T
E
L
K
O
M
N
I
K
A
T
elec
o
m
m
un
ica
t
io
n Co
m
pu
t
ing
E
lect
ro
nics
a
nd
Co
ntr
o
l
Vo
l.
20
,
No
.
1
,
Feb
r
u
ar
y
20
22
,
p
p
.
89
~
97
I
SS
N:
1
6
9
3
-
6
9
3
0
,
DOI
: 1
0
.
1
2
9
2
8
/TE
L
KOM
NI
KA.
v
20
i
1
.
1
8
6
3
0
89
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//telko
mn
ika
.
u
a
d
.
a
c.
i
d
Na
tural la
ng
ua
g
e process
ing
a
nd m
a
chine learning
b
a
sed
cy
berbully
ing
det
ection for
Ba
ng
la
a
nd
Ro
ma
nized B
a
ng
la
tex
ts
M
d.
T
o
f
a
el
Ahm
ed
1
,
M
a
q
s
u
du
r
Ra
hm
a
n
1
,
Sh
a
f
a
y
et
Nur
2
,
Abu Z
a
f
o
r
M
uh
a
m
m
a
d T
o
uh
idu
l
I
s
la
m
3
,
Dipa
nk
a
r
Da
s
4
1
D
e
p
a
r
t
me
n
t
o
f
I
n
f
o
r
mat
i
o
n
a
n
d
C
o
m
mu
n
i
c
a
t
i
o
n
Te
c
h
n
o
l
o
g
y
,
C
o
mi
l
l
a
U
n
i
v
e
r
si
t
y
,
C
o
mi
l
l
a
,
B
a
n
g
l
a
d
e
s
h
2
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
&
En
g
i
n
e
e
r
i
n
g
,
P
o
r
t
C
i
t
y
I
n
t
e
r
n
a
t
i
o
n
a
l
U
n
i
v
e
r
si
t
y
,
C
h
a
t
t
o
g
r
a
m
,
B
a
n
g
l
a
d
e
s
h
3
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
i
c
a
l
a
n
d
El
e
c
t
r
o
n
i
c
s E
n
g
i
n
e
e
r
i
n
g
,
U
n
i
v
e
r
si
t
y
o
f
R
a
j
s
h
a
h
i
,
R
a
j
sh
a
h
i
,
B
a
n
g
l
a
d
e
sh
4
D
e
p
a
r
t
me
n
t
o
f
I
n
f
o
r
mat
i
o
n
a
n
d
C
o
m
mu
n
i
c
a
t
i
o
n
En
g
i
n
e
e
r
i
n
g
,
U
n
i
v
e
r
s
i
t
y
o
f
R
a
j
sh
a
h
i
,
R
a
j
s
h
a
h
i
,
B
a
n
g
l
a
d
e
s
h
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
No
v
18
,
2
0
2
0
R
ev
is
ed
Dec
08
,
2
0
2
1
Acc
ep
ted
Dec
18
,
2
0
2
1
Th
e
p
o
p
u
lari
ty
o
f
s
o
c
ial
m
e
d
ia
h
a
s
b
e
e
n
in
c
re
a
sin
g
trem
e
n
d
o
u
sly
in
re
c
e
n
t
ti
m
e
s
a
n
d
t
h
u
s
c
y
b
e
r
b
u
ll
y
i
n
g
to
wa
rd
s
p
e
o
p
le
h
a
s
a
lso
in
c
re
a
se
d
a
t
a
n
a
larm
in
g
ra
te.
M
a
n
y
c
y
b
e
rb
u
ll
y
in
g
tex
ts
c
a
n
b
e
f
o
u
n
d
i
n
t
h
e
c
o
m
m
e
n
t
se
c
ti
o
n
s
o
f
m
a
n
y
we
ll
-
k
n
o
w
n
Ba
n
g
lad
e
sh
i
s
o
c
ia
l
m
e
d
ia
p
e
rso
n
a
li
ti
e
s
Yo
u
T
u
b
e
v
id
e
o
s.
It
h
a
s
th
e
p
o
te
n
ti
a
l
to
c
a
u
se
se
v
e
re
e
m
o
ti
o
n
a
l
a
n
d
p
sy
c
h
o
l
o
g
ica
l
d
istres
s.
Th
e
re
fo
re
,
tex
ts
c
o
n
tain
i
n
g
c
y
b
e
r
b
u
ll
y
in
g
s
h
o
u
ld
b
e
d
e
tec
ted
a
t
t
h
e
e
a
rli
e
st
sta
g
e
a
n
d
p
re
v
e
n
ted
fr
o
m
b
e
i
n
g
d
isp
lay
e
d
.
In
t
h
is
st
u
d
y
,
we
u
se
n
a
t
u
ra
l
lan
g
u
a
g
e
p
ro
c
e
ss
in
g
(NLP
)
tec
h
n
iq
u
e
s
a
n
d
v
a
ri
o
u
s
m
a
c
h
in
e
lea
rn
in
g
c
las
sifiers
a
n
d
p
re
se
n
ted
m
o
d
e
l
f
o
r
c
y
b
e
r
b
u
ll
y
i
n
g
d
e
tec
ti
o
n
in
Ba
n
g
la
a
n
d
R
o
m
a
n
iz
e
d
Ba
n
g
la
tex
ts
o
b
tai
n
e
d
fro
m
Yo
u
Tu
b
e
v
id
e
o
c
o
m
m
e
n
ts.
We
d
e
v
e
l
o
p
e
d
o
u
r
o
w
n
d
a
tas
e
ts
u
sin
g
Yo
u
Tu
b
e
a
p
p
li
c
a
ti
o
n
p
ro
g
ra
m
m
in
g
in
terfa
c
e
(API)
v
e
rsio
n
3
.
0
.
We
c
o
ll
e
c
ted
5
0
0
0
Ba
n
g
la
c
o
m
m
e
n
ts,
a
s
we
ll
a
s
7
0
0
0
R
o
m
a
n
ize
d
Ba
n
g
la
c
o
m
m
e
n
ts
fro
m
v
id
e
o
s
o
f
d
iffere
n
t
we
ll
-
k
n
o
wn
so
c
ial
m
e
d
ia
p
e
rs
o
n
a
ls.
Th
e
se
two
d
a
tas
e
ts,
a
s
we
ll
a
s
a
th
ir
d
d
a
tas
e
t
o
f
1
2
0
0
0
te
x
ts
wh
ic
h
wa
s
th
e
c
o
m
b
in
a
ti
o
n
o
f
t
h
e
first
two
d
a
tas
e
ts
we
re
u
se
d
to
train
th
e
c
las
sifi
e
rs.
Th
e
se
d
a
tas
e
ts
we
re
u
se
d
to
train
m
a
c
h
in
e
lea
rn
in
g
c
las
sifiers
a
fter
b
e
in
g
p
re
p
ro
c
e
ss
e
d
u
sin
g
NLP
tec
h
n
iq
u
e
s.
Wi
th
a
n
a
c
c
u
ra
c
y
sc
o
re
o
f
7
6
%
,
su
p
p
o
r
t
v
e
c
to
r
m
a
c
h
in
e
(S
VM)
o
u
tp
e
rf
o
r
m
e
d
th
e
o
th
e
r
c
las
sifiers
fo
r
th
e
fi
rst
d
a
tas
e
t.
Th
e
h
i
g
h
e
st
a
c
c
u
ra
c
y
sc
o
re
s
fo
r
t
h
e
se
c
o
n
d
a
n
d
t
h
ird
d
a
tas
e
ts
we
re
8
4
%
a
n
d
8
0
%
,
re
sp
e
c
ti
v
e
l
y
,
wh
ich
we
re
b
o
th
a
c
h
iev
e
d
b
y
m
u
lt
in
o
m
ial
n
a
iv
e
Ba
y
e
s.
K
ey
w
o
r
d
s
:
C
y
b
er
b
u
lly
in
g
Ma
ch
in
e
lear
n
in
g
Natu
r
al
lan
g
u
ag
e
p
r
o
ce
s
s
in
g
Yo
u
T
u
b
e
c
o
m
m
en
ts
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Ma
q
s
u
d
u
r
R
ah
m
an
Dep
ar
tm
en
t o
f
I
n
f
o
r
m
atio
n
an
d
C
o
m
m
u
n
icatio
n
T
ec
h
n
o
lo
g
y
Co
m
il
la Un
i
v
e
rsity
C
o
m
illa,
B
an
g
lad
esh
E
m
ail:
m
r
r
ajo
n
@
co
u
.
ac
.
b
d
1.
I
NT
RO
D
UCT
I
O
N
On
e
o
f
th
e
m
o
s
t
co
m
m
o
n
d
i
g
ital
h
o
b
b
ies
th
ese
d
ay
s
is
s
p
en
d
in
g
tim
e
o
n
s
o
cial
m
ed
ia
s
ites
s
u
ch
a
s
Face
b
o
o
k
,
I
n
s
tag
r
am
,
T
witter
,
an
d
Yo
u
T
u
b
e
[
1
]
.
Alm
o
s
t
3
.
6
b
illi
o
n
p
e
o
p
le
ar
e
u
s
in
g
s
o
cial
m
ed
ia
in
2
0
2
0
wh
ich
h
as
an
escalatio
n
r
ate
o
f
4
9
%
as
o
f
J
an
u
ar
y
2
0
2
0
.
On
a
d
aily
b
asis
,
p
eo
p
le
s
p
en
d
an
av
e
r
ag
e
o
f
1
4
4
m
in
u
tes
o
n
s
o
cial
m
ed
ia.
T
h
is
h
u
g
e
in
cr
ea
s
e
in
t
h
e
u
s
e
o
f
s
o
cial
m
ed
ia
h
as
its
h
u
g
e
a
d
v
an
tag
es
as
well
as
d
is
ad
v
an
tag
es.
On
e
o
f
th
e
m
o
s
t ser
io
u
s
d
r
awb
ac
k
s
is
th
e
r
is
e
o
f
cy
b
e
r
b
u
lly
i
n
g
o
n
v
ar
i
o
u
s
s
o
cial
m
ed
ia
s
ites
.
C
y
b
er
b
u
lly
in
g
is
d
escr
ib
ed
a
s
th
e
in
ten
tio
n
al
an
d
r
ep
ea
te
d
in
f
lictio
n
o
f
h
a
r
m
th
r
o
u
g
h
elec
tr
o
n
ic
m
ed
ia
[
2
]
.
Ov
e
r
8
0
%
o
f
c
h
ild
r
en
o
wn
s
a
m
o
b
ile
p
h
o
n
e
an
d
u
s
es
s
o
cial
n
etwo
r
k
s
ites
o
f
w
h
ich
5
7
%
ad
m
itted
th
e
ex
p
er
ien
ce
o
f
cy
b
er
b
u
lly
i
n
g
an
d
also
6
0
%
ch
ild
r
en
an
d
y
o
u
n
g
p
e
o
p
le
h
av
e
witn
ess
ed
b
u
lly
in
g
o
n
s
o
cial
m
ed
ia.
T
h
is
h
o
r
r
ib
le
e
x
p
er
ien
c
e
u
n
d
er
m
in
es
a
p
er
s
o
n
’
s
f
r
ee
d
o
m
to
u
s
e
o
n
lin
e
r
eso
u
r
ce
s
an
d
also
ca
u
s
es
s
ev
er
al
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
TEL
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
20
,
No
.
1
,
Feb
r
u
ar
y
20
22
:
89
-
97
90
p
s
y
ch
o
lo
g
ical
ef
f
ec
ts
[
3
]
.
C
y
b
er
b
u
lly
in
g
v
ictim
s
ar
e
1
.
9
tim
es
m
o
r
e
lik
ely
in
co
m
m
itt
in
g
s
u
icid
e
an
d
f
u
r
th
er
m
o
r
e
en
d
u
r
es c
er
eb
r
al
p
r
o
b
lem
s
lik
e
au
tis
m
7
5
%,
s
o
m
atic
f
au
lts
7
0
% a
n
d
lear
n
in
g
co
m
p
licatio
n
s
5
2
%.
T
h
is
in
cr
ea
s
e
in
cy
b
er
b
u
lly
in
g
h
as
also
escalate
d
th
e
n
ec
ess
ity
o
f
p
r
ev
en
tio
n
o
f
c
y
b
er
b
u
lly
in
g
.
T
h
e
d
etec
tio
n
o
f
cy
b
er
b
u
lly
in
g
ca
n
b
e
a
s
ig
n
if
ican
t
m
a
n
eu
v
e
r
in
p
r
e
v
en
tin
g
cy
b
er
b
u
lly
in
g
.
I
f
th
e
tex
ts
th
at
co
n
tain
s
cy
b
er
b
u
lly
in
g
ca
n
b
e
d
etec
ted
at
th
e
ea
r
lies
t stag
e,
th
ey
ca
n
b
e
p
r
ev
en
te
d
f
r
o
m
b
ein
g
co
m
m
en
ted
.
Ma
ch
in
e
lear
n
in
g
b
ased
class
if
icatio
n
m
o
d
els
ca
n
b
e
o
f
g
r
ea
t
p
ly
in
d
etec
tin
g
cy
b
e
r
b
u
lly
in
g
.
Ov
er
th
e
y
ea
r
s
m
ac
h
in
e
lea
r
n
in
g
m
o
d
els
h
av
e
p
r
o
v
ed
th
eir
ef
f
icien
c
y
in
p
r
ed
ictio
n
an
d
d
etec
tio
n
.
T
h
er
e
is
a
h
u
g
e
am
o
u
n
t
o
f
r
esear
ch
a
v
ailab
le
wh
ich
u
tili
ze
s
m
ac
h
in
e
lear
n
in
g
b
ased
p
r
ed
ictio
n
an
d
d
etec
tio
n
.
Bh
a
ra
t
et
a
l.
[
4
]
u
s
ed
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
f
o
r
p
r
ed
ic
tio
n
o
f
b
r
ea
s
t
ca
n
ce
r
an
d
also
d
iag
n
o
s
ed
b
r
ea
s
t
ca
n
ce
r
u
s
in
g
m
ac
h
in
e
le
ar
n
in
g
alg
o
r
ith
m
s
an
d
Kau
r
an
d
Ku
m
ar
i
[
5
]
class
if
ied
d
iab
etic
an
d
n
o
n
-
d
ia
b
etic
p
atien
ts
u
s
in
g
m
ac
h
in
e
lear
n
in
g
ap
p
r
o
ac
h
.
Far
h
an
a
et
a
l.
[
6
]
u
tili
ze
d
d
ee
p
lear
n
i
n
g
ap
p
r
o
ac
h
to
d
etec
t
in
tr
u
s
io
n
f
o
r
p
ac
k
et
an
d
f
l
o
w
-
b
ased
n
et
wo
r
k
s
an
d
in
[
7
]
,
ag
ai
n
p
r
esen
ted
m
ac
h
in
e
lear
n
in
g
m
o
d
els
f
o
r
au
to
m
ated
tr
af
f
ic
class
if
icatio
n
an
d
a
p
p
licatio
n
id
en
tific
atio
n
.
Als
o
,
Ho
s
s
ain
et
a
l.
[
8
]
u
s
ed
m
ac
h
in
e
lear
n
in
g
al
g
o
r
ith
m
s
to
p
r
ed
ict
r
atin
g
o
f
p
r
o
d
u
ct
r
ev
ie
ws
an
d
in
[
9
]
,
t
h
e
au
th
o
r
s
p
r
o
p
o
s
ed
a
m
et
h
o
d
o
f
tr
ac
k
in
g
an
d
d
etec
tin
g
v
eh
icles
f
r
o
m
r
ea
l tim
e
v
id
eo
s
tr
ea
m
i
n
g
u
s
in
g
b
l
o
b
tr
ac
k
er
alg
o
r
ith
m
.
T
h
er
e
ar
e
also
m
an
y
r
esear
ch
es
o
n
tex
t
-
b
ased
m
ac
h
in
e
lear
n
in
g
class
i
f
icatio
n
m
eth
o
d
s
lik
e
I
k
o
n
o
m
ak
is
et
a
l.
[
1
0
]
u
s
ed
m
ac
h
in
e
lear
n
in
g
tech
n
i
q
u
es
to
co
n
d
u
ct
te
x
t
class
if
icatio
n
.
B
o
iy
an
d
Mo
e
n
s
[
1
1
]
u
s
ed
m
ac
h
in
e
lea
r
n
in
g
to
ev
al
u
ate
s
en
tim
en
t
in
E
n
g
lis
h
,
Du
tch
,
an
d
Fre
n
ch
te
x
ts
.
T
h
ese
k
in
d
s
o
f
tex
t
-
b
ased
class
if
icatio
n
m
o
d
e
ls
ca
n
b
e
a
g
r
ea
t
u
s
e
in
class
if
y
in
g
cy
b
er
b
u
lly
in
g
tex
ts
f
o
r
m
r
e
g
u
lar
te
x
ts
.
Similar
k
in
d
o
f
wo
r
k
was
p
r
esen
te
d
b
y
Haid
ar
et
a
l.
[
1
2
]
wh
er
e
th
e
d
etec
ted
c
y
b
er
b
u
lly
in
g
f
r
o
m
Ar
ab
ic
an
d
E
n
g
lis
h
tex
ts
u
s
in
g
m
ac
h
in
e
lear
n
in
g
m
o
d
els
an
d
in
[
1
3
]
,
cy
b
e
r
b
u
lly
in
g
f
r
o
m
twitter
o
f
Sp
an
is
h
lan
g
u
ag
e
was
d
etec
ted
u
s
in
g
m
ac
h
in
e
lear
n
in
g
ap
p
r
o
ac
h
.
S
im
ilar
ly
,
Gr
ee
v
y
a
n
d
Sm
ea
to
n
[
1
4
]
d
ev
elo
p
ed
a
s
y
s
tem
to
d
etec
t
r
ac
is
m
u
s
in
g
m
ac
h
in
e
lear
n
in
g
tech
n
iq
u
es.
T
h
er
e
is
also
r
esear
ch
o
n
b
u
ll
y
in
g
d
etec
tio
n
o
n
B
an
g
la
tex
t
s
wh
er
e
Al
-
Mam
u
n
an
d
Ak
h
ter
[
1
5
]
p
r
o
p
o
s
ed
m
ac
h
in
e
lear
n
in
g
b
ased
a
p
p
r
o
ac
h
.
T
h
e
r
em
ain
in
g
p
a
p
er
is
laid
o
u
t
as
f
o
llo
ws
:
s
ec
tio
n
2
in
clu
d
es
s
ev
er
al
wo
r
k
s
th
at
ar
e
r
ele
v
an
t
to
o
u
r
s
tu
d
y
.
T
h
e
m
eth
o
d
o
l
o
g
y
is
p
r
e
s
en
ted
in
s
ec
tio
n
3
.
T
h
e
f
in
d
i
n
g
s
ar
e
d
is
cu
s
s
ed
in
s
ec
tio
n
4
,
an
d
th
e
co
n
clu
s
io
n
an
d
f
u
t
u
r
e
wo
r
k
ca
n
b
e
f
o
u
n
d
in
s
ec
tio
n
5
.
2.
R
E
L
AT
E
D
WO
RK
S
As
m
an
y
r
esear
ch
er
s
ar
e
wo
r
k
in
g
h
ar
d
to
d
etec
t
c
y
b
er
b
u
lly
in
g
in
s
ev
er
al
la
n
g
u
a
g
es,
th
er
e
ar
e
s
o
m
e
p
r
ev
io
u
s
r
esear
ch
es
av
ailab
le
t
h
is
f
ield
.
I
n
th
is
s
ec
tio
n
,
we
will
d
is
cu
s
s
ab
o
u
t
s
o
m
e
o
f
th
e
wo
r
k
s
th
at
ar
e
r
elev
an
t
to
o
u
r
s
tu
d
ies.
Haid
ar
et
a
l.
[
1
2
]
p
r
o
p
o
s
ed
a
s
o
lu
tio
n
f
o
r
d
etec
tin
g
cy
b
er
b
u
lly
in
g
u
s
in
g
m
ac
h
in
e
lear
n
in
g
.
I
n
th
eir
r
esear
ch
,
th
e
y
u
s
ed
b
o
th
E
n
g
lis
h
an
d
R
o
m
an
ize
d
tex
ts
a
n
d
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM)
h
ad
th
e
h
ig
h
es
t
ove
r
all
p
r
ec
is
io
n
(
9
3
.
4
%).
Us
in
g
SVM
an
d
n
aiv
e
B
ay
es
class
if
ier
s
,
Dalv
i
et
a
l.
[
1
6
]
p
r
o
p
o
s
ed
a
m
ac
h
in
e
lear
n
in
g
m
o
d
el
to
id
en
tify
an
d
elim
in
ate
cy
b
er
b
u
lly
in
g
.
T
h
e
d
ata
was
o
b
tain
ed
f
r
o
m
T
witter
th
r
o
u
g
h
th
e
T
witter
ap
p
licatio
n
p
r
o
g
r
am
m
in
g
in
te
r
f
ac
e
(
API
)
.
SVM
h
ad
a
h
ig
h
e
r
ac
cu
r
ac
y
o
f
7
1
.
2
5
%
in
t
h
eir
an
aly
s
is
th
an
n
ai
v
e
B
ay
es,
wh
ich
h
ad
a
5
2
.
7
0
% a
cc
u
r
ac
y
.
T
h
er
e
ar
e
also
s
ev
er
al
o
t
h
er
r
e
s
ea
r
ch
f
o
r
c
y
b
er
b
u
lly
in
g
d
etec
tio
n
,
s
u
ch
as
Par
ed
es
et
a
l.
[
1
3
]
r
etr
iev
ed
Sp
an
is
h
tex
ts
f
r
o
m
T
witter
an
d
ac
h
iev
ed
a
9
3
%
ac
cu
r
ac
y
r
ate
u
s
in
g
m
ac
h
in
e
l
ea
r
n
in
g
alg
o
r
ith
m
s
.
B
an
er
jee
et
a
l.
[
1
7
]
in
tr
o
d
u
ce
d
a
n
o
v
el
d
ee
p
n
e
u
r
al
n
etwo
r
k
a
p
p
r
o
ac
h
f
o
r
cy
b
er
b
u
lly
in
g
d
ete
ctio
n
,
an
d
th
e
C
NN
m
eth
o
d
r
ec
eiv
ed
a
m
ax
im
u
m
o
f
9
3
.
9
7
%
test
in
g
ac
cu
r
ac
y
.
Ali
an
d
Sy
ed
[
1
8
]
also
u
s
in
g
m
ac
h
i
n
e
lear
n
in
g
tech
n
iq
u
es.
I
n
th
eir
r
esear
c
h
,
th
ey
u
s
ed
th
r
ee
d
atasets
an
d
SVM
h
ad
th
e
h
ig
h
est av
er
ag
e
a
cc
u
r
ac
y
o
f
8
0
%.
W
e
wer
e
in
s
p
ir
ed
b
y
th
ese
ex
ce
lle
n
t e
f
f
o
r
ts
o
f
cy
b
e
r
b
u
lly
i
n
g
d
et
ec
tio
n
in
B
an
g
la
an
d
R
o
m
an
iz
ed
B
an
g
la
tex
ts
.
Ma
ch
in
e
lear
n
in
g
is
also
u
tili
ze
d
in
B
an
g
la
C
y
b
er
b
u
ll
y
in
g
d
etec
tio
n
d
o
m
ain
.
Ma
m
u
n
an
d
Ak
h
ter
[
1
5
]
s
u
g
g
ested
u
s
in
g
m
ac
h
in
e
lear
n
in
g
t
o
d
etec
t
c
y
b
er
b
u
lly
in
g
in
B
an
g
la
tex
t.
T
h
ey
co
llected
2
4
0
0
s
tatu
s
f
r
o
m
Face
b
o
o
k
a
n
d
T
witter
an
d
ap
p
lied
m
ac
h
in
e
l
ea
r
n
in
g
alg
o
r
ith
m
s
in
two
p
h
ases
.
T
h
eir
h
ig
h
est
ac
cu
r
ac
y
was 9
7
.
2
7
% a
cc
u
r
ac
y
an
d
it wa
s
g
ain
e
d
b
y
SVM.
C
h
ak
r
ab
o
r
ty
a
n
d
Sed
d
iq
u
i
[
1
9
]
u
s
ed
m
ac
h
in
e
an
d
d
ee
p
l
ea
r
n
in
g
to
class
if
y
B
an
g
la
tex
ts
,
with
SVM
p
er
f
o
r
m
in
g
b
est
with
7
8
%
ac
cu
r
a
cy
.
Similar
ly
,
a
m
ax
im
u
m
o
f
7
2
%
ac
cu
r
ac
y
was
ac
h
iev
e
d
b
y
Ah
am
m
ed
et
a
l.
[
2
0
]
.
T
h
ey
g
ath
er
ed
th
eir
B
en
g
ali
d
ata
f
r
o
m
Face
b
o
o
k
.
T
h
ese
wo
r
k
s
f
o
r
th
e
d
etec
tio
n
o
f
cy
b
er
b
u
lly
in
g
in
B
an
g
la
m
o
tiv
ated
u
s
to
wo
r
k
with
B
an
g
la
d
ata
co
llected
f
r
o
m
Yo
u
T
u
b
e.
Als
o
,
th
er
e
is
a
v
er
y
f
ew
wo
r
k
s
av
ailab
le
w
h
ich
u
s
ed
R
o
m
an
ized
B
an
g
la
tex
ts
li
k
e
T
r
ip
t
o
a
n
d
Ali
[
2
1
]
.
T
h
eir
r
esear
ch
class
if
ied
s
en
tim
en
t
o
f
B
an
g
la,
R
o
m
an
ized
B
an
g
la
an
d
E
n
g
lis
h
tex
ts
co
llected
f
r
o
m
Yo
u
T
u
b
e
[
2
1
]
.
T
h
ey
u
s
ed
lo
n
g
s
h
o
r
t
-
ter
m
m
em
o
r
y
(
L
STM
)
,
c
o
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
(
C
NN)
,
n
aiv
e
B
ay
es
an
d
SVM
an
d
s
h
o
wed
an
ac
c
u
r
ac
y
o
f
6
5
%
f
o
r
L
STM
.
Similar
ly
,
Hass
an
et
a
l.
[
2
2
]
u
s
ed
B
an
g
l
a
an
d
R
o
m
an
ized
B
an
g
la
tex
ts
.
Usi
n
g
th
ese
tex
ts
,
th
ey
tr
ain
e
d
a
d
ee
p
r
ec
u
r
r
en
t
m
o
d
el
wh
ic
h
g
a
v
e
t
h
em
a
h
ig
h
est
o
f
7
8
%
ac
cu
r
a
cy
.
B
ec
au
s
e
th
e
n
u
m
b
er
o
f
wo
r
k
s
f
o
r
R
o
m
a
n
ized
B
an
g
la
tex
ts
i
s
m
in
im
al,
we
d
ec
id
e
d
to
co
n
d
u
ct
o
u
r
r
esear
ch
u
s
in
g
R
o
m
a
n
ized
B
an
g
la
tex
ts
co
llected
f
r
o
m
Yo
u
T
u
b
e.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
N
a
tu
r
a
l la
n
g
u
a
g
e
p
r
o
ce
s
s
in
g
a
n
d
ma
c
h
in
e
lea
r
n
in
g
b
a
s
ed
c
yb
erb
u
llyin
g
d
etec
tio
n
fo
r
… (
Md
.
To
fa
el
A
h
med
)
91
3.
M
E
T
H
O
DO
L
O
G
Y
3
.
1
.
Wo
r
k
f
l
o
w
Usi
n
g
n
atu
r
al
lan
g
u
a
g
e
p
r
o
ce
s
s
in
g
tech
n
iq
u
es
a
n
d
m
ac
h
in
e
lear
n
in
g
class
if
ier
s
,
we
aim
t
o
id
en
tif
y
cy
b
er
b
u
lly
in
g
tex
ts
o
b
tain
e
d
f
r
o
m
Yo
u
T
u
b
e
v
id
eo
co
m
m
en
t
s
ec
tio
n
s
.
T
h
r
o
u
g
h
o
u
t
t
h
is
r
esear
ch
,
a
to
tal
o
f
th
r
ee
d
atasets
wer
e
u
s
ed
.
T
h
e
d
atas
ets
wer
e
p
r
ep
r
o
ce
s
s
ed
u
s
in
g
n
atu
r
al
lan
g
u
ag
e
p
r
o
ce
s
s
in
g
(
N
L
P)
tech
n
iq
u
es
an
d
th
en
wer
e
u
s
ed
to
tr
ai
n
th
e
m
ac
h
in
e
lear
n
in
g
class
if
ier
s
.
Fin
ally
,
th
e
p
er
f
o
r
m
an
ce
an
aly
s
i
s
was
p
er
f
o
r
m
e
d
in
ter
m
s
o
f
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
f
1
-
s
co
r
e
an
d
ar
ea
u
n
d
er
th
e
cu
r
v
e
o
f
r
ec
eiv
er
ch
a
r
ac
ter
is
tic
o
p
er
ato
r
(
AUC
-
R
O
C
)
cu
r
v
e.
Fig
u
r
e
1
d
ep
icts
th
e
p
r
o
p
o
s
ed
m
eth
o
d
o
lo
g
y
.
Fig
u
r
e
1
.
Pro
p
o
s
ed
m
et
h
o
d
o
lo
g
y
3
.
2
.
Da
t
a
s
et
T
h
e
m
o
s
t
im
p
o
r
tan
t
p
h
ase
o
f
o
u
r
r
esear
ch
is
th
e
co
llectio
n
o
f
d
ata.
Fo
r
t
h
is
v
er
y
p
u
r
p
o
s
e,
we
co
llected
d
ata
f
r
o
m
Yo
u
T
u
b
e.
Fo
r
th
is
,
we
u
tili
ze
d
th
e
Yo
u
T
u
b
e
API
.
T
h
e
v
id
e
o
s
,
wh
ich
in
clu
d
ed
a
f
ew
well
-
k
n
o
wn
s
o
cial
m
ed
ia
p
er
s
o
n
alities
f
r
o
m
B
an
g
lad
esh
,
wer
e
h
an
d
-
p
i
ck
e
d
.
B
an
g
la
an
d
R
o
m
an
ize
d
B
an
g
la
tex
ts
wer
e
in
clu
d
ed
in
th
e
te
x
ts
.
T
h
e
tex
t
s
wer
e
d
iv
id
ed
in
to
two
d
atase
ts
.
T
h
er
e
wer
e
5
0
0
0
B
an
g
la
te
x
ts
in
Data
s
et
1
an
d
7
0
0
0
R
o
m
a
n
ized
B
an
g
la
tex
ts
in
Data
s
et
2
.
Af
ter
th
at,
th
e
f
ir
s
t
two
d
atasets
wer
e
co
m
b
in
ed
to
cr
ea
te
a
n
ew
d
ataset
with
a
to
tal
o
f
1
2
0
0
0
te
x
ts
.
Fo
llo
win
g
th
at,
we
an
n
o
ta
ted
all
th
e
d
atasets
in
to
2
ca
teg
o
r
ies:
b
u
lly
in
g
an
d
non
-
b
u
lly
in
g
.
So
m
e
o
f
th
e
a
n
n
o
tated
d
ata
is
p
r
esen
ted
in
T
ab
le
1
.
T
ab
le
1
.
Sam
p
le
o
f
an
n
o
tated
d
ata
Te
x
t
s
La
n
g
u
a
g
e
La
b
e
l
ফাউ
ল
ম
হ
িল
া
লা
ক
র
ে
না
হি
হি
(
T
h
e
f
o
u
l
w
o
m
a
n
i
s
n
o
t
a
s
h
a
m
e
d
)
B
a
n
g
l
a
1
(
B
u
l
l
y
i
n
g
)
অর
ে
বাল
ত
া
র
ক
হ
ি
হ
নই
না
.
.
.
.
ু
ই
ব
ার
ল
ে
হ
ি
াে
(
I
d
o
n
’
t
e
v
e
n
k
n
o
w
w
h
o
y
o
u
a
r
e
.
Y
o
u
a
re
a
h
o
rri
b
l
e
s
i
n
g
e
r.
)
B
a
n
g
l
a
1
(
B
u
l
l
y
i
n
g
)
ি
হ
িকা
ে
অ
র
থ
ে
ওম
ে
ি
ানী
ভ
াই
খ
ুব
ই
একট
া
অি
াধ
াে
ণ
ভ
ার
ল
া
ম
া
নুষ
(
T
ru
l
y
sp
e
a
k
i
n
g
O
m
o
r
S
a
n
i
b
r
o
t
h
e
r i
s
a
n
a
w
e
s
o
m
e
m
a
n
.
)
B
a
n
g
l
a
0
(
N
o
t
B
u
l
l
y
i
n
g
)
ভ
ার
ল
া
ল
া
গর
ল
া
,
ি
ব
া
ই
আম
া
ে
হ
ি
য়
ম
ানু
ষ
,
ি
ব
া
ই
অ
র
নক
ণ
ব
ান।
(
I
t
’
s g
o
o
d
,
e
v
e
ry
o
n
e
i
s my
f
a
v
o
u
ri
t
e
,
h
a
v
i
n
g
g
o
o
d
q
u
a
l
i
t
y
.
)
B
a
n
g
l
a
0
(
N
o
t
B
u
l
l
y
i
n
g
)
3
r
d
c
l
a
ss
q
u
a
l
i
t
y
r
2
p
e
r
s
o
n
(
Bo
t
h
o
f
t
h
e
m
a
r
e
t
h
i
rd
c
l
a
ss
q
u
a
l
i
t
y
p
e
rs
o
n
s
)
R
o
m
a
n
i
z
e
d
1
(
B
u
l
l
y
i
n
g
)
S
o
b
g
u
l
a
h
i
j
r
a
m
a
g
i
r
c
h
a
o
a
a
l
.
(
A
l
l
o
f
t
h
e
m
a
re
b
a
s
t
a
r
d
s
)
R
o
m
a
n
i
z
e
d
1
(
B
u
l
l
y
i
n
g
)
V
l
o
l
a
g
l
o
(
F
e
e
l
s
g
o
o
d
.
)
R
o
m
a
n
i
z
e
d
0
(
N
o
t
B
u
l
l
y
i
n
g
)
B
a
c
h
a
a
d
i
g
i
t
a
k
e
b
e
s
h
i
v
a
l
o
l
a
g
e
(
I
l
i
k
e
c
h
i
l
d
h
o
o
d
d
i
g
h
i
e
v
e
n
m
o
re
.
)
R
o
m
a
n
i
z
e
d
0
(
N
o
t
B
u
l
l
y
i
n
g
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
TEL
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
20
,
No
.
1
,
Feb
r
u
ar
y
20
22
:
89
-
97
92
3.
3
.
P
re
pro
ce
s
s
ing
W
e
s
tar
ted
th
e
p
r
ep
r
o
ce
s
s
in
g
b
y
r
em
o
v
i
n
g
a
n
y
d
u
p
licate
d
a
ta
f
r
o
m
o
u
r
d
atasets
.
All
th
r
e
e
d
atasets
wer
e
th
en
s
tr
ip
p
ed
o
f
d
ig
its
,
e
m
o
tico
n
s
,
p
u
n
ct
u
atio
n
m
ar
k
s
,
lin
k
s
,
u
s
er
tag
s
,
u
n
if
o
r
m
r
eso
u
r
ce
lo
ca
to
r
(
UR
L
)
’
s
,
elo
n
g
ated
wo
r
d
s
a
n
d
u
s
er
m
en
tio
n
s
.
So
m
e
o
f
th
e
tex
ts
c
o
n
s
is
ted
o
f
b
o
th
B
an
g
la
an
d
R
o
m
an
ized
B
an
g
la
wh
ic
h
wer
e
r
em
o
v
ed
f
r
o
m
th
e
d
ataset
in
o
r
d
er
to
o
b
tain
r
eliab
le
r
esu
lts
.
Als
o
,
we
d
id
n
o
t
p
er
f
o
r
m
an
y
s
to
p
wo
r
d
r
em
o
v
al,
s
tem
m
in
g
o
n
o
u
r
d
at
asets
s
in
ce
th
e
tex
ts
wer
e
m
ai
n
ly
in
lo
ca
l la
n
g
u
ag
e
.
3.
4
.
F
e
a
t
ure
ex
t
r
a
ct
io
n
W
e
u
s
e
d
t
e
r
m
f
r
e
q
u
e
n
c
y
-
i
n
v
e
r
s
e
d
o
c
u
m
e
n
t
f
r
e
q
u
e
n
c
y
(
T
F
-
I
D
F
)
t
o
e
x
t
r
a
ct
f
e
at
u
r
e
s
f
r
o
m
th
e
d
a
t
as
e
ts
.
TF
-
I
D
F
a
p
o
w
e
r
f
u
l
f
e
a
t
u
r
e
ex
t
r
a
c
t
i
o
n
t
e
c
h
n
i
q
u
e
w
h
i
c
h
i
d
e
n
t
i
f
i
es
i
m
p
o
r
t
a
n
t
w
o
r
d
s
i
n
te
x
t
u
a
l
d
a
t
a
[
2
3
]
.
I
t
t
r
a
n
s
f
o
r
m
s
s
t
r
i
n
g
s
i
n
t
o
n
u
m
e
r
i
ca
l
v
a
l
u
es
,
a
ll
o
w
i
n
g
m
ac
h
i
n
e
le
ar
n
i
n
g
c
l
a
s
s
i
f
ie
r
s
t
o
u
s
e
t
h
e
m
.
T
h
e
n
u
m
b
e
r
o
f
t
i
m
es
a
w
o
r
d
a
p
p
e
a
r
s
i
n
a
d
o
c
u
m
e
n
t
d
i
v
i
d
e
d
b
y
t
h
e
t
o
t
a
l
n
u
m
b
e
r
o
f
w
o
r
d
s
i
n
t
h
e
d
o
c
u
m
e
n
t
y
i
e
l
d
s
te
r
m
f
r
e
q
u
e
n
c
y
(
T
F
)
.
=
ℎ
ℎ
(
1
)
I
DF id
en
tifie
s
th
e
weig
h
ts
o
f
e
s
s
en
tial w
o
r
d
s
in
a
d
o
cu
m
e
n
t.
I
t is m
ea
s
u
r
ed
u
s
in
g
(
2
)
.
=
l
og
2
(
ℎ
)
(
2
)
Fin
ally
,
b
o
th
th
e
ter
m
f
r
eq
u
en
c
y
an
d
in
v
er
s
e
d
o
cu
m
en
t
f
r
e
q
u
e
n
cy
ca
n
b
e
m
u
ltip
lied
to
o
b
tain
th
e
T
F
-
I
DF
wh
ich
will h
av
e
n
o
r
m
alize
d
weig
h
ts
.
I
t is ca
lcu
lated
with
(
3
)
.
−
=
∗
(
3
)
3.
5
.
M
a
chine le
a
rning
cla
s
s
i
f
iers
Ma
ch
in
e
lear
n
in
g
class
if
ier
s
ar
e
wid
ely
u
tili
ze
d
t
o
p
r
ed
ict
c
ateg
o
r
ical
d
ata.
T
o
d
a
y
m
ac
h
i
n
e
lear
n
in
g
is
u
s
ed
to
b
u
ild
d
if
f
er
e
n
t
in
tellig
en
t
s
y
s
tem
s
th
at
m
ak
es
d
ec
is
s
io
n
m
ak
in
g
ea
s
ier
.
Ma
ch
in
e
lear
n
in
g
is
a
b
r
o
a
d
ter
m
th
at
en
c
o
m
p
ass
es
s
u
p
er
v
is
ed
,
u
n
s
u
p
er
v
is
ed
an
d
r
ein
f
o
r
ce
m
en
t
lear
n
i
n
g
.
I
n
th
is
r
ese
ar
ch
,
we
u
s
ed
f
o
u
r
s
u
p
er
v
is
ed
m
ac
h
in
e
lear
n
in
g
c
lass
if
ier
s
.
3
.
5
.
1
.
M
ultino
m
ia
l na
iv
e
B
a
y
es
T
h
e
m
u
ltin
o
m
ial
n
aiv
e
B
ay
e
s
alg
o
r
ith
m
is
a
p
r
o
b
ab
ilis
tic
lear
n
in
g
m
eth
o
d
p
o
p
u
lar
in
NL
P.
T
h
e
alg
o
r
ith
m
p
r
e
d
icts
u
s
in
g
th
e
B
ay
es
th
eo
r
em
[
2
4
]
.
I
t
ca
lcu
lates
p
r
o
b
ab
ilit
y
f
o
r
a
g
iv
e
n
s
am
p
le
an
d
o
u
tp
u
ts
th
e
v
alu
e
with
th
e
h
i
g
h
est p
r
o
b
ab
i
lity
u
s
in
g
(
4
)
.
(
|
)
=
(
|
)
∗
(
)
/
(
)
(
4
)
3
.
5
.
2
.
Su
pp
o
rt
v
ec
t
o
r
m
a
chi
ne
(
SVM
)
Su
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SV
M)
is
v
astl
y
u
s
ed
f
o
r
class
if
icatio
n
p
r
o
b
lem
s
.
I
t
class
if
ies
d
ata
b
y
g
en
er
atin
g
a
d
ec
is
io
n
b
o
u
n
d
a
r
y
o
r
h
y
p
e
r
p
lan
e
in
an
n
-
d
im
en
s
io
n
al
s
p
ac
e
[
2
5
]
.
T
o
ch
o
o
s
e
t
h
e
b
est
p
lan
e
am
o
n
g
n
u
m
er
o
u
s
p
o
s
s
ib
le
p
lan
es,
th
e
v
alu
e
th
at
h
as th
e
h
ig
h
est m
ar
g
in
is
ch
o
s
en
.
I
t
h
as a
n
e
d
g
e
o
v
er
o
th
e
r
class
if
ier
s
b
ec
au
s
e
to
its
f
aster
p
r
o
ce
s
s
in
g
s
p
ee
d
an
d
g
r
ea
ter
p
er
f
o
r
m
a
n
ce
with
less
s
am
p
les
.
3
.
5
.
3
.
L
o
g
is
t
ic
re
g
re
s
s
io
n
Fo
r
b
in
ar
y
class
if
icatio
n
,
lo
g
is
tic
r
eg
r
ess
io
n
is
a
co
m
m
o
n
ly
u
s
ed
class
if
ier
.
T
o
class
if
y
d
ata,
lo
g
is
tic
r
eg
r
ess
io
n
u
s
es
a
s
ig
m
o
id
f
u
n
ctio
n
.
T
h
e
f
u
n
ctio
n
co
n
v
er
ts
an
y
r
ea
l
v
al
u
e
b
etwe
en
0
to
1
[
2
6
]
.
T
h
e
s
ig
m
o
id
f
u
n
ctio
n
is
s
h
o
wn
in
(
5
)
.
(
)
=
1
1
+
−
(
5
)
T
h
e
v
alu
es
th
at
th
e
f
u
n
ctio
n
r
etu
r
n
s
,
is
co
n
v
er
te
d
in
to
0
o
r
1
.
T
o
d
o
s
o
,
a
th
r
esh
o
ld
v
alu
e
is
s
et.
T
h
e
v
alu
es
ab
o
v
e
th
e
th
r
esh
o
ld
v
alu
e
ar
e
class
if
ied
as c
las
s
1
an
d
b
elo
w
ar
e
class
if
ied
as c
lass
0
.
3
.
5
.
4
.
XG
B
o
o
s
t
XGBo
o
s
t is an
en
s
em
b
le
o
f
d
ec
is
io
n
tr
ee
s
[
2
7
]
.
I
t is a
m
ac
h
i
n
e
lear
n
in
g
class
if
ier
th
at
u
s
es a
g
r
ad
ien
t
b
o
o
s
tin
g
al
g
o
r
ith
m
.
XGBo
o
s
t
is
k
n
o
wn
f
o
r
its
f
aster
ex
ec
u
tio
n
s
p
ee
d
a
n
d
h
ig
h
e
r
m
o
d
el
p
er
f
o
r
m
an
ce
.
XGBo
o
s
t
is
ex
tr
em
ely
u
s
ef
u
l f
o
r
ac
h
ie
v
in
g
g
o
o
d
r
esu
lts
with
m
in
im
al
r
eso
u
r
ce
s
an
d
tim
e.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
N
a
tu
r
a
l la
n
g
u
a
g
e
p
r
o
ce
s
s
in
g
a
n
d
ma
c
h
in
e
lea
r
n
in
g
b
a
s
ed
c
yb
erb
u
llyin
g
d
etec
tio
n
fo
r
… (
Md
.
To
fa
el
A
h
med
)
93
3.
6
.
P
er
f
o
r
m
a
nce
ev
a
lua
t
io
n
T
o
an
al
y
ze
th
e
p
er
f
o
r
m
a
n
ce
o
f
an
y
q
u
alif
ied
m
ac
h
in
e
lear
n
i
n
g
class
if
ier
,
p
e
r
f
o
r
m
an
ce
ev
a
lu
atio
n
is
cr
itical.
W
e
co
n
s
id
er
ed
co
n
f
u
s
io
n
m
atr
i
x
,
p
r
ec
is
io
n
,
r
ec
all,
f
1
-
s
co
r
e,
ac
cu
r
ac
y
an
d
AUC
-
R
OC
cu
r
v
e
[
2
8
]
,
[
2
9
]
f
o
r
p
er
f
o
r
m
an
ce
ev
alu
atio
n
.
W
e
also
s
h
o
wed
h
o
w
m
a
n
y
p
r
ed
ictio
n
s
wer
e
c
o
r
r
ec
tly
o
r
in
c
o
r
r
ec
tly
d
o
n
e
b
y
th
e
class
if
ier
s
.
T
h
e
co
n
f
u
s
io
n
m
atr
ix
is
a
v
er
y
im
p
o
r
tan
t
p
er
f
o
r
m
a
n
ce
ev
alu
a
tio
n
p
ar
am
eter
.
I
t
is
a
co
m
b
in
a
tio
n
o
f
f
o
r
d
is
tin
ct
ac
tu
al
an
d
p
r
ed
icted
v
alu
es.
C
o
n
f
u
s
io
n
m
atr
ix
p
la
y
s
a
v
er
y
v
ital
r
o
le
in
c
o
m
p
u
tin
g
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
f
1
-
s
co
r
e
an
d
th
e
AUC
-
R
OC
cu
r
v
e.
T
h
e
r
atio
o
f
ac
c
u
r
ate
p
r
ed
ictio
n
s
to
t
h
e
to
tal
n
u
m
b
er
o
f
in
p
u
t
s
am
p
les
d
eter
m
in
es th
e
ac
cu
r
ac
y
r
ate
[
2
8
]
an
d
is
ca
lcu
late
d
u
s
in
g
(
6
)
.
=
+
+
+
+
(
6
)
T
h
e
n
u
m
b
er
o
f
ac
c
u
r
ate
p
o
s
itiv
e
p
r
e
d
ictio
n
s
d
i
v
id
ed
b
y
th
e
to
tal
n
u
m
b
e
r
o
f
p
o
s
itiv
e
p
r
e
d
ictio
n
s
m
ad
e
b
y
a
class
if
ier
y
ield
s
th
e
p
r
ec
is
io
n
v
alu
e
[
2
8
]
an
d
it is
ca
lcu
lated
u
s
in
g
(
7
)
.
=
+
(
7
)
T
h
e
n
u
m
b
e
r
o
f
ac
cu
r
ate
p
o
s
itiv
e
p
r
e
d
ictio
n
s
d
iv
id
ed
b
y
th
e
to
tal
n
u
m
b
er
o
f
ac
tu
al
p
o
s
itiv
e
s
am
p
les
y
ield
s
th
e
r
ec
all
v
alu
e
[
2
8
]
.
I
t is ca
lcu
lated
u
s
in
g
(
8
)
.
=
+
(
8
)
T
h
e
h
ar
m
o
n
ic
m
ea
n
o
f
p
r
ec
i
s
io
n
an
d
r
ec
all
is
th
e
f
1
-
s
co
r
e
[
2
8
]
.
B
etter
o
u
tp
u
t
is
ass
o
c
iated
with
a
h
i
g
h
er
f1
-
s
co
r
e.
1
−
=
2
∗
∗
+
(
9
)
T
h
e
AUC
-
R
O
C
cu
r
v
e
tells
u
s
h
o
w
g
o
o
d
a
m
o
d
el
is
at
d
i
s
ti
n
g
u
is
h
in
g
b
etwe
en
class
es
[
2
9
]
.
A
h
ig
h
er
AUC
in
d
icate
s
th
at
th
e
m
o
d
el
is
b
ett
er
at
p
r
ed
ictio
n
.
4.
RE
SU
L
T
S
AND
D
I
SCU
SS
I
O
N
W
e
d
iv
id
ed
o
u
r
d
atasets
in
to
8
0
%
f
o
r
tr
ain
in
g
an
d
2
0
%
f
o
r
test
in
g
.
Af
ter
tr
ain
in
g
th
e
class
if
ier
s
with
8
0
%
d
ata,
we
u
s
ed
t
h
e
2
0
%
tes
tin
g
s
ets
to
ev
alu
ate
p
er
f
o
r
m
a
n
ce
.
T
ab
le
2
d
is
p
lay
s
th
e
to
tal
n
u
m
b
er
o
f
co
r
r
ec
tly
an
d
i
n
co
r
r
ec
tly
r
ec
o
g
n
ized
in
s
tan
ce
s
f
o
r
ea
ch
class
if
ier
ac
r
o
s
s
all
th
e
d
atasets
.
Fig
u
r
e
2
s
h
o
ws
th
at
SVM
co
r
r
ec
tly
id
en
tifie
d
7
5
7
in
s
tan
ce
s
7
5
.
7
%
in
Data
s
et
1
.
m
u
l
tin
o
m
ial
n
aiv
e
B
ay
es
h
as
th
e
h
ig
h
est
n
u
m
b
er
o
f
co
r
r
ec
tly
class
if
ied
in
s
tan
ce
s
in
Data
s
ets
2
an
d
3
,
with
1
1
8
0
8
4
.
2
8
%
an
d
1
9
2
8
8
0
.
3
3
%
r
esp
ec
tiv
ely
.
W
ith
th
e
g
r
ea
test
n
u
m
b
er
o
f
co
r
r
ec
tly
class
if
ied
in
s
tan
ce
s
,
SVM
o
u
tp
er
f
o
r
m
ed
all
o
th
e
r
alg
o
r
ith
m
s
in
Data
s
et
1
.
Similar
ly
,
m
u
ltin
o
m
ial
Naïv
e
B
ay
es
s
to
o
d
o
u
t
m
o
s
t
f
o
r
D
ataset
2
an
d
Data
s
et
3
.
T
ab
le
2
also
s
h
o
ws
th
e
co
n
f
u
s
io
n
m
atr
ix
o
f
th
e
b
est p
er
f
o
r
m
in
g
alg
o
r
ith
m
s
f
o
r
ea
ch
d
ataset.
Fro
m
T
ab
le
2
,
it
ca
n
b
e
s
ee
n
th
at
3
6
3
cy
b
er
b
u
lly
i
n
g
tex
ts
an
d
3
9
4
n
o
n
-
cy
b
er
b
u
lly
i
n
g
tex
ts
o
f
Data
s
et
1
ar
e
clas
s
if
ied
co
r
r
ec
tly
b
y
SVM
f
o
r
Data
s
et
2
,
6
6
1
cy
b
er
b
u
lly
in
g
tex
ts
an
d
5
1
9
n
o
n
-
cy
b
er
b
u
lly
in
g
tex
ts
ar
e
class
if
ied
co
r
r
ec
tly
b
y
m
u
ltin
o
m
ial
n
ai
v
e
B
ay
es.
Fo
r
Data
s
et
3
,
1
0
8
6
cy
b
er
b
u
lly
in
g
tex
ts
an
d
8
4
2
non
-
c
y
b
er
b
u
lly
in
g
tex
ts
ar
e
class
if
ied
co
r
r
ec
tly
.
T
ab
le
3
s
h
o
ws
th
e
p
r
ec
is
io
n
,
r
ec
all
an
d
f
1
-
s
co
r
e
o
f
all
alg
o
r
ith
m
s
f
o
r
all
d
a
tasets
in
d
etails an
d
Fig
u
r
e
3
s
h
o
ws th
e
ac
cu
r
ac
y
o
f
all
alg
o
r
it
h
m
s
.
T
ab
le
3
an
d
Fig
u
r
e
3
s
h
o
w
th
at
f
o
r
Data
s
et
1
,
SVM
ac
h
iev
ed
p
r
ec
is
io
n
,
r
ec
all
an
d
f
1
-
s
co
r
e
o
f
0
.
7
6
ea
ch
,
as
well
a
s
an
o
v
er
all
ac
c
u
r
ac
y
o
f
7
6
%,
th
e
h
ig
h
est
o
f
all
th
e
alg
o
r
ith
m
s
.
Fo
r
Data
s
et
2
,
m
u
ltin
o
m
ial
n
aiv
e
B
ay
es
ac
h
iev
ed
p
r
ec
is
io
n
,
r
ec
all
an
d
f
1
-
s
co
r
e
o
f
0
.
8
4
ea
ch
,
as
well
as
o
v
er
all
ac
cu
r
ac
y
o
f
8
4
%,
th
e
h
i
g
h
e
s
t
o
f
a
l
l
t
h
e
al
g
o
r
i
t
h
m
s
.
Fi
n
a
l
l
y
,
m
u
l
t
i
n
o
m
i
a
l
n
a
i
v
e
B
a
y
e
s
a
g
a
i
n
o
u
t
p
e
r
f
o
r
m
e
d
a
l
l
o
t
h
e
r
al
g
o
r
i
t
h
m
s
f
o
r
D
a
t
as
e
t
3
b
y
a
c
h
i
e
v
i
n
g
p
r
e
c
i
s
i
o
n
o
f
0
.
8
1
,
r
e
c
a
l
l
a
n
d
f
1
-
s
c
o
r
e
o
f
0
.
8
0
e
a
c
h
,
a
s
w
el
l
as
a
n
o
v
e
r
a
l
l
ac
c
u
r
a
c
y
o
f
8
0
%
.
An
o
th
er
p
er
f
o
r
m
an
ce
an
al
y
s
is
is
th
e
R
OC
ar
ea
.
T
h
e
lar
g
er
th
e
R
OC
ar
ea
,
th
e
m
o
r
e
ac
c
u
r
ate
ly
a
m
o
d
el
ca
n
id
en
tify
in
s
tan
ce
s
.
Fig
u
r
e
4
s
h
o
ws
th
e
R
OC
cu
r
v
e
o
f
S
VM
f
o
r
Da
taset
1
as
well
as
t
h
e
R
OC
cu
r
v
es
o
f
m
u
ltin
o
m
ial
Naïv
e
B
ay
es
f
o
r
Data
s
et
2
an
d
3
as
th
ese
two
alg
o
r
ith
m
s
p
er
f
o
r
m
e
d
b
est
am
o
n
g
all
f
o
u
r
alg
o
r
ith
m
s
.
As
s
h
o
wn
in
Fig
u
r
e
4
,
it
is
clea
r
th
at
th
e
h
i
g
h
est
p
er
f
o
r
m
in
g
alg
o
r
ith
m
is
m
u
ltin
o
m
ial
n
aiv
e
B
ay
es.
I
t
p
er
f
o
r
m
s
b
est f
o
r
Data
s
et
2
an
d
3
.
I
t a
ls
o
p
er
f
o
r
m
s
r
ea
s
o
n
a
b
ly
well
f
o
r
Data
s
et
1
b
u
t w
as o
u
tp
er
f
o
r
m
ed
b
y
SVM.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
TEL
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
20
,
No
.
1
,
Feb
r
u
ar
y
20
22
:
89
-
97
94
(
a)
(
b
)
Fig
u
r
e
2
.
Nu
m
b
er
o
f
co
r
r
ec
tly
an
d
in
co
r
r
ec
tly
class
if
ied
in
s
tan
ce
s
:
(
a)
n
u
m
b
er
o
f
co
r
r
ec
tly
class
if
ied
in
s
tan
ce
s
an
d
(
b
)
n
u
m
b
er
o
f
in
co
r
r
ec
tly
class
if
ied
in
s
tan
ce
s
T
ab
le
2
.
C
o
n
f
u
s
io
n
m
atr
ix
o
f
b
est p
er
f
o
r
m
in
g
alg
o
r
ith
m
s
f
o
r
ea
ch
d
ataset
D
a
t
a
s
e
t
A
l
g
o
r
i
t
h
m
TP
FP
FN
TN
D
a
t
a
s
e
t
1
S
V
M
3
6
3
1
4
1
1
0
2
3
9
4
D
a
t
a
s
e
t
2
M
u
l
t
i
n
o
m
i
a
l
N
a
ï
v
e
B
a
y
e
s
6
6
1
78
1
4
2
5
1
9
D
a
t
a
s
e
t
3
M
u
l
t
i
n
o
m
i
a
l
N
a
ï
v
e
B
a
y
e
s
1
0
8
6
1
5
8
3
1
4
8
4
2
T
ab
le
3
.
Pre
cisi
o
n
,
r
ec
all
an
d
F1
-
s
co
r
es b
y
class
o
f
all
alg
o
r
i
th
m
s
f
o
r
all
d
atasets
D
a
t
a
s
e
t
s
A
l
g
o
r
i
t
h
m
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F1
-
sc
o
r
e
D
a
t
a
s
e
t
1
M
u
l
t
i
n
o
m
i
a
l
N
a
ï
v
e
B
a
y
e
s
0
.
7
4
0
.
7
4
0
.
7
4
S
V
M
0
.
7
6
0
.
7
6
0
.
7
6
Lo
g
i
s
t
i
c
R
e
g
r
e
ssi
o
n
0
.
7
6
0
.
7
5
0
.
7
5
X
G
B
o
o
st
0
.
7
5
0
.
7
4
0
.
7
4
D
a
t
a
s
e
t
2
M
u
l
t
i
n
o
m
i
a
l
N
a
ï
v
e
B
a
y
e
s
0
.
8
4
0
.
8
4
0
.
8
4
S
V
M
0
.
8
3
0
.
8
3
0
.
8
3
Lo
g
i
s
t
i
c
R
e
g
r
e
ssi
o
n
0
.
8
3
0
.
8
3
0
.
8
2
X
G
B
o
o
st
0
.
8
0
0
.
7
8
0
.
7
8
D
a
t
a
s
e
t
3
M
u
l
t
i
n
o
m
i
a
l
N
a
ï
v
e
B
a
y
e
s
0
.
8
1
0
.
8
0
0
.
8
0
S
V
M
0
.
7
9
0
.
7
9
0
.
7
9
Lo
g
i
s
t
i
c
R
e
g
r
e
ssi
o
n
0
.
7
9
0
.
7
9
0
.
7
9
X
G
B
o
o
st
0
.
7
6
0
.
7
4
0
.
7
4
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
N
a
tu
r
a
l la
n
g
u
a
g
e
p
r
o
ce
s
s
in
g
a
n
d
ma
c
h
in
e
lea
r
n
in
g
b
a
s
ed
c
yb
erb
u
llyin
g
d
etec
tio
n
fo
r
… (
Md
.
To
fa
el
A
h
med
)
95
Fig
u
r
e
3
.
Acc
u
r
ac
y
o
f
all
th
e
c
lass
if
ier
s
Fig
u
r
e
4
.
R
OC
cu
r
v
es f
o
r
th
e
b
est p
er
f
o
r
m
in
g
alg
o
r
ith
m
s
5.
CO
NCLU
SI
O
N
AND
F
U
T
U
RE
WO
RK
W
ith
th
e
in
cr
ea
s
e
u
s
e
o
f
d
if
f
er
en
t
s
o
cial
m
ed
ia
s
ites
p
eo
p
le
a
r
e
in
ter
ac
tin
g
with
ea
c
h
o
t
h
er
m
o
r
e
o
f
te
n
wh
ich
h
as
also
b
r
o
u
g
h
t
an
in
c
r
ea
s
e
in
th
e
am
o
u
n
t
o
f
cy
b
er
b
u
lly
in
g
.
T
o
d
etec
t
th
ese
cy
b
er
b
u
lly
in
g
tex
ts
,
we
d
ev
elo
p
e
d
m
o
d
els
b
ased
o
n
NL
P
tech
n
iq
u
es
an
d
m
ac
h
in
e
lear
n
in
g
.
W
e
co
llected
a
to
tal
o
f
1
2
0
0
0
te
x
ts
f
r
o
m
Yo
u
T
u
b
e
an
d
p
r
ep
a
r
ed
th
r
ee
d
atasets
.
T
h
ese
d
atase
ts
we
r
e
u
s
ed
to
tr
ain
th
e
m
o
d
els.
Af
ter
test
in
g
two
alg
o
r
ith
m
s
s
to
o
d
o
u
t
m
o
s
t.
SVM
p
er
f
o
r
m
ed
b
est
f
o
r
1
s
t
d
a
taset
wi
th
ac
cu
r
ac
y
o
f
7
6
%
an
d
m
u
ltin
o
m
i
al
n
aiv
e
B
ay
es
p
r
o
d
u
ce
d
b
est
r
esu
lts
f
o
r
2
n
d
an
d
3
r
d
d
ataset
with
ac
cu
r
ac
y
o
f
8
4
%
an
d
8
0
%.
SVM
also
o
b
tain
ed
p
r
ec
is
io
n
o
f
7
4
%,
r
ec
all
7
5
%
an
d
f
1
-
s
co
r
e
7
4
%
f
o
r
f
ir
s
t
d
ataset.
Fo
r
th
e
s
ec
o
n
d
d
ataset
m
u
ltin
o
m
ial
n
aiv
e
B
ay
es
g
o
t
8
5
%
p
r
ec
is
io
n
,
8
4
%
r
ec
all
an
d
8
4
%
f
1
-
s
co
r
e
a
n
d
f
o
r
th
e
th
ir
d
d
ataset
m
u
ltin
o
m
ial
Naïv
e
B
ay
es
g
o
t
8
1
%
p
r
ec
is
io
n
,
8
0
%
r
ec
all
an
d
f
1
-
s
co
r
es.
T
h
e
o
th
er
two
al
g
o
r
ith
m
s
,
lo
g
is
tic
r
eg
r
ess
io
n
an
d
XGBo
o
s
t
also
p
er
f
o
r
m
ed
r
ea
s
o
n
a
b
ly
well
b
u
t
was
s
lig
h
tly
o
u
tp
er
f
o
r
m
e
d
b
y
SVM
an
d
XGBo
o
s
t.
T
h
ese
tr
ai
n
ed
m
o
d
els
ca
n
b
e
u
s
ed
to
d
etec
t
B
an
g
la
a
n
d
R
o
m
an
ized
B
an
g
la
cy
b
er
b
u
lly
in
g
tex
ts
o
f
Yo
u
T
u
b
e
at
an
ea
r
ly
s
tag
e
an
d
s
to
p
th
e
m
b
ein
g
co
m
m
en
ted
.
I
n
th
e
f
u
tu
r
e
we
wan
t to
wo
r
k
with
m
o
r
e
d
ata
an
d
also
m
o
r
e
v
i
d
eo
s
o
f
d
if
f
er
en
t c
ateg
o
r
ies.
RE
F
E
R
E
NC
E
S
[1
]
A
.
W
h
i
t
i
n
g
a
n
d
D
.
W
i
l
l
i
a
ms
,
“
W
h
y
p
e
o
p
l
e
u
se
s
o
c
i
a
l
m
e
d
i
a
:
a
u
ses
a
n
d
g
r
a
t
i
f
i
c
a
t
i
o
n
s
a
p
p
r
o
a
c
h
,
”
Q
u
a
l
.
M
a
rk
.
R
e
s.
I
n
t
.
J
.
,
v
o
l
.
1
6
,
n
o
.
4
,
p
p
.
3
6
2
–
3
6
9
,
2
0
1
3
,
d
o
i
:
1
0
.
1
1
0
8
/
Q
M
R
-
06
-
2
0
1
3
-
0
0
4
1
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
TEL
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
20
,
No
.
1
,
Feb
r
u
ar
y
20
22
:
89
-
97
96
[2
]
S
.
H
i
n
d
u
j
a
a
n
d
J.
W
.
P
a
t
c
h
i
n
,
“
C
y
b
e
r
b
u
l
l
y
i
n
g
:
A
n
e
x
p
l
o
r
a
t
o
r
y
a
n
a
l
y
si
s
o
f
f
a
c
t
o
r
s
r
e
l
a
t
e
d
t
o
o
f
f
e
n
d
i
n
g
a
n
d
v
i
c
t
i
m
i
z
a
t
i
o
n
,
”
De
v
i
a
n
t
b
e
h
a
v
i
o
r
,
v
o
l
.
2
9
,
n
o
.
2
,
p
.
1
2
9
–
1
5
6
,
2
0
0
8
,
d
o
i
:
1
0
.
1
0
8
0
/
0
1
6
3
9
6
2
0
7
0
1
4
5
7
8
1
6
.
[3
]
A
.
N
.
D
o
a
n
e
,
M
.
L
.
K
e
l
l
e
y
,
E.
S
.
C
h
i
a
n
g
,
a
n
d
M
.
A
.
P
a
d
i
l
l
a
,
“
D
e
v
e
l
o
p
me
n
t
o
f
t
h
e
C
y
b
e
r
b
u
l
l
y
i
n
g
Ex
p
e
r
i
e
n
c
e
s
S
u
r
v
e
y
,
”
Em
e
rg
.
Ad
u
l
t
h
o
o
d
,
v
o
l
.
1
,
n
o
.
3
,
p
p
.
2
0
7
–
2
1
8
,
2
0
1
3
,
d
o
i
:
1
0
.
1
1
7
7
/
2
1
6
7
6
9
6
8
1
3
4
7
9
5
8
4
.
[4
]
A
.
B
h
a
r
a
t
,
N
.
P
o
o
j
a
,
a
n
d
R
.
A
.
R
e
d
d
y
,
"
U
si
n
g
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
m
s
f
o
r
b
r
e
a
s
t
c
a
n
c
e
r
r
i
sk
p
r
e
d
i
c
t
i
o
n
a
n
d
d
i
a
g
n
o
s
i
s,"
2
0
1
8
3
rd
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
C
i
rc
u
i
t
s,
C
o
n
t
ro
l
,
C
o
m
m
u
n
i
c
a
t
i
o
n
a
n
d
C
o
m
p
u
t
i
n
g
(
I
4
C
)
,
2
0
1
8
,
p
p
.
1
-
4
,
d
o
i
:
1
0
.
1
1
0
9
/
C
I
M
C
A
.
2
0
1
8
.
8
7
3
9
6
9
6
.
[5
]
H
.
K
a
u
r
a
n
d
V
.
K
u
m
a
r
i
,
"
P
r
e
d
i
c
t
i
v
e
mo
d
e
l
l
i
n
g
a
n
d
a
n
a
l
y
t
i
c
s
f
o
r
d
i
a
b
e
t
e
s
u
si
n
g
a
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
,
"
A
p
p
l
i
e
d
C
o
m
p
u
t
i
n
g
a
n
d
I
n
f
o
rm
a
t
i
c
s
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
c
i
.
2
0
1
8
.
1
2
.
0
0
4
.
[6
]
K
.
F
a
r
h
a
n
a
,
M
.
R
a
h
ma
n
,
a
n
d
M
.
A
h
med
,
"
A
n
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
s
y
st
e
m
f
o
r
p
a
c
k
e
t
a
n
d
f
l
o
w
b
a
se
d
n
e
t
w
o
r
k
s
u
si
n
g
d
e
e
p
n
e
u
r
a
l
n
e
t
w
o
r
k
a
p
p
r
o
a
c
h
,
"
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
t
ri
c
a
l
a
n
d
C
o
m
p
u
t
e
r
En
g
i
n
e
e
ri
n
g
(
I
J
E
C
E)
,
v
o
l
.
1
0
,
n
o
.
5
,
p
p
.
5
5
1
4
-
5
5
2
5
,
2
0
2
0
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
c
e
.
v
1
0
i
5
.
p
p
5
5
1
4
-
5
5
2
5
.
[7
]
S
.
Za
n
d
e
r
,
T.
N
g
u
y
e
n
a
n
d
G
.
A
r
mi
t
a
g
e
,
"
A
u
t
o
m
a
t
e
d
t
r
a
f
f
i
c
c
l
a
ssi
f
i
c
a
t
i
o
n
a
n
d
a
p
p
l
i
c
a
t
i
o
n
i
d
e
n
t
i
f
i
c
a
t
i
o
n
u
s
i
n
g
mac
h
i
n
e
l
e
a
r
n
i
n
g
,
"
T
h
e
I
EEE
C
o
n
f
e
r
e
n
c
e
o
n
L
o
c
a
l
C
o
m
p
u
t
e
r
N
e
t
w
o
r
k
s
3
0
t
h
A
n
n
i
v
e
rs
a
ry
(
L
C
N
'
0
5
)
l
,
2
0
0
5
,
p
p
.
2
5
0
-
2
5
7
,
d
o
i
:
1
0
.
1
1
0
9
/
L
C
N
.
2
0
0
5
.
3
5
.
[8
]
M
.
H
o
ss
a
i
n
,
M
.
R
a
h
ma
n
,
M
.
A
h
me
d
,
M
.
R
a
h
ma
n
,
a
n
d
A
.
I
sl
a
m,
"
R
a
t
i
n
g
P
r
e
d
i
c
t
i
o
n
o
f
P
r
o
d
u
c
t
R
e
v
i
e
w
s
o
f
B
a
n
g
l
a
L
a
n
g
u
a
g
e
u
si
n
g
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
A
l
g
o
r
i
t
h
m
s
,
"
2
0
2
1
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
Me
c
h
a
t
ro
n
i
c
s
S
y
s
t
e
m
s
(
AI
MS
)
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
0
9
/
a
i
ms
5
2
4
1
5
.
2
0
2
1
.
9
4
6
6
0
2
2
.
[9
]
M
.
R
a
h
m
a
n
,
M
.
T.
A
h
me
d
,
a
n
d
A
.
Z.
M
.
T.
I
sl
a
m
,
"
O
n
D
e
t
e
c
t
i
o
n
a
n
d
Tr
a
c
k
i
n
g
o
f
t
h
e
V
e
h
i
c
l
e
s
f
r
o
m t
h
e
R
e
a
l
Ti
me
V
i
d
e
o
S
t
r
e
a
m
u
si
n
g
B
a
c
k
g
r
o
u
n
d
S
u
b
t
r
a
c
t
i
o
n
P
r
o
c
e
s
s
w
i
t
h
B
l
o
b
Tr
a
c
k
e
r
A
l
g
o
r
i
t
h
m,"
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
Am
b
i
e
n
t
S
y
s
t
e
m
s
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
(
I
J
AS
A)
,
v
o
l
.
0
8
,
p
p
.
2
1
-
3
0
1
0
5
1
2
1
,
2
0
2
0
,
d
o
i
:
1
0
.
5
1
2
1
/
i
j
a
s
a
.
2
0
2
0
.
8
2
0
3
.
[1
0
]
M
.
I
k
o
n
o
m
a
k
i
s,
S
.
K
o
t
s
i
a
n
t
i
s,
a
n
d
V
.
Ta
mp
a
k
a
s,
“
Te
x
t
c
l
a
ssi
f
i
c
a
t
i
o
n
u
s
i
n
g
mac
h
i
n
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s
,
”
W
S
E
A
S
T
ra
n
s
a
c
t
i
o
n
s
o
n
C
o
m
p
u
t
e
rs
,
v
o
l
.
4
,
n
o
.
8
,
p
p
.
9
6
6
–
9
7
4
,
2
0
0
5
.
[1
1
]
E.
B
o
i
y
a
n
d
M
.
-
F
.
M
o
e
n
s,
"
A
mac
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
t
o
s
e
n
t
i
m
e
n
t
a
n
a
l
y
si
s i
n
m
u
l
t
i
l
i
n
g
u
a
l
W
e
b
t
e
x
t
s,"
I
n
f
o
r
m
a
t
i
o
n
Re
t
ri
e
v
a
l
,
v
o
l
.
1
2
,
n
o
.
5
,
p
p
.
5
2
6
-
5
5
8
,
2
0
0
8
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
0
7
9
1
-
0
0
8
-
9
0
7
0
-
z.
[1
2
]
B
.
H
a
i
d
a
r
,
M
.
C
h
a
m
o
u
n
,
a
n
d
A
.
S
e
r
h
r
o
u
c
h
n
i
,
"
A
M
u
l
t
i
l
i
n
g
u
a
l
S
y
s
t
e
m
f
o
r
C
y
b
e
r
b
u
l
l
y
i
n
g
D
e
t
e
c
t
i
o
n
:
A
r
a
b
i
c
C
o
n
t
e
n
t
D
e
t
e
c
t
i
o
n
u
s
i
n
g
M
a
c
h
i
n
e
Le
a
r
n
i
n
g
,
"
A
d
v
a
n
c
e
s
i
n
S
c
i
e
n
c
e
,
T
e
c
h
n
o
l
o
g
y
a
n
d
En
g
i
n
e
e
r
i
n
g
S
y
st
e
m
s
J
o
u
r
n
a
l
,
v
o
l
.
2
,
n
o
.
6
,
p
p
.
2
7
5
-
2
8
4
,
2
0
1
7
.
d
o
i
:
1
0
.
2
5
0
4
6
/
a
j
0
2
0
6
3
4
.
[1
3
]
G
.
A
.
L
e
ó
n
-
P
a
r
e
d
e
s
e
t
a
l
.
,
"
P
r
e
su
m
p
t
i
v
e
D
e
t
e
c
t
i
o
n
o
f
C
y
b
e
r
b
u
l
l
y
i
n
g
o
n
Tw
i
t
t
e
r
t
h
r
o
u
g
h
N
a
t
u
r
a
l
La
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
Le
a
r
n
i
n
g
i
n
t
h
e
S
p
a
n
i
sh
L
a
n
g
u
a
g
e
,
"
2
0
1
9
I
EE
E
C
H
I
L
EAN
C
o
n
f
e
re
n
c
e
o
n
El
e
c
t
ri
c
a
l
,
El
e
c
t
ro
n
i
c
s
En
g
i
n
e
e
ri
n
g
,
I
n
f
o
rm
a
t
i
o
n
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
T
e
c
h
n
o
l
o
g
i
e
s (
C
H
I
L
EC
O
N)
,
2
0
1
9
,
p
p
.
1
-
7
,
d
o
i
:
1
0
.
1
1
0
9
/
C
H
I
LEC
O
N
4
7
7
4
6
.
2
0
1
9
.
8
9
8
7
6
8
4
.
[1
4
]
E.
G
r
e
e
v
y
a
n
d
A
.
S
m
e
a
t
o
n
,
"
C
l
a
ss
i
f
y
i
n
g
r
a
c
i
s
t
t
e
x
t
s
u
si
n
g
a
su
p
p
o
r
t
v
e
c
t
o
r
mac
h
i
n
e
,
"
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
2
7
t
h
a
n
n
u
a
l
i
n
t
e
r
n
a
t
i
o
n
a
l
c
o
n
f
e
re
n
c
e
o
n
R
e
se
a
rc
h
a
n
d
d
e
v
e
l
o
p
m
e
n
t
i
n
i
n
f
o
r
m
a
t
i
o
n
re
t
ri
e
v
a
l
-
S
I
G
I
R
'
0
4
,
2
0
0
4
,
p
p
.
4
6
8
–
4
6
9
,
d
o
i
:
1
0
.
1
1
4
5
/
1
0
0
8
9
9
2
.
1
0
0
9
0
7
4
.
[1
5
]
A
.
-
Al
-
M
a
m
u
n
a
n
d
S
.
A
k
h
t
e
r
,
"
S
o
c
i
a
l
med
i
a
b
u
l
l
y
i
n
g
d
e
t
e
c
t
i
o
n
u
si
n
g
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
o
n
B
a
n
g
l
a
t
e
x
t
,
"
2
0
1
8
1
0
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
El
e
c
t
ri
c
a
l
a
n
d
C
o
m
p
u
t
e
r E
n
g
i
n
e
e
r
i
n
g
(
I
C
EC
E)
,
2
0
1
8
,
d
o
i
:
1
0
.
1
1
0
9
/
i
c
e
c
e
.
2
0
1
8
.
8
6
3
6
7
9
7
.
[1
6
]
R
.
R
.
D
a
l
v
i
,
S
.
B
.
C
h
a
v
a
n
,
a
n
d
A
.
H
a
l
b
e
,
"
D
e
t
e
c
t
i
n
g
A
Tw
i
t
t
e
r
C
y
b
e
r
b
u
l
l
y
i
n
g
U
si
n
g
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
,
"
2
0
2
0
4
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
I
n
t
e
l
l
i
g
e
n
t
C
o
m
p
u
t
i
n
g
a
n
d
C
o
n
t
r
o
l
S
y
st
e
m
s
(
I
C
I
C
C
S
)
,
2
0
2
0
,
p
p
.
2
9
7
-
3
0
1
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
I
C
C
S
4
8
2
6
5
.
2
0
2
0
.
9
1
2
0
8
9
3
.
[1
7
]
V
.
B
a
n
e
r
j
e
e
,
J.
T
e
l
a
v
a
n
e
,
P
.
G
a
i
k
w
a
d
,
a
n
d
P
.
V
a
r
t
a
k
,
"
D
e
t
e
c
t
i
o
n
o
f
C
y
b
e
r
b
u
l
l
y
i
n
g
U
si
n
g
D
e
e
p
N
e
u
r
a
l
N
e
t
w
o
r
k
,
"
2
0
1
9
5
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
Ad
v
a
n
c
e
d
C
o
m
p
u
t
i
n
g
&
C
o
m
m
u
n
i
c
a
t
i
o
n
S
y
s
t
e
m
s
(
I
C
AC
C
S
)
,
2
0
1
9
,
p
p
.
6
0
4
-
6
0
7
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
A
C
C
S
.
2
0
1
9
.
8
7
2
8
3
7
8
.
[1
8
]
A
.
A
l
i
a
n
d
A
.
M
.
S
y
e
d
,
“
C
y
b
e
r
b
u
l
l
y
i
n
g
d
e
t
e
c
t
i
o
n
u
s
i
n
g
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
,
”
Pa
k
i
s
t
a
n
J
o
u
r
n
a
l
o
f
En
g
i
n
e
e
r
i
n
g
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
3
,
n
o
.
2
,
p
p
.
4
5
–
5
0
,
2
0
2
0
,
d
o
i
:
1
0
.
5
1
8
4
6
/
v
o
l
3
i
ss
2
p
p
4
5
-
5
0
.
[1
9
]
P
.
C
h
a
k
r
a
b
o
r
t
y
a
n
d
M
.
H
.
S
e
d
d
i
q
u
i
,
"
Th
r
e
a
t
a
n
d
A
b
u
s
i
v
e
La
n
g
u
a
g
e
D
e
t
e
c
t
i
o
n
o
n
S
o
c
i
a
l
M
e
d
i
a
i
n
B
e
n
g
a
l
i
La
n
g
u
a
g
e
,
"
2
0
1
9
1
st
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
Ad
v
a
n
c
e
s
i
n
S
c
i
e
n
c
e
,
E
n
g
i
n
e
e
ri
n
g
a
n
d
R
o
b
o
t
i
c
s
T
e
c
h
n
o
l
o
g
y
(
I
C
A
S
ERT)
,
2
0
1
9
,
p
p
.
1
-
6
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
A
S
ER
T.
2
0
1
9
.
8
9
3
4
6
0
9
.
[2
0
]
S
.
A
h
a
mm
e
d
,
M
.
R
a
h
ma
n
,
M
.
H
.
N
i
l
o
y
,
a
n
d
S
.
M
.
M
.
H
.
C
h
o
w
d
h
u
r
y
,
"
I
mp
l
e
me
n
t
a
t
i
o
n
o
f
M
a
c
h
i
n
e
Le
a
r
n
i
n
g
t
o
D
e
t
e
c
t
H
a
t
e
S
p
e
e
c
h
i
n
B
a
n
g
l
a
L
a
n
g
u
a
g
e
,
"
2
0
1
9
8
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
S
y
st
e
m
M
o
d
e
l
i
n
g
a
n
d
A
d
v
a
n
c
e
m
e
n
t
i
n
R
e
se
a
rc
h
T
re
n
d
s
(
S
MA
RT)
,
2
0
1
9
,
p
p
.
3
1
7
-
3
2
0
,
d
o
i
:
1
0
.
1
1
0
9
/
S
M
A
R
T
4
6
8
6
6
.
2
0
1
9
.
9
1
1
7
2
1
4
.
[2
1
]
N
.
I
.
Tr
i
p
t
o
a
n
d
M
.
E
.
A
l
i
,
"
D
e
t
e
c
t
i
n
g
M
u
l
t
i
l
a
b
e
l
S
e
n
t
i
me
n
t
a
n
d
E
mo
t
i
o
n
s
f
r
o
m
B
a
n
g
l
a
Y
o
u
T
u
b
e
C
o
mm
e
n
t
s,"
2
0
1
8
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
B
a
n
g
l
a
S
p
e
e
c
h
a
n
d
L
a
n
g
u
a
g
e
Pr
o
c
e
ssi
n
g
(
I
C
BS
L
P)
,
2
0
1
8
,
p
p
.
1
-
6
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
B
S
LP.2
0
1
8
.
8
5
5
4
8
7
5
.
[2
2
]
A
.
H
a
ssa
n
,
M
.
R
.
A
m
i
n
,
A
.
K
.
A
.
A
z
a
d
,
a
n
d
N
.
M
o
h
a
mm
e
d
,
"
S
e
n
t
i
m
e
n
t
a
n
a
l
y
si
s o
n
b
a
n
g
l
a
a
n
d
r
o
m
a
n
i
z
e
d
b
a
n
g
l
a
t
e
x
t
u
si
n
g
d
e
e
p
r
e
c
u
r
r
e
n
t
mo
d
e
l
s,"
2
0
1
6
I
n
t
e
r
n
a
t
i
o
n
a
l
W
o
r
k
sh
o
p
o
n
C
o
m
p
u
t
a
t
i
o
n
a
l
I
n
t
e
l
l
i
g
e
n
c
e
(
I
WC
I
)
,
2
0
1
6
,
p
p
.
5
1
-
5
6
,
d
o
i
:
1
0
.
1
1
0
9
/
I
W
C
I
.
2
0
1
6
.
7
8
6
0
3
3
8
.
[2
3
]
L.
Ji
n
g
,
H
.
H
u
a
n
g
,
a
n
d
H
.
S
h
i
,
"
I
mp
r
o
v
e
d
f
e
a
t
u
r
e
se
l
e
c
t
i
o
n
a
p
p
r
o
a
c
h
TFI
D
F
i
n
t
e
x
t
m
i
n
i
n
g
,
"
Pr
o
c
e
e
d
i
n
g
s
.
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
M
a
c
h
i
n
e
L
e
a
rn
i
n
g
a
n
d
C
y
b
e
r
n
e
t
i
c
s
.
d
o
i
:
1
0
.
1
1
0
9
/
i
c
m
l
c
.
2
0
0
2
.
1
1
7
4
5
2
2
.
[2
4
]
S
.
X
u
,
Y
.
Li
,
a
n
d
Z
.
W
a
n
g
,
"
B
a
y
e
si
a
n
M
u
l
t
i
n
o
m
i
a
l
N
a
ï
v
e
B
a
y
e
s
C
l
a
ss
i
f
i
e
r
t
o
Te
x
t
C
l
a
ssi
f
i
c
a
t
i
o
n
,
"
L
e
c
t
u
re
N
o
t
e
s
i
n
El
e
c
t
ri
c
a
l
En
g
i
n
e
e
ri
n
g
,
2
0
1
7
,
p
p
.
3
4
7
-
3
5
2
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
981
-
10
-
5
0
4
1
-
1
_
5
7
.
[2
5
]
Q
.
W
u
a
n
d
D
.
-
X
.
Zh
o
u
,
“
A
n
a
l
y
si
s
o
f
su
p
p
o
r
t
v
e
c
t
o
r
ma
c
h
i
n
e
c
l
a
ss
i
f
i
c
a
t
i
o
n
,
”
Jo
u
r
n
a
l
o
f
C
o
m
p
u
t
a
t
i
o
n
a
l
An
a
l
y
s
i
s
&
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
8
,
n
o
.
2
,
2
0
0
6
.
[2
6
]
R
.
E
.
W
r
i
g
h
t
,
“
Lo
g
i
s
t
i
c
r
e
g
r
e
ss
i
o
n
,
”
A
m
e
ri
c
a
n
Ps
y
c
h
o
l
o
g
i
c
a
l
Ass
o
c
i
a
t
i
o
n
,
p
p
.
2
1
7
-
2
4
4
,
1
9
9
5
.
[
2
7
]
T
.
C
h
e
n
a
n
d
C
.
G
u
e
s
t
r
i
n
,
“
X
g
b
o
o
s
t
:
A
s
c
a
l
a
b
l
e
t
r
e
e
b
o
o
s
t
i
n
g
s
y
s
t
e
m
,
”
P
r
o
c
e
e
d
i
n
g
s
o
f
t
h
e
2
2
n
d
a
c
m
s
i
g
k
d
d
i
n
t
e
r
n
a
t
i
o
n
a
l
c
o
n
f
e
r
e
n
c
e
o
n
k
n
o
w
l
e
d
g
e
d
i
s
c
o
v
e
r
y
a
n
d
d
a
t
a
m
i
n
i
n
g
,
2
0
1
6
,
p
p
.
7
8
5
–
7
9
4
,
d
o
i
:
1
0
.
1
1
4
5
/
2
9
3
9
6
7
2
.
2
9
3
9
7
8
5
.
[2
8
]
B
.
Ju
b
a
a
n
d
H
.
S
.
L
e
,
“
P
r
e
c
i
s
i
o
n
-
r
e
c
a
l
l
v
e
r
s
u
s
a
c
c
u
r
a
c
y
a
n
d
t
h
e
r
o
l
e
o
f
l
a
r
g
e
d
a
t
a
set
s,”
Pr
o
c
.
C
o
n
f
.
AA
AI
Art
i
f
.
I
n
t
e
l
l
.
,
v
o
l
.
3
3
,
n
o
.
0
1
,
2
0
1
9
,
p
p
.
4
0
3
9
–
4
0
4
8
,
d
o
i
:
1
0
.
1
6
0
9
/
a
a
a
i
.
v
3
3
i
0
1
.
3
3
0
1
4
0
3
9
.
[2
9
]
A
.
P
.
B
r
a
d
l
e
y
,
“
T
h
e
u
se
o
f
t
h
e
a
r
e
a
u
n
d
e
r
t
h
e
R
O
C
c
u
r
v
e
i
n
t
h
e
e
v
a
l
u
a
t
i
o
n
o
f
mac
h
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
ms,
”
Pa
t
t
e
r
n
R
e
c
o
g
n
i
t
.
,
v
o
l
.
3
0
,
n
o
.
7
,
p
p
.
1
1
4
5
–
1
1
5
9
,
1
9
9
7
,
d
o
i
:
1
0
.
1
0
1
6
/
S
0
0
3
1
-
3
2
0
3
(
9
6
)
0
0
1
4
2
-
2
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
N
a
tu
r
a
l la
n
g
u
a
g
e
p
r
o
ce
s
s
in
g
a
n
d
ma
c
h
in
e
lea
r
n
in
g
b
a
s
ed
c
yb
erb
u
llyin
g
d
etec
tio
n
fo
r
… (
Md
.
To
fa
el
A
h
med
)
97
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Md
.
T
o
fa
e
l
Ah
m
e
d
As
so
c
iate
P
ro
fe
ss
o
r
o
f
ICT,
Co
m
il
la
Un
iv
e
rsity
(Ba
n
g
lad
e
sh
).
Re
se
a
rc
h
a
re
a
in
c
lu
d
e
s
c
y
b
e
rb
u
ll
i
n
g
d
e
tec
ti
o
n
,
d
a
ta m
in
in
g
,
tex
t
a
n
a
ly
ti
c
,
b
ig
d
a
ta,
o
n
li
n
e
s
o
c
ial
n
e
two
rk
a
n
a
ly
sis
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
to
fa
e
l@c
o
u
.
a
c
.
b
d
Ma
q
sud
u
r
Ra
h
m
a
n
Lec
tu
re
r
a
t
De
p
a
rtme
n
t
o
f
IC
T,
C
o
m
il
la
Un
iv
e
rsity
(Ba
n
g
lad
e
sh
).
Tea
c
h
i
n
g
a
re
a
in
c
l
u
d
e
s
c
o
m
p
u
ter p
ro
g
ra
m
m
in
g
,
c
o
m
p
u
ter
n
e
two
r
k
s,
d
a
ta
c
o
m
m
u
n
ica
ti
o
n
&
m
o
b
il
e
c
o
m
m
u
n
ica
ti
o
n
.
re
se
a
rc
h
in
ter
e
sts
li
e
a
t
d
ig
it
a
l
c
o
m
m
u
n
ica
ti
o
n
,
Io
T
a
n
d
n
e
t
wo
rk
se
c
u
rit
y
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
m
rra
jo
n
@c
o
u
.
a
c
.
b
d
S
h
a
f
a
y
e
t
Nur
Tea
c
h
in
g
As
sista
n
t
a
t
P
o
rt
Cit
y
In
ter
n
a
ti
o
n
a
l
Un
iv
e
rsity
(Ba
n
g
lad
e
sh
).
Re
se
a
rc
h
in
tere
st
i
n
c
lu
d
e
s
m
a
c
h
i
n
e
lea
rn
i
n
g
,
d
e
e
p
l
e
a
rn
in
g
a
p
p
r
o
a
c
h
e
s,
n
a
tu
ra
l
lan
g
u
a
g
e
p
r
o
c
e
ss
in
g
,
p
e
rfo
rm
a
n
c
e
a
n
a
ly
sis
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
sh
a
fa
y
e
tn
u
r.
a
b
ir@g
m
a
il
.
c
o
m
Abu
Za
fo
r
Mu
h
a
m
m
a
d
To
u
h
i
d
u
l
Is
la
m
P
ro
fe
ss
o
r
o
f
El
e
c
tri
c
a
l
a
n
d
El
e
c
tro
n
ics
En
g
i
n
e
e
rin
g
,
Ra
jsh
a
h
i
Un
iv
e
rsit
y
,
Ba
n
g
lad
e
sh
.
Co
m
p
le
ted
P
h
D
fr
o
m
Ja
p
a
n
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
t
o
u
h
id
.
e
e
e
@ru
.
a
c
.
b
d
Dip
a
n
k
a
r
Da
s
P
ro
fe
ss
o
r
o
f
De
p
a
rtme
n
t
o
f
I
n
fo
rm
a
ti
o
n
a
n
d
Co
m
m
u
n
ica
ti
o
n
En
g
i
n
e
e
rin
g
,
Un
iv
e
rsity
o
f
Ra
jsh
a
h
i,
Ba
n
g
lad
e
sh
.
Co
m
p
lete
d
P
h
D
f
ro
m
Ja
p
a
n
.
Re
se
a
rc
h
in
tere
st
i
n
c
lu
d
e
s
h
u
m
a
n
r
o
b
o
t
in
te
ra
c
ti
o
n
,
c
o
m
p
u
ter
v
isio
n
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
d
ip
a
n
k
a
r@ru
.
a
c
.
b
d
Evaluation Warning : The document was created with Spire.PDF for Python.