I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
p
ute
r
E
ng
in
ee
ring
(
I
J
E
CE
)
Vo
l.
8
,
No
.
5
,
Octo
b
e
r
2
0
1
8
,
p
p
.
3
5
2
3
~
3
5
3
5
I
SS
N:
2
0
8
8
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v8
i
5
.
p
p
3
5
2
3
-
3535
3523
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
co
r
e
.
co
m/
jo
u
r
n
a
ls
/in
d
ex
.
p
h
p
/
I
JE
C
E
Predict
i
ng
the
B
rand
Populari
t
y
f
rom
the
Bran
d
Metadat
a
B
ha
rg
a
v
i
K
1
,
Sa
t
his
h B
a
bu
B
2
,
S.
S.
I
y
eng
a
r
3
1
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter S
c
ien
c
e
a
n
d
E
n
g
in
e
e
rin
g
,
S
i
d
d
a
g
a
n
g
a
In
stit
u
te o
f
T
e
c
h
n
o
lo
g
y
,
T
u
m
k
u
r,
Ka
rn
a
tak
a
,
In
d
ia
2
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter S
c
ien
c
e
a
n
d
E
n
g
in
e
e
rin
g
,
R
V
C
o
ll
e
g
e
o
f
En
g
in
e
e
rin
g
,
Be
n
g
a
lu
ru
,
Ka
rn
a
t
a
k
a
,
In
d
ia
3
F
lo
ri
d
a
In
tern
a
t
io
n
a
l
Un
iv
e
rsity
,
M
iam
i,
F
lo
rid
a
,
USA
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Ma
y
2
1
,
2
0
1
8
R
ev
i
s
ed
A
u
g
2
0
,
2
0
1
8
A
cc
ep
ted
A
u
g
2
8
,
2
0
1
8
S
o
c
ial
n
e
tw
o
rk
s h
a
v
e
b
e
c
o
m
e
o
n
e
o
f
th
e
p
ri
m
a
r
y
so
u
rc
e
s o
f
b
ig
d
a
t
a
,
w
h
e
re
a
v
a
riet
y
o
f
p
o
sts
re
late
d
to
b
ra
n
d
s
a
re
li
k
e
d
,
sh
a
re
d
,
a
n
d
c
o
m
m
e
n
ted
,
w
h
ich
a
re
c
o
ll
e
c
ti
v
e
l
y
c
a
ll
e
d
a
s
b
ra
n
d
m
e
tad
a
ta.
Du
e
to
th
e
in
c
re
a
se
d
b
o
o
m
in
E/
M
-
c
o
m
m
e
rc
e
,
b
u
y
e
rs
o
f
ten
re
f
e
r
th
e
b
ra
n
d
m
e
tad
a
ta
a
s
a
v
a
lu
a
b
le
so
u
rc
e
o
f
in
f
o
rm
a
ti
o
n
to
m
a
k
e
th
e
ir
p
u
rc
h
a
sin
g
d
e
c
isio
n
.
F
ro
m
th
e
li
tera
tu
re
stu
d
y
,
w
e
f
o
u
n
d
th
a
t
th
e
re
a
re
n
o
t
m
a
n
y
w
o
rk
s o
n
p
re
d
ictin
g
th
e
p
o
p
u
larity
o
f
th
e
b
ra
n
d
b
a
se
d
o
n
th
e
c
o
m
b
in
a
ti
o
n
o
f
b
ra
n
d
m
e
tad
a
ta
a
n
d
c
o
m
m
e
n
t’s
th
o
u
g
h
tf
u
ln
e
ss
a
n
a
ly
sis.
T
h
is
p
a
p
e
r
p
ro
p
o
se
s
a
n
o
v
e
l
f
ra
m
e
w
o
rk
to
c
la
ss
i
fy
th
e
c
o
m
m
e
n
t’s
a
s
th
o
u
g
h
tf
u
l
f
a
v
o
re
d
o
r
d
isfa
v
o
re
d
c
o
m
m
e
n
t’s,
a
n
d
late
r
c
o
m
b
in
e
s
th
e
m
w
it
h
th
e
b
ra
n
d
m
e
tad
a
ta
to
f
o
re
c
a
st
th
e
p
o
p
u
larity
o
f
th
e
b
ra
n
d
in
n
e
a
r
f
u
tu
re
.
T
h
e
p
e
rf
o
r
m
a
n
c
e
o
f
th
e
p
ro
p
o
se
d
f
ra
m
e
w
o
r
k
is
c
o
m
p
a
re
d
w
it
h
so
m
e
o
f
th
e
re
c
e
n
t
w
o
rk
s
w
.
r.
t.
th
o
u
g
h
tf
u
l
c
o
m
m
e
n
t’s
id
e
n
ti
f
ica
ti
o
n
a
c
c
u
ra
c
y
,
e
x
e
c
u
ti
o
n
ti
m
e
,
p
re
d
ictio
n
a
c
c
u
ra
c
y
a
n
d
p
re
d
ictio
n
ti
m
e
,
th
e
re
su
lt
s
o
b
tain
e
d
a
re
f
o
u
n
d
to
b
e
v
e
ry
e
n
c
o
u
ra
g
in
g
.
K
ey
w
o
r
d
:
B
r
an
d
m
etad
ata
B
r
an
d
p
o
p
u
lar
ity
Ma
p
R
ed
u
ce
So
cial
n
et
w
o
r
k
s
T
h
o
u
g
h
t
f
u
l c
o
m
m
e
n
t
Co
p
y
rig
h
t
©
2
0
1
8
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
:
B
h
ar
g
av
i
K
,
Facu
lt
y
o
f
C
o
m
p
u
ter
Scien
ce
an
d
E
n
g
in
ee
r
i
n
g
,
Sid
d
ag
an
g
a
I
n
s
tit
u
te
o
f
T
ec
h
n
o
lo
g
y
,
T
u
m
k
u
r
,
I
n
d
ia.
P
h
o
n
e:
+9
1
-
9
8
8
6
2
8
0
9
3
1
E
m
ail:
b
h
ar
g
av
i.tu
m
k
u
r
@
g
m
a
il.c
o
m
1.
I
NT
RO
D
UCT
I
O
N
So
cial
n
et
w
o
r
k
s
h
av
e
co
n
n
ec
t
ed
b
illi
o
n
s
o
f
p
eo
p
le
all
o
v
er
th
e
w
o
r
ld
,
w
h
o
g
e
n
er
ate
b
ig
d
ata
in
th
e
f
o
r
m
o
f
tex
t,
i
m
a
g
e,
an
d
a
u
d
io
/v
id
eo
.
T
h
is
d
ata
w
o
u
ld
s
er
v
e
as
a
v
al
u
ab
le
s
o
u
r
ce
o
f
i
n
f
o
r
m
atio
n
f
o
r
m
an
y
b
i
g
d
ata
r
esear
ch
er
s
[
1
]
-
[
3
]
.
Du
e
t
o
in
cr
ea
s
ed
p
en
etr
atio
n
o
f
s
o
ci
aliza
tio
n
in
d
ail
y
liv
e
s
,
th
e
s
o
c
ial
n
et
w
o
r
k
i
n
g
h
as
tu
r
n
ed
o
u
t
to
b
e
a
p
r
o
m
in
e
n
t
p
latf
o
r
m
f
o
r
b
r
an
d
ad
v
er
t
is
e
m
en
ts
.
T
h
is
ad
v
er
ti
s
e
m
e
n
t
co
u
ld
b
e
p
er
s
o
n
alize
d
b
ased
o
n
th
e
cu
s
to
m
er
p
r
o
f
ile,
d
e
m
o
g
r
ap
h
y
s
p
ec
i
f
ic
in
ter
e
s
ts
,
cu
s
to
m
er
f
ee
d
b
ac
k
,
an
d
o
t
h
e
r
p
ar
am
eter
s
.
I
t h
a
s
b
ee
n
an
al
y
ze
d
t
h
at
ad
v
er
tis
er
s
w
ill
s
p
e
n
d
o
v
er
5
0
b
illi
o
n
d
o
l
l
ar
s
o
n
s
o
cial
m
ed
i
a
ad
v
er
tis
i
n
g
b
y
2
0
2
0
[
4
]
-
[
6
]
.
Sin
ce
s
u
b
s
cr
ib
er
s
h
a
v
e
th
e
f
r
ee
d
o
m
to
ex
p
r
ess
t
h
eir
o
p
in
i
o
n
s
o
n
s
o
cial
n
et
w
o
r
k
s
ite
s
,
th
e
p
latf
o
r
m
ca
n
b
e
m
is
u
s
ed
to
p
o
s
t m
ea
n
i
n
g
less
o
r
n
o
t
-
t
h
o
u
g
h
t
f
u
l c
o
m
m
e
n
t
s
o
v
er
th
e
b
r
an
d
s
[
7
]
-
[
1
0
]
.
I
n
th
is
w
o
r
k
,
a
f
r
a
m
e
w
o
r
k
i
s
p
r
o
p
o
s
ed
to
p
r
ed
ict
th
e
b
r
an
d
p
o
p
u
lar
it
y
b
ased
o
n
b
r
an
d
m
e
tad
ata
an
d
co
m
m
e
n
t
’
s
a
n
al
y
s
i
s
.
T
h
is
f
r
a
m
e
w
o
r
k
id
en
ti
f
ies
t
h
o
u
g
h
tf
u
l
co
m
m
e
n
ts
f
r
o
m
t
h
e
b
r
an
d
co
m
m
e
n
t
co
r
p
u
s
a
n
d
u
s
e
s
th
e
co
m
m
en
t
s
to
ev
al
u
at
e
th
e
cu
r
r
e
n
t
p
o
p
u
lar
it
y
o
f
th
e
b
r
an
d
.
T
h
en
p
er
f
o
r
m
p
r
ed
icti
v
e
an
al
y
tic
s
o
n
t
h
e
n
u
m
b
er
o
f
li
k
es,
t
h
e
n
u
m
b
er
o
f
s
h
ar
es,
a
n
d
t
h
e
n
u
m
b
er
o
f
id
en
ti
f
ied
t
h
o
u
g
h
t
f
u
l c
o
m
m
en
t
s
to
p
r
ed
ict
th
e
b
r
an
d
p
o
p
u
lar
it
y
s
tatu
s
in
t
h
e
n
ea
r
f
u
tu
r
e.
Ov
er
all
t
h
e
b
r
an
d
p
o
p
u
l
ar
it
y
p
r
ed
ictio
n
ai
m
s
to
an
s
w
e
r
th
e
q
u
est
io
n
,
i.e
.
,
"
W
h
a
t
p
o
p
u
lar
it
y
lev
el
t
h
e
B
r
an
d
B
w
i
ll
b
e
at
f
u
t
u
r
e
ti
m
e
T
?"
.
T
h
e
p
r
o
p
o
s
ed
th
o
u
g
h
t
f
u
l
co
m
m
e
n
t
s
id
en
ti
f
ie
r
p
r
ep
r
o
ce
s
s
es
t
h
e
co
m
m
e
n
ts
u
s
in
g
A
p
ac
h
e
Op
en
N
L
P
p
ar
s
er
,
an
d
th
e
o
p
en
n
lp
.
g
r
o
k
.
m
l.d
ec
tr
ee
class
i
s
u
s
ed
to
id
en
ti
f
y
th
e
th
o
u
g
h
t
f
u
l
co
m
m
en
ts
[
1
1
]
.
Oth
er
m
e
ta
d
ata
f
ie
ld
s
li
k
e
n
u
m
b
er
o
f
lik
e
s
,
n
u
m
b
er
o
f
s
h
ar
e
s
ar
e
co
m
b
i
n
ed
w
i
th
t
h
e
n
u
m
b
er
o
f
id
en
tif
ied
th
o
u
g
h
t
f
u
l
co
m
m
e
n
ts
to
f
o
r
ec
ast
th
e
p
o
p
u
lar
it
y
o
f
th
e
b
r
an
d
in
n
ea
r
f
u
tu
r
e.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
5
,
Octo
b
er
2
0
1
8
:
3
5
2
3
–
3
5
3
5
3524
T
h
e
p
r
o
ce
s
s
o
f
id
e
n
ti
f
y
i
n
g
t
h
o
u
g
h
t
f
u
l
co
m
m
e
n
t
s
f
r
o
m
th
e
s
o
cial
m
ed
ia
is
d
is
c
u
s
s
ed
in
[
1
2
]
,
h
er
e
t
h
e
p
r
o
b
lem
s
r
elate
d
to
t
h
e
t
h
o
u
g
h
t
f
u
l
co
m
m
e
n
ts
id
en
ti
f
icatio
n
is
d
is
c
u
s
s
ed
f
ir
s
t
t
h
e
n
t
h
e
te
x
tu
al
f
ea
t
u
r
es
o
f
t
h
e
co
m
m
e
n
t
s
alo
n
g
w
it
h
t
h
e
d
is
co
u
r
s
e
r
elatio
n
s
h
ip
a
m
o
n
g
t
h
e
w
o
r
d
s
i
n
t
h
e
co
m
m
e
n
ts
ar
e
u
s
ed
to
p
r
ed
ict
th
e
th
o
u
g
h
tf
u
l
co
m
m
en
t
s
u
s
i
n
g
lo
g
is
t
ic
r
eg
r
es
s
io
n
m
o
d
el.
T
h
e
ac
cu
r
ac
y
o
f
th
e
p
r
ed
ictio
n
f
all
s
b
elo
w
av
er
a
g
e
as
th
e
m
o
d
el
f
ails
to
id
en
t
if
y
t
h
e
n
o
n
-
li
n
ea
r
tex
t
u
al
r
elatio
n
s
h
ip
in
th
e
lar
g
e
co
r
p
u
s
o
f
co
m
m
e
n
ts
.
On
li
n
e
co
n
te
n
t
s
b
ased
p
o
p
u
la
r
it
y
p
r
ed
ictio
n
m
o
d
el
w
a
s
co
n
s
tr
u
cted
in
[
1
3
]
;
th
e
C
o
x
p
r
o
p
o
r
tio
n
al
h
az
ar
d
r
eg
r
ess
io
n
m
o
d
el
w
as
u
s
ed
f
o
r
p
r
ed
ictio
n
p
u
r
p
o
s
e
b
y
co
n
s
id
er
in
g
p
u
b
licl
y
a
v
ailab
le
m
etr
ic
s
o
f
o
n
li
n
e
co
n
ten
t
lik
e
th
r
ea
d
li
f
eti
m
e,
t
h
e
n
u
m
b
er
o
f
co
m
m
e
n
t
s
,
an
d
th
e
n
u
m
b
er
o
f
v
ie
w
s
.
Ho
w
e
v
er
,
th
e
p
o
p
u
lar
it
y
p
r
ed
ictio
n
w
as
r
e
n
d
er
ed
b
y
co
n
s
id
er
in
g
o
n
l
y
t
h
e
p
u
b
licl
y
a
v
ailab
le
m
e
tr
ics
d
u
r
in
g
th
e
f
ir
s
t
h
o
u
r
o
f
th
e
o
n
l
in
e
co
n
ten
t p
u
b
lica
tio
n
,
w
h
ich
l
i
m
its
th
e
s
ca
lab
ilit
y
o
f
t
h
e
ap
p
r
o
ac
h
.
A
tte
n
tio
n
p
r
ed
ictio
n
o
n
b
r
an
d
p
ag
es
in
s
o
cial
m
ed
ia
w
as
d
is
cu
s
s
ed
in
[
1
4
]
,
b
o
th
co
n
ten
t
an
d
n
et
w
o
r
k
f
ea
t
u
r
e
s
o
f
t
h
e
g
en
er
ated
co
m
m
en
t’
s
w
er
e
u
s
ed
to
d
eter
m
i
n
e
th
e
b
r
an
d
p
o
p
u
lar
it
y
.
T
h
e
att
en
tio
n
g
ai
n
ed
b
y
t
h
e
b
r
an
d
p
o
s
t
w
as
d
eter
m
i
n
ed
u
s
i
n
g
r
e
g
r
ess
io
n
a
n
d
class
i
f
icat
io
n
m
eth
o
d
o
lo
g
y
.
Ho
w
e
v
er
,
d
etailed
d
is
cu
s
s
io
n
o
n
th
e
ag
g
r
e
g
ated
an
al
y
s
i
s
o
f
u
s
er
-
s
p
ec
if
ic
in
ter
e
s
t
a
n
d
its
i
n
f
l
u
en
ce
o
n
th
e
p
o
p
u
lar
it
y
o
f
b
r
an
d
p
ag
es
w
er
e
n
o
t
p
r
o
v
id
ed
.
T
h
e
p
o
p
u
lar
it
y
o
f
t
h
e
n
e
w
s
ite
m
i
n
s
o
cial
m
ed
ia
w
a
s
d
is
c
u
s
s
ed
in
[
1
5
]
.
T
h
e
s
o
cial
m
ed
ia
c
o
n
s
id
er
ed
w
a
s
T
w
itter
;
t
h
e
p
o
p
u
lar
it
y
o
f
t
h
e
n
e
w
s
ite
m
w
a
s
d
eter
m
i
n
ed
b
y
e
x
tr
ac
tin
g
co
n
ten
t
f
e
at
u
r
es
f
r
o
m
t
h
e
n
e
w
s
ar
ticles.
T
h
e
m
u
lt
id
i
m
e
n
s
io
n
a
l
f
ea
t
u
r
e
s
p
ac
e
m
o
d
el
f
o
r
ev
e
r
y
n
e
w
s
ite
m
i
n
ar
ticle
s
w
as
co
n
s
tr
u
cted
,
w
h
ich
s
er
v
ed
as
a
p
r
o
m
i
n
e
n
t
in
d
icat
o
r
o
f
b
r
an
d
p
o
p
u
lar
ity
.
St
ill,
t
h
e
in
f
l
u
en
ce
o
f
in
d
i
v
id
u
al
p
r
o
p
ag
ato
r
s
o
n
b
r
an
d
p
o
p
u
lar
it
y
w
as
n
o
t c
lear
l
y
d
is
c
u
s
s
ed
.
T
h
e
[
1
6
]
,
f
o
cu
s
es
m
ain
l
y
o
n
t
h
e
s
o
cio
-
p
o
liti
ca
l
ar
ea
,
w
h
ich
h
ig
h
li
g
h
ts
t
h
e
is
s
u
es
r
elate
d
to
m
in
i
n
g
th
e
d
ata
f
r
o
m
t
h
e
s
o
cial
w
eb
,
ex
tr
ac
tio
n
o
f
o
p
in
io
n
s
r
elat
ed
to
th
e
to
p
ic,
an
d
id
en
ti
f
ic
atio
n
o
f
t
h
o
u
g
h
tf
u
l
o
p
in
io
n
s
f
r
o
m
t
h
e
co
m
m
e
n
t
’
s
av
a
ilab
le
i
n
t
h
e
s
o
cio
-
p
o
lit
ical
w
eb
s
ite
s
.
Her
e
t
h
e
Ku
l
l
b
ac
k
-
L
eib
ler
(
K
L
)
d
iv
er
g
e
n
ce
alg
o
r
it
h
m
w
as
u
s
ed
to
d
eter
m
i
n
e
r
elev
a
n
ce
wo
r
d
s
in
th
e
co
m
m
e
n
t
s
,
b
u
t
it
s
p
er
f
o
r
m
an
ce
w
as
f
o
u
n
d
to
b
e
lo
w
er
w
h
e
n
u
s
er
ab
b
r
ev
iated
w
o
r
d
s
in
t
h
e
co
m
m
en
t
s
as
t
h
e
alg
o
r
it
h
m
d
id
n
o
t
u
s
e
an
y
to
p
ic
m
o
d
eli
n
g
to
ch
ec
k
r
elev
a
n
ce
b
et
w
ee
n
t
h
e
co
m
m
en
ts
.
T
h
e
p
o
p
u
lar
it
y
o
f
t
h
e
n
e
w
s
in
s
o
cial
m
ed
ia
is
p
r
ed
icted
in
[
1
7
]
o
n
th
e
b
asis
o
f
th
e
n
u
m
b
e
r
o
f
lik
es,
s
h
ar
es,
a
n
d
t
h
e
co
m
m
e
n
t
’
s
t
h
e
ar
ticle
g
et
s
p
r
io
r
to
th
e
p
u
b
lic
atio
n
.
A
g
r
ad
ien
t
b
o
o
s
ti
n
g
m
a
ch
in
e
is
d
e
v
elo
p
ed
to
p
r
ed
ict
th
e
p
o
p
u
lar
it
y
o
f
th
e
ar
ticle
in
n
ea
r
f
u
tu
r
e
b
y
u
s
i
n
g
th
e
m
etad
ata
o
f
th
e
ar
ticle.
Her
e,
f
o
r
p
r
e
d
ictio
n
p
u
r
p
o
s
e,
th
e
o
r
ig
i
n
al
m
etad
a
ta
av
ailab
le
a
f
ter
ar
tic
le
p
u
b
l
icatio
n
i
s
n
o
t
tak
e
n
in
to
co
n
s
id
er
atio
n
a
n
d
th
e
co
m
m
e
n
t
’
s
ar
e
n
o
t
v
er
if
ied
f
o
r
th
o
u
g
h
t
f
u
l
n
es
s
s
o
th
e
p
r
ed
icti
o
n
r
ate
f
alls
o
n
a
d
ec
en
t scale
.
A
p
r
o
ac
tiv
e
s
y
s
te
m
is
p
r
o
p
o
s
ed
to
f
o
r
ec
ast
th
e
p
o
p
u
lar
it
y
o
f
th
e
o
n
lin
e
n
e
w
s
in
[
1
8
]
,
th
e
r
o
llin
g
w
i
n
d
o
w
ev
al
u
atio
n
f
o
llo
w
ed
b
y
h
il
l
cli
m
b
in
g
lo
ca
l
s
ea
r
ch
is
ex
p
lo
r
ed
o
n
a
la
r
g
e
co
llecti
o
n
o
f
n
e
w
s
d
ata
s
et
th
en
r
a
n
d
o
m
f
o
r
est
alg
o
r
it
h
m
w
a
s
r
u
n
to
class
i
f
y
t
h
e
n
e
w
s
co
n
ten
t
a
s
p
o
p
u
lar
o
r
n
o
t
p
o
p
u
lar
.
B
u
t
th
e
n
e
w
s
co
n
ten
t
s
co
llected
f
o
r
p
o
p
u
lar
it
y
p
r
ed
ictio
n
ar
e
s
tatic
in
n
atu
r
e
an
d
is
s
u
e
s
r
elate
d
to
n
atu
r
al
lan
g
u
ag
e
p
r
o
ce
s
s
in
g
o
f
t
h
e
n
e
w
s
te
x
t
ar
e
n
o
t
ta
k
e
n
i
n
to
co
n
s
id
er
at
io
n
t
h
is
li
m
it
s
t
h
e
p
r
ac
tical
ap
p
licab
ilit
y
o
f
t
h
e
s
y
s
te
m
.
T
h
e
[
1
9
]
,
d
escr
ib
es
s
ev
er
al
s
tati
s
tical
s
el
f
-
lear
n
i
n
g
f
r
a
m
e
w
o
r
k
s
u
s
e
f
u
l
f
o
r
co
n
te
n
t
p
o
p
u
lar
it
y
p
r
ed
ictio
n
.
T
w
o
m
et
h
o
d
s
ar
e
u
s
ed
to
ass
ess
th
e
co
n
te
n
t
p
o
p
u
lar
it
y
o
n
e
is
r
eg
r
ess
io
n
b
ased
an
d
o
th
er
is
class
i
f
icatio
n
b
ased
.
I
t
h
as
b
ee
n
s
aid
th
at
t
h
e
er
r
o
r
r
ate
i
n
co
n
te
n
t
p
o
p
u
lar
it
y
p
r
ed
ictio
n
is
lo
w
er
i
n
th
e
g
en
er
alize
d
ad
d
itiv
e
m
o
d
e
r
eg
r
ess
io
n
m
o
d
el
a
n
d
r
an
d
o
m
f
o
r
ec
ast
class
if
ica
tio
n
m
o
d
el
b
u
t
b
o
th
o
f
th
e
m
o
d
e
ls
s
u
f
f
er
s
f
r
o
m
o
v
er
f
i
tti
n
g
p
r
o
b
lem
d
u
e
to
th
e
p
r
ese
n
ce
o
f
n
o
i
s
e
in
th
e
co
n
ten
t
s
a
m
p
le
s
.
T
h
e
p
o
p
u
lar
ity
o
f
t
h
e
n
e
w
s
a
r
ticles
is
f
o
r
ec
asted
i
n
[
2
0
]
,
b
ased
o
n
th
e
ar
ticle
m
etad
ata
,
co
n
ten
t,
s
en
ti
m
e
n
t,
r
ea
d
ab
ilit
y
,
an
d
n
a
m
ed
en
tit
y
f
ea
tu
r
es.
Her
e
th
e
p
r
o
b
lem
o
f
p
o
p
u
lar
it
y
p
r
ed
ictio
n
is
co
n
s
id
er
ed
as
a
r
eg
r
ess
io
n
p
r
o
b
lem
a
n
d
p
r
ed
icts
th
e
n
u
m
b
er
o
f
v
ie
w
s
o
f
th
e
ar
ticle
i
n
t
h
e
f
u
t
u
r
e.
T
h
e
o
n
l
y
s
o
cial
m
ed
i
a
f
ea
t
u
r
e
co
n
s
id
er
ed
f
o
r
p
o
p
u
lar
it
y
p
r
ed
ictio
n
w
as
t
w
itter
,
th
e
s
o
cial
m
ed
ia
-
b
ased
f
ea
t
u
r
es
lik
e
t
h
e
n
u
m
b
er
o
f
r
et
w
ee
t
s
,
an
d
th
e
n
u
m
b
er
o
f
f
o
llo
w
er
s
is
u
s
ed
as
p
o
p
u
lar
it
y
m
ea
s
u
r
e
m
e
n
t
m
etr
ics.
T
h
e
p
o
p
u
lar
it
y
p
r
ed
ictio
n
m
o
d
el
s
u
f
f
er
s
f
r
o
m
s
ca
lab
ilit
y
is
s
u
e
as
o
n
l
y
s
o
cial
m
ed
ia
co
n
s
id
er
ed
f
o
r
an
al
y
s
i
s
is
t
w
i
tter
an
d
all
t
w
ee
ts
ar
e
co
n
s
id
er
ed
f
o
r
an
al
y
s
is
w
i
th
o
u
t in
v
e
s
ti
g
ati
n
g
its
t
h
o
u
g
h
tf
u
l
n
e
s
s
.
T
h
e
f
ac
to
r
s
af
f
ec
ti
n
g
th
e
p
o
p
u
lar
it
y
o
f
th
e
b
r
an
d
p
o
s
ts
ar
e
s
u
r
v
e
y
ed
in
[
2
1
]
.
T
h
e
f
ac
to
r
s
id
en
ti
f
ied
ar
e
n
u
m
b
er
o
f
li
k
es,
n
u
m
b
er
o
f
co
m
m
e
n
t
’
s
,
an
d
n
u
m
b
er
o
f
s
h
ar
es
a
n
d
th
e
r
elatio
n
s
h
i
p
b
etw
ee
n
f
a
n
s
a
n
d
b
r
an
d
p
o
s
ts
ar
e
m
ea
s
u
r
ed
in
te
r
m
s
o
f
n
u
m
b
er
o
f
n
e
w
lik
e
s
a
n
d
n
u
m
b
er
o
f
u
n
l
ik
e
s
o
v
er
th
e
p
o
s
ts
.
T
h
e
an
al
y
s
i
s
s
h
o
w
s
th
at
g
r
ee
ti
n
g
p
o
s
ts
r
ec
eiv
ed
m
o
r
e
n
u
m
b
er
o
f
li
k
es,
p
h
o
to
s
w
ith
m
e
s
s
a
g
es
r
ec
ei
v
e
d
m
o
r
e
n
u
m
b
er
o
f
s
h
ar
es,
a
n
d
w
ee
k
e
n
d
p
o
s
ts
r
ec
eiv
ed
m
o
r
e
n
u
m
b
er
o
f
co
m
m
e
n
ts
.
T
h
e
in
f
l
u
en
ce
o
f
li
k
es,
s
h
ar
es,
an
d
co
m
m
e
n
t
’
s
to
w
ar
d
s
s
p
r
ea
d
in
g
o
f
Face
b
o
o
k
m
es
s
ag
e
s
is
d
escr
ib
ed
in
[
2
2
]
.
T
h
e
s
tu
d
y
r
e
v
ea
ls
th
a
t
if
a
p
er
s
o
n
h
a
s
li
k
ed
a
m
e
s
s
a
g
e
t
h
e
n
t
h
e
c
h
an
ce
s
o
f
t
h
at
p
er
s
o
n
co
m
m
e
n
ti
n
g
o
r
s
h
ar
i
n
g
t
h
e
m
es
s
ag
e
w
ill
b
e
h
ig
h
a
s
th
er
e
e
x
is
t
s
a
h
ig
h
co
r
r
elatio
n
b
et
w
ee
n
li
k
i
n
g
,
s
h
ar
in
g
o
r
co
m
m
e
n
ti
n
g
ac
tiv
itie
s
o
f
t
h
e
en
d
u
s
er
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
P
r
ed
ictin
g
th
e
B
r
a
n
d
P
o
p
u
la
r
ity
fr
o
m
th
e
B
r
a
n
d
Meta
d
a
t
a
(
B
h
a
r
g
a
vi
K
)
3525
I
n
liter
atu
r
e,
s
ep
ar
ate
w
o
r
k
s
ex
is
t
w
it
h
r
esp
ec
t
to
th
o
u
g
h
t
f
u
l
co
m
m
e
n
ts
id
e
n
ti
f
icatio
n
an
d
b
r
an
d
p
o
p
u
lar
it
y
p
r
ed
ictio
n
p
u
r
p
o
s
e.
T
h
e
ex
is
ti
n
g
w
o
r
k
s
r
eg
ar
d
in
g
t
h
o
u
g
h
tf
u
l
co
m
m
e
n
ts
i
d
en
tific
atio
n
f
ail
to
id
en
ti
f
y
n
o
n
-
li
n
ea
r
te
x
t
u
al
a
n
al
y
s
i
s
i
n
t
h
e
co
m
m
e
n
t
s
an
d
t
h
e
ex
is
ti
n
g
w
o
r
k
s
r
eg
ar
d
in
g
p
o
p
u
lar
it
y
p
r
ed
ictio
n
s
u
f
f
er
s
f
r
o
m
o
v
er
f
i
tti
n
g
p
r
o
b
le
m
.
He
n
ce
t
h
e
p
r
o
p
o
s
ed
w
o
r
k
f
ir
s
t
id
en
ti
f
ie
s
t
h
o
u
g
h
tf
u
l
c
o
m
m
e
n
t
s
a
n
d
th
e
n
u
s
e
s
it
as
o
n
e
o
f
t
h
e
a
ttrib
u
te
s
alo
n
g
w
i
th
th
e
o
th
er
m
e
tad
ata
attr
ib
u
te
s
o
f
b
r
an
d
f
o
r
p
r
ed
ictio
n
p
u
r
p
o
s
e
w
h
ic
h
ef
f
icien
tl
y
d
ea
l
s
w
it
h
b
o
th
n
o
n
-
li
n
ea
r
tex
t
u
al
a
n
al
y
s
is
a
n
d
o
v
er
f
itti
n
g
p
r
o
b
le
m
.
2.
RE
S
E
ARCH
M
E
T
H
O
D
C
o
n
s
id
er
a
C
o
m
m
e
n
t
’
s
L
o
g
co
n
s
is
tin
g
o
f
co
m
m
e
n
t
’
s
o
n
s
e
v
er
al
B
r
an
d
Mo
d
els
(
B
Ms)
,
let
is
a
s
ter
eo
t
y
p
e
o
f
b
r
an
d
co
n
ta
in
in
g
s
e
v
er
al
B
r
an
d
P
o
s
ts
(
B
P
s
)
,
an
d
let
is
a
f
o
r
m
o
f
p
r
o
m
o
tio
n
co
m
p
r
is
i
n
g
o
f
B
r
an
d
P
o
s
t
id
en
ti
f
ier
(
)
,
s
et
o
f
B
r
an
d
P
o
s
t
C
o
m
m
e
n
t’
s
(
B
P
C
)
,
Nu
m
b
er
-
of
-
lik
e
s
(
)
,
an
d
Nu
m
b
er
-
of
-
s
h
ar
es (
)
.
Fig
u
r
e
1
.
T
h
o
u
g
h
t
f
u
l c
o
m
m
e
n
ts
id
en
ti
f
ier
f
r
a
m
e
w
o
r
k
T
h
e
Fig
u
r
e
1
d
ep
icts
th
e
f
r
a
m
e
w
o
r
k
u
s
ed
f
o
r
th
o
u
g
h
t
f
u
l
co
m
m
en
t
s
id
en
ti
f
ica
tio
n
.
T
h
e
p
r
ep
r
o
ce
s
s
o
r
clea
n
s
t
h
e
B
P
C
o
f
ev
er
y
B
P
s
in
C
L
b
y
r
e
m
o
v
i
n
g
s
to
p
w
o
r
d
s
,
n
o
is
y
tex
t,
p
u
n
ctu
a
tio
n
,
an
d
s
o
o
n
to
y
ield
p
r
ep
r
o
ce
s
s
ed
co
m
m
e
n
ts
a
s
o
u
tp
u
t,
r
elev
a
n
ce
c
h
ec
k
er
an
a
l
y
z
es
co
m
m
e
n
t
s
f
o
r
th
eir
r
el
ev
an
ce
le
v
el,
an
d
th
o
u
g
h
tf
u
l
n
es
s
i
n
v
e
s
ti
g
ato
r
in
s
p
ec
ts
th
e
th
o
u
g
h
t
f
u
ln
e
s
s
o
f
th
e
co
m
m
en
t.
A
T
h
o
u
g
h
t
f
u
l
C
o
m
m
en
t
co
m
p
r
is
e
s
o
f
t
h
o
u
g
h
tf
u
l
Qu
est
io
n
i
n
g
(
Q)
,
R
ea
s
o
n
i
n
g
(
R
)
,
E
x
p
er
ien
ce
(
E
)
,
an
d
C
o
m
p
ar
is
o
n
(
C
)
w
o
r
d
s
,
t
h
e
wo
r
d
s
m
a
y
b
e
in
f
a
v
o
r
o
r
d
is
f
av
o
r
o
f
th
e
b
r
an
d
.
T
h
e
T
h
o
u
g
h
t
f
u
l
Fav
o
r
C
o
m
m
e
n
t
i.e
.
,
ex
p
er
ien
ce
w
o
r
d
s
f
o
llo
w
ed
b
y
p
o
s
itiv
e
r
ea
s
o
n
in
g
w
o
r
d
s
o
r
co
m
p
ar
is
o
n
w
o
r
d
s
f
o
llo
w
ed
b
y
p
o
s
iti
v
e
q
u
e
s
tio
n
in
g
w
o
r
d
s
,
an
d
T
h
o
u
g
h
tf
u
l
Dis
-
Fav
o
r
C
o
m
m
en
t
i.e
.
,
ex
p
er
ien
ce
w
o
r
d
s
f
o
llo
w
ed
b
y
n
e
g
ati
v
e
r
ea
s
o
n
in
g
w
o
r
d
s
o
r
c
o
m
p
ar
is
o
n
w
o
r
d
s
f
o
llo
w
ed
b
y
n
e
g
ati
v
e
q
u
esti
o
n
in
g
w
o
r
d
s
ar
e
id
en
tif
ied
b
ased
o
n
t
h
e
s
eq
u
en
c
e
o
f
ap
p
ea
r
an
ce
o
f
th
o
u
g
h
tf
u
l
w
o
r
d
s
i
n
th
e
co
m
m
en
t.
T
h
e
s
eq
u
en
ce
b
ei
n
g
co
n
s
id
er
ed
is
i
n
s
p
ir
ed
b
y
th
e
t
h
o
u
g
h
t
p
r
o
ce
s
s
o
f
t
h
e
h
u
m
a
n
b
ein
g
s
w
h
ile
f
r
a
m
i
n
g
m
ea
n
in
g
f
u
l se
n
ten
ce
s
.
Ma
p
R
e
d
u
ce
m
o
d
el
[
2
3
]
is
u
s
ed
to
p
r
ep
r
o
ce
s
s
an
d
class
if
y
th
e
co
m
m
e
n
t
’
s
a
s
U
n
i
-
g
r
a
m
(
o
n
e
w
o
r
d
in
t
h
e
co
m
m
e
n
t)
,
B
i
-
g
r
a
m
(
t
w
o
w
o
r
d
s
i
n
t
h
e
co
m
m
en
t)
,
o
r
N
-
g
r
a
m
(
N
w
o
r
d
s
in
t
h
e
co
m
m
en
t)
.
T
h
e
p
icto
r
ical
r
ep
r
esen
tatio
n
o
f
class
i
f
icatio
n
alo
n
g
w
i
th
a
n
ex
a
m
p
le
is
s
h
o
w
n
i
n
Fig
u
r
e
2
an
d
Fig
u
r
e
3
.
As
a
s
a
m
p
le
ca
s
e,
w
e
h
av
e
co
n
s
id
er
ed
o
v
er
t
w
o
h
u
n
d
r
ed
th
o
u
s
an
d
co
m
m
e
n
t
p
o
s
ts
o
n
H
y
u
n
d
ai
ca
r
s
f
o
r
an
al
y
s
is
.
T
h
e
p
o
s
t
s
i
n
cl
u
d
e
v
ar
io
u
s
m
o
d
els o
f
H
y
u
n
d
ai
li
k
e
N
g
u
y
e
n
Van
Sa
n
g
,
H
y
u
n
d
ai
i1
0
,
H
y
u
n
d
ai
i2
0
,
A
ze
r
a,
Ge
n
esi
s
G9
0
,
Velo
s
t
er
T
u
r
b
o
,
So
n
ata
H
y
b
r
id
,
Sa
n
taFe,
a
n
d
s
o
o
n
.
A
s
ca
tter
p
lo
t
d
ep
ictin
g
t
h
e
d
is
tr
ib
u
tio
n
o
f
d
if
f
er
en
t
t
y
p
e
o
f
co
m
m
en
t’
s
i.e
.
,
u
n
i
g
r
a
m
,
b
ig
r
a
m
,
o
r
m
u
lti
g
r
a
m
co
m
m
e
n
ts
g
e
n
er
ated
o
v
er
v
ar
io
u
s
m
o
d
els
o
f
H
y
u
n
d
ai
with
r
esp
ec
t
to
ti
m
e
i
s
s
h
o
w
n
i
n
Fi
g
u
r
e
4
(
a)
.
T
h
e
p
r
ep
r
o
ce
s
s
ed
o
u
tp
u
t
o
f
th
o
s
e
co
m
m
e
n
t
s
is
d
ep
icted
in
Fi
g
u
r
e
4
(
b
)
,
w
h
ic
h
s
h
o
w
s
t
h
at
m
o
s
t
o
f
th
e
co
m
m
e
n
ts
w
er
e
m
u
lti
g
r
a
m
b
y
n
a
tu
r
e.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
5
,
Octo
b
er
2
0
1
8
:
3
5
2
3
–
3
5
3
5
3526
Fig
u
r
e
2
.
A
s
a
m
p
le
Ma
p
R
ed
u
c
e
f
r
a
m
e
w
o
r
k
f
o
r
co
m
m
en
t p
r
ep
r
o
ce
s
s
in
g
i
n
p
r
ep
r
o
ce
s
s
o
r
Fig
u
r
e
3
.
C
o
m
m
en
t p
r
ep
r
o
ce
s
s
in
g
ex
a
m
p
le
Fig
u
r
e
4
.
(
a)
T
im
e
v
er
s
u
s
n
u
m
b
er
o
f
h
y
u
n
d
ai
b
r
an
d
m
o
d
els (
b
)
T
y
p
es o
f
co
m
m
e
n
t
s
af
ter
p
r
ep
r
o
ce
s
s
in
g
v
er
s
u
s
n
u
m
b
er
o
f
co
m
m
e
n
ts
I
n
r
elev
a
n
ce
c
h
ec
k
er
,
B
P
id
,
a
n
d
P
r
ep
r
o
ce
s
s
ed
C
o
m
m
e
n
t
s
in
P
r
ep
r
o
ce
s
s
ed
C
o
m
m
en
ts
s
et
;
th
e
lex
ical
d
en
s
it
y
is
co
m
p
u
te
d
b
ased
o
n
th
e
n
u
m
b
er
o
f
lex
i
ca
l
w
o
r
d
s
in
to
p
r
o
d
u
ce
Hig
h
L
e
x
ical
De
n
s
it
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
P
r
ed
ictin
g
th
e
B
r
a
n
d
P
o
p
u
la
r
ity
fr
o
m
th
e
B
r
a
n
d
Meta
d
a
t
a
(
B
h
a
r
g
a
vi
K
)
3527
C
o
m
m
e
n
ts
s
et
[
2
4
]
-
[
2
9
]
.
T
h
e
ar
e
f
u
r
th
er
e
v
al
u
ated
f
o
r
r
elev
an
ce
u
s
i
n
g
th
e
L
ate
n
t
Di
r
ich
l
et
al
lo
ca
tio
n
to
p
ic
m
o
d
elin
g
tec
h
n
iq
u
e.
Du
r
i
n
g
to
p
ic
m
o
d
eli
n
g
p
r
o
ce
s
s
,
th
e
R
ele
v
a
n
ce
C
o
m
m
e
n
t
s
s
et
is
in
itial
ized
w
i
th
a
li
s
t
o
f
r
ele
v
a
n
ce
to
p
ic
w
o
r
d
s
f
r
o
m
th
e
co
m
m
en
ts
.
T
h
e
p
r
o
b
ab
ilit
y
o
f
co
m
m
en
ts
b
elo
n
g
i
n
g
to
i.e
.
,
is
d
eter
m
i
n
ed
,
an
d
t
h
en
e
v
er
y
ex
h
ib
iti
n
g
h
i
g
h
er
p
r
o
b
ab
ilit
y
to
w
ar
d
s
i
s
ag
g
r
e
g
ated
to
o
u
tp
u
t
.
A
n
e
x
a
m
p
le
f
o
r
id
en
ti
f
y
in
g
r
elev
a
n
t
co
m
m
e
n
t
s
f
r
o
m
is
s
h
o
w
n
in
Fi
g
u
r
e
5
,
an
d
f
r
eq
u
e
n
tl
y
o
cc
u
r
r
in
g
to
p
ic
r
elev
a
n
t
w
o
r
d
s
a
lo
n
g
w
it
h
t
h
e
to
p
ic
ir
r
elev
a
n
t
w
o
r
d
s
w
it
h
r
esp
ec
t
to
H
y
u
n
d
ai
ca
r
s
co
m
m
en
ts
is
s
h
o
w
n
in
F
ig
u
r
e
6
[
1
1
]
.
T
h
e
R
C
is
f
u
r
t
h
er
clas
s
i
f
ied
in
to
f
av
o
r
ed
o
r
d
is
f
av
o
r
ed
b
y
t
h
o
u
g
h
t
f
u
l
n
e
s
s
i
n
v
e
s
ti
g
ato
r
u
s
i
n
g
d
ec
is
io
n
tr
ee
p
r
ed
ictio
n
m
o
d
el.
T
h
e
d
ec
is
io
n
tr
ee
lo
g
ic
w
o
r
k
s
i
n
t
w
o
p
h
ases
i.e
.
,
tr
ain
in
g
an
d
te
s
ti
n
g
.
D
u
r
i
n
g
tr
ai
n
i
n
g
p
h
ase;
,
an
d
,
if
th
e
is
N
ULL
,
th
e
co
m
m
e
n
t
is
co
n
s
id
er
ed
as
n
o
t
-
t
h
o
u
g
h
t
f
u
l
co
m
m
e
n
t
w
h
ic
h
is
d
is
ca
r
d
ed
,
if
t
h
e
ex
h
ib
its
p
o
s
iti
v
e
s
eq
u
e
n
ce
o
f
w
o
r
d
s
i.e
.
,
th
en
i
t
is
co
n
s
id
er
ed
as
else
if
it
ex
h
ib
its
th
e
n
eg
at
iv
e
s
eq
u
e
n
ce
o
f
w
o
r
d
s
i.e
.
,
th
en
it
is
co
n
s
id
er
ed
as
.
Du
r
in
g
test
i
n
g
p
h
ase,
th
e
s
et
an
d
s
et
ar
e
en
u
m
er
ated
.
A
n
e
x
a
m
p
le
f
o
r
id
en
tify
i
n
g
t
h
o
u
g
h
tf
u
l
co
m
m
en
t
s
f
r
o
m
is
s
h
o
w
n
i
n
F
ig
u
r
e
7
.
T
h
e
m
o
s
t
f
r
eq
u
e
n
tl
y
o
cc
u
r
r
in
g
p
o
s
itiv
e
a
n
d
n
e
g
ati
v
e
t
h
o
u
g
h
tf
u
l
w
o
r
d
s
o
n
H
y
u
n
d
ai
s
o
cial
m
ed
ia
p
ag
es a
r
e
g
iv
e
n
i
n
T
ab
le
1
.
Fig
u
r
e
5
E
x
a
m
p
le
f
o
r
id
en
ti
f
y
i
n
g
r
ele
v
an
t c
o
m
m
en
t
Fig
u
r
e
6
T
o
p
ic
r
elev
an
t a
n
d
ir
r
elev
an
t
w
o
r
d
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
5
,
Octo
b
er
2
0
1
8
:
3
5
2
3
–
3
5
3
5
3528
Fig
u
r
e
7
.
T
h
o
u
g
h
t
f
u
l c
o
m
m
e
n
ts
id
en
ti
f
icatio
n
ex
a
m
p
le
T
h
e
th
o
u
g
h
t
f
u
ln
e
s
s
o
f
th
e
co
m
m
en
t
is
h
ig
h
l
y
i
n
f
l
u
e
n
ce
d
b
y
th
e
co
r
r
elatio
n
b
et
w
ee
n
th
e
n
u
m
b
er
o
f
w
o
r
d
s
i
n
t
h
e
co
m
m
e
n
t.
I
t
is
m
ea
s
u
r
ed
o
n
a
s
ca
le
o
f
0
to
1
,
is
d
ep
en
d
en
t
o
n
t
h
e
n
u
m
b
er
o
f
w
o
r
d
s
i
n
t
h
e
co
m
m
e
n
t,
a
s
s
h
o
w
n
i
n
t
h
e
s
ca
tter
p
lo
t
in
F
i
g
u
r
e
8
(
a)
.
T
h
e
n
u
m
b
er
o
f
t
h
o
u
g
h
tf
u
l
f
a
v
o
r
o
r
d
is
f
a
v
o
r
co
m
m
en
ts
o
n
H
y
u
n
d
ai
ca
r
s
d
u
r
i
n
g
th
e
i
n
ter
v
al
1
/1
/2
0
1
6
to
1
/3
/2
0
1
7
(
1
2
0
w
ee
k
s
)
i
s
g
i
v
en
in
F
i
g
u
r
e
8
(
b
)
.
T
w
o
t
h
ir
d
o
f
th
e
th
o
u
g
h
t
f
u
l c
o
m
m
e
n
ts
p
o
s
t
ed
w
er
e
d
is
f
a
v
o
r
ed
as
m
o
s
t o
f
th
e
m
e
x
p
r
ess
ed
t
h
e
n
e
g
ati
v
e
o
p
in
io
n
.
T
ab
l
e
1
.
T
o
p
1
0
E
x
tr
ac
ted
W
o
r
d
s
C
a
t
e
g
o
r
i
e
s o
f
w
o
r
d
s
Ex
t
r
a
c
t
e
d
w
o
r
d
s
Ex
p
e
r
i
e
n
c
e
t
r
i
a
l
,
o
b
se
r
v
a
t
i
o
n
,
so
p
h
i
st
i
c
a
t
e
,
p
r
a
c
t
i
c
e
,
f
a
mi
l
i
a
r
,
d
e
a
l
,
sp
a
r
k
,
p
r
o
d
u
c
e
,
c
h
e
c
k
,
t
r
i
e
d
C
o
mp
a
r
i
so
n
si
mi
l
a
r
,
a
l
i
k
e
,
c
o
r
r
e
l
a
t
e
,
a
n
a
l
y
z
e
,
e
x
a
mp
l
e
,
e
st
i
ma
t
i
o
n
,
me
a
su
r
e
,
c
o
l
l
a
t
e
,
t
o
g
e
t
h
e
r
,
l
i
k
e
Q
u
e
st
i
o
n
i
n
g
w
h
i
c
h
,
h
o
w
,
w
h
e
n
,
h
a
v
e
n
’
t
,
d
i
d
n
’
t
,
d
o
n
’
t
,
d
o
e
s i
t
,
w
o
n
t
,
w
o
u
l
d
,
sh
a
l
l
R
e
a
so
n
i
n
g
c
l
e
a
r
,
o
b
v
i
o
u
s,
t
h
o
u
g
h
t
,
o
p
i
n
i
o
n
,
t
h
i
n
k
,
l
i
mi
t
,
a
r
g
u
e
,
a
l
so
,
t
h
i
n
k
,
c
o
n
c
l
u
s
i
o
n
Fig
u
r
e
8
.
(
a)
Nu
m
b
er
o
f
w
o
r
d
s
in
th
e
co
m
m
e
n
t a
n
d
th
o
u
g
h
t
f
u
ln
e
s
s
o
f
th
e
co
m
m
e
n
t (
b
)
T
y
p
es o
f
co
m
m
e
n
t
s
af
ter
th
o
u
g
h
t
f
u
ln
e
s
s
i
n
v
esti
g
at
io
n
v
er
s
u
s
n
u
m
b
er
o
f
co
m
m
e
n
ts
2
.1
.
B
ra
nd
po
pu
la
rit
y
predict
o
r
T
h
e
b
r
an
d
p
o
p
u
lar
it
y
p
r
ed
icto
r
f
o
r
ec
asts
th
e
P
o
p
u
lar
it
y
I
n
d
ex
o
f
th
e
b
r
an
d
b
ased
o
n
th
e
b
r
an
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
P
r
ed
ictin
g
th
e
B
r
a
n
d
P
o
p
u
la
r
ity
fr
o
m
th
e
B
r
a
n
d
Meta
d
a
t
a
(
B
h
a
r
g
a
vi
K
)
3529
m
etad
ata
i.e
.
,
.
W
eig
h
ts
ar
e
ass
i
g
n
ed
to
ea
ch
o
f
t
h
e
m
e
tad
ata
e
m
p
ir
icall
y
,
f
o
r
e.
g
.
,
T
h
e
b
r
an
d
o
w
n
er
s
i
n
itiate
s
s
e
v
er
al
p
r
o
m
o
tio
n
a
l
ev
e
n
ts
r
elat
ed
to
n
e
w
r
elea
s
es
li
k
e
ad
d
in
g
B
M
v
id
eo
s
,
cr
ea
tin
g
r
ec
en
t
r
elea
s
es
co
v
er
p
ag
e,
f
r
eq
u
e
n
tl
y
u
p
d
atin
g
s
ta
tu
s
o
f
b
r
an
d
,
u
s
i
n
g
h
as
h
tag
s
w
it
h
th
e
pr
o
d
u
ct
n
a
m
es,
ti
m
el
y
r
esp
o
n
s
e
to
cu
s
to
m
er
s
,
a
s
k
i
n
g
q
u
esti
o
n
s
r
elate
d
to
v
ar
io
u
s
m
o
d
els
o
f
th
e
b
r
an
d
,
co
n
d
u
cti
n
g
co
n
tes
ts
o
n
t
h
e
n
e
w
m
o
d
el
la
u
n
c
h
,
cr
ea
ti
n
g
a
p
o
llin
g
s
e
s
s
io
n
o
v
er
t
h
e
b
r
an
d
p
o
s
ter
s
,
an
d
s
o
o
n
.
T
h
e
b
r
an
d
attr
ib
u
tes
{
}
in
f
l
u
e
n
c
es
th
e
p
o
p
u
lar
it
y
o
f
t
h
e
BM
o
n
o
cc
u
r
r
en
ce
o
f
E
v
en
t
j
at
ti
m
e
t,
E
j,
t
,
th
e
I
n
f
lu
en
ce
Valu
e
i
s
g
iv
e
n
b
y
∑
(
(
)
)
T
h
e
P
o
s
itiv
e
P
o
p
u
lar
ity
I
n
d
e
x
o
f
th
e
BM
(
)
∑
(
]
is
in
f
l
u
e
n
ce
d
b
y
an
d
PPI
o
f
th
e
R
elate
d
B
r
an
d
Mo
d
els
(
R
B
M
)
,
w
h
e
r
e
th
e
is
g
iv
e
n
b
y
∑
∑
[
(
)
]
.
T
h
e
Neg
ativ
e
P
o
p
u
lar
ity
I
n
d
ex
o
f
th
e
BM
(
)
(
)
is
in
f
l
u
e
n
ce
d
b
y
,
an
d
N
P
I
o
f
R
B
Ms
,
w
h
er
e
th
e
is
g
i
v
en
b
y
∑
(
)
.
As
th
e
B
r
an
d
B
co
n
tain
s
s
ev
er
al
B
M
s
,
th
e
P
P
I
an
d
N
P
I
o
f
ea
ch
o
f
th
e
BM
u
p
o
n
is
co
n
s
id
er
ed
to
d
eter
m
in
e
th
e
o
v
er
all
PPI
an
d
NP
I
o
f
th
e
b
r
an
d
,
th
er
ef
o
r
e
th
e
(
)
∑
an
d
(
)
∑
.
T
h
e
o
v
er
all
o
f
th
e
u
p
o
n
is
d
eter
m
in
ed
b
y
co
m
p
ar
in
g
t
h
e
PPI
an
d
N
P
I
o
f
B
u
p
o
n
(
)
{
(
)
(
)
A
h
y
b
r
id
ti
m
e
s
er
ies
m
o
d
el
w
it
h
A
R
I
M
A
(
A
u
to
r
eg
r
ess
i
v
e
I
n
te
g
r
ated
Mo
v
i
n
g
Av
er
ag
e)
an
d
n
eu
r
al
n
et
w
o
r
k
[
3
0
]
,
[
3
1
]
is
u
s
ed
to
co
m
p
u
te
Fo
r
ec
ast
P
o
p
u
lar
it
y
I
n
d
ex
o
f
E
v
en
t
j
at
ti
m
e
t
+
k
,
)
.
T
h
e
AR
I
M
A
p
r
ed
ictio
n
m
o
d
el
is
u
s
ed
to
d
eter
m
i
n
e
th
e
l
in
ea
r
r
el
atio
n
s
h
ip
o
v
er
th
e
p
ast
o
b
s
er
v
atio
n
s
o
f
p
o
p
u
lar
it
y
in
d
ex
an
d
th
e
n
e
u
r
al
n
et
w
o
r
k
m
o
d
el
is
u
s
ed
o
n
t
h
e
r
es
id
u
es
o
f
th
e
A
R
I
M
As
o
u
tp
u
t
to
p
r
ed
ict
th
e
n
o
n
-
lin
ea
r
r
elatio
n
s
h
ip
in
t
h
e
p
ast
o
b
s
er
v
atio
n
s
o
f
PI
,
a
1
0
0
*
5
0
*
1
0
n
e
u
r
al
n
et
w
o
r
k
m
o
d
el
u
s
ed
f
o
r
p
o
p
u
lar
it
y
p
r
ed
ictio
n
is
s
h
o
w
n
in
Fi
g
u
r
e
9
.
I
n
th
e
f
ir
s
t
s
ta
g
e,
th
e
A
R
I
M
A
p
o
p
u
lar
it
y
i
n
d
ex
is
co
m
p
u
ted
(
)
∑
[
(
)
]
∑
(
)
w
h
er
e
an
d
ar
e
e
m
p
ir
ica
l
co
n
s
ta
n
ts
.
T
h
e
n
b
y
u
s
in
g
c
o
m
p
u
ted
th
e
r
es
id
u
al
o
u
tp
u
t
o
f
AR
I
M
A
is
ca
lc
u
lated
(
)
(
)
(
)
I
n
t
h
e
s
ec
o
n
d
s
ta
g
e,
t
h
e
n
e
u
r
al
n
et
w
o
r
k
p
o
p
u
lar
it
y
i
n
d
ex
i
s
ca
lcu
lated
u
s
i
n
g
(
)
o
u
tp
u
ts
(
)
(
)
(
)
(
)
w
h
er
e
f
is
a
n
o
n
-
li
n
ea
r
n
e
u
r
al
n
et
w
o
r
k
f
u
n
ctio
n
,
is
th
e
r
an
d
o
m
er
r
o
r
an
d
th
e
f
i
n
al
co
m
p
u
ted
f
o
r
ec
asted
p
o
p
u
lar
ity
i
n
d
ex
is
o
b
tain
ed
(
)
(
)
(
)
Fig
u
r
e
9
.
1
0
0
*
5
0
*
1
0
n
e
u
r
al
n
et
w
o
r
k
m
o
d
el
f
o
r
p
r
ed
ictio
n
2
.
2
.
E
x
ec
utio
n t
i
m
e
a
na
ly
s
is
o
f
t
he
pro
po
s
ed
f
ra
m
ew
o
rk
T
h
e
to
tal
ex
ec
u
tio
n
ti
m
e
is
co
m
p
u
ted
u
s
i
n
g
p
r
ep
r
o
ce
s
s
in
g
ti
m
e
,
r
elev
an
ce
ch
ec
k
i
n
g
ti
m
e
,
an
d
th
o
u
g
h
t
f
u
l
n
e
s
s
i
n
v
esti
g
atio
n
ti
m
e
i.e
.
,
.
Si
n
ce
,
th
e
p
r
o
p
o
s
ed
f
r
a
m
e
w
o
r
k
m
a
k
e
u
s
e
o
f
Ma
p
R
ed
u
ce
p
ar
allel
p
r
o
g
r
am
m
i
n
g
m
o
d
el,
th
e
is
g
iv
e
n
b
y
th
e
s
u
m
o
f
m
ap
p
i
n
g
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
5
,
Octo
b
er
2
0
1
8
:
3
5
2
3
–
3
5
3
5
3530
ti
m
e
,
s
h
u
f
f
li
n
g
ti
m
e
,
an
d
r
e
d
u
cin
g
ti
m
e
i.e
.
,
.
T
h
e
∑
*
+
w
h
er
e,
is
th
e
co
m
m
e
n
t
lo
ad
o
n
m
ap
o
p
er
atio
n
,
is
co
s
t
o
f
m
ap
o
p
er
atio
n
,
an
d
is
th
e
ca
p
ac
it
y
o
f
m
ap
o
p
er
atio
n
.
T
h
e
∑
[
]
,
w
h
er
e
is
th
e
co
m
m
e
n
t
lo
ad
o
u
tp
u
t
o
f
m
ap
o
p
er
atio
n
,
M
is
th
e
n
u
m
b
er
o
f
m
ap
p
in
g
u
n
i
ts
,
is
th
e
ca
p
ac
it
y
o
f
s
h
u
f
f
le
o
p
er
atio
n
,
an
d
S
is
t
h
e
n
u
m
b
er
o
f
s
h
u
f
f
li
n
g
u
n
it
s
.
T
h
e
∑
[
]
,
w
h
er
e
is
th
e
co
m
m
en
t
lo
ad
o
u
t
p
u
t
o
f
s
h
u
f
f
le
o
p
er
atio
n
,
is
ca
p
ac
it
y
o
f
r
ed
u
ce
o
p
er
atio
n
,
an
d
R
is
th
e
n
u
m
b
er
o
f
r
ed
u
cin
g
u
n
its
.
T
h
e
ti
m
e
to
co
m
p
u
t
e
lex
ical
d
en
s
it
y
o
f
co
m
m
e
n
t
,
co
s
t
o
f
r
ele
v
an
c
e
ch
ec
k
er
,
an
d
ca
p
ac
it
y
o
f
r
elev
an
ce
c
h
ec
k
er
is
u
s
ed
to
co
m
p
u
te
∑
,
w
h
er
e
C
is
t
h
e
n
u
m
b
er
o
f
p
r
ep
r
o
ce
s
s
ed
co
m
m
en
t
s
.
T
h
e
∑
,
w
h
er
e
is
th
e
tr
ain
in
g
ti
m
e,
is
th
e
test
i
n
g
ti
m
e,
an
d
T
in
d
icate
s
th
e
n
u
m
b
er
o
f
tr
ain
i
n
g
s
a
m
p
le
s
.
2
.
3
.
P
a
ra
llel e
f
f
iciency
o
f
t
he
pro
po
s
ed
f
ra
m
ew
o
r
k
T
h
e
p
ar
allel
ef
f
icien
c
y
o
f
Ma
p
R
ed
u
ce
i
n
t
h
e
p
r
o
p
o
s
ed
f
r
a
m
e
w
o
r
k
CT
p
)
,
w
h
er
e,
CT
is
th
e
to
tal
co
m
p
u
tatio
n
ti
m
e,
a
n
d
CT
p
is
t
h
e
co
m
p
u
tatio
n
t
i
m
e
o
n
P
m
ac
h
i
n
es.
C
o
n
s
id
er
a
s
in
g
le
m
ac
h
i
n
e
en
v
ir
o
n
m
e
n
t,
w
h
er
e
a
g
e
n
e
r
al
co
m
p
u
ta
tio
n
tas
k
is
p
er
f
o
r
m
ed
o
n
CL
o
f
s
ize
S
.
T
h
e
(
)
(
)
,
w
h
ic
h
is
d
ec
o
m
p
o
s
ed
in
to
m
ap
,
s
h
u
f
f
le,
an
d
r
ed
u
ce
s
ta
g
es.
A
t
f
ir
s
t
co
m
p
u
tatio
n
i
s
p
er
f
o
r
m
ed
in
m
ap
s
ta
g
e
to
y
ield
as
o
u
tp
u
t.
T
h
en
co
m
p
u
tatio
n
is
p
er
f
o
r
m
ed
i
n
s
h
u
f
f
le
s
ta
g
e
to
y
ield
as
o
u
tp
u
t.
L
as
tl
y
co
m
p
u
tatio
n
i
s
p
er
f
o
r
m
ed
in
r
ed
u
ce
s
tag
e
to
y
ield
o
u
tp
u
t.
C
o
n
s
id
er
m
ap
,
s
h
u
f
f
le,
an
d
r
e
d
u
ce
s
ta
g
es
o
n
P
m
ac
h
i
n
es
p
a
r
allel
en
v
ir
o
n
m
en
t,
w
h
er
e
e
v
e
r
y
m
ac
h
i
n
e
p
er
f
o
r
m
s
m
ap
,
s
h
u
f
f
le,
a
n
d
r
e
d
u
ce
o
p
er
atio
n
s
.
T
h
e
,
in
w
h
ic
h
t
h
e
m
ap
s
tag
e
p
r
o
d
u
ce
s
as
o
u
tp
u
t,
s
h
u
f
f
le
s
ta
g
e
p
r
o
d
u
ce
s
as
o
u
tp
u
t,
an
d
r
ed
u
ce
s
tag
e
p
r
o
d
u
ce
s
as
o
u
tp
u
t.
E
x
a
m
p
le:
L
et
t
h
er
e
b
e
N
B
Ms
,
X
co
m
m
e
n
ts
o
n
ev
er
y
BM
,
an
d
Y
w
o
r
d
s
in
ev
er
y
co
m
m
en
t.
T
h
e
Ma
p
R
ed
u
ce
m
o
d
el
p
er
f
o
r
m
s
C
o
m
m
e
n
t
p
r
ep
r
o
ce
s
s
in
g
(
)
,
g
r
a
m
co
u
n
t
b
ased
Sh
u
f
f
lin
g
(
)
,
an
d
Du
p
licate
C
o
m
m
e
n
ts
r
e
m
o
v
a
l
(
)
o
p
er
at
io
n
s
.
T
h
e
ca
lc
u
lated
C
T=
N
X
Y
,
an
d
,
w
h
er
e
is
t
h
e
ti
m
e
ta
k
e
n
to
r
ea
d
p
ar
tial o
u
tp
u
t f
r
o
m
e
v
er
y
m
a
p
p
er
an
d
is
th
e
ti
m
e
ta
k
en
f
o
r
s
h
u
f
f
l
in
g
.
T
h
e
p
ar
allel
i
m
p
le
m
e
n
tatio
n
o
f
Ma
p
R
ed
u
ce
i
s
s
ca
lab
le
b
ec
au
s
e
th
e
e
f
f
icien
c
y
in
cr
ea
s
e
s
w
it
h
th
e
i
n
cr
ea
s
e
in
co
m
m
e
n
t
s
lo
g
s
ize
a
n
d
th
e
n
u
m
b
er
o
f
m
ac
h
i
n
es.
3.
RE
SU
L
T
S
A
ND
AN
AL
Y
SI
S
T
h
is
s
ec
tio
n
p
r
o
v
id
es
in
f
o
r
m
atio
n
ab
o
u
t
th
e
e
x
p
er
i
m
e
n
tal
r
esu
lts
o
f
th
e
p
r
o
p
o
s
ed
w
o
r
k
in
t
h
r
ee
s
tag
e
s
,
f
ir
s
t
t
h
e
b
r
an
d
m
etad
at
a
s
o
u
r
ce
a
n
d
d
u
r
atio
n
o
f
d
ata
co
llectio
n
i
s
i
n
d
icate
d
,
s
ec
o
n
d
th
e
e
x
p
er
i
m
e
n
tal
s
etu
p
o
f
Op
en
Stac
k
e
n
v
ir
o
n
m
en
t
f
o
r
e
x
p
er
i
m
e
n
t
p
u
r
p
o
s
e
is
d
is
c
u
s
s
ed
an
d
th
ir
d
th
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
p
r
o
p
o
s
ed
w
o
r
k
is
e
v
al
u
ated
in
t
w
o
s
ta
g
es
o
n
e
is
w
i
th
r
esp
ec
t
to
th
o
u
g
h
t
f
u
l
co
m
m
e
n
ts
id
e
n
tif
icat
io
n
an
d
o
th
er
is
w
it
h
r
esp
ec
t to
b
r
an
d
p
o
p
u
lar
it
y
p
r
ed
ictio
n
.
3
.
1
.
H
y
un
da
i bra
nd
m
et
a
da
t
a
T
h
e
co
m
m
en
t’
s
o
n
H
y
u
n
d
ai
b
r
an
d
is
o
b
tain
ed
f
r
o
m
th
e
E
d
m
u
n
d
w
eb
s
ite
w
h
ic
h
is
p
u
b
licl
y
av
ailab
le
at
[
1
1
]
.
I
t
is
o
n
e
o
f
th
e
p
o
p
u
lar
Am
er
i
ca
n
o
n
li
n
e
r
eso
u
r
ce
r
ep
o
s
ito
r
ies
f
o
r
au
to
m
o
b
ile
in
f
o
r
m
ati
o
n
,
w
h
ich
p
r
o
v
id
e
s
in
f
o
r
m
atio
n
ab
o
u
t
ca
r
ev
en
t
s
,
d
ea
ler
s
,
r
ev
ie
w
s
,
o
w
n
er
s
h
ip
a
n
d
s
o
o
n
.
T
h
e
Hy
u
n
d
ai
ca
r
d
ata
s
et
co
n
s
is
ti
n
g
o
f
ar
o
u
n
d
5
0
ca
teg
o
r
ies
w
i
th
s
ev
er
al
h
eter
o
g
en
eo
u
s
f
ea
tu
r
e
s
d
u
r
in
g
t
h
e
s
p
an
1
/1
/2
0
1
6
to
1
/
3
/2
0
1
7
is
co
n
s
id
er
ed
f
o
r
ev
alu
a
tio
n
p
u
r
p
o
s
e.
Mo
r
e
p
r
ec
is
el
y
th
r
ee
v
ar
ie
ties
o
f
c
o
m
m
e
n
t
s
w
er
e
ex
tr
ac
ted
lik
e
u
n
i
-
g
r
a
m
,
b
i
-
g
r
a
m
,
a
n
d
m
u
l
ti
-
g
r
a
m
.
I
n
o
r
d
er
to
li
m
it
th
e
b
o
u
n
d
less
n
es
s
a
n
d
h
i
g
h
er
o
r
d
er
s
p
an
n
in
g
n
at
u
r
e
o
f
t
h
e
u
s
er
co
m
m
en
ts
,
th
e
co
m
m
e
n
ts
u
n
d
er
g
o
p
o
w
er
tr
an
s
f
o
r
m
atio
n
p
r
o
ce
s
s
.
T
h
e
s
y
m
b
o
lic
f
ea
t
u
r
es
o
f
t
h
e
co
m
m
en
ts
ar
e
n
o
r
m
alize
d
u
s
i
n
g
Nai
v
e
B
a
y
es
e
n
co
d
in
g
tech
n
iq
u
e
w
h
ic
h
ef
f
icie
n
tl
y
h
an
d
les
t
h
e
m
u
l
tip
le
ca
teg
o
r
ie
s
o
f
co
m
m
e
n
ts
.
I
n
ad
d
itio
n
,
L
D
A
m
e
th
o
d
is
u
s
ed
to
ex
tr
ac
t b
r
an
d
r
elev
a
n
t
w
o
r
d
s
,
an
d
ev
e
n
g
et
t
h
e
r
ate
o
f
r
elev
an
t
an
d
ir
r
ele
v
an
t
w
o
r
d
s
[
3
2
]
.
3.
2
.
E
x
peri
m
e
nta
l set
up
T
h
e
p
r
o
p
o
s
ed
w
o
r
k
is
v
alid
at
ed
u
s
i
n
g
Op
e
n
Stac
k
p
r
i
v
ate
c
lo
u
d
test
b
ed
o
n
Op
en
C
ir
r
u
s
test
b
ed
av
ailab
le
in
HP
lab
w
eb
s
ite
[
3
3
]
,
[
3
4
]
.
I
n
o
r
d
er
to
p
r
o
ce
s
s
th
e
lar
g
e
v
o
lu
m
e
o
f
co
m
m
e
n
t
’
s
th
e
So
lid
Stat
e
Dr
iv
es
(
S
SD)
is
co
m
b
in
ed
w
i
t
h
t
h
e
Har
d
Dis
k
Dr
i
v
es
(
HD
D
)
to
ac
ce
ler
ate
th
e
co
m
m
e
n
t
’
s
p
r
ep
r
o
ce
s
s
in
g
a
n
d
class
i
f
icatio
n
r
ate.
T
h
e
co
m
m
e
n
ts
ar
e
p
r
o
ce
s
s
ed
i
n
p
ar
allel
f
r
o
m
HDF
S (
Had
o
o
p
Dis
tr
ib
u
te
d
Fil
e
S
y
s
te
m
)
an
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
P
r
ed
ictin
g
th
e
B
r
a
n
d
P
o
p
u
la
r
ity
fr
o
m
th
e
B
r
a
n
d
Meta
d
a
t
a
(
B
h
a
r
g
a
vi
K
)
3531
th
e
class
if
ied
co
m
m
e
n
ts
ar
e
w
r
itte
n
b
ac
k
to
HDFS.
T
h
e
p
er
f
o
r
m
a
n
ce
is
e
v
alu
a
ted
u
s
i
n
g
A
p
ac
h
e
Had
o
o
p
-
3
.
1
.
0
co
m
p
o
s
ed
o
f
H
DD
a
n
d
SS
D,
w
h
ich
is
ca
p
ab
le
o
f
p
e
r
f
o
r
m
i
n
g
r
ea
d
,
s
h
u
f
f
le
a
n
d
r
e
d
u
ce
o
p
er
atio
n
s
o
n
1
MB
o
f
b
r
an
d
co
m
m
e
n
t
s
.
T
o
h
an
d
le
t
h
e
v
ar
iab
ilit
y
i
n
th
e
m
etad
ata,
f
o
u
r
-
f
o
ld
v
alid
atio
n
is
ca
r
r
ied
o
u
t
b
y
d
o
in
g
d
i
f
f
er
e
n
t
le
v
el
s
o
f
p
ar
ti
tio
n
s
a
n
d
t
h
en
r
o
u
n
d
i
n
g
t
h
e
m
o
f
f
to
esti
m
ate
th
e
p
o
p
u
lar
it
y
o
f
th
e
b
r
an
d
.
T
h
e
ex
p
er
i
m
e
n
t
i
s
r
ep
ea
ted
f
o
r
1
0
ti
m
e
s
u
s
i
n
g
d
if
f
er
en
t
s
ee
d
v
alu
es
in
o
r
d
er
to
ac
h
ie
v
e
t
h
e
ac
cu
r
ate
p
r
ed
ictio
n
v
alu
e.
T
h
e
p
o
p
u
lar
it
y
p
r
ed
ictio
n
is
co
n
s
id
er
ed
as
a
b
in
ar
y
c
lass
i
f
icatio
n
p
r
o
b
lem
,
t
h
e
m
ai
n
g
o
al
is
to
p
r
ed
ict
w
h
et
h
er
th
e
p
o
p
u
lar
it
y
o
f
t
h
e
b
r
an
d
is
lo
w
o
r
h
i
g
h
i
n
t
h
e
n
ea
r
f
u
t
u
r
e.
T
h
e
co
n
tr
ib
u
t
io
n
o
f
e
v
er
y
tr
ait
o
f
t
h
e
p
r
ed
ictio
n
m
o
d
el
i
s
e
s
ti
m
ated
u
s
in
g
Xb
o
o
s
t
p
ac
k
a
g
e
i
n
R
,
an
d
t
h
e
AR
I
M
A
p
r
ed
ictio
n
m
o
d
el
is
tr
ai
n
ed
w
it
h
lear
n
in
g
r
ate=
0
.
0
0
1
5
2
,
s
am
p
li
n
g
s
ize=
0
.
6
,
s
ize
o
f
t
h
e
n
et
wo
r
k
=1
0
0
*
5
0
*
1
0
,
an
d
th
e
n
u
m
b
er
o
f
iter
atio
n
s
is
ch
o
s
en
b
y
f
o
u
r
-
f
o
ld
v
alid
atio
n
tech
n
iq
u
e.
T
h
e
id
en
t
if
ica
tio
n
o
f
t
h
o
u
g
h
tf
u
l
f
a
v
o
r
o
r
d
is
f
av
o
r
co
m
m
e
n
t
s
ad
d
ed
m
o
r
e
p
o
w
er
to
th
e
p
r
ed
ictio
n
ac
cu
r
ac
y
.
I
n
liter
atu
r
e,
s
ep
ar
ate
w
o
r
k
s
w
er
e
f
o
u
n
d
w
it
h
r
e
s
p
ec
t
to
t
h
o
u
g
h
t
f
u
l
co
m
m
e
n
t
s
id
en
t
if
ica
tio
n
a
n
d
p
o
p
u
lar
it
y
p
r
ed
ictio
n
.
B
u
t
in
t
h
e
p
r
o
p
o
s
ed
w
o
r
k
w
e
d
o
b
o
th
th
o
u
g
h
t
f
u
l
co
m
m
e
n
t
s
id
en
ti
f
icat
io
n
a
n
d
p
o
p
u
lar
it
y
p
r
ed
ictio
n
,
s
o
th
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
p
r
o
p
o
s
ed
p
r
ed
ictio
n
m
o
d
el
i
s
co
m
p
ar
ed
w
it
h
th
e
ex
i
s
ti
n
g
w
o
r
k
s
[
1
2
]
,
[
1
7
]
in
t
w
o
p
er
s
p
e
ctiv
es,
o
n
e
is
to
w
ar
d
s
th
o
u
g
h
t
f
u
l
co
m
m
e
n
t
s
id
en
ti
f
icatio
n
a
n
d
o
th
er
is
to
war
d
s
th
e
b
r
an
d
p
o
p
u
lar
ity
p
r
ed
ictio
n
.
3
.
3
.
T
ho
ug
htf
ul co
mm
e
nts ide
ntif
ica
t
io
n
Her
e
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
p
r
o
p
o
s
ed
th
o
u
g
h
t
f
u
l
co
m
m
en
t
s
id
en
ti
f
icatio
n
f
r
a
m
e
w
o
r
k
is
co
m
p
ar
ed
w
it
h
th
e
e
x
is
ti
n
g
w
o
r
k
d
is
c
u
s
s
ed
in
[
1
2
]
.
T
h
e
co
m
p
ar
is
o
n
i
s
m
ad
e
w
it
h
r
esp
ec
t
to
p
ar
a
m
eter
s
li
k
e
t
h
o
u
g
h
t
f
u
l
co
m
m
e
n
t
’
s
id
e
n
ti
f
icatio
n
ac
c
u
r
ac
y
,
ca
tc
h
r
ate
an
d
m
is
s
r
ate
o
f
th
o
u
g
h
t
f
u
l c
o
m
m
en
t
’
s
,
an
d
ex
ec
u
t
io
n
ti
m
e.
A
g
r
ap
h
o
f
ti
m
e
v
er
s
u
s
th
o
u
g
h
t
f
u
l
co
m
m
e
n
t
’
s
id
en
t
if
ica
tio
n
ac
cu
r
ac
y
is
s
h
o
w
n
i
n
F
i
g
u
r
e
1
0
.
I
t
is
o
b
s
er
v
ed
f
r
o
m
t
h
e
g
r
ap
h
th
a
t
th
e
ac
c
u
r
a
c
y
in
id
en
ti
f
y
i
n
g
t
h
e
t
h
o
u
g
h
tf
u
ln
e
s
s
o
f
t
h
e
co
m
m
e
n
ts
i
s
h
ig
h
er
in
t
h
e
p
r
o
p
o
s
ed
w
o
r
k
co
m
p
ar
ed
to
th
e
e
x
is
t
in
g
o
n
e.
T
h
e
p
r
o
p
o
s
ed
w
o
r
k
is
t
r
ain
ed
to
id
en
tify
t
h
o
u
g
h
tf
u
l
c
o
m
m
e
n
t
’
s
b
ased
o
n
th
e
o
cc
u
r
r
en
ce
o
f
f
o
u
r
t
h
o
u
g
h
t
f
u
l b
ag
s
o
f
w
o
r
d
s
E
,
R
,
C
,
an
d
Q;
m
o
s
t o
f
t
h
e
tar
g
et
s
p
ec
i
f
ic
co
m
m
e
n
ti
n
g
ter
m
s
in
s
o
cial
m
ed
ia
w
ill b
e
ei
th
er
i
n
p
o
s
iti
v
e
o
r
n
e
g
ati
v
e
i
n
cli
n
ati
o
n
o
f
t
h
e
co
n
s
id
er
ed
b
ag
o
f
wo
r
d
s
as a
r
esu
lt,
t
h
e
th
o
u
g
h
tf
u
l
co
m
m
e
n
t
’
s
id
en
ti
f
i
ca
tio
n
ac
cu
r
ac
y
in
cr
ea
s
es
o
v
e
r
ti
m
e.
B
u
t
t
h
e
ex
i
s
ti
n
g
w
o
r
k
r
elies
o
n
lo
g
i
s
tic
m
o
d
el
f
o
r
co
m
m
en
t’
s
clas
s
i
f
i
ca
tio
n
,
as
t
h
e
m
o
d
el
d
ea
ls
w
it
h
a
lo
t
o
f
i
n
d
ep
en
d
en
t
s
en
t
i
m
en
t
w
o
r
d
s
,
th
e
y
ar
e
v
u
l
n
er
ab
le
to
o
v
er
f
it
tin
g
p
r
o
b
le
m
,
th
er
e
f
o
r
e,
th
e
th
o
u
g
h
t
f
u
l
c
o
m
m
e
n
t
’
s
id
e
n
ti
f
icatio
n
ac
c
u
r
ac
y
d
ec
r
ea
s
es.
A
g
r
ap
h
o
f
co
m
m
en
t’
s
lo
g
s
iz
e
v
er
s
u
s
t
h
o
u
g
h
tf
u
l
co
m
m
en
t
’
s
ca
tch
r
ate,
m
is
s
r
ate
o
f
p
r
o
p
o
s
ed
w
o
r
k
an
d
ex
i
s
tin
g
w
o
r
k
ar
e
s
h
o
w
n
in
F
i
g
u
r
e
1
1
(
a)
an
d
Fig
u
r
e
1
1
(
b
)
.
I
t
is
o
b
s
er
v
ed
f
r
o
m
t
h
e
g
r
ap
h
t
h
at
t
h
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
p
r
o
p
o
s
ed
w
o
r
k
w
it
h
r
esp
ec
t
to
th
e
ca
tc
h
r
ate
an
d
m
i
s
s
r
ate
o
f
th
o
u
g
h
t
f
u
l
co
m
m
en
ts
i
s
f
o
u
n
d
to
b
e
g
o
o
d
.
T
h
e
ac
cu
r
ac
y
i
n
id
en
ti
f
y
i
n
g
t
h
e
t
h
o
u
g
h
tf
u
l
co
m
m
e
n
t
’
s
i
n
t
h
e
p
r
o
p
o
s
ed
w
o
r
k
is
h
ig
h
as
o
n
l
y
h
ig
h
r
elev
a
n
ce
co
m
m
e
n
t
’
s
ar
e
co
n
s
id
er
ed
f
o
r
th
o
u
g
h
t
f
u
l
n
es
s
in
v
e
s
ti
g
atio
n
a
n
d
len
g
t
h
ier
co
m
m
en
t
’
s
w
ith
v
er
y
lo
w
r
ele
v
a
n
ce
f
ac
to
r
ar
e
d
is
ca
r
d
ed
f
r
o
m
e
v
alu
a
tio
n
a
f
ter
to
p
ic
m
o
d
eli
n
g
,
th
is
f
ea
t
u
r
e
e
x
h
ib
its
h
i
g
h
co
r
r
elatio
n
w
it
h
t
h
e
q
u
alit
y
o
f
t
h
e
co
m
m
e
n
t
w
h
ich
i
n
f
lu
e
n
ce
s
t
h
e
ca
tc
h
r
ate
o
f
th
e
th
o
u
g
h
t
f
u
l
co
m
m
e
n
t
to
b
e
h
ig
h
er
an
d
th
e
m
is
s
r
ate
to
b
e
lo
w
er
.
Ho
w
e
v
er
,
in
t
h
e
ex
i
s
ti
n
g
w
o
r
k
,
t
h
e
th
o
u
g
h
t
f
u
l
co
m
m
en
t’
s
id
en
tif
icat
io
n
ac
cu
r
ac
y
is
lo
w
er
a
s
it
u
s
es
KL
d
iv
er
g
en
ce
alg
o
r
it
h
m
f
o
r
co
m
m
e
n
t
’
s
c
lass
i
f
icatio
n
,
w
h
ic
h
e
x
h
ib
it
s
v
er
y
lo
w
co
r
r
elatio
n
w
it
h
t
h
e
q
u
alit
y
o
f
th
e
co
m
m
e
n
ts
.
He
n
ce
,
t
h
e
m
is
s
r
ate
o
f
th
e
th
o
u
g
h
t
f
u
l
co
m
m
e
n
t
is
h
i
g
h
er
an
d
ca
tch
r
ate
is
lo
w
er
.
Fig
u
r
e
1
0
.
T
im
e
v
er
s
u
s
t
h
o
u
g
h
t
f
u
l c
o
m
m
en
t
s
id
en
ti
f
icat
io
n
ac
cu
r
ac
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
5
,
Octo
b
er
2
0
1
8
:
3
5
2
3
–
3
5
3
5
3532
Fig
u
r
e
1
1
.
(
a)
C
o
m
m
en
t
s
lo
g
s
ize
v
er
s
u
s
t
h
o
u
g
h
t
f
u
l c
o
m
m
e
n
ts
ca
tch
r
ate
(
b
)
C
o
m
m
en
ts
lo
g
s
ize
v
er
s
u
s
th
o
u
g
h
tf
u
l c
o
m
m
e
n
t
s
m
i
s
s
r
at
e
A
g
r
ap
h
o
f
co
m
m
en
t’
s
lo
g
s
iz
e
v
er
s
u
s
ex
ec
u
tio
n
ti
m
e
is
s
h
o
w
n
i
n
F
i
g
u
r
e
1
2
.
I
t
i
s
o
b
s
er
v
e
d
f
r
o
m
th
e
g
r
ap
h
th
at
t
h
e
to
tal
ex
ec
u
tio
n
ti
m
e
o
f
th
e
p
r
o
p
o
s
ed
th
o
u
g
h
t
f
u
l
co
m
m
e
n
ts
id
e
n
ti
f
ier
f
r
a
m
e
w
o
r
k
is
lo
w
er
co
m
p
ar
ed
to
th
e
e
x
i
s
ti
n
g
w
o
r
k
.
T
h
e
p
r
o
p
o
s
ed
f
r
am
e
w
o
r
k
u
s
es
p
ar
allel
p
r
o
g
r
a
m
m
in
g
m
o
d
el
(
Ma
p
R
ed
u
ce
)
f
o
r
co
m
m
e
n
t
’
s
p
r
ep
r
o
ce
s
s
in
g
a
n
d
th
o
u
g
h
t
f
u
l
w
o
r
d
s
s
eq
u
e
n
c
e
m
atc
h
in
g
ap
p
r
o
ac
h
f
o
r
th
o
u
g
h
t
f
u
l
co
m
m
en
t
s
id
en
ti
f
icatio
n
,
w
h
ic
h
s
p
ee
d
s
u
p
th
e
e
x
ec
u
tio
n
o
f
th
e
f
r
a
m
e
w
o
r
k
w
i
th
th
e
i
n
cr
ea
s
e
i
n
c
o
m
m
e
n
t
’
s
lo
g
s
ize.
W
h
er
ea
s
in
th
e
e
x
is
t
in
g
w
o
r
k
s
eq
u
en
tial
s
tep
s
ar
e
f
o
llo
w
ed
f
o
r
co
m
m
en
t
’
s
clea
n
i
n
g
,
to
p
ic
m
o
d
eli
n
g
,
d
is
co
u
r
s
e
r
elatio
n
s
h
ip
es
ti
m
ati
o
n
,
an
d
lo
g
is
tic
r
eg
r
e
s
s
io
n
w
h
ich
i
n
cr
ea
s
e
th
e
ex
ec
u
tio
n
ti
m
e
w
it
h
t
h
e
i
n
cr
ea
s
e
in
co
m
m
e
n
t
’
s
lo
g
s
ize.
Fig
u
r
e
1
2
.
C
o
m
m
e
n
t
s
lo
g
s
ize
v
er
s
u
s
ex
ec
u
tio
n
ti
m
e
3
.
3
.
P
o
pu
la
rit
y
predict
io
n
Her
e
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
p
r
o
p
o
s
ed
p
o
p
u
lar
it
y
p
r
ed
ictio
n
m
o
d
el
is
co
m
p
ar
ed
w
it
h
th
e
ex
is
ti
n
g
w
o
r
k
d
i
s
cu
s
s
ed
i
n
[
1
7
]
.
T
h
e
co
m
p
ar
is
o
n
i
s
m
ad
e
w
i
th
r
esp
ec
t
to
p
ar
am
e
ter
s
li
k
e
p
r
ed
ictio
n
ef
f
icie
n
c
y
,
p
r
ed
ictio
n
ac
cu
r
ac
y
,
an
d
p
r
e
d
ictio
n
ti
m
e.
A
g
r
ap
h
o
f
tr
ain
i
n
g
ep
o
ch
s
v
er
s
u
s
e
f
f
ic
ie
n
c
y
o
f
th
e
v
ar
io
u
s
p
r
ed
ictio
n
m
o
d
els
is
s
h
o
w
n
i
n
F
ig
u
r
e
1
3
.
T
h
e
p
r
o
p
o
s
ed
w
o
r
k
u
s
es
AR
I
M
A
-
Ne
u
r
al
-
Net
w
o
r
k
(
A
R
I
M
A
-
NN)
m
o
d
el
f
o
r
p
r
ed
ictio
n
p
u
r
p
o
s
e,
th
e
ef
f
icie
n
c
y
o
f
it
is
co
m
p
ar
ed
w
it
h
th
e
w
ell
-
k
n
o
w
n
p
r
ed
ictio
n
m
o
d
el
li
k
e
Neu
r
al
Net
w
o
r
k
(
NN)
,
Gr
ad
i
en
t
B
o
o
s
tin
g
T
r
ee
(
GB
T
)
,
an
d
AR
I
M
A
.
I
t
i
s
o
b
s
er
v
ed
f
r
o
m
t
h
e
g
r
ap
h
t
h
at
th
e
ef
f
icien
c
y
o
f
th
e
NN
m
o
d
el
i
s
v
er
y
lo
w
a
s
it id
en
t
if
ies o
n
l
y
n
o
n
-
li
n
ea
r
r
elatio
n
s
h
ip
s
i
n
th
e
b
r
an
d
m
etad
ata,
t
h
e
ef
f
icien
c
y
o
f
th
e
GB
M
is
m
o
d
er
ate
as
it
ca
n
n
o
t
e
x
tr
ap
o
late
to
u
n
k
n
o
w
n
r
elatio
n
s
h
ip
s
o
f
m
etad
ata
s
a
m
p
les,
th
e
ef
f
icie
n
c
y
o
f
th
e
AR
I
M
A
al
s
o
f
alls
i
n
m
o
d
er
ate
r
an
g
e
as
it
id
en
ti
f
ies
o
n
l
y
li
n
ea
r
r
elatio
n
s
h
ip
in
th
e
b
r
an
d
m
etad
ata,
w
h
er
ea
s
t
h
e
ef
f
icie
n
c
y
o
f
A
R
I
M
A
-
NN
is
h
i
g
h
er
b
ec
au
s
e
it
is
ab
le
to
id
en
tify
lin
ea
r
,
n
o
n
-
li
n
ea
r
,
an
d
u
n
k
n
o
w
n
r
elatio
n
s
h
ip
s
i
n
th
e
b
r
an
d
m
e
tad
ata.
Evaluation Warning : The document was created with Spire.PDF for Python.