T
E
L
KO
M
N
I
KA
T
e
lec
om
m
u
n
icat
ion
,
Com
p
u
t
i
n
g,
E
lec
t
r
on
ics
an
d
Cont
r
ol
Vol.
1
8
,
No.
3
,
J
une
2020
,
pp.
1
2
6
8
~
1
2
7
4
I
S
S
N:
1693
-
6930,
a
c
c
r
e
dit
e
d
F
ir
s
t
G
r
a
de
by
Ke
me
nr
is
tekdikti
,
De
c
r
e
e
No:
21/E
/KP
T
/2018
DO
I
:
10.
12928/
T
E
L
KO
M
NI
KA
.
v1
8
i
3
.
14028
1268
Jou
r
n
al
h
omepage
:
ht
tp:
//
jour
nal.
uad
.
ac
.
id/
index
.
php/T
E
L
K
OM
N
I
K
A
D
e
si
gn
of
o
p
t
imal s
e
a
r
c
h
e
n
gi
n
e
u
si
n
g t
e
xt
su
m
m
ar
iz
at
io
n
t
h
r
o
u
gh
a
r
t
ifi
c
ia
l
in
t
e
ll
ig
e
n
c
e
t
e
c
h
n
i
q
u
e
s
Kau
s
h
ik
S
e
k
ar
an
1
,
P
.
Cha
n
d
an
a
2
,
J.
Re
t
h
n
a
Vir
gil
Je
n
y
3
,
M
ay
t
h
am
N.
M
e
q
d
ad
4
,
S
e
if
e
d
in
e
Kad
r
y
5
1
,
2,
3
D
ep
ar
t
men
t
o
f
Co
m
p
u
t
er
Sci
en
ce
a
n
d
E
n
g
i
n
eeri
n
g
,
V
i
g
n
a
n
In
s
t
i
t
u
t
e
o
f
T
ec
h
n
o
l
o
g
y
an
d
Sci
en
ce,
In
d
i
a
4
Al
-
M
u
s
t
aq
b
al
U
n
i
v
ers
i
t
y
Co
l
l
e
g
e,
Iraq
5
D
ep
ar
t
men
t
o
f
Mat
h
emat
i
cs
an
d
Co
m
p
u
t
er
Sci
ec
n
e,
Facu
l
t
y
o
f
Sc
i
en
ce,
Bei
r
u
t
A
ra
b
U
n
i
v
eri
s
t
y
,
L
eb
an
o
n
Ar
t
icle
I
n
f
o
AB
S
T
RA
CT
A
r
ti
c
le
h
is
tor
y
:
R
e
c
e
ived
S
e
p
2
,
2019
R
e
vis
e
d
J
a
n
18
,
2020
Ac
c
e
pted
F
e
b
23
,
2020
N
at
u
ral
l
a
n
g
u
ag
e
p
ro
ce
s
s
i
n
g
i
s
t
h
e
t
re
n
d
i
n
g
t
o
p
i
c
i
n
t
h
e
l
at
es
t
res
earc
h
areas
,
w
h
i
ch
al
l
o
w
s
t
h
e
d
ev
e
l
o
p
ers
t
o
crea
t
e
t
h
e
h
u
man
-
c
o
m
p
u
t
er
i
n
t
erac
t
i
o
n
s
t
o
co
me
i
n
t
o
ex
i
s
t
en
ce.
T
h
e
n
a
t
u
ra
l
l
a
n
g
u
ag
e
p
r
o
ces
s
i
n
g
i
s
a
n
i
n
t
eg
ra
t
i
o
n
o
f
art
i
f
i
ci
a
l
i
n
t
e
l
l
i
g
e
n
ce,
co
mp
u
t
er
s
c
i
en
ce
an
d
co
mp
u
t
er
l
i
n
g
u
i
s
t
i
cs
.
T
h
e
re
s
earch
t
o
w
ard
s
n
a
t
u
ra
l
L
an
g
u
a
g
e
Pro
ce
s
s
i
n
g
i
s
fo
cu
s
ed
o
n
crea
t
i
n
g
i
n
n
o
v
at
i
o
n
s
t
o
w
ar
d
s
crea
t
i
n
g
t
h
e
d
ev
i
ces
o
r
mach
i
n
e
s
w
h
i
ch
o
p
erat
es
b
a
s
i
n
g
o
n
t
h
e
s
i
n
g
l
e
co
mma
n
d
o
f
a
h
u
ma
n
.
It
al
l
o
w
s
v
ar
i
o
u
s
Bo
t
creat
i
o
n
s
t
o
i
n
n
o
v
at
e
t
h
e
i
n
s
t
r
u
ct
i
o
n
s
fr
o
m
t
h
e
mo
b
i
l
e
d
ev
i
ces
t
o
co
n
t
ro
l
t
h
e
p
h
y
s
i
ca
l
d
ev
i
ces
b
y
a
l
l
o
w
i
n
g
t
h
e
s
p
ee
ch
-
t
ag
g
i
n
g
.
In
o
u
r
p
ap
er,
w
e
d
es
i
g
n
a
s
earc
h
en
g
i
n
e
w
h
i
c
h
n
o
t
o
n
l
y
d
i
s
p
l
ay
s
t
h
e
d
a
t
a
acco
r
d
i
n
g
t
o
u
s
er
q
u
er
y
b
u
t
al
s
o
p
erfo
rm
s
t
h
e
d
et
a
i
l
e
d
d
i
s
p
l
ay
o
f
t
h
e
co
n
t
e
n
t
o
r
t
o
p
i
c
u
s
er
i
s
i
n
t
ere
s
t
e
d
fo
r
u
s
i
n
g
t
h
e
s
u
mmari
za
t
i
o
n
c
o
n
ce
p
t
.
W
e
fi
n
d
t
h
e
d
e
s
i
g
n
ed
s
earc
h
en
g
i
n
e
i
s
h
av
i
n
g
o
p
t
i
mal
re
s
p
o
n
s
e
t
i
me
f
o
r
t
h
e
u
s
er
q
u
eri
e
s
b
y
a
n
al
y
z
i
n
g
w
i
t
h
n
u
mb
er
o
f
t
ran
s
act
i
o
n
s
as
i
n
p
u
t
s
.
A
l
s
o
,
t
h
e
res
u
l
t
fi
n
d
i
n
g
s
i
n
t
h
e
p
erfo
rma
n
ce
an
al
y
s
i
s
s
h
o
w
t
h
at
t
h
e
t
e
x
t
s
u
mmar
i
zat
i
o
n
met
h
o
d
h
as
b
ee
n
an
effi
ci
e
n
t
w
a
y
fo
r
i
mp
r
o
v
i
n
g
t
h
e
re
s
p
o
n
s
e
t
i
me
i
n
t
h
e
s
earc
h
en
g
i
n
e
o
p
t
i
m
i
zat
i
o
n
s
.
K
e
y
w
o
r
d
s
:
Ar
ti
f
icia
l
int
e
ll
igenc
e
B
ot
c
r
e
a
ti
on
Na
tur
a
l
langua
ge
pr
oc
e
s
s
ing
S
e
a
r
c
h
e
ngine
T
e
xt
s
umm
a
r
iza
ti
on
Th
i
s
i
s
a
n
o
p
en
a
c
ces
s
a
r
t
i
c
l
e
u
n
d
e
r
t
h
e
CC
B
Y
-
SA
l
i
ce
n
s
e
.
C
or
r
e
s
pon
din
g
A
u
th
or
:
S
e
if
e
dine
Ka
dr
y
,
De
pa
r
tm
e
nt
of
M
a
thema
ti
c
s
a
nd
C
omput
e
r
S
c
iec
ne
,
F
a
c
ult
y
o
f
S
c
ienc
e
,
B
e
ir
ut
Ar
a
b
Unive
r
is
ty,
B
e
ir
ut
,
L
e
ba
non
.
E
mail:
s
.
ka
dr
y@ba
u
.
e
du.
lb
1.
I
NT
RODU
C
T
I
ON
Na
tur
a
l
langua
ge
p
r
oc
e
s
s
ing
(
NL
P
)
[1
-
3]
is
the
tr
e
nding
a
r
e
a
o
f
r
e
s
e
a
r
c
h
whic
h
a
ll
ows
mac
hines
to
unde
r
s
tand
humans
in
a
s
mar
ter
wa
y.
I
t
a
ll
ows
c
omput
e
r
s
to
a
na
lyze
,
unde
r
s
tand
a
nd
de
r
ive
the
mea
ning
f
r
om
human
ins
tr
uc
ti
ons
.
Us
ing
na
tur
a
l
langua
ge
pr
oc
e
s
s
ing
the
pr
ogr
a
mm
e
r
s
c
ould
or
ga
nize
a
nd
de
s
ign
the
knowle
dge
ba
s
e
d
r
e
pr
e
s
e
ntation
s
to
pe
r
f
or
m
the
tas
ks
li
ke
s
umm
a
r
iza
ti
on
,
s
e
nti
ment
a
na
lys
i
s
,
s
pe
e
c
h
r
e
c
ognit
ion
a
nd
topi
c
s
e
gmenta
ti
on
.
T
he
na
tur
a
l
la
ngua
ge
pr
oc
e
s
s
ing
c
ons
ider
s
the
hier
a
r
c
hy
of
the
l
a
ngua
ge
s
tar
ti
ng
with
s
e
ntenc
e
then
phr
a
s
e
s
a
nd
f
inally
s
e
ntenc
e
s
.
B
y
a
na
lyzing
the
langua
ge
with
their
mea
nings
,
NL
P
pr
oc
e
s
s
e
s
the
s
e
ntenc
e
s
by
c
or
r
e
c
ti
ng
the
gr
a
mm
a
r
,
s
pe
e
c
h
to
text
c
onve
r
s
ion
by
a
uto
matica
ll
y
tr
a
ns
lating
the
langua
ge
s
.
Na
tur
a
l
langua
ge
pr
oc
e
s
s
ing
is
us
e
d
to
a
na
lyze
text
a
nd
a
ll
ows
mac
hine
to
unde
r
s
tand
the
ins
tr
uc
ti
ons
or
c
omm
a
nds
given
by
humans
i
n
the
f
o
r
m
o
f
s
pe
e
c
h.
T
he
na
tur
a
l
langua
ge
pr
oc
e
s
s
ing
is
ge
ne
r
a
ll
y
us
e
d
f
or
a
utom
a
ti
c
mac
hine
a
ns
we
r
ing,
mac
hine
tr
a
ns
lation
[
4
]
a
nd
text
mi
ning
.
I
t
is
c
ons
ider
e
d
a
s
the
toughes
t
r
e
s
e
a
r
c
h
a
r
e
a
to
a
na
lyze
the
human
s
pe
e
c
h
int
o
text
f
o
r
m
ins
pit
e
of
thei
r
pr
onunc
iati
ons
a
nd
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
De
s
ign
of
opti
mal
s
e
ar
c
h
e
ngine
us
ing
tex
t
s
umm
ar
iz
ati
on
thr
ough
ar
ti
fi
c
ial
…
(
K
aus
hik
Se
k
ar
an
)
1269
ne
e
d
to
int
e
r
pr
e
t
their
mea
nings
a
nd
unde
r
s
tand
the
a
s
s
oc
iation
s
be
twe
e
n
wor
ds
to
c
r
e
a
te
a
mea
ning.
T
he
r
e
f
or
e
,
it
c
a
n
be
a
s
a
id
that
thi
s
is
the
toughes
t
r
e
s
e
a
r
c
h
a
r
e
a
f
or
the
pr
ogr
a
mm
e
r
s
to
de
s
ign.
T
he
s
umm
a
r
iza
ti
on
[
5
,
6
]
tec
hnique
is
the
idea
of
da
ta
mi
ning
,
whic
h
f
oc
us
on
the
c
e
ntr
a
li
z
ing
the
idea
of
topi
c
by
c
ons
ider
ing
the
e
nti
r
e
text
i
n
the
doc
ument
given
a
s
a
n
input
.
I
t
c
o
nve
ys
im
por
tant
inf
or
mation
in
the
or
igi
na
l
text
in
s
umm
a
r
ize
d
f
o
r
m.
Automatic
t
e
xt
s
umm
a
r
iza
ti
on
pr
oduc
e
s
the
s
ho
r
ter
view
of
text
with
the
s
e
mantics
whic
h
r
e
duc
e
s
the
r
e
a
di
ng
ti
me.
T
he
c
las
s
if
ica
ti
on
of
text
s
umm
a
r
iza
ti
on
include
s
I
nf
or
mative
a
nd
I
ndica
t
ive.
I
ndica
ti
ve
p
r
e
s
e
nts
the
main
idea
of
text
to
the
us
e
r
whic
h
c
ove
r
s
10
p
e
r
c
e
nt
of
the
main
text.
I
nf
or
mative
s
c
he
me
p
r
e
s
e
nts
the
c
onc
is
e
inf
or
mation
o
f
the
main
text
whic
h
c
ove
r
s
3
0
pe
r
c
e
nt
of
the
o
r
igi
na
l
text
.
T
he
S
e
a
r
c
h
e
ngines
play
a
vit
a
l
r
ole
in
a
huma
n
li
f
e
to
obtain
the
in
f
or
ma
ti
on
in
a
f
r
a
c
ti
on
of
s
e
c
onds
,
f
or
d
if
f
e
r
e
nt
us
e
r
s
que
r
ies
,
ther
e
a
r
e
f
a
mous
s
e
a
r
c
h
e
ngines
li
ke
Google
,
Ya
hoo
,
B
ing
,
As
k
e
tc
whic
h
pe
r
f
or
ms
the
tas
k
of
s
e
a
r
c
h
ope
r
a
ti
ons
with
r
e
s
pe
c
t
to
the
us
e
r
que
r
y
a
nd
d
is
plays
the
r
e
l
e
va
n
t
c
ontent
withi
n
a
f
r
a
c
ti
on
of
s
e
c
onds
.
T
he
s
e
a
r
c
h
e
ngines
a
r
e
de
s
igned
to
dis
play
the
c
ontents
f
or
the
que
r
y
in
the
f
or
mat
of
ti
tl
e
,
UR
L
a
nd
de
s
c
r
ipt
ion
f
or
mat
whic
h
c
ons
is
t
of
s
ubs
c
r
ipt
e
d
text
with
r
e
a
d
m
or
e
opti
on
in
the
de
s
c
r
ipt
ion
pa
r
t
whic
h
a
ll
ows
the
us
e
r
to
r
e
a
d
o
r
know
the
in
f
or
mation
a
bout
that
c
ontent
a
f
te
r
c
li
c
king
on
that
UR
L
.
I
n
our
pa
pe
r
,
we
make
a
n
a
tt
e
mpt
to
de
s
ign
a
s
e
a
r
c
h
e
ngine
whic
h
pr
ovides
the
s
umm
a
r
ize
d
c
ontent
a
bout
the
da
ta
we
a
r
e
looki
ng
ins
tea
d
of
c
li
c
king
on
UR
L
to
know
th
e
in
f
or
matio
n
a
bout
s
e
lec
ted
r
e
s
ult
.
2.
RE
L
AT
E
D
WORK
Atif
Kha
n
e
t
al
.,
[
7
]
p
r
e
s
e
nts
a
s
e
mantic
gr
a
ph
a
ppr
oa
c
h
with
im
pr
ove
d
r
a
nking
a
lgor
it
hm
f
or
a
bs
tr
a
c
ti
ve
s
umm
a
r
iza
ti
on
of
mul
ti
-
doc
uments
.
T
he
pr
e
dica
te
a
r
gument
s
tr
uc
tur
e
s
a
r
e
the
gr
a
p
h
node
s
c
ons
tr
uc
ted
f
r
om
the
s
our
c
e
doc
uments
c
oll
e
c
te
d
f
r
om
mul
ti
ple
s
our
c
e
s
whe
r
e
the
pr
e
dica
te
a
r
gument
s
tr
uc
tur
e
s
(
P
ASs
)
c
a
n
be
r
e
f
e
r
r
e
d
a
s
the
s
e
mantic
s
tr
uc
tur
e
of
s
e
ntenc
e
,
whic
h
is
a
utom
a
ti
c
a
ll
y
ident
if
ied
by
us
ing
s
e
mantic
r
ole
labe
li
ng;
while
gr
a
ph
e
dge
s
r
e
pr
e
s
e
nt
s
im
il
a
r
it
y
we
ight
,
whic
h
is
c
omput
e
d
f
r
o
m
P
ASs
s
e
mantic
s
im
il
a
r
it
y.
H
e
c
onduc
ted
e
xpe
r
im
e
nt
us
ing
DU
C
-
2002,
a
s
tanda
r
d
da
tas
e
t
f
or
d
oc
ument
s
umm
a
r
iza
ti
on.
E
xpe
r
im
e
ntal
r
e
s
ult
s
pr
e
s
e
nt
the
s
upe
r
ior
pe
r
f
or
manc
e
than
o
ther
s
umm
a
r
iza
ti
on
a
ppr
oa
c
he
s
.
S
a
mr
a
t
B
a
ba
r
[
8]
,
pr
opos
e
d
a
n
a
utom
a
ti
c
s
umm
a
r
i
z
a
ti
on
text
f
or
a
bs
or
bing
the
r
e
leva
nt
c
ontent
f
r
om
the
number
of
doc
uments
a
va
il
a
ble.
T
he
a
uthor
di
s
c
us
s
e
d
a
bout
the
im
por
tanc
e
of
a
utom
a
ti
c
s
umm
a
r
iza
ti
on
with
the
ba
s
ic
de
f
ini
t
ions
of
text
s
umm
a
r
iza
ti
on.
He
dis
c
us
s
e
d
a
bout
the
va
r
ious
r
e
s
e
a
r
c
h
a
r
e
a
s
f
or
c
ons
ider
ing
the
a
utom
a
ti
c
s
umm
a
r
iza
ti
on
l
ike
m
a
c
hine
lea
r
ning,
Na
tur
a
l
langua
ge
pr
oc
e
s
s
ing
.
T
h
e
a
uthor
e
xplaine
d
the
im
por
tant
e
xtr
a
c
ti
on
a
nd
di
f
f
e
r
e
nc
e
s
be
twe
e
n
both
e
xtr
a
c
ti
ve
a
nd
a
bs
tr
a
c
ti
ve
s
umm
a
r
iza
ti
ons
with
two
gr
oups
o
f
text
s
umm
a
r
iza
ti
on
na
mely
ind
ica
ti
ve
a
nd
inductive
s
umm
a
r
iza
ti
ons
[
9
]
.
De
e
pa
li
K.
Ga
ikwa
d,
e
t
al
.,
[
10]
,
p
r
opos
e
d
the
im
p
or
tanc
e
of
text
s
umm
a
r
iza
ti
on
[
11]
a
s
a
br
a
nc
h
of
na
tur
a
l
langua
ge
p
r
oc
e
s
s
ing
with
the
a
bs
tr
a
c
t
pr
e
s
e
ntation
of
in
f
or
ma
ti
on
a
va
il
a
ble
in
the
int
e
r
n
e
t.
T
he
y
pr
e
s
e
nted
a
bout
the
de
tails
of
both
the
e
xtr
a
c
ti
ve
a
nd
a
bs
tr
a
c
ti
ve
a
ppr
oa
c
he
s
a
long
with
the
tec
hniques
us
e
d,
it
s
pe
r
f
or
manc
e
a
c
hieve
d,
a
long
with
a
dva
ntage
s
a
nd
dis
a
dva
ntage
s
of
e
a
c
h
a
ppr
oa
c
h.
T
he
y
pr
e
s
e
nted
a
ll
the
de
tails
of
both
the
e
xtr
a
c
ti
ve
a
nd
a
bs
tr
a
c
ti
ve
a
ppr
oa
c
he
s
a
long
with
the
tec
hniques
u
s
e
d,
it
s
pe
r
f
or
manc
e
a
c
hieve
d,
a
long
with
a
dva
ntage
s
a
nd
dis
a
dva
ntage
s
of
e
a
c
h
a
ppr
oa
c
h.
T
e
xt
s
umm
a
r
iza
ti
on
ha
s
it
s
im
por
tanc
e
in
both
c
omm
e
r
c
ial
a
s
we
ll
a
s
r
e
s
e
a
r
c
h
c
omm
unit
y
.
As
a
bs
tr
a
c
ti
ve
s
umm
a
r
iza
ti
on
r
e
quir
e
s
mo
r
e
lea
r
ning
a
nd
r
e
a
s
oning,
it
is
bit
c
ompl
e
x
then
e
xtr
a
c
ti
ve
a
ppr
oa
c
h
but
,
a
bs
tr
a
c
ti
ve
s
umm
a
r
iza
ti
on
pr
ovid
e
s
mor
e
mea
ningf
ul
a
nd
a
ppr
opr
iate
s
umm
a
r
y
c
ompar
e
to
e
xtr
a
c
ti
ve
.
T
he
y
pr
e
s
e
nted
va
r
ious
types
of
text
s
umm
a
r
iza
ti
on
tec
hniques
with
va
r
ious
f
o
r
ms
of
a
ppr
oa
c
he
s
.
R
a
s
hmi
Kur
mi
,
e
t
al
.,
[
12
]
im
pleme
nted
a
metho
d
to
r
e
duc
e
c
os
t
a
nd
ti
me.
T
he
method
wor
ks
on
the
pr
incipa
l
o
f
maximal
mar
ginal
s
igni
f
ica
nc
e
be
twe
e
n
wor
d
a
nd
s
e
ntenc
e
.
T
he
maximal
mar
gina
l
s
igni
f
ica
nc
e
is
de
c
ided
by
the
unit
s
tep
f
unc
ti
on
us
e
d
whic
h
c
ontains
da
taba
s
e
with
of
us
e
les
s
wor
ds
or
wor
ds
whic
h
c
a
n’
t
im
pa
c
t
the
mea
ning
of
doc
ument
a
r
e
maintaine
d.
T
he
input
doc
ument
is
tr
a
ve
r
s
e
d
a
nd
wor
ds
c
ontaining
in
the
da
taba
s
e
a
r
e
e
li
m
inate
d
f
r
o
m
ini
t
i
a
l
pos
it
i
on
of
the
s
e
ntenc
e
to
the
e
nd
.
L
uc
ia
no
C
a
br
a
l
e
l
a
l
.,
[
13
]
pr
opos
e
d
an
a
ut
omatic
s
umm
a
r
iza
ti
on
method
whic
h
dis
plays
the
s
umm
a
r
ies
of
ne
ws
pa
ge
s
on
Andr
o
id
-
e
na
bled
mobi
le
de
vice
s
to
the
dif
f
e
r
e
nt
f
or
ms
of
us
e
r
s
.
T
he
method
c
ontain
two
ba
s
ic
a
ppr
oa
c
he
s
whe
r
e
the
f
ir
s
t
a
ppr
oa
c
h
pr
e
pr
oc
e
s
s
e
s
we
b
pa
ge
s
by
r
e
f
or
matti
ng
or
a
da
pti
ng
them
to
a
mor
e
a
ppr
op
r
iate
wa
y
of
view
ing
on
s
mall
s
c
r
e
e
ns
,
wi
thout
a
lt
e
r
ing
the
or
igi
na
l
c
ontent
.
S
e
c
ond
a
ppr
oa
c
h
pr
e
s
e
nts
the
mos
t
im
por
tant
a
nd
r
e
leva
nt
c
ontent
of
a
pa
ge
to
the
us
e
r
,
with
r
e
s
pe
c
t
to
the
n
e
e
d
f
or
gr
a
s
ping
the
ba
s
ic
inf
or
mation
.
3.
AR
CHI
T
E
C
T
UR
E
T
he
Ar
c
hit
e
c
tur
e
f
or
the
pr
opos
e
d
f
r
a
mew
or
k
is
g
iven
in
F
igur
e
1
.
T
he
a
r
c
hit
e
c
tur
e
o
f
the
pr
opos
e
d
wor
k
s
tar
t
with
a
gr
a
phica
l
us
e
r
int
e
r
f
a
c
e
de
s
ign
whic
h
a
ll
ows
us
e
r
to
input
his
que
r
y
a
s
in
a
ge
ne
r
a
l
s
e
a
r
c
h
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
1693
-
6930
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
,
Vol.
1
8
,
No
.
3
,
J
une
2020:
1
2
6
8
-
1
2
7
4
1270
e
ngine,
whe
r
e
the
s
e
a
r
c
h
pr
ogr
a
m
f
e
tche
s
a
nd
e
xt
r
a
c
ts
da
ta
f
r
om
mul
ti
ple
s
e
a
r
c
h
e
ngines
li
ke
Goog
le,
B
ing
a
nd
Ya
hoo
us
ing
the
s
pider
a
nd
r
obo
t
p
r
ogr
a
m
s
f
or
the
given
us
e
r
que
r
y
a
nd
s
tor
e
s
in
the
d
a
taba
s
e
.
T
he
e
xtr
a
c
tor
pr
ogr
a
m
in
the
GU
I
pe
r
f
o
r
ms
the
a
utom
a
ti
c
ke
ywor
d
e
xtr
a
c
ti
on
[
14
-
17]
f
o
r
the
o
btaine
d
da
taba
s
e
to
pr
e
s
e
nt
the
vis
ibi
li
ty
of
f
r
e
que
nt
s
e
a
r
c
h
ter
ms
f
or
the
us
e
r
whic
h
a
ll
ows
us
e
r
to
r
e
f
r
a
me
t
he
que
r
y
he
would
li
ke
to
r
e
que
s
t.
T
he
o
btaine
d
da
ta
is
pr
e
s
e
nt
e
d
in
the
f
or
m
of
t
i
tl
e
,
UR
L
a
nd
d
e
s
c
r
ipt
ion
f
or
mat
a
s
in
ge
ne
r
a
l
s
e
a
r
c
h
e
ngines
.
T
he
dis
play
of
r
e
s
ult
s
is
pr
e
s
e
nted
with
the
li
nk
to
view
the
ove
r
view
of
c
ontent
whic
h
pr
e
s
e
nts
the
s
umm
a
r
ize
d
text
of
the
s
e
le
c
ted
r
e
s
ult
without
na
vi
ga
ti
ng
to
the
pa
ge
of
t
he
UR
L
.
T
he
c
ontent
s
umm
a
r
iza
ti
on
is
pe
r
f
o
r
med
by
th
e
s
umm
a
r
iza
ti
on
tec
hnique
[
18
-
22]
of
na
tu
r
a
l
l
a
ngua
ge
pr
oc
e
s
s
ing
[
23
-
25]
.
I
n
f
utur
e
,
the
s
c
ope
of
the
NL
P
c
ould
be
e
xtende
d
towa
r
ds
c
loud
ba
s
e
d
[
26
-
30]
pr
oc
e
s
s
ing
of
AI
tec
hniques
.
F
ig
ur
e
1.
T
he
f
r
a
mew
or
k
f
o
r
the
s
e
a
r
c
h
e
ngine
de
s
ign
4.
AL
GO
RI
T
HM
Input: D Database =
{
1
,
2
.
.
.
}
∀
{
=
1
,
2
…
}
Output: Result Display with Summarized Text
Parameters: Swi =
Array of Stop Words
attsi = Description attribute
KWi = Words stemmed from
Description attribute
WFi = Word frequencies after
stemming
Sti = Summarized text of
selected result
Method:
-
Consider the list of stop words to be removed from description attribute
Swi= {
};
-
Perform the following operations, for each tuple Ti in Database D,
Consider the Description attributes in D and name it as ‘attsi’
-
Stem
the
stop
words
from
the
text
of
de
scription
attribute
and
separate
the
keywords
as
follows:
Wk=Separate (attsi, Swi)
;
-
The
word
frequency
(freq
(w)),
word
deg
ree
(deg
(w))
are
considered
for
calculat
ing
the
word score WS that is, the ratio of degree to frequency (deg(w)/freq(w)).
a.
Wordf(word)=words(D,attsi,WK)
b.
Wordd(word)=words(D, attsi)
c.
WS=
Wordf/Wordd
-
The
frequency
(f)
is
considered
as
t
he
highest
score
receive
d
for
the
word
s
after
stemming
the description attribute.
f=highest_score (WS)
-
Gener
ate the list of words extracted along with frequency
in descending order.
-
Select
the
result
fo
r
display,
feed
the
url
to
summarizer
to
generate
the
overview
of
the
text.
Result= Summarize(url)
(Or) Res=Sti(attsi)
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
De
s
ign
of
opti
mal
s
e
ar
c
h
e
ngine
us
ing
tex
t
s
umm
ar
iz
ati
on
thr
ough
ar
ti
fi
c
ial
…
(
K
aus
hik
Se
k
ar
an
)
1271
5.
RE
S
UL
T
S
L
ot
of
AP
I
’
s
a
nd
tool
s
[
20
]
ther
e
to
a
na
lyze
the
c
ur
r
e
nt
texts
a
nd
a
ls
o
the
c
ontextua
l
da
taba
s
e
s
,
ye
t
a
nother
s
igni
f
ica
nt
a
c
hieve
ment
in
the
f
ield
o
f
NL
P
,
i
ntellexe
r
API
is
one
o
f
the
p
r
omi
ne
nt
pla
tf
or
m
f
or
ge
tt
ing
the
s
olut
ions
on
text
mi
ning.
I
t
c
ould
be
e
a
s
il
y
incor
por
a
ted
with
the
pr
ogr
a
mm
ing
langua
ge
s
li
ke
python,
php
a
nd
java
whic
h
he
lp
de
ve
loper
s
to
pr
e
s
e
nt
the
innovative
a
nd
ins
tant
s
olut
i
ons
f
or
the
inf
or
mation
s
e
a
r
c
h,
e
xt
r
a
c
ti
on
a
nd
s
e
mantic
modeling.
W
e
ha
ve
a
na
lyze
d
us
ing
thi
s
AP
I
a
nd
c
a
me
up
w
it
h
the
r
e
s
ult
s
a
s
s
hown
in
F
ig
ur
e
2
.
T
he
pr
opos
e
d
wor
k
is
p
r
e
s
e
nted
with
the
r
e
s
ult
s
a
s
f
oll
ows
:
F
igur
e
2.
T
he
r
e
s
ult
s
f
or
the
inpu
t
que
r
y
is
c
oll
e
c
ted
f
r
om
s
e
a
r
c
h
e
ngines
li
ke
Google
,
Ya
hoo
a
nd
B
ing
us
ing
the
c
r
a
wle
r
pr
og
r
a
ms
T
he
us
e
r
int
e
r
f
a
c
e
i
s
de
s
igned
a
s
a
s
im
ple
s
e
a
r
c
h
e
ngine
whic
h
c
oll
e
c
ts
da
ta
us
ing
we
b
c
r
a
wle
r
s
a
nd
f
e
tche
s
inf
or
mation
f
r
om
the
popular
s
e
a
r
c
h
e
ngines
li
ke
Google
,
Ya
hoo,
B
ing
s
e
a
r
c
h
e
ngines
in
the
f
or
mat
of
<
T
i
tl
e
,
Ur
l
,
De
s
c
r
ipt
ion>
a
nd
s
tor
e
d
in
the
da
ta
ba
s
e
in
the
o
r
de
r
t
he
y
ha
ve
f
e
tche
d.
T
he
da
ta
s
c
r
a
pe
d
f
r
om
the
da
taba
s
e
may
c
ontain
s
ome
mi
s
s
ing,
ir
r
e
leva
nt
va
lues
a
nd
ther
e
f
or
e
it
is
f
il
ter
e
d
a
nd
pr
un
e
d
f
r
om
the
da
taba
s
e
whic
h
c
ontains
only
the
r
e
leva
nt
i
nf
or
mation
a
nd
maintaine
d
with
the
index
a
s
s
hown
in
F
igur
e
3
.
Th
e
da
ta
c
oll
e
c
ted
f
r
o
m
the
c
r
a
wle
r
p
r
ogr
a
m
is
pr
une
d
a
nd
de
s
c
r
ipt
ion
pa
r
t
c
oll
e
c
ted
is
f
ur
ther
s
pli
tt
e
d
int
o
s
e
t
o
f
ke
ywor
ds
f
o
r
pr
opa
ga
ti
ng
ge
ne
r
a
l
s
e
a
r
c
h
ter
ms
[
17]
to
pr
ovide
c
onve
nienc
e
f
o
r
t
he
us
e
r
,
while
s
e
a
r
c
hing
f
or
the
c
ontent
.
T
he
r
e
s
ult
s
a
r
e
pr
e
s
e
nted
to
the
us
e
r
in
a
tabula
r
f
o
r
m
whe
r
e
the
us
e
r
s
e
lec
ts
the
r
e
s
ult
s
to
ge
t
ove
r
view
of
the
c
ontent
he
de
s
ir
e
d
to
view
a
s
s
hown
in
F
igu
r
e
4
.
T
he
ge
ne
r
a
li
z
e
d
s
umm
a
r
y
of
the
r
e
s
ult
s
e
lec
ted
by
the
us
e
r
is
pr
e
s
e
nted
a
s
a
doc
ument
s
umm
a
r
y
whic
h
c
ontains
ove
r
view
o
f
t
he
topi
c
on
whic
h
their
we
bs
it
e
is
de
s
igned
a
bout.
I
t
pr
e
s
e
nts
the
ove
r
view
inf
or
mation
f
or
whic
h
the
us
e
r
ha
s
s
e
lec
ted
a
s
s
hown
in
F
igur
e
5
.
F
igur
e
3.
T
he
R
e
s
ult
s
f
e
tche
d
us
ing
c
r
a
wle
r
pr
og
r
a
m
s
pli
ts
the
e
nti
r
e
text
in
the
de
s
c
r
ipt
ion
pa
r
t
int
o
s
e
t
o
f
ke
ywor
ds
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
1693
-
6930
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
,
Vol.
1
8
,
No
.
3
,
J
une
2020:
1
2
6
8
-
1
2
7
4
1272
F
igur
e
4.
Dis
play
of
r
e
s
ult
s
pr
e
s
e
nti
ng
or
the
us
e
r
t
o
s
e
lec
t
the
s
umm
a
r
ize
d
text
F
igur
e
5.
Ge
ne
r
a
tes
the
s
umm
a
r
y
o
f
the
c
ontent
f
o
r
the
r
e
s
ult
s
e
lec
ted
by
the
us
e
r
6.
P
E
RF
ORM
AN
CE
AN
AL
YSI
S
T
he
pe
r
f
or
manc
e
e
va
luation
o
f
the
s
e
a
r
c
h
E
ngi
ne
is
a
s
s
e
s
s
e
d
by
c
ons
ider
ing
the
da
taba
s
e
s
ize
,
e
xe
c
uti
on
ti
me
a
nd
r
e
s
pons
e
ti
me
f
o
r
the
numb
e
r
que
r
ies
e
xe
c
uted
whe
r
e
the
r
e
s
pons
e
ti
me
is
dir
e
c
tl
y
pr
opor
ti
ona
l
to
the
ne
two
r
k
late
nc
y
a
nd
ba
ndwidth
a
s
s
hown
in
F
ig
u
r
e
6.
F
igur
e
6.
P
e
r
f
or
manc
e
e
va
luation
o
f
s
e
a
r
c
h
e
ngine
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
De
s
ign
of
opti
mal
s
e
ar
c
h
e
ngine
us
ing
tex
t
s
umm
ar
iz
ati
on
thr
ough
ar
ti
fi
c
ial
…
(
K
aus
hik
Se
k
ar
an
)
1273
7.
CONC
L
USI
ON
Th
e
s
e
a
r
c
h
e
ngines
play
a
vit
a
l
r
ole
in
inf
or
ma
ti
on
r
e
tr
ieva
l
pr
oc
e
s
s
f
or
a
ny
f
o
r
mat
of
que
r
ies
pr
e
s
e
nted
by
us
e
r
.
T
hus
it
is
c
r
uc
ial
s
tep
in
a
s
e
a
r
c
h
e
ngine
de
s
ign
to
int
e
r
pr
e
t
the
que
r
y
a
nd
s
houl
d
pr
e
s
e
nt
the
r
e
s
ult
s
in
a
n
e
f
f
e
c
ti
ve
manne
r
s
o
that
us
e
r
s
ho
uld
vis
ua
li
z
e
a
gr
e
a
t
look
a
nd
f
e
e
l
a
bout
the
i
nter
f
a
c
e
.
I
n
or
de
r
to
pr
ov
ide
mor
e
f
lexible
us
a
ge
of
s
e
a
r
c
h
e
ngine,
the
s
e
a
r
c
h
r
e
s
ult
s
a
r
e
f
ur
ther
e
labor
a
ted
by
pr
ovidi
ng
the
s
mall
s
umm
a
r
y
c
ontent
o
f
the
r
e
s
ult
they
a
r
e
looki
ng
by
c
li
c
king
on
the
s
e
lec
t
op
ti
on
f
or
e
a
c
h
r
e
s
ult
dis
playe
d
r
a
the
r
than
na
vigating
to
the
we
bpa
ge
by
c
li
c
king
the
u
r
l
in
the
r
e
s
ult
pa
ge
.
I
n
o
ur
pa
pe
r
,
we
pr
opos
e
d
the
f
r
a
mew
or
k
f
o
r
de
s
igni
ng
the
s
e
a
r
c
h
e
ngine
us
ing
a
utom
a
ti
c
ke
ywor
d
e
xtr
a
c
ti
on
a
nd
s
umm
a
r
iza
ti
on
tec
hniques
.
T
he
pe
r
f
o
r
manc
e
of
o
ur
s
e
a
r
c
h
e
ngine
is
e
va
luate
d
with
r
e
s
pe
c
t
to
the
da
taba
s
e
s
ize
,
R
e
s
pons
e
ti
me
a
nd
E
xe
c
uti
on
ti
me/
thr
oughp
ut
ti
me.
RE
F
E
RE
NC
E
S
[1
]
Co
l
l
o
b
ert
,
Ro
n
a
n
,
J
as
o
n
W
e
s
t
o
n
,
L
éo
n
Bo
t
t
o
u
,
Mi
ch
ae
l
K
arl
e
n
,
K
o
r
ay
K
av
u
k
c
u
o
g
l
u
,
an
d
Pav
el
K
u
k
s
a,
"
N
at
u
ra
l
l
an
g
u
a
g
e
p
r
o
ces
s
i
n
g
(a
l
mo
s
t
)
fro
m
s
crat
c
h
,
"
Jo
u
r
n
a
l
o
f
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
r
es
e
a
r
c
h
,
v
o
l
.
1
2
,
n
o
.
1
2
,
p
p
.
2
4
9
3
-
2
5
3
7
,
March
2
0
1
1
.
[2
]
G
ro
s
z,
Barb
ara
J
.
,
K
aren
Sp
ar
c
k
J
o
n
e
s
,
an
d
B
o
n
n
i
e
L
y
n
n
W
e
b
b
er,
"
Read
i
n
g
s
i
n
n
a
t
u
ra
l
l
an
g
u
a
g
e
p
r
o
ces
s
i
n
g
,
"
M
o
r
g
a
n
Ka
u
f
m
a
n
n
,
A
u
g
u
s
t
1
9
8
6
.
[3
]
Ch
o
w
d
h
u
r
y
,
G
o
b
i
n
d
a
G
.
,
"
N
at
u
ral
l
an
g
u
a
g
e
p
ro
ces
s
i
n
g
,
"
A
n
n
u
a
l
r
evi
ew
o
f
i
n
f
o
r
m
a
t
i
o
n
s
c
i
en
ce
a
n
d
t
ech
n
o
l
o
g
y,
v
o
l
.
3
7
,
n
o
.
1
,
p
p
.
5
1
-
8
9
,
2
0
0
3
.
[4
]
Sark
ar,
K
ama
l
,
Mi
t
a
N
a
s
i
p
u
r
i
,
an
d
Su
ra
n
j
a
n
G
h
o
s
e,
"
U
s
i
n
g
mac
h
i
n
e
l
ear
n
i
n
g
fo
r
me
d
i
ca
l
d
o
c
u
men
t
s
u
mm
ar
i
zat
i
o
n
,
"
In
t
er
n
a
t
i
o
n
a
l
Jo
u
r
n
a
l
o
f
D
a
t
a
b
a
s
e
Th
e
o
r
y
a
n
d
A
p
p
l
i
ca
t
i
o
n
,
v
o
l
.
4
,
n
o
.
1
,
p
p
.
3
1
-
4
8
,
March
2
0
1
1
.
[5
]
Pad
maL
ah
ari
,
E
.
,
D
.
V
.
N
.
Si
v
a
K
u
mar,
an
d
S
h
i
v
a
P
ras
ad
,
"
A
u
t
o
ma
t
i
c
t
e
x
t
s
u
mmar
i
zat
i
o
n
w
i
t
h
s
t
at
i
s
t
i
ca
l
an
d
l
i
n
g
u
i
s
t
i
c
feat
u
re
s
u
s
i
n
g
s
u
cc
es
s
i
v
e
t
h
re
s
h
o
l
d
s
,
"
2
0
1
4
IE
E
E
In
t
e
r
n
a
t
i
o
n
a
l
Co
n
f
e
r
en
ce
o
n
A
d
v
a
n
ce
d
Co
m
m
u
n
i
c
a
t
i
o
n
s
,
Co
n
t
r
o
l
a
n
d
Co
m
p
u
t
i
n
g
Tec
h
n
o
l
o
g
i
es
,
Raman
at
h
a
p
u
ram,
p
p
.
1
5
1
9
-
1
5
2
4
,
2
0
1
4
.
[6
]
S
aran
y
am
o
l
,
C.
S.
,
an
d
L
.
Si
n
d
h
u
,
"
A
s
u
r
v
ey
o
n
a
u
t
o
ma
t
i
c
t
e
x
t
s
u
mmar
i
zat
i
o
n
,
"
In
t
er
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
Co
m
p
u
t
e
r
S
ci
e
n
ce
a
n
d
In
f
o
r
m
a
t
i
o
n
Tech
n
o
l
o
g
i
es
,
v
o
l
.
5
,
n
o
.
6
,
p
p
.
7
8
8
9
-
7
8
9
3
,
2
0
1
4
.
[7
]
Cab
ral
,
L
u
ci
a
n
o
,
R
i
n
a
l
d
o
L
i
ma,
Rafael
L
i
n
s
,
Man
o
el
N
et
o
,
Rafael
Ferrei
ra,
S
t
ev
e
n
Si
m
s
k
e
,
an
d
Marce
l
o
R
i
s
s
,
"
A
u
t
o
mat
i
c
Su
mmari
za
t
i
o
n
o
f
N
e
w
s
A
r
t
i
c
l
es
i
n
Mo
b
i
l
e
D
e
v
i
ce
s
,
"
2
0
1
5
F
o
u
r
t
ee
n
t
h
M
exi
c
a
n
In
t
er
n
a
t
i
o
n
a
l
Co
n
f
er
e
n
ce
o
n
A
r
t
i
f
i
c
i
a
l
In
t
e
l
l
i
g
e
n
ce
(M
IC
A
I)
,
Cu
ern
a
v
a
ca,
p
p
.
8
-
1
3
,
2
0
1
5
.
[8
]
Imam,
Ib
rah
i
m,
N
i
h
a
l
N
o
u
n
o
u
,
A
l
aa
H
amo
u
d
a
,
H
e
b
at
A
l
l
ah
,
a
n
d
A
b
d
u
l
K
h
a
l
ek
,
"
Q
u
ery
Ba
s
e
d
A
ra
b
i
c
T
ex
t
Su
mmari
za
t
i
o
n
,
"
In
t
e
r
n
a
t
i
o
n
a
l
Jo
u
r
n
a
l
o
f
Co
m
p
u
t
e
r
S
c
i
en
ce
a
n
d
Tech
n
o
l
o
g
y
,
v
o
l
.
4
,
n
o
.
2
,
p
p
.
3
5
-
3
9
,
J
u
n
e
2
0
1
3
.
[9
]
Reev
e,
L
aw
ren
ce
H
.
,
H
y
o
i
l
H
an
,
Say
a
V
.
N
ag
o
r
i
,
J
o
n
a
t
h
a
n
C.
Y
an
g
,
T
amara
A
.
Sch
w
i
mmer,
an
d
A
ri
D
.
Bro
o
k
s
,
"
Co
n
cep
t
freq
u
e
n
cy
d
i
s
t
ri
b
u
t
i
o
n
i
n
b
i
o
med
i
cal
t
e
x
t
s
u
mmari
zat
i
o
n
,
"
P
r
o
cee
d
i
n
g
s
o
f
t
h
e
1
5
th
A
CM
i
n
t
e
r
n
a
t
i
o
n
a
l
co
n
f
er
e
n
ce
o
n
I
n
f
o
r
m
a
t
i
o
n
a
n
d
k
n
o
w
l
ed
g
e
m
a
n
a
g
em
e
n
t
,
p
p
.
6
0
4
-
6
1
1
,
2
0
0
6
.
[1
0
]
K
h
a
n
,
A
t
i
f,
an
d
N
a
o
mi
e
Sal
i
m
,
"
A
rev
i
ew
o
n
ab
s
t
rac
t
i
v
e
s
u
mmari
za
t
i
o
n
met
h
o
d
s
,
"
Jo
u
r
n
a
l
o
f
Th
e
o
r
e
t
i
c
a
l
a
n
d
A
p
p
l
i
ed
In
f
o
r
m
a
t
i
o
n
Tech
n
o
l
o
g
y
,
v
o
l
.
5
9
,
n
o
.
1
,
p
p
.
6
4
-
7
2
,
J
an
u
ary
2
0
1
4
.
[1
1
]
G
ai
k
w
a
d
,
D
eep
al
i
K
.
,
an
d
C.
N
amrat
a
Mah
e
n
d
er,
"
A
rev
i
ew
p
a
p
er
o
n
t
e
x
t
s
u
mmari
za
t
i
o
n
,
"
In
t
e
r
n
a
t
i
o
n
a
l
Jo
u
r
n
a
l
o
f
A
d
v
a
n
ce
d
R
es
e
a
r
c
h
i
n
Co
m
p
u
t
er
a
n
d
Co
m
m
u
n
i
c
a
t
i
o
n
E
n
g
i
n
ee
r
i
n
g
,
v
o
l
.
5
,
n
o
.
3
,
p
p
.
1
5
4
-
1
6
0
,
March
2
0
1
6
.
[1
2
]
V
i
s
h
a
l
G
u
p
t
a,
“A
Su
r
v
ey
o
f
Recen
t
K
ey
w
o
r
d
s
an
d
T
o
p
i
c
E
x
t
rac
t
i
o
n
S
y
s
t
ems
fo
r
In
d
i
a
n
L
an
g
u
ag
e
s
,
”
In
t
e
r
n
a
t
i
o
n
a
l
Jo
u
r
n
a
l
o
f
E
n
g
i
n
ee
r
i
n
g
Tr
e
n
d
s
a
n
d
Tech
n
o
l
o
g
y
(IJE
TT)
,
v
o
l
.
6
,
n
o
.
6
,
p
p
.
3
4
0
-
3
4
3
,
D
ecemb
er
2
0
1
3
.
[1
3
]
G
u
p
t
a
V
.
A
n
d
L
eh
al
G
.
S.
,
“A
Su
r
v
ey
o
f
T
e
x
t
S
u
m
ma
ri
zat
i
o
n
E
x
t
rac
t
i
v
e
T
ec
h
n
i
q
u
es
,
”
J
o
u
r
n
a
l
o
f
E
m
e
r
g
i
n
g
Tech
n
o
l
o
g
i
es
i
n
W
e
b
In
t
e
l
l
i
g
e
n
ce
,
v
o
l
.
2
,
n
o
.
3
,
p
p
.
2
5
8
-
268
,
A
u
g
u
s
t
2
0
1
0
.
[1
4
]
Cab
ral
,
L
u
ci
a
n
o
,
R
i
n
a
l
d
o
L
i
ma,
Rafael
L
i
n
s
,
Man
o
el
N
et
o
,
Rafael
Ferrei
ra,
S
t
ev
e
n
Si
m
s
k
e,
an
d
Marce
l
o
R
i
s
s
,
"
A
u
t
o
mat
i
c
Su
mmari
za
t
i
o
n
o
f
N
e
w
s
A
r
t
i
c
l
es
i
n
Mo
b
i
l
e
D
e
v
i
ce
s
,
"
2
0
1
5
F
o
u
r
t
ee
n
t
h
M
exi
c
a
n
In
t
er
n
a
t
i
o
n
a
l
Co
n
f
er
e
n
ce
o
n
A
r
t
i
f
i
c
i
a
l
In
t
e
l
l
i
g
e
n
ce
(M
IC
A
I)
,
Cu
ern
a
v
a
ca,
p
p
.
8
-
1
3
,
2
0
1
5
.
[1
5
]
G
u
p
t
a
V
i
s
h
al
,
“A
S
u
rv
e
y
o
f
T
e
x
t
Su
mmar
i
zer
s
fo
r
I
n
d
i
an
L
an
g
u
ag
e
s
an
d
C
o
mp
ar
i
s
o
n
o
f
t
h
e
i
r
Perfo
rma
n
ce”
Jo
u
r
n
a
l
o
f
E
m
er
g
i
n
g
Tech
n
o
l
o
g
i
e
s
i
n
W
eb
I
n
t
e
l
l
i
g
e
n
ce
,
v
o
l
.
5
,
n
o
.
4
,
p
p
.
3
6
1
-
3
6
6
,
N
o
v
em
b
er
2
0
1
3
.
[1
6
]
Y
o
u
n
g
,
T
o
m,
D
ev
ama
n
y
u
H
azari
k
a,
So
u
j
a
n
y
a
P
o
ri
a,
an
d
E
ri
k
Camb
r
i
a,
"
Recen
t
t
re
n
d
s
i
n
d
ee
p
l
earn
i
n
g
b
a
s
e
d
n
at
u
ral
l
a
n
g
u
ag
e
p
ro
ce
s
s
i
n
g
,
"
IE
E
E
C
o
m
p
u
t
a
t
i
o
n
a
l
In
t
el
l
i
g
en
ce
M
a
g
a
z
i
n
e
,
v
o
l
.
1
3
,
n
o
.
3
,
p
p
.
5
5
-
7
5
,
A
u
g
.
2
0
1
8
.
[1
7
]
V
i
s
w
a
n
at
h
Meg
h
a
n
a,
“T
h
es
i
s
:
O
n
t
o
l
o
g
y
-
Bas
e
d
A
u
t
o
ma
t
i
c
T
e
x
t
S
u
mmari
za
t
i
o
n
,
”
M.
Sc
T
h
es
i
s
,
V
i
s
h
w
e
s
h
w
ara
i
ah
In
s
t
i
t
u
t
e
o
f
T
ec
h
n
o
l
o
g
y
,
In
d
i
a,
2
0
0
9
.
[1
8
]
Man
n
i
n
g
,
Ch
ri
s
t
o
p
h
er,
Mi
h
a
i
Su
rd
ea
n
u
,
J
o
h
n
Bau
er,
J
en
n
y
Fi
n
k
e
l
,
St
e
v
en
Bet
h
ard
,
an
d
D
a
v
i
d
McCl
o
s
k
y
,
"
T
h
e
St
an
f
o
rd
Co
re
N
L
P
n
at
u
ra
l
l
an
g
u
a
g
e
p
ro
ce
s
s
i
n
g
t
o
o
l
k
i
t
,
"
In
P
r
o
ceed
i
n
g
s
o
f
5
2
n
d
a
n
n
u
a
l
m
eet
i
n
g
o
f
t
h
e
a
s
s
o
ci
a
t
i
o
n
f
o
r
co
m
p
u
t
a
t
i
o
n
a
l
l
i
n
g
u
i
s
t
i
c
s
:
s
ys
t
em
d
e
m
o
n
s
t
r
a
t
i
o
n
s
,
p
p
.
5
5
-
6
0
,
J
u
n
e
2
0
1
4
.
[1
9
]
K
h
a
n
,
A
t
i
f,
N
a
o
mi
e
Sa
l
i
m,
H
al
eem
Farma
n
,
Mu
ra
d
K
h
an
,
Bi
l
al
J
a
n
,
A
w
a
i
s
A
h
m
ad
,
Imran
A
h
med
,
an
d
A
n
an
d
Pau
l
,
"
A
b
s
t
ract
i
v
e
t
ex
t
s
u
mmar
i
zat
i
o
n
b
a
s
e
d
o
n
i
m
p
r
o
v
e
d
s
ema
n
t
i
c
g
rap
h
ap
p
ro
a
c
h
,
"
In
t
er
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
P
a
r
a
l
l
e
l
P
r
o
g
r
a
m
m
i
n
g
,
v
o
l
.
4
6
,
n
o
.
5
,
p
p
.
9
9
2
-
1
0
1
6
,
Fe
b
ru
ar
y
2
0
1
8
.
[2
0
]
Bab
ar,
Samrat
,
an
d
M.
T
ech
-
C
s
e,
"
T
ex
t
s
u
mmar
i
zat
i
o
n
:
A
n
o
v
erv
i
ew
,
"
Res
each
G
at
e,
2
0
1
3
.
[o
n
l
i
n
e].
A
v
a
i
l
ab
l
e
fro
m:
h
t
t
p
s
:
/
/
w
w
w
.
res
earc
h
g
a
t
e.
n
et
/
p
u
b
l
i
c
at
i
o
n
/
2
5
7
9
4
7
5
2
8
_
T
e
x
t
_
Su
mmar
i
zat
i
o
n
A
n
_
O
v
er
v
i
e
w
.
[2
1
]
Man
n
e,
Su
n
eet
h
a,
Z
ah
eer
Parv
ez
S
h
ai
k
Mo
h
d
,
a
n
d
S.
Sameen
Fa
t
i
ma,
"
E
x
t
rac
t
i
o
n
b
as
ed
a
u
t
o
mat
i
c
t
ex
t
s
u
mma
r
i
zat
i
o
n
s
y
s
t
em
w
i
t
h
H
MM
t
a
g
g
er,
"
In
t
e
r
n
a
t
i
o
n
a
l
Jo
u
r
n
a
l
o
f
S
o
f
t
Co
m
p
u
t
i
n
g
a
n
d
E
n
g
i
n
ee
r
i
n
g
(IJS
CE
)
,
v
o
l
.
1
,
n
o
.
3
,
p
p
.
1
1
8
-
1
2
3
,
J
u
l
y
2
0
1
1
.
[2
2
]
L
i
u
,
Pet
er
J
.
,
Mo
h
ammad
Sal
eh
,
E
t
i
e
n
n
e
Po
t
,
Ben
G
o
o
d
r
i
ch
,
Ry
an
Sep
a
s
s
i
,
L
u
k
as
z
K
ai
s
er,
an
d
N
o
am
Sh
az
eer,
"
G
en
erat
i
n
g
w
i
k
i
p
ed
i
a
b
y
s
u
mmari
z
i
n
g
l
o
n
g
s
eq
u
en
ce
s
,
"
c
o
n
f
er
e
n
ce
p
a
p
e
r
a
t
ICLR
2
0
1
8
,
J
an
u
ary
2
0
1
8
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
1693
-
6930
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
,
Vol.
1
8
,
No
.
3
,
J
une
2020:
1
2
6
8
-
1
2
7
4
1274
[2
3
]
K
h
a
i
ru
d
d
i
n
K
h
a
l
i
d
,
A
za
h
M
o
h
amed
,
Rami
z
i
M
o
h
ame
d
,
H
u
s
s
ai
n
Sh
areef,
"
Perfo
rm
an
ce
C
o
mp
ar
i
s
o
n
o
f
A
r
t
i
f
i
ci
al
In
t
e
l
l
i
g
e
n
ce
T
ech
n
i
q
u
e
s
fo
r
N
o
n
-
i
n
t
ru
s
i
v
e
E
l
ec
t
ri
ca
l
L
o
ad
Mo
n
i
t
o
ri
n
g
,
"
B
u
l
l
e
t
i
n
o
f
E
l
ec
t
r
i
ca
l
E
n
g
i
n
eer
i
n
g
a
n
d
In
f
o
r
m
a
t
i
c
s
,
v
o
l
.
7
,
n
o
.
2
,
p
p
.
1
4
3
-
1
5
2
,
J
u
n
e
2
0
1
8
.
[2
4
]
V
i
n
n
ara
s
u
A
.
,
D
eep
a
V
.
J
o
s
e
,
“
Sp
eec
h
t
o
t
e
x
t
c
o
n
v
ers
i
o
n
an
d
s
u
mmar
i
zat
i
o
n
fo
r
effect
i
v
e
n
d
ers
t
an
d
i
n
g
an
d
d
o
c
u
men
t
at
i
o
n
,
”
In
t
er
n
a
t
i
o
n
a
l
Jo
u
r
n
a
l
o
f
E
l
ect
r
i
c
a
l
a
n
d
Co
m
p
u
t
er
E
n
g
i
n
ee
r
i
n
g
(IJ
E
CE
)
,
v
o
l
.
9
,
n
o
.
5
,
p
p
.
3
6
4
2
-
3
6
4
8
,
O
ct
o
b
er
2
0
1
9
.
[2
5
]
H
u
s
n
i
T
h
amr
i
n
,
G
u
n
a
w
a
n
A
ri
y
an
t
o
,
I
r
ma
Y
u
l
i
an
a,
W
a
w
an
J
o
k
o
Pran
o
t
o
,
“
Cro
w
d
s
o
u
rc
i
n
g
i
n
d
e
v
el
o
p
i
n
g
rep
o
s
i
t
o
ry
o
f
p
h
ras
e
d
efi
n
i
t
i
o
n
i
n
Bah
a
s
a
In
d
o
n
es
i
a,
”
TE
LK
O
M
NI
KA
Tel
ec
o
m
m
u
n
i
c
a
t
i
o
n
Co
m
p
u
t
i
n
g
E
l
ect
r
o
n
i
c
s
a
n
d
Co
n
t
r
o
l
,
v
o
l
.
1
7
,
n
o
.
5
,
p
p
.
2
3
2
1
-
2
3
2
6
,
O
ct
o
b
er
2
0
1
9
.
[2
6
]
Sek
aran
K
au
s
h
i
k
an
d
P.
V
en
k
a
t
a
K
ri
s
h
n
a,
"
Bi
g
C
l
o
u
d
:
a
h
y
b
r
i
d
c
l
o
u
d
m
o
d
e
l
fo
r
s
ecu
re
d
at
a
s
t
o
rag
e
t
h
r
o
u
g
h
cl
o
u
d
s
p
ace,
"
In
t
e
r
n
a
t
i
o
n
a
l
Jo
u
r
n
a
l
o
f
A
d
v
a
n
ce
d
In
t
e
l
l
i
g
e
n
ce
P
a
r
a
d
i
g
m
s
,
v
o
l
.
8
,
n
o
.
2
,
p
p
.
2
2
9
-
2
4
1
,
J
a
n
u
ar
y
2
0
1
6
.
[2
7
]
Sek
ar
an
,
K
.
,
&
K
ri
s
h
n
a,
P.
V
.
,
“
Cro
s
s
reg
i
o
n
l
o
ad
b
al
a
n
ci
n
g
o
f
t
a
s
k
s
u
s
i
n
g
reg
i
o
n
-
b
a
s
ed
rer
o
u
t
i
n
g
o
f
l
o
a
d
s
i
n
cl
o
u
d
co
mp
u
t
i
n
g
en
v
i
r
o
n
me
n
t
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
Jo
u
r
n
a
l
o
f
A
d
v
a
n
ce
d
In
t
el
l
i
g
en
ce
P
a
r
a
d
i
g
m
s
,
v
o
l
.
9
,
n
o
.
5
-
6
,
p
p
.
5
8
9
-
6
0
3
,
J
an
u
ary
2
0
1
7
.
[2
8
]
Sek
aran
,
K
.
,
K
h
an
,
M.
S.
,
Pat
a
n
,
R.
,
G
an
d
o
mi
,
A
.
H
.
,
K
ri
s
h
n
a,
P.
V
.
,
&
K
a
l
l
am,
S.
,
“
Imp
r
o
v
i
n
g
t
h
e
Res
p
o
n
s
e
T
i
me
o
f
M
-
L
earn
i
n
g
an
d
Cl
o
u
d
C
o
mp
u
t
i
n
g
E
n
v
i
r
o
n
m
en
t
s
U
s
i
n
g
a
D
o
m
i
n
a
n
t
Fi
ref
l
y
A
p
p
ro
ac
h
,
”
in
I
E
E
E
A
cces
s
,
v
o
l
.
7
,
p
p
.
3
0
2
0
3
-
3
0
2
1
2
,
2
0
1
9
.
[2
9
]
K
au
s
h
i
k
,
S.
,
Si
n
g
h
,
S.
,
&
Pat
h
an
,
R.
K
.
,
“
D
es
i
g
n
o
f
n
o
v
el
cl
o
u
d
arch
i
t
ec
t
u
re
fo
r
en
er
g
y
aw
are
c
o
s
t
co
mp
u
t
a
t
i
o
n
i
n
cl
o
u
d
co
m
p
u
t
i
n
g
e
n
v
i
ro
n
men
t
,
”
2
0
1
7
In
n
o
v
a
t
i
o
n
s
i
n
P
o
we
r
a
n
d
A
d
va
n
ced
C
o
m
p
u
t
i
n
g
Tec
h
n
o
l
o
g
i
es
(
i
-
P
A
CT)
,
V
el
l
o
re,
p
p
.
1
-
6
,
2
0
1
7
.
[3
0
]
Sek
aran
,
K
.
,
V
i
k
ram,
G
.
R.
,
Ch
o
w
d
ar,
B.
V
.
,
&
Raj
u
,
U
.
N
.
P.
,
“
Co
mb
at
i
n
g
D
i
s
t
ri
b
u
t
ed
D
e
n
i
a
l
o
f
Serv
i
ce
A
t
t
ack
s
U
s
i
n
g
L
o
ad
Ba
l
an
ce
d
H
ad
o
o
p
Cl
u
s
t
eri
n
g
i
n
Cl
o
u
d
Co
mp
u
t
i
n
g
E
n
v
i
ro
n
men
t
,
”
In
P
r
o
cee
d
i
n
g
s
o
f
t
h
e
2
nd
In
t
e
r
n
a
t
i
o
n
a
l
Co
n
f
er
e
n
ce
o
n
D
i
g
i
t
a
l
Tech
n
o
l
o
g
y
i
n
E
d
u
c
a
t
i
o
n
,
p
p
.
7
7
-
8
1
,
O
ct
o
b
er
2
0
1
8
.
Evaluation Warning : The document was created with Spire.PDF for Python.