TELK
O
M
N
I
K
A
Te
l
e
c
ommu
n
i
c
at
i
on
,
C
omp
u
ti
n
g
,
El
e
c
t
r
on
i
c
s
an
d
C
on
tr
o
l
V
ol
.
18
,
N
o
.
2
,
A
pri
l
2020
,
p
p.
799
~
8
0
6
IS
S
N
:
1693
-
6930
,
a
c
c
r
e
di
t
e
d
F
i
rs
t
G
r
a
de
by
K
e
m
e
n
ri
s
t
e
kd
i
k
t
i
,
D
e
c
r
e
e
N
o:
21
/
E
/
K
P
T
/
2
018
D
O
I:
10
.
129
28/
T
E
L
K
O
M
N
IK
A
.
v1
8i
2
.
1474
4
799
Jou
r
n
al
h
o
m
e
pa
ge
:
ht
t
p:
/
/
j
our
nal
.
uad
.
a
c
.
i
d/
i
nd
e
x
.
php
/
T
E
L
K
O
MNIKA
Ho
a
x
c
l
a
ss
i
f
i
c
a
t
i
o
n
a
n
d
s
e
n
t
i
m
e
n
t
a
n
a
l
y
si
s
o
f
I
n
d
o
n
e
s
i
a
n
n
e
w
s
u
si
n
g
N
a
i
v
e
B
a
y
e
s
o
p
t
i
mi
z
a
t
i
o
n
H
e
r
u
A
gu
s
S
an
tos
o
,
Ek
o
H
a
r
i
R
ac
h
m
aw
an
to,
A
d
h
i
tya
N
u
gr
ah
a,
A
k
b
ar
A
ji
N
u
gr
oh
o
,
D
e
R
os
al
I
gn
at
i
u
s
M
os
e
s
S
e
ti
ad
i
,
R
u
r
i
S
u
k
o
Bas
u
k
i
D
e
pa
r
t
m
e
n
t
of
I
nf
or
m
a
t
i
c
s
E
ng
i
n
e
e
r
i
n
g,
F
a
c
u
l
t
y
o
f
C
om
pu
t
e
r
S
c
i
e
n
c
e
,
D
i
a
n
N
us
w
a
nt
or
o
U
n
i
ve
r
s
i
t
y
,
I
ndo
ne
s
i
a
A
r
ti
c
l
e
I
n
fo
A
BS
TR
A
C
T
Ar
t
i
c
l
e
h
i
s
t
or
y
:
Re
c
e
i
v
e
d
J
ul
4
,
201
9
Re
vi
s
e
d
J
a
n
17
,
202
0
A
c
c
e
pt
e
d
F
e
b
15
,
202
0
C
ur
r
e
nt
l
y
,
t
he
s
p
r
e
a
d
of
h
oa
x
n
e
w
s
h
a
s
i
nc
r
e
a
s
e
d
s
i
gn
i
f
i
c
a
n
t
l
y,
e
s
p
e
c
i
a
l
l
y
on
s
oc
i
a
l
m
e
d
i
a
ne
t
w
or
ks
.
H
o
a
x
n
e
w
s
i
s
ve
r
y
da
n
ge
r
ous
a
n
d
c
a
n
pr
o
voke
r
e
a
d
e
r
s
.
S
o,
t
h
i
s
r
e
qu
i
r
e
s
s
pe
c
i
a
l
h
a
nd
l
i
n
g.
T
hi
s
r
e
s
e
a
r
c
h
pr
opos
e
d
a
ho
a
x
ne
w
s
de
t
e
c
t
i
on
s
y
s
t
e
m
us
i
ng
s
e
a
r
c
h
i
n
g
,
s
n
i
ppe
t
a
n
d
c
os
i
ne
s
i
m
i
l
a
r
i
t
y
m
e
t
ho
ds
t
o
c
l
a
s
s
i
f
y
hoa
x
n
e
w
s
.
T
h
i
s
m
e
t
hod
i
s
p
r
op
os
e
d
be
c
a
us
e
t
he
s
e
a
r
c
hi
ng
m
e
t
ho
d
do
e
s
no
t
r
e
qu
i
r
e
t
r
a
i
n
i
ng
da
t
a
,
s
o
i
t
i
s
p
r
a
c
t
i
c
a
l
t
o
us
e
a
nd
a
l
w
a
ys
u
p
t
o
d
a
t
e
.
I
n
a
dd
i
t
i
on
,
one
of
t
h
e
d
r
a
w
ba
c
ks
of
t
h
e
e
x
i
s
t
i
ng
a
ppr
oa
c
he
s
i
s
t
h
e
y
a
r
e
n
ot
e
qu
i
pp
e
d
w
i
t
h
a
s
e
nt
i
m
e
n
t
a
n
a
l
ys
i
s
f
e
a
t
ur
e
.
I
n
ou
r
s
ys
t
e
m
,
s
e
n
t
i
m
e
n
t
a
n
a
l
y
s
i
s
i
s
c
a
r
r
i
e
d
ou
t
a
f
t
e
r
ho
a
x
ne
w
s
i
s
d
e
t
e
c
t
e
d
.
T
he
go
a
l
i
s
t
o
e
x
t
r
a
c
t
t
h
e
t
r
ue
hi
dde
n
s
e
n
t
i
m
e
n
t
i
n
s
i
de
ho
a
x
w
he
t
h
e
r
pos
i
t
i
v
e
s
e
n
t
i
m
e
nt
o
r
ne
g
a
t
i
ve
s
e
n
t
i
m
e
nt
.
I
n
t
he
p
r
oc
e
s
s
o
f
s
e
nt
i
m
e
n
t
a
n
a
l
y
s
i
s
,
t
he
N
a
ï
v
e
B
a
ye
s
(
N
B
)
m
e
t
hod
w
a
s
us
e
d
w
h
i
c
h
w
a
s
op
t
i
m
i
z
e
d
u
s
i
n
g
t
he
P
a
r
t
i
c
l
e
S
w
a
r
m
O
pt
i
m
i
z
a
t
i
o
n
(
P
S
O
)
m
e
t
hod
.
B
a
s
e
d
on
t
he
r
e
s
u
l
t
s
of
e
x
pe
r
i
m
e
n
t
on
30
ho
a
x
ne
w
s
s
a
m
p
l
e
s
t
ha
t
a
r
e
w
i
d
e
l
y
s
p
r
e
a
d
on
s
oc
i
a
l
m
e
d
i
a
n
e
t
w
o
r
k
s
,
t
he
a
v
e
r
a
ge
of
h
oa
x
ne
w
s
de
t
e
c
t
i
on
r
e
a
c
h
e
s
77
%
of
a
c
c
u
r
a
c
y,
w
he
r
e
e
a
c
h
n
e
w
s
i
s
c
o
r
r
e
c
t
l
y
i
d
e
nt
i
f
i
e
d
a
s
a
ho
a
x
i
n
t
he
r
a
n
ge
be
t
w
e
e
n
66
%
a
n
d
91
%
of
a
c
c
u
r
a
c
y.
I
n
a
dd
i
t
i
o
n,
t
he
p
r
opo
s
e
d
s
e
n
t
i
m
e
nt
a
na
l
ys
i
s
m
e
t
h
od
p
r
ov
e
d
t
o
h
a
s
a
b
e
t
t
e
r
pe
r
f
or
m
a
nc
e
t
h
a
n
t
he
pr
e
v
i
o
us
a
na
l
ys
i
s
s
e
nt
i
m
e
n
t
m
e
t
hod
.
Ke
y
w
or
d
s
:
Cos
i
ne
s
i
m
i
l
a
r
i
t
y
H
oa
x
ne
w
s
d
e
t
e
c
t
i
o
n
N
a
ï
v
e
B
a
y
e
s
P
a
rt
i
c
l
e
s
w
a
rm
op
t
i
m
i
z
a
t
i
on
s
e
nt
i
m
e
nt
a
n
a
l
ys
i
s
T
hi
s
i
s
an
op
e
n
ac
c
e
s
s
ar
t
i
c
l
e
u
nde
r
t
he
C
C
B
Y
-
SA
l
i
c
e
n
s
e
.
Cor
r
e
s
pon
di
n
g
Au
t
h
or
:
H
e
ru
A
g
us
S
a
n
t
os
o
,
D
e
pa
r
t
m
e
nt
of
Info
rm
a
t
i
c
s
E
ng
i
ne
e
ri
ng,
F
a
c
ul
t
y
of
Com
put
e
r
S
c
i
e
nc
e
,
D
i
a
n
N
us
w
a
nt
oro
U
ni
v
e
rs
i
t
y,
Indo
ne
s
i
a
.
E
m
a
i
l
:
he
ru
.
a
gus
.
s
a
nt
os
o
@
ds
n
.
di
n
us
.
a
c
.
i
d
1.
I
N
TR
O
D
U
C
TI
O
N
Curre
n
t
l
y
,
t
he
i
m
pa
c
t
of
s
o
c
i
a
l
m
e
di
a
i
n
our
da
i
l
y
l
i
f
e
t
e
n
ds
t
o
i
n
c
re
a
s
e
.
M
or
e
ov
e
r,
s
oc
i
a
l
m
e
di
a
c
a
n
ha
ve
g
ood
or
b
a
d
i
m
pa
c
t
s
t
oge
t
he
r.
S
o
c
i
a
l
m
e
di
a
s
uc
h
a
s
T
w
i
t
t
e
r
or
F
a
c
e
b
ook
prod
uc
e
ne
w
s
or
i
nfor
m
a
t
i
on
w
hi
c
h
c
a
n
be
e
a
s
i
l
y
s
pr
e
a
d
a
roun
d
t
h
e
gl
obe
.
In
t
e
r
m
s
of
ho
a
x,
t
he
i
nfor
m
a
t
i
on
w
i
l
l
b
e
good
i
f
i
t
i
s
g
e
nu
i
ne
a
nd
h
a
s
g
one
t
hr
ough
good
re
a
s
on
i
ng.
Bu
t
t
he
fa
c
t
pe
opl
e
s
p
re
a
d
fa
l
s
e
i
nf
orm
a
t
i
on
t
o
ga
i
n
p
a
rt
i
c
ul
a
r
b
e
ne
fi
t
s
i
s
i
n
c
re
a
s
i
n
g
e
ve
r
y
y
e
a
r
a
nd
p
re
c
i
s
e
l
y
It
ha
s
i
n
c
re
a
s
e
d
s
ha
rpl
y
i
n
t
h
e
p
a
s
t
t
w
o
ye
a
rs
[1
,
2]
.
A
c
t
i
v
e
s
o
c
i
a
l
m
e
d
i
a
a
c
c
ount
s
a
l
s
o
i
n
c
re
a
s
e
e
ve
ry
y
e
a
r
,
i
n
c
l
u
di
ng
t
he
on
e
s
prod
uc
i
ng
hoa
x
i
nfor
m
a
t
i
o
n.
T
hi
s
c
a
us
e
p
e
opl
e
c
onn
e
c
t
e
d
t
o
s
o
c
i
a
l
m
e
di
a
ha
v
e
di
ff
i
c
ul
t
i
e
s
t
o
d
e
t
e
rm
i
n
e
w
he
t
h
e
r
t
h
e
y
re
a
d
ge
n
ui
n
e
o
r
f
a
l
s
e
i
nfor
m
a
t
i
o
n.
T
he
s
i
t
u
a
t
i
on
w
o
rs
e
ns
a
s
ho
a
x
s
pr
e
a
d
ov
e
r
s
oc
i
a
l
m
e
d
i
a
ne
t
w
orks
re
a
d
by
m
ore
a
nd
m
o
re
pe
op
l
e
,
e
s
pe
c
i
a
l
l
y
i
n
In
done
s
i
a
t
h
e
c
ou
nt
ry
w
i
t
h
t
he
t
hi
rd
-
l
a
rge
s
t
s
oc
i
a
l
m
e
di
a
p
e
ne
t
ra
t
i
on
i
n
t
he
w
or
l
d
i
n
2018
[3]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
1693
-
6930
T
E
L
K
O
M
N
IK
A
T
e
l
e
c
om
m
un
Co
m
put
E
l
Con
t
rol
,
V
ol
.
18
,
N
o.
2
,
A
pri
l
2
020:
799
-
80
6
800
H
oa
x
i
s
fa
l
s
e
i
nf
orm
a
t
i
on
t
ha
t
i
s
c
ons
i
d
e
re
d
c
orre
c
t
a
n
d
c
a
n
m
i
s
l
e
a
d
hum
a
n
pe
r
c
e
pt
i
on
[4
,
5]
.
S
pre
a
d
i
ng
hoa
x
i
nfor
m
a
t
i
on
us
u
a
l
l
y
h
a
s
m
ul
t
i
p
l
e
purpos
e
s
,
w
i
t
h
t
h
e
a
i
m
of
p
e
rs
ua
d
i
ng
or
m
a
ni
pu
l
a
t
i
n
g
pu
bl
i
c
opi
ni
on.
T
h
e
s
pre
a
d
of
hoa
x
e
s
i
s
us
u
a
l
l
y
a
c
c
o
m
p
a
ni
e
d
by
fr
a
ud
a
nd
e
ve
n
t
hre
a
t
s
.
In
2
016,
t
h
e
re
w
e
re
a
rou
nd
800,
00
0
hoa
x
s
i
t
e
s
t
h
a
t
pro
duc
e
fa
l
s
e
i
nfor
m
a
t
i
on
w
hi
c
h
w
i
d
e
l
y
di
s
t
ri
bu
t
e
d
ove
r
s
o
c
i
a
l
m
e
d
i
a
,
s
uc
h
a
s
T
w
i
t
t
e
r
,
F
a
c
e
book
a
nd
o
t
h
e
rs
[6]
,
e
v
e
n
h
oa
x
ne
w
s
i
s
i
nc
r
e
a
s
i
n
gl
y
pre
va
l
e
n
t
w
i
t
h
e
v
i
l
pol
i
t
i
c
a
l
goa
l
s
[3]
.
T
he
s
p
r
e
a
d
of
ho
a
xe
s
h
a
s
a
v
e
ry
bro
a
d
i
m
pa
c
t
a
nd
e
v
e
n
i
t
h
a
s
m
a
ny
pot
e
nt
i
a
l
s
o
f
c
a
us
i
ng
da
nge
r
ous
hori
z
ont
a
l
c
onf
l
i
c
t
s
for
t
he
s
t
a
b
i
l
i
t
y
of
t
he
w
hol
e
c
o
unt
ry
.
T
h
us
,
a
hoa
x
de
t
e
c
t
i
on
s
ys
t
e
m
i
s
ne
e
de
d
t
o
a
ut
o
m
a
t
i
c
a
l
l
y
h
e
l
p
c
i
t
i
z
e
n
a
nd
gov
e
rn
m
e
n
t
f
i
l
t
e
r
i
ng
i
nf
orm
a
t
i
on.
Re
s
e
a
rc
h
o
n
t
he
h
oa
x
d
e
t
e
c
t
i
on
s
ys
t
e
m
ha
s
be
e
n
c
a
rr
i
e
d
ou
t
i
n
r
e
c
e
nt
y
e
a
rs
s
uc
h
a
s
t
h
e
one
s
in
[2
,
4,
5,
7
-
11]
.
T
he
s
e
s
t
udi
e
s
prop
os
e
c
l
a
s
s
i
fi
c
a
t
i
on
o
r
l
e
a
rni
n
g
t
e
c
hn
i
qu
e
s
,
w
h
e
re
t
h
i
s
t
e
c
hn
i
que
a
l
w
a
ys
re
qui
re
s
up
-
to
-
da
t
e
t
ra
i
ni
n
g
da
t
a
t
o
m
a
i
nt
a
i
n
t
he
a
c
c
ur
a
c
y
of
t
h
e
de
t
e
c
t
i
on
.
O
n
t
he
o
t
he
r
ha
nd
,
t
h
e
s
e
a
rc
h
i
ng
t
e
c
hni
q
ue
t
o
de
t
e
c
t
h
oa
x
n
e
w
s
c
a
n
be
done
us
i
ng
a
s
ni
pp
e
t
a
s
pr
e
s
e
nt
e
d
i
n
t
h
e
fol
l
ow
i
ng
s
t
ud
i
e
s
[12
-
14]
.
S
e
a
rc
hi
ng
t
e
c
hni
que
s
ha
ve
t
h
e
a
dv
a
n
t
a
g
e
of
be
i
ng
m
o
re
up
t
o
d
a
t
e
a
nd
m
ore
pr
a
c
t
i
c
a
l
i
n
us
e
.
T
h
e
re
fo
re
,
t
hi
s
pa
pe
r
prop
os
e
s
ho
a
x
n
e
w
s
d
e
t
e
c
t
i
on
t
e
c
hn
i
que
s
e
m
pl
o
yi
n
g
s
e
a
r
c
hi
ng
t
e
c
hni
q
ue
s
t
ha
t
a
re
c
o
m
bi
n
e
d
w
i
t
h
c
l
a
s
s
i
fi
e
r
m
e
t
hods
t
o
i
m
pr
ove
a
c
c
ur
a
c
y.
A
fur
t
he
r
dra
w
ba
c
k
of
t
h
e
e
xi
s
t
i
ng
hoa
x
de
t
e
c
t
i
on
s
ys
t
e
m
i
s
t
h
e
y
a
re
not
e
qui
ppe
d
w
i
t
h
s
e
nt
i
m
e
nt
a
n
a
l
ys
i
s
fe
a
t
u
re
s
.
T
o
a
ddre
s
s
t
he
probl
e
m
,
S
e
nt
i
m
e
n
t
a
na
l
ys
i
s
f
e
a
t
ur
e
i
s
propos
e
d
.
In
our
s
ys
t
e
m
,
s
e
nt
i
m
e
nt
a
na
l
ys
i
s
i
s
c
a
rr
i
e
d
o
ut
a
ft
e
r
hoa
x
ne
w
s
i
s
d
e
t
e
c
t
e
d
.
S
e
n
t
i
m
e
n
t
a
n
a
l
ys
i
s
c
a
n
e
xt
r
a
c
t
t
h
e
t
rue
hi
dd
e
n
s
e
nt
i
m
e
nt
i
ns
i
d
e
ho
a
x
w
h
e
t
h
e
r
p
os
i
t
i
ve
s
e
n
t
i
m
e
n
t
or
ne
g
a
t
i
ve
s
e
nt
i
m
e
n
t
.
T
hi
s
fe
a
t
ur
e
he
l
ps
us
t
o
fur
t
h
e
r
e
xt
r
a
c
t
t
h
e
m
ot
i
va
t
i
on
of
t
he
h
oa
x
w
hi
c
h
c
a
n
be
f
or
bl
a
c
k
c
a
m
pa
i
gns
or
n
ot
.
H
e
n
c
e
i
t
i
s
ne
c
e
s
s
a
ry
t
o
kn
ow
i
t
s
s
e
nt
i
m
e
nt
c
l
a
s
s
i
f
i
c
a
t
i
on
i
n
re
s
p
ons
e
t
o
t
he
hoa
x
ne
w
s
.
S
o
m
e
m
e
t
hods
t
ha
t
a
re
w
i
de
l
y
us
e
d
t
o
c
l
a
s
s
i
fy
t
e
x
t
a
n
d
c
on
duc
t
s
e
n
t
i
m
e
n
t
a
na
l
ys
i
s
a
r
e
N
a
ï
v
e
Ba
y
e
s
[15
-
19]
,
S
uppor
t
V
e
c
t
or
M
a
c
h
i
n
e
[20
-
22]
,
a
nd
K
N
N
[23
,
24]
.
I
n
t
hi
s
re
s
e
a
r
c
h,
N
a
ï
v
e
B
a
y
e
s
m
e
t
hod
w
a
s
c
hos
e
n
t
o
c
a
rry
ou
t
c
l
a
s
s
i
fi
c
a
t
i
o
n
a
nd
s
e
nt
i
m
e
nt
a
na
l
ys
i
s
o
n
H
oa
x
n
e
w
s
.
N
a
ï
v
e
B
a
ye
s
a
s
a
m
a
c
h
i
n
e
l
e
a
r
ni
ng
prob
a
bi
l
i
s
t
i
c
a
ppr
oa
c
h
t
e
nds
t
o
w
or
ks
w
e
l
l
f
or
h
a
ndl
i
ng
t
ra
i
ni
n
g
s
e
t
s
t
ha
t
c
ha
ng
e
ov
e
r
t
i
m
e
.
F
urt
h
e
rm
o
re
,
i
t
w
a
s
c
hos
e
n
be
c
a
us
e
N
a
ï
ve
B
a
y
e
s
ha
s
prove
n
t
o
prod
uc
e
goo
d,
fa
s
t
a
c
c
ura
c
y
a
nd
c
a
n
w
ork
w
e
l
l
on
t
he
ve
r
i
fi
c
a
t
i
o
n
of
s
e
n
t
i
m
e
n
t
a
na
l
ys
i
s
w
i
t
h
re
l
a
t
i
ve
l
y
f
e
w
t
ra
i
ni
ng
da
t
a
[15
,
2
5,
26]
.
In
s
e
ve
ra
l
pre
vi
ous
t
e
x
t
c
l
a
s
s
i
fi
c
a
t
i
on
a
nd
h
oa
x
de
t
e
c
t
i
on
s
t
udi
e
s
,
t
h
e
p
e
rfor
m
a
nc
e
of
c
l
a
s
s
i
fi
c
a
t
i
on
m
e
t
hods
c
a
n
b
e
o
pt
i
m
i
z
e
d
b
y
us
i
ng
f
e
a
t
ur
e
s
e
l
e
c
t
i
on
m
e
t
h
ods
s
uc
h
a
s
pa
r
t
i
c
l
e
s
w
a
rm
opt
i
m
i
z
a
t
i
on
(P
S
O
),
i
nf
or
m
a
t
i
on
gr
a
i
n
(IG
)
a
nd
ge
n
e
t
i
c
a
l
gori
t
h
m
(G
A
)
[5,
22
,
2
7,
28
]
.
In
t
he
pr
e
vi
o
us
s
t
ud
i
e
s
,
w
e
c
o
nc
l
ude
t
h
a
t
P
S
O
ha
s
s
e
v
e
ra
l
a
dv
a
nt
a
ge
s
ov
e
r
o
t
he
r
m
e
t
hods
,
s
uc
h
a
s
e
a
s
y
t
o
i
m
p
l
e
m
e
n
t
,
i
t
c
a
n
a
l
s
o
s
e
a
rc
h
for
op
t
i
m
a
l
va
l
u
e
s
a
nd
h
a
ve
a
l
gor
i
t
h
m
i
c
m
ode
l
s
t
ha
t
c
a
n
b
e
fur
t
he
r
i
m
prov
e
d
.
P
S
O
i
s
a
l
s
o
w
i
de
l
y
e
m
p
l
oye
d
i
n
t
he
prob
l
e
m
o
f
c
l
a
s
s
i
fi
c
a
t
i
on
,
c
l
us
t
e
r
i
ng
,
a
nd
s
e
l
e
c
t
i
on
of
t
e
x
t
f
e
a
t
ure
s
[29
-
31
]
.
A
ft
e
r
c
ondu
c
t
i
ng
a
n
a
l
ys
i
s
a
nd
hypot
h
e
s
i
s
ba
s
e
d
on
pr
e
vi
o
us
r
e
s
e
a
rc
h
,
t
h
i
s
pa
p
e
r
prop
os
e
d
a
l
gor
i
t
h
m
for
d
e
ve
l
op
i
ng
a
h
oa
x
n
e
w
s
d
e
t
e
c
t
i
o
n
s
ys
t
e
m
,
w
i
t
h
t
h
e
c
om
bi
n
a
t
i
on
o
f
s
e
a
rc
hi
ng
t
e
c
h
ni
qu
e
s
a
nd
i
t
s
opt
i
m
i
z
a
t
i
o
n,
a
nd
a
l
s
o
e
qui
ppe
d
w
i
t
h
s
e
nt
i
m
e
nt
a
na
l
ys
i
s
.
2.
R
ES
EA
R
C
H
M
ET
H
O
D
T
he
r
e
a
r
e
s
e
v
e
ra
l
m
e
t
h
ods
w
e
h
a
d
s
t
ud
i
e
d
i
n
t
h
e
ki
n
ds
o
f
l
i
t
e
ra
t
ure
.
T
hi
s
l
e
a
d
t
o
t
h
e
c
o
nc
l
us
i
on
t
ha
t
s
e
a
rc
h
t
e
c
hni
qu
e
i
s
m
o
re
pra
c
t
i
c
a
l
t
ha
n
l
e
a
r
ni
ng
t
e
c
hn
i
qu
e
for
ho
a
x
d
e
t
e
c
t
i
o
n.
T
hus
,
t
hi
s
p
a
pe
r
pro
pos
e
s
s
e
a
rc
hi
ng
t
e
c
h
ni
qu
e
s
t
o
c
l
a
s
s
i
fy
hoa
x
n
e
w
s
i
n
a
m
ore
pra
c
t
i
c
a
l
a
nd
up
t
o
da
t
e
m
a
nne
r
b
e
c
a
us
e
c
r
a
w
l
i
ng
proc
e
s
s
e
s
c
a
n
be
c
a
rr
i
e
d
out
e
ve
r
y
t
i
m
e
by
c
he
c
ki
n
g
t
h
e
ne
w
s
.
T
he
a
c
c
ur
a
c
y
of
t
h
e
re
s
u
l
t
s
i
s
m
uc
h
be
t
t
e
r
for
fre
qu
e
nt
s
e
a
rc
h
i
ng
a
s
t
h
e
q
ue
ry
ov
e
r
w
e
b
c
a
n
be
pos
e
d
e
ve
r
y
t
i
m
e
.
T
he
c
l
a
s
s
i
f
i
c
a
t
i
on
proc
e
s
s
i
s
don
e
us
i
ng
t
he
c
os
i
n
e
s
i
m
i
l
a
r
i
t
y
m
e
t
ri
c
.
F
urt
h
e
r
m
ore
,
t
he
n
e
w
s
w
a
r
e
t
he
n
fur
t
he
r
pro
c
e
s
s
e
d
by
t
he
s
e
nt
i
m
e
nt
a
n
a
l
ys
i
s
proc
e
s
s
us
i
ng
N
a
ï
ve
Ba
ye
s
.
T
hi
s
a
l
gori
t
hm
i
s
t
h
e
n
opt
i
m
i
z
e
d
by
t
h
e
P
S
O
.
T
o
b
e
fo
c
us
e
d,
s
e
nt
i
m
e
n
t
a
na
l
ys
i
s
i
s
c
a
rri
e
d
ou
t
on
l
y
for
ne
w
s
t
h
a
t
w
a
s
de
t
e
c
t
e
d
a
s
ho
a
x
ba
s
e
d
on
t
h
e
s
e
a
r
c
hi
ng
a
ppro
a
c
h.
O
n
t
h
e
o
t
he
r
h
a
nd
,
our
a
pp
roa
c
h
c
ra
w
l
s
d
a
t
a
fro
m
s
o
c
i
a
l
m
e
d
i
a
T
w
i
t
t
e
r
a
n
d
F
a
c
e
book
.
T
h
e
e
xpl
a
na
t
i
o
n
of
t
he
a
ppro
a
c
h
i
s
e
l
a
bora
t
e
d
w
i
t
h
t
h
e
f
ol
l
ow
i
ng
F
i
g
u
re
1.
2.
1
.
H
oax
d
e
t
e
c
t
i
on
Be
for
e
t
h
e
hoa
x
d
e
t
e
c
t
i
o
n
proc
e
s
s
i
s
c
a
r
ri
e
d
out
,
i
np
ut
qu
e
ri
e
s
a
r
e
pe
r
form
e
d
by
t
he
us
e
r,
qu
e
ri
e
s
from
t
h
e
us
e
r
c
o
nt
a
i
n
i
ng
t
h
e
k
e
yw
or
d
n
e
w
s
t
ha
t
w
i
l
l
b
e
s
e
a
r
c
he
d
.
Inpu
t
q
ue
r
i
e
s
a
r
e
us
e
d
t
o
c
ol
l
e
c
t
ne
w
s
da
t
a
.
D
a
t
a
c
ol
l
e
c
t
i
on
i
s
do
ne
by
c
ra
w
l
i
ng
t
o
re
t
ri
e
v
e
Indon
e
s
i
a
n
l
a
ngu
a
ge
n
e
w
s
s
n
i
pp
e
t
s
t
hr
ough
s
e
a
rc
hi
ng
fa
c
i
l
i
t
i
e
s
prov
i
d
e
d
by
G
oogl
e
by
ut
i
l
i
z
i
ng
t
h
e
G
oo
gl
e
A
P
I.
G
oogl
e
Cus
t
o
m
S
e
a
r
c
h
m
a
k
e
s
i
t
pos
s
i
b
l
e
t
o
m
a
ke
s
e
a
rc
h
e
ng
i
ne
s
a
s
de
s
i
r
e
d
.
W
h
e
r
e
t
h
e
w
e
b
s
n
i
ppe
t
pro
c
e
s
s
w
i
l
l
b
e
di
r
e
c
t
e
d
t
o
t
he
t
urnb
a
c
kh
oa
x
.
i
d
w
e
bs
i
t
e
,
s
t
ophoa
x.
i
d,
o
pe
r
a
i
n
.
b
l
ogs
pot
.
c
om
,
a
nd
a
y
om
a
j
u
t
e
r
us
.
b
l
ogs
p
ot
.
c
om
.
N
e
xt
,
t
h
e
s
i
m
i
l
a
ri
t
y
of
d
oc
u
m
e
nt
s
e
a
rc
h
re
s
ul
t
s
a
nd
t
e
x
t
i
np
ut
i
s
c
a
l
c
u
l
a
t
e
d
us
i
ng
c
os
i
n
e
s
i
m
i
l
a
r
i
t
y
.
T
he
re
s
u
l
t
s
of
t
he
c
a
l
c
ul
a
t
i
on
of
c
os
i
n
e
s
i
m
i
l
a
ri
t
y
w
i
l
l
prod
uc
e
a
p
e
r
c
e
n
t
a
g
e
of
hoa
x
r
e
s
ul
t
s
.
Cos
i
n
e
s
i
m
i
l
a
ri
t
y
(
c
s
)
c
a
n
be
c
a
l
c
u
l
a
t
e
d
by
t
h
e
form
ul
a
(1
)
[3
2]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
K
O
M
N
IK
A
T
e
l
e
c
om
m
un
Co
m
put
E
l
Con
t
rol
H
oax
c
l
as
s
i
f
i
c
at
i
on
and
s
e
n
t
i
m
e
nt
ana
l
y
s
i
s
of
Ind
one
s
i
an
ne
ws
us
i
ng
na
i
v
e
…
(
H
e
r
u
A
gus
San
t
os
o
)
801
=
c
o
s
(
)
=
.
‖
‖
‖
‖
=
∑
=
√
∑
2
=
√
∑
2
=
(1)
F
i
gure
1
.
T
he
w
orkf
l
ow
of
t
h
e
pr
opos
e
d
m
e
t
hod
2.
2
.
S
e
n
t
i
m
e
n
t
a
n
a
l
ys
i
s
T
o
fi
n
d
out
t
h
e
s
e
nt
i
m
e
n
t
t
ow
a
rds
t
h
e
n
e
w
s
,
t
he
n
t
he
da
t
a
s
e
a
rc
h
i
s
ba
s
e
d
on
t
he
s
e
l
e
c
t
e
d
n
e
w
s
.
N
e
xt
i
s
c
r
a
w
l
i
n
g
s
e
n
t
i
m
e
n
t
on
s
oc
i
a
l
m
e
di
a
w
e
bs
i
t
e
s
s
uc
h
a
s
F
a
c
e
bo
ok
a
nd
T
w
i
t
t
e
r
.
P
r
e
pro
c
e
s
s
i
ng
re
s
u
l
t
s
fro
m
c
ra
w
l
i
ng
da
t
a
t
o
op
t
i
m
i
z
e
fe
a
t
ur
e
e
x
t
ra
c
t
i
on
a
nd
c
l
a
s
s
i
f
i
c
a
t
i
o
n
re
s
u
l
t
s
.
P
re
proc
e
s
s
i
ng
c
o
ns
i
s
t
s
o
f
fo
l
di
n
g
c
a
s
e
s
,
fi
l
t
e
ri
ng,
t
oke
ni
z
i
ng
,
a
nd
s
t
e
m
m
i
ng
[33]
,
w
h
e
r
e
a
s
,
fo
r
d
a
t
a
a
g
gre
g
a
t
i
on
t
e
c
h
ni
qu
e
,
t
hi
s
w
ork
re
l
i
e
s
on
o
ur
pre
vi
ous
w
ork
pr
e
s
e
nt
e
d
i
n
[34
-
36
]
.
Ca
s
e
fo
l
di
n
g
i
s
don
e
t
o
c
ha
n
ge
a
l
l
l
e
t
t
e
rs
t
o
l
ow
e
rc
a
s
e
.
F
i
l
t
e
ri
ng
o
r
of
t
e
n
re
fe
r
re
d
t
o
a
s
s
t
o
p
w
ord
re
m
ov
a
l
i
s
us
e
d
t
o
d
e
l
e
t
e
w
ords
t
ha
t
a
r
e
no
t
t
oo
i
m
po
rt
a
nt
,
t
o
ke
ni
z
i
ng
i
s
us
e
d
t
o
bre
a
k
t
he
i
npu
t
of
t
he
qu
e
ry
i
nt
o
w
ords
pe
r
w
ord
,
a
nd
s
t
e
m
m
i
ng
i
s
us
e
d
t
o
r
e
m
ove
w
ord
a
dd
i
t
i
ons
s
o
t
h
a
t
t
h
e
ba
s
i
c
w
o
rds
a
r
e
a
t
t
a
c
h
e
d.
F
r
om
t
he
r
e
s
ul
t
s
of
pr
e
proc
e
s
s
i
ng
r
e
s
ul
t
s
,
i
t
i
s
c
a
l
c
ul
a
t
e
d
t
h
e
nu
m
b
e
r
of
o
c
c
ur
re
n
c
e
s
of
e
a
c
h
w
or
d
i
n
e
a
c
h
doc
um
e
nt
a
nd
t
he
n
c
a
l
c
u
l
a
t
e
T
F
ID
F
for
e
a
c
h
w
ord
w
i
t
h
t
he
for
m
u
l
a
(2)
[28]
.
=
∗
l
o
g
(2)
W
he
r
e
,
i
s
t
h
e
w
e
i
g
ht
of
,
i
s
t
he
nu
m
b
e
r
of
oc
c
urr
e
n
c
e
s
fro
m
,
t
h
e
nu
m
be
r
of
doc
um
e
nt
s
c
ont
a
i
n
i
ng
,
a
nd
i
s
t
he
t
o
t
a
l
n
u
m
be
r
of
doc
u
m
e
nt
s
.
A
ft
e
r
t
h
e
t
e
rm
w
e
i
ght
i
ng
va
l
ue
i
s
ob
t
a
i
ne
d
,
t
he
n
t
hi
s
w
e
i
gh
t
v
a
l
u
e
i
s
us
e
d
a
s
a
re
fe
r
e
nc
e
f
or
P
S
O
p
a
r
t
i
c
l
e
s
.
T
he
fi
r
s
t
s
t
e
p
i
n
P
S
O
i
s
a
n
i
nput
of
popu
l
a
t
i
on
nu
m
be
rs
.
E
a
c
h
pop
ul
a
t
i
o
n
i
n
i
t
i
a
l
i
z
e
s
p
a
rt
i
c
l
e
s
t
ha
t
re
p
re
s
e
n
t
e
a
c
h
fe
a
t
u
re
/
w
o
rd
w
i
t
h
pos
i
t
i
o
n
=
r
a
ndo
m
n
um
b
e
rs
fro
m
0
-
1
a
nd
ve
l
oc
i
t
y
=
0
.
T
h
e
n
s
o
rt
by
t
h
e
h
i
gh
e
s
t
pos
i
t
i
on
va
l
u
e
.
N
e
xt
,
c
a
l
c
ul
a
t
e
t
he
N
B
c
a
t
e
gor
i
z
a
t
i
on
w
i
t
h
t
h
e
r
e
du
c
e
d
f
e
a
t
u
re
ba
s
e
d
on
t
h
e
hi
gh
e
s
t
pa
rt
i
c
l
e
pos
i
t
i
on
va
l
u
e
.
A
t
e
r
m
w
i
t
h
l
ow
v
a
l
u
e
w
i
l
l
no
t
be
us
e
d
for
c
l
a
s
s
i
fi
c
a
t
i
on.
I
t
m
e
a
ns
t
ha
t
pa
rt
i
c
l
e
v
a
l
u
e
s
a
r
e
re
s
t
r
i
c
t
e
d
t
o
a
c
e
r
t
a
i
n
ra
nk
,
for
e
x
a
m
p
l
e
,
i
f
t
h
e
re
a
r
e
32
pa
r
t
i
c
l
e
s
,
a
nd
a
re
l
i
m
i
t
e
d
t
o
2
0
p
a
rt
i
c
l
e
s
w
i
t
h
t
he
hi
g
he
s
t
v
a
l
u
e
t
he
n
p
a
rt
i
c
l
e
s
i
n
t
h
e
or
de
r
of
21
t
o
32
a
r
e
not
us
e
d.
N
e
x
t
,
do
t
he
p
roba
bi
l
i
t
y
c
a
l
c
u
l
a
t
i
on
us
i
n
g
fo
rm
u
l
a
(3)
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
1693
-
6930
T
E
L
K
O
M
N
IK
A
T
e
l
e
c
om
m
un
Co
m
put
E
l
Con
t
rol
,
V
ol
.
18
,
N
o.
2
,
A
pri
l
2
020:
799
-
80
6
802
(
|
)
=
1
√
2
e
x
p
[
−
(
−
)
2
2
(
)
2
(3)
P
roba
bi
l
i
t
y
r
e
s
ul
t
s
w
i
l
l
ge
t
a
c
a
t
e
go
ry
or
c
l
a
s
s
f
rom
e
a
c
h
do
c
um
e
nt
,
t
h
e
n
r
e
pe
a
t
t
h
i
s
c
a
l
c
u
l
a
t
i
on
on
a
l
l
doc
u
m
e
nt
s
t
o
c
a
l
c
ul
a
t
e
t
h
e
a
c
c
ura
c
y
i
n
t
he
n
e
xt
proc
e
s
s
.
Ca
l
c
u
l
a
t
e
t
he
N
a
ï
ve
B
a
y
e
s
a
c
c
ur
a
c
y
of
e
a
c
h
popul
a
t
i
on
by
f
orm
ul
a
(4)
.
=
ℎ
(4)
T
he
n
c
a
l
c
u
l
a
t
e
w
he
t
h
e
r
t
he
N
a
ï
v
e
Ba
y
e
s
a
c
c
ur
a
c
y
i
s
b
e
t
t
e
r
t
h
a
n
t
h
e
be
s
t
a
c
c
ur
a
c
y
a
nd
t
h
e
be
s
t
a
c
c
ura
c
y
.
If
t
h
e
a
c
c
ura
c
y
of
N
a
ï
v
e
B
a
y
e
s
i
n
t
h
e
c
urr
e
nt
popul
a
t
i
on
i
s
be
t
t
e
r
t
ha
n
P
b
e
s
t
a
nd
G
b
e
s
t
t
h
e
n
t
he
p
opul
a
t
i
on
i
s
now
us
e
d
a
s
t
h
e
n
e
w
P
be
s
t
a
nd
G
b
e
s
t
.
T
o
c
a
l
c
ul
a
t
e
t
h
e
s
pe
e
d
a
nd
upd
a
t
e
pos
i
t
i
on
of
p
a
rt
i
c
l
e
pos
i
t
i
ons
us
i
n
g
for
m
u
l
a
(5)
a
nd
for
p
a
rt
i
c
l
e
,
p
os
i
t
i
on
upda
t
e
s
us
i
ng
fo
rm
u
l
a
(6)
.
T
o
s
e
e
m
or
e
c
l
e
a
rl
y
a
bout
t
h
e
fl
ow
o
f
t
he
pr
o
c
e
s
s
a
t
t
hi
s
s
t
a
g
e
c
a
n
s
e
e
F
i
gur
e
2
.
F
i
gure
2
.
D
e
t
a
i
l
s
pro
c
e
s
s
of
N
B
C
l
a
s
s
i
f
i
e
r
w
i
t
h
P
S
O
opt
i
m
i
z
a
t
i
on
f
or
h
oa
x
d
e
t
e
c
t
i
on
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
K
O
M
N
IK
A
T
e
l
e
c
om
m
un
Co
m
put
E
l
Con
t
rol
H
oax
c
l
as
s
i
f
i
c
at
i
on
and
s
e
n
t
i
m
e
nt
ana
l
y
s
i
s
of
Ind
one
s
i
an
ne
ws
us
i
ng
na
i
v
e
…
(
H
e
r
u
A
gus
San
t
os
o
)
803
(
)
=
(
−
1
)
+
1
1
[
−
(
)
]
+
2
2
[
−
(
)
]
(5
)
(
)
=
(
−
1
)
+
(
)
(6
)
Re
pe
a
t
t
he
s
t
e
ps
t
o
c
a
l
c
u
l
a
t
e
t
he
prob
a
b
i
l
i
t
i
e
s
of
t
he
P
Be
s
t
a
nd
G
B
e
s
t
unt
i
l
t
h
e
i
t
e
r
a
t
i
on
i
s
c
o
m
pl
e
t
e
a
nd
p
rodu
c
e
a
m
od
e
l
w
i
t
h
G
b
e
s
t
a
c
c
ur
a
c
y.
W
h
e
n
t
h
e
i
t
e
r
a
t
i
on
i
s
c
om
p
l
e
t
e
,
t
he
s
e
qu
e
nt
i
a
l
p
opul
a
t
i
on
i
s
obt
a
i
ne
d
fro
m
t
h
e
h
i
gh
e
s
t
t
o
l
ow
e
s
t
N
a
ï
ve
B
a
ye
s
a
c
c
ur
a
c
y,
a
nd
t
h
e
G
Be
s
t
va
l
ue
i
s
us
e
d
a
s
a
fe
a
t
ur
e
m
od
e
l
t
o
produc
e
a
n
a
l
y
t
i
c
a
l
s
e
n
t
i
m
e
nt
.
T
o
s
e
e
t
h
e
fl
ow
of
t
h
e
a
l
gori
t
h
m
m
o
re
c
l
e
a
rl
y
you
c
a
n
s
e
e
F
i
g
ure
2.
T
he
n
pe
rfor
m
a
N
a
ï
ve
Ba
y
e
s
pr
oba
b
i
l
i
t
y
c
a
l
c
ul
a
t
i
on
i
n
t
hre
e
c
l
a
s
s
e
s
,
n
a
m
e
l
y
pos
i
t
i
v
e
,
ne
ga
t
i
v
e
a
nd
n
e
ut
ra
l
.
F
urt
he
r
m
or
e
,
t
he
re
s
u
l
t
s
of
c
l
a
s
s
c
l
a
s
s
i
f
i
c
a
t
i
on
a
r
e
w
r
i
t
t
e
n
w
i
t
h
t
h
e
hi
g
h
e
s
t
prob
a
bi
l
i
t
y
va
l
ue
.
3.
R
ES
U
LTS
A
N
D
D
I
S
C
U
S
S
I
O
N
In
t
hi
s
re
s
e
a
r
c
h
us
e
d
hoa
x
n
e
w
s
da
t
a
t
h
a
t
i
s
w
i
de
l
y
di
s
s
e
m
i
na
t
e
d
t
hr
ough
s
oc
i
a
l
m
e
di
a
by
c
r
a
w
l
i
ng
da
t
a
,
fro
m
t
he
da
t
a
t
ha
t
h
a
v
e
b
e
e
n
s
e
a
rc
h
e
d
f
or
30
ne
w
s
hoa
x
s
a
m
pl
e
s
,
n
e
w
s
s
a
m
pl
e
s
c
a
n
be
s
e
e
n
i
n
T
a
b
l
e
1.
Fr
om
a
l
l
t
he
n
e
w
s
d
a
t
a
,
c
os
i
ne
s
i
m
i
l
a
r
i
t
y
i
s
don
e
a
nd
t
h
e
c
a
l
c
ul
a
t
i
on
v
a
l
u
e
c
a
n
b
e
s
e
e
n
i
n
T
a
b
l
e
2
.
T
a
b
l
e
1
.
S
a
m
p
l
e
of
i
nd
one
s
i
a
n
ne
w
s
a
n
d
i
t
s
c
a
t
e
g
ory
No
Title
1
W
e
b
K
P
U
D
i
re
t
a
s
,
T
e
m
u
a
n
M
e
n
g
e
j
u
t
a
n
!!
J
o
k
o
w
i
A
n
g
k
a
t
Is
u
P
K
I
2
Bo
c
o
ra
n
In
fo
rm
a
s
i
P
e
n
t
i
n
g
V
a
l
i
d
P
o
l
a
K
e
c
u
ra
n
g
a
n
S
i
s
t
e
m
P
e
n
g
h
i
t
u
n
g
a
n
S
u
a
ra
K
P
U
D
e
n
g
a
n
M
o
d
u
s
N
o
m
o
r
0
1
d
a
n
0
2
3
K
e
rt
a
s
s
u
a
ra
P
e
m
i
l
u
d
i
b
a
k
a
r
s
e
p
e
rt
i
s
a
m
p
a
h
,
k
e
c
u
ra
n
g
a
n
i
n
i
m
a
u
d
i
d
i
a
m
k
a
n
k
a
re
n
a
d
i
l
i
n
d
u
n
g
i
o
l
e
h
a
p
a
ra
t
d
a
n
p
e
j
a
b
a
t
?
4
M
e
n
p
o
ra
Im
a
m
N
a
h
ra
w
i
M
u
n
d
u
r
D
a
ri
J
a
b
a
t
a
n
n
y
a
5
G
a
m
b
a
r
Ra
n
c
a
n
g
a
n
G
e
d
u
n
g
Is
t
a
n
a
N
e
g
a
ra
d
i
P
a
l
a
n
g
k
a
ra
y
a
6
M
e
g
a
w
a
t
i
S
o
e
k
a
rn
o
p
u
t
ri
D
i
ra
w
a
t
d
i
Ru
m
a
h
S
a
k
i
t
k
a
re
n
a
S
t
ro
k
e
…
…
30
s
i
m
p
a
t
i
s
a
n
p
k
i
b
a
c
o
k
s
e
o
ra
n
g
u
l
a
m
a
d
i
d
a
e
ra
h
b
a
n
t
e
n
T
a
b
l
e
2
.
Cos
i
ne
s
i
m
i
l
a
r
i
t
y
re
s
u
l
t
s
No
Co
s
i
n
e
S
i
m
i
l
a
ri
t
y
No
Co
s
i
n
e
S
i
m
i
l
a
ri
t
y
No
Co
s
i
n
e
S
i
m
i
l
a
ri
t
y
(%
)
1
8
9
.
5
6
6
9
11
7
4
.
8
4
5
5
21
6
6
.
2
2
6
6
2
6
6
.
7
4
2
4
12
7
2
.
6
2
7
3
22
8
6
.
0
6
6
3
3
8
8
.
6
4
0
5
13
9
0
.
8
6
8
8
23
8
4
.
6
3
2
7
4
6
7
.
8
8
4
4
14
7
1
.
9
0
9
2
24
6
8
.
1
3
8
5
5
7
5
.
0
5
8
7
15
7
1
.
4
4
3
5
25
7
6
.
7
3
6
6
6
9
1
.
6
3
4
2
16
8
1
.
1
1
0
7
26
6
7
.
3
4
3
5
7
6
6
.
9
4
3
9
17
7
5
.
3
7
7
8
27
7
6
.
3
3
2
3
8
6
7
.
1
9
3
7
18
7
7
.
4
0
7
0
28
8
5
.
5
2
3
6
9
7
3
.
7
2
1
0
19
7
4
.
7
2
6
5
29
7
7
.
6
8
9
9
10
8
6
.
7
2
2
7
20
6
9
.
2
3
0
8
30
8
9
.
9
6
4
7
F
rom
t
he
30
d
a
t
a
a
bov
e
,
t
h
e
p
e
r
c
e
n
t
a
g
e
v
a
l
u
e
o
f
t
he
a
v
e
ra
g
e
c
os
i
ne
s
i
m
i
l
a
r
i
t
y
c
a
l
c
ul
a
t
i
on
i
s
a
ro
und
77,
077
%.
T
he
m
e
a
ni
n
g
a
pp
e
a
rs
t
h
a
t
t
hi
s
m
e
t
ho
d
c
a
n
i
de
n
t
i
f
y
ho
a
x
ne
w
s
w
e
l
l
,
w
h
e
re
t
he
h
i
ghe
s
t
c
s
va
l
ue
i
s
91.
634
2%
a
nd
t
h
e
l
ow
e
s
t
i
s
66
.
226
6%
,
a
l
t
h
ough
t
h
e
a
ve
r
a
ge
pe
rc
e
nt
a
g
e
v
a
l
u
e
i
s
not
hi
gh
a
l
l
c
a
l
c
ul
a
t
i
ons
l
e
a
d
t
o
t
h
e
c
orre
c
t
c
l
a
s
s
i
fi
c
a
t
i
o
n.
In
t
h
e
ne
xt
pr
oc
e
s
s
,
t
h
e
c
a
l
c
u
l
a
t
i
on
of
s
e
n
t
i
m
e
n
t
a
na
l
ys
i
s
on
t
he
ne
w
s
w
a
s
c
a
rr
i
e
d
out
us
i
ng
t
he
n
a
ï
v
e
Ba
y
e
s
a
nd
P
S
O
m
e
t
ho
ds
t
ha
t
ha
d
b
e
e
n
pro
pos
e
d
pr
e
vi
ous
l
y
.
S
e
nt
i
m
e
n
t
a
n
a
l
ys
i
s
w
a
s
di
vi
d
e
d
i
n
t
o
t
hre
e
c
a
t
e
g
ori
e
s
,
na
m
e
l
y
p
os
i
t
i
ve
,
ne
g
a
t
i
ve
a
nd
ne
ut
r
a
l
.
T
a
b
l
e
3
s
how
s
t
h
e
r
e
s
ul
t
s
of
t
he
s
e
n
t
i
m
e
n
t
a
na
l
ys
i
s
of
t
h
e
p
ropos
e
d
m
e
t
hod
.
F
rom
t
h
e
s
e
re
s
u
l
t
s
,
i
t
c
a
n
b
e
c
onc
l
ud
e
d
t
h
a
t
t
he
r
e
a
r
e
19
r
e
s
ul
t
s
of
t
he
c
orr
e
c
t
c
hi
l
d
s
e
n
t
i
m
e
nt
.
A
l
t
houg
h
t
he
l
e
v
e
l
of
a
c
c
ura
c
y
of
t
h
e
s
e
nt
i
m
e
nt
i
s
pr
oba
b
l
y
no
t
ve
ry
h
i
gh
,
t
he
a
c
c
ura
c
y
of
t
h
e
s
e
n
t
i
m
e
n
t
a
na
l
ys
i
s
i
s
s
t
i
l
l
be
t
t
e
r
a
nd
fa
s
t
e
r
c
o
m
pa
r
e
d
t
o
ot
h
e
r
m
e
t
hod
s
s
uc
h
a
s
t
h
e
K
N
N
.
T
he
pro
c
e
s
s
of
c
a
l
c
ul
a
t
i
ng
t
he
s
e
nt
i
m
e
nt
a
n
a
l
ys
i
s
f
or
e
a
c
h
do
c
um
e
nt
i
s
a
l
s
o
f
a
s
t
e
r
t
h
a
n
t
h
a
t
of
t
he
K
N
N
w
h
e
re
t
h
e
N
B
m
e
t
hod
c
a
n
c
a
l
c
u
l
a
t
e
t
h
e
a
ve
r
a
ge
of
e
a
c
h
do
c
um
e
nt
0
.
4733
s
e
c
onds
a
nd
t
he
K
N
N
c
a
l
c
u
l
a
t
e
s
t
he
a
v
e
r
a
ge
of
e
a
c
h
doc
u
m
e
n
t
6
,
213
s
e
c
onds
w
i
t
h
t
he
s
a
m
e
c
o
m
pu
t
e
r
s
pe
c
i
fi
c
a
t
i
on.
T
a
b
l
e
4
s
how
s
t
h
e
c
om
p
a
ri
s
on
of
t
h
e
r
e
s
ul
t
s
of
t
h
e
c
l
a
s
s
i
f
i
c
a
t
i
on
s
e
nt
i
m
e
nt
a
na
l
ys
i
s
b
e
t
w
e
e
n
t
he
N
a
ï
ve
Ba
ye
s
m
e
t
h
od,
N
a
ï
v
e
B
a
ye
s
+
P
S
O
,
K
N
N
on
l
y
,
a
nd
K
N
N
+
P
S
O
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
1693
-
6930
T
E
L
K
O
M
N
IK
A
T
e
l
e
c
om
m
un
Co
m
put
E
l
Con
t
rol
,
V
ol
.
18
,
N
o.
2
,
A
pri
l
2
020:
799
-
80
6
804
T
a
b
l
e
3
.
S
e
nt
i
m
e
nt
a
n
a
l
ys
i
s
r
e
s
ul
t
s
N
u
m
b
e
r
o
f
N
e
w
s
S
e
n
t
i
m
e
n
t
c
l
a
s
s
S
e
n
t
i
m
e
n
t
Re
s
u
l
t
s
1
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
2
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
3
n
e
u
t
ra
l
p
o
s
i
t
i
v
e
4
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
5
n
e
u
t
ra
l
n
e
u
t
ra
l
6
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
7
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
8
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
9
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
10
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
11
n
e
u
t
ra
l
p
o
s
i
t
i
v
e
12
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
13
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
14
n
e
u
t
ra
l
n
e
g
a
t
i
v
e
15
p
o
s
i
t
i
v
e
n
e
u
t
ra
l
16
n
e
u
t
ra
l
p
o
s
i
t
i
v
e
17
n
e
u
t
ra
l
n
e
u
t
ra
l
18
p
o
s
i
t
i
v
e
n
e
g
a
t
i
v
e
19
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
20
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
21
p
o
s
i
t
i
v
e
n
e
g
a
t
i
v
e
22
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
23
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
24
n
e
g
a
t
i
v
e
n
e
u
t
ra
l
25
n
e
g
a
t
i
v
e
p
o
s
i
t
i
v
e
26
n
e
u
t
ra
l
n
e
u
t
ra
l
27
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
28
n
e
g
a
t
i
v
e
p
o
s
i
t
i
v
e
29
n
e
g
a
t
i
v
e
p
o
s
i
t
i
v
e
30
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
T
a
b
l
e
4
.
Co
m
p
a
ri
s
on
s
e
n
t
i
m
e
nt
a
n
a
l
ys
i
s
r
e
s
ul
t
s
of
e
a
c
h
m
e
t
h
od
N
u
m
b
e
r
o
f
N
e
w
s
S
e
n
t
i
m
e
n
t
c
l
a
s
s
KNN
N
a
ï
v
e
Ba
y
e
s
K
N
N
+
P
S
O
N
a
ï
v
e
Ba
y
e
s
+
PSO
(P
ro
p
o
s
e
d
M
e
t
h
o
d
)
1
n
e
g
a
t
i
v
e
p
o
s
i
t
i
v
e
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
2
n
e
g
a
t
i
v
e
n
e
u
t
ra
l
n
e
g
a
t
i
v
e
n
e
u
t
ra
l
n
e
g
a
t
i
v
e
3
n
e
u
t
ra
l
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
p
o
s
i
t
i
v
e
4
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
5
n
e
u
t
ra
l
p
o
s
i
t
i
v
e
n
e
g
a
t
i
v
e
n
e
u
t
ra
l
n
e
u
t
ra
l
6
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
7
n
e
g
a
t
i
v
e
n
e
u
t
ra
l
n
e
g
a
t
i
v
e
n
e
u
t
ra
l
n
e
g
a
t
i
v
e
8
n
e
g
a
t
i
v
e
n
e
u
t
ra
l
n
e
g
a
t
i
v
e
n
e
u
t
ra
l
n
e
g
a
t
i
v
e
9
n
e
g
a
t
i
v
e
n
e
u
t
ra
l
n
e
g
a
t
i
v
e
n
e
u
t
ra
l
n
e
g
a
t
i
v
e
10
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
11
n
e
u
t
ra
l
n
e
u
t
ra
l
p
o
s
i
t
i
v
e
n
e
u
t
ra
l
p
o
s
i
t
i
v
e
12
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
13
n
e
g
a
t
i
v
e
p
o
s
i
t
i
v
e
n
e
g
a
t
i
v
e
p
o
s
i
t
i
v
e
n
e
g
a
t
i
v
e
14
n
e
u
t
ra
l
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
15
p
o
s
i
t
i
v
e
n
e
u
t
ra
l
n
e
u
t
ra
l
n
e
u
t
ra
l
n
e
u
t
ra
l
16
n
e
u
t
ra
l
n
e
u
t
ra
l
p
o
s
i
t
i
v
e
n
e
u
t
ra
l
p
o
s
i
t
i
v
e
17
n
e
u
t
ra
l
n
e
u
t
ra
l
n
e
u
t
ra
l
n
e
u
t
ra
l
n
e
u
t
ra
l
18
p
o
s
i
t
i
v
e
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
19
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
20
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
21
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
22
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
23
n
e
g
a
t
i
v
e
p
o
s
i
t
i
v
e
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
24
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
n
e
u
t
ra
l
n
e
g
a
t
i
v
e
n
e
u
t
ra
l
25
n
e
g
a
t
i
v
e
n
e
u
t
ra
l
p
o
s
i
t
i
v
e
n
e
u
t
ra
l
p
o
s
i
t
i
v
e
26
n
e
u
t
ra
l
n
e
u
t
ra
l
n
e
u
t
ra
l
n
e
u
t
ra
l
n
e
u
t
ra
l
27
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
28
n
e
g
a
t
i
v
e
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
p
o
s
i
t
i
v
e
29
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
p
o
s
i
t
i
v
e
n
e
g
a
t
i
v
e
p
o
s
i
t
i
v
e
30
n
e
g
a
t
i
v
e
n
e
u
t
ra
l
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
n
e
g
a
t
i
v
e
N
u
m
b
e
r
o
f
c
o
rre
c
t
c
l
a
s
s
i
fi
c
a
t
i
o
n
17
18
18
19
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
K
O
M
N
IK
A
T
e
l
e
c
om
m
un
Co
m
put
E
l
Con
t
rol
H
oax
c
l
as
s
i
f
i
c
at
i
on
and
s
e
n
t
i
m
e
nt
ana
l
y
s
i
s
of
Ind
one
s
i
an
ne
ws
us
i
ng
na
i
v
e
…
(
H
e
r
u
A
gus
San
t
os
o
)
805
4.
C
O
N
C
LU
S
I
O
N
In
t
hi
s
r
e
s
e
a
rc
h
,
ho
a
x
d
e
t
e
c
t
i
on
m
e
t
ho
ds
ha
v
e
b
e
e
n
p
ropos
e
d
us
i
ng
s
e
a
rc
hi
ng
m
e
t
hods
,
us
e
rs
e
nt
e
r
que
ri
e
s
t
o
s
e
a
rc
h
f
or
ne
w
s
t
h
a
t
i
s
c
ons
i
d
e
r
e
d
hoa
xe
s
.
A
f
t
e
r
t
he
ho
a
x
ne
w
s
t
i
t
l
e
i
s
ob
t
a
i
ne
d,
c
l
a
s
s
i
fi
c
a
t
i
on
i
s
done
us
i
ng
t
h
e
s
e
a
rc
h
i
ng
m
e
t
hod
us
i
ng
G
oog
l
e
c
us
t
o
m
s
e
a
rc
h
a
nd
s
n
i
ppe
t
.
T
h
e
re
s
u
l
t
s
a
re
c
l
a
s
s
i
fi
e
d
by
t
he
c
os
i
n
e
s
i
m
i
l
a
r
i
t
y
m
e
t
hod
,
b
a
s
e
d
on
t
he
r
e
s
ul
t
s
of
t
e
s
t
i
ng
of
30
n
e
w
s
,
t
he
a
ve
ra
g
e
ho
a
x
i
s
77%
,
w
he
r
e
a
l
l
t
he
ne
w
s
i
s
de
t
e
c
t
e
d
a
s
a
h
oa
x
w
i
t
h
a
m
i
ni
m
um
p
e
rc
e
n
t
a
g
e
of
a
bou
t
66
%
a
nd
a
m
a
x
i
m
u
m
of
9
1%
.
T
hi
s
s
how
s
t
ha
t
t
h
e
pe
rf
orm
a
nc
e
of
t
h
e
propos
e
d
m
e
t
ho
d
i
s
r
e
l
i
a
bl
e
e
n
ough
t
o
d
e
t
e
c
t
hoa
x
n
e
w
s
.
T
hi
s
s
ys
t
e
m
i
s
a
l
s
o
e
qu
i
pp
e
d
w
i
t
h
s
e
nt
i
m
e
nt
a
na
l
ys
i
s
pro
c
e
s
s
us
i
ng
N
a
ï
v
e
B
a
y
e
s
w
h
i
c
h
i
s
o
pt
i
m
i
z
e
d
by
t
h
e
P
S
O
m
e
t
h
od,
ba
s
e
d
on
t
h
e
r
e
s
ul
t
s
of
t
e
s
t
i
n
g
t
he
s
e
n
t
i
m
e
n
t
a
na
l
ys
i
s
m
e
t
h
od
of
t
h
e
p
ropos
e
d
s
e
nt
i
m
e
nt
w
orks
b
e
t
t
e
r
t
ha
n
t
he
ot
h
e
r
m
e
t
h
ods
pr
opos
e
d
e
a
r
l
i
e
r.
A
C
K
N
O
WL
ED
G
E
M
EN
TS
T
he
a
u
t
hor
w
oul
d
l
i
k
e
t
o
gr
a
t
e
ful
l
y
a
c
know
l
e
dg
e
t
he
M
i
n
i
s
t
ry
of
Re
s
e
a
r
c
h
a
nd
H
i
gh
e
r
E
du
c
a
t
i
on
of
t
h
e
Re
p
ubl
i
c
of
Indon
e
s
i
a
u
nde
r
t
he
Re
s
e
a
r
c
h
G
ra
nd
w
i
t
h
c
ont
r
a
c
t
num
be
r
:
0
60/
A
38
.
04
/
U
D
N
-
09/
V
/
2019
.
R
EF
ER
EN
C
ES
[
1]
A
.
B
ond
i
e
l
l
i
a
nd
F
.
M
a
r
c
e
l
l
on
i
,
“
A
S
ur
v
e
y
o
n
F
a
ke
N
e
w
s
a
nd
R
u
m
ou
r
D
e
t
e
c
t
i
on
T
e
c
hni
que
s
,
”
I
nf
.
S
c
i
.
(
N
y
)
,
v
ol
.
497,
p
p.
38
–
55
,
2019
.
[
2]
P
.
A
s
s
i
r
oj
,
M
e
yl
i
a
na
,
A
.
N
.
H
i
da
ya
nt
o,
H
.
P
r
a
b
ow
o,
a
nd
H
.
L
.
H
.
S
.
W
a
r
na
r
s
,
“
H
oa
x
N
e
w
s
D
e
t
e
c
t
i
on
on
S
o
c
i
a
l
M
e
d
i
a
:
A
S
u
r
ve
y,
”
I
nd
o
.
A
s
s
oc
i
.
f
or
P
a
t
t
e
r
n
R
e
c
o
gni
t
i
on
I
nt
.
C
onf
.
pp.
18
6
–
19
1
,
20
18
[
3]
F
.
E
.
N
a
s
t
i
t
i
,
R
.
A
.
P
r
a
s
t
y
a
nt
i
,
R
.
B
.
T
a
r
u
no,
a
nd
D
.
H
a
r
i
ya
d
i
,
“
S
o
c
i
a
l
M
e
d
i
a
W
a
r
f
a
r
e
i
n
I
ndon
e
s
i
a
P
o
l
i
t
i
c
a
l
C
a
m
p
a
i
gn:
A
S
u
r
ve
y
,
”
I
nt
.
C
on
f
.
on
I
nf
or
m
a
t
i
on
T
e
c
hno
l
ogy
,
I
nf
o
r
.
S
ys
.
a
nd
E
l
e
c
t.
E
n
gi
n
e
e
,
pp
.
4
9
-
53
,
20
18
.
[
4]
A
.
B
.
P
r
a
s
e
t
i
j
o
,
e
t
al
.
,
“
H
oa
x
D
e
t
e
c
t
i
on
S
ys
t
e
m
on
I
ndon
e
s
i
a
n
N
e
w
s
S
i
t
e
s
B
a
s
e
d
o
n
T
e
x
t
C
l
a
s
s
i
f
i
c
a
t
i
on
u
s
i
n
g
S
V
M
a
nd
S
G
D
,
”
4
t
h
I
nt
.
C
onf
.
on
I
nf
o
r
m
.
T
e
c
hn
ol
ogy
,
C
om
p,
a
nd
E
l
e
c
t
.
E
ngi
n
e
e
.,
p
p.
45
-
49
,
201
7.
[
5]
A
.
P
.
W
i
j
a
ya
a
nd
H
.
A
.
S
a
n
t
os
o
,
“
I
m
pr
o
vi
n
g
t
he
A
c
c
ur
a
c
y
o
f
N
a
ï
v
e
B
a
ye
s
A
l
gor
i
t
h
m
f
o
r
H
o
a
x
C
l
a
s
s
i
f
i
c
a
t
i
on
U
s
i
ng
P
a
r
t
i
c
l
e
S
w
a
r
m
O
pt
i
m
i
z
a
t
i
o
n,
”
I
n
t
.
S
e
m
i
n
.
A
pp
l
.
T
e
c
hn
ol
.
I
n
f
.
C
om
m
un
.,
p
p.
48
2
-
4
87
,
2
018
.
[
6]
A
.
B
.
P
r
a
t
a
m
a
,
“
A
da
80
0
R
i
bu
S
i
t
u
s
P
e
ny
e
ba
r
H
oa
x
di
I
ndone
s
i
a
,
”
C
N
N
I
ndon
e
s
i
a
,
A
v
a
l
a
bl
e
a
t
:
c
nni
ndon
e
s
i
a
.
c
o
m
/
t
e
kn
ol
o
gi
/
20
1612
2917
0130
-
18
5
-
1
8295
6/
a
da
-
800
-
r
i
bu
-
s
i
t
u
s
-
pe
ny
e
ba
r
-
hoa
x
-
di
-
i
n
done
s
i
a
,
2
016
.
[
7]
I
.
Y
.
R
.
P
r
a
t
i
w
i
,
R
.
A
.
A
s
m
a
r
a
,
a
nd
F
.
R
a
hut
om
o,
“
S
t
ud
y
o
f
H
o
a
x
N
e
w
s
D
e
t
e
c
t
i
on
us
i
ng
N
a
ï
ve
B
a
y
e
s
C
l
a
s
s
i
f
i
e
r
i
n
I
ndone
s
i
a
n
L
a
ngu
a
ge
,
”
11
t
h
I
nt
.
C
on
f
.
on
I
nf
or
m
.
C
om
m
u
T
e
c
hn
ol
o
gy
a
nd
Sy
s
,
pp
.
73
–
78
,
2
017
.
[
8]
S
.
A
f
r
o
z
,
M
.
B
r
e
n
na
n
,
a
nd
R
.
G
r
e
e
ns
t
a
d
t
,
“
D
e
t
e
c
t
i
ng
H
oa
x
e
s
,
F
r
a
uds
,
a
nd
D
e
c
e
pt
i
on
i
n
W
r
i
t
i
ng
S
t
y
l
e
O
nl
i
n
e
,
”
I
E
E
E
Sy
m
po
s
i
um
on
Se
c
ur
i
t
y
a
nd
P
r
i
v
a
c
y
,
p
p.
46
1
-
475
,
2012
.
[
9]
J
.
C
ha
e
,
e
t
a
l
.
,
“
V
i
s
u
a
l
A
na
l
y
t
i
c
s
f
o
r
I
nv
e
s
t
i
g
a
t
i
v
e
A
n
a
l
y
s
i
s
of
H
o
a
x
D
i
s
t
r
e
s
s
C
a
l
l
s
us
i
ng
S
o
c
i
a
l
M
e
d
i
a
,
”
I
E
E
E
Sy
m
po
s
i
um
on
T
e
c
hno
l
og
i
e
s
f
or
H
om
e
l
a
nd
Se
c
ur
i
t
y
,
pp
.
1
-
6
,
2016
.
[
10]
I
.
S
a
nt
os
o
,
e
t
al
.
,
“
E
a
r
l
y
I
nv
e
s
t
i
ga
t
i
o
n
of
P
r
op
os
e
d
H
oa
x
D
e
t
e
c
t
i
on
f
or
D
e
c
r
e
a
s
i
ng
H
o
a
x
i
n
S
oc
i
a
l
M
e
d
i
a
,
”
I
E
E
E
I
nt
.
C
o
nf
.
on
C
ybe
r
ne
t
i
c
s
a
nd
C
o
m
pu
t
a
t
i
o
na
l
I
n
t
e
l
l
i
ge
nc
e
,
pp
.
175
-
17
9
,
20
17
.
[
11]
N
.
A
.
M
i
f
t
a
hu
l
H
ud
a
a
nd
I
.
S
e
m
b
i
r
i
n
g,
“
T
h
e
U
s
e
o
f
S
of
t
S
y
s
t
e
m
s
M
e
t
hod
ol
o
gy
t
o
R
e
s
ol
v
e
H
o
a
x
N
e
w
s
P
r
ob
l
e
m
s
i
n
I
ndone
s
i
a
,
”
3
r
d
I
n
t
.
C
o
nf
.
on
I
n
f
o
r
m
.
T
e
c
hno
l
og
y,
I
nf
o
r
m
.
S
y
s
.
a
nd
E
l
e
c
t
.
E
ngi
ne
e
,
pp
.
65
-
68
,
2
018
.
[
12]
F
.
D
e
l
V
i
gna
,
e
t
al
.
,
“
W
ho
F
r
a
m
e
d
R
o
ge
r
R
e
i
nde
e
r
?
D
e
-
c
e
ns
o
r
s
hi
p
of
F
a
c
e
boo
k
P
o
s
t
s
by
S
n
i
ppe
t
C
l
a
s
s
i
f
i
c
a
t
i
on
,
”
O
nl
i
ne
Soc
.
N
e
t
w
or
k
s
M
e
d
i
a
,
v
ol
.
6,
pp
.
41
-
57
,
201
8.
[
13]
P
.
M
u
r
ug
e
s
a
n
a
nd
K
.
M
a
l
a
t
hi
,
“
E
f
f
i
c
i
e
n
t
S
e
a
r
c
h
E
ng
i
n
e
A
p
pr
o
a
c
h
f
or
M
e
a
s
ur
i
n
g
S
i
m
i
l
a
r
i
t
y
B
e
t
w
e
e
n
W
o
r
ds
:
u
s
i
ng
P
a
ge
C
oun
t
a
n
d
S
ni
p
pe
t
s
,
”
O
nl
i
n
e
I
n
t
.
C
on
f
.
o
n
G
r
e
e
n
E
ngi
ne
e
.
a
n
d
T
e
c
hno
l
og
i
e
s
(
I
C
-
G
E
T
)
,
pp
.
1
-
5
,
201
5.
[
14]
R
.
M
e
na
ha
,
A
.
U
.
S
u
r
ya
,
K
.
N
a
ndh
ni
,
a
n
d
M
.
I
s
hw
a
r
y
a
,
“
Q
ue
s
t
i
o
n
A
ns
w
e
r
i
ng
S
ys
t
e
m
us
i
n
g
W
e
b
S
n
i
pp
e
t
s
,
”
I
nt
.
C
on
f
.
on
I
-
S
M
A
C
(
I
oT
i
n
Soc
i
a
l
,
M
ob
i
l
e
,
A
n
al
y
t
i
c
s
a
nd
C
l
o
ud)
,
pp.
38
7
-
3
90
,
2
017
.
[
15]
B
.
S
e
t
i
yo
,
H
.
C
.
S
i
ho
m
b
i
n
g
,
a
nd
F
.
R
.
I
.
M
,
“
D
e
pr
e
s
s
i
o
n
a
nd
A
n
xi
e
t
y
D
e
t
e
c
t
i
on
T
hr
o
ugh
t
h
e
C
l
os
e
d
-
L
oop
M
e
t
h
od
us
i
n
g
D
A
S
S
-
2
1,
”
T
E
L
K
O
M
N
I
K
A
.
,
vo
l
.
1
7,
no
.
4,
p
p.
20
87
-
2097
,
2
018.
[
16]
A
.
D
.
P
o
e
r
no
m
o
a
nd
S
.
S
uh
a
r
j
i
t
o
,
“
I
ndo
.
O
n
l
i
n
e
T
r
a
ve
l
A
ge
nt
S
e
nt
i
m
e
n
t
A
na
l
y
s
i
s
us
i
ng
M
a
c
h
i
ne
L
e
a
r
ni
ng
M
e
t
h
od
s
,
”
I
nd
o.
J
.
E
l
e
c
t
r
.
E
ng.
C
om
p
ut
.
Sc
i
.
,
vo
l
.
14
,
no
.
1,
p
p
.
113
-
117
,
201
9.
[
17]
R
.
B
ha
l
l
a
a
nd
A
.
B
a
gg
a
,
“
O
pi
n
i
o
n
M
i
ni
ng
F
r
a
m
e
w
o
r
k
us
i
ng
P
r
o
pos
e
d
rb
-
ba
y
e
s
M
od
e
l
f
or
T
e
xt
C
l
a
s
s
i
f
i
c
a
t
i
on
,
”
I
nt
.
J
.
E
l
e
c
t
r
.
C
om
p
ut
.
E
ng
.
,
vo
l
.
9
,
no
.
1,
pp
.
477
-
48
5,
2
019
.
[
18]
M
.
A
.
A
h
m
e
d,
R
.
A
.
H
a
s
a
n
,
A
.
H
.
A
l
i
,
a
nd
M
.
A
.
M
oha
m
m
e
d,
“
U
s
i
ng
M
a
c
hi
ne
L
e
a
r
n
i
ng
f
or
t
he
C
l
a
s
s
i
f
i
c
a
t
i
o
n
of
t
he
M
ode
r
n
A
r
a
bi
c
P
oe
t
r
y
,
”
T
E
L
K
O
M
N
I
K
A
.,
vo
l
.
17
,
no
.
5,
20
19
.
[
19]
N
.
M
.
S
a
m
s
u
di
n
,
e
t
al
.
,
“
Y
o
ut
u
be
S
pa
m
D
e
t
e
c
t
i
on
F
r
a
m
e
w
o
r
k
us
i
n
g
N
a
ï
ve
B
a
y
e
s
a
nd
L
og
i
s
t
i
c
R
e
g
r
e
s
s
i
on
,
”
I
ndo
ne
s
.
J
.
E
l
e
c
t
r
.
E
ng
.
C
om
pu
t
.
S
c
i
.
,
v
ol
.
14
,
no.
3
,
pp
.
1508
-
15
17
,
201
9.
[
20]
Y
.
A
l
A
m
r
a
ni
,
M
.
L
a
z
a
a
r
,
a
nd
K
.
E
.
E
l
K
a
d
i
r
i
,
“
A
N
o
ve
l
H
yb
r
i
d
C
l
a
s
s
i
f
i
c
a
t
i
on
A
ppr
oa
c
h
f
or
S
e
n
t
i
m
e
n
t
A
na
l
y
s
i
s
o
f
T
e
xt
D
oc
u
m
e
n
t
,
”
I
n
t
.
J
.
E
l
e
c
t
r
.
C
om
put
.
E
ng
.
,
v
ol
.
8,
n
o.
6
,
pp.
4
55
4
–
456
7,
D
e
c
.
201
8.
[
21]
C
.
J
.
R
a
m
e
s
hbh
a
i
a
nd
J
.
P
a
u
l
os
e
,
“
O
p
i
n
i
on
m
i
n
i
ng
on
ne
w
s
pa
p
e
r
he
a
dl
i
ne
s
us
i
ng
S
V
M
a
ndN
L
P
,
”
I
nt
.
J
.
E
l
e
c
t
r
.
C
om
put
.
E
ng
.
,
vo
l
.
9,
no
.
3,
p
p.
21
52
–
2
163
,
J
un
.
201
9.
[
22]
A
.
R
i
z
a
l
dy
a
nd
H
.
A
.
S
a
n
t
o
s
o
,
“
P
e
r
f
o
r
m
a
n
c
e
I
m
p
r
ove
m
e
nt
o
f
S
u
ppor
t
V
e
c
t
or
M
a
c
hi
n
e
(
S
V
M
)
w
i
t
h
In
f
o
r
m
a
t
i
on
G
a
i
n
on
C
a
t
e
g
or
i
z
a
t
i
on
of
I
ndon
e
s
i
a
n
N
e
w
s
D
o
c
u
m
e
n
t
s
,
”
I
nt
.
Se
m
i
n
.
A
p
pl
.
T
e
c
hno
l
.
I
nf
.
C
om
m
un
.
,
vol
.
2018
,
pp
.
227
-
231
,
201
8.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
1693
-
6930
T
E
L
K
O
M
N
IK
A
T
e
l
e
c
om
m
un
Co
m
put
E
l
Con
t
rol
,
V
ol
.
18
,
N
o.
2
,
A
pri
l
2
020:
799
-
80
6
806
[
23]
A
.
M
ol
da
gu
l
o
va
a
nd
R
.
B
.
S
ul
a
i
m
a
n,
“
D
oc
um
e
n
t
C
l
a
s
s
i
f
i
c
a
t
i
o
n
B
a
s
e
d
on
K
N
N
A
l
g
or
i
t
h
m
by
T
e
r
m
V
e
c
t
or
S
pa
c
e
R
e
duc
t
i
on
,
”
I
n
t
.
C
o
nf
.
on
C
on
t
r
o
l
,
A
u
t
om
at
i
on
an
d
S
y
s
t
e
m
s
,
201
8.
[
24]
Y
.
T
a
n
,
“
A
n
I
m
p
r
ov
e
d
K
N
N
T
e
x
t
C
l
a
s
s
i
f
i
c
a
t
i
on
A
l
go
r
i
t
h
m
B
a
s
e
d
on
K
-
M
e
doi
ds
a
nd
R
o
ugh
S
e
t
,”
10t
h
I
nt
.
C
on
f
.
on
I
nt
e
l
l
i
ge
n
t
H
u
m
a
n
-
M
a
c
h
i
ne
S
y
s
t
e
m
s
a
n
d
C
y
be
r
n
e
t
i
c
s
,
pp
.
109
-
113
,
201
8.
[
25]
G
.
F
e
n
g,
J
.
G
u
o,
B
.
-
Y
.
J
i
ng
,
a
nd
T
.
S
un
,
“
F
e
a
t
ur
e
S
ub
s
e
t
S
e
l
e
c
t
i
on
us
i
ng
N
a
i
v
e
B
a
y
e
s
f
or
T
e
x
t
C
l
a
s
s
i
f
i
c
a
t
i
on
,
”
P
at
t
e
r
n
R
e
c
og
ni
t
.
L
e
t
t
.
,
vo
l
.
6
5,
pp
.
109
-
11
5,
20
15.
[
26]
Z
.
E
.
R
a
s
j
i
d
a
nd
R
.
S
e
t
i
a
w
a
n,
“
P
e
r
f
o
r
m
a
n
c
e
C
o
m
pa
r
i
s
on
a
nd
O
pt
i
m
i
z
a
t
i
on
of
T
e
xt
D
oc
um
e
n
t
C
l
a
s
s
i
f
i
c
a
t
i
on
us
i
ng
k
-
N
N
a
nd
N
a
ï
ve
B
a
y
e
s
C
l
a
s
s
i
f
i
c
a
t
i
o
n
T
e
c
h
ni
q
ue
s
,
”
P
r
oc
e
di
a
C
om
p
ut
.
S
c
i
.
,
v
ol
.
116
,
pp
.
107
-
112
,
201
7.
[
27]
Y
.
L
u
a
nd
Y
.
P
e
ng
,
“
F
e
a
t
ur
e
W
e
i
g
ht
i
ng
I
m
pr
ove
m
e
nt
o
f
W
e
b
T
e
xt
C
a
t
e
gor
i
z
a
t
i
on
B
a
s
e
d
on
P
a
r
t
i
c
l
e
S
w
a
r
m
O
pt
i
m
i
z
a
t
i
on
A
l
gor
i
t
h
m
,
”
J
.
C
om
pu
t
.
,
vo
l
.
10,
n
o.
4
,
pp.
2
60
-
267
,
2
015.
[
28]
O
.
S
o
m
a
n
t
r
i
a
nd
D
.
A
pr
i
l
i
a
ni
,
“
O
p
i
n
i
on
M
i
ni
ng
on
C
ul
i
n
a
r
y
F
ood
C
us
t
o
m
e
r
S
a
t
i
s
f
a
c
t
i
on
u
s
i
ng
N
a
ï
ve
B
a
ye
s
B
a
s
e
d
-
on
H
yb
r
i
d
F
e
a
t
u
r
e
S
e
l
e
c
t
i
on
,
”
I
n
don
e
s
.
J
.
E
l
e
c
t
r
.
E
ng
.
C
om
p
ut
.
S
c
i
.
,
v
ol
.
15
,
no.
1
,
pp
.
468
-
475
,
201
9.
[
29]
M
.
J
.
B
a
s
h
a
a
nd
K
.
P
.
K
a
l
i
y
a
m
ur
t
h
i
e
,
“
A
n
I
m
pr
ove
d
S
i
m
i
l
a
r
i
t
y
M
a
t
c
h
i
ng
ba
s
e
d
C
l
us
t
e
r
i
n
g
F
r
a
m
e
w
o
r
k
f
or
S
h
or
t
a
nd
S
e
n
t
e
n
c
e
L
e
v
e
l
T
e
xt
,
”
I
nt
.
J
.
E
l
e
c
t
r
.
C
om
pu
t
.
E
ng
.
,
vo
l
.
7
,
no
.
1
,
pp.
5
51
-
558
,
2017
.
[
30]
Y
.
L
u
,
M
.
L
i
a
ng,
Z
.
Y
e
,
a
nd
L
.
C
a
o,
“
I
m
pr
o
ve
d
P
a
r
t
i
c
l
e
S
w
a
r
m
O
p
t
i
m
i
z
a
t
i
on
A
l
gor
i
t
h
m
a
nd
i
t
s
A
ppl
i
c
a
t
i
on
i
n
T
e
x
t
F
e
a
t
ur
e
S
e
l
e
c
t
i
o
n
,
”
A
ppl
.
So
f
t
C
om
p
ut
.
,
v
ol
.
35
,
pp.
6
29
-
636
,
2015
.
[
31]
R
.
J
a
n
a
n
i
a
nd
S
.
V
i
j
a
y
a
r
a
ni
,
“
T
e
xt
d
oc
u
m
e
nt
c
l
u
s
t
e
r
i
ng
u
s
i
ng
S
pe
c
t
r
a
l
C
l
us
t
e
r
i
ng
a
l
go
r
i
t
h
m
w
i
t
h
P
a
r
t
i
c
l
e
S
w
a
r
m
O
pt
i
m
i
z
a
t
i
on
,
”
E
x
p
e
r
t
Sy
s
t
.
A
ppl
.
,
vol
.
134
,
pp
.
192
–
200
,
N
ov
.
2
01
9.
[
32]
P
.
G
oku
l
,
B
.
K
.
A
kh
i
l
,
a
n
d
K
.
K
.
M
.
S
hi
v
a
,
“
S
e
n
t
e
n
c
e
S
i
m
i
l
a
r
i
t
y
D
e
t
e
c
t
i
on
i
n
M
a
l
a
ya
l
a
m
L
a
n
gua
ge
us
i
ng
c
os
i
ne
s
i
m
i
l
a
r
i
t
y
,
”
2n
d
I
E
E
E
I
nt
.
C
onf
.
on
R
e
c
e
n
t
T
r
e
n
ds
i
n
E
l
e
c
t
r
o
,
I
nf
or
m
&
C
o
m
m
u
T
e
c
hno
l
og
y
,
pp
.
221
-
22
5
,
20
17
.
[
33]
A
.
N
u
r
s
e
t
yo
,
D
.
R
.
I
.
M
.
S
e
t
i
a
di
,
a
nd
E
.
R
.
S
ubh
i
ya
kt
o
,
“
S
m
a
r
t
C
h
a
t
bo
t
S
ys
t
e
m
f
o
r
E
-
C
o
m
m
e
r
c
e
A
s
s
i
t
a
nc
e
ba
s
e
d
on
A
I
M
L
,
”
i
n
I
n
t
e
r
na
t
i
o
na
l
S
e
m
i
nar
on
R
e
s
e
a
r
c
h
o
f
I
n
f
o
r
m
at
i
on
T
e
c
hno
l
og
y
a
nd
I
n
t
e
l
l
i
g
e
nt
S
y
s
t
e
m
s
,
201
8.
[
34]
H
.
A
.
S
a
n
t
o
s
o,
J
.
Z
e
ni
a
r
j
a
,
A
.
L
u
t
h
f
i
a
r
t
a
,
a
nd
B
.
J
.
W
i
j
a
ya
,
“
A
n
O
nt
o
l
og
i
c
a
l
C
r
a
w
l
i
ng
A
pp
r
o
a
c
h
f
or
I
m
pr
o
vi
n
g
I
nf
or
m
a
t
i
on
A
ggr
e
ga
t
i
on
ov
e
r
e
-
G
ove
r
n
m
e
nt
We
bs
i
t
e
s
,
”
J
.
C
om
pu
t
.
Sc
i
.
,
201
6.
[
35]
H.
-
A
.
S
a
n
t
o
s
o
,
S
.
-
C
.
H
a
w
,
a
n
d
C
.
-
S
.
L
e
e
,
“
S
o
f
t
w
a
r
e
R
e
us
e
:
M
D
A
-
B
a
s
e
d
O
n
t
o
l
ogy
D
e
ve
l
op
m
e
nt
t
o
S
upp
or
t
D
a
t
a
A
c
c
e
s
s
ove
r
L
e
g
a
c
y
A
pp
l
i
c
a
t
i
o
ns
,
”
Sp
r
i
nge
r
,
B
e
r
l
i
n
,
H
e
i
d
e
l
be
r
g,
p
p.
130
-
14
2
,
20
11.
[
36]
H
.
A
.
S
a
nt
o
s
o
,
Z
.
T
.
A
b
du
l
-
M
e
h
di
,
a
nd
S
.
-
C
.
H
a
w
,
“
S
e
m
a
n
t
i
c
E
n
ha
nc
e
m
e
n
t
F
r
a
m
e
w
o
r
k
f
o
r
e
-
G
ov
e
r
nm
e
n
t
u
s
i
n
g
O
nt
o
l
ogy
V
e
r
s
i
on
i
ng
A
pp
r
o
a
c
h
,
”
i
n
I
n
t
.
C
on
f
.
o
n
I
n
f
or
m
a
t
i
on
T
e
c
h
nol
o
gy
a
nd
A
pp
l
i
c
a
t
i
ons
,
pp
.
296
-
301
,
200
9.
Evaluation Warning : The document was created with Spire.PDF for Python.