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.
10
,
No
.
5
,
Octo
b
er
2
0
2
0
,
p
p
.
5
5
0
7
~
5
5
1
3
I
SS
N:
2
0
8
8
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v
1
0
i
5
.
pp
5
5
0
7
-
5
5
1
3
5507
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m/in
d
ex
.
p
h
p
/I
JE
C
E
Co
m
pa
ra
tive s
tu
dy
on
m
a
chi
ne
l
e
a
rning
alg
o
rith
m
s for
ea
rly
f
ire
f
o
rest
d
etec
ti
o
n
s
y
ste
m
using
g
eo
da
t
a
Z
o
uite
n M
o
ha
mm
ed
1
,
C
ha
a
o
ua
n H
a
na
e
2
,
Set
t
i La
rbi
3
1
De
p
a
r
tme
n
t
L
a
b
o
ra
to
ry
EH3
D,
F
a
c
u
lt
y
P
o
ly
d
isc
ip
li
n
a
ry
TA
Z
A
,
Un
iv
e
rsit
y
S
id
i
M
o
h
a
m
e
d
b
e
n
A
b
d
e
lah
o
f
F
EZ
,
M
o
ro
c
c
o
2,
3
L
a
b
o
ra
to
ry
S
c
ien
c
e
s a
n
d
A
d
v
a
n
c
e
d
T
e
c
h
n
o
lo
g
ies
,
F
a
c
u
lt
y
p
o
ly
d
isc
ip
li
n
a
ry
o
f
L
a
ra
c
h
e
,
Un
iv
e
rsit
y
A
b
d
e
lm
a
le
k
Essa
d
i
T
e
to
u
a
n
,
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
Feb
27
,
2
0
20
R
ev
i
s
ed
A
p
r
2
5
,
2
0
2
0
A
cc
ep
ted
Ma
y
1
0
,
2
0
2
0
F
o
re
st
f
ires
h
a
v
e
b
e
c
o
m
e
a
g
re
a
t
risk
f
o
r
c
o
u
n
tri
e
s.
T
o
m
in
i
m
ize
th
e
ir
im
p
a
c
t
a
n
d
p
re
v
e
n
t
t
h
is
p
h
e
n
o
m
e
n
o
n
,
sc
ien
ti
f
ic
m
e
th
o
d
s
h
a
v
e
e
m
e
r
g
e
d
.
No
ta
b
ly
m
a
c
h
in
e
lea
rn
in
g
a
lg
o
rit
h
m
s
a
n
d
d
e
c
isio
n
-
m
a
k
in
g
Ge
o
g
ra
p
h
ica
l
In
f
o
r
m
a
ti
o
n
S
y
st
e
m
s.
T
h
e
re
f
o
re
,
a
c
o
m
p
e
ti
ti
v
e
s
p
a
ti
a
l
p
re
d
ictio
n
m
o
d
e
l
f
o
r
e
a
rl
y
f
ire
f
o
re
st
d
e
tec
ti
o
n
sy
ste
m
u
sin
g
g
e
o
d
a
ta
c
a
n
b
e
p
ro
p
o
s
e
d
.
T
h
is
m
o
d
e
l
c
a
n
h
e
l
p
re
se
a
rc
h
e
rs
to
p
re
d
ict
f
o
re
st
f
ires
a
n
d
i
d
e
n
ti
f
y
ris
k
z
o
n
a
s.
S
y
s
tem
u
sin
g
m
a
c
h
in
e
lea
rn
in
g
a
lg
o
rit
h
m
o
n
g
e
o
d
a
ta
w
il
l
b
e
a
b
le
t
o
n
o
ti
fy
in
re
a
l
ti
m
e
th
e
in
tere
ste
d
p
a
rts
a
n
d
a
u
th
o
rit
ies
b
y
p
ro
v
id
in
g
a
lerts
a
n
d
p
re
se
n
ti
n
g
o
n
m
a
p
s
b
a
se
d
o
n
g
e
o
g
ra
p
h
ica
l
trea
tme
n
ts
f
o
r
m
o
re
e
ff
ic
a
c
it
y
a
n
d
a
n
a
l
y
z
in
g
o
f
th
e
situ
a
ti
o
n
.
T
h
is
re
se
a
r
c
h
e
x
t
e
n
d
s
th
e
a
p
p
li
c
a
ti
o
n
o
f
m
a
c
h
in
e
lea
rn
in
g
a
lg
o
rit
h
m
s
f
o
r
e
a
rl
y
f
ire
f
o
re
st
p
re
d
ictio
n
to
d
e
tec
ti
o
n
a
n
d
re
p
re
s
e
n
tatio
n
in
g
e
o
g
ra
p
h
ica
l
in
f
o
rm
a
ti
o
n
sy
ste
m
(G
IS
)
m
a
p
s.
K
ey
w
o
r
d
s
:
Fire
f
o
r
est
d
etec
tio
n
Ma
ch
i
n
e
l
ea
r
n
i
n
g
R
an
d
o
m
f
o
r
est
Su
p
p
o
r
t
v
ec
to
r
m
ac
h
i
n
e
Vo
r
o
n
o
i
Co
p
y
rig
h
t
©
2
0
2
0
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
o
u
i
ten
Mo
h
a
m
m
ed
,
Dep
ar
t
m
en
t o
f
Geo
g
r
ap
h
y
FP
T
T
A
Z
A
,
Un
i
v
er
s
it
y
Sid
i M
o
h
a
m
m
ed
B
en
A
b
d
elah
o
f
Fez,
FP
T
B
P
1
2
2
3
,
T
A
Z
A
,
M
o
r
o
cc
o
.
E
m
ail:
m
o
h
a
m
m
ed
.
zo
u
ite
n
@
u
s
m
b
a.
ac
.
m
a
1.
I
NT
RO
D
UCT
I
O
N
Fo
r
est
f
ir
es
h
a
v
e
b
ec
o
m
e
a
cr
is
is
f
ac
to
r
in
th
e
w
o
r
ld
.
T
o
m
ea
s
u
r
e
th
e
i
m
p
ac
t
o
f
f
o
r
est
f
ir
es
,
co
u
n
tr
ies
to
u
ch
th
e
ec
o
n
o
m
ic,
h
u
m
an
,
e
n
v
ir
o
n
m
e
n
tal,
s
o
cial
s
id
e
f
ac
t
o
r
s
.
Sev
er
al
ca
u
s
es
ar
e
at
t
h
e
r
o
o
t
o
f
th
i
s
p
r
o
b
le
m
lik
e
ig
n
o
r
an
ce
o
f
p
eo
p
le
w
h
o
h
a
v
e
co
n
tact
w
i
th
f
o
r
est
s
,
g
l
o
b
al
w
ar
m
in
g
an
d
n
at
u
r
al
f
ac
to
r
s
.
I
n
p
ar
ticu
lar
,
P
o
r
tu
g
al
is
a
n
af
f
ec
ted
co
u
n
t
r
y
b
y
t
h
is
k
i
n
d
o
f
d
is
a
s
ter
[
1
]
.
B
etw
ee
n
1
9
8
0
an
d
2
0
0
5
,
al
m
o
s
t
2
.
7
m
i
llio
n
h
ec
tar
es
o
f
f
o
r
est
w
er
e
d
estro
y
ed
.
I
n
p
ar
ticu
lar
,
th
e
f
ir
e
s
o
f
2
0
0
3
an
d
2
0
0
5
w
h
ich
a
f
f
e
cted
4
.
6
%
an
d
3
.
1
%
o
f
th
e
ter
r
ito
r
y
an
d
w
h
ich
w
er
e
t
r
ag
ic
[
2
]
.
P
r
ed
ictin
g
th
i
s
p
h
e
n
o
m
e
n
o
n
i
s
th
e
s
o
lu
tio
n
to
m
i
n
i
m
ize
t
h
e
d
a
m
a
g
e.
As
a
r
es
u
lt,
h
u
m
a
n
in
ter
v
en
t
io
n
alo
n
e
is
i
n
s
u
f
f
icie
n
t.
T
h
er
ef
o
r
e,
it
is
n
ec
es
s
ar
y
to
r
ely
o
n
tech
n
o
lo
g
ica
l
to
o
ls
[
3
]
:
s
atelli
tes,
to
p
o
g
r
ap
h
y
d
r
o
n
e
s
an
d
s
en
s
o
r
s
.
E
ac
h
co
u
n
tr
y
c
h
o
o
s
es
t
h
e
ap
p
r
o
p
r
iat
e
m
eth
o
d
ac
co
r
d
in
g
to
th
ese
m
ea
n
s
.
Ot
h
er
m
ea
n
s
ca
n
also
b
e
u
s
ed
to
m
ea
s
u
r
e
n
o
n
-
s
ta
tio
n
ar
y
f
ac
to
r
s
s
u
c
h
a
s
m
eteo
r
o
lo
g
y
[
4
]
.
P
o
r
tu
g
al
h
as 1
6
2
s
tatio
n
s
p
r
o
v
id
in
g
d
ata
to
b
e
an
al
y
ze
d
i
n
r
ea
l ti
m
e
[
5
]
.
Fo
r
est
w
ea
t
h
er
s
tr
u
c
tu
r
es
p
r
o
v
id
e
n
u
m
er
ical
i
n
d
ices
f
o
r
p
r
ev
en
t
in
g
an
d
w
ar
n
i
n
g
p
r
o
b
ab
i
lit
y
o
f
f
ir
e
s
u
c
h
a
s
t
h
e
C
an
ad
ia
n
Fire
W
ea
th
er
I
n
d
ex
(
FW
I
)
[
6
]
.
I
t
is
a
s
y
s
te
m
f
o
r
i
n
d
ex
e
s
ca
l
cu
latio
n
s
b
ased
o
n
:
te
m
p
er
atu
r
e,
r
elati
v
e
h
u
m
id
it
y
,
r
ain
,
etc.
)
.
T
h
is
s
y
s
te
m
i
s
n
o
t
o
n
l
y
u
s
ed
i
n
C
an
ad
a
b
u
t
it
h
a
s
also
b
ee
n
u
s
ed
i
n
s
o
m
e
E
u
r
o
p
ea
n
co
u
n
tr
ies
in
c
l
u
d
in
g
P
o
r
tu
g
a
l
[
7
]
.
r
ec
en
tl
y
t
h
ese
in
d
ices
h
a
v
e
b
ec
o
m
e
p
ar
t
o
f
m
eteo
r
o
lo
g
ical
d
atab
ases
.
T
h
e
y
ar
e
s
u
b
j
ec
t st
u
d
y
f
o
r
d
i
f
f
er
e
n
t c
o
n
te
x
ts
,
th
e
m
ai
n
o
n
e
b
ein
g
th
e
ex
tr
ac
tio
n
o
f
k
n
o
w
led
g
e
f
r
o
m
d
ata
b
ased
o
n
th
e
n
o
tio
n
s
o
f
d
ata
m
i
n
i
n
g
[
8
]
.
C
er
tai
n
l
y
,
t
h
e
s
e
d
atab
ases
ar
e
v
er
y
i
m
p
o
r
tan
t
b
u
t f
ac
ed
w
it
h
t
h
eir
v
o
lu
m
es
ar
e
litt
le
e
x
p
lo
ited
.
T
h
er
ef
o
r
e,
d
ec
is
io
n
m
ak
er
s
m
u
s
t
n
o
t
b
e
s
ati
s
f
ied
w
it
h
s
i
m
p
le
s
tatis
tical
an
al
y
s
es.
F
o
r
an
al
y
s
i
n
g
a
n
d
u
n
d
er
s
ta
n
d
in
g
o
f
t
h
is
p
h
e
n
o
m
en
o
n
.
E
m
er
g
i
n
g
m
ac
h
i
n
e
lear
n
in
g
m
eth
o
d
s
w
ill
r
ep
lace
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.
10
,
No
.
5
,
Octo
b
e
r
2
0
2
0
:
5
5
0
7
-
5513
5508
co
n
v
e
n
tio
n
al
m
et
h
o
d
s
.
Ma
ch
in
e
lear
n
i
n
g
alg
o
r
it
h
m
s
ar
e
u
s
ed
f
o
r
tr
ain
i
n
g
an
d
lear
n
i
n
g
al
g
o
r
ith
m
s
f
o
r
p
r
ed
ictio
n
o
f
f
ir
es
f
o
r
ests
[
9
]
.
Usi
n
g
m
ac
h
in
e
lear
n
in
g
a
l
g
o
r
ith
m
s
,
a
p
r
ed
ictiv
e
m
o
d
el
b
as
ed
o
n
g
eo
g
r
ap
h
ical
an
d
m
ete
o
r
o
lo
g
ical
ex
p
licati
v
e
v
ar
iab
les
ca
n
p
r
ed
ict
f
ir
e
p
r
o
p
ag
atio
n
ar
ea
f
r
o
m
a
lo
ca
lized
f
ir
e
o
n
e.
Se
v
er
al
m
ac
h
in
e
lear
n
i
n
g
alg
o
r
ith
m
s
ar
e
u
s
ed
i
n
o
u
r
a
p
p
r
o
ac
h
an
d
co
m
p
ar
ed
f
o
llo
w
i
n
g
a
r
eg
r
es
s
io
n
o
f
d
a
ta
co
llected
f
r
o
m
co
n
tr
o
l
zo
n
as.
Ov
er
t
h
e
p
ast
y
ea
r
s
,
s
ev
er
al
s
t
u
d
ies
h
a
v
e
b
ee
n
c
o
n
d
u
cted
o
n
f
ir
e
d
etec
tio
n
.
An
ad
ap
tiv
e
f
la
m
e
r
ec
o
g
n
itio
n
an
d
s
e
g
m
en
tatio
n
alg
o
r
ith
m
w
a
s
u
s
ed
to
d
etec
t
f
ir
e
d
etec
tio
n
i
n
lar
g
e
b
u
ild
i
n
g
s
[
1
0
]
.
A
n
i
m
a
g
e
p
r
o
ce
s
s
in
g
m
et
h
o
d
w
a
s
test
ed
to
d
etec
t
s
m
o
k
e
in
v
id
eo
s
[
1
1
]
.
T
h
e
f
ield
o
f
o
p
tical
r
em
o
t
e
s
en
s
i
n
g
h
a
s
s
ee
n
m
u
c
h
p
r
o
g
r
es
s
.
Ob
j
ec
t
d
etec
tio
n
f
r
o
m
i
m
a
g
es
h
as
b
ec
o
m
e
m
o
r
e
ac
ce
s
s
ib
le
[
1
2
]
.
I
n
th
i
s
w
o
r
k
w
e
ar
e
u
s
i
n
g
d
if
f
er
e
n
t
m
ac
h
in
e
lear
n
i
n
g
al
g
o
r
ith
m
s
f
o
r
f
o
r
est
f
ir
e
p
r
ed
icti
o
n
.
T
h
e
f
ir
s
t
o
n
e
i
s
t
h
e
S
u
p
p
o
r
t
Vec
to
r
Ma
ch
i
n
e
w
h
ic
h
is
a
s
u
p
er
v
i
s
ed
class
i
f
i
ca
tio
n
m
o
d
el.
T
h
e
r
eg
r
ess
io
n
m
e
th
o
d
is
co
n
s
id
er
ed
to
b
e
m
o
r
e
e
f
f
ic
ien
t
a
n
d
m
o
r
e
s
u
itab
le
f
o
r
f
o
r
est f
ir
es g
iv
en
t
h
e
d
iv
is
io
n
in
to
cl
u
s
ter
s
o
f
all
th
e
ar
ea
s
li
k
el
y
to
b
e
af
f
ec
ted
[
1
3
]
.
W
e
also
m
ak
e
co
m
p
ar
is
o
n
w
i
th
d
ec
is
io
n
tr
ee
s
a
n
d
n
eu
r
o
n
al
n
et
w
o
r
k
s
w
h
ic
h
ar
e
w
id
el
y
u
s
ed
in
o
u
r
co
n
tex
t
a
n
d
th
at
s
ev
er
al
r
ec
e
n
t
s
t
u
d
ies
h
a
v
e
s
h
o
w
n
t
h
eir
p
er
f
o
r
m
a
n
ce
co
m
p
ar
ed
to
o
th
er
m
et
h
o
d
s
[
1
4
]
.
H
y
b
r
id
Me
t
h
o
d
s
ar
e
s
u
itab
le
f
o
r
f
o
r
est
f
ir
es
b
ec
a
u
s
e
th
e
y
u
s
e
s
i
m
u
lta
n
eo
u
s
l
y
t
h
e
co
n
ce
p
ts
o
f
clas
s
i
f
icatio
n
an
d
r
eg
r
es
s
io
n
:
n
ai
v
e
B
a
y
e
s
a
n
d
d
ec
is
io
n
tr
ee
s
w
h
ic
h
i
n
cr
ea
s
es
t
h
e
p
r
ec
is
io
n
o
f
th
i
s
m
et
h
o
d
.
A
m
o
b
ile
a
g
e
n
t
in
a
w
ir
ele
s
s
s
e
n
s
o
r
n
et
w
o
r
k
co
u
ld
b
e
u
s
ed
to
p
r
ed
i
ct
f
o
r
est
f
ir
es
d
u
r
in
g
th
e
ir
s
u
r
v
eilla
n
ce
[
1
5
]
.
Fin
all
y
,
w
e
a
ls
o
d
is
c
u
s
s
ev
e
n
t
d
etec
tio
n
w
h
ich
r
eq
u
ir
es
a
d
i
f
f
er
en
t
m
et
h
o
d
o
f
cl
u
s
ter
i
n
g
a
n
d
Su
p
p
o
r
t
Vec
to
r
Ma
ch
i
n
e
(
SVM)
r
elati
n
g
to
t
h
e
p
r
o
p
ag
atio
n
o
f
f
o
r
est
f
ir
es
f
o
llo
w
i
n
g
a
f
ir
e
s
tar
ted
[
1
6
]
.
R
ec
e
n
tl
y
s
e
v
er
al
h
ec
tar
es
o
f
f
o
r
est
ar
e
t
h
r
ea
ten
ed
b
y
f
o
r
es
t
f
ir
es.
T
h
is
is
d
u
e
to
s
ev
er
al
f
ac
to
r
s
.
W
e
e
s
p
ec
iall
y
f
o
c
u
s
o
n
th
e
n
e
g
lect
o
f
f
o
r
est
u
s
er
s
,
p
o
llu
tio
n
,
g
lo
b
al
w
ar
m
i
n
g
a
n
d
o
th
er
en
v
ir
o
n
m
en
tal
f
ac
to
r
s
[
1
7
-
1
9
]
.
M
o
d
ellin
g
th
is
t
y
p
e
o
f
p
h
e
n
o
m
e
n
o
n
is
n
o
t
al
w
a
y
s
a
n
ea
s
y
t
h
i
n
g
.
T
h
e
ca
u
s
es
co
n
s
tit
u
te
n
o
n
-
lin
ea
r
v
ec
to
r
s
f
o
r
th
e
tr
an
s
f
o
r
m
atio
n
i
n
to
a
m
o
d
el
g
iv
e
n
t
h
e
p
ar
ticu
lar
itie
s
an
d
th
e
d
iv
er
s
it
y
o
f
t
h
ese
f
ac
to
r
s
[
2
0
]
.
Sev
er
al
d
i
s
cip
lin
e
s
ca
n
co
m
e
in
to
p
la
y
f
o
r
t
h
e
tr
ea
t
m
en
t
o
f
t
h
i
s
k
in
d
o
f
p
r
o
b
le
m
s
.
A
s
a
r
esu
l
t
o
f
th
e
i
n
ter
s
ec
tio
n
o
f
co
m
p
u
ter
s
cien
ce
;
g
eo
g
r
ap
h
y
,
g
eo
lo
g
y
,
p
h
y
s
ics
a
n
d
s
tati
s
tics
;
i
s
a
m
e
an
s
f
o
r
o
p
ti
m
izi
n
g
th
e
r
es
u
lt
s
o
b
tain
ed
[
2
1
-
2
5
]
.
I
n
p
ar
ticu
lar
,
f
o
r
f
o
r
est
f
ir
e
s
an
d
g
iv
e
n
t
h
eir
co
m
p
le
x
a
n
d
s
p
a
tio
te
m
p
o
r
al
n
at
u
r
e;
m
ac
h
in
e
lear
n
i
n
g
al
g
o
r
ith
m
s
p
r
o
v
e
to
b
e
th
e
m
o
s
t
j
u
d
icio
u
s
m
ea
n
s
[
2
6
]
.
T
h
e
liter
atu
r
e
c
o
n
tain
s
ca
s
es
u
s
in
g
ar
tif
icial
n
e
u
r
al
n
et
w
o
r
k
s
[
21
,
2
7
,
2
8
]
r
an
d
o
m
f
o
r
ests
(
R
F)
[
2
9
-
3
1
]
o
t
h
er
s
u
s
e
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
i
n
e
(
SVM)
[
3
2
]
,
th
e
p
er
ce
p
tr
o
n
m
u
lt
ila
y
er
n
e
u
r
al
n
et
w
o
r
k
(
ML
P
)
[
2
8
,
3
3
]
lo
g
is
tic
r
eg
r
e
s
s
io
n
o
f
th
e
n
u
cle
u
s
(
KL
R
)
[
3
4
,
3
5
]
Na
iv
e
B
ay
e
s
[
3
6
,
3
7
]
.
A
s
tu
d
y
p
a
n
o
r
a
m
a
w
a
s
also
s
t
u
d
ied
to
s
h
o
w
t
h
e
p
o
ten
tial
o
f
ea
c
h
o
f
th
e
m
et
h
o
d
s
[
2
0
,
31
,
38
-
4
0
]
.
T
h
er
ef
o
r
e,
it
is
c
lear
t
h
at
t
h
e
m
et
h
o
d
s
m
en
t
io
n
ed
ab
o
v
e
ar
e
th
e
m
o
s
t
s
u
i
tab
le
f
o
r
s
o
lv
i
n
g
th
e
p
r
o
b
le
m
s
o
f
f
o
r
es
t
f
ir
es,
f
o
r
est
f
ir
es
in
p
ar
tic
u
la
r
g
iv
e
n
t
h
e
p
o
s
s
ib
il
it
y
o
f
a
n
al
y
s
i
n
g
t
h
e
p
ix
el
s
o
f
th
e
i
m
ag
e
s
[
4
1
]
.
I
n
ad
d
itio
n
,
w
i
th
o
u
t
a
n
y
e
x
tr
a
ctio
n
o
f
t
h
e
e
n
titi
e
s
,
t
h
e
cla
s
s
if
ier
s
d
ir
ec
tl
y
u
s
e
t
h
e
in
p
u
t
d
ata
w
h
ich
ac
ts
d
ir
ec
tl
y
an
d
p
o
s
itiv
e
l
y
o
n
th
e
ac
cu
r
ac
y
o
f
th
e
clas
s
if
ica
tio
n
.
Fo
r
m
u
ch
m
o
r
e
c
o
m
p
le
x
p
r
o
b
lem
s
,
s
y
s
te
m
p
er
f
o
r
m
a
n
ce
ca
n
b
e
i
m
p
r
o
v
ed
b
y
u
s
i
n
g
lear
n
in
g
-
to
-
lear
n
(
DL
)
f
o
r
t
h
e
i
m
p
r
e
s
s
i
v
e
r
esu
lt
s
t
h
at
ca
n
b
e
o
b
tain
ed
[
4
2
,
4
3
]
.
T
h
is
d
ee
p
lear
n
in
g
g
o
es
f
u
r
t
h
er
t
h
an
th
e
u
s
e
o
f
i
m
ag
er
y
to
al
s
o
r
ea
ch
th
e
r
ec
o
g
n
itio
n
o
f
o
b
j
ec
ts
,
s
o
u
n
d
s
w
h
i
c
h
w
ill
cle
ar
l
y
h
elp
i
n
o
p
ti
m
izi
n
g
t
h
e
p
r
ed
ictio
n
p
r
esen
ted
f
o
r
o
u
r
p
r
o
b
le
m
o
f
f
o
r
est
f
ir
es
[
4
3
]
.
T
h
e
co
n
v
o
lu
tio
n
a
l
n
eu
r
a
l
n
et
w
o
r
k
(
C
NN)
is
o
n
e
o
f
th
e
m
o
s
t
f
o
r
m
id
ab
le
d
ee
p
lear
n
in
g
al
g
o
r
ith
m
s
f
o
r
f
o
r
est
f
ir
es
[
4
4
,
4
5
]
ch
ar
ac
ter
ized
b
y
a
b
etter
c
lass
if
icati
on
o
f
r
e
m
o
te
s
e
n
s
ed
i
m
a
g
e
s
[
4
1
,
4
6
]
as
w
el
l
a
s
ca
r
to
g
r
ap
h
y
s
e
n
s
i
tiv
it
y
to
ter
r
estrial
tr
an
s
lat
io
n
s
[
4
7
]
.
Un
f
o
r
tu
n
atel
y
,
n
o
n
e
o
f
th
ese
s
tu
d
ies
h
as
ev
a
lu
ate
d
C
NN
'
s
p
er
f
o
r
m
an
ce
i
n
p
r
ed
ictin
g
f
o
r
est
f
ir
es.
T
h
e
f
ir
s
t
la
w
o
f
g
eo
g
r
ap
h
y
[
4
8
]
f
o
cu
s
es
o
n
t
h
e
p
ix
els,
o
n
th
e
o
th
er
h
a
n
d
f
o
r
f
o
r
est
f
ir
es
ea
ch
p
ix
e
l
o
f
f
ir
es
is
a
s
p
ar
k
,
in
a
s
p
an
o
f
ti
m
e
th
e
p
ix
el
ca
n
g
e
n
er
ate
ad
j
ac
en
t
p
ix
els
[
4
1
]
.
T
h
e
p
er
f
o
r
m
an
ce
o
f
t
h
e
p
r
o
p
o
s
ed
m
o
d
el
tes
ted
u
s
i
n
g
Ma
t
h
w
o
r
k
s
an
d
T
o
o
ls
B
o
x
,
w
h
ic
h
is
a
n
en
v
ir
o
n
m
en
t f
o
r
t
h
e
co
n
s
tr
u
ctio
n
a
n
d
ev
al
u
atio
n
o
f
m
ac
h
in
e
lear
n
i
n
g
al
g
o
r
ith
m
s
.
2.
M
E
T
H
O
D
T
h
is
s
t
u
d
y
u
s
ed
t
h
e
co
m
b
i
n
a
tio
n
o
f
GI
S
an
d
al
g
o
r
ith
m
s
o
f
m
ac
h
i
n
e
lear
n
i
n
g
to
d
etec
t
o
r
p
r
ed
ict
a
s
p
atio
te
m
p
o
r
al
d
y
n
a
m
ic
s
o
f
f
ir
e
Fo
r
est
A
r
ea
v
u
l
n
er
ab
ilit
y
in
t
h
e
n
o
r
t
h
ea
s
ter
n
r
eg
io
n
o
f
P
o
r
tu
g
al.
No
r
th
e
r
n
P
o
r
tu
g
al
is
th
e
m
o
s
t
p
o
p
u
lo
u
s
r
eg
io
n
i
n
P
o
r
tu
g
al
,
ah
ea
d
o
f
L
is
b
o
an
,
an
d
t
h
e
th
ir
d
m
o
s
t
ex
ten
s
iv
e
b
y
ar
ea
.
A
ca
r
to
g
r
ap
h
ic
r
ep
r
esen
ta
tio
n
b
y
f
u
zz
y
s
u
r
f
ac
es
f
o
r
a
f
o
r
est
r
eg
io
n
w
as
d
ev
el
o
p
ed
an
d
ev
alu
ated
b
y
co
m
p
ar
i
n
g
t
h
e
g
r
o
u
n
d
tr
u
t
h
f
o
r
t
w
o
f
o
r
est
p
ar
a
m
e
ter
s
:
t
h
e
b
asal
s
u
r
f
ac
e
a
n
d
p
o
p
u
latio
n
.
T
h
e
r
ep
r
esen
tatio
n
b
ased
o
n
Vo
r
o
n
o
i
alg
o
r
ith
m
an
d
b
lu
r
r
ed
s
u
r
f
ac
es
w
itc
h
es
m
ak
e
s
b
etter
est
i
m
a
tio
n
o
f
t
h
ese
v
ar
iab
les
th
a
n
a
co
n
v
en
tio
n
al
th
e
m
ati
c
m
ap
as
s
h
o
w
n
i
n
Fi
g
u
r
e
1
.
Si
n
ce
t
h
e
r
ep
r
esen
tatio
n
b
ased
o
n
b
l
u
r
r
ed
s
u
r
f
ac
es
o
f
f
er
s
th
e
p
o
s
s
ib
ilit
y
o
f
h
ig
h
li
g
h
ti
n
g
lo
ca
l
v
ar
ia
tio
n
s
a
n
d
o
f
r
ep
r
e
s
en
ti
n
g
t
h
e
b
o
r
d
er
s
b
et
w
ee
n
f
o
r
est
t
y
p
es
s
u
c
h
as
tr
an
s
itio
n
zo
n
e
s
,
it
s
ee
m
s
th
a
t
it
co
n
s
t
itu
tes
a
r
ep
r
esen
ta
t
io
n
o
f
t
h
e
r
ea
l
w
o
r
ld
w
h
ic
h
i
s
m
o
r
e
r
ea
li
s
tic
a
n
d
m
o
r
e
u
s
e
f
u
l
t
h
a
n
co
n
v
en
t
io
n
a
l
th
e
m
atic
m
ap
s
.
As
a
r
es
u
lt,
s
y
s
te
m
u
s
er
s
w
il
l
b
e
ab
le
to
an
al
y
ze
d
ata
th
at
is
clo
s
er
to
ac
tu
al
f
ield
co
n
d
itio
n
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
C
o
mp
a
r
a
tive
s
tu
d
y
o
n
ma
c
h
in
e
lea
r
n
in
g
a
lg
o
r
ith
ms fo
r
ea
r
ly
fir
e
fo
r
es
t d
etec
tio
n
.
..
(
Zo
u
iten
Mo
h
a
mme
d
)
5509
Fig
u
r
e
1
.
Vo
r
o
n
o
i
a
lg
o
r
ith
m
o
n
p
o
p
u
latio
n
an
d
f
o
r
est d
ata
C
las
s
i
f
icatio
n
s
ar
e
o
n
e
o
f
t
h
e
m
aj
o
r
p
r
o
b
lem
s
th
at
m
a
n
y
r
e
s
ea
r
ch
er
s
f
ac
e
w
h
e
n
w
o
r
k
i
n
g
o
n
co
m
m
o
n
b
u
s
i
n
ess
p
r
o
b
lem
s
i
n
all
s
ec
to
r
s
.
I
n
th
i
s
ar
ticle,
w
e
w
i
ll
co
m
p
ar
e
th
r
ee
m
aj
o
r
tech
n
i
q
u
es
a
m
o
n
g
m
an
y
,
R
an
d
o
m
Fo
r
est,
SVM
an
d
KN
N.
-
R
an
d
o
m
Fo
r
est
is
a
co
llectio
n
o
f
d
ec
is
io
n
tr
ee
s
ap
p
lied
to
av
o
id
th
e
i
n
s
tab
ili
t
y
a
n
d
r
is
k
o
f
o
v
er
tr
ai
n
in
g
th
at
ca
n
o
cc
u
r
w
it
h
a
s
i
n
g
le
tr
ee
.
I
t
co
n
s
is
t
s
i
n
s
u
p
p
r
es
s
in
g
t
h
e
d
ec
i
s
io
n
n
o
d
es
w
it
h
o
u
t
r
ed
u
cin
g
th
e
o
v
er
all
p
r
ec
is
io
n
o
f
th
e
tr
ee
[
4
9
]
.
C
h
ar
ac
ter
ized
b
y
a
n
ad
j
u
s
t
m
e
n
t
o
f
o
n
l
y
t
w
o
p
ar
a
m
e
t
er
s
w
h
ic
h
ar
e
th
e
n
u
m
b
er
o
f
tr
ee
s
an
d
t
h
e
s
et
o
f
at
tr
ib
u
tes
to
b
e
c
h
o
s
en
d
u
r
in
g
t
h
e
co
n
s
tr
u
ctio
n
o
f
ea
ch
n
o
d
e,
w
h
ich
s
i
m
p
li
f
ie
s
th
e
g
e
n
er
atio
n
o
f
d
ec
is
io
n
f
o
r
ests
[
5
0
]
-
Su
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
s
ar
e
a
class
i
f
icatio
n
m
e
th
o
d
th
a
t
tr
an
s
f
o
r
m
s
a
lin
ea
r
p
r
o
b
le
m
in
to
a
h
i
g
h
er
d
i
m
en
s
io
n
al
s
p
ac
e
e
n
tit
y
.
T
h
e
y
m
an
a
g
e
n
o
n
-
li
n
ea
r
d
ec
is
io
n
li
m
it
s
a
n
d
t
h
e
ap
p
licatio
n
o
f
li
m
it
ca
s
es
allo
w
s
th
e
m
to
m
a
n
a
g
e
m
is
s
in
g
d
ata
[
5
1
]
.
Fo
r
a
b
in
ar
y
cla
s
s
i
f
icat
io
n
o
n
t
h
e
d
ata,
a
clas
s
i
f
icati
o
n
h
y
p
er
p
lan
e
i
s
u
s
ed
f
o
r
s
a
m
p
li
n
g
:
(
1
,
1
)
,
(
2
,
2
)
,
(
3
,
3
)
,
…
.
,
(
,
)
,
w
h
er
e
∈
,
∈
{
−
1
,
+
1
}
,
an
d
th
e
v
ec
to
r
1
is
th
e
v
ec
to
r
d
ir
ec
tl
y
cr
ea
ted
b
y
s
o
m
e
f
ea
t
u
r
es
o
f
t
h
e
s
a
m
p
le
[
3
5
]
.
T
h
e
k
ey
o
f
th
e
SVM
alg
o
r
ith
m
is
to
f
i
n
d
a
f
u
n
ctio
n
F.
So
th
at
X,
ap
ar
t f
r
o
m
t
h
e
s
a
m
p
le,
ca
n
o
b
t
ain
t
h
e
co
r
r
esp
o
n
d
in
g
Y
b
y
F
af
ter
t
h
e
tr
ain
in
g
o
f
th
e
s
a
m
p
l
e,
it is
t
h
en
p
o
s
s
ib
le
to
f
in
d
a
h
y
p
er
p
lan
e
i
n
d
icate
d
b
y
F
a
f
ter
th
e
tr
ai
n
i
n
g
;
it
c
an
d
iv
id
e
t
h
e
lear
n
i
n
g
s
a
m
p
les
in
to
p
o
s
iti
v
e
an
d
n
eg
at
i
v
e
ca
te
g
o
r
ies
a
n
d
t
h
en
s
ep
ar
ate
t
h
e
o
t
h
er
X
f
r
o
m
t
h
e
s
a
m
p
le.
I
f
th
e
d
ata
i
s
n
o
t
lin
ea
r
l
y
s
ep
ar
ab
le,
th
e
alg
o
r
it
h
m
ac
t
s
b
y
m
ap
p
i
n
g
th
e
d
ata
to
a
h
ig
h
er
-
d
i
m
e
n
s
io
n
al
f
ea
t
u
r
e
s
p
ac
e
ad
o
p
tin
g
a
n
o
n
-
lin
ea
r
k
er
n
el
f
u
n
ctio
n
Φ
(
X)
,
a
n
d
t
h
en
an
o
p
ti
m
ized
h
y
p
er
p
lan
e
is
p
r
o
d
u
ce
d
in
th
e
s
a
m
e
s
p
ac
e.
T
h
e
alg
o
r
it
h
m
ca
n
b
e
w
r
itte
n
as b
elo
w
.
,
,
1
2
+
+
∑
=
1
(
1
)
Su
b
j
ec
t to
(
∅
(
)
+
)
≥
1
−
,
≥
0
,
=
1
,
…
,
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.
10
,
No
.
5
,
Octo
b
e
r
2
0
2
0
:
5
5
0
7
-
5513
5510
I
n
th
is
al
g
o
r
ith
m
,
∅
(
)
+
=
0
d
ef
in
es
t
h
e
s
ep
ar
atin
g
h
y
p
er
la
n
e,
w
is
n
o
r
m
al
v
ec
to
r
o
f
h
y
p
er
p
lan
e,
b
is
o
f
f
s
et
o
f
h
y
p
er
p
la
n
e.
T
h
e
C
>
0
is
t
h
e
p
e
n
alt
y
p
ar
a
m
eter
o
f
t
h
e
er
r
o
r
ter
m
a
n
d
w
ar
e
t
h
e
w
e
ig
h
t
co
ef
fi
cie
n
ts
o
f
t
h
e
h
y
p
er
p
lan
e.
T
h
e
K
-
NN
alg
o
r
it
h
m
ca
n
b
e
u
s
ed
to
f
i
n
d
t
h
e
k
tr
ai
n
in
g
s
a
m
p
les
c
lo
s
est
to
th
e
tar
g
et
o
b
j
ec
t
b
ein
g
tau
g
h
t.
I
t
f
i
n
d
s
d
o
m
i
n
an
ce
f
r
o
m
t
h
e
k
lear
n
in
g
s
a
m
p
les;
t
h
en
as
s
i
g
n
t
h
ese
d
o
m
i
n
an
t
class
es
to
th
e
tar
g
e
t
o
b
j
ec
t,
w
h
er
e
k
is
th
e
n
u
m
b
er
o
f
tr
ain
i
n
g
s
a
m
p
les.
T
h
e
b
asic
ele
m
e
n
t
o
f
K
-
NN
is
th
at
all
s
a
m
p
les
h
av
e
th
e
s
a
m
e
p
r
o
p
er
ties
w
h
en
t
h
e
y
ar
e
class
if
ied
in
th
e
s
a
m
e
clas
s
in
f
u
n
c
tio
n
al
s
p
ac
e,
th
i
s
clas
s
co
m
p
r
is
in
g
th
e
k
clo
s
est
s
a
m
p
les
[
5
2
]
.
I
n
w
h
ic
h
X
u
b
elo
n
g
s
to
t
h
e
ca
te
g
o
r
y
o
f
(
1
)
.
T
h
e
m
ac
h
i
n
e
lear
n
in
g
m
eth
o
d
s
d
e
f
i
n
ed
ab
o
v
e
ar
e
ap
p
lied
ac
co
r
d
in
g
to
th
e
m
o
d
el
b
elo
w
i
n
F
ig
u
r
e
2
.
T
h
e
m
o
d
els
ar
e
d
ev
e
lo
p
ed
an
d
test
ed
b
y
u
s
in
g
Ma
th
w
o
r
k
s
T
o
o
ls
B
o
x
,
w
h
ic
h
i
s
an
en
v
ir
o
n
m
e
n
t
f
o
r
b
u
ild
in
g
an
d
ev
al
u
ati
n
g
m
ac
h
in
e
-
lear
n
i
n
g
al
g
o
r
ith
m
s
.
Fig
u
r
e
2
.
Sch
e
m
atic
i
ll
u
s
tr
atio
n
o
f
p
r
ed
ictiv
e
m
ac
h
i
n
e
l
ea
r
n
i
n
g
m
o
d
els
3.
RE
S
E
ARCH
M
E
T
H
O
D
T
h
e
FW
I
s
y
s
te
m
i
s
o
u
r
r
eso
u
r
ce
d
ata
o
f
th
e
ar
ea
s
b
u
r
n
ed
d
u
r
in
g
t
h
e
f
ir
e
s
b
et
w
ee
n
2
0
0
0
a
n
d
2
0
0
3
in
P
o
r
tu
g
al.
T
h
ey
co
n
tai
n
a
clea
r
d
escr
ip
tio
n
o
f
th
e
cli
m
atic
co
n
d
itio
n
s
.
T
h
ese
d
ata
ar
e
d
if
f
ic
u
lt
to
co
llect
f
r
o
m
lo
ca
l
s
en
s
o
r
s
av
ai
lab
le
in
P
o
r
tu
g
a
l
g
i
v
en
t
h
e
n
u
m
b
er
o
f
s
t
atio
n
s
.
T
h
e
y
also
c
o
n
ta
i
n
ad
d
itio
n
al
ti
m
e
v
al
u
es
s
u
c
h
as
d
a
y
s
,
m
o
n
t
h
s
,
a
n
d
co
o
r
d
in
ates
o
f
b
u
r
n
ed
ar
ea
s
.
T
h
e
ca
lcu
lated
v
al
u
es
o
f
th
e
in
d
ices
b
y
t
h
e
FW
I
s
y
s
te
m
ar
e
a
d
ir
ec
t
in
d
icato
r
o
f
th
e
i
n
ten
s
it
y
o
f
t
h
e
f
ir
e.
B
y
e
x
a
m
in
i
n
g
th
e
d
ata,
w
e
ca
n
s
a
y
t
h
at
w
h
e
n
th
e
w
i
n
d
b
lo
w
s
ar
o
u
n
d
1
5
k
m
/ h
o
u
r
,
th
e
r
is
k
o
f
f
ir
e
is
h
i
g
h
,
f
o
r
ex
a
m
p
le.
Ou
r
m
eth
o
d
i
s
m
ain
l
y
b
ased
o
n
d
iv
i
s
io
n
o
f
d
ata
i
n
to
s
e
v
er
al
eq
u
al
s
ize
clas
s
es.
E
ac
h
d
a
ta
ite
m
is
tr
ea
ted
s
ep
ar
atel
y
.
T
h
er
ef
o
r
e,
w
e
ca
n
u
s
e
th
e
n
ea
r
est
n
ei
g
h
b
o
r
m
e
th
o
d
o
r
th
e
a
v
er
ag
e
o
f
t
h
e
v
al
u
es
i
n
o
r
d
er
to
s
to
p
th
e
task
.
C
o
n
s
id
er
th
e
o
u
t
p
u
t
v
ar
iab
le
is
t
h
e
ar
ea
.
W
e
f
in
d
th
at
it
h
as
a
p
o
s
itiv
e
b
ias
.
T
h
e
m
aj
o
r
it
y
o
f
ar
ea
v
alu
e
s
is
n
u
l
l.
T
h
e
p
o
s
iti
v
e
t
ilt
illu
s
tr
ates
th
e
m
aj
o
r
ity
o
f
f
o
r
e
s
t
f
ir
es.
T
h
e
a
s
y
m
m
etr
ic
ch
ar
a
cter
s
y
s
te
m
i
s
a
ls
o
av
ailab
le
i
n
o
t
h
er
c
o
u
n
tr
ies
[
5
3
]
.
T
h
e
co
n
s
tr
ain
t
is
to
i
n
cr
ea
s
e
p
r
ec
is
io
n
a
n
d
d
ec
r
ea
s
e
a
s
y
m
m
etr
y
.
W
e
ad
d
a
class
co
lu
m
n
as
r
esp
o
n
s
e
v
a
r
iab
le,
w
h
ic
h
co
n
tai
n
s
t
w
o
v
a
lu
es
0
f
o
r
ar
ea
s
o
f
f
ir
e
less
th
an
5
0
h
a
an
d
1
f
o
r
ar
ea
s
g
r
ea
ter
th
a
n
5
0
h
a.
I
n
o
r
d
er
to
f
in
d
th
e
m
ea
n
i
n
g
f
u
l
at
tr
ib
u
te,
th
e
co
r
r
elatio
n
m
atr
ix
i
s
u
s
ed
.
W
e
n
o
te
th
at
th
e
attr
ib
u
te
s
DC
a
n
d
ar
ea
h
av
e
a
m
o
r
e
p
o
s
itiv
e
co
r
r
elatio
n
w
it
h
t
h
e
r
esp
o
n
s
e
v
ar
iab
le
an
d
L
e
R
H
h
a
s
a
m
o
r
e
n
eg
at
iv
e
co
r
r
elatio
n
w
it
h
t
h
e
o
u
tp
u
t
v
ar
iab
le.
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
I
n
th
i
s
s
tep
,
w
e
m
u
s
t
ch
o
o
s
e
th
e
b
est
p
r
ed
ictiv
e
m
o
d
el
to
u
s
e.
T
h
e
b
asic
co
m
p
ar
is
o
n
p
a
r
a
m
eter
is
ac
cu
r
ac
y
.
T
h
e
r
esu
lt
s
o
f
t
h
e
d
if
f
er
en
t
m
o
d
el
s
as f
o
llo
w
s
:
I
n
o
r
d
er
to
b
etter
s
it
u
ate
t
h
e
p
r
ed
ictiv
e
m
ac
h
i
n
e
lear
n
i
n
g
m
o
d
el
s
,
w
e
s
tar
t
b
y
t
h
e
co
n
f
u
s
io
n
m
atr
i
x
w
h
ic
h
h
elp
u
s
ca
lc
u
late
t
h
e
ac
cu
r
ac
y
o
f
th
e
m
o
d
el.
T
h
e
f
o
r
m
u
la
to
ca
lcu
late
is
g
iv
e
n
b
el
o
w
.
=
(
T
r
u
e
p
o
s
itiv
e
+T
r
u
e
Neg
ativ
e)
/
(
T
r
u
e
p
o
s
itiv
e
+T
r
u
e
Neg
ativ
e
+
Fals
e
p
o
s
itiv
e
+
Fals
e
Ne
g
ativ
e)
.
C
o
n
f
u
s
io
n
Ma
tr
i
x
is
a
tab
le
s
h
o
w
s
ac
tu
al
v
s
p
r
ed
icted
v
alu
e
s
.
I
t
is
o
n
e
o
f
t
h
e
e
asies
t
w
a
y
s
to
f
i
n
d
ac
cu
r
ac
y
an
d
also
it
h
elp
s
to
av
o
id
o
v
er
f
itti
n
g
.
T
h
e
F
ig
u
r
e
3
p
r
esen
ts
th
e
co
n
f
u
s
io
n
m
a
tr
ix
v
al
u
es
f
o
r
ea
ch
M
L
m
o
d
el
.
R
F
m
o
d
el
p
r
o
d
u
ce
s
1
0
0
%
o
f
p
o
s
itiv
e
p
r
ed
ictiv
e
v
alu
e
w
h
er
e
th
e
r
ate
o
f
b
o
th
s
m
all
(
C
l
ass
=0
)
an
d
lar
g
er
(
C
lass
=1
)
f
ir
e
p
r
ed
ictio
n
is
1
0
0
%
w
h
ile
t
h
e
f
al
s
e
d
is
co
v
er
y
r
ate
–
er
r
o
r
ty
p
e
–
is
0
%.
Fo
r
SVM
an
d
KNN,
t
h
e
r
ate
o
f
er
r
o
r
t
h
at
t
h
e
y
p
r
o
d
u
ce
r
esp
ec
ti
v
el
y
3
5
%,
4
5
% f
o
r
t
h
e
s
m
all
f
ir
e
an
d
2
9
%
,
4
5
% f
o
r
t
h
e
lar
g
e
f
ir
e.
I
n
co
n
s
eq
u
e
n
ce
,
th
e
p
er
f
o
r
m
an
ce
cla
s
s
i
f
i
ca
tio
n
r
ate
o
f
th
e
t
w
o
m
o
d
els
SVM
a
n
d
KNN
d
ec
r
ea
s
e.
T
h
e
p
r
e
d
ictio
n
ac
cu
r
ac
y
o
f
r
a
n
d
o
m
f
o
r
est is
i
n
ter
esti
n
g
.
Hen
ce
,
it r
ed
u
ce
s
th
e
n
o
i
s
e
in
t
h
e
d
ata
s
et.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
C
o
mp
a
r
a
tive
s
tu
d
y
o
n
ma
c
h
in
e
lea
r
n
in
g
a
lg
o
r
ith
ms fo
r
ea
r
ly
fir
e
fo
r
es
t d
etec
tio
n
.
..
(
Zo
u
iten
Mo
h
a
mme
d
)
5511
(
a)
(
b
)
(
c)
Fig
u
r
e
3
.
C
o
n
f
u
s
io
n
m
atr
ix
f
o
r
R
F,
SVM
an
d
KNN
,
(
a)
c
o
n
f
u
s
io
n
m
atr
i
x
f
o
r
r
an
d
o
m
f
o
r
est
,
(
b
)
c
o
n
f
u
s
io
n
m
atr
i
x
f
o
r
SVM
m
eth
o
d
,
(
c)
c
o
n
f
u
s
io
n
m
atr
i
x
f
o
r
KNN
m
et
h
o
d
O
v
er
all
ac
c
u
r
ac
y
o
f
R
F
i
s
1
0
0
%.
T
h
is
s
h
o
w
s
th
at
R
F
h
as
t
h
e
b
est
p
r
ed
ictio
n
r
es
u
lts
co
m
p
ar
in
g
w
it
h
SVM
an
d
KNN
w
h
ic
h
th
e
y
r
esp
ec
tiv
el
y
g
et
6
7
.
7
%
an
d
5
4
.
9
%
(
u
ci)
.
T
h
e
r
ec
eiv
er
o
p
e
r
atin
g
c
u
r
v
e
(
R
O
C
cu
r
v
e)
w
i
ll
s
u
m
m
ar
ize
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
m
o
d
el
b
y
ass
e
s
s
i
n
g
t
h
e
tr
ad
e
-
o
f
f
b
et
wee
n
s
en
s
iti
v
it
y
a
n
d
s
p
ec
if
icit
y
.
W
e
m
u
s
t
al
w
a
y
s
t
h
in
k
o
f
p
>
0
.
5
w
h
e
n
w
e
d
r
aw
t
h
e
R
OC
,
b
ec
au
s
e
w
e
ar
e
co
n
ce
r
n
ed
ab
o
u
t
th
e
s
u
cc
e
s
s
r
ate.
A
r
ea
u
n
d
er
t
h
e
c
u
r
v
e
(
AUC)
o
r
co
n
co
r
d
an
ce
i
n
d
ex
,
i
s
a
m
e
tr
ic
o
f
t
h
e
e
x
ce
ll
en
t
p
er
f
o
r
m
an
ce
o
f
th
e
R
O
C
c
u
r
v
e.
T
h
e
ac
cu
r
ac
y
r
esu
lt
s
o
f
t
h
e
th
r
ee
m
eth
o
d
s
ar
e
r
ep
r
esen
ted
as f
o
llo
w
s
:
I
n
th
i
s
ar
ticle,
w
e
h
av
e
s
i
m
u
la
ted
th
r
ee
m
ac
h
in
e
lear
n
in
g
al
g
o
r
ith
m
s
u
s
in
g
d
ata
f
r
o
m
t
h
e
M
o
n
tesi
n
h
o
P
ar
k
in
P
o
r
tu
g
al.
A
cc
o
r
d
in
g
t
o
T
ab
le
1
,
th
e
class
if
icatio
n
p
er
f
o
r
m
a
n
ce
o
f
t
h
e
th
r
ee
m
et
h
o
d
s
is
as
f
o
llo
w
s
:
th
e
r
an
d
o
m
f
o
r
est
is
at
1
0
0
%
w
h
ile
th
e
SVM
p
r
o
v
i
d
es
7
4
%
an
d
K
-
NN
o
f
f
er
s
5
8
%
w
h
ich
r
ep
r
esen
ts
a
li
m
it
i
n
ter
m
s
o
f
cla
s
s
i
f
icat
io
n
.
T
ab
le
1
.
A
cc
u
r
ac
y
f
o
r
R
F,
SV
M
an
d
KNN
m
et
h
o
d
M
e
t
h
o
d
RF
S
V
M
K
N
N
A
c
c
u
r
a
c
y
1
0
.
7
4
0
.
5
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.
10
,
No
.
5
,
Octo
b
e
r
2
0
2
0
:
5
5
0
7
-
5513
5512
5.
CO
NCLU
SI
O
N
T
h
e
f
ield
o
f
d
ata
s
cie
n
ce
is
b
o
o
m
i
n
g
.
T
h
is
p
u
s
h
es
r
esear
ch
er
s
to
d
ev
elo
p
in
cr
ea
s
in
g
l
y
co
m
p
le
x
p
r
o
b
lem
-
s
o
lv
in
g
m
et
h
o
d
s
.
Ou
r
ap
p
r
o
ac
h
b
ased
m
ai
n
l
y
o
n
ex
tr
ac
ti
n
g
d
ata
f
r
o
m
ex
i
s
ti
n
g
d
atab
ases
u
s
ed
th
r
ee
d
if
f
er
e
n
t
m
ac
h
i
n
e
lear
n
in
g
al
g
o
r
ith
m
s
.
B
et
w
ee
n
th
e
K
-
N
N
S
u
p
p
o
r
t
Vec
to
r
Ma
ch
in
e
a
n
d
R
F
alg
o
r
it
h
m
w
e
h
av
e
s
h
o
w
n
t
h
at
t
h
e
K
-
N
N
h
a
s
th
e
b
est
ac
cu
r
ac
y
.
T
h
is
al
g
o
r
ith
m
h
as
a
s
et
o
f
d
ata
d
etec
tio
n
an
d
r
ec
o
g
n
itio
n
ass
ets
w
h
ic
h
m
a
k
es
s
p
atial
m
a
n
ip
u
lat
io
n
m
u
c
h
ea
s
ier
f
o
r
th
e
d
etec
tio
n
o
f
at
-
r
is
k
o
r
b
u
r
n
t
a
r
ea
s
.
T
h
e
ch
o
ice
o
f
th
e
alg
o
r
it
h
m
o
f
t
h
e
h
i
g
h
es
t
p
r
ec
is
io
n
is
j
u
s
ti
f
ied
b
y
t
h
e
s
i
m
u
latio
n
o
f
th
e
d
i
f
f
er
en
t
al
g
o
r
ith
m
s
a
n
d
th
e
co
m
p
ar
is
o
n
o
f
th
e
e
x
p
er
i
m
en
tal
r
es
u
lt
s
o
b
tain
ed
.
I
n
ad
d
itio
n
,
th
e
s
y
s
te
m
,
u
s
i
n
g
d
ata
co
llectin
g
s
e
n
s
o
r
s
,
g
en
er
ate
s
lar
g
e
d
ata
w
h
ic
h
is
an
al
y
ze
d
b
y
v
ar
io
u
s
m
ac
h
i
n
e
l
ea
r
n
in
g
m
et
h
o
d
s
ci
ted
an
d
co
m
p
ar
ed
in
th
is
s
t
u
d
y
to
p
r
ed
ict
w
i
th
h
ig
h
ac
c
u
r
ac
y
th
e
a
m
o
u
n
t
o
f
la
n
d
b
u
r
n
ed
i
n
a
f
o
r
est.
In
th
e
n
o
r
th
ea
s
t
r
eg
io
n
o
f
P
o
r
tu
g
al.
T
h
e
u
s
e
o
f
tec
h
n
o
lo
g
y
is
t
h
e
n
t
h
e
s
tr
e
n
g
t
h
o
f
th
is
p
r
ed
icti
o
n
s
y
s
te
m
.
Geo
g
r
ap
h
ic
in
f
o
r
m
atio
n
s
y
s
te
m
s
a
n
d
m
ac
h
in
e
lear
n
i
n
g
ca
n
h
elp
d
ec
is
io
n
m
a
k
er
s
to
m
in
i
m
ize
th
e
n
atu
r
al
an
d
h
u
m
an
d
a
m
a
g
e
ca
u
s
ed
b
y
f
o
r
est
f
ir
es.
T
h
e
u
s
e
o
f
t
h
ese
m
et
h
o
d
s
is
i
n
cr
ea
s
i
n
g
l
y
o
p
ti
m
izi
n
g
t
h
e
tr
e
at
m
e
n
t
o
f
t
h
is
p
h
en
o
m
e
n
o
n
a
n
d
th
o
s
e
o
f
its
k
i
n
d
.
T
h
e
p
lay
er
s
i
n
th
e
s
ec
to
r
s
in
q
u
esti
o
n
ar
e
th
e
n
in
v
ited
to
j
o
in
h
an
d
s
i
n
f
ig
h
ti
n
g
ag
ai
n
s
t la
te
in
ter
v
e
n
tio
n
s
.
RE
F
E
R
E
NC
E
S
[1
]
P
.
M
a
teu
s
a
n
d
P
.
M
.
F
e
rn
a
n
d
e
s
,
“
F
o
re
st
F
ires
i
n
P
o
rt
u
g
a
l:
Dy
n
a
m
ics
,
Ca
u
se
s
a
n
d
P
o
li
c
ies
,
”
in
F
.
Re
b
o
re
d
o
(e
d
s),
“
F
o
re
st Co
n
tex
t
a
n
d
P
o
li
c
ies
i
n
P
o
rtu
g
a
l
,
”
W
o
rld
Fo
re
sts,
S
p
rin
g
e
r
,
v
o
l.
1
9
,
p
p
.
9
7
-
1
1
5
,
2
0
1
4
.
[2
]
M
.
L
.
S
h
a
h
re
z
a
,
e
t
a
l.
,
“
A
n
o
m
a
l
y
d
e
tec
ti
o
n
u
sin
g
a
se
lf
-
o
rg
a
n
izin
g
m
a
p
a
n
d
p
a
rti
c
le
sw
a
r
m
o
p
ti
m
iz
a
ti
o
n
,
”
S
c
ien
ti
a
Ira
n
ica
,
v
o
l.
18
,
n
o
.
6
,
p
p
.
1
4
6
0
-
1
4
6
8
,
2
0
1
1
.
[3
]
B.
C.
A
rru
e
,
e
t
a
l.
,
“
A
n
in
telli
g
e
n
t
sy
ste
m
f
o
r
f
a
lse
a
lar
m
r
e
d
u
c
ti
o
n
i
n
i
n
f
ra
re
d
f
o
re
st
-
f
ire
d
e
tec
ti
o
n
,
”
IE
EE
In
tell
ig
e
n
t
S
y
st
e
ms
a
n
d
t
h
e
ir A
p
p
l
ica
ti
o
n
s
,
v
o
l.
15
,
n
o
.
3
,
p
p
.
64
-
73
,
2
0
0
0
.
[4
]
J
.
P
i
ñ
o
l,
e
t
a
l.
,
“
Cli
m
a
te
wa
r
m
i
n
g
,
w
il
d
f
ire
h
a
z
a
rd
,
a
n
d
w
il
d
f
ire
o
c
c
u
rre
n
c
e
in
c
o
a
sta
l
e
a
ste
rn
S
p
a
in
,
”
Cli
m
a
ti
c
Ch
a
n
g
e
,
v
o
l
.
38
,
n
o
.
3
,
p
p
.
3
4
5
-
3
5
7
,
1
9
9
8
.
[5
]
Q.
Ng
u
y
e
n
a
n
d
G
.
Ch
a
k
ra
b
o
rty
,
“
P
re
d
icti
n
g
f
o
re
st
f
ire
o
c
c
u
rre
n
c
e
a
n
d
in
c
r
e
m
e
n
tal
f
ire
r
a
te
u
sin
g
S
A
S
®
9
.
4
a
n
d
S
A
S
®
E
n
terp
rise
M
i
n
e
r
TM
1
4
.
1
,
”
Pro
c
e
e
d
in
g
s o
f
S
A
S
Glo
b
al
Fo
r
u
m 2
0
1
6
,
p
p
.
1
-
12
,
2
0
1
6
.
[6
]
S
.
W
.
Ta
y
lo
r
a
n
d
M
.
E.
A
le
x
a
n
d
e
r,
“
S
c
ien
c
e
,
tec
h
n
o
lo
g
y
,
a
n
d
h
u
m
a
n
fa
c
to
rs
in
f
ire
d
a
n
g
e
r
ra
ti
n
g
:
T
h
e
Ca
n
a
d
ian
e
x
p
e
rien
c
e
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
W
il
d
la
n
d
Fi
re
,
v
o
l.
15
,
n
o
.
1
,
p
p
.
1
2
1
-
135
,
2
0
0
6
.
[7
]
D.
X
.
V
ieg
a
s,
e
t
a
l.
,
“
Co
m
p
a
ra
ti
v
e
stu
d
y
o
f
v
a
rio
u
s
m
e
th
o
d
s
o
f
f
ire
d
a
n
g
e
r
e
v
a
lu
a
ti
o
n
in
s
o
u
th
e
rn
Eu
r
o
p
e
,
”
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
W
il
d
la
n
d
Fi
re
,
v
o
l.
9
,
n
o
.
4
,
p
p
.
2
3
5
-
2
4
6
,
1
9
9
9
.
[8
]
U.
M
.
F
e
y
y
a
d
,
“
Da
ta
m
in
in
g
a
n
d
k
n
o
w
led
g
e
d
isc
o
v
e
r
y
:
m
a
k
in
g
se
n
se
o
u
t
o
f
d
a
ta,
”
in
IEE
E
Exp
e
rt
,
v
o
l.
11
,
n
o
.
5
,
p
p
.
20
-
2
5
,
1
9
9
6
.
[9
]
D.
J.
Ha
n
d
,
e
t
a
l.
,
“
Da
ta m
in
in
g
f
o
r
f
u
n
a
n
d
p
ro
f
it
,”
S
t
a
t
isti
c
a
l
S
c
i
e
n
c
e
,
v
o
l.
15
,
n
o
.
2
,
p
p
.
1
1
1
-
1
2
6
,
2
0
0
0
.
[1
0
]
Y.
Jia
,
e
t
a
l.
,
“
L
ig
h
t
Co
n
d
it
i
o
n
Es
ti
m
a
ti
o
n
Ba
se
d
o
n
Vid
e
o
F
ire
De
tec
ti
o
n
i
n
S
p
a
c
io
u
s
Bu
il
d
i
n
g
s,”
Ara
b
ia
n
J
o
u
rn
a
l
fo
r
S
c
i
e
n
c
e
a
n
d
En
g
i
n
e
e
rin
g
,
v
o
l.
41
,
p
p
.
1
0
3
1
-
1
0
4
1
,
2
0
1
6
.
[1
1
]
C.
E
.
P
re
m
a
,
e
t
a
l.
,
“
M
u
lt
i
F
e
a
t
u
r
e
A
n
a
l
y
sis
o
f
S
m
o
k
e
in
YU
V
Co
lo
r
S
p
a
c
e
f
o
r
Earl
y
F
o
re
st
F
ire
De
tec
ti
o
n
,
”
Fi
re
T
e
c
h
n
o
l
ogy
,
v
o
l.
52
,
n
o
.
5
,
p
p
.
1
3
1
9
-
1
3
4
2
,
2
0
1
6
.
[1
2
]
D.
P
.
Ku
m
a
r,
e
t
a
l.
,
“
M
a
c
h
in
e
le
a
rn
in
g
a
lg
o
rit
h
m
s
f
o
r
w
irele
ss
se
n
so
r
n
e
tw
o
rk
s:
A
su
rv
e
y
,
”
In
f
o
rm
a
ti
o
n
Fu
si
o
n
,
v
o
l.
49
,
p
p
.
1
-
2
5
,
2
0
1
9
.
[1
3
]
G
.
Ch
e
n
g
a
n
d
J.
Ha
n
,
“
A
su
rv
e
y
o
n
o
b
jec
t
d
e
tec
ti
o
n
in
o
p
ti
c
a
l
re
m
o
te
s
e
n
sin
g
i
m
a
g
e
s,”
IS
P
RS
J
o
u
rn
a
l
o
f
Ph
o
t
o
g
r
a
mm
e
try
a
n
d
Rem
o
te
S
e
n
sin
g
,
v
o
l.
1
1
7
,
p
p
.
11
-
2
8
,
2
0
1
6
.
[1
4
]
M
.
O.
Ola
d
ime
ji
,
e
t
a
l.
,
“
A
n
e
w
a
p
p
r
o
a
c
h
f
o
r
e
v
e
n
t
d
e
tec
ti
o
n
u
sin
g
k
-
m
e
a
n
s
c
lu
ste
rin
g
a
n
d
n
e
u
ra
l
n
e
tw
o
rk
s,
”
2
0
1
5
In
ter
n
a
t
io
n
a
l
J
o
in
t
Co
n
fer
e
n
c
e
o
n
Ne
u
ra
l
Ne
two
rk
s (
IJ
CNN)
,
Kill
a
rn
e
y
,
p
p
.
1
-
5
,
2
0
1
5
.
[1
5
]
K.
T
riv
e
d
i
a
n
d
A
.
K.
S
riv
a
sta
v
a
,
“
A
n
e
n
e
r
g
y
e
ff
icie
n
t
f
r
a
m
e
w
o
rk
f
o
r
d
e
tec
ti
o
n
a
n
d
m
o
n
it
o
r
in
g
o
f
fo
re
st
f
ire
u
sin
g
m
o
b
il
e
a
g
e
n
t
in
w
irele
ss
se
n
so
r
n
e
t
w
o
rk
s,
”
2
0
1
4
IEE
E
I
n
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
C
o
mp
u
ta
t
io
n
a
l
In
telli
g
e
n
c
e
a
n
d
Co
mp
u
t
in
g
Res
e
a
rc
h
,
C
o
im
b
a
to
re
,
p
p
.
1
-
4
,
2
0
1
4
.
[1
6
]
Y.
S
in
g
h
,
e
t
a
l.
,
“
Distrib
u
ted
e
v
e
n
t
d
e
tec
ti
o
n
in
w
irele
ss
se
n
so
r
n
e
tw
o
rk
s
f
o
r
f
o
re
st
f
ires
,
”
2
0
1
3
UKS
im
1
5
th
In
ter
n
a
t
io
n
a
l
C
o
n
fer
e
n
c
e
o
n
C
o
mp
u
ter
M
o
d
e
ll
in
g
a
n
d
S
im
u
la
t
io
n
,
Ca
m
b
rid
g
e
,
p
p
.
6
3
4
-
6
3
9
,
2
0
1
3
.
[1
7
]
M
.
A
.
Cri
m
m
in
s,
“
S
y
n
o
p
ti
c
c
li
m
a
to
lo
g
y
o
f
e
x
tre
m
e
f
ire
-
we
a
th
e
r
c
o
n
d
i
ti
o
n
s
a
c
r
o
ss
th
e
so
u
t
h
w
e
st
Un
it
e
d
S
tate
s
,”
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
Cl
ima
t
o
l
o
g
y
,
v
o
l
.
26
,
n
o
.
8
,
p
p
.
1
0
0
1
-
1
0
1
6
,
2
0
0
6
.
[1
8
]
S.
Ha
n
tso
n
,
e
t
a
l.
,
“
G
lo
b
a
l
f
ire
siz
e
d
istri
b
u
ti
o
n
is
d
ri
v
e
n
b
y
h
u
m
a
n
i
m
p
a
c
t
a
n
d
c
li
m
a
te
,”
Glo
b
a
l
Eco
lo
g
y
a
n
d
Bi
o
g
e
o
g
ra
p
h
y
,
v
o
l.
24
,
n
o
.
1
,
p
p
.
77
-
86
,
2
015
.
[1
9
]
S
.
W
.
Ru
n
n
i
n
g
,
“
Is
g
lo
b
a
l
w
a
r
m
in
g
c
a
u
sin
g
m
o
re
,
larg
e
r
w
il
d
f
ir
e
s?
,
”
S
c
ien
c
e
,
v
o
l.
3
1
3
,
n
o
.
5
7
8
9
,
p
p
.
9
2
7
-
9
2
8
,
2
0
0
6
.
[2
0
]
N.
N
.
T
h
a
c
h
,
e
t
a
l.
,
“
S
p
a
ti
a
l
p
a
tt
e
rn
a
ss
e
ss
m
e
n
t
o
f
tro
p
ica
l
f
o
re
st
f
ire
d
a
n
g
e
r
a
t
T
h
u
a
n
Ch
a
u
a
re
a
(V
ietn
a
m
)
u
sin
g
G
IS
-
b
a
se
d
a
d
v
a
n
c
e
d
m
a
c
h
in
e
lea
rn
in
g
a
lg
o
rit
h
m
s:
A
c
o
m
p
a
ra
ti
v
e
stu
d
y
,”
Eco
lo
g
ica
l
I
n
fo
rm
a
ti
c
s
,
v
o
l.
46
,
pp.
74
-
85
,
2
0
1
8
.
[2
1
]
L
.
A
.
Di
m
u
c
c
io
,
e
t
a
l.
,
“
Re
g
io
n
a
l
f
o
re
st
-
f
ire
su
sc
e
p
ti
b
il
it
y
a
n
a
l
y
sis
in
c
e
n
tral
P
o
rtu
g
a
l
u
sin
g
a
p
ro
b
a
b
il
isti
c
ra
ti
n
g
s
p
ro
c
e
d
u
re
a
n
d
a
rti
f
icia
l
n
e
u
ra
l
n
e
tw
o
rk
w
e
i
g
h
ts
a
ss
ig
n
m
e
n
t
,”
In
ter
n
a
ti
o
n
a
l
J
o
u
r
n
a
l
o
f
W
il
d
la
n
d
Fi
re
,
v
o
l.
20
,
n
o
.
6
,
pp.
7
7
6
-
791
,
2
0
1
1
.
[2
2
]
H.
Ho
n
g
,
e
t
a
l.
,
“
P
re
d
icti
n
g
sp
a
ti
a
l
p
a
tt
e
rn
s
o
f
w
il
d
f
ire
su
sc
e
p
ti
b
il
i
ty
in
th
e
Hu
ich
a
n
g
Co
u
n
ty
,
Ch
i
n
a
:
A
n
in
teg
ra
ted
m
o
d
e
l
to
a
n
a
ly
sis o
f
lan
d
sc
a
p
e
in
d
ica
to
rs
,”
Eco
l
o
g
ic
a
l
I
n
d
ica
to
rs
,
v
o
l.
1
0
1
,
p
p
.
8
7
8
-
8
9
1
,
2
0
1
9
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
C
o
mp
a
r
a
tive
s
tu
d
y
o
n
ma
c
h
in
e
lea
r
n
in
g
a
lg
o
r
ith
ms fo
r
ea
r
ly
fir
e
fo
r
es
t d
etec
tio
n
.
..
(
Zo
u
iten
Mo
h
a
mme
d
)
5513
[2
3
]
A
.
J
a
a
f
a
ri,
e
t
a
l.
,
“
H
y
b
rid
a
rti
f
icia
l
in
t
e
ll
ig
e
n
c
e
m
o
d
e
ls
b
a
se
d
o
n
a
n
e
u
r
o
-
f
u
z
z
y
s
y
ste
m
a
n
d
m
e
tah
e
u
risti
c
o
p
ti
m
iza
ti
o
n
a
lg
o
ri
th
m
s
f
o
r
sp
a
ti
a
l
p
re
d
ictio
n
o
f
w
il
d
f
ire
p
ro
b
a
b
il
it
y
,”
Ag
ric
u
lt
u
r
a
l
a
n
d
Fo
re
st
M
e
teo
ro
lo
g
y
,
v
o
l.
2
6
6
-
2
6
7
,
p
p
.
1
9
8
-
2
0
7
,
2
0
1
9
.
[2
4
]
M.
L
e
u
e
n
b
e
rg
e
r,
e
t
a
l.
,
“
W
il
d
f
ire
su
sc
e
p
ti
b
i
li
t
y
m
a
p
p
in
g
:
De
term
i
n
isti
c
v
s.
sto
c
h
a
stic
a
p
p
r
o
a
c
h
e
s
,”
En
v
iro
n
me
n
ta
l
M
o
d
e
ll
in
g
a
n
d
S
o
ft
wa
re
,
v
o
l.
1
0
1
,
p
p
.
1
9
4
-
2
0
3
,
2
0
1
8
.
[2
5
]
D.
T
.
Bu
i,
e
t
a
l.
,
“
A
h
y
b
rid
a
rti
f
icia
l
in
telli
g
e
n
c
e
a
p
p
ro
a
c
h
u
sin
g
G
IS
-
b
a
se
d
n
e
u
ra
l
-
f
u
z
z
y
in
f
e
r
e
n
c
e
sy
ste
m
a
n
d
p
a
rti
c
le
sw
a
r
m
o
p
ti
m
iza
ti
o
n
f
o
r
f
o
re
st
f
ire
su
sc
e
p
ti
b
il
it
y
m
o
d
e
li
n
g
a
t
a
tro
p
ica
l
a
re
a
,”
Ag
ric
u
lt
u
r
a
l
a
n
d
F
o
re
st
M
e
teo
ro
lo
g
y
,
v
o
l
.
2
3
3
,
p
p
.
32
-
44
,
2
0
1
7
.
[2
6
]
A
.
B
.
M
a
ss
a
d
a
,
e
t
a
l.
,
“
W
il
d
f
ir
e
ig
n
it
io
n
-
d
istri
b
u
ti
o
n
m
o
d
e
ll
in
g
:
a
c
o
m
p
a
ra
ti
v
e
stu
d
y
in
th
e
H
u
ro
n
–
M
a
n
istee
Na
ti
o
n
a
l
F
o
re
st,
M
ic
h
ig
a
n
,
USA
,”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
W
il
d
la
n
d
F
ire
,
v
o
l.
22
,
n
o
.
2
,
p
p
.
1
7
4
-
1
8
3
,
2
0
1
2
.
[2
7
]
M.
Bisq
u
e
rt,
e
t
a
l.
,
“
A
p
p
li
c
a
ti
o
n
o
f
a
rti
f
icia
l
n
e
u
ra
l
n
e
tw
o
rk
s
a
n
d
l
o
g
isti
c
re
g
re
ss
io
n
to
th
e
p
re
d
icti
o
n
o
f
f
o
re
st
f
ire
d
a
n
g
e
r
in
G
a
li
c
ia u
sin
g
M
OD
IS
d
a
ta
,”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
W
il
d
la
n
d
Fi
re
,
v
o
l.
21
,
n
o
.
8
,
p
p
.
1
0
2
5
-
1
0
2
9
,
2
0
1
2
.
[2
8
]
O.
S
a
ti
r,
e
t
a
l.
,
“
M
a
p
p
i
n
g
re
g
io
n
a
l
f
o
re
st
f
ire
p
ro
b
a
b
il
it
y
u
sin
g
a
rti
f
icia
l
n
e
u
ra
l
n
e
tw
o
rk
m
o
d
e
l
in
a
M
e
d
it
e
rra
n
e
a
n
f
o
re
st ec
o
s
y
st
e
m
,”
Ge
o
ma
ti
c
s,
Na
tu
ra
l
H
a
za
rd
s
a
n
d
Ri
sk
,
v
o
l.
7
,
n
o
.
5
,
p
p
.
1
6
4
5
-
1
6
5
8
,
2
0
1
6
.
[2
9
]
A
.
A
rp
a
c
i,
e
t
a
l.
,
“
Us
in
g
m
u
lt
i
v
a
riate
d
a
ta
m
in
in
g
tec
h
n
iq
u
e
s
f
o
r
e
stim
a
ti
n
g
f
ire
su
s
c
e
p
ti
b
il
it
y
o
f
Ty
ro
lea
n
f
o
re
sts
,”
Ap
p
li
e
d
Ge
o
g
r
a
p
h
y
,
v
o
l.
53
,
p
p
.
2
5
8
-
2
7
0
,
2
0
1
4
.
[3
0
]
S.
Oliv
e
ira,
e
t
a
l.
,
“
M
o
d
e
li
n
g
sp
a
t
ial
p
a
tt
e
rn
s o
f
f
ire
o
c
c
u
rre
n
c
e
in
M
e
d
it
e
rra
n
e
a
n
E
u
ro
p
e
u
si
n
g
M
u
lt
ip
le
Re
g
re
ss
io
n
a
n
d
Ra
n
d
o
m
F
o
re
st
,”
Fo
re
st E
c
o
l
o
g
y
a
n
d
M
a
n
a
g
e
me
n
t
,
v
o
l.
2
7
5
,
p
p
.
1
1
7
-
1
2
9
,
2
0
1
2
.
[3
1
]
Z.
S
.
P
o
u
rtag
h
i,
e
t
a
l.
,
“
In
v
e
stig
a
ti
o
n
o
f
g
e
n
e
ra
l
in
d
ica
to
rs
i
n
f
lu
e
n
c
in
g
o
n
f
o
re
st
f
ire
a
n
d
it
s
su
sc
e
p
ti
b
il
it
y
m
o
d
e
li
n
g
u
sin
g
d
if
f
e
re
n
t
d
a
ta m
in
in
g
tec
h
n
i
q
u
e
s
,”
Eco
l
o
g
ic
a
l
I
n
d
ica
to
rs
,
v
o
l
.
64
,
p
p
.
72
-
84
,
2
0
1
6
.
[3
2
]
H.
Ho
n
g
,
e
t
a
l.
,
“
A
p
p
ly
in
g
g
e
n
e
t
ic
a
lg
o
rit
h
m
s
to
se
t
t
h
e
o
p
ti
m
a
l
c
o
m
b
in
a
ti
o
n
o
f
f
o
re
st
f
ire
re
late
d
v
a
riab
les
a
n
d
m
o
d
e
l
f
o
re
st
f
ire
su
sc
e
p
ti
b
il
it
y
b
a
se
d
o
n
d
a
ta
m
in
in
g
m
o
d
e
ls
,”
S
c
ien
c
e
o
f
th
e
T
o
t
a
l
E
n
v
iro
n
me
n
t
,
v
o
l.
6
3
0
,
pp.
1
0
4
4
-
1
0
5
6
,
2
0
1
8
.
[3
3
]
M
.
J.
P
.
V
a
sc
o
n
c
e
lo
s,
e
t
a
l.
,
“
S
p
a
ti
a
l
P
re
d
icti
o
n
o
f
F
ire
Ig
n
it
io
n
P
ro
b
a
b
i
li
ti
e
s:
Co
m
p
a
rin
g
L
o
g
isti
c
Re
g
re
ss
io
n
a
n
d
Ne
u
ra
l
Ne
tw
o
rk
s
,”
Ph
o
to
g
ra
mm
e
tric E
n
g
i
n
e
e
rin
g
&
Rem
o
te
S
e
n
sin
g
,
v
o
l.
67
,
n
o
.
1
,
p
p
.
73
-
81
,
2
0
0
1
.
[3
4
]
D.
T
.
Bu
i,
e
t
a
l.
,
“
T
ro
p
ica
l
f
o
re
s
t
f
ire
su
sc
e
p
ti
b
il
it
y
m
a
p
p
in
g
a
t
th
e
Ca
t
Ba
Na
ti
o
n
a
l
P
a
rk
a
re
a
,
Ha
i
P
h
o
n
g
Cit
y
,
V
ietn
a
m
,
u
sin
g
G
IS
-
b
a
se
d
Ke
rn
e
l
lo
g
isti
c
re
g
re
ss
io
n
,”
Rem
o
te S
e
n
sin
g
,
v
o
l.
8
,
n
o
.
4
,
p
p
.
3
4
7
-
3
6
1
,
2
0
1
6
.
[3
5
]
H.
Ho
n
g
,
e
t
a
l.
,
“
S
p
a
ti
a
l
p
re
d
icti
o
n
o
f
lan
d
slid
e
h
a
z
a
rd
a
t
t
h
e
Yih
u
a
n
g
a
re
a
(Ch
in
a
)
u
sin
g
tw
o
-
c
las
s
k
e
rn
e
l
lo
g
isti
c
re
g
re
ss
io
n
,
a
lt
e
rn
a
ti
n
g
d
e
c
isio
n
tr
e
e
a
n
d
su
p
p
o
r
t
v
e
c
to
r
m
a
c
h
in
e
s
,”
Ca
ten
a
,
v
o
l.
1
3
3
,
p
p
.
2
6
6
-
2
8
1
,
2
0
1
5
.
[3
6
]
C.
El
m
a
s
a
n
d
Y.
S
o
n
m
e
z
,
“
A
d
a
t
a
f
u
sio
n
f
ra
m
e
w
o
rk
w
it
h
n
o
v
e
l
h
y
b
rid
a
lg
o
rit
h
m
f
o
r
m
u
lt
i
-
a
g
e
n
t
D
e
c
isio
n
S
u
p
p
o
r
t
S
y
st
e
m
f
o
r
F
o
re
st F
ire
,”
Ex
p
e
rt S
y
ste
ms
wit
h
Ap
p
li
c
a
ti
o
n
s
,
v
o
l.
38
,
n
o
.
8
,
p
p
.
9
2
2
5
-
9
2
3
6
,
2
0
1
1
.
[3
7
]
A
.
Ja
a
f
a
ri,
e
t
a
l.
,
“
W
il
d
f
ire
sp
a
ti
a
l
p
a
tt
e
rn
a
n
a
ly
sis
in
th
e
Zag
ro
s
M
o
u
n
tain
s,
Ira
n
:
A
c
o
m
p
a
ra
ti
v
e
s
tu
d
y
o
f
d
e
c
isio
n
tree
b
a
se
d
c
las
sif
ier
s
,”
Eco
lo
g
ic
a
l
In
fo
rm
a
ti
c
s
,
v
o
l.
43
,
p
p
.
2
0
0
-
2
1
1
,
2
0
1
8
.
[3
8
]
S.
S
a
c
h
d
e
v
a
,
e
t
a
l.
,
“
G
IS
-
b
a
se
d
e
v
o
lu
ti
o
n
a
ry
o
p
ti
m
ize
d
G
r
a
d
ien
t
Bo
o
ste
d
De
c
isio
n
T
re
e
s
fo
r
f
o
r
e
st
f
ire
su
sc
e
p
ti
b
il
it
y
m
a
p
p
in
g
,”
Na
tu
ra
l
Ha
za
rd
s
,
v
o
l
.
92
,
n
o
.
3
,
p
p
.
1
3
9
9
-
1
4
1
8
,
2
0
1
8
.
[3
9
]
Y.
Ca
o
,
e
t
a
l.
,
“
W
il
d
f
ire
su
sc
e
p
ti
b
i
li
ty
a
ss
e
ss
m
e
n
t
in
S
o
u
th
e
rn
Ch
i
n
a
:
A
c
o
m
p
a
riso
n
o
f
m
u
lt
ip
le
m
e
th
o
d
s
,”
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
Dis
a
ste
r R
isk
S
c
ien
c
e
,
v
o
l.
8
,
n
o
.
2
,
p
p
.
1
6
4
-
1
8
1
,
2
0
1
7
.
[4
0
]
S
.
J.
Kim
,
e
t
a
l
.,
“
M
u
lt
i
-
tem
p
o
r
a
l
a
n
a
ly
sis
o
f
f
o
re
st
f
ire
p
ro
b
a
b
il
it
y
u
sin
g
so
c
io
-
e
c
o
n
o
m
ic
a
n
d
e
n
v
iro
n
m
e
n
tal
v
a
riab
les
,”
Rem
o
te S
e
n
sin
g
,
v
o
l.
11
,
n
o
.
1
,
p
p
.
8
6
-
1
0
4
,
2
0
1
9
.
[4
1
]
C.
Zh
a
n
g
,
e
t
a
l.
,
“
A
h
y
b
rid
M
L
P
-
CNN
c
las
sif
ier
f
o
r
v
e
r
y
f
in
e
re
so
lu
ti
o
n
re
m
o
t
e
ly
se
n
se
d
ima
g
e
c
las
sif
i
c
a
ti
o
n
,”
IS
PR
S
J
o
u
r
n
a
l
o
f
Ph
o
to
g
ra
mm
e
try
a
n
d
Rem
o
te
S
e
n
sin
g
,
v
o
l
.
1
4
0
,
p
p
.
1
3
3
-
1
4
4
,
2
0
1
8
.
[4
2
]
G
.
E.
Hin
to
n
a
n
d
R.
R.
S
a
lak
h
u
td
i
n
o
v
,
“
Re
d
u
c
i
n
g
th
e
d
im
e
n
si
o
n
a
li
ty
o
f
d
a
ta
w
it
h
n
e
u
ra
l
n
e
tw
o
rk
s
,”
S
c
ien
c
e
,
v
o
l.
3
1
3
,
n
o
.
5
7
8
6
,
p
p
.
5
0
4
-
5
0
7
,
2
0
0
6
.
[4
3
]
Y.
L
e
c
u
n
,
e
t
a
l.
,
“
De
e
p
lea
rn
i
n
g
,”
Na
tu
re
,
v
o
l.
5
2
1
,
n
o
.
7
5
5
3
,
p
p
.
4
3
6
-
4
4
4
,
2
0
1
5
.
[4
4
]
K.
M
u
h
a
m
m
a
d
,
e
t
a
l.
,
“
Earl
y
f
ir
e
d
e
tec
ti
o
n
u
si
n
g
c
o
n
v
o
l
u
ti
o
n
a
l
n
e
u
ra
l
n
e
tw
o
rk
s
d
u
rin
g
su
rv
e
il
lan
c
e
f
o
r
e
ff
e
c
ti
v
e
d
isa
ste
r
m
a
n
a
g
e
m
e
n
t
,”
Ne
u
ro
c
o
m
p
u
ti
n
g
,
v
o
l.
2
8
8
,
p
p
.
30
-
42
,
2
0
1
8
.
[4
5
]
A
.
V
e
tri
v
e
l,
e
t
a
l.
,
“
Di
sa
ste
r
d
a
m
a
g
e
d
e
tec
ti
o
n
th
ro
u
g
h
sy
n
e
r
g
isti
c
u
se
o
f
d
e
e
p
l
e
a
rn
in
g
a
n
d
3
D
p
o
in
t
c
lo
u
d
f
e
a
tu
re
s
d
e
riv
e
d
f
ro
m
v
e
r
y
h
ig
h
re
so
lu
ti
o
n
o
b
li
q
u
e
a
e
rial
im
a
g
e
s,
a
n
d
m
u
lt
ip
le
-
k
e
rn
e
l
-
lea
rn
in
g
,”
I
S
P
RS
J
o
u
rn
a
l
o
f
Ph
o
t
o
g
r
a
mm
e
try
a
n
d
Rem
o
te
S
e
n
sin
g
,
v
o
l.
1
4
0
,
p
p
.
45
-
59
,
2
0
1
8
.
[4
6
]
T
.
L
iu
a
n
d
A
.
A
.
El
ra
h
m
a
n
,
“
De
e
p
c
o
n
v
o
lu
ti
o
n
a
l
n
e
u
ra
l
n
e
tw
o
rk
tr
a
in
in
g
e
n
rich
m
e
n
t
u
sin
g
m
u
lt
i
-
v
ie
w o
b
jec
t
-
b
a
se
d
a
n
a
ly
sis
o
f
Un
m
a
n
n
e
d
A
e
rial
s
y
st
e
m
s
i
m
a
g
e
r
y
f
o
r
w
e
tl
a
n
d
s
c
las
si
f
i
c
a
ti
o
n
,”
IS
PR
S
J
o
u
rn
a
l
o
f
P
h
o
t
o
g
ra
mm
e
try
a
n
d
Rem
o
te S
e
n
si
n
g
,
v
o
l.
1
3
9
,
p
p
.
1
5
4
-
1
7
0
,
2
0
1
8
.
[4
7
]
Y.
W
a
n
g
,
e
t
a
l.
,
“
Co
m
p
a
riso
n
o
f
c
o
n
v
o
l
u
ti
o
n
a
l
n
e
u
ra
l
n
e
tw
o
rk
s
fo
r
lan
d
slid
e
su
sc
e
p
ti
b
il
i
ty
m
a
p
p
in
g
in
Ya
n
sh
a
n
Co
u
n
ty
,
Ch
in
a
,”
S
c
ien
c
e
o
f
th
e
T
o
ta
l
E
n
v
iro
n
me
n
t
,
v
o
l
.
6
6
6
,
p
p
.
9
7
5
-
9
9
3
,
2
0
1
9
.
[4
8
]
W
.
R.
T
o
b
ler,
“
A
c
o
m
p
u
ter
m
o
v
ie
sim
u
latin
g
u
rb
a
n
g
ro
w
th
in
th
e
De
tro
i
t
re
g
io
n
,”
Eco
n
o
mi
c
Ge
o
g
ra
p
h
y
,
v
o
l.
46
,
n
o
.
1
,
p
p
.
2
3
4
-
2
4
0
,
1
9
7
0
.
[4
9
]
Y.
M
iao
,
e
t
a
l.
,
“
Co
m
p
re
h
e
n
siv
e
a
n
a
l
y
sis
o
f
n
e
t
w
o
rk
tra
ff
ic
d
a
ta,”
in
2
0
1
6
IEE
E
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
Co
mp
u
ter
a
n
d
In
fo
rm
a
t
io
n
T
e
c
h
n
o
lo
g
y
,
CIT
2
0
16
,
p
p
.
4
2
3
-
4
3
0
,
201
6
.
[5
0
]
A
.
Cu
tl
e
r,
e
t
a
l.
,
“
R
a
n
d
o
m
f
o
re
s
ts,”
in
C.
Zh
a
n
g
a
n
d
Y.
M
a
(e
d
s),
“
En
se
m
b
le
M
a
c
h
in
e
L
e
a
rn
in
g
:
M
e
th
o
d
s
a
n
d
A
p
p
li
c
a
ti
o
n
s
,
”
S
p
rin
g
e
r
,
pp
.
1
5
7
-
1
7
5
,
2
0
1
2
.
[5
1
]
P
.
W
.
W
a
n
g
a
n
d
C.
J.
L
in
,
“
S
u
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
s,”
in
C.
C.
Ag
g
a
r
wa
l
(e
d
),
“
Da
ta
Clas
si
f
ica
t
io
n
:
A
lg
o
rit
h
m
s
a
n
d
A
p
p
li
c
a
ti
o
n
s
,
”
CRC
Pre
ss
,
p
p
.
1
8
7
-
2
0
1
,
2
0
1
4
.
[5
2
]
S
.
D.
Ja
d
h
a
v
a
n
d
H.
P
.
Ch
a
n
n
e
,
“
Co
m
p
a
ra
ti
v
e
S
tu
d
y
o
f
K
-
NN
,
Na
iv
e
B
a
y
e
s
a
n
d
De
c
i
sio
n
T
re
e
Clas
si
f
ica
ti
o
n
T
e
c
h
n
iq
u
e
s,”
In
t
e
rn
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
S
c
i
e
n
c
e
a
n
d
Res
e
a
rc
h
,
v
o
l
.
5
,
n
o
.
1
,
p
p
.
1
8
4
2
-
1
8
4
5
,
2
0
1
6
.
[5
3
]
K.
M
a
larz
,
e
t
a
l.
,
“
A
r
e
F
o
re
st
F
ires
P
re
d
icta
b
le
?”
In
t
e
rn
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
M
o
d
e
rn
Ph
y
s
ics
C
,
v
o
l.
13
,
n
o
.
8
,
p
p
.
1
0
1
7
-
1
0
3
1
,
2
0
0
2
.
Evaluation Warning : The document was created with Spire.PDF for Python.