I
nte
rna
t
io
na
l J
o
urna
l o
f
E
v
a
lua
t
io
n a
nd
Resea
rc
h in
E
du
ca
t
io
n (
I
J
E
RE
)
Vo
l.
5
,
No
.
3
,
Sep
tem
b
er
201
6
,
p
p
.
2
2
7
~
2
3
4
I
SS
N:
2252
-
8822
227
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
JE
R
E
Do
m
a
in
a
nd
Intell
ig
ence
Ba
sed
M
ulti
m
ed
ia
Q
u
estio
n
Answ
ering
S
y
stem
K
.
M
a
g
esh
K
u
m
a
r
,
P
.
Va
la
rm
a
t
hie
Co
m
p
u
ter S
c
ien
c
e
a
n
d
E
n
g
in
e
e
ri
n
g
,
S
a
v
e
e
th
a
En
g
in
e
e
rin
g
C
o
ll
e
g
e
,
Ch
e
n
n
a
i,
T
a
m
il
N
a
d
u
,
In
d
ia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
u
n
0
1
,
2
0
1
6
R
ev
i
s
ed
A
u
g
2
7
,
2
0
1
6
A
cc
ep
ted
A
u
g
3
0
,
2
0
1
6
M
u
lt
im
e
d
ia
q
u
e
stio
n
a
n
sw
e
rin
g
s
y
ste
m
s
h
a
v
e
b
e
c
o
m
e
v
e
r
y
p
o
p
u
la
r
o
v
e
r
th
e
p
a
st
f
e
w
y
e
a
rs.
It
a
ll
o
w
s
u
se
rs
t
o
sh
a
re
th
e
ir
th
o
u
g
h
ts
b
y
a
n
s
w
e
rin
g
g
iv
e
n
q
u
e
stio
n
o
r
o
b
tain
i
n
f
o
rm
a
ti
o
n
f
r
o
m
a
s
e
t
o
f
a
n
s
w
e
r
e
d
q
u
e
stio
n
s.
Ho
w
e
v
e
r,
e
x
isti
n
g
Q
A
s
y
ste
m
s
su
p
p
o
rt
o
n
l
y
te
x
tu
a
l
a
n
sw
e
r
w
h
ich
is
n
o
t
s
o
in
stru
c
ti
v
e
f
o
r
m
a
n
y
u
se
rs.
T
h
e
u
se
r’s
d
isc
u
ss
io
n
c
a
n
b
e
e
n
h
a
n
c
e
d
b
y
a
d
d
i
n
g
su
it
a
b
le
m
u
lt
i
m
e
d
ia
d
a
ta.
M
u
lt
im
e
d
ia
a
n
s
w
e
r
s
o
ffe
r
in
tu
it
iv
e
in
f
o
rm
a
ti
o
n
w
it
h
m
o
re
su
it
a
b
le
im
a
g
e
,
v
o
ice
a
n
d
v
id
e
o
.
T
h
is
s
y
ste
m
in
c
lu
d
e
s
a
se
t
o
f
in
f
o
rm
a
ti
o
n
a
s
w
e
ll
a
s
c
l
a
ss
i
f
ica
ti
o
n
o
f
q
u
e
sti
o
n
a
n
d
a
n
sw
e
r,
q
u
e
ry
g
e
n
e
ra
ti
o
n
,
m
u
lt
i
m
e
d
ia
d
a
ta
se
lec
ti
o
n
a
n
d
p
re
se
n
tatio
n
.
T
h
is
sy
ste
m
w
il
l
ta
k
e
a
ll
k
in
d
s
o
f
m
e
d
ia
su
c
h
a
s
tex
t,
i
m
a
g
e
s,
v
id
e
o
s,
a
n
d
v
id
e
o
s
w
h
ich
w
il
l
b
e
c
o
m
b
in
e
d
w
it
h
a
tex
tu
a
l
a
n
sw
e
r.
In
a
wa
y
,
it
a
u
to
m
a
ti
c
a
ll
y
c
o
ll
e
c
ts
in
f
o
r
m
a
ti
o
n
f
ro
m
th
e
u
se
r
to
im
p
ro
v
isin
g
th
e
a
n
s
w
e
r.
T
h
is
m
e
th
o
d
c
o
n
sists
o
f
ra
n
k
in
g
f
o
r
a
n
sw
e
rs
to
se
le
c
t
th
e
b
e
st
a
n
s
w
e
r.
B
y
d
e
a
li
n
g
o
u
t
a
h
u
g
e
se
t
o
f
Q
A
p
a
irs
a
n
d
a
d
d
i
n
g
th
e
m
to
a
d
a
tab
a
se
,
m
u
lt
im
e
d
ia
q
u
e
stio
n
a
n
sw
e
rin
g
a
p
p
ro
a
c
h
f
o
r
u
se
rs
w
h
ich
f
in
d
s
m
u
lt
im
e
d
ia
a
n
s
w
e
r
s
b
y
m
a
tch
in
g
th
e
ir
q
u
e
sti
o
n
s
w
it
h
th
o
se
in
th
e
d
a
tab
a
se
.
T
h
e
e
ff
e
c
ti
v
e
n
e
ss
o
f
M
u
lt
im
e
d
ia
s
y
ste
m
is
d
e
ter
m
in
e
d
b
y
ra
n
k
in
g
o
f
tex
t,
ima
g
e
,
a
u
d
io
a
n
d
v
id
e
o
i
n
u
se
rs
a
n
sw
e
r.
T
h
e
a
n
s
w
e
r
w
h
ich
is
g
iv
e
n
b
y
th
e
u
se
r
it
’s
p
ro
c
e
ss
e
d
b
y
S
e
m
a
n
ti
c
m
a
tch
a
l
g
o
rit
h
m
a
n
d
th
e
b
e
st
a
n
s
w
e
rs
c
a
n
b
e
v
iew
e
d
b
y
Na
iv
e
Ba
y
e
sia
n
ra
n
k
in
g
sy
ste
m
.
K
ey
w
o
r
d
:
Mu
lti
m
ed
ia
An
s
w
er
s
Naiv
e
B
a
y
es
ian
R
an
k
i
n
g
S
y
s
te
m
Qu
est
io
n
An
s
w
er
i
n
g
S
y
s
te
m
Qu
est
io
n
C
lass
if
ica
tio
n
R
an
k
i
n
g
T
ex
tu
al
A
n
s
w
er
s
Co
p
y
rig
h
t
©
2
0
1
6
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
:
K.
Ma
g
es
h
Ku
m
ar
,
P
G
Sch
o
lar
,
C
o
m
p
u
ter
Scie
n
c
e
an
d
E
n
g
i
n
ee
r
i
n
g
,
Sav
ee
t
h
a
E
n
g
i
n
ee
r
in
g
C
o
lla
g
e
,
C
h
e
n
n
ai,
T
am
il Na
d
u
,
I
n
d
ia.
E
m
ail:
k
u
m
ar
.
m
ag
e
s
h
5
5
4
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
Q
u
est
io
n
A
n
s
w
er
i
n
g
(
Q
A
)
is
a
t
y
p
e
o
f
i
n
f
o
r
m
atio
n
r
etr
iev
a
l
p
r
o
ce
s
s
in
g
.
T
h
e
s
y
s
te
m
r
etr
ie
v
e
an
s
w
er
s
to
q
u
esti
o
n
s
p
o
s
ed
i
n
n
at
u
r
al
lan
g
u
a
g
e.
Q
A
i
s
co
n
s
id
er
ed
as
r
eq
u
ir
i
n
g
m
o
r
e
m
u
lti
f
ac
ete
d
n
atu
r
al
lan
g
u
ag
e
p
r
o
ce
s
s
in
g
(
N
L
P
)
tech
n
iq
u
e
s
t
h
an
o
t
h
er
m
et
h
o
d
o
f
i
n
f
o
r
m
ati
o
n
r
etr
iev
al
s
u
c
h
as
d
o
cu
m
en
t
r
etr
iev
al,
an
d
it
i
s
o
cc
asio
n
all
y
r
e
g
ar
d
ed
as
th
e
n
ex
t
s
tep
b
e
y
o
n
d
s
ea
r
ch
e
n
g
i
n
es.
T
h
er
e
ar
e
t
w
o
t
y
p
es
o
f
QA
s
y
s
te
m
,
w
h
ic
h
is
c
lo
s
ed
d
o
m
ain
Q
A
s
y
s
te
m
a
n
d
Op
en
d
o
m
ai
n
Q
A
s
y
s
te
m
.
T
h
e
C
lo
s
ed
d
o
m
ai
n
q
u
esti
o
n
an
s
w
er
in
g
s
y
s
te
m
w
o
r
k
s
w
it
h
q
u
esti
o
n
s
u
n
d
er
a
s
p
ec
if
ic
d
o
m
a
in
b
u
t
i
n
a
n
Op
e
n
d
o
m
ai
n
q
u
e
s
tio
n
an
s
w
er
in
g
s
y
s
te
m
w
o
r
k
s
w
it
h
q
u
esti
o
n
s
ab
o
u
t a
n
y
s
u
b
j
ec
t
ar
ea
an
d
it c
an
b
e
f
o
u
n
d
ed
o
n
o
n
to
lo
g
ies a
n
d
w
o
r
ld
k
n
o
w
led
g
e
.
I
n
th
i
s
p
r
o
j
ec
t
th
e
Do
m
ain
a
n
d
I
n
tell
ig
e
n
ce
B
as
ed
Mu
lti
m
ed
ia
Qu
es
tio
n
An
s
w
er
i
n
g
(
DI
MQ
A
)
s
y
s
te
m
is
f
o
cu
s
ed
a
s
clo
s
ed
dom
ai
n
q
u
esti
o
n
an
s
w
er
i
n
g
s
y
s
te
m
h
as
to
b
e
d
ev
elo
p
ed
f
o
r
th
e
s
t
u
d
en
t
’
s
ed
u
ca
tio
n
in
o
r
d
er
to
en
h
a
n
ce
t
h
eir
k
n
o
w
led
g
e.
I
f
u
s
er
ca
n
h
av
e
a
n
y
d
o
u
b
ts
o
n
s
u
b
j
ec
ts
,
th
e
n
th
e
y
ca
n
s
ea
r
c
h
t
h
e
q
u
esti
o
n
i
n
DI
MQ
A
s
y
s
te
m
.
I
f
th
e
a
n
s
w
er
is
alr
ea
d
y
i
n
Q
A
s
y
s
te
m
,
t
h
en
it
w
i
ll
r
et
u
r
n
t
h
e
an
s
w
er
au
to
m
atica
ll
y
to
t
h
e
ir
q
u
esti
o
n
s
.
I
f
t
h
e
an
s
w
er
i
s
n
o
t
f
o
u
n
d
in
th
e
Q
A
s
y
s
te
m
m
ea
n
s
a
n
o
th
er
o
p
ti
o
n
av
ai
lab
le,
th
at
is
v
ar
io
u
s
f
ac
u
lt
y
o
r
u
s
e
r
s
w
il
l
g
iv
e
a
n
s
w
er
to
th
e
q
u
est
io
n
i
n
th
e
f
o
r
m
o
f
te
x
t,
i
m
a
g
e,
v
o
ice
an
d
v
id
eo
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8822
I
J
E
R
E
Vo
l.
5
,
No
.
3
,
Sep
tem
b
er
20
1
6
:
227
–
2
3
4
228
Qu
est
io
n
a
n
s
w
er
i
n
g
(
Q
A
)
is
a
s
y
s
te
m
f
o
r
a
u
to
m
at
icall
y
r
ep
ly
in
g
a
q
u
esti
o
n
t
h
at
is
p
o
s
ted
in
n
atu
r
a
l
lan
g
u
a
g
e.
C
o
m
p
ar
in
g
to
s
ea
r
ch
s
y
s
te
m
s
b
ased
o
n
k
e
y
w
o
r
d
s
,
it
ex
tr
e
m
el
y
f
ac
ilit
ate
s
t
h
e
co
m
m
u
n
icatio
n
a
m
o
n
g
co
m
p
u
ter
s
y
s
te
m
s
a
n
d
h
u
m
a
n
s
.
I
t
also
a
v
o
id
s
th
e
ac
h
in
g
j
o
b
o
f
b
r
o
w
s
in
g
t
h
e
v
er
y
en
o
r
m
o
u
s
a
m
o
u
n
t
o
f
ed
u
ca
tio
n
al
co
n
ten
t
w
h
ic
h
is
r
etu
r
n
ed
as
e
x
ac
t
an
s
w
er
s
b
y
s
ea
r
ch
e
n
g
i
n
es.
Ho
w
e
v
er
,
f
u
ll
y
co
m
p
u
ter
ized
QA
is
s
till
f
ac
i
n
g
c
h
alle
n
g
es
w
h
ic
h
ar
e
n
o
t
s
i
m
p
le
to
h
a
n
d
le,
s
u
c
h
as
k
ee
n
k
i
n
d
o
f
co
m
p
l
ex
q
u
es
tio
n
s
an
d
t
h
e
s
o
p
h
is
tic
ated
s
y
n
tax
,
s
e
m
a
n
ti
cs
an
d
co
n
tex
t
u
al
p
r
o
ce
s
s
in
g
to
attain
p
r
ed
ict
ab
le
an
s
w
er
s
.
I
t
is
ex
p
er
im
e
n
ted
th
at,
m
o
s
t
l
y
au
to
m
ated
ap
p
r
o
ac
h
is
n
o
t
s
k
illed
o
f
o
b
tain
in
g
t
h
e
co
n
s
eq
u
e
n
ce
s
t
h
at
ar
e
as
g
o
o
d
as
th
o
s
e
g
en
er
ated
b
y
h
u
m
a
n
in
telli
g
en
ce
.
T
h
e
DI
MQ
A
s
y
s
te
m
al
lo
w
s
t
h
e
u
s
er
s
to
a
n
s
w
er
i
n
m
u
lti
m
ed
ia
f
ea
t
u
r
e
alo
n
g
w
it
h
te
x
t.
So
m
eti
m
es,
T
ex
tu
al
an
s
w
er
s
m
a
y
n
o
t p
r
o
v
id
e
s
u
f
f
i
cien
t a
n
d
ea
s
il
y
u
n
d
er
s
tan
d
ab
l
e
in
f
o
r
m
atio
n
.
T
h
e
s
y
s
te
m
ap
p
r
o
ac
h
es
w
o
u
ld
h
e
lp
p
r
o
v
id
e
an
s
w
er
to
u
s
e
r
s
i
n
f
o
r
m
o
f
m
u
lti
m
ed
ia
.
B
ec
au
s
e
p
ictu
r
e
s
p
ea
k
s
a
th
o
u
s
an
d
s
o
f
w
o
r
d
s
,
w
e
ar
e
s
i
g
n
i
f
y
i
n
g
a
b
asic
id
ea
f
r
o
m
t
h
i
s
s
y
s
te
m
th
at
n
o
t
o
n
l
y
co
n
cise
tex
tu
a
l
in
f
o
r
m
atio
n
b
u
t
al
s
o
o
th
er
a
u
d
io
,
v
id
eo
an
d
i
m
ag
e
i
n
f
o
r
m
atio
n
ca
n
b
e
tea
m
ed
u
p
w
i
th
t
h
is
tex
t
u
al
a
n
s
w
er
to
b
etter
e
m
p
h
a
s
ize
it a
n
d
t
h
u
s
p
r
o
v
id
e
a
b
etter
ex
p
er
ien
ce
to
th
e
all
u
s
er
s
.
T
h
e
DI
MQ
A
s
y
s
te
m
o
b
tai
n
s
in
f
o
r
m
atio
n
o
n
li
n
e
a
lo
n
g
w
i
t
h
p
ar
ticu
lar
q
u
es
tio
n
o
n
s
o
m
e
to
p
ic
an
d
o
b
tain
s
e
x
ac
t
a
n
s
w
er
f
o
r
m
o
t
h
er
p
ar
ticip
an
t.
I
n
T
h
is
s
y
s
te
m
u
s
er
s
s
h
ar
e
th
e
ir
k
n
o
w
led
g
e
ac
co
r
d
in
g
to
t
h
eir
in
ter
est
w
h
ich
is
h
a
v
i
n
g
d
is
s
i
m
ilar
ca
te
g
o
r
ies
a
n
d
u
s
er
s
ca
n
s
ea
r
c
h
f
o
r
an
s
w
er
to
q
u
e
s
ti
o
n
f
r
o
m
t
h
e
m
.
T
h
e
r
esu
lt
o
b
tain
s
f
r
o
m
th
e
DI
MQ
A
s
y
s
te
m
f
o
r
u
m
ar
e
i
m
p
r
o
v
ed
an
s
w
er
b
ec
au
s
e
th
at
a
n
s
w
er
g
en
er
ated
b
y
h
u
m
a
n
clev
er
n
e
s
s
.
O
v
er
th
e
y
ea
r
,
th
eir
h
u
g
e
a
m
o
u
n
t
s
o
f
an
s
w
er
an
d
q
u
esti
o
n
h
a
v
e
b
e
en
ac
cu
m
u
lated
to
o
f
f
er
th
e
f
ac
ilit
y
li
k
e
co
n
s
er
v
atio
n
an
d
s
ea
r
ch
o
f
a
n
s
w
e
r
ed
q
u
e
s
tio
n
.
T
h
e
p
r
o
b
lem
w
i
th
e
x
is
tin
g
s
y
s
te
m
is
t
h
at
t
h
e
y
s
u
p
p
o
r
t
o
n
l
y
te
x
tu
a
l
an
s
w
er
an
d
w
h
ic
h
is
n
o
t
r
elev
an
t
f
o
r
m
a
n
y
ti
m
es,
i
f
w
e
ad
d
ass
o
ciate
d
m
u
lti
m
ed
ia
co
n
ten
t
s
u
c
h
as
i
m
a
g
e
o
r
v
id
eo
an
d
au
d
io
to
s
h
o
w
t
h
e
p
r
o
ce
s
s
w
h
ich
p
r
o
v
id
e
b
etter
r
esu
lt.
T
h
e
o
b
tain
ab
le
s
y
s
te
m
o
f
co
m
m
u
n
i
t
y
b
ased
q
u
esti
o
n
an
s
w
er
in
g
s
y
s
te
m
s
u
c
h
as
s
tac
k
o
v
er
f
lo
w
,
y
a
h
o
o
an
s
w
er
,
w
i
k
i
a
n
s
w
e
r
an
d
ask
.
co
m
p
r
o
v
id
e
an
s
w
er
o
n
l
y
in
te
x
t
u
al
f
o
r
m
b
u
t
a
f
e
w
q
u
esti
o
n
s
u
c
h
as
Ho
w
to
in
s
tal
l
W
in
d
o
w
s
OS?
.
I
n
th
is
ca
s
e,
if
p
r
o
v
id
e
an
s
w
er
i
n
te
x
t
u
al
f
o
r
m
w
h
ich
is
n
o
t
i
n
f
o
r
m
ati
v
e
f
o
r
m
a
n
y
u
s
er
.
Ass
o
ciate
d
v
id
eo
o
r
i
m
a
g
es
p
r
o
v
id
e
b
etter
r
esu
lt,
in
f
ac
t
s
o
m
e
co
m
m
u
n
it
y
f
o
r
u
m
p
r
o
v
id
e
a
b
alan
cin
g
li
n
k
t
o
d
em
o
n
s
tr
ate
th
e
p
r
o
ce
s
s
.
I
t c
o
n
f
ir
m
s
th
at
m
u
lti
m
ed
ia
co
n
te
n
t a
r
e
i
m
p
o
r
tan
t to
s
h
o
w
th
e
p
r
o
ce
s
s
.
2.
L
I
T
E
R
AT
U
RE
SU
RVE
Y
2
.
1
.
F
ro
m
T
e
x
t
ua
l
Q
A
t
o
M
ultim
edia
Q
A
T
h
e
ea
r
ly
ex
a
m
i
n
atio
n
o
f
Q
A
s
y
s
te
m
s
s
tar
ted
f
r
o
m
1
9
6
1
an
d
m
a
in
l
y
f
o
cu
s
ed
o
n
s
k
illed
s
y
s
te
m
s
i
n
s
p
ec
if
ic
d
o
m
a
in
s
.
T
ex
t
b
ased
QA
h
a
s
g
a
in
ed
it
s
r
esear
c
h
r
ep
u
tatio
n
s
i
n
ce
th
e
o
r
g
an
izatio
n
o
f
a
Q
A
tr
ac
k
i
n
T
R
E
C
in
th
e
late
1
9
9
0
s
[
1
]
.
B
ased
o
n
th
e
k
in
d
o
f
q
u
est
io
n
s
an
d
p
r
ed
ictab
le
an
s
w
er
s
,
w
e
ca
n
r
o
u
g
h
l
y
s
u
m
m
ar
ize
th
e
s
o
r
ts
o
f
Q
A
in
to
Op
en
-
Do
m
ai
n
Q
A
[
2
]
,
R
es
tr
icted
-
Do
m
ai
n
Q
A
[
2
]
,
Def
i
n
itio
n
al
Q
A
[
3
]
an
d
L
is
t
Q
A
[
4
]
.
On
th
e
o
th
er
h
a
n
d
,
in
s
p
ite
o
f
th
e
attai
n
m
en
t
as
d
escr
ib
ed
ab
o
v
e,
a
u
to
m
atic
Q
A
s
till
h
a
s
s
o
m
e
d
if
f
ic
u
lt
ies
i
n
an
s
w
er
i
n
g
co
m
p
o
s
ite
q
u
esti
o
n
s
.
A
l
l
alo
n
g
w
it
h
th
e
b
lo
o
m
in
g
o
f
W
eb
2
.
0
,
C
o
m
m
u
n
it
y
q
u
es
tio
n
an
s
w
er
s
b
ec
o
m
es
a
n
al
ter
n
ati
v
e
ap
p
r
o
ac
h
.
I
t
is
a
h
u
g
e
a
n
d
v
ar
io
u
s
q
u
e
s
tio
n
-
a
n
s
w
er
d
i
s
cu
s
s
io
n
s
,
ac
tin
g
a
s
n
o
t
o
n
l
y
a
q
u
a
n
tit
y
f
o
r
s
h
ar
in
g
tec
h
n
ica
l
k
n
o
w
led
g
e
b
u
t
al
s
o
a
p
lace
w
h
er
e
o
n
e
ca
n
s
ee
k
ad
v
ic
e
an
d
o
p
in
io
n
s
[
3
]
,
[
5
]
.
Sti
ll,
n
ea
r
l
y
all
o
f
t
h
e
o
b
tain
ab
le
cQ
A
(
C
o
m
m
u
n
it
y
q
u
e
s
tio
n
a
n
s
w
er
s
)
s
y
s
te
m
s
,
s
u
ch
as
Ya
h
o
o
!
An
s
w
er
s
,
W
ik
i
A
n
s
w
er
s
a
n
d
A
s
k
.
co
m
,
Stack
o
v
er
f
lo
w
,
o
n
l
y
s
u
p
p
o
r
t
p
u
r
e
tex
t
-
b
ased
an
s
w
er
s
,
wh
ich
m
a
y
n
o
t
g
i
v
e
in
t
u
iti
v
e
an
d
en
o
u
g
h
i
n
f
o
r
m
ati
o
n
.
So
m
e
ex
a
m
i
n
e
e
f
f
o
r
ts
h
av
e
b
ee
n
p
u
t
o
n
m
u
lti
m
ed
ia
QA
,
w
h
ic
h
is
a
n
s
w
er
q
u
e
s
t
io
n
s
u
s
i
n
g
m
u
lti
m
ed
ia
d
ata.
C
h
u
a
et
al.
[
6
]
p
r
o
j
ec
ted
a
co
m
p
r
eh
en
s
iv
e
ap
p
r
o
ac
h
to
ex
ten
d
tex
t
-
b
ased
QA
to
m
u
lt
i
m
ed
i
a
QA
f
o
r
a
r
an
g
e
o
f
f
ac
to
id
,
d
ef
in
itio
n
an
d
“
h
o
w
-
to
”
q
u
e
s
tio
n
s
.
T
h
eir
s
y
s
te
m
w
a
s
p
r
ep
ar
ed
to
f
in
d
m
u
lti
m
ed
ia
an
s
w
er
s
f
r
o
m
w
eb
-
s
ca
le
m
e
d
ia
r
eso
u
r
ce
s
s
u
ch
a
s
Fli
c
k
er
an
d
Yo
u
T
u
b
e.
Ho
w
ev
er
,
ar
ticle
r
eg
ar
d
in
g
m
u
lti
m
ed
ia
Q
A
is
s
till
m
o
d
er
atel
y
th
in
.
A
u
to
m
atic
m
u
l
ti
m
ed
ia
QA
o
n
l
y
w
o
r
k
s
in
s
p
ec
i
f
ic
d
o
m
ai
n
s
an
d
ca
n
b
ar
ely
h
a
n
d
le
m
u
lti
f
ac
eted
q
u
esti
o
n
s
.
Dif
f
er
e
n
t
f
r
o
m
t
h
ese
w
o
r
k
s
,
o
u
r
ap
p
r
o
ac
h
is
b
u
il
t
b
ased
o
n
c
Q
A
.
A
s
a
n
alter
n
ati
v
e
o
f
d
ir
ec
tl
y
co
llect
in
g
m
u
l
ti
m
ed
ia
f
iles
f
o
r
a
n
s
w
er
i
n
g
q
u
es
tio
n
s
,
o
u
r
m
et
h
o
d
o
n
l
y
f
i
n
d
s
i
m
a
g
e,
au
d
io
an
d
v
id
eo
to
en
r
ich
th
e
tex
t
u
al
an
s
w
er
s
p
r
o
v
id
ed
b
y
u
s
er
s
.
I
t
m
a
k
es
o
u
r
ap
p
r
o
ac
h
ca
p
ab
le
to
d
ea
l
w
it
h
m
o
r
e
co
m
m
o
n
q
u
est
io
n
s
a
n
d
t
o
r
ea
ch
b
etter
p
er
f
o
r
m
an
ce
.
2
.
2
.
M
ultim
e
dia
Sea
rc
h
A
p
p
r
o
p
r
iate
to
th
e
r
is
i
n
g
q
u
an
tit
y
o
f
d
i
g
ital
in
f
o
r
m
at
io
n
s
to
r
ed
o
v
er
th
e
w
eb
,
p
en
et
r
atin
g
f
o
r
p
r
ef
er
r
ed
in
f
o
r
m
atio
n
h
a
s
b
ec
o
m
e
a
n
e
s
s
e
n
tial
ta
s
k
.
T
h
e
r
esear
ch
i
n
th
i
s
ar
ea
s
tar
ted
f
r
o
m
t
h
e
1
9
8
0
s
[
7
]
b
y
ad
d
r
ess
in
g
th
e
co
m
m
o
n
p
r
o
b
le
m
o
f
d
ec
is
io
n
i
m
a
g
es
f
r
o
m
a
f
i
x
ed
d
atab
ase.
W
ith
th
e
q
u
ick
d
ev
elo
p
m
e
n
t
o
f
co
n
ten
t
a
n
al
y
s
i
s
tec
h
n
o
lo
g
y
i
n
th
e
1
9
9
0
s
,
th
e
s
e
ef
f
o
r
ts
r
a
p
id
ly
e
x
p
an
d
ed
to
atte
m
p
t
th
e
v
id
eo
an
d
au
d
io
r
etr
iev
al
p
r
o
b
lem
s
[
7
]
,
[
8
].
I
n
g
en
er
al,
m
u
lti
m
ed
ia
s
ee
k
s
e
f
f
o
r
ts
ca
n
b
e
class
i
f
ied
i
n
to
t
wo
ca
teg
o
r
ies:
tex
t
-
b
ased
s
ea
r
ch
a
n
d
co
n
te
n
t
-
b
as
ed
s
ea
r
ch
.
T
h
e
tex
t
-
b
ased
s
ea
r
ch
[
9
]
ap
p
r
o
ac
h
es
w
o
r
k
s
w
it
h
te
x
tu
a
l
q
u
er
ies,
a
ter
m
-
b
a
s
ed
r
eq
u
ir
e
m
e
n
t
o
f
t
h
e
d
esire
d
m
ed
ia
e
n
titi
e
s
,
to
s
ea
r
ch
f
o
r
m
ed
ia
d
ata
b
y
m
atc
h
in
g
th
e
m
w
it
h
t
h
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
R
E
I
SS
N:
2252
-
8822
Do
ma
in
a
n
d
I
n
tellig
en
ce
B
a
s
e
d
Mu
ltimed
ia
Qu
esti
o
n
A
n
s
w
erin
g
S
ystem
(
K
.
Ma
g
esh
K
u
ma
r
)
229
n
eig
h
b
o
r
in
g
te
x
t
u
al
d
escr
ip
tio
n
s
.
T
o
i
m
p
r
o
v
e
t
h
e
p
er
f
o
r
m
a
n
ce
o
f
tex
t
-
b
ased
s
ea
r
ch
,
s
o
m
e
m
ac
h
i
n
e
lear
n
i
n
g
tech
n
iq
u
es
t
h
at
ai
m
to
m
ec
h
an
ica
ll
y
a
n
n
o
tate
m
ed
i
u
m
en
titi
e
s
h
a
v
e
b
ee
n
p
r
o
p
o
s
e
d
in
th
e
m
u
lti
m
ed
ia
co
m
m
u
n
it
y
[
5
]
,
[
1
0
],
[1
1]
.
A
d
d
itio
n
al
ly
,
a
n
u
m
b
er
o
f
s
o
cial
m
ed
ia
w
eb
s
ite
s
,
s
u
c
h
as
Fli
c
k
r
an
d
Face
b
o
o
k
,
h
av
e
e
m
er
g
ed
to
b
u
ild
u
p
m
a
n
u
all
y
a
n
n
o
tated
m
ed
i
u
m
e
n
tit
ies
b
y
e
x
p
lo
r
in
g
t
h
e
g
r
ass
r
o
o
t
I
n
ter
n
et
u
s
er
s
,
w
h
ich
also
f
ac
ilit
a
te
th
e
te
x
t
-
b
ased
s
ea
r
ch
.
C
o
n
v
er
s
el
y
,
u
s
er
-
p
r
o
v
i
d
ed
tex
t
d
ef
in
i
tio
n
f
o
r
m
ed
ia
d
a
ta
ar
e
o
f
ten
b
iased
to
w
ar
d
s
i
n
d
iv
id
u
al
p
er
s
p
ec
tiv
e
an
d
co
n
te
x
t
cu
e
s
,
an
d
t
h
u
s
t
h
er
e
is
a
b
r
ea
k
b
et
w
ee
n
t
h
ese
tag
s
a
n
d
t
h
e
co
n
te
n
t
o
f
t
h
e
m
ed
iu
m
en
ti
ties
th
a
t
co
m
m
o
n
u
s
er
s
ar
e
i
n
ter
ested
in
.
T
o
atte
m
p
t
t
h
is
is
s
u
e,
c
o
n
ten
t
-
b
ased
m
ed
ia
r
etr
iev
al
[
6
]
p
er
f
o
r
m
s
ex
p
lo
r
atio
n
b
y
a
n
al
y
zi
n
g
th
e
co
n
te
n
ts
o
f
m
ed
iu
m
d
ata
r
ath
er
th
an
t
h
e
m
etad
ata.
Desp
ite
th
e
m
ar
v
elo
u
s
i
m
p
r
o
v
e
m
en
t
in
co
n
te
n
t
-
b
ased
r
etr
iev
al,
s
till
it
h
a
s
s
e
v
er
al
li
m
itati
o
n
s
,
s
u
ch
a
s
h
ig
h
co
m
p
u
tatio
n
al
co
s
t,
tr
o
u
b
le
in
f
in
d
i
n
g
v
i
s
u
a
l
q
u
er
ies,
a
n
d
th
e
lar
g
e
b
r
ea
k
b
et
w
ee
n
lo
w
-
le
v
el
v
i
s
u
al
d
escr
ip
tio
n
s
an
d
u
s
er
’
s
s
e
m
a
n
tic
an
tic
ip
atio
n
.
As
a
r
esu
lt,
k
e
y
w
o
r
d
-
b
ased
s
ea
r
ch
en
g
i
n
es
ar
e
s
till
b
r
o
ad
ly
u
s
ed
f
o
r
m
ed
ia
e
x
p
lo
r
atio
n
.
Ho
w
e
v
er
,
th
e
in
h
er
e
n
t
li
m
ita
tio
n
o
f
tex
t
-
b
ased
ap
p
r
o
ac
h
es
b
u
ild
th
at
all
t
h
e
p
r
esen
t
co
m
m
er
cial
m
ed
ia
s
ea
r
ch
en
g
in
e
s
tr
ick
y
to
lin
k
th
e
g
ap
b
et
w
ee
n
tex
t
u
al
q
u
er
ie
s
an
d
m
u
lti
m
ed
ia
d
ata,
p
ar
ticu
lar
l
y
f
o
r
w
o
r
d
y
q
u
es
tio
n
s
i
n
n
at
u
r
al
la
n
g
u
a
g
es.
2
.
3
.
M
ultim
e
dia
Sea
rc
h Re
-
ra
n
k
i
ng
As
b
ef
o
r
e
m
en
tio
n
ed
,
p
r
esen
t
m
ed
ia
s
ea
r
c
h
e
n
g
i
n
es
ar
e
t
y
p
icall
y
b
u
i
lt
u
p
o
n
th
e
tex
t
i
n
f
o
r
m
atio
n
lin
k
ed
w
it
h
m
u
lt
i
m
ed
ia
e
n
tit
ie
s
,
s
u
c
h
a
s
,
AL
T
tex
ts
,
a
n
d
s
u
r
r
o
u
n
d
i
n
g
te
x
t
s
o
n
m
u
ltip
le
w
eb
p
ag
e.
B
u
t
t
h
e
tex
t
in
f
o
r
m
atio
n
t
y
p
icall
y
d
o
es
n
o
t
ex
ac
tl
y
e
x
p
r
ess
t
h
e
co
n
te
n
t
o
f
th
e
i
m
a
g
es
a
n
d
v
id
eo
s
,
an
d
th
is
in
f
o
r
m
atio
n
ca
n
cr
u
ell
y
d
e
g
r
ad
e
s
ea
r
ch
r
o
u
tin
e
[
1
2
]
.
R
e
-
r
an
k
i
n
g
i
s
a
tech
n
iq
u
e
th
at
i
m
p
r
o
v
e
s
s
ee
k
s
s
i
g
n
i
f
ic
an
ce
b
y
m
in
i
n
g
th
e
v
is
u
al
i
n
f
o
r
m
atio
n
o
f
i
m
a
g
e
s
an
d
v
id
eo
s
.
Ob
tain
ab
le
r
e
-
r
a
n
k
in
g
al
g
o
r
ith
m
s
ca
n
m
o
s
tl
y
b
e
ca
teg
o
r
ized
in
to
t
w
o
m
eth
o
d
s
,
o
n
e
i
s
p
s
eu
d
o
r
elev
an
ce
f
ee
d
b
ac
k
a
n
d
th
e
o
t
h
e
r
is
g
r
ap
h
-
b
ased
r
e
-
r
an
k
i
n
g
.
T
h
e
p
s
eu
d
o
r
elev
an
ce
f
ee
d
b
ac
k
ap
p
r
o
ac
h
[
9
]
,
[
1
1
]
,
[
1
3
]
r
eg
ar
d
s
to
p
co
n
s
eq
u
en
ce
s
as
ap
p
licab
le
s
a
m
p
les
an
d
t
h
e
n
it
co
llect
s
s
o
m
e
s
a
m
p
les
t
h
at
ar
e
u
n
s
p
ec
i
f
ied
to
b
e
ir
r
elev
a
n
t.
A
ca
te
g
o
r
izatio
n
o
r
r
an
k
i
n
g
m
o
d
el
is
ed
u
ca
ted
b
ased
o
n
t
h
e
p
s
eu
d
o
ap
p
licab
le
an
d
i
m
m
a
ter
ial
s
a
m
p
le
s
an
d
t
h
e
r
ep
r
ese
n
tatio
n
is
t
h
e
n
u
s
e
d
to
r
e
-
r
an
k
t
h
e
o
r
ig
i
n
al
s
ee
k
r
e
s
u
lt
s
.
I
t
i
s
i
n
d
is
t
in
g
u
is
h
ed
to
r
elev
an
ce
f
ee
d
b
ac
k
w
h
er
e
u
s
er
s
clea
r
l
y
p
r
o
v
id
e
o
p
in
io
n
b
y
ca
talo
g
i
n
g
th
e
r
es
u
lts
as r
elev
a
n
t o
r
ir
r
elev
an
t.
T
h
e
g
r
ap
h
-
b
ased
r
e
-
r
an
k
i
n
g
a
p
p
r
o
ac
h
[
1
2
],
[1
4
]
-
[1
6
]
r
eg
u
lar
l
y
f
o
llo
w
s
t
w
o
as
s
u
m
p
ti
o
n
s
.
First,
th
e
d
is
ag
r
ee
m
e
n
t
b
et
w
ee
n
t
h
e
f
ir
s
t
r
an
k
i
n
g
lis
t
an
d
th
e
r
ef
i
n
ed
r
an
k
i
n
g
lis
t
s
h
o
u
ld
b
e
s
m
all.
S
ec
o
n
d
,
th
e
r
an
k
i
n
g
p
o
s
itio
n
s
o
f
v
is
u
all
y
r
elate
d
s
a
m
p
les
s
h
o
u
ld
b
e
clo
s
e.
Usu
all
y
,
t
h
is
ap
p
r
o
ac
h
co
n
s
tr
u
cts
a
g
r
ap
h
w
h
er
e
t
h
e
v
er
tices
ar
e
i
m
a
g
es
o
r
v
id
eo
s
an
d
th
e
ed
g
es
i
m
itate
t
h
eir
p
air
-
w
i
s
e
s
i
m
ilar
itie
s
.
A
g
r
a
p
h
-
b
ased
lear
n
i
n
g
p
r
o
ce
s
s
is
th
e
n
f
o
r
m
u
lated
b
as
ed
o
n
a
r
eg
u
lar
izatio
n
s
tr
u
ct
u
r
e.
B
o
th
o
f
th
e
t
w
o
ap
p
r
o
ac
h
es
r
el
y
o
n
th
e
v
i
s
u
al
s
i
m
ilar
itie
s
b
et
w
ee
n
m
ed
i
u
m
e
n
titi
e
s
.
C
o
n
s
er
v
ativ
e
m
et
h
o
d
s
u
s
u
a
ll
y
ca
lcu
late
th
e
s
i
m
ilar
i
ties
b
ased
o
n
a
f
i
x
ed
s
et
o
f
f
ea
t
u
r
es
ex
tr
ac
ted
f
r
o
m
m
ed
i
u
m
e
n
ti
ties
,
s
u
c
h
as
co
lo
r
,
tex
t
u
r
e,
s
h
ap
e
an
d
b
ag
-
of
-
v
i
s
u
al
w
o
r
d
s
.
Ho
w
e
v
er
,
th
e
r
e
s
e
m
b
la
n
ce
est
i
m
atio
n
ac
tu
a
ll
y
s
h
o
u
ld
b
e
q
u
esti
o
n
ad
ap
tiv
e.
Fo
r
ex
a
m
p
le,
if
w
e
w
a
n
t
to
f
in
d
a
p
er
s
o
n
,
w
e
s
h
o
u
ld
ca
lcu
late
th
e
s
i
m
ilar
itie
s
o
f
f
ac
ial
s
k
i
n
tex
tu
r
e
i
n
s
tead
o
f
th
e
f
ea
t
u
r
es
ex
tr
ac
ted
f
r
o
m
t
h
e
w
h
o
le
i
m
a
g
es
[
1
7
]
.
I
t
is
s
en
s
ib
le
as
i
n
f
o
r
m
atio
n
s
ee
k
er
s
ar
e
f
u
t
u
r
e
to
f
i
n
d
a
p
er
s
o
n
r
ath
er
th
an
o
th
er
o
b
j
ec
ts
.
3.
P
RO
P
O
SE
D
SYS
T
E
M
T
h
e
d
o
m
ai
n
a
n
d
i
n
tell
ig
e
n
ce
b
ased
Mu
lti
m
ed
ia
Q
A
s
y
s
te
m
h
a
v
i
n
g
th
e
f
o
llo
w
in
g
f
ea
t
u
r
es
;
th
e
s
e
ar
e
Sear
ch
a
n
d
P
o
s
t
Qu
e
s
tio
n
s
,
Do
cu
m
e
n
t
R
etr
ie
v
al,
An
s
w
er
s
E
x
tr
ac
tio
n
,
An
s
w
er
s
E
v
al
u
atio
n
s
,
An
s
w
er
in
g
Mo
d
e
an
d
R
an
k
in
g
.
T
h
e
m
u
lt
i
m
ed
ia
Q
A
s
y
s
te
m
c
o
n
s
o
les
a
n
d
h
elp
s
th
e
s
t
u
d
en
t
s
an
d
p
r
o
f
ess
o
r
s
b
y
p
r
o
v
id
in
g
t
h
eir
n
ee
d
s
.
I
n
th
is
s
y
s
te
m
s
t
u
d
en
t
s
a
n
d
p
r
o
f
ess
o
r
s
ar
e
co
n
s
id
er
as
u
s
er
s
.
I
f
u
s
er
s
r
eq
u
ir
e
a
n
s
w
er
f
o
r
an
y
q
u
e
s
tio
n
,
t
h
e
y
ca
n
s
ee
k
t
h
e
an
s
w
er
in
Q
A
s
y
s
te
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8822
I
J
E
R
E
Vo
l.
5
,
No
.
3
,
Sep
tem
b
er
20
1
6
:
227
–
2
3
4
230
Fig
u
r
e
1
.
A
r
ch
itectu
r
e
o
f
s
t
u
d
en
t q
u
es
tio
n
a
n
s
w
er
i
n
g
s
y
s
te
m
I
f
t
h
e
an
s
w
er
is
alr
ea
d
y
i
n
t
h
e
d
atab
ase
th
e
u
s
er
s
ca
n
r
etr
ie
v
e
th
e
d
o
cu
m
e
n
ts
a
n
d
e
x
tr
ac
t
t
h
e
an
s
w
er
f
r
o
m
w
h
ic
h
th
e
y
ca
n
e
v
al
u
ate
it.
A
t
last
th
e
u
s
er
s
ca
n
v
ie
w
an
d
u
tili
ze
t
h
e
a
n
s
w
er
.
I
f
i
n
ca
s
e
n
o
a
n
s
w
er
is
f
o
u
n
d
in
d
atab
ase
o
th
er
u
s
er
c
an
p
o
s
t
th
e
a
n
s
w
er
t
o
t
h
e
ir
q
u
esti
o
n
i
n
t
h
e
f
o
r
m
o
f
te
x
t,
i
m
a
g
e,
v
o
ice
an
d
v
id
eo
.
T
h
e
b
est
an
s
w
er
s
ar
e
r
a
n
k
ed
b
y
r
an
k
i
n
g
m
et
h
o
d
o
lo
g
y
.
At
l
ast
t
h
e
u
s
er
e
v
al
u
ates
a
n
d
v
ie
w
s
t
h
e
b
est
an
s
w
er
s
.
T
h
e
f
o
llo
w
i
n
g
f
ac
to
r
s
ar
e
u
s
e
d
to
d
ev
elo
p
a
DI
MQ
A
(
Do
m
ain
an
d
I
n
te
llig
e
n
ce
b
ased
M
u
lti
m
ed
ia
Qu
e
s
tio
n
An
s
w
er
i
n
g
)
s
y
s
te
m
.
T
h
er
e
ar
e
s
ev
er
al
a
n
s
w
er
in
g
m
o
d
e
s
a
v
ailab
le
to
u
s
er
s
,
e
s
p
ec
iall
y
t
h
is
s
y
s
te
m
s
u
p
p
o
r
t
m
u
lti
m
ed
ia
f
ea
tu
r
es
in
An
s
w
e
r
in
g
m
o
d
e.
T
ex
tu
al
a
n
s
w
er
s
m
a
y
n
o
t
al
w
a
y
s
o
f
f
er
s
u
f
f
ic
ien
t
n
atu
r
al
a
n
d
s
i
m
p
l
y
ac
ce
p
t
ab
le
in
f
o
r
m
at
io
n
.
T
h
e
s
y
s
t
e
m
’
s
ap
p
r
o
ac
h
w
o
u
ld
h
e
lp
g
iv
e
a
n
s
w
er
g
ai
n
er
s
m
o
r
e
co
n
cise,
co
m
p
r
eh
e
n
s
i
v
e
in
f
o
r
m
atio
n
a
n
d
en
h
an
ce
d
e
x
p
er
ien
ce
.
As
i
m
a
g
e
s
p
ea
k
s
a
th
o
u
s
a
n
d
s
o
f
w
o
r
d
s
,
th
is
s
y
s
te
m
t
h
at
n
o
t
o
n
l
y
co
n
cise
te
x
t
u
al
in
f
o
r
m
atio
n
b
u
t
also
o
th
er
m
u
lti
m
ed
ia
in
f
o
r
m
atio
n
ca
n
b
e
tea
m
ed
u
p
w
it
h
t
h
is
te
x
t
u
al
an
s
w
er
to
b
etter
h
ig
h
l
ig
h
t
it
a
n
d
th
u
s
p
r
o
v
id
e
a
b
etter
ex
p
er
ien
ce
to
th
e
co
m
m
o
n
u
s
er
s
.
A
n
s
w
e
r
in
g
m
o
d
e
is
h
elp
to
d
eter
m
in
e
w
h
ich
t
y
p
e
o
f
m
ed
iu
m
is
r
eq
u
ir
e
d
to
i
m
p
r
o
v
e
t
h
e
tex
t
u
al
an
s
w
er
.
Fo
r
e
x
a
m
p
le,
“
W
h
at
is
m
ea
n
b
y
R
i
n
g
to
p
o
lo
g
y
?
”
t
h
is
q
u
esti
o
n
o
n
l
y
n
ee
d
s
p
u
r
e
te
x
t
u
al
a
n
s
w
er
s
.
B
u
t
s
o
m
e
q
u
e
s
tio
n
s
m
a
y
b
e
l
ik
e,
“Ho
w
to
co
n
n
ec
t
th
e
s
y
s
te
m
s
i
n
to
R
in
g
T
o
p
o
lo
g
y
?
”
p
r
o
v
id
e
th
e
tex
t
u
al
a
n
s
w
er
w
it
h
a
n
i
m
a
g
e
o
f
R
in
g
to
p
o
lo
g
y
,
it
w
i
ll
b
e
m
o
r
e
i
n
f
o
r
m
ati
v
e.
So
m
eti
m
es
t
h
e
q
u
e
s
tio
n
s
m
a
y
b
e
lik
e
th
i
s
,
”
Ho
w
th
e
s
y
s
te
m
s
co
m
m
u
n
icate
i
n
r
in
g
to
p
o
lo
g
y
?
”
T
h
e
an
s
w
er
is
ex
p
lain
ed
w
i
th
a
v
id
eo
th
a
t s
h
o
w
s
h
o
w
th
e
s
y
s
te
m
co
m
m
u
n
i
ca
tes
w
it
h
ea
c
h
o
t
h
er
,
a
n
d
th
e
n
it
w
ill
b
e
ea
s
ier
to
u
n
d
er
s
ta
n
d
.
So
ea
ch
q
u
e
s
tio
n
n
ee
d
s
d
if
f
er
en
t
m
ed
i
u
m
to
i
m
p
r
o
v
e
th
e
tex
t
u
al
d
ata.
B
as
ed
o
n
th
i
s
an
al
y
s
is
we
ca
n
class
if
y
t
h
e
an
s
w
er
s
b
ased
o
n
th
e
m
ed
i
u
m
as
:
a.
T
ex
t
b
.
T
ex
t +
I
m
a
g
e
c.
T
ex
t +
Vid
eo
d
.
T
ex
t +
I
m
a
g
e
+V
id
eo
F
ig
u
r
e
2
.
Mo
d
e
o
f
A
n
s
w
er
in
g
an
d
P
r
o
ce
s
s
in
g
a.
T
ex
t: it
m
ea
n
s
th
a
t u
n
iq
u
e
te
x
t
u
al
an
s
w
er
s
ar
e
en
o
u
g
h
.
b.
T
ex
t+
i
m
ag
e
:
it
m
ea
n
s
t
h
at
te
x
tu
a
l
i
n
f
o
r
m
atio
n
is
n
o
t
e
n
o
u
g
h
to
u
s
er
s
o
i
m
ag
e
in
f
o
r
m
atio
n
m
u
s
t
b
e
ad
d
ed
.
c.
T
ex
t
+v
id
eo
: it
m
ea
n
s
t
h
at
tex
t
u
al
in
f
o
r
m
a
tio
n
a
n
d
v
id
eo
in
f
o
r
m
atio
n
m
u
s
t b
e
ad
d
ed
.
d.
T
ex
t+i
m
ag
e+
v
id
eo
:
it
m
ea
n
s
th
at
w
e
ad
d
b
o
th
i
m
a
g
e
an
d
v
id
eo
i
n
f
o
r
m
atio
n
alo
n
g
w
it
h
tex
t
u
a
l
in
f
o
r
m
atio
n
.
T
h
e
ab
o
v
e
an
s
w
er
i
n
g
m
o
d
es
ar
e
u
s
ed
to
g
i
v
e
b
r
ief
a
n
s
w
er
s
to
th
e
f
ac
u
lt
y
a
n
d
s
t
u
d
en
t
to
g
ain
m
o
r
e
k
n
o
w
led
g
e
th
e
v
id
eo
an
s
w
er
s
ar
e
ex
p
lain
ed
in
d
etai
l
m
a
n
n
er
.
W
h
en
c
o
m
p
ar
in
g
w
it
h
t
h
e
tex
t
u
al
an
s
w
er
i
n
g
m
o
d
e
t
h
e
m
u
lt
i
m
ed
ia
a
n
s
w
er
i
n
g
m
o
d
es p
r
o
v
id
e
th
e
es
s
en
tia
l in
f
o
r
m
atio
n
to
t
h
e
u
s
er
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
R
E
I
SS
N:
2252
-
8822
Do
ma
in
a
n
d
I
n
tellig
en
ce
B
a
s
e
d
Mu
ltimed
ia
Qu
esti
o
n
A
n
s
w
erin
g
S
ystem
(
K
.
Ma
g
esh
K
u
ma
r
)
231
4.
M
E
T
H
O
D
4
.
1
.
Se
m
a
ntic
M
a
t
ch
Alg
o
rit
h
m
T
h
e
Sem
a
n
tic
m
atc
h
(
S
-
m
atc
h
)
alg
o
r
ith
m
is
tr
y
i
n
g
to
clo
s
e
th
e
g
ap
b
et
w
ee
n
u
s
er
co
m
m
a
n
d
an
d
th
e
n
ee
d
f
o
r
h
y
p
er
li
n
k
ac
ce
s
s
ib
il
it
y
.
T
h
e
DI
MQ
A
s
y
s
te
m
s
tar
ts
in
o
n
e
d
o
cu
m
en
t
a
n
d
th
e
n
m
o
v
e
th
r
o
u
g
h
an
u
n
e
n
d
in
g
s
e
t
o
f
d
o
cu
m
e
n
t
s
,
w
h
ic
h
co
n
n
ec
ted
b
y
to
p
ic
.
T
h
is
in
f
o
r
m
atio
n
n
et
w
o
r
k
in
g
is
b
as
ed
o
n
th
e
p
r
o
p
o
s
al
o
f
s
e
m
a
n
tic
ass
o
ciat
io
n
s
,
wh
er
e
o
n
e
u
n
it
(
n
o
d
e)
is
co
n
n
ec
ted
to
a
n
o
th
er
u
n
it
(
n
o
d
e)
b
y
m
ea
n
s
o
f
a
r
elatio
n
s
h
ip
(
an
ed
g
e)
.
Mo
s
t
s
ea
r
ch
e
n
g
in
e
s
r
etr
iev
e
in
f
o
r
m
atio
n
ac
cu
r
atel
y
b
y
ex
p
lo
iti
n
g
k
e
y
co
n
ten
t
o
f
ass
o
ciatio
n
s
in
Se
m
a
n
tic
r
eso
u
r
ce
s
,
o
r
r
elatio
n
s
.
T
h
e
Se
m
a
n
tic
b
ased
s
ea
r
ch
en
g
i
n
es
w
h
ich
r
el
y
o
n
in
f
o
r
m
atio
n
t
h
at
co
u
ld
b
e
ex
tr
ac
tin
g
f
r
o
m
u
s
er
q
u
er
y
a
n
d
th
e
o
n
to
lo
g
y
f
o
r
a
g
i
v
en
d
o
cu
m
en
t
.
T
h
e
id
ea
is
to
u
s
e
s
u
r
v
i
v
i
n
g
r
elatio
n
s
i
n
th
e
o
n
to
lo
g
y
n
a
m
e
s
“
v
ir
t
u
al
li
n
k
s
”
alo
n
g
w
it
h
ap
p
l
y
th
e
m
to
a
s
et
o
f
d
o
c
u
m
e
n
ts
to
in
cr
ea
s
e
th
e
p
r
o
b
ab
ilit
ies
o
f
f
i
n
d
in
g
t
h
e
in
h
er
en
t
as
s
o
ciatio
n
s
m
ad
e
b
y
th
e
u
s
er
at
t
h
e
ti
m
e
o
f
t
h
e
q
u
er
y
.
T
h
e
id
ea
o
f
ex
p
lo
itin
g
o
n
to
lo
g
y
-
b
ased
an
n
o
tatio
n
s
f
o
r
in
f
o
r
m
at
io
n
is
n
o
t
late
s
t;
s
e
m
a
n
tic
s
e
ar
ch
en
g
in
e
w
o
u
ld
co
n
s
id
er
k
e
y
w
o
r
d
co
n
ce
p
t
ass
o
ciatio
n
s
an
d
w
o
u
ld
r
etu
r
n
a
d
o
cu
m
e
n
t
o
n
l
y
i
f
k
e
y
w
o
r
d
s
(
o
r
s
y
n
o
n
y
m
s
,
h
o
m
o
n
y
m
s
,
etc.
)
A
r
e
f
o
u
n
d
with
i
n
t
h
e
d
o
cu
m
e
n
t
a
n
d
r
elate
to
ass
o
ciate
co
n
ce
p
ts
.
T
h
e
s
e
m
an
tic
al
g
o
r
ith
m
u
s
ed
to
p
r
o
d
u
ce
th
e
m
i
n
i
m
al
r
esu
lt
in
g
s
i
m
ilar
a
n
s
w
er
s
i
n
e
f
f
ec
tiv
e
Q
A
s
y
s
te
m
.
W
ith
in
n
at
u
r
al
lan
g
u
a
g
e
w
e
u
s
e
a
v
o
ca
b
u
lar
y
o
f
tin
y
e
x
p
r
ess
io
n
s
an
d
a
g
r
a
m
m
ar
to
b
u
ild
w
e
ll
-
f
o
r
m
ed
an
d
m
ea
n
i
n
g
f
u
l
ex
p
r
e
s
s
io
n
s
an
d
s
e
n
ten
ce
s
.
I
n
th
e
f
r
a
m
e
w
o
r
k
o
f
a
n
o
n
to
lo
g
y
lan
g
u
ag
e
t
h
e
v
o
ca
b
u
lar
y
is
ca
lled
s
ig
n
at
u
r
e.
I
t c
an
b
e
d
ef
i
n
ed
as f
o
llo
w
s
.
4
.
1
.
1
.
Def
ini
t
io
n o
f
Sig
na
t
ure
A
s
ig
n
at
u
r
e
K
is
a
q
u
ad
r
u
p
le
K
=
(
C
,
P
,
R
,
I
)
w
h
er
e
C
is
a
s
et
o
f
co
n
ce
p
t
n
a
m
es,
P
is
a
s
et
o
f
o
b
j
e
ct
p
r
o
p
er
ty
n
a
m
es,
R
is
a
s
et
o
f
d
ata
p
r
o
p
er
ty
n
a
m
e
s
,
a
n
d
I
i
s
a
s
e
t
o
f
in
d
i
v
id
u
al
n
a
m
e
s
.
T
h
e
u
n
io
n
P
∪
R
is
r
ef
er
r
ed
to
as th
e
s
et
o
f
p
r
o
p
er
t
y
n
a
m
e
s
.
4
.
1
.
2
.
Def
ini
t
io
n o
f
Si
m
ila
rit
y
Sea
rc
h Alg
o
rit
h
m
σ
Giv
e
n
t
w
o
o
n
to
lo
g
ies
O1
a
n
d
O2
an
d
th
eir
s
i
g
n
at
u
r
es
S1
=
{C1
,
P
1
,
R
1
,
I
1
i}
an
d
S2
=
{C2
,
P
2
,
R
2
,
I
2
},
a
s
i
m
ilar
it
y
s
ea
r
ch
a
lg
o
r
i
th
m
σ
is
d
ef
in
ed
as
σ
(
S,
Si
m
l
r
Strin
g
)
→
T
w
h
er
e
S
=
C
2
|
P
2
|
R
2
|
I
2
is
th
e
se
ar
ch
s
p
ac
e
s
u
c
h
th
at
T
∈
S.
Si
m
lr
Strin
g
∈
S1
is
a
s
ea
r
ch
s
tr
in
g
.
T
t
y
p
e
s
h
o
u
ld
b
e
s
a
m
e
as
Si
m
lr
Stri
n
g
,
i.e
.
SlrStr
in
g
∈
C
1
w
ill
lead
to
T
∈
C
2
a
n
d
s
o
o
n
.
B
y
r
ed
u
c
in
g
th
e
p
r
o
b
le
m
w
ith
j
u
s
t
co
n
s
id
er
in
g
o
n
e
n
a
m
e
f
r
o
m
S1
as
Si
m
lr
S
tr
in
g
,
w
e
tr
ied
to
k
ee
p
th
e
al
g
o
r
it
h
m
m
o
r
e
g
e
n
e
r
al,
s
o
it
co
u
ld
b
e
u
s
ed
b
y
o
t
h
er
ap
p
licatio
n
s
s
u
c
h
as
s
ea
r
ch
e
n
g
in
e
s
,
w
h
ic
h
n
ee
d
to
f
in
d
a
co
n
ce
p
t
in
o
n
to
lo
g
y
s
i
m
ilar
to
a
s
ea
r
ch
te
x
t.
F
o
r
th
e
s
ak
e
o
f
th
e
s
i
m
p
lic
it
y
,
in
t
h
e
f
o
llo
w
i
n
g
s
,
w
e
o
n
l
y
r
e
f
er
to
co
n
ce
p
t
s
b
u
t
s
i
m
ilar
m
et
h
o
d
s
co
u
l
d
b
e
ap
p
lied
to
s
ea
r
ch
f
o
r
o
th
er
p
ar
ts
o
f
s
ig
n
at
u
r
es.
4
.
1
.
3
.
Alg
o
rit
h
m
o
f
Si
m
i
la
rit
y
Sea
rc
h Alg
o
rit
h
m
FIN
DI
NGSI
MI
L
AR
I
T
Y
(
Si
m
lr
Strin
g
,
On
to
Sear
c
h
L
is
t
)
1
: Fir
s
t tr
ies to
f
i
n
d
r
eso
u
r
ce
t
h
at
ar
e
s
i
m
ilar
to
Si
m
lr
Strin
g
d
ir
ec
tl
y
2
: Si
m
lr
O
n
tR
e
s
←
FIN
DL
E
XI
C
AL
SIM
I
L
A
R
(
Sl
r
Strin
g
,
On
to
Sear
ch
L
is
t,
I
s
u
b
T
h
r
s
h
ld
)
3
: if
Si
m
lr
On
tR
e
s
6
=
NI
L
t
h
en
4
: if
SEM
A
NT
I
C
FIL
T
E
R
AC
C
E
P
T
S (
SlrOn
tR
es.
L
o
ca
lN
a
m
e
,
Si
m
lr
Stri
n
g
)
t
h
e
n
5
: r
etu
r
n
Si
m
lr
O
n
tR
e
s
6
: e
n
d
if
7
: e
n
d
if
8
: B
C
r
ea
tin
g
Sear
c
h
Ma
tr
ix
9
: M
←
W
OR
DNE
T
NUM
B
E
R
OFME
ANI
NG
(
Si
m
lr
Stri
n
g
)
1
0
: Si
m
lr
Ma
tr
ix
←
B
UI
L
DE
MPT
YSI
MI
L
A
R
I
T
YM
A
T
R
I
X
(
M)
1
1
: f
o
r
i ←
0
to
M
−
1
d
o
1
2
:A
DDT
OR
OW
(
Si
m
lr
Ma
tr
i
x
,
i,W
OR
DNE
T
GE
T
SYNONYM
S(Si
m
lr
Stri
n
g
,
i)
)
1
3
:A
P
P
E
NDT
OR
OW
(
Sim
lr
M
atr
ix
,
i,W
OR
DNE
T
GE
T
HYP
E
R
NYM
S(
Si
m
Stri
n
g
,
i)
)
1
4
: e
n
d
f
o
r
1
5
: B
C
alcu
late
Mo
s
t Si
m
ilar
s
1
6
:
C
AL
C
U
L
A
T
E
SIM
I
L
A
R
I
T
I
E
S (
On
t0
Sear
ch
L
is
t,
Sear
ch
Ma
tr
ix
)
1
7
: Can
d
id
ateA
r
r
a
y
←
B
UI
L
DAR
R
A
Y
(
M)
1
8
: f
o
r
i ←
0
to
M
−
1
d
o
1
9
: Can
d
id
ateA
r
r
a
y
[
i]
←
FIN
DC
A
NDI
D
A
T
E
(
Sear
ch
Ma
tr
i
x
,
i
)
2
0
: e
n
d
f
o
r
2
1
: B
W
o
r
d
Sen
s
e
Dis
a
m
b
i
g
u
atio
n
2
2
: p
r
ef
er
r
ed
Me
an
in
g
←
W
S
D
(
Sear
ch
Ma
tr
ix
[
i]
)
2
3
: if
C
an
d
id
ate
A
r
r
a
y
[
p
r
ef
er
r
ed
Me
an
in
g
]
[
i]
6
=
NI
L
th
e
n
2
4
: r
etu
r
n
C
an
d
id
ate
A
r
r
a
y
[
p
r
ef
er
r
ed
Me
an
in
g
]
.
Mo
s
tS
i
m
ilar
On
t
R
es
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8822
I
J
E
R
E
Vo
l.
5
,
No
.
3
,
Sep
tem
b
er
20
1
6
:
227
–
2
3
4
232
2
5
: e
n
d
if
2
6
: B
I
f
W
SD f
ailed
2
7
: f
o
r
i ←
0
to
M
−
1
d
o
2
8
: if
C
an
d
id
ate
A
r
r
a
y
[
i]
6
=
NI
L
th
e
n
2
9
: r
etu
r
n
C
an
d
id
ate
A
r
r
a
y
[
i]
.
Mo
s
tSi
m
ilar
O
n
tR
e
s
3
0
: e
n
d
if
3
1
: e
n
d
f
o
r
3
2
: B
No
t f
o
u
n
d
3
3
: r
etu
r
n
NI
L
4
.
2
.
Na
iv
e
B
a
y
esia
n
Ra
nk
ing
Alg
o
rit
h
m
T
h
e
Naiv
e
B
ay
esia
n
R
a
n
k
in
g
A
l
g
o
r
ith
m
h
e
lp
s
to
r
an
k
th
e
au
d
io
an
d
v
id
eo
b
y
th
e
u
s
er
b
ased
r
ev
ie
w
.
T
h
is
is
o
n
e
o
f
t
h
e
m
o
s
t
ef
f
ec
tiv
e
al
g
o
r
it
h
m
s
to
r
an
k
th
e
a
u
d
io
an
d
v
id
eo
f
ile
s
.
T
h
e
d
escr
ib
ed
p
r
o
b
lem
o
f
r
an
k
i
n
g
an
d
s
u
g
g
esti
n
g
t
h
i
n
g
s
ar
e
ar
is
e
in
a
v
ar
iet
y
o
f
ap
p
li
ca
tio
n
s
i
n
cl
u
d
e
i
n
ter
ac
tiv
e
co
m
p
u
tat
io
n
al
s
ch
e
m
e
f
o
r
h
elp
i
n
g
p
eo
p
le
to
p
o
w
er
s
o
cial
in
f
o
r
m
at
io
n
;
in
tec
h
n
o
lo
g
ical
th
e
s
e
s
y
s
te
m
s
ar
e
ca
lled
s
o
cial
n
a
v
i
g
atio
n
s
y
s
te
m
s
.
T
h
ese
s
o
cial
n
av
ig
a
tio
n
s
y
s
te
m
s
h
elp
ea
ch
in
d
i
v
id
u
al
an
d
th
eir
p
er
f
o
r
m
a
n
ce
an
d
th
eir
d
ec
is
io
n
m
ak
in
g
o
v
er
s
elec
t
in
g
an
s
w
er
s
.
B
ased
o
n
th
e
ea
ch
p
er
s
o
n
al
ities
r
ep
l
y
t
h
e
r
an
k
in
g
an
d
s
u
g
g
e
s
ti
n
g
o
f
p
o
p
u
lar
ite
m
s
w
as
d
o
n
e.
T
h
e
p
er
s
o
n
’
s
o
p
in
io
n
m
ig
h
t
b
e
o
b
tain
ed
b
y
d
is
p
la
y
i
n
g
a
s
e
t
o
f
s
u
g
g
e
s
ted
an
s
w
er
s
,
w
h
er
e
t
h
e
s
elec
tio
n
o
f
a
n
s
w
er
s
is
b
ased
o
n
th
e
li
k
in
g
o
f
th
e
en
t
it
y
.
T
h
e
p
lan
is
to
p
r
o
p
o
s
e
ac
ce
p
ted
ite
m
s
b
y
r
ap
id
l
y
s
tu
d
y
in
g
t
h
e
tr
u
e
p
o
p
u
lar
it
y
r
an
k
i
n
g
o
f
an
s
w
er
s
.
B
y
th
i
s
m
et
h
o
d
p
r
o
p
o
s
ed
in
,
w
h
ic
h
d
ef
i
n
es
a
s
co
r
e
f
o
r
a
q
u
er
y
b
ased
o
n
th
e
r
elati
v
e
en
t
r
o
p
y
b
et
w
ee
n
t
h
e
q
u
er
y
a
n
d
c
o
llectio
n
lan
g
u
ag
e
m
o
d
els.
(
1
)
W
h
er
e
V
ci
i
s
t
h
e
e
n
tire
v
o
ca
b
u
lar
y
o
f
t
h
e
co
llectio
n
C
i
,
a
n
d
i
=
1
;
2
;
3
r
ep
r
esen
t
tex
t
,
i
m
a
g
e
a
n
d
v
id
eo
,
r
esp
ec
tiv
el
y
.
T
h
e
T
er
m
s
P
(
w
|
q
)
an
d
P
(
w
|
C
i
)
ar
e
th
e
q
u
er
y
an
d
co
llectio
n
lan
g
u
ag
e
m
o
d
els,
r
esp
ec
tiv
el
y
.
T
h
e
C
lar
it
y
v
alu
e
b
ec
o
m
es
s
m
al
ler
as
th
e
to
p
r
an
k
ed
d
o
cu
m
e
n
ts
ap
p
r
o
ac
h
a
r
an
d
o
m
s
a
m
p
le
f
r
o
m
t
h
e
co
llectio
n
.
T
h
e
q
u
er
y
la
n
g
u
a
g
e
m
o
d
el
is
esti
m
ated
f
r
o
m
th
e
to
p
d
o
cu
m
e
n
ts
,
R
,
as th
e
f
o
llo
w
i
n
g
f
o
r
m
u
la,
(
2
)
an
d
ᵶ
is
d
ef
i
n
ed
as,
(
3
)
W
h
er
e
P
(
q
|
D
)
is
th
e
q
u
er
y
li
k
elih
o
o
d
s
co
r
e
o
f
d
o
cu
m
en
t
D
.
W
e
ap
p
ly
th
i
s
m
eth
o
d
to
ca
lcu
late,
(
4
)
I
n
th
i
s
w
o
r
k
,
f
o
r
a
q
u
er
y
g
en
er
ated
f
r
o
m
a
g
i
v
en
Q
A
p
air
,
w
e
u
s
e
m
u
ltip
le
d
o
cu
m
e
n
t
s
(
f
o
r
s
ev
er
a
l
co
m
p
le
x
q
u
er
ies,
t
h
er
e
m
a
y
b
e
less
th
a
n
2
0
r
esu
lts
r
etu
r
n
ed
)
to
esti
m
ate
th
e
r
etr
iev
a
l
ef
f
e
ctiv
e
n
ess
f
o
r
ea
ch
m
ed
iu
m
t
y
p
e,
i
n
cl
u
d
in
g
tex
t,
i
m
a
g
e
a
n
d
v
id
eo
.
T
h
e
Naiv
e
B
ay
e
s
ian
A
p
p
r
o
ac
h
r
ep
r
esen
t
s
t
h
e
cla
s
s
-
s
p
ec
i
f
ic
r
elate
d
w
o
r
d
s
in
m
u
l
tip
le
f
o
r
m
ats.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
R
E
I
SS
N:
2252
-
8822
Do
ma
in
a
n
d
I
n
tellig
en
ce
B
a
s
e
d
Mu
ltimed
ia
Qu
esti
o
n
A
n
s
w
erin
g
S
ystem
(
K
.
Ma
g
esh
K
u
ma
r
)
233
T
ab
le
1
.
R
ep
r
esen
tativ
e
C
las
s
-
Sp
ec
if
ic
R
e
lated
W
o
r
d
s
C
a
t
e
g
o
r
i
e
s
C
l
a
ss
-
S
p
e
c
i
f
i
c
R
e
l
a
t
e
d
W
o
r
k
L
i
st
T
e
x
t
n
a
me
,
p
o
p
u
l
a
t
i
o
n
,
p
e
r
i
o
d
,
t
i
me
s,
c
o
u
n
t
r
y
,
h
e
i
g
h
t
,
w
e
b
si
t
e
,
b
i
r
t
h
d
a
y
,
a
g
e
,
d
a
t
e
,
r
a
t
e
,
d
i
st
a
n
c
e
,
s
p
e
e
d
,
r
e
l
i
g
i
o
n
s,
n
u
mb
e
r
,
e
t
c
T
e
x
t
+
I
mag
e
c
o
l
o
u
r
,
p
e
t
,
c
l
o
t
h
e
s,
l
o
o
k
l
i
k
e
,
w
h
o
,
i
mag
e
,
p
i
c
t
u
r
e
s,
a
p
p
e
a
r
a
n
c
e
,
l
a
r
g
e
st
,
b
a
n
d
,
p
h
o
t
o
,
s
u
r
f
a
c
e
,
c
a
p
i
t
a
l
,
f
i
g
u
r
e
,
w
h
a
t
i
s
a
,
sy
mb
o
l
,
w
h
o
m,
l
o
g
o
,
p
l
a
c
e
,
e
t
c
T
e
x
t
+
V
i
d
e
o
h
o
w
t
o
,
h
o
w
d
o
,
h
o
w
c
a
n
,
i
n
v
e
n
t
e
d
,
st
o
r
y
,
f
i
l
m,
t
e
l
l
,
so
n
g
s,
mu
s
i
c
,
r
e
c
i
p
e
,
d
i
f
f
e
r
e
n
c
e
s,
w
a
y
s,
st
e
p
s,
d
a
n
c
e
,
f
i
r
st
,
s
a
i
d
,
e
t
c
T
e
x
t
+
I
mag
e
+
V
i
d
e
o
p
r
e
si
d
e
n
t
,
k
i
n
g
,
p
r
i
me
m
i
n
i
s
t
e
r
,
k
i
l
l
,
i
s
su
e
,
n
u
c
l
e
a
r
,
e
a
r
t
h
q
u
a
k
e
,
si
n
g
e
r
m
b
a
t
t
l
e
,
e
v
e
n
t
,
w
a
r
,
h
a
p
p
e
n
e
d
,
e
t
c
5.
E
XP
E
R
I
M
E
NT
A
L
RE
SUL
T
S AN
D
A
NALY
SI
S
T
h
e
Data
s
et
f
o
r
ex
p
er
im
en
ts
h
av
e
t
w
o
s
u
b
s
et
s
.
First,
ar
b
itra
r
il
y
c
o
llect
s
o
m
e
q
u
esti
o
n
f
r
o
m
w
i
k
i
a
n
s
w
er
s
an
d
f
o
r
s
ec
o
n
d
,
co
lle
ct
a
f
e
w
q
u
es
tio
n
a
n
d
t
h
eir
r
el
ativ
e
a
n
s
w
er
s
f
r
o
m
Y
!
A
.
R
an
k
in
g
m
et
h
o
d
is
u
s
ed
to
f
in
d
/s
elec
t
th
e
b
est
o
r
f
in
es
t
an
s
w
er
f
r
o
m
t
h
e
d
atab
ase
an
d
h
er
e
f
o
r
r
an
k
in
g
t
h
e
v
o
te
is
co
llected
f
r
o
m
th
e
u
s
er
b
ased
o
n
Naiv
e
B
a
y
esia
n
alg
o
r
ith
m
.
T
o
ca
lcu
late
o
u
r
an
s
w
er
,
t
h
e
m
ed
iu
m
s
elec
tio
n
ap
p
r
o
ac
h
is
u
s
ed
h
er
e
it
m
e
n
tio
n
s
t
h
e
lab
el
th
at
in
v
o
l
v
ed
i
n
t
h
e
g
r
o
u
n
d
tr
u
t
h
l
ab
elin
g
p
r
o
ce
s
s
.
T
h
e
y
ar
e
h
el
p
f
u
l
in
an
s
w
er
m
ed
iu
m
i
n
f
o
r
m
atio
n
,
f
o
r
e
x
a
m
p
l
e
Ho
w
to
co
n
n
ec
t
a
s
y
s
te
m
u
s
in
g
r
i
n
g
to
p
o
lo
g
y
.
Fo
r
th
i
s
q
u
esti
o
n
,
i
f
t
h
e
an
s
w
er
f
o
r
r
elev
an
t
q
u
est
io
n
i
s
av
ailab
le
in
t
h
e
d
atab
ase
th
e
n
th
e
Q
A
s
y
s
te
m
ea
s
il
y
r
etr
i
ev
e
th
e
a
n
s
w
e
r
f
r
o
m
th
e
d
ata
b
ase.
Oth
er
w
is
e
t
h
e
q
u
esti
o
n
is
p
o
s
ted
i
n
t
h
e
Q
A
s
y
s
te
m
b
y
th
e
u
s
er
f
o
r
th
is
p
o
s
ted
q
u
esti
o
n
t
h
e
u
s
er
ca
n
p
o
s
t
th
e
an
s
w
er
s
i
n
t
h
e
w
a
y
o
f
tex
t,
v
id
eo
a
n
d
i
m
a
g
es.
I
n
t
h
is
s
y
s
te
m
a
s
et
o
f
u
s
er
w
il
l
ex
p
ec
t
m
u
lti
m
ed
ia
a
n
s
w
er
b
ec
au
s
e
w
h
e
n
co
m
p
ar
i
n
g
to
th
e
te
x
tu
a
l
an
s
w
er
s
t
h
e
v
id
eo
an
s
w
er
s
ar
e
m
o
r
e
i
n
f
o
r
m
ati
v
e
.
I
n
th
is
tab
l
e
it
s
h
o
w
s
th
at
m
o
r
e
th
an
5
0
%
o
f
t
h
e
q
u
e
s
tio
n
ca
n
b
e
an
s
w
er
ed
b
y
ad
d
in
g
m
u
lti
m
ed
ia
co
n
te
n
t
s
in
s
tead
o
f
p
u
r
el
y
te
x
t.
So
w
e
ca
n
co
n
clu
d
e
t
h
at
m
u
lti
m
ed
ia
ap
p
r
o
ac
h
h
i
g
h
l
y
p
r
ef
er
r
ed
.
A
c
co
r
d
in
g
to
t
h
e
r
es
u
lt,
t
h
e
c
o
m
p
ar
is
o
n
b
et
w
ee
n
o
r
ig
in
al
te
x
t
u
al
an
s
w
er
an
d
m
ed
ia
an
s
w
er
th
e
m
u
l
ti
m
ed
i
a
an
s
w
er
i
n
g
s
y
s
te
m
i
s
m
o
r
e
u
s
e
f
u
l
b
ec
au
s
e
th
e
tex
t
u
al
a
n
s
w
er
h
a
v
e
o
n
l
y
te
x
t
a
n
d
it
is
les
s
p
r
e
f
er
ab
le
b
y
t
h
e
u
s
er
b
u
t
t
h
e
an
s
w
er
s
w
it
h
t
h
e
tex
t
an
d
m
u
lti
m
e
d
ia
f
ea
t
u
r
es
ar
e
m
o
r
e
ef
f
ec
tiv
e
a
n
d
u
n
d
er
s
ta
n
d
ab
le
.
T
h
e
tab
le
s
h
o
w
s
th
e
ac
t
u
al
Ass
i
s
tan
ce
o
f
o
n
l
y
tex
t
u
al
a
n
s
w
er
a
n
d
te
x
tu
a
l
a
n
s
w
er
s
w
it
h
m
u
lti
m
ed
ia
f
ea
t
u
r
es.
T
h
e
ex
p
er
i
m
en
tal
s
et
tin
g
s
p
r
ese
n
t
t
h
e
u
s
er
s
tu
d
y
r
es
u
lt
i
n
T
ab
le
1
.
A
cc
o
r
d
in
g
to
s
t
u
d
y
m
o
r
e
th
a
n
7
0
%
p
eo
p
le
p
r
ef
er
s
m
ed
ia
an
s
w
er
alo
n
g
w
it
h
te
x
tu
al
an
s
w
er
.
I
t
is
m
o
r
e
i
m
p
o
r
tan
t
c
o
m
m
u
n
it
y
m
e
m
b
er
to
p
r
o
v
id
e
th
e
an
s
w
er
w
it
h
m
ed
ia
d
ata
t
o
b
etter
u
n
d
er
s
tan
d
th
e
q
u
es
tio
n
.
T
ab
le
2
.
C
o
m
p
ar
is
o
n
o
f
S
y
s
te
m
Usab
ili
t
y
P
r
e
f
e
r
me
d
i
a
a
n
sw
e
r
N
o
a
n
sw
e
r
P
r
e
f
e
r
o
r
i
g
i
n
a
l
t
e
x
t
u
a
l
a
n
sw
e
r
7
0
%
5%
2
5
%
T
h
e
f
o
llo
w
i
n
g
tab
le
(
T
ab
le
3
)
c
o
m
p
ar
es
t
h
e
m
u
lti
m
ed
ia
f
e
atu
r
e
w
it
h
Y
!
A
a
n
d
W
ik
i
An
s
w
er
s
,
th
e
Y!
A
an
d
W
ik
i
An
s
w
er
s
d
o
es
n
o
t
s
u
p
p
o
r
t
m
u
lti
m
ed
ia
an
s
wer
s
b
u
t
th
e
DI
MQ
A
p
r
o
v
id
e
t
h
e
9
5
%
o
f
ef
f
ec
ti
v
e
an
s
w
er
s
i
n
m
u
lti
m
ed
ia
f
o
r
m
a
t.
T
ab
le
3
.
DI
MQ
A
w
i
th
M
u
lti
m
ed
ia
Featu
r
e
M
e
t
h
o
d
Y
!
A
W
i
k
i
A
n
sw
e
r
D
I
M
Q
A
T
e
x
t
-
B
a
se
me
t
h
o
d
8
2
.
1
7
%
8
5
.
2
6
%
9
2
.
4
5
%
M
u
l
t
i
me
d
i
a
a
p
p
r
o
a
c
h
N
I
L
N
I
L
9
5
%
T
h
e
f
o
llo
w
in
g
c
h
ar
t
(
F
ig
u
r
e
1
)
s
h
o
w
s
th
e
r
es
u
lt
o
f
DI
MQ
A
s
y
s
te
m
u
s
in
g
o
n
l
y
te
x
t
u
al
an
s
w
er
a
n
d
tex
t
u
al
w
it
h
m
u
lt
i
m
ed
ia
a
n
s
wer
.
0
50
10
0
o
n
ly
textu
al
an
s
w
e
rs
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8822
I
J
E
R
E
Vo
l.
5
,
No
.
3
,
Sep
tem
b
er
20
1
6
:
227
–
2
3
4
234
Fig
u
r
e
3.
DI
MQ
A
s
y
s
te
m
p
er
f
o
r
m
a
n
ce
ev
a
lu
atio
n
T
h
e
an
s
w
er
i
n
g
m
o
d
es
ar
e
u
s
ed
to
g
iv
e
b
r
ief
an
s
w
er
s
to
th
e
f
ac
u
lt
y
a
n
d
s
tu
d
e
n
t
to
g
ain
m
o
r
e
k
n
o
w
led
g
e
an
d
t
h
e
v
id
eo
an
s
w
er
s
ar
e
ex
p
lai
n
ed
i
n
a
d
et
ailed
m
a
n
n
er
.
W
h
e
n
co
m
p
ar
i
n
g
w
it
h
t
h
e
tex
tu
al
an
s
w
er
i
n
g
m
o
d
e,
th
e
m
u
l
ti
m
e
d
ia
an
s
w
er
in
g
m
o
d
e
s
p
r
o
v
id
es th
e
ess
e
n
tial i
n
f
o
r
m
atio
n
to
th
e
u
s
er
s
.
6.
CO
NCLU
SI
O
N
AND
F
U
T
U
RE
E
NH
A
NCEM
E
NT
T
h
e
QA
s
y
s
te
m
d
e
v
elo
p
ed
b
y
th
e
s
e
m
a
n
tic
m
atc
h
al
g
o
r
ith
m
a
n
d
Nai
v
e
B
a
y
esia
n
r
a
n
k
i
n
g
al
g
o
r
ith
m
w
h
ic
h
allo
w
s
t
h
e
m
u
ltip
le
u
s
er
s
to
s
h
ar
e
th
e
ir
an
s
w
er
i
n
th
e
w
a
y
o
f
te
x
t,
i
m
a
g
e,
a
u
d
io
an
d
v
id
eo
.
T
h
e
ef
f
ec
tiv
e
w
a
y
o
f
an
s
w
e
r
i
n
g
m
o
d
e
s
is
e
v
al
u
ated
w
it
h
s
e
m
an
tic
m
atc
h
an
d
n
aïv
e
B
a
y
esi
an
alg
o
r
it
h
m
s
a
n
d
it
p
r
o
v
id
es
th
e
b
est
a
n
s
w
er
f
o
r
t
h
e
u
s
er
s
.
W
ith
t
h
e
h
elp
o
f
t
h
i
s
s
y
s
te
m
e
v
er
y
s
t
u
d
en
t
an
d
f
a
cu
lt
y
w
i
ll
g
et
g
o
o
d
k
n
o
w
led
g
e
o
n
v
a
r
io
u
s
s
u
b
j
ec
ts
.
T
h
is
s
y
s
te
m
ca
n
b
e
f
u
r
th
er
ex
ten
d
ed
to
g
e
n
er
ate
t
h
e
a
u
to
m
atic
Qu
e
s
tio
n
b
an
k
f
o
r
a
s
t
u
d
en
t
a
n
d
g
e
n
er
ate
t
h
e
q
u
esti
o
n
p
ap
er
s
f
o
r
co
lleg
e
ex
a
m
in
at
io
n
.
Ma
i
n
l
y
,
th
e
s
y
s
t
e
m
f
u
r
t
h
er
ex
ten
d
s
w
it
h
e
-
p
r
o
f
ess
o
r
m
o
d
e
to
g
et
i
m
m
ed
iate
s
o
lu
tio
n
f
o
r
th
e
q
u
e
s
tio
n
w
h
ic
h
m
ea
n
s
it
w
i
ll d
esi
g
n
f
o
r
o
n
li
n
e
ch
at
s
.
RE
F
E
R
E
NC
E
S
[1
]
L
.
Nie
,
e
t
a
l.
,
“
Be
y
o
n
d
tex
t
QA
:
M
u
lt
im
e
d
ia
a
n
sw
e
r
g
e
n
e
ra
ti
o
n
b
y
h
a
rv
e
stin
g
W
e
b
in
f
o
rm
a
ti
o
n
,”
M
u
lt
ime
d
i
a
,
I
EE
E
T
ra
n
sa
c
ti
o
n
s
o
n
,
v
o
l/
iss
u
e
:
15
(
2
),
p
p
.
4
2
6
-
4
4
1
,
2
0
1
3
.
[2
]
A
.
M
o
sc
h
it
ti
a
n
d
S
.
Q
u
a
rt
e
ro
n
i
,
“
L
in
g
u
isti
c
k
e
rn
e
ls
f
o
r
a
n
s
w
e
r
re
-
ra
n
k
in
g
in
q
u
e
stio
n
a
n
sw
e
rin
g
s
y
ste
m
s
,”
In
fo
rm
a
t
io
n
Pro
c
e
ss
in
g
&
M
a
n
a
g
e
me
n
t
,
v
o
l/
issu
e
:
4
(7
,
6
)
p
p
.
8
2
5
-
8
4
2
,
2
0
1
1
.
[3
]
C.
H.
Hs
u
,
e
t
a
l.
,
“
Us
in
g
d
o
m
a
in
o
n
t
o
lo
g
y
to
im
p
le
m
e
n
t
a
f
r
e
q
u
e
n
tl
y
a
s
k
e
d
q
u
e
stio
n
s
sy
ste
m
,”
Co
mp
u
ter
S
c
ien
c
e
a
n
d
In
f
o
rm
a
ti
o
n
En
g
in
e
e
rin
g
,
2
0
0
9
W
RI
W
o
rl
d
Co
n
g
re
ss
o
n
.
,
v
o
l.
4
,
2
0
0
9
.
[4
]
R.
C.
W
a
n
g
,
N.
S
c
h
lae
f
e
r,
W
.
W
.
Co
h
e
n
,
E.
Ny
b
e
rg
,
“
Au
to
ma
t
ic
se
t
e
x
p
a
n
sio
n
f
o
r
li
st
q
u
e
sti
o
n
a
n
swe
rin
g
,
”
i
n
P
r
o
c
.
I
n
t.
C
o
n
f
.
Em
p
iri
c
a
l
M
e
th
o
d
s in
Na
t
u
ra
l
L
a
n
g
u
a
g
e
P
ro
c
e
ss
in
g
,
2
0
0
8
.
[5
]
E.
P
a
rz
e
n
a
n
d
F
.
H
o
ti
,
“
On
Esti
m
a
ti
o
n
o
f
a
P
ro
b
a
b
i
li
ty
De
n
sit
y
F
u
n
c
ti
o
n
a
n
d
M
o
d
e
,
”
An
n
a
ls
o
f
M
a
th
e
ma
ti
c
a
l
S
ta
ti
st
ics
,
v
o
l
/i
ss
u
e
:
33
(
3
)
,
1
9
6
2
.
[6
]
T
.
S
.
Ch
u
a
,
e
t
a
l.
,
“
F
ro
m
tex
t
q
u
e
stio
n
-
a
n
sw
e
rin
g
to
m
u
lt
im
e
d
ia Q
A
o
n
w
e
b
-
sc
a
le
m
e
d
ia res
o
u
rc
e
s
,”
Pro
c
e
e
d
in
g
s
o
f
th
e
Fi
rs
t
ACM
wo
rk
sh
o
p
o
n
L
a
r
g
e
-
sc
a
le mu
lt
ime
d
ia
re
triev
a
l
a
n
d
min
in
g
,
ACM
,
2
0
0
9
.
[7
]
M
.
W
a
n
g
a
n
d
X
.
S
.
Hu
a
,
“
A
c
ti
v
e
lea
rn
in
g
in
m
u
lt
im
e
d
ia
a
n
n
o
tatio
n
a
n
d
re
tri
e
v
a
l:
A
su
rv
e
y
,
”
AC
M
T
ra
n
s.
In
tel
l
.
S
y
st.
T
e
c
h
n
o
l.
,
v
o
l
/i
ss
u
e
:
2
(
2
)
,
p
p
.
1
0
–
31
,
2
0
1
1
.
[8
]
Y.
Ga
o
,
M
.
W
a
n
g
,
Z.
J.
Zh
a
,
Q.
T
ian
,
Q.
D
a
i,
N.
Zh
a
n
g
,
“
L
e
ss
i
s
m
o
re
:
E
ff
icie
n
t
3
d
o
b
jec
t
re
tri
e
v
a
l
w
it
h
q
u
e
r
y
v
ie
w
se
lec
ti
o
n
,
”
IEE
E
T
ra
n
s.
M
u
l
ti
me
d
ia
,
v
o
l
/i
ss
u
e
:
13
(
5
)
,
p
p
.
1
0
0
7
–
1
0
1
8
,
2
0
1
1
.
[9
]
I.
A
h
m
a
d
a
n
d
T
.
S
.
Ja
n
g
,
“
Ol
d
f
a
s
h
io
n
tex
t
-
b
a
se
d
im
a
g
e
re
tri
e
v
a
l
u
sin
g
F
CA
,
”
in
Pro
c
.
ICI
P
,
2
0
0
3
.
[1
0
]
D.
L
iu
,
e
t
a
l
.
, “
T
a
g
Ra
n
k
in
g
,
”
P
ro
c
.
1
8
t
h
I
n
t‟l
Co
n
f
.
W
o
rld
W
id
e
W
e
b
,
A
CM
P
re
ss
,
p
p
.
3
5
1
-
3
6
0
,
2
0
0
9
.
[1
1
]
X
.
T
ian
,
L
.
Y
a
n
g
,
J.
Wan
g
,
Y.
Ya
n
g
,
X
.
W
u
,
X.
S
.
Hu
a
,
“
Ba
y
e
sia
n
v
id
e
o
se
a
rc
h
re
ra
n
k
in
g
,
”
in
P
ro
c
.
A
CM
In
t.
Co
n
f
.
M
u
l
ti
m
e
d
ia
,
2
0
0
8
.
[1
2
]
S.
L
iu
,
e
t
a
l.
,
“
S
o
c
ial
v
isu
a
l
im
a
g
e
ra
n
k
in
g
f
o
r
w
e
b
im
a
g
e
se
a
rc
h
,
”
A
d
v
a
n
c
e
s
in
M
u
lt
im
e
d
ia
M
o
d
e
li
n
g
,
S
p
r
in
g
e
r
Be
rli
n
He
id
e
lb
e
rg
,
p
p
.
2
3
9
-
2
4
9
,
2
0
1
3
.
[1
3
]
H.
F
e
n
g
,
A
.
Ch
a
n
d
ra
sh
e
k
h
a
ra
,
T
.
S
.
Ch
u
a
,
“
T
a
mr
a
:
a
n
Au
t
o
ma
ti
c
T
e
mp
o
ra
l
M
u
lt
ire
so
lu
ti
o
n
An
a
ly
sis
Fra
me
wo
rk
fo
r S
h
o
t
Bo
u
n
d
a
ry
De
tec
ti
o
n
,
”
P
r
o
c
.
In
t‟l
Co
n
f
.
M
u
lt
im
e
d
ia
M
o
d
e
li
n
g
(M
M
M
)
,
A
CM
P
re
ss
,
2
0
0
3
.
[1
4
]
S
.
L
a
z
e
b
n
ik
,
C.
S
c
h
m
id
,
J.
P
o
n
c
e
,
“
Bey
o
n
d
Ba
g
s
o
f
Fea
tu
re
s:
S
p
a
ti
a
l
Pyr
a
mid
M
a
tch
i
n
g
f
o
r
Rec
o
g
n
izin
g
Na
tu
r
a
l
S
c
e
n
e
Ca
teg
o
rie
s
,
”
P
ro
c
.
IEE
E
Co
m
p
u
ter
S
o
c
iety
Co
n
f
.
Co
m
p
u
ter
V
isio
n
a
n
d
P
a
tt
e
r
n
Re
c
o
g
n
it
io
n
(CVP
R),
IE
EE
CS
P
re
ss
,
2
0
0
6
.
[1
5
]
H.
Ba
y
,
e
t
a
l.
,
“
S
p
e
e
d
e
d
-
Up
R
o
b
u
st
F
e
a
t
u
re
s
(S
URF),
”
Co
m
p
u
t
e
r
Vi
sio
n
a
n
d
Ima
g
e
Un
d
e
rs
ta
n
d
in
g
,
v
o
l
/
issu
e
:
110
(
3
)
,
p
p
.
3
4
6
-
3
5
9
,
2
0
0
8
.
[1
6
]
J.
L
.
S
o
n
g
,
“
S
c
a
b
le
Ima
g
e
Retrie
v
a
l
Ba
se
d
o
n
Fea
tu
re
Fo
re
st
,
”
P
ro
c
.
A
si
a
n
Co
n
f
.
Co
m
p
u
ter V
isio
n
,
S
p
rin
g
e
r
P
re
ss
,
2
0
0
9
.
[1
7
]
D.
R.
Ra
d
e
v
,
e
t
a
l
.
,
“
Eva
lu
a
ti
n
g
W
e
b
-
b
a
se
d
Qu
e
stio
n
An
swe
rin
g
S
y
ste
ms
,
”
P
ro
c
.
In
t‟l
C
o
n
f
.
L
a
n
g
u
a
g
e
Re
so
u
rc
e
s
a
n
d
Ev
a
lu
a
ti
o
n
,
2
0
0
2
.
Evaluation Warning : The document was created with Spire.PDF for Python.