I
AE
S In
t
er
na
t
io
na
l J
o
urna
l o
f
Art
if
icia
l In
t
ellig
ence
(
I
J
-
AI
)
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
26
,
p
p
.
2838
~
2
8
4
8
I
SS
N:
2
2
5
2
-
8
9
3
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijai.v
15
.i
3
.
p
p
2
8
3
8
-
2
8
4
8
2838
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
Identifica
tion o
f
a
reas
of inf
lu
ence
using
a thema
tic
mo
deling
a
ppro
a
ch and bel
ief
f
unction
theor
y
F
a
t
im
a
-
Z
a
hra
e
Sifi
1
,
Wa
f
a
e
Sa
bb
a
r
1
,
Am
a
l El
M
za
bi
2
1
La
b
o
r
a
t
o
r
y
o
f
M
a
c
h
i
n
e
I
n
t
e
l
l
i
g
e
n
c
e
,
F
a
c
u
l
t
y
o
f
S
c
i
e
n
c
e
s
a
n
d
Te
c
h
n
o
l
o
g
y
,
H
a
ssan
I
I
U
n
i
v
e
r
si
t
y
,
M
o
h
a
m
me
d
i
a
,
M
o
r
o
c
c
o
2
La
b
o
r
a
t
o
r
y
o
f
E
c
o
n
o
mi
c
P
e
r
f
o
r
ma
n
c
e
s a
n
d
L
o
g
i
st
i
c
s
,
F
a
c
u
l
t
y
o
f
L
a
w
,
Ec
o
n
o
m
i
c
a
n
d
S
o
c
i
a
l
S
c
i
e
n
c
e
s
,
H
a
ss
a
n
I
I
U
n
i
v
e
r
si
t
y
,
M
o
h
a
mm
e
d
i
a
,
M
o
r
o
c
c
o
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
u
l
14
,
2
0
2
5
R
ev
is
ed
Mar
17
,
2
0
2
6
Acc
ep
ted
Ap
r
20
,
2
0
2
6
S
o
c
ial
n
e
two
rk
s
h
a
v
e
b
e
c
o
m
e
k
e
y
p
latfo
rm
s
fo
r
th
e
sp
re
a
d
o
f
i
n
fo
rm
a
ti
o
n
a
n
d
t
h
e
e
x
c
h
a
n
g
e
o
f
id
e
a
s.
T
h
e
y
c
h
a
n
g
e
th
e
wa
y
in
fl
u
e
n
c
e
rs
in
tera
c
t
a
n
d
sh
o
w
in
fl
u
e
n
c
e
o
v
e
r
p
e
o
p
le.
On
p
latfo
rm
s
su
c
h
a
s
Twit
ter,
F
a
c
e
b
o
o
k
,
a
n
d
In
sta
g
ra
m
,
in
flu
e
n
c
e
e
m
e
rg
e
s
o
v
e
r
so
c
ial
in
tera
c
ti
o
n
s
su
c
h
a
s
re
twe
e
ts,
li
k
e
s,
m
e
n
ti
o
n
s,
c
o
m
m
e
n
ts
,
a
n
d
sh
a
re
s.
Th
e
se
a
c
ti
v
it
ies
h
a
v
e
a
m
a
jo
r
ro
le
in
m
e
ss
a
g
e
a
m
p
li
fica
ti
o
n
a
n
d
o
p
i
n
io
n
f
o
rm
a
ti
o
n
.
T
h
ro
u
g
h
t
h
e
a
n
a
ly
sis
o
f
in
flu
e
n
c
e
rs’
p
ra
c
ti
c
e
s,
it
b
e
c
o
m
e
s
p
o
ss
ib
le
t
o
o
u
tl
in
e
t
h
e
a
re
a
s
w
h
e
re
th
e
ir
imp
a
c
t
is
p
a
rti
c
u
larly
si
g
n
ifi
c
a
n
t.
Ho
we
v
e
r,
t
h
is
tas
k
re
m
a
in
s
c
o
m
p
lex
.
Eff
e
c
ti
v
e
a
n
a
ly
sis
re
q
u
ires
ro
b
u
st
m
e
th
o
d
s
th
a
t
c
a
n
in
c
o
r
p
o
ra
te
d
iv
e
rse
fo
rm
s
o
f
s
o
c
ial
e
n
g
a
g
e
m
e
n
t
a
n
d
n
a
v
i
g
a
te
t
h
e
u
n
c
e
rtain
ti
e
s
p
o
se
d
b
y
h
e
tero
g
e
n
e
o
u
s
d
a
ta.
I
n
th
is
c
o
n
t
e
x
t,
we
p
ro
p
o
se
a
m
e
th
o
d
to
i
d
e
n
ti
fy
t
h
e
d
o
m
a
in
s
o
f
in
fl
u
e
n
c
e
o
f
a
so
c
ial
m
e
d
ia
in
flu
e
n
c
e
r.
T
h
is
a
p
p
ro
a
c
h
c
o
m
b
in
e
s
th
e
m
a
ti
c
m
o
d
e
l
late
n
t
d
iri
c
h
let
a
l
lo
c
a
ti
o
n
(LDA)
with
b
e
li
e
f
fu
n
c
ti
o
n
th
e
o
ry
(BF
T)
to
a
n
a
ly
z
e
so
c
ial
i
n
tera
c
ti
o
n
s
a
n
d
d
o
m
in
a
n
t
t
o
p
ics
o
f
in
t
e
re
st.
Th
e
met
h
o
d
e
x
a
m
in
e
s
in
d
ica
to
rs
s
u
c
h
a
s
re
twe
e
ts,
li
k
e
s
,
a
n
d
m
e
n
ti
o
n
s
a
n
d
p
ro
v
id
e
s
a
ro
b
u
st
fra
m
e
wo
rk
to
m
e
a
su
re
th
e
in
flu
e
n
c
e
r'
s
imp
a
c
t
a
c
ro
ss
d
iffere
n
t
d
o
m
a
in
s.
I
t
t
h
u
s
o
ffe
rs
p
re
c
ise
to
o
ls
f
o
r
re
se
a
rc
h
e
rs,
p
ra
c
ti
ti
o
n
e
rs
a
n
d
d
e
c
isio
n
-
m
a
k
e
rs wh
o
a
im to
b
e
tt
e
r
u
n
d
e
rsta
n
d
th
e
se
c
o
m
p
le
x
d
y
n
a
m
ics
.
K
ey
w
o
r
d
s
:
B
elief
f
u
n
ctio
n
th
eo
r
y
Dem
p
s
ter
-
Sh
af
er
th
eo
r
y
I
n
f
lu
en
ce
r
an
aly
s
is
L
aten
t d
ir
ich
let
allo
ca
tio
n
Natu
r
al
lan
g
u
ag
e
p
r
o
ce
s
s
in
g
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
:
Fatim
a
-
Z
ah
r
ae
Sifi
L
ab
o
r
ato
r
y
o
f
Ma
ch
in
e
I
n
tellig
en
ce
,
Facu
lty
o
f
Scien
ce
s
an
d
T
ec
h
n
o
lo
g
y
,
Hass
an
I
I
Un
iv
er
s
ity
Mo
h
am
m
ed
ia
2
8
8
0
6
Mo
r
o
cc
o
E
m
ail:
f
atim
az
ah
r
ae
.
s
if
i@
u
n
iv
h
2
c.
m
a
1.
I
NT
RO
D
UCT
I
O
N
I
n
r
ec
en
t
y
ea
r
s
,
p
latf
o
r
m
s
lik
e
s
o
cial
m
ed
ia
h
av
e
b
ec
o
m
e
ce
n
tr
al
to
th
e
cir
cu
latio
n
o
f
in
f
o
r
m
atio
n
an
d
p
u
b
lic
d
ialo
g
u
e
[
1
]
.
T
h
ei
r
wid
esp
r
ea
d
u
s
e
h
as
ch
a
n
g
e
d
th
e
way
we
co
m
m
u
n
icate
.
I
t
p
r
o
v
id
es
u
s
er
s
to
s
h
ar
e
co
n
ten
t
in
s
tan
tly
[
2
]
an
d
r
ea
ch
lar
g
e
d
iv
er
s
e
au
d
ien
ce
s
[
3
]
.
I
n
th
is
o
n
lin
e
en
v
ir
o
n
m
en
t,
in
f
lu
en
ce
r
s
–
in
d
iv
id
u
als
wh
o
co
m
m
an
d
s
u
b
s
tan
tial
atten
tio
n
an
d
en
g
ag
em
e
n
t
[
4
]
–
ex
er
t
co
n
s
id
er
ab
le
im
p
ac
t
o
n
o
p
i
n
io
n
s
,
tr
en
d
s
an
d
b
eh
av
io
r
s
ac
r
o
s
s
v
ar
io
u
s
s
ec
to
r
s
.
Platfo
r
m
s
s
u
ch
as
T
witter
,
Face
b
o
o
k
,
an
d
I
n
s
tag
r
a
m
f
ac
ilit
ate
th
is
in
f
lu
en
ce
th
r
o
u
g
h
in
ter
ac
tio
n
f
ea
tu
r
es
lik
e
r
et
wee
ts
,
lik
es,
m
en
tio
n
s
,
co
m
m
en
ts
,
an
d
s
h
ar
es
[
5
]
.
T
h
ese
m
ec
h
an
is
m
s
co
n
tr
ib
u
te
to
th
e
am
p
lific
atio
n
an
d
p
r
o
p
a
g
atio
n
o
f
m
ess
ag
es
[
6
]
.
T
o
u
n
d
er
s
tan
d
h
o
w
in
f
l
u
en
ce
is
ex
er
cised
is
a
ce
n
tr
al
ch
alle
n
g
e
f
o
r
r
esear
ch
e
r
s
an
d
p
r
ac
ti
tio
n
er
s
.
I
t
is
im
p
o
r
tan
t
to
i
d
en
tify
t
h
e
d
o
m
ain
s
o
f
g
r
ea
test
in
f
lu
en
c
er
im
p
ac
t
[
7
]
.
T
r
a
d
itio
n
al
in
f
l
u
en
ce
m
ea
s
u
r
em
e
n
t
tech
n
iq
u
es
f
r
e
q
u
en
tl
y
ig
n
o
r
e
t
wo
cr
itical
asp
ec
ts
.
T
h
e
f
i
r
s
t
is
th
e
th
em
atic
n
u
an
ce
s
o
f
co
n
ten
t
[
8
]
.
T
h
e
s
ec
o
n
d
is
th
e
in
tr
in
s
ic
u
n
ce
r
tain
ty
an
d
am
b
ig
u
ity
in
s
o
cial
en
g
ag
e
m
en
t
d
ata
[
9
]
.
T
h
is
r
esear
ch
a
r
ea
s
u
n
d
e
r
s
co
r
es
th
e
n
ee
d
f
o
r
ad
v
a
n
ce
d
an
al
y
tical
m
o
d
els.
T
h
ese
f
r
am
ewo
r
k
s
s
h
o
u
ld
co
m
b
in
e
m
ac
h
i
n
e
lear
n
in
g
-
b
ased
to
p
ic
m
o
d
elin
g
with
r
o
b
u
s
t m
eth
o
d
s
f
o
r
u
n
ce
r
tain
ty
m
an
ag
em
en
t.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
I
d
en
tifi
ca
tio
n
o
f
a
r
ea
s
o
f in
flu
en
ce
u
s
in
g
a
th
ema
tic
mo
d
elin
g
a
p
p
r
o
a
ch
a
n
d
b
elief
…
(
F
a
ti
ma
-
Za
h
r
a
e
S
ifi)
2839
T
o
ad
d
r
ess
th
is
g
ap
,
we
p
r
o
p
o
s
e
a
n
o
v
el
ar
tific
ial
in
tellig
en
ce
(
AI
)
ap
p
r
o
ac
h
.
I
t
u
s
es
laten
t
d
ir
ich
let
allo
ca
tio
n
(
L
DA)
[
1
0
]
,
a
g
e
n
er
ativ
e
p
r
o
b
ab
ilis
tic
m
o
d
el
ap
p
lied
in
m
ac
h
in
e
lear
n
in
g
f
o
r
to
p
ic
d
is
co
v
er
y
.
I
t
also
em
p
lo
y
s
b
elief
f
u
n
ctio
n
th
eo
r
y
(
B
FT)
,
s
p
ec
if
ically
th
e
Dem
p
s
ter
-
Sh
af
er
f
r
am
ew
o
r
k
[
1
1
]
,
to
ad
r
ess
u
n
ce
r
tain
ty
in
in
f
lu
en
ce
r
b
e
h
av
io
r
a
n
aly
s
is
.
T
h
is
in
teg
r
atio
n
is
o
u
r
m
ain
co
n
tr
ib
u
tio
n
to
AI
.
I
t
co
m
b
in
es
a
p
r
o
b
a
b
ilis
tic
g
en
er
ativ
e
m
o
d
el
with
ev
id
en
ce
f
u
s
io
n
tech
n
iq
u
es.
T
h
e
g
o
al
is
to
ac
h
ie
v
e
a
m
o
r
e
r
o
b
u
s
t
id
en
tific
atio
n
o
f
d
o
m
ai
n
-
s
p
e
cif
ic
in
f
lu
en
ce
.
O
u
r
ap
p
r
o
ac
h
in
co
r
p
o
r
ates
ess
en
tial
en
g
ag
em
en
t
in
d
icato
r
s
,
s
u
ch
as
r
etwe
ets,
lik
es,
an
d
m
en
tio
n
s
.
I
t
p
r
o
v
id
es
a
m
u
ltid
im
en
s
io
n
al
an
d
co
n
tex
t
-
s
en
s
itiv
e
p
er
s
p
ec
tiv
e
o
n
in
f
l
u
en
ce
.
T
h
e
p
r
o
p
o
s
ed
f
r
am
ewo
r
k
co
n
tr
ib
u
tes
b
o
th
th
eo
r
etica
lly
an
d
p
r
ac
tically
.
Fro
m
a
r
esear
ch
p
er
s
p
ec
tiv
e,
it
in
tr
o
d
u
ce
s
an
in
n
o
v
ativ
e
c
o
m
b
in
atio
n
o
f
p
r
o
b
a
b
ilis
tic
m
o
d
elin
g
an
d
b
elief
f
u
s
io
n
.
T
h
is
ap
p
r
o
ac
h
ad
d
r
ess
es u
n
ce
r
tain
t
y
in
s
o
cial
in
f
lu
en
ce
an
aly
s
is
.
Fro
m
an
ap
p
licatio
n
p
er
s
p
ec
tiv
e,
it su
p
p
o
r
ts
m
o
r
e
ef
f
ec
tiv
e
s
tr
ateg
ies
in
m
ar
k
eti
n
g
,
o
p
in
io
n
m
i
n
in
g
an
d
d
ig
it
al
co
m
m
u
n
icatio
n
.
I
t
also
h
el
p
s
au
d
ien
ce
s
b
etter
u
n
d
er
s
tan
d
h
o
w
in
f
lu
en
ce
r
s
g
u
id
e
p
u
b
lic
d
is
co
u
r
s
e.
T
h
e
s
tr
u
ctu
r
e
o
f
th
is
p
ap
er
i
s
as
f
o
llo
ws.
Sectio
n
2
r
ev
i
ews
p
r
ev
io
u
s
r
esear
ch
in
t
h
is
d
o
m
ain
.
Sectio
n
3
d
escr
ib
es
th
e
p
r
o
p
o
s
ed
m
eth
o
d
.
I
t
in
te
g
r
ates
L
DA
an
d
B
FT
to
id
en
tify
an
d
ev
alu
ate
th
e
d
o
m
ai
n
s
o
f
g
r
ea
test
in
f
lu
en
ce
r
im
p
ac
t.
Se
ctio
n
4
p
r
esen
ts
an
d
d
is
cu
s
s
es
r
esu
lts
f
r
o
m
two
ca
s
e
s
tu
d
ies.
Fin
ally
,
s
ec
tio
n
5
co
n
clu
d
es th
e
s
tu
d
y
an
d
o
u
tlin
es p
r
o
s
p
ec
t d
ir
ec
tio
n
s
f
o
r
f
u
tu
r
e
r
esear
ch
.
2.
RE
L
AT
E
D
WO
RK
S
R
ec
en
t
ad
v
an
ce
s
in
AI
h
av
e
a
u
g
m
en
ted
th
e
n
ee
d
f
o
r
co
n
s
is
ten
t
m
eth
o
d
s
to
m
an
ag
e
u
n
ce
r
t
ain
ty
.
T
h
is
is
tr
u
e
f
o
r
in
co
m
p
lete,
am
b
i
g
u
o
u
s
,
o
r
in
co
n
s
is
ten
t
d
ata.
Pro
b
ab
ilis
tic
f
r
am
ewo
r
k
s
[
1
2
]
ar
e
co
m
m
o
n
,
b
u
t
th
ey
f
r
eq
u
e
n
tly
f
ail
to
r
ep
r
esen
t
ig
n
o
r
an
ce
o
r
p
ar
tial
in
f
o
r
m
atio
n
clea
r
ly
.
T
h
is
lim
itatio
n
h
as
ex
ten
d
ed
in
ter
est
i
n
ap
p
r
o
ac
h
es su
ch
as De
m
p
s
ter
-
Sh
af
er
th
eo
r
y
(
DST)
[
1
3
]
.
T
h
e
DST
f
r
am
ewo
r
k
was
in
tr
o
d
u
ce
d
b
y
Dem
p
s
ter
an
d
f
o
r
m
alize
d
b
y
Sh
af
er
.
I
t
s
u
p
p
o
r
ts
b
elief
to
b
e
ap
p
o
in
ted
to
s
ets
o
f
h
y
p
o
th
e
s
es
in
s
tead
o
f
ex
ac
t
p
r
o
b
ab
il
ities
.
T
h
is
m
ak
es
it
ef
f
ec
tiv
e
in
th
e
in
ter
est
o
f
b
o
th
u
n
ce
r
tain
ty
an
d
ig
n
o
r
an
ce
in
ar
ea
s
s
u
c
h
as
s
en
s
o
r
f
u
s
io
n
[
1
4
]
,
p
atter
n
r
ec
o
g
n
itio
n
[
1
5
]
,
a
n
d
c
o
m
p
lex
d
ec
is
io
n
-
m
ak
in
g
[
1
6
]
.
I
ts
k
ey
co
m
p
o
n
e
n
ts
b
asic
b
elief
ass
i
g
n
m
en
t
(
B
B
A)
,
b
elief
an
d
p
l
au
s
ib
ilit
y
m
ea
s
u
r
es
an
d
Dem
p
s
ter
’
s
r
u
le
p
r
o
v
id
e
a
s
tr
u
ctu
r
ed
way
t
o
m
an
a
g
e
am
b
ig
u
ity
an
d
in
co
m
p
atib
le
in
f
o
r
m
atio
n
[
1
7
]
.
R
ec
en
t
wo
r
k
h
as
f
o
cu
s
ed
t
o
r
e
in
f
o
r
ce
th
e
t
h
eo
r
etica
l
f
o
u
n
d
at
io
n
s
o
f
DST.
On
e
s
tu
d
y
[
1
8
]
i
n
tr
o
d
u
ce
d
n
ew
d
ef
in
itio
n
s
o
f
co
n
d
itio
n
al
b
elief
f
u
n
ctio
n
s
a
n
d
g
r
ap
h
ical
m
o
d
els
s
im
ilar
to
B
ay
esian
n
etwo
r
k
s
.
T
h
is
im
p
r
o
v
e
d
co
n
d
itio
n
al
r
ea
s
o
n
in
g
.
T
h
e
au
th
o
r
s
o
f
[
1
9
]
ex
a
m
in
ed
th
e
co
n
ce
p
t
o
f
“d
is
tin
ct”
b
elief
f
u
n
ctio
n
s
.
T
h
ey
clar
if
ied
th
e
in
d
ep
e
n
d
en
ce
co
n
d
itio
n
s
n
ee
d
e
d
to
p
r
o
p
er
l
y
ap
p
ly
Dem
p
s
ter
’
s
co
m
b
in
atio
n
r
u
le.
Dem
p
s
ter
’
s
o
r
ig
in
al
r
u
le
is
eleg
an
t b
u
t p
er
f
o
r
m
s
p
o
o
r
ly
in
h
i
g
h
-
co
n
f
lict sit
u
atio
n
s
.
T
o
ad
d
r
ess
th
is
,
alter
n
ativ
e
co
m
b
in
atio
n
r
u
les
an
d
en
tr
o
p
y
-
b
ased
m
ea
s
u
r
es
h
av
e
b
ee
n
d
ev
elo
p
e
d
.
On
e
ap
p
r
o
ac
h
is
a
R
én
y
i
-
s
ty
le
b
elief
en
tr
o
p
y
[
2
0
]
.
I
t
im
p
r
o
v
es
u
n
ce
r
tain
ty
m
ea
s
u
r
em
e
n
t
to
s
u
p
p
o
r
t
d
ec
is
io
n
-
m
ak
in
g
.
I
n
p
a
r
a
llel,
r
ec
en
t
m
o
d
els
th
at
co
m
b
in
e
DST
with
to
p
o
lo
g
ical
ev
id
en
ce
f
r
a
m
ewo
r
k
s
p
r
o
v
id
e
b
etter
co
n
f
lict
m
an
ag
e
m
en
t.
T
h
ey
ap
p
r
o
v
e
b
elief
co
m
p
u
tatio
n
to
a
d
ju
s
t in
ac
co
r
d
an
ce
with
d
ec
is
io
n
p
r
io
r
ities
[
2
1
]
.
Fu
r
th
er
r
esear
ch
h
as
in
teg
r
ated
DST
with
to
p
o
lo
g
ical
m
o
d
e
ls
to
im
p
r
o
v
e
co
n
f
lict
m
an
ag
e
m
en
t.
T
h
e
au
th
o
r
o
f
[
2
2
]
i
n
tr
o
d
u
ce
s
f
lex
i
b
le
b
elief
ar
ch
itectu
r
es.
T
h
ese
ar
ch
itectu
r
es
a
d
ju
s
t
b
elief
ca
lcu
latio
n
s
b
ased
o
n
d
ec
is
io
n
-
m
ak
in
g
p
r
io
r
ities
.
T
h
is
f
ac
ilit
ate
s
DST
to
ad
ap
t
to
p
ar
ticu
lar
g
o
als,
s
u
ch
as
r
ed
u
c
e
f
alse
p
o
s
itiv
es
o
r
f
alse n
eg
ativ
es.
I
t p
r
o
v
id
es m
o
r
e
f
o
cu
s
ed
s
o
lu
ti
o
n
s
in
co
m
p
le
x
d
ec
is
io
n
-
m
a
k
in
g
c
o
n
tex
ts
.
I
n
d
ec
is
io
n
th
eo
r
y
,
DST
h
as
in
f
lu
en
ce
d
m
o
d
els
th
a
t
s
ep
ar
ate
b
elief
a
g
g
r
e
g
atio
n
f
r
o
m
d
ec
is
io
n
-
m
ak
in
g
.
T
h
e
tr
an
s
f
er
ab
le
b
elief
m
o
d
el
(
T
B
M)
[
2
3
]
d
ef
in
es
two
s
tag
es:
th
e
c
r
ed
a
l
lev
el
b
elief
s
an
d
th
e
p
ig
n
is
tic
lev
el
s
u
p
p
o
r
ts
d
e
cisi
o
n
-
m
ak
in
g
.
A
m
o
d
er
n
DST
-
b
ased
m
o
d
el
ad
d
r
ess
es
ev
id
en
tial
u
n
ce
r
tain
ty
.
I
t
p
r
o
v
id
es
a
s
tr
u
ctu
r
ed
m
a
n
n
er
to
s
u
p
p
o
r
t
r
atio
n
al
d
ec
is
io
n
s
in
s
itu
atio
n
s
n
ee
d
s
q
u
ick
ju
d
g
m
en
t
[
2
4
]
.
T
h
is
ex
ten
d
s
DST’
s
u
s
e
to
s
ce
n
ar
io
s
with
in
co
m
p
lete
o
r
a
m
b
ig
u
o
u
s
in
f
o
r
m
atio
n
[
2
5
]
.
DST
h
as
b
ee
n
u
s
ed
b
r
o
a
d
ly
in
p
r
ac
tical
f
ield
s
.
I
n
s
en
s
o
r
n
et
wo
r
k
s
an
d
th
e
in
te
r
n
et
o
f
t
h
in
g
s
(
I
o
T
)
,
it
s
u
p
p
o
r
ts
co
m
p
lex
ev
en
t
a
n
al
y
s
is
.
I
t
co
m
b
in
es
d
ata
f
r
o
m
m
u
ltip
le
s
en
s
o
r
s
an
d
co
n
tr
o
ls
u
n
ce
r
tain
ty
an
d
co
n
f
lict
in
d
y
n
am
ic
e
n
v
ir
o
n
m
en
ts
[
2
6
]
.
DST
is
also
u
s
ed
i
n
d
ee
p
lea
r
n
in
g
f
o
r
m
u
ltimo
d
al
lear
n
in
g
.
I
t
h
elp
s
m
o
d
els
in
teg
r
ate
in
f
o
r
m
atio
n
f
r
o
m
s
o
u
r
ce
s
lik
e
im
ag
es
an
d
tex
t,
p
r
in
cip
ally
wh
e
n
d
at
a
is
in
co
m
p
lete
o
r
co
n
cu
r
r
en
t.
T
h
is
is
u
s
ef
u
l in
ta
s
k
s
s
u
ch
as im
ag
e
d
escr
ip
tio
n
s
an
d
s
en
tim
en
t a
n
al
y
s
is
[
2
7
]
.
I
n
co
n
clu
s
io
n
,
DST
co
n
tin
u
es
to
g
r
o
w
as
a
p
o
wer
f
u
l
m
ec
h
an
is
m
f
o
r
u
n
ce
r
tain
ty
m
a
n
ag
e
m
en
t.
I
t
is
u
s
ef
u
l
wh
er
e
tr
ad
itio
n
al
p
r
o
b
ab
ilit
y
m
o
d
els
f
all
s
h
o
r
t.
I
t
s
f
lex
ib
ilit
y
s
u
p
p
o
r
ts
ap
p
licatio
n
s
in
in
f
er
en
ce
,
co
n
f
lict r
eso
lu
tio
n
a
n
d
d
ec
is
io
n
-
m
ak
in
g
.
T
h
is
d
em
o
n
s
tr
ates its
r
is
e
r
o
le
in
AI
an
d
d
ec
is
io
n
s
cien
ce
.
3.
M
E
T
H
O
D
T
h
i
s
s
t
u
d
y
p
r
o
p
o
s
e
s
a
r
i
g
o
r
o
u
s
a
n
d
a
u
t
o
m
a
t
e
d
a
p
p
r
o
a
c
h
t
o
i
d
e
n
t
i
f
y
t
h
e
s
p
e
c
i
f
i
c
d
o
m
a
in
s
w
h
e
r
e
i
n
f
l
u
e
n
c
e
r
s
h
a
v
e
t
h
e
g
r
e
at
e
s
t
im
p
a
c
t
.
I
t
c
o
m
b
i
n
e
s
L
D
A
[
2
8
]
to
e
x
t
r
a
c
t
t
h
e
m
a
i
n
t
o
p
i
c
s
f
r
o
m
i
n
f
l
u
e
n
c
e
r
-
g
e
n
e
r
a
t
e
d
c
o
n
t
e
n
t
,
s
p
e
c
i
f
i
c
a
ll
y
t
w
e
et
s
.
I
t
a
l
s
o
u
s
es
B
F
T
[
2
3
]
t
o
e
v
a
l
u
a
t
e
u
s
e
r
e
n
g
a
g
e
m
e
n
t
.
Fig
u
r
e
1
ill
u
s
tr
ates
th
e
g
lo
b
al
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
26
:
2
8
3
8
-
2
8
4
8
2840
ar
ch
itectu
r
e
o
f
th
e
p
r
o
p
o
s
ed
d
o
m
ain
in
f
lu
en
ce
id
en
tific
atio
n
p
r
o
ce
s
s
.
I
t
s
h
o
ws
ea
c
h
s
tep
f
r
o
m
d
ata
co
llectio
n
to
th
e
id
en
tific
atio
n
o
f
th
e
m
ain
d
o
m
ain
o
f
in
f
lu
e
n
ce
.
L
DA
id
en
tifie
s
th
e
m
ain
to
p
ics
in
in
f
lu
en
ce
r
s
’
d
is
co
u
r
s
e.
B
elief
th
eo
r
y
an
aly
ze
s
in
ter
ac
tio
n
s
s
u
ch
as
r
etwe
ets,
m
en
tio
n
s
,
an
d
lik
es
to
m
e
asu
r
e
co
n
f
id
e
n
ce
in
ea
ch
to
p
ic.
T
h
e
co
m
b
in
atio
n
o
f
th
ese
m
eth
o
d
s
ex
p
o
s
es
b
o
th
th
e
co
n
te
n
t
an
d
h
o
w
i
t
r
eso
n
ates
in
th
e
co
m
m
u
n
ity
.
T
h
is
p
r
o
v
id
es a
p
r
ec
is
e
r
ep
r
esen
tatio
n
o
f
in
f
lu
e
n
ce
r
s
’
ar
ea
s
o
f
i
n
f
lu
en
ce
.
Fig
u
r
e
1
.
Glo
b
al
ar
ch
itectu
r
e
f
o
r
in
f
lu
e
n
ce
r
ec
o
g
n
itio
n
with
i
n
th
e
d
o
m
ain
3
.
1
.
F
o
rma
l
f
ra
m
ewo
r
k
o
f
o
ur
m
et
ho
d
L
et
Ω
r
e
p
r
esen
t
th
e
s
et
o
f
p
o
ten
tial
to
p
ics
o
r
th
em
atic
a
r
e
as
r
elev
an
t
to
a
g
iv
en
in
f
l
u
en
ce
r
.
Fo
r
a
s
p
ec
if
ic
in
f
lu
en
ce
r
,
Ω
in
cl
u
d
e
to
p
ics
s
u
ch
as
{
p
o
liti
cs,
tech
n
o
lo
g
y
,
ec
o
n
o
m
y
,
an
d
h
ea
lth
}
.
W
e
d
ef
in
e
a
s
et
o
f
twee
ts
=
{
1
,
2
,
.
.
.
,
}
p
o
s
ted
with
in
a
tim
e
win
d
o
w
[
,
]
,
wh
er
e
ea
ch
is
a
twee
t
p
u
b
lis
h
ed
at
a
tim
estam
p
with
in
th
is
in
ter
v
al
as in
(
1
)
.
=
{
|
≤
(
)
≤
}
(
1
)
Fo
r
ea
ch
twee
t
we
c
o
n
s
id
er
en
g
ag
em
e
n
t
m
etr
ics
,
wh
e
r
e
th
e
n
u
m
b
er
o
f
r
etwe
ets
=
(
)
,
lik
es
=
(
)
an
d
m
e
n
tio
n
s
=
(
)
.
T
h
ese
m
etr
ic
s
q
u
an
tify
b
o
th
t
h
e
in
f
lu
e
n
ce
r
'
s
en
g
ag
em
en
t
an
d
th
e
twee
t'
s
im
p
ac
t o
n
th
e
s
o
cial
n
etwo
r
k
.
3
.
2
.
T
hem
a
t
ic
m
o
delin
g
us
ing
la
t
ent
dirichlet
a
llo
ca
t
io
n
T
o
p
r
ep
a
r
e
th
e
d
ata
f
o
r
a
n
aly
s
is
,
we
f
ir
s
t
p
r
ep
r
o
ce
s
s
twee
ts
b
y
r
em
o
v
in
g
r
etwe
ets,
UR
L
s
,
h
ash
tag
s
,
p
u
n
ctu
atio
n
an
d
s
to
p
w
o
r
d
s
.
T
h
e
tex
t
is
th
en
lo
wer
ca
s
ed
,
lem
m
atize
d
an
d
to
k
en
ized
.
T
h
is
clea
n
ed
d
ataset
f
o
r
m
s
th
e
b
asis
to
u
s
e
L
DA
to
id
en
tify
an
d
class
if
y
p
r
im
ar
y
to
p
ics
in
th
e
twee
ts
.
Fig
u
r
e
2
illu
s
tr
ates
th
e
p
r
o
ce
s
s
in
g
p
ip
elin
e
.
Fig
u
r
e
2
.
Seq
u
en
tial p
r
o
ce
s
s
in
g
f
o
r
t
h
e
an
aly
s
is
an
d
i
n
ter
p
r
et
atio
n
o
f
T
wee
ts
3
.
2
.
1
.
Da
t
a
prepro
ce
s
s
ing
Pre
p
r
o
ce
s
s
in
g
[
2
9
]
s
tar
ts
b
y
r
e
m
o
v
in
g
r
etwe
ets to
f
o
cu
s
o
n
l
y
on
o
r
ig
in
al
co
n
ten
t
,
as sh
o
wn
in
(
2
)
.
=
{
∈
/
∉
}
(
2
)
Su
b
s
eq
u
en
t step
s
in
clu
d
e
:
–
UR
L
r
em
o
v
al:
=
{
∈
∣
∣
}
–
R
em
o
v
al
o
f
h
ash
tag
s
a
n
d
s
p
ec
ial
ch
ar
ac
ter
s
:
ℎ
=
{
∈
∣
∣
ℎ
}
–
Pu
n
ctu
atio
n
r
em
o
v
al:
=
{
∈
ℎ
∣
}
–
Sto
p
wo
r
d
r
em
o
v
al:
=
{
∈
∣
}
–
L
o
wer
ca
s
in
g
:
=
{
∈
∣
∣
(
)
}
–
L
em
m
atiza
tio
n
/
s
tem
m
in
g
:
=
{
∈
∣
(
)
}
–
T
o
k
en
izatio
n
:
=
{
∈
∣
∣
(
)
}
–
No
is
e
r
ed
u
ctio
n
:
=
{
∈
|
ℎ
}
3
.
2
.
2
.
T
wee
t
cla
s
s
if
ica
t
io
n us
ing
la
t
ent
dirichlet
a
llo
ca
t
io
n
L
DA
is
ap
p
lied
to
d
is
co
v
er
la
ten
t
to
p
ics
with
in
twee
ts
[
3
0
]
.
L
DA
tr
ea
ts
ea
ch
twee
t
as
a
m
ix
tu
r
e
o
f
to
p
ics
[
3
1
]
a
n
d
ea
ch
to
p
ic
as
a
d
is
tr
ib
u
tio
n
o
v
er
th
e
v
o
ca
b
u
lar
y
.
E
ac
h
t
o
p
ic
s
is
ass
o
ciate
d
with
a
p
r
o
b
a
b
ilit
y
d
is
tr
ib
u
tio
n
o
v
er
wo
r
d
s
as in
(
3
)
.
T
w
e
e
t
s
P
r
e
p
r
o
c
e
ssi
n
g
L
DA
mo
d
e
l
i
n
g
F
u
sio
n
w
i
t
h
b
e
l
i
e
f
t
h
e
o
r
y
Ra
n
k
i
n
g
an
d
i
n
t
e
r
p
r
e
t
at
i
o
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
I
d
en
tifi
ca
tio
n
o
f
a
r
ea
s
o
f in
flu
en
ce
u
s
in
g
a
th
ema
tic
mo
d
elin
g
a
p
p
r
o
a
ch
a
n
d
b
elief
…
(
F
a
ti
ma
-
Za
h
r
a
e
S
ifi)
2841
=
(
,
1
,
,
2
,
…
,
,
|
|
)
∼
ℎ
(
)
(
3
)
W
h
er
e
co
n
tr
o
ls
to
p
ic
-
wo
r
d
s
p
ar
s
ity
an
d
r
ef
er
s
to
th
e
v
o
ca
b
u
lar
y
.
E
ac
h
twee
t
∈
is
m
o
d
eled
as a
to
p
ic
d
is
tr
ib
u
tio
n
as in
(
4
)
.
=
(
,
1
,
,
2
,
…
,
,
|
Ω
|
)
∼
ℎ
(
)
(
4
)
W
ith
is
u
s
ed
t
o
c
o
n
tr
o
l
to
p
ic
d
iv
er
s
ity
with
in
twee
ts
.
T
we
ets
ar
e
th
u
s
class
if
ied
b
y
th
ei
r
d
o
m
in
an
t
to
p
ics
b
ased
o
n
th
r
esh
o
l
d
ed
to
p
ic
p
r
o
b
ab
ilit
ies.
T
h
is
f
r
am
ewo
r
k
e
n
ab
les
u
s
to
m
ap
twee
ts
to
m
ea
n
in
g
f
u
l
to
p
ics.
I
t
p
r
o
v
id
es in
f
o
r
m
atio
n
ab
o
u
t in
f
lu
en
ce
r
s
’
to
p
ic
p
r
ef
er
e
n
ce
s
an
d
d
is
co
u
r
s
e
p
atter
n
s
.
3
.
3
.
T
hem
a
t
ic
influence
a
nd
eng
a
g
em
ent
m
et
rics
On
ce
L
DA
h
as
b
ee
n
a
p
p
lied
f
o
r
to
p
ic
m
o
d
elin
g
,
we
q
u
an
tif
y
au
d
ien
ce
im
p
ac
t
by
m
ea
n
s
o
f
m
etr
ics.
T
h
em
atic
m
ass
m
ea
s
u
r
es
to
p
i
c
p
r
ev
alen
ce
,
wh
ile
en
g
ag
em
e
n
t
m
ass
es
(
r
etwe
et
s
,
lik
es,
an
d
m
en
tio
n
s
)
r
ef
lect
au
d
ien
ce
in
ter
ac
ti
o
n
.
T
o
h
o
l
d
u
n
ce
r
tain
ty
in
s
o
cial
in
ter
ac
tio
n
s
,
we
u
s
e
B
FT
to
m
o
d
el
co
n
f
i
d
en
ce
in
in
f
o
r
m
atio
n
s
o
u
r
ce
s
.
T
h
is
en
ab
les a
m
o
r
e
r
o
b
u
s
t e
v
alu
atio
n
o
f
to
p
ic
in
f
lu
en
ce
.
3
.
3
.
1
.
T
hem
a
t
ic
ma
s
s
estim
a
t
io
n
On
ce
th
e
L
DA
p
ar
am
eter
s
ar
e
esti
m
ated
,
ea
ch
twee
t
∈
is
r
ep
r
esen
ted
b
y
a
to
p
ic
d
is
tr
ib
u
tio
n
=
(
(
1
)
,
(
2
)
,
.
.
.
,
(
)
)
,
wh
er
e
(
)
d
en
o
tes
th
e
p
r
o
p
o
r
tio
n
o
f
co
n
ten
t
in
twee
t
attr
ib
u
te
d
to
to
p
ic
∈
,
s
atis
f
y
in
g
∑
(
)
∈
Ω
=
1
.
T
o
q
u
an
tify
th
e
o
v
er
all
p
r
o
m
in
e
n
ce
o
f
a
g
iv
en
to
p
ic
with
in
th
e
co
r
p
u
s
,
we
d
ef
in
e
th
e
t
h
em
atic
m
ass
(
)
.
T
h
is
m
ea
s
u
r
e
ag
g
r
eg
ate
m
e
asu
r
es
th
e
weig
h
te
d
p
r
esen
ce
o
f
th
e
t
o
p
ic
ac
r
o
s
s
all
twee
ts
as in
(
5
)
.
(
)
=
∑
(
)
∈
∑
∑
(
′
)
′
∈
Ω
∈
(
5
)
T
h
is
v
alu
e
ca
p
tu
r
es
th
e
r
elati
v
e
im
p
o
r
ta
n
ce
an
d
p
r
e
v
alen
ce
o
f
to
p
ic
in
th
e
d
ataset.
Hig
h
er
v
alu
es
in
d
icat
e
g
r
ea
ter
th
em
atic
d
o
m
in
an
ce
.
3
.
3
.
2
.
Ret
wee
t
m
a
s
s
estim
a
t
i
o
n
L
et
r
ep
r
esen
t
t
h
e
n
u
m
b
e
r
o
f
r
etwe
ets
ass
o
ciate
d
with
twee
t
.
T
o
e
v
alu
ate
th
e
lev
el
o
f
a
u
d
ien
ce
en
g
ag
em
e
n
t w
ith
ea
ch
to
p
ic,
we
d
ef
in
e
th
e
r
etwe
et
m
ass
(
)
as
in
(
6
)
.
(
)
=
∑
(
)
∈
∑
∈
(
6
)
T
h
is
m
etr
ic
q
u
a
n
tifie
s
th
e
r
el
ativ
e
s
h
ar
e
o
f
r
etwe
et
-
d
r
i
v
en
atten
tio
n
allo
ca
ted
to
to
p
ic
ac
r
o
s
s
th
e
co
r
p
u
s
.
I
n
co
n
tr
ast
t
o
th
e
m
atic
m
ass
,
r
etwe
et
m
ass
em
p
h
asizes
th
e
d
if
f
u
s
io
n
p
o
ten
tial
o
f
to
p
ics.
A
h
ig
h
(
)
in
d
icate
s
th
at
to
p
ic
is
f
r
eq
u
en
tly
ass
o
ciate
d
with
twee
ts
.
I
t
t
h
at
g
en
er
ate
s
u
b
s
tan
tial
en
g
a
g
em
en
t,
r
eg
ar
d
less
o
f
th
e
to
tal
n
u
m
b
e
r
o
f
twee
ts
.
3
.
3
.
3
.
L
ik
e
ma
s
s
estim
a
t
io
n
C
o
r
r
esp
o
n
d
in
g
ly
,
let
r
ep
r
esen
t
th
e
n
u
m
b
er
o
f
lik
es
r
ec
eiv
ed
b
y
twee
t
.
T
o
q
u
a
n
tify
th
e
lev
el
o
f
p
o
s
itiv
e
au
d
ien
ce
e
n
g
ag
e
m
en
t
with
ea
ch
to
p
ic,
we
d
ef
in
e
th
e
lik
e
m
ass
(
)
as
in
(
7
)
.
(
)
=
∑
(
)
∈
∑
∈
(
7
)
T
h
is
m
etr
ic
ca
p
tu
r
es
th
e
em
o
tio
n
al
o
r
af
f
ir
m
ativ
e
r
esp
o
n
s
e
elicited
b
y
to
p
ic
ac
r
o
s
s
th
e
c
o
r
p
u
s
.
A
h
ig
h
er
(
)
v
alu
e
in
d
icate
s
th
at
twee
ts
as
s
o
ciate
d
with
to
p
ic
ten
d
to
r
ec
eiv
e
g
r
ea
ter
u
s
er
a
p
p
r
ec
iatio
n
.
T
h
is
s
u
g
g
ests
s
tr
o
n
g
er
r
eso
n
a
n
ce
o
r
a
g
r
ee
m
en
t w
ith
in
th
e
a
u
d
ien
ce
.
3
.
3
.
4
.
M
ent
io
n m
a
s
s
estim
a
t
i
o
n
T
h
e
m
en
tio
n
m
ass
(
)
q
u
an
tifie
s
th
e
s
ig
n
if
ican
ce
o
f
to
p
ic
b
ased
o
n
th
e
f
r
eq
u
en
c
y
o
f
its
ex
p
licit
r
ef
er
en
ce
s
,
s
u
c
h
as
h
ash
tag
s
o
r
k
e
y
wo
r
d
s
,
with
in
th
e
twee
t
co
r
p
u
s
.
Fo
r
ea
c
h
twee
t
,
we
co
m
b
in
e
its
to
p
i
c
p
r
o
p
o
r
tio
n
(
)
with
th
e
n
u
m
b
er
o
f
m
en
tio
n
s
,
an
d
d
ef
in
e
th
e
m
e
n
tio
n
m
ass
as
in
(
8
)
.
(
)
=
∑
(
)
∈
∑
∈
(
8
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
26
:
2
8
3
8
-
2
8
4
8
2842
T
h
is
m
etr
ic
co
m
b
in
es
b
o
th
th
e
th
em
atic
r
elev
an
ce
o
f
to
p
ic
with
in
in
d
iv
id
u
al
twee
ts
an
d
th
e
in
ten
s
ity
o
f
its
ex
p
licit
m
e
n
tio
n
.
T
wee
ts
with
h
i
g
h
(
)
v
alu
e
s
an
d
n
u
m
er
o
u
s
m
en
tio
n
s
c
o
n
tr
ib
u
te
m
o
r
e
s
ig
n
if
ican
tly
to
(
)
.
No
r
m
aliza
tio
n
b
y
to
tal
m
en
tio
n
s
en
s
u
r
es
v
a
lid
co
m
p
ar
is
o
n
s
ac
r
o
s
s
to
p
ics.
3
.
4
.
M
a
s
s
f
us
io
n v
ia
s
equent
ia
l c
o
njunct
iv
e
f
us
io
n r
ule
T
o
o
b
tain
a
co
m
p
r
e
h
en
s
iv
e
m
ea
s
u
r
e
o
f
to
p
ic
s
ig
n
if
ican
ce
,
we
co
m
b
in
e
th
e
m
ass
f
u
n
ctio
n
s
—
(
)
,
(
)
,
(
)
,
an
d
(
)
—
th
r
o
u
g
h
th
e
s
eq
u
en
ti
al
co
n
ju
n
ctiv
e
f
u
s
io
n
r
u
le
f
r
o
m
b
elief
th
eo
r
y
.
T
h
is
m
eth
o
d
in
teg
r
ates
ev
id
en
ce
ac
r
o
s
s
d
if
f
er
en
t
d
im
en
s
io
n
s
.
I
t
im
p
r
o
v
es
th
e
alig
n
m
en
t
b
etwe
e
n
th
em
at
ic
an
d
en
g
a
g
em
en
t
m
etr
ics.
T
h
e
s
tep
-
by
-
s
tep
f
u
s
io
n
p
r
o
ce
s
s
u
s
es
co
n
f
lict
c
o
ef
f
icien
ts
to
tack
le
u
n
ce
r
tai
n
ty
.
T
h
is
y
ield
s
a
d
ep
en
d
a
b
le
g
lo
b
al
in
f
lu
en
ce
s
co
r
e
f
o
r
ea
ch
to
p
ic.
i)
Fu
s
i
o
n
o
f
t
h
em
ati
c
an
d
r
etw
ee
t
m
ass
es
as
i
n
(
9
)
.
(
)
=
(
)
⊕
(
)
=
1
1
−
∑
(
)
.
(
)
∩
=
(
9
)
W
ith
c
o
n
f
l
ict
c
o
ef
f
i
cie
n
t
as
(
1
0
)
.
=
∑
(
)
.
(
)
∩
=
∅
(
1
0
)
ii)
Fu
s
i
o
n
w
it
h
l
ik
e
as
i
n
(
1
1
)
a
n
d
(
1
2
)
.
(
)
=
(
)
⊕
(
)
=
1
1
−
∑
(
)
.
(
)
∩
=
(
1
1
)
=
∑
(
)
.
(
)
∩
=
∅
(
1
2
)
iii)
Fu
s
i
o
n
w
it
h
m
e
n
t
io
n
m
ass
as i
n
(
1
3
)
a
n
d
(
1
4
)
.
(
)
=
(
)
⊕
(
)
=
1
1
−
∑
(
)
.
(
)
∩
=
(
1
3
)
=
∑
(
)
.
(
)
∩
=
∅
(
1
4
)
At
ea
ch
f
u
s
io
n
s
tep
,
th
e
co
n
f
lict
co
ef
f
icien
t
q
u
an
tifie
s
th
e
d
eg
r
ee
o
f
i
n
co
n
s
is
ten
cy
b
et
wee
n
th
e
d
if
f
er
en
t
m
ass
s
o
u
r
ce
s
.
T
h
is
co
ef
f
icien
t
en
ab
les
th
e
r
u
le
t
o
ad
ju
s
t
co
n
f
lict
ev
id
en
ce
.
I
t
en
s
u
r
es
a
b
alan
ce
d
in
teg
r
atio
n
o
f
th
e
m
etr
ics.
T
h
e
f
in
al
m
ass
(
)
,
r
ef
lects
th
e
ag
g
r
eg
ated
in
f
lu
e
n
ce
o
f
t
o
p
ic
.
T
h
e
to
p
ic
with
th
e
h
ig
h
est
(
)
is
id
en
tifie
d
as
th
e
d
o
m
in
an
t
in
f
lu
en
ce
with
in
th
e
co
r
p
u
s
,
r
ef
lectin
g
its
o
v
er
all
p
r
o
m
in
e
n
ce
ac
r
o
s
s
all
d
im
en
s
i
o
n
s
o
f
en
g
ag
em
e
n
t.
3
.
5
.
P
i
g
nis
t
ic
pro
ba
bil
it
y
T
o
r
a
n
k
to
p
ics
b
y
th
eir
in
f
lu
e
n
ce
,
we
co
m
p
u
te
th
e
p
ig
n
is
tic
p
r
o
b
ab
ilit
y
(
(
)
)
,
wh
ich
co
n
v
er
ts
th
e
f
in
al
b
elief
m
ass
in
to
a
p
r
o
b
ab
ilit
y
d
is
tr
ib
u
ti
o
n
o
v
er
in
d
iv
id
u
al
to
p
ic
.
T
h
e
p
ig
n
is
tic
p
r
o
b
a
b
ilit
y
is
d
ef
in
ed
as
in
(
1
5
)
.
(
)
=
∑
(
)
|
|
∋
(
1
5
)
W
h
er
e
∋
d
en
o
tes all
s
u
b
s
ets co
n
tain
in
g
to
p
ic
an
d
∣
∣
r
ep
r
esen
ts
th
e
ca
r
d
in
ality
o
f
s
u
b
s
et
.
T
h
is
m
ea
s
u
r
e
r
ef
lects
th
e
ex
p
ec
ted
b
elief
a
r
atio
n
al
ag
en
t
wo
u
ld
ass
ig
n
to
to
p
ic
wh
en
m
ak
in
g
d
ec
is
io
n
s
u
n
d
er
u
n
ce
r
tain
ty
.
A
h
ig
h
er
(
)
in
d
icate
s
a
g
r
ea
ter
in
f
er
r
ed
in
f
lu
e
n
ce
o
f
to
p
ic
.
T
h
er
ef
o
r
e,
th
e
d
o
m
in
an
t i
n
f
lu
en
ce
d
o
m
ai
n
is
d
eter
m
in
ed
as th
e
t
o
p
ic
with
t
h
e
m
ax
im
u
m
(
)
.
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
I
n
th
is
s
ec
tio
n
,
it
is
ex
p
lain
ed
th
e
r
esu
lts
o
f
r
esear
c
h
an
d
at
t
h
e
s
am
e
tim
e
is
g
iv
en
th
e
co
m
p
r
eh
e
n
s
iv
e
d
is
cu
s
s
io
n
.
R
esu
lts
ca
n
b
e
p
r
esen
ted
in
f
ig
u
r
es,
g
r
ap
h
s
,
tab
les
,
an
d
o
t
h
er
s
th
at
m
ak
e
th
e
r
ea
d
e
r
u
n
d
er
s
tan
d
ea
s
ily
[
1
4
]
,
[
1
5
]
.
T
h
e
d
is
cu
s
s
io
n
ca
n
b
e
m
ad
e
in
s
ev
er
al
s
u
b
-
s
ec
tio
n
s
.
T
h
is
s
tu
d
y
u
tili
ze
s
two
T
wi
tter
d
atasets
f
r
o
m
in
f
lu
en
tial
f
ig
u
r
es
—
J
ac
k
Do
r
s
ey
a
n
d
C
r
is
tian
o
R
o
n
ald
o
—
en
ab
lin
g
a
co
m
p
a
r
a
tiv
e
an
aly
s
is
o
f
co
m
m
u
n
icatio
n
s
ty
les an
d
s
o
cial
m
ed
ia
en
g
a
g
em
en
t.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
I
d
en
tifi
ca
tio
n
o
f
a
r
ea
s
o
f in
flu
en
ce
u
s
in
g
a
th
ema
tic
mo
d
elin
g
a
p
p
r
o
a
ch
a
n
d
b
elief
…
(
F
a
ti
ma
-
Za
h
r
a
e
S
ifi)
2843
i)
J
ac
k
Do
r
s
ey
d
ataset:
d
u
e
to
ap
p
licatio
n
p
r
o
g
r
am
m
in
g
in
te
r
f
ac
e
(
API
)
r
estrictio
n
s
f
o
llo
win
g
T
witter
’
s
tr
an
s
itio
n
to
"X"
in
ea
r
ly
2
0
2
3
,
we
m
an
u
ally
co
llected
4
0
0
twee
ts
f
r
o
m
Do
r
s
ey
’
s
ac
co
u
n
t.
T
h
e
d
ataset
co
n
s
is
ts
o
f
o
r
i
g
in
al
twee
ts
,
r
e
p
lies
an
d
r
etwe
ets
s
elec
ted
to
en
s
u
r
e
b
o
th
tem
p
o
r
al
a
n
d
th
e
m
atic
v
ar
iety
.
E
ac
h
twee
t
is
ac
c
o
m
p
an
ie
d
b
y
s
tr
u
ctu
r
e
d
m
eta
d
ata,
wh
ic
h
in
clu
d
es
I
D,
d
ate,
tex
t,
lik
es,
an
d
r
etwe
ets.
All
d
ata
ar
e
s
to
r
ed
in
c
o
m
m
a
-
s
ep
ar
ated
v
alu
e
s
(
C
SV)
f
o
r
m
at.
T
h
is
m
an
u
al
c
o
llectio
n
m
eth
o
d
h
elp
ed
m
ain
tain
th
e
ac
cu
r
ac
y
an
d
r
elev
an
ce
o
f
th
e
d
ata.
T
h
e
d
atas
et
is
u
s
ed
f
o
r
th
em
atic
an
aly
s
is
with
L
DA
ap
p
lied
to
id
en
tif
y
to
p
ics
in
Do
r
s
ey
’
s
d
ig
ital
d
is
co
u
r
s
e.
Ad
d
itio
n
ally
,
it
also
s
u
p
p
o
r
ts
th
e
m
ea
s
u
r
em
en
t
o
f
au
d
ie
n
ce
en
g
a
g
em
en
t.
ii)
C
r
is
tian
o
R
o
n
ald
o
d
ataset:
co
llected
v
ia
th
e
T
witter
API
th
r
o
u
g
h
T
wee
p
y
in
2
0
1
9
,
th
i
s
d
ataset
[
3
2
]
co
n
tain
s
2
,
5
0
7
o
r
i
g
in
al
twee
ts
b
y
C
r
is
tian
o
R
o
n
ald
o
o
v
e
r
s
i
x
y
ea
r
s
.
R
etwe
ets
wer
e
ex
clu
d
ed
to
f
o
cu
s
o
n
u
s
er
-
g
en
er
ate
d
co
n
ten
t.
E
ac
h
twee
t
in
clu
d
es
m
etad
ata
s
u
c
h
as
tim
estam
p
,
tex
t,
lan
g
u
ag
e,
lik
es
,
an
d
r
etwe
ets
s
to
r
ed
in
C
SV
f
o
r
m
at.
T
h
e
d
ataset
r
ef
lects
d
iv
er
s
e
co
m
m
u
n
icativ
e
s
ty
les,
f
r
o
m
p
r
o
m
o
tio
n
al
to
p
er
s
o
n
al
m
ess
ag
es.
I
t
s
er
v
es
to
ex
p
lo
r
e
d
i
g
ital
id
en
tity
co
n
s
tr
u
ctio
n
an
d
au
d
ien
ce
i
n
ter
ac
tio
n
with
L
DA
-
b
ased
to
p
ic
m
o
d
elin
g
a
n
d
en
g
a
g
em
en
t a
n
al
y
s
is
ap
p
lied
to
th
e
d
ata.
4
.
1
.
Resul
t
s
f
ro
m t
he
J
a
c
k
D
o
rsey
da
t
a
s
et
A
co
m
p
r
eh
e
n
s
iv
e
th
em
atic
a
n
aly
s
is
was
co
n
d
u
cted
o
n
t
wee
ts
f
r
o
m
J
ac
k
Do
r
s
ey
,
co
-
f
o
u
n
d
er
o
f
T
witter
an
d
k
ey
f
ig
u
r
e
i
n
tech
n
o
lo
g
y
an
d
c
r
y
p
to
c
u
r
r
e
n
cy
.
T
h
e
o
b
jectiv
e
was
to
id
en
tif
y
d
o
m
in
an
t
to
p
ics
an
d
ev
alu
ate
th
eir
im
p
o
r
tan
ce
.
Pro
b
ab
ilis
tic
to
p
ic
m
o
d
elin
g
L
DA
was
co
m
b
in
e
d
with
B
FT
f
o
r
th
is
task
.
T
h
e
p
r
e
p
r
o
ce
s
s
in
g
s
tep
s
in
v
o
lv
ed
clea
n
i
n
g
th
e
tex
t,
r
e
m
o
v
in
g
s
to
p
wo
r
d
s
,
lem
m
atizin
g
an
d
n
o
r
m
alizin
g
tem
p
o
r
al
a
n
d
n
u
m
er
ic
e
n
titi
es.
T
wee
ts
wer
e
to
k
en
ized
an
d
b
i
g
r
am
s
wer
e
in
co
r
p
o
r
ated
to
i
m
p
r
o
v
e
th
e
an
aly
s
is
b
ef
o
r
e
v
ec
to
r
izatio
n
.
T
h
e
L
DA
m
o
d
el
was
tr
ain
ed
o
n
th
e
p
r
e
p
r
o
ce
s
s
ed
co
r
p
u
s
u
s
in
g
Gen
s
im
.
T
h
e
n
u
m
b
e
r
o
f
to
p
ics
was
s
et
to
7
,
b
ased
o
n
co
h
e
r
en
ce
o
p
tim
izatio
n
.
T
h
e
Dir
ich
let
p
r
i
o
r
s
wer
e
co
n
f
ig
u
r
ed
as
α
=
‘
a
u
to
’
(
o
p
tim
ized
p
er
d
o
cu
m
e
n
t)
an
d
β =0
.
0
1
.
T
h
is
c
o
n
f
ig
u
r
atio
n
e
n
s
u
r
ed
b
alan
ce
d
to
p
ic
s
p
ar
s
ity
an
d
i
n
ter
p
r
etab
i
lity
.
Mo
d
el
p
er
f
o
r
m
an
ce
was
ev
alu
ated
u
s
in
g
p
e
r
p
lex
ity
an
d
to
p
ic
co
h
er
e
n
ce
m
etr
ics.
T
h
e
m
o
d
el
ac
h
iev
ed
a
p
er
p
lex
ity
s
co
r
e
o
f
-
7
.
3
5
an
d
a
co
h
er
en
ce
s
co
r
e
o
f
0
.
6
7
.
T
h
ese
r
esu
lts
in
d
icate
s
tr
o
n
g
in
ter
n
al
co
n
s
is
ten
cy
am
o
n
g
to
p
ics.
Ad
d
itio
n
ally
,
t
h
e
UM
ass
co
h
e
r
en
ce
v
alu
e
o
f
-
2
0
.
2
1
r
ef
lects
a
s
t
ab
le
d
is
tr
ib
u
tio
n
o
f
s
em
an
tically
co
h
er
en
t to
p
ic
-
w
o
r
d
ass
o
ciatio
n
s
ac
r
o
s
s
r
u
n
s
.
T
ab
le
1
p
r
esen
ts
th
e
k
ey
to
p
ics
id
en
tifie
d
in
th
e
J
ac
k
Do
r
s
ey
d
ataset
alo
n
g
with
th
eir
d
o
m
in
an
t
k
ey
wo
r
d
s
.
E
ac
h
to
p
ic
was
allo
ca
ted
a
lab
el
b
ased
o
n
its
d
o
m
in
an
t
k
e
y
wo
r
d
s
.
T
h
e
m
o
s
t
r
ep
r
esen
tativ
e
to
p
ics
in
clu
d
ed
b
itco
in
,
m
o
n
ey
,
b
an
k
,
d
ata,
p
r
o
to
co
l,
a
n
d
f
in
an
ce
.
T
ab
le
1
.
Key
t
o
p
ics
an
d
d
o
m
i
n
an
t k
ey
wo
r
d
s
in
th
e
J
ac
k
Do
r
s
ey
d
ataset
To
p
i
c
D
o
mi
n
a
n
t
k
e
y
w
o
r
d
s
B
i
t
c
o
i
n
b
i
t
c
o
i
n
,
t
o
o
l
,
c
o
n
t
r
o
l
,
g
l
o
b
a
l
,
d
e
c
l
a
r
a
t
i
o
n
,
c
a
s
h
…
M
o
n
e
y
mo
n
e
y
,
v
a
l
u
e
,
i
n
f
l
a
t
i
o
n
,
p
r
i
v
a
c
y
,
sa
v
i
n
g
…
B
a
n
k
b
a
n
k
,
c
r
i
si
s
,
i
n
f
r
a
s
t
r
u
c
t
u
r
e
,
mo
n
e
y
…
D
a
t
a
d
a
t
a
,
c
o
n
t
e
n
t
,
c
o
n
t
e
n
t
,
t
e
c
h
n
o
l
o
g
y
…
P
r
o
t
o
c
o
l
p
r
o
t
o
c
o
l
,
p
l
a
t
f
o
r
m,
si
g
n
a
l
,
c
o
r
r
u
p
t
i
o
n
…
F
i
n
a
n
c
e
f
i
n
a
n
c
e
,
c
o
n
t
r
o
l
,
w
o
r
l
d
,
p
e
r
m
i
ssi
o
n
,
b
a
n
k
…
T
h
e
"
b
itco
in
"
to
p
ic
s
tan
d
s
o
u
t
with
a
th
em
atic
m
ass
o
f
0
.
7
6
3
.
I
t
s
u
r
p
ass
es
o
th
er
to
p
i
cs
s
u
ch
as
m
o
n
ey
(
0
.
0
6
3
)
,
d
ata
(
0
.
0
5
3
)
an
d
f
in
an
ce
(
0
.
0
3
8
)
an
d
d
e
m
o
n
s
tr
ates
a
clea
r
f
o
cu
s
o
n
cr
y
p
to
cu
r
r
en
c
y
.
T
o
ev
alu
ate
au
d
ien
ce
im
p
ac
t,
we
an
aly
ze
d
t
h
e
e
n
g
ag
em
e
n
t
ass
o
ciate
d
with
ea
ch
to
p
ic
an
d
we
co
m
p
a
r
ed
r
etwe
et
an
d
lik
e
d
is
tr
ib
u
tio
n
s
.
T
a
b
le
2
p
r
esen
ts
th
e
t
h
em
atic
m
ass
an
d
au
d
ien
ce
e
n
g
ag
em
e
n
t
m
e
tr
ics
(
r
etwe
ets
an
d
lik
es)
f
o
r
ea
ch
id
en
tifie
d
t
o
p
ic
in
th
e
J
ac
k
Do
r
s
ey
d
ataset.
T
ab
le
2
.
T
o
p
ic
an
d
en
g
a
g
em
e
n
t m
etr
ics
in
th
e
J
ac
k
Do
r
s
ey
d
ataset
To
p
i
c
Th
e
m
a
t
i
c
mass
R
e
t
w
e
e
t
mas
s
Li
k
e
m
a
ss
B
i
t
c
o
i
n
0
.
7
6
3
0
.
5
2
5
0
.
5
0
8
M
o
n
e
y
0
.
0
6
3
0
.
0
3
6
0
.
0
3
4
P
r
o
t
o
c
o
l
0
.
0
4
8
0
.
0
3
1
0
.
0
2
6
D
a
t
a
0
.
0
5
3
0
.
0
2
6
0
.
0
3
7
B
a
n
k
0
.
0
3
5
0
.
0
2
4
0
.
0
2
6
F
i
n
a
n
c
e
0
.
0
3
8
0
.
0
2
6
0
.
0
2
4
T
h
e
“
b
itco
in
”
to
p
ic
is
th
e
m
o
s
t
p
r
o
m
in
en
t
a
n
d
attr
ac
ts
th
e
lar
g
est
au
d
ien
ce
,
with
h
ig
h
er
co
n
ten
t
f
o
cu
s
an
d
a
u
d
ien
ce
i
n
ter
ac
tio
n
th
an
o
th
er
to
p
ics.
T
o
co
m
b
in
e
in
f
o
r
m
atio
n
f
r
o
m
s
em
an
tic,
tem
p
o
r
al
a
n
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
26
:
2
8
3
8
-
2
8
4
8
2844
en
g
ag
em
e
n
t d
im
en
s
io
n
s
,
we
u
s
ed
a
s
eq
u
en
tial m
ass
f
u
s
io
n
m
eth
o
d
b
ased
o
n
B
FT.
T
h
is
a
p
p
r
o
ac
h
co
n
s
o
lid
ates
th
em
atic
im
p
o
r
tan
ce
an
d
ac
c
o
u
n
ts
f
o
r
u
n
ce
r
tain
ty
ac
r
o
s
s
m
e
tr
ics.
Af
ter
f
u
s
io
n
,
t
h
e
"
b
itco
in
"
t
o
p
ic
h
ad
a
n
o
tab
ly
h
ig
h
f
u
s
ed
m
ass
o
f
0
.
3
0
0
.
All
o
th
e
r
to
p
i
cs
s
h
o
wed
n
ea
r
-
ze
r
o
v
alu
es.
T
h
e
B
etP
c
o
n
f
ir
m
e
d
b
itco
in
'
s
d
o
m
in
a
n
ce
,
with
a
v
alu
e
o
f
0
.
2
8
5
,
h
i
g
h
er
th
an
o
th
er
to
p
ics
(
≈
0
.
1
4
3
)
.
T
h
is
r
esu
lt
r
ein
f
o
r
ce
s
its
ce
n
tr
al
r
o
le
in
th
e
d
ig
ital
d
is
co
u
r
s
e.
T
ab
le
3
p
r
e
s
en
ts
th
e
p
ig
n
is
tic
p
r
o
b
a
b
ilit
y
(
B
etP)
v
alu
es a
f
ter
s
eq
u
en
tial f
u
s
io
n
f
o
r
ea
c
h
to
p
ic.
T
ab
le
3
.
Pig
n
is
tic
s
co
r
e
co
m
p
ar
is
o
n
To
p
i
c
B
e
t
P
(
se
q
u
e
n
t
i
a
l
f
u
si
o
n
)
B
i
t
c
o
i
n
0
.
2
8
5
5
6
5
7
0
8
2
M
o
n
e
y
0
.
1
4
2
9
1
1
4
7
9
3
P
r
o
t
o
c
o
l
0
.
1
4
2
8
8
3
7
2
8
6
D
a
t
a
0
.
1
4
2
8
9
2
5
2
5
5
B
a
n
k
0
.
1
4
2
8
7
2
7
2
5
1
F
i
n
a
n
c
e
0
.
1
4
2
8
7
3
8
3
3
4
T
h
e
r
esu
lts
s
h
o
w
th
at
J
ac
k
Do
r
s
ey
’
s
T
witter
d
is
co
u
r
s
e
f
o
cu
s
es
o
n
cr
y
p
t
o
cu
r
r
e
n
cy
.
B
itco
in
s
er
v
es
as
a
k
ey
elem
en
t
o
f
h
is
d
ig
ital
p
r
esen
ce
.
Oth
er
to
p
ics
ap
p
ea
r
,
b
u
t
th
eir
lo
wer
B
etP
s
co
r
es
in
d
icate
s
ec
o
n
d
ar
y
im
p
o
r
tan
ce
.
T
h
is
r
ef
lects
a
f
o
cu
s
ed
co
m
m
u
n
icatio
n
s
tr
ateg
y
th
at
g
i
v
es
p
r
io
r
ity
cr
y
p
to
-
r
elate
d
n
ar
r
ativ
es,
d
esp
ite
Do
r
s
ey
’
s
r
o
o
ts
in
web
tech
n
o
lo
g
ies.
T
o
v
alid
ate
t
h
e
to
p
ic
m
o
d
eli
n
g
r
esu
lts
o
b
tain
e
d
with
L
D
A,
th
e
m
eth
o
d
u
s
es
tr
an
s
f
o
r
m
er
-
b
ased
em
b
ed
d
in
g
s
f
o
r
r
e
p
r
esen
tatio
n
o
f
to
p
ics
(
B
E
R
T
o
p
ic)
co
m
b
i
n
ed
with
clu
s
ter
in
g
an
d
d
im
en
s
io
n
ality
r
ed
u
ctio
n
.
T
h
is
m
o
d
el
en
ab
les
a
m
o
r
e
p
r
ec
is
e
d
etec
tio
n
o
f
s
em
an
tically
co
h
er
en
t
to
p
ics,
esp
ec
ially
f
o
r
s
h
o
r
t
,
co
n
tex
t
-
r
ich
m
ess
ag
es
s
u
ch
as
twee
ts
.
Af
ter
p
r
ep
r
o
ce
s
s
in
g
th
e
s
am
e
co
r
p
u
s
,
B
E
R
T
o
p
ic
au
to
m
atica
lly
id
en
tifie
d
s
ix
d
o
m
in
a
n
t
to
p
ic
s
.
T
h
ey
c
o
r
r
esp
o
n
d
to
th
e
s
am
e
th
em
atic
f
a
m
ilies
as
L
DA:
b
itco
in
,
m
o
n
ey
,
f
in
an
ce
,
p
r
o
to
c
o
l,
d
ata,
a
n
d
b
an
k
.
T
ab
le
4
p
r
esen
ts
th
e
B
E
R
T
o
p
ic
r
esu
lts
an
d
in
clu
d
e
r
ep
r
esen
tativ
e
k
ey
wo
r
d
s
,
to
p
ic
f
r
e
q
u
en
c
y
,
an
d
to
p
ic
p
r
o
b
a
b
ilit
y
f
o
r
ea
c
h
id
en
tifie
d
to
p
ic.
T
ab
le
4
.
B
E
R
T
o
p
ic
r
esu
lts
o
n
J
ac
k
Do
r
s
ey
d
ataset
To
p
i
c
R
e
p
r
e
se
n
t
a
t
i
v
e
k
e
y
w
o
r
d
s
F
r
e
q
u
e
n
c
y
(
%)
To
p
i
c
p
r
o
b
a
b
i
l
i
t
y
B
i
t
c
o
i
n
b
i
t
c
o
i
n
,
mo
n
e
y
,
sy
s
t
e
m
,
t
r
u
s
t
l
e
ss,
n
e
t
w
o
r
k
4
1
.
6
0
.
7
2
F
i
n
a
n
c
e
b
a
n
k
,
s
y
st
e
m,
c
e
n
t
r
a
l
,
f
a
i
l
u
r
e
,
t
r
a
d
i
t
i
o
n
a
l
1
9
.
4
0
.
6
1
P
r
o
t
o
c
o
l
p
r
o
t
o
c
o
l
,
p
l
a
t
f
o
r
m,
c
o
d
e
,
l
a
y
e
r
,
p
r
i
v
a
c
y
1
3
.
2
0
.
5
8
D
a
t
a
d
a
t
a
,
c
o
n
t
e
n
t
,
t
e
c
h
n
o
l
o
g
y
,
u
s
e
r
,
c
o
n
t
r
o
l
1
1
.
8
0
.
5
5
M
o
n
e
y
mo
n
e
y
,
i
n
f
l
a
t
i
o
n
,
v
a
l
u
e
,
s
a
v
i
n
g
,
c
u
r
r
e
n
c
y
8
.
9
0
.
4
9
B
a
n
k
b
a
n
k
,
s
y
st
e
m,
c
e
n
t
r
a
l
,
f
a
i
l
u
r
e
,
t
r
a
d
i
t
i
o
n
a
l
5
.
1
0
.
4
5
A
cr
o
s
s
-
co
m
p
ar
is
o
n
o
f
b
o
th
m
eth
o
d
s
(
L
DA
v
s
.
B
E
R
T
o
p
ic)
in
d
icate
s
th
at
th
ey
co
n
v
er
g
e
o
n
th
e
s
am
e
k
ey
to
p
ics,
p
a
r
ticu
lar
ly
b
itco
in
,
f
in
an
ce
,
an
d
p
r
o
t
o
co
ls
.
T
h
is
r
esu
lt
d
em
o
n
s
tr
ates
th
e
r
o
b
u
s
tn
ess
o
f
th
e
d
is
co
u
r
s
e
s
tr
u
ctu
r
e.
W
h
ile
L
DA
q
u
an
tifie
s
to
p
ic
d
is
tr
ib
u
ti
o
n
b
ased
o
n
wo
r
d
f
r
eq
u
en
c
y
,
B
E
R
T
o
p
ic
clu
s
ter
s
s
em
an
tically
r
elate
d
twee
ts
a
cc
o
r
d
in
g
to
m
ea
n
in
g
an
d
co
n
tex
t.
T
h
is
cr
ea
tes
r
ef
in
ed
d
is
t
in
ctio
n
s
with
in
th
e
s
am
e
co
n
ce
p
tu
al
f
a
m
ilies
.
B
o
th
an
aly
s
es
co
n
f
ir
m
b
itco
in
’
s
th
em
atic
ce
n
tr
ality
,
th
e
r
ec
u
r
r
e
n
ce
o
f
f
in
an
cial
an
d
tech
n
o
lo
g
ical
is
s
u
es
an
d
th
e
lim
ited
p
r
esen
ce
o
f
s
ec
o
n
d
ar
y
to
p
ics
s
u
ch
as
d
ata
a
n
d
b
an
k
.
R
ath
er
t
h
an
o
p
p
o
s
in
g
b
o
th
ap
p
r
o
ac
h
es,
th
eir
co
n
v
er
g
en
ce
s
tr
en
g
th
e
n
s
th
e
in
ter
p
r
etativ
e
v
alid
ity
o
f
th
e
f
in
d
in
g
s
:
Do
r
s
ey
’
s
T
witter
ac
tiv
ity
is
ce
n
ter
ed
o
n
cr
y
p
to
c
u
r
r
en
c
y
id
eo
lo
g
y
b
u
t
also
in
ter
lin
k
s
tech
n
o
lo
g
y
,
d
ec
en
tr
aliza
tio
n
an
d
f
in
an
cial
in
d
ep
e
n
d
en
ce
th
at
r
e
f
lect
a
co
h
er
e
n
t d
ig
ital id
en
tity
ac
r
o
s
s
m
eth
o
d
s
.
4
.
2
.
Resul
t
s
f
ro
m t
he
Cristia
no
Ro
na
ldo
da
t
a
s
et
A
th
em
atic
an
aly
s
is
o
f
twee
t
s
f
r
o
m
C
r
is
tian
o
R
o
n
ald
o
,
a
g
lo
b
ally
in
f
lu
e
n
tial
f
o
o
tb
all
f
ig
u
r
e,
was
p
er
f
o
r
m
ed
to
id
en
tify
r
ec
u
r
r
in
g
co
n
ten
t
to
p
ics
an
d
ass
ess
th
eir
r
elativ
e
p
r
o
m
in
en
ce
.
T
h
e
an
aly
s
is
co
m
b
in
ed
L
DA
with
B
FT
an
d
also
im
p
lem
en
ted
in
Gen
s
im
t
o
c
o
n
s
o
lid
ate
f
in
d
in
g
s
f
r
o
m
b
o
t
h
s
em
an
tic
s
tr
u
ctu
r
e
an
d
au
d
ien
ce
en
g
a
g
em
en
t.
T
h
e
m
o
d
el
was
co
n
f
ig
u
r
e
d
with
1
2
to
p
ics,
ch
o
s
en
to
b
alan
ce
g
r
an
u
lar
ity
an
d
in
ter
p
r
etab
ilit
y
,
with
α
au
to
m
atica
lly
o
p
tim
ized
p
er
d
o
c
u
m
en
t
a
n
d
β
s
et
to
0
.
0
1
to
en
co
u
r
ag
e
s
p
ar
s
e
to
p
ic
-
wo
r
d
d
is
tr
ib
u
tio
n
s
.
M
o
d
el
ev
alu
atio
n
y
ield
ed
a
p
er
p
le
x
ity
o
f
-
1
0
.
3
9
,
a
co
h
e
r
en
ce
s
c
o
r
e
o
f
0
.
6
0
a
n
d
a
UM
ass
co
h
er
en
ce
o
f
-
1
7
.
5
0
,
in
d
icatin
g
r
ea
s
o
n
ab
le
s
em
an
tic
co
h
er
e
n
ce
ac
r
o
s
s
to
p
ics.
L
DA
id
en
tifie
d
f
o
u
r
d
o
m
in
an
t
to
p
ics
b
ased
o
n
s
em
an
tic
p
r
o
x
im
ity
.
T
a
b
le
5
lis
ts
th
e
m
ain
to
p
ics
id
en
tifie
d
in
th
e
C
r
is
tian
o
R
o
n
ald
o
d
ataset
alo
n
g
with
th
eir
d
o
m
in
a
n
t k
ey
wo
r
d
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
I
d
en
tifi
ca
tio
n
o
f
a
r
ea
s
o
f in
flu
en
ce
u
s
in
g
a
th
ema
tic
mo
d
elin
g
a
p
p
r
o
a
ch
a
n
d
b
elief
…
(
F
a
ti
ma
-
Za
h
r
a
e
S
ifi)
2845
T
ab
le
5
.
Key
t
o
p
ics an
d
d
o
m
i
n
an
t k
ey
wo
r
d
s
in
th
e
C
r
is
tian
o
R
o
n
ald
o
d
ataset
To
p
i
c
D
o
mi
n
a
n
t
k
e
y
w
o
r
d
s
G
a
me
g
a
m
e
,
m
a
t
c
h
,
w
o
r
l
d
c
u
p
,
p
o
r
t
u
g
a
l
.
.
.
M
a
d
r
i
d
mad
r
i
d
,
g
o
a
l
,
t
e
a
m,
f
a
v
o
r
i
t
e
,
m
a
t
c
h
.
.
.
Th
a
n
k
s
f
a
n
,
p
h
o
t
o
,
s
u
p
p
o
r
t
,
v
i
d
e
o
.
.
.
P
o
r
t
u
g
a
l
P
o
r
t
u
g
a
l
,
g
a
me
,
m
a
t
c
h
,
p
a
r
a
b
e
n
s…
T
h
e
Ma
d
r
i
d
to
p
ic
em
er
g
ed
as
th
e
m
o
s
t
p
r
o
m
in
e
n
t,
with
a
th
em
atic
m
ass
o
f
0
.
3
3
1
,
f
o
llo
we
d
b
y
g
am
e
(
0
.
2
5
4
)
,
th
an
k
s
(
0
.
2
2
1
)
,
an
d
Po
r
tu
g
al
(
0
.
1
9
5
)
.
T
h
ese
r
esu
lts
s
h
o
w
a
s
tr
o
n
g
ass
o
ciatio
n
with
R
ea
l
Ma
d
r
id
,
a
ce
n
tr
al
elem
en
t
o
f
R
o
n
al
d
o
'
s
p
u
b
lic
id
en
tity
.
T
ab
le
6
p
r
esen
ts
th
e
th
em
atic
m
ass
alo
n
g
s
id
e
r
etwe
et
an
d
lik
e
en
g
ag
em
e
n
t m
etr
ics f
o
r
ea
c
h
t
o
p
ic.
T
ab
le
6
.
T
o
p
ic
an
d
en
g
a
g
em
e
n
t m
etr
ics in
th
e
C
r
is
tian
o
R
o
n
ald
o
d
ataset
To
p
i
c
Th
e
m
a
t
i
c
mass
R
e
t
w
e
e
t
mas
s
Li
k
e
m
a
ss
M
a
d
r
i
d
0
.
3
3
1
0
.
2
5
7
0
.
2
1
5
G
a
me
0
.
2
5
4
0
.
2
1
1
0
.
2
1
0
Th
a
n
k
s
0
.
2
2
1
0
.
1
9
9
0
.
2
1
6
P
o
r
t
u
g
a
l
0
.
1
9
5
0
.
1
4
0
0
.
1
5
2
T
o
r
e
f
in
e
t
h
e
an
aly
s
is
,
s
o
cial
en
g
a
g
em
en
t
m
etr
ics
was
in
t
eg
r
ated
.
T
h
e
Ma
d
r
id
to
p
ic
g
u
id
ed
i
n
to
au
d
ien
ce
r
ec
ep
tio
n
with
th
e
h
ig
h
est
r
etwe
ets
an
d
lik
es.
Gam
e,
th
an
k
s
,
an
d
Po
r
tu
g
al
also
h
ad
s
ig
n
if
ican
t
en
g
ag
em
e
n
t
b
u
t
wer
e
s
ec
o
n
d
a
r
y
.
T
h
ese
r
esu
lts
u
n
d
er
s
co
r
e
R
ea
l
Ma
d
r
id
’
s
s
tr
ateg
ic
p
r
o
m
i
n
en
ce
in
R
o
n
al
d
o
’
s
d
ig
ital c
o
m
m
u
n
icatio
n
,
wh
ich
co
n
f
ir
m
s
a
co
n
s
is
ten
t id
en
tity
tied
to
clu
b
leg
ac
y
an
d
f
an
in
t
er
ac
tio
n
.
A
to
p
ic
weig
h
t
b
ased
o
n
a
u
d
i
en
ce
en
g
ag
em
e
n
t,
m
ea
s
u
r
ed
t
h
r
o
u
g
h
"lik
es"
an
d
"r
etwe
ets"
was
also
in
tr
o
d
u
ce
d
.
T
h
is
cr
o
s
s
-
ev
alu
at
io
n
co
n
f
ir
m
ed
th
e
co
n
tin
u
ed
d
o
m
in
an
ce
o
f
th
e
Ma
d
r
i
d
to
p
ic
with
th
e
h
ig
h
est
en
g
ag
em
e
n
t
v
alu
es
(
r
etwe
et
m
ass
:
0
.
2
5
7
,
lik
e
m
ass
:
0
.
2
1
5
)
.
I
n
c
o
n
tr
ast,
g
am
e
(
r
etwe
ets:
0
.
2
1
1
;
lik
es:
0
.
2
1
0
)
,
th
an
k
s
(
r
etwe
ets:
0
.
1
9
9
;
lik
e
s
:
0
.
2
1
6
)
,
an
d
Po
r
tu
g
al
(
r
et
wee
ts
:
0
.
1
4
0
;
lik
es:
0
.
1
5
2
)
h
ad
lo
wer
lev
els
o
f
atten
tio
n
b
u
t
r
em
ain
e
d
alig
n
e
d
with
th
e
o
v
e
r
all
th
em
atic
f
r
am
ewo
r
k
.
T
o
c
o
n
s
o
lid
ate
th
es
e
in
ter
p
r
etatio
n
s
,
a
s
eq
u
en
tial
f
u
s
io
n
s
tr
ateg
y
b
ased
o
n
B
FT
was
ap
p
lied
.
T
ab
le
7
p
r
esen
ts
th
e
B
etP
v
alu
es
o
b
tain
ed
f
r
o
m
th
e
s
eq
u
en
tial f
u
s
io
n
p
r
o
ce
s
s
.
T
ab
le
7
.
Pig
n
is
tic
s
co
r
e
co
m
p
ar
is
o
n
To
p
i
c
B
e
t
P
(
se
q
u
e
n
t
i
a
l
f
u
si
o
n
)
M
a
d
r
i
d
0
.
2
8
4
8
G
a
me
0
.
2
5
2
1
Th
a
n
k
s
0
.
2
4
3
8
P
o
r
t
u
g
a
l
0
.
2
1
9
1
T
h
e
p
ig
n
is
tic
p
r
o
b
ab
ilit
y
f
o
r
th
e
Ma
d
r
id
to
p
ic
was
0
.
2
8
4
,
m
u
ch
h
ig
h
er
th
an
o
th
e
r
to
p
ics.
T
h
is
co
n
f
ir
m
s
its
ce
n
tr
al
r
o
le
in
R
o
n
ald
o
’
s
d
ig
ital
c
o
m
m
u
n
icatio
n
.
R
ea
l
Ma
d
r
id
r
ef
er
e
n
ce
s
ar
e
b
o
th
th
em
atica
ll
y
d
o
m
in
an
t
a
n
d
s
tr
o
n
g
ly
co
n
n
e
cted
with
h
is
au
d
ien
ce
.
T
h
ey
r
ef
lect
a
co
n
s
is
ten
t
id
en
tity
b
u
ilt
ar
o
u
n
d
clu
b
leg
ac
y
an
d
p
er
s
o
n
al
b
r
an
d
.
T
h
e
L
DA
co
m
b
i
n
ed
with
B
FT
s
h
o
ws
th
e
Ma
d
r
id
t
o
p
ic
as
ce
n
tr
al
to
C
r
is
tian
o
R
o
n
ald
o
’
s
d
ig
ital
d
is
co
u
r
s
e
an
d
g
iv
es
s
tr
u
ct
u
r
e
to
h
is
T
witter
co
m
m
u
n
icatio
n
.
I
t
r
ef
lects
h
is
h
is
to
r
y
with
R
ea
l
Ma
d
r
id
an
d
h
is
s
tr
ateg
y
to
s
u
p
p
o
r
t
h
is
p
u
b
lic
im
ag
e
in
ath
letic
ac
h
iev
e
m
en
t
s
.
Un
lik
e
o
th
e
r
p
u
b
lic
f
ig
u
r
es
wh
o
d
iv
e
r
s
if
y
in
t
o
en
tr
ep
r
en
e
u
r
s
h
ip
,
p
o
liti
cs
o
r
p
h
ilan
th
r
o
p
y
,
R
o
n
ald
o
k
ee
p
s
a
s
p
o
r
ts
-
ce
n
ter
ed
n
ar
r
ativ
e.
Seco
n
d
ar
y
to
p
ics
lik
e
g
am
e,
th
a
n
k
s
,
an
d
Po
r
tu
g
al
s
u
p
p
o
r
t
en
g
a
g
em
en
t
an
d
co
m
m
u
n
ity
b
u
t
s
tay
alig
n
ed
with
h
is
f
o
o
tb
all
id
e
n
tity
.
T
h
ey
m
ain
tain
a
f
o
cu
s
ed
b
r
an
d
s
tr
ateg
y
r
o
o
ted
in
h
is
a
th
letic
leg
ac
y
.
B
ey
o
n
d
th
e
L
DA
f
r
am
ewo
r
k
,
a
co
m
p
lem
en
tar
y
e
x
p
er
im
en
t
with
B
E
R
T
o
p
ic
was
co
n
d
u
ct
ed
to
u
n
c
o
v
er
s
em
an
tic
lay
e
r
s
with
in
C
r
is
tian
o
R
o
n
ald
o
’
s
twee
ts
.
T
ab
le
8
s
u
m
m
ar
izes
th
e
B
E
R
T
o
p
ic
to
p
ics
ex
tr
ac
ted
f
r
o
m
th
e
C
r
is
tian
o
R
o
n
ald
o
d
ataset.
I
t
in
clu
d
ed
r
e
p
r
esen
tativ
e
k
e
y
wo
r
d
s
,
f
r
eq
u
en
cies a
n
d
t
o
p
ic
p
r
o
b
ab
ilit
ies.
T
h
e
m
o
d
el
p
r
o
d
u
ce
d
f
iv
e
c
o
h
esiv
e
th
em
atic
clu
s
ter
s
;
ea
ch
o
n
e
r
ep
r
esen
ts
a
k
e
y
d
im
en
s
io
n
o
f
R
o
n
ald
o
’
s
co
m
m
u
n
icatio
n
.
T
h
e
aim
was
n
o
t
to
r
ep
lace
L
DA
b
u
t
to
co
m
p
ar
e
th
e
th
em
atic
s
eg
m
en
tatio
n
o
f
b
o
th
m
o
d
els.
T
h
is
en
s
u
r
es
in
ter
p
r
etiv
e
c
o
n
s
is
ten
cy
ac
r
o
s
s
co
m
p
u
tatio
n
al
ap
p
r
o
ac
h
es.
B
o
th
L
DA
an
d
B
E
R
T
o
p
ic
co
n
v
er
g
e
o
n
th
e
s
am
e
co
r
e
th
em
atic
ar
ea
s
.
Ma
d
r
id
,
g
am
e,
a
n
d
Po
r
tu
g
al
ar
e
th
e
m
ain
ax
es
o
f
R
o
n
ald
o
’
s
d
ig
ital
d
is
co
u
r
s
e.
W
h
ile
L
DA
d
escr
ib
es
th
e
p
r
o
b
ab
ilis
tic
s
tr
u
ctu
r
e
o
f
to
p
i
cs
b
ased
o
n
wo
r
d
f
r
eq
u
e
n
cy
,
B
E
R
T
o
p
ic
r
ein
f
o
r
c
es sem
an
tic
co
h
esio
n
.
I
t g
r
o
u
p
s
r
elate
d
twee
ts
u
n
d
er
b
r
o
ad
er
clu
s
ter
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
26
:
2
8
3
8
-
2
8
4
8
2846
T
h
is
co
m
p
lem
en
tar
y
p
e
r
s
p
ec
tiv
e
co
n
f
ir
m
s
th
at
R
o
n
ald
o
’
s
co
m
m
u
n
icatio
n
s
tr
ateg
y
is
ce
n
ter
ed
i
n
clu
b
id
en
tity
an
d
s
p
o
r
ts
p
er
f
o
r
m
an
ce
.
I
t
also
in
teg
r
ates
p
er
s
o
n
al
an
d
em
o
tio
n
al
n
ar
r
ativ
es
(
th
an
k
s
an
d
m
o
tiv
atio
n
)
th
at
s
tr
en
g
th
en
f
an
p
r
o
x
im
ity
an
d
e
n
g
ag
em
e
n
t.
T
h
e
co
n
v
er
g
e
n
ce
o
f
r
esu
lts
f
r
o
m
b
o
th
m
o
d
els
s
u
p
p
o
r
ts
th
e
r
o
b
u
s
tn
ess
an
d
co
n
s
is
ten
cy
o
f
th
e
th
em
atic
in
ter
p
r
etatio
n
.
C
r
is
tian
o
R
o
n
ald
o
’
s
T
witter
d
is
co
u
r
s
e,
th
er
ef
o
r
e,
r
ef
lects
a
co
h
er
en
t
an
d
m
u
ltid
im
en
s
io
n
al
n
ar
r
ativ
e
in
wh
ic
h
p
r
o
f
ess
io
n
al
e
x
ce
l
len
ce
an
d
p
e
r
s
o
n
al
au
th
en
ticity
co
ex
is
t to
s
u
s
tain
au
d
ien
ce
co
n
n
ec
tio
n
.
T
ab
le
8
.
B
E
R
T
o
p
ic
to
p
ics ex
tr
ac
ted
f
r
o
m
th
e
C
r
is
tian
o
R
o
n
a
ld
o
d
ataset
To
p
i
c
R
e
p
r
e
se
n
t
a
t
i
v
e
k
e
y
w
o
r
d
s
F
r
e
q
u
e
n
c
y
(
%)
To
p
i
c
p
r
o
b
a
b
i
l
i
t
y
M
a
d
r
i
d
mad
r
i
d
,
g
o
a
l
,
c
h
a
m
p
i
o
n
s,
t
e
a
m,
c
l
u
b
3
3
.
7
0
.
7
0
G
a
me
mat
c
h
,
w
i
n
,
w
o
r
l
d
c
u
p
,
t
r
a
i
n
i
n
g
,
v
i
c
t
o
r
y
2
7
.
1
0
.
6
5
P
o
r
t
u
g
a
l
p
o
r
t
u
g
a
l
,
sel
e
c
a
o
,
f
a
n
s,
n
a
t
i
o
n
,
p
r
i
d
e
1
8
.
4
0
.
6
1
Th
a
n
k
s
t
h
a
n
k
,
su
p
p
o
r
t
,
l
o
v
e
,
p
h
o
t
o
,
f
a
m
i
l
y
1
2
.
3
0
.
5
7
M
o
t
i
v
a
t
i
o
n
p
r
o
u
d
,
d
r
e
a
m,
t
o
g
e
t
h
e
r
,
mo
m
e
n
t
,
h
o
m
e
8
.
5
0
.
5
4
5.
CO
NCLU
SI
O
N
T
h
is
s
tu
d
y
p
r
esen
ts
a
f
r
am
ew
o
r
k
th
at
co
m
b
i
n
es
L
DA
with
B
FT
to
an
aly
ze
s
o
cial
m
ed
ia
d
is
co
u
r
s
e.
T
o
p
ics
ar
e
m
o
d
eled
p
r
o
b
a
b
ilis
tically
an
d
au
d
ie
n
ce
en
g
a
g
em
en
t,
s
u
ch
as
lik
es
an
d
r
etwe
ets,
ar
e
in
co
r
p
o
r
ated
.
T
h
is
ap
p
r
o
ac
h
p
r
o
v
id
es
a
clea
r
er
v
iew
o
f
wh
at
d
r
i
v
es
co
m
m
u
n
icatio
n
am
o
n
g
in
f
l
u
en
ce
r
s
.
T
h
e
k
ey
ad
v
an
tag
e
o
f
th
is
m
eth
o
d
o
lo
g
y
r
esid
es
in
it
s
d
ata
f
u
s
io
n
s
tr
ate
g
y
.
Seq
u
en
tial
b
elief
m
ass
f
u
s
io
n
r
ec
o
n
ciles
in
co
n
s
is
ten
cies
b
etwe
en
co
n
ten
t
r
elev
an
ce
an
d
p
u
b
lic
r
ec
e
p
t
io
n
.
I
t
p
r
o
v
id
es
a
n
u
a
n
ce
d
v
ie
w
o
f
to
p
ic
s
alien
ce
u
n
d
er
u
n
ce
r
tain
ty
.
B
ey
o
n
d
it
s
cu
r
r
en
t
a
p
p
licatio
n
,
th
e
f
r
am
ewo
r
k
is
r
e
m
ar
k
ab
l
y
ad
a
p
tab
le.
I
t
c
o
u
ld
b
e
ex
ten
d
ed
to
d
o
m
ain
s
s
u
ch
as
s
en
tim
en
t
an
aly
s
is
,
tr
e
n
d
d
etec
t
io
n
o
r
s
cien
tific
tex
t
m
in
in
g
.
I
t
also
co
u
ld
b
en
ef
it
f
r
o
m
f
u
r
t
h
er
r
ef
in
em
e
n
ts
,
p
ar
ticu
lar
ly
in
tem
p
o
r
al
d
y
n
am
i
cs
an
d
f
u
s
io
n
r
u
le
o
p
tim
izatio
n
.
B
etter
co
n
f
lict
m
an
ag
em
en
t
with
in
b
elief
f
u
s
io
n
,
f
o
r
ex
a
m
p
le,
wo
u
ld
im
p
r
o
v
e
s
en
s
itiv
ity
to
r
e
f
in
e
p
atter
n
s
an
d
c
o
r
r
elate
d
to
p
ics.
Ov
er
all,
th
is
ap
p
r
o
ac
h
p
r
o
v
id
es
a
f
lex
ib
le,
tr
a
n
s
f
er
ab
le
an
d
th
e
o
r
etica
lly
p
r
ac
tical
t
o
o
l
f
o
r
q
u
a
n
titativ
e
an
aly
s
is
o
f
d
ig
ital
d
is
co
u
r
s
e.
I
t
ex
ten
s
iv
ely
r
elev
an
t
ac
r
o
s
s
f
ield
s
th
at
wo
r
k
with
m
as
s
iv
e
tex
tu
al
d
ata
an
d
o
n
lin
e
in
f
lu
e
n
ce
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
Au
th
o
r
s
s
tate
n
o
f
u
n
d
in
g
in
v
o
lv
ed
.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
T
h
is
jo
u
r
n
al
u
s
es
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT)
to
r
ec
o
g
n
ize
in
d
iv
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
llab
o
r
atio
n
.
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
Fatim
a
-
Z
ah
r
ae
Sifi
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
W
af
ae
Sab
b
ar
✓
✓
✓
✓
✓
✓
✓
✓
Am
al
E
l M
za
b
i
✓
✓
✓
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
T
h
e
au
t
h
o
r
s
d
ec
lar
e
th
at
th
er
e
is
n
o
co
n
f
lict
o
f
in
ter
est
r
eg
a
r
d
in
g
th
e
p
u
b
licatio
n
o
f
th
is
a
r
ticle.
T
h
e
au
th
o
r
s
co
n
f
ir
m
ed
th
at
th
e
p
a
p
er
was f
r
ee
o
f
p
lag
iar
is
m
.
I
NF
O
RM
E
D
CO
NS
E
N
T
T
h
is
s
tu
d
y
d
o
es
n
o
t
in
v
o
lv
e
h
u
m
an
p
ar
ticip
an
ts
o
r
p
er
s
o
n
al
d
ata.
T
h
er
e
f
o
r
e,
in
f
o
r
m
ed
co
n
s
en
t
is
n
o
t
r
eq
u
ir
ed
in
th
is
s
tu
d
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
I
d
en
tifi
ca
tio
n
o
f
a
r
ea
s
o
f in
flu
en
ce
u
s
in
g
a
th
ema
tic
mo
d
elin
g
a
p
p
r
o
a
ch
a
n
d
b
elief
…
(
F
a
ti
ma
-
Za
h
r
a
e
S
ifi)
2847
E
T
H
I
CAL AP
P
RO
V
AL
T
h
i
s
r
e
s
e
a
r
c
h
d
o
e
s
n
o
t
i
n
v
o
l
v
e
h
u
m
a
n
s
u
b
j
e
c
t
s
o
r
a
n
i
m
a
l
s
.
T
h
e
r
e
f
o
r
e
,
e
t
h
i
c
a
l
a
p
p
r
o
v
a
l
w
a
s
n
o
t
r
e
q
u
i
r
e
d
.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
au
th
o
r
s
co
n
f
ir
m
th
at
th
e
d
ata
s
u
p
p
o
r
tin
g
th
e
f
in
d
in
g
s
o
f
th
is
s
tu
d
y
ar
e
av
ailab
le
with
in
th
e
ar
ticle
[
an
d
/o
r
its
s
u
p
p
lem
en
tar
y
m
ater
ials
]
.
RE
F
E
R
E
NC
E
S
[
1
]
K
.
K
.
K
a
p
o
o
r
,
K
.
Ta
m
i
l
m
a
n
i
,
N
.
P
.
R
a
n
a
,
P
.
P
a
t
i
l
,
Y
.
K
.
D
w
i
v
e
d
i
,
a
n
d
S
.
N
e
r
u
r
,
“
A
d
v
a
n
c
e
s
i
n
so
c
i
a
l
me
d
i
a
r
e
se
a
r
c
h
:
p
a
s
t
,
p
r
e
s
e
n
t
a
n
d
f
u
t
u
r
e
,
”
I
n
f
o
rm
a
t
i
o
n
S
y
st
e
m
s F
r
o
n
t
i
e
rs
,
v
o
l
.
2
0
,
n
o
.
3
,
p
p
.
5
3
1
–
5
5
8
,
J
u
n
.
2
0
1
8
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
0
7
9
6
-
0
1
7
-
9
8
1
0
-
y.
[
2
]
V
.
B
a
r
g
e
r
,
J.
W
.
P
e
l
t
i
e
r
,
a
n
d
D
.
E.
S
c
h
u
l
t
z
,
“
S
o
c
i
a
l
m
e
d
i
a
a
n
d
c
o
n
s
u
m
e
r
e
n
g
a
g
e
me
n
t
:
a
r
e
v
i
e
w
a
n
d
r
e
se
a
r
c
h
a
g
e
n
d
a
,
”
J
o
u
r
n
a
l
o
f
Re
se
a
rc
h
i
n
I
n
t
e
r
a
c
t
i
v
e
Ma
r
k
e
t
i
n
g
,
v
o
l
.
1
0
,
n
o
.
4
,
p
p
.
2
6
8
–
2
8
7
,
O
c
t
.
2
0
1
6
,
d
o
i
:
1
0
.
1
1
0
8
/
JR
I
M
-
06
-
2
0
1
6
-
0
0
6
5
.
[
3
]
S
.
Z
h
o
u
,
M
.
B
l
a
z
q
u
e
z
,
H
.
M
c
C
o
r
mi
c
k
,
a
n
d
L.
B
a
r
n
e
s,
“
H
o
w
s
o
c
i
a
l
m
e
d
i
a
i
n
f
l
u
e
n
c
e
r
s’
n
a
r
r
a
t
i
v
e
s
t
r
a
t
e
g
i
e
s
b
e
n
e
f
i
t
c
u
l
t
i
v
a
t
i
n
g
i
n
f
l
u
e
n
c
e
r
mark
e
t
i
n
g
:
t
a
c
k
l
i
n
g
i
ssu
e
s
o
f
c
u
l
t
u
r
a
l
b
a
r
r
i
e
r
s
,
c
o
m
merc
i
a
l
i
s
e
d
c
o
n
t
e
n
t
,
a
n
d
s
p
o
n
s
o
r
sh
i
p
d
i
s
c
l
o
s
u
r
e
,
”
J
o
u
r
n
a
l
o
f
Bu
si
n
e
ss R
e
se
a
rc
h
,
v
o
l
.
1
3
4
,
p
p
.
1
2
2
–
1
4
2
,
S
e
p
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
b
u
sr
e
s
.
2
0
2
1
.
0
5
.
0
1
1
.
[
4
]
M
.
El
M
a
r
r
a
k
c
h
i
,
H
.
B
e
n
sa
i
d
,
a
n
d
M
.
B
e
l
l
a
f
k
i
h
,
“
E
-
r
e
p
u
t
a
t
i
o
n
p
r
e
d
i
c
t
i
o
n
mo
d
e
l
i
n
o
n
l
i
n
e
s
o
c
i
a
l
n
e
t
w
o
r
k
s
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
I
n
t
e
l
l
i
g
e
n
t
S
y
st
e
m
s
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
9
,
n
o
.
1
1
,
p
p
.
1
7
–
2
5
,
N
o
v
.
2
0
1
7
,
d
o
i
:
1
0
.
5
8
1
5
/
i
j
i
s
a
.
2
0
1
7
.
1
1
.
0
3
.
[
5
]
Y
.
Y
a
n
,
F
.
T
o
r
i
u
mi
,
a
n
d
T.
S
u
g
a
w
a
r
a
,
“
U
n
d
e
r
s
t
a
n
d
i
n
g
h
o
w
r
e
t
w
e
e
t
s
i
n
f
l
u
e
n
c
e
t
h
e
b
e
h
a
v
i
o
r
s
o
f
s
o
c
i
a
l
n
e
t
w
o
r
k
i
n
g
serv
i
c
e
u
s
e
r
s v
i
a
a
g
e
n
t
-
b
a
se
d
si
mu
l
a
t
i
o
n
,
”
C
o
m
p
u
t
a
t
i
o
n
a
l
S
o
c
i
a
l
N
e
t
w
o
r
k
s
,
v
o
l
.
8
,
n
o
.
1
,
D
e
c
.
2
0
2
1
,
d
o
i
:
1
0
.
1
1
8
6
/
s
4
0
6
4
9
-
0
2
1
-
0
0
0
9
9
-
8.
[
6
]
B
.
W
i
l
d
e
r
,
N
.
I
mm
o
r
l
i
c
a
,
E.
R
i
c
e
,
a
n
d
M
.
Ta
m
b
e
,
“
M
a
x
i
m
i
z
i
n
g
i
n
f
l
u
e
n
c
e
i
n
a
n
u
n
k
n
o
w
n
s
o
c
i
a
l
n
e
t
w
o
r
k
,
”
P
ro
c
e
e
d
i
n
g
s
o
f
t
h
e
AAAI C
o
n
f
e
re
n
c
e
o
n
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
3
2
,
n
o
.
1
,
A
p
r
.
2
0
1
8
,
d
o
i
:
1
0
.
1
6
0
9
/
a
a
a
i
.
v
3
2
i
1
.
1
1
5
8
5
.
[
7
]
J.
Ta
n
g
a
n
d
H
.
Li
u
,
“
U
n
s
u
p
e
r
v
i
se
d
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
f
o
r
l
i
n
k
e
d
s
o
c
i
a
l
me
d
i
a
d
a
t
a
,
”
i
n
Pro
c
e
e
d
i
n
g
s
o
f
t
h
e
1
8
t
h
AC
M
S
I
G
K
D
D
i
n
t
e
r
n
a
t
i
o
n
a
l
c
o
n
f
e
re
n
c
e
o
n
K
n
o
w
l
e
d
g
e
d
i
sc
o
v
e
ry
a
n
d
d
a
t
a
m
i
n
i
n
g
,
A
u
g
.
2
0
1
2
,
p
p
.
9
0
4
–
9
1
2
,
d
o
i
:
1
0
.
1
1
4
5
/
2
3
3
9
5
3
0
.
2
3
3
9
6
7
3
.
[
8
]
M
.
C
h
a
,
H
.
H
a
d
d
a
d
i
,
F
.
B
e
n
e
v
e
n
u
t
o
,
a
n
d
K
.
G
u
mm
a
d
i
,
“
M
e
a
s
u
r
i
n
g
u
se
r
i
n
f
l
u
e
n
c
e
i
n
Tw
i
t
t
e
r
:
t
h
e
mi
l
l
i
o
n
f
o
l
l
o
w
e
r
f
a
l
l
a
c
y
,
”
Pro
c
e
e
d
i
n
g
s
o
f
t
h
e
I
n
t
e
rn
a
t
i
o
n
a
l
AAAI
C
o
n
f
e
re
n
c
e
o
n
We
b
a
n
d
S
o
c
i
a
l
Me
d
i
a
,
v
o
l
.
4
,
n
o
.
1
,
p
p
.
1
0
–
1
7
,
M
a
y
2
0
1
0
,
d
o
i
:
1
0
.
1
6
0
9
/
i
c
w
sm
.
v
4
i
1
.
1
4
0
3
3
.
[
9
]
S
.
Jen
d
o
u
b
i
a
n
d
A
.
M
a
r
t
i
n
,
“
A
r
e
l
i
a
b
i
l
i
t
y
-
b
a
s
e
d
a
p
p
r
o
a
c
h
f
o
r
i
n
f
l
u
e
n
c
e
max
i
mi
z
a
t
i
o
n
u
s
i
n
g
t
h
e
e
v
i
d
e
n
c
e
t
h
e
o
r
y
,
”
i
n
B
i
g
D
a
t
a
An
a
l
y
t
i
c
s
a
n
d
K
n
o
w
l
e
d
g
e
D
i
s
c
o
v
e
r
y
(
D
a
W
a
K
2
0
1
7
)
,
2
0
1
7
,
p
p
.
3
1
3
–
326
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
3
1
9
-
6
4
2
8
3
-
3
_
2
3
.
[
1
0
]
H
.
Jel
o
d
a
r
e
t
a
l
.
,
“
La
t
e
n
t
d
i
r
i
c
h
l
e
t
a
l
l
o
c
a
t
i
o
n
(
LD
A
)
a
n
d
t
o
p
i
c
mo
d
e
l
i
n
g
:
mo
d
e
l
s
,
a
p
p
l
i
c
a
t
i
o
n
s,
a
s
u
r
v
e
y
,
”
Mu
l
t
i
m
e
d
i
a
T
o
o
l
s
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
7
8
,
n
o
.
1
1
,
p
p
.
1
5
1
6
9
–
1
5
2
1
1
,
J
u
n
.
2
0
1
9
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
0
4
2
-
018
-
6
8
9
4
-
4.
[
1
1
]
T.
D
e
n
œ
u
x
,
“
D
e
c
i
s
i
o
n
-
ma
k
i
n
g
w
i
t
h
b
e
l
i
e
f
f
u
n
c
t
i
o
n
s:
a
r
e
v
i
e
w
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
A
p
p
r
o
x
i
m
a
t
e
Re
a
so
n
i
n
g
,
v
o
l
.
1
0
9
,
p
p
.
8
7
–
1
1
0
,
Ju
n
.
2
0
1
9
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
j
a
r
.
2
0
1
9
.
0
3
.
0
0
9
.
[
1
2
]
Y
.
H
u
a
n
g
,
Q
.
Li
u
,
J.
Li
u
,
a
n
d
Y
.
H
u
,
“
To
p
i
c
d
i
s
c
o
v
e
r
y
i
n
s
c
i
e
n
t
i
f
i
c
l
i
t
e
r
a
t
u
r
e
,
”
i
n
C
o
m
p
u
t
e
r
S
u
p
p
o
r
t
e
d
C
o
o
p
e
ra
t
i
v
e
W
o
r
k
a
n
d
S
o
c
i
a
l
C
o
m
p
u
t
i
n
g
(
C
h
i
n
e
s
e
C
S
C
W
2
0
2
2
)
,
2
0
2
3
,
p
p
.
4
8
1
–
491
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
981
-
99
-
2
3
5
6
-
4
_
3
8
.
[
1
3
]
R
.
J
i
r
o
u
še
k
,
V
.
K
r
a
t
o
c
h
v
í
l
,
a
n
d
P
.
P
.
S
h
e
n
o
y
,
“
C
o
mp
u
t
i
n
g
t
h
e
d
e
c
o
m
p
o
sa
b
l
e
e
n
t
r
o
p
y
o
f
b
e
l
i
e
f
-
f
u
n
c
t
i
o
n
g
r
a
p
h
i
c
a
l
mo
d
e
l
s,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
A
p
p
ro
x
i
m
a
t
e
Re
a
s
o
n
i
n
g
,
v
o
l
.
1
6
1
,
O
c
t
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
j
a
r
.
2
0
2
3
.
1
0
8
9
8
4
.
[
1
4
]
E.
K
o
k
s
a
l
m
i
s
a
n
d
Ö
.
K
a
b
a
k
,
“
S
e
n
s
o
r
f
u
si
o
n
b
a
s
e
d
o
n
D
e
m
p
s
t
e
r
‐
S
h
a
f
e
r
t
h
e
o
r
y
o
f
e
v
i
d
e
n
c
e
u
s
i
n
g
a
l
a
r
g
e
sc
a
l
e
g
r
o
u
p
d
e
c
i
si
o
n
mak
i
n
g
a
p
p
r
o
a
c
h
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
I
n
t
e
l
l
i
g
e
n
t
S
y
s
t
e
m
s
,
v
o
l
.
3
5
,
n
o
.
7
,
p
p
.
1
1
2
6
–
1
1
6
2
,
J
u
l
.
2
0
2
0
,
d
o
i
:
1
0
.
1
0
0
2
/
i
n
t
.
2
2
2
3
7
.
[
1
5
]
J.
N
i
,
J.
L
u
o
,
a
n
d
W
.
L
i
u
,
“
3
D
p
a
l
mp
r
i
n
t
r
e
c
o
g
n
i
t
i
o
n
u
s
i
n
g
D
e
mp
s
t
e
r
-
S
h
a
f
e
r
f
u
si
o
n
t
h
e
o
r
y
,
”
J
o
u
rn
a
l
o
f
S
e
n
so
rs
,
v
o
l
.
2
0
1
5
,
p
p
.
1
–
7
,
2
0
1
5
,
d
o
i
:
1
0
.
1
1
5
5
/
2
0
1
5
/
2
5
2
0
8
6
.
[
1
6
]
N
.
J.
J.
S
m
i
t
h
,
“
A
c
t
i
n
g
o
n
b
e
l
i
e
f
f
u
n
c
t
i
o
n
s,
”
T
h
e
o
ry
a
n
d
D
e
c
i
si
o
n
,
v
o
l
.
9
5
,
n
o
.
4
,
p
p
.
5
7
5
–
6
2
1
,
N
o
v
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
2
3
8
-
023
-
0
9
9
3
7
-
9.
[
1
7
]
Z.
A
.
S
o
s
n
o
w
s
k
i
a
n
d
J.
S
.
W
a
l
i
j
e
w
s
k
i
,
“
F
u
z
z
y
D
e
mp
s
t
e
r
-
S
h
a
f
e
r
m
o
d
e
l
l
i
n
g
a
n
d
d
e
c
i
si
o
n
r
u
l
e
s
,
”
i
n
C
o
m
p
u
t
e
r
I
n
f
o
rm
a
t
i
o
n
S
y
s
t
e
m
s
a
n
d
I
n
d
u
s
t
ri
a
l
M
a
n
a
g
e
m
e
n
t
:
1
5
t
h
I
FI
P
T
C
8
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
,
2
0
1
6
,
p
p
.
5
1
6
–
5
2
9
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
3
1
9
-
4
5
3
7
8
-
1
_
4
6
.
[
1
8
]
R
.
J
i
r
o
u
še
k
,
V
.
K
r
a
t
o
c
h
v
í
l
,
a
n
d
P
.
P
.
S
h
e
n
o
y
,
“
O
n
c
o
n
d
i
t
i
o
n
a
l
b
e
l
i
e
f
f
u
n
c
t
i
o
n
s
i
n
d
i
r
e
c
t
e
d
g
r
a
p
h
i
c
a
l
m
o
d
e
l
s
i
n
t
h
e
D
e
m
p
st
e
r
-
S
h
a
f
e
r
t
h
e
o
r
y
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
Ap
p
r
o
x
i
m
a
t
e
Re
a
so
n
i
n
g
,
v
o
l
.
1
6
0
,
S
e
p
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
j
a
r
.
2
0
2
3
.
1
0
8
9
7
6
.
[
1
9
]
P
.
P
.
S
h
e
n
o
y
,
“
O
n
d
i
st
i
n
c
t
b
e
l
i
e
f
f
u
n
c
t
i
o
n
s
i
n
t
h
e
D
e
m
p
s
t
e
r
-
S
h
a
f
e
r
t
h
e
o
r
y
,
”
i
n
Pro
c
e
e
d
i
n
g
s
o
f
t
h
e
T
h
i
r
t
e
e
n
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
S
y
m
p
o
s
i
u
m
o
n
I
m
p
rec
i
se
Pr
o
b
a
b
i
l
i
t
y
:
T
h
e
o
r
i
e
s
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
,
2
0
2
3
,
p
p
.
4
2
6
–
4
3
7
.
[
2
0
]
Z.
L
i
u
,
Y
.
C
a
o
,
X
.
Y
a
n
g
,
a
n
d
L.
Li
u
,
“
A
n
e
w
u
n
c
e
r
t
a
i
n
t
y
m
e
a
s
u
r
e
v
i
a
b
e
l
i
e
f
R
é
n
y
i
e
n
t
r
o
p
y
i
n
D
e
m
p
st
e
r
-
S
h
a
f
e
r
t
h
e
o
r
y
a
n
d
i
t
s
a
p
p
l
i
c
a
t
i
o
n
t
o
d
e
c
i
si
o
n
m
a
k
i
n
g
,
”
C
o
m
m
u
n
i
c
a
t
i
o
n
s
i
n
S
t
a
t
i
st
i
c
s
-
T
h
e
o
ry
a
n
d
Me
t
h
o
d
s
,
v
o
l
.
5
3
,
n
o
.
1
9
,
p
p
.
6
8
5
2
–
6
8
6
8
,
O
c
t
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
8
0
/
0
3
6
1
0
9
2
6
.
2
0
2
3
.
2
2
5
3
3
4
2
.
[
2
1
]
K
.
Q
u
,
G
.
Z
h
a
o
,
Y
.
W
u
,
a
n
d
L
.
T
o
n
g
,
“
R
e
s
e
a
r
c
h
o
n
a
i
r
sp
a
c
e
c
o
n
f
l
i
c
t
d
e
t
e
c
t
i
o
n
me
t
h
o
d
b
a
se
d
o
n
s
p
h
e
r
i
c
a
l
d
i
scre
t
e
g
r
i
d
r
e
p
r
e
s
e
n
t
a
t
i
o
n
,
”
Ap
p
l
i
e
d
S
c
i
e
n
c
e
s
,
v
o
l
.
1
3
,
n
o
.
1
1
,
M
a
y
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
a
p
p
1
3
1
1
6
4
9
3
.
[
2
2
]
D
.
P
.
P
r
i
e
t
o
,
R
.
d
e
H
a
a
n
,
a
n
d
A
.
Ö
z
g
ü
n
,
“
A
b
e
l
i
e
f
m
o
d
e
l
f
o
r
c
o
n
f
l
i
c
t
i
n
g
a
n
d
u
n
c
e
r
t
a
i
n
e
v
i
d
e
n
c
e
:
c
o
n
n
e
c
t
i
n
g
D
e
mp
s
t
e
r
-
S
h
a
f
e
r
t
h
e
o
r
y
a
n
d
t
h
e
t
o
p
o
l
o
g
y
o
f
e
v
i
d
e
n
c
e
,
”
i
n
Pro
c
e
e
d
i
n
g
s
o
f
t
h
e
T
w
e
n
t
i
e
t
h
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
Pr
i
n
c
i
p
l
e
s
o
f
K
n
o
w
l
e
d
g
e
Re
p
r
e
se
n
t
a
t
i
o
n
a
n
d
R
e
a
s
o
n
i
n
g
,
S
e
p
.
2
0
2
3
,
p
p
.
5
5
2
–
5
6
1
,
d
o
i
:
1
0
.
2
4
9
6
3
/
k
r
.
2
0
2
3
/
5
4
.
[
2
3
]
P
.
S
me
t
s
a
n
d
R
.
K
e
n
n
e
s
,
“
Th
e
t
r
a
n
sf
e
r
a
b
l
e
b
e
l
i
e
f
m
o
d
e
l
,
”
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
6
6
,
n
o
.
2
,
p
p
.
1
9
1
–
2
3
4
,
A
p
r
.
1
9
9
4
,
d
o
i
:
1
0
.
1
0
1
6
/
0
0
0
4
-
3
7
0
2
(
9
4
)
9
0
0
2
6
-
4.
[
2
4
]
K
.
S
e
n
t
z
a
n
d
S
.
F
e
r
so
n
,
“
C
o
mb
i
n
a
t
i
o
n
o
f
e
v
i
d
e
n
c
e
i
n
D
e
m
p
st
e
r
-
S
h
a
f
e
r
t
h
e
o
r
y
,
”
Re
p
o
r
t
s
,
A
l
b
u
q
u
e
r
q
u
e
,
N
M
,
a
n
d
L
i
v
e
r
mo
r
e
,
C
A
(
U
n
i
t
e
d
S
t
a
t
e
s)
,
A
p
r
.
2
0
0
2
,
d
o
i
:
1
0
.
2
1
7
2
/
8
0
0
7
9
2
.
[
2
5
]
B
.
S
u
o
,
L.
Z
h
a
o
,
a
n
d
Y
.
Y
a
n
,
“
A
n
o
v
e
l
D
e
mp
s
t
e
r
-
S
h
a
f
e
r
t
h
e
o
r
y
-
b
a
s
e
d
a
p
p
r
o
a
c
h
w
i
t
h
w
e
i
g
h
t
e
d
a
v
e
r
a
g
e
f
o
r
f
a
i
l
u
r
e
m
o
d
e
a
n
d
e
f
f
e
c
t
s
a
n
a
l
y
s
i
s
u
n
d
e
r
u
n
c
e
r
t
a
i
n
t
y
,
”
J
o
u
r
n
a
l
o
f
L
o
ss
Pre
v
e
n
t
i
o
n
i
n
t
h
e
Pr
o
c
e
ss
I
n
d
u
st
r
i
e
s
,
v
o
l
.
6
5
,
M
a
y
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
l
p
.
2
0
2
0
.
1
0
4
1
4
5
.
[
2
6
]
P
.
K
u
mari
,
S
.
R
.
N
.
R
e
d
d
y
,
a
n
d
R
.
Y
a
d
a
v
,
“
A
p
p
l
i
c
a
t
i
o
n
o
f
D
e
mp
s
t
e
r
-
S
h
a
f
e
r
t
h
e
o
r
y
i
n
se
n
s
o
r
d
a
t
a
f
u
si
o
n
,
”
i
n
P
ro
c
e
e
d
i
n
g
s
o
f
t
h
e
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
,
D
e
e
p
L
e
a
r
n
i
n
g
a
n
d
C
o
m
p
u
t
a
t
i
o
n
a
l
I
n
t
e
l
l
i
g
e
n
c
e
f
o
r
W
i
rel
e
ss
C
o
m
m
u
n
i
c
a
t
i
o
n
,
2
0
2
4
,
p
p
.
9
1
–
1
0
2
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
0
3
1
-
4
7
9
4
2
-
7
_
9
.
[
2
7
]
L.
Q
i
y
a
n
h
u
i
,
“
A
f
r
a
mew
o
r
k
f
o
r
m
u
l
t
i
-
mo
d
a
l
f
u
s
i
o
n
u
si
n
g
D
e
m
p
s
t
e
r
-
S
h
a
f
e
r
t
h
e
o
r
y
i
n
c
o
m
p
u
t
e
r
v
i
si
o
n
a
p
p
l
i
c
a
t
i
o
n
s
,
”
i
n
2
0
2
4
I
E
EE
6
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
Po
w
e
r,
I
n
t
e
l
l
i
g
e
n
t
C
o
m
p
u
t
i
n
g
a
n
d
S
y
st
e
m
s
(
I
C
PI
C
S
)
,
J
u
l
.
2
0
2
4
,
p
p
.
2
3
9
–
2
4
5
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
P
I
C
S
6
2
0
5
3
.
2
0
2
4
.
1
0
7
9
7
1
0
5
.
Evaluation Warning : The document was created with Spire.PDF for Python.