I
nd
o
ne
s
ia
n J
o
urna
l o
f
E
lect
rica
l En
g
ineering
a
nd
Co
m
pu
t
er
Science
Vo
l.
3
9
,
No
.
3
,
Sep
tem
b
er
2
0
2
5
,
p
p
.
1
8
3
7
~
1
845
I
SS
N:
2
5
0
2
-
4
7
5
2
,
DOI
: 1
0
.
1
1
5
9
1
/ijeecs.v
3
9
.i
3
.
pp
1
8
3
7
-
1
8
4
5
1837
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ee
cs.ia
esco
r
e.
co
m
A
hy
brid app
ro
a
ch t
o
behav
io
ra
l sp
a
m review de
tec
ti
o
n on
e
-
co
mm
erce
platf
o
rms
using
aprio
ri and
CN
N
G
a
nes
h Wa
y
a
l
,
Vij
a
y
B
ha
nd
a
ri
D
e
p
a
r
t
me
n
t
o
f
C
o
mp
u
t
e
r
S
c
i
e
n
c
e
a
n
d
En
g
i
n
e
e
r
i
n
g
,
M
a
d
h
y
a
n
c
h
a
l
P
r
o
f
e
ss
i
o
n
a
l
U
n
i
v
e
r
si
t
y
,
B
h
o
p
a
l
,
I
n
d
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
No
v
12
,
2
0
2
4
R
ev
is
ed
Ap
r
8
,
2
0
2
5
Acc
ep
ted
J
u
l
2
,
2
0
2
5
S
p
a
m
re
v
iew
s
si
g
n
if
ica
n
tl
y
u
n
d
e
rm
in
e
t
h
e
c
re
d
i
b
il
it
y
o
f
o
n
li
n
e
re
v
iew
sy
ste
m
s
o
n
e
-
c
o
m
m
e
rc
e
we
b
sites
.
Th
is
p
a
p
e
r
p
re
se
n
ts
a
h
y
b
rid
m
e
th
o
d
o
l
o
g
y
th
a
t
c
o
m
b
in
e
s
t
h
e
Ap
rio
ri
a
l
g
o
rit
h
m
a
n
d
c
o
n
v
o
lu
ti
o
n
a
l
n
e
u
ra
l
n
e
two
rk
s
(CNN
)
to
e
fficie
n
tl
y
i
d
e
n
ti
f
y
a
n
d
m
it
ig
a
te
s
p
a
m
re
v
iew
s.
B
y
e
x
a
m
in
in
g
u
se
r
b
e
h
a
v
i
o
r,
i
n
c
lu
d
in
g
a
c
ti
v
i
ty
p
a
tt
e
rn
s,
re
v
iew
e
r
re
p
u
tatio
n
,
tem
p
o
ra
l
d
y
n
a
m
ics
,
a
n
d
se
n
ti
m
e
n
t
c
o
n
sist
e
n
c
y
,
we
p
ro
p
o
se
a
c
o
m
p
re
h
e
n
siv
e
m
o
d
e
l
fo
r
u
n
d
e
rsta
n
d
i
n
g
u
se
r
i
n
tera
c
ti
o
n
s
a
n
d
e
n
g
a
g
e
m
e
n
t.
T
o
e
x
trac
t
imp
o
rtan
t
in
fo
rm
a
ti
o
n
a
n
d
b
u
il
d
p
re
c
ise
sp
a
m
d
e
tec
ti
o
n
m
o
d
e
ls,
we
u
se
d
a
ta
m
in
in
g
a
n
d
m
a
c
h
in
e
lea
rn
in
g
a
p
p
r
o
a
c
h
e
s.
F
u
rth
e
rm
o
re
,
c
o
n
tex
t
u
a
l
a
n
d
d
o
m
a
in
-
sp
e
c
ifi
c
a
n
a
ly
se
s
a
re
c
o
n
d
u
c
ted
to
im
p
ro
v
e
d
e
tec
ti
o
n
stra
teg
ies
.
Th
e
stu
d
y
h
ig
h
li
g
h
ts
t
h
e
sig
n
i
fica
n
c
e
o
f
h
y
b
rid
tec
h
n
i
q
u
e
s
i
n
p
re
se
rv
i
n
g
t
h
e
i
n
teg
rit
y
o
f
e
-
c
o
m
m
e
rc
e
p
latfo
rm
s
th
r
o
u
g
h
su
c
c
e
ss
fu
l
in
d
u
stry
imp
lem
e
n
tat
io
n
s
a
n
d
p
re
se
n
ts ev
a
lu
a
ti
o
n
m
e
tri
c
s,
p
ro
b
l
e
m
s,
a
n
d
fu
t
u
re
re
se
a
rc
h
o
b
jec
ti
v
e
s.
K
ey
w
o
r
d
s
:
B
eh
av
io
r
al
an
aly
s
is
C
o
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
s
E
-
co
m
m
er
ce
we
b
s
ites
Mitig
ati
o
n
s
tr
ateg
ies
R
ev
iewe
r
r
ep
u
tatio
n
m
o
d
elin
g
Sen
tim
en
t a
n
aly
s
is
Sp
am
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
:
Gan
esh
W
ay
al
Dep
ar
tm
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
an
d
E
n
g
in
ee
r
in
g
,
Ma
d
h
y
a
n
ch
al
Pro
f
ess
io
n
al
Un
iv
er
s
ity
B
h
o
p
al,
I
n
d
ia
E
m
ail:
g
an
esh
w2
0
0
6
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
I
n
th
e
d
ig
ital
e
r
a,
e
-
co
m
m
e
r
ce
web
s
ites
h
av
e
r
ev
o
lu
tio
n
i
ze
d
s
h
o
p
p
in
g
b
y
o
f
f
er
in
g
co
n
v
en
ien
ce
,
v
ar
iety
,
a
n
d
ac
ce
s
s
to
a
g
l
o
b
al
m
ar
k
etp
lace
.
A
cr
u
cial
asp
ec
t
in
f
lu
en
ci
n
g
c
o
n
s
u
m
er
d
ec
is
io
n
-
m
ak
in
g
o
n
th
ese
p
latf
o
r
m
s
is
o
n
lin
e
r
ev
iews.
P
o
ten
tial
clien
ts
ca
n
m
ak
e
well
-
in
f
o
r
m
e
d
p
u
r
c
h
asin
g
s
elec
tio
n
s
with
th
e
h
elp
o
f
o
n
lin
e
r
e
v
iews,
wh
ich
o
f
f
er
in
s
ig
h
tf
u
l
co
m
m
en
tar
y
f
r
o
m
p
ast
cu
s
to
m
er
s
.
Ho
wev
er
,
th
e
au
th
en
ticity
a
n
d
tr
u
s
two
r
th
in
ess
o
f
th
ese
r
e
v
iews
h
av
e
b
ee
n
s
ig
n
if
ica
n
tly
co
m
p
r
o
m
is
ed
b
y
th
e
p
r
esen
ce
o
f
r
ev
iew
s
p
am
[
1
]
,
[
2
]
.
R
ev
iew
s
p
am
r
ef
er
s
to
th
e
d
elib
er
ate
m
an
ip
u
latio
n
o
f
r
ev
iews
f
o
r
p
r
o
m
o
tio
n
al,
m
al
icio
u
s
,
o
r
d
ec
ep
tiv
e
p
u
r
p
o
s
es.
I
t
en
c
o
m
p
ass
es
a
wid
e
r
an
g
e
o
f
d
ec
e
p
tiv
e
tactics,
in
clu
d
in
g
f
a
k
e
r
e
v
i
ews,
o
p
in
io
n
s
p
am
,
an
d
Sy
b
il
attac
k
s
[
3
]
.
Fak
e
r
e
v
iews
ar
e
co
m
m
e
n
ts
wr
itten
b
y
p
e
o
p
le
o
r
o
r
g
an
izatio
n
s
w
ith
th
e
in
ten
tio
n
o
f
h
ar
m
in
g
r
iv
al
b
u
s
in
ess
es
o
r
p
r
o
m
o
tin
g
th
eir
o
wn
g
o
o
d
s
.
Su
b
m
is
s
io
n
s
o
f
f
alse
r
ev
ie
ws
th
at
d
is
to
r
t
th
e
r
ev
iewe
r
'
s
ac
tu
al
o
p
in
io
n
s
ar
e
k
n
o
wn
as
o
p
in
io
n
s
p
am
.
Sy
b
il
attac
k
s
in
v
o
lv
e
th
e
cr
ea
tio
n
o
f
m
u
ltip
le
f
ak
e
id
en
titi
es
to
d
ec
eiv
e
r
ea
d
er
s
a
n
d
m
a
n
ip
u
late
th
e
o
v
e
r
all
r
ati
n
g
o
f
a
p
r
o
d
u
ct
o
r
s
er
v
ice
[
4
]
.
A
s
er
io
u
s
th
r
ea
t
to
th
e
tr
u
s
two
r
th
in
ess
an
d
d
ep
en
d
ab
ilit
y
o
f
e
-
co
m
m
er
ce
p
latf
o
r
m
s
is
th
e
p
r
o
life
r
atio
n
o
f
r
e
v
iew
s
p
am
.
C
o
n
s
u
m
er
s
r
ely
o
n
r
e
v
iews
to
m
ak
e
i
n
f
o
r
m
ed
d
ec
is
io
n
s
,
a
n
d
wh
en
th
ese
r
ev
iews
ar
e
ta
in
ted
b
y
s
p
am
,
th
e
tr
u
s
two
r
th
in
ess
o
f
th
e
en
tire
s
y
s
tem
is
co
m
p
r
o
m
is
ed
[
5
]
−
[
7
]
.
T
h
is
s
tu
d
y
ad
d
r
ess
es
th
e
g
ap
s
in
ex
is
tin
g
r
esear
ch
o
n
s
p
am
r
ev
iew
d
etec
tio
n
.
W
h
ile
p
r
e
v
io
u
s
s
tu
d
ies
h
av
e
in
d
i
v
id
u
ally
an
aly
ze
d
eith
er
b
e
h
av
io
r
al
p
atter
n
s
o
r
tex
tu
al
f
ea
tu
r
es,
th
e
y
h
a
v
e
n
o
t
a
d
eq
u
ately
e
x
p
lo
r
e
d
in
teg
r
atin
g
b
o
t
h
asp
ec
ts
co
m
p
r
eh
en
s
iv
ely
.
Sp
ec
if
ically
,
ea
r
lier
ap
p
r
o
ac
h
es
h
av
e
o
v
er
lo
o
k
ed
th
e
co
m
b
i
n
ed
ef
f
ec
t
o
f
u
s
er
b
eh
a
v
io
r
an
aly
s
is
(
e.
g
.
,
p
o
s
tin
g
f
r
eq
u
e
n
cy
,
r
ev
iewe
r
cr
ed
ib
ilit
y
)
an
d
co
n
ten
t
-
b
ased
an
aly
s
is
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
5
2
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
9
,
No
.
3
,
Sep
tem
b
er
20
25
:
1
8
3
7
-
1
8
4
5
1838
(
e.
g
.
,
s
en
tim
en
t
in
c
o
n
s
is
ten
cies,
lin
g
u
is
tic
cu
es).
Ou
r
r
esear
ch
f
ills
th
is
g
ap
b
y
p
r
o
p
o
s
in
g
a
n
o
v
el
h
y
b
r
i
d
m
o
d
el
u
s
in
g
th
e
Ap
r
i
o
r
i
alg
o
r
ith
m
an
d
co
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
s
(
C
NN)
,
ac
h
iev
in
g
e
n
h
an
ce
d
d
etec
tio
n
accu
r
ac
y
an
d
r
eliab
ilit
y
co
m
p
ar
ed
to
s
tan
d
alo
n
e
ap
p
r
o
ac
h
es.
T
h
e
o
b
jectiv
e
o
f
th
is
s
tu
d
y
is
to
p
r
o
v
id
e
a
c
o
m
p
r
e
h
en
s
iv
e
an
aly
s
is
o
f
v
a
r
io
u
s
tec
h
n
iq
u
es
an
d
s
tr
ateg
ies
f
o
r
d
etec
tin
g
a
n
d
m
itig
atin
g
r
ev
iew
s
p
am
o
n
e
-
co
m
m
er
ce
p
latf
o
r
m
s
.
B
y
e
x
am
in
in
g
t
h
e
latest
r
esear
ch
ad
v
an
ce
m
en
ts
an
d
i
n
s
ig
h
ts
,
we
aim
to
co
n
tr
ib
u
te
to
th
e
d
ev
elo
p
m
en
t
o
f
ef
f
ec
t
iv
e
s
p
am
d
etec
tio
n
s
y
s
tem
s
an
d
p
r
o
m
o
te
tr
u
s
t
an
d
au
th
en
ticity
in
o
n
lin
e
r
e
v
ie
ws
[
8
]
.
T
h
is
p
ap
er
s
p
ec
if
ically
f
o
cu
s
es
o
n
a
h
y
b
r
id
ap
p
r
o
ac
h
u
s
in
g
t
h
e
Ap
r
io
r
i
alg
o
r
ith
m
an
d
C
NN
f
o
r
en
h
an
ce
d
s
p
am
d
etec
tio
n
.
I
n
v
esti
g
ate
th
e
ty
p
es
an
d
ch
a
r
ac
ter
is
tics
o
f
r
ev
iew
s
p
am
:
w
e
will
d
elv
e
in
to
th
e
d
if
f
er
e
n
t
f
o
r
m
s
o
f
r
ev
iew
s
p
am
,
in
clu
d
in
g
f
ak
e
r
ev
iews,
o
p
in
io
n
s
p
am
,
a
n
d
Sy
b
il
attac
k
s
.
B
y
u
n
d
er
s
tan
d
in
g
th
e
n
atu
r
e
o
f
th
ese
s
p
am
m
in
g
tech
n
i
q
u
es,
w
e
ca
n
b
etter
d
e
v
is
e
ap
p
r
o
p
r
iate
d
etec
tio
n
an
d
m
itig
atio
n
ap
p
r
o
ac
h
es
[
9
]
.
E
x
p
lo
r
e
b
eh
av
io
r
al
a
n
aly
s
is
ap
p
r
o
ac
h
es
f
o
r
s
p
am
d
etec
tio
n
:
B
eh
a
v
io
r
al
an
aly
s
is
f
o
cu
s
es
o
n
id
en
tify
in
g
a
b
n
o
r
m
al
r
ev
iew
p
o
s
tin
g
b
eh
av
i
o
r
s
,
ass
ess
in
g
th
e
cr
ed
ib
ilit
y
o
f
r
ev
iewe
r
s
,
an
aly
z
in
g
th
e
tim
in
g
an
d
f
r
eq
u
e
n
cy
o
f
r
ev
iew
s
u
b
m
is
s
io
n
s
,
an
d
d
etec
tin
g
in
co
n
s
is
ten
cies
in
r
ev
iewe
r
s
en
tim
en
ts
.
W
e
will
ex
am
i
n
e
th
e
ef
f
ec
tiv
en
ess
o
f
th
ese
ap
p
r
o
ac
h
es a
n
d
th
eir
p
o
ten
tial in
m
itig
atin
g
r
ev
iew
s
p
am
[
1
0
]
.
R
ev
iew
s
p
am
h
as
b
ee
n
d
is
co
v
er
ed
u
s
in
g
m
ac
h
in
e
lear
n
in
g
ap
p
r
o
ac
h
es
f
o
r
s
p
am
d
etec
tio
n
alg
o
r
ith
m
s
.
T
o
f
in
d
s
p
am
t
r
en
d
s
an
d
in
c
r
ea
s
e
d
etec
tio
n
ac
cu
r
ac
y
,
s
u
p
er
v
is
ed
lear
n
i
n
g
m
o
d
els,
an
o
m
aly
d
etec
tio
n
tech
n
iq
u
es,
an
d
u
n
s
u
p
er
v
is
ed
lear
n
in
g
ar
e
ap
p
lied
.
E
x
am
in
in
g
co
n
ten
t
-
b
ased
a
p
p
r
o
ac
h
es
to
s
p
am
d
etec
tio
n
,
s
u
ch
as
tex
t
m
in
in
g
,
lin
g
u
is
tic
an
aly
s
is
,
an
d
n
atu
r
al
lan
g
u
ag
e
p
r
o
ce
s
s
in
g
alg
o
r
it
h
m
s
,
ca
n
h
elp
f
in
d
lin
g
u
is
tic
clu
es
an
d
s
p
am
ten
d
en
cies
in
r
ev
iews.
W
e
will
i
n
v
esti
g
ate
h
o
w
well
th
ese
co
n
ten
t
-
b
ased
m
eth
o
d
s
id
en
tify
r
ev
iew
s
p
am
.
E
x
am
in
e
h
y
b
r
id
ap
p
r
o
ac
h
es
f
o
r
s
p
am
d
etec
tio
n
th
at
lev
er
ag
e
th
e
s
tr
en
g
th
s
o
f
m
u
ltip
le
tech
n
iq
u
es,
s
u
ch
as
co
m
b
in
in
g
b
eh
av
io
r
al
an
aly
s
is
with
m
a
ch
in
e
lear
n
in
g
o
r
co
n
ten
t
-
b
ased
m
eth
o
d
s
.
W
e
will
d
is
cu
s
s
th
e
b
en
ef
its
an
d
ch
allen
g
es
o
f
th
ese
h
y
b
r
id
ap
p
r
o
ac
h
es
an
d
p
r
o
v
id
e
e
x
am
p
les
o
f
s
u
cc
e
s
s
f
u
l
im
p
lem
en
tatio
n
s
[
1
1
]
.
E
x
p
lo
r
e
m
itig
atio
n
s
tr
ateg
ies
f
o
r
r
ev
iew
s
p
am
:
Mitig
atio
n
s
tr
ateg
ies
f
o
cu
s
o
n
r
e
d
u
cin
g
th
e
im
p
ac
t
o
f
r
ev
iew
s
p
am
o
n
e
-
co
m
m
er
ce
p
latf
o
r
m
s
.
W
e
will
ex
am
in
e
u
s
er
r
ep
u
tatio
n
m
o
d
elin
g
,
co
m
m
u
n
ity
f
ee
d
b
ac
k
an
d
in
f
lu
en
ce
,
an
d
th
e
r
o
le
o
f
p
la
tf
o
r
m
p
o
licies
an
d
g
u
id
elin
es
in
p
r
ev
en
tin
g
an
d
m
itig
atin
g
r
ev
iew
s
p
am
[
1
2
]
.
Dis
cu
s
s
f
u
tu
r
e
d
ir
ec
tio
n
s
an
d
ch
allen
g
es:
W
e
wil
l
id
en
tify
th
e
ev
o
lv
in
g
c
h
allen
g
es
in
s
p
am
d
etec
tio
n
an
d
m
itig
atio
n
,
in
clu
d
in
g
ad
v
er
s
ar
ial
attac
k
s
an
d
eth
ical
c
o
n
s
id
e
r
atio
n
s
.
Fu
r
th
er
m
o
r
e,
we
will d
is
cu
s
s
th
e
n
ee
d
f
o
r
r
ea
l
-
tim
e
d
etec
tio
n
an
d
th
e
s
ca
lab
ilit
y
o
f
s
p
am
d
etec
tio
n
s
y
s
tem
s
.
Ad
d
itio
n
ally
,
we
will
h
ig
h
lig
h
t
th
e
im
p
o
r
tan
ce
o
f
o
n
g
o
i
n
g
r
esea
r
ch
an
d
co
llab
o
r
atio
n
b
etwe
e
n
r
esear
ch
er
s
,
p
latf
o
r
m
o
p
er
ato
r
s
,
an
d
u
s
er
s
to
co
m
b
at
r
ev
iew
s
p
am
ef
f
ec
tiv
e
ly
[
1
3
]
.
T
h
e
r
em
ain
in
g
s
ec
tio
n
s
ar
e
s
tr
u
ctu
r
ed
as f
o
llo
ws:
I
n
o
r
d
er
to
en
h
an
ce
s
p
am
r
ev
iew
id
en
tifi
ca
tio
n
,
th
e
p
r
o
p
o
s
ed
m
eth
o
d
o
lo
g
y
ex
p
lai
n
ed
in
Sectio
n
2
in
teg
r
ates
C
NN
[
1
4
]
f
o
r
co
n
te
n
t
an
aly
s
is
an
d
th
e
Ap
r
io
r
i
alg
o
r
ith
m
[
1
5
]
f
o
r
b
eh
a
v
io
u
r
a
l
an
aly
s
is
.
R
esu
lts
an
d
d
is
cu
s
s
io
n
ar
e
p
r
esen
ted
in
Sectio
n
3
,
with
an
em
p
h
asis
o
n
p
e
r
f
o
r
m
an
ce
ev
alu
atio
n
u
ti
lis
in
g
m
etr
ics
s
u
ch
as
p
r
ec
is
io
n
,
ac
c
u
r
ac
y
,
an
d
r
ec
all
as
w
ell
as
a
co
m
p
ar
is
o
n
with
cu
r
r
e
n
t
ap
p
r
o
ac
h
es.
T
h
e
wo
r
k
is
c
o
n
clu
d
e
d
i
n
Sectio
n
4
,
w
h
ich
s
u
m
m
ar
is
es
th
e
m
a
in
co
n
cl
u
s
io
n
s
an
d
s
u
g
g
ests
f
u
tu
r
e
r
esear
ch
t
o
p
ics to
en
h
a
n
ce
s
p
am
d
etec
tio
n
o
n
e
-
co
m
m
er
ce
p
latf
o
r
m
s
.
2.
M
E
T
H
O
D
I
t
p
r
esen
ts
a
h
y
b
r
id
ap
p
r
o
ac
h
co
m
b
in
in
g
th
e
Ap
r
io
r
i
alg
o
r
it
h
m
f
o
r
b
e
h
av
io
r
al
an
aly
s
is
an
d
C
NN
f
o
r
co
n
ten
t
an
al
y
s
is
.
Ap
r
io
r
i
d
et
ec
ts
s
u
s
p
icio
u
s
u
s
er
p
atter
n
s
,
wh
ile
C
NN
an
aly
ze
s
r
ev
ie
w
tex
t
f
o
r
s
p
am
in
d
icato
r
s
.
T
h
eir
o
u
t
p
u
ts
ar
e
in
teg
r
ated
in
to
a
u
n
if
ied
m
o
d
e
l
to
en
h
an
ce
s
p
am
d
etec
tio
n
a
cc
u
r
ac
y
.
T
h
e
f
i
n
al
ar
ch
itectu
r
e
en
s
u
r
es e
f
f
ec
tiv
e
u
s
e
o
f
b
o
t
h
b
eh
a
v
io
r
al
a
n
d
tex
tu
al
f
ea
tu
r
es.
2
.
1
.
I
ntr
o
du
ct
io
n t
o
hy
brid
a
pp
ro
a
ches
Hy
b
r
id
s
tr
ateg
ies
in
co
r
p
o
r
ate
a
v
ar
iety
o
f
tactics
to
m
ax
im
i
ze
th
eir
u
n
iq
u
e
ad
v
a
n
tag
es
an
d
m
in
im
ize
th
eir
d
r
awb
ac
k
s
.
I
n
th
e
co
n
t
ex
t
o
f
s
p
am
r
e
v
iew
d
etec
tio
n
,
a
h
y
b
r
id
ap
p
r
o
ac
h
u
tili
zin
g
b
o
th
th
e
Ap
r
io
r
i
alg
o
r
ith
m
an
d
C
NN
[
1
4
]
ca
n
p
r
o
v
id
e
a
c
o
m
p
r
e
h
en
s
iv
e
an
d
r
o
b
u
s
t
s
o
lu
tio
n
.
T
h
e
Ap
r
io
r
i
alg
o
r
ith
m
ex
ce
ls
at
id
en
tify
in
g
f
r
eq
u
e
n
t
item
s
ets
an
d
ass
o
ciatio
n
s
in
b
eh
av
io
r
al
d
ata,
wh
er
ea
s
C
NN
s
ar
e
p
o
wer
f
u
l
f
o
r
an
al
y
zin
g
th
e
tex
tu
al
co
n
ten
t
o
f
r
ev
iews
.
T
h
is
s
ec
tio
n
d
elv
es
f
u
r
th
er
in
to
th
e
v
ar
io
u
s
way
s
in
wh
ich
tech
n
iq
u
es
ca
n
b
e
co
m
b
in
ed
to
im
p
r
o
v
e
th
e
p
r
ec
i
s
io
n
an
d
d
e
p
en
d
a
b
ilit
y
o
f
s
p
a
m
d
etec
tio
n
s
y
s
tem
s
.
2
.
2
.
Aprio
ri
a
lg
o
rit
hm
f
o
r
b
eha
v
io
ra
l a
na
ly
s
is
T
h
e
class
ic
d
ata
m
in
in
g
tec
h
n
iq
u
e
k
n
o
wn
as
ap
r
i
o
r
i
l
o
c
ates
f
r
eq
u
e
n
tly
r
ec
u
r
r
i
n
g
item
s
ets
an
d
estab
lis
h
es
as
s
o
ciatio
n
r
u
les.
E
x
am
in
in
g
u
s
er
b
eh
a
v
io
r
a
n
d
lo
o
k
in
g
f
o
r
p
atter
n
s
th
at
s
u
g
g
est
th
e
f
o
llo
win
g
ac
tio
n
s
is
p
ar
ticu
lar
ly
b
en
e
f
icial
wh
en
it c
o
m
es to
s
p
am
i
d
en
tific
atio
n
:
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4
7
5
2
A
h
yb
r
id
a
p
p
r
o
a
c
h
to
b
e
h
a
vio
r
a
l sp
a
m
r
ev
iew
d
etec
tio
n
…
(
Ga
n
esh
Wa
ya
l
)
1839
−
Data
p
r
ep
r
o
ce
s
s
in
g
:
t
h
e
f
ir
s
t step
in
v
o
lv
es p
r
ep
r
o
ce
s
s
in
g
th
e
u
s
er
b
eh
av
io
r
d
ata,
s
u
c
h
as r
ev
iewin
g
p
o
s
tin
g
p
atter
n
s
,
tem
p
o
r
al
d
y
n
am
ics,
an
d
in
ter
a
ctio
n
m
etr
ics.
I
n
o
r
d
er
to
m
ak
e
th
e
d
ata
ac
ce
p
ta
b
le
f
o
r
Ap
r
i
o
r
i
an
aly
s
is
,
it is
tr
an
s
lated
i
n
to
a
tr
an
s
ac
tio
n
-
lik
e
s
tr
u
ctu
r
e.
−
Fre
q
u
en
t
item
s
et
m
in
in
g
:
p
r
e
p
r
o
ce
s
s
ed
d
ata
is
s
u
b
jecte
d
to
t
h
e
Ap
r
io
r
i
alg
o
r
ith
m
to
f
in
d
f
r
eq
u
en
t
item
s
ets
co
m
b
in
atio
n
s
o
f
u
s
er
b
e
h
av
io
u
r
p
atter
n
s
th
at
h
ap
p
e
n
t
o
g
eth
er
f
r
eq
u
e
n
tly
.
T
h
ese
it
em
s
ets
aid
in
co
m
p
r
eh
e
n
d
in
g
ty
p
ic
al
ac
tio
n
s
o
f
s
p
am
m
er
s
.
−
Ass
o
ciatio
n
r
u
le
lear
n
in
g
:
t
h
e
item
s
ets
th
at
o
cc
u
r
f
r
eq
u
e
n
tly
ar
e
u
s
ed
to
co
n
s
tr
u
ct
ass
o
ciatio
n
r
u
les.
T
h
ese
r
u
les
s
h
o
w
th
e
r
elatio
n
s
h
ip
s
b
etwe
en
s
ev
er
al
b
eh
av
io
u
r
al
te
n
d
en
cies,
ex
am
p
le
as
u
s
in
g
s
ev
er
al
ac
co
u
n
ts
o
r
f
r
e
q
u
en
tly
p
u
b
lis
h
in
g
ev
alu
atio
n
s
in
a
litt
le
p
er
io
d
o
f
tim
e
.
−
Patter
n
an
aly
s
is
:
t
h
e
g
en
er
ate
d
ass
o
ciatio
n
r
u
les
ar
e
an
aly
ze
d
to
id
en
tify
p
atter
n
s
an
d
an
o
m
alies
in
u
s
er
b
eh
av
io
r
.
Patter
n
s
th
at
d
ev
iate
s
ig
n
if
ican
tly
f
r
o
m
n
o
r
m
al
u
s
e
r
b
eh
av
i
o
r
ar
e
f
la
g
g
ed
as
p
o
te
n
tial
in
d
icato
r
s
o
f
s
p
am
.
S
am
p
le
r
u
le
is
:
"User
s
wh
o
p
o
s
t
m
o
r
e
th
a
n
1
0
r
e
v
iews
p
er
d
ay
an
d
u
s
e
m
u
ltip
le
ac
co
u
n
t
s
f
r
o
m
th
e
s
am
e
I
P
ad
d
r
ess
ar
e
lik
ely
to
b
e
s
p
am
m
er
s
.
"
T
h
is
r
u
le
ca
n
t
h
en
b
e
u
s
ed
to
f
ilter
a
n
d
m
o
n
i
to
r
u
s
er
ac
tiv
ities
o
n
th
e
p
latf
o
r
m
.
2
.
3
.
CNN
f
o
r
co
nte
nt
a
na
ly
s
is
C
NNs
ar
e
h
ig
h
ly
ef
f
ec
tiv
e
i
n
p
r
o
ce
s
s
in
g
an
d
an
aly
zin
g
tex
t
u
al
d
ata.
T
h
e
y
ca
n
d
etec
t
s
u
b
t
le
p
atter
n
s
an
d
in
co
n
s
is
ten
cies in
th
e
co
n
t
en
t o
f
r
e
v
iews th
at
m
ig
h
t in
d
ic
ate
s
p
am
s
tep
s
ar
e
as f
o
llo
ws:
−
Data
p
r
ep
r
o
ce
s
s
in
g
:
ev
er
y
r
ev
iew'
s
te
x
t
u
n
d
er
g
o
es
p
r
ep
r
o
ce
s
s
in
g
,
wh
ic
h
in
cl
u
d
es
v
ec
to
r
izatio
n
,
to
k
en
izatio
n
,
an
d
s
to
p
wo
r
d
r
em
o
v
al.
I
n
o
r
d
er
to
f
ee
d
th
e
t
ex
tu
al
d
ata
i
n
to
th
e
C
NN,
th
i
s
p
h
ase
tu
r
n
s
it
in
to
a
n
u
m
er
ical
r
ep
r
esen
tatio
n
.
−
C
NN
ar
ch
itectu
r
e
:
to
id
en
tify
b
o
th
lo
ca
l
an
d
g
lo
b
al
tr
e
n
d
s
i
n
th
e
r
ev
iew
tex
t,
a
C
NN
m
o
d
el
is
b
u
ilt
with
m
an
y
co
n
v
o
lu
tio
n
al
lay
er
s
.
T
y
p
ical
lay
er
s
in
clu
d
e
em
b
ed
d
i
n
g
lay
er
s
,
co
n
v
o
lu
tio
n
al
lay
er
s
with
d
if
f
er
en
t
f
ilter
s
izes,
p
o
o
lin
g
lay
er
s
,
a
n
d
f
u
lly
co
n
n
ec
ted
lay
e
r
s
[
1
4
]
.
−
T
r
ain
in
g
:
th
e
lab
elled
d
ataset,
wh
ich
in
clu
d
es
b
o
th
leg
itima
te
an
d
s
p
am
r
ev
iews,
is
u
s
ed
t
o
tr
ain
th
e
C
NN
m
o
d
el.
Du
r
in
g
tr
ain
in
g
,
th
e
m
o
d
el
p
ick
s
u
p
ch
ar
ac
ter
is
tics
co
m
m
o
n
to
s
p
am
r
e
v
iews,
s
u
ch
o
v
er
ly
p
o
s
itiv
e
o
r
n
eg
ativ
e
p
h
r
asin
g
,
r
e
p
ea
tin
g
p
h
r
ases
,
an
d
a
lack
o
f
s
p
ec
if
ic
d
etails,
wh
ich
h
elp
it
lea
r
n
to
d
is
tin
g
u
is
h
b
etwe
en
th
e
two
[
1
5
]
.
−
E
v
alu
atio
n
:
p
e
r
f
o
r
m
an
ce
m
etr
ics
f
o
r
th
e
C
NN
m
o
d
el
in
clu
d
e
r
ec
all,
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
an
d
F1
-
s
co
r
e.
T
h
e
p
r
e
d
ictio
n
s
m
ad
e
b
y
th
e
s
y
s
tem
ar
e
co
m
p
a
r
ed
to
a
test
s
et
to
s
ee
h
o
w
well
it d
etec
ts
s
p
am
r
ev
iews.
2
.
4
.
I
nte
g
ra
t
ing
a
prio
ri
a
nd
CNN
f
o
r
enha
nced
det
ec
t
io
n
T
h
e
in
teg
r
atio
n
o
f
Ap
r
io
r
i a
n
d
C
NN
tech
n
iq
u
es c
o
m
b
in
es b
eh
av
io
r
al
an
d
c
o
n
ten
t a
n
aly
s
is
,
p
r
o
v
id
i
n
g
a
h
o
lis
tic
ap
p
r
o
ac
h
to
s
p
am
d
e
tectio
n
Step
s
ar
e
as f
o
llo
ws:
−
Featu
r
e
f
u
s
io
n
:
f
ea
tu
r
es
e
x
tr
a
cted
f
r
o
m
th
e
Ap
r
io
r
i
alg
o
r
ith
m
(
b
eh
av
io
r
al
p
atter
n
s
)
an
d
C
NN
(
tex
tu
al
f
ea
tu
r
es)
ar
e
co
m
b
in
ed
.
C
o
n
c
aten
atin
g
f
ea
tu
r
e
v
ec
to
r
s
o
r
e
m
p
lo
y
in
g
a
f
ea
tu
r
e
s
elec
tio
n
t
ec
h
n
iq
u
e
to
p
ick
th
e
m
o
s
t p
er
tin
en
t c
h
ar
ac
ter
is
tics
f
r
o
m
b
o
t
h
s
ets ar
e
two
way
s
to
ac
co
m
p
lis
h
th
is
f
u
s
io
n
.
−
Hy
b
r
id
m
o
d
el
tr
ain
in
g
:
th
e
f
e
atu
r
e
s
ets
ar
e
co
m
b
in
ed
to
p
r
o
d
u
ce
a
h
y
b
r
id
m
o
d
el.
T
h
is
f
ea
tu
r
e
s
et
ca
n
b
e
u
s
ed
to
tr
ain
m
ac
h
in
e
lear
n
i
n
g
class
if
ier
s
,
s
u
ch
as
d
ec
is
io
n
tr
ee
s
,
r
an
d
o
m
f
o
r
ests
,
an
d
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
es (
SVM)
,
to
d
is
tin
g
u
is
h
b
et
wee
n
au
t
h
en
tic
an
d
f
r
au
d
u
len
t r
ev
iews.
−
Dete
ctio
n
an
d
m
itig
atio
n
:
th
e
h
y
b
r
id
m
o
d
el
is
d
ep
lo
y
e
d
i
n
r
ea
l
-
tim
e
to
d
etec
t
an
d
m
itig
at
e
s
p
am
r
ev
iews.
R
ev
iews f
lag
g
ed
as sp
am
ca
n
b
e
s
u
b
jecte
d
to
f
u
r
th
e
r
v
er
if
ica
tio
n
o
r
a
u
to
m
atica
lly
f
ilter
ed
o
u
t [
1
6
]
.
2
.
5
.
Desig
n o
f
t
he
hy
brid m
o
del a
rc
hite
ct
ure
Ou
r
h
y
b
r
id
s
p
am
d
etec
tio
n
m
o
d
el
u
tili
ze
s
b
o
t
h
b
e
h
av
io
u
r
al
an
d
co
n
te
n
t
-
b
ased
a
n
aly
s
is
th
r
o
u
g
h
t
h
e
co
m
b
in
atio
n
o
f
th
e
Ap
r
io
r
i
al
g
o
r
ith
m
an
d
C
NN
in
its
ar
ch
i
tectu
r
e.
T
h
e
ar
ch
itectu
r
e
m
o
d
el
f
o
r
h
y
b
r
id
s
p
am
r
ev
iew
d
etec
ti
o
n
is
s
h
o
wn
in
Fig
u
r
e
1
.
T
h
is
m
o
d
el
is
d
esig
n
ed
to
im
p
r
o
v
e
th
e
ac
cu
r
ac
y
an
d
e
f
f
icien
cy
o
f
d
etec
tin
g
s
p
am
r
e
v
iews
o
n
e
-
co
m
m
er
ce
p
latf
o
r
m
s
[
1
7
]
.
B
y
in
teg
r
atin
g
t
h
ese
two
p
o
wer
f
u
l
tech
n
iq
u
es,
t
h
e
m
o
d
el
ca
n
id
en
tify
s
u
b
tle
p
at
ter
n
s
th
at
m
ay
b
e
m
is
s
ed
b
y
s
in
g
le
-
m
eth
o
d
ap
p
r
o
ac
h
es.
T
h
is
h
y
b
r
id
ap
p
r
o
ac
h
en
s
u
r
es
a
co
m
p
r
e
h
en
s
iv
e
a
n
aly
s
is
o
f
b
o
t
h
u
s
er
b
e
h
av
io
r
an
d
r
ev
iew
co
n
ten
t.
T
h
e
f
o
llo
win
g
elem
e
n
ts
co
m
p
r
is
e
th
e
m
o
d
el
ar
ch
itectu
r
e:
−
Data
in
p
u
t
lay
er
:
t
h
e
s
y
s
tem
ac
ce
p
ts
two
ty
p
es
o
f
in
p
u
t
d
ata:
b
eh
a
v
io
r
al
d
ata
(
e.
g
.
,
r
ev
iew
p
o
s
tin
g
p
atter
n
s
,
u
s
er
in
ter
ac
tio
n
m
etr
i
cs)
an
d
r
ev
iew
tex
t
d
ata.
−
B
eh
av
io
r
al
an
al
y
s
is
m
o
d
u
le
(
Ap
r
io
r
i
al
g
o
r
ith
m
):
to
d
eter
m
i
n
e
f
r
eq
u
e
n
tly
o
cc
u
r
r
i
n
g
item
s
ets
an
d
p
r
o
d
u
ce
ass
o
ciatio
n
r
u
les,
th
is
m
o
d
u
le
an
aly
s
es
b
eh
av
i
o
u
r
al
d
ata.
T
h
is
m
o
d
u
le
p
r
o
d
u
ce
s
o
u
tp
u
t
with
b
eh
av
io
u
r
al
ch
ar
ac
ter
is
tics
th
at
ar
e
s
u
g
g
esti
v
e
o
f
s
p
am
ac
tiv
ity
.
−
C
o
n
ten
t
an
aly
s
is
m
o
d
u
le:
by
p
ass
in
g
tex
tu
al
d
ata
th
r
o
u
g
h
a
s
er
ies
o
f
n
eu
r
al
lay
er
s
,
th
is
m
o
d
u
le
ex
tr
ac
ts
ch
ar
ac
ter
is
tics
th
at
p
in
p
o
in
t
s
u
b
tle
tr
en
d
s
in
r
ev
iew
co
n
t
en
t.
Sp
am
f
ea
tu
r
es
in
th
e
r
ev
iew
tex
t
ar
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
5
2
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
9
,
No
.
3
,
Sep
tem
b
er
20
25
:
1
8
3
7
-
1
8
4
5
1840
id
en
tifie
d
b
y
th
e
C
NN
ar
ch
itectu
r
e,
wh
ich
co
n
s
is
ts
o
f
a
n
e
m
b
ed
d
in
g
lay
e
r
,
m
u
ltip
le
co
n
v
o
lu
tio
n
al
la
y
er
s
,
p
o
o
lin
g
lay
er
s
,
an
d
f
u
lly
c
o
n
n
ec
ted
lay
er
s
.
−
Featu
r
e
f
u
s
io
n
lay
er
:
th
e
f
ea
t
u
r
es
ex
tr
ac
ted
f
r
o
m
t
h
e
b
eh
a
v
io
r
al
an
aly
s
is
an
d
co
n
ten
t
an
aly
s
is
m
o
d
u
les
ar
e
f
u
s
ed
to
g
eth
er
.
T
h
is
ca
n
b
e
a
ch
iev
ed
th
r
o
u
g
h
co
n
ca
ten
atio
n
o
r
a
f
ea
tu
r
e
s
elec
tio
n
p
r
o
ce
s
s
,
en
s
u
r
in
g
th
at
th
e
m
o
s
t r
elev
an
t f
ea
t
u
r
es f
r
o
m
b
o
th
m
o
d
u
les ar
e
u
s
ed
f
o
r
t
h
e
f
in
al
class
if
icatio
n
.
−
C
las
s
if
icatio
n
lay
er
:
one
ca
n
u
s
e
a
m
ac
h
in
e
lear
n
in
g
class
if
ier
s
u
ch
a
r
a
n
d
o
m
f
o
r
est,
d
ec
is
io
n
tr
ee
,
o
r
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
to
p
r
e
d
ict
th
e
lik
elih
o
o
d
th
at
a
r
ev
iew
is
s
p
am
g
iv
en
th
e
f
u
s
ed
d
ata.
T
o
tr
ain
th
e
class
if
ier
,
a
tag
g
ed
d
ataset
o
f
r
ev
iews
b
o
th
s
p
am
an
d
n
o
n
-
s
p
a
m
is
u
tili
ze
d
.
−
Ou
tp
u
t
lay
er
:
th
e
f
in
al
o
u
tp
u
t
is
a
b
in
ar
y
class
if
icatio
n
th
at
in
d
icate
s
wh
eth
er
a
r
ev
iew
is
r
eg
ar
d
ed
as
s
p
am
.
Per
f
o
r
m
an
ce
in
d
icato
r
s
in
clu
d
in
g
F1
-
s
co
r
e
,
AUC
-
R
O
C
,
p
r
ec
is
io
n
,
r
ec
all,
an
d
ac
cu
r
ac
y
ar
e
u
s
ed
to
ass
es
s
th
e
m
o
d
el'
s
ef
f
ec
tiv
en
ess
.
Fig
u
r
e
1
.
A
r
c
h
itectu
r
e
m
o
d
el
o
f
h
y
b
r
id
s
p
am
r
ev
iew
d
etec
ti
o
n
2
.
6
.
J
us
t
if
ica
t
io
n
f
o
r
cho
s
en
m
et
ho
ds
a
nd
a
lg
o
rit
hm
s
T
h
e
s
elec
tio
n
o
f
th
e
Ap
r
io
r
i
alg
o
r
ith
m
an
d
C
NN
f
o
r
t
h
is
h
y
b
r
id
ap
p
r
o
ac
h
is
d
r
iv
e
n
b
y
t
h
eir
co
m
p
lem
en
tar
y
s
tr
en
g
th
s
in
a
n
aly
zin
g
d
if
f
er
en
t a
s
p
ec
ts
o
f
r
ev
iew
d
ata
:
−
Ap
r
io
r
i
alg
o
r
ith
m
:
b
ec
au
s
e
it
i
s
ex
ce
llen
t
at
r
ec
o
g
n
is
in
g
f
r
eq
u
en
t
item
s
ets
an
d
p
r
o
d
u
cin
g
a
s
s
o
ciatio
n
r
u
les
b
o
th
n
ec
ess
ar
y
f
o
r
s
p
o
ttin
g
p
a
tter
n
s
s
u
g
g
esti
v
e
o
f
s
p
am
th
e
Ap
r
io
r
i
alg
o
r
ith
m
is
a
g
o
o
d
f
i
t
f
o
r
an
aly
s
in
g
b
eh
av
io
u
r
al
d
ata.
T
h
is
alg
o
r
it
h
m
,
f
o
r
ex
a
m
p
le,
ef
f
ec
tiv
ely
c
ap
tu
r
es
tr
en
d
s
lik
e
th
e
u
s
e
o
f
s
ev
er
al
ac
co
u
n
ts
f
r
o
m
a
s
in
g
le
I
P
ad
d
r
ess
o
r
th
e
s
u
b
m
is
s
io
n
o
f
r
ev
iews
r
e
p
e
ated
ly
with
in
s
h
o
r
t
p
er
io
d
s
o
f
tim
e.
Ap
r
io
r
i'
s
ca
p
ac
ity
to
id
en
tify
th
ese
co
r
r
elatio
n
s
m
ak
es
it
an
ef
f
e
ctiv
e
to
o
l
f
o
r
b
eh
a
v
io
u
r
al
a
n
aly
s
is
in
s
p
am
id
en
tific
atio
n
.
−
C
NN:
b
ec
au
s
e
o
f
its
tr
ac
k
r
ec
o
r
d
o
f
id
en
tify
i
n
g
b
o
th
lo
ca
l
an
d
g
lo
b
al
p
atter
n
s
in
tex
tu
al
i
n
f
o
r
m
atio
n
,
C
NNs
ar
e
u
s
ed
f
o
r
te
x
t
an
al
y
s
is
task
s
.
R
ep
etitiv
e
p
h
r
ase
s
,
s
tr
o
n
g
em
o
tio
n
s
,
o
r
g
e
n
er
al
lan
g
u
a
g
e
a
r
e
ex
am
p
les
o
f
s
u
b
tle
s
ig
n
s
o
f
s
p
am
th
at
C
NNs
ar
e
esp
ec
ially
g
o
o
d
at
s
p
o
ttin
g
in
r
e
v
iew
m
ater
ial.
As
C
NNs
ca
n
lear
n
h
ier
ar
ch
ical
r
e
p
r
ese
n
tatio
n
s
o
f
th
e
tex
t,
th
ey
ar
e
id
ea
l
f
o
r
id
en
tify
i
n
g
co
m
p
le
x
p
atter
n
s
th
at
s
im
p
ler
m
o
d
els m
ig
h
t m
is
s
b
e
ca
u
s
e
o
f
th
eir
lay
e
r
ed
ar
c
h
itectu
r
e.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4
7
5
2
A
h
yb
r
id
a
p
p
r
o
a
c
h
to
b
e
h
a
vio
r
a
l sp
a
m
r
ev
iew
d
etec
tio
n
…
(
Ga
n
esh
Wa
ya
l
)
1841
−
By
in
teg
r
atin
g
th
e
b
en
e
f
its
o
f
b
eh
av
io
u
r
al
an
d
co
n
ten
t
an
al
y
s
is
,
th
ese
two
m
eth
o
d
s
ca
n
b
e
in
teg
r
ated
to
p
r
o
v
id
e
a
m
o
r
e
th
o
r
o
u
g
h
a
n
a
ly
s
is
.
B
ec
au
s
e
it
ca
p
tu
r
es
a
lar
g
er
r
a
n
g
e
o
f
in
d
icato
r
s
th
at
wo
u
ld
n
o
t
b
e
v
is
ib
le
wh
en
em
p
lo
y
in
g
a
s
in
g
le
m
eth
o
d
alo
n
e,
th
is
h
y
b
r
id
ap
p
r
o
ac
h
im
p
r
o
v
es
th
e
r
o
b
u
s
tn
ess
an
d
ac
cu
r
ac
y
o
f
s
p
am
id
e
n
tific
atio
n
.
2
.
7
.
Da
t
a
des
cr
iptio
n a
nd
prepro
ce
s
s
ing
T
h
e
ef
f
ec
tiv
e
n
ess
o
f
th
e
h
y
b
r
id
s
p
am
d
etec
tio
n
alg
o
r
ith
m
is
s
ig
n
if
ican
tly
i
n
f
lu
en
ce
d
b
y
th
e
co
m
p
leten
ess
an
d
ca
lib
r
e
o
f
t
h
e
d
ata
u
tili
s
ed
.
T
h
is
s
ec
tio
n
o
u
tlin
es
th
e
p
r
o
ce
d
u
r
es
u
s
e
d
f
o
r
d
ata
co
llectin
g
,
p
r
ep
r
o
ce
s
s
in
g
,
an
d
p
r
ep
ar
atio
n
to
g
u
a
r
an
tee
th
e
c
o
r
r
ec
tn
ess
an
d
d
e
p
en
d
a
b
ilit
y
o
f
th
e
m
o
d
el.
Data
co
llectio
n
:
th
e
u
s
er
r
ev
iews
u
tili
s
ed
i
n
th
i
s
in
v
esti
g
atio
n
wer
e
tak
e
n
f
r
o
m
a
p
u
b
licly
ac
ce
s
s
ib
le
d
atase
t
o
f
an
e
-
co
m
m
er
ce
p
latf
o
r
m
(
Fig
u
r
e
2
)
.
d
e
p
icts
d
ata
f
r
o
m
Am
az
o
n
,
n
o
tab
ly
t
h
e
Am
az
o
n
r
ev
iew
p
o
lar
ity
d
ataset.
Fig
u
r
e
2
.
v
iew
o
f
d
ata
u
s
ed
T
h
is
d
ataset
h
as
b
o
th
au
th
en
t
ic
an
d
f
r
au
d
u
le
n
t
r
ev
iews,
wh
ich
ar
e
u
s
ed
to
tr
ain
an
d
ev
alu
ate
th
e
h
y
b
r
id
m
o
d
el.
W
ith
a
s
ig
n
if
ican
t
n
u
m
b
er
o
f
r
ev
iews
s
p
an
n
in
g
m
u
ltip
le
p
r
o
d
u
ct
ca
teg
o
r
ies,
th
e
d
ataset
p
r
o
v
id
es a
b
r
o
a
d
r
an
g
e
o
f
u
s
er
b
eh
av
io
u
r
an
d
r
e
v
iew
co
n
ten
t.
2
.
8
.
Da
t
a
prepro
ce
s
s
ing
B
ef
o
r
e
ap
p
ly
i
n
g
a
n
aly
tical
o
r
m
ac
h
i
n
e
lear
n
i
n
g
tech
n
iq
u
es,
th
e
r
aw
d
ata
u
n
d
er
wen
t
ess
en
tial
p
r
ep
r
o
ce
s
s
in
g
s
tep
s
to
en
s
u
r
e
co
n
s
is
ten
cy
,
q
u
ality
,
an
d
s
u
itab
ilit
y
f
o
r
an
aly
s
is
.
T
h
ese
p
r
o
ce
d
u
r
es
h
elp
tr
an
s
f
o
r
m
u
n
s
tr
u
ct
u
r
ed
in
p
u
ts
i
n
to
a
s
tr
u
ctu
r
ed
f
o
r
m
at
th
at
en
h
an
ce
s
m
o
d
el
ac
cu
r
ac
y
an
d
r
eliab
ilit
y
.
A
f
ew
p
r
ep
r
o
ce
s
s
in
g
p
r
o
ce
d
u
r
es we
r
e
u
s
ed
to
g
et
th
e
d
ata
r
ea
d
y
f
o
r
an
aly
s
is
,
in
clu
d
in
g
:
2
.
8
.
1
.
B
eha
v
io
ra
l
da
t
a
prepro
ce
s
s
ing
−
Po
s
tin
g
p
atter
n
s
f
o
r
r
ev
iews:
I
n
f
o
r
m
atio
n
ab
o
u
t
th
e
n
u
m
b
er
a
n
d
tim
e
o
f
s
u
b
m
is
s
io
n
s
f
o
r
r
ev
iews
was
tak
en
o
u
t.
W
e
lo
o
k
ed
at
tr
e
n
d
s
in
clu
d
in
g
th
e
q
u
an
tity
o
f
r
ev
iews
s
u
b
m
itted
d
aily
,
th
e
len
g
th
o
f
tim
e
b
etwe
en
r
ev
iews,
an
d
h
o
w
r
ev
iews
wer
e
s
p
r
ea
d
am
o
n
g
v
ar
io
u
s
g
o
o
d
s
.
I
n
te
r
ac
tio
n
Me
tr
ics:
User
in
ter
ac
tio
n
d
ata,
in
clu
d
i
n
g
th
e
n
u
m
b
er
o
f
h
elp
f
u
l
v
o
tes
r
e
ce
iv
ed
,
u
s
er
ac
co
u
n
t
ag
e,
an
d
o
v
er
all
ac
tiv
ity
lev
els,
wer
e
co
llected
to
id
en
ti
f
y
p
o
te
n
tial sp
am
m
er
s
.
−
T
r
an
s
f
o
r
m
atio
n
:
b
e
h
av
io
r
al
[
1
8
]
d
ata
was
tr
an
s
f
o
r
m
ed
i
n
to
a
tr
an
s
ac
tio
n
-
lik
e
f
o
r
m
at
s
u
itab
le
f
o
r
t
h
e
Ap
r
io
r
i
alg
o
r
ith
m
.
T
h
is
in
v
o
l
v
ed
en
co
d
in
g
u
s
er
ac
tio
n
s
an
d
in
ter
ac
tio
n
s
in
to
a
s
tr
u
ctu
r
ed
d
ataset
th
at
co
u
ld
b
e
m
in
ed
f
o
r
f
r
eq
u
en
t it
em
s
ets an
d
ass
o
ciatio
n
r
u
les.
2
.
8
.
2
.
Te
x
tu
a
l
da
t
a
prepro
ce
s
s
ing
−
T
o
k
en
izatio
n
:
t
o
k
en
is
in
g
ea
ch
r
ev
iew
tex
t
in
t
o
in
d
i
v
id
u
al
w
o
r
d
s
o
r
p
h
r
ases
allo
wed
th
e
m
o
d
el
to
e
x
am
in
e
wo
r
d
u
s
ag
e
a
n
d
p
atter
n
s
.
−
Sto
p
wo
r
d
s
r
em
o
v
al:
co
m
m
o
n
s
to
p
wo
r
d
s
th
at
d
o
n
o
t
co
n
tr
i
b
u
te
to
s
p
am
d
etec
tio
n
(
e.
g
.
,
"
an
d
,
""
th
e,
""
is
")
wer
e
r
em
o
v
e
d
to
f
o
cu
s
o
n
m
o
r
e
m
ea
n
in
g
f
u
l c
o
n
te
n
t [
1
9
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
5
2
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
9
,
No
.
3
,
Sep
tem
b
er
20
25
:
1
8
3
7
-
1
8
4
5
1842
−
Vec
to
r
izatio
n
:
wo
r
d
e
m
b
ed
d
in
g
s
an
d
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
e
n
cy
(
TF
-
I
DF
)
wer
e
u
s
ed
to
tr
an
s
f
o
r
m
th
e
tex
tu
al
in
p
u
t in
t
o
a
n
u
m
er
ical
r
ep
r
esen
tatio
n
s
o
th
at
th
e
C
NN
co
u
ld
p
r
o
ce
s
s
it.
−
Sen
tim
en
t
an
aly
s
is
:
to
d
eter
m
in
e
th
e
g
e
n
er
al
s
en
tim
en
t
o
f
ea
ch
r
e
v
iew,
p
r
elim
in
ar
y
s
e
n
tim
en
t
an
aly
s
is
was c
ar
r
ied
o
u
t.
T
h
is
ca
n
b
e
a
h
elp
f
u
l to
o
l f
o
r
id
en
tify
i
n
g
r
e
v
iews th
at
ar
e
m
is
lead
in
g
o
r
o
v
er
ly
d
r
am
atic.
−
Data
s
p
litt
in
g
:
a
s
tan
d
ar
d
s
p
lit
r
ati
o
o
f
7
0
:3
0
was
u
s
ed
to
en
s
u
r
e
th
at
th
e
m
o
d
el
was
tr
ain
ed
o
n
a
s
am
p
le
an
d
th
at
it wa
s
also
v
alid
ated
a
n
d
test
ed
o
n
v
ar
io
u
s
d
ata
to
m
i
n
im
ize
o
v
er
f
itti
n
g
.
3.
RE
SU
L
T
S AN
D
AN
AL
Y
SI
S
3
.
1
.
M
ea
s
ures o
f
ev
a
lua
t
i
o
n
Stan
d
ar
d
m
etr
ics
lik
e
p
r
ec
is
io
n
,
r
ec
all,
ac
cu
r
ac
y
,
F1
-
s
co
r
e,
an
d
th
e
ar
ea
u
n
d
e
r
th
e
r
ec
eiv
e
r
o
p
er
atin
g
ch
ar
ac
ter
is
tic
cu
r
v
e
(
AUC
-
R
OC
)
wer
e
u
s
ed
t
o
ass
ess
o
u
r
h
y
b
r
id
ap
p
r
o
ac
h
em
p
lo
y
in
g
t
h
e
Ap
r
io
r
i
alg
o
r
ith
m
an
d
C
NN.
W
ith
th
e
f
ewe
s
t
f
alse
p
o
s
itiv
es
an
d
n
eg
at
iv
es,
th
ese
m
etr
ics
o
f
f
er
a
th
o
r
o
u
g
h
ev
alu
atio
n
o
f
th
e
m
o
d
el'
s
p
er
f
o
r
m
a
n
ce
i
n
c
o
r
r
e
ctly
class
if
y
in
g
s
p
am
r
ev
iews
.
B
en
ch
m
ar
k
in
g
th
ese
m
etr
ics
ag
ain
s
t
c
o
m
p
etin
g
m
o
d
els v
alid
ates o
u
r
h
y
b
r
id
s
tr
ateg
y
'
s
co
n
s
is
ten
t o
u
tp
er
f
o
r
m
an
ce
ac
r
o
s
s
all
m
ain
f
ac
to
r
s
:
−
Pre
cisi
o
n
:
t
h
e
p
er
ce
n
tag
e
o
f
s
p
am
r
e
v
iews
th
at
wer
e
s
u
cc
ess
f
u
lly
id
en
tifie
d
a
m
o
n
g
th
o
s
e
th
at
wer
e
m
ar
k
ed
as su
ch
.
−
R
ec
all:
t
h
e
p
er
ce
n
tag
e
o
f
g
en
u
in
e
s
p
am
r
ev
iews th
at
wer
e
ac
cu
r
ately
class
if
ied
.
−
F1
-
Sco
r
e:
t
h
e
p
r
ec
is
io
n
an
d
r
e
ca
ll h
ar
m
o
n
ic
m
ea
n
,
wh
ic
h
b
a
lan
ce
s
b
o
t
h
m
ea
s
u
r
es.
−
Pre
cisi
o
n
:
t
h
e
to
tal
p
er
ce
n
tag
e
o
f
r
ev
iews th
at
ar
e
ac
cu
r
ately
ca
teg
o
r
ized
.
−
AUC
-
R
O
C
:
t
h
e
m
o
d
el'
s
ca
p
ac
ity
to
d
is
cr
im
in
ate
b
etwe
en
leg
itima
te
an
d
s
p
am
r
ev
iews;
v
alu
es
n
ea
r
1
d
en
o
te
s
u
p
e
r
io
r
p
e
r
f
o
r
m
an
ce
.
3
.
2
.
Co
m
pa
ra
t
iv
e
a
na
ly
s
is
W
e
ev
alu
ated
o
u
r
p
r
o
p
o
s
ed
h
y
b
r
id
m
eth
o
d
a
g
ain
s
t
s
tan
d
alo
n
e
b
e
h
av
io
r
al
an
aly
s
is
,
C
NN
-
b
ased
co
n
ten
t
an
aly
s
is
,
an
d
tr
ad
iti
o
n
al
m
ac
h
in
e
lear
n
in
g
tech
n
iq
u
es
s
u
ch
as
SVM
an
d
d
ec
is
io
n
tr
ee
s
[
2
0
]
.
C
o
m
p
ar
ativ
e
r
esu
lts
ar
e
s
u
m
m
ar
ized
in
T
a
b
le
1
an
d
v
is
u
al
ized
in
Fig
u
r
e
3
.
Fig
u
r
e
3
(
a)
co
m
p
ar
es
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
an
d
F1
-
s
co
r
e
ac
r
o
s
s
all
m
eth
o
d
s
,
h
ig
h
lig
h
tin
g
th
e
s
u
p
er
io
r
p
er
f
o
r
m
a
n
ce
o
f
o
u
r
Ap
r
io
r
i
-
C
NN
h
y
b
r
id
m
o
d
el,
wh
ile
Fig
u
r
e
3
(
b
)
s
h
o
ws
th
e
v
ar
iab
ilit
y
o
f
th
e
s
e
m
etr
ics
u
s
in
g
a
b
o
x
p
lo
t
o
f
in
ter
q
u
ar
tile
r
a
n
g
e
(
I
QR
)
.
T
h
e
in
teg
r
atio
n
o
f
Ap
r
i
o
r
i
an
d
C
NN
ef
f
ec
tiv
ely
i
d
en
ti
f
ies
s
u
b
tle
s
p
am
in
d
icato
r
s
m
i
s
s
ed
b
y
s
tan
d
al
o
n
e
ap
p
r
o
ac
h
es,
s
ig
n
if
ican
tly
im
p
r
o
v
in
g
d
etec
tio
n
ac
cu
r
ac
y
an
d
r
eliab
ilit
y
[
2
1
]
,
[
2
2
]
.
T
h
is
h
y
b
r
id
s
tr
ateg
y
o
u
tp
er
f
o
r
m
ed
in
d
iv
id
u
al
m
eth
o
d
s
an
d
tr
a
d
i
tio
n
al
alg
o
r
ith
m
s
ac
r
o
s
s
all
cr
iter
ia,
p
r
o
d
u
cin
g
a
m
o
r
e
co
n
s
is
ten
t
an
d
s
ca
lab
le
d
etec
tio
n
s
y
s
tem
f
o
r
r
ea
l
-
tim
e
d
ep
lo
y
m
en
t o
n
e
-
co
m
m
er
ce
p
latf
o
r
m
s
.
(
a)
(
b
)
Fig
u
r
e
3
.
C
o
m
p
a
r
ativ
e
p
e
r
f
o
r
m
an
ce
o
f
s
p
a
m
d
etec
tio
n
m
eth
o
d
s
: (
a)
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
e
ca
ll
,
an
d
F1
-
s
co
r
e
an
d
(
b
)
in
ter
q
u
ar
tile
r
an
g
e
(
I
Q
R
)
o
f
m
etr
ics ac
r
o
s
s
m
eth
o
d
s
T
ab
le
1
.
Per
f
o
r
m
an
ce
co
m
p
a
r
is
o
n
o
f
d
i
f
f
er
en
t sp
am
d
etec
tio
n
m
eth
o
d
s
M
e
t
h
o
d
AUC
-
R
O
C
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F1
-
S
c
o
r
e
A
c
c
u
r
a
c
y
B
e
h
a
v
i
o
u
r
a
l
a
n
a
l
y
s
i
s
0
.
8
5
0
.
8
2
0
.
7
6
0
.
7
9
0
.
8
1
C
o
n
t
e
n
t
a
n
a
l
y
si
s
(
C
N
N
)
0
.
8
7
0
.
8
5
0
.
7
8
0
.
8
1
0
.
8
3
S
V
M
0
.
8
3
0
.
8
0
0
.
7
5
0
.
7
7
0
.
8
0
D
e
c
i
s
i
o
n
t
r
e
e
s
0
.
8
2
0
.
7
8
0
.
7
4
0
.
7
6
0
.
7
9
H
y
b
r
i
d
a
p
p
r
o
a
c
h
(
A
p
r
i
o
r
i
+
C
N
N
)
0
.
9
2
0
.
8
9
0
.
8
4
0
.
8
6
0
.
8
8
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4
7
5
2
A
h
yb
r
id
a
p
p
r
o
a
c
h
to
b
e
h
a
vio
r
a
l sp
a
m
r
ev
iew
d
etec
tio
n
…
(
Ga
n
esh
Wa
ya
l
)
1843
3
.
3
.
Dis
cus
s
io
n a
nd
a
na
ly
s
is
Ou
r
s
tu
d
y
d
e
m
o
n
s
tr
ates
im
p
r
o
v
ed
p
e
r
f
o
r
m
an
ce
c
o
m
p
ar
e
d
to
p
r
ev
io
u
s
m
eth
o
d
s
f
o
cu
s
in
g
s
o
lely
o
n
eith
er
b
eh
a
v
io
r
al
o
r
tex
tu
al
f
ea
tu
r
es.
W
h
ile
s
tan
d
alo
n
e
b
eh
av
io
r
al
a
n
aly
s
is
ac
h
iev
ed
m
o
d
er
ate
d
etec
tio
n
ac
cu
r
ac
y
(
0
.
8
1
)
an
d
co
n
ten
t
-
b
ased
C
NN
an
aly
s
is
attain
ed
s
lig
h
tly
b
etter
p
er
f
o
r
m
a
n
ce
(
0
.
8
3
)
,
o
u
r
h
y
b
r
id
m
eth
o
d
s
ig
n
i
f
ican
tly
im
p
r
o
v
e
d
d
etec
tio
n
ac
cu
r
ac
y
(
0
.
8
8
)
,
p
r
ec
is
io
n
(
0
.
8
9
)
,
r
ec
all
(
0
.
8
4
)
,
an
d
F1
-
s
co
r
e
(
0
.
8
6
)
.
T
h
ese
r
esu
lts
s
u
r
p
ass
p
r
ev
io
u
s
ap
p
r
o
ac
h
es
r
e
p
o
r
te
d
b
y
Ott
et
a
l
.
[
1
]
an
d
Ku
m
ar
et
a
l
.
[
7
]
,
em
p
h
asizin
g
th
e
ad
v
an
tag
e
o
f
in
te
g
r
atin
g
A
p
r
io
r
i’
s
b
eh
a
v
io
r
al
p
atter
n
r
e
co
g
n
itio
n
with
C
NN’
s
tex
tu
al
an
aly
s
is
.
Su
ch
in
teg
r
atio
n
ef
f
ec
tiv
ely
ad
d
r
ess
es
lim
itatio
n
s
id
en
tifie
d
in
p
r
i
o
r
r
esear
ch
,
p
ar
ticu
lar
ly
in
ca
p
tu
r
in
g
s
u
b
tle
s
p
am
in
d
icato
r
s
th
at
in
d
iv
i
d
u
al
m
eth
o
d
s
f
a
il to
d
etec
t.
B
y
ex
am
in
in
g
b
o
th
b
eh
a
v
io
r
a
l
an
d
co
n
ten
t
p
atter
n
s
,
th
e
h
y
b
r
id
m
o
d
el
m
ad
e
it
h
ar
d
er
f
o
r
s
p
am
m
er
s
to
ev
ad
e
d
etec
tio
n
.
I
ts
s
ca
lab
i
lity
allo
ws
h
an
d
lin
g
lar
g
e
d
at
asets
,
m
ak
in
g
it
id
ea
l
f
o
r
e
-
c
o
m
m
er
ce
p
latf
o
r
m
s
with
h
ig
h
r
e
v
iew
v
o
lu
m
es.
Par
allel
p
r
o
ce
s
s
in
g
an
d
i
n
cr
em
en
tal
lear
n
in
g
ca
n
h
el
p
o
v
er
co
m
e
r
ea
l
-
tim
e
p
r
o
ce
s
s
in
g
ch
allen
g
es
[
2
3
]
.
Hig
h
-
q
u
ality
,
well
-
lab
eled
d
atas
ets
[
2
4
]
ar
e
cr
u
cial
f
o
r
m
ain
ta
in
in
g
p
er
f
o
r
m
an
ce
,
as
p
o
o
r
d
ata
q
u
ality
ca
n
r
ed
u
ce
ac
cu
r
ac
y
.
T
h
e
ef
f
ec
tiv
e
n
ess
o
f
o
u
r
h
y
b
r
id
s
tr
ateg
y
i
s
f
u
r
th
er
s
u
p
p
o
r
ted
b
y
T
ab
le
2
,
wh
ich
s
h
o
ws
th
at
th
e
m
o
d
el
ca
n
r
eliab
ly
d
is
tin
g
u
i
s
h
b
etwe
en
r
ev
iews
th
at
a
r
e
s
p
am
an
d
t
h
o
s
e
th
at
ar
e
n
o
t
.
T
h
is
c
o
m
p
r
eh
en
s
iv
e
a
n
d
s
ca
lab
le
s
o
l
u
tio
n
[
2
5
]
,
[
2
6
]
en
h
a
n
ce
s
th
e
i
n
teg
r
ity
o
f
o
n
lin
e
r
ev
iew
s
y
s
tem
s
b
y
d
eliv
er
i
n
g
m
o
r
e
ac
cu
r
ate
s
p
am
d
etec
tio
n
.
T
ab
le
2
.
S
am
p
le
o
u
tp
u
ts
co
m
p
ar
in
g
s
p
am
a
n
d
n
o
n
-
s
p
am
s
en
t
en
ce
s
as id
en
tifie
d
R
e
v
i
e
w
I
D
R
e
v
i
e
w
t
e
x
t
G
r
o
u
n
d
t
r
u
t
h
P
r
e
d
i
c
t
e
d
l
a
b
e
l
1
"
Th
i
s
i
s
t
h
e
b
e
s
t
p
r
o
d
u
c
t
e
v
e
r
!
I
l
o
v
e
i
t
s
o
m
u
c
h
,
h
i
g
h
l
y
r
e
c
o
mm
e
n
d
t
o
e
v
e
r
y
o
n
e
!
"
S
p
a
m
S
p
a
m
2
"
Th
e
q
u
a
l
i
t
y
w
a
s
p
o
o
r
a
n
d
n
o
t
a
s
d
e
sc
r
i
b
e
d
.
V
e
r
y
d
i
sa
p
p
o
i
n
t
e
d
.
"
N
o
n
-
s
p
a
m
N
o
n
-
s
p
a
m
3
"
A
maz
i
n
g
ser
v
i
c
e
a
n
d
f
a
st
d
e
l
i
v
e
r
y
.
F
i
v
e
st
a
r
s!
"
S
p
a
m
S
p
a
m
4
"
P
r
o
d
u
c
t
a
r
r
i
v
e
d
o
n
t
i
me
a
n
d
a
s
d
e
scri
b
e
d
.
S
a
t
i
sf
i
e
d
w
i
t
h
t
h
e
p
u
r
c
h
a
se.
"
N
o
n
-
s
p
a
m
N
o
n
-
s
p
a
m
5
"
Ex
c
e
l
l
e
n
t
v
a
l
u
e
f
o
r
m
o
n
e
y
.
W
i
l
l
b
u
y
a
g
a
i
n
!
"
S
p
a
m
S
p
a
m
6
"
I
t
e
m w
a
s fau
l
t
y
a
n
d
c
u
st
o
mer s
e
r
v
i
c
e
w
a
s
u
n
h
e
l
p
f
u
l
.
W
o
u
l
d
n
o
t
b
u
y
f
r
o
m
h
e
r
e
a
g
a
i
n
.
"
N
o
n
-
s
p
a
m
N
o
n
-
s
p
a
m
4.
CO
NCLU
SI
O
N
AND
F
U
T
U
RE
WO
RK
Ou
r
s
tu
d
y
p
r
o
p
o
s
ed
a
h
y
b
r
i
d
s
p
am
d
etec
tio
n
ap
p
r
o
ac
h
i
n
teg
r
atin
g
b
eh
av
io
r
al
an
aly
s
is
(
Ap
r
io
r
i
alg
o
r
ith
m
)
an
d
co
n
ten
t
an
aly
s
is
(
C
NN
)
to
en
h
an
ce
t
h
e
d
ete
ctio
n
o
f
s
p
am
r
e
v
iews
o
n
e
-
c
o
m
m
er
ce
p
latf
o
r
m
s
.
T
h
e
r
esu
lts
p
r
o
v
id
e
s
tr
o
n
g
ev
id
en
ce
th
at
co
m
b
i
n
in
g
b
eh
a
v
io
r
al
an
d
tex
tu
al
an
aly
s
es
s
ig
n
if
ican
tly
im
p
r
o
v
es
d
etec
tio
n
,
ac
h
iev
in
g
an
AUC
-
R
OC
o
f
0
.
9
2
,
alo
n
g
with
p
r
e
cisi
o
n
(
0
.
8
9
)
,
r
ec
all
(
0
.
8
4
)
,
F1
-
s
co
r
e
(
0
.
8
6
)
,
an
d
ac
cu
r
ac
y
(
0
.
8
8
)
.
T
h
is
clea
r
ly
in
d
icate
s
th
at
in
teg
r
atin
g
th
ese
ap
p
r
o
ac
h
es
ef
f
ec
tiv
el
y
ca
p
tu
r
es
s
u
b
tle
s
p
am
in
d
icato
r
s
,
o
f
ten
m
is
s
ed
wh
en
u
s
in
g
b
eh
av
io
r
al
o
r
co
n
te
n
t
-
ba
s
ed
m
eth
o
d
s
alo
n
e.
T
h
is
s
tu
d
y
ex
p
lo
r
ed
a
co
m
p
r
eh
e
n
s
iv
e
h
y
b
r
id
ap
p
r
o
a
ch
in
teg
r
atin
g
b
eh
av
i
o
r
al
p
atter
n
s
v
ia
t
h
e
Ap
r
io
r
i
al
g
o
r
ith
m
an
d
tex
t
u
al
an
al
y
s
is
th
r
o
u
g
h
C
NN.
Ho
wev
er
,
ce
r
t
ain
lim
itatio
n
s
r
em
ain
.
First,
th
e
m
o
d
el'
s
p
er
f
o
r
m
a
n
ce
h
ea
v
ily
d
ep
en
d
s
o
n
th
e
qu
ality
an
d
la
b
elin
g
ac
cu
r
ac
y
o
f
th
e
d
ataset;
m
is
lab
eled
r
ev
iews
o
r
s
u
b
tle
s
p
am
i
n
d
icato
r
s
m
ay
af
f
ec
t
d
etec
tio
n
ac
cu
r
ac
y
.
Seco
n
d
ly
,
th
e
co
m
p
u
tatio
n
al
c
o
m
p
lex
it
y
o
f
th
e
Ap
r
i
o
r
i
alg
o
r
ith
m
m
ig
h
t
lim
it
s
ca
lab
ilit
y
with
ex
tr
em
ely
lar
g
e
d
atasets
o
r
r
ea
l
-
tim
e
s
ce
n
a
r
io
s
with
o
u
t
ad
d
itio
n
al
o
p
tim
izatio
n
.
Fu
r
th
er
an
d
m
o
r
e
ex
ten
s
iv
e
ev
alu
atio
n
s
o
n
d
iv
e
r
s
e
d
atasets
ar
e
n
ec
ess
ar
y
to
co
n
f
ir
m
th
e
g
en
e
r
aliza
b
ilit
y
an
d
s
ca
lab
ilit
y
o
f
o
u
r
p
r
o
p
o
s
ed
h
y
b
r
id
ap
p
r
o
ac
h
,
es
p
ec
ially
r
eg
ar
d
in
g
h
a
n
d
lin
g
h
ig
h
-
v
o
l
u
m
e
r
ev
iew
p
latf
o
r
m
s
an
d
a
d
ap
tiv
e
s
p
am
b
eh
av
io
r
s
.
Ou
r
s
tu
d
y
d
em
o
n
s
tr
ates
th
at
in
teg
r
atin
g
b
eh
av
i
o
r
al
p
atter
n
s
an
d
tex
tu
al
f
ea
tu
r
es
s
ig
n
if
ican
tly
im
p
r
o
v
es
s
p
a
m
d
etec
tio
n
ac
cu
r
ac
y
o
n
e
-
c
o
m
m
er
ce
p
latf
o
r
m
s
.
Fu
tu
r
e
r
esear
c
h
ca
n
f
u
r
th
er
en
h
a
n
ce
t
h
is
ap
p
r
o
ac
h
b
y
e
x
p
lo
r
i
n
g
p
r
iv
a
cy
-
p
r
eser
v
i
n
g
m
eth
o
d
s
s
u
ch
as
f
ed
er
ated
lear
n
in
g
to
s
ec
u
r
ely
lev
er
a
g
e
u
s
er
b
eh
av
io
r
d
ata
with
o
u
t
c
o
m
p
r
o
m
is
in
g
u
s
er
p
r
iv
ac
y
.
A
d
d
itio
n
ally
,
f
u
tu
r
e
s
tu
d
ies
co
u
ld
f
o
cu
s
o
n
d
ev
elo
p
i
n
g
r
o
b
u
s
t
tech
n
iq
u
es
r
esil
ien
t
to
ad
v
er
s
ar
ial
attac
k
s
,
wh
er
e
s
p
am
m
er
s
co
n
ti
n
u
o
u
s
ly
ev
o
lv
e
th
eir
s
tr
ateg
ies
to
ev
ad
e
d
etec
tio
n
.
L
astl
y
,
o
p
ti
m
izin
g
th
e
f
ea
tu
r
e
f
u
s
io
n
p
r
o
ce
s
s
an
d
ex
ten
d
in
g
th
e
h
y
b
r
i
d
ap
p
r
o
ac
h
to
cr
o
s
s
-
p
latf
o
r
m
s
p
am
d
etec
tio
n
co
u
ld
s
ig
n
if
ican
tly
im
p
r
o
v
e
p
r
ac
tica
l
ef
f
ec
tiv
en
ess
an
d
ad
ap
tab
ilit
y
ac
r
o
s
s
v
ar
io
u
s
e
-
co
m
m
e
r
ce
en
v
ir
o
n
m
en
ts
.
ACK
NO
WL
E
DG
E
M
E
NT
S
W
e
wo
u
ld
lik
e
to
th
an
k
e
v
er
y
o
n
e
wh
o
h
elp
ed
with
th
is
r
esear
ch
an
d
f
o
r
th
eir
cr
u
cial
a
d
v
ice
an
d
ass
is
tan
ce
.
W
e
th
an
k
team
at
Ma
d
h
y
an
c
h
al
Pro
f
ess
io
n
al
Un
iv
er
s
ity
in
B
h
o
p
al,
I
n
d
i
a,
as
well
as
o
u
r
co
lleag
u
es
in
th
e
Dep
ar
tm
en
t
o
f
C
o
m
p
u
ter
E
n
g
in
ee
r
i
n
g
,
f
o
r
th
eir
h
elp
f
u
l
talk
s
an
d
ass
is
tan
ce
.
W
e
ar
e
also
g
r
atef
u
l
f
o
r
o
u
r
in
s
titu
tio
n
'
s
r
eso
u
r
ce
s
an
d
f
ac
ilit
ies,
wh
ich
wer
e
cr
u
cial
to
th
e
ac
co
m
p
lis
h
m
en
t
o
f
th
is
s
tu
d
y
.
L
astl
y
,
we
ex
p
r
ess
o
u
r
g
r
atitu
d
e
to
o
u
r
f
r
ie
n
d
s
an
d
f
am
ily
f
o
r
th
eir
s
u
p
p
o
r
t a
n
d
p
atien
ce
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
5
2
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
9
,
No
.
3
,
Sep
tem
b
er
20
25
:
1
8
3
7
-
1
8
4
5
1844
F
UNDING
I
NF
O
R
M
A
T
I
O
N
T
h
e
au
t
h
o
r
s
c
o
n
f
ir
m
th
at
th
is
r
esear
ch
was
co
n
d
u
cted
with
o
u
t
an
y
f
in
an
cial
s
u
p
p
o
r
t
f
r
o
m
f
u
n
d
in
g
ag
en
cies
in
th
e
p
u
b
lic,
co
m
m
er
cial,
o
r
n
o
t
-
f
o
r
-
p
r
o
f
it
s
ec
to
r
s
.
No
ex
ter
n
al
g
r
an
ts
o
r
in
s
tit
u
tio
n
al
f
u
n
d
s
wer
e
u
tili
ze
d
d
u
r
in
g
th
e
s
tu
d
y
.
All
r
eso
u
r
ce
s
an
d
to
o
ls
u
s
ed
in
th
e
r
esear
ch
wer
e
s
elf
-
s
u
p
p
o
r
ted
b
y
th
e
au
th
o
r
s
.
T
h
is
d
ec
lar
atio
n
en
s
u
r
es tr
a
n
s
p
ar
en
cy
a
n
d
c
o
n
f
ir
m
s
th
e
in
d
e
p
en
d
en
t
n
atu
r
e
o
f
th
e
s
tu
d
y
.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
Au
th
o
r
s
h
av
e
co
n
tr
i
b
u
ted
s
ig
n
if
ican
tly
to
th
e
wo
r
k
u
s
in
g
th
e
C
R
ed
iT
tax
o
n
o
m
y
.
B
o
th
au
th
o
r
s
h
av
e
r
ev
iewe
d
an
d
a
p
p
r
o
v
ed
th
e
f
in
al
v
er
s
io
n
o
f
t
h
e
m
an
u
s
cr
ip
t
an
d
ag
r
ee
to
b
e
ac
co
u
n
tab
le
f
o
r
all
asp
ec
ts
o
f
th
e
wo
r
k
.
T
h
eir
c
o
n
tr
ib
u
tio
n
s
r
e
f
lect
s
u
b
s
tan
tial
in
v
o
lv
e
m
en
t
in
th
e
r
ese
ar
c
h
p
r
o
ce
s
s
,
in
clu
d
i
n
g
d
esig
n
,
a
n
aly
s
is
,
an
d
in
ter
p
r
etatio
n
.
T
h
e
y
co
lla
b
o
r
ated
th
r
o
u
g
h
o
u
t
t
h
e
wr
itin
g
,
r
ev
is
io
n
,
an
d
f
in
aliza
tio
n
o
f
th
e
m
an
u
s
cr
ip
t
to
en
s
u
r
e
its
q
u
ality
an
d
in
teg
r
ity
as p
er
:
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
Gan
esh
W
ay
al
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
Vijay
B
h
an
d
ar
i
✓
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
T
h
e
au
th
o
r
s
d
ec
lar
e
th
at
th
er
e
ar
e
n
o
co
n
f
licts
o
f
in
ter
est
r
elate
d
to
th
is
r
esear
ch
.
No
p
er
s
o
n
al,
p
r
o
f
ess
io
n
al,
o
r
f
in
an
cial
r
el
atio
n
s
h
ip
s
in
f
lu
en
ce
d
th
e
o
u
tco
m
es
o
f
th
is
s
tu
d
y
.
All
i
n
ter
p
r
etatio
n
s
an
d
co
n
clu
s
io
n
s
p
r
esen
ted
ar
e
s
o
lely
th
o
s
e
o
f
th
e
au
t
h
o
r
s
.
T
h
is
s
tatem
en
t
en
s
u
r
es
tr
an
s
p
ar
en
cy
an
d
u
p
h
o
l
d
s
th
e
in
teg
r
ity
o
f
th
e
r
esear
ch
p
r
o
ce
s
s
.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
d
ata
th
at
s
u
p
p
o
r
t
th
e
f
in
d
in
g
s
o
f
th
is
s
tu
d
y
ar
e
o
p
en
l
y
av
ailab
le
f
r
o
m
th
e
Am
az
o
n
Pro
d
u
ct
R
ev
iew
Data
s
et,
cu
r
ated
an
d
p
u
b
lis
h
ed
b
y
J
u
lian
Mc
Au
ley
[
2
]
an
d
co
lleag
u
es.
T
h
e
d
ata
s
et
ca
n
b
e
ac
ce
s
s
ed
th
r
o
u
g
h
t
h
e
UC
SD D
ata
R
ep
o
s
ito
r
y
.
RE
F
E
R
E
NC
E
S
[
1
]
M
.
O
t
t
,
C
.
C
a
r
d
i
e
,
a
n
d
J
.
H
a
n
c
o
c
k
,
“
Est
i
m
a
t
i
n
g
t
h
e
p
r
e
v
a
l
e
n
c
e
o
f
d
e
c
e
p
t
i
o
n
i
n
o
n
l
i
n
e
r
e
v
i
e
w
c
o
mm
u
n
i
t
i
e
s
,
”
i
n
Pro
c
e
e
d
i
n
g
s
o
f
t
h
e
2
1
s
t
i
n
t
e
rn
a
t
i
o
n
a
l
c
o
n
f
e
re
n
c
e
o
n
Wo
r
l
d
W
i
d
e
We
b
,
N
e
w
Y
o
r
k
,
N
Y
,
U
S
A
:
A
C
M
,
A
p
r
.
2
0
1
2
,
p
p
.
2
0
1
–
2
1
0
.
d
o
i
:
1
0
.
1
1
4
5
/
2
1
8
7
8
3
6
.
2
1
8
7
8
6
4
.
[
2]
J.
M
c
A
u
l
e
y
a
n
d
J.
L
e
sk
o
v
e
c
,
“
H
i
d
d
e
n
f
a
c
t
o
r
s
a
n
d
h
i
d
d
e
n
t
o
p
i
c
s
:
U
n
d
e
r
s
t
a
n
d
i
n
g
r
a
t
i
n
g
d
i
m
e
n
si
o
n
s
w
i
t
h
r
e
v
i
e
w
t
e
x
t
,
”
i
n
Re
c
S
y
s
2
0
1
3
-
Pro
c
e
e
d
i
n
g
s o
f
t
h
e
7
t
h
A
C
M
C
o
n
f
e
r
e
n
c
e
o
n
Re
c
o
m
m
e
n
d
e
r
S
y
st
e
m
s
,
N
e
w
Y
o
r
k
,
N
Y
,
U
S
A
:
A
C
M
,
O
c
t
.
2
0
1
3
,
p
p
.
1
6
5
–
1
7
2
.
d
o
i
:
1
0
.
1
1
4
5
/
2
5
0
7
1
5
7
.
2
5
0
7
1
6
3
.
[
3
]
N
.
Ji
n
d
a
l
a
n
d
B
.
Li
u
,
“
O
p
i
n
i
o
n
sp
a
m
a
n
d
a
n
a
l
y
s
i
s,
”
i
n
W
S
D
M’0
8
-
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
2
0
0
8
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
W
e
b
S
e
a
rc
h
a
n
d
D
a
t
a
Mi
n
i
n
g
,
N
e
w
Y
o
r
k
,
N
e
w
Y
o
r
k
,
U
S
A
:
A
C
M
P
r
e
ss
,
2
0
0
8
,
p
p
.
2
1
9
–
2
2
9
.
d
o
i
:
1
0
.
1
1
4
5
/
1
3
4
1
5
3
1
.
1
3
4
1
5
6
0
.
[
4
]
W
.
M
i
n
,
W
.
L
i
a
n
g
,
H
.
Y
i
n
,
Z.
W
a
n
g
,
M
.
L
i
,
a
n
d
A
.
La
l
,
“
Ex
p
l
a
i
n
a
b
l
e
d
e
e
p
b
e
h
a
v
i
o
r
a
l
s
e
q
u
e
n
c
e
c
l
u
st
e
r
i
n
g
f
o
r
t
r
a
n
sa
c
t
i
o
n
f
r
a
u
d
d
e
t
e
c
t
i
o
n
,
”
Ja
n
.
2
0
2
1
,
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
:
/
/
a
r
x
i
v
.
o
r
g
/
a
b
s/
2
1
0
1
.
0
4
2
8
5
[
5
]
M
.
Z
a
g
o
e
t
a
l
.
,
“
S
c
r
e
e
n
i
n
g
o
u
t
s
o
c
i
a
l
b
o
t
s
i
n
t
e
r
f
e
r
e
n
c
e
:
a
r
e
t
h
e
r
e
a
n
y
si
l
v
e
r
b
u
l
l
e
t
s
?
,
”
I
EE
E
C
o
m
m
u
n
i
c
a
t
i
o
n
s
M
a
g
a
z
i
n
e
,
v
o
l
.
5
7
,
n
o
.
8
,
p
p
.
9
8
–
1
0
4
,
A
u
g
.
2
0
1
9
,
d
o
i
:
1
0
.
1
1
0
9
/
M
C
O
M
.
2
0
1
9
.
1
8
0
0
5
2
0
.
[
6
]
M
.
C
r
a
w
f
o
r
d
,
T
.
M
.
K
h
o
s
h
g
o
f
t
a
a
r
,
J
.
D
.
P
r
u
s
a
,
A
.
N
.
R
i
c
h
t
e
r
,
a
n
d
H
.
A
l
N
a
j
a
d
a
,
“
S
u
r
v
e
y
o
f
r
e
v
i
e
w
s
p
a
m d
e
t
e
c
t
i
o
n
u
s
i
n
g
mac
h
i
n
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s,
”
J
o
u
rn
a
l
o
f
Bi
g
D
a
t
a
,
v
o
l
.
2
,
n
o
.
1
,
p
.
2
3
,
D
e
c
.
2
0
1
5
,
d
o
i
:
1
0
.
1
1
8
6
/
s
4
0
5
3
7
-
015
-
0
0
2
9
-
9.
[
7
]
N
.
K
u
mar,
D
.
V
e
n
u
g
o
p
a
l
,
L.
Q
i
u
,
a
n
d
S
.
K
u
mar,
“
D
e
t
e
c
t
i
n
g
r
e
v
i
e
w
m
a
n
i
p
u
l
a
t
i
o
n
o
n
o
n
l
i
n
e
p
l
a
t
f
o
r
ms w
i
t
h
h
i
e
r
a
r
c
h
i
c
a
l
s
u
p
e
r
v
i
se
d
l
e
a
r
n
i
n
g
,
”
J
o
u
r
n
a
l
o
f
M
a
n
a
g
e
m
e
n
t
I
n
f
o
rm
a
t
i
o
n
S
y
st
e
m
s
,
v
o
l
.
3
5
,
n
o
.
1
,
p
p
.
3
5
0
–
3
8
0
,
2
0
1
8
,
d
o
i
:
1
0
.
1
0
8
0
/
0
7
4
2
1
2
2
2
.
2
0
1
8
.
1
4
4
0
7
5
8
.
[
8
]
C
.
V
a
n
D
i
n
h
,
S
.
T
.
L
u
u
,
a
n
d
A
.
G
.
T.
N
g
u
y
e
n
,
“
D
e
t
e
c
t
i
n
g
sp
a
m
r
e
v
i
e
w
s
o
n
v
i
e
t
n
a
mes
e
e
-
c
o
mm
e
r
c
e
w
e
b
s
i
t
e
s
,
”
i
n
L
e
c
t
u
re
N
o
t
e
s
i
n
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
(
i
n
c
l
u
d
i
n
g
s
u
b
seri
e
s
L
e
c
t
u
re
N
o
t
e
s
i
n
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
L
e
c
t
u
re
N
o
t
e
s
i
n
Bi
o
i
n
f
o
rm
a
t
i
c
s)
,
v
o
l
.
1
3
7
5
7
LN
A
I
,
2
0
2
2
,
p
p
.
5
9
5
–
6
0
7
.
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
0
3
1
-
2
1
7
4
3
-
2
_
4
8
.
[
9
]
G
.
M
.
S
h
a
h
a
r
i
a
r
,
S
.
B
i
sw
a
s,
F
.
O
m
a
r
,
F
.
M
.
S
h
a
h
,
a
n
d
S
.
B
.
H
a
ss
a
n
,
“
S
p
a
m
r
e
v
i
e
w
d
e
t
e
c
t
i
o
n
u
si
n
g
d
e
e
p
l
e
a
r
n
i
n
g
,
”
2
0
1
9
I
EEE
1
0
t
h
A
n
n
u
a
l
I
n
f
o
rm
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
,
El
e
c
t
ro
n
i
c
s
a
n
d
M
o
b
i
l
e
C
o
m
m
u
n
i
c
a
t
i
o
n
C
o
n
f
e
r
e
n
c
e
,
I
EM
C
O
N
2
0
1
9
,
p
p
.
2
7
–
3
3
,
2
0
1
9
,
d
o
i
:
1
0
.
1
1
0
9
/
I
EM
C
O
N
.
2
0
1
9
.
8
9
3
6
1
4
8
.
[
1
0
]
Y
.
Le
C
u
n
,
L.
B
o
t
t
o
u
,
Y
.
B
e
n
g
i
o
,
a
n
d
P
.
H
a
f
f
n
e
r
,
“
G
r
a
d
i
e
n
t
-
b
a
s
e
d
l
e
a
r
n
i
n
g
a
p
p
l
i
e
d
t
o
d
o
c
u
m
e
n
t
r
e
c
o
g
n
i
t
i
o
n
,
”
P
ro
c
e
e
d
i
n
g
s
o
f
t
h
e
I
EEE
,
v
o
l
.
8
6
,
n
o
.
1
1
,
p
p
.
2
2
7
8
–
2
3
2
3
,
1
9
9
8
,
d
o
i
:
1
0
.
1
1
0
9
/
5
.
7
2
6
7
9
1
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4
7
5
2
A
h
yb
r
id
a
p
p
r
o
a
c
h
to
b
e
h
a
vio
r
a
l sp
a
m
r
ev
iew
d
etec
tio
n
…
(
Ga
n
esh
Wa
ya
l
)
1845
[
1
1
]
R
.
A
g
r
a
w
a
l
a
n
d
R
.
S
r
i
k
a
n
t
,
“
F
a
s
t
a
l
g
o
r
i
t
h
ms
f
o
r
m
i
n
i
n
g
a
ss
o
c
i
a
t
i
o
n
r
u
l
e
s
,
”
Pro
c
.
o
f
2
0
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
V
e
ry
L
a
r
g
e
D
a
t
a
B
a
ses
,
{
VLD
B’
9
4
}
,
p
p
.
4
8
7
–
4
9
9
,
1
9
9
4
,
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
c
i
t
e
s
e
e
r
.
i
s
t
.
p
su
.
e
d
u
/
a
g
r
a
w
a
l
9
4
f
a
s
t
.
h
t
m
l
[
1
2
]
J.
Y
a
o
,
Y
.
Z
h
e
n
g
,
a
n
d
H
.
J
i
a
n
g
,
“
A
n
e
n
s
e
mb
l
e
m
o
d
e
l
f
o
r
f
a
k
e
o
n
l
i
n
e
r
e
v
i
e
w
d
e
t
e
c
t
i
o
n
b
a
se
d
o
n
d
a
t
a
r
e
sam
p
l
i
n
g
,
f
e
a
t
u
r
e
p
r
u
n
i
n
g
,
a
n
d
p
a
r
a
m
e
t
e
r
o
p
t
i
m
i
z
a
t
i
o
n
,
”
I
EE
E
A
c
c
e
ss
,
v
o
l
.
9
,
p
p
.
1
6
9
1
4
–
1
6
9
2
7
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
1
.
3
0
5
1
1
7
4
.
[
1
3
]
D
.
D
e
B
a
r
r
a
n
d
H
.
W
e
c
h
sl
e
r
,
“
U
si
n
g
s
o
c
i
a
l
n
e
t
w
o
r
k
a
n
a
l
y
si
s
f
o
r
s
p
a
m
d
e
t
e
c
t
i
o
n
,
”
i
n
L
e
c
t
u
r
e
N
o
t
e
s
i
n
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
(
i
n
c
l
u
d
i
n
g
su
b
s
e
ri
e
s
L
e
c
t
u
re
N
o
t
e
s
i
n
Ar
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
L
e
c
t
u
re
N
o
t
e
s
i
n
B
i
o
i
n
f
o
rm
a
t
i
c
s)
,
v
o
l
.
6
0
0
7
LN
C
S
,
2
0
1
0
,
p
p
.
6
2
–
6
9
.
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
6
4
2
-
1
2
0
7
9
-
4
_
1
0
.
[
1
4
]
S
.
B
a
n
e
r
j
e
e
,
S
.
B
h
a
t
t
a
c
h
a
r
y
y
a
,
a
n
d
I
.
B
o
se
,
“
W
h
o
se
o
n
l
i
n
e
r
e
v
i
e
w
s
t
o
t
r
u
st
?
U
n
d
e
r
st
a
n
d
i
n
g
r
e
v
i
e
w
e
r
t
r
u
s
t
w
o
r
t
h
i
n
e
ss
a
n
d
i
t
s
i
mp
a
c
t
o
n
b
u
si
n
e
ss,”
D
e
c
i
si
o
n
S
u
p
p
o
r
t
S
y
st
e
m
s
,
v
o
l
.
9
6
,
p
p
.
1
7
–
2
6
,
A
p
r
.
2
0
1
7
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
d
ss.
2
0
1
7
.
0
1
.
0
0
6
.
[
1
5
]
R
.
T.
S
i
k
o
r
a
a
n
d
K
.
C
h
a
u
h
a
n
,
“
Es
t
i
mat
i
n
g
se
q
u
e
n
t
i
a
l
b
i
a
s
i
n
o
n
l
i
n
e
r
e
v
i
e
w
s:
A
K
a
l
m
a
n
f
i
l
t
e
r
i
n
g
a
p
p
r
o
a
c
h
,
”
K
n
o
w
l
e
d
g
e
-
B
a
se
d
S
y
s
t
e
m
s
,
v
o
l
.
2
7
,
p
p
.
3
1
4
–
3
2
1
,
M
a
r
.
2
0
1
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
k
n
o
s
y
s.
2
0
1
1
.
1
0
.
0
1
1
.
[
1
6
]
M
.
G
o
l
d
s
t
e
i
n
a
n
d
S
.
U
c
h
i
d
a
,
“
B
e
h
a
v
i
o
r
a
n
a
l
y
si
s
u
si
n
g
u
n
s
u
p
e
r
v
i
se
d
a
n
o
mal
y
d
e
t
e
c
t
i
o
n
,
”
T
h
e
1
0
t
h
J
o
i
n
t
W
o
r
k
sh
o
p
o
n
M
a
c
h
i
n
e
Pe
rc
e
p
t
i
o
n
a
n
d
R
o
b
o
t
i
c
s
,
n
o
.
O
c
t
o
b
e
r
,
2
0
1
4
.
[
1
7
]
H
.
A
.
A
l
-
K
a
b
b
i
,
M
.
R
.
F
e
i
z
i
-
D
e
r
a
k
h
s
h
i
,
a
n
d
S
.
P
a
s
h
a
z
a
d
e
h
,
“
M
u
l
t
i
-
t
y
p
e
f
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
a
n
d
e
a
r
l
y
f
u
si
o
n
f
r
a
me
w
o
r
k
f
o
r
S
M
S
sp
a
m
d
e
t
e
c
t
i
o
n
,”
I
E
EE
Ac
c
e
ss
,
v
o
l
.
1
1
,
p
p
.
1
2
3
7
5
6
–
1
2
3
7
6
5
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
3
.
3
3
2
7
8
9
7
.
[
1
8
]
D
.
S
a
v
a
g
e
,
X
.
Z
h
a
n
g
,
X
.
Y
u
,
P
.
C
h
o
u
,
a
n
d
Q
.
W
a
n
g
,
“
D
e
t
e
c
t
i
o
n
o
f
o
p
i
n
i
o
n
s
p
a
m
b
a
se
d
o
n
a
n
o
ma
l
o
u
s
r
a
t
i
n
g
d
e
v
i
a
t
i
o
n
,
”
Ex
p
e
r
t
S
y
s
t
e
m
s
w
i
t
h
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
4
2
,
n
o
.
2
2
,
p
p
.
8
6
5
0
–
8
6
5
7
,
D
e
c
.
2
0
1
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
sw
a
.
2
0
1
5
.
0
7
.
0
1
9
.
[
1
9
]
A
.
M
u
k
h
e
r
j
e
e
,
B
.
Li
u
,
a
n
d
N
.
G
l
a
n
c
e
,
“
S
p
o
t
t
i
n
g
f
a
k
e
r
e
v
i
e
w
e
r
g
r
o
u
p
s
i
n
c
o
n
su
mer
r
e
v
i
e
w
s,”
i
n
WW
W’1
2
-
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
2
1
s
t
A
n
n
u
a
l
C
o
n
f
e
re
n
c
e
o
n
W
o
rl
d
W
i
d
e
W
e
b
,
N
e
w
Y
o
r
k
,
N
Y
,
U
S
A
:
A
C
M
,
A
p
r
.
2
0
1
2
,
p
p
.
1
9
1
–
2
0
0
.
d
o
i
:
1
0
.
1
1
4
5
/
2
1
8
7
8
3
6
.
2
1
8
7
8
6
3
.
[
2
0
]
S
.
R
a
o
,
A
.
K
.
V
e
r
ma
,
a
n
d
T.
B
h
a
t
i
a
,
“
H
y
b
r
i
d
e
n
sem
b
l
e
f
r
a
mew
o
r
k
w
i
t
h
s
e
l
f
-
a
t
t
e
n
t
i
o
n
m
e
c
h
a
n
i
s
m
f
o
r
s
o
c
i
a
l
sp
a
m
d
e
t
e
c
t
i
o
n
o
n
i
mb
a
l
a
n
c
e
d
d
a
t
a
,
”
E
x
p
e
rt
S
y
st
e
m
s w
i
t
h
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
2
1
7
,
p
.
1
1
9
5
9
4
,
M
a
y
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
sw
a
.
2
0
2
3
.
1
1
9
5
9
4
.
[
2
1
]
L.
H
e
,
G
.
X
u
,
S
.
Jam
e
e
l
,
X
.
W
a
n
g
,
a
n
d
H
.
C
h
e
n
,
“
G
r
a
p
h
-
a
w
a
r
e
d
e
e
p
f
u
s
i
o
n
n
e
t
w
o
r
k
s
f
o
r
o
n
l
i
n
e
sp
a
m
r
e
v
i
e
w
d
e
t
e
c
t
i
o
n
,”
I
EEE
T
ra
n
s
a
c
t
i
o
n
s
o
n
C
o
m
p
u
t
a
t
i
o
n
a
l
S
o
c
i
a
l
S
y
st
e
m
s
,
v
o
l
.
1
0
,
n
o
.
5
,
p
p
.
2
5
5
7
–
2
5
6
5
,
O
c
t
.
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
T
C
S
S
.
2
0
2
2
.
3
1
8
9
8
1
3.
[
2
2
]
Y
.
R
e
n
a
n
d
D
.
J
i
,
“
N
e
u
r
a
l
n
e
t
w
o
r
k
s
f
o
r
d
e
c
e
p
t
i
v
e
o
p
i
n
i
o
n
sp
a
m
d
e
t
e
c
t
i
o
n
:
A
n
e
m
p
i
r
i
c
a
l
st
u
d
y
,
”
I
n
f
o
rm
a
t
i
o
n
S
c
i
e
n
c
e
s
,
v
o
l
.
3
8
5
–
3
8
6
,
p
p
.
2
1
3
–
2
2
4
,
A
p
r
.
2
0
1
7
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
n
s.
2
0
1
7
.
0
1
.
0
1
5
.
[
2
3
]
T.
R
.
S
r
e
e
a
n
d
R
.
Tr
i
p
a
t
h
i
,
“
F
a
k
e
r
e
v
i
e
w
d
e
t
e
c
t
i
o
n
u
si
n
g
e
v
i
d
e
n
t
i
a
l
c
l
a
ss
i
f
i
e
r
,
”
i
n
2
0
2
3
2
n
d
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
A
d
v
a
n
c
e
s
i
n
C
o
m
p
u
t
a
t
i
o
n
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
,
I
C
A
C
I
C
2
0
2
3
,
I
EEE,
D
e
c
.
2
0
2
3
,
p
p
.
1
–
5
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
A
C
I
C
5
9
4
5
4
.
2
0
2
3
.
1
0
4
3
5
3
4
3
.
[
2
4
]
X
.
Li
a
n
d
L.
C
h
e
n
,
“
F
a
k
e
r
e
v
i
e
w
d
e
t
e
c
t
i
o
n
u
s
i
n
g
d
e
e
p
n
e
u
r
a
l
n
e
t
w
o
r
k
s wit
h
mu
l
t
i
m
o
d
a
l
f
e
a
t
u
r
e
f
u
s
i
o
n
m
e
t
h
o
d
,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
P
a
ra
l
l
e
l
a
n
d
D
i
st
r
i
b
u
t
e
d
S
y
s
t
e
m
s
-
I
C
PAD
S
,
I
EEE,
D
e
c
.
2
0
2
3
,
p
p
.
2
8
6
9
–
2
8
7
2
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
P
A
D
S
6
0
4
5
3
.
2
0
2
3
.
0
0
4
1
1
.
[
2
5
]
S
.
R
a
y
a
n
a
a
n
d
L
.
A
k
o
g
l
u
,
“
C
o
l
l
e
c
t
i
v
e
o
p
i
n
i
o
n
s
p
a
m
d
e
t
e
c
t
i
o
n
:
B
r
i
d
g
i
n
g
r
e
v
i
e
w
n
e
t
w
o
r
k
s
a
n
d
m
e
t
a
d
a
t
a
,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
AC
M
S
I
G
K
D
D
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
K
n
o
w
l
e
d
g
e
D
i
sc
o
v
e
ry
a
n
d
D
a
t
a
M
i
n
i
n
g
,
N
e
w
Y
o
r
k
,
N
Y
,
U
S
A
:
A
C
M
,
A
u
g
.
2
0
1
5
,
p
p
.
9
8
5
–
9
9
4
.
d
o
i
:
1
0
.
1
1
4
5
/
2
7
8
3
2
5
8
.
2
7
8
3
3
7
0
.
[
26]
J.
K
.
R
o
u
t
,
A
.
D
a
l
mi
a
,
K
.
K
.
R
.
C
h
o
o
,
S
.
B
a
k
sh
i
,
a
n
d
S
.
K
.
Je
n
a
,
“
R
e
v
i
s
i
t
i
n
g
sem
i
-
su
p
e
r
v
i
s
e
d
l
e
a
r
n
i
n
g
f
o
r
o
n
l
i
n
e
d
e
c
e
p
t
i
v
e
r
e
v
i
e
w
d
e
t
e
c
t
i
o
n
,
”
I
EEE
Ac
c
e
ss
,
v
o
l
.
5
,
p
p
.
1
3
1
9
–
1
3
2
7
,
2
0
1
7
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
E
S
S
.
2
0
1
7
.
2
6
5
5
0
3
2
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
G
a
n
e
sh
Wa
y
a
l
is
P
h
.
D.
(
S
c
h
o
lar)
a
t
M
a
d
h
y
a
n
c
h
a
l
P
r
o
fe
ss
io
n
a
l
U
n
iv
e
rsit
y
,
Bh
o
p
a
l
,
M
a
d
h
y
a
P
ra
d
e
sh
,
In
d
ia
.
He
h
o
l
d
s
a
n
M
.
Tec
h
a
n
d
B.
E.
in
Co
m
p
u
ter
En
g
in
e
e
ri
n
g
.
His
re
se
a
rc
h
in
tere
sts
fo
c
u
s
o
n
AI
a
n
d
M
L
,
c
o
n
tri
b
u
ti
n
g
to
a
d
v
a
n
c
e
m
e
n
ts
in
t
h
e
se
field
s.
C
u
rre
n
tl
y
p
u
rsu
i
n
g
h
is
d
o
c
t
o
ra
l
stu
d
ies
,
h
e
c
o
n
ti
n
u
e
s
to
b
u
il
d
e
x
p
e
rti
se
i
n
t
h
e
in
ters
e
c
ti
o
n
o
f
c
o
m
p
u
ter
sc
ien
c
e
a
n
d
in
telli
g
e
n
t
s
y
ste
m
s.
F
o
r
f
u
rth
e
r
i
n
fo
rm
a
ti
o
n
o
r
c
o
ll
a
b
o
ra
ti
o
n
,
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
g
a
n
e
sh
w2
0
0
6
@
g
m
a
il
.
c
o
m
.
Vija
y
B
h
a
n
d
a
r
i
re
c
e
iv
e
d
h
is
P
h
.
D.
i
n
Co
m
p
u
ter
En
g
in
e
e
ri
n
g
a
l
o
n
g
wit
h
a
n
M
.
Tec
h
.
a
n
d
B.
E.
in
C
o
m
p
u
ter
En
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
Cu
rr
e
n
tl
y
wo
r
k
i
n
g
a
s
As
so
c
iate
P
ro
fe
ss
o
r
a
t
M
a
d
h
y
a
n
c
h
a
l
P
r
o
fe
ss
io
n
a
l
U
n
iv
e
rsit
y
,
B
h
o
p
a
l,
M
a
d
h
y
a
P
ra
d
e
sh
.
His
a
c
a
d
e
m
ic
b
a
c
k
g
ro
u
n
d
f
o
c
u
se
s
o
n
a
d
v
a
n
c
e
d
c
o
m
p
u
tati
o
n
a
l
tec
h
n
o
lo
g
ies
,
p
a
rt
icu
larly
AI
a
n
d
M
L,
w
h
ich
a
re
c
e
n
tral
to
h
is
re
se
a
rc
h
a
n
d
e
x
p
e
rti
se
.
Dr.
Bh
a
n
d
a
ri'
s
wo
rk
in
teg
ra
tes
th
e
se
c
u
tt
i
n
g
-
e
d
g
e
tec
h
n
o
l
o
g
ies
wit
h
p
ra
c
ti
c
a
l
a
p
p
l
ica
ti
o
n
s
i
n
c
o
m
p
u
ter
sc
ien
c
e
.
F
o
r
a
n
y
i
n
q
u
iri
e
s
o
r
p
ro
fe
ss
io
n
a
l
c
o
ll
a
b
o
ra
ti
o
n
,
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
b
h
a
n
d
a
riv
i
jay
3
1
4
@
g
m
a
il
.
c
o
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.