I
nte
rna
t
io
na
l J
o
urna
l o
f
I
nfo
r
m
a
t
ics a
nd
Co
mm
u
n
ica
t
io
n T
ec
hn
o
lo
g
y
(
I
J
-
I
CT
)
Vo
l.
3
,
No
.
2
,
Ju
ne
201
4
,
p
p
.
88
~
96
I
SS
N:
2252
-
8776
88
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
jo
u
r
n
a
l.c
o
m/o
n
lin
e/in
d
ex
.
p
h
p
/I
J
I
C
T
Decisio
n Su
ppo
rt
Sys
te
m
f
o
r
a
Cus
to
m
e
r Relat
io
nship
M
a
na
g
em
e
nt
Ca
se Study
Ö
zg
e
K
a
rt
*
,
Alp K
ut
*
,
Vla
di
m
i
r
Ra
dev
s
k
i
*
*
*
D
e
p
a
rte
m
e
n
t
o
f
Co
m
p
u
ter E
n
g
in
e
e
rin
g
,
Do
k
u
z
Ey
lu
l
Un
iv
e
rsity
*
*
F
a
c
u
lt
y
o
f
Co
n
tem
p
o
ra
r
y
S
c
ie
n
c
e
s an
d
T
e
c
h
n
o
l
o
g
ies
,
S
EE
U
n
iv
e
rsit
y
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Feb
2
8
,
2
0
1
4
R
ev
i
s
ed
Ma
y
1
9
,
2
0
1
4
A
cc
ep
ted
Ma
y
2
5
,
2
0
1
4
Da
ta
m
in
in
g
is
a
c
o
m
p
u
tatio
n
a
l
a
p
p
ro
a
c
h
a
im
in
g
to
d
isc
o
v
e
r
h
id
d
e
n
a
n
d
v
a
lu
a
b
le
in
f
o
rm
a
ti
o
n
in
larg
e
d
a
tas
e
ts.
It
h
a
s
g
a
in
e
d
im
p
o
rtan
c
e
r
e
c
e
n
tl
y
in
th
e
w
id
e
a
re
a
o
f
c
o
m
p
u
tatio
n
a
l
a
m
o
n
g
w
h
ich
m
a
n
y
in
th
e
d
o
m
a
in
o
f
Bu
sin
e
ss
In
f
o
rm
a
ti
c
s.
T
h
is
p
a
p
e
r
f
o
c
u
se
s
o
n
a
p
p
li
c
a
ti
o
n
s
o
f
d
a
ta
m
in
in
g
in
Cu
sto
m
e
r
Re
latio
n
sh
ip
M
a
n
a
g
e
m
e
n
t
(CRM
).
T
h
e
c
o
re
o
f
o
u
r
a
p
p
li
c
a
ti
o
n
is
a
c
las
si
f
ier
b
a
se
d
o
n
th
e
n
a
iv
e
Ba
y
e
sia
n
c
las
si
f
ica
ti
o
n
.
T
h
e
a
c
c
u
ra
c
y
ra
te
o
f
th
e
m
o
d
e
l
is
d
e
ter
m
in
e
d
b
y
d
o
in
g
c
ro
ss
v
a
li
d
a
ti
o
n
.
T
h
e
re
su
lt
s
d
e
m
o
n
stra
ted
th
e
a
p
p
li
c
a
b
il
i
ty
a
n
d
e
ffe
c
ti
v
e
n
e
s
s
o
f
th
e
p
ro
p
o
se
d
m
o
d
e
l.
Na
iv
e
Ba
y
e
sia
n
c
las
si
f
ier
re
p
o
rted
h
ig
h
a
c
c
u
ra
c
y
.
S
o
th
e
c
las
sif
ica
ti
o
n
ru
les
c
a
n
b
e
u
se
d
to
su
p
p
o
rt
d
e
c
isio
n
m
a
k
in
g
in
CRM
f
ield
.
T
h
e
a
i
m
o
f
th
is
stu
d
y
is
to
a
p
p
ly
th
e
d
a
ta
m
in
in
g
m
o
d
e
l
to
th
e
b
a
n
k
in
g
se
c
to
r
a
s
e
x
a
m
p
le
c
a
se
stu
d
y
.
T
h
is
w
o
r
k
a
lso
c
o
n
tai
n
s
a
n
e
x
a
m
p
le
d
a
ta
se
t
re
late
d
w
it
h
c
u
sto
m
e
rs
to
p
re
d
ict
if
th
e
c
li
e
n
t
w
il
l
su
b
sc
rib
e
a
term
d
e
p
o
sit.
T
h
e
re
su
lt
s
o
f
th
e
im
p
le
m
e
n
tatio
n
a
re
a
v
a
il
a
b
le o
n
a
m
o
b
il
e
p
latf
o
rm
.
K
ey
w
o
r
d
:
CRM
C
u
s
to
m
er
r
elatio
n
s
h
ip
Ma
n
ag
e
m
e
n
t
Dec
is
io
n
s
u
p
p
o
r
t
Co
p
y
rig
h
t
©
2
0
1
4
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
:
First
Au
th
o
r
,
Dep
ar
te
m
en
t o
f
C
o
m
p
u
ter
E
n
g
in
ee
r
i
n
g
,
Do
k
u
z
E
y
lu
l
Un
i
v
er
s
i
t
y
,
T
in
az
tep
e
C
a
m
p
u
s
,
B
u
ca
3
5
3
9
0
I
zm
ir
,
T
u
r
k
e
y
.
E
m
ail:
o
zg
e
@
c
s
.
d
eu
.
ed
u
.
tr
1.
I
NT
RO
D
UCT
I
O
N
I
n
th
i
s
n
e
w
er
a,
co
m
p
a
n
ies
h
a
v
e
b
eg
u
n
to
g
iv
e
m
o
r
e
atte
n
ti
o
n
to
cu
s
to
m
er
s
’
p
er
s
o
n
al
p
r
ef
er
en
ce
s
.
A
h
i
g
h
co
n
s
id
er
atio
n
o
f
th
e
c
u
s
t
o
m
er
s
’
p
er
s
o
n
a
l
p
r
eferen
ce
s
i
s
co
n
s
id
er
ed
to
b
e
a
v
er
y
i
m
p
o
r
tan
t
asp
ec
t
o
f
t
h
e
n
o
w
ad
a
y
s
f
u
n
c
tio
n
i
n
g
o
f
t
h
e
co
m
p
an
ie
s
.
C
R
M
f
o
cu
s
e
s
to
th
e
m
an
a
g
er
ial
a
s
p
ec
ts
o
f
o
r
g
an
iza
tio
n
al
co
m
m
u
n
icatio
n
to
th
e
c
u
s
to
m
er
s
an
d
p
r
o
s
p
ec
ts
.
R
aisi
n
g
th
e
cu
s
to
m
er
s
ati
s
f
ac
t
io
n
is
o
n
e
o
f
t
h
e
m
ai
n
s
u
b
j
ec
ts
o
f
in
ter
est
i
n
t
h
is
ar
ea
.
O
n
e
to
o
n
e
m
ar
k
eti
n
g
s
tr
ate
g
ies
b
eg
an
to
co
m
e
to
th
e
f
o
r
e.
T
h
ese
ar
e
o
n
l
y
f
e
w
co
n
te
m
p
o
r
ar
y
ac
t
u
al
ex
a
m
p
les
p
u
ttin
g
t
h
e
w
h
o
le
C
R
M
co
n
c
ep
t i
n
th
e
f
o
cu
s
o
f
m
a
n
y
r
esea
r
ch
er
s
[
1
]
[
2
]
.
“CR
M
is
a
n
in
te
g
r
ated
in
f
o
r
m
atio
n
s
y
s
te
m
t
h
at
is
u
s
ed
to
p
lan
,
s
ch
ed
u
le
an
d
co
n
tr
o
l
th
e
p
r
e
-
s
ales
an
d
p
o
s
t
-
s
ales
ac
tiv
ities
i
n
a
n
o
r
g
a
n
iz
atio
n
.
C
R
M
e
m
b
r
ac
es
all
a
s
p
ec
ts
o
f
d
ea
li
n
g
w
i
th
p
r
o
s
p
ec
ts
a
n
d
cu
s
to
m
er
s
,
in
cl
u
d
in
g
th
e
ca
l
l
ce
n
tr
e,
s
al
es
-
f
o
r
ce
,
m
ar
k
et
in
g
,
tech
n
ical
s
u
p
p
o
r
t
an
d
f
ield
s
er
v
ice.
T
h
e
p
r
i
m
ar
y
g
o
al
o
f
C
R
M
i
s
to
i
m
p
r
o
v
e
lo
n
g
-
ter
m
g
r
o
w
t
h
a
n
d
p
r
o
f
itab
ilit
y
th
r
o
u
g
h
a
b
etter
u
n
d
er
s
ta
n
d
in
g
o
f
cu
s
to
m
er
b
eh
a
v
io
r
.
C
R
M
ai
m
s
to
p
r
o
v
id
e
m
o
r
e
ef
f
ec
tiv
e
f
ee
d
b
ac
k
an
d
i
m
p
r
o
v
ed
in
teg
r
atio
n
to
b
etter
g
au
g
e
t
h
e
r
etu
r
n
o
n
in
v
est
m
e
n
t (
R
OI
)
in
t
h
ese
ar
ea
s
.
”
[
3
]
Data
m
in
i
n
g
h
as
a
g
r
ea
t
co
n
t
r
ib
u
tio
n
to
t
h
e
e
x
tr
ac
tio
n
o
f
k
n
o
w
led
g
e
a
n
d
in
f
o
r
m
at
io
n
w
h
ic
h
h
av
e
b
ee
n
h
id
d
en
in
a
lar
g
e
v
o
l
u
m
e
o
f
d
ata[
4
]
.
T
h
e
co
n
ce
p
t
o
f
cu
s
to
m
er
s
atis
f
ac
tio
n
an
d
lo
y
alt
y
(
C
S
&
L
)
h
as
attr
ac
ted
m
u
c
h
atte
n
tio
n
i
n
r
ec
en
t
y
ea
r
s
.
A
k
e
y
m
o
ti
v
atio
n
f
o
r
th
e
f
a
s
t
g
r
o
w
i
n
g
e
m
p
h
a
s
is
o
n
C
S
&
L
ca
n
b
e
attr
ib
u
ted
to
th
e
f
ac
t
th
at
h
i
g
h
er
cu
s
to
m
er
s
atis
f
ac
tio
n
a
n
d
lo
y
alt
y
ca
n
lead
to
s
tr
o
n
g
er
co
m
p
etiti
v
e
p
o
s
itio
n
r
esu
lti
n
g
in
lar
g
er
m
ar
k
et
s
h
ar
e
an
d
p
r
o
f
itab
ilit
y
[
5
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
I
C
T
I
SS
N:
2252
-
8776
Dec
is
io
n
S
u
p
p
o
r
t S
ystem
F
o
r
A
C
u
s
to
mer …
(
Özg
e
K
a
r
t
)
89
Z
e
n
g
y
o
u
He
e
t
al.
[
6
]
i
m
p
le
m
en
ted
class
o
u
tlier
f
ac
to
r
s
as
l
o
y
al
t
y
s
co
r
es
f
o
r
f
in
d
i
n
g
cu
s
t
o
m
er
s
w
h
o
ar
e
ab
o
u
t
to
lo
s
e
t
h
e
lo
y
alt
y
s
e
g
m
e
n
t.
I
n
th
eir
s
t
u
d
y
,
th
e
y
co
n
s
id
er
ed
th
e
clas
s
o
u
tlier
d
etec
tio
n
p
r
o
b
le
m
„
g
iv
e
n
a
s
et
o
f
o
b
s
er
v
at
io
n
s
with
clas
s
lab
els,
f
in
d
t
h
o
s
e
th
a
t
ar
o
u
s
e
s
u
s
p
icio
n
s
,
tak
i
n
g
in
t
o
a
cc
o
u
n
t
th
e
cla
s
s
lab
els‟
.
A
s
e
m
a
n
tic
o
u
tlier
is
a
d
ata
p
o
in
t,
w
h
ic
h
lo
o
k
s
r
e
g
u
l
ar
ac
co
r
d
in
g
to
d
ata
p
o
in
ts
i
n
an
o
th
er
cla
s
s
w
h
ile
s
ee
m
s
ir
r
eg
u
lar
ac
co
r
d
in
g
to
d
ata
p
o
in
ts
in
t
h
e
s
a
m
e
class
.
T
h
ey
d
ev
elo
p
ed
th
e
id
ea
o
f
class
o
u
tlier
a
n
d
p
r
o
p
o
s
ed
n
e
w
s
o
l
u
tio
n
a
s
an
ex
ten
s
io
n
o
f
w
el
l
k
n
o
w
n
o
u
tl
ier
d
etec
tio
n
alg
o
r
it
h
m
s
to
t
h
i
s
ca
s
e.
S
h
i
n
-
Yu
a
n
Hu
n
g
et
a
l.
[
7
]
co
m
p
ar
ed
s
ev
e
r
al
d
ata
m
in
in
g
tec
h
n
iq
u
es
th
a
t c
an
g
iv
e
a
„
p
r
o
p
en
s
it
y
-
to
-
c
h
u
r
n
‟
p
o
in
t a
t r
e
g
u
lar
in
ter
v
a
ls
to
e
v
er
y
m
o
b
ile
o
p
er
ato
r
cu
s
to
m
er
s
.
T
h
e
y
u
s
e
d
cu
s
to
m
er
d
e
m
o
g
r
ap
h
ic
s
,
b
illi
n
g
i
n
f
o
r
m
atio
n
,
co
n
tr
ac
t/s
er
v
ice
s
ta
tu
s
,
ca
ll
d
e
tail
r
ec
o
r
d
s
,
an
d
s
er
v
ice
c
h
a
n
g
e
lo
g
a
s
c
u
s
to
m
er
d
ata.
A
s
a
r
esu
lt,
b
o
th
d
ec
is
io
n
tr
ee
an
d
n
eu
r
al
n
et
w
o
r
k
tec
h
n
iq
u
e
s
g
a
v
e
s
u
cc
es
s
f
u
l
c
h
u
r
n
p
r
ed
ictio
n
m
o
d
els.
S.M
.
S.
Ho
s
s
ei
n
i
et
al.
[
8
]
p
r
o
p
o
s
ed
a
n
e
w
p
r
o
ce
d
u
r
e,
w
h
ich
i
s
an
ex
p
an
s
io
n
o
f
R
FM
(
R
ec
en
c
y
Fre
q
u
en
c
y
Mo
n
etar
y
)
m
o
d
el
b
y
ad
d
in
g
o
n
e
p
ar
a
m
eter
,
in
s
er
tin
g
W
R
FM
-
b
ased
m
et
h
o
d
to
K
-
m
ea
n
s
alg
o
r
it
h
m
i
m
p
le
m
e
n
ted
i
n
D
M
w
ith
K
-
o
p
ti
m
u
m
w
it
h
r
esp
ec
t
to
Dav
ie
s
–
B
o
u
ld
in
I
n
d
ex
,
an
d
t
h
en
cla
s
s
i
f
y
i
n
g
cu
s
to
m
er
p
r
o
d
u
ct
lo
y
alt
y
in
B
2
B
co
n
ce
p
t.
T
h
e
r
esu
lt
s
p
r
o
v
id
ed
a
h
ig
h
er
ab
il
it
y
to
th
e
co
m
p
a
n
y
to
s
p
ec
if
y
its
cu
s
to
m
er
lo
y
alt
y
i
n
m
ar
k
etin
g
s
tr
ate
g
y
.
R
.
S.
C
h
e
n
et
al.
[
9
]
d
ev
elo
p
ed
class
i
f
icatio
n
o
f
c
h
o
s
en
c
u
s
to
m
e
r
s
in
to
cl
u
s
ter
s
u
s
in
g
R
FM
m
o
d
el
to
d
eter
m
i
n
e
h
ig
h
-
p
r
o
f
it,
g
o
ld
cu
s
to
m
er
s
.
T
h
e
y
u
s
ed
d
ata
m
i
n
i
n
g
tec
h
n
iq
u
es a
n
d
d
is
co
v
er
ed
t
h
e
ac
t
u
al
c
o
n
s
u
m
p
tio
n
p
atter
n
o
f
cu
s
to
m
er
s
a
n
d
b
eh
av
io
r
al
c
h
an
g
es
i
n
tr
en
d
s
,
w
h
ic
h
w
ill
a
llo
w
m
an
a
g
e
m
e
n
t
to
d
etec
t
p
o
ten
tial
ch
a
n
g
es
o
f
cu
s
to
m
er
p
r
ef
er
en
ce
,
an
d
to
p
r
ev
en
t
cu
s
to
m
er
lo
s
es.
C
h
ao
-
T
o
n
Su
et
al.
[
1
0
]
p
r
o
p
o
s
ed
an
E
-
C
KM
m
o
d
el
tak
i
n
g
ad
v
a
n
tag
e
o
f
e
s
ti
m
ati
n
g
m
et
h
o
d
s
b
ased
o
n
d
ata
m
i
n
in
g
,
f
o
r
t
h
e
d
e
v
elo
p
m
e
n
t
o
f
i
n
n
o
v
ati
v
e
p
r
o
d
u
cts
th
at
m
ee
t
p
o
ten
tial
r
eq
u
ir
e
m
e
n
ts
o
f
c
u
s
to
m
er
s
.
T
h
e
y
u
s
ed
w
eb
-
b
ased
s
u
r
v
e
y
s
an
d
d
ata
m
i
n
in
g
tech
n
iq
u
e
s
t
o
o
b
tain
cu
s
to
m
er
k
n
o
w
led
g
e
f
r
o
m
d
i
f
f
er
en
t
m
ar
k
et
s
eg
m
e
n
ts
.
T
h
e
co
r
e
p
ar
t
o
f
C
R
M
ac
tiv
itie
s
is
to
u
n
d
er
s
tan
d
cu
s
to
m
er
r
eq
u
ir
e
m
e
n
ts
a
n
d
r
etain
p
r
o
f
itab
le
cu
s
to
m
er
s
.
Data
m
i
n
i
n
g
tech
n
iq
u
es
s
u
ch
as
cla
s
s
i
f
icatio
n
,
clu
s
ter
i
n
g
etc
,
h
a
v
e
i
m
p
o
r
ta
n
t
r
o
le
to
p
lay
in
C
R
M
ap
p
licatio
n
s
.
W
it
h
d
ata
m
i
n
i
n
g
ap
p
licatio
n
s
,
d
atab
ases
,
r
e
co
r
d
s
in
lar
g
e
co
m
p
a
n
ie
s
ca
n
b
e
co
n
v
er
ted
in
to
m
ea
n
in
g
f
u
l
in
f
o
r
m
a
tio
n
.
I
n
f
a
ct
th
r
o
u
g
h
th
e
s
e
p
r
o
ce
s
s
es
i
m
p
o
r
tan
t
k
n
o
w
led
g
e
a
n
d
in
f
o
r
m
atio
n
ar
e
ex
tr
ac
te
d
f
r
o
m
t
h
e
lar
g
e
v
o
lu
m
e
o
f
d
ata,
w
h
er
e
t
h
e
y
h
a
v
e
b
ee
n
h
id
d
en
p
r
ev
io
u
s
l
y
[
1
1
]
[
5
]
.
T
h
e
m
ai
n
p
u
r
p
o
s
e
o
f
t
h
is
s
t
u
d
y
i
s
a
n
i
m
p
le
m
en
tatio
n
o
f
Da
t
a
Min
i
n
g
al
g
o
r
ith
m
f
o
r
e
x
tr
ac
t
in
g
h
id
d
en
in
f
o
r
m
atio
n
f
r
o
m
t
h
e
co
r
p
o
r
a
te
d
atab
ases
an
d
d
atasets
th
at
co
m
p
an
ies
ca
n
u
s
e
i
n
d
ec
is
i
o
n
m
ak
in
g
p
r
o
ce
s
s
.
T
h
is
v
alu
ab
le
in
f
o
r
m
a
tio
n
is
a
cc
ess
ed
v
ia
a
W
C
F
s
er
v
ice
an
d
p
r
esen
ted
o
n
a
m
o
b
ile
p
latf
o
r
m
[
1
2
]
.
A
s
a
ca
s
e
s
tu
d
y
,
cu
s
to
m
er
d
ataset
p
r
o
v
id
ed
f
r
o
m
a
b
a
n
k
w
as
u
s
ed
to
p
r
ed
ict
if
t
h
e
cl
ien
t
w
il
l
s
u
b
s
c
r
ib
e
a
ter
m
d
ep
o
s
i
t
B
ay
e
s
ia
n
C
las
s
i
f
icatio
n
is
i
m
p
le
m
en
ted
o
n
th
i
s
d
ata
s
et
.
A
d
ec
is
io
n
s
u
p
p
o
r
t
s
y
s
te
m
is
g
e
n
er
ated
to
h
elp
t
h
e
in
s
t
itu
tio
n
to
p
r
ed
ict
th
e
b
e
h
a
v
io
r
o
f
a
n
e
w
cu
s
to
m
er
.
T
h
is
p
r
ed
ictio
n
is
p
r
ese
n
ted
o
n
a
m
o
b
ile
p
latf
o
r
m
.
T
h
e
d
ec
is
io
n
s
u
p
p
o
r
t s
y
s
te
m
is
ac
c
ess
ed
v
ia
W
C
F ser
v
ice.
2.
P
RO
P
O
SE
D
M
E
T
H
O
D/AL
G
O
RI
T
HM
S
w
if
t
[
1
3
]
d
escr
ib
ed
C
R
M
d
i
m
e
n
s
io
n
s
a
s
:
C
u
s
to
m
er
I
d
en
ti
f
icatio
n
,
C
u
s
to
m
er
Attr
ac
tio
n
,
C
u
s
to
m
er
R
eten
tio
n
a
n
d
C
u
s
to
m
er
De
v
elo
p
m
e
n
t.
C
R
M
s
tar
ts
w
it
h
cu
s
to
m
er
id
en
tific
atio
n
.
T
h
is
f
ir
s
t
s
tep
is
ab
o
u
t
d
is
co
v
er
in
g
th
e
e
n
titi
e
s
th
at
a
r
e
ten
d
to
b
ec
o
m
e
c
u
s
to
m
er
s
o
r
w
h
o
b
r
in
g
t
h
e
m
o
s
t
p
r
o
f
it
to
th
e
co
m
p
a
n
y
.
I
n
ad
d
itio
n
,
it
co
n
tain
s
a
n
al
y
zin
g
cu
s
to
m
er
s
w
h
o
ar
e
ab
o
u
t
to
l
o
s
t
to
t
h
e
co
m
p
eti
tio
n
a
n
d
h
o
w
th
e
y
ca
n
b
e
w
o
n
ag
ai
n
.
Af
ter
id
en
ti
f
y
in
g
th
e
s
eg
m
en
t
s
o
f
p
o
ten
tial
cu
s
to
m
e
r
s
,
co
m
p
an
ies
m
a
y
co
n
s
u
m
e
ef
f
o
r
t
an
d
r
eso
u
r
ce
s
f
o
r
attr
ac
tin
g
th
e
p
o
ten
tial c
u
s
t
o
m
er
s
eg
m
e
n
ts
.
Fig
u
r
e
1
.
C
R
M
Di
m
e
n
s
io
n
s
a
n
d
Data
Min
i
n
g
Mo
d
els
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8776
IJ
-
I
C
T
Vo
l.
3
,
No
.
2
,
J
u
n
e
20
1
4
:
88
–
96
90
I
n
C
R
M,
C
u
s
to
m
er
r
eten
ti
o
n
is
t
h
e
m
ain
in
ter
e
s
t.
C
u
s
to
m
er
s
ati
s
f
ac
tio
n
,
w
h
ic
h
m
ea
n
s
t
h
e
co
m
p
ar
is
o
n
o
f
c
u
s
to
m
er
s
’
n
ee
d
s
an
d
h
o
w
m
u
c
h
th
e
y
ar
e
s
ati
s
f
ied
,
i
s
t
h
e
m
aj
o
r
f
ac
to
r
f
o
r
r
etain
i
n
g
cu
s
to
m
er
s
.
C
o
m
p
o
n
e
n
ts
o
f
c
u
s
to
m
er
r
eten
tio
n
co
n
tai
n
lo
y
alt
y
p
r
o
g
r
a
m
s
,
o
n
e
-
to
-
o
n
e
m
ar
k
eti
n
g
an
d
co
m
p
la
in
t
s
m
a
n
ag
e
m
e
n
t.
C
u
s
to
m
er
d
ev
e
lo
p
m
e
n
t
co
n
tain
s
tr
an
s
ac
tio
n
d
en
s
it
y
,
v
al
u
es
o
f
tr
an
s
ac
ti
o
n
s
an
d
c
u
s
to
m
er
p
r
o
f
i
tab
ilit
y
.
C
o
m
p
o
n
e
n
t
s
o
f
c
u
s
to
m
er
d
ev
elo
p
m
e
n
t
co
n
tai
n
u
p
/cr
o
s
s
s
e
lli
n
g
a
n
d
m
ar
k
et
b
a
s
k
et
a
n
al
y
s
is
[
1
5
]
.
Ah
m
ed
[
1
4
]
d
escr
ib
ed
th
e
t
y
p
es
o
f
d
ata
m
i
n
in
g
m
o
d
el
as
As
s
o
ciatio
n
,
C
lass
if
icatio
n
,
C
l
u
s
t
er
in
g
,
Fo
r
ec
asti
n
g
,
R
eg
r
es
s
io
n
,
Seq
u
e
n
ce
Dis
co
v
er
y
a
n
d
Vi
s
u
al
iza
tio
n
(
F
ig
u
r
e
1
)
.
I
n
th
is
s
t
u
d
y
,
B
a
y
es
ian
cla
s
s
i
f
icatio
n
w
h
ich
is
o
n
e
o
f
th
e
w
el
l k
n
o
w
n
cla
s
s
i
f
i
ca
tio
n
alg
o
r
it
h
m
s
is
i
m
p
le
m
e
n
ted
.
B
ay
e
s
ia
n
class
if
ica
tio
n
ca
n
p
r
ed
ict
class
m
e
m
b
er
s
h
ip
p
r
o
b
ab
ilit
ies.
Naï
v
e
B
a
y
esia
n
clas
s
if
ier
s
ar
e
b
ased
o
n
B
a
y
es
’
t
h
eo
r
e
m
.
T
h
e
y
ar
e
s
h
o
w
n
to
b
e
v
er
y
p
o
w
er
f
u
l
to
o
l
in
d
ata
m
i
n
i
n
g
a
n
d
d
ec
is
io
n
s
u
p
p
o
r
t
s
y
s
te
m
s
co
n
s
eq
u
e
n
tl
y
.
I
n
t
h
is
ap
p
r
o
ac
h
lear
n
in
g
i
s
f
o
r
m
u
lat
ed
as
a
f
o
r
m
o
f
p
r
o
b
ab
ilis
tic
i
n
f
er
e
n
ce
,
u
s
i
n
g
t
h
e
o
b
s
er
v
atio
n
s
to
u
p
d
ate
a
p
r
io
r
d
is
tr
ib
u
tio
n
o
v
er
h
y
p
o
t
h
eses
i
n
B
a
y
es c
la
s
s
i
f
icatio
n
[
4
]
.
Giv
e
n
a
h
y
p
o
t
h
esi
s
h
a
n
d
d
ata
D
w
h
ic
h
co
n
ce
r
n
ed
w
it
h
t
h
e
h
y
p
o
t
h
esi
s
:
(
1
)
P
(
h
)
: in
d
ep
en
d
en
t p
r
o
b
ab
ilit
y
o
f
h
(
p
r
io
r
p
r
o
b
ab
ilit
y
)
P
(
D)
: in
d
ep
en
d
en
t p
r
o
b
ab
ilit
y
o
f
D
(
d
ata
ev
id
en
ce
)
P
(
D|
h
)
: c
o
n
d
itio
n
al
p
r
o
b
ab
ilit
y
o
f
D
g
iv
e
n
h
(
li
k
eli
h
o
o
d
)
P
(
h
|
D)
: c
o
n
d
itio
n
al
p
r
o
b
ab
ilit
y
o
f
h
g
i
v
en
D
(
p
o
s
ter
io
r
p
r
o
b
ab
ilit
y
)
T
h
e
g
o
al
o
f
B
ay
e
s
T
h
eo
r
em
i
s
to
s
p
ec
if
y
t
h
e
m
o
s
t p
r
o
b
ab
le
h
y
p
o
t
h
esi
s
f
r
o
m
t
h
e
g
iv
e
n
d
at
a
D.
P
r
io
r
p
r
o
b
a
b
ilit
y
o
f
h
,
P
(
h
)
:
is
th
e
p
r
o
b
a
b
ilit
y
o
f
b
ei
n
g
h
is
a
co
r
r
ec
t
h
y
p
o
th
e
s
is
.
P
r
io
r
p
r
o
b
ab
ilit
y
o
f
D,
P
(
D)
:
is
t
h
e
p
r
o
b
ab
ilit
y
o
f
tr
ai
n
i
n
g
d
ata
D
w
ill
b
e
o
b
s
er
v
ed
.
C
o
n
d
i
tio
n
al
P
r
o
b
ab
ilit
y
o
f
o
b
s
er
v
at
i
o
n
D,
P
(
D
|
h
)
:
is
t
h
e
p
r
o
b
a
b
ilit
y
o
f
o
b
s
er
v
i
n
g
d
ata
D
g
i
v
en
s
o
m
e
w
o
r
ld
in
w
h
ich
h
y
p
o
t
h
esi
s
h
h
o
ld
s
[
1
1
]
.
Naiv
e
B
a
y
es
ai
m
s
to
s
i
m
p
li
f
y
th
e
esti
m
a
tio
n
p
r
o
b
lem
b
y
as
s
u
m
i
n
g
t
h
at
th
e
d
if
f
er
e
n
t
in
p
u
t
f
ea
tu
r
e
s
ar
e
co
n
d
itio
n
all
y
i
n
d
ep
en
d
en
t
.
T
h
at
is
,
th
e
y
ar
e
a
s
s
u
m
ed
to
b
e
in
d
ep
en
d
en
t
w
h
e
n
co
n
d
iti
o
n
ed
o
n
t
h
e
clas
s
.
Ma
th
e
m
atica
ll
y
,
f
o
r
in
p
u
ts
x
∈
R
^d
,
it is
ex
p
r
ess
ed
as:
(
2
)
Fo
r
th
is
r
e
aso
n
,
it
is
o
n
l
y
n
e
ed
ed
to
g
et
P
(
Xi
|
C
)
f
o
r
ev
e
r
y
p
o
s
s
ib
le
co
u
p
le
o
f
a
ca
teg
o
r
y
an
d
a
f
ea
t
u
r
e
-
v
al
u
e.
B
ay
e
s
ian
ca
te
g
o
r
izatio
n
n
o
r
m
all
y
d
ea
ls
w
it
h
ca
teg
o
r
ical
d
ata.
B
u
t
in
o
u
r
ca
s
e,
co
n
tin
u
o
u
s
attr
ib
u
tes
ex
is
t
i
n
d
ataset.
T
o
class
if
y
co
n
ti
n
u
o
u
s
attr
ib
u
te
s
,
Gau
s
s
ia
n
d
is
tr
ib
u
tio
n
i
s
u
s
ed
.
I
f
attr
ib
u
t
e
X
h
as
co
n
tin
u
o
u
s
v
alu
e
s
in
s
tead
o
f
ca
teg
o
r
i
ca
l
v
alu
e
s
,
f
o
r
ca
lcu
la
tin
g
P
(
Xi
|
Y)
,
Gau
s
s
ia
n
d
is
tr
ib
u
t
io
n
is
u
s
ed
to
ca
lcu
late
t
h
e
p
r
o
b
a
b
ilit
y
o
f
X
(
P
(
X
|
Y)
)
.
Fo
r
ea
ch
co
m
b
in
atio
n
o
f
a
co
n
ti
n
u
o
u
s
v
alu
e
X
i
an
d
a
class
v
alu
e
f
o
r
Y,
y
k
,
h
as
a
m
ea
n
,
μ
ik
,
an
d
s
tan
d
ar
d
d
ev
iatio
n
(
v
ar
ian
ce
)
σ
ik
b
ased
o
n
v
a
lu
e
s
Xi
i
n
clas
s
y
k
.
Fo
r
est
i
m
a
tio
n
P
(
Xi
|
Y=
y
k
)
o
f
t
h
is
e
x
a
m
p
le,
Gau
s
s
ia
n
d
is
tr
ib
u
tio
n
o
f
Xi
i
s
d
ef
in
ed
b
y
μ
ik
(
m
ea
n
)
an
d
σ
ik
(
v
ar
ian
ce
)
d
ep
en
d
s
o
n
Y:
(
3
)
)
(
)
(
)
|
(
)
|
(
D
P
h
P
h
D
P
D
h
P
p
o
st
e
r
i
o
r
p
r
o
b
a
b
i
l
i
t
y
d
a
t
a
e
v
i
d
e
n
c
e
l
i
k
e
l
i
h
o
o
d
p
r
i
o
r
p
r
o
p
a
p
i
l
i
t
y
d
i
i
C
x
p
C
X
P
1
|
)
|
(
2
2
2
)
(
e
x
p
2
1
)
|
(
ik
ik
i
ik
k
i
X
y
Y
X
P
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
I
C
T
I
SS
N:
2252
-
8776
Dec
is
io
n
S
u
p
p
o
r
t S
ystem
F
o
r
A
C
u
s
to
mer …
(
Özg
e
K
a
r
t
)
91
3.
RE
S
E
ARCH
M
E
T
H
OD
T
h
e
d
ataset
u
s
ed
i
n
t
h
i
s
ca
s
e
s
tu
d
y
,
is
ab
o
u
t
a
m
ar
k
eti
n
g
ca
m
p
a
ig
n
o
f
a
b
an
k
.
I
t
co
n
s
i
s
ts
o
f
3
0
9
0
2
r
ec
o
r
d
s
.
A
ttrib
u
tes
o
f
t
h
e
d
ata
s
et
ar
e
r
elate
d
to
cu
s
to
m
er
in
f
o
r
m
at
io
n
s
u
ch
a
s
ag
e,
j
o
b
,
m
a
r
ital
s
tatu
s
etc
an
d
r
elate
d
w
it
h
th
e
las
t
co
n
tact
o
f
th
e
cu
r
r
en
t
ca
m
p
ai
g
n
s
u
c
h
as
co
n
tact
(
co
m
m
u
n
icatio
n
t
y
p
e)
,
d
u
r
atio
n
(
las
t
co
n
tact
d
u
r
atio
n
,
i
n
s
ec
o
n
d
s
)
a
n
d
ca
m
p
aig
n
(
n
u
m
b
er
o
f
co
n
ta
cts
m
ad
e
d
u
r
i
n
g
t
h
e
ca
m
p
ai
g
n
f
o
r
th
e
clie
n
t
)
.
B
ay
e
s
ian
clas
s
i
f
icatio
n
m
et
h
o
d
is
i
m
p
le
m
en
ted
to
an
a
l
y
ze
t
h
is
d
ataset.
T
h
e
cla
s
s
i
f
ier
w
ill
p
r
ed
ict
th
e
cu
s
to
m
er
s
b
elo
n
g
s
to
w
h
ic
h
c
lass
t
h
at
s
h
o
u
ld
h
a
v
e
h
ig
h
es
t
p
o
s
ter
io
r
p
r
o
b
ab
ilit
y
.
T
h
e
cu
s
to
m
er
in
f
o
r
m
at
io
n
ac
cu
m
u
lated
b
y
a
P
o
r
tu
g
u
ese
b
an
k
i
n
g
i
n
s
tit
u
tio
n
i
s
u
s
ed
to
id
en
tify
c
u
s
to
m
er
s
an
d
p
r
o
v
id
e
d
ec
is
io
n
s
u
p
p
o
r
t.
A
d
ata
m
o
d
el
i
s
g
e
n
er
ated
b
a
s
ed
u
p
o
n
t
h
e
h
i
s
to
r
y
o
f
th
e
cu
s
to
m
er
s
in
th
e
b
an
k
.
I
n
t
h
i
s
a
p
p
licatio
n
,
th
e
d
ataset
is
o
b
tain
ed
f
r
o
m
t
h
e
UC
I
m
ac
h
i
n
e
lear
n
i
n
g
r
ep
o
s
ito
r
y
(
h
ttp
://ar
ch
iv
e.
ic
s
.
u
ci.
e
d
u
/
m
l
/
)
.
A
i
m
o
f
t
h
e
class
i
f
icatio
n
is
to
p
r
ed
ict
if
th
e
clien
t
w
ill
s
u
b
s
cr
ib
e
a
ter
m
d
ep
o
s
it o
r
n
o
t
[
1
6
]
.
No
te:
Un
k
n
o
w
n
v
al
u
es a
r
e
o
m
itted
f
r
o
m
d
ataset.
B
an
k
s
h
a
v
e
n
u
m
er
o
u
s
i
n
d
iv
id
u
al
r
etail
c
u
s
to
m
er
s
.
T
h
e
y
u
s
e
s
C
R
M
b
ec
au
s
e
o
f
its
a
n
al
y
tic
al
ab
ilit
ies.
C
R
M
h
elp
s
th
e
b
a
n
k
s
to
in
cr
e
ase
th
e
cr
o
s
s
s
ell
p
er
f
o
r
m
a
n
ce
an
d
m
a
n
a
g
e
t
h
e
ch
u
r
n
r
ate
s
(
cu
s
to
m
er
d
ef
ec
tio
n
r
ates)
.
Data
Min
in
g
m
o
d
els
m
a
y
b
e
u
s
ed
to
d
ef
in
e
th
e
c
u
s
to
m
er
s
w
h
ich
ar
e
ea
g
er
to
co
n
f
i
r
m
cr
o
s
s
s
ell
o
f
f
er
s
,
w
h
ic
h
ar
e
ab
o
u
t to
b
e
lo
s
t a
n
d
w
h
at
ca
n
b
e
d
o
n
e
to
w
in
t
h
e
m
ag
ain
.
Mic
r
o
s
o
f
t
SQ
L
Ser
v
er
2
0
0
8
i
s
u
s
ed
as
d
atab
ase
m
a
n
a
g
e
m
e
n
t
s
y
s
te
m
.
T
h
e
Data
Min
i
n
g
o
p
er
atio
n
is
ap
p
lied
to
d
ata.
T
h
is
o
p
er
atio
n
is
co
n
v
er
ted
to
a
W
C
F ser
v
ic
e.
A
r
c
h
itect
u
r
al
d
esig
n
o
f
th
e
ap
p
licatio
n
is
as
in
th
e
F
ig
u
r
e
2
.
W
CF
s
er
v
ice
[
1
7
]
c
o
n
n
ec
ts
t
o
d
atab
ase
an
d
r
e
ad
s
d
ata
f
r
o
m
d
atab
ase
o
r
w
r
ite
s
in
to
th
e
d
atab
ase.
Fro
m
th
e
ap
p
licatio
n
,
W
C
F
s
er
v
ice
is
ca
lled
f
o
r
co
m
p
u
ti
n
g
t
h
e
B
a
y
esia
n
clas
s
if
ica
tio
n
.
P
h
o
n
e
g
ap
[
1
8
]
w
h
i
ch
is
a
“w
r
ite
o
n
ce
,
r
u
n
ev
e
r
y
w
h
er
e”
p
latf
o
r
m
co
n
n
ec
t
s
to
W
C
F
s
er
v
ices.
I
t
en
ab
les
to
r
u
n
th
e
ap
p
lica
tio
n
o
n
all
o
p
er
atin
g
s
y
s
te
m
s
lik
e
I
OS,
An
d
r
o
id
,
W
in
d
o
w
s
m
o
b
ile
etc.
.
.
Fig
u
r
e
2
.
A
r
ch
itectu
r
al
d
esi
g
n
T
h
e
ap
p
licatio
n
is
d
ev
elo
p
ed
o
n
An
d
r
o
id
p
latf
o
r
m
(
An
d
r
o
id
4
.
1
.
2
)
u
s
in
g
P
h
o
n
eGa
p
(
v
er
s
io
n
2
.
1
.
0
)
tech
n
o
lo
g
y
to
co
n
n
ec
t
to
th
e
s
er
v
ice
an
d
r
et
u
r
n
t
h
e
r
es
u
lt.
P
h
o
n
eGa
p
is
an
o
p
en
-
s
o
u
r
ce
d
ev
elo
p
m
e
n
t
to
o
l
f
o
r
m
o
b
ile
cr
o
s
s
-
p
lat
f
o
r
m
A
p
p
p
u
b
licatio
n
t
h
at
u
s
e
s
d
ev
ice
-
a
g
n
o
s
tic
w
r
ap
p
er
s
lik
e
HT
ML
,
J
av
ascr
ip
t,
a
n
d
C
SS
,
th
at
ca
n
b
e
r
ap
id
l
y
d
ep
lo
y
ed
o
n
A
n
d
r
o
id
,
B
lack
b
er
r
y
,
a
n
d
iP
h
o
n
e
p
lat
f
o
r
m
s
.
B
ay
e
s
ian
ca
lcu
latio
n
i
s
d
o
n
e
b
y
u
s
in
g
t
h
e
d
ataset
an
d
t
h
e
p
r
o
b
ab
ili
ti
es
o
f
ea
ch
v
ar
iab
les
ar
e
w
r
itte
n
in
to
d
atab
ase
w
h
e
n
th
e
m
et
h
o
d
B
ay
esia
n
C
alcu
latio
n
(
)
is
ca
l
led
f
r
o
m
W
C
F ser
v
ice.
A
t
f
ir
s
t,
m
o
d
el
i
s
co
n
s
tr
u
cted
w
it
h
tr
ain
in
g
s
et
a
n
d
test
ed
w
it
h
tes
t
s
et.
Fi
g
u
r
e
3
s
h
o
w
s
th
e
m
o
d
e
l
co
n
s
tr
u
ct
io
n
d
iag
r
a
m
.
I
n
tr
ai
n
in
g
s
e
t,
o
u
tp
u
t
cla
s
s
e
s
o
f
s
a
m
p
les
ar
e
k
n
o
w
n
.
T
h
er
e
ar
e
ca
teg
o
r
ical
an
d
co
n
tin
u
o
u
s
attr
ib
u
te
s
in
d
atase
t.
Fig
u
r
e
3
in
clu
d
es
s
o
m
e
o
f
t
h
e
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8776
IJ
-
I
C
T
Vo
l.
3
,
No
.
2
,
J
u
n
e
20
1
4
:
88
–
96
92
Fig
u
r
e
3
.
Mo
d
el
co
n
s
tr
u
ctio
n
T
ab
le
1
.
P
r
o
b
ab
ilit
ies o
f
ca
teg
o
r
ical
attr
ib
u
tes
T
ab
le
2
.
Me
an
a
n
d
v
ar
i
an
ce
o
f
co
n
ti
n
u
o
u
s
attr
ib
u
tes
So
m
e
o
f
t
h
e
i
n
p
u
t
s
s
u
ch
a
s
ag
e,
b
alan
ce
,
d
u
r
atio
n
a
n
d
ca
m
p
aig
n
h
av
e
c
o
n
tin
u
o
u
s
v
a
l
u
es
i
n
t
h
e
d
ataset.
So
t
h
e
y
ar
e
h
a
n
d
led
b
y
Ga
u
s
s
ia
n
d
is
tr
ib
u
tio
n
.
P
r
o
b
ab
ilit
ies
o
f
all
ca
te
g
o
r
ical
attr
ib
u
tes
w
h
e
n
o
u
tp
u
t
class
y
is
„
No
‟
an
d
o
u
tp
u
t
cla
s
s
y
is
„
Ye
s
‟
ar
e
ca
lc
u
lated
a
n
d
s
h
o
w
n
i
n
T
ab
le
1
f
o
r
th
e
d
ataset
u
s
ed
i
n
t
h
e
ap
p
licatio
n
.
Me
an
a
n
d
v
ar
ian
c
e
o
f
ea
ch
co
n
tin
u
o
u
s
attr
ib
u
te
s
w
h
e
n
o
u
tp
u
t c
lass
y
i
s
„
No
‟
a
n
d
o
u
tp
u
t c
la
s
s
y
i
s
„
Yes
‟
ar
e
ca
l
cu
la
ted
an
d
s
h
o
wn
in
T
ab
le
2.
As an
e
x
a
m
p
le,
f
o
llo
w
in
g
s
a
m
p
le
cu
s
to
m
er
d
ata
is
a
n
ev
id
e
n
ce
f
o
r
B
ay
es
ian
C
las
s
i
f
ier
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
I
C
T
I
SS
N:
2252
-
8776
Dec
is
io
n
S
u
p
p
o
r
t S
ystem
F
o
r
A
C
u
s
to
mer …
(
Özg
e
K
a
r
t
)
93
T
ab
le
3
.
E
x
am
p
le
c
u
s
to
m
er
d
ata
P
r
o
b
ab
ilit
ies
o
f
ea
ch
clas
s
es
f
o
r
g
iv
e
n
s
a
m
p
le
cu
s
to
m
er
d
ata
(
s
h
o
w
n
i
n
T
ab
le
3
)
an
d
Gau
s
s
ia
n
d
is
tr
ib
u
tio
n
s
f
o
r
co
n
ti
n
u
o
u
s
attr
ib
u
tes
an
d
ea
c
h
g
e
n
er
al
class
p
r
o
b
ab
ilit
ies
(
P
(
y
es)
a
n
d
P
(
n
o
)
)
ar
e
all
m
u
ltip
lied
f
o
r
f
i
n
d
in
g
t
h
e
p
r
o
b
ab
le
class
o
f
g
i
v
en
c
u
s
to
m
er
.
A
ll
v
alu
e
s
an
d
r
es
u
lt
o
f
m
u
l
t
i
p
licatio
n
ar
e
s
h
o
wn
in
T
ab
le
4
.
A
cc
o
r
d
in
g
to
th
e
ta
b
le,
P
r
o
b
ab
ilit
y
o
f
No
is
h
i
g
h
e
r
th
an
P
r
o
b
ab
ilit
y
o
f
Yes.
T
ab
le
4
.
T
h
e
ter
m
s
f
o
r
ca
lcu
l
atin
g
p
er
ce
n
tag
e
o
f
r
esu
lts
T
h
e
p
er
ce
n
tag
e
o
f
r
esu
l
ts
m
a
y
b
e
ca
lcu
lated
as f
o
llo
w
i
n
g
:
(
X
*
1
0
0
)
/(
X+
Y)
W
h
en
it i
s
ap
p
lied
to
th
e
s
a
m
p
le
d
ata:
(
3
,
8
5
4
5
1
E
-
1
4
*
1
0
0
)
/ (
3
,
8
5
4
5
1
E
-
1
4
+
8
,
0
1
8
7
8
E
-
1
6
)
A
cc
o
r
d
in
g
to
r
es
u
lt
o
f
th
e
c
alcu
latio
n
it
ca
n
b
e
ex
p
r
ess
e
d
th
at
t
h
e
o
u
tp
u
t
clas
s
o
f
t
h
e
s
a
m
p
le
cu
s
to
m
er
is
n
o
w
it
h
9
7
% p
r
o
b
ab
ilit
y
.
W
h
en
t
h
e
ap
p
lic
atio
n
i
s
r
u
n
,
f
o
llo
w
in
g
s
cr
ee
n
s
h
o
w
n
in
Fig
u
r
e
4
ap
p
ea
r
s
an
d
u
s
er
en
ter
s
t
h
e
p
ar
am
eter
s
.
Fig
u
r
e
4
.
A
p
p
licatio
n
‟
s
I
n
itial
Scr
ee
n
Fig
u
r
e
5
.
R
esu
l
t o
f
t
h
e
class
i
f
icatio
n
an
d
p
r
o
b
ab
ilit
y
o
f
th
e
r
es
u
lt
Af
ter
clic
k
i
n
g
t
h
e
S
h
o
w
R
e
s
u
lt
b
u
tto
n
,
p
r
o
g
r
a
m
e
v
al
u
ates
t
h
e
v
al
u
e
s
en
ter
ed
b
y
u
s
er
an
d
s
h
o
w
s
t
h
e
class
i
f
icatio
n
r
es
u
lt.
“Cl
as
s
i
f
y
”
m
et
h
o
d
is
ca
lled
f
r
o
m
W
C
F
w
h
e
n
t
h
e
b
u
t
to
n
i
s
clic
k
ed
.
An
d
th
e
s
er
v
ice
r
etu
r
n
s
t
h
e
r
es
u
lt.
Fi
g
u
r
e
5
s
h
o
w
s
th
e
r
es
u
lt:
4.
RE
SU
L
T
S
A
ND
AN
AL
Y
SI
S
A
cc
u
r
ac
y
r
ate
o
f
th
e
m
o
d
el
u
s
ed
in
th
i
s
ap
p
licatio
n
is
co
m
p
u
ted
b
y
C
r
o
s
s
Valid
atio
n
.
C
r
o
s
s
v
alid
atio
n
allo
w
s
u
s
i
n
g
t
h
e
wh
o
le
d
ataset
i
n
co
m
p
u
ti
n
g
.
T
h
e
d
ataset
i
s
d
i
v
id
ed
i
n
to
t
w
o
p
ar
ts
r
an
d
o
m
l
y
.
Firs
t
p
ar
t
is
u
s
ed
f
o
r
m
o
d
el
co
n
s
tr
u
ctio
n
.
T
h
e
m
o
d
el
is
te
s
ted
w
i
t
h
s
ec
o
n
d
p
ar
t
o
f
d
ataset
an
d
th
e
a
cc
u
r
ac
y
r
ate
i
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8776
IJ
-
I
C
T
Vo
l.
3
,
No
.
2
,
J
u
n
e
20
1
4
:
88
–
96
94
co
m
p
u
ted
.
Af
ter
t
h
at,
a
m
o
d
e
l
is
co
n
s
tr
u
cted
w
it
h
s
ec
o
n
d
p
ar
t
an
d
test
ed
w
i
th
f
ir
s
t
p
ar
t
.
A
cc
u
r
ac
y
r
ate
is
co
m
p
u
ted
.
Fi
n
all
y
,
t
h
e
m
o
d
el
is
co
n
s
tr
u
cted
u
s
i
n
g
w
h
o
le
d
ataset.
Av
er
ag
e
o
f
t
h
e
co
m
p
u
te
d
ac
cu
r
ac
y
r
ates
i
s
th
e
ac
cu
r
ac
y
r
ate
o
f
th
e
c
o
n
s
tr
u
cted
m
o
d
el.
Fig
u
r
e
6
s
h
o
w
s
t
h
e
C
r
o
s
s
Va
lid
atio
n
:
Fig
u
r
e
6
.
C
r
o
s
s
Valid
atio
n
T
ab
le
5
.
First cr
o
s
s
v
alid
atio
n
r
esu
lts
I
n
th
i
s
s
tu
d
y
,
th
e
f
ir
s
t
h
al
f
o
f
d
ataset
is
u
s
ed
to
tr
ai
n
an
d
s
e
co
n
d
h
al
f
is
u
s
ed
to
te
s
t.
Af
te
r
th
at,
th
e
s
ec
o
n
d
h
al
f
o
f
d
atase
t
is
u
s
ed
to
tr
ai
n
an
d
f
ir
s
t
h
al
f
i
s
u
s
ed
to
test
th
e
al
g
o
r
ith
m
.
C
r
o
s
s
v
alid
ati
o
n
s
h
o
w
ed
t
h
at
t
h
e
ac
cu
r
ac
y
r
ate
o
f
t
h
e
m
o
d
el
is
8
4
%.
First
test
r
es
u
lt
s
ar
e
s
h
o
w
ed
i
n
T
ab
le
5
.
T
h
e
f
ir
s
t
p
ar
t
o
f
d
ataset
is
u
s
ed
f
o
r
tes
tin
g
.
T
h
er
e
ar
e
9
0
6
s
a
m
p
les
w
h
i
c
h
b
elo
n
g
to
ac
t
u
al
clas
s
v
alu
e
„
y
e
s
‟
.
Af
ter
t
h
e
test
,
5
5
5
o
f
t
h
e
m
ar
e
cla
s
s
i
f
ie
d
as
„
y
e
s
‟
w
h
ic
h
i
s
th
e
r
ig
h
t
r
es
u
lt
an
d
3
5
1
o
f
th
em
ar
e
class
i
f
ied
as
„
n
o
‟
.
T
h
er
e
ar
e
also
1
4
5
4
5
s
a
m
p
les
w
h
i
ch
b
elo
n
g
to
ac
tu
a
l
class
v
a
lu
e
„
n
o
.
Af
ter
th
e
test
,
1
3
5
4
0
o
f
th
e
m
ar
e
clas
s
i
f
ied
as
„
n
o
‟
w
h
ich
i
s
th
e
r
i
g
h
t
r
es
u
lt
an
d
1
0
0
5
o
f
th
e
m
ar
e
class
if
ied
as „
y
es
‟
.
As a
r
esu
lt,
ac
c
u
r
ac
y
r
ate
o
f
f
i
r
s
t tr
ain
i
n
g
i
s
;
T
r
u
ly
cla
s
s
i
f
ied
s
a
m
p
les:
5
5
5
+1
3
5
4
0
=1
4
0
9
5
A
ll
s
a
m
p
le
s
in
d
ata
s
et:
1
5
4
5
1
A
cc
u
r
ac
y
r
ate:
(
1
4
0
9
5
*
1
0
0
)
/ 1
5
4
5
1
~=
9
1
%
Seco
n
d
test
r
es
u
lts
ar
e
s
h
o
w
e
d
in
T
ab
le
6
.
T
h
e
s
ec
o
n
d
p
ar
t
o
f
d
ataset
is
u
s
ed
f
o
r
test
i
n
g
.
T
h
er
e
ar
e
3
6
0
6
s
am
p
les
w
h
ic
h
b
elo
n
g
t
o
ac
tu
al
class
v
al
u
e
„
y
e
s
‟
.
Af
ter
th
e
test
,
5
5
2
o
f
th
e
m
ar
e
class
i
f
ied
as
„
y
e
s
‟
w
h
ic
h
is
t
h
e
r
ig
h
t
r
esu
lt
an
d
3
0
5
4
o
f
th
e
m
ar
e
cla
s
s
i
f
ied
as
„
n
o
‟
.
T
h
er
e
ar
e
also
1
1
8
4
6
s
a
m
p
les
w
h
ic
h
b
elo
n
g
to
ac
tu
al
clas
s
v
alu
e
„
n
o
.
Af
te
r
th
e
test
,
1
1
5
1
4
o
f
t
h
e
m
ar
e
c
lass
i
f
ied
a
s
„
n
o
‟
w
h
ic
h
i
s
t
h
e
r
ig
h
t
r
es
u
lt
an
d
3
3
2
o
f
th
e
m
ar
e
class
i
f
ied
as
„
y
es
‟
.
As a
r
esu
lt,
ac
c
u
r
ac
y
r
ate
o
f
s
e
co
n
d
tr
ain
in
g
is
;
T
r
u
ly
cla
s
s
i
f
ied
s
a
m
p
les:
5
5
2
+1
1
5
1
4
=1
2
0
66
A
ll
s
a
m
p
le
s
in
d
ata
s
et:
1
5
4
5
2
A
cc
u
r
ac
y
r
ate:
(
1
2
0
6
6
*
1
0
0
)
/
1
5
4
5
2
~=
7
8
%
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
I
C
T
I
SS
N:
2252
-
8776
Dec
is
io
n
S
u
p
p
o
r
t S
ystem
F
o
r
A
C
u
s
to
mer …
(
Özg
e
K
a
r
t
)
95
T
ab
le
6
.
Seco
n
d
cr
o
s
s
v
alid
ati
o
n
r
esu
lt
s
Av
er
ag
e
ac
c
u
r
ac
y
r
ates o
f
f
ir
s
t
an
d
s
ec
o
n
d
test
s
g
i
v
e
th
e
ac
c
u
r
ac
y
o
f
th
e
i
m
p
le
m
en
tatio
n
w
h
ich
is
8
4
,
5
%.
5.
CO
NCLU
SI
O
N
T
o
d
ay
i
n
f
o
r
m
atio
n
i
s
v
i
tal
f
o
r
co
m
p
a
n
ies
a
n
d
C
u
s
to
m
er
R
ela
tio
n
s
h
ip
Ma
n
a
g
e
m
en
t
a
n
d
Dat
a
Min
i
n
g
tech
n
iq
u
es
a
i
m
to
d
eter
m
i
n
e
th
e
c
u
s
to
m
er
s
w
it
h
h
i
g
h
p
r
o
f
itab
ilit
y
.
T
h
e
i
n
f
o
r
m
atio
n
o
b
t
ain
ed
b
y
ap
p
l
y
i
n
g
s
o
p
h
is
ticated
Data
m
in
i
n
g
te
ch
n
iq
u
es
to
C
R
M
p
r
o
b
lem
s
h
av
e
a
s
tr
ate
g
ic
i
m
p
o
r
tan
ce
f
o
r
co
m
p
a
n
ies.
A
n
i
m
p
o
r
tan
t c
o
m
p
etit
iv
e
ad
v
a
n
ta
g
e
is
e
v
id
en
t a
s
i
m
p
licat
io
n
o
f
u
s
i
n
g
s
u
ch
s
y
s
te
m
s
at
t
h
e
d
ec
is
io
n
m
ak
i
n
g
lev
e
l.
I
n
th
i
s
p
ap
er
w
e
p
r
esen
t
an
an
al
y
s
is
o
f
a
m
a
s
s
i
v
e
v
o
lu
m
e
o
f
c
u
s
to
m
er
d
ata
w
h
ich
is
af
ter
w
ar
d
s
class
i
f
ie
d
b
ased
o
n
th
e
c
u
s
to
m
er
b
eh
av
io
r
s
.
Naiv
e
B
a
y
e
s
ia
n
class
i
f
icatio
n
is
u
s
ed
as
cla
s
s
i
f
ier
to
p
r
ed
ict
if
th
e
clien
t
w
ill
s
u
b
s
cr
ib
e
a
ter
m
d
e
p
o
s
it.
T
h
e
class
i
f
ier
r
ep
o
r
ted
h
ig
h
l
y
ac
ce
p
tab
le
ac
c
u
r
ac
y
li
k
e
8
4
.
5
%
f
o
r
all
t
h
e
test
ed
d
ata
b
y
d
o
in
g
cr
o
s
s
v
ali
d
atio
n
.
T
h
e
ac
cu
r
ac
y
i
s
d
eter
m
i
n
ed
as
t
h
e
p
er
ce
n
ta
g
e
o
f
t
h
e
co
r
r
ec
tl
y
cla
s
s
i
f
ied
in
s
ta
n
ce
s
f
r
o
m
t
h
e
test
s
et.
I
n
o
t
h
er
w
o
r
d
s
,
clas
s
i
f
icatio
n
ce
n
ter
s
ar
o
u
n
d
ex
p
lo
r
in
g
th
r
o
u
g
h
d
ata
o
b
j
ec
ts
(
tr
ain
in
g
s
et)
to
f
in
d
a
s
et
o
f
r
u
les
w
h
ich
d
eter
m
i
n
e
t
h
e
cla
s
s
o
f
ea
c
h
o
b
j
ec
t
ac
co
r
d
in
g
to
its
attr
ib
u
tes.
Sin
ce
th
e
ac
cu
r
ac
y
o
f
th
e
m
o
d
el
is
a
cc
ep
tab
le,
th
e
m
o
d
el
ca
n
b
e
u
s
ed
to
class
if
y
d
ata
tu
p
le
s
w
h
o
s
e
class
lab
els
ar
e
n
o
t
k
n
o
w
n
.
T
h
e
class
if
ica
tio
n
r
u
les
ca
n
b
e
u
s
ed
to
s
u
p
p
o
r
t
d
ec
is
io
n
m
a
k
in
g
f
o
r
ac
h
iev
i
n
g
a
g
o
o
d
C
R
M
f
o
r
b
u
s
i
n
ess
e
s
.
I
n
th
i
s
p
ap
er
,
th
e
d
ata
o
b
tain
ed
f
r
o
m
a
b
an
k
is
a
n
al
y
ze
d
.
B
ay
e
s
ian
cla
s
s
i
f
icat
io
n
is
ap
p
lied
to
th
e
d
ata.
B
ay
esia
n
cla
s
s
i
f
icat
io
n
is
i
m
p
le
m
en
ted
a
s
a
W
C
F
s
er
v
i
ce
.
Fro
m
a
n
An
d
r
o
id
d
ev
ice,
th
is
s
er
v
ice
i
s
ca
lle
d
an
d
r
esu
lt o
f
class
if
ica
tio
n
o
f
a
n
e
w
cu
s
to
m
er
d
ata
is
s
h
o
w
e
d
.
Data
s
et
i
s
s
to
r
ed
i
n
Mic
r
o
s
o
f
t
SQ
L
Ser
v
er
2
0
0
8
.
B
ay
esia
n
class
i
f
icatio
n
m
et
h
o
d
is
i
m
p
le
m
en
ted
o
n
a
W
C
F
(
W
in
d
o
w
s
C
o
m
m
u
n
i
ca
tio
n
Fo
u
n
d
atio
n
)
s
er
v
ice
p
r
o
j
ec
t
u
s
in
g
Mic
r
o
s
o
f
t
V
is
u
al
Stu
d
io
2
0
1
0
.
T
h
e
r
esu
lt o
f
t
h
e
i
m
p
le
m
en
tat
io
n
is
p
r
es
en
ted
o
n
An
d
r
o
id
p
latf
o
r
m
u
s
i
n
g
P
h
o
n
e
g
ap
tech
n
o
lo
g
y
.
T
h
e
d
ev
elo
p
ed
s
er
v
ice
ca
n
h
a
n
d
le
d
if
f
er
en
t
d
ataset
s
w
it
h
d
if
f
er
en
t
n
u
m
b
er
o
f
attr
ib
u
te
s
.
Her
eb
y
t
h
i
s
m
o
d
el
ca
n
also
b
e
u
s
ed
f
o
r
o
th
er
ty
p
e
o
f
co
m
p
an
ie
s
to
p
r
ed
ict
th
eir
cu
s
to
m
er
b
eh
av
io
r
s
w
ith
s
p
ec
i
f
ic
h
is
to
r
ical
d
ata.
I
n
th
is
ap
p
licatio
n
,
th
e
b
an
k
ca
n
p
r
ed
ict
if
a
cu
s
to
m
er
w
ill
s
u
b
s
cr
ib
e
a
ter
m
d
ep
o
s
it
o
r
n
o
t.
I
t
ca
n
m
an
a
g
e
d
ir
ec
t
m
ar
k
eti
n
g
ca
m
p
ai
g
n
s
u
s
i
n
g
t
h
is
p
r
ed
ic
tio
n
.
T
h
e
r
esu
lts
ca
n
b
e
ac
c
ess
ed
an
y
ti
m
e
a
n
d
an
y
w
h
er
e
t
h
r
o
u
g
h
a
m
o
b
ile
d
e
v
ice.
W
ith
ap
p
licatio
n
o
f
C
R
M
in
m
o
b
ile
p
lat
f
o
r
m
,
ch
a
n
g
es
an
d
u
p
d
ates
ca
n
b
e
d
o
n
e
s
ea
m
le
s
s
l
y
f
r
o
m
an
y
w
h
er
e
an
d
a
n
y
t
i
m
e
a
n
d
w
it
h
n
o
d
ela
y
.
RE
F
E
R
E
NC
E
S
[1
]
G
a
o
,
Hu
a
;
,
“
Cu
st
o
m
e
r
Re
latio
n
sh
ip
M
a
n
a
g
e
m
e
n
t
Ba
se
d
o
n
Da
ta M
in
i
n
g
T
e
c
h
n
iq
u
e
,
”
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
E
-
Bu
si
n
e
s
s
a
n
d
E
–
Go
v
e
rn
me
n
t
,
1
–
4,
2
0
1
1
[2
]
P
a
d
m
a
n
a
b
h
a
n
,
B.
;
T
u
z
h
il
i
n
,
A
.
;
,
“
On
th
e
Us
e
o
f
Op
ti
m
iz
a
ti
o
n
f
o
r
Da
ta M
in
in
g
:
T
h
e
o
re
ti
c
a
l
In
tera
c
t
io
n
s an
d
e
CRM
Op
p
o
r
tu
n
it
ies
[
J
]
,
”
M
a
n
a
g
e
me
n
t
S
c
ien
c
e
,
49
,
1
0
,
2
7
1
–
3
4
3
,
2
0
0
3
[3
]
P
e
ters
e
n
,
G
.
;
,
“
C
u
sto
m
e
r
Re
latio
n
sh
ip
M
a
n
a
g
e
m
e
n
t
In
ROI:
Bu
il
d
in
g
th
e
CRM
Bu
sin
e
ss
Ca
se
,
”
X
l
i
b
ris,
2
0
0
3
[4
]
Ha
n
,
J.;
Ka
m
b
e
r,
M
.
;
,
“
Da
ta
M
in
i
n
g
:
C
o
n
c
e
p
ts
a
n
d
T
e
c
h
n
i
q
u
e
s
[M
]
,
”
CA
:
M
o
rg
a
n
Ka
u
f
m
a
n
n
P
u
b
l
ish
e
rs.
S
a
n
F
ra
n
c
isc
o
,
2
0
0
1
[5
]
Ku
y
k
e
n
d
a
ll
,
L
.
;
,
“
T
h
e
d
a
ta
-
m
in
in
g
to
o
l
b
o
x
[
J
]
,
”
Cre
d
it
C
a
rd
M
a
n
a
g
e
me
n
t
,
1
2
,
6
,
30
–
4
0
,
1
9
9
9
[6
]
He
Z.
;
X
u
X
.
;
Hu
a
n
g
J.Z
.
;
De
n
g
S
.
;
,
“
M
in
in
g
c
las
s
o
u
tl
iers
:
c
o
n
c
e
p
ts,
a
lg
o
rit
h
m
s
a
n
d
a
p
p
li
c
a
ti
o
n
s
in
CRM
,”
Exp
e
rt S
y
ste
ms
wit
h
A
p
p
l
ica
ti
o
n
s
,
2
7
,
6
8
1
–
6
9
7
,
2
0
0
4
[7
]
Hu
n
g
S
.
Y.;
Ye
n
D.C.
;
Hs
i
u
-
Yu
W
a
n
g
;
,
“
A
p
p
l
y
in
g
d
a
ta
m
in
in
g
to
tele
c
o
m
c
h
u
rn
m
a
n
a
g
e
m
e
n
t,
”
Exp
e
rt
S
y
ste
ms
wit
h
A
p
p
l
ica
ti
o
n
s
3
1
,
5
1
5
–
5
2
4
,
2
0
0
6
[8
]
Ho
ss
e
in
i
S
.
M
.
S
.
;
M
a
lek
i
A
.
;
G
h
o
lam
ian
M
.
R.
;
,
“
Clu
ste
r
a
n
a
ly
si
s
u
sin
g
d
a
ta
m
in
in
g
a
p
p
r
o
a
c
h
to
d
e
v
e
lo
p
CR
M
m
e
th
o
d
o
lo
g
y
to
a
ss
e
ss
th
e
c
u
sto
m
e
r
lo
y
a
lt
y
,”
Exp
e
rt S
y
ste
ms
wi
th
Ap
p
li
c
a
ti
o
n
s
,
3
7
,
5
2
5
9
–
5
2
6
4
,
2
0
1
0
[9
]
Ch
e
n
R.
S
.
;
W
u
R.
C.
;
Ch
e
n
J.Y.;
,
“
Da
ta M
in
in
g
A
p
p
li
c
a
ti
o
n
i
n
Cu
s
to
m
e
r
Re
l
a
ti
o
n
sh
i
p
M
a
n
a
g
e
m
e
n
t
Of
Cr
e
d
it
Ca
rd
Bu
sin
e
ss
,
”
In
ter
n
a
ti
o
n
a
l
Co
m
p
u
te
r S
o
ft
w
a
re
a
n
d
Ap
p
li
c
a
ti
o
n
s C
o
n
f
e
re
n
c
e
on
,
2
0
0
5
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8776
IJ
-
I
C
T
Vo
l.
3
,
No
.
2
,
J
u
n
e
20
1
4
:
88
–
96
96
[1
0
]
S
u
a
C.
T
.
;
Ch
e
n
Y.H.;
S
h
a
D.Y.;
,
“
L
in
k
in
g
in
n
o
v
a
ti
v
e
p
ro
d
u
c
t
d
e
v
e
lo
p
m
e
n
t
w
it
h
c
u
sto
m
e
r
k
n
o
wle
d
g
e
:
a
d
a
ta
-
m
in
in
g
a
p
p
ro
a
c
h
,”
T
e
c
h
n
o
v
a
ti
o
n
,
2
6
,
7
8
4
–
7
9
5
,
2
0
0
6
[1
1
]
Ca
rrier,
C.
G
.
;
P
o
v
e
l,
O.;
,
“
Ch
a
r
a
c
terisin
g
d
a
ta m
in
in
g
so
f
t
w
a
re
,
”
I
n
telli
g
e
n
t
Da
t
a
A
n
a
lys
is
,
7
,
1
8
1
–
192
,
2
0
0
3
[1
2
]
Ch
u
n
g
,
H
.
M
.
;
G
ra
y
,
P
.
;
,
“
Da
ta m
in
in
g
[
J
]
,
”
J
o
u
rn
a
l
o
f
M
IS
,
1
6
(
1
),
1
1
–
13
,
1
9
9
9
[1
3
]
S
w
i
f
t,
R.
S
.
;
,
“
A
c
c
e
l
e
ra
ti
n
g
c
u
sto
m
e
r
re
latio
n
sh
ip
s:
Us
in
g
CR
M
a
n
d
re
latio
n
sh
i
p
tec
h
n
o
lo
g
ies
,
”
Up
p
e
r
S
a
d
d
le
R
iv
e
r.
N.J.:
P
re
n
ti
c
e
Ha
ll
P
T
R.
,
2
0
0
1
[1
4
]
A
h
m
e
d
,
S
.
R.
;
,
“
A
p
p
li
c
a
ti
o
n
s
o
f
d
a
ta
m
in
in
g
in
re
t
a
il
b
u
sin
e
ss
,
”
In
fo
rm
a
t
io
n
T
e
c
h
n
o
l
o
g
y
:
Co
d
in
g
a
n
d
Co
m
p
u
t
in
g
,
2
,
4
5
5
–
4
5
9
,
2
0
0
4
[1
5
]
Ng
a
i
E.
W
.
T
.
;
X
iu
L
.
;
Ch
a
u
D.C.
K.;
,
“
A
p
p
li
c
a
ti
o
n
o
f
d
a
ta
m
in
in
g
tec
h
n
iq
u
e
s in
c
u
sto
m
e
r
re
latio
n
sh
ip
m
a
n
a
g
e
m
e
n
t:
A
li
ter
a
tu
re
re
v
ie
w
a
n
d
c
las
si
f
ica
t
io
n
,
”
Exp
e
rt S
y
ste
ms
wit
h
Ap
p
li
c
a
ti
o
n
s
,
3
6
,
2
5
9
2
–
2
6
0
2
,
2
0
0
9
[1
6
]
M
o
ro
,
S
.
;
L
a
u
re
a
n
o
,
R.
;
Co
rtez
,
P
.
;
,
“
Us
in
g
Da
ta
M
in
in
g
f
o
r
Ba
n
k
Dire
c
t
M
a
r
k
e
ti
n
g
:
A
n
A
p
p
li
c
a
t
io
n
o
f
th
e
CRIS
P
-
DM
M
e
t
h
o
d
o
l
o
g
y
,
”
Eu
ro
p
e
a
n
S
imu
la
ti
o
n
a
n
d
M
o
d
e
ll
i
n
g
C
o
n
fer
e
n
c
e
on
,
2
0
1
1
[1
7
]
L
ö
wy
,
J.;
,
“
W
CF
Esse
n
ti
a
ls
In
P
r
o
g
ra
m
m
in
g
W
CF
S
e
rv
ic
e
s (1
st.
e
d
.
)
(1
)
,
”
USA
:
O‟Re
il
l
y
,
2
0
0
7
[1
8
]
G
h
a
to
l,
R.
;
Pa
tel,
Y.;
,
“
Be
g
in
n
in
g
P
h
o
n
e
G
a
p
,”
A
p
re
ss
,
2
0
1
2
BI
O
G
RAP
H
I
E
S
O
F
AUTH
O
RS
Öz
g
e
Ka
rt
re
c
e
iv
e
d
h
e
r
m
a
ste
r
d
e
g
re
e
in
c
o
m
p
u
ter
e
n
g
in
e
e
rin
g
f
r
o
m
Do
k
u
z
E
y
lu
l
Un
iv
e
rsit
y
in
2
0
1
3
.
C
u
rre
n
tl
y
,
sh
e
is
a
re
se
a
rc
h
a
ss
istan
t
a
t
th
e
De
p
a
rtm
e
n
t
o
f
Co
m
p
u
ter
En
g
in
e
e
rin
g
,
Do
k
u
z
Ey
lu
l
Un
iv
e
rsit
y
in
T
u
rk
e
y
.
He
r
r
e
se
a
rc
h
in
tere
sts
in
c
lu
d
e
d
a
ta
m
in
in
g
in
larg
e
d
a
tab
a
se
s,
so
c
ial
m
e
d
ia m
in
in
g
,
d
e
c
isio
n
s
u
p
p
o
rt
s
y
ste
m
s
,
a
n
d
in
d
u
strial
a
p
p
li
c
a
ti
o
n
s
.
A
lp
Ku
t
is
a
f
u
ll
p
ro
f
e
ss
o
r
o
f
c
o
m
p
u
ter en
g
in
e
e
rin
g
a
t
Do
k
u
z
E
y
l
u
l
Un
iv
e
rsity
.
H
e
is
h
e
a
d
o
f
th
e
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter
En
g
in
e
e
rin
g
o
f
Do
k
u
z
E
y
lu
l
Un
iv
e
rsity
sin
c
e
th
e
f
a
ll
o
f
2
0
0
3
.
His
w
o
rk
s
in
c
lu
d
e
d
a
ta
m
in
in
g
in
d
a
t
a
b
a
se
s,
d
a
tab
a
se
m
a
n
a
g
e
m
e
n
t
s
y
s
tem
s
a
n
d
d
istri
b
u
te
d
sy
ste
m
s.
He
h
a
s
m
a
n
y
p
u
b
li
c
a
ti
o
n
s
o
n
a
v
a
riet
y
o
to
p
ics
,
in
c
l
u
d
i
n
g
,
w
e
b
-
b
a
se
d
sy
ste
m
s
a
n
d
p
a
ra
ll
e
l
s
y
ste
m
s.
Dr
.
V
lad
im
ir
Ra
d
e
v
s
k
i
is
A
ss
o
c
iate
P
ro
f
e
ss
o
r
a
t
th
e
F
a
c
u
lt
y
o
f
Co
n
tem
p
o
ra
r
y
S
c
ien
c
e
s
a
n
d
T
e
c
h
n
o
lo
g
ies
,
S
o
u
th
Eas
t
Eu
r
o
p
e
a
n
Un
iv
e
rsity
in
Teto
v
o
,
Re
p
u
b
li
c
o
f
M
a
c
e
d
o
n
ia,
a
n
d
d
irec
to
r
o
f
it
s
S
k
o
p
je
S
tu
d
y
Ce
n
ter.
He
re
c
e
iv
e
d
h
is
P
h
D
f
ro
m
Un
iv
e
rsit
y
o
f
P
a
r
i
s
1
3
,
F
ra
n
c
e
a
n
d
h
a
s
b
e
e
n
tea
c
h
in
g
a
t
se
v
e
r
a
l
in
tern
a
ti
o
n
a
l
p
ro
g
ra
m
s
a
n
d
u
n
iv
e
rsiti
e
s.
His
re
se
a
rc
h
in
tere
sts
a
r
e
in
th
e
d
o
m
a
in
s o
f
se
m
a
n
ti
c
w
e
b
a
n
d
a
rti
f
icia
l
in
telli
g
e
n
c
e
.
Evaluation Warning : The document was created with Spire.PDF for Python.