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
8
,
No
.
1
,
A
p
r
il
20
2
5
,
p
p
.
478
~
485
I
SS
N:
2
502
-
4
7
52
,
DOI
: 1
0
.
1
1
5
9
1
/ijee
cs
.v
3
8
.
i
1
.
pp
478
-
4
8
5
478
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ee
cs
.
ia
esco
r
e.
co
m
Fake revi
ew
det
ec
tion usin
g
enhanc
ed ensembl
e supp
o
rt
v
ector
ma
chine sy
stem o
n e
-
co
mm
er
ce pla
tf
o
rm
Seenia
J
o
s
eph,
S.
H
em
a
la
t
ha
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
,
K
a
r
p
a
g
a
m A
c
a
d
e
m
y
o
f
H
i
g
h
e
r
Ed
u
c
a
t
i
o
n
,
C
o
i
m
b
a
t
o
r
e
,
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
Ju
n
18
,
2
0
2
4
R
ev
is
ed
Oct
21
,
2
0
2
4
Acc
ep
ted
Oct
30
,
2
0
2
4
Du
e
to
th
e
q
u
ick
g
ro
wt
h
o
f
o
n
li
n
e
m
a
rk
e
ti
n
g
tran
sa
c
ti
o
n
s,
in
c
l
u
d
i
n
g
b
u
y
in
g
a
n
d
se
ll
in
g
,
fa
k
e
re
v
iew
s
a
re
c
re
a
ted
to
p
ro
m
o
te
th
e
p
r
o
d
u
c
t
m
a
rk
e
t
a
n
d
m
islea
d
n
e
w
c
u
sto
m
e
rs.
E
-
c
o
m
m
e
rc
e
c
u
sto
m
e
rs
c
a
n
p
o
st
re
v
iew
s
a
n
d
c
o
m
m
e
n
ts
o
n
th
e
g
o
o
d
s
o
r
s
e
rv
ice
s
th
e
y
o
b
tain
e
d
.
Be
f
o
re
m
a
k
in
g
a
p
u
rc
h
a
se
,
n
e
w
c
u
sto
m
e
rs
fre
q
u
e
n
tl
y
re
a
d
th
e
fe
e
d
b
a
c
k
a
n
d
c
o
m
m
e
n
ts
p
o
ste
d
o
n
th
e
we
b
site.
No
wa
d
a
y
s c
u
sto
m
e
rs fi
n
d
it
v
e
ry
d
iff
icu
lt
t
o
id
e
n
t
ify
wh
e
th
e
r
th
e
re
v
iew
s
a
re
fa
k
e
o
r
n
o
t,
b
u
t
d
o
in
g
so
is
e
ss
e
n
ti
a
l.
S
o
,
i
t'
s
v
e
ry
c
ru
c
ial
to
d
e
v
e
lo
p
a
n
o
n
li
n
e
sp
a
m
d
e
tec
ti
o
n
sy
ste
m
t
o
h
e
lp
b
o
th
c
o
n
su
m
e
rs
a
n
d
p
ro
d
u
c
e
rs
in
th
e
ir
d
e
c
isio
n
-
m
a
k
i
n
g
.
Th
e
re
v
iew
e
r'
s
b
e
h
a
v
io
u
r
a
n
d
i
m
p
o
rtan
t
re
v
iew
c
h
a
ra
c
teristics
c
a
n
h
e
lp
y
o
u
i
d
e
n
ti
f
y
fa
k
e
re
v
iew
s.
Th
e
imp
o
rtan
c
e
o
f
th
is
stu
d
y
is
to
d
e
v
e
l
o
p
a
fa
k
e
re
v
iew
d
e
tec
ti
o
n
sy
ste
m
o
n
e
-
c
o
m
m
e
rc
e
p
latfo
rm
s
u
si
n
g
a
n
e
n
h
a
n
c
e
d
e
n
se
m
b
le
su
p
p
o
r
t
v
e
c
to
r
m
a
c
h
i
n
e
sy
ste
m
in
wh
ich
th
e
Eu
c
l
id
e
a
n
d
ista
n
c
e
is
re
p
lac
e
d
with
th
e
M
a
h
a
lan
o
b
is
d
istan
c
e
m
e
tri
c
.
Re
v
iew
tex
ts
c
o
ll
e
c
ted
fr
o
m
Am
a
z
o
n
a
n
d
Ye
lp
we
re
g
iv
e
n
a
s
in
p
u
t
d
a
ta se
ts i
n
to
th
e
c
o
n
stru
c
te
d
m
o
d
e
l
a
n
d
c
las
sified
a
s fa
k
e
o
r
re
a
l.
K
ey
w
o
r
d
s
:
Dis
tan
ce
m
etr
ic
E
n
s
em
b
le
m
eth
o
d
Fak
e
r
ev
iews
d
etec
tio
n
Su
p
er
v
is
ed
m
ac
h
i
n
e
lear
n
in
g
Featu
r
e
s
elec
tio
n
Su
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
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
:
Seen
ia
J
o
s
ep
h
Dep
ar
tm
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
,
Kar
p
ag
a
m
Aca
d
em
y
o
f
H
ig
h
er
E
d
u
ca
tio
n
C
o
im
b
ato
r
e
,
I
n
d
ia
E
m
ail:
s
ee
n
iajo
s
ep
h
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
T
h
e
in
ter
n
et
h
as
b
ec
o
m
e
a
co
n
ten
t
cr
ea
tio
n
p
latf
o
r
m
w
h
er
e
p
eo
p
le
ex
p
r
ess
th
eir
o
p
in
io
n
s
an
d
ex
p
er
ien
ce
s
,
s
ig
n
if
ican
tly
im
p
ac
tin
g
cu
s
to
m
er
s
a
n
d
b
u
s
in
ess
es.
Po
ten
tial
cu
s
to
m
er
s
o
f
ten
ch
ec
k
r
ev
iews
b
ef
o
r
e
m
ak
i
n
g
a
p
u
r
c
h
ase.
R
ev
iews
h
elp
p
o
ten
tial
cu
s
to
m
er
s
b
etter
u
n
d
er
s
ta
n
d
o
th
e
r
p
eo
p
le'
s
ex
p
er
ien
ce
s
,
esp
ec
ially
wh
en
ch
o
o
s
in
g
b
et
wee
n
p
u
r
c
h
asin
g
a
p
r
o
d
u
ct
o
r
s
er
v
ice.
C
h
ad
ch
an
k
ar
et
a
l.
[
1
]
o
b
s
er
v
e
d
th
at
8
1
%
o
f
in
d
iv
id
u
als
r
esear
ch
p
r
o
d
u
cts
o
n
lin
e
b
ef
o
r
e
b
u
y
in
g
,
an
d
if
v
er
if
ied
b
u
y
er
s
r
ep
o
r
t
b
a
d
cu
s
to
m
er
s
er
v
ice,
o
v
er
5
8
%
wi
ll
ce
ase
tr
an
s
ac
tio
n
s
.
T
h
is
u
n
d
e
r
s
co
r
es
th
e
s
ig
n
if
ican
t
in
f
lu
e
n
ce
o
f
b
u
y
er
f
ee
d
b
ac
k
o
n
p
u
r
c
h
asin
g
d
ec
is
io
n
s
[
2
]
.
Ho
wev
er
,
n
o
t
al
l
r
ev
iews
o
f
th
e
p
r
o
d
u
ct
o
n
th
e
in
ter
n
et
ar
e
g
en
u
in
e.
Ma
licio
u
s
u
s
er
s
o
f
ten
p
o
s
t
f
ak
e
r
e
v
iews
to
m
is
lead
c
u
s
to
m
er
s
in
to
p
r
o
m
o
tin
g
o
r
d
o
wn
g
r
ad
in
g
a
ta
r
g
et
p
r
o
d
u
ct
o
r
s
e
r
v
ice.
Fak
e
r
ev
iews
o
n
e
-
c
o
m
m
er
ce
p
latf
o
r
m
s
m
i
s
lead
co
n
s
u
m
er
s
,
lea
d
in
g
t
h
e
m
to
m
a
k
e
p
o
o
r
l
y
in
f
o
r
m
e
d
p
u
r
ch
asin
g
d
ec
is
io
n
s
an
d
p
o
te
n
tially
r
ec
eiv
e
s
u
b
p
ar
p
r
o
d
u
cts.
T
h
is
er
o
s
io
n
o
f
tr
u
s
t
ca
n
d
am
ag
e
c
o
n
s
u
m
er
co
n
f
id
e
n
ce
in
th
e
p
latf
o
r
m
an
d
th
e
r
e
p
u
tatio
n
o
f
h
o
n
est
b
u
s
in
ess
es.
T
h
is
ar
ti
cle
f
o
cu
s
es
o
n
d
ev
elo
p
in
g
an
ef
f
icien
t
m
eth
o
d
to
id
en
tify
f
a
k
e
r
ev
iews
o
n
e
-
c
o
m
m
er
ce
p
latf
o
r
m
s
to
h
elp
b
o
th
co
n
s
u
m
e
r
s
an
d
p
r
o
d
u
ce
r
s
in
th
eir
d
ec
is
io
n
-
m
ak
in
g
.
T
o
en
s
u
r
e
th
e
i
n
teg
r
ity
o
f
o
n
lin
e
r
ev
iews,
it
is
cr
u
cial
a
n
d
n
ec
ess
ar
y
to
c
r
ea
te
ef
f
ici
en
t
to
o
ls
to
id
en
tify
o
n
lin
e
r
e
v
iewe
r
s
.
T
h
e
ty
p
e
o
f
r
ev
iew
an
d
f
ea
tu
r
e
s
m
en
tio
n
ed
th
at
ar
e
n
o
t
d
ir
e
ctly
r
elate
d
to
th
e
co
n
ten
t
p
lay
a
r
o
le
in
id
e
n
tif
y
in
g
f
ak
e
r
ev
iews.
H
o
wev
er
,
f
ak
e
r
ev
iews
m
ay
r
eq
u
i
r
e
t
h
e
cr
ea
tio
n
o
f
o
th
er
ch
ar
ac
ter
is
tics
r
elate
d
to
th
e
r
ev
iewe
r
h
im
s
elf
,
s
u
ch
as
th
e
tim
e/d
ate
o
f
th
e
ass
es
s
m
en
t
o
r
h
is
wr
itin
g
s
ty
le.
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:
2
5
0
2
-
4
7
52
F
a
ke
r
ev
iew
d
etec
tio
n
u
s
in
g
e
n
h
a
n
ce
d
en
s
emb
le
s
u
p
p
o
r
t v
ec
to
r
…
(
S
ee
n
ia
J
o
s
ep
h
)
479
T
h
er
ef
o
r
e,
th
e
s
u
cc
ess
f
u
l
f
ea
tu
r
e
ex
tr
ac
tio
n
o
f
r
ev
iews
lead
s
to
th
e
s
u
cc
ess
f
u
l
r
ec
o
g
n
itio
n
o
f
f
ak
e
r
ev
iews.
T
h
is
s
tu
d
y
aim
s
to
cr
ea
te
a
f
ak
e
r
ev
iew
d
etec
tio
n
s
y
s
tem
f
o
r
e
-
co
m
m
er
ce
p
latf
o
r
m
s
b
y
u
t
ilizin
g
an
ad
v
an
ce
d
en
s
em
b
le
s
u
p
p
o
r
t
v
ec
to
r
m
a
ch
in
e
(
SVM)
m
o
d
el,
wh
ic
h
r
ep
lace
s
th
e
E
u
clid
ea
n
d
is
tan
ce
m
etr
ic
with
th
e
Ma
h
alan
o
b
is
d
is
tan
ce
m
etr
ic.
E
u
clid
ea
n
d
is
tan
ce
is
a
co
m
m
o
n
m
et
r
i
c
f
o
r
m
ea
s
u
r
in
g
th
e
d
i
s
tan
ce
b
etwe
en
two
p
o
in
ts
in
a
f
ea
tu
r
e
s
p
ac
e.
Ho
wev
er
,
it
ass
u
m
es
th
at
th
e
f
ea
tu
r
es
ar
e
u
n
c
o
r
r
elate
d
a
n
d
h
a
v
e
th
e
s
am
e
v
ar
ian
ce
.
On
th
e
o
th
er
h
an
d
,
Ma
h
alan
o
b
is
d
is
tan
ce
ac
co
u
n
ts
f
o
r
t
h
e
co
r
r
elatio
n
b
etwe
en
f
ea
tu
r
e
s
an
d
th
e
v
ar
ian
ce
wi
th
in
th
e
d
ata.
T
h
is
m
a
k
es
it
m
o
r
e
s
u
itab
le
f
o
r
ca
s
es
with
co
r
r
elate
d
f
ea
tu
r
es
o
r
d
if
f
er
in
g
s
ca
les.
B
y
r
ep
lacin
g
E
u
clid
ea
n
d
is
tan
ce
with
Ma
h
alan
o
b
is
d
is
tan
ce
in
SVM,
we
aim
to
ac
h
iev
e
a
m
o
r
e
n
u
a
n
ce
d
an
d
ac
cu
r
ate
d
is
tan
ce
m
ea
s
u
r
em
e
n
t,
im
p
r
o
v
in
g
th
e
d
etec
tio
n
o
f
f
ak
e
r
ev
iews.
T
h
is
en
h
a
n
ce
d
SVM
is
th
en
en
s
em
b
led
with
d
if
f
e
r
en
t c
lass
if
ier
s
f
o
r
b
etter
r
esu
lts
th
an
th
e
co
n
v
e
n
tio
n
al
m
eth
o
d
.
Gr
ee
n
g
r
ad
[
3
]
b
eliev
es
th
at
im
p
lem
en
tin
g
n
ew
n
o
v
el
a
lg
o
r
ith
m
s
an
d
id
ea
s
ca
n
in
cr
ea
s
e
th
e
p
er
f
o
r
m
an
ce
o
f
a
s
p
am
d
etec
t
io
n
s
y
s
te
m
.
Alter
n
ativ
ely
,
C
h
av
o
lla
et
a
l.
[
4
]
an
d
C
lu
n
e
[
5
]
ar
g
u
e
th
at
r
ath
e
r
th
an
lo
o
k
i
n
g
f
o
r
n
ew
id
ea
s
,
it
is
m
o
r
e
u
s
ef
u
l
to
im
p
r
o
v
e
th
e
f
u
n
ctio
n
i
n
g
o
f
ex
is
tin
g
s
y
s
tem
s
.
T
h
e
d
ev
elo
p
ed
s
y
s
tem
in
v
o
lv
es
en
h
an
cin
g
t
h
e
SVM
class
if
ier
b
y
o
p
tim
i
zin
g
th
e
s
p
ee
d
b
y
r
em
o
v
i
n
g
ir
r
elev
an
t
s
u
p
p
o
r
t
v
ec
to
r
s
to
r
ed
u
ce
th
e
n
u
m
b
er
o
f
co
m
p
u
tatio
n
s
in
v
o
lv
ed
a
n
d
u
tili
zin
g
Ma
h
alan
o
b
is
d
is
tan
ce
m
atr
ic
to
im
p
r
o
v
e
th
e
co
n
s
tr
u
ctio
n
o
f
h
y
p
e
r
p
lan
e
in
class
if
icat
io
n
.
Am
o
n
g
o
n
lin
e
m
er
ch
an
ts
,
Am
az
o
n
h
as
b
ee
n
ch
o
s
en
f
o
r
th
e
s
tu
d
y
'
s
ap
p
licatio
n
s
ec
tio
n
.
B
ec
au
s
e
o
f
Am
az
o
n
'
s
d
o
m
in
an
ce
in
o
n
lin
e
r
etailin
g
,
we
ch
o
s
e
th
eir
d
ataset.
Am
az
o
n
is
a
s
izab
le,
well
-
estab
lis
h
ed
o
n
lin
e
r
etailer
th
at
o
f
f
er
s
a
v
a
r
iety
o
f
d
atasets
f
o
r
m
ac
h
in
e
lear
n
in
g
ap
p
licatio
n
s
.
T
h
e
Yelp
d
ataset
is
p
r
o
v
id
ed
as
th
e
s
tu
d
ies
'
s
ec
o
n
d
d
ataset.
Yelp
.
co
m
i
s
a
u
s
er
-
g
e
n
er
ated
web
s
ite
th
at
r
ev
iews
n
ea
r
b
y
estab
lis
h
m
en
ts
an
d
r
esem
b
les
s
o
cial
n
etwo
r
k
in
g
s
ites
in
t
h
at
it
en
ab
les
u
s
er
co
m
m
u
n
icatio
n
.
E
lm
u
r
n
g
i
an
d
Gh
er
b
i
[
6
]
h
av
e
tak
en
m
o
v
ie
r
e
v
iews
as
a
d
ataset.
T
ex
t
class
if
icatio
n
an
d
s
en
tim
en
t
an
aly
s
is
(
SA
)
m
eth
o
d
s
ar
e
u
s
ed
o
n
th
e
r
ea
l
d
ataset
o
f
m
o
v
ie
r
ev
iews.
T
h
ey
h
a
v
e
ap
p
lied
two
d
if
f
er
en
t
ap
p
r
o
ac
h
es,
with
an
d
with
o
u
t
s
to
p
wo
r
d
s
in
th
at
th
ey
h
av
e
co
m
p
ar
e
d
Naiv
e
B
ay
es
(
NB
)
,
SVM,
K
-
n
ea
r
est
n
eig
h
b
o
r
s
(
KNN
-
I
B
K)
,
a
n
d
d
ec
is
io
n
tr
ee
(D
T
-
J
4
8
)
f
o
r
s
en
tim
en
t
class
if
icatio
n
o
f
r
ev
ie
ws.
T
h
e
m
ea
s
u
r
ed
o
u
tco
m
es
s
h
o
w
th
at
th
e
SVM
alg
o
r
ith
m
o
u
t
p
er
f
o
r
m
s
r
iv
al
alg
o
r
ith
m
s
f
o
r
b
o
th
tech
n
iq
u
e
s
an
d
ac
h
iev
es
th
e
h
ig
h
est
lev
el
o
f
ac
cu
r
ac
y
i
n
t
ex
t
class
if
icatio
n
an
d
th
e
d
ete
ctio
n
o
f
f
a
k
e
r
ev
iews.
Ab
r
i
et
a
l.
[
7
]
th
o
r
o
u
g
h
ly
ex
am
in
ed
lin
g
u
is
tic
tr
aits
th
at
d
if
f
er
e
n
tiate
f
r
au
d
u
len
t
f
r
o
m
r
eliab
le
in
ter
n
et
r
ev
iews.
Af
te
r
ex
am
in
i
n
g
f
if
teen
ch
ar
ac
ter
is
tics
,
th
ey
d
is
co
v
e
r
ed
th
at
f
ak
e
r
ev
iews
f
r
eq
u
en
tly
em
p
lo
y
m
o
r
e
p
au
s
es,
len
g
th
ier
p
h
r
ases
,
an
d
d
u
p
licate
ter
m
in
o
lo
g
y
.
Usi
n
g
th
ese
tr
aits
with
v
ar
io
u
s
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
,
th
ey
ac
cu
r
ately
d
is
tin
g
u
is
h
ed
f
ak
e
f
r
o
m
r
ea
l
r
ev
iews.
Similar
ly
,
W
an
g
et
a
l.
[
8
]
em
p
l
o
y
ed
s
u
p
er
v
is
ed
m
ac
h
in
e
lear
n
in
g
t
o
p
r
o
p
o
s
e
two
f
ea
tu
r
e
ty
p
es
s
u
b
ject
f
ea
tu
r
es
an
d
r
ea
d
ab
ilit
y
c
h
ar
ac
ter
is
tics
f
o
r
class
if
y
in
g
Yelp
r
ev
iews.
T
h
eir
r
esu
lts
s
h
o
wed
th
ese
f
ea
tu
r
es
o
u
tp
er
f
o
r
m
e
d
n
-
g
r
am
s
in
id
en
tify
in
g
f
r
au
d
u
len
t
r
ev
iews,
an
d
in
co
r
p
o
r
atin
g
r
ev
iewe
r
s
'
b
eh
av
io
r
al
tr
aits
s
ig
n
if
ican
tly
im
p
r
o
v
ed
class
if
icat
io
n
ac
cu
r
ac
y
f
o
r
ac
tu
al
Yelp
o
p
in
io
n
s
p
am
d
ata.
B
ir
im
et
a
l.
[
9
]
s
tu
d
ied
wh
i
ch
f
ea
tu
r
e
co
m
b
in
atio
n
em
o
t
io
n
s
co
r
es,
to
p
ic
d
is
tr
ib
u
tio
n
s
,
clu
s
ter
d
is
tr
ib
u
tio
n
s
,
an
d
a
b
ag
o
f
wo
r
d
s
m
o
s
t
ef
f
ec
tiv
ely
d
etec
t
f
r
au
d
u
len
t
r
e
v
iews.
T
h
e
r
esear
ch
ad
d
r
ess
es
th
e
s
ig
n
if
ican
t
is
s
u
e
o
f
f
ak
e
r
ev
ie
ws
in
f
lu
en
cin
g
cu
s
to
m
er
p
u
r
c
h
ase
d
ec
is
io
n
s
,
u
s
in
g
Am
az
o
n
.
co
m
r
ev
iew
d
ata
an
d
v
ar
i
o
u
s
s
en
tim
en
t a
n
aly
s
is
f
ea
tu
r
es.
Fin
d
in
g
s
s
h
o
w
th
at
b
eh
av
io
r
-
r
elate
d
f
ea
tu
r
es,
p
ar
tic
u
lar
ly
th
e
v
e
r
if
ied
p
u
r
ch
ase
f
ea
tu
r
e,
s
ig
n
if
ican
tl
y
im
p
ac
t
th
e
class
if
icat
io
n
o
f
f
r
au
d
u
le
n
t
r
ev
iews
wh
en
co
m
b
in
ed
with
tex
t
-
r
elate
d
f
ea
tu
r
es.
Als
u
b
ar
i
et
a
l.
[
1
0
]
ex
am
in
e
d
a
Yelp
d
ataset,
ap
p
ly
in
g
m
eth
o
d
s
lik
e
s
en
ti
m
en
t
an
aly
s
is
,
p
ar
t
-
of
-
s
p
ee
ch
(
POS
)
tag
g
i
n
g
,
lin
g
u
is
tic
in
q
u
ir
y
an
d
wo
r
d
c
o
u
n
t
(
L
I
W
C
)
,
an
d
s
u
b
jectiv
ity
f
o
r
f
ea
tu
r
e
ex
tr
ac
tio
n
.
T
h
ey
e
x
tr
ac
ted
v
ar
io
u
s
attr
i
b
u
tes,
in
clu
d
in
g
co
u
n
ts
o
f
ad
ject
iv
es,
v
er
b
s
,
n
o
u
n
s
,
ad
v
er
b
s
,
p
o
lar
ity
,
o
b
jectiv
ity
,
an
d
s
u
b
jectiv
ity
.
Usi
n
g
in
f
o
r
m
atio
n
g
ain
(
I
G)
,
th
ey
s
elec
ted
th
e
m
o
s
t
v
alu
ab
le
f
ea
tu
r
es.
DT
s
,
r
an
d
o
m
f
o
r
est
,
an
d
ad
ap
tiv
e
b
o
o
s
tin
g
wer
e
e
m
p
lo
y
ed
to
class
if
y
r
ev
iews
as
f
a
ls
e
o
r
r
eliab
le,
ac
h
iev
in
g
ac
cu
r
ac
ies
o
f
9
6
%,
9
4
%,
an
d
9
7
%,
r
esp
ec
tiv
ely
.
Gu
tier
r
ez
-
E
s
p
in
o
za
et
a
l.
[
1
1
]
s
tu
d
ied
en
s
em
b
le
lear
n
in
g
ap
p
r
o
ac
h
es
f
o
r
d
etec
tin
g
f
a
ls
e
o
n
lin
e
co
n
ten
t,
s
p
ec
if
ically
f
a
k
e
r
estau
r
an
t
r
ev
iews.
T
h
eir
r
esu
lts
s
h
o
wed
th
at
th
ese
m
eth
o
d
s
o
u
t
p
er
f
o
r
m
tr
ad
itio
n
al
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
.
Stan
d
-
alo
n
e
m
u
ltil
ay
er
p
er
c
ep
tr
o
n
(
MLP
)
class
if
ier
s
ac
h
iev
ed
u
p
to
6
8
.
2
%
ac
cu
r
ac
y
,
wh
ile
an
Ad
aBo
o
s
t
en
s
em
b
le
o
f
ML
Ps
r
ea
ch
e
d
7
7
.
3
%.
L
iu
et
a
l.
[
1
2
]
,
R
o
u
t
et
a
l.
[
1
3
]
,
an
d
Yo
u
et
a
l.
[
1
4
]
,
u
s
ed
o
u
tlier
d
etec
tio
n
tech
n
iq
u
es
to
class
if
y
r
ev
iews
as
s
p
am
o
r
ac
cu
r
ate.
Ou
tlier
d
etec
tio
n
,
a
co
m
m
o
n
d
ata
an
aly
s
is
to
p
ic,
f
o
cu
s
es
o
n
id
en
tify
in
g
an
o
m
alies
in
d
ataset
s
[
1
5
]
an
d
is
ap
p
lied
in
f
au
lt
d
etec
tio
n
,
in
tr
u
s
io
n
d
etec
tio
n
,
an
d
f
r
au
d
d
etec
tio
n
.
C
u
r
r
en
t
o
u
tlier
d
etec
tio
n
m
eth
o
d
s
f
all
in
to
f
o
u
r
ca
teg
o
r
ies:
s
tatis
t
ical
d
is
tr
ib
u
tio
n
-
b
ased
,
d
is
tan
ce
-
b
ased
,
d
e
n
s
ity
-
b
ased
,
an
d
s
u
b
s
p
ac
e
lear
n
in
g
-
b
ased
[
1
6
]
.
Ad
d
itio
n
ally
,
s
tu
d
ies
u
s
in
g
p
r
e
-
tr
ain
ed
lan
g
u
ag
e
m
o
d
els
lik
e
b
id
ir
ec
tio
n
al
en
co
d
er
r
ep
r
esen
tatio
n
s
f
r
o
m
t
r
an
s
f
o
r
m
e
r
s
(
B
E
R
T
)
an
d
XL
Net
with
laten
t
d
ir
ich
let
allo
ca
tio
n
(
L
D
A
)
to
p
ic
d
is
tr
ib
u
tio
n
s
f
o
u
n
d
th
em
ef
f
ec
tiv
e
f
o
r
id
en
tify
in
g
f
ak
e
C
OVI
D
-
1
9
n
ews
[
1
7
]
.
Mu
h
am
m
a
d
an
d
Ah
m
ed
[
1
8
]
d
em
o
n
s
tr
ated
g
o
o
d
ac
c
u
r
ac
y
in
id
en
tify
in
g
f
alse
r
ev
iews
u
s
in
g
s
p
ar
s
e
m
atr
ices
o
f
ter
m
f
r
eq
u
en
c
y
-
in
v
e
r
s
e
d
o
cu
m
en
t
f
r
eq
u
en
c
y
(
TF
-
I
DF
)
,
co
u
n
t
v
ec
to
r
izer
(
CV
)
,
an
d
n
-
g
r
am
f
ea
tu
r
es
in
a
p
r
i
n
cip
al
co
m
p
o
n
en
t
an
aly
s
is
(
PC
A
)
f
ea
tu
r
e
s
et.
Als
u
b
ar
i
et
a
l.
[
1
9
]
u
tili
ze
d
a
co
n
v
o
l
u
tio
n
al
n
eu
r
al
n
etwo
r
k
-
lo
n
g
s
h
o
r
t
-
ter
m
m
em
o
r
y
(
C
N
N
-
L
STM
)
mode
l
o
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
8
,
No
.
1
,
Ap
r
il
20
2
5
:
4
7
8
-
4
8
5
480
a
m
u
lti
-
d
o
m
ain
d
ataset,
ac
h
i
ev
in
g
i
n
-
d
o
m
ain
ac
cu
r
ac
y
r
at
es
o
f
8
7
%,
8
6
%,
8
5
%,
a
n
d
7
7
%,
with
a
cr
o
s
s
-
d
o
m
ain
ac
c
u
r
ac
y
o
f
8
9
%,
s
u
r
p
ass
in
g
p
r
ev
io
u
s
s
tu
d
ies.
2.
P
RO
P
O
SE
D
M
E
T
H
O
D
T
h
e
m
ain
g
o
al
o
f
th
is
s
tu
d
y
is
to
cr
ea
te
a
m
eth
o
d
f
o
r
d
etec
tin
g
s
p
am
o
n
lin
e
r
e
v
iews
th
at
is
b
o
th
ef
f
ec
tiv
e
an
d
e
f
f
icien
t.
Sev
e
r
al
s
tu
d
ies
h
av
e
b
ee
n
ca
r
r
ie
d
o
u
t
to
o
f
f
er
d
etec
tin
g
m
eth
o
d
s
th
at
will
s
o
lv
e
th
e
ab
o
v
e
-
m
e
n
tio
n
ed
d
esira
b
le
f
e
atu
r
es.
Fu
r
th
er
m
o
r
e
,
ex
is
tin
g
s
o
lu
tio
n
s
h
av
e
a
h
ig
h
f
alse
-
p
o
s
itiv
e
r
ate,
a
lo
n
g
tim
e
to
id
en
tify
s
p
a
m
r
ev
ie
ws,
a
lar
g
e
g
ap
b
etwe
en
th
e
in
s
tallatio
n
o
f
s
p
am
d
etec
ti
o
n
m
eth
o
d
s
an
d
th
e
g
u
ar
an
tee
o
f
a
p
o
s
itiv
e
r
esu
l
t,
an
d
s
o
o
n
.
T
h
e
p
r
o
p
o
s
ed
s
y
s
tem
o
v
er
co
m
es
th
ese
d
if
f
i
cu
lties
an
d
it
f
ir
s
t
p
r
o
p
o
s
es
en
h
an
ce
d
alg
o
r
ith
m
s
to
im
p
r
o
v
e
th
e
wo
r
k
in
g
o
f
ea
ch
s
tep
,
f
r
o
m
wh
ic
h
th
e
b
est
wo
r
k
in
g
m
eth
o
d
is
co
m
b
in
ed
to
f
o
r
m
th
e
e
n
h
an
ce
d
en
s
em
b
le
id
en
tific
atio
n
s
y
s
tem
.
Featu
r
es o
f
th
e
p
r
o
p
o
s
ed
s
y
s
tem
ar
e
−
T
o
p
er
f
o
r
m
f
ea
t
u
r
e
en
g
in
ee
r
i
n
g
co
n
s
tr
u
ct
a
f
ea
tu
r
e
v
ec
t
o
r
h
av
in
g
o
n
ly
o
p
tim
al
f
ea
tu
r
es
ex
tr
ac
ted
f
r
o
m
m
u
ltip
le
en
titi
es,
wh
ich
h
elp
s
to
im
p
r
o
v
e
th
e
p
e
r
f
o
r
m
an
ce
o
f
h
am
/s
p
am
d
etec
tio
n
s
y
s
tem
s
.
−
T
o
d
esig
n
en
h
a
n
ce
d
class
if
icatio
n
alg
o
r
ith
m
s
to
im
p
r
o
v
e
th
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
o
n
lin
e
r
ev
iew
s
p
am
d
etec
tio
n
s
y
s
tem
.
−
T
o
d
esig
n
a
n
en
h
a
n
ce
d
e
n
s
em
b
le
class
if
ier
s
y
s
tem
to
in
cr
ea
s
e
th
e
ac
cu
r
ac
y
o
f
s
p
am
d
etec
ti
o
n
.
I
n
th
is
r
esear
ch
m
eth
o
d
,
t
h
e
o
p
tim
izatio
n
o
f
th
e
SVM
class
if
ier
is
d
o
n
e
i
n
two
m
an
n
er
s
.
T
h
e
f
ir
s
t
s
tep
is
to
r
em
o
v
e
ir
r
elev
a
n
t
s
u
p
p
o
r
t
v
ec
to
r
s
with
n
o
r
elev
an
cy
d
u
r
in
g
class
if
icatio
n
.
T
h
is
less
en
s
th
e
q
u
an
tity
o
f
co
m
p
u
tatio
n
s
an
d
th
u
s
s
o
lv
es
th
e
h
ig
h
tr
ain
in
g
tim
e
r
e
q
u
i
r
ed
.
T
h
e
s
ec
o
n
d
is
to
r
ep
lace
th
e
co
n
v
e
n
tio
n
ally
u
s
ed
E
u
clid
ea
n
d
is
tan
ce
with
th
e
Ma
h
alan
o
b
is
d
is
tan
ce
m
ea
s
u
r
e.
3.
M
E
T
H
O
D
AND
F
I
N
DING
S
T
h
e
s
u
g
g
ested
s
p
am
o
n
lin
e
r
e
v
iew
d
etec
tio
n
is
p
ar
t
o
f
th
e
o
n
lin
e
r
ev
iew
s
ec
u
r
ity
co
m
p
o
n
en
t
s
in
ce
it
p
r
o
tects
u
s
er
s
(
o
r
cu
s
to
m
e
r
s
)
ag
ain
s
t
f
alse
d
etails.
T
h
e
s
u
g
g
ested
s
y
s
tem
em
p
lo
y
s
f
ea
tu
r
e
en
g
in
ee
r
in
g
,
class
if
icatio
n
,
an
d
cl
u
s
ter
in
g
m
eth
o
d
s
to
en
h
a
n
ce
o
n
lin
e
s
p
am
d
etec
tio
n
.
T
h
e
p
r
im
ar
y
g
o
al
o
f
th
e
s
p
am
r
ev
iew
d
etec
tio
n
s
y
s
tem
(
S
R
D)
is
to
id
en
tif
y
all
s
p
am
r
e
v
iews
u
s
in
g
m
ac
h
i
n
e
lear
n
i
n
g
tech
n
iq
u
es.
T
h
e
alg
o
r
ith
m
s
tar
ts
b
y
m
ap
p
in
g
a
ll
tr
ain
in
g
f
ea
tu
r
es
in
to
t
h
e
SVM
v
ec
to
r
s
p
ac
e
an
d
c
o
m
p
u
t
in
g
th
e
m
ar
g
in
s
f
o
r
ea
ch
ca
teg
o
r
y
.
T
h
e
s
m
allest
M
m
ar
g
in
s
ar
e
s
elec
ted
as
r
elev
an
t
s
u
p
p
o
r
t
v
ec
t
o
r
s
(
SVs
)
,
with
th
e
r
est
d
is
ca
r
d
ed
as
n
o
is
e.
T
h
e
id
en
tifie
d
SVs
ar
e
th
en
o
u
tlin
ed
in
th
e
p
r
o
to
ty
p
al
v
ec
to
r
s
p
ac
e
an
d
r
e
m
ap
p
ed
to
th
e
o
r
ig
i
n
al
s
p
ac
e.
Ma
h
alan
o
b
is
d
is
tan
ce
i
s
u
s
ed
to
ca
lc
u
late
th
e
av
er
ag
e
d
is
tan
ce
b
etwe
en
n
ew
f
ea
tu
r
es
an
d
ea
c
h
SV
s
et.
T
h
e
ca
teg
o
r
y
with
th
e
clo
s
est
SVs
is
a
s
s
ig
n
ed
to
th
e
n
ew
f
ea
tu
r
e.
T
h
e
r
ev
iew
d
atasets
o
f
s
m
ar
tp
h
o
n
es
co
llected
f
r
o
m
Am
az
o
n
an
d
Yelp
ar
e
tak
en
in
to
co
n
s
id
er
atio
n
.
T
h
e
d
atasets
r
ec
eiv
ed
wer
e
p
r
e
-
p
r
o
ce
s
s
ed
to
m
ee
t
th
e
r
eq
u
ir
em
en
ts
o
f
th
is
s
tu
d
y
.
T
h
e
f
in
al
p
r
e
-
p
r
o
ce
s
s
ed
d
ataset
th
er
ef
o
r
e
in
clu
d
ed
d
etails
s
to
r
ed
in
a
way
th
at
th
e
alg
o
r
ith
m
s
co
u
ld
s
im
p
ly
ac
ce
s
s
.
T
h
is
f
in
al
d
ataset
in
clu
d
es 6
7
,
9
8
6
r
ev
iews f
r
o
m
Am
az
o
n
an
d
1
2
8
7
6
f
r
o
m
Yelp
an
d
is
g
iv
en
as in
p
u
t to
th
e
s
y
s
tem
d
ev
elo
p
ed
.
T
h
e
SR
D
is
a
p
r
o
g
r
am
m
e
th
at
u
s
es
a
m
a
ch
in
e
lear
n
in
g
alg
o
r
ith
m
,
A,
to
d
eter
m
i
n
e
if
a
r
e
v
iew
R
,
is
s
p
am
o
r
n
o
t.
(
,
)
=
{
ℎ
ℎ
I
n
th
e
ab
o
v
e
eq
u
atio
n
,
R
is
th
e
o
n
lin
e
r
ev
iew
th
at
h
as
to
b
e
ca
teg
o
r
ized
as
s
p
am
o
r
h
am
,
F
is
a
f
ea
tu
r
e
v
ec
t
o
r
th
at
r
ep
r
esen
ts
th
e
v
ar
io
u
s
c
h
ar
ac
ter
is
tics
o
f
R
.
T
h
e
d
etec
ti
o
n
alg
o
r
ith
m
,
A,
u
s
es
a
lear
n
i
n
g
alg
o
r
ith
m
(
Q
)
to
tr
ain
th
e
m
ac
h
in
e
lear
n
i
n
g
a
lg
o
r
ith
m
with
th
e
d
ataset
th
at
h
as
f
ea
tu
r
es
(
F)
p
r
e
-
co
lle
cted
f
r
o
m
r
e
v
iews
=
(
,
)
,
wh
er
e
=
{
1
,
…
,
}
an
d
C
i
s
th
e
s
e
t o
f
tar
g
et
lab
els,
wh
ich
is
{sp
am
,
h
am
}
in
th
is
r
esear
ch
.
T
h
e
d
etec
tio
n
alg
o
r
ith
m
,
A,
h
an
d
les
o
n
e
r
ev
iew
at
a
tim
e
an
d
class
if
ies
th
em
as
h
am
o
r
s
p
am
,
d
ep
e
n
d
in
g
o
n
th
e
r
esu
lt (
q
)
o
b
tain
ed
f
r
o
m
(
Q)
.
T
wo
ac
tio
n
s
ar
e
ta
k
en
f
r
o
m
th
e
r
esu
lt.
{
=
=
Fig
u
r
e
1
p
r
o
v
id
es a
d
etailed
m
eth
o
d
o
lo
g
y
f
o
r
th
e
s
p
am
r
ev
ie
w
d
etec
tio
n
m
o
d
el,
with
d
if
f
e
r
en
t p
h
ases
.
Ph
ase
I
d
escr
ib
es f
ea
tu
r
e
en
g
in
ee
r
in
g
,
an
d
p
h
ase
I
I
in
clu
d
es d
esig
n
in
g
an
en
h
an
ce
d
en
s
em
b
le
class
if
icatio
n
s
y
s
tem
.
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:
2
5
0
2
-
4
7
52
F
a
ke
r
ev
iew
d
etec
tio
n
u
s
in
g
e
n
h
a
n
ce
d
en
s
emb
le
s
u
p
p
o
r
t v
ec
to
r
…
(
S
ee
n
ia
J
o
s
ep
h
)
481
Fig
u
r
e
1
.
Dif
f
e
r
en
t p
h
ases
o
f
m
eth
o
d
s
3
.
1
.
P
ha
s
e
I
:
f
e
a
t
ure
eng
ineering
3
.
1
.
1
.
F
ea
t
ure
e
ng
ineering
T
h
e
ap
p
r
o
ac
h
es
p
u
t
f
o
r
th
i
n
P
h
ase
I
in
v
o
lv
e
f
ea
tu
r
e
e
n
g
in
ee
r
in
g
t
o
cr
ea
te
a
f
e
atu
r
e
v
ec
to
r
with
o
n
ly
th
e
b
est
f
ea
tu
r
es
an
d
t
o
im
p
r
o
v
e
r
ev
iew
s
p
am
d
etec
tio
n
.
F
ea
tu
r
e
en
g
in
ee
r
in
g
is
d
escr
ib
e
d
as
th
e
p
r
o
ce
s
s
o
f
cr
ea
tin
g
o
r
e
x
tr
ac
tin
g
f
ea
tu
r
e
s
f
r
o
m
d
ata
s
o
u
r
ce
s
[
2
0
]
.
T
h
is
is
ac
co
m
p
lis
h
ed
in
Ph
ase
I
th
r
o
u
g
h
th
e
two
p
r
o
ce
s
s
es o
f
f
ea
tu
r
e
e
x
tr
ac
tio
n
an
d
f
ea
tu
r
e
s
elec
tio
n
.
3
.
1
.
2.
F
ea
t
ure
e
x
t
ra
ct
i
o
n
T
h
e
f
ea
tu
r
e
ex
tr
ac
tio
n
tech
n
i
q
u
e,
wh
ich
co
n
d
en
s
es
r
aw
d
ata,
is
th
e
m
o
s
t
s
ig
n
if
ican
t
co
m
p
o
n
e
n
t
o
f
th
e
s
y
s
tem
f
o
r
d
etec
tin
g
r
ev
ie
w
s
p
am
.
T
h
is
r
ed
u
ce
d
d
ata
is
r
ef
er
r
e
d
to
as
f
ea
tu
r
e
v
ec
to
r
s
o
r
f
ea
tu
r
e
s
p
ac
es.
T
h
e
th
r
ee
asp
ec
ts
to
ev
alu
ate
in
an
o
n
lin
e
r
ev
iew
ar
e
co
n
ten
t,
r
ev
iewe
r
,
an
d
p
r
o
d
u
ct
a
r
e
also
th
e
to
p
ic
o
f
s
ev
er
al
ch
ar
ac
ter
is
tics
o
b
tain
ed
f
o
r
th
is
s
tu
d
y
.
I
n
th
e
f
ir
s
t
s
tag
e
o
f
Ph
ase
I
,
a
to
tal
o
f
5
3
s
ets
o
f
f
ea
tu
r
es
ar
e
ex
tr
ac
ted
.
Ou
t
o
f
5
3
f
ea
tu
r
es,
th
er
e
ar
e
3
8
r
e
v
iew
-
ce
n
tr
ic
f
e
atu
r
es,1
3
r
e
v
iewe
r
-
ce
n
t
r
ic
f
ea
tu
r
es
an
d
2
p
r
o
d
u
ct
-
ce
n
tr
ic
f
ea
tu
r
es d
escr
ib
e
d
in
T
ab
le
1
.
T
ab
le
1
.
Featu
r
es
e
x
tr
ac
te
d
R
e
v
i
e
w
c
e
n
t
r
i
c
f
e
a
t
u
r
e
s
(
3
8
)
Te
x
t
u
a
l
f
e
a
t
u
r
e
s (
9
)
,
m
e
t
a
d
a
t
a
(
8
)
,
c
o
n
t
e
n
t
s
i
mi
l
a
r
i
t
y
u
si
n
g
b
a
g
o
f
w
o
r
d
s
(
5
)
,
P
O
S
t
a
g
s (
9
)
,
n
-
g
r
a
m
s (4
)
,
r
a
t
i
n
g
(
1
)
,
se
n
t
i
me
n
t
(
1
)
,
b
u
r
s
t
p
a
t
t
e
r
n
s (
1
)
R
e
v
i
e
w
e
r
c
e
n
t
r
i
c
f
e
a
t
u
r
e
s
(
1
3
)
R
e
v
i
e
w
e
r
a
c
t
i
v
i
t
i
e
s
,
ma
x
i
mu
m
n
u
m
b
e
r
o
f
r
e
v
i
e
w
s,
p
e
r
c
e
n
t
a
g
e
o
f
p
o
s
i
t
i
v
e
r
e
v
i
e
w
s,
r
e
v
i
e
w
l
e
n
g
t
h
,
r
e
v
i
e
w
e
r
d
e
v
i
a
t
i
o
n
,
b
u
r
st
r
e
v
i
e
w
r
a
t
i
o
,
r
a
t
i
o
o
f
v
e
r
i
f
i
e
d
p
u
r
c
h
a
ses,
r
e
v
i
e
w
e
r
b
u
r
s
t
i
n
e
s
s,
e
x
t
r
e
me
r
a
t
i
n
g
s,
r
e
v
i
e
w
e
r
a
v
e
r
a
g
e
p
r
o
l
i
f
e
r
a
t
i
o
n
,
r
e
v
i
e
w
e
r
sp
a
mi
c
i
t
y
,
%
o
f
p
o
si
t
i
v
e
r
e
v
i
e
w
s,
%
o
f
n
e
g
a
t
i
v
e
r
e
v
i
e
w
s
P
r
o
d
u
c
t
c
e
n
t
r
i
c
f
e
a
t
u
r
e
s
(
2
)
R
a
n
k
i
n
s
a
l
e
,
a
v
e
r
a
g
e
r
a
t
i
n
g
3
.
1
.
3
.
F
ea
t
ure
s
elec
t
io
n
T
h
e
f
ea
tu
r
es
r
et
r
iev
ed
i
n
s
tep
1
m
ay
co
n
tain
ir
r
elev
an
t
,
n
o
i
s
y
,
o
r
d
u
p
licated
p
r
o
p
er
ties
.
Ov
er
f
itti
n
g
an
d
ex
tr
a
ca
lcu
latio
n
s
ar
e
two
class
if
icat
io
n
d
if
f
icu
lties
th
at
r
esu
lt
in
in
cr
ea
s
ed
te
m
p
o
r
al
co
m
p
le
x
ity
.
A
f
ea
tu
r
e
s
elec
tio
n
alg
o
r
ith
m
o
p
er
ates
o
n
th
e
p
r
em
is
e
th
at
n
o
t
all
ex
tr
ac
ted
f
ea
tu
r
es
ar
e
c
r
u
cial
f
o
r
d
etec
tin
g
r
ev
iew
s
p
am
.
I
d
en
tify
i
n
g
an
d
r
em
o
v
i
n
g
th
ese
ir
r
elev
an
t
f
e
atu
r
es
ca
n
en
h
an
ce
t
h
e
p
er
f
o
r
m
an
ce
o
f
th
e
s
p
am
d
etec
tio
n
s
y
s
tem
[
2
1
]
.
I
n
th
e
f
ir
s
t
s
tag
e,
an
en
h
an
ce
d
m
ax
i
m
u
m
r
elev
an
ce
m
in
im
u
m
r
ed
u
n
d
a
n
c
y
(
MRMR
)
f
ea
tu
r
e
s
elec
tio
n
ap
p
r
o
ac
h
is
em
p
lo
y
ed
to
lo
ca
te
d
is
tin
g
u
is
h
i
n
g
an
d
s
ig
n
if
ican
t c
h
a
r
ac
ter
is
tics
.
Acc
o
r
d
in
g
to
th
e
ex
p
er
im
e
n
t
al
r
esu
lts
,
u
s
in
g
an
y
f
ea
tu
r
e
s
elec
tio
n
tech
n
iq
u
e
m
ig
h
t
in
cr
ea
s
e
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
s
p
am
r
ev
iew
d
etec
tio
n
(
SR
D
)
s
y
s
te
m
.
Usi
n
g
th
e
s
u
g
g
ested
f
ea
tu
r
e
s
elec
tio
n
u
s
in
g
th
e
m
o
d
if
ied
g
en
etic
alg
o
r
ith
m
(
FS
_
MG
A
)
m
eth
o
d
,
th
e
SR
D
s
y
s
tem
ac
h
iev
ed
a
p
r
ec
is
io
n
o
f
7
9
.
2
5
%,
7
6
.
9
4
%
,
an
d
7
4
.
4
3
%
(
Am
az
o
n
d
ataset)
an
d
7
8
.
0
4
%,
7
5
.
0
4
%,
an
d
7
2
.
5
1
%
(
Yelp
d
ataset)
.
Similar
ly
,
th
e
SR
D
s
y
s
tem
with
th
e
MG
A
f
ea
t
u
r
e
s
elec
t
io
n
m
eth
o
d
h
ad
a
h
ig
h
r
ec
all
v
alu
e
o
f
8
0
.
2
5
%
(
SVM)
,
7
7
.
8
8
%
(
KNN)
,
an
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
8
,
No
.
1
,
Ap
r
il
20
2
5
:
4
7
8
-
4
8
5
482
7
5
.
7
6
%
(
NB
)
f
o
r
th
e
Am
az
o
n
d
ataset
an
d
7
8
.
5
3
%
(
SVM)
,
7
6
.
1
2
%
(
KNN)
,
an
d
7
3
.
8
7
%
(
NB
)
f
o
r
th
e
Yelp
d
ataset
.
C
o
d
es u
s
ed
in
th
e
SR
D
s
y
s
tem
f
o
r
d
if
f
e
r
en
t f
ea
tu
r
e
s
elec
tio
n
s
ar
e
d
escr
ib
ed
in
T
a
b
le
2
.
T
ab
le
2
.
C
o
d
in
g
s
ch
em
e
u
s
ed
–
f
ea
tu
r
e
s
elec
tio
n
C
o
d
e
A
l
g
o
r
i
t
h
m
N
F
S
N
o
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
FS_
M
R
M
R
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
u
s
i
n
g
M
R
M
R
a
l
g
o
r
i
t
h
m
F
S
_
M
I
M
G
F
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
u
s
i
n
g
M
R
M
R
_
I
G
a
n
d
M
R
M
R
_
M
I
a
l
g
o
r
i
t
h
m
F
S
_
A
C
O
F
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
u
s
i
n
g
A
C
O
a
l
g
o
r
i
t
h
m
F
S
_
A
C
O
+
G
A
F
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
u
s
i
n
g
A
C
O
+
G
A
a
l
g
o
r
i
t
h
m
F
S
_
M
G
A
F
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
u
s
i
n
g
F
S
_
M
I
M
G
a
n
d
A
C
O
+
G
A
a
l
g
o
r
i
t
h
m
Per
f
o
r
m
an
ce
m
atr
ices
f
o
r
d
if
f
er
en
t
f
ea
tu
r
e
s
elec
tio
n
s
a
r
e
g
iv
e
n
b
elo
w.
Fig
u
r
e
2
s
h
o
ws
th
e
p
er
ce
n
tag
e
o
f
p
r
ec
is
io
n
wh
e
n
im
p
lem
en
ted
in
SVM,
K
NN
an
d
NB
with
Am
az
o
n
an
d
Yelp
d
ata
s
ets.
Similar
ly
,
Fig
u
r
e
3
s
h
o
ws p
e
r
f
o
r
m
an
ce
o
f
R
ec
all.
T
h
e
MG
A
m
eth
o
d
g
en
er
ate
d
an
F
-
m
ea
s
u
r
e
o
f
7
9
.
7
5
%
(
SV
M)
,
7
7
.
4
1
%
(
KNN)
,
an
d
7
5
.
0
9
%
(
NB
)
wh
en
u
tili
zin
g
th
e
Am
az
o
n
d
a
taset
s
h
o
wn
in
F
ig
u
r
e
4
.
W
h
en
ev
alu
ated
u
s
in
g
th
e
Yelp
d
ataset,
th
e
F
-
m
ea
s
u
r
e
o
f
th
e
MG
A
m
eth
o
d
was
7
8
.
2
9
%,
7
5
.
5
8
%,
an
d
7
3
.
1
8
%,
r
es
p
ec
tiv
ely
,
wh
e
n
test
ed
with
S
VM
,
KNN,
an
d
NB
class
if
ier
s
.
W
h
en
ev
alu
ated
u
s
in
g
SVM,
KNN,
an
d
NB
class
if
ier
s
,
th
e
p
r
o
p
o
s
ed
M
GA
m
eth
o
d
h
a
d
a
m
ax
im
u
m
ac
cu
r
ac
y
o
f
8
0
.
0
8
%,
7
8
.
4
0
%,
an
d
7
6
.
9
2
%
o
n
t
h
e
Am
az
o
n
an
d
7
8
.
0
7
%,
7
6
.
7
6
%,
an
d
7
4
.
9
4
%
o
n
th
e
Yelp
d
ata
s
ets as sh
o
wn
in
Fig
u
r
e
5
.
Fig
u
r
e
2
.
Pre
cisi
o
n
Fig
u
r
e
3
.
R
ec
all
Fig
u
r
e
4
. F
-
m
ea
s
u
r
e
Fig
u
r
e
5
.
Acc
u
r
ac
y
3
.
2
.
P
H
ASE
2
:
des
ig
n o
f
enha
nced
ens
em
ble c
la
s
s
if
ica
t
io
n sy
s
t
em
T
h
e
p
r
o
p
o
s
ed
s
y
s
tem
is
d
esig
n
ed
in
two
s
tep
s
.
Step
1
:
e
n
h
a
n
ce
a
class
if
icatio
n
alg
o
r
ith
m
an
d
s
tep
2
:
d
esig
n
en
s
em
b
le
s
y
s
tem
s
u
s
in
g
th
e
class
if
ier
en
h
an
ce
d
in
s
tep
1
.
SVM
was
s
elec
ted
as
th
e
class
if
ier
b
ec
au
s
e
it
is
th
e
m
o
s
t
wid
ely
u
s
ed
m
e
th
o
d
f
o
r
class
if
icatio
n
an
d
p
r
e
d
ictio
n
an
d
also
h
as
an
ex
ce
ll
en
t
tr
ac
k
r
ec
o
r
d
o
f
s
u
cc
ess
in
ac
h
iev
in
g
h
ig
h
p
e
r
f
o
r
m
an
ce
[
2
2
]
-
[
2
4
]
wh
en
co
m
p
ar
ed
to
v
a
r
io
u
s
o
th
er
class
if
ier
s
.
I
n
th
is
r
esear
ch
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:
2
5
0
2
-
4
7
52
F
a
ke
r
ev
iew
d
etec
tio
n
u
s
in
g
e
n
h
a
n
ce
d
en
s
emb
le
s
u
p
p
o
r
t v
ec
to
r
…
(
S
ee
n
ia
J
o
s
ep
h
)
483
s
tu
d
y
,
th
e
SVM
class
if
ier
is
o
p
tim
ized
in
two
d
if
f
er
en
t w
ay
s
.
T
h
e
f
ir
s
t step
is
to
elim
in
ate
s
u
p
p
o
r
t v
ec
t
o
r
s
th
at
ar
e
n
o
t
r
elev
an
t
f
o
r
class
if
icat
io
n
an
d
ar
e
i
r
r
elev
an
t.
T
h
e
s
ec
o
n
d
is
to
u
tili
ze
th
e
Ma
h
ala
n
o
b
is
d
is
tan
ce
m
etr
ic
in
p
lace
o
f
th
e
c
o
m
m
o
n
ly
u
s
ed
E
u
clid
ea
n
d
is
tan
ce
.
T
h
is
en
h
an
ce
d
SVM
is
tak
en
as
th
e
p
r
im
a
r
y
b
ase
class
if
ier
to
d
esig
n
an
e
n
s
em
b
le
class
if
ier
.
Acc
o
r
d
in
g
to
Yild
ır
ım
et
a
l.
[
2
5
]
,
s
ev
er
al
m
eth
o
d
s
f
o
r
cr
ea
tin
g
en
s
em
b
les
h
av
e
b
ee
n
p
r
o
p
o
s
e
d
lik
e
k
n
o
wled
g
e
-
b
ased
m
eth
o
d
s
an
d
r
an
d
o
m
izatio
n
m
eth
o
d
s
.
F
o
llo
win
g
th
e
estab
lis
h
m
en
t
o
f
th
e
en
s
em
b
le
s
y
s
tem
wi
th
b
ase
cla
s
s
if
ier
s
,
th
e
s
u
b
s
eq
u
en
t
p
h
ase
in
v
o
lv
es
em
p
lo
y
in
g
a
tech
n
iq
u
e
t
o
co
m
b
in
e
t
h
e
o
u
tco
m
es
o
f
th
ese
b
ase
class
if
ier
s
.
T
h
is
p
r
o
ce
s
s
em
p
lo
y
s
two
ap
p
r
o
ac
h
es:
in
teg
r
atio
n
(
f
u
s
io
n
)
m
et
h
o
d
s
an
d
s
elec
tio
n
m
eth
o
d
s
[
2
6
]
.
T
h
e
s
ec
o
n
d
m
et
h
o
d
is
u
s
ed
i
n
th
is
r
esear
ch
.
T
h
e
o
p
tim
al
f
ea
tu
r
e
v
ec
to
r
p
r
o
d
u
c
ed
b
y
th
e
MG
A
alg
o
r
ith
m
is
u
s
ed
to
tr
ain
an
d
test
all
th
e
class
if
ier
s
.
C
o
d
in
g
s
ch
em
es u
s
ed
in
en
h
a
n
ce
d
cla
s
s
if
icatio
n
s
y
s
tem
s
f
o
r
d
if
f
er
e
n
t a
lg
o
r
ith
m
s
ar
e
s
h
o
wn
in
T
a
b
le
3
.
T
h
e
ad
d
itio
n
o
f
a
s
p
ee
d
o
p
tim
izatio
n
s
tr
ateg
y
to
th
e
s
in
g
le
SVM
class
if
ier
in
cr
ea
s
ed
its
p
er
f
o
r
m
a
n
ce
with
th
e
Am
az
o
n
d
ataset
b
y
4
.
5
%
(
p
r
ec
is
io
n
an
d
r
ec
all)
,
4
.
4
7
%
(
F
-
m
ea
s
u
r
e
)
,
an
d
3
.
6
6
%
(
ac
cu
r
ac
y
)
.
T
h
e
ef
f
icien
cy
im
p
r
o
v
em
en
t
g
ain
e
d
wh
en
ev
alu
ated
with
th
e
Yelp
d
ataset
was
3
.
7
1
%
(
p
r
ec
is
io
n
)
,
4
.
6
9
%
(
r
ec
all)
,
4
.
2
0
%
(
F
-
m
ea
s
u
r
e)
,
an
d
3
.
3
4
%
(
ac
cu
r
ac
y
)
.
W
h
en
e
v
alu
a
ted
with
th
e
Am
az
o
n
d
ataset,
th
e
o
p
tim
izatio
n
ap
p
r
o
ac
h
u
s
in
g
th
e
e
n
s
em
b
le
s
y
s
tem
im
p
r
o
v
ed
p
er
f
o
r
m
an
ce
b
y
0
.
7
9
%
(
p
r
ec
is
io
n
)
,
0
.
8
6
%
(
r
e
ca
ll),
0
.
8
3
%
(F
-
Me
asu
r
e)
,
a
n
d
0
.
6
1
%
(
ac
c
u
r
ac
y
)
.
T
h
e
ef
f
icien
c
y
im
p
r
o
v
e
m
en
t
f
o
r
t
h
e
Yelp
d
ataset
was
1
.
3
4
%
(
p
r
ec
is
io
n
)
,
1
.
0
4
%
(
r
ec
all)
,
1
.
1
9
%
(
F
-
m
e
asu
r
e)
,
an
d
1
.
1
4
s
ec
o
n
d
s
(
ac
cu
r
ac
y
)
.
Pre
cisi
o
n
ev
alu
atio
n
wh
en
u
s
ed
with
Am
az
o
n
an
d
Yelp
d
atasets
is
g
iv
en
in
Fig
u
r
e
6
.
Per
f
o
r
m
a
n
c
e
o
f
R
ec
all
wh
en
u
s
ed
with
d
if
f
er
en
t a
lg
o
r
ith
m
s
is
g
iv
en
in
Fig
u
r
e
7
.
F
m
ea
s
u
r
e
m
atr
ix
wh
en
u
s
ed
with
Am
a
zo
n
an
d
Yelp
d
atasets
is
g
iv
e
n
in
Fig
u
r
e
8
.
T
h
e
ac
cu
r
ac
y
o
f
m
o
d
els wh
en
u
s
ed
with
d
if
f
er
en
t a
l
g
o
r
ith
m
s
is
p
r
esen
ted
in
Fig
u
r
e
9
.
T
ab
le
3
.
C
o
d
in
g
s
ch
em
e
u
s
ed
–
en
h
an
ce
d
class
if
icatio
n
s
y
s
tem
C
o
d
e
A
l
g
o
r
i
t
h
m
S
S
V
M
C
l
a
s
si
f
i
c
a
t
i
o
n
s
y
s
t
e
m
ES
En
se
mb
l
e
S
V
M
c
l
a
ssi
f
i
c
a
t
i
o
n
s
y
st
e
m
ES_
S
O
En
h
a
n
c
e
d
e
n
sem
b
l
e
S
V
M
c
l
a
ss
i
f
i
c
a
t
i
o
n
sy
s
t
e
m
w
i
t
h
sp
e
e
d
o
p
t
i
mi
z
e
r
s
ES_
S
O
+
ED
En
h
a
n
c
e
d
e
n
sem
b
l
e
S
V
M
c
l
a
ss
i
f
i
c
a
t
i
o
n
sy
s
t
e
m
w
i
t
h
sp
e
e
d
o
p
t
i
mi
z
e
r
s a
n
d
Eu
c
l
i
d
e
a
n
d
i
st
a
n
c
e
me
a
su
r
e
ES_
S
O
+
M
D
En
h
a
n
c
e
d
e
n
sem
b
l
e
S
V
M
c
l
a
ss
i
f
i
c
a
t
i
o
n
sy
s
t
e
m
w
i
t
h
sp
e
e
d
o
p
t
i
mi
z
e
r
s
a
n
d
M
a
h
a
l
a
n
o
b
i
s
d
i
s
t
a
n
c
e
me
a
s
u
r
e
Fig
u
r
e
6
.
Pre
cisi
o
n
Fig
u
r
e
7
.
R
ec
all
Fig
u
r
e
8
. F
-
m
ea
s
u
r
e
Fig
u
r
e
9
.
Acc
u
r
ac
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
8
,
No
.
1
,
Ap
r
il
20
2
5
:
4
7
8
-
4
8
5
484
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
I
n
th
is
r
esear
ch
,
we
h
av
e
d
ev
elo
p
ed
a
f
r
au
d
u
len
t
r
e
v
iew
id
en
tific
atio
n
m
o
d
el
u
s
in
g
a
n
o
n
lin
e
s
p
am
d
etec
tio
n
s
y
s
tem
in
wh
ich
e
n
h
an
ce
d
e
n
s
em
b
le
SVM
is
u
s
ed
,
wh
ich
ca
n
h
elp
cu
s
to
m
er
s
an
d
m
ar
k
etin
g
m
an
ag
er
s
id
en
tify
o
p
in
io
n
s
p
a
m
m
er
s
an
d
th
eir
s
u
s
p
icio
u
s
b
eh
av
io
r
wh
en
m
a
k
in
g
d
ec
is
io
n
s
.
T
h
e
o
p
tim
izatio
n
ac
co
m
p
lis
h
ed
v
ia
th
e
u
s
ag
e
o
f
th
e
d
is
tan
ce
m
ea
s
u
r
e
d
em
o
n
s
tr
ated
th
at
th
e
Ma
h
alan
o
b
is
d
is
tan
ce
m
ea
s
u
r
e
o
u
tp
er
f
o
r
m
ed
th
e
E
u
clid
ea
n
d
is
tan
ce
in
ter
m
s
o
f
class
if
icatio
n
p
er
f
o
r
m
a
n
ce
.
W
h
en
co
m
p
ar
ed
t
o
E
S
an
d
u
tili
zin
g
th
e
Am
az
o
n
d
ataset,
th
e
s
y
s
tem
E
S_
SO+M
D
d
em
o
n
s
tr
ated
an
av
er
a
g
e
e
f
f
icien
c
y
im
p
r
o
v
em
e
n
t
o
f
4
.
6
7
%
(
p
r
ec
is
io
n
)
,
5
.
9
5
%
(
r
e
ca
ll),
5
.
3
0
%
(
F
-
Me
asu
r
e)
,
an
d
5
.
0
9
%
(
ac
c
u
r
ac
y
)
in
ter
m
s
o
f
p
r
ec
is
io
n
,
r
ec
al
l,
F
-
m
ea
s
u
r
e,
an
d
ac
cu
r
ac
y
.
Usi
n
g
th
e
Yelp
d
ataset,
th
e
s
am
e
ap
p
r
o
ac
h
d
em
o
n
s
tr
ated
ef
f
icie
n
cy
g
ain
s
o
f
5
.
3
4
%,
6
.
1
1
%,
5
.
7
2
%,
an
d
3
.
9
7
%
i
n
ter
m
s
o
f
p
r
ec
is
io
n
,
r
ec
all,
F
-
m
ea
s
u
r
e,
an
d
ac
cu
r
ac
y
,
r
esp
ec
tiv
ely
.
Ph
ase
I
ex
p
er
im
en
tal
r
esu
lts
d
em
o
n
s
t
r
ated
th
at
em
p
l
o
y
in
g
an
y
f
ea
tu
r
e
s
elec
tio
n
alg
o
r
ith
m
p
o
s
itiv
ely
im
p
ac
ted
t
h
e
p
er
f
o
r
m
an
ce
o
f
o
n
lin
e
s
p
a
m
r
ev
iew
d
etec
tio
n
,
with
th
e
p
r
o
p
o
s
ed
alg
o
r
ith
m
y
ie
ld
in
g
th
e
g
r
ea
test
im
p
r
o
v
em
e
n
t.
Sp
ec
if
ically
,
th
e
co
m
b
in
atio
n
o
f
MRMR
wit
h
m
u
tu
al
in
f
o
r
m
atio
n
(
MI
)
a
n
d
MRMR
with
I
G,
e
n
h
an
ce
d
b
y
a
n
t
co
lo
n
y
o
p
tim
izatio
n
an
d
g
en
etic
alg
o
r
ith
m
s
,
ac
h
iev
ed
a
9
.
1
1
%
ef
f
icien
cy
g
ain
i
n
ac
cu
r
ac
y
f
o
r
Am
az
o
n
a
n
d
a
9
.
0
8
%
g
ai
n
f
o
r
Yelp
c
o
m
p
ar
e
d
to
class
i
f
ier
s
with
o
u
t
f
ea
tu
r
e
s
elec
tio
n
.
I
n
Ph
ase
I
I
,
it
was
co
n
f
ir
m
e
d
th
at
th
e
o
p
tim
izati
o
n
m
eth
o
d
s
in
te
g
r
ated
in
to
t
h
e
SVM
class
if
ier
wer
e
ef
f
ec
tiv
e.
T
h
e
e
n
h
an
ce
d
en
s
em
b
le
s
y
s
tem
,
u
s
in
g
th
e
i
m
p
r
o
v
e
d
SVM
as
t
h
e
b
ase
class
if
ier
,
ac
h
iev
ed
h
ig
h
ac
cu
r
a
cies
o
f
8
6
.
7
9
%
f
o
r
Am
az
o
n
a
n
d
8
3
.
2
0
%
f
o
r
Yelp
,
an
d
also
r
ed
u
ce
d
tim
e
co
m
p
l
ex
ity
s
ig
n
if
ican
tly
.
W
h
ile
th
e
co
n
v
en
t
io
n
al
SVM
class
if
ier
to
o
k
2
2
.
0
3
s
ec
o
n
d
s
f
o
r
Am
az
o
n
an
d
1
7
.
3
7
s
ec
o
n
d
s
f
o
r
Yelp
,
t
h
e
o
p
tim
ize
d
en
s
em
b
le
s
y
s
tem
cu
t
th
ese
tim
es to
1
8
.
0
4
s
ec
o
n
d
s
a
n
d
1
3
.
5
3
s
ec
o
n
d
s
,
r
esp
ec
tiv
ely
.
5.
CO
NCLU
SI
O
N
T
h
e
o
n
lin
e
s
p
am
d
etec
tio
n
s
y
s
tem
d
ev
elo
p
ed
in
th
is
r
e
s
ea
r
ch
en
h
an
ce
d
th
e
SVM
s
y
s
tem
b
y
r
ep
lacin
g
E
u
clid
ea
n
d
is
tan
ce
with
th
e
Ma
h
ala
n
o
b
is
d
is
tan
ce
m
ea
s
u
r
e
an
d
th
en
en
s
em
b
le
d
with
a
class
if
ier
g
iv
in
g
a
b
etter
r
esu
lt
th
an
a
c
o
n
v
en
tio
n
al
m
eth
o
d
.
T
h
e
r
esu
lts
o
b
tain
ed
s
h
o
w
th
at
th
e
c
o
m
b
in
atio
n
o
f
s
p
ee
d
o
p
tim
izatio
n
an
d
th
e
h
y
p
e
r
p
l
an
e
co
n
s
tr
u
ctio
n
wh
ile
u
s
in
g
th
e
Ma
h
alan
o
b
is
d
is
tan
ce
m
ea
s
u
r
e
h
as
a
h
ig
h
im
p
ac
t
o
n
th
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
SR
D
s
y
s
tem
b
o
th
in
ter
m
s
o
f
class
if
icatio
n
,
ac
cu
r
ac
y
an
d
s
p
ee
d
co
m
p
ar
e
d
with
class
ical
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
class
if
ier
.
T
h
e
p
r
o
p
o
s
ed
s
y
s
tem
s
ca
n
b
e
f
u
r
th
er
im
p
r
o
v
e
d
b
y
in
clu
d
in
g
an
o
u
tlier
d
etec
tio
n
alg
o
r
ith
m
,
th
at
ca
n
d
etec
t
ab
n
o
r
m
al
b
e
h
av
io
r
s
in
r
ev
iews.
Dif
f
er
en
t
l
in
g
u
is
tic
co
n
s
tr
u
cts
s
u
ch
as
m
o
d
if
ier
s
,
n
eg
atio
n
s
,
em
o
jis
,
an
d
ir
o
n
ic
wo
r
d
s
ar
e
n
o
t
tak
e
n
in
t
o
c
o
n
s
id
er
atio
n
,
b
u
t
th
e
y
ca
n
all
b
e
u
s
ed
to
im
p
r
o
v
e
th
e
ef
f
ec
tiv
en
ess
o
f
th
e
p
r
o
p
o
s
ed
s
y
s
tem
.
RE
F
E
R
E
NC
E
S
[
1
]
A
.
S
.
C
h
a
d
c
h
a
n
k
a
r
,
K
.
B
o
r
a
t
e
,
a
n
d
P
.
G
o
m
b
h
i
,
“
O
p
i
n
i
o
n
m
i
n
i
n
g
f
r
o
m
o
n
l
i
n
e
p
r
o
d
u
c
t
r
e
v
i
e
w
a
n
d
o
n
l
i
n
e
p
r
o
d
u
c
t
’
s
r
a
t
i
n
g
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
S
c
i
e
n
t
i
f
i
c
a
n
d
T
e
c
h
n
o
l
o
g
y
Re
se
a
r
c
h
,
v
o
l
.
9
,
n
o
.
1
,
p
p
.
3
3
1
7
–
3
3
2
0
,
2
0
2
0
.
[
2
]
J.
Z
h
a
n
g
,
W
.
Z
h
e
n
g
,
a
n
d
S
.
W
a
n
g
,
“
Th
e
s
t
u
d
y
o
f
t
h
e
e
f
f
e
c
t
o
f
o
n
l
i
n
e
r
e
v
i
e
w
o
n
p
u
r
c
h
a
se
b
e
h
a
v
i
o
r
:
c
o
mp
a
r
i
n
g
t
h
e
t
w
o
r
e
se
a
r
c
h
met
h
o
d
s,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
C
r
o
w
d
S
c
i
e
n
c
e
,
v
o
l
.
4
,
n
o
.
1
,
p
p
.
7
3
–
8
6
,
2
0
2
0
,
d
o
i
:
1
0
.
1
1
0
8
/
I
JC
S
-
10
-
2
0
1
9
-
0
0
2
7
.
[
3
]
Bar
-
I
l
a
n
U
n
i
v
e
r
s
i
t
y
,
“
T
h
e
b
r
a
i
n
i
n
s
p
i
r
e
s a
n
e
w
t
y
p
e
o
f
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
,
”
S
c
i
e
n
c
e
N
e
w
s.
S
c
i
e
n
c
e
D
a
i
l
y
,
2
0
1
9
.
[
4
]
E.
C
h
a
v
o
l
l
a
,
A
.
V
a
l
d
i
v
i
a
,
P
.
D
i
a
z
,
D
.
Za
l
d
i
v
a
r
,
E
.
C
u
e
v
a
s,
a
n
d
M
.
A
.
P
e
r
e
z
,
“
I
mp
r
o
v
e
d
u
n
su
p
e
r
v
i
se
d
c
o
l
o
r
se
g
men
t
a
t
i
o
n
u
s
i
n
g
a
mo
d
i
f
i
e
d
H
S
V
c
o
l
o
r
m
o
d
e
l
a
n
d
a
b
a
g
g
i
n
g
p
r
o
c
e
d
u
r
e
i
n
K
-
me
a
n
s++
a
l
g
o
r
i
t
h
m
,
”
Ma
t
h
e
m
a
t
i
c
a
l
Pr
o
b
l
e
m
s
i
n
En
g
i
n
e
e
ri
n
g
,
v
o
l
.
2
0
1
8
,
n
o
.
1
,
p
.
2
7
8
6
9
5
2
,
2
0
1
8
,
d
o
i
:
1
0
.
1
1
5
5
/
2
0
1
8
/
2
7
8
6
9
5
2
.
[
5
]
J.
C
l
u
n
e
,
“
A
I
-
G
A
s:
A
I
-
g
e
n
e
r
a
t
i
n
g
a
l
g
o
r
i
t
h
ms
,
a
n
a
l
t
e
r
n
a
t
e
p
a
r
a
d
i
g
m
f
o
r
p
r
o
d
u
c
i
n
g
g
e
n
e
r
a
l
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
,
”
a
rX
i
v
p
re
p
r
i
n
t
1
9
0
5
.
1
0
9
8
5
,
2
0
1
9
,
d
o
i
:
1
0
.
4
8
5
5
0
/
a
r
X
i
v
.
1
9
0
5
.
1
0
9
8
5
.
[
6
]
E.
El
mu
r
n
g
i
a
n
d
A
.
G
h
e
r
b
i
,
“
A
n
e
mp
i
r
i
c
a
l
s
t
u
d
y
o
n
d
e
t
e
c
t
i
n
g
f
a
k
e
r
e
v
i
e
w
s
u
si
n
g
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s
,
”
i
n
2
0
1
7
S
e
v
e
n
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
I
n
n
o
v
a
t
i
v
e
C
o
m
p
u
t
i
n
g
T
e
c
h
n
o
l
o
g
y
(
I
N
T
EC
H
)
,
A
u
g
.
2
0
1
7
,
p
p
.
1
0
7
–
1
1
4
,
d
o
i
:
1
0
.
1
1
0
9
/
I
N
TEC
H
.
2
0
1
7
.
8
1
0
2
4
4
2
.
[
7
]
F
.
A
b
r
i
,
L
.
F
.
G
u
t
i
e
r
r
e
z
,
A
.
S
.
N
a
mi
n
,
K
.
S
.
Jo
n
e
s
,
a
n
d
D
.
R
.
W
.
S
e
a
r
s,
“
L
i
n
g
u
i
st
i
c
f
e
a
t
u
r
e
s
f
o
r
d
e
t
e
c
t
i
n
g
f
a
k
e
r
e
v
i
e
w
s,
”
i
n
2
0
2
0
1
9
t
h
I
EE
E
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
(
I
C
MLA
)
,
D
e
c
.
2
0
2
0
,
p
p
.
3
5
2
–
3
5
9
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
M
LA
5
1
2
9
4
.
2
0
2
0
.
0
0
0
6
3
.
[
8
]
X
.
W
a
n
g
,
X
.
Z
h
a
n
g
,
C
.
Ji
a
n
g
,
a
n
d
H
.
L
i
u
,
“
I
d
e
n
t
i
f
i
c
a
t
i
o
n
o
f
f
a
k
e
r
e
v
i
e
w
s
u
s
i
n
g
sem
a
n
t
i
c
a
n
d
b
e
h
a
v
i
o
r
a
l
f
e
a
t
u
r
e
s,”
i
n
2
0
1
8
4
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
I
n
f
o
rm
a
t
i
o
n
Ma
n
a
g
e
m
e
n
t
(
I
C
I
M)
,
M
a
y
2
0
1
8
,
p
p
.
9
2
–
9
7
,
d
o
i
:
1
0
.
1
1
0
9
/
I
N
F
O
M
A
N
.
2
0
1
8
.
8
3
9
2
8
1
6
.
[
9
]
Ş
.
Ö
.
B
i
r
i
m,
I
.
K
a
z
a
n
c
o
g
l
u
,
S
.
K
.
M
a
n
g
l
a
,
A
.
K
a
h
r
a
m
a
n
,
S
.
K
u
mar,
a
n
d
Y
.
K
a
z
a
n
c
o
g
l
u
,
“
D
e
t
e
c
t
i
n
g
f
a
k
e
r
e
v
i
e
w
s
t
h
r
o
u
g
h
t
o
p
i
c
mo
d
e
l
l
i
n
g
,
”
J
o
u
r
n
a
l
o
f
B
u
s
i
n
e
ss R
e
se
a
rc
h
,
v
o
l
.
1
4
9
,
p
p
.
8
8
4
–
9
0
0
,
O
c
t
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
b
u
sr
e
s.
2
0
2
2
.
0
5
.
0
8
1
.
[
1
0
]
S
.
N
.
A
l
s
u
b
a
r
i
,
M
.
B
.
S
h
e
l
k
e
,
a
n
d
S
.
N
.
D
e
s
h
m
u
k
h
,
“
F
a
k
e
r
e
v
i
e
w
s
i
d
e
n
t
i
f
i
c
a
t
i
o
n
b
a
se
d
o
n
d
e
e
p
c
o
m
p
u
t
a
t
i
o
n
a
l
l
i
n
g
u
i
st
i
c
f
e
a
t
u
r
e
s,”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
A
d
v
a
n
c
e
d
S
c
i
e
n
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
2
9
,
n
o
.
8
s
,
p
p
.
3
8
4
6
–
3
8
5
6
,
2
0
2
0
.
[
1
1
]
L.
G
u
t
i
e
r
r
e
z
-
Es
p
i
n
o
z
a
,
F
.
A
b
r
i
,
A
.
S
.
N
a
mi
n
,
K
.
S
.
J
o
n
e
s,
a
n
d
D
.
R
.
W
.
S
e
a
r
s
,
“
F
a
k
e
r
e
v
i
e
w
s
d
e
t
e
c
t
i
o
n
t
h
r
o
u
g
h
e
n
s
e
m
b
l
e
l
e
a
r
n
i
n
g
,
”
a
r
Xi
v
p
re
p
ri
n
t
2
0
0
6
.
0
7
9
1
2
,
Ju
n
.
2
0
2
0
,
d
o
i
:
1
0
.
4
8
5
5
0
/
a
r
X
i
v
.
2
0
0
6
.
0
7
9
1
2
.
[
1
2
]
W
.
L
i
u
,
J.
H
e
,
S
.
H
a
n
,
F
.
C
a
i
,
Z
.
Y
a
n
g
,
a
n
d
N
.
Z
h
u
,
“
A
me
t
h
o
d
f
o
r
t
h
e
d
e
t
e
c
t
i
o
n
o
f
f
a
k
e
r
e
v
i
e
w
s
b
a
se
d
o
n
t
e
mp
o
r
a
l
f
e
a
t
u
r
e
s
o
f
r
e
v
i
e
w
s
a
n
d
c
o
mm
e
n
t
s,
”
I
EEE
En
g
i
n
e
e
ri
n
g
M
a
n
a
g
e
m
e
n
t
R
e
v
i
e
w
,
v
o
l
.
4
7
,
n
o
.
4
,
p
p
.
6
7
–
7
9
,
D
e
c
.
2
0
1
9
,
d
o
i
:
1
0
.
1
1
0
9
/
E
M
R
.
2
0
1
9
.
2
9
2
8
9
6
4
.
[
1
3
]
J.
K
.
R
o
u
t
,
A
.
D
a
l
mi
a
,
K
.
-
K
.
R
.
C
h
o
o
,
S
.
B
a
k
s
h
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
.
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:
2
5
0
2
-
4
7
52
F
a
ke
r
ev
iew
d
etec
tio
n
u
s
in
g
e
n
h
a
n
ce
d
en
s
emb
le
s
u
p
p
o
r
t v
ec
to
r
…
(
S
ee
n
ia
J
o
s
ep
h
)
485
[
1
4
]
L.
Y
o
u
,
Q
.
P
e
n
g
,
Z.
X
i
o
n
g
,
D
.
H
e
,
M
.
Q
i
u
,
a
n
d
X
.
Z
h
a
n
g
,
“
I
n
t
e
g
r
a
t
i
n
g
a
sp
e
c
t
a
n
a
l
y
si
s
a
n
d
l
o
c
a
l
o
u
t
l
i
e
r
f
a
c
t
o
r
f
o
r
i
n
t
e
l
l
i
g
e
n
t
r
e
v
i
e
w
s
p
a
m
d
e
t
e
c
t
i
o
n
,
”
F
u
t
u
re
G
e
n
e
r
a
t
i
o
n
C
o
m
p
u
t
e
r
S
y
s
t
e
m
s
,
v
o
l
.
1
0
2
,
p
p
.
1
6
3
–
1
7
2
,
J
a
n
.
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
f
u
t
u
r
e
.
2
0
1
9
.
0
7
.
0
4
4
.
[
1
5
]
I
.
C
h
a
t
t
e
r
j
e
e
,
M
.
Z
h
o
u
,
A
.
A
b
u
s
o
r
r
a
h
,
K
.
S
e
d
r
a
o
u
i
,
a
n
d
A
.
A
l
a
b
d
u
l
w
a
h
a
b
,
“
S
t
a
t
i
st
i
c
s
-
b
a
se
d
o
u
t
l
i
e
r
d
e
t
e
c
t
i
o
n
a
n
d
c
o
r
r
e
c
t
i
o
n
met
h
o
d
f
o
r
A
ma
z
o
n
c
u
st
o
mer r
e
v
i
e
w
s
,
”
E
n
t
ro
p
y
,
v
o
l
.
2
3
,
n
o
.
1
2
,
p
.
1
6
4
5
,
D
e
c
.
2
0
2
1
,
d
o
i
:
1
0
.
3
3
9
0
/
e
2
3
1
2
1
6
4
5
.
[
1
6
]
S
.
A
f
z
a
l
,
A
.
A
f
z
a
l
,
M
.
A
mi
n
,
S
.
S
a
l
e
e
m,
N
.
A
l
i
,
a
n
d
M
.
S
a
j
i
d
,
“
A
n
o
v
e
l
a
p
p
r
o
a
c
h
f
o
r
o
u
t
l
i
e
r
d
e
t
e
c
t
i
o
n
i
n
m
u
l
t
i
v
a
r
i
a
t
e
d
a
t
a
,
”
Ma
t
h
e
m
a
t
i
c
a
l
Pr
o
b
l
e
m
s i
n
En
g
i
n
e
e
ri
n
g
,
p
p
.
1
–
1
2
,
O
c
t
.
2
0
2
1
,
d
o
i
:
1
0
.
1
1
5
5
/
2
0
2
1
/
1
8
9
9
2
2
5
.
[
1
7
]
A
.
G
a
u
t
a
m
,
V
.
V
e
n
k
t
e
sh
,
a
n
d
S
.
M
a
s
u
d
,
“
F
a
k
e
n
e
w
s
d
e
t
e
c
t
i
o
n
s
y
st
e
m
u
si
n
g
X
LN
e
t
m
o
d
e
l
w
i
t
h
t
o
p
i
c
d
i
st
r
i
b
u
t
i
o
n
s
:
C
O
N
S
T
R
A
I
N
T
@
A
A
A
I
2
0
2
1
s
h
a
r
e
d
t
a
s
k
,
”
i
n
C
o
m
m
u
n
i
c
a
t
i
o
n
s
i
n
C
o
m
p
u
t
e
r
a
n
d
I
n
f
o
r
m
a
t
i
o
n
S
c
i
e
n
c
e
,
S
p
r
i
n
g
e
r
,
2
0
2
1
,
p
p
.
1
8
9
–
2
0
0
.
[
1
8
]
F
.
M
u
h
a
m
ma
d
a
n
d
S
.
A
h
me
d
,
“
F
a
k
e
r
e
v
i
e
w
d
e
t
e
c
t
i
o
n
u
s
i
n
g
p
r
i
n
c
i
p
a
l
c
o
mp
o
n
e
n
t
a
n
a
l
y
si
s
a
n
d
a
c
t
i
v
e
l
e
a
r
n
i
n
g
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
C
o
m
p
u
t
e
r A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
1
7
8
,
n
o
.
4
8
,
p
p
.
4
2
–
4
8
,
2
0
1
9
,
d
o
i
:
1
0
.
5
1
2
0
/
i
j
c
a
2
0
1
9
9
1
9
4
1
8
.
[
1
9
]
S
.
N
.
A
l
s
u
b
a
r
i
,
S
.
N
.
D
e
s
h
m
u
k
h
,
M
.
H
.
A
l
-
A
d
h
a
i
l
e
h
,
F
.
W
.
A
l
s
a
a
d
e
,
a
n
d
T.
H
.
H
.
A
l
d
h
y
a
n
i
,
“
D
e
v
e
l
o
p
me
n
t
o
f
i
n
t
e
g
r
a
t
e
d
n
e
u
r
a
l
n
e
t
w
o
r
k
m
o
d
e
l
f
o
r
i
d
e
n
t
i
f
i
c
a
t
i
o
n
o
f
f
a
k
e
r
e
v
i
e
w
s i
n
e
-
c
o
mm
e
r
c
e
u
s
i
n
g
m
u
l
t
i
d
o
ma
i
n
d
a
t
a
se
t
s,”
A
p
p
l
i
e
d
Bi
o
n
i
c
s
a
n
d
B
i
o
m
e
c
h
a
n
i
c
s
,
p
p
.
1
–
1
1
,
A
p
r
.
2
0
2
1
,
d
o
i
:
1
0
.
1
1
5
5
/
2
0
2
1
/
5
5
2
2
5
7
4
.
[
2
0
]
R
.
M
.
N
a
b
i
,
S
.
A
.
S
a
e
e
d
,
a
n
d
A
.
M
.
W
.
A
b
d
i
,
“
F
e
a
t
u
r
e
e
n
g
i
n
e
e
r
i
n
g
f
o
r
s
t
o
c
k
p
r
i
c
e
p
r
e
d
i
c
t
i
o
n
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
Ad
v
a
n
c
e
d
S
c
i
e
n
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
2
9
,
n
o
.
1
2
s,
p
p
.
2
4
8
6
–
2
4
9
6
,
2
0
2
0
,
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
:
/
/
sersc
.
o
r
g
/
j
o
u
r
n
a
l
s/
i
n
d
e
x
.
p
h
p
/
I
JA
S
T/
a
r
t
i
c
l
e
/
v
i
e
w
/
2
4
7
2
2
.
[
2
1
]
K
.
S
.
R
e
d
d
y
a
n
d
E
.
S
.
R
e
d
d
y
,
“
U
si
n
g
r
e
d
u
c
e
d
set
o
f
f
e
a
t
u
r
e
s
t
o
d
e
t
e
c
t
sp
a
m
i
n
t
w
i
t
t
e
r
d
a
t
a
w
i
t
h
d
e
c
i
s
i
o
n
t
r
e
e
a
n
d
K
N
N
c
l
a
ssi
f
i
e
r
a
l
g
o
r
i
t
h
ms,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
I
n
n
o
v
a
t
i
v
e
T
e
c
h
n
o
l
o
g
y
a
n
d
E
x
p
l
o
r
i
n
g
E
n
g
i
n
e
e
r
i
n
g
,
v
o
l
.
8
,
n
o
.
9
,
p
p
.
6
–
1
2
,
2
0
1
9
,
d
o
i
:
1
0
.
3
5
9
4
0
/
i
j
i
t
e
e
.
f
3
6
1
6
.
0
7
8
9
1
9
.
[
2
2
]
T.
V
.
D
i
n
h
,
H
.
N
g
u
y
e
n
,
X
.
-
L.
Tr
a
n
,
a
n
d
N
.
-
D
.
H
o
a
n
g
,
“
P
r
e
d
i
c
t
i
n
g
r
a
i
n
f
a
l
l
-
i
n
d
u
c
e
d
s
o
i
l
e
r
o
s
i
o
n
b
a
se
d
o
n
a
h
y
b
r
i
d
i
z
a
t
i
o
n
o
f
a
d
a
p
t
i
v
e
d
i
f
f
e
r
e
n
t
i
a
l
e
v
o
l
u
t
i
o
n
a
n
d
su
p
p
o
r
t
v
e
c
t
o
r
m
a
c
h
i
n
e
c
l
a
ssi
f
i
c
a
t
i
o
n
,
”
Ma
t
h
e
m
a
t
i
c
a
l
Pr
o
b
l
e
m
s i
n
E
n
g
i
n
e
e
r
i
n
g
,
p
p
.
1
–
2
0
,
F
e
b
.
2
0
2
1
,
d
o
i
:
1
0
.
1
1
5
5
/
2
0
2
1
/
6
6
4
7
8
2
9
.
[
2
3
]
B
.
G
a
y
e
,
D
.
Z
h
a
n
g
,
a
n
d
A
.
W
u
l
a
m
u
,
“
I
mp
r
o
v
e
me
n
t
o
f
su
p
p
o
r
t
v
e
c
t
o
r
m
a
c
h
i
n
e
a
l
g
o
r
i
t
h
m
i
n
b
i
g
d
a
t
a
b
a
c
k
g
r
o
u
n
d
,
”
M
a
t
h
e
m
a
t
i
c
a
l
Pro
b
l
e
m
s i
n
En
g
i
n
e
e
ri
n
g
,
p
p
.
1
–
9
,
J
u
n
.
2
0
2
1
,
d
o
i
:
1
0
.
1
1
5
5
/
2
0
2
1
/
5
5
9
4
8
9
9
.
[
2
4
]
F
.
R
.
L
u
m
b
a
n
r
a
j
a
,
E.
F
i
t
r
i
,
A
r
d
i
a
n
s
y
a
h
,
A
.
J
u
n
a
i
d
i
,
a
n
d
R
.
P
r
a
b
o
w
o
,
“
A
b
st
r
a
c
t
c
l
a
ssi
f
i
c
a
t
i
o
n
u
s
i
n
g
s
u
p
p
o
r
t
v
e
c
t
o
r
m
a
c
h
i
n
e
a
l
g
o
r
i
t
h
m (c
a
se
s
t
u
d
y
:
a
b
s
t
r
a
c
t
i
n
a
c
o
mp
u
t
e
r
s
c
i
e
n
c
e
j
o
u
r
n
a
l
)
,
”
J
o
u
r
n
a
l
o
f
Ph
y
s
i
c
s:
C
o
n
f
e
re
n
c
e
S
e
r
i
e
s
,
v
o
l
.
1
7
5
1
,
n
o
.
1
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
8
8
/
1
7
4
2
-
6
5
9
6
/
1
7
5
1
/
1
/
0
1
2
0
4
2
.
[
2
5
]
P
.
Y
ı
l
d
ı
r
ı
m,
U
.
K
.
B
i
r
a
n
t
,
a
n
d
D
.
B
i
r
a
n
t
,
“
E
B
O
C
:
e
n
se
mb
l
e
-
b
a
se
d
o
r
d
i
n
a
l
c
l
a
ssi
f
i
c
a
t
i
o
n
i
n
t
r
a
n
s
p
o
r
t
a
t
i
o
n
,
”
J
o
u
rn
a
l
o
f
A
d
v
a
n
c
e
d
T
ra
n
s
p
o
rt
a
t
i
o
n
,
p
p
.
1
–
1
7
,
M
a
r
.
2
0
1
9
,
d
o
i
:
1
0
.
1
1
5
5
/
2
0
1
9
/
7
4
8
2
1
3
8
.
[
2
6
]
H
.
B
.
A
h
m
a
d
,
“
E
n
sem
b
l
e
c
l
a
ss
i
f
i
e
r
b
a
se
d
s
p
e
c
t
r
u
m
se
n
si
n
g
i
n
c
o
g
n
i
t
i
v
e
r
a
d
i
o
n
e
t
w
o
r
k
s
,
”
W
i
re
l
e
ss
C
o
m
m
u
n
i
c
a
t
i
o
n
s
a
n
d
M
o
b
i
l
e
C
o
m
p
u
t
i
n
g
,
p
p
.
1
–
1
6
,
Ja
n
.
2
0
1
9
,
d
o
i
:
1
0
.
1
1
5
5
/
2
0
1
9
/
9
2
5
0
5
6
2
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
S
e
e
n
ia
J
o
se
p
h
is
a
re
se
a
rc
h
sc
h
o
lar
a
t
Ka
rp
a
g
a
m
Ac
a
d
e
m
y
o
f
Hig
h
e
r
E
d
u
c
a
ti
o
n
,
Co
imb
a
to
re
,
a
n
d
a
n
As
sista
n
t
P
r
o
fe
ss
o
r
a
t
Bh
a
ra
ta
M
a
ta
Co
l
leg
e
.
S
h
e
o
b
tain
e
d
h
e
r
m
a
ste
r'
s
d
e
g
re
e
in
Co
m
p
u
ter
A
p
p
li
c
a
ti
o
n
s
fro
m
Ca
li
c
u
t
Un
i
v
e
rsity
in
2
0
0
7
.
He
r
re
se
a
rc
h
fo
c
u
se
s
o
n
d
a
ta
m
in
in
g
te
c
h
n
iq
u
e
s
a
p
p
li
e
d
a
c
ro
ss
v
a
rio
u
s
d
o
m
a
in
s.
To
d
a
te,
s
h
e
h
a
s
a
u
th
o
re
d
5
p
a
p
e
rs
in
n
a
ti
o
n
a
l
a
n
d
i
n
tern
a
ti
o
n
a
l
j
o
u
r
n
a
l
s
a
n
d
c
o
n
tri
b
u
ted
t
o
2
b
o
o
k
c
h
a
p
t
e
rs.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
se
e
n
iajo
se
p
h
@g
m
a
il
.
c
o
m
.
Dr
.
S
.
H
e
m
a
la
th
a
is
p
re
se
n
tl
y
a
n
As
so
c
iate
P
ro
fe
ss
o
r
a
n
d
Re
se
a
rc
h
G
u
id
e
a
t
Ka
rp
a
g
a
m
Ac
a
d
e
m
y
o
f
Hi
g
h
e
r
Ed
u
c
a
ti
o
n
in
Co
imb
a
t
o
re
.
He
r
r
e
se
a
rc
h
in
tere
sts
e
n
c
o
m
p
a
ss
d
a
ta
m
in
in
g
,
ima
g
e
p
r
o
c
e
ss
in
g
,
a
n
d
n
e
two
rk
i
n
g
.
Wi
th
a
p
r
o
li
fic
sc
h
o
larly
o
u
t
p
u
t,
sh
e
h
a
s
a
u
th
o
re
d
o
v
e
r
1
0
p
a
p
e
rs,
wh
ich
h
a
v
e
b
e
e
n
p
u
b
li
sh
e
d
in
S
c
o
p
u
s
-
i
n
d
e
x
e
d
jo
u
rn
a
ls,
a
l
o
n
g
with
a
n
a
d
d
i
ti
o
n
a
l
5
p
a
p
e
rs
in
p
e
e
r
-
re
v
iew
e
d
jo
u
r
n
a
ls.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
d
rh
e
m
a
sh
a
n
m
u
g
a
m
@g
m
a
il
.
c
o
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.