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
.
6
,
Dec
em
b
er
20
25
,
p
p
.
4
9
2
3
~
4
9
3
2
I
SS
N:
2
2
5
2
-
8
9
3
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijai.v
14
.i
6
.
p
p
4
9
2
3
-
4
9
3
2
4923
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
Enha
ncing
c
ro
ss
-
site scrip
ting
a
tt
a
ck det
ection
by
us
ing
Fas
tTex
t
a
s
wo
rd
embedding
s
a
nd
l
o
ng
-
sho
rt
term
m
emo
ry
M
uh
a
m
m
a
d Alk
ha
iri
M
a
s
h
uri
1
,
Nico
Su
ra
ntha
1,
2
1
M
a
s
t
e
r
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
B
I
N
U
S
G
r
a
d
u
a
t
e
P
r
o
g
r
a
m,
B
i
n
a
N
u
s
a
n
t
a
r
a
U
n
i
v
e
r
s
i
t
y
,
J
a
k
a
r
t
a
,
I
n
d
o
n
e
s
i
a
2
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
i
c
a
l
,
E
l
e
c
t
r
o
n
i
c
a
n
d
C
o
m
mu
n
i
c
a
t
i
o
n
E
n
g
i
n
e
e
r
i
n
g
,
F
a
c
u
l
t
y
o
f
En
g
i
n
e
e
r
i
n
g
,
To
k
y
o
C
i
t
y
U
n
i
v
e
r
si
t
y
,
To
k
y
o
,
J
a
p
a
n
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Au
g
2
0
,
2
0
2
4
R
ev
is
ed
J
u
l 7
,
2
0
2
5
Acc
ep
ted
Au
g
6
,
2
0
2
5
Cro
ss
-
site
sc
rip
ti
n
g
(XS
S
)
is
o
n
e
o
f
t
h
e
d
a
n
g
e
ro
u
s
c
y
b
e
r
-
a
tt
a
c
k
s
a
n
d
t
h
e
n
u
m
b
e
r
o
f
a
tt
a
c
k
s
c
o
n
ti
n
u
e
s
to
in
c
re
a
se
.
Th
is
stu
d
y
tak
e
s
a
n
e
w
a
p
p
ro
a
c
h
to
d
e
tec
t
a
tt
a
c
k
s
b
y
u
ti
li
z
in
g
F
a
stTe
x
t
a
s
wo
rd
e
m
b
e
d
d
in
g
,
a
n
d
l
o
n
g
-
sh
o
rt
term
m
e
m
o
ry
(LS
TM
)
,
wh
ich
a
ims
t
o
imp
ro
v
e
th
e
p
e
rfo
rm
a
n
c
e
o
f
d
e
e
p
lea
rn
in
g
.
Th
is
m
e
th
o
d
is
p
r
o
p
o
se
d
t
o
c
a
p
t
u
re
t
h
e
b
r
o
a
d
e
r
m
e
a
n
i
n
g
a
n
d
c
o
n
tex
t
o
f
t
h
e
d
a
ta
u
se
d
,
lea
d
in
g
to
b
e
tt
e
r
fe
a
tu
re
e
x
trac
ti
o
n
a
n
d
m
o
d
e
l
p
e
rfo
rm
a
n
c
e
.
Th
is
stu
d
y
n
o
t
o
n
l
y
im
p
ro
v
e
s
th
e
d
e
tec
ti
o
n
o
f
XS
S
a
tt
a
c
k
s,
b
u
t
a
lso
h
i
g
h
li
g
h
ts
t
h
e
p
o
ten
ti
a
l
fo
r
b
e
t
ter
tex
t
p
r
o
c
e
ss
in
g
tec
h
n
i
q
u
e
s.
T
h
e
re
su
lt
s
o
b
tai
n
e
d
sh
o
wi
n
g
th
is
m
e
th
o
d
a
c
h
iev
e
s
h
i
g
h
e
r
re
su
l
ts
th
a
n
o
t
h
e
r
m
e
th
o
d
s,
wi
th
a
n
a
c
c
u
ra
c
y
o
f
9
9
.
8
9
%
.
K
ey
w
o
r
d
s
:
C
r
o
s
s
-
s
ite
s
cr
ip
tin
g
C
y
b
er
s
ec
u
r
ity
Dee
p
lear
n
in
g
Fas
tTe
x
t
L
o
n
g
-
s
h
o
r
t te
r
m
m
e
m
o
r
y
W
o
r
d
em
b
ed
d
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
:
Mu
h
am
m
ad
Alk
h
air
i M
ash
u
r
i
Ma
s
ter
o
f
C
o
m
p
u
ter
Scie
n
ce
,
B
I
NUS
Gr
ad
u
ate
Pro
g
r
am
,
B
in
a
Nu
s
an
tar
a
Un
iv
er
s
ity
J
ak
ar
ta,
I
n
d
o
n
esia
E
m
ail:
m
.
m
ash
u
r
i@
b
in
u
s
.
ac
.
i
d
1.
I
NT
RO
D
UCT
I
O
N
T
h
e
g
r
o
win
g
u
s
e
o
f
web
ap
p
licatio
n
s
h
as
r
esu
lted
in
a
r
is
e
i
n
s
ec
u
r
ity
v
u
ln
er
a
b
ilit
ies,
with
c
r
o
s
s
-
s
ite
s
cr
ip
tin
g
(
XSS)
attac
k
s
p
o
s
in
g
a
m
ajo
r
r
is
k
to
d
ata
in
teg
r
i
ty
an
d
p
r
iv
ac
y
.
T
h
ese
attac
k
s
tak
e
ad
v
an
tag
e
o
f
clien
t
-
s
id
e
s
cr
ip
tin
g
wea
k
n
es
s
es,
m
ak
in
g
web
ap
p
licatio
n
s
a
f
r
eq
u
en
t
tar
g
et
[
1
]
.
T
h
e
co
n
s
eq
u
en
ce
s
o
f
a
s
u
cc
ess
f
u
l
XSS
attac
k
ca
n
in
cl
u
d
e
s
ess
io
n
h
ijack
in
g
,
s
en
s
itiv
e
d
ata
ex
p
o
s
u
r
e,
an
d
im
p
e
r
s
o
n
atio
n
o
f
th
e
v
ictim
an
d
ca
n
ev
e
n
lead
to
c
o
d
e
ex
e
cu
tio
n
o
n
t
h
e
s
er
v
er
,
d
ep
en
d
in
g
o
n
th
e
a
p
p
licatio
n
an
d
u
s
er
ac
co
u
n
t
p
r
i
v
ileg
es
[
2
]
.
XSS
attac
k
s
ca
n
b
e
g
e
n
er
ally
d
iv
id
ed
in
to
t
h
r
ee
g
r
o
u
p
s
:
p
er
s
is
ten
t,
n
o
n
-
p
er
s
is
ten
t,
an
d
d
o
c
u
m
en
t
o
b
ject
m
o
d
el
(
DOM
)
b
ased
attac
k
s
[
3
]
.
No
n
-
p
er
s
is
ten
t
attac
k
s
a
r
e
also
ca
lled
r
ef
lecte
d
XSS,
wh
ich
ar
e
attac
k
s
ca
r
r
ied
o
u
t
b
y
in
s
er
tin
g
m
alic
io
u
s
s
cr
ip
ts
in
to
th
e
u
n
if
o
r
m
r
eso
u
r
ce
lo
ca
to
r
(
UR
L
)
.
W
h
ile
p
er
s
is
ten
t
attac
k
s
ar
e
also
ca
lled
s
to
r
ed
XSS,
wh
ich
ar
e
attac
k
s
th
at
in
s
er
t m
alicio
u
s
s
cr
ip
ts
in
to
s
y
s
tem
f
iles
,
d
atab
ases
,
o
r
o
th
er
lo
ca
tio
n
s
m
an
ag
ed
b
y
th
e
s
er
v
er
an
d
will b
e
d
is
p
lay
ed
to
u
s
e
r
s
.
DOM
b
ased
XSS
is
an
atta
ck
th
at
ch
an
g
es th
e
DOM
en
v
ir
o
n
m
e
n
t.
T
h
e
o
p
e
n
web
ap
p
licatio
n
s
ec
u
r
ity
p
r
o
ject
(
OW
ASP)
s
t
ates
th
at
XSS
is
o
n
e
o
f
th
e
to
p
te
n
v
u
ln
er
ab
ilit
ies
o
n
we
b
s
ites
an
d
XSS
is
r
an
k
ed
s
ev
en
t
h
,
in
a
d
d
itio
n
it
is
n
o
te
d
th
at
v
u
ln
er
ab
ilit
ies
to
XSS
ar
e
in
ab
o
u
t
two
-
th
ir
d
s
o
f
all
web
ap
p
licatio
n
s
.
I
n
2
0
2
1
,
XSS
is
in
clu
d
ed
in
o
n
e
o
f
th
e
attac
k
s
in
th
e
i
n
jectio
n
ca
teg
o
r
y
wh
ich
is
r
a
n
k
ed
th
ir
d
[
4
]
.
I
n
an
XSS
attac
k
,
attac
k
er
s
ex
p
lo
it
v
u
l
n
er
ab
ilit
ies
b
y
in
jectin
g
m
alicio
u
s
s
cr
ip
ts
in
to
h
y
p
er
te
x
t
m
ar
k
u
p
lan
g
u
ag
e
(
HT
ML
)
web
p
ag
es.
W
h
en
u
s
er
s
ac
ce
s
s
th
ese
co
m
p
r
o
m
is
ed
p
a
g
es,
th
e
b
r
o
wser
ex
ec
u
tes
th
e
m
alicio
u
s
co
d
e,
allo
win
g
attac
k
er
s
to
h
ijack
web
s
ess
io
n
s
.
R
ec
en
tl
y
,
th
e
p
r
e
v
alen
ce
o
f
XSS
th
r
ea
ts
h
as
g
r
o
wn
s
ig
n
if
i
ca
n
tly
,
d
r
awin
g
in
cr
ea
s
in
g
att
en
tio
n
f
r
o
m
b
o
th
in
d
u
s
tr
y
an
d
ac
ad
em
ia.
T
h
is
h
as
m
ad
e
XSS
v
u
ln
er
ab
ilit
ies
a
k
ey
f
o
c
u
s
in
cy
b
e
r
s
ec
u
r
ity
r
esear
ch
,
with
n
u
m
er
o
u
s
s
tu
d
ies
d
ed
icate
d
to
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.
14
,
No
.
6
,
Dec
em
b
er
20
25
:
4
9
2
3
-
4
9
3
2
4924
en
h
an
cin
g
d
etec
tio
n
a
n
d
m
itig
atio
n
s
tr
ateg
ies.
C
o
n
s
eq
u
en
tl
y
,
th
e
r
is
in
g
s
ev
er
ity
o
f
XSS
attac
k
s
u
n
d
er
s
co
r
es
th
e
u
r
g
en
t
n
ee
d
f
o
r
r
o
b
u
s
t
d
e
f
en
s
e
m
ec
h
an
is
m
s
[
5
]
.
XSS
ca
n
also
b
e
ex
p
lo
ited
in
co
n
j
u
n
ctio
n
with
o
th
er
v
u
ln
er
ab
ilit
ies,
s
u
ch
as
cr
o
s
s
-
s
ite
r
eq
u
est
f
o
r
g
er
y
(
C
SR
F),
an
d
r
em
o
te
co
d
e
ex
ec
u
tio
n
(
R
C
E
)
,
p
o
ten
tially
lead
in
g
to
m
o
r
e
s
ig
n
if
ican
t
th
r
ea
ts
an
d
h
ar
m
t
o
th
e
v
ictim
'
s
in
ter
n
al
n
etwo
r
k
[
6
]
.
I
n
Ma
r
c
h
2
0
2
4
,
t
h
e
Secu
r
e
L
i
s
t
b
y
Kasp
er
s
k
y
web
s
ite
p
u
b
l
is
h
ed
a
r
ep
o
r
t
with
r
esear
ch
d
ata
f
r
o
m
2
0
2
1
to
2
0
2
3
s
tatin
g
th
at
XSS
v
u
ln
er
ab
ilit
ies
wer
e
f
o
u
n
d
o
n
6
1
%
o
f
t
h
e
an
aly
ze
d
we
b
s
ites
[
7
]
.
XSS
r
an
k
ed
2
nd
o
u
t
o
f
t
h
e
2
5
m
o
s
t
d
an
g
er
o
u
s
s
o
f
twar
e
v
u
ln
er
a
b
ilit
ies
in
2
0
2
3
[
8
]
,
a
n
d
h
as
r
ec
ei
v
ed
th
e
s
am
e
an
d
h
ig
h
e
r
r
an
k
in
g
in
p
r
e
v
io
u
s
y
ea
r
s
b
y
tak
in
g
f
ir
s
t
p
lace
in
2
0
2
0
w
ith
th
e
h
ig
h
est
o
v
er
all
s
co
r
e
in
p
r
e
v
alen
ce
a
n
d
s
ev
er
ity
.
Acc
o
r
d
in
g
to
C
is
co
's
2
0
1
8
an
n
u
al
s
ec
u
r
ity
r
ep
o
r
t,
it
s
h
o
ws
th
at
all
we
b
ap
p
l
icatio
n
s
th
at
th
ey
an
aly
ze
d
is
h
av
in
g
at
least
o
n
e
v
u
ln
er
ab
ilit
y
.
T
h
e
r
ep
o
r
t
a
ls
o
s
h
o
win
g
th
e
web
attac
k
s
ar
e
b
ec
o
m
in
g
m
o
r
e
f
r
eq
u
e
n
t,
m
o
r
e
s
p
ec
ialized
,
an
d
m
o
r
e
tech
n
ically
s
o
p
h
is
ticated
.
Ad
d
itio
n
ally
,
4
0
%
o
f
all
attac
k
attem
p
ts
in
v
o
lv
ed
lead
to
a
tec
h
n
iq
u
e
k
n
o
wn
as
XSS
,
m
ak
in
g
it
o
n
e
o
f
th
e
m
o
s
t w
id
ely
u
s
ed
tech
n
iq
u
e
[
9
]
.
A
v
ar
iety
o
f
p
r
ev
e
n
tio
n
an
d
m
itig
atio
n
m
eth
o
d
s
h
av
e
b
ee
n
p
r
o
p
o
s
ed
to
co
m
b
at
an
ex
is
tin
g
XSS
attac
k
s
,
wh
ich
wer
e
class
if
ied
in
to
th
e
f
o
llo
win
g
ca
t
eg
o
r
ies:
m
ac
h
in
e
lear
n
in
g
tec
h
n
iq
u
e,
clien
t
-
s
ize
tech
n
iq
u
e,
p
r
o
x
y
-
s
id
e
tech
n
iq
u
e,
s
er
v
er
-
s
ize
tech
n
iq
u
e,
a
n
d
co
m
b
in
ed
tech
n
iq
u
e
[
1
0
]
.
Ho
wev
er
,
th
e
p
r
o
p
o
s
ed
s
o
lu
tio
n
s
ar
e
b
ec
o
m
in
g
in
s
u
f
f
icien
t
d
u
e
to
t
h
e
in
cr
ea
s
in
g
l
y
s
o
p
h
is
ticated
f
o
r
m
o
f
XSS
p
ay
lo
ad
s
o
v
e
r
tim
e
[
1
1
]
,
r
esu
ltin
g
in
n
o
n
-
n
eg
lig
ib
le
f
alse
p
o
s
itiv
e
ca
s
es.
XSS
b
ec
am
e
th
e
m
o
s
t
wid
esp
r
ea
d
f
o
r
m
o
f
attac
k
v
ec
to
r
in
2
0
1
9
.
R
esear
ch
b
y
p
r
ec
is
e
s
ec
u
r
ity
[
1
2
]
s
tates
th
at
XSS
ac
co
u
n
ted
f
o
r
alm
o
s
t
4
0
%
o
f
all
attac
k
s
r
ec
o
r
d
e
d
b
y
s
ec
u
r
ity
ex
p
e
r
ts
.
Ar
o
u
n
d
7
5
%
o
f
co
m
p
a
n
ies
ac
r
o
s
s
No
r
th
Am
er
ica
an
d
E
u
r
o
p
e
h
av
e
al
s
o
b
ee
n
tar
g
eted
b
y
th
is
attac
k
d
u
r
in
g
2
0
1
9
.
I
n
ad
d
itio
n
,
th
e
Natio
n
al
Vu
ln
er
ab
i
liti
es
Data
b
ase
al
s
o
s
tated
th
at
th
e
to
tal
n
u
m
b
er
o
f
XSS
attac
k
s
in
2
0
1
9
in
cr
ea
s
ed
b
y
3
0
.
2
%
co
m
p
ar
ed
to
2
0
1
8
,
an
d
7
9
.
2
%
co
m
p
a
r
ed
to
2
0
1
7
.
T
h
e
th
r
ea
t
o
f
XSS
attac
k
s
is
wid
esp
r
ea
d
,
r
esu
ltin
g
in
s
ev
er
al
m
ajo
r
in
cid
en
ts
.
L
ik
e
in
2
0
1
8
,
B
r
itis
h
Air
way
s
was
co
m
p
r
o
m
is
ed
Ma
g
ec
ar
t,
a
h
ac
k
er
g
r
o
u
p
s
p
e
cializin
g
in
cr
ed
it
ca
r
d
th
ef
t.
T
h
e
attac
k
er
s
ex
p
lo
ited
a
XSS
v
u
ln
er
ab
ilit
y
in
th
e
J
av
ascr
ip
t
lib
r
ar
y
th
at
u
tili
ze
d
o
n
th
e
B
r
itis
h
Air
way
s
web
s
ite,
it
en
ab
lin
g
th
e
attac
k
er
to
u
s
e
th
e
s
cr
ip
t
to
ex
f
iltra
te
cu
s
to
m
er
d
ata
to
th
e
attac
k
er
'
r
em
o
te
s
er
v
er
,
th
is
b
r
ea
ch
led
to
th
e
th
ef
t
o
r
f
cr
ed
i
t
ca
r
d
s
in
f
o
r
m
atio
n
f
r
o
m
ap
p
r
o
x
im
ately
3
8
0
,
0
0
0
b
o
o
k
i
n
g
tr
a
n
s
ac
tio
n
s
b
ef
o
r
e
it
was
d
etec
ted
.
Similar
ly
in
2
0
1
9
,
Fo
r
t
n
ite,
a
m
u
ltip
lay
er
g
am
e
with
m
o
r
e
t
h
an
2
0
0
m
illi
o
n
u
s
er
s
,
was
f
o
u
n
d
to
h
av
e
an
XSS
v
u
ln
e
r
ab
i
lity
o
n
an
u
n
s
ec
u
r
ed
web
p
ag
e.
Fu
r
th
er
m
o
r
e,
i
n
ea
r
ly
2
0
1
6
,
eBay
id
en
tifie
d
a
f
atal
XSS
v
u
ln
er
ab
ilit
y
ca
u
s
ed
b
y
im
p
r
o
p
er
ly
v
alid
ated
UR
L
p
ar
am
eter
,
wh
i
ch
allo
wed
attac
k
er
g
ai
n
f
u
ll
a
cc
ess
to
th
e
s
eller
ac
co
u
n
t,
co
n
d
u
ct
u
n
au
th
o
r
ized
tr
an
s
ac
tio
n
s
,
an
d
s
tealin
g
p
a
y
m
en
t in
f
o
r
m
atio
n
d
etails
[
1
3
]
.
Sev
er
al
s
tu
d
ies
h
av
e
b
ee
n
c
o
n
d
u
cted
u
s
in
g
ar
tific
ial
in
tellig
en
ce
(
AI
)
to
p
e
r
f
o
r
m
d
etec
tio
n
,
s
u
ch
as
th
e
u
s
e
o
f
m
ac
h
in
e
lear
n
in
g
a
n
d
d
ee
p
lear
n
in
g
.
Dee
p
lea
r
n
i
n
g
ca
n
b
e
ap
p
lied
as
o
n
e
o
f
t
h
e
ef
f
o
r
ts
to
p
r
ev
e
n
t
XSS
attac
k
s
.
Dee
p
lear
n
in
g
h
as
h
ig
h
er
ca
p
ab
ilit
ies
an
d
f
lex
ib
ilit
y
in
p
r
o
ce
s
s
in
g
a
n
u
m
b
er
o
f
f
ea
tu
r
es
in
u
n
s
tr
u
ctu
r
ed
d
ata.
I
n
alg
o
r
ith
m
s
in
d
ee
p
lear
n
in
g
,
d
ata
is
p
r
o
ce
s
s
ed
th
r
o
u
g
h
m
u
ltip
le
lay
e
r
s
,
wh
er
e
ea
ch
lay
er
is
ab
le
p
r
o
g
r
ess
iv
ely
ex
tr
ac
t
an
d
r
ef
in
es
f
ea
tu
r
es
b
ef
o
r
e
p
ass
in
g
th
em
o
n
to
th
e
n
ex
t
lay
er
.
Var
io
u
s
ar
ch
itectu
r
es
ca
n
b
e
u
s
ed
t
o
im
p
lem
en
t
d
ee
p
lear
n
in
g
,
in
clu
d
in
g
u
n
s
u
p
e
r
v
is
ed
p
r
e
-
tr
ain
ed
n
etwo
r
k
s
,
co
n
v
o
l
u
tio
n
al
n
e
u
r
al
n
etwo
r
k
s
(
C
NN)
,
r
ec
u
r
r
en
t
n
e
u
r
al
n
etwo
r
k
s
(
R
NN)
,
an
d
r
ec
u
r
s
iv
e
n
e
u
r
al
n
etwo
r
k
s
[
1
4
]
.
Ho
wev
er
,
alth
o
u
g
h
AI
ca
n
s
ig
n
if
ican
tly
s
o
lv
e
th
is
p
r
o
b
lem
,
th
er
e
ar
e
s
till
f
u
n
d
a
m
en
tal
s
h
o
r
tco
m
in
g
s
s
u
ch
as
th
e
f
alse
n
e
g
ativ
e
r
atio
.
Fals
e
n
eg
ativ
e
is
a
b
ig
g
e
r
p
r
o
b
lem
th
an
f
alse
p
o
s
itiv
e
,
b
ec
a
u
s
e
m
an
y
attac
k
s
tr
y
to
av
o
id
th
e
d
etec
tio
n
s
y
s
tem
,
w
h
ich
u
ltima
tely
r
esu
lts
in
r
ea
l t
h
r
ea
ts
an
d
co
m
p
r
o
m
is
ed
s
ec
u
r
ity
s
y
s
tem
s
[
1
5
]
.
Ma
n
y
o
f
s
tu
d
ies
u
s
in
g
th
e
W
o
r
d
2
Vec
wo
r
d
em
b
e
d
d
in
g
m
eth
o
d
.
W
o
r
d
2
Vec
is
an
alg
o
r
ith
m
th
at
m
ap
s
ea
ch
wo
r
d
in
a
tex
t
in
to
a
v
ec
to
r
f
o
r
m
.
T
h
is
alg
o
r
ith
m
was
in
tr
o
d
u
ce
d
in
2
0
1
3
with
t
wo
m
ain
m
eth
o
d
s
,
n
am
ely
s
k
ip
-
g
r
am
a
n
d
co
n
tin
u
o
u
s
b
ag
o
f
w
o
r
d
s
(
C
B
OW
)
[
1
6
]
.
Un
til
n
o
w,
th
is
wo
r
d
em
b
ed
d
in
g
m
o
d
el
h
as
b
ee
n
wid
ely
u
s
ed
in
n
atu
r
al
l
an
g
u
ag
e
p
r
o
ce
s
s
in
g
(
NL
P)
r
e
s
ea
r
ch
.
T
h
e
W
o
r
d
2
Vec
m
et
h
o
d
g
en
e
r
ates
wo
r
d
em
b
ed
d
in
g
s
u
s
in
g
a
d
en
s
e
r
ep
r
esen
tatio
n
.
As
a
p
r
ed
ictiv
e
m
o
d
el,
it
ass
ig
n
s
p
r
o
b
ab
ilit
ies
t
o
wo
r
d
s
,
m
ak
in
g
it
h
ig
h
ly
e
f
f
ec
tiv
e
in
wo
r
d
s
im
ilar
ity
task
s
[
1
7
]
.
Ho
wev
er
,
W
o
r
d
2
Vec
is
u
n
a
b
le
to
lea
r
n
s
y
n
t
ac
tic
r
elatio
n
s
h
ip
s
,
th
e
s
tr
u
ctu
r
e
o
f
W
o
r
d
2
Vec
p
r
o
v
es
to
b
e
ex
tr
em
ely
d
ep
e
n
d
e
n
t
o
n
t
h
e
co
r
p
u
s
,
m
a
k
in
g
s
em
an
tic
p
r
o
x
im
ity
o
n
ly
a
s
id
e
ef
f
ec
t o
f
its
tr
u
e
o
b
jecti
v
e
f
u
n
ctio
n
[
1
8
]
.
I
n
ad
d
itio
n
to
W
o
r
d
2
Vec
,
t
h
e
r
e
ar
e
o
th
er
wo
r
d
em
b
ed
d
i
n
g
tech
n
iq
u
es,
n
am
el
y
Glo
Ve
wh
ich
was
in
tr
o
d
u
ce
d
in
2
0
1
4
an
d
Fas
tTe
x
t
wh
ich
was
in
tr
o
d
u
ce
d
in
2
0
1
7
.
Glo
Ve
is
d
if
f
er
en
t
f
r
o
m
W
o
r
d
2
Vec
wh
ich
o
n
ly
r
elies
o
n
lo
ca
l
in
f
o
r
m
atio
n
f
r
o
m
wo
r
d
s
with
lo
ca
l
co
n
te
x
t
win
d
o
ws
(
s
k
ip
-
g
r
am
a
n
d
C
B
OW
)
,
Glo
Ve
a
ls
o
co
m
b
in
es
wo
r
d
c
o
-
o
cc
u
r
r
en
ce
in
f
o
r
m
atio
n
o
r
g
lo
b
al
s
tatis
tic
s
to
o
b
tain
s
em
an
tic
r
elatio
n
s
h
ip
s
b
etwe
en
wo
r
d
s
[
1
9
]
.
W
h
ile
Fas
tTe
x
t
is
a
d
ev
elo
p
m
en
t
o
f
W
o
r
d
2
Vec
,
th
is
m
eth
o
d
lear
n
s
wo
r
d
s
b
y
co
n
s
i
d
er
in
g
in
f
o
r
m
atio
n
f
r
o
m
s
y
llab
les.
E
ac
h
wo
r
d
is
r
ep
r
esen
ted
as
a
s
et
o
f
n
-
g
r
a
m
ch
ar
ac
ter
s
.
T
h
is
h
elp
s
in
t
h
e
s
en
s
e
o
f
s
h
o
r
ter
wo
r
d
s
an
d
allo
ws em
b
e
d
d
in
g
t
o
u
n
d
e
r
s
tan
d
th
e
s
u
f
f
ix
an
d
p
r
ef
ix
o
f
t
h
e
wo
r
d
[
2
0
]
.
Fas
tTe
x
t
h
as
th
e
ad
v
an
tag
e
o
f
b
ein
g
ab
le
t
o
p
r
o
v
id
e
a
r
ep
r
e
s
en
tatio
n
o
f
wo
r
d
s
th
at
d
o
n
o
t
ap
p
ea
r
in
th
e
tr
ain
in
g
d
ata,
o
r
b
ein
g
ab
le
to
o
v
er
co
m
e
th
e
p
r
o
b
lem
o
f
o
u
t
o
f
v
o
ca
b
u
lar
y
.
I
f
it
d
o
es
n
o
t
f
in
d
a
wo
r
d
th
a
t
is
n
o
t
in
th
e
tr
ain
in
g
p
r
o
ce
s
s
,
th
en
th
e
wo
r
d
will
b
e
b
r
o
k
en
d
o
wn
in
to
n
-
g
r
am
s
in
th
e
f
o
r
m
o
f
a
co
llectio
n
o
f
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
E
n
h
a
n
ci
n
g
cro
s
s
-
s
ite
s
crip
tin
g
a
tta
ck
d
etec
tio
n
b
y
u
s
in
g
F
a
s
tText
a
s
w
o
r
d
…
(
Mu
h
a
mma
d
A
lkh
a
ir
i
)
4925
s
y
llab
le
s
eq
u
en
ce
s
to
g
et
its
em
b
ed
d
in
g
v
ec
to
r
.
Fas
tTe
x
t
h
as
g
o
o
d
p
e
r
f
o
r
m
an
ce
,
an
d
c
an
tr
ain
m
o
d
els
o
n
lar
g
e
d
atasets
q
u
ick
ly
,
a
n
d
is
a
b
le
to
p
r
o
v
id
e
a
r
ep
r
esen
tatio
n
o
f
wo
r
d
s
th
at
d
o
n
o
t a
p
p
ea
r
in
th
e
tr
ain
in
g
d
ata.
T
h
er
e
ar
e
s
ev
er
al
p
r
ev
io
u
s
s
tu
d
ies
r
elate
d
to
XSS
d
etec
tio
n
.
I
n
g
en
er
al,
t
h
e
s
o
lu
tio
n
s
o
f
f
er
ed
ar
e
s
im
ilar
to
ea
ch
o
th
e
r
.
R
esear
ch
b
y
B
ak
ir
an
d
B
ak
ir
[
2
1
]
,
co
n
d
u
cted
XSS
attac
k
d
etec
tio
n
u
s
in
g
s
ev
e
r
al
m
ac
h
in
e
lear
n
in
g
an
d
d
ee
p
l
ea
r
n
in
g
m
o
d
els
,
co
m
b
in
ed
w
ith
W
o
r
d
2
Vec
wo
r
d
em
b
ed
d
in
g
an
d
u
n
iv
e
r
s
al
s
en
ten
ce
en
co
d
er
(
USE)
.
T
h
e
r
esu
lts
o
f
d
ee
p
lear
n
in
g
an
d
W
o
r
d
2
Vec
,
lo
n
g
-
s
h
o
r
t
ter
m
m
em
o
r
y
(
L
STM
)
o
b
tain
ed
a
n
ac
c
u
r
ac
y
o
f
9
3
.
9
4
%,
an
d
an
ac
c
u
r
ac
y
o
f
9
7
.
6
6
%
with
C
NN.
T
h
e
r
esu
lts
o
f
d
ee
p
lear
n
in
g
a
n
d
USE
an
d
W
o
r
d
2
Vec
,
o
b
tain
ed
an
ac
cu
r
ac
y
o
f
9
8
.
4
7
% with
L
STM
,
an
d
9
9
.
1
6
% with
Van
i
lla
NN
2
.
R
esear
ch
co
n
d
u
cted
b
y
T
a
d
h
an
i
et
a
l.
[
2
2
]
,
p
r
o
p
o
s
ed
a
s
o
lu
tio
n
u
s
in
g
C
NN
an
d
L
STM
to
d
etec
t
SQL
in
jectio
n
an
d
XSS.
T
h
e
p
r
o
ce
s
s
b
eg
in
s
with
p
r
ep
r
o
ce
s
s
in
g
in
p
u
t
d
ata,
wh
ich
i
n
clu
d
es
d
ec
o
d
i
n
g
,
to
k
en
izatio
n
,
an
d
g
en
e
r
aliza
tio
n
tech
n
iq
u
es.
T
h
en
,
th
e
p
r
o
c
ess
ed
d
ata
is
f
ed
in
to
th
e
C
N
N
m
o
d
el
f
o
r
f
ea
tu
r
e
ex
tr
ac
tio
n
,
a
n
d
th
e
e
x
tr
ac
ted
f
ea
tu
r
es
ar
e
u
s
ed
f
o
r
tr
ain
in
g
th
e
L
STM
m
o
d
el.
Data
is
t
ak
en
f
r
o
m
v
ar
io
u
s
s
o
u
r
ce
s
,
s
u
ch
as
th
e
Natio
n
al
Vu
ln
er
ab
ilit
y
Data
b
ase,
an
d
also
OW
A
SP
.
T
h
e
W
o
r
d
2
Vec
m
o
d
el
is
u
s
ed
f
o
r
f
ea
tu
r
e
ex
tr
ac
tio
n
f
r
o
m
i
n
p
u
t
d
ata.
T
h
e
r
esu
lts
o
f
th
is
s
tu
d
y
,
th
e
C
NN
m
o
d
el
h
as
an
ac
cu
r
ac
y
o
f
9
9
.
5
%,
an
d
L
STM
9
8
.
6
9
%,
a
n
d
C
NN
with
L
STM
g
ets
an
ac
cu
r
ac
y
o
f
9
9
.
8
4
%
with
its
o
wn
test
b
ed
d
a
taset.
T
h
e
r
esu
lts
o
f
th
is
s
tu
d
y
also
s
u
g
g
est
th
at
f
u
r
th
er
s
tu
d
ies
ca
n
lo
o
k
at
o
th
er
ar
ch
itectu
r
es
an
d
m
eth
o
d
s
to
co
m
b
in
e
C
NN
an
d
L
STM
m
o
d
els to
im
p
r
o
v
e
m
o
d
el
p
r
ec
is
io
n
an
d
r
o
b
u
s
tn
ess
.
I
n
a
s
t
u
d
y
c
o
n
d
u
c
t
e
d
b
y
Y
o
u
n
as
e
t
a
l
.
[
1
3
]
t
r
a
i
n
e
d
a
s
et
o
f
1
3
,
6
8
6
r
e
c
o
r
d
s
d
a
t
a
f
r
o
m
K
a
g
g
l
e
,
p
r
e
s
e
n
t
i
n
g
a
n
a
p
p
r
o
a
c
h
t
h
a
t
u
s
e
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
t
h
at
i
n
c
l
u
d
es
L
S
T
M
t
o
c
r
ea
t
e
e
a
r
ly
d
e
t
e
c
t
i
o
n
o
f
XS
S
a
t
t
a
c
k
s
,
b
y
c
o
m
p
a
r
i
n
g
a
n
d
ap
p
l
y
i
n
g
t
e
r
m
f
r
e
q
u
e
n
c
y
-
i
n
v
e
r
s
e
d
o
c
u
m
e
n
t
f
r
e
q
u
e
n
c
y
(
T
F
I
D
F
)
t
o
e
v
a
l
u
a
te
t
h
e
r
e
s
u
l
ts
.
L
S
T
M
-
T
F
I
D
F
f
o
r
f
e
a
tu
r
e
e
x
t
r
a
c
t
i
o
n
,
u
s
i
n
g
b
o
t
h
t
e
m
p
o
r
a
l
f
e
a
t
u
r
e
s
a
n
d
T
F
I
D
F
f
r
o
m
t
h
e
XS
S
d
a
t
as
e
t,
r
e
s
u
l
t
i
n
g
i
n
a
n
ew
s
et
o
f
f
e
a
t
u
r
e
s
.
T
h
e
r
e
a
r
e
a
ls
o
s
e
v
e
r
a
l
a
l
g
o
r
i
t
h
m
s
u
s
e
d
i
n
t
h
i
s
s
t
u
d
y
,
s
u
ch
a
s
r
a
n
d
o
m
f
o
r
e
s
t
(
R
F
)
,
l
o
g
is
ti
c
r
e
g
r
e
s
s
i
o
n
,
Ga
u
s
s
i
a
n
n
a
i
v
e
B
a
y
es
(
G
NB
)
,
d
e
cis
i
o
n
t
r
e
e
(
DT
)
,
g
a
t
e
d
r
e
c
u
r
r
e
n
t
u
n
i
t
.
W
i
t
h
f
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
f
r
o
m
L
S
T
M
-
T
F
I
DF
,
a
n
a
c
c
u
r
a
c
y
o
f
7
4
%
w
a
s
o
b
t
a
i
n
e
d
w
it
h
t
h
e
L
ST
M
m
o
d
e
l
.
T
o
p
e
r
f
o
r
m
d
at
a
e
x
t
r
a
c
t
i
o
n
,
t
h
e
s
t
u
d
y
c
o
n
d
u
c
t
e
d
w
i
t
h
t
h
e
b
a
g
o
f
w
o
r
d
s
(
B
o
W
)
te
c
h
n
i
q
u
e
f
o
r
f
e
a
t
u
r
e
e
x
t
r
a
c
ti
o
n
.
B
y
a
p
p
l
y
i
n
g
t
h
es
e
B
o
W
f
ea
t
u
r
e
s
,
t
h
e
r
e
s
u
lt
s
o
f
GN
B
,
D
T
,
a
n
d
R
F
g
e
tt
i
n
g
w
e
a
k
p
e
r
f
o
r
m
a
n
c
e
s
c
o
r
e
s
.
B
o
W
l
i
m
it
a
t
i
o
n
is
l
a
r
g
e
l
y
d
u
e
t
o
i
n
a
b
i
l
i
t
y
t
o
c
a
p
t
u
r
e
h
i
g
h
-
l
e
v
e
l
o
r
c
o
m
p
le
x
f
e
a
t
u
r
es
,
r
es
u
lt
in
g
i
n
l
o
w
p
e
r
f
o
r
m
a
n
c
e
s
c
o
r
e
s
f
o
r
m
a
n
y
m
e
t
h
o
d
s
.
F
o
r
f
u
r
t
h
e
r
i
m
p
r
o
v
e
m
e
n
t
,
a
l
t
e
r
n
a
t
i
v
e
e
x
p
l
o
r
a
t
i
o
n
o
f
f
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
t
e
c
h
n
i
q
u
es
is
n
e
e
d
e
d
.
Nex
t,
th
er
e
is
a
s
tu
d
y
co
n
d
u
ct
ed
b
y
Alao
u
i
an
d
Nf
ao
u
i
[
2
3
]
,
th
is
s
tu
d
y
p
r
o
p
o
s
es
an
ap
p
r
o
ac
h
b
ased
o
n
W
o
r
d
2
Vec
as
wo
r
d
em
b
e
d
d
in
g
,
an
d
a
s
tack
ed
g
en
er
atio
n
al
en
s
em
b
le
m
o
d
el
f
o
r
L
STM
to
d
etec
t
m
alicio
u
s
HT
T
P
R
eq
u
ests
.
W
o
r
d
2
Vec
i
s
a
two
-
lay
e
r
n
e
u
r
al
n
etwo
r
k
t
h
at
tak
es
a
tex
t
co
r
p
u
s
as
in
p
u
t
an
d
o
u
tp
u
ts
a
s
et
o
f
v
ec
t
o
r
s
.
I
t
in
clu
d
es
two
m
o
d
els,
C
B
OW
an
d
s
k
ip
-
g
r
a
m
.
T
h
is
p
a
p
er
u
s
es
th
e
C
B
OW
m
o
d
el
b
ec
au
s
e
it
is
u
s
u
ally
f
aster
an
d
m
o
r
e
ac
c
u
r
ate
th
an
th
e
s
k
ip
-
g
r
am
m
o
d
el.
T
h
is
s
tu
d
y
m
o
d
if
ies
th
e
tr
ain
in
g
p
ar
am
ete
r
v
alu
es,
n
am
ely
win
d
o
w
s
ize
an
d
em
b
ed
d
in
g
s
ize,
to
o
b
tain
d
if
f
er
en
t
v
ec
to
r
r
ep
r
esen
tatio
n
s
o
f
wo
r
d
s
with
th
e
s
am
e
wo
r
d
.
W
ith
th
e
s
tack
in
g
en
s
em
b
le
o
f
L
STM
m
o
d
els,
th
e
h
ig
h
est ac
cu
r
ac
y
r
esu
lt is
7
8
.
9
5
%.
T
h
er
e
is
also
a
s
tu
d
y
co
n
d
u
ct
ed
b
y
Go
g
o
i
et
a
l.
[
2
]
t
h
at
p
r
o
p
o
s
es
XSS
attac
k
d
etec
tio
n
u
s
in
g
th
e
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM)
alg
o
r
ith
m
.
T
h
is
s
tu
d
y
u
s
es
T
FID
F
to
p
er
f
o
r
m
f
ea
tu
r
e
ex
tr
ac
tio
n
f
r
o
m
to
k
en
ized
tex
t.
W
ith
lin
ea
r
an
d
n
o
n
-
lin
ea
r
SVM,
th
e
class
if
icatio
n
r
esu
lts
g
et
a
p
r
ec
is
io
n
o
f
0
.
9
7
,
r
ec
all
0
.
9
9
,
an
d
F1
s
co
r
e
0
.
9
7
.
T
h
en
th
er
e
is
an
o
th
er
s
tu
d
y
c
o
n
d
u
cte
d
b
y
L
en
te
et
a
l.
[
2
4
]
,
th
is
s
tu
d
y
co
m
b
i
n
es
C
NN
wi
th
L
STM
,
an
d
ac
h
iev
es
9
9
.
3
6
%
ac
cu
r
ac
y
wh
en
p
r
ed
ictin
g
wh
eth
e
r
a
n
ew
UR
L
co
r
r
esp
o
n
d
s
to
a
n
XSS
attac
k
.
T
h
e
d
ataset
co
n
s
is
ts
o
f
s
am
p
les
o
f
3
3
,
4
2
6
XSS
attac
k
s
ex
tr
ac
ted
f
r
o
m
th
e
XSSed
d
atab
ase
at
h
ttp
://www.
x
s
s
ed
.
co
m
/ a
n
d
3
1
,
4
0
7
r
eg
u
lar
UR
L
s
ex
tr
ac
ted
f
r
o
m
th
e
DM
OZ
d
atab
ase
at
h
tt
p
://d
m
o
zto
o
ls
.
n
et/
.
T
h
is
s
tu
d
y
co
n
s
is
ts
o
f
f
o
u
r
m
ain
s
tep
s
:
p
r
o
ce
s
s
in
g
in
p
u
t
d
ata,
W
o
r
d
2
Vec
tr
an
s
f
o
r
m
,
c
o
n
v
o
lu
tio
n
,
an
d
L
STM
.
B
ased
o
n
p
r
ev
io
u
s
r
esear
ch
,
th
is
s
tu
d
y
p
r
o
p
o
s
es
a
XSS
attac
k
d
etec
to
r
th
at
u
s
es
d
ee
p
lear
n
in
g
with
th
e
L
STM
m
eth
o
d
wh
ic
h
is
a
ty
p
e
o
f
R
NN.
L
STM
s
to
r
es
in
f
o
r
m
atio
n
a
b
o
u
t
d
ata
p
atter
n
s
an
d
ca
n
als
o
lear
n
wh
ich
d
ata
s
h
o
u
ld
b
e
k
e
p
t
o
r
d
is
ca
r
d
ed
,
b
ec
au
s
e
ea
ch
L
STM
n
eu
r
o
n
h
a
s
s
ev
er
al
g
ates
th
at
r
eg
u
late
th
e
m
em
o
r
y
o
f
ea
c
h
n
eu
r
o
n
its
elf
[
2
5
]
.
T
h
e
f
ea
tu
r
e
ex
tr
ac
tio
n
m
eth
o
d
u
s
ed
is
Fa
s
tTe
x
t
wh
ich
aim
s
to
p
r
o
v
id
e
in
n
o
v
atio
n
in
s
cr
ip
t
f
ea
tu
r
es.
Fas
tTe
x
t
ex
ten
d
s
W
o
r
d
2
Vec
b
y
in
clu
d
in
g
s
u
b
wo
r
d
in
f
o
r
m
atio
n
.
T
h
is
f
ea
tu
r
e
r
e
p
r
esen
ts
wo
r
d
s
as
a
co
llectio
n
o
f
n
-
g
r
am
c
h
ar
ac
ter
s
,
allo
win
g
it
to
p
r
o
d
u
ce
em
b
e
d
d
in
g
s
th
at
ca
p
tu
r
e
m
o
r
p
h
o
l
o
g
ical
v
ar
iatio
n
s
an
d
h
an
d
le
r
a
r
e
o
r
m
is
s
p
elled
wo
r
d
s
m
o
r
e
ef
f
ec
tiv
ely
.
T
h
is
f
e
atu
r
e
is
v
er
y
u
s
ef
u
l
f
o
r
XSS
d
etec
tio
n
,
b
ec
au
s
e
attac
k
er
s
o
f
ten
u
s
e
cr
ea
tiv
e
o
b
f
u
s
ca
tio
n
tech
n
iq
u
es
i
n
s
cr
ip
ts
to
b
y
p
ass
s
ec
u
r
ity
f
ilter
s
.
So
t
h
at
it
ca
n
in
c
r
ea
s
e
th
e
ac
cu
r
ac
y
in
d
etec
tin
g
XSS
attac
k
s
.
W
ith
th
i
s
m
eth
o
d
,
it
is
ex
p
ec
ted
to
in
cr
ea
s
e
th
e
ac
cu
r
ac
y
in
d
etec
tin
g
XSS attac
k
s
in
th
is
s
tu
d
y
.
2.
M
E
T
H
O
D
T
h
e
r
esear
ch
s
tag
e
b
eg
in
s
with
d
ata
co
llectio
n
.
XSS
u
r
l
d
ata
is
tak
en
f
r
o
m
Kag
g
le,
w
h
ich
co
n
s
is
ts
o
f
1
3
,
6
8
6
d
ata
th
at
h
av
e
b
ee
n
lab
eled
as
XS
S
attac
k
s
an
d
n
o
n
-
XSS
attac
k
s
.
T
h
e
d
ata
is
tak
en
f
r
o
m
Po
r
tSwig
g
er
,
an
d
OW
ASP C
h
ea
t Sh
ee
t
s
f
o
r
XSS attac
k
s
,
wh
ich
co
n
s
is
ts
o
f
7
,
3
7
3
XSS attac
k
s
,
an
d
6
,
3
1
3
n
o
n
-
XSS attac
k
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.
14
,
No
.
6
,
Dec
em
b
er
20
25
:
4
9
2
3
-
4
9
3
2
4926
T
h
en
co
n
tin
u
ed
with
th
e
p
r
e
p
r
o
ce
s
s
in
g
s
tag
e,
th
e
d
ataset
is
c
lean
ed
f
r
o
m
n
o
is
e
o
r
th
in
g
s
th
at
h
av
e
n
o
ef
f
ec
t.
T
h
e
p
r
o
ce
s
s
es
th
at
will
b
e
ca
r
r
ied
o
u
t
ar
e
d
ec
o
d
e,
g
en
er
aliza
tio
n
,
an
d
to
k
en
izatio
n
.
Fu
r
th
er
m
o
r
e,
th
e
r
esu
lts
o
f
th
e
p
r
ep
r
o
ce
s
s
in
g
will
b
e
wo
r
d
em
b
ed
d
i
n
g
u
s
in
g
Fas
tTe
x
t.
T
h
e
m
o
d
el
th
at
will
b
e
u
s
ed
f
o
r
class
if
icatio
n
is
L
STM
,
an
d
th
en
all
th
e
r
esu
lts
will
b
e
an
aly
ze
d
to
m
ea
s
u
r
e
th
e
p
er
f
o
r
m
an
ce
an
d
ac
c
u
r
ac
y
lev
el
o
f
th
e
tex
t
class
if
icatio
n
m
o
d
el
u
s
in
g
L
STM
.
2
.
1
.
Da
t
a
s
et
T
h
e
d
ataset
u
s
ed
in
th
is
s
tu
d
y
u
s
es
a
d
ataset
f
o
r
d
ee
p
lear
n
in
g
p
u
b
lis
h
ed
o
n
th
e
Kag
g
le
r
ep
o
s
ito
r
y
at
h
ttp
s
://www.
k
ag
g
le.
co
m
/d
atasets
/s
y
ed
s
aq
lain
h
u
s
s
ain
/cr
o
s
s
-
s
ite
-
s
cr
ip
tin
g
-
x
s
s
-
d
ataset
-
f
o
r
-
d
ee
p
-
lear
n
in
g
.
T
h
e
d
ataset
co
n
s
is
ts
o
f
1
3
,
6
8
6
e
n
t
r
ies
co
llected
f
r
o
m
Po
r
tSwig
g
er
an
d
OW
ASP
C
h
ea
t
Sh
ee
ts
f
o
r
XSS
attac
k
s
.
T
h
is
d
ataset
alr
ea
d
y
in
clu
d
es
a
v
ar
iety
o
f
XSS
attac
k
v
ec
to
r
s
th
at
allo
w
th
is
s
tu
d
y
to
ass
e
s
s
th
e
ef
f
ec
tiv
en
ess
o
f
th
e
m
o
d
el
u
s
ed
in
d
etec
tin
g
v
ar
io
u
s
attac
k
s
ce
n
ar
i
o
s
.
T
h
e
r
e
ar
e
2
lab
els
in
th
e
d
ataset,
n
a
m
ely
XSS
an
d
n
o
t
XSS.
Of
th
e
1
3
,
6
8
6
e
n
tr
ies,
th
er
e
ar
e
7
,
3
7
3
en
t
r
ies
th
at
ar
e
XSS
attac
k
s
,
an
d
6
,
3
1
3
th
at
a
r
e
n
o
t
XSS
attac
k
s
,
o
r
5
3
.
8
% XSS a
ttack
s
an
d
4
6
.
2
% n
o
t X
SS
attac
k
s
.
Sam
p
les o
f
th
e
d
ata
ca
n
b
e
s
ee
n
in
T
ab
l
e
1
.
T
ab
le
1
.
Sam
p
le
d
ataset
S
e
n
t
e
n
c
e
X
S
S
<
d
i
v
d
r
a
g
g
a
b
l
e
=
"
t
r
u
e
"
c
o
n
t
e
n
t
e
d
i
t
a
b
l
e
>
d
r
a
g
m
e
<
/
d
i
v
>
<
h
e
a
d
e
r
o
n
d
r
o
p
=
a
l
e
r
t
(
1
)
c
o
n
t
e
n
t
e
d
i
t
a
b
l
e
>
d
r
o
p
h
e
r
e
<
/
h
e
a
d
e
r
>
1
<
i
n
s
o
n
c
u
t
=
"
a
l
e
r
t
(
1
)
"
c
o
n
t
e
n
t
e
d
i
t
a
b
l
e
>
t
e
s
t
<
/
i
n
s>
1
<
/
d
i
v
>
<
/
t
d
>
<
/
t
r
>
<
t
r
>
<
t
h
sc
o
p
e
=
"
r
o
w
"
c
l
a
ss
=
"
n
a
v
b
o
x
-
g
r
o
u
p
"
st
y
l
e
=
"
w
i
d
t
h
:
1
%">
<
a
h
r
e
f
=
"
/
w
i
k
i
/
A
l
g
o
r
i
t
h
m"
t
i
t
l
e
=
"
A
l
g
o
r
i
t
h
m"
>
A
l
g
o
r
i
t
h
ms
<
/
a
>
<
/
t
h
>
<
t
d
c
l
a
ss
=
"
n
a
v
b
o
x
-
l
i
st
n
a
v
b
o
x
-
e
v
e
n
h
l
i
s
t
"
st
y
l
e
=
"
t
e
x
t
-
a
l
i
g
n
:
l
e
f
t
;
b
o
r
d
e
r
-
l
e
f
t
-
w
i
d
t
h
:
2
p
x
;
b
o
r
d
e
r
-
l
e
f
t
-
st
y
l
e
:
s
o
l
i
d
;
w
i
d
t
h
:
1
0
0
%;
p
a
d
d
i
n
g
:
0
p
x
"
>
0
<
/
s
p
a
n
>
<
l
i
n
k
r
e
l
=
"
mw
-
d
e
d
u
p
l
i
c
a
t
e
d
-
i
n
l
i
n
e
-
st
y
l
e
"
h
r
e
f
=
"
mw
-
d
a
t
a
:
Te
mp
l
a
t
e
S
t
y
l
e
s:
r
9
3
5
2
4
3
6
0
8
"
/
>
0
<
su
mm
a
r
y
o
n
d
b
l
c
l
i
c
k
=
"
a
l
e
r
t
(
1
)
"
>
t
e
s
t
<
/
s
u
mm
a
r
y
>
1
2
.
2.
P
re
pro
ce
s
s
ing
Pre
p
r
o
ce
s
s
in
g
is
th
e
f
ir
s
t
s
ta
g
e
ca
r
r
ied
o
u
t
af
ter
o
b
tain
in
g
d
ata,
an
d
it
p
lay
s
a
cr
itic
al
r
o
le
in
p
r
ep
ar
in
g
th
e
r
aw
in
p
u
t
f
o
r
a
n
aly
s
is
.
T
h
is
s
tag
e
co
n
s
is
ts
o
f
th
r
ee
k
ey
s
tep
s
:
d
ec
o
d
in
g
,
g
en
er
aliza
tio
n
,
an
d
to
k
en
izatio
n
,
wh
ich
ar
e
e
x
p
la
in
ed
in
th
e
f
o
llo
win
g
s
u
b
s
ec
t
io
n
s
.
E
ac
h
s
tep
s
er
v
es
t
o
cle
an
,
tr
an
s
f
o
r
m
,
an
d
s
tr
u
ctu
r
e
th
e
d
ata,
en
s
u
r
in
g
it
is
in
a
s
u
itab
le
f
o
r
m
at
f
o
r
th
e
m
o
d
el
to
p
r
o
ce
s
s
ef
f
ec
tiv
ely
.
Pro
p
er
p
r
ep
r
o
ce
s
s
in
g
h
elp
s
m
i
n
im
iz
e
n
o
is
e,
r
ed
u
ce
ir
r
elev
a
n
t
in
f
o
r
m
atio
n
,
an
d
m
ai
n
tain
c
o
n
s
is
ten
cy
ac
r
o
s
s
th
e
d
ataset.
T
h
is
n
o
t
o
n
ly
e
n
h
an
c
es
th
e
p
er
f
o
r
m
an
ce
o
f
m
ac
h
in
e
lear
n
in
g
m
o
d
els
b
u
t
also
en
s
u
r
es
th
at
im
p
o
r
tan
t
f
ea
tu
r
es r
elate
d
to
th
e
d
etec
tio
n
o
f
XSS attac
k
s
ar
e
p
r
eser
v
e
d
an
d
h
ig
h
lig
h
ted
d
u
r
in
g
t
h
e
tr
ain
in
g
p
h
ase.
2
.
2.
1
.
Dec
o
din
g
Attack
er
s
ca
n
av
o
id
f
ilter
s
o
r
v
alid
atio
n
b
ased
o
n
r
eg
u
la
r
ex
p
r
ess
io
n
s
b
y
u
s
in
g
en
co
d
in
g
tech
n
iq
u
es
s
u
ch
as
Hex
en
co
d
in
g
,
UR
L
en
co
d
in
g
,
u
n
ico
d
e
en
co
d
i
n
g
,
a
n
d
HT
ML
en
tity
en
co
d
in
g
.
So
d
ec
o
d
in
g
is
n
ee
d
ed
to
r
esto
r
e
th
e
o
r
ig
in
al
v
alu
e.
T
h
e
T
ab
le
2
is
an
ex
am
p
le
o
f
a
co
m
p
ar
is
o
n
o
f
d
ata
b
e
f
o
r
e
an
d
af
ter
d
ec
o
d
in
g
f
r
o
m
th
e
d
ataset
u
s
ed
.
T
h
e
d
ata
u
s
es
en
co
d
i
n
g
in
in
s
er
tin
g
s
cr
ip
t
tag
s
,
an
d
is
s
u
cc
ess
f
u
lly
r
etu
r
n
ed
to
th
e
ac
tu
al
s
cr
ip
t ta
g
f
o
r
m
af
ter
d
ec
o
d
in
g
.
T
ab
le
2
.
T
h
e
co
m
p
ar
is
o
n
o
f
d
e
co
d
ed
r
esu
lts
O
r
i
g
i
n
a
l
t
e
x
t
D
e
c
o
d
e
d
t
e
x
t
<
A
H
R
EF="
h
t
t
p
:
/
/
%
7
7
%7
7
%7
7
%
2
E
%6
7
%6
F
%6
F
%6
7
%
6
C
%
6
5
%2
E%
6
3
%
6
F
%
6
D
"
>
<
/
A
>
<
A
H
R
EF="
h
t
t
p
:
/
/
w
w
w
.
g
o
o
g
l
e
.
c
o
m"
>
<
/
A
>
2
.
2
.
2
.
G
ener
a
liza
t
io
n
T
h
is
p
r
o
ce
s
s
is
ca
r
r
ied
o
u
t
af
ter
th
e
d
ec
o
d
in
g
p
r
o
ce
s
s
,
with
th
e
aim
o
f
r
ed
u
cin
g
in
ter
f
er
en
ce
f
r
o
m
r
ed
u
n
d
an
cy
a
n
d
ir
r
elev
a
n
t
in
f
o
r
m
atio
n
.
All
p
ar
ts
s
u
ch
as
d
o
m
ain
s
,
I
P
ad
d
r
ess
es,
an
d
ex
te
n
s
io
n
s
co
n
tain
ed
in
th
e
UR
L
ar
e
r
ep
lace
d
t
o
en
s
u
r
e
a
u
n
if
o
r
m
an
d
s
tan
d
ar
d
i
ze
d
f
o
r
m
at.
Gen
er
aliza
tio
n
h
elp
s
to
s
tr
ip
awa
y
u
n
n
ec
ess
ar
y
d
etails
th
at
co
u
l
d
in
tr
o
d
u
ce
n
o
is
e
in
to
th
e
an
aly
s
is
,
allo
win
g
th
e
m
o
d
el
to
f
o
cu
s
o
n
th
e
c
o
r
e
s
tr
u
ctu
r
e
an
d
b
eh
av
i
o
r
o
f
th
e
UR
L
o
r
s
cr
ip
t.
T
h
is
s
tep
i
s
p
a
r
ticu
lar
ly
im
p
o
r
tan
t
in
d
etec
ti
n
g
p
atter
n
s
co
m
m
o
n
in
m
alicio
u
s
UR
L
s
,
as
it
r
e
d
u
ce
s
v
ar
iab
ilit
y
an
d
h
ig
h
lig
h
ts
s
u
s
p
icio
u
s
p
atter
n
s
th
at
m
ig
h
t
o
th
er
wis
e
b
e
o
v
er
lo
o
k
ed
.
B
y
d
o
in
g
s
o
,
g
en
er
aliza
tio
n
en
h
a
n
ce
s
th
e
m
o
d
el's
ab
ilit
y
to
d
etec
t
XSS
attac
k
s
m
o
r
e
ac
c
u
r
ately
d
u
r
in
g
s
u
b
s
eq
u
e
n
t
s
tag
es
o
f
p
r
ep
r
o
ce
s
s
in
g
a
n
d
f
ea
tu
r
e
e
x
tr
a
ctio
n
.
T
h
e
r
esu
lts
o
f
g
en
er
aliza
tio
n
o
n
o
n
e
o
f
th
e
d
ata
th
at
h
as
g
o
n
e
th
r
o
u
g
h
th
e
d
ec
o
d
in
g
p
r
o
ce
s
s
ca
n
b
e
s
ee
n
in
T
ab
le
3
.
Do
m
ain
is
o
n
e
o
f
th
e
p
ar
ts
th
at
n
ee
d
s
to
b
e
g
en
er
alize
d
f
r
o
m
t
h
e
d
at
a
f
r
o
m
th
e
r
esu
lts
o
f
th
e
d
ec
o
d
in
g
p
r
o
ce
s
s
.
An
d
af
ter
th
e
g
e
n
er
aliza
tio
n
p
r
o
ce
s
s
is
ca
r
r
ied
o
u
t,
"
h
ttp
://www.
g
o
o
g
le.
co
m
"
is
ch
a
n
g
ed
t
o
"d
o
m
ain
".
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
E
n
h
a
n
ci
n
g
cro
s
s
-
s
ite
s
crip
tin
g
a
tta
ck
d
etec
tio
n
b
y
u
s
in
g
F
a
s
tText
a
s
w
o
r
d
…
(
Mu
h
a
mma
d
A
lkh
a
ir
i
)
4927
T
ab
le
3
.
T
h
e
co
m
p
ar
is
o
n
o
f
g
e
n
er
aliza
tio
n
r
esu
lts
O
r
i
g
i
n
a
l
t
e
x
t
D
e
c
o
d
e
d
t
e
x
t
<
A
H
R
EF
=
"
h
t
t
p
:
/
/
w
w
w
.
g
o
o
g
l
e
.
c
o
m"
>
<
/
A
>
<
A
H
R
EF
=
"
d
o
ma
i
n
"
>
<
/
A
>
2
.
2
.
3
.
T
o
k
eniza
t
io
n
T
h
is
s
tag
e
h
elp
s
s
tan
d
ar
d
ize
t
ex
t
b
y
r
ep
lacin
g
m
u
ltip
le
f
o
r
m
s
o
f
th
e
s
am
e
tag
with
co
n
s
i
s
ten
t
to
k
en
s
.
T
o
k
en
izatio
n
r
ed
u
ce
s
th
e
co
m
p
lex
ity
o
f
tex
t
b
y
co
n
v
er
tin
g
it
in
to
a
f
o
r
m
th
at
is
ea
s
ier
to
an
aly
ze
an
d
p
r
o
ce
s
s
.
I
t
en
s
u
r
es
th
at
s
im
ilar
p
atter
n
s
,
s
u
ch
as
v
ar
iatio
n
s
in
s
y
n
tax
o
r
ca
s
e,
ar
e
tr
ea
ted
u
n
if
o
r
m
ly
,
en
ab
lin
g
th
e
m
o
d
el
to
f
o
cu
s
o
n
co
n
ten
t
r
ath
er
th
an
f
o
r
m
attin
g
d
if
f
er
e
n
ce
s
.
Ad
d
itio
n
ally
,
to
k
e
n
izatio
n
p
lay
s
a
cr
u
cial
r
o
le
in
h
an
d
lin
g
c
o
m
p
lex
d
ata
lik
e
U
R
L
s
o
r
s
cr
ip
ts
,
b
r
ea
k
in
g
th
em
d
o
wn
in
to
m
an
a
g
ea
b
le
co
m
p
o
n
e
n
ts
.
Af
ter
th
is
s
tag
e,
th
e
to
k
e
n
ized
d
ata
will
b
e
u
s
ed
f
o
r
th
e
wo
r
d
e
m
b
ed
d
in
g
p
r
o
ce
s
s
u
s
in
g
Fas
tTe
x
t,
wh
ich
ca
p
tu
r
es
co
n
tex
tu
al
r
elatio
n
s
h
ip
s
b
etwe
en
to
k
en
s
f
o
r
d
o
wn
s
tr
ea
m
task
s
s
u
ch
as c
lass
if
icatio
n
o
r
p
r
ed
ictio
n
.
T
ab
le
4
ex
p
lain
s
th
e
c
o
m
p
ar
is
o
n
o
f
d
ata
b
e
f
o
r
e
an
d
af
ter
th
e
to
k
en
izatio
n
p
r
o
ce
s
s
.
T
h
e
g
en
er
alize
d
tex
t
s
h
o
ws
th
at
th
e
o
p
en
in
g
an
d
clo
s
in
g
tag
s
ar
e
s
till
in
o
n
e
u
n
it,
an
d
af
te
r
th
e
to
k
en
izatio
n
p
r
o
ce
s
s
,
th
e
o
p
en
in
g
an
d
cl
o
s
in
g
tag
s
ar
e
s
ep
ar
ated
.
At
th
is
s
tag
e,
th
e
attr
ib
u
tes
o
n
t
h
e
tag
ar
e
m
ain
tain
ed
b
ec
au
s
e
attr
ib
u
tes ca
n
co
n
tain
XSS attac
k
s
,
s
u
ch
as a
ler
t,
J
av
ascr
ip
t
,
an
d
ev
al
.
T
ab
le
4
.
T
h
e
co
m
p
ar
is
o
n
o
f
to
k
en
izatio
n
r
esu
lts
O
r
i
g
i
n
a
l
t
e
x
t
D
e
c
o
d
e
d
t
e
x
t
<
A
H
R
EF
=
"
d
o
ma
i
n
"
>
<
/
A
>
<
A
H
R
EF="
d
o
m
a
i
n
"
>
<
/
A
>
2
.
3
.
Wo
rd
e
m
bedd
ing
W
o
r
d
em
b
ed
d
in
g
is
a
s
et
o
f
NL
P
m
o
d
els
th
at
r
ep
r
esen
t
wo
r
d
s
in
th
e
f
o
r
m
o
f
lo
w
-
d
i
m
en
s
io
n
al
v
ec
to
r
s
ca
lled
em
b
ed
d
in
g
s
,
with
th
e
aim
o
f
im
p
r
o
v
in
g
t
h
e
p
er
f
o
r
m
an
ce
o
f
m
ac
h
in
e
l
ea
r
n
in
g
n
etwo
r
k
s
in
tex
tu
al
an
aly
s
is
o
f
d
ata
s
et
s
.
T
h
is
tech
n
iq
u
e
tr
an
s
f
o
r
m
s
tex
t d
ata
in
to
n
u
m
er
ic
r
ep
r
esen
tati
o
n
s
th
at
ca
n
b
e
u
s
ed
b
y
m
ac
h
in
e
lear
n
in
g
m
o
d
els.
I
n
Fas
tTe
x
t,
ea
ch
wo
r
d
is
r
ep
r
esen
ted
as
th
e
av
er
ag
e
o
f
th
e
n
-
g
r
am
ch
ar
ac
ter
v
ec
to
r
r
e
p
r
esen
tatio
n
s
a
n
d
t
h
e
wo
r
d
its
elf
.
Fo
r
e
x
a
m
p
le,
i
n
th
e
w
o
r
d
“e
q
u
al”
an
d
n
=
3
,
th
e
wo
r
d
will
b
e
r
ep
r
esen
ted
b
y
th
e
n
-
g
r
am
ch
ar
ac
ter
s
:
<e
q
,
eq
u
,
q
u
a,
u
al,
a
l>
an
d
<e
q
u
al
>.
So
,
th
e
wo
r
d
em
b
ed
d
in
g
f
o
r
t
h
e
wo
r
d
‘
e
q
u
al’
ca
n
b
e
g
i
v
en
as
th
e
s
u
m
o
f
all
th
e
v
ec
to
r
r
ep
r
esen
tatio
n
s
o
f
all
th
e
n
-
g
r
am
ch
ar
ac
ter
s
an
d
th
e
wo
r
d
its
elf
.
Fas
tTe
x
t
ca
n
al
s
o
ap
p
ly
h
y
p
er
p
a
r
am
eter
t
u
n
in
g
.
I
n
th
is
s
tu
d
y
,
th
e
h
y
p
er
p
ar
am
eter
v
alu
es
d
escr
ib
ed
in
T
a
b
le
5
.
T
ab
le
5
.
Hy
p
er
p
ar
a
m
eter
co
n
f
ig
u
r
atio
n
s
f
o
r
wo
r
d
em
b
e
d
d
in
g
P
a
r
a
me
t
e
r
V
a
l
u
e
M
o
d
e
l
C
B
O
W
,
s
k
i
p
-
g
r
a
m
D
i
me
n
si
o
n
1
0
0
,
1
5
0
Le
a
r
n
i
n
g
r
a
t
e
0
.
0
5
,
0
.
1
Ep
o
c
h
5
0
,
1
0
0
2
.
4
.
L
o
ng
-
s
ho
rt
t
er
m
m
emo
ry
T
h
e
L
STM
wo
r
k
s
b
y
u
s
in
g
u
n
its
co
n
s
is
tin
g
o
f
s
ev
er
al
g
a
tes,
n
am
ely
f
o
r
g
et
g
ate,
in
p
u
t
g
ate,
ce
ll
g
ate,
an
d
o
u
tp
u
t
g
ate.
T
h
e
d
e
tectio
n
m
o
d
el
u
s
ed
in
clu
d
es
L
STM
l
ay
er
,
d
r
o
p
o
u
t
lay
er
,
a
n
d
s
o
f
tm
ax
o
u
tp
u
t
lay
er
.
L
STM
lay
er
is
th
e
co
r
e
lay
er
o
f
th
is
m
o
d
el
d
etec
ti
o
n
,
d
r
o
p
o
u
t
lay
er
is
ch
o
s
en
t
o
r
ed
u
ce
o
v
er
f
itti
n
g
p
r
o
b
lem
s
,
an
d
s
o
f
tm
a
x
o
u
tp
u
t
lay
er
f
o
r
XSS
attac
k
p
r
ed
ictio
n
.
T
h
e
L
STM
m
o
d
el
will
u
s
e
a
h
id
d
en
lay
er
s
ize
o
f
1
2
8
,
d
r
o
p
o
u
t
0
.
2
,
an
d
th
e
f
in
al
o
u
tp
u
t
lay
e
r
is
ac
tiv
ated
u
s
in
g
s
ig
m
o
id
.
Fo
r
th
e
lo
s
s
f
u
n
ctio
n
,
it
will
u
s
e
b
in
ar
y
c
r
o
s
s
-
en
tr
o
p
y
.
Deta
ils
o
f
th
e
h
y
p
er
p
ar
am
eter
s
u
s
ed
c
an
b
e
s
ee
n
in
T
ab
le
6
.
T
ab
le
6
.
Hy
p
er
p
ar
a
m
eter
co
n
f
ig
u
r
atio
n
s
f
o
r
L
STM
P
a
r
a
me
t
e
r
V
a
l
u
e
Ep
o
c
h
s
5
,
7
,
9
,
1
0
B
a
t
c
h
si
z
e
3
2
,
6
4
V
a
l
i
d
a
t
i
o
n
s
p
l
i
t
0
.
2
H
i
d
d
e
n
l
a
y
e
r
s
i
z
e
1
2
8
D
r
o
p
o
u
t
r
a
t
e
0
.
2
D
e
n
se
u
n
i
t
32
Le
a
r
n
i
n
g
r
a
t
e
0
.
0
0
1
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.
14
,
No
.
6
,
Dec
em
b
er
20
25
:
4
9
2
3
-
4
9
3
2
4928
2
.
5
.
E
v
a
lua
t
i
o
n
I
n
t
h
i
s
s
t
u
d
y
,
t
h
e
e
v
al
u
a
t
i
o
n
o
f
t
h
e
XS
S
a
tt
a
c
k
d
et
e
c
t
i
o
n
m
o
d
e
l
u
s
i
n
g
d
e
e
p
le
a
r
n
i
n
g
is
b
as
ed
o
n
t
h
r
e
e
k
e
y
p
e
r
f
o
r
m
a
n
c
e
m
e
t
r
i
c
s
:
a
c
cu
r
a
c
y
,
r
e
c
a
l
l
,
a
n
d
F
1
s
c
o
r
e
.
T
h
e
s
e
m
e
t
r
i
cs
a
r
e
c
h
o
s
e
n
t
o
c
o
m
p
r
e
h
e
n
s
i
v
e
l
y
a
s
s
es
s
t
h
e
m
o
d
e
l
’
s
a
b
i
li
t
y
t
o
c
o
r
r
e
c
tl
y
i
d
e
n
t
i
f
y
m
a
l
i
ci
o
u
s
i
n
p
u
t
s
w
h
i
le
m
i
n
i
m
iz
i
n
g
m
is
c
l
ass
i
f
i
ca
t
i
o
n
.
T
h
e
e
v
a
l
u
at
i
o
n
is
s
u
p
p
o
r
t
e
d
b
y
a
c
o
n
f
u
s
i
o
n
m
at
r
ix
t
h
a
t
c
o
m
p
r
i
s
es
f
o
u
r
cl
as
s
es
:
tr
u
e
p
o
s
i
ti
v
e
cl
a
s
s
r
e
p
r
es
e
n
ti
n
g
t
h
e
n
u
m
b
e
r
o
f
XS
S
s
a
m
p
l
es
c
o
r
r
e
c
tl
y
i
d
e
n
ti
f
i
e
d
as
XS
S
,
f
a
l
s
e
p
o
s
it
i
v
e
cl
a
s
s
i
n
d
i
c
a
t
i
n
g
t
h
e
n
u
m
b
e
r
o
f
non
-
X
S
S
s
a
m
p
l
es
t
h
at
a
r
e
m
i
s
t
a
k
e
n
l
y
c
l
as
s
i
f
i
e
d
as
X
SS
,
t
r
u
e
n
e
g
a
t
i
v
e
c
la
s
s
w
h
i
c
h
i
n
c
l
u
d
e
s
t
h
e
n
u
m
b
e
r
o
f
non
-
X
SS
s
a
m
p
le
s
c
o
r
r
e
ct
l
y
c
l
a
s
s
i
f
ie
d
,
a
n
d
F
a
l
s
e
N
e
g
a
ti
v
e
c
l
a
s
s
r
e
p
r
e
s
e
n
ti
n
g
X
S
S
s
a
m
p
l
es
t
h
a
t
a
r
e
i
n
c
o
r
r
e
c
t
l
y
c
l
as
s
i
f
i
e
d
a
s
non
-
X
S
S
.
A
h
i
g
h
t
r
u
e
p
o
s
i
t
i
v
e
r
at
e
i
n
d
i
c
at
es
s
t
r
o
n
g
d
e
t
e
c
t
i
o
n
c
a
p
a
b
i
l
it
y
,
w
h
i
le
a
l
o
w
f
a
ls
e
n
e
g
a
t
i
v
e
r
at
e
is
e
s
s
e
n
t
i
a
l
t
o
a
v
o
i
d
m
i
s
s
i
n
g
a
c
t
u
a
l
a
tt
a
c
k
s
.
At
t
h
e
s
a
m
e
t
i
m
e
,
r
e
d
u
c
i
n
g
F
a
ls
e
P
o
s
i
ti
v
e
s
is
i
m
p
o
r
t
a
n
t
t
o
p
r
e
v
e
n
t
d
i
s
r
u
p
t
i
n
g
n
o
r
m
a
l
u
s
e
r
i
n
p
u
t
s
.
B
y
a
n
a
l
y
z
i
n
g
t
h
es
e
m
e
tr
i
c
s
,
we
c
a
n
d
e
t
e
r
m
i
n
e
h
o
w
w
el
l
t
h
e
m
o
d
e
l
c
a
n
cl
a
s
s
i
f
y
t
h
e
XS
S
at
t
a
c
k
s
.
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
3
.
1
.
Da
t
a
s
et
s
et
t
ing
T
h
e
d
ataset
is
d
i
v
id
ed
in
to
t
h
r
ee
p
ar
ts
,
co
n
s
is
tin
g
o
f
tr
ain
in
g
d
ata,
test
d
ata,
an
d
v
alid
atio
n
d
ata.
T
h
e
s
ep
ar
atio
n
in
to
th
ese
th
r
e
e
p
ar
ts
is
d
o
n
e
u
s
in
g
a
6
0
:2
0
:
2
0
p
o
r
tio
n
,
wh
er
e
6
0
%
o
f
th
e
d
ata
is
allo
ca
ted
f
o
r
tr
ain
in
g
,
2
0
%
f
o
r
test
in
g
,
an
d
2
0
%
f
o
r
v
alid
atio
n
.
T
h
is
d
iv
is
io
n
is
es
s
en
tial
f
o
r
en
s
u
r
in
g
th
at
th
e
m
o
d
el
is
p
r
o
p
er
l
y
tr
ain
ed
,
e
v
alu
ated
,
a
n
d
f
in
e
-
tu
n
ed
.
T
h
e
tr
ain
in
g
d
a
ta
is
u
s
ed
to
teac
h
th
e
m
o
d
el,
allo
win
g
it
to
lear
n
p
atter
n
s
an
d
r
elatio
n
s
h
ip
s
wi
th
in
th
e
d
ata.
T
h
e
test
d
ata
p
r
o
v
id
es
an
in
itial
ass
ess
m
en
t
o
f
th
e
m
o
d
el’
s
p
er
f
o
r
m
an
ce
,
h
elp
in
g
to
id
en
t
if
y
an
y
o
v
er
f
itti
n
g
o
r
u
n
d
er
f
it
tin
g
is
s
u
es.
Fin
ally
,
th
e
v
alid
atio
n
d
ata
s
er
v
es
a
s
an
u
n
b
iased
d
ataset
to
f
i
n
e
-
tu
n
e
th
e
m
o
d
el
’
s
p
ar
am
ete
r
s
,
en
s
u
r
in
g
th
at
th
e
m
o
d
el
g
en
e
r
alize
s
well
to
u
n
s
ee
n
d
ata.
B
y
m
ain
tain
in
g
th
is
s
ep
ar
atio
n
,
th
e
m
o
d
el’
s
ac
cu
r
ac
y
a
n
d
r
o
b
u
s
tn
ess
ca
n
b
e
ev
alu
ate
d
m
o
r
e
ef
f
ec
tiv
ely
,
lead
in
g
to
m
o
r
e
r
eliab
le
a
n
d
tr
u
s
two
r
th
y
p
r
e
d
ictio
n
s
in
r
ea
l
-
wo
r
ld
ap
p
licatio
n
s
.
3.
2
.
H
y
perpa
ra
m
e
t
er
T
h
e
r
esu
lts
o
f
t
h
e
ex
p
er
im
en
t
s
h
ig
h
lig
h
t
th
e
im
p
o
r
tan
ce
o
f
ca
r
ef
u
lly
s
elec
tin
g
h
y
p
e
r
p
ar
a
m
eter
s
f
o
r
b
o
th
Fas
tTe
x
t
an
d
L
STM
m
o
d
els
in
th
e
co
n
tex
t
o
f
XSS
attac
k
d
etec
tio
n
.
Am
o
n
g
th
e
1
2
8
h
y
p
er
p
a
r
am
eter
co
m
b
in
atio
n
s
test
ed
,
th
e
co
n
f
ig
u
r
atio
n
with
Fas
tTe
x
t
u
s
in
g
t
h
e
C
B
OW
m
o
d
el,
a
d
im
en
s
io
n
o
f
1
5
0
,
a
lea
r
n
in
g
r
ate
o
f
0
.
0
5
,
a
n
d
1
0
0
ep
o
ch
s
,
p
air
ed
with
an
L
STM
s
etu
p
o
f
6
4
b
atch
s
ize
an
d
7
ep
o
ch
s
,
y
ield
ed
th
e
h
ig
h
est
ac
cu
r
ac
y
o
f
9
9
.
8
9
%.
T
h
is
c
o
m
b
in
atio
n
co
n
s
is
ten
tly
o
u
tp
er
f
o
r
m
e
d
o
t
h
er
c
o
n
f
ig
u
r
atio
n
s
,
d
em
o
n
s
tr
atin
g
a
b
alan
ce
b
etwe
en
th
e
r
ich
n
ess
o
f
th
e
wo
r
d
em
b
ed
d
in
g
s
g
en
er
ated
b
y
Fas
tTe
x
t
an
d
th
e
s
eq
u
en
tial
lear
n
in
g
ca
p
ab
ilit
ies
o
f
L
STM
.
T
h
e
h
ig
h
ac
cu
r
ac
y
,
co
u
p
le
d
with
s
tr
o
n
g
p
r
ec
is
io
n
,
r
ec
all,
an
d
F1
s
co
r
es,
r
ef
lects
th
e
r
o
b
u
s
tn
ess
o
f
th
e
m
o
d
el
i
n
class
if
y
in
g
XSS
attac
k
s
.
T
h
e
r
esu
lts
also
em
p
h
asize
th
at
in
cr
ea
s
in
g
th
e
d
im
en
s
io
n
ality
o
f
em
b
ed
d
i
n
g
s
an
d
ad
ju
s
tin
g
lear
n
in
g
r
ate
s
ca
n
s
ig
n
if
ican
tly
im
p
ac
t
m
o
d
el
p
er
f
o
r
m
an
ce
,
p
ar
ticu
lar
ly
in
ca
s
es
wh
er
e
co
m
p
lex
p
atter
n
s
s
u
ch
as
m
alici
o
u
s
s
cr
ip
ts
n
ee
d
to
b
e
id
en
tifie
d
.
Ad
d
itio
n
ally
,
th
e
v
alid
atio
n
an
d
test
r
esu
lts
co
n
f
ir
m
ed
t
h
at
th
e
m
o
d
el
g
e
n
er
alize
s
well
to
u
n
s
ee
n
d
ata,
m
ak
in
g
it
a
r
eliab
le
ap
p
r
o
ac
h
f
o
r
p
r
ac
tical
XSS
d
etec
tio
n
s
ce
n
ar
io
s
.
T
h
is
s
ec
tio
n
also
d
is
cu
s
s
es
th
e
r
o
le
o
f
p
r
ep
r
o
ce
s
s
in
g
s
tep
s
,
s
u
ch
as
to
k
en
izatio
n
an
d
g
en
er
aliza
tio
n
,
wh
ich
co
n
tr
ib
u
te
d
to
en
h
a
n
cin
g
m
o
d
el
p
er
f
o
r
m
an
ce
b
y
r
e
d
u
cin
g
n
o
is
e
an
d
p
r
eser
v
in
g
k
ey
f
ea
tu
r
es in
th
e
d
ata.
3.
3
.
E
v
a
lua
t
i
o
n
T
h
e
ev
alu
atio
n
r
esu
lts
ca
n
b
e
s
ee
n
in
th
e
Fig
u
r
e
1
,
b
ased
o
n
th
e
p
r
ed
ictio
n
r
esu
lts
,
th
e
Fas
tTe
x
t
an
d
L
STM
m
o
d
els
ac
h
iev
ed
an
a
cc
u
r
ac
y
o
f
9
9
.
8
9
%.
T
h
is
ac
cu
r
ac
y
in
d
icate
s
th
at
th
e
m
o
d
el
is
ab
le
to
co
r
r
ec
tly
class
if
y
ap
p
r
o
x
im
ately
9
9
%
o
f
th
e
to
tal
test
in
g
d
ata,
d
em
o
n
s
tr
atin
g
its
h
ig
h
r
eliab
ilit
y
in
d
etec
tin
g
XSS
attac
k
s
.
Fu
r
th
er
m
o
r
e,
th
e
m
o
d
el
ac
h
iev
ed
a
p
er
f
ec
t
p
r
ec
is
io
n
s
co
r
e
o
f
1
0
0
%,
m
ea
n
i
n
g
t
h
at
all
th
e
s
am
p
les
p
r
ed
icted
as
XSS
wer
e
in
d
ee
d
tr
u
e
XSS
s
am
p
les,
with
n
o
f
alse
p
o
s
itiv
es
.
T
h
e
r
ec
all
s
co
r
e
r
ea
ch
e
d
9
9
.
7
9
%,
in
d
icatin
g
th
at
n
ea
r
ly
all
ac
tu
al
XSS
s
am
p
les
wer
e
s
u
cc
es
s
f
u
lly
id
en
tifie
d
,
with
o
n
l
y
a
m
in
im
al
n
u
m
b
er
o
f
f
alse
n
eg
ativ
es
.
T
h
e
F1
s
co
r
e,
wh
ich
r
ep
r
esen
ts
th
e
h
ar
m
o
n
i
c
m
ea
n
o
f
p
r
ec
is
io
n
a
n
d
r
ec
all,
s
to
o
d
at
9
9
.
8
9
%,
r
ef
lectin
g
a
n
o
p
tim
al
b
alan
ce
b
etwe
en
th
e
m
o
d
el'
s
ab
ilit
y
to
d
etec
t
XSS
attac
k
s
an
d
m
in
im
ize
m
is
class
if
icatio
n
.
T
h
ese
r
esu
lt
s
h
ig
h
lig
h
t
th
e
ef
f
ec
tiv
e
n
ess
o
f
co
m
b
i
n
in
g
Fas
tTe
x
t
wo
r
d
e
m
b
ed
d
in
g
s
with
an
L
STM
in
ca
p
tu
r
in
g
th
e
p
atter
n
s
o
f
XSS p
ay
lo
ad
s
.
Fro
m
th
e
co
n
f
u
s
io
n
m
atr
ix
r
e
s
u
lts
in
Fig
u
r
e
2
,
it
ca
n
b
e
s
ee
n
th
at
th
er
e
ar
e
1
,
2
6
0
d
ata
th
at
ar
e
n
o
t
XSS
attac
k
s
an
d
ar
e
class
if
ied
co
r
r
ec
tly
o
r
t
r
u
e
p
o
s
itiv
e.
T
h
er
e
is
n
o
d
ata
th
at
is
n
o
t
a
n
XSS
attac
k
an
d
is
class
if
ied
as
an
XSS
attac
k
o
r
f
alse
n
eg
ativ
e
.
T
h
en
t
h
er
e
a
r
e
3
XSS
attac
k
d
ata
th
at
ar
e
class
if
ied
as
n
o
t
an
XSS
attac
k
o
r
f
alse
p
o
s
itiv
e
;
an
d
1
,
4
7
5
XSS
attac
k
d
ata
th
at
ar
e
class
if
ied
as
XSS,
o
r
tr
u
e
n
e
g
ativ
e
.
T
h
ese
r
esu
lts
in
d
icate
a
h
ig
h
lev
el
o
f
ac
cu
r
ac
y
in
d
etec
tin
g
b
o
th
XSS
an
d
n
o
n
-
XSS
attac
k
s
,
with
m
in
im
al
m
is
class
if
icatio
n
.
T
h
e
lo
w
n
u
m
b
er
o
f
Fals
e
Po
s
itiv
es
an
d
a
b
s
en
ce
o
f
f
alse
n
eg
ativ
es
d
em
o
n
s
tr
ates
th
e
m
o
d
el'
s
ef
f
ec
tiv
en
ess
in
m
in
im
izin
g
cl
ass
if
icatio
n
er
r
o
r
s
,
wh
ich
is
cr
u
cial
f
o
r
r
ea
l
-
w
o
r
ld
XSS
d
etec
tio
n
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
E
n
h
a
n
ci
n
g
cro
s
s
-
s
ite
s
crip
tin
g
a
tta
ck
d
etec
tio
n
b
y
u
s
in
g
F
a
s
tText
a
s
w
o
r
d
…
(
Mu
h
a
mma
d
A
lkh
a
ir
i
)
4929
T
h
e
s
tr
o
n
g
p
er
f
o
r
m
an
ce
in
co
r
r
ec
tly
class
if
y
in
g
m
alicio
u
s
a
ttack
s
h
ig
h
lig
h
ts
th
e
r
eliab
ilit
y
o
f
th
e
m
o
d
el
f
o
r
s
ec
u
r
ity
task
s
,
en
s
u
r
in
g
th
at
tr
u
e
th
r
ea
ts
ar
e
id
en
tifie
d
with
o
u
t g
en
er
atin
g
u
n
n
ec
ess
ar
y
aler
ts
.
Fig
u
r
e
1
.
C
lass
if
icatio
n
r
esu
lt
Fig
u
r
e
2
.
C
o
n
f
u
s
io
n
m
atr
i
x
r
e
s
u
lt
Fig
u
r
e
3
s
h
o
ws
th
e
m
o
d
el
ac
cu
r
ac
y
ac
r
o
s
s
tr
ain
in
g
a
n
d
v
al
id
atio
n
.
T
h
e
tr
ai
n
in
g
ac
cu
r
ac
y
in
cr
ea
s
es
s
tead
ily
f
r
o
m
ar
o
u
n
d
0
.
9
8
5
t
o
0
.
9
9
8
,
in
d
icatin
g
th
e
m
o
d
e
l
is
lear
n
in
g
ef
f
ec
tiv
ely
f
r
o
m
th
e
tr
ain
in
g
d
ata.
T
h
e
v
alid
atio
n
ac
cu
r
ac
y
th
en
s
h
o
ws
a
s
im
i
lar
u
p
war
d
tr
en
d
b
u
t
with
s
o
m
e
f
lu
ctu
atio
n
s
.
I
t
s
tar
ts
at
ar
o
u
n
d
0
.
9
9
6
a
n
d
p
ea
k
s
ar
o
u
n
d
0
.
9
9
8
b
ef
o
r
e
d
r
o
p
p
in
g
s
lig
h
tly
.
T
h
is
in
d
icate
s
th
e
m
o
d
el
is
g
en
er
alizin
g
well,
an
d
lik
ely
ex
p
er
ien
ci
n
g
a
b
it o
f
o
v
er
f
itti
n
g
to
war
d
s
th
e
e
n
d
o
f
th
e
v
alid
atio
n
p
r
o
ce
s
s
.
Fig
u
r
e
4
s
h
o
ws
th
e
lo
s
s
m
o
d
el.
T
h
e
s
ig
n
if
ican
tly
d
ec
r
ea
s
i
n
g
tr
ain
in
g
lin
e
i
n
d
icate
s
th
at
th
e
m
o
d
el
m
ak
es
f
ewe
r
er
r
o
r
s
o
n
th
e
tr
ai
n
in
g
d
ata
as
th
e
n
u
m
b
er
o
f
ep
o
ch
s
in
cr
ea
s
es,
s
u
g
g
esti
n
g
ef
f
ec
tiv
e
lear
n
in
g
an
d
co
n
v
er
g
en
ce
d
u
r
in
g
tr
ain
in
g
.
Similar
ly
,
th
e
v
alid
atio
n
lo
s
s
lin
e
also
s
h
o
ws
a
d
ec
r
ea
s
in
g
tr
en
d
,
wh
ich
r
ef
lects
th
at
th
e
m
o
d
el
is
g
e
n
er
alizin
g
wel
l
.
Ho
wev
er
,
th
er
e
ar
e
s
till
s
o
m
e
f
lu
ctu
atio
n
s
in
t
h
e
v
al
id
atio
n
lin
e,
wh
ic
h
m
ay
r
ef
lect
s
en
s
itiv
ity
to
s
p
ec
if
ic
v
alid
atio
n
s
am
p
les
o
r
s
lig
h
t
o
v
er
f
itti
n
g
d
u
r
in
g
ce
r
tain
ep
o
ch
s
.
Desp
ite
th
is
,
th
e
o
v
e
r
all
d
o
wn
war
d
tr
en
d
i
n
b
o
th
tr
ain
in
g
an
d
v
alid
atio
n
lo
s
s
s
u
g
g
ests
th
at
t
h
e
m
o
d
el
r
em
ain
s
s
tab
le
an
d
p
er
f
o
r
m
s
well.
T
h
is
b
eh
av
io
r
f
u
r
th
er
s
u
p
p
o
r
ts
th
e
r
o
b
u
s
tn
ess
o
f
th
e
Fas
tTe
x
t
an
d
L
STM
ar
ch
itectu
r
e
u
s
ed
in
th
e
ex
p
er
im
e
n
t.
Fig
u
r
e
3
.
Mo
d
el
ac
cu
r
ac
y
Fig
u
r
e
4
.
Mo
d
el
lo
s
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.
14
,
No
.
6
,
Dec
em
b
er
20
25
:
4
9
2
3
-
4
9
3
2
4930
3.
4
.
Co
m
pa
riso
n wit
h o
t
her
m
et
ho
ds
R
ef
er
r
in
g
to
p
r
e
v
io
u
s
r
esear
ch
[
2
1
]
th
at
also
co
n
d
u
cte
d
r
ese
ar
ch
o
n
m
ac
h
in
e
lear
n
in
g
,
d
ee
p
lear
n
in
g
an
d
wo
r
d
em
b
ed
d
i
n
g
u
s
in
g
W
o
r
d
2
Vec
an
d
USE
,
an
d
also
u
s
in
g
th
e
s
am
e
d
ataset.
T
h
e
ac
cu
r
ac
y
o
b
tain
ed
in
th
is
s
tu
d
y
is
q
u
ite
f
ar
in
cr
e
ased
wh
en
co
m
p
a
r
ed
to
d
et
ec
tio
n
u
s
in
g
o
th
e
r
m
eth
o
d
s
in
p
r
ev
io
u
s
s
tu
d
ies.
Deta
iled
r
esu
lts
ca
n
b
e
s
ee
n
in
T
ab
le
7.
T
ab
le
7
.
T
h
e
co
m
p
ar
is
o
n
with
o
th
er
m
o
d
els
M
e
t
o
d
e
W
o
r
d
e
m
b
e
d
d
i
n
g
A
c
c
u
r
a
c
y
(
%)
F
1
s
c
o
r
e
R
e
c
a
l
l
P
r
e
c
i
s
i
o
n
V
a
n
i
l
l
a
N
N
1
W
o
r
d
2
V
e
c
9
6
.
6
4
0
.
9
6
8
3
0
.
9
8
8
0
0
.
9
4
9
3
V
a
n
i
l
l
a
N
N
2
W
o
r
d
2
V
e
c
9
7
.
2
6
0
.
9
7
5
1
0
.
9
9
5
5
8
0
.
9
9
5
3
G
R
U
W
o
r
d
2
V
e
c
9
3
.
6
1
0
.
9
4
3
7
0
.
8
9
9
9
0
.
9
9
1
9
C
N
N
W
o
r
d
2
V
e
c
9
7
.
6
6
0
.
9
7
8
8
0
.
9
6
0
3
0
.
9
9
8
0
LSTM
W
o
r
d
2
V
e
c
9
3
.
9
4
0
.
9
4
3
5
0
.
9
4
9
9
0
.
9
3
7
1
V
a
n
i
l
l
a
N
N
1
W
o
r
d
2
V
e
c
a
n
d
U
S
E
9
9
.
1
2
0
.
9
9
1
9
0
.
9
9
3
9
0
.
9
8
9
9
V
a
n
i
l
l
a
N
N
2
W
o
r
d
2
V
e
c
a
n
d
U
S
E
9
9
.
1
6
0
.
9
9
2
2
0
.
9
8
9
9
0
.
9
9
4
6
G
R
U
W
o
r
d
2
V
e
c
a
n
d
U
S
E
9
4
.
4
9
0
.
9
4
9
7
0
.
9
8
3
8
0
.
9
6
8
4
C
N
N
W
o
r
d
2
V
e
c
a
n
d
U
S
E
9
9
.
0
5
0
.
9
9
1
2
0
.
9
3
5
1
0
.
9
8
7
8
LSTM
W
o
r
d
2
V
e
c
a
n
d
U
S
E
9
8
.
4
7
0
.
9
8
5
8
0
.
9
9
4
6
0
.
9
8
7
8
LSTM
(
p
r
o
p
o
se
d
)
F
a
st
T
e
x
t
(
p
r
o
p
o
s
e
d
)
9
9
.
8
9
0
.
9
9
8
9
0
.
9
9
7
9
1
L
STM
co
m
b
in
ed
with
W
o
r
d
2
Vec
ac
h
iev
ed
an
ac
cu
r
ac
y
o
f
9
3
.
9
4
%,
wh
ile
th
e
co
m
b
i
n
atio
n
o
f
L
STM
,
W
o
r
d
2
Vec
,
an
d
t
h
e
USE
r
ea
ch
ed
9
8
.
4
7
%
ac
cu
r
a
cy
.
I
n
th
is
s
tu
d
y
,
L
STM
p
ai
r
ed
with
Fas
tTe
x
t
ac
h
iev
ed
a
n
o
tab
l
y
h
ig
h
er
ac
cu
r
ac
y
o
f
9
9
.
8
9
%,
o
u
tp
er
f
o
r
m
in
g
b
o
th
p
r
ev
io
u
s
m
eth
o
d
s
.
T
h
is
im
p
r
o
v
em
en
t
h
ig
h
lig
h
ts
th
e
s
u
p
er
i
o
r
ity
o
f
Fas
tTe
x
t
in
ca
p
tu
r
in
g
wo
r
d
r
elatio
n
s
h
ip
s
,
esp
ec
ially
f
o
r
ta
s
k
s
lik
e
XSS
at
tack
d
etec
tio
n
,
wh
er
e
u
n
d
er
s
tan
d
i
n
g
s
u
b
tle
v
a
r
iatio
n
s
in
in
p
u
t
p
atter
n
s
is
cr
u
cial.
Fas
tTe
x
t'
s
ab
ilit
y
to
m
o
d
el
s
u
b
wo
r
d
in
f
o
r
m
atio
n
g
iv
es
it
a
s
ig
n
if
ican
t
ad
v
an
tag
e
o
v
er
W
o
r
d
2
Vec
,
wh
ich
o
n
l
y
co
n
s
id
er
s
wh
o
le
wo
r
d
s
,
an
d
ev
e
n
th
e
USE
,
wh
ich
o
p
er
ates a
t a
s
en
ten
ce
lev
el
b
u
t m
i
g
h
t m
is
s
in
tr
icate
d
etails at
th
e
to
k
en
lev
el.
Ad
d
itio
n
ally
,
th
e
p
er
f
o
r
m
an
ce
g
ain
u
s
in
g
Fas
tTe
x
t
d
e
m
o
n
s
tr
ates
its
ab
il
ity
to
h
an
d
le
o
u
t
-
of
-
v
o
ca
b
u
lar
y
wo
r
d
s
m
o
r
e
ef
f
ec
tiv
ely
,
wh
ich
is
p
ar
ticu
lar
ly
i
m
p
o
r
tan
t
i
n
th
e
d
y
n
am
ic
an
d
ev
o
lv
in
g
n
at
u
r
e
o
f
web
s
ec
u
r
ity
,
wh
er
e
n
ew
f
o
r
m
s
o
f
XSS
attac
k
s
co
n
tin
u
o
u
s
ly
em
er
g
e
.
B
y
lev
er
ag
i
n
g
th
is
em
b
ed
d
i
n
g
tech
n
iq
u
e,
th
e
m
o
d
el
in
th
is
s
t
u
d
y
n
o
t
o
n
l
y
ac
h
iev
ed
h
ig
h
er
ac
cu
r
ac
y
b
u
t
also
d
em
o
n
s
tr
ate
d
b
etter
r
o
b
u
s
tn
ess
an
d
g
e
n
er
aliza
tio
n
.
T
h
is
s
u
g
g
ests
th
at
Fas
tTe
x
t,
wh
en
co
m
b
in
ed
with
L
STM
,
o
f
f
e
r
s
a
m
o
r
e
r
eliab
le
s
o
lu
tio
n
f
o
r
r
ea
l
-
tim
e
XSS
d
etec
tio
n
co
m
p
ar
ed
to
tr
ad
itio
n
al
em
b
ed
d
in
g
m
eth
o
d
s
,
p
o
s
itio
n
in
g
it
as
a
m
o
r
e
s
u
itab
le
ch
o
ice
f
o
r
f
u
tu
r
e
r
esear
ch
a
n
d
p
r
ac
tical
im
p
lem
en
tatio
n
s
in
c
y
b
er
s
ec
u
r
ity
.
4.
CO
NCLU
SI
O
N
Fro
m
th
e
r
esu
lts
o
f
t
h
e
r
esear
ch
co
n
d
u
cted
an
d
co
m
p
ar
is
o
n
s
with
s
ev
er
al
o
th
e
r
m
eth
o
d
s
an
d
wo
r
d
em
b
ed
d
in
g
s
in
class
if
y
in
g
XSS
attac
k
s
,
th
e
f
in
d
in
g
s
r
ev
e
aled
th
at
th
e
L
STM
an
d
Fas
tTe
x
t
co
m
b
in
atio
n
ac
h
iev
ed
th
e
h
ig
h
est
p
er
f
o
r
m
an
ce
in
d
etec
tin
g
attac
k
s
.
A
f
ter
co
n
d
u
ctin
g
s
ev
er
al
ex
p
er
i
m
en
ts
with
d
if
f
er
e
n
t
co
m
b
in
atio
n
s
o
f
h
y
p
er
p
ar
a
m
eter
s
,
it
was
p
r
o
v
en
th
at
th
e
L
STM
an
d
Fas
tTe
x
t
m
eth
o
d
s
s
ig
n
if
ican
tly
s
tr
en
g
th
en
ed
t
h
e
d
etec
tio
n
c
ap
ab
ilit
ies,
ac
h
iev
in
g
an
im
p
r
ess
iv
e
ac
cu
r
ac
y
o
f
9
9
.
8
9
%,
r
ec
all
o
f
9
9
.
7
9
%,
p
r
ec
is
io
n
o
f
1
0
0
%,
an
d
a
n
F1
s
co
r
e
o
f
9
9
.
8
9
%.
T
h
ese
r
esu
lts
d
em
o
n
s
tr
ate
n
o
t
o
n
ly
th
e
r
o
b
u
s
tn
ess
an
d
p
r
ec
is
io
n
o
f
th
e
m
o
d
el
b
u
t
also
its
ab
ilit
y
to
ef
f
ec
tiv
el
y
m
in
i
m
ize
f
alse
p
o
s
itiv
es
an
d
f
alse
n
eg
ativ
es
,
wh
ich
is
cr
itical
f
o
r
p
r
ac
tical
s
ec
u
r
it
y
ap
p
licatio
n
s
.
T
h
e
s
u
p
er
io
r
p
er
f
o
r
m
a
n
ce
co
m
p
ar
ed
to
o
th
er
em
b
ed
d
in
g
tech
n
iq
u
es
s
u
ch
as
W
o
r
d
2
Ve
c
an
d
USE
f
u
r
th
e
r
em
p
h
asizes
th
e
im
p
o
r
ta
n
ce
o
f
s
elec
tin
g
th
e
r
ig
h
t
m
o
d
el
an
d
h
y
p
er
p
ar
am
eter
co
n
f
ig
u
r
atio
n
s
.
T
h
is
s
tu
d
y
h
ig
h
lig
h
ts
th
at
t
h
e
co
m
b
in
atio
n
o
f
L
STM
with
Fas
tTe
x
t
ca
n
b
e
co
n
s
id
er
ed
a
h
ig
h
l
y
ef
f
ec
tiv
e
ap
p
r
o
ac
h
f
o
r
r
ea
l
-
tim
e
XSS
d
etec
tio
n
an
d
o
f
f
er
s
a
p
r
o
m
is
in
g
s
o
lu
tio
n
f
o
r
im
p
r
o
v
in
g
web
s
ec
u
r
ity
f
r
a
m
e
wo
r
k
s
.
Fu
tu
r
e
wo
r
k
co
u
ld
e
x
p
lo
r
e
s
ca
lin
g
th
is
ap
p
r
o
ac
h
t
o
h
an
d
le
m
o
r
e
d
iv
er
s
e
attac
k
p
atter
n
s
o
r
in
te
g
r
ate
it i
n
to
co
m
p
r
eh
en
s
iv
e
t
h
r
ea
t d
ete
ctio
n
s
y
s
tem
s
f
o
r
ev
en
b
r
o
ad
er
ap
p
licatio
n
s
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
T
h
is
wo
r
k
was
s
u
p
p
o
r
ted
b
y
B
I
NUS
Gr
ad
u
ate
Pro
g
r
am
–
Ma
s
ter
o
f
C
o
m
p
u
ter
Scien
ce
,
Dep
ar
tm
en
t
o
f
C
o
m
p
u
ter
Scien
ce
,
B
in
a
Nu
s
an
tar
a
Un
iv
er
s
ity
.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
T
h
is
jo
u
r
n
al
u
s
es
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT)
to
r
ec
o
g
n
ize
in
d
iv
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
llab
o
r
atio
n
.
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
E
n
h
a
n
ci
n
g
cro
s
s
-
s
ite
s
crip
tin
g
a
tta
ck
d
etec
tio
n
b
y
u
s
in
g
F
a
s
tText
a
s
w
o
r
d
…
(
Mu
h
a
mma
d
A
lkh
a
ir
i
)
4931
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
Mu
h
am
m
ad
Alk
h
air
i
Ma
s
h
u
r
i
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
Nico
Su
r
an
th
a
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
Au
th
o
r
s
s
tate
n
o
co
n
f
lict o
f
in
t
er
est.
I
NF
O
RM
E
D
CO
NS
E
N
T
T
h
is
s
tu
d
y
d
o
es
n
o
t
in
v
o
lv
e
in
d
iv
id
u
al
p
er
s
o
n
al
d
ata
o
r
id
en
tifia
b
le
h
u
m
a
n
s
u
b
jects.
T
h
er
ef
o
r
e,
in
f
o
r
m
e
d
co
n
s
en
t w
as n
o
t
r
eq
u
ir
ed
.
E
T
H
I
CAL AP
P
RO
V
AL
T
h
is
s
tu
d
y
d
id
n
o
t
i
n
v
o
lv
e
h
u
m
an
p
ar
ticip
a
n
ts
o
r
an
im
als
,
an
d
th
er
e
f
o
r
e
d
id
n
o
t
r
eq
u
i
r
e
eth
ical
ap
p
r
o
v
al.
All p
r
o
ce
d
u
r
es we
r
e
co
n
d
u
cted
in
ac
c
o
r
d
an
ce
with
r
elev
an
t in
s
titu
tio
n
al
an
d
n
ati
o
n
al
g
u
i
d
elin
es.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
d
ata
th
at
s
u
p
p
o
r
t
th
e
f
i
n
d
in
g
s
o
f
th
is
s
tu
d
y
ar
e
o
p
e
n
ly
av
ailab
le
in
Kag
g
le
R
ep
o
s
ito
r
y
at
h
ttp
s
://www.
k
ag
g
le.
co
m
/d
atasets
/s
y
ed
s
aq
lain
h
u
s
s
ain
/cr
o
s
s
-
s
ite
-
s
cr
ip
tin
g
-
x
s
s
-
d
ataset
-
f
o
r
-
d
ee
p
-
lear
n
in
g
.
RE
F
E
R
E
NC
E
S
[
1
]
A
.
A
.
S
a
l
a
m
a
,
E.
-
S
.
F
.
A
b
o
e
l
f
o
t
o
h
,
H
.
E.
K
h
a
l
i
d
,
A
.
K
.
Essa
,
H
.
M
.
E
l
-
B
a
k
r
y
,
a
n
d
D
.
S
.
E
l
-
M
o
r
sh
e
d
y
,
“
I
n
t
e
g
r
a
t
i
n
g
n
e
u
t
r
o
s
o
p
h
i
c
l
o
g
i
c
w
i
t
h
A
S
P
.
N
ET
t
o
p
r
e
v
e
n
t
X
S
S
a
t
t
a
c
k
s,”
N
e
u
t
ro
s
o
p
h
i
c
O
p
t
i
m
i
z
a
t
i
o
n
a
n
d
I
n
t
e
l
l
i
g
e
n
t
S
y
s
t
e
m
s
,
v
o
l
.
5
,
p
p
.
1
4
–
2
8
,
2
0
2
5
,
d
o
i
:
1
0
.
6
1
3
5
6
/
j
.
n
o
i
s.
2
0
2
5
.
5
4
5
5
.
[
2
]
B
.
G
o
g
o
i
,
T.
A
h
me
d
,
a
n
d
H
.
K
.
S
a
i
k
i
a
,
“
D
e
t
e
c
t
i
o
n
o
f
X
S
S
a
t
t
a
c
k
s
i
n
w
e
b
a
p
p
l
i
c
a
t
i
o
n
s:
a
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
I
n
n
o
v
a
t
i
v
e
R
e
se
a
rc
h
i
n
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
&
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
9
,
n
o
.
1
,
p
p
.
1
–
1
0
,
2
0
2
1
,
d
o
i
:
1
0
.
2
1
2
7
6
/
i
j
i
r
c
st
.
2
0
2
1
.
9
.
1
.
1
.
[
3
]
A
.
A
l
a
n
d
a
,
D
.
S
a
t
r
i
a
,
a
n
d
H
.
A
.
M
o
o
d
u
t
o
,
“
C
r
o
ss
-
si
t
e
s
c
r
i
p
t
i
n
g
(
X
S
S
)
v
u
l
n
e
r
a
b
i
l
i
t
i
e
s
i
n
mo
d
e
r
n
w
e
b
a
p
p
l
i
c
a
t
i
o
n
s,”
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
El
e
c
t
ri
c
a
l
E
n
g
i
n
e
e
ri
n
g
,
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
I
n
f
o
rm
a
t
i
c
s
(
EE
C
S
I
)
,
p
p
.
2
7
0
–
2
7
6
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
0
9
/
EEC
S
I
6
3
4
4
2
.
2
0
2
4
.
1
0
7
7
6
4
6
1
.
[
4
]
O
W
A
S
P
,
“
O
W
A
S
P
TO
P
Te
n
,
”
T
h
e
O
p
e
n
We
b
A
p
p
l
i
c
a
t
i
o
n
S
e
c
u
r
i
t
y
Pro
j
e
c
t
,
2
0
2
1
.
(
A
c
c
e
ss
e
d
:
M
a
r
.
0
1
,
2
0
2
5
)
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s:
/
/
o
w
a
s
p
.
o
r
g
/
w
w
w
-
p
r
o
j
e
c
t
-
t
o
p
-
t
e
n
/
[
5
]
G
.
H
.
Lu
u
e
t
a
l
.
,
“
X
S
S
h
i
e
l
d
:
a
n
o
v
e
l
d
a
t
a
set
a
n
d
l
i
g
h
t
w
e
i
g
h
t
h
y
b
r
i
d
d
e
e
p
l
e
a
r
n
i
n
g
mo
d
e
l
f
o
r
X
S
S
a
t
t
a
c
k
d
e
t
e
c
t
i
o
n
,
”
R
e
su
l
t
s
i
n
En
g
i
n
e
e
ri
n
g
,
v
o
l
.
2
4
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
r
i
n
e
n
g
.
2
0
2
4
.
1
0
3
3
6
3
.
[
6
]
Y
.
F
a
n
g
,
Y
.
L
i
,
L
.
L
i
u
,
a
n
d
C
.
H
u
a
n
g
,
“
D
e
e
p
X
S
S
:
c
r
o
ss
s
i
t
e
scri
p
t
i
n
g
d
e
t
e
c
t
i
o
n
b
a
s
e
d
o
n
d
e
e
p
l
e
a
r
n
i
n
g
,
”
A
C
M
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
Pr
o
c
e
e
d
i
n
g
S
e
r
i
e
s
,
p
p
.
4
7
–
5
1
,
2
0
1
8
,
d
o
i
:
1
0
.
1
1
4
5
/
3
1
9
4
4
5
2
.
3
1
9
4
4
6
9
.
[
7
]
S
e
c
u
r
e
Li
s
t
b
y
K
a
s
p
e
r
s
k
y
,
“
T
o
p
1
0
w
e
b
a
p
p
l
i
c
a
t
i
o
n
v
u
l
n
e
r
a
b
i
l
i
t
i
e
s
i
n
2
0
2
1
–
2
0
2
3
,
”
S
e
c
u
re
l
i
s
t
,
2
0
2
3
.
(
A
c
c
e
sse
d
:
M
a
r
.
1
2
,
2
0
2
4
)
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s:
/
/
s
e
c
u
r
e
l
i
st
.
c
o
m
/
t
o
p
-
10
-
w
e
b
-
a
p
p
-
v
u
l
n
e
r
a
b
i
l
i
t
i
e
s
/
1
1
2
1
4
4
/
.
[
8
]
C
W
E
,
“
2
0
2
0
C
W
E
T
o
p
2
5
,
”
C
o
m
m
o
n
W
e
a
k
n
e
ss
En
u
m
e
r
a
t
i
o
n
,
2
0
2
0
.
(
A
c
c
e
sse
d
:
M
a
r
.
1
,
2
0
2
5
)
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s
:
/
/
c
w
e
.
mi
t
r
e
.
o
r
g
/
t
o
p
2
5
/
a
r
c
h
i
v
e
/
2
0
2
0
/
2
0
2
0
_
c
w
e
_
t
o
p
2
5
.
h
t
ml
.
[
9
]
G
.
E.
R
o
d
r
í
g
u
e
z
,
J.
G
.
To
r
r
e
s
,
P
.
F
l
o
r
e
s,
a
n
d
D
.
E.
B
e
n
a
v
i
d
e
s
,
“
C
r
o
ss
-
s
i
t
e
scri
p
t
i
n
g
(
X
S
S
)
a
t
t
a
c
k
s
a
n
d
mi
t
i
g
a
t
i
o
n
:
A
su
r
v
e
y
,
”
C
o
m
p
u
t
e
r
N
e
t
w
o
r
k
s
,
v
o
l
.
1
6
6
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
o
mn
e
t
.
2
0
1
9
.
1
0
6
9
6
0
.
[
1
0
]
S
.
J.
Y
.
W
e
a
m
i
e
,
“
C
r
o
ss
-
s
i
t
e
s
c
r
i
p
t
i
n
g
a
t
t
a
c
k
s
a
n
d
d
e
f
e
n
s
i
v
e
t
e
c
h
n
i
q
u
e
s
:
A
c
o
m
p
r
e
h
e
n
si
v
e
su
r
v
e
y
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
C
o
m
m
u
n
i
c
a
t
i
o
n
s,
N
e
t
w
o
r
k
a
n
d
S
y
st
e
m
S
c
i
e
n
c
e
s
,
v
o
l
.
1
5
,
n
o
.
8
,
p
p
.
1
2
6
–
1
4
8
,
2
0
2
2
,
d
o
i
:
1
0
.
4
2
3
6
/
i
j
c
n
s.
2
0
2
2
.
1
5
8
0
1
0
.
[
1
1
]
Y
.
W
a
n
g
e
t
a
l
.
,
“
A
c
o
m
p
a
r
i
so
n
o
f
w
o
r
d
e
m
b
e
d
d
i
n
g
s
f
o
r
t
h
e
b
i
o
me
d
i
c
a
l
n
a
t
u
r
a
l
l
a
n
g
u
a
g
e
p
r
o
c
e
ss
i
n
g
,
”
J
o
u
r
n
a
l
o
f
Bi
o
m
e
d
i
c
a
l
I
n
f
o
rm
a
t
i
c
s
,
v
o
l
.
8
7
,
p
p
.
1
2
–
2
0
,
2
0
1
8
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
b
i
.
2
0
1
8
.
0
9
.
0
0
8
.
[
1
2
]
J.
I
l
i
c
,
“
C
r
o
ss
-
si
t
e
scri
p
t
i
n
g
(
X
S
S
)
m
a
k
e
s
n
e
a
r
l
y
4
0
%
o
f
a
l
l
c
y
b
e
r
a
t
t
a
c
k
s
i
n
2
0
1
9
,
”
Pre
c
i
se
S
e
c
u
ri
t
y
,
2
0
1
9
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s
:
/
/
w
w
w
.
p
r
e
c
i
s
e
se
c
u
r
i
t
y
.
c
o
m
/
a
r
t
i
c
l
e
s/
c
r
o
ss
-
s
i
t
e
-
scri
p
t
i
n
g
-
x
ss
-
ma
k
e
s
-
n
e
a
r
l
y
-
40
-
of
-
a
l
l
-
c
y
b
e
r
-
a
t
t
a
c
k
s
-
in
-
2
0
1
9
/
.
[
1
3
]
F
.
Y
o
u
n
a
s,
A
.
R
a
z
a
,
N
.
T
h
a
l
j
i
,
L
.
A
b
u
a
l
i
g
a
h
,
R
.
A
.
Zi
t
a
r
,
a
n
d
H
.
J
i
a
,
“
A
n
e
f
f
i
c
i
e
n
t
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
a
p
p
r
o
a
c
h
f
o
r
e
a
r
l
y
d
e
t
e
c
t
i
o
n
o
f
c
r
o
ss
-
si
t
e
s
c
r
i
p
t
i
n
g
a
t
t
a
c
k
s,”
D
e
c
i
s
i
o
n
An
a
l
y
t
i
c
s
J
o
u
r
n
a
l
,
v
o
l
.
1
1
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
d
a
j
o
u
r
.
2
0
2
4
.
1
0
0
4
6
6
.
[
1
4
]
A
.
M
a
t
h
e
w
,
P
.
A
m
u
d
h
a
,
a
n
d
S
.
S
i
v
a
k
u
mar
i
,
“
D
e
e
p
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s
:
a
n
o
v
e
r
v
i
e
w
,
”
A
d
v
a
n
c
e
s
i
n
I
n
t
e
l
l
i
g
e
n
t
S
y
s
t
e
m
s
a
n
d
C
o
m
p
u
t
i
n
g
,
v
o
l
.
1
1
4
1
,
p
p
.
5
9
9
–
6
0
8
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
9
8
1
-
15
-
3
3
8
3
-
9
_
5
4
.
[
1
5
]
F
.
M
.
M
.
M
o
k
b
a
l
,
W
.
D
a
n
,
W
.
X
i
a
o
x
i
,
Z.
W
e
n
b
i
n
,
a
n
d
F
.
Li
h
u
a
,
“
X
G
B
X
S
S
:
a
n
e
x
t
r
e
m
e
g
r
a
d
i
e
n
t
b
o
o
s
t
i
n
g
d
e
t
e
c
t
i
o
n
f
r
a
m
e
w
o
r
k
f
o
r
c
r
o
ss
-
s
i
t
e
scri
p
t
i
n
g
a
t
t
a
c
k
s
b
a
se
d
o
n
h
y
b
r
i
d
f
e
a
t
u
r
e
se
l
e
c
t
i
o
n
a
p
p
r
o
a
c
h
a
n
d
p
a
r
a
m
e
t
e
r
s
o
p
t
i
mi
z
a
t
i
o
n
,
”
J
o
u
r
n
a
l
o
f
I
n
f
o
rm
a
t
i
o
n
S
e
c
u
r
i
t
y
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
5
8
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
i
s
a
.
2
0
2
1
.
1
0
2
8
1
3
.
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.
14
,
No
.
6
,
Dec
em
b
er
20
25
:
4
9
2
3
-
4
9
3
2
4932
[
1
6
]
T.
M
i
k
o
l
o
v
,
K
.
C
h
e
n
,
G
.
C
o
r
r
a
d
o
,
a
n
d
J.
D
e
a
n
,
“
D
i
s
t
r
i
b
u
t
e
d
r
e
p
r
e
s
e
n
t
a
t
i
o
n
s
o
f
w
o
r
d
s
a
n
d
p
h
r
a
ses
a
n
d
t
h
e
i
r
c
o
mp
o
s
i
t
i
o
n
a
l
i
t
y
,
”
N
e
u
ra
l
i
n
f
o
rm
a
t
i
o
n
p
r
o
c
e
ssi
n
g
sy
st
e
m
s
,
v
o
l
.
1
,
p
p
.
1
–
9
,
2
0
0
6
.
[
1
7
]
S
.
S
.
B
i
r
u
n
d
a
a
n
d
R
.
K
a
n
n
i
g
a
D
e
v
i
,
“
A
r
e
v
i
e
w
o
n
w
o
r
d
e
m
b
e
d
d
i
n
g
t
e
c
h
n
i
q
u
e
s
f
o
r
t
e
x
t
c
l
a
ssi
f
i
c
a
t
i
o
n
,
”
I
n
n
o
v
a
t
i
v
e
D
a
t
a
C
o
m
m
u
n
i
c
a
t
i
o
n
T
e
c
h
n
o
l
o
g
i
e
s
a
n
d
Ap
p
l
i
c
a
t
i
o
n
,
v
o
l
.
5
9
,
p
p
.
2
6
7
–
2
8
1
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
9
8
1
-
15
-
9
6
5
1
-
3
_
2
3
.
[
1
8
]
G
.
D
i
G
e
n
n
a
r
o
,
A
.
B
u
o
n
a
n
n
o
,
a
n
d
F
.
A
.
N
.
P
a
l
m
i
e
r
i
,
“
C
o
n
s
i
d
e
r
a
t
i
o
n
s
a
b
o
u
t
l
e
a
r
n
i
n
g
W
o
r
d
2
V
e
c
,
”
J
o
u
r
n
a
l
o
f
S
u
p
e
r
c
o
m
p
u
t
i
n
g
,
v
o
l
.
7
7
,
n
o
.
1
1
,
p
p
.
1
2
3
2
0
–
1
2
3
3
5
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
s1
1
2
2
7
-
0
2
1
-
0
3
7
4
3
-
2.
[
1
9
]
J.
P
e
n
n
i
n
g
t
o
n
,
R
.
S
o
c
h
e
r
,
a
n
d
C
.
D
.
M
a
n
n
i
n
g
,
“
G
l
o
V
e
:
G
l
o
b
a
l
v
e
c
t
o
r
s
f
o
r
w
o
r
d
r
e
p
r
e
s
e
n
t
a
t
i
o
n
,
”
i
n
E
M
N
L
P
2
0
1
4
-
2
0
1
4
C
o
n
f
e
re
n
c
e
o
n
Em
p
i
r
i
c
a
l
Me
t
h
o
d
s
i
n
N
a
t
u
r
a
l
L
a
n
g
u
a
g
e
Pro
c
e
ss
i
n
g
,
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
C
o
n
f
e
re
n
c
e
,
2
0
1
4
,
p
p
.
1
5
3
2
–
1
5
4
3
,
d
o
i
:
1
0
.
3
1
1
5
/
v
1
/
d
1
4
-
1
1
6
2
.
[
2
0
]
P
.
B
o
j
a
n
o
w
s
k
i
,
E.
G
r
a
v
e
,
A
.
J
o
u
l
i
n
,
a
n
d
T
.
M
i
k
o
l
o
v
,
“
En
r
i
c
h
i
n
g
w
o
r
d
v
e
c
t
o
r
s
w
i
t
h
su
b
w
o
r
d
i
n
f
o
r
ma
t
i
o
n
,
”
T
r
a
n
s
a
c
t
i
o
n
s
o
f
t
h
e
Asso
c
i
a
t
i
o
n
f
o
r
C
o
m
p
u
t
a
t
i
o
n
a
l
L
i
n
g
u
i
st
i
c
s
,
v
o
l
.
5
,
p
p
.
1
3
5
–
1
4
6
,
2
0
1
7
,
d
o
i
:
1
0
.
1
1
6
2
/
t
a
c
l
_
a
_
0
0
0
5
1
.
[
2
1
]
R
.
B
a
k
ı
r
a
n
d
H
.
B
a
k
ı
r
,
“
S
w
i
f
t
d
e
t
e
c
t
i
o
n
o
f
X
S
S
a
t
t
a
c
k
s:
e
n
h
a
n
c
i
n
g
X
S
S
a
t
t
a
c
k
d
e
t
e
c
t
i
o
n
b
y
l
e
v
e
r
a
g
i
n
g
h
y
b
r
i
d
s
e
ma
n
t
i
c
e
mb
e
d
d
i
n
g
s
a
n
d
A
I
t
e
c
h
n
i
q
u
e
s,”
A
ra
b
i
a
n
J
o
u
rn
a
l
f
o
r
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
r
i
n
g
,
v
o
l
.
5
0
,
n
o
.
2
,
p
p
.
1
1
9
1
–
1
2
0
7
,
2
0
2
5
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
3
3
6
9
-
024
-
0
9
1
4
0
-
0.
[
2
2
]
J.
R
.
T
a
d
h
a
n
i
,
V
.
V
e
k
a
r
i
y
a
,
V
.
S
o
r
a
t
h
i
y
a
,
S
.
A
l
sh
a
t
h
r
i
,
a
n
d
W
.
El
-
S
h
a
f
a
i
,
“
S
e
c
u
r
i
n
g
w
e
b
a
p
p
l
i
c
a
t
i
o
n
s
a
g
a
i
n
s
t
X
S
S
a
n
d
S
Q
Li
a
t
t
a
c
k
s
u
si
n
g
a
n
o
v
e
l
d
e
e
p
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
,
”
S
c
i
e
n
t
i
f
i
c
R
e
p
o
rt
s
,
v
o
l
.
1
4
,
n
o
.
1
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
3
8
/
s
4
1
5
9
8
-
023
-
4
8
8
4
5
-
4.
[
2
3
]
R
.
L.
A
l
a
o
u
i
a
n
d
E.
H
.
N
f
a
o
u
i
,
“
W
e
b
a
t
t
a
c
k
s
d
e
t
e
c
t
i
o
n
u
s
i
n
g
st
a
c
k
e
d
g
e
n
e
r
a
l
i
z
a
t
i
o
n
e
n
s
e
mb
l
e
f
o
r
LST
M
s
a
n
d
w
o
r
d
e
m
b
e
d
d
i
n
g
,
”
Pro
c
e
d
i
a
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
2
1
5
,
p
p
.
6
8
7
–
6
9
6
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
r
o
c
s.
2
0
2
2
.
1
2
.
0
7
0
.
[
2
4
]
C
.
Le
n
t
e
,
R
.
H
i
r
a
t
a
Jr
.
,
a
n
d
D
.
M
.
B
a
t
i
s
t
a
,
“
A
n
i
m
p
r
o
v
e
d
t
o
o
l
f
o
r
d
e
t
e
c
t
i
o
n
o
f
X
S
S
a
t
t
a
c
k
s
b
y
c
o
m
b
i
n
i
n
g
C
N
N
w
i
t
h
LS
TM
,
”
An
a
i
s
Es
t
e
n
d
i
d
o
s
d
o
S
i
m
p
ó
si
o
Bra
s
i
l
e
i
r
o
d
e
S
e
g
u
ra
n
ç
a
d
a
I
n
f
o
rm
a
ç
ã
o
e
d
e
S
i
st
e
m
a
s
C
o
m
p
u
t
a
c
i
o
n
a
i
s,
p
p
.
1
–
8
,
2
0
2
1
,
d
o
i
:
1
0
.
5
7
5
3
/
s
b
se
g
_
e
st
e
n
d
i
d
o
.
2
0
2
1
.
1
7
3
3
3
.
[
2
5
]
W
.
Zh
a
n
g
,
Z.
L
i
n
,
a
n
d
X
.
L
i
u
,
“
S
h
o
r
t
-
t
e
r
m
o
f
f
s
h
o
r
e
w
i
n
d
p
o
w
e
r
f
o
r
e
c
a
s
t
i
n
g
-
a
h
y
b
r
i
d
m
o
d
e
l
b
a
s
e
d
o
n
d
i
scr
e
t
e
w
a
v
e
l
e
t
t
r
a
n
sf
o
r
m
(
D
W
T)
,
sea
s
o
n
a
l
a
u
t
o
r
e
g
r
e
ss
i
v
e
i
n
t
e
g
r
a
t
e
d
m
o
v
i
n
g
a
v
e
r
a
g
e
(
S
A
R
I
M
A
)
,
a
n
d
d
e
e
p
-
l
e
a
r
n
i
n
g
-
b
a
s
e
d
l
o
n
g
sh
o
r
t
-
t
e
r
m
mem
o
r
y
(
LSTM
)
,
”
R
e
n
e
w
a
b
l
e
E
n
e
r
g
y
,
v
o
l
.
1
8
5
,
p
p
.
6
1
1
–
6
2
8
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
r
e
n
e
n
e
.
2
0
2
1
.
1
2
.
1
0
0
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Mu
h
a
m
m
a
d
Alk
h
a
iri
Ma
shu
r
i
h
o
ld
s
a
Ba
c
h
e
lo
r
o
f
Ap
p
l
ied
S
c
ien
c
e
fr
o
m
P
o
li
tek
n
i
k
Ca
lt
e
x
Riau
,
In
d
o
n
e
sia
in
2
0
1
8
.
He
is
c
u
rre
n
tl
y
p
u
rsu
in
g
m
a
ste
r’s
d
e
g
re
e
a
t
Bin
a
Nu
sa
n
tara
Un
i
v
e
rsity
.
He
is
a
ls
o
wo
rk
in
g
a
s
an
As
sista
n
t
Vic
e
P
re
sid
e
n
t
(AV
P
)
-
F
u
l
l
S
tac
k
S
o
ftwa
re
En
g
in
e
e
r
a
t
OCBC
S
d
n
Bh
d
,
M
a
lay
sia
.
His
re
se
a
rc
h
in
tere
st
is
in
so
ftwa
re
e
n
g
in
e
e
rin
g
a
n
d
c
y
b
e
r
se
c
u
rit
y
.
H
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
m
.
m
a
sh
u
ri@b
in
u
s.a
c
.
id
.
Nic
o
S
u
r
a
n
th
a
re
c
e
iv
e
d
th
e
B.
En
g
.
a
n
d
M
.
En
g
.
d
e
g
re
e
s
f
ro
m
t
h
e
Ba
n
d
u
n
g
In
stit
u
te
o
f
Tec
h
n
o
lo
g
y
,
In
d
o
n
e
sia
a
n
d
th
e
P
h
.
D.
d
e
g
re
e
fro
m
th
e
Ky
u
sh
u
In
sti
tu
te
o
f
Tec
h
n
o
l
o
g
y
,
Ja
p
a
n
.
He
is
a
le
c
tu
re
r
with
th
e
De
p
a
rtme
n
t
o
f
El
e
c
tri
c
a
l,
El
e
c
tro
n
ics
,
a
n
d
Co
m
m
u
n
ica
ti
o
n
En
g
in
e
e
rin
g
,
T
o
k
y
o
Ci
ty
Un
i
v
e
rsity
,
Ja
p
a
n
.
He
is
a
lso
a
n
As
so
c
iate
P
ro
fe
ss
o
r
with
th
e
M
a
ste
r
o
f
C
o
m
p
u
ter
S
c
ien
c
e
De
p
a
rtme
n
t,
BINU
S
G
r
a
d
u
a
te
P
ro
g
ra
m
,
Bin
a
Nu
sa
n
tara
Un
iv
e
rsity
,
In
d
o
n
e
sia
.
He
wa
s
w
it
h
th
e
Au
to
n
o
m
o
u
s
Un
i
v
e
rsity
o
f
Ba
rc
e
lo
n
a
(S
p
a
in
)
a
n
d
th
e
Un
iv
e
rsity
o
f
Tec
h
n
o
l
o
g
y
S
y
d
n
e
y
(Au
stra
li
a
)
in
2
0
2
4
a
s
a
v
is
it
in
g
p
ro
fe
ss
o
r
a
n
d
v
isit
in
g
re
se
a
rc
h
e
r,
re
sp
e
c
ti
v
e
ly
.
His
re
se
a
rc
h
in
tere
sts
in
c
lu
d
e
u
b
iq
u
it
o
u
s
c
o
m
p
u
ti
n
g
,
i
n
telli
g
e
n
t
sy
ste
m
s,
th
e
i
n
tern
e
t
o
f
t
h
in
g
s,
a
n
d
d
i
g
it
a
l
h
e
a
lt
h
.
He
is
c
u
rre
n
t
ly
a
re
se
a
rc
h
e
x
p
e
rt
c
o
m
m
it
tee
m
e
m
b
e
r
a
t
th
e
IEI
CE
S
m
a
rt
In
fo
M
e
d
ia
S
y
ste
m
S
o
c
iety
a
n
d
t
h
e
De
v
e
lo
p
m
e
n
t
o
f
Dig
it
a
l
Hu
m
a
n
Re
so
u
rc
e
s
Or
g
a
n
iza
ti
o
n
i
n
Ja
p
a
n
(DD
HR)
a
ss
o
c
iatio
n
.
He
is
a
se
n
i
o
r
m
e
m
b
e
r
o
f
I
EE
E
a
n
d
a
m
e
m
b
e
r
o
f
IEI
C
E
Ja
p
a
n
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
n
ico
.
su
ra
n
th
a
@b
i
n
u
s
.
a
c
.
id
.
Evaluation Warning : The document was created with Spire.PDF for Python.