I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
p
ute
r
E
ng
in
ee
ring
(
I
J
E
CE
)
Vo
l.
8
,
No
.
2
,
A
p
r
il
201
8
,
p
p
.
1122
~
1
1
3
0
I
SS
N:
2
0
8
8
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v8
i
2
.
p
p
1
1
2
2
-
1130
1122
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
co
r
e
.
co
m/
jo
u
r
n
a
ls
/in
d
ex
.
p
h
p
/
I
JE
C
E
Ba
t
-
Cluste
r
:
A B
a
t
Alg
o
rith
m
-
ba
sed Au
to
m
a
t
ed
G
r
a
ph
Clustering
Appro
a
ch
Z
a
k
a
ria
B
o
ulo
ua
rd
1
,
A
m
i
ne
E
l H
a
dd
a
di
2
,
F
a
dw
a
B
o
uh
a
f
er
3
,
Ana
s
s
E
l H
a
dd
a
d
i
4
,
L
a
h
ce
n K
o
utt
i
5
,
B
er
na
rd
Do
us
s
et
6
1,
5
L
a
b
S
IV
,
F
a
c
u
lt
y
o
f
S
c
ien
c
e
s
,
Ib
n
Z
o
h
r
Un
iv
e
rsity
,
M
o
ro
c
c
o
2,
6
S
IG
,
IRIT
,
P
a
u
l
S
a
b
a
ti
e
r
U
n
iv
e
rsity
,
F
ra
n
c
e
3,
4
De
p
a
rtm
e
n
t
o
f
M
a
th
e
m
a
ti
c
s an
d
IT
,
ENS
A
,
M
o
h
a
m
e
d
1
st
Un
iv
e
rsity
,
M
o
ro
c
c
o
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Oct
9
,
2
0
1
7
R
ev
i
s
ed
J
an
3
0
,
2
0
1
8
A
cc
ep
ted
Feb
9
,
2
0
1
8
De
f
in
in
g
th
e
c
o
rre
c
t
n
u
m
b
e
r
o
f
c
lu
ste
rs
is
o
n
e
o
f
th
e
m
o
st
f
u
n
d
a
m
e
n
tal
tas
k
s
in
g
ra
p
h
c
lu
ste
rin
g
.
W
h
e
n
it
c
o
m
e
s
to
larg
e
g
ra
p
h
s,
th
is
tas
k
b
e
c
o
m
e
s
m
o
re
c
h
a
ll
e
n
g
in
g
b
e
c
a
u
se
o
f
th
e
lac
k
o
f
p
rio
r
in
f
o
rm
a
ti
o
n
.
T
h
is
p
a
p
e
r
p
re
se
n
ts
a
n
a
p
p
ro
a
c
h
t
o
so
lv
e
th
is
p
r
o
b
lem
b
a
se
d
o
n
th
e
Ba
t
A
lg
o
rit
h
m
,
o
n
e
o
f
th
e
m
o
st
p
ro
m
isin
g
s
w
a
r
m
in
telli
g
e
n
c
e
b
a
se
d
a
lg
o
rit
h
m
s.
W
e
c
h
o
se
to
c
a
ll
o
u
r
so
lu
ti
o
n
,
“
Ba
t
-
Clu
ste
r
(BC).
”
T
h
is
a
p
p
ro
a
c
h
a
ll
o
w
s
a
n
a
u
to
m
a
ti
o
n
o
f
g
ra
p
h
c
lu
ste
rin
g
b
a
se
d
o
n
a
b
a
lan
c
e
b
e
tw
e
e
n
g
lo
b
a
l
a
n
d
lo
c
a
l
se
a
rc
h
p
ro
c
e
ss
e
s.
T
h
e
sim
u
latio
n
o
f
f
o
u
r
b
e
n
c
h
m
a
rk
g
ra
p
h
s
o
f
d
if
fe
re
n
t
siz
e
s
sh
o
ws
th
a
t
o
u
r
p
ro
p
o
se
d
a
lg
o
rit
h
m
is
e
ff
icie
n
t
a
n
d
c
a
n
p
r
o
v
id
e
h
ig
h
e
r
p
re
c
isio
n
a
n
d
e
x
c
e
e
d
so
m
e
b
e
st
-
k
n
o
w
n
v
a
lu
e
s.
K
ey
w
o
r
d
:
Au
to
m
a
ted
clu
s
ter
in
g
B
at
a
lg
o
r
tih
m
B
at
-
clu
s
ter
L
ar
g
e
g
r
ap
h
s
S
w
ar
m
i
n
telli
g
e
n
ce
Co
p
y
rig
h
t
©
2
0
1
8
In
stit
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
Al
l
rig
h
ts
re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Z
a
k
ar
ia
B
o
u
lo
u
ar
d
,
L
a
v
SIV
,
Facu
lt
y
o
f
Sc
ien
ce
s
,
I
b
n
Z
o
h
r
U
n
iv
er
s
it
y
,
B
P
8
1
0
6
,
C
ite
Dak
h
la,
Ag
ad
ir
,
Mo
r
o
cc
o
.
E
m
ail:
zb
o
u
lo
u
ar
d
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
Gr
ap
h
s
en
ab
le
u
s
to
v
is
u
alize
co
n
n
ec
ted
d
ata
an
d
to
r
el
y
o
n
v
is
u
al
p
r
o
w
e
s
s
to
d
ec
ip
h
er
v
a
l
u
ab
le
an
d
i
m
p
o
r
tan
t
h
id
d
en
k
n
o
w
led
g
e,
w
h
ic
h
co
u
ld
b
e
u
s
ed
to
i
m
p
r
o
v
e
th
e
d
ec
is
io
n
-
m
a
k
i
n
g
p
r
o
ce
s
s
o
f
an
o
r
g
an
izatio
n
.
Vis
u
aliz
in
g
lar
g
e
g
r
ap
h
s
b
ased
o
n
t
h
e
co
n
t
in
u
o
u
s
l
y
g
r
o
w
i
n
g
a
m
o
u
n
t
o
f
av
ailab
le
d
ata
h
a
s
b
ec
o
m
e
a
v
er
y
co
m
p
le
x
tas
k
an
d
h
as
o
u
tp
ac
ed
th
e
h
u
m
a
n
‟
s
ab
ilit
y
to
p
r
o
ce
s
s
,
an
al
y
ze
,
v
is
u
alize
an
d
ev
e
n
,
u
n
d
er
s
ta
n
d
t
h
e
m
.
T
h
er
ef
o
r
e,
a
p
r
o
ce
s
s
o
f
r
ed
u
ci
n
g
lar
g
e
g
r
ap
h
s
in
to
s
m
aller
,
m
o
r
e
r
e
p
r
esen
tati
v
e
o
n
e
s
i
s
n
ee
d
ed
.
T
o
ad
d
r
ess
th
is
c
h
alle
n
g
e,
g
r
ap
h
clu
s
ter
i
n
g
h
as i
m
p
o
s
ed
its
elf
late
l
y
as a
p
r
o
m
i
s
in
g
r
esear
ch
ar
ea
.
Gr
ap
h
clu
s
ter
i
n
g
ca
n
b
e
d
ef
i
n
ed
as
t
h
e
p
r
o
b
le
m
o
f
co
llecti
n
g
s
i
m
ilar
n
o
d
es
i
n
to
s
a
m
e
g
r
o
u
p
s
ca
lled
“
cl
u
s
ter
s
.
”
I
t
i
s
a
w
id
el
y
k
n
o
w
n
tec
h
n
iq
u
e
w
it
h
ap
p
lica
tio
n
s
in
v
ar
io
u
s
f
ield
s
s
u
c
h
a
s
s
o
cial
m
ed
ia
[
1
]
,
W
eb
s
ea
r
ch
r
es
u
lt
s
o
p
ti
m
izatio
n
[
2
]
,
w
ir
ele
s
s
s
en
s
o
r
n
et
w
o
r
k
s
[
3
]
an
d
a
ls
o
in
b
io
ch
e
m
ical
n
eu
r
al
n
et
w
o
r
k
s
[
4
]
a
m
o
n
g
o
th
er
s
.
I
n
m
o
s
t
ca
s
es,
t
h
e
n
u
m
b
er
o
f
cl
u
s
ter
s
to
f
o
r
m
is
alr
ea
d
y
k
n
o
w
n
a
n
d
is
g
i
v
e
n
as
an
e
n
tr
y
to
t
h
e
clu
s
ter
i
n
g
a
lg
o
r
ith
m
.
Ho
w
ev
e
r
,
w
ith
t
h
e
p
r
ev
alen
ce
o
f
B
ig
Data
,
it
b
ec
am
e
h
ar
d
er
to
h
av
e
a
p
r
i
o
r
id
ea
o
n
th
e
n
u
m
b
er
o
f
clu
s
ter
s
.
T
h
i
s
also
ap
p
lies
to
th
e
lar
g
e
g
r
ap
h
clu
s
ter
in
g
w
h
er
e
t
h
e
d
ec
id
er
‟
s
v
i
s
u
al
p
r
o
w
es
s
i
s
n
o
t
s
u
f
f
icie
n
t
en
o
u
g
h
to
p
r
o
v
id
e
h
i
m
w
i
th
a
n
ap
p
r
o
x
i
m
ate
p
r
io
r
id
ea
o
n
th
e
ev
en
tu
al
n
u
m
b
er
o
f
clu
s
ter
s
.
T
h
er
ef
o
r
e,
it
b
ec
am
e
i
m
p
er
a
tiv
e
to
p
r
o
p
o
s
e
s
o
lu
tio
n
s
w
h
er
e
th
e
clu
s
ter
i
n
g
al
g
o
r
ith
m
ca
n
au
to
m
a
ticall
y
“g
u
e
s
s
”
th
e
co
r
r
ec
t
n
u
m
b
er
o
f
cl
u
s
ter
s
b
ef
o
r
e
p
r
o
ce
ed
in
g
w
it
h
th
e
cl
u
s
ter
in
g
o
p
er
atio
n
.
T
h
is
r
esear
ch
f
ield
,
ca
lled
“Au
to
m
atic
Gr
ap
h
C
l
u
s
ter
in
g
,
”
s
tar
ted
in
t
h
e
late
1
9
9
0
‟
s
b
u
t
co
u
ld
n
‟
t
b
lo
s
s
o
m
u
n
til
t
h
e
late
2
0
0
0
‟
s
ea
r
l
y
2
0
1
0
‟
s
w
it
h
th
e
in
tr
o
d
u
ctio
n
o
f
th
e
ar
tific
ial
i
n
telli
g
e
n
ce
co
n
ce
p
ts
s
u
c
h
as
n
atu
r
e
-
i
n
s
p
ir
ed
alg
o
r
ith
m
s
[
5
]
,
[
6
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
B
a
t
-
C
lu
s
ter:
A
B
a
t A
lg
o
r
ith
m
-
b
a
s
ed
A
u
to
ma
te
d
Gra
p
h
C
lu
s
teri
n
g
A
p
p
r
o
a
c
h
(
Za
ka
r
ia
B
o
u
l
o
u
a
r
d
)
1123
I
n
m
o
s
t
o
f
th
e
p
ap
er
s
in
th
e
li
ter
atu
r
e
r
elate
d
to
au
to
m
atic
g
r
ap
h
clu
s
ter
i
n
g
b
ased
o
n
n
at
u
r
e
-
i
n
s
p
ir
ed
alg
o
r
ith
m
s
,
th
e
m
ai
n
id
ea
is
t
o
d
ef
in
e
a
“
S
i
m
i
lar
it
y
Me
as
u
r
e,
”
th
en
s
et
t
h
e
cl
u
s
ter
s
ac
co
r
d
in
g
to
it.
Sev
er
al
p
ap
er
s
ad
o
p
ted
a
d
is
cr
ete
f
o
r
m
u
latio
n
o
f
th
is
a
n
d
p
r
o
p
o
s
ed
ad
ap
tatio
n
s
o
f
t
h
e
b
as
icall
y
co
n
ti
n
u
o
u
s
n
at
u
r
e
in
s
p
ir
ed
alg
o
r
it
h
m
s
to
s
o
lv
e
it.
I
n
t
h
e
p
r
esen
t
w
o
r
k
,
w
e
w
ill
p
r
o
ce
ed
d
if
f
er
en
tl
y
s
in
ce
w
e
w
il
l
ad
ap
t
t
h
e
g
r
ap
h
cl
u
s
ter
i
n
g
p
r
o
b
le
m
its
el
f
s
o
it
ca
n
b
e
r
ep
r
esen
te
d
as
a
co
n
ti
n
u
o
u
s
p
r
o
b
le
m
.
T
h
is
ad
ap
tatio
n
w
ill
b
e
u
s
i
n
g
“
B
at
-
C
l
u
s
ter
,
”
a
co
m
b
i
n
atio
n
b
et
w
ee
n
“
F
FDP
,
”
a
lar
g
e
g
r
ap
h
v
is
u
aliza
tio
n
a
lg
o
r
ith
m
w
e
d
ev
elo
p
ed
b
y
o
u
r
tea
m
[
7
]
,
an
d
“
B
at
A
l
g
o
r
ith
m
,
”
a
n
at
u
r
e
in
s
p
ir
ed
o
p
tim
izatio
n
al
g
o
r
ith
m
d
e
v
elo
p
ed
b
y
Xi
n
-
Sh
e
Yan
g
[
8
]
b
ased
o
n
th
e
b
eh
a
v
io
r
o
f
b
ats.
F
FDP
w
ill
s
et
an
eq
u
ilib
r
iu
m
p
o
s
i
tio
n
i
n
g
o
f
th
e
l
ar
g
e
g
r
ap
h
;
t
h
en
it
w
ill
p
r
o
v
i
d
e
th
e
n
o
d
es
f
in
a
l
p
o
s
itio
n
s
as
a
v
ec
to
r
o
f
co
o
r
d
i
n
ates.
B
at
alg
o
r
ith
m
w
ill
ta
k
e
th
is
v
ec
to
r
in
to
co
n
s
id
er
atio
n
an
d
tr
y
to
f
i
n
d
th
e
b
est clu
s
ter
i
n
g
co
n
f
ig
u
r
atio
n
p
o
s
s
ib
le.
Af
ter
r
e
v
ie
w
i
n
g
t
h
e
r
elate
d
wo
r
k
s
i
n
Sec
tio
n
2
,
w
e
w
ill
d
es
cr
ib
e,
in
Sect
io
n
3
,
t
h
e
s
i
m
ilar
it
y
m
ea
s
u
r
e
w
e
u
s
ed
as
an
o
b
j
ec
tiv
e
f
u
n
c
t
io
n
an
d
w
e
w
ill
d
escr
ib
e
h
o
w
it
w
ill
b
e
o
p
tim
ized
b
y
t
h
e
“B
at
-
C
l
u
s
ter
”
(
B
C
)
alg
o
r
ith
m
.
T
h
e
test
in
g
a
n
d
r
es
u
lts
o
f
t
h
e
clu
s
ter
i
n
g
p
r
o
v
id
ed
b
y
“
B
at
-
C
lu
s
ter
”
co
m
p
ar
ed
w
i
th
o
t
h
er
well
-
k
n
o
w
n
s
o
lu
tio
n
s
,
s
u
ch
a
s
P
SO,
Dif
f
er
en
tial E
v
o
lu
tio
n
an
d
An
t Co
lo
n
y
Op
ti
m
izat
io
n
,
w
ill b
e
d
is
c
u
s
s
ed
in
Sec
tio
n
4
.
Sectio
n
5
co
n
clu
d
e
s
th
e
p
ap
er
an
d
p
r
esen
ts
a
n
id
ea
o
f
o
u
r
f
u
t
u
r
e
w
o
r
k
s
.
2.
RE
L
AT
E
D
WO
RK
S
I
n
t
h
is
s
ec
tio
n
,
w
e
w
ill
e
x
p
lo
r
e
s
o
m
e
o
f
th
e
m
o
s
t
i
m
p
o
r
tan
t
n
atu
r
e
in
s
p
ir
ed
s
o
lu
tio
n
s
u
s
ed
to
an
s
w
er
th
e
is
s
u
e
o
f
au
to
m
ated
g
r
ap
h
c
lu
s
ter
i
n
g
b
ef
o
r
e
m
o
v
i
n
g
to
in
tr
o
d
u
cin
g
B
at
-
C
lu
s
ter
in
Sec
tio
n
3
.
2
.1
.
P
a
rt
icle
Sw
a
rm
O
pti
m
i
za
t
io
n
T
h
e
liter
atu
r
e
co
n
tain
s
s
ev
er
a
l
ap
p
r
o
ac
h
es
to
u
s
in
g
P
SO
i
n
g
r
ap
h
cl
u
s
ter
i
n
g
,
o
f
ten
r
ef
e
r
r
ed
to
as
“Co
m
m
u
n
i
t
y
Dete
ctio
n
.
”
Mo
s
t
o
f
th
ese
ap
p
r
o
ac
h
es
ar
e
b
ase
d
o
n
th
e
id
ea
o
f
ad
ap
tin
g
th
e
P
SO,
an
alg
o
r
ith
m
o
r
ig
in
all
y
d
esig
n
ed
to
s
o
l
v
e
co
n
tin
u
o
u
s
o
p
ti
m
izatio
n
p
r
o
b
le
m
s
s
o
th
at
it
w
o
u
ld
b
e
ab
l
e
to
s
o
lv
e
d
is
cr
ete
p
r
o
b
lem
s
.
C
a
i
et
al.
p
r
o
p
o
s
ed
in
[
9
]
an
d
[
1
0
]
an
alter
atio
n
o
f
th
e
d
ef
i
n
it
io
n
o
f
t
h
e
p
o
s
iti
o
n
an
d
th
e
v
elo
cit
y
ter
m
s
w
h
er
e
th
e
p
o
s
itio
n
v
ec
to
r
r
ep
r
esen
ts
a
p
ar
titi
o
n
o
f
a
s
i
g
n
ed
n
et
w
o
r
k
an
d
t
h
e
v
elo
cit
y
r
ep
r
ese
n
ts
a
n
ev
en
t
u
al
p
er
m
u
tatio
n
o
f
t
h
e
p
ar
titi
o
n
.
Su
g
a
n
t
h
i
a
n
d
R
aj
ag
o
p
alan
[
1
1
]
h
av
e
ap
p
lied
P
SO
in
its
co
n
ti
n
u
o
u
s
s
tate,
b
u
t
t
h
e
y
s
u
g
g
es
ted
u
s
i
n
g
a
m
u
ltip
le
p
o
p
u
lati
o
n
s
w
a
r
m
in
s
tead
o
f
u
s
i
n
g
t
h
e
s
ta
n
d
ar
d
P
SO
w
it
h
o
n
e
p
o
p
u
latio
n
.
R
ej
in
a
P
ar
v
in
an
d
Vasan
th
a
n
a
y
ak
i
[
1
2
]
u
s
ed
P
SO
to
p
r
ev
en
t
r
esid
u
al
n
o
d
es
in
w
ir
eles
s
s
en
s
o
r
n
et
w
o
r
k
s
(
n
o
d
es
th
a
t
d
o
n
‟
t
b
elo
n
g
to
an
y
cl
u
s
ter
)
.
T
h
eir
id
ea
h
as
b
ee
n
ap
p
lied
to
o
p
tim
ize
e
n
er
g
y
co
n
s
u
m
p
tio
n
,
th
r
o
u
g
h
p
u
t,
p
ac
k
et
d
eliv
er
y
r
atio
,
an
d
n
et
w
o
r
k
li
f
eti
m
e
o
f
t
h
e
w
ir
eles
s
s
en
s
o
r
n
et
w
o
r
k
s
.
2
.
2.
Ant
Co
lo
ny
O
pti
m
iza
t
io
n
Ma
n
d
ala
et
al.
[
1
3
]
p
r
o
p
o
s
ed
an
A
C
O
b
ased
tech
n
iq
u
e
f
o
r
g
r
ap
h
clu
s
ter
in
g
an
d
ap
p
lied
it
in
d
etec
tin
g
cu
s
to
m
er
co
m
m
u
n
it
i
es i
n
t
h
e
e
-
m
ar
k
eti
n
g
f
ield
.
J
i e
t a
l.
[
1
4
]
s
u
g
g
e
s
ted
a
s
o
l
u
tio
n
f
o
r
t
h
e
p
r
o
b
le
m
o
f
co
m
p
le
x
co
m
m
u
n
it
y
d
etec
t
io
n
in
lar
g
e
g
r
ap
h
s
b
ased
o
n
th
e
s
tr
ateg
y
o
f
an
t p
h
er
o
m
o
n
e
d
i
f
f
u
s
io
n
a
n
d
u
p
d
ate
to
s
ea
r
ch
f
o
r
an
o
p
ti
m
al
g
r
ap
h
p
ar
titi
o
n
in
g
.
Z
h
o
u
et
a
l.
[
1
5
]
f
o
llo
w
ed
a
s
i
m
ilar
p
r
o
ce
s
s
,
b
u
t
th
e
y
to
o
k
t
h
e
o
v
er
lap
p
in
g
i
s
s
u
e
o
f
t
h
e
lar
g
e
co
m
m
u
n
ities
in
to
co
n
s
id
er
atio
n
.
Mo
r
ad
i
an
d
R
o
s
ta
m
i
[
1
6
]
u
s
ed
AC
O
alo
n
g
w
it
h
f
ea
tu
r
e
s
elec
tio
n
to
d
ef
in
e
clu
s
ter
s
o
f
f
ea
t
u
r
es.
Gao
et
al.
[
1
7
]
p
r
o
p
o
s
ed
a
co
m
b
in
a
tio
n
b
et
w
ee
n
A
C
O
an
d
K
-
Me
a
n
s
a
s
a
s
o
l
u
tio
n
to
th
e
d
y
n
a
m
ic
lo
ca
tio
n
r
o
u
ti
n
g
p
r
o
b
lem
.
K
-
Me
a
n
s
is
u
s
ed
to
d
ef
in
e
t
h
e
lo
ca
tio
n
o
f
d
ep
o
ts
(
clu
s
ter
ce
n
ter
s
)
w
h
i
le
AC
O
is
u
tili
ze
d
to
h
a
n
d
le
th
e
VR
P
in
d
y
n
a
m
ic
e
n
v
ir
o
n
m
e
n
ts
.
2
.
3.
Dif
f
er
ent
ia
l Ev
o
lutio
n
P
ater
lin
i
et
al.
[
1
8
]
p
r
o
p
o
s
ed
a
d
ir
ec
t
ap
p
licatio
n
o
f
DE
to
s
o
lv
e
t
h
e
p
r
o
b
lem
o
f
g
r
ap
h
p
ar
titi
o
n
i
n
g
an
d
a
co
m
p
ar
ativ
e
s
t
u
d
y
w
it
h
th
e
Ge
n
etic
A
l
g
o
r
i
th
m
(
G
A
)
s
h
o
w
ed
th
at
DE
w
a
s
m
o
r
e
ef
f
ic
ien
t.
C
ai
et
a
l.
[
1
9
]
p
r
o
p
o
s
ed
an
ad
ap
tatio
n
o
f
D
E
in
s
p
ir
ed
b
y
t
h
e
i
m
itatio
n
o
f
t
h
e
p
h
e
n
o
m
e
n
o
n
o
f
s
o
cial
lear
n
i
n
g
i
n
a
n
i
m
al
s
o
cieti
es.
T
h
e
y
i
m
p
r
o
v
ed
th
e
tr
ad
itio
n
al
DE
b
y
i
n
tr
o
d
u
cin
g
t
h
e
s
tr
ate
g
ic
A
S
L
s
ele
ctio
n
.
I
t
allo
w
s
t
h
e
alg
o
r
ith
m
to
r
el
y
o
n
t
h
e
i
n
f
o
r
m
atio
n
e
x
tr
ac
ted
f
r
o
m
t
h
e
n
ei
g
h
b
o
r
h
o
o
d
r
elatio
n
s
h
ip
s
o
f
it
s
p
o
p
u
latio
n
in
d
iv
id
u
als
to
g
u
id
e
th
e
s
elec
t
io
n
o
f
t
h
e
eli
g
ib
le
p
ar
en
ts
f
o
r
th
e
cr
o
s
s
o
v
er
.
H
y
b
r
id
izatio
n
a
tte
m
p
ts
o
f
DE
w
it
h
o
th
er
alg
o
r
ith
m
s
ca
n
b
e
f
o
u
n
d
in
r
ec
en
t
liter
atu
r
e.
Fo
r
in
s
ta
n
ce
,
Z
o
r
ar
p
ac
i
an
d
Özil
[
2
0
]
s
u
g
g
e
s
ted
a
co
m
b
i
n
atio
n
b
et
w
ee
n
DE
a
n
d
th
e
A
r
ti
f
icial
B
ee
C
o
lo
n
y
alg
o
r
ith
m
a
n
d
ap
p
lied
it
to
s
o
lv
e
th
e
p
r
o
b
le
m
o
f
f
ea
t
u
r
e
s
elec
tio
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
2
,
A
p
r
il 2
0
1
8
:
1
1
2
2
–
1130
1124
3.
P
RO
P
O
SE
D
SO
L
UT
I
O
N
:
“
B
AT
-
CL
US
T
E
R”
3
.
1
.
O
bje
ct
iv
e
F
un
ct
io
n
T
h
e
o
b
j
ec
tiv
e
f
u
n
ctio
n
f
o
r
t
h
e
al
g
o
r
ith
m
is
th
e
q
u
alit
y
m
ea
s
u
r
e
t
h
at
w
ill
h
elp
i
t
d
e
cid
e
w
h
at
clu
s
ter
i
n
g
co
n
f
i
g
u
r
atio
n
is
t
h
e
b
est.
Nan
d
a
an
d
P
an
d
a
[
2
1
]
p
r
o
v
id
ed
a
lis
t
o
f
s
ev
er
al
clu
s
ter
in
g
q
u
a
lit
y
m
etr
ic
s
av
ailab
le
i
n
t
h
e
liter
atu
r
e.
W
h
at
w
e
w
an
t
i
s
a
cl
u
s
ter
in
g
a
b
le
to
h
i
g
h
lig
h
t,
o
n
t
h
e
o
n
e
h
an
d
,
t
h
e
clo
s
en
e
s
s
b
et
w
ee
n
s
i
m
ilar
n
o
d
es,
an
d
o
n
th
e
o
th
er
h
an
d
,
t
h
e
s
ep
ar
atio
n
b
et
w
ee
n
d
if
f
er
en
t
n
o
d
es.
T
h
er
ef
o
r
e,
th
e
d
is
ta
n
ce
s
h
o
u
ld
h
av
e
a
f
u
n
d
a
m
en
tal
r
o
le
in
ch
o
o
s
in
g
o
u
r
q
u
alit
y
m
e
tr
ic.
Ho
w
e
v
er
,
r
el
y
i
n
g
o
n
t
h
e
d
is
tan
ce
f
r
o
m
t
h
e
clu
s
ter
ce
n
ter
alo
n
e
as
i
n
th
e
tr
ad
itio
n
al
K
-
Me
a
n
s
,
o
r
th
e
d
is
tan
ce
b
et
w
ee
n
cl
u
s
ter
ce
n
ter
s
m
a
y
n
o
t
b
e
s
u
f
f
icie
n
t.
W
e
n
ee
d
a
m
etr
ic
ab
le
to
p
r
o
v
id
e
a
c
o
m
b
i
n
atio
n
of
t
h
ese
t
wo
m
etr
ics
s
o
th
a
t
it
w
o
u
ld
ass
u
r
e
th
at
th
e
s
i
m
ilar
n
o
d
es a
r
e
clo
s
e
to
ea
ch
o
th
er
an
d
f
ar
f
r
o
m
t
h
e
n
o
d
es th
at
ar
e
d
if
f
er
en
t
f
r
o
m
t
h
e
m
.
On
e
o
f
th
e
m
o
s
t
p
o
p
u
lar
m
et
r
ics
in
th
e
liter
at
u
r
e
is
ca
lled
“
DB
I
n
d
ex
”
[
2
2
]
.
I
t
w
as
d
ev
elo
p
ed
b
y
Dav
ies
a
n
d
B
o
u
ld
in
,
an
d
it
p
r
o
v
id
es
a
r
atio
b
etw
ee
n
th
e
i
n
tr
a
-
cl
u
s
ter
d
is
ta
n
ce
(
th
e
d
is
tan
c
e
b
et
w
ee
n
t
h
e
n
o
d
es
in
th
e
s
a
m
e
clu
s
ter
)
an
d
th
e
i
n
t
er
-
clu
s
ter
d
is
ta
n
ce
(
th
e
d
is
ta
n
c
e
b
et
w
ee
n
t
h
e
ce
n
ter
s
o
f
ea
c
h
clu
s
ter
).
DB
I
n
d
ex
is
d
ef
i
n
ed
as:
1
1
c
n
i
i
c
D
B
R
n
(
1
)
W
h
er
e:
1
.
.
.
,
m
a
x
(
)
,
1
.
.
.
c
i
i
j
c
j
n
i
j
R
R
i
n
ij
ij
ij
ss
R
d
1
(
,
)
,
(
,
)
i
i
j
i
j
i
i
xc
i
d
d
g
g
s
d
x
g
c
(
,
)
d
x
y
is
th
e
E
u
clid
ian
d
i
s
tan
ce
b
et
wee
n
x
an
d
y
.
i
c
is
th
e
cl
u
s
ter
i
.
is
th
e
ce
n
ter
o
f
t
h
e
clu
s
ter
.
is
th
e
n
o
r
m
o
f
i
c
.
A
cc
o
r
d
in
g
to
Dav
ie
s
a
n
d
B
o
u
ld
in
[
2
2
]
,
a
co
r
r
ec
t
clu
s
ter
in
g
m
i
n
i
m
ize
s
t
h
e
DB
I
n
d
e
x
as
d
ep
icted
in
E
q
u
atio
n
(
1
)
.
T
h
at
b
ein
g
s
aid
,
th
e
o
b
j
ec
tiv
e
f
u
n
ctio
n
f
o
r
o
u
r
clu
s
ter
i
n
g
alg
o
r
it
h
m
s
h
o
u
ld
b
e:
(2
)
To
s
o
lv
e
E
q
u
atio
n
(
2
)
,
w
e
p
r
o
p
o
s
e
a
h
y
b
r
id
izatio
n
o
f
t
h
e
s
t
an
d
ar
d
B
at
Alg
o
r
it
h
m
b
y
Yan
g
[
8
]
w
it
h
th
e
FF
DP
alg
o
r
ith
m
th
a
t
w
e
d
ev
elo
p
ed
in
a
p
r
ev
io
u
s
w
o
r
k
[
7
]
.
W
e
ch
o
s
e
to
ca
ll
th
is
h
y
b
r
id
izatio
n
“
B
at
-
C
lu
s
ter
,
”
o
r
B
C
.
3.
2.
B
a
t
-
Clus
t
er
B
at
-
C
l
u
s
ter
,
o
r
B
C
,
is
a
co
m
b
in
atio
n
o
f
t
w
o
al
g
o
r
ith
m
s
,
F
FDP
an
d
B
at
A
lg
o
r
it
h
m
.
F
F
DP
w
ill
r
u
n
f
ir
s
t
to
s
et
a
n
o
p
ti
m
ized
eq
u
i
li
b
r
iu
m
p
o
s
itio
n
i
n
g
o
f
t
h
e
n
o
d
e
s
o
f
t
h
e
g
r
ap
h
.
T
h
ese
n
o
d
e
p
o
s
itio
n
s
w
ill
th
e
n
b
e
ass
i
g
n
ed
to
th
e
B
at
A
l
g
o
r
ith
m
.
B
A
w
ill
s
tar
t
b
y
g
en
er
ati
n
g
a
p
o
p
u
latio
n
o
f
b
ats.
E
ac
h
o
f
th
ese
b
ats
w
il
l
h
a
v
e
its
o
w
n
in
itia
l
lo
u
d
n
es
s
,
p
u
l
s
e
r
ate,
p
o
s
it
io
n
an
d
v
elo
cit
y
.
T
h
e
i
n
itial
b
ats
p
o
s
itio
n
s
r
ep
r
ese
n
t
t
h
e
i
n
it
i
al
clu
s
ter
ce
n
ter
s
.
W
h
en
t
h
e
alg
o
r
it
h
m
s
tar
ts
r
u
n
n
i
n
g
,
ea
ch
b
at
w
ill
b
e
ass
ig
n
ed
to
a
clu
s
ter
ce
n
ter
lo
ca
ti
o
n
.
Fo
r
ea
ch
clu
s
te
r
i
g
i
i
c
1
1
m
in
c
n
i
i
c
D
B
R
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
B
a
t
-
C
lu
s
ter:
A
B
a
t A
lg
o
r
ith
m
-
b
a
s
ed
A
u
to
ma
te
d
Gra
p
h
C
lu
s
teri
n
g
A
p
p
r
o
a
c
h
(
Za
ka
r
ia
B
o
u
l
o
u
a
r
d
)
1125
ce
n
ter
,
th
e
alg
o
r
ith
m
w
ill
ca
lc
u
late
th
e
m
ea
n
v
a
lu
e
o
f
t
h
e
clo
s
est
n
o
d
es
to
it.
T
h
e
clu
s
ter
ce
n
ter
‟
s
p
o
s
itio
n
is
th
en
u
p
d
ated
,
an
d
t
h
e
o
b
j
ec
tiv
e
f
u
n
ctio
n
i
s
ca
lc
u
lated
a
s
i
n
t
h
e
E
q
u
atio
n
(
2
)
.
I
f
t
h
e
v
al
u
e
o
f
t
h
e
o
b
j
ec
tiv
e
f
u
n
ctio
n
h
as
co
n
v
er
g
ed
,
we
r
etu
r
n
th
e
clu
s
ter
ce
n
ter
lo
c
atio
n
s
;
o
th
er
w
i
s
e,
w
e
r
ea
s
s
i
g
n
ea
c
h
b
at
to
th
e
co
r
r
esp
o
n
d
in
g
clu
s
ter
ce
n
ter
o
n
ce
ag
ai
n
.
I
f
th
e
r
an
d
o
m
v
alu
e
r
a
n
d
is
g
r
ea
ter
th
an
t
h
e
b
ats
p
u
ls
e
r
ate,
th
e
al
g
o
r
ith
m
s
elec
t
s
a
s
o
lu
tio
n
a
m
o
n
g
th
e
b
est
s
o
l
u
tio
n
s
an
d
g
en
er
at
es
a
lo
ca
l
s
o
lu
tio
n
ar
o
u
n
d
t
h
e
s
elec
ted
b
est
s
o
lu
tio
n
.
I
f
t
h
e
r
an
d
o
m
v
a
lu
e
r
a
n
d
is
s
m
aller
t
h
a
n
t
h
e
lo
u
d
n
es
s
i
A
,
an
d
th
e
v
al
u
e
o
f
th
e
o
b
j
ec
tiv
e
f
u
n
ct
io
n
f
o
r
th
e
c
u
r
r
en
t
b
at
p
o
s
itio
n
i
s
b
etter
(
s
m
aller
i
n
o
u
r
ca
s
e)
t
h
a
n
th
e
v
alu
e
o
f
t
h
e
b
est
s
o
l
u
tio
n
f
o
u
n
d
s
o
f
ar
,
*
(
)
(
)
i
f
x
f
x
,
th
e
n
e
w
s
o
l
u
tio
n
is
ac
ce
p
ted
,
th
e
b
ats
p
u
ls
e
r
ate
i
s
i
n
cr
ea
s
ed
,
an
d
th
e
lo
u
d
n
e
s
s
is
d
ec
r
ea
s
ed
.
T
h
e
s
o
l
u
tio
n
s
f
o
u
n
d
ar
e
s
o
r
ted
,
an
d
th
e
cu
r
r
en
t b
est s
o
lu
tio
n
i
s
s
to
r
ed
.
T
h
e
alg
o
r
ith
m
k
ee
p
s
r
u
n
n
i
n
g
u
n
t
il
th
e
s
to
p
cr
iter
io
n
is
r
esp
ec
ted
.
I
n
o
u
r
ca
s
e,
th
e
alg
o
r
ith
m
s
h
o
u
ld
s
to
p
if
th
e
iter
at
io
n
n
u
m
b
er
t
b
ec
o
m
e
s
eq
u
al
to
t
h
e
m
a
x
i
m
u
m
n
u
m
b
er
o
f
iter
atio
n
s
M
.
T
h
e
p
s
eu
d
o
co
d
e
o
f
th
e
B
at
-
C
l
u
s
ter
al
g
o
r
ith
m
w
ill
b
e
t
h
en
d
e
s
cr
ib
ed
as f
o
llo
w
s
:
A
l
g
o
r
i
t
h
m
B
a
t
C
l
u
s
t
e
r
(
G
,
X
,
to
l
,
K
,
M
,
N
,
0
A
,
0
r
)
i
n
p
u
t
:
G
r
a
p
h
G
,
N
o
d
e
s i
n
i
t
i
a
l
p
o
s
i
t
i
o
n
s
X
,
t
o
l
e
r
a
n
c
e
to
l
,
n
o
mi
n
a
l
e
d
g
e
l
e
n
g
t
h
K
,
M
a
x
i
mu
m
i
t
e
r
a
t
i
o
n
s
n
u
m
b
e
r
M
,
B
a
t
s
t
o
t
a
l
p
o
p
u
l
a
t
i
o
n
N
,
i
n
i
t
i
a
l
l
o
u
d
n
e
ss
0
A
a
n
d
i
n
i
t
i
a
l
p
u
l
se
r
a
t
e
0
r
/
/
S
e
t
t
h
e
g
r
a
p
h
‟
s
p
o
s
i
t
i
o
n
i
n
g
(
,
,
,
)
c
o
o
r
d
s
F
F
D
P
G
X
t
o
l
K
/
/
c
o
o
r
d
s
i
s
2
-
D
o
r
3
-
D
v
e
c
t
o
r
/
*
I
n
i
t
i
a
l
i
z
e
t
h
e
b
a
t
s
p
o
si
t
i
o
n
s
i
x
,
v
e
l
o
c
i
t
i
e
s
i
v
a
n
d
f
r
e
q
u
e
n
c
i
e
s
i
f
*/
Fo
r
a
l
l
b
a
t
s
i
d
o
{
0
0
0
0
i
i
i
i
i
AA
rr
f
v
x
r
a
n
d
/
*
C
a
l
c
u
l
a
t
e
t
h
e
so
l
u
t
i
o
n
o
f
t
h
e
o
b
j
e
c
t
i
v
e
f
u
n
c
t
i
o
n
f
o
r
t
h
e
b
a
t
i
a
c
c
o
r
d
i
n
g
t
o
t
h
e
e
q
u
a
t
i
o
n
(
2
)
*
/
(
,
)
i
D
B
x
c
o
o
r
d
s
}
/
/
S
e
l
e
c
t
t
h
e
b
e
st
so
l
u
t
i
o
n
W
h
i
l
e
(
tM
)
{
/
*
G
e
n
e
r
a
t
e
a
n
e
w
so
l
u
t
i
o
n
b
y
a
d
j
u
st
i
n
g
t
h
e
f
r
e
q
u
e
n
c
i
e
s
o
f
t
h
e
b
a
t
s
a
c
c
o
r
d
i
n
g
t
o
t
h
e
e
q
u
a
t
i
o
n
(
4
)
,
u
p
d
a
t
i
n
g
t
h
e
v
e
l
o
c
i
t
i
e
s a
n
d
t
h
e
p
o
s
i
t
i
o
n
s o
f
t
h
e
b
a
t
s
a
c
c
o
r
d
i
n
g
t
o
t
h
e
e
q
u
a
t
i
o
n
s (5
)
a
n
d
(
6
)
*
/
Fo
r
a
l
l
b
a
t
s
i
d
o
{
m
i
n
m
a
x
m
i
n
()
i
f
f
f
f
1*
()
t
t
t
i
i
i
i
v
v
x
x
f
11
t
t
t
i
i
i
x
x
v
If
(
i
r
a
n
d
r
)
{
/
*
G
e
n
e
r
a
t
e
a
n
e
w
so
l
u
t
i
o
n
a
r
o
u
n
d
t
h
e
b
e
st
so
l
u
t
i
o
n
a
c
c
o
r
d
i
n
g
t
o
t
h
e
e
q
u
a
t
i
o
n
(
7
)
*
/
t
n
e
w
o
l
d
x
x
A
}
}
If
(
*
(
,
)
(
,
)
ii
r
a
n
d
A
a
n
d
D
B
x
c
o
o
r
d
s
D
B
x
c
o
o
r
d
s
)
{
/
/
A
c
c
e
p
t
t
h
e
n
e
w
so
l
u
t
i
o
n
s
*
i
xx
/
*
U
p
d
a
t
e
i
A
a
n
d
i
r
a
c
c
o
r
d
i
n
g
t
o
e
q
u
a
t
i
o
n
s
(
8
)
a
n
d
(
9
)
*
/
1
tt
ii
AA
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
2
,
A
p
r
il 2
0
1
8
:
1
1
2
2
–
1130
1126
10
1
t
t
ii
r
r
e
}
}
/
/
S
e
l
e
c
t
t
h
e
c
u
r
r
e
n
t
b
e
st
so
l
u
t
i
o
n
1
tt
}
4.
T
E
ST
I
N
G
AND
R
E
SU
L
T
S
4
.
1
.
T
esting
E
nv
iro
n
m
ent
Fo
r
s
i
m
u
latio
n
p
u
r
p
o
s
es,
w
e
r
an
th
e
B
at
-
C
l
u
s
ter
alg
o
r
ith
m
i
n
a
co
m
p
u
ter
w
it
h
t
h
e
f
o
llo
w
i
n
g
tech
n
ical
ch
ar
ac
ter
is
tics
:
a.
E
n
v
ir
o
n
m
e
n
t: J
av
a
8
+
Ma
tlab
R
2
0
1
7
a
b.
OS: W
in
d
o
w
s
7
c.
C
P
U:
I
n
tel
i5
2
4
5
0
M
2
.
5
Gh
z
d.
R
A
M:
4
Gb
T
o
test
th
e
p
er
f
o
r
m
a
n
ce
s
o
f
B
C
,
w
e
p
u
t it
i
n
co
m
p
ar
is
o
n
w
i
t
h
th
r
ee
d
is
ti
n
g
u
is
h
ed
alg
o
r
ith
m
s
:
e.
P
ar
ticle
S
w
ar
m
Op
ti
m
iza
tio
n
f.
An
t Co
lo
n
y
Op
ti
m
izat
io
n
g.
Dif
f
er
en
t
ial
E
v
o
l
u
tio
n
W
e
w
ill
u
s
e
t
h
e
co
n
t
in
u
o
u
s
a
s
p
ec
t
o
f
all
th
e
s
e
al
g
o
r
ith
m
s
,
an
d
th
e
f
u
n
ctio
n
to
o
p
ti
m
ize
w
il
l
b
e
th
e
DB
I
n
d
ex
as
d
ep
icted
in
th
e
E
q
u
atio
n
(2
)
.
T
h
is
ap
p
r
o
ac
h
w
il
l
en
ab
le
u
s
to
co
m
p
ar
e
t
h
e
p
er
f
o
r
m
a
n
ce
s
o
f
t
h
es
e
alg
o
r
ith
m
s
o
n
an
eq
u
al
f
o
o
t.
4
.
2.
B
ench
m
a
r
k
G
ra
p
hs
T
h
e
g
r
ap
h
s
t
h
at
w
e
w
ill
u
s
e
i
n
o
u
r
test
s
ar
e
t
h
r
ee
b
en
ch
m
ar
k
g
r
ap
h
s
o
f
d
i
f
f
er
en
t
s
izes a
n
d
co
m
e
f
r
o
m
d
if
f
er
e
n
t d
o
m
ai
n
s
.
T
h
e
s
e
g
r
ap
h
s
ar
e
av
ailab
le
i
n
t
h
e
Gep
h
i s
tan
d
ar
d
d
ataset
ac
ce
s
s
ib
le
in
t
h
e
f
o
llo
w
i
n
g
lin
k
:
h
ttp
s
:/
/g
i
th
u
b
.
co
m
/
m
ed
ialab
/b
en
ch
m
ar
k
F
o
r
ce
Atlas2
/b
lo
b
/
m
aster
/d
ataset.
zip
(
last
ch
ec
k
ed
:
Dec
e
m
b
er
2
1
th
,
2
0
1
7
)
.
T
h
e
4
g
r
ap
h
s
ar
e:
a.
f
ac
eb
o
o
k
_
eg
o
_
6
8
6
(
1
6
8
n
o
d
es
an
d
1
6
5
6
lin
k
s
)
b.
y
ea
s
t (
2
3
6
1
n
o
d
es a
n
d
7
1
8
2
lin
k
s
)
c.
ar
x
iv
_
g
en
er
al_
r
elati
v
it
y
(
5
2
4
2
n
o
d
es a
n
d
2
8
9
8
0
lin
k
s
)
d.
o
r
eg
o
n
2
_
0
1
0
3
3
1
(
1
0
9
0
0
n
o
d
e
s
an
d
3
1
1
8
0
lin
k
s
)
T
ab
le
1
d
is
p
lay
s
t
h
e
la
y
o
u
t
s
o
f
th
e
4
b
en
ch
m
ar
k
g
r
ap
h
s
.
T
ab
le
1
.
T
h
e
B
en
ch
m
ar
k
Gr
ap
h
s
L
a
y
o
u
ts
f
a
c
e
b
o
o
k
_
e
g
o
_
6
8
6
y
e
a
st
a
r
x
i
v
_
g
e
n
e
r
a
l
_
r
e
l
a
t
i
v
i
t
y
o
r
e
g
o
n
2
_
0
1
0
3
3
1
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
B
a
t
-
C
lu
s
ter:
A
B
a
t A
lg
o
r
ith
m
-
b
a
s
ed
A
u
to
ma
te
d
Gra
p
h
C
lu
s
teri
n
g
A
p
p
r
o
a
c
h
(
Za
ka
r
ia
B
o
u
l
o
u
a
r
d
)
1127
4
.
3
.
P
a
ra
m
et
er
s
Set
t
ing
Def
i
n
i
n
g
t
h
e
co
r
r
ec
t
p
ar
am
ete
r
s
f
o
r
a
n
atu
r
e
in
s
p
ir
ed
alg
o
r
i
th
m
,
i
n
g
e
n
er
al,
r
eq
u
ir
es
r
ig
o
r
o
u
s
p
r
io
r
test
i
n
g
.
T
h
e
s
a
m
e
g
o
es
f
o
r
B
C
an
d
all
o
f
th
e
al
g
o
r
ith
m
s
th
at
w
e
w
i
ll test it
a
g
ai
n
s
t
.
Af
ter
s
e
v
er
al
ex
p
er
i
m
en
t
s
,
th
e
p
ar
am
eter
s
w
e
f
o
u
n
d
ab
le
to
an
s
w
er
o
u
r
n
ee
d
s
co
r
r
ec
tl
y
ar
e
th
e
f
o
llo
w
i
n
g
:
a.
B
at
-
C
l
u
s
ter
:
Ma
x
i
m
u
m
I
ter
at
io
n
s
Nu
m
b
er
200
M
,
B
at
P
o
p
u
latio
n
Size
50
N
,
I
n
itial
L
o
u
d
n
ess
0
1
.
1
A
,
I
n
itial P
u
ls
e
R
ate
0
0
.
5
r
b.
P
ar
ticle
S
w
ar
m
Op
ti
m
iza
tio
n
:
Ma
x
i
m
u
m
I
ter
atio
n
s
Nu
m
b
er
200
M
,
P
ar
ticle
Po
p
u
latio
n
Size
50
N
,
I
n
er
tia
W
eig
h
t
1
w
,
I
n
er
tia
W
eig
h
t
Da
m
p
in
g
R
atio
0
.
9
9
w
d
a
m
p
,
P
er
s
o
n
al
L
ea
r
n
i
n
g
C
o
e
f
f
icien
t
1
1
.
5
c
,
Glo
b
al
L
ea
r
n
i
n
g
C
o
e
f
f
icien
t
2
2
c
c.
An
t
C
o
lo
n
y
Op
ti
m
izatio
n
:
Ma
x
i
m
u
m
I
ter
atio
n
s
Nu
m
b
e
r
200
M
,
An
t
P
o
p
u
latio
n
Size
10
N
,
Sa
m
p
le
Size
40
n
S
a
m
p
l
e
,
I
n
ten
s
i
f
icatio
n
Facto
r
0
.
5
q
,
Dev
iatio
n
Dis
ta
n
c
e
R
atio
1
d.
Dif
f
er
en
t
ial
E
v
o
l
u
tio
n
:
Ma
x
i
m
u
m
I
ter
atio
n
s
Nu
m
b
er
200
M
,
I
n
d
iv
id
u
al
P
o
p
u
latio
n
Size
50
N
,
C
r
o
s
s
o
v
er
P
r
o
b
ab
ilit
y
0
.
2
p
C
R
4
.
3
.
E
x
peri
m
e
nta
l R
esu
lt
s
T
h
e
T
a
b
le
2
to
T
a
b
le
5
s
h
o
w
th
e
p
er
f
o
r
m
a
n
ce
s
o
f
th
e
B
at
-
C
lu
s
ter
co
m
p
ar
ed
w
it
h
ea
c
h
o
f
t
h
e
o
t
h
er
af
o
r
esaid
alg
o
r
ith
m
s
o
n
t
h
e
f
o
u
r
b
en
ch
m
ar
k
g
r
ap
h
s
.
T
ab
le
2
.
f
ac
eb
o
o
k
_
eg
o
_
6
8
6
A
l
g
o
r
i
t
h
m
N
u
mb
e
r
o
f
C
l
u
st
e
r
s
D
B
I
n
d
e
x
V
a
l
u
e
s
BC
4
0
,
7
2
6
5
7
PSO
3
0
,
7
2
7
3
1
A
C
O
3
0
,
8
0
2
0
9
DE
3
0
,
7
3
8
8
4
I
n
“f
ac
eb
o
o
k
_
eg
o
_
6
8
6
”,
B
at
-
C
lu
s
ter
p
r
o
v
id
ed
th
e
s
m
allest
o
p
tim
a
l v
al
u
e
f
o
r
th
e
DB
I
n
d
ex
,
clo
s
el
y
s
ec
o
n
d
ed
b
y
P
SO.
Yet,
B
C
w
a
s
th
e
o
n
l
y
al
g
o
r
ith
m
ab
le
to
p
r
o
v
id
e
th
r
ee
clu
s
ter
s
w
h
ile
t
h
e
o
th
er
alg
o
r
ith
m
s
p
r
o
v
id
ed
o
n
ly
3
clu
s
ter
s
.
T
ab
le
3
.
y
ea
s
t
A
l
g
o
r
i
t
h
m
N
u
mb
e
r
o
f
C
l
u
st
e
r
s
D
B
I
n
d
e
x
V
a
l
u
e
s
BC
4
0
,
6
9
3
3
2
PSO
5
0
,
6
9
4
2
3
A
C
O
3
0
,
8
1
1
5
2
DE
3
0
,
7
1
9
6
8
T
h
e
r
esu
lts
i
n
t
h
e
“y
ea
s
t”
g
r
a
p
h
ca
n
b
e
d
eb
atab
le
at
f
ir
s
t.
I
n
d
ee
d
,
b
ased
o
n
th
e
DB
I
n
d
ex
alo
n
e,
w
e
w
il
l
s
a
y
t
h
at
B
C
w
a
s
t
h
e
b
est,
b
u
t
s
ee
i
n
g
th
a
t
P
SO
w
as
ab
le
to
p
r
o
v
id
e
m
o
r
e
cl
u
s
ter
s
ca
n
o
p
en
th
e
p
o
s
s
ib
ili
t
y
th
at
P
SO
m
a
y
b
e
ab
le
to
f
i
n
d
a
b
etter
v
al
u
e
f
o
r
t
h
e
DB
I
n
d
ex
.
Ho
w
e
v
er
,
ac
co
r
d
in
g
to
t
h
e
ev
o
lu
tio
n
o
f
t
h
e
b
es
t
v
alu
e
s
as
d
ep
icted
in
Fig
u
r
e
1
,
w
e
w
il
l
s
ee
t
h
at
P
SO
s
tar
ted
s
tag
n
ati
n
g
af
ter
t
h
e
iter
atio
n
1
0
0
in
a
DB
I
n
d
ex
v
alu
e
h
i
g
h
er
th
a
n
t
h
e
o
n
e
p
r
o
v
id
ed
b
y
B
C
.
T
h
is
co
n
cl
u
d
es
to
th
e
f
ac
t
th
at
h
a
v
i
n
g
5
cl
u
s
t
er
s
m
a
y
n
o
t
b
e
t
h
e
b
est clu
s
ter
i
n
g
s
ce
n
ar
io
.
T
ab
le
4
.
ar
x
iv
_
g
en
er
al_
r
elati
v
it
y
A
l
g
o
r
i
t
h
m
N
u
mb
e
r
o
f
C
l
u
st
e
r
s
D
B
I
n
d
e
x
V
a
l
u
e
s
BC
6
0
,
7
4
7
9
7
PSO
4
0
,
7
5
5
8
A
C
O
3
0
,
8
1
4
7
1
DE
3
0
,
7
6
0
4
9
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
2
,
A
p
r
il 2
0
1
8
:
1
1
2
2
–
1130
1128
Fig
u
r
e
1
.
T
h
e
ev
o
lu
tio
n
o
f
t
h
e
b
est DB
I
n
d
ex
v
al
u
es p
r
o
v
id
ed
b
y
P
SO f
o
r
t
h
e
“y
ea
s
t”
g
r
ap
h
I
n
th
e
“
ar
x
i
v
_
g
e
n
er
al_
r
elati
v
it
y
”
g
r
ap
h
,
B
C
g
av
e
t
h
e
s
m
alle
s
t v
alu
e
s
o
f
DB
I
n
d
e
x
an
d
m
u
c
h
m
o
r
e
clu
s
ter
s
(
6
ag
ain
s
t 4
p
r
o
v
id
ed
b
y
t
h
e
f
ir
s
t r
u
n
n
er
-
u
p
P
SO)
.
T
ab
le
5
.
o
r
eg
o
n
2
_
0
1
0
3
3
1
A
l
g
o
r
i
t
h
m
N
u
mb
e
r
o
f
C
l
u
st
e
r
s
D
B
I
n
d
e
x
V
a
l
u
e
s
BC
3
0
,
5
8
0
9
3
PSO
3
0
,
5
8
1
0
1
A
C
O
2
0
,
7
1
0
3
7
DE
2
0
,
6
2
9
0
2
R
eg
ar
d
in
g
th
e
“
o
r
eg
o
n
2
_
0
1
0
3
3
1
”
g
r
ap
h
,
B
C
an
d
P
SO
w
e
r
e
ab
le
t
o
p
r
o
v
id
e
3
clu
s
ter
s
,
w
h
ile
th
e
o
th
er
t
w
o
co
u
ld
o
n
l
y
p
r
o
v
id
e
2
clu
s
ter
s
.
T
h
e
DB
I
n
d
ex
v
al
u
es
o
f
B
C
an
d
P
SO
w
er
e
v
er
y
clo
s
e,
w
it
h
a
s
m
al
l
ad
v
an
ta
g
e
f
o
r
B
C
.
Ov
er
all,
B
at
-
C
lu
s
ter
w
a
s
ab
le
to
p
r
o
v
id
e
th
e
b
est
v
alu
es
o
f
DB
I
n
d
ex
o
n
all
th
e
b
en
ch
m
a
r
k
g
r
ap
h
s
.
B
ein
g
c
lo
s
el
y
s
ec
o
n
d
ed
b
y
P
SO
s
h
o
w
s
th
e
ab
ilit
y
o
f
t
h
e
S
w
ar
m
Op
ti
m
izat
io
n
al
g
o
r
ith
m
s
to
tac
k
le
t
h
i
s
k
in
d
o
f
p
r
o
b
le
m
s
.
Ho
w
e
v
er
,
t
h
e
r
esu
lt
s
p
r
o
v
id
ed
b
y
A
C
O
w
e
r
e
p
o
o
r
er
th
an
e
x
p
ec
ted
.
W
h
en
w
e
lo
o
k
at
t
h
e
ev
o
lu
tio
n
o
f
t
h
e
b
es
t
v
alu
e
p
r
o
v
id
ed
b
y
AC
O
o
n
“f
ac
eb
o
o
k
_
eg
o
_
6
8
6
,
”
as
in
Fi
g
u
r
e
2
,
f
o
r
ex
a
m
p
le,
w
e
ca
n
s
ee
th
at
t
h
e
al
g
o
r
ith
m
k
ep
t
f
i
n
d
in
g
b
etter
v
al
u
e
s
.
T
h
is
ca
n
b
e
ex
p
lain
ed
b
y
t
h
e
f
ac
t
t
h
at
t
h
e
co
n
f
i
g
u
r
atio
n
w
e
g
av
e
to
AC
O
m
a
y
p
r
o
b
ab
ly
n
o
t b
e
th
e
b
est
.
Fig
u
r
e
2
.
T
h
e
ev
o
lu
tio
n
o
f
t
h
e
b
est DB
I
n
d
ex
v
al
u
es p
r
o
v
id
ed
b
y
AC
O
f
o
r
th
e
“f
ac
eb
o
o
k
_
e
g
o
_
6
8
6
”
g
r
ap
h
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
B
a
t
-
C
lu
s
ter:
A
B
a
t A
lg
o
r
ith
m
-
b
a
s
ed
A
u
to
ma
te
d
Gra
p
h
C
lu
s
teri
n
g
A
p
p
r
o
a
c
h
(
Za
ka
r
ia
B
o
u
l
o
u
a
r
d
)
1129
5.
CO
NCLU
SI
O
N
T
h
is
p
ap
er
p
r
esen
ted
th
e
B
at
-
C
l
u
s
ter
(
BC
)
alg
o
r
ith
m
.
I
t
is
a
co
m
b
in
at
io
n
o
f
th
e
F
FDP
alg
o
r
ith
m
d
ev
elo
p
ed
by
o
u
r
tea
m
[
7
]
an
d
th
e
B
at
A
l
g
o
r
it
h
m
d
e
v
elo
p
ed
b
y
Xi
n
-
S
h
e
Ya
n
g
[
8
]
.
B
C
is
a
n
al
g
o
r
ith
m
d
esig
n
ed
to
an
s
w
er
t
h
e
n
ee
d
f
o
r
au
to
m
ated
lar
g
e
g
r
ap
h
clu
s
ter
i
n
g
.
I
n
co
n
tr
ast
w
it
h
s
ev
er
al
clu
s
ter
i
n
g
alg
o
r
ith
m
s
a
v
ailab
le
i
n
th
e
lite
r
atu
r
e,
B
C
w
a
s
ab
le
to
tr
an
s
lat
e
th
e
au
to
m
ated
lar
g
e
g
r
ap
h
cl
u
s
ter
i
n
g
i
s
s
u
e
i
n
to
a
co
n
tin
u
o
u
s
p
r
o
b
l
e
m
,
w
h
ile
t
h
e
o
th
er
s
o
lu
t
io
n
s
ten
d
to
f
o
r
m
u
late
it
a
s
a
d
i
s
cr
ete
p
r
o
b
le
m
.
T
h
e
id
ea
h
er
e
w
a
s
to
r
u
n
a
lar
g
e
g
r
ap
h
la
y
o
u
t
a
lg
o
r
ith
m
,
th
e
FF
DP
,
an
d
m
a
k
e
it
p
r
o
v
id
e
th
e
co
o
r
d
in
ates
o
f
th
e
eq
u
ilib
r
iu
m
p
o
s
itio
n
s
o
f
th
e
g
r
ap
h
‟
s
n
o
d
es
.
Hav
i
n
g
t
h
e
s
e
co
o
r
d
in
ates
en
a
b
led
u
s
to
tr
an
s
late
t
h
e
g
r
ap
h
to
a
s
tan
d
ar
d
r
ea
l
v
alu
ed
v
ec
to
r
ea
s
il
y
s
o
lv
ab
le
w
it
h
t
h
e
co
n
ti
n
u
o
u
s
v
er
s
io
n
o
f
th
e
B
at
A
l
g
o
r
it
h
m
.
T
h
e
q
u
alit
y
m
e
tr
ic
w
e
u
s
ed
to
m
ea
s
u
r
e
th
e
q
u
alit
y
o
f
o
u
r
clu
s
ter
i
n
g
w
as
t
h
e
DB
I
n
d
ex
b
y
Da
v
ies
a
n
d
B
o
u
ld
in
[
2
2
]
.
T
h
e
B
at
-
C
lu
s
ter
alg
o
r
ith
m
w
a
s
tes
ted
o
n
f
o
u
r
b
en
ch
m
ar
k
g
r
ap
h
s
o
f
d
i
f
f
er
e
n
t
s
izes
a
n
d
f
r
o
m
d
if
f
er
en
t
d
o
m
ai
n
s
.
B
C
p
r
o
v
ed
to
b
e
a
g
o
o
d
alter
n
ativ
e
s
o
lu
tio
n
to
s
o
lv
e
th
e
au
to
m
ated
lar
g
e
g
r
ap
h
clu
s
ter
in
g
p
r
o
b
lem
w
h
e
n
co
m
p
ar
ed
to
alg
o
r
ith
m
s
co
n
s
id
er
ed
a
m
o
n
g
th
e
b
est i
n
th
e
li
ter
atu
r
e.
T
h
e
B
at
-
C
l
u
s
ter
al
g
o
r
it
h
m
will
b
e
in
teg
r
ated
i
n
to
XE
W
G
r
ap
h
[
2
3
]
,
th
e
lar
g
e
g
r
ap
h
v
i
s
u
aliza
tio
n
s
er
v
ice
o
f
th
e
C
o
m
p
etiti
v
e
I
n
telli
g
e
n
ce
to
o
l
Xp
lo
r
E
v
er
y
W
h
er
e
[
2
4
]
.
C
o
u
p
led
w
i
th
t
h
e
o
u
t
o
f
th
e
b
o
x
ca
teg
o
r
izatio
n
p
r
o
v
id
ed
b
y
X
E
W
Gr
ap
h
‟
s
h
y
p
er
g
r
ap
h
ap
p
r
o
ac
h
,
B
C
w
ill
e
n
ab
le
t
h
e
u
s
er
t
o
h
av
e
lar
g
e
g
r
ap
h
s
clu
s
ter
ed
an
d
ex
p
a
n
d
ed
o
n
d
em
an
d
f
o
r
b
o
th
th
e
w
eb
an
d
th
e
m
o
b
ile
o
r
ien
ted
in
ter
f
ac
e
s
o
f
XE
W
Gr
ap
h
.
RE
F
E
R
E
NC
E
S
[1
]
C.
Virm
a
n
i,
A
.
P
il
lai
,
a
n
d
D.
J
u
n
e
ja,
“
Clu
ste
rin
g
i
n
A
g
g
re
g
a
ted
Us
e
r
P
r
o
f
il
e
s
a
c
ro
ss
M
u
lt
i
p
le
S
o
c
ial
Ne
tw
o
rk
s,”
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
E
lec
trica
l
a
n
d
C
o
mp
u
ter
En
g
in
e
e
rin
g
(
IJ
ECE
)
,
v
o
l.
7
,
n
o
.
6
,
p
p
.
3
6
9
2
–
3
6
9
9
,
De
c
.
2
0
1
7
.
[2
]
S
.
Jin
a
ra
t,
C.
Ha
ru
e
c
h
a
iy
a
sa
k
,
a
n
d
A
.
Ru
n
g
sa
w
a
n
g
,
“
G
r
a
p
h
-
Ba
se
d
Co
n
c
e
p
t
Clu
ste
ri
n
g
f
o
r
Web
S
e
a
rc
h
Re
su
lt
s,”
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
E
lec
trica
l
a
n
d
C
o
mp
u
ter
En
g
in
e
e
rin
g
(
IJ
ECE
)
,
v
o
l.
5
,
n
o
.
6
,
p
p
.
1
5
3
6
–
1
5
4
4
,
De
c
.
2
0
1
5
.
[3
]
A
.
M
a
h
b
o
u
b
,
M
.
A
rio
u
a
,
a
n
d
E.
M
.
En
-
Na
im
i,
“
En
e
rg
y
-
Eff
i
c
ien
t
Hy
b
rid
K
-
M
e
a
n
s
A
lg
o
rit
h
m
f
o
r
Clu
ste
re
d
W
irele
ss
S
e
n
so
r
Ne
tw
o
rk
s,”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
trica
l
a
n
d
C
o
mp
u
ter
En
g
in
e
e
rin
g
(
IJ
ECE
)
,
v
o
l.
7
,
n
o
.
4
,
p
p
.
2
0
5
4
–
2
0
6
0
,
A
u
g
.
2
0
1
7
.
[4
]
H.
Zh
o
u
a
n
d
R.
L
ip
o
w
s
k
y
,
“
Dy
n
a
mic
p
a
t
ter
n
e
v
o
lu
ti
o
n
o
n
sc
a
le
-
fre
e
n
e
two
rk
s
,”
P
ro
c
e
e
d
in
g
s
o
f
th
e
Na
ti
o
n
a
l
A
c
a
d
e
m
y
o
f
S
c
ien
c
e
s o
f
th
e
Un
it
e
d
S
tate
s o
f
Am
e
ri
c
a
,
v
o
l.
1
0
2
,
n
o
.
2
9
,
p
p
.
1
0
0
5
2
–
7
,
J
u
l.
2
0
0
5
.
[5
]
E
.
R.
Hru
sc
h
k
a
,
R.
J.
G
.
B.
Ca
m
p
e
ll
o
,
A
.
A
.
F
re
it
a
s,
a
n
d
A
.
C.
P
.
L
.
F
.
d
e
Ca
rv
a
lh
o
,
“
A
su
rv
e
y
o
f
e
v
o
lu
ti
o
n
a
ry
a
lg
o
rit
h
m
s
f
o
r
c
lu
ste
rin
g
,
”
IEE
E
T
ra
n
sa
c
ti
o
n
s o
n
S
y
ste
ms
,
M
a
n
a
n
d
Cy
b
e
rn
e
ti
c
s P
a
rt C:
Ap
p
li
c
a
ti
o
n
s a
n
d
Rev
iews
,
v
o
l.
3
9
,
n
o
.
2
,
p
p
.
1
3
3
–
1
5
5
,
2
0
0
9
.
[6
]
D.
Ca
m
a
c
h
o
,
“
Bi
o
-
i
n
sp
ire
d
Cl
u
s
ter
in
g
:
b
a
sic
fea
t
u
re
s
a
n
d
f
u
t
u
re
tre
n
d
s
in
t
h
e
e
ra
o
f
Bi
g
Da
t
a
,
”
in
Cy
b
e
rn
e
ti
c
s
IEE
E
Co
n
f
o
n
,
2
0
1
5
.
[7
]
Z.
Bo
u
l
o
u
a
rd
,
L
.
Ko
u
tt
i,
A
.
El
Ha
d
d
a
d
i
,
a
n
d
B.
Do
u
ss
e
t,
“
„F
o
rc
e
d
‟
F
o
rc
e
Dire
c
ted
P
lac
e
m
e
n
t:
a
Ne
w
A
lg
o
rit
h
m
f
o
r
L
a
r
g
e
G
r
a
p
h
V
isu
a
li
z
a
ti
o
n
,
”
In
ter
n
a
ti
o
n
a
l
Rev
iew o
n
Co
mp
u
ter
s a
n
d
S
o
f
twa
re
(
IRE
COS
)
,
v
o
l.
1
2
,
n
o
.
2
,
p
p
.
7
5
–
8
3
,
M
a
r.
2
0
1
7
.
[8
]
X.
-
S
.
Ya
n
g
,
“
A
Ne
w
M
e
ta
h
e
u
ristic
Ba
t
-
In
s
p
ire
d
Al
g
o
rit
h
m,”
i
n
Na
tu
re
In
sp
ired
C
o
o
p
e
ra
ti
v
e
S
trate
g
ies
f
o
r
Op
ti
m
iza
ti
o
n
(NICSO
2
0
1
0
)
,
2
0
1
0
t
h
e
d
.
,
v
o
l.
2
8
4
,
J.
R.
G
o
n
z
a
lez
,
D.
A
.
P
e
lt
a
,
C.
Cr
u
z
,
T
.
Ge
r
m
a
n
,
a
n
d
N.
Kra
sn
o
g
o
r,
E
d
s.
G
ra
n
a
d
a
:
S
p
ri
n
g
e
r
S
p
rin
g
e
r,
2
0
1
0
,
p
p
.
6
5
–
7
4
.
[9
]
Q.
Ca
i,
M
.
G
o
n
g
,
B.
S
h
e
n
,
L
.
M
a
,
a
n
d
L
.
Jia
o
,
“
Disc
re
te
p
a
rti
c
le
sw
a
r
m
o
p
ti
m
iza
ti
o
n
f
o
r
id
e
n
ti
f
y
in
g
c
o
m
m
u
n
it
y
stru
c
tu
re
s in
sig
n
e
d
so
c
ial
n
e
tw
o
rk
s,”
Ne
u
ra
l
Ne
two
rk
s
,
v
o
l.
5
8
,
p
p
.
4
–
1
3
,
Oc
t.
2
0
1
4
.
[1
0
]
Q.
Ca
i,
M
.
G
o
n
g
,
L
.
M
a
,
S
.
R
u
a
n
,
F
.
Yu
a
n
,
a
n
d
L
.
Jia
o
,
“
G
re
e
d
y
d
isc
re
te
p
a
rti
c
le
sw
a
r
m
o
p
ti
m
iz
a
ti
o
n
f
o
r
larg
e
-
sc
a
le so
c
ial
n
e
tw
o
rk
c
lu
ste
rin
g
,
”
In
fo
rm
a
t
io
n
S
c
ien
c
e
s
,
v
o
l
.
3
1
6
,
p
p
.
5
0
3
–
5
1
6
,
S
e
p
.
2
0
1
5
.
[1
1
]
S
.
S
u
g
a
n
t
h
i
a
n
d
S
.
P
.
Ra
jag
o
p
a
la
n
,
“
M
u
lt
i
-
S
w
a
r
m
P
a
rti
c
le
S
w
a
r
m
Op
ti
m
iza
ti
o
n
f
o
r
En
e
rg
y
-
Eff
e
c
ti
v
e
Clu
ste
rin
g
in
W
irele
ss
S
e
n
so
r
Ne
t
w
o
rk
s,”
W
ire
les
s P
e
rs
o
n
a
l
C
o
mm
u
n
ic
a
ti
o
n
s
,
v
o
l.
9
4
,
n
o
.
4
,
p
p
.
2
4
8
7
–
2
4
9
7
,
Ju
n
.
2
0
1
7
.
[1
2
]
J.
Re
ji
n
a
P
a
rv
in
a
n
d
C.
V
a
sa
n
t
h
a
n
a
y
a
k
i,
“
P
a
rti
c
le
S
w
a
r
m
Op
ti
m
iza
ti
o
n
-
Ba
se
d
Clu
ste
rin
g
b
y
P
re
v
e
n
ti
n
g
Re
sid
u
a
l
No
d
e
s in
W
irele
ss
S
e
n
so
r
Ne
tw
o
r
k
s,”
IEE
E
S
e
n
so
rs
J
o
u
rn
a
l
,
v
o
l.
1
5
,
n
o
.
8
,
p
p
.
4
2
6
4
–
4
2
7
4
,
A
u
g
.
2
0
1
5
.
[1
3
]
S
.
R.
M
a
n
d
a
la,
S
.
R
.
T
.
Ku
m
a
ra
,
C.
R.
Ra
o
,
a
n
d
R.
A
lb
e
rt,
“
Clu
ste
rin
g
so
c
ial
n
e
tw
o
rk
s
u
sin
g
a
n
t
c
o
lo
n
y
o
p
ti
m
iza
ti
o
n
,
”
Op
e
ra
ti
o
n
a
l
Res
e
a
rc
h
,
v
o
l.
1
3
,
n
o
.
1
,
p
p
.
4
7
–
6
5
,
A
p
r.
2
0
1
3
.
[1
4
]
J.
Ji,
X
.
S
o
n
g
,
C.
L
iu
,
a
n
d
X
.
Zh
a
n
g
,
“
A
n
t
c
o
lo
n
y
c
lu
ste
rin
g
w
it
h
f
it
n
e
ss
p
e
rc
e
p
ti
o
n
a
n
d
p
h
e
ro
m
o
n
e
d
if
f
u
sio
n
f
o
r
c
o
m
m
u
n
it
y
d
e
tec
ti
o
n
i
n
c
o
m
p
lex
n
e
tw
o
rk
s,”
P
h
y
sic
a
A:
S
t
a
ti
st
ica
l
M
e
c
h
a
n
ics
a
n
d
it
s
Ap
p
li
c
a
t
io
n
s
,
v
o
l.
3
9
2
,
n
o
.
1
5
,
p
p
.
3
2
6
0
–
3
2
7
2
,
A
u
g
.
2
0
1
3
.
[1
5
]
X
.
Zh
o
u
,
Y.
L
iu
,
J.
Zh
a
n
g
,
T
.
Li
u
,
a
n
d
D.
Zh
a
n
g
,
“
A
n
a
n
t
c
o
lo
n
y
b
a
s
e
d
a
lg
o
rit
h
m
f
o
r
o
v
e
rlap
p
in
g
c
o
m
m
u
n
it
y
d
e
tec
ti
o
n
in
c
o
m
p
lex
n
e
tw
o
rk
s,”
Ph
y
sic
a
A:
S
t
a
ti
stic
a
l
M
e
c
h
a
n
ics
a
n
d
it
s
Ap
p
li
c
a
t
io
n
s
,
v
o
l
.
4
2
7
,
p
p
.
2
8
9
–
3
0
1
,
Ju
n
.
2
0
1
5
.
[1
6
]
P
.
M
o
ra
d
i
a
n
d
M
.
Ro
sta
m
i,
“
In
t
e
g
ra
ti
o
n
o
f
g
ra
p
h
c
lu
ste
rin
g
w
it
h
a
n
t
c
o
lo
n
y
o
p
ti
m
iz
a
ti
o
n
f
o
r
f
e
a
t
u
re
se
lec
ti
o
n
,
”
Kn
o
wled
g
e
-
Ba
se
d
S
y
ste
ms
,
v
o
l.
8
4
,
p
p
.
1
4
4
–
1
6
1
,
2
0
1
5
.
[1
7
]
S
.
G
a
o
,
Y.
W
a
n
g
,
J.
Ch
e
n
g
,
Y.
In
a
z
u
m
i,
a
n
d
Z.
T
a
n
g
,
“
A
n
t
c
o
lo
n
y
o
p
ti
m
iza
ti
o
n
w
it
h
c
lu
ste
rin
g
f
o
r
so
lv
in
g
th
e
d
y
n
a
m
ic l
o
c
a
ti
o
n
ro
u
ti
n
g
p
ro
b
lem
,
”
Ap
p
li
e
d
M
a
t
h
e
ma
ti
c
s a
n
d
Co
m
p
u
t
a
ti
o
n
,
v
o
l.
2
8
5
,
p
p
.
1
4
9
–
1
7
3
,
Ju
l.
2
0
1
6
.
[1
8
]
S
.
P
a
terli
n
i
a
n
d
T
.
Kri
n
k
,
“
Dif
f
e
r
e
n
ti
a
l
e
v
o
lu
ti
o
n
a
n
d
p
a
rti
c
le
sw
a
r
m
o
p
ti
m
isa
ti
o
n
in
p
a
rti
ti
o
n
a
l
c
lu
ste
ri
n
g
,
”
Co
mp
u
t
a
ti
o
n
a
l
S
ta
t
isti
c
s
&
Da
ta
An
a
lys
is
,
v
o
l
.
5
0
,
n
o
.
5
,
p
p
.
1
2
2
0
–
1
2
4
7
,
M
a
r.
2
0
0
6
.
[1
9
]
Y.
Ca
i,
J.
L
iao
,
T
.
W
a
n
g
,
Y.
Ch
e
n
,
a
n
d
H.
T
ian
,
“
S
o
c
ial
lea
rn
in
g
d
iff
e
re
n
ti
a
l
e
v
o
lu
ti
o
n
,
”
I
n
fo
rm
a
ti
o
n
S
c
ien
c
e
s
,
Oc
t
.
2
0
1
6
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
2
,
A
p
r
il 2
0
1
8
:
1
1
2
2
–
1130
1130
[2
0
]
E.
Zo
ra
r
p
a
c
ı
a
n
d
S
.
A
.
Öz
e
l,
“
A
h
y
b
rid
a
p
p
r
o
a
c
h
o
f
d
if
f
e
re
n
ti
a
l
e
v
o
lu
ti
o
n
a
n
d
a
rti
f
icia
l
b
e
e
c
o
l
o
n
y
f
o
r
fe
a
tu
re
se
lec
ti
o
n
,
”
Exp
e
rt S
y
ste
ms
wit
h
A
p
p
li
c
a
ti
o
n
s
,
v
o
l
.
6
2
,
p
p
.
9
1
–
1
0
3
,
No
v
.
2
0
1
6
.
[2
1
]
S
.
J.
Na
n
d
a
a
n
d
G
.
P
a
n
d
a
,
“
A
su
r
v
e
y
o
n
n
a
tu
re
in
s
p
ired
m
e
tah
e
u
ri
stic
a
lg
o
rit
h
m
s
f
o
r
p
a
rti
ti
o
n
a
l
c
lu
ste
rin
g
,
”
S
wa
rm
a
n
d
Evo
l
u
ti
o
n
a
ry
Co
mp
u
ta
t
io
n
,
v
o
l.
1
6
,
p
p
.
1
–
1
8
,
2
0
1
4
.
[2
2
]
D.
L
.
Da
v
ies
a
n
d
D.
W
.
Bo
u
ld
in
,
“
DBIn
d
e
x
:
A
Clu
ste
r
S
e
p
a
ra
ti
o
n
M
e
a
su
re
,
”
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
P
a
tt
e
r
n
An
a
lys
is
a
n
d
M
a
c
h
i
n
e
In
tell
ig
e
n
c
e
,
v
o
l.
P
A
M
I
-
1
,
n
o
.
2
,
p
p
.
2
2
4
–
2
2
7
,
A
p
r.
1
9
7
9
.
[2
3
]
Z.
Bo
u
lo
u
a
rd
,
L
.
Ko
u
tt
i,
A
.
E.
H
a
d
d
a
d
i,
A
.
E.
Ha
d
d
a
d
i,
a
n
d
A
.
F
e
n
n
a
n
,
“
X
EW
G
ra
p
h
:
A
to
o
l
f
o
r
v
i
su
a
li
z
a
ti
o
n
a
n
d
a
n
a
ly
sis
o
f
h
y
p
e
rg
ra
p
h
s
f
o
r
a
c
o
m
p
e
ti
ti
v
e
in
telli
g
e
n
c
e
s
y
st
e
m
,
”
in
S
II
E
2
0
1
5
-
6
t
h
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
“
In
fo
rm
a
ti
o
n
S
y
ste
ms
a
n
d
Eco
n
o
mic
In
telli
g
e
n
c
e
,
”
2
0
1
5
,
p
p
.
6
6
–
7
0
.
[2
4
]
A
.
El
Ha
d
d
a
d
i,
“
F
o
u
il
le
m
u
lt
id
i
m
e
n
sio
n
n
e
ll
e
su
r
les
d
o
n
n
é
e
s
tex
tu
e
ll
e
s
v
isa
n
t
à
e
x
tr
a
ire
les
ré
s
e
a
u
x
so
c
iau
x
e
t
sé
m
a
n
ti
q
u
e
s p
o
u
r
le
u
r
e
x
p
lo
i
tatio
n
v
ia l
a
télé
p
h
o
n
ie m
o
b
il
e
,
”
U
n
iv
e
rsité d
e
T
o
u
lo
u
se
III,
P
a
u
l
S
a
b
a
ti
e
r,
2
0
1
1
.
Evaluation Warning : The document was created with Spire.PDF for Python.