I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
p
ute
r
E
ng
in
ee
ring
(
I
J
E
CE
)
Vo
l.
11
,
No
.
3
,
J
u
n
e
202
1
,
p
p
.
2
3
1
5
~
2
3
2
6
I
SS
N:
2
0
8
8
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v
1
1
i
3
.
p
p
2
3
1
5
-
2
3
2
6
2315
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m
A sy
ste
m
a
tic
revi
e
w
on sequence
-
to
-
sequence
learni
ng
w
ith
neura
l net
w
o
rk
a
nd it
s
m
o
dels
H
a
na
Yo
us
uf
1
,
M
icha
el
L
a
h
zi
2
,
Sa
id A.
Sa
llo
u
m
3
,
K
ha
le
d Sha
a
la
n
4
1
,
2,
3,
4
F
a
c
u
lt
y
o
f
En
g
in
e
e
rin
g
a
n
d
I
T
,
th
e
Brit
ish
U
n
iv
e
rsity
in
Du
b
a
i
,
Un
it
e
d
A
ra
b
Em
irate
s
3
Re
se
a
rc
h
In
stit
u
te o
f
S
c
ien
c
e
s a
n
d
En
g
i
n
e
e
rin
g
,
Un
iv
e
rsity
o
f
S
h
a
rjah
,
Un
i
ted
A
ra
b
Em
irate
s
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Mar
24
,
2
0
2
0
R
ev
i
s
ed
Sep
1
8
,
2
0
2
0
A
cc
ep
ted
Oct
5
,
2
0
2
0
W
e
d
e
v
e
lo
p
a
p
re
c
ise
w
rit
in
g
su
rv
e
y
o
n
se
q
u
e
n
c
e
-
to
-
se
q
u
e
n
c
e
lea
rn
in
g
w
it
h
n
e
u
ra
l
n
e
tw
o
rk
a
n
d
it
s
m
o
d
e
ls.
Th
e
p
rim
a
r
y
a
i
m
o
f
th
is
re
p
o
rt
is
t
o
e
n
h
a
n
c
e
th
e
k
n
o
w
led
g
e
o
f
th
e
se
q
u
e
n
c
e
-
to
-
se
q
u
e
n
c
e
n
e
u
ra
l
n
e
tw
o
rk
a
n
d
t
o
lo
c
a
te
th
e
b
e
st
w
a
y
to
d
e
a
l
w
it
h
e
x
e
c
u
ti
n
g
it
.
T
h
re
e
m
o
d
e
ls
a
re
m
o
stly
u
se
d
in
se
q
u
e
n
c
e
-
to
-
se
q
u
e
n
c
e
n
e
u
ra
l
n
e
t
w
o
rk
a
p
p
li
c
a
ti
o
n
s,
n
a
m
e
l
y
:
re
c
u
r
re
n
t
n
e
u
ra
l
n
e
tw
o
rk
s
(RNN
),
c
o
n
n
e
c
ti
o
n
ist
t
e
m
p
o
ra
l
c
las
sif
i
c
a
ti
o
n
(CT
C),
a
n
d
a
tt
e
n
ti
o
n
m
o
d
e
l.
T
h
e
e
v
id
e
n
c
e
we
a
d
o
p
ted
in
c
o
n
d
u
c
ti
n
g
th
is
s
u
rv
e
y
in
c
lu
d
e
d
u
ti
li
z
in
g
th
e
e
x
a
m
in
a
ti
o
n
i
n
q
u
iri
e
s
o
r
re
se
a
rc
h
q
u
e
stio
n
s
to
d
e
ter
m
in
e
k
e
y
w
o
rd
s,
w
h
ich
w
e
r
e
u
se
d
to
se
a
rc
h
f
o
r
b
i
t
s
o
f
p
e
e
r
-
re
v
ie
we
d
p
a
p
e
rs,
a
rti
c
le
s,
o
r
b
o
o
k
s
a
t
sc
h
o
las
ti
c
d
irec
to
ries
.
T
h
ro
u
g
h
in
tro
d
u
c
t
o
ry
h
u
n
ts,
7
9
0
p
a
p
e
rs,
a
n
d
sc
h
o
larly
w
o
rk
s
w
e
r
e
f
o
u
n
d
,
a
n
d
w
it
h
th
e
a
ss
i
sta
n
c
e
o
f
c
h
o
ice
c
r
it
e
ria
a
n
d
P
RI
S
M
A
m
e
th
o
d
o
l
o
g
y
,
th
e
n
u
m
b
e
r
o
f
p
a
p
e
rs
r
e
v
ie
we
d
d
e
c
re
a
se
d
to
1
6
.
Ev
e
r
y
o
n
e
o
f
th
e
1
6
a
rti
c
les
w
a
s
c
a
te
g
o
rize
d
b
y
th
e
ir
c
o
n
tri
b
u
t
i
o
n
t
o
e
a
c
h
e
x
a
m
in
a
ti
o
n
q
u
e
stio
n
,
a
n
d
th
e
y
w
e
r
e
b
ro
k
e
n
d
o
w
n
.
A
t
las
t,
th
e
e
x
a
m
in
a
ti
o
n
p
a
p
e
rs
e
x
p
e
rien
c
e
d
a
q
u
a
li
ty
a
p
p
ra
isa
l
w
h
e
re
th
e
su
b
se
q
u
e
n
t
ra
n
g
e
w
a
s
f
ro
m
8
3
.
3
%
to
1
0
0
%
.
T
h
e
p
r
o
p
o
se
d
sy
ste
m
a
ti
c
re
v
ie
w
e
n
a
b
led
u
s
to
c
o
l
lec
t,
e
v
a
lu
a
te,
a
n
a
l
y
z
e
,
a
n
d
e
x
p
lo
re
d
iff
e
re
n
t
a
p
p
ro
a
c
h
e
s
o
f
i
m
p
lem
e
n
ti
n
g
se
q
u
e
n
c
e
-
to
-
se
q
u
e
n
c
e
n
e
u
ra
l
n
e
tw
o
rk
m
o
d
e
ls
a
n
d
p
o
i
n
ted
o
u
t
th
e
m
o
st
c
o
m
m
o
n
u
se
in
m
a
c
h
in
e
lea
rn
in
g
.
We
f
o
ll
o
w
e
d
a
m
e
th
o
d
o
l
o
g
y
t
h
a
t
sh
o
w
s
th
e
p
o
ten
ti
a
l
o
f
a
p
p
ly
in
g
th
e
se
m
o
d
e
ls t
o
re
a
l
-
w
o
rld
a
p
p
li
c
a
ti
o
n
s.
K
ey
w
o
r
d
s
:
C
o
n
n
ec
tio
n
i
s
t te
m
p
o
r
al
class
i
f
icatio
n
s
R
ec
u
r
r
en
t n
e
u
r
al
n
et
w
o
r
k
s
atten
tio
n
m
o
d
els
Seq
u
en
ce
-
to
-
s
eq
u
e
n
ce
m
o
d
el
s
S
y
s
te
m
a
tic
r
ev
ie
w
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
:
Said
A
.
Sal
lo
u
m
R
esear
ch
I
n
s
ti
tu
te
o
f
Scie
n
ce
s
an
d
E
n
g
in
ee
r
i
n
g
Un
i
v
er
s
it
y
o
f
S
h
ar
j
ah
A
ca
d
e
m
ic
cit
y
R
o
ad
,
P
.
O.
B
o
x
2
7
2
7
2
Sh
ar
j
ah
,
Un
ited
A
r
ab
E
m
ir
ates
T
el:
+9
7
1
6
5
5
8
5
0
0
0
,
Fax
: +
9
7
1
6
5
5
8
5
0
9
9
E
m
ail:
s
s
allo
u
m
@
s
h
ar
j
ah
.
ac
.
ae
1.
I
NT
RO
D
UCT
I
O
N
Ma
ch
i
n
e
lear
n
i
n
g
(
M
L
)
is
a
lo
g
ical
in
v
esti
g
atio
n
o
f
ca
lcu
latio
n
s
an
d
ac
cu
r
ate
m
o
d
e
ls
w
ith
i
n
co
m
p
u
tatio
n
al
f
r
a
m
e
w
o
r
k
s
th
at
ac
t
w
it
h
o
u
t
u
tili
zi
n
g
clea
r
g
u
id
elin
e
s
,
d
ep
en
d
i
n
g
o
n
e
x
a
m
p
les
an
d
s
u
r
m
i
s
in
g
.
I
t
is
v
ie
w
ed
a
s
a
s
u
b
s
et
o
f
co
m
p
u
ter
ized
r
ea
s
o
n
in
g
[
1
-
4
]
.
P
er
f
o
r
m
i
n
g
M
L
i
n
cl
u
d
es
m
ak
in
g
a
m
o
d
el,
w
h
ich
is
p
r
ep
ar
e
d
o
n
s
o
m
e
p
r
ep
ar
atio
n
i
n
f
o
r
m
a
t
i
o
n
a
n
d
a
f
t
e
r
w
a
r
d
c
a
n
p
r
o
c
e
s
s
e
x
t
r
a
i
n
f
o
r
m
a
t
i
o
n
t
o
m
a
k
e
f
o
r
e
c
a
s
t
s
[5
-
7
]
.
Dif
f
er
en
t
k
in
d
s
o
f
m
o
d
els
h
a
v
e
b
ee
n
u
t
ilized
an
d
in
v
est
ig
ated
f
o
r
M
L
f
r
a
m
e
w
o
r
k
s
.
T
h
ese
m
o
d
els
i
n
clu
d
e
n
eu
r
al
n
e
t
w
o
r
k
s
,
d
ec
is
io
n
tr
ee
s
,
r
eg
r
ess
io
n
an
a
l
y
s
is
an
d
h
a
v
e
a
m
as
s
i
v
e
ap
p
licatio
n
th
at
in
clu
d
e
s
s
p
ee
ch
an
d
o
b
j
ec
t
r
ec
o
g
n
itio
n
[8
-
1
5
]
.
T
h
e
s
co
p
e
o
f
th
is
p
ap
er
is
f
o
cu
s
ed
o
n
n
e
u
r
al
n
e
t
w
o
r
k
s
an
d
th
e
ir
s
u
b
s
ets,
p
ar
ticu
lar
l
y
n
e
u
r
al
n
et
w
o
r
k
s
a
n
d
s
eq
u
en
ce
-
to
-
s
eq
u
e
n
ce
lear
n
in
g
.
Neu
r
al
s
y
s
te
m
s
o
r
co
n
n
e
ctio
n
is
t
f
r
a
m
e
w
o
r
k
s
ar
e
r
eg
is
ter
in
g
f
r
a
m
e
w
o
r
k
s
d
u
b
io
u
s
l
y
m
o
t
iv
ated
b
y
t
h
e
o
r
g
an
ic
n
eu
r
al
s
y
s
te
m
s
t
h
a
t
estab
lis
h
cr
ea
tu
r
e
ce
r
eb
r
u
m
s
.
S
u
c
h
f
r
a
m
e
w
o
r
k
s
"
lear
n
"
to
p
er
f
o
r
m
as
s
i
g
n
m
e
n
ts
b
y
t
h
in
k
i
n
g
ab
o
u
t
m
o
d
el
s
,
f
o
r
th
e
m
o
s
t
p
ar
t,
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
11
,
No
.
3
,
J
u
n
e
2
0
2
1
:
2
3
1
5
-
2326
2316
w
it
h
o
u
t
b
ein
g
m
o
d
if
ied
w
it
h
task
-
e
x
p
licit
g
u
id
eli
n
es.
T
h
e
ir
ap
p
licatio
n
s
ar
e
m
as
s
i
v
e,
a
n
d
th
e
y
h
a
v
e
b
ee
n
d
ev
elo
p
ed
to
p
e
r
f
o
r
m
m
a
n
y
d
i
f
f
icu
l
t ta
s
k
s
,
m
ain
l
y
i
n
cl
u
d
in
g
p
r
ed
ictio
n
an
d
class
i
f
icatio
n
[
1
6
]
.
Seq
u
en
ce
-
to
-
Seq
u
e
n
ce
lear
n
i
n
g
is
a
p
ar
t
o
f
M
L
a
n
d
a
m
eth
o
d
o
f
n
e
u
r
al
n
et
w
o
r
k
s
t
h
a
t
is
m
o
s
tl
y
u
tili
ze
d
i
n
la
n
g
u
a
g
e
p
r
o
ce
s
s
i
n
g
m
o
d
els
[
1
7
-
2
3
]
.
I
t
ca
n
b
e
i
m
p
le
m
en
ted
w
it
h
r
ec
u
r
r
e
n
t
n
e
u
r
al
n
e
t
w
o
r
k
s
(
R
NNs)
u
s
i
n
g
en
co
d
er
-
d
ec
o
d
er
b
ased
m
ac
h
i
n
e
i
n
ter
p
r
etatio
n
th
at
m
ap
s
a
n
i
n
p
u
t
s
eq
u
e
n
ce
to
a
y
ield
o
f
a
s
u
cc
e
s
s
io
n
o
f
o
u
tp
u
t
s
eq
u
e
n
ce
w
i
th
a
tag
a
n
d
co
n
s
id
er
atio
n
estee
m
.
T
h
e
id
ea
is
to
ex
p
lo
it
t
w
o
R
NN
t
h
at
w
i
ll
co
o
p
er
ate
w
i
th
a
u
n
iq
u
e
to
k
en
a
n
d
atte
m
p
t
to
an
ticip
ate
th
e
f
o
llo
w
i
n
g
s
tate
ar
r
an
g
e
m
en
t
f
r
o
m
t
h
e
p
as
t
s
u
cc
e
s
s
io
n
.
Seq
u
e
n
ce
-
to
-
Seq
u
en
ce
m
o
d
el
s
ca
n
al
s
o
b
e
i
m
p
l
e
m
en
ted
th
r
o
u
g
h
w
h
at
is
k
n
o
w
n
a
s
co
n
n
ec
tio
n
is
t
te
m
p
o
r
al
class
i
f
icat
io
n
(
C
T
C
)
an
d
atten
tio
n
-
b
ased
m
o
d
els
.
T
h
e
s
eq
u
en
ce
-
to
-
s
eq
u
en
ce
m
o
d
el
w
a
s
i
n
itiall
y
cr
ea
ted
b
y
Go
o
g
le
f
o
r
m
ac
h
in
e
tr
an
s
latio
n
an
d
w
as
in
tr
o
d
u
c
ed
to
tr
ain
th
e
m
o
d
el
w
it
h
a
s
i
n
g
le
co
m
m
a
n
d
.
T
h
e
m
o
d
el
w
as
al
s
o
m
ad
e
to
b
e
e
asil
y
r
ep
r
o
d
u
cib
le
an
d
ex
ten
d
ab
le
s
u
ch
t
h
at
th
e
co
d
e
f
iles
w
er
e
o
r
g
an
ized
in
a
m
ea
s
u
r
ed
m
a
n
n
er
a
n
d
th
at
s
i
m
p
le
to
ex
p
an
d
u
p
o
n
an
d
r
ec
r
ea
te.
T
h
e
ex
ten
t
o
f
t
h
i
s
p
ap
er
in
co
r
p
o
r
ates
n
eu
r
al
s
y
s
te
m
s
a
n
d
t
h
eir
s
u
b
s
et
s
,
esp
ec
iall
y
r
ec
u
r
r
en
t
n
eu
r
al
n
et
w
o
r
k
s
(
R
NN)
,
co
n
n
ec
t
io
n
is
t
te
m
p
o
r
al
class
if
ica
tio
n
(
C
T
C
)
,
an
d
atten
tio
n
-
b
ased
m
o
d
els
to
d
eter
m
i
n
e
w
h
ic
h
i
s
t
h
e
b
est
-
s
u
ited
ap
p
r
o
ac
h
to
i
m
p
le
m
e
n
t
w
it
h
s
eq
u
en
ce
-
to
-
s
eq
u
e
n
ce
lear
n
i
n
g
.
T
h
is
p
ap
er
ai
m
s
to
ex
p
lo
r
e
th
e
w
o
r
k
in
g
o
f
s
eq
u
e
n
ce
-
to
-
s
eq
u
en
ce
lear
n
in
g
an
d
th
eir
d
if
f
er
en
t
ap
p
licatio
n
s
,
w
h
ich
i
s
r
ef
lecte
d
i
n
th
e
r
esear
ch
q
u
e
s
tio
n
s
o
f
t
h
e
p
ap
er
.
T
h
is
p
ap
er
w
ill
co
n
d
u
ct
a
s
y
s
te
m
atic
liter
at
u
r
e
r
ev
ie
w
o
n
s
eq
u
e
n
ce
-
to
-
s
eq
u
en
ce
n
e
u
r
al
n
et
w
o
r
k
th
r
o
u
g
h
ex
p
lo
r
in
g
d
i
f
f
er
en
t
a
ca
d
em
ic
r
esear
c
h
d
ir
ec
to
r
ies
to
lo
o
k
f
o
r
p
ee
r
-
r
ev
ie
w
ed
co
n
ten
t.
Mo
r
e
o
n
th
e
m
et
h
o
d
o
f
co
n
d
u
ctin
g
t
h
is
s
y
s
te
m
atic
liter
at
u
r
e
r
ev
ie
w
is
d
is
cu
s
s
ed
in
Sectio
n
3
.
T
h
e
s
tr
u
ct
u
r
e
o
f
th
i
s
p
ap
er
is
as
: Sec
t
io
n
2
w
ill
tal
k
ab
o
u
t
a
n
d
d
is
s
ec
t t
h
e
ac
ce
s
s
ib
le
w
r
iti
n
g
ab
o
u
t
R
NN
s
,
C
T
C
,
an
d
att
en
tio
n
-
b
ased
m
o
d
els
lear
n
in
g
;
i
t
w
ill
e
x
p
lai
n
t
h
e
wo
r
k
in
g
o
f
ea
c
h
n
et
w
o
r
k
a
n
d
d
is
cu
s
s
t
h
eir
s
tr
en
g
th
s
a
n
d
w
ea
k
n
e
s
s
e
s
.
Sectio
n
3
w
il
l
s
t
u
d
y
t
h
e
e
x
a
m
in
a
tio
n
m
e
th
o
d
o
lo
g
y
u
s
ed
to
lead
th
e
s
y
s
te
m
atic
l
iter
atu
r
e
r
ev
ie
w
,
w
h
i
ch
i
n
co
r
p
o
r
ates
th
e
d
etailin
g
an
d
f
o
r
m
u
latio
n
o
f
t
h
e
r
esear
ch
q
u
es
tio
n
s
.
Sectio
n
4
w
i
ll
u
tili
ze
th
e
ac
ad
e
m
ic
wr
itin
g
s
to
co
n
d
u
ct
a
q
u
alit
y
a
s
s
e
s
s
m
en
t,
r
esp
o
n
d
to
th
e
q
u
es
tio
n
s
ad
d
r
ess
ed
,
an
d
last
l
y
,
s
ec
tio
n
5
w
il
l
h
a
v
e
a
s
h
o
r
t
co
n
clu
s
io
n
to
th
e
p
ap
er
.
2.
L
I
T
E
R
AT
U
RE
R
E
VI
E
W
Neu
r
al
n
et
w
o
r
k
s
is
an
estab
li
s
h
ed
m
o
d
el
o
f
m
ac
h
i
n
e
lear
n
i
n
g
a
n
d
h
as
h
ad
ex
te
n
s
i
v
e
r
ese
ar
ch
d
o
n
e
o
n
it
o
v
er
th
e
y
ea
r
s
;
Seq
u
e
n
c
e
-
to
-
s
eq
u
e
n
ce
n
e
u
r
al
n
et
w
o
r
k
is
a
n
e
w
lear
n
i
n
g
tec
h
n
iq
u
e
[
9
,
2
4
,
2
5
]
.
Desp
ite
th
is
,
th
er
e
is
s
til
l
q
u
i
te
a
s
u
b
s
tan
tial
s
o
m
e
o
f
r
esear
ch
d
o
n
e
o
n
b
o
th
m
o
d
els
an
d
tech
n
iq
u
es,
w
h
ic
h
w
ill
b
e
ex
p
an
d
ed
o
n
in
t
h
is
s
ec
tio
n
.
2
.
1
.
B
a
ck
g
ro
un
d
Neu
r
al
n
et
w
o
r
k
s
ar
e
in
s
p
ir
ed
b
y
b
io
lo
g
ical
n
e
u
r
al
n
e
t
w
o
r
k
s
s
y
s
te
m
s
th
a
t
co
m
p
r
is
e
an
i
m
a
ls
'
b
r
ain
s
;
th
e
y
ar
e
d
esig
n
ed
to
d
ev
el
o
p
,
p
r
o
g
r
ess
,
an
d
s
o
lv
e
co
m
p
lex
p
r
o
b
le
m
s
t
h
at
r
eq
u
ir
e
a
h
i
g
h
lev
el
o
f
co
m
p
r
e
h
en
s
io
n
to
p
er
f
o
r
m
.
T
h
er
e
ar
e
m
an
y
t
y
p
es
o
f
n
e
u
r
a
l
n
et
w
o
r
k
s
th
at
ar
e
f
o
u
n
d
to
p
er
f
o
r
m
ex
tr
e
m
el
y
w
ell
w
i
th
s
u
c
h
d
if
f
ic
u
lt
tas
k
s
s
u
c
h
as
s
p
ee
ch
r
ec
o
g
n
itio
n
a
n
d
m
ac
h
i
n
e
tr
an
s
la
tio
n
;
o
n
e
o
f
s
u
c
h
n
et
w
o
r
k
s
ar
e
th
e
r
ec
u
r
r
en
t
n
e
u
r
al
n
et
w
o
r
k
s
(
R
NN)
.
T
h
e
w
o
r
k
i
n
g
p
r
in
cip
le
b
eh
in
d
R
NNs
is
es
s
en
tiall
y
co
n
s
tr
u
cted
ar
o
u
n
d
n
eu
r
al
n
e
t
w
o
r
k
m
o
d
els
t
h
at
in
co
r
p
o
r
ate
an
en
co
d
er
-
d
ec
o
d
er
f
r
a
m
e
w
o
r
k
th
a
t
ca
n
b
e
u
s
ed
an
d
tr
ain
ed
en
d
-
to
-
en
d
to
m
ap
in
p
u
t
s
eq
u
e
n
ce
s
in
to
o
u
tp
u
t
tar
g
e
t
s
eq
u
e
n
ce
s
[
2
6
]
.
A
w
id
e
-
r
an
g
i
n
g
d
ef
in
i
ti
o
n
o
f
s
eq
u
en
ce
-
to
-
s
eq
u
en
ce
m
o
d
el
s
ca
n
b
e
s
aid
to
"
r
ef
er
s
to
t
h
e
b
r
o
ad
er
class
o
f
m
o
d
els
th
a
t
i
n
cl
u
d
e
all
m
o
d
els
th
at
m
ap
o
n
e
s
eq
u
en
ce
to
a
n
o
th
er
"
[
2
7
]
.
T
h
u
s
,
b
y
co
m
p
ar
i
n
g
it
to
t
h
e
d
ef
in
i
tio
n
o
f
[
2
6
]
,
it
is
clea
r
to
s
ee
th
e
r
elatio
n
b
et
w
ee
n
th
e
t
w
o
d
ef
in
i
tio
n
s
.
C
o
n
n
ec
tio
n
i
s
t
te
m
p
o
r
al
clas
s
if
icatio
n
s
(
C
T
C
)
is
a
k
i
n
d
o
f
n
eu
r
al
s
y
s
te
m
y
ield
r
elate
d
to
s
co
r
in
g
ca
p
ac
it
y
,
f
o
r
p
r
e
p
ar
in
g
i
n
ter
m
itten
t
n
e
u
r
al
s
y
s
te
m
s
to
h
a
n
d
l
e
g
r
o
u
p
i
n
g
is
s
u
e
s
w
h
er
e
t
h
e
t
i
m
i
n
g
is
v
ar
iab
le.
I
t
m
a
y
b
e
u
ti
lized
f
o
r
as
s
i
g
n
m
e
n
ts
lik
e
o
n
li
n
e
p
en
m
a
n
s
h
ip
r
ec
o
g
n
itio
n
o
r
p
er
ce
iv
i
n
g
p
h
o
n
e
m
e
s
i
n
d
is
co
u
r
s
e
s
o
u
n
d
.
C
T
C
w
a
s
p
r
esen
ted
i
n
2
0
0
6
an
d
allu
d
ed
to
th
e
y
ie
l
d
s
an
d
s
co
r
in
g
an
d
is
a
u
to
n
o
m
o
u
s
o
f
t
h
e
h
id
d
en
n
eu
r
al
s
y
s
te
m
s
tr
u
ct
u
r
e
[
2
8
]
.
His
to
r
icall
y
,
C
T
C
h
a
s
b
ee
n
u
s
ed
f
o
r
t
h
e
clas
s
i
f
icatio
n
o
f
u
n
s
e
g
m
en
ted
s
eq
u
en
ce
s
w
it
h
R
NN
s
,
s
u
ch
as
ca
s
es
o
f
h
an
d
w
r
iti
n
g
o
r
s
p
ee
ch
r
ec
o
g
n
itio
n
.
R
NN
s
o
n
th
eir
o
w
n
w
er
e
n
o
t
s
u
f
f
icie
n
t
f
o
r
th
e
ta
s
k
as
t
h
ei
r
s
tan
d
ar
d
n
e
u
r
al
s
y
s
te
m
tar
g
et
ca
p
ac
ities
ar
e
c
h
ar
ac
ter
ized
in
d
ep
en
d
en
tl
y
f
o
r
ea
ch
p
o
in
t
i
n
th
e
p
r
ep
ar
atio
n
ar
r
an
g
e
m
en
t
;
b
asicall
y
,
R
N
Ns
m
u
s
t
b
e
p
r
ep
ar
ed
t
o
m
a
k
e
a
p
r
o
g
r
ess
io
n
o
f
au
to
n
o
m
o
u
s
m
ar
k
o
r
d
er
s
.
I
n
o
r
d
er
to
r
em
o
v
e
th
i
s
d
ep
e
n
d
en
c
y
a
n
d
en
ab
le
R
NNs
to
p
er
f
o
r
m
th
i
s
tas
k
,
th
e
n
et
wo
r
k
h
ad
to
d
ec
o
d
e
th
e
s
y
s
te
m
o
u
tp
u
t
s
as
a
lik
elih
o
o
d
ap
p
r
o
p
r
iatio
n
o
v
er
all
p
o
s
s
ib
le
m
ar
k
s
u
cc
ess
io
n
s
,
ad
ap
ted
o
n
a
g
iv
e
n
i
n
p
u
t
g
r
o
u
p
i
n
g
.
Gi
v
e
n
th
is
d
i
s
p
er
s
io
n
,
a
tar
g
et
ca
p
ac
it
y
ca
n
b
e
d
eter
m
i
n
ed
th
at
s
tr
ai
g
h
t
f
o
r
w
ar
d
l
y
ex
p
an
d
s
t
h
e
p
r
o
b
ab
ilit
ies
o
f
t
h
e
r
ig
h
t
m
ar
k
i
n
g
.
S
in
ce
t
h
e
tar
g
et
w
o
r
k
i
s
d
if
f
er
e
n
tiab
le,
th
e
s
y
s
te
m
w
o
u
ld
t
h
e
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
A
s
ystema
tic
r
ev
ie
w
o
n
s
eq
u
en
ce
-
to
-
s
eq
u
en
ce
le
a
r
n
in
g
w
ith
n
eu
r
a
l n
etw
o
r
k
a
n
d
its
mo
d
els
(
Ha
n
a
Yo
u
s
u
f
)
2317
b
e
ab
le
to
b
e
p
r
ep
ar
ed
w
it
h
s
t
an
d
ar
d
b
ac
k
p
r
o
p
ag
atio
n
t
h
r
o
u
g
h
ti
m
e
[
2
9
]
.
T
h
u
s
,
u
s
in
g
th
i
s
co
n
ce
p
t,
t
h
er
ef
o
r
e,
u
s
e
o
f
R
NN
s
in
t
h
is
w
a
y
w
a
s
k
n
o
w
n
as
C
T
C
.
T
h
e
atten
tio
n
m
ec
h
a
n
i
s
m
i
s
a
t
y
p
e
o
f
n
e
u
r
al
n
e
t
w
o
r
k
th
a
t
al
lo
w
s
t
h
e
d
ec
o
d
er
asp
ec
t
o
f
th
e
n
et
w
o
r
k
to
f
o
cu
s
o
n
ce
r
tai
n
p
ar
ts
o
f
t
h
e
s
eq
u
en
ce
w
h
ile
t
h
e
o
u
tp
u
t
is
g
en
er
ated
.
A
tte
n
tio
n
-
b
a
s
ed
m
o
d
els
h
elp
r
e
m
o
v
e
an
y
d
ep
en
d
en
cie
s
o
n
v
ar
iab
le
-
le
n
g
t
h
in
p
u
ts
w
it
h
o
u
t
co
m
p
r
ess
i
n
g
t
h
e
m
i
n
to
f
i
x
ed
v
ec
to
r
s
b
y
u
s
i
n
g
v
ar
iab
le
-
len
g
th
m
e
m
o
r
y
w
h
er
e
t
h
e
n
,
th
e
m
o
d
el
is
f
r
ee
to
u
s
e
t
h
i
s
m
e
m
o
r
y
i
n
a
tr
u
l
y
ad
ap
tab
le
w
a
y
to
cr
ea
te
th
e
o
u
tp
u
t
s
u
cc
e
s
s
io
n
.
I
n
ad
d
itio
n
,
v
ar
io
u
s
p
iece
s
o
f
m
e
m
o
r
y
ca
n
b
e
o
b
tain
ed
at
m
u
lt
ip
le
y
ield
ti
m
e
s
t
ep
s
.
T
h
ese
m
o
d
el
s
ar
e
w
ell
-
p
er
s
u
ad
ed
in
lig
h
t
o
f
t
h
e
f
ac
t
t
h
at
d
ata
is
lo
s
t
b
y
co
m
p
ac
ti
n
g
lo
n
g
f
ac
to
r
len
g
th
g
r
o
u
p
in
g
s
i
n
to
a
f
i
x
ed
-
s
ize
v
ec
to
r
,
an
d
ch
o
o
s
i
n
g
th
e
p
r
ess
u
r
e
is
a
n
ex
tr
a
as
s
i
g
n
m
e
n
t to
co
m
p
r
eh
e
n
d
[
3
0
]
.
2
.
2
.
Sequ
ence
-
to
-
s
eque
nce
m
o
del
s
T
h
er
e
ar
e
s
ev
er
al
ap
p
r
o
ac
h
es
u
s
ed
to
i
m
p
le
m
e
n
t
s
eq
u
e
n
ce
-
to
-
s
eq
u
e
n
ce
al
g
o
r
ith
m
m
o
d
els.
T
h
e
m
o
s
t
co
m
m
o
n
m
o
d
els ar
e
th
e
co
n
n
e
ctio
n
is
t te
m
p
o
r
al
class
i
f
icat
io
n
(
C
T
C
)
,
R
NNs,
an
d
atten
t
io
n
-
b
ased
m
o
d
el.
2
.
2
.
1
.
Co
nn
ec
t
io
nis
t
t
e
m
po
ra
l c
la
s
s
if
ica
t
io
n
Th
e
C
T
C
alg
o
r
ith
m
p
r
o
p
o
s
ed
b
y
[
2
8
]
.
T
h
is
alg
o
r
ith
m
is
a
m
eth
o
d
o
f
p
r
ep
ar
in
g
s
tar
t to
f
in
i
s
h
m
o
d
els
w
it
h
o
u
t
a
r
eq
u
ir
e
m
e
n
t
o
f
ca
s
i
n
g
le
v
el
ar
r
an
g
e
m
e
n
t
o
f
t
h
e
o
b
j
ec
tiv
e
n
a
m
es
f
o
r
a
p
r
ep
ar
ati
o
n
ar
ticu
latio
n
.
C
T
C
d
ef
in
e
s
t
h
e
p
r
o
b
ab
ilit
y
o
f
t
h
e
o
u
tp
u
t
co
n
d
itio
n
,
e
s
ti
m
ated
to
u
s
e
r
ec
u
r
r
en
t
n
eu
r
al
n
et
w
o
r
k
s
,
s
i
m
p
l
y
k
n
o
w
n
as
en
co
d
er
s
[
3
1
]
.
I
n
ad
d
itio
n
,
C
T
C
u
s
es
f
o
r
s
eq
u
e
n
ce
-
to
-
s
eq
u
en
ce
m
e
th
o
d
to
h
elp
to
ad
d
r
ess
an
y
is
s
u
e
s
r
elate
d
to
th
e
le
n
g
th
o
f
t
h
e
o
u
tp
u
t
l
ab
els
w
h
e
n
it
is
s
h
o
r
ter
th
a
n
th
e
le
n
g
t
h
o
f
t
h
e
i
n
p
u
t
s
eq
u
en
c
e
s
i
n
ce
"
C
T
C
in
tr
o
d
u
ce
s
a
s
p
ec
ial
b
la
n
k
lab
el
an
d
allo
w
s
f
o
r
r
ep
etitio
n
o
f
lab
els
to
f
o
r
ce
th
e
o
u
tp
u
t
a
n
d
in
p
u
t
s
eq
u
en
ce
s
to
h
av
e
th
e
s
a
m
e
len
g
t
h
.
C
T
C
o
u
tp
u
t
s
ar
e
u
s
u
a
ll
y
d
o
m
in
a
ted
b
y
b
la
n
k
s
y
m
b
o
ls
"
[
3
2
]
.
T
h
is
g
iv
e
s
C
T
C
a
m
aj
o
r
ad
v
an
ta
g
e
w
h
e
n
u
s
i
n
g
s
eq
u
e
n
c
e
-
to
-
s
eq
u
e
n
ce
m
o
d
els i
n
m
an
y
ap
p
licatio
n
s
s
u
c
h
as tr
an
s
lati
o
n
.
2
.
2
.
2
.
Rec
urre
nt
neura
l net
wo
rk
s
T
h
e
id
ea
b
eh
in
d
s
eq
u
e
n
ce
-
to
-
s
eq
u
en
ce
m
o
d
els
u
s
in
g
t
h
e
R
NN
ap
p
r
o
ac
h
u
tili
ze
s
t
w
o
R
N
N
th
at
w
ill
co
o
p
er
ate
w
i
th
a
u
n
iq
u
e
to
k
en
a
n
d
atte
m
p
t
to
an
ticip
ate
th
e
f
o
ll
o
w
i
n
g
s
tate
ar
r
an
g
e
m
en
t
f
r
o
m
t
h
e
p
as
t
s
u
cc
e
s
s
io
n
.
T
h
e
R
NN
tr
an
s
d
u
ce
r
d
if
f
er
s
o
n
t
h
e
e
n
co
d
er
u
s
ag
e
f
r
o
m
t
h
e
C
T
C
alig
n
m
e
n
t
m
o
d
el
b
y
d
i
f
f
er
e
n
t
r
ep
ea
t
lease
ex
p
ec
tatio
n
ar
r
an
g
e
m
e
n
t
o
v
er
th
e
o
u
tp
u
t
s
eq
u
e
n
ce
s
.
I
n
s
ti
n
cti
v
el
y
,
t
h
e
en
co
d
e
r
ca
n
b
e
th
o
u
g
h
t
o
f
as
an
a
co
u
s
t
ic
m
o
d
el,
w
h
ile
th
e
ex
p
ec
tatio
n
ar
r
an
g
es
p
r
ac
ticall
y
eq
u
i
v
ale
n
t
to
a
lan
g
u
a
g
e
m
o
d
el.
T
h
e
ex
p
ec
tatio
n
ar
r
an
g
e
g
et
s
as i
n
f
o
an
d
p
r
o
ce
s
s
es a
n
o
u
tp
u
t
v
ec
t
o
r
,
s
u
b
j
ec
t to
th
e
w
h
o
le
s
eq
u
e
n
ce
o
f
lab
els [
3
1
]
.
2
.
2
.
3
.
At
t
ent
io
n
m
o
del
I
t
is
a
co
n
s
id
er
atio
n
-
b
as
ed
m
o
d
el
co
n
tain
s
a
n
e
n
co
d
er
o
r
g
a
n
ize,
a
s
i
n
th
e
R
N
N
tr
a
n
s
d
u
ce
r
m
o
d
el.
I
n
an
y
ca
s
e,
i
n
co
n
tr
as
t
to
th
e
R
NN
tr
an
s
d
u
ce
r
,
in
w
h
ic
h
t
h
e
e
n
co
d
er
an
d
th
e
e
x
p
ec
tatio
n
ar
r
an
g
e
ar
e
d
is
p
la
y
ed
au
to
n
o
m
o
u
s
l
y
a
n
d
co
n
s
o
lid
at
ed
in
th
e
j
o
in
t
s
y
s
te
m
,
a
co
n
s
id
er
atio
n
b
ased
m
o
d
el
u
s
es
a
s
o
litar
y
d
ec
o
d
er
to
d
eliv
er
an
ap
p
r
o
p
r
iatio
n
o
v
er
th
e
m
ar
k
s
m
o
ld
ed
o
n
th
e
f
u
ll
g
r
o
u
p
in
g
o
f
p
ast
f
o
r
ec
asts
a
n
d
th
e
ac
o
u
s
tic
s
[
3
1
]
.
T
h
e
d
ec
o
d
er
n
et
w
o
r
k
co
n
s
is
t
s
o
f
s
ev
er
al
r
ec
u
r
r
e
n
t
la
y
er
s
.
T
h
e
atten
tio
n
asp
ec
t
o
f
t
h
e
m
o
d
el
p
u
ts
a
h
i
g
h
er
w
ei
g
h
t o
n
ce
r
tai
n
la
y
er
s
to
p
r
o
d
u
ce
an
o
u
tp
u
t u
s
in
g
t
h
e
en
d
-
to
-
en
d
s
eq
u
e
n
ce
m
eth
o
d
.
3.
M
E
T
H
O
DO
L
O
G
Y
T
h
e
p
r
in
cip
le
o
f
r
esear
ch
m
e
th
o
d
o
lo
g
y
i
s
d
ev
elo
p
ed
b
ased
o
n
a
s
y
s
te
m
atic
li
ter
atu
r
e
r
e
v
ie
w
.
T
h
e
p
ap
er
f
o
llo
w
s
t
h
e
s
y
s
te
m
a
tic
r
ev
ie
w
m
eth
o
d
o
lo
g
y
ill
u
s
tr
ate
d
b
y
[
3
3
]
t
o
d
ir
ec
t
th
e
d
elib
er
ate
liter
atu
r
e
a
u
d
it.
T
h
e
r
ea
s
o
n
f
o
r
s
elec
tin
g
t
h
e
s
y
s
te
m
atic
r
ev
ie
w
f
o
r
s
eq
u
en
ce
-
to
-
s
eq
u
e
n
ce
n
e
u
r
al
n
et
w
o
r
k
i
s
th
at
n
o
s
y
s
te
m
atic
r
ev
ie
w
f
o
cu
s
e
s
o
n
s
eq
u
e
n
ce
-
to
-
s
eq
u
e
n
ce
n
eu
r
al
n
et
w
o
r
k
u
s
ag
e,
li
m
itatio
n
s
,
a
n
d
ap
p
licati
o
n
s
.
Mo
r
eo
v
er
,
th
i
s
m
et
h
o
d
o
lo
g
y
e
n
ab
led
u
s
to
co
llect,
ev
al
u
ate,
a
n
al
y
ze
,
an
d
ex
p
lo
r
e
d
if
f
er
en
t
ap
p
r
o
ac
h
es
to
i
m
p
le
m
e
n
ti
n
g
s
eq
u
en
ce
-
to
-
s
eq
u
e
n
ce
n
e
u
r
al
n
et
w
o
r
k
m
o
d
els
a
n
d
f
i
n
d
th
e
m
o
s
t
co
m
m
o
n
u
s
e
i
n
m
ac
h
i
n
e
lear
n
in
g
.
T
h
e
in
itia
l
s
tep
to
th
is
m
et
h
o
d
is
to
f
ig
u
r
e
o
u
t th
e
r
esea
r
ch
h
y
p
o
t
h
esi
s
o
f
o
u
r
p
ap
er
.
3
.
1
.
Resea
rc
h hy
po
t
hes
i
s
T
h
e
r
esear
ch
h
y
p
o
th
e
s
is
d
ev
e
l
o
p
ed
f
o
r
th
e
p
ap
er
w
er
e
as
:
1.
W
h
at
ar
e
th
e
d
if
f
er
en
t a
p
p
licat
io
n
s
o
f
s
eq
u
en
ce
-
to
-
s
eq
u
en
ce
n
eu
r
al
n
et
w
o
r
k
m
o
d
el
s
?
2.
Ho
w
h
as t
h
is
m
o
d
el
b
ee
n
i
m
p
l
e
m
en
ted
an
d
d
ev
elo
p
ed
?
3.
W
h
at
ar
e
th
e
ad
v
a
n
tag
e
s
an
d
l
i
m
itat
io
n
s
o
f
i
m
p
le
m
e
n
ti
n
g
s
e
q
u
en
ce
-
to
-
s
eq
u
e
n
ce
m
o
d
els?
4.
W
h
at
is
t
h
e
b
est
m
o
d
el
to
ap
p
r
o
ac
h
s
eq
u
e
n
ce
-
to
-
s
eq
u
en
ce
i
m
p
le
m
en
ta
tio
n
?
5.
W
h
at
ar
e
th
e
co
u
n
tr
ies t
h
at
co
n
tr
ib
u
ted
to
th
e
d
ev
e
lo
p
m
en
t a
n
d
i
m
p
le
m
en
tatio
n
o
f
s
eq
u
e
n
c
e
-
to
-
s
eq
u
e
n
ce
?
T
ab
le
1
s
h
o
w
s
t
h
e
r
esear
ch
h
y
p
o
th
esis
,
i
n
clu
d
i
n
g
th
eir
m
o
ti
v
atio
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
11
,
No
.
3
,
J
u
n
e
2
0
2
1
:
2
3
1
5
-
2326
2318
T
ab
le
1
.
R
esear
ch
m
o
ti
v
atio
n
an
d
h
y
p
o
t
h
esi
s
R
Q
#
R
e
se
a
r
c
h
H
y
p
o
t
h
e
si
s
M
o
t
i
v
a
t
i
o
n
R
Q
1
W
h
a
t
a
r
e
t
h
e
d
i
f
f
e
r
e
n
t
a
p
p
l
i
c
a
t
i
o
n
s o
f
se
q
u
e
n
c
e
-
to
-
se
q
u
e
n
c
e
mo
d
e
l
s?
T
h
i
s
q
u
e
st
i
o
n
h
e
l
p
s
g
a
i
n
a
b
r
o
a
d
e
r
u
n
d
e
r
st
a
n
d
i
n
g
o
f
a
l
l
t
h
e
a
p
p
l
i
c
a
t
i
o
n
s
o
f
se
q
u
e
n
c
e
-
to
-
se
q
u
e
n
c
e
mo
d
e
l
s
t
o
r
e
l
i
e
v
e
a
l
o
t
o
f
a
mb
i
g
u
i
t
y
su
r
r
o
u
n
d
i
n
g
t
h
e
d
e
f
i
n
i
t
i
o
n
.
T
h
i
s
w
i
l
l
l
a
t
e
r
h
e
l
p
i
d
e
n
t
i
f
y
t
h
e
b
e
st
p
o
ssi
b
l
e
a
p
p
r
o
a
c
h
t
o
i
m
p
l
e
me
n
t
t
h
e
mo
d
e
l
b
a
se
d
o
n
t
h
e
b
e
st
a
d
v
a
n
t
a
g
e
s i
t
h
a
s
o
n
e
a
c
h
a
p
p
l
i
c
a
t
i
o
n
.
R
Q
2
H
o
w
h
a
s
t
h
i
s mo
d
e
l
b
e
e
n
i
mp
l
e
me
n
t
e
d
a
n
d
d
e
v
e
l
o
p
e
d
?
T
h
i
s
q
u
e
st
i
o
n
a
l
l
o
w
s
a
d
e
e
p
e
r
u
n
d
e
r
st
a
n
d
i
n
g
o
f
t
h
e
d
e
v
e
l
o
p
me
n
t
o
f
t
h
e
mo
d
e
l
.
M
o
r
e
o
v
e
r
,
I
t
w
i
l
l
h
i
g
h
l
i
g
h
t
i
t
s sh
o
r
t
c
o
mi
n
g
s
t
o
i
d
e
n
t
i
f
y
t
h
e
a
d
d
i
t
i
o
n
a
l
w
o
r
k
n
e
e
d
s t
o
b
e
d
o
n
e
o
n
t
h
e
f
i
e
l
d
a
n
d
w
h
a
t
i
s
t
h
e
b
e
st
a
n
d
l
e
a
st
p
r
o
b
l
e
mat
i
c
a
p
p
r
o
a
c
h
t
o
b
e
u
se
d
w
i
t
h
c
o
mm
o
n
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
l
i
c
a
t
i
o
n
s l
i
k
e
sp
e
e
c
h
r
e
c
o
g
n
i
t
i
o
n
a
n
d
t
r
a
n
sl
a
t
i
o
n
.
R
Q
3
W
h
a
t
a
r
e
t
h
e
a
d
v
a
n
t
a
g
e
s a
n
d
l
i
mi
t
a
t
i
o
n
s
of
i
mp
l
e
me
n
t
i
n
g
se
q
u
e
n
c
e
-
to
-
se
q
u
e
n
c
e
mo
d
e
l
s?
T
h
i
s
q
u
e
st
i
o
n
a
d
d
r
e
sse
s
t
h
e
si
g
n
i
f
i
c
a
n
c
e
o
f
t
h
e
mo
d
e
l
a
n
d
h
o
w
t
h
e
i
n
s
t
a
l
l
a
t
i
o
n
h
a
s
h
e
l
p
e
d
i
n
r
e
l
i
e
v
i
n
g
so
me
o
f
t
h
e
p
r
o
b
l
e
ms
t
h
a
t
w
e
r
e
i
n
v
o
l
v
e
d
i
n
t
h
e
f
i
e
l
d
o
f
mac
h
i
n
e
l
e
a
r
n
i
n
g
p
r
i
o
r
t
o
i
t
s
d
e
v
e
l
o
p
me
n
t
.
M
o
r
e
o
v
e
r
,
t
h
i
s
h
e
l
p
s
t
o
h
i
g
h
l
i
g
h
t
so
me
c
h
a
l
l
e
n
g
e
s
t
h
a
t
a
r
e
st
i
l
l
n
o
t
so
l
v
e
d
a
n
d
n
e
w
c
h
a
l
l
e
n
g
e
s t
h
a
t
d
e
v
e
l
o
p
e
d
w
i
t
h
t
h
e
n
e
w
mo
d
e
l
.
R
Q
4
W
h
a
t
i
s
t
h
e
b
e
st
mo
d
e
l
t
o
a
p
p
r
o
a
c
h
se
q
u
e
n
c
e
-
to
-
se
q
u
e
n
c
e
i
mp
l
e
me
n
t
a
t
i
o
n
?
T
h
i
s
q
u
e
st
i
o
n
w
i
l
l
h
i
g
h
l
i
g
h
t
t
h
e
a
d
v
a
n
t
a
g
e
s
a
n
d
d
i
sa
d
v
a
n
t
a
g
e
s
o
f
e
v
e
r
y
a
p
p
r
o
a
c
h
.
I
t
w
i
l
l
h
e
l
p
c
h
o
o
se
t
h
e
b
e
st
a
p
p
r
o
a
c
h
f
o
r
t
h
e
p
u
r
p
o
se
s
o
f
su
g
g
e
st
i
n
g
a
st
a
n
d
a
r
d
t
o
u
se
w
h
e
n
i
mp
l
e
me
n
t
i
n
g
t
h
i
s mo
d
e
l
i
n
c
o
mm
o
n
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
l
i
c
a
t
i
o
n
s
.
R
Q
5
W
h
a
t
a
r
e
t
h
e
c
o
u
n
t
r
i
e
s
t
h
a
t
c
o
n
t
r
i
b
u
t
e
d
to
t
h
e
d
e
v
e
l
o
p
me
n
t
a
n
d
i
mp
l
e
me
n
t
a
t
i
o
n
o
f
se
q
u
e
n
c
e
-
to
-
se
q
u
e
n
c
e
?
T
h
i
s
q
u
e
st
i
o
n
w
i
l
l
h
i
g
h
l
i
g
h
t
t
h
e
mo
st
c
o
u
n
t
r
i
e
s
c
o
n
t
r
i
b
u
t
e
d
to
t
h
e
i
mp
l
e
me
n
t
a
t
i
o
n
a
n
d
d
e
v
e
l
o
p
me
n
t
S
e
q
u
e
n
c
e
-
to
-
S
e
q
u
e
n
c
e
3
.
2
.
Resea
rc
h str
a
t
eg
y
T
h
e
p
r
in
cip
le
o
f
r
esear
ch
s
tr
a
teg
y
i
n
t
h
is
p
ap
er
is
to
co
n
d
u
ct
a
ca
r
ef
u
l
c
h
ec
k
in
g
o
n
th
e
s
u
b
s
eq
u
e
n
t
d
atab
ase,
p
ee
r
-
r
ev
ie
w
ed
j
o
u
r
n
als,
an
d
p
er
i
o
d
icals.
Mo
s
t
o
f
th
e
p
ap
er
s
u
t
ilized
f
r
o
m
ar
Xi
v
,
Go
o
g
le
Sc
h
o
lar
,
Scien
ce
Dir
ec
t,
I
E
E
E
Xp
lo
r
e,
an
d
Sp
r
in
g
er
co
m
p
le
te
j
o
u
r
n
als
.
W
h
e
n
t
h
e
m
o
s
t
s
i
g
n
i
f
ica
n
t
r
esear
ch
r
e
g
is
tr
ie
s
w
er
e
c
h
o
s
e
n
,
i
m
p
o
r
tan
t
k
e
y
w
o
r
d
s
r
etr
iev
ed
f
r
o
m
th
e
r
e
s
ea
r
ch
h
y
p
o
th
es
is
w
er
e
u
til
ize
d
in
s
ea
r
c
h
t
h
r
o
u
g
h
p
er
tin
en
t
titl
e
s
a
n
d
m
o
d
if
ied
w
o
r
k
s
o
f
p
ap
er
s
a
n
d
ar
ticle
s
.
A
B
o
o
lean
tec
h
n
iq
u
e
ad
ap
ted
to
s
tr
in
g
to
g
et
h
er
t
h
e
i
m
p
o
r
tan
t
s
ea
r
ch
ter
m
s
to
l
o
ca
te
th
e
m
o
s
t
s
i
g
n
i
f
ican
t
p
ee
r
-
ev
al
u
ated
ite
m
s
f
o
r
t
h
i
s
p
ap
er
.
Sp
ec
if
ic
p
er
s
p
ec
tiv
es
w
er
e
v
ie
w
ed
w
h
en
u
t
ilizi
n
g
t
h
e
s
ea
r
ch
ter
m
s
,
s
u
c
h
as
eq
u
iv
ale
n
t
w
o
r
d
s
o
r
p
lu
r
als
o
f
ter
m
s
.
T
ab
le
2
s
h
o
w
s
th
e
s
ea
r
c
h
ter
m
s
u
s
ed
in
v
ar
io
u
s
d
atab
ase
s
an
d
th
e
o
u
tco
m
es
it
ac
co
m
p
li
s
h
ed
.
O
v
er
all,
t
h
er
e
w
er
e
8
7
1
p
a
p
er
s
ap
p
ea
r
ed
t
h
r
o
u
g
h
th
e
i
n
itial
s
ea
r
c
h
i
n
g
s
tag
e.
Fro
m
th
o
s
e,
8
6
p
ap
er
s
ap
p
ea
r
e
d
to
b
e
d
u
p
licated
,
an
d
f
iv
e
p
ap
er
s
w
er
e
f
o
u
n
d
th
r
o
u
g
h
citatio
n
s
o
f
f
o
u
n
d
ac
ad
e
m
ic
p
ap
er
s
,
w
h
ic
h
b
r
o
u
g
h
t
th
e
to
tal
n
u
m
b
er
o
f
p
ap
er
s
to
7
9
0
.
T
h
ese
p
ap
er
s
w
er
e
r
e
f
in
ed
.
T
h
e
f
i
n
al
p
iece
s
o
f
liter
at
u
r
e
u
s
ed
in
th
is
p
ap
er
w
er
e
s
cr
ee
n
i
n
g
ac
co
r
d
in
g
to
th
e
s
e
lectio
n
cr
iter
ia
an
d
th
e
p
r
ef
er
r
ed
r
ep
o
r
tin
g
ite
m
s
f
o
r
s
y
s
te
m
atic
r
ev
ie
w
s
a
n
d
m
eta
-
an
a
l
y
s
es (
P
R
I
SM
A
)
Stat
e
m
en
t [
3
4
]
.
Fig
u
r
e
1
s
h
o
w
s
t
h
e
P
R
I
SMA
f
lo
w
ch
ar
t
.
T
ab
le
2
.
B
o
o
lean
tech
n
iq
u
e
s
e
ar
ch
r
esu
lt
s
D
a
t
a
b
a
se
S
e
a
r
c
h
S
t
r
i
n
g
N
u
mb
e
r
o
f
r
e
su
l
t
s
G
o
o
g
l
e
S
c
h
o
l
a
r
(
"
C
o
n
n
e
c
t
i
o
n
i
st
t
e
mp
o
r
a
l
c
l
a
ss
i
f
i
c
a
t
i
o
n
s
"
)
a
n
d
(
"
se
q
u
e
n
c
e
-
to
-
se
q
u
e
n
c
e
"
)
a
n
d
(
"
R
N
N
"
)
a
n
d
(
"
C
T
C
"
)
56
S
c
i
e
n
c
e
D
i
r
e
c
t
((
"
A
t
t
e
n
t
i
o
n
mo
d
e
l
"
o
r
"
a
t
t
e
n
t
i
o
n
me
c
h
a
n
i
sm
"
)
a
n
d
(
"
se
q
u
e
n
c
e
-
to
-
S
e
q
u
e
n
c
e
"
))
68
I
EEE
X
p
l
o
r
e
"
S
e
q
u
e
n
c
e
-
to
-
se
q
u
e
n
c
e
"
a
n
d
"
n
e
u
r
a
l
n
e
t
w
o
r
k
"
a
n
d
"
c
h
a
l
l
e
n
g
e
s
"
o
r
"
l
i
mi
t
a
t
i
o
n
"
2
8
4
S
p
r
i
n
g
e
r
C
o
mp
l
e
t
e
Jo
u
r
n
a
l
s
(
"
S
e
q
u
e
n
c
e
-
to
-
se
q
u
e
n
c
e
"
)
a
n
d
(
"
r
e
c
u
r
r
e
n
t
n
e
u
r
a
l
n
e
t
w
o
r
k
s
"
o
r
"
c
o
n
n
e
c
t
i
o
n
i
st
t
e
mp
o
r
a
l
c
l
a
ssi
f
i
c
a
t
i
o
n
s
"
o
r
"
a
t
t
e
n
t
i
o
n
mo
d
e
l
"
)
4
6
3
3
.
3
.
I
nclu
s
io
n a
nd
e
x
clu
s
io
n c
rit
er
ia
I
n
clu
s
io
n
an
d
ex
cl
u
s
io
n
cr
iter
ia
w
er
e
ch
o
s
en
cr
iter
ia
to
m
ea
s
u
r
e
th
e
m
o
s
t
f
o
r
m
al
w
r
iti
n
g
f
o
r
th
e
ex
ten
t o
f
t
h
i
s
p
ap
er
.
I
n
v
ie
w
o
f
t
h
ese
cr
iter
ia,
t
h
e
p
ap
er
s
w
h
i
ch
f
o
llo
w
ed
t
h
e
r
esear
ch
p
er
s
p
ec
tiv
e
cr
iter
ia
w
er
e
in
cl
u
d
ed
f
o
r
th
e
ex
te
n
t o
f
r
ese
ar
ch
.
T
ab
le
3
s
h
o
w
s
th
e
i
n
cl
u
s
io
n
an
d
ex
cl
u
s
io
n
cr
iter
ia.
T
ab
le
3
.
I
n
clu
s
io
n
a
n
d
ex
cl
u
s
i
o
n
cr
iter
ia
I
n
c
l
u
si
o
n
c
r
i
t
e
r
i
a
Ex
c
i
si
o
n
c
r
i
t
e
r
i
a
M
u
s
t
b
e
p
e
e
r
-
r
e
v
i
e
w
e
d
j
o
u
r
n
a
l
p
a
p
e
r
s,
a
r
t
i
c
l
e
s
,
o
r
b
o
o
k
s
A
r
t
i
c
l
e
s
a
r
e
w
r
i
t
t
e
n
i
n
an
o
t
h
e
r
l
a
n
g
u
a
g
e
S
h
o
u
l
d
i
n
c
l
u
d
e
t
h
e
d
e
v
e
l
o
p
me
n
t
a
n
d
i
mp
l
e
me
n
t
a
t
i
o
n
o
f
t
h
e
se
q
u
e
n
c
e
-
to
-
se
q
u
e
n
c
e
mo
d
e
l
a
n
d
n
e
u
r
a
l
n
e
t
w
o
r
k
s
P
a
p
e
r
s fo
c
u
s
o
n
s
p
e
c
i
a
l
i
z
e
d
p
a
r
t
s o
r
w
o
r
k
i
n
g
t
h
e
se
q
u
e
n
c
e
-
to
-
se
q
u
e
n
c
e
mo
d
e
l
a
n
d
n
e
u
r
a
l
n
e
t
w
o
r
k
M
u
s
t
b
e
w
r
i
t
t
e
n
i
n
E
n
g
l
i
sh
D
i
scu
se
s a
s
a
s
u
b
t
o
p
i
c
i
n
S
e
q
u
e
n
c
e
-
to
-
S
e
q
u
e
n
c
e
,
R
N
N
,
C
T
C
,
a
n
d
a
t
t
e
n
t
i
o
n
mo
d
e
l
s.
A
ca
d
e
m
ic
p
ap
er
s
th
at
f
a
iled
to
m
ee
t
t
h
e
m
en
t
io
n
ed
cr
iter
ia
w
er
e
ex
c
lu
d
ed
f
r
o
m
r
esear
c
h
.
T
h
e
in
itia
l
s
cr
ee
n
i
n
g
w
as
d
o
n
e
o
n
all
t
h
e
e
x
tr
ac
ted
7
9
0
r
ec
o
r
d
s
in
o
r
d
er
to
n
ar
r
o
w
th
e
r
es
u
lt.
T
h
ese
r
ec
o
r
d
s
w
er
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
A
s
ystema
tic
r
ev
ie
w
o
n
s
eq
u
en
ce
-
to
-
s
eq
u
en
ce
le
a
r
n
in
g
w
ith
n
eu
r
a
l n
etw
o
r
k
a
n
d
its
mo
d
els
(
Ha
n
a
Yo
u
s
u
f
)
2319
co
m
p
ar
ed
ag
ai
n
s
t
p
r
e
-
s
elec
tio
n
cr
iter
ia
th
a
t
i
n
cl
u
d
ed
r
ef
in
i
n
g
th
e
s
ea
r
ch
f
ield
s
to
en
g
i
n
ee
r
in
g
a
n
d
co
m
p
u
ter
s
cien
ce
,
li
m
it
in
g
t
h
e
r
esu
lts
to
th
e
n
o
o
ld
er
th
a
n
ten
y
ea
r
s
.
T
h
is
r
es
u
lted
i
n
t
h
e
ev
al
u
ated
n
u
m
b
er
s
to
d
ec
r
ea
s
e
to
4
1
3
p
a
p
er
s
,
w
h
ich
w
er
e
a
s
s
ess
ed
b
y
t
h
e
s
elec
tio
n
cr
iter
i
a
m
e
n
tio
n
ed
.
Fi
g
u
r
e
1
r
ep
r
esen
ts
th
e
p
r
o
ce
s
s
f
o
r
b
o
th
th
e
p
r
e
-
s
elec
tio
n
a
n
d
th
e
s
elec
tio
n
cr
iter
ia.
T
h
is
p
r
o
ce
s
s
r
es
u
lted
i
n
1
6
p
ap
er
s
to
b
e
in
cl
u
d
ed
in
t
h
e
s
y
s
te
m
a
tic
r
ev
ie
w
i
n
s
eq
u
en
ce
-
to
-
s
eq
u
e
n
ce
n
e
u
r
al
n
et
w
o
r
k
.
Fig
u
r
e
1
.
P
R
I
SMA
f
lo
w
ch
ar
t
3
.
4
.
Q
ua
lity
a
s
s
es
s
m
e
nt
I
n
clu
s
io
n
q
u
alit
y
as
s
ess
m
e
n
t
is
o
n
e
o
f
th
e
m
o
s
t
es
s
en
t
i
al
an
d
cr
itical
p
ar
ts
o
f
an
y
s
y
s
te
m
a
tic
r
ev
ie
w
[
3
5
-
3
9
]
.
T
h
e
ass
es
s
m
e
n
t
q
u
ali
t
y
a
s
s
u
r
an
ce
c
h
ec
k
lis
t
f
o
r
o
u
r
s
y
s
te
m
atic
r
e
v
ie
w
co
n
s
is
t
s
o
f
6
q
u
e
s
tio
n
s
f
o
r
th
e
1
6
ch
o
s
en
p
ap
er
s
,
as
s
h
o
w
n
i
n
T
ab
le
4
.
T
h
e
s
co
r
in
g
o
f
th
is
p
r
o
ce
s
s
is
d
o
n
e
b
ase
d
o
n
th
e
w
o
r
k
o
f
[
4
0
]
as
:
a
'
Ye
s
'
to
th
e
q
u
e
s
tio
n
o
f
th
e
q
u
alit
y
as
s
es
s
m
e
n
t
i
s
in
d
i
ca
ted
b
y
a
1
,
a
'
No
'
w
as
i
n
d
icate
d
b
y
a
0
,
an
d
a
'
P
ar
tiall
y
'
w
a
s
in
d
icate
d
b
y
a
0
.
5
.
A
s
s
ee
n
b
y
t
h
e
r
esu
lt
s
in
T
a
b
le
5
,
all
th
e
ch
o
s
en
p
a
p
er
s
h
av
e
p
ass
ed
t
h
e
q
u
alit
y
a
s
s
e
s
s
m
en
t.
T
ab
le
4
.
Qu
alit
y
as
s
u
r
a
n
ce
q
u
esti
o
n
s
Q
u
e
s.
Q
u
a
l
i
t
y
A
ssu
r
a
n
c
e
Q
u
e
st
i
o
n
1
A
r
e
t
h
e
r
e
se
a
r
c
h
a
i
ms s
p
e
c
i
f
i
e
d
c
l
e
a
r
l
y
?
2
I
s t
h
e
i
n
f
o
r
mat
i
o
n
p
r
e
se
n
t
e
d
c
l
e
a
r
a
n
d
c
o
n
c
i
se
?
3
D
o
e
s t
h
e
st
u
d
y
p
r
o
v
i
d
e
e
n
o
u
g
h
e
x
p
l
a
n
a
t
i
o
n
of
i
t
s
me
t
h
o
d
o
l
o
g
y
?
4
D
o
t
h
e
f
i
n
d
i
n
g
s
o
f
t
h
e
s
t
u
d
y
a
d
d
t
o
t
h
e
u
n
d
e
r
st
a
n
d
i
n
g
o
f
se
q
u
e
n
c
e
-
to
-
se
q
u
e
n
c
e
mo
d
e
l
s?
5
A
r
e
t
h
e
c
o
n
c
l
u
si
o
n
s
c
l
e
a
r
l
y
i
d
e
n
t
i
f
i
e
d
?
6
A
r
e
t
h
e
c
o
n
c
l
u
si
o
n
s
l
o
g
i
c
a
l
a
n
d
c
o
n
c
i
se
w
i
t
h
t
h
e
f
l
o
w
o
f
t
h
e
p
a
p
e
r
?
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
11
,
No
.
3
,
J
u
n
e
2
0
2
1
:
2
3
1
5
-
2326
2320
T
ab
le
5
.
Qu
alit
y
as
s
ess
m
e
n
t r
e
s
u
lt
s
S
t
u
d
y
Q1
Q2
Q3
Q4
Q5
Q6
T
o
t
a
l
P
e
r
c
e
n
t
a
g
e
S1
1
1
0
.
5
1
1
1
5
.
5
9
1
.
6
7
%
S2
1
1
1
1
0
.
5
1
5
.
5
9
1
.
6
7
%
S3
1
1
1
0
1
1
5
8
3
.
3
3
%
S4
1
1
1
0
.
5
1
0
.
5
5
8
3
.
3
3
%
S5
1
1
1
0
.
5
1
1
5
.
5
9
1
.
6
7
%
S6
1
1
0
.
5
0
.
5
1
1
5
8
3
.
3
3
%
S7
1
1
1
1
1
1
6
1
0
0
%
S8
1
1
0
.
5
1
1
0
.
5
5
8
3
.
3
3
%
S9
1
1
1
0
1
1
5
8
3
.
3
3
%
S
1
0
1
1
1
0
1
1
5
8
3
.
3
3
%
S
1
1
1
1
1
1
1
1
6
1
0
0
%
S
1
2
1
1
1
1
0
.
5
1
5
.
5
9
1
.
6
7
%
S
1
3
1
1
0
.
5
1
0
.
5
1
5
.
5
9
1
.
6
7
%
S
1
4
1
0
.
5
1
0
.
5
1
1
5
.
5
9
1
.
6
7
%
S
1
5
1
1
0
1
1
1
5
8
3
.
3
3
%
S
1
6
1
1
1
1
1
0
.
5
5
.
5
9
1
.
6
7
%
4.
RE
SU
L
T
S
A
ND
D
I
SCU
SS
I
O
N
Utilizi
n
g
t
h
e
r
esear
ch
p
r
o
ce
d
u
r
e
o
u
tli
n
ed
in
t
h
e
p
r
ev
io
u
s
s
ec
tio
n
,
th
e
r
esear
c
h
q
u
est
io
n
s
th
at
w
er
e
ten
d
ed
to
ea
ch
p
ap
er
ar
e
class
if
ied
an
d
an
al
y
ze
d
b
ased
o
n
th
eir
co
n
tr
ib
u
tio
n
to
ea
ch
q
u
e
s
ti
o
n
.
4
.
1
.
Cla
s
s
if
ica
t
io
ns
a
nd
a
na
ly
s
i
s
o
f
s
t
ud
ies
B
y
s
t
u
d
y
in
g
all
th
e
1
6
r
esear
c
h
p
ap
er
s
in
clu
d
ed
in
th
e
s
y
s
te
m
atic
r
ev
ie
w
,
a
class
i
f
ica
tio
n
s
y
s
te
m
w
a
s
co
m
p
r
is
ed
b
ased
o
n
t
h
eir
co
n
t
r
ib
u
tio
n
to
a
n
s
w
er
i
n
g
t
h
e
r
ese
ar
ch
q
u
es
tio
n
s
.
Ma
r
k
i
n
g
s
w
er
e
m
ad
e
i
f
t
h
e
m
ai
n
f
o
cu
s
o
f
th
e
p
ap
er
w
as
r
elate
d
to
a
p
ar
ticu
lar
ca
teg
o
r
y
.
M
o
s
t
p
ap
er
s
,
f
o
r
ex
am
p
le,
tal
k
e
d
in
b
r
ief
ab
o
u
t
th
e
d
if
f
er
e
n
t
ap
p
licatio
n
s
o
f
s
eq
u
en
ce
-
to
-
s
eq
u
en
ce
(
s
eq
2
s
eq
)
m
o
d
el
s
.
Ho
w
e
v
er
,
o
n
l
y
s
tu
d
i
es
t
h
at
w
er
e
m
ai
n
l
y
f
o
cu
s
ed
o
n
a
ce
r
tai
n
ap
p
licati
o
n
o
r
d
is
cu
s
s
ed
s
e
v
er
al
ap
p
licatio
n
s
i
n
d
ep
th
w
er
e
ca
teg
o
r
iz
ed
as
'A
p
p
l
icatio
n
s
o
f
s
eq
2
s
eq
'
.
T
h
e
clas
s
i
f
icatio
n
r
esu
lt
s
ca
n
b
e
s
ee
n
i
n
T
ab
le
6
.
A
d
d
itio
n
al
l
y
,
ea
c
h
s
t
u
d
y
w
as
an
al
y
ze
d
in
d
etail,
an
d
th
e
r
es
u
lts
o
f
th
is
d
etailed
s
tu
d
y
ar
e
o
u
tli
n
ed
in
T
ab
le
7
(
s
ee
in
A
p
p
en
d
i
x
)
.
Fig
u
r
e
2
s
h
o
w
s
t
h
e
p
u
b
licatio
n
d
is
tr
ib
u
tio
n
s
co
u
n
tr
y
w
i
s
e.
A
s
th
e
r
e
s
u
l
ts
s
h
o
w
,
th
er
e
h
as
b
ee
n
a
n
i
n
cr
ea
s
e
i
n
in
ter
e
s
t
i
n
th
is
to
p
ic
o
v
er
t
h
e
last
t
w
o
d
ec
ad
es,
as
in
d
icate
d
b
y
t
h
e
i
n
cr
ea
s
e
i
n
a
n
u
m
b
er
o
f
p
u
b
licatio
n
s
s
i
n
ce
1
9
9
0
.
Mo
s
t
o
f
th
i
s
r
esear
ch
is
f
o
cu
s
ed
o
n
th
e
US,
as
m
e
n
tio
n
ed
.
Fig
u
r
e
2
.
P
u
b
lic
atio
n
d
is
tr
ib
u
ti
o
n
co
u
n
tr
y
w
is
e
4
.
2
.
Q
ua
lity
a
s
s
es
s
m
e
nt
re
s
ults
Usi
n
g
t
h
e
q
u
esti
o
n
s
o
u
tli
n
ed
in
T
ab
le
4
,
a
s
co
r
e
w
as
g
i
v
e
n
to
ea
ch
o
f
t
h
e
1
6
s
t
u
d
ies
u
s
ed
in
t
h
i
s
p
ap
er
.
T
h
e
m
a
x
i
m
u
m
s
co
r
e
th
at
a
p
ap
er
ca
n
h
av
e
is
6
.
T
h
e
r
esu
lt
s
o
f
ea
c
h
p
ap
er
ar
e
p
r
esen
te
d
in
T
ab
le
6
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
A
s
ystema
tic
r
ev
ie
w
o
n
s
eq
u
en
ce
-
to
-
s
eq
u
en
ce
le
a
r
n
in
g
w
ith
n
eu
r
a
l n
etw
o
r
k
a
n
d
its
mo
d
els
(
Ha
n
a
Yo
u
s
u
f
)
2321
T
ab
le
6
.
C
lass
if
icatio
n
o
f
l
iter
atu
r
e
r
ev
ie
w
S
t
u
d
y
S
o
u
r
c
e
A
p
p
l
i
c
a
t
i
o
n
s
o
f
se
q
2
se
q
S
e
q
2
se
q
v
i
a
R
N
N
S
e
q
2
se
q
v
i
a
C
T
C
S
e
q
2
se
q
v
i
a
a
t
t
e
n
t
i
o
n
mo
d
e
l
s
C
h
a
l
l
e
n
g
e
s o
f
se
q
2
se
q
S1
[
4
1
]
x
x
S2
[
3
2
]
x
x
S3
[
2
8
]
x
x
x
S4
[
4
2
]
x
x
S5
[
4
3
]
x
x
S6
[
2
7
]
x
x
S7
[
1
6
]
x
x
x
S8
[
3
1
]
x
x
x
S9
[
2
6
]
x
x
S
1
0
[
4
4
]
x
S
1
1
[
4
5
]
x
x
S
1
2
[
8
]
x
x
S
1
3
[
4
6
]
x
x
S
1
4
[
2
9
]
x
x
S
1
5
[
4
7
]
x
x
S
1
6
[
4
8
]
x
x
4
.
3
.
Ans
w
er
s
t
o
r
esea
rc
h
q
ues
t
io
ns
a.
R
Q1
.
W
h
at
ar
e
th
e
d
if
f
er
en
t a
p
p
licatio
n
s
o
f
t
h
e
s
eq
u
e
n
ce
-
to
-
s
eq
u
en
ce
n
eu
r
al
n
et
w
o
r
k
m
o
d
el
?
As
s
ee
n
f
r
o
m
E
rr
o
r!
Ref
er
e
nce
s
o
urce
no
t
f
o
un
d.
,
1
1
,
o
r
6
8
.
7
5
%,
o
f
th
e
s
tu
d
ie
s
w
er
e
r
elev
an
t
to
th
e
ap
p
licatio
n
s
o
f
s
eq
u
en
ce
-
to
-
s
eq
u
e
n
ce
n
eu
r
al
n
et
w
o
r
k
m
o
d
el
s
,
in
d
icati
n
g
n
o
t
o
n
l
y
t
h
e
r
elev
a
n
ce
o
f
t
h
i
s
q
u
esti
o
n
b
u
t
also
it
s
w
id
e
s
p
r
ea
d
in
ter
est
in
t
h
e
f
ield
.
T
h
e
g
en
er
al
co
n
s
en
s
u
s
w
a
s
t
h
at
s
eq
u
en
ce
-
to
-
s
eq
u
e
n
c
e
m
o
d
el
s
w
er
e
b
est
u
tili
ze
d
f
o
r
s
p
ee
ch
r
ec
o
g
n
itio
n
an
d
g
e
n
e
r
al
lin
g
u
i
s
tics
,
as
s
u
g
g
e
s
ted
b
y
[
1
6
,
3
1
,
4
8
]
.
I
n
ad
d
itio
n
,
s
eq
u
e
n
ce
-
to
-
s
eq
u
en
c
e
m
o
d
els
ca
n
b
e
u
s
ed
f
o
r
v
i
d
eo
to
tex
t
co
n
v
er
s
io
n
[
4
6
]
an
d
h
a
n
d
li
n
g
lar
g
e
v
o
ca
b
u
lar
ies,
o
p
ti
m
iz
in
g
tr
an
s
latio
n
p
er
f
o
r
m
a
n
ce
,
an
d
m
u
lti
-
lin
g
u
al
lear
n
i
n
g
[
2
7
]
.
b.
R
Q2
.
Ho
w
h
as t
h
is
m
o
d
el
b
ee
n
i
m
p
le
m
e
n
ted
an
d
d
ev
elo
p
ed
?
A
ll
1
6
o
f
t
h
e
p
ap
er
s
u
s
ed
i
n
t
h
e
s
y
s
te
m
atic
r
e
v
ie
w
m
en
tio
n
ed
d
if
f
er
en
t
ap
p
r
o
ac
h
es
to
i
m
p
le
m
en
t
in
g
an
d
d
ev
elo
p
in
g
t
h
e
s
eq
u
e
n
ce
-
to
-
s
eq
u
e
n
ce
n
e
u
r
al
n
et
w
o
r
k
m
o
d
el
s
.
Fo
r
ex
a
m
p
le:
R
.
P
r
a
b
h
av
al
k
ar
et
a
l
.
[
3
1
]
s
u
m
m
ar
is
e
s
th
r
ee
m
et
h
o
d
s
o
f
i
m
p
le
m
en
ta
tio
n
,
w
h
ich
i
n
cl
u
d
e:
R
NN
s
,
C
o
n
n
ec
tio
n
i
s
t
T
em
p
o
r
al
C
lass
i
f
icatio
n
s
(
C
T
C
)
,
an
d
A
tte
n
tio
n
m
o
d
els.
c.
R
Q3
.
W
h
at
ar
e
th
e
ad
v
a
n
tag
e
s
an
d
li
m
ita
tio
n
s
o
f
i
m
p
le
m
e
n
ti
n
g
s
eq
u
e
n
ce
-
to
-
s
eq
u
en
ce
m
o
d
els?
I
.
Su
ts
k
e
v
er
et
a
l
.
[8
]
an
d
Y.
H.
C
h
a
n
et
a
l.
[
41]
talk
ed
m
o
s
tl
y
ab
o
u
t
i
m
p
le
m
e
n
tatio
n
s
u
s
in
g
R
N
N
s
an
d
d
is
c
u
s
s
ed
m
a
n
y
ad
v
a
n
ta
g
es
an
d
d
i
s
ad
v
a
n
tag
e
s
o
f
ap
p
ly
in
g
th
i
s
m
o
d
el,
w
h
ile
i
n
[
2
8
,
4
2
]
d
is
cu
s
s
ed
t
h
e
i
m
p
le
m
en
ta
tio
n
t
h
r
o
u
g
h
C
T
C
in
d
etails
w
i
th
t
h
e
li
m
ita
tio
n
s
i
n
i
m
p
le
m
e
n
tatio
n
.
Si
m
ilar
l
y
,
in
[
3
2
,
4
5
]
d
is
cu
s
s
ed
th
e
li
m
ita
tio
n
s
o
f
th
e
atten
tio
n
m
o
d
el
.
d.
R
Q4
.
W
h
at
is
t
h
e
b
est
m
o
d
el
t
o
ap
p
r
o
ac
h
s
eq
u
en
ce
-
to
-
s
eq
u
e
n
ce
i
m
p
le
m
e
n
tatio
n
?
T
h
e
m
aj
o
r
ity
o
f
t
h
e
p
ap
er
s
(
6
2
.
5
%)
talk
ab
o
u
t
R
NNs
an
d
th
eir
im
p
le
m
e
n
tatio
n
,
li
m
it
atio
n
s
,
an
d
en
d
o
r
s
e
m
e
n
t.
No
tab
l
y
,
t
h
e
w
o
r
k
o
f
[
3
1
]
co
m
p
ar
ed
th
e
th
r
ee
d
i
f
f
er
en
t
ap
p
r
o
ac
h
es
an
d
f
o
u
n
d
th
e
m
o
s
t
p
r
o
m
i
s
in
g
ap
p
r
o
ac
h
to
b
e
"
t
h
e
R
N
N
tr
an
s
d
u
ce
r
,
atte
n
tio
n
-
b
a
s
ed
m
o
d
els,
an
d
a
n
o
v
e
l
R
NN
tr
a
n
s
d
u
ce
r
au
g
m
e
n
ted
w
it
h
atten
tio
n
.
"
5.
CO
NCLU
SI
O
N
I
n
co
n
cl
u
s
io
n
,
th
e
p
ap
er
ai
m
ed
to
co
n
d
u
ct
a
s
y
s
te
m
a
tic
r
ev
ie
w
o
n
th
e
to
p
ic
o
f
t
h
e
s
e
q
u
en
ce
-
to
-
s
eq
u
en
ce
n
eu
r
al
n
e
t
w
o
r
k
an
d
its
m
o
d
els.
T
h
e
m
ai
n
ai
m
o
f
th
e
r
ev
ie
w
to
g
a
in
i
n
s
i
g
h
t
in
to
th
e
s
eq
u
en
ce
-
to
-
s
eq
u
en
ce
n
e
u
r
al
n
e
t
w
o
r
k
m
o
d
els
an
d
to
f
i
n
d
th
e
b
est
ap
p
r
o
ac
h
to
i
m
p
le
m
e
n
t
it.
T
h
r
ee
s
u
ch
ap
p
r
o
ac
h
es
w
er
e
f
o
u
n
d
:
th
r
o
u
g
h
r
ec
u
r
r
en
t
n
e
u
r
al
n
et
w
o
r
k
s
,
co
n
n
ec
t
io
n
is
t
t
e
m
p
o
r
al
class
if
icatio
n
s
(
C
T
C
)
,
an
d
atten
tio
n
m
o
d
el
s
.
T
h
e
r
esear
c
h
q
u
esti
o
n
d
er
iv
ed
f
o
r
t
h
e
l
iter
atu
r
e
r
ev
ie
w
w
er
e
e
n
co
m
p
as
s
i
n
g
th
e
ap
p
licatio
n
s
o
f
s
eq
u
en
ce
-
to
-
s
eq
u
e
n
ce
m
o
d
els
,
th
eir
ad
v
a
n
ta
g
es
a
n
d
d
is
ad
v
an
ta
g
e
s
,
as
w
el
l
as
t
h
e
b
e
s
t
i
m
p
le
m
e
n
tatio
n
ap
p
r
o
ac
h
f
o
r
th
e
m
.
T
h
e
p
r
o
ce
d
u
r
e
d
o
n
e
to
co
n
d
u
ct
t
h
is
s
y
s
te
m
atic
l
iter
atu
r
e
r
e
v
ie
w
in
cl
u
d
ed
u
s
in
g
t
h
e
r
esear
ch
q
u
est
io
n
s
to
d
er
iv
e
k
e
y
w
o
r
d
s
th
at
w
er
e
t
h
e
n
u
s
ed
t
o
lo
o
k
at
th
e
s
u
b
s
eq
u
e
n
t
d
atab
ase,
p
ee
r
-
r
ev
ie
w
ed
j
o
u
r
n
als,
an
d
p
er
io
d
icals.
Mo
s
t
o
f
p
ap
er
s
u
tili
ze
d
f
r
o
m
ar
Xi
v
,
Go
o
g
le
Sc
h
o
lar
,
Scien
ce
D
ir
ec
t,
I
E
E
E
Xp
lo
r
e
an
d
Sp
r
in
g
er
co
m
p
lete
j
o
u
r
n
a
ls
.
T
h
r
o
u
g
h
i
n
itia
l
s
ea
r
c
h
es,
7
9
0
p
ap
er
s
an
d
ac
ad
e
m
ic
w
o
r
k
s
w
er
e
f
o
u
n
d
,
an
d
w
it
h
th
e
h
elp
o
f
s
elec
tio
n
cr
it
er
ia
an
d
P
R
I
SMA
p
r
o
ce
d
u
r
e,
th
e
n
u
m
b
er
o
f
p
ap
er
s
r
ev
ie
w
e
d
in
th
is
p
ap
e
r
w
a
s
r
ed
u
ce
d
to
1
6
.
E
ac
h
o
f
t
h
e
1
6
p
ap
er
s
w
as
ca
teg
o
r
ized
b
y
th
eir
co
n
tr
ib
u
tio
n
to
ea
ch
r
es
ea
r
ch
q
u
es
tio
n
,
a
n
d
th
e
y
w
er
e
an
al
y
ze
d
.
Fin
all
y
,
t
h
e
r
esear
ch
p
ap
er
s
u
n
d
er
w
e
n
t
a
q
u
ality
as
s
es
s
m
e
n
t
w
h
er
e
th
e
r
esu
lti
n
g
r
an
g
e
w
a
s
f
r
o
m
8
3
.
3
% to
1
0
0
%.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
11
,
No
.
3
,
J
u
n
e
2
0
2
1
:
2
3
1
5
-
2326
2322
AP
P
E
NDI
X
Ta
b
le
7
.
A
n
al
y
s
i
s
o
f
th
e
l
iter
at
u
r
e
r
ev
ie
w
S
t
u
d
y
S
o
u
r
c
e
P
u
r
p
o
se
C
o
u
n
t
r
y
D
a
t
a
b
a
se
F
i
n
d
i
n
g
s
S1
[
4
1
]
T
h
e
a
u
t
h
o
r
s
su
g
g
e
st
e
d
a
n
d
e
x
p
e
r
i
me
n
t
e
d
w
i
t
h
a
n
a
l
t
e
r
n
a
t
i
v
e
a
p
p
r
o
a
c
h
t
o
e
mb
e
d
d
e
d
e
mo
t
i
o
n
a
l
i
n
f
o
r
mat
i
o
n
a
t
t
h
e
e
n
c
o
d
e
r
st
a
g
e
o
f
se
q
u
e
n
c
e
t
o
se
q
u
e
n
c
e
b
a
se
d
e
mo
t
i
o
n
a
l
g
e
n
e
r
a
t
i
o
n
.
H
o
n
g
K
o
n
g
I
EEE
X
p
l
o
r
e
D
i
f
f
e
r
e
n
t
me
t
h
o
d
s
w
e
r
e
t
e
st
e
d
o
n
e
mo
t
i
o
n
a
l
e
n
c
o
d
i
n
g
i
n
f
o
r
mat
i
o
n
f
o
r
se
q
u
e
n
c
e
t
o
se
q
u
e
n
c
e
g
e
n
e
r
a
t
i
o
n
r
e
sp
o
n
se
a
n
d
e
v
a
l
u
a
t
e
d
t
h
e
r
e
su
l
t
,
w
h
i
c
h
w
a
s
f
o
u
n
d
t
o
h
a
v
e
a
p
o
si
t
i
v
e
e
f
f
e
c
t
o
n
a
se
n
t
e
n
c
e
l
e
v
e
l
.
S2
[
3
2
]
T
h
e
p
a
p
e
r
p
r
o
p
o
se
d
a
n
e
w
me
t
h
o
d
u
se
d
i
n
n
e
u
r
a
l
sp
e
e
c
h
r
e
c
o
g
n
i
t
i
o
n
b
y
a
l
i
g
n
i
n
g
t
h
e
a
t
t
e
n
t
i
o
n
mo
d
e
l
i
n
g
w
i
t
h
i
n
t
h
e
C
T
C
f
r
a
me
w
o
r
k
C
a
n
a
d
a
I
EEE
X
p
l
o
r
e
T
h
e
p
r
o
p
o
se
d
me
t
h
o
d
b
o
o
st
e
d
t
h
e
e
n
d
-
to
-
e
n
d
a
c
o
u
s
t
i
c
-
to
-
w
o
r
d
C
T
C
mo
d
e
l
a
n
d
a
c
h
i
e
v
e
d
b
e
t
t
e
r
W
ER
t
h
a
n
t
h
e
t
r
a
d
i
t
i
o
n
a
l
c
o
n
t
e
x
t
-
d
e
p
e
n
d
e
n
t
p
h
o
n
e
m
e
C
T
C
m
o
d
e
l
d
e
c
o
d
e
d
w
i
t
h
a
v
e
r
y
l
a
r
g
e
-
si
z
e
d
l
a
n
g
u
a
g
e
mo
d
e
l
S3
[
2
8
]
T
h
e
p
a
p
e
r
p
r
e
se
n
t
e
d
a
n
o
v
e
l
me
t
h
o
d
f
o
r
t
r
a
i
n
i
n
g
R
N
N
s
t
o
l
a
b
e
l
u
n
se
g
me
n
t
e
d
se
q
u
e
n
c
e
s.
"
A
n
e
x
p
e
r
i
me
n
t
o
n
t
h
e
TI
M
I
T
sp
e
e
c
h
c
o
r
p
u
s
d
e
mo
n
st
r
a
t
e
s
i
t
s
a
d
v
a
n
t
a
g
e
s
o
v
e
r
b
o
t
h
a
b
a
s
e
l
i
n
e
H
M
M
a
n
d
a
h
y
b
r
i
d
H
M
M
-
R
N
N
"
.
S
w
i
t
z
e
r
l
a
n
d
A
C
M
D
i
g
i
t
a
l
L
i
b
r
a
r
i
e
s
T
h
e
me
t
h
o
d
i
s
d
e
r
i
v
e
d
f
r
o
m
p
r
o
b
a
b
i
l
i
st
i
c
p
r
i
n
c
i
p
l
e
s
a
n
d
f
i
t
s
t
h
e
f
r
a
mew
o
r
k
o
f
t
h
e
n
e
u
r
a
l
n
e
t
w
o
r
k
c
l
a
ssi
f
i
e
r
.
I
t
a
l
l
o
w
s
t
h
e
n
e
t
w
o
r
k
t
o
b
e
t
r
a
i
n
e
d
d
i
r
e
c
t
l
y
f
o
r
se
q
u
e
n
c
e
l
a
b
e
l
i
n
g
b
y
r
e
mo
v
i
n
g
t
h
e
r
e
q
u
i
r
e
me
n
t
o
f
p
r
e
-
se
g
m
e
n
t
e
d
d
a
t
a
.
S4
[
4
2
]
T
h
e
a
u
t
h
o
r
s
d
e
v
e
l
o
p
e
d
a
t
e
c
h
n
i
q
u
e
t
o
b
a
se
t
h
e
R
N
N
-
T
sy
st
e
m
t
o
w
a
r
d
s
a
s
p
e
c
i
f
i
c
k
e
y
w
o
r
d
.
U
S
A
I
EEE
X
p
l
o
r
e
T
h
e
p
a
p
e
r
d
e
v
e
l
o
p
e
d
a
st
r
e
a
mi
n
g
k
e
y
w
o
r
d
b
y
u
si
n
g
(
R
N
N
-
T
)
mo
d
e
l
t
o
p
r
e
d
i
c
t
e
i
t
h
e
r
p
h
o
n
e
me
s
o
r
g
r
a
p
h
e
m
e
s
a
s
s
u
b
-
w
o
r
d
,
w
h
i
c
h
a
l
l
o
w
s
d
e
t
e
c
t
i
n
g
r
a
n
d
o
m
e
x
p
r
e
ssi
o
n
s.
S5
[
4
3
]
T
h
e
a
u
t
h
o
r
s p
r
o
p
o
se
d
a
n
a
r
t
i
f
i
c
i
a
l
B
a
n
g
l
a
t
e
x
t
g
e
n
e
r
a
t
o
r
w
i
t
h
L
S
T
M
a
n
d
mo
d
e
l
i
t
t
o
v
a
l
i
d
a
t
e
t
h
e
a
c
c
u
r
a
c
y
o
f
t
h
e
t
e
x
t
g
e
n
e
r
a
t
o
r
s.
B
a
n
g
l
a
d
e
sh
S
c
i
e
n
c
e
D
i
r
e
c
t
T
h
e
p
a
p
e
r
w
o
r
k
e
d
w
i
t
h
R
N
N
s
t
r
u
c
t
u
r
e
s
a
n
d
L
T
S
M
n
e
t
w
o
r
k
s
t
o
b
e
a
b
l
e
t
o
f
u
l
f
i
l
l
t
h
e
i
r
p
u
r
p
o
s
e
.
T
h
e
y
w
e
r
e
a
b
l
e
t
o
"
mo
d
e
l
f
o
r
a
c
h
i
e
v
i
n
g
m
u
l
t
i
-
t
a
s
k
s
e
q
u
e
n
c
e
t
o
s
e
q
u
e
n
c
e
t
e
x
t
g
e
n
e
r
a
t
i
o
n
a
n
d
m
u
l
t
i
-
w
a
y
t
r
a
n
s
l
a
t
i
o
n
l
i
k
e
B
e
n
g
a
l
i
a
r
t
i
c
l
e
s,
c
a
p
t
i
o
n
g
e
n
e
r
a
t
i
o
n
"
S6
[
2
7
]
T
h
e
a
u
t
h
o
r
i
n
t
r
o
d
u
c
e
d
'n
e
u
r
a
l
mac
h
i
n
e
t
r
a
n
sl
a
t
i
o
n
'
o
r
'n
e
u
r
a
l
se
q
u
e
n
c
e
-
to
-
se
q
u
e
n
c
e
mo
d
e
l
s'
U
S
A
a
r
X
i
v
T
h
e
p
a
p
e
r
c
o
v
e
r
e
d
t
h
e
b
a
si
c
s
o
f
n
e
u
r
a
l
mac
h
i
n
e
t
r
a
n
sl
a
t
i
o
n
a
n
d
se
q
u
e
n
c
e
t
o
se
q
u
e
n
c
e
mo
d
e
l
s.
I
t
c
o
v
e
r
e
d
se
v
e
r
a
l
a
p
p
l
i
c
a
t
i
o
n
s
u
c
h
a
s
H
a
n
d
l
i
n
g
l
a
r
g
e
v
o
c
a
b
u
l
a
r
i
e
s,
O
p
t
i
mi
z
i
n
g
t
r
a
n
sl
a
t
i
o
n
p
e
r
f
o
r
ma
n
c
e
,
a
n
d
M
u
l
t
i
-
l
i
n
g
u
a
l
l
e
a
r
n
i
n
g
.
S7
[
1
6
]
T
h
i
s
p
a
p
e
r
c
a
r
r
i
e
s
o
u
t
a
c
o
mp
r
e
h
e
n
si
v
e
r
e
v
i
e
w
o
f
a
r
t
i
c
l
e
s
t
h
a
t
i
n
v
o
l
v
e
a
c
o
mp
a
r
a
t
i
v
e
st
u
d
y
o
f
f
e
e
d
f
o
r
w
a
r
d
n
e
u
r
a
l
n
e
t
w
o
r
k
s
a
n
d
s
t
a
t
i
st
i
c
a
l
t
e
c
h
n
i
q
u
e
s u
se
d
f
o
r
p
r
e
d
i
c
t
i
o
n
a
n
d
c
l
a
ss
i
f
i
c
a
t
i
o
n
p
r
o
b
l
e
m
s
i
n
v
a
r
i
o
u
s
a
r
e
a
s
o
f
a
p
p
l
i
c
a
t
i
o
n
s
.
T
a
b
u
l
a
r
p
r
e
s
e
n
t
a
t
i
o
n
s
h
i
g
h
l
i
g
h
t
i
n
g
t
h
e
i
m
p
o
r
t
a
n
t
f
e
a
t
u
r
e
s
o
f
t
h
e
s
e
a
r
t
i
c
l
e
s a
r
e
a
l
so
p
r
o
v
i
d
e
d
I
n
d
i
a
S
c
i
e
n
c
e
D
i
r
e
c
t
"
N
e
u
r
a
l
n
e
t
w
o
r
k
s
a
r
e
b
e
i
n
g
u
se
d
i
n
a
r
e
a
s
o
f
p
r
e
d
i
c
t
i
o
n
a
n
d
c
l
a
ss
i
f
i
c
a
t
i
o
n
,
t
h
e
a
r
e
a
s
w
h
e
r
e
st
a
t
i
s
t
i
c
a
l
me
t
h
o
d
s
h
a
v
e
t
r
a
d
i
t
i
o
n
a
l
l
y
b
e
e
n
u
se
d
".
"
O
n
e
o
f
t
h
e
i
mp
o
r
t
a
n
t
a
d
v
a
n
t
a
g
e
s
o
f
n
e
u
r
a
l
n
e
t
w
o
r
k
s
c
i
t
e
d
i
n
t
h
e
l
i
t
e
r
a
t
u
r
e
i
s
t
h
a
t
i
t
c
a
n
a
u
t
o
ma
t
i
c
a
l
l
y
a
p
p
r
o
x
i
m
a
t
e
a
n
y
n
o
n
l
i
n
e
a
r
ma
t
h
e
ma
t
i
c
a
l
f
u
n
c
t
i
o
n
".
S8
[
3
1
]
T
h
e
a
u
t
h
o
r
s
c
o
n
d
u
c
t
e
d
a
d
e
t
a
i
l
e
d
e
v
a
l
u
a
t
i
o
n
o
f
se
q
u
e
n
c
e
t
o
se
q
u
e
n
c
e
mo
d
e
l
s
w
h
i
c
h
a
r
e
u
s
e
d
i
n
t
h
e
t
a
s
k
o
f
s
p
e
e
c
h
r
e
c
o
g
n
i
t
i
o
n
s
u
c
h
a
s
c
o
n
n
e
c
t
i
o
n
i
s
t
t
e
m
p
o
r
a
l
c
l
a
s
s
i
f
i
c
a
t
i
o
n
(
C
T
C
)
,
t
h
e
r
e
c
u
r
r
e
n
t
n
e
u
r
a
l
n
e
t
w
o
r
k
(
R
N
N
)
t
r
a
n
s
d
u
c
e
r
,
a
n
a
t
t
e
n
t
i
o
n
-
b
a
se
d
mo
d
e
l
,
a
n
d
a
mo
d
e
l
w
h
i
c
h
a
u
g
me
n
t
s
t
h
e
R
N
N
t
r
a
n
s
d
u
c
e
r
w
i
t
h
a
n
a
t
t
e
n
t
i
o
n
me
c
h
a
n
i
s
m.
U
S
A
G
o
o
g
l
e
S
c
h
o
l
a
r
"
[
t
h
e
y
]
c
o
mp
a
r
e
d
a
n
u
m
b
e
r
o
f
s
e
q
u
e
n
c
e
-
to
-
se
q
u
e
n
c
e
mo
d
e
l
i
n
g
a
p
p
r
o
a
c
h
e
s
o
n
a
n
L
V
C
S
R
t
a
s
k
.
I
n
e
x
p
e
r
i
men
t
a
l
e
v
a
l
u
a
t
i
o
n
s
,
w
e
f
i
n
d
t
h
a
t
t
h
e
R
N
N
t
r
a
n
s
d
u
c
e
r
,
a
t
t
e
n
t
i
o
n
-
b
a
se
d
mo
d
e
l
s
a
n
d
a
n
o
v
e
l
R
N
N
t
r
a
n
sd
u
c
e
r
a
u
g
me
n
t
e
d
w
i
t
h
a
t
t
e
n
t
i
o
n
a
r
e
c
o
m
p
a
r
a
b
l
e
i
n
p
e
r
f
o
r
man
c
e
t
o
a
st
r
o
n
g
st
a
t
e
-
of
-
t
h
e
-
a
r
t
b
a
s
e
l
i
n
e
o
n
a
d
i
c
t
a
t
i
o
n
t
e
s
t
s
e
t
,
e
v
e
n
w
h
e
n
e
v
a
l
u
a
t
e
d
w
i
t
h
o
u
t
t
h
e
u
s
e
o
f
a
n
e
x
t
e
r
n
a
l
p
r
o
n
u
n
c
i
a
t
i
o
n
o
r
l
a
n
g
u
a
g
e
mo
d
e
l
"
S9
[
2
6
]
"
W
e
i
n
t
r
o
d
u
c
e
a
n
e
n
c
o
d
e
r
-
d
e
c
o
d
e
r
r
e
c
u
r
r
e
n
t
n
e
u
r
a
l
n
e
t
w
o
r
k
mo
d
e
l
c
a
l
l
e
d
r
e
c
u
r
r
e
n
t
n
e
u
r
a
l
a
l
i
g
n
e
r
(
R
N
A
)
t
h
a
t
c
a
n
b
e
u
s
e
d
f
o
r
t
h
e
s
e
q
u
e
n
c
e
t
o
s
e
q
u
e
n
c
e
m
a
p
p
i
n
g
t
a
s
k
s
.
L
i
k
e
c
o
n
n
e
c
t
i
o
n
i
s
t
t
e
m
p
o
r
a
l
c
l
a
s
s
i
f
i
c
a
t
i
o
n
(
C
T
C
)
m
o
d
e
l
s
,
R
N
A
d
e
f
i
n
e
s
a
p
r
o
b
a
b
i
l
i
t
y
d
i
st
r
i
b
u
t
i
o
n
o
v
e
r
t
a
r
g
e
t
l
a
b
e
l
s
e
q
u
e
n
c
e
s,
i
n
c
l
u
d
i
n
g
b
l
a
n
k
l
a
b
e
l
s
c
o
r
r
e
sp
o
n
d
i
n
g
t
o
e
a
c
h
t
i
me
st
e
p
i
n
t
h
e
i
n
p
u
t
".
U
S
A
R
e
se
a
r
c
h
G
a
t
e
"
W
e
p
r
e
se
n
t
e
d
t
h
e
R
N
A
mo
d
e
l
,
w
h
i
c
h
i
s
a
r
e
c
u
r
r
e
n
t
n
e
u
r
a
l
n
e
t
w
o
r
k
mo
d
e
l
i
n
t
h
e
e
n
c
o
d
e
r
-
d
e
c
o
d
e
r
f
r
a
mew
o
r
k
.
W
e
a
p
p
l
i
e
d
i
t
t
o
e
n
d
-
to
-
e
n
d
sp
e
e
c
h
r
e
c
o
g
n
i
t
i
o
n
a
n
d
sh
o
w
e
d
i
n
i
t
i
a
l
e
x
p
e
r
i
me
n
t
a
l
r
e
su
l
t
s
o
n
Y
o
u
T
u
b
e
t
r
a
n
s
c
r
i
p
t
i
o
n
a
n
d
m
o
b
i
l
e
d
i
c
t
a
t
i
o
n
t
a
s
k
s
.
W
e
f
o
u
n
d
t
h
a
t
t
h
e
R
N
A
g
r
a
p
h
e
m
e
m
o
d
e
l
c
a
n
n
o
t
p
e
r
f
o
r
m
a
s
w
e
l
l
a
s
C
T
C
C
D
p
h
o
n
e
mo
d
e
l
s
f
o
r
mo
b
i
l
e
d
i
c
t
a
t
i
o
n
t
a
s
k
w
h
e
r
e
t
h
e
v
o
c
a
b
u
l
a
r
y
s
i
z
e
i
s
l
a
r
g
e
r
,
a
n
d
t
h
e
t
r
a
i
n
i
n
g
d
a
t
a
i
s
r
e
l
a
t
i
v
e
l
y
smal
l
e
r
.
"
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
A
s
ystema
tic
r
ev
ie
w
o
n
s
eq
u
en
ce
-
to
-
s
eq
u
en
ce
le
a
r
n
in
g
w
ith
n
eu
r
a
l n
etw
o
r
k
a
n
d
its
mo
d
els
(
Ha
n
a
Yo
u
s
u
f
)
2323
T
ab
le
7
.
A
n
al
y
s
i
s
o
f
th
e
l
iter
at
u
r
e
r
ev
i
ew
(
C
o
n
tin
u
e
)
S
t
u
d
y
S
o
u
r
c
e
P
u
r
p
o
se
C
o
u
n
t
r
y
D
a
t
a
b
a
se
F
i
n
d
i
n
g
s
S
1
0
[
4
4
]
“
W
e
p
r
o
p
o
se
S
A
N
-
C
T
C
,
a
d
e
e
p
,
f
u
l
l
y
s
e
l
f
-
a
t
t
e
n
t
i
o
n
a
l
n
e
t
w
o
r
k
f
o
r
C
T
C
,
a
n
d
sh
o
w
i
t
i
s
t
r
a
c
t
a
b
l
e
a
n
d
c
o
mp
e
t
i
t
i
v
e
f
o
r
e
n
d
-
to
-
e
n
d
s
p
e
e
c
h
r
e
c
o
g
n
i
t
i
o
n
.
W
e
mo
t
i
v
a
t
e
t
h
e
a
r
c
h
i
t
e
c
t
u
r
e
f
o
r
sp
e
e
c
h
,
e
v
a
l
u
a
t
e
t
h
e
p
o
s
i
t
i
o
n
a
n
d
d
o
w
n
-
s
a
m
p
l
i
n
g
a
p
p
r
o
a
c
h
e
s
,
a
n
d
e
x
p
l
o
r
e
h
o
w
t
o
l
a
b
e
l
a
l
p
h
a
b
e
t
s
(
c
h
a
r
a
c
t
e
r
,
p
h
o
n
e
m
e
,
s
u
b
w
o
r
d
)
a
f
f
e
c
t
a
t
t
e
n
t
i
o
n
h
e
a
d
s
a
n
d
p
e
r
f
o
r
man
c
e
”
.
UK
I
EEE
X
p
l
o
r
e
"
S
A
N
-
C
T
C
t
r
a
i
n
s
q
u
i
c
k
l
y
a
n
d
o
u
t
p
e
r
f
o
r
ms
e
x
i
st
i
n
g
C
T
C
mo
d
e
l
s
a
n
d
mo
st
e
n
c
o
d
e
r
-
d
e
c
o
d
e
r
mo
d
e
l
s,
w
i
t
h
c
h
a
r
a
c
t
e
r
e
r
r
o
r
r
a
t
e
s
(
C
ER
s)
o
f
4
.
7
%
i
n
1
d
a
y
o
n
W
S
J
e
v
a
l
9
2
a
n
d
2
.
8
%
i
n
1
w
e
e
k
o
n
L
i
b
r
i
S
p
e
e
c
h
t
e
st
-
c
l
e
a
n
,
w
i
t
h
a
f
i
x
e
d
a
r
c
h
i
t
e
c
t
u
r
e
a
n
d
o
n
e
G
P
U
.
S
i
mi
l
a
r
i
mp
r
o
v
e
me
n
t
s
h
o
l
d
f
o
r
W
ER
s
a
f
t
e
r
L
M
d
e
c
o
d
i
n
g
.
S
1
1
[
4
5
]
T
h
e
w
o
r
k
i
s
f
o
c
u
s
e
d
o
n
u
s
i
n
g
se
q
u
e
n
c
e
t
o
s
e
q
u
e
n
c
e
t
h
e
a
t
t
e
n
t
i
o
n
m
o
d
e
l
b
y
t
h
e
G
o
o
g
l
e
b
r
a
i
n
t
e
a
m
t
o
g
e
n
e
r
a
t
e
t
h
e
a
b
s
t
r
a
c
t
o
f
r
e
s
e
a
r
c
h
p
a
p
e
r
s
.
M
o
r
e
o
v
e
r
,
t
e
m
p
o
r
a
l
a
t
t
e
n
t
i
o
n
m
e
c
h
a
n
i
s
m
h
a
s
b
e
e
n
u
s
e
d
i
n
r
e
p
l
a
c
e
m
e
n
t
t
o
t
h
e
g
l
o
b
a
l
a
t
t
e
n
t
i
o
n
t
o
c
a
t
e
r
t
o
t
h
e
p
r
o
b
l
e
m
o
f
r
e
p
e
t
i
t
i
v
e
w
o
r
d
s.
P
a
k
i
st
a
n
I
EEE
X
p
l
o
r
e
T
h
e
r
e
su
l
t
i
n
d
i
c
a
t
e
s
t
h
a
t
t
h
e
t
e
mp
o
r
a
l
a
t
t
e
n
t
i
o
n
mo
d
e
l
i
s
a
u
se
f
u
l
m
e
t
h
o
d
f
o
r
g
e
n
e
r
a
t
i
n
g
su
mm
a
r
i
e
s.
M
o
r
e
o
v
e
r
,
r
e
s
u
l
t
s
i
n
d
i
c
a
t
e
t
h
a
t
w
i
t
h
t
h
e
i
n
c
r
e
a
se
i
n
d
a
t
a
se
t
si
z
e
,
t
h
e
a
c
c
u
r
a
c
y
o
f
t
h
e
r
e
su
l
t
s
a
l
so
i
n
c
r
e
a
se
s.
S
1
2
[
8
]
T
h
e
p
a
p
e
r
u
se
s
a
mu
l
t
i
l
a
y
e
r
e
d
l
o
n
g
sh
o
r
t
-
t
e
r
m
me
mo
r
y
(
L
S
T
M
)
t
o
map
t
h
e
i
n
p
u
t
se
q
u
e
n
c
e
t
o
a
v
e
c
t
o
r
o
f
f
i
x
e
d
d
i
me
n
si
o
n
a
l
i
t
y
.
T
h
e
n
,
a
n
o
t
h
e
r
d
e
e
p
L
S
T
M
t
o
d
e
c
o
d
e
t
h
e
t
a
r
g
e
t
se
q
u
e
n
c
e
f
r
o
m
t
h
e
v
e
c
t
o
r
.
"
C
a
n
a
d
a
G
o
o
g
l
e
S
c
h
o
l
a
r
T
h
e
mai
n
r
e
su
l
t
i
s
t
h
a
t
o
n
a
n
En
g
l
i
s
h
t
o
F
r
e
n
c
h
t
r
a
n
sl
a
t
i
o
n
t
a
s
k
f
r
o
m
t
h
e
W
M
T
-
14
d
a
t
a
s
e
t
,
t
h
e
t
r
a
n
s
l
a
t
i
o
n
s
p
r
o
d
u
c
e
d
b
y
t
h
e
L
S
T
M
a
c
h
i
e
v
e
a
B
L
E
U
s
c
o
r
e
o
f
3
4
.
8
o
n
t
h
e
e
n
t
i
r
e
t
e
s
t
s
e
t
,
w
h
e
r
e
t
h
e
L
S
T
M
'
s
B
L
E
U
s
c
o
r
e
w
a
s
p
e
n
a
l
i
z
e
d
o
n
o
u
t
-
of
-
v
o
c
a
b
u
l
a
r
y
w
o
r
d
s
.
A
d
d
i
t
i
o
n
a
l
l
y
,
t
h
e
L
S
T
M
d
i
d
n
o
t
h
a
v
e
d
i
f
f
i
c
u
l
t
y
i
n
l
o
n
g
se
n
t
e
n
c
e
s.
S1
3
[
4
6
]
"
W
e
p
r
o
p
o
se
a
n
o
v
e
l
e
n
d
-
to
-
e
n
d
se
q
u
e
n
c
e
-
to
-
se
q
u
e
n
c
e
mo
d
e
l
t
o
g
e
n
e
r
a
t
e
c
a
p
t
i
o
n
s
f
o
r
v
i
d
e
o
s.
F
o
r
t
h
i
s,
w
e
e
x
p
l
o
i
t
r
e
c
u
r
r
e
n
t
n
e
u
r
a
l
n
e
t
-
w
o
r
k
s,
sp
e
c
i
f
i
c
a
l
l
y
L
S
T
M
s,
w
h
i
c
h
h
a
v
e
d
e
mo
n
st
r
a
t
e
d
st
a
t
e
-
of
-
t
h
e
-
a
r
t
p
e
r
f
o
r
man
c
e
i
n
i
m
a
g
e
c
a
p
t
i
o
n
g
e
n
e
r
a
t
i
o
n
.
"
U
S
A
I
EEE
X
p
l
o
r
e
[
T
h
i
s
p
a
p
e
r
]
c
o
n
s
t
r
u
c
t
e
d
d
e
scri
p
t
i
o
n
s
u
si
n
g
a
se
q
u
e
n
c
e
t
o
se
q
u
e
n
c
e
mo
d
e
l
,
w
h
e
r
e
f
r
a
me
s
a
r
e
f
i
r
st
t
o
r
e
a
d
se
q
u
e
n
t
i
a
l
l
y
,
a
n
d
t
h
e
n
w
o
r
d
s
a
r
e
g
e
n
e
r
a
t
e
d
se
q
u
e
n
t
i
a
l
l
y
.
T
h
i
s
a
l
l
o
w
s
u
s
t
o
h
a
n
d
l
e
v
a
r
i
a
b
l
e
-
l
e
n
g
t
h
i
n
p
u
t
a
n
d
o
u
t
p
u
t
w
h
i
l
e
si
mu
l
t
a
n
e
o
u
sl
y
mo
d
e
l
i
n
g
t
e
mp
o
r
a
l
st
r
u
c
t
u
r
e
.
O
u
r
m
o
d
e
l
a
c
h
i
e
v
e
s
s
t
a
t
e
-
of
-
t
h
e
-
a
r
t
p
e
r
f
o
r
m
a
n
c
e
o
n
t
h
e
M
S
V
D
d
a
t
a
s
e
t
,
a
n
d
o
u
t
p
e
r
f
o
r
m
s
r
e
l
a
t
e
d
w
o
r
k
o
n
t
w
o
l
a
r
g
e
a
n
d
c
h
a
l
l
e
n
g
i
n
g
m
o
v
i
e
-
d
e
s
c
r
i
p
t
i
o
n
d
a
t
a
se
t
s."
S
1
4
[
2
9
]
"
B
a
si
c
b
a
c
k
p
r
o
p
a
g
a
t
i
o
n
,
w
h
i
c
h
i
s
a
si
mp
l
e
me
t
h
o
d
n
o
w
b
e
i
n
g
w
i
d
e
l
y
u
se
d
i
n
a
r
e
a
s
l
i
k
e
p
a
t
t
e
r
n
r
e
c
o
g
n
i
t
i
o
n
a
n
d
f
a
u
l
t
d
i
a
g
n
o
si
s,
i
s
r
e
v
i
e
w
e
d
.
T
h
e
b
a
si
c
e
q
u
a
t
i
o
n
s
f
o
r
b
a
c
k
p
r
o
p
a
g
a
t
i
o
n
t
h
r
o
u
g
h
t
i
me
,
a
n
d
a
p
p
l
i
c
a
t
i
o
n
s
t
o
a
r
e
a
s
l
i
k
e
p
a
t
t
e
r
n
r
e
c
o
g
n
i
t
i
o
n
i
n
v
o
l
v
i
n
g
d
y
n
a
m
i
c
s
y
s
t
e
m
s
,
s
y
s
t
e
m
s
i
d
e
n
t
i
f
i
c
a
t
i
o
n
,
a
n
d
c
o
n
t
r
o
l
a
r
e
d
i
scu
sse
d
.
"
U
S
A
I
EEE
X
p
l
o
r
e
T
h
i
s
p
a
p
e
r
p
r
e
se
n
t
s
t
h
e
k
e
y
e
q
u
a
t
i
o
n
s
o
f
b
a
c
k
p
r
o
p
a
g
a
t
i
o
n
,
a
s
a
p
p
l
i
e
d
t
o
n
e
u
r
a
l
n
e
t
w
o
r
k
s
o
f
v
a
r
y
i
n
g
d
e
g
r
e
e
s
o
f
c
o
mp
l
e
x
i
t
y
.
I
t
h
a
s
a
l
so
d
i
sc
u
sse
d
o
t
h
e
r
p
a
p
e
r
s
w
h
i
c
h
e
l
a
b
o
r
a
t
e
o
n
t
h
e
e
x
t
e
n
si
o
n
s
o
f
t
h
i
s
me
t
h
o
d
t
o
m
o
r
e
g
e
n
e
r
a
l
a
p
p
l
i
c
a
t
i
o
n
s
a
n
d
so
me
o
f
t
h
e
t
r
a
d
e
o
f
f
s i
n
v
o
l
v
e
d
S
1
5
[
4
7
]
"
T
h
i
s
p
a
p
e
r
e
x
a
mi
n
e
s
t
h
e
mo
st
p
o
p
u
l
a
r
D
N
N
s
a
p
p
r
o
a
c
h
e
s:
L
S
T
M
,
E
n
c
o
d
e
r
-
D
e
c
o
d
e
r
n
e
t
w
o
r
k
a
n
d
M
e
mo
r
y
n
e
t
w
o
r
k
i
n
se
q
u
e
n
c
e
p
r
e
d
i
c
t
i
o
n
f
i
e
l
d
t
o
h
a
n
d
l
e
t
h
e
so
f
t
w
a
r
e
se
q
u
e
n
c
e
l
e
a
r
n
i
n
g
a
n
d
p
r
e
d
i
c
t
i
o
n
t
a
s
k
.
"
C
h
i
n
a
I
EEE
X
p
l
o
r
e
"
O
u
r
r
e
su
l
t
s
d
e
mo
n
st
r
a
t
e
t
h
a
t
a
t
t
e
n
t
i
o
n
me
c
h
a
n
i
sm
d
o
e
s
n
o
t
f
i
t
a
l
l
se
q
2
se
q
p
r
o
b
l
e
ms,
e
sp
e
c
i
a
l
l
y
i
n
a
w
e
a
k
map
p
i
n
g
r
e
l
a
t
i
o
n
sh
i
p
.
A
n
d
a
d
d
i
t
i
o
n
a
l
i
n
f
o
r
mat
i
o
n
c
a
n
b
e
n
e
f
i
t
se
q
u
e
n
c
e
p
r
e
d
i
c
t
i
o
n
i
n
n
e
u
r
a
l
n
e
t
w
o
r
k
s."
S
1
6
[
4
8
]
T
h
e
p
a
p
e
r
p
r
e
se
n
t
e
d
a
c
h
a
r
a
c
t
e
r
l
e
v
e
l
se
q
u
e
n
c
e
t
o
se
q
u
e
n
c
e
t
h
e
l
e
a
r
n
i
n
g
me
t
h
o
d
.
T
h
e
a
u
t
h
o
r
se
t
i
n
a
n
R
N
N
i
n
t
o
a
n
e
n
c
o
d
e
r
-
d
e
c
o
d
e
r
f
r
a
mew
o
r
k
a
n
d
g
e
n
e
r
a
t
e
t
h
e
c
h
a
r
a
c
t
e
r
-
l
e
v
e
l
s
e
q
u
e
n
c
e
i
n
r
e
p
r
e
se
n
t
a
t
i
o
n
a
s
a
n
i
n
p
u
t
C
h
i
n
a
I
EEE
X
p
l
o
r
e
Ex
p
e
r
i
me
n
t
a
l
r
e
su
l
t
s
b
y
t
h
e
a
u
t
h
o
r
s
c
o
n
f
i
r
me
d
t
h
a
t
t
h
e
p
r
o
p
o
se
d
a
p
p
r
o
a
c
h
a
c
h
i
e
v
e
d
p
e
r
f
o
r
man
c
e
c
l
o
se
t
o
c
o
n
v
e
n
t
i
o
n
a
l
w
o
r
d
a
n
d
p
h
r
a
se
-
b
a
se
d
t
r
a
n
sl
a
t
i
o
n
sy
st
e
ms.
T
h
e
p
r
o
p
o
se
d
me
t
h
o
d
s
a
l
l
o
w
s
t
o
"
r
e
a
d
s
q
u
a
n
t
i
z
e
d
c
h
a
r
a
c
t
e
r
s
i
n
t
o
t
h
e
t
r
a
n
sl
a
t
i
o
n
sy
st
e
m,
i
n
s
t
e
a
d
o
f
u
si
n
g
a
p
r
e
d
e
f
i
n
e
d
v
o
c
a
b
u
l
a
r
y
w
i
t
h
a
l
i
mi
t
e
d
n
u
m
b
e
r
o
f
w
o
r
d
s"
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
11
,
No
.
3
,
J
u
n
e
2
0
2
1
:
2
3
1
5
-
2326
2324
ACK
NO
WL
E
D
G
E
M
E
NT
S
T
h
is
w
o
r
k
i
s
a
p
ar
t o
f
a
p
r
o
j
ec
t u
n
d
er
ta
k
en
at
t
h
e
B
r
itis
h
Un
i
v
er
s
it
y
in
Du
b
ai.
RE
F
E
R
E
NC
E
S
[1
]
S.
A
.
S
a
ll
o
u
m
,
M
.
A
lsh
u
ri
d
e
h
,
A
.
El
n
a
g
a
r,
a
n
d
K
.
S
h
a
a
lan
,
"
M
a
c
h
i
n
e
L
e
a
rn
in
g
a
n
d
De
e
p
L
e
a
rn
in
g
T
e
c
h
n
iq
u
e
s
f
o
r
C
y
b
e
rse
c
u
rit
y
:
A
Re
v
ie
w
,
"
Pro
c
e
e
d
in
g
s
o
f
t
h
e
I
n
ter
n
a
t
io
n
a
l
Co
n
fer
e
n
c
e
o
n
Arti
fi
c
ia
l
In
telli
g
e
n
c
e
a
n
d
C
o
mp
u
ter
Vi
sio
n
(
AICV
2
0
2
0
)
,
v
o
l.
1
1
5
3
,
2
0
2
0
,
p
p
.
5
0
-
57
.
[2
]
S.
A
.
S
a
ll
o
u
m
,
M
.
A
lsh
u
rid
e
h
,
A
.
El
n
a
g
a
r,
a
n
d
K.
S
h
a
a
lan
,
"
M
in
in
g
in
Ed
u
c
a
ti
o
n
a
l
Da
ta:
Re
v
i
e
w
a
n
d
F
u
tu
r
e
Dire
c
ti
o
n
s,"
Pro
c
e
e
d
in
g
s
o
f
th
e
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
Arti
fi
c
ia
l
In
tell
ig
e
n
c
e
a
n
d
C
o
mp
u
ter
Vi
si
o
n
(
AICV
2
0
2
0
)
,
v
o
l.
1
1
5
3
,
2
0
2
0
,
p
p
.
9
2
-
1
0
2
.
[3
]
S.
A
.
S
a
ll
o
u
m
,
R.
Kh
a
n
,
a
n
d
K.
S
h
a
a
lan
,
"
A
S
u
rv
e
y
o
f
S
e
m
a
n
ti
c
A
n
a
l
y
sis
A
p
p
ro
a
c
h
e
s,"
Pr
o
c
e
e
d
in
g
s
o
f
th
e
In
ter
n
a
t
io
n
a
l
C
o
n
fer
e
n
c
e
o
n
Arti
f
icia
l
In
telli
g
e
n
c
e
a
n
d
Co
m
p
u
ter
V
isio
n
(
AICV
2
0
2
0
)
,
v
o
l
.
1
1
5
3
,
2
0
2
0
,
p
p
.
6
1
-
70
.
[4
]
A.
Y.
Zain
a
l,
H.
Yo
u
su
f
,
a
n
d
S
.
A
.
S
a
ll
o
u
m
,
"
Di
m
e
n
sio
n
s
o
f
A
g
il
it
y
C
a
p
a
b
il
it
ies
Org
a
n
iza
ti
o
n
a
l
C
o
m
p
e
ti
ti
v
e
n
e
ss
in
S
u
sta
in
i
n
g
,
"
Pro
c
e
e
d
in
g
s
o
f
th
e
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
Ar
ti
fi
c
i
a
l
I
n
telli
g
e
n
c
e
a
n
d
C
o
mp
u
ter
V
isio
n
(
AICV
2
0
2
0
)
,
v
o
l.
1
1
5
3
,
2
0
2
0
,
p
p
.
7
6
2
-
7
7
2
.
[5
]
H.
A
l
G
h
a
n
e
m
,
M
.
S
h
a
n
a
a
,
S
.
S
a
ll
o
u
m
,
a
n
d
K.
S
h
a
a
lan
,
"
T
h
e
Ro
le
o
f
KM
in
En
h
a
n
c
in
g
A
I
A
l
g
o
rit
h
m
s
a
n
d
S
y
st
e
m
s,"
Ad
v
a
n
c
e
s is
S
c
ien
c
e
,
T
e
c
h
n
o
l
o
g
y
a
n
d
En
g
i
n
e
e
rin
g
S
y
ste
m
J
o
u
rn
a
l
,
v
o
l
.
5
,
n
o
.
4
,
p
p
.
3
8
8
-
3
9
6
,
2
0
2
0
.
[6
]
H.
Yo
u
su
f
,
A
.
Y
.
Zain
a
l,
M
.
A
lsh
u
rid
e
h
,
a
n
d
S
.
A
.
S
a
ll
o
u
m
,
"
A
rti
f
icia
l
In
telli
g
e
n
c
e
M
o
d
e
ls
in
P
o
w
e
r
S
y
ste
m
A
n
a
l
y
si
s,"
in
Arti
fi
c
ia
l
In
telli
g
e
n
c
e
fo
r
S
u
sta
in
a
b
le
De
v
e
lo
p
me
n
t:
T
h
e
o
ry
,
Pr
a
c
ti
c
e
a
n
d
F
u
tu
r
e
Ap
p
li
c
a
t
io
n
s
,
v
o
l.
9
1
2
,
p
p
.
2
3
1
-
2
4
2
,
2
0
2
0
.
[7
]
M
.
A
lS
h
a
m
si,
S
.
A
.
S
a
ll
o
u
m
,
M
.
A
lsh
u
rid
e
h
,
a
n
d
S
.
A
b
d
a
ll
a
h
,
"
A
rti
f
i
c
ial
In
telli
g
e
n
c
e
a
n
d
Blo
c
k
c
h
a
in
f
o
r
T
ra
n
sp
a
re
n
c
y
in
G
o
v
e
rn
a
n
c
e
,
"
in
Arti
fi
c
ia
l
I
n
telli
g
e
n
c
e
fo
r
S
u
sta
in
a
b
le
De
v
e
lo
p
me
n
t:
T
h
e
o
ry
,
Pr
a
c
t
ice
a
n
d
F
u
t
u
re
Ap
p
li
c
a
ti
o
n
s
,
v
o
l
.
9
1
2
,
p
p
.
2
1
9
-
2
3
0
,
2
0
2
0
.
[8
]
I
.
S
u
tsk
e
v
e
r,
O.
V
in
y
a
ls,
a
n
d
Q.
V
.
L
e
,
"
S
e
q
u
e
n
c
e
to
se
q
u
e
n
c
e
l
e
a
rn
in
g
w
it
h
n
e
u
ra
l
n
e
tw
o
rk
s,"
i
n
Ad
v
a
n
c
e
s
i
n
n
e
u
ra
l
in
f
o
rm
a
ti
o
n
p
ro
c
e
ss
in
g
sy
ste
ms
,
p
p
.
3
1
0
4
-
3
1
1
2
,
2
0
1
4
.
[9
]
T.
-
S
.
Ng
u
y
e
n
,
S
.
S
tü
k
e
r,
J.
Nie
h
u
e
s,
a
n
d
A
.
W
a
ib
e
l,
"
I
m
p
ro
v
in
g
se
q
u
e
n
c
e
-
to
-
se
q
u
e
n
c
e
sp
e
e
c
h
re
c
o
g
n
it
io
n
train
i
n
g
w
it
h
o
n
-
th
e
-
f
ly
d
a
ta
a
u
g
m
e
n
tatio
n
,
"
ICAS
S
P
2
0
2
0
-
2
0
2
0
IE
EE
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
Aco
u
st
ics
,
S
p
e
e
c
h
a
n
d
S
ig
n
a
l
Pro
c
e
ss
in
g
(
ICAS
S
P)
,
Ba
r
c
e
lo
n
a
,
S
p
a
in
,
2
0
2
0
,
p
p
.
7
6
8
9
-
7
6
9
3
.
[1
0
]
I.
A
re
l,
J.
B.
L
o
o
k
s,
A
.
Z
iae
i,
a
n
d
M
.
L
e
fk
o
w
it
z
,
"
Re
c
o
n
c
il
iatio
n
b
e
tw
e
e
n
si
m
u
lato
r
a
n
d
sp
e
e
c
h
re
c
o
g
n
it
io
n
o
u
tp
u
t
u
si
n
g
se
q
u
e
n
c
e
-
to
-
se
q
u
e
n
c
e
m
a
p
p
in
g
,
"
G
o
o
g
le P
a
ten
ts
,
2
0
2
0
.
[1
1
]
M
.
Blaa
u
w
a
n
d
J.
Bo
n
a
d
a
,
"
S
e
q
u
e
n
c
e
-
to
-
S
e
q
u
e
n
c
e
S
in
g
in
g
S
y
n
th
e
sis
Us
in
g
th
e
F
e
e
d
-
F
o
rw
a
rd
Tran
sf
o
r
m
e
r,
"
in
ICAS
S
P
2
0
2
0
-
2
0
2
0
I
EE
E
I
n
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
Aco
u
st
ics
,
S
p
e
e
c
h
a
n
d
S
ig
n
a
l
Pro
c
e
ss
in
g
(
ICAS
S
P
)
,
Ba
rc
e
lo
n
a
,
S
p
a
i
n
,
2
0
2
0
,
p
p
.
7
2
2
9
-
7
2
3
3
.
[1
2
]
S.
A
.
S
a
ll
o
u
m
,
M
.
A
l
-
E
m
r
a
n
,
A
.
A
.
M
o
n
e
m
,
a
n
d
K.
S
h
a
a
lan
,
"
A
su
rv
e
y
o
f
tex
t
m
in
in
g
in
so
c
ial
m
e
d
ia:
f
a
c
e
b
o
o
k
a
n
d
tw
it
ter
p
e
rsp
e
c
ti
v
e
s,"
Ad
v
a
n
c
e
s
is
S
c
ien
c
e
,
T
e
c
h
n
o
l
o
g
y
a
n
d
En
g
in
e
e
rin
g
S
y
ste
m
J
o
u
rn
a
l
,
v
o
l.
2
,
n
o
.
1
,
p
p
.
1
2
7
-
1
3
3
,
2
0
1
7
.
[1
3
]
S.
A
.
S
a
ll
o
u
m
,
A
.
Q.
A
lHa
m
a
d
,
M
.
A
l
-
E
m
ra
n
,
a
n
d
K.
S
h
a
a
lan
,
“
A
su
rv
e
y
o
f
A
r
a
b
ic
tex
t
m
in
in
g
,
”
In
telli
g
e
n
t
Na
tu
ra
l
L
a
n
g
u
a
g
e
Pr
o
c
e
ss
in
g
:
T
r
e
n
d
s a
n
d
Ap
p
li
c
a
t
io
n
s
,
v
o
l.
7
4
0
,
pp
.
4
1
7
-
431
,
2
0
1
7
.
[1
4
]
S.
A
.
S
a
ll
o
u
m
,
M
.
A
l
-
E
m
ra
n
,
A
.
A
.
M
o
n
e
m
,
a
n
d
K.
S
h
a
a
lan
,
“
Us
in
g
te
x
t
m
in
in
g
tec
h
n
iq
u
e
s
f
o
r
e
x
trac
ti
n
g
in
f
o
rm
a
ti
o
n
f
ro
m
re
se
a
rc
h
a
rti
c
le
s,
”
In
telli
g
e
n
t
N
a
tu
r
a
l
L
a
n
g
u
a
g
e
Pro
c
e
ss
in
g
:
T
re
n
d
s
a
n
d
Ap
p
li
c
a
t
io
n
s
,
v
o
l.
7
40
,
pp
.
3
7
3
-
3
9
7
,
2
0
1
7
.
[1
5
]
S.
A
.
S
a
ll
o
u
m
,
M
.
A
l
-
E
m
ra
n
,
S
.
A
b
d
a
ll
a
h
,
a
n
d
K.
S
h
a
a
lan
,
"
A
n
a
l
y
z
in
g
th
e
A
ra
b
G
u
lf
N
e
w
sp
a
p
e
rs
Us
in
g
T
e
x
t
M
in
i
n
g
T
e
c
h
n
iq
u
e
s,"
Pro
c
e
e
d
in
g
s
o
f
t
h
e
In
ter
n
a
ti
o
n
a
l
Co
n
f
e
re
n
c
e
o
n
Ad
v
a
n
c
e
d
I
n
telli
g
e
n
t
S
y
ste
ms
a
n
d
In
fo
rm
a
t
ics
,
v
o
l.
6
3
9
,
2
0
1
7
,
p
p
.
3
9
6
-
4
0
5
.
[1
6
]
M
.
P
a
li
w
a
l
a
n
d
U.
A
.
Ku
m
a
r,
"
Ne
u
ra
l
n
e
tw
o
rk
s
a
n
d
sta
ti
stica
l
tec
h
n
i
q
u
e
s:
A
re
v
iew
o
f
a
p
p
li
c
a
ti
o
n
s,"
Exp
e
r
t
S
y
ste
ms
wit
h
Ap
p
li
c
a
ti
o
n
s
,
v
o
l
.
3
6
,
n
o
.
1
,
p
p
.
2
-
1
7
,
2
0
0
9
.
[1
7
]
P
.
S
e
ra
i,
A
.
S
ti
f
f
,
a
n
d
E.
F
o
sle
r
-
L
u
ss
ier,
"
En
d
to
E
n
d
S
p
e
e
c
h
Re
c
o
g
n
it
io
n
Err
o
r
P
re
d
icti
o
n
w
it
h
S
e
q
u
e
n
c
e
t
o
S
e
q
u
e
n
c
e
L
e
a
rn
in
g
,
"
ICAS
S
P
2
0
2
0
-
2
0
2
0
IE
EE
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
Aco
u
stics
,
S
p
e
e
c
h
a
n
d
S
ig
n
a
l
Pro
c
e
ss
in
g
(
ICAS
S
P)
,
Ba
rc
e
lo
n
a
,
S
p
a
in
,
2
0
2
0
,
p
p
.
6
3
3
9
-
6
3
4
3
.
[1
8
]
K.
P
a
las
u
n
d
ra
m
,
N.
M
.
S
h
a
re
f
,
K.
A
.
Ka
s
m
iran
,
a
n
d
A
.
Az
m
a
n
,
"
En
h
a
n
c
e
m
e
n
ts
to
th
e
S
e
q
u
e
n
c
e
-
to
-
S
e
q
u
e
n
c
e
-
Ba
se
d
Na
tu
ra
l
A
n
s
w
e
r
G
e
n
e
ra
ti
o
n
M
o
d
e
ls,"
IEE
E
Acc
e
ss
,
v
o
l.
8
,
p
p
.
4
5
7
3
8
-
4
5
7
5
2
,
2
0
2
0
.
[1
9
]
I.
S
a
it
o
,
K.
Nish
id
a
,
K.
Nis
h
id
a
,
a
n
d
J.
T
o
m
it
a
,
"
A
b
stra
c
ti
v
e
S
u
m
m
a
ri
z
a
ti
o
n
w
it
h
Co
m
b
in
a
ti
o
n
o
f
P
re
-
train
e
d
S
e
q
u
e
n
c
e
-
to
-
S
e
q
u
e
n
c
e
a
n
d
S
a
li
e
n
c
y
M
o
d
e
ls,"
a
rXiv P
re
p
r.
a
rXiv2
0
0
3
.
1
3
0
2
8
,
2
0
2
0
.
[2
0
]
M
.
S
u
rd
e
a
n
u
,
"
A
Ge
n
tl
e
In
tro
d
u
c
ti
o
n
t
o
De
e
p
L
e
a
rn
in
g
f
o
r
Na
tu
ra
l
L
a
n
g
u
a
g
e
P
ro
c
e
ss
in
g
,
"
g
o
o
g
le
S
c
h
o
l
a
r
,
p
p
.
1
-
6
3
,
2
0
2
0
.
[2
1
]
D.
Kin
g
,
A
.
D.
S
ims
,
a
n
d
M
.
El
sn
e
r,
"
In
terp
re
ti
n
g
se
q
u
e
n
c
e
-
to
-
se
q
u
e
n
c
e
m
o
d
e
ls
f
o
r
Ru
ss
ian
in
f
lec
ti
o
n
a
l
m
o
rp
h
o
lo
g
y
,
"
Pro
c
e
e
d
in
g
s o
f
th
e
S
o
c
iety
fo
r C
o
mp
u
ta
ti
o
n
i
n
L
i
n
g
u
isti
c
s
,
v
o
l.
3
,
p
p
.
4
0
2
-
4
1
1
,
2
0
2
0
.
[2
2
]
Y.
Ya
n
e
t
a
l.
,
"
P
ro
p
h
e
tNe
t:
P
r
e
d
ictin
g
F
u
t
u
re
N
-
g
ra
m
f
o
r
S
e
q
u
e
n
c
e
-
to
-
S
e
q
u
e
n
c
e
P
re
-
train
i
n
g
,
"
a
rXiv
Pre
p
r.
a
rXiv2
0
0
1
.
0
4
0
6
3
,
2
0
2
0
.
[2
3
]
A
.
S
u
n
,
J.
Wan
g
,
N.
Ch
e
n
g
,
H.
P
e
n
g
,
Z.
Zen
g
,
a
n
d
J.
X
iao
,
"
G
ra
p
h
T
T
S
:
g
r
a
p
h
-
to
-
se
q
u
e
n
c
e
m
o
d
e
ll
in
g
in
n
e
u
ra
l
tex
t
-
to
-
sp
e
e
c
h
,
"
ICAS
S
P
2
0
2
0
-
2
0
2
0
IEE
E
In
ter
n
a
ti
o
n
a
l
C
o
n
fer
e
n
c
e
o
n
Aco
u
stics
,
S
p
e
e
c
h
a
n
d
S
i
g
n
a
l
Pro
c
e
ss
in
g
(
ICAS
S
P)
,
Ba
rc
e
lo
n
a
,
S
p
a
i
n
,
2
0
2
0
,
p
p
.
6
7
1
9
-
6
7
2
3
.
Evaluation Warning : The document was created with Spire.PDF for Python.