I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
p
ute
r
E
ng
in
ee
ring
(
I
J
E
CE
)
Vo
l.
8
,
No
.
4
,
A
u
g
u
s
t
201
8
,
p
p
.
2
3
6
7
~2
3
8
3
I
SS
N:
2
0
8
8
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v8
i
4
.
p
p
2
3
6
7
-
2383
2367
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
co
r
e
.
co
m/
jo
u
r
n
a
ls
/in
d
ex
.
p
h
p
/
I
JE
C
E
Clustering
Predi
c
tion Techniq
ues i
n Def
i
ning
and
Pr
edicting
Custo
m
ers D
efec
t
io
n:
The
Ca
se o
f
E
-
Co
m
m
e
rce
Co
ntext
Ait
Da
qu
d Ra
chid
1
,
A
m
ine
Abdella
h
2
,
B
o
ui
k
ha
le
ne
B
ela
id
3
,
L
bi
bb
Ra
chi
d
4
1,
4
De
p
a
rtm
e
n
t
o
f
P
h
y
sic
s,
F
S
T
,
S
u
lt
a
n
M
o
u
lay
S
li
m
a
n
e
Un
iv
e
rsit
y
,
M
o
r
o
c
c
o
2
De
p
a
rtme
n
t
o
f
A
p
p
li
e
d
M
a
th
e
m
a
ti
c
s,
F
S
T
,
S
u
lt
a
n
M
o
u
lay
S
li
m
a
n
e
Un
iv
e
rsity
,
M
o
ro
c
c
o
3
De
p
a
rtme
n
t
o
f
M
a
th
e
m
a
ti
c
s an
d
In
f
o
rm
a
ti
c
s,
P
F
,
S
u
lt
a
n
M
o
u
lay
S
li
m
a
n
e
Un
iv
e
rsit
y
,
M
o
ro
c
c
o
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Au
g
3
1
,
2
0
1
7
R
ev
i
s
ed
Feb
1
0
,
2
0
1
8
A
cc
ep
ted
Feb
1
7
,
2
0
1
8
W
it
h
th
e
g
ro
w
th
o
f
th
e
e
-
c
o
m
m
e
rc
e
se
c
to
r,
c
u
sto
m
e
rs
h
a
v
e
m
o
re
c
h
o
ice
s,
a
f
a
c
t
w
h
ich
e
n
c
o
u
ra
g
e
s
th
e
m
to
d
iv
id
e
th
e
ir
p
u
rc
h
a
se
s
a
m
o
n
g
st
se
v
e
r
a
l
e
-
c
o
m
m
e
rc
e
site
s
a
n
d
c
o
m
p
a
re
th
e
ir
c
o
m
p
e
ti
to
rs‟
p
ro
d
u
c
ts,
y
e
t
th
is
in
c
re
a
se
s
h
ig
h
risk
s
o
f
c
h
u
rn
in
g
.
A
re
v
ie
w
o
f
th
e
li
tera
tu
re
o
n
c
u
sto
m
e
r
c
h
u
rn
i
n
g
m
o
d
e
ls
re
v
e
a
ls
th
a
t
n
o
p
rio
r
re
s
e
a
rc
h
h
a
d
c
o
n
sid
e
re
d
b
o
t
h
p
a
rti
a
l
a
n
d
to
tal
d
e
f
e
c
ti
o
n
in
n
o
n
-
c
o
n
trac
tu
a
l
o
n
li
n
e
e
n
v
iro
n
m
e
n
ts.
In
ste
a
d
,
th
e
y
f
o
c
u
se
d
e
it
h
e
r
o
n
a
to
tal
o
r
p
a
rti
a
l
d
e
f
e
c
t.
T
h
is
stu
d
y
p
ro
p
o
se
s
a
c
u
sto
m
e
r
c
h
u
rn
p
re
d
ictio
n
m
o
d
e
l
in
a
n
e
-
c
o
m
m
e
rc
e
c
o
n
tex
t,
w
h
e
r
e
in
a
c
lu
ste
rin
g
p
h
a
se
is
b
a
se
d
o
n
t
h
e
i
n
teg
ra
ti
o
n
o
f
th
e
k
-
m
e
a
n
s
m
e
th
o
d
a
n
d
t
h
e
L
e
n
g
th
-
Re
c
e
n
c
y
-
F
re
q
u
e
n
c
y
-
M
o
n
e
tary
(
L
RF
M
)
m
o
d
e
l.
T
h
is
p
h
a
se
is
e
m
p
lo
y
e
d
to
d
e
f
in
e
c
h
u
rn
f
o
ll
o
w
e
d
b
y
a
m
u
lt
i
-
c
las
s
p
re
d
ictio
n
p
h
a
se
b
a
se
d
o
n
th
re
e
c
las
sif
ic
a
ti
o
n
tec
h
n
iq
u
e
s: S
im
p
le d
e
c
isio
n
tree
,
A
rti
f
icia
l
n
e
u
ra
l
n
e
tw
o
rk
s an
d
De
c
isio
n
tree
e
n
se
m
b
le,
in
w
h
ich
t
h
e
d
e
p
e
n
d
e
n
t
v
a
riab
le
c
las
sif
ies
a
p
a
rti
c
u
lar
c
u
sto
m
e
r
in
to
a
c
u
sto
m
e
r
c
o
n
ti
n
u
in
g
lo
y
a
l
b
u
y
in
g
p
a
tt
e
rn
s
(No
n
-
c
h
u
rn
e
d
)
,
a
p
a
rti
a
l
d
e
f
e
c
to
r
(P
a
rti
a
ll
y
-
c
h
u
rn
e
d
),
a
n
d
a
to
tal
d
e
f
e
c
to
r
(T
o
tall
y
-
c
h
u
rn
e
d
).
M
a
c
ro
-
a
v
e
ra
g
in
g
m
e
a
su
re
s
in
c
lu
d
in
g
a
v
e
ra
g
e
a
c
c
u
ra
c
y
,
m
a
c
ro
-
a
v
e
ra
g
e
o
f
P
re
c
isio
n
,
Re
c
a
ll
,
a
n
d
F
-
1
a
re
u
se
d
t
o
e
v
a
lu
a
te
c
las
si
f
iers
‟
p
e
r
f
o
r
m
a
n
c
e
o
n
10
-
f
o
ld
c
ro
ss
v
a
li
d
a
ti
o
n
.
Us
in
g
re
a
l
d
a
ta
f
ro
m
a
n
o
n
li
n
e
sto
re
,
th
e
re
su
l
ts
sh
o
w
th
e
e
ff
ici
e
n
c
y
o
f
d
e
c
isio
n
tree
e
n
s
e
m
b
le
m
o
d
e
l
o
v
e
r
th
e
o
th
e
r
m
o
d
e
ls
in
id
e
n
ti
f
y
in
g
b
o
th
f
u
tu
re
p
a
rti
a
l
a
n
d
to
tal
d
e
f
e
c
ti
o
n
.
K
ey
w
o
r
d
:
C
h
u
r
n
p
r
ed
ictio
n
C
lu
s
ter
i
n
g
C
u
s
to
m
er
c
h
u
r
n
E
-
co
m
m
er
ce
L
R
FM
m
o
d
el
P
r
ed
ictio
n
Co
p
y
rig
h
t
©
2
0
1
8
In
stit
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
Al
l
rig
h
ts
re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
A
it
Daq
u
d
R
ac
h
id
,
Dep
ar
t
m
en
t o
f
P
h
y
s
ics,
F
ST
,
Su
lta
n
Mo
u
la
y
Sl
i
m
a
n
e
U
n
i
v
e
r
s
it
y
,
P
B
5
2
3
,
B
en
i M
ellal,
Mo
r
o
cc
o
.
E
m
ail:
d
ao
u
d
.
r
ac
h
id
@
g
m
ail.
c
o
m
1.
I
NT
RO
D
UCT
I
O
N
A
cc
o
r
d
in
g
to
th
e
f
ig
u
r
es
r
elea
s
ed
b
y
t
h
e
I
n
ter
b
an
k
E
lectr
o
n
i
c
B
an
k
i
n
g
C
en
tr
e
(
I
E
B
C
)
,
th
e
m
er
ch
a
n
t
s
ites
h
a
v
e
r
ea
lized
3
.
1
m
illi
o
n
o
n
lin
e
p
a
y
m
en
t
tr
an
s
ac
t
io
n
s
w
it
h
cr
ed
it
ca
r
d
s
f
o
r
a
to
tal
o
f
MA
D
1
.
3
b
illi
o
n
in
2
0
1
7
.
T
h
is
y
ea
r
k
n
e
w
an
i
n
cr
ea
s
e
o
f
+8
1
,
7
%
i
n
n
u
m
b
er
a
n
d
+
5
2
,
3
%
in
t
h
e
a
m
o
u
n
t
s
p
en
t
as
o
p
p
o
s
ed
to
2
0
1
6
[
1
]
.
Mo
r
eo
v
er
,
th
e
g
r
o
w
t
h
o
f
e
-
co
m
m
er
ce
m
ea
n
s
an
i
n
cr
ea
s
e
i
n
o
n
li
n
e
co
m
p
etit
io
n
.
Ho
w
e
v
er
,
w
i
th
t
h
e
ex
p
an
s
io
n
o
f
s
u
c
h
Mo
r
o
cc
an
ac
tiv
e
e
-
co
m
m
er
ce
s
ites
,
it
b
e
co
m
e
s
d
if
f
ic
u
lt
f
o
r
a
co
m
p
an
y
to
d
is
tin
g
u
is
h
it
s
elf
f
r
o
m
o
th
er
co
m
p
eti
to
r
s
–
b
e
th
e
y
n
atio
n
al
o
r
in
ter
n
a
tio
n
al.
T
h
e
g
r
o
w
in
g
n
u
m
b
er
o
f
m
er
ch
an
t
s
ites
h
av
e
p
r
o
v
i
d
ed
cu
s
to
m
er
s
w
it
h
m
o
r
e
ch
o
ice
s
.
I
n
th
e
f
ir
s
t
p
lace
,
it
i
n
cites
t
h
e
m
to
d
i
v
id
e
th
eir
p
u
r
ch
a
s
es
a
m
o
n
g
s
t
s
e
v
er
al
e
-
co
m
m
er
ce
s
ites
.
Nex
t,
it
e
n
ab
les
th
e
m
to
co
m
p
ar
e
as
w
ell
a
s
co
n
tr
ast
c
o
m
p
eti
n
g
p
r
o
d
u
cts
[
2
]
.
Nev
er
th
eles
s
,
th
e
e
-
co
m
m
er
ce
s
ec
to
r
s
u
f
f
er
s
f
r
o
m
t
h
e
k
in
d
o
f
r
elatio
n
s
h
ip
b
et
w
ee
n
co
m
p
a
n
ies
a
n
d
th
e
ir
c
u
s
to
m
e
r
s
(
No
n
-
co
n
tr
ac
t
u
al
s
etti
n
g
)
.
I
n
s
u
c
h
a
ca
s
e,
th
e
cu
s
to
m
er
s
h
av
e
t
h
e
p
o
s
s
ib
ilit
y
to
ch
an
g
e
t
h
eir
p
u
r
ch
a
s
i
n
g
p
a
tter
n
s
w
i
th
o
u
t i
n
f
o
r
m
i
n
g
s
eller
s
.
Face
d
w
ith
th
e
s
e
co
n
s
tr
ain
ts
,
cu
s
to
m
er
s
‟
r
eten
t
i
o
n
n
e
ce
s
s
itates
d
is
allo
w
i
n
g
th
e
m
to
s
w
itch
s
o
m
e
o
f
th
eir
p
u
r
ch
a
s
es
to
an
o
th
er
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2088
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
4
,
A
u
g
u
s
t 2
0
1
8
:
2
3
6
7
–
2
3
8
3
2368
co
n
cu
r
r
en
t.
T
h
is
i
s
a
co
m
m
o
n
ch
alle
n
g
e
f
o
r
an
y
o
n
li
n
e
co
m
p
an
y
.
W
e
th
er
e
f
o
r
e
ad
d
r
ess
th
e
f
o
llo
w
in
g
q
u
e
s
tio
n
to
b
etter
r
etain
cu
s
to
m
er
s
:
Ho
w
ca
n
w
e
a
v
o
id
r
ed
u
cin
g
c
u
s
t
o
m
er
r
ete
n
tio
n
r
ate
?
P
u
t
d
if
f
e
r
en
tl
y
,
h
o
w
ca
n
w
e
r
ed
u
ce
cu
s
to
m
er
ch
u
r
n
i
n
e
-
co
m
m
er
ce
co
n
te
x
t?
A
cc
o
r
d
in
g
to
Nesli
n
et
a
l
.
,
[
3
]
an
d
B
u
re
z
[
4
],
tw
o
b
asi
c
ap
p
r
o
ac
h
es
ex
is
t
f
o
r
r
eso
l
v
in
g
t
h
is
is
s
u
e:
On
th
e
o
n
e
h
an
d
,
„
u
n
tar
g
eted
ap
p
r
o
ac
h
es
‟
t
h
at
r
el
y
o
n
s
u
p
er
io
r
p
r
o
d
u
ct
an
d
m
as
s
a
d
v
er
ti
s
i
n
g
to
i
n
cr
ea
s
e
b
r
an
d
lo
y
alt
y
a
n
d
r
etain
c
u
s
to
m
er
s
.
A
g
o
o
d
ex
a
m
p
le
o
f
th
is
i
s
A
O
L
‟
s
e
f
f
o
r
ts
to
d
ec
r
ea
s
e
ch
u
r
n
th
r
o
u
g
h
b
etter
s
o
f
t
w
ar
e
an
d
co
n
ten
t
[
5
]
.
On
t
h
e
o
th
er
h
a
n
d
,
„
tar
g
ete
d
ap
p
r
o
ac
h
es‟
th
a
t
r
el
y
o
n
id
en
ti
f
y
in
g
p
o
ten
tia
l
ch
u
r
n
er
s
in
o
r
d
er
to
a
v
o
id
d
ef
ec
tio
n
b
y
tar
g
eti
n
g
s
u
c
h
cu
s
to
m
er
s
w
i
th
d
ir
ec
t in
ce
n
ti
v
es [
4
]
,
[6
-
9]
.
I
n
th
i
s
s
tu
d
y
,
w
e
ar
e
co
n
ce
r
n
ed
w
it
h
t
h
e
s
ec
o
n
d
ap
p
r
o
ac
h
.
Fo
r
th
at
alo
n
e,
w
e
i
n
v
e
s
ti
g
ate
w
h
e
th
er
o
r
n
o
t
w
e
ar
e
ab
le
to
id
en
tify
t
h
e
m
o
m
en
t
w
h
e
n
cu
s
to
m
er
s
b
eg
in
to
d
is
co
n
t
in
u
e
th
e
ir
r
elatio
n
s
h
ip
w
it
h
e
-
co
m
m
er
ce
w
eb
s
ite
in
o
r
d
er
to
tar
g
et
th
e
m
b
y
r
ete
n
tio
n
p
r
o
g
r
a
m
s
to
av
o
id
th
eir
to
tal
d
ef
ec
tio
n
.
C
u
s
to
m
er
r
elatio
n
s
h
ip
m
a
n
a
g
e
m
en
t,
an
d
cu
s
to
m
er
ch
u
r
n
p
r
ed
ictio
n
in
p
ar
ticu
lar
,
h
av
e
r
ec
eiv
ed
a
g
r
o
w
i
n
g
atten
tio
n
d
u
r
i
n
g
t
h
e
l
ast
d
ec
ad
e.
T
ab
le
1
s
u
m
m
ar
izes
cu
s
to
m
er
ch
u
r
n
p
r
ed
ictio
n
m
o
d
el
s
r
ep
o
r
ted
in
th
e
l
iter
atu
r
e
in
r
ec
en
t
y
ea
r
s
.
T
h
e
d
is
tin
ct
iv
e
ch
ar
ac
ter
i
s
tics
o
f
ea
c
h
s
tu
d
y
i
n
ter
m
s
o
f
t
h
e
s
ec
to
r
s
,
en
v
ir
o
n
m
en
t
s
etti
n
g
s
,
d
ef
ec
t
io
n
t
y
p
es,
an
d
ch
u
r
n
d
ef
i
n
itio
n
s
ar
e
p
r
o
v
id
ed
.
As
ca
n
b
e
s
ee
n
f
r
o
m
T
ab
le
1
,
th
er
e
ar
e
t
w
o
m
aj
o
r
r
em
ar
k
s
:
(
1
)
E
n
v
ir
o
n
m
en
t
s
etti
n
g
s
:
g
r
ea
t
n
u
m
b
er
o
f
s
tu
d
ie
s
ar
e
in
th
e
co
n
tr
ac
tu
al
s
ett
in
g
,
th
a
t
is
ch
ar
ac
ter
ized
b
y
t
h
e
ex
is
te
n
c
e
o
f
a
co
n
tr
ac
t
b
et
w
ee
n
th
e
f
i
r
m
an
d
th
e
c
u
s
to
m
er
,
in
s
u
c
h
a
ca
s
e,
th
e
d
ate
o
f
ch
u
r
n
is
clea
r
l
y
k
n
o
w
n
,
a
n
d
it
m
atc
h
es
u
p
w
i
th
t
h
e
co
n
tr
ac
t
ca
n
ce
llatio
n
d
ate.
(
2
)
P
ar
tial
o
r
to
tal
d
ef
ec
tio
n
:
m
o
s
t
o
f
th
o
s
e
s
t
u
d
ies
co
n
s
id
er
to
tal
d
ef
ec
tio
n
,
w
h
ils
t
o
n
l
y
f
e
w
s
t
u
d
ies
u
s
e
p
r
ed
ictio
n
m
o
d
els
to
id
en
ti
f
y
p
ar
tial d
ef
ec
tio
n
[
6
]
,
[
10
]
-
[
12
]
.
Mo
r
eo
v
er
,
ea
ch
o
f
t
h
o
s
e
s
t
u
d
ies d
ef
i
n
es
cu
s
to
m
er
c
h
u
r
n
d
i
f
f
er
en
tl
y
,
t
h
i
s
r
aise
s
th
e
f
o
llo
w
i
n
g
q
u
est
io
n
: W
h
ic
h
o
n
e
is
m
o
r
e
ap
p
r
o
p
r
iate
?
T
ab
le
1
r
ev
ea
ls
th
at
t
h
e
c
h
u
r
n
is
s
u
e
h
a
s
b
ee
n
u
n
d
er
-
r
e
s
ea
r
ch
ed
in
th
e
e
-
co
m
m
er
ce
s
ec
to
r
.
Mo
r
eo
v
er
,
all
an
al
y
s
e
s
in
t
h
i
s
s
ec
to
r
co
n
s
id
er
to
tal
d
ef
ec
tio
n
(
d
ef
ec
t
io
n
co
lu
m
n
)
.
T
o
d
is
co
v
er
b
o
t
h
p
ar
tial
an
d
to
tal
d
ef
ec
tio
n
i
n
e
-
co
m
m
er
ce
s
ec
t
o
r
,
th
is
s
tu
d
y
co
n
tr
ib
u
tes
to
th
e
ex
ta
n
t
liter
at
u
r
e
i
n
t
w
o
i
m
p
o
r
tan
t
w
a
y
s
.
Fir
s
t,
it
co
m
b
i
n
es
L
R
F
M
m
o
d
el
a
n
d
c
lu
s
ter
i
n
g
tec
h
n
iq
u
es
d
u
r
in
g
a
ca
lib
r
at
io
n
p
er
io
d
(
T
1
)
to
s
eg
m
en
t
all
c
u
s
to
m
er
s
in
to
h
o
m
o
g
e
n
eo
u
s
cl
u
s
ter
s
,
th
en
an
L
R
FM
p
atter
n
w
il
l
b
e
ass
i
g
n
ed
to
ea
ch
cl
u
s
ter
[
1
3
]
.
C
h
an
g
e
i
n
th
e
L
R
FM
p
atter
n
(
Mo
v
in
g
a
cu
s
to
m
er
f
r
o
m
o
n
e
clu
s
ter
w
it
h
an
i
m
p
o
r
t
an
t
v
a
lu
e
i
n
T
1
to
an
o
th
er
g
r
o
u
p
o
f
les
s
v
al
u
e
i
n
p
r
ed
ictio
n
p
er
io
d
(
T
2
)
)
m
a
y
b
e
a
p
ar
tial
o
r
to
tal
d
ef
ec
tio
n
s
i
g
n
a
l.
Seco
n
d
,
it
i
n
tr
o
d
u
ce
s
cla
s
s
i
f
icatio
n
tech
n
iq
u
es
f
o
r
b
u
ild
in
g
p
r
ed
ictio
n
m
o
d
els
to
p
r
ed
ict
b
o
th
p
ar
tial
an
d
to
tal
d
ef
ec
tio
n
in
o
r
d
er
to
m
in
i
m
ize
th
e
r
is
k
o
f
c
h
u
r
n
.
On
t
h
e
o
th
er
h
a
n
d
,
co
n
tr
ar
y
t
o
r
esear
ch
th
a
t
s
ee
k
s
to
r
etai
n
o
n
l
y
p
r
o
f
i
tab
le
cu
s
to
m
er
s
[
6
]
,
[
7
]
,
[
1
4
]
,
[
1
5
]
o
r
th
o
s
e
th
at
s
p
e
n
d
m
an
y
e
f
f
o
r
ts
f
o
r
th
e
e
n
tire
c
u
s
to
m
er
b
ase
[
9
]
,
[
1
6
]
,
[
1
7
]
,
o
u
r
s
tu
d
y
i
s
ce
n
ter
ed
n
o
t
o
n
l
y
o
n
t
h
e
cu
s
to
m
er
s
w
h
o
b
elo
n
g
to
th
e
cl
u
s
ter
s
r
ep
r
esen
tin
g
th
e
co
r
e
cu
s
to
m
er
s
,
b
u
t
also
o
n
th
o
s
e
w
h
o
d
em
o
n
s
tr
ate
p
o
s
iti
v
e
ch
a
n
g
e
in
th
e
ir
p
u
r
ch
ase
b
eh
a
v
io
r
ev
en
i
f
t
h
e
y
ar
e
g
r
o
u
p
ed
in
c
lu
s
ter
s
th
at
d
o
n
o
t
co
n
tr
ib
u
te
p
o
s
iti
v
el
y
to
p
r
o
f
it
s
.
T
h
e
cr
ea
tio
n
o
f
a
r
eten
tio
n
p
r
o
g
r
a
m
t
h
at
tar
g
ets
all
t
y
p
es
o
f
cu
s
to
m
er
s
w
ill
b
e
v
er
y
co
s
t
l
y
f
o
r
th
e
co
m
p
a
n
y
.
B
y
ad
o
p
tin
g
a
m
e
th
o
d
th
at
f
o
c
u
s
e
s
o
n
l
y
o
n
p
r
o
f
itab
le
cu
s
to
m
er
s
,
co
m
p
a
n
ie
s
,
esp
ec
iall
y
t
h
o
s
e
w
o
r
k
i
n
g
in
e
-
co
m
m
er
ce
f
iel
d
,
ca
n
lo
s
e
s
o
m
e
cu
s
to
m
er
s
.
T
h
is
co
u
ld
b
e
ascr
ib
ed
to
th
e
lack
o
f
t
h
eir
en
g
a
g
e
m
en
t
w
it
h
t
h
e
b
en
e
f
ici
ar
ies
o
f
t
h
e
r
eten
tio
n
p
r
o
g
r
am
s
,
w
h
ic
h
w
ill
lead
to
in
cr
ea
s
ed
cu
s
to
m
er
ch
u
r
n
r
ate
f
o
llo
w
ed
b
y
a
d
ec
r
ea
s
e
i
n
p
r
o
f
its
.
T
h
ese
c
u
s
to
m
er
s
r
e
all
y
d
eser
v
e
a
tten
tio
n
f
r
o
m
t
h
e
co
m
p
a
n
y
;
s
o
th
e
y
s
h
o
u
ld
n
o
t
b
e
eli
m
i
n
ated
,
b
u
t
th
e
y
s
h
o
u
ld
b
e
p
lace
d
in
a
n
o
th
er
ca
te
g
o
r
y
.
T
h
is
is
an
i
m
p
o
r
tan
t
p
o
in
t
b
ec
au
s
e
n
o
co
m
p
an
y
w
a
n
t
s
to
m
is
s
t
h
e
o
p
p
o
r
tu
n
it
y
o
f
co
n
v
er
ti
n
g
a
p
r
ev
io
u
s
l
y
d
is
s
ati
s
f
ied
cu
s
to
m
er
in
to
a
lo
y
al
cu
s
to
m
er
.
T
h
ese
cu
s
to
m
er
s
ar
e
th
o
s
e
t
h
at
d
e
m
o
n
s
tr
ate
p
o
s
iti
v
e
ch
a
n
g
e
i
n
t
h
eir
p
u
r
ch
a
s
e
b
e
h
av
io
r
e
v
en
if
th
e
y
ar
e
g
r
o
u
p
ed
in
cl
u
s
t
er
s
th
at
d
o
n
o
t
co
n
tr
ib
u
te
p
o
s
iti
v
el
y
to
p
r
o
f
its
.
T
h
e
id
en
ti
f
icatio
n
o
f
t
h
ese
c
u
s
to
m
er
s
w
i
ll
b
e
d
is
cu
s
s
ed
i
n
th
e
f
o
llo
w
in
g
s
ec
tio
n
s
.
Fo
r
ex
a
m
p
le,
in
a
s
itu
atio
n
w
h
er
e
th
e
g
o
al
o
f
a
co
m
p
an
y
is
to
r
etain
o
n
ly
p
r
o
f
itab
le
cu
s
t
o
m
er
s
,
th
e
co
m
p
a
n
y
s
h
o
u
ld
d
is
co
v
er
w
h
y
cu
s
to
m
er
s
leav
e
a
n
d
g
o
to
co
m
p
eti
to
r
s
.
A
c
h
u
r
n
a
n
al
y
s
i
s
f
o
r
th
eir
p
r
o
f
itab
le
cu
s
to
m
er
‟
s
s
eg
m
e
n
t
s
h
o
w
s
t
h
at
s
o
m
e
cu
s
to
m
er
s
lea
v
e
th
e
e
-
co
m
m
er
ce
w
eb
s
ite
b
ec
au
s
e
d
eliv
er
y
c
h
ar
g
e
s
ar
e
n
o
t
f
r
ee
.
S
u
b
s
eq
u
e
n
tl
y
,
t
h
e
c
o
m
p
a
n
y
d
ec
id
es
to
r
ed
u
ce
d
eliv
er
y
co
s
ts
f
o
r
th
e
m
o
s
t
p
r
o
f
itab
le
c
u
s
to
m
er
in
o
r
d
er
to
r
etain
th
e
m
.
Ho
w
e
v
e
r
,
th
e
les
s
p
r
o
f
itab
le
cu
s
to
m
er
s
ar
e
n
o
t
s
er
v
ed
w
it
h
t
h
is
r
ed
u
ce
;
o
n
l
y
p
r
o
f
itab
le
cu
s
to
m
er
s
ar
e
s
at
is
f
ied
.
T
h
er
ef
o
r
e,
tar
g
etin
g
o
n
l
y
p
r
o
f
it
ab
le
cu
s
to
m
er
s
is
n
o
t
a
n
o
p
ti
m
al
s
tr
ateg
y
f
o
r
in
cr
ea
s
i
n
g
r
ete
n
tio
n
r
ate
b
ec
au
s
e
a
g
r
o
u
p
o
f
cu
s
to
m
er
s
w
a
s
p
r
o
f
itab
le
in
th
e
p
ast,
d
o
esn
‟
t
m
ea
n
it
w
il
l
co
n
tin
u
e
to
b
e
s
o
in
th
e
f
u
tu
r
e
[
18
].
T
h
e
r
est
o
f
t
h
i
s
p
ap
er
is
o
r
g
a
n
ized
b
y
4
o
t
h
er
s
ec
t
io
n
s
,
th
e
r
esear
ch
m
et
h
o
d
in
cl
u
d
in
g
s
eg
m
e
n
ti
n
g
m
et
h
o
d
s
an
d
d
ata
m
in
i
n
g
t
ec
h
n
iq
u
es
u
s
ed
in
t
h
is
s
tu
d
y
ar
e
b
r
iefly
d
escr
ib
ed
in
Sectio
n
2
f
o
llo
w
ed
b
y
a
n
e
m
p
ir
ical
s
tu
d
y
i
n
Sectio
n
3
t
o
d
e
m
o
n
s
tr
ate
h
o
w
th
is
p
r
ed
i
ctio
n
ap
p
r
o
ac
h
w
o
r
k
s
i
n
p
r
ac
t
ice,
w
h
er
ea
s
i
n
t
h
e
f
o
u
r
t
h
s
ec
tio
n
w
e
d
is
c
u
s
s
t
h
e
r
esu
lt
s
.
T
h
e
p
ap
er
f
i
n
is
h
es
w
i
th
t
h
e
co
n
c
lu
s
io
n
,
li
m
itatio
n
a
n
d
s
o
m
e
i
s
s
u
es
f
o
r
f
u
tu
r
e
r
esear
ch
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
C
lu
s
ter
in
g
P
r
ed
ictio
n
Tech
n
iq
u
es in
Defin
in
g
a
n
d
P
r
ed
ictin
g
C
u
s
to
mers
Defe
ct
io
n
:
…
(
A
it Da
q
u
d
R
a
ch
id
)
2369
T
ab
le
1
.
L
iter
atu
r
e
R
ev
ie
w
C
h
u
r
n
d
e
f
i
n
i
t
i
o
n
C
u
s
t
o
mer s
h
i
f
t
t
r
a
n
s
a
c
t
i
o
n
p
a
t
t
e
r
n
.
A
c
h
u
r
n
e
r
i
s
d
e
f
i
n
e
d
a
s a
su
b
s
c
r
i
b
e
r
w
h
o
i
s
v
o
l
u
n
t
a
r
y
t
o
l
e
a
v
e
C
h
a
n
g
e
i
n
c
u
s
t
o
mer'
s s
t
a
t
u
s
.
S
u
b
s
c
r
i
b
e
r
d
o
e
s
n
o
t
r
e
n
e
w
o
r
p
a
y
.
N
o
t
m
e
n
t
i
o
n
e
d
.
C
u
s
t
o
mer s
t
o
p
s
d
o
i
n
g
b
u
s
i
n
e
ss.
C
u
s
t
o
mers
w
h
o
,
f
r
o
m
a
c
e
r
t
a
i
n
p
e
r
i
o
d
,
d
i
d
n
o
t
b
u
y
a
n
y
t
h
i
n
g
o
r
t
h
o
se
w
h
o
i
n
a
l
l
su
b
s
e
q
u
e
n
t
p
e
r
i
o
d
s
s
p
e
n
t
l
e
ss t
h
a
n
4
0
%
o
f
t
h
e
a
m
o
u
n
t
s
p
e
n
t
i
n
t
h
e
r
e
f
e
r
e
n
c
e
p
e
r
i
o
d
.
N
o
t
m
e
n
t
i
o
n
e
d
.
G
a
mb
l
e
r
d
o
e
s
n
o
t
p
l
a
y
d
u
r
i
n
g
a
p
e
r
i
o
d
.
S
e
t
o
f
d
e
f
i
n
i
t
i
o
n
s
o
f
c
h
u
r
n
.
U
ser l
e
a
v
e
s
c
o
m
p
a
n
y
.
I
n
a
c
t
i
v
i
t
y
i
n
s
e
c
o
n
d
h
a
l
f
o
f
t
h
e
y
e
a
r
.
C
u
s
t
o
mers
w
h
o
q
u
i
t
t
h
e
s
e
r
v
i
c
e
o
r
t
r
a
n
sf
e
r
t
o
c
o
mp
e
t
i
t
o
r
s a
t
a
d
a
t
e
c
l
o
se
t
o
a
s
p
e
c
i
f
i
c
d
a
t
e
.
S
h
i
f
t
i
n
g
l
o
y
a
l
t
i
e
s fr
o
m
o
n
e
ser
v
i
c
e
p
r
o
v
i
d
e
r
t
o
a
n
o
t
h
e
r
C
u
s
t
o
mer
d
e
c
l
a
r
e
d
c
h
u
r
n
e
d
i
f
h
e
r
p
o
r
t
f
o
l
i
o
s
i
z
e
f
a
l
l
s
b
e
l
o
w
a
sp
e
c
i
f
i
c
t
h
r
e
s
h
o
l
d
v
a
l
u
e
a
n
d
s
t
a
y
s
t
h
a
t
w
a
y
f
o
r
si
x
c
o
n
sec
u
t
i
v
e
mo
n
t
h
s.
C
u
s
t
o
mer
w
i
t
h
f
e
w
s
e
ssi
o
n
s
d
u
r
i
n
g
t
h
e
8
–
1
2
m
o
n
t
h
s
a
f
t
e
r
r
e
g
i
s
t
r
a
t
i
o
n
.
W
h
e
n
t
h
e
c
o
m
p
a
n
y
h
a
s
b
e
e
n
a
c
t
i
v
e
(
i
.
e
.
h
a
s
a
t
l
e
a
s
t
o
n
e
t
r
a
n
s
a
c
t
i
o
n
i
n
t
h
e
y
e
a
r
)
b
u
t
h
a
s
n
o
a
c
t
i
v
i
t
y
(
i
.
e
.
p
u
r
c
h
a
s
e
)
i
n
t
h
e
y
e
a
r
.
C
u
s
t
o
mer
w
i
t
h
t
h
e
i
n
s
u
r
e
r
a
t
t
h
e
s
t
a
r
t
o
f
t
h
e
y
e
a
r
b
u
t
i
s
n
o
l
o
n
g
e
r
a
t
t
h
e
e
n
d
o
f
t
h
e
y
e
a
r
.
C
u
s
t
o
mer s
h
i
f
t
s L
R
F
M
p
a
t
t
e
r
n
,
c
h
a
n
g
e
s i
n
p
u
r
c
h
a
se
b
e
h
a
v
i
o
r
.
D
e
f
e
c
t
i
o
n
P
a
r
t
i
a
l
To
t
a
l
P
a
r
t
i
a
l
a
n
d
To
t
a
l
To
t
a
l
To
t
a
l
To
t
a
l
P
a
r
t
i
a
l
To
t
a
l
To
t
a
l
P
a
r
t
i
a
l
To
t
a
l
To
t
a
l
To
t
a
l
To
t
a
l
To
t
a
l
To
t
a
l
To
t
a
l
To
t
a
l
P
a
r
t
i
a
l
a
n
d
To
t
a
l
En
v
i
r
o
n
m
e
n
t
s
e
t
t
i
n
g
O
f
f
l
i
n
e
e
n
v
i
r
o
n
m
e
n
t
/
N
o
n
-
c
o
n
t
r
a
c
t
u
a
l
se
t
t
i
n
g
O
f
f
l
i
n
e
e
n
v
i
r
o
n
m
e
n
t
/
C
o
n
t
r
a
c
t
u
a
l
s
e
t
t
i
n
g
O
f
f
l
i
n
e
e
n
v
i
r
o
n
m
e
n
t
/
C
o
n
t
r
a
c
t
u
a
l
s
e
t
t
i
n
g
O
f
f
l
i
n
e
e
n
v
i
r
o
n
m
e
n
t
/
C
o
n
t
r
a
c
t
u
a
l
s
e
t
t
i
n
g
O
f
f
l
i
n
e
e
n
v
i
r
o
n
m
e
n
t
/
C
o
n
t
r
a
c
t
u
a
l
s
e
t
t
i
n
g
O
n
l
i
n
e
e
n
v
i
r
o
n
m
e
n
t
/
N
o
n
-
c
o
n
t
r
a
c
t
u
a
l
set
t
i
n
g
O
f
f
l
i
n
e
e
n
v
i
r
o
n
m
e
n
t
/
N
o
n
-
c
o
n
t
r
a
c
t
u
a
l
se
t
t
i
n
g
O
f
f
l
i
n
e
e
n
v
i
r
o
n
m
e
n
t
/
C
o
n
t
r
a
c
t
u
a
l
s
e
t
t
i
n
g
O
n
l
i
n
e
e
n
v
i
r
o
n
m
e
n
t
/
N
o
n
-
c
o
n
t
r
a
c
t
u
a
l
set
t
i
n
g
O
f
f
l
i
n
e
e
n
v
i
r
o
n
m
e
n
t
/
N
o
n
-
c
o
n
t
r
a
c
t
u
a
l
se
t
t
i
n
g
O
f
f
l
i
n
e
e
n
v
i
r
o
n
m
e
n
t
/
C
o
n
t
r
a
c
t
u
a
l
s
e
t
t
i
n
g
O
n
l
i
n
e
e
n
v
i
r
o
n
m
e
n
t
/
N
o
n
-
c
o
n
t
r
a
c
t
u
a
l
set
t
i
n
g
O
f
f
l
i
n
e
e
n
v
i
r
o
n
m
e
n
t
/
C
o
n
t
r
a
c
t
u
a
l
s
e
t
t
i
n
g
O
f
f
l
i
n
e
e
n
v
i
r
o
n
m
e
n
t
/
C
o
n
t
r
a
c
t
u
a
l
s
e
t
t
i
n
g
O
f
f
l
i
n
e
e
n
v
i
r
o
n
m
e
n
t
/
C
o
n
t
r
a
c
t
u
a
l
s
e
t
t
i
n
g
O
n
l
i
n
e
e
n
v
i
r
o
n
m
e
n
t
/
N
o
n
-
c
o
n
t
r
a
c
t
u
a
l
set
t
i
n
g
O
n
l
i
n
e
e
n
v
i
r
o
n
m
e
n
t
/
N
o
n
-
c
o
n
t
r
a
c
t
u
a
l
set
t
i
n
g
O
f
f
l
i
n
e
e
n
v
i
r
o
n
m
e
n
t
/
C
o
n
t
r
a
c
t
u
a
l
s
e
t
t
i
n
g
O
n
l
i
n
e
e
n
v
i
r
o
n
m
e
n
t
/
N
o
n
-
c
o
n
t
r
a
c
t
u
a
l
set
t
i
n
g
S
e
c
t
o
r
s
R
e
t
a
i
l
i
n
g
Te
l
e
c
o
mm
u
n
i
c
a
t
i
o
n
s
Te
l
e
c
o
mm
u
n
i
c
a
t
i
o
n
s
P
a
y
-
TV
s
u
b
scri
p
t
i
o
n
F
i
n
a
n
c
i
a
l
s
e
r
v
i
c
e
s
E
-
c
o
mm
e
r
c
e
R
e
t
a
i
l
i
n
g
Te
l
e
c
o
mm
u
n
i
c
a
t
i
o
n
s
O
n
l
i
n
e
g
a
m
b
l
i
n
g
R
e
t
a
i
l
i
n
g
Te
l
e
c
o
mm
u
n
i
c
a
t
i
o
n
s
E
-
c
o
mm
e
r
c
e
(
B
2
B
)
Te
l
e
c
o
mm
u
n
i
c
a
t
i
o
n
s
F
i
n
a
n
c
i
a
l
s
e
r
v
i
c
e
s
F
i
n
a
n
c
i
a
l
s
e
r
v
i
c
e
s
E
-
c
o
mm
e
r
c
e
E
-
c
o
mm
e
r
c
e
(
B
2
B
)
F
i
n
a
n
c
i
a
l
s
e
r
v
i
c
e
s
E
-
c
o
mm
e
r
c
e
Li
t
e
r
a
t
u
r
e
r
e
v
i
e
w
B
u
c
k
i
n
x
a
n
d
v
a
n
d
e
n
P
o
e
l
(
2
0
0
5
)
[
6
]
S
h
i
n
-
Y
u
a
n
H
u
n
g
,
e
t
al
.
(
2
0
0
6
)
[
1
4
]
Jae
-
H
y
e
o
n
A
h
n
a
,
e
t
al
.
(
2
0
0
6
)
[
1
2
]
B
u
r
e
z
a
n
d
v
a
n
d
e
n
P
o
e
l
(
2
0
0
7
)
[
4
]
X
i
e
,
e
t
a
l
.
(
2
0
0
9
)
[
1
9
]
X
i
a
o
b
i
n
g
Y
u
,
et
al
.
(
2
0
1
1
)
[
1
7
]
V
.
L.
M
i
g
u
é
i
s,
e
t
a
l
.
(
2
0
1
2
)
[
1
0
]
B
i
n
g
q
u
a
n
H
u
a
n
g
,
e
t
e
l
.
(
2
0
1
2
)
[
2
0
]
K
r
i
st
o
f
C
o
u
ss
e
me
n
t
,
e
t
a
l
.
(
2
0
1
3
)
[
5
7
]
M
.
C
l
e
me
n
t
e
-
C
í
sc
a
r
,
a
l
.
(
2
0
1
4
)
[
1
1
]
F
a
r
i
s (2
0
1
4
)
[
2
1
]
A
.
T.
Ja
h
r
o
mi
,
e
t
a
l
.
(
2
0
1
4
)
[
7
]
K
.
K
y
o
u
n
g
o
k
e
t
a
l
.
(
2
0
1
4
)
[
2
2
]
M
.
A
.
H
.
F
a
r
q
u
a
d
,
e
t
a
l
(
2
0
1
4
)
[
1
5
]
O
z
d
e
n
G
u
r
A
l
i
a
n
d
U
mu
t
A
r
ı
t
u
r
k
(
2
0
1
4
)
[
8
]
S
su
-
H
a
n
C
h
e
n
(
2
0
1
6
)
[
2
3
]
N
i
c
c
o
l
ò
G
o
r
d
i
n
i
a
,
V
a
l
e
r
i
o
V
e
g
l
i
o
b
(
2
0
1
7
)
[
9
]
N
.
H
o
l
t
r
o
p
,
e
t
a
l
.
(
2
0
1
7
)
[
2
4
]
Th
i
s
st
u
d
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2088
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
4
,
A
u
g
u
s
t 2
0
1
8
:
2
3
6
7
–
2
3
8
3
2370
2.
RE
S
E
ARCH
M
E
T
H
O
D
T
h
e
p
u
r
p
o
s
e
o
f
th
is
s
t
u
d
y
is
t
o
b
u
ild
a
cu
s
to
m
er
‟
s
c
h
u
r
n
p
r
ed
ictio
n
m
o
d
el
in
e
-
co
m
m
er
c
e
s
ec
to
r
b
y
u
s
i
n
g
clu
s
ter
i
n
g
a
n
d
p
r
ed
ictio
n
tech
n
iq
u
e
s
to
p
r
ed
ict
th
o
s
e
cu
s
to
m
er
s
w
h
o
ar
e
lik
el
y
t
o
ch
u
r
n
in
t
h
e
n
ea
r
f
u
tu
r
e
i
n
o
r
d
er
to
m
in
i
m
ize
th
e
r
is
k
o
f
c
h
u
r
n
.
2
.
1
.
Cus
t
o
m
er
p
ro
f
ilin
g
Ma
r
k
et
s
e
g
m
en
tatio
n
is
t
h
e
p
r
o
ce
s
s
o
f
id
e
n
ti
f
y
i
n
g
k
e
y
g
r
o
u
p
s
w
it
h
i
n
t
h
e
g
e
n
er
al
m
ar
k
et
th
at
s
h
ar
e
s
p
ec
if
ic
ch
ar
ac
ter
i
s
tics
a
n
d
co
n
s
u
m
in
g
h
ab
it
s
[
2
5
]
.
R
FM
m
o
d
el,
w
h
ic
h
w
as
p
r
o
p
o
s
ed
b
y
Hu
g
h
es
(
1
9
9
4
)
[
2
6
]
,
is
o
n
e
o
f
th
e
m
o
s
t
co
m
m
o
n
m
eth
o
d
s
f
o
r
s
eg
m
e
n
ti
n
g
a
n
d
id
en
ti
f
y
in
g
cu
s
to
m
er
v
a
lu
e
s
in
co
m
p
a
n
ie
s
.
C
l
u
s
ter
in
g
tech
n
iq
u
es
h
a
v
e
b
ee
n
w
id
el
y
u
s
ed
to
s
eg
m
e
n
t
c
u
s
to
m
er
s
w
h
en
u
s
in
g
R
FM
m
o
d
el
[
1
3
]
,
[
2
5
]
,
[
2
7
]
-
[
29]
.
I
n
th
is
s
ec
tio
n
,
w
e
d
is
cu
s
s
k
-
m
ea
n
s
as
clu
s
ter
i
n
g
tech
n
iq
u
e
a
n
d
th
e
L
R
F
M
m
o
d
el
as
th
e
ex
ten
d
ed
v
er
s
io
n
o
f
R
FM
m
o
d
el
t
h
at
co
n
s
i
d
er
cu
s
to
m
er
r
elatio
n
s
h
ip
len
g
t
h
(
L
)
w
e
u
s
e
f
o
r
cu
s
to
m
er
p
r
o
f
ilin
g
tas
k
.
2
.
1
.
1
.
RF
M
a
nd
L
RF
M
m
o
de
ls
R
FM
m
o
d
el
i
s
an
e
f
f
ec
ti
v
e
m
e
th
o
d
o
f
s
eg
m
e
n
ti
n
g
an
d
it
is
l
i
k
e
w
is
e
a
b
eh
a
v
io
r
al
an
a
l
y
s
is
t
h
at
ca
n
b
e
e
m
p
lo
y
ed
f
o
r
m
ar
k
et
s
e
g
m
e
n
tatio
n
[
30
]
,
[
31
]
.
A.
Hu
g
h
es
[
30
]
d
escr
ib
es
th
at
th
e
m
ai
n
ass
et
o
f
th
e
R
FM
m
et
h
o
d
is
,
o
n
th
e
o
n
e
h
an
d
,
to
o
b
tain
cu
s
to
m
er
s
‟
b
eh
av
io
r
al
an
al
y
s
is
i
n
o
r
d
er
t
o
g
r
o
u
p
th
e
m
in
to
h
o
m
o
g
en
eo
u
s
cl
u
s
ter
s
,
an
d
,
o
n
t
h
e
o
th
er
h
an
d
,
to
d
ev
elo
p
a
m
ar
k
et
in
g
p
la
n
tailo
r
ed
to
ea
ch
s
p
ec
i
f
ic
m
ar
k
e
t
s
eg
m
e
n
t.
R
FM
a
n
al
y
s
is
i
m
p
r
o
v
es
th
e
m
ar
k
et
s
e
g
m
en
tatio
n
b
y
ex
a
m
i
n
in
g
t
h
e
w
h
e
n
(
r
ec
en
c
y
)
,
h
o
w
o
f
te
n
(
f
r
eq
u
en
c
y
)
,
an
d
th
e
m
o
n
e
y
s
p
en
t
(
m
o
n
etar
y
)
i
n
a
p
ar
ticu
la
r
ite
m
o
r
s
er
v
ice
[
32
]
.
A
.
Ya
n
g
[
32
]
s
u
m
m
ar
ized
th
at
c
u
s
to
m
er
s
w
h
o
h
ad
b
o
u
g
h
t
m
o
s
t
r
ec
e
n
tl
y
,
m
o
s
t
f
r
eq
u
en
t
l
y
,
a
n
d
h
ad
s
p
en
t
t
h
e
m
o
s
t
m
o
n
e
y
w
o
u
ld
b
e
m
u
c
h
m
o
r
e
li
k
el
y
to
r
ea
ct
to
th
e
f
u
tu
r
e
p
r
o
m
o
tio
n
s
.
So
m
e
r
esear
ch
er
s
tr
y
to
d
ev
elo
p
n
e
w
R
FM
m
o
d
els
b
y
ad
d
in
g
s
o
m
e
ad
d
itio
n
al
p
ar
am
eter
s
to
it
s
o
as
to
ex
a
m
in
e
w
h
et
h
er
th
e
y
ac
h
ie
v
e
g
o
o
d
r
esu
lt
s
t
h
an
t
h
e
b
asi
c
R
FM
m
o
d
el
o
r
n
o
t
[
3
3
]
-
[
35]
.
Fo
r
ex
am
p
le
,
C
h
an
g
an
d
T
s
a
y
[
3
6
]
p
r
o
p
o
s
e
th
e
L
R
FM
m
o
d
el,
b
y
tak
in
g
th
e
cu
s
to
m
er
r
elatio
n
len
g
th
i
n
to
ac
co
u
n
t,
in
o
r
d
er
to
r
eso
lv
e
R
FM
m
o
d
el
p
r
o
b
lem
r
elate
d
to
th
e
d
if
f
ic
u
lt
y
o
f
d
is
tin
g
u
is
h
i
n
g
b
et
w
ee
n
cu
s
to
m
er
s
,
w
h
o
h
a
v
e
lo
n
g
-
ter
m
o
r
s
h
o
r
t
-
ter
m
r
ela
tio
n
s
h
ip
s
w
it
h
th
e
co
m
p
an
y
.
I
n
ad
d
itio
n
,
S.
C
h
o
w
an
d
R
.
H
o
ld
en
[
3
7
]
s
u
g
g
e
s
t
th
a
t
th
e
cu
s
to
m
er
‟
s
lo
y
alt
y
an
d
p
r
o
f
itab
ilit
y
d
ep
en
d
o
n
th
e
r
elatio
n
s
h
ip
b
et
w
ee
n
a
co
m
p
an
y
a
n
d
its
cu
s
to
m
er
s
.
I
n
t
h
is
r
eg
ar
d
,
in
o
r
d
er
t
o
id
en
tify
m
o
s
t
lo
y
al
c
u
s
t
o
m
er
s
,
it
is
n
ec
es
s
ar
y
to
co
n
s
id
er
th
e
c
u
s
to
m
er
‟
s
r
elatio
n
le
n
g
th
(
L
)
,
w
h
er
e
L
i
s
d
ef
i
n
ed
as
th
e
n
u
m
b
er
o
f
ti
m
e
p
er
io
d
s
(
s
u
c
h
as d
a
y
s
)
f
r
o
m
t
h
e
f
ir
s
t p
u
r
ch
a
s
e
to
th
e
last
p
u
r
ch
ase
i
n
th
e
d
atab
ase.
2
.
1
.
2
.
K
-
m
ea
ns
m
et
ho
d
K
-
m
ea
n
s
cl
u
s
ter
i
n
g
is
th
e
m
o
s
t
co
m
m
o
n
al
g
o
r
ith
m
u
s
ed
to
c
lu
s
ter
n
v
ec
to
r
s
b
ased
o
n
at
tr
i
b
u
tes
i
n
to
k
p
ar
titi
o
n
s
,
w
h
er
e
k
<
n
,
d
e
p
en
d
in
g
on
s
o
m
e
m
ea
s
u
r
es.
T
h
e
n
a
m
e
co
m
e
s
f
r
o
m
th
e
f
a
ct
th
at
k
cl
u
s
ter
s
ar
e
id
en
ti
f
ied
,
an
d
t
h
e
ce
n
ter
o
f
a
clu
s
ter
is
t
h
e
m
ea
n
o
f
all
v
ec
to
r
s
w
it
h
in
t
h
is
clu
s
ter
.
T
h
e
al
g
o
r
ith
m
s
tar
ts
w
it
h
ch
o
o
s
in
g
k
r
a
n
d
o
m
in
i
tial
ce
n
t
r
o
id
s
,
th
en
as
s
i
g
n
s
v
ec
to
r
s
to
t
h
e
n
ea
r
est
ce
n
tr
o
id
u
s
i
n
g
E
u
cl
id
ea
n
d
is
ta
n
ce
a
n
d
r
ec
alcu
lates
t
h
e
n
e
w
ce
n
tr
o
id
s
as
m
ea
n
s
o
f
t
h
e
as
s
i
g
n
ed
d
ata
v
ec
to
r
s
.
T
h
is
p
r
o
ce
s
s
is
r
ep
ea
ted
m
a
n
y
ti
m
es
u
n
t
il v
ec
to
r
s
n
o
lo
n
g
er
alter
ed
clu
s
ter
s
b
et
w
ee
n
iter
atio
n
s
[
3
8
]
.
Ho
w
e
v
er
,
i
n
t
h
e
k
-
m
ea
n
s
tec
h
n
iq
u
e,
th
e
n
u
m
b
er
o
f
c
lu
s
ter
s
is
r
an
d
o
m
l
y
s
elec
ted
,
w
h
ic
h
m
ea
n
s
t
h
at
th
e
cl
u
s
ter
i
n
g
r
es
u
lt
w
ill
b
ec
o
m
e
u
n
r
eliab
le
i
f
th
e
s
u
p
p
o
s
ed
n
u
m
b
er
o
f
t
h
e
clu
s
ter
s
i
s
in
co
r
r
ec
t
[
3
9
]
,
[
4
0
]
,
th
is
r
aises
th
e
f
o
llo
w
i
n
g
f
u
n
d
a
m
en
tal
q
u
esti
o
n
: H
o
w
to
ch
o
o
s
e
th
e
r
ig
h
t
n
u
m
b
er
o
f
ex
p
ec
ted
cl
u
s
ter
s
(
k
)
?
.
So
m
e
t
y
p
es
o
f
e
f
f
icie
n
t
clu
s
te
r
in
g
q
u
al
it
y
i
n
d
ex
e
s
ca
n
h
elp
d
eter
m
in
e
t
h
e
b
est
n
u
m
b
er
.
I
n
th
is
s
t
u
d
y
,
w
e
h
a
v
e
u
s
ed
t
w
o
m
eth
o
d
s
f
o
r
d
eter
m
i
n
i
n
g
t
h
e
o
p
ti
m
al
n
u
m
b
er
o
f
cl
u
s
ter
s
f
o
r
k
-
m
ea
n
s
.
T
h
ese
m
e
th
o
d
s
co
n
s
is
t
o
f
o
p
ti
m
izi
n
g
a
cr
iter
i
o
n
,
s
u
c
h
as
th
e
w
i
th
i
n
cl
u
s
ter
s
u
m
s
o
f
s
q
u
ar
es
an
d
th
e
a
v
er
ag
e
s
il
h
o
u
ette.
T
h
e
co
r
r
esp
o
n
d
in
g
m
e
th
o
d
s
ar
e
n
a
m
ed
elb
o
w
a
n
d
s
il
h
o
u
ette
m
et
h
o
d
s
,
r
esp
ec
tiv
el
y
.
I
n
t
h
i
s
s
t
u
d
y
,
t
h
e
s
u
m
o
f
s
q
u
ar
ed
er
r
o
r
s
(
SS
E
)
an
d
th
e
av
er
ag
e
s
i
lh
o
u
ette
co
ef
f
icie
n
t
w
h
ic
h
ar
e
s
h
o
w
n
i
n
th
e
E
q
u
atio
n
s
(
1
)
an
d
(
2
)
r
esp
ec
tiv
el
y
,
ar
e
co
m
b
i
n
ed
t
o
m
ea
s
u
r
e
t
h
e
q
u
al
it
y
o
f
clu
s
ter
in
g
an
d
to
d
eter
m
i
n
e
t
h
e
o
p
tim
a
l
clu
s
ter
in
g
n
u
m
b
er
.
Sp
ec
if
icall
y
,
w
e
ap
p
lied
k
-
m
ea
n
s
tech
n
iq
u
e
u
n
d
er
d
if
f
er
e
n
t
k
v
al
u
e
s
an
d
th
en
we
p
lo
t
th
e
cu
r
v
es
o
f
th
e
SS
E
a
n
d
av
er
a
g
e
s
il
h
o
u
et
te
co
ef
f
ic
ien
t
a
g
ai
n
s
t
t
h
e
n
u
m
b
er
o
f
clu
s
ter
s
to
a
n
al
y
ze
t
h
e
t
w
o
c
u
r
v
es
a
n
d
to
id
en
ti
f
y
t
h
e
o
p
ti
m
al
n
u
m
b
er
o
f
cl
u
s
ter
.
T
h
e
o
p
ti
m
al
c
lu
s
ter
i
n
g
n
u
m
b
er
ca
n
b
e
f
o
u
n
d
i
n
a
d
ata
s
et
b
y
lo
o
k
i
n
g
f
o
r
th
e
n
u
m
b
er
o
f
cl
u
s
ter
s
at
w
h
ic
h
a
k
n
ee
,
p
ea
k
,
o
r
d
ip
ex
is
t
s
i
n
th
e
p
lo
t
o
f
t
h
e
e
v
alu
a
tio
n
m
ea
s
u
r
e
w
h
e
n
p
lo
tted
ag
ain
s
t t
h
e
n
u
m
b
er
o
f
clu
s
ter
s
[
4
1
]
.
∑
∑
‖
‖
(
1
)
W
h
er
e
k
is
t
h
e
n
u
m
b
er
o
f
cl
u
s
t
er
s
,
y
i
is
t
h
e
j
th
o
b
j
ec
t in
clu
s
te
r
C
i
an
d
c
i
is
t
h
e
ce
n
ter
o
f
cl
u
s
t
er
C
i
.
(
2
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
C
lu
s
ter
in
g
P
r
ed
ictio
n
Tech
n
iq
u
es in
Defin
in
g
a
n
d
P
r
ed
ictin
g
C
u
s
to
mers
Defe
ct
io
n
:
…
(
A
it Da
q
u
d
R
a
ch
id
)
2371
W
h
er
e
a
i
is
th
e
a
v
er
a
g
e
d
i
s
tan
ce
o
f
o
b
j
ec
t
I
to
all
o
th
er
o
b
j
e
cts
in
its
cl
u
s
ter
;
f
o
r
o
b
j
ec
t
I
a
n
d
an
y
cl
u
s
ter
n
o
t
co
n
tain
i
n
g
it,
ca
lc
u
late
th
e
a
v
er
ag
e
d
is
ta
n
ce
o
f
t
h
e
o
b
j
ec
t
to
all
th
e
o
b
j
ec
ts
in
th
e
g
i
v
e
n
clu
s
ter
,
a
n
d
b
i
is
t
h
e
m
i
n
i
m
u
m
o
f
s
u
c
h
v
a
lu
e
s
w
it
h
r
esp
ec
t
to
all
clu
s
ter
s
.
T
h
e
d
etails
o
f
SS
E
an
d
Sil
h
o
u
ette
ca
n
b
e
f
o
u
n
d
r
esp
ec
tiv
el
y
i
n
[
42
]
,
[
4
3
]
.
T
h
is
s
t
u
d
y
co
m
b
in
e
s
K
-
m
ea
n
s
an
d
L
R
FM
m
o
d
el
i
n
e
-
co
m
m
er
ce
s
ec
to
r
to
d
i
v
id
e
th
e
cu
s
to
m
er
b
ase
u
p
in
to
h
o
m
o
g
e
n
eo
u
s
cl
u
s
ter
s
ac
co
r
d
in
g
to
th
eir
L
,
R
,
F
an
d
M
v
alu
e
s
.
Si
m
ilar
l
y
to
C
h
an
g
an
d
T
s
ay
[
3
6
]
,
w
e
w
il
l
u
s
e
a
v
er
ag
e
L
R
FM
v
alu
e
s
o
f
ea
c
h
c
lu
s
ter
to
co
m
p
ar
e
with
t
h
e
to
tal
av
er
a
g
e
L
R
FM
v
a
lu
es
o
f
al
l
cl
u
s
ter
s
.
I
f
t
h
e
a
v
er
ag
e
(
L
,
R
,
F,M
)
v
al
u
e
o
f
a
cl
u
s
ter
is
g
r
ea
ter
t
h
a
n
t
h
e
to
tal
av
er
a
g
e,
a
n
o
v
er
b
ar
a
p
p
ea
r
s
.
Ho
w
ev
er
,
if
th
e
av
er
ag
e
(
L
,
R
,
F,M
)
v
alu
e
o
f
a
clu
s
ter
is
less
th
a
n
th
e
to
tal
av
er
ag
e,
an
u
n
d
er
b
ar
ap
p
ea
r
s
.
(
i.e
.
,
:
Hig
h
er
R
v
alu
e
;
cu
s
to
m
er
h
a
v
e
r
ec
en
tl
y
m
ad
e
a
p
u
r
ch
ase,
:
L
o
w
er
R
v
alu
e;
cu
s
to
m
er
h
a
v
e
n
o
t
b
u
y
o
n
th
e
o
n
li
n
e
s
to
r
e
f
o
r
a
lo
n
g
ti
m
e)
.
C
h
a
n
g
an
d
T
s
a
y
[
3
6
]
b
ased
o
n
Ha
an
d
P
ar
k
[
4
4
]
f
u
r
th
er
p
r
o
p
o
s
ed
cu
s
to
m
er
class
i
f
icatio
n
b
y
s
u
m
m
i
n
g
t
h
e
s
i
x
tee
n
co
m
b
i
n
atio
n
s
o
f
L
R
FM
m
o
d
el
to
f
i
v
e
k
i
n
d
s
o
f
c
u
s
to
m
er
g
r
o
u
p
s
ac
co
r
d
in
g
to
th
eir
L
R
FM
p
atter
n
s
,
s
u
c
h
a
s
co
r
e
cu
s
to
m
er
s
,
p
o
ten
t
ial
c
u
s
to
m
er
s
,
lo
s
t
c
u
s
to
m
er
s
,
n
e
w
cu
s
to
m
er
s
,
a
n
d
r
eso
u
r
ce
-
co
n
s
u
m
p
tio
n
c
u
s
to
m
er
s
.
Sp
e
cif
icall
y
,
co
r
e
c
u
s
to
m
er
s
i
n
cl
u
d
e
L
↑
R
↑
F
↑
M
↑
,
L
↑
R
↑
F
↑
M
↓
,
an
d
L
↑
R
↑
F
↓
M
↑
.
P
o
ten
tial
cu
s
t
o
m
er
s
co
n
s
i
s
t
o
f
L
↑
R
↓
F
↑
M
↑
,
L
↑
R
↓
F
↑
M
↓
,
a
n
d
L
↑
R
↓
F
↓
M
↑
.
L
o
s
t
c
u
s
to
m
e
r
s
ar
e
co
m
p
o
s
ed
o
f
L
↓
R
↓
F
↑
M
↑
,
L
↓
R
↓
F
↑
M
↓
,
L
↓
R
↓
F
↓
M
↑
,
a
n
d
L
↓
R
↓
F
↓
M
↓
.
N
e
w
c
u
s
to
m
er
s
co
m
p
r
is
e
L
↓
R
↑
F
↓
M
↓
,
L
↓
R
↑
F
↑
M
↓
,
L
↓
R
↑
F
↓
M
↑
,
a
n
d
L
↓
R
↑
F
↑
M
↑
.
Fin
all
y
,
r
eso
u
r
ce
-
co
n
s
u
m
p
tio
n
cu
s
to
m
er
s
ar
e
L
↑
R
↑
F
↓
M
↓
a
n
d
L
↑
R
↓
F
↓
M
↓
.
W
h
en
d
if
f
er
en
t
L
R
FM
co
m
b
i
n
atio
n
s
ar
e
id
en
tif
ied
d
u
r
i
n
g
a
p
er
io
d
T
,
cu
s
to
m
er
s
ca
n
b
e
class
i
f
ied
in
to
ap
p
r
o
p
r
iate
g
r
o
u
p
s
s
u
c
h
as
co
r
e
cu
s
to
m
er
s
,
p
o
ten
tial
cu
s
to
m
er
s
,
lo
s
t
c
u
s
to
m
er
s
,
n
e
w
c
u
s
to
m
er
s
,
an
d
r
eso
u
r
ce
-
co
n
s
u
m
p
tio
n
cu
s
to
m
er
s
.
First,
w
e
f
o
cu
s
o
n
cu
s
to
m
er
s
b
elo
n
g
to
co
r
e
cu
s
to
m
er
s
,
n
e
w
cu
s
to
m
er
s
(
n
o
co
m
p
a
n
y
w
a
n
t
to
m
is
s
n
e
w
c
u
s
to
m
er
s
)
,
s
ec
o
n
d
,
w
e
tak
e
i
n
to
ac
co
u
n
t
th
o
s
e
b
elo
n
g
i
n
g
d
u
r
in
g
t
h
e
p
er
io
d
T
t
o
o
th
er
r
em
a
in
i
n
g
g
r
o
u
p
s
,
an
d
w
h
ic
h
ar
e
s
u
b
s
eq
u
e
n
tl
y
co
n
v
e
r
ted
in
to
co
r
e
cu
s
to
m
er
s
i
n
T
+
1
.
Mo
r
e
s
p
ec
if
ical
l
y
,
th
e
cu
s
to
m
er
s
in
o
u
r
cl
u
s
ter
s
o
f
atte
n
tio
n
b
elo
n
g
to
th
e
f
o
llo
w
i
n
g
p
atter
n
s
:
a.
,
,
,
,
an
d
d
u
r
in
g
a
p
er
io
d
T
.
b.
C
u
s
to
m
er
s
w
h
o
d
o
n
o
t
b
elo
n
g
in
p
er
io
d
T
to
th
e
p
atter
n
s
lis
ted
in
(
1
)
,
b
u
t
in
th
e
T
+1
p
er
i
o
d
,
th
eir
L
R
F
M
p
atter
n
tr
an
s
f
o
r
m
ed
i
n
to
o
n
e
o
f
th
e
p
atter
n
s
m
en
tio
n
ed
in
(
1
)
.
C
u
s
to
m
er
s
w
h
o
w
er
e
clu
s
ter
e
d
d
u
r
in
g
th
e
p
er
io
d
T
w
it
h
p
o
ten
tial,
lo
s
t
o
r
r
eso
u
r
ce
-
co
n
s
u
m
p
tio
n
c
u
s
to
m
er
s
,
an
d
th
at
ar
e
s
ta
y
ed
in
t
h
e
s
a
m
e
g
r
o
u
p
o
r
ar
e
tr
an
s
f
o
r
m
ed
to
a
lo
w
er
v
al
u
e
g
r
o
u
p
in
T
+1
,
th
ey
w
ill b
e
r
e
m
o
v
ed
.
2
.2
.
P
a
rt
ia
l a
nd
t
o
t
a
l c
hu
rning
Am
o
n
g
th
e
f
ir
s
t
m
ain
h
u
r
d
les
w
h
ich
f
ac
e
o
n
th
e
cu
s
to
m
er
s
ch
u
r
n
p
r
ed
ictio
n
in
t
h
e
n
o
n
-
c
o
n
tr
ac
tu
a
l
b
u
s
i
n
ess
e
s
i
s
t
h
e
d
i
f
f
icu
l
t
y
o
f
d
ef
i
n
in
g
c
h
u
r
n
b
ec
a
u
s
e
th
e
c
h
ar
ac
ter
is
tic
s
t
h
at
s
h
o
u
ld
b
e
o
b
s
er
v
ed
to
s
a
y
i
n
g
th
at
a
cu
s
to
m
er
h
as to
tall
y
o
r
p
ar
tiall
y
d
ef
ec
ted
ar
e
n
o
t c
lear
l
y
d
ef
i
n
ed
[
1
1
]
.
Fo
r
s
o
lv
i
n
g
t
h
e
p
r
o
b
le
m
s
a
b
o
v
e,
(
d
ef
in
i
tio
n
o
f
c
u
s
to
m
e
r
ch
u
r
n
)
L
R
FM
m
o
d
el
a
n
d
clu
s
ter
in
g
tech
n
iq
u
e
(
k
-
m
ea
n
s
)
ar
e
co
m
b
in
ed
.
T
h
is
s
t
u
d
y
p
r
o
p
o
s
es a
n
e
w
p
r
o
ce
d
u
r
e
by
j
o
in
i
n
g
th
e
q
u
an
titati
v
e
v
al
u
e
s
o
f
th
e
L
R
FMa
ttrib
u
te
s
,
ex
tr
ac
te
d
d
u
r
in
g
a
p
er
io
d
T
,
in
to
K
-
m
ea
n
s
alg
o
r
it
h
m
to
id
en
t
if
y
t
h
e
d
if
f
er
en
t
t
y
p
e
s
o
f
cu
s
to
m
er
p
r
o
f
iles
(
d
if
f
er
en
t
L
R
FM
p
atter
n
s
)
.
W
e
th
en
d
ef
i
n
e
a
cu
s
to
m
er
‟
s
L
R
FM
p
atter
n
ch
an
g
e
f
r
o
m
a
co
r
e
(
,
,
)
o
r
a
n
e
w
cu
s
to
m
er
(
,
,
)
to
p
o
ten
tial
cu
s
to
m
er
(
,
,
)
o
r
to
lo
w
c
o
n
s
u
m
in
g
r
eso
u
r
c
e
cu
s
to
m
er
g
r
o
u
p
s
(
)
as
p
ar
tial
d
ef
ec
tio
n
.
B
y
th
e
s
a
m
e
to
k
e
n
,
if
a
cu
s
to
m
er
ch
an
g
es
h
er
L
R
FM
m
o
d
el
f
r
o
m
o
n
e
o
f
t
w
o
f
o
llo
w
i
n
g
t
y
p
es
o
f
cu
s
to
m
er
:
co
r
e
cu
s
to
m
er
s
(
,
,
)
o
r
n
e
w
c
u
s
to
m
er
(
,
,
)
to
th
e
lo
s
t
cu
s
to
m
er
s
(
,
,
,
)
o
r
to
h
ig
h
c
o
n
s
u
m
in
g
r
eso
u
r
ce
cu
s
to
m
er
g
r
o
u
p
s
(
)
,
in
th
is
ca
s
e,
w
e
ar
e
talk
i
n
g
ab
o
u
t
to
tal
d
ef
ec
tio
n
.
T
h
is
w
o
u
ld
in
d
icat
e
th
at
a
cu
s
to
m
er
‟
s
ch
a
n
g
e
i
n
L
R
FM
p
atter
n
s
i
s
an
ea
r
l
y
s
ig
n
al
o
f
eit
h
er
p
ar
tial
o
r
to
tal
d
ef
ec
tio
n
.
W
h
er
ea
s
c
u
s
to
m
er
s
w
h
o
s
ta
y
i
n
g
tr
u
e
to
th
eir
ex
is
ti
n
g
p
o
s
iti
v
e
p
atter
n
s
ar
e
lik
e
l
y
to
s
ta
y
.
Fo
r
th
is
p
u
r
p
o
s
e,
as
s
h
o
w
n
in
Fi
g
u
r
e
1
,
w
e
co
n
s
id
er
t
wo
eq
u
al
s
u
b
-
p
er
io
d
s
T
1
an
d
T
2
.
T
1
is
u
s
ed
to
d
eter
m
in
e
t
h
e
d
if
f
er
e
n
t
cu
s
t
o
m
er
g
r
o
u
p
s
(
d
if
f
er
e
n
t
L
R
F
M
p
atter
n
s
)
an
d
ass
ig
n
ea
c
h
cu
s
to
m
er
to
its
a
p
p
r
o
p
r
iate
g
r
o
u
p
.
T
h
e
p
er
i
o
d
T
2
is
u
s
ed
to
d
eter
m
i
n
e
p
ar
tial
o
r
to
tal
d
ef
ec
tio
n
.
Fig
u
r
e
2
illu
s
tr
ates
o
u
r
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
to
d
ef
in
in
g
p
ar
tial a
n
d
to
tal
d
ef
ec
tio
n
,
an
d
t
h
e
f
u
ll p
r
o
ce
s
s
is
s
u
m
m
ar
iz
ed
in
Fig
u
r
e
3
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2088
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
4
,
A
u
g
u
s
t 2
0
1
8
:
2
3
6
7
–
2
3
8
3
2372
Fig
u
r
e
1
.
P
er
io
d
o
f
o
b
s
er
v
atio
n
.
T
1
t
h
is
p
er
io
d
o
f
eig
h
t
m
o
n
t
h
s
(
f
r
o
m
No
v
e
m
b
er
2
0
1
3
t
o
J
u
n
e
2
0
1
4
)
w
as a
ls
o
u
s
ed
to
d
er
iv
e
th
e
i
n
d
ep
en
d
en
t
v
ar
iab
les (
ca
lib
r
atio
n
p
er
io
d
)
o
f
th
e
m
o
d
el.
T
2
th
is
p
er
io
d
o
f
eig
h
t
m
o
n
th
s
(
f
r
o
m
J
u
l
y
,
2
0
1
4
to
Feb
r
u
ar
y
,
2
0
1
5
)
w
a
s
u
s
ed
to
d
er
iv
e
th
e
d
ep
en
d
e
n
t v
ar
iab
le
(
p
r
ed
ictio
n
p
er
io
d
)
Fig
u
r
e
2
.
Dev
iatio
n
o
f
cu
s
to
m
er
s
L
R
FM
p
atter
n
o
v
e
r
ti
m
e
to
d
ef
in
e
c
u
s
to
m
er
ch
u
r
n
Ob
se
rv
a
ti
o
n
m
id
p
o
in
t
T
1
.1
T
1
.
2
T
1
:
P
er
i
o
d
to
i
d
en
ti
f
y
d
iff
er
en
t cu
s
to
m
er
g
r
o
u
p
s
T
2
:
P
er
i
o
d
to
d
eter
m
i
n
e
p
ar
ti
al an
d
to
tal d
ef
ectio
n
No
v
-
20
13
…
…
Ju
n
-
2
0
1
4
Ju
l
-
2
0
1
4
…
…
F
e
b
-
20
15
T1
T
2
Po
ten
tia
l cu
s
to
m
e
rs
Core
cu
s
to
m
e
r
s
Re
s
o
u
rc
e
-
con
s
u
m
p
tio
n
cu
s
to
m
e
rs
Re
s
o
u
rc
e
-
con
s
u
m
p
tio
n
cu
s
to
m
e
rs
N
e
w
cu
s
to
m
e
r
s
Lo
s
t
cu
s
to
m
e
r
s
T1
T
2
A1
A2
B
1
B
2
B
2
A1:
P
artial
defec
tio
n
A2:
To
tal defe
cti
o
n
B1
:
P
arti
al
d
efe
cti
o
n
B
2
:
To
tal def
ect
i
o
n
T
i
me
x
x
Po
ten
tia
l cu
s
to
m
e
rs
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
C
lu
s
ter
in
g
P
r
ed
ictio
n
Tech
n
iq
u
es in
Defin
in
g
a
n
d
P
r
ed
ictin
g
C
u
s
to
mers
Defe
ct
io
n
:
…
(
A
it Da
q
u
d
R
a
ch
id
)
2373
Fig
u
r
e
3
.
Def
i
n
in
g
c
h
u
r
n
in
n
o
n
-
co
n
tr
ac
tu
a
l setti
n
g
s
m
et
h
o
d
o
lo
g
y
b
ased
o
n
L
R
FM
m
o
d
el
an
d
K
-
m
ea
n
s
tech
n
iq
u
e
2
.3
.
Cla
s
s
if
ica
t
io
n
t
ec
hn
iq
ues
T
h
e
o
b
j
ec
tiv
e
o
f
th
is
r
e
s
ea
r
ch
w
a
s
to
d
e
v
elo
p
a
p
r
ed
ictiv
e
m
o
d
el
f
o
r
c
u
s
to
m
er
c
h
u
r
n
i
n
a
n
o
n
-
co
n
tr
ac
tu
al
s
e
tti
n
g
w
h
ic
h
w
o
u
ld
b
e
ab
le
to
d
is
tin
g
u
i
s
h
b
et
w
e
en
cu
s
to
m
er
s
w
h
o
ar
e
lik
el
y
to
p
ar
tiall
y
o
r
to
tall
y
ch
u
r
n
i
n
th
e
n
ea
r
f
u
t
u
r
e
an
d
t
h
e
o
n
es
w
h
o
ar
e
li
k
el
y
to
s
ta
y
w
it
h
t
h
e
co
m
p
a
n
y
b
ased
o
n
h
i
s
to
r
ical
tr
an
s
ac
t
io
n
s
an
d
ch
ar
ac
ter
is
tics
o
f
a
c
u
s
to
m
er
.
T
o
r
ea
ch
th
is
g
o
al
t
h
r
ee
m
o
d
el
s
ar
e
p
r
o
p
o
s
ed
,
th
e
f
ir
s
t
is
b
ased
o
n
d
ec
i
s
io
n
tr
ee
tech
n
iq
u
e
s
(
DT
)
,
th
e
s
ec
o
n
d
o
n
ar
ti
f
icial
n
e
u
r
al
n
et
w
o
r
k
s
(
A
N
N)
an
d
t
h
e
t
h
ir
d
b
a
s
ed
o
n
an
en
s
e
m
b
le
o
f
d
ec
is
io
n
tr
ee
s
.
W
e
n
o
te
th
at
all
o
u
r
m
o
d
els
ar
e
co
n
s
tr
u
cte
d
u
s
in
g
KNI
ME
An
al
y
tic
s
P
latf
o
r
m
3
.
3
.
2
.
T
h
e
f
o
llo
w
in
g
i
s
th
e
s
h
o
r
t d
escr
ip
tio
n
f
o
r
th
ese
k
n
o
w
n
d
ata
m
in
i
n
g
tech
n
iq
u
e
s
u
s
ed
f
o
r
th
i
s
tas
k
.
2
.
3
.
1
.
Art
if
icia
l
neura
l net
w
o
rk
s
(
ANN)
Un
li
k
e
to
co
n
v
en
t
io
n
al
s
tati
s
ti
c
al
m
e
th
o
d
s
,
ar
ti
f
icia
l
n
eu
r
al
n
et
w
o
r
k
s
d
o
n
o
t
n
ee
d
an
y
h
y
p
o
th
esi
s
o
n
th
e
v
ar
iab
les,
th
e
y
ar
e
w
ell
-
s
u
ited
to
h
an
d
le
u
n
s
tr
u
c
tu
r
ed
co
m
p
lex
p
r
o
b
le
m
s
,
i.e
is
s
u
e
s
o
n
w
h
ic
h
t
h
er
e
is
n
o
a
p
r
io
r
i sp
ec
if
y
th
e
f
o
r
m
o
f
r
elat
io
n
s
h
i
p
s
b
et
w
ee
n
v
ar
iab
les.
Neu
r
al
n
et
w
o
r
k
s
ca
n
b
e
d
is
ti
n
g
u
i
s
h
ed
in
to
s
i
n
g
le
-
la
y
er
p
er
ce
p
tr
o
n
an
d
m
u
l
tila
y
er
p
er
ce
p
tr
o
n
(
ML
P
)
,
in
t
h
is
p
ap
er
,
w
e
u
s
e
t
h
e
M
L
P
s
tr
u
ct
u
r
e
th
at
al
lo
w
s
r
ea
lizi
n
g
th
e
m
o
s
t
d
iv
er
s
e
ap
p
licatio
n
s
.
An
M
L
P
n
et
w
o
r
k
is
g
e
n
er
all
y
co
m
p
o
s
ed
o
f
a
f
i
n
ite
s
e
t
o
f
ce
l
ls
(
n
e
u
r
o
n
s
)
,
o
r
g
an
ized
i
n
s
u
cc
es
s
iv
e
la
y
er
s
.
T
h
e
f
ir
s
t
la
y
er
co
m
p
r
is
in
g
s
e
v
er
al
n
e
u
r
o
n
s
is
ca
lled
th
e
in
p
u
t
la
y
er
,
th
e
la
s
t
la
y
er
i
s
th
e
o
u
tp
u
t
la
y
er
,
an
d
th
e
i
n
ter
m
ed
iate
la
y
er
s
(
i
f
an
y
)
ar
e
th
e
h
id
d
en
la
y
er
s
.
Neu
r
o
n
s
i
n
d
if
f
er
en
t
la
y
er
s
ar
e
co
n
n
ec
ted
b
y
s
i
g
m
o
id
o
r
h
y
p
er
b
o
lic
tan
g
e
n
t
f
u
n
ctio
n
s
t
h
at
ar
e
u
s
e
d
as
ac
ti
v
atio
n
f
u
n
c
tio
n
s
i
n
M
u
lti
-
la
y
er
p
er
ce
p
tio
n
.
T
h
e
d
etails
o
f
M
L
P
ca
n
b
e
f
o
u
n
d
in
[
4
5
]
.
2
.
3
.
2
.
Si
m
ple
decisi
o
n t
re
e
(
DT
)
Dec
is
io
n
tr
ee
(
DT
)
is
o
n
e
o
f
t
h
e
m
o
s
t
d
ata
m
in
i
n
g
tech
n
iq
u
es
f
o
r
k
n
o
w
led
g
e
d
is
co
v
er
y
a
n
d
it
u
s
ed
u
s
u
all
y
f
o
r
th
e
p
u
r
p
o
s
e
o
f
cl
ass
i
f
icatio
n
a
n
d
p
r
ed
ictio
n
[
4
6
]
.
T
h
e
s
i
m
p
licit
y
a
n
d
ea
s
e
o
f
i
n
ter
p
r
etin
g
t
h
e
Data s
e
t
Se
le
ct
cu
s
to
m
e
r
s
with
d
at
e
_
s
e
s
s
ion
in
T1
Se
le
ct
cu
s
to
m
e
r
s
with
d
at
e
_
s
e
s
s
ion
in
T
2
E
xt
ra
ct
th
e
v
alu
e
s
o
f
L
,
R,
F
an
d
M
for
e
ach
cu
s
to
m
e
r
E
xt
ra
ct
th
e
v
alu
e
s
o
f
L
,
R,
F
an
d
M
for
e
ach
cu
s
to
m
e
r
N
o
rm
aliz
e
t
h
e
v
alu
e
s
o
f
a
ll v
ar
iab
le
s
(Z
-
s
cor
e
N
o
r
m
aliz
at
ion
)
N
o
rm
aliz
e
t
h
e
v
alu
e
s
o
f
L
,
R,
F
an
d
M
in
T2
acco
rd
in
g t
o
t
h
e
n
o
rm
aliz
at
ion
p
ar
ame
ters
as giv
e
n
in
T1
De
ter
m
in
e
t
h
e
b
e
s
t
n
u
m
b
e
r
o
f
c
lu
s
ter
(K
)
b
y
u
s
in
g SSE
a
n
d
si
lh
o
u
e
tt
e
m
e
th
o
d
s
Ap
p
lic
at
ion
o
f
clu
s
ter a
s
s
i
gn
e
r
t
h
at
a
s
s
igns
e
xis
tin
g cu
s
to
m
e
rs
in
T
2 t
o
t
h
e
exi
s
tin
g
cluste
r,
wh
i
ch
a
re
o
b
ta
in
e
d
b
y
k
-
m
e
an
s
in
T1
U
s
in
g t
h
e
n
u
m
b
e
r
(K
) a
s
in
p
u
t
p
ar
ame
ter
o
f
K
-
m
e
an
s
t
o
s
e
g
m
e
n
t
all
cu
s
to
m
e
r
s
in
to
(K
) clu
s
t
e
r
s
ac
cord
in
g t
o
t
h
e
ir L
,
R,
F
an
d
M
v
alu
e
s
K clu
s
ter
s
a
re
d
e
t
e
rm
in
e
d
,
a
n
d
each
cu
s
to
m
e
r
is
a
s
s
ign
e
d
t
o
it
s
a
p
p
ro
p
riat
e
cluste
r
E
ach
cu
s
to
m
e
r
w
ill
b
e
a
s
s
ign
e
d
t
o
it
s
n
e
ar
e
s
t
cluste
r
LRFM
p
at
tern
ch
an
ge
fro
m
a
co
re
o
r
n
e
w
cu
s
to
m
e
r
(
T
1) to
p
o
ten
ti
al
cu
s
to
m
e
r
o
r
to
low con
s
u
m
in
g
re
s
o
u
rc
e
cu
s
to
m
e
r
grou
p
s
(
T
2
)
LRFM
p
at
tern
ch
an
ge
fro
m
a
co
re
o
r
n
e
w
cu
s
to
m
e
r
(
T
1) to
lo
s
t
o
r
to
h
igh
con
s
u
m
in
g re
s
o
u
rc
e
cu
s
to
m
e
r
grou
p
s
(T
2
)
P
artial
d
efe
c
tio
n
To
tal
d
efe
c
tio
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2088
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
4
,
A
u
g
u
s
t 2
0
1
8
:
2
3
6
7
–
2
3
8
3
2374
o
b
s
er
v
ed
r
esu
lts
b
y
d
ec
is
io
n
m
ak
er
s
ar
e
th
e
m
ai
n
r
ea
s
o
n
s
f
o
r
its
p
o
p
u
lar
i
t
y
in
b
u
s
i
n
es
s
co
m
p
ar
ed
to
o
th
er
p
r
ed
ictio
n
tech
n
iq
u
e
s
[
4
7
]
.
D
T
d
ev
elo
p
m
e
n
t
u
s
u
a
ll
y
co
n
s
is
ts
o
f
t
w
o
d
is
ti
n
ct
s
ta
g
e
s
,
tr
ee
b
u
ild
in
g
,
a
n
d
tr
ee
p
r
u
n
i
n
g
.
A
t
f
ir
s
t,
t
h
e
tec
h
n
iq
u
es
s
tar
t
to
s
ea
r
ch
i
n
t
h
e
tr
ain
i
n
g
s
et
an
at
tr
ib
u
te
o
f
f
er
i
n
g
t
h
e
b
est
in
f
o
r
m
atio
n
ga
in
at
t
h
e
r
o
o
t
n
o
d
e
lev
el,
an
d
th
en
d
iv
id
in
g
th
e
tr
ee
in
to
s
u
b
-
tr
ee
s
.
T
h
e
s
a
m
e
p
r
o
ce
d
u
r
e
is
u
s
ed
to
r
ec
u
r
s
iv
e
l
y
p
ar
titi
o
n
ed
th
e
s
u
b
-
tr
ee
f
o
llo
w
i
n
g
th
e
s
a
m
e
r
u
le,
th
en
th
e
p
ar
titi
o
n
in
g
s
to
p
s
w
h
en
th
e
lea
f
n
o
d
e
is
r
ea
ch
ed
.
On
ce
t
h
e
tr
ee
is
cr
ea
ted
,
r
u
les
ca
n
be
ex
tr
ac
ted
b
y
tr
a
v
er
s
i
n
g
t
h
r
o
u
g
h
th
e
tr
ee
u
n
t
il
a
lea
f
n
o
d
e
is
r
ea
ch
ed
.
Sev
er
al
a
lg
o
r
it
h
m
s
s
u
ch
as
C
4
.
5
,
C
5
.
0
,
C
H
A
I
D
a
n
d
C
AR
T
ar
e
u
s
ed
to
p
r
o
d
u
ce
th
e
tr
ee
s
,
i
n
t
h
i
s
s
tu
d
y
w
e
co
n
s
id
er
C
4
.
5
alg
o
r
it
h
m
.
T
h
e
d
etails o
f
DT
ca
n
b
e
f
o
u
n
d
in
[
48
]
,
[
4
9
]
.
2
.
3
.
3
.
D
ec
is
io
n
t
re
e
ens
e
m
ble
(
DT
E
)
Desp
ite
t
h
e
ad
v
an
ta
g
es
o
f
th
e
d
ec
is
io
n
tr
ee
m
et
h
o
d
m
e
n
tio
n
ed
ab
o
v
e,
it
also
h
as
s
o
m
e
d
is
ad
v
an
ta
g
es.
Fo
r
e
x
a
m
p
le,
Du
d
o
it,
Frid
l
y
a
n
d
,
a
n
d
Sp
ee
d
[
5
0
]
n
o
te
s
o
m
e
o
f
i
ts
d
i
s
a
d
v
an
ta
g
es;
e.
g
.
its
s
u
b
o
p
ti
m
al
p
er
f
o
r
m
a
n
ce
an
d
t
h
e
lack
o
f
r
o
b
u
s
t
n
ess
.
Am
o
n
g
th
e
b
est
w
a
y
s
to
s
o
lv
e
t
h
e
m
i
s
th
e
cr
ea
tio
n
o
f
th
e
en
s
e
m
b
le
o
f
tr
ee
s
f
o
llo
w
ed
b
y
a
v
o
te
f
o
r
th
e
m
o
s
t
p
o
p
u
lar
class
[
5
1
]
.
T
h
is
s
o
lu
tio
n
i
s
th
e
r
esu
l
t
o
f
s
o
m
e
r
esear
ch
er
s
w
h
o
o
p
ti
m
ized
th
e
D
ec
is
io
n
tr
ee
tech
n
iq
u
e.
I
n
th
is
r
e
g
ar
d
,
w
e
u
s
e
t
h
e
b
o
th
T
r
ee
E
n
s
e
m
b
le
L
ea
r
n
er
an
d
th
e
T
r
ee
E
n
s
e
m
b
le
P
r
ed
icto
r
n
o
d
es
o
f
Kn
i
m
e
to
b
u
ild
o
u
r
th
ir
d
m
o
d
e
l th
at
is
b
ased
o
n
d
ec
is
io
n
tr
ee
en
s
e
m
b
le.
T
h
e
T
r
ee
E
n
s
e
m
b
le
L
ea
r
n
er
n
o
d
e
b
u
ild
s
an
en
s
e
m
b
le
o
f
d
ec
is
io
n
tr
ee
s
,
as
a
v
ar
ia
n
t
o
f
t
h
e
r
an
d
o
m
f
o
r
est.
E
ac
h
o
f
t
h
e
d
ec
is
io
n
tr
ee
m
o
d
els
is
tr
ai
n
ed
o
n
a
d
if
f
e
r
en
t
s
u
b
s
et
o
f
r
o
w
s
a
n
d
/o
r
o
n
a
d
if
f
er
e
n
t
s
u
b
s
et
o
f
co
lu
m
n
s
,
r
a
n
d
o
m
l
y
s
elec
ted
a
t
ea
ch
iter
atio
n
.
T
h
e
o
u
tp
u
t
m
o
d
el
i
s
t
h
en
a
n
en
s
e
m
b
le
o
f
d
if
f
er
en
tl
y
tr
ai
n
ed
d
ec
is
io
n
tr
ee
m
o
d
els.
T
h
e
d
ec
is
io
n
tr
ee
s
lear
n
i
n
g
p
ar
a
m
e
t
er
s
ar
e
s
i
m
i
lar
to
th
e
R
a
n
d
o
m
Fo
r
est
cla
s
s
i
f
ier
d
escr
ib
ed
b
y
L
eo
B
r
ei
m
a
n
[
51
]
.
T
h
e
T
r
ee
E
n
s
e
m
b
le
P
r
ed
icto
r
n
o
d
e
ap
p
lies
all
d
ec
is
io
n
tr
ee
s
to
ea
c
h
d
ata
r
o
w
an
d
u
s
e
s
th
e
s
i
m
p
le
m
aj
o
r
it
y
v
o
te
f
o
r
p
r
ed
ictio
n
.
3.
E
M
P
I
RICAL
S
T
UDY
3
.
1
.
G
ener
a
l
T
h
e
d
ata
an
al
y
ze
d
in
t
h
is
r
e
s
ea
r
ch
h
a
v
e
b
ee
n
p
r
o
v
id
ed
f
r
o
m
o
n
e
o
f
th
e
b
i
g
g
e
s
t
o
n
li
n
e
r
etailer
s
s
p
ec
ialized
in
e
lectr
o
n
ics,
f
a
s
h
io
n
,
h
o
m
e
ap
p
lian
ce
s
an
d
c
h
ild
r
en
's
ite
m
s
i
n
Mo
r
o
cc
o
.
W
h
en
cu
s
to
m
er
s
v
i
s
it
th
e
w
eb
s
ite,
t
h
e
s
y
s
te
m
r
ec
o
r
d
s
th
e
ir
lo
g
i
n
,
lo
g
o
u
t,
s
h
o
p
p
in
g
p
r
o
ce
s
s
a
n
d
t
h
e
f
i
n
al
s
tate
o
f
ea
c
h
s
es
s
io
n
.
A
cu
s
to
m
er
ca
n
m
a
k
e
f
o
u
r
t
y
p
es
o
f
ev
en
t
s
,
n
a
m
e
l
y
“
Se
s
s
io
n
w
ith
P
r
o
d
u
ct
Vie
w
s
”,
“
Se
s
s
io
n
w
it
h
A
d
d
to
C
ar
t”,
“
Se
s
s
io
n
w
it
h
C
h
ec
k
-
O
u
t”,
a
n
d
“
Se
s
s
io
n
w
it
h
T
r
an
s
ac
tio
n
s
”.
T
h
e
d
ataset
co
n
s
is
ts
o
f
2
7
8
3
cu
s
to
m
er
s
w
h
o
v
is
i
ted
th
e
e
-
co
m
m
er
ce
w
eb
s
i
te.
Sp
ec
if
ical
l
y
,
t
h
e
d
atase
t
c
o
n
s
is
ts
o
f
i
n
f
o
r
m
atio
n
a
t
t
h
e
i
n
d
iv
id
u
al
c
u
s
to
m
er
lev
el,
s
u
ch
a
s
cu
s
to
m
er
r
eg
i
s
ter
,
lo
g
in
,
s
e
s
s
io
n
,
tr
an
s
ac
ti
o
n
an
d
w
e
b
lo
g
in
t
h
e
e
-
c
o
m
m
er
ce
w
eb
s
ite
.
T
r
an
s
ac
tio
n
al
r
ec
o
r
d
s
o
f
cu
s
t
o
m
er
s
f
o
r
th
e
p
er
io
d
No
v
e
m
b
er
1
,
2
0
1
3
th
r
o
u
g
h
Feb
r
u
ar
y
2
8
,
2
0
1
5
h
av
e
b
ee
n
u
tili
ze
d
.
C
u
s
to
m
er
s
h
av
e
f
o
u
r
m
o
d
es
o
f
p
ay
m
e
n
t:
C
a
s
h
o
n
d
eliv
er
y
,
o
n
lin
e
cr
ed
it
ca
r
d
,
b
an
k
tr
an
s
f
er
an
d
p
ay
m
e
n
t i
n
th
r
ee
i
n
s
tal
l
m
e
n
t
s
.
T
h
e
T
r
an
s
ac
tio
n
al
r
ec
o
r
d
s
f
o
r
ea
ch
cu
s
to
m
er
m
u
s
t
b
e
tr
an
s
f
o
r
m
ed
to
a
u
s
ab
le
f
o
r
m
at
f
o
r
th
e
L
R
FM
m
o
d
el.
Fro
m
th
e
i
n
te
g
r
ated
d
ataset,
th
e
L
,
R
,
F a
n
d
M
v
ar
iab
les
w
er
e
ex
tr
ac
ted
f
o
r
ea
ch
c
u
s
to
m
er
.
T
h
e
d
ef
in
itio
n
o
f
L
R
FM
m
o
d
el
u
s
ed
in
t
h
i
s
s
t
u
d
y
i
s
s
h
o
w
n
i
n
T
ab
le
2
.
T
h
e
d
escr
ip
tiv
e
s
tatis
tic
s
f
o
r
th
e
v
ar
iab
les (
L
R
FM)
in
T
1
ar
e
p
r
o
v
id
e
d
in
T
ab
le
3.
Ta
b
le
2
.
T
h
e
D
ef
in
it
io
n
s
o
f
L
R
FM
M
o
d
el
A
t
t
r
i
b
u
t
e
n
a
me
D
a
t
a
c
o
n
t
e
n
t
L
e
n
g
t
h
(
L
)
R
e
f
e
r
s t
o
t
h
e
n
u
m
b
e
r
o
f
d
a
y
s fr
o
m t
h
e
f
i
r
st
t
o
t
h
e
l
a
st
p
u
r
c
h
a
se
R
e
c
e
n
c
y
(
R
)
R
e
f
e
r
s t
o
t
h
e
n
u
m
b
e
r
o
f
d
a
y
s b
e
t
w
e
e
n
t
h
e
f
i
r
s
t
d
a
y
o
f
st
u
d
y
p
e
r
i
o
d
a
n
d
t
h
e
d
a
y
o
f
t
h
e
l
a
st
p
u
r
c
h
a
se
.
F
r
e
q
u
e
n
c
y
(
F
)
R
e
f
e
r
s t
o
t
h
e
n
u
m
b
e
r
o
f
t
r
a
n
sac
t
i
o
n
o
b
se
r
v
e
d
i
n
t
h
e
p
e
r
i
o
d
a
n
a
l
y
z
e
d
M
o
n
e
t
a
r
y
(
M
)
R
e
f
e
r
s t
o
t
h
e
t
o
t
a
l
a
mo
u
n
t
s
p
e
n
t
b
y
c
u
st
o
me
r
s i
n
t
h
e
p
e
r
i
o
d
a
n
a
l
y
z
e
d
.
(
M
o
r
o
c
c
a
n
d
i
r
h
a
ms)
T
ab
le
3
.
T
h
e
Descr
ip
tio
n
s
o
f
L
e
n
g
t
h
,
R
ec
e
n
c
y
,
Fre
q
u
en
c
y
a
n
d
Mo
n
etar
y
i
n
T1
V
a
r
i
a
b
l
e
s
M
a
x
M
i
n
A
v
e
r
a
g
e
S
t
a
n
d
a
r
d
d
e
v
i
a
t
i
o
n
L
e
n
g
t
h
(
L
)
8
1
3
2
6
5
6
.
6
8
1
9
2
.
8
7
R
e
c
e
n
c
y
(
R
)
2
4
1
1
1
6
4
.
7
7
7
6
.
0
5
F
r
e
q
u
e
n
c
y
(
F
)
17
1
8
.
6
7
4
.
9
9
M
o
n
e
t
a
r
y
(
M
)
1
3
,
7
2
3
.
0
0
8
7
.
0
0
4
4
3
1
.
1
5
4
3
2
7
.
7
2
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
C
lu
s
ter
in
g
P
r
ed
ictio
n
Tech
n
iq
u
es in
Defin
in
g
a
n
d
P
r
ed
ictin
g
C
u
s
to
mers
Defe
ct
io
n
:
…
(
A
it Da
q
u
d
R
a
ch
id
)
2375
3
.2
.
Clus
t
er
ing
by
K
-
m
ea
ns
b
a
s
e
d
o
n L
RF
M
v
a
ri
a
bles
T
h
e
f
ir
s
t
ei
g
h
t
-
m
o
n
t
h
s
p
er
io
d
o
f
t
h
e
av
a
ilab
le
d
ata,
f
r
o
m
No
v
e
m
b
er
,
2
0
1
3
to
J
u
n
e,
2
0
1
4
(
T
1
)
,
is
u
s
ed
to
id
en
ti
f
y
t
h
e
d
if
f
er
en
t
cu
s
to
m
er
g
r
o
u
p
s
(
d
if
f
er
en
t
L
R
FM
p
atter
n
s
)
.
C
o
n
s
eq
u
e
n
tl
y
,
2
6
9
2
cu
s
to
m
er
s
h
o
w
v
is
i
ted
th
e
e
-
co
m
m
er
ce
w
eb
s
it
e
in
th
i
s
p
er
io
d
ar
e
s
elec
ted
.
A
cc
o
r
d
in
g
to
t
h
e
p
r
o
p
o
s
ed
m
o
d
el
d
escr
ib
ed
in
S
ec
tio
n
3
,
KNI
ME
A
n
al
y
tic
s
P
latf
o
r
m
3
.
3
.
2
is
u
s
ed
.
C
o
n
s
eq
u
en
tl
y
,
w
e
f
i
n
d
s
e
v
en
c
lu
s
ter
s
o
f
c
u
s
to
m
er
s
t
h
at
h
av
e
a
d
if
f
er
e
n
t
L
R
FM
b
eh
a
v
io
r
.
T
h
e
o
p
ti
m
al
n
u
m
b
e
r
o
f
clu
s
ter
(
k
=7
)
is
o
b
tain
ed
b
a
s
ed
o
n
elb
o
w
an
d
s
il
h
o
u
ette
m
et
h
o
d
s
.
Fig
u
r
e
4
s
h
o
w
s
t
h
e
p
lo
ts
o
f
th
e
SS
E
a
n
d
av
er
ag
e
s
il
h
o
u
ette
co
ef
f
icie
n
t
v
er
s
u
s
th
e
n
u
m
b
er
o
f
c
lu
s
ter
s
f
o
r
k
-
m
ea
n
s
.
A
d
i
s
tin
c
t
k
n
e
e
in
th
e
SS
E
an
d
a
d
is
tin
ct
p
ea
k
in
t
h
e
s
il
h
o
u
e
tte
co
ef
f
icie
n
t a
r
e
p
r
esen
t
w
h
e
n
t
h
e
n
u
m
b
er
o
f
cl
u
s
ter
s
is
eq
u
al
to
7
.
Fig
u
r
e
4
.
E
lb
o
w
a
n
d
A
v
er
a
g
e
s
ilh
o
u
ette
m
et
h
o
d
s
f
o
r
d
eter
m
i
n
in
g
t
h
e
o
p
ti
m
al
n
u
m
b
er
o
f
cl
u
s
ter
T
ab
le
4
is
a
s
u
m
m
ar
y
o
f
t
h
e
c
lu
s
ter
i
n
g
o
f
th
e
s
e
s
ev
e
n
cl
u
s
ter
s
,
ea
ch
w
i
th
th
e
co
r
r
esp
o
n
d
in
g
n
u
m
b
er
o
f
cu
s
to
m
er
s
,
av
er
a
g
e
le
n
g
th
(
L
)
,
av
er
a
g
e
r
ec
en
c
y
(
R
)
,
av
er
a
g
e
f
r
eq
u
en
c
y
(
F),
a
v
er
ag
e
m
o
n
etar
y
(
M)
a
n
d
t
h
e
last
co
lu
m
n
s
h
o
w
s
t
h
e
L
R
F
M
p
atter
n
f
o
r
ea
ch
c
lu
s
ter
.
Mo
s
t
o
f
t
h
e
cu
s
to
m
er
s
ar
e
in
C
lu
s
ter
s
1
,
3
an
d
5
.
W
h
er
ea
s
,
clu
s
ter
6
in
clu
d
es t
h
e
m
in
i
m
u
m
n
u
m
b
er
cu
s
to
m
er
s
(
o
n
l
y
7
7
cu
s
to
m
er
s
)
.
As
m
e
n
tio
n
ed
ea
r
lier
,
w
e
f
o
cu
s
o
u
r
s
t
u
d
y
o
n
c
u
s
to
m
er
s
b
elo
n
g
to
co
r
e
cu
s
to
m
er
s
,
an
d
(
C
lu
s
ter
2
,
3
an
d
4
)
a
n
d
th
e
h
i
gh
-
v
al
u
e
n
e
w
c
u
s
to
m
er
s
(
C
lu
s
ter
0
)
,
th
e
b
o
th
r
ep
r
esen
t
5
1
.
2
3
%
o
f
th
e
to
tal
a
v
ailab
le
cu
s
to
m
er
d
atab
ase.
T
ab
le
4
.
Descr
ip
tiv
e
Statis
tic
s
o
f
Sev
e
n
C
lu
s
ter
s
b
ased
o
n
K
-
Me
an
s
Me
t
h
o
d
i
n
T1
C
l
u
st
e
r
C
o
u
n
t
M
e
a
n
(
L
)
M
e
a
n
(
R
)
M
e
a
n
(
F
)
M
e
a
n
(
M
)
P
a
t
t
e
r
n
c
l
u
st
e
r
_
0
3
3
2
2
8
2
.
31
2
1
1
.
26
12
.
41
8
2
0
4
.
28
c
l
u
st
e
r
_
1
7
6
0
7
5
2
.
87
48
.
31
4
.
47
8
5
7
.
62
c
l
u
st
e
r
_
2
3
7
5
7
0
7
.
71
2
0
6
.
59
13
.
34
2
1
8
7
.
08
c
l
u
st
e
r
_
3
5
0
9
7
4
2
.
81
2
1
0
.
91
14
.
65
1
0
8
1
7
.
28
c
l
u
st
e
r
_
4
2
1
0
7
4
1
.
10
2
0
9
.
95
6
.
97
8
2
6
6
.
72
c
l
u
st
e
r
_
5
4
2
8
6
9
9
.
51
2
1
2
.
89
3
.
93
1
0
6
3
.
69
c
l
u
st
e
r
_
6
77
35
.
52
2
1
4
.
66
2
.
91
4
0
5
.
12
I
n
th
e
s
ec
o
n
d
p
er
io
d
T
2
(
f
r
o
m
J
u
l
y
,
2
0
1
4
to
Feb
r
u
ar
y
,
2
0
1
5
)
,
w
e
in
tr
o
d
u
ce
th
e
cl
u
s
ter
as
s
i
g
n
er
n
o
d
e
(
th
at
ass
i
g
n
s
ex
is
ti
n
g
cu
s
to
m
er
s
in
T
2
t
o
th
e
ex
is
ti
n
g
g
r
o
u
p
s
,
w
h
ic
h
ar
e
o
b
tain
ed
b
y
k
-
m
ea
n
s
i
n
T
1
)
to
d
eter
m
in
e
c
u
s
to
m
er
w
h
o
h
as
m
o
v
ed
f
r
o
m
t
h
e
co
r
e
cu
s
to
m
er
in
T
1
to
th
e
d
ef
ec
to
r
cu
s
to
m
er
d
u
r
in
g
t
h
e
s
u
b
s
eq
u
en
t
p
er
io
d
o
f
eig
h
t
m
o
n
th
s
.
A
p
p
l
y
i
n
g
o
u
r
p
ar
tial
-
to
ta
l
ch
u
r
n
d
ef
i
n
itio
n
d
escr
ib
ed
in
3
.
2
s
ec
tio
n
r
esu
lts
in
2
5
4
p
a
r
tial
d
ef
ec
tio
n
s
(
1
7
.
8
1
%
=
2
5
4
/1
4
2
6
)
an
d
3
6
3
to
t
al
d
ef
ec
tio
n
s
(
2
5
.
4
5
%
=
3
6
3
/1
4
2
6
)
,
w
h
er
e
1
4
2
6
r
ep
r
esen
ts
t
h
e
n
u
m
b
er
o
f
cu
s
to
m
er
s
u
n
d
er
in
v
est
ig
atio
n
(
clu
s
ter
_
2
+
clu
s
ter
_
3
+
clu
s
ter
_
4
+
clu
s
ter
_
0
=
1
4
2
6
)
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2088
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
4
,
A
u
g
u
s
t 2
0
1
8
:
2
3
6
7
–
2
3
8
3
2376
3
.3
.
Va
ria
bles
o
pera
t
io
na
liza
t
io
n
3
.
3
.
1
.
P
re
dict
o
rs (
i
nd
epende
nt
v
a
ria
bles
)
A
m
aj
o
r
p
ar
t
o
f
th
e
ex
is
ti
n
g
s
t
u
d
ies
r
elate
d
to
t
h
e
p
r
ed
ictio
n
o
f
cu
s
to
m
er
c
h
u
r
n
f
o
cu
s
es
o
n
t
h
e
in
co
r
p
o
r
atio
n
o
f
t
w
o
g
r
o
u
p
s
o
f
in
f
o
r
m
a
tio
n
:
b
eh
a
v
io
r
al
in
f
o
r
m
atio
n
a
n
d
cu
s
to
m
er
d
e
m
o
g
r
ap
h
ics.
A
cc
o
r
d
in
g
to
s
ev
er
al
s
tu
d
ies
s
u
c
h
as
C
o
u
s
s
e
m
e
n
t
an
d
Van
d
en
P
o
el
[
5
2
]
,
Gu
ad
ag
n
i
a
n
d
L
ittl
e
[
5
3
]
;
R
o
s
s
i
et
a
l
.,
[
5
4
]
,
an
d
T
a
m
ad
d
o
n
i
J
ah
r
o
m
i
et
a
l
.,
[
7
]
d
em
o
g
r
ap
h
ic
d
ata
(
i.e
g
en
d
er
,
a
g
e,
ad
d
r
ess
,
p
r
o
f
e
s
s
i
o
n
,
etc)
h
av
e
les
s
i
m
p
ac
t
o
n
c
h
u
r
n
p
r
ed
ictio
n
.
F
o
r
th
is
,
o
u
r
s
t
u
d
y
w
ill
b
e
b
ase
d
o
n
l
y
o
n
b
e
h
av
io
r
al
i
n
f
o
r
m
at
io
n
at
t
h
e
lev
e
l
o
f
th
e
i
n
d
iv
id
u
al
c
u
s
to
m
er
(
in
d
e
p
en
d
en
t
v
ar
iab
les),
th
i
s
w
ill
a
llo
w
u
s
to
k
ee
p
th
e
m
o
d
els
i
n
th
eir
s
i
m
p
les
t
f
o
r
m
an
d
,
o
n
th
e
o
th
er
h
an
d
,
to
m
a
x
i
m
ize
t
h
eir
p
r
ed
ictiv
e
p
o
w
er
.
C
o
m
p
ar
ed
w
it
h
tr
ad
itio
n
al
tr
an
s
ac
tio
n
m
et
h
o
d
s
,
th
e
b
i
g
g
est
ad
v
an
ta
g
e
o
f
e
-
co
m
m
er
ce
is
th
at
all
t
h
e
n
av
i
g
atio
n
d
ata
o
f
all
t
h
e
v
i
s
it
s
m
ad
e
b
y
c
u
s
to
m
er
s
o
n
t
h
e
e
-
co
m
m
er
ce
s
i
te
ar
e
s
to
r
ed
in
th
e
s
er
v
er
s
.
Fro
m
t
h
i
s
b
eh
av
io
r
al
a
n
d
tr
an
s
a
ctio
n
al
i
n
f
o
r
m
atio
n
at
t
h
e
le
v
el
o
f
t
h
e
in
d
i
v
id
u
al
cu
s
to
m
er
(
p
a
g
e
v
i
e
w
ed
,
s
eq
u
en
ce
o
f
v
is
i
ts
,
p
u
r
ch
a
s
e
p
r
o
ce
s
s
,
n
u
m
b
er
o
f
tr
an
s
ac
tio
n
s
,
etc
.
.
.
)
an
d
in
ad
d
itio
n
to
R
FM
v
ar
iab
les,
m
a
n
y
i
n
d
icato
r
s
ca
n
b
e
e
x
tr
ac
ted
[
5
5
]
,
an
d
u
s
ed
as
p
r
ed
icto
r
v
ar
iab
les
by
o
u
r
m
o
d
els
to
i
m
p
r
o
v
e
t
h
ei
r
d
is
tin
c
tio
n
p
o
w
er
b
et
w
ee
n
cu
s
to
m
er
s
to
tall
y
c
h
u
r
n
an
d
th
o
s
e
w
h
o
p
ar
tiall
y
d
ef
ec
t
an
d
th
o
s
e
w
h
o
r
e
m
ai
n
l
o
y
al.
An
o
v
er
v
ie
w
o
f
all
ex
tr
ac
ted
v
ar
iab
les
u
s
ed
in
th
is
s
t
u
d
y
is
p
r
ese
n
ted
in
T
ab
le
5
.
T
ab
le
6
s
u
m
m
ar
ie
s
all
b
eh
av
i
o
u
r
al
in
d
ep
en
d
en
t
v
ar
iab
les
s
u
p
p
o
r
ted
b
y
p
r
ev
io
u
s
r
esear
ch
i
n
th
e
b
o
th
o
f
f
li
n
e
an
d
o
n
li
n
e
en
v
ir
o
n
m
e
n
ts
.
T
h
e
r
ec
en
c
y
,
f
r
eq
u
e
n
c
y
a
n
d
m
o
n
etar
y
v
ar
iab
les
ar
e
th
o
s
e
th
at
h
av
e
m
o
r
e
p
o
p
u
lar
it
y
in
p
r
ed
icti
n
g
c
u
s
to
m
er
ch
u
r
n
i
n
t
h
e
o
n
l
in
e
e
n
v
ir
o
n
m
e
n
ts
.
T
h
e
v
ar
iab
le
s
th
a
t
d
escr
ib
e
th
e
d
r
o
p
o
u
t
r
ates
at
ea
ch
s
tep
o
f
t
h
e
b
u
y
i
n
g
p
r
o
ce
s
s
,
th
e
len
g
t
h
o
f
r
elati
o
n
s
h
ip
(
L
)
,
av
er
a
g
e
i
n
ter
p
u
r
c
h
ase
t
i
m
e
(
I
T
P
)
an
d
m
o
d
e
o
f
p
a
y
m
e
n
t (
Mo
p
a
y
m
e
n
t)
ar
e
v
ar
iab
les in
f
r
eq
u
e
n
tl
y
u
s
ed
in
p
r
io
r
r
esear
ch
.
T
h
er
ef
o
r
e,
in
o
r
d
er
to
ass
ess
th
eir
i
m
p
o
r
tan
ce
i
n
p
r
ed
ictin
g
cu
s
to
m
er
ch
u
r
n
,
w
e
w
ill ta
k
e
t
h
e
m
i
n
to
ac
co
u
n
t.
T
ab
le
5
.
P
r
ed
icto
r
V
ar
iab
les
V
a
r
i
a
b
l
e
t
y
p
e
V
a
r
i
a
b
l
e
n
a
me
D
e
scri
p
t
i
o
n
R
e
c
e
n
c
y
R
N
u
mb
e
r
o
f
d
a
y
s b
e
t
w
e
e
n
t
h
e
f
i
r
st
d
a
y
o
f
t
h
e
st
u
d
y
p
e
r
i
o
d
a
n
d
t
h
e
d
a
y
o
f
t
h
e
l
a
st
p
u
r
c
h
a
se
i
n
c
a
l
i
b
r
a
t
i
o
n
p
e
r
i
o
d
(
0
<
=
R
<
=
T
1
)
.
F
r
e
q
u
e
n
c
y
F
N
u
mb
e
r
o
f
p
u
r
c
h
a
se
s o
b
se
r
v
i
n
g
d
u
r
i
n
g
t
h
e
c
a
l
i
b
r
a
t
i
o
n
p
e
r
i
o
d
(
T
1
)
.
R
_
c
h
a
n
g
e
.
F
R
e
l
a
t
i
v
e
c
h
a
n
g
e
i
n
n
u
mb
e
r
o
f
p
u
r
c
h
a
s
e
s
i
n
t
h
e
se
c
o
n
d
h
a
l
f
o
f
t
h
e
c
a
l
i
b
r
a
t
i
o
n
p
e
r
i
o
d
F
.
T
1
.
2
w
h
e
n
c
o
mp
a
r
e
d
w
i
t
h
t
h
e
f
i
r
st
h
a
l
f
o
f
t
h
e
c
a
l
i
b
r
a
t
i
o
n
p
e
r
i
o
d
F
.
T
1
.
1
,
i
.
e
.
R
_
c
h
a
n
g
e
.
F
=
(
F
.
T
1
.
2
-
F
.
T
1
.
1
)
/
F
.
T
.
11
M
o
n
e
t
a
r
y
M
T
o
t
a
l
mo
n
e
t
a
r
y
a
mo
u
n
t
o
f
p
u
r
c
h
a
se
s i
n
c
a
l
i
b
r
a
t
i
o
n
p
e
r
i
o
d
(
T
1
)
.
L
e
n
g
t
h
o
f
r
e
l
a
t
i
o
n
sh
i
p
L
N
u
mb
e
r
o
f
d
a
y
s fr
o
m t
h
e
f
i
r
st
t
o
t
h
e
l
a
st
p
u
r
c
h
a
se
.
I
n
t
e
r
p
u
r
c
h
a
se
t
i
me
I
TP
A
v
e
r
a
g
e
n
u
m
b
e
r
o
f
d
a
y
s b
e
t
w
e
e
n
p
u
r
c
h
a
se
s
.
R
_
c
h
a
n
g
e
.
I
T
P
R
e
l
a
t
i
v
e
c
h
a
n
g
e
i
n
i
n
t
e
r
p
u
r
c
h
a
se
t
i
me
i
n
t
h
e
se
c
o
n
d
h
a
l
f
o
f
t
h
e
c
a
l
i
b
r
a
t
i
o
n
p
e
r
i
o
d
I
T
P
.
T
1
.
2
w
h
e
n
c
o
mp
a
r
e
d
w
i
t
h
t
h
e
f
i
r
st
h
a
l
f
o
f
t
h
e
c
a
l
i
b
r
a
t
i
o
n
p
e
r
i
o
d
I
TP.
T
1
.
1
,
i
.
e
.
R
_
c
h
a
n
g
e
.
I
T
P
=
(
I
TP.
T
1
.
2
-
I
T
P
.
T
1
.
1
)
/
I
T
P
.
T
.
11
M
o
d
e
o
f
p
a
y
me
n
t
M
o
p
a
y
me
n
t
I
n
d
i
c
a
t
e
s t
h
e
mo
st
mo
d
e
o
f
p
a
y
me
n
t
u
se
d
i
n
t
h
e
l
a
st
t
h
r
e
e
t
r
a
n
sa
c
t
i
o
n
s
.
D
r
o
p
o
u
t
r
a
t
e
L
a
st
_
se
ssi
o
n
_
a
b
a
n
d
o
n
e
d
I
n
d
i
c
a
t
e
s w
h
e
t
h
e
r
t
h
e
l
a
s
t
se
ssi
o
n
w
a
s
a
b
a
n
d
o
n
e
d
a
t
c
h
e
c
k
o
u
t
st
e
p
(
y
e
s,n
o
n
)
.
a
b
a
n
d
_
r
a
t
e
(
a
l
l
v
i
si
t
t
o
p
r
o
d
u
c
t
v
i
e
w
s)
T
1
.
2
T
h
e
p
e
r
c
e
n
t
o
f
se
ssi
o
n
s
t
h
a
t
a
b
a
n
d
o
n
e
d
t
h
e
b
u
y
i
n
g
p
r
o
c
e
ss a
t
t
h
e
"
P
r
o
d
u
c
t
v
i
e
w
s
"
st
e
p
i
n
t
h
e
se
c
o
n
d
h
a
l
f
o
f
t
h
e
c
a
l
i
b
r
a
t
i
o
n
p
e
r
i
o
d
T
1
.
2
.
a
b
a
n
d
_
r
a
t
e
(
p
r
o
d
u
c
t
v
i
e
w
s
t
o
a
d
d
c
a
r
t
)
T
1
.
2
A
b
a
n
d
o
n
me
n
t
r
a
t
e
f
o
r
a
c
u
s
t
o
me
r
w
h
e
n
mo
v
i
n
g
f
r
o
m "
P
r
o
d
u
c
t
v
i
e
w
s"
st
e
p
t
o
"
A
d
d
t
o
c
a
r
t
"
s
t
e
p
i
n
t
h
e
se
c
o
n
d
h
a
l
f
o
f
t
h
e
c
a
l
i
b
r
a
t
i
o
n
p
e
r
i
o
d
T
1
.
2
.
a
b
a
n
d
_
r
a
t
e
(
a
d
d
c
a
r
t
t
o
c
h
e
c
k
o
u
t
)
T
1
.
2
A
b
a
n
d
o
n
me
n
t
r
a
t
e
f
o
r
a
c
u
s
t
o
me
r
w
h
e
n
mo
v
i
n
g
f
r
o
m "
A
d
d
t
o
c
a
r
t
"
st
e
p
t
o
"
C
h
e
c
k
-
o
u
t
"
s
t
e
p
i
n
t
h
e
se
c
o
n
d
h
a
l
f
o
f
t
h
e
c
a
l
i
b
r
a
t
i
o
n
p
e
r
i
o
d
T
1
.
2
.
a
b
a
n
d
_
r
a
t
e
(
c
h
e
c
k
o
u
t
t
o
t
r
a
n
s
a
c
t
i
o
n
)
T
1
.
2
A
b
a
n
d
o
n
me
n
t
r
a
t
e
f
o
r
a
c
u
s
t
o
me
r
w
h
e
n
mo
v
i
n
g
f
r
o
m C
h
e
c
k
-
o
u
t
s
t
e
p
t
o
T
r
a
n
sac
t
i
o
n
st
e
p
i
n
t
h
e
se
c
o
n
d
h
a
l
f
o
f
t
h
e
c
a
l
i
b
r
a
t
i
o
n
p
e
r
i
o
d
T
1
.
2
.
a
b
a
n
d
_
r
a
t
e
(
a
l
l
v
i
si
t
t
o
t
r
a
n
s
a
c
t
i
o
n
)
T
1
.
2
T
h
e
p
e
r
c
e
n
t
o
f
se
ssi
o
n
s
t
h
a
t
a
b
a
n
d
o
n
e
d
t
h
e
b
u
y
i
n
g
p
r
o
c
e
ss a
t
C
h
e
c
k
-
o
u
t
s
t
e
p
i
n
T
1
.
2
p
e
r
i
o
d
.
T
ab
le
6
.
B
eh
av
io
u
r
al
P
r
ed
icto
r
s
o
f
Def
ec
tio
n
an
d
T
y
p
e
o
f
T
ar
g
et
Var
iab
le
i
n
P
r
io
r
R
esear
ch
P
r
e
d
i
c
t
o
r
v
a
r
i
a
b
l
e
s
T
a
r
g
e
t
v
a
r
i
a
b
l
e
L
R
F
M
I
P
T
P
r
o
d
u
c
t
c
a
t
e
g
o
r
i
e
s
M
o
d
e
o
f
p
a
y
me
n
t
F
a
i
l
u
r
e
D
r
o
p
o
u
t
r
a
t
e
V
a
r
i
a
b
l
e
t
y
p
e
O
f
f
l
i
n
e
e
n
v
i
r
o
n
m
e
n
t
B
u
c
k
i
n
x
a
n
d
v
a
n
d
e
n
P
o
e
l
(
2
0
0
5
)
[
6
]
X
X
X
X
X
X
X
B
i
n
a
r
y
(
C
h
u
r
n
e
r
,
N
o
n
-
c
h
u
r
n
e
r
)
V
.
L
.
M
i
g
u
é
i
s,
e
t
al
(
2
0
1
2
)
[
1
0
]
X
X
X
X
B
i
n
a
r
y
(
P
a
r
t
i
a
l
l
y
c
h
u
r
n
e
d
,
N
o
n
-
c
h
u
r
n
e
r
)
M
o
z
e
r
,
e
t
a
l
.
(
2
0
0
0
)
[
1
6
]
X
X
X
X
X
X
B
i
n
a
r
y
(
C
h
u
r
n
e
r
,
N
o
n
-
c
h
u
r
n
e
r
)
Evaluation Warning : The document was created with Spire.PDF for Python.