I
A
E
S
I
n
t
e
r
n
at
io
n
al
Jou
r
n
al
of
A
r
t
if
ic
ia
l
I
n
t
e
ll
ig
e
n
c
e
(
I
J
-
AI
)
V
ol
.
14
, N
o.
3
,
J
une
2025
, pp.
2001
~
2011
I
S
S
N
:
2252
-
8938
,
D
O
I
:
10.11591/
ij
a
i.
v
14
.i
3
.pp
2001
-
2011
2001
Jou
r
n
al
h
om
e
page
:
ht
tp
:
//
ij
ai
.
ia
e
s
c
or
e
.c
om
H
yb
r
i
d
se
m
a
n
t
i
c
m
od
e
l
b
a
se
d
on
m
a
c
h
i
n
e
l
e
ar
n
i
n
g f
or
se
n
t
i
m
e
n
t
c
l
ass
i
f
i
c
at
i
on
of
c
on
su
m
e
r
r
e
vi
e
w
s
P
al
an
ir
aj
R
aj
i
d
u
r
ai
P
ar
vat
h
y
1
,
N
agar
aj
a
n
M
oh
an
k
u
m
ar
2
, R
aj
e
n
d
r
an
S
h
ob
ig
a
3
,
G
ou
r
S
u
n
d
ar
M
it
r
a T
h
ak
u
r
4
,
M
am
at
h
a
B
an
d
ar
u
5
, V
e
lu
s
am
y
S
u
j
at
h
a
6
,
S
h
an
m
u
gam
S
u
j
at
h
a
7
1
P
r
oj
e
c
t
M
a
na
ge
r
, M
pha
s
i
s
C
or
por
a
t
i
on, C
ha
ndl
e
r
,
U
ni
t
e
d S
t
a
t
e
s
of
A
m
e
r
i
c
a
2
S
ym
bi
os
i
s
I
ns
t
i
t
ut
e
of
T
e
c
hnol
ogy, S
ym
bi
os
i
s
I
nt
e
r
na
t
i
ona
l
(
D
e
e
m
e
d U
ni
ve
r
s
i
t
y)
,
P
une
, I
ndi
a
3
D
e
pa
r
t
m
e
nt
of
E
l
e
c
t
r
oni
c
s
a
nd C
om
m
uni
c
a
t
i
on E
ngi
ne
e
r
i
ng, J
.
J.
C
ol
l
e
ge
of
E
ngi
ne
e
r
i
ng a
nd T
e
c
hnol
ogy, T
i
r
uc
hi
r
a
ppa
l
l
i
,
I
ndi
a
4
D
e
pa
r
t
m
e
nt
of
C
om
put
e
r
S
c
i
e
nc
e
a
nd E
ngi
ne
e
r
i
ng (
A
I
M
L
)
,
D
r
. B
. C
. R
oy E
ngi
ne
e
r
i
ng C
ol
l
e
ge
, D
ur
ga
pur
, I
ndi
a
5
D
e
pa
r
t
m
e
nt
of
E
l
e
c
t
r
oni
c
s
a
nd C
om
m
uni
c
a
t
i
on E
ngi
ne
e
r
i
ng, K
one
r
u L
a
ks
hm
a
i
a
h E
duc
a
t
i
on F
ounda
t
i
on, G
unt
ur
, I
ndi
a
6
D
e
pa
r
t
m
e
nt
of
E
l
e
c
t
r
oni
c
s
a
nd C
om
m
uni
c
a
t
i
on E
ngi
ne
e
r
i
ng, S
hr
e
e
S
a
t
hya
m
C
ol
l
e
ge
of
E
ngi
ne
e
r
i
ng a
nd T
e
c
hnol
ogy, S
a
nga
gi
r
i
, I
nd
i
a
7
D
e
pa
r
t
m
e
nt
of
B
i
om
e
di
c
a
l
E
ngi
ne
e
r
i
ng, S
a
ve
e
t
ha
S
c
hool
of
E
ngi
ne
e
r
i
ng, S
a
ve
e
t
ha
I
ns
t
i
t
ut
e
of
M
e
di
c
a
l
a
nd T
e
c
hni
c
a
l
S
c
i
e
nc
e
s
,
S
a
ve
e
t
ha
U
ni
ve
r
s
i
t
y, C
he
nna
i
, I
ndi
a
A
r
t
ic
le
I
n
f
o
A
B
S
T
R
A
C
T
A
r
ti
c
le
h
is
to
r
y
:
R
e
c
e
iv
e
d
F
e
b 27, 2024
R
e
vi
s
e
d
D
e
c
20, 2024
A
c
c
e
pt
e
d
J
a
n 27, 2025
Digital
information
is
regularly
produced
from
a
variety
of
s
ources,
including
social
media
and
customer
service
reviews.
For
the
purp
ose
of
increasin
g customer
happiness, th
is
written da
ta
must be
processe
d to
extract
user
comments.
Consumers
typically
share
comments
and
thoughts
about
consumab
le
items,
technolo
gical
goods,
and
services
suppli
ed
for
pa
yment
in
the
modern
period
of
consumerism
with
simple
access
to
social
networking
globe.
Each
object
has
a
plethora
of
remarks
or
though
ts
that
demand
specia
l
attention
due
to
their
sentimental
worth,
especially
in
the
written
portions.
The
goal
of
the
current
project
is
to
do
sentiment
pre
diction
on
the
Amazon
Electronics
,
Kindle,
and
Gift
Card
datas
ets.
In
or
der
to
predict sentiment and evaluate utilizing many
execution
s
evaluates
ad
mitti
ng
accuracy,
recall,
and
F1
-
score,
a
hybrid
soft
voting
ensemble
meth
od
that
combines
lexical
and
ensemble
methodo
logies
is
proposed
in
this
st
udy.
In
additio
n
to
calculati
ng
a
subjecti
vity
score
and
sentimen
t
score,
this
stu
dy
also
suggests
a
non
-
interpretiv
e
sentiment
cl
ass
label
that
may
be
u
sed
to
assess
the
sign
of
the
evaluations
applying
suggested
method
for
sen
timent
categorizati
on.
The
effectivenes
s
of
our
suggest
ed
ensemble
model
is
examined
using
datasets
from
Amazon
customer
product
reviews,
a
nd
we
found an im
provement of
2
-
5% in
accuracy c
ompared to the current
st
ate
-
of
-
the
-
art ensemb
le metho
d.
K
e
y
w
o
r
d
s
:
C
ons
um
e
r
r
e
vi
e
w
s
E
ns
e
m
bl
e
c
a
te
gor
iz
a
ti
on
L
e
xi
c
on mode
l
S
e
nt
im
e
nt
a
na
ly
s
is
S
e
nt
im
e
nt
s
c
or
e
This is an
open
acce
ss artic
le unde
r the
CC BY
-
SA
license.
C
or
r
e
s
pon
di
n
g A
u
th
or
:
N
a
ga
r
a
ja
n M
oha
nkuma
r
S
ym
bi
os
is
I
ns
ti
tu
te
of
T
e
c
hnol
ogy, S
ym
bi
os
is
I
nt
e
r
na
ti
ona
l
(
D
e
e
m
e
d U
ni
ve
r
s
it
y)
N
a
gpur
C
a
m
pus
,
P
une
, I
ndi
a
E
m
a
il
:
nm
kpr
of
e
s
s
or
@
gm
a
il
.c
om
1.
I
N
T
R
O
D
U
C
T
I
O
N
N
um
e
r
ous
s
e
c
to
r
s
f
or
da
ta
a
na
ly
s
is
ha
v
e
e
m
e
r
ge
d
a
s
a
r
e
s
ul
t
of
th
e
e
nor
m
ous
a
m
ount
of
de
ta
il
s
c
r
e
a
te
d
by
popula
r
s
o
c
ia
l
m
e
di
a
pl
a
tf
or
m
s
. T
he
goa
l
of
s
e
nt
im
e
nt
a
na
ly
s
is
(
S
A
)
,
in
pa
r
ti
c
ul
a
r
,
is
to
gl
e
a
n
us
e
r
th
oughts
a
bout
a
good
or
s
e
r
vi
c
e
f
r
om
th
e
r
e
le
va
nt
te
xt
.
T
h
e
r
a
w
te
xt
m
us
t
be
a
na
ly
z
e
d
us
in
g
m
a
c
hi
ne
le
a
r
ni
ng
(
ML
)
a
lg
or
it
hm
s
a
lo
ng
w
it
h
na
tu
r
a
l
la
ngua
ge
pr
oc
e
s
s
in
g
(
N
L
P
)
te
c
hni
que
s
due
to
th
e
e
nor
m
ous
a
m
ount
of
in
f
or
m
a
ti
on
ge
ne
r
a
te
d
[
1]
.
S
e
nt
im
e
nt
e
xa
m
in
a
ti
on
m
a
de
pos
s
ib
le
th
r
ough
th
e
de
ta
il
th
a
t
s
oc
ia
l
m
e
di
a
is
a
n
e
c
e
s
s
a
r
y
d
e
vi
c
e
f
or
p
e
opl
e
a
nd
th
a
t
th
e
y
r
e
gul
a
r
ly
di
s
tr
ib
ut
e
th
e
ir
not
io
ns
.
E
xpr
e
s
s
io
ns
of
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
. 14, No. 3, J
une
2025
:
2001
-
2011
2002
s
e
nt
im
e
nt
in
te
xt
s
a
r
e
pr
e
di
c
te
d
by
S
A
s
tu
di
e
s
.
P
e
opl
e
'
s
te
xt
s
a
r
e
e
va
lu
a
te
d
f
or
th
e
ir
pos
it
iv
it
y,
ne
ga
ti
vi
ty
,
or
ne
ut
r
a
li
ty
.
B
us
in
e
s
s
e
s
c
a
n
do
a
pr
e
li
m
in
a
r
y
a
na
ly
s
is
of
ne
w
goods
,
a
nd
f
il
m
s
th
a
nks
to
S
A
[
2
]
.
O
n
th
e
ot
he
r
ha
nd,
in
or
de
r
to
a
c
hi
e
ve
a
s
ig
ni
f
ic
a
nt
in
c
r
e
a
s
e
in
a
c
c
ur
a
c
y,
th
is
m
e
th
odol
ogy
is
c
os
tl
y,
ti
m
e
-
ut
il
iz
in
g,
a
nd
m
a
y ne
e
d a
dva
nc
e
d t
e
c
hnol
ogy
[
3]
.
S
A
a
nd
N
L
P
a
r
e
be
c
om
in
g
in
c
r
e
a
s
in
gl
y
popula
r
a
s
a
r
e
s
ul
t.
S
A
us
e
s
to
ol
s
a
nd
m
e
th
ods
f
r
om
N
L
P
,
ML
,
a
nd
s
ta
ti
s
ti
c
s
to
de
te
r
m
in
e
th
e
c
ont
e
xt
ua
l
pol
a
r
it
y
of
te
xt
c
ont
e
nt
[
4]
.
T
he
in
te
r
ne
t
is
one
of
th
e
in
f
or
m
a
ti
on
s
our
c
e
s
f
or
S
A
th
a
t
c
ont
a
in
s
a
n
e
ve
r
-
gr
ow
in
g
a
m
o
unt
of
da
ta
in
a
va
r
ie
ty
of
m
ul
ti
m
e
di
a
f
or
m
a
ts
th
a
t
c
a
n
be
us
e
f
ul
to
gove
r
nm
e
nt
a
l
bodi
e
s
,
c
om
m
e
r
c
ia
l
or
ga
n
iz
a
ti
ons
,
a
nd
in
di
vi
dua
l
de
c
is
io
n
-
m
a
ke
r
s
[
5]
.
I
ni
ti
a
ll
y,
a
qui
c
k
s
tu
dy
w
a
s
c
onduc
te
d
to
id
e
nt
if
y
a
ge
ne
r
a
li
z
e
d
c
ons
um
e
r
r
e
vi
e
w
pa
tt
e
r
n
on
te
xt
ua
l
c
us
to
m
e
r
c
om
m
e
nt
s
s
ha
r
e
d
f
or
th
e
pr
oduc
ts
th
e
y
bough
t
onl
in
e
[
6]
.
T
he
s
e
onl
in
e
r
e
vi
e
w
s
a
r
e
pos
te
d
on
s
oc
ia
l
m
e
di
a
pl
a
tf
or
m
s
or
on
th
e
w
e
bs
it
e
s
of
r
e
put
a
bl
e
c
om
pa
ni
e
s
.
A
s
in
di
c
a
te
d
in
F
ig
ur
e
1,
c
ons
um
e
r
r
e
vi
e
w
s
pa
tt
e
r
n
c
a
n
be
di
vi
de
d
in
to
two
di
s
ti
nc
t
c
a
te
gor
ie
s
ba
s
e
d
on
th
e
in
c
li
na
ti
ons
of
th
e
r
e
vi
e
w
e
r
s
'
poi
nt
s
of
vi
e
w
:
pol
a
r
r
e
vi
e
w
s
a
nd non
-
pol
a
r
r
e
vi
e
w
s
.
F
ig
ur
e
1
. I
ll
us
tr
a
ti
on of
c
ons
um
e
r
r
e
vi
e
w
s
pa
tt
e
r
n
D
ue
to
th
e
ir
a
tt
it
udi
na
l
vi
e
w
poi
nt
s
la
nt
,
pol
a
r
e
va
lu
a
ti
ons
in
c
lu
de
bot
h
pos
it
iv
e
a
nd
n
e
ga
ti
ve
r
e
vi
e
w
s
, e
ve
n i
f
bot
h o
f
t
he
m
a
r
e
i
nt
e
r
p
r
e
ta
ti
ve
a
nd us
e
f
ul
f
r
om
t
he
pe
r
s
pe
c
ti
ve
of
t
he
t
e
xt
m
a
te
r
ia
l.
A
lt
hough
th
e
y
a
r
e
in
th
e
non
-
pol
a
r
r
e
vi
e
w
c
la
s
s
,
n
e
ut
r
a
l
e
va
lu
a
ti
ons
c
ont
a
in
e
duc
a
ti
ona
l
la
ngua
g
e
.
T
he
r
e
f
or
e
,
a
c
a
te
gor
y
of
r
e
vi
e
w
s
known
a
s
in
f
or
m
a
ti
ve
r
e
vi
e
w
s
in
c
lu
de
s
a
ll
th
r
e
e
ty
pe
s
of
e
va
lu
a
ti
on
s
:
pos
it
iv
e
,
ne
ga
ti
ve
,
a
nd
ne
ut
r
a
l
.
A
f
te
r
di
s
c
ove
r
in
g
a
la
c
k
of
in
te
r
pr
e
ti
ve
c
om
pone
nt
s
in
th
e
r
e
vi
e
w
te
xt
,
non
-
pol
a
r
r
e
vi
e
w
s
w
e
r
e
f
in
a
ll
y
gi
ve
n
a
s
e
pa
r
a
te
e
nt
it
y
f
or
th
e
f
ir
s
t
t
im
e
known
a
s
non
-
in
te
r
pr
e
ti
ve
r
e
vi
e
w
s
.
A
ddi
ti
ona
ll
y,
th
e
y
a
r
e
la
be
ll
e
d
a
s
"
uni
nf
or
m
a
ti
ve
r
e
vi
e
w
s
"
.
S
e
nt
e
nc
e
-
le
ve
l
a
na
ly
s
is
s
e
e
ks
to
d
e
c
id
e
if
a
s
p
e
c
if
ic
s
e
nt
e
nc
e
r
e
pr
e
s
e
nt
s
pos
it
iv
e
,
unf
a
vor
a
bl
e
(
ne
ga
ti
ve
)
o
r
ne
ut
r
a
l
opi
ni
ons
.
H
ow
e
ve
r
,
f
ur
th
e
r
in
-
de
pt
h
e
xa
m
in
a
ti
on
a
t
a
n
e
nt
it
y
o
r
a
s
pe
c
t
le
v
e
l
m
ig
ht
c
om
pl
e
te
ly
e
xpos
e
a
tt
it
ude
s
or
opi
ni
ons
a
bo
ut
m
a
ny
c
ha
r
a
c
te
r
is
ti
c
s
or
a
s
pe
c
t
s
of
a
s
pe
c
if
ic
e
nt
it
y.
S
om
e
of
th
e
n
e
w
ly
de
te
r
m
in
e
d
non
-
in
te
r
pr
e
ti
ve
r
e
vi
e
w
s
a
r
e
a
ls
o
gi
ve
n
s
e
pa
r
a
te
ly
w
it
h
c
us
to
m
e
r
c
om
m
e
nt
s
on
A
m
a
z
on
K
in
dl
e
a
nd
E
le
c
tr
oni
c
s
.
T
a
bl
e
1
pr
ovi
d
e
s
a
br
ie
f
s
um
m
a
r
y
of
te
xt
c
ont
e
nt
tr
e
nds
f
o
r
w
id
e
ly
a
c
c
e
pt
e
d
s
e
nt
im
e
nt
gr
oups
,
na
m
e
ly
po
s
it
iv
e
,
ne
g
a
ti
ve
,
a
nd
ne
ut
r
a
l.
E
xa
m
pl
e
s
of
r
e
c
e
nt
ly
di
s
c
ove
r
e
d
non
-
in
te
r
pr
e
ti
ve
r
e
vi
e
w
s
a
r
e
a
ls
o
pr
ovi
de
d
in
is
ol
a
ti
on
w
it
h
c
ons
um
e
r
c
om
m
e
nt
s
f
r
om
th
e
A
m
a
z
on
K
in
dl
e
da
ta
s
e
t
a
nd
th
e
e
le
c
tr
oni
c
pr
oduc
t
da
ta
s
e
t
[
7]
r
e
s
pe
c
ti
ve
ly
.
P
r
o
duc
t
r
e
vi
e
w
s
s
houl
d
a
lwa
ys
r
a
ti
ona
ll
y
r
e
la
te
to
a
pr
oduc
t'
s
qua
li
ty
or
ut
il
it
y va
lu
e
f
r
om
th
e
pe
r
s
pe
c
ti
ve
of
t
he
c
ons
um
e
r
. I
t
m
ig
ht
a
ls
o r
e
pr
e
s
e
nt
hi
s
r
a
nki
ng of
m
e
r
it
s
a
nd
de
m
e
r
it
s
.
I
n
a
ddi
ti
on,
th
e
e
xt
r
e
m
e
pol
a
r
r
e
vi
e
w
r
e
m
a
r
ks
a
r
e
li
ke
w
is
e
de
s
e
r
vi
ng
of
a
hi
g
h
s
ubj
e
c
ti
vi
ty
r
a
ti
ng.
T
a
bl
e
1.
E
xa
m
pl
e
of
i
nf
or
m
a
ti
ve
c
ons
um
e
r
r
e
vi
e
w
t
a
ke
n f
r
om
A
m
a
z
on kindl
e
da
ta
s
e
t
C
ons
um
e
r
r
e
vi
e
w
s
a
nd c
om
m
e
nt
s
S
e
nt
i
m
e
nt
c
l
a
s
s
i
f
i
c
a
t
i
on
G
r
e
a
t
r
e
s
ul
t
s
. I
l
ove
d i
t
.
P
os
i
t
i
ve
N
i
c
e
s
t
uf
f
a
l
l
a
r
ound f
a
nt
a
s
t
i
c
t
a
bl
e
t
f
or
t
he
m
one
y. T
he
na
vi
ga
t
i
on i
s
s
i
m
pl
e
.
P
os
i
t
i
ve
P
oor
vi
s
ua
l
s
, uns
t
a
bl
e
w
i
r
e
l
e
s
s
c
onne
c
t
i
on, a
nd ov
e
r
a
l
l
di
s
s
a
t
i
s
f
a
c
t
i
on m
a
de
m
e
r
e
gr
e
t
pur
c
ha
s
i
ng t
hi
s
pr
oduc
t
.
N
e
ga
t
i
ve
I
w
on'
t
e
ve
r
s
ugge
s
t
t
hi
s
t
o a
nyone
.
N
e
ga
t
i
ve
S
om
e
t
i
m
e
s
i
t
s
i
m
pl
y r
e
m
a
i
ns
i
n s
l
e
e
p m
ode
, r
e
qui
r
i
ng a
r
e
s
t
a
r
t
.
N
on
-
I
nt
e
r
pr
e
t
i
ve
N
ot
hi
ng t
o s
a
y a
bout
t
he
pr
oduc
t
N
on
-
I
nt
e
r
pr
e
t
i
ve
C
oul
d not
unde
r
s
t
a
nd i
t
N
on
-
I
nt
e
r
pr
e
t
i
ve
V
e
r
y l
a
ggi
ng a
nd s
ui
t
a
bl
e
f
or
ki
ds
onl
y
N
e
ut
r
a
l
C
ons
i
de
r
a
n A
ndr
oi
d t
a
bl
e
t
i
f
you r
e
qui
r
e
m
or
e
t
ha
n ba
s
i
c
m
ode
l
.
N
e
ut
r
a
l
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
H
y
br
id
s
e
m
ant
ic
m
ode
l
bas
e
d on
m
ac
hi
ne
l
e
ar
ni
ng f
o
r
s
e
nt
ime
nt
…
(
P
al
ani
r
aj
R
aj
id
ur
ai
P
ar
v
at
hy
)
2003
A
s
a
r
e
s
ul
t,
it
is
s
ugge
s
te
d
in
th
is
r
e
s
e
a
r
c
h
th
a
t
th
e
r
e
vi
e
w
s
a
r
e
a
dde
d
to
a
n
e
w
c
a
te
gor
y
of
"
non
-
in
te
r
pr
e
ti
ve
"
r
e
vi
e
w
s
.
R
e
vi
e
w
s
th
a
t
f
a
ll
in
to
th
e
"
non
-
in
te
r
pr
e
ti
ve
"
c
a
te
gor
y
a
r
e
"
non
-
pol
a
r
"
a
nd
de
voi
d
of
in
f
or
m
a
ti
on
.
A
s
a
r
e
s
ul
t,
th
e
y
a
r
e
a
ls
o
c
a
te
gor
iz
e
d
a
s
"
u
ni
n
f
or
m
a
ti
ve
"
r
e
vi
e
w
s
.
T
he
f
ol
lo
w
in
g
is
a
l
is
t
of
th
is
s
tu
dy'
s
m
a
in
c
ont
r
ib
ut
io
ns
:
i)
to
de
te
r
m
in
e
how
s
ubj
e
c
ti
ve
l
y
th
e
uns
tr
uc
tu
r
e
d
c
on
s
um
e
r
r
e
vi
e
w
w
or
ds
a
r
e
e
xpr
e
s
s
e
d
;
ii
)
to
de
v
e
lo
p
a
ne
w
s
e
nt
im
e
nt
c
a
te
gor
y
c
a
ll
e
d
"
non
-
in
te
r
pr
e
ti
ve
s
e
nt
im
e
nt
"
ba
s
e
d
on
th
e
s
ubj
e
c
ti
ve
e
xpr
e
s
s
io
n
s
c
or
e
obt
a
in
e
d
f
r
om
r
e
vi
e
w
s
;
a
nd
ii
i)
by
m
e
r
gi
ng
L
e
xi
c
on
-
ba
s
e
d
a
s
w
e
ll
a
s
e
ns
e
m
bl
e
le
a
r
ni
ng
w
it
h
s
of
t
vot
in
g
f
or
s
e
nt
im
e
nt
c
a
te
gor
iz
a
ti
on,
pr
o
pos
e
a
uni
que
hybr
id
m
ode
l
f
or
s
e
nt
im
e
nt
pr
e
di
c
ti
on.
2.
R
E
L
A
T
E
D
WORK
N
ow
a
da
ys
,
s
e
nt
im
e
nt
e
va
lu
a
ti
on
is
in
c
r
e
a
s
in
gl
y
popula
r
w
it
h
bus
in
e
s
s
e
s
,
gove
r
nm
e
nt
s
,
a
nd
or
ga
ni
z
a
ti
ons
[
8]
.
ML
m
e
th
od
s
r
e
a
c
h
s
up
e
r
io
r
a
c
c
ur
a
c
y,
s
e
nt
i
m
e
nt
c
a
te
gor
iz
a
ti
on
[
9]
.
B
a
la
ji
e
t
al
.
[
10]
us
e
d
pow
e
r
f
ul
ML
te
c
hni
que
s
to
th
or
oughly
e
xa
m
in
e
th
e
va
r
i
ous
s
oc
ia
l
m
e
di
a
a
na
ly
s
is
a
nd
a
lg
or
it
hm
s
.
O
ha
na
a
nd
T
ie
r
ne
y
[
11]
e
va
lu
a
te
d
how
w
e
ll
th
e
s
e
nt
i
-
w
or
dne
t
opi
ni
on
le
xi
c
on
c
la
s
s
if
ie
d
th
e
s
e
nt
im
e
nt
in
m
ovi
e
r
e
vi
e
w
s
.
J
a
in
e
t
al
.
[
12]
de
s
c
r
ib
e
s
th
e
ge
n
e
r
a
ti
on
of
s
e
ve
r
a
l
c
la
s
s
if
ie
r
s
th
r
ough
tr
a
in
in
g
w
it
h
va
r
io
us
f
e
a
tu
r
e
s
e
ts
,
f
ol
lo
w
e
d
by
th
e
s
e
le
c
ti
on
a
nd
c
om
bi
na
ti
on
of
c
o
m
pone
nt
c
la
s
s
if
ie
r
s
us
in
g
a
num
be
r
of
pr
e
s
e
t
c
om
bi
na
ti
on
c
r
it
e
r
ia
.
A
us
e
f
ul
c
la
s
s
if
ic
a
ti
on
m
e
th
od
f
or
m
a
ny
dom
a
in
s
is
th
e
e
ns
e
m
bl
e
m
e
th
odol
ogy,
w
hi
c
h
in
te
gr
a
te
s
th
e
r
e
s
ul
ts
of
va
r
io
us
ba
s
e
c
la
s
s
if
ic
a
ti
on
m
ode
ls
to
pr
ovi
de
a
n
in
te
gr
a
te
d
ou
tp
ut
[
13
]
.
A
c
om
bi
na
ti
on
of
va
r
io
us
c
la
s
s
if
ic
a
ti
on
a
lg
or
it
hm
s
(
K
-
ne
a
r
e
s
t
ne
i
ghbor
(
K
N
N
)
a
nd
B
a
ye
s
ia
n
c
la
s
s
if
ie
r
s
)
yi
e
ld
s
s
upe
r
io
r
r
e
s
ul
ts
th
a
n
a
ny
s
in
gl
e
ty
pe
o
f
c
la
s
s
if
ie
r
in
th
e
e
a
r
ly
w
or
k
[
14]
.
T
ur
ki
s
h
S
A
te
c
hni
que
s
f
or
c
us
to
m
e
r
r
e
vi
e
w
s
of
hot
e
ls
w
e
r
e
de
ve
lo
pe
d
by
E
r
ş
a
hi
n
e
t
al
.
[
15]
us
in
g
a
c
om
bi
na
ti
on
of
s
e
nt
im
e
nt
di
c
ti
ona
r
ie
s
a
nd
M
L
a
lg
or
it
hm
s
.
C
oba
n
e
t
al
.
[
16]
r
e
c
e
nt
ly
di
d
a
S
A
s
tu
dy
us
in
g
th
e
in
nova
ti
ve
e
xt
r
e
m
e
le
a
r
ni
ng
m
a
c
hi
ne
(
E
L
M
)
t
o e
xt
r
a
c
t
opi
ni
ons
f
r
o
m
T
ur
ki
s
h t
w
e
e
ts
[
16]
.
T
o
in
c
r
e
a
s
e
th
e
pr
e
c
is
io
n
of
th
e
opi
ni
on
m
in
in
g
pr
oc
e
s
s
,
A
lf
r
ja
ni
e
t
al
.
[
17]
s
ugge
s
ts
a
hybr
id
s
e
m
a
nt
ic
knowle
dge
-
ba
s
e
d
ML
te
c
hni
que
.
B
hoi
r
a
nd
K
ol
te
[
1
8]
pr
e
s
e
nt
s
a
r
ul
e
-
ba
s
e
d
a
ppr
oa
c
h
e
m
pl
oyi
ng
two
te
c
hni
que
s
:
a
le
xi
c
on
a
ppl
ie
d
to
in
te
r
ne
t
m
ovi
e
d
a
ta
ba
s
e
(
I
M
D
b)
m
ovi
e
r
e
vi
e
w
s
a
nd
a
n
a
ïv
e
B
a
y
e
s
c
la
s
s
if
ie
r
on
ML
.
T
he
n
a
ïv
e
B
a
ye
s
c
la
s
s
if
ie
r
is
de
f
e
nde
d
in
th
e
s
tu
dy
[
19]
.
A
r
e
in
f
or
c
e
m
e
nt
le
a
r
ni
ng
te
c
hni
que
ba
s
is
of
doc
um
e
nt
le
ve
l
a
s
pe
c
t
-
ba
s
e
d
s
e
nt
im
e
nt
c
a
te
gor
iz
a
ti
on
(
D
A
S
C
)
,
H
a
s
a
n
e
t
al
.
[
20]
s
ugg
e
s
ts
a
hi
e
r
a
r
c
hi
c
a
l
r
e
in
f
or
c
e
m
e
nt
le
a
r
ni
ng
s
tr
a
te
g
y.
T
he
a
c
c
ur
a
c
y
of
th
e
e
ns
e
m
bl
e
le
a
r
ni
ng
m
ode
ls
[
21]
c
a
n
be
s
ig
ni
f
ic
a
nt
ly
im
pr
ove
d
by
us
in
g
th
e
r
ig
ht
e
ns
e
m
bl
e
le
a
r
ni
ng
a
lg
or
it
hm
[
22
]
,
[
23
]
.
F
or
th
e
pur
pos
e
o
f
c
la
s
s
if
yi
ng
twe
e
t
s
e
nt
im
e
nt
,
R
e
hi
oui
a
nd
I
dr
is
s
i
[
24]
s
ugg
e
s
ts
a
br
a
nd
-
ne
w
c
lu
s
te
r
in
g
te
c
hni
que
th
a
t
m
a
k
e
s
us
e
of
K
-
m
e
a
ns
a
nd
D
e
nc
lu
e
.
A
l
-
S
a
qq
a
e
t
al
.
[
25]
of
f
e
r
s
c
ut
ti
ng
-
e
dge
m
e
th
odol
ogy
us
in
g
a
gr
oup
of
c
la
s
s
if
ie
r
s
t
o a
s
c
e
r
ta
in
t
he
pol
a
r
it
y of
t
he
A
r
a
bi
c
t
e
xt
'
s
opi
ni
ons
.
B
e
h
e
r
a
e
t
a
l.
[
2
6]
t
a
l
k
s
o
n
nu
m
e
r
o
u
s
e
n
s
e
m
b
l
e
l
e
a
r
ni
n
g
t
e
c
h
n
iq
u
e
s
u
s
e
d
t
o
e
n
h
a
nc
e
a
m
o
d
e
l
'
s
p
e
r
f
or
m
a
n
c
e
i
n t
e
r
m
s
of
c
l
a
s
s
i
f
i
c
a
t
io
n
,
f
un
c
ti
on
a
p
pr
o
x
im
a
t
io
n
,
pr
e
di
c
ti
on
, a
nd p
e
r
f
or
m
a
n
c
e
.
I
n
o
r
d
e
r
t
o i
m
pr
o
v
e
t
h
e
a
c
c
u
r
a
c
y,
p
e
r
f
o
r
m
a
n
c
e
,
a
n
d
s
p
e
e
d
a
t
w
h
ic
h
th
e
a
lg
or
it
hm
i
s
e
xe
c
ut
e
d,
S
a
d
h
a
s
i
v
a
m
a
n
d
K
a
li
v
a
r
a
d
h
a
n
[
27]
s
ugge
s
ts
a
m
a
jo
r
it
y voti
ng
-
ba
s
e
d e
ns
e
m
bl
e
t
e
c
hni
que
t
ha
t
c
om
bi
ne
s
s
uppor
t
ve
c
to
r
m
a
c
hi
ne
(
S
V
M
)
[
28]
w
it
h
n
a
ïv
e
B
a
y
e
s
.
I
n
or
de
r
to
a
ddr
e
s
s
th
e
bi
na
r
y
c
l
a
s
s
if
ic
a
ti
on
pr
obl
e
m
ba
s
e
d
on
twe
e
ts
,
m
ovi
e
r
e
vi
e
w
s
,
a
nd
pr
oduc
t
r
e
vi
e
w
s
,
R
a
je
s
w
a
r
i
e
t
al
.
[
29]
pr
opos
e
d
a
hybr
id
s
tr
a
te
gy
e
m
pl
oyi
ng
le
xi
c
a
l
a
nd
M
L
m
e
th
od
s
.
T
o
a
s
c
e
r
ta
in
th
e
s
e
nt
im
e
nt
of
T
r
ip
A
dvi
s
or
us
e
r
r
e
vi
e
w
s
,
M
a
he
s
w
a
r
i
e
t
al
.
[
30
]
us
e
hom
oge
ne
ous
e
ns
e
m
bl
e
c
la
s
s
if
ie
r
s
in
c
lu
di
ng
ba
ggi
ng
de
c
i
s
io
n
tr
e
e
,
ba
gge
d
m
ul
ti
la
ye
r
pe
r
c
e
pt
r
ons
(
M
L
P
)
,
r
a
ndom
f
o
r
e
s
ts
(
R
F
)
,
a
nd
lo
gi
s
ti
c
m
ode
l
tr
e
e
s
.
T
he
r
e
s
e
a
r
c
h
c
la
im
s
th
a
t
th
e
hybr
id
te
c
hni
que
is
m
or
e
e
f
f
ic
ie
nt
a
nd
g
e
ts
ove
r
th
e
dr
a
w
ba
c
ks
of
e
a
c
h
of
th
e
or
ig
in
a
l
le
xi
c
on
m
e
th
od
a
nd
e
ns
e
m
bl
e
m
e
th
od
-
ba
s
e
d
m
e
th
ods
.
A
hybr
id
bi
di
r
e
c
ti
ona
l
lo
ng
s
hor
t
-
te
r
m
m
e
m
or
y
(
Bi
-
L
S
T
M
)
-
a
r
ti
f
ic
ia
l
ne
ur
a
l
ne
twor
k
(
ANN
)
m
ode
l
b
a
s
e
d
on
bot
h
te
m
por
a
l
a
nd
hype
r
nym
c
ha
r
a
c
te
r
is
ti
c
s
is
im
pl
e
m
e
nt
e
d
by
S
r
id
ha
r
a
nd
S
a
na
ga
va
r
a
pu
[
31]
.
T
he
pr
e
vi
ous
r
e
s
e
a
r
c
h
e
m
pl
oys
a
hybr
id
a
ppr
oa
c
h
to
SA
,
c
om
bi
n
in
g
th
e
s
e
nt
im
e
nt
le
xi
c
on
w
it
h
ML
a
lg
or
it
hm
s
li
ke
S
V
M
[
32]
a
nd
K
N
N
,
a
nd
f
in
ds
th
a
t
S
V
M
e
xe
c
ut
e
s
b
e
tt
e
r
on
ne
w
s
c
om
m
e
nt
s
th
a
n
K
N
N
[
33]
,
[
34]
.
I
n
th
is
w
or
k,
a
hybr
id
m
ode
l
ba
s
e
d
on
e
ns
e
m
bl
e
a
nd
le
xi
c
on
m
ode
l
s
u
s
in
g
s
of
t
vot
in
g
m
e
th
od
is
de
ve
lo
pe
d
f
or
s
e
nt
im
e
nt
c
la
s
s
if
ic
a
ti
on
tr
a
in
e
d
on
A
m
a
z
on
K
in
dl
e
a
nd
E
le
c
tr
oni
c
s
d
a
t
a
s
e
t.
C
lo
ud
c
om
put
in
g
a
ll
ow
s
th
e
pr
opo
s
e
d
s
e
iz
ur
e
pr
e
di
c
ti
on
m
e
c
ha
ni
s
m
i
s
m
or
e
a
c
c
e
s
s
ib
le
a
nd
s
c
a
la
bl
e
[
35]
.
S
V
M
’
s
c
la
s
s
if
ic
a
ti
on
a
lg
or
it
hm
to
c
a
te
gor
iz
e
s
p
a
ti
e
nt
s
’
r
is
ks
a
nd
f
or
e
c
a
s
t
th
e
r
a
py. R
e
c
ur
r
e
nt
ne
ur
a
l
ne
twor
ks
a
lg
or
it
hm
to
r
e
c
ogoniz
e
s
tr
e
nds
in
pa
ti
e
nt
s
ym
pt
om
s
a
nd
dr
ug
[
36]
.
C
lo
ud
-
ba
s
e
d
d
e
c
is
io
n
-
s
uppo
r
t
s
ys
te
m
[
37]
a
ppl
yi
ng
KNN
a
lg
or
it
hm
w
it
h
I
oT
s
tr
uc
tu
r
e
s
ys
te
m
i
nt
e
gr
a
te
s
I
oT
de
vi
c
e
s
t
o c
ol
le
c
t
a
nd t
r
a
ns
m
it
t
he
da
ta
[
38]
.
3.
P
R
O
P
O
S
E
D
M
E
T
H
O
D
T
hi
s
pa
pe
r
pr
opos
e
s
a
hybr
id
m
ode
l
ba
s
e
d
s
of
t
vot
in
g
m
ode
l
by
c
om
bi
ni
ng
le
xi
c
on
a
ppr
oa
c
h
a
nd
a
n
e
ns
e
m
bl
in
g
te
c
hni
que
a
s
s
how
n
in
F
ig
ur
e
2.
T
he
le
xi
c
on
a
nd
ML
to
c
r
e
a
te
pr
e
di
c
ti
ons
by
w
e
ig
ht
e
d
a
ve
r
a
ge
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
. 14, No. 3, J
une
2025
:
2001
-
2011
2004
T
he
ope
n
-
s
our
c
e
li
br
a
r
y
na
m
e
d
N
um
P
y,
S
C
I
K
I
T
le
a
r
n
[
39
]
s
e
nt
i
-
w
or
dne
t,
a
nd
na
tu
r
a
l
la
ngua
ge
to
ol
ki
t
(
N
L
T
K
)
a
r
e
a
ls
o i
ns
ta
ll
e
d t
o pe
r
f
or
m
t
hi
s
s
im
ul
a
ti
on.
F
ig
ur
e
2
. H
ybr
id
e
ns
e
m
bl
e
m
e
th
od a
r
c
hi
te
c
tu
r
e
3.1. Dat
as
e
t
s
T
hi
s
w
or
k
is
done
by
c
ons
um
e
r
e
le
c
tr
oni
c
pr
oduc
t
r
e
vi
e
w
(
C
E
P
R
)
f
r
om
K
a
ggl
e
w
e
bs
it
e
.
T
he
C
E
P
R
da
ta
s
e
t
c
on
s
is
t
of
th
e
li
s
t
of
34
,
661
A
m
a
z
on
E
le
c
tr
oni
c
c
ons
u
m
e
r
pr
oduc
ts
r
e
vi
e
w
s
s
pe
c
if
ic
a
ll
y
K
in
dl
e
a
nd
F
ir
e
T
V
i
n .c
s
v f
o
r
m
a
t.
I
t
a
dm
it
s
t
he
pr
oduc
t
de
ta
il
s
, c
om
m
e
nt
s
t
e
xt
r
e
vi
e
w
,
a
nd
r
a
te
. L
ik
e
w
is
e
, t
he
e
le
c
tr
oni
c
pr
oduc
ts
a
nd
G
if
t
C
a
r
ds
da
ta
s
e
ts
i
nc
lu
de
2
,
972
a
s
w
e
ll
a
s
2
,
375
r
e
vi
e
w
s
c
or
r
e
s
pondingl
y.
3.2. P
r
op
os
e
d
m
od
e
l
S
e
nt
i
-
w
or
dne
t
a
nd
e
ns
e
m
bl
e
-
ba
s
e
d
M
L
te
c
hni
qu
e
s
a
r
e
us
e
d
to
c
r
e
a
te
a
s
ugge
s
te
d
hybr
id
e
ns
e
m
bl
e
m
ode
l.
T
he
in
tr
oduc
e
d
m
e
th
od
r
e
a
c
he
s
im
pr
ove
d
f
unc
ti
o
ni
ng
a
nd
in
c
r
e
a
s
e
d
pr
e
c
is
io
n
a
r
e
ge
ne
r
a
ll
y
e
xpl
ic
a
te
d by the
f
ol
lo
w
in
g t
w
o s
ta
ge
s
:
3.
2
.
1.
L
e
xi
c
on
-
b
as
e
d
t
e
c
h
n
iq
u
e
T
hi
s
s
t
a
ge
in
vol
ve
s
pr
e
-
pr
oc
e
s
s
in
g
th
e
pr
ovi
de
d
da
t
a
s
e
t
s
to
r
e
m
ove
th
e
ke
yw
or
ds
ut
il
iz
in
g
to
ke
ni
z
a
ti
on
a
s
w
e
ll
a
s
le
m
m
a
ti
z
a
ti
on
f
or
ke
yw
or
d
r
e
c
ogni
t
io
n
a
nd
r
e
m
ova
l
of
e
xt
r
a
ne
ous
s
to
p
w
or
ds
,
de
te
r
m
in
a
nt
s
,
a
nd
pr
e
pos
it
io
n
s
.
U
s
in
g
th
e
N
L
T
K
p
a
c
ka
ge
,
w
hi
c
h
is
ba
s
e
d
on
N
L
P
,
th
e
pa
r
t
-
of
-
s
pe
e
c
h
(
P
O
S
)
ta
ggi
ng
is
c
a
lc
ul
a
te
d.
F
in
a
ll
y,
us
in
g
th
e
s
e
m
a
nt
ic
in
f
or
m
a
ti
on
gl
e
a
ne
d
f
r
om
th
e
s
e
nt
i
-
w
or
dne
t
di
c
ti
ona
r
y,
th
e
s
e
nt
im
e
nt
(
S
S
)
a
nd
s
ubj
e
c
ti
vi
ty
(
S
T
)
s
c
or
e
is
be
in
g
c
a
lc
ul
a
te
d.
T
he
c
om
put
a
ti
ona
l
m
e
th
ods
r
e
qui
r
e
d
to
c
a
lc
ul
a
te
th
e
e
nt
ir
e
S
S
s
c
or
e
a
s
w
e
ll
a
s
S
T
s
c
or
e
,
c
or
r
e
s
po
ndi
ngl
y,
by
th
e
c
ont
e
nt
of
A
m
a
z
on
r
e
vi
e
w
c
om
m
e
nt
s
a
r
e
d
e
s
c
r
ib
e
d
in
(
1
)
.
T
he
in
it
ia
l
s
pl
it
of
th
e
da
ta
s
e
t
is
90:
10
to
th
e
tr
a
in
a
nd
80:
20
f
or
th
e
te
s
t
da
ta
s
e
t.
T
h
e
s
e
nt
i
-
w
or
dne
t
di
c
ti
ona
r
y'
s
in
te
r
na
l
ke
yw
or
ds
'
s
e
m
a
nt
ic
pr
ope
r
ti
e
s
a
r
e
us
e
d
to
f
or
e
c
a
s
t
e
a
c
h
pe
r
s
on'
s
s
e
nt
im
e
nt
s
c
or
e
.
I
n
or
de
r
to
de
te
r
m
in
e
th
e
pol
a
r
it
y
of
opi
ni
ons
,
s
uc
h
tr
a
in
da
ta
s
e
t
is
ut
il
iz
e
d
to
c
a
lc
ul
a
te
s
e
nt
im
e
nt
a
s
w
e
ll
a
s
S
T
s
c
or
e
s
a
s
w
e
ll
a
s
to
id
e
nt
if
y
s
ubj
e
c
ti
ve
s
ta
te
m
e
nt
s
th
a
t
ha
v
e
a
s
ig
ni
f
ic
a
nt
im
pa
c
t
on
th
e
S
S
s
c
or
e
a
s
a
w
hol
e
.
T
o
c
a
lc
ul
a
te
th
e
S
S
s
c
or
e
a
nd
e
s
ta
bl
is
h
th
e
S
S
pol
a
r
i
ty
of
th
e
s
e
r
e
vi
e
w
c
onde
m
na
ti
ons
,
th
e
tr
a
in
in
g
da
ta
i
s
e
m
pl
oye
d.
In
(
1)
is
u
s
e
d
to
c
a
lc
ul
a
t
e
e
a
c
h
r
e
vi
e
w
s
e
nt
e
n
c
e
'
s
ov
e
r
a
ll
s
e
nt
im
e
nt
s
c
or
e
(
S
S
)
.
=
√
∑
(
p
)
3
−
∑
(
n
)
3
n
w
=
1
n
w
=
1
n
3
(
1)
W
he
r
e
is
th
e
pos
it
iv
e
S
S
s
c
or
e
a
s
c
om
put
e
d
f
r
om
th
e
s
e
nt
i
-
w
or
dne
t
di
c
ti
ona
r
y;
is
th
e
ne
ga
ti
ve
S
S
s
c
or
e
a
s
c
om
put
e
d
f
r
om
th
e
s
e
nt
i
-
w
or
dne
t
di
c
ti
ona
r
y.
T
he
ove
r
a
ll
S
T
pe
r
r
e
vi
e
w
s
e
nt
e
nc
e
c
a
n
be
c
om
put
e
d
us
in
g i
n (
2)
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
H
y
br
id
s
e
m
ant
ic
m
ode
l
bas
e
d on
m
ac
hi
ne
l
e
ar
ni
ng f
o
r
s
e
nt
ime
nt
…
(
P
al
ani
r
aj
R
aj
id
ur
ai
P
ar
v
at
hy
)
2005
=
∑
1
−
(
_
)
=
1
(
2)
S
im
il
a
r
to
th
e
obj
e
c
ti
ve
s
c
or
e
,
s
e
nt
i
-
w
or
dne
t
di
c
ti
ona
r
y'
s
s
e
m
a
nt
ic
pr
ope
r
ti
e
s
a
r
e
us
e
d
to
c
a
lc
ul
a
te
th
e
s
ubj
e
c
ti
ve
s
c
or
e
ba
s
e
d
on
th
e
obj
e
c
ti
ve
s
c
or
e
.
I
n
or
de
r
to
de
te
r
m
in
e
if
th
e
S
S
c
onne
c
te
d
to
e
a
c
h
r
e
vi
e
w
s
e
nt
e
nc
e
a
r
e
good,
ne
ut
r
a
l,
or
ne
ga
ti
ve
,
th
e
in
di
vi
dua
l
s
c
or
e
s
a
r
e
c
om
bi
ne
d
to
pr
ovi
de
a
n
ove
r
a
ll
s
e
nt
e
nc
e
S
S
s
c
or
e
.
3.
2
.
2.
E
n
s
e
m
b
le
l
e
a
r
n
in
g t
e
c
h
n
iq
u
e
I
n t
hi
s
m
odul
e
, i
t
a
ppl
ie
s
a
n e
ns
e
m
bl
e
l
e
a
r
ni
ng t
o buil
d a
S
S
pr
e
di
c
ti
ons
. T
hi
s
m
e
th
od de
c
id
e
s
t
he
S
S
f
or
e
c
a
s
ti
ng
on
c
om
m
e
nt
s
e
m
pl
oyi
ng
th
e
A
m
a
z
on
r
e
vi
e
w
s
da
ta
s
e
t
of
c
u
s
to
m
e
r
e
le
c
tr
oni
c
pr
oduc
ts
vi
a
c
a
lc
ul
a
ti
ng
th
e
w
e
ig
ht
e
d
a
v
e
r
a
ge
vot
in
g
of
in
di
vi
dua
l
c
la
s
s
if
ie
r
.
I
n
th
e
in
tr
oduc
e
d
w
e
ig
ht
s
a
r
e
a
ll
ot
te
d
to
r
e
le
va
nt
f
iv
e
c
la
s
s
if
ie
r
s
a
nd t
he
w
e
ig
ht
e
d a
ve
r
a
ge
of
po
s
s
ib
il
it
y voti
ng i
s
be
in
g c
a
lc
ul
a
te
d a
s
i
n (
3)
.
=
w
1
pc
1
+
w
2
pc
2
+
w
3
pc
3
w1
+
w2
+
w3
(
3)
W
he
r
e
,
pc
1
r
e
pr
e
s
e
nt
s
th
e
pr
e
di
c
te
d
pos
s
ib
il
it
y
de
c
id
e
d
by
S
V
M
c
la
s
s
if
ie
r
(
s
upe
r
vi
s
e
d
le
a
r
ni
ng)
,
pc
2
r
e
pr
e
s
e
nt
s
th
e
pr
e
di
c
te
d
po
s
s
ib
il
it
y
de
c
id
e
d
by
lo
gi
s
ti
c
r
e
g
r
e
s
s
io
n
(
L
R
)
,
c
la
s
s
if
ie
r
(
s
upe
r
vi
s
e
d
le
a
r
ni
ng)
,
pc
3
is
pr
e
di
c
te
d
pos
s
ib
il
it
y
de
c
id
e
d
by
K
N
N
c
la
s
s
if
ie
r
(
uns
upe
r
vi
s
e
d
le
a
r
ni
ng)
,
w
1
in
di
c
a
te
s
th
e
w
e
ig
ht
a
ll
ot
te
d
f
or
L
R
c
la
s
s
if
ie
r
,
w
2
de
not
e
s
th
e
w
e
ig
ht
a
ll
ot
te
d
f
or
S
V
M
c
la
s
s
if
ie
r
,
w
3
in
di
c
a
te
s
th
e
w
e
ig
ht
a
ll
ot
te
d
f
or
K
N
N
c
la
s
s
if
ie
r
.
4.
R
E
S
U
L
T
S
A
N
D
D
I
S
C
U
S
S
I
O
N
T
he
pr
opos
e
d
m
e
c
ha
ni
s
m
i
s
e
m
pl
oye
d
to
c
a
lc
ul
a
te
S
S
s
tu
d
y
f
or
A
m
a
z
on’
s
c
us
to
m
e
r
pr
oduc
ts
a
ppr
oxi
m
a
ti
ng
G
if
t
C
a
r
ds
,
M
a
ga
z
in
e
s
,
F
ir
e
T
V
,
a
nd
K
in
dl
e
.
T
he
S
T
a
nd
S
S
s
c
or
e
s
a
r
e
c
a
l
c
ul
a
te
d
in
th
e
pr
opos
e
d
m
e
th
od
a
nd
th
e
r
e
s
ul
t
r
e
c
e
iv
e
d
e
qu
a
te
d
w
it
h
T
e
x
tB
lo
b
li
br
a
r
y
th
a
t
a
ls
o
of
f
e
r
s
na
ti
ve
A
P
I
to
c
om
put
e
th
e
S
S
a
nd
S
T
s
c
or
e
.
T
e
xt
B
lo
b
is
a
c
c
e
pt
e
d
pyt
hon
li
b
r
a
r
y
in
te
nde
d
on
le
xi
c
on
-
ba
s
e
d
m
e
th
od
a
nd
is
ha
bi
tu
a
ll
y
a
ppl
ie
d
f
or
te
xt
m
in
in
g
a
nd
tr
e
a
ti
ng
of
th
e
te
xt
ua
l
da
ta
.
T
h
e
e
xpe
r
im
e
nt
i
s
c
a
r
r
ie
d
out
in
G
oogl
e
C
ol
a
b
P
r
o+
w
hi
c
h
is
c
om
pl
e
te
ly
c
lo
ud
-
ba
s
e
d
pl
a
tf
or
m
w
it
h
N
V
I
D
I
A
T
e
s
la
T
4
G
P
U
s
a
nd
8
G
B
of
hi
gh
-
R
A
M
.
T
he
c
om
p
a
r
is
on
of
S
S
a
nd
S
T
s
c
or
e
s
a
r
e
s
how
n
in
T
a
bl
e
2.
T
he
pr
opos
e
d
m
e
c
ha
ni
s
m
e
xe
c
ut
e
s
th
e
e
ns
e
m
bl
e
l
e
a
r
ni
ng t
hr
ough c
om
pounding t
hr
e
e
c
la
s
s
if
ie
r
s
e
xpl
i
c
it
ly
L
R
, S
V
M
,
a
nd
K
N
N
c
or
r
e
s
pondingl
y.
T
a
bl
e
2
. C
om
pa
r
is
on of
S
S
/S
T
s
c
or
e
c
a
lc
ul
a
te
d ut
il
iz
in
g pr
opos
e
d e
ns
e
m
bl
e
m
ode
l
a
nd
T
e
xt
B
lo
b
R
e
vi
e
w
c
om
m
e
nt
s
S
S
s
c
or
e
S
T
s
c
or
e
S
S
s
c
or
e
T
e
xt
B
l
ob
S
T
s
c
or
e
T
e
xt
B
l
ob
G
r
e
a
t
pr
oduc
t
f
or
s
uc
h pr
i
c
e
0.3012
0.6531
0.6
0.4475
I
s
t
r
ongl
y s
ugge
s
t
t
o ot
he
r
s
0.4472
0.7343
0.7
0.81
T
he
t
a
b qui
t
t
e
d f
unc
t
i
oni
ng a
f
t
e
r
f
unc
t
i
on
i
ng f
or
t
w
o w
e
e
ks
.
0.0521
0.452
0.1853
0.3317
A
bs
ol
ut
e
w
a
s
t
e
of
m
one
y. I
t
di
d not
j
oi
n t
o W
i
-
F
i
a
t
m
y hom
e
. J
unk!
-
0.0026
0.5124
0.1853
0.3802
T
he
w
e
ig
ht
s
a
r
e
c
ur
r
e
nt
ly
a
ll
ot
te
d
to
th
e
c
la
s
s
if
ie
r
s
de
r
iv
e
d
f
r
o
m
how
w
e
ll
e
a
c
h
in
di
vi
dua
l
c
la
s
s
if
ie
r
pe
r
f
or
m
e
d,
w
it
h
th
e
be
s
t
pe
r
f
or
m
in
g
c
la
s
s
if
ie
r
r
e
c
e
iv
in
g
m
or
e
w
e
ig
ht
s
.
T
he
tr
a
in
in
g
da
ta
s
e
t'
s
r
e
vi
e
w
c
om
m
e
nt
s
s
e
r
ve
a
s
th
e
ba
s
i
s
f
or
c
a
lc
ul
a
ti
ng
opi
ni
on
pol
a
r
it
y,
w
hi
c
h
is
th
e
n
us
e
d
to
tr
a
in
th
e
s
ugge
s
te
d
m
od
e
l.
S
e
ve
r
a
l
c
ur
r
e
nt
c
om
m
on
e
xe
c
ut
io
n
m
e
a
s
ur
e
s
to
e
va
lu
a
te
th
e
in
f
or
m
a
ti
on
r
e
c
ove
r
y
m
e
th
od
in
c
lu
de
pr
e
c
is
io
n,
r
e
c
a
ll
, a
nd F
1
-
s
c
or
e
. T
he
c
om
put
a
ti
on f
or
m
ul
a
ti
ons
f
or
t
he
s
e
e
va
lu
a
ti
on i
ndi
c
e
s
a
r
e
l
is
te
d i
n (
3)
t
o (
7)
.
(
)
=
+
+
+
+
(
4)
(
)
=
+
(
5)
(
)
=
+
(
6)
1
−
=
2
×
×
+
(
7)
W
he
r
e
T
P
is
tr
ue
pos
it
iv
e
s
, T
N
i
s
tr
ue
ne
ga
ti
ve
s
, F
P
is
f
a
ls
e
pos
it
iv
e
s
,
a
nd F
N
is
f
a
ls
e
n
e
ga
ti
ve
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
. 14, No. 3, J
une
2025
:
2001
-
2011
2006
T
a
bl
e
3
s
how
s
th
a
t
th
e
F
1
-
s
c
or
e
of
pr
opos
e
d
m
ode
l
s
,
w
hi
c
h
is
ba
s
e
d
on
s
of
t
vot
in
g,
out
pe
r
f
or
m
s
th
e
c
ur
r
e
nt
s
ta
te
-
of
-
th
e
-
a
r
t
e
ns
e
m
bl
e
m
ode
l.
T
he
"
non
-
in
te
r
pr
e
ta
ti
v
e
s
e
nt
im
e
nt
"
is
us
e
d
in
th
e
pr
opos
e
d
a
ppr
oa
c
h
to
c
a
te
gor
iz
e
r
e
vi
e
w
c
om
m
e
nt
s
th
a
t
la
c
k
a
ny
s
ub
s
ta
nt
iv
e
opi
ni
o
ns
.
F
or
our
s
ugge
s
te
d
m
ode
l
to
pe
r
f
or
m
be
tt
e
r
a
nd
a
tt
a
in
gr
e
a
te
r
a
c
c
ur
a
c
y
in
s
e
nt
im
e
nt
c
la
s
s
if
ic
a
ti
on,
a
la
r
ge
num
be
r
of
da
ta
s
e
t
s
m
u
s
t
be
us
e
d
f
or
tr
a
in
in
g.
I
n
or
de
r
to
gi
ve
th
e
s
lu
ggi
s
h
a
nd
und
e
r
a
c
hi
e
vi
ng
le
a
r
ne
r
s
le
s
s
w
e
ig
ht
,
w
e
m
us
t
u
s
e
e
ns
e
m
bl
e
le
a
r
ni
ng
w
it
h
w
e
ig
ht
e
d
vot
in
g.
T
he
c
la
s
s
la
be
l
w
it
h
th
e
hi
ghe
s
t
a
ve
r
a
ge
pos
s
ib
il
it
ie
s
f
r
om
th
e
gr
oup
o
f
c
la
s
s
if
ie
r
s
is
r
e
tu
r
ne
d by the
w
e
ig
ht
e
d a
ve
r
a
ge
.
T
a
bl
e
3
. R
e
la
ti
ve
a
na
ly
s
i
s
us
e
pr
e
c
is
io
n, r
e
c
a
ll
, a
nd F
1
-
s
c
or
e
f
o
r
pr
opos
e
d m
ode
l
D
a
t
a
s
e
t
A
l
gor
i
t
hm
P
r
e
c
i
s
i
on
R
e
c
a
l
F1
-
s
c
or
e
K
i
ndl
e
B
a
ggi
ng e
ns
e
m
bl
e
0.63
0.51
0.5636
B
oos
t
i
ng e
ns
e
m
bl
e
0.65
0.52
0.5777
P
r
opos
e
d e
ns
e
m
bl
e
m
ode
l
0.81
0.52
0.6333
G
i
f
t
-
C
a
r
ds
B
a
ggi
ng
e
ns
e
m
bl
e
0.87
0.67
0.7570
B
oos
t
i
ng e
ns
e
m
bl
e
0.74
0.59
0.6565
P
r
opos
e
d e
ns
e
m
bl
e
m
ode
l
0.98
0.61
0.7519
E
l
e
c
t
r
oni
c
pr
oduc
t
s
B
a
ggi
ng e
ns
e
m
bl
e
0.76
0.62
0.6828
B
oos
t
i
ng e
ns
e
m
bl
e
0.79
0.61
0.6884
P
r
opos
e
d e
ns
e
m
bl
e
m
ode
l
0.81
0.61
0.6959
4
.1.
P
e
r
f
or
m
an
c
e
e
val
u
at
io
n
T
he
a
c
c
ur
a
c
y
s
c
or
e
s
how
s
a
s
li
ght
in
c
r
e
a
s
e
th
r
oughout
th
e
e
va
lu
a
ti
on
of
th
e
s
ugge
s
te
d
m
ode
l'
s
pe
r
f
or
m
a
nc
e
in
c
om
pa
r
is
on
to
th
e
c
ur
r
e
nt
e
ns
e
m
bl
e
-
ba
s
e
d
m
od
e
.
T
he
s
ugge
s
t
e
d
e
ns
e
m
bl
e
m
ode
l
ha
s
th
e
be
s
t
a
c
c
ur
a
c
y
s
c
or
e
of
m
or
e
th
a
n
80%
a
c
r
os
s
di
ve
r
s
e
da
ta
s
e
ts
,
a
s
s
how
n
in
th
e
F
ig
ur
e
3.
T
he
c
ur
r
e
nt
ly
e
m
pl
oye
d
e
ns
e
m
bl
e
c
la
s
s
if
ie
r
i
s
e
s
ta
bl
is
he
d on ma
jo
r
it
y voti
ng, a
nd i
t
ty
pi
c
a
ll
y f
or
e
c
a
s
ts
t
a
r
ge
t
c
la
s
s
l
a
be
l
s
ba
s
e
d on the
m
ode
of
a
nt
ic
ip
a
te
d i
ndi
vi
dua
l
c
la
s
s
l
a
be
ls
.
F
ig
ur
e
3. A
c
c
ur
a
c
y of
pr
opos
e
d e
ns
e
m
bl
e
m
e
th
od f
or
S
A
t
r
a
in
e
d on Ama
z
on K
in
dl
e
da
ta
s
e
t
A
s
il
lu
s
tr
a
te
d
in
F
ig
ur
e
4,
th
e
pr
opos
e
d
hybr
id
e
ns
e
m
bl
e
m
e
th
od
is
a
s
s
e
s
s
e
d
ut
il
iz
in
g
th
e
r
e
c
e
iv
in
g
ope
r
a
ti
ng
c
ha
r
a
c
te
r
is
ti
c
(
R
O
C
)
c
ur
ve
on
m
a
ny
c
la
s
s
l
a
be
ls
,
in
c
lu
di
ng
pos
it
iv
e
,
ne
ut
r
a
l,
a
nd
ne
ga
ti
ve
S
S
a
s
w
e
ll
a
s
non
-
in
te
r
pr
e
ti
ve
s
e
nt
im
e
nt
.
A
pos
s
ib
il
it
y
-
ba
s
e
d
c
ur
v
e
a
m
ong
tr
ue
pos
it
iv
e
r
a
te
(
T
P
R
)
a
nd
f
a
ls
e
pos
it
iv
e
r
a
te
(
F
P
R
)
a
t
d
if
f
e
r
e
nt
th
r
e
s
hol
ds
is
known
a
s
th
e
R
O
C
c
ur
ve
.
T
he
c
la
s
s
if
ic
a
ti
on
s
ys
te
m
'
s
e
f
f
ic
a
c
y
is
e
xpl
a
in
e
d by the
s
ha
r
pe
r
s
lo
pe
b
e
twe
e
n
T
P
R
a
nd
FPR
.
A
c
c
or
di
ng
to
F
ig
ur
e
4,
th
e
a
r
e
a
unde
r
th
e
c
ur
ve
(
A
U
C
)
va
lu
e
s
f
or
pos
it
iv
e
,
ne
ut
r
a
l,
ne
ga
ti
ve
,
a
nd
non
-
in
te
r
pr
e
ti
ve
s
e
nt
im
e
nt
f
or
th
e
K
in
dl
e
da
ta
s
e
t
r
e
s
pe
c
ti
ve
ly
.
T
he
T
P
R
a
nd
FPR
va
lu
e
s
be
twe
e
n
0.0
to
1.0.
B
e
tt
e
r
pe
r
f
or
m
a
nc
e
out
c
om
e
s
c
a
n
be
e
xpl
a
in
e
d
by
A
U
C
'
s
hi
ghe
r
va
lu
e
.
C
om
pa
r
e
to
a
ll
c
ur
ve
m
ic
r
o
a
ve
r
a
ge
R
O
C
c
ur
ve
ha
s
0.92.
A
c
c
or
di
ng
to
F
ig
ur
e
5,
th
e
A
U
C
va
lu
e
s
f
or
pos
it
iv
e
,
ne
ut
r
a
l,
ne
ga
ti
ve
,
a
nd
non
-
in
te
r
p
r
e
ti
ve
s
e
nt
im
e
nt
f
or
t
he
G
if
t
C
a
r
ds
da
ta
s
e
t
a
r
e
0.92, 0.89, 0.89, a
nd 0.
91
, r
e
s
pe
c
ti
ve
ly
. B
e
tt
e
r
pe
r
f
or
m
a
nc
e
out
c
om
e
s
c
a
n be
e
xpl
a
in
e
d by AU
C
'
s
hi
ghe
r
va
lu
e
.
T
he
m
ic
r
o a
ve
r
a
ge
R
O
C
c
ur
ve
ha
s
0.97 v
a
lu
e
t
ha
n a
ll
ot
he
r
c
ur
ve
s
.
0
20
40
60
80
100
E
l
e
c
t
r
oni
c
P
r
oduc
t
s
K
i
ndl
e
G
i
f
t
c
a
r
ds
A
c
c
u
r
ac
y
i
n
(
%
)
Dat
a
s
e
t
s
A
c
c
u
r
ac
y
c
om
p
ar
i
s
on
B
oos
t
i
ng e
ns
e
m
bl
e
B
a
ggi
ng e
ns
e
m
bl
e
P
r
opos
e
d m
ode
l
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
H
y
br
id
s
e
m
ant
ic
m
ode
l
bas
e
d on
m
ac
hi
ne
l
e
ar
ni
ng f
o
r
s
e
nt
ime
nt
…
(
P
al
ani
r
aj
R
aj
id
ur
ai
P
ar
v
at
hy
)
2007
F
ig
ur
e
4. R
O
C
c
ur
ve
of
pr
opos
e
d e
ns
e
m
bl
in
g
a
ppr
oa
c
h us
in
g K
in
dl
e
da
ta
s
e
t
a
nd
G
if
t
C
a
r
d
da
ta
s
e
t
F
ig
ur
e
5. R
O
C
c
ur
ve
of
pr
opos
e
d e
ns
e
m
bl
e
a
ppr
oa
c
h us
in
g
G
if
t
C
a
r
d
da
ta
s
e
t
5.
C
O
N
C
L
U
S
I
O
N
T
he
pur
pos
e
of
th
is
s
tu
dy
is
c
a
ll
e
d
a
hybr
id
e
ns
e
m
bl
e
th
a
t
c
o
m
bi
ne
s
le
xi
c
on
a
nd
e
ns
e
m
bl
e
le
a
r
ni
ng
w
it
h
th
e
us
e
of
s
of
t
vot
in
g
a
nd
a
n
a
ppr
opr
ia
te
di
s
tr
ib
ut
io
n
of
w
e
ig
ht
s
to
th
e
c
la
s
s
if
ie
r
s
.
T
he
f
e
a
tu
r
e
s
a
r
e
th
e
n
r
e
m
ove
d
f
r
om
th
e
r
e
vi
e
w
te
xt
ut
il
iz
in
g
te
r
m
f
r
e
que
nc
y
-
in
ve
r
s
e
doc
um
e
nt
f
r
e
que
nc
y
ve
c
to
r
iz
e
r
w
or
d
e
m
be
ddi
ngs
ut
il
iz
in
g
a
m
ix
tu
r
e
of
uni
gr
a
m
a
s
w
e
ll
a
s
bi
gr
a
m
te
xt
f
e
a
tu
r
e
s
,
f
ol
lo
w
in
g
th
e
in
it
ia
l
pr
e
-
pr
oc
e
s
s
in
g.
T
he
s
ubj
e
c
ti
vi
ty
s
c
or
e
,
w
hi
c
h
ha
s
be
e
n
ut
il
i
z
e
d
a
s
a
n
a
ddi
ti
ona
l
m
e
a
s
ur
e
to
pr
e
di
c
t
th
e
s
ubj
e
c
ti
ve
e
xpr
e
s
s
io
ns
to
id
e
nt
if
y
opi
ni
ona
te
d
m
a
te
r
ia
l,
ha
s
be
e
n
c
om
put
e
d
us
in
g
a
nove
l
a
ppr
oa
c
h
th
a
t
ha
s
be
e
n
pr
opos
e
d
in
th
is
r
e
s
e
a
r
c
h.
A
s
e
pa
r
a
te
c
la
s
s
la
be
l
c
a
ll
e
d
"
non
-
in
te
r
pr
e
ti
ve
s
e
nt
im
e
nt
"
is
a
ppl
ie
d
to
r
e
vi
e
w
c
om
m
e
nt
s
th
a
t
la
c
k
s
ubs
ta
nt
iv
e
opi
ni
ons
ba
s
e
d
on
th
e
s
ubj
e
c
ti
vi
ty
s
c
or
e
.
T
h
e
ke
y
di
f
f
ic
ul
ty
in
th
e
pr
opos
e
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
. 14, No. 3, J
une
2025
:
2001
-
2011
2008
hybr
id
e
ns
e
m
bl
e
m
ode
l
is
f
ig
ur
in
g
out
a
n
e
f
f
e
c
ti
ve
m
e
th
od
f
or
gi
vi
ng
th
e
be
s
t
-
pe
r
f
or
m
in
g
c
la
s
s
if
ie
r
th
e
r
ig
ht
w
e
ig
ht
s
dur
in
g
th
e
e
ns
e
m
bl
e
le
a
r
ni
ng
pr
oc
e
s
s
.
W
e
a
ls
o
poi
nt
out
th
a
t
s
ynt
a
c
ti
c
r
e
la
ti
ons
a
r
e
im
por
ta
n
t
c
ha
r
a
c
te
r
is
ti
c
s
f
or
s
e
nt
im
e
nt
c
a
te
gor
iz
a
ti
on,
but
th
e
y
a
ls
o
pr
e
s
e
nt
a
hi
gh
c
om
put
a
ti
ona
l
c
om
pl
e
xi
ty
is
s
u
e
.
F
ut
ur
e
r
e
s
e
a
r
c
h
w
il
l
b
r
oa
de
n
it
s
f
oc
us
a
s
w
e
s
e
e
k
to
a
dva
nc
e
e
nha
nc
e
th
e
in
te
nd
of
th
e
e
xi
s
ti
ng
m
e
th
od
a
n
d
de
ve
lo
p de
e
p ne
ur
a
l
ne
twor
ks
f
or
ha
ndl
in
g s
pa
m
, dupli
c
a
t
e
c
on
te
nt
, a
nd f
r
a
udul
e
nt
r
e
vi
e
w
s
.
F
U
N
D
I
N
G
I
N
F
O
R
M
A
T
I
O
N
F
undi
ng i
nf
or
m
a
ti
on i
s
not
a
va
il
a
bl
e
.
A
U
T
H
O
R
C
O
N
T
R
I
B
U
T
I
O
N
S
S
T
A
T
E
M
E
N
T
T
hi
s
jo
ur
na
l
us
e
s
th
e
C
ont
r
ib
ut
or
R
ol
e
s
T
a
xonomy
(
C
R
e
di
T
)
to
r
e
c
ogni
z
e
in
di
vi
dua
l
a
ut
hor
c
ont
r
ib
ut
io
ns
, r
e
duc
e
a
ut
hor
s
hi
p di
s
put
e
s
,
a
nd f
a
c
il
it
a
te
c
ol
la
bo
r
a
ti
on.
N
am
e
o
f
A
u
t
h
or
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
P
a
la
ni
r
a
j
R
a
ji
dur
a
i
P
a
r
va
th
y
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
N
a
ga
r
a
ja
n M
oha
nkuma
r
✓
✓
✓
✓
✓
✓
✓
✓
✓
R
a
je
ndr
a
n S
hobi
ga
✓
✓
✓
✓
✓
✓
✓
G
our
S
unda
r
M
it
r
a
T
ha
kur
✓
✓
✓
✓
✓
✓
✓
M
a
m
a
th
a
B
a
nda
r
u
✓
✓
✓
✓
✓
✓
✓
V
e
lu
s
a
m
y S
uj
a
th
a
✓
✓
✓
✓
✓
✓
✓
S
ha
nm
uga
m
S
uj
a
th
a
✓
✓
✓
✓
✓
C
:
C
onc
e
pt
ua
l
i
z
a
t
i
on
M
:
M
e
t
hodol
ogy
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
da
t
i
on
Fo
:
Fo
r
m
a
l
a
na
l
ys
i
s
I
:
I
nve
s
t
i
ga
t
i
on
R
:
R
e
s
our
c
e
s
D
:
D
a
t
a
C
ur
a
t
i
on
O
:
W
r
i
t
i
ng
-
O
r
i
gi
na
l
D
r
a
f
t
E
:
W
r
i
t
i
ng
-
R
e
vi
e
w
&
E
di
t
i
ng
Vi
:
Vi
s
ua
l
i
z
a
t
i
on
Su
:
Su
pe
r
vi
s
i
on
P
:
P
r
oj
e
c
t
a
dm
i
ni
s
t
r
a
t
i
on
Fu
:
Fu
ndi
ng a
c
qui
s
i
t
i
on
C
O
N
F
L
I
C
T
O
F
I
N
T
E
R
E
S
T
S
T
A
T
E
M
E
N
T
T
he
a
ut
hor
s
ha
v
e
no c
onf
li
c
t
of
i
nt
e
r
e
s
t
r
e
le
va
nt
t
o t
hi
s
pa
pe
r
.
D
A
T
A
A
V
A
I
L
A
B
I
L
I
T
Y
T
he
da
ta
th
a
t
s
uppor
t
th
e
f
in
di
ngs
of
th
is
s
tu
dy
a
r
e
a
va
il
a
bl
e
on
r
e
que
s
t
f
r
om
th
e
c
or
r
e
s
ponding
a
ut
hor
,
[
NM
]
.
T
he
da
ta
,
w
hi
c
h
c
ont
a
in
in
f
or
m
a
ti
on
th
a
t
c
oul
d
c
om
pr
om
is
e
th
e
pr
iv
a
c
y
of
r
e
s
e
a
r
c
h
pa
r
ti
c
ip
a
nt
s
, a
r
e
not
publi
c
ly
a
va
il
a
bl
e
due
t
o c
e
r
ta
in
r
e
s
tr
ic
ti
ons
.
R
E
F
E
R
E
N
C
E
S
[
1]
R
.
A
.
J
a
c
obs
,
M
.
I
.
J
or
da
n,
S
.
J
.
N
ow
l
a
n,
a
nd
G
.
E
.
H
i
nt
on,
“
A
da
pt
i
ve
m
i
xt
ur
e
of
l
oc
a
l
e
xpe
r
t
s
,”
N
e
ur
al
C
o
m
put
at
i
on
,
vol
.
3,
no.
1, pp. 79
–
87, 1991, doi
:
10.1162/
ne
c
o.1991.3.1.79.
[
2]
L
.
A
ugus
t
yni
a
k,
P
.
S
z
ym
á
ns
ki
,
T
.
K
a
j
da
now
i
c
z
,
a
nd
W
.
T
ul
i
gl
ow
i
c
z
,
“
C
om
pr
e
he
ns
i
ve
s
t
udy
on
l
e
xi
c
on
-
ba
s
e
d
e
n
s
e
m
bl
e
c
l
a
s
s
i
f
i
c
a
t
i
on s
e
nt
i
m
e
nt
a
n
a
l
ys
i
s
,
”
E
nt
r
opy
, vol
. 18, no. 1, 2016, doi
:
10.3390/
e
18010004.
[
3]
W
.
M
e
dha
t
,
A
.
H
a
s
s
a
n,
a
nd
H
.
K
or
a
s
hy,
“
S
e
nt
i
m
e
nt
a
na
l
ys
i
s
a
l
gor
i
t
hm
s
a
nd
a
ppl
i
c
a
t
i
ons
:
a
s
ur
ve
y,”
A
i
n
Sham
s
E
ngi
ne
e
r
i
ng
J
our
nal
, vol
. 5, no. 4, pp. 1093
–
1113, 2014, doi
:
10.1016/
j
.a
s
e
j
.2014.04.011.
[
4]
D
.
T
i
w
a
r
i
,
B
.
N
a
gpa
l
,
B
.
S
.
B
ha
t
i
,
A
.
M
i
s
hr
a
,
a
nd
M
.
K
um
a
r
,
“
A
s
ys
t
e
m
a
t
i
c
r
e
vi
e
w
of
s
oc
i
a
l
ne
t
w
or
k
s
e
nt
i
m
e
nt
a
na
l
ys
i
s
w
i
t
h
c
om
pa
r
a
t
i
ve
s
t
udy
of
e
ns
e
m
bl
e
-
ba
s
e
d
t
e
c
hni
que
s
,”
A
r
t
i
f
i
c
i
al
I
nt
e
l
l
i
ge
nc
e
R
e
v
i
e
w
,
vol
.
56,
no.
11,
pp.
13407
–
13461,
2023,
doi
:
10.1007/
s
10462
-
023
-
10472
-
w.
[
5]
A
.
S
a
l
i
h,
M
.
S
a
l
i
h,
a
nd
A
.
A
br
a
ha
m
,
“
N
ove
l
e
ns
e
m
bl
e
de
c
i
s
i
on
s
uppor
t
a
nd
h
e
a
l
t
h
c
a
r
e
m
oni
t
or
i
ng
s
ys
t
e
m
,”
J
our
nal
of
N
e
t
w
or
k
and I
nnov
at
i
v
e
C
om
put
i
ng
, vol
. 2, no. 2014, pp. 41
–
51, 2014.
[
6]
R
.
X
i
a
,
C
.
Z
ong,
a
nd
S
.
L
i
,
“
E
ns
e
m
bl
e
of
f
e
a
t
ur
e
s
e
t
s
a
nd
c
l
a
s
s
i
f
i
c
a
t
i
on
a
l
gor
i
t
hm
s
f
or
s
e
nt
i
m
e
nt
c
l
a
s
s
i
f
i
c
a
t
i
on,”
I
nf
or
m
at
i
on
Sc
i
e
nc
e
s
, vol
. 181, no. 6, pp. 1138
–
1152, 2011, doi
:
10.1016/
j
.i
ns
.2010.11.023.
[
7]
B
.
S
r
i
gi
r
i
r
a
j
u,
“
A
m
a
z
on
r
e
vi
e
w
s
:
K
i
ndl
e
s
t
or
e
c
a
t
e
gor
y,”
K
aggl
e
.
2018.
A
c
c
e
s
s
e
d:
J
un.
14,
2023.
[
O
nl
i
ne
]
.
A
va
i
l
a
bl
e
:
ht
t
ps
:
/
/
w
w
w
.ka
ggl
e
.c
om
/
da
t
a
s
e
t
s
/
bha
r
a
dw
a
j
6/
ki
ndl
e
-
r
e
vi
e
w
s
[
8]
M
.
W
a
nkha
d
e
,
A
.
C
.
S
.
R
a
o,
a
nd
C
.
K
ul
ka
r
ni
,
“
A
s
ur
ve
y
on
s
e
nt
i
m
e
nt
a
na
l
ys
i
s
m
e
t
hods
,
a
ppl
i
c
a
t
i
ons
,
a
nd
c
ha
l
l
e
ng
e
s
,”
A
r
t
i
f
i
c
i
al
I
nt
e
l
l
i
ge
nc
e
R
e
v
i
e
w
, vol
. 55, no. 7, pp. 5731
–
5780, 2022, doi
:
10.1007/
s
10462
-
022
-
10144
-
1.
[
9]
M
.
B
.
I
m
a
ni
,
M
.
R
.
K
e
yva
npour
,
a
nd
R
.
A
z
m
i
,
“
A
nov
e
l
e
m
be
dde
d
f
e
a
t
u
r
e
s
e
l
e
c
t
i
on
m
e
t
hod:
A
c
om
pa
r
a
t
i
ve
s
t
udy
i
n
t
h
e
a
ppl
i
c
a
t
i
on
of
t
e
xt
c
a
t
e
gor
i
z
a
t
i
on,”
A
ppl
i
e
d
A
r
t
i
f
i
c
i
al
I
nt
e
l
l
i
ge
nc
e
,
vol
.
27,
no.
5,
pp.
408
–
427,
2013,
do
i
:
10.1080/
08839514.2013.774211.
[
10]
T
.
K
.
B
a
l
a
j
i
,
C
.
S
.
R
.
A
nna
va
r
a
pu,
a
nd
A
.
B
a
bl
a
ni
,
“
M
a
c
hi
ne
l
e
a
r
ni
ng
a
l
gor
i
t
hm
s
f
or
s
oc
i
a
l
m
e
di
a
a
na
l
ys
i
s
:
A
s
ur
ve
y,”
C
om
put
e
r
Sc
i
e
nc
e
R
e
v
i
e
w
, vol
. 40, 2021, doi
:
10.1016/
j
.c
os
r
e
v.2021.100395.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
H
y
br
id
s
e
m
ant
ic
m
ode
l
bas
e
d on
m
ac
hi
ne
l
e
ar
ni
ng f
o
r
s
e
nt
ime
nt
…
(
P
al
ani
r
aj
R
aj
id
ur
ai
P
ar
v
at
hy
)
2009
[
11]
B
.
O
ha
na
a
nd
B
.
T
i
e
r
ne
y,
“
S
e
nt
i
m
e
nt
c
l
a
s
s
i
c
a
t
i
on
of
r
e
vi
e
w
s
us
i
ng
S
e
nt
i
W
or
dN
e
t
,”
i
n
T
he
9t
h.
I
T
&
T
C
onf
e
r
e
nc
e
,
2009
,
doi
:
10.21427/
D
77S
56.
[
12]
P
.
K
.
J
a
i
n,
R
.
P
a
m
ul
a
,
a
nd
G
.
S
r
i
va
s
t
a
va
,
“
A
s
ys
t
e
m
a
t
i
c
l
i
t
e
r
a
t
ur
e
r
e
vi
e
w
on
m
a
c
hi
ne
l
e
a
r
ni
ng
a
ppl
i
c
a
t
i
ons
f
or
c
ons
um
e
r
s
e
nt
i
m
e
nt
a
na
l
y
s
i
s
u
s
i
ng
onl
i
ne
r
e
vi
e
w
s
,”
C
om
put
e
r
Sc
i
e
nc
e
R
e
v
i
e
w
,
vol
.
41,
pp.
1
-
17,
A
ug.
2021,
doi
:
10.1016/
j
.c
os
r
e
v.2021.100413.
[
13]
S
.
L
i
,
C
.
Z
ong,
a
nd
X
.
W
a
ng,
“
S
e
nt
i
m
e
nt
c
l
a
s
s
i
f
i
c
a
t
i
on
t
hr
ough
c
om
bi
ni
ng
c
l
a
s
s
i
f
i
e
r
s
w
i
t
h
m
ul
t
i
pl
e
f
e
a
t
ur
e
s
e
t
s
,”
i
n
200
7
I
nt
e
r
nat
i
onal
C
onf
e
r
e
nc
e
on
N
at
u
r
al
L
anguage
P
r
oc
e
s
s
i
ng
and
K
no
w
l
e
dge
E
ngi
ne
e
r
i
ng
,
2007,
pp.
135
–
140
,
doi
:
10.1109/
N
L
P
K
E
.2007.4368024.
[
14]
B
.
H
.
J
ua
ng
a
nd
S
.
K
a
t
a
gi
r
i
,
“
D
i
s
c
r
i
m
i
na
t
i
ve
l
e
a
r
ni
ng
f
or
m
i
ni
m
um
e
r
r
or
c
l
a
s
s
i
f
i
c
a
t
i
on,”
I
E
E
E
T
r
ans
ac
t
i
ons
on
Si
gnal
P
r
oc
e
s
s
i
ng
, vol
. 40, no. 12, pp. 3043
–
3054, 1992, doi
:
10.1109/
78.175747.
[
15]
B
.
E
r
ş
a
hi
n,
Ö
.
A
kt
a
ş
,
D
.
K
i
l
i
nç
,
a
nd
M
. E
r
ş
a
hi
n,
“
A
hybr
i
d
s
e
nt
i
m
e
nt
a
na
l
ys
i
s
m
e
t
hod
f
or
T
ur
ki
s
h,”
T
ur
k
i
s
h
J
our
nal
of
E
l
e
c
t
r
i
c
al
E
ngi
ne
e
r
i
ng and C
om
put
e
r
Sc
i
e
n
c
e
s
, vol
. 27, no. 3, pp. 1780
–
1793, 2019, doi
:
10.3906/
e
l
k
-
1808
-
189.
[
16]
O
.
C
oba
n,
B
.
M
.
O
z
yi
l
di
r
i
m
,
a
nd
S
.
A
.
O
z
e
l
,
“
A
n
e
m
pi
r
i
c
a
l
s
t
udy
of
t
he
e
xt
r
e
m
e
l
e
a
r
ni
ng
m
a
c
hi
ne
f
or
T
w
i
t
t
e
r
s
e
nt
i
m
e
nt
a
na
l
ys
i
s
,”
I
nt
e
r
nat
i
onal
J
ou
r
nal
of
I
nt
e
l
l
i
ge
nt
Sy
s
t
e
m
s
and
A
ppl
i
c
at
i
ons
i
n
E
ngi
ne
e
r
i
ng
,
vol
.
3,
no.
6,
pp.
178
–
184,
2018,
doi
:
10.18201/
i
j
i
s
a
e
.2018644774.
[
17]
R
.
A
l
f
r
j
a
ni
,
T
.
O
s
m
a
n,
a
nd
G
.
C
o
s
m
a
,
“
A
hybr
i
d
s
e
m
a
nt
i
c
know
l
e
dge
ba
s
e
-
m
a
c
hi
ne
l
e
a
r
ni
ng
a
ppr
oa
c
h
f
or
opi
ni
on
m
i
ni
ng,”
D
at
a
and K
now
l
e
dge
E
ngi
ne
e
r
i
ng
, vol
. 121, pp. 88
–
108, 2019, doi
:
10.1016/
j
.da
t
a
k.
2019.05.002.
[
18]
P
.
B
hoi
r
a
nd
S
.
K
ol
t
e
,
“
S
e
nt
i
m
e
nt
a
na
l
ys
i
s
of
m
ovi
e
r
e
vi
e
w
s
us
i
ng
l
e
xi
c
on
a
ppr
oa
c
h,”
i
n
2015
I
E
E
E
I
nt
e
r
nat
i
onal
C
onf
e
r
e
nc
e
on
C
om
put
at
i
onal
I
nt
e
l
l
i
ge
nc
e
and
C
om
put
i
ng
R
e
s
e
ar
c
h
(
I
C
C
I
C
)
,
M
a
dur
a
i
,
I
ndi
a
,
D
e
c
.
2015,
pp.
1
-
6,
doi
:
10.1109/
I
C
C
I
C
.2015.7435796.
[
19]
E
.
C
a
m
br
i
a
,
“
A
n
i
nt
r
oduc
t
i
on
t
o
c
onc
e
pt
-
l
e
ve
l
s
e
nt
i
m
e
nt
a
na
l
ys
i
s
,”
i
n
A
dv
anc
e
s
i
n
Sof
t
C
om
put
i
ng
and
I
t
s
A
ppl
i
c
at
i
ons
,
B
e
r
l
i
n,
H
e
i
de
l
be
r
g:
S
pr
i
nge
r
, 2013, pp. 478
–
483
, doi
:
10.1007/
978
-
3
-
642
-
45111
-
9_41.
[
20]
M
.
D
.
A
.
H
a
s
a
n,
K
.
B
a
l
a
s
uba
dr
a
,
G
.
V
a
di
ve
l
,
N
.
A
r
unf
r
e
d,
M
.
V
.
I
s
hw
a
r
ya
,
a
nd
S
.
M
ur
uga
n,
“
I
oT
-
d
r
i
ve
n
i
m
a
ge
r
e
c
ogni
t
i
on
f
o
r
m
i
c
r
opl
a
s
t
i
c
a
na
l
ys
i
s
i
n
w
a
t
e
r
s
ys
t
e
m
s
us
i
ng
c
onvol
ut
i
ona
l
ne
ur
a
l
ne
t
w
or
ks
,”
i
n
2024
2nd I
nt
e
r
nat
i
onal
C
onf
e
r
e
nc
e
on
C
om
put
e
r
,
C
om
m
uni
c
at
i
on and C
ont
r
ol
(
I
C
4)
, 2024, pp. 1
–
6
, doi
:
10.1109/
I
C
457434.2024.10486490.
[
21]
K
.
R
a
vi
a
nd
V
.
R
a
vi
,
“
A
s
ur
ve
y
on
opi
ni
on
m
i
ni
ng
a
nd
s
e
nt
i
m
e
nt
a
na
l
y
s
i
s
:
T
a
s
ks
,
a
ppr
oa
c
he
s
a
nd
a
ppl
i
c
a
t
i
on
s
,”
K
no
w
l
e
dge
-
B
as
e
d Sy
s
t
e
m
s
, vol
. 89, pp. 14
–
46, 2015, doi
:
10.1016/
j
.knos
ys
.2015.06.015.
[
22]
M
.
W
hi
t
e
he
a
d
a
nd
L
.
Y
a
e
ge
r
,
“
S
e
nt
i
m
e
nt
m
i
ni
ng
u
s
i
ng
e
n
s
e
m
bl
e
c
l
a
s
s
i
f
i
c
a
t
i
o
n
m
ode
l
s
,”
I
nnov
at
i
ons
and
A
dv
anc
e
s
i
n
C
om
put
e
r
Sc
i
e
nc
e
s
and E
ngi
ne
e
r
i
ng
, pp. 509
–
514, 2010, doi
:
10.1007/
978
-
90
-
481
-
3658
-
2_89.
[
23]
R
.
I
.
P
e
r
m
a
t
a
s
a
r
i
,
M
.
A
.
F
a
uz
i
,
P
.
P
.
A
di
ka
r
a
,
a
nd
E
.
D
.
L
.
S
a
r
i
,
“
T
w
i
t
t
e
r
s
e
nt
i
m
e
nt
a
na
l
ys
i
s
of
m
ovi
e
r
e
vi
e
w
s
u
s
i
ng
e
ns
e
m
bl
e
f
e
a
t
ur
e
s
ba
s
e
d
na
ï
ve
B
a
ye
s
,”
i
n
2018
I
nt
e
r
nat
i
onal
C
onf
e
r
e
nc
e
on
Su
s
t
ai
nabl
e
I
nf
or
m
at
i
on
E
ngi
ne
e
r
i
ng
and
T
e
c
hnol
ogy
(
SI
E
T
)
,
2018, pp. 92
–
95
, doi
:
10.1109/
S
I
E
T
.2018.8693195.
[
24]
H
.
R
e
hi
oui
a
nd
A
.
I
dr
i
s
s
i
,
“
N
e
w
c
l
u
s
t
e
r
i
ng
a
l
gor
i
t
hm
s
f
or
t
w
i
t
t
e
r
s
e
nt
i
m
e
nt
a
na
l
ys
i
s
,”
I
E
E
E
Sy
s
t
e
m
s
J
our
nal
,
vol
.
14,
no.
1,
pp
.
530
–
537, 2020, doi
:
10.1109/
J
S
Y
S
T
.2019.2912759.
[
25]
S
.
A
l
-
S
a
qqa
,
N
.
O
be
i
d,
a
nd
A
.
A
w
a
j
a
n,
“
S
e
nt
i
m
e
nt
a
na
l
y
s
i
s
f
or
A
r
a
bi
c
t
e
xt
us
i
ng
e
ns
e
m
bl
e
l
e
a
r
ni
ng,”
i
n
2018
I
E
E
E
/
A
C
S
15t
h
I
nt
e
r
nat
i
onal
C
onf
e
r
e
nc
e
on
C
o
m
put
e
r
Sy
s
t
e
m
s
and
A
ppl
i
c
at
i
ons
(
A
I
C
C
SA
)
,
A
qa
ba
,
J
or
da
n,
2018,
pp.
1
-
7,
doi
:
10.1109/
A
I
C
C
S
A
.2018.8612804.
[
26]
R
.
N
.
B
e
he
r
a
,
M
.
R
oy,
a
nd
S
.
D
a
s
h,
“
E
ns
e
m
bl
e
ba
s
e
d
hybr
i
d
m
a
c
hi
ne
l
e
a
r
ni
ng
a
ppr
oa
c
h
f
or
s
e
nt
i
m
e
nt
c
l
a
s
s
i
f
i
c
a
t
i
on
–
a
r
e
vi
e
w
,”
I
nt
e
r
nat
i
onal
J
our
nal
of
C
om
put
e
r
A
ppl
i
c
at
i
ons
, vol
. 146, no. 6, pp. 31
–
36, 20
16, doi
:
10.5120/
i
j
c
a
2016910813.
[
27]
J
.
S
a
dha
s
i
va
m
a
nd
R
.
B
.
K
a
l
i
va
r
a
dha
n,
“
S
e
nt
i
m
e
nt
a
na
l
ys
i
s
of
A
m
a
z
on
pr
oduc
t
s
us
i
ng
e
ns
e
m
bl
e
m
a
c
hi
ne
l
e
a
r
ni
ng
a
l
gor
i
t
hm
,”
I
nt
e
r
nat
i
onal
J
our
nal
of
M
at
he
m
at
i
c
al
,
E
ngi
ne
e
r
i
ng
and
M
anage
m
e
nt
Sc
i
e
nc
e
s
,
vol
.
4,
no.
2,
pp.
508
–
520,
2019,
doi
:
10.33889/
i
j
m
e
m
s
.2019.4.2
-
041.
[
28]
K
. S
. R
a
j
e
s
h, S
. K
a
l
a
i
s
e
l
vi
, P
. S
a
t
hya
na
t
ha
n, K
. S
udha
, G
. K
a
vi
t
ha
, a
nd S
. M
ur
uga
n, “
S
m
a
r
t
ve
ndi
ng m
a
c
hi
ne
s
:
I
m
pr
ovi
ng s
e
r
vi
c
e
qua
l
i
t
y
w
i
t
h
i
nt
e
gr
a
t
e
d
c
ont
r
ol
l
e
r
s
a
nd
S
V
M
a
l
gor
i
t
hm
,”
i
n
2024
6t
h
I
nt
e
r
nat
i
onal
C
onf
e
r
e
nc
e
on
E
ne
r
gy
,
P
ow
e
r
an
d
E
nv
i
r
onm
e
nt
(
I
C
E
P
E
)
, 2024, pp. 1
–
6
, doi
:
10.1109/
I
C
E
P
E
63236.2024.106688
78.
[
29]
A
.
M
.
R
a
j
e
s
w
a
r
i
,
M
.
M
a
ha
l
a
ks
hm
i
,
R
.
N
i
t
hya
s
hr
e
e
,
a
nd
G
.
N
a
l
i
ni
,
“
S
e
nt
i
m
e
nt
a
na
l
ys
i
s
f
or
p
r
e
di
c
t
i
ng
c
us
t
om
e
r
r
e
vi
e
w
s
us
i
ng
a
hybr
i
d
a
ppr
oa
c
h,”
i
n
2020
A
dv
anc
e
d
C
om
put
i
ng
and
C
om
m
uni
c
at
i
on
T
e
c
hnol
ogi
e
s
f
or
H
i
gh
P
e
r
f
or
m
anc
e
A
ppl
i
c
at
i
ons
(
A
C
C
T
H
P
A
)
, 2020, pp. 200
–
205
, do
i
:
10.1109/
A
C
C
T
H
P
A
49271.2020.9213236.
[
30]
P
.
M
a
he
s
w
a
r
i
,
S
.
G
ow
r
i
s
w
a
r
i
,
S
.
B
a
l
a
s
ubr
a
m
a
ni
,
A
.
R
.
B
a
bu,
N
.
K
.
J
i
j
i
t
h,
a
n
d
S
.
M
ur
uga
n,
“
I
nt
e
l
l
i
ge
nt
he
a
dl
i
ght
s
f
o
r
a
da
pt
i
ng
be
a
m
pa
t
t
e
r
ns
w
i
t
h
r
a
s
pbe
r
r
y
pi
a
nd
c
onvol
ut
i
ona
l
ne
ur
a
l
ne
t
w
or
ks
,”
i
n
2024
2nd
I
nt
e
r
nat
i
onal
C
onf
e
r
e
nc
e
on
D
e
v
i
c
e
I
nt
e
l
l
i
ge
nc
e
,
C
om
put
i
ng
and
C
om
m
uni
c
at
i
on
T
e
c
hnol
ogi
e
s
(
D
I
C
C
T
)
,
2024,
pp.
182
–
187
,
doi
:
10.1109/
D
I
C
C
T
61038.2024.10533159.
[
31]
S
.
S
r
i
dha
r
a
nd
S
.
S
a
na
ga
va
r
a
pu,
“
S
e
nt
i
m
e
nt
a
na
l
ys
i
s
us
i
ng
e
n
s
e
m
bl
e
-
hybr
i
d
m
ode
l
w
i
t
h
hype
r
nym
ba
s
e
d
f
e
a
t
ur
e
e
ngi
ne
e
r
i
ng,”
i
n
2020
F
O
R
T
E
I
-
I
nt
e
r
nat
i
onal
C
onf
e
r
e
nc
e
on
E
l
e
c
t
r
i
c
al
E
ngi
ne
e
r
i
ng
(
F
O
R
T
E
I
-
I
C
E
E
)
,
2020,
pp.
13
–
18
,
doi
:
10.1109/
F
O
R
T
E
I
-
I
C
E
E
50915.2020.9249945.
[
32]
B
.
J
.
G
a
ne
s
h,
P
.
V
i
j
a
ya
n,
V
.
V
a
i
de
hi
,
S
.
M
ur
uga
n,
R
.
M
e
e
na
ks
hi
,
a
nd
M
.
R
a
j
m
oha
n,
“
S
V
M
-
ba
s
e
d
pr
e
di
c
t
i
ve
m
ode
l
i
ng
of
dr
ow
s
i
ne
s
s
i
n
hos
pi
t
a
l
s
t
a
f
f
f
or
oc
c
upa
t
i
ona
l
s
a
f
e
t
y
s
ol
ut
i
on
vi
a
I
o
T
i
nf
r
a
s
t
r
uc
t
ur
e
,”
i
n
2024
2nd
I
nt
e
r
nat
i
onal
C
onf
e
r
e
nc
e
on
C
om
put
e
r
, C
om
m
uni
c
at
i
on and C
ont
r
ol
(
I
C
4)
, 2024, pp. 1
–
5
, doi
:
10.1109/
I
C
457434.2024.10486429.
[
33]
A
.
M
ukw
a
z
vur
e
a
nd
K
.
.
S
upr
e
e
t
hi
,
“
A
hybr
i
d
a
ppr
oa
c
h
t
o
s
e
nt
i
m
e
nt
a
na
l
ys
i
s
of
ne
w
s
c
om
m
e
nt
s
,”
i
n
2015
4t
h
I
nt
e
r
nat
i
ona
l
C
onf
e
r
e
nc
e
on
R
e
l
i
abi
l
i
t
y
,
I
n
f
oc
om
T
e
c
hnol
ogi
e
s
and
O
pt
i
m
i
z
at
i
on
(
I
C
R
I
T
O
)
(
T
r
e
nds
and
F
ut
ur
e
D
i
r
e
c
t
i
ons
)
,
2015,
pp.
1
–
6
,
doi
:
10.1109/
I
C
R
I
T
O
.2015.7359282.
[
34]
V
.
V
.
B
a
s
ka
r
,
S
.
S
e
ka
r
,
K
.
S
.
R
a
j
e
s
h,
N
.
C
.
S
e
ndhi
l
kum
a
r
,
T
.
R
,
a
nd
S
.
M
u
r
uga
n,
“
C
l
oud
-
ba
s
e
d
de
c
i
s
i
on
s
uppor
t
s
ys
t
e
m
s
f
or
s
e
c
ur
i
ng
f
a
r
m
-
to
-
t
a
bl
e
t
r
a
c
e
a
bi
l
i
t
y
us
i
ng
I
o
T
a
nd
K
N
N
a
l
gor
i
t
hm
,”
i
n
2024
Se
c
ond
I
nt
e
r
nat
i
onal
C
onf
e
r
e
nc
e
on
I
nt
e
l
l
i
ge
nt
C
y
be
r
P
hy
s
i
c
al
Sy
s
t
e
m
s
and
I
nt
e
r
ne
t
of
T
hi
ng
s
(
I
C
oI
C
I
)
,
C
oi
m
ba
t
or
e
,
I
ndi
a
,
2024,
pp.
443
-
448,
doi
:
10.1109/
I
C
oI
C
I
62503.2024.10696659.
[
35]
G
.
T
ha
hni
ya
t
h
e
t
al
.
,
“
C
l
oud
ba
s
e
d
pr
e
di
c
t
i
on
of
e
pi
l
e
pt
i
c
s
e
i
z
ur
e
s
us
i
ng
r
e
a
l
-
t
i
m
e
e
l
e
c
t
r
oe
nc
e
pha
l
ogr
a
m
s
a
na
l
ys
i
s
,”
I
nt
e
r
nat
i
onal
J
our
nal
of
E
l
e
c
t
r
i
c
al
and
C
om
put
e
r
E
ngi
ne
e
r
i
ng
(
I
J
E
C
E
)
,
vol
.
14
,
no.
5,
pp.
6047
–
6056,
O
c
t
.
2024,
doi
:
10.11591/
i
j
e
c
e
.v14i
5.pp6047
-
6056.
[
36]
N
.
M
oha
nkum
a
r
e
t
al
.
,
“
A
dva
nc
i
ng
c
hr
oni
c
pa
i
n
r
e
l
i
e
f
c
l
oud
-
ba
s
e
d
r
e
m
ot
e
m
a
na
ge
m
e
nt
w
i
t
h
m
a
c
hi
ne
l
e
a
r
ni
ng
i
n
he
a
l
t
hc
a
r
e
,”
I
ndone
s
i
an
J
our
nal
of
E
l
e
c
t
r
i
c
al
E
ngi
ne
e
r
i
ng
and
C
om
put
e
r
Sc
i
e
n
c
e
,
vol
.
37,
no.
2,
pp.
1042
–
1052,
2025,
doi
:
10.11591/
i
j
e
e
c
s
.v37.i
2.pp1042
-
1052.
[
37]
J
.
R
a
m
a
s
a
m
y,
E
.
S
r
i
vi
dhya
,
V
.
V
a
i
de
hi
,
S
.
V
i
m
a
l
a
de
vi
,
N
.
M
oha
nkum
a
r
,
a
n
d
S
.
M
ur
uga
n,
“
C
l
oud
-
e
na
bl
e
d
i
s
ol
a
t
i
on
f
or
e
s
t
f
or
a
nom
a
l
y
de
t
e
c
t
i
on
i
n
U
A
V
-
ba
s
e
d
pow
e
r
l
i
ne
i
ns
pe
c
t
i
on,”
i
n
2024
2nd
I
nt
e
r
nat
i
onal
C
onf
e
r
e
nc
e
on
N
e
t
w
or
k
i
ng
and
C
om
m
uni
c
at
i
ons
(
I
C
N
W
C
)
, 2024, pp. 1
–
6
, doi
:
10.1109/
I
C
N
W
C
60771.2024.1
0537407.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
. 14, No. 3, J
une
2025
:
2001
-
2011
2010
[
38]
J
.
J
e
ga
n,
M
.
R
.
S
uguna
,
M
.
S
hoba
na
,
H
.
A
z
a
t
h,
S
.
M
ur
uga
n,
a
nd
M
.
R
a
j
m
oha
n
,
“
I
oT
-
e
na
bl
e
d
bl
a
c
k
box
f
or
d
r
i
ve
r
be
ha
vi
or
a
na
l
ys
i
s
us
i
ng
c
l
oud
c
om
put
i
ng
,”
2024
I
nt
e
r
nat
i
onal
C
onf
e
r
e
nc
e
on
A
dv
anc
e
s
i
n
D
at
a
E
ngi
ne
e
r
i
ng
and
I
nt
e
l
l
i
ge
nt
C
om
put
i
ng
Sy
s
t
e
m
s
(
A
D
I
C
S)
, C
he
nna
i
, I
ndi
a
, 2024, pp. 1
-
6, doi
:
10.1109/
A
D
I
C
S
58448.2024.10533471
.
[
39]
A
.
A
.
K
.
T
a
hi
r
a
nd
S
.
A
nghe
l
us
,
“
I
m
p
r
ovi
n
g
i
r
i
s
r
e
c
ogn
i
t
i
o
n
a
c
c
u
r
a
c
y
us
i
ng
ga
bor
ke
r
ne
l
s
w
i
t
h
ne
a
r
-
ho
r
i
z
on
t
a
l
o
r
i
e
n
t
a
t
i
ons
,”
I
nt
e
r
na
t
i
ona
l
J
our
na
l
of
A
dv
anc
e
s
i
n
Si
g
nal
a
nd
I
m
age
Sc
i
e
nc
e
s
, vo
l
. 8
, no.
1, p
p. 25
–
3
9, 20
22, d
oi
:
10.29
284/
i
j
a
s
i
s
.8.
1.202
2.25
-
39.
B
I
O
G
R
A
P
H
I
E
S
O
F
A
U
T
H
O
R
S
Mr.
Palaniraj
Rajidurai
Parvathy
is
a
Project
Manager
at
Mphas
is
Corporation
in
Chandler,
Arizona,
USA.
He
has
16+
years
of
IT
experience
in
the BI
and
analytics
domain
with
a
focus
on
data
modeling,
integration
and
visualization
(Snowfl
ake,
Azure,
AWS,
GCP,
Azure
Data
Factory,
Databricks,
Tableau,
Power
BI,
Python,
R,
SA
P
BO,
Altreyx,
Xceptor
(RPA)).
Rewarded
from
customer
for
providi
ng
“customer
value
ad
dition
”
for
performance
tuning
on
Schedule.
He
has
been
rewarded
the
“
Star
performer
”
a
w
ard
of
the
quarter
for
a
support
project
by
Hexaware
leadership
.
Also,
he
received
the
“Stat
p
erformer”
award
of
the
quarter
for
migration
project
fro
m
Hexaware
leadersh
ip
.
Moreover,
r
ewarded
“Most
valuable
player
”
support
project
from
wipro
-
best
buy
account
leadership.
F
urthermore
,
r
ewarded
a
“Feather
in
my
cap”
for
outstanding
contribution
to
the
project
b
usiness
group
hierarchy
itera
tion.
He
was
rewarded
with
a
“Feather
in
my
cap”
awa
rd
for
his
outstanding
contribution
to
Project
Business
Group
Hierarchy
Iteration
1.
He
can
b
e
contacted
at
email:
palanirajrps@
gmail.com.
Dr.
Nagarajan
Mohankumar
was
born
in
India
in
1978.
He
received
his
B.E.
d
egree
from
Bharathiyar
University,
Tamilnadu,
India
in
2000
and
M.E.
and
Ph.D
.
d
egree
from
Jadavpur
University,
Kolkata
in
2004
&
2010.
He
joined
the
Nano
Device
Simulation
Laboratory
in
2007
and
worked
as
a
senior
research
f
ellow
under
CSIR
direct
Scheme
till
Septemb
er
2009.
Later
he
joined
SKP
Engine
ering
College
as
a
Profe
ssor
to
develo
p
resea
rc
h
activit
ies
in
the
field
of
VLSI
and
NANO
technolo
gy.
He
is
current
ly
working
as
Researc
h
Profes
sor
at
Symbiosis
Institute
of
Techn
ology,
Nagpur
Campus,
(
Symbiosis
Inter
nationa
l
Deemed
University
)
,
Pune,
India.
He
is
a
senior
member
of
I
EEE.
He
has
about
85
internationa
l
journal
publications
in
reputed
journals
and
about
50
internationa
l
confere
nce
proceedings.
He
received
the
carrier
award
for
young
teachers
(CA
YT)
from
AICTE,
New
Delhi
in
the
year
of
2012
-
2014.
His
research
interest
includes
modeli
ng
and
simulation
study
of
HEMTs,
optimization
of
devices
for
RF
applications
and
characteriza
tion
of
advanced
HEMT
architecture,
Terahertz
electroni
cs,
high
frequency
i
maging,
sensors
and
communi
cation
.
He can be contacted at email:
nmkprofessor@
gmail.com.
Ms.
Rajendran
Shobiga
received
her
B.E.
in
electroni
cs
and
c
ommuni
cation
engine
ering
from
Greentech
College
of
Engineering
f
o
r
Womens
(A
nna
University)
in
2014
and
her
M.E.
from
Karpagam
College
of
Engineering
(Anna
Univers
ity)
in
2016.
At
present;
she
is
working
as
an
Assistant
Professor
at
J.
J
.
College
of
Engin
eering
and
Technol
ogy,
Trichy.
The
author
has
internationa
l
approved
master
engineer
in
co
mmunication
system
by
world
education
ser
vices
Canada.
She
has
working
experience
as
admi
nistrator
cum
planner
in
how
united
services
LLC,
Doha,
Qatar
a
nd
Planner
cum
estimat
or
in
spectrum
cube
technical
service
s LLC, Dubai, UAE. She
can be
contacted
at email
: shobi
garaj123@
gmail.co
m.
Dr.
Gour
Sundar
Mitra
Thakur
is
educator
,
AI
enthusiast,
r
esearcher
,
and
mentor
.
He
is
AI
enthusiast
with
16
years
of
teaching
and
research
ex
perience.
After
pursuing
B.Tech
.
and
M.Tech
.
in
computer
science
and
engineerin
g
,
he
co
mpleted
Ph
.
D
.
,
with
th
e
thesis
titled
“Intelligent
systems
in
manifold
decision
making
”
fro
m
National
Institute
of
Technology
Durgapur,
India.
C
ur
rently
,
he
associ
ated
with
Dr.
B.
C.
Roy
Engineering
College
Durgapur
as
Associate
Professor
in
the
D
epartment
of
Comput
er
S
cience
and
Engineeri
ng
(AIML).
He
was
able
to
publish
research
findings
in
22
internation
al
journal
s,
national
an
d
internationa
l
confere
nces
and
book
chapters
till
date.
He
working
as
a
reviewer
of
many
reputed
internatio
nal
journals
like,
Knowledge
Based
Systems,
Applied
Soft
Computi
ng,
International
Journal
of
Fuzzy
Systems,
and
Frontie
rs
in
Oncolo
g
y.
He
passionate
about
research
and
curious
about
applying
different
AI/ML/DL
algorithm
s
t
o
solve
various
real
-
life
problems
in
biology,
agriculture,
environment
and
manufacturing
indu
stries.
He
has
the
ability
to
collaborate
with
other
experts
from
differen
t
fields
to
achieve
a
co
mmon
goal.
He
believes
in
and
is
commit
ted
to
develop
AI
systems
that
are
fair,
transparen
t,
and
unbiased.
He
has
developed
various
modern
ML/DL
models
using
Tensorflow
and
Pyto
rch
and
developed
some
expertise
in
ideating
ML
pipelin
es
for
complex
real
-
life
problems.
He
can
be
contacted
at
email
: gour.mitrath
akur@bcrec.ac.in.
Evaluation Warning : The document was created with Spire.PDF for Python.