I
nte
rna
t
io
na
l J
o
urna
l o
f
Rec
o
nfig
ura
ble a
nd
E
m
be
dd
e
d Sy
s
t
e
m
s
(
I
J
R
E
S)
Vo
l.
1
5
,
No
.
1
,
Ma
r
ch
2
0
2
6
,
p
p
.
31
~
41
I
SS
N:
2089
-
4864
,
DOI
:
1
0
.
1
1
5
9
1
/i
j
r
es.v
1
5
.
i1
.
p
p
31
-
41
31
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
r
es.ia
esco
r
e.
co
m
O
pti
m
i
zing
call c
enter
a
g
en
t
ef
ficie
ncy
t
hro
ug
h de
ep
learning
-
ba
sed cla
ss
ificatio
ns usi
ng
SMFC
C
AE
Ra
m
a
cha
n
dra
n
P
er
iy
a
s
a
m
y
1
,
M
a
nik
a
nd
a
n
G
o
v
ind
a
ra
j
i
2
,
I
.
Na
s
urull
a
3
,
V.
Srin
iv
a
s
a
n
4
,
K
.
Ra
m
a
Dev
i
5
1
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
A
p
p
l
i
c
a
t
i
o
n
,
F
I
T
C
S
,
P
a
r
u
l
U
n
i
v
e
r
si
t
y
V
a
d
o
d
a
r
a
,
I
n
d
i
a
2
D
e
p
a
r
t
me
n
t
o
f
M
a
st
e
r
o
f
C
o
m
p
u
t
e
r
A
p
p
l
i
c
a
t
i
o
n
s
,
P
r
i
y
a
d
a
r
s
h
i
E
n
g
i
n
e
e
r
i
n
g
C
o
l
l
e
g
e
,
V
a
n
i
y
a
mb
a
d
i
,
I
n
d
i
a
3
D
e
p
a
r
t
me
n
t
o
f
M
a
st
e
r
o
f
C
o
m
p
u
t
e
r
A
p
p
l
i
c
a
t
i
o
n
s
,
V
E
M
U
I
n
st
i
t
u
t
e
o
f
T
e
c
h
n
o
l
o
g
y
,
C
h
i
t
t
o
o
r
,
I
n
d
i
a
4
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
A
p
p
l
i
c
a
t
i
o
n
s,
D
a
y
a
n
a
n
d
a
S
a
g
a
r
C
o
l
l
e
g
e
o
f
En
g
i
n
e
e
r
i
n
g
,
B
e
n
g
a
l
u
r
u
,
I
n
d
i
a
5
D
e
p
a
r
t
me
n
t
o
f
I
n
f
o
r
mat
i
o
n
T
e
c
h
n
o
l
o
g
y
,
P
a
n
i
mal
a
r
E
n
g
i
n
e
e
r
i
n
g
C
o
l
l
e
g
e
,
A
n
n
a
U
n
i
v
e
r
si
t
y
,
C
h
e
n
n
a
i
,
I
n
d
i
a
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Au
g
1
8
,
2
0
2
4
R
ev
i
s
ed
J
an
1
7
,
2
0
2
5
A
cc
ep
ted
Ma
r
2
9
,
2
0
2
5
Ca
ll
c
e
n
ters
a
re
v
it
a
l
to
b
u
sin
e
ss
o
p
e
ra
ti
o
n
s
w
o
rld
w
id
e
,
a
c
ti
n
g
a
s
t
h
e
p
rim
a
r
y
in
terf
a
c
e
b
e
t
w
e
e
n
c
o
m
p
a
n
ies
a
n
d
th
e
ir
c
u
sto
m
e
rs.
T
h
e
y
h
a
n
d
le
c
u
sto
m
e
r
in
q
u
iri
e
s,
m
a
n
a
g
e
c
o
m
p
lain
ts,
a
n
d
f
a
c
il
it
a
te
tele
p
h
o
n
ic
sa
les
,
m
a
k
in
g
th
e
m
e
ss
e
n
ti
a
l
to
c
u
sto
m
e
r
s
e
rv
ic
e
.
H
o
w
e
v
e
r,
e
n
su
rin
g
q
u
a
li
ty
in
th
e
c
a
ll
c
e
n
ter
in
d
u
stry
re
m
a
in
s
c
h
a
ll
e
n
g
in
g
,
p
r
im
a
ril
y
d
u
e
to
th
e
h
e
a
v
y
re
li
a
n
c
e
o
n
c
a
ll
c
e
n
ter
re
p
re
se
n
tativ
e
s
(CS
Rs) w
h
o
m
a
n
a
g
e
h
ig
h
v
o
lu
m
e
s
o
f
c
a
ll
s.
T
ra
d
it
io
n
a
l
m
e
th
o
d
s
o
f
e
v
a
lu
a
ti
n
g
CS
R
p
e
rf
o
rm
a
n
c
e
o
f
ten
re
l
y
o
n
m
a
n
u
a
l
a
ss
e
ss
m
e
n
ts
o
f
s
m
a
ll
c
a
ll
sa
m
p
le
s,
w
h
ich
c
a
n
b
e
ti
m
e
-
c
o
n
su
m
in
g
a
n
d
li
m
it
e
d
in
sc
o
p
e
.
W
it
h
th
e
a
d
v
a
n
c
e
m
e
n
t
o
f
d
e
e
p
lea
rn
i
n
g
tec
h
n
i
q
u
e
s
(DL
T
s),
th
e
re
is
a
n
o
p
p
o
rtu
n
it
y
to
m
o
re
a
c
c
u
ra
tel
y
a
ss
e
ss
CS
R
p
e
rf
o
rm
a
n
c
e
.
T
h
is
stu
d
y
in
tr
o
d
u
c
e
s
th
e
se
lec
ti
n
g
m
in
i
m
a
l
fe
a
tu
r
e
s
f
o
r
c
a
ll
c
e
n
ter
a
g
e
n
ts
e
f
f
ici
e
n
c
y
(S
M
F
CCE)
a
p
p
ro
a
c
h
,
w
h
ich
o
p
ti
m
ize
s
f
e
a
t
u
re
se
lec
ti
o
n
f
ro
m
CS
R
d
a
ta
t
o
e
n
h
a
n
c
e
c
las
si
f
ica
ti
o
n
a
c
c
u
ra
c
y
a
n
d
sp
e
e
d
.
T
h
e
p
ro
p
o
se
d
m
e
th
o
d
a
c
h
iev
e
s
a
p
p
ro
x
im
a
tel
y
8
5
%
a
c
c
u
ra
c
y
,
o
ff
e
rin
g
v
a
lu
a
b
le
in
sig
h
ts
a
n
d
re
c
o
m
m
e
n
d
a
ti
o
n
s f
o
r
im
p
ro
v
in
g
o
v
e
ra
ll
c
a
ll
c
e
n
ter o
p
e
ra
ti
o
n
s.
K
ey
w
o
r
d
s
:
C
all
ce
n
tr
e
C
all
ce
n
tr
e
ag
en
ts
Dee
p
lear
n
in
g
E
n
s
e
m
b
les
Mo
tiv
atio
n
XGB
o
o
s
t
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
:
Ma
n
i
k
an
d
a
n
Go
v
i
n
d
ar
aj
i
Dep
ar
t
m
en
t o
f
Ma
s
ter
o
f
C
o
m
p
u
ter
A
p
p
licatio
n
s
,
P
r
i
y
ad
ar
s
h
i E
n
g
i
n
ee
r
i
n
g
C
o
lleg
e
Van
i
y
a
m
b
ad
i,
T
am
iln
ad
u
6
3
5
7
5
1
,
I
n
d
ia
E
m
ail: d
r
m
a
n
i
k
an
d
a
n
g
@
p
r
i
y
a
d
ar
s
h
in
i.
n
et.
i
n
1.
I
NT
RO
D
UCT
I
O
N
T
h
e
p
r
im
ar
y
o
b
j
ec
tiv
e
o
f
m
a
n
ag
i
n
g
cu
s
to
m
er
m
ar
k
ets
i
s
to
cu
ltiv
ate
lo
y
al
c
u
s
to
m
er
s
a
n
d
estab
lis
h
s
tr
o
n
g
r
elatio
n
s
h
ip
s
[
1
]
.
B
u
s
in
ess
es
ai
m
to
m
ai
n
tai
n
th
ese
r
e
latio
n
s
h
ip
s
b
y
u
n
d
er
s
ta
n
d
in
g
c
u
s
to
m
er
n
ee
d
s
an
d
r
esp
o
n
d
in
g
e
f
f
ec
ti
v
el
y
,
e
n
s
u
r
i
n
g
c
u
s
to
m
er
lo
y
alt
y
.
W
h
ile
f
u
n
d
a
m
e
n
tal
m
ar
k
et
in
g
s
tr
ate
g
ie
s
ai
m
ed
at
attr
ac
ti
n
g
n
e
w
c
u
s
to
m
er
s
ca
n
b
e
e
x
p
en
s
iv
e
a
n
d
ti
m
e
-
co
n
s
u
m
i
n
g
,
t
h
er
e
h
as
b
ee
n
a
g
r
o
w
in
g
f
o
cu
s
o
n
s
er
v
ices
an
d
t
h
eir
m
ar
k
et
in
g
[
2
]
.
T
h
is
s
h
i
f
t
h
as
l
ed
to
th
e
w
id
e
s
p
r
ea
d
ad
o
p
ti
o
n
o
f
ca
ll
ce
n
ter
s
in
lar
g
e
o
r
g
an
i
za
tio
n
s
.
C
a
ll
ce
n
ter
r
ep
r
esen
tativ
e
s
(
C
SR
s
)
ar
e
ce
n
tr
al
to
cu
s
to
m
er
in
ter
ac
tio
n
s
,
r
esp
o
n
s
ib
le
f
o
r
r
ec
o
r
d
in
g
tr
an
s
ac
tio
n
s
,
h
a
n
d
lin
g
u
s
er
ca
s
e
s
,
an
d
ad
d
r
ess
i
n
g
i
s
s
u
e
s
.
Fi
g
u
r
e
1
d
ep
icts
th
e
g
r
o
w
i
n
g
g
lo
b
al
m
ar
k
et
f
o
r
ca
ll
ce
n
ter
o
u
ts
o
u
r
cin
g
,
r
ef
lecti
n
g
t
h
e
i
n
d
u
s
tr
y
'
s
e
x
p
an
d
in
g
r
o
le
in
c
u
s
to
m
er
s
er
v
ice.
T
h
e
co
m
p
etiti
v
e
la
n
d
s
ca
p
e
o
f
m
o
d
er
n
b
u
s
i
n
e
s
s
es
p
lace
s
s
i
g
n
i
f
ican
t
e
m
p
h
asi
s
o
n
m
ai
n
tai
n
in
g
s
tr
o
n
g
cu
s
to
m
er
r
elatio
n
s
h
ip
s
,
w
h
ic
h
is
ess
en
tial
f
o
r
s
u
s
tai
n
i
n
g
cu
s
to
m
er
lo
y
alt
y
a
n
d
b
u
s
i
n
ess
g
r
o
w
th
.
C
al
l
ce
n
ter
s
h
av
e
b
ec
o
m
e
cr
itical
co
m
p
o
n
en
ts
o
f
t
h
is
cu
s
to
m
er
e
n
g
ag
e
m
en
t
s
tr
ate
g
y
,
s
er
v
in
g
a
s
th
e
p
r
im
ar
y
to
u
c
h
p
o
in
t
b
et
w
ee
n
co
m
p
an
ies an
d
th
eir
clien
t
s
.
T
h
ese
ce
n
ter
s
ar
e
r
esp
o
n
s
ib
le
f
o
r
m
a
n
a
g
i
n
g
cu
s
to
m
e
r
in
q
u
ir
ies,
h
an
d
li
n
g
co
m
p
lai
n
t
s
,
an
d
ex
ec
u
ti
n
g
te
l
ep
h
o
n
ic
s
ale
s
,
th
er
eb
y
d
ir
ec
tl
y
in
f
l
u
e
n
cin
g
cu
s
to
m
er
s
atis
f
a
ctio
n
an
d
r
eten
tio
n
.
Ho
w
e
v
er
,
th
e
ca
ll c
en
ter
in
d
u
s
tr
y
f
ac
es se
v
er
al
ch
allen
g
e
s
,
p
ar
ticu
lar
l
y
i
n
en
s
u
r
in
g
th
e
co
n
s
is
te
n
t p
er
f
o
r
m
an
c
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
9
-
4864
I
n
t J
R
ec
o
n
f
i
g
u
r
ab
le
&
E
m
b
ed
d
ed
Sy
s
t
,
Vo
l.
1
5
,
No
.
1
,
Ma
r
c
h
2
0
2
6
:
31
-
41
32
o
f
C
S
R
s
.
T
r
ad
itio
n
al
m
eth
o
d
s
o
f
ev
a
lu
at
in
g
C
S
R
p
er
f
o
r
m
a
n
ce
t
y
p
icall
y
i
n
v
o
l
v
e
m
an
u
al
q
u
alitat
iv
e
ass
es
s
m
en
t
s
,
w
h
ic
h
ar
e
o
f
ten
li
m
ited
b
y
th
eir
r
elian
ce
o
n
s
m
al
l
s
a
m
p
le
s
izes
a
n
d
th
e
s
u
b
j
ec
tiv
e
n
at
u
r
e
o
f
h
u
m
a
n
j
u
d
g
m
e
n
t.
T
h
ese
m
et
h
o
d
s
ar
e
n
o
t
o
n
l
y
t
i
m
e
-
co
n
s
u
m
i
n
g
b
u
t
a
ls
o
p
r
o
n
e
to
in
co
n
s
is
te
n
cie
s
,
m
ak
in
g
it
d
i
f
f
ic
u
lt
to
ac
h
iev
e
a
h
o
lis
tic
e
v
al
u
atio
n
o
f
C
S
R
ef
f
ic
ien
c
y
.
E
x
i
s
ti
n
g
s
o
l
u
tio
n
s
f
o
r
en
h
an
c
in
g
ca
ll
ce
n
te
r
o
p
er
atio
n
s
in
clu
d
e
au
to
m
ated
ca
ll
r
ec
o
r
d
in
g
s
y
s
te
m
s
,
in
ter
ac
t
iv
e
v
o
ice
r
esp
o
n
s
e
(
I
V
R
)
tech
n
o
lo
g
ie
s
,
a
n
d
cu
s
to
m
er
f
ee
d
b
ac
k
s
u
r
v
e
y
s
[3
]
-
[
6]
.
W
h
ile
th
ese
t
o
o
ls
h
elp
s
tr
ea
m
lin
e
ce
r
tai
n
a
s
p
ec
ts
o
f
ca
ll
ce
n
ter
m
a
n
a
g
e
m
en
t,
th
e
y
o
f
te
n
f
all
s
h
o
r
t
in
p
r
o
v
id
i
n
g
a
co
m
p
r
eh
en
s
i
v
e
as
s
es
s
m
e
n
t
o
f
C
S
R
p
e
r
f
o
r
m
an
ce
.
Fo
r
in
s
ta
n
ce
,
au
to
m
ated
s
y
s
te
m
s
ca
n
h
an
d
le
b
asic
tas
k
s
b
u
t
m
a
y
n
o
t
ac
cu
r
atel
y
m
ea
s
u
r
e
th
e
q
u
alit
y
o
f
h
u
m
an
i
n
ter
ac
tio
n
s
,
w
h
ic
h
is
cr
u
cial
f
o
r
cu
s
to
m
er
s
atis
f
ac
tio
n
[7
]
-
[
12]
.
T
h
e
m
aj
o
r
co
n
s
tr
ain
ts
i
n
t
h
is
f
ield
in
clu
d
e
t
h
e
h
i
g
h
v
o
l
u
m
e
o
f
ca
lls
t
h
at
m
u
s
t
b
e
m
an
a
g
ed
,
t
h
e
p
r
ess
u
r
e
o
n
C
S
R
s
to
p
er
f
o
r
m
co
n
s
i
s
te
n
tl
y
u
n
d
e
r
d
em
a
n
d
i
n
g
co
n
d
itio
n
s
,
a
n
d
t
h
e
n
ee
d
f
o
r
ac
c
u
r
ate
an
d
ti
m
el
y
p
er
f
o
r
m
a
n
ce
as
s
e
s
s
m
en
ts
.
Mo
r
eo
v
er
,
th
e
v
ar
ia
b
ilit
y
i
n
cu
s
to
m
er
ex
p
ec
tat
io
n
s
a
n
d
th
e
d
y
n
a
m
ic
n
atu
r
e
o
f
cu
s
to
m
er
i
n
ter
ac
tio
n
s
ad
d
f
u
r
t
h
er
co
m
p
lex
i
t
y
to
e
v
alu
ati
n
g
C
S
R
e
f
f
ec
t
iv
e
n
es
s
.
T
h
r
o
u
g
h
o
u
r
r
es
ea
r
c
h
,
w
e
ai
m
to
ad
d
r
ess
th
ese
ch
alle
n
g
e
s
b
y
in
tr
o
d
u
ci
n
g
a
n
o
v
el
ap
p
r
o
ac
h
n
a
m
ed
s
elec
tin
g
m
in
i
m
al
f
ea
t
u
r
es
f
o
r
ca
ll
ce
n
ter
ag
e
n
ts
e
f
f
icien
c
y
(
SMF
C
C
E
)
.
T
h
is
ap
p
r
o
ac
h
lev
er
ag
e
s
d
ee
p
lear
n
in
g
tec
h
n
iq
u
e
s
(
DL
T
s
)
to
o
p
tim
ize
t
h
e
f
ea
t
u
r
e
s
elec
tio
n
p
r
o
ce
s
s
f
r
o
m
C
S
R
d
ata,
r
esu
lti
n
g
i
n
f
aster
an
d
m
o
r
e
ac
cu
r
ate
cla
s
s
i
f
icatio
n
o
f
C
SR
p
er
f
o
r
m
a
n
ce
.
Ou
r
o
b
j
ec
tiv
e
is
to
en
h
an
ce
t
h
e
o
v
er
all
ef
f
icie
n
c
y
o
f
ca
ll
ce
n
ter
o
p
er
atio
n
s
b
y
p
r
o
v
id
in
g
ac
tio
n
ab
le
in
s
i
g
h
ts
a
n
d
r
ec
o
m
m
en
d
atio
n
s
th
a
t
ca
n
lead
to
s
ig
n
i
f
ic
an
t
i
m
p
r
o
v
e
m
e
n
ts
i
n
cu
s
to
m
er
s
er
v
ice
q
u
alit
y
.
Sp
ec
if
icall
y
,
w
e
ai
m
to
ac
h
ie
v
e
h
ig
h
er
ac
cu
r
ac
y
i
n
p
er
f
o
r
m
a
n
ce
class
i
f
icat
io
n
,
th
er
eb
y
e
n
a
b
lin
g
ca
ll c
en
ter
s
to
m
ak
e
d
ata
-
d
r
i
v
en
d
ec
is
io
n
s
in
m
an
a
g
i
n
g
an
d
tr
ain
in
g
th
eir
C
SR
s
.
T
h
is
w
o
r
k
p
r
o
p
o
s
es
SMFC
C
E
s
c
h
e
m
a
w
h
ic
h
ef
f
ec
tiv
e
l
y
cla
s
s
i
f
ies
C
S
R
s
q
u
alitie
s
f
o
r
p
r
o
v
id
in
g
s
u
g
g
e
s
t
io
n
s
th
at
ca
n
en
h
a
n
ce
ca
ll
ce
n
tr
e’
s
ef
f
ec
ti
v
en
e
s
s
.
Fo
llo
w
i
n
g
t
h
is
in
tr
o
d
u
cto
r
y
s
e
ctio
n
,
s
ec
tio
n
2
d
is
c
u
s
s
e
s
t
h
e
l
iter
atu
r
e
r
ev
ie
w
.
Sectio
n
3
d
is
cu
s
s
es
t
h
e
m
et
h
o
d
in
tr
o
d
u
ce
d
in
th
e
s
y
s
te
m
.
Sect
io
n
4
d
etails
o
n
o
b
t
ain
ed
r
esu
lts
w
i
th
d
is
c
u
s
s
io
n
s
w
h
ile
s
ec
t
i
o
n
5
co
n
clu
d
es
th
is
p
ap
er
.
Fig
u
r
e
1
.
Glo
b
al
ca
ll c
en
ter
o
u
ts
o
u
r
cin
g
m
ar
k
et
[
1
3]
C
u
s
to
m
er
s
o
f
te
n
r
ea
ch
o
u
t
to
a
co
m
p
a
n
y
's
h
elp
d
esk
,
w
h
ic
h
is
m
an
a
g
ed
b
y
C
S
R
s
at
ca
ll
ce
n
t
er
s
.
T
h
ese
in
ter
ac
tio
n
s
,
in
c
lu
d
i
n
g
ca
lls
a
n
d
d
is
cu
s
s
io
n
s
,
ar
e
a
u
to
m
atica
ll
y
r
ec
o
r
d
ed
f
o
r
f
u
t
u
r
e
r
ef
er
en
c
e.
C
SR
s
ar
e
tas
k
ed
w
it
h
i
m
m
ed
iate
l
y
as
s
is
ti
n
g
cu
s
to
m
er
s
,
w
h
o
m
a
y
al
s
o
in
ter
ac
t
w
it
h
au
to
m
ated
s
y
s
te
m
s
li
k
e
I
VR
s
th
a
t
g
r
ee
t
th
e
m
an
d
o
f
f
er
o
p
tio
n
s
s
u
c
h
as
li
s
te
n
in
g
to
ad
v
er
tis
e
m
e
n
t
s
o
r
p
r
o
m
o
tio
n
al
o
f
f
er
s
.
De
s
p
ite
th
e
s
e
s
y
s
te
m
s
,
cu
s
to
m
er
s
f
r
eq
u
en
tl
y
e
x
p
er
ien
ce
f
r
u
s
tr
at
io
n
d
u
e
to
lo
n
g
w
ait
ti
m
es,
i
n
co
n
s
is
te
n
t
s
er
v
ice,
an
d
i
n
ad
eq
u
ate
s
o
lu
t
io
n
s
.
A
co
m
m
o
n
co
m
p
lai
n
t
is
t
h
e
lack
o
f
co
n
s
i
s
ten
c
y
i
n
th
e
s
er
v
ice,
a
s
cu
s
to
m
er
s
ca
n
n
o
t
p
r
ed
ict
w
h
i
ch
C
S
R
w
ill
h
a
n
d
le
th
eir
ca
ll.
C
u
s
to
m
er
s
h
a
v
e
ex
p
r
ess
ed
d
is
s
at
is
f
ac
tio
n
f
o
r
v
ar
io
u
s
r
ea
s
o
n
s
,
i
n
cl
u
d
in
g
d
i
f
f
icu
l
t
y
h
ea
r
in
g
o
r
u
n
d
er
s
ta
n
d
in
g
th
e
C
SR
,
b
ein
g
d
is
co
n
n
ec
ted
d
u
r
in
g
tr
an
s
f
er
s
,
o
r
r
ec
eiv
in
g
p
o
o
r
s
er
v
ice
d
u
e
to
th
e
u
s
e
o
f
j
ar
g
o
n
,
ex
ten
d
ed
h
o
ld
t
i
m
e
s
,
o
r
r
u
d
e
b
eh
av
io
r
.
T
h
ese
ch
a
llen
g
es
h
i
g
h
li
g
h
t
th
e
n
ee
d
to
i
m
p
r
o
v
e
t
h
e
s
tan
d
ar
d
s
o
f
C
S
R
s
o
p
er
atin
g
i
n
ca
ll
ce
n
ter
s
.
C
u
s
t
o
m
er
s
w
a
n
t
to
k
n
o
w
th
e
m
o
s
t
i
m
p
o
r
tan
t
f
ac
to
r
s
t
h
at
co
n
tr
ib
u
te
to
C
S
R
s
u
cc
es
s
to
en
h
a
n
ce
b
u
s
i
n
ess
o
p
er
atio
n
s
u
s
i
n
g
co
ld
ca
ll
d
ata.
T
h
is
s
tu
d
y
ai
m
s
to
p
r
o
v
id
e
s
o
l
u
tio
n
s
to
i
m
p
r
o
v
e
t
h
e
q
u
alit
y
o
f
C
S
R
s
,
t
h
er
eb
y
en
h
a
n
cin
g
t
h
e
o
v
er
all
ef
f
icien
c
y
o
f
ca
l
l c
en
ter
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
R
ec
o
n
f
i
g
u
r
ab
le
&
E
m
b
ed
d
ed
Sy
s
t
I
SS
N:
2089
-
4864
Op
timiz
in
g
ca
ll c
en
ter a
g
en
t e
fficien
cy
th
r
o
u
g
h
d
ee
p
lea
r
n
in
g
-
b
a
s
ed
…
(
R
a
ma
c
h
a
n
d
r
a
n
P
eriya
s
a
my
)
33
2.
L
I
T
E
R
AT
U
RE
R
E
VI
E
W
C
all
ce
n
tr
es
ca
n
b
e
ca
teg
o
r
ized
as
o
r
g
an
izatio
n
b
ased
o
r
as
in
d
ep
en
d
en
t
b
u
s
in
e
s
s
e
s
.
Mo
r
eo
v
er
,
ca
ll
ce
n
tr
es
ca
n
b
e
ca
teg
o
r
ized
ac
co
r
d
in
g
to
th
eir
o
r
g
an
izatio
n
al
s
tr
u
ctu
r
e
i
n
clu
d
i
n
g
s
in
g
u
lar
ce
n
tr
alize
d
ca
ll
ce
n
tr
es
o
r
s
ev
er
al
ca
ll
ce
n
tr
es
s
ca
tter
ed
ac
r
o
s
s
s
ev
er
al
s
ites
.
E
ac
h
f
o
r
m
o
f
ca
ll
ce
n
tr
es
h
as
t
h
eir
o
w
n
b
en
e
f
its
an
d
d
r
a
w
b
ac
k
s
a
n
d
h
en
ce
it
b
ec
o
m
es
i
m
p
er
ativ
e
to
co
m
p
r
eh
e
n
d
id
ea
s
,
p
r
o
ce
d
u
r
es,
an
d
h
az
ar
d
s
in
v
o
l
v
ed
w
ith
ea
c
h
f
o
r
m
.
Or
g
a
n
izatio
n
s
t
h
at
o
w
n
an
d
r
u
n
t
h
eir
o
w
n
ca
l
l
ce
n
tr
e
s
b
en
ef
it
i
n
a
v
ar
iet
y
o
f
w
a
y
s
.
W
ith
o
n
l
y
o
n
e
t
y
p
e
o
f
in
s
tit
u
tio
n
as t
h
eir
clie
n
ts
,
i
n
-
h
o
u
s
e
co
n
tact
ce
n
tr
e
s
h
a
v
e
a
d
v
an
ta
g
es o
f
b
ein
g
ab
le
to
ca
t
er
to
ea
ch
ca
ller
th
e
atten
tio
n
th
e
y
n
ee
d
.
C
S
R
s
r
ec
eiv
e
tr
ain
i
n
g
to
b
ec
o
m
e
d
ep
en
d
ab
le
b
r
an
d
am
b
ass
ad
o
r
s
an
d
ad
d
u
n
iq
u
e
to
u
c
h
es
to
o
f
f
er
ed
clie
n
t
s
er
v
ices.
T
h
e
lev
el
o
f
c
u
s
to
m
er
s
er
v
ices
m
a
y
b
e
co
m
p
r
o
m
is
ed
w
h
e
n
co
n
tact
ce
n
tr
es
ar
e
o
u
ts
o
u
r
ce
d
w
h
er
e
e
m
p
lo
y
ee
d
ev
o
tio
n
s
m
a
y
b
e
lo
w
.
I
n
-
h
o
u
s
e
ca
ll
ce
n
tr
es
r
ed
u
ce
th
e
d
an
g
er
s
o
f
d
is
clo
s
i
n
g
h
ig
h
l
y
p
r
o
tecte
d
p
er
s
o
n
al
d
ata
d
u
e
to
th
e
i
n
v
o
l
v
e
m
e
n
t
o
f
th
ir
d
p
ar
ties
.
I
n
s
p
ite
o
f
q
u
alit
y
c
h
ec
k
lis
t
s
f
o
r
p
er
f
o
r
m
a
n
ce
m
ea
s
u
r
e
m
e
n
t
s
t
h
at
ca
n
e
n
s
u
r
e
co
n
s
is
te
n
c
y
o
f
p
r
o
v
id
ed
s
er
v
ices
b
y
C
S
R
s
,
th
e
s
t
u
d
y
i
n
[
1
4
]
r
ec
o
g
n
is
ed
t
h
at
in
ab
ili
t
y
to
m
ai
n
tai
n
ef
f
icie
n
c
y
a
n
d
p
r
o
v
id
e
co
n
s
u
m
er
s
w
it
h
q
u
alit
y
s
er
v
ices
h
a
m
p
er
ed
o
p
er
atio
n
s
.
Qu
ali
t
y
ca
n
i
n
s
ti
ll
g
r
ea
ter
co
n
f
id
en
ce
to
co
n
tact
w
it
h
d
ir
ec
t
r
ep
r
esen
tativ
e
s
.
Ad
d
itio
n
all
y
,
in
ter
n
al
ca
ll
ce
n
tr
es
ar
e
m
o
r
e
ad
ap
tab
l
e
th
an
ca
ll
ce
n
tr
es
th
at
ar
e
o
u
t
s
o
u
r
ce
d
as
th
e
y
ar
e
d
ir
ec
tl
y
u
n
d
er
th
e
m
an
a
g
e
m
e
n
t
o
f
o
r
g
an
i
s
atio
n
s
,
an
y
ch
a
n
g
e
s
to
b
u
s
in
e
s
s
p
r
o
ce
d
u
r
es c
o
u
ld
b
e
h
an
d
led
r
ig
h
t a
w
a
y
[
1
5
]
.
A
lt
h
o
u
g
h
i
n
ter
n
al
co
n
tact
ce
n
t
r
es
co
u
ld
b
e
m
o
r
e
d
ep
en
d
ab
le,
s
ec
u
r
e,
an
d
ad
ap
tab
le
th
an
o
u
ts
o
u
r
ce
d
ca
ll
ce
n
tr
es.
Ma
n
y
o
r
g
a
n
izatio
n
s
ar
e
t
u
r
n
i
n
g
to
w
ar
d
s
o
u
t
s
o
u
r
ce
d
co
n
tact
ce
n
tr
es
as
t
h
e
y
ar
e
m
o
r
e
co
s
t
-
e
f
f
ec
ti
v
e
o
p
tio
n
s
th
a
n
i
n
-
h
o
u
s
e
ca
ll
ce
n
t
r
es,
s
in
ce
t
h
e
y
r
eq
u
ir
e
less
er
m
ain
te
n
an
ce
s
th
a
n
i
n
-
h
o
u
s
e
ca
ll
ce
n
tr
es.
O
u
ts
o
u
r
ce
d
ca
ll
ce
n
ter
s
also
eli
m
i
n
ate
o
r
g
an
iza
tio
n
al
n
ee
d
s
f
o
r
in
v
esti
n
g
i
n
tr
ain
i
n
g
o
f
C
S
R
s
a
n
d
w
h
e
n
ca
ll
v
o
lu
m
e
s
in
cr
ea
s
e,
s
ca
lab
ilit
y
o
f
ca
ll
ce
n
ter
s
ize
s
g
et
ea
s
il
y
b
e
ad
j
u
s
t
ed
to
m
ee
t
c
u
s
to
m
er
s
’
r
eq
u
ir
em
en
ts
.
C
all
ce
n
tr
e
s
s
er
v
e
as
co
m
p
lete
i
n
f
o
r
m
ati
o
n
n
et
w
o
r
k
s
an
d
h
a
v
e
th
e
a
d
v
an
ta
g
es
o
f
r
eq
u
ir
i
n
g
s
i
n
g
u
lar
co
m
m
u
n
ica
tio
n
p
latf
o
r
m
s
,
s
i
n
g
u
lar
c
u
s
to
m
er
r
elatio
n
s
h
ip
m
an
a
g
e
m
en
t
s
y
s
t
e
m
s
,
a
n
d
s
in
g
u
lar
r
ea
l
esta
te
e
x
p
en
d
it
u
r
es.
C
a
ll
ce
n
tr
es
m
a
y
b
e
g
i
v
e
n
au
to
n
o
m
y
,
w
it
h
ea
ch
s
ite
p
r
o
ce
s
s
i
n
g
j
u
s
t
ce
r
tain
k
i
n
d
s
o
f
ca
lls
.
Fo
r
ex
a
m
p
le
,
tec
h
n
ical
ass
is
tan
ce
m
a
y
f
al
l
u
n
d
er
th
e
p
u
r
v
ie
w
o
f
o
n
e
s
ite,
w
h
ile
b
il
lin
g
,
m
a
y
f
all
u
n
d
er
th
e
p
u
r
v
i
e
w
o
f
an
o
t
h
er
s
ite.
R
o
u
te
p
ar
ticu
lar
ca
lls
to
o
n
e
lo
ca
tio
n
,
th
en
u
s
e
o
th
er
lo
ca
tio
n
s
w
h
en
t
h
er
e
is
an
o
v
er
f
lo
w
o
r
w
h
e
n
it
is
b
e
y
o
n
d
b
u
s
i
n
ess
h
o
u
r
s
as
a
n
o
th
er
o
p
tio
n
f
o
r
d
ec
en
tr
alize
d
ca
ll
ce
n
tr
e
m
o
d
el
s
.
E
ac
h
lo
ca
tio
n
m
a
y
als
o
b
e
tr
ea
te
d
eq
u
ally
,
s
i
m
ilar
to
ce
n
tr
alize
d
lo
ca
tio
n
s
,
r
o
u
tin
g
ca
lls
t
o
th
e
n
e
x
t
av
a
il
ab
le
C
SR
[
1
6
]
.
T
h
e
o
p
tim
al
ap
p
r
o
ac
h
w
ill
d
ep
en
d
o
n
th
e
n
ee
d
s
o
f
ea
ch
h
ea
l
th
ca
r
e
in
s
tit
u
tio
n
,
alth
o
u
g
h
b
o
th
ce
n
tr
alize
d
an
d
d
ec
en
tr
alize
d
co
n
tact
ce
n
tr
es
ca
n
b
e
eith
er
in
-
h
o
u
s
e
o
r
o
u
ts
o
u
r
ce
d
.
A
h
ea
lt
h
ca
r
e
o
r
g
an
izatio
n
is
i
m
p
ac
ted
in
n
u
m
er
o
u
s
w
a
y
s
b
y
th
e
i
m
p
le
m
e
n
tatio
n
o
f
a
ca
ll
ce
n
tr
e
s
tr
ateg
y
.
T
h
e
i
m
p
ac
t
th
at
ca
ll
ce
n
tr
e
m
et
h
o
d
s
h
av
e
o
n
h
ea
lth
ca
r
e
f
ir
m
s
'
o
v
er
all
cu
s
to
m
er
s
er
v
ice
s
tan
d
ar
d
s
m
a
y
b
e
th
e
m
o
s
t
ev
id
en
t
e
f
f
ec
t.
A
s
c
u
s
to
m
er
s
er
v
ice
q
u
alit
y
r
is
es,
s
o
d
o
o
u
tco
m
es,
cu
s
to
m
er
ac
ce
s
s
ib
ilit
y
to
s
er
v
ices,
an
d
o
r
g
an
is
at
io
n
al
ex
p
e
n
s
es.
A
n
e
w
l
y
d
ep
lo
y
ed
ca
ll
ce
n
tr
e
s
tr
at
eg
y
's
o
r
g
an
i
s
atio
n
al
i
m
p
licatio
n
s
o
n
a
n
u
n
id
en
tifie
d
in
f
o
r
m
at
io
n
d
eliv
er
y
s
y
s
te
m
w
er
e
as
s
ess
ed
i
n
th
e
s
t
u
d
y
i
n
[
1
7
]
.
A
cc
o
r
d
in
g
to
th
e
s
tu
d
y
,
co
m
p
an
y
m
o
d
els
t
h
at
p
r
io
r
itis
e
co
s
t
-
c
u
tti
n
g
o
v
er
cu
s
to
m
er
s
ati
s
f
ac
tio
n
an
d
s
er
v
ice
q
u
alit
y
ar
e
m
o
r
e
s
u
cc
e
s
s
f
u
l t
h
a
n
th
o
s
e
th
a
t d
o
n
o
t.
F
u
r
th
er
m
o
r
e,
it d
is
co
v
er
ed
th
at
lo
y
a
l c
o
n
s
u
m
er
s
ar
e
th
o
s
e
t
h
at
r
ec
eiv
e
o
u
t
s
ta
n
d
in
g
cu
s
to
m
er
s
er
v
ice.
T
h
e
ad
o
p
tio
n
o
f
au
to
m
ated
ca
ll
d
is
tr
i
b
u
tio
n
tec
h
n
o
lo
g
ies
i
m
p
r
o
v
ed
co
n
tact
ce
n
tr
es
'
le
v
els
o
f
c
u
s
to
m
er
s
er
v
ice,
s
u
ch
as
th
e
g
e
n
er
al
h
ap
p
i
n
ess
o
f
p
atien
t
s
in
h
ea
lt
h
ca
r
e
s
y
s
te
m
s
,
w
h
ile
s
i
m
u
lta
n
eo
u
s
l
y
i
n
cr
ea
s
i
n
g
th
eir
co
s
t
-
e
f
f
ec
ti
v
en
e
s
s
.
I
n
th
i
s
in
s
tan
ce
,
t
h
e
av
er
ag
e
ca
ll
r
esp
o
n
s
e
ti
m
e
w
as 3
0
s
ec
o
n
d
s
,
an
d
th
e
n
u
m
b
er
o
f
ca
ller
s
w
h
o
h
u
n
g
u
p
w
a
s
u
n
d
er
5
%.
T
h
ese
m
etr
ic
s
w
er
e
u
s
ed
to
s
h
o
w
h
o
w
t
h
e
q
u
ali
t
y
o
f
t
h
e
clien
t
s
er
v
ices
g
i
v
e
n
h
ad
i
m
p
r
o
v
ed
.
An
o
th
er
ex
a
m
p
le
o
f
a
b
u
s
i
n
e
s
s
lev
er
ag
i
n
g
its
ca
ll
ce
n
tr
e
s
er
v
ice
s
to
en
h
a
n
ce
q
u
alit
y
w
a
s
th
e
i
m
p
ac
t
o
f
r
ed
esi
g
n
i
n
g
a
n
e
t
w
o
r
k
'
s
co
n
tact
ce
n
tr
e
o
n
a
p
ar
ticu
lar
u
r
o
lo
g
y
cli
n
ic
w
i
th
i
n
th
e
n
et
wo
r
k
.
B
y
i
m
p
le
m
en
ti
n
g
L
E
A
N
m
et
h
o
d
o
lo
g
y
to
r
eo
r
g
an
is
e
s
taf
f
ac
co
r
d
in
g
to
ca
ll
v
o
lu
m
e,
cr
ea
te
a
b
ac
k
u
p
ca
ll
co
v
er
ag
e
s
y
s
te
m
d
u
r
i
n
g
d
o
w
n
ti
m
e,
m
o
v
e
o
f
f
-
s
ite
ca
ll
ce
n
tr
e
ag
en
t
s
to
a
ce
n
tr
a
l
lo
ca
tio
n
,
h
ir
e
a
r
eg
i
s
ter
ed
n
u
r
s
e
to
i
m
p
le
m
e
n
t
a
tr
ia
g
e
li
n
e,
an
d
s
et
n
e
w
p
er
f
o
r
m
a
n
ce
s
ta
n
d
ar
d
s
,
th
e
ca
ll
ce
n
ter
's
ef
f
icien
c
y
w
a
s
s
ig
n
i
f
ica
n
tl
y
i
n
cr
ea
s
ed
.
A
p
p
l
y
i
n
g
L
E
A
N
a
n
d
6
Sig
m
a
m
et
h
o
d
o
lo
g
ies
to
e
n
h
an
ce
co
n
tact
ce
n
tr
e
p
er
f
o
r
m
a
n
ce
led
to
g
r
ea
ter
p
atien
t sat
is
f
ac
tio
n
a
n
d
en
h
an
ce
d
p
atien
t c
ar
e
[
1
8
]
.
C
u
s
to
m
er
s
m
a
y
h
a
v
e
m
o
r
e
ac
ce
s
s
to
th
eir
r
eq
u
ir
ed
s
er
v
ices
w
h
en
ca
ll
ce
n
tr
es
ar
e
u
s
ed
e
f
f
ec
ti
v
el
y
.
C
all
ce
n
tr
e
s
c
h
ed
u
li
n
g
tech
n
iq
u
es
al
s
o
p
la
y
i
m
p
o
r
tan
t
r
o
les
i
n
i
m
p
r
o
v
in
g
ac
ce
s
s
to
ti
m
el
y
s
er
v
ice.
T
h
e
d
em
a
n
d
f
o
r
u
n
s
u
itab
le
s
er
v
ice
s
g
et
s
r
ed
u
ce
d
w
h
ile
ac
ce
s
s
to
ap
p
r
o
p
r
iate
s
er
v
ices
d
u
r
in
g
p
r
o
p
er
ti
m
e
an
d
lo
ca
tio
n
ar
e
i
m
p
r
o
v
ed
w
ith
t
h
e
i
n
te
g
r
atio
n
o
f
i
n
f
o
r
m
atio
n
tec
h
n
o
lo
g
y
ap
p
licatio
n
s
w
it
h
in
ca
l
l
ce
n
tr
es.
A
s
a
r
es
u
lt
,
b
u
s
i
n
ess
e
s
ar
e
b
etter
eq
u
ip
p
e
d
to
b
alan
ce
th
e
n
ee
d
s
o
f
b
o
th
cu
s
to
m
er
s
a
n
d
p
r
o
v
id
er
s
.
Ma
n
y
f
i
n
d
in
g
s
f
r
o
m
ex
p
er
i
m
e
n
ts
a
n
d
o
b
s
er
v
atio
n
s
s
u
g
g
e
s
t
t
h
at
ca
r
e
f
u
l
m
o
n
ito
r
in
g
p
r
ac
tis
es
ar
e
cr
u
cial
m
ea
s
u
r
e
s
to
g
e
n
er
ati
n
g
g
o
o
d
p
er
f
o
r
m
a
n
ce
s
[
1
9
]
.
P
er
f
o
r
m
a
n
ce
m
o
n
ito
r
in
g
,
w
h
ic
h
en
ab
les e
m
p
h
a
s
is
o
n
ag
e
n
ts
'
w
o
r
k
b
eh
av
io
u
r
,
in
cl
u
d
es
ca
ll
lis
te
n
in
g
a
n
d
o
b
s
er
v
atio
n
[
2
0
]
.
A
cc
o
r
d
in
g
to
R
ic
h
ar
d
s
o
n
an
d
B
elt
(
2
0
0
1
)
,
m
o
n
ito
r
in
g
is
d
o
n
e
to
m
a
k
e
s
u
r
e
th
e
task
i
s
b
ein
g
d
o
n
e
at
th
e
p
r
o
p
er
r
ate.
A
ls
o
,
f
r
o
m
t
h
e
clie
n
t
's
p
er
s
p
ec
tiv
e,
p
er
f
o
r
m
a
n
ce
m
o
n
it
o
r
in
g
is
s
ee
n
a
s
o
n
e
o
f
th
e
p
r
i
m
ar
y
m
a
n
ag
er
ia
l
ab
ilit
ies
[
2
1
]
.
Me
asu
r
e
m
en
t
o
f
ev
er
y
a
s
p
ec
t
o
f
c
u
s
to
m
er
s
er
v
ice
r
ep
r
esen
tat
iv
e
s
'
(
C
SR
s
'
)
in
ter
ac
tio
n
s
w
it
h
clie
n
ts
w
h
ile
p
r
o
v
id
in
g
s
er
v
ices
i
s
v
ital
f
o
r
co
n
tact
ce
n
tr
e
m
a
n
a
g
er
s
an
d
s
u
p
er
v
i
s
o
r
s
[
2
2
]
.
I
n
o
r
d
er
to
s
u
s
tain
a
n
d
en
h
an
ce
cl
ien
t
lo
y
alt
y
a
n
d
r
ete
n
ti
o
n
.
I
t
is
cr
u
cial
th
at
co
n
tact
ce
n
tr
es
p
r
ac
tis
e
q
u
alit
y
ass
u
r
an
ce
an
d
m
o
n
ito
r
in
g
o
n
a
d
aily
a
n
d
h
o
u
r
l
y
b
asis
.
Fee
d
b
ac
k
f
r
o
m
m
o
n
ito
r
in
g
m
u
s
t
also
b
e
s
en
t
to
ca
ll
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
9
-
4864
I
n
t J
R
ec
o
n
f
i
g
u
r
ab
le
&
E
m
b
ed
d
ed
Sy
s
t
,
Vo
l.
1
5
,
No
.
1
,
Ma
r
c
h
2
0
2
6
:
31
-
41
34
ce
n
tr
e
ag
en
ts
d
u
r
in
g
an
o
f
f
li
n
e
s
ess
io
n
.
T
h
e
r
esear
ch
Au
d
r
e
y
in
[
2
3
]
ca
m
e
to
th
e
co
n
clu
s
io
n
th
at
co
n
tact
ce
n
tr
e
m
an
a
g
e
m
e
n
t
s
h
o
u
ld
tak
e
a
m
o
r
e
ac
tiv
e
r
o
le
b
y
co
llab
o
r
atin
g
m
o
r
e
clo
s
el
y
w
i
th
t
h
eir
ca
ll
ce
n
tr
e
ag
e
n
ts
.
A
s
a
r
esu
lt,
m
a
n
a
g
er
s
o
r
s
u
p
er
v
is
o
r
s
w
o
u
ld
b
e
ab
le
to
d
eter
m
i
n
e
th
e
p
r
ec
is
e
co
ac
h
in
g
a
n
d
tr
ain
i
n
g
n
ee
d
ed
f
o
r
ea
ch
ag
en
t
a
n
d
r
ec
eiv
e
a
id
in
e
n
h
a
n
cin
g
t
h
e
ab
ilit
ies
o
f
th
eir
p
er
s
o
n
n
e
l.
A
cc
o
r
d
in
g
to
d
atab
r
id
g
e
m
ar
k
et
r
esear
c
h
(
2
0
2
4
)
,
th
e
g
lo
b
al
ca
ll
ce
n
ter
o
u
ts
o
u
r
ci
n
g
m
ar
k
et
is
e
x
p
er
ien
ci
n
g
s
i
g
n
i
f
ica
n
t
g
r
o
w
t
h
,
d
r
iv
en
b
y
in
cr
ea
s
ed
d
em
a
n
d
f
o
r
cu
s
to
m
er
s
er
v
ice
s
o
lu
tio
n
s
[
2
4
]
.
3.
M
E
T
H
O
D
A
ca
l
l
ce
n
ter
is
a
n
es
s
e
n
tial
co
m
p
o
n
en
t
o
f
c
u
s
to
m
er
s
er
v
ice
o
p
er
atio
n
s
,
w
h
er
e
C
S
R
s
m
a
n
ag
e
in
co
m
in
g
an
d
o
u
tg
o
i
n
g
ca
ll
s
,
ad
d
r
ess
in
g
in
q
u
ir
ies
ab
o
u
t
p
r
o
d
u
cts
o
r
s
er
v
ices,
r
eso
lv
i
n
g
co
m
p
la
i
n
ts
,
an
d
p
r
o
v
id
in
g
s
u
p
p
o
r
t.
T
h
e
ef
f
ec
ti
v
e
n
es
s
o
f
C
S
R
s
d
ir
ec
tl
y
i
m
p
ac
t
s
c
u
s
to
m
e
r
ex
p
er
ie
n
ce
,
m
a
k
i
n
g
it
cr
u
cial
f
o
r
th
e
m
to
b
e
k
n
o
w
led
g
ea
b
le,
e
m
p
at
h
etic,
an
d
r
esp
o
n
s
i
v
e.
T
h
is
s
t
u
d
y
i
n
tr
o
d
u
ce
s
t
h
e
SM
FC
C
E
s
c
h
e
m
a,
a
f
r
a
m
e
w
o
r
k
d
esig
n
ed
to
ev
alu
ate
th
e
q
u
a
li
t
y
an
d
p
r
o
d
u
ctiv
it
y
o
f
C
S
R
s
u
s
in
g
D
L
T
s
.
T
h
e
SMFC
C
E
m
e
th
o
d
o
lo
g
y
in
c
lu
d
es
s
ev
er
al
s
ta
g
e
s
,
s
u
c
h
as
e
x
p
lo
r
ato
r
y
d
ata
a
n
al
y
s
i
s
(
E
D
A
)
,
d
a
ta
clea
n
i
n
g
,
f
ea
tu
r
e
e
x
tr
ac
tio
n
,
tr
ain
/tes
t
s
p
lit
tin
g
,
an
d
class
i
f
icatio
n
u
s
i
n
g
e
n
s
e
m
b
le
m
et
h
o
d
s
.
Fi
g
u
r
e
2
ill
u
s
tr
ates
th
e
o
v
er
all
m
et
h
o
d
o
lo
g
y
f
o
llo
w
ed
b
y
t
h
e
SMFC
C
E
s
c
h
e
m
a.
Fig
u
r
e
2
.
SMFC
C
E
s
c
h
e
m
a
’
s
m
et
h
o
d
o
lo
g
y
3
.
1
.
E
x
plo
ra
t
o
ry
da
t
a
a
na
l
y
s
is
a
nd
da
t
a
clea
nin
g
E
DA
i
s
a
v
ital
d
ata
an
al
y
s
is
te
ch
n
iq
u
e
th
a
t
o
f
te
n
e
m
p
lo
y
s
v
i
s
u
al
to
o
ls
to
th
o
r
o
u
g
h
l
y
ex
a
m
i
n
e
d
atasets
.
T
h
is
s
tep
h
e
lp
s
id
en
t
if
y
s
ig
n
i
f
i
ca
n
t
p
atter
n
s
,
r
elatio
n
s
h
ip
s
b
et
w
ee
n
v
ar
iab
les,
an
d
p
o
ten
tial
a
n
o
m
alie
s
t
h
at
co
u
ld
af
f
ec
t th
e
ac
c
u
r
ac
y
o
f
m
ac
h
in
e
lear
n
in
g
m
o
d
els.
Data
clea
n
i
n
g
is
cr
u
cial
i
n
t
h
is
p
h
a
s
e,
as it
in
v
o
l
v
es r
e
m
o
v
i
n
g
in
co
r
r
ec
t
v
ar
iab
les,
h
a
n
d
li
n
g
m
is
s
i
n
g
v
al
u
es,
a
n
d
eli
m
i
n
ati
n
g
o
u
tlier
s
th
at
co
u
ld
d
is
to
r
t
th
e
r
esu
lt
s
.
Fo
r
in
s
ta
n
ce
,
co
lu
m
n
s
w
i
th
m
o
r
e
th
a
n
1
5
% m
is
s
i
n
g
d
ata
ar
e
t
y
p
icall
y
ex
c
l
u
d
ed
f
r
o
m
th
e
a
n
al
y
s
i
s
.
3
.
2
.
Descript
iv
e
s
t
a
t
is
t
ics
T
o
s
u
m
m
ar
ize
th
e
ce
n
tr
al
ten
d
en
c
y
o
f
t
h
e
o
b
s
er
v
ed
d
ata,
th
e
ar
ith
m
etic
m
ea
n
(
av
er
ag
e)
is
e
m
p
lo
y
ed
,
w
h
ic
h
r
ep
r
esen
ts
th
e
s
u
m
o
f
all
in
d
iv
id
u
al
o
b
s
er
v
atio
n
s
d
iv
id
ed
b
y
th
e
to
tal
n
u
m
b
er
o
f
s
a
m
p
les.
I
t
is
m
at
h
e
m
a
ticall
y
ex
p
r
ess
ed
as
:
Me
an
(
av
er
ag
e
)
:
̅
=
1
∑
=
1
W
h
er
e
̅
is
th
e
m
ea
n
,
ar
e
th
e
d
ata
p
o
in
ts
,
an
d
n
n
n
is
t
h
e
n
u
m
b
er
o
f
d
ata
p
o
in
ts
.
Me
d
ian
:
th
e
m
id
d
le
v
al
u
e
w
h
e
n
d
ata
p
o
in
ts
ar
e
ar
r
an
g
ed
in
a
s
ce
n
d
i
n
g
o
r
d
er
.
Mo
d
e:
t
h
e
m
o
s
t
f
r
eq
u
en
tl
y
o
cc
u
r
r
in
g
v
al
u
e
i
n
th
e
d
atase
t.
Stan
d
ar
d
d
ev
iatio
n
(σ
)
:
=
√
1
∑
(
)
−
=
1
̅
)
2
W
h
er
e
σ
m
ea
s
u
r
es th
e
a
m
o
u
n
t
o
f
v
ar
iatio
n
o
r
d
is
p
er
s
io
n
o
f
a
s
et
o
f
v
al
u
es.
Dete
ctin
g
o
u
tl
ier
s
:
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
R
ec
o
n
f
i
g
u
r
ab
le
&
E
m
b
ed
d
ed
Sy
s
t
I
SS
N:
2089
-
4864
Op
timiz
in
g
ca
ll c
en
ter a
g
en
t e
fficien
cy
th
r
o
u
g
h
d
ee
p
lea
r
n
in
g
-
b
a
s
ed
…
(
R
a
ma
c
h
a
n
d
r
a
n
P
eriya
s
a
my
)
35
Z
-
s
co
r
e:
=
−
W
h
er
e
x
is
a
d
ata
p
o
in
t,
μ
is
th
e
m
ea
n
,
an
d
σ
is
th
e
s
tan
d
ar
d
d
ev
iatio
n
.
A
Z
-
s
co
r
e
g
r
ea
ter
th
an
3
o
r
less
th
an
-
3
is
t
y
p
icall
y
co
n
s
id
er
ed
an
o
u
tli
er
.
I
n
ter
q
u
ar
tile r
an
g
e
(
I
QR
)
: I
Q
R
=Q
3
−Q
1
Ou
tlier
s
ar
e
o
f
te
n
d
ef
in
ed
as
v
alu
es b
elo
w
Q1
−
1
.
5
×I
QR
o
r
a
b
o
v
e
Q3
+1
.
5
×I
QR
.
Han
d
lin
g
m
is
s
i
n
g
v
al
u
es
:
I
m
p
u
tatio
n
w
it
h
m
ea
n
/
m
ed
ian
: r
ep
lace
m
i
s
s
i
n
g
v
al
u
es
w
i
th
t
h
e
m
ea
n
o
r
m
ed
ian
o
f
th
e
co
l
u
m
n
.
I
f
is
m
is
s
in
g
,
r
ep
lace
w
it
h
x
ˉ
o
r
th
e
m
ed
ian
.
Dr
o
p
p
in
g
r
o
w
s
/co
lu
m
n
s
:
r
e
m
o
v
e
r
o
w
s
o
r
co
lu
m
n
s
w
i
th
m
i
s
s
in
g
v
alu
e
s
b
e
y
o
n
d
a
ce
r
tain
t
h
r
es
h
o
ld
(
e.
g
.
,
m
o
r
e
th
an
1
5
%
m
is
s
i
n
g
d
ata)
.
No
r
m
a
lizatio
n
:
m
i
n
-
m
a
x
s
ca
li
n
g
:
′
=
−
m
i
n
(
)
m
ax
(
)
−
m
i
n
(
)
W
h
er
e
x
′ is
t
h
e
n
o
r
m
alize
d
v
al
u
e.
−
Z
-
s
co
r
e
n
o
r
m
al
izatio
n
:
′
=
−
Featu
r
e
e
x
tr
ac
tio
n
s
:
f
e
a
tu
r
e
e
x
tr
ac
tio
n
r
ed
u
ce
s
t
h
e
d
i
m
e
n
s
i
o
n
alit
y
o
f
t
h
e
d
ataset
b
y
s
u
m
m
ar
izi
n
g
t
h
e
o
r
ig
i
n
a
l
f
ea
t
u
r
es
in
to
a
m
o
r
e
co
m
p
a
ct
s
et.
T
ec
h
n
iq
u
es
s
u
c
h
as
p
r
in
cip
al
co
m
p
o
n
e
n
t
an
al
y
s
is
(
P
C
A
)
an
d
lin
ea
r
d
is
cr
i
m
i
n
an
t a
n
al
y
s
is
(
L
D
A
)
ar
e
em
p
lo
y
ed
to
ac
h
iev
e
t
h
is
:
−
P
C
A
:
i
n
v
o
l
v
es
ca
lc
u
lati
n
g
t
h
e
co
v
ar
ian
ce
m
atr
ix
,
s
o
l
v
i
n
g
f
o
r
eig
en
v
a
lu
e
s
an
d
eig
e
n
v
ec
to
r
s
,
an
d
s
elec
tin
g
p
r
in
cip
al
co
m
p
o
n
en
t
s
.
−
L
D
A
:
f
o
c
u
s
es
o
n
m
ax
i
m
izin
g
th
e
r
atio
o
f
b
et
w
ee
n
-
clas
s
v
a
r
ian
ce
to
w
it
h
in
-
cla
s
s
v
ar
ia
n
c
e
f
o
r
i
m
p
r
o
v
ed
s
ep
ar
ab
ilit
y
.
R
eg
u
lar
izatio
n
tech
n
iq
u
es,
in
c
lu
d
in
g
L
1
(
L
as
s
o
)
an
d
L
2
(
R
id
g
e)
r
eg
u
lar
izatio
n
,
ar
e
u
s
ed
to
p
r
ev
en
t o
v
er
f
it
tin
g
b
y
ad
d
in
g
p
en
altie
s
to
t
h
e
lo
s
s
f
u
n
ct
io
n
.
Feat
u
r
e
s
e
lectio
n
f
u
r
th
er
r
ef
i
n
es
th
e
d
ata
s
et
b
y
p
r
io
r
itizin
g
s
i
g
n
if
ican
t
f
ea
t
u
r
es
w
h
ile
eli
m
i
n
atin
g
i
r
r
e
lev
an
t o
n
es.
P
C
A
:
−
C
o
v
ar
ian
ce
m
atr
i
x
:
=
1
−
1
∑
(
−
̅
)
2
)
(
−
̅
)
)
=
1
W
h
er
e
Σ
is
th
e
co
v
ar
ian
ce
m
at
r
ix
,
ar
e
th
e
d
ata
p
o
in
ts
,
an
d
is
th
e
m
ea
n
v
ec
to
r
.
−
E
ig
en
v
al
u
es
a
n
d
eig
e
n
v
ec
to
r
s
:
s
o
lv
e
f
o
r
λ
a
n
d
v
in
=
T
h
e
eig
en
v
ec
to
r
s
co
r
r
esp
o
n
d
in
g
to
t
h
e
lar
g
es
t e
ig
e
n
v
alu
e
s
a
r
e
u
s
ed
as th
e
p
r
in
cip
al
co
m
p
o
n
en
t
s
.
3
.
3
.
L
inea
r
dis
cr
i
m
ina
nt
a
n
a
ly
s
is
T
o
q
u
an
tify
th
e
d
is
p
er
s
io
n
o
f
class
-
w
i
s
e
m
ea
n
v
ec
to
r
s
r
elati
v
e
to
th
e
o
v
er
all
d
ata
m
ea
n
,
th
e
b
et
w
ee
n
-
class
s
ca
t
ter
m
atr
i
x
is
co
m
p
u
te
d
,
ca
p
tu
r
in
g
t
h
e
s
ep
ar
ab
ilit
y
a
m
o
n
g
d
if
f
er
en
t c
lass
e
s
.
I
t is d
ef
i
n
ed
as:
B
et
w
ee
n
-
cla
s
s
s
ca
tter
m
atr
i
x
(
S_
B
)
:
∑
=
1
(
−
̅
)
(
−
̅
)
)
W
h
er
e
c
is
th
e
n
u
m
b
er
o
f
clas
s
es,
is
th
e
n
u
m
b
er
o
f
s
a
m
p
le
s
in
class
i,
is
th
e
m
ea
n
v
ec
to
r
o
f
class
i,
an
d
μ
is
th
e
o
v
er
all
m
ea
n
v
ec
to
r
.
W
ith
in
-
cla
s
s
s
ca
tter
m
atr
i
x
(
S
_
W
)
:
∑
∑
=
1
(
−
̅
)
(
−
̅
)
)
W
h
er
e
ar
e
th
e
s
a
m
p
les i
n
cla
s
s
i.
R
eg
u
lar
izatio
n
tech
n
iq
u
e
s
:
−
L
1
r
eg
u
lar
izatio
n
(
L
as
s
o
)
:
a
d
d
s
a
p
en
alt
y
eq
u
al
to
th
e
ab
s
o
l
u
te
v
alu
e
o
f
th
e
m
a
g
n
itu
d
e
o
f
c
o
ef
f
icie
n
t
s
.
L
o
s
s
f
u
n
c
tio
n
:
=
∑
(
−
̿
=
1
)
2
+
∑
|
|
=
1
−
L
2
r
eg
u
lar
izatio
n
(
R
id
g
e)
:
ad
d
s
a
p
en
alt
y
eq
u
al
to
th
e
s
q
u
ar
e
o
f
th
e
m
ag
n
it
u
d
e
o
f
co
ef
f
icie
n
ts
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
9
-
4864
I
n
t J
R
ec
o
n
f
i
g
u
r
ab
le
&
E
m
b
ed
d
ed
Sy
s
t
,
Vo
l.
1
5
,
No
.
1
,
Ma
r
c
h
2
0
2
6
:
31
-
41
36
L
o
s
s
f
u
n
c
tio
n
:
=
∑
(
−
̿
=
1
)
2
+
∑
2
=
1
Featu
r
e
s
elec
tio
n
is
an
o
t
h
er
o
f
te
n
e
m
p
lo
y
ed
m
e
th
o
d
f
o
r
m
i
n
i
m
is
i
n
g
th
e
a
m
o
u
n
t
o
f
f
ea
tu
r
es
in
a
d
ataset.
I
n
co
n
tr
ast
to
f
ea
tu
r
e
e
x
tr
ac
tio
n
,
th
e
o
b
j
ec
tiv
e
o
f
f
ea
tu
r
e
s
e
lecti
o
n
is
to
p
r
io
r
itis
e
th
e
s
i
g
n
if
ica
n
ce
o
f
t
h
e
d
ataset
's
cu
r
r
en
t
f
ea
tu
r
e
s
an
d
r
e
m
o
v
e
a
n
y
t
h
at
ar
e
u
n
i
m
p
o
r
tan
t (
n
o
n
e
w
f
ea
t
u
r
es a
r
e
ad
d
ed
)
.
3
.
4
.
T
est/
t
ra
in s
pli
t
s
A
m
et
h
o
d
f
o
r
as
s
es
s
i
n
g
a
m
a
ch
in
e
lear
n
i
n
g
s
y
s
te
m
's
p
er
f
o
r
m
an
ce
is
t
h
e
tr
ain
-
tes
t
s
p
lit.
P
r
o
b
lem
s
in
v
o
l
v
i
n
g
clas
s
if
icatio
n
,
r
eg
r
e
s
s
io
n
,
o
r
an
y
o
t
h
er
s
u
p
er
v
is
ed
lear
n
in
g
tec
h
n
iq
u
e
m
a
y
b
e
r
eso
lv
ed
u
s
i
n
g
i
t.
T
h
e
d
ataset
n
ee
d
s
to
b
e
s
p
lit
in
to
t
w
o
ca
teg
o
r
ies
f
i
r
s
t.
T
h
e
tr
ain
in
g
d
atase
t
s
er
v
e
s
as
th
e
i
n
iti
al
s
u
b
s
et
f
o
r
m
o
d
el
f
itti
n
g
.
I
n
s
tead
o
f
tr
ain
i
n
g
t
h
e
m
o
d
el
o
n
t
h
e
s
ec
o
n
d
s
u
b
s
et,
t
h
e
d
ataset
's
i
n
p
u
t
co
m
p
o
n
e
n
ts
ar
e
g
i
v
en
to
it,
a
n
d
its
p
r
ed
ictio
n
s
ar
e
th
en
cr
ea
te
d
an
d
co
m
p
ar
ed
to
p
r
ed
icted
v
alu
e
s
.
T
h
e
s
ec
o
n
d
in
q
u
esti
o
n
is
th
e
te
s
t
d
ataset.
T
h
e
tr
ain
-
test
s
p
lit
tech
n
iq
u
e
is
a
r
ap
id
an
d
s
im
p
le
p
r
o
ce
d
u
r
e,
an
d
th
e
r
esu
lts
o
f
f
er
ass
e
s
s
m
en
ts
o
f
th
e
ef
f
icac
y
o
f
m
ac
h
i
n
e
lear
n
i
n
g
alg
o
r
it
h
m
s
f
o
r
ce
r
tain
p
r
ed
ictiv
e
m
o
d
ell
in
g
p
r
o
b
le
m
s
.
T
h
e
g
o
al
is
to
ass
es
s
th
e
m
ac
h
i
n
e
l
ea
r
n
in
g
m
o
d
el
's
p
er
f
o
r
m
a
n
ce
u
s
i
n
g
f
r
es
h
d
ata
th
a
t
w
er
en
'
t
u
s
ed
to
tr
ain
th
e
m
o
d
el.
I
f
tar
g
e
t
v
alu
e
s
o
r
p
r
o
j
ec
ted
o
u
tco
m
es
ar
e
ab
s
en
t,
test
/tra
i
n
f
it
o
n
e
x
is
t
in
g
d
ata
w
ith
k
n
o
w
n
in
p
u
t
s
an
d
o
u
tp
u
t
s
is
u
s
ed
to
g
en
er
ate
p
r
ed
ictio
n
s
o
n
f
r
e
s
h
e
x
a
m
p
les
in
th
e
f
u
t
u
r
e
.
W
h
en
a
lar
g
e
e
n
o
u
g
h
d
ataset
i
s
av
a
ilab
le,
th
e
tr
ain
-
te
s
t
tec
h
n
i
q
u
e
is
ap
p
r
o
p
r
iate.
Ma
ch
i
n
e
lear
n
i
n
g
m
o
d
els
f
o
r
class
i
f
icatio
n
o
r
r
eg
r
ess
io
n
m
a
y
b
e
ev
alu
ated
u
s
i
n
g
a
tr
ai
n
-
test
s
p
lit.
T
est
d
ataset
s
tan
d
s
i
n
co
n
tr
ast
to
th
e
tr
ain
d
ataset
,
w
h
ic
h
is
u
s
ed
to
as
s
e
s
s
h
o
w
w
e
ll
t
h
e
m
ac
h
i
n
e
lear
n
in
g
m
o
d
el
f
it
s
t
h
e
d
ata.
3
.
5
.
Cla
s
s
if
ica
t
io
ns
I
n
s
tatis
t
ics
an
d
m
ac
h
in
e
lear
n
i
n
g
,
clas
s
i
f
icatio
n
is
a
s
u
p
er
v
is
ed
lear
n
in
g
tech
n
iq
u
e
in
w
h
ic
h
co
m
p
u
ter
p
r
o
g
r
am
m
e
s
m
ak
e
n
e
w
o
b
s
e
r
v
atio
n
s
o
r
ca
teg
o
r
is
e
e
x
i
s
ti
n
g
d
ata
d
ep
en
d
i
n
g
o
n
w
h
at
t
h
e
y
h
a
v
e
lear
n
ed
.
I
n
co
m
i
n
g
d
ata
is
m
ap
p
ed
to
p
r
ed
ef
in
ed
ca
te
g
o
r
ies
u
s
i
n
g
cla
s
s
i
f
ier
al
g
o
r
ith
m
s
.
C
la
s
s
i
f
icat
i
o
n
m
o
d
els
u
s
e
t
h
e
tr
ain
i
n
g
d
ata
as
in
p
u
t
to
f
o
r
ec
ast
class
e
s
o
r
ca
teg
o
r
ies
o
r
to
d
r
aw
co
n
cl
u
s
io
n
s
.
B
in
ar
y
,
m
u
lt
ip
le
-
cla
s
s
,
o
r
m
u
ltip
le
-
l
ab
el
ca
teg
o
r
izatio
n
s
ca
n
all
b
e
u
s
ed
in
class
if
icat
io
n
s
.
B
in
ar
y
clas
s
if
icatio
n
s
ar
e
class
i
f
icatio
n
k
i
n
d
s
th
at
h
av
e
t
w
o
p
o
s
s
ib
le
r
esu
lt
s
(
tr
u
e
o
r
f
alse
)
.
Sa
m
p
les
ar
e
as
s
i
g
n
ed
to
tar
g
et
c
lass
e
s
i
n
m
u
lt
i
-
class
cla
s
s
i
f
icatio
n
s
if
t
h
er
e
ar
e
m
o
r
e
th
a
n
t
w
o
cla
s
s
e
s
.
I
n
m
u
lti
-
lab
el
cla
s
s
i
f
icati
o
n
s
,
m
a
n
y
lab
els
o
r
g
o
als
ar
e
ap
p
lied
to
th
e
s
a
m
e
s
a
m
p
le.
L
az
y
o
r
en
t
h
u
s
ia
s
tic
l
ea
r
n
er
s
ca
n
b
e
ch
ar
ac
ter
i
s
ed
i
n
clas
s
i
f
icatio
n
lear
n
in
g
.
L
ea
r
n
er
s
s
av
e
d
ata
w
h
ich
is
d
iv
id
ed
in
to
ca
te
g
o
r
ies
b
ase
d
o
n
th
e
m
o
s
t
i
m
p
o
r
tan
t
in
f
o
r
m
ati
o
n
.
W
h
e
n
co
m
p
ar
ed
to
e
n
t
h
u
s
ias
tic
p
u
p
il
s
,
t
h
e
y
h
av
e
m
o
r
e
ti
m
e
f
o
r
p
r
ed
ictio
n
.
as
in
k
-
n
ea
r
e
s
t
n
eig
h
b
o
u
r
s
(
K
NNs).
B
ef
o
r
e
r
ec
eiv
i
n
g
d
ata
f
o
r
p
r
ed
ictio
n
s
,
ea
g
er
lear
n
er
s
b
u
ild
a
clas
s
i
f
icatio
n
m
o
d
el
u
s
i
n
g
th
e
tr
ain
i
n
g
d
ata
t
h
at
i
s
alr
ea
d
y
a
v
ailab
le.
I
t
m
u
s
t
b
e
ab
le
to
f
o
llo
w
a
s
in
g
le
t
h
eo
r
y
th
at
ap
p
lies
to
th
e
w
h
o
le
f
ield
.
T
h
ey
s
p
en
d
l
ess
ti
m
e
m
a
k
i
n
g
p
r
ed
ictio
n
s
s
in
ce
th
e
y
p
r
ac
tis
e
a
m
u
c
h
.
E
x
a
m
p
le
s
i
n
cl
u
d
e
ar
tific
ial
n
e
u
r
al
n
et
w
o
r
k
s
(
A
NNs)
,
d
ec
is
io
n
tr
ee
s
,
a
n
d
n
ai
v
e
b
a
y
es.
B
o
o
s
tin
g
i
s
a
n
en
s
e
m
b
le
m
o
d
elli
n
g
tec
h
n
iq
u
e
d
esig
n
ed
to
tu
r
n
a
lar
g
e
n
u
m
b
er
o
f
p
o
o
r
class
if
ier
s
in
to
a
f
e
w
p
o
w
er
f
u
l
o
n
es.
W
ea
k
m
o
d
els
ar
e
u
s
ed
s
eq
u
e
n
tiall
y
to
co
n
s
tr
u
ct
m
o
d
els
i
n
o
r
d
er
to
ac
h
iev
e
th
is
.
T
h
is
w
o
r
k
u
s
es
b
o
o
tin
g
(
XGB
o
o
s
t)
f
o
r
class
if
icat
io
n
s
an
d
ev
alu
a
tio
n
s
o
f
s
elec
ted
f
ea
tu
r
es
b
y
t
h
e
p
r
o
p
o
s
ed
SMFC
C
E
s
c
h
e
m
a
.
T
h
e
s
eq
u
en
tial
tr
ee
co
n
s
tr
u
ct
io
n
m
eth
o
d
is
ap
p
r
o
ac
h
ed
b
y
XGB
o
o
s
t
u
s
in
g
a
p
ar
allelize
d
im
p
l
e
m
en
tatio
n
.
T
h
is
i
s
m
ad
e
p
o
s
s
ib
le
b
y
t
h
e
i
n
ter
ch
a
n
g
ea
b
ilit
y
o
f
th
e
t
w
o
i
n
n
er
lo
o
p
s
t
h
at
co
m
p
u
te
t
h
e
f
ea
t
u
r
es
a
n
d
th
e
o
u
ter
lo
o
p
th
at
co
u
n
t
s
th
e
lea
f
n
o
d
es
o
f
a
tr
ee
w
h
e
n
cr
ea
tin
g
b
ase
lear
n
er
s
.
T
h
is
s
tack
in
g
o
f
lo
o
p
s
p
r
ev
en
t
s
p
ar
alleliza
tio
n
s
i
n
ce
th
e
o
u
ter
lo
o
p
,
w
h
ic
h
is
th
e
m
o
r
e
co
m
p
u
ta
tio
n
all
y
e
x
p
en
s
iv
e
o
f
th
e
t
w
o
,
ca
n
n
o
t
b
e
b
eg
u
n
u
n
t
il
t
h
e
in
n
er
lo
o
p
,
w
h
ic
h
is
t
h
e
m
o
r
e
e
x
p
en
s
iv
e
o
f
th
e
t
w
o
,
h
as
b
ee
n
f
i
n
is
h
ed
.
T
h
e
lo
o
p
s
ar
e
r
ec
o
n
f
ig
u
r
ed
u
s
in
g
i
n
itializa
tio
n
,
a
g
lo
b
al
s
ca
n
o
f
all
in
s
ta
n
ce
s
,
an
d
s
o
r
tin
g
w
ith
p
ar
allel
th
r
ea
d
s
to
s
av
e
r
u
n
ti
m
e.
T
h
is
d
ec
is
io
n
i
m
p
r
o
v
e
s
alg
o
r
ith
m
ic
e
f
f
icie
n
c
y
b
y
b
al
an
cin
g
a
n
y
p
ar
alleliza
t
io
n
o
v
er
h
ea
d
s
in
co
m
p
u
ta
tio
n
.
B
elo
w
is
a
l
is
t
o
f
th
e
XGB
o
o
s
t a
lg
o
r
ith
m
.
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
is
s
ec
tio
n
p
r
esen
t
s
th
e
e
x
p
er
i
m
en
tal
f
i
n
d
in
g
s
f
r
o
m
t
h
e
p
r
o
p
o
s
ed
SMFC
C
E
s
c
h
e
m
e,
w
h
ic
h
w
a
s
i
m
p
le
m
en
ted
o
n
an
ad
v
an
ce
d
m
icr
o
d
ev
ices
(
A
M
D)
A
th
lo
n
C
P
U
w
ith
4
GB
o
f
R
A
M
u
s
i
n
g
P
y
th
o
n
3
.
9
.
T
h
e
d
ataset
u
s
ed
f
o
r
t
h
e
e
x
p
er
i
m
e
n
ts
,
k
n
o
w
n
as
th
e
ca
r
in
s
u
r
a
n
ce
co
ld
ca
lls
r
ep
o
r
t
,
w
as
o
b
tain
ed
f
r
o
m
Kag
g
le.
T
h
is
d
ataset
w
as
s
o
u
r
ce
d
f
r
o
m
a
US
b
an
k
t
h
at
o
f
f
er
s
au
to
i
n
s
u
r
an
ce
alo
n
g
s
id
e
it
s
s
ta
n
d
ar
d
s
er
v
ice
s
.
T
h
e
b
an
k
r
eg
u
lar
l
y
co
n
d
u
cts
m
ar
k
e
tin
g
ca
m
p
aig
n
s
to
attr
ac
t
n
e
w
c
u
s
t
o
m
er
s
b
y
co
n
tacti
n
g
p
o
ten
tial
clien
t
s
to
p
r
o
m
o
te
v
ar
io
u
s
v
e
h
icle
i
n
s
u
r
an
ce
o
p
tio
n
s
.
T
h
e
d
ataset
i
n
clu
d
e
s
g
e
n
e
r
al
clien
t
i
n
f
o
r
m
atio
n
(
s
u
ch
a
s
ag
e
an
d
e
m
p
lo
y
m
e
n
t
s
tatu
s
)
an
d
s
p
ec
i
f
ic
d
etail
s
ab
o
u
t
o
n
g
o
in
g
an
d
p
ast
i
n
s
u
r
a
n
c
e
s
ales
ca
m
p
a
i
g
n
s
(
s
u
c
h
as
co
m
m
u
n
icatio
n
t
y
p
e,
last
co
n
tact
d
ate,
an
d
th
e
n
u
m
b
er
o
f
p
r
ev
io
u
s
atte
m
p
t
s
)
.
Data
s
et
o
v
er
v
ie
w
: t
h
e
d
ataset
's f
ea
t
u
r
es a
r
e
as f
o
llo
w
s
:
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
R
ec
o
n
f
i
g
u
r
ab
le
&
E
m
b
ed
d
ed
Sy
s
t
I
SS
N:
2089
-
4864
Op
timiz
in
g
ca
ll c
en
ter a
g
en
t e
fficien
cy
th
r
o
u
g
h
d
ee
p
lea
r
n
in
g
-
b
a
s
ed
…
(
R
a
ma
c
h
a
n
d
r
a
n
P
eriya
s
a
my
)
37
−
I
d
:
u
n
iq
u
e
I
D
n
u
m
b
er
.
−
Ag
e:
a
g
e
o
f
th
e
cl
ien
t.
−
J
o
b
:
clien
t'
s
j
o
b
.
−
Ma
r
ital:
m
ar
ital
s
tat
u
s
o
f
t
h
e
c
lien
t.
−
E
d
u
ca
tio
n
:
clie
n
t
'
s
ed
u
ca
tio
n
l
ev
el.
−
Def
a
u
lt:
w
h
et
h
er
th
e
clie
n
t is
a
d
ef
au
lter
.
−
B
alan
ce
:
a
v
er
ag
e
y
ea
r
l
y
b
ala
n
ce
.
−
HHI
n
s
u
r
an
ce
:
w
h
et
h
er
th
e
h
o
u
s
e
h
o
ld
is
in
s
u
r
ed
.
−
C
ar
L
o
a
n
:
w
h
e
th
er
t
h
e
clien
t h
as a
ca
r
lo
an
.
−
C
o
m
m
u
n
ica
tio
n
:
t
y
p
e
o
f
co
m
m
u
n
icat
io
n
.
−
L
ast
C
o
n
tactM
o
n
t
h
:
m
o
n
t
h
o
f
t
h
e
last
co
n
tact.
−
L
ast
C
o
n
tactDa
y
:
d
a
y
o
f
t
h
e
la
s
t c
o
n
tact.
−
C
allStar
t:
s
tar
t ti
m
e
o
f
t
h
e
last
ca
ll (
HH:M
M)
.
−
C
allE
n
d
:
e
n
d
ti
m
e
o
f
th
e
la
s
t c
all
(
HH:M
M)
.
−
No
Of
C
o
n
tacts
:
n
u
m
b
er
o
f
co
n
tacts d
u
r
in
g
t
h
is
ca
m
p
ai
g
n
f
o
r
th
is
cl
ien
t.
−
Da
y
s
P
ass
ed
:
n
u
m
b
er
o
f
d
a
y
s
s
in
ce
th
e
la
s
t c
o
n
tact
f
r
o
m
a
p
r
ev
io
u
s
ca
m
p
ai
g
n
.
−
P
r
ev
A
tte
m
p
ts
:
n
u
m
b
er
o
f
co
n
t
ac
ts
p
er
f
o
r
m
ed
b
ef
o
r
e
th
is
ca
m
p
a
ig
n
.
−
Ou
tco
m
e
:
o
u
tco
m
e
o
f
th
e
p
r
ev
io
u
s
m
ar
k
eti
n
g
ca
m
p
ai
g
n
.
−
C
ar
I
n
s
u
r
an
ce
:
w
h
et
h
er
th
e
clie
n
t s
u
b
s
cr
ib
ed
to
ca
r
in
s
u
r
an
ce
.
4
.
1
.
Select
ing
m
i
ni
m
a
l f
ea
t
u
re
s
f
o
r
ca
ll c
ent
er
a
g
ent
s
ef
f
iciency
s
che
m
a
’
s
e
x
plo
ra
t
o
ry
da
t
a
a
na
ly
s
is
E
DA
p
la
y
s
a
cr
u
cial
r
o
le
in
u
n
co
v
er
i
n
g
n
e
w
i
n
s
ig
h
ts
an
d
d
ev
elo
p
in
g
a
d
ee
p
er
u
n
d
er
s
ta
n
d
in
g
o
f
th
e
d
ata.
T
h
e
in
itial
s
tep
in
v
o
lv
ed
ex
a
m
in
i
n
g
th
e
s
h
ap
e
o
f
t
h
e
d
at
a
s
et
an
d
its
co
l
u
m
n
s
,
d
atat
y
p
es
,
an
d
b
asic
s
tatis
t
ics.
T
h
e
n
u
m
er
ical
co
lu
m
n
s
—
Def
a
u
lt
,
HHI
n
s
u
r
a
n
ce
,
C
a
r
Lo
a
n
,
an
d
C
a
r
I
n
s
u
r
a
n
ce
—
co
n
tai
n
ed
b
in
ar
y
v
al
u
es
(
0
s
an
d
1
s
)
.
C
ateg
o
r
ical
f
ea
t
u
r
es
w
er
e
also
as
s
es
s
ed
.
SMFC
C
E
d
ata
clea
n
i
n
g
:
h
a
n
d
l
in
g
m
is
s
in
g
v
al
u
es
i
s
a
s
i
g
n
i
f
ic
an
t
c
h
alle
n
g
e
i
n
d
ata
a
n
al
y
s
i
s
,
as
m
is
s
in
g
d
ata
ca
n
i
m
p
ed
e
ca
lcu
latio
n
s
a
n
d
v
i
s
u
al
izatio
n
s
.
I
n
o
u
r
d
atase
t,
th
e
r
e
s
u
lt
an
d
co
mmu
n
ica
tio
n
f
ield
s
w
er
e
p
r
o
n
e
to
m
i
s
s
i
n
g
v
al
u
es,
w
it
h
n
u
m
er
o
u
s
m
i
s
s
i
n
g
en
tr
ies
i
n
th
e
jo
b
an
d
ed
u
ca
tio
n
ca
teg
o
r
ies.
T
h
e
d
at
a
clea
n
in
g
p
r
o
ce
s
s
in
v
o
l
v
ed
i
m
p
u
ti
n
g
m
is
s
i
n
g
v
a
lu
es
f
o
r
t
h
e
jo
b
an
d
ed
u
ca
tio
n
f
ield
s
u
s
in
g
P
y
t
h
o
n
’
s
b
ac
k
f
ill
/f
r
o
n
t
f
ill
m
e
th
o
d
s
.
Fo
r
f
ield
s
w
it
h
e
x
ten
s
i
v
e
m
is
s
in
g
d
ata,
s
u
ch
as
r
esu
lt
a
n
d
c
o
mmu
n
ica
tio
n
,
th
e
m
is
s
i
n
g
v
a
lu
es
w
er
e
r
ep
lace
d
w
it
h
"
No
n
e.
"
SMFC
C
E
f
ea
t
u
r
e
ex
tr
ac
tio
n
s
:
f
ea
t
u
r
e
en
g
i
n
ee
r
i
n
g
is
v
ital
f
o
r
en
h
a
n
cin
g
th
e
p
er
f
o
r
m
a
n
ce
o
f
m
ac
h
i
n
e
lear
n
in
g
al
g
o
r
ith
m
s
.
C
o
n
ti
n
u
o
u
s
v
ar
iab
les
li
k
e
A
g
e
an
d
B
a
la
n
ce
w
er
e
b
in
n
ed
in
to
f
iv
e
b
u
ck
ets
u
s
in
g
t
h
e
q
u
ar
til
e
cu
t
m
et
h
o
d
.
T
h
e
C
a
llS
ta
r
t
a
n
d
C
a
llEn
d
ti
m
es,
i
n
itia
ll
y
s
to
r
ed
as
o
b
j
ec
t
v
ar
iab
les,
w
er
e
co
n
v
er
ted
in
to
d
ateti
m
e
f
o
r
m
at
to
ca
lcu
late
th
e
ac
tu
a
l
C
a
llT
ime
.
T
h
is
ca
lcu
lated
C
a
l
lTi
me
w
a
s
th
e
n
b
in
n
ed
,
an
d
th
e
o
r
i
g
in
al
co
lu
m
n
s
w
er
e
r
e
m
o
v
ed
.
C
ateg
o
r
ical
v
a
r
iab
les
w
er
e
en
co
d
ed
as
d
u
m
m
y
v
ar
iab
les
to
b
e
in
clu
d
ed
in
th
e
m
o
d
el
-
b
u
ild
i
n
g
p
r
o
ce
s
s
,
an
d
co
r
r
elatio
n
s
b
et
w
ee
n
v
ar
iab
les
w
er
e
e
x
a
m
i
n
e
d
u
s
i
n
g
a
h
ea
t
m
ap
.
No
tab
l
y
,
a
p
o
s
itiv
e
co
r
r
elatio
n
w
a
s
o
b
s
er
v
ed
b
et
w
ee
n
Da
ysP
a
s
s
e
d
an
d
P
r
ev
A
ttemp
ts
.
Fi
g
u
r
e
3
illu
s
tr
ates t
h
e
co
r
r
elatio
n
s
b
et
w
ee
n
v
ar
iab
les.
Fig
u
r
e
3
.
C
o
r
r
elatio
n
s
b
et
w
ee
n
v
ar
iab
les
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
9
-
4864
I
n
t J
R
ec
o
n
f
i
g
u
r
ab
le
&
E
m
b
ed
d
ed
Sy
s
t
,
Vo
l.
1
5
,
No
.
1
,
Ma
r
c
h
2
0
2
6
:
31
-
41
38
SMFC
C
E
test
/tra
i
n
s
p
lits
:
to
en
s
u
r
e
s
u
f
f
icien
t
d
ata
f
o
r
m
o
d
el
tr
ain
in
g
an
d
ev
al
u
atio
n
,
th
e
tr
ain
-
tes
t
s
p
lit
m
et
h
o
d
w
as
e
m
p
lo
y
ed
.
T
h
e
k
-
f
o
ld
cr
o
s
s
-
v
alid
atio
n
tech
n
iq
u
e
w
as
also
u
s
ed
as
a
m
o
d
el
ass
e
s
s
m
e
n
t
alter
n
ati
v
e,
p
ar
ticu
lar
l
y
b
en
e
f
i
cial
w
h
en
d
ata
is
li
m
ited
o
r
w
h
en
m
o
d
els
ar
e
co
m
p
u
tatio
n
all
y
ex
p
en
s
i
v
e
to
tr
ain
.
T
h
e
tr
ain
-
tes
t sp
lit
w
as c
o
n
f
ig
u
r
ed
w
it
h
s
tan
d
ar
d
s
izes to
ev
alu
ate
m
o
d
e
l p
er
f
o
r
m
a
n
ce
ac
c
u
r
atel
y
.
4
.
2
.
Cla
s
s
if
ier
ev
a
lua
t
io
ns
On
ce
a
clas
s
i
f
ier
is
d
e
v
elo
p
ed
,
ev
alu
a
tin
g
it
s
ac
c
u
r
ac
y
a
n
d
ef
f
icien
c
y
is
cr
itica
l.
Var
io
u
s
m
et
h
o
d
s
w
er
e
e
m
p
lo
y
ed
to
ass
ess
t
h
e
class
i
f
ier
s
,
in
cl
u
d
i
n
g
ac
c
u
r
ac
y
s
co
r
es,
cr
o
s
s
-
v
alid
atio
n
s
co
r
es,
cl
ass
i
f
icatio
n
r
ep
o
r
ts
(
p
r
ec
is
io
n
,
r
ec
all
,
F1
-
s
co
r
e,
s
u
p
p
o
r
t)
,
R
OC
cu
r
v
e
s
,
a
n
d
co
n
f
u
s
io
n
m
atr
ices.
T
h
e
m
o
d
el
s
wer
e
tr
ain
ed
u
s
in
g
th
e
A
d
aB
o
o
s
t
(
w
it
h
p
ar
a
m
eter
s
n
_
esti
m
ato
r
s
=4
0
0
,
lear
n
in
g
_
r
ate=
0
.
1
)
an
d
XGBo
o
s
t
(
w
it
h
p
ar
am
e
ter
s
n
_
esti
m
ato
r
s
=1
0
0
0
,
lear
n
in
g
_
r
ate=
0
.
0
1
)
class
if
ier
s
.
T
h
e
r
esu
lts
s
h
o
w
ed
t
h
e
clas
s
i
f
icatio
n
ac
cu
r
ac
y
ac
h
iev
e
d
b
y
t
h
e
clas
s
i
f
ier
s
o
n
t
h
e
o
u
tp
u
t
s
o
f
th
e
p
r
o
p
o
s
ed
s
ch
e
m
e,
a
s
d
ep
icted
in
Fig
u
r
e
4
.
Fig
u
r
e
4
. O
b
tain
ed
class
i
f
icati
o
n
s
ac
cu
r
ac
ie
s
b
y
clas
s
if
ier
s
o
n
th
e
o
u
tp
u
t
s
o
f
th
e
p
r
o
p
o
s
ed
s
ch
e
m
e
T
h
e
tr
u
e
p
o
s
itiv
e
r
ate
an
d
tr
u
e
n
eg
ati
v
e
r
ate
w
er
e
also
ca
lcu
lated
,
w
ith
F1
-
s
co
r
es
r
ep
r
esen
ti
n
g
t
h
e
w
ei
g
h
ted
av
er
ag
e
s
o
f
p
r
ec
is
io
n
an
d
r
ec
all.
T
h
e
co
n
f
u
s
io
n
m
a
tr
ix
,
s
h
o
w
n
i
n
Fi
g
u
r
e
5
,
p
r
o
v
id
ed
a
co
m
p
r
eh
en
s
i
v
e
p
er
f
o
r
m
a
n
ce
s
tati
s
tic
f
o
r
th
e
m
ac
h
in
e
lear
n
i
n
g
clas
s
i
f
icatio
n
s
ce
n
ar
io
s
.
T
h
e
a
m
o
u
n
t
o
f
ac
cu
r
ate
f
o
r
ec
ast
s
t
h
at
th
e
o
cc
u
r
r
en
ce
is
p
o
s
itiv
e
is
k
n
o
w
n
as
th
e
tr
u
e
p
o
s
itiv
e
r
ate.
T
h
e
r
ea
l
n
eg
ati
v
e
is
d
ef
in
ed
as
th
e
o
u
tco
m
es
th
a
t
w
er
e
ac
c
u
r
atel
y
ex
p
ec
ted
to
b
e
n
eg
a
tiv
e.
F1
-
s
co
r
es
ar
e
w
ei
g
h
t
ed
av
er
a
g
es
o
f
p
r
ec
is
io
n
an
d
r
ec
all,
w
h
er
e
r
ec
all
(
T
P
/(
T
P
+FN)
)
is
th
e
p
er
ce
n
tag
e
o
f
p
er
tin
e
n
t
in
s
ta
n
ce
s
t
h
at
h
av
e
b
ee
n
f
o
u
n
d
a
m
o
n
g
all
o
f
th
e
in
s
tan
ce
s
,
an
d
p
r
ec
is
io
n
(
T
P
/(
T
P
+FP
)
)
is
th
e
p
r
o
p
o
r
tio
n
o
f
p
er
tin
en
t in
s
ta
n
c
es f
o
u
n
d
a
m
o
n
g
th
e
r
etr
ie
v
ed
in
s
ta
n
ce
s
.
T
h
e
y
ar
e
ess
e
n
tiall
y
u
s
ed
as
a
r
elev
an
ce
ass
es
s
m
en
t,
w
it
h
t
h
e
in
te
n
d
ed
r
esu
lt
b
ein
g
n
e
g
ati
v
e
b
u
t
b
ein
g
f
alse.
Fo
r
m
ac
h
in
e
lear
n
in
g
clas
s
i
f
icatio
n
s
ce
n
ar
i
o
s
w
h
er
e
th
e
o
u
tp
u
t
m
a
y
b
e
t
w
o
o
r
m
o
r
e
class
es,
t
h
e
co
n
f
u
s
io
n
m
atr
i
x
is
a
p
er
f
o
r
m
a
n
ce
s
ta
tis
t
ic.
W
ith
o
u
t
d
ata
v
i
s
u
a
lis
atio
n
,
it
w
o
u
ld
b
e
d
if
f
ic
u
lt
to
r
ea
d
il
y
ar
r
iv
e
at
a
co
n
clu
s
io
n
i
n
d
ata
s
ci
en
ce
.
E
v
e
n
if
th
e
o
u
tco
m
e
is
estab
li
s
h
ed
b
y
tab
le
s
,
it
m
i
g
h
t b
e
d
if
f
ic
u
lt t
o
an
al
y
s
e
ea
ch
f
i
g
u
r
e
an
d
d
r
aw
co
n
cl
u
s
io
n
s
.
E
v
e
n
a
n
o
n
-
tec
h
n
ica
l
in
d
i
v
id
u
a
l
m
a
y
co
m
p
lete
t
h
o
s
e
d
u
ties
w
i
th
ea
s
e
b
y
u
s
i
n
g
c
h
ar
ts
an
d
g
r
ap
h
s
.
E
x
ec
u
t
iv
e
s
an
d
m
an
a
g
er
s
e
n
j
o
y
lo
o
k
i
n
g
at
r
ep
o
r
ts
th
at
h
av
e
v
is
u
ali
s
atio
n
s
b
ec
au
s
e
it
m
a
k
es
it
ea
s
ier
f
o
r
th
e
m
to
m
a
k
e
co
m
p
lica
ted
ch
o
ices
[
2
5
]
,
[
2
6
]
.
T
h
e
p
air
p
lo
t
th
at
p
air
s
an
d
p
l
o
ts
f
ield
s
o
f
i
n
ter
est
i
s
s
h
o
w
n
b
elo
w
.
T
h
e
h
ea
t
m
ap
is
u
s
ed
to
ch
o
o
s
e
th
e
f
ac
to
r
s
f
o
r
th
e
P
ai
r
p
lo
t
th
at
in
f
lu
e
n
ce
t
h
e
r
esu
lt.
T
h
e
d
im
e
n
s
io
n
alit
y
r
ed
u
ctio
n
tech
n
iq
u
e
o
f
f
ea
tu
r
e
ex
tr
ac
tio
n
r
ed
u
ce
s
i
n
itial set
s
o
f
r
a
w
d
ata
to
lev
els
s
u
itab
le
f
o
r
p
r
o
ce
s
s
in
g
[
2
7
]
.
T
h
e
lar
g
e
n
u
m
b
er
o
f
v
ar
iab
les
in
th
e
s
e
en
o
r
m
o
u
s
d
ata
s
ets
m
ak
e
s
th
e
m
d
i
f
f
icu
lt
to
p
r
o
ce
s
s
co
m
p
u
tatio
n
all
y
.
"Fea
tu
r
e
ex
tr
ac
tio
n
"
r
ef
er
s
to
m
et
h
o
d
s
f
o
r
s
elec
tin
g
an
d
/o
r
co
m
b
in
i
n
g
v
ar
iab
les
in
to
f
ea
t
u
r
es,
w
h
ic
h
g
r
ea
tl
y
d
ec
r
ea
s
es
th
e
a
m
o
u
n
t
o
f
d
ata
th
at
m
u
s
t
b
e
p
r
o
ce
s
s
ed
w
h
ile
ac
cu
r
ate
l
y
a
n
d
co
m
p
lete
l
y
d
escr
ib
in
g
in
it
ial
d
ata
s
et.
W
h
en
le
s
s
p
r
o
ce
s
s
in
g
p
o
w
er
is
r
eq
u
ir
ed
w
it
h
o
u
t
lo
s
in
g
cr
u
cial
o
r
p
er
tin
e
n
t
d
ata,
th
e
f
ea
t
u
r
e
ex
tr
ac
tio
n
s
tr
ate
g
y
is
ad
v
a
n
tag
eo
u
s
.
R
ed
u
ce
d
d
u
p
licate
d
ata
m
i
g
h
t
h
elp
an
a
n
al
y
s
i
s
b
y
u
s
i
n
g
f
ea
t
u
r
e
ex
tr
ac
tio
n
.
T
h
e
d
ata
r
ed
u
ctio
n
an
d
co
m
p
u
ter
-
g
en
er
ated
atte
m
p
t
s
to
co
m
b
i
n
e
v
ar
iab
les
in
to
f
ea
t
u
r
es
s
p
ee
d
u
p
th
e
lear
n
in
g
an
d
g
e
n
er
ali
s
atio
n
s
ta
g
es
o
f
t
h
e
m
ac
h
in
e
lear
n
i
n
g
p
r
o
ce
s
s
.
T
h
e
m
o
s
t
s
i
g
n
i
f
ican
t
ch
ar
ac
ter
is
t
ics
id
en
ti
f
ied
in
t
h
i
s
in
v
es
tig
a
t
io
n
w
er
e
ca
ll
ti
m
e
s
,
las
t
co
n
tacte
d
d
ay
,
b
alan
ce
s
p
ay
ab
le,
co
n
tacts
co
u
n
t,
s
u
cc
es
s
es
o
f
o
u
tco
m
e
s
,
ag
es,
in
s
u
r
a
n
ce
s
,
n
o
n
-
co
m
m
u
n
icatio
n
s
,
d
a
y
s
p
as
s
ed
af
ter
ca
ll,
a
n
d
n
o
n
et
o
u
tco
m
e
s
.
Fi
g
u
r
e
6
ill
u
s
tr
ates
t
h
e
c
u
m
u
lati
v
e
g
ai
n
c
h
ar
t a
s
ap
p
lied
to
th
e
SMFC
C
AE
s
ch
e
m
a,
s
h
o
w
ca
s
i
n
g
t
h
e
g
a
i
n
s
ac
h
iev
ed
b
y
le
v
er
ag
in
g
t
h
is
p
r
ed
ictiv
e
m
o
d
el.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
R
ec
o
n
f
i
g
u
r
ab
le
&
E
m
b
ed
d
ed
Sy
s
t
I
SS
N:
2089
-
4864
Op
timiz
in
g
ca
ll c
en
ter a
g
en
t e
fficien
cy
th
r
o
u
g
h
d
ee
p
lea
r
n
in
g
-
b
a
s
ed
…
(
R
a
ma
c
h
a
n
d
r
a
n
P
eriya
s
a
my
)
39
Fig
u
r
e
5
.
C
o
n
f
u
s
io
n
m
atr
ix
o
f
th
e
SMF
C
C
A
E
s
ch
e
m
a
Fig
u
r
e
6
.
C
h
ar
t o
f
c
u
m
u
lat
iv
e
g
ain
s
o
b
tain
ed
b
y
SMFC
C
A
E
s
c
h
e
m
a
T
h
e
f
o
llo
w
in
g
s
tep
s
ca
n
b
e
tak
en
to
im
p
r
o
v
e
th
e
s
k
ill
s
o
f
C
S
R
s
:
i)
p
r
o
v
id
e
ca
ll
ce
n
tr
e
em
p
lo
y
ee
s
w
ith
p
eo
p
le
s
k
ills
tr
ain
i
n
g
s
o
th
a
t
th
e
y
ca
n
b
e
f
r
ien
d
lier
an
d
m
o
r
e
en
g
a
g
i
n
g
w
it
h
cu
s
to
m
er
s
d
u
r
in
g
ca
lls
;
ii)
k
ee
p
a
tr
ac
k
er
th
at
r
e
m
in
d
s
o
f
f
o
llo
w
-
u
p
s
s
o
th
a
t
t
h
e
r
ep
r
esen
tat
iv
e
ca
n
s
p
ea
k
w
it
h
th
e
cu
s
to
m
er
ag
ai
n
an
d
tr
y
to
p
er
s
u
ad
e
th
e
m
to
b
u
y
ca
r
in
s
u
r
an
ce
; iii)
c
h
o
o
s
e
c
u
s
to
m
er
s
w
i
th
g
o
o
d
cr
ed
it sco
r
es a
n
d
ac
co
u
n
t b
ala
n
ce
s
s
o
th
at
t
h
e
ti
m
e
in
v
ested
i
n
th
e
m
is
u
s
e
f
u
l; a
n
d
i
v
)
f
o
cu
s
o
n
o
ld
er
p
eo
p
le
w
it
h
a
h
i
g
h
e
r
i
n
co
m
e.
5.
CO
NCLU
SI
O
N
T
h
is
s
t
u
d
y
s
ig
n
i
f
ica
n
tl
y
ad
v
a
n
ce
s
o
u
r
u
n
d
er
s
ta
n
d
in
g
o
f
e
v
al
u
ati
n
g
C
S
R
s
in
ca
ll
ce
n
ter
s
t
h
r
o
u
g
h
th
e
ap
p
licatio
n
o
f
DL
T
s
.
B
y
in
tr
o
d
u
cin
g
t
h
e
SMFC
C
E
f
r
a
m
e
w
o
r
k
,
w
e
h
av
e
d
e
m
o
n
s
tr
ated
th
at
it
is
p
o
s
s
ib
le
to
ac
h
iev
e
a
n
i
m
p
r
ess
i
v
e
clas
s
i
f
ic
atio
n
ac
cu
r
ac
y
o
f
8
5
%
u
s
in
g
a
m
i
n
i
m
al
s
e
t
o
f
f
ea
tu
r
e
s
,
h
i
g
h
l
ig
h
ti
n
g
th
e
ef
f
icien
c
y
o
f
DL
T
s
in
i
m
p
r
o
v
i
n
g
ca
ll
ce
n
ter
p
er
f
o
r
m
an
ce
.
T
h
e
f
i
n
d
in
g
s
in
d
icate
t
h
at
th
e
S
MFC
C
E
f
r
a
m
e
w
o
r
k
en
h
a
n
ce
s
ass
e
s
s
m
e
n
t
ac
c
u
r
ac
y
an
d
o
p
er
atio
n
al
ef
f
icie
n
c
y
,
p
r
o
v
id
in
g
ac
tio
n
ab
le
i
n
s
ig
h
t
s
i
n
t
o
C
SR
p
r
o
d
u
ctiv
it
y
.
T
h
is
ap
p
r
o
ac
h
o
f
f
er
s
p
r
ac
tica
l
b
en
ef
its
f
o
r
b
u
s
in
e
s
s
e
s
,
lead
in
g
to
b
etter
cu
s
to
m
er
s
atis
f
a
ctio
n
an
d
r
ete
n
tio
n
w
h
ile
ad
d
r
ess
i
n
g
cr
it
ical
h
u
m
an
r
eso
u
r
ce
ch
a
llen
g
es
s
u
c
h
as
s
tr
es
s
an
d
tu
r
n
o
v
er
.
T
h
e
s
tu
d
y
's
i
m
p
lica
tio
n
s
ex
ten
d
to
th
e
r
esear
c
h
f
ield
b
y
s
h
o
w
ca
s
i
n
g
th
e
p
o
ten
tia
l o
f
ad
v
an
ce
d
an
al
y
tic
s
in
o
p
er
atio
n
al
m
an
a
g
e
m
e
n
t a
n
d
s
u
g
g
e
s
ti
n
g
f
u
r
th
er
e
x
p
lo
r
atio
n
o
f
its
ap
p
licatio
n
s
in
v
ar
io
u
s
in
d
u
s
tr
ies.
Fo
r
th
e
co
m
m
u
n
it
y
,
i
m
p
le
m
e
n
ti
n
g
th
e
SMFC
C
E
f
r
a
m
e
w
o
r
k
ca
n
lea
d
to
s
ig
n
if
ica
n
t
i
m
p
r
o
v
e
m
e
n
t
s
in
s
er
v
ice
q
u
ali
t
y
an
d
w
o
r
k
f
o
r
ce
m
a
n
a
g
e
m
en
t.
Fu
t
u
r
e
r
esear
ch
co
u
ld
e
x
p
lo
r
e
b
r
o
ad
e
r
ap
p
licatio
n
s
o
f
th
e
f
r
a
m
e
w
o
r
k
,
i
n
teg
r
at
io
n
w
it
h
e
m
e
r
g
in
g
tec
h
n
o
lo
g
ies,
an
d
lo
n
g
-
ter
m
ef
f
ec
t
s
o
n
C
SR
p
er
f
o
r
m
a
n
ce
an
d
ca
ll
ce
n
ter
o
p
er
atio
n
s
.
Ov
er
all,
th
i
s
r
esea
r
ch
u
n
d
er
s
co
r
es
th
e
tr
an
s
f
o
r
m
ati
v
e
p
o
ten
tial
o
f
DL
T
s
an
d
e
m
p
h
a
s
izes
t
h
e
i
m
p
o
r
tan
ce
o
f
ad
d
r
ess
in
g
h
u
m
an
r
eso
u
r
ce
f
ac
to
r
s
to
en
h
a
n
ce
ca
ll c
e
n
ter
o
p
er
atio
n
s
.
F
UNDIN
G
I
NF
O
RM
AT
I
O
N
Au
t
h
o
r
s
s
tate
n
o
f
u
n
d
i
n
g
i
n
v
o
l
v
ed
.
AUTHO
R
CO
NT
RIB
UT
I
O
NS ST
A
T
E
M
E
NT
T
h
is
jo
u
r
n
al
u
s
e
s
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT
)
t
o
r
ec
o
g
n
ize
in
d
i
v
id
u
al
au
t
h
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
t
h
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
lla
b
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
R
a
m
ac
h
a
n
d
r
an
P
er
iy
asa
m
y
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
Ma
n
i
k
an
d
a
n
Go
v
in
d
ar
aj
i
✓
✓
✓
✓
✓
✓
✓
I
.
Nasu
r
u
lla
✓
✓
✓
✓
✓
✓
✓
✓
V.
Srin
i
v
asa
n
✓
✓
✓
✓
✓
✓
✓
✓
K.
R
a
m
a
De
v
i
✓
✓
✓
✓
✓
✓
✓
✓
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
9
-
4864
I
n
t J
R
ec
o
n
f
i
g
u
r
ab
le
&
E
m
b
ed
d
ed
Sy
s
t
,
Vo
l.
1
5
,
No
.
1
,
Ma
r
c
h
2
0
2
6
:
31
-
41
40
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
si
s
I
:
I
n
v
e
st
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
si
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
t
h
o
r
s
s
tate
n
o
co
n
f
lic
t o
f
i
n
t
er
est.
DATA AV
AI
L
AB
I
L
I
T
Y
Data
av
ailab
ilit
y
i
s
n
o
t a
p
p
licab
le
to
th
is
p
ap
er
as n
o
n
e
w
d
ata
w
er
e
cr
ea
ted
o
r
an
al
y
ze
d
in
th
is
s
t
u
d
y
.
RE
F
E
R
E
NC
E
S
[
1
]
B
.
D
o
n
a
l
d
so
n
a
n
d
T
.
O
’
T
o
o
l
e
,
S
t
ra
t
e
g
i
c
M
a
rk
e
t
Re
l
a
t
i
o
n
s
h
i
p
s:
Fro
m
S
t
r
a
t
e
g
y
t
o
I
m
p
l
e
m
e
n
t
a
t
i
o
n
.
S
h
a
n
n
o
n
,
E
n
g
l
a
n
d
:
Jo
h
n
W
i
l
e
y
&
S
o
n
s
L
t
d
.
,
2
0
0
7
.
[
2
]
K
.
D
.
H
o
f
f
man
a
n
d
J
.
E
.
G
.
B
a
t
e
so
n
,
S
e
r
v
i
c
e
s
Ma
rke
t
i
n
g
:
C
o
n
c
e
p
t
s,
S
t
r
a
t
e
g
i
e
s
&
C
a
ses
,
3
r
d
e
d
.
U
S
A
:
T
h
o
mso
n
S
o
u
t
h
-
W
e
st
e
r
n
,
2
0
0
6
.
[
3
]
D
.
T
h
e
r
o
n
,
C
.
B
o
t
h
m
a
,
a
n
d
M
.
D
.
T
o
i
t
,
F
o
c
u
s
i
n
g
o
n
C
u
s
t
o
m
e
r
S
e
r
v
i
c
e
i
n
S
o
u
t
h
e
r
n
A
f
ri
c
a
.
P
r
e
t
o
r
i
a
:
P
r
i
n
c
e
P
u
b
l
i
sh
e
r
s,
2
0
0
3
.
[
4
]
B
a
l
t
o
,
C
a
l
l
C
e
n
t
e
r
C
o
a
c
h
i
n
g
:
8
T
o
p
T
i
p
s
a
n
d
Be
s
t
Pra
c
t
i
c
e
s
,
B
a
l
t
o
,
2
0
2
5
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s:
/
/
w
w
w
.
b
a
l
t
o
.
a
i
/
b
l
o
g
/
c
a
l
l
-
c
e
n
t
e
r
-
c
o
a
c
h
i
n
g
-
b
e
st
-
p
r
a
c
t
i
c
e
s/
,
(
A
c
c
e
sse
d
:
Jan
.
2
6
,
2
0
2
5
].
[
5
]
Z
e
n
d
e
sk
,
5
T
i
p
s
f
o
r
T
ra
i
n
i
n
g
C
a
l
l
C
e
n
t
e
r
A
g
e
n
t
s
,
Z
e
n
d
e
sk
,
2
0
2
6
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s:
/
/
w
w
w
.
z
e
n
d
e
sk
.
c
o
m/
b
l
o
g
/
5
-
t
i
p
s
-
t
r
a
i
n
i
n
g
-
c
a
l
l
-
c
e
n
t
e
r
-
a
g
e
n
t
s/
,
(
A
c
c
e
ssed
:
Ja
n
.
2
6
,
2
0
2
5
)
.
[
6
]
S
Q
M
G
r
o
u
p
,
W
h
a
t
S
k
i
l
l
s
T
o
d
a
y
’
s
C
a
l
l
C
e
n
t
e
r
A
g
e
n
t
s
N
e
e
d
t
o
S
u
c
c
e
e
d
—
a
n
d
H
o
w
t
o
T
r
a
i
n
T
h
e
m
,
S
Q
M
G
r
o
u
p
,
2
0
2
5
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s:
/
/
w
w
w
.
sq
mg
r
o
u
p
.
c
o
m/
r
e
so
u
r
c
e
s/
l
i
b
r
a
r
y
/
b
l
o
g
/
w
h
a
t
-
sk
i
l
l
s
-
t
o
d
a
y
s
-
c
a
l
l
-
c
e
n
t
e
r
-
a
g
e
n
t
s
-
n
e
e
d
-
to
-
su
c
c
e
e
d
-
a
n
d
-
h
o
w
-
to
-
t
r
a
i
n
-
t
h
e
m/
,
(
A
c
c
e
sse
d
:
Jan
.
2
6
,
2
0
2
5
)
.
[
7
]
P
.
R
.
T
i
mm
,
C
u
st
o
m
e
r
S
e
rv
i
c
e
:
C
a
re
e
r S
u
c
c
e
ss T
h
r
o
u
g
h
C
u
s
t
o
m
e
r L
o
y
a
l
t
y
,
4
t
h
e
d
.
U
S
A
:
P
e
a
r
so
n
P
r
e
n
t
i
c
e
H
a
l
l
,
2
0
0
8
.
[
8
]
S
mart
R
o
l
e
,
C
a
l
l
C
e
n
t
e
r
T
ra
i
n
i
n
g
G
u
i
d
e
(
2
0
2
5
)
:
AI
-
Po
w
e
re
d
&
S
c
a
l
a
b
l
e
,
S
mar
t
R
o
l
e
,
2
0
2
5
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s:
/
/
w
w
w
.
smar
t
r
o
l
e
.
a
i
/
b
l
o
g
/
t
h
e
-
c
o
mp
l
e
t
e
-
g
u
i
d
e
-
to
-
mo
d
e
r
n
-
c
a
l
l
-
c
e
n
t
e
r
-
t
r
a
i
n
i
n
g
-
in
-
2
0
2
5
,
(
A
c
c
e
sse
d
:
Jan
.
2
6
,
2
0
2
6
)
.
[
9
]
K
.
B
a
i
l
e
y
a
n
d
K
.
L
e
l
a
n
d
,
C
u
s
t
o
m
e
r
S
e
rvi
c
e
i
n
a
n
I
n
s
t
a
n
t
:
6
0
W
a
y
s
t
o
Wi
n
C
u
st
o
m
e
rs
a
n
d
K
e
e
p
T
h
e
m
C
o
m
i
n
g
B
a
c
k
.
F
r
a
n
k
l
i
n
L
a
k
e
s:
C
a
r
e
e
r
P
r
e
ss
I
n
c
.
,
2
0
0
8
.
[
1
0
]
D
.
C
h
i
c
u
,
M
.
d
e
l
M
a
r
P
à
mi
e
s,
G
.
R
y
a
n
,
a
n
d
C
.
C
r
o
ss,
“
Ex
p
l
o
r
i
n
g
t
h
e
i
n
f
l
u
e
n
c
e
o
f
t
h
e
h
u
ma
n
f
a
c
t
o
r
o
n
c
u
s
t
o
me
r
sat
i
sf
a
c
t
i
o
n
i
n
c
a
l
l
c
e
n
t
r
e
s,”
BRQ
B
u
si
n
e
ss R
e
se
a
rc
h
Q
u
a
rt
e
rl
y
,
v
o
l
.
2
2
,
n
o
.
2
,
p
p
.
8
3
–
9
5
,
A
p
r
.
–
Ju
n
.
2
0
1
9
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
b
r
q
.
2
0
1
8
.
0
8
.
0
0
4
.
[
1
1
]
K
.
N
.
G
r
i
f
f
i
t
h
,
D
.
L
i
,
M
.
L
.
D
a
v
i
e
s,
S
.
D
.
P
i
z
e
r
,
a
n
d
J.
C
.
P
r
e
n
t
i
c
e
,
“
C
a
l
l
C
e
n
t
e
r
P
e
r
f
o
r
man
c
e
A
f
f
e
c
t
s
P
a
t
i
e
n
t
P
e
r
c
e
p
t
i
o
n
s
o
f
A
c
c
e
ss
a
n
d
S
a
t
i
sf
a
c
t
i
o
n
,
”
Am
e
ri
c
a
n
J
o
u
r
n
a
l
o
f
M
a
n
a
g
e
d
C
a
r
e
,
v
o
l
.
2
5
,
n
o
.
9
,
p
p
.
e
2
8
2
–
e
2
8
7
,
S
e
p
.
2
0
1
9
,
d
o
i
:
1
0
.
3
7
7
6
5
/
a
j
mc.
2
0
1
9
.
4
3
6
9
6
.
[
1
2
]
T
.
P
r
o
e
n
ç
a
a
n
d
H
.
R
o
d
r
i
g
u
e
s,
“
Emp
o
w
e
r
me
n
t
i
n
c
a
l
l
c
e
n
t
e
r
s
a
n
d
c
u
st
o
me
r
sat
i
sf
a
c
t
i
o
n
,
”
Ma
n
a
g
e
m
e
n
t
Re
s
e
a
r
c
h
:
T
h
e
J
o
u
r
n
a
l
o
f
t
h
e
I
b
e
r
o
a
m
e
r
i
c
a
n
A
c
a
d
e
m
y
o
f
M
a
n
a
g
e
m
e
n
t
,
v
o
l
.
1
9
,
n
o
.
2
,
p
p
.
1
4
3
–
1
6
1
,
J
u
n
.
2
0
2
1
,
d
o
i
:
1
0
.
1
1
0
8
/
M
R
JI
A
M
-
01
-
2
0
2
1
-
1
1
2
9
.
[
1
3
]
D
a
t
a
B
r
i
d
g
e
M
a
r
k
e
t
R
e
se
a
r
c
h
,
“
G
l
o
b
a
l
c
a
l
l
c
e
n
t
e
r
o
u
t
so
u
r
c
i
n
g
mar
k
e
t
–
i
n
d
u
st
r
y
t
r
e
n
d
s
a
n
d
f
o
r
e
c
a
st
,
”
D
a
t
a
B
r
i
d
g
e
M
a
r
k
e
t
R
e
se
a
r
c
h
,
2
0
2
4
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s:
/
/
w
w
w
.
d
a
t
a
b
r
i
d
g
e
mark
e
t
r
e
se
a
r
c
h
.
c
o
m
/
r
e
p
o
r
t
s/
g
l
o
b
a
l
-
c
a
l
l
-
c
e
n
t
e
r
-
o
u
t
so
u
r
c
i
n
g
-
mark
e
t
,
(
A
c
c
e
sse
d
:
Jan
.
2
6
,
2
0
2
6
)
.
[
1
4
]
C
.
G
.
A
r
mi
st
e
a
d
a
n
d
G
.
C
l
a
r
k
e
,
C
u
s
t
o
m
e
r S
e
rv
i
c
e
a
n
d
S
u
p
p
o
rt
:
I
m
p
l
e
m
e
n
t
i
n
g
Ef
f
e
c
t
i
v
e
S
t
r
a
t
e
g
i
e
s
.
U
K
:
P
i
t
t
ma
n
P
u
b
l
i
s
h
i
n
g
,
1
9
9
2
.
[
1
5
]
S
.
F
r
e
n
k
e
l
a
n
d
L
.
D
o
n
o
g
h
u
e
,
“
R
e
-
c
o
n
st
i
t
u
t
i
n
g
w
o
r
k
:
t
r
e
n
d
s
t
o
w
a
r
d
s
k
n
o
w
l
e
d
g
e
w
o
r
k
a
n
d
i
n
f
o
-
n
o
r
ma
t
i
v
e
c
o
n
t
r
o
l
,
”
Wo
rk,
Em
p
l
o
y
m
e
n
t
a
n
d
S
o
c
i
e
t
y
,
v
o
l
.
9
,
n
o
.
4
,
p
p
.
7
7
3
-
7
9
6
,
1
9
9
6
,
d
o
i
:
1
0
.
1
1
7
7
/
0
9
5
0
0
1
7
0
9
5
9
4
.
[
1
6
]
J.
A
n
i
st
o
n
,
In
-
H
o
u
se
v
s O
u
t
s
o
u
rc
e
C
a
l
l
C
e
n
t
e
r:
A
d
v
a
n
t
a
g
e
s
a
n
d
D
i
s
a
d
v
a
n
t
a
g
e
s
,
2
0
1
7
.
[
1
7
]
G
.
M
.
K
o
o
l
e
a
n
d
S
.
L
i
,
“
A
p
r
a
c
t
i
c
e
-
o
r
i
e
n
t
e
d
o
v
e
r
v
i
e
w
o
f
c
a
l
l
c
e
n
t
e
r
w
o
r
k
f
o
r
c
e
p
l
a
n
n
i
n
g
,
”
S
t
o
c
h
a
st
i
c
S
y
st
e
m
s
,
v
o
l
.
1
3
,
n
o
.
1
,
p
p
.
1
1
4
–
1
4
0
,
2
0
2
3
,
d
o
i
:
1
0
.
1
2
8
7
/
s
t
sy
.
2
0
2
1
.
0
0
0
8
.
[
1
8
]
T
h
e
G
l
o
b
a
l
H
e
a
l
t
h
D
e
l
i
v
e
ry
Pr
o
j
e
c
t
,
H
a
r
v
a
r
d
T
.
H
.
C
h
a
n
S
c
h
o
o
l
o
f
P
u
b
l
i
c
H
e
a
l
t
h
,
B
o
st
o
n
,
M
A
,
U
S
A
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s:
/
/
h
s
p
h
.
h
a
r
v
a
r
d
.
e
d
u
/
r
e
se
a
r
c
h
/
g
l
o
b
a
l
-
h
e
a
l
t
h
-
d
e
l
i
v
e
r
y
-
p
r
o
j
e
c
t
/
.
(
A
c
c
e
sse
d
:
Jan
.
2
6
,
2
0
2
5
)
.
[
1
9
]
T
.
T
i
e
u
,
B
.
S
t
e
v
e
n
so
n
,
T
.
B
a
l
d
i
n
i
,
T
.
K
o
h
l
e
r
,
C
.
G
o
n
z
a
l
e
z
,
a
n
d
K
.
M
c
V
a
r
y
,
“
M
P
3
2
-
0
9
U
si
n
g
L
E
A
N
M
e
t
h
o
d
o
l
o
g
y
f
o
r
U
r
o
l
o
g
y
P
a
t
i
e
n
t
C
a
l
l
C
e
n
t
e
r
I
mp
r
o
v
e
me
n
t
a
t
a
S
i
n
g
l
e
A
c
a
d
e
mi
c
I
n
st
i
t
u
t
i
o
n
,
”
Am
e
ri
c
a
n
U
r
o
l
o
g
i
c
a
l
Ass
o
c
i
a
t
i
o
n
,
v
o
l
.
1
9
3
,
n
o
.
4
,
p
.
3
6
7
,
A
p
r
.
2
0
1
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
u
r
o
.
2
0
1
5
.
0
2
.
1
4
0
5
.
[
2
0
]
S
.
C
.
L
.
K
o
h
,
A
.
G
u
n
a
se
k
a
r
a
n
,
A
.
T
h
o
mas,
a
n
d
S
.
A
r
u
n
a
c
h
a
l
a
m,
“
T
h
e
A
p
p
l
i
c
a
t
i
o
n
o
f
K
n
o
w
l
e
d
g
e
M
a
n
a
g
e
me
n
t
i
n
C
a
l
l
C
e
n
t
e
r
s,”
J
o
u
rn
a
l
o
f
K
n
o
w
l
e
d
g
e
M
a
n
a
g
e
m
e
n
t
,
v
o
l
.
9
,
n
o
.
4
,
p
p
.
5
6
-
6
9
,
2
0
0
4
,
d
o
i
:
1
0
.
1
1
0
8
/
1
3
6
7
3
2
7
0
5
1
0
6
1
0
3
3
2
.
[
2
1
]
T
.
N
.
G
a
r
a
v
a
n
e
t
a
l
.,
“
M
a
p
p
i
n
g
t
h
e
C
o
n
t
e
x
t
a
n
d
P
r
a
c
t
i
c
e
o
f
T
r
a
i
n
i
n
g
D
e
v
e
l
o
p
me
n
t
a
n
d
H
R
D
i
n
Eu
r
o
p
e
a
n
C
a
l
l
C
e
n
t
e
r
s,”
J
o
u
r
n
a
l
o
f
Eu
r
o
p
e
a
n
I
n
d
u
s
t
ri
a
l
T
ra
i
n
i
n
g
,
v
o
l
.
3
2
,
n
o
.
8
/
9
,
p
p
.
6
1
2
-
7
2
8
,
2
0
0
8
,
d
o
i
:
1
0
.
1
1
0
8
/
0
3
0
9
0
5
9
0
8
1
0
9
1
8
7
6
4
.
[
2
2
]
K
.
B
.
C
.
S
a
x
e
n
a
a
n
d
S
.
B
h
a
r
a
d
w
a
j
,
“
B
u
s
i
n
e
ss
P
r
o
c
e
ss
O
u
t
so
u
r
c
i
n
g
f
o
r
S
t
r
a
t
e
g
i
c
A
d
v
a
n
t
a
g
e
,
”
S
l
o
a
n
Ma
n
a
g
e
m
e
n
t
R
e
v
i
e
w
,
v
o
l
.
4
0
,
n
o
.
1
2
,
p
p
.
9
-
2
1
,
2
0
0
9
.
[
2
3
]
A
.
M
.
D
e
a
n
,
“
S
e
r
v
i
c
e
Q
u
a
l
i
t
y
i
n
C
a
l
l
C
e
n
t
e
r
s:
I
mp
l
i
c
a
t
i
o
n
s
f
o
r
C
u
s
t
o
me
r
L
o
y
a
l
t
y
,
”
Ma
n
a
g
i
n
g
S
e
r
v
i
c
e
Q
u
a
l
i
t
y
,
v
o
l
.
1
2
,
n
o
.
6
,
p
p
.
4
1
4
-
4
2
3
,
2
0
0
2
,
d
o
i
:
1
0
.
1
1
0
8
/
0
9
6
0
4
5
2
0
2
1
0
4
5
1
8
9
4
.
[
2
4
]
A
.
G
i
l
m
o
r
e
,
“
C
a
l
l
C
e
n
t
r
e
M
a
n
a
g
e
me
n
t
:
I
s
S
e
r
v
i
c
e
Q
u
a
l
i
t
y
a
P
r
i
o
r
i
t
y
?
”
M
a
n
a
g
i
n
g
S
e
r
v
i
c
e
Q
u
a
l
i
t
y
:
A
n
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
,
v
o
l
.
1
1
n
o
.
3
,
p
p
.
1
5
3
-
1
5
9
,
2
0
0
1
,
d
o
i
:
1
0
.
1
1
0
8
/
0
9
6
0
4
5
2
0
1
1
0
3
9
1
3
1
5
2
0
0
1
.
[
2
5
]
K
.
O
.
Y
a
m
a
k
,
“
I
m
p
a
c
t
o
f
o
u
t
so
u
rc
i
n
g
o
n
e
f
f
e
c
t
i
v
e
n
e
ss o
f
c
a
l
l
c
e
n
t
e
r se
r
v
i
c
e
s
,
”
J
o
u
r
n
a
l
o
f
Asi
a
n
R
e
se
a
rc
h
,
v
o
l
.
5
,
n
o
.
4
,
p
p
.
2
9
–
4
2
,
D
e
c
.
2
0
2
1
,
d
o
i
:
1
0
.
2
2
1
5
8
/
j
a
r
.
v
5
n
4
p
2
9
.
[
2
6
]
A
.
U
l
l
a
h
e
t
a
l
.,
“
C
u
st
o
me
r
A
n
a
l
y
si
s
U
si
n
g
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
-
B
a
se
d
C
l
a
ss
i
f
i
c
a
t
i
o
n
A
l
g
o
r
i
t
h
ms
f
o
r
Ef
f
e
c
t
i
v
e
S
e
g
me
n
t
a
t
i
o
n
U
si
n
g
R
e
c
e
n
c
y
,
F
r
e
q
u
e
n
c
y
,
M
o
n
e
t
a
r
y
,
a
n
d
T
i
me
,
”
S
e
n
s
o
rs
,
v
o
l
.
2
3
,
n
o
.
6
,
p
.
3
1
8
0
,
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
s
2
3
0
6
3
1
8
0
.
[
2
7
]
L
.
O
.
I
h
e
m
e
a
n
d
Ş
.
O
z
a
n
,
“
A
N
o
v
e
l
S
e
mi
-
S
u
p
e
r
v
i
se
d
F
r
a
me
w
o
r
k
f
o
r
C
a
l
l
C
e
n
t
e
r
A
g
e
n
t
M
a
l
p
r
a
c
t
i
c
e
D
e
t
e
c
t
i
o
n
v
i
a
N
e
u
r
a
l
F
e
a
t
u
r
e
L
e
a
r
n
i
n
g
,
”
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
.
2
0
8
,
p
.
1
1
8
1
7
3
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
sw
a
.
2
0
2
2
.
1
1
8
1
7
3
.
Evaluation Warning : The document was created with Spire.PDF for Python.