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.
14
,
No
.
6
,
Dec
em
b
er
2
0
2
5
,
p
p
.
4
6
6
1
~
4
6
7
4
I
SS
N:
2
2
5
2
-
8
9
3
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijai.v
14
.i
6
.
p
p
4
6
6
1
-
4
6
7
4
4661
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
Enha
ncing
com
munica
tion a
nd in
t
eractio
n in
the
m
o
v
ie
i
ndus
try ba
sed
Sp
a
rkM
Llib'
s reco
mm
enda
tion sy
st
em
Sa
id Cha
k
o
uk
1
,
Abdelk
er
im
Z
it
o
un
i
1
,
Na
zif
T
cha
g
a
f
o
2
,
Ahio
d B
ela
id
2
1
D
e
p
a
r
t
me
n
t
o
f
C
o
mm
u
n
i
c
a
t
i
o
n
S
c
i
e
n
c
e
s
a
n
d
Te
c
h
n
i
q
u
e
s,
M
o
h
a
mm
e
d
V
U
n
i
v
e
r
s
i
t
y
i
n
R
a
b
a
t
,
R
a
b
a
t
,
M
o
r
o
c
c
o
2
D
e
p
a
r
t
me
n
t
o
f
S
c
i
e
n
c
e
P
h
y
si
c
,
F
a
c
u
l
t
y
o
f
S
c
i
e
n
c
e
s,
M
o
h
a
mm
e
d
V
U
n
i
v
e
r
s
i
t
y
i
n
R
a
b
a
t
,
R
a
b
a
t
,
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
Ap
r
6
,
2
0
2
5
R
ev
is
ed
J
u
l
26
,
2
0
2
5
Acc
ep
ted
Au
g
6
,
2
0
2
5
In
th
e
e
v
e
r
-
e
v
o
lv
i
n
g
la
n
d
sc
a
p
e
o
f
stre
a
m
in
g
p
latf
o
rm
s,
re
c
o
m
m
e
n
d
a
ti
o
n
sy
ste
m
s
c
o
n
tri
b
u
te
sig
n
ifi
c
a
n
tl
y
t
o
e
n
h
a
n
c
in
g
th
e
u
se
r
e
x
p
e
rien
c
e
.
Th
is
a
rti
c
le
e
x
a
m
in
e
s
th
e
sig
n
ifi
c
a
n
c
e
o
f
th
e
s
e
sy
ste
m
s
in
su
g
g
e
stin
g
m
o
v
ies
,
a
n
a
ly
z
in
g
th
e
ir
imp
a
c
t
o
n
u
se
r
sa
ti
sfa
c
ti
o
n
a
n
d
p
latfo
rm
p
e
rf
o
rm
a
n
c
e
.
Util
izin
g
S
p
a
rk
M
Ll
ib
,
a
p
o
we
rfu
l
t
o
o
l
fo
r
larg
e
-
sc
a
le
d
a
ta
p
r
o
c
e
ss
in
g
,
we
e
x
p
l
o
re
v
a
rio
u
s
re
c
o
m
m
e
n
d
a
ti
o
n
tec
h
n
iq
u
e
s,
in
c
l
u
d
i
n
g
c
o
ll
a
b
o
ra
ti
v
e
f
il
t
e
rin
g
a
n
d
c
o
n
ten
t
-
b
a
se
d
fi
lt
e
rin
g
.
We
h
ig
h
li
g
h
t
t
h
e
d
ime
n
sio
n
o
f
d
ig
it
a
l
c
o
m
m
u
n
ica
ti
o
n
to
fu
rt
h
e
r
e
n
h
a
n
c
e
th
e
a
c
c
u
ra
c
y
o
f
re
c
o
m
m
e
n
d
a
ti
o
n
s
a
n
d
fo
ste
r
g
re
a
ter
u
se
r
e
n
g
a
g
e
m
e
n
t.
Ou
r
stu
d
y
a
lso
a
d
d
re
ss
e
s
th
e
c
h
a
ll
e
n
g
e
s
a
n
d
fu
tu
re
o
p
p
o
rtu
n
it
ies
re
late
d
to
re
c
o
m
m
e
n
d
a
ti
o
n
sy
ste
m
s,
e
m
p
h
a
siz
in
g
th
e
n
e
e
d
fo
r
tra
n
sp
a
re
n
c
y
a
n
d
e
th
ica
l
a
l
g
o
rit
h
m
s.
Th
is
re
se
a
rc
h
h
ig
h
li
g
h
ts
th
e
p
o
t
e
n
ti
a
l
fo
r
re
c
o
m
m
e
n
d
a
ti
o
n
sy
ste
m
s to
re
v
o
l
u
ti
o
n
ize
t
h
e
d
ig
i
tal
e
n
terta
in
m
e
n
t
lan
d
sc
a
p
e
a
n
d
sh
a
p
e
th
e
f
u
tu
re
o
f
t
h
e
m
o
v
ie
in
d
u
stry
.
K
ey
w
o
r
d
s
:
Dig
ital c
o
m
m
u
n
icatio
n
Mo
v
ie
in
d
u
s
tr
y
R
ec
o
m
m
en
d
atio
n
s
y
s
tem
s
Sp
ar
k
ML
lib
Stre
am
in
g
p
latf
o
r
m
s
User
s
atis
f
ac
tio
n
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Said
C
h
ak
o
u
k
Dep
ar
tm
en
t o
f
C
o
m
m
u
n
icatio
n
Scien
ce
s
an
d
T
ec
h
n
iq
u
es
,
M
o
h
am
m
ed
V
Un
iv
er
s
ity
in
R
ab
at
Av
en
u
e
d
es Na
tio
n
s
Un
ies,
A
g
d
al,
R
ab
at
,
Ma
r
o
c
B
.
P:
8
0
0
7
.
N.
U
.
,
R
ab
at,
Mo
r
o
cc
o
E
m
ail:
s
.
ch
ak
o
u
k
@
u
m
5
r
.
ac
.
m
a
1.
I
NT
RO
D
UCT
I
O
N
I
n
to
d
ay
'
s
d
ata
-
d
r
iv
en
wo
r
ld
,
t
h
e
d
elu
g
e
o
f
in
f
o
r
m
atio
n
p
r
esen
ts
a
s
ig
n
if
ican
t
ch
allen
g
e:
h
o
w
to
ex
tr
ac
t
m
ea
n
in
g
f
u
l
in
s
ig
h
ts
an
d
p
r
o
v
i
d
e
p
er
s
o
n
alize
d
e
x
p
er
ien
ce
s
at
s
ca
le
[
1
]
,
[
2
]
.
T
h
is
p
r
o
b
lem
is
p
ar
ticu
lar
ly
r
elev
a
n
t
in
th
e
r
ea
lm
o
f
r
ec
o
m
m
en
d
ati
o
n
s
y
s
tem
s
,
wh
e
r
e
u
s
er
s
ar
e
o
f
ten
o
v
er
wh
elm
e
d
b
y
v
ast
ca
t
alo
g
s
o
f
p
r
o
d
u
cts
o
r
co
n
ten
t
[
3
]
.
Ad
d
r
ess
in
g
th
is
g
lo
b
al
n
ee
d
f
o
r
ef
f
icien
t
an
d
s
c
alab
le
p
er
s
o
n
alize
d
r
ec
o
m
m
en
d
atio
n
s
,
esp
ec
ially
with
in
th
e
co
n
tex
t
o
f
lar
g
e
d
at
asets
,
is
th
e
co
r
e
f
o
cu
s
o
f
t
h
is
p
ap
er
[
4
]
.
Ou
r
ap
p
r
o
ac
h
is
b
ased
o
n
ar
tif
icial
in
tellig
en
ce
an
d
co
llab
o
r
ativ
e
f
ilter
in
g
,
an
d
we
s
p
ec
if
ic
ally
f
o
cu
s
u
s
in
g
Ap
ac
h
e
Sp
ar
k
ML
ib
f
o
r
b
ig
d
ata
a
n
aly
tics
[
5
]
.
W
e
ad
d
r
ess
th
e
ch
allen
g
e
o
f
b
u
ild
in
g
a
s
ca
lab
le
an
d
h
ig
h
-
q
u
ality
m
o
v
ie
r
ec
o
m
m
en
d
ati
o
n
s
y
s
tem
f
o
r
lar
g
e
-
s
ca
le
u
s
er
-
item
in
ter
ac
tio
n
d
ata.
No
w
th
at
we
h
av
e
h
u
g
e
d
atasets
,
we
ca
n
u
s
e
th
e
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
o
f
Sp
ar
k
ML
ib
to
p
r
o
ce
s
s
th
em
an
d
a
n
aly
ze
th
em
to
g
iv
e
o
u
t a
cc
u
r
ate
a
n
d
p
e
r
s
o
n
alize
d
m
o
v
ie
r
ec
o
m
m
en
d
ati
o
n
s
[
6
]
.
P
r
e
v
i
o
u
s
r
e
s
e
ar
c
h
i
n
th
i
s
a
r
ea
o
f
t
en
i
n
v
e
s
t
i
g
a
te
s
d
i
f
f
e
r
e
n
t
c
o
l
l
a
b
o
r
a
t
i
v
e
f
i
l
t
e
r
in
g
a
p
p
r
o
a
c
h
e
s
t
h
a
t
l
e
v
e
r
ag
e
c
l
a
s
s
i
c
m
a
ch
i
n
e
l
ea
r
n
i
n
g
l
i
b
r
a
r
i
e
s
[
7
]
.
A
l
th
o
u
g
h
t
h
e
s
e
m
e
t
h
o
d
s
h
av
e
p
er
f
o
r
m
e
d
w
e
l
l
o
n
s
m
a
l
l
d
a
t
a
s
e
t
s
,
t
h
ey
d
o
n
o
t
s
c
a
l
e
we
l
l
w
i
t
h
th
e
e
x
p
o
n
e
n
t
i
al
g
r
o
wt
h
o
f
u
s
er
d
a
t
a
[
8
]
.
B
a
s
e
d
o
n
t
h
e
s
e
f
o
u
n
d
a
t
i
o
n
s
,
t
h
i
s
p
ap
e
r
s
o
l
v
e
s
th
e
s
e
l
im
it
a
t
i
o
n
s
w
i
th
t
h
e
u
s
e
o
f
t
h
e
d
i
s
t
r
i
b
u
t
ed
p
r
o
ce
s
s
i
n
g
c
a
p
ab
i
l
i
t
i
e
s
o
f
f
er
e
d
b
y
S
p
a
r
k
ML
i
b
[
9
]
.
W
e
co
n
tr
ib
u
te
to
th
e
ap
p
licatio
n
o
f
Sp
a
r
k
ML
ib
i
n
c
r
ea
tin
g
h
ig
h
-
p
er
f
o
r
m
a
n
ce
m
o
v
ie
r
ec
o
m
m
en
d
atio
n
s
y
s
tem
.
Yo
u
will
lear
n
h
o
w
to
co
n
f
ig
u
r
e
an
d
tu
n
e
Sp
ar
k
ML
i
b
alg
o
r
ith
m
s
to
m
ak
e
ac
c
u
r
ate
r
ec
o
m
m
en
d
atio
n
s
o
n
r
ea
l
-
wo
r
l
d
d
atasets
.
I
n
ad
d
itio
n
,
we
will
lo
o
k
in
to
b
e
n
ef
its
S
p
ar
k
ML
ib
p
r
o
v
id
es
with
its
d
is
tr
ib
u
ted
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.
14
,
No
.
6
,
Dec
em
b
er
20
25
:
4
6
6
1
-
4
6
7
4
4662
p
r
o
ce
s
s
in
g
ca
p
ab
ilit
ies
an
d
its
ju
s
tific
atio
n
with
r
esp
ec
t
to
lar
g
e
s
ca
le
r
ec
o
m
m
en
d
atio
n
task
s
[
1
0
]
.
T
h
e
r
esear
ch
p
r
o
v
id
es
a
v
al
u
ab
le
r
eso
u
r
ce
f
o
r
d
e
v
elo
p
er
s
a
n
d
r
esear
ch
er
s
in
ter
ested
in
b
u
ild
in
g
s
ca
lab
le
an
d
ef
f
icien
t
r
ec
o
m
m
en
d
atio
n
s
y
s
tem
s
with
b
ig
d
ata
tech
n
o
lo
g
y
.
2.
RO
L
E
O
F
RE
CO
M
M
E
N
D
AT
I
O
N
S
YS
T
E
M
S I
N
T
H
E
M
O
VI
E
I
ND
UST
RY
2
.
1
.
Sig
nifica
nce
o
f
re
co
m
mend
a
t
io
n sy
s
t
em
s
R
ec
o
m
m
en
d
atio
n
s
y
s
tem
s
ar
e
o
n
e
o
f
th
e
k
ey
co
m
p
o
n
e
n
ts
f
o
r
im
p
r
o
v
in
g
u
s
er
ex
p
er
ien
ce
o
n
s
tr
ea
m
in
g
p
latf
o
r
m
s
b
y
p
r
o
v
id
in
g
th
e
u
s
er
with
m
o
v
ies
th
e
y
m
ay
en
jo
y
[
1
1
]
,
[
1
2
]
,
an
d
a
h
y
b
r
id
r
ec
o
m
m
en
d
atio
n
s
y
s
tem
ca
n
en
h
a
n
ce
m
o
v
ie
s
u
g
g
esti
o
n
s
[
1
3
]
.
B
ased
o
n
an
aly
s
is
o
f
u
s
er
p
r
ef
e
r
en
ce
s
an
d
b
e
h
av
io
r
o
n
d
ig
ital
m
ed
iu
m
s
o
f
co
m
m
u
n
icatio
n
,
th
ese
s
y
s
tem
s
s
u
p
p
ly
p
er
s
o
n
alize
d
r
ec
o
m
m
en
d
atio
n
s
th
at
r
ef
lect
in
d
iv
id
u
al
tast
e,
ea
s
in
g
th
e
task
o
f
an
ev
e
r
-
ex
p
an
d
in
g
u
n
iv
er
s
e
o
f
n
ew
c
o
n
ten
t u
s
er
s
ar
e
lik
ely
to
en
jo
y
[
1
4
]
.
T
h
is
,
in
tu
r
n
,
lead
s
to
g
r
ea
te
r
u
s
er
s
atis
f
ac
tio
n
,
ex
ten
d
e
d
v
ie
win
g
s
ess
io
n
s
,
an
d
im
p
r
o
v
ed
r
eten
tio
n
o
n
th
e
p
ar
t
o
f
s
tr
ea
m
i
n
g
p
latf
o
r
m
s
[
1
5
]
.
R
ec
en
t
ad
v
an
ce
m
e
n
ts
h
av
e
in
clu
d
ed
t
h
e
in
teg
r
atio
n
o
f
d
ee
p
lear
n
in
g
tech
n
iq
u
es
with
tr
ad
i
tio
n
al
co
llab
o
r
ativ
e
f
ilter
in
g
m
eth
o
d
s
,
en
h
a
n
cin
g
t
h
e
ef
f
ec
tiv
en
ess
o
f
t
h
ese
s
y
s
tem
s
[
1
6
]
.
2
.
2
.
Dig
it
a
l c
o
m
m
un
ica
t
io
n
a
nd
re
co
m
m
enda
t
io
n sy
s
t
ems
O
u
r
v
i
e
w
o
n
d
a
t
a
co
l
l
e
c
t
io
n
f
r
o
m
t
h
e
d
i
g
i
ta
l
c
o
m
m
u
n
i
c
a
t
io
n
f
o
r
o
u
r
m
o
v
i
e
r
e
c
o
m
m
en
d
a
t
io
n
p
r
o
je
c
t
is
b
a
s
ed
o
n
o
n
l
i
n
e
co
m
m
u
n
i
ca
t
i
o
n
s
l
i
k
e
m
o
v
ie
r
e
v
i
e
w
s
,
r
a
ti
n
g
s
,
a
n
d
s
o
c
i
a
l
m
ed
i
a
a
c
t
iv
i
ti
e
s
[
1
7
]
,
[
1
8
]
.
T
h
i
s
e
n
a
b
le
d
u
s
t
o
f
e
e
d
o
u
r
m
o
d
e
ls
w
i
t
h
t
h
e
r
e
l
e
v
a
n
t
c
o
n
t
ex
t
u
a
l
a
n
d
b
eh
a
v
io
r
a
l
d
a
t
a
,
im
p
r
o
v
in
g
t
h
e
p
r
e
c
i
s
io
n
o
f
t
h
e
r
e
c
o
m
m
en
d
a
t
io
n
s
a
n
d
d
i
f
f
e
r
en
t
m
e
t
h
o
d
s
l
ik
e
c
o
l
l
ab
o
r
at
i
v
e
f
i
l
t
e
r
in
g
,
co
n
t
en
t
-
b
a
s
e
d
f
il
t
e
r
i
n
g
,
a
n
d
h
y
b
r
i
d
a
p
p
r
o
a
ch
e
s
[
1
9
]
,
[
2
0
]
.
W
e
i
m
p
l
e
m
en
t
e
d
r
e
co
m
m
e
n
d
a
t
io
n
m
o
d
e
l
s
u
s
i
n
g
th
e
P
y
th
o
n
p
r
o
g
r
a
m
m
i
n
g
l
an
g
u
a
g
e
a
n
d
u
s
i
n
g
A
p
ac
h
e
S
p
a
r
k
f
r
am
e
w
o
r
k
[
2
1
]
,
[
2
2
]
to
en
ab
l
e
f
a
s
t
p
r
o
c
e
s
s
i
n
g
o
f
l
ar
g
e
d
a
t
a
s
e
t
s
[
2
3
]
.
T
h
e
l
a
r
g
e
l
i
b
r
ar
i
e
s
a
v
a
il
a
b
l
e
i
n
Py
t
h
o
n
f
o
r
d
a
t
a
a
n
a
ly
s
i
s
a
n
d
m
ac
h
in
e
l
e
a
r
n
i
n
g
,
a
lo
n
g
w
i
th
t
h
e
d
i
s
t
r
i
b
u
te
d
co
m
p
u
t
in
g
a
b
i
l
i
t
y
o
f
Sp
a
r
k
[
2
4
]
,
a
l
lo
w
e
d
u
s
t
o
i
m
p
l
e
m
e
n
t
s
c
a
la
b
l
e
r
e
co
m
m
en
d
a
t
i
o
n
s
y
s
t
e
m
s
o
n
h
u
g
e
v
o
lu
m
e
o
f
m
o
v
i
e
d
a
t
a
[
2
5
]
,
[
2
6
]
.
2
.
3
.
Dig
it
a
l c
o
m
m
un
ica
t
io
n
a
nd
re
co
m
m
enda
t
io
n sy
s
t
ems
R
ec
o
m
m
en
d
atio
n
s
y
s
tem
s
wer
e
r
ig
o
r
o
u
s
ly
test
ed
an
d
an
aly
z
ed
,
with
m
eth
o
d
s
u
s
in
g
p
r
ec
is
io
n
,
r
ec
all
,
an
d
m
ea
n
av
er
a
g
e
p
r
ec
is
io
n
(
MA
P)
as
ev
alu
atio
n
m
etr
ics
[
2
7
]
,
[
2
8
]
.
T
o
e
n
s
u
r
e
h
ig
h
ac
cu
r
ac
y
an
d
r
elev
an
ce
o
f
m
o
v
ie
r
ec
o
m
m
en
d
ati
o
n
s
,
we
co
n
d
u
cte
d
iter
ativ
e
o
p
tim
izatio
n
as
well
as
ex
p
er
im
en
tatio
n
to
f
in
e
-
tu
n
e
o
u
r
m
o
d
els
[
2
9
]
.
H
o
wev
er
,
b
y
in
te
g
r
atin
g
d
ig
ital
in
ter
ac
tio
n
s
,
we
u
tili
ze
d
tech
n
iq
u
es
lik
e
cr
o
s
s
-
v
alid
atio
n
an
d
A/B
test
in
g
to
v
alid
ate
th
e
r
eliab
ilit
y
an
d
e
f
f
ec
tiv
en
ess
o
f
th
e
r
ec
o
m
m
en
d
atio
n
alg
o
r
ith
m
s
[
3
0
]
.
2
.
4
.
I
m
pa
c
t
o
f
re
co
m
m
enda
t
io
n sy
s
t
em
s
o
n dig
it
a
l us
er
eng
a
g
em
ent
T
h
e
in
clu
s
io
n
o
f
r
ec
o
m
m
en
d
atio
n
s
y
s
tem
s
an
d
o
n
lin
e
co
m
m
u
n
icatio
n
in
s
tr
ea
m
in
g
p
l
atf
o
r
m
s
h
as
tr
an
s
f
o
r
m
ed
h
o
w
th
e
u
s
er
s
id
en
tify
an
d
co
n
s
u
m
e
f
ilm
s
[
3
1
]
.
T
h
ese
s
y
s
tem
s
im
p
r
o
v
e
u
s
er
i
n
ter
ac
tio
n
,
in
c
r
ea
s
e
tim
e
s
p
en
t
o
n
t
h
e
p
latf
o
r
m
,
a
n
d
d
ev
elo
p
a
s
en
s
e
o
f
lo
y
alty
to
war
d
s
s
tr
ea
m
in
g
p
latf
o
r
m
s
b
y
p
r
o
v
id
in
g
s
p
ec
if
ic
s
u
g
g
esti
o
n
s
b
ased
o
n
th
e
u
s
er
p
r
ef
er
e
n
ce
s
an
d
h
is
/h
er
r
esp
ec
tiv
e
d
ig
ital
f
o
o
t
p
r
in
t
[
3
2
]
,
[
3
3
]
.
I
n
ad
d
itio
n
,
r
ec
o
m
m
en
d
atio
n
s
y
s
tem
s
h
el
p
p
o
p
u
lar
ize
d
iv
er
s
e
co
n
ten
t
av
ailab
ilit
y
[
2
9
]
,
wh
ic
h
ex
t
en
d
s
th
e
r
ea
c
h
o
f
f
ilm
m
ak
er
s
an
d
p
r
o
m
o
tes h
ete
r
o
g
en
eity
a
n
d
in
n
o
v
atio
n
in
th
e
f
ilm
s
ec
to
r
[
2
8
]
,
[
3
4
]
.
2
.
5
.
Cha
lleng
es a
nd
f
uture
direct
io
ns
Alth
o
u
g
h
b
en
e
f
icial,
r
ec
o
m
m
e
n
d
atio
n
s
y
s
tem
s
also
s
u
f
f
er
f
r
o
m
p
r
i
v
ac
y
is
s
u
es,
alg
o
r
ith
m
ic
a
d
v
er
s
ar
ial
b
iases
[
3
5
]
,
an
d
th
e
n
ec
ess
it
y
to
tr
ac
e
th
e
p
r
ev
ailin
g
tr
en
d
s
o
f
u
s
er
p
r
ef
er
e
n
ce
s
o
v
er
t
im
e
with
d
y
n
am
ic
ad
ap
tatio
n
ac
r
o
s
s
v
ar
io
u
s
d
o
m
ain
s
an
d
i
n
d
u
s
tr
ies
[
3
4
]
,
[
3
6
]
.
So
l
v
in
g
th
ese
is
s
u
es
will
n
ec
ess
itate
th
e
co
o
p
er
atio
n
o
f
d
ata
s
cien
tis
ts
,
en
g
in
ee
r
s
,
an
d
in
d
u
s
tr
y
p
r
o
f
ess
io
n
als
to
en
ab
le
f
air
er
a
n
d
m
o
r
e
tr
an
s
p
a
r
en
t
r
ec
o
m
m
en
d
atio
n
alg
o
r
ith
m
s
[
3
5
]
,
[
3
7
]
.
I
n
th
e
lo
n
g
r
u
n
,
n
ew
u
s
ab
le
f
in
d
i
n
g
s
(
e.
g
.
,
k
e
y
lear
n
in
g
p
o
in
ts
)
will
b
e
ex
p
lo
r
ed
b
y
u
s
in
g
i
n
n
o
v
ativ
e
s
y
s
tem
lear
n
in
g
p
r
o
ce
d
u
r
es
l
ik
e
p
r
o
f
o
u
n
d
lear
n
in
g
an
d
s
u
p
p
o
r
t
lea
r
n
in
g
[
3
8
]
(
e.
g
.
p
io
n
ee
r
f
in
d
i
n
g
o
f
th
e
clien
t,
s
ellin
g
m
o
r
e
th
an
o
n
e
th
in
g
with
o
u
t
a
m
o
m
e
n
t'
s
d
elay
to
h
elp
p
r
ec
is
io
n
an
d
p
er
tin
en
t p
r
o
p
o
s
al)
[
3
9
]
,
[
4
0
]
.
3.
M
E
T
H
O
D
T
h
is
s
ec
tio
n
o
v
er
v
iews
th
e
m
eth
o
d
o
lo
g
y
th
at
h
as
b
ee
n
u
s
e
d
th
r
o
u
g
h
o
u
t
th
e
p
r
o
ject.
First
o
f
all,
we
will
d
ef
in
e
th
e
p
r
o
ce
s
s
es
o
f
d
ata
co
llectio
n
an
d
d
ata
p
r
e
p
r
o
ce
s
s
in
g
an
d
th
en
d
ea
l
with
th
e
co
n
cr
ete
d
atasets
an
d
th
eir
s
p
ec
if
ic
ch
ar
ac
ter
is
tics
th
at
wer
e
u
s
e
d
.
Fin
ally
,
we
will
o
u
tlin
e
th
e
d
ata
clea
n
in
g
p
r
o
ce
s
s
to
s
h
o
w
th
e
way
th
e
v
alid
ity
o
f
th
e
an
al
y
s
is
h
as
b
ee
n
s
u
s
tain
ed
.
T
h
is
lo
g
ical
ch
ain
is
d
esig
n
ed
to
en
s
u
r
e
th
e
h
ig
h
lev
el
o
f
r
eliab
ilit
y
o
f
th
e
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
E
n
h
a
n
ci
n
g
co
mmu
n
ica
tio
n
a
n
d
in
tera
ctio
n
in
th
e
mo
vie
in
d
u
s
tr
y
b
a
s
ed
…
(
S
a
id
C
h
a
ko
u
k
)
4663
3
.
1
.
Da
t
a
c
o
llect
io
n a
nd
prepro
ce
s
s
ing
Fo
r
th
is
p
r
o
ject,
we
p
u
lled
m
o
v
ie
d
ata
f
r
o
m
th
r
ee
C
SV
f
iles
:
m
o
v
ies.csv
,
r
atin
g
s
.
csv
,
a
n
d
tag
s
.
csv
.
[
4
1
]
.
Sp
ec
if
ically
,
th
e
th
r
ee
f
i
les
co
n
s
is
t
o
f
in
f
o
ab
o
u
t
m
o
v
ies,
r
atin
g
s
f
r
o
m
u
s
er
s
o
n
th
e
m
o
v
ies,
an
d
tag
s
g
en
er
ated
b
y
u
s
er
s
o
n
th
e
m
o
v
ies
[
4
2
]
.
Af
ter
co
llectin
g
an
d
p
r
ep
r
o
ce
s
s
in
g
t
h
e
d
ata,
it
is
im
p
o
r
tan
t to
p
r
o
v
id
e
a
d
etailed
d
escr
ip
tio
n
o
f
th
e
c
o
n
ten
ts
o
f
ea
ch
f
ile
u
s
ed
in
th
is
s
tu
d
y
.
3
.
2
.
Da
t
a
des
cr
iptio
n
T
h
is
s
ec
tio
n
p
r
esen
ts
a
d
etaile
d
s
u
m
m
ar
y
o
f
d
ata
r
eg
ar
d
i
n
g
d
if
f
er
en
t
s
o
u
r
ce
s
,
with
th
e
f
o
c
u
s
o
n
th
e
in
ter
p
r
etatio
n
o
f
th
eir
n
atu
r
e
a
n
d
th
e
ty
p
es
o
f
in
f
o
r
m
atio
n
th
a
t
th
ey
co
u
ld
p
o
s
s
ess
.
T
h
e
in
f
o
r
m
atio
n
is
n
ec
ess
ar
y
to
ac
q
u
ir
e
a
b
asic
u
n
d
er
s
tan
d
in
g
o
f
th
e
ch
ar
ac
ter
is
tics
o
f
d
ata
u
s
ed
in
th
e
p
r
o
ject
an
d
p
r
o
ce
ed
with
th
eir
an
aly
s
is
.
i)
Mo
v
ies.csv
:
a
C
SV
f
ile
with
d
ata
ab
o
u
t th
e
m
o
v
ies an
d
th
e
f
o
llo
win
g
co
lu
m
n
s
[
4
3
]
:
‒
m
o
v
ieI
d
:
a
n
id
en
tifie
r
f
o
r
m
o
v
ies.
‒
titl
e:
t
itle o
f
th
e
f
ilm
.
‒
g
en
r
es:
g
en
r
es
o
f
th
e
f
ilm
.
ii)
R
atin
g
s
.
csv
:
w
e
h
av
e
a
C
SV
f
ile
th
at
h
as
u
s
er
r
atin
g
s
f
o
r
v
ar
io
u
s
m
o
v
ies
th
at
in
clu
d
es
th
e
f
o
llo
win
g
co
lu
m
n
s
:
‒
u
s
er
I
d
:
t
o
id
e
n
tify
u
s
er
s
u
n
iq
u
ely
.
‒
m
o
v
ieI
d
:
m
o
v
ie
’
s
u
n
iq
u
e
id
e
n
tifie
r
.
‒
r
atin
g
: 0
.
5
-
5
.
0
b
ased
o
n
u
s
er
v
o
tes f
o
r
th
e
f
ilm
.
‒
tim
estam
p
: d
atetim
e
wh
en
th
e
r
atin
g
was c
r
ea
ted
.
iii)
T
ag
s
.
csv
: T
h
e
C
SV
f
ile
is
ab
o
u
t u
s
er
p
r
o
v
id
ed
tag
s
f
o
r
m
o
v
i
es a
n
d
co
n
tain
s
th
e
c
o
lu
m
n
s
:
‒
u
s
er
I
d
:
e
v
er
y
u
s
er
h
as a
u
n
iq
u
e
I
D.
‒
m
o
v
ieI
d
:
m
o
v
ie
u
n
iq
u
e
i
d
en
tif
ier
.
‒
t
ag
a
u
s
er
,
b
u
t it
ca
n
b
e
ad
d
e
d
o
n
th
e
im
a
g
e
in
th
e
m
o
v
ie
.
‒
t
im
estam
p
o
f
wh
en
t
h
e
tag
is
c
r
ea
ted
.
Fo
llo
win
g
th
e
p
r
esen
tatio
n
o
f
th
e
d
if
f
e
r
en
t
d
ata
s
o
u
r
ce
s
an
d
th
eir
c
o
n
ten
ts
,
a
th
o
r
o
u
g
h
d
ata
clea
n
in
g
p
r
o
ce
s
s
was c
ar
r
ied
o
u
t to
e
n
s
u
r
e
th
e
q
u
ality
an
d
c
o
n
s
is
ten
cy
o
f
th
e
d
ataset.
3
.
3
.
Da
t
a
clea
nin
g
T
h
e
m
o
v
ie
d
ata
p
r
o
v
id
ed
was
clea
n
ed
p
r
io
r
t
o
p
e
r
f
o
r
m
in
g
t
h
e
an
aly
s
is
in
o
r
d
e
r
to
m
ain
tain
d
ata
q
u
ality
an
d
co
n
s
is
ten
cy
[
4
4
]
.
T
h
is
clea
n
in
g
p
r
o
ce
s
s
en
s
u
r
ed
th
e
r
e
liab
ilit
y
o
f
th
e
s
u
b
s
eq
u
en
t
r
esu
lts
.
T
h
e
f
o
llo
win
g
s
tep
s
wer
e
ca
r
r
ied
o
u
t
d
u
r
in
g
t
h
is
p
h
ase:
‒
Du
p
licate
d
etec
tio
n
:
a
ll
C
SV
f
iles
wer
e
ch
ec
k
ed
f
o
r
d
u
p
licate
s
an
d
r
em
o
v
ed
u
n
n
ec
ess
ar
y
d
u
p
licate
en
tr
ies.
‒
Miss
in
g
v
al
u
es:
m
is
s
in
g
o
r
NaN
v
alu
es
in
th
e
d
ataset
was
ch
ec
k
ed
an
d
h
an
d
le
d
v
ia
v
ar
io
u
s
m
eth
o
d
s
(
im
p
u
tatio
n
o
r
r
em
o
v
al
o
f
r
ec
o
r
d
s
with
m
is
s
in
g
v
alu
es).
‒
T
im
estam
p
co
n
v
er
s
io
n
:
t
he
ti
m
estam
p
v
alu
es
o
f
th
e
r
atin
g
s
.
csv
an
d
tag
s
.
csv
→
C
SV
h
u
m
an
r
ea
d
ab
le
in
ter
p
r
etab
le
C
SV
f
iles
wer
e
co
n
v
er
ted
to
ea
s
ily
r
ea
d
a
b
le
in
t
er
p
r
etab
le
C
SV
f
iles
f
o
r
b
etter
an
aly
s
is
.
3
.
4
.
Da
t
a
prepro
ce
s
s
ing
T
h
e
m
o
v
ie
d
ata
was
th
en
p
r
e
p
r
o
ce
s
s
ed
b
y
im
p
o
r
tin
g
it
u
s
in
g
Py
Sp
ar
k
an
d
,
wh
en
n
ec
ess
ar
y
,
Pan
d
as
[
4
5
]
.
T
h
is
p
r
ep
r
o
ce
s
s
in
g
h
elp
e
d
ef
f
icien
tly
p
r
e
p
ar
e
th
e
d
ata
f
o
r
f
u
r
th
er
an
aly
s
is
.
Sev
er
al
k
e
y
task
s
wer
e
ca
r
r
ied
o
u
t a
s
p
ar
t o
f
th
is
p
r
o
ce
s
s
,
s
u
c
h
as:
‒
Featu
r
e
en
g
in
ee
r
in
g
:
t
h
e
n
ew
f
ea
tu
r
es
ar
e
d
er
iv
e
d
f
r
o
m
t
h
e
f
ea
tu
r
es
th
at
alr
ea
d
y
e
x
is
t
o
r
t
h
e
o
ld
f
ea
tu
r
e
,
wh
ich
is
tr
an
s
f
o
r
m
ed
in
t
o
a
n
ew
f
ea
tu
r
e
,
co
n
s
id
er
i
n
g
th
e
r
e
lev
an
t
in
f
o
r
m
atio
n
to
b
e
u
s
ed
f
o
r
m
o
d
elin
g
p
u
r
p
o
s
es.
T
h
is
m
ea
n
s
th
at
s
o
m
e
o
f
th
e
f
ea
tu
r
es
wer
e
ca
lcu
l
ated
to
p
r
o
v
id
e
e
x
tr
a
d
ata
,
e.
g
.
,
m
o
v
ie
r
atin
g
s
co
u
n
t p
e
r
u
s
er
an
d
av
e
r
ag
e
m
o
v
ie
r
atin
g
.
‒
No
r
m
aliza
tio
n
:
w
e
n
o
r
m
alize
d
an
y
n
u
m
er
ic
f
ea
tu
r
es
in
th
e
v
ar
iab
le
4
2
to
en
s
u
r
e
th
at
we
wer
e
o
n
th
e
s
am
e
s
ca
le
ac
r
o
s
s
th
e
v
ar
iab
les.
Su
ch
p
r
ep
r
o
ce
s
s
in
g
is
p
ar
ticu
lar
ly
im
p
o
r
ta
n
t
in
en
h
a
n
c
in
g
alg
o
r
ith
m
p
er
f
o
r
m
an
ce
o
r
ac
cu
r
ac
y
,
s
p
e
cif
ically
o
n
m
o
d
els
th
at
ar
e
s
en
s
itiv
e
to
th
e
s
ca
les
o
f
f
ea
tu
r
es
,
s
u
ch
as
i
n
co
llab
o
r
ativ
e
f
ilter
in
g
alg
o
r
ith
m
s
.
‒
E
n
co
d
in
g
ca
teg
o
r
ical
v
a
r
iab
l
es
:
c
ateg
o
r
ical
f
ea
tu
r
es
lik
e
m
o
v
ie
g
en
r
es
wer
e
co
n
v
e
r
ted
to
n
u
m
e
r
ical
r
ep
r
esen
tatio
n
u
s
in
g
m
eth
o
d
s
lik
e
o
n
e
-
h
o
t
en
co
d
in
g
.
B
y
th
is
co
n
v
er
s
io
n
,
we
ca
n
u
s
e
ca
teg
o
r
ical
d
ata
as
in
p
u
t v
ar
ia
b
les f
o
r
m
ac
h
in
e
le
ar
n
in
g
alg
o
r
ith
m
s
th
at
r
e
q
u
ir
e
n
u
m
er
ical
in
p
u
t.
T
h
e
p
r
ep
r
o
ce
s
s
ed
d
ata
with
r
en
d
er
e
d
f
ea
tu
r
es,
n
o
r
m
alize
d
n
u
m
er
ical
v
a
r
iab
les,
an
d
e
n
co
d
ed
ca
teg
o
r
ical
v
ar
iab
les we
r
e
in
p
u
t t
o
co
n
s
tr
u
ct
r
ec
o
m
m
e
n
d
atio
n
m
o
d
els a
n
d
an
aly
s
es
[
4
6
]
.
3
.
5
.
Alg
o
rit
hm
s
elec
t
io
n f
o
r
m
o
v
ie
re
c
o
mm
enda
t
io
n
3
.
5
.
1
.
O
v
er
v
iew
I
n
th
is
s
ec
tio
n
,
we
ex
p
lain
th
e
alg
o
r
ith
m
s
th
at
we
u
s
e
in
o
u
r
c
o
llab
o
r
ativ
e
f
ilter
in
g
m
o
v
ie
r
ec
o
m
m
en
d
atio
n
s
y
s
tem
an
d
d
is
cu
s
s
th
eir
p
er
f
o
r
m
an
ce
ag
ain
s
t
ea
ch
o
th
e
r
.
A
n
aly
s
is
o
f
c
o
llab
o
r
ativ
e
f
ilter
in
g
,
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.
14
,
No
.
6
,
Dec
em
b
er
20
25
:
4
6
6
1
-
4
6
7
4
4664
co
n
ten
t
-
b
ased
f
ilter
in
g
,
an
d
h
y
b
r
id
m
eth
o
d
s
o
n
t
h
eir
r
o
l
e
in
p
r
o
v
id
i
n
g
u
s
er
s
with
p
er
s
o
n
alize
d
m
o
v
ie
r
ec
o
m
m
en
d
atio
n
s
.
Fu
ll
a
n
aly
s
is
o
f
b
o
th
ap
p
r
o
ac
h
es
an
d
t
h
eir
ad
v
an
tag
es/d
is
ad
v
an
tag
e
s
in
th
e
c
o
n
tex
t
o
f
o
u
r
s
y
s
tem
.
3
.
5
.
2
.
Alg
o
rit
hm
des
cr
iptio
n
C
o
llab
o
r
ativ
e
fi
lter
in
g
alter
n
at
in
g
least
s
q
u
ar
es
(
AL
S):
c
o
lla
b
o
r
ativ
e
f
ilter
in
g
is
a
p
o
p
u
lar
tech
n
iq
u
e
f
o
r
b
u
ild
in
g
r
ec
o
m
m
en
d
atio
n
s
y
s
tem
s
th
at
u
s
e
th
e
wid
er
p
o
p
u
latio
n
to
p
r
e
d
ict
wh
at
an
y
s
in
g
le
u
s
er
will
lik
e.
Ap
ac
h
e
Sp
ar
k
in
clu
d
es
a
n
im
p
lem
en
tatio
n
o
f
th
e
AL
S
alg
o
r
it
h
m
f
o
r
co
llab
o
r
ativ
e
f
ilter
in
g
i
n
its
ML
lib
lib
r
ar
y
,
wh
ich
is
o
n
e
o
f
th
e
m
o
s
t
p
o
p
u
lar
im
p
lem
en
tatio
n
s
o
f
th
is
te
ch
n
iq
u
e.
AL
S
f
ac
t
o
r
izes
th
e
u
s
er
-
item
in
ter
ac
tio
n
m
atr
ix
in
to
two
lo
w
-
r
a
n
k
m
atr
ices,
o
n
e
p
er
tain
in
g
t
o
th
e
late
n
t
f
ea
tu
r
es
o
f
th
e
u
s
er
s
a
n
d
th
e
o
th
er
to
th
e
laten
t
f
ea
tu
r
es
o
f
item
s
.
AL
S
m
in
im
izes
th
e
s
q
u
ar
ed
er
r
o
r
b
etwe
en
th
e
p
r
ed
icted
v
alu
es
an
d
r
ea
l
r
atin
g
s
,
cr
ea
tin
g
r
ec
o
m
m
en
d
atio
n
s
f
o
r
p
er
s
o
n
a
lized
item
s
to
u
s
er
s
b
y
iter
ati
v
ely
ad
ju
s
tin
g
th
ese
m
atr
ices
[
4
7
]
,
[
4
8
]
.
R
ec
en
t
s
tu
d
ies
h
av
e
also
ev
alu
ated
th
e
p
er
f
o
r
m
a
n
ce
o
f
m
ac
h
in
e
lear
n
in
g
f
r
am
ewo
r
k
s
s
u
ch
as
Py
T
o
r
ch
an
d
T
en
s
o
r
Flo
w
in
th
e
co
n
tex
t
o
f
b
ig
d
ata
r
ec
o
m
m
en
d
atio
n
s
y
s
tem
s
[
2
2
]
,
a
n
d
in
teg
r
atin
g
th
ese
m
eth
o
d
s
with
co
llab
o
r
ativ
e
f
ilter
in
g
ca
n
e
n
h
an
ce
r
ec
o
m
m
en
d
atio
n
ac
c
u
r
ac
y
a
n
d
ef
f
icien
cy
[
4
9
]
.
C
o
n
t
e
n
t
b
a
s
ed
f
i
l
t
er
i
n
g
:
co
n
t
en
t
-
b
a
s
e
d
f
i
l
t
e
r
in
g
r
e
co
m
m
e
n
d
a
t
i
o
n
s
ar
e
m
a
d
e
f
o
r
u
s
e
r
s
b
a
s
e
d
o
n
th
e
c
h
a
r
ac
t
e
r
i
s
t
i
c
s
/
f
ea
t
u
r
e
s
o
f
t
h
e
i
t
e
m
s
.
F
o
r
e
x
a
m
p
le
,
c
o
n
t
en
t
-
b
a
s
e
d
f
i
l
t
e
r
in
g
in
m
o
v
i
e
r
e
co
m
m
en
d
a
t
i
o
n
u
s
e
s
m
o
v
ie
m
e
t
ad
a
t
a
(
e
.
g
.
,
g
e
n
r
e,
c
a
s
t
,
d
i
r
ec
t
o
r
,
an
d
p
l
o
t
s
u
m
m
a
r
y
)
to
r
e
t
r
i
ev
e
s
i
m
i
la
r
i
ti
e
s
b
e
t
w
e
en
i
t
e
m
s
.
A
f
t
e
r
w
ar
d
s
,
i
t
r
e
co
m
m
en
d
s
y
o
u
m
o
v
i
e
s
s
i
m
i
la
r
t
o
t
h
e
o
n
es
y
o
u
e
n
jo
y
ed
w
a
tc
h
in
g
e
a
r
l
ie
r
.
C
o
n
t
en
t
-
b
a
s
e
d
f
i
l
t
e
r
in
g
d
o
e
s
n
o
t
d
e
p
en
d
o
n
th
e
in
t
e
r
a
c
t
i
o
n
b
e
t
w
e
en
u
s
e
r
an
d
i
t
e
m
;
h
o
w
ev
e
r
i
t
ta
k
e
s
t
h
e
a
s
p
e
c
t
o
f
i
t
em
s
an
d
u
s
e
r
p
r
ef
e
r
en
c
e
s
to
i
t
s
e
lf
[
5
0
]
.
3
.
5
.
3
.
Alg
o
rit
hm
c
o
m
pa
riso
n
T
h
is
an
aly
s
is
co
n
tain
s
two
tab
les,
p
r
o
v
id
in
g
p
e
r
s
p
ec
tiv
es
in
to
d
if
f
e
r
en
t
r
ec
o
m
m
en
d
atio
n
al
g
o
r
ith
m
s
.
E
v
en
wh
en
th
e
b
est
alg
o
r
ith
m
s
ca
n
b
e
ap
p
lied
to
th
e
d
at
a,
th
e
q
u
ality
o
f
t
h
at
d
ata
h
a
s
a
lar
g
e
im
p
ac
t
o
n
ef
f
ec
tiv
en
ess
.
Fin
ally
,
s
en
tim
en
t
an
aly
s
is
ca
n
b
e
a
u
s
ef
u
l
in
ass
ess
in
g
th
e
q
u
ality
o
f
th
e
d
ata
p
r
o
v
id
ed
b
y
r
ec
o
m
m
en
d
atio
n
s
y
s
tem
s
u
s
in
g
o
n
lin
e
p
an
els
[
5
1
]
.
T
ab
le
1
o
f
f
er
s
a
d
etailed
c
o
m
p
ar
is
o
n
o
f
d
if
f
er
e
n
t
r
ec
o
m
m
en
d
atio
n
alg
o
r
ith
m
s
b
ased
o
n
th
ei
r
p
er
s
o
n
aliza
tio
n
ca
p
ab
ilit
ies
an
d
s
ca
lab
ilit
y
.
C
o
llab
o
r
ativ
e
f
ilter
in
g
AL
S
is
n
o
te
d
f
o
r
its
h
ig
h
p
er
s
o
n
aliza
tio
n
b
u
t
o
n
ly
m
o
d
e
r
ate
s
ca
lab
ilit
y
.
I
n
co
n
tr
ast,
co
n
ten
t
-
b
ased
f
il
ter
in
g
e
x
ce
ls
in
s
ca
lab
ilit
y
b
u
t
p
r
o
v
i
d
es
lo
wer
p
er
s
o
n
aliza
tio
n
.
H
y
b
r
id
a
p
p
r
o
ac
h
es,
wh
ich
b
le
n
d
elem
e
n
ts
f
r
o
m
b
o
th
tech
n
iq
u
es,
ac
h
iev
e
h
ig
h
p
er
s
o
n
aliza
tio
n
an
d
m
o
d
er
ate
to
h
ig
h
s
ca
lab
ili
ty
.
T
ab
le
1
.
C
o
m
p
a
r
is
o
n
o
f
r
ec
o
m
m
en
d
atio
n
al
g
o
r
ith
m
s
(
p
ar
t 1
)
A
l
g
o
r
i
t
h
m
P
e
r
so
n
a
l
i
z
a
t
i
o
n
S
c
a
l
a
b
i
l
i
t
y
C
o
l
l
a
b
o
r
a
t
i
v
e
f
i
l
t
e
r
i
n
g
A
LS
H
i
g
h
M
o
d
e
r
a
t
e
C
o
n
t
e
n
t
-
b
a
se
d
f
i
l
t
e
r
i
n
g
Lo
w
H
i
g
h
H
y
b
r
i
d
a
p
p
r
o
a
c
h
e
s
H
i
g
h
M
o
d
e
r
a
t
e
-
h
i
g
h
T
h
e
d
ata
r
eq
u
ir
em
e
n
ts
o
f
th
e
alg
o
r
ith
m
s
ar
e
lis
ted
in
T
ab
le
2
.
C
o
llab
o
r
ativ
e
f
ilter
in
g
AL
S
is
b
ased
o
n
u
s
er
-
item
in
ter
ac
tio
n
d
ata.
T
h
is
ty
p
e
o
f
co
n
ten
t
-
b
ased
f
ilter
in
g
r
elies
o
n
m
etad
ata
lik
e
d
ata
ab
o
u
t
m
o
v
ies
ch
ar
ac
ter
is
tics
o
r
attr
ib
u
tes
r
elate
d
to
p
r
o
d
u
cts.
Hy
b
r
id
m
e
th
o
d
s
u
s
e
b
o
th
in
te
r
ac
tio
n
d
a
ta
an
d
m
etad
ata
to
ex
p
lo
it th
e
ad
v
an
tag
es o
f
th
e
o
th
er
two
m
eth
o
d
s
to
in
cr
ea
s
e
r
elev
an
ce
an
d
q
u
ality
.
T
ab
le
2
.
C
o
m
p
a
r
is
o
n
o
f
r
ec
o
m
m
en
d
atio
n
al
g
o
r
ith
m
s
(
p
ar
t 2
)
A
l
g
o
r
i
t
h
m
D
a
t
a
r
e
q
u
i
r
e
me
n
t
s
C
o
l
l
a
b
o
r
a
t
i
v
e
f
i
l
t
e
r
i
n
g
A
LS
U
ser
-
i
t
e
m
i
n
t
e
r
a
c
t
i
o
n
C
o
n
t
e
n
t
-
b
a
se
d
f
i
l
t
e
r
i
n
g
M
o
v
i
e
m
e
t
a
d
a
t
a
H
y
b
r
i
d
a
p
p
r
o
a
c
h
e
s
C
o
m
b
i
n
e
d
i
n
t
e
r
a
c
t
i
o
n
d
a
t
a
a
n
d
m
e
t
a
d
a
t
a
3
.
5
.
4
.
Alg
o
rit
hm
s
elec
t
io
n
Ou
r
m
o
v
ie
r
ec
o
m
m
e
n
d
atio
n
s
y
s
tem
u
s
es
a
h
y
b
r
id
a
p
p
r
o
ac
h
,
wh
ich
is
a
c
o
m
b
in
atio
n
o
f
c
o
l
lab
o
r
ativ
e
f
ilter
in
g
an
d
c
o
n
ten
t
-
b
ased
f
il
ter
in
g
.
T
h
is
m
eth
o
d
u
tili
ze
s
t
h
e
ad
v
an
tag
es
o
f
b
o
th
m
eth
o
d
s
.
W
e
s
elec
ted
th
is
ap
p
r
o
ac
h
to
in
cr
ea
s
e
r
ec
o
m
m
e
n
d
atio
n
p
r
ec
is
io
n
an
d
o
v
e
r
co
m
e
d
ef
icien
cies o
f
in
d
iv
id
u
al
alg
o
r
ith
m
s
.
Her
e
ar
e
th
e
r
ea
s
o
n
s
wh
y
we
m
a
d
e
th
is
ch
o
ice:
‒
Flex
ib
ilit
y
/
pe
r
s
o
n
aliza
tio
n
:
h
y
b
r
id
m
eth
o
d
s
ar
e
f
lex
i
b
le
an
d
c
an
co
m
b
in
e
th
e
s
tr
en
g
th
s
o
f
b
o
th
d
ata
-
d
r
iv
e
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
E
n
h
a
n
ci
n
g
co
mmu
n
ica
tio
n
a
n
d
in
tera
ctio
n
in
th
e
mo
vie
in
d
u
s
tr
y
b
a
s
ed
…
(
S
a
id
C
h
a
ko
u
k
)
4665
an
d
k
n
o
wled
g
e
-
d
r
iv
e
n
ap
p
r
o
a
ch
es.
‒
Scalab
ilit
y
:
m
o
d
er
ately
s
ca
lab
le
f
o
r
co
llab
o
r
ativ
e
f
ilter
in
g
A
L
S;
h
ig
h
s
ca
lab
le
f
o
r
co
n
ten
t
-
b
ased
f
ilter
in
g
,
s
o
th
is
h
y
b
r
id
s
y
s
tem
ca
n
d
ea
l
with
lar
g
e
s
ca
le
m
o
v
ies.
‒
Data
r
eq
u
ir
e
m
en
ts
:
h
y
b
r
id
m
eth
o
d
s
u
tili
ze
u
s
er
-
item
in
ter
a
ctio
n
s
an
d
m
o
v
ie
m
etad
ata;
t
h
er
ef
o
r
e
,
th
e
y
ty
p
ically
h
av
e
m
ed
i
u
m
to
h
ig
h
d
ata
r
eq
u
ir
e
m
en
ts
y
et
th
ey
ca
n
p
r
o
v
id
e
m
o
r
e
i
n
ter
p
r
etab
l
e
an
d
d
iv
er
s
e
r
ec
o
m
m
en
d
atio
n
s
.
W
e
d
ev
elo
p
ed
a
h
y
b
r
id
r
ec
o
m
m
en
d
atio
n
s
y
s
tem
th
at
co
m
b
in
es
co
llab
o
r
ativ
e
an
d
co
n
ten
t
-
b
ased
f
ilter
in
g
ap
p
r
o
ac
h
es
to
p
r
o
v
id
e
p
er
s
o
n
alize
d
m
o
v
ie
r
ec
o
m
m
e
n
d
ati
o
n
s
to
u
s
er
s
wh
ile
o
v
e
r
co
m
in
g
th
e
s
ca
lab
ilit
y
lim
itatio
n
s
o
f
s
tan
d
ar
d
c
o
llab
o
r
ativ
e
f
ilter
in
g
tech
n
iq
u
es
[
5
2
]
,
[
5
3
]
.
3
.
6
.
M
o
del e
v
a
lua
t
i
o
n
Fo
r
th
e
p
u
r
p
o
s
e
o
f
ev
alu
atin
g
th
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
r
ec
o
m
m
en
d
atio
n
m
o
d
els,
we
em
p
lo
y
ed
s
ev
er
al
ev
alu
atio
n
m
ea
s
u
r
es.
T
h
ese
m
ea
s
u
r
es
s
h
o
w
h
o
w
p
r
ec
is
e
an
d
h
o
w
ac
cu
r
ate
th
e
m
o
d
els
ar
e
in
r
ec
o
m
m
e
n
d
in
g
m
o
v
ies.
Her
e’
s
h
o
w
we
e
v
alu
ated
th
em
:
i)
Data
s
p
litt
in
g
:
w
e
d
iv
id
e
d
th
e
d
ataset
in
to
two
p
ar
ts
:
tr
ain
in
g
an
d
test
in
g
s
ets.
W
e
u
s
ed
m
o
s
t
o
f
t
h
e
d
at
a
(
ab
o
u
t
8
0
)
.
ii)
Mo
d
el
tr
ain
in
g
:
w
e
tr
ai
n
ed
t
h
e
r
ec
o
m
m
en
d
atio
n
m
o
d
els,
i
n
clu
d
in
g
c
o
llab
o
r
ativ
e
f
ilter
in
g
AL
S
,
an
d
h
y
b
r
id
ap
p
r
o
ac
h
es,
u
s
in
g
th
e
tr
ain
in
g
d
ataset.
W
e
u
s
ed
t
h
e
r
i
g
h
t
alg
o
r
ith
m
s
an
d
s
ettin
g
s
to
tr
ain
th
em
.
Fo
r
ex
am
p
le,
th
e
AL
S
alg
o
r
ith
m
le
ar
n
ed
f
r
o
m
th
e
in
ter
ac
tio
n
s
b
et
wee
n
u
s
er
s
an
d
m
o
v
ies
to
m
a
k
e
p
er
s
o
n
alize
d
r
ec
o
m
m
en
d
atio
n
s
.
iii)
Pre
d
ictio
n
g
en
er
atio
n
:
o
n
ce
t
r
ain
ed
,
th
e
m
o
d
els
wer
e
p
u
t
to
wo
r
k
g
en
er
atin
g
p
r
e
d
ictio
n
s
f
o
r
m
o
v
ie
r
ec
o
m
m
en
d
atio
n
s
o
n
th
e
test
in
g
d
ataset.
Fo
r
in
s
tan
ce
,
th
e
y
s
u
g
g
ested
m
o
v
ies th
at
u
s
er
s
h
a
d
n
’
t r
ated
y
et.
iv
)
E
v
alu
atio
n
m
etr
ics
:
‒
W
e
u
s
ed
s
ev
er
al
ev
alu
atio
n
m
etr
ics to
ju
d
g
e
th
e
q
u
ality
o
f
th
e
r
ec
o
m
m
en
d
atio
n
s
m
ad
e
b
y
t
h
e
m
o
d
els
.
‒
Pre
cisi
o
n
:
t
h
is
m
ea
s
u
r
e
h
o
w
m
an
y
o
f
th
e
r
ec
o
m
m
en
d
ed
item
s
ar
e
ac
tu
ally
r
elev
an
t t
o
th
e
u
s
er
.
‒
R
ec
all:
i
t sh
o
ws h
o
w
m
an
y
o
f
th
e
r
elev
an
t item
s
wer
e
s
u
cc
ess
f
u
lly
r
ec
o
m
m
e
n
d
ed
t
o
th
e
u
s
e
r
.
‒
MA
P
:
t
h
i
s
ca
lcu
lates
th
e
av
er
ag
e
p
r
ec
is
io
n
ac
r
o
s
s
all
u
s
er
s
,
g
iv
in
g
u
s
a
g
o
o
d
o
v
er
all
v
iew
o
f
r
ec
o
m
m
en
d
atio
n
q
u
ality
[
5
4
]
.
‒
W
e
ca
lcu
lated
th
ese
m
etr
ic
s
f
o
r
ea
ch
m
o
d
el
to
s
ee
h
o
w
ac
c
u
r
ate
an
d
r
elev
an
t
th
eir
r
ec
o
m
m
en
d
atio
n
s
wer
e.
v)
C
o
m
p
ar
is
o
n
an
d
an
aly
s
is
:
‒
W
e
co
m
p
ar
ed
an
d
an
aly
ze
d
th
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
r
ec
o
m
m
en
d
atio
n
m
o
d
els
b
ased
o
n
th
e
s
e
ev
alu
atio
n
m
etr
ics.
T
h
is
h
elp
e
d
u
s
u
n
d
er
s
tan
d
th
e
s
tr
en
g
th
s
a
n
d
wea
k
n
e
s
s
es
o
f
ea
ch
m
o
d
el,
s
u
ch
as
h
o
w
well
th
e
y
co
u
ld
p
e
r
s
o
n
alize
r
ec
o
m
m
en
d
atio
n
s
,
h
an
d
le
lar
g
e
am
o
u
n
ts
o
f
d
ata,
an
d
m
o
r
e.
‒
B
y
r
ig
o
r
o
u
s
ly
e
v
alu
atin
g
th
e
m
o
d
els
an
d
an
al
y
zin
g
th
e
r
esu
lts
,
we
g
o
t
im
p
o
r
tan
t
i
n
s
ig
h
ts
in
to
h
o
w
ef
f
ec
tiv
e
th
ey
wer
e
an
d
w
h
er
e
o
u
r
m
o
v
ie
r
ec
o
m
m
en
d
atio
n
s
y
s
tem
co
u
ld
b
e
im
p
r
o
v
ed
.
4.
R
E
SU
L
T
S
AND
DI
SCUS
SI
O
N
T
h
e
ev
alu
atio
n
o
f
th
e
r
ec
o
m
m
en
d
atio
n
s
y
s
tem
p
r
o
d
u
ce
d
v
alu
a
b
le
r
esu
lts
.
T
h
ese
f
i
n
d
in
g
s
ar
e
illu
s
tr
ated
in
th
e
f
o
llo
win
g
v
is
u
aliza
tio
n
s
.
E
ac
h
v
is
u
aliza
tio
n
h
elp
s
h
ig
h
lig
h
t
d
if
f
er
e
n
t
asp
ec
ts
o
f
th
e
s
y
s
tem
’
s
ef
f
ec
tiv
en
ess
an
d
p
e
r
f
o
r
m
an
ce
.
4
.
1
.
Rec
o
mm
enda
t
io
n r
esu
lt
s
T
a
b
l
e
3
s
h
o
w
s
t
h
e
r
e
co
m
m
en
d
a
t
io
n
s
cr
e
a
t
ed
f
o
r
t
h
e
u
s
e
r
w
i
t
h
u
s
er
I
d
=1
,
a
lo
n
g
w
i
t
h
t
h
e
m
o
v
i
e
I
D
s
a
n
d
t
h
e
p
r
e
d
i
c
te
d
r
a
t
i
n
g
s
.
T
h
i
s
t
ab
l
e
g
i
v
e
s
t
h
e
u
s
e
r
a
n
i
d
e
a
ab
o
u
t
th
e
l
i
s
t
o
f
m
o
v
i
e
I
D
s
a
n
d
p
r
e
d
ic
t
e
d
r
a
t
i
n
g
f
o
r
r
e
co
m
m
e
n
d
e
d
m
o
v
ie
s
.
H
e
n
ce
,
i
t
s
er
v
e
s
a
s
a
g
r
e
a
t
i
n
s
i
g
h
t
i
n
to
th
e
r
e
c
o
m
m
en
d
a
ti
o
n
s
o
f
th
e
s
y
s
t
em
f
o
r
t
h
e
u
s
e
r
.
4
.
2
.
Dis
t
ributio
n o
f
ra
t
ing
s
Un
d
er
s
tan
d
in
g
th
e
d
is
tr
ib
u
tio
n
o
f
m
o
v
ies
b
ased
o
n
th
ei
r
r
atin
g
s
o
f
f
e
r
s
in
f
o
r
m
atio
n
r
e
g
ar
d
in
g
u
s
er
p
r
ef
er
en
ce
s
an
d
th
e
o
v
e
r
all
q
u
ality
o
f
th
e
d
ataset.
As
s
h
o
wn
in
Fig
u
r
e
1
,
th
e
m
o
v
ies
ar
e
a
r
r
an
g
ed
ac
co
r
d
in
g
to
th
eir
r
atin
g
s
,
wh
ich
clea
r
ly
h
i
g
h
lig
h
ts
th
e
lev
els
o
f
p
o
p
u
la
r
ity
an
d
u
s
er
s
atis
f
ac
tio
n
.
T
h
is
i
n
f
o
r
m
atio
n
h
elp
s
to
id
en
tify
tr
en
d
s
an
d
p
atter
n
s
with
in
th
e
d
ataset.
T
h
e
d
is
tr
ib
u
tio
n
o
f
r
atin
g
s
allo
ws
u
s
to
id
e
n
tify
tr
e
n
d
s
in
u
s
er
p
r
ef
er
e
n
ce
s
.
Fo
r
ex
am
p
le,
ex
tr
ac
tin
g
tr
en
d
s
f
r
o
m
u
s
er
r
atin
g
s
,
i
f
all
o
r
m
o
s
t
o
f
t
h
e
r
atin
g
s
ar
e
c
o
n
c
en
tr
ated
o
n
th
e
h
ig
h
er
s
id
e
o
f
th
e
s
ca
le
th
at
m
ea
n
s
th
e
u
s
er
s
g
e
n
er
ally
lik
e
th
e
m
o
v
ies
th
ey
watc
h
,
th
e
n
th
e
co
lle
ctio
n
o
f
th
e
m
o
v
ies
was
well
r
ec
eiv
ed
.
I
n
c
o
n
tr
ast,
a
d
is
tr
ib
u
tio
n
th
at
is
m
o
r
e
e
v
en
ly
s
p
r
ea
d
o
u
t
with
s
ev
er
al
r
atin
g
s
lo
w
d
o
wn
m
ay
im
p
ly
wid
er
v
ar
iatio
n
in
q
u
ality
o
r
lev
els o
f
s
atis
f
ac
tio
n
am
o
n
g
s
t u
s
er
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.
14
,
No
.
6
,
Dec
em
b
er
20
25
:
4
6
6
1
-
4
6
7
4
4666
T
ab
le
3
.
R
ec
o
m
m
e
n
d
atio
n
r
esu
lts
mo
v
i
e
I
d
U
ser
ID
p
r
e
d
i
c
t
i
o
n
6
1
4
.
4
5
0
5
1
2
4
1
0
1
1
4
.
2
9
3
9
2
3
4
1
5
1
1
4
.
3
9
9
8
6
6
6
2
3
1
1
3
.
8
5
5
8
2
3
3
4
9
1
3
.
8
6
0
4
7
5
3
4
2
3
1
2
.
8
9
3
9
6
9
5
4
3
1
4
.
2
2
1
2
8
4
4
5
9
6
1
4
.
5
1
7
4
1
8
9
2
3
1
4
.
1
9
4
7
1
8
4
9
4
0
1
4
.
4
2
3
7
8
1
9
4
3
1
3
.
6
4
1
2
9
0
2
1
0
0
9
1
3
.
2
4
1
9
9
1
1
0
2
4
1
4
.
2
6
4
5
8
7
1
0
3
1
1
4
.
6
7
4
8
0
3
7
1
0
8
9
1
4
.
8
8
0
2
9
1
1
9
7
1
5
.
0
1
2
2
0
1
4
.
5
9
0
9
7
0
5
1
5
7
3
1
3
.
7
7
2
4
6
7
4
1
8
0
5
1
3
.
7
7
0
3
6
7
4
1
9
6
7
1
4
.
3
8
8
1
3
5
Fig
u
r
e
1
.
Dis
tr
ib
u
tio
n
o
f
r
at
in
g
s
4
.
3
.
Av
er
a
g
e
predict
io
n by
g
enre
plo
t
T
h
is
im
p
o
r
ta
n
t
p
l
o
t
illu
s
tr
ates
h
o
w
th
e
av
e
r
ag
e
p
r
ed
ictio
n
s
c
o
r
es
v
ar
y
ac
r
o
s
s
d
if
f
e
r
en
t
m
o
v
ie
g
e
n
r
es
an
d
h
elp
s
ass
ess
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
r
ec
o
m
m
e
n
d
atio
n
s
y
s
tem
.
Fig
u
r
e
2
,
a
p
ie
ch
ar
t
o
f
th
e
av
er
ag
e
p
r
ed
ictio
n
s
co
r
es,
p
r
o
v
id
es
a
v
is
u
al
o
v
e
r
v
iew
o
f
th
e
r
e
c
o
m
m
en
d
atio
n
s
g
en
er
ated
b
y
th
e
s
y
s
tem
.
I
t
allo
ws
f
o
r
b
etter
in
ter
p
r
etatio
n
o
f
wh
ich
g
en
r
es
ar
e
p
r
io
r
itized
o
r
f
a
v
o
r
e
d
b
ase
d
o
n
th
e
s
y
s
tem
’
s
p
r
ed
ictio
n
s
.
I
n
th
is
p
lo
t c
h
ar
t w
e
s
ee
th
e
f
r
ac
tio
n
o
f
r
ec
o
m
m
en
d
atio
n
s
f
o
r
ea
ch
m
o
v
ie
g
en
r
e
.
I
f
a
g
en
r
e
h
as a
lar
g
e
s
lice,
it
m
ea
n
s
th
at
th
e
r
ec
o
m
m
en
d
atio
n
s
y
s
tem
h
a
d
r
ec
o
m
m
e
n
d
ed
m
o
r
e
m
o
v
ies
to
it.
T
h
is
d
is
tr
ib
u
tio
n
ca
n
g
iv
e
s
o
m
e
in
s
ig
h
ts
in
to
wh
at
u
s
er
s
ar
e
ac
tu
ally
lis
ten
in
g
to
an
d
th
en
ca
n
h
elp
m
a
k
e
th
e
r
ec
o
m
m
en
d
atio
n
s
y
s
tem
m
o
r
e
u
s
er
p
r
ef
er
e
n
ce
ce
n
ter
e
d
.
4
.
4
.
Wo
rd
clo
ud
o
f
m
o
v
ie
t
it
les
T
h
is
v
is
u
aliza
tio
n
cr
ea
tes
a
w
o
r
d
clo
u
d
o
f
th
e
titl
es
o
f
m
o
v
ies,
wh
er
e
th
e
s
ize
o
f
th
e
wo
r
d
in
d
icate
s
h
o
w
o
f
ten
th
ey
a
p
p
ea
r
.
As
we
ca
n
s
ee
in
Fig
u
r
e
3
,
it
p
r
esen
ts
th
e
wo
r
d
clo
u
d
a
n
d
m
o
s
t
f
r
eq
u
e
n
tly
ap
p
ea
r
s
illu
s
tr
atin
g
titl
es in
o
u
r
d
ataset.
I
t g
iv
es y
o
u
an
i
d
ea
,
wh
at
m
o
v
ie
titl
es a
r
e
m
en
tio
n
ed
th
e
m
o
s
t.
T
h
is
is
a
w
o
r
d
clo
u
d
to
s
h
o
w
h
o
w
f
r
eq
u
e
n
tly
a
m
o
v
ie
-
titl
e
a
p
p
ea
r
s
in
o
u
r
d
ataset.
T
h
e
b
ig
g
er
wo
r
d
s
m
ea
n
m
o
r
e
f
r
eq
u
e
n
t
u
s
e
o
f
t
h
o
s
e
titl
es,
h
in
tin
g
at
th
e
p
o
p
u
lar
m
o
v
ies
o
r
m
o
v
ies
th
at
ar
e
m
o
s
t
co
m
m
o
n
l
y
watc
h
ed
.
T
h
is
v
is
u
aliza
tio
n
is
a
f
ir
s
t
s
tep
to
s
ee
in
g
m
o
r
e
p
o
p
u
lar
m
o
v
ies
an
d
th
e
u
s
er
-
p
r
e
f
e
r
en
ce
s
b
ased
o
n
th
e
titl
e
f
r
eq
u
en
cy
[
5
5
]
,
[
5
6
]
.
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
E
n
h
a
n
ci
n
g
co
mmu
n
ica
tio
n
a
n
d
in
tera
ctio
n
in
th
e
mo
vie
in
d
u
s
tr
y
b
a
s
ed
…
(
S
a
id
C
h
a
ko
u
k
)
4667
Fig
u
r
e
2
.
Av
e
r
ag
e
p
r
ed
ictio
n
b
y
g
en
r
e
p
lo
t
Fig
u
r
e
3
.
W
o
r
d
c
lo
u
d
o
f
m
o
v
i
e
titl
es
4
.
5
.
M
o
v
ie
c
o
nn
ec
t
io
ns
net
wo
rk
g
r
a
ph
A
n
etwo
r
k
g
r
ap
h
in
d
icatin
g
c
o
n
n
ec
tio
n
s
b
etwe
en
m
o
v
ies
b
ased
o
n
eith
er
u
s
er
in
ter
ac
tio
n
s
o
r
o
th
er
cr
iter
ia
.
S
tated
d
if
f
er
en
tly
,
th
ese
ar
e
co
-
o
cc
u
r
r
en
ce
s
o
r
s
im
ilar
ities
,
an
d
th
ese
r
elatio
n
s
h
ip
s
ar
e
s
h
o
wn
in
Fig
u
r
e
4
.
T
h
is
h
elp
s
u
s
to
in
f
o
r
m
an
d
g
u
id
e
in
b
u
ild
in
g
ef
f
ec
t
iv
e
r
ec
o
m
m
e
n
d
atio
n
s
tr
ateg
y
.
T
h
e
n
etwo
r
k
g
r
ap
h
r
ep
r
esen
ts
r
elatio
n
s
h
ip
s
b
etwe
en
m
o
v
ies,
with
n
o
d
es
r
ep
r
esen
tin
g
m
o
v
ies
an
d
ed
g
es
in
d
i
ca
tin
g
co
n
n
ec
tio
n
s
b
etwe
en
th
em
.
M
o
v
ies
th
at
ar
e
clo
s
ely
co
n
n
ec
ted
in
t
h
e
g
r
a
p
h
ar
e
lik
ely
to
h
av
e
s
im
ilar
c
h
ar
ac
ter
is
tics
o
r
b
e
f
r
eq
u
e
n
tly
watc
h
ed
to
g
eth
e
r
b
y
u
s
er
s
.
B
y
r
ec
o
m
m
en
d
i
n
g
s
im
ilar
m
o
v
ies
to
u
s
er
s
b
ased
o
n
th
eir
v
iewin
g
h
is
to
r
y
,
we
ca
n
lev
er
ag
e
t
h
is
in
f
o
r
m
ati
o
n
to
e
n
h
an
ce
r
ec
o
m
m
e
n
d
atio
n
alg
o
r
ith
m
s
.
4
.
6
.
B
o
x
plo
t
o
f
predict
io
n sco
re
s
by
g
enre
T
h
e
b
o
x
p
lo
t
p
r
o
v
id
es
u
s
a
s
u
m
m
ar
y
o
f
p
r
ed
ictio
n
s
co
r
e
d
is
tr
ib
u
tio
n
b
y
g
e
n
r
e.
I
n
co
n
tr
ast,
Fig
u
r
e
5
s
h
o
ws th
e
d
is
tr
ib
u
tio
n
an
d
ce
n
tr
al
ten
d
en
cy
o
f
th
e
p
r
e
d
ictio
n
s
co
r
es b
y
to
p
ic.
T
h
ese
two
v
i
s
u
aliza
tio
n
s
h
elp
to
u
n
d
er
s
tan
d
t
h
e
p
r
e
d
ictio
n
d
a
ta
v
ar
iatio
n
an
d
k
ey
te
n
d
en
cy
.
T
h
e
b
o
x
p
lo
t
d
is
p
lay
s
t
h
e
p
r
ed
ictio
n
s
co
r
e
d
is
tr
ib
u
tio
n
o
f
m
o
v
ie
ca
teg
o
r
i
es
th
is
s
h
ed
s
lig
h
t
o
n
th
e
s
p
r
e
ad
o
f
p
r
ed
ictio
n
s
co
r
es
f
o
r
ea
c
h
g
en
r
e
,
an
d
it
m
ay
ass
is
t
in
h
ig
h
lig
h
tin
g
th
e
o
u
tli
er
s
o
r
g
e
n
r
es
th
at
h
a
v
e
m
o
r
e
o
r
f
ewe
r
p
r
ed
ictio
n
s
co
r
e
s
.
I
t
h
elp
s
to
m
ak
e
th
e
r
ec
o
m
m
en
d
atio
n
s
y
s
tem
b
etter
an
d
m
o
r
e
p
r
ec
is
e.
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.
14
,
No
.
6
,
Dec
em
b
er
20
25
:
4
6
6
1
-
4
6
7
4
4668
Fig
u
r
e
4
.
Mo
v
ie
co
n
n
ec
tio
n
s
n
etwo
r
k
g
r
a
p
h
Fig
u
r
e
5
.
B
o
x
p
lo
t o
f
p
r
e
d
ictio
n
s
co
r
es b
y
g
en
r
e
4
.
7
.
Co
rr
el
a
t
io
n c
o
e
f
f
icient
s
bet
wee
n t
a
g
s
a
nd
predict
io
n
s
co
re
s
T
h
e
r
elatio
n
s
h
ip
b
etwe
en
u
s
er
-
g
en
er
ated
tag
s
an
d
r
esu
ltin
g
p
r
ed
ictio
n
s
co
r
es
ca
n
s
h
ed
lig
h
t
o
n
h
o
w
ef
f
ec
tiv
e
tag
-
b
ased
r
ec
o
m
m
en
d
atio
n
s
ar
e.
T
h
e
co
r
r
elatio
n
c
o
ef
f
icien
ts
b
etwe
en
tag
s
an
d
p
r
ed
ictio
n
s
co
r
es
ar
e
s
h
o
wn
,
with
th
e
tag
s
s
tr
o
n
g
ly
ass
o
ciate
d
with
h
ig
h
o
r
lo
w
p
r
ed
ictio
n
s
co
r
es
,
in
Fig
u
r
e
6
.
T
h
e
p
lo
t
s
h
o
ws
th
e
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
E
n
h
a
n
ci
n
g
co
mmu
n
ica
tio
n
a
n
d
in
tera
ctio
n
in
th
e
mo
vie
in
d
u
s
tr
y
b
a
s
ed
…
(
S
a
id
C
h
a
ko
u
k
)
4669
co
r
r
elatio
n
co
ef
f
icien
ts
b
etwe
en
p
r
e
d
ictio
n
s
co
r
es
an
d
tag
s
g
en
er
ated
b
y
u
s
er
s
.
T
a
g
s
h
av
i
n
g
r
elativ
el
y
h
i
g
h
e
r
p
o
s
itiv
e
co
r
r
elatio
n
co
ef
f
icien
t
s
ar
e
m
o
r
e
s
tr
o
n
g
ly
ass
o
ciate
d
with
h
ig
h
p
r
e
d
ictio
n
s
co
r
es
(
lik
ely
r
elev
an
t)
an
d
r
ig
h
tly
r
ec
o
m
m
en
d
co
r
r
ec
t
tag
s
.
On
th
e
f
lip
s
id
e,
tag
s
w
ith
n
eg
ativ
e
co
r
r
elatio
n
co
e
f
f
icien
ts
m
ay
b
e
o
r
th
o
g
o
n
al
o
r
p
er
h
ap
s
less
p
r
ed
ictiv
e.
Fig
u
r
e
6
.
C
o
r
r
elatio
n
co
ef
f
icien
ts
b
etwe
en
tag
s
an
d
p
r
ed
ictio
n
s
co
r
es
4
.
8
.
Dis
t
ributio
n o
f
m
o
v
ie
r
a
t
ing
s
by
ho
ur
o
f
t
he
da
y
T
h
e
d
is
tr
ib
u
tio
n
o
f
m
o
v
ie
r
atin
g
s
ac
co
r
d
in
g
to
tim
e
o
f
d
ay
u
n
v
eils
s
o
m
e
f
ascin
atin
g
in
f
o
r
m
atio
n
ab
o
u
t
h
o
w
u
s
er
s
co
n
s
u
m
e
th
eir
co
n
t
en
t
an
d
th
eir
p
r
ef
e
r
en
ce
s
.
T
h
i
s
d
is
tr
ib
u
tio
n
s
h
o
ws
wh
er
e
v
iews
p
ea
k
an
d
wh
en
en
g
ag
em
e
n
t
d
r
o
p
s
as
ca
n
b
e
s
ee
n
in
Fig
u
r
e
7
.
W
ith
th
ese
in
s
ig
h
ts
,
co
n
ten
t
r
elea
s
e
tim
in
g
ca
n
b
e
m
a
x
im
ized
,
an
d
u
s
er
ex
p
er
ien
ce
ca
n
b
e
o
p
tim
ized
.
Fig
u
r
e
7
.
Dis
tr
ib
u
tio
n
o
f
m
o
v
i
e
r
atin
g
s
b
y
h
o
u
r
o
f
t
h
e
d
a
y
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.
14
,
No
.
6
,
Dec
em
b
er
20
25
:
4
6
6
1
-
4
6
7
4
4670
T
h
i
s
v
i
s
u
a
l
i
z
a
t
io
n
s
h
o
w
s
h
o
w
m
o
v
i
e
r
a
t
i
n
g
s
a
r
e
d
i
s
tr
i
b
u
t
e
d
b
y
th
e
t
im
e
o
f
d
ay
i
n
wh
i
c
h
u
s
e
r
s
w
a
t
c
h
ed
m
o
v
i
e
s
.
T
h
i
s
d
i
s
t
r
ib
u
t
i
o
n
a
l
lo
w
s
u
s
to
id
e
n
t
if
y
wh
a
t
th
e
p
ea
k
h
o
u
r
s
ar
e
f
o
r
w
at
c
h
i
n
g
te
l
e
v
i
s
i
o
n
,
a
s
w
e
l
l
a
s
h
o
w
u
s
er
s
'
v
i
e
w
er
s
h
ip
b
e
h
av
i
o
r
i
s
d
u
r
in
g
d
if
f
er
e
n
t
p
ar
t
s
o
f
t
h
e
d
a
y
.
T
h
i
s
in
f
o
r
m
a
t
i
o
n
i
s
e
x
tr
e
m
e
ly
v
a
l
u
ab
l
e
w
h
en
d
e
t
er
m
in
i
n
g
w
h
e
n
t
o
r
e
l
e
a
s
e
c
o
n
t
en
t
o
r
h
o
w
to
f
r
am
e
m
ar
k
e
t
in
g
ef
f
o
r
ts
s
o
th
a
t
p
e
o
p
l
e
a
r
e
m
o
r
e
l
i
k
e
l
y
t
o
e
n
g
a
g
e
w
i
th
o
r
b
e
co
n
v
e
r
t
ed
b
y
y
o
u
r
s
er
v
ic
e
.
I
n
a
d
d
i
t
i
o
n
,
u
n
l
i
k
e
o
th
e
r
m
o
d
e
l
s
t
h
a
t
s
o
l
e
ly
d
e
p
en
d
o
n
ex
i
s
t
i
n
g
u
s
e
r
d
a
t
a
f
o
r
g
en
er
a
t
i
n
g
t
h
e
m
o
d
e
l
,
t
h
i
s
r
e
c
o
m
m
e
n
d
a
t
i
o
n
s
y
s
t
e
m
i
s
n
o
t
d
e
p
en
d
en
t
o
n
a
n
y
o
f
th
e
s
e
d
a
t
a.
T
h
e
s
y
s
t
e
m
c
an
a
l
s
o
s
u
g
g
e
s
t
p
e
r
s
o
n
a
l
i
ze
d
m
o
v
i
e
s
f
o
r
e
v
e
n
n
o
n
r
e
g
i
s
t
er
e
d
u
s
e
r
s
i
f
t
h
ey
ar
e
n
o
t
a
l
r
e
a
d
y
p
r
e
s
en
t
in
t
h
e
d
a
ta
b
a
s
e
,
t
h
e
s
u
g
g
e
s
t
i
o
n
ca
n
b
e
b
a
s
ed
o
n
th
e
t
i
m
e
o
f
th
e
d
a
y
,
g
en
r
e
p
r
e
f
er
e
n
ce
o
r
th
e
p
o
p
u
l
ar
i
t
y
o
f
th
e
m
o
v
i
e.
S
u
ch
a
b
i
l
i
ty
a
d
d
s
t
o
th
e
en
j
o
y
m
en
t
o
f
a
m
o
v
i
e
an
d
m
a
y
in
c
r
e
as
e
u
s
e
r
s
a
t
i
s
f
a
c
t
io
n
a
n
d
en
g
ag
e
m
e
n
t
[
5
3
]
.
R
e
co
m
m
e
n
d
a
t
i
o
n
s
y
s
t
e
m
s
a
r
e
p
o
we
r
f
u
l
t
o
o
l
s
f
o
r
h
ar
n
e
s
s
i
n
g
t
h
e
v
as
t
a
m
o
u
n
t
s
o
f
d
a
t
a
p
r
o
d
u
c
e
d
in
th
e
b
i
g
d
a
t
a
e
r
a
an
d
cr
e
a
t
in
g
p
e
r
s
o
n
a
l
iz
e
d
u
s
e
r
e
x
p
er
i
e
n
c
e
s
.
T
h
i
s
c
h
a
p
t
er
g
iv
e
s
u
s
a
n
id
e
a
a
b
o
u
t
h
o
w
r
e
co
m
m
e
n
d
a
t
i
o
n
a
l
g
o
r
i
th
m
s
f
u
n
c
t
i
o
n
u
n
d
er
th
e
h
o
o
d
,
an
o
v
er
v
ie
w
o
f
t
h
e
ir
im
p
o
r
t
a
n
c
e
,
an
d
a
b
a
s
i
c
u
n
d
e
r
s
t
a
n
d
i
n
g
o
f
t
h
e
ir
f
u
t
u
r
e
w
i
t
h
i
n
b
i
g
d
a
ta
.
4
.
9
.
Understa
nd
ing
us
er
pre
f
er
ence
s
:
t
he
ba
ck
bo
ne
o
f
re
co
m
mend
a
t
io
n sy
s
t
em
s
At
its
co
r
e,
a
r
ec
o
m
m
en
d
atio
n
s
y
s
tem
is
r
ea
d
in
g
b
etwe
en
th
e
lin
es a
n
d
m
ak
i
n
g
in
tellig
en
t
p
r
ed
ictio
n
s
b
ased
o
n
a
g
r
a
n
u
lar
p
atter
n
h
i
d
d
en
u
n
d
er
b
i
g
o
ce
an
s
o
f
d
ata.
T
h
ese
s
y
s
tem
s
id
en
tify
p
atter
n
s
ac
r
o
s
s
m
ass
iv
e
d
ata
s
ets
co
m
p
r
is
in
g
u
s
er
s
'
in
ter
ac
tio
n
s
,
v
iewin
g
h
is
to
r
y
,
a
n
d
c
o
m
m
en
ts
to
d
etec
t
m
o
r
e
n
u
an
ce
d
p
atter
n
s
o
f
b
eh
av
io
r
an
d
p
r
ef
er
en
ce
with
th
e
ai
d
o
f
b
ig
d
ata
an
aly
tics
.
R
ec
o
m
m
en
d
atio
n
s
y
s
tem
s
p
e
r
f
o
r
m
a
lar
g
e
-
s
ca
le
an
aly
s
is
o
f
u
s
er
b
e
h
a
v
io
r
s
u
s
i
n
g
th
e
ad
v
an
ce
d
m
ac
h
in
e
lea
r
n
in
g
wiza
r
d
r
y
o
f
co
llab
o
r
ativ
e
f
ilter
in
g
an
d
d
ee
p
lear
n
in
g
an
d
p
r
ed
ict
with
h
ig
h
ac
cu
r
ac
y
th
e
n
ex
t
p
iece
o
f
c
o
n
ten
t th
e
u
s
er
is
lik
ely
to
c
o
n
s
u
m
e.
4
.
1
0
.
Driv
ing
b
us
ines
s
o
utc
o
m
es:
t
he
p
o
wer
o
f
p
er
s
o
na
liz
a
t
io
n
T
h
e
r
ec
o
m
m
e
n
d
atio
n
s
y
s
tem
s
u
s
in
g
b
ig
d
ata
an
aly
tics
f
o
r
Ne
tf
lix
,
Dis
n
ey
,
an
d
Am
az
o
n
Pri
m
e
Vid
eo
g
ian
ts
ar
e
n
o
t ju
s
t a
tech
n
o
lo
g
ical
m
ar
v
el
b
u
t a
b
u
s
in
ess
g
r
o
w
-
or
-
d
ie
s
tr
ateg
ic
ass
et.
W
ith
ac
ce
s
s
to
m
o
u
n
tain
s
o
f
d
ata,
cr
ea
ted
b
y
m
illi
o
n
s
o
f
u
s
er
s
ar
o
u
n
d
th
e
wo
r
ld
,
th
ese
c
o
m
p
an
ies
ca
n
p
r
o
v
id
e
h
y
p
e
r
-
p
er
s
o
n
alize
d
co
n
ten
t
r
ec
o
m
m
en
d
atio
n
s
b
ased
o
n
p
er
s
o
n
al
in
ter
ests
an
d
v
iewin
g
h
is
to
r
ies.
Su
ch
a
h
ig
h
d
eg
r
e
e
o
f
p
er
s
o
n
aliza
tio
n
in
cr
ea
s
es
u
s
er
ac
tiv
ity
,
r
eten
ti
o
n
an
d
,
in
tu
r
n
,
th
e
b
o
tto
m
lin
e,
allo
win
g
th
em
to
co
n
tin
u
e
ass
er
tin
g
m
ar
k
et
lead
er
s
h
ip
in
th
e
h
ig
h
ly
co
m
p
etitiv
e
ar
en
a
o
f
d
ig
ital e
n
ter
tai
n
m
en
t
[
5
1
]
,
[
5
2
]
.
4
.
1
1
.
B
us
ine
s
s
im
pa
ct
o
f
re
c
o
m
m
enda
t
i
o
n sy
s
t
em
s
T
h
e
in
f
lu
e
n
ce
o
f
b
ig
d
ata
an
aly
tics
-
p
o
wer
ed
r
ec
o
m
m
en
d
at
io
n
s
y
s
tem
s
d
ep
lo
y
e
d
wid
ely
b
y
d
ig
ital
en
ter
tain
m
en
t
c
o
m
p
a
n
ies
o
n
f
i
r
m
s
in
a
n
d
ar
o
u
n
d
th
e
d
ig
ital
e
n
ter
tain
m
en
t
s
p
ac
e
is
v
e
r
y
p
r
o
f
o
u
n
d
.
Per
s
o
n
alize
d
r
ec
o
m
m
en
d
atio
n
s
g
o
p
ast
co
n
v
en
ien
ce
(
Mo
b
ile
p
ay
m
e
n
t
API
s
lead
to
g
r
ea
te
r
d
elig
h
t)
;
o
n
a
m
ac
r
o
lev
el,
p
er
s
o
n
alize
d
r
ec
o
m
m
e
n
d
atio
n
s
h
av
e
p
r
o
f
o
u
n
d
p
o
wer
o
v
er
c
o
n
s
u
m
er
b
eh
a
v
io
r
,
d
r
iv
in
g
h
o
w
d
ig
ital
co
m
m
u
n
icatio
n
tak
es
p
lace
ac
r
o
s
s
m
illi
o
n
s
o
f
u
s
er
s
.
A
cu
s
to
m
co
n
ten
t
ca
tal
o
g
d
r
iv
en
b
y
b
i
g
d
ata
an
al
y
tics
will
b
r
in
g
c
o
m
p
an
ies
clo
s
er
to
co
n
s
u
m
er
s
,
in
cr
ea
s
in
g
s
u
b
s
cr
ip
tio
n
r
en
ewa
ls
,
cu
s
to
m
e
r
life
tim
e
v
alu
e,
an
d
b
r
an
d
lo
y
alty
.
Ad
d
itio
n
ally
,
r
ec
o
m
m
en
d
atio
n
s
y
s
tem
s
ar
e
p
o
wer
f
u
l
co
n
ten
t
d
is
co
v
er
y
to
o
ls
th
at
u
p
lift
n
ich
e
titl
es
an
d
p
r
o
m
o
te
d
iv
er
s
e
co
n
ten
t c
o
n
s
u
m
p
tio
n
ac
r
o
s
s
d
if
f
er
en
t
g
lo
b
al
au
d
ien
ce
s
[
5
2
]
.
4
.
1
2
.
Cha
rt
ing
t
he
pa
t
h f
o
r
wa
rd:
f
uture
direct
io
ns
in re
co
m
mend
a
t
io
n sy
s
t
em
s
As
we
ch
ar
t
th
e
co
u
r
s
e
f
o
r
wa
r
d
in
th
e
r
ea
lm
o
f
b
ig
d
at
a
-
d
r
i
v
en
r
ec
o
m
m
en
d
at
io
n
s
y
s
t
em
s
,
a
m
y
r
iad
o
f
p
o
s
s
ib
il
i
ti
es
b
ec
k
o
n
s
u
s
to
ex
p
lo
r
e
n
ew
f
r
o
n
t
ier
s
an
d
p
u
s
h
th
e
b
o
u
n
d
ar
ie
s
o
f
in
n
o
v
at
io
n
.
W
h
il
e
th
e
cu
r
r
en
t
s
t
at
e
o
f
r
ec
o
m
m
en
d
at
io
n
alg
o
r
ith
m
s
s
ta
n
d
s
a
s
a
t
es
ta
m
en
t
t
o
th
e
ir
ef
f
i
ca
cy
,
th
er
e
is
s
am
p
le
r
o
o
m
f
o
r
f
u
r
th
e
r
en
h
an
c
em
en
t
an
d
ev
o
l
u
t
io
n
.
L
o
o
k
in
g
ah
e
ad
,
s
e
v
er
a
l k
ey
ar
e
as
w
ar
r
an
t
at
ten
tio
n
an
d
in
v
e
s
tm
en
t:
‒
E
n
h
an
c
ed
p
er
s
o
n
al
iz
at
io
n
a
n
d
co
n
tex
tu
al
u
n
d
e
r
s
ta
n
d
in
g
:
c
o
n
t
in
u
o
u
s
ly
r
ef
in
in
g
al
g
o
r
it
h
m
s
to
f
u
r
th
e
r
an
a
ly
z
e
u
s
er
-
p
r
ef
er
en
ce
s
,
in
co
r
p
o
r
at
in
g
co
n
tex
tu
a
l
in
f
o
r
m
a
ti
o
n
s
u
ch
a
s
ti
m
e,
l
o
ca
t
io
n
,
a
n
d
d
ev
i
ce
ty
p
e
to
d
e
liv
er
h
y
p
er
-
p
er
s
o
n
a
li
ze
d
r
ec
o
m
m
en
d
at
io
n
s
tai
lo
r
ed
to
in
d
iv
i
d
u
a
l
p
r
ef
er
en
c
e
s
an
d
cir
cu
m
s
t
an
c
es
.
‒
Mu
lti
-
m
o
d
al
r
ec
o
m
m
en
d
atio
n
s
:
e
m
b
r
ac
in
g
a
m
u
ltimo
d
al
ap
p
r
o
ac
h
th
at
tr
an
s
ce
n
d
s
tr
ad
itio
n
al
b
o
u
n
d
ar
ies,
co
m
b
in
in
g
tex
t,
i
m
ag
e,
au
d
io
,
a
n
d
o
th
er
f
o
r
m
s
o
f
d
ata
t
o
o
f
f
er
r
ich
er
,
m
o
r
e
im
m
er
s
iv
e
co
n
ten
t r
ec
o
m
m
en
d
ati
o
n
s
th
at
r
eso
n
ate
with
d
iv
er
s
e
a
u
d
ien
c
es a
cr
o
s
s
g
lo
b
al
m
ar
k
ets.
‒
I
n
ter
p
r
etab
le
an
d
tr
an
s
p
ar
en
t
m
o
d
els:
d
ev
elo
p
in
g
r
ec
o
m
m
e
n
d
atio
n
m
o
d
els
th
at
ar
e
n
o
t
o
n
ly
ac
cu
r
ate
b
u
t
also
in
ter
p
r
etab
le,
p
r
o
v
id
in
g
u
s
er
s
with
in
s
ig
h
ts
in
to
th
e
r
atio
n
ale
b
eh
in
d
ea
ch
r
ec
o
m
m
en
d
a
tio
n
,
th
er
e
b
y
f
o
s
ter
in
g
tr
u
s
t,
tr
a
n
s
p
ar
en
cy
,
a
n
d
u
s
er
en
g
ag
em
e
n
t a
t scale
.
‒
C
o
n
t
i
n
u
o
u
s
e
v
a
l
u
a
t
i
o
n
a
n
d
f
ee
d
b
ac
k
l
o
o
p
:
i
n
s
t
i
t
u
t
i
n
g
r
o
b
u
s
t
m
e
c
h
a
n
i
s
m
s
f
o
r
co
n
t
in
u
o
u
s
e
v
a
lu
a
t
io
n
o
f
r
e
c
o
m
m
en
d
a
t
io
n
p
e
r
f
o
r
m
an
c
e
,
g
a
t
h
e
r
in
g
u
s
e
r
f
e
e
d
b
a
ck
,
a
n
d
i
t
er
a
t
iv
e
l
y
r
e
f
in
i
n
g
t
h
e
s
y
s
t
e
m
t
o
ad
a
p
t
t
o
e
v
o
l
v
in
g
u
s
e
r
p
r
e
f
er
e
n
ce
s
,
c
u
l
t
u
r
a
l
n
u
an
c
e
s
,
an
d
in
d
u
s
t
r
y
t
r
en
d
s
w
i
th
i
n
th
e
d
y
n
a
m
i
c
la
n
d
s
c
ap
e
o
f
b
i
g
d
a
t
a
[
5
2
]
,
[
5
3
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.