I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
pu
t
er
E
ng
ineering
(
I
J
E
CE
)
Vo
l.
1
6
,
No
.
3
,
J
u
n
e
20
2
6
,
p
p
.
1608
~
1
6
2
5
I
SS
N:
2088
-
8
7
0
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijece.
v
1
6
i
3
.
pp
1
6
0
8
-
1
6
2
5
1608
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m
A
multi
-
mo
da
l f
r
a
mewo
rk f
o
r imp
ro
v
ing
t
he accura
cy
of
phish
ing
ema
il de
tect
io
n
L
a
m
ee
s
M
o
ha
m
ed
F
a
ra
j
1
,
S
a
y
ed
Abdel
-
G
a
ber
2
,
H
a
na
n F
a
hm
y
2
1
S
c
h
o
o
l
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
C
a
n
a
d
i
a
n
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
l
l
a
g
e
,
C
a
i
r
o
,
E
g
y
p
t
2
D
e
p
a
r
t
me
n
t
o
f
I
n
f
o
r
mat
i
o
n
S
y
s
t
e
ms
,
F
a
c
u
l
t
y
o
f
C
o
m
p
u
t
e
r
s a
n
d
A
r
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
,
C
a
p
i
t
a
l
U
n
i
v
e
r
si
t
y
,
H
e
l
w
a
n
,
E
g
y
p
t
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Dec
1
,
2
0
2
5
R
ev
is
ed
Feb
1
9
,
2
0
2
6
Acc
ep
ted
Ap
r
2
6
,
2
0
2
6
P
h
ish
i
n
g
e
m
a
il
s
c
o
n
t
in
u
e
to
p
o
se
a
si
g
n
ifi
c
a
n
t
c
y
b
e
rse
c
u
rit
y
t
h
re
a
t,
p
a
rti
c
u
larly
t
h
ro
u
g
h
t
h
e
i
n
c
re
a
sin
g
u
se
o
f
m
a
li
c
io
u
s
a
tt
a
c
h
m
e
n
ts
to
e
v
a
d
e
trad
it
io
n
a
l
te
x
t
-
b
a
se
d
d
e
tec
ti
o
n
sy
ste
m
s.
M
o
st
e
x
isti
n
g
a
p
p
ro
a
c
h
e
s
fo
c
u
s
p
rima
ril
y
o
n
e
m
a
il
c
o
n
ten
t,
c
r
e
a
ti
n
g
a
b
li
n
d
s
p
o
t
in
a
tt
a
c
h
m
e
n
t
-
a
wa
re
p
h
ish
in
g
d
e
tec
ti
o
n
.
Th
is
p
a
p
e
r
p
ro
p
o
se
s
a
m
u
lt
i
-
m
o
d
a
l
p
h
ish
in
g
e
m
a
il
c
las
sifica
ti
o
n
m
o
d
e
l
t
h
a
t
i
n
teg
ra
te
s
e
m
a
il
h
e
a
d
e
r
fe
a
tu
re
s,
b
o
d
y
te
x
t
a
n
a
ly
sis,
a
n
d
a
tt
a
c
h
m
e
n
t
in
sp
e
c
ti
o
n
wi
th
in
a
n
e
n
se
m
b
le
lea
rn
in
g
fr
a
m
e
wo
rk
.
In
d
e
p
e
n
d
e
n
t
m
a
c
h
i
n
e
lea
rn
in
g
c
las
sifiers
a
re
e
m
p
lo
y
e
d
f
o
r
e
a
c
h
e
m
a
il
c
o
m
p
o
n
e
n
t,
a
n
d
a
m
a
jo
r
it
y
v
o
ti
n
g
m
e
c
h
a
n
ism
is
u
se
d
to
d
e
term
in
e
th
e
fin
a
l
c
las
sifica
ti
o
n
d
e
c
isio
n
.
Th
e
p
ro
p
o
se
d
m
o
d
e
l
is
e
v
a
lu
a
ted
u
si
n
g
p
u
b
li
c
l
y
a
v
a
il
a
b
le
e
m
a
il
a
n
d
a
tt
a
c
h
m
e
n
t
d
a
tas
e
ts
th
a
t
a
re
c
o
m
b
in
e
d
t
o
sim
u
late
a
tt
a
c
h
m
e
n
t
-
b
e
a
rin
g
p
h
ish
i
n
g
e
m
a
il
s.
Ex
p
e
rime
n
tal
re
su
lt
s
d
e
m
o
n
stra
te
stro
n
g
d
e
tec
ti
o
n
p
e
rfo
rm
a
n
c
e
a
c
ro
ss
m
u
lt
ip
le
e
v
a
lu
a
ti
o
n
m
e
tri
c
s.
Ne
v
e
rth
e
les
s,
th
e
stu
d
y
a
c
k
n
o
wl
e
d
g
e
s
th
e
li
m
it
a
ti
o
n
o
f
u
si
n
g
sy
n
th
e
ti
c
a
ll
y
p
a
ired
e
m
a
il
b
o
d
ies
a
n
d
a
tt
a
c
h
m
e
n
ts,
wh
ic
h
m
a
y
n
o
t
f
u
ll
y
c
a
p
tu
re
r
e
a
l
-
wo
rld
se
m
a
n
ti
c
re
latio
n
sh
i
p
s.
T
h
e
fin
d
i
n
g
s
h
i
g
h
li
g
h
t
t
h
e
imp
o
r
tan
c
e
o
f
in
c
o
rp
o
ra
ti
n
g
a
tt
a
c
h
m
e
n
t
-
a
wa
re
a
n
a
ly
sis
in
t
o
p
h
ish
i
n
g
d
e
tec
ti
o
n
sy
ste
m
s
a
n
d
p
ro
v
id
e
a
fo
u
n
d
a
ti
o
n
f
o
r
f
u
t
u
re
r
e
se
a
rc
h
o
n
se
m
a
n
ti
c
c
o
n
siste
n
c
y
m
o
d
e
li
n
g
a
n
d
tran
sf
o
rm
e
r
-
b
a
se
d
a
rc
h
it
e
c
tu
r
e
s.
K
ey
w
o
r
d
s
:
Attach
m
en
t
-
b
ased
an
aly
s
is
E
m
ail
s
ec
u
r
ity
Ma
ch
in
e
lear
n
in
g
Mu
lti
-
m
o
d
el
d
etec
tio
n
Neu
r
al
n
etwo
r
k
Ph
is
h
in
g
em
ail
d
etec
tio
n
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
L
am
ee
s
Mo
h
am
ed
Sch
o
o
l o
f
C
o
m
p
u
ter
Scien
ce
,
C
an
ad
ian
I
n
ter
n
atio
n
al
C
o
llag
e
C
air
o
,
E
g
y
p
t
E
m
ail: L
am
ee
s
_
m
_
f
r
j@
cic
-
ca
ir
o
.
co
m
1.
I
NT
RO
D
UCT
I
O
N
Ph
is
h
in
g
em
ails
co
n
tin
u
e
to
p
o
s
e
a
m
ajo
r
cy
b
er
s
ec
u
r
ity
t
h
r
ea
t,
with
attac
k
s
in
cr
ea
s
in
g
l
y
lev
er
ag
in
g
m
alicio
u
s
attac
h
m
en
ts
to
b
y
p
ass
tr
ad
itio
n
al
tex
t
-
b
ased
d
etec
tio
n
s
y
s
tem
s
.
R
ec
en
t
r
ep
o
r
ts
in
d
icate
th
at
o
v
er
9
0
%
o
f
cy
b
er
attac
k
s
in
2
0
2
3
in
v
o
lv
ed
s
o
m
e
f
o
r
m
o
f
s
o
cial
en
g
in
ee
r
i
n
g
,
in
clu
d
in
g
p
h
is
h
in
g
ca
m
p
aig
n
s
tar
g
etin
g
b
o
th
in
d
iv
id
u
als
an
d
o
r
g
a
n
izatio
n
s
(
Mic
r
o
s
o
f
t
Secu
r
ity
I
n
tellig
en
ce
R
ep
o
r
t,
2
0
2
3
)
.
Desp
ite
n
u
m
er
o
u
s
ad
v
a
n
ce
s
in
e
m
ail
s
ec
u
r
ity
,
c
u
r
r
en
t
d
etec
tio
n
s
y
s
tem
s
p
r
im
ar
ily
f
o
cu
s
o
n
tex
tu
a
l
co
n
ten
t,
cr
ea
tin
g
a
b
lin
d
s
p
o
t
in
attac
h
m
en
t
-
awa
r
e
p
h
is
h
in
g
d
etec
tio
n
.
T
h
is
s
tu
d
y
a
d
d
r
ess
es
th
is
g
a
p
b
y
p
r
o
p
o
s
in
g
a
m
u
lti
-
m
o
d
al
class
if
icatio
n
m
o
d
el
th
at
in
t
eg
r
ates
h
ea
d
er
f
ea
tu
r
es,
b
o
d
y
tex
t
an
aly
s
is
,
an
d
attac
h
m
en
t
in
s
p
ec
tio
n
[
1
]
.
Ph
is
h
in
g
is
a
c
y
b
er
attac
k
th
at
co
m
b
i
n
es
s
o
cial
en
g
in
ee
r
in
g
tech
n
iq
u
es
with
tech
n
ical
m
a
n
ip
u
latio
n
to
o
b
tain
s
en
s
itiv
e
u
s
er
in
f
o
r
m
atio
n
.
I
n
th
is
ty
p
e
o
f
attac
k
,
th
e
a
d
v
er
s
ar
y
s
en
d
s
d
ec
e
p
tiv
e
e
m
ails
d
esig
n
ed
to
m
is
lead
r
ec
ip
ien
ts
in
to
r
ev
ea
lin
g
co
n
f
i
d
en
tial
d
ata.
T
h
is
in
f
o
r
m
atio
n
m
ay
b
e
ex
p
o
s
ed
th
r
o
u
g
h
d
if
f
e
r
en
t
m
eth
o
d
s
,
s
u
ch
as
click
in
g
o
n
h
ar
m
f
u
l
lin
k
s
t
h
at
lead
to
f
r
au
d
u
len
t
web
s
ites
wh
er
e
lo
g
in
d
etails
ar
e
en
ter
ed
,
d
ir
ec
tly
s
h
a
r
in
g
th
e
d
ata,
o
r
d
o
wn
lo
a
d
in
g
at
tach
m
en
ts
em
b
ed
d
e
d
with
m
alicio
u
s
s
o
f
twar
e
[
2
]
.
Ph
is
h
in
g
o
f
te
n
ex
p
l
o
its
wea
k
n
ess
es
in
s
y
s
tem
p
r
o
ce
s
s
es,
an
d
f
r
o
m
th
e
e
n
d
u
s
er
'
s
p
e
r
s
p
ec
tiv
e,
d
ef
en
d
in
g
ag
ain
s
t
it
lar
g
ely
d
e
p
en
d
s
o
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
A
mu
lti
-
mo
d
a
l fra
mewo
r
k
fo
r
imp
r
o
vin
g
th
e
a
cc
u
r
a
cy
o
f p
h
i
s
h
in
g
ema
il d
etec
tio
n
(
La
mee
s
Mo
h
a
med
F
a
r
a
j
)
1609
th
eir
k
n
o
wled
g
e
an
d
awa
r
en
e
s
s
,
wh
ich
ca
n
v
ar
y
wid
ely
a
m
o
n
g
in
d
iv
i
d
u
als
[
1
]
.
E
v
en
with
a
well
-
s
ec
u
r
ed
s
y
s
tem
,
an
u
n
awa
r
e
u
s
er
m
ay
b
e
tr
ick
ed
in
to
r
ev
ea
lin
g
p
er
s
o
n
al
o
r
f
in
an
cial
d
etails,
wh
ich
ca
n
f
u
r
th
er
jeo
p
ar
d
ize
th
e
o
r
g
an
izatio
n
'
s
s
ec
u
r
ity
b
y
g
r
an
tin
g
u
n
a
u
th
o
r
iz
ed
ac
ce
s
s
to
its
s
en
s
itiv
e
d
ata.
Mc
Af
ee
esti
m
ates
th
at
cy
b
er
cr
im
e
-
r
elate
d
lo
s
s
es
co
s
t
th
e
g
lo
b
al
ec
o
n
o
m
y
ap
p
r
o
x
im
ately
o
n
e
tr
illi
o
n
US
d
o
llar
s
an
n
u
ally
[
1
]
,
[
3
]
.
Ov
e
r
all,
p
h
is
h
in
g
co
n
tin
u
es to
b
e
a
s
ig
n
if
ican
t th
r
ea
t a
f
f
ec
tin
g
b
o
th
in
d
iv
id
u
als an
d
b
u
s
in
ess
e
s
[
4
]
.
R
ec
en
tly
,
p
h
is
h
in
g
attac
k
e
r
s
h
av
e
in
cr
ea
s
in
g
ly
r
elied
o
n
m
a
licio
u
s
em
ail
attac
h
m
en
ts
,
s
u
ch
as
PDF
d
o
cu
m
e
n
ts
,
im
ag
es,
an
d
co
m
p
r
ess
ed
f
iles
,
to
b
y
p
ass
tr
ad
i
tio
n
al
tex
t
-
b
ased
d
etec
tio
n
s
y
s
tem
s
,
r
ev
ea
lin
g
a
cr
itical
b
lin
d
s
p
o
t
in
m
a
n
y
ex
is
tin
g
p
h
is
h
in
g
d
etec
tio
n
ap
p
r
o
ac
h
es
[
4
]
.
A
s
tu
d
y
co
n
d
u
cte
d
b
y
B
ir
th
r
iy
a
a
n
d
J
ain
o
n
p
h
is
h
in
g
e
m
ail
d
etec
tio
n
in
d
icate
s
th
at
p
h
is
h
in
g
attac
k
s
m
o
s
t
f
r
eq
u
en
tly
tar
g
et
s
p
e
cif
ic
s
ec
to
r
s
,
with
th
e
SaaS
in
d
u
s
tr
y
ac
co
u
n
tin
g
f
o
r
3
4
.
7
%,
f
o
llo
wed
b
y
f
in
an
cial
in
s
titu
tio
n
s
at
1
8
%,
p
ay
m
en
t
s
er
v
ices
at
1
1
.
8
%,
s
o
cial
m
e
d
ia
p
latf
o
r
m
s
at
1
0
.
8
%,
a
n
d
e
-
c
o
m
m
er
ce
b
u
s
in
ess
es
at
7
.
5
%
[
5
]
.
Acc
o
r
d
in
g
to
r
ec
e
n
t
cy
b
er
s
ec
u
r
ity
r
ep
o
r
ts
,
p
h
is
h
i
n
g
r
e
m
ain
s
o
n
e
o
f
th
e
m
o
s
t
p
r
ev
alen
t
attac
k
v
ec
to
r
s
wo
r
ld
wid
e,
with
a
co
n
tin
u
o
u
s
in
cr
ea
s
e
in
em
ail
-
b
ased
attac
k
s
lev
er
ag
in
g
s
o
cial
en
g
in
ee
r
in
g
an
d
m
alicio
u
s
at
tach
m
en
ts
.
R
ep
o
r
ts
p
u
b
lis
h
ed
i
n
2
0
2
3
an
d
2
0
2
4
in
d
icate
th
at
p
h
is
h
in
g
em
ails
co
n
tin
u
e
to
ac
c
o
u
n
t
f
o
r
a
s
ig
n
if
ican
t
p
o
r
tio
n
o
f
s
u
cc
ess
f
u
l
cy
b
er
attac
k
s
ac
r
o
s
s
b
o
th
p
e
r
s
o
n
al
an
d
o
r
g
an
izatio
n
al
en
v
ir
o
n
m
e
n
ts
[
4
]
.
Giv
e
n
th
e
v
ast
v
o
lu
m
e
o
f
p
o
ten
tially
m
alicio
u
s
em
ails
,
alo
n
g
with
t
h
eir
in
cr
ea
s
in
g
d
i
v
er
s
ity
,
th
er
e
is
a
p
r
ess
in
g
n
e
ed
f
o
r
s
o
lu
tio
n
s
th
at
en
h
an
ce
th
e
to
o
ls
an
d
tech
n
iq
u
es
em
ail
s
er
v
ice
p
r
o
v
id
er
s
u
s
e
to
d
etec
t
h
ar
m
f
u
l
em
ails
[
4
]
.
E
m
ail
m
ess
ag
es
ty
p
ically
co
n
s
is
t o
f
th
r
ee
m
ain
co
m
p
o
n
en
ts
:
a.
T
h
e
m
ess
ag
e
en
v
elo
p
e,
wh
ich
r
ep
r
esen
ts
th
e
em
ail'
s
elec
tr
o
n
ic
f
o
r
m
at.
b.
T
h
e
m
ess
ag
e
h
ea
d
er
,
wh
ich
co
n
tain
s
d
etails
s
u
ch
as
t
h
e
em
ail
s
u
b
ject
lin
e
an
d
s
en
d
er
/r
ec
ip
ien
t
in
f
o
r
m
atio
n
.
c.
T
h
e
m
ess
ag
e
b
o
d
y
,
wh
ich
in
cl
u
d
es tex
t,
im
ag
es,
an
d
f
ile
atta
ch
m
en
ts
.
Fig
u
r
e
1
illu
s
tr
ates
an
ex
a
m
p
le
o
f
th
e
s
tan
d
ar
d
s
tr
u
ct
u
r
e
o
f
a
m
alicio
u
s
s
p
am
e
m
ail.
Su
ch
em
ails
o
f
ten
em
p
l
o
y
s
o
cial
en
g
in
ee
r
in
g
tech
n
iq
u
es
in
th
eir
s
u
b
je
ct
lin
e
o
r
b
o
d
y
tex
t
to
d
ec
ei
v
e
r
ec
ip
ien
ts
in
to
in
ter
ac
tin
g
with
th
e
m
ess
ag
e.
Fo
r
ex
am
p
le,
t
h
e
r
ec
ip
ien
t
m
ay
b
e
in
s
tr
u
cted
to
d
o
wn
l
o
ad
a
co
m
p
r
ess
ed
f
ile,
p
r
eten
d
in
g
to
r
etr
iev
e
a
f
ak
e
u
n
d
eliv
er
ed
item
.
T
h
ese
f
iles
ar
e
in
ten
tio
n
ally
m
is
lead
in
g
an
d
o
f
ten
u
s
e
m
u
ltip
le
f
ile
ex
ten
s
io
n
s
to
o
b
s
cu
r
e
th
eir
m
alicio
u
s
n
atu
r
e
[
6
]
.
Fig
u
r
e
1
.
An
ex
am
p
le
o
f
(
f
ak
e
)
s
p
am
em
ail
with
a
p
o
ten
tial m
alicio
u
s
attac
h
m
en
t a
n
d
UR
L
[
7
]
T
h
e
g
o
al
o
f
t
h
is
co
m
p
r
ess
ed
f
ile
in
Fig
u
r
e
1
is
to
h
i
d
e
m
al
war
e
ex
ec
u
tab
le
f
r
o
m
v
ir
u
s
s
c
an
n
er
s
r
u
n
b
y
o
r
ap
p
lied
b
y
m
ail
s
er
v
er
s
.
I
n
th
is
ex
am
p
le,
th
e
UR
L
ac
ts
as
a
s
ec
o
n
d
ar
y
m
et
h
o
d
to
s
p
r
ea
d
h
ar
m
f
u
l
m
ater
ial.
Similar
to
m
alicio
u
s
f
iles
,
m
alicio
u
s
UR
L
s
ca
n
h
id
e
a
h
ar
m
f
u
l
web
s
ite
(
s
u
ch
a
s
ex
a
mp
le.
co
m
)
b
y
ad
d
in
g
s
u
b
d
o
m
ain
s
th
at
r
ese
m
b
le
a
tr
u
s
ted
an
d
s
ec
u
r
e
we
b
s
ite
(
e.
g
.
,
tr
a
ck
in
g
.
yo
u
r
p
o
s
to
ffice
)
.
T
h
is
ex
am
p
le
also
illu
s
tr
ates
a
p
o
s
s
ib
le
s
p
a
m
tem
p
late
wh
er
e
UR
L
s
o
r
attac
h
m
en
ts
m
ay
h
av
e
d
if
f
er
e
n
t
n
am
es
b
u
t
s
till
s
er
v
e
th
e
s
am
e
m
alicio
u
s
p
u
r
p
o
s
e
[
8
]
.
Desp
ite
th
e
ef
f
ec
tiv
en
ess
o
f
ex
is
tin
g
m
ac
h
in
e
lear
n
in
g
-
b
a
s
ed
p
h
is
h
in
g
d
etec
tio
n
tech
n
i
q
u
es,
m
o
s
t
p
r
o
p
o
s
ed
s
o
lu
tio
n
s
p
r
im
a
r
ily
f
o
cu
s
o
n
a
n
aly
zin
g
em
ail
h
ea
d
er
s
o
r
b
o
d
y
te
x
t,
wh
ile
th
e
an
aly
s
is
o
f
m
alicio
u
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
1
6
,
No
.
3
,
J
u
n
e
20
2
6
:
1
6
0
8
-
1
6
2
5
1610
attac
h
m
en
ts
is
o
f
ten
tr
ea
ted
as
a
s
ec
o
n
d
ar
y
o
r
in
d
e
p
en
d
e
n
t
task
[
8
]
,
[
9
]
.
Var
i
o
u
s
tech
n
iq
u
es
ar
e
em
p
lo
y
ed
f
o
r
f
ilter
in
g
u
n
s
o
licited
em
ails
,
in
clu
d
in
g
k
n
o
wled
g
e
-
b
ased
m
eth
o
d
s
,
lear
n
in
g
-
b
ased
ap
p
r
o
ac
h
es,
clu
s
ter
in
g
tech
n
iq
u
es,
h
e
u
r
is
tic
p
r
o
ce
d
u
r
es,
an
d
o
th
e
r
s
.
Ho
wev
er
,
th
ese
tech
n
iq
u
es
o
f
ten
s
tr
u
g
g
le
to
p
r
ev
e
n
t
b
y
p
ass
attac
k
s
.
T
o
ad
d
r
ess
th
is
lim
i
tatio
n
,
th
is
p
ap
er
in
tr
o
d
u
ce
s
th
e
p
r
o
p
o
s
ed
p
h
is
h
in
g
em
a
il
class
if
ier
m
o
d
el
(
PECM)
d
esig
n
ed
to
class
if
y
e
m
ail
s
eg
m
en
ts
as
eith
er
s
p
am
o
r
leg
itima
te
(
h
am
)
.
T
h
e
m
o
d
el
id
en
tifie
s
g
ap
s
in
ex
is
tin
g
em
ail
f
ilter
in
g
s
y
s
tem
s
an
d
an
aly
ze
s
p
h
is
h
in
g
em
ail
lin
k
s
an
d
m
alicio
u
s
attac
h
m
e
n
ts
.
Sp
am
em
ails
co
n
tain
in
g
m
alicio
u
s
attac
h
m
en
ts
o
r
UR
L
s
aim
to
d
ec
eiv
e
r
ec
ip
ien
ts
in
to
o
p
en
in
g
attac
h
m
en
t
o
r
click
i
n
g
o
n
lin
k
s
.
T
h
e
s
u
b
ject
lin
es
an
d
b
o
d
y
tex
t
o
f
th
ese
em
ails
ar
e
cr
a
f
ted
to
tem
p
t
o
r
alar
m
th
e
r
ec
ip
ien
t
in
to
r
esp
o
n
d
in
g
to
th
e
d
is
g
u
is
ed
m
alicio
u
s
c
o
n
ten
t.
Scan
n
in
g
UR
L
s
an
d
attac
h
m
en
ts
ag
ain
s
t
v
ir
u
s
s
ca
n
n
er
s
o
r
b
lack
lis
ts
,
o
f
ten
r
ev
ea
lin
g
th
e
n
at
u
r
e
an
d
s
co
p
e
o
f
th
e
m
alicio
u
s
co
n
ten
t
in
th
ese
h
ar
m
f
u
l
em
ails
.
Ho
wev
er
,
th
ese
s
ca
n
n
in
g
p
r
o
ce
d
u
r
es
u
s
u
ally
r
eq
u
ir
e
ex
ter
n
al
r
eso
u
r
ce
s
,
wh
ich
ca
n
b
e
co
s
tly
to
co
m
p
u
te
an
d
h
a
r
d
to
m
ai
n
tain
.
W
h
ile
m
an
y
s
p
am
d
etec
tio
n
an
d
f
ilter
in
g
s
y
s
tem
s
ty
to
ad
ap
t
to
ch
a
n
g
in
g
s
p
am
m
in
g
s
tr
ateg
ies,
th
e
y
ar
e
o
f
ten
to
o
i
n
f
lex
ib
le
t
o
ef
f
ec
tiv
ely
h
a
n
d
le
th
e
ev
o
l
v
in
g
n
atu
r
e
o
f
s
p
am
em
ails
[
8
]
.
T
h
er
e
h
as
b
ee
n
lim
ited
r
e
s
ea
r
ch
o
n
d
etec
tin
g
m
alicio
u
s
co
n
ten
t
—
s
p
ec
if
ically
attac
h
m
en
ts
o
r
UR
L
s
—
in
s
p
am
em
ails
.
T
h
is
wo
r
k
ad
d
r
ess
es
th
at
g
ap
b
y
f
o
cu
s
in
g
o
n
class
if
y
in
g
an
d
p
h
is
h
in
g
e
m
ails
b
ased
o
n
th
eir
UR
L
s
an
d
attac
h
m
en
ts
.
Un
lik
e
s
im
ilar
ap
p
r
o
ac
h
es,
th
e
p
r
o
p
o
s
ed
PECM is
s
elf
-
co
n
tain
ed
an
d
d
o
es n
o
t r
ely
o
n
ex
ter
n
al
r
eso
u
r
ce
s
s
u
ch
as
b
lack
lis
ts
.
Giv
en
th
e
r
ap
i
d
ad
o
p
tio
n
o
f
n
ew
tactics
b
y
attac
k
er
s
,
it
is
cr
u
cial
to
im
p
r
o
v
e
th
e
ac
c
u
r
ac
y
o
f
p
h
is
h
in
g
em
ail
d
etec
tio
n
s
y
s
tem
s
an
d
to
m
itig
ate
b
y
p
ass
attem
p
ts
,
as
th
e
co
n
ten
t
o
f
m
alicio
u
s
attac
h
m
en
ts
an
d
p
h
i
s
h
in
g
lin
k
s
f
r
eq
u
en
tly
c
h
an
g
es
.
T
h
er
ef
o
r
e,
th
is
p
a
p
er
aim
s
to
im
p
r
o
v
e
p
h
is
h
in
g
em
ail
d
etec
tio
n
ac
cu
r
ac
y
b
y
p
r
o
p
o
s
in
g
a
class
if
icatio
n
m
o
d
el
th
at
jo
i
n
tly
an
aly
ze
s
em
ail
h
ea
d
er
s
,
b
o
d
y
co
n
ten
t,
an
d
attac
h
m
en
ts
u
s
in
g
m
ac
h
in
e
lear
n
in
g
a
n
d
e
n
s
em
b
le
-
b
ased
d
ec
is
io
n
m
ec
h
an
is
m
s
.
T
h
e
m
ain
co
n
tr
ib
u
tio
n
s
o
f
th
is
p
ap
er
ca
n
b
e
s
u
m
m
ar
ized
as f
o
llo
ws:
a.
W
e
p
r
o
p
o
s
e
a
p
h
is
h
in
g
d
etec
t
io
n
m
o
d
el
th
at
an
aly
s
es
em
ail
attac
h
m
en
ts
in
ad
d
itio
n
to
h
e
ad
er
s
an
d
b
o
d
y
co
n
ten
t.
b.
W
e
ad
d
r
ess
a
cr
itical
b
lin
d
s
p
o
t in
ex
is
tin
g
s
y
s
tem
s
b
y
in
c
o
r
p
o
r
atin
g
m
alicio
u
s
attac
h
m
en
t
f
ea
tu
r
es.
c.
W
e
ev
alu
ate
th
e
m
o
d
el
o
n
a
b
en
ch
m
ar
k
d
ataset,
d
e
m
o
n
s
tr
ati
n
g
im
p
r
o
v
e
d
d
etec
tio
n
ac
cu
r
a
cy
co
m
p
ar
ed
to
b
aselin
e
m
eth
o
d
s
.
d.
W
e
p
r
esen
t a
m
o
d
u
lar
f
r
am
ew
o
r
k
th
at
ca
n
b
e
in
te
g
r
ated
in
t
o
ex
is
tin
g
em
ail
s
ec
u
r
ity
s
y
s
tem
s
.
T
h
e
r
est
o
f
th
is
p
ap
er
is
o
r
g
a
n
ized
as
f
o
llo
ws.
T
h
e
s
ec
tio
n
titl
ed
“
r
elate
d
wo
r
k
”
in
tr
o
d
u
c
es
th
e
k
ey
d
ef
in
itio
n
s
an
d
r
e
v
iews
p
r
io
r
s
tu
d
ies
in
th
is
ar
ea
.
I
n
“
th
e
p
r
o
p
o
s
ed
p
h
is
h
in
g
em
ail
class
if
ier
m
o
d
el
(
PECM)
,
”
th
e
ad
o
p
ted
m
eth
o
d
o
l
o
g
y
an
d
th
e
p
r
o
p
o
s
ed
m
o
d
el
ar
e
d
escr
ib
ed
.
T
h
e
“
ev
alu
atio
n
an
d
r
esu
lts
”
s
ec
tio
n
p
r
o
v
id
es
an
a
n
aly
s
is
o
f
th
e
o
u
tco
m
es
to
m
ea
s
u
r
e
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
p
r
o
p
o
s
ed
ap
p
r
o
a
ch
.
T
h
is
is
f
o
llo
wed
b
y
a
d
is
cu
s
s
io
n
o
f
th
e
s
tu
d
y
’
s
lim
itatio
n
s
.
Fin
ally
,
t
h
e
“
c
o
n
clu
s
io
n
”
s
u
m
m
ar
izes
th
e
m
ain
f
in
d
in
g
s
a
n
d
h
ig
h
lig
h
ts
p
o
te
n
tial d
ir
ec
tio
n
s
f
o
r
f
u
tu
r
e
r
esear
c
h
.
2.
RE
L
AT
E
D
WO
RK
T
h
is
s
ec
tio
n
s
u
m
m
ar
izes
r
elat
ed
s
tu
d
ies
th
at
h
ig
h
lig
h
t
m
eth
o
d
o
lo
g
ies
b
ased
o
n
m
ac
h
in
e
lear
n
in
g
f
o
r
s
p
am
d
etec
tio
n
.
T
h
e
f
o
llo
win
g
r
elate
d
s
tu
d
ies
ar
e
d
iv
id
ed
i
n
to
th
r
ee
s
u
b
s
ec
tio
n
s
.
First,
th
e
s
tu
d
ies
th
at
f
o
cu
s
on
em
ail
s
p
am
d
etec
tio
n
.
Seco
n
d
th
e
s
tu
d
ies th
at
f
o
cu
s
o
n
m
alicio
u
s
attac
h
m
en
t d
etec
tio
n
.
Fin
ally
,
th
e
s
tu
d
ies
th
at
f
o
cu
s
o
n
m
alicio
u
s
UR
L
d
etec
tio
n
[
1
0
]
.
2
.1
.
E
ma
il
s
pa
m
f
ilte
ring
a
n
d det
ec
t
io
n
T
h
is
s
u
b
s
ec
tio
n
r
ev
iews
m
eth
o
d
s
f
o
r
d
etec
tin
g
m
alicio
u
s
e
m
ails
an
d
in
tr
o
d
u
ce
s
a
tax
o
n
o
m
y
o
f
th
ese
tech
n
iq
u
es,
as
illu
s
tr
ated
in
Fig
u
r
e
2
.
T
h
e
tax
o
n
o
m
y
is
d
iv
i
d
ed
in
to
two
lev
els:
E
m
ail
co
m
p
o
n
en
ts
n
ee
d
ed
f
o
r
f
ea
tu
r
e
ex
tr
ac
tio
n
,
s
u
ch
as
th
e
h
ea
d
er
,
b
o
d
y
,
o
r
attac
h
m
en
t,
ar
e
id
en
tifie
d
b
y
L
ay
er
1
.
T
h
e
ac
cu
r
ac
y
o
f
class
if
icatio
n
is
in
f
lu
en
ce
d
d
if
f
er
en
tly
b
y
ea
ch
c
o
m
p
o
n
en
t.
L
ay
er
2
:
d
if
f
er
e
n
tiates
b
etwe
en
co
m
m
o
n
m
ac
h
in
e
lear
n
in
g
(
ML
)
an
d
d
ee
p
lea
r
n
i
n
g
(
DL
)
tec
h
n
iq
u
es f
o
r
id
e
n
tif
y
in
g
f
r
au
d
u
le
n
t a
n
d
b
en
ig
n
em
ails
[
1
1
]
.
B
ea
m
an
an
d
I
s
ah
[
1
2
]
f
o
c
u
s
ed
o
n
an
o
m
aly
d
etec
tio
n
i
n
em
ail
h
ea
d
er
s
,
b
y
u
s
in
g
f
ea
tu
r
es
s
u
ch
as
m
is
s
in
g
f
ield
s
,
b
lack
lis
ted
SMT
P
s
er
v
er
s
,
an
d
tim
e
zo
n
e
s
.
T
h
e
o
n
e
-
class
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM)
ac
h
iev
ed
8
7
%
ac
cu
r
ac
y
i
n
d
et
ec
tin
g
p
h
is
h
in
g
em
ails
,
wh
ile
a
s
tack
ed
en
s
em
b
le
(
r
a
n
d
o
m
f
o
r
est
(
RF
)
,
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM)
,
m
u
l
tilay
er
p
er
ce
p
tr
o
n
(
ML
P)
d
a
n
K
-
n
ea
r
est
n
eig
h
b
o
r
s
(
KNN)
)
ac
h
iev
ed
9
7
%
ac
cu
r
ac
y
in
p
h
is
h
in
g
d
etec
tio
n
.
Ab
d
u
lNab
i
an
d
Yaseen
[
1
3
]
e
x
am
in
ed
th
e
ef
f
ec
tiv
en
ess
o
f
wo
r
d
em
b
e
d
d
in
g
tech
n
iq
u
es
i
n
d
etec
tin
g
s
p
am
em
ails
.
T
h
ey
en
h
an
ce
d
th
e
p
er
f
o
r
m
an
ce
o
f
b
id
ir
ec
ti
o
n
al
en
co
d
e
r
r
ep
r
esen
tatio
n
s
f
r
o
m
tr
an
s
f
o
r
m
er
s
(
B
E
R
T
)
,
a
p
r
e
-
tr
ai
n
ed
tr
a
n
s
f
o
r
m
er
-
b
ased
m
o
d
el,
to
class
if
y
e
m
ails
in
to
s
p
am
an
d
leg
itima
t
e
(
h
am
)
ca
teg
o
r
ies.
Owin
g
to
its
atten
tio
n
m
ec
h
an
is
m
,
B
E
R
T
is
ca
p
ab
le
o
f
ca
p
t
u
r
in
g
co
n
tex
t
u
al
r
elatio
n
s
h
i
p
s
with
in
tex
t,
wh
ich
co
n
tr
ib
u
tes to
im
p
r
o
v
ed
class
i
f
icatio
n
p
er
f
o
r
m
an
ce
.
Fo
r
co
m
p
ar
is
o
n
p
u
r
p
o
s
es,
a
b
aselin
e
d
ee
p
n
eu
r
al
n
etwo
r
k
(
DNN)
ar
ch
itectu
r
e
was
im
p
lem
en
ted
,
co
n
s
is
tin
g
o
f
two
f
u
lly
co
n
n
ec
te
d
lay
er
s
alo
n
g
with
a
b
id
ir
ec
tio
n
al
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
A
mu
lti
-
mo
d
a
l fra
mewo
r
k
fo
r
imp
r
o
vin
g
th
e
a
cc
u
r
a
cy
o
f p
h
i
s
h
in
g
ema
il d
etec
tio
n
(
La
mee
s
Mo
h
a
med
F
a
r
a
j
)
1611
lo
n
g
s
h
o
r
t
-
ter
m
m
em
o
r
y
(
B
iLST
M)
lay
er
.
Ad
d
itio
n
ally
,
t
h
e
s
tu
d
y
e
v
alu
ated
t
r
ad
itio
n
a
l
m
ac
h
in
e
lear
n
in
g
ap
p
r
o
ac
h
es su
ch
as
n
aiv
e
B
ay
es (
NB
)
an
d
K
NN.
T
h
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
ac
h
ie
v
ed
a
p
ea
k
a
cc
u
r
ac
y
o
f
9
8
.
6
7
%
an
d
an
F1
-
s
co
r
e
o
f
9
8
.
6
6
%
.
T
h
ese
r
esu
lts
wer
e
o
b
tain
ed
u
s
in
g
two
p
u
b
licly
av
ailab
le
d
atasets
—
o
n
e
d
esig
n
ated
f
o
r
tr
ain
i
n
g
a
n
d
th
e
o
th
er
f
o
r
test
in
g
—
to
ass
ess
th
e
m
o
d
el’
s
g
en
er
aliza
t
io
n
ca
p
a
b
ilit
y
an
d
r
o
b
u
s
tn
ess
wh
en
h
a
n
d
lin
g
u
n
s
ee
n
d
ata.
Fig
u
r
e
2
.
T
a
x
o
n
o
m
y
o
f
m
alici
o
u
s
em
ail
d
etec
tio
n
s
tu
d
ies
Sh
ar
af
f
an
d
Srin
iv
asar
a
o
[
1
4
]
a
n
aly
ze
d
th
e
co
r
r
elatio
n
b
etwe
en
th
e
wo
r
d
s
u
s
ed
in
em
ail
s
u
b
ject
lin
es
an
d
th
eir
c
o
r
r
esp
o
n
d
in
g
m
es
s
ag
e
co
n
ten
t
to
d
eter
m
in
e
t
h
e
n
atu
r
e
o
f
th
e
e
m
ails
.
T
h
ey
ap
p
lied
a
f
ea
tu
r
e
s
elec
tio
n
p
r
o
ce
s
s
to
id
en
tify
th
e
m
o
s
t
s
ig
n
if
ican
t
ter
m
s
f
r
o
m
th
e
en
tire
v
o
ca
b
u
lar
y
,
wh
ic
h
wer
e
th
en
u
tili
ze
d
to
co
n
s
tr
u
ct
N
-
g
r
am
f
ea
tu
r
es.
T
h
e
e
f
f
ec
tiv
en
ess
o
f
th
ese
N
-
g
r
am
r
ep
r
esen
tatio
n
s
was
ev
alu
ated
u
s
in
g
f
o
u
r
m
ac
h
in
e
lear
n
i
n
g
class
if
ier
s
:
d
ec
is
io
n
tr
ee
,
m
u
ltin
o
m
ial
NB
,
RF
,
an
d
lin
ea
r
SVM.
T
h
e
ex
p
er
im
en
tal
ev
alu
atio
n
was c
o
n
d
u
cted
u
s
in
g
th
e
L
in
g
-
Sp
am
d
ataset.
Pen
g
et
a
l.
[
1
5
]
c
o
n
ce
n
tr
ate
d
o
n
u
s
in
g
s
em
an
tic
an
aly
s
i
s
to
id
en
tify
m
alicio
u
s
in
ten
t
in
em
ail
ass
au
lts
b
y
ex
am
in
in
g
th
e
n
a
tu
r
al
lan
g
u
ag
e
te
x
t.
A
s
izab
le
b
en
ch
m
ar
k
s
et
o
f
p
h
is
h
in
g
e
m
ails
was
u
s
ed
to
s
h
o
w
th
e
ef
f
icac
y
o
f
th
is
s
tr
ateg
y
.
Alh
o
g
ail
an
d
Als
ab
ih
[
7
]
p
r
o
p
o
s
ed
a
p
h
is
h
in
g
em
ail
class
if
icatio
n
m
o
d
el
b
ased
o
n
d
ee
p
lear
n
in
g
tech
n
iq
u
es.
T
h
eir
ap
p
r
o
ac
h
e
n
h
an
ce
s
d
etec
tio
n
ac
cu
r
ac
y
b
y
ap
p
ly
in
g
n
atu
r
al
lan
g
u
ag
e
p
r
o
ce
s
s
in
g
(
NL
P)
to
an
aly
ze
em
ail
co
n
te
n
t a
n
d
i
n
c
o
r
p
o
r
ati
n
g
a
g
r
ap
h
co
n
v
o
lu
tio
n
al
n
etwo
r
k
(
GC
N)
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
1
6
,
No
.
3
,
J
u
n
e
20
2
6
:
1
6
0
8
-
1
6
2
5
1612
2
.2
.
Cla
s
s
if
ica
t
io
n o
f
ma
licio
us
a
t
t
a
chm
ent
s
Mu
r
alid
h
ar
an
a
n
d
Nis
s
im
[
6
]
.
p
r
o
p
o
s
ed
a
n
au
to
m
ated
m
alicio
u
s
em
ail
d
etec
tio
n
f
r
am
ew
o
r
k
u
s
in
g
d
ee
p
en
s
em
b
le
lea
r
n
in
g
,
th
at
an
aly
ze
s
all
s
eg
m
en
ts
o
f
a
n
em
ail.
T
h
eir
ap
p
r
o
ac
h
d
e
m
o
n
s
tr
ated
th
at
a
n
en
s
em
b
le
f
r
am
ewo
r
k
o
f
d
ee
p
lear
n
in
g
class
if
ier
s
,
ea
ch
tr
ai
n
ed
o
n
d
if
f
er
en
t
p
o
r
tio
n
s
o
f
an
em
ail,
ca
n
o
f
f
er
b
etter
g
en
er
aliza
tio
n
th
an
tr
a
d
i
tio
n
al
m
eth
o
d
s
th
at
o
n
ly
a
n
aly
ze
s
p
ec
if
ic
s
ec
tio
n
s
o
f
an
em
ai
l.
L
is
tík
et
a
l.
[
1
6
]
d
e
v
elo
p
ed
a
m
o
d
el
ar
ch
itectu
r
e
t
h
at
is
b
ased
o
n
co
n
v
o
lu
tio
n
al
n
eu
r
a
l
n
etwo
r
k
s
(
C
NN)
,
wh
ich
h
av
e
d
em
o
n
s
tr
ated
im
p
r
ess
iv
e
p
ictu
r
e
ca
teg
o
r
izatio
n
p
er
f
o
r
m
a
n
ce
.
T
h
ey
ev
alu
ated
th
is
ar
ch
itectu
r
e
o
n
two
p
u
b
licly
a
cc
ess
ib
le
d
ataset
s
an
d
wer
e
ab
le
to
class
if
y
em
ail
im
ag
es
as
s
p
am
with
s
tate
-
of
-
th
e
-
ar
t
p
er
f
o
r
m
an
ce
.
On
b
o
th
t
h
e
Prin
ce
to
n
an
d
Dr
ed
ze
d
atas
ets,
th
e
s
u
g
g
ested
C
N
N
ar
ch
it
ec
tu
r
e
ac
h
iev
ed
an
F1
-
s
co
r
e
o
f
9
6
%
an
d
9
9
%,
r
e
s
p
ec
tiv
ely
.
Ad
d
itio
n
ally
,
u
s
in
g
th
e
E
m
ail.
cz
im
ag
e
s
p
am
v
1
d
ataset,
it
o
b
tain
ed
an
F1
-
s
co
r
e
o
f
u
p
to
9
6
%.
T
r
an
et
a
l.
[
8
]
p
r
o
p
o
s
ed
a
n
u
m
b
er
o
f
n
o
v
el
f
ea
tu
r
es
f
o
r
id
en
ti
f
y
in
g
d
an
g
er
o
u
s
UR
L
s
an
d
att
ac
h
m
en
ts
in
s
p
am
em
ails
.
Acc
o
r
d
in
g
t
o
th
eir
h
y
p
o
th
esis
,
th
e
b
est
b
asis
f
o
r
d
etec
tin
g
s
p
am
em
ails
with
h
ar
m
f
u
l
attac
h
m
en
ts
o
r
UR
L
s
is
th
e
tex
t
co
n
te
n
t
in
th
e
em
ail
s
u
b
ject
an
d
b
o
d
y
.
Sin
g
h
et
a
l.
[
1
7
]
aim
ed
to
en
h
a
n
ce
th
e
class
if
icatio
n
r
esu
lts
b
y
ap
p
ly
i
n
g
n
eu
r
al
n
etwo
r
k
s
an
d
DNN
to
th
e
s
p
am
im
ag
e
class
if
icatio
n
p
r
o
b
lem
.
San
ju
s
h
r
ee
an
d
Me
e
n
a
[
1
8
]
u
s
ed
p
ar
ticle
s
war
m
o
p
tim
iz
atio
n
(
PS
O)
an
d
SVM
o
n
th
r
ee
tex
tu
al
f
ea
tu
r
es
an
d
ten
m
etad
ata
f
ea
tu
r
es
to
ca
teg
o
r
ize
im
ag
e
s
p
am
.
On
th
e
Dr
ed
ze
d
ataset,
w
h
ich
co
n
tain
ed
3
0
0
tr
ain
in
g
im
ag
es
a
n
d
3
8
0
test
i
m
ag
es,
th
is
co
m
b
in
atio
n
o
f
m
eth
o
d
s
p
r
o
d
u
ce
d
an
ac
c
u
r
ac
y
o
f
9
0
%.
W
ith
clien
t
an
d
s
er
v
er
m
o
d
els,
th
e
au
th
o
r
s
ad
d
itio
n
ally
em
p
lo
y
clu
s
ter
-
b
ased
f
ilter
in
g
,
claim
in
g
a
9
9
%
ac
cu
r
ac
y
r
ate.
So
r
an
am
ag
eswar
i
et
a
l.
[
1
9
]
ac
h
iev
ed
9
2
.
8
2
%
ac
c
u
r
ac
y
u
s
in
g
a
co
lo
r
f
ea
tu
r
e
-
b
ased
b
ac
k
p
r
o
p
ag
atio
n
n
eu
r
a
l
n
etwo
r
k
.
T
h
ey
also
tr
ied
d
iv
i
d
in
g
th
e
im
ag
e
in
to
p
iece
s
a
n
d
u
s
ed
th
o
s
e
b
lo
ck
s
as
f
ea
t
u
r
es;
h
o
wev
er
,
th
is
ap
p
r
o
ac
h
o
n
l
y
p
r
o
d
u
ce
d
an
ac
cu
r
ac
y
o
f
8
9
.
3
2
%.
2
.3
.
Cla
s
s
if
ica
t
io
n o
f
ma
licio
us
UR
L
s
R
esear
ch
o
n
d
etec
tin
g
m
alici
o
u
s
UR
L
s
h
as
ex
p
an
d
e
d
b
ey
o
n
d
s
p
am
e
m
ail
f
ilter
in
g
,
d
r
i
v
en
b
y
th
e
ex
ten
s
iv
e
u
s
e
o
f
UR
L
s
ac
r
o
s
s
web
co
n
ten
t
an
d
d
ig
ital
co
m
m
u
n
icatio
n
s
.
Alth
o
u
g
h
b
lac
k
lis
tin
g
r
em
ain
s
a
co
m
m
o
n
l
y
u
s
ed
ap
p
r
o
ac
h
to
r
estrict
ac
ce
s
s
to
h
ar
m
f
u
l
lin
k
s
,
its
ef
f
ec
tiv
en
ess
d
ep
e
n
d
s
o
n
p
r
e
v
io
u
s
ly
id
en
tify
in
g
a
n
d
ca
talo
g
in
g
s
u
c
h
UR
L
s
[
2
0
]
.
Sm
ad
i
et
a
l.
[
2
1
]
p
r
o
p
o
s
ed
a
n
eu
r
al
n
etwo
r
k
–
b
ased
a
p
p
r
o
ac
h
th
at
u
tili
ze
s
f
o
u
r
k
e
y
co
m
p
o
n
en
ts
o
f
an
em
ail:
th
e
m
ess
ag
e
b
o
d
y
(
n
in
e
f
ea
tu
r
es),
HT
ML
s
tr
u
ctu
r
e
(
1
0
f
ea
tu
r
es),
em
b
ed
d
ed
UR
L
s
(
1
2
f
ea
t
u
r
es),
an
d
h
ea
d
er
in
f
o
r
m
atio
n
(
1
9
f
ea
tu
r
es).
T
h
ey
also
in
tr
o
d
u
ce
d
f
ea
tu
r
e
ev
alu
atio
n
a
n
d
r
e
d
u
ctio
n
(
FEa
R
)
,
a
tailo
r
ed
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
d
esi
g
n
ed
to
id
en
tif
y
th
e
m
o
s
t
s
u
itab
le
s
u
b
s
et
o
f
f
ea
tu
r
es
b
ased
o
n
th
e
d
ataset’
s
s
iz
e
an
d
ch
ar
ac
te
r
is
tics
.
Am
ee
n
o
et
a
l.
[
2
2
]
wer
e
a
m
o
n
g
th
e
ea
r
ly
r
esear
ch
er
s
to
ad
d
r
ess
p
h
is
h
in
g
d
etec
tio
n
th
r
o
u
g
h
th
e
an
aly
s
is
o
f
m
alicio
u
s
lin
k
s
.
T
h
eir
wo
r
k
in
c
o
r
p
o
r
ated
v
ar
io
u
s
f
ea
tu
r
es,
i
n
clu
d
in
g
I
P
a
d
d
r
ess
ch
ar
ac
ter
is
tics
,
W
HOI
S
r
ec
o
r
d
s
,
d
o
m
ain
-
r
ela
ted
attr
ib
u
tes,
b
lack
lis
t
in
clu
s
io
n
,
g
eo
g
r
ap
h
ic
lo
ca
tio
n
,
an
d
co
n
n
ec
tio
n
s
p
ee
d
.
T
h
e
s
tu
d
y
f
u
r
th
e
r
r
ev
ea
led
th
a
t
o
n
lin
e
lear
n
in
g
tech
n
i
q
u
es
o
u
tp
er
f
o
r
m
ed
b
atch
lear
n
in
g
ap
p
r
o
ac
h
es,
with
th
e
co
n
f
id
en
ce
-
weig
h
ted
o
n
lin
e
m
eth
o
d
ac
h
ie
v
in
g
b
etter
r
esu
lts
th
an
s
ev
er
al
SVM
v
ar
ian
ts
.
Mo
r
e
r
ec
en
tly
,
p
h
is
h
in
g
em
a
il
d
etec
tio
n
h
as
b
en
ef
ited
f
r
o
m
th
e
ad
o
p
tio
n
o
f
tr
an
s
f
o
r
m
er
-
b
ased
ar
ch
itectu
r
es
s
u
ch
as
B
E
R
T
[
2
3
]
.
T
h
ese
m
o
d
els
ex
ce
l
at
c
ap
tu
r
in
g
co
n
tex
t
u
al
an
d
s
em
a
n
tic
in
f
o
r
m
atio
n
in
tex
t
th
r
o
u
g
h
d
ee
p
b
id
ir
ec
tio
n
al
r
ep
r
esen
tatio
n
s
,
o
f
te
n
s
u
r
p
ass
in
g
tr
ad
itio
n
al
m
ac
h
in
e
le
ar
n
in
g
m
et
h
o
d
s
in
p
er
f
o
r
m
an
ce
.
Nev
er
th
eless
,
t
h
eir
h
i
g
h
c
o
m
p
u
tatio
n
al
d
e
m
an
d
s
an
d
e
x
ten
d
ed
tr
ain
in
g
tim
es
ca
n
p
o
s
e
ch
allen
g
es f
o
r
d
ep
lo
y
m
en
t in
r
ea
l
-
tim
e
s
y
s
tem
s
o
r
en
v
ir
o
n
m
en
ts
with
lim
ited
r
eso
u
r
ce
s
[
2
4
]
.
3.
M
E
T
H
O
D
T
h
is
s
ec
tio
n
o
u
tlin
es
th
e
p
r
o
p
o
s
ed
PECM,
an
d
th
e
s
tep
s
in
v
o
lv
ed
in
co
n
d
u
ctin
g
th
e
m
o
d
el,
t
h
e
p
r
o
p
o
s
ed
PECM
em
p
lo
y
s
m
a
ch
in
e
lear
n
in
g
alg
o
r
ith
m
s
to
a
n
aly
ze
all
em
ail
co
m
p
o
n
en
ts
—
b
o
d
y
,
h
ea
d
er
,
a
n
d
attac
h
m
en
ts
.
E
n
h
a
n
cin
g
p
h
is
h
in
g
d
etec
tio
n
ac
c
u
r
ac
y
to
p
r
ev
en
t
u
n
wan
ted
an
d
p
o
ten
ti
ally
h
ar
m
f
u
l
em
ails
f
r
o
m
r
ea
c
h
in
g
t
h
e
u
s
er
'
s
in
b
o
x
.
T
h
e
p
r
o
ce
s
s
o
f
b
u
ild
in
g
th
e
p
r
o
p
o
s
ed
PECM
in
v
o
lv
es
th
r
ee
m
ain
s
tep
s
.
First
s
tep
,
d
ata
f
r
o
m
b
o
th
p
h
is
h
in
g
an
d
leg
itima
te
e
m
ails
ar
e
co
lle
cted
an
d
p
r
e
p
ar
ed
f
o
r
p
r
e
-
p
r
o
c
ess
in
g
s
tag
e.
Nex
t,
th
e
p
r
o
p
o
s
ed
PECM
em
p
lo
y
s
b
ag
-
of
-
wo
r
d
s
(
B
o
W
)
an
d
T
F
-
I
DF
m
eth
o
d
s
to
ex
tr
ac
t
m
ea
n
in
g
f
u
l
f
ea
tu
r
es.
T
h
en
,
th
e
p
r
o
p
o
s
ed
PECM
is
d
ev
elo
p
e
d
u
s
in
g
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
,
with
t
h
e
tr
ai
n
in
g
d
ata
b
ein
g
f
e
d
in
to
th
e
class
if
ier
to
co
n
s
tr
u
ct
th
e
m
o
d
el
an
d
p
r
ep
a
r
e
it
f
o
r
ev
alu
atio
n
.
T
h
e
f
i
n
al
p
h
ase
i
n
v
o
lv
es
test
in
g
th
e
p
r
o
p
o
s
ed
PECM w
ith
th
e
test
i
n
g
d
ata
to
v
alid
ate
th
e
m
o
d
el'
s
p
er
f
o
r
m
an
ce
.
I
n
th
is
w
o
r
k
,
th
e
p
r
o
p
o
s
ed
p
h
is
h
in
g
d
etec
tio
n
m
o
d
el
is
p
r
im
ar
ily
d
esig
n
ed
to
o
p
er
ate
as
a
s
elf
-
co
n
tain
ed
s
y
s
tem
b
ased
o
n
m
ac
h
in
e
lear
n
in
g
tech
n
i
q
u
es.
Ho
wev
er
,
Vir
u
s
T
o
tal
is
u
tili
ze
d
o
n
ly
d
u
r
i
n
g
th
e
d
ataset
lab
elin
g
a
n
d
v
alid
atio
n
p
h
ase,
n
o
t
d
u
r
in
g
th
e
r
ea
l
-
ti
m
e
d
etec
tio
n
p
r
o
ce
s
s
.
T
h
e
r
ef
o
r
e,
t
h
e
m
o
d
el
d
o
es
n
o
t r
ely
o
n
an
y
ex
ter
n
al
t
h
ir
d
-
p
ar
ty
s
er
v
ices d
u
r
i
n
g
d
e
p
lo
y
m
en
t,
en
s
u
r
in
g
in
d
ep
e
n
d
en
ce
,
ef
f
icien
cy
,
an
d
f
aster
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
A
mu
lti
-
mo
d
a
l fra
mewo
r
k
fo
r
imp
r
o
vin
g
th
e
a
cc
u
r
a
cy
o
f p
h
i
s
h
in
g
ema
il d
etec
tio
n
(
La
mee
s
Mo
h
a
med
F
a
r
a
j
)
1613
r
esp
o
n
s
e
tim
e
in
r
ea
l
-
wo
r
ld
ap
p
licatio
n
s
.
W
h
ile
th
e
p
r
o
p
o
s
ed
PECM
an
aly
ze
s
ea
ch
em
ail
co
m
p
o
n
e
n
t
s
ep
ar
ately
,
th
e
f
in
al
class
if
icatio
n
d
ec
is
io
n
is
m
a
d
e
u
s
in
g
a
m
aj
o
r
ity
v
o
tin
g
en
s
e
m
b
le
m
ec
h
a
n
is
m
.
Sp
ec
if
ically
,
s
ep
ar
ate
class
if
ier
s
ar
e
tr
ain
e
d
f
o
r
th
e
h
ea
d
er
,
b
o
d
y
,
an
d
attac
h
m
en
t
f
ea
tu
r
es,
an
d
ea
ch
class
if
ier
p
r
o
d
u
ce
s
a
b
in
ar
y
o
u
tp
u
t
(
p
h
i
s
h
in
g
o
r
le
g
itima
te)
.
T
h
e
f
in
al
d
ec
is
io
n
is
d
eter
m
in
ed
b
y
th
e
class
th
at
r
ec
eiv
es
th
e
m
ajo
r
ity
o
f
v
o
tes.
{
,
,
:
>
ℎ
>
,
}
=
T
h
is
en
s
em
b
le
ap
p
r
o
ac
h
was
ch
o
s
en
d
u
e
to
its
ea
s
e
o
f
im
p
lem
en
tatio
n
,
clea
r
in
ter
p
r
eta
b
ilit
y
,
an
d
ef
f
ec
tiv
en
ess
in
in
teg
r
atin
g
d
i
v
er
s
e
f
ea
tu
r
e
s
ets.
T
h
e
p
r
o
p
o
s
ed
PECM
i
s
s
h
o
wn
in
Fig
u
r
e
3
,
an
d
a
d
etailed
d
escr
ip
tio
n
o
f
ea
ch
s
tep
is
p
r
o
v
id
ed
in
t
h
e
f
o
llo
win
g
s
u
b
s
ec
tio
n
.
Fig
u
r
e
3
.
Ph
is
h
in
g
em
ail
class
if
ier
m
o
d
el
(
PECM)
3
.1
.
Da
t
a
prepro
ce
s
s
ing
I
n
o
r
d
e
r
to
en
s
u
r
e
th
at
th
e
d
at
a
is
clea
n
an
d
ap
p
r
o
p
r
iate
f
o
r
an
aly
s
is
,
d
ata
p
r
ep
ar
atio
n
is
an
ess
en
tial
s
tep
in
d
ev
elo
p
in
g
th
e
s
u
g
g
ested
PECM.
Un
n
ec
ess
ar
y
wo
r
d
s
an
d
ch
ar
ac
ter
s
ar
e
r
em
o
v
ed
d
u
r
in
g
th
is
p
r
o
ce
s
s
.
On
ce
clea
n
ed
,
f
o
r
m
al
f
ea
tu
r
e
ex
tr
ac
tio
n
tec
h
n
iq
u
es
ca
n
b
e
a
p
p
lied
.
T
h
is
s
tu
d
y
u
tili
ze
d
s
ev
er
al
p
r
e
p
r
o
ce
s
s
in
g
tech
n
iq
u
es,
in
clu
d
in
g
to
k
en
iza
tio
n
,
s
to
p
-
wo
r
d
r
em
o
v
al,
a
n
d
e
m
ail
clea
n
in
g
.
E
ac
h
m
eth
o
d
is
d
escr
ib
ed
b
el
o
w:
3
.
1
.
1
.
Rem
o
v
a
l o
f
white
s
pa
ce
C
lean
tex
t
is
ess
en
tial
f
o
r
m
ac
h
in
e
lear
n
in
g
m
o
d
els
an
d
t
y
p
ically
c
o
n
s
is
ts
o
f
to
k
en
s
o
r
a
lis
t
o
f
wo
r
d
s
.
T
h
is
p
r
o
ce
s
s
in
v
o
lv
es
co
n
v
e
r
tin
g
r
aw
tex
t
in
t
o
a
s
tr
u
ctu
r
ed
lis
t
o
f
wo
r
d
s
.
A
s
im
p
le
ap
p
r
o
ac
h
to
ac
h
iev
e
th
is
is
to
s
p
lit
t
h
e
tex
t
b
ased
o
n
wh
ites
p
ac
e,
in
clu
d
i
n
g
s
p
ac
es,
n
ewlin
es,
an
d
tab
s
.
I
n
Py
t
h
o
n
,
th
is
ca
n
b
e
d
o
n
e
u
s
in
g
th
e
(
)
f
u
n
ctio
n
o
n
th
e
lo
ad
ed
s
tr
in
g
.
3
.
1
.
2
.
Rem
o
v
a
l o
f
pu
nct
ua
t
io
n
T
o
r
em
o
v
e
p
u
n
ctu
atio
n
,
a
p
r
e
d
ef
in
ed
s
tr
in
g
o
f
p
u
n
ctu
atio
n
ch
ar
ac
ter
s
is
u
tili
ze
d
.
A
f
o
r
lo
o
p
iter
ates
th
r
o
u
g
h
th
e
p
r
o
v
id
ed
tex
t,
a
n
d
ea
ch
ch
ar
ac
ter
is
ch
ec
k
ed
f
o
r
m
em
b
er
s
h
ip
in
th
e
p
u
n
ctu
atio
n
s
tr
in
g
.
I
f
th
e
ch
ar
ac
ter
is
n
o
t
p
u
n
ctu
atio
n
,
i
t
is
co
n
ca
ten
ated
to
a
n
em
p
ty
s
tr
in
g
.
T
h
is
r
esu
lts
in
a
clea
n
ed
tex
t
s
tr
in
g
with
p
u
n
ctu
atio
n
r
em
o
v
ed
.
3
.
1
.
3
.
T
o
k
eniza
t
io
n
T
o
k
en
izatio
n
is
th
e
p
r
o
ce
s
s
o
f
d
iv
id
in
g
ea
ch
em
ail
in
to
s
ep
ar
ate
u
n
its
,
k
n
o
wn
as
to
k
en
s
,
ty
p
ically
b
y
s
p
litt
in
g
th
e
tex
t b
ased
o
n
wh
i
tesp
ac
e.
I
n
th
is
wo
r
k
,
th
is
s
tep
is
im
p
lem
en
ted
u
s
in
g
t
h
e
(
)
f
u
n
ctio
n
.
3
.
1
.
4
.
Rem
o
v
a
l o
f
s
t
o
p wo
rds
Sto
p
wo
r
d
s
ar
e
wo
r
d
s
th
at
h
o
ld
litt
le
s
ig
n
if
ican
ce
in
s
ea
r
ch
q
u
er
ies
o
r
tex
t
an
aly
s
is
.
Mo
s
t
s
ea
r
ch
en
g
in
es
ar
e
p
r
o
g
r
am
m
ed
to
d
is
r
eg
ar
d
s
to
p
wo
r
d
s
.
T
h
ese
wo
r
d
s
f
r
eq
u
e
n
tly
o
cc
u
r
in
tex
t
an
d
m
ain
ly
in
clu
d
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
1
6
,
No
.
3
,
J
u
n
e
20
2
6
:
1
6
0
8
-
1
6
2
5
1614
co
n
ju
n
ctio
n
s
,
ar
ticles,
p
r
ep
o
s
itio
n
s
,
an
d
o
th
er
f
u
n
ctio
n
al
wo
r
d
s
th
at
s
u
p
p
o
r
t
th
e
s
tr
u
ctu
r
e
o
f
a
s
en
ten
ce
b
u
t
lack
s
tan
d
alo
n
e
m
ea
n
in
g
.
T
ab
l
e
1
illu
s
tr
ates sam
p
le
o
f
s
to
p
wo
r
d
s
.
T
ab
le
1
.
Sam
p
le
o
f
s
to
p
wo
r
d
s
W
o
r
d
s
my
se
l
f
t
h
e
i
r
s
l
e
t
’
s
w
i
t
h
an
b
e
i
n
g
at
i
n
t
o
c
o
u
l
d
a
t
h
a
t
’
s
b
e
t
w
e
e
n
w
h
a
t
’
s
t
h
e
w
h
o
’
s
a
g
a
i
n
st
a
b
o
u
t
b
u
t
at
b
e
f
o
r
e
t
h
r
o
u
g
h
f
o
r
by
d
u
r
i
n
g
3
.
1
.
5
.
Ste
mm
ing
Stem
m
in
g
aim
s
to
r
e
d
u
ce
v
ar
iatio
n
s
in
in
f
lecte
d
f
o
r
m
s
o
f
a
wo
r
d
b
y
tr
an
s
f
o
r
m
i
n
g
th
em
in
to
th
eir
r
o
o
t
f
o
r
m
.
T
h
er
e
ar
e
two
p
r
i
m
ar
y
m
eth
o
d
s
f
o
r
co
m
p
letin
g
th
is
p
r
o
ce
s
s
:
th
e
Po
r
ter
s
tem
m
in
g
alg
o
r
ith
m
an
d
th
e
d
ictio
n
ar
y
-
b
ased
m
eth
o
d
.
I
n
th
e
d
ataset
u
s
ed
f
o
r
th
is
s
tu
d
y
,
n
u
m
e
r
o
u
s
im
ag
es
w
er
e
p
r
esen
t
in
v
ar
io
u
s
f
o
r
m
ats,
with
ap
p
r
o
x
im
ately
6
0
-
7
0
%
in
GI
F
f
o
r
m
at.
T
o
e
n
h
an
ce
an
d
p
r
ep
r
o
ce
s
s
th
e
d
ataset,
th
e
GI
F
im
ag
es
wer
e
h
an
d
led
b
y
ex
tr
ac
tin
g
th
eir
in
itial
f
r
a
m
e
an
d
co
n
v
e
r
tin
g
it
in
to
PNG
f
o
r
m
at.
T
h
e
p
r
o
ce
d
u
r
e
f
o
ll
o
wed
to
ac
co
m
p
lis
h
th
is
is
o
u
tlin
ed
b
elo
w:
a.
C
o
n
v
er
s
io
n
to
PNG
f
o
r
m
at
All
.
GI
F
im
ag
es
wer
e
tr
an
s
f
o
r
m
ed
in
to
.
PNG
f
o
r
m
at
to
e
n
s
u
r
e
co
n
s
is
ten
t
p
r
o
ce
s
s
in
g
.
Fro
m
ea
ch
.
GI
F
f
ile,
o
n
ly
th
e
f
ir
s
t
f
r
am
e
was
ex
tr
ac
ted
an
d
r
etain
ed
as
th
e
s
p
am
im
ag
e,
wh
ile
th
e
r
e
m
ain
in
g
f
r
am
es
th
at
d
id
n
o
t c
o
n
tain
s
p
am
-
r
elate
d
i
n
f
o
r
m
atio
n
wer
e
d
is
ca
r
d
e
d
.
b.
R
em
o
v
al
o
f
co
r
r
u
p
ted
f
iles
An
y
co
r
r
u
p
ted
im
a
g
e
f
iles
wer
e
id
en
tifie
d
an
d
r
em
o
v
ed
f
r
o
m
th
e
d
ataset.
T
o
im
p
lem
e
n
t
th
ese
s
tep
s
,
p
r
o
o
f
-
of
-
co
n
ce
p
t
s
cr
ip
ts
wer
e
ex
ec
u
ted
.
C
u
s
to
m
B
ash
s
cr
ip
ts
wer
e
d
ev
el
o
p
ed
to
e
x
ec
u
te
th
ese
task
s
an
d
r
ec
o
r
d
th
e
o
u
tp
u
t
at
ea
ch
s
tag
e,
en
s
u
r
i
n
g
th
e
d
ataset
o
f
s
p
am
im
ag
es
was
p
r
o
p
er
ly
clea
n
ed
an
d
en
h
an
ce
d
.
C
o
n
s
eq
u
en
tly
,
th
e
tar
g
et
v
ar
ia
b
le
f
o
r
lab
elin
g
was
en
co
d
e
d
i
n
to
a
b
in
ar
y
f
o
r
m
at,
wh
er
e
s
p
am
was
ass
ig
n
ed
a
v
alu
e
o
f
1
an
d
HAM
was a
s
s
i
g
n
ed
a
v
alu
e
o
f
0
.
Fo
r
th
is
in
itial
in
v
esti
g
atio
n
,
th
is
s
tu
d
y
f
o
cu
s
es
o
n
em
ai
ls
co
n
tain
in
g
attac
h
m
en
ts
o
r
UR
L
s
to
id
en
tify
o
r
a
n
ticip
ate
h
ar
m
f
u
l
co
n
ten
t.
Ad
d
itio
n
ally
,
s
ig
n
if
ican
t
ter
m
s
th
at
f
r
eq
u
e
n
tly
a
p
p
ea
r
in
th
e
s
p
am
em
ail
d
ataset
ar
e
ex
am
in
ed
.
Fig
u
r
e
4
illu
s
tr
ates
th
e
m
o
s
t
co
m
m
o
n
s
p
am
-
r
elate
d
wo
r
d
s
f
r
o
m
th
e
c
o
m
b
in
e
d
d
atasets
,
wh
ich
will h
elp
in
id
en
tify
in
g
s
p
am
em
ails
.
Fig
u
r
e
4
.
T
h
e
m
o
s
t f
r
e
q
u
en
t s
p
am
m
in
g
w
o
r
d
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
A
mu
lti
-
mo
d
a
l fra
mewo
r
k
fo
r
imp
r
o
vin
g
th
e
a
cc
u
r
a
cy
o
f p
h
i
s
h
in
g
ema
il d
etec
tio
n
(
La
mee
s
Mo
h
a
med
F
a
r
a
j
)
1615
3
.2
.
F
e
a
t
ure
ex
t
r
a
ct
io
n
Af
ter
p
r
ep
r
o
ce
s
s
in
g
,
th
e
T
F
-
I
DF
tech
n
iq
u
e
is
ap
p
lied
f
o
r
f
ea
tu
r
e
ex
tr
ac
tio
n
.
T
e
r
m
f
r
eq
u
en
cy
(
T
F)
r
ep
r
esen
ts
h
o
w
f
r
eq
u
en
tly
a
w
o
r
d
o
cc
u
r
s
in
a
d
o
cu
m
en
t,
wh
er
ea
s
in
v
er
s
e
d
o
cu
m
e
n
t
f
r
eq
u
en
cy
(
I
DF)
r
ef
lects
h
o
w
u
n
c
o
m
m
o
n
th
at
wo
r
d
is
ac
r
o
s
s
th
e
en
tire
co
r
p
u
s
b
y
r
el
atin
g
th
e
to
tal
n
u
m
b
er
o
f
d
o
c
u
m
en
ts
to
th
o
s
e
th
at
in
clu
d
e
th
e
ter
m
.
I
n
th
is
s
tu
d
y
,
I
DF
is
ca
lcu
lated
u
s
in
g
th
e
“
T
f
id
f
T
r
an
s
f
o
r
m
er
”
m
o
d
u
le,
w
h
ich
tr
an
s
f
o
r
m
s
tex
t
in
to
a
n
u
m
er
ical
f
o
r
m
at
ap
p
r
o
p
r
iate
f
o
r
m
ac
h
i
n
e
lear
n
in
g
alg
o
r
ith
m
s
.
T
y
p
ically
,
f
r
eq
u
en
tl
y
o
cc
u
r
r
i
n
g
wo
r
d
s
ar
e
ass
ig
n
ed
v
alu
es
b
etwe
en
0
an
d
1
,
wh
e
r
e
lo
wer
v
alu
es
in
d
icate
th
at
th
e
ter
m
is
co
m
m
o
n
a
n
d
ca
r
r
ies
less
d
is
cr
im
in
ativ
e
im
p
o
r
ta
n
ce
.
T
h
is
m
ak
es
it
p
o
s
s
ib
le
f
o
r
t
h
e
alg
o
r
ith
m
s
to
r
ea
d
th
e
d
ata.
E
q
u
atio
n
(
1
)
,
wh
er
e
(
,
)
r
ep
r
esen
ts
th
e
ter
m
f
r
eq
u
e
n
cy
(
)
with
in
a
d
o
cu
m
en
t
(
)
,
ca
n
b
e
u
s
ed
t
o
d
eter
m
i
n
e
th
e
T
F
-
I
DF:
−
(
,
)
=
(
,
)
×
(
)
(
1
)
w
h
er
e
I
DF is ca
lcu
lated
b
y
th
e
(
2
)
,
g
iv
en
n
is
th
e
n
u
m
b
er
o
f
d
o
cu
m
e
n
ts
.
(
)
=
(
(
)
)
+
1
(
2
)
T
h
e
B
o
W
ap
p
r
o
ac
h
is
em
p
lo
y
ed
to
ex
tr
ac
t
f
ea
tu
r
es
f
r
o
m
tex
tu
al
d
ata,
af
ter
w
h
ich
d
if
f
er
e
n
t
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
ar
e
ap
p
l
ied
to
class
if
y
s
p
am
em
ails
.
T
h
e
em
ail
co
n
ten
t
is
co
n
v
er
ted
in
to
a
v
ec
to
r
r
ep
r
esen
tatio
n
u
s
in
g
th
e
C
o
u
n
tVe
cto
r
iz
er
f
r
o
m
th
e
Scik
it
-
le
ar
n
lib
r
ar
y
.
T
h
is
m
eth
o
d
r
ec
o
r
d
s
th
e
f
r
eq
u
en
c
y
o
f
ea
ch
wo
r
d
ap
p
ea
r
in
g
in
th
e
t
ex
t,
h
elp
in
g
h
i
g
h
lig
h
t
im
p
o
r
t
an
t
f
ea
tu
r
es
f
o
r
s
p
am
d
etec
ti
o
n
.
Sin
ce
m
ac
h
in
e
lear
n
in
g
al
g
o
r
ith
m
s
ca
n
n
o
t
p
r
o
ce
s
s
r
aw
tex
t
i
n
p
u
t
d
ir
ec
tl
y
,
s
en
ten
ce
s
a
n
d
w
o
r
d
s
in
e
m
ails
n
ee
d
to
b
e
co
n
v
er
ted
i
n
to
n
u
m
er
ical
r
e
p
r
esen
tatio
n
s
o
r
f
ea
tu
r
e
v
ec
t
o
r
s
.
T
h
is
p
r
o
ce
s
s
,
k
n
o
wn
as
f
ea
tu
r
e
ex
tr
ac
tio
n
,
in
v
o
lv
es
tr
an
s
f
o
r
m
in
g
tex
tu
al
d
ata
in
to
a
f
o
r
m
th
at
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
ca
n
u
s
e
.
I
n
th
is
p
ap
e
r
,
th
e
B
o
W
m
o
d
el
is
em
p
lo
y
e
d
as
t
h
e
f
ea
tu
r
e
ex
tr
ac
to
r
.
I
t
is
o
n
e
o
f
th
e
s
im
p
lest
an
d
m
o
s
t
wid
ely
u
s
ed
tech
n
iq
u
es
f
o
r
f
ea
t
u
r
e
ex
t
r
ac
tio
n
in
in
f
o
r
m
atio
n
r
etr
ie
v
al.
R
ath
er
th
a
n
co
n
s
id
er
in
g
th
e
o
r
d
er
o
r
p
o
s
itio
n
o
f
wo
r
d
s
i
n
s
en
ten
ce
s
,
th
e
B
o
W
m
o
d
el
f
o
cu
s
es
s
o
lely
o
n
th
e
f
r
eq
u
e
n
cy
o
f
wo
r
d
o
cc
u
r
r
e
n
ce
s
as
th
e
k
ey
f
ea
tu
r
e
f
o
r
class
if
icatio
n
.
I
n
o
u
r
ex
p
er
i
m
en
ts
,
d
if
f
er
en
t
f
ea
tu
r
e
ex
tr
ac
tio
n
m
eth
o
d
s
we
r
e
ap
p
lied
d
ep
e
n
d
in
g
o
n
t
h
e
n
a
tu
r
e
o
f
th
e
em
ail
co
m
p
o
n
en
t.
Sp
ec
if
ically
,
th
e
B
o
W
m
et
h
o
d
was
em
p
lo
y
ed
f
o
r
p
r
o
ce
s
s
in
g
h
ea
d
e
r
f
ield
s
an
d
UR
L
-
r
elate
d
tex
tu
al
in
f
o
r
m
atio
n
,
as
t
h
ese
co
m
p
o
n
en
ts
b
e
n
ef
it
f
r
o
m
s
im
p
le
to
k
e
n
o
cc
u
r
r
en
ce
r
ep
r
ese
n
tatio
n
.
I
n
co
n
tr
ast,
th
e
ter
m
f
r
e
q
u
en
c
y
–
in
v
e
r
s
e
d
o
cu
m
en
t f
r
eq
u
en
cy
(TF
-
I
DF)
m
eth
o
d
was
ap
p
lied
to
th
e
b
o
d
y
tex
t
an
d
to
tex
tu
al
co
n
ten
t
ex
tr
ac
te
d
f
r
o
m
d
o
cu
m
en
t
attac
h
m
en
ts
(
e.
g
.
,
DOC,
PDF)
to
b
etter
ca
p
tu
r
e
th
e
i
m
p
o
r
tan
ce
o
f
ter
m
s
r
elativ
e
to
th
e
e
n
tire
d
ataset.
T
h
is
s
ep
ar
atio
n
o
f
m
eth
o
d
s
en
s
u
r
es
th
at
ea
c
h
c
o
m
p
o
n
en
t’
s
u
n
iq
u
e
ch
a
r
ac
ter
is
tics
ar
e
ap
p
r
o
p
r
iately
r
ep
r
esen
ted
f
o
r
th
e
class
if
icatio
n
task
.
T
o
en
h
an
ce
s
em
an
tic
co
n
s
is
ten
cy
b
etwe
en
em
ail
co
m
p
o
n
en
ts
,
an
ad
d
itio
n
al
v
er
if
icatio
n
s
tep
was
in
tr
o
d
u
ce
d
to
ev
alu
ate
t
h
e
c
o
n
tex
tu
al
alig
n
m
e
n
t
b
etwe
en
th
e
em
ail
s
u
b
ject
an
d
th
e
a
ttach
m
en
t
f
ilen
am
e.
Key
wo
r
d
m
atch
in
g
was
p
e
r
f
o
r
m
ed
af
te
r
tex
t
n
o
r
m
aliza
tio
n
a
n
d
to
k
en
izatio
n
,
wh
er
e
co
m
m
o
n
p
h
is
h
in
g
-
r
elate
d
ter
m
s
(
e.
g
.
,
in
v
o
ice,
p
ay
m
e
n
t,
r
ec
eip
t,
d
o
cu
m
e
n
t)
ap
p
ea
r
i
n
g
in
th
e
s
u
b
ject
lin
e
wer
e
co
m
p
ar
ed
a
g
ain
s
t
attac
h
m
en
t
f
ilen
am
es.
T
h
is
lig
h
tweig
h
t
s
em
an
tic
co
n
s
is
ten
cy
ch
ec
k
c
o
n
tr
ib
u
tes
to
d
etec
tin
g
s
u
s
p
icio
u
s
em
ails
wh
er
e
m
is
lead
in
g
o
r
m
is
m
atch
ed
attac
h
m
en
ts
ar
e
e
m
p
lo
y
e
d
.
3
.
2
.
1
.
H
ea
der
f
e
a
t
ure
cla
s
s
T
h
e
em
ail
h
ea
d
er
,
wh
ich
is
ty
p
ically
n
o
t
v
is
ib
le
to
en
d
u
s
er
s
,
co
n
tain
s
im
p
o
r
tan
t
m
etad
at
a
s
u
ch
as
th
e
tr
an
s
m
is
s
io
n
p
ath
,
s
en
d
in
g
an
d
r
ec
eiv
in
g
s
er
v
er
s
,
tim
e
s
tam
p
s
,
an
d
m
ess
ag
e
f
o
r
m
at.
Fro
m
th
is
h
ea
d
er
,
s
ev
er
al
in
f
o
r
m
ativ
e
f
ea
tu
r
es
ca
n
b
e
d
er
iv
ed
,
in
clu
d
i
n
g
th
e
ex
is
ten
ce
o
f
a
m
ess
ag
e
I
D,
id
en
tific
atio
n
o
f
b
lack
lis
ted
ter
m
s
in
th
e
s
u
b
je
ct
lin
e,
wh
eth
er
th
e
m
ess
ag
e
is
in
p
lain
-
tex
t
f
o
r
m
at,
co
n
s
is
ten
cy
b
etwe
en
th
e
s
en
d
er
an
d
r
etu
r
n
-
p
ath
d
o
m
ain
s
,
d
is
cr
ep
an
cies
in
I
P
ad
d
r
ess
es
ac
r
o
s
s
s
u
cc
ess
iv
e
R
ec
eiv
ed
f
ield
s
,
a
n
d
v
er
if
icatio
n
u
s
in
g
Do
m
ain
Ke
y
s
I
d
en
tifie
d
Ma
il
(
DKI
M
)
.
A
s
u
b
s
et
o
f
th
ese
attr
ib
u
tes
was
ex
tr
ac
ted
f
r
o
m
th
e
em
ail
h
ea
d
er
,
a
n
d
th
eir
d
etailed
d
escr
ip
tio
n
s
ar
e
p
r
o
v
id
ed
in
T
ab
le
2
.
3
.
2
.
2
.
B
o
dy
f
ea
t
ures c
la
s
s
E
m
ail
co
n
ten
t
m
ay
co
n
tain
U
R
L
s
an
d
b
e
d
is
p
lay
ed
in
a
v
ar
iety
o
f
f
o
r
m
s
,
in
clu
d
i
n
g
HT
M
L
an
d
tex
t.
C
o
n
s
eq
u
en
tly
,
th
e
b
o
d
y
class
is
d
iv
id
ed
in
t
o
two
s
u
b
class
es:
th
e
UR
L
s
u
b
class
an
d
th
e
te
x
t
s
u
b
class
,
wh
ich
in
clu
d
es scr
ip
ts
an
d
tex
t c
o
n
te
n
t e
x
tr
ac
ted
f
r
o
m
HT
ML
.
a.
T
ex
t f
ea
tu
r
e
class
T
ex
t
an
aly
s
is
an
d
an
e
x
am
in
at
io
n
o
f
th
e
em
ail'
s
HT
ML
co
n
ten
t
ca
n
b
e
u
s
ed
to
ex
tr
ac
t
th
es
e
f
ea
tu
r
es.
E
x
am
p
les
in
clu
d
e
th
e
u
s
ag
e
o
f
d
if
f
icu
lt
p
h
r
ases
in
th
e
tex
t,
th
e
f
r
eq
u
en
cy
o
r
p
r
esen
ce
o
f
tag
s
lik
e
“
h
r
e
f
”
an
d
“
o
n
m
o
u
s
eo
v
er
,
”
a
n
d
t
h
e
o
cc
u
r
r
en
ce
o
f
wo
r
d
s
th
at
ar
e
o
n
a
b
lack
lis
t
o
r
th
at
elicit
an
ac
tio
n
.
T
h
e
p
r
im
ar
y
tex
t
th
at
th
e
r
ea
d
er
m
ay
s
ee
is
lo
ca
ted
in
a
s
ep
ar
ate
ar
ea
o
f
th
e
e
m
ail
b
o
d
y
.
T
h
is
s
tag
e
in
v
o
lv
e
s
ex
tr
ac
tin
g
th
e
tex
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
1
6
,
No
.
3
,
J
u
n
e
20
2
6
:
1
6
0
8
-
1
6
2
5
1616
th
at
th
e
em
ail
r
ec
ip
ien
t
ca
n
s
e
e
f
r
o
m
th
e
co
n
ten
t.
So
m
e
o
f
t
h
ese
f
ea
tu
r
es
wer
e
tak
en
f
r
o
m
t
h
e
em
ail
b
o
d
y
,
a
n
d
T
ab
le
3
lis
ts
th
eir
d
escr
ip
tio
n
s
.
T
ab
le
2
.
E
x
am
p
le
o
f
s
elec
ted
h
ea
d
er
f
ea
t
u
r
es.
F
e
a
t
u
r
e
F
e
a
t
u
r
e
D
e
scri
p
t
i
o
n
H
0
1
Th
i
s
f
e
a
t
u
r
e
i
s
u
se
d
a
s a
b
i
n
a
r
y
i
n
d
i
c
a
t
o
r
t
o
d
e
t
e
r
mi
n
e
w
h
e
t
h
e
r
t
h
e
d
o
m
a
i
n
n
a
mes
d
e
r
i
v
e
d
f
r
o
m
t
h
e
e
ma
i
l
se
n
d
e
r
m
a
t
c
h
t
h
o
se
f
r
o
m
t
h
e
m
e
ssa
g
e
I
D
b
y
c
o
mp
a
r
i
n
g
t
h
e
se
n
d
e
r
's
d
o
ma
i
n
w
i
t
h
t
h
e
d
o
m
a
i
n
f
r
o
m
t
h
e
mess
a
g
e
I
D
.
H
0
2
A
b
i
n
a
r
y
f
e
a
t
u
r
e
t
h
a
t
i
n
d
i
c
a
t
e
s
i
f
t
h
e
c
o
n
t
e
n
t
t
y
p
e
o
f
t
h
e
e
m
a
i
l
i
s se
t
t
o
H
T
M
L
o
r
Te
x
t
.
H
0
3
A
b
i
n
a
r
y
f
e
a
t
u
r
e
t
h
a
t
d
e
t
e
r
mi
n
e
s
i
f
t
h
e
e
ma
i
l
's
h
e
a
d
e
r
s
p
e
c
i
f
i
e
s
t
e
x
t
o
r
p
l
a
i
n
a
s t
h
e
c
o
n
t
e
n
t
t
y
p
e
.
H
0
4
D
e
t
e
r
m
i
n
e
s
h
o
w
ma
n
y
r
e
c
i
p
i
e
n
t
s a
r
e
l
i
st
e
d
i
n
t
h
e
e
m
a
i
l
h
e
a
d
e
r
's “
f
r
o
m”.
H
0
5
C
h
e
c
k
s
i
f
t
h
e
t
e
r
m
“
b
a
n
k
”
a
p
p
e
a
r
s i
n
t
h
e
e
ma
i
l
su
b
j
e
c
t
a
n
d
r
e
t
u
r
n
s
“
t
r
u
e
”
i
f
i
t
d
o
e
s
o
r
“
f
a
l
se
”
o
t
h
e
r
w
i
s
e
.
H
0
6
C
h
e
c
k
s
i
f
t
h
e
t
e
r
m
“
d
e
b
i
t
”
a
p
p
e
a
r
s
i
n
t
h
e
e
m
a
i
l
s
u
b
j
e
c
t
a
n
d
r
e
t
u
r
n
s “
t
r
u
e
”
i
f
i
t
d
o
e
s
o
r
“
f
a
l
s
e
”
o
t
h
e
r
w
i
se
.
H
0
7
D
e
t
e
r
m
i
n
e
s i
f
t
h
e
w
o
r
d
“
F
w
d
:
”
a
p
p
e
a
r
s i
n
t
h
e
e
ma
i
l
su
b
j
e
c
t
a
n
d
r
e
t
u
r
n
s
“
t
r
u
e
”
i
f
i
t
d
o
e
s
o
r
“
f
a
l
s
e
”
o
t
h
e
r
w
i
se.
.
H
0
8
C
h
e
c
k
s
t
o
see
i
f
t
h
e
w
o
r
d
“
R
e
:
”
a
p
p
e
a
r
s i
n
t
h
e
e
ma
i
l
su
b
j
e
c
t
a
n
d
r
e
t
u
r
n
s
“
t
r
u
e
”
i
f
i
t
d
o
e
s
o
r
“
f
a
l
s
e
”
o
t
h
e
r
w
i
se.
H
0
9
I
d
e
n
t
i
f
i
e
s
w
h
e
t
h
e
r
t
h
e
w
o
r
d
“
v
e
r
i
f
y
”
a
p
p
e
a
r
s
i
n
t
h
e
e
m
a
i
l
s
u
b
j
e
c
t
a
n
d
r
e
t
u
r
n
s “
t
r
u
e
”
i
f
i
t
d
o
e
s
o
r
“
f
a
l
se
”
o
t
h
e
r
w
i
s
e
.
H
1
0
D
e
t
e
r
m
i
n
e
s
h
o
w
ma
n
y
c
h
a
r
a
c
t
e
r
s t
h
e
r
e
a
r
e
i
n
t
h
e
s
u
b
j
e
c
t
a
r
e
a
a
n
d
g
i
v
e
s
b
a
c
k
t
h
e
a
p
p
r
o
p
r
i
a
t
e
n
u
m
b
e
r
.
H
1
1
D
e
t
e
r
m
i
n
e
s t
h
e
t
o
t
a
l
w
o
r
d
c
o
u
n
t
i
n
t
h
e
su
b
j
e
c
t
f
i
e
l
d
a
n
d
p
r
o
v
i
d
e
s
t
h
e
o
u
t
c
o
me.
H
1
2
C
h
e
c
k
s
t
o
see
i
f
t
h
e
e
ma
i
l
se
n
d
e
r
ma
t
c
h
e
s t
h
e
“
R
e
p
l
y
To
”
f
i
e
l
d
;
i
f
n
o
t
,
r
e
t
u
r
n
s (f
a
l
se);
o
t
h
e
r
w
i
se
,
r
e
t
u
r
n
s (
t
r
u
e
)
.
H
1
3
D
e
t
e
r
m
i
n
e
s
h
o
w
ma
n
y
r
e
c
i
p
i
e
n
t
s a
r
e
l
i
st
e
d
i
n
t
h
e
e
m
a
i
l
h
e
a
d
e
r
's “
T
o
:
”
f
i
e
l
d
.
H
1
4
D
e
t
e
r
m
i
n
e
s
h
o
w
ma
n
y
r
e
c
i
p
i
e
n
t
s a
r
e
l
i
st
e
d
i
n
t
h
e
e
m
a
i
l
h
e
a
d
e
r
's “
C
c
:
”
f
i
e
l
d
.
H
1
5
D
e
t
e
r
m
i
n
e
s
h
o
w
ma
n
y
p
e
o
p
l
e
a
r
e
l
i
s
t
e
d
i
n
t
h
e
e
m
a
i
l
h
e
a
d
e
r
's “
B
c
c
:
”
f
i
e
l
d
.
T
ab
le
3
.
E
x
am
p
le
o
f
s
elec
ted
b
o
d
y
c
o
n
ten
t
f
ea
tu
r
es
F
e
a
t
u
r
e
F
e
a
t
u
r
e
D
e
scri
p
t
i
o
n
B
0
1
C
h
e
c
k
s
t
o
see
w
h
e
t
h
e
r
t
h
e
r
e
i
s
a
n
H
T
M
L
f
o
r
m e
l
e
m
e
n
t
i
n
t
h
e
e
m
a
i
l
m
e
ssa
g
e
.
B
0
2
c
h
e
c
k
s t
o
s
e
e
w
h
e
t
h
e
r
t
h
e
r
e
i
s
a
Ja
v
a
S
c
r
i
p
t
p
o
p
-
u
p
i
n
t
h
e
e
mai
l
.
B
0
3
D
e
t
e
r
m
i
n
e
s
h
o
w
ma
n
y
p
i
c
t
u
r
e
s a
r
e
u
s
e
d
a
s
h
y
p
e
r
l
i
n
k
s.
B
0
4
D
e
t
e
r
m
i
n
e
s w
h
e
t
h
e
r
t
h
e
d
o
ma
i
n
n
a
me
s h
a
v
e
a
c
o
r
r
e
sp
o
n
d
i
n
g
r
e
v
e
r
se
D
N
S
r
e
c
o
r
d
;
i
f
s
o
,
i
t
r
e
t
u
r
n
s (t
r
u
e
)
;
i
f
n
o
t
,
i
t
r
e
t
u
r
n
s
(
f
a
l
se)
.
B
0
5
A
f
u
n
c
t
i
o
n
t
h
a
t
r
e
t
u
r
n
s (t
r
u
e
)
i
f
a
n
e
m
a
i
l
c
o
n
t
a
i
n
s a
o
n
C
l
i
c
k
Ja
v
a
S
c
r
i
p
t
e
v
e
n
t
a
n
d
(
f
a
l
s
e
)
o
t
h
e
r
w
i
s
e
.
B
0
6
R
e
t
u
r
n
s
t
h
e
e
m
a
i
l
m
e
ssa
g
e
's
b
y
t
e
s
i
z
e
.
B
0
7
b
i
n
a
r
y
f
e
a
t
u
r
e
c
h
e
c
k
s f
o
r
t
h
e
o
c
c
u
r
r
e
n
c
e
o
f
t
h
e
w
o
r
d
“
D
e
a
r
”
i
n
t
h
e
e
ma
i
l
b
o
d
y
,
a
s
si
g
n
i
n
g
a
v
a
l
u
e
o
f
t
r
u
e
i
f
i
t
i
s
p
r
e
se
n
t
a
n
d
f
a
l
se
i
f
i
t
i
s
n
o
t
.
B
0
8
D
e
t
e
r
m
i
n
e
s t
h
e
e
m
a
i
l
b
o
d
y
's
c
h
a
r
a
c
t
e
r
c
o
u
n
t
.
B
0
9
D
e
t
e
r
m
i
n
e
s
h
o
w
ma
n
y
w
o
r
d
s
a
r
e
i
n
t
h
e
b
o
d
y
o
f
t
h
e
e
ma
i
l
.
B
1
0
d
e
t
e
r
mi
n
e
s
h
o
w
m
a
n
y
o
f
t
h
e
s
p
e
c
i
f
i
e
d
k
e
y
w
o
r
d
s
a
r
e
i
n
t
h
e
b
o
d
y
o
f
t
h
e
e
ma
i
l
.
A
c
c
o
u
n
t
,
c
l
i
c
k
,
i
d
e
n
t
i
f
y
,
i
n
c
o
n
v
e
n
i
e
n
t
,
i
n
f
o
r
mat
i
o
n
,
l
i
mi
t
e
d
,
l
o
g
,
m
i
n
u
t
e
s,
p
a
s
sw
o
r
d
,
r
e
c
e
n
t
l
y
,
r
i
s
k
,
s
o
c
i
a
l
,
sec
u
r
i
t
y
,
serv
i
c
e
,
a
n
d
s
u
sp
e
n
d
e
d
a
r
e
a
f
e
w
e
x
a
mp
l
e
s
o
f
t
h
e
se
t
e
r
ms
.
B
1
1
C
h
e
c
k
s
t
h
e
e
m
a
i
l
b
o
d
y
f
o
r
t
h
e
w
o
r
d
“
su
sp
e
n
si
o
n
,
”
r
e
t
u
r
n
i
n
g
t
r
u
e
i
f
i
t
i
s
p
r
e
s
e
n
t
a
n
d
f
a
l
se
i
f
i
t
i
s
n
o
t
.
B
1
2
Th
e
p
h
r
a
s
e
“
v
e
r
i
f
y
y
o
u
r
a
c
c
o
u
n
t
”
i
s c
h
e
c
k
e
d
f
o
r
i
n
t
h
e
e
ma
i
l
t
e
x
t
u
si
n
g
a
b
i
n
a
r
y
i
n
d
i
c
a
t
o
r
,
w
h
i
c
h
r
e
t
u
r
n
s
t
r
u
e
i
f
i
t
i
s
p
r
e
se
n
t
a
n
d
f
a
l
se
o
t
h
e
r
w
i
se
.
b.
E
m
ail
UR
L
an
d
DOC f
ea
tu
r
e
class
es
E
m
ails
m
ay
co
n
tain
a
v
ar
iab
le
n
u
m
b
e
r
o
f
UR
L
s
,
n
ec
ess
itatin
g
th
e
ag
g
r
eg
atio
n
o
f
UR
L
ch
ar
ac
ter
is
tics
to
p
r
ev
e
n
t
f
ea
tu
r
e
v
ec
to
r
s
with
in
co
n
s
is
ten
t
len
g
th
s
.
Ag
g
r
eg
ated
f
ea
tu
r
es
co
llect
s
p
ec
if
ic
in
f
o
r
m
atio
n
f
r
o
m
ev
er
y
UR
L
in
an
em
ail
an
d
p
r
o
v
id
e
co
n
ti
n
u
o
u
s
o
r
B
o
o
lean
r
esu
lts
.
T
h
e
n
u
m
b
er
o
f
UR
L
s
u
s
in
g
a
s
ec
u
r
e
co
n
n
ec
tio
n
,
th
e
n
u
m
b
e
r
o
f
UR
L
s
h
av
in
g
an
I
P
ad
d
r
ess
,
an
d
th
e
n
u
m
b
e
r
o
f
UR
L
s
with
d
o
m
ain
s
o
th
er
th
an
th
e
s
en
d
er
'
s
ad
d
r
ess
ar
e
a
f
ew
ex
am
p
les.
T
ab
les
4
an
d
T
ab
le
5
p
r
o
v
i
d
e
th
e
co
m
p
lete
lis
t
o
f
f
ea
tu
r
es
ex
tr
ac
ted
f
r
o
m
UR
L
s
an
d
d
o
cu
m
en
t
attac
h
m
en
ts
(
DOCs
)
,
r
esp
ec
tiv
ely
.
T
h
ese
f
ea
tu
r
es
wer
e
u
s
ed
in
th
e
ex
p
er
im
en
tal
s
etu
p
o
f
th
e
p
r
o
p
o
s
ed
PECM f
o
r
tr
ain
in
g
an
d
e
v
alu
atio
n
.
T
ab
le
4
.
E
x
am
p
le
o
f
s
elec
ted
UR
L
f
ea
tu
r
es
F
e
a
t
u
r
e
F
e
a
t
u
r
e
D
e
scri
p
t
i
o
n
U
R
L
L
e
n
g
t
h
To
t
a
l
n
u
m
b
e
r
o
f
c
h
a
r
a
c
t
e
r
s
i
n
t
h
e
U
R
L
N
u
mb
e
r
o
f
D
o
t
s
C
o
u
n
t
o
f
“
.
”
c
h
a
r
a
c
t
e
r
s i
n
t
h
e
U
R
L
P
r
e
sen
c
e
o
f
I
P
A
d
d
r
e
ss
B
o
o
l
e
a
n
f
l
a
g
i
n
d
i
c
a
t
i
n
g
i
f
t
h
e
U
R
L
c
o
n
t
a
i
n
s a
n
I
P
a
d
d
r
e
ss i
n
s
t
e
a
d
o
f
a
d
o
m
a
i
n
N
u
mb
e
r
o
f
S
u
b
d
o
m
a
i
n
s
C
o
u
n
t
o
f
s
u
b
d
o
ma
i
n
se
g
m
e
n
t
s
i
n
t
h
e
U
R
L.
H
TTPS
U
sa
g
e
B
o
o
l
e
a
n
f
l
a
g
i
n
d
i
c
a
t
i
n
g
i
f
t
h
e
U
R
L
s
t
a
r
t
s wit
h
H
TTPS
S
p
e
c
i
a
l
C
h
a
r
a
c
t
e
r
s
C
o
u
n
t
N
u
mb
e
r
o
f
sp
e
c
i
a
l
c
h
a
r
a
c
t
e
r
s (@
,
#
,
?
,
=
,
e
t
c
.
)
S
u
sp
i
c
i
o
u
s
K
e
y
w
o
r
d
P
r
e
se
n
c
e
B
o
o
l
e
a
n
f
l
a
g
i
n
d
i
c
a
t
i
n
g
i
f
t
h
e
U
R
L
c
o
n
t
a
i
n
s s
u
s
p
i
c
i
o
u
s
k
e
y
w
o
r
d
s
(
e
.
g
.
,
“
l
o
g
i
n
”
,
“
v
e
r
i
f
y
”
,
“
u
p
d
a
t
e
”
)
U
R
L
E
n
t
r
o
p
y
M
e
a
su
r
e
o
f
r
a
n
d
o
m
n
e
ss
i
n
t
h
e
U
R
L
st
r
i
n
g
R
e
d
i
r
e
c
t
i
o
n
C
o
u
n
t
N
u
mb
e
r
o
f
r
e
d
i
r
e
c
t
i
o
n
s
t
h
e
U
R
L
p
e
r
f
o
r
ms
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
A
mu
lti
-
mo
d
a
l fra
mewo
r
k
fo
r
imp
r
o
vin
g
th
e
a
cc
u
r
a
cy
o
f p
h
i
s
h
in
g
ema
il d
etec
tio
n
(
La
mee
s
Mo
h
a
med
F
a
r
a
j
)
1617
T
ab
le
5
.
ex
a
m
p
le
o
f
s
elec
ted
DOCS
f
ea
tu
r
es
F
e
a
t
u
r
e
F
e
a
t
u
r
e
D
e
scri
p
t
i
o
n
F
i
l
e
S
i
z
e
(
K
B
)
S
i
z
e
o
f
t
h
e
d
o
c
u
me
n
t
i
n
k
i
l
o
b
y
t
e
s
N
u
mb
e
r
o
f
P
a
g
e
s
To
t
a
l
p
a
g
e
c
o
u
n
t
i
n
t
h
e
d
o
c
u
m
e
n
t
W
o
r
d
C
o
u
n
t
N
u
mb
e
r
o
f
w
o
r
d
s i
n
t
h
e
d
o
c
u
me
n
t
Emb
e
d
d
e
d
L
i
n
k
s
C
o
u
n
t
N
u
mb
e
r
o
f
h
y
p
e
r
l
i
n
k
s
i
n
si
d
e
t
h
e
d
o
c
u
men
t
.
M
a
c
r
o
P
r
e
se
n
c
e
B
o
o
l
e
a
n
f
l
a
g
i
n
d
i
c
a
t
i
n
g
i
f
m
a
c
r
o
s a
r
e
e
mb
e
d
d
e
d
S
u
sp
i
c
i
o
u
s
M
a
c
r
o
K
e
y
w
o
r
d
s
B
o
o
l
e
a
n
f
l
a
g
f
o
r
s
u
s
p
i
c
i
o
u
s ma
c
r
o
k
e
y
w
o
r
d
s (
e
.
g
.
,
“
S
h
e
l
l
”
,
“
C
r
e
a
t
e
O
b
j
e
c
t
”
)
S
u
sp
i
c
i
o
u
s
K
e
y
w
o
r
d
P
r
e
se
n
c
e
B
o
o
l
e
a
n
f
l
a
g
i
n
d
i
c
a
t
i
n
g
i
f
t
h
e
U
R
L
c
o
n
t
a
i
n
s s
u
s
p
i
c
i
o
u
s
k
e
y
w
o
r
d
s
(
e
.
g
.
,
“
l
o
g
i
n
”
,
“
v
e
r
i
f
y
”
,
“
u
p
d
a
t
e
”
)
F
i
l
e
E
x
t
e
n
s
i
o
n
Ty
p
e
F
i
l
e
f
o
r
ma
t
e
x
t
e
n
si
o
n
(
e
.
g
.
,
.
d
o
c
,
.
d
o
c
x
,
.
p
d
f
)
M
e
t
a
d
a
t
a
A
n
o
ma
l
i
e
s
B
o
o
l
e
a
n
f
l
a
g
i
f
d
o
c
u
me
n
t
me
t
a
d
a
t
a
i
s
su
sp
i
c
i
o
u
s
Ex
t
e
r
n
a
l
O
b
j
e
c
t
E
mb
e
d
d
i
n
g
B
o
o
l
e
a
n
f
l
a
g
i
f
t
h
e
f
i
l
e
c
o
n
t
a
i
n
s e
m
b
e
d
d
e
d
e
x
t
e
r
n
a
l
o
b
j
e
c
t
s
c.
E
m
ail
im
ag
e
f
ea
tu
r
e
class
C
o
lo
r
,
in
f
o
r
m
atio
n
,
s
h
ap
e
,
tex
tu
r
e,
a
n
d
n
o
is
e
ch
a
r
ac
ter
is
tics
ar
e
t
h
e
f
i
v
e
m
ai
n
ca
teg
o
r
ies
i
n
to
wh
ich
im
ag
e
f
ea
tu
r
es
ar
e
co
m
m
o
n
ly
d
iv
id
e
d
.
T
h
ese
ca
teg
o
r
ies
p
r
o
v
id
e
a
co
m
p
r
eh
e
n
s
iv
e
r
e
p
r
esen
tatio
n
o
f
d
if
f
er
e
n
t
v
is
u
al
p
r
o
p
er
ties
th
at
ca
n
b
e
ex
tr
ac
ted
f
r
o
m
im
ag
es
f
o
r
a
n
aly
s
is
.
T
ab
le
6
p
r
esen
ts
a
d
et
ailed
o
v
er
v
iew
o
f
s
ev
er
al
r
ep
r
esen
tativ
e
f
ea
tu
r
e
g
r
o
u
p
s
alo
n
g
with
th
ei
r
co
r
r
esp
o
n
d
in
g
d
escr
ip
tio
n
s
.
T
ab
le
6
.
E
x
am
p
le
o
f
s
elec
ted
im
ag
e
f
ea
tu
r
es
F
e
a
t
u
r
e
F
e
a
t
u
r
e
D
e
scri
p
t
i
o
n
M
e
t
a
d
a
t
a
F
e
a
t
u
r
e
Th
i
s
f
e
a
t
u
r
e
i
n
c
l
u
d
e
s
t
h
e
i
ma
g
e
's
h
e
i
g
h
t
,
w
i
d
t
h
,
b
i
t
d
e
p
t
h
,
a
sp
e
c
t
r
a
t
i
o
a
n
d
c
o
mp
r
e
ssi
o
n
r
a
t
i
o
.
C
o
l
o
r
F
e
a
t
u
r
e
Th
i
s
f
e
a
t
u
r
e
e
n
c
o
m
p
a
sses
t
h
e
m
e
a
n
,
s
k
e
w
n
e
ss,
v
a
r
i
a
n
c
e
,
a
n
d
e
n
t
r
o
p
y
v
a
l
u
e
s
o
f
v
a
r
i
o
u
s
i
ma
g
e
a
t
t
r
i
b
u
t
e
s,
s
u
c
h
a
s R
G
B
c
o
l
o
r
c
h
a
n
n
e
l
s.
Te
x
t
u
r
e
F
e
a
t
u
r
e
Th
i
s
f
e
a
t
u
r
e
d
e
t
e
r
mi
n
e
s si
mi
l
a
r
i
t
y
a
n
d
i
n
f
o
r
ma
t
i
o
n
o
f
a
d
j
a
c
e
n
t
p
i
x
e
l
s.
S
h
a
p
e
F
e
a
t
u
r
e
Th
i
s
F
e
a
t
u
r
e
M
e
a
su
r
e
s
t
h
e
v
a
r
i
a
t
i
o
n
s
i
n
i
n
t
e
n
si
t
y
g
r
a
d
i
e
n
t
s
w
i
t
h
i
n
a
n
i
ma
g
e
.
N
o
i
se
F
e
a
t
u
r
e
Th
i
s
f
e
a
t
u
r
e
c
o
m
p
r
i
s
e
s
t
h
e
s
i
g
n
a
l
-
to
-
n
o
i
s
e
r
a
t
i
o
(
S
N
R
)
a
n
d
n
o
i
s
e
e
n
t
r
o
p
y
.
3
.3
.
M
o
del
t
ra
ini
ng
I
n
t
h
e
m
o
d
e
l
t
r
a
i
n
i
n
g
a
n
d
e
v
a
l
u
a
t
i
o
n
p
h
a
s
e
s
,
w
e
i
n
i
t
i
a
l
l
y
p
r
o
v
i
d
e
d
a
g
e
n
e
r
a
l
o
v
e
r
v
i
e
w
o
f
t
h
e
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
m
o
d
e
l
s
u
s
e
d
,
a
l
o
n
g
w
i
t
h
a
c
o
n
f
u
s
i
o
n
m
a
t
r
i
x
t
o
i
l
l
u
s
t
r
a
t
e
c
l
a
s
s
i
f
i
c
a
t
i
o
n
p
e
r
f
o
r
m
a
n
c
e
.
T
o
e
n
h
a
n
c
e
t
h
e
r
e
p
r
o
d
u
c
i
b
i
l
i
t
y
o
f
o
u
r
s
t
u
d
y
a
n
d
e
n
s
u
r
e
t
h
e
v
a
l
i
d
i
t
y
o
f
o
u
r
f
i
n
d
i
n
g
s
,
w
e
n
o
w
p
r
e
s
e
n
t
d
e
t
a
i
l
e
d
c
o
n
f
i
g
u
r
a
t
i
o
n
s
e
t
t
i
n
g
s
.
P
y
t
h
o
n
3
.
1
0
a
n
d
t
h
e
S
c
i
k
i
t
-
l
e
a
r
n
1
.
3
l
i
b
r
a
r
y
w
e
r
e
u
s
e
d
t
o
i
m
p
l
e
m
e
n
t
e
a
c
h
m
o
d
e
l
.
W
e
e
m
p
l
o
y
e
d
a
n
R
B
F
k
e
r
n
e
l
w
i
t
h
“
C
=
1
.
0
”
a
n
d
“
g
a
m
m
a
=
s
c
a
l
e
”
f
o
r
t
h
e
S
V
M
.
T
h
e
r
a
n
d
o
m
f
o
r
e
s
t
c
l
a
s
s
i
f
i
e
r
w
a
s
s
e
t
u
p
w
i
t
h
“
c
r
i
t
e
r
i
o
n
=
g
i
n
i
,
”
“
m
a
x
_
d
e
p
t
h
=
10,
”
a
n
d
“
n
_
e
s
t
i
m
a
t
o
r
s
=
100.
”
T
h
e
G
a
u
s
s
i
a
n
v
a
r
i
a
n
t
w
i
t
h
d
e
f
a
u
l
t
s
e
t
t
i
n
g
s
w
a
s
u
s
e
d
b
y
t
h
e
n
a
ï
v
e
B
a
y
e
s
m
o
d
e
l
.
T
h
e
d
a
t
a
s
e
t
w
a
s
s
p
l
i
t
i
n
t
o
8
0
%
f
o
r
t
r
a
i
n
i
n
g
a
n
d
2
0
%
f
o
r
t
e
s
t
i
n
g
.
A
d
d
i
t
i
o
n
a
l
l
y
,
a
5
-
f
o
l
d
c
r
o
s
s
-
v
a
l
i
d
a
t
i
o
n
a
p
p
r
o
a
c
h
w
a
s
u
s
e
d
t
o
e
n
s
u
r
e
r
e
l
i
a
b
l
e
e
v
a
l
u
a
t
i
o
n
.
E
a
c
h
e
x
p
e
r
i
m
e
n
t
w
a
s
r
e
p
e
a
t
e
d
t
e
n
t
i
m
e
s
,
a
n
d
t
h
e
r
e
s
u
l
t
s
,
i
n
c
l
u
d
i
n
g
t
h
e
c
o
n
f
u
s
i
o
n
m
a
t
r
i
x
,
r
e
p
r
e
s
e
n
t
t
h
e
a
v
e
r
a
g
e
a
c
r
o
s
s
t
h
e
s
e
r
u
n
s
.
T
h
i
s
m
e
t
h
o
d
o
l
o
g
y
w
a
s
a
d
o
p
t
e
d
t
o
m
i
t
i
g
a
t
e
t
h
e
i
m
p
a
c
t
o
f
r
a
n
d
o
m
v
a
r
i
a
t
i
o
n
s
a
n
d
t
o
p
r
o
v
i
d
e
s
t
a
b
l
e
p
e
r
f
o
r
m
a
n
c
e
m
e
t
r
i
c
s
.
I
n
t
h
e
p
r
o
p
o
s
e
d
P
E
C
M
,
a
n
e
m
a
i
l
u
n
d
e
r
g
o
i
n
g
c
l
a
s
s
i
f
i
c
a
t
i
o
n
i
s
d
i
v
i
d
e
d
i
n
t
o
i
t
s
t
h
r
e
e
p
r
i
m
a
r
y
s
e
g
m
e
n
t
s
:
t
h
e
h
e
a
d
e
r
,
b
o
d
y
a
n
d
a
t
t
a
c
h
m
e
n
t
s
.
E
a
c
h
o
f
t
h
e
s
e
s
e
g
m
e
n
t
s
i
s
t
r
a
i
n
e
d
u
s
i
n
g
a
d
i
s
t
i
n
c
t
c
l
a
s
s
i
f
i
e
r
.
3
.
3
.
1
.
P
his
hin
g
em
a
il c
la
s
s
if
i
ca
t
io
n ba
s
ed
o
n t
he
hea
der
o
f
a
n e
m
a
il
E
m
ail
m
etad
ata,
s
u
ch
as
th
e
s
en
d
er
'
s
I
P
ad
d
r
ess
,
co
n
te
n
t
ty
p
e,
r
ec
ip
ien
t'
s
em
ail
ad
d
r
e
s
s
,
MI
ME
v
er
s
io
n
,
an
d
o
th
er
attr
ib
u
tes,
is
s
to
r
ed
in
th
e
em
ail
h
ea
d
er
.
T
h
e
s
ize
o
f
th
e
h
ea
d
er
a
n
d
th
e
n
u
m
b
er
o
f
ar
g
u
m
en
ts
it c
o
n
tain
s
ca
n
v
ar
y
s
ig
n
if
ican
tly
b
etwe
en
em
ails
.
C
las
s
if
ier
:
T
h
e
r
an
d
o
m
f
o
r
est
alg
o
r
ith
m
is
u
s
ed
f
o
r
h
ea
d
er
cl
ass
if
icatio
n
.
a.
R
an
d
o
m
f
o
r
est
(
R
F)
R
F
alg
o
r
ith
m
is
u
s
ed
to
clas
s
if
y
em
ail
h
ea
d
er
s
.
T
h
e
r
a
n
d
o
m
f
o
r
est
C
lass
if
ier
is
an
en
s
em
b
le
o
f
d
ec
is
io
n
tr
ee
s
,
ea
ch
v
ar
y
i
n
g
i
n
s
ize
an
d
s
tr
u
ctu
r
e.
Du
r
i
n
g
tr
ain
in
g
,
th
e
d
ata
is
s
am
p
led
r
a
n
d
o
m
ly
,
an
d
at
ea
ch
tr
ee
n
o
d
e,
s
u
b
s
ets
o
f
i
n
p
u
t
f
ea
tu
r
es
ar
e
s
elec
ted
r
an
d
o
m
ly
.
T
h
is
r
an
d
o
m
izatio
n
r
ed
u
ce
s
c
o
r
r
elatio
n
s
b
etwe
en
in
d
iv
id
u
al
tr
ee
s
,
r
esu
ltin
g
in
im
p
r
o
v
e
d
g
e
n
er
aliza
tio
n
an
d
r
ed
u
ce
d
en
s
em
b
le
e
r
r
o
r
.
B
y
en
s
u
r
in
g
th
at
th
e
f
ea
tu
r
es
u
s
ed
i
n
tr
ee
s
p
lits
d
if
f
er
ac
r
o
s
s
tr
ee
s
,
th
e
m
o
d
el
b
ec
o
m
es
m
o
r
e
r
o
b
u
s
t
an
d
b
etter
s
u
ited
f
o
r
em
ail
h
ea
d
er
class
if
icatio
n
.
3
.
3
.
2
.
P
his
hin
g
em
a
il c
la
s
s
if
i
ca
t
io
n ba
s
ed
o
n t
he
bo
dy
o
f
a
n e
m
a
il
E
m
ail
b
o
d
ies
o
f
te
n
co
n
tain
p
l
ain
tex
t
as
well
as
ad
d
itio
n
al
elem
en
ts
s
u
ch
as
UR
L
s
,
HT
ML
co
n
ten
t
an
d
o
th
er
lin
k
s
.
T
h
e
p
lain
te
x
t
ty
p
ically
in
clu
d
es
th
e
m
ain
m
ess
ag
e
in
ten
d
ed
f
o
r
th
e
r
ec
ip
ien
t,
wh
ich
m
ay
co
n
s
is
t
o
f
s
im
p
le
wo
r
d
s
,
im
a
g
es,
J
av
aScr
ip
t,
an
d
lin
k
s
to
in
ter
n
al
o
r
e
x
ter
n
al
r
eso
u
r
ce
s
.
HT
ML
co
n
ten
t
is
Evaluation Warning : The document was created with Spire.PDF for Python.