I
AE
S
I
nte
rna
t
io
na
l J
o
urna
l o
f
Art
if
icia
l In
t
ellig
ence
(
I
J
-
AI
)
Vo
l.
3
,
No
.
3
,
Sep
tem
b
er
201
4
,
p
p
.
105
~
111
I
SS
N:
2252
-
8938
105
J
o
ur
na
l
ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
jo
u
r
n
a
l.c
o
m/o
n
lin
e/in
d
ex
.
p
h
p
/I
J
A
I
H
y
brid Mo
del
of
Auto
m
a
ted
Ana
p
ho
ra
Reso
lution
K
a
ly
a
ni K
a
m
un
e
,
Av
ina
s
h A
g
ra
wa
l
De
p
a
rtme
n
t
Of
Co
m
p
u
ter S
c
ien
c
e
,
RKN
EC,
Na
g
p
u
r,
I
n
d
ia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Ma
y
2
1
,
2
0
1
4
R
ev
i
s
ed
A
u
g
7
,
2
0
1
4
A
cc
ep
ted
A
u
g
1
9
,
2
0
1
4
A
n
a
p
h
o
ra
re
so
lu
t
io
n
h
a
s
p
ro
v
e
n
to
b
e
a
v
e
ry
d
iff
icu
lt
p
ro
b
lem
o
f
n
a
tu
ra
l
lan
g
u
a
g
e
p
ro
c
e
ss
in
g
,
a
n
d
i
t
is
u
se
f
u
l
in
d
isc
o
u
rse
a
n
a
ly
sis,
lan
g
u
a
g
e
u
n
d
e
rsta
n
d
i
n
g
a
n
d
p
ro
c
e
ss
in
g
,
in
f
o
rm
a
ti
o
n
e
x
a
c
ti
o
n
,
m
a
c
h
in
e
tran
sla
ti
o
n
a
n
d
m
a
n
y
m
o
re
.
T
h
is
p
a
p
e
r
re
p
re
se
n
ts
a
s
y
ste
m
th
a
t
in
ste
a
d
o
f
u
sin
g
a
m
o
n
o
li
th
ic
a
rc
h
it
e
c
tu
re
f
o
r
re
so
lv
in
g
a
n
a
p
h
o
ra
,
u
se
th
e
h
y
b
rid
m
o
d
e
l
w
h
ich
c
o
m
b
in
e
s
th
e
c
o
n
stra
i
n
t
-
b
a
se
d
a
n
d
p
re
f
e
re
n
c
e
s
-
b
a
se
d
a
rc
h
it
e
c
tu
re
s,
e
a
c
h
u
se
s
a
d
if
fe
re
n
t
so
u
rc
e
o
f
k
n
o
w
led
g
e
,
a
n
d
p
ro
v
e
s
e
ff
e
c
ti
v
e
o
n
th
e
o
r
e
ti
c
a
l
a
n
d
co
m
p
u
tatio
n
a
l
b
a
sis.
A
n
a
lg
o
rit
h
m
id
e
n
ti
f
ies
b
o
th
in
ter
-
se
n
te
n
ti
a
l
a
n
d
in
tra
-
se
n
ten
ti
a
l
a
n
tec
e
d
e
n
ts
o
f
“
T
h
ird
p
e
rso
n
p
ro
n
o
u
n
a
n
a
p
h
o
rs”
,
“
P
le
o
n
a
stic
it
”
,
a
n
d
“
L
e
x
i
c
a
l
n
o
u
n
p
h
ra
se
a
n
a
p
h
o
ra
”
.
T
h
e
a
lg
o
rit
h
m
u
se
Ch
a
rn
iak
p
a
rse
r
(p
a
rse
r0
5
A
u
g
1
6
)
a
s
a
n
a
ss
o
c
iate
d
to
o
l,
a
n
d
it
re
lay
s
o
n
th
e
o
u
tp
u
t
g
e
n
e
ra
ted
b
y
it
.
S
a
li
e
n
c
e
m
e
a
su
re
s
d
e
riv
e
d
fro
m
p
a
rse
tre
e
,
in
o
rd
e
r
to
f
in
d
o
u
t
a
c
c
u
ra
te
a
n
tec
e
d
e
n
ts
f
ro
m
th
e
li
st
o
f
p
o
ten
ti
a
l
a
n
tec
e
d
e
n
ts.
W
e
h
a
v
e
tes
ted
t
h
e
sy
st
e
m
e
x
ten
siv
e
l
y
o
n
'
Re
u
ters
Ne
w
sp
a
p
e
r
c
o
rp
u
s'
.
K
ey
w
o
r
d
:
An
a
p
h
o
r
a
An
ap
h
o
r
a
r
eso
lu
tio
n
An
tece
d
en
t
Dis
co
u
r
s
e
Mo
d
elli
n
g
P
r
o
n
o
m
i
n
al
R
e
s
o
lu
t
io
n
Co
p
y
rig
h
t
©
2
0
1
4
In
stit
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
Al
l
rig
h
ts
re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Kal
y
a
n
i K
a
m
u
n
e,
Dep
ar
te
m
en
t o
f
C
o
m
p
u
ter
Sci
en
ce
,
Sh
r
i Ra
m
d
eo
B
ab
a
C
o
lleg
e
o
E
n
g
i
n
ee
r
i
n
g
a
n
d
Ma
n
a
g
m
en
t,
Kato
l Rd
,
Nag
p
u
r
,
Ma
h
ar
as
h
tr
a
,
I
n
d
ia
.
E
m
ail:
k
al
y
a
n
i
k
a
m
u
n
e2
4
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
I
n
ter
p
r
etin
g
a
n
ap
h
o
r
ic
e
x
p
r
ess
io
n
s
is
o
n
e
o
f
t
h
e
m
o
s
t
f
u
n
d
a
m
e
n
tal
asp
ec
t
s
o
f
lan
g
u
a
g
e
in
ter
p
r
etatio
n
.
T
h
e
s
tu
d
y
o
f
an
ap
h
o
r
a
an
d
an
ap
h
o
r
a
h
as
b
r
o
u
g
h
t
ab
o
u
t
m
an
y
f
u
n
d
a
m
e
n
t
al
d
ev
elo
p
m
e
n
t
s
in
th
eo
r
etica
l
lin
g
u
is
t
ics
an
d
co
m
p
u
tat
io
n
al
li
n
g
u
i
s
tics
a
n
d
h
as
i
m
p
o
r
t
an
t
p
r
ac
tical
ap
p
licatio
n
s
i
n
w
o
r
k
o
n
in
f
o
r
m
atio
n
ex
tr
ac
tio
n
,
s
u
m
m
ar
izatio
n
,
a
n
d
e
n
tit
y
d
is
a
m
b
ig
u
atio
n
.
An
ap
h
o
r
a
r
eso
lu
t
i
o
n
is
a
co
m
p
licated
p
r
o
b
lem
in
Nat
u
r
al
L
a
n
g
u
a
g
e
P
r
o
ce
s
s
in
g
an
d
h
as
attr
ac
t
ed
th
e
atten
tio
n
o
f
m
a
n
y
r
es
ea
r
ch
er
s
.
An
ap
h
o
r
a
d
escr
ib
es
th
e
lan
g
u
ag
e
p
h
e
n
o
m
e
n
o
n
o
f
r
ef
er
r
in
g
to
a
p
r
ev
io
u
s
l
y
m
e
n
tio
n
ed
en
tit
y
(
a
ls
o
ca
lled
o
b
j
ec
t
o
r
ev
en
t
)
;
a
n
ap
h
o
r
a
r
eso
lu
tio
n
i
s
th
e
p
r
o
ce
s
s
o
f
f
in
d
i
n
g
t
h
at
p
r
ev
io
u
s
ite
m
.
C
o
n
s
id
er
th
e
f
o
llo
w
i
n
g
clar
if
y
i
n
g
ex
a
m
p
le
f
r
o
m
a
B
r
itis
h
W
o
r
ld
W
ar
I
I
an
ti
-
r
aid
leaf
let:
“
I
f
an
i
n
ce
n
d
iar
y
b
o
m
b
d
r
o
p
s
n
ex
t
to
y
o
u
,
d
o
n
‟
t
lo
s
e
y
o
u
r
h
ea
d
.
P
u
t
it
i
n
an
y
b
u
c
k
et
a
n
d
co
v
er
it
w
it
h
s
o
m
e
s
a
n
d
.
”
I
f
th
i
s
r
aised
e
y
eb
r
o
w
s
-
d
o
n
‟
t
w
o
r
r
y
-
it
is
m
ea
n
t
to
.
I
n
d
ee
d
“
it”
co
u
ld
s
tan
d
f
o
r
(
o
r
r
efer
to
)
eith
er
o
f
th
e
t
w
o
o
b
j
ec
ts
m
en
t
io
n
ed
b
ef
o
r
e
it,
“
B
o
m
b
”
an
d
“h
ea
d
”.
T
h
e
au
th
o
r
s
m
ea
n
t
th
e
f
o
r
m
er
,
b
u
t
th
e
r
u
les
o
f
lan
g
u
ag
e
h
av
e
a
te
n
d
en
c
y
to
b
ias
r
ea
d
er
s
to
p
ick
in
g
th
e
latter
.
B
u
t
t
h
en
“h
ea
d
‟
s
”
ar
e
n
o
t
t
h
e
u
s
u
al
th
i
n
g
s
o
n
e
p
u
ts
in
b
u
ck
et
s
an
d
co
v
er
s
w
it
h
s
an
d
.
W
h
at
a
n
ap
h
o
r
a
r
eso
lu
tio
n
,
wh
en
d
o
n
e
co
r
r
ec
tl
y
,
e
n
ab
les
u
s
an
d
s
y
s
te
m
s
to
d
o
,
is
to
m
er
g
e
t
h
e
p
r
ev
io
u
s
i
n
f
o
r
m
atio
n
ab
o
u
t
a
n
en
tit
y
w
it
h
th
e
n
e
w
i
n
f
o
r
m
atio
n
w
e
en
co
u
n
ter
.
So
th
in
k
o
f
an
ap
h
o
r
a
as
th
e
i
n
tr
icate
b
alan
ce
b
et
w
ee
n
co
n
ci
s
en
e
s
s
o
f
co
m
m
u
n
icatio
n
a
n
d
th
e
ab
ilit
y
o
f
h
u
m
a
n
s
t
o
un
d
er
s
ta
n
d
ea
ch
o
th
er
[
1
]
.
T
h
e
r
em
ain
in
g
s
ec
tio
n
s
o
f
t
h
is
p
ap
er
ar
e
o
r
g
an
ized
as
f
o
llo
w
s
.
I
n
s
ec
tio
n
2
,
w
e
r
ep
r
esen
t
th
e
L
iter
at
u
r
e
S
u
r
v
e
y
o
f
An
ap
h
o
r
a
R
eso
l
u
tio
n
.
I
n
s
ec
tio
n
3
,
w
e
r
ep
r
esen
t
i
n
d
etails
h
o
w
ea
ch
s
tep
o
f
“Au
to
m
ated
An
ap
h
o
r
a
R
eso
l
u
tio
n
”
alg
o
r
i
th
m
is
i
m
p
l
e
m
en
ted
.
Sectio
n
4
,
ex
p
lain
s
th
e
ar
c
h
itect
u
r
e
o
f
“Au
to
m
a
ted
An
ap
h
o
r
a
R
eso
l
u
tio
n
”.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
3
,
No
.
3
,
Sep
tem
b
er
20
1
4
:
1
0
5
–
1
1
1
106
2.
L
I
T
E
R
AT
U
RE
SU
RVE
Y
I
n
t
h
e
p
r
o
ce
s
s
o
f
a
n
ap
h
o
r
a
r
e
s
o
lu
tio
n
,
a
n
tece
d
en
t
s
ca
n
b
e
n
o
u
n
o
r
v
er
b
p
h
r
ases
,
an
y
c
l
au
s
e
s
,
an
y
s
en
te
n
ce
o
r
e
v
e
n
p
ar
ag
r
ap
h
s
/
d
is
co
u
r
s
e
s
eg
m
e
n
ts
as
a
n
tece
d
e
n
ts
.
Fin
d
i
n
g
a
n
tece
d
en
t
s
a
s
a
n
o
u
n
p
h
r
a
s
e
i
s
co
m
p
ar
ati
v
el
y
a
n
ea
s
y
tas
k
t
h
an
t
h
at
o
f
f
i
n
d
in
g
r
ests
as
a
n
an
tece
d
e
n
t.
Gen
er
all
y
,
all
n
o
u
n
p
h
r
ases
(
NP
s
)
p
r
ec
ed
in
g
an
a
n
ap
h
o
r
ar
e
in
iti
all
y
co
n
s
id
er
ed
as
p
o
ten
tial
c
an
d
id
ates
f
o
r
an
tece
d
en
t
s
.
Sea
r
ch
li
m
it
h
as
t
o
b
e
p
r
ed
ef
in
ed
.
An
“
id
ea
l”
an
ap
h
o
r
a
r
eso
lu
tio
n
s
y
s
te
m
h
as
t
h
e
s
ea
r
ch
li
m
i
t
o
f
1
7
s
en
te
n
ce
s
a
w
a
y
f
r
o
m
th
e
s
en
te
n
ce
i
n
w
h
ic
h
a
n
ap
h
o
r
is
p
r
esen
t
[
4
]
.
A
ll
t
h
e
p
o
ten
tial
a
n
tece
d
en
ts
w
it
h
i
n
t
h
e
s
ea
r
ch
li
m
it
p
r
ec
ed
in
g
a
n
ap
h
o
r
a
ar
e
f
in
d
i
n
g
o
u
t,
an
d
v
ar
io
u
s
an
ap
h
o
r
a
r
eso
lu
t
io
n
f
ac
to
r
s
ar
e
u
s
ed
to
f
i
n
d
t
h
e
co
r
r
ec
t a
n
te
ce
d
en
t f
o
r
t
h
e
p
ar
tic
u
lar
an
ap
h
o
r
a.
Var
io
u
s
f
ac
to
r
s
u
s
ed
ca
n
b
e
"
eli
m
i
n
ato
r
y
"
i.e
.
d
o
esn
'
t
co
u
n
t
ce
r
tai
n
n
o
u
n
p
h
r
ases
f
r
o
m
t
h
e
s
et
o
f
p
o
ten
t
i
al
an
tece
d
en
t
s
(
s
u
c
h
as
g
e
n
d
er
,
n
u
m
b
er
,
p
eo
p
le
co
n
s
tr
ai
n
ts
)
o
r
"
p
r
ef
er
en
tial"
,
ass
ig
n
i
n
g
m
o
r
e
p
r
ef
er
e
n
ce
to
s
o
m
e
p
o
ten
tial
an
tece
d
en
t
s
an
d
les
s
to
o
th
er
s
(
s
u
c
h
as salie
n
ce
)
.
[
5
]
S
y
s
te
m
m
u
s
t
d
ef
i
n
e
t
h
e
te
x
t
s
eg
m
e
n
ts
i
n
w
h
ic
h
t
h
e
an
tec
ed
en
t
ca
n
b
e
f
o
u
n
d
,
w
h
ic
h
i
s
ca
lled
as
s
ea
r
ch
li
m
it
o
r
an
ap
h
o
r
ic
ac
ce
s
s
ib
ilit
y
s
p
ac
e.
T
h
is
s
tep
is
v
e
r
y
i
m
p
o
r
tan
t,
f
o
r
f
u
r
th
er
p
r
o
ce
s
s
i
n
g
.
Kee
p
i
n
g
th
e
s
ea
r
ch
li
m
it
to
o
n
ar
r
o
w
r
es
u
lt
s
in
th
e
ex
c
lu
s
io
n
o
f
v
al
id
an
t
ec
ed
en
ts
an
d
k
ee
p
in
g
th
e
s
ea
r
ch
li
m
i
t
to
o
b
r
o
ad
r
esu
lt
s
in
th
e
lar
g
e
ca
n
d
id
ate
lis
ts
w
h
ich
ar
e
u
lti
m
atel
y
lead
s
to
er
r
o
n
eo
u
s
r
esu
lt
s
.
On
ce
th
e
lis
t
o
f
all
p
o
s
s
ib
le
ca
n
d
id
ates
i
s
f
o
u
n
d
,
s
ev
er
al
c
o
n
s
tr
ain
ts
ca
n
b
e
ap
p
lied
in
o
r
d
er
to
r
em
o
v
e
in
co
m
p
atib
le
an
tece
d
en
t
s
.
I
t
m
a
y
b
e
p
o
s
s
ib
le
th
at
af
ter
ap
p
l
y
i
n
g
co
n
s
tr
ain
ts
,
s
o
m
e
a
n
ap
h
o
r
a
ca
n
s
till
h
a
v
e
m
o
r
e
th
a
n
o
n
e
p
o
s
s
ib
le
an
tece
d
e
n
t
,
in
s
u
ch
ca
s
e
w
e
ca
n
g
et
ac
cu
r
a
te
an
tece
d
en
t
b
y
u
s
in
g
p
r
ef
er
e
n
ce
s
.
T
w
o
t
y
p
es
o
f
ar
c
h
itect
u
r
e
ar
e
p
r
esen
t
co
n
s
tr
ai
n
t
-
b
as
ed
ar
ch
itect
u
r
es
a
n
d
p
r
ef
er
en
ce
s
-
b
ased
ar
ch
itect
u
r
es,
b
ased
o
n
t
h
e
f
ac
to
r
s
(
co
n
s
tr
ai
n
t
-
b
ased
o
r
p
r
ef
e
r
en
ce
s
-
b
ased
)
w
h
ic
h
w
e
ar
e
u
s
in
g
i
n
th
e
p
r
o
ce
s
s
o
f
an
ap
h
o
r
a
r
eso
lu
tio
n
.
I
n
s
tea
d
o
f
u
s
in
g
a
m
o
n
o
lit
h
ic
ar
ch
it
ec
tu
r
e
f
o
r
r
eso
lv
in
g
an
ap
h
o
r
a,
in
th
e
“Au
to
m
ated
An
ap
h
o
r
a
R
e
s
o
lu
tio
n
”
s
y
s
te
m
w
e
u
s
e
t
h
e
co
m
b
i
n
atio
n
o
f
co
n
s
tr
ai
n
t
-
b
a
s
ed
an
d
p
r
ef
er
en
ce
s
-
b
ased
ar
ch
itect
u
r
es;
ea
c
h
u
s
es
a
d
if
f
er
e
n
t
s
o
u
r
ce
o
f
k
n
o
w
le
d
g
e,
an
d
p
r
o
v
es
e
f
f
ec
ti
v
e
o
n
th
eo
r
etica
l
a
n
d
co
m
p
u
tatio
n
al
b
asi
s
.
He
n
ce
th
e
s
y
s
te
m
m
ai
n
l
y
w
o
r
k
s
o
n
3
s
tep
s
g
iv
e
n
a
s
:
(
1
)
d
ef
i
n
in
g
a
n
a
n
ap
h
o
r
i
c
ac
ce
s
s
ib
ilit
y
s
p
ac
e
o
r
Sear
ch
L
i
m
it,
(
2
)
ap
p
ly
co
n
s
tr
ain
ts
,
a
n
d
th
e
n
(
3
)
ap
p
ly
p
r
ef
er
en
ce
s
.
3.
‘
H
YB
RID M
O
DE
L
O
F
AU
T
O
M
AT
E
D
ANAP
H
O
RA
R
E
SO
L
UT
I
O
N’
I
N
DE
T
A
I
L
I
f
w
e
r
eso
l
v
e
an
ap
h
o
r
a
co
r
r
e
ctl
y
,
it
i
t
s
i
g
n
i
f
ican
t
l
y
in
cr
ea
s
es
th
e
p
er
f
o
r
m
an
ce
o
f
t
h
e
d
o
w
n
s
tr
ea
m
Natu
r
al
L
a
n
g
u
a
g
e
P
r
o
ce
s
s
in
g
ap
p
licatio
n
s
.
He
n
ce
,
to
ad
d
r
ess
t
h
is
p
r
o
b
lem
o
f
r
eso
l
v
i
n
g
an
ap
h
o
r
a
co
r
r
e
ctl
y
,
w
e
h
a
v
e
i
m
p
le
m
e
n
ted
a
J
av
a
-
b
ased
s
y
s
te
m
w
h
ic
h
u
s
e
s
a
h
y
b
r
id
ap
p
r
o
ac
h
f
o
r
r
eso
lv
i
n
g
an
ap
h
o
r
a.
A
s
y
s
te
m
u
s
ed
f
o
r
id
en
ti
f
y
in
g
b
o
th
i
n
te
r
-
s
e
n
ten
tial
an
d
i
n
tr
a
-
s
en
te
n
ti
al
an
tece
d
en
t
s
o
f
th
ir
d
p
er
s
o
n
p
r
o
n
o
u
n
s
in
t
h
eir
n
o
m
i
n
ati
v
e,
ac
cu
s
a
tiv
e
o
r
p
o
s
s
es
s
i
v
e
c
ase
an
d
p
leo
n
asti
c
an
ap
h
o
r
a.
I
n
o
u
r
s
y
s
te
m
,
w
e
h
av
e
co
n
s
id
er
th
e
s
ea
r
ch
li
m
it/
An
ap
h
o
r
ic
A
cc
e
s
s
ib
ilit
y
Sp
ac
e
o
f
3
s
en
ten
ce
s
,
h
en
ce
f
o
r
an
y
a
n
ap
h
o
r
a,
s
y
s
t
e
m
w
ill
f
i
n
d
all
th
e
p
o
ten
tial
an
tece
d
en
ts
f
r
o
m
t
h
e
3
s
en
te
n
ce
s
p
r
ec
ed
in
g
t
h
e
s
e
n
ten
ce
i
n
w
h
ic
h
an
ap
h
o
r
a
is
p
r
esen
t,
i
n
cl
u
d
in
g
t
h
e
s
en
te
n
ce
i
n
w
h
ich
an
ap
h
o
r
a
is
p
r
esen
t.
S
y
s
te
m
u
s
e
s
C
h
ar
n
i
ak
p
ar
s
er
(
p
ar
s
er
0
5
A
u
g
1
6
)
as
an
as
s
o
ciate
d
to
o
l,
an
d
it r
ela
y
s
o
n
t
h
e
o
u
tp
u
t g
e
n
er
ated
b
y
it.
I
n
s
tead
o
f
u
s
i
n
g
a
s
in
g
le
m
o
n
o
lith
ic
ar
ch
itect
u
r
e
s
y
s
te
m
u
s
es
th
e
h
y
b
r
i
d
ap
p
r
o
ac
h
w
h
ic
h
co
m
b
i
n
es
co
n
s
tr
ain
t
-
b
ased
an
d
p
r
e
f
er
en
ce
s
-
b
ased
ar
c
h
itect
u
r
es
S
y
s
te
m
r
ea
d
te
x
t
f
ile
a
s
a
n
i
n
p
u
t
an
d
g
iv
e
it
to
t
h
e
s
en
te
n
ce
s
p
lit
ter
.
Sen
te
n
ce
Sp
litt
er
u
s
ed
b
y
th
e
s
y
s
te
m
as
a
n
as
s
o
ciate
d
to
o
l
s
p
lit
th
e
s
en
ten
ce
s
a
n
d
p
u
t
t
h
e
tag
s
li
k
e
<S>a
n
d
</S>
b
ef
o
r
e
a
n
d
a
f
ter
ea
ch
s
e
n
te
n
ce
r
esp
ec
t
iv
el
y
,
ac
co
r
d
in
g
to
t
h
e
r
eq
u
ir
em
en
t
o
f
t
h
e
p
ar
s
er
.
T
h
e
o
u
tp
u
t
g
en
er
ated
b
y
t
h
e
s
en
te
n
ce
s
p
litt
er
is
th
e
n
g
iv
e
n
to
th
e
p
ar
s
er
0
5
au
g
1
6
f
o
r
f
u
r
th
er
p
r
o
ce
s
s
in
g
.
A
S
y
n
tactic
r
ep
r
esen
tatio
n
i
s
g
e
n
er
ated
b
y
th
e
p
ar
s
er
c
alled
as
p
ar
s
e
tr
ee
.
T
h
is
s
y
n
tact
ic
r
ep
r
esen
tatio
n
cr
ea
ted
b
y
th
e
p
ar
s
er
p
la
y
s
a
n
v
ital
r
o
le
f
o
r
th
e
f
u
r
th
er
p
r
o
ce
s
s
i
n
g
.
Nex
t
s
tep
is
to
cr
ea
te
th
e
li
s
t
o
f
a
n
ap
h
o
r
a
a
n
d
an
tece
d
en
t.
No
w
,
ea
ch
an
ap
h
o
r
a
f
o
r
m
a
p
air
w
it
h
all
th
e
p
o
ten
tial
an
tece
d
en
ts
co
m
e
s
in
its
An
ap
h
o
r
ic
A
cc
e
s
s
ib
ilit
y
Sp
ac
e
i.e
.
W
ith
i
n
3
s
en
ten
ce
f
r
o
m
th
e
s
en
te
n
c
e
in
w
h
ic
h
an
ap
h
o
r
a
is
p
r
esen
t
.
Fo
r
ea
ch
p
r
o
n
o
u
n
s
an
d
n
o
u
n
p
h
r
a
s
e
in
ea
c
h
p
air
f
i
n
d
ag
r
ee
m
en
t
f
ea
t
u
r
es
(
N
u
m
b
er
,
P
eo
p
le
an
d
Gen
d
er
)
.
E
ac
h
cr
ea
ted
p
air
o
f
an
ap
h
o
r
a
an
d
n
o
u
n
p
h
r
ases
i
s
t
h
en
c
h
ec
k
s
f
o
r
t
h
e
a
g
r
ee
m
en
t
f
ea
tu
r
e
(
P
er
s
o
n
,
Ge
n
d
e
r
,
an
d
Nu
m
b
er
)
in
ag
r
ee
m
e
n
t
f
i
lter
.
I
f
t
h
e
g
i
v
en
p
air
f
u
l
f
ill
s
t
h
e
a
g
r
ee
m
en
t
f
ea
th
er
s
th
e
n
allo
w
p
as
s
i
n
g
f
o
r
th
e
f
u
r
t
h
er
p
r
o
ce
s
s
i
n
g
,
else p
air
is
d
is
ca
r
d
ed
b
y
th
e
s
y
s
te
m
.
T
h
e
r
es
u
lted
p
air
s
ar
e
th
en
f
ilter
ed
th
r
o
u
g
h
f
u
r
t
h
er
f
il
t
er
in
g
p
r
o
ce
s
s
to
g
et
th
e
co
r
r
ec
t a
n
tece
d
en
t
f
o
r
th
e
an
ap
h
o
r
a.
A
ll
t
h
e
i
n
f
o
r
m
atio
n
r
eq
u
ir
ed
b
y
th
e
s
y
s
te
m
is
n
o
t
u
s
ed
g
e
n
er
ated
b
y
t
h
e
p
ar
s
er
,
s
y
s
te
m
h
av
e
to
ex
tr
ac
t
ce
r
tain
r
eq
u
ir
ed
in
f
o
r
m
atio
n
f
r
o
m
th
e
o
u
tp
u
t
g
e
n
er
ated
b
y
th
e
p
ar
s
er
.
Nex
t
s
tep
is
to
d
er
iv
e
r
eq
u
ir
e
d
s
alien
ce
m
ea
s
u
r
es
f
r
o
m
p
ar
s
e
tr
ee
,
w
h
ic
h
is
u
s
ed
f
o
r
th
e
f
u
r
t
h
er
p
r
o
ce
s
s
in
g
.
A
p
p
l
y
“P
leo
n
asti
c
P
r
o
n
o
u
n
Fil
ter
”
to
f
i
n
d
p
leo
n
asti
c
p
r
o
n
o
u
n
s
(
I
t
w
ill
ta
k
e
“L
i
s
t
o
f
An
ap
h
o
r
s
”
as
a
n
i
n
p
u
t)
.
A
p
p
l
y
„
P
er
s
o
n
al
P
r
o
n
o
u
n
Fil
ter
‟
f
o
r
r
eso
lv
i
n
g
„
T
h
ir
d
p
er
s
o
n
P
r
o
n
o
u
n
s
‟
,
w
h
ic
h
w
ill
tak
e
li
s
t
o
f
n
o
u
n
p
h
r
ase
s
an
d
p
r
o
n
o
u
n
as
i
n
p
u
t.
P
o
ten
tial
C
an
d
id
ates
f
o
r
an
tec
ed
en
t
ar
e
r
an
k
ed
b
y
th
e
ir
“
s
alien
ce
w
ei
g
h
t
s
”
an
d
th
e
to
p
o
n
e
is
p
r
o
p
o
s
ed
as
th
e
ac
cu
r
ate
a
n
tece
d
en
t.
Ge
n
er
ate
o
u
tp
u
t a
s
“Co
-
r
e
f
er
en
tial P
air
”.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
Hyb
r
id
Mo
d
el
o
f A
u
to
ma
te
d
A
n
a
p
h
o
r
a
R
eso
lu
tio
n
(
K
a
lya
n
i
K
a
mu
n
e
)
107
3
.
1
.
Ag
re
em
ent
F
ilte
r
W
e
g
et
n
u
m
b
er
o
f
p
air
s
o
f
n
o
u
n
a
n
d
p
r
o
n
o
u
n
f
r
o
m
t
h
e
li
s
t
o
f
n
o
u
n
p
h
r
a
s
e
a
n
d
p
r
o
n
o
u
n
p
h
r
ase
,
w
h
ic
h
ar
e
th
e
n
allo
w
ed
to
f
i
l
ter
th
r
o
u
g
h
„
Ag
r
ee
m
e
n
t
f
ilter
‟
.
T
h
e
ag
r
ee
m
e
n
t
f
ea
t
u
r
es
‟
co
m
p
a
tib
ilit
y
o
f
ea
c
h
an
d
ev
er
y
p
air
o
f
p
r
o
n
o
u
n
an
d
a
n
o
u
n
p
h
r
ase
is
tes
ted
b
y
a
A
g
r
ee
men
t
filt
er
.
I
t
s
tates
n
o
u
n
a
n
d
p
r
o
n
o
u
n
p
air
as
n
o
n
-
m
a
tch
i
n
g
i
n
th
eir
ag
r
e
e
m
en
t
f
ea
t
u
r
es
o
n
l
y
i
f
at
least
o
n
e
a
g
r
ee
m
e
n
t
f
ea
tu
r
e
d
o
es
n
‟
t
a
g
r
ee
s
,
o
r
el
s
e
i
t
s
tates
t
h
e
m
as
m
atch
in
g
.
T
h
e
v
alu
e
“
u
n
k
n
o
w
n
”
is
r
eg
ar
d
e
d
to
ag
r
ee
w
it
h
an
y
v
al
u
e
o
f
th
e
f
ea
t
u
r
e.
T
h
e
co
n
s
tr
ain
t
s
y
s
te
m
co
n
s
is
t
s
o
f
co
n
d
itio
n
s
t
h
at
m
u
s
t
b
e
m
e
t,
an
d
ca
n
d
id
ates
t
h
at
d
o
n
o
t
f
u
l
f
ill
t
h
ese
co
n
d
itio
n
s
w
il
l n
o
t b
e
co
n
s
id
er
ed
p
o
s
s
ib
le
an
tece
d
en
t
s
f
o
r
th
e
a
n
ap
h
o
r
.
C
o
n
s
id
er
b
ello
w
ex
a
m
p
le:
“Ci
n
c
in
n
ati
B
ell
I
n
c
s
aid
it h
a
s
s
tar
ted
its
p
r
ev
io
u
s
l
y
-
f
ea
t
h
er
an
n
o
u
n
ce
d
1
5
.
7
5
d
lr
p
er
s
h
ar
e
ten
d
er
o
f
f
er
f
o
r
all
s
h
ar
es o
f
A
u
x
to
n
C
o
m
p
u
ter
E
n
ter
p
r
is
es I
n
c.
”
I
n
th
e
ab
o
v
e
ex
a
m
p
le,
'
C
i
n
ci
n
n
ati
B
ell
I
n
c'
a
n
d
'
it
'
s
at
is
f
ies
th
e
co
n
d
itio
n
s
o
f
ag
r
ee
m
e
n
t
f
ea
tu
r
e
an
d
p
ass
f
o
r
th
e
f
at
h
er
f
ilter
i
n
g
.
1.
R
eq
u
ir
ed
I
n
f
o
r
m
atio
n
Ob
tai
n
e
d
b
y
I
n
s
p
ec
tin
g
th
e
P
ar
s
e
T
r
ee
Stru
ct
u
r
e
A
ll
t
h
e
i
n
f
o
r
m
atio
n
r
eq
u
ir
e
d
b
y
th
e
s
y
s
te
m
is
n
o
t
g
i
v
en
b
y
th
e
p
ar
s
er
;
“Au
to
m
at
ed
A
n
ap
h
o
r
a
R
eso
l
u
tio
n
”
s
y
s
te
m
r
ec
o
v
er
s
th
e
m
b
y
u
s
in
g
s
tr
u
ct
u
r
e
in
f
o
r
m
atio
n
o
f
th
e
v
er
b
/n
o
u
n
p
h
r
as
es.
[
2
]
2.
P
leo
n
asti
c
P
r
o
n
o
u
n
s
F
ilter
S
y
s
te
m
al
s
o
id
en
tifie
s
p
leo
n
a
s
tic
p
r
o
n
o
u
n
s
.
I
n
ca
s
e
o
f
P
leo
n
asti
c
An
ap
h
o
r
a,
th
e
p
r
o
n
o
u
n
d
o
esn
'
t
h
a
v
e
an
y
r
e
f
er
en
t.
S
y
s
te
m
u
s
es
t
h
e
lis
t
o
f
m
o
d
al
ad
j
ec
tiv
e
an
d
a
co
g
n
iti
v
e
v
er
b
in
o
r
d
er
to
f
i
n
d
p
leo
n
asti
c
p
r
o
n
o
u
n
s
.
3.
P
er
s
o
n
al
P
r
o
n
o
u
n
Fil
ter
an
d
L
ex
ical
Fil
ter
“Au
to
m
a
ted
An
ap
h
o
r
a
R
eso
l
u
tio
n
”
s
y
s
te
m
i
s
u
s
ed
to
f
in
d
o
u
t
th
e
“
T
h
ir
d
p
er
s
o
n
p
r
o
n
o
u
n
s
”
in
t
h
eir
n
o
m
i
n
ati
v
e,
ac
cu
s
ati
v
e
o
r
p
o
s
s
ess
i
v
e
ca
s
e
a
n
d
“L
e
x
ical
a
n
a
p
h
o
r
a”
.
I
n
f
o
r
m
atio
n
Ob
tain
ed
b
y
I
n
s
p
ec
ti
n
g
th
e
P
ar
s
e
T
r
ee
Stru
ctu
r
e
p
la
y
s
a
v
ital
r
o
le
f
o
r
id
en
tify
i
n
g
th
e
“
T
h
ir
d
p
er
s
o
n
p
r
o
n
o
u
n
a
n
ap
h
o
r
a”
an
d
“L
e
x
ical
An
ap
h
o
r
a”
.
I
t
is
th
e
m
o
s
t
w
id
e
s
p
r
ea
d
t
y
p
e
o
f
an
ap
h
o
r
a.
P
r
o
n
o
m
i
n
al
a
n
ap
h
o
r
a
o
cc
u
r
s
at
t
h
e
lev
el
o
f
p
er
s
o
n
a
l
p
r
o
n
o
u
n
s
,
p
o
s
s
ess
iv
e
p
r
o
n
o
u
n
s
,
r
e
f
lex
iv
e
p
r
o
n
o
u
n
s
.
T
h
e
s
e
t
o
f
an
ap
h
o
r
ic
p
r
o
n
o
u
n
s
co
n
s
is
t o
f
all
t
h
ir
d
p
er
s
o
n
p
er
s
o
n
al
(
h
e,
h
i
m
,
s
h
e,
h
er
,
it,
th
e
y
,
th
e
m
)
,
p
o
s
s
e
s
s
i
v
e
(
h
is
,
h
er
,
h
er
s
,
its
,
th
e
ir
,
th
eir
s
)
,
r
ef
lex
i
v
e
(
h
i
m
s
el
f
,
h
e
r
s
elf
,
it
s
el
f
,
th
e
m
s
elv
e
s
)
p
r
o
n
o
u
n
s
in
b
o
t
h
s
i
n
g
u
lar
an
d
p
lu
r
al.
W
h
er
ea
s
,
f
ir
s
t a
n
d
s
ec
o
n
d
p
er
s
o
n
p
r
o
n
o
u
n
ca
n
o
f
te
n
b
e
n
o
n
-
a
n
ap
h
o
r
ic
.
S
y
s
te
m
i
s
u
s
ed
to
f
i
n
d
o
u
t
th
e
“
T
h
ir
d
p
er
s
o
n
p
r
o
n
o
u
n
s
”
i
n
t
h
eir
n
o
m
i
n
ati
v
e,
ac
cu
s
a
tiv
e
o
r
p
o
s
s
ess
i
v
e
ca
s
e.
I
n
f
o
r
m
at
io
n
o
b
tain
ed
f
r
o
m
th
e
P
ar
s
e
T
r
ee
Stru
ctu
r
e
p
la
y
s
a
v
ita
l
r
o
le
f
o
r
id
en
tify
in
g
th
e
“
T
h
ir
d
p
er
s
o
n
p
r
o
n
o
u
n
an
ap
h
o
r
a”
.
Fo
r
E
x
a
m
p
le
:
“
S
u
m
ito
m
o
P
r
esid
en
t K
o
h
Ko
m
ats
u
to
ld
R
e
u
ter
s
h
e
i
s
co
n
f
i
d
en
t h
i
s
b
an
k
ca
n
q
u
ic
k
l
y
r
eg
a
in
its
p
o
s
itio
n
.
”
I
n
th
e
ab
o
v
e
ex
a
m
p
le,
'h
e
'
an
d
'h
i
s
'
b
o
th
r
e
f
er
s
to
t
h
e
'
S
u
m
ito
m
o
P
r
esid
en
t
Ko
h
Ko
m
at
s
u
'
w
er
e
co
r
r
ec
tl
y
id
en
ti
f
ied
b
y
th
e
p
er
s
o
n
al
p
r
o
n
o
u
n
f
ilter
.
3.
2
.
Sa
lience
w
eig
hts
Af
ter
ap
p
l
y
i
n
g
all
ab
o
v
e
m
e
n
ti
o
n
ed
f
ilter
s
,
t
h
er
e
is
a
p
o
s
s
ib
ili
t
y
th
a
t f
o
r
a
p
ar
ticu
lar
an
ap
h
o
r
a
m
o
r
e
th
an
o
n
e
p
o
ten
tial a
n
tece
d
en
t
ar
e
p
r
esen
t.
Hen
ce
,
in
o
r
d
er
to
g
et
f
i
n
al
a
n
tece
d
en
t
s
alie
n
ce
w
ei
g
h
t is
u
s
ed
.
[
2
]
T
h
e
m
ai
n
te
x
t
f
o
r
m
at
co
n
s
i
s
t
s
o
f
a
f
lat
le
f
t
-
r
ig
h
t
co
lu
m
n
s
o
n
A
4
p
ap
er
(
q
u
ar
to
)
.
T
h
e
m
ar
g
in
te
x
t
f
r
o
m
t
h
e
lef
t
an
d
to
p
ar
e
2
.
5
cm
,
r
i
g
h
t
an
d
b
o
tto
m
ar
e
2
c
m
.
T
h
e
m
an
u
s
cr
ip
t
is
w
r
i
tten
i
n
Mic
r
o
s
o
f
t
W
o
r
d
,
s
in
g
le
s
p
ac
e,
T
im
e
Ne
w
R
o
m
an
1
0
p
t
an
d
m
ax
i
m
u
m
1
2
p
ag
es,
w
h
ic
h
ca
n
b
e
d
o
w
n
lo
ad
ed
at
th
e
w
eb
s
ite:
h
ttp
://
www
.
iaesj
o
u
r
n
al.
co
m
/o
n
li
n
e/i
n
d
ex
.
p
h
p
/I
J
E
C
E
A
titl
e
o
f
ar
tic
le
s
h
o
u
ld
b
e
t
h
e
f
e
w
es
t
p
o
s
s
ib
le
w
o
r
d
s
th
a
t
ac
cu
r
atel
y
d
escr
ib
e
t
h
e
co
n
t
en
t
o
f
t
h
e
p
ap
er
.
Om
it
al
l
w
a
s
te
w
o
r
d
s
s
u
c
h
as
"
A
s
tu
d
y
o
f
.
.
.
",
"
I
n
ve
s
tig
a
tio
n
s
o
f
.
.
.
",
"
I
mp
l
eme
n
ta
tio
n
o
f
.
.
.
”
,
"
Ob
s
erva
tio
n
s
o
n
.
.
.
"
,
"
E
ffect
o
f..
.
.
.
",
“
A
n
a
lysi
s
o
f
…
”,
“De
s
ig
n
o
f
…”
etc.
I
n
d
ex
i
n
g
a
n
d
a
b
s
tr
ac
tin
g
s
er
v
ices
d
ep
en
d
o
n
t
h
e
ac
c
u
r
ac
y
o
f
t
h
e
titl
e,
e
x
tr
ac
ti
n
g
f
r
o
m
i
t
k
e
y
w
o
r
d
s
u
s
e
f
u
l
i
n
cr
o
s
s
-
r
e
f
er
en
cin
g
a
n
d
co
m
p
u
ter
s
ea
r
ch
i
n
g
.
An
i
m
p
r
o
p
er
l
y
titl
e
d
p
ap
er
m
a
y
n
e
v
er
r
ea
ch
t
h
e
a
u
d
ien
ce
f
o
r
w
h
ic
h
it
w
as i
n
te
n
d
ed
,
s
o
b
e
s
p
ec
if
ic.
T
h
e
I
n
tr
o
d
u
ctio
n
s
h
o
u
ld
p
r
o
v
id
e
a
clea
r
b
ac
k
g
r
o
u
n
d
,
a
clea
r
s
tate
m
e
n
t
o
f
th
e
p
r
o
b
lem
,
t
h
e
r
elev
an
t
liter
atu
r
e
o
n
t
h
e
s
u
b
j
ec
t,
th
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
o
r
s
o
lu
ti
o
n
,
an
d
th
e
n
e
w
v
alu
e
o
f
r
esear
ch
w
h
ic
h
it
i
s
in
n
o
v
atio
n
.
I
t
s
h
o
u
ld
b
e
u
n
d
er
s
tan
d
ab
le
to
co
lleag
u
e
s
f
r
o
m
a
b
r
o
ad
r
an
g
e
o
f
s
cien
ti
f
ic
d
is
cip
li
n
es.
Or
g
an
iza
tio
n
a
n
d
citatio
n
o
f
t
h
e
b
ib
lio
g
r
ap
h
y
ar
e
m
ad
e
i
n
Van
co
u
v
er
s
t
y
le
i
n
s
ig
n
[
1
]
,
[
2
]
an
d
s
o
o
n
.
T
h
e
ter
m
s
i
n
f
o
r
eig
n
la
n
g
u
a
g
es
ar
e
w
r
itte
n
i
talic
(
italic)
.
T
h
e
t
ex
t
s
h
o
u
ld
b
e
d
iv
id
ed
in
to
s
e
ctio
n
s
,
ea
c
h
w
it
h
a
s
ep
ar
ate
h
ea
d
in
g
a
n
d
n
u
m
b
er
ed
co
n
s
ec
u
t
iv
el
y
.
T
h
e
s
ec
tio
n
/
s
u
b
s
ec
tio
n
h
ea
d
i
n
g
s
s
h
o
u
ld
b
e
t
y
p
ed
o
n
a
s
ep
ar
ate
lin
e
,
e.
g
.
,
1
.
I
ntr
o
du
ct
io
n
[
3
]
.
A
u
t
h
o
r
s
ar
e
s
u
g
g
es
ted
to
p
r
esen
t
th
eir
ar
ticles
i
n
th
e
s
ec
tio
n
s
tr
u
c
tu
r
e
:
I
ntr
o
du
ct
io
n
-
t
he
co
m
prehe
ns
iv
e
t
h
eo
re
t
ica
l
ba
s
is
a
nd
/o
r
t
he
P
ro
po
s
ed
M
e
t
ho
d/Alg
o
rit
h
m
-
Resea
rc
h
M
e
t
ho
d
-
Resul
t
s
a
nd
Dis
c
us
s
io
n
–
Co
nclus
io
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
3
,
No
.
3
,
Sep
tem
b
er
20
1
4
:
1
0
5
–
1
1
1
108
L
iter
at
u
r
e
r
ev
ie
w
th
at
h
as
b
ee
n
d
o
n
e
au
th
o
r
u
s
ed
in
t
h
e
ch
ap
ter
"
I
n
tr
o
d
u
ctio
n
"
to
e
x
p
lain
t
h
e
d
if
f
er
e
n
ce
o
f
th
e
m
a
n
u
s
cr
ip
t
w
it
h
o
th
er
p
ap
er
s
,
th
a
t
it
is
i
n
n
o
v
ati
v
e,
it
ar
e
u
s
ed
in
t
h
e
ch
ap
ter
"
R
esear
ch
Me
th
o
d
"
to
d
escr
ib
e
th
e
s
tep
o
f
r
esear
ch
an
d
u
s
ed
in
t
h
e
ch
ap
ter
"
R
esu
lt
s
an
d
Dis
c
u
s
s
i
o
n
"
to
s
u
p
p
o
r
t
th
e
an
al
y
s
is
o
f
t
h
e
r
es
u
lt
s
[
2
]
.
I
f
t
h
e
m
an
u
s
cr
ip
t
w
as
w
r
it
ten
r
ea
ll
y
h
av
e
h
i
g
h
o
r
ig
i
n
ali
t
y
,
w
h
i
ch
p
r
o
p
o
s
ed
a
n
e
w
m
et
h
o
d
o
r
alg
o
r
ith
m
,
t
h
e
ad
d
itio
n
al
c
h
ap
ter
af
ter
th
e
"
I
n
tr
o
d
u
ctio
n
"
ch
ap
ter
a
n
d
b
ef
o
r
e
th
e
"
R
esear
c
h
Me
th
o
d
"
ch
ap
ter
ca
n
b
e
ad
d
ed
to
ex
p
lain
b
r
ief
l
y
th
e
t
h
eo
r
y
a
n
d
/o
r
th
e
p
r
o
p
o
s
ed
m
eth
o
d
/
al
g
o
r
ith
m
[
4
]
.
4.
SYST
E
M
ARCH
I
T
E
CT
U
R
E
I
n
s
tead
o
f
u
s
i
n
g
a
s
in
g
le
m
o
n
o
lith
ic
ar
ch
itect
u
r
e
s
y
s
te
m
u
s
es
th
e
h
y
b
r
id
ap
p
r
o
ac
h
w
h
ic
h
co
m
b
i
n
es
co
n
s
tr
ain
t
-
b
ased
a
n
d
p
r
ef
er
en
ce
s
-
b
ased
ar
ch
itec
tu
r
es,
as
s
h
o
w
n
in
t
h
e
F
i
g
u
r
e
1
.
S
y
s
te
m
tak
es
i
n
p
u
t
i
n
th
e
f
o
r
m
o
f
tex
t
f
i
les
a
n
d
ass
i
g
n
s
co
n
tr
o
l
to
th
e
s
e
n
te
n
c
e
s
p
litt
er
.
Sen
te
n
ce
s
p
litt
er
s
p
its
t
h
e
s
e
n
ten
ce
s
a
n
d
ass
i
g
n
s
tag
s
i
n
t
h
e
b
eg
i
n
n
in
g
a
n
d
en
d
o
f
ea
ch
s
e
n
te
n
ce
as
p
er
r
eq
u
ir
ed
b
y
t
h
e
p
ar
s
er
.
Ou
tp
u
t
o
f
t
h
e
s
en
te
n
ce
s
p
litt
er
i
s
th
en
g
iv
e
n
to
th
e
p
ar
s
er
.
P
ar
s
er
u
s
ed
b
y
t
h
e
s
y
s
te
m
is
C
h
ar
n
iak
'
s
P
ar
s
er
(
p
ar
s
er
0
5
A
u
g
1
6
)
w
h
ic
h
ta
g
s
th
e
te
x
t
an
d
g
e
n
er
ates
p
ar
s
e
tr
ee
.
Fro
m
t
h
e
s
y
n
tactic
s
tr
u
ctu
r
al
g
e
n
er
ated
b
y
th
e
p
ar
s
er
,
s
y
s
te
m
g
en
er
ates
t
w
o
lis
t
s
o
f
an
ap
h
o
r
a
an
d
n
o
u
n
p
h
r
ase
s
.
Fro
m
t
h
e
lis
t
o
f
p
r
o
n
o
u
n
a
n
d
n
o
u
n
p
h
r
a
s
es,
all
p
o
s
s
ib
le
p
air
s
o
f
a
n
ap
h
o
r
a
an
d
an
tece
d
en
t
s
ar
e
g
en
er
ate
s
.
E
ac
h
p
air
is
th
en
f
ilter
ed
th
r
o
u
g
h
ag
r
ee
m
en
t
f
ilter
w
h
ich
c
h
ec
k
s
f
o
r
co
m
p
atib
ilit
y
o
f
ea
ch
p
air
o
n
th
e
b
asi
s
o
f
a
g
r
ee
m
e
n
t
f
ea
t
u
r
es.
All
t
h
e
in
f
o
r
m
at
io
n
r
eq
u
ir
ed
b
y
t
h
e
s
y
s
t
e
m
i
s
n
o
t
g
en
er
ate
d
b
y
th
e
p
ar
s
er
;
h
e
n
ce
r
e
m
a
in
i
n
g
r
eq
u
ir
ed
in
f
o
r
m
a
ti
o
n
i
s
e
v
a
lu
ated
b
y
t
h
e
s
y
s
te
m
f
o
r
th
e
f
u
r
t
h
er
p
r
o
ce
s
s
i
n
g
.
Nex
t
s
tep
is
to
ap
p
l
y
p
leo
n
as
tic
p
r
o
n
o
u
n
f
i
lter
w
h
ich
w
ill
tak
e
li
s
t
o
f
an
ap
h
o
r
a
as
a
n
i
n
p
u
t.
A
f
ter
ap
p
l
y
i
n
g
p
leo
n
asti
c
f
ilter
,
p
er
s
o
n
al
p
r
o
n
o
u
n
f
ilter
i
s
ap
p
lied
;
it
co
n
s
i
d
er
s
th
e
li
s
t
o
f
a
n
ap
h
o
r
a
an
d
lis
t
o
f
n
o
u
n
p
h
r
ase
cr
ea
ted
b
y
th
e
s
y
s
te
m
.
T
h
er
e
is
a
p
o
s
s
ib
ilit
y
t
h
at
f
o
r
a
p
ar
ticu
lar
an
ap
h
o
r
a
m
o
r
e
th
a
n
o
n
e
p
o
ten
tial
ca
n
d
id
ate
an
tece
d
en
t
ca
n
b
e
p
r
esen
t.
So
,
f
r
o
m
t
h
e
li
s
t
o
f
all
p
o
ten
tial
ca
n
d
id
ate
an
tece
d
e
n
ts
,
f
i
n
al
a
n
tece
d
en
t
is
ch
o
s
e
n
w
it
h
th
e
h
elp
o
f
s
alie
n
ce
w
ei
g
h
t.
T
h
e
ar
ch
i
tectu
r
e
o
f
“
P
r
o
n
o
m
i
n
al
An
ap
h
o
r
a
R
eso
l
u
tio
n
”
i
s
g
iv
e
n
i
n
Fi
g
u
r
e
1
in
th
e
p
icto
r
ial
f
o
r
m
ab
o
v
e
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
Hyb
r
id
Mo
d
el
o
f A
u
to
ma
te
d
A
n
a
p
h
o
r
a
R
eso
lu
tio
n
(
K
a
lya
n
i
K
a
mu
n
e
)
109
5.
I
SS
U
E
S
I
N
ANAP
H
O
RA
R
E
SO
L
UT
I
O
N
T
h
e
B
asically
,
t
h
er
e
ar
e
t
w
o
m
a
in
ap
p
r
o
ac
h
es
f
o
r
r
eso
lv
i
n
g
a
n
ap
h
o
r
a:
(
1
)
T
r
ad
itio
n
al
A
p
p
r
o
ac
h
,
w
h
ic
h
u
s
u
all
y
d
ep
en
d
s
u
p
o
n
lin
g
u
i
s
tics
k
n
o
w
led
g
e,
an
d
(
2
)
th
e
Dis
co
u
r
s
e
-
o
r
ie
n
ted
Ap
p
r
o
ac
h
,
h
er
e
th
e
r
esear
ch
er
tr
ies
to
m
o
d
el
co
m
p
lex
d
is
co
u
r
s
e
s
tr
u
ct
u
r
e
an
d
th
en
u
s
es
s
tr
u
ctu
r
e
s
f
o
r
th
e
p
r
o
ce
s
s
o
f
a
n
ap
h
o
r
a
r
eso
lu
tio
n
.
T
r
ad
itio
n
al
ap
p
r
o
ac
h
es
ap
p
l
y
lin
g
u
is
tics
k
n
o
w
led
g
e,
in
t
h
e
w
a
y
o
f
“
P
r
ef
er
en
ce
s
”
a
n
d
“Co
n
s
tr
ain
t
s
”,
in
w
h
ic
h
s
y
s
te
m
s
ca
n
b
e
p
r
o
p
o
s
ed
as
a
tech
n
iq
u
e
f
o
r
co
m
b
i
n
i
n
g
v
ar
io
u
s
i
n
f
o
r
m
atio
n
s
o
u
r
ce
s
.
T
r
a
d
itio
n
al
ap
p
r
o
ac
h
es
ar
e
m
a
in
l
y
w
o
r
k
s
i
n
b
asic
t
h
r
ee
s
t
ep
s
as:
(
1
)
Dec
id
in
g
s
ea
r
ch
li
m
it
o
r
an
ap
h
o
r
ic
ac
ce
s
s
ib
ilit
y
s
p
ac
e,
(
2
)
ap
p
ly
v
ar
io
u
s
co
n
s
tr
ai
n
ts
,
a
n
d
th
e
n
(
3
)
ap
p
ly
p
r
ef
er
en
ce
s
.
[
3
]
A
.
Sear
c
h
li
m
it o
r
An
ap
h
o
r
ic
A
cc
e
s
s
ib
ilit
y
Sp
ac
e:
T
h
is
r
ep
r
esen
ts
a
li
m
i
t
f
o
r
s
e
ar
ch
in
g
al
l
p
o
s
s
ib
le
ca
n
d
id
ate
an
tece
d
en
ts
f
o
r
a
p
ar
tic
u
lar
an
ap
h
o
r
a.
S
y
s
te
m
d
ef
i
n
es
th
e
tex
t
s
e
g
m
en
ts
w
it
h
in
w
h
ich
an
tece
d
e
n
t
f
o
r
a
p
ar
ticu
lar
a
n
ap
h
o
r
a
ca
n
b
e
f
o
u
n
d
.
Fi
n
d
in
g
'An
ap
h
o
r
ic
A
cc
e
s
s
ib
ili
t
y
Sp
ac
e'
i
s
a
v
er
y
cr
u
cia
l
p
h
a
s
e
i
n
t
h
e
p
r
o
ce
s
s
o
f
a
n
ap
h
o
r
a
r
eso
l
u
ti
o
n
.
B
ec
au
s
e,
s
m
a
ll
s
ea
r
ch
li
m
it
r
es
u
lt
s
i
n
t
h
e
ex
cl
u
s
io
n
o
f
v
alid
an
tece
d
en
ts
a
n
d
to
o
b
r
o
a
d
s
ea
r
ch
li
m
i
t
r
esu
lts
in
lar
g
e
ca
n
d
id
ate
lis
ts
,
w
h
ic
h
u
l
ti
m
atel
y
r
e
s
u
l
t
s
in
er
r
o
n
eo
u
s
an
ap
h
o
r
a
r
eso
lu
tio
n
.
Ge
n
er
all
y
,
s
ea
r
ch
li
m
it
o
r
'An
ap
h
o
r
ic
A
cc
e
s
s
ib
ilit
y
Sp
ac
e
'
i
s
d
e
f
in
ed
as
'n
'
p
r
ev
io
u
s
s
e
n
te
n
ce
to
an
ap
h
o
r
a,
w
h
er
e
v
al
u
e
o
f
'n
'
v
ar
i
es
ac
co
r
d
in
g
to
t
h
e
k
in
d
o
f
an
ap
h
o
r
a.
A
cc
o
r
d
in
g
t
o
R
u
s
la
n
Mitk
o
v
(
2
0
0
8
)
,
th
e
i
d
ea
l
an
ap
h
o
r
a
r
eso
lu
tio
n
s
y
s
te
m
,
v
al
u
e
o
f
'n
'
is
1
7
i.e
.
S
y
s
te
m
c
h
ec
k
1
7
s
e
n
te
n
ce
s
a
w
a
y
f
r
o
m
t
h
e
s
en
t
en
ce
in
w
h
ich
an
ap
h
o
r
a
is
p
r
esen
t.
'An
ap
h
o
r
ic
A
cc
e
s
s
ib
ilit
y
Sp
ac
e
'
is
p
r
ed
ef
i
n
ed
b
y
th
e
d
ev
e
lo
p
er
,
if
an
y
a
n
ap
h
o
r
ic
w
o
r
d
is
r
ec
o
g
n
ized
b
y
t
h
e
s
y
s
te
m
t
h
e
n
lis
t
o
f
all
p
o
s
s
ib
le
ca
n
d
id
ates
f
o
r
an
tece
d
en
t
w
it
h
i
n
th
e
p
r
ed
ef
i
n
ed
s
ea
r
ch
li
m
it
is
f
o
u
n
d
o
u
t
f
o
r
th
at
p
ar
ticu
lar
an
ap
h
o
r
a.
B
.
C
o
n
s
tr
ain
t
s
Af
ter
g
et
tin
g
t
h
e
lis
t
o
f
all
p
o
s
s
ib
le
an
tece
d
en
t
s
,
s
ev
er
al
co
n
s
tr
ain
t
s
ar
e
ap
p
lied
f
o
r
r
e
m
o
v
i
n
g
th
e
in
co
m
p
atib
le
a
n
tece
d
en
t.
Us
u
all
y
,
co
n
s
tr
ai
n
t
s
h
o
ld
ce
r
tain
co
n
d
itio
n
s
t
h
at
m
u
s
t
b
e
f
u
lf
i
ll
ed
b
y
th
e
ca
n
d
id
ate
a
n
tece
d
en
t,
if
it
f
ai
ls
,
t
h
e
n
ca
n
d
id
ate
w
ill
n
o
t
b
e
co
n
s
id
er
ed
p
o
s
s
ib
le
an
tece
d
e
n
ts
f
o
r
t
h
e
an
ap
h
o
r
.
Var
io
u
s
in
f
o
r
m
atio
n
l
ik
e
s
y
n
tactica
l,
s
e
m
an
tic,
m
o
r
p
h
o
lo
g
ical,
a
n
d
lex
ical
ar
e
u
s
ed
to
d
ef
i
n
e
v
ar
io
u
s
co
n
s
tr
ain
ts
.
C
.
P
r
ef
er
en
ce
s
Af
ter
r
e
m
o
v
i
n
g
all
i
n
co
m
p
atib
le
ca
n
d
id
ates
f
o
r
an
tece
d
e
n
t,
i
f
t
h
e
r
e
m
a
in
i
n
g
li
s
t
o
f
an
tece
d
en
t
co
n
ta
in
s
m
o
r
e
t
h
an
o
n
e
an
tece
d
en
t,
th
en
p
r
ef
er
e
n
ce
s
ar
e
ap
p
lied
f
o
r
s
elec
tin
g
o
n
l
y
o
n
e
p
o
ten
tial
an
tece
d
en
t.
P
r
ef
er
en
ce
s
y
s
te
m
m
u
s
t b
e
d
ev
elo
p
in
g
b
y
k
ee
p
i
n
g
in
m
i
n
d
th
at
o
n
l
y
a
s
i
n
g
le
ca
n
d
id
ate
m
u
s
t r
e
m
ai
n
at
t
h
e
en
d
o
f
p
r
o
ce
s
s
.
An
d
t
h
ese
f
i
n
al
ca
n
d
id
ates
g
i
v
en
b
y
t
h
e
p
r
e
f
er
en
ce
s
y
s
te
m
w
ill
b
e
co
n
s
id
er
ed
as
a
p
o
ten
tia
l
ca
n
d
id
ate
f
o
r
th
at
p
ar
ticu
lar
an
ap
h
o
r
a.
Var
io
u
s
in
f
o
r
m
a
tio
n
lik
e
s
y
n
tactica
l,
s
e
m
a
n
tic,
m
o
r
p
h
o
lo
g
ical,
an
d
lex
ical
ar
e
u
s
ed
to
d
ef
in
e
p
r
ef
er
en
ce
s
y
s
te
m
.
A
tex
t
co
n
tain
s
l
in
g
u
is
t
ic
d
ata
in
m
an
y
f
o
r
m
s
u
c
h
a
s
s
y
n
tact
ic
p
ar
alleli
s
m
,
an
tece
d
en
t
p
r
o
x
i
m
it
y
,
g
en
d
er
a
n
d
n
u
m
b
er
ag
r
ee
m
en
t,
le
x
ical
r
ep
etitio
n
o
r
c
-
co
m
m
a
n
d
r
estricti
o
n
s
p
la
y
s
an
v
ital
r
o
le
in
t
h
e
an
ap
h
o
r
a
r
eso
lu
tio
n
p
r
o
ce
s
s
.
Var
io
u
s
m
et
h
o
d
o
lo
g
ies s
u
ch
as
s
tati
s
tical
a
n
d
p
r
o
b
ab
ilis
tic
m
o
d
els,
K
n
o
w
led
g
e
-
p
o
o
r
s
o
lu
tio
n
s
,
u
s
in
g
co
r
p
u
s
-
d
r
iv
e
n
m
et
h
o
d
o
lo
g
ies
ar
e
p
r
ef
er
r
ed
f
o
r
r
eso
lv
in
g
a
n
ap
h
o
r
a.
Op
p
o
s
ite
to
th
e
p
u
r
e
s
tatis
t
ical
m
o
d
el,
s
o
m
e
s
tr
ate
g
ical
a
p
p
r
o
ac
h
es
h
av
e
also
b
ee
n
p
r
o
p
o
s
ed
f
o
r
tr
a
ck
in
g
t
h
e
an
tece
d
e
n
t,
w
h
ic
h
c
an
b
e
f
o
r
m
alize
d
in
ter
m
s
o
f
r
u
les
b
ased
o
n
'
p
r
ef
er
en
ce
s
'
an
d
'
co
n
s
tr
ai
n
t
s
'
.
Su
c
h
s
tr
ate
g
ical
ap
p
r
o
ac
h
u
s
u
all
y
co
m
b
in
e
s
o
f
'
co
n
s
tr
ai
n
ts
'
a
n
d
'
p
r
ef
er
e
n
ce
s
'
.
T
h
e
w
o
r
k
i
n
g
o
f
a
n
ap
h
o
r
a
r
eso
l
u
tio
n
s
y
s
te
m
r
elies
o
n
th
e
s
et
o
f
v
ar
io
u
s
a
n
ap
h
o
r
a
r
eso
lu
tio
n
f
ac
to
r
s
.
T
h
ese
f
ac
to
r
s
ca
n
b
e
eith
er
“
e
li
m
i
n
ati
n
g
”
i.e
.
it
d
o
es
n
o
t
co
u
n
t
s
o
m
e
ca
n
d
id
ates
f
r
o
m
t
h
e
lis
t
o
f
a
ll
p
o
s
s
ib
le
ca
n
d
id
ates
o
r
“
p
r
ef
er
en
t
ial”
i
.
e.
I
t
g
iv
e
m
o
r
e
p
r
ef
er
en
ce
t
o
s
o
m
e
ca
n
d
id
at
es
t
h
an
r
e
m
ain
i
n
g
ca
n
d
id
ates.
P
ar
titi
o
n
o
f
an
ap
h
o
r
a
r
eso
lu
ti
o
n
f
ac
to
r
s
in
to
p
r
ef
er
en
ce
s
an
d
co
n
s
tr
ai
n
ts
is
r
esp
o
n
s
ib
le
f
o
r
th
e
p
r
ef
er
en
ce
s
-
b
ased
an
d
co
n
s
tr
ain
ts
-
b
ased
ar
ch
itect
u
r
e
in
an
ap
h
o
r
a
r
eso
lu
tio
n
.
I
n
s
tead
o
f
u
s
in
g
s
i
n
g
le
p
r
ef
er
en
ce
s
-
b
ased
o
r
co
n
s
tr
a
in
ts
-
b
ased
ar
ch
itect
u
r
e
,
in
o
u
r
s
y
s
te
m
w
e
h
av
e
u
s
e
th
e
co
m
b
i
n
atio
n
o
f
p
r
ef
er
en
ce
s
-
b
ased
an
d
co
n
s
tr
ain
ts
-
b
ased
ar
ch
itec
tu
r
e,
in
o
r
d
er
to
g
et
t
h
e
m
o
r
e
e
f
f
i
cien
t
r
es
u
lt
s
.
W
e
h
a
v
e
u
s
ed
t
h
is
ar
c
h
itect
u
r
e
f
o
r
o
u
r
s
y
s
te
m
b
ec
a
u
s
e
s
t
u
d
y
s
h
o
w
s
t
h
at
an
ap
h
o
r
a
r
eso
lu
ti
o
n
s
y
s
te
m
s
b
ased
o
n
co
n
s
tr
ai
n
ts
a
n
d
p
r
ef
er
en
ce
s
ca
n
g
iv
e
a
s
u
cc
e
s
s
f
u
l r
esu
l
t
w
h
e
n
ap
p
lied
to
n
o
n
-
d
ialo
g
u
e
tex
t
s
.
6.
E
VA
L
UA
T
I
O
N
T
esti
n
g
o
f
S
y
s
te
m
p
er
f
o
r
m
s
m
an
u
all
y
.
Fo
r
t
h
e
te
s
ti
n
g
p
u
r
p
o
s
e,
w
e
h
av
e
u
s
e
'
R
e
u
ter
s
Ne
w
s
p
ap
er
co
r
p
u
s
'
.
I
n
R
eu
ter
s
Ne
w
s
p
ap
er
co
r
p
u
s
'
,
to
tal
4
0
2
4
f
ile
s
ar
e
p
r
esen
t
in
t
h
e
te
s
t
s
ec
tio
n
an
d
to
tal
1
1
,
4
1
3
f
iles
ar
e
p
r
esen
t
i
n
t
h
e
T
r
ain
i
n
g
s
ec
tio
n
.
I
n
'
R
e
u
ter
s
Ne
w
s
p
ap
er
co
r
p
u
s
'
,
f
ile
s
ar
e
p
r
ese
n
t
i
n
to
tal
9
1
v
ar
io
u
s
ca
te
g
o
r
ies
lik
e
h
o
u
s
in
g
,
i
n
co
m
e,
j
o
b
s
,
m
o
n
e
y
-
s
u
p
p
l
y
,
li
v
esto
c
k
,
r
etai
l,
in
ter
est,
s
il
v
er
,
tr
ad
e
an
d
s
o
o
n
in
b
o
t
h
test
a
n
d
tr
ain
i
n
g
s
ec
tio
n
.
I
n
th
e
'R
e
u
ter
s
Ne
w
s
p
ap
er
co
r
p
u
s
'
,
to
tal
1
5
,
4
3
7
f
iles
ar
e
p
r
ese
n
t
d
is
tr
ib
u
te
d
o
v
er
9
1
d
if
f
er
e
n
t
ca
teg
o
r
ies
i
n
b
o
th
test
a
n
d
tr
ain
i
n
g
s
ec
tio
n
.
Fil
e
s
i
n
t
h
e
'R
eu
ter
s
Ne
w
s
p
ap
er
co
r
p
u
s
'
a
r
e
p
r
esen
t
i
n
T
XT
f
o
r
m
at,
o
f
m
i
n
i
m
u
m
s
ize
6
9
3
b
y
te
a
n
d
m
a
x
i
m
u
m
s
ize
o
f
4
.
3
k
b
ap
p
r
o
x
im
a
tel
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
3
,
No
.
3
,
Sep
tem
b
er
20
1
4
:
1
0
5
–
1
1
1
110
„
T
ab
le
I
I
I
‟
s
h
o
w
s
t
h
at
test
in
g
o
f
s
y
s
te
m
is
p
er
f
o
r
m
ed
o
n
t
o
tal
1
2
0
f
ile
o
f
d
if
f
er
en
t
ca
te
g
o
r
ies
o
f
'R
eu
ter
s
Ne
w
s
p
ap
er
co
r
p
u
s
'
.
I
n
th
e
s
e
1
2
0
f
iles
,
to
tal
4
9
2
an
ap
h
o
r
a
-
an
tece
d
e
n
t
p
air
s
w
er
e
f
o
u
n
d
o
u
t
o
f
w
h
ic
h
4
0
4
p
air
s
w
er
e
co
r
r
ec
tly
id
en
t
if
ied
b
y
th
e
s
y
s
te
m
.
He
n
ce
,
w
e
ca
n
s
a
y
t
h
at
ef
f
icie
n
c
y
o
f
th
e
s
y
s
te
m
is
8
2
.
1
1
%.
1
2
0
f
iles
w
h
ic
h
w
e
h
a
v
e
u
s
ed
f
o
r
th
e
te
s
tin
g
p
u
r
p
o
s
e
ar
e
p
r
esen
t
i
n
T
XT
f
o
r
m
at,
o
f
m
i
n
i
m
u
m
s
ize
6
9
3
b
y
t
e
an
d
m
ax
i
m
u
m
s
ize
o
f
4
.
3
k
b
ap
p
r
o
x
im
a
tel
y
.
W
e
ca
n
s
a
y
t
h
at
th
e
av
er
ag
e
s
ize
o
f
ea
c
h
f
il
e
is
2
.
5
k
b
.
On
an
av
er
ag
e,
w
e
ca
n
s
a
y
t
h
at
ea
c
h
f
ile
co
n
ta
in
s
ap
p
r
o
x
i
m
atel
y
4
p
air
s
o
f
an
ap
h
o
r
a
-
a
n
tece
d
en
t p
air
.
T
ab
le
3
.
Sh
o
w
s
t
h
at
T
esti
n
g
o
f
S
y
s
te
m
I
n
to
tal
4
9
2
p
air
s
o
f
An
ap
h
o
r
a
-
an
tece
d
e
n
t,
all
t
h
ir
d
p
er
s
o
n
p
er
s
o
n
al
(
h
e,
h
i
m
,
s
h
e,
h
er
,
it,
th
e
y
,
th
e
m
)
,
p
o
s
s
ess
iv
e
(
h
i
s
,
h
er
,
h
e
r
s
,
its
,
t
h
eir
,
t
h
eir
s
)
,
an
d
r
e
f
le
x
iv
e
(
h
i
m
s
el
f
,
h
er
s
el
f
,
its
el
f
,
t
h
e
m
s
el
v
es)
p
r
o
n
o
u
n
s
in
b
o
th
s
i
n
g
u
lar
a
n
d
p
lu
r
al,
ar
e
test
ed
alo
n
g
w
it
h
t
h
e
o
cc
u
r
r
en
ce
o
f
p
leo
n
asti
c
it
a
n
d
th
e
lex
ical
an
ap
h
o
r
a.
Fro
m
t
h
e
r
es
u
lt
co
m
i
n
g
o
u
t
f
r
o
m
t
h
e
e
x
p
er
i
m
en
t,
s
y
s
te
m
co
r
r
ec
tly
f
o
u
n
d
s
to
tal
2
7
p
a
ir
s
o
f
th
ir
d
p
er
s
o
n
p
er
s
o
n
al
f
r
o
m
4
5
p
air
s
,
1
4
4
p
air
s
o
f
p
o
s
s
ess
iv
e
p
r
o
n
o
u
n
s
f
r
o
m
1
6
7
p
air
s
,
an
d
1
9
1
o
cc
u
r
r
en
ce
o
f
„
I
t
‟
p
r
o
n
o
u
n
s
f
r
o
m
2
3
0
p
air
s
in
b
o
th
s
in
g
u
lar
an
d
p
lu
r
al
f
o
r
m
,
also
4
2
p
air
s
o
f
lex
ical
p
r
o
n
o
u
n
s
f
r
o
m
5
0
p
air
s
as
s
h
o
w
s
i
n
f
i
g
u
r
e
3
.
Hen
ce
w
e
ca
n
s
a
y
t
h
at
s
y
s
te
m
h
a
v
e
6
0
%
ef
f
icie
n
c
y
to
f
in
d
th
ir
d
p
er
s
o
n
al,
8
6
.
2
2
%
ef
f
icien
c
y
to
f
in
d
p
o
s
s
e
s
s
i
v
e
p
r
o
n
o
u
n
,
8
3
.
0
4
%
ef
f
icie
n
c
y
in
f
in
d
i
n
g
„
I
t
‟
p
r
o
n
o
u
n
,
a
n
d
s
y
s
te
m
h
a
v
e
8
4
%
ef
f
icien
c
y
to
f
in
d
le
x
ical
an
ap
h
o
r
a.
Fig
u
r
e
3
.
P
er
f
o
r
m
a
n
ce
o
f
S
y
s
t
e
m
W
e
ca
n
s
a
y
t
h
at
s
y
s
te
m
a
l
m
o
s
t f
o
u
n
d
s
2
7
o
f
t
h
ir
d
p
er
s
o
n
p
er
s
o
n
al,
1
4
4
p
air
s
o
f
p
o
s
s
ess
iv
e,
1
9
1
p
air
s
o
f
„
it‟
p
r
o
n
o
u
n
s
a
n
d
4
2
p
air
s
o
f
lex
ical
p
r
o
n
o
u
n
.
I
n
o
r
d
er
to
i
m
p
r
o
v
e
th
e
p
er
f
o
r
m
an
ce
o
f
t
h
e
s
y
s
te
m
p
o
w
er
f
u
l
p
er
s
o
n
al
p
r
o
n
o
u
n
an
d
ag
r
ee
m
en
t
f
i
lter
p
la
y
s
a
n
i
m
p
o
r
tan
t
r
o
le.
T
h
e
w
a
y
i
n
w
h
ich
i
m
p
le
m
en
tatio
n
e
x
tr
ac
t
th
e
g
r
a
m
m
atica
l
r
o
les
b
y
ap
p
l
y
i
n
g
ce
r
tain
h
a
n
d
-
cr
a
f
ted
r
u
le
s
o
n
th
e
p
ar
s
e
tr
ee
a
ls
o
a
f
f
ec
t
t
h
e
o
v
er
all
p
er
f
o
r
m
a
n
ce
o
f
th
e
s
y
s
te
m
.
U
s
e
o
f
k
n
o
w
led
g
e
r
ich
C
h
ar
n
ak
p
ar
s
er
(
p
ar
s
er
05
au
g
1
6
)
in
s
y
s
te
m
,
a
l
s
o
co
n
tr
ib
u
te
s
in
o
v
er
all
p
er
f
o
r
m
a
n
ce
o
f
t
h
e
s
y
s
te
m
.
A
lo
n
g
w
i
th
t
h
e
in
tr
a
-
s
e
n
te
n
tia
l
an
ap
h
o
r
a
s
y
s
te
m
al
s
o
f
i
n
d
s
i
n
ter
-
s
en
ten
t
ial
an
ap
h
o
r
a.
Ou
t
o
f
th
e
to
tal
4
9
2
p
air
s
o
f
an
ap
h
o
r
a
an
d
a
n
t
ec
ed
en
t,
to
tal
3
8
7
p
air
s
w
er
e
i
n
tr
a
-
s
en
ten
t
ial
a
n
d
to
tal
1
0
5
p
air
s
w
e
r
e
f
o
u
n
d
to
b
e
in
ter
-
s
e
n
te
n
tial.
Ou
t
o
f
t
o
tal
3
8
7
in
tr
a
-
s
e
n
ten
tial
p
air
s
,
s
y
s
te
m
f
o
u
n
d
s
3
1
4
an
ap
h
o
r
a
-
an
tece
d
en
t
p
air
co
r
r
ec
tly
a
n
d
o
u
t o
f
to
tal
1
0
5
i
n
ter
-
s
en
ten
t
ial
p
air
s
y
s
te
m
f
i
n
d
s
9
0
p
air
s
co
r
r
ec
tly
.
He
n
ce
,
e
f
f
icien
c
y
o
f
s
y
s
te
m
to
f
i
n
d
in
ter
-
s
e
n
te
n
tial
a
n
d
i
n
t
r
a
-
s
e
n
ten
tial
a
n
ap
h
o
r
a
-
a
n
tece
d
en
t
p
air
s
is
8
5
.
7
1
an
d
8
1
.
1
3
r
esp
ec
tiv
el
y
,
w
h
ich
is
s
h
o
w
n
b
y
F
ig
u
r
e
4
in
a
p
icto
r
ial
f
o
r
m
.
T
o
t
a
l
F
i
l
e
c
o
n
si
d
e
r
e
d
T
o
t
a
l
N
u
mb
e
r
O
f
A
n
a
p
h
o
r
a
-
a
n
t
e
c
e
d
e
n
t
P
a
i
r
s Pr
e
se
n
t
N
u
mb
e
r
o
f
c
a
se
s t
h
a
t
t
h
e
sy
st
e
m re
so
l
v
e
s
c
o
r
r
e
c
t
l
y
A
c
c
u
r
a
c
y
%
1
2
0
4
9
2
4
0
4
8
2
.
1
1
%
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
Hyb
r
id
Mo
d
el
o
f A
u
to
ma
te
d
A
n
a
p
h
o
r
a
R
eso
lu
tio
n
(
K
a
lya
n
i
K
a
mu
n
e
)
111
Fig
u
r
e
4
.
E
f
f
icie
n
c
y
o
f
S
y
s
te
m
in
P
er
ce
n
ta
g
e
7.
CO
NCLU
SI
O
N
S
y
s
te
m
w
h
ic
h
u
s
e
s
th
e
co
m
b
in
atio
n
o
f
co
n
s
tr
ai
n
t
-
b
ased
an
d
p
r
ef
er
en
ce
s
-
b
ased
ar
ch
ite
ctu
r
es,
f
o
r
r
eso
lv
in
g
an
ap
h
o
r
s
is
r
ep
r
esen
ted
as
ab
o
v
e.
I
t
is
u
s
ed
to
id
en
tify
i
n
ter
-
s
e
n
te
n
tial
a
n
d
in
tr
a
-
s
e
n
ten
t
ial
an
tece
d
en
t
s
o
f
“
T
h
ir
d
p
er
s
o
n
p
r
o
n
o
u
n
a
n
ap
h
o
r
s
”,
“
P
leo
n
a
s
t
ic
it
”,
a
n
d
“L
e
x
ical
n
o
u
n
p
h
r
ase
a
n
ap
h
o
r
a”
.
A
n
alg
o
r
ith
m
at
f
ir
s
t
d
ef
in
e
s
an
an
ap
h
o
r
ic
ac
ce
s
s
ib
ilit
y
s
p
ac
e,
th
e
n
ap
p
l
y
i
n
g
co
n
s
tr
ain
t
s
,
a
n
d
f
i
n
all
y
ap
p
l
y
in
g
p
r
ef
er
en
ce
s
.
A
n
ap
h
o
r
ic
ac
ce
s
s
ib
ilit
y
s
p
ac
e
is
s
elec
ted
ca
r
ef
u
ll
y
w
h
ich
is
n
o
t
to
o
b
r
o
ad
o
r
to
o
s
m
a
ll.
Sev
er
al
co
n
s
tr
ain
ts
ar
e
ap
p
lied
in
o
r
d
er
to
r
em
o
v
e
i
n
co
m
p
atib
le
a
n
tece
d
en
ts
f
o
r
a
p
ar
ticu
lar
a
n
a
p
h
o
r
.
Fin
all
y
,
a
f
ter
r
e
m
o
v
i
n
g
in
co
m
p
a
tib
le
ca
n
d
i
d
ates,
if
t
h
e
r
e
m
ai
n
i
n
g
lis
t
co
n
tai
n
s
m
o
r
e
th
a
n
o
n
e
a
n
tece
d
en
t,
s
alie
n
ce
w
ei
g
h
ts
ar
e
ap
p
lied
in
o
r
d
e
r
to
ch
o
o
s
e
a
s
in
g
le
an
tece
d
en
t.
T
h
e
alg
o
r
ith
m
ap
p
lies
to
th
e
o
u
tp
u
t
g
en
er
ated
b
y
C
h
ar
n
ia
k
p
ar
s
er
(
p
ar
s
er
0
5
A
u
g
1
6
)
an
d
r
elies
o
n
s
alie
n
ce
m
ea
s
u
r
es
d
er
iv
ed
f
r
o
m
p
ar
s
e
tr
ee
.
Var
io
u
s
f
ac
to
r
s
s
u
ch
as
"
elim
i
n
ato
r
y
"
i.e
.
d
o
esn
‟
t
co
u
n
t
ce
r
tain
n
o
u
n
p
h
r
ase
s
f
r
o
m
th
e
s
et
o
f
p
o
ten
t
ial
a
n
tece
d
en
t
s
(
s
u
c
h
as
g
en
d
er
,
n
u
m
b
er
,
p
eo
p
le
co
n
s
tr
ain
t
s
)
an
d
"
p
r
ef
er
en
tial"
i.e
.
ass
i
g
n
i
n
g
m
o
r
e
p
r
ef
er
e
n
ce
to
s
o
m
e
p
o
ten
tial
a
n
tece
d
en
ts
an
d
less
to
o
th
er
s
(
s
u
c
h
as
s
alien
ce
)
is
u
s
ed
in
o
r
d
er
t
o
r
eso
lv
e
th
e
a
n
ap
h
o
r
a.
W
e
h
av
e
test
ed
th
e
s
y
s
te
m
ex
ten
s
i
v
el
y
o
n
'
R
e
u
ter
s
Ne
w
s
p
ap
er
co
r
p
u
s
'
a
n
d
ef
f
icie
n
c
y
o
f
th
e
s
y
s
te
m
is
f
o
u
n
d
to
b
e
8
2
.
1
1
%.
RE
F
E
R
E
NC
E
S
[1
]
Ru
sla
n
M
it
k
o
v
,
ANA
PHORA
RE
S
OLUT
ION:
T
HE
S
T
AT
E
OF
T
HE
A
RT
,
In
tern
a
t
io
n
a
l
Co
n
f
e
re
n
c
e
o
n
M
a
th
e
m
a
ti
c
a
l
L
in
g
u
isti
c
s
,
2
0
0
8
[2
]
S
h
a
lo
m
L
a
p
p
in
a
n
d
He
rb
e
rt
J.
L
e
a
ss
,
A
n
A
lg
o
rit
h
m
f
o
r
P
r
o
n
o
m
in
a
l
A
n
a
p
h
o
ra
Re
so
l
u
ti
o
n
,
1
9
9
4
[3
]
Ch
in
a
tsu
A
o
n
e
a
n
d
S
c
o
tt
W
il
li
a
m
Be
n
n
e
tt
,
Eva
l
u
a
t
in
g
Au
t
o
ma
te
d
a
n
d
M
a
n
u
a
l
Acq
u
isit
i
o
n
o
f
An
a
p
h
o
ra
Res
o
lu
t
i
o
n
S
tra
teg
ies
,
In
tern
a
ti
o
n
a
l
W
o
rk
sh
o
p
o
n
S
h
a
ra
b
le Na
tu
ra
l
L
a
n
g
u
a
g
e
Re
so
u
rc
e
s (S
NL
R),
2
0
0
0
[4
]
M
it
k
o
v
a
n
d
R
u
sla
n
,
A
n
a
p
h
o
ra
re
so
lu
ti
o
n
in
Na
t
u
ra
l
L
a
n
g
u
a
g
e
P
ro
c
e
ss
in
g
a
n
d
M
a
c
h
in
e
T
ra
n
sla
ti
o
n
.
W
o
rk
in
g
p
a
p
e
r.
S
a
a
rb
rü
c
k
e
n
:
IA
I,
1
9
9
5
a
.
[5
]
M
it
k
o
v
,
Ru
sla
n
,
"
F
a
c
to
rs
in
a
n
a
p
h
o
r
a
re
so
lu
ti
o
n
:
th
e
y
a
re
n
o
t
th
e
o
n
ly
th
in
g
s
th
a
t
ma
t
ter
.
A
c
a
se
stu
d
y
b
a
se
d
o
n
two
d
if
fer
e
n
t
a
p
p
r
o
a
c
h
e
s"
P
r
o
c
e
e
d
in
g
s
o
f
th
e
A
C
L
'
9
7
/E
A
C
L'
9
7
w
o
rk
sh
o
p
o
n
Op
e
ra
ti
o
n
a
l
f
a
c
to
rs
in
p
ra
c
ti
c
a
l,
ro
b
u
st an
a
p
h
o
ra
re
so
l
u
ti
o
n
.
14
-
2
1
.
M
a
d
ri
d
,
S
p
a
in
,
1
9
9
7
b
.
Evaluation Warning : The document was created with Spire.PDF for Python.