T
E
L
KO
M
NIK
A
,
V
ol
.
17
,
No.
5,
O
c
tob
er
20
1
9,
p
p.
2
66
7
~
2674
IS
S
N: 1
69
3
-
6
93
0
,
accr
ed
ited
F
irst
Gr
ad
e b
y K
em
en
r
istekdikti,
Decr
ee
No: 2
1/E/
K
P
T
/20
18
DOI:
10.12928/TE
LK
OM
N
IK
A
.v
1
7
i
5
.
12646
◼
26
67
Rec
ei
v
ed
J
an
u
ary
5
, 2
01
9
;
Rev
i
s
ed
Ma
r
c
h 1
7
, 2
01
9
; A
c
c
ep
ted
A
pri
l
2
2
, 2
01
9
Th
e
cla
ssific
ati
on
o
f t
he m
od
ern ara
bic
po
e
tr
y
usin
g m
ac
hin
e l
earnin
g
M
u
n
ef
A
b
d
u
lla
h
A
h
m
ed*
1
,
Rae
d
A
b
d
u
l
ka
r
ee
m H
as
a
n
2
,
A
h
m
ed Hu
ss
ein
A
li
3
,
M
o
staf
a
A
b
d
u
lgh
afo
o
r
M
o
h
a
mm
ed
4
1
Fa
c
u
l
ty
o
f
A
l
Ha
w
i
j
a
T
e
c
h
n
i
c
a
l
i
n
s
ti
t
u
te
,
Nor
th
e
r
n
T
e
c
h
n
i
c
a
l
Uni
v
e
rs
i
ty
,
M
o
s
e
l
,
4
1
0
0
2
,
Ir
a
q
2
Fa
c
u
l
ty
o
f
A
l
-
d
o
u
r
T
e
c
h
n
i
c
a
l
i
n
s
ti
t
u
te
,
N
o
rth
e
r
n
T
e
c
h
n
i
c
a
l
U
n
i
v
e
rs
i
ty
,
M
o
s
e
l
,
4
1
0
0
2
,
Ir
a
q
3
AL
Sa
l
a
m
Un
i
v
e
rs
i
ty
Co
l
l
e
g
e
Com
p
u
t
e
r Sc
i
e
n
c
e
De
p
a
rt
m
e
n
t
Ba
g
h
d
a
d
,
I
ra
q
4
G
re
a
t
I
m
a
m
U
n
i
v
e
rs
i
ty
Co
l
l
e
g
e
,
Ba
g
h
d
a
d
,
1
0
0
5
3
,
Ira
q
4
Fa
c
u
l
ty
o
f
Au
to
m
a
ti
c
Co
n
tr
o
l
a
n
d
Co
m
p
u
t
e
rs
,
Uni
v
e
rs
i
ty
Po
l
y
te
c
h
n
i
c
o
f
Bu
c
h
a
re
s
t,
0
6
0
0
4
2
,
Rom
a
n
i
a
*C
o
rre
s
p
o
n
d
i
n
g
a
u
th
o
r,
e
-
m
a
i
l
:
d
rm
u
n
e
f6
9
@gm
a
i
l
.
c
o
m
1
,
ra
e
d
.i
s
c
.s
a
@
g
m
a
i
l
.
c
o
m
2
m
s
c
.a
h
m
e
d
.
h
.a
l
i
@g
m
a
i
l
.c
o
m
3
,
a
l
q
a
i
s
y
8
6
@g
m
a
i
l
.
c
o
m
4
Ab
strac
t
In
re
c
e
n
t
y
e
a
r
s
,
wor
k
i
n
g
o
n
t
e
x
t
c
l
a
s
s
i
fi
c
a
ti
o
n
a
n
d
a
n
a
l
y
s
i
s
o
f
Ara
b
i
c
t
e
x
t
s
u
s
i
n
g
m
a
c
h
i
n
e
l
e
a
rn
i
n
g
h
a
s
s
e
e
n
s
o
m
e
p
ro
g
re
s
s
,
b
u
t
m
o
s
t
o
f
th
i
s
re
s
e
a
r
c
h
h
a
s
n
o
t
fo
c
u
s
e
d
o
n
Ara
b
i
c
p
o
e
tr
y
.
B
e
c
a
u
s
e
o
f
s
o
m
e
d
i
ff
i
c
u
l
ti
e
s
i
n
th
e
a
n
a
l
y
s
i
s
o
f
A
ra
b
i
c
p
o
e
try
,
i
t
wa
s
re
q
u
i
re
d
t
h
e
u
s
e
o
f
s
t
a
n
d
a
r
d
Ara
b
i
c
l
a
n
g
u
a
g
e
o
n
wh
i
c
h
“Al
Aru
d
”
,
th
e
s
c
i
e
n
c
e
o
f
s
tu
d
y
i
n
g
p
o
e
try
i
s
b
a
s
e
d
.
Th
i
s
p
a
p
e
r
p
re
s
e
n
ts
a
n
a
p
p
ro
a
c
h
th
a
t
u
s
e
s
m
a
c
h
i
n
e
l
e
a
rn
i
n
g
f
o
r
th
e
c
l
a
s
s
i
fi
c
a
t
i
o
n
o
f
m
o
d
e
r
n
Ar
a
b
i
c
p
o
e
try
i
n
to
f
o
u
r
ty
p
e
s
:
l
o
v
e
p
o
e
m
s
,
I
s
l
a
m
i
c
p
o
e
m
s
,
s
o
c
i
a
l
p
o
e
m
s
,
a
n
d
p
o
l
i
t
i
c
a
l
p
o
e
m
s
.
Ea
c
h
o
f
th
e
s
e
s
p
e
c
i
e
s
u
s
u
a
l
l
y
h
a
s
fe
a
t
u
re
s
th
a
t
i
n
d
i
c
a
t
e
th
e
c
l
a
s
s
o
f
th
e
p
o
e
m
.
Des
p
i
te
th
e
c
h
a
l
l
e
n
g
e
s
g
e
n
e
ra
t
e
d
b
y
th
e
d
i
ff
i
c
u
l
ty
o
f
th
e
ru
l
e
s
o
f
th
e
Ara
b
i
c
l
a
n
g
u
a
g
e
o
n
wh
i
c
h
th
i
s
c
l
a
s
s
i
fi
c
a
ti
o
n
d
e
p
e
n
d
s
,
w
e
p
ro
p
o
s
e
d
a
n
e
w
a
u
to
m
a
ti
c
way
o
f
m
o
d
e
r
n
Ara
b
i
c
p
o
e
m
s
c
l
a
s
s
i
fi
c
a
ti
o
n
t
o
s
o
l
v
e
th
e
s
e
i
s
s
u
e
s
.
Th
e
re
c
o
m
m
e
n
d
e
d
m
e
th
o
d
i
s
s
u
i
t
a
b
l
e
fo
r
th
e
a
b
o
v
e
-
m
e
n
t
i
o
n
e
d
c
l
a
s
s
e
s
o
f
p
o
e
m
s
.
T
h
i
s
s
tu
d
y
u
s
e
d
Naï
v
e
Ba
y
e
s
,
Su
p
p
o
rt
V
e
c
t
o
r
M
a
c
h
i
n
e
s
,
a
n
d
L
i
n
e
a
r
Su
p
p
o
rt
Ve
c
to
r
fo
r
t
h
e
c
l
a
s
s
i
f
i
c
a
ti
o
n
p
ro
c
e
s
s
e
s
.
Dat
a
p
re
p
r
o
c
e
s
s
i
n
g
was
a
n
i
m
p
o
rta
n
t
s
te
p
o
f
th
e
a
p
p
r
o
a
c
h
i
n
th
i
s
p
a
p
e
r,
a
s
i
t
i
n
c
re
a
s
e
d
th
e
a
c
c
u
r
a
c
y
o
f
t
h
e
c
l
a
s
s
i
fi
c
a
t
i
o
n
.
Key
w
ords
:
c
l
a
s
s
i
f
i
c
a
ti
o
n
o
f
a
r
a
b
i
c
p
o
e
m
s
,
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
th
m
s
,
m
o
d
e
rn
a
r
a
b
i
c
p
o
e
m
s
Copy
righ
t
©
2
0
1
9
Uni
v
e
rsi
t
a
s
Ahm
a
d
D
a
hl
a
n.
All
rig
ht
s
r
e
s
e
rve
d
.
1.
Int
r
o
d
u
ctio
n
Des
pi
t
e
the
n
um
be
r
of
a
pp
r
oa
c
h
es
on
the
a
uto
m
ati
c
c
l
as
s
i
f
i
c
ati
on
of
the
E
ng
l
i
s
h
l
an
gu
a
ge
a
nd
ot
he
r
l
an
g
u
ag
es
,
th
e
A
r
a
bi
c
l
an
g
ua
g
e
s
ti
l
l
ne
e
ds
a
l
o
t
of
r
es
ea
r
c
h,
es
pe
c
i
a
l
l
y
r
el
ate
d
to
A
r
a
bi
c
po
etr
y
.
T
hi
s
i
s
du
e
to
th
e
nu
m
be
r
of
d
ete
r
m
i
na
nts
i
n
th
e
l
an
gu
ag
e,
i
nc
l
ud
i
ng
i
ts
di
f
f
i
c
ul
t
y
an
d
th
e
n
ee
d
to
m
as
ter
the
r
ul
es
of
the
l
an
g
u
ag
e
whe
n
s
tud
y
i
ng
po
etr
y
.
T
he
r
e
i
s
al
s
o
a
ne
ed
f
or
a
f
ul
l
u
nd
ers
ta
nd
i
ng
of
the
t
he
or
y
of
“
A
l
A
r
u
d
”
,
whi
c
h
s
pe
c
i
a
l
i
z
es
i
n
th
e
s
tud
y
of
A
r
ab
i
c
po
etr
y
[
1
]
w
he
t
he
r
as
a
r
e
gu
l
ar
tex
t
or
p
oe
m
,
f
oc
us
e
d
on
t
he
t
op
i
c
or
on
t
he
ef
f
ec
ts
[
2
]
.
F
ew
s
tud
i
es
ha
v
e
us
ed
s
en
t
i
m
e
n
t
an
a
l
y
s
i
s
to
c
l
as
s
i
f
y
A
r
ab
i
c
tex
ts
[
3
]
.
In
thi
s
s
tud
y
,
we
us
ed
Naï
v
e
B
a
y
es
(
N
B
)
,
S
up
po
r
t
V
ec
t
or
Ma
c
h
i
n
es
(
S
V
M),
an
d
L
i
ne
ar
S
up
p
ort
V
ec
tor
c
l
as
s
i
f
i
c
ati
o
n
(
S
V
C)
f
or the
c
l
as
s
i
f
i
c
ati
on
t
as
k
.
T
he
ne
x
t
s
ec
ti
o
n
of
thi
s
pa
pe
r
c
o
v
ers
a
r
ev
i
e
w
of
the
r
e
l
at
ed
wor
k
,
f
ol
l
o
wed
b
y
the
i
ntrod
uc
ti
o
n
of
the
f
ou
r
c
ate
go
r
i
es
of
m
od
ern
A
r
ab
i
c
po
e
tr
y
.
A
f
ter
tha
t
,
the
d
ata
s
et
of
the
w
ork
i
s
pres
en
ted
,
f
ol
l
o
w
ed
b
y
th
e
da
ta
pre
proc
es
s
i
ng
s
tep
whi
c
h
h
as
a
d
i
r
ec
t
ef
f
ec
t
on
the
ac
c
urac
y
of
the
c
l
as
s
i
f
i
c
ati
on
pr
oc
es
s
.
T
he
s
i
x
th
an
d
s
e
v
e
nth
s
ec
ti
on
s
f
oc
us
on
f
ea
ture
s
el
ec
ti
on
an
d
the
m
ac
hi
ne
l
ea
r
n
i
n
g
a
l
go
r
i
thm
s
us
ed
.
T
he
s
e
s
ec
ti
on
s
are
f
ol
l
o
w
ed
b
y
tho
s
e
th
at
di
s
c
us
s
th
e m
eth
od
o
l
o
g
y
,
r
es
ul
ts
, a
nd
c
o
nc
l
us
i
on
s
f
r
om
th
e s
tud
y
.
2.
S
t
ate
of
t
h
e
A
RT
S
e
v
era
l
m
eth
od
s
h
av
e
b
e
en
us
e
d
i
n
the
E
ng
l
i
s
h
l
a
ng
ua
ge
f
or
t
he
c
l
as
s
i
f
i
c
ati
on
of
em
oti
on
s
.
S
om
e
of
the
s
e
s
t
ud
i
es
de
p
en
d
ed
on
k
e
y
wor
ds
s
po
tt
i
ng
or
u
na
m
bi
gu
ou
s
w
ords
l
i
k
e
“
ha
pp
y
”
an
d
“
s
ad
”
[
4
]
.
T
he
l
ex
i
c
al
af
f
i
ni
t
y
f
r
om
the
ef
f
ec
ti
v
e
r
es
e
arc
h
i
n
th
i
s
f
i
e
l
d
de
pe
n
de
d
o
n
the
em
oti
o
n
of
the
arb
i
tr
ar
y
term
or
wor
ds
.
In
ge
ne
r
a
l
,
th
i
s
m
eth
od
i
s
be
t
ter
th
a
n
the
k
e
y
wor
d
Evaluation Warning : The document was created with Spire.PDF for Python.
◼
IS
S
N: 16
93
-
6
93
0
T
E
L
KO
M
NIK
A
V
ol
.
17
,
No
.
5,
O
c
tob
er 20
19
:
26
6
7
-
26
74
2668
s
po
tti
ng
m
eth
od
as
i
t
c
an
n
ot
be
us
e
d
as
an
i
nd
e
pe
n
de
nt
m
od
el
[
5
]
.
T
he
r
e
are
oth
er
m
eth
od
s
whi
c
h
r
e
l
y
o
n
a
de
e
p
u
nd
ers
tan
di
ng
of
th
e
l
an
g
ua
ge
an
d
s
em
an
ti
c
s
[
5
]
.
Rel
i
a
nc
e
on
ps
y
c
ho
l
og
i
c
al
t
he
or
y
i
n
de
t
erm
i
ni
ng
de
s
i
r
es
,
go
a
l
s
,
an
d
ne
e
ds
was
on
e
of
the
m
od
e
l
s
us
ed
i
n
the
c
l
as
s
i
f
i
c
ati
on
[
6
]
.
T
he
m
a
c
hi
ne
l
ea
r
n
i
ng
t
ec
hn
i
qu
es
us
ed
i
n
th
e
c
l
as
s
i
f
i
c
ati
on
of
c
l
as
s
i
c
al
A
r
ab
i
c
po
etr
y
de
p
en
d
ed
on
th
e
em
oti
on
[
7
]
.
T
hi
s
wor
k
c
l
as
s
i
f
i
ed
the
A
r
a
bi
c
po
etr
y
i
n
to
F
ak
hr,
Reth
a,
G
ha
z
a
l
,
a
n
d
Hei
j
a.
T
he
p
ol
y
n
om
i
al
ne
t
w
ork
s
w
ere
us
e
d
i
n
t
he
A
r
a
bi
c
tex
t
c
l
as
s
i
f
i
c
ati
on
[
8
]
.
S
e
v
era
l
c
l
as
s
i
f
i
c
ati
on
al
g
orit
hm
s
ha
v
e
be
e
n
us
e
d
i
n
the
c
l
as
s
i
f
i
c
ati
o
n
of
A
r
ab
i
c
tex
t,
s
uc
h
as
S
V
M
[
8
,
9
]
,
t
h
e
N
B
[
10
]
,
K
-
Near
es
t
N
ei
g
hb
or
(
K
NN)
[
11
]
,
A
r
ti
f
i
c
i
al
N
eu
r
al
Ne
t
w
ork
(
A
NN)
[
12
]
, a
nd
t
he
R
oc
c
hi
o f
ee
db
ac
k
al
go
r
i
thm
[
13
]
.
3.
Cat
ego
r
ies
of
M
o
d
er
n
A
r
abic
P
o
et
r
y
T
he
m
od
ern A
r
a
bi
c
p
oe
tr
y
i
n g
e
ne
r
a
l
c
on
s
i
s
ts
of
th
e
f
ol
l
o
wi
ng
t
y
p
es
[
14
]
:
−
Lo
v
e
p
oe
m
s
:
It
i
s
a
po
e
ti
c
art
us
ed
to
ex
pres
s
the
f
ee
l
i
ng
s
be
t
ween
l
o
v
ers
.
T
he
p
oe
t
d
eri
v
es
the
m
ea
ni
n
gs
of
hi
s
r
e
l
at
i
on
s
h
i
p
wi
t
h
th
e
s
ub
j
ec
t,
hi
s
ou
t
l
oo
k
,
the
i
n
f
l
ue
nc
e
of
the
e
nv
i
r
on
m
en
t, a
nd
t
he
r
e
al
i
t
y
of
th
os
e f
ee
l
i
n
gs
.
−
Is
l
am
i
c
(
r
el
i
g
i
ou
s
)
po
em
s
:
T
he
po
ets
be
ne
f
i
te
d
f
r
o
m
t
he
s
tori
es
c
on
ta
i
n
ed
i
n
t
he
Hol
y
Q
uran;
s
o,
th
e
y
to
ok
the
prec
ep
t
s
,
r
ul
i
ng
s
,
an
d
s
em
an
ti
c
s
an
d
em
pl
o
y
e
d
t
he
m
i
n
t
he
i
r
po
etr
y
,
tr
ea
ti
ng
c
om
m
un
i
t
y
i
s
s
ue
s
an
d
prob
l
em
s
th
at
s
pread
i
n t
h
e
i
r
c
ou
n
tr
y
at
th
e t
i
m
e.
−
S
oc
i
a
l
po
em
s
:
S
oc
i
al
po
em
s
ai
m
to
r
ep
ai
r
ba
d
s
oc
i
a
l
c
on
di
t
i
o
ns
b
y
d
i
ag
no
s
i
ng
th
e
probl
em
,
i
de
nti
f
y
i
ng
i
ts
c
au
s
e,
an
d
de
s
c
r
i
b
i
ng
i
ts
r
es
ol
uti
on
.
T
he
po
ets
r
es
ort
to
the
m
eth
od
of
en
c
ou
r
a
ge
m
en
t
an
d
m
oti
v
ati
o
n
w
he
n
the
y
w
a
nt
the
i
r
pe
op
l
e
to
c
on
tr
i
bu
te
to
th
e
prom
oti
on
an
d
pro
gres
s
an
d
av
oi
d
the
pe
s
ts
a
nd
c
on
di
t
i
o
ns
tha
t
u
nd
erm
i
ne
the
f
ou
nd
at
i
on
s
of
i
ts
r
en
a
i
s
s
an
c
e.
−
P
ol
i
t
i
c
al
po
em
s
:
T
hi
s
t
y
p
e
of
po
etr
y
ex
pres
s
es
c
ertai
n
po
l
i
t
i
c
al
orie
n
tat
i
on
s
an
d
t
he
p
ers
on
a
l
v
i
e
w
s
of
po
ets
w
h
i
l
e
pre
s
erv
i
n
g
the
wa
y
po
e
tr
y
i
s
w
r
i
tt
en
,
t
he
v
al
ue
s
of
l
i
terar
y
an
d
arti
s
ti
c
p
oe
tr
y
.
4.
T
h
e Da
t
as
et
T
he
A
r
ab
i
c
l
a
ng
u
ag
e
r
es
e
arc
h
us
i
ng
N
atu
r
a
l
La
ng
u
a
ge
P
r
oc
es
s
i
n
g
(
NL
P
)
i
s
di
f
f
erent
f
r
o
m
the
E
ng
l
i
s
h
l
a
ng
ua
g
e
i
n
term
s
of
the
nu
m
be
r
an
d
s
i
z
e
of
th
e
da
tas
ets
us
ed
.
Due
to
the
l
i
m
i
ted
nu
m
be
r
of
fr
ee
av
a
i
l
a
bl
e
da
t
as
ets
i
n
th
e
A
r
ab
i
c
l
an
g
ua
ge
(
w
hi
c
h
i
s
an
ob
s
t
ac
l
e
i
n
the
w
a
y
of
r
es
ea
r
c
he
r
s
)
,
m
os
t
r
es
ea
r
c
he
r
s
r
el
y
on
a
c
ol
l
ec
ti
on
of
da
tas
ets
tak
en
fr
o
m
m
ag
az
i
ne
s
,
ne
w
s
s
tat
i
o
ns
,
an
d
w
e
bs
i
tes
.
S
o
m
e
r
es
ea
r
c
he
r
s
de
pe
nd
ed
o
n
S
au
d
i
ne
w
s
pa
pe
r
s
[
11
]
.
In
t
he
A
r
a
bi
c
r
es
ea
r
c
h,
s
e
v
era
l
s
c
ho
o
l
s
of
tho
ug
ht
ha
v
e
c
l
as
s
i
f
i
e
d
the
da
tas
e
ts
i
nto
tr
ai
ni
n
g
a
nd
tes
t
i
n
g
gr
ou
ps
.
In
ou
r
wor
k
,
the
bi
g
pr
ob
l
em
i
s
f
i
nd
i
ng
t
he
d
ata
s
ets
f
or
tun
i
ng
an
d
tes
t
i
ng
b
ec
au
s
e
i
t
i
s
t
he
f
i
r
s
t
wor
k
on
us
i
ng
m
ac
hi
ne
l
ea
r
n
i
ng
f
or
c
l
as
s
i
f
y
i
ng
t
he
m
od
ern
A
r
ab
i
c
po
etr
y
.
W
e
de
pe
nd
ed
on
th
e
w
eb
s
i
te
f
or
da
tas
ets
to
tr
ai
n
a
nd
tes
t
th
e
c
ate
go
r
i
es
of
m
od
ern A
r
ab
i
c
p
oe
tr
y
.
5.
Dat
a
P
r
e
-
P
r
o
ce
ss
ing
T
he
A
r
ab
i
c
l
an
gu
ag
e
i
s
d
i
f
f
i
c
ul
t
bo
t
h
i
n
s
pe
ak
i
ng
an
d
wr
i
ti
n
g.
It
c
on
s
i
s
ts
of
29
l
ett
er
s
(
أ
ب
ت
ث
ج
ح
خ
د
ذ
ر
ز
س
ش
ص
ض
ط
ظ
ع
غ
ف
ق
ك
ل
م
ن
ه
و
ي
)
a
nd
th
e
”
Ham
z
a
”
(
ء
)
whi
c
h
a
r
e
di
v
i
de
d
i
nt
o
two
t
y
p
es
.
T
he
f
i
r
s
t
t
y
p
e
i
s
c
al
l
e
d
l
on
g
v
o
wel
s
,
wh
i
c
h
i
nc
l
ud
es
thre
e
l
ett
ers
(
ا
,
و
,
ي
)
;
the
ot
he
r
i
s
c
al
l
e
d
c
on
s
ta
nt
l
ett
ers
.
In
thi
s
l
an
g
ua
g
e,
t
he
r
e
are
s
ev
era
l
k
i
nd
s
of
di
ac
r
i
ti
c
s
u
s
ed
,
s
uc
h
as
“
s
u
k
oo
n”,
“
da
m
m
ah
”
,
“
K
as
r
a”,
“
F
ath
a”,
“
tan
ween
f
ath
a
”
,
“
tan
ween
k
as
r
a”,
“
tan
we
en
da
m
m
ah
”
,
“
s
ha
dd
e”,
an
d
“
m
ad
”
.
T
he
s
e
s
ho
r
t
v
o
wel
s
gi
v
e
c
orr
ec
t
pron
un
c
i
ati
on
an
d
m
ea
ni
ng
.
T
ab
l
e
1
i
l
l
us
tr
ate
s
the
s
ho
r
t
v
o
w
e
l
s
an
d
pro
nu
nc
i
ati
on
s
t
o
t
he
wor
ds
th
at
ha
v
e
th
e
s
am
e
l
ett
ers
b
ut
di
f
f
erent p
r
on
u
nc
i
at
i
o
n a
nd
m
ea
ni
ng
as
s
ho
wn
i
n
T
ab
l
e
2.
A
r
ab
i
c
w
r
i
ti
ng
s
are
d
i
f
f
erent
f
r
o
m
tho
s
e
us
i
ng
th
e
La
ti
n
al
ph
a
be
t,
du
e
t
o
th
e
di
r
e
c
ti
on
of
w
r
i
ti
n
g
f
r
om
r
i
gh
t
to
l
ef
t.
S
om
e
l
ett
ers
i
n
A
r
ab
i
c
al
s
o
tak
e
s
ev
eral
f
orm
s
de
pe
nd
i
ng
on
the
l
oc
at
i
on
of
th
e
c
ha
r
a
c
ter
on
the
w
ord.
T
he
s
e
f
ea
tures
m
us
t
be
c
on
s
i
de
r
ed
i
n
th
i
s
wor
k
as
s
h
ow
n
i
n
T
ab
l
e 3
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NIK
A
IS
S
N: 1
69
3
-
6
93
0
◼
T
he
c
l
as
s
i
f
i
c
at
i
on
of
t
he
m
o
de
r
n a
r
a
bi
c
p
oe
tr
y
us
i
ng
… (
Mu
ne
f
A
b
du
l
l
ah
A
hm
ed
)
2669
T
ab
l
e 1
.
T
he
D
i
ac
r
i
t
i
c
s
i
n
M
od
ern
A
r
ab
i
c
P
oe
m
The
s
h
o
r
t
v
o
w
e
l
The
S
ign
A
p
p
li
e
d
t
o
t
h
e
let
t
e
r
P
r
o
n
u
n
c
iat
ion
“
s
u
k
o
o
n
”
ْ
ل
-
س
S
-
L
“
d
a
mm
a
h
”
ْ
ل
-
س
S
u
-
Lu
“
K
a
s
r
a
”
ْ
ل
-
س
S
i
-
Li
“
Fatha
”
ْ
ل
-
س
S
a
-
La
“
t
a
n
w
e
e
n
f
a
t
h
a
”
ْ
ل
-
س
S
a
n
-
Lan
“
t
a
n
w
e
e
n
k
a
s
r
a
”
ْ
ل
-
س
S
in
-
L
in
“
t
a
n
w
e
e
n
d
a
mm
a
h
”
ْ
ل
-
س
S
o
n
-
Lon
“
s
h
a
d
d
e
”
ْ
ل
–
س
S
s
–
Ll
M
a
d
~
آ
Aa
T
ab
l
e 2
.
E
x
am
pl
e f
or the
E
f
f
ec
t th
e Di
ac
r
i
ti
c
s
on
th
e A
r
ab
i
c
W
ord
T
ab
l
e 3
.
T
he
E
f
f
ec
t o
f
a
P
os
i
ti
o
ni
n
g o
n
the
f
orm
o
f
a
Le
tte
r
The
w
o
r
d
The
m
e
a
n
ing
م
ل
س
H
e
ll
o
م
ل
س
Ladder
م
ل
س
W
a
s
d
e
li
v
e
r
e
d
م
ل
س
S
a
f
e
t
y
م
ل
س
S
a
v
e
d
The
let
t
e
r
The
A
r
a
b
i
c
w
o
r
d
The
m
e
a
n
ing
ـه
هي
د
ه
Gif
t
ـ
ـه
ــ
م
ا
ه
لا
I
m
p
o
r
t
a
n
t
ه
ــ
ـ
هل
For
h
i
m
ه
هر
ك
A
b
a
ll
T
he
A
r
ab
i
c
l
an
g
ua
g
e
h
as
t
wo
t
y
pe
s
of
ge
nres
,
m
as
c
ul
i
ne
an
d
f
em
i
ni
ne
.
E
ac
h
t
y
p
e
i
n
the
A
r
ab
i
c
l
a
ng
u
ag
e
ha
s
di
f
f
erent
qu
a
l
i
ti
es
an
d
f
ea
t
ures
i
n
A
r
ab
i
c
gram
m
ar.
T
he
r
e
are
thre
e
c
l
as
s
es
i
n
the
A
r
ab
i
c
l
an
g
ua
ge
,
the
f
i
r
s
t
i
s
s
i
ng
u
l
ar,
t
he
s
ec
on
d
i
s
du
a
l
,
an
d
pl
u
r
al
w
hi
c
h
al
s
o
ha
s
t
w
o
t
y
pe
s
(
r
e
gu
l
ar
a
nd
br
ok
en
)
.
T
he
A
r
ab
i
c
l
an
gu
ag
e
c
on
t
ai
ns
m
a
n
y
r
am
i
f
i
c
ati
on
s
i
n
gram
m
a
r
.
It
i
s
a
v
er
y
r
i
c
h
l
an
g
ua
g
e,
an
d
th
i
s
m
ak
es
i
t
di
f
f
i
c
ul
t
a
nd
a
c
h
al
l
en
ge
to
r
ea
c
h
the
r
eq
ui
r
e
d a
c
c
urac
y
i
n
th
e c
l
as
s
i
f
i
c
ati
on
of
m
od
ern A
r
ab
i
c
po
etr
y
.
P
r
e
-
proc
es
s
i
ng
of
da
ta
i
s
an
i
m
po
r
tan
t
th
i
n
g
to
do
when
bu
i
l
d
i
n
g
c
l
as
s
i
f
i
c
at
i
on
s
y
s
t
em
s
us
i
ng
m
a
c
hi
ne
l
an
gu
a
ge
f
or th
e f
ol
l
o
w
i
ng
r
e
as
on
s
:
−
It rem
ov
es
no
i
s
e
f
r
o
m
th
e t
ex
t u
s
ed
i
n t
he
c
l
as
s
i
f
i
c
ati
o
n.
−
It redu
c
es
t
he
t
erm
s
or c
ha
r
ac
teri
s
ti
c
s
on
wh
i
c
h
w
e
ba
s
e o
ur c
l
as
s
i
f
i
c
ati
o
n.
−
It h
e
l
ps
r
ed
uc
i
n
g t
h
e a
m
ou
nt
of
m
e
m
ory
r
eq
u
i
r
ed
f
or t
he
c
l
as
s
i
f
i
c
at
i
on
.
−
It h
e
l
ps
i
nc
r
ea
s
i
n
g t
h
e a
c
c
u
r
ac
y
of
th
e c
l
as
s
i
f
i
c
ati
o
n.
W
e
ap
pl
i
ed
t
he
f
ol
l
o
wi
ng
pr
e
-
proc
es
s
i
ng
on
t
he
d
ata
u
s
ed
i
n
ou
r
wor
k
:
−
T
o
k
en
i
z
at
i
o
n:
W
e
di
v
i
de
d
t
he
da
ta
i
nto
p
arts
an
d
ba
s
ed
on
c
h
arac
teri
s
t
i
c
s
an
d
r
ec
og
n
i
ti
on
of
de
l
i
m
i
ters
l
i
k
e t
he
pu
n
c
tu
ati
o
n o
f
s
pe
c
i
a
l
c
ha
r
ac
ters
an
d
whi
t
e s
pa
c
e.
−
W
e
r
e
m
ov
ed
no
n
-
A
r
ab
i
c
te
r
m
s
, w
ords
, n
um
be
r
s
, p
un
c
tua
ti
on
s
, a
nd
an
y
ot
he
r
s
i
n
g
e.
−
T
he
s
top
w
ords
l
i
k
e
pron
ou
ns
,
pr
ep
os
i
ti
o
ns
,
a
nd
c
on
j
un
c
ti
on
s
w
er
e
a
l
s
o
r
e
m
ov
ed
;
w
e
de
ep
en
e
d t
h
e l
i
s
t a
do
p
ted
b
y
K
ho
j
a
an
d Gar
s
i
d
e
[
15
,
16
]
.
−
S
tem
m
i
ng
:
T
he
m
aj
or
ai
m
of
s
te
m
m
i
ng
i
s
to
de
c
r
ea
s
e
an
i
nf
l
at
ed
d
ata
s
et.
I
n
A
r
a
bi
c
,
m
an
y
wor
ds
c
an
be
c
om
po
s
ed
f
r
om
the
s
a
m
e
s
te
m
.
T
hu
s
,
we
c
an
r
ed
uc
e
the
nu
m
be
r
of
ter
m
s
us
ed
i
n
t
he
da
t
as
et
a
nd
the
c
om
pl
ex
i
t
y
of
tex
t
c
l
a
s
s
i
f
i
c
ati
on
.
T
hi
s
i
s
a
l
s
o
a
s
torag
e
r
eq
ui
r
em
en
t f
or c
l
as
s
i
f
i
c
ati
o
n s
y
s
tem
s
[
1
7
,
18
]
.
6.
F
ea
t
u
r
es
S
el
ec
t
ion
In
m
ac
hi
ne
l
ea
r
n
i
ng
,
c
on
s
tr
uc
ti
ng
or
r
ep
r
es
en
t
i
ng
v
ec
t
ors
of
f
ea
tures
i
s
a
v
er
y
i
m
po
r
tan
t
an
d
c
r
i
t
i
c
al
po
i
nt
an
d
ha
s
a
s
i
gn
i
f
i
c
an
t
i
m
pa
c
t
on
t
he
r
es
ul
ts
of
the
m
ac
hi
n
e
l
ea
r
n
i
ng
al
go
r
i
t
hm
.
E
ac
h o
bj
ec
t s
ho
u
l
d
be
r
e
pres
en
te
d
w
i
th
i
ts
o
w
n
f
ea
tur
es
.
=
1
2
…
.
(
1)
=
1
2
…
(
2)
̈
=
(
)
(
3)
where
D
i
s
a
do
c
um
en
t,
i
s
a
w
ord,
a
nd
i
s
t
he
f
un
c
ti
on
r
e
pres
en
t
i
n
g
t
he
r
el
ati
o
n
b
et
w
e
en
the
d
om
ai
n
of
do
c
um
en
ts
an
d
f
ea
tures
.
m
a
y
be
a
l
i
ne
ar
or
no
n
l
i
ne
ar
e
qu
a
ti
o
n.
T
he
nu
m
be
r
of
Evaluation Warning : The document was created with Spire.PDF for Python.
◼
IS
S
N: 16
93
-
6
93
0
T
E
L
KO
M
NIK
A
V
ol
.
17
,
No
.
5,
O
c
tob
er 20
19
:
26
6
7
-
26
74
2670
c
l
as
s
es
i
s
r
ep
r
es
e
nte
d
b
y
an
d
the
nu
m
be
r
of
f
ea
tures
i
s
r
ep
r
es
e
nte
d
b
y
.
*
i
s
a
f
ea
ture
v
ec
tor
l
en
g
t
h.
W
e
pe
r
f
o
r
m
ed
the
m
utu
al
l
y
de
d
u
c
ted
oc
c
urr
en
c
e
as
f
ol
l
o
w
s
:
=
(
)
r
ep
r
es
en
te
d
th
e
pro
ba
b
i
l
i
t
y
of
oc
c
urr
en
c
e
of
f
ea
ture
i
n
c
a
teg
or
y
or
c
l
as
s
c
.
T
he
r
ef
ore,
the
m
utu
al
l
y
de
du
c
ted
c
ou
n
t f
ea
ture b
ec
am
e a
s
f
ol
l
o
w
s
:
(
)
=
(
)
−
(
)
,
where
≠
,
(
4)
whi
c
h
r
ef
ers
to
t
he
nu
m
be
r
of
ap
pe
ara
nc
es
of
a
n
y
c
ha
r
ac
teri
s
t
i
c
or
f
ea
ture
i
n
a
n
y
c
ate
go
r
y
de
du
c
t
ed
f
r
om
the
nu
m
be
r
of
ap
p
ea
r
a
nc
es
of
the
s
a
m
e
c
ha
r
ac
teri
s
ti
c
i
n
a
l
l
oth
er
c
ate
g
orie
s
.
T
he
f
ea
ture
v
ec
tor
was
us
ed
f
or
bu
i
l
d
i
ng
do
c
um
en
t
on
c
e.
W
he
n
f
ou
nd
an
y
f
ea
ture,
the
B
oo
l
ea
n
f
l
ag
was
us
e
d.
T
he
B
oo
l
ea
n
v
ec
tor
m
od
el
us
ed
i
n
th
i
s
t
y
pe
of
c
l
as
s
i
f
i
c
ati
on
i
s
b
ett
er
tha
n
the
c
o
un
t m
od
el
[
1
9
,
20
]
.
7.
M
ac
h
ine
L
ea
r
n
ing
A
l
g
o
r
it
h
ms
In
o
ur
ap
proac
h,
thre
e
m
ac
hi
ne
l
e
arni
ng
al
go
r
i
t
hm
s
wer
e
s
el
ec
ted
f
or
the
c
l
as
s
i
f
i
c
ati
o
n
of
m
od
ern
A
r
a
bi
c
p
oe
tr
y
.
T
he
s
e
a
l
g
orit
hm
s
ha
v
e
be
en
prov
en
s
uc
c
es
s
f
ul
i
n
the
c
l
as
s
i
f
i
c
ati
on
of
the
E
ng
l
i
s
h
te
x
t.
T
he
f
i
r
s
t a
l
go
r
i
thm
i
s
S
up
p
ort V
ec
tor
Ma
c
hi
ne
s
, t
he
s
ec
o
nd
i
s
N
aïv
e B
a
y
es
, a
nd
the
th
i
r
d
i
s
L
i
ne
ar
S
u
pp
ort
V
ec
tor
Cl
as
s
i
f
i
c
ati
o
n.
T
he
da
tas
ets
c
on
s
i
s
t
of
f
ou
r
gr
ou
ps
(
f
ol
de
r
s
)
:
Is
l
am
i
c
c
on
tai
ns
2
3
f
i
l
es
,
L
ov
e
c
o
nta
i
ns
2
5
f
i
l
es
,
P
o
l
i
t
i
c
c
on
tai
ns
22
f
i
l
es
,
an
d
S
o
c
i
al
c
on
tai
ns
22
f
i
l
es
,
as
i
l
l
us
tr
at
ed
i
n
T
ab
l
e
4
.
C
l
as
s
i
f
i
er
pe
r
f
orm
an
c
e
i
s
ev
al
ua
t
ed
b
y
c
om
pu
ti
ng
i
ts
prec
i
s
i
o
n [
2
1],
r
ec
a
l
l
[
16
],
a
nd
f
-
m
ea
s
ure [22
].
T
ab
l
e 4
.
T
he
D
ata
s
ets
f
or t
he
C
l
as
s
i
f
i
c
ati
o
n
The
f
o
lde
r
n
a
me
N
u
m
b
e
r
o
f
f
il
e
s
N
u
m
b
e
r
o
f
v
e
r
s
e
s
I
s
la
mi
c
23
600
L
o
v
e
25
600
P
o
li
t
i
c
22
500
s
o
c
ia
l
22
550
7.1
.
S
u
p
p
o
r
t
V
ec
t
o
r
M
ac
h
ines
S
V
M
i
s
a
c
om
pu
tat
i
on
al
l
y
k
erne
l
-
ba
s
e
d
al
go
r
i
thm
f
or
r
eg
r
es
s
i
on
an
d
bi
na
r
y
da
t
a
c
l
as
s
i
f
i
c
ati
on
pu
r
po
s
es
[
1
7
,
18
]
.
B
as
e
d
on
th
e
s
tr
uc
tural
r
i
s
k
m
i
ni
m
i
z
ati
on
th
eo
r
y
,
the
S
V
M
ha
s
be
en
prov
en
s
uc
c
es
s
f
ul
i
n
s
ol
v
i
ng
bo
th
l
oc
a
l
m
i
ni
m
u
m
an
d
h
i
gh
di
m
en
s
i
on
a
l
i
t
y
pr
ob
l
em
s
.
It
ha
s
a
b
ett
er
ge
n
eral
i
z
at
i
on
p
erf
or
m
an
c
e
c
o
m
pa
r
ed
to
ot
h
er
ML
m
eth
od
s
s
uc
h
as
A
NNs
[
19
,
20
]
.
S
V
M
h
as
s
o
f
ar
be
e
n
ex
c
e
l
l
en
t
i
n
s
o
l
v
i
n
g
s
e
v
era
l
r
ea
l
-
wor
l
d
da
t
a
m
i
ni
ng
pre
di
c
ti
v
e
prob
l
em
s
l
i
k
e
ti
m
e
s
erie
s
pred
i
c
ti
on
,
tex
t
c
ate
g
ori
z
at
i
on
,
i
m
ag
e
pro
c
es
s
i
ng
,
an
d
pa
t
tern
r
ec
og
ni
t
i
on
[
21
,
22
]
.
Des
pi
t
e
th
e
r
em
ar
k
ab
l
e
ac
hi
e
v
em
en
ts
of
the
S
V
M,
th
ere
are
s
t
i
l
l
c
erta
i
n
dra
wba
c
k
s
tha
t
ne
ed
t
o
be
ad
dres
s
ed
,
s
uc
h
as
pro
bl
em
s
on
the
r
el
ati
on
s
hi
p
o
f
the
s
tat
i
s
ti
c
a
l
l
e
arni
ng
th
e
or
y
w
i
th
ot
he
r
the
oret
i
c
a
l
f
r
a
m
ew
ork
s
,
bi
g
da
ta
proc
es
s
i
n
g,
pa
r
am
ete
r
s
s
el
ec
ti
on
,
a
nd
th
e
ge
n
eral
i
z
a
ti
o
n
ab
i
l
i
t
y
of
a
gi
v
e
n
prob
l
em
[
23
,
24
]
.
W
i
th
the
r
ate
of
de
v
e
l
o
pm
en
t
of
i
nf
or
m
ati
o
n
s
y
s
tem
s
,
hi
g
h
-
di
m
en
s
i
on
al
,
d
y
n
am
i
c
an
d c
om
pl
ex
da
ta
are
ea
s
i
l
y
g
en
erat
ed
[
25
,
26]
.
7.2
. N
aïv
e Ba
ye
s
T
he
NB
m
eth
od
i
s
a
c
l
as
s
i
f
i
c
ati
o
n
s
c
he
m
e
whi
c
h
r
e
l
i
es
on
t
he
B
a
y
es
’
th
eo
r
em
.
T
hi
s
tec
hn
i
qu
e
as
s
um
es
the
i
nd
ep
en
de
nc
e
of
i
ts
pred
i
c
tor
s
.
S
i
m
pl
y
,
the
NB
c
l
as
s
i
f
i
er
as
s
um
es
tha
t
the
r
e
i
s
no
r
el
a
ti
o
ns
hi
p
be
t
ween
th
e e
x
i
s
ten
c
e
of
c
erta
i
n f
ea
t
ures
i
n
a c
l
as
s
an
d t
h
at
of
a
n
y
ot
he
r
f
ea
ture
[
2
7
-
30
]
.
T
hi
s
the
or
y
w
as
a
do
p
ted
i
n
d
ete
r
m
i
ni
ng
t
he
c
l
as
s
of
the
do
c
um
en
t
on
the
f
ol
l
o
wi
ng
eq
u
at
i
on
:
∗
=
(
|
)
(
5)
where c
r
ep
r
es
en
ts
th
e c
l
as
s
an
d d
r
ep
r
es
en
t th
e d
oc
u
m
en
t.
∗
=
(
|
)
∗
(
)
(
)
(
6)
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NIK
A
IS
S
N: 1
69
3
-
6
93
0
◼
T
he
c
l
as
s
i
f
i
c
at
i
on
of
t
he
m
o
de
r
n a
r
a
bi
c
p
oe
tr
y
us
i
ng
… (
Mu
ne
f
A
b
du
l
l
ah
A
hm
ed
)
2671
B
ec
au
s
e
p(d)
h
as
no
ef
f
ec
t
or r
ol
e,
th
e
eq
ua
t
i
on
s
b
ec
o
m
e:
∗
=
(
|
)
∗
(
)
(
7)
T
he
i
m
po
r
tan
t
h
y
p
oth
es
i
s
i
n
t
hi
s
al
go
r
i
t
hm
i
s
tha
t
ea
c
h
pr
op
ert
y
or
f
ea
t
u
r
e
i
n
the
do
c
um
en
t
do
es
no
t
de
pe
nd
o
n
the
oth
er
's
f
ea
ture
s
,
an
d
as
s
um
pti
on
s
produc
e
the
f
ol
l
o
w
i
ng
eq
ua
ti
o
n:
∗
=
(
|
)
∏
(
)
∗
(
)
⁄
(
8)
7.3
. Lin
e
ar
S
u
p
p
o
r
t
V
ec
t
o
r
Cla
ss
if
ic
atio
n
Li
n
ea
r
S
V
C
i
s
a
t
y
p
e
of
m
ac
hi
ne
l
e
arni
ng
al
go
r
i
t
hm
s
s
i
m
i
l
ar
to
th
e
S
V
M.
S
om
e
f
ea
tures
of
thi
s
al
go
r
i
t
hm
are
th
e
f
l
ex
i
bi
l
i
t
y
i
n
s
e
l
ec
t
i
on
a
nd
l
o
s
s
of
f
un
c
ti
on
s
.
It
i
s
s
u
i
ta
b
l
e
f
or
a
hu
g
e
nu
m
be
r
of
s
a
m
pl
es
.
F
r
om
the
t
es
ti
n
g
of
th
i
s
m
od
el
o
n
da
ta,
r
es
e
arc
he
r
s
ha
v
e
f
ou
nd
i
t
us
i
ng
one
-
a
ga
i
ns
t
-
r
es
t
a
pp
r
oa
c
h
c
om
pa
r
ed
to
S
V
M
whi
c
h
us
es
on
e
-
ag
ai
ns
t
-
o
ne
a
pp
r
oa
c
h.
T
hi
s
m
od
el
i
s
us
ed
i
n
s
ev
eral
ap
p
l
i
c
at
i
on
s
l
i
k
e
the
c
l
as
s
i
f
i
c
ati
on
of
tex
t
do
c
um
en
ts
us
i
ng
s
pa
r
s
e
f
ea
tures
[
22
-
24
]
.
8.
M
eth
o
d
o
log
y
F
i
gu
r
e
1
pres
en
ts
th
e
o
utl
i
n
e
of
o
ur
w
ork
.
In
the
be
gi
n
ni
n
g,
we
c
h
oo
s
e
the
da
tas
et
us
e
d
i
n
o
ur
wor
k
;
af
ter
tha
t,
we
s
eg
m
en
ted
i
t
i
nt
o
w
ords
a
nd
a
l
l
the
s
te
ps
of
da
ta
pr
ep
r
oc
es
s
i
ng
w
er
e
ap
p
l
i
e
d,
i
nc
l
u
di
ng
f
ea
tures
ex
tr
ac
ti
on
.
W
e
us
ed
three
m
a
c
hi
ne
l
ea
r
n
i
n
g
al
go
r
i
thm
s
(
S
V
M,
LS
V
C,
an
d N
B
)
i
n t
r
a
i
n
i
ng
an
d
te
s
t
i
ng
.
F
i
gu
r
e
1.
B
l
oc
k
di
a
gram
of
t
he
pro
po
s
ed
m
eth
od
9.
Re
sult
s
T
he
w
ork
was
do
n
e
wi
th
the
P
y
t
ho
n
l
an
g
ua
g
e
us
i
n
g
the
m
ac
hi
ne
c
on
f
i
g
urati
on
as
f
ol
l
o
w
s
:
O
S
:
W
i
nd
ow
s
7,
CP
U
S
pe
ed
:
3.2
0
G
H
z
,
P
r
oc
es
s
or:
Int
el
Cor
e
i
7
,
R
A
M:
4G
B
.
W
i
t
h
the
i
n
ten
t
i
o
n
of
s
c
r
uti
ni
z
i
n
g
th
e
s
ug
g
es
ted
w
ork
’
s
p
erf
or
m
an
c
e,
di
f
f
erent
p
ara
m
ete
r
s
s
uc
h
as
prec
i
s
i
o
n,
r
ec
a
l
l
,
a
nd
f
-
m
ea
s
ure
wer
e
m
ea
s
ured
f
or
al
l
t
y
p
es
of
m
od
ern
A
r
ab
i
c
po
em
.
Evaluation Warning : The document was created with Spire.PDF for Python.
◼
IS
S
N: 16
93
-
6
93
0
T
E
L
KO
M
NIK
A
V
ol
.
17
,
No
.
5,
O
c
tob
er 20
19
:
26
6
7
-
26
74
2672
T
he
pe
r
f
or
m
an
c
e
of
the
pr
op
os
ed
m
eth
od
i
s
pr
es
en
te
d
i
n
T
ab
l
es
5
to
8
an
d
F
i
g
ures
2
to
5,
as
de
s
c
r
i
be
d
b
el
o
w.
T
he
f
i
r
s
t
t
y
p
e
of
m
ac
hi
ne
l
ea
r
n
i
ng
a
l
go
r
i
thm
us
ed
was
N
aïv
e
B
a
y
es
.
T
ab
l
e
5
i
l
l
us
tr
ate
s
t
he
prec
i
s
i
o
n,
r
ec
al
l
,
an
d
f
-
m
ea
s
ure
f
or
thi
s
a
l
go
r
i
thm
.
T
he
m
ax
i
m
u
m
v
al
u
e
f
or
prec
i
s
i
o
n
w
as
f
or
the
po
l
i
ti
c
c
l
as
s
,
w
h
i
l
e
f
or
the
r
ec
al
l
,
the
m
ax
i
m
u
m
v
al
ue
was
f
or
l
ov
e
c
l
as
s
.
F
-
m
ea
s
ure
w
as
hi
g
he
s
t
i
n
the
s
oc
i
a
l
an
d
p
ol
i
t
i
c
c
l
as
s
es
.
T
he
r
es
ul
ts
f
or
thi
s
a
l
go
r
i
thm
w
er
e
c
o
m
pa
r
ed
to
the
r
es
ul
ts
of
oth
er m
ac
hi
ne
l
ea
r
ni
n
g a
l
g
orit
hm
s
.
T
ab
l
e
6
pres
e
nts
t
he
r
es
u
l
ts
of
th
e
S
V
M
al
go
r
i
t
hm
.
F
r
o
m
the
r
es
u
l
ts
,
the
m
ax
i
m
u
m
v
a
l
ue
s
of
prec
i
s
i
o
n,
r
ec
a
l
l
,
an
d
f
-
m
ea
s
ure
wer
e
al
l
f
or
the
Is
l
am
i
c
c
l
as
s
.
T
hi
s
r
es
ul
t
was
a
l
s
o
c
o
m
pa
r
ed
to
th
e
r
es
ul
ts
of
the
ot
he
r
m
ac
hi
ne
l
e
arni
ng
f
r
a
m
ew
ork
s
.
T
ab
l
e
7
i
l
l
us
tr
ate
s
the
r
es
u
l
t
of
th
e
c
l
as
s
i
f
i
c
ati
on
proc
es
s
us
i
ng
l
i
ne
ar
S
V
C
a
l
g
orit
hm
. Fr
om
th
e res
ul
ts
,
th
e m
ax
i
m
u
m
v
al
ue
of
prec
i
s
i
o
n
w
as
f
or
the
s
oc
i
a
l
c
l
as
s
w
h
i
l
e
the
m
ax
i
m
u
m
v
a
l
ue
s
f
or
r
ec
al
l
an
d
f
-
m
ea
s
ure
w
ere
f
or
l
o
v
e
c
l
as
s
.
T
ab
l
e
8
i
l
l
us
tr
ate
s
the
a
v
er
ag
e
v
a
l
ue
f
o
r
prec
i
s
i
on
,
r
ec
a
l
l
,
an
d
f
-
m
ea
s
ure
f
or
al
l
the
m
ac
hi
ne
l
e
arni
ng
a
l
g
ori
thm
s
us
ed
i
n
th
e
c
l
as
s
i
f
i
c
ati
on
of
ou
r
d
ata
s
et.
F
r
om
the
tab
l
e,
l
i
ne
ar
S
V
C
al
go
r
i
thm
w
as
f
ou
nd
t
o h
a
v
e
the
m
ax
i
m
u
m
prec
i
s
i
on
,
r
ec
al
l
, a
nd
f
-
m
ea
s
ure v
al
u
es
.
F
i
gu
r
e
2
i
l
l
us
tr
ate
s
t
he
prec
i
s
i
on
f
or
al
l
t
y
p
es
of
m
od
ern
A
r
ab
i
c
po
em
us
i
ng
three
m
a
c
hi
ne
l
ea
r
ni
ng
a
l
go
r
i
thm
s
.
F
r
om
the
f
i
gu
r
e
,
t
he
m
ax
i
m
u
m
v
al
u
e
of
prec
i
s
i
on
f
or
m
os
t
t
y
pe
s
of
the
m
od
ern
p
oe
m
w
as
pre
s
en
ted
b
y
th
e
l
i
ne
ar
S
V
C
al
g
orit
hm
w
h
i
l
e
t
he
m
i
ni
m
um
v
al
ue
was
pres
en
te
d
b
y
the
S
V
M
a
l
g
orit
hm
.
W
he
n
w
e
c
om
pa
r
ed
the
r
ec
a
l
l
f
or
ou
r
da
t
as
et
as
c
al
c
ul
at
ed
us
i
ng
the
tr
ee
m
ac
hi
ne
l
ea
r
ni
n
g
a
l
go
r
i
thm
s
,
w
e
f
ou
nd
t
he
m
ax
i
m
u
m
r
ec
al
l
v
a
l
ue
i
n
bo
t
h
N
B
a
nd
LS
V
C
a
l
go
r
i
thm
s
w
h
i
l
e
th
e
m
i
ni
m
u
m
r
ec
al
l
v
a
l
ue
w
as
f
ou
nd
i
n
S
V
M
a
l
go
r
i
th
m
as
s
ho
wn
i
n
F
i
gu
r
e
3.
F
i
g
ur
es
4
i
l
l
us
tr
at
es
the
f
-
m
ea
s
ure
f
or
ou
r
d
a
tas
et.
T
he
s
e
qu
e
nc
e
of
v
a
l
ue
s
f
r
o
m
top
t
o
bo
tto
m
i
n
t
he
s
e
al
go
r
i
t
hm
s
was
as
f
ol
l
o
w
s
:
LS
V
C,
N
B
,
an
d
S
V
M
al
go
r
i
thm
.
F
i
gu
r
e
5
i
l
l
us
tr
at
es
the
a
v
er
ag
e
v
al
u
e f
or our
d
ata
s
et.
T
he
be
s
t res
u
l
t
was
f
ou
nd
i
n t
h
e L
S
V
C
al
g
ori
t
h
m
,
f
ol
l
o
w
e
d b
y
the
N
B
a
l
go
r
i
thm
an
d
S
V
M
al
g
orit
hm
.
T
ab
l
e 5
.
C
l
as
s
i
f
i
c
at
i
on
of
ou
r
Data
s
et
us
i
ng
Naïv
e B
a
y
es
T
ab
l
e 6
.
C
l
as
s
i
f
i
c
at
i
on
of
ou
r
Data
s
et
us
i
ng
S
up
po
r
t
V
ec
tor M
ac
hi
ne
p
r
e
c
is
ion
r
e
c
a
ll
F
-
m
e
a
s
u
r
e
I
s
la
mi
c
0
.
1
4
0
.
5
0
.
2
2
L
o
v
e
0
.
5
7
0
.
8
0
.
6
7
P
o
li
t
i
c
1
0
.
5
0
.
6
7
S
o
c
ial
0
.
5
0
.
1
7
0
.
2
5
A
v
e
r
a
g
e
0
.
6
4
0
.
4
7
0
.
4
9
p
r
e
c
is
ion
r
e
c
a
ll
F
-
m
e
a
s
u
r
e
I
s
la
mi
c
0
.
5
0
.
2
5
0
.
3
3
L
o
v
e
0
.
0
2
0
.
1
0
.
2
P
o
li
t
i
c
0
.
0
7
0
.
0
5
0
.
0
9
S
o
c
ial
0
.
1
2
0
.
1
6
0
.
1
A
v
e
r
a
g
e
0
.
1
7
7
5
0
.
1
4
0
.
1
8
T
ab
l
e 7
.
C
l
as
s
i
f
i
c
at
i
on
of
ou
r
Data
s
et
us
i
ng
Li
n
ea
r
S
up
p
ort V
ec
tor C
l
as
s
i
f
i
c
ati
on
T
ab
l
e 8
.
A
v
erag
e res
ul
ts
of
ou
r
Dat
as
et
us
i
ng
T
hree M
ac
hi
ne
L
ea
r
n
i
ng
A
l
go
r
i
thm
s
p
r
e
c
is
ion
r
e
c
a
ll
F
-
m
e
a
s
u
r
e
I
s
la
mi
c
0
.
1
7
0
.
5
0
.
2
5
L
o
v
e
0
.
8
3
0
.
7
1
0
.
7
7
P
o
li
t
i
c
0
.
2
0
.
3
3
0
.
2
5
S
o
c
ial
1
0
.
2
9
0
.
4
4
A
v
e
r
a
g
e
0
.
7
2
0
.
4
7
0
.
5
1
p
r
e
c
is
ion
r
e
c
a
ll
F
-
m
e
a
s
u
r
e
N
a
ï
v
e
B
a
y
e
s
0
.
6
4
0
.
4
7
0
.
4
9
S
u
p
p
o
r
t
V
e
c
t
o
r
M
a
c
h
ine
0
.
1
7
7
5
0
.
1
4
0
.
1
8
L
ine
a
r
S
u
p
p
o
r
t
V
e
c
t
o
r
C
las
s
i
f
i
c
a
t
ion
0
.
7
2
0
.
4
7
0
.
5
1
F
i
gu
r
e
2
.
T
he
prec
i
s
i
on
f
or
ou
r
da
t
as
et
us
i
ng
three m
ac
hi
ne
l
ea
r
ni
n
g a
l
g
orit
hm
s
F
i
gu
r
e
3
.
T
he
r
ec
al
l
f
or our
da
tas
et
us
i
n
g
three m
ac
hi
ne
l
ea
r
ni
n
g a
l
g
orit
hm
s
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NIK
A
IS
S
N: 1
69
3
-
6
93
0
◼
T
he
c
l
as
s
i
f
i
c
at
i
on
of
t
he
m
o
de
r
n a
r
a
bi
c
p
oe
tr
y
us
i
ng
… (
Mu
ne
f
A
b
du
l
l
ah
A
hm
ed
)
2673
F
i
gu
r
e
4
.
T
he
F
-
m
ea
s
ure f
o
r
ou
r
da
t
as
et
us
i
ng
three m
ac
hi
ne
l
ea
r
ni
n
g a
l
g
orit
hm
s
F
i
gu
r
e
5
.
A
v
erage
r
es
ul
ts
o
f
ou
r
da
tas
et
us
i
n
g
three m
ac
hi
ne
l
ea
r
ni
n
g a
l
g
orit
hm
s
10
. Con
c
lusio
n
In
th
i
s
pa
p
er,
w
e
us
e
d
S
u
pp
ort
V
ec
t
or
M
ac
hi
n
e,
l
i
ne
ar
S
up
p
ort
V
ec
tor
C
l
as
s
i
f
i
c
ati
on
,
an
d
Naï
v
e
B
a
y
es
f
or
th
e
c
l
as
s
i
f
i
c
ati
on
of
m
od
ern
A
r
a
bi
c
po
em
s
.
T
he
m
a
c
hi
ne
l
ea
r
n
i
ng
al
g
orit
hm
s
prov
e
d
t
o
b
e
g
o
od
t
oo
l
s
f
or
tex
t
c
l
as
s
i
f
i
c
at
i
on
.
F
r
om
the
c
om
pa
r
i
s
on
of
the
r
es
u
l
t
of
the
prec
i
s
i
on
,
r
ec
a
l
l
,
a
nd
f
-
m
ea
s
ure
f
or
al
l
t
y
p
es
of
the
m
od
ern
A
r
ab
i
c
po
em
,
the
be
s
t
r
es
ul
t
w
as
f
ou
nd
w
h
en
us
i
n
g
l
i
n
ea
r
S
up
po
r
t
V
ec
tor
Cl
as
s
i
f
i
c
at
i
on
a
nd
Naï
v
e
B
a
y
es
.
O
n
e
of
the
m
ai
n
r
ea
s
on
s
f
or
thi
s
d
i
s
pa
r
i
t
y
i
n
pe
r
f
or
m
an
c
e
c
ou
l
d
be
t
he
s
i
z
e
o
f
the
d
ata
s
et
s
i
nc
e
s
o
m
e
m
ac
hi
ne
l
ea
r
n
i
n
g
al
g
orit
hm
s
c
an
wo
r
k
be
tte
r
wi
th
f
ew
d
ata
s
ets
.
A
l
s
o,
th
e
prepr
oc
es
s
i
n
g
of
ou
r
da
tas
e
t
was
an
i
m
po
r
tan
t
s
te
p
as
i
t
i
nc
r
ea
s
ed
th
e
ac
c
urac
y
of
t
he
c
l
as
s
i
f
i
c
ati
on
an
d
r
e
du
c
ed
the
r
eq
ui
r
e
d
m
e
m
or
y
s
i
z
e
f
or
the
c
l
as
s
i
f
i
c
a
ti
on
proc
e
s
s
.
T
hi
s
m
eth
od
of
c
l
as
s
i
f
i
c
ati
on
c
an
b
e
f
urther i
m
prov
ed
f
or the
oth
er t
y
p
es
of
A
r
ab
i
c
po
etr
y
.
Refe
ren
c
e
s
[1
]
M
A
Ah
m
e
d
,
S
T
ra
u
s
a
n
-
M
a
tu
.
Us
i
n
g
n
a
tu
ra
l
l
a
n
g
u
a
g
e
p
ro
c
e
s
s
i
n
g
fo
r
a
n
a
l
y
z
i
n
g
Ara
b
i
c
p
o
e
try
rh
y
t
h
m
.
in
Net
w
o
rk
i
n
g
i
n
Ed
u
c
a
ti
o
n
a
n
d
Re
s
e
a
rc
h
(Ro
Ed
u
Ne
t),
2
0
1
7
1
6
th
Ro
E
d
u
Net
C
o
n
fe
r
e
n
c
e
.
2017
:
1
-
5
.
[2
]
S
Al
-
Harb
i
,
A
A
l
m
u
h
a
re
b
,
A
Al
-
T
h
u
b
a
i
ty
,
M
Kh
o
rs
h
e
e
d
,
A
Al
-
Raj
e
h
.
Au
to
m
a
ti
c
Ara
b
i
c
te
x
t
c
l
a
s
s
i
fi
c
a
ti
o
n
.
J
ADT
2
0
0
8
:
9
e
s
J
o
u
rn
é
e
s
i
n
te
r
n
a
ti
o
n
a
l
e
s
d
’
An
a
l
y
s
e
s
ta
t
i
s
ti
q
u
e
d
e
s
Don
n
é
e
s
T
e
x
tu
e
l
l
e
s
.
2
0
0
8
:
77
-
83.
[3
]
M
Ab
d
u
l
-
M
a
g
e
e
d
,
M
T
Dia
b
,
M
Ko
ra
y
e
m
.
Su
b
j
e
c
ti
v
i
ty
a
n
d
s
e
n
t
i
m
e
n
t
a
n
a
l
y
s
i
s
o
f
m
o
d
e
rn
s
t
a
n
d
a
r
d
Ara
b
i
c
.
i
n
Pro
c
e
e
d
i
n
g
s
o
f
th
e
4
9
th
An
n
u
a
l
M
e
e
ti
n
g
o
f
t
h
e
A
s
s
o
c
i
a
ti
o
n
f
o
r
Com
p
u
ta
t
i
o
n
a
l
L
i
n
g
u
i
s
t
i
c
s
:
Hum
a
n
L
a
n
g
u
a
g
e
T
e
c
h
n
o
l
o
g
i
e
s
.
2
0
1
1
;
2
:
5
8
7
-
5
9
1
.
[4
]
A O
rto
n
y
,
G
L
Cl
o
re
,
A
Col
l
i
n
s
.
T
h
e
c
o
g
n
i
ti
v
e
s
tru
c
tu
re
o
f
e
m
o
ti
o
n
s
:
Ca
m
b
ri
d
g
e
u
n
i
v
e
rs
i
ty
p
r
e
s
s
.
1
9
9
0
.
[5
]
H
L
i
u
,
H
L
i
e
b
e
rm
a
n
,
T
Se
l
k
e
r
.
A
m
o
d
e
l
o
f
te
x
tu
a
l
a
ff
e
c
t
s
e
n
s
i
n
g
u
s
i
n
g
re
a
l
-
w
o
rl
d
k
n
o
wle
d
g
e
.
in
Pro
c
e
e
d
i
n
g
s
o
f
th
e
8
th
i
n
t
e
rn
a
t
i
o
n
a
l
c
o
n
fe
r
e
n
c
e
o
n
In
te
l
l
i
g
e
n
t
u
s
e
r
i
n
te
r
fa
c
e
s
.
2
0
0
3
:
1
2
5
-
1
3
2
.
[6
]
M
G
Dy
e
r
.
Em
o
ti
o
n
s
a
n
d
th
e
i
r
c
o
m
p
u
ta
t
i
o
n
s
:
T
h
re
e
c
o
m
p
u
t
e
r
m
o
d
e
l
s
.
Cog
n
i
ti
o
n
a
n
d
e
m
o
ti
o
n
.
1987
;
1
(
3
):
3
2
3
-
3
4
7
.
[7
]
O
Al
s
h
a
r
i
f,
D
Al
s
h
a
m
a
a
,
N
G
h
n
e
i
m
.
Em
o
ti
o
n
c
l
a
s
s
i
fi
c
a
ti
o
n
i
n
Ara
b
i
c
p
o
e
try
u
s
i
n
g
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
.
In
te
rn
a
ti
o
n
a
l
J
o
u
r
n
a
l
o
f
Com
p
u
te
r Ap
p
l
i
c
a
ti
o
n
s
.
2
0
1
3
;
5
6
(1
6
)
:1
0
-
15
.
[8
]
M
M
Al
-
T
a
h
ra
w
i
,
SN
Al
-
Kh
a
ti
b
.
Ara
b
i
c
t
e
x
t
c
l
a
s
s
i
fi
c
a
ti
o
n
u
s
i
n
g
Po
l
y
n
o
m
i
a
l
Net
w
o
rk
s
.
J
o
u
rn
a
l
o
f
Ki
n
g
Sa
u
d
Un
i
v
e
r
s
i
t
y
-
Com
p
u
te
r
a
n
d
I
n
fo
rm
a
t
i
o
n
S
c
i
e
n
c
e
s
.
2
0
1
5
;
27
(
4
):
437
-
4
4
9
.
[9
]
S
Al
s
a
l
e
e
m
.
Au
to
m
a
te
d
Ara
b
i
c
T
e
x
t
Cat
e
g
o
ri
z
a
ti
o
n
u
s
i
n
g
SV
M
a
n
d
NB
.
i
n
In
t.
Ara
b
J
.
e
-
Te
c
h
n
o
l
.
2011
;
2
(
2
):
124
-
128.
[1
0
]
R
Be
l
k
e
b
i
r,
A
G
u
e
s
s
o
u
m
.
A
h
y
b
ri
d
BSO
-
Chi
2
-
SVM
a
p
p
ro
a
c
h
t
o
Ara
b
i
c
te
x
t
c
a
te
g
o
ri
z
a
t
i
o
n
.
i
n
AC
S
In
te
rn
a
ti
o
n
a
l
Co
n
fe
re
n
c
e
o
n
C
o
m
p
u
te
r S
y
s
te
m
s
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
(AI
CC
SA)
.
2
0
1
3
:
1
-
7.
[1
1
]
J
Ab
a
b
n
e
h
,
O
Al
m
o
m
a
n
i
,
W
Had
i
,
NKT
El
-
O
m
a
ri
,
A
Al
-
Ib
r
a
h
i
m
.
Ve
c
t
o
r
s
p
a
c
e
m
o
d
e
l
s
t
o
c
l
a
s
s
i
fy
Ara
b
i
c
te
x
t
.
In
te
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
Com
p
u
te
r
Tre
n
d
s
a
n
d
Te
c
h
n
o
l
o
g
y
(I
J
CTT)
.
2
0
1
4
;
7
(
4
)
:
219
-
2
2
3
.
[1
2
]
S
Kh
o
rs
h
e
e
d
,
AO
Al
-
T
h
u
b
a
i
ty
.
Com
p
a
ra
t
i
v
e
e
v
a
l
u
a
t
i
o
n
o
f
te
x
t
c
l
a
s
s
i
fi
c
a
t
i
o
n
te
c
h
n
i
q
u
e
s
u
s
i
n
g
a
l
a
r
g
e
d
i
v
e
rs
e
Ar
a
b
i
c
d
a
ta
s
e
t
.
La
n
g
u
a
g
e
re
s
o
u
r
c
e
s
a
n
d
e
v
a
l
u
a
ti
o
n
.
2013
;
47
(
2
):
5
1
3
-
5
3
8
.
[1
3
]
L
Fo
d
i
l
,
H
Sa
y
o
u
d
,
S
O
u
a
m
o
u
r
.
T
h
e
m
e
c
l
a
s
s
i
fi
c
a
ti
o
n
o
f
Ara
b
i
c
te
x
t:
A
s
ta
t
i
s
t
i
c
a
l
a
p
p
ro
a
c
h
.
T
e
rm
i
n
o
l
o
g
y
a
n
d
K
n
o
w
l
e
d
g
e
En
g
i
n
e
e
ri
n
g
.
2
0
1
4
:
0
1
0
0
5
8
7
3
.
Evaluation Warning : The document was created with Spire.PDF for Python.
◼
IS
S
N: 16
93
-
6
93
0
T
E
L
KO
M
NIK
A
V
ol
.
17
,
No
.
5,
O
c
tob
er 20
19
:
26
6
7
-
26
74
2674
[1
4
]
C Hol
e
s
.
M
o
d
e
rn
Ara
b
i
c
:
Str
u
c
tu
re
s
,
f
u
n
c
ti
o
n
s
,
a
n
d
v
a
ri
e
t
i
e
s
:
G
e
o
rg
e
to
w
n
Un
i
v
e
rs
i
ty
Pre
s
s
.
2004.
[1
5
]
S
Kh
o
j
a
,
R
G
a
rs
i
d
e
.
St
e
m
m
i
n
g
a
r
a
b
i
c
te
x
t
.
L
a
n
c
a
s
te
r,
UK,
Com
p
u
ti
n
g
Dep
a
rt
m
e
n
t,
L
a
n
c
a
s
t
e
r
Uni
v
e
rs
i
ty
.
1
9
9
9
.
[1
6
]
B
Pa
n
g
,
L
L
e
e
,
S
V
a
i
th
y
a
n
a
th
a
n
.
T
h
u
m
b
s
u
p
?
:
s
e
n
ti
m
e
n
t
c
l
a
s
s
i
f
i
c
a
t
i
o
n
u
s
i
n
g
m
a
c
h
i
n
e
l
e
a
rn
i
n
g
te
c
h
n
i
q
u
e
s
.
i
n
Pr
o
c
e
e
d
i
n
g
s
o
f
th
e
ACL
-
0
2
c
o
n
fe
re
n
c
e
o
n
Em
p
i
ri
c
a
l
m
e
th
o
d
s
i
n
n
a
tu
r
a
l
l
a
n
g
u
a
g
e
p
ro
c
e
s
s
i
n
g
.
2
0
0
2
;
10
:
79
-
86
.
[1
7
]
C
Su
d
h
e
e
r,
R
M
a
h
e
s
w
a
ra
n
,
B
K
Pa
n
i
g
ra
h
i
,
S
M
a
th
u
r
.
A
h
y
b
ri
d
SVM
-
PSO
m
o
d
e
l
fo
r
f
o
re
c
a
s
ti
n
g
m
o
n
t
h
l
y
s
tre
a
m
f
l
o
w
.
Neu
ra
l
C
o
m
p
u
ti
n
g
a
n
d
Ap
p
l
i
c
a
ti
o
n
s
.
2
0
1
4
;
24
(
6
):
1
3
8
1
-
1
3
8
9
.
[1
8
]
X
Zh
a
n
g
,
S
Din
g
,
Y
X
u
e
.
An
i
m
p
ro
v
e
d
m
u
l
t
i
p
l
e
b
i
rt
h
s
u
p
p
o
r
t
v
e
c
to
r
m
a
c
h
i
n
e
f
o
r
p
a
tt
e
rn
c
l
a
s
s
i
f
i
c
a
ti
o
n
.
Neu
ro
c
o
m
p
u
ti
n
g
.
2
0
1
7
;
225
:
1
1
9
-
1
2
8
.
[1
9
]
Z
Che
n
,
Z
Q
i
,
B
W
a
n
g
,
L
Cui
,
F
M
e
n
g
,
Y
Sh
i
.
L
e
a
rn
i
n
g
w
i
th
l
a
b
e
l
p
r
o
p
o
rt
i
o
n
s
b
a
s
e
d
o
n
n
o
n
p
a
ra
l
l
e
l
s
u
p
p
o
rt
v
e
c
t
o
r m
a
c
h
i
n
e
s
.
Kn
o
wle
d
g
e
-
B
a
s
e
d
Sy
s
te
m
s
.
2
0
1
7
;
119
:
126
-
141.
[2
0
]
W
J
i
a
n
g
,
D
-
S
Hua
n
g
,
S
L
i
.
Ran
d
o
m
w
a
l
k
-
b
a
s
e
d
s
o
l
u
t
i
o
n
to
tr
i
p
l
e
l
e
v
e
l
s
to
c
h
a
s
ti
c
p
o
i
n
t
l
o
c
a
t
i
o
n
p
ro
b
l
e
m
.
IEEE
tra
n
s
a
c
ti
o
n
s
o
n
c
y
b
e
rn
e
t
i
c
s
.
2
0
1
6
;
46
(
6
):
1
4
3
8
-
1
4
5
1
.
[2
1
]
T
J
o
a
c
h
i
m
s
.
Te
x
t
c
a
te
g
o
ri
z
a
ti
o
n
wit
h
s
u
p
p
o
rt
v
e
c
t
o
r
m
a
c
h
i
n
e
s
:
L
e
a
rn
i
n
g
wit
h
m
a
n
y
r
e
l
e
v
a
n
t
fe
a
tu
r
e
s
.
i
n
Eu
ro
p
e
a
n
c
o
n
fe
re
n
c
e
o
n
m
a
c
h
i
n
e
l
e
a
rn
i
n
g
.
1
9
9
8
:
1
3
7
-
1
4
2
.
[2
2
]
F
Deb
o
l
e
,
F
Se
b
a
s
ti
a
n
i
.
A
n
a
n
a
l
y
s
i
s
o
f
th
e
r
e
l
a
t
i
v
e
h
a
rd
n
e
s
s
o
f
Reu
te
r
s
‐
2
1
5
7
8
s
u
b
s
e
t
s
.
J
o
u
rn
a
l
o
f
t
h
e
As
s
o
c
i
a
ti
o
n
f
o
r I
n
fo
rm
a
ti
o
n
S
c
i
e
n
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
.
2
0
0
5
;
56
(
6
):
5
8
4
-
5
9
6
.
[2
3
]
RA
Has
a
n
,
M
A M
o
h
a
m
m
e
d
,
Z
H Sa
l
i
h
,
M
AB Am
e
e
d
e
e
n
,
N Ţ
ă
p
u
ş
,
M
N
M
o
h
a
m
m
e
d
.
HSO
:
A Hy
b
ri
d
Sw
a
rm
Opt
i
m
i
z
a
ti
o
n
Al
g
o
ri
t
h
m
f
o
r Re
d
u
c
i
n
g
En
e
rg
y
Co
n
s
u
m
p
ti
o
n
i
n
t
h
e
Cl
o
u
d
l
e
ts
.
TEL
KO
M
NIKA
Te
l
e
c
o
m
m
u
n
i
c
a
ti
o
n
,
Com
p
u
t
i
n
g
,
El
e
c
tro
n
i
c
s
a
n
d
Co
n
tro
l
.
2
0
1
8
;
16
(
5
)
:
2
1
4
4
-
2
1
5
4
.
[2
4
]
RA
Has
a
n
,
M
A
M
o
h
a
m
m
e
d
,
N
Ţ
ă
p
u
ş
,
O
A
Ha
m
m
o
o
d
.
A
c
o
m
p
re
h
e
n
s
i
v
e
s
tu
d
y
:
A
n
t
Col
o
n
y
O
p
ti
m
i
z
a
ti
o
n
(ACO
)
fo
r
f
a
c
i
l
i
t
y
l
a
y
o
u
t
p
ro
b
l
e
m
.
i
n
2
0
1
7
1
6
t
h
RoEd
u
Ne
t
Con
fe
re
n
c
e
:
Net
w
o
rk
i
n
g
i
n
Ed
u
c
a
t
i
o
n
a
n
d
Re
s
e
a
rc
h
(Ro
E
d
u
Net
)
.
2
0
1
7
:
1
-
8.
[2
5
]
M
A
M
o
h
a
m
m
e
d
,
ZH
Sa
l
i
h
,
N
Ţ
ă
p
u
ş
,
RAK
Has
a
n
.
Se
c
u
ri
t
y
a
n
d
a
c
c
o
u
n
ta
b
i
l
i
ty
fo
r
s
h
a
ri
n
g
th
e
d
a
t
a
s
to
re
d
i
n
th
e
c
l
o
u
d
.
i
n
2
0
1
6
1
5
th
R
o
Ed
u
Net
Con
f
e
re
n
c
e
:
Net
w
o
rk
i
n
g
i
n
Ed
u
c
a
ti
o
n
a
n
d
Res
e
a
rc
h
.
2016
:
1
-
5.
[2
6
]
MA
M
o
h
a
m
m
e
d
,
N
Ţ
ĂPUŞ
.
A
Nov
e
l
Ap
p
r
o
a
c
h
o
f
R
e
d
u
c
i
n
g
En
e
r
g
y
Con
s
u
m
p
ti
o
n
b
y
Uti
l
i
z
i
n
g
En
th
a
l
p
y
i
n
M
o
b
i
l
e
Cl
o
u
d
C
o
m
p
u
ti
n
g
.
Stu
d
i
e
s
i
n
I
n
f
o
rm
a
ti
c
s
a
n
d
Co
n
tro
l
.
2
0
1
7
;
26
:
4
2
5
-
4
3
4
.
[2
7
]
MA
M
o
h
a
m
m
e
d
,
RA
Ha
s
a
n
.
Pa
rti
c
l
e
s
warm
o
p
ti
m
i
z
a
ti
o
n
fo
r
f
a
c
i
l
i
t
y
l
a
y
o
u
t
p
r
o
b
l
e
m
s
FL
P
—
A
c
o
m
p
re
h
e
n
s
i
v
e
s
tu
d
y
.
in
2
0
1
7
1
3
t
h
IEEE
In
te
rn
a
ti
o
n
a
l
Con
fe
re
n
c
e
o
n
In
te
l
l
i
g
e
n
t
Com
p
u
te
r
Com
m
u
n
i
c
a
ti
o
n
a
n
d
Pro
c
e
s
s
i
n
g
(I
CC
P)
.
2
0
1
7
:
93
-
99.
[2
8
]
ZH
Sa
l
i
h
,
G
T
Has
a
n
,
M
A
M
o
h
a
m
m
e
d
.
I
n
v
e
s
ti
g
a
te
a
n
d
a
n
a
l
y
z
e
th
e
l
e
v
e
l
s
o
f
e
l
e
c
t
ro
m
a
g
n
e
ti
c
ra
d
i
a
t
i
o
n
s
e
m
i
tt
e
d
fr
o
m
u
n
d
e
rg
ro
u
n
d
p
o
w
e
r
c
a
b
l
e
s
e
x
t
e
n
d
e
d
i
n
m
o
d
e
rn
c
i
t
i
e
s
.
i
n
2
0
1
7
9
t
h
In
te
rn
a
ti
o
n
a
l
Co
n
fe
re
n
c
e
o
n
E
l
e
c
tro
n
i
c
s
,
Co
m
p
u
te
rs
a
n
d
Art
i
f
i
c
i
a
l
I
n
te
l
l
i
g
e
n
c
e
(ECAI),
2
0
1
7
.
[2
9
]
RA
Has
a
n
,
M
N
M
o
h
a
m
m
e
d
.
A
k
ri
l
l
h
e
rd
b
e
h
a
v
i
o
u
r
i
n
s
p
i
re
d
l
o
a
d
b
a
l
a
n
c
i
n
g
o
f
ta
s
k
s
i
n
c
l
o
u
d
c
o
m
p
u
ti
n
g
.
St
u
d
i
e
s
i
n
I
n
fo
rm
a
t
i
c
s
a
n
d
Co
n
tro
l
.
2
0
1
7
;
26
:
413
-
424.
[3
0
]
M
A
M
o
h
a
m
m
e
d
,
RA
Ha
s
a
n
,
M
A
Ah
m
e
d
,
N
T
a
p
u
s
,
M
A
Sh
a
n
a
n
,
M
K
Kh
a
l
e
e
l
,
e
t
a
l
.
A
Fo
c
a
l
l
o
a
d
b
a
l
a
n
c
e
r
b
a
s
e
d
a
l
g
o
r
i
th
m
fo
r
ta
s
k
a
s
s
i
g
n
m
e
n
t
i
n
c
l
o
u
d
e
n
v
i
ro
n
m
e
n
t
.
i
n
2
0
1
8
1
0
th
In
t
e
rn
a
ti
o
n
a
l
Con
fe
re
n
c
e
o
n
El
e
c
tr
o
n
i
c
s
,
Co
m
p
u
t
e
rs
a
n
d
Art
i
fi
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
(ECAI)
.
2018
:
1
-
4.
Evaluation Warning : The document was created with Spire.PDF for Python.