IA
E
S
In
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
Ar
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
(I
J
-
AI
)
Vo
l
.
3
, N
o
.
4
,
De
c
e
m
b
e
r
201
4
, p
p
.
145
~
14
9
IS
S
N
:
2252
-
8938
r
145
Jou
r
n
al
h
om
e
page
:
ht
t
p:
/
/
i
ae
s
j
our
nal
.
c
om
/
onl
i
ne
/
i
nde
x
.
php/
I
J
AI
Pr
e
di
c
t
i
o
n
o
f
D
a
i
l
y
N
e
t
w
o
r
k
T
r
a
f
f
i
c
ba
s
e
d
o
n
R
a
di
a
l
B
a
s
i
s
Func
t
i
o
n
N
e
ur
a
l
N
e
t
w
o
r
k
Ha
v
i
l
u
d
d
i
n
*
,
Im
a
m
T
a
h
y
u
d
i
n
*
*
*
De
p
a
r
t
me
n
t
o
f
C
o
mp
u
t
e
r
Sc
i
e
n
c
e
,
Fa
c
u
l
t
y
o
f
M
a
t
h
e
m
a
t
i
c
a
n
d
N
a
t
u
r
a
l
Sc
i
e
n
c
e
,
M
u
l
a
w
a
r
m
a
n
Un
i
v
e
r
s
i
t
y
, I
n
d
o
n
e
s
ia
**
De
p
a
r
t
me
n
t
o
f
In
fo
rm
a
t
i
o
n
Sy
s
t
em
, S
T
M
I
K
A
MI
K
O
M
P
u
r
w
o
k
e
r
t
o
,
I
n
d
o
n
e
s
i
a
Ar
t
i
c
l
e
I
n
f
o
AB
S
T
RACT
Ar
t
i
c
l
e
h
i
s
t
o
r
y
:
Re
c
e
i
v
e
d
Ju
l
2,
201
4
Re
v
i
s
e
d
N
o
v
8
, 2
0
1
4
Ac
c
e
p
t
e
d
No
v
2
0
, 2
0
1
4
Th
i
s
p
a
p
e
r
p
r
e
s
e
n
t
s
a
n
a
p
p
r
o
a
c
h
f
o
r
p
r
e
d
i
c
t
i
n
g
d
a
i
l
y
n
e
t
w
o
r
k
t
r
a
f
f
i
c
u
s
i
n
g
ar
t
i
f
i
ci
al
n
eu
r
al
n
et
w
o
r
k
s
(
A
N
N
)
,
na
m
e
l
y
r
a
di
a
l
ba
s
i
s
f
unc
t
i
on
ne
ur
a
l
ne
t
w
or
k
(
R
B
F
N
N
)
m
e
t
hod.
Th
e
d
a
t
a
i
s
g
a
i
n
e
d
f
r
o
m
21
-
24
J
une
201
3
(1
9
2
sam
p
l
es
ser
i
es
d
at
a)
in
I
C
T
U
n
it
U
n
iv
e
r
s
ita
s
M
u
la
w
a
r
m
a
n
,
E
a
s
t
K
a
lim
a
n
ta
n
,
In
d
o
n
e
s
i
a
.
T
h
e
re
s
u
l
t
s
o
f
m
e
a
s
u
re
m
e
n
t
a
re
u
s
i
n
g
s
t
a
t
i
s
t
i
c
a
l
an
al
y
si
s,
e.
g
.
s
um
of
s
qua
r
e
e
r
r
or
(
S
S
E
)
,
m
e
a
n
of
s
qua
r
e
e
r
r
or
(
M
S
E
)
,
m
e
a
n
of
pe
r
c
e
nt
a
ge
e
r
r
or
(M
P
E
),
m
e
a
n
o
f
a
b
s
o
l
u
t
e
p
e
rc
e
n
t
a
g
e
e
rro
r
(M
A
P
E
),
a
n
d
m
e
a
n
o
f
a
b
s
o
l
u
t
e
de
vi
a
t
i
on
(
M
A
D
)
.
T
h
e
r
e
s
u
lts
s
h
o
w
th
a
t
va
l
ue
s
ar
e
t
h
e
sam
e
,
wi
t
h
di
f
f
e
r
e
nt
goa
l
s
t
ha
t
ha
ve
be
e
n
s
e
t
ar
e
0.
0
01
,
0.
002
,
an
d
0.
003
,
a
n
d
s
p
r
e
a
d
2
0
0
.
T
h
e
sm
al
l
est
MS
E
v
a
l
u
e
i
n
d
i
c
a
t
e
s
a
g
o
o
d
me
t
h
o
d
fo
r
a
c
c
u
ra
c
y
.
T
h
e
re
fo
re
,
t
h
e
RBF
N
N
m
o
d
e
l
i
l
l
u
s
t
r
a
t
e
s
t
h
e
p
r
o
p
o
s
e
d
b
e
s
t
m
o
d
e
l
t
o
p
r
e
d
i
c
t
d
a
i
l
y
n
e
t
w
o
r
k
tr
a
f
f
ic
.
Ke
y
wo
r
d
:
Ne
t
wo
r
k
t
r
a
f
f
i
c
RBF
N
N
SSE
MS
E
MA
P
E
MA
D
Co
p
y
r
i
ght
©
2014
In
s
t
i
t
u
t
e
o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
r
i
n
g
a
n
d
S
c
i
e
n
c
e
.
Al
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
Co
r
r
e
s
p
o
n
d
i
n
g
Au
t
h
o
r
:
Ha
v
i
l
u
d
d
i
n
,
De
p
a
r
t
me
n
t
o
f
C
o
mp
u
t
e
r
S
c
i
e
n
c
e
,
Fa
c
u
l
t
y
o
f
M
a
t
h
e
m
a
t
i
c
s
a
n
d
N
a
t
u
r
a
l
Sc
i
e
n
c
e
,
Mu
l
a
w
a
r
m
a
n
U
n
i
v
e
r
s
i
t
y
, I
n
d
o
n
e
s
ia
.
Em
a
i
l
:
h
a
v
i
l
u
d
d
i
n
@
u
n
m
u
l
.
a
c
.
i
d
1.
IN
T
R
O
D
U
C
T
IO
N
Th
e
m
a
n
a
g
e
m
e
n
t
o
f
tr
a
f
f
ic
quot
a
is
a
n
im
p
o
r
ta
n
t
p
a
r
t
,
es
p
eci
al
l
y
f
o
r
t
h
e
or
ga
ni
z
a
t
i
ons
t
ha
t
us
e
in
f
o
r
m
a
tio
n
te
c
h
n
o
lo
g
y
.
Su
b
s
e
q
u
e
n
t
l
y
,
f
or
th
e
le
a
d
e
r
s
h
ip
,
ma
n
a
g
e
me
n
t
tr
a
f
f
ic
quot
a
wi
l
l
h
e
l
p
in
m
a
k
in
g
de
c
i
s
i
ons
th
a
t
wi
l
l
b
e
n
e
f
i
t
t
h
e
ef
f
i
ci
en
c
y
an
d
ef
f
ect
i
v
e
fo
r
o
rg
a
n
i
z
a
t
i
o
n
s
in
c
lu
d
in
g
uni
ve
r
s
i
t
i
e
s
.
Th
e
p
r
e
d
i
c
t
i
n
g
act
i
v
i
t
i
es
i
s
a
p
ar
t
o
f
o
r
g
an
i
zat
i
o
n
m
an
ag
em
en
t
.
Su
b
s
e
q
u
e
n
t
l
y
,
t
h
e
d
a
i
l
y
n
e
t
w
o
r
k
t
r
a
f
f
i
c
p
r
e
d
i
c
t
i
o
n
i
s
a
l
s
o
a
pr
oc
e
s
s
of
a
na
l
yz
i
ng
a
nd
de
t
e
r
m
i
ni
ng
t
he
quot
a
of
ba
ndw
i
dt
h
i
n
a
ne
t
w
or
k
in
th
e
f
u
tu
r
e
,
in
w
h
ic
h
a
te
c
h
n
ic
a
l
an
al
y
s
i
s
ap
p
r
o
ach
u
s
ag
e
d
at
a
t
r
af
f
i
c.
F
u
r
t
h
er
m
o
r
e,
t
h
e
p
r
ed
i
ct
i
n
g
t
ech
n
i
q
u
es
u
s
ed
i
n
t
h
e
l
i
t
er
at
u
r
e
can
b
e
cl
as
s
i
f
i
ed
i
n
t
o
t
w
o
cat
eg
o
r
i
es
:
s
t
at
i
s
t
i
cal
an
d
s
o
f
t
-
co
m
p
u
t
i
n
g
m
o
d
el
s
.
T
h
e
s
t
at
i
s
t
i
cal
m
o
d
el
s
in
c
lu
d
e
si
m
p
l
e
re
g
r
es
s
i
o
n
l
i
n
ear
(
S
R
L
)
,
ex
p
o
n
en
t
i
al
s
m
o
o
t
h
i
n
g
,
t
h
e
au
t
o
r
eg
r
es
s
i
v
e
m
o
v
i
n
g
av
er
ag
e
(
A
R
M
A
)
,
au
t
o
r
eg
r
es
s
i
v
e
in
te
g
r
a
te
d
m
o
v
in
g
a
v
e
r
a
g
e
(
A
R
I
M
A
)
a
n
d
g
e
n
e
r
a
liz
e
d
a
u
to
r
e
g
r
e
s
s
iv
e
c
o
n
d
itio
n
a
l
h
e
te
r
o
s
k
e
d
a
s
tic
ity
(G
A
R
C
H
)
m
o
d
e
l
s
.
N
e
v
e
rt
h
e
l
e
s
s
,
t
h
e
s
e
m
o
d
e
l
s
a
re
fo
c
u
s
e
d
a
ro
u
nd
t
he
s
uppos
i
t
i
on
t
ha
t
t
he
da
t
a
of
va
r
i
ous
tim
e
s
e
r
ie
s
lin
e
a
r
ly
c
o
r
r
e
la
te
a
n
d
p
r
o
v
id
e
p
o
o
r
p
r
e
d
ic
tio
n
p
e
r
f
o
r
m
a
n
c
e
[
1
-
4
]
.
Me
a
n
w
h
i
l
e
,
t
h
e
d
a
i
l
y
n
e
t
w
o
r
k
t
r
a
f
f
i
c
da
t
a
a
r
e
nonl
i
ne
a
r
a
nd
non
-
st
a
t
i
o
n
a
r
y
i
n
n
a
t
u
r
e
.
T
o
o
v
e
r
c
o
m
e
th
is
lim
ita
tio
n
,
th
e
s
e
c
o
n
d
m
o
d
e
l
is
s
o
f
t
-
co
m
p
u
t
i
n
g
m
et
h
o
d
s
h
av
e
b
een
s
u
g
g
es
t
ed
.
F
u
r
t
h
er
m
o
r
e,
m
o
d
el
i
n
g
us
i
ng
t
he
a
r
t
i
f
i
c
i
a
l
ne
ur
a
l
ne
t
w
or
k
(
A
N
N
)
m
ode
l
c
a
n
pr
ovi
de
be
t
t
e
r
an
al
y
t
i
cal
r
es
u
l
t
s
,
an
d
i
t
is
e
f
f
e
c
tiv
e
f
o
r
fo
re
c
a
s
t
i
n
g
,
i
n
w
h
i
c
h
t
h
i
s
m
e
t
h
o
d
i
s
a
b
l
e
t
o
w
o
rk
w
e
l
l
o
n
t
h
e
n
o
n
-
lin
e
a
r
tim
e
-
se
r
i
e
s
d
a
t
a
[
3
,
5
-
7
]
.
Th
e
r
e
f
o
r
e
,
th
is
p
a
p
e
r
w
ill
s
tu
d
y
o
n
e
o
f
th
e
A
N
N
m
o
d
e
ls
,
n
a
m
e
ly
th
e
R
a
d
ia
l
B
a
s
is
F
u
n
c
tio
n
N
e
u
r
a
l
N
e
tw
o
r
k
(
R
B
F
N
N
)
,
in
o
r
d
e
r
to
a
d
d
r
e
s
s
th
e
is
s
u
e
o
f
n
e
tw
o
r
k
tr
a
f
f
ic
tim
e
s
e
r
ie
s
d
a
ta
th
a
t
h
a
s
n
o
n
-
lin
e
a
r
c
h
a
r
a
c
te
r
is
tic
s
.
T
h
is
p
a
p
e
r
co
n
s
i
s
t
s
of
fo
u
r
se
c
t
i
o
ns
.
In
t
ro
d
u
c
t
i
o
n
s
e
c
t
i
o
n
is
th
e
m
o
tiv
a
tio
n
to
d
o
th
e
w
r
itin
g
of
t
he
a
r
t
i
c
l
e
.
Ne
x
t
,
t
h
e
m
et
h
o
d
o
l
o
g
y
i
s
de
s
c
r
i
bi
ng
of
m
ode
l
.
T
hi
r
d
s
e
c
t
i
on
i
s
t
he
an
al
y
s
i
s
an
d
d
i
s
cu
s
s
i
o
n
r
es
u
l
t
s
,
an
d
fi
n
a
l
l
y
co
n
cl
u
s
i
o
n
se
c
t
i
o
n
is
re
s
e
a
rc
h
su
m
m
a
r
i
e
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
r
IS
S
N
:
2252
-
8938
IJ
-
AI
Vo
l
.
3
, N
o
.
4
,
De
c
e
m
b
e
r
201
4
:
145
–
149
146
2.
RE
S
E
ARCH
M
E
T
H
O
D
Th
e
R
B
F
N
N
i
s
t
h
e
a
b
b
r
e
v
i
a
t
i
o
n
o
f
r
a
d
i
a
l
b
a
s
i
s
f
u
n
c
t
i
o
n
n
e
u
r
a
l
n
e
t
wo
r
k
wh
i
c
h
i
s
b
a
s
e
d
o
n
t
h
e
fu
n
c
t
i
o
n
a
p
p
ro
x
i
m
a
t
i
o
n
t
h
e
o
ry
o
r
s
u
p
e
rv
i
s
e
d
a
n
d
u
n
s
u
p
e
rv
i
s
e
d
m
a
n
n
e
r
w
e
re
u
s
e
d
t
o
g
e
t
h
e
r.
S
u
b
s
e
q
u
e
n
t
l
y
,
i
t
ha
s
a
uni
que
t
r
a
i
ni
ng
a
l
gor
i
t
hm
c
a
l
l
e
d
hybr
i
d
m
e
t
hod
t
ha
t
e
m
e
r
ge
d
a
s
a
v
a
r
i
a
n
t
o
f
N
N
i
n
l
a
t
e
8
0
’
s.
T
h
i
s
mo
d
e
l
i
s
a
k
i
n
d
o
f
f
e
e
d
-
fo
rw
a
rd
n
e
u
ra
l
n
e
t
w
o
rk
(F
F
N
N
)
i
n
w
h
i
c
h
i
n
c
l
u
d
e
s
a
n
i
n
p
u
t
l
a
y
e
r,
a
h
i
d
d
e
n
l
a
y
e
r,
a
n
d
an
o
u
t
p
u
t
l
ay
er
[
8
]
. A
s
sh
o
w
n
in
F
ig
ur
e
1.
Fi
g
u
r
e
1
.
RBF
n
e
u
r
a
l
n
e
t
w
o
r
k
s
t
r
u
c
t
u
r
e
[
9
]
In
g
e
n
e
ra
l
,
R
B
F
N
N
p
ro
c
e
s
s
t
h
e
fi
rs
t
p
h
a
s
e
i
s
u
n
s
u
p
e
rv
i
s
e
d
l
e
a
rn
i
n
g
b
e
t
w
e
e
n
i
n
p
u
t
l
a
y
e
r
a
n
d
h
i
d
d
e
n
la
y
e
r
th
a
t
n
o
n
-
lin
e
a
r
r
a
d
ia
l
-
ba
s
e
d
a
c
t
i
va
t
i
on
f
unc
t
i
on,
c
om
m
onl
y
G
a
us
s
i
a
n
f
unc
t
i
on.
S
e
c
ond
pha
s
e
i
s
su
p
e
r
v
i
se
d
l
e
a
r
n
i
n
g
b
e
t
w
e
e
n
h
i
d
d
e
n
l
a
y
e
r
a
n
d
o
u
t
p
u
t
l
a
y
e
r
w
i
t
h
a
l
i
n
e
a
r
.
−
=
−
1
−
1
W
he
r
e
1
−
is
th
e
E
u
c
lid
e
a
n
d
is
ta
n
c
e
,
c
is
th
e
c
e
n
te
r
o
f
G
a
u
s
s
ia
n
f
u
n
c
tio
n
=
,
,
…
,
,
…
,
is
th
e
qt
h
in
p
u
t d
a
ta
.
He
n
c
e
,
t
h
e
a
r
c
h
i
t
e
c
t
u
r
e
o
f
R
B
F
NN
a
s
s
h
o
wn
i
n
F
i
g
ur
e
1,
a
nd
t
he
e
qua
t
i
on
i
s
=
.
wh
e
r
e
:
Y
=
o
u
t
p
u
t
v
a
l
u
e
;
φ
=
h
i
d
d
e
n
v
a
l
u
e
;
W
=
w
e
i
g
h
t
s
(
0
-
1)
Th
e
a
l
g
o
r
i
t
h
m
o
f
R
B
F
N
N
t
o
a
n
a
l
y
ze
w
i
t
h
i
n
t
i
m
e
s
er
i
es
d
at
a
ch
ar
act
er
i
s
t
i
cs
i
s
:
1.
In
i
t
i
a
l
i
z
a
t
i
o
n
o
f
t
h
e
n
e
t
w
o
rk
.
2.
De
t
e
r
m
i
n
i
n
g
th
e
in
p
u
t s
ig
n
a
l to
h
id
d
e
n
la
y
e
r
,
an
d
f
i
n
d
is
a
d
is
ta
n
c
e
d
a
ta
to
wh
e
r
e
,
=
1
,
2
,
…
,
=
(
−
)
3.
Fi
n
d
1
is
a
re
s
u
l
t
a
c
t
i
v
a
t
i
o
n
fro
m
d
i
s
t
a
n
c
e
d
a
t
a
m
u
l
t
i
p
l
y
b
i
a
s
.
1
=
(
∗
)
x
1
=
(
.
)
4.
Fi
n
d
w
e
i
g
h
t
a
n
d
b
i
a
s
l
a
y
e
r
s
,
2
an
d
2
, in
e
a
c
h
=
1
,
2
,
…
,
.
a.
De
t
e
r
m
i
n
i
n
g
t
r
a
i
n
i
n
g
s
a
m
p
l
e
s
a
n
d
t
e
s
t
s
a
m
p
l
e
s
.
In
t
h
i
s
s
tu
d
y
,
th
e
d
a
ta
w
e
r
e
c
o
lle
c
te
d
f
r
o
m
I
C
T
s
e
r
v
e
r
U
n
iv
e
r
s
ita
s
M
u
la
w
a
r
m
a
n
.
T
h
e
n
,
th
e
d
a
ta
w
e
r
e
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
IS
S
N
:
2252
-
8938
r
Pr
e
d
i
c
t
i
o
n
o
f
D
a
i
l
y
N
e
t
w
o
r
k
T
r
a
f
f
i
c
b
a
s
e
d
o
n
Ra
d
i
a
l
Ba
s
i
s
Fu
n
c
t
i
o
n
N
e
u
r
a
l
N
e
t
w
o
r
k
(
Ha
v
ilu
d
d
in
)
147
co
l
l
ect
ed
f
r
o
m
21
-
24
J
une
2014
(
192
s
a
m
pl
e
s
s
e
r
i
e
s
da
t
a
)
.
T
h
e
n
,
each
n
et
w
o
r
k
t
r
af
f
i
c
d
at
a
w
as
cap
t
u
r
ed
b
y
th
e
C
A
C
T
I
s
o
f
tw
a
r
e
.
Th
e
d
a
i
l
y
n
e
t
w
o
r
k
t
r
a
f
f
i
c
d
a
t
a
w
a
s
a
n
a
l
y
zed
u
s
i
n
g
M
A
T
L
A
B
R
2
0
1
3
b
.
Si
n
c
e
t
h
e
im
p
lic
it
f
u
n
c
tio
n
o
f
R
B
F
N
N
is
G
a
u
s
s
ia
n
f
u
n
c
tio
n
,
in
w
h
ic
h
g
e
n
e
r
a
l
r
e
q
u
ir
e
s
f
o
r
in
p
u
t
v
a
lu
e
b
e
tw
e
e
n
0
a
n
d
1
.
Th
e
d
a
i
l
y
n
e
t
w
o
r
k
t
r
a
f
f
i
c
d
a
t
a
n
e
e
d
n
o
r
m
a
l
i
z
e
d
u
s
i
n
g
s
t
a
t
i
s
t
i
c
a
l
d
a
t
a
n
o
r
m
a
l
i
z
a
t
i
o
n
,
w
h
i
c
h
i
s
u
s
u
a
l
l
y
ex
p
r
es
s
ed
as
:
=
−
−
W
he
r
e
:
X
i
s
t
he
a
c
t
ua
l
va
l
ue
of
s
a
m
pl
e
;
ta
k
e
s
a
la
r
g
e
v
a
lu
e
,
a
n
d
ta
k
e
s
a
s
a
m
p
le
s
o
f
d
a
ta
is
le
s
s
th
a
n
th
e
m
in
im
u
m
v
a
lu
e
to
e
n
s
u
r
e
n
o
r
m
a
liz
e
d
v
a
lu
e
is
n
o
t
c
lo
s
e
to
0
.
L
a
te
r
,
p
r
o
c
es
s
i
n
v
er
s
e
t
r
an
s
f
o
r
m
to
g
e
t
th
e
a
c
tu
a
l
v
a
lu
e
is
o
b
ta
in
e
d
.
Fu
r
t
h
e
r
m
o
r
e
,
i
n
t
hi
s
e
xpe
r
i
m
e
nt
w
e
us
e
d
t
he
s
um
of
s
qua
r
e
e
r
r
or
(
S
S
E
)
,
me
a
n
o
f
s
q
u
a
r
e
e
r
r
o
r
(
M
S
E
)
,
me
a
n
o
f
a
b
s
o
l
u
t
e
p
e
r
c
e
n
t
a
g
e
e
r
r
o
r
(
M
A
P
E
)
,
a
n
d
me
a
n
o
f
a
b
s
o
l
u
t
e
d
e
v
i
a
t
i
o
n
(M
A
D
)
w
e
re
en
g
ag
ed
th
e
p
r
ed
i
ct
ed
o
u
t
p
u
t
w
i
t
h
t
h
e
d
es
i
r
ed
o
u
t
p
u
t
.
3.
RE
S
UL
T
S
A
ND
ANAL
YS
I
S
Th
i
s
s
e
c
t
i
o
n
p
r
e
s
e
n
t
s
t
h
e
b
e
s
t
a
c
h
i
e
v
e
d
r
e
s
u
l
t
b
y
t
h
e
R
B
F
N
N
a
l
g
o
r
i
t
h
m
.
In
t
h
i
s
e
x
p
e
ri
m
e
n
t
,
t
h
e
tr
a
in
in
g
d
a
ta
w
e
r
e
s
e
le
c
te
d
f
r
o
m
21
-
24
J
une
2014
(
192
s
a
m
pl
e
s
s
e
r
i
e
s
da
t
a
)
w
a
s
c
a
pt
ur
e
d.
Ba
s
e
d
o
n
NNs
ru
l
e
s
,
t
h
e
d
a
t
a
w
e
re
d
i
v
i
d
e
d
i
n
t
o
t
ra
i
n
i
n
g
a
n
d
t
e
s
t
i
n
g
d
a
t
a
t
h
e
n
,
th
e
d
a
ta
s
e
ts
c
o
n
s
is
t
o
f
1
4
4
(
9
0
%
)
s
a
m
p
le
s
f
o
r
da
t
a
t
r
a
i
ni
ng
a
nd
48
(
10%
)
s
a
m
pl
e
s
f
or
d
at
a
t
es
t
i
n
g
o
r
t
w
o
n
eu
r
o
n
s
,
P
=
[p
(t
-
1)
,
p(
t
)
]
,
a
nd
t
he
num
be
r
of
out
put
ne
ur
ons
i
s
one
,
p'
(
t
)
,
a
s
s
h
ow
n
i
n
T
a
bl
e
1
.
In
o
rd
e
r
t
o
t
e
s
t
a
n
d
v
a
l
i
d
a
t
e
t
h
e
d
i
ffe
re
n
t
e
rro
r
g
o
a
l
s
,
fo
u
r
st
a
t
i
st
i
c
a
l
;
S
S
E
,
M
S
E
,
M
A
P
E
,
a
n
d
M
A
D
t
e
st
w
e
r
e
c
a
r
r
i
e
d
o
u
t
.
T
h
e
r
e
s
ul
t
s
obt
a
i
ne
d
a
r
e
s
um
m
a
r
i
z
e
d
i
n
Ta
b
l
e
2
A
a
n
d
2
B
.
F
r
o
m
th
e
s
im
u
la
tio
n
s
c
a
r
r
ie
d
o
u
t,
it
w
a
s
c
r
e
a
te
d
a
p
r
e
c
is
e
n
e
u
ra
l
n
e
t
w
o
rk
b
y
ne
w
r
b
(P
,
T
,
e
r
r
o
r
_
g
o
a
l
,
s
p
r
e
a
d
)
fu
n
c
t
i
o
n
,
w
h
i
c
h
i
s
t
h
i
s
fu
n
c
t
i
o
n
c
re
a
t
e
s
R
B
F
N
N
s
t
ru
c
t
u
re
,
a
u
t
o
m
a
t
i
c
a
l
l
y
s
e
l
e
c
t
e
d
t
h
e
num
be
r
of
hi
dde
n
l
a
ye
r
a
nd
m
a
de
t
he
e
r
r
or
t
o
0.
F
or
t
he
e
r
r
or
goa
l
va
l
ue
s
w
e
r
e
0.
01,
0.
02,
a
nd
0.
03.
W
he
r
e
,
sp
re
a
d
is
th
e
d
e
n
s
ity
o
f
b
a
s
is
f
u
n
c
tio
n
v
a
lu
e
o
f
2
0
0
w
a
s
s
e
ttle
d
.
Th
e
ex
p
er
i
m
en
t
h
as
s
h
o
w
n
v
al
u
es
ar
e
t
h
e
s
am
e,
w
i
t
h
d
i
f
f
er
en
t
g
o
al
s
w
er
e
0
.
0
0
1
,
0
.
0
0
2
,
an
d
0
.
003
ha
ve
be
e
n
s
e
t
.
T
he
va
l
ue
s
in
th
e
tr
a
in
in
g
R
B
F
N
N
a
r
e
S
S
E
v
a
lu
e
is
0
9
8
6
,
M
S
E
v
a
lu
e
is
0
.0
0
7
,
M
A
P
E
v
a
lu
e
is
0.
043,
an
d
M
A
D
v
al
u
e
i
s
0
0
6
3
t
h
en
,
t
h
e
v
al
u
es
of
t
he
R
B
F
N
N
t
e
s
t
i
ng
a
r
e
S
S
E
va
l
ue
i
s
0343,
M
S
E
va
l
ue
i
s
0.
007,
M
A
P
E
va
l
ue
i
s
0.
036,
a
nd
M
A
D
va
l
ue
i
s
0.
064.
Ta
b
l
e
1.
Th
e
D
a
i
l
y
N
e
t
w
o
r
k
D
a
t
a
A
f
t
e
r
N
o
r
m
a
l
i
z
e
d
Gr
o
u
p
In
p
u
t
n
e
u
ro
n
s
P
=
[p
(t
-
1)
,
p(
t
)
]
Ou
t
p
u
t
n
e
u
r
o
n
T
=
p
’
(
t
)]
Gr
o
u
p
In
p
u
t
n
e
u
ro
n
s
P
=
[p
(t
-
1)
,
p(
t
)
]
Ou
t
p
u
t
n
e
u
r
o
n
T
=
p
’
(
t
)
]
t
-
1
t
t+
1
t
-
1
t
t+
1
Gr
o
u
p
T
r
a
i
n
1
0.
20
0.
16
0.
17
Gr
o
u
p
T
e
s
t
142
0.
27
0.
23
0.
16
2
0.
16
0.
17
0.
12
143
0.
14
0.
11
0.
19
3
0.
17
0.
12
0.
07
144
0.
11
0.
19
0.
18
4
0.
12
0.
07
0.
05
145
0.
19
0.
1
8
0.
14
5
0.
07
0.
05
0.
08
146
0.
18
0.
14
0.
08
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
137
0.
41
0.
42
0.
45
184
0.
22
0.
32
0.
31
138
0.
42
0.
45
0.
48
185
0.
32
0.
31
0.
27
139
0.
45
0.
48
0.
43
186
0.
31
0.
27
0.
32
140
0.
48
0.
43
0.
27
187
0.
27
0.
32
0.
20
141
0.
43
0.
27
0.
23
188
0.
32
0.
20
0.
19
Ta
b
l
e
2
A.
T
h
e
R
B
F
NN
T
r
a
i
n
i
n
g
R
e
s
u
l
t
s
RBF
N
N
Tr
a
i
n
i
n
g
SSE
MS
E
MA
P
E
MA
D
er
r
o
r
g
o
al
=
0
.
0
0
1
SPR
EA
D
2
0
0
0.
986
0.
007
0.
043
0.
063
er
r
o
r
g
o
al
=
0
.
0
0
2
SPR
EA
D
200
0.
986
0.
007
0.
043
0.
063
er
r
o
r
g
o
al
=
0
.
0
0
3
SPR
EA
D
2
0
0
0.
986
0.
007
0.
043
0.
063
Evaluation Warning : The document was created with Spire.PDF for Python.
r
IS
S
N
:
2252
-
8938
IJ
-
AI
Vo
l
.
3
, N
o
.
4
,
De
c
e
m
b
e
r
201
4
:
145
–
149
148
Ta
b
l
e
2
B
. T
h
e
R
B
F
N
N
T
e
s
tin
g
R
e
s
u
lts
RBF
N
N
Te
s
t
i
n
g
SSE
MS
E
MA
P
E
MA
D
er
r
o
r
g
o
al
=
0
.
0
0
1
SPR
EA
D
2
0
0
0.
343
0.
007
0.
036
0.
064
er
r
o
r
g
o
al
=
0
.
0
0
2
SPR
EA
D
2
0
0
0
.3
4
3
0.
007
0.
036
0.
064
er
r
o
r
g
o
al
=
0
.
0
0
3
SPR
EA
D
2
0
0
0.
343
0.
007
0.
036
0.
064
4.
CO
NCL
US
I
O
N
In
t
h
i
s
p
a
p
e
r,
t
h
e
a
n
a
l
y
s
i
s
u
s
i
n
g
R
B
F
N
N
t
e
c
h
n
i
q
u
e
t
o
a
c
h
i
e
v
e
t
h
e
m
o
d
e
l
o
f
d
a
i
l
y
n
e
t
w
o
rk
t
ra
ffi
c
act
i
v
i
t
i
es
h
av
e
b
een
co
n
d
u
ct
ed
i
n
t
h
e
I
C
T
U
n
i
t
,
U
n
i
v
er
s
i
t
as
M
u
l
aw
ar
m
an
.
A
cco
r
d
i
n
g
to
F
ig
u
r
e
2
a
nd
3,
t
he
re
s
u
l
t
s
o
f
R
B
F
N
N
t
ra
i
n
i
n
g
s
h
o
w
s
t
h
a
t
fo
r
e
rro
r
g
o
a
l
v
a
l
u
e
i
s
0
.
0
0
1
t
h
e
n
S
S
E
v
a
l
u
e
i
s
0
9
8
6
,
M
S
E
v
a
l
u
e
i
s
0.
007,
M
A
P
E
va
l
ue
i
s
0.
043,
a
nd
M
A
D
va
l
ue
is
0
0
6
3
A
f
te
r
w
a
r
d
,
th
e
v
a
lu
e
s
of
t
he
R
B
F
N
N
t
e
s
t
i
ng
a
r
e
S
S
E
va
l
ue
i
s
0343,
M
S
E
va
l
ue
i
s
0.
007,
M
A
P
E
va
l
ue
i
s
0.
036,
a
nd
M
A
D
va
l
ue
i
s
0.
064.
In
d
i
c
a
t
o
r
t
e
s
t
re
s
u
l
t
o
f
da
t
a
i
s
t
he
s
m
a
l
l
e
s
t
e
r
r
or
va
l
ue
,
w
he
r
e
va
l
ue
i
ndi
c
a
t
i
ng
a
n
e
r
r
or
t
e
s
t
i
ng
i
s
t
he
be
s
t
m
ode
l
[
9
]
.
Th
e
r
e
f
o
r
e
,
t
h
e
d
e
t
e
r
m
i
n
a
t
i
o
n
o
f
t
h
e
b
e
s
t
m
o
d
e
l
i
s
d
e
t
e
r
m
i
n
e
d
b
y
s
e
l
e
c
t
i
n
g
t
h
e
s
m
a
l
l
e
s
t
v
a
l
u
e
o
f
te
s
tin
g
e
r
r
o
r
.
B
a
s
e
d
o
n
th
e
r
es
u
l
t
s
o
f
R
B
F
N
N
ar
e
co
n
s
i
d
er
ed
cl
o
s
er
t
o
t
h
e
act
u
al
v
al
u
e.
I
n
o
t
h
er
w
o
r
d
s
,
t
h
e
RBF
N
N
m
o
d
e
l
w
i
t
h
d
i
f
f
e
r
e
n
t
s
p
r
e
a
d
v
a
l
u
e
s
i
l
l
u
s
t
r
a
t
e
s
t
h
e
p
r
o
p
o
s
e
d
b
e
s
t
m
o
d
e
l
t
o
p
r
e
d
i
c
t
d
a
i
l
y
n
e
t
w
o
r
k
tr
a
f
f
ic
a
c
tiv
itie
s
.
(a
)
e
rro
r
g
o
a
l
0
.
0
0
1
(b
)
e
rro
r
g
o
a
l
0
.
0
0
2
(c
)
e
rro
r
g
o
a
l
0
.
0
0
3
Fi
g
u
r
e
2
.
Th
e
R
B
F
N
N
t
r
a
i
n
i
n
g
r
e
s
u
l
t
s
c
u
r
v
e
s
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
IS
S
N
:
2252
-
8938
r
Pr
e
d
i
c
t
i
o
n
o
f
D
a
i
l
y
N
e
t
w
o
r
k
T
r
a
f
f
i
c
b
a
s
e
d
o
n
Ra
d
i
a
l
Ba
s
i
s
Fu
n
c
t
i
o
n
N
e
u
r
a
l
N
e
t
w
o
r
k
(
Ha
v
ilu
d
d
in
)
149
(a
)
e
rro
r
g
o
a
l
0
.
0
0
1
(b
)
e
rro
r
g
o
a
l
0
.
0
0
2
(
c)
er
r
o
r
g
o
al
0
.
0
0
3
Fi
g
u
r
e
3
.
Th
e
R
B
F
N
N
t
e
s
t
i
n
g
r
e
s
u
l
t
s
c
u
r
v
e
s
ACK
NO
W
L
E
DG
E
M
E
NT
S
Th
i
s
s
t
u
d
y
ha
s
be
e
n
co
m
p
l
et
ed
th
a
n
k
s
t
o
t
h
e
h
e
l
p
an
d
s
u
p
p
o
r
t
fro
m
va
r
i
ous
pa
r
t
i
e
s
th
a
t
can
n
o
t
b
e
me
n
t
i
o
n
e
d
one
by
one
.
Es
p
e
c
i
a
l
l
y
th
a
n
k
y
o
u
to
our
fa
m
i
l
y
,
ch
i
l
d
r
en
an
d
wi
f
e
.
Re
s
e
a
r
c
h
e
r
s
sa
y
a
b
i
g
th
a
n
k
you
t
o
fa
m
i
l
y
of
M
ul
a
w
a
r
m
a
n
U
ni
ve
r
s
i
t
y
a
nd
ST
M
I
K
AM
I
KOM
Pu
r
w
o
k
e
r
t
o
wh
o
h
a
s
g
i
v
e
n
su
p
p
o
r
t
to
co
m
p
l
et
e
th
is
s
tu
d
y
.
Ho
p
e
f
u
l
l
y
th
is
re
s
e
a
rc
h
can
b
e
u
s
ef
u
l
.
RE
F
E
RE
NCE
S
[1
]
Ha
v
i
l
u
d
d
i
n
,
a
n
d
R
.
Al
f
r
e
d
,
“
F
o
r
e
c
a
s
t
i
n
g
Ne
t
wo
r
k
Ac
t
i
v
i
t
i
e
s
Us
i
n
g
AR
I
M
A
M
e
t
h
o
d
,
”
Jo
u
rn
a
l
o
f
A
d
va
n
ces
i
n
Co
m
p
u
t
e
r
Ne
t
wo
r
k
s
,
vol
.
2,
no.
3,
S
e
pt
e
m
be
r
2014,
pp.
173
-
179,
2014.
[2
]
B.
M
a
j
h
i,
M
.
R
o
u
t,
a
n
d
V
.
B
a
g
h
e
l,
“
O
n
th
e
d
e
v
e
lo
p
m
e
n
t
a
n
d
p
e
r
f
o
r
m
a
n
c
e
e
v
a
lu
a
tio
n
o
f
a
m
u
ltio
b
je
c
tiv
e
G
A
-
ba
s
e
d
RBF
a
d
a
p
t
i
v
e
m
o
d
e
l
f
o
r
t
h
e
p
r
e
d
i
c
t
i
o
n
o
f
s
t
o
c
k
i
n
d
i
c
e
s
,
”
Jo
ur
nal
of
K
i
ng
Saud
U
ni
v
e
r
s
i
t
y
-
Co
m
p
u
t
e
r
a
n
d
In
f
o
r
m
a
t
i
o
n
S
c
i
e
n
c
e
s
,
2014.
[3
]
Y.
B
i
r
d
i
,
T
.
A
u
r
o
r
a
,
a
n
d
P
.
A
r
o
r
a
,
“
S
tu
d
y
o
f
A
r
tif
ic
ia
l
N
e
u
r
a
l
N
e
tw
o
r
k
s
a
n
d
N
e
u
r
a
l
I
m
p
la
n
ts
,”
In
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
on
R
e
c
e
nt
and
I
nnov
at
i
on
T
r
e
nds
i
n
C
om
put
i
ng
and
C
om
m
uni
c
at
i
on,
vol
.
1,
no.
4,
2013.
[4
]
U.
Yo
l
c
u
,
E
.
E
g
r
i
o
g
l
u
,
a
n
d
C
.
H.
Al
a
d
a
g
,
“
A
n
e
w
l
i
n
e
a
r
&
n
o
n
l
i
n
e
a
r
ar
t
i
f
i
ci
al
n
eu
r
al
n
et
w
o
r
k
m
o
d
el
f
o
r
t
i
m
e
ser
i
es
fo
re
c
a
s
t
i
n
g
,
”
De
c
i
s
i
o
n
S
u
p
p
o
r
t
S
y
s
t
e
ms
,
vol
.
54,
no.
2013,
pp.
1340
-
1347,
2013.
[5
]
O.
C
l
a
v
e
r
i
a
,
a
n
d
S
.
T
o
r
r
a
,
“
F
o
r
e
c
a
s
t
i
n
g
t
o
u
r
i
s
m
d
e
ma
n
d
t
o
C
a
t
a
l
o
n
i
a
:
Ne
u
r
a
l
n
e
t
wo
r
k
s
v
s
.
t
i
me
s
e
r
i
e
s
mo
d
e
l
s
,
”
Ec
o
n
o
m
i
c
M
o
d
e
l
l
i
n
g,
vol
.
36,
no.
2014,
pp.
220
-
228,
2014.
[6
]
J.
-
Z.
W
a
n
g
,
J
.
-
J.
W
an
g
,
Z
.
-
G.
Z
h
a
n
g
et
a
l
.
,
“
F
o
r
e
c
a
s
tin
g
s
to
c
k
in
d
ic
e
s
w
ith
b
a
c
k
p
r
o
p
a
g
a
tio
n
n
e
u
r
a
l
n
e
tw
o
r
k
,”
Ex
p
e
r
t
S
y
s
t
e
m
s
w
i
t
h
Ap
p
l
i
c
a
t
i
o
n
s
,
vol
.
38,
no.
2011,
pp.
14346
-
14355,
2011.
[7
]
G.
C
h
e
n
,
K.
F
u
,
Z
.
L
i
a
n
g
et
a
l
.
,
“
T
h
e
g
e
n
e
tic
a
lg
o
r
ith
m
b
a
s
e
d
b
a
c
k
p
r
o
p
a
g
a
tio
n
n
e
u
r
a
l
n
e
tw
o
r
k
f
o
r
M
M
P
p
r
e
d
ic
tio
n
in
C
O
2
-
EO
R
p
r
o
c
e
s
s
,
”
Fu
e
l
,
vol
.
126,
no.
2014,
pp.
202
-
212,
2014.
[8
]
W.
J
i
a
,
D
.
Z
h
a
o
,
T
.
S
h
e
n
et
a
l
.
,
“
A
N
e
w
O
p
tim
iz
e
d
G
A
-
RBF
N
e
u
r
a
l
N
e
t
w
o
r
k
A
l
g
o
r
i
t
h
m
,
”
Co
m
p
u
t
a
t
i
o
n
a
l
In
t
el
l
i
g
en
ce
a
n
d
N
eu
ro
sci
en
ce
, n
o
. 2
0
1
4
, p
p
. 1
-
6,
2014.
[9
]
J.
W
u
-
Yu
,
a
n
d
J
.
Yu
,
“
R
a
i
n
f
a
l
l
t
i
me
s
e
r
i
e
s
f
o
r
e
c
a
s
t
i
n
g
b
a
s
e
d
o
n
M
o
d
u
l
a
r
R
B
F
Ne
u
r
a
l
Ne
t
wo
r
k
mo
d
e
l
c
o
u
p
l
e
d
wi
t
h
SSA
a
n
d
PLS,
”
Jo
u
rn
a
l
o
f
T
h
eo
ret
i
ca
l
a
n
d
A
p
p
l
i
ed
C
o
m
p
u
t
er
S
ci
en
ce,
vol
.
6,
no.
2,
pp.
3
-
12,
2
012
.
Evaluation Warning : The document was created with Spire.PDF for Python.