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.
1
5
,
No
.
4
,
A
u
g
u
s
t 2
0
2
6
,
p
p
.
3
4
0
2
~
3
4
1
0
I
SS
N:
2
2
5
2
-
8
9
3
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijai.v
15
.i
4
.
p
p
3
4
0
2
-
3
4
1
0
3402
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
An empiri
ca
l com
pa
riso
n of cluste
r
ing
appro
a
ches fo
r re
cen
cy
,
frequen
cy
,
a
nd m
o
netary
custom
er
seg
menta
tion
Upek
k
ha
L
a
u,
M
edit
y
a
Wa
s
esa
S
c
h
o
o
l
o
f
B
u
s
i
n
e
ss
a
n
d
M
a
n
a
g
e
me
n
t
,
I
n
st
i
t
u
t
Te
k
n
o
l
o
g
i
B
a
n
d
u
n
g
,
B
a
n
d
u
n
g
,
I
n
d
o
n
e
si
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Ma
y
1
5
,
2
0
2
5
R
ev
is
ed
Ma
y
2
5
,
2
0
2
6
Acc
ep
ted
J
u
n
1
9
,
2
0
2
6
Th
is
stu
d
y
e
v
a
lu
a
tes
e
ffe
c
ti
v
e
n
e
s
s
o
f
th
re
e
c
lu
ste
rin
g
tec
h
n
i
q
u
e
s
—
k
-
m
e
a
n
s,
h
iera
rc
h
ica
l
c
lu
ste
rin
g
,
a
n
d
d
e
n
sity
-
b
a
se
d
sp
a
ti
a
l
c
lu
ste
ri
n
g
o
f
a
p
p
li
c
a
ti
o
n
s
with
n
o
ise
(DBSCAN)
—
a
p
p
li
e
d
to
th
e
re
c
e
n
c
y
-
fre
q
u
e
n
c
y
-
m
o
n
e
tary
(RF
M
)
m
o
d
e
l
fo
r
c
u
st
o
m
e
r
se
g
m
e
n
tatio
n
in
th
e
re
tail
se
c
to
r.
Us
in
g
sa
les
tran
sa
c
ti
o
n
d
a
ta
fro
m
a
d
istr
ib
u
to
r
o
f
c
o
m
p
u
ter
a
c
c
e
ss
o
ries
a
n
d
p
rin
t
in
g
p
r
o
d
u
c
ts.
Th
e
re
su
lt
s
sh
o
w
th
a
t
k
-
m
e
a
n
s
a
c
h
iev
e
d
th
e
b
e
st
c
l
u
ste
rin
g
v
a
li
d
a
ti
o
n
s
c
o
re
s
a
n
d
e
ffe
c
ti
v
e
ly
id
e
n
ti
fied
h
i
g
h
-
v
a
lu
e
c
u
sto
m
e
rs,
h
iera
rc
h
ica
l
c
lu
ste
ri
n
g
g
e
n
e
ra
te
d
les
s
m
e
a
n
in
g
fu
l
g
r
o
u
p
i
n
g
s
t
h
a
n
k
-
m
e
a
n
s,
a
n
d
DBSCAN
m
isc
las
sified
k
e
y
c
u
sto
m
e
rs
a
s
n
o
ise
.
Th
e
se
fin
d
in
g
s
h
i
g
h
li
g
h
t
k
-
m
e
a
n
s
a
s
th
e
m
o
st
su
it
a
b
l
e
tec
h
n
iq
u
e
fo
r
RF
M
-
b
a
se
d
se
g
m
e
n
tatio
n
in
t
h
is
re
tail
b
u
sin
e
ss
c
o
n
tex
t.
T
h
e
stu
d
y
o
ffe
rs
p
ra
c
ti
c
a
l
i
n
sig
h
ts
fo
r
re
tail
a
n
d
d
istri
b
u
ti
o
n
b
u
si
n
e
ss
e
s
a
imin
g
t
o
a
d
o
p
t
d
a
ta
-
d
ri
v
e
n
c
u
sto
m
e
r
str
a
teg
ies
a
n
d
su
g
g
e
sts
fu
tu
re
re
se
a
rc
h
to
e
n
h
a
n
c
e
se
g
m
e
n
tatio
n
ro
b
u
st
n
e
ss
a
n
d
re
fin
e
t
h
e
RF
M
fra
m
e
wo
rk
.
K
ey
w
o
r
d
s
:
B
u
s
in
ess
an
aly
tics
C
u
s
to
m
er
s
eg
m
en
tatio
n
DB
S
C
AN
clu
s
ter
in
g
Hier
ar
ch
ical
clu
s
ter
in
g
K
-
m
ea
n
s
clu
s
ter
in
g
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Up
ek
k
h
a
L
a
u
Sch
o
o
l o
f
B
u
s
in
ess
an
d
Ma
n
a
g
em
en
t,
I
n
s
titu
t T
ek
n
o
lo
g
i Ban
d
u
n
g
B
an
d
u
n
g
,
I
n
d
o
n
esia
E
m
ail:
u
p
k
lau
2
1
2
1
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
C
o
m
p
etin
g
o
n
an
aly
tics
h
as
b
ec
o
m
e
an
in
cr
ea
s
in
g
ly
p
o
p
u
lar
s
tr
ateg
y
f
o
r
b
u
s
in
ess
es
th
at
g
en
er
ate
lar
g
e
v
o
lu
m
es
o
f
tr
an
s
ac
tio
n
al
d
ata
[
1
]
.
T
h
is
s
tu
d
y
ex
p
l
o
r
es
cu
s
to
m
er
s
eg
m
en
tatio
n
in
th
e
d
is
tr
ib
u
tio
n
b
u
s
in
ess
u
s
in
g
th
e
r
ec
en
cy
-
f
r
eq
u
en
cy
-
m
o
n
eta
r
y
(
R
FM)
m
o
d
el
co
m
b
i
n
ed
with
clu
s
ter
in
g
an
aly
s
is
.
C
u
s
to
m
er
s
eg
m
en
tatio
n
is
a
m
ar
k
etin
g
s
tr
ateg
y
t
h
at
g
r
o
u
p
s
cu
s
to
m
er
w
ith
s
im
ilar
ch
ar
ac
te
r
is
tics
in
to
th
e
s
am
e
ca
teg
o
r
y
,
wh
ich
th
ey
r
esp
o
n
d
to
m
ar
k
etin
g
ef
f
o
r
ts
in
a
s
im
ilar
way
[
2
]
.
C
lu
s
ter
in
g
is
a
m
ac
h
in
e
lear
n
in
g
tech
n
iq
u
e
th
at
g
r
o
u
p
s
a
s
et
o
f
p
h
y
s
ical
o
r
ab
s
tr
ac
t
o
b
jects
in
to
clu
s
ter
s
b
ased
o
n
s
im
ilar
ity
[
3
]
.
I
t
is
p
ar
ticu
lar
ly
u
s
ef
u
l
f
o
r
id
en
tify
in
g
p
atter
n
s
in
u
n
lab
eled
d
ata.
I
n
th
e
co
n
tex
t
o
f
cu
s
to
m
er
s
eg
m
en
tatio
n
,
th
e
"o
b
jects"
r
ep
r
esen
t
cu
s
to
m
er
s
,
an
d
s
im
ilar
ity
is
ty
p
ically
m
ea
s
u
r
ed
b
ased
o
n
p
u
r
c
h
asin
g
b
eh
a
v
io
r
,
d
e
m
o
g
r
ap
h
ics,
o
r
o
th
er
r
elev
an
t
attr
ib
u
tes.
B
y
lev
e
r
ag
in
g
clu
s
ter
in
g
tec
h
n
iq
u
es,
b
u
s
in
ess
es
ca
n
d
esig
n
tar
g
eted
m
a
r
k
etin
g
ca
m
p
aig
n
s
,
p
r
o
v
id
e
p
er
s
o
n
alize
d
r
ec
o
m
m
en
d
atio
n
s
,
a
n
d
i
m
p
lem
en
t
tailo
r
ed
c
u
s
to
m
er
s
er
v
ice
s
tr
ateg
ies
f
o
r
ea
ch
id
en
tifie
d
cu
s
to
m
er
s
eg
m
en
t.
T
h
e
R
FM
m
o
d
el
is
a
p
o
p
u
lar
ap
p
r
o
ac
h
f
o
r
cu
s
to
m
e
r
s
eg
m
en
tatio
n
.
I
t
ev
alu
ates
c
u
s
to
m
e
r
b
eh
a
v
io
r
b
ased
o
n
th
r
ee
k
ey
d
im
e
n
s
io
n
s
:
r
ec
en
cy
,
f
r
eq
u
e
n
cy
,
an
d
m
o
n
etar
y
v
alu
e
d
er
i
v
ed
f
r
o
m
s
ales
d
ata
[
4
]
,
[
5
]
.
B
y
clu
s
ter
in
g
th
ese
R
FM
v
ar
iab
le
s
,
b
u
s
in
ess
es
ca
n
id
en
tify
d
is
t
in
ct
cu
s
to
m
er
g
r
o
u
p
s
b
ased
o
n
th
eir
en
g
a
g
em
en
t
p
atter
n
s
.
T
h
is
s
tu
d
y
ap
p
lies
th
r
ee
clu
s
ter
in
g
tech
n
iq
u
es
—
k
-
m
ea
n
s
,
h
ier
ar
ch
ical
clu
s
ter
in
g
,
an
d
d
en
s
ity
-
b
ased
s
p
at
ial
cl
u
s
t
er
i
n
g
o
f
a
p
p
lic
ati
o
n
s
wi
th
n
o
is
e
(
DB
SC
AN)
—
to
d
et
e
r
m
i
n
e
t
h
e
m
o
s
t
s
u
ita
b
le
m
eth
o
d
f
o
r
an
aly
zin
g
m
aster
d
ea
ler
tr
an
s
ac
tio
n
al
d
ata.
T
h
e
k
-
m
ea
n
s
alg
o
r
ith
m
p
ar
titi
o
n
s
d
ata
in
to
n
clu
s
ter
s
b
y
m
in
im
izin
g
in
tr
a
-
clu
s
ter
v
ar
ian
ce
,
also
k
n
o
wn
as
in
er
tia
o
r
with
in
-
clu
s
ter
s
u
m
o
f
s
q
u
ar
es
(
W
C
S
S)
[
6
]
.
Hier
ar
ch
ical
clu
s
ter
in
g
,
o
n
th
e
o
th
er
h
a
n
d
,
cr
ea
tes
a
h
ier
ar
ch
y
o
f
clu
s
ter
s
wi
th
o
u
t
r
eq
u
ir
in
g
a
p
r
e
d
ef
in
ed
n
u
m
b
er
o
f
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
A
n
emp
ir
ica
l c
o
mp
a
r
is
o
n
o
f c
l
u
s
teri
n
g
a
p
p
r
o
a
c
h
es fo
r
r
ec
en
cy
,
fr
eq
u
en
cy
,
a
n
d
mo
n
eta
r
y
…
(
Up
ek
kh
a
La
u
)
3403
clu
s
ter
s
[
4
]
.
DB
SC
AN
g
r
o
u
p
s
d
ata
p
o
in
ts
b
ased
o
n
d
en
s
ity
,
allo
win
g
it
to
d
etec
t
clu
s
ter
s
o
f
ar
b
itra
r
y
s
h
ap
e
an
d
id
en
tif
y
n
o
is
e
o
r
o
u
tlier
s
,
with
o
u
t n
ee
d
in
g
to
s
p
ec
if
y
th
e
n
u
m
b
e
r
o
f
clu
s
ter
s
in
ad
v
an
ce
[
7
]
.
T
h
is
r
esear
ch
ad
d
r
ess
es
th
e
p
r
o
b
lem
o
f
d
eter
m
in
in
g
th
e
m
o
s
t
s
u
itab
le
clu
s
ter
in
g
m
eth
o
d
f
o
r
cu
s
to
m
er
s
eg
m
en
tatio
n
in
a
r
e
tail
d
is
tr
ib
u
tio
n
b
u
s
in
ess
,
wh
er
e
u
n
d
er
s
tan
d
i
n
g
cu
s
to
m
er
p
at
ter
n
s
is
cr
u
cial
f
o
r
s
tr
ateg
ic
d
ec
is
io
n
-
m
ak
in
g
.
T
h
e
s
tu
d
y
c
o
m
p
ar
es
th
r
ee
wid
ely
u
s
ed
clu
s
ter
in
g
tech
n
iq
u
es
—
k
-
m
ea
n
s
,
h
ier
ar
ch
ical
,
an
d
DB
SC
AN
—
t
o
ev
alu
ate
th
eir
e
f
f
ec
tiv
en
ess
in
s
eg
m
en
tin
g
c
u
s
to
m
er
s
b
ased
o
n
R
FM
an
aly
s
is
.
T
o
en
s
u
r
e
th
e
r
eliab
ilit
y
an
d
v
alid
ity
o
f
th
e
cl
u
s
ter
in
g
r
es
u
lts
,
th
e
s
tu
d
y
em
p
lo
y
s
th
r
ee
in
ter
n
al
v
alid
atio
n
m
etr
ics:
s
ilh
o
u
ette
s
co
r
e
,
Da
v
ies
-
B
o
u
ld
in
in
d
ex
(
DB
I
)
,
an
d
C
alin
s
k
i
-
Har
ab
asz
in
d
ex
(
C
HI
)
.
B
ased
o
n
th
e
ev
alu
atio
n
o
u
tco
m
es,
th
e
r
ese
ar
ch
n
o
t o
n
ly
id
e
n
tifie
s
th
e
b
e
s
t
-
p
er
f
o
r
m
in
g
clu
s
ter
in
g
m
et
h
o
d
b
u
t
also
p
r
o
v
id
es
an
in
ter
p
r
etatio
n
o
f
t
h
e
r
esu
lti
n
g
cu
s
to
m
er
s
eg
m
en
ts
to
o
f
f
er
p
r
ac
tical
in
s
ig
h
ts
f
o
r
th
e
b
u
s
i
n
ess
.
T
ab
le
1
s
u
m
m
ar
izes
p
r
io
r
s
tu
d
ies
o
n
cu
s
to
m
er
s
eg
m
en
t
atio
n
em
p
lo
y
in
g
clu
s
ter
in
g
t
ec
h
n
iq
u
es.
Am
o
n
g
th
e
ten
r
e
v
iewe
d
s
tu
d
i
es,
s
ev
en
ap
p
lied
k
-
m
ea
n
s
clu
s
ter
in
g
,
f
o
u
r
u
s
ed
h
ier
ar
c
h
ical
clu
s
ter
in
g
,
an
d
two
em
p
lo
y
ed
DB
SC
AN.
Alth
o
u
g
h
k
-
m
ea
n
s
ap
p
ea
r
s
to
b
e
th
e
m
o
s
t
f
r
eq
u
en
tly
u
s
ed
m
et
h
o
d
,
its
p
r
ev
alen
ce
d
o
es
n
o
t
im
p
l
y
m
eth
o
d
o
l
o
g
ical
s
u
p
er
io
r
ity
.
R
ath
er
,
k
-
m
ea
n
s
is
o
f
ten
ad
o
p
ted
d
u
e
to
its
s
im
p
licity
,
s
ca
lab
ilit
y
,
an
d
g
en
er
al
ap
p
licab
ilit
y
ac
r
o
s
s
d
i
v
er
s
e
s
eg
m
en
tatio
n
co
n
tex
ts
.
T
h
e
au
th
o
r
s
[
8
]
,
[
9
]
b
o
th
ap
p
lied
th
e
R
FM
m
o
d
el
with
k
-
m
ea
n
s
clu
s
ter
in
g
to
s
eg
m
en
t
cu
s
to
m
er
s
b
ased
o
n
p
u
r
ch
asin
g
b
e
h
av
io
r
,
aim
in
g
to
en
h
a
n
ce
d
ata
-
d
r
iv
en
m
ar
k
etin
g
an
d
id
en
tif
y
h
i
g
h
-
v
alu
e
cu
s
to
m
er
s
.
Ak
an
d
e
et
a
l
.
[
9
]
u
s
ed
o
n
lin
e
r
etail
d
ata
f
r
o
m
th
e
UC
I
Ma
ch
in
e
L
ea
r
n
i
n
g
R
ep
o
s
ito
r
y
an
d
id
en
tifie
d
f
iv
e
cu
s
to
m
er
ca
teg
o
r
ies
—
p
latin
u
m
,
g
o
ld
,
s
ilv
er
,
b
r
o
n
ze
,
an
d
b
ad
—
u
s
in
g
th
e
e
lb
o
w
m
eth
o
d
to
d
eter
m
in
e
th
e
o
p
tim
al
n
u
m
b
er
o
f
clu
s
ter
s
.
Similar
ly
,
C
h
r
is
ty
et
a
l.
[
1
0
]
u
tili
ze
d
R
FM
wi
th
k
-
m
ea
n
s
,
f
u
zz
y
c
-
m
ea
n
s
,
an
d
r
ep
etitiv
e
m
ed
ia
n
b
ased
k
-
m
e
an
s
(
R
M
k
-
m
ea
n
s
)
clu
s
ter
in
g
,
f
in
d
in
g
R
M
k
-
m
ea
n
s
to
b
e
th
e
m
o
s
t
ef
f
ec
tiv
e
in
ter
m
s
o
f
co
n
v
er
g
en
ce
a
n
d
clu
s
ter
co
m
p
ac
tn
ess
,
r
ein
f
o
r
cin
g
th
e
v
alu
e
o
f
R
FM
-
b
ased
s
eg
m
en
tatio
n
in
cu
s
to
m
er
r
elatio
n
s
h
ip
m
an
ag
e
m
en
t.
Me
an
wh
ile,
Su
tr
am
ian
i
et
a
l.
[
1
1
]
co
m
p
ar
e
d
k
-
m
ea
n
s
an
d
DB
S
C
AN
f
o
r
clu
s
ter
in
g
m
icr
o
,
s
m
all
,
an
d
m
ed
iu
m
e
n
ter
p
r
is
es
(
MSM
E
s
)
b
ased
o
n
ass
et
v
alu
e
an
d
t
u
r
n
o
v
er
.
K
-
m
ea
n
s
o
u
tp
er
f
o
r
m
ed
DB
SC
AN
in
cl
u
s
ter
co
m
p
ac
tn
ess
(
DB
I
)
,
th
o
u
g
h
DB
SC
AN
was
s
u
p
er
io
r
in
id
en
tify
i
n
g
o
u
tlier
s
an
d
d
id
n
o
t
r
e
q
u
ir
e
a
p
r
e
d
e
f
in
ed
n
u
m
b
er
o
f
clu
s
ter
s
.
T
h
is
s
tu
d
y
co
n
tr
ib
u
tes
n
o
v
elt
y
b
y
s
y
s
tem
atica
lly
co
m
p
ar
in
g
k
-
m
ea
n
s
,
h
ier
ar
c
h
ical
clu
s
ter
in
g
,
an
d
DB
SC
AN
u
s
in
g
m
u
ltip
le
in
ter
n
al
v
alid
atio
n
m
etr
ics
(
s
ilh
o
u
ette
,
DB
I
,
an
d
C
HI
)
with
in
a
r
ea
l
-
wo
r
ld
d
is
tr
ib
u
ti
o
n
b
u
s
in
ess
co
n
tex
t
to
id
en
ti
f
y
th
e
m
o
s
t
s
u
itab
le
m
eth
o
d
f
o
r
ac
tio
n
ab
le
cu
s
to
m
er
s
eg
m
en
tatio
n
.
T
ab
le
1
.
Pre
v
i
o
u
s
s
tu
d
ies o
n
c
u
s
to
m
er
s
eg
m
en
tatio
n
cl
u
s
ter
in
g
an
aly
s
is
A
r
t
i
c
l
e
B
u
s
i
n
e
ss
c
o
n
t
e
x
t
F
e
a
t
u
r
e
s
C
l
u
st
e
r
i
n
g
me
t
h
o
d
V
a
l
i
d
a
t
i
o
n
me
t
h
o
d
[
4
]
R
e
t
a
i
l
R
F
M
H
i
e
r
a
r
c
h
i
c
a
l
D
B
I
a
n
d
si
l
h
o
u
e
t
t
e
[
1
2
]
R
e
t
a
i
l
M
a
r
k
e
t
b
a
s
k
e
t
K
-
me
a
n
s
S
u
m
o
f
s
q
u
a
r
e
d
e
r
r
o
r
(
S
S
E)
,
si
l
h
o
u
e
t
t
e
,
a
n
d
g
a
p
st
a
t
i
st
i
c
[
8
]
O
n
l
i
n
e
r
e
t
a
i
l
R
F
M
K
-
me
a
n
s
S
i
l
h
o
u
e
t
t
e
[
1
0
]
O
n
l
i
n
e
r
e
t
a
i
l
R
F
M
K
-
me
a
n
s
,
R
M
k
-
m
e
a
n
s,
a
n
d
f
u
z
z
y
c
-
mea
n
s
S
i
l
h
o
u
e
t
t
e
[
1
3
]
O
n
l
i
n
e
r
e
t
a
i
l
R
F
M
i
n
t
e
r
p
u
r
c
h
a
se
t
i
me
H
i
e
r
a
r
c
h
i
c
a
l
C
H
I
a
n
d
D
B
I
[
1
4
]
I
n
t
e
r
n
e
t
b
a
n
k
i
n
g
R
F
M
K
-
me
a
n
s
a
n
d
D
B
S
C
A
N
D
B
I
[
1
5
]
R
e
t
a
i
l
R
F
M
(
t
i
m
e
)
H
i
e
r
a
r
c
h
i
c
a
l
a
n
d
k
-
sh
a
p
e
S
i
l
h
o
u
e
t
t
e
a
n
d
C
H
I
[
1
6
]
O
n
l
i
n
e
r
e
t
a
i
l
R
d
F
d
M
d
K
-
me
a
n
s
D
B
I
[
1
7
]
R
e
t
a
i
l
R
F
M
D
e
n
si
t
y
p
e
a
k
(
DP
)
-
g
e
n
e
t
i
c
a
l
g
o
r
i
t
h
m (G
A
)
-
p
o
ssi
b
i
l
i
s
t
i
c
f
u
z
z
y
c
-
mea
n
s
(
P
F
C
M
)
a
n
d
k
-
mea
n
s
A
R
I
,
N
M
I
,
a
n
d
D
u
n
n
-
B
o
n
f
e
r
r
o
n
i
[
1
8
]
D
i
st
r
i
b
u
t
i
o
n
a
n
d
r
e
t
a
i
l
R
F
M
F
u
z
z
y
c
-
me
a
n
s
P
a
r
t
i
t
i
o
n
c
o
e
f
f
i
c
i
e
n
t
Th
i
s
s
t
u
d
y
D
i
st
r
i
b
u
t
i
o
n
R
F
M
K
-
me
a
n
s
,
h
i
e
r
a
r
c
h
i
c
a
l
,
a
n
d
D
B
S
C
A
N
S
i
l
h
o
u
e
t
t
e
,
D
B
I
,
a
n
d
C
H
I
2.
M
E
T
H
O
D
Fig
u
r
e
1
o
u
tlin
e
th
e
r
esear
c
h
f
r
am
ewo
r
k
o
f
th
is
s
tu
d
y
.
I
t
is
ad
ap
ted
f
r
o
m
p
r
e
v
io
u
s
r
esear
ch
[
1
9
]
.
T
h
e
f
r
am
ewo
r
k
c
o
n
s
is
ts
o
f
f
o
u
r
s
tag
es:
tr
an
s
ac
tio
n
al
d
ata
co
llectio
n
,
d
ata
p
r
e
p
r
o
ce
s
s
in
g
an
d
p
r
ep
ar
atio
n
,
clu
s
ter
in
g
an
d
v
alid
atio
n
,
an
d
lab
ellin
g
an
d
an
al
y
s
is
.
I
n
th
is
s
tu
d
y
,
th
e
tr
an
s
ac
tio
n
al
d
ataset
was
alr
ea
d
y
in
a
clea
n
an
d
s
tr
u
ctu
r
ed
f
o
r
m
at,
w
ith
n
o
m
is
s
in
g
v
alu
es o
r
in
co
n
s
is
ten
cies id
en
tifie
d
; th
er
ef
o
r
e
,
n
o
ad
d
itio
n
al
d
ata
clea
n
in
g
was
r
eq
u
ir
e
d
.
Pre
p
r
o
ce
s
s
in
g
was
co
n
d
u
cted
b
y
tr
a
n
s
f
o
r
m
in
g
th
e
r
aw
d
ata
in
t
o
r
ec
en
cy
,
f
r
eq
u
en
cy
,
an
d
m
o
n
etar
y
v
ar
iab
les.
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.
1
5
,
No
.
4
,
Au
g
u
s
t 2
0
2
6
:
3
4
0
2
-
3
4
1
0
3404
Fig
u
r
e
1
.
T
h
e
r
esear
ch
f
r
am
e
wo
r
k
2
.
1
.
Clus
t
er
ing
a
nd
v
a
lid
a
t
i
o
n
2
.
1
.
1
.
Clus
t
er
ing
T
h
is
s
tu
d
y
em
p
lo
y
s
th
r
ee
wid
ely
u
s
ed
clu
s
ter
in
g
t
ec
h
n
iq
u
es
—
k
-
m
ea
n
s
,
h
ier
ar
c
h
ical
,
an
d
DB
S
C
AN
—
s
elec
ted
f
o
r
th
eir
d
is
tin
ct
u
n
d
er
ly
in
g
p
r
in
cip
les:
ce
n
tr
o
id
-
b
ased
(
k
-
m
ea
n
s
)
,
h
ier
ar
ch
ical
(
ag
g
lo
m
e
r
ativ
e)
,
an
d
d
e
n
s
ity
-
b
ased
(
DB
SC
AN)
.
i)
K
-
m
ea
n
s
clu
s
ter
in
g
:
t
h
e
k
-
m
ea
n
s
alg
o
r
ith
m
p
ar
titi
o
n
s
d
at
a
in
to
k
clu
s
ter
s
b
y
m
in
im
izin
g
th
e
s
p
r
ea
d
with
in
ea
ch
cl
u
s
ter
[
2
0
]
.
I
t
s
ta
r
ts
b
y
r
an
d
o
m
ly
s
elec
tin
g
k
ce
n
tr
o
id
s
an
d
ass
ig
n
in
g
ea
ch
d
at
a
p
o
in
t
to
t
h
e
n
ea
r
est
o
n
e.
T
h
e
n
,
it
iter
ativ
ely
u
p
d
ates
ce
n
tr
o
i
d
s
b
y
ca
l
cu
latin
g
th
e
m
ea
n
o
f
ass
ig
n
ed
p
o
in
ts
an
d
r
ea
s
s
ig
n
s
p
o
in
ts
u
n
til ce
n
tr
o
id
m
o
v
em
en
t
f
alls
u
n
d
er
a
th
r
esh
o
ld
,
in
d
icatin
g
clu
s
ter
s
tab
ilit
y
.
=
∑
∑
‖
−
‖
2
∈
=
1
(
1
)
W
h
er
e
is
n
u
m
b
e
r
o
f
cl
u
s
ter
s
;
is
s
et
o
f
p
o
in
ts
in
clu
s
ter
;
is
d
ata
p
o
in
t
in
clu
s
ter
;
is
th
e
ce
n
tr
o
id
o
f
clu
s
ter
; a
n
d
‖
−
‖
2
is
s
q
u
ar
ed
E
u
clid
ea
n
d
is
tan
ce
b
etwe
en
a
d
ata
p
o
in
t a
n
d
its
ce
n
tr
o
i
d
.
ii)
Hier
ar
ch
ical
c
lu
s
ter
in
g
:
a
g
g
lo
m
er
ativ
e
(
h
ier
ar
c
h
ical)
clu
s
ter
in
g
u
s
es
a
b
o
tto
m
-
u
p
ap
p
r
o
ac
h
,
m
er
g
in
g
in
d
iv
id
u
al
d
ata
p
o
in
ts
in
to
clu
s
ter
s
s
u
cc
ess
iv
ely
to
f
o
r
m
a
tr
ee
-
lik
e
s
tr
u
ctu
r
e
ca
lled
a
d
en
d
r
o
g
r
am
[
2
0
]
.
Un
lik
e
k
-
m
ea
n
s
,
it
d
o
esn
'
t
r
eq
u
ir
e
s
p
ec
if
y
in
g
th
e
n
u
m
b
er
o
f
clu
s
ter
s
in
ad
v
an
ce
.
T
h
e
p
r
o
ce
s
s
s
tar
ts
with
ea
ch
p
o
i
n
t
as
its
o
wn
clu
s
ter
,
m
er
g
in
g
th
e
clo
s
est
p
air
s
u
n
til
all
p
o
in
ts
f
o
r
m
a
s
in
g
le
clu
s
t
er
.
T
h
is
p
a
p
er
u
s
es War
d
’
s
m
eth
o
d
,
wh
ich
m
in
im
izes th
e
s
u
m
o
f
s
q
u
a
r
ed
d
i
f
f
er
en
ce
s
with
in
clu
s
ter
s
.
(
,
)
=
|
|
|
|
+
|
|
‖
−
‖
2
(
2
)
W
h
er
e
(
,
)
is
d
is
tan
ce
b
etwe
en
clu
s
ter
an
d
;
|
|
an
d
|
|
is
n
u
m
b
e
r
o
f
p
o
in
ts
i
n
clu
s
ter
an
d
;
an
d
is
ce
n
tr
o
id
o
f
clu
s
ter
an
d
;
an
d
‖
−
‖
2
is
s
q
u
ar
ed
E
u
cli
d
ea
n
d
is
tan
ce
b
etwe
en
th
e
ce
n
tr
o
id
s
.
iii)
DB
S
C
AN
:
DB
S
C
AN
g
r
o
u
p
s
d
en
s
e
r
eg
io
n
s
o
f
d
ata
wh
ile
tr
ea
tin
g
s
p
ar
s
e
ar
ea
s
as
n
o
is
e
[
1
1
]
.
Un
lik
e
k
-
m
ea
n
s
,
it
ca
n
d
etec
t
clu
s
ter
s
o
f
ar
b
itra
r
y
s
h
ap
es
with
o
u
t
r
eq
u
ir
in
g
t
h
e
n
u
m
b
er
o
f
clu
s
te
r
s
in
ad
v
an
ce
.
I
t
s
tar
ts
with
an
u
n
v
is
ited
p
o
i
n
t
an
d
ch
ec
k
s
h
o
w
m
an
y
n
eig
h
b
o
r
s
lie
with
in
a
s
p
ec
if
ied
d
is
tan
ce
(
ep
s
)
.
I
f
th
is
n
u
m
b
er
m
ee
ts
o
r
ex
ce
e
d
s
m
in
_
s
am
p
les,
th
e
p
o
in
t
b
ec
o
m
es
a
co
r
e
s
am
p
le
an
d
in
itiates
a
clu
s
ter
.
T
h
e
clu
s
ter
ex
p
an
d
s
b
y
r
ec
u
r
s
iv
ely
in
clu
d
i
n
g
n
eig
h
b
o
r
i
n
g
co
r
e
an
d
b
o
r
d
er
p
o
in
ts
.
T
h
is
p
r
o
ce
s
s
co
n
tin
u
es u
n
til all
p
o
in
ts
ar
e
c
lass
if
ied
.
(
)
=
{
∈
|
(
,
)
≤
}
(
3
)
W
h
er
e
(
)
is
n
eig
h
b
o
r
h
o
o
d
o
f
p
o
in
t
with
in
r
ad
i
u
s
;
(
,
)
is
d
is
tan
ce
b
etwe
en
p
o
in
ts
an
d
;
is
n
eig
h
b
o
r
h
o
o
d
r
a
d
iu
s
p
ar
am
eter
; a
n
d
A
p
o
in
t p
is
co
r
e
p
o
i
n
t if
|
(
)
|
≥
.
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
A
n
emp
ir
ica
l c
o
mp
a
r
is
o
n
o
f c
l
u
s
teri
n
g
a
p
p
r
o
a
c
h
es fo
r
r
ec
en
cy
,
fr
eq
u
en
cy
,
a
n
d
mo
n
eta
r
y
…
(
Up
ek
kh
a
La
u
)
3405
2
.
1
.
2
.
Va
lid
a
t
io
n
Fo
r
v
alid
atio
n
,
th
is
s
tu
d
y
u
tili
ze
s
in
tr
o
d
u
ce
s
:
s
ilh
o
u
ette
s
co
r
e,
DB
I
,
an
d
C
HI
.
As
s
h
o
wn
i
n
T
ab
le
1
,
th
ese
m
etr
ics
wer
e
s
elec
ted
f
o
r
th
eir
wid
esp
r
ea
d
u
s
e
in
cu
s
to
m
er
s
eg
m
en
tatio
n
a
n
d
t
h
eir
av
ailab
ilit
y
in
t
h
e
s
cik
it
-
lear
n
lib
r
ar
y
,
en
s
u
r
i
n
g
e
asy
r
ep
r
o
d
u
cib
ilit
y
.
Fu
r
th
er
m
o
r
e,
p
r
io
r
r
esear
ch
id
en
tifie
s
s
ilh
o
u
ette
,
DB
I
,
an
d
C
HI
as a
m
o
n
g
th
e
to
p
-
p
er
f
o
r
m
in
g
in
d
ices w
ith
in
a
g
r
o
u
p
o
f
r
ec
o
m
m
e
n
d
ed
clu
s
ter
in
g
v
ali
d
atio
n
m
etr
ics
[
2
1
]
.
i)
Sil
h
o
u
ette
s
co
r
e
:
t
h
e
s
ilh
o
u
ett
e
s
co
r
e
ev
alu
ates
clu
s
ter
in
g
q
u
ality
b
y
c
o
m
p
a
r
in
g
h
o
w
wel
l
a
d
ata
p
o
in
t
f
its
with
in
it
s
o
wn
clu
s
ter
r
ela
tiv
e
to
th
e
clo
s
est
n
eig
h
b
o
r
in
g
clu
s
ter
[
2
2
]
.
T
h
is
is
d
o
n
e
b
y
m
ea
s
u
r
in
g
th
e
d
is
tan
ce
b
etwe
en
a
p
o
in
t
an
d
all
o
th
er
p
o
in
ts
in
its
clu
s
ter
,
as
well
a
s
i
ts
d
is
tan
ce
to
th
e
n
ea
r
est
clu
s
ter
.
C
alcu
latin
g
s
ilh
o
u
ette
s
co
r
es
r
eq
u
ir
es
two
th
i
n
g
s
,
th
e
clu
s
ter
in
g
r
esu
lt
f
r
o
m
a
ch
o
s
en
alg
o
r
ith
m
an
d
th
e
p
air
wis
e
p
r
o
x
im
ities
b
etwe
en
s
am
p
les
[
2
3
]
.
(
)
=
(
)
−
(
)
{
(
)
,
(
)
}
(
4
)
W
h
er
e
(
)
is
m
ea
n
in
tr
a
-
clu
s
ter
d
is
tan
ce
;
(
)
is
m
ea
n
n
ea
r
est
-
clu
s
t
er
d
is
tan
ce
;
a
n
d
{
(
)
,
(
)
}
is
n
o
r
m
alize
th
e
s
co
r
e
s
o
th
at
s
co
r
e
r
em
ain
with
in
-
1
a
n
d
1
.
ii)
DB
I
:
t
h
e
D
B
I
ev
alu
ates c
lu
s
te
r
in
g
q
u
ality
b
ased
o
n
clu
s
ter
c
o
m
p
ac
tn
ess
an
d
s
ep
ar
atio
n
[
2
4
]
.
I
t m
ea
s
u
r
es
h
o
w
tig
h
tly
g
r
o
u
p
ed
th
e
p
o
in
t
s
in
ea
ch
clu
s
ter
a
r
e
a
n
d
h
o
w
d
is
tin
ct
th
e
clu
s
ter
s
ar
e
f
r
o
m
o
n
e
an
o
th
er
.
L
o
wer
DB
I
v
alu
es in
d
icate
b
et
ter
clu
s
ter
in
g
.
(
)
=
1
∑
≠
+
(
̅
,
̅
)
=
1
(
5
)
W
h
er
e
an
d
is
m
ea
s
u
r
e
th
e
c
o
m
p
ac
tn
ess
o
f
cl
u
s
ter
an
d
;
an
d
(
̅
,
̅
)
is
m
ea
s
u
r
e
h
o
w
f
ar
a
p
ar
t
clu
s
ter
an
d
.
iii)
C
HI
:
t
h
e
C
HI
,
o
r
v
ar
ian
ce
r
a
tio
cr
iter
io
n
,
m
ea
s
u
r
es
clu
s
te
r
in
g
q
u
ality
b
y
c
o
m
p
ar
in
g
b
e
twee
n
-
clu
s
ter
s
ep
ar
atio
n
to
with
in
-
clu
s
ter
c
o
m
p
ac
tn
ess
[
2
5
]
.
A
h
ig
h
e
r
C
HI
in
d
icate
s
d
en
s
er
,
well
-
s
ep
ar
at
ed
clu
s
ter
s
.
(
)
=
/
(
−
1
)
/
(
−
)
(
6
)
W
h
er
e
b
etwe
en
-
clu
s
ter
s
u
m
o
f
s
q
u
ar
es
(
B
C
SS
)
is
m
ea
s
u
r
es
h
o
w
f
ar
ea
c
h
clu
s
ter
ce
n
tr
o
id
is
f
r
o
m
th
e
g
lo
b
al
ce
n
tr
o
id
;
a
n
d
W
C
SS
i
s
m
ea
s
u
r
es
h
o
w
c
o
m
p
ac
t
th
e
cl
u
s
ter
s
ar
e
b
y
s
u
m
m
in
g
th
e
s
q
u
ar
ed
d
is
tan
ce
s
o
f
p
o
i
n
ts
to
th
eir
clu
s
ter
ce
n
tr
o
id
.
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
3
.
1
.
Resul
t
s
T
h
e
an
aly
s
is
b
eg
in
s
with
d
ata
v
is
u
aliza
tio
n
to
u
n
d
e
r
s
tan
d
c
u
s
to
m
er
b
eh
a
v
io
r
.
R
ec
en
cy
v
alu
es
wer
e
r
ev
er
s
ed
f
o
r
co
n
s
is
ten
cy
with
f
r
e
q
u
en
c
y
a
n
d
m
o
n
eta
r
y
m
etr
ics,
wh
e
r
e
h
ig
h
er
v
al
u
es
in
d
icate
b
etter
cu
s
to
m
er
s
.
Fig
u
r
e
2
s
h
o
ws
th
a
t
o
v
er
h
alf
o
f
th
e
cu
s
to
m
er
s
h
av
e
h
ig
h
r
ec
en
cy
,
in
d
icatin
g
s
tr
o
n
g
lo
y
alty
in
t
h
e
p
ast
two
y
ea
r
s
.
Fig
u
r
e
3
r
ev
ea
ls
m
o
s
t
cu
s
to
m
er
s
h
av
e
lo
w
tr
an
s
ac
tio
n
f
r
eq
u
en
cy
,
wit
h
f
ew
h
av
in
g
h
ig
h
f
r
eq
u
e
n
cy
.
Similar
ly
,
Fig
u
r
e
4
s
h
o
ws
m
o
s
t
cu
s
to
m
er
s
h
a
v
e
lo
w
tr
an
s
ac
tio
n
v
alu
es.
T
o
p
r
o
tect
s
en
s
itiv
e
in
f
o
r
m
atio
n
,
th
e
cu
s
to
m
er
m
o
n
etar
y
v
alu
es
d
is
tr
ib
u
tio
n
p
l
o
t
is
n
o
r
m
alize
d
as
r
e
q
u
ested
b
y
th
e
d
ata
o
wn
er
.
Fig
u
r
e
5
s
h
o
w
th
e
co
r
r
elati
o
n
b
etwe
en
r
ec
en
c
y
an
d
f
r
eq
u
en
cy
(
Fig
u
r
e
5
(
a
)
)
,
r
ec
e
n
cy
an
d
m
o
n
etar
y
(
Fig
u
r
e
5
(
b
)
)
,
an
d
f
r
eq
u
en
cy
a
n
d
m
o
n
etar
y
(
Fig
u
r
e
5
(
c)
)
v
al
u
e.
i)
R
ec
en
cy
v
s
.
f
r
eq
u
en
cy
:
th
e
s
ca
tter
p
lo
t
s
h
o
ws
th
at
m
o
s
t
cu
s
to
m
er
s
h
av
e
lo
w
f
r
eq
u
en
cy
v
alu
es,
r
eg
ar
d
less
o
f
t
h
eir
r
ec
e
n
cy
.
H
o
wev
er
,
th
e
r
e
is
a
g
r
o
u
p
o
f
cu
s
to
m
er
s
with
h
ig
h
r
ec
en
cy
v
al
u
es
wh
o
also
ex
h
ib
it sl
ig
h
tly
h
ig
h
er
f
r
e
q
u
en
cy
,
s
u
g
g
esti
n
g
th
at
cu
s
to
m
er
s
wh
o
h
av
e
n
o
t e
n
g
ag
ed
r
ec
e
n
tly
ten
d
to
h
av
e
lo
wer
in
ter
ac
tio
n
lev
els.
ii)
R
ec
en
cy
v
s
.
m
o
n
etar
y
:
s
im
ilar
to
th
e
r
ec
en
cy
-
f
r
eq
u
en
cy
r
e
latio
n
s
h
ip
,
th
e
m
aj
o
r
ity
o
f
cu
s
to
m
er
s
with
h
ig
h
r
ec
e
n
cy
s
co
r
es
h
av
e
l
o
w
m
o
n
etar
y
v
alu
es.
A
s
m
all
p
o
r
tio
n
o
f
cu
s
to
m
er
s
with
h
ig
h
r
ec
en
cy
s
co
r
es
also
h
av
e
h
ig
h
er
m
o
n
etar
y
v
a
lu
es,
in
d
icatin
g
th
at
wh
ile
m
o
s
t
in
ac
tiv
e
cu
s
to
m
er
s
h
av
e
lo
w
s
p
en
d
in
g
,
a
f
ew
ex
ce
p
tio
n
s
ex
is
t.
iii)
Fre
q
u
en
cy
v
s
.
m
o
n
etar
y
:
th
e
p
lo
t
r
ev
ea
ls
a
p
o
s
itiv
e
co
r
r
ela
tio
n
,
wh
er
e
cu
s
to
m
er
s
with
h
i
g
h
er
p
u
r
ch
ase
f
r
eq
u
e
n
cies
ten
d
to
h
av
e
h
ig
h
er
m
o
n
etar
y
v
alu
es.
Ho
we
v
er
,
a
c
o
n
ce
n
tr
atio
n
o
f
cu
s
to
m
er
s
with
lo
w
f
r
eq
u
e
n
cy
a
n
d
l
o
w
m
o
n
etar
y
v
alu
es
s
h
o
ws
a
s
ig
n
i
f
ican
t
p
o
r
tio
n
o
f
th
e
c
u
s
to
m
er
b
ase
co
n
s
is
ts
o
f
in
f
r
eq
u
e
n
t a
n
d
lo
w
-
s
p
en
d
i
n
g
b
u
y
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.
1
5
,
No
.
4
,
Au
g
u
s
t 2
0
2
6
:
3
4
0
2
-
3
4
1
0
3406
Fig
u
r
e
2
.
Dis
tr
ib
u
tio
n
o
f
c
u
s
to
m
er
r
ec
en
c
y
v
alu
es
Fig
u
r
e
3
.
Dis
tr
ib
u
tio
n
o
f
c
u
s
to
m
er
f
r
eq
u
en
cy
v
alu
es
Fig
u
r
e
4
.
Dis
tr
ib
u
tio
n
o
f
c
u
s
to
m
er
m
o
n
eta
r
y
v
al
u
es
(
a)
(
b
)
(
c)
Fig
u
r
e
5
.
Pair
wis
e
co
r
r
elatio
n
s
am
o
n
g
(
a)
r
ec
e
n
cy
v
s
.
f
r
eq
u
e
n
cy
,
(
b
)
r
ec
e
n
cy
v
s
.
m
o
n
etar
y
,
an
d
(
c)
f
r
e
q
u
en
c
y
v
s
.
m
o
n
etar
y
T
ab
le
2
p
r
esen
ts
th
e
iter
ativ
e
r
esu
lts
o
f
th
e
k
-
m
ea
n
s
clu
s
t
er
in
g
,
v
alid
ated
u
s
in
g
th
r
ee
ev
alu
atio
n
m
etr
ics
f
o
r
ea
ch
n
u
m
b
er
o
f
c
lu
s
ter
s
.
T
o
id
en
tify
th
e
o
p
tim
al
clu
s
ter
in
g
s
o
lu
tio
n
,
th
e
v
al
id
atio
n
s
co
r
es
ar
e
n
o
r
m
alize
d
an
d
ass
ig
n
ed
eq
u
al
weig
h
ts
o
f
0
.
3
3
.
E
q
u
al
wei
g
h
ts
wer
e
ass
ig
n
ed
to
th
e
n
o
r
m
alize
d
s
ilh
o
u
ette,
DB
I
,
an
d
C
HI
to
e
n
s
u
r
e
a
b
alan
ce
d
e
v
alu
atio
n
,
as
ea
c
h
m
et
r
ic
ca
p
tu
r
es
co
m
p
lem
en
tar
y
a
s
p
ec
ts
o
f
clu
s
ter
in
g
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
A
n
emp
ir
ica
l c
o
mp
a
r
is
o
n
o
f c
l
u
s
teri
n
g
a
p
p
r
o
a
c
h
es fo
r
r
ec
en
cy
,
fr
eq
u
en
cy
,
a
n
d
mo
n
eta
r
y
…
(
Up
ek
kh
a
La
u
)
3407
q
u
ality
an
d
n
o
s
in
g
le
m
et
r
ic
is
th
eo
r
etica
lly
d
o
m
in
an
t f
o
r
th
e
g
iv
en
ap
p
licatio
n
[
2
1
]
.
T
h
e
cl
u
s
ter
co
n
f
ig
u
r
atio
n
with
th
e
h
ig
h
est
c
o
m
b
in
e
d
we
ig
h
ted
s
co
r
e
is
s
elec
ted
as
th
e
b
est
-
p
er
f
o
r
m
in
g
s
o
lu
tio
n
.
Alth
o
u
g
h
k
=
2
y
ield
s
th
e
h
ig
h
est co
m
b
in
ed
weig
h
ted
s
co
r
e,
th
is
s
o
lu
tio
n
p
r
o
v
i
d
es lim
ited
an
aly
tical
in
s
ig
h
t d
u
e
t
o
th
e
o
v
e
r
ly
co
a
r
s
e
s
eg
m
en
tatio
n
.
C
o
n
s
eq
u
en
tly
,
k
=7
is
s
elec
ted
as
th
e
o
p
tim
al
n
u
m
b
er
o
f
clu
s
ter
s
,
ac
h
iev
in
g
a
co
m
b
i
n
ed
weig
h
ted
s
co
r
e
o
f
0
.
4
9
.
Fig
u
r
e
6
illu
s
tr
ates th
e
r
esu
ltin
g
k
-
m
ea
n
s
clu
s
ter
in
g
s
o
lu
tio
n
.
Similar
to
T
a
b
le
2
,
T
a
b
le
3
p
r
esen
ts
th
e
r
esu
lts
o
f
th
e
h
ier
a
r
ch
ical
clu
s
ter
in
g
iter
atio
n
s
,
e
v
alu
ated
f
o
r
clu
s
ter
n
u
m
b
er
s
r
a
n
g
in
g
f
r
o
m
two
to
ten
.
Fo
r
ea
ch
co
n
f
ig
u
r
atio
n
,
th
e
v
alid
atio
n
m
etr
ics,
n
o
r
m
alize
d
v
alid
atio
n
s
co
r
es,
an
d
co
m
b
in
ed
weig
h
ted
s
co
r
es
a
r
e
r
ep
o
r
ted
.
B
ased
o
n
t
h
ese
r
esu
lts
,
k
=3
is
s
elec
ted
as
t
h
e
o
p
tim
al
n
u
m
b
er
o
f
clu
s
ter
s
f
o
r
th
e
h
ier
a
r
ch
ical
clu
s
ter
in
g
an
aly
s
is
.
Fig
u
r
e
7
illu
s
tr
ates
th
e
h
ier
ar
c
h
ical
clu
s
ter
in
g
s
o
lu
tio
n
.
T
ab
le
2
.
K
-
m
ea
n
s
clu
s
ter
in
g
r
esu
lts
k
S
i
l
h
o
u
e
t
t
e
D
B
I
C
H
I
S
i
l
h
o
u
e
t
t
e
_
n
o
r
m
D
B
I
_
n
o
r
m
C
H
I
_
n
o
r
m
W
e
i
g
h
t
e
d
_
sc
o
r
e
2
0
.
7
2
5
6
4
0
6
8
1
0
.
4
3
8
0
4
6
9
0
3
3
8
0
.
4
3
2
1
6
1
1
0
.
2
0
3
2
8
1
7
4
0
.
7
2
7
0
8
2
9
7
7
0
.
6
1
8
5
3
9
9
4
1
0
.
4
6
8
3
9
8
5
8
3
8
6
9
.
6
4
4
6
1
0
.
2
6
4
0
0
8
7
0
0
.
7
6
4
6
4
1
2
4
5
0
.
4
5
9
7
1
3
1
0
0
.
4
9
1
1
5
9
8
0
6
0
.
6
2
0
6
2
0
3
4
3
0
.
4
6
4
4
0
8
5
6
3
5
0
2
.
4
2
2
1
4
0
.
2
7
8
3
0
5
1
3
0
.
7
9
5
5
8
1
3
7
4
0
.
2
6
7
2
2
5
4
5
0
.
4
4
2
5
6
6
9
4
9
0
.
5
9
4
3
7
5
8
1
5
0
.
5
3
5
2
9
7
2
3
4
8
5
2
.
0
5
6
2
7
0
.
0
9
7
9
5
3
9
5
0
.
2
4
5
8
8
2
8
7
2
0
.
9
7
4
6
6
5
5
7
0
.
4
3
5
1
0
5
7
9
3
0
.
6
6
4
0
0
9
6
9
9
0
.
4
9
0
8
2
3
4
5
3
1
6
0
.
2
1
8
0
5
0
.
5
7
6
4
7
4
7
6
0
.
5
9
0
7
5
0
0
0
8
0
.
0
8
7
8
5
1
7
2
0
.
4
1
4
1
7
5
2
4
T
ab
le
3
.
Hier
ar
c
h
ical
clu
s
ter
in
g
r
esu
lts
k
S
i
l
h
o
u
e
t
t
e
D
B
I
C
H
I
S
i
l
h
o
u
e
t
t
e
_
n
o
r
m
D
B
I
_
n
o
r
m
C
H
I
_
n
o
r
m
W
e
i
g
h
t
e
d
_
sc
o
r
e
2
0
.
7
0
0
4
4
9
1
3
0
.
4
8
5
5
5
9
1
2
2
9
8
0
.
2
9
0
0
4
7
1
1
0
.
0
4
4
7
3
0
9
7
4
0
.
6
7
4
7
6
1
2
2
3
0
.
6
5
7
9
8
4
3
1
0
.
4
9
1
8
9
9
6
8
3
1
2
1
.
7
2
1
2
6
2
0
.
7
9
4
6
6
2
5
1
9
0
.
9
6
2
5
9
8
2
0
8
0
.
1
2
3
6
9
2
4
8
5
0
.
6
2
0
7
1
4
5
6
10
0
.
5
0
2
4
3
4
2
9
0
.
5
5
4
2
4
1
0
5
4
6
9
1
.
3
1
1
7
5
2
0
.
0
4
2
5
0
4
7
3
2
0
.
5
9
4
8
5
7
7
6
6
1
0
.
5
4
0
3
2
9
6
2
9
0
.
5
1
9
7
6
6
1
4
0
.
5
5
0
6
6
7
9
1
4
0
3
3
.
0
4
7
7
7
3
0
.
1
2
6
3
1
2
3
9
7
0
.
6
1
5
9
3
5
0
7
5
0
.
6
3
2
4
8
9
0
6
3
0
.
4
5
3
6
6
3
0
5
4
0
.
6
7
1
0
8
3
7
1
0
.
5
7
1
1
7
1
8
7
2
9
0
0
.
1
7
0
5
5
7
0
.
8
5
8
0
0
4
3
0
8
0
.
4
9
4
9
8
5
8
8
7
0
0
.
4
4
6
4
8
6
7
6
Fig
u
r
e
6
.
K
-
m
ea
n
s
clu
s
ter
in
g
r
esu
lt
Fig
u
r
e
7
.
Hier
ar
c
h
ical
clu
s
ter
in
g
r
esu
lt
T
h
e
DB
SC
AN
co
n
f
ig
u
r
atio
n
d
if
f
er
s
f
r
o
m
th
o
s
e
o
f
k
-
m
ea
n
s
a
n
d
h
ier
ar
ch
ical
clu
s
ter
in
g
,
as
i
t
d
o
es n
o
t
r
eq
u
ir
e
p
r
io
r
s
p
ec
if
icatio
n
o
f
th
e
clu
s
ter
s
n
u
m
b
er
.
I
n
s
tead
,
DB
S
C
AN
r
elies
o
n
two
p
ar
a
m
eter
s
,
ε
(
ep
s
ilo
n
)
an
d
m
in
_
s
am
p
les.
I
n
th
is
s
tu
d
y
,
ε
was
v
ar
ied
f
r
o
m
0
.
1
to
5
.
0
with
in
cr
em
en
ts
o
f
0
.
1
,
wh
ile
m
in
_
s
am
p
les
r
an
g
ed
f
r
o
m
6
to
2
0
with
in
cr
em
en
ts
o
f
1
.
T
ab
le
4
p
r
esen
ts
th
e
to
p
ten
p
ar
am
eter
co
m
b
in
a
tio
n
s
o
b
tain
ed
f
r
o
m
th
is
iter
ativ
e
p
r
o
ce
s
s
,
f
r
o
m
wh
ich
ε=
1
.
6
a
n
d
m
in
_
s
am
p
les
=7
wer
e
id
en
tifie
d
as
p
r
o
d
u
cin
g
th
e
b
est
clu
s
ter
in
g
p
er
f
o
r
m
an
ce
.
T
h
e
r
esu
ltin
g
D
B
S
C
AN
clu
s
ter
in
g
s
tr
u
ctu
r
e
b
ased
o
n
th
ese
p
ar
a
m
eter
s
is
illu
s
tr
ated
in
Fig
u
r
e
8.
T
ab
le
4
.
T
o
p
1
0
DB
SC
AN
clu
s
ter
in
g
r
esu
lts
e
p
s
mi
n
_
sa
mp
l
e
s
C
l
u
st
e
r
s
S
i
l
h
o
u
e
t
t
e
D
B
I
C
H
I
S
i
l
h
o
u
e
t
t
e
_
n
o
r
m
D
B
I
_
n
o
r
m
C
H
I
_
n
o
r
m
W
e
i
g
h
t
e
d
_
s
c
o
r
e
1
.
6
7
3
0
.
4
8
9
0
2
1
.
1
3
4
4
0
1
7
0
6
.
92
0
.
9
3
3
8
7
0
.
7
3
1
5
8
1
0
.
8
7
9
6
0
2
2
.
1
12
3
0
.
5
3
1
5
1
1
.
1
9
8
8
5
1
4
9
6
.
54
0
.
9
7
0
2
5
0
.
7
1
1
0
5
0
.
8
7
4
9
8
5
0
.
8
4
3
5
7
6
2
10
3
0
.
5
1
9
2
3
1
.
2
4
7
8
5
1
4
6
4
.
34
0
.
9
5
9
7
3
0
.
6
9
5
4
4
0
.
8
5
5
8
5
3
0
.
8
2
8
6
4
2
1
.
4
6
4
0
.
4
6
1
7
7
1
.
0
4
3
0
8
1
3
8
9
.
07
0
.
9
1
0
5
5
0
.
7
6
0
6
7
0
.
8
1
1
1
2
3
0
.
8
1
9
1
7
5
1
.
9
9
3
0
.
5
3
0
3
2
1
.
2
9
5
3
3
1
3
9
3
.
93
0
.
9
6
9
2
3
0
.
6
8
0
3
2
0
.
8
1
4
0
1
2
0
.
8
1
2
9
7
6
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.
1
5
,
No
.
4
,
Au
g
u
s
t 2
0
2
6
:
3
4
0
2
-
3
4
1
0
3408
Fig
u
r
e
8
.
DB
SC
AN
clu
s
ter
in
g
r
esu
lt
3
.
2
.
Dis
cus
s
io
n
T
h
is
s
u
b
s
ec
tio
n
ev
alu
ates
th
e
clu
s
ter
in
g
r
esu
lts
u
s
in
g
v
alid
a
tio
n
m
etr
ics
an
d
ex
ten
d
s
th
e
a
n
aly
s
is
to
s
tr
ateg
ic
cu
s
to
m
er
in
s
ig
h
ts
.
Fi
g
u
r
es
2
to
4
s
h
o
w
th
at
th
e
f
r
e
q
u
en
cy
an
d
m
o
n
etar
y
v
ar
ia
b
les
ar
e
lef
t
-
s
k
ewe
d
,
wh
ile
r
ec
en
cy
is
r
ig
h
t
-
s
k
ewe
d
—
an
ex
p
ec
ted
p
atter
n
alig
n
ed
with
th
e
Par
eto
p
r
in
ci
p
le,
wh
ich
s
tates
th
at
a
s
m
all
p
r
o
p
o
r
tio
n
o
f
cu
s
to
m
e
r
s
co
n
tr
i
b
u
tes
to
th
e
m
ajo
r
ity
o
f
r
ev
en
u
e
[
2
6
]
.
I
n
th
is
d
ataset,
8
4
%
o
f
th
e
r
ev
en
u
e
is
g
en
er
ated
b
y
ju
s
t
1
6
%
o
f
th
e
cu
s
to
m
er
s
.
I
d
en
tify
in
g
th
e
s
e
h
ig
h
-
v
alu
e
"Par
eto
cu
s
to
m
er
s
"
is
e
s
s
en
tial
f
o
r
ef
f
ec
tiv
e
s
eg
m
en
tatio
n
an
d
tar
g
eted
m
ar
k
etin
g
.
C
lu
s
ter
in
g
tech
n
iq
u
es
th
at
ca
n
ac
c
u
r
ately
is
o
late
th
ese
cu
s
to
m
er
s
p
r
o
v
i
d
e
s
ig
n
if
ican
t
b
u
s
in
ess
v
alu
e.
T
h
e
ex
p
e
r
im
en
tal
r
esu
lts
,
s
h
o
wn
in
T
a
b
le
5
,
in
d
icate
th
at
k
-
m
ea
n
s
ac
h
iev
ed
th
e
b
est
p
er
f
o
r
m
an
ce
o
n
th
e
DB
I
an
d
C
HI
,
wh
ile
h
ier
ar
ch
ical
clu
s
ter
in
g
h
ad
th
e
h
i
g
h
est
s
ilh
o
u
ette
s
co
r
e
.
DB
SC
AN
p
er
f
o
r
m
ed
th
e
wo
r
s
t
ac
r
o
s
s
all
m
etr
ics.
Vis
u
aliza
tio
n
s
in
Fig
u
r
e
s
6
to
8
s
u
p
p
o
r
t
th
ese
f
in
d
in
g
s
:
k
-
m
e
an
s
is
th
e
o
n
ly
m
eth
o
d
th
at
clea
r
ly
s
ep
ar
ates
Par
eto
cu
s
to
m
er
s
in
to
d
is
tin
ct
clu
s
ter
s
(
e.
g
.
,
p
u
r
p
le
an
d
c
y
an
p
o
in
ts
with
h
ig
h
v
al
u
es).
Hier
ar
ch
ical
c
lu
s
ter
in
g
f
ails
to
d
is
tin
g
u
is
h
th
ese
cu
s
to
m
er
s
,
g
r
o
u
p
in
g
th
em
with
l
o
wer
-
v
alu
e
s
eg
m
en
ts
.
Alth
o
u
g
h
DB
SC
AN
is
ef
f
ec
t
iv
e
at
id
en
tify
in
g
n
o
is
e,
in
t
h
is
co
n
tex
t
it
m
is
class
if
ies
P
ar
eto
cu
s
to
m
er
s
as
o
u
tlier
s
—
lim
itin
g
its
u
s
ef
u
ln
ess
wh
en
h
ig
h
-
v
al
u
e
cu
s
to
m
er
s
m
ay
ap
p
ea
r
as sp
ar
s
e
b
u
t c
r
iti
ca
l c
ases
in
d
ata.
T
ab
le
5
.
C
o
m
p
a
r
is
o
n
o
f
clu
s
ter
in
g
p
e
r
f
o
r
m
an
ce
ac
r
o
s
s
m
eth
o
d
s
A
l
g
o
r
i
t
h
m
k
S
i
l
h
o
u
e
t
t
e
D
B
I
C
H
I
K
-
me
a
n
s
7
0
.
6
1
8
5
3
9
9
4
0
.
4
6
8
3
9
8
5
8
3
.
8
6
9
.
6
4
4
.
617
H
i
e
r
a
r
c
h
i
c
a
l
3
0
.
6
5
7
9
8
4
3
1
0
.
4
9
1
8
9
9
6
8
3
.
1
2
1
.
7
2
1
.
262
D
B
S
C
A
N
3
0
.
4
8
9
0
2
7
6
5
1
1
3
.
4
4
0
.
359
1
.
7
0
6
.
9
2
4
.
696
T
h
is
s
tu
d
y
h
as
s
ev
er
al
lim
it
atio
n
s
.
First,
th
e
an
aly
s
is
is
b
ased
o
n
a
s
in
g
le
d
ataset
f
r
o
m
o
n
e
d
is
tr
ib
u
tio
n
b
u
s
in
ess
,
wh
ich
m
ay
lim
it
th
e
g
en
er
aliza
b
ilit
y
o
f
t
h
e
f
in
d
in
g
s
to
o
th
e
r
in
d
u
s
tr
ies
o
r
cu
s
to
m
er
s
tr
u
ctu
r
es.
Fu
tu
r
e
r
esear
ch
s
h
o
u
ld
v
alid
ate
th
e
p
r
o
p
o
s
ed
a
p
p
r
o
ac
h
ac
r
o
s
s
m
u
ltip
le
d
atasets
an
d
s
ec
to
r
s
to
en
h
an
ce
ex
ter
n
al
v
alid
ity
.
Se
co
n
d
,
th
e
s
tu
d
y
em
p
lo
y
s
eq
u
al
weig
h
tin
g
ac
r
o
s
s
clu
s
ter
in
g
v
alid
atio
n
m
etr
ics
with
o
u
t
co
n
d
u
ctin
g
s
en
s
itiv
ity
an
aly
s
is
;
alth
o
u
g
h
th
is
e
n
s
u
r
es
a
b
alan
ce
d
ev
alu
atio
n
,
th
e
ab
s
en
ce
o
f
r
o
b
u
s
tn
ess
test
in
g
m
ay
in
f
lu
e
n
ce
th
e
s
tab
ilit
y
o
f
th
e
f
in
al
clu
s
ter
in
g
s
elec
tio
n
u
n
d
e
r
alter
n
ativ
e
weig
h
tin
g
s
ch
em
es.
Fu
tu
r
e
wo
r
k
s
h
o
u
ld
i
n
co
r
p
o
r
ate
s
en
s
itiv
ity
an
aly
s
is
to
f
u
r
t
h
er
ass
ess
th
e
co
n
s
is
ten
cy
o
f
th
e
r
esu
lts
.
4.
CO
NCLU
SI
O
N
T
h
is
s
tu
d
y
ev
alu
ated
k
-
m
ea
n
s
,
h
ier
ar
ch
ical
clu
s
ter
in
g
,
an
d
DB
SC
A
N
f
o
r
R
FM
-
b
ased
cu
s
to
m
er
s
eg
m
en
tatio
n
u
s
in
g
th
e
s
ilh
o
u
ette
s
co
r
e,
DB
I
,
an
d
C
HI
.
T
h
e
r
esu
lts
s
h
o
w
th
at
k
-
m
ea
n
s
co
n
s
is
ten
tly
o
u
tp
er
f
o
r
m
s
o
th
e
r
m
eth
o
d
s
a
n
d
is
u
n
iq
u
ely
ef
f
ec
tiv
e
in
id
en
tify
in
g
h
ig
h
-
v
alu
e
Par
eto
c
u
s
to
m
er
s
,
m
ak
in
g
it
p
ar
ticu
lar
ly
s
u
itab
le
f
o
r
d
ata
-
d
r
iv
en
m
ar
k
etin
g
ap
p
licatio
n
s
.
Aca
d
em
ically
,
th
e
s
tu
d
y
co
n
tr
ib
u
tes
a
s
y
s
tem
ati
c
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
A
n
emp
ir
ica
l c
o
mp
a
r
is
o
n
o
f c
l
u
s
teri
n
g
a
p
p
r
o
a
c
h
es fo
r
r
ec
en
cy
,
fr
eq
u
en
cy
,
a
n
d
mo
n
eta
r
y
…
(
Up
ek
kh
a
La
u
)
3409
co
m
p
ar
is
o
n
o
f
cl
u
s
ter
in
g
t
ec
h
n
iq
u
es
th
at
i
n
teg
r
ates
q
u
an
titativ
e
v
alid
atio
n
m
etr
ics
with
b
u
s
in
ess
in
ter
p
r
etab
ilit
y
,
em
p
h
asizin
g
t
h
e
im
p
o
r
tan
ce
o
f
alig
n
in
g
m
e
th
o
d
o
lo
g
ical
ch
o
ices
with
a
n
a
ly
tical
o
b
jectiv
es.
Pra
ctica
lly
,
th
e
f
in
d
in
g
s
d
em
o
n
s
tr
ate
th
at
k
-
m
ea
n
s
o
f
f
e
r
s
a
r
eliab
le
an
d
ac
tio
n
ab
le
ap
p
r
o
ac
h
f
o
r
cu
s
to
m
er
s
eg
m
en
tatio
n
to
s
u
p
p
o
r
t
tar
g
eted
m
ar
k
etin
g
an
d
r
es
o
u
r
ce
allo
ca
tio
n
d
ec
is
io
n
s
.
Nev
er
th
eless
,
th
e
g
en
er
aliza
b
ilit
y
o
f
th
e
r
esu
lts
is
lim
ited
b
y
th
e
d
ataset
an
d
m
eth
o
d
s
u
s
ed
,
an
d
f
u
tu
r
e
r
esear
ch
s
h
o
u
ld
e
x
p
lo
r
e
ad
d
itio
n
al
clu
s
ter
in
g
al
g
o
r
ith
m
s
an
d
en
h
a
n
ce
d
R
FM
m
o
d
el
s
tailo
r
ed
to
o
th
er
b
u
s
in
ess
co
n
tex
ts
.
ACK
NO
WL
E
DG
M
E
N
T
S
T
h
e
r
esear
ch
e
r
s
wo
u
ld
lik
e
t
o
ex
p
r
ess
th
eir
ap
p
r
ec
iatio
n
t
o
th
e
o
wn
er
o
f
Me
g
ac
o
m
p
f
o
r
g
r
an
tin
g
p
er
m
is
s
io
n
to
u
s
e
th
e
c
o
m
p
an
y
’
s
s
ales d
ata.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
Au
th
o
r
s
s
tate
n
o
f
u
n
d
in
g
in
v
o
lv
ed
.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
T
h
is
jo
u
r
n
al
u
s
es
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT)
to
r
ec
o
g
n
ize
in
d
iv
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
llab
o
r
atio
n
.
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
Up
ek
k
h
a
L
a
u
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
Me
d
ity
a
W
ases
a
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
Au
th
o
r
s
s
tate
n
o
co
n
f
lict o
f
in
t
er
est.
I
NF
O
RM
E
D
CO
NS
E
N
T
W
e
h
av
e
o
b
tain
ed
in
f
o
r
m
ed
c
o
n
s
en
t f
r
o
m
all
in
d
iv
id
u
als in
c
lu
d
ed
in
t
h
is
s
tu
d
y
.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
d
ata
t
h
at
s
u
p
p
o
r
t
th
e
f
in
d
in
g
s
o
f
t
h
is
s
tu
d
y
ar
e
av
ailab
l
e
f
r
o
m
Me
g
ac
o
m
p
.
R
estrictio
n
s
ap
p
ly
t
o
th
e
av
ailab
ilit
y
o
f
t
h
ese
d
ata,
wh
ich
wer
e
u
s
ed
u
n
d
er
p
er
m
is
io
n
f
o
r
th
is
s
tu
d
y
.
Data
a
r
e
av
ailab
le
f
r
o
m
th
e
au
th
o
r
s
with
th
e
p
er
m
is
s
io
n
o
f
Me
g
ac
o
m
p
.
RE
F
E
R
E
NC
E
S
[
1
]
T.
H
.
D
a
v
e
n
p
o
r
t
,
“
C
o
m
p
e
t
i
n
g
o
n
a
n
a
l
y
t
i
c
s,
”
H
a
rv
a
r
d
B
u
si
n
e
ss
R
e
v
i
e
w
,
v
o
l
.
8
4
,
n
o
.
1
,
p
p
.
9
8
–
1
0
7
,
2
0
0
6
.
[
2
]
B
.
C
o
o
i
l
,
L
.
A
k
s
o
y
,
a
n
d
T.
L.
K
e
i
n
i
n
g
h
a
m,
“
A
p
p
r
o
a
c
h
e
s
t
o
c
u
s
t
o
m
e
r
s
e
g
men
t
a
t
i
o
n
,
”
J
o
u
r
n
a
l
o
f
Re
l
a
t
i
o
n
sh
i
p
Ma
r
k
e
t
i
n
g
,
v
o
l
.
6
,
n
o
.
3
–
4
,
p
p
.
9
–
3
9
,
Ja
n
.
2
0
0
8
,
d
o
i
:
1
0
.
1
3
0
0
/
J
3
6
6
v
0
6
n
0
3
_
0
2
.
[
3
]
J.
H
a
n
,
M
.
K
a
mb
e
r
,
a
n
d
J
.
P
e
i
,
D
a
t
a
m
i
n
i
n
g
:
c
o
n
c
e
p
t
s
a
n
d
t
e
c
h
n
i
q
u
e
s
.
M
a
ssac
h
u
se
t
t
s,
U
n
i
t
e
d
S
t
a
t
e
s
:
M
o
r
g
a
n
K
a
u
f
ma
n
n
,
2
0
1
2
,
d
o
i
:
1
0
.
1
0
1
6
/
C
2
0
0
9
-
0
-
6
1
8
1
9
-
5.
[
4
]
C
.
R
u
n
g
r
u
a
n
g
,
P
.
R
i
y
a
p
a
n
,
A
.
I
n
t
a
r
a
si
t
,
K
.
C
h
u
a
r
k
h
a
m,
a
n
d
J.
M
u
a
n
g
p
r
a
t
h
u
b
,
“
R
F
M
mo
d
e
l
c
u
st
o
mer
se
g
m
e
n
t
a
t
i
o
n
b
a
se
d
o
n
h
i
e
r
a
r
c
h
i
c
a
l
a
p
p
r
o
a
c
h
u
s
i
n
g
F
C
A
,
”
E
x
p
e
r
t
S
y
s
t
e
m
s
w
i
t
h
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
2
3
7
,
M
a
r
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
sw
a
.
2
0
2
3
.
1
2
1
4
4
9
.
[
5
]
T.
H
o
,
S
.
N
g
u
y
e
n
,
H
.
N
g
u
y
e
n
,
N
.
N
g
u
y
e
n
,
D
.
-
S
.
M
a
n
,
a
n
d
T.
-
G
.
Le
,
“
A
n
e
x
t
e
n
d
e
d
R
F
M
mo
d
e
l
f
o
r
c
u
s
t
o
me
r
b
e
h
a
v
i
o
u
r
a
n
d
d
e
m
o
g
r
a
p
h
i
c
a
n
a
l
y
si
s
i
n
r
e
t
a
i
l
i
n
d
u
st
r
y
,
”
B
u
si
n
e
ss
S
y
st
e
m
s
Re
se
a
rc
h
J
o
u
r
n
a
l
,
v
o
l
.
1
4
,
n
o
.
1
,
p
p
.
2
6
–
5
3
,
S
e
p
.
2
0
2
3
,
d
o
i
:
1
0
.
2
4
7
8
/
b
sr
j
-
2
0
2
3
-
0
0
0
2
.
[
6
]
L.
B
u
i
t
i
n
c
k
e
t
a
l
.
,
“
A
P
I
d
e
si
g
n
f
o
r
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
s
o
f
t
w
a
r
e
:
e
x
p
e
r
i
e
n
c
e
s
f
r
o
m
t
h
e
s
c
i
k
i
t
-
l
e
a
r
n
p
r
o
j
e
c
t
,
”
i
n
E
u
ro
p
e
a
n
C
o
n
f
e
r
e
n
c
e
o
n
M
a
c
h
i
n
e
L
e
a
rn
i
n
g
a
n
d
Pr
i
n
c
i
p
l
e
s
a
n
d
Pra
c
t
i
c
e
s
o
f
K
n
o
w
l
e
d
g
e
D
i
sc
o
v
e
r
y
i
n
D
a
t
a
b
a
s
e
s
,
2
0
1
3
,
p
p
.
1
–
1
5
.
[
7
]
P
.
B
í
r
ó
,
B
.
B
.
H
.
K
o
v
á
c
s
,
T
.
N
o
v
á
k
,
a
n
d
M
.
Er
d
é
l
y
i
,
“
C
l
u
s
t
e
r
p
a
r
a
me
t
e
r
-
b
a
se
d
D
B
S
C
A
N
m
a
p
s
f
o
r
i
m
a
g
e
c
h
a
r
a
c
t
e
r
i
z
a
t
i
o
n
,
”
C
o
m
p
u
t
a
t
i
o
n
a
l
a
n
d
S
t
ru
c
t
u
r
a
l
Bi
o
t
e
c
h
n
o
l
o
g
y
J
o
u
r
n
a
l
,
v
o
l
.
2
7
,
p
p
.
9
2
0
–
9
2
7
,
J
a
n
.
2
0
2
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
s
b
j
.
2
0
2
5
.
0
2
.
0
3
7
.
[
8
]
P
.
A
n
i
t
h
a
a
n
d
M
.
M
.
P
a
t
i
l
,
“
R
F
M
mo
d
e
l
f
o
r
c
u
s
t
o
mer
p
u
r
c
h
a
se
b
e
h
a
v
i
o
r
u
s
i
n
g
k
-
m
e
a
n
s
a
l
g
o
r
i
t
h
m,
”
J
o
u
rn
a
l
o
f
K
i
n
g
S
a
u
d
U
n
i
v
e
rsi
t
y
-
C
o
m
p
u
t
e
r
a
n
d
I
n
f
o
rm
a
t
i
o
n
S
c
i
e
n
c
e
s
,
v
o
l
.
3
4
,
n
o
.
5
,
p
p
.
1
7
8
5
–
1
7
9
2
,
M
a
y
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
k
su
c
i
.
2
0
1
9
.
1
2
.
0
1
1
.
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.
1
5
,
No
.
4
,
Au
g
u
s
t 2
0
2
6
:
3
4
0
2
-
3
4
1
0
3410
[
9
]
O
.
N
.
A
k
a
n
d
e
,
H
.
B
.
A
k
a
n
d
e
,
E
.
O
.
A
s
a
n
i
,
a
n
d
B
.
T
.
D
a
u
t
a
r
e
,
“
C
u
s
t
o
m
e
r
s
e
g
m
e
n
t
a
t
i
o
n
t
h
r
o
u
g
h
R
F
M
a
n
a
l
y
s
i
s
a
n
d
k
-
m
e
a
n
s
c
l
u
s
t
e
r
i
n
g
:
l
e
v
e
r
a
g
i
n
g
d
a
t
a
-
d
r
i
v
e
n
i
n
s
i
g
h
t
s
f
o
r
e
f
f
e
c
t
i
v
e
m
a
r
k
e
t
i
n
g
s
t
r
a
t
e
g
y
,
”
i
n
2
0
2
4
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
S
c
i
e
n
c
e
,
E
n
g
i
n
e
e
r
i
n
g
a
n
d
B
u
s
i
n
e
s
s
f
o
r
D
r
i
v
i
n
g
S
u
s
t
a
i
n
a
b
l
e
D
e
v
e
l
o
p
m
e
n
t
G
o
a
l
s
(
S
E
B
4
S
D
G
)
,
2
0
2
4
,
p
p
.
1
–
8
,
d
o
i
:
1
0
.
1
1
0
9
/
S
E
B
4
S
D
G
6
0
8
7
1
.
2
0
2
4
.
1
0
6
3
0
0
5
2
.
[
1
0
]
A
.
J.
C
h
r
i
st
y
,
A
.
U
ma
ma
k
e
sw
a
r
i
,
L.
P
r
i
y
a
t
h
a
r
si
n
i
,
a
n
d
A
.
N
e
y
a
a
,
“
R
F
M
r
a
n
k
i
n
g
–
a
n
e
f
f
e
c
t
i
v
e
a
p
p
r
o
a
c
h
t
o
c
u
s
t
o
me
r
seg
m
e
n
t
a
t
i
o
n
,
”
J
o
u
r
n
a
l
o
f
K
i
n
g
S
a
u
d
U
n
i
v
e
rsi
t
y
-
C
o
m
p
u
t
e
r
a
n
d
I
n
f
o
rm
a
t
i
o
n
S
c
i
e
n
c
e
s
,
v
o
l
.
3
3
,
n
o
.
1
0
,
p
p
.
1
2
5
1
–
1
2
5
7
,
D
e
c
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
k
s
u
c
i
.
2
0
1
8
.
0
9
.
0
0
4
.
[
1
1
]
N
.
P
.
S
u
t
r
a
mi
a
n
i
,
I
.
M
.
T.
A
r
t
h
a
n
a
,
P
.
F
.
La
m
p
u
n
g
,
S
.
A
u
r
e
l
i
a
,
M
.
F
a
u
z
i
,
a
n
d
I
.
W
.
A
.
S
.
D
a
r
ma,
“
T
h
e
p
e
r
f
o
r
ma
n
c
e
c
o
m
p
a
r
i
s
o
n
o
f
D
B
S
C
A
N
a
n
d
k
-
me
a
n
s
c
l
u
st
e
r
i
n
g
f
o
r
M
S
M
Es
g
r
o
u
p
i
n
g
b
a
se
d
o
n
a
sse
t
v
a
l
u
e
a
n
d
t
u
r
n
o
v
e
r
,
”
J
o
u
rn
a
l
o
f
I
n
f
o
rm
a
t
i
o
n
S
y
st
e
m
s
En
g
i
n
e
e
ri
n
g
a
n
d
B
u
s
i
n
e
ss I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
1
0
,
n
o
.
1
,
p
p
.
1
3
–
2
4
,
F
e
b
.
2
0
2
4
,
d
o
i
:
1
0
.
2
0
4
7
3
/
j
i
se
b
i
.
1
0
.
1
.
1
3
-
2
4
.
[
1
2
]
A
.
G
r
i
v
a
,
C
.
B
a
r
d
a
k
i
,
K
.
P
r
a
ma
t
a
r
i
,
a
n
d
D
.
P
a
p
a
k
i
r
i
a
k
o
p
o
u
l
o
s,
“
R
e
t
a
i
l
b
u
s
i
n
e
ss
a
n
a
l
y
t
i
c
s
:
c
u
st
o
m
e
r
v
i
si
t
se
g
me
n
t
a
t
i
o
n
u
si
n
g
mark
e
t
b
a
s
k
e
t
d
a
t
a
,
”
Ex
p
e
r
t
S
y
st
e
m
s
w
i
t
h
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
1
0
0
,
p
p
.
1
–
1
6
,
Ju
n
.
2
0
1
8
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
sw
a
.
2
0
1
8
.
0
1
.
0
2
9
.
[
1
3
]
J.
Zh
o
u
,
J.
W
e
i
,
a
n
d
B
.
X
u
,
“
C
u
st
o
mer
seg
m
e
n
t
a
t
i
o
n
b
y
w
e
b
c
o
n
t
e
n
t
m
i
n
i
n
g
,
”
J
o
u
r
n
a
l
o
f
R
e
t
a
i
l
i
n
g
a
n
d
C
o
n
su
m
e
r
S
e
rv
i
c
e
s
,
v
o
l
.
6
1
,
Ju
l
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
r
e
t
c
o
n
s
e
r
.
2
0
2
1
.
1
0
2
5
8
8
.
[
1
4
]
M
.
H
o
ss
e
i
n
i
,
N
.
A
b
d
o
l
v
a
n
d
,
a
n
d
S
.
R
.
H
a
r
a
n
d
i
,
“
Tw
o
-
d
i
me
n
s
i
o
n
a
l
a
n
a
l
y
si
s
o
f
c
u
st
o
mer
b
e
h
a
v
i
o
r
i
n
t
r
a
d
i
t
i
o
n
a
l
a
n
d
e
l
e
c
t
r
o
n
i
c
b
a
n
k
i
n
g
,
”
D
i
g
i
t
a
l
B
u
si
n
e
ss
,
v
o
l
.
2
,
n
o
.
2
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
d
i
g
b
u
s
.
2
0
2
2
.
1
0
0
0
3
0
.
[
1
5
]
H
.
A
b
b
a
si
m
e
h
r
a
n
d
A
.
B
a
h
r
i
n
i
,
“
A
n
a
n
a
l
y
t
i
c
a
l
f
r
a
mew
o
r
k
b
a
se
d
o
n
t
h
e
r
e
c
e
n
c
y
,
f
r
e
q
u
e
n
c
y
,
a
n
d
mo
n
e
t
a
r
y
mo
d
e
l
a
n
d
t
i
m
e
s
e
r
i
e
s
c
l
u
st
e
r
i
n
g
t
e
c
h
n
i
q
u
e
s
f
o
r
d
y
n
a
mi
c
se
g
me
n
t
a
t
i
o
n
,
”
Ex
p
e
r
t
S
y
s
t
e
m
s
w
i
t
h
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
1
9
2
,
A
p
r
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
sw
a
.
2
0
2
1
.
1
1
6
3
7
3
.
[
1
6
]
S
.
H
a
g
h
i
g
h
a
t
n
i
a
,
N
.
A
b
d
o
l
v
a
n
d
,
a
n
d
S
.
R
.
H
a
r
a
n
d
i
,
“
Ev
a
l
u
a
t
i
n
g
d
i
sc
o
u
n
t
s a
s
a
d
i
me
n
si
o
n
o
f
c
u
s
t
o
m
e
r
b
e
h
a
v
i
o
r
a
n
a
l
y
si
s
,
”
J
o
u
r
n
a
l
o
f
M
a
rke
t
i
n
g
C
o
m
m
u
n
i
c
a
t
i
o
n
s
,
v
o
l
.
2
4
,
n
o
.
4
,
p
p
.
3
2
1
–
3
3
6
,
M
a
y
2
0
1
8
,
d
o
i
:
1
0
.
1
0
8
0
/
1
3
5
2
7
2
6
6
.
2
0
1
7
.
1
4
1
0
2
1
0
.
[
1
7
]
R
.
J
.
K
u
o
,
M
.
N
.
A
l
f
a
r
e
z
a
,
a
n
d
T.
P
.
Q
.
N
g
u
y
e
n
,
“
G
e
n
e
t
i
c
b
a
s
e
d
d
e
n
si
t
y
p
e
a
k
p
o
ssi
b
i
l
i
st
i
c
f
u
z
z
y
c
-
m
e
a
n
s
a
l
g
o
r
i
t
h
ms
t
o
c
l
u
s
t
e
r
a
n
a
l
y
si
s
-
a
c
a
se
st
u
d
y
o
n
c
u
st
o
m
e
r
se
g
me
n
t
a
t
i
o
n
,
”
En
g
i
n
e
e
r
i
n
g
S
c
i
e
n
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
,
a
n
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
rn
a
l
,
v
o
l
.
4
7
,
N
o
v
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
e
s
t
c
h
.
2
0
2
3
.
1
0
1
5
2
5
.
[
1
8
]
S
.
M
o
n
a
l
i
s
a
,
P
.
N
a
d
y
a
,
a
n
d
R
.
N
o
v
i
t
a
,
“
A
n
a
l
y
s
i
s
f
o
r
c
u
st
o
mer
l
i
f
e
t
i
m
e
v
a
l
u
e
c
a
t
e
g
o
r
i
z
a
t
i
o
n
w
i
t
h
R
F
M
mo
d
e
l
,
”
Pr
o
c
e
d
i
a
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
1
6
1
,
p
p
.
8
3
4
–
8
4
0
,
2
0
1
9
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
r
o
c
s.
2
0
1
9
.
1
1
.
1
9
0
.
[
1
9
]
H
.
H
.
H
a
m
i
d
i
a
n
d
B
.
H
a
g
h
i
,
“
A
n
a
p
p
r
o
a
c
h
b
a
se
d
o
n
d
a
t
a
m
i
n
i
n
g
a
n
d
g
e
n
e
t
i
c
a
l
g
o
r
i
t
h
m
t
o
o
p
t
i
m
i
z
i
n
g
t
i
me
se
r
i
e
s
c
l
u
s
t
e
r
i
n
g
f
o
r
e
f
f
i
c
i
e
n
t
se
g
m
e
n
t
a
t
i
o
n
o
f
c
u
st
o
mer
b
e
h
a
v
i
o
r
,
”
C
o
m
p
u
t
e
rs
i
n
H
u
m
a
n
Be
h
a
v
i
o
r
R
e
p
o
r
t
s
,
v
o
l
.
1
6
,
D
e
c
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
h
b
r
.
2
0
2
4
.
1
0
0
5
2
0
.
[
2
0
]
S
c
i
k
i
t
Le
a
r
n
,
“
2
.
3
.
C
l
u
s
t
e
r
i
n
g
,
”
sci
k
i
t
-
l
e
a
rn
.
o
r
g
.
A
c
c
e
sse
d
:
M
a
r
.
2
2
,
2
0
2
5
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s
:
/
/
sc
i
k
i
t
-
l
e
a
r
n
.
o
r
g
/
st
a
b
l
e
/
mo
d
u
l
e
s/
c
l
u
st
e
r
i
n
g
.
h
t
ml
[
2
1
]
O
.
A
r
b
e
l
a
i
t
z
,
I
.
G
u
r
r
u
t
x
a
g
a
,
J
.
M
u
g
u
e
r
z
a
,
J
.
M
.
P
é
r
e
z
,
a
n
d
I
.
P
e
r
o
n
a
,
“
A
n
e
x
t
e
n
s
i
v
e
c
o
m
p
a
r
a
t
i
v
e
st
u
d
y
o
f
c
l
u
st
e
r
v
a
l
i
d
i
t
y
i
n
d
i
c
e
s,”
Pa
t
t
e
r
n
Re
c
o
g
n
i
t
i
o
n
,
v
o
l
.
4
6
,
n
o
.
1
,
p
p
.
2
4
3
–
2
5
6
,
Ja
n
.
2
0
1
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
a
t
c
o
g
.
2
0
1
2
.
0
7
.
0
2
1
.
[
2
2
]
L.
L
o
v
m
a
r
,
A
.
A
h
l
f
o
r
d
,
M
.
J
o
n
ss
o
n
,
a
n
d
A
.
-
C
.
S
y
v
ä
n
e
n
,
“
S
i
l
h
o
u
e
t
t
e
s
c
o
r
e
s
f
o
r
a
sse
ssm
e
n
t
o
f
S
N
P
g
e
n
o
t
y
p
e
c
l
u
st
e
r
s,”
B
MC
G
e
n
o
m
i
c
s
,
v
o
l
.
6
,
n
o
.
1
,
M
a
r
.
2
0
0
5
,
d
o
i
:
1
0
.
1
1
8
6
/
1
4
7
1
-
2
1
6
4
-
6
-
3
5
.
[
2
3
]
P
.
J.
R
o
u
sse
e
u
w
,
“
S
i
l
h
o
u
e
t
t
e
s:
a
g
r
a
p
h
i
c
a
l
a
i
d
t
o
t
h
e
i
n
t
e
r
p
r
e
t
a
t
i
o
n
a
n
d
v
a
l
i
d
a
t
i
o
n
o
f
c
l
u
s
t
e
r
a
n
a
l
y
si
s,
”
J
o
u
rn
a
l
o
f
C
o
m
p
u
t
a
t
i
o
n
a
l
a
n
d
Ap
p
l
i
e
d
Ma
t
h
e
m
a
t
i
c
s
,
v
o
l
.
2
0
,
p
p
.
5
3
–
6
5
,
N
o
v
.
1
9
8
7
,
d
o
i
:
1
0
.
1
0
1
6
/
0
3
7
7
-
0
4
2
7
(
8
7
)
9
0
1
2
5
-
7.
[
2
4
]
F
.
R
o
s,
R
.
R
i
a
d
,
a
n
d
S
.
G
u
i
l
l
a
u
me
,
“
P
D
B
I
:
a
p
a
r
t
i
t
i
o
n
i
n
g
D
a
v
i
e
s
-
B
o
u
l
d
i
n
i
n
d
e
x
f
o
r
c
l
u
st
e
r
i
n
g
e
v
a
l
u
a
t
i
o
n
,
”
N
e
u
ro
c
o
m
p
u
t
i
n
g
,
v
o
l
.
5
2
8
,
p
p
.
1
7
8
–
1
9
9
,
A
p
r
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
n
e
u
c
o
m
.
2
0
2
3
.
0
1
.
0
4
3
.
[
2
5
]
T.
C
a
l
i
ń
s
k
i
a
n
d
J.
H
a
r
a
b
a
s
z
,
“
A
d
e
n
d
r
i
t
e
m
e
t
h
o
d
f
o
r
c
l
u
st
e
r
a
n
a
l
y
si
s
,
”
C
o
m
m
u
n
i
c
a
t
i
o
n
s
i
n
S
t
a
t
i
st
i
c
s
,
v
o
l
.
3
,
n
o
.
1
,
p
p
.
1
–
2
7
,
Ja
n
.
1
9
7
4
,
d
o
i
:
1
0
.
1
0
8
0
/
0
3
6
1
0
9
2
7
4
0
8
8
2
7
1
0
1
.
[
2
6
]
R
.
S
a
n
d
e
r
s
,
“
Th
e
p
a
r
e
t
o
p
r
i
n
c
i
p
l
e
:
I
t
s
u
s
e
a
n
d
a
b
u
se
,
”
J
o
u
r
n
a
l
o
f
Pro
d
u
c
t
&
Br
a
n
d
M
a
n
a
g
e
m
e
n
t
,
v
o
l
.
1
,
n
o
.
2
,
p
p
.
3
7
–
4
0
,
1
9
9
2
,
d
o
i
:
1
0
.
1
1
0
8
/
e
b
0
2
4
7
0
6
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Upekk
h
a
La
u
g
ra
d
u
a
ted
with
a
d
e
g
re
e
i
n
I
n
fo
rm
a
ti
c
s
fr
o
m
P
a
ra
h
y
a
n
g
a
n
Ca
th
o
li
c
Un
iv
e
rsity
i
n
2
0
2
2
.
He
b
e
g
a
n
h
is
p
r
o
fe
ss
io
n
a
l
jo
u
rn
e
y
a
s
a
d
a
ta
a
n
a
ly
st
a
t
a
re
tail
d
istri
b
u
t
io
n
c
o
m
p
a
n
y
,
wh
e
re
h
e
g
a
in
e
d
h
a
n
d
s
-
o
n
e
x
p
e
rien
c
e
in
lev
e
ra
g
i
n
g
d
a
ta
to
su
p
p
o
r
t
b
u
si
n
e
ss
d
e
c
isio
n
s.
His
wo
r
k
in
v
o
l
v
e
d
a
n
a
ly
z
in
g
sa
les
tren
d
s
a
n
d
m
o
n
it
o
r
in
g
in
v
e
n
to
r
y
p
e
rf
o
rm
a
n
c
e
.
In
2
0
2
5
,
h
e
p
u
rsu
e
d
fu
rt
h
e
r
e
d
u
c
a
ti
o
n
b
y
e
n
ro
ll
in
g
i
n
th
e
M
a
ste
r
o
f
S
c
ien
c
e
M
a
n
a
g
e
m
e
n
t
p
ro
g
ra
m
a
t
th
e
S
c
h
o
o
l
o
f
Bu
si
n
e
ss
a
n
d
M
a
n
a
g
e
m
e
n
t,
In
stit
u
t
Tek
n
o
lo
g
i
Ba
n
d
u
n
g
,
wit
h
a
fo
c
u
s
on
b
u
si
n
e
ss
a
n
a
ly
ti
c
s
.
His
c
u
rre
n
t
a
c
a
d
e
m
ic
in
tere
sts
in
c
lu
d
e
d
a
ta
-
d
riv
e
n
d
e
c
isio
n
m
a
k
in
g
,
c
lu
ste
rin
g
,
a
n
d
t
h
e
in
te
g
ra
ti
o
n
o
f
a
n
a
ly
ti
c
s
in
to
stra
teg
ic
b
u
si
n
e
ss
p
ro
c
e
ss
e
s.
He
c
a
n
b
e
c
o
n
tac
ted
at
e
m
a
il
:
u
p
k
la
u
2
1
2
1
@
g
m
a
il
.
c
o
m
.
Medi
ty
a
W
a
se
sa
is
a
n
a
ss
o
c
iate
p
ro
fe
ss
o
r
a
t
th
e
S
c
h
o
o
l
o
f
Bu
si
n
e
ss
a
n
d
M
a
n
a
g
e
m
e
n
t,
I
n
stit
u
t
Tek
n
o
l
o
g
i
Ba
n
d
u
n
g
.
He
h
o
l
d
s
a
P
h
.
D.
in
M
a
n
a
g
e
m
e
n
t
I
n
fo
rm
a
ti
o
n
S
y
ste
m
s
fro
m
Ro
tt
e
rd
a
m
S
c
h
o
o
l
o
f
M
a
n
a
g
e
m
e
n
t,
Eras
m
u
s
Un
i
v
e
rsity
,
Ne
th
e
rlan
d
s,
a
n
M
.
S
c
.
in
L
o
g
isti
c
s
En
g
in
e
e
rin
g
fr
o
m
Du
isb
u
r
g
-
Esse
n
U
n
iv
e
rsit
y
,
G
e
rm
a
n
y
,
a
n
d
a
Ba
c
h
e
lo
r
o
f
M
e
c
h
a
n
ica
l
En
g
in
e
e
rin
g
(S
.
T.
)
fr
o
m
In
stit
u
t
Tek
n
o
l
o
g
i
Ba
n
d
u
n
g
.
In
In
d
o
n
e
sia
,
h
e
h
a
s
se
rv
e
d
a
s
a
n
e
x
e
c
u
ti
v
e
,
c
o
n
su
l
tan
t,
a
n
d
t
ra
in
e
r
fo
r
a
d
iv
e
rse
in
tern
a
ti
o
n
a
l
a
n
d
n
a
ti
o
n
a
l
o
r
g
a
n
iza
ti
o
n
s.
His
p
rima
ry
in
tere
st
re
v
o
l
v
e
s
a
ro
u
n
d
lev
e
ra
g
in
g
a
d
v
a
n
c
e
d
in
f
o
r
m
a
ti
o
n
sy
ste
m
s
t
o
im
p
ro
v
e
b
u
sin
e
ss
d
e
c
isio
n
s,
wit
h
a
p
a
rti
c
u
lar
e
m
p
h
a
sis
o
n
b
u
sin
e
ss
a
n
a
ly
ti
c
s.
His
wo
rk
h
a
s
b
e
e
n
fe
a
tu
re
d
in
re
n
o
wn
e
d
i.
e
.
,
IEE
E
Ac
c
e
ss
,
Jo
u
rn
a
l
o
f
En
ter
p
rise
In
fo
rm
a
ti
o
n
M
a
n
a
g
e
m
e
n
t,
Jo
u
rn
a
l
o
f
M
a
n
a
g
e
m
e
n
t
An
a
l
y
ti
c
s,
Re
so
u
rc
e
s
P
o
li
c
y
,
Jo
u
rn
a
l
o
f
Clea
n
e
r
P
ro
d
u
c
ti
o
n
,
a
n
d
De
c
isio
n
S
u
p
p
o
rt
S
y
ste
m
s.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
m
e
d
it
y
a
.
wa
se
sa
@itb
.
a
c
.
id
.
Evaluation Warning : The document was created with Spire.PDF for Python.