I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
p
ute
r
E
ng
in
ee
ring
(
I
J
E
CE
)
Vo
l.
10
,
No
.
6
,
Dec
em
b
er
2
0
2
0
,
p
p
.
5
9
9
2
~
6
0
0
0
I
SS
N:
2
0
8
8
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v
1
0
i
6
.
pp
5
9
9
2
-
6
0
0
0
5992
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m/in
d
ex
.
p
h
p
/I
JE
C
E
M
ulti
-
l
ing
ua
l
T
w
i
tt
er
s
enti
m
e
nt
a
n
a
ly
sis
u
sing
m
a
ch
ine
l
ea
rning
K.
Arun
1
,
A.
Srina
g
esh
2
1
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter S
c
ien
c
e
a
n
d
E
n
g
in
e
e
rin
g
,
A
c
h
a
r
y
a
Na
g
a
rju
n
a
Un
iv
e
rsit
y
,
In
d
ia
2
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter S
c
ien
c
e
a
n
d
E
n
g
in
e
e
rin
g
,
Ra
y
a
p
a
ti
V
e
n
k
a
ta Ran
g
a
Ra
o
a
n
d
Ja
g
a
rla
m
u
d
i
Ch
a
n
d
ra
m
o
u
li
(R.
V
.
R
.
a
n
d
J.C.
)
C
o
l
leg
e
o
f
En
g
in
e
e
rin
g
,
A
c
h
a
r
y
a
N
a
g
a
rju
n
a
Un
iv
e
rsit
y
,
In
d
ia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Au
g
2
,
2
0
1
9
R
ev
i
s
ed
Ma
y
2
,
2020
A
cc
ep
ted
Ma
y
1
7
,
2
0
2
0
Tw
it
ter
S
e
n
ti
m
e
n
t
A
n
a
l
y
sis
is
o
n
e
o
f
th
e
lea
d
i
n
g
re
se
a
rc
h
f
ield
s
n
o
w
a
d
a
y
s
.
M
o
st
o
f
th
e
re
se
a
rc
h
e
rs
h
a
v
e
c
o
n
tri
b
u
te
d
to
th
e
re
se
a
rc
h
in
tw
it
ter
se
n
ti
m
e
n
t
a
n
a
ly
sis
in
En
g
li
sh
tw
e
e
ts,
b
u
t
f
e
w
re
s
e
a
rc
h
e
rs
h
a
v
e
f
o
c
u
s
ed
on
th
e
m
u
lt
il
in
g
u
a
l
tw
it
ter
se
n
ti
m
e
n
t
a
n
a
ly
si
s.
S
ti
ll
,
so
m
e
m
o
re
c
h
a
l
len
g
e
s
a
re
p
re
se
n
t
a
n
d
n
o
t
y
e
t
a
d
d
re
ss
e
d
in
th
e
d
o
m
a
in
o
f
m
u
lt
il
in
g
u
a
l
tw
it
ter
se
n
ti
m
e
n
t
a
n
a
ly
sis
(
M
LT
S
A
).
R
e
se
a
r
c
h
is
h
ig
h
ly
wa
rra
n
ted
in
th
e
se
u
n
e
x
p
lo
re
d
a
re
a
s
.
T
h
is
stu
d
y
p
re
se
n
ts
th
e
im
p
le
m
e
n
tatio
n
o
f
se
n
ti
m
e
n
t
a
n
a
l
y
sis
in
m
u
lt
il
in
g
u
a
l
tw
it
ter
d
a
ta
a
n
d
im
p
ro
v
e
s
th
e
d
a
ta
c
las
si
f
ica
ti
o
n
u
p
t
o
th
e
a
d
e
q
u
a
t
e
lev
e
l
o
f
a
c
c
u
ra
c
y
.
Tw
it
ter
is
th
e
si
x
th
lea
d
in
g
so
c
ial
n
e
tw
o
rk
in
g
site
in
th
e
w
o
rld
.
A
c
ti
v
e
u
se
rs
f
o
r
tw
it
ter
in
a
m
o
n
t
h
a
re
3
3
0
m
il
li
o
n
.
P
e
o
p
le
c
a
n
tw
e
e
t
o
r
re
twe
e
t
in
th
e
ir
lan
g
u
a
g
e
s
a
n
d
a
ll
o
w
u
se
rs
to
u
se
e
m
o
ji
’s,
a
b
b
re
v
i
a
ti
o
n
s,
c
o
n
trac
ti
o
n
w
o
rd
s,
m
is
sp
e
ll
in
g
s,
a
n
d
sh
o
rtc
u
t
w
o
rd
s.
T
h
e
b
e
st
p
latf
o
r
m
f
o
r
se
n
ti
m
e
n
t
a
n
a
l
y
sis
is
t
w
it
ter.
M
u
lt
il
in
g
u
a
l
tw
e
e
ts
a
n
d
d
a
ta
s
p
a
rsity
a
re
th
e
tw
o
m
a
i
n
c
h
a
ll
e
n
g
e
s.
In
th
is
p
a
p
e
r,
t
h
e
M
L
T
S
A
a
l
g
o
rit
h
m
g
i
v
e
s
th
e
so
l
u
ti
o
n
f
o
r
th
e
se
tw
o
c
h
a
ll
e
n
g
e
s.
M
LT
S
A
a
l
g
o
rit
h
m
is
d
iv
id
e
d
i
n
to
tw
o
p
a
rts.
On
e
f
o
r
d
e
tec
ti
n
g
a
n
d
tran
sl
a
ti
n
g
n
o
n
-
En
g
li
s
h
tw
e
e
ts i
n
to
En
g
li
sh
u
si
n
g
n
a
tu
ra
l
lan
g
u
a
g
e
p
ro
c
e
ss
in
g
(NL
P
)
a
n
d
t
h
e
se
c
o
n
d
o
n
e
is
an
a
p
p
ro
p
riate
p
re
-
p
ro
c
e
ss
in
g
m
e
th
o
d
w
it
h
NLP
s
u
p
p
o
rt
t
h
a
t
c
a
n
re
d
u
c
e
th
e
d
a
ta
sp
a
rsity
.
T
h
e
re
su
lt
o
f
th
e
M
L
T
S
A
w
it
h
S
V
M
a
c
h
iev
e
s g
o
o
d
a
c
c
u
ra
c
y
b
y
u
p
to
9
5
%
.
K
ey
w
o
r
d
s
:
Ma
ch
i
n
e
lear
n
i
n
g
a
l
g
o
r
it
h
m
s
Mu
ltil
in
g
u
al
NL
P
T
w
it
ter
s
en
t
i
m
e
n
t a
n
al
y
s
i
s
Co
p
y
rig
h
t
©
2
0
2
0
In
stit
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
Al
l
rig
h
ts re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
K.
A
r
u
n
,
Dep
ar
t
m
en
t
o
f
C
o
m
p
u
ter
Scie
n
ce
E
n
g
i
n
ee
r
in
g
,
A
c
h
ar
y
a
Na
g
ar
j
u
n
a
U
n
i
v
er
s
it
y
,
NH1
6
,
Nag
ar
j
u
n
a
Nag
ar
,
Gu
n
tu
r
,
An
d
h
r
a
P
r
ad
esh
,
I
n
d
ia.
E
m
ail:
k
ar
u
n
0
1
4
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
T
w
it
ter
is
m
icr
o
b
lo
g
g
i
n
g
,
w
h
ich
allo
w
s
a
m
a
x
i
m
u
m
o
f
2
8
0
ch
ar
ac
te
r
s
p
er
t
w
ee
t;
t
w
i
tte
r
is
o
n
e
o
f
th
e
b
est
s
o
cial
n
e
t
w
o
r
k
s
.
T
w
it
ter
h
as
3
3
0
m
il
lio
n
ac
ti
v
e
u
s
e
r
s
p
er
m
o
n
t
h
[
1
]
.
I
t
p
r
o
v
id
es
a
g
o
o
d
p
latf
o
r
m
to
s
h
ar
e
u
s
er
'
s
o
p
in
io
n
s
a
n
d
v
ie
w
s
ab
o
u
t
t
h
e
tr
e
n
d
in
g
to
p
ics.
T
w
i
tter
h
a
s
n
o
lo
ca
l
g
r
o
u
p
s
;
an
y
u
s
er
ca
n
p
o
s
t
t
w
ee
t
s
an
d
h
as o
p
en
ac
ce
s
s
.
T
w
it
ter
p
r
o
v
id
es
a
g
o
o
d
en
v
i
r
o
n
m
e
n
t
f
o
r
s
e
n
ti
m
e
n
t
a
n
al
y
s
is
[
2
]
.
Sen
ti
m
en
t
clas
s
i
f
icati
o
n
ca
n
b
e
d
o
n
e
o
n
t
w
i
tter
d
ata
lik
e
p
o
s
it
iv
e,
n
e
g
ati
v
e
an
d
n
e
u
tr
al.
Se
n
ti
m
e
n
t
an
a
l
y
s
is
ca
n
b
e
ap
p
lied
to
m
o
v
ie
r
ev
ie
w
s
,
p
r
o
d
u
ct
r
ev
ie
w
s
,
m
ar
k
e
t
r
ev
i
e
w
a
n
al
y
s
is
,
t
w
i
tter
tr
en
d
in
g
n
e
w
s
[
3
]
,
r
estau
r
an
t
r
ev
ie
w
an
a
l
y
s
is
[
4
]
,
ag
e,
g
en
d
er
s
en
t
i
m
e
n
t
clas
s
i
f
icatio
n
[
5
]
,
f
u
t
u
r
e
p
r
ed
ictio
n
s
li
k
e
elec
tio
n
p
r
ed
ictio
n
,
f
o
r
ec
ast
a
p
p
licatio
n
s
[
6
,
7
]
,
an
d
f
o
r
ec
asti
n
g
s
to
ck
m
ar
k
et
m
o
v
e
m
e
n
ts
[
8
]
.
Mu
lti
lin
g
u
al
t
w
ee
t
s
ca
n
b
e
s
ee
n
i
n
tr
en
d
in
g
n
e
w
s
i
n
t
h
e
t
w
itter
d
ata.
E
x
ten
d
in
g
t
h
e
s
e
n
ti
m
en
t
an
al
y
s
is
i
n
to
m
u
lt
ili
n
g
u
al
[
9
]
t
w
ee
t
s
is
v
er
y
es
s
en
tial
to
i
m
p
r
o
v
e
th
e
ac
cu
r
ac
y
in
s
en
t
i
m
e
n
t
an
al
y
s
is
w
it
h
o
u
t
l
o
s
in
g
a
n
y
u
s
er
t
w
ee
t
s
.
T
h
e
e
f
f
icien
c
y
o
f
t
h
is
an
al
y
s
is
w
as
i
m
p
r
o
v
ed
b
y
th
e
b
est
p
r
e
-
p
r
o
ce
s
s
in
g
tech
n
iq
u
e
s
[
1
0
]
.
I
n
cr
ea
s
in
g
th
e
ac
cu
r
ac
y
i
n
s
en
ti
m
en
t
a
n
al
y
s
is
is
a
r
esear
ch
c
h
alle
n
g
e
in
b
ig
d
ata.
B
y
r
ed
u
cin
g
th
e
d
ata
s
p
ar
s
it
y
an
d
tr
an
s
lat
in
g
t
h
e
m
u
lt
ili
n
g
u
al
t
w
ee
t
s
ac
cu
r
ac
y
ca
n
b
e
in
cr
ea
s
ed
[
7
]
.
I
n
th
e
s
en
ti
m
e
n
t
an
al
y
s
is
,
p
r
e
-
p
r
o
ce
s
s
in
g
p
la
y
s
an
i
m
p
o
r
tan
t
r
o
le
[
1
0
,
1
1
]
.
Mo
s
tl
y
s
e
n
ti
m
en
t
an
al
y
s
is
is
ap
p
lied
o
n
l
y
o
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
Mu
lti
-
lin
g
u
a
l tw
itter
s
en
timen
t
a
n
a
lysi
s
u
s
in
g
ma
ch
in
e
lea
r
n
in
g
(
K
.
A
r
u
n
)
5993
E
n
g
l
is
h
t
w
ee
ts
.
I
n
t
h
e
p
r
e
-
p
r
o
ce
s
s
i
n
g
le
v
el
n
o
n
-
E
n
g
l
is
h
t
w
e
ets
ar
e
a
v
o
id
ed
,
i.e
.
m
u
lt
ili
n
g
u
al
t
w
ee
ts
ar
e
n
o
t
co
n
s
id
er
ed
in
t
h
is
an
a
l
y
s
is
.
On
t
h
is
o
cc
asio
n
,
th
e
ac
c
u
r
ac
y
o
f
th
e
s
e
n
ti
m
e
n
t
a
n
al
y
s
i
s
g
ets
to
d
r
o
p
.
Ver
y
v
al
u
ab
le
t
w
ee
ts
an
d
o
p
in
io
n
s
o
f
th
e
p
u
b
lic
ar
e
n
o
t
co
n
s
id
er
e
d
h
er
e.
I
n
t
w
itter
s
en
ti
m
en
t
an
al
y
s
i
s
,
m
u
ltil
i
n
g
u
al
t
w
ee
t
s
m
u
s
t
b
e
tr
an
s
lated
in
to
E
n
g
li
s
h
t
w
ee
t
s
[
1
2
]
,
an
d
n
o
is
e
d
ata
lik
e
m
i
s
s
p
ellin
g
s
,
e
m
p
h
asized
w
o
r
d
s
,
co
d
e
-
m
i
x
ed
t
w
ee
t
s
n
ee
d
to
b
e
s
o
lv
ed
[
1
3
]
.
M
u
ltil
in
g
u
al
t
w
i
tter
s
e
n
ti
m
e
n
t
an
al
y
s
i
s
i
s
s
till
p
r
o
m
in
e
n
t
in
r
ea
s
er
ch
ar
ea
[
14
,
1
5
].
I
n
t
h
i
s
co
n
te
x
t
,
an
al
y
s
is
o
f
m
u
ltil
in
g
u
al
t
w
ee
ts
is
n
o
t
a
d
ir
ec
t
p
r
o
ce
s
s
.
T
h
e
f
ir
s
t
s
tep
in
t
h
is
s
eq
u
en
ce
is
th
at
it
r
eq
u
ir
es
a
tr
an
s
latio
n
f
r
o
m
n
o
n
-
E
n
g
l
is
h
t
w
ee
ts
to
t
h
e
E
n
g
li
s
h
la
n
g
u
ag
e
an
d
th
e
co
m
p
le
m
e
n
tar
y
s
t
ep
s
ar
e
ap
p
lied
f
o
r
th
e
s
e
n
ti
m
e
n
t
clas
s
i
f
icatio
n
.
Sp
ellin
g
m
is
ta
k
es,
e
m
p
h
a
s
ize
d
w
o
r
d
s
,
co
n
tr
ac
tio
n
w
o
r
d
s
ar
e
ca
lled
as
d
ata
s
p
ar
s
it
y
.
Data
s
p
ar
s
it
y
ca
n
b
e
r
ed
u
ce
d
in
p
r
e
-
p
r
o
ce
s
s
in
g
u
s
in
g
N
L
P
.
I
n
th
is
w
a
y
,
th
e
ac
cu
r
ac
y
o
f
th
e
clas
s
i
f
icatio
n
ca
n
b
e
i
m
p
r
o
v
ed
d
r
asti
ca
ll
y
t
h
r
o
u
g
h
m
u
ltil
i
n
g
u
a
l t
w
i
tter
s
en
ti
m
en
t a
n
al
y
s
is
[
1
6
,
1
7
].
I
n
th
i
s
p
ap
er
,
ML
T
SA
(
A
l
g
o
r
i
th
m
-
1
)
is
a
n
e
w
p
r
o
p
o
s
ed
alg
o
r
ith
m
,
w
h
ic
h
i
s
u
s
ed
to
d
ef
in
e
an
d
s
o
lv
e
th
e
d
ata
s
p
ar
s
it
y
b
y
u
s
in
g
e
f
f
i
cien
t
p
r
ep
r
o
ce
s
s
in
g
m
eth
o
d
s
a
n
d
tr
an
s
latio
n
f
r
o
m
n
o
n
-
E
n
g
l
i
s
h
to
E
n
g
l
is
h
is
b
y
u
s
i
n
g
Go
o
g
le
tr
a
n
s
lato
r
.
Fi
n
a
ll
y
,
t
h
e
s
e
n
ti
m
e
n
t
an
a
l
y
s
is
p
r
o
ce
s
s
h
a
s
b
ee
n
ap
p
lied
in
t
h
i
s
alg
o
r
it
h
m
.
I
n
t
h
is
p
r
o
p
o
s
ed
s
y
s
te
m
ac
cu
r
ac
y
is
i
n
cr
ea
s
ed
u
p
to
9
5
%.
S
en
ti
m
e
n
t
a
n
al
y
s
is
f
o
r
S
w
is
s
p
o
liti
cian
t
w
ee
t
s
;
in
th
i
s
s
t
u
d
y
t
w
ee
t
s
ar
e
av
a
ilab
le
i
n
m
u
l
tip
le
n
atio
n
a
l
lan
g
u
a
g
es
li
k
e
S
w
i
s
s
.
T
h
e
lan
g
u
a
g
e
w
as
tr
an
s
lated
f
r
o
m
S
w
i
s
s
to
E
n
g
lis
h
w
it
h
th
e
h
elp
o
f
m
ac
h
in
e
lear
n
in
g
an
d
a
lex
ico
n
-
b
ased
ap
p
r
o
ac
h
[
1
8
].
No
r
ah
Fah
ad
A
ls
h
a
m
m
ar
i
[
1
9
]
Sen
ti
m
e
n
t
an
a
l
y
s
is
cla
s
s
if
ica
tio
n
f
o
r
t
w
i
tter
d
ata
f
r
o
m
A
r
ab
ic
to
E
n
g
l
is
h
w
i
th
d
ee
p
lear
n
i
n
g
m
o
d
el.
A
co
m
p
lete
r
e
v
ie
w
o
f
A
r
ab
ic
s
e
n
ti
m
en
t
an
al
y
s
is
i
s
g
iv
e
n
b
y
A
b
d
u
lla
tif
Gh
allab
[
2
0
]
.
Md
.
A
l
-
Am
i
n
i
m
p
r
o
v
e
s
th
e
ac
cu
r
ac
y
o
f
th
e
B
en
g
ali
co
m
m
e
n
ts
u
p
to
7
5
.
5
%,
w
it
h
th
e
co
m
b
in
atio
n
o
f
w
o
r
d
2
v
ec
an
d
s
e
n
ti
m
e
n
t
ex
tr
a
ctio
n
o
f
w
o
r
d
s
.
T
h
is
ap
p
r
o
ac
h
p
er
f
o
r
m
a
n
ce
is
n
o
t
co
m
p
r
o
m
is
in
g
i
n
in
cr
ea
s
in
g
t
h
e
s
ize
o
f
t
h
e
d
ataset
[
2
1
].
Nar
r
,
P
r
esen
ts
m
u
ltil
i
n
g
u
a
l
t
w
ee
t
s
s
en
t
i
m
e
n
t
a
n
al
y
s
is
w
it
h
t
h
e
lan
g
u
a
g
e
in
d
ep
en
d
en
t c
lass
if
ica
tio
n
ap
p
r
o
ac
h
.
Her
e
f
o
u
r
la
n
g
u
a
g
es
w
e
r
e
tr
an
s
lated
u
s
i
n
g
th
e
A
m
az
o
n
Me
c
h
an
ical
T
u
r
k
s
u
c
h
as
E
n
g
lis
h
,
Ger
m
a
n
,
F
r
en
ch
,
a
n
d
P
o
r
tu
g
u
e
s
e.
L
a
n
g
u
a
g
e
t
w
ee
t
s
ar
e
in
d
ep
en
d
e
n
tl
y
tr
an
s
lated
an
d
class
i
f
ied
o
n
h
u
m
an
-
a
n
n
o
tat
e
d
t
w
ee
ts
w
it
h
g
o
o
d
ac
cu
r
ac
y
[
2
2
]
.
Nan
k
a
n
i
H.
,
Du
tta
H
f
o
cu
s
e
s
o
n
s
en
t
i
m
e
n
t
an
al
y
s
is
o
f
v
ar
io
u
s
l
o
w
r
e
s
o
u
r
ce
s
m
u
lt
i
-
lan
g
u
ag
e
s
[
2
3
]
h
av
in
g
li
m
ited
s
e
n
ti
m
e
n
t
an
al
y
s
i
s
r
eso
u
r
ce
s
s
u
ch
as
an
n
o
tated
d
ataset
s
,
w
o
r
d
e
m
b
e
d
d
in
g
s
a
n
d
s
en
ti
m
e
n
t le
x
ico
n
s
,
alo
n
g
w
it
h
th
e
E
n
g
lis
h
la
n
g
u
ag
e
.
2.
RE
S
E
ARCH
M
E
T
H
O
D
Mu
ltil
in
g
u
al
t
w
itter
s
e
n
ti
m
e
n
t
an
al
y
s
is
(
M
L
T
S
A
)
co
n
tain
s
t
h
e
f
o
llo
w
in
g
s
tep
s
.
-
P
r
e
-
p
r
o
ce
s
s
in
g
.
-
L
a
n
g
u
a
g
e
tr
an
s
latio
n
f
o
r
ea
ch
m
u
lti
-
la
n
g
u
a
g
e
t
w
ee
ts
.
T
h
ese
t
w
o
m
e
th
o
d
s
ar
e
v
er
y
c
l
ea
r
l
y
d
ef
i
n
ed
in
t
h
e
f
lo
w
c
h
ar
t
Fig
u
r
e
1
an
d
alg
o
r
ith
m
as
f
o
ll
o
w
s
.
Fig
u
r
e
1
.
Flo
w
c
h
ar
t f
o
r
m
u
lti
li
n
g
u
al
t
w
i
tter
s
en
ti
m
en
t
a
n
al
y
s
is
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
10
,
No
.
6
,
Dec
em
b
er
2
0
2
0
:
5
9
9
2
-
6
0
0
0
5994
A
l
g
o
r
ith
m
1
.
Mu
lt
ili
n
g
u
al
tw
i
t
ter
s
en
ti
m
e
n
t a
n
al
y
s
is
(
M
L
T
SA
)
ALGORITHM:
-
“Multilingual Twitter Sentiment Analysis (MLTSA)”
Purpose:
-
“Prediction of sentiment Analysis from the Multilingual and sparsity of tweets”.
Inp
ut:
-
Tweets ᴦ, Tweet_language
T_L, Sentiment_
Lexicon L
Output:
-
Tweet_Sentiment TS {P, Neg, NT}, Sentiment_Score SS
Where P: Positive, Neg: Negative, and Nt: Neutral.
Initialize:
-
POS,NEG, and NT=0,E=”English Language”
1.
While Tokenize Tweets ᴦ into Tweet(t
i
)
2.
Remove RT
, Numbers, URL, Hyperlinks from t
i
3.
if T_L(t
i
) ≠ English then
4.
Each t
i
∈
ᴦ is Tokenize into word set W
i
5.
//translating into English using Google Translator(GNMT)//
6.
t
i
=P(E
i
|W
i
)=
∏
=
1
(E
i
|E
0
,E
1
,E
3
…E
i
-
1
;W
1
,W
2
,W
3
…W
i
)
7.
Tokenize t
i
∈
ᴦ into word set W
i
8.
Remove stop words, punctuation symbols and special symbols from W
i
9.
if contraction words in W
i
then
10.
Replace with the complete word.
11.
else if emphasized words in W
i
then
12.
Replace wit
h the proper word.
13.
//Lexicon values from the Lexicon Dictionary for each word//
14.
Search for W
i
in L
15.
if W
i
∈
L.POS then
16.
POS
POS+L.
val
17.
else if W
i
∈
L.NEG then
18.
NEG
NEG+L
.val
19.
else
20.
NT
NT+1
21.
// Tweets Sentiments and Sentiments scores calculation //
22.
if POS>|NEG| then
23.
TS=P, SS=POS|(POS+NEG)
24.
else if POS<|NEG| then
25.
TS=Neg, SS=NEG|(POS+NEG)
26.
else
27.
TS=Nt
28.
End
2
.
1
.
P
re
-
pro
ce
s
s
ing
In
m
u
lti
-
li
n
g
u
al
t
w
i
tter
s
e
n
ti
m
en
t
an
al
y
s
i
s
,
p
r
e
-
p
r
o
ce
s
s
i
n
g
p
la
y
s
an
i
m
p
o
r
tan
t
k
e
y
r
o
le
.
T
h
e
p
r
e
-
p
r
o
ce
s
s
in
g
s
tep
is
u
s
ed
to
p
r
e
p
ar
e
th
e
r
a
w
d
ata
in
t
o
s
u
itab
le
d
ata
f
o
r
an
al
y
s
is
.
T
h
e
b
asic
m
eth
o
d
o
f
th
is
p
r
o
ce
s
s
is
clea
n
i
n
g
a
n
d
r
ep
lacin
g
t
h
e
r
eq
u
ir
ed
d
ata.
T
h
e
p
r
e
-
p
r
o
ce
s
s
in
g
m
et
h
o
d
d
ir
ec
tly
a
f
f
ec
ts
th
e
ac
cu
r
ac
y
an
d
e
f
f
icie
n
c
y
o
f
s
en
ti
m
e
n
t
a
n
al
y
s
i
s
.
Her
e
s
o
m
e
o
f
th
e
p
r
e
-
p
r
o
ce
s
s
i
n
g
s
tep
s
a
r
e
u
s
ed
f
o
r
th
e
b
e
s
t
an
al
y
s
is
.
In
th
e
F
ig
u
r
e
1
an
d
A
l
g
o
r
ith
m
1
d
ef
i
n
es
m
a
n
y
s
te
p
s
in
p
r
e
-
p
r
o
ce
s
s
i
n
g
.
-
R
e
m
o
v
e
“RT
”:
I
n
th
e
t
w
itt
er
d
ata
s
et
“RT
”
r
e
-
t
w
ee
ts
ar
e
co
m
m
o
n
,
b
u
t
it
is
n
o
t
r
eq
u
ir
ed
f
o
r
th
e
p
r
o
ce
s
s
in
g
,
s
o
it is
r
e
m
o
v
e
d
f
r
o
m
th
e
d
ata.
-
R
e
m
o
v
e
n
u
m
er
ic
v
a
lu
e
s
:
Nu
m
b
er
s
ar
e
n
o
t
r
eq
u
ir
ed
in
th
is
an
al
y
s
is
p
ar
t,
s
o
n
u
m
b
er
s
a
r
e
also
r
em
o
v
ed
f
r
o
m
t
h
e
t
w
ee
ts
i
n
t
h
e
d
ata
s
et.
-
R
e
m
o
v
e
I
s
o
lated
w
o
r
d
s
:
T
h
ese
w
o
r
d
s
h
a
v
e
n
o
s
i
g
n
i
f
ic
an
t
i
m
p
ac
t
o
n
s
en
t
i
m
e
n
t
a
n
al
y
s
i
s
.
Su
c
h
as
{i,
a,
y
}
w
h
ic
h
ar
e
s
in
g
le
c
h
ar
ac
ter
s
.
-
R
e
m
o
v
e
u
n
d
ef
i
n
ed
lan
g
u
a
g
e
t
w
ee
t
s
f
r
o
m
t
h
e
d
ata
s
et:
I
n
t
h
e
t
w
itter
d
ata
s
et,
ev
er
y
t
w
e
et
is
w
r
itte
n
in
a
s
p
ec
if
ic
la
n
g
u
a
g
e.
B
u
t
s
o
m
e
o
f
th
e
t
w
ee
t
s
ar
e
p
o
s
ted
in
th
e
u
n
d
e
f
in
ed
la
n
g
u
a
g
e,
th
i
s
t
y
p
e
o
f
t
w
ee
ts
w
il
l
n
o
t
s
u
p
p
o
r
t
f
o
r
t
h
e
la
n
g
u
a
g
e
tr
an
s
latio
n
s
a
n
d
s
e
n
ti
m
e
n
t
an
al
y
s
is
.
So
i
t
i
s
b
etter
to
a
v
o
id
u
n
d
ef
in
e
d
lan
g
u
a
g
e
t
w
ee
t
s
.
-
R
e
m
o
v
e
p
u
n
ct
u
atio
n
s
an
d
s
p
ec
ial
s
y
m
b
o
ls
:
P
u
n
ct
u
atio
n
s
y
m
b
o
l
s
d
o
n
o
t
af
f
ec
t
th
e
a
n
al
y
s
is
;
p
u
n
ct
u
atio
n
s
ar
e
r
em
o
v
ed
f
r
o
m
ea
c
h
an
d
ev
er
y
t
w
ee
t
.
T
h
e
s
y
m
b
o
ls
ar
e
[
!
"#
$
%
&
\
'
(
)
*
+,
-
.
/:
;<=
>
?
@
[
\
\
]
^_
`
{
|
}~
]
,
n
o
t
o
n
l
y
t
h
ese
s
y
m
b
o
ls
s
o
m
e
m
o
r
e
s
p
ec
ial
s
y
m
b
o
ls
ar
e
av
ai
lab
le
r
eg
u
lar
l
y
t
h
at
ar
e
u
s
ed
in
th
e
t
w
ee
ts
s
u
c
h
as [
:“
‘
’
…]
,
th
ese
ar
e
also
r
e
m
o
v
ed
f
r
o
m
th
e
t
w
ee
ts
.
-
R
e
m
o
v
e
h
y
p
er
lin
k
s
: H
y
p
er
li
n
k
s
d
o
n
o
t a
f
f
ec
t t
h
e
an
al
y
s
is
;
f
r
o
m
ea
c
h
t
w
ee
t,
h
y
p
er
lin
k
s
ar
e
r
em
o
v
ed
.
-
R
e
m
o
v
e
T
ag
s
:
th
er
e
ar
e
d
if
f
er
en
t
t
y
p
es
o
f
ta
g
s
u
s
ed
i
n
t
h
e
t
w
ee
t
s
s
u
c
h
a
s
@
,
a
n
d
&
.
T
h
is
w
ill
n
o
t
s
h
o
w
an
y
e
f
f
ec
t o
n
t
h
e
an
al
y
s
is
.
-
R
e
m
o
v
e
Sto
p
w
o
r
d
s
:
T
h
er
e
is
On
e
H
u
n
d
r
ed
an
d
Sev
e
n
t
y
-
N
in
e
(
179
)
lis
t
o
f
E
n
g
lis
h
la
n
g
u
ag
e
s
to
p
w
o
r
d
s
th
at
w
er
e
s
u
p
p
o
r
ted
in
t
h
e
N
L
T
K.
A
ll t
h
ese
ar
e
u
s
ed
in
r
e
g
u
lar
E
n
g
li
s
h
s
e
n
te
n
ce
s
,
b
u
t
t
h
e
r
e
is
n
o
e
f
f
ec
t i
n
th
e
s
en
t
i
m
e
n
t
a
n
al
y
s
i
s
s
o
,
i
t
is
b
etter
to
r
e
m
o
v
e
th
e
s
t
o
p
w
o
r
d
s
.
I
n
t
h
e
s
to
p
w
o
r
d
lis
t
,
s
o
m
e
o
f
th
e
n
e
g
ati
v
e
w
o
r
d
s
s
u
c
h
as
{
'n
o
r
'
,
'
n
o
t
'
,
}
h
as
an
e
f
f
ec
t
i
n
th
e
s
en
ti
m
e
n
t
clas
s
i
f
icatio
n
,
s
o
t
h
ese
w
o
r
d
s
m
u
s
t
b
e
ex
clu
d
ed
f
r
o
m
t
h
e
s
to
p
w
o
r
d
lis
t
,
o
th
er
w
is
e
n
eg
at
iv
e
t
w
ee
t
s
w
ill b
e
class
if
ied
i
n
to
p
o
s
iti
v
e.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
Mu
lti
-
lin
g
u
a
l tw
itter
s
en
timen
t
a
n
a
lysi
s
u
s
in
g
ma
ch
in
e
lea
r
n
in
g
(
K
.
A
r
u
n
)
5995
-
R
ep
lace
th
e
co
n
tr
ac
tio
n
w
o
r
d
s
:
C
o
n
tr
ac
tio
n
w
o
r
d
s
ar
e
co
m
m
o
n
l
y
u
s
ed
i
n
t
h
e
t
w
ee
ts
.
T
h
e
co
n
tr
ac
tio
n
w
o
r
d
s
ar
e
{I
’
ll,
d
o
n
’
t,
I
’
m
,
w
o
n
’
t,
t
h
i
s
’
l
l,
etc}
.
B
u
t
th
is
t
y
p
e
o
f
co
n
tr
ac
tio
n
w
o
r
d
s
ar
e
n
o
t
d
ir
ec
tl
y
p
r
o
ce
s
s
ed
,
it n
ee
d
s
to
b
e
r
ep
la
ce
d
w
it
h
th
e
p
r
o
p
er
w
o
r
d
s
in
s
t
ea
d
o
f
co
n
tr
ac
tio
n
w
o
r
d
s
.
L
i
k
e
{I
’
ll
I
w
ill,
d
o
n
’
t
d
o
n
o
t,
I
’
m
I
a
m
,
w
o
n
’
t
w
o
u
l
d
n
o
t,
th
is
’
ll
t
h
is
w
i
ll,
.
.
etc}
-
Sp
ellin
g
co
r
r
ec
tio
n
s
:
Mo
r
e
t
h
an
1
0
%
o
f
t
w
ee
t
s
co
n
tai
n
m
i
s
s
p
elli
n
g
s
,
w
h
ich
d
ec
r
ea
s
e
s
t
h
e
ef
f
icie
n
c
y
o
f
class
i
f
icatio
n
[
2
4
]
.
Her
e
ev
er
y
t
w
ee
t
a
n
d
ev
er
y
w
o
r
d
in
t
h
e
t
w
ee
t
is
v
er
i
f
ied
.
I
f
an
y
s
p
ellin
g
m
i
s
ta
k
e
s
o
cc
u
r
,
th
at
m
is
s
p
ell
in
g
s
ar
e
co
r
r
ec
ted
b
y
u
s
in
g
d
ictio
n
ar
y
-
b
a
s
ed
w
h
ic
h
is
p
r
o
p
o
s
ed
b
y
P
eter
No
r
v
ig
[
9
].
-
R
ep
lace
t
h
e
e
m
p
h
a
s
ized
w
o
r
d
s
:
E
m
p
h
asized
w
o
r
d
s
ar
e
u
s
ed
f
o
r
e
x
p
r
ess
i
v
e
w
o
r
d
s
an
d
i
t
i
s
v
er
y
co
m
m
o
n
to
e
x
p
r
ess
t
h
e
f
ee
li
n
g
s
i
n
t
h
e
te
x
t
f
o
r
m
.
T
h
ese
w
o
r
d
s
ar
e
n
o
t
s
u
p
p
o
r
ted
f
r
o
m
a
n
y
d
ictio
n
ar
y
,
w
it
h
t
h
e
h
elp
o
f
NL
T
K,
r
eg
u
l
ar
ex
p
r
ess
io
n
s
an
d
s
p
ellin
g
co
r
r
ec
ti
o
n
s
ca
n
b
e
r
ep
lace
d
w
it
h
p
r
o
p
er
w
o
r
d
s
,
{‘
co
o
o
o
o
o
o
o
l’
’
co
o
l
’
,
’
p
er
r
r
r
r
r
r
r
r
r
r
f
f
f
f
f
ec
t
’
’
p
er
f
ec
t’
,
’
g
o
o
o
o
o
o
d
’
’
g
o
o
d
’
,
’
h
o
o
o
o
o
t’
’
h
o
t’
,
’
to
o
o
o
o
o
’
’
to
o
’
}.
2
.
2
.
L
a
ng
ua
g
e
t
ra
ns
la
t
io
n
L
a
n
g
u
a
g
e
tr
an
s
latio
n
i
s
u
s
e
d
to
tr
an
s
la
te
n
o
n
-
E
n
g
li
s
h
lan
g
u
a
g
e
t
w
ee
ts
in
to
E
n
g
lis
h
t
w
ee
ts
.
T
h
e
lan
g
u
a
g
e
o
f
t
h
e
t
w
ee
ts
is
s
p
ec
if
ied
in
t
h
e
"
lan
g
"
attr
ib
u
te
alo
n
g
w
it
h
th
e
"
tex
t"
.
T
h
r
ee
t
y
p
es
o
f
lab
els
ar
e
d
ef
in
i
n
g
t
h
e
la
n
g
u
a
g
e
o
f
t
w
e
ets.
T
h
e
lab
els
ar
e
1
)
E
n
g
lis
h
,
2
)
No
n
-
E
n
g
l
is
h
,
3
)
U
n
d
ef
i
n
ed
.
E
x
a
m
p
le
"
en
"
,
"
h
i"
,
"
u
n
d
"
in
th
e
F
i
g
u
r
e
1
a
n
d
A
lg
o
r
it
h
m
1
ex
p
lai
n
s
t
h
e
tr
an
s
latio
n
p
r
o
ce
d
u
r
e.
I
f
th
e
’
lan
g
'
t
y
p
e
i
s
'
u
n
d
'
,
T
h
ese
t
y
p
es o
f
t
w
ee
ts
p
r
o
v
id
e
h
y
p
er
li
n
k
s
,
v
id
eo
s
,
s
tic
k
er
s
,
or
an
y
g
r
ap
h
ic
s
b
u
t
n
o
t t
h
e
tex
t
u
al
in
f
o
r
m
atio
n
an
d
th
er
e
is
n
o
u
s
e
i
n
th
e
s
en
t
i
m
e
n
t a
n
al
y
s
is
p
r
o
ce
s
s
.
I
f
th
e
la
n
g
u
a
g
e
is
in
t
h
e
'
No
n
-
E
n
g
li
s
h
'
t
y
p
e,
it
r
eq
u
ir
e
s
E
n
g
lis
h
tr
a
n
s
lat
io
n
.
A
cc
o
r
d
in
g
to
th
e
g
i
v
e
n
alg
o
r
ith
m
(
M
L
T
S
A
)
,
A
lg
o
r
it
h
m
1
u
s
es
o
n
e
m
o
s
t
Go
o
g
le
tr
an
s
lato
r
t
h
at
i
s
G
NM
T
(
g
o
o
g
le
n
eu
r
al
m
ac
h
i
n
e
tr
an
s
lato
r
)
.
T
h
is
GNM
T
[
2
5
-
28
]
u
s
e
s
t
h
e
L
ST
M
R
N
N
(
lo
n
g
-
s
h
o
r
t
ter
m
m
e
m
o
r
y
r
ec
u
r
r
en
t
n
e
u
r
al
n
et
w
o
r
k
s
)
[
29
]
.
T
h
is
GNM
T
ar
ch
itect
u
r
e
d
iv
id
es
i
n
to
f
o
u
r
p
ar
ts
1
)
Dec
o
d
er
L
ST
Ms,
2
)
E
n
co
d
er
L
ST
Ms,
3
)
A
tte
n
tio
n
,
4
)
So
f
t
m
ax
.
E
n
co
d
er
L
ST
Ms
en
co
d
es
th
e
g
i
v
e
n
in
p
u
t
s
et
o
f
w
o
r
d
s
f
r
o
m
t
h
e
s
o
u
r
ce
s
en
te
n
ce
i
n
to
f
i
x
ed
-
s
ized
v
ec
to
r
s
i
n
it
s
h
id
d
en
la
y
er
s
,
Dec
o
d
er
L
ST
Ms
an
d
So
f
t
m
a
x
ca
n
d
ec
o
d
e
in
to
tar
g
et
la
n
g
u
a
g
e
w
o
r
d
s
s
y
m
b
o
ls
,
a
n
d
atte
n
tio
n
p
r
o
v
i
d
es
th
e
g
e
n
er
ated
tar
g
et
w
o
r
d
s
et
an
d
co
m
m
u
n
icatio
n
b
e
t
w
ee
n
e
n
co
d
er
an
d
d
ec
o
d
er
h
id
d
en
lay
er
s
.
I
n
th
i
s
p
r
o
ce
s
s
,
T
w
itter
d
atase
t
ᴦ
is
to
k
en
ized
i
n
to
s
in
g
le
t
wee
ts
ti
,
an
d
ea
ch
t
w
ee
t
h
a
s
its
lan
g
u
a
g
e,
t
h
e
L
a
n
g
u
a
g
e
o
f
ti
is
'
en
g
'
o
r
'u
n
d
'
n
o
n
ee
d
o
f
tr
a
n
s
lat
io
n
,
o
th
er
w
is
e,
tr
a
n
s
la
tio
n
i
s
n
ee
d
ed
f
o
r
an
y
s
o
u
r
ce
lan
g
u
a
g
e
in
to
E
n
g
l
is
h
o
n
l
y
.
I
n
th
i
s
p
r
o
ce
s
s
ti to
k
e
n
ized
in
to
w
o
r
d
s
W
i.
V1
,
V2
,
V3
,
.
.
.
.
.
Vn
=
E
n
co
d
er
L
ST
M(
W
1
,
W
2
,
W
3
,
.
.
.
.
.
W
n
)
(
1
)
w
h
er
e
Vn
i
s
th
e
w
o
r
d
s
y
m
b
o
l
v
ec
to
r
is
th
e
o
u
tp
u
t o
f
t
h
e
e
n
c
o
d
er
L
ST
M
h
id
d
en
lay
er
s
.
T
h
en
d
ec
o
d
e
r
L
MST
an
d
So
f
t
m
ax
m
o
d
u
le
s
ca
n
m
a
k
e
it
in
to
E
n
g
li
s
h
w
o
r
d
v
ec
to
r
b
y
u
s
in
g
th
e
co
n
d
itio
n
a
l p
r
o
b
ab
ilit
y
m
o
d
el.
P
(
E
|
V)
=
P
(
E
|
V1
,
V2
,
V3
,
.
.
.
.
.
Vn
)
(
2
)
=
∏
P
n
i
=
1
(
E
i|
E
0
,
E
1
,
E
3
…E
i
-
1
;
V1
,
V2
,
V3
,
.
.
.
.
.
Vn
)
(
3
)
w
h
er
e
E
0
is
th
e
s
p
ec
ial
s
y
m
b
o
l
as
th
e
b
eg
in
n
in
g
o
f
t
h
e
s
en
te
n
ce
in
a
n
y
tar
g
et
lan
g
u
ag
e.
E
i
tr
an
s
lated
w
o
r
d
v
ec
to
r
.
T
h
is
s
y
s
te
m
ca
n
d
o
th
e
tr
a
n
s
la
tio
n
tas
k
f
o
r
th
e
v
alid
i
n
p
u
ts
.
I
f
W
i is
a
n
o
n
-
E
n
g
lis
h
w
o
r
d
an
d
w
h
ich
i
s
d
ef
in
ed
in
a
d
ictio
n
ar
y
o
f
a
l
an
g
u
a
g
e
,
th
at
w
o
r
d
(
s
)
w
il
l
go
f
o
r
tr
an
s
latio
n
.
B
u
t
i
n
o
th
er
ca
s
es
w
h
er
e
W
i
is
n
o
n
-
E
n
g
li
s
h
b
u
t le
x
icall
y
E
n
g
l
is
h
,
tr
an
s
latio
n
is
n
o
t p
o
s
s
ib
le.
3.
RE
SU
L
T
S
3
.
1
.
Da
t
a
s
et
T
h
e
d
ata
s
et
is
co
llec
ted
f
r
o
m
th
e
T
w
itter
A
P
I
.
T
h
e
to
p
ic
th
at
h
as
b
ee
n
ac
ce
s
s
ed
f
r
o
m
T
w
it
ter
li
v
e
st
r
ea
m
is
t
h
e
h
o
n
o
u
r
ab
le
C
M
,
h
e
is
th
e
An
d
h
r
a
P
r
ad
esh
h
o
n
o
u
r
ab
le
C
M,
an
d
2
0
0
tw
ee
ts
w
er
e
co
llected
.
T
w
ee
t
s
ar
e
in
lo
ca
l
la
n
g
u
a
g
es
,
an
d
m
o
s
tl
y
in
E
n
g
lis
h
.
T
h
e
r
est
o
f
t
h
e
t
w
ee
t
s
ar
e
i
n
I
n
d
ia
n
lo
ca
l
la
n
g
u
a
g
es
s
u
c
h
as Hi
n
d
i,
an
d
T
elu
g
u
.
w
h
ich
is
i
n
t
h
e
F
i
g
u
r
e
2.
I
n
itiall
y
,
ea
ch
t
w
ee
t
co
n
tain
s
6
9
attr
ib
u
tes,
s
u
ch
a
s
u
s
ed
_
id
,
u
s
er
_
s
tr
_
id
,
tex
t,
lan
g
u
a
g
e,
co
u
n
tr
y
,
o
u
t
o
f
th
i
s
p
r
o
ce
s
s
it
r
eq
u
ir
es
o
n
l
y
t
w
o
attr
ib
u
te
s
{tex
t,la
n
g
}.
Her
e
tex
t
co
n
tain
s
t
w
ee
t
s
,
r
e
-
t
w
ee
t
s
an
d
lan
g
is
r
ef
er
r
ed
to
as th
e
lan
g
u
a
g
e
o
f
e
ac
h
t
w
ee
t.
T
h
e
d
i
m
en
s
io
n
al
it
y
o
f
t
h
e
d
ata
s
et
w
as r
ed
u
ce
d
u
p
to
t
w
o
f
ea
tu
r
es.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
10
,
No
.
6
,
Dec
em
b
er
2
0
2
0
:
5
9
9
2
-
6
0
0
0
5996
On
e
o
f
th
e
n
e
w
f
ea
t
u
r
e
ad
d
ed
to
th
e
d
ata
s
et
is
"
s
en
ti
m
en
t"
as
a
t
h
ir
d
f
ea
t
u
r
e.
T
h
is
s
en
t
i
m
e
n
t
attr
ib
u
te
co
n
tai
n
s
th
r
ee
lab
els
i.e
.
"
p
o
s
itiv
e,
n
e
g
ati
v
e,
n
e
u
tr
a
l"
.
A
n
a
u
to
lab
el
is
as
s
i
g
n
ed
w
it
h
t
h
e
s
u
p
p
o
r
t
o
f
a
lex
ico
n
-
b
ased
ap
p
r
o
ac
h
.
T
r
an
s
latio
n
o
f
d
i
f
f
er
e
n
t
la
n
g
u
a
g
es
in
to
E
n
g
lis
h
i
s
d
o
n
e
u
s
i
n
g
a
p
y
t
h
o
n
p
r
o
g
r
am
m
i
n
g
lan
g
u
ag
e
w
it
h
NL
P
s
u
p
p
o
r
t.
Fin
all
y
d
ata
s
et
co
n
tain
s
t
h
r
ee
attr
ib
u
te
s
.
I
n
F
ig
u
r
e
3
,
6
5
-
8
0
%
o
f
t
w
ee
t
s
ar
e
in
th
e
E
n
g
lis
h
la
n
g
u
ag
e
(
en
)
,
a
n
d
r
e
m
ain
in
g
ar
e
in
t
h
e
lo
ca
l
la
n
g
u
a
g
e
[
te:
T
elu
g
u
,
h
i:
Hi
n
d
i,
b
e:
B
en
g
ali,
ar
:
A
r
ab
ic,
u
r
:
Ur
d
u
]
alo
n
g
w
it
h
t
h
e
u
n
d
ef
i
n
ed
(
u
n
d
)
lan
g
u
ag
e.
Un
d
e
f
in
ed
lan
g
u
ag
e
t
w
ee
t
s
ar
e
n
o
t
co
n
s
id
er
ed
in
th
i
s
p
r
o
ce
s
s
.
Fig
u
r
e
2
.
Sa
m
p
le
d
ata
s
et
f
r
o
m
t
w
itter
li
v
e
Fig
u
r
e
3
.
Mu
ltil
i
n
g
u
al
t
w
ee
ts
[
E
n
g
l
is
h
,
T
elu
g
u
,
Hi
n
d
i,
u
n
d
]
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
Mu
lti
-
lin
g
u
a
l tw
itter
s
en
timen
t
a
n
a
lysi
s
u
s
in
g
ma
ch
in
e
lea
r
n
in
g
(
K
.
A
r
u
n
)
5997
3
.
2
.
P
re
-
pro
ce
s
s
ing
T
h
e
p
r
e
p
r
o
ce
s
s
in
g
m
eth
o
d
co
n
tain
s
1
1
s
tep
s
as
m
en
t
io
n
e
d
ab
o
v
e.
T
h
e
d
ata
s
et
co
n
tain
s
a
h
u
g
e
a
m
o
u
n
t
o
f
n
o
is
e
d
ata,
ev
e
n
th
o
u
g
h
th
e
at
tr
ib
u
tes
w
er
e
r
ed
u
ce
d
u
p
to
th
r
ee
.
T
h
e
n
o
is
e
d
a
ta
m
u
s
t
b
e
clea
r
ed
b
ef
o
r
e
an
al
y
zi
n
g
it.
P
y
th
o
n
lan
g
u
a
g
e
co
m
p
letel
y
s
u
p
p
o
r
ts
t
h
e
p
r
e
-
p
r
o
ce
s
s
i
n
g
b
ec
au
s
e
o
f
its
r
ich
s
et
o
f
p
ac
k
ag
es.
UR
L
's,
n
u
m
b
er
s
,
H
T
ML
tag
s
,
s
to
p
w
o
r
d
s
,
p
u
n
ct
u
atio
n
s
y
m
b
o
l
s
,
s
p
ec
ial
s
y
m
b
o
ls
,
an
d
e
m
o
tico
n
s
w
er
e
r
e
m
o
v
ed
f
r
o
m
ea
c
h
an
d
ev
er
y
t
w
ee
t.
A
d
d
itio
n
all
y
,
th
e
p
r
e
-
p
r
o
ce
s
s
i
n
g
m
o
d
u
le
r
e
p
lace
s
co
n
tr
ac
tio
n
w
o
r
d
s
w
it
h
ac
tu
a
l
w
o
r
d
s
,
N
eg
atio
n
w
o
r
d
s
w
it
h
co
m
p
let
e
w
o
r
d
an
d
m
i
s
s
p
elli
n
g
w
o
r
d
s
w
er
e
r
ec
ti
f
ied
.
Fin
all
y
,
t
h
e
t
w
o
d
ata
s
e
ts
w
er
e
p
r
e
-
p
r
o
ce
s
s
ed
v
er
y
ef
f
ec
ti
v
el
y
.
Sa
m
p
le
p
r
e
-
p
r
o
ce
s
s
i
n
g
s
tep
s
:
I
n
p
u
t
-
1
: RT
@
r
av
i
:
A
p
p
le
p
h
o
n
e
it is
v
er
r
r
r
y
co
s
t
llll
ll
llll
lll
llll
llll
lll
y
Ou
tp
u
t
-
1
: a
p
p
le
p
h
o
n
e
co
s
tl
y
.
E
m
p
h
as
ized
w
o
r
d
s
an
d
s
p
elli
n
g
co
r
r
ec
tio
n
s
in
t
h
e
t
w
ee
t
s
ar
e
v
er
r
r
r
r
y
,
co
s
tlll
llll
ll
llll
ll
y
.
T
h
ese
w
o
r
d
s
ca
n
r
ed
u
ce
th
e
le
n
g
th
li
k
e
v
er
y
,
co
s
tl
y
b
y
u
s
in
g
s
o
m
e
o
f
t
h
e
r
eg
u
lar
e
x
p
r
ess
io
n
s
.
T
h
e
r
em
ai
n
i
n
g
co
r
r
ec
tio
n
s
d
ep
en
d
o
n
th
e
s
p
ell
in
g
co
r
r
ec
tio
n
s
.
3
.
3
.
L
a
ng
ua
g
e
t
ra
ns
la
t
io
n
A
cc
o
r
d
in
g
to
th
e
s
p
ec
if
ied
alg
o
r
ith
m
M
L
T
S
A
,
d
if
f
er
en
t
la
n
g
u
a
g
e
(
n
o
n
-
E
n
g
lis
h
)
t
w
ee
t
s
ar
e
tr
an
s
lated
.
T
h
is
tr
an
s
latio
n
i
s
s
u
p
p
o
r
ted
b
y
o
n
e
o
f
t
h
e
p
y
t
h
o
n
m
o
d
u
le
s
,
an
d
th
at
m
o
d
u
le
d
ep
en
d
s
o
n
th
e
Go
o
g
le
tr
an
s
lato
r
.
I
n
th
i
s
a
lg
o
r
ith
m
,
it c
o
n
tai
n
s
3
ca
s
es.
3
.
3
.
1
.
T
ra
ns
la
t
ing
no
n
-
E
ng
li
s
h into
E
ng
li
s
h
T
w
ee
t
i
n
Hi
n
d
i:
पुल
वा
म
ा
ह
म
ल
े
पर
बो
ल
े
भा
र
त
ी
य
बैड
म
म
ट
न
ट
ी
म
के
र
ा
ी
य
क
ो
च
पु
ल
े
ल
ा
गो
पी
चद
पा
क
के
स
ा
थ
क
क
स
ी
त
र
ह
के
T
r
an
s
lated
to
E
n
g
lis
h
:
P
u
llela
Go
p
ich
an
d
,
n
atio
n
al
co
ac
h
o
f
th
e
I
n
d
ia
n
b
ad
m
in
to
n
tea
m
,
s
p
o
k
e
o
n
t
he
P
u
l
w
a
m
a
attac
k
.
T
w
ee
t
in
B
en
g
ali:
R
T
@
I
a
m
So
u
r
av
_
b
:
যখন
আ
জ
ক
া
লকার
জ
ী
বন
ম
ু
খ
ী
বা
ল
া
চ
লি
চ
ি
ত
্
র
বা
গ
া
ন
ে
গ
া
ল
ি
গ
া
লা
জ
ব্যব
হ
া
র
ক
র
া
হ
য়
ত
খ
ন
তো
@
M
y
An
an
d
aB
az
ar
এর
গ
া
য়ে
ল
া
গ
ে
ন
া
দ
ে
খ
ি
|
ক
ি
ন
্ত
ু
…
T
r
an
s
lated
to
E
n
g
lis
h
:
W
h
e
n
w
e
ar
e
n
o
w
u
s
i
n
g
th
e
l
y
r
ic
s
o
f
th
e
li
f
e
-
c
h
a
n
g
in
g
B
an
g
la
f
il
m
o
r
s
o
n
g
,
w
e
d
o
n
o
t see
it
T
w
ee
t i
n
T
elu
g
u
an
d
T
r
an
s
lati
o
n
:
3
.
3
.
2
.
T
ra
ns
la
t
ing
m
i
x
e
d la
n
g
ua
g
e
w
o
rds
into
E
ng
lis
h
T
r
an
s
latin
g
i
n
to
E
n
g
lis
h
:
C
o
u
n
tr
y
u
n
i
v
er
s
al
elec
tio
n
n
o
is
e
t
h
er
e
f
ar
m
i
n
g
at
tach
ed
m
i
n
e
th
e
p
r
o
b
lem
s
j
u
d
iciar
y
h
is
m
aj
esti
ca
ll
y
li
g
h
t
lo
s
in
g
Var
u
k
ir
ata.
3
.
3
.
3
.
Co
de
-
m
i
x
ed
w
o
rds
A
t
w
ee
t
i
s
t
y
p
ed
in
o
n
e
lan
g
u
ag
e
b
u
t
t
h
e
m
ea
n
i
n
g
is
in
o
t
h
er
lan
g
u
a
g
es,
t
h
is
t
y
p
e
o
f
t
w
ee
t
i
s
n
o
t
g
etti
n
g
tr
an
s
lated
(
C
o
d
e
m
i
x
e
d
)
.
T
w
ee
t:
“An
t
h
a
R
aj
an
n
a
m
ah
i
m
a
A
p
p
atlo
an
n
a
g
ar
in
i
d
e
v
u
d
u
n
i
ela
c
h
esar
o
ip
u
d
u
y
s
r
n
i
d
ev
u
d
u
ch
e
y
ad
am
c
h
u
s
t
h
u
n
te
ar
d
am
aip
o
t
h
”.
T
h
is
t
w
ee
t
i
s
w
r
it
ten
i
n
E
n
g
lis
h
,
b
u
t
ea
ch
an
d
ev
er
y
m
ea
n
in
g
o
f
t
h
e
w
o
r
d
is
r
elate
d
to
T
elu
g
u
lan
g
u
a
g
e.
So
t
h
is
t
y
p
e
o
f
t
w
ee
t i
s
n
o
t c
o
n
v
er
ted
.
3
.
4
.
T
ra
ini
ng
a
nd
t
esting
B
ag
-
of
-
w
o
r
d
s
v
ec
to
r
is
cr
ea
t
ed
af
ter
th
e
p
r
e
-
p
r
o
ce
s
s
in
g
an
d
tr
an
s
latio
n
s
b
y
u
s
in
g
t
h
e
co
u
n
ter
v
ec
to
r
izatio
n
.
Data
s
et
s
p
lits
i
n
to
7
:3
r
atio
o
u
t
o
f
th
at
7
0
%
is
f
o
r
th
e
tr
ai
n
d
ata,
an
d
3
0
%
is
f
o
r
th
e
tes
t
d
ata.
B
y
u
s
i
n
g
m
ac
h
i
n
e
lear
n
i
n
g
alg
o
r
ith
m
s
[
3
0
]
,
th
e
tr
ai
n
i
n
g
an
d
te
s
ti
n
g
p
r
o
ce
s
s
i
s
i
m
p
l
e
m
en
ted
.
T
h
e
ML
alg
o
r
ith
m
s
ar
e
-
Mu
lti
n
o
m
ial
Naï
v
e
B
a
y
es
(
M
NB
)
,
-
L
o
g
i
s
tic
R
eg
r
es
s
io
n
(
L
R
)
,
-
Su
p
p
o
r
t V
ec
to
r
Ma
ch
in
es (
SV
M)
,
-
Dec
is
io
n
T
r
ee
(
D
T
)
,
-
k
-
Nea
r
es
t N
ei
g
h
b
o
u
r
(
k
NN)
,
-
R
an
d
o
m
Fo
r
est (
R
F).
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
10
,
No
.
6
,
Dec
em
b
er
2
0
2
0
:
5
9
9
2
-
6
0
0
0
5998
P
er
f
o
r
m
a
n
ce
m
ea
s
u
r
es a
r
e
co
m
p
u
ted
fo
r
all
th
e
al
g
o
r
ith
m
s
.
P
er
f
o
r
m
a
n
ce
m
ea
s
u
r
es
s
u
c
h
as
p
r
ec
is
io
n
,
r
ec
all,
f
1
-
s
co
r
es.
F
i
g
u
r
e
4
to
F
ig
u
r
e
6
ar
e
u
s
ed
to
r
ep
r
esen
t t
h
e
s
e
n
ti
m
en
t c
lass
if
icatio
n
w
i
t
h
p
r
e
-
p
r
o
ce
s
s
es a
n
d
ML
T
SA
al
g
o
r
it
h
m
.
Fig
u
r
e
4
.
P
r
ec
is
io
n
in
p
re
-
p
r
o
ce
s
s
v
s
.
ML
T
S
A
clas
s
i
f
icatio
n
Fig
u
r
e
5
.
R
ec
all
in
p
re
-
p
r
o
ce
s
s
v
s
.
M
L
T
SA
cla
s
s
i
f
ica
tio
n
Fig
u
r
e
6
.
F1
-
Sco
r
e
in
p
re
-
p
r
o
ce
s
s
v
s
.
ML
T
S
A
clas
s
i
f
icatio
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
Mu
lti
-
lin
g
u
a
l tw
itter
s
en
timen
t
a
n
a
lysi
s
u
s
in
g
ma
ch
in
e
lea
r
n
in
g
(
K
.
A
r
u
n
)
5999
I
n
t
h
e
F
ig
u
r
e
s
4
-
6
t
h
e
r
es
u
lt
s
s
h
o
w
th
e
co
m
p
ar
is
o
n
b
et
wee
n
p
r
e
-
p
r
o
ce
s
s
ed
t
w
ee
ts
an
d
ML
T
S
A
alg
o
r
ith
m
.
I
n
all
t
h
e
ca
s
e
s
,
M
L
T
SA
lead
s
to
g
o
o
d
an
d
ac
c
u
r
ate
r
es
u
lt
s
.
P
r
ec
is
io
n
,
R
ec
al
l
an
d
F1
-
s
co
r
es
ar
e
th
e
av
er
ag
e
s
f
o
r
th
e
p
o
s
itiv
e,
n
eg
at
iv
e,
n
e
u
tr
als.
M
L
T
SA
li
n
e
is
in
cr
ea
s
ed
ab
o
v
e
th
e
lin
e
o
f
p
r
e
-
p
r
o
ce
s
s
in
g
.
T
r
ain
in
g
a
n
d
te
s
ti
n
g
s
co
r
e
ac
cu
r
ac
y
in
m
ac
h
i
n
e
lear
n
i
n
g
a
lg
o
r
ith
m
s
f
o
r
SVM
(
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
i
n
e)
i
s
9
5
% a
n
d
f
o
r
R
F (
r
an
d
o
m
f
o
r
e
s
t)
,
it is
9
3
%.
4.
CO
NCLU
SI
O
N
AND
F
U
T
U
RE
WO
RK
I
n
th
is
p
ap
er
,
s
e
n
ti
m
en
ts
ar
e
ex
tr
ac
ted
f
r
o
m
t
h
e
m
u
ltil
i
n
g
u
al
t
w
ee
ts
.
T
h
i
s
al
g
o
r
i
th
m
p
r
o
v
id
e
s
s
u
f
f
icie
n
t
p
r
e
-
p
r
o
ce
s
s
i
n
g
tech
n
iq
u
e
s
th
a
t
w
er
e
ap
p
lied
to
s
et
th
e
d
ataset.
Her
e
to
tall
y
1
1
p
r
e
-
p
r
o
ce
s
s
in
g
tech
n
iq
u
es
w
er
e
i
m
p
le
m
e
n
ted
to
i
m
p
r
o
v
e
t
h
e
ac
c
u
r
ac
y
i
n
t
h
e
s
e
n
ti
m
e
n
t
a
n
al
y
s
is
.
Mu
l
ti
-
l
in
g
u
al
t
w
ee
ts
d
ea
l
w
it
h
m
u
ltip
le
n
u
m
b
er
s
o
f
in
t
er
n
a
tio
n
al
o
r
lo
ca
l
lan
g
u
a
g
es
.
P
y
th
o
n
lan
g
u
ag
e
tr
a
n
s
lato
r
m
o
d
u
le,
w
h
ic
h
i
s
s
u
p
p
o
r
ted
b
y
t
h
e
Go
o
g
le
tr
a
n
s
lato
r
,
is
u
s
ed
to
tr
an
s
late
th
e
n
o
E
n
g
lis
h
t
w
ee
ts
i
n
to
E
n
g
l
is
h
,
an
d
t
h
e
n
s
en
ti
m
e
n
ts
w
er
e
ex
tr
ac
ted
f
r
o
m
th
e
E
n
g
l
is
h
d
ata.
ML
T
SA
alg
o
r
ith
m
is
a
b
etter
p
r
e
-
pr
o
c
ess
i
n
g
a
n
d
lan
g
u
a
g
e
tr
an
s
latio
n
tech
n
iq
u
e.
Ma
ch
i
n
e
lear
n
i
n
g
al
g
o
r
ith
m
s
ar
e
ap
p
lied
in
th
e
tr
ain
ed
an
d
te
s
t
d
ata.
Sen
ti
m
e
n
t
class
i
f
icatio
n
s
ar
e
i
m
p
r
o
v
ed
o
n
t
h
e
tr
a
n
s
lated
d
ata.
Ma
c
h
i
n
e
lear
n
i
n
g
al
g
o
r
ith
m
s
ar
e
Mu
l
ti
n
o
m
ial
n
a
iv
e
b
a
y
es
(
MN
B
)
,
l
o
g
is
tic
r
eg
r
es
s
io
n
(
L
R
)
,
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
i
n
e
(
SVM)
,
d
ec
is
io
n
tr
ee
(
DT
)
,
K
-
n
ea
r
est
n
e
ig
h
b
o
r
(
KNN)
,
an
d
r
an
d
o
m
f
o
r
est
(
R
F)
ap
p
lied
.
T
h
e
p
er
f
o
r
m
a
n
ce
m
ea
s
u
r
es
in
m
ac
h
in
e
lear
n
in
g
ar
e
p
r
ec
is
io
n
,
r
ec
all,
an
d
f
1
-
s
co
r
e.
A
cc
ep
tab
le
i
m
p
r
o
v
e
m
e
n
ts
w
er
e
r
ec
o
r
d
ed
af
ter
th
e
tr
a
n
s
la
tio
n
a
n
d
SVM
is
t
h
e
b
est
cla
s
s
i
f
ier
in
m
u
ltil
i
n
g
u
al
t
w
i
tter
s
en
ti
m
e
n
t
an
al
y
s
is
.
A
cc
u
r
ac
y
is
i
m
p
r
o
v
ed
b
y
u
p
to
9
5
%.
I
n
th
is
w
o
r
k
,
s
o
m
e
ch
a
llen
g
e
s
n
ee
d
to
b
e
s
o
lv
ed
i
n
s
itu
a
tio
n
s
l
ik
e
f
o
r
ex
a
m
p
le,
ac
c
u
r
ac
y
d
r
o
p
s
w
h
e
n
t
h
e
t
w
ee
ts
co
n
ta
in
co
d
e
-
m
i
x
ed
an
d
co
d
e
-
s
w
itc
h
ed
w
o
r
d
s
an
d
s
e
n
t
en
ce
s
.
RE
F
E
R
E
NC
E
S
[1
]
J.
Clem
e
n
t,
“
Tw
it
ter:
n
u
m
b
e
r
o
f
a
c
ti
v
e
u
se
rs
2
0
1
0
-
2
0
1
9
,
”
S
ta
ti
sta
,
A
u
g
.
2
0
1
9
.
[
O
n
li
n
e
]
,
A
v
a
il
a
b
le:
h
tt
p
s:/
/www
.
sta
ti
sta
.
c
o
m
/statisti
c
s/2
8
2
0
8
7
/
n
u
m
b
e
r
-
of
-
m
o
n
th
ly
-
a
c
ti
v
e
t
w
it
ter
-
u
se
rs.
[2
]
E
.
Ko
u
l
o
u
m
p
is
,
e
t
a
l.
,
“
T
w
it
ter
S
e
n
ti
m
e
n
t
A
n
a
l
y
sis:
T
h
e
G
o
o
d
th
e
Ba
d
a
n
d
t
h
e
OMG!
”
in
Pro
c
e
e
d
in
g
s
o
f
Fi
ft
h
In
ter
n
a
t
io
n
a
l
AA
AI
C
o
n
fer
e
n
c
e
o
n
W
e
b
lo
g
s a
n
d
S
o
c
i
a
l
M
e
d
i
a
,
p
p
.
5
3
8
-
5
4
1
,
2
0
1
1
.
[3
]
K.
A
ru
n
,
e
t
a
l.
,
“
Tw
it
ter
S
e
n
ti
m
e
n
t
A
n
a
l
y
sis
o
n
De
m
o
n
e
ti
z
a
ti
o
n
t
w
e
e
ts
in
In
d
ia
Us
in
g
R
lan
g
u
a
g
e
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Co
m
p
u
ter
E
n
g
i
n
e
e
rin
g
in
Res
e
a
rc
h
T
re
n
d
s
,
v
o
l.
4
,
no.
6
,
p
p
.
2
5
2
-
2
5
8
,
2
0
1
7
.
[4
]
S
.
A
.
El
Ra
h
m
a
n
,
e
t
a
l.
,
“
S
e
n
t
im
e
n
t
A
n
a
l
y
sis
o
f
Tw
it
ter
Da
ta
,
”
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
Co
mp
u
ter
a
n
d
In
fo
rm
a
t
io
n
S
c
ien
c
e
s (
ICCIS
),
S
a
u
d
i
A
ra
b
ia,
p
p
.
1
-
4
,
2
0
1
9
.
[5
]
S
.
Ku
m
a
r
,
e
t
a
l.
,
“
E
x
p
lo
rin
g
I
m
p
a
c
t
o
f
Ag
e
a
n
d
G
e
n
d
e
r
o
n
S
e
n
ti
m
e
n
t
A
n
a
l
y
sis
Us
in
g
M
a
c
h
in
e
L
e
a
rn
in
g
,
”
El
e
c
tro
n
ics
,
v
o
l.
9
,
n
o
.
2
,
p
p.
1
-
1
4
,
2
0
2
0
.
[6
]
H.
S
a
if
,
e
t
a
l.
,
“
S
e
m
a
n
ti
c
se
n
ti
m
e
n
t
a
n
a
l
y
sis
o
f
t
w
it
ter
,
”
In
ter
n
a
ti
o
n
a
l
S
e
ma
n
ti
c
W
e
b
Co
n
fer
e
n
c
e
,
p
p.
5
0
8
-
5
2
4
,
2
0
1
2
.
[7
]
S
.
S
h
a
y
a
a
,
e
t
a
l.
,
“
S
e
n
ti
m
e
n
t
A
n
a
ly
sis
o
f
Big
Da
ta:
M
e
th
o
d
s,
A
p
p
l
ica
ti
o
n
s,
a
n
d
Op
e
n
Ch
a
ll
e
n
g
e
s
,
”
in
IEE
E
Acc
e
ss
,
v
o
l.
6
,
p
p
.
3
7
8
0
7
-
3
7
8
2
7
,
2
0
1
8
.
[8
]
T
.
M.
Nisa
r
a
n
d
M.
Ye
u
n
g
,
“
Twit
ter
a
s
a
T
o
o
l
f
o
r
F
o
re
c
a
stin
g
S
to
c
k
M
a
rk
e
t
M
o
v
e
m
e
n
ts:
A
S
h
o
rt
-
w
in
d
o
w
Ev
e
n
t
S
tu
d
y
,
”
T
h
e
J
o
u
rn
a
l
o
f
Fi
n
a
n
c
e
a
n
d
D
a
ta
S
c
ien
c
e
,
v
o
l.
4
,
n
o
.
2
,
p
p
.
1
0
1
-
1
1
9
,
20
18
.
[9
]
A
.
K
.
S
o
n
i,
“
M
u
lt
i
-
L
in
g
u
a
l
S
e
n
ti
m
e
n
t
A
n
a
l
y
sis
o
f
Tw
it
ter
d
a
ta
b
y
u
sin
g
c
las
sif
i
c
a
ti
o
n
a
lg
o
rit
h
m
s
,
”
2
0
0
7
S
e
c
o
n
d
In
ter
n
a
t
io
n
a
l
C
o
n
fer
e
n
c
e
o
n
E
lec
trica
l,
Co
m
p
u
ter
a
n
d
Co
mm
u
n
ica
ti
o
n
T
e
c
h
n
o
lo
g
ies
,
p
p
.
1
-
5
,
2
0
1
7
.
[1
0
]
Z.
Jia
n
q
ian
g
a
n
d
G
.
X
iao
li
n
,
“
Co
m
p
a
riso
n
Re
se
a
rc
h
o
n
T
e
x
t
P
re
-
p
ro
c
e
ss
in
g
M
e
th
o
d
s
o
n
Tw
it
ter
S
e
n
ti
m
e
n
t
A
n
a
l
y
si
s,
”
in
IEE
E
Acc
e
ss
,
v
o
l.
5
,
p
p
.
2
8
7
0
-
2
8
7
9
,
2
0
1
7
.
[1
1
]
A
.
De
sh
w
a
l
a
n
d
S
.
K.
S
h
a
rm
a
,
“
Tw
it
ter
S
e
n
ti
m
e
n
t
A
n
a
l
y
sis
u
sin
g
V
a
ri
o
u
s
Clas
sif
ica
ti
o
n
A
lg
o
rit
h
m
s,
”
2
0
1
6
5
t
h
In
ter
n
a
t
io
n
a
l
C
o
n
fer
e
n
c
e
o
n
Reli
a
b
il
i
ty,
In
f
o
c
o
m
T
e
c
h
n
o
l
o
g
ies
a
n
d
Op
ti
miz
a
ti
o
n
(
T
re
n
d
s
a
n
d
Fu
t
u
re
Dire
c
ti
o
n
s
)
(
ICRIT
O
)
,
No
i
d
a
,
p
p
.
2
5
1
-
2
5
7
,
2
0
1
6
.
[1
2
]
A
.
Ba
lah
u
r
a
n
d
M
.
T
u
rc
h
i
,
“
Im
p
ro
v
in
g
se
n
ti
m
e
n
t
a
n
a
ly
sis
in
tw
it
ter
u
sin
g
m
u
lt
il
i
n
g
u
a
l
m
a
c
h
in
e
tran
sla
ted
d
a
ta
,
”
Pro
c
e
e
d
in
g
s
o
f
t
h
e
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
Rec
e
n
t
Ad
v
a
n
c
e
s in
Na
tu
ra
l
L
a
n
g
u
a
g
e
Pr
o
c
e
ss
in
g
,
p
p.
49
-
5
5
,
2
0
1
3
.
[1
3
]
S
.
K.
M
a
h
a
ta,
e
t
a
l.
,
“
A
n
a
l
y
z
in
g
Co
d
e
-
S
w
it
c
h
in
g
Ru
les
f
o
r
En
g
li
sh
–
Hin
d
i
Co
d
e
-
M
ix
e
d
Tex
t
,
”
Eme
rg
in
g
T
e
c
h
n
o
l
o
g
y
i
n
M
o
d
e
ll
in
g
a
n
d
Gr
a
p
h
ics
,
p
p
.
1
3
7
-
1
4
5
,
2
0
2
0
.
[1
4
]
K
.
Da
sh
ti
p
o
u
r
,
e
t
a
l.
,
“
M
u
lt
il
i
n
g
u
a
l
S
e
n
ti
m
e
n
t
A
n
a
l
y
sis:
S
tate
o
f
th
e
A
rt
a
n
d
In
d
e
p
e
n
d
e
n
t
Co
m
p
a
riso
n
o
f
T
e
c
h
n
iq
u
e
s
,
”
Co
g
n
it
ive
C
o
mp
u
t
a
t
io
n
,
v
o
l.
8
,
p
p
.
7
5
7
-
7
7
1
,
2
0
1
6
.
[1
5
]
I
.
M
o
z
e
ti
č
,
e
t
a
l.
,
“
M
u
lt
il
in
g
u
a
l
Tw
it
ter
S
e
n
ti
m
e
n
t
Clas
sif
ic
a
ti
o
n
:
T
h
e
Ro
le
o
f
Hu
m
a
n
A
n
n
o
tat
o
r
s
,
”
PL
OS
ONE
,
v
o
l.
1
1
,
n
o
.
5
,
2
0
1
6
.
[1
6
]
M
.
E.
Ba
siri
a
n
d
A
.
Ka
b
iri
,
“
Un
in
o
rm
o
p
e
ra
to
rs
f
o
r
se
n
ten
c
e
-
lev
e
l
s
c
o
re
a
g
g
re
g
a
ti
o
n
in
se
n
t
im
e
n
t
a
n
a
l
y
sis,
”
4
th
I
n
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
W
e
b
Res
e
a
rc
h
(
ICW
R
)
,
Teh
ra
n
,
p
p
.
97
-
1
0
2
,
2
0
1
8
.
[1
7
]
S.
S
a
h
u
,
S
.
K.
Ro
u
t
a
n
d
D.
M
o
h
a
n
ty
,
“
Tw
it
ter
S
e
n
ti
m
e
n
t
A
n
a
l
y
si
s
-
A
M
o
re
En
h
a
n
c
e
d
W
a
y
o
f
Cl
a
ss
if
i
c
a
ti
o
n
a
n
d
S
c
o
rin
g
,
”
2
0
1
5
IE
EE
In
ter
n
a
ti
o
n
a
l
S
y
mp
o
siu
m
o
n
Na
n
o
e
lec
tro
n
ic
a
n
d
In
f
o
rm
a
ti
o
n
S
y
ste
ms
,
p
p
.
67
-
72
,
2
0
1
5
.
[1
8
]
L
.
Brö
n
n
im
a
n
n
,
“
M
u
lt
il
a
n
g
u
a
g
e
se
n
ti
m
e
n
t
-
a
n
a
ly
sis
o
f
Tw
it
ter
d
a
ta
o
n
th
e
e
x
a
m
p
le
o
f
S
w
i
ss
p
o
li
ti
c
ian
s
,
”
M
.
S
c
.
T
h
e
sis
,
Un
iv
e
rsity
o
f
A
p
p
li
e
d
S
c
ien
c
e
s N
o
rth
w
e
ste
rn
S
w
it
z
e
rlan
d
,
2
0
1
4
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
10
,
No
.
6
,
Dec
em
b
er
2
0
2
0
:
5
9
9
2
-
6
0
0
0
6000
[1
9
]
N.
F
.
A
lsh
a
m
m
a
ri
a
n
d
A
.
A
.
A
l
M
a
n
so
u
r,
“
S
tate
-
of
-
th
e
-
a
rt
re
v
ie
w
o
n
Tw
it
ter
S
e
n
ti
m
e
n
t
A
n
a
l
y
sis,
”
2
n
d
In
ter
n
a
t
io
n
a
l
C
o
n
fer
e
n
c
e
o
n
Co
mp
u
ter
Ap
p
li
c
a
t
io
n
s
&
In
fo
rm
a
t
io
n
S
e
c
u
rity
(
ICCAIS
),
Riy
a
d
h
,
S
a
u
d
i
A
ra
b
ia,
p
p
.
1
-
8
,
2
0
1
9
.
[2
0
]
A
.
G
h
a
ll
a
b
,
e
t
a
l.
,
“
A
ra
b
ic
S
e
n
ti
m
e
n
t
A
n
a
l
y
sis:
A
S
y
st
e
m
a
ti
c
L
it
e
ra
tu
re
R
e
v
ie
w
,
”
Ap
p
li
e
d
Co
mp
u
t
a
ti
o
n
a
l
In
telli
g
e
n
c
e
a
n
d
S
o
ft
Co
mp
u
ti
n
g
,
pp.
1
-
2
1
,
2
0
2
0
.
[2
1
]
M
.
Al
-
Am
in
,
e
t
a
l.
,
“
S
e
n
ti
m
e
n
t
A
n
a
l
y
si
s
o
f
Be
n
g
a
li
Co
m
m
e
n
ts
w
it
h
w
o
rd
s2
V
CE
a
n
d
S
e
n
ti
m
e
n
t
In
f
o
rm
a
ti
o
n
o
f
W
o
rd
s
,
”
2
0
1
7
In
ter
n
a
ti
o
n
a
l
C
o
n
fer
e
n
c
e
o
n
El
e
c
trica
l,
c
o
mp
u
t
e
r
a
n
d
Co
mm
u
n
ica
ti
o
n
E
n
g
i
n
e
e
rin
g
(
ECCE
)
,
pp
.
1
8
6
-
1
90
,
2
0
1
7
.
[2
2
]
S.
Na
rr,
e
t
a
l.
,
“
L
a
n
g
u
a
g
e
-
In
d
e
p
e
n
d
e
n
t
T
w
it
ter
S
e
n
ti
m
e
n
t
A
n
a
l
y
sis
,
”
W
o
rk
sh
o
p
o
n
K
n
o
wle
d
g
e
Disc
o
v
e
ry
,
Da
ta
M
in
in
g
a
n
d
M
a
c
h
in
e
L
e
a
rn
i
n
g
(
KDM
L
2
0
1
2
)
,
2
0
1
2
.
[2
3
]
Na
n
k
a
n
i
H.,
e
t
a
l
.
,
“
M
u
lt
il
in
g
u
a
l
S
e
n
ti
m
e
n
t
A
n
a
l
y
sis
,
”
i
n
A
g
a
rwa
l
B.
,
e
t
a
l.
(e
d
s)
,
“
De
e
p
L
e
a
rn
in
g
-
Ba
s
e
d
A
p
p
ro
a
c
h
e
s f
o
r
S
e
n
ti
m
e
n
t
A
n
a
l
y
s
is
,
”
Al
g
o
rith
ms
f
o
r In
tell
ig
e
n
t
S
y
s
tem
s
,
p
p
.
1
9
3
-
2
3
6
,
2
0
2
0
.
[2
4
]
H.
S
a
if
,
e
t
a
l.
,
“
A
ll
e
v
iatin
g
d
a
ta
sp
a
rsity
f
o
r
Tw
it
ter
se
n
ti
m
e
n
t
a
n
a
ly
sis
,
”
2
n
d
W
o
rk
sh
o
p
o
n
M
a
k
in
g
S
e
n
se
o
f
M
icr
o
p
o
sts
,
p
p
.
2
-
9
,
2
0
1
2
.
[
2
5
]
Y.
W
u
,
e
t
a
l.
,
“
G
o
o
g
le'
s
Ne
u
ra
l
M
a
c
h
in
e
T
ra
n
sla
ti
o
n
S
y
ste
m
:
B
rid
g
in
g
th
e
G
a
p
b
e
t
w
e
e
n
Hu
m
a
n
a
n
d
M
a
c
h
i
n
e
T
ra
n
sla
ti
o
n
,
”
a
rXiv:
1
6
0
9
.
0
8
1
4
4
,
2
0
1
6
.
[2
6
]
Q
.
V
.
L
e
a
n
d
M
.
S
c
h
u
ste
r,
“
A
Ne
u
ra
l
Ne
t
w
o
rk
f
o
r
M
a
c
h
in
e
T
ra
n
sla
ti
o
n
,
a
t
P
ro
d
u
c
ti
o
n
S
c
a
le
,
”
G
o
o
g
le
A
I
Blo
g
,
S
e
p
2
0
1
6
.
[
O
n
li
n
e
],
A
v
a
il
a
b
le:
h
tt
p
s://
a
i.
g
o
o
g
leb
lo
g
.
c
o
m
/2
0
1
6
/
0
9
/a
-
n
e
u
ra
l
-
n
e
tw
o
rk
-
f
o
r
-
m
a
c
h
in
e
.
h
tm
l
.
[2
7
]
S
.
S
t
r
a
s
s
e
l
,
e
t
a
l.
,
“
H
a
n
d
b
o
o
k
o
f
N
a
t
u
r
a
l
L
a
n
g
u
a
g
e
P
r
o
c
e
s
s
i
n
g
a
n
d
M
a
c
h
i
n
e
T
r
a
n
s
l
a
t
i
o
n
:
D
A
R
P
A
G
l
o
b
a
l
A
u
t
o
n
o
m
o
u
s
L
a
n
g
u
a
g
e
E
x
p
l
o
i
t
a
t
i
o
n
,
”
S
p
r
i
n
g
e
r
-
V
e
r
l
a
g
,
2
0
1
1
.
[
2
8
]
L
.
S
p
e
c
ia
,
e
t
a
l.
,
“
Q
u
a
l
i
t
y
E
s
t
i
m
a
t
i
o
n
f
o
r
M
a
c
h
i
n
e
T
r
a
n
s
l
a
t
i
o
n
,
”
M
o
r
g
a
n
&
C
l
a
y
p
o
o
l
P
u
b
l
i
s
h
e
r
s
,
2
0
1
8
.
[
2
9
]
C
h
r
i
s
t
o
p
h
e
r
O
l
a
h
,
“
U
n
d
e
r
s
t
a
n
d
i
n
g
L
S
T
M
N
e
t
w
o
r
k
s
,
”
2
0
1
5
.
[
O
n
l
i
n
e
]
,
A
v
a
i
l
a
b
l
e
:
h
t
tp
s://
c
o
lah
.
g
it
h
u
b
.
io
/
p
o
sts/2
0
1
5
-
08
-
Un
d
e
rsta
n
d
i
n
g
-
L
S
T
M
s/
.
[
3
0
]
J
e
s
u
s
R
o
g
e
l
-
S
a
l
a
z
a
r
,
“
D
a
t
a
S
c
i
e
n
c
e
a
n
d
A
n
a
l
y
t
i
c
s
w
i
t
h
P
y
t
h
o
n
,
”
1
s
t
.
e
d
.
,
C
h
a
p
m
a
n
&
H
a
l
l
/
C
R
C
,
2
0
1
7
Evaluation Warning : The document was created with Spire.PDF for Python.