I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
p
ute
r
E
ng
in
ee
ring
(
I
J
E
CE
)
Vo
l.
10
,
No
.
5
,
Octo
b
er
2
0
2
0
,
p
p
.
5
3
3
5
~
5
3
4
6
I
SS
N:
2
0
8
8
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v
10
i
5
.
pp
5
3
3
5
-
5
3
4
6
5335
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m/in
d
ex
.
p
h
p
/I
JE
C
E
An ada
ptive a
no
m
a
ly
requ
est d
ete
ction fra
m
ew
o
rk
ba
sed
o
n dyn
a
m
ic w
eb appli
ca
tion pro
fil
es
Cho
Do
Xua
n
1
,
Na
m
Ng
uy
en
2
,
H
o
a
Ng
uy
en
Dinh
3
1,
3
In
f
o
rm
a
ti
o
n
S
e
c
u
rit
y
De
p
a
rt
m
e
n
t
P
o
sts a
n
d
T
e
lec
o
m
m
u
n
ica
ti
o
n
s
I
n
stit
u
te o
f
T
e
c
h
n
o
lo
g
y
,
V
ietn
a
m
1,
2
In
f
o
rm
a
ti
o
n
A
ss
u
ra
n
c
e
De
p
a
rt
m
e
n
t
F
P
T
Un
iv
e
rsity
,
V
ietn
a
m
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
No
v
1
5
,
2
0
1
9
R
ev
i
s
ed
Mar
2
3
,
2
0
2
0
A
cc
ep
ted
A
p
r
1
3
,
2
0
2
0
W
e
b
a
p
p
li
c
a
ti
o
n
f
ire
w
a
ll
is
a
h
ig
h
ly
e
ffe
c
ti
v
e
a
p
p
li
c
a
ti
o
n
in
p
ro
tec
ti
n
g
th
e
a
p
p
li
c
a
ti
o
n
lay
e
r
a
n
d
d
a
tab
a
se
la
y
e
r
o
f
w
e
b
sites
f
ro
m
a
tt
a
c
k
a
c
c
e
ss
.
T
h
is
p
a
p
e
r
p
ro
p
o
se
s
a
n
e
w
w
e
b
a
p
p
li
c
a
ti
o
n
f
irew
a
ll
d
e
p
lo
y
in
g
m
e
t
h
o
d
b
a
se
d
on
d
y
n
a
m
ic
w
e
b
a
p
p
li
c
a
ti
o
n
p
r
o
f
il
in
g
(D
WA
P
)
a
n
a
ly
sis
tec
h
n
iq
u
e
.
T
h
is
is
a
m
e
th
o
d
to
d
e
p
l
o
y
a
f
ire
wa
ll
b
a
se
d
o
n
a
n
a
ly
z
in
g
we
b
site
a
c
c
e
ss
d
a
ta.
DWA
P
is
i
m
p
ro
v
e
d
to
in
teg
ra
te
d
e
e
p
ly
in
to
th
e
stru
c
tu
re
o
f
th
e
w
e
b
site
to
in
c
re
a
se
th
e
c
o
m
p
a
ti
b
il
it
y
o
f
th
e
a
n
o
m
a
l
y
d
e
tec
ti
o
n
sy
ste
m
in
to
e
a
c
h
w
e
b
site,
th
e
re
b
y
i
m
p
r
o
v
i
n
g
t
h
e
a
b
i
l
i
t
y
t
o
d
e
t
e
c
t
a
b
n
o
r
m
a
l
r
e
q
u
e
s
t
s
.
T
o
i
m
p
r
o
v
e
t
h
e
c
o
m
p
a
t
i
b
i
l
i
t
y
o
f
th
e
w
e
b
a
p
p
li
c
a
ti
o
n
f
irew
a
ll
w
it
h
p
ro
tec
t
e
d
o
b
jec
ts,
th
e
p
r
o
p
o
se
d
sy
ste
m
c
o
n
sists
o
f
tw
o
p
a
rts
w
it
h
th
e
m
a
in
tas
k
s
a
re
:
i)
D
e
tec
t
a
b
n
o
rm
a
l
a
c
c
e
ss
in
w
e
b
a
p
p
li
c
a
ti
o
n
(W
A
)
a
c
c
e
s
s
;
ii
)
S
e
m
i
-
a
u
to
m
a
ti
c
u
p
d
a
te
th
e
a
tt
a
c
k
d
a
ta
to
th
e
a
b
n
o
rm
a
l
a
c
c
e
s
s
d
e
tec
ti
o
n
sy
ste
m
d
u
rin
g
WA
a
c
c
e
s
s.
T
h
is
n
e
w
m
e
th
o
d
is
a
p
p
li
c
a
b
le
in
re
a
l
-
ti
m
e
d
e
tec
ti
o
n
s
y
ste
m
s
w
h
e
r
e
u
p
d
a
ti
n
g
o
f
n
e
w
a
tt
a
c
k
d
a
ta
is
e
ss
e
n
ti
a
l
sin
c
e
w
e
b
a
tt
a
c
k
s
a
re
in
c
re
a
sin
g
l
y
c
o
m
p
le
x
a
n
d
so
p
h
ist
ica
ted
.
K
ey
w
o
r
d
s
:
An
o
m
al
y
r
eq
u
e
s
t
Data
clu
s
ter
in
g
D
y
n
a
m
ic
p
r
o
f
il
in
g
W
eb
ap
p
licatio
n
f
ir
e
w
all
Co
p
y
rig
h
t
©
2
0
2
0
In
stit
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
Al
l
rig
h
ts re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
C
h
o
Do
Xu
a
n
,
Dep
ar
t
m
en
t o
f
I
n
f
o
r
m
atio
n
Se
cu
r
it
y
,
P
o
s
ts
an
d
T
elec
o
m
m
u
n
icatio
n
s
I
n
s
ti
tu
te
o
f
T
ec
h
n
o
lo
g
y
,
1
2
2
Ho
an
g
Qu
o
c
Viet,
C
a
u
Gi
a
y
Dis
tr
ict,
Ha
n
o
i,
Viet
n
a
m
.
E
m
ail:
c
h
o
d
x
@
p
ti
t.e
d
u
.
v
n
1.
I
NT
RO
D
UCT
I
O
N
C
u
r
r
en
tl
y
,
w
eb
ap
p
licatio
n
s
e
cu
r
it
y
h
as
b
ee
n
a
h
o
t
to
p
ic
f
o
r
m
a
n
y
r
e
s
ea
r
ch
er
s
a
n
d
in
ter
n
et
s
er
v
ice
p
r
o
v
id
er
s
.
A
cc
o
r
d
in
g
to
th
e
S
y
m
a
n
tec
r
ep
o
r
t
[
1
]
,
w
eb
-
b
ased
attac
k
s
ac
co
u
n
ted
f
o
r
1
0
p
er
ce
n
t
o
f
to
ta
l
m
alicio
u
s
r
eq
u
est
s
d
etec
ted
i
n
2
0
1
8
,
an
d
w
eb
s
ite
s
ec
u
r
it
y
is
a
n
ec
es
s
ar
y
is
s
u
e
at
p
r
ese
n
t.
I
n
[
2
]
,
Mo
o
k
h
e
y
p
r
esen
ted
th
e
c
h
ar
ac
ter
is
t
ics,
co
m
p
o
s
itio
n
s
an
d
o
p
er
atio
n
p
r
in
cip
les
o
f
WA
.
B
esid
es,
s
o
m
e
o
th
er
w
o
r
k
s
in
[
2
-
4
]
h
a
v
e
s
h
o
w
n
s
e
v
er
al
v
u
l
n
er
ab
ilit
ie
s
an
d
t
h
r
ea
ts
t
h
at
attac
k
er
s
co
u
ld
ex
p
lo
it
t
o
attac
k
th
e
w
eb
ap
p
licatio
n
s
.
A
cc
o
r
d
in
g
to
t
h
e
s
u
r
v
e
y
s
i
n
[
3
]
an
d
[
4
]
,
th
e
v
u
l
n
er
ab
ilit
ie
s
o
f
t
h
e
p
r
o
to
co
l
h
y
p
er
te
x
t
tr
an
s
f
er
p
r
o
to
co
l
(
H
T
T
P
)
o
r
h
y
p
er
tex
t
tr
an
s
f
er
p
r
o
to
co
l
s
ec
u
r
e
(
H
T
T
P
S)
ar
e
o
f
ten
p
r
ef
er
ab
le
to
b
e
ex
p
lo
ited
b
y
attac
k
er
s
.
HT
T
P
an
d
H
T
T
PS
ar
e
th
e
t
w
o
m
o
s
t
p
o
p
u
lar
p
r
o
to
c
o
ls
to
co
m
m
u
n
icat
e
f
o
r
en
d
u
s
er
s
.
B
ef
o
r
e
r
etu
r
n
in
g
th
e
co
n
te
n
ts
to
d
is
p
lay
o
n
th
e
w
eb
b
r
o
w
s
er
,
w
eb
ap
p
licatio
n
s
p
r
o
ce
s
s
th
e
co
n
ten
t
o
f
th
e
u
s
er
r
eq
u
ests
.
A
cc
o
r
d
in
g
to
t
h
e
s
ta
n
d
ar
d
d
escr
ib
ed
in
[
5
]
,
th
e
s
tr
u
ctu
r
al
co
m
p
o
n
en
t
s
o
f
an
HT
T
P
o
r
H
T
T
P
S p
ac
k
et
in
cl
u
d
e
r
eq
u
est
li
n
e,
s
tatu
s
li
n
e,
h
ea
d
er
f
ield
s
,
m
es
s
ag
e
b
o
d
y
,
an
d
s
o
m
e
o
th
er
co
m
p
o
n
en
t
s
.
Her
e,
in
o
r
d
er
to
attac
k
w
eb
ap
p
licatio
n
s
,
atta
ck
er
s
w
i
ll
tr
y
to
ch
a
n
g
e
t
h
e
co
n
te
n
t
o
f
t
h
ese
co
m
p
o
n
e
n
t
s
,
t
h
er
eb
y
cr
ea
ti
n
g
a
v
u
l
n
er
ab
ilit
y
i
n
t
h
e
p
r
o
ce
s
s
o
f
p
r
o
ce
s
s
i
n
g
r
eq
u
e
s
t.
As
a
r
esu
lt,
t
h
e
w
eb
ap
p
licatio
n
s
r
e
tu
r
n
t
h
e
o
u
tp
u
ts
a
s
attac
k
er
d
esire
d
.
So
f
ar
,
th
er
e
ar
e
t
w
o
m
a
in
m
eth
o
d
s
o
f
d
etec
tin
g
w
eb
ap
p
licatio
n
attac
k
s
:
s
i
g
n
atu
r
e
-
b
ase
d
m
et
h
o
d
s
b
ased
o
n
a
s
et
o
f
p
r
ed
ef
in
ed
r
u
les,
an
d
an
o
m
al
y
-
b
a
s
ed
m
et
h
o
d
s
th
at
r
el
y
o
n
d
ata
an
al
y
zi
n
g
a
n
d
s
tatis
tics
to
f
i
n
d
ab
n
o
r
m
al
c
h
ar
ac
ter
is
tics
in
t
h
e
r
eq
u
est
s
.
B
o
th
s
i
g
n
a
tu
r
e
-
b
ased
an
d
an
o
m
al
y
-
b
a
s
ed
m
eth
o
d
s
h
a
v
e
ce
r
tai
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
10
,
No
.
5
,
Octo
b
e
r
2
0
2
0
:
5
3
3
5
-
5346
5336
ad
v
an
ta
g
es
a
n
d
d
is
ad
v
an
ta
g
es.
I
n
g
en
er
al,
s
o
l
u
tio
n
s
ap
p
l
y
in
g
s
ig
n
atu
r
e
-
b
ased
tech
n
iq
u
e
s
[
2
,
3
]
a
r
e
n
o
t
ab
le
to
d
etec
t
u
n
u
s
u
al
r
eq
u
est
s
s
i
n
ce
th
ese
m
eth
o
d
s
ar
e
m
ai
n
l
y
b
ased
o
n
f
ix
ed
r
u
li
n
g
s
y
s
te
m
s
.
An
o
m
al
y
-
b
a
s
ed
m
et
h
o
d
s
ar
e
ca
p
ab
le
o
f
d
etec
t
in
g
ab
n
o
r
m
al
r
eq
u
ests
b
ec
au
s
e
th
e
y
u
t
ilize
tech
n
iq
u
e
s
to
an
al
y
ze
an
d
ev
al
u
ate
th
e
b
eh
av
io
r
s
o
f
th
e
r
eq
u
est
s
.
T
h
er
e
h
av
e
b
ee
n
m
a
n
y
s
tu
d
ies
u
s
i
n
g
an
o
m
al
y
-
b
ased
an
al
y
s
i
s
tech
n
iq
u
e
s
to
d
etec
t
u
n
u
s
u
al
r
eq
u
est
s
.
Hab
ee
b
a
et
al.
[
6
]
lis
ted
a
n
u
m
b
er
o
f
is
s
u
es
i
n
d
etec
tin
g
ab
n
o
r
m
alities
i
n
c
y
b
er
s
p
ac
e
in
g
e
n
er
al
an
d
in
w
eb
ac
ce
s
s
in
p
ar
ticu
lar
.
E
s
p
ec
iall
y
,
t
h
e
y
m
ad
e
a
s
u
r
v
e
y
o
n
m
et
h
o
d
s
an
d
tech
n
iq
u
e
s
co
m
m
o
n
l
y
ap
p
lied
in
b
ig
d
ata
an
al
y
s
is
,
a
n
d
p
o
in
t
ed
o
u
t
m
a
n
y
b
en
e
f
its
to
an
o
m
a
l
y
-
b
ased
d
etec
tio
n
ap
p
r
o
ac
h
es
.
Ho
w
ev
er
,
an
o
r
m
a
l
y
-
b
ased
m
et
h
o
d
s
al
s
o
h
a
v
e
s
o
m
e
p
r
o
b
lem
s
p
r
esen
ted
as
f
o
ll
o
w
s
:
a.
A
ll a
n
o
r
m
al
y
r
eq
u
es
ts
’
ch
ar
ac
ter
is
tics
co
u
ld
n
o
t b
e
f
o
u
n
d
in
tr
ain
i
n
g
d
ataset
s
A
cc
o
r
d
in
g
l
y
,
d
ataset
s
f
o
r
u
n
u
s
u
al
ac
ce
s
s
d
etec
tio
n
u
s
ed
in
a
n
u
m
b
er
o
f
p
r
ev
io
u
s
s
t
u
d
ies
w
er
e
co
llected
th
r
o
u
g
h
t
h
e
test
r
es
u
lts
o
f
a
v
ailab
le
s
ec
u
r
it
y
to
o
ls
,
f
ir
e
w
all
s
,
etc.
As
a
r
es
u
lt,
th
o
s
e
d
atasets
co
n
tai
n
al
m
o
s
t
all
attac
k
r
eq
u
est
s
[
7
]
.
Mo
r
e
o
v
er
,
in
r
ea
lity
,
th
er
e
ar
e
also
m
a
n
y
ac
ce
s
s
e
s
,
w
h
ic
h
d
o
n
o
t
co
n
tain
attac
k
co
n
ten
t
s
,
h
a
v
i
n
g
th
e
s
a
m
e
ch
a
r
ac
ter
is
tics
a
n
d
s
tr
u
ct
u
r
e
as
ab
n
o
r
m
al
r
eq
u
e
s
ts
.
A
cc
o
r
d
in
g
to
th
is
s
t
u
d
y
,
n
o
r
m
al
r
eq
u
ests
ar
e
o
n
es
t
h
at
p
er
f
o
r
m
le
g
al
o
p
er
atio
n
s
a
n
d
co
n
tai
n
s
in
f
o
r
m
at
io
n
th
at
f
o
llo
w
s
t
h
e
p
r
escr
ib
ed
s
tan
d
ar
d
s
.
I
n
co
n
tr
as
t,
i
f
r
eq
u
e
s
ts
ar
e
d
i
f
f
er
en
t
f
r
o
m
t
h
e
s
p
ec
if
ied
cr
iter
ia,
t
h
e
y
w
ill
b
e
co
n
s
i
d
er
ed
as a
b
n
o
r
m
al
r
eq
u
ests
.
Ge
n
er
all
y
,
ab
n
o
r
m
al
r
eq
u
ests
ca
n
b
e
ex
p
r
ess
ed
in
t
w
o
ca
s
e
s
:
W
eb
s
ite
attac
k
r
eq
u
est
s
,
T
h
e
r
eq
u
ests
id
en
ti
f
y
a
n
d
ex
p
l
o
it
w
eb
s
ite
v
u
ln
er
ab
ilit
ies.
I
t
c
a
n
b
e
s
a
i
d
t
h
a
t
a
t
t
a
c
k
i
n
g
r
e
q
u
e
s
t
s
a
r
e
ju
s
t
o
n
e
p
a
r
t
o
f
a
b
n
o
r
m
a
l
r
e
q
u
e
s
t
s
.
F
r
o
m
t
h
e
i
n
c
o
m
p
l
e
t
e
d
e
f
i
n
i
t
i
o
n
o
f
a
b
n
o
r
m
a
l
a
c
c
e
s
s
e
s
,
i
t
m
a
y
l
e
a
d
t
o
t
h
e
l
a
c
k
o
f
o
b
je
c
t
i
v
i
t
y
w
h
e
n
b
u
i
l
d
i
n
g
p
r
o
p
e
r
t
i
e
s
f
o
r
t
h
e
m
o
d
e
l
f
o
r
a
t
t
a
c
k
r
e
q
u
e
s
t
d
e
t
e
c
t
i
o
n
f
r
o
m
p
r
e
v
i
o
u
s
s
t
u
d
i
e
s
.
S
i
n
c
e
a
n
o
m
a
l
o
u
s
a
t
t
r
i
b
u
t
e
s
a
r
e
b
u
i
l
t
m
a
i
n
l
y
o
n
t
h
e
a
t
t
a
c
k
i
n
g
r
e
q
u
e
s
t
d
a
t
a
,
t
h
o
s
e
a
t
t
r
i
b
u
t
e
s
o
n
l
y
f
o
c
u
s
o
n
r
e
p
r
e
s
e
n
t
i
n
g
t
h
e
c
h
a
r
a
c
t
e
r
i
s
t
i
c
s
o
f
t
h
e
a
t
t
a
c
k
r
e
q
u
e
s
t
s
.
b.
Featu
r
e
ex
tr
ac
tio
n
m
eth
o
d
s
co
u
ld
n
o
t p
r
esen
t t
h
e
ch
ar
ac
ter
i
s
t
ics o
f
w
eb
ap
p
licatio
n
s
P
r
e
v
i
o
u
s
s
t
u
d
i
e
s
o
n
a
b
n
o
r
m
a
l
d
e
t
e
c
t
i
o
n
s
y
s
t
em
s
m
a
i
n
l
y
u
s
e
d
w
e
b
s
i
t
e
d
a
t
a
,
b
u
t
d
i
d
n
o
t
c
h
a
r
a
c
t
e
r
i
z
e
e
a
c
h
U
n
i
f
o
r
m
R
e
s
o
u
r
c
e
I
d
e
n
t
i
f
i
e
r
(
U
R
I
)
.
T
h
e
U
R
I
i
s
t
h
e
p
a
t
h
t
h
a
t
i
d
e
n
t
i
f
i
e
s
a
w
e
b
s
i
t
e
'
s
r
e
s
o
u
r
c
e
s
.
T
h
e
r
e
s
o
u
r
c
e
s
in
ea
c
h
U
R
I
c
a
n
b
e
i
n
f
o
r
m
a
t
i
o
n
i
n
f
o
r
m
o
f
H
T
M
L
,
o
r
i
t
c
a
n
b
e
l
o
g
i
n
t
a
s
k
s
,
r
e
g
i
s
t
r
a
t
i
o
n
,
i
n
f
o
r
m
a
t
i
o
n
s
e
a
r
c
h
.
I
f
t
h
e
U
R
I
ch
ar
ac
ter
is
t
ic
s
ar
e
n
o
t c
ar
ef
u
ll
y
ex
p
lo
ited
f
o
r
attac
k
r
eq
u
e
s
t
d
etec
tio
n
,
s
o
m
e
is
s
u
es
m
a
y
e
m
er
g
e
as
f
o
llo
w
s
:
Mo
d
el
m
a
y
n
o
t
d
ig
in
to
th
e
i
n
s
i
g
h
t
f
u
l
ch
ar
ac
ter
i
s
tics
o
f
ea
ch
U
R
I
s
i
n
ce
th
e
ex
tr
ac
ted
f
e
atu
r
es
ar
e
m
ad
e
to
p
r
esen
t
t
h
e
ch
ar
ac
te
r
is
tics
o
f
all
U
R
I
s
.
Fo
r
ex
a
m
p
le,
th
e
f
ea
t
u
r
e
p
r
esen
t
in
g
t
h
e
UR
I
in
f
o
r
m
at
io
n
len
g
th
u
s
ed
in
p
r
ev
io
u
s
s
t
u
d
i
es
is
u
s
u
al
l
y
s
h
o
r
t
f
o
r
UR
I
s
h
av
i
n
g
in
f
o
r
m
a
tio
n
i
n
HT
M
L
f
o
r
m
,
w
h
ile
it
i
s
u
s
u
all
y
lo
n
g
f
o
r
UR
I
s
in
a
r
eq
u
est
f
o
r
m
f
o
llo
w
i
n
g
GE
T
m
et
h
o
d
.
T
h
is
lead
s
to
th
e
f
ac
t
th
at
w
h
e
n
i
n
itialize
t
h
is
v
ar
iab
le
w
ith
t
h
e
en
tire
UR
I
,
its
v
alu
e
b
ec
o
m
e
s
s
p
ar
e
w
it
h
a
lar
g
e
v
ar
ian
ce
,
a
n
d
m
a
y
n
o
t
p
r
ese
n
t
th
e
ab
n
o
r
m
ali
t
y
w
h
en
u
s
ed
in
th
e
m
o
d
el.
C
o
n
s
eq
u
e
n
tl
y
,
th
e
m
o
d
el
w
i
ll
ca
u
s
e
co
n
f
u
s
io
n
in
d
etec
ti
n
g
ab
n
o
r
m
al
r
eq
u
est
s
at
d
i
f
f
er
e
n
t
UR
I
s
.
Fi
g
u
r
e
1
b
elo
w
s
h
o
w
s
an
ex
a
m
p
le
o
f
th
e
co
n
te
n
ts
o
f
a
n
o
r
m
a
l
ac
ce
s
s
1
(
a)
an
d
a
ab
n
o
r
m
al
ac
ce
s
s
1
(
b
)
.
(
a)
(
b
)
Fig
u
r
e
1
.
Sa
m
p
le
co
n
te
n
ts
o
f
n
o
r
m
al
r
eq
u
e
s
t (
a)
an
d
an
ab
n
o
r
m
al
r
eq
u
est (
b
)
I
n
F
i
g
u
r
e
1
,
it
ca
n
b
e
s
ee
n
t
h
at
in
n
o
r
m
al
r
eq
u
est
s
,
all
ac
c
ep
ted
v
alu
e
s
ar
e
“tex
t
/h
t
m
l,a
p
p
licatio
n
/
x
h
t
m
l+
x
m
l,a
p
p
licatio
n
/
x
m
l;q
=0
.
9
,
*
/
*
;q
=0
.
8
,”
w
h
ile
i
n
o
p
e
r
atio
n
s
t
h
at
ar
e
n
o
t
l
ik
e
n
o
r
m
al
u
s
er
s
,
m
a
y
b
e
d
u
e
to
th
e
s
ca
n
n
er
to
o
ls
o
r
i
n
ten
t
io
n
al
c
h
an
g
e
o
f
attac
k
er
,
th
e
ac
c
ep
ted
v
alu
e
is
s
et
to
“*
/
*
”
(
ac
ce
p
ted
all)
.
T
h
is
is
an
i
m
p
o
r
tan
t
c
h
ar
ac
ter
is
tic
to
d
ete
r
m
in
e
th
e
ab
n
o
r
m
al
r
eq
u
e
s
ts
i
n
F
ig
u
r
e
1
(
b
)
.
T
h
er
e
ar
e
tw
o
t
y
p
es
o
f
m
is
s
ed
c
lass
i
f
icatio
n
:
Fals
e
Ne
g
ati
v
e
(
FN)
an
d
Fals
e
Po
s
itiv
e
(
F
P
)
.
FN
is
w
h
e
n
an
ab
n
o
r
m
a
l
r
eq
u
est
is
m
is
s
ed
class
i
f
ied
as
n
o
r
m
a
l.
T
h
is
u
s
u
all
y
h
ap
p
en
s
w
h
e
n
th
e
n
u
m
b
e
r
o
f
r
eq
u
est
s
a
m
p
les
h
av
i
n
g
v
alu
e
A
cc
ep
t
b
ein
g
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
A
n
a
d
a
p
tive
a
n
o
ma
ly
r
eq
u
est
d
etec
tio
n
fr
a
mewo
r
k
b
a
s
ed
…
(
C
h
o
Do
X
u
a
n
)
5337
“*
/
*
”
i
n
t
h
e
w
h
o
le
w
eb
s
ite
i
s
co
n
s
id
er
ab
le
g
r
ea
ter
t
h
a
n
o
t
h
e
r
t
y
p
es
o
f
r
eq
u
e
s
ts
.
I
n
t
h
i
s
s
itu
atio
n
,
t
h
e
ab
n
o
r
m
al
ch
ar
ac
ter
is
tic
o
f
th
e
ab
n
o
r
m
a
l
r
eq
u
est
w
ill
b
e
i
g
n
o
r
ed
a
n
d
t
h
at
r
eq
u
e
s
t
is
d
ec
id
ed
a
s
n
o
r
m
al.
On
t
h
e
co
n
tr
ar
y
,
FP
p
r
esen
ts
n
o
r
m
al
r
eq
u
e
s
t
m
is
s
ed
class
if
ied
as
ab
n
o
r
m
al
o
n
es.
Fo
r
ex
a
m
p
le,
in
ca
s
e
t
h
e
w
eb
s
i
te
h
as
a
f
o
r
m
f
illi
n
g
f
u
n
ct
io
n
a
n
d
w
h
e
n
t
h
e
co
n
te
n
t
o
f
t
h
e
f
o
r
m
co
n
tai
n
s
th
e
k
e
y
w
o
r
d
f
o
u
n
d
to
b
e
ab
n
o
r
m
al
b
ased
on
r
u
le
-
b
ased
s
y
s
te
m
s
lik
e
Mo
d
-
s
ec
u
r
it
y
,
it
w
i
ll
b
e
b
lo
ck
ed
.
Ho
w
e
v
er
,
w
h
en
th
at
co
n
te
n
t
is
ac
ce
p
tab
le
f
o
r
th
e
w
eb
s
ite,
t
h
e
ad
m
in
is
tr
at
o
r
n
ee
d
s
to
ed
it
t
h
e
f
ilter
to
ad
d
an
e
x
ce
p
tio
n
to
th
i
s
ca
s
e
a
n
d
n
ee
d
s
to
e
n
r
ic
h
th
e
r
u
le
li
s
t.
T
h
is
p
r
o
ce
s
s
s
o
m
eti
m
e
s
is
co
m
p
licated
a
n
d
ca
n
m
ak
e
w
eb
ap
p
licatio
n
p
r
o
tecti
o
n
less
e
f
f
ec
tiv
e.
T
h
e
is
s
u
es
d
is
c
u
s
s
ed
in
a)
an
d
b)
a
r
e
o
u
r
p
r
ac
tical
ex
p
er
ien
ce
w
h
e
n
i
m
p
le
m
e
n
ti
n
g
d
if
f
er
en
t
u
n
u
s
u
al
ac
ce
s
s
d
etec
tio
n
m
o
d
els.
T
h
e
ef
f
ec
tiv
e
n
e
s
s
in
p
r
ac
tice
o
f
p
r
ev
io
u
s
s
tu
d
ie
s
i
n
d
etec
ti
n
g
u
n
u
s
u
al
r
eq
u
e
s
ts
is
n
o
t
h
ig
h
.
I
n
f
ac
t,
w
h
en
t
h
o
s
e
ap
p
r
o
ac
h
es
ar
e
ap
p
lied
o
n
ab
n
o
r
m
al
d
ata
u
s
i
n
g
Mo
d
Secu
r
it
y
t
o
o
l
[
8
]
,
th
e
h
ig
h
e
s
t
r
ec
a
ll sco
r
e
is
o
n
l
y
ab
o
u
t 3
0
%
.
I
n
o
r
d
er
to
o
v
er
co
m
e
t
h
e
w
ea
k
n
e
s
s
e
s
t
h
at
h
a
v
e
b
ee
n
p
o
in
te
d
o
u
t,
in
t
h
i
s
p
ap
er
,
a
n
e
w
m
et
h
o
d
to
b
u
ild
a
w
e
b
f
ir
e
w
all
b
ased
o
n
D
y
n
a
m
ic
W
eb
A
p
p
licatio
n
P
r
o
f
ili
n
g
(
DW
A
P
)
an
al
y
s
i
s
i
s
p
r
o
p
o
s
ed
.
DW
A
P
is
a
m
et
h
o
d
to
s
u
m
m
ar
ize
th
e
c
h
ar
ac
ter
is
t
ics
o
f
a
s
p
ec
i
f
ic
w
eb
s
i
te’
s
UR
I
s
.
T
h
ese
p
r
o
p
e
r
ties
in
cl
u
d
e
m
et
h
o
d
s
(
GE
T
,
P
OST
)
,
h
ea
d
er
s
an
d
p
ar
am
eter
s
o
f
th
e
U
R
I
s
.
B
ased
o
n
th
e
DW
A
P
ap
p
licatio
n
an
al
y
s
is
,
t
h
e
f
o
llo
w
in
g
co
n
tr
ib
u
tio
n
s
ar
e
p
r
esen
ted
in
t
h
is
p
a
p
er
:
A
p
p
l
y
in
g
DW
A
P
to
ab
n
o
r
m
al
d
etec
tio
n
s
y
s
te
m
s
.
T
h
e
p
r
o
b
lem
s
d
is
c
u
s
s
ed
ab
o
v
e
ca
n
b
e
s
o
lv
ed
if
th
e
d
etec
tio
n
m
o
d
el
is
tr
ain
ed
u
s
i
n
g
ea
ch
UR
I
s
in
ce
t
h
e
v
ar
iab
le
v
alu
e
s
g
en
er
ated
f
r
o
m
e
ac
h
UR
I
ar
e
n
o
lo
n
g
er
s
p
ar
e
an
d
an
o
m
alie
s
ar
e
ea
s
il
y
r
ec
o
g
n
ized
i
f
t
h
e
ab
n
o
r
m
al
f
ea
t
u
r
e
d
escr
ib
ed
in
t
h
e
e
x
a
m
p
le
in
b)
is
d
etec
ted
.
Mo
r
eo
v
er
,
b
y
d
ev
elo
p
in
g
a
s
ep
ar
ate
m
o
d
el
f
o
r
ea
ch
UR
I
,
it
is
p
o
s
s
ib
le
to
ex
tr
ac
t
n
e
w
f
ea
t
u
r
es
p
r
esen
tin
g
t
h
e
c
h
ar
ac
ter
is
tic
s
o
f
th
e
m
et
h
o
d
,
h
ea
d
er
an
d
p
ar
a
m
eter
s
o
f
ea
ch
UR
I
.
T
h
is
co
u
ld
n
o
t
b
e
d
o
n
e
in
p
r
ev
io
u
s
ap
p
r
o
ac
h
es.
T
h
ese
f
ea
t
u
r
es
f
u
l
f
i
ll
t
h
e
u
lti
m
a
te
p
u
r
p
o
s
e
o
f
DW
A
P
ap
p
licatio
n
th
a
t
is
to
o
p
tim
ize
th
e
ab
n
o
r
m
al
r
eq
u
est
d
etec
tio
n
o
n
ea
ch
U
R
I
.
A
p
ar
t
f
r
o
m
ap
p
l
y
i
n
g
DW
A
P
to
d
etec
t
ab
n
o
r
m
al
r
eq
u
e
s
ts
,
a
r
ea
l
-
ti
m
e
o
p
ti
m
izatio
n
f
o
r
m
o
d
el
u
p
d
ate
m
et
h
o
d
is
a
ls
o
p
r
ese
n
ted
.
T
h
is
is
s
u
e
p
la
y
s
a
v
er
y
i
m
p
o
r
ta
n
t
r
o
le
in
th
e
w
eb
attac
k
d
etec
tio
n
m
o
d
el
as
w
ell
as
an
o
m
alo
u
s
ac
ce
s
s
d
etec
tio
n
b
ased
o
n
a
n
o
m
al
y
-
b
a
s
ed
al
g
o
r
ith
m
s
.
Ho
w
e
v
er
,
p
r
ev
io
u
s
w
o
r
k
s
d
id
n
o
t
p
a
y
m
u
c
h
atte
n
tio
n
to
th
is
p
r
o
b
le
m
.
A
ll
c
u
r
r
en
t
s
ec
u
r
it
y
ap
p
licatio
n
s
n
ee
d
to
b
e
co
n
s
ta
n
tl
y
u
p
d
ated
to
a
cc
o
m
m
o
d
ate
n
e
w
attac
k
s
.
T
h
at
is
th
e
m
ai
n
r
ea
s
o
n
w
h
y
Mo
d
Secu
r
it
y
i
s
s
till
a
p
o
p
u
lar
s
e
cu
r
it
y
d
etec
tio
n
to
o
l to
d
ay
b
ec
au
s
e
it
s
r
u
le
s
y
s
t
e
m
i
s
k
ep
t
u
p
to
d
ate
an
d
m
ai
n
tain
ed
b
y
co
m
m
u
n
it
y
co
n
tr
ib
u
tio
n
s
.
I
t c
a
n
b
e
s
ee
n
t
h
at
a
n
o
m
al
y
-
b
ased
m
o
d
els n
ee
d
to
b
e
tr
ai
n
ed
b
ased
o
n
t
h
e
d
ata
f
r
o
m
t
h
e
s
p
ec
i
f
ic
co
n
ce
r
n
ed
w
eb
s
ite
.
I
n
f
ac
t,
th
e
n
u
m
b
er
o
f
u
n
u
s
u
a
l
r
eq
u
ests
is
m
u
ch
s
m
al
ler
th
a
n
n
o
r
m
al
r
eq
u
est
s
,
w
h
ic
h
g
e
n
er
ates
a
b
u
r
d
en
j
o
b
f
o
r
ad
m
i
n
is
tr
ato
r
s
i
n
co
m
p
o
s
in
g
tr
ai
n
in
g
d
ata.
I
n
o
r
d
er
to
tack
le
th
is
p
r
o
b
le
m
,
a
r
eq
u
est
g
r
o
u
p
i
n
g
m
et
h
o
d
is
p
r
o
p
o
s
ed
to
s
u
p
p
o
r
t
th
e
d
ata
class
if
icat
io
n
p
r
o
ce
s
s
.
T
h
is
m
et
h
o
d
ca
n
h
elp
r
ed
u
ce
ad
m
in
i
s
tr
ato
r
’
s
d
ata
co
m
p
o
s
i
n
g
ti
m
e
b
y
5
0
-
7
0
%,
th
u
s
m
a
k
in
g
o
u
r
p
r
o
p
o
s
ed
an
o
m
al
y
-
b
ased
d
etec
tio
n
m
o
d
el
ea
s
y
to
d
ep
lo
y
in
p
r
ac
ti
ce
.
E
x
p
er
i
m
e
n
tal
r
e
s
u
lt
s
o
n
t
h
e
s
a
m
e
d
ata
s
et
s
h
o
w
t
h
at
t
h
er
e
i
s
a
s
i
g
n
i
f
ican
t
i
m
p
r
o
v
e
m
en
t
i
n
d
etec
tio
n
p
er
f
o
r
m
a
n
ce
o
f
o
u
r
m
e
th
o
d
.
T
h
e
r
ec
all
in
d
ex
o
f
th
e
n
e
w
a
p
p
r
o
ac
h
ca
n
r
ea
ch
9
0
%.
T
h
e
f
o
llo
w
in
g
co
n
te
n
t
o
f
th
e
p
ap
er
is
o
r
g
an
ized
as
f
o
llo
w
s
.
Sectio
n
2
p
r
esen
t
s
all
r
elate
d
w
o
r
k
s
o
n
ab
n
o
r
m
al
r
eq
u
est
d
etec
tio
n
tech
n
iq
u
es.
T
h
e
n
e
w
l
y
p
r
o
p
o
s
ed
m
et
h
o
d
is
p
r
ese
n
ted
i
n
s
ec
tio
n
3
.
Sectio
n
4
in
tr
o
d
u
ce
s
s
o
m
e
m
ai
n
ap
p
licatio
n
s
o
f
t
h
e
n
e
w
f
r
a
m
e
w
o
r
k
.
E
x
p
er
i
m
e
n
tal
r
es
u
lt
s
an
d
all
d
is
c
u
s
s
io
n
s
ar
e
i
n
c
lu
d
ed
in
s
ec
tio
n
5
.
Sectio
n
6
co
n
clu
d
e
s
w
h
at
h
a
v
e
b
ee
n
d
o
n
e
an
d
d
is
cu
s
s
es so
m
e
s
u
g
g
e
s
tio
n
s
f
o
r
th
e
f
u
tu
r
e
w
o
r
k
s
.
2.
RE
L
AT
E
D
WO
RK
2
.
1
.
Web
a
t
t
a
ck
det
ec
t
io
n r
esea
rc
h
T
h
er
e
ar
e
t
w
o
m
ai
n
t
y
p
es
o
f
w
eb
at
tack
d
etec
tio
n
s
y
s
te
m
s
.
T
h
e
class
if
ica
tio
n
is
m
a
in
l
y
b
ased
o
n
th
e
d
etec
tio
n
m
ec
h
a
n
i
s
m
o
f
th
e
m
eth
o
d
s
.
Sig
n
at
u
r
e
-
b
ased
m
et
h
o
d
s
[
3
,
4
]
:
th
i
s
i
s
a
w
ell
-
k
n
o
w
n
a
p
p
r
o
ac
h
an
d
h
a
s
b
ee
n
in
v
e
s
t
ig
ated
b
y
m
a
n
y
r
esear
ch
er
s
.
So
f
ar
,
th
e
r
esear
ch
co
m
m
u
n
it
y
o
f
w
eb
attac
k
d
etec
tio
n
h
a
s
b
u
il
t
u
p
a
co
m
p
lete
C
o
r
e
R
u
l
e
Set [
9
]
to
s
u
p
p
o
r
t n
et
w
o
r
k
u
s
e
r
s
.
C
u
r
r
en
t
l
y
,
t
h
e
C
o
r
e
R
u
l
e
S
et
is
u
s
ed
in
m
o
s
t o
f
t
h
e
w
eb
f
i
r
e
w
all
s
[
3
].
An
o
m
al
y
-
b
ased
m
eth
o
d
s
:
t
h
er
e
h
av
e
b
ee
n
m
an
y
d
i
f
f
er
e
n
t
a
n
o
m
a
l
y
b
ased
ap
p
r
o
ac
h
es
o
n
n
e
t
w
o
r
k
s
ec
u
r
it
y
.
On
e
o
f
t
h
o
s
e
ap
p
r
o
ac
h
es
is
b
ased
o
n
t
h
e
m
a
n
u
al
f
ea
t
u
r
e
e
x
t
r
ac
tio
n
tec
h
n
iq
u
es.
Sh
i
a
n
d
et
al.
[
10
]
p
r
esen
t
a
lis
t
o
f
f
ea
t
u
r
es
f
o
r
q
u
er
ie
s
t
h
at
i
n
cl
u
d
e
U
R
I
’
s
p
r
o
p
er
ties
,
s
u
c
h
a
s
le
n
g
th
,
q
u
an
tit
y
,
t
y
p
e
an
d
d
an
g
er
o
u
s
lev
els
o
f
ea
c
h
f
ea
t
u
r
e.
A
f
ter
t
h
at,
t
h
e
y
ap
p
lied
Naïv
e
B
a
y
es
,
Dec
is
io
n
tr
ee
an
d
SVM
al
g
o
r
ith
m
o
n
th
o
s
e
f
ea
t
u
r
es
to
d
etec
t
ab
n
o
r
m
al
r
eq
u
ests
.
A
o
th
er
ap
p
r
o
ac
h
is
b
ased
o
n
th
e
n
at
u
r
al
lan
g
u
ag
e
p
r
o
ce
s
s
in
g
.
Z
h
an
g
M
an
d
et
al.
[
11
]
in
tr
o
d
u
ce
d
a
m
eth
o
d
th
a
t
u
s
es
C
N
N
to
class
i
f
y
t
h
e
attac
k
s
.
W
o
r
d
2
v
ec
m
o
d
el
is
u
s
ed
to
tr
a
n
s
f
o
r
m
th
e
r
a
w
r
eq
u
est
in
to
a
m
atr
i
x
,
an
d
th
e
n
a
C
NN
i
s
ad
o
p
ted
to
ex
tr
ac
t
r
eq
u
est
’
s
f
ea
tu
r
e
s
.
T
h
e
r
esear
ch
[
12
]
in
tr
o
d
u
ce
s
an
o
th
er
ap
p
r
o
ac
h
u
s
i
n
g
Gate
d
r
ec
u
r
r
en
t
u
n
it
(
G
R
U)
to
a
n
al
y
ze
t
h
e
co
n
ten
t
s
o
f
th
e
r
eq
u
est
s
.
E
v
er
y
ch
ar
ac
t
er
in
t
h
e
r
eq
u
est
is
co
n
v
er
ted
i
n
to
a
o
n
e
-
h
o
t v
ec
to
r
w
it
h
1
2
9
d
i
m
en
s
io
n
s
,
an
d
ev
er
y
ce
ll
i
n
GR
U
is
u
s
ed
to
a
n
al
y
ze
t
h
is
r
eq
u
est
’
s
co
n
te
n
t.
Yan
g
[
13
]
also
atte
m
p
ts
a
s
i
m
ilar
m
et
h
o
d
t
h
a
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
10
,
No
.
5
,
Octo
b
e
r
2
0
2
0
:
5
3
3
5
-
5346
5338
u
s
e
s
GR
U
to
clas
s
i
f
y
r
eq
u
e
s
ts
.
I
n
th
is
r
esear
c
h
,
h
e
u
s
e
s
an
e
n
co
d
in
g
m
et
h
o
d
w
h
ic
h
r
ec
o
n
s
tr
u
cts
a
ch
ar
ac
ter
in
to
a
2
-
d
i
m
e
n
s
io
n
al
m
atr
i
x
.
T
h
e
au
t
h
o
r
s
o
f
t
h
e
r
esear
ch
[
14
]
u
s
e
N
-
g
r
a
m
a
n
d
Gen
er
ic
Feat
u
r
e
Selectio
n
al
g
o
r
ith
m
s
to
ex
tr
a
ct
f
ea
t
u
r
es
f
r
o
m
D
AR
P
A
an
d
E
C
M
L
/P
KDD2
0
0
7
d
atasets
[
4
]
.
I
n
o
r
d
e
r
to
d
etec
t
ab
n
o
r
m
al
r
eq
u
est
s
,
t
h
e
y
ap
p
lied
s
o
m
e
cl
u
s
ter
in
g
al
g
o
r
ith
m
s
li
k
e
C
4
.
5
,
C
AR
T
,
R
a
n
d
o
m
f
o
r
est
o
r
r
an
d
o
m
tr
ee
.
Asi
d
e
f
r
o
m
ap
p
ly
in
g
a
n
o
m
al
y
-
b
ased
to
d
etec
t
ab
n
o
r
m
al
r
eq
u
est
s
i
n
g
e
n
er
al
,
th
er
e
ar
e
also
s
o
m
e
o
th
er
r
esear
ch
es
f
o
cu
s
in
g
o
n
d
etec
ti
n
g
s
o
m
e
co
m
m
o
n
attac
k
s
o
n
w
eb
ap
p
licatio
n
[
15
,
16
].
I
n
p
ar
ticu
lar
,
Na
g
ar
j
u
n
a
n
d
Ah
a
m
ad
[
1
7
]
p
r
esen
ted
a
n
attac
k
d
etec
tio
n
m
e
th
o
d
b
ased
o
n
i
m
ag
e
p
r
o
ce
s
s
in
g
tech
n
iq
u
e
to
d
etec
t
s
p
ec
ial
ch
a
r
ac
ter
s
th
at
r
ep
r
esen
t
X
SS
atta
ck
s
.
Yo
n
g
Ya
n
g
[
1
8
]
in
tr
o
d
u
c
ed
a
n
ap
p
r
o
ac
h
to
d
etec
t
an
o
m
a
lies
b
y
a
n
al
y
z
in
g
t
h
e
s
eq
u
en
ce
o
f
w
eb
ac
ce
s
s
b
eh
a
v
io
r
s
.
I
n
ad
d
itio
n
,
J
ag
d
is
h
et
al.
[
1
9
]
d
esig
n
ed
an
an
o
m
a
l
y
d
etec
ti
o
n
s
y
s
te
m
i
n
E
-
co
m
m
er
ec
s
y
s
te
m
s
b
ased
o
n
f
ea
tu
r
e
s
s
h
o
w
i
n
g
b
u
s
i
n
es
s
ch
ar
ac
ter
is
tic
s
s
u
c
h
as
p
r
ice,
g
o
o
d
s
,
etc.
T
h
es f
ea
t
u
r
es a
r
e
al
s
o
ad
o
p
ted
in
th
i
s
p
ap
er
,
b
u
t a
t
a
m
o
r
e
g
e
n
er
al
lev
el
an
d
t
h
e
ex
tr
ac
tio
n
p
r
o
ce
s
s
o
f
th
e
s
e
f
ea
t
u
r
es i
s
i
m
p
le
m
e
n
ted
au
to
m
atica
ll
y
.
2
.
2
.
Da
t
a
up
da
t
ing
a
nd
m
o
nito
ri
ng
re
s
ea
rc
h
T
o
o
v
er
co
m
e
d
ata
i
m
b
ala
n
ce
p
r
o
b
lem
in
t
h
e
tr
ai
n
i
n
g
p
r
o
ce
s
s
a
s
w
el
l
a
s
i
n
t
h
e
ab
n
o
r
m
al
r
eq
u
est
d
etec
tio
n
p
r
o
ce
s
s
,
th
er
e
h
a
v
e
b
ee
n
s
o
m
e
r
esear
ch
e
s
an
d
p
r
o
p
o
s
als.
Hu
Y
[
20
]
p
r
o
p
o
s
es
a
h
u
m
a
n
-
m
ac
h
i
n
e
s
y
s
te
m
to
i
m
p
r
o
v
e
d
etec
tio
n
m
o
d
el
s
.
On
t
h
i
s
s
y
s
te
m
,
t
h
e
r
o
le
o
f
th
e
ex
p
er
t
is
to
r
e
-
c
la
s
s
i
f
y
th
e
d
ata
a
f
ter
r
u
n
n
i
n
g
t
h
e
u
n
lab
eled
clas
s
if
i
ca
tio
n
.
T
h
e
au
th
o
r
u
s
es
K
-
m
ea
n
to
class
i
f
y
th
e
d
atase
t
in
to
t
w
o
g
r
o
u
p
s
a
n
d
s
elec
tes
a
c
er
tain
p
er
ce
n
tag
e
f
r
o
m
t
h
o
s
e
2
g
r
o
u
p
s
to
r
ec
lass
if
y
.
I
n
t
h
e
r
esear
ch
[
21
]
,
Do
n
g
an
d
et
al.
p
r
esen
te
a
s
o
lu
tio
n
to
r
ec
las
s
i
f
y
r
eq
u
e
s
t
s
w
h
ic
h
ar
e
n
o
t in
t
h
e
b
o
u
n
d
ar
y
tr
ain
ed
b
y
t
h
e
SVM
al
g
o
r
ith
m
.
3.
F
E
AT
U
RE
E
XT
RAC
T
I
O
N
USI
N
G
DW
AP
3
.
1
.
Dy
na
m
ic
w
eb
a
pp
lica
t
io
n pr
o
f
ilin
g
DW
A
P
r
ep
r
esen
ts
t
h
e
UR
I
t
h
at
m
ap
s
a
w
eb
ap
p
licatio
n
in
t
o
a
tr
ee
.
DW
A
P
co
n
tain
s
i
n
f
o
r
m
atio
n
o
f
all
UR
I
s
in
a
w
eb
p
ag
e,
i
n
clu
d
in
g
s
tatic
UR
I
a
n
d
d
y
n
a
m
ic
U
R
I
.
Static
U
R
I
is
a
f
ile
p
ath
o
f
s
tatic
f
ile
s
s
u
ch
as
m
ed
ia
f
ile
[
*
.
j
p
g
,
*
.
p
n
g
,
*
.
g
if
]
o
r
f
iles
b
ein
g
u
s
ed
to
d
is
p
l
a
y
w
eb
p
ag
e
[
*
.
cs
s
,
*
.
j
s
,
*
.
h
t
m
l]
.
D
y
n
a
m
ic
UR
I
tr
an
s
f
er
s
p
ar
a
m
eter
s
to
t
h
e
web
ap
p
licatio
n
th
at
it
p
r
o
ce
s
s
e
s
.
Fig
u
r
e
2
illu
s
tr
ates
a
n
ex
a
m
p
le
o
f
a
DW
A
P
o
f
a
w
eb
ap
p
licatio
n
.
T
h
er
e
ar
e
2
ex
a
m
p
les
o
f
d
y
n
a
m
ic
U
R
I
,
w
h
ich
ar
e
“
h
t
tp
s
://
w
eb
.
co
m
/
l
o
g
in
.
j
s
p
”
s
h
o
w
i
n
g
th
e
p
u
r
p
o
s
e
o
f
lo
g
i
n
,
a
n
d
“h
ttp
s
://
w
eb
.
co
m
/p
r
o
d
u
ct/
v
ie
w
.
j
s
p
”
s
h
o
w
i
n
g
t
h
e
p
u
r
p
o
s
e
o
f
v
ie
w
i
n
g
p
r
o
d
u
cts.
T
h
ese
UR
I
s
h
a
v
e
d
i
f
f
er
en
t
q
u
er
ies.
Fo
r
e
x
a
m
p
le,
th
e
lo
g
i
n
.
j
s
p
w
i
ll
n
ee
d
th
e
v
a
lu
e
o
f
u
s
er
’
s
id
a
n
d
u
s
er
’
s
p
ass
w
o
r
d
,
w
h
ile
t
h
e
v
ie
w
.
j
s
p
w
ill
n
ee
d
in
f
o
r
m
a
tio
n
ab
o
u
t
th
e
d
is
p
la
y
ed
p
r
o
d
u
ct
id
s
.
T
h
e
p
r
o
p
er
ty
o
f
t
h
i
s
in
f
o
r
m
atio
n
is
d
i
f
f
er
en
t
in
ter
m
s
o
f
v
al
u
e,
d
ata
t
y
p
e,
etc.
I
t
is
n
ec
ess
ar
y
to
b
u
ild
in
d
i
v
id
u
al
m
o
d
el
s
f
o
r
ea
ch
U
R
I
t
o
b
e
a
b
l
e
t
o
r
e
c
o
g
n
i
z
e
t
h
e
s
m
a
l
l
e
s
t
a
b
n
o
r
m
a
l
c
h
a
n
g
e
s
t
o
a
v
o
i
d
c
o
n
f
u
s
i
o
n
w
h
e
n
a
n
a
l
y
z
i
n
g
m
u
l
t
i
p
l
e
U
R
I
’
s
i
n
f
o
r
m
a
t
i
o
n
a
t
t
h
e
s
a
m
e
t
im
e
.
I
n
t
h
i
s
r
e
s
e
a
r
c
h
,
w
e
a
r
e
f
o
c
u
s
i
n
g
o
n
a
n
a
l
y
z
i
n
g
a
n
d
e
v
a
l
u
a
t
i
n
g
t
h
e
b
e
h
a
v
i
o
r
o
f
th
e
d
y
n
a
m
ic
UR
I
.
Fig
u
r
e
2
.
D
y
n
a
m
ic
w
eb
ap
p
licatio
n
p
r
o
f
ilin
g
s
a
m
p
le
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
A
n
a
d
a
p
tive
a
n
o
ma
ly
r
eq
u
est
d
etec
tio
n
fr
a
mewo
r
k
b
a
s
ed
…
(
C
h
o
Do
X
u
a
n
)
5339
3
.
2
.
F
ea
t
ure
ex
t
ra
ct
io
n in D
WAP
a
na
ly
s
is
T
h
e
r
eq
u
est’
s
f
ea
t
u
r
es
in
D
W
A
P
an
al
y
s
is
m
et
h
o
d
ar
e
b
u
ilt
to
d
etec
t
th
e
ab
n
o
r
m
a
l
r
eq
u
ests
a
t
th
e
co
m
p
o
n
e
n
t
le
v
el.
B
y
a
n
al
y
zin
g
o
n
ea
ch
UR
I
,
ev
er
y
r
eq
u
est
’
s
co
m
p
o
n
e
n
t
li
k
e
h
ea
d
er
o
r
s
et
o
f
p
ar
a
m
eter
s
is
d
ee
p
ly
a
n
al
y
ze
d
.
I
n
o
r
d
er
to
d
o
th
at,
th
e
f
ea
tu
r
e
s
et
is
d
iv
i
d
ed
in
to
2
g
r
o
u
p
s
.
T
h
e
f
ir
s
t
g
r
o
u
p
is
u
s
ed
to
lo
o
k
f
o
r
ab
n
o
r
m
a
l
ch
ar
ac
ter
is
tic
s
ap
p
ea
r
ed
in
th
e
attac
k
s
.
T
h
e
s
ec
o
n
d
g
r
o
u
p
is
ex
p
lo
ited
to
an
al
y
ze
ab
n
o
r
m
al
co
n
ten
t i
n
ea
ch
co
m
p
o
n
en
t o
f
th
e
r
eq
u
est.
3
.
2
.
1
.
M
a
licio
us
k
ey
w
o
r
ds
f
ea
t
ure
a.
At
t
a
ck
k
ey
w
o
rds
Ke
y
w
o
r
d
is
th
e
m
ai
n
id
en
ti
f
i
ca
tio
n
ch
ar
ac
ter
is
tic
o
f
s
o
m
e
t
y
p
es
o
f
attac
k
s
.
Fo
r
ex
a
m
p
l
e,
in
SQL
I
n
j
ec
tio
n
attac
k
,
t
h
e
attac
k
er
s
tr
y
to
f
in
d
a
w
a
y
to
i
n
s
er
t
th
eir
SQ
L
q
u
er
ies
i
n
to
t
h
e
d
ata
s
en
t
to
s
er
v
er
.
T
h
e
ap
p
ea
r
an
ce
o
f
th
o
s
e
k
e
y
w
o
r
d
s
in
t
h
e
r
eq
u
est
i
s
a
s
ig
n
to
d
eter
m
i
n
e
w
h
e
th
er
a
r
eq
u
es
t
is
an
attac
k
o
r
n
o
t.
T
h
e
k
e
y
w
o
r
d
s
lis
t
ed
in
T
ab
le
1
ar
e
s
u
m
m
ed
u
p
f
r
o
m
O
S
W
A
P
’
s
d
o
cu
m
e
n
t
ab
o
u
t
w
eb
attac
k
s
[
9
]
.
I
n
th
o
s
e
attac
k
s
,
th
e
k
e
y
w
o
r
d
is
t
h
e
m
o
s
t
i
m
p
o
r
ta
n
t
co
m
p
o
n
en
t
to
i
n
s
er
t
illeg
a
l
q
u
er
ie
s
.
T
h
ese
k
e
y
w
o
r
d
s
ar
e
p
u
t
i
n
to
OSW
A
P
’
s
r
u
le
s
to
d
etec
t
th
e
attac
k
s
.
Ho
w
e
v
er
,
th
e
ev
al
u
ati
o
n
b
ased
o
n
l
y
o
n
th
e
ap
p
ea
r
an
ce
o
f
th
e
k
e
y
w
o
r
d
m
a
y
lead
to
t
h
e
in
co
r
r
ec
t
w
ar
n
in
g
b
ec
au
s
e
s
o
m
e
w
eb
s
i
tes
a
llo
w
th
e
ex
is
te
n
ce
o
f
t
h
o
s
e
k
e
y
w
o
r
d
s
.
T
h
er
ef
o
r
e,
w
h
e
n
b
u
ild
in
g
th
e
U
R
I
’
s
f
ea
t
u
r
es,
th
e
ap
p
ea
r
an
ce
o
f
th
ese
k
e
y
w
o
r
d
s
as
a
f
ea
t
u
r
e
s
h
o
u
ld
b
e
c
o
m
b
i
n
ed
w
it
h
o
th
er
f
ac
to
r
s
in
o
r
d
er
to
s
u
cc
ess
f
u
l
l
y
co
n
cl
u
d
e
w
h
e
th
er
t
h
e
r
eq
u
est
i
s
a
n
attac
k
o
r
n
o
t.
T
ab
le
1
s
u
m
m
ar
izes
s
o
m
e
Ma
licio
u
s
k
e
y
w
o
r
d
s
.
T
ab
le
1
.
Ma
licio
u
s
k
e
y
w
o
r
d
s
G
r
o
u
p
L
i
st
o
f
k
e
y
w
o
r
d
S
Q
L
i
n
j
e
c
t
i
o
n
C
r
e
a
t
e
,
d
r
o
p
,
a
l
t
e
r
,
t
r
u
n
c
a
t
e
,
se
l
e
c
t
,
i
n
se
r
t
,
u
p
d
a
t
e
,
d
e
l
e
t
e
,
o
r
,
a
n
d
,
u
n
i
o
n
,
j
o
i
n
,
w
h
e
r
e
,
i
f
,
c
o
n
v
e
r
t
,
c
a
st
,
h
e
x
t
o
r
a
w
O
S
C
o
mm
a
n
d
I
n
j
e
c
t
i
o
n
C
a
t
,
l
s,
mv
,
c
p
,
r
m
,
p
e
r
l
,
p
y
t
h
o
n
,
r
u
b
y
,
l
u
a
X
M
L
I
n
j
e
c
t
i
o
n
l
t
,
g
t
,
n
o
d
e
,
x
ml
,
e
n
c
o
d
i
n
g
,
c
d
a
t
a
,
d
o
c
t
y
p
e
,
e
l
e
m
e
n
t
C
r
o
ss
-
si
t
e
S
c
r
i
p
t
i
n
g
O
n
l
o
a
d
,
o
n
mo
u
se
h
o
v
e
r
,
o
n
e
r
r
o
r
,
scri
p
t
,
i
f
r
a
me
,
v
a
r
,
a
l
e
r
t
b.
Ano
m
a
ly
no
n
-
let
t
er
No
n
-
le
tter
f
ea
t
u
r
es
ar
e
o
n
e
p
a
r
t
o
f
t
h
e
attac
k
s
ig
n
s
w
h
en
t
h
e
y
ac
co
m
p
a
n
y
w
i
th
t
h
e
k
e
y
w
o
r
d
s
s
h
o
w
n
in
T
ab
le
1
.
I
n
o
r
d
er
to
i
n
s
er
t
m
a
licio
u
s
k
e
y
w
o
r
d
s
,
a
ttack
er
s
m
u
s
t
f
i
n
d
t
h
e
w
a
y
to
p
ass
t
h
e
ap
p
licatio
n
's
co
m
p
iler
.
Fo
r
ex
a
m
p
le,
th
e
y
m
a
y
i
n
s
er
t
co
m
m
e
n
t
s
o
r
ch
ar
a
cter
s
to
d
ec
eiv
e
t
h
e
i
n
p
u
t
s
tr
u
ctu
r
e.
I
n
t
h
i
s
p
ap
er
,
th
e
f
r
eq
u
e
n
c
y
o
f
t
h
o
s
e
ch
ar
a
cter
s
in
t
h
e
q
u
er
i
es
is
u
s
ed
t
o
v
er
if
y
t
h
e
i
n
p
u
t
i
n
f
o
r
m
atio
n
.
T
h
ese
ch
ar
ac
ter
g
r
o
u
p
s
ar
e
lis
ted
in
T
ab
le
2
.
T
ab
le
2
.
No
n
letter
k
e
y
w
o
r
d
G
r
o
u
p
L
i
st
o
f
n
o
n
-
l
e
t
t
e
r
c
h
a
r
a
c
t
e
r
s
T
h
e
c
h
a
r
a
c
t
e
r
s
u
se
d
t
o
st
a
r
t
a
c
o
mm
e
n
t
o
r
t
o
mark
a
b
l
o
c
k
c
o
mm
e
n
t
--
[
st
a
r
t
c
o
mm
e
n
t
]
/
*
[
s
t
a
r
t
c
o
mm
e
n
t
]
*
/
[
e
n
d
c
o
mm
e
n
t
]
*
[
A
st
e
r
i
sk
]
#
[
h
a
sh
]
|
|
[
d
o
u
b
l
e
v
e
r
t
i
c
a
l
b
a
r
]
&
&
[
d
o
u
b
l
e
A
mp
e
r
san
d
]
;
[
S
e
mi
c
o
l
o
n
]
`
[
G
r
a
v
e
A
c
c
e
n
t
]
C
h
a
r
a
c
t
e
r
s u
se
d
i
n
c
o
mm
a
n
d
s
&
[
A
mp
e
r
san
d
]
&
&
[
d
o
u
b
l
e
A
mp
e
r
s
a
n
d
]
|
[
v
e
r
t
i
c
a
l
b
a
r
]
|
|
[
d
o
u
b
l
e
v
e
r
t
i
c
a
l
b
a
r
]
`
[
G
r
a
v
e
A
c
c
e
n
t
]
;
[
S
e
mi
c
o
l
o
n
]
$
[
D
o
l
l
a
r
S
i
g
n
]
C
h
a
r
a
c
t
e
r
s t
o
i
n
se
r
t
st
a
t
e
me
n
t
s
>
[
G
r
e
a
t
e
r
-
T
h
a
n
S
i
g
n
]
<
[
G
r
e
a
t
e
r
-
T
h
a
n
S
i
g
n
]
>
>
[
d
o
u
b
l
e
G
r
e
a
t
e
r
-
T
h
a
n
S
i
g
n
]
#
[
h
a
s
h
]
!
[
Ex
c
l
a
m
a
t
i
o
n
M
a
r
k
]
-
[
H
y
p
h
e
n
-
M
i
n
u
s
]
?
[
H
y
p
h
e
n
-
M
i
n
u
s
]
=
[
Eq
u
a
l
s
S
i
g
n
]
[
[
L
e
f
t
S
q
u
a
r
e
B
r
a
c
k
e
t
]
]
[
R
i
g
h
t
S
q
u
a
r
e
B
r
a
c
k
e
t
]
~
[
T
i
l
d
e
]
.
[
d
o
t
]
,
[
C
o
mm
a
]
%
[
P
e
r
c
e
n
t
S
i
g
n
]
:
[
C
o
l
o
n
]
+
[
P
l
u
s
s
i
g
n
]
(
[
L
e
f
t
B
r
a
c
k
e
t
]
)
[
R
i
g
t
h
B
r
a
c
k
e
t
]
{
[
L
e
f
t
C
u
r
l
y
B
r
a
c
k
e
t
]
}
[
R
i
g
h
t
C
u
r
l
y
B
r
a
c
k
e
t
]
/
>
[
b
a
c
k
sl
a
sh
g
r
e
a
t
e
r
]
.
.
/
[
d
o
t
d
o
t
sl
a
s
h
]
.
.
\
[
d
o
t
d
o
t
b
a
c
k
sl
a
s
h
]
C
h
a
r
a
c
t
e
r
s c
a
n
n
o
t
b
e
d
i
s
p
l
a
y
e
d
C
h
a
r
a
c
t
e
r
s w
i
t
h
U
n
i
c
o
d
e
c
o
d
e
sm
a
l
l
e
r
t
h
a
n
3
2
3
.
2
.
2
.
Ano
m
a
ly
re
qu
est
co
nt
ent
f
ea
t
ure
I
n
p
r
ev
io
u
s
r
esear
c
h
es
o
n
ab
n
o
r
m
al
ac
ce
s
s
d
etec
tio
n
,
t
h
e
co
n
ten
t
v
alu
e
s
o
f
ea
ch
r
eq
u
est
c
o
m
p
o
n
en
t
ar
e
n
o
t th
o
r
o
u
g
h
l
y
an
al
y
ze
d
.
T
h
e
m
ai
n
r
ea
s
o
n
is
ab
o
u
t t
h
e
d
if
f
er
en
ce
b
et
w
ee
n
th
e
f
u
n
ct
io
n
s
o
f
d
if
f
er
en
t U
R
I
s
.
E
ac
h
U
R
I
u
s
es
d
if
f
er
en
t
co
n
f
ig
u
r
at
io
n
f
o
r
its
h
e
ad
er
a
s
well
as
its
q
u
er
y
v
al
u
es.
T
h
is
ca
u
s
e
s
d
if
f
ic
u
lt
y
i
n
p
r
o
p
o
s
in
g
a
n
e
f
f
ec
tiv
e
m
et
h
o
d
b
ased
o
n
an
al
y
z
in
g
m
u
ltip
le
UR
I
s
to
d
etec
t
an
o
m
alies
s
u
ch
as
ex
p
lo
iti
n
g
lo
g
ic
er
r
o
r
s
o
r
d
etec
tin
g
r
eq
u
ests
f
r
o
m
s
ca
n
n
i
n
g
to
o
ls
.
I
n
th
is
p
ap
er
,
a
s
et
o
f
n
e
w
f
ea
t
u
r
es
e
x
tr
ac
ted
th
r
o
u
g
h
th
e
co
n
te
n
t
an
a
l
y
zi
n
g
p
r
o
ce
s
s
f
o
r
ea
ch
co
m
p
o
n
e
n
t
o
f
a
r
eq
u
est
is
u
til
ized
f
o
r
an
o
m
al
y
d
etec
tio
n
p
r
o
ce
s
s
.
Usi
n
g
t
h
is
f
ea
t
u
r
e
ex
tr
ac
tio
n
h
elp
s
r
ev
ie
w
a
n
d
ev
al
u
ate
t
h
e
v
alu
e
s
o
f
t
h
e
r
eq
u
est
co
m
p
o
n
en
t
s
i
n
d
etails
an
d
clea
r
l
y
.
Fo
r
i
n
s
ta
n
ce
,
i
n
r
ea
l
s
ce
n
ar
io
s
,
ab
n
o
r
m
a
l
r
eq
u
es
ts
c
an
b
e
id
e
n
ti
f
ied
i
f
t
h
er
e
i
s
a
s
t
r
an
g
e
UR
I
(
s
u
c
h
a
s
w
eb
s
h
ell)
,
o
r
a
s
tr
an
g
e
f
ield
in
th
e
h
ea
d
er
(
s
u
ch
as
X
-
Fo
r
w
ar
d
ed
-
Fo
r
u
s
ed
to
b
y
p
ass
f
i
r
e
w
all)
,
o
r
a
n
e
w
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
10
,
No
.
5
,
Octo
b
e
r
2
0
2
0
:
5
3
3
5
-
5346
5340
p
ar
am
eter
t
h
at
s
eld
o
m
o
r
n
e
v
er
ap
p
ea
r
s
in
r
eq
u
ests
b
y
n
o
r
m
al
u
s
e
r
s
.
T
h
e
p
r
o
p
o
s
ed
f
ea
tu
r
es
ca
n
b
e
u
s
ed
to
b
u
ild
t
h
e
p
r
o
f
ile
tr
ee
.
Fu
r
t
h
er
,
th
i
s
tr
ee
is
u
s
ed
to
d
etec
t
ab
n
o
r
m
al
r
eq
u
est
s
b
y
co
m
p
ar
i
n
g
th
eir
co
n
ten
t
s
w
i
th
th
e
d
ata
in
t
h
e
tr
ee
.
I
f
t
h
er
e
i
s
an
y
m
is
m
atc
h
i
n
g
b
et
w
ee
n
t
h
e
r
eq
u
est
s
a
n
d
t
h
e
p
r
o
f
ile
tr
ee
,
th
o
s
e
r
eq
u
e
s
ts
m
a
y
b
e
co
n
s
id
er
ed
as
ab
n
o
r
m
al.
T
h
is
i
s
t
h
e
i
m
p
r
o
v
e
m
e
n
t
o
f
t
h
e
DW
A
P
an
al
y
s
is
b
ased
m
et
h
o
d
co
m
p
ar
ed
w
it
h
p
r
ev
io
u
s
m
eth
o
d
s
.
a.
An
o
m
al
y
h
ea
d
er
v
alu
e
Hea
d
er
s
ar
e
im
p
o
r
tan
t
an
d
ar
e
f
r
eq
u
en
tl
y
ch
a
n
g
ed
tar
g
ets.
I
n
th
i
s
p
ap
er
,
f
o
u
r
m
ai
n
h
ea
d
er
f
ield
s
ar
e
in
v
e
s
ti
g
ated
,
w
h
ich
ar
e
C
o
n
ten
t
-
T
y
p
e,
A
cc
ep
t,
A
cc
ep
t
-
C
h
ar
s
et,
A
cc
ep
t
-
E
n
co
d
in
g
.
T
h
e
v
al
u
es
o
f
t
h
o
s
e
h
ea
d
er
s
ar
e
ex
tr
ac
ted
an
d
n
o
r
m
alize
d
to
f
o
r
m
a
v
ec
to
r
.
I
n
o
r
d
er
to
f
ac
ilit
ate
th
e
f
ea
t
u
r
e
class
i
f
icatio
n
p
r
o
ce
s
s
,
th
e
f
ea
tu
r
es a
r
e
d
iv
id
ed
in
to
g
r
o
u
p
s
as b
elo
w
:
Gr
o
u
p
1
:
in
cl
u
d
in
g
t
h
e
v
al
u
es
o
f
C
o
n
te
n
t
-
T
y
p
e
[
5
]
an
d
A
cc
ep
t
[
5
]
.
I
ts
s
tr
u
ctu
r
e
co
n
tai
n
s
t
y
p
e
an
d
s
u
b
-
t
y
p
e.
T
h
e
v
alu
e
o
f
th
is
t
y
p
e
an
d
s
u
b
t
y
p
e
ar
e
co
m
p
ar
ed
w
it
h
th
at
o
f
t
y
p
e
a
n
d
s
u
b
t
y
p
e
lis
t
in
n
o
r
m
a
l
r
eq
u
ests
.
Me
t
h
o
d
s
an
d
p
r
o
ce
d
u
r
es o
f
t
h
is
i
n
v
esti
g
atio
n
ar
e
d
escr
ib
ed
in
A
l
g
o
r
it
h
m
1
:
Algorithm
1
. Check value group 1
Input: Content
-
type, Accept
Output: Vector represent existing value of header
1: fu
nction: CHECK_VALUE_GROUP_1 (header):
2: types/subtypes <
-
header
3: normal_types <
-
list type of header in normal requests
4: normal_subtypes <
-
list subtype of header in normal requests
5: type_feature <
-
a size of normal_types
-
array of 0
6: subtype
_feature <
-
a size of normal_subtypes
-
array of 0
7: for i <
-
0 to size of types:
8: if types_i exist in normal_types
9: type_feature [position of types_i in normal_types] = 1
10: for i <
-
0 to size of subtypes:
11: if subtypes_i exist in normal_
subtypes
12: subtype_feature [position of subtypes_i in normal_subtypes]=1
13: return type_feature+subtype_feature
Gr
o
u
p
2
:
in
clu
d
in
g
th
e
v
al
u
es
o
f
A
cc
ep
t
-
C
h
ar
s
et
[
5
]
an
d
A
cc
ep
t
-
E
n
co
d
in
g
[
5
]
.
T
h
e
f
ea
tu
r
es
o
f
Gr
o
u
p
2
ar
e
ex
tr
ac
ted
b
y
co
m
p
ar
i
n
g
t
h
ese
v
al
u
es
w
it
h
t
h
o
s
e
i
n
t
h
e
n
o
r
m
al
r
eq
u
e
s
ts
.
Me
t
h
o
d
s
an
d
p
r
o
ce
d
u
r
es
to
ex
tr
ac
t f
ea
t
u
r
e
v
al
u
e
s
ar
e
d
escr
ib
ed
in
th
e
A
lg
o
r
it
h
m
2
:
Algorithm
2
. Check value group 2
Input: Accept
-
Encoding, Accept
-
Charset,
Output: Vector represent existing value of header
1:
function: CHECK_VALUE_GROUP_2 (header):
2: values <
-
header
3: normal_values <
-
list normal value of header in normal requests
4: values_feature <
-
a size of normal_values
-
array of 0
5: for i <
-
0 to size of normal_values:
6: if value_i exist in nor
mal_values:
7: values_feature [position of value_i in normal_values] = 1
8: return values_feature
b.
An
o
m
al
y
p
ar
a
m
eter
Algorithm
3
. Get Structured data feature
Input: request
Output: key
-
value feature
1: function GET_
STRUCTED_DATA_NAME(
structed_data)
2: normal_key <
-
list normal key in normal requests
3: request_name <
-
list name in request
4: num_abnormal_key <
-
0
5: name_value_feature <
-
[]
6: for name in request_key
7: if key not in normal_key
8: num_abnormal_key +=1
9:
else
10
:
ke
y_
va
lu
e_
fe
at
ur
e
<
-
key_value_feature
+
[raito
of
alphabet
character
in
value,
raito of digit in value, raito of other character in value]
11: key_value_feature <
-
key_value_feature + [num_abnormal_key]
12: return num_abnormal_key
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
A
n
a
d
a
p
tive
a
n
o
ma
ly
r
eq
u
est
d
etec
tio
n
fr
a
mewo
r
k
b
a
s
ed
…
(
C
h
o
Do
X
u
a
n
)
5341
User
r
eq
u
est
s
u
s
u
al
l
y
co
n
tai
n
i
m
p
o
r
ta
n
t
i
n
f
o
r
m
atio
n
f
o
r
w
eb
-
s
er
v
er
to
p
r
o
ce
s
s
.
T
h
e
co
n
ten
t
o
f
th
e
r
eq
u
est
m
a
y
b
e
p
r
esen
ted
i
n
a
f
o
r
m
o
f
s
tr
u
ct
u
r
ed
d
ata
s
u
ch
as
t
h
e
q
u
er
y
in
GE
T
m
et
h
o
d
o
r
th
e
p
ay
lo
ad
in
P
OST
/
P
U
T
m
et
h
o
d
,
o
r
u
n
s
tr
u
ctu
r
ed
d
ata
lik
e
d
o
cu
m
e
n
ts
,
f
i
les,
etc.
Fo
r
s
tr
u
ctu
r
ed
d
ata,
th
e
v
al
u
es
o
f
le
n
g
th
,
r
atio
o
f
letter
s
an
d
n
u
m
b
er
s
in
ea
ch
i
n
p
u
t
s
tr
i
n
g
ar
e
e
x
t
r
ac
ted
.
A
d
d
itio
n
all
y
,
th
e
e
x
is
ten
ce
o
f
ab
n
o
r
m
al
p
ar
am
eter
s
is
also
ch
ec
k
ed
.
Me
th
o
d
s
an
d
p
r
o
ce
d
u
r
es
to
t
o
ex
a
m
i
n
e
an
o
m
a
l
y
p
ar
a
m
ete
r
s
ar
e
d
escr
ib
e
d
in
th
e
A
l
g
o
r
ith
m
3
ab
o
v
e.
4.
AP
P
L
I
CA
T
I
O
N
O
F
DW
AP
ANALY
SI
S O
N
W
E
B
AP
P
L
I
CAT
I
O
N
S
E
CUR
I
T
Y
4
.
1
.
DWAP
a
na
ly
s
is
f
o
r
a
no
m
a
ly
re
qu
est
det
ec
t
io
n
B
ased
o
n
th
e
f
ea
tu
r
es
o
b
tai
n
ed
f
r
o
m
t
h
e
DW
A
P
an
al
y
s
is
tec
h
n
iq
u
e
ap
p
lied
o
n
th
e
r
eq
u
est
co
m
p
o
n
e
n
t
p
r
esen
ted
in
Sect
io
n
3
.
2
,
f
u
r
th
er
p
r
o
ce
s
s
in
g
s
t
ep
s
ar
e
n
ee
d
ed
t
o
d
is
cr
im
i
n
a
te
n
o
r
m
al
ac
ce
s
s
e
s
f
r
o
m
ab
n
o
r
m
al
o
n
es.
I
n
t
h
i
s
p
ap
er
,
R
an
d
o
m
Fo
r
est
cla
s
s
if
ier
s
[
22
]
ar
e
ad
o
p
ted
to
d
is
ti
n
g
u
is
h
b
et
w
ee
n
ab
n
o
r
m
al
a
n
d
n
o
r
m
al
r
eq
u
est
s
.
R
a
n
d
o
m
Fo
r
est
is
a
n
e
n
s
e
m
b
le
clas
s
i
f
icatio
n
m
e
th
o
d
[
23
]
.
T
h
is
al
g
o
r
ith
m
i
s
b
ased
o
n
an
en
s
e
m
b
le
o
f
cl
ass
i
f
ier
s
,
w
h
ich
n
o
r
m
all
y
ar
e
Dec
is
io
n
T
r
ee
s
to
m
a
k
e
th
e
f
i
n
al
p
r
ed
ictio
n
.
T
h
e
th
eo
r
etica
l
f
o
u
n
d
atio
n
o
f
th
i
s
alg
o
r
it
h
m
is
b
ased
o
n
J
en
s
en
'
s
i
n
eq
u
al
it
y
[
24
]
.
A
c
co
r
d
in
g
to
J
en
s
en
's
in
eq
u
ali
t
y
ap
p
lied
to
th
e
c
las
s
if
ica
tio
n
p
r
o
b
le
m
s
,
it
i
s
s
h
o
w
n
t
h
at
th
e
co
m
b
in
a
tio
n
o
f
m
an
y
m
o
d
els
m
a
y
p
r
o
d
u
ce
less
er
r
o
r
r
ate
th
an
th
at
o
f
ea
ch
i
n
d
iv
id
u
al
m
o
d
el.
4
.2
.
DWAP
a
na
ly
s
is
f
o
r
co
ns
t
ruc
t
ing
t
ra
ini
ng
da
t
a
s
et
s
T
h
e
m
ain
c
h
ar
ac
ter
is
t
ic
o
f
t
h
e
ab
n
o
r
m
al
r
eq
u
es
t
d
etec
tio
n
m
et
h
o
d
u
s
i
n
g
DW
A
P
an
al
y
s
i
s
m
et
h
o
d
is
th
at
it
d
o
es
n
o
t
u
s
e
ex
i
s
ti
n
g
d
atasets
f
o
r
tr
ain
in
g
d
ata,
b
u
t
it
u
tili
ze
s
t
h
e
d
ata
o
f
th
e
d
ep
lo
y
ed
w
eb
s
ite.
I
n
f
ac
t,
th
e
n
u
m
b
er
o
f
a
n
o
m
al
y
r
eq
u
est
s
is
m
u
ch
s
m
aller
t
h
an
th
at
o
f
n
o
r
m
a
l
r
eq
u
est
s
in
t
h
e
w
h
o
le
d
ata
s
et.
As
a
r
es
u
lt,
it
i
s
n
ec
e
s
s
ar
y
to
h
av
e
s
u
itab
le
s
a
m
p
li
n
g
m
et
h
o
d
s
an
d
tec
h
n
iq
u
es
to
cr
ea
te
a
g
o
o
d
tr
ain
in
g
d
ata
s
et
th
at
h
e
lp
s
th
ea
b
n
o
r
m
al
r
eq
u
e
s
t
d
etec
tio
n
p
r
o
ce
s
s
b
ec
o
m
e
m
o
r
e
e
f
f
ec
tiv
e.
Fro
m
th
is
p
o
in
t,
a
n
e
w
s
a
m
p
li
n
g
m
et
h
o
d
b
ased
-
o
n
DW
A
P
an
al
y
s
i
s
tech
n
iq
u
e
an
d
u
n
s
u
p
er
v
is
ed
lear
n
in
g
al
g
o
r
ith
m
is
p
r
o
p
o
s
ed
.
T
h
is
m
eth
o
d
f
ir
s
tl
y
d
iv
id
es
t
h
e
d
ata
in
to
d
i
f
f
er
en
t
clu
s
ter
s
.
T
h
en
,
it
s
elec
ts
r
eq
u
ests
f
r
o
m
t
h
e
n
e
w
l
y
d
i
v
id
ed
d
ata
clu
s
ter
s
.
T
h
e
co
m
b
i
n
atio
n
o
f
t
h
e
DW
AP
an
al
y
s
is
tec
h
n
iq
u
e
a
n
d
t
h
e
u
n
s
u
p
er
v
is
ed
lear
n
i
n
g
alg
o
r
it
h
m
s
n
o
t
o
n
l
y
e
n
s
u
r
es
th
e
r
an
d
o
m
n
e
s
s
o
f
s
a
m
p
li
n
g
,
b
u
t
also
i
n
cr
ea
s
es
th
e
r
ate
o
f
ab
n
o
r
m
al
r
eq
u
est
s
t
h
at
ap
p
ea
r
in
t
h
e
s
a
m
p
le
d
ata.
C
o
n
s
eq
u
en
tl
y
,
t
h
i
s
h
e
lp
s
g
e
n
e
r
ate
a
m
o
r
e
b
alan
ce
d
tr
ain
in
g
d
ataset,
an
d
r
ed
u
ce
ti
m
e
a
n
d
ef
f
o
r
t
to
s
ea
r
ch
f
o
r
ab
n
o
r
m
al
r
eq
u
est
s
.
T
h
e
p
r
o
p
o
s
ed
m
et
h
o
d
ca
n
b
e
s
u
m
m
ar
ize
d
as f
o
llo
w
s
:
Step
1
:
Data
clu
s
ter
in
g
:
t
h
i
s
s
tep
ai
m
s
at
a
g
g
r
e
g
ati
n
g
r
e
q
u
ests
th
at
h
a
v
e
s
i
m
ilar
c
h
ar
ac
ter
is
tics
.
Data
C
lu
s
ter
i
n
g
is
k
n
o
w
n
as
a
m
e
th
o
d
to
g
ath
er
co
r
r
elate
d
o
b
s
er
v
atio
n
s
in
to
s
ep
ar
ate
g
r
o
u
p
s
.
T
h
is
m
et
h
o
d
h
as
b
ee
n
ad
o
p
ted
b
y
R
i
y
az
[
2
5
]
f
o
r
d
ep
lo
y
m
e
n
t
o
n
lar
g
e
d
atab
ases
a
n
d
h
a
s
s
h
o
w
n
t
h
at
p
r
ac
tical
ap
p
licatio
n
s
of
th
ese
clu
s
ter
i
n
g
al
g
o
r
ith
m
s
ar
e
p
r
o
m
is
s
i
n
g
.
Si
n
ce
th
e
f
ea
tu
r
es
ar
e
ex
tr
ac
ted
s
u
ch
t
h
at
t
h
e
y
ca
n
d
is
t
in
g
u
is
h
b
et
w
ee
n
n
o
r
m
al
a
n
d
ab
n
o
r
m
al
r
eq
u
ests
,
t
h
e
cl
u
s
ter
in
g
p
r
o
ce
s
s
o
f
t
h
ese
f
ea
t
u
r
es
n
o
t o
n
l
y
s
e
p
ar
ates n
o
r
m
al
a
n
d
ab
n
o
r
m
al
r
eq
u
est
s
,
b
u
t
also
c
lass
i
f
ies
t
h
e
attac
k
r
eq
u
est
s
i
n
d
if
f
er
en
t
f
o
r
m
s
.
T
h
e
r
em
ai
n
in
g
i
s
s
u
e
is
to
f
in
d
th
e
o
p
ti
m
u
m
n
u
m
b
er
o
f
cl
u
s
te
r
s
f
o
r
th
e
d
ata.
I
n
t
h
is
p
ap
er
,
K
-
m
ea
n
al
g
o
r
ith
m
is
ad
o
p
ted
f
o
r
clu
s
ter
i
n
g
tas
k
.
T
h
is
clu
s
ter
in
g
m
eth
o
d
is
b
as
ed
o
n
th
e
m
in
i
m
izatio
n
o
f
th
e
d
is
tan
ce
s
f
r
o
m
a
ll
d
ata
p
o
in
ts
w
it
h
i
n
ea
c
h
clu
s
te
r
to
th
e
c
lu
s
ter
ce
n
tr
o
id
[
26
]
.
I
n
o
r
d
er
to
f
in
d
th
e
n
u
m
b
er
o
f
t
h
e
cl
u
s
ter
s
f
o
r
th
e
K
-
m
ea
n
al
g
o
r
ith
m
,
t
h
e
E
lb
o
w
m
et
h
o
d
is
u
s
ed
.
T
h
is
m
et
h
o
d
is
b
ased
o
n
t
h
e
g
r
ap
h
p
r
esen
t
in
g
t
h
e
co
r
r
elatio
n
b
et
w
ee
n
th
e
t
o
tal
d
is
tan
ce
s
f
r
o
m
all
d
ata
p
o
in
ts
in
ea
c
h
clu
s
ter
to
th
eir
clu
s
ter
ce
n
tr
o
id
an
d
th
e
n
u
m
b
er
o
f
clu
s
ter
s
.
T
h
e
E
lb
o
w
cr
iter
io
n
is
m
et
w
h
e
n
t
h
e
n
u
m
b
er
o
f
cl
u
s
ter
s
N
is
c
h
o
s
e
n
s
u
c
h
t
h
at
t
h
e
r
atio
b
et
w
ee
n
t
h
e
to
tal
d
is
ta
n
ce
w
it
h
N
g
r
o
u
p
s
a
n
d
th
a
t
o
f
N
+1
g
r
o
u
p
s
i
s
s
m
alle
s
t.
T
h
e
E
lb
o
w
m
et
h
o
d
is
s
u
m
m
ar
ize
d
as f
o
llo
w
s
:
L
et
Δ
S
S
E
i
is
th
e
to
tal
s
u
m
o
f
s
q
u
ar
ed
er
r
o
r
d
is
tan
ce
s
o
f
i
cl
u
s
ter
s
L
et
r
i
i
s
th
e
r
atio
b
et
w
ee
n
Δ
S
S
E
i
an
d
Δ
S
S
E
i+
1
=
∆
+
1
∆
(
1
)
T
h
e
o
p
tim
al
n
u
m
b
er
o
f
clu
s
ter
N
co
r
r
esp
o
n
d
s
to
th
e
s
m
a
lles
t
r
i
:
=
{
}
(
2
)
Step
2
:
Sa
m
p
lin
g
d
ata
f
r
o
m
c
l
u
s
ter
s
: th
e
p
r
o
ce
s
s
to
tak
e
M
s
a
m
p
les
f
r
o
m
N
g
r
o
u
p
s
:
I
f
th
e
n
u
m
b
er
o
f
s
a
m
p
le
s
i
n
o
n
e
p
ar
ticu
lar
cl
u
s
ter
i
s
s
m
alle
r
th
an
,
th
e
n
all
s
a
m
p
les
in
t
h
at
cl
u
s
ter
ar
e
s
elec
ted
.
T
h
e
r
ea
s
o
n
f
o
r
t
h
is
i
s
t
h
at
t
h
e
n
u
m
b
er
o
f
ab
n
o
r
m
a
l
r
eq
u
est
s
i
s
v
er
y
s
m
all
co
m
p
ar
ed
to
n
o
r
m
al
o
n
es,
an
d
d
u
e
to
a
n
o
m
al
y
c
h
ar
ac
ter
is
tics
,
t
h
e
y
ar
e
u
s
u
all
y
n
o
t
in
t
h
e
s
a
m
e
ca
te
g
o
r
y
a
s
n
o
r
m
al
r
eq
u
e
s
ts
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
10
,
No
.
5
,
Octo
b
e
r
2
0
2
0
:
5
3
3
5
-
5346
5342
As
a
r
es
u
lt,
ab
n
o
r
m
a
l
r
eq
u
est
s
ten
d
to
b
e
s
ep
ar
ated
in
s
m
al
l
clu
s
ter
s
.
So
a
f
ter
th
i
s
p
r
o
ce
s
s
,
th
e
r
e
m
ai
n
in
g
n
u
m
b
er
o
f
s
a
m
p
les
n
ee
d
to
b
e
tak
en
i
s
M
1
a
n
d
th
e
r
e
m
a
in
i
n
g
n
u
m
b
er
o
f
clu
s
ter
s
i
s
N
1
.
R
ep
ea
t
(
i)
w
ith
th
e
n
u
m
b
er
o
f
s
a
m
p
les
n
ee
d
to
b
e
tak
e
n
as
M
1
an
d
cl
u
s
ter
n
u
m
b
er
i
s
N
1
.
T
h
e
s
a
m
p
lin
g
p
r
o
ce
s
s
w
ill
en
d
a
f
ter
i
i
ter
ati
o
n
s
w
h
e
n
t
h
e
n
u
m
b
er
s
o
f
s
a
m
p
les
i
n
all
r
e
m
a
in
i
n
g
cl
u
s
ter
s
ar
e
g
r
ea
ter
t
h
a
n
th
e
r
atio
.
Fro
m
ea
ch
o
f
r
e
m
ai
n
in
g
cl
u
s
te
r
s
,
s
a
m
p
les ar
e
r
an
d
o
m
l
y
s
ele
cted
.
T
h
e
w
h
o
le
p
r
o
ce
s
s
is
p
r
ese
n
te
d
in
Alg
o
r
it
h
m
4
:
Algorithm
4
. Sampling data
Input: clustered_data
Output: sampled data
1: function DATA_SAMPLING (clustered_data, M)
2: sampled_data []
<
-
empty list
3: clusters [] <
-
list of cluster in clustered_data
4: number_of_data_in_cluster [] <
-
number of data in each cluster
5: N <
-
number of cluster
6: while existing clusters[i] which has number of data is smaller than M/N:
7: sampled_data
= sampled_data + data in clusters[i]
8: M = M
–
number_of_data_in_cluster[i]
9: N = N
-
1
10: clusters pop i
11: for cluster in clusers:
12: sampled_data <
-
sampled_data
+ random choice M/N data in cluster
13: return sampled_data
Dis
cu
s
s
io
n
:
I
f
th
e
r
ate
o
f
t
h
e
ab
n
o
r
m
al
r
eq
u
est,
K
1
,
i
n
o
n
e
d
ataset
is
v
er
y
s
m
all,
i.e
.
K
1
<<
1
,
th
e
n
a
m
o
n
g
M
s
elec
ted
s
a
m
p
les,
t
h
e
r
ate
o
f
a
n
o
m
al
y
r
eq
u
es
t
is
s
till
K
1
.
Mo
r
eo
v
er
,
an
o
m
al
y
r
eq
u
ests
ar
e
u
s
u
all
y
s
ep
ar
ated
f
r
o
m
n
o
r
m
a
l
r
eq
u
e
s
ts
a
f
ter
th
e
clu
s
ter
i
n
g
m
eth
o
d
.
A
lt
h
o
u
g
h
t
h
er
e
i
s
n
o
g
u
ar
an
tee
t
h
at
all
d
ata
s
a
m
p
les
i
n
ea
ch
cl
u
s
ter
h
av
e
t
h
e
s
a
m
e
lab
el
b
u
t
if
K
1
<<
1
,
th
er
e
is
a
g
r
ea
t
ch
a
n
ce
th
at
t
h
e
al
m
o
s
t
all
n
u
m
b
er
o
f
an
o
m
al
y
r
eq
u
es
ts
ar
e
s
ele
cted
f
r
o
m
s
m
all
clu
s
ter
s
.
A
s
a
r
esu
lt,
th
e
cl
u
s
ter
i
n
g
m
et
h
o
d
co
m
b
i
n
ed
w
it
h
s
a
m
p
li
n
g
al
g
o
r
ith
m
p
r
o
p
o
s
ed
in
t
h
i
s
p
ap
er
ca
n
ef
f
icie
n
tl
y
f
i
lter
o
u
t
a
l
m
o
s
t
all
ab
n
o
r
m
a
l
r
eq
u
ests
,
w
h
ich
ca
n
h
elp
r
ed
u
ce
t
h
e
b
u
ild
i
n
g
ti
m
e
o
f
t
h
e
d
ata
s
e
t
f
o
r
t
h
e
DW
A
P
a
n
al
y
s
is
.
T
h
is
s
a
m
p
li
n
g
p
r
o
ce
s
s
h
as
m
o
r
e
ad
v
an
ta
g
es t
h
a
n
r
an
d
o
m
s
a
m
p
lin
g
ap
p
r
o
ac
h
.
5.
E
XP
E
R
I
M
E
NT
5
.
1
.
Da
t
a
s
et
I
n
o
r
d
er
to
ev
alu
ate
th
e
ef
f
icie
n
c
y
o
f
th
e
p
r
o
p
o
s
ed
alg
o
r
ith
m
s
,
t
w
o
2
d
atasets
ar
e
u
s
ed
.
-
Data
s
et
1
:
T
h
e
f
ir
s
t
d
ataset
is
C
SIC
2
0
1
0
[
7
]
,
w
h
ic
h
i
s
d
ev
e
lo
p
ed
b
y
C
ar
m
e
n
.
T
h
e
d
atase
t
in
cl
u
d
es
ab
o
u
t
3
6
0
0
0
n
o
r
m
al
r
eq
u
e
s
ts
a
n
d
2
5
0
0
0
ab
n
o
r
m
al
r
eq
u
est
s
.
Si
n
ce
m
o
s
t
o
f
th
e
s
a
m
p
le
s
i
n
th
e
C
S
I
C
2
0
1
0
d
ataset
ar
e
attac
k
i
n
g
r
eq
u
e
s
ts
,
i
t
m
a
y
n
o
t
b
e
s
u
i
tab
le
f
o
r
ev
al
u
lat
in
g
th
e
d
etec
tio
n
o
f
t
h
e
ab
n
o
r
m
al
r
eq
u
ests
.
C
SI
C
2
0
1
0
d
ataset
is
f
ilter
ed
an
d
d
i
v
id
ed
in
to
8
m
ai
n
UR
I
g
r
o
u
p
s
as p
r
esen
ted
in
T
ab
le
3
.
T
ab
le
3
.
Statis
tics
o
f
t
h
e
n
u
m
b
er
o
f
n
o
r
m
al
r
eq
u
est
s
an
d
ab
n
o
r
m
al
r
eq
u
e
s
ts
i
n
t
h
e
d
ataset
1
O
r
d
e
r
U
R
I
N
o
r
mal
A
b
n
o
r
mal
1
/
t
i
e
n
d
a
1
/
p
u
b
l
i
c
o
/
a
n
a
d
i
r
.
j
s
p
4
0
0
0
2
8
2
1
2
/
t
i
e
n
d
a
1
/
p
u
b
l
i
c
o
/
a
u
t
e
n
t
i
c
a
r
.
j
s
p
4
0
0
0
2
7
8
3
3
/
t
i
e
n
d
a
1
/
p
u
b
l
i
c
o
/
c
a
r
a
c
t
e
r
i
st
i
c
a
s.j
sp
4
0
0
0
1
9
5
7
4
/
t
i
e
n
d
a
1
/
p
u
b
l
i
c
o
/
e
n
t
r
a
r
.
j
sp
4
0
0
0
1
8
3
5
5
/
t
i
e
n
d
a
1
/
p
u
b
l
i
c
o
/
p
a
g
a
r
.
j
s
p
4
0
0
0
2
7
2
2
6
/
t
i
e
n
d
a
1
/
p
u
b
l
i
c
o
/
r
e
g
i
st
r
o
.
j
s
p
4
0
0
0
2
7
8
1
7
/
t
i
e
n
d
a
1
/
p
u
b
l
i
c
o
/
v
a
c
i
a
r
.
j
s
p
4
0
0
0
1
8
8
4
8
/
t
i
e
n
d
a
1
/
mi
e
m
b
r
o
s/
e
d
i
t
a
r
.
j
sp
4
0
0
0
2
7
7
4
Data
s
et
2
:
T
h
e
s
ec
o
n
d
d
ataset
is
m
ad
e
b
y
u
s
i
n
g
s
o
m
e
s
ec
u
r
it
y
to
o
ls
li
k
e
A
c
u
n
et
ix
,
B
u
r
p
Su
ite,
S
QL
Ma
p
to
s
ca
n
t
h
e
v
u
l
n
er
ab
ilit
ie
s
f
r
o
m
o
u
r
p
r
o
to
ty
p
e
w
eb
s
i
tes.
T
h
o
s
e
s
ca
n
n
in
g
to
o
ls
s
ea
r
ch
an
d
e
x
p
lo
it
v
u
l
n
er
ab
ilit
ie
s
in
b
o
th
t
h
e
q
u
er
y
a
n
d
th
e
r
eq
u
est
’
s
h
ea
d
er
s
.
T
h
e
co
llect
ed
d
ata
is
class
if
ied
f
o
llo
w
i
n
g
th
e
d
ef
i
n
ed
cr
iter
ia
in
t
h
e
p
r
ev
io
u
s
s
ec
tio
n
.
B
esid
es
th
e
ab
n
o
r
m
al
r
eq
u
e
s
ts
co
llected
b
y
s
ca
n
n
i
n
g
to
o
ls
,
w
e
m
ad
e
s
o
m
e
n
o
r
m
al
r
eq
u
es
ts
b
y
n
o
r
m
all
y
o
p
er
ate
o
n
t
h
e
s
a
m
e
UR
I
s
.
E
ac
h
U
R
I
co
n
tai
n
s
5
0
0
0
n
o
r
m
al
r
eq
u
ests
a
n
d
5
0
0
0
a
b
n
o
r
m
al
r
e
q
u
ests
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
A
n
a
d
a
p
tive
a
n
o
ma
ly
r
eq
u
est
d
etec
tio
n
fr
a
mewo
r
k
b
a
s
ed
…
(
C
h
o
Do
X
u
a
n
)
5343
5
.2
.
Cla
s
s
if
ica
t
io
n
m
ea
s
ure
s
5
.
2
.
1
.
E
v
a
lua
t
io
n c
rit
er
ia
t
o
det
ec
t
a
bn
o
r
m
a
l r
eque
s
t
I
n
th
i
s
r
esear
ch
,
t
h
r
ee
ev
al
u
ati
o
n
m
etr
ics ar
e
u
s
ed
as f
o
llo
w
s
:
P
r
ec
is
io
n
is
d
ef
i
n
ed
as
th
e
r
a
tio
b
et
w
ee
n
t
h
e
n
u
m
b
er
o
f
tr
u
e
p
o
s
itiv
e
alar
m
s
(
TP
)
an
d
all
th
e
s
a
m
p
les
class
i
f
ied
as
p
o
s
iti
v
e
(
TP
+
FP
)
.
T
h
e
h
ig
h
er
t
h
e
p
r
ec
is
io
n
s
co
r
e,
th
e
m
o
r
e
n
u
m
b
er
o
f
p
o
s
itiv
e
alar
m
s
ar
e
co
r
r
ec
t.
=
+
(
3
)
R
ec
all
is
d
e
f
in
ed
as t
h
e
r
atio
o
f
tr
u
e
p
o
s
iti
v
e
alar
m
s
a
m
o
n
g
all
s
a
m
p
le
s
th
at
ar
e
ac
t
u
all
y
p
o
s
iti
v
e.
=
+
(
4
)
F1
-
s
co
r
e
is
t
h
e
h
ar
m
o
n
ic
m
ea
n
o
f
p
r
ec
is
io
n
a
n
d
r
ec
all.
1
=
2
∗
∗
+
(
5
)
w
h
er
e,
TP
-
T
r
u
e
p
o
s
iti
v
e:
i
s
th
e
n
u
m
b
er
o
f
r
ec
o
r
d
s
th
at
ar
e
co
r
r
ec
tly
lab
eled
as
“
ab
n
o
r
m
al
r
eq
u
e
s
ts
”;
F
N
-
Fal
s
e
n
e
g
ati
v
e:
i
s
th
e
n
u
m
b
er
o
f
r
ec
o
r
d
s
th
at
ar
e
ac
t
u
all
y
“
ab
n
o
r
m
al
r
eq
u
e
s
ts
”
b
u
t
ar
e
class
if
ied
as
“n
o
r
m
al
r
eq
u
est
s
”;
TN
-
T
r
u
e
n
e
g
ati
v
e:
i
s
t
h
e
n
u
m
b
er
o
f
r
ec
o
r
d
s
th
at
ar
e
co
r
r
ec
tl
y
l
ab
eled
as
“
n
o
r
m
al
r
eq
u
ests
”;
FP
-
Fals
e
p
o
s
iti
v
e:
i
s
th
e
n
u
m
b
er
o
f
r
ec
o
r
d
s
t
h
at
ar
e
ac
tu
a
ll
y
“
n
o
r
m
al
r
eq
u
ests
”
b
u
t
ar
e
m
is
c
lass
if
ied
to
“
ab
n
o
r
m
al
r
eq
u
est
s
”.
5
.
2
.
2
.
Crit
er
ia
f
o
r
ev
a
lua
t
ing
t
he
ef
f
ec
t
iv
eness
o
f
a
pp
ly
i
ng
DWAP
a
na
ly
s
i
s
f
o
r
co
n
s
t
ruct
ing
t
ra
ini
ng
da
t
a
s
et
I
n
o
r
d
er
to
ev
alu
ate
th
e
e
f
f
ici
en
c
y
o
f
t
h
e
s
a
m
p
li
n
g
m
et
h
o
d
in
th
e
co
n
s
tr
u
c
tio
n
o
f
t
h
e
tr
ai
n
in
g
d
ata,
th
e
i
m
b
ala
n
ce
i
n
s
a
m
p
lin
g
p
r
o
ce
s
s
b
et
w
ee
n
t
h
e
r
a
n
d
o
m
s
a
m
p
lin
g
m
e
th
o
d
a
n
d
t
h
e
n
e
w
l
y
p
r
o
p
o
s
ed
s
a
m
p
lin
g
m
et
h
o
d
.
T
h
is
v
alu
e
i
s
ex
p
r
ess
ed
b
y
p
ar
a
m
eter
s
K
1
an
d
K
2
a
s
f
o
llo
w
s
:
T
h
e
p
ar
am
eter
K
1
r
ep
r
esen
ts
th
e
p
r
o
p
o
r
tio
n
o
f
ab
n
o
r
m
al
r
eq
u
ests
in
t
h
e
d
ata
r
ec
o
g
n
i
ze
d
b
y
r
an
d
o
m
s
a
m
p
li
n
g
m
et
h
o
d
s
.
K
2
is
t
h
e
r
atio
o
f
ab
n
o
r
m
al
r
eq
u
est
s
in
t
h
i
s
p
r
o
p
o
s
e
d
s
am
p
li
n
g
m
et
h
o
d
.
I
n
th
i
s
p
ap
er
,
th
ese
t
w
o
v
al
u
e
s
ar
e
co
m
p
ar
ed
w
it
h
r
esp
ec
t
t
o
d
if
f
er
en
t
K
1
v
al
u
es
a
n
d
th
e
n
u
m
b
er
o
f
s
elec
ted
n
u
m
b
er
o
f
s
a
m
p
les
M
.
5
.3
.
E
x
peri
m
e
nta
l
re
s
ults a
nd
co
mm
e
nts
5
.
3
.
1
.
Crit
er
ia
ex
peri
m
ent
a
l scena
rio
s
a
nd
ex
peri
m
e
nta
l r
esu
lt
s
f
o
r
det
ec
t
ing
a
bn
o
r
m
a
l r
eques
t
a.
E
x
p
er
i
m
e
n
ta
l sce
n
ar
io
s
T
h
e
ef
f
icie
n
c
y
o
f
DW
A
P
an
al
y
s
i
s
tech
n
iq
u
e
s
in
d
etec
tin
g
ab
n
o
r
m
al
r
eq
u
ests
u
s
i
n
g
R
an
d
o
m
Fo
r
est
clu
s
ter
i
n
g
al
g
o
r
ith
m
is
ev
al
u
ated
b
ased
o
n
b
o
th
d
ataset
s
d
escr
ib
ed
in
s
ec
tio
n
5
.
1
.
A
ll
th
r
ee
p
er
f
o
r
m
an
c
e
me
tr
ics
ar
e
r
ec
o
r
ed
.
E
ac
h
d
ata
s
et
i
s
d
i
v
id
ed
in
to
t
w
o
s
u
b
s
et
s
:
th
e
tr
ai
n
in
g
d
ata
co
n
tai
n
i
n
g
8
0
%
o
f
t
h
e
d
atase
t
is
u
s
ed
f
o
r
tr
ain
i
n
g
th
e
c
lass
i
f
icatio
n
m
o
d
el;
r
e
m
a
in
i
n
g
2
0
%
o
f
t
h
e
d
ata
is
u
s
ed
f
o
r
test
in
g
.
T
h
e
n
u
m
b
er
o
f
tr
ee
s
f
o
r
R
an
d
o
m
Fo
r
est alg
o
r
i
th
m
is
s
e
t a
t 3
0
0
.
b.
E
x
p
er
i
m
e
n
tal
r
es
u
lts
a
n
d
co
m
m
en
ts
E
x
p
er
i
m
e
n
tal
r
es
u
lt
s
o
f
u
s
i
n
g
DW
A
P
a
n
al
y
s
is
tech
n
iq
u
e
to
d
etec
t
ab
n
o
r
m
al
r
eq
u
e
s
ts
o
n
d
atasets
1
an
d
2
ar
e
s
h
o
w
n
i
n
T
ab
les
4
an
d
5
.
T
h
e
r
esu
lts
in
T
ab
le
4
s
h
o
w
t
h
at
u
s
i
n
g
DW
A
P
an
al
y
s
is
tec
h
n
iq
u
es
ca
n
ac
cu
r
atel
y
a
n
d
ef
f
ic
ien
t
l
y
d
ete
ct
ab
n
o
r
m
al
r
eq
u
es
ts
.
I
n
p
ar
ticu
lar
,
P
r
ec
is
io
n
s
co
r
es
ac
r
o
s
s
all
d
ata
r
an
g
e
f
r
o
m
9
9
.
4
6
% to
1
0
0
%.
T
h
is
r
esu
lt s
h
o
w
s
t
h
at
th
e
p
o
s
iti
v
e
alar
m
o
f
th
is
m
eth
o
d
is
v
er
y
r
eliab
le.
T
ab
le
5
s
h
o
w
s
t
h
at
ev
e
n
wh
en
t
h
e
d
ataset
co
n
ta
in
s
a
h
ig
h
er
p
er
ce
n
ta
g
e
o
f
n
o
r
m
al
r
eq
u
est
a
s
illu
s
tr
ated
i
n
d
ataset
2
t
h
e
n
e
w
DW
A
P
an
al
y
s
i
s
tech
n
iq
u
e
is
s
till
h
i
g
h
l
y
e
f
f
ec
tiv
e,
w
h
ile
tr
ad
itio
n
al
to
o
ls
et
u
s
i
n
g
Mo
d
Secu
r
it
y
r
u
les
ar
e
n
o
t
ef
f
icie
n
t.
Sp
ec
if
icall
y
,
r
ec
al
l
v
alu
e
o
f
t
h
e
to
o
ls
et
is
j
u
s
t
3
0
%
w
h
ile
t
h
at
s
co
r
e
o
f
th
e
p
r
o
p
o
s
ed
m
et
h
o
d
is
m
o
r
e
th
an
9
0
%.
B
esid
es,
th
e
n
e
w
m
et
h
o
d
ca
n
o
b
tain
p
er
f
ec
t
p
r
ec
is
io
n
s
co
r
e
o
n
all
UR
I
s
et
s
.
T
h
e
F1
s
co
r
e
o
f
t
h
e
p
r
o
p
o
s
ed
m
e
th
o
d
is
also
m
u
c
h
h
i
g
h
er
,
w
h
ic
h
i
s
o
v
er
9
5
%,
co
m
p
ar
ed
to
th
e
to
o
ls
et.
T
h
e
r
esu
lts
s
h
o
w
n
in
tab
les
4
an
d
5
d
e
m
o
n
s
tr
ate
th
at
DW
A
P
an
al
y
s
is
tec
h
n
iq
u
es
ar
e
n
o
t
o
n
l
y
ab
le
to
ef
f
icie
n
tl
y
d
etec
t r
eq
u
est
s
at
tack
s
,
b
u
t t
h
e
y
ar
e
also
ca
p
ab
le
o
f
co
r
r
ec
tly
d
etec
ti
n
g
ab
n
o
r
m
al
r
eq
u
est
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
10
,
No
.
5
,
Octo
b
e
r
2
0
2
0
:
5
3
3
5
-
5346
5344
T
ab
le
4
.
E
x
p
er
im
e
n
tal
r
esu
lts
o
f
ab
n
o
r
m
al
r
eq
u
est
s
d
etec
tio
n
u
s
in
g
DW
A
P
an
al
y
s
i
s
m
et
h
o
d
o
n
d
ataset
1
O
r
d
e
r
U
R
I
P
r
e
c
i
si
o
n
R
e
c
a
l
l
F
1
S
c
o
r
e
1
/
t
i
e
n
d
a
1
/
p
u
b
l
i
c
o
/
a
n
a
d
i
r
.
j
s
p
1
0
0
%
1
0
0
%
1
0
0
%
2
/
t
i
e
n
d
a
1
/
p
u
b
l
i
c
o
/
a
u
t
e
n
t
i
c
a
r
.
j
s
p
9
9
.
4
6
%
9
9
.
2
8
%
9
9
.
3
7
%
3
/
t
i
e
n
d
a
1
/
p
u
b
l
i
c
o
/
c
a
r
a
c
t
e
r
i
st
i
c
a
s.j
sp
1
0
0
%
1
0
0
%
1
0
0
%
4
/
t
i
e
n
d
a
1
/
p
u
b
l
i
c
o
/
e
n
t
r
a
r
.
j
sp
1
0
0
%
1
0
0
%
1
0
0
%
5
/
t
i
e
n
d
a
1
/
p
u
b
l
i
c
o
/
p
a
g
a
r
.
j
s
p
1
0
0
%
1
0
0
%
1
0
0
%
6
/
t
i
e
n
d
a
1
/
p
u
b
l
i
c
o
/
r
e
g
i
st
r
o
.
j
s
p
9
9
.
6
3
%
9
5
.
3
3
%
9
7
.
9
4
%
7
/
t
i
e
n
d
a
1
/
p
u
b
l
i
c
o
/
v
a
c
i
a
r
.
j
s
p
1
0
0
%
1
0
0
%
1
0
0
%
8
/
t
i
e
n
d
a
1
/
mi
e
m
b
r
o
s/
e
d
i
t
a
r
.
j
sp
9
9
.
6
3
%
9
6
.
0
4
%
9
8
.
2
3
%
T
ab
le
5
.
A
n
o
m
al
y
d
etec
tio
n
p
er
f
o
r
m
an
ce
co
m
p
ar
i
s
o
n
b
et
w
e
en
DW
A
P
an
al
y
s
is
m
et
h
o
d
s
an
d
Mo
d
Secu
r
it
y
to
o
l a
p
p
lied
o
n
d
ataset
2
O
r
d
e
r
U
R
I
D
W
A
P
a
n
a
l
y
si
s +
R
a
n
d
o
m F
o
r
e
st
M
o
d
S
e
c
u
r
i
t
y
[
8
]
P
r
e
c
i
si
o
n
R
e
c
a
l
l
F
1
S
c
o
r
e
P
r
e
c
i
si
o
n
R
e
c
a
l
l
F
1
S
c
o
r
e
1
/
si
g
n
i
n
1
0
0
%
9
6
.
7
7
%
9
8
.
3
6
%
9
8
%
3
0
%
4
6
%
2
/
si
g
n
u
p
1
0
0
%
9
1
.
4
1
%
9
5
.
5
1
%
9
9
%
4
0
%
5
7
%
5
.
3
.
2
.
E
x
peri
m
ent
a
l set
up
t
o
ev
a
lua
t
e
t
he
co
ns
t
ruct
io
n o
f
t
ra
ini
ng
da
t
a
us
ing
DWAP
a
na
ly
s
is
m
et
ho
d
a.
E
x
p
er
i
m
e
n
tal
s
et
u
p
T
h
e
ef
f
ec
tiv
e
n
es
s
o
f
t
h
e
d
at
a
co
n
s
tr
u
ctio
n
p
r
o
ce
s
s
i
s
e
v
alu
ated
b
y
u
s
i
n
g
th
e
ch
a
n
g
i
n
g
h
y
p
er
p
ar
am
eter
s
as f
o
llo
w
s
:
Dis
tr
ib
u
tio
n
r
ate
o
f
a
n
o
m
al
y
r
eq
u
est
s
i
n
d
ata
:
d
if
f
er
en
t
v
alu
e
s
o
f
K
1
r
a
n
g
in
g
f
r
o
m
1
%
to
3
0
%
o
n
th
e
U
R
I
s
o
f
th
e
C
SIC
2
0
1
0
d
ataset
i
s
u
s
ed
,
b
ased
o
n
w
h
ic
h
th
e
ev
al
u
atio
n
o
f
K
2
ca
n
b
e
o
b
tain
ed
t
h
r
o
u
g
h
th
e
p
r
o
p
o
s
ed
s
am
p
li
n
g
p
r
o
ce
s
s
.
T
h
e
f
ea
tu
r
es e
x
tr
ac
ted
in
s
e
ctio
n
3
.
2
ar
e
u
s
ed
to
r
ep
r
esen
t
all
d
ata
p
o
in
ts
.
Nu
m
b
er
o
f
s
a
m
p
les
M
:
Si
n
c
e
th
e
v
alu
e
o
f
K
2
d
ep
en
d
s
m
ain
l
y
o
n
t
h
e
n
u
m
b
er
o
f
M
s
e
lecte
d
s
a
m
p
le
s
,
th
e
in
f
l
u
en
ce
a
n
d
d
ep
en
d
en
ce
o
f
K
2
w
h
e
n
th
e
v
alu
e
o
f
M
c
h
an
g
es
ca
n
b
e
u
s
ed
to
s
h
o
w
t
h
e
ef
f
icie
n
c
y
o
f
th
e
s
a
m
p
li
n
g
m
et
h
o
d
.
T
h
is
f
ac
to
r
p
lay
s
a
n
i
m
p
o
r
tan
t
r
o
le
in
s
a
m
p
lin
g
p
r
o
ce
s
s
s
u
ch
t
h
at
th
e
ab
n
o
r
m
a
l
r
eq
u
est d
is
tr
ib
u
tio
n
in
t
h
e
d
ata
s
et
is
o
p
ti
m
ized
.
Fig
u
r
e
3
p
r
esen
ts
th
e
clu
s
ter
i
n
g
r
esu
lt
s
b
ased
o
n
th
e
v
al
u
e
o
f
th
e
SS
E
o
f
t
h
e
clu
s
ter
s
.
Fro
m
F
i
g
u
r
e
3
it
ca
n
b
e
s
ee
n
th
at
th
e
SS
E
v
al
u
e
v
ar
ies
a
lo
t
w
h
en
N
=
2
an
d
N
=
3
,
r
esu
ltin
g
to
th
e
r
atio
s
r
2
an
d
r
3
ar
e
alm
o
s
t
eq
u
al
to
1
.
W
h
e
n
N
=
4
,
t
h
e
v
ar
iatio
n
o
f
SSE
d
ec
r
ea
s
es
s
i
g
n
i
f
ica
n
tl
y
,
s
o
is
t
h
e
v
al
u
e
o
f
r
4
.
SSE
r
et
u
r
n
s
to
a
litt
le
v
ar
iatio
n
w
h
en
N
>
4
r
esu
lti
n
g
to
r
5
an
d
r
6
ar
e
alm
o
s
t
eq
u
al
to
1
.
T
h
er
ef
o
r
e,
N
=
4
is
ch
o
s
en
as
th
e
clu
s
ter
n
u
m
b
er
o
f
t
h
e
d
ata.
Fig
u
r
e
4
illu
s
tr
ates
t
h
e
d
is
tr
ib
u
tio
n
o
f
n
o
r
m
al
r
eq
u
est
s
an
d
ab
n
o
r
m
al
r
eq
u
ests
i
n
ea
ch
cl
u
s
ter
af
ter
th
e
K
-
m
ea
n
alg
o
r
it
h
m
.
T
h
e
d
ata
is
cla
s
s
i
f
ied
i
n
to
4
cl
u
s
ter
s
an
d
t
h
e
d
is
tr
ib
u
tio
n
o
f
a
ll
l
ab
els
ar
e
s
h
o
w
n
o
n
th
e
g
r
ap
h
.
T
h
e
r
atio
o
f
ab
n
o
r
m
al
r
eq
u
e
s
ts
i
n
t
h
is
s
ce
n
ar
io
is
K
1
=
3
1
%.
W
h
e
n
ap
p
l
y
i
n
g
t
h
e
p
r
o
p
o
s
ed
s
a
m
p
lin
g
m
et
h
o
d
w
it
h
M
=
1
0
0
0
,
th
e
p
er
ce
n
tag
e
o
f
ab
n
o
r
m
al
r
eq
u
es
ts
in
th
e
s
a
m
p
led
d
ata
r
ea
ch
es
K
2
=
7
1
%.
Fig
u
r
e
3
.
E
x
a
m
p
le
o
f
cl
u
s
ter
n
u
m
b
er
s
elec
t
io
n
Fig
u
r
e
4
.
T
h
e
r
eq
u
est d
is
tr
ib
u
tio
n
o
f
a
UR
I
i
n
clu
s
ter
s
b.
E
x
p
er
i
m
e
n
tal
r
es
u
lts
a
n
d
co
m
m
en
ts
T
h
e
r
esu
lts
s
h
o
w
t
h
e
ef
f
ec
ti
v
e
n
es
s
o
f
th
e
p
r
o
p
o
s
ed
s
am
p
li
n
g
m
et
h
o
d
co
m
p
ar
ed
to
r
an
d
o
m
s
a
m
p
lin
g
m
et
h
o
d
.
A
d
d
itio
n
all
y
,
th
e
co
m
p
ar
i
s
o
n
b
et
w
ee
n
th
e
d
is
tr
ib
u
tio
n
o
f
a
n
o
m
al
y
r
eq
u
es
ts
i
n
th
e
d
ata
o
f
b
o
th
m
et
h
o
d
s
ar
e
also
r
ec
o
r
d
ed
.
T
a
b
le
6
s
h
o
w
s
t
h
e
c
h
an
g
e
in
K
2
v
alu
e
w
h
en
t
h
e
v
alu
e
o
f
K
1
v
a
r
ie
s
f
r
o
m
1
to
3
0
%.
T
h
e
r
esu
lt o
f
T
ab
le
6
s
h
o
w
s
t
h
at
K
2
is
g
r
ea
ter
th
a
n
K
1
i
n
d
if
f
er
en
t
K
1
d
is
tr
ib
u
tio
n
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.