ig
h
er
F
-
m
ea
s
u
r
e
r
ate
v
alu
es
o
f
0
.
9
5
5
%.
W
h
ile,
f
o
r
Py
th
o
n
,
I
B
K
h
as
th
e
h
ig
h
est
F
-
m
ea
s
u
r
e
r
ate
v
alu
e
at
0
.
9
8
1
%
an
d
n
aïv
e
B
ay
es
alg
o
r
ith
m
h
as th
e
lo
west F
-
m
ea
s
u
r
e
r
ate
v
alu
e
at
0
.
9
0
3
%.
Fig
u
r
e
5
.
F
-
Me
asu
r
e
r
ate
o
f
n
u
m
er
o
u
s
class
if
icatio
n
alg
o
r
ith
m
s
0.95
7
0.95
7
0.95
7
0.95
7
0.95
7
0
.9
5
9
0.95
9
0
.8
2
3
0.97
7
1
1
0.97
7
1
1
0
0.2
0.4
0.6
0.8
1
1.2
N
aï
v
e
B
ay
e
s
R
an
do
m
F
o
r
e
st
IB
K
R
e
g
r
e
ss
i
o
n
J
48
M
ul
t
i
l
ay
e
r
P
e
r
c
e
pt
r
o
n
S
M
O
P
e
r
c
e
n
t
a
g
e
M
ac
hi
ne
Le
ar
ni
ng
Al
g
o
r
i
t
hm
s
Re
c
a
ll
W
E
K
A
R
e
c
al
l
P
Y
T
H
O
N
R
e
c
al
l
0.95
3
0.95
3
0.95
3
0.95
3
0
.9
5
3
0.95
5
0.95
5
0
.9
0
3
0.96
8
0.98
1
0.93
4
0.97
4
0.93
4
0.93
4
0.86
0.88
0.9
0.92
0.94
0.96
0.98
1
Naï
v
e
B
ay
es
R
an
do
m
F
o
r
e
st
IB
K
R
e
g
r
e
ss
i
o
n
J
48
M
ul
t
i
l
ay
e
r
P
e
r
c
e
pt
r
o
n
S
M
O
P
e
r
c
e
n
t
a
g
e
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
A
l
g
o
r
i
t
h
ms
F
-
M
e
a
su
r
e
W
E
K
A
F-
M
e
as
ur
e
P
Y
T
H
O
N
F-
M
e
as
u
r
e
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
A
co
mp
a
r
a
tive
s
tu
d
y
o
f m
a
ch
i
n
e
lea
r
n
in
g
to
o
ls
fo
r
d
etec
tin
g
Tr
o
ja
n
h
o
r
s
e
…
(
Ha
s
a
n
K
a
n
a
ke
r
)
6649
5
.
4
.
Acc
ura
cy
ra
t
e
re
s
ults in We
k
a
a
nd
P
y
t
ho
n
Acc
u
r
ac
y
is
o
f
ten
r
e
f
er
r
ed
to
as
co
r
r
ec
t
class
if
icatio
n
.
T
h
e
p
er
ce
n
ta
g
e
o
f
ac
cu
r
ate
p
r
e
d
ictio
n
s
is
ex
p
r
ess
ed
b
y
th
e
p
e
r
f
o
r
m
an
c
e
s
tatis
tic
ac
cu
r
ac
y
.
T
a
b
le
1
an
d
Fig
u
r
e
6
s
h
o
w
t
h
e
ac
c
u
r
ac
y
r
ates
f
o
r
th
e
d
if
f
er
en
t c
lass
if
ier
s
.
T
h
e
ex
p
er
im
en
tal
r
esu
lts
f
o
r
th
e
v
a
r
io
u
s
class
if
ier
s
in
th
is
s
tu
d
y
h
av
e
s
h
o
wn
t
h
at
f
o
r
W
E
KA,
m
u
ltil
ay
er
p
e
r
ce
p
tr
o
n
a
n
d
SMO
h
av
e
s
lig
h
tly
h
ig
h
er
ac
cu
r
a
cy
r
ates
at
9
5
.
8
%.
I
n
co
n
tr
ast,
all
o
th
e
r
class
if
ier
s
ac
cu
r
ac
y
r
ates
ar
e
eq
u
al,
a
n
d
th
eir
ac
cu
r
ac
y
r
ates
ar
e
eq
u
a
l
to
9
5
.
6
%.
W
h
ile,
f
o
r
Py
th
o
n
,
th
e
ex
p
e
r
im
en
tal
r
esu
lts
h
av
e
s
h
o
w
n
th
at
th
e
I
B
K
h
as
th
e
h
i
g
h
est
ac
cu
r
ac
y
r
ate
v
alu
e
o
f
9
6
.
6
%
a
n
d
t
h
e
n
aïv
e
B
ay
es
h
as
t
h
e
lo
west a
cc
u
r
ac
y
r
ate
v
alu
e
at
8
4
.
5
%.
Dep
en
d
in
g
o
n
th
e
f
in
d
in
g
s
in
T
ab
le
1
an
d
Fig
u
r
es
3
to
6
,
it
ca
n
b
e
co
n
clu
d
ed
th
at
f
o
r
W
E
KA,
all
th
e
alg
o
r
ith
m
s
h
av
e
th
e
s
am
e
p
r
ec
is
io
n
,
r
ec
all,
F
-
m
ea
s
u
r
e,
an
d
ac
cu
r
ac
y
r
ate
v
alu
es
o
f
9
5
.
7
%
ex
ce
p
t
f
o
r
m
u
ltil
ay
er
p
er
ce
p
tr
o
n
an
d
S
MO
,
wh
ich
h
a
v
e
s
lig
h
tly
h
ig
h
er
p
r
ec
is
io
n
,
r
ec
all,
a
n
d
F
-
m
ea
s
u
r
e
r
ate
v
alu
es
at
9
5
.
9
%
an
d
ac
c
u
r
ac
y
r
ate
v
al
u
es
at
9
5
.
8
%.
On
th
e
o
th
e
r
h
an
d
,
f
o
r
Py
th
o
n
,
n
aïv
e
B
ay
es
h
as
th
e
h
ig
h
est
p
r
ec
is
io
n
r
ate
v
al
u
e
at
1
0
0
%,
wh
ile
I
B
K,
r
eg
r
ess
io
n
,
SMO
an
d
m
u
ltil
ay
er
p
er
ce
p
t
r
o
n
,
h
a
v
e
th
e
h
ig
h
est
r
ec
all
r
ate
v
alu
e
at
1
0
0
%.
I
B
K
h
as
th
e
h
ig
h
est
F
-
m
ea
s
u
r
e
a
n
d
ac
cu
r
ac
y
r
ate
v
alu
es
at
9
8
.
1
%
an
d
9
6
.
6
%,
r
esp
ec
tiv
ely
.
Ho
wev
er
,
r
e
g
r
ess
io
n
,
m
u
ltil
ay
er
p
er
ce
p
tio
n
,
a
n
d
SMO
h
av
e
th
e
l
o
west
p
r
e
cisi
o
n
r
ate
v
alu
e
at
8
7
.
6
%.
Naïv
e
B
ay
es
h
as
th
e
l
o
west
r
ec
all,
F
-
m
ea
s
u
r
e,
an
d
ac
cu
r
ac
y
r
ate
v
alu
es
at
8
2
.
3
%,
9
0
.
3
%,
an
d
8
4
.
5
%,
r
esp
ec
tiv
ely
.
Fig
u
r
e
6
.
Acc
u
r
ac
y
r
ate
o
f
n
u
m
er
o
u
s
class
if
icatio
n
alg
o
r
ith
m
s
Dif
f
er
en
t
m
ac
h
in
e
lear
n
in
g
a
lg
o
r
ith
m
s
p
er
f
o
r
m
d
if
f
er
e
n
tly
in
T
r
o
jan
h
o
r
s
e
d
etec
tio
n
.
Mu
ltil
ay
er
p
er
ce
p
tr
o
n
an
d
s
eq
u
e
n
tial
m
in
im
al
o
p
tim
izatio
n
class
if
ier
s
h
av
e
s
h
o
wn
b
etter
r
esu
lts
in
W
ek
a,
wh
ile
I
B
K
an
d
J
4
8
class
if
ier
s
h
av
e
p
er
f
o
r
m
e
d
b
etter
in
Py
th
o
n
.
Ov
er
all,
th
e
f
in
d
in
g
s
s
h
o
w
th
at
m
ac
h
in
e
lear
n
in
g
tech
n
iq
u
es
ar
e
ef
f
ec
tiv
e
at
f
in
d
in
g
T
r
o
ja
n
h
o
r
s
es,
with
h
ig
h
r
ec
all,
p
r
ec
is
io
n
,
an
d
F
-
Me
asu
r
e
s
co
r
es.
T
h
e
ch
o
ice
o
f
alg
o
r
ith
m
an
d
to
o
l
(
W
E
KA
o
r
Py
th
o
n
)
m
ay
d
ep
e
n
d
o
n
th
e
d
ataset,
s
p
ec
if
ic
r
eq
u
i
r
em
en
t
s
an
d
co
n
s
tr
ain
ts
o
f
th
e
ap
p
licatio
n
.
A
co
m
p
ar
is
o
n
b
etwe
en
Py
th
o
n
an
d
W
ek
a
f
o
r
v
ar
io
u
s
m
ac
h
in
e
lear
n
i
n
g
al
g
o
r
ith
m
s
i
n
t
er
m
o
f
th
e
ex
ec
u
tio
n
tim
e
is
s
h
o
wn
in
T
ab
le
2
.
W
e
ca
n
s
ee
th
at
Py
t
h
o
n
h
as
o
u
t
p
er
f
o
r
m
W
ek
a
in
l
o
wer
f
o
r
m
o
s
t
o
f
th
e
alg
o
r
ith
m
s
.
Py
th
o
n
e
x
ec
u
tio
n
tim
e
f
o
r
n
aïv
e
B
ay
es,
r
an
d
o
m
f
o
r
est
,
I
B
K,
J
4
8
,
SMO
is
s
m
aller
th
an
W
ek
a'
s
.
B
u
t
th
e
m
u
ltil
ay
er
p
e
r
ce
p
tr
o
n
alg
o
r
ith
m
h
as
s
h
o
wn
m
o
r
e
ex
ec
u
tio
n
tim
e
in
W
ek
a.
T
h
e
r
ea
s
o
n
f
o
r
th
a
t
b
ec
au
s
e
W
ek
a’
s
im
p
lem
e
n
tatio
n
o
f
m
u
ltil
ay
er
p
er
ce
p
tr
o
n
h
a
s
co
m
p
lex
co
m
p
u
tatio
n
s
th
a
n
in
Py
th
o
n
.
Ov
e
r
all,
th
e
ex
p
er
im
en
tal
f
i
n
d
in
g
s
d
e
m
o
n
s
tr
ate
th
at
Py
th
o
n
is
b
ett
er
in
ter
m
s
o
f
ex
ec
u
tio
n
tim
e
f
o
r
th
e
m
ajo
r
ity
o
f
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
.
Ho
wev
er
,
we
s
h
o
u
ld
n
o
t
f
o
r
g
et
th
at
th
e
ex
ec
u
tio
n
tim
e
c
an
af
f
ec
ted
b
y
th
e
h
ar
d
war
e
a
n
d
s
o
f
twar
e
c
o
n
f
ig
u
r
atio
n
s
o
f
t
h
e
co
m
p
u
ter
s
y
s
te
m
s
u
s
ed
f
o
r
th
e
an
aly
s
is
.
T
h
is
r
esear
ch
em
p
l
o
y
s
th
e
d
e
f
au
lt
p
ar
am
eter
co
n
f
ig
u
r
atio
n
s
f
o
r
b
o
th
Py
t
h
o
n
-
C
o
lab
an
d
W
ek
a.
T
h
e
T
ab
le
s
3
t
o
9
d
is
p
lay
th
e
p
ar
a
m
eter
s
ettin
g
s
f
o
r
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
u
s
ed
in
t
h
is
s
tu
d
y
.
T
h
e
y
s
h
o
w
n
aiv
e
B
ay
es
,
r
an
d
o
m
f
o
r
est
,
I
B
K
,
r
eg
r
ess
io
n
,
J
4
8
,
m
u
ltil
ay
e
r
p
er
ce
p
tr
o
n
,
a
n
d
s
eq
u
e
n
tial m
in
im
al
o
p
tim
izatio
n
class
if
ier
s
u
s
in
g
Py
th
o
n
-
C
o
lab
an
d
W
ek
a,
r
esp
ec
tiv
ely
.
95.6
95.6
95.6
95.6
95.6
95.8
95.8
84.5
94.3
96.6
87.6
95.4
87.6
87.6
78
80
82
84
86
88
90
92
94
96
98
N
aï
v
e
B
ay
e
s
R
an
do
m
F
o
r
e
st
IB
K
R
e
g
r
e
ss
i
o
n
J
48
M
ul
t
i
l
ay
e
r
P
e
r
c
ept
r
on
S
M
O
P
e
r
ce
n
t
ag
e
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
A
l
g
o
r
i
t
h
ms
Ac
c
u
r
a
c
y
(%
)
W
E
K
A
Ac
c
ur
ac
y
(
%)
P
Y
T
H
O
N
Ac
c
ur
a
c
y
(
%)
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.
14
,
No
.
6
,
Dec
em
b
e
r
20
24
:
6
6
4
2
-
6
6
5
5
6650
T
ab
le
2
.
T
im
e
c
o
m
p
a
r
is
o
n
b
et
wee
n
Py
th
o
n
a
n
d
W
ek
a
Mac
h
i
n
e
l
ear
n
i
n
g
al
g
o
ri
t
h
m
Py
t
h
o
n
e
x
e
c
u
t
i
o
n
t
i
m
e
(
s
)
W
e
k
a
e
x
ec
u
t
i
o
n
t
i
me
(s
)
N
a
ï
v
e
B
ay
es
0
.
0
0
3
0
.
0
2
Ran
d
o
m
f
o
re
s
t
0
.
0
0
1
0
.
0
5
IBK
0
.
0
0
2
0
.
0
1
Reg
res
s
i
o
n
0
.
0
0
4
0
.
0
7
J
4
8
0
.
0
0
2
0
.
0
1
Mu
l
t
i
l
a
y
er
p
erce
p
t
r
o
n
0
.
0
0
3
0
.
7
9
SM
O
0
.
0
0
2
0
.
0
3
T
ab
le
3
.
Naiv
e
B
ay
es
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
p
ar
am
eter
s
in
Py
th
o
n
an
d
W
ek
a
Py
t
h
o
n
W
e
k
a
Param
et
e
r
V
a
l
u
e
Param
et
e
r
V
a
l
u
e
p
r
i
o
rs
N
o
n
e
U
s
eKe
r
n
e
l
E
s
t
i
m
a
t
o
r
Fal
s
e
va
r
_
s
m
o
o
t
h
i
n
g
1e
-
9
U
s
e
S
u
p
e
r
v
i
s
e
d
D
i
s
c
r
e
t
i
z
a
t
i
o
n
Fal
s
e
T
ab
le
4
.
R
an
d
o
m
f
o
r
est m
ac
h
i
n
e
lear
n
in
g
alg
o
r
ith
m
p
ar
a
m
eter
s
in
Py
th
o
n
a
n
d
W
ek
a
Py
t
h
o
n
W
e
k
a
Param
et
e
r
V
a
l
u
e
Param
et
e
r
V
a
l
u
e
n
_
es
t
i
m
a
t
o
r
s
5
0
0
b
a
g
S
i
z
eP
er
ce
n
t
1
0
0
cr
i
t
e
r
i
o
n
g
i
n
i
m
a
x
D
e
p
t
h
0
m
a
x
_
d
ep
t
h
N
o
n
e
n
u
m
F
e
a
t
u
r
e
s
0
m
i
n
_
s
a
m
p
l
e
s
_
s
p
l
i
t
2
s
ee
d
1
m
i
n
_
s
a
m
p
l
e
s
_
l
e
a
f
1
n
u
m
E
x
ec
u
t
i
o
n
S
l
o
t
s
1
m
i
n
_
w
ei
g
h
t
_
f
r
a
c
t
i
o
n
_
l
e
a
f
0
b
a
t
c
h
S
i
z
e
1
0
0
m
a
x
_
f
ea
t
u
r
es
au
t
o
d
e
b
u
g
Fal
s
e
m
a
x
_
l
ea
f
_
n
o
d
e
s
N
o
n
e
d
o
N
o
t
C
h
eckC
a
p
a
b
i
l
i
t
i
es
Fal
s
e
m
i
n
_
i
m
p
u
r
i
t
y
_
d
ec
r
e
a
s
e
0
b
r
ea
kT
i
es
R
a
n
d
o
m
l
y
Fal
s
e
b
o
o
t
s
t
r
a
p
T
r
u
e
p
r
i
n
t
Cl
a
s
s
i
f
i
er
s
Fal
s
e
o
o
b
_
s
c
o
r
e
Fal
s
e
p
r
i
n
t
Tr
ee
s
Fal
s
e
n
_
j
o
b
s
N
o
n
e
i
n
B
a
g
O
u
t
O
f
B
a
g
E
v
a
l
u
a
t
i
o
n
Fal
s
e
r
a
n
d
o
m
_
s
t
a
t
e
1
ver
b
o
s
e
0
wa
r
m
_
s
t
a
r
t
Fal
s
e
cl
a
s
s
_
we
i
g
h
t
N
o
n
e
ccp
_
a
l
p
h
a
0
.
0
m
a
x
_
s
a
m
p
l
es
N
o
n
e
T
ab
le
5
.
I
B
K
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
p
ar
am
ete
r
s
in
Py
th
o
n
an
d
W
ek
a
Py
t
h
o
n
W
e
k
a
Param
et
e
r
V
a
l
u
e
Param
et
e
r
V
a
l
u
e
n
_
n
ei
g
h
b
o
r
s
5
KNN
1
wei
g
h
t
s
u
n
i
f
o
r
m
D
i
s
t
a
n
ceW
e
i
g
h
t
i
n
g
N
o
d
i
s
t
a
n
ce
w
ei
g
h
t
i
n
g
a
l
g
o
r
i
t
h
m
au
t
o
W
i
n
d
o
w
S
i
z
e
0
l
e
a
f
_
s
i
z
e
30
D
i
s
t
a
n
ce
F
u
n
c
t
i
o
n
E
u
cl
i
d
ea
n
D
i
s
t
a
n
ce
p
2
Cr
o
s
s
V
a
l
i
d
a
t
e
Fal
s
e
m
e
t
r
i
c
mi
n
k
o
w
s
k
i
M
e
a
n
S
q
u
a
r
ed
Fal
s
e
m
e
t
r
i
c
_
p
a
r
a
m
s
N
o
n
e
D
e
b
u
g
Fal
s
e
n
_
j
o
b
s
N
o
n
e
d
o
N
o
t
C
h
eckC
a
p
a
b
i
l
i
t
i
es
Fal
s
e
T
ab
le
6
.
R
eg
r
ess
io
n
m
ac
h
i
n
e
l
ea
r
n
in
g
alg
o
r
ith
m
p
ar
am
eter
s
in
Py
th
o
n
a
n
d
W
ek
a
Py
t
h
o
n
W
e
k
a
Param
et
e
r
V
a
l
u
e
Param
et
e
r
V
a
l
u
e
p
e
n
a
l
t
y
l2
R
i
d
g
e
1
.
0
E
-
8
d
u
a
l
Fal
s
e
M
a
xI
t
s
-
1
t
o
l
1e
-
4
D
e
b
u
g
Fal
s
e
C
1
0
0
d
o
N
o
t
C
h
eckC
a
p
a
b
i
l
i
t
i
es
Fal
s
e
f
i
t
_
i
n
t
e
r
ce
p
t
T
r
u
e
B
a
t
c
h
S
i
z
e
1
0
0
i
n
t
e
r
ce
p
t
_
s
ca
l
i
n
g
1
Nu
m
D
ec
i
m
a
l
P
l
a
ce
s
2
cl
a
s
s
_
we
i
g
h
t
N
o
n
e
r
a
n
d
o
m
_
s
t
a
t
e
1
s
o
l
v
er
l
b
fg
s
m
a
x
_
i
t
e
r
1
0
0
m
u
l
t
i
_
cl
a
s
s
o
v
r
ver
b
o
s
e
0
wa
r
m
_
s
t
a
r
t
Fal
s
e
n
_
j
o
b
s
N
o
n
e
l
1
_
r
a
t
i
o
N
o
n
e
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
A
co
mp
a
r
a
tive
s
tu
d
y
o
f m
a
ch
i
n
e
lea
r
n
in
g
to
o
ls
fo
r
d
etec
tin
g
Tr
o
ja
n
h
o
r
s
e
…
(
Ha
s
a
n
K
a
n
a
ke
r
)
6651
T
ab
le
7
.
J
4
8
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
p
a
r
am
eter
s
in
Py
th
o
n
an
d
W
ek
a
P
y
t
h
o
n
W
e
k
a
P
a
r
a
me
t
e
r
V
a
l
u
e
P
a
r
a
me
t
e
r
V
a
l
u
e
C
r
i
t
e
r
i
o
n
g
i
n
i
C
r
i
t
e
r
i
o
n
e
n
t
r
o
p
y
S
p
l
i
t
t
e
r
b
e
s
t
M
a
x
d
e
p
t
h
N
o
n
e
M
a
x
d
e
p
t
h
N
o
n
e
M
i
n
s
a
mp
l
e
s
sp
l
i
t
2
M
i
n
s
a
mp
l
e
s
sp
l
i
t
2
M
i
n
s
a
mp
l
e
s
l
e
a
f
1
M
i
n
s
a
mp
l
e
s
l
e
a
f
1
C
C
P
a
l
p
h
a
0
.
2
5
M
i
n
w
e
i
g
h
t
f
r
a
c
t
i
o
n
l
e
a
f
0
.
0
S
u
b
t
r
e
e
r
a
i
s
i
n
g
T
r
u
e
M
a
x
f
e
a
t
u
r
e
s
N
o
n
e
U
se
La
p
l
a
c
e
F
a
l
se
R
a
n
d
o
m s
t
a
t
e
N
o
n
e
B
i
n
a
r
y
s
p
l
i
t
s
F
a
l
se
M
a
x
l
e
a
f
n
o
d
e
s
N
o
n
e
S
a
v
e
i
n
st
a
n
c
e
d
a
t
a
F
a
l
se
M
i
n
i
mp
u
r
i
t
y
d
e
c
r
e
a
se
0
.
0
C
l
a
s
s we
i
g
h
t
N
o
n
e
C
C
P
a
l
p
h
a
0
.
0
T
ab
le
8
.
Mu
ltil
ay
er
p
er
ce
p
t
r
o
n
m
ac
h
in
e
lear
n
i
n
g
alg
o
r
ith
m
p
ar
am
eter
s
in
Py
th
o
n
an
d
W
ek
a
P
y
t
h
o
n
W
e
k
a
p
e
n
a
l
t
y
N
o
n
e
l
e
a
rn
i
n
g
R
a
t
e
0
.
3
a
l
p
h
a
0
.
0
0
0
1
m
o
m
e
n
t
u
m
0
.
2
f
i
t
_
i
n
t
e
r
c
e
p
t
Tr
u
e
h
i
d
d
e
n
L
a
y
e
rs
'a'
m
a
x
_
i
t
e
r
1
0
0
0
t
ra
i
n
i
n
g
T
i
m
e
5
0
0
t
o
l
1e
-
3
v
a
l
i
d
a
t
i
o
n
T
h
r
e
sh
o
l
d
20
sh
u
f
f
l
e
Tr
u
e
see
d
0
v
e
r
b
o
s
e
0
l
e
a
rn
i
n
g
R
a
t
e
D
e
c
a
y
F
a
l
se
e
t
a
0
0
.
1
c
o
n
v
e
rg
e
E
p
o
c
h
s
0
n
_
j
o
b
s
N
o
n
e
c
o
n
v
e
rg
e
T
h
r
e
sh
o
l
d
0
.
0
0
1
ra
n
d
o
m
_
s
t
a
t
e
1
n
o
rm
a
l
i
ze
At
t
ri
b
u
t
e
s
Tr
u
e
e
a
r
l
y
_
st
o
p
p
i
n
g
F
a
l
se
n
o
rm
a
l
i
ze
C
l
a
ss
Tr
u
e
v
a
l
i
d
a
t
i
o
n
_
f
ra
c
t
i
o
n
0
.
1
d
e
c
a
y
F
a
l
se
n
_
i
t
e
r_
n
o
_
c
h
a
n
g
e
5
rese
t
Tr
u
e
c
l
a
ss_
w
e
i
g
h
t
N
o
n
e
n
o
m
i
n
a
l
T
o
B
i
n
a
r
y
Fi
l
t
e
r
Tr
u
e
w
a
rm
_
st
a
rt
F
a
l
se
d
e
b
u
g
F
a
l
se
T
ab
le
9
.
SMO
m
ac
h
in
e
lear
n
i
n
g
alg
o
r
ith
m
p
a
r
am
eter
s
in
Py
th
o
n
an
d
W
ek
a
P
y
t
h
o
n
W
e
k
a
C
1
.
0
C
1
.
0
k
e
r
n
e
l
l
i
n
e
a
r
k
e
r
n
e
l
P
o
l
y
K
e
r
n
e
l
d
e
g
re
e
3
f
i
l
t
e
rT
y
p
e
2
g
a
m
m
a
'sca
l
e
'
t
o
l
e
r
a
n
c
e
P
a
r
a
m
e
t
e
r
0
.
0
0
1
c
o
e
f
0
0
.
0
e
p
s
i
l
o
n
1
.
0
E
-
12
sh
ri
n
k
i
n
g
Tr
u
e
n
u
m
F
o
l
d
s
-
1
p
ro
b
a
b
i
l
i
t
y
F
a
l
se
ra
n
d
o
m
S
e
e
d
1
t
o
l
1e
-
3
b
u
i
l
d
L
o
g
i
st
i
c
Mo
d
e
l
s
F
a
l
se
c
a
c
h
e
_
s
i
ze
2
0
0
d
e
b
u
g
F
a
l
se
c
l
a
ss_
w
e
i
g
h
t
N
o
n
e
d
o
N
o
t
C
h
e
c
k
C
a
p
a
b
i
l
i
t
i
e
s
F
a
l
se
v
e
r
b
o
s
e
F
a
l
se
m
a
x
_
i
t
e
r
-
1
d
e
c
i
si
o
n
_
f
u
n
c
t
i
o
n
_
sh
a
p
e
'o
v
r
'
b
re
a
k
_
t
i
e
s
F
a
l
se
ra
n
d
o
m
_
s
t
a
t
e
1
6.
CO
NCLU
SI
O
N
T
h
is
s
tu
d
y
e
x
p
lo
r
ed
th
e
s
ec
u
r
i
ty
r
is
k
s
an
d
b
en
ef
its
ass
o
ciate
d
with
cl
o
u
d
co
m
p
u
tin
g
an
d
t
h
e
u
s
ag
e
o
f
m
ac
h
in
e
lear
n
in
g
f
o
r
d
etec
tin
g
T
r
o
jan
h
o
r
s
e
in
f
ec
tio
n
s
.
W
ek
a
an
d
Py
th
o
n
C
o
lab
,
two
w
ell
-
k
n
o
wn
m
ac
h
in
e
lear
n
in
g
to
o
ls
,
wer
e
test
ed
in
th
e
s
tu
d
y
to
s
ee
h
o
w
well
th
ey
p
er
f
o
r
m
ed
at
id
en
tify
i
n
g
T
r
o
ja
n
h
o
r
s
e
in
f
ec
tio
n
s
.
T
h
e
r
esu
lts
s
h
o
w
th
at
b
o
th
to
o
ls
ar
e
ef
f
ec
tiv
e
in
d
etec
tin
g
T
r
o
jan
h
o
r
s
e
in
f
ec
tio
n
s
,
Ho
w
ev
er
,
th
e
in
s
tr
u
m
en
t
s
elec
ted
will
r
ely
o
n
th
e
p
ar
ticu
lar
n
ee
d
s
an
d
r
eso
u
r
ce
s
o
f
f
er
ed
.
Fu
r
th
er
m
o
r
e,
th
e
s
tu
d
y
ev
alu
ated
t
h
e
co
m
p
u
tatio
n
al
r
eso
u
r
c
es
r
eq
u
i
r
ed
b
y
ea
c
h
to
o
l
an
d
f
o
u
n
d
t
h
at
Py
th
o
n
C
o
lab
g
en
er
ally
h
ad
f
aster
e
x
ec
u
tio
n
tim
es
th
an
W
ek
a.
T
h
is
s
tu
d
y
h
elp
s
r
esear
ch
er
s
an
d
p
r
ac
titi
o
n
er
s
in
c
y
b
er
s
ec
u
r
ity
f
ield
to
ad
o
p
t
th
e
s
u
itab
le
m
ac
h
in
e
lear
n
i
n
g
to
o
ls
to
d
etec
t
th
e
m
alwa
r
e
in
f
ec
tio
n
ef
f
ec
tiv
ely
b
ased
o
n
p
er
f
o
r
m
an
ce
an
d
ac
cu
r
ac
y
d
etec
tio
n
r
ate
u
s
in
g
th
is
s
tu
d
y
in
s
ig
h
ts
in
to
th
e
s
tr
en
g
th
s
an
d
lim
itatio
n
s
o
f
W
ek
a
an
d
Py
th
o
n
C
o
lab
f
o
r
b
u
ild
in
g
m
ac
h
in
e
lear
n
in
g
m
o
d
els an
d
d
etec
tin
g
T
r
o
jan
h
o
r
s
e
in
f
ec
tio
n
s
.
T
h
is
s
tu
d
y
lim
ited
b
y
ex
tr
ac
tin
g
n
in
e
f
ea
tu
r
es
f
r
o
m
th
e
s
am
p
les.
Ho
wev
er
,
it
is
cr
u
cial
to
em
p
h
asize
th
at
ad
d
itio
n
al
s
tu
d
y
is
r
e
q
u
ir
e
d
,
s
u
ch
as
b
y
i
n
cr
ea
s
in
g
f
ea
tu
r
e
ex
tr
ac
tio
n
,
in
o
r
d
er
to
in
cr
ea
s
e
th
e
s
u
cc
ess
r
ate
Evaluation Warning : The document was created with Spire.PDF for Python.