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
.
1
,
Feb
r
u
ar
y
2
0
2
5
,
p
p
.
810
~
821
I
SS
N:
2
2
5
2
-
8
9
3
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijai.v
14
.i
1
.
p
p
8
1
0
-
8
2
1
810
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
M
ethod for
dev
el
o
ping
and pa
rtit
i
o
ning
gra
ph
-
ba
sed da
ta
wa
reho
uses
using
as
so
cia
tion rules
Redo
ua
ne
L
a
bzio
ui,
K
ha
dij
a
L
et
ra
che,
M
o
ha
m
m
ed
Ra
m
da
ni
LI
M
La
b
o
r
a
t
o
r
y
,
F
a
c
u
l
t
y
o
f
S
c
i
e
n
c
e
s
a
n
d
T
e
c
h
n
i
q
u
e
s
o
f
M
o
h
a
m
me
d
i
a
,
U
n
i
v
e
r
si
t
y
H
a
ss
a
n
I
I
,
C
a
sa
b
l
a
n
c
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
M
ar
2
3
,
2
0
2
4
R
ev
is
ed
O
ct
1
7
,
2
0
2
4
Acc
ep
ted
Oct
2
1
,
2
0
2
4
Th
e
e
v
o
lu
ti
o
n
o
f
m
o
d
e
r
n
d
a
tab
a
se
s
h
a
s
led
t
o
a
v
a
riety
o
f
n
o
t
o
n
l
y
stru
c
tu
re
d
q
u
e
ry
lan
g
u
a
g
e
(No
S
QL
)
m
o
d
e
ls,
p
a
rti
c
u
larl
y
g
ra
p
h
-
o
rien
ted
-
d
a
tab
a
se
s.
Th
is
g
ro
wt
h
h
a
s
e
n
c
o
u
ra
g
e
d
b
u
si
n
e
ss
e
s
to
e
x
p
l
o
re
g
ra
p
h
-
b
a
se
d
b
u
si
n
e
ss
in
telli
g
e
n
c
e
(BI)
s
o
lu
ti
o
n
s.
Th
is
p
a
p
e
r
e
x
p
lo
re
s
th
re
e
e
ss
e
n
ti
a
l
a
sp
e
c
ts
in
t
h
e
d
o
m
a
in
o
f
g
ra
p
h
wa
re
h
o
u
se
:
th
e
e
sta
b
li
sh
m
e
n
t
o
f
e
fficie
n
t
g
ra
p
h
wa
re
h
o
u
se
s,
th
e
sig
n
if
ica
n
c
e
o
f
d
a
ta
h
isto
r
i
z
a
ti
o
n
,
a
n
d
th
e
d
e
v
e
lo
p
m
e
n
t
o
f
e
ffe
c
ti
v
e
stra
teg
ies
fo
r
g
ra
p
h
p
a
rti
ti
o
n
in
g
.
It
st
a
rts
b
y
b
u
il
d
in
g
a
BI
sy
s
tem
with
in
a
g
ra
p
h
d
a
tab
a
se
.
S
u
b
se
q
u
e
n
tl
y
,
t
h
e
p
a
p
e
r
e
m
p
h
a
siz
e
s
th
e
p
i
v
o
tal
r
o
le
o
f
d
a
ta
h
isto
riza
ti
o
n
,
h
i
g
h
l
ig
h
ti
n
g
t
h
e
sl
o
wly
g
ra
p
h
c
h
a
n
g
in
g
d
ime
n
sio
n
(S
G
CD
)
a
p
p
ro
a
c
h
a
s
a
v
e
rsa
ti
le
fra
m
e
wo
rk
f
o
r
a
c
c
o
m
m
o
d
a
ti
n
g
v
a
ried
d
i
m
e
n
sio
n
a
l
c
h
a
n
g
e
s,
a
d
d
it
io
n
a
ll
y
;
th
e
p
a
p
e
r
in
tro
d
u
c
e
s
a
n
o
v
e
l
p
a
rti
ti
o
n
in
g
stra
teg
y
u
ti
li
z
i
n
g
a
ss
o
c
iati
o
n
ru
les
a
l
g
o
r
it
h
m
s,
f
o
r
o
p
ti
m
ize
d
a
n
d
sc
a
lab
le
g
ra
p
h
wa
re
h
o
u
se
m
a
n
a
g
e
m
e
n
t.
K
ey
w
o
r
d
s
:
Ass
o
ciatio
n
r
u
les
B
u
s
in
ess
in
tell
ig
en
ce
Gr
ap
h
war
eh
o
u
s
e
Gr
ap
h
-
o
r
ien
ted
-
d
atab
ases
N
o
t
o
n
l
y
s
t
r
u
c
t
u
r
e
d
q
u
e
r
y
l
a
n
g
u
a
g
e
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
:
R
ed
o
u
an
e
L
ab
zi
o
u
i
L
I
M
L
ab
o
r
at
o
r
y
,
Facu
lty
o
f
S
cien
ce
s
an
d
T
ec
h
n
iq
u
es o
f
Mo
h
am
m
ed
ia
,
Un
i
v
er
s
ity
Hass
an
I
I
C
asab
lan
ca
,
Mo
r
r
o
co
E
m
ail:
r
ed
o
u
a
n
e.
lab
zio
u
i
-
etu
@
etu
.
u
n
iv
h
2
c.
m
a
1.
I
NT
RO
D
UCT
I
O
N
E
x
p
an
d
i
n
g
d
ata
s
ets
h
av
e
s
ig
n
if
ican
tly
alter
ed
th
e
lan
d
s
ca
p
e
o
f
m
o
d
e
r
n
d
atab
ases
,
p
ar
ti
cu
lar
ly
th
e
em
er
g
en
ce
an
d
ex
p
an
s
io
n
o
f
v
ar
io
u
s
n
o
t
o
n
ly
s
tr
u
ctu
r
ed
q
u
er
y
lan
g
u
a
g
e
(
No
SQL)
m
o
d
els
lik
e
d
o
cu
m
en
t,
co
lu
m
n
,
k
ey
-
v
alu
e,
an
d
g
r
ap
h
d
atab
ases
[
1
]
,
[
2
]
.
Am
o
n
g
th
ese,
g
r
a
p
h
-
o
r
ien
ted
d
atab
a
s
es
h
av
e
g
ar
n
e
r
e
d
s
u
b
s
tan
tial
atten
tio
n
,
th
e
y
o
f
f
e
r
a
u
n
iq
u
e
ab
s
tr
ac
tio
n
to
h
an
d
le
d
en
s
ely
c
o
n
n
ec
ted
d
ata,
allo
win
g
f
o
r
c
o
m
p
lex
d
o
m
ain
m
o
d
elin
g
a
n
d
ex
ec
u
tio
n
o
f
in
tr
icate
q
u
er
ies
[
3
]
,
[
4
]
.
T
h
e
in
cr
ea
s
ed
atten
tio
n
h
as
led
to
a
n
u
m
b
er
o
f
b
u
s
in
ess
es st
ar
tin
g
p
r
o
jects to
d
ev
elo
p
b
u
s
in
ess
in
tellig
en
ce
(
B
I
)
s
o
lu
tio
n
s
with
g
r
ap
h
s
[
5
]
,
[
6
]
.
Ho
wev
er
,
d
ata
h
is
to
r
izatio
n
-
t
h
e
p
r
o
ce
s
s
o
f
co
n
s
er
v
in
g
d
ata
ac
r
o
s
s
tim
e
-
is
b
ec
o
m
in
g
m
o
r
e
an
d
m
o
r
e
n
ec
ess
ar
y
d
u
e
to
t
h
e
in
cr
ea
s
in
g
co
m
p
le
x
ity
o
f
d
ata.
Fo
r
th
e
p
u
r
p
o
s
e
o
f
m
o
n
ito
r
in
g
ch
an
g
es
o
v
e
r
tim
e,
ex
am
in
in
g
p
ast
tr
en
d
s
,
ad
h
e
r
in
g
to
leg
al
o
b
lig
atio
n
s
,
an
d
m
ak
in
g
wis
e
d
ec
is
io
n
s
[
7
]
.
I
n
g
r
ap
h
-
b
ased
d
ata
war
eh
o
u
s
es,
wh
er
e
d
i
m
en
s
io
n
s
ar
e
f
r
eq
u
e
n
tly
u
s
ed
to
d
escr
ib
e
en
titi
es
with
in
th
e
g
r
ap
h
,
th
is
b
ec
o
m
es
v
er
y
im
p
o
r
tan
t,
u
n
d
er
s
tan
d
i
n
g
th
e
ev
o
lu
tio
n
o
f
th
e
g
r
a
p
h
an
d
its
co
m
p
o
n
e
n
ts
b
ec
o
m
es
d
if
f
icu
l
t
in
th
e
ab
s
en
ce
o
f
d
ata
h
is
to
r
izatio
n
[
8
]
.
I
n
ad
d
itio
n
to
h
is
to
r
izatio
n
,
th
is
p
ap
er
ad
d
r
ess
es
an
o
th
er
cr
itical
ch
allen
g
e:
d
ata
p
ar
titi
o
n
in
g
with
in
th
ese
g
r
ap
h
-
b
ased
d
ata
war
eh
o
u
s
es.
Gr
ap
h
s
,
as
co
m
p
lex
s
tr
u
ctu
r
es,
r
eq
u
ir
e
a
n
u
an
ce
d
ap
p
r
o
ac
h
to
ef
f
ec
tiv
ely
m
an
ag
e
p
ar
titi
o
n
s
[
9
]
.
W
ell
-
m
an
ag
ed
p
ar
titi
o
n
s
ar
e
ess
en
tial
f
o
r
o
p
tim
izin
g
q
u
er
ies,
en
s
u
r
in
g
ef
f
icien
t
s
to
r
ag
e,
an
d
en
ab
lin
g
q
u
ick
ac
ce
s
s
to
r
elev
an
t
d
ata
[
1
0
]
,
[
1
1
]
.
T
h
er
ef
o
r
e,
th
is
p
ap
er
aim
s
to
co
n
cu
r
r
en
tly
ad
d
r
ess
th
r
ee
cr
u
cial
asp
ec
ts
:
th
e
cr
ea
tio
n
o
f
ef
f
icien
t
d
ata
wa
r
eh
o
u
s
es
u
n
d
er
g
r
a
p
h
d
atab
ase
s
y
s
tem
s
,
th
e
im
p
o
r
tan
ce
o
f
d
ata
h
is
to
r
izatio
n
,
an
d
th
e
ef
f
ec
tiv
e
m
an
ag
em
e
n
t o
f
d
ata
p
ar
titi
o
n
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
Meth
o
d
fo
r
d
ev
elo
p
i
n
g
a
n
d
p
a
r
titi
o
n
in
g
g
r
a
p
h
-
b
a
s
ed
d
a
ta
w
a
r
eh
o
u
s
es u
s
in
g
…
(
R
e
d
o
u
a
n
e
La
b
z
io
u
i
)
811
First,
we
co
n
s
tr
u
ct
a
B
I
s
y
s
tem
with
in
a
g
r
ap
h
d
atab
ase.
W
e
th
en
ex
am
in
e
t
h
e
im
p
o
r
ta
n
ce
o
f
d
ata
h
is
to
r
izatio
n
,
f
o
c
u
s
in
g
o
n
th
e
s
lo
wly
g
r
ap
h
c
h
an
g
i
n
g
d
im
e
n
s
io
n
(
SGC
D)
ap
p
r
o
ac
h
,
wh
i
ch
is
ad
ap
tab
le
to
d
if
f
er
en
t
t
y
p
es
o
f
d
im
en
s
io
n
al
ch
an
g
es
o
v
er
tim
e.
T
h
is
f
o
cu
s
o
n
h
is
to
r
izatio
n
ad
d
r
ess
es
t
h
e
g
ap
in
p
r
ev
io
u
s
r
esear
ch
b
y
d
etailin
g
s
tr
ateg
ies
f
o
r
tr
ac
k
in
g
an
d
m
an
ag
i
n
g
c
h
a
n
g
es
o
v
er
tim
e
in
g
r
ap
h
d
ata
w
ar
eh
o
u
s
es.
Fin
ally
,
we
e
x
p
lo
r
e
s
tr
ateg
ies
an
d
alg
o
r
ith
m
s
b
ased
o
n
ass
o
ciatio
n
r
u
les
f
o
r
e
f
f
ec
tiv
ely
m
an
a
g
in
g
p
ar
ti
tio
n
s
in
th
e
co
n
tex
t o
f
g
r
a
p
h
-
b
ased
d
ata
w
ar
eh
o
u
s
es.
T
h
e
r
est
o
f
th
is
d
o
cu
m
en
t
is
s
tr
u
ctu
r
ed
as
o
u
tlin
ed
b
elo
w:
s
ec
tio
n
2
ex
p
lo
r
es t
h
e
e
x
is
tin
g
li
ter
atu
r
e
o
n
g
r
ap
h
d
ata
war
e
h
o
u
s
in
g
.
I
n
s
ec
tio
n
3
,
we
p
r
o
v
id
e
th
e
n
ec
ess
ar
y
b
ac
k
g
r
o
u
n
d
o
f
th
e
a
p
p
r
o
ac
h
.
Sectio
n
4
d
el
v
es
in
to
th
e
im
p
lem
en
tatio
n
o
f
t
h
e
s
tr
ateg
y
.
Sectio
n
5
d
is
cu
s
s
es
th
e
p
r
in
cip
al
r
esu
lts
o
f
th
e
s
tu
d
y
.
Fin
ally
,
th
is
wo
r
k
is
co
n
clu
d
ed
in
s
ec
tio
n
6
,
w
h
ich
also
s
u
g
g
ests
o
p
tio
n
s
f
o
r
f
u
tu
r
e
r
esear
ch
.
2.
RE
L
AT
E
D
WO
RK
S
I
n
th
e
liter
atu
r
e,
d
is
cu
s
s
io
n
s
ab
o
u
t d
ata
p
ar
titi
o
n
i
n
g
an
d
h
is
to
r
izatio
n
h
av
e
p
r
ed
o
m
in
an
tly
f
o
cu
s
ed
o
n
r
elatio
n
al
d
ata
war
eh
o
u
s
es,
with
a
lim
ited
ex
p
lo
r
atio
n
in
to
g
r
ap
h
war
eh
o
u
s
e.
Ho
wev
e
r
,
r
ec
en
t
y
ea
r
s
h
a
v
e
witn
ess
ed
a
g
r
o
win
g
in
ter
est
in
in
teg
r
atin
g
BI
tech
n
o
lo
g
y
with
g
r
ap
h
d
atab
ases
,
r
esu
ltin
g
in
v
ar
io
u
s
p
r
o
p
o
s
ed
m
eth
o
d
o
l
o
g
ies
d
o
cu
m
en
ted
i
n
th
e
liter
atu
r
e.
Z
h
ao
et
a
l
.
[
1
2
]
h
av
e
p
r
esen
ted
a
b
r
an
d
-
n
ew
id
ea
k
n
o
wn
as
th
e
g
r
ap
h
c
u
b
e
,
th
at
is
a
n
ew
d
ata
war
eh
o
u
s
e
p
ar
a
d
ig
m
in
te
n
d
ed
to
h
an
d
le
m
u
ltid
im
en
s
io
n
al
q
u
e
r
ies
in
s
id
e
lar
g
e
-
s
ca
le
m
u
ltid
im
en
s
io
n
al
n
etwo
r
k
s
.
T
h
is
ap
p
r
o
ac
h
o
r
g
a
n
izes
d
im
en
s
io
n
s
b
ased
o
n
n
o
d
e
attr
ib
u
tes
wh
ile
em
p
lo
y
in
g
co
m
p
u
ted
m
ea
s
u
r
e
s
to
ag
g
r
eg
ate
th
ese
n
o
d
e
attr
i
b
u
tes,
au
th
o
r
s
d
id
n
o
t a
d
d
r
ess
th
e
p
ar
titi
o
n
in
g
a
n
d
h
is
to
r
izatio
n
asp
ec
t
with
in
th
e
g
r
ap
h
war
e
h
o
u
s
e.
C
astellt
o
r
t
an
d
L
au
r
e
n
t
[
1
3
]
in
tr
o
d
u
c
ed
a
m
eth
o
d
o
lo
g
y
s
u
g
g
esti
n
g
th
e
u
s
e
o
f
g
r
ap
h
s
tr
u
ctu
r
e
f
o
r
o
n
lin
e
a
n
aly
tical
p
r
o
ce
s
s
in
g
(
OL
AP
)
q
u
er
ies,
lev
er
ag
in
g
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
g
r
ap
h
d
ata
b
ase
f
o
r
s
to
r
ag
e
an
d
q
u
er
y
ti
m
e
p
r
o
ce
s
s
in
g
.
T
h
is
ap
p
r
o
ac
h
in
v
o
l
v
es
co
n
v
er
tin
g
m
ea
s
u
r
es
an
d
d
i
m
en
s
io
n
s
in
to
n
o
d
es
with
in
th
e
g
r
a
p
h
,
u
s
in
g
ar
cs
to
estab
lis
h
r
elatio
n
s
h
ip
s
b
etwe
en
d
im
en
s
io
n
s
an
d
m
ea
s
u
r
em
en
ts
.
Hier
ar
ch
i
ca
l
in
ter
ac
tio
n
s
am
o
n
g
n
o
d
e
s
r
etain
h
ier
ar
ch
ical
d
im
en
s
io
n
s
.
Ho
wev
er
,
th
is
s
tr
ateg
y
is
co
n
f
in
ed
s
o
lely
to
t
h
e
s
n
o
wf
lak
e
m
o
d
el,
p
o
ten
tial
ly
li
m
itin
g
its
ap
p
licab
ilit
y
to
o
th
er
d
ata
m
o
d
els.
Mu
ttip
ati
an
d
Pad
m
aja
[
1
4
]
o
f
f
er
a
co
m
p
r
eh
en
s
iv
e
o
v
er
v
ie
w
o
f
ex
is
tin
g
to
o
ls
an
d
ap
p
r
o
ac
h
es
f
o
r
g
r
ap
h
p
ar
titi
o
n
in
g
a
n
d
f
r
eq
u
en
t
s
u
b
-
g
r
a
p
h
ex
tr
ac
tio
n
.
I
t
ex
p
lo
r
es
a
wid
e
ar
r
ay
o
f
to
p
ics,
en
co
m
p
ass
in
g
g
r
ap
h
-
p
ar
titi
o
n
in
g
tech
n
iq
u
es,
f
r
eq
u
en
t
s
u
b
-
g
r
ap
h
m
in
in
g
a
lg
o
r
ith
m
s
,
p
ar
allel
p
r
o
ce
s
s
in
g
f
r
am
ewo
r
k
s
,
an
d
m
an
ag
in
g
s
u
b
s
tan
tial
v
o
lu
m
es
o
f
g
r
a
p
h
d
ata.
Ho
wev
er
,
it
is
wo
r
th
n
o
tin
g
th
at
th
e
s
tu
d
y
m
ig
h
t
lack
in
-
d
ep
th
ex
p
er
ien
ce
a
n
d
em
p
i
r
ical
r
esu
lts
to
s
u
b
s
tan
t
iate
it
s
f
in
d
in
g
s
f
u
lly
.
W
h
ile
its
b
r
o
ad
s
co
p
e
p
r
o
v
id
es
a
wea
lth
o
f
in
f
o
r
m
atio
n
,
it
m
i
g
h
t
f
all
s
h
o
r
t
in
th
o
r
o
u
g
h
ly
ex
am
in
i
n
g
s
p
ec
if
ic
asp
ec
ts
o
r
o
f
f
e
r
in
g
d
e
tailed
in
s
ig
h
ts
in
to
ce
r
tain
alg
o
r
ith
m
s
o
r
p
ar
allel
f
r
am
ewo
r
k
s
.
Ad
d
itio
n
ally
,
it
d
o
es
n
o
t
d
elv
e
in
to
p
ar
titi
o
n
in
g
o
f
th
e
g
r
ap
h
war
eh
o
u
s
e,
lim
itin
g
its
ex
p
lo
r
atio
n
to
ce
r
tain
a
r
ea
s
wit
h
in
th
e
r
ea
lm
o
f
g
r
ap
h
d
atab
ase
s
.
Dai
et
a
l.
[
1
5
]
in
tr
o
d
u
ce
d
th
e
in
cr
em
e
n
tal
o
n
l
in
e
g
r
a
p
h
p
ar
titi
o
n
in
g
(
I
OGP)
alg
o
r
ith
m
,
aim
ed
at
ad
d
r
ess
in
g
p
er
f
o
r
m
an
ce
n
ee
d
s
in
d
is
tr
ib
u
ted
g
r
a
p
h
d
atab
ases
.
I
OGP
d
y
n
am
ically
ad
ju
s
ts
th
r
o
u
g
h
th
r
ee
o
p
e
r
atio
n
al
s
tag
es,
ef
f
icien
tly
ac
co
m
m
o
d
atin
g
c
o
n
tin
u
o
u
s
g
r
ap
h
c
h
an
g
es.
I
t
s
wif
tly
p
r
o
d
u
ce
s
o
p
tim
ized
p
ar
titi
o
n
ed
g
r
ap
h
s
,
p
r
o
f
ici
en
tl
y
s
er
v
in
g
c
o
m
p
lex
tr
av
e
r
s
als.
I
m
p
lem
en
tatio
n
d
etails,
in
clu
d
in
g
in
-
m
em
o
r
y
d
ata
s
tr
u
ct
u
r
es
lik
e
ed
g
e
co
u
n
ter
s
,
f
ac
ilit
ate
r
ap
id
o
n
lin
e
g
r
a
p
h
p
ar
titi
o
n
in
g
.
E
x
ten
s
iv
e
ev
a
lu
atio
n
s
ac
r
o
s
s
d
iv
er
s
e
g
r
ap
h
s
af
f
ir
m
I
OGP's
ad
v
an
tag
es,
aid
i
n
g
in
estab
lis
h
in
g
k
e
y
p
a
r
am
eter
s
elec
tio
n
g
u
id
elin
es.
Ak
id
et
a
l.
[
1
6
]
s
u
g
g
est
g
u
id
elin
es
f
o
r
tr
an
s
f
o
r
m
in
g
a
m
u
ltid
im
en
s
io
n
al
d
ata
m
o
d
el
i
n
to
a
g
r
a
p
h
d
ata
m
o
d
el
(
MD
M2
G)
a
n
d
co
m
p
ar
e
th
e
p
er
f
o
r
m
a
n
ce
o
f
s
n
o
wf
lak
e
m
o
d
el
an
d
s
tar
d
esig
n
s
in
b
o
th
g
r
ap
h
d
atab
ases
an
d
r
elatio
n
al,
f
o
c
u
s
in
g
o
n
s
ize
an
d
d
im
en
s
io
n
ality
.
T
h
eir
co
m
p
ar
i
s
o
n
s
u
g
g
ests
th
at
em
p
lo
y
in
g
a
g
r
ap
h
-
b
ased
im
p
lem
en
tatio
n
f
o
r
a
m
u
lti
-
tab
le
d
ata
war
eh
o
u
s
e
h
as
g
r
ea
ter
ef
f
icien
cy
th
an
a
r
elatio
n
al
ap
p
r
o
ac
h
.
Mo
r
eo
v
e
r
,
with
in
g
r
a
p
h
d
ata
b
ases
,
a
s
tar
m
o
d
el
d
em
o
n
s
tr
ates sim
ilar
p
er
f
o
r
m
a
n
ce
to
a
s
n
o
wf
la
k
e
m
o
d
el.
An
d
r
iam
am
p
ian
i
n
a
et
a
l.
[
1
7
]
s
u
g
g
ested
a
co
n
ce
p
tu
al
f
r
a
m
ewo
r
k
f
o
r
tem
p
o
r
al
g
r
ap
h
s
d
ep
ictin
g
ev
o
lv
in
g
g
r
ap
h
d
ata.
I
t
ca
p
t
u
r
es
h
o
w
d
ata
ch
an
g
es
o
v
er
tim
e
with
o
u
t
s
ac
r
if
icin
g
in
f
o
r
m
at
io
n
o
r
in
tr
o
d
u
ci
n
g
r
ed
u
n
d
an
cy
,
w
h
ich
d
if
f
e
r
s
f
r
o
m
h
o
w
s
n
ap
s
h
o
t
-
b
ased
m
o
d
els
h
an
d
le
th
is
ev
o
lu
tio
n
.
Firstl
y
,
it
is
tailo
r
ed
to
war
d
s
b
u
s
in
ess
n
ee
d
s
,
o
f
f
er
i
n
g
n
o
n
-
ex
p
er
t
u
s
er
s
a
co
m
p
lete
u
n
d
er
s
tan
d
in
g
o
f
d
ata
an
d
its
ev
o
lv
in
g
elem
en
ts
.
Seco
n
d
ly
,
it
is
v
er
s
atile,
ca
p
a
b
le
o
f
r
e
p
r
esen
tin
g
v
ar
io
u
s
ty
p
es
o
f
ch
an
g
es
in
g
r
a
p
h
d
ata;
in
clu
d
in
g
t
o
p
o
lo
g
y
s
h
if
ts
an
d
alter
a
tio
n
s
in
attr
ib
u
te
s
ets
an
d
v
alu
es.
L
astl
y
,
it
ac
cu
r
ately
ca
p
tu
r
es
th
e
tem
p
o
r
al
ev
o
lu
tio
n
o
f
d
ata,
p
r
eser
v
in
g
in
f
o
r
m
atio
n
an
d
a
v
o
id
in
g
r
e
d
u
n
d
an
cy
,
u
n
lik
e
s
n
ap
s
h
o
t
-
b
ased
m
o
d
els.
Ho
wev
er
,
wh
en
ap
p
lied
to
lar
g
e
d
atasets
,
tem
p
o
r
al
co
n
te
x
t
an
aly
s
is
an
d
m
an
ip
u
la
tio
n
ca
n
o
cc
asio
n
ally
r
esu
lt
in
s
lo
wer
p
r
o
ce
s
s
in
g
tim
es
an
d
r
ed
u
ce
d
p
e
r
f
o
r
m
an
ce
.
M
o
r
eo
v
e
r
,
tem
p
o
r
al
g
r
a
p
h
s
ca
n
q
u
ick
ly
g
r
o
w
in
s
ize
with
ad
d
ed
d
ata
o
v
er
tim
e.
Ma
n
ag
in
g
th
ese
lar
g
e
d
atasets
ca
n
b
e
co
s
tly
in
ter
m
s
o
f
s
to
r
a
g
e
an
d
p
r
o
ce
s
s
in
g
,
r
eq
u
ir
i
n
g
a
d
d
itio
n
al
r
eso
u
r
ce
s
.
Z
h
o
u
et
a
l.
[
1
8
]
d
escr
ib
e
a
n
o
v
el
m
eth
o
d
f
o
r
r
e
p
r
esen
tin
g
d
at
a
an
d
q
u
er
ies
u
s
in
g
g
r
a
p
h
to
p
o
lo
g
ies,
p
r
o
v
id
in
g
a
d
is
tin
ctiv
e
way
to
v
is
u
alize
co
lu
m
n
c
o
r
r
elatio
n
s
.
I
t
ch
o
o
s
es
p
ar
titi
o
n
in
g
k
ey
s
b
y
u
tili
zin
g
g
r
ap
h
em
b
e
d
d
in
g
’
s
,
wh
ich
m
ay
in
cr
ea
s
e
th
e
ef
f
ec
t
iv
en
ess
o
f
d
ata
p
ar
ti
tio
n
in
g
.
I
t
also
s
u
g
g
ests
a
lear
n
in
g
ass
e
s
s
m
en
t
m
o
d
el
th
at
esti
m
ates
th
e
p
er
f
o
r
m
an
ce
o
f
p
ar
titi
o
n
in
g
s
tr
ateg
ies
with
o
u
t
p
h
y
s
ically
d
iv
id
in
g
d
ata,
p
o
t
en
tially
co
n
s
er
v
in
g
co
m
p
u
tin
g
p
o
wer
.
B
en
h
is
s
en
et
a
l.
[
1
9
]
s
u
g
g
est
a
m
eth
o
d
ce
n
ter
ed
ar
o
u
n
d
a
p
r
o
g
r
ess
iv
e
s
ch
em
a
m
o
d
el
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
.
1
,
Feb
r
u
ar
y
20
25
:
810
-
8
2
1
812
f
ea
tu
r
in
g
m
u
ltip
le
v
er
s
io
n
s
,
w
h
er
e
a
g
r
ap
h
d
ata
war
eh
o
u
s
e
a
cc
o
m
m
o
d
ates d
ata
in
s
tan
ce
s
a
lig
n
ed
with
d
is
tin
ct
s
ch
em
a
v
er
s
io
n
s
.
T
h
e
h
a
n
d
lin
g
o
f
th
ese
v
er
s
io
n
s
is
m
ad
e
ea
s
ier
u
s
in
g
a
d
e
d
icate
d
m
eta
-
m
o
d
el
f
o
r
war
eh
o
u
s
e
s
ch
em
a
v
er
s
io
n
s
,
alo
n
g
with
th
e
in
te
g
r
atio
n
o
f
ev
o
l
u
tio
n
ar
y
f
u
n
ctio
n
s
i
n
tr
o
d
u
ce
d
at
th
e
s
ch
em
a
lev
el
.
Nev
er
th
eless
,
th
e
r
esear
ch
d
o
e
s
n
o
t d
elv
e
in
to
th
e
asp
ec
t o
f
OL
AP c
u
b
e
p
ar
titi
o
n
in
g
with
in
g
r
ap
h
s
.
I
t
is
clea
r
f
r
o
m
th
e
p
r
ev
io
u
s
o
v
er
v
iew
t
h
at
th
e
m
ajo
r
ity
o
f
th
ese
p
ap
e
r
s
p
r
im
a
r
ily
co
n
ce
n
tr
ate
o
n
tr
an
s
f
o
r
m
in
g
class
ical
d
ata
war
eh
o
u
s
e
in
to
g
r
ap
h
s
.
T
h
is
em
p
h
asis
co
u
ld
p
o
ten
tially
r
estrict
th
e
th
o
r
o
u
g
h
ex
p
lo
r
atio
n
an
d
u
tili
za
tio
n
o
f
th
e
co
m
p
lete
b
en
ef
its
p
r
o
v
id
e
d
b
y
g
r
ap
h
d
atab
ases
.
Ad
d
itio
n
ally
,
th
ese
s
tu
d
ies
o
v
er
lo
o
k
th
e
asp
ec
t
o
f
h
is
to
r
i
ca
l
d
ata
an
d
p
ar
titi
o
n
in
g
g
r
ap
h
war
eh
o
u
s
es,
cr
ea
t
in
g
a
n
o
th
e
r
s
ig
n
if
ican
t
g
a
p
in
th
eir
co
v
e
r
ag
e.
I
n
c
o
n
tr
ast,
o
u
r
s
u
g
g
ested
m
eth
o
d
m
a
k
es
ad
v
a
n
tag
e
o
f
g
r
a
p
h
f
ea
tu
r
es
to
b
u
ild
g
r
ap
h
war
eh
o
u
s
es,
ad
r
ess
es a
ls
o
th
e
asp
ec
t o
f
d
at
a
h
is
to
r
izatio
n
,
an
d
u
s
es a
s
s
o
ciatio
n
r
u
les f
o
r
p
ar
titi
o
n
in
g
g
r
ap
h
war
eh
o
u
s
es.
3.
O
UR
AP
P
RO
ACH
Ou
r
m
eth
o
d
en
tails
f
ir
s
t
b
u
ild
i
n
g
a
g
r
a
p
h
war
eh
o
u
s
e,
with
a
p
ar
ticu
lar
em
p
h
asis
o
n
th
e
SGC
D
m
eth
o
d
f
o
r
d
ata
h
is
to
r
izatio
n
.
T
h
is
m
e
th
o
d
is
s
p
ec
if
ically
d
esig
n
e
d
t
o
m
an
ag
e
ch
an
g
es in
g
r
ap
h
d
i
m
en
s
io
n
s
o
v
er
tim
e,
en
s
u
r
in
g
th
at
th
e
h
is
to
r
ical
ev
o
lu
tio
n
o
f
d
ata
is
ac
cu
r
ately
ca
p
tu
r
ed
.
T
h
e
SGC
D
en
ab
les
th
e
war
eh
o
u
s
e
to
m
ain
tain
a
co
m
p
r
eh
en
s
iv
e
h
i
s
to
r
ical
r
ec
o
r
d
o
f
th
e
d
im
e
n
s
io
n
s
,
wh
ich
is
cr
u
cial
f
o
r
a
n
aly
zin
g
tr
en
d
s
an
d
p
er
f
o
r
m
in
g
tim
e
-
b
ased
q
u
er
i
es
[
2
0
]
.
Su
b
s
eq
u
en
tly
,
we
u
s
e
u
s
er
OL
AP
q
u
er
ies
as
in
p
u
t
f
o
r
two
cr
u
cial
alg
o
r
ith
m
s
:
T
h
e
Ap
r
i
o
r
i
alg
o
r
ith
m
[
2
1
]
,
wh
ich
i
d
en
tifie
s
f
r
e
q
u
en
tly
o
c
cu
r
r
i
n
g
item
s
ets
in
th
e
g
r
ap
h
cu
b
e,
an
d
a
r
u
le
-
b
ased
ass
o
ciatio
n
alg
o
r
i
th
m
is
em
p
lo
y
ed
to
d
etec
t
th
e
m
o
s
t
u
s
ed
p
ar
titi
o
n
s
[
2
2
]
.
T
h
e
p
r
im
ar
y
o
b
jectiv
e
o
f
th
is
tech
n
iq
u
e
is
to
o
p
tim
ize
o
u
r
OL
AP
cu
b
e
f
o
r
m
ax
im
u
m
p
er
f
o
r
m
a
n
ce
,
en
s
u
r
in
g
ef
f
icien
t
h
an
d
lin
g
o
f
u
s
er
q
u
er
ies an
d
d
ata
an
aly
s
is
.
T
h
e
r
e
ar
e
f
o
u
r
s
tep
s
in
th
is
ap
p
r
o
a
ch
:
‒
B
u
ild
in
g
a
g
r
a
p
h
war
eh
o
u
s
e.
‒
I
n
teg
r
ate
a
d
ata
h
is
to
r
izatio
n
co
m
p
o
n
en
t
in
to
th
e
p
r
o
ce
s
s
,
em
p
h
asizin
g
th
e
SGC
D
m
et
h
o
d
,
t
h
is
s
tep
en
s
u
r
es
a
co
m
p
r
e
h
en
s
iv
e
v
iew
o
f
d
ata
ch
a
n
g
es
o
v
er
tim
e,
h
i
g
h
lig
h
tin
g
a
d
ap
tab
ilit
y
to
v
ar
io
u
s
d
im
en
s
io
n
al
m
o
d
if
icatio
n
s
.
‒
Utilizin
g
u
s
er
q
u
er
ies
as
in
p
u
t
f
o
r
th
e
Ap
r
io
r
i
alg
o
r
ith
m
to
i
d
en
tify
f
r
eq
u
en
tly
o
cc
u
r
r
in
g
it
em
s
ets
with
in
th
e
g
r
ap
h
cu
b
e
.
‒
Usi
n
g
a
r
u
le
-
b
ased
ass
o
ciatio
n
alg
o
r
ith
m
to
id
en
tify
th
e
m
o
s
t
f
r
eq
u
e
n
tly
u
s
ed
p
ar
titi
o
n
s
.
3
.
1
.
G
ra
ph
wa
re
ho
us
e
Ou
r
m
eth
o
d
b
u
ild
s
th
e
g
r
ap
h
war
eh
o
u
s
e
b
y
co
m
b
in
in
g
th
e
in
h
er
en
t
f
lex
ib
ilit
y
o
f
g
r
ap
h
s
tr
u
ctu
r
es
with
th
e
co
n
ce
p
ts
o
f
m
u
ltid
im
en
s
io
n
al
m
o
d
elin
g
.
I
n
t
h
is
co
n
tex
t,
d
im
en
s
io
n
s
ar
e
m
o
d
eled
as
n
o
d
es,
an
d
f
ac
ts
,
wh
ich
co
n
tain
m
ea
s
u
r
em
e
n
ts
,
ar
e
r
e
p
r
esen
ted
as
n
o
d
e
s
,
cr
ea
tin
g
a
clea
r
an
d
n
av
ig
ab
le
s
tr
u
ct
u
r
e
f
o
r
q
u
er
y
i
n
g
an
d
an
aly
zin
g
d
ata
[
2
3
]
.
T
h
e
ed
g
e
s
in
th
is
g
r
ap
h
m
o
d
el
ar
e
cr
u
c
ial
as
th
ey
d
ef
in
e
th
e
r
elatio
n
s
h
ip
s
b
etwe
en
f
ac
ts
an
d
d
im
e
n
s
io
n
s
,
ca
p
tu
r
in
g
b
o
th
th
e
s
tr
u
ctu
r
e
o
f
th
e
d
ata
an
d
th
e
lin
k
s
b
etwe
en
d
if
f
er
en
t e
n
titi
es.
Dim
en
s
io
n
D
S
t
h
e
n
o
d
e
i
d
e
n
t
if
i
ca
ti
o
n
i
n
t
h
is
a
r
c
h
it
ec
t
u
r
e
is
u
s
e
d
t
o
r
e
p
r
ese
n
t
n
o
d
es
(
L
N
,
P
N
)
[
2
4
]
wh
e
r
e:
‒
L
N
in
d
icate
s
th
e
n
o
d
e'
s
n
am
e.
‒
P
N
in
d
icate
s
d
im
en
s
io
n
s
’
attr
ib
u
tes.
Fact:
T
h
e
f
ac
t
n
o
d
e
is
r
ep
r
es
en
ted
as
a
n
o
d
e
co
n
n
ec
ted
b
y
ed
g
es
to
d
im
en
s
io
n
n
o
d
es
[
2
5
]
.
Fu
r
th
er
m
o
r
e,
p
r
o
p
er
ties
r
elate
d
to
th
e
m
ea
s
u
r
em
en
ts
,
s
u
ch
as
ap
p
lied
ag
g
r
e
g
ate
f
u
n
ctio
n
s
o
r
ac
tu
al
v
al
u
es,
m
ay
b
e
p
r
esen
t
in
th
e
f
ac
t n
o
d
e
[
2
6
]
.
T
h
e
f
ac
t n
o
d
e
is
d
eter
m
in
ed
b
y
(
M
F
,
N
F
)
wh
er
e:
‒
N
F
: th
e
n
am
e
o
f
t
h
e
f
ac
t.
‒
M
F
:
it
co
n
s
is
ts
o
f
m
u
ltip
le
m
e
asu
r
em
en
ts
f
u
n
ctio
n
in
g
as
attr
ib
u
tes
f
o
r
n
o
d
es,
with
ea
ch
o
n
e
lin
k
ed
to
an
ag
g
r
eg
atio
n
f
u
n
ctio
n
.
T
h
e
lin
k
b
etwe
en
a
f
ac
t
n
o
d
e
an
d
its
ass
o
ciate
d
d
im
en
s
io
n
s
:
E
d
g
es
in
th
e
m
o
d
el
r
ep
r
es
en
t
th
e
r
elatio
n
s
h
ip
b
etwe
en
a
f
ac
t a
n
d
its
ass
o
ciat
ed
d
im
en
s
io
n
s
; th
is
r
elatio
n
s
h
i
p
is
d
ef
in
ed
b
y
(
L
E
, N
F
, N
D
, P
E
)
[
2
7
]
,
wh
er
e:
‒
L
E
is
th
e
r
elatio
n
s
h
ip
'
s
lab
el.
‒
N
F
is
th
e
f
ac
t n
o
d
e.
‒
N
D
th
e
n
o
d
e
r
ep
r
esen
tin
g
th
e
ass
o
ciate
d
d
im
en
s
io
n
.
‒
P
E
d
ef
in
e
th
e
ch
ar
ac
ter
is
tics
o
f
th
e
co
n
n
ec
tio
n
.
T
h
e
attr
i
b
u
tes
in
clu
d
e
k
ey
-
v
alu
e
p
ai
r
s
u
s
ed
to
s
to
r
e
r
elatio
n
s
h
ip
-
r
elate
d
d
ata
[
2
8
]
.
3
.
2
.
Da
t
a
his
t
o
riza
t
io
n in t
h
e
g
ra
ph
wa
re
ho
us
e
Up
o
n
th
e
estab
lis
h
m
en
t
o
f
o
u
r
g
r
ap
h
war
eh
o
u
s
e,
o
u
r
n
e
x
t
s
tr
ateg
ic
s
tep
in
v
o
lv
es
th
e
im
p
le
m
en
tatio
n
o
f
th
e
SGC
D
m
eth
o
d
o
lo
g
y
.
T
h
is
ap
p
r
o
ac
h
aim
s
to
en
h
a
n
ce
o
u
r
d
ata
war
eh
o
u
s
in
g
ca
p
a
b
ilit
ies
b
y
en
ab
lin
g
th
e
s
y
s
tem
atic
tr
ac
k
in
g
an
d
m
an
a
g
em
en
t
o
f
ch
a
n
g
in
g
d
im
en
s
io
n
s
with
in
th
e
g
r
ap
h
s
tr
u
ctu
r
e
[
2
9
]
.
B
y
ad
o
p
tin
g
SGC
D,
we
p
lan
to
m
eticu
lo
u
s
ly
ca
p
tu
r
e
an
d
p
r
eser
v
e
h
is
to
r
ic
al
d
ata
ch
an
g
es,
en
s
u
r
in
g
a
co
m
p
r
eh
en
s
iv
e
r
ec
o
r
d
o
f
ev
o
lv
in
g
in
f
o
r
m
atio
n
[
3
0
]
.
T
h
is
im
p
lem
en
tatio
n
will
en
ab
le
u
s
to
ef
f
icien
tly
h
an
d
le
v
ar
iatio
n
s
in
d
ata
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
Meth
o
d
fo
r
d
ev
elo
p
i
n
g
a
n
d
p
a
r
titi
o
n
in
g
g
r
a
p
h
-
b
a
s
ed
d
a
ta
w
a
r
eh
o
u
s
es u
s
in
g
…
(
R
e
d
o
u
a
n
e
La
b
z
io
u
i
)
813
d
im
en
s
io
n
s
o
v
e
r
tim
e,
en
s
u
r
in
g
ac
cu
r
ac
y
a
n
d
p
r
ec
is
io
n
i
n
a
n
aly
zin
g
h
is
to
r
ical
tr
en
d
s
an
d
p
atter
n
s
[
3
1
]
.
W
ith
SGC
D
in
p
lace
,
o
u
r
g
r
ap
h
war
eh
o
u
s
e
will
p
o
s
s
ess
a
r
o
b
u
s
t
f
o
u
n
d
atio
n
f
o
r
m
ain
tain
in
g
d
ata
in
teg
r
ity
an
d
f
ac
ilit
atin
g
in
-
d
ep
th
h
is
to
r
ical
an
aly
s
es c
r
u
cial
f
o
r
in
f
o
r
m
e
d
d
ec
is
io
n
-
m
ak
in
g
.
I
n
o
u
r
ap
p
r
o
ac
h
,
we
will
im
p
lem
en
t
SGC
D
b
y
u
tili
zin
g
Alg
o
r
ith
m
1
:
Ma
n
ag
in
g
h
is
to
r
izatio
n
d
im
en
s
io
n
s
in
g
r
ap
h
war
eh
o
u
s
es
.
T
h
is
alg
o
r
ith
m
is
d
esig
n
ed
to
ef
f
ec
tiv
ely
m
an
a
g
e
ch
an
g
es in
d
im
en
s
io
n
d
ata
b
y
cr
ea
tin
g
n
ew
v
er
s
io
n
s
wh
en
n
ec
ess
ar
y
an
d
u
p
d
atin
g
c
u
r
r
en
t
v
e
r
s
io
n
s
.
Ad
d
itio
n
ally
,
i
t
m
ain
tain
s
p
r
o
p
er
r
elatio
n
s
h
ip
s
with
in
th
e
g
r
a
p
h
-
b
ased
d
ata
war
eh
o
u
s
e
,
en
s
u
r
in
g
d
ata
in
teg
r
ity
an
d
c
o
n
s
is
ten
cy
.
Alg
o
r
ith
m
1
: M
a
n
ag
in
g
h
is
to
r
izatio
n
d
im
en
s
io
n
in
g
r
a
p
h
wa
r
eh
o
u
s
e
1.
R
eq
u
ir
e:
New
d
im
en
s
io
n
d
ata:
New
Data
2.
E
n
s
u
r
e:
Su
cc
ess
f
u
l o
p
er
atio
n
:
Su
cc
ess
3.
Fu
n
ctio
n
Ma
n
ag
eSGC
DFo
r
Dim
en
s
io
n
(
New
Data
)
:
4.
E
x
is
tin
g
R
ec
o
r
d
←Sear
ch
Dim
en
s
io
n
R
ec
o
r
d
B
y
Key
(
New
Data
.
Key
)
5.
I
f
E
x
is
tin
g
R
ec
o
r
d
is
n
o
t e
m
p
t
y
:
6.
I
f
E
x
is
tin
g
R
ec
o
r
d
.
Valu
e
≠
New
Data
.
Valu
e:
7.
Up
d
ateCu
r
r
en
tVer
s
io
n
(
E
x
is
tin
g
R
ec
o
r
d
)
8.
E
ls
e
Do
No
th
in
g
(
)
9.
E
n
d
I
f
10.
E
ls
e
C
r
ea
teNe
wVe
r
s
io
n
(
New
Data
)
11.
E
n
d
I
f
12.
R
etu
r
n
Su
cc
ess
13.
E
n
d
Fu
n
ctio
n
14.
Fu
n
ctio
n
Sear
ch
Dim
en
s
io
n
R
ec
o
r
d
B
y
Key
(
Key
)
:
15.
R
etu
r
n
Dim
en
s
io
n
R
ec
o
r
d
16.
E
n
d
Fu
n
ctio
n
17.
Fu
n
ctio
n
Up
d
ateCu
r
r
e
n
tVer
s
io
n
(
R
ec
o
r
d
)
:
18.
R
ec
o
r
d
.
Valid
T
o
←
C
u
r
r
en
tTi
m
estam
p
(
)
19.
New
Ver
s
io
n
←
C
r
ea
teNe
wVe
r
s
io
n
(
R
ec
o
r
d
)
20.
R
ec
o
r
d
.
C
u
r
r
en
tVer
s
io
n
←
Ne
wVe
r
s
io
n
21.
E
n
d
Fu
n
ctio
n
22.
Fu
n
ctio
n
C
r
ea
teNe
wVe
r
s
io
n
(
Data
)
:
23.
New
No
d
e
←
C
r
ea
teNe
wNo
d
e(
Data
)
24.
New
No
d
e.
Valid
Fro
m
←
C
u
r
r
en
tTim
estam
p
(
)
25.
R
etu
r
n
New
No
d
e
26.
E
n
d
Fu
n
ctio
n
27.
Fu
n
ctio
n
Ver
s
io
n
in
g
R
elatio
n
s
h
ip
(
Dim
en
s
io
n
R
ec
o
r
d
,
Ver
s
io
n
)
:
28.
C
r
ea
teR
elatio
n
s
h
ip
(
Dim
en
s
io
n
R
ec
o
r
d
,
Ver
s
io
n
)
29.
E
n
d
Fu
n
ctio
n
30.
Fu
n
ctio
n
Do
No
th
in
g
(
)
:
31.
// T
h
is
f
u
n
ctio
n
d
o
es n
o
t
h
in
g
,
u
s
ed
wh
en
th
e
r
ec
o
r
d
h
as n
o
t
ch
an
g
ed
32.
E
n
d
Fu
n
ctio
n
3
.
3
.
G
ener
a
t
e
f
re
qu
ent
ly
us
ed
it
em
s
et
s
T
h
e
n
ex
t
s
tag
e
in
th
is
p
r
o
ce
s
s
en
tails
r
etr
iev
in
g
u
s
er
q
u
er
ies
f
r
o
m
th
e
OL
AP
s
y
s
tem
lo
g
s
s
u
b
s
eq
u
en
t
to
th
e
co
n
s
tr
u
ctio
n
o
f
o
u
r
g
r
a
p
h
war
eh
o
u
s
e.
T
h
is
r
etr
iev
al
is
ess
en
tial
as
it
allo
w
s
u
s
to
an
aly
ze
u
s
er
b
eh
a
v
io
r
an
d
tailo
r
t
h
e
s
y
s
tem
to
b
etter
m
ee
t
th
eir
n
ee
d
s
.
Nex
t,
we
u
s
e
th
e
Ap
r
io
r
i
tech
n
iq
u
e
to
f
in
d
co
m
m
o
n
item
s
ets
o
f
p
r
e
d
icate
s
[
3
2
]
.
T
h
e
Alg
o
r
ith
m
2
:
Gen
e
r
ate
f
r
e
q
u
en
tly
u
s
ed
item
s
et
h
as
a
s
ig
n
if
ican
t
im
p
ac
t
in
o
u
r
g
r
a
p
h
c
u
b
e
m
o
d
el
f
o
r
d
eter
m
in
in
g
f
r
e
q
u
en
t
item
s
ets
with
in
OL
AP
q
u
er
ies.
I
ts
p
r
im
ar
y
o
b
jectiv
e
is
to
g
e
n
er
ate
th
ese
f
r
eq
u
en
t
item
s
ets
f
r
o
m
tr
an
s
ac
tio
n
al
d
ata
b
y
ev
alu
atin
g
ite
m
o
cc
u
r
r
en
ce
f
r
eq
u
e
n
cy
an
d
f
ilter
in
g
th
o
s
e
s
u
r
p
ass
in
g
a
m
in
im
u
m
s
u
p
p
o
r
t th
r
esh
o
ld
(
m
in
-
s
u
p
)
.
T
h
e
p
r
o
ce
s
s
co
m
p
r
i
s
es two
m
ain
s
tep
s
:
‒
I
n
itial
g
en
er
atio
n
o
f
u
n
iq
u
e
i
tem
s
et:
T
h
e
alg
o
r
it
h
m
tr
a
v
er
s
es
tr
an
s
ac
tio
n
s
to
estab
lis
h
th
e
o
cc
u
r
r
en
c
e
f
r
eq
u
e
n
cy
o
f
ea
ch
item
.
I
t th
en
f
o
r
m
s
a
s
et
o
f
f
r
eq
u
en
t item
s
ets b
ased
o
n
th
e
m
i
n
im
al
s
u
p
p
o
r
t th
r
esh
o
ld
.
‒
E
x
ten
s
io
n
o
f
item
s
ets:
Star
tin
g
f
r
o
m
f
r
eq
u
e
n
t
item
s
ets
o
f
s
ize
1
,
t
h
e
alg
o
r
ith
m
p
r
o
g
r
ess
iv
ely
g
en
e
r
ates
lar
g
er
item
s
ets.
I
t
cr
ea
tes
p
o
ten
tial
ca
n
d
id
ates
b
y
c
o
m
b
in
in
g
p
r
ev
io
u
s
ly
id
e
n
tifie
d
f
r
e
q
u
en
t
item
s
ets.
T
h
ese
n
ew
s
ets
ar
e
ass
ess
ed
with
in
tr
an
s
ac
tio
n
s
t
o
d
ete
r
m
in
e
t
h
eir
f
r
eq
u
en
cy
an
d
ar
e
r
etain
e
d
o
n
ly
if
th
ey
s
u
r
p
ass
th
e
m
in
im
al
s
u
p
p
o
r
t th
r
esh
o
ld
.
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
.
1
,
Feb
r
u
ar
y
20
25
:
810
-
8
2
1
814
Alg
o
r
ith
m
2
: G
en
e
r
ate
f
r
e
q
u
e
n
tly
u
s
ed
item
s
et
1.
f
r
eq
u
e
n
tI
tem
s
ets
←
{}
2.
p
r
ev
io
u
s
Fre
q
u
en
tI
tem
s
ets ←
Gen
er
ateI
n
itialF
r
eq
u
en
tI
tem
s
e
ts
3.
wh
ile
p
r
ev
io
u
s
Fre
q
u
en
tI
tem
s
e
ts
is
n
o
t e
m
p
ty
d
o
4.
C
u
r
r
en
tFre
q
u
en
tI
tem
s
ets←g
en
er
ateCan
d
id
ates(
p
r
ev
io
u
s
Fre
q
u
en
tI
tem
s
ets)
5.
f
o
r
ea
ch
p
ar
titi
o
n
in
g
r
ap
h
c
u
b
e
p
ar
titi
o
n
s
d
o
6.
f
o
r
ea
c
h
ca
n
d
id
ate
in
cu
r
r
en
tFre
q
u
en
tI
tem
s
ets d
o
7.
if
ca
n
d
id
ate
is
em
p
lo
y
ed
in
p
ar
titi
o
n
th
en
8.
ca
n
d
id
ate.
f
r
eq
u
en
cy
←
ca
n
d
id
ate.
f
r
e
q
u
en
c
y
+1
9.
en
d
if
10.
e
n
d
f
o
r
11.
en
d
f
o
r
12.
p
r
ev
io
u
s
Fre
q
u
en
tI
tem
s
ets ←
cu
r
r
en
tFre
q
u
en
tI
tem
s
ets
13.
cu
r
r
en
tFre
q
u
en
tI
tem
s
ets ←
f
ilter
C
an
d
id
ates (
cu
r
r
en
tFre
q
u
en
tI
tem
s
ets,
m
in
Su
p
)
14.
f
r
eq
u
e
n
tI
tem
s
ets ←
f
r
eq
u
en
tI
t
em
s
ets
∪
cu
r
r
en
tFre
q
u
e
n
tI
tem
s
ets
15.
en
d
wh
ile
16.
f
u
n
ctio
n
Gen
er
ateI
n
itialF
r
eq
u
en
tI
tem
s
ets
()
17.
in
itialF
r
eq
u
en
tI
tem
s
ets ←
{}
18.
f
o
r
ea
c
h
item
in
all
item
s
d
o
19.
if
item
.
f
r
eq
u
e
n
cy
≥
m
in
Su
p
th
en
20.
in
itialF
r
eq
u
en
tI
tem
s
ets ←
in
itialF
r
eq
u
en
tI
tem
s
ets
∪
{item
}
21.
en
d
if
22.
en
d
f
o
r
23.
r
etu
r
n
i
n
itialF
r
eq
u
en
tI
tem
s
ets
24.
en
d
f
u
n
ctio
n
25.
f
u
n
ctio
n
Ge
n
er
ateCan
d
id
ates (
item
s
ets)
26.
ca
n
d
id
ates ←
{}
27.
i ←
0
28.
wh
ile
i <
s
ize
(
item
s
ets)
d
o
29.
item
s
et
←
item
s
ets[i]
30.
j ←
0
31.
wh
ile
j <
s
ize(
item
s
et)
d
o
32.
elem
en
t ←
item
s
et[
j]
33.
ca
n
d
i
d
ate
←
item
s
et
with
o
u
t e
lem
en
t
34.
if
ca
n
d
id
ate
is
n
o
t a
l
r
ea
d
y
in
clu
d
ed
am
o
n
g
th
e
ca
n
d
id
ates th
en
35.
Ad
d
ca
n
d
id
ate
to
th
e
s
et
o
f
ca
n
d
id
ates
36.
e
n
d
if
37.
j ←
j +
1
38.
en
d
wh
ile
39.
i ←
i +
1
40.
en
d
wh
ile
41.
r
etu
r
n
ca
n
d
id
ates
42.
en
d
f
u
n
ctio
n
43.
f
u
n
ctio
n
Fil
ter
C
an
d
id
ates (
ca
n
d
id
ates,
m
in
Su
p
)
44.
f
r
eq
u
en
tC
an
d
id
ates ←
{}
45.
i ←
0
46.
wh
ile
i <
s
ize
(
ca
n
d
i
d
ates)
d
o
47.
ca
n
d
id
ate
←
ca
n
d
id
ates[
i]
48.
if
ca
n
d
id
ate.
f
r
eq
u
en
cy
≥
m
in
Su
p
th
en
49.
Ad
d
ca
n
d
id
ate
to
f
r
e
q
u
en
tC
an
d
id
ates
50.
e
n
d
if
51.
i ←
i +
1
52.
en
d
w
h
ile
53.
r
etu
r
n
f
r
e
q
u
en
tC
an
d
id
ates
54.
en
d
f
u
n
ctio
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
Meth
o
d
fo
r
d
ev
elo
p
i
n
g
a
n
d
p
a
r
titi
o
n
in
g
g
r
a
p
h
-
b
a
s
ed
d
a
ta
w
a
r
eh
o
u
s
es u
s
in
g
…
(
R
e
d
o
u
a
n
e
La
b
z
io
u
i
)
815
3
.
4
.
G
ener
a
t
e
pa
rt
it
io
ns
us
ing
a
s
s
o
cia
t
io
n r
ules
T
h
e
n
ex
t
alg
o
r
ith
m
in
o
u
r
a
p
p
r
o
ac
h
u
tili
ze
s
th
e
attr
ib
u
tes
an
d
r
elatio
n
s
h
ip
s
id
en
tifie
d
with
in
th
e
f
r
eq
u
e
n
t
item
s
ets
to
estab
li
s
h
ass
o
ciatio
n
r
u
les.
T
h
ese
r
u
les
s
er
v
e
as
th
e
b
asis
f
o
r
p
ar
titi
o
n
in
g
a
g
r
ap
h
b
ased
o
n
th
ese
attr
ib
u
tes
an
d
r
elatio
n
s
h
ip
s
.
T
h
is
p
ar
titi
o
n
in
g
allo
w
s
f
o
r
m
o
r
e
ef
f
icien
t
d
ata
r
etr
iev
al
an
d
an
aly
s
is
,
u
ltima
tely
en
h
an
cin
g
th
e
p
er
f
o
r
m
an
ce
o
f
o
u
r
g
r
a
p
h
-
b
ased
s
y
s
tem
.
T
h
e
Alg
o
r
ith
m
3
:
Gen
er
ate
p
ar
titi
o
n
s
u
s
in
g
ass
o
ciatio
n
r
u
les
b
eg
in
s
b
y
iter
atin
g
th
r
o
u
g
h
ea
c
h
f
r
e
q
u
en
t
item
s
et.
Fo
r
ea
ch
item
s
et,
it
g
en
er
ates
all
p
o
s
s
ib
le
s
u
b
s
ets
.
Su
b
s
eq
u
en
tly
,
f
o
r
ea
c
h
s
u
b
s
et,
it
co
n
s
tr
u
cts
an
ass
o
ciatio
n
r
u
le,
wh
er
e
th
e
an
t
ec
ed
en
t
co
m
p
r
is
es
th
e
s
u
b
s
et,
an
d
th
e
c
o
n
s
eq
u
en
t
co
n
s
is
ts
o
f
th
e
co
m
p
lem
e
n
t
o
f
th
e
s
u
b
s
et
with
in
th
e
f
r
eq
u
en
t
item
s
et.
T
h
is
s
y
s
tem
at
ic
p
r
o
ce
s
s
allo
ws f
o
r
th
e
cr
ea
tio
n
o
f
ass
o
cia
tio
n
r
u
les th
at
ca
p
tu
r
e
th
e
in
h
er
en
t
p
atter
n
s
a
n
d
co
r
r
elatio
n
s
with
in
th
e
d
ata
.
Alg
o
r
ith
m
3
: G
en
e
r
ate
p
ar
titi
o
n
s
u
s
in
g
ass
o
ciatio
n
r
u
les
1.
// I
n
itializatio
n
2.
m
in
Su
p
: th
e
m
in
im
u
m
s
u
p
p
o
r
t
th
r
esh
o
ld
3.
Pre
d
icate
s
I
tem
s
ets:
l
is
t o
f
f
r
eq
u
en
t p
r
e
d
icate
item
s
ets
4.
ass
o
ciatio
n
R
u
les:
li
s
t o
f
r
u
les
5.
// I
ter
ate
th
r
o
u
g
h
ea
c
h
f
r
e
q
u
en
t item
s
et
6.
f
o
r
all
f
r
eq
u
en
tI
tem
s
et
I
n
Pr
ed
icate
s
I
tem
s
ets d
o
7.
s
u
b
s
ets ←
GE
NE
R
AT
E
S
UB
SET
S(f
r
eq
u
en
tI
tem
s
et)
8.
r
u
les ←
[
]
9.
// C
r
ea
te
r
u
les b
ased
o
n
co
n
f
id
en
ce
10.
f
o
r
all
s
u
b
s
et
I
n
s
u
b
s
ets d
o
11.
an
tece
d
e
n
t ←
s
u
b
s
et
12.
co
n
s
eq
u
en
t ←
f
r
e
q
u
en
t
I
tem
s
et
-
s
u
b
s
et
13.
co
n
f
id
en
ce
←
C
AL
C
UL
AT
E
C
ONFI
DE
N
C
E
(
an
tece
d
en
t,
f
r
eq
u
en
tI
tem
s
et)
14.
if
co
n
f
id
en
ce
≥
m
in
Su
p
th
en
15.
r
u
le
←
{
an
tece
d
e
n
t: a
n
tece
d
en
t,
c
o
n
s
eq
u
en
t: c
o
n
s
eq
u
en
t
c
o
n
f
id
e
n
ce
:
co
n
f
id
en
ce
}
16.
r
u
les.ad
d
(
r
u
le)
17.
en
d
i
f
18.
en
d
f
o
r
19.
// Ge
n
er
ate
ass
o
ciatio
n
r
u
les f
o
r
th
is
f
r
eq
u
en
t item
s
et
20.
ass
o
ciatio
n
R
u
les ←
ass
o
ci
atio
n
R
u
les
∪
r
u
les
21.
en
d
f
o
r
22.
// Par
titi
o
n
g
r
ap
h
b
ased
o
n
ass
o
ciatio
n
r
u
les
23.
f
u
n
ctio
n
Par
titi
o
n
Gr
a
p
h
B
ased
On
R
u
les(g
r
ap
h
,
ass
o
ciatio
n
R
u
les)
24.
p
ar
titi
o
n
s
←
{}
25.
f
o
r
all
ass
o
ciatio
n
R
u
le
I
N
ass
o
ciatio
n
R
u
les d
o
26.
an
tece
d
e
n
tNo
d
es ←
FIN
DNOD
E
S
W
I
T
HP
R
OPE
R
T
I
E
S(g
r
ap
h
,
ass
o
ciatio
n
R
u
le.
an
tece
d
en
t)
27.
c
o
n
s
eq
u
e
n
tNo
d
es ←
FIN
DNOD
E
S
W
I
T
HPR
OPERTI
E
S(g
r
ap
h
,
ass
o
ciatio
n
R
u
le.
co
n
s
eq
u
en
t)
28.
p
ar
titi
o
n
←
C
R
E
AT
E
P
AR
T
I
T
I
ON(
an
tece
d
en
tNo
d
es,
co
n
s
eq
u
en
tNo
d
es)
29.
p
ar
titi
o
n
s
.
a
d
d
(
p
a
r
titi
o
n
)
30.
en
d
f
o
r
r
etu
r
n
p
ar
titi
o
n
s
31.
en
d
f
u
n
ctio
n
4.
M
E
T
H
O
DO
L
O
G
Y
I
M
P
L
E
M
E
N
T
A
T
I
O
N
4
.
1
.
I
m
ple
m
ent
a
t
io
n o
f
t
he
g
ra
ph
wa
re
ho
us
e
Neo
4
j
(
v
er
s
io
n
5
.
1
.
0
)
is
th
e
d
atab
ase
th
at
we
u
s
ed
t
o
cr
ea
te
o
u
r
g
r
ap
h
war
eh
o
u
s
e.
Ne
o
4
j
is
a
g
r
a
p
h
d
atab
ase
m
an
ag
em
en
t
s
y
s
tem
th
at
ef
f
icien
tly
s
to
r
es
an
d
q
u
er
y
in
g
co
m
p
lex
,
in
ter
c
o
n
n
ec
te
d
d
ata
b
y
r
ep
r
esen
tin
g
an
d
s
to
r
in
g
d
ata
i
n
g
r
a
p
h
s
tr
u
ctu
r
es
co
n
tain
in
g
n
o
d
es,
r
elati
o
n
s
h
ip
s
,
an
d
ch
ar
ac
te
r
is
tics
[
3
3
]
.
Ne
o
4
j
o
f
f
er
s
a
co
n
f
ig
u
r
ab
le
d
ata
f
o
r
m
at,
e
n
ab
les
AC
I
D
-
co
m
p
lian
t
tr
an
s
ac
tio
n
s
an
d
u
tili
ze
s
a
q
u
er
y
lan
g
u
ag
e
k
n
o
w
n
as
C
y
p
h
e
r
s
p
ec
if
ically
d
esig
n
ed
f
o
r
h
an
d
lin
g
g
r
ap
h
d
ata
[
3
4
]
.
Ad
d
itio
n
ally
,
we
em
p
lo
y
e
d
a
co
m
m
a
-
s
ep
ar
ated
v
alu
es
(
C
SV
)
f
ile
in
clu
d
in
g
d
ata
f
r
o
m
a
f
lat
m
eta
-
m
o
d
el
an
d
th
e
tr
an
s
ac
tio
n
p
r
o
ce
s
s
in
g
p
er
f
o
r
m
an
ce
co
u
n
cil
-
H
(
T
PC
-
H
)
d
atab
ase
as o
u
r
s
o
u
r
ce
f
ile
[
3
5
]
.
T
h
e
f
o
llo
win
g
a
r
e
th
e
m
o
d
el'
s
d
im
en
s
io
n
s
:
‒
Pro
d
u
ct
d
im
e
n
s
io
n
with
th
e
T
y
p
es o
f
Pro
d
u
cts lev
el.
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
.
1
,
Feb
r
u
ar
y
20
25
:
810
-
8
2
1
816
‒
C
u
s
to
m
er
d
im
en
s
io
n
with
th
e
two
lev
els R
eg
io
n
C
u
s
to
m
er
an
d
city
C
u
s
to
m
er
.
‒
Dim
en
s
io
n
o
f
th
e
Su
p
p
lier
u
s
in
g
th
e
two
le
v
els city
Su
p
p
lier
an
d
R
eg
io
n
Su
p
p
lier
.
‒
Dim
en
s
io
n
tim
e
with
th
e
lev
el
s
Yea
r
an
d
Mo
n
t
h
.
L
is
tin
g
1
'
s
s
cr
ip
t
im
p
o
r
ts
an
d
m
an
ag
es
s
u
p
p
lier
d
ata
f
r
o
m
a
C
SV
f
ile
in
to
a
Neo
4
j
g
r
ap
h
d
atab
ase.
I
t
ef
f
ec
tiv
ely
h
a
n
d
les SGC
D
d
ata
b
y
m
ain
tain
i
n
g
h
is
to
r
ical
v
e
r
s
io
n
s
an
d
en
s
u
r
in
g
d
ata
in
te
g
r
i
ty
.
‒
Data
lo
ad
in
g
: T
h
e
L
OAD
C
S
V
W
I
T
H
HE
ADE
R
S c
lau
s
e
lo
ad
s
th
e
C
SV d
ata
f
r
o
m
th
e
s
p
ec
if
ied
f
ile
in
to
a
tem
p
o
r
ar
y
v
ar
ia
b
le
n
am
e
d
r
o
w.
‒
No
d
e
m
er
g
i
n
g
: T
h
e
ME
R
GE
clau
s
e
cr
ea
tes o
r
u
p
d
ates n
o
d
es
in
th
e
g
r
a
p
h
d
atab
ase.
‒
No
d
e
cr
ea
tio
n
:
I
f
a
n
o
d
e
with
th
e
s
p
ec
if
ied
SUPP
L
I
E
R
-
I
D
d
o
es
n
o
t
ex
is
t,
th
e
ME
R
GE
clau
s
e
cr
ea
tes
a
n
ew
SUPP
L
I
E
R
n
o
d
e
with
th
e
p
r
o
v
i
d
ed
p
r
o
p
e
r
ties
.
L
is
tin
g
1
.
T
h
e
Su
p
p
lier
Dim
en
s
io
n
L
OAD
C
SV
W
I
T
H
HE
ADE
R
S
FR
OM
"f
ile:
///s
cd
.
csv
"
AS
r
o
w
FIE
L
DT
E
R
MI
NAT
OR
";"
ME
R
GE
(
S: SU
PP
L
I
E
R
{SU
PP
L
I
E
R
_
I
D:
r
o
w.
SUPP
L
I
E
R
_
I
D})
ON
C
R
E
AT
E
SET
S.
SUPP
L
I
E
R
_
NAM
E
=
r
o
w.
SUPP
L
I
E
R
_
NAM
E
,
S.
SUPP
L
I
E
R
_
C
I
T
Y
=
r
o
w.
SUPP
L
I
E
R
_
C
I
T
Y,
S.
SUPP
L
I
E
R
_
R
E
GI
ON
=
r
o
w.
SUPP
L
I
E
R
_
R
E
GI
ON,
S.
SUPP
L
I
E
R
_
C
ODE
=
r
o
w.
SUPP
L
I
E
R
_
C
ODE
,
S.
START_
DAT
E
=
r
o
w.
START_
DAT
E
,
S.
E
ND_
DAT
E
=
(
C
ASE
W
HE
N
r
o
w.
I
S_
C
UR
R
E
NT
=
'
1
'
T
HE
N
d
ate
(
'
9
9
9
9
-
12
-
31'
)
E
L
SE
d
ate
(
r
o
w.
START_
DAT
E
)
E
ND)
,
S.
I
S_
C
UR
R
E
NT
=
r
o
w.
I
S_
C
UR
R
E
NT
,
S.
VE
R
SIO
N
=
1
;
W
e
u
s
e
a
s
im
ilar
m
eth
o
d
o
lo
g
y
to
g
e
n
er
ate
th
e
d
im
en
s
io
n
s
"
C
USTO
ME
R
"
"T
im
e"
an
d
"P
R
ODU
C
T
"
ju
s
t
as
we
d
id
f
o
r
th
e
"SUPPLI
E
R
"
d
im
en
s
io
n
.
T
h
e
f
ac
t
n
o
d
e,
wh
ich
s
to
r
es
th
e
m
ea
s
u
r
es,
m
u
s
t
b
e
cr
ea
ted
af
ter
all
th
e
d
im
en
s
io
n
n
o
d
es.
L
is
tin
g
2
h
as
th
e
s
cr
ip
t
th
at
s
h
o
ws
h
o
w
to
cr
ea
te
a
f
ac
t
n
o
d
e
in
Ne
o
4
j
an
d
h
o
w
to
lin
k
a
f
ac
t
n
o
d
e
to
a
d
im
en
s
io
n
n
o
d
e.
T
h
e
s
cr
ip
t
cr
ea
tes
a
n
o
d
e
n
am
ed
"FAC
T
"
with
t
h
e
attr
ib
u
tes
"
I
D,
"
"QU
ANT
I
T
Y,
"
an
d
"Pr
ice"
th
at
ar
e
o
b
tain
ed
f
r
o
m
th
e
a
p
p
r
o
p
r
iate
co
lu
m
n
s
in
th
e
C
SV f
ile
u
s
in
g
th
e
ME
R
GE
cl
au
s
e.
T
h
e
"SUPPLI
E
R
"
n
o
d
e
with
th
e
m
atch
in
g
"SUPPLI
E
R
-
I
D"
an
d
th
e
"FAC
T
"
n
o
d
e
with
th
e
m
atch
in
g
"I
D"
p
r
o
p
e
r
ty
ar
e
lo
ca
te
d
f
o
r
ea
ch
r
o
w
b
y
th
e
s
cr
ip
t
u
s
in
g
th
e
MA
T
C
H
clau
s
es.
T
h
e
"
FAC
T
-
SU
PP
L
I
E
R
"
r
elatio
n
s
h
ip
is
estab
lis
h
ed
b
y
t
h
e
ME
R
GE
clau
s
e
b
etwe
en
th
e
m
atch
ed
"FAC
T
"
an
d
"SUPPLI
E
R
"
n
o
d
es.
L
is
tin
g
2
.
T
h
e
Fact
No
d
e
L
OAD
C
SV
W
I
T
H
HE
ADE
R
S
FR
OM
"f
ile:
///s
cd
.
csv
"
AS
r
o
w
FIE
L
DT
E
R
MI
NAT
OR
";"
ME
R
GE
(
m
ea
s
:Me
asu
r
e
{m
id
: r
o
w.
I
NT
E
GR
AT
I
ON_
I
D})
ON
C
R
E
AT
E
SET
m
ea
s
.
Pric
e
=
to
Flo
at(
r
o
w.
O_
T
OT
AL
PR
I
C
E
)
,
m
ea
s
.
QUAN
T
I
T
Y
=
to
I
n
teg
er
(
r
o
w.
L
_
QUAN
T
I
T
Y)
;
// R
elatio
n
s
h
ip
FAC
T
/SUP
PLI
E
R
L
OAD
C
SV
W
I
T
H
HE
ADE
R
S
FR
OM
"f
ile:
///s
cd
.
csv
"
AS
r
o
w
FIE
L
DT
E
R
MI
NAT
OR
";"
W
I
T
H
r
o
w
MA
T
C
H
(
S: SU
PP
L
I
E
R
{S
U
PP
L
I
E
R
_
I
D:
r
o
w.
SUPP
L
I
E
R
I
D})
MA
T
C
H
(
m
ea
s
:Me
asu
r
e
{m
id
: r
o
w.
I
NT
E
GR
AT
I
ON_
I
D})
ME
R
GE
(
m
ea
s
)
-
[
:FAC
T
_
SU
PP
L
I
E
R
{
START_
DAT
E
:r
o
w.
START_
DAT
E
,
E
ND_
DAT
E
: (
C
ASE
W
HE
N
r
o
w.
I
S_
C
UR
R
E
NT
=
'
1
'
T
HE
N
d
ate
(
'
9
9
9
9
-
12
-
31'
)
E
L
SE
d
ate
(
r
o
w.
START_
DAT
E
)
E
ND)
,
AC
T
I
VE
: (
C
ASE
W
HE
N
r
o
w.
E
ND_
DAT
E
=
d
ate
(
'
9
9
9
9
-
12
-
3
1
'
)
T
HE
N
1
E
L
SE
r
o
w.
I
S
_
C
UR
R
E
NT
E
ND)
}]
-
>(
S);
W
e
estab
li
s
h
co
n
n
ec
tio
n
s
b
etw
ee
n
th
e
f
ac
t
an
d
th
e
d
im
e
n
s
io
n
s
(
T
I
ME
,
C
USTO
ME
R
,
an
d
P
R
ODU
C
T
)
with
a
s
im
ilar
m
eth
o
d
.
T
o
e
f
f
e
ctiv
ely
im
p
o
r
t
d
ata
f
r
o
m
a
C
SV
f
ile
an
d
co
n
s
tr
u
ct
th
e
ass
o
ciatio
n
s
b
etwe
en
th
e
"FAC
T
"
n
o
d
es
an
d
its
a
s
s
o
cia
ted
n
o
d
es
in
th
e
"PR
ODU
C
T
,
"
"T
I
ME
,
"
an
d
"COS
T
UM
E
R
"
d
im
en
s
io
n
s
in
th
e
Neo
4
j d
atab
ase
.
Fig
u
r
e
1
s
tan
d
s
f
o
r
th
e
r
ea
lizatio
n
o
f
g
r
a
p
h
war
eh
o
u
s
e
u
s
in
g
Ne
o
4
j.
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
Meth
o
d
fo
r
d
ev
elo
p
i
n
g
a
n
d
p
a
r
titi
o
n
in
g
g
r
a
p
h
-
b
a
s
ed
d
a
ta
w
a
r
eh
o
u
s
es u
s
in
g
…
(
R
e
d
o
u
a
n
e
La
b
z
io
u
i
)
817
Fig
u
r
e
1
.
Ou
r
m
o
d
el
p
r
esen
ted
in
th
e
s
am
p
le
s
tu
d
y
4
.
2
.
I
m
ple
m
ent
ing
his
t
o
rica
l dim
ens
io
n m
a
na
g
em
ent
wit
h c
y
ph
er
qu
er
ie
s
T
o
im
p
lem
e
n
t
th
e
SGC
D
co
n
c
ep
t
in
Ne
o
4
j,
we
em
p
l
o
y
a
cy
p
h
er
q
u
er
y
-
b
ased
ap
p
r
o
ac
h
to
ef
f
ec
tiv
ely
m
an
ag
e
ch
an
g
in
g
d
im
e
n
s
io
n
s
.
On
e
co
m
m
o
n
s
ce
n
ar
i
o
in
d
ata
m
an
ag
em
en
t
is
th
e
n
ee
d
to
tr
ac
k
ch
an
g
es
in
en
tity
attr
ib
u
tes
o
v
er
tim
e.
I
n
th
e
co
n
t
ex
t
o
f
a
g
r
ap
h
d
ata
b
ase,
th
ese
en
titi
es
ca
n
b
e
r
ep
r
esen
ted
b
y
n
o
d
es
with
ch
an
g
in
g
p
r
o
p
er
ties
,
s
u
ch
as
cu
s
to
m
er
s
o
r
p
r
o
d
u
cts.
Fo
r
illu
s
tr
atio
n
,
c
o
n
s
id
er
t
h
e
ca
s
e
o
f
a
SUPP
L
I
E
R
d
im
en
s
io
n
in
a
g
r
ap
h
d
ata
war
eh
o
u
s
e.
W
h
en
a
SUPP
L
I
E
R
ch
an
g
es
th
eir
n
am
e
o
r
o
th
er
in
f
o
r
m
atio
n
,
it
is
ess
en
tial
to
tr
ac
k
th
ese
ch
an
g
es
wh
ile
m
ain
tain
in
g
a
h
is
to
r
y
o
f
p
r
ev
io
u
s
v
er
s
io
n
s
.
T
h
e
f
o
llo
win
g
cy
p
h
er
q
u
er
y
d
e
m
o
n
s
tr
ates
h
o
w
th
is
task
ca
n
b
e
ac
co
m
p
lis
h
ed
:
L
is
tin
g
3
.
His
to
r
ical
Dim
en
s
io
n
Ma
n
ag
em
e
n
t
L
OAD
C
SV
W
I
T
H
HE
ADE
R
S
FR
OM
"f
ile:
///s
cd
.
csv
"
AS
r
o
w
FIE
L
DT
E
R
MI
NAT
OR
";"
W
I
T
H
r
o
w
MA
T
C
H
(
S: SU
PP
L
I
E
R
{S
U
PP
L
I
E
R
_
I
D:
r
o
w.
SUPP
L
I
E
R
_
I
D})
W
HE
R
E
S.
S
UPPLI
E
R
_
NA
ME
<>
r
o
w.
SUPP
L
I
E
R
_
NAM
E
an
d
S.
SUPP
L
I
E
R
_
C
I
T
Y
<>
r
o
w.
SUPP
L
I
E
R
_
C
I
T
Y
an
d
S.
SUPP
L
I
E
R
_
R
E
GI
ON
<>
r
o
w.
SUPP
L
I
E
R
_
R
E
GI
O
N
SET
S.I
S_
C
UR
R
E
NT
=
0
,
S.E
ND_
DAT
E
=
d
atetim
e(
)
C
R
E
AT
E
(
n
ewS:
SU
PP
L
I
E
R
{
SUPP
L
I
E
R
_
I
D:
r
o
w.
SUPP
L
I
E
R
_
I
D,
SUPP
L
I
E
R
_
NAM
E
: r
o
w.
SUPP
L
I
E
R
_
NAM
E
,
SUPP
L
I
E
R
_
C
ODE
: r
o
w.
S
UPPLI
E
R
_
C
O
DE
,
START_
DAT
E
: d
atetim
e(
)
,
E
ND_
DAT
E
: d
ate
(
'
9
9
9
9
-
12
-
31'
),
I
S_
C
UR
R
E
NT
: 1
})
C
R
E
AT
E
(
S)
-
[
: V
E
R
SIO
N_
1
]
-
>(
n
ewS)
;
T
h
is
c
y
p
h
er
q
u
e
r
y
m
a
n
ag
es
c
h
an
g
in
g
d
im
en
s
io
n
s
in
o
u
r
d
a
ta
g
r
ap
h
.
I
t
h
an
d
les
s
u
p
p
lier
u
p
d
ates
b
y
m
ar
k
in
g
th
e
cu
r
r
en
t
v
e
r
s
io
n
as
o
b
s
o
lete,
cr
ea
tin
g
a
n
ew
v
er
s
io
n
with
th
e
u
p
d
ated
d
ata,
an
d
estab
lis
h
in
g
a
r
elatio
n
s
h
ip
b
etwe
en
th
e
o
ld
a
n
d
n
ew
v
er
s
io
n
s
to
e
n
ab
le
h
is
to
r
ical
tr
ac
k
in
g
.
T
h
is
ap
p
r
o
ac
h
en
s
u
r
es
th
at
ch
a
n
g
es
in
d
im
en
s
io
n
s
ar
e
ap
p
r
o
p
r
iatel
y
ac
co
u
n
ted
f
o
r
,
wh
ic
h
is
c
r
u
c
ial
f
o
r
m
ain
tain
in
g
d
ata
co
n
s
is
ten
cy
an
d
e
n
ab
lin
g
h
is
to
r
ical
an
aly
s
is
in
o
u
r
g
r
ap
h
d
ata
war
eh
o
u
s
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
.
1
,
Feb
r
u
ar
y
20
25
:
810
-
8
2
1
818
4
.
3
.
P
a
rt
it
io
nin
g
g
r
a
ph
wa
re
ho
us
e
On
ce
th
e
f
r
eq
u
e
n
tly
u
s
ed
q
u
e
r
ies
wer
e
g
en
er
ated
,
th
e
alg
o
r
ith
m
b
ased
o
n
Ap
r
io
r
i
was
em
p
lo
y
ed
to
id
en
tify
th
e
m
o
s
t
u
s
ed
item
s
ets.
Su
b
s
eq
u
en
tly
,
th
e
s
ec
o
n
d
alg
o
r
ith
m
was
ex
ec
u
ted
to
g
en
er
ate
p
r
e
v
alen
t
co
m
b
in
atio
n
s
f
o
r
p
a
r
titi
o
n
cr
ea
tio
n
.
Ou
r
p
a
r
am
eter
s
ettin
g
s
in
clu
d
ed
a
s
u
p
p
o
r
t
th
r
esh
o
ld
o
f
0
.
3
an
d
a
co
n
f
id
en
ce
lev
el
o
f
0
.
4
.
T
h
ese
co
m
b
in
atio
n
s
o
f
th
e
m
o
s
t c
o
m
m
o
n
ly
u
tili
ze
d
p
ar
titi
o
n
s
ar
e
illu
s
tr
ated
in
Fig
u
r
e
2
.
Fig
u
r
e
2
.
T
h
e
co
m
b
in
atio
n
s
f
r
eq
u
en
tly
u
tili
ze
d
f
o
r
p
ar
titi
o
n
i
n
g
T
h
e
p
ar
titi
o
n
s
ar
e
s
o
r
ted
in
d
ec
r
ea
s
in
g
o
r
d
er
o
f
s
u
p
p
o
r
t.
Su
p
p
o
r
t
is
a
m
ea
s
u
r
e
o
f
th
e
f
r
eq
u
en
cy
o
f
o
cc
u
r
r
e
n
ce
o
f
a
p
ar
titi
o
n
i
n
th
e
d
ataset.
T
h
e
m
o
s
t c
o
m
m
o
n
ly
u
s
ed
p
ar
titi
o
n
s
ar
e
as f
o
llo
ws:
‒
P1
: (
PR
ODU
C
T
{Bra
n
d
#
2
2
},
C
USTO
ME
R
)
.
‒
P2
: (
YE
AR
{2
0
1
5
},
C
USTO
ME
R
)
.
‒
P3
:(
PR
ODU
C
T
{Bra
n
d
#
2
2
},
SUPP
L
I
E
R
-
R
E
GI
ON
{AF
R
I
C
A})
.
‒
P4
: (
PR
ODU
C
T
{Bra
n
d
#
2
2
},
YE
AR
{2
0
1
5
})
.
‒
P5
:(
PR
ODU
C
T
{Bra
n
d
#
2
2
},
YE
AR
{2
0
1
5
},
C
USTO
ME
R
)
.
T
h
e
n
u
m
b
er
o
f
p
ar
titi
o
n
s
d
ef
i
n
ed
d
ep
e
n
d
s
o
n
th
e
m
in
im
u
m
s
u
p
p
o
r
t a
n
d
m
in
im
u
m
c
o
n
f
i
d
e
n
ce
s
p
ec
if
ied
b
y
th
e
u
s
er
.
A
h
ig
h
er
m
in
im
u
m
s
u
p
p
o
r
t
m
ea
n
s
th
at
p
ar
titi
o
n
s
m
u
s
t
b
e
m
o
r
e
f
r
eq
u
e
n
t
in
th
e
d
ataset
to
b
e
in
clu
d
ed
in
th
e
r
esu
lts
.
B
y
ad
ju
s
tin
g
t
h
e
v
a
lu
es
o
f
th
e
m
in
im
u
m
s
u
p
p
o
r
t
a
n
d
m
in
im
u
m
c
o
n
f
id
e
n
ce
,
th
e
u
s
er
ca
n
co
n
tr
o
l
th
e
n
u
m
b
er
o
f
p
ar
titi
o
n
s
d
ef
in
e
d
a
n
d
th
e
q
u
ality
o
f
th
e
r
esu
lts
.
5.
DIS
CU
SS
I
O
N
W
e
p
er
f
o
r
m
e
d
a
s
eq
u
e
n
ce
o
f
e
x
p
er
im
en
ts
to
v
alid
ate
o
u
r
s
tr
a
teg
y
an
d
ass
ess
th
e
im
p
ac
t
o
f
p
ar
titi
o
n
in
g
OL
AP
cu
b
e
with
in
th
e
g
r
ap
h
.
T
h
ese
ex
p
er
im
e
n
ts
in
v
o
lv
ed
ev
alu
atin
g
p
e
r
f
o
r
m
an
ce
lev
e
ls
b
ef
o
r
e
an
d
a
f
ter
im
p
lem
en
tin
g
o
u
r
p
ar
titi
o
n
ed
m
eth
o
d
.
W
e
m
ea
s
u
r
e
d
q
u
er
y
e
x
ec
u
tio
n
tim
es
f
o
r
b
o
th
th
e
o
r
i
g
in
al
cu
b
e
an
d
th
e
p
ar
titi
o
n
ed
cu
b
e.
W
e
p
er
f
o
r
m
e
d
o
u
r
test
in
g
o
n
a
m
ac
h
in
e
e
q
u
ip
p
ed
with
an
i7
p
r
o
ce
s
s
o
r
,
1
6
GB
r
an
d
o
m
ac
ce
s
s
m
em
o
r
y
(
R
AM
)
,
an
d
1
T
B
s
to
r
ag
e
ca
p
ac
ity
.
Alo
n
g
s
id
e,
we
u
tili
ze
d
th
e
T
PC
-
H
d
atab
ase
at
a
s
ca
le
f
ac
to
r
o
f
SF
1
,
SF
5
an
d
SF
1
0
wh
ich
wh
i
ch
co
r
r
esp
o
n
d
t
o
s
izes
o
f
1
GB
,
5
GB
,
an
d
1
0
GB
.
I
n
T
ab
le
1
,
we
u
s
e
th
e
q
u
e
r
ies
f
r
o
m
th
e
5
d
e
r
iv
ed
p
ar
titi
o
n
s
to
s
u
b
s
eq
u
en
tly
c
o
m
p
a
r
e
th
e
e
x
ec
u
t
io
n
tim
e.
T
h
e
Fig
u
r
e
3
illu
s
tr
ates
th
e
q
u
er
y
ex
ec
u
tio
n
tim
e
co
m
p
ar
is
o
n
,
d
is
p
lay
in
g
th
e
tim
es
b
ef
o
r
e
p
ar
titi
o
n
in
g
th
e
g
r
ap
h
wa
r
eh
o
u
s
e
an
d
af
te
r
im
p
lem
en
tin
g
t
h
e
p
ar
titi
o
n
i
n
g
ap
p
r
o
ac
h
u
s
in
g
a
s
ca
le
o
f
1
GB
.
T
h
e
f
in
d
in
g
s
r
ev
ea
l
a
d
ec
r
ea
s
e
in
q
u
er
y
e
x
ec
u
tio
n
tim
es
f
o
llo
win
g
th
e
o
p
tim
izatio
n
an
d
u
tili
za
tio
n
o
f
p
ar
titi
o
n
s
.
As
an
illu
s
tr
atio
n
,
in
th
e
i
n
itial
q
u
er
y
,
th
e
ex
ec
u
tio
n
tim
e
was
ar
o
u
n
d
3
m
s
m
illi
s
ec
o
n
d
s
in
th
e
i
n
itial
cu
b
e
OL
AP,
an
d
th
e
s
am
e
q
u
er
y
it
d
r
o
p
p
e
d
to
2
m
s
wh
en
u
s
in
g
p
ar
titi
o
n
cu
b
e.
T
h
is
s
ig
n
if
ies
a
n
o
tab
le
p
er
ce
n
tag
e
en
h
an
ce
m
en
t
o
f
ab
o
u
t
3
3
%.
Similar
ly
,
in
th
e
q
u
er
y
5
,
s
ig
n
if
ican
t
im
p
r
o
v
em
en
ts
wer
e
r
ea
lized
as
th
e
ex
ec
u
tio
n
tim
e
d
ec
r
ea
s
ed
f
r
o
m
1
4
to
1
2
m
s
.
T
h
ese
s
u
b
s
tan
tial
p
er
f
o
r
m
a
n
ce
g
ain
s
clea
r
ly
h
ig
h
lig
h
t
th
e
ef
f
icac
y
o
f
o
u
r
a
p
p
r
o
ac
h
,
r
en
d
er
in
g
th
e
s
y
s
tem
n
ea
r
ly
1
6
tim
es
f
aster
th
an
its
p
r
ev
io
u
s
s
tate.
T
h
e
Fig
u
r
e
4
illu
s
tr
ates
th
e
c
o
m
p
ar
is
o
n
o
f
q
u
er
y
ex
ec
u
tio
n
tim
es
f
o
r
th
e
s
am
e
q
u
er
y
,
d
is
p
lay
i
n
g
th
e
tim
es
b
ef
o
r
e
p
ar
titi
o
n
in
g
th
e
g
r
ap
h
war
eh
o
u
s
e
an
d
af
ter
im
p
lem
en
tin
g
th
e
p
ar
titi
o
n
in
g
ap
p
r
o
ac
h
u
s
in
g
s
c
ales o
f
5
GB
an
d
1
0
GB
r
esp
ec
tiv
ely
.
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
Meth
o
d
fo
r
d
ev
elo
p
i
n
g
a
n
d
p
a
r
titi
o
n
in
g
g
r
a
p
h
-
b
a
s
ed
d
a
ta
w
a
r
eh
o
u
s
es u
s
in
g
…
(
R
e
d
o
u
a
n
e
La
b
z
io
u
i
)
819
T
ab
le
1
.
Qu
e
r
y
u
s
ed
i
n
o
u
r
ca
s
e
s
tu
d
y
Q
u
e
r
y
D
i
me
n
si
o
n
M
e
a
su
r
e
Q1
‒
P
R
O
D
U
C
T
{
B
r
a
n
d
#
2
2
}
‒
C
U
S
TO
M
E
R
S
u
m(
P
r
i
c
e
)
Q2
‒
P
R
O
D
U
C
T{B
r
a
n
d
#
2
2
}
‒
S
U
P
P
LI
ER
‒
R
EG
I
O
N
{A
F
R
I
C
A
}
S
u
m(
P
r
i
c
e
)
Q3
‒
P
R
O
D
U
C
T{B
r
a
n
d
#
2
2
}
‒
S
U
P
P
LI
ER
‒
R
EG
I
O
N
{A
F
R
I
C
A
}
S
u
m(
P
r
i
c
e
)
Q4
‒
P
R
O
D
U
C
T{B
r
a
n
d
#
2
2
}
‒
Y
EA
R
{
2
0
1
5
}
S
u
m(
P
r
i
c
e
)
Q5
‒
P
R
O
D
U
C
T
{
B
r
a
n
d
#
2
2
}
‒
Y
EA
R
{
2
0
1
5
}
‒
C
U
S
TO
M
E
R
S
u
m(
P
r
i
c
e
)
Fig
u
r
e
3
.
Qu
e
r
y
ex
ec
u
tio
n
tim
e
f
o
r
1
GB
Fig
u
r
e
4
.
Qu
e
r
y
ex
ec
u
tio
n
tim
e
f
o
r
5
GB
an
d
1
0
GB
d
ataset
W
h
en
u
s
in
g
a
s
ca
le
o
f
5
GB
f
o
r
th
e
s
am
e
q
u
e
r
y
1
,
th
er
e
is
a
n
o
tab
le
p
er
c
en
tag
e
en
h
a
n
ce
m
en
t
o
f
a
b
o
u
t
4
6
%.
Similar
ly
,
wh
en
u
s
in
g
a
s
ca
le
o
f
1
0
GB
,
th
e
ex
ec
u
tio
n
tim
e
f
o
r
th
e
in
itial q
u
er
y
was a
r
o
u
n
d
3
0
m
s
in
th
e
o
r
ig
in
al
OL
AP
cu
b
e.
H
o
wev
er
,
with
th
e
s
am
e
q
u
e
r
y
,
it
d
r
o
p
p
ed
to
1
4
m
s
wh
en
u
s
in
g
th
e
p
ar
titi
o
n
ed
cu
b
e.
T
h
is
s
ig
n
if
ies
a
n
o
tab
le
p
er
ce
n
tag
e
en
h
an
ce
m
e
n
t
o
f
ab
o
u
t
5
3
%.
T
h
e
co
n
s
id
er
ab
le
d
ec
r
e
ase
in
ex
ec
u
tio
n
tim
e
d
em
o
n
s
tr
ates th
e
ab
ilit
y
o
f
th
is
ap
p
r
o
ac
h
to
s
u
b
s
tan
tially
im
p
r
o
v
e
th
e
s
y
s
tem
'
s
s
p
ee
d
co
m
p
ar
ed
to
its
p
r
ev
io
u
s
s
tate.
T
h
is
en
h
an
ce
m
en
t
co
n
tr
ib
u
tes
to
m
o
r
e
ef
f
icien
t
d
ata
an
aly
s
is
wi
th
in
th
e
g
r
ap
h
war
eh
o
u
s
e.
Fu
r
th
er
m
o
r
e,
th
e
ad
o
p
tio
n
o
f
p
ar
titi
o
n
in
g
in
th
e
g
r
a
p
h
p
r
o
v
es
to
b
e
a
cr
u
cial
f
ac
to
r
i
n
ac
h
iev
i
n
g
th
ese
s
ig
n
if
ican
t
im
p
r
o
v
em
e
n
ts
.
B
y
s
tr
ateg
ica
lly
o
r
g
an
izi
n
g
an
d
m
a
n
ag
in
g
d
ata
th
r
o
u
g
h
p
a
r
titi
o
n
s
,
th
e
s
y
s
tem
n
o
t
o
n
l
y
ex
p
er
ien
ce
s
a
s
u
b
s
tan
tial
b
o
o
s
t
in
s
p
ee
d
b
u
t
also
lay
s
th
e
g
r
o
u
n
d
wo
r
k
f
o
r
s
tr
ea
m
lin
ed
an
d
o
p
tim
ize
d
d
ata
p
r
o
ce
s
s
in
g
.
Evaluation Warning : The document was created with Spire.PDF for Python.