I
n
t
e
r
n
at
io
n
al
Jou
r
n
al
of
A
d
van
c
e
s
i
n
A
p
p
li
e
d
S
c
ie
n
c
e
s
(
I
JA
A
S
)
V
ol
.
14
, N
o.
4
,
D
e
c
e
m
be
r
20
25
, pp.
1146
~
1154
I
S
S
N
:
2252
-
8814
,
D
O
I
:
10.11591/
ij
a
a
s
.
v14.
i
4
.
pp1146
-
1154
1146
Jou
r
n
al
h
om
e
page
:
ht
tp
:
//
ij
aas
.i
ae
s
c
or
e
.c
om
Pr
e
d
i
c
t
i
on
i
n
d
e
x d
r
ou
gh
t
u
se
n
e
u
r
al
n
e
t
w
or
k
b
ase
d
r
ai
n
f
al
l
N
u
r
N
af
ii
yah
1,
2
, A
li
M
ok
h
t
ar
2
1
D
e
pa
r
t
m
e
nt
of
I
nf
or
m
a
t
i
c
s
, F
a
c
ul
t
y of
S
c
i
e
nc
e
a
nd T
e
c
hnol
ogy
, U
ni
ve
r
s
i
t
a
s
I
s
l
a
m
L
a
m
onga
n, L
a
m
onga
n, I
ndone
s
i
a
2
P
r
of
e
s
s
i
ona
l
E
ngi
ne
e
r
P
r
ogr
a
m
, U
ni
ve
r
s
i
t
a
s
M
uha
m
m
a
di
ya
h M
a
l
a
ng,
M
a
l
a
ng,
I
ndone
s
i
a
A
r
t
ic
le
I
n
f
o
A
B
S
T
R
A
C
T
A
r
ti
c
le
h
is
to
r
y
:
R
e
c
e
iv
e
d
J
un
24
,
2025
R
e
vi
s
e
d
O
c
t
21
,
2025
A
c
c
e
pt
e
d
N
ov
4
,
2025
Prolonge
d
dry
season
s
compar
ed
to
rainy
season
s
often
lead
to
dr
ought,
making
drought
index
observations
essential.
In
Indonesia,
d
rought
monitoring
commonly
uses
the
standardized
precipitation
index
(SP
I),
yet
there is no
common
standard
for droug
ht
index mea
surement.
Theref
o
re, this
research
applies
the
Z
-
score
ind
ex
(ZSI)
and
China
-
Z
ind
ex
(CZI),
which,
like
SPI,
are
rainfall
-
based
drought
indices
but
have
rarely
been
explored
in
previous
research.
To
predict
ZSI
and
CZI,
this
research
compar
es
the
weigh
ted
moving
average
(WMA)
and
multilayer
percep
tron
(MLP)
methods.
Two
input
scenario
s
are
tested:
the
previous
two
periods
(t
-
2,
t
-
1)
and
the
previous
three
periods
(t
-
3,
t
-
2,
t
-
1)
.
The
results
show
that
MLP
outperforms
WMA,
with
the
best
perform
ance
achieved
by
the
MLP
model
at
a
mean
absolute
percentage
error
(MAPE)
of
4.177%
using
the
three
-
variable input scenario and MLP arc
hitecture 3
-
6
-
10
-
1.
K
e
y
w
o
r
d
s
:
C
hi
na
-
z
in
de
x
I
nde
x dr
ought
M
ul
ti
la
ye
r
pe
r
c
e
pt
r
on
W
e
ig
ht
e
d m
ovi
ng a
ve
r
a
ge
Z
-
s
c
or
e
i
nde
x
This is an
open
acce
ss artic
le unde
r the
CC BY
-
SA
license.
C
or
r
e
s
pon
di
n
g A
u
th
or
:
N
ur
N
a
f
ii
ya
h
D
e
pa
r
tm
e
nt
of
I
nf
or
m
a
ti
c
s
, F
a
c
ul
ty
of
S
c
ie
nc
e
a
nd T
e
c
hnol
ogy
,
U
ni
ve
r
s
it
a
s
I
s
la
m
L
a
m
onga
n
S
t.
V
e
te
r
a
n
N
o.
53A
,
L
a
m
onga
n
62211, I
ndone
s
ia
E
m
a
il
:
m
yna
f
f
@
uni
s
la
.a
c
.i
d
1.
I
N
T
R
O
D
U
C
T
I
O
N
C
li
m
a
te
c
ha
nge
is
a
r
e
s
e
a
r
c
h
to
pi
c
th
a
t
c
ont
in
ue
s
to
be
r
e
s
e
a
r
c
he
d
by
a
c
a
de
m
ic
s
a
nd
pr
a
c
ti
ti
one
r
s
be
c
a
us
e
a
lm
os
t
th
e
e
nt
ir
e
w
or
ld
is
e
xp
e
r
ie
nc
in
g
pr
obl
e
m
s
,
na
m
e
ly
r
is
in
g
s
e
a
le
ve
l
s
,
te
m
pe
r
a
tu
r
e
,
e
xc
e
s
s
iv
e
r
a
in
f
a
ll
,
a
nd
dr
ought.
C
li
m
a
te
c
ha
nge
in
c
r
e
a
s
e
s
th
e
f
r
e
que
nc
y
of
e
xt
r
e
m
e
hyd
r
ol
ogy
,
s
uc
h
a
s
f
lo
ods
a
nd
dr
oughts
.
T
he
pr
obl
e
m
of
la
c
k
of
r
a
in
f
a
ll
c
a
u
s
e
s
c
li
m
a
te
pr
o
bl
e
m
s
a
nd
ha
s
a
n
im
pa
c
t
on
th
e
e
c
os
ys
t
e
m
.
E
xt
r
e
m
e
w
e
a
th
e
r
,
s
uc
h
a
s
dr
ought
,
is
in
f
lu
e
nc
e
d
by
r
a
in
f
a
ll
[
1]
,
[
2]
.
I
n
r
e
c
e
nt
ye
a
r
s
,
e
xt
r
e
m
e
w
e
a
th
e
r
ha
s
of
te
n
oc
c
ur
r
e
d.
D
r
ought
is
a
n
e
nvi
r
onm
e
nt
a
l
pr
obl
e
m
in
s
e
ve
r
a
l
c
ount
r
ie
s
,
in
c
lu
di
ng
I
ndone
s
i
a
.
S
e
ve
r
e
dr
ought
c
a
n
da
m
a
ge
va
r
io
us
f
ie
ld
s
s
uc
h
a
s
a
gr
ic
ul
tu
r
e
(
dr
y
la
nd
r
e
s
ul
ti
ng
in
c
r
op
f
a
il
ur
e
)
,
th
e
e
nvi
r
onm
e
nt
,
in
dus
tr
y
,
a
nd
hum
a
n
li
f
e
(
la
c
k
of
w
a
te
r
o
r
de
hydr
a
ti
on)
[
3]
,
[
4]
.
T
he
dr
ought
in
r
e
c
e
nt
ye
a
r
s
ha
s
c
ont
in
ue
d
to
in
c
r
e
a
s
e
,
c
a
us
in
g
a
s
hor
ta
ge
of
w
a
te
r
s
our
c
e
s
due
to
th
e
la
c
k
of
r
a
in
f
a
ll
[
5]
.
I
n
a
ddi
t
io
n
to
r
e
duc
e
d
r
a
in
f
a
ll
,
hum
a
n a
c
ti
vi
ti
e
s
c
a
n a
l
s
o a
f
f
e
c
t
dr
ought.
S
e
ve
r
a
l
c
ount
r
ie
s
a
r
e
s
tr
uggl
in
g
w
it
h
th
e
im
pa
c
ts
of
dr
ought.
T
he
im
pa
c
ts
of
dr
ought
in
c
lu
de
w
a
te
r
s
hor
ta
ge
s
or
w
a
te
r
a
va
il
a
bi
li
ty
,
de
c
r
e
a
s
e
d
a
gr
ic
ul
tu
r
a
l
pr
oduc
ti
vi
ty
,
f
ood
s
e
c
ur
it
y,
e
nvi
r
onm
e
nt
a
l
de
gr
a
da
ti
on
,
a
nd
ot
he
r
lo
s
s
e
s
[
4]
,
[
6]
.
D
r
ought
is
le
s
s
th
a
n
a
v
e
r
a
ge
r
a
i
nf
a
ll
in
a
pl
a
c
e
ove
r
a
lo
ng
pe
r
io
d
of
ti
m
e
[
7]
,
[
8]
.
T
ype
s
of
dr
ought
in
c
lu
de
:
m
e
t
e
or
ol
ogi
c
a
l
dr
ought,
w
hi
c
h
is
le
s
s
th
a
n
a
ve
r
a
g
e
r
a
in
f
a
ll
in
a
c
e
r
ta
in
a
r
e
a
,
hydr
ol
ogi
c
a
l
dr
ought
is
a
la
c
k
of
s
ur
f
a
c
e
a
nd
gr
oundwa
te
r
f
or
w
a
te
r
s
uppl
y,
a
nd
a
gr
ic
ul
tu
r
a
l
dr
ought
is
w
he
n c
r
ops
do not f
in
d
t
he
w
a
te
r
s
uppl
y t
he
y ne
e
d
[
9
]
, [
10
]
. O
ne
of
t
he
i
m
pa
c
ts
of
c
li
m
a
te
c
ha
nge
i
s
dr
ought
;
th
e
r
e
f
or
e
,
it
is
ve
r
y
ne
c
e
s
s
a
r
y
to
id
e
nt
if
y,
obs
e
r
ve
a
nd
a
na
ly
z
e
t
e
m
por
a
l
,
a
nd
s
pa
ti
a
l
dr
ought
pr
e
di
c
ti
ons
[
11]
,
[
12]
.
S
ig
ni
f
ic
a
nt
m
a
na
ge
m
e
nt
of
w
a
te
r
a
nd
la
nd
r
e
s
our
c
e
s
c
a
n
r
e
duc
e
e
nvi
r
onm
e
nt
a
l
da
m
a
ge
,
one
of
w
hi
c
h
i
s
m
oni
to
r
in
g
a
nd
p
r
e
di
c
ti
ng
dr
ough
ts
[
13]
,
[
14]
.
M
e
th
ods
th
a
t
a
r
e
of
te
n
us
e
d
to
pr
e
di
c
t
w
it
h
good
r
e
s
ul
ts
a
r
e
r
a
ndom f
or
e
s
t
[
15]
,
m
ul
ti
-
la
ye
r
pe
r
c
e
pt
r
on a
r
ti
f
ic
ia
l
ne
ur
a
l
ne
tw
or
k
(
M
L
P
-
ANN)
[
16]
, [
17]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
dv A
ppl
S
c
i
I
S
S
N
:
2252
-
8814
P
r
e
di
c
ti
on i
nde
x
dr
ought us
e
ne
ur
al
ne
t
w
or
k
ba
s
e
d r
ai
nf
al
l
(
N
ur
N
af
ii
y
ah)
1147
M
it
ig
a
ti
on
s
tr
a
te
gi
e
s
th
a
t
c
a
n
be
us
e
d
to
r
e
duc
e
th
e
im
pa
c
t
of
dr
ought
in
c
lu
de
dr
ought
m
oni
to
r
in
g
a
nd
a
s
s
e
s
s
m
e
nt
[
18]
.
D
r
ought
m
oni
to
r
in
g
to
ol
s
th
a
t
c
a
n
be
us
e
d
a
r
e
dr
ought
in
di
c
e
s
in
c
lu
di
ng
th
e
s
ta
nda
r
di
z
e
d
pr
e
c
ip
it
a
ti
on
in
de
x
(
S
P
I
)
,
Z
-
s
c
or
e
in
de
x
(
Z
S
I
)
,
C
hi
na
-
Z
in
de
x
(
C
Z
I
)
,
r
a
in
f
a
ll
a
nom
a
ly
in
de
x
(
R
A
I
)
,
r
a
in
f
a
ll
de
pa
r
tu
r
e
(
R
D
)
,
pr
e
c
ip
it
a
ti
on
de
c
il
e
s
(
P
D
)
,
de
c
il
e
s
in
de
x
(
D
I
)
,
a
nd
pe
r
c
e
nt
of
nor
m
a
l
in
de
x
(
P
N
I
)
w
hi
c
h
a
r
e
dr
ought
in
di
c
e
s
ba
s
e
d
on
r
a
in
f
a
ll
[
2]
,
[
19]
–
[
21]
.
T
he
s
ta
nda
r
di
z
e
d
pr
e
c
ip
it
a
ti
on
e
va
pot
r
a
ns
pi
r
a
ti
on
in
de
x
(
S
P
E
I
)
is
a
dr
ought
in
de
x
b
a
s
e
d
o
n
te
m
pe
r
a
tu
r
e
[
22]
.
T
h
e
in
de
x
f
or
a
s
s
e
s
s
in
g
dr
ought
s
it
ua
ti
ons
is
th
e
P
a
lm
e
r
dr
ought
s
e
ve
r
it
y
in
de
x
(
P
D
S
I
)
,
a
nd
a
s
ta
ti
s
ti
c
a
l
dow
ns
c
a
li
ng
m
ode
l
(
S
D
S
M
)
is
ve
r
y
good
f
or
pr
e
di
c
ti
ng
dr
ought
in
a
r
id
a
nd
s
e
m
i
-
a
r
id
a
r
e
a
s
[
23]
.
T
he
dr
ought
in
de
x
c
a
n
pr
ovi
de
in
f
or
m
a
ti
on
r
e
ga
r
di
ng
th
e
a
r
e
a
,
s
e
ve
r
it
y,
dur
a
ti
on
,
a
nd
f
r
e
que
n
c
y
of
dr
ought
[
24]
.
T
he
dr
ought
in
de
x
th
a
t
is
f
r
e
que
nt
ly
obs
e
r
ve
d,
a
na
ly
z
e
d,
a
nd
pr
e
di
c
te
d
i
s
th
e
S
P
I
[
25]
–
[
27]
.
S
e
ve
r
a
l
m
e
th
ods
us
e
d
to
pr
e
di
c
t
th
e
dr
ought
in
de
x
in
c
lu
de
ne
ur
a
l
ne
twor
ks
,
f
uz
z
y
lo
gi
c
,
a
nd
lo
ng
s
hor
t
-
te
r
m
m
e
m
or
y
(
L
S
T
M
)
[
28]
–
[
30]
.
C
om
pa
r
in
g
th
e
e
m
pi
r
ic
a
l
m
ode
de
c
om
pos
it
io
n
(
E
M
D
)
,
de
tr
e
nd
e
d
f
lu
c
tu
a
ti
on
a
na
ly
s
is
(
D
F
A
)
a
nd
de
e
p
be
li
e
f
ne
twor
k
(
D
B
N
)
,
a
nd
m
ul
ti
la
ye
r
pe
r
c
e
pt
r
on
(
M
L
P
)
m
ode
ls
in
pr
e
di
c
ti
ng
S
P
I
,
w
it
h
th
e
E
M
D
-
D
F
A
r
e
s
ul
ts
pr
ovi
di
ng
a
c
c
ur
a
te
dr
ought
in
de
x
pr
e
di
c
ti
on
r
e
s
ul
ts
[
31]
.
T
h
e
c
li
m
a
te
ha
z
a
r
ds
gr
oup
in
f
r
a
r
e
d
pr
e
c
ip
it
a
ti
on
w
it
h
s
ta
ti
on
da
ta
(
C
H
I
R
P
S
)
r
a
in
f
a
ll
da
ta
s
e
t
s
how
s
th
a
t
th
e
dr
o
ught
in
di
c
e
s
C
Z
I
,
S
P
I
,
a
nd
Z
S
I
c
a
n
e
f
f
e
c
ti
ve
ly
de
te
c
t
dr
ought,
a
nd
th
e
r
e
s
ul
ts
of
s
pa
ti
ot
e
m
por
a
l
dr
ought
c
o
ndi
ti
on
a
na
ly
s
is
c
a
n
be
u
s
e
d
f
or
pol
ic
y
a
nd
s
us
ta
in
a
bl
e
de
ve
lo
pm
e
nt
[
32]
.
A
na
ly
z
in
g
dr
ought
vul
ne
r
a
bi
li
ty
us
in
g
th
e
a
na
ly
ti
c
hi
e
r
a
r
c
hy
pr
oc
e
s
s
(
AHP
)
ba
s
e
d
on
gr
oundwa
t
e
r
r
e
s
our
c
e
s
in
de
x,
w
a
te
r
w
a
y
de
ns
it
y
in
de
x,
c
li
m
a
te
in
de
x,
la
nd
us
e
in
d
e
x,
a
nd
to
pogr
a
phy
in
de
x
in
di
c
a
to
r
s
,
r
e
s
e
a
r
c
h
r
e
s
ul
ts
[
33]
th
a
t
la
nd
us
e
a
f
f
e
c
ts
dr
ought
vul
ne
r
a
bi
li
ty
a
nd
r
is
k.
H
ybr
i
d
m
ode
ls
f
or
a
na
ly
z
in
g
th
e
r
e
c
onna
is
s
a
nc
e
dr
ought
in
de
x
(
R
D
I
)
,
na
m
e
ly
s
uppor
t
ve
c
to
r
r
e
gr
e
s
s
io
n
(
S
V
R
)
a
nd
w
a
ve
le
t
a
na
ly
s
i
s
(
W
)
c
a
n
be
us
e
d
to
pr
e
di
c
t
good
dr
ought
w
it
h
r
oot
m
e
a
n
s
qua
r
e
e
r
r
or
(
R
M
S
E
)
=
0.301,
m
e
a
n
a
bs
ol
ut
e
e
r
r
or
(
M
A
E
)
=
0.166,
W
il
l
m
ot
t
in
de
x
(
W
I
)
=
0.910,
N
a
s
h
–
S
ut
c
li
f
f
e
e
f
f
ic
ie
nc
y
(
N
S
E
)
=
0.936
[
34]
.
T
he
de
c
is
io
n
tr
e
e
(
D
T
)
m
e
th
od
ha
s
good
r
e
s
ul
t
s
in
pr
e
di
c
ti
ng
th
e
S
P
I
dr
ought
in
de
x
a
nd
a
s
s
e
s
s
in
g
dr
ought
m
it
ig
a
ti
on
[
35]
.
T
he
nonl
in
e
a
r
a
ut
or
e
gr
e
s
s
iv
e
ne
ur
a
l
n
e
twor
k
(
N
A
R
N
N
)
m
ode
l
is
th
e
be
s
t
a
lg
or
it
hm
f
or
pr
e
di
c
ti
ng t
he
S
P
I
dr
ought i
nde
x w
it
h R
M
S
E
=
0.997
[
36]
.
I
ndone
s
ia
is
a
tr
opi
c
a
l
c
ount
r
y
w
it
h
two
s
e
a
s
ons
,
na
m
e
ly
th
e
r
a
in
y
a
nd
dr
y
s
e
a
s
ons
.
I
n
r
e
c
e
nt
y
e
a
r
s
,
c
li
m
a
te
c
ha
nge
ha
s
di
s
r
upt
e
d
th
e
us
u
a
l
s
e
a
s
ona
l
pa
tt
e
r
ns
,
l
e
a
di
ng
to
pr
ol
onge
d
dr
y
or
r
a
in
y
s
e
a
s
ons
.
A
lo
nge
r
dr
y
s
e
a
s
on
c
a
n
c
a
u
s
e
dr
ought,
hi
ghl
ig
ht
in
g
th
e
im
por
ta
nc
e
of
dr
ought
in
de
x
obs
e
r
va
ti
ons
.
I
n
I
ndone
s
ia
,
dr
ought m
oni
to
r
in
g
of
te
n us
e
s
t
he
S
P
I
, but
t
he
r
e
i
s
no ge
ne
r
a
l
s
ta
nda
r
d f
or
dr
ought
i
nde
x m
e
a
s
ur
e
m
e
nt
. T
hus
,
th
is
r
e
s
e
a
r
c
h a
dopt
s
th
e
Z
S
I
a
nd
C
Z
I
,
w
hi
c
h,
li
ke
S
P
I
,
a
r
e
r
a
in
f
a
ll
-
ba
s
e
d
dr
ought
in
di
c
e
s
but
ha
ve
r
a
r
e
ly
be
e
n
in
ve
s
ti
ga
te
d
in
pr
e
di
c
ti
ve
r
e
s
e
a
r
c
h
[
19]
.
P
r
e
vi
ous
r
e
s
e
a
r
c
h
ha
s
m
a
in
ly
f
oc
u
s
e
d
on
S
P
I
pr
e
di
c
ti
on
us
in
g
m
e
th
ods
s
uc
h
a
s
w
a
ve
le
t
-
de
c
om
po
s
e
d
hybr
id
m
ode
ls
(
W
B
R
F
)
,
bi
-
di
r
e
c
ti
ona
l
lo
ng
s
hor
t
-
te
r
m
m
e
m
or
y
(
B
i
-
L
S
T
M
)
[
27]
;
c
om
pl
e
m
e
nt
a
r
y
e
ns
e
m
bl
e
e
m
pi
r
ic
a
l
m
ode
de
c
om
pos
it
io
n
(
C
E
E
M
D
)
w
it
h
L
S
T
M
[
25]
;
a
nd
E
M
D
-
e
xt
r
e
m
e
le
a
r
ni
ng
m
a
c
hi
ne
(
E
L
M
)
hybr
id
m
ode
ls
[
20]
.
T
he
s
e
a
ppr
oa
c
he
s
s
how
th
a
t
ne
ur
a
l
ne
twor
ks
a
nd
ti
m
e
s
e
r
ie
s
m
ode
ls
a
c
hi
e
ve
good
r
e
s
ul
ts
.
H
ow
e
ve
r
,
r
e
s
e
a
r
c
h
th
a
t
pr
e
di
c
ts
Z
S
I
a
nd
C
Z
I
in
d
ic
e
s
r
e
m
a
in
s
li
m
it
e
d,
e
s
p
e
c
ia
ll
y
u
s
in
g
r
e
la
ti
ve
ly
s
im
pl
e
m
ode
ls
f
or
c
om
pa
r
is
on.
T
o
a
ddr
e
s
s
th
is
g
a
p,
th
is
r
e
s
e
a
r
c
h
pr
opos
e
s
th
e
w
e
ig
ht
e
d
m
ovi
ng
a
ve
r
a
ge
(
W
M
A
)
a
nd
M
L
P
m
e
th
ods
f
or
pr
e
di
c
ti
ng
Z
S
I
a
nd
C
Z
I
in
di
c
e
s
.
T
w
o
pr
e
di
c
ti
on
s
c
e
na
r
io
s
a
r
e
a
ppl
ie
d,
ba
s
e
d
on
th
e
pr
e
vi
ous
two
pe
r
io
ds
(
−
2
,
−
1
)
a
nd
th
e
pr
e
vi
ous
th
r
e
e
pe
r
io
ds
(
−
3
,
−
2
,
−
1
)
.
T
hi
s
r
e
s
e
a
r
c
h
a
im
s
to
id
e
nt
if
y
th
e
m
os
t
a
ppr
opr
ia
te
m
ode
l
a
nd
in
put
s
c
e
n
a
r
io
f
or
a
c
c
ur
a
te
ly
pr
e
di
c
ti
ng Z
S
I
a
nd C
Z
I
d
r
ought i
ndi
c
e
s
.
2.
M
E
T
H
O
D
2.1.
D
at
as
e
t
D
a
ta
w
a
s
dow
nl
oa
de
d f
r
om
t
he
w
e
bs
it
e
ht
tp
s
:
//
hi
dr
ol
ogi
.dpuai
r
.j
at
impr
ov
.go.i
d/
pe
la
y
anan/
,
th
e
a
r
e
a
r
e
s
e
a
r
c
he
d
w
a
s
P
a
nda
nl
a
r
a
s
s
ta
ti
on,
K
r
uc
il
s
ub
-
di
s
tr
ic
t,
P
r
obol
in
ggo
di
s
tr
ic
t,
E
a
s
t
J
a
va
pr
ovi
nc
e
.
D
a
ta
in
th
e
f
or
m
of
r
a
in
f
a
ll
(
m
m
)
e
a
c
h
m
ont
h
f
r
om
2003
to
2023.
D
a
ta
f
r
om
th
is
r
e
s
e
a
r
c
h
a
r
e
a
s
in
F
ig
ur
e
1,
w
it
h
a
n
a
ve
r
a
ge
va
lu
e
of
287.19
m
m
;
s
ta
nda
r
d
de
vi
a
ti
on
197.15
m
m
;
m
in
im
um
3
m
m
;
m
a
xi
m
u
m
860
m
m
;
a
nd
m
e
di
a
n
281
m
m
.
R
a
in
f
a
ll
da
ta
is
th
e
n
p
r
oc
e
s
s
e
d
to
c
a
lc
ul
a
t
e
th
e
dr
ought
in
de
x
Z
S
I
a
nd
C
Z
I
,
w
it
h
th
e
f
or
m
ul
a
s
Z
S
I
a
s
i
n
(
1)
a
nd C
Z
I
(
2)
f
r
om
t
he
r
e
s
e
a
r
c
h
[
2]
.
=
(
−
)
(
1)
=
(
6
(
2
+
1
)
1
3
⁄
)
−
(
6
+
6
)
(
2)
D
e
s
c
r
ip
ti
on
of
(
1
)
,
µ
is
th
e
a
ve
r
a
ge
,
w
it
h
(
3
)
,
σ
is
th
e
s
ta
nda
r
d
de
vi
a
ti
on
,
a
nd
w
it
h
(
4
)
.
I
n
(
2
)
,
is
th
e
s
ke
w
ne
s
s
c
o
e
f
f
ic
ie
nt
w
it
h
(
5
)
,
is
t
he
s
ta
nd
a
r
d va
r
ia
te
w
it
h
(
6
)
.
=
∑
=
1
(
3)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8814
I
nt
J
A
dv A
ppl
S
c
i
,
V
ol
.
14
, N
o.
4
,
D
e
c
e
m
be
r
20
25
:
1146
-
1154
1148
=
√
∑
(
−
)
2
=
1
(
4)
=
∑
(
−
)
3
=
1
3
(
5)
=
−
(
6)
F
ig
ur
e
1
.
R
a
in
f
a
ll
da
ta
T
he
r
e
s
ul
ts
of
th
e
c
a
lc
ul
a
ti
on
of
th
e
Z
S
I
dr
ought
in
de
x
,
a
s
s
how
n
in
F
ig
ur
e
2
.
C
Z
I
a
s
s
how
n
in
F
ig
ur
e
3,
w
it
h
th
e
m
e
a
n,
s
ta
nda
r
d
de
vi
a
ti
on,
m
in
im
um
,
m
a
xi
m
um
,
a
nd
m
e
di
a
n
va
lu
e
s
in
T
a
bl
e
1.
B
a
s
e
d
on
T
a
bl
e
1,
th
e
s
ta
nd
a
r
d
de
vi
a
ti
on
va
lu
e
is
1,
m
e
a
ni
ng
a
s
m
a
ll
di
s
pe
r
s
io
n
va
lu
e
w
it
h
a
di
s
ta
nc
e
v
a
lu
e
of
1
f
r
om
th
e
a
ve
r
a
ge
.
F
ig
ur
e
2. Z
S
I
d
r
ought i
nde
x
F
ig
ur
e
3. C
Z
I
dr
ought i
nde
x
T
a
bl
e
1. D
r
ought i
nde
x s
ta
ti
s
ti
c
a
l
va
lu
e
s
V
a
l
ue
Z
S
I
C
Z
I
M
e
a
n
1
.
19
×
10
−
16
-
0.124
S
t
a
nda
r
d de
vi
a
t
i
on
1
0.992
M
i
ni
m
um
-
1.442
-
1.656
M
a
xi
m
um
2.905
2.435
M
e
di
a
n
-
0.031
-
0.094
2.2.
P
r
op
os
e
d
m
e
t
h
od
T
hi
s
r
e
s
e
a
r
c
h
pr
e
di
c
ts
th
e
Z
S
I
a
nd
C
Z
I
dr
ought
in
d
ic
e
s
us
in
g
th
e
W
M
A
a
nd
M
L
P
m
e
th
ods
.
U
nl
ik
e
pr
e
vi
ous
r
e
s
e
a
r
c
h
[
20]
,
w
hi
c
h
f
oc
us
e
d
on
S
P
I
pr
e
di
c
ti
on,
th
is
r
e
s
e
a
r
c
h
a
ppl
ie
s
di
f
f
e
r
e
nt
dr
ought
in
di
c
e
s
(
Z
S
I
a
nd
C
Z
I
)
a
nd
c
om
pa
r
e
s
a
s
ta
ti
s
ti
c
a
l
a
ppr
oa
c
h
,
W
M
A
,
w
it
h
a
ne
ur
a
l
ne
twor
k
a
ppr
oa
c
h
,
M
L
P
.
T
he
in
put
da
ta
a
r
e
Z
S
I
a
nd
C
Z
I
in
di
c
e
s
de
r
iv
e
d
f
r
om
r
a
in
f
a
ll
r
e
c
or
ds
.
T
w
o
in
put
s
c
e
na
r
io
s
a
r
e
te
s
te
d:
s
c
e
na
r
io
1
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
dv A
ppl
S
c
i
I
S
S
N
:
2252
-
8814
P
r
e
di
c
ti
on i
nde
x
dr
ought us
e
ne
ur
al
ne
t
w
or
k
ba
s
e
d r
ai
nf
al
l
(
N
ur
N
af
ii
y
ah)
1149
(
two
in
put
s
)
:
d
r
ought
in
di
c
e
s
a
t
(
1
)
-
1,
(
2
)
-
2
.
S
c
e
na
r
io
2
(
th
r
e
e
in
put
s
)
:
dr
ought
in
di
c
e
s
a
t
(
1
)
−
3
,
(
2
)
−
2
,
a
nd
(
3
)
−
1
.
F
or
th
e
W
M
A
m
e
th
od,
th
e
pr
e
di
c
ti
on
us
e
s
a
W
M
A
,
w
he
r
e
th
e
dr
ought
in
di
c
e
s
a
t
(
1
)
−
3
,
(
2
)
−
2
,
a
nd
(
3
)
−
1
a
r
e
m
ul
ti
pl
ie
d
by
w
e
ig
ht
s
w
=
[
1,
2,
3]
a
s
in
(
7
)
[
37]
.
F
or
th
e
M
L
P
m
e
th
od,
th
e
in
put
va
r
ia
bl
e
s
a
r
e
th
e
dr
ought
in
di
c
e
s
(
1
)
−
3
,
(
2
)
−
2
,
(
3
)
−
1
,
de
p
e
ndi
ng
on
th
e
s
c
e
na
r
io
. T
he
M
L
P
a
r
c
hi
te
c
tu
r
e
i
s
opt
im
iz
e
d t
o m
in
im
iz
e
pr
e
di
c
ti
on e
r
r
or
, w
it
h
t
he
out
put
be
in
g t
he
p
r
e
di
c
te
d
Z
S
I
or
C
Z
I
va
lu
e
a
t
ti
m
e
t.
T
he
w
or
kf
lo
w
of
th
e
pr
opos
e
d
m
e
t
hod
is
s
how
n
in
F
ig
ur
e
4,
w
h
e
r
e
Z
S
I
a
nd
C
Z
I
in
de
x
da
ta
a
r
e
pr
oc
e
s
s
e
d
a
s
in
put
s
in
to
bot
h
W
M
A
a
nd
M
L
P
m
ode
ls
unde
r
th
e
de
f
in
e
d
s
c
e
n
a
r
io
s
.
T
a
bl
e
2
i
s
a
n e
xa
m
pl
e
of
pr
e
di
c
ti
ng
W
M
A
w
it
h t
w
o i
nput
a
nd t
h
r
e
e
i
nput
s
c
e
na
r
io
s
. T
h
e
w
e
ig
ht
s
i
n t
hi
s
r
e
s
e
a
r
c
h a
r
e
t
he
r
e
s
ul
ts
of
e
xpe
r
im
e
nt
s
t
ha
t
ha
ve
t
he
b
e
s
t
pr
e
di
c
ti
on a
c
c
ur
a
c
y.
=
−
1
+
−
2
+
−
3
∑
(
7)
F
ig
ur
e
4. R
e
s
e
a
r
c
h pr
opos
e
d
T
a
bl
e
2
.
W
M
A
pr
e
di
c
ti
ons
A
c
t
ua
l
da
t
a
P
r
e
di
c
t
i
on da
t
a
(
=
1
.
−
1
+
2
−
2
3
)
P
r
e
di
c
t
i
on da
t
a
(
=
1
.
−
1
+
2
−
2
+
3
−
3
6
)
-
0.448
0.965
0.265
=
(
1
∗
0
.
9
6
5
)
+
(
2
∗
−
0
.
4
4
8
)
3
=0.667
-
1.429
=
(
1
∗
0
.
2
6
5
)
+
(
2
∗
0
.
9
6
5
)
3
=0.909
=
(
1
∗
0
.
2
6
5
)
+
(
2
∗
0
.
9
6
5
)
+
(
3
∗
−
0
.
4
4
8
)
6
=1.972
-
0.976
=
(
1
∗
−
1
.
4
2
9
)
+
(
2
∗
0
.
2
6
5
)
3
=
-
1.252
=
(
1
∗
−
1
.
4
2
9
)
+
(
2
∗
0
.
2
6
5
)
+
(
3
∗
0
.
9
6
5
)
6
=
-
0.899
-
1.551
=
(
1
∗
−
0
.
9
7
6
)
+
(
2
∗
−
1
.
4
2
9
)
3
=
-
1.929
=
(
1
∗
−
0
.
9
7
6
)
+
(
2
∗
−
1
.
4
2
9
)
+
(
3
∗
0
.
265
)
6
=
-
3.834
-
1.551
=
(
1
∗
−
1
.
5
5
1
)
+
(
2
∗
−
0
.
9
7
6
)
3
=
-
2.202
=
(
1
∗
−
1
.
5
5
1
)
+
(
2
∗
−
0
.
9
7
6
)
+
(
3
∗
−
1
.
4
2
9
)
6
=
-
3.504
-
1.057
=
(
1
∗
−
1
.
5
5
1
)
+
(
2
∗
−
1
.
5
5
1
)
3
=
-
2.585
=
(
1
∗
−
1
.
5
5
1
)
+
(
2
∗
−
0
.
9
7
6
)
+
(
3
∗
−
1
.
4
2
9
)
6
=
-
4.653
T
he
M
L
P
m
e
th
od
in
th
is
r
e
s
e
a
r
c
h
us
e
s
th
e
be
s
t
a
r
c
hi
te
c
tu
r
a
l
s
c
e
na
r
io
ba
s
e
d
on
e
xpe
r
im
e
nt
s
,
a
s
s
how
n
in
F
ig
ur
e
5.
T
he
M
L
P
m
e
th
od
u
s
e
s
two
in
put
s
w
it
h
th
e
a
r
c
hi
te
c
tu
r
e
[
2
-
4
-
6
-
1]
(
F
ig
ur
e
5(
a
)
)
,
na
m
e
ly
two
ne
ur
ons
in
th
e
in
put
la
ye
r
,
f
our
ne
ur
ons
in
th
e
hi
dde
n
l
a
ye
r
,
s
ix
ne
ur
ons
in
th
e
hi
dde
n
la
ye
r
,
a
nd
one
ne
ur
on
in
th
e
out
put
la
ye
r
,
a
nd
th
r
e
e
in
put
s
w
it
h
th
e
a
r
c
hi
te
c
tu
r
e
[
3
-
6
-
10
-
1]
(
F
ig
ur
e
5(
b
)
)
,
na
m
e
ly
th
r
e
e
ne
ur
ons
in
th
e
in
put
la
y
e
r
,
s
ix
ne
ur
ons
in
th
e
hi
dde
n
la
ye
r
,
te
n ne
ur
ons
in
th
e
hi
dde
n
la
ye
r
,
a
nd
on
e
ne
ur
on
in
th
e
out
put
la
ye
r
.
T
he
M
L
P
m
e
th
od
in
th
is
r
e
s
e
a
r
c
h
us
e
s
2
hi
dde
n
la
ye
r
s
,
a
nd
th
e
out
put
la
ye
r
ha
s
1
ne
ur
on.
E
a
c
h
hi
dde
n
la
ye
r
a
nd
out
put
l
a
ye
r
ha
s
two
pr
oc
e
s
s
e
s
,
na
m
e
ly
t
he
di
r
e
c
ti
on
of
th
e
in
c
om
in
g
a
r
r
ow
c
a
ll
e
d
,
a
nd
th
e
di
r
e
c
ti
on
of
th
e
a
r
r
ow
is
c
a
ll
e
d
out
,
.
A
ll
lo
gi
n
pr
oc
e
s
s
e
s
(
,
)
c
a
lc
ul
a
te
th
e
in
put
w
it
h
w
e
ig
ht
s
a
s
in
(
8)
,
de
s
c
r
ip
ti
on
is
th
e
w
e
ig
ht
a
nd
is
th
e
bi
a
s
.
T
o
opt
im
iz
e
le
a
r
ni
ng,
th
e
M
L
P
m
ode
l
a
ppl
ie
s
th
e
m
e
a
n
a
b
s
ol
ut
e
pe
r
c
e
nt
a
g
e
e
r
r
or
(
M
A
P
E
)
a
s
th
e
l
os
s
f
unc
ti
on,
w
it
h
th
e
R
M
S
pr
op
opt
im
iz
e
r
to
a
c
c
e
le
r
a
te
c
onve
r
ge
nc
e
,
le
a
r
ni
ng
r
a
te
=
0.01.
T
he
m
ode
l
is
tr
a
i
ne
d
us
in
g
100
e
poc
hs
a
nd
a
ba
tc
h
s
iz
e
of
2,
w
hi
c
h
w
e
r
e
f
ound
to
pr
ovi
de
s
ta
bl
e
c
onve
r
ge
nc
e
a
nd
r
e
li
a
bl
e
pr
e
di
c
ti
on
pe
r
f
or
m
a
nc
e
a
c
r
os
s
di
f
f
e
r
e
nt
in
put
s
c
e
na
r
io
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8814
I
nt
J
A
dv A
ppl
S
c
i
,
V
ol
.
14
, N
o.
4
,
D
e
c
e
m
be
r
20
25
:
1146
-
1154
1150
(
a
)
(
b)
F
ig
ur
e
5
.
M
L
P
a
r
c
hi
te
c
tu
r
e
of
(
a
)
2
-
in
put
s
a
nd
(
b)
3
-
in
put
s
=
+
∑
∗
=
1
(
8)
T
hi
s
r
e
s
e
a
r
c
h
pr
e
di
c
ts
th
e
Z
S
I
a
nd
C
Z
I
dr
ought
in
di
c
e
s
us
in
g
s
e
ve
r
a
l
m
e
th
ods
.
T
he
m
os
t
a
ppr
opr
ia
te
m
e
th
od
f
or
pr
e
di
c
ti
ng t
he
Z
S
I
a
nd C
Z
I
dr
ought i
nd
ic
e
s
i
s
e
va
lu
a
te
d us
in
g t
he
M
A
P
E
(
9
)
[
27]
.
=
1
∑
|
ℎ
−
ℎ
|
|
ℎ
|
=
1
∗
100
(
9)
3.
R
E
S
U
L
T
S
A
N
D
D
I
S
C
U
S
S
I
O
N
T
hi
s
r
e
s
e
a
r
c
h
pr
e
di
c
ts
th
e
Z
S
I
a
nd
C
Z
I
dr
ought
in
di
c
e
s
us
in
g
th
e
W
M
A
a
nd
M
L
P
m
e
th
ods
.
T
he
in
put
s
of
th
is
r
e
s
e
a
r
c
h
a
r
e
th
e
Z
S
I
a
nd
C
Z
I
dr
ought
in
di
c
e
s
in
t
he
pe
r
io
d
(
−
1
,
−
2
,
−
3
)
.
D
r
ought
in
de
x
va
lu
e
f
r
om
r
a
in
f
a
ll
da
ta
c
a
lc
ul
a
ti
on.
F
ig
ur
e
6
is
th
e
r
e
s
ul
t
of
W
M
A
pr
e
di
c
ti
on
b
a
s
e
d
on
2
-
m
ont
h
a
nd
3
-
m
ont
h
pe
r
io
d
s
on
th
e
Z
S
I
a
nd
C
Z
I
dr
ought
in
de
x
;
th
e
o
r
a
nge
c
ol
or
gr
a
ph
is
th
e
pr
e
di
c
ti
on
r
e
s
ul
t,
a
nd
bl
ue
is
th
e
a
c
tu
a
l
da
ta
.
F
ig
ur
e
6(
a
)
s
how
s
th
e
Z
S
I
W
M
A
2
ti
m
e
s
,
F
ig
ur
e
6
(
b)
s
how
s
th
e
Z
S
I
W
M
A
3
ti
m
e
s
,
F
ig
ur
e
6(
c
)
s
how
s
th
e
C
Z
I
W
M
A
2
ti
m
e
s
,
a
nd
F
ig
ur
e
6(
d)
s
how
s
th
e
C
Z
I
W
M
A
3
ti
m
e
s
.
T
a
bl
e
3
is
th
e
s
ta
ti
s
ti
c
a
l
va
lu
e
of
th
e
Z
S
I
a
nd
C
Z
I
in
de
x
pr
e
di
c
ti
on
r
e
s
ul
t
w
it
h
W
M
A
.
T
a
bl
e
4
is
th
e
M
A
P
E
va
lu
e
f
r
om
th
e
Z
S
I
a
nd
C
Z
I
in
de
x pr
e
di
c
ti
on r
e
s
ul
t
w
it
h W
M
A
. B
a
s
e
d on T
a
bl
e
4
,
th
e
W
M
A
m
e
th
od ha
s
t
he
l
ow
e
s
t
pe
r
c
e
nt
a
ge
e
r
r
or
w
it
h
2 t
im
e
s
2
-
pe
r
io
d
(
−
1
,
−
2
)
, a
nd i
n F
ig
ur
e
6
,
th
e
or
a
nge
gr
a
ph i
s
a
lm
os
t
c
l
os
e
t
o t
he
bl
ue
gr
a
ph.
T
a
bl
e
5
is
th
e
M
A
P
E
va
lu
e
of
th
e
M
L
P
m
e
th
od
in
pr
e
di
c
ti
ng
th
e
dr
ought
in
de
x.
T
h
e
M
L
P
m
e
th
od
c
onduc
te
d
tr
a
in
in
g
a
nd
e
va
lu
a
ti
on
e
xpe
r
im
e
nt
s
f
iv
e
ti
m
e
s
,
b
y
s
e
tt
in
g
th
e
opt
im
iz
e
r
to
R
M
S
pr
op,
w
it
h
a
le
a
r
ni
ng
r
a
te
of
0.01,
e
poc
hs
of
100,
a
nd
ba
tc
h
s
iz
e
2.
B
a
s
e
d
on
T
a
bl
e
5,
th
e
M
L
P
m
e
th
od
ha
s
th
e
lo
w
e
s
t
M
A
P
E
w
it
h
th
r
e
e
in
put
va
r
ia
bl
e
s
1
=
−
3
,
2
=
−
2
,
3
=
−
1
.
B
a
s
e
d
on
T
a
bl
e
s
4
a
nd
5,
t
he
s
m
a
ll
e
s
t
M
A
P
E
va
lu
e
of
th
e
W
M
A
a
nd
M
L
P
m
e
th
ods
is
th
e
M
L
P
m
e
th
od,
w
hi
c
h
is
4.177%
,
m
e
a
ni
ng
th
e
e
r
r
or
va
lu
e
is
4%
a
ga
in
s
t
th
e
a
c
tu
a
l
da
ta
.
F
ig
ur
e
7
s
how
s
th
e
r
e
s
ul
t
of
t
he
M
L
P
pr
e
di
c
ti
on
w
it
h
two
v
a
r
ia
bl
e
s
1
=
−
2
,
2
=
−
1
,
a
nd
th
r
e
e
va
r
ia
bl
e
s
1
=
−
3
,
2
=
−
2
,
3
=
−
1
;
F
ig
ur
e
7
(
a
)
s
how
s
th
e
Z
S
I
2
va
r
ia
bl
e
s
,
F
ig
ur
e
7(
b)
s
how
s
th
e
Z
S
I
3
v
a
r
ia
bl
e
s
,
F
ig
ur
e
7(
c
)
s
how
s
th
e
C
Z
I
2
va
r
ia
bl
e
s
,
a
nd
F
ig
ur
e
7(
d)
s
how
s
th
e
C
Z
I
3
va
r
ia
bl
e
s
.
B
a
s
e
d
on
T
a
bl
e
5,
th
e
M
L
P
m
e
th
od
ha
s
th
e
lo
w
e
s
t
pe
r
c
e
nt
a
ge
e
r
r
or
w
it
h
th
r
e
e
va
r
ia
bl
e
s
1
=
−
3
,
2
=
−
2
,
3
=
−
1
,
a
nd
in
F
ig
ur
e
7,
th
e
or
a
nge
a
nd
bl
ue
gr
a
phs
s
how
th
a
t
th
e
pr
e
di
c
ti
on r
e
s
ul
ts
a
r
e
a
lm
os
t
a
c
c
ur
a
te
.
T
hi
s
r
e
s
e
a
r
c
h
pr
e
di
c
ts
th
e
Z
S
I
a
nd
C
Z
I
dr
ought
in
di
c
e
s
us
in
g
th
e
W
M
A
a
nd
M
L
P
m
e
th
ods
.
B
a
s
e
d
on
th
e
r
e
s
ul
ts
of
th
e
M
A
P
E
e
va
lu
a
ti
on,
th
e
m
e
th
od
th
a
t
is
c
lo
s
e
s
t
to
th
e
a
c
c
ur
a
c
y
in
pr
e
di
c
ti
ng
th
e
Z
S
I
a
nd
C
Z
I
i
ndi
c
e
s
i
s
M
L
P
w
it
h a
s
c
e
n
a
r
io
of
t
hr
e
e
i
nput
va
r
ia
bl
e
s
, a
nd t
he
M
L
P
a
r
c
hi
te
c
tu
r
e
i
s
3
-
6
-
10
-
1. T
he
W
M
A
a
nd
M
L
P
m
e
th
ods
ha
ve
th
e
s
a
m
e
in
put
,
na
m
e
ly
th
e
Z
S
I
a
nd
C
Z
I
in
di
c
e
s
f
r
om
th
e
pr
e
vi
ous
pe
r
io
d
da
ta
−
3
,
−
2
,
−
1
,
th
e
di
f
f
e
r
e
n
c
e
i
s
th
a
t
th
e
W
M
A
m
e
th
od
onl
y
c
a
lc
ul
a
te
s
th
e
a
v
e
r
a
ge
of
th
e
a
c
c
um
ul
a
ti
on
of
th
e
m
ul
ti
pl
ic
a
ti
on
of
th
e
dr
ought
in
de
x
a
nd
w
e
ig
ht
.
T
he
M
L
P
m
e
th
od
a
c
c
um
ul
a
te
s
th
e
m
ul
ti
pl
ic
a
ti
on
of
e
a
c
h
va
r
ia
bl
e
a
nd
w
e
ig
ht
,
th
e
n
th
e
r
e
is
a
w
e
ig
ht
im
pr
ove
m
e
nt
pr
oc
e
s
s
to
ge
t
a
m
or
e
pr
e
c
is
e
v
a
lu
e
in
th
e
pr
e
di
c
ti
on. T
he
pr
oc
e
s
s
i
s
c
a
r
r
ie
d out r
e
pe
a
te
dl
y. S
o, t
he
M
L
P
m
e
th
od ha
s
a
s
m
a
ll
e
r
pr
e
di
c
ti
on e
r
r
or
.
R
e
s
e
a
r
c
h
[
2]
pr
e
di
c
te
d
S
P
E
I
,
C
Z
I
,
S
P
I
,
Z
S
I
,
D
I
,
P
N
I
,
a
nd
R
A
I
us
in
g
a
r
ti
f
ic
ia
l
ne
ur
a
l
ne
twor
k
(
A
N
N
)
,
L
S
T
M
,
S
V
M
,
r
a
ndom
f
or
e
s
t,
a
nd
k
-
ne
a
r
e
s
t
n
e
ig
hbor
s
(
k
-
N
N
)
m
e
th
ods
;
th
e
be
s
t
m
ode
l
in
pr
e
di
c
ti
ng
w
a
s
li
ne
a
r
ke
r
ne
l
S
V
M
.
P
r
e
di
c
ti
on
S
P
I
,
Z
S
I
w
it
h
ge
ne
ti
c
pr
og
r
a
m
m
in
g
(
G
P
)
m
ode
ls
s
how
th
a
t
th
e
m
ode
l
is
a
bl
e
t
o pr
e
di
c
t
dr
ought we
ll
[
7]
. T
he
l
in
e
a
r
r
e
gr
e
s
s
io
n m
e
th
od i
s
a
bl
e
t
o pr
e
di
c
t
S
P
I
, Z
S
I
, R
A
I
, S
P
E
I
, a
nd R
D
I
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
dv A
ppl
S
c
i
I
S
S
N
:
2252
-
8814
P
r
e
di
c
ti
on i
nde
x
dr
ought us
e
ne
ur
al
ne
t
w
or
k
ba
s
e
d r
ai
nf
al
l
(
N
ur
N
af
ii
y
ah)
1151
w
e
ll
[
12]
.
T
r
a
di
ti
ona
l
s
ta
ti
s
ti
c
a
l
m
ode
ls
s
uc
h
a
s
A
R
I
M
A
ha
ve
a
ls
o
be
e
n
w
id
e
ly
a
ppl
ie
d
f
or
dr
ought
f
or
e
c
a
s
ti
ng,
pa
r
ti
c
ul
a
r
ly
f
o
r
S
P
I
,
due
to
th
e
ir
a
bi
li
ty
to
c
a
pt
ur
e
te
m
por
a
l
de
pe
nde
nc
ie
s
.
H
ow
e
ve
r
,
A
R
I
M
A
is
li
m
it
e
d
in
ha
ndl
in
g
nonl
in
e
a
r
pa
tt
e
r
ns
c
om
m
onl
y
pr
e
s
e
nt
in
c
li
m
a
te
a
nd
r
a
in
f
a
ll
da
ta
.
S
im
il
a
r
ly
,
D
T
m
ode
ls
pr
ovi
de
in
te
r
pr
e
ta
bl
e
r
e
s
ul
ts
a
nd
c
a
n
c
a
pt
ur
e
s
im
pl
e
nonl
in
e
a
r
r
e
la
ti
ons
hi
ps
,
but
th
e
ir
pr
e
di
c
ti
on
a
c
c
ur
a
c
y
is
of
te
n l
ow
e
r
t
ha
n e
ns
e
m
bl
e
m
e
th
ods
s
uc
h a
s
r
a
ndom f
or
e
s
t
or
gr
a
di
e
nt
boos
ti
ng. C
om
pa
r
e
d t
o t
he
s
e
t
r
a
di
ti
ona
l
a
ppr
oa
c
he
s
,
th
e
M
L
P
m
ode
l
in
th
is
r
e
s
e
a
r
c
h
a
c
hi
e
ve
d
lo
w
e
r
e
r
r
or
va
lu
e
s
(
M
A
P
E
=
4.177%
)
,
in
di
c
a
ti
ng
th
a
t
ne
ur
a
l
ne
twor
k
–
ba
s
e
d
m
e
th
ods
a
r
e
m
or
e
e
f
f
e
c
ti
ve
in
c
a
pt
ur
in
g
th
e
nonl
in
e
a
r
c
ha
r
a
c
te
r
is
ti
c
s
of
dr
ought
in
di
c
e
s
s
uc
h
a
s
Z
S
I
a
nd C
Z
I
.
D
e
s
p
it
e
t
he
pr
om
i
s
i
ng
r
e
s
u
lt
s
,
t
h
is
r
e
s
e
a
r
c
h
ha
s
s
e
v
e
r
a
l
li
m
it
a
ti
o
n
s
.
F
ir
s
t
, t
he
r
a
in
f
a
ll
d
a
t
a
s
e
t
u
s
e
d
i
n
t
h
i
s
r
e
s
e
a
r
c
h
i
s
r
e
l
a
t
iv
e
ly
li
m
it
e
d
i
n
te
r
m
s
of
t
e
m
por
a
l
c
ov
e
r
a
g
e
a
nd
s
p
a
t
ia
l
r
e
s
ol
ut
i
on,
w
hi
c
h
m
a
y
a
f
f
e
c
t
t
he
r
obu
s
t
ne
s
s
of
t
he
m
od
e
l.
S
e
c
on
d,
th
e
u
s
e
o
f
ne
ur
a
l
ne
two
r
k
s
s
uc
h a
s
M
L
P
c
a
r
r
i
e
s
a
n
i
nh
e
r
e
nt
r
i
s
k
o
f
ov
e
r
f
it
ti
n
g,
e
s
p
e
c
i
a
l
ly
w
h
e
n
tr
a
in
i
ng
w
it
h
s
p
a
r
s
e
da
ta
.
A
lt
h
oug
h
m
e
a
s
ur
e
s
s
uc
h
a
s
i
np
ut
s
c
e
na
r
io
t
e
s
ti
ng
w
e
r
e
a
p
pl
i
e
d
,
t
he
pot
e
nt
ia
l
r
i
s
k
c
a
nn
ot
be
f
ul
l
y
e
li
m
i
n
a
te
d.
T
hi
r
d
,
t
he
de
ve
lo
p
e
d
m
od
e
l
s
w
e
r
e
tr
a
i
ne
d
a
nd
v
a
l
id
a
t
e
d
onl
y
f
or
a
s
p
e
c
if
i
c
r
e
gi
on
a
nd
f
or
t
w
o
r
a
in
f
a
l
l
-
b
a
s
e
d
in
di
c
e
s
(
Z
S
I
a
n
d
C
Z
I
)
.
T
he
r
e
f
or
e
,
th
e
g
e
n
e
r
a
li
z
a
ti
on
of
th
e
r
e
s
ul
t
s
t
o
ot
h
e
r
r
e
g
io
n
s
,
c
li
m
a
te
c
ond
it
i
on
s
,
or
dr
o
ug
ht
in
d
ic
e
s
m
a
y r
e
q
ui
r
e
r
e
tr
a
in
in
g or
f
ur
th
e
r
a
d
a
p
ta
t
io
n.
(
a
)
(
b)
(
c
)
(
d)
F
ig
ur
e
6
.
P
r
e
di
c
ti
on W
M
A
of
(
a
)
Z
S
I
W
M
A
2 t
im
e
s
,
(
b)
Z
S
I
W
M
A
3 t
im
e
s
,
(
c
)
C
Z
I
W
M
A
2 t
im
e
s
,
a
nd
(
d)
C
Z
I
W
M
A
3 t
im
e
s
T
a
bl
e
3
.
W
M
A
pr
e
di
c
ti
on s
ta
ti
s
ti
c
s
V
a
l
ue
Z
S
I
C
Z
I
2 t
i
m
e
s
3 t
i
m
e
s
2 t
i
m
e
s
3 t
i
m
e
s
M
i
ni
m
um
-
2.332
-
4.794
-
2.674
-
4.758
M
a
xi
m
um
3.415
7.116
2.934
5.797
S
t
a
nda
r
d
d
e
vi
a
t
i
on
1.458
2.928
1.460
2.663
M
e
di
a
n
0.077
0.207
-
0.058
-
0.334
M
e
a
n
0.003
0.007
-
0.204
-
0.370
T
a
bl
e
4
.
M
A
P
E
W
M
A
pr
e
di
c
ti
on (
%
)
I
nde
x
2 t
i
m
e
s
3 t
i
m
e
s
Z
S
I
609.37
1475.39
C
Z
I
347.48
348.43
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8814
I
nt
J
A
dv A
ppl
S
c
i
,
V
ol
.
14
, N
o.
4
,
D
e
c
e
m
be
r
20
25
:
1146
-
1154
1152
T
a
bl
e
5.
M
A
P
E
s
ta
ti
s
ti
c
s
of
M
L
P
(
%
)
V
a
l
ue
Z
S
I
C
Z
I
t
w
o va
r
i
a
bl
e
s
t
hr
e
e
va
r
i
a
bl
e
s
t
w
o va
r
i
a
bl
e
s
t
hr
e
e
va
r
i
a
bl
e
s
M
i
ni
m
um
176.375
4.177
128.984
10.030
M
a
xi
m
um
326.313
17.381
644.352
56.035
S
t
a
nda
r
d
de
vi
a
t
i
on
283.499
8.643
213.315
17.722
M
e
di
a
n
263.734
10.553
243.341
21.350
M
e
a
n
78.682
6.733
132.256
12.412
(
a
)
(
b)
(
c
)
(
d)
F
ig
ur
e
7
.
M
L
P
pr
e
di
c
t
io
n
of
(
a
)
Z
S
I
2 va
r
ia
bl
e
s
,
(
b)
Z
S
I
3
va
r
ia
bl
e
s
,
(
c
)
C
Z
I
2 va
r
ia
bl
e
s
,
a
nd
(
d)
C
Z
I
3 va
r
ia
bl
e
s
4.
C
O
N
C
L
U
S
I
O
N
T
hi
s
r
e
s
e
a
r
c
h
in
ve
s
ti
ga
te
d
th
e
pr
e
di
c
ti
on
of
Z
S
I
a
nd
C
Z
I
dr
ought
in
di
c
e
s
u
s
in
g
th
e
W
M
A
a
nd
M
L
P
m
e
th
ods
unde
r
two
in
put
s
c
e
na
r
io
s
:
two
pr
e
vi
ou
s
pe
r
io
ds
(
−
2
,
−
1
)
a
nd
th
r
e
e
pr
e
vi
ous
pe
r
io
ds
(
−
3
,
−
2
,
−
1
)
.
T
he
r
e
s
ul
ts
s
how
th
a
t
M
L
P
c
ons
i
s
te
nt
ly
out
pe
r
f
or
m
s
W
M
A
in
pr
e
di
c
ti
on
a
c
c
ur
a
c
y.
T
he
be
s
t
pe
r
f
or
m
a
nc
e
w
a
s
a
c
hi
e
ve
d
by
th
e
M
L
P
m
ode
l
w
it
h
a
M
A
P
E
of
4.177%
us
in
g
th
e
th
r
e
e
-
va
r
ia
bl
e
in
put
s
c
e
na
r
io
. T
he
opt
im
a
l
M
L
P
a
r
c
hi
te
c
tu
r
e
obt
a
in
e
d i
s
3
-
6
-
10
-
1. T
he
s
e
f
in
di
ngs
de
m
ons
tr
a
te
t
ha
t
M
L
P
c
a
n
s
e
r
ve
a
s
a
r
e
li
a
bl
e
m
ode
l
f
or
pr
e
di
c
ti
ng
r
a
in
f
a
ll
-
ba
s
e
d
dr
ought
in
di
c
e
s
be
yond
S
P
I
,
pa
r
ti
c
ul
a
r
ly
Z
S
I
a
nd
C
Z
I
,
w
hi
c
h
ha
ve
be
e
n
r
a
r
e
ly
e
xa
m
in
e
d
in
pr
e
vi
ous
r
e
s
e
a
r
c
h.
F
or
f
ut
ur
e
r
e
s
e
a
r
c
h,
s
e
ve
r
a
l
pot
e
nt
ia
l
e
xt
e
ns
io
ns
c
a
n
be
c
ons
id
e
r
e
d
to
e
nh
a
nc
e
th
e
a
ppl
ic
a
bi
li
ty
of
dr
ought
pr
e
di
c
t
io
n
m
ode
ls
.
F
ir
s
t,
m
ul
ti
m
oda
l
in
put
s
s
u
c
h
a
s
c
om
bi
ni
ng
r
a
in
f
a
ll
w
it
h
te
m
pe
r
a
tu
r
e
,
hum
id
it
y,
or
s
oi
l
m
o
is
tu
r
e
da
ta
c
oul
d
im
pr
ove
pr
e
di
c
ti
on
r
obus
tn
e
s
s
.
S
e
c
ond,
r
e
a
l
-
ti
m
e
or
ne
a
r
-
r
e
a
l
-
ti
m
e
dr
ought
pr
e
di
c
ti
on
s
ys
te
m
s
c
oul
d
be
de
ve
lo
pe
d
to
s
uppor
t
e
a
r
ly
w
a
r
ni
ng
a
nd r
a
pi
d r
e
s
pons
e
.
F
U
N
D
I
N
G
I
N
F
O
R
M
A
T
I
O
N
T
ha
nk
s
to
U
ni
ve
r
s
it
a
s
I
s
la
m
L
a
m
onga
n
f
or
f
undi
ng
th
e
r
e
s
e
a
r
c
h.
A
U
T
H
O
R
C
O
N
T
R
I
B
U
T
I
O
N
S
S
T
A
T
E
M
E
N
T
T
hi
s
jo
ur
na
l
us
e
s
th
e
C
ont
r
ib
ut
or
R
ol
e
s
T
a
xonomy
(
C
R
e
di
T
)
to
r
e
c
ogni
z
e
in
di
vi
dua
l
a
ut
hor
c
ont
r
ib
ut
io
ns
, r
e
duc
e
a
ut
hor
s
hi
p di
s
put
e
s
,
a
nd f
a
c
il
it
a
te
c
ol
la
bo
r
a
ti
on.
N
am
e
o
f
A
u
t
h
or
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
N
ur
N
a
f
ii
ya
h
✓
✓
✓
✓
✓
✓
✓
✓
A
li
M
okht
a
r
✓
✓
✓
✓
✓
✓
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
dv A
ppl
S
c
i
I
S
S
N
:
2252
-
8814
P
r
e
di
c
ti
on i
nde
x
dr
ought us
e
ne
ur
al
ne
t
w
or
k
ba
s
e
d r
ai
nf
al
l
(
N
ur
N
af
ii
y
ah)
1153
C
:
C
onc
e
pt
ua
l
i
z
a
t
i
on
M
:
M
e
t
hodol
ogy
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
da
t
i
on
Fo
:
Fo
r
m
a
l
a
na
l
ys
i
s
I
:
I
nve
s
t
i
ga
t
i
on
R
:
R
e
s
our
c
e
s
D
:
D
a
t
a
C
ur
a
t
i
on
O
:
W
r
i
t
i
ng
-
O
r
i
gi
na
l
D
r
a
f
t
E
:
W
r
i
t
i
ng
-
R
e
vi
e
w
&
E
di
t
i
ng
Vi
:
Vi
s
ua
l
i
z
a
t
i
on
Su
:
Su
pe
r
vi
s
i
on
P
:
P
r
oj
e
c
t
a
dm
i
ni
s
t
r
a
t
i
on
Fu
:
Fu
ndi
ng a
c
qui
s
i
t
i
on
C
O
N
F
L
I
C
T
O
F
I
N
T
E
R
E
S
T
S
T
A
T
E
M
E
N
T
A
ll
r
e
s
e
a
r
c
he
r
s
do not ha
ve
a
c
onf
li
c
t.
D
A
T
A
A
V
A
I
L
A
B
I
L
I
T
Y
D
a
ta
w
a
s
do
w
nl
oa
de
d
f
r
om
th
e
w
e
bs
it
e
H
id
r
ol
ogi
S
D
A
J
a
w
a
T
im
ur
a
t
ht
tp
s
:/
/h
id
r
ol
ogi
.dpua
ir
.j
a
ti
m
p
r
ov.go.id/
pe
la
ya
na
n/
.
R
E
F
E
R
E
N
C
E
S
[
1]
N
.
N
a
ndgude
,
T
.
P
.
S
i
ngh,
S
.
N
a
ndgude
,
a
nd
M
.
T
i
w
a
r
i
,
“
D
r
ought
pr
e
di
c
t
i
on:
a
c
om
pr
e
he
ns
i
ve
r
e
vi
e
w
of
di
f
f
e
r
e
nt
dr
ought
pr
e
di
c
t
i
on m
ode
l
s
a
nd a
dopt
e
d t
e
c
hnol
ogi
e
s
,
”
Sus
t
ai
nabi
l
i
t
y
, vol
. 15, no. 5, 202
3, doi
:
10.3390/
s
u151511684.
[
2]
V
.
K
a
r
t
a
l
,
O
.
M
.
K
a
t
i
poğl
u,
E
.
K
a
r
a
koyun,
O
.
S
i
m
s
e
k,
V
.
S
.
Y
a
vuz
,
a
nd
S
.
A
r
i
m
a
n,
“
P
r
e
di
c
t
i
on
of
g
r
oundw
a
t
e
r
dr
ought
ba
s
e
d
on
hydr
o
-
m
e
t
e
or
ol
ogi
c
a
l
i
ns
i
ght
s
vi
a
m
a
c
hi
ne
l
e
a
r
ni
ng
a
ppr
oa
c
he
s
,”
P
hy
s
i
c
s
and
C
he
m
i
s
t
r
y
of
t
he
E
ar
t
h
,
vol
.
136,
no.
S
e
pt
e
m
be
r
,
2024, doi
:
10.1016/
j
.pc
e
.2024.103757.
[
3]
P
.
M
a
hm
oudi
,
A
.
R
i
gi
,
a
nd
M
.
M
.
K
a
m
a
k,
“
A
c
om
pa
r
a
t
i
ve
s
t
udy
of
pr
e
c
i
pi
t
a
t
i
on
-
ba
s
e
d
dr
ought
i
ndi
c
e
s
w
i
t
h
t
he
a
i
m
of
s
e
l
e
c
t
i
ng
t
he
be
s
t
i
nde
x
f
or
dr
ought
m
oni
t
or
i
ng
i
n
I
r
a
n,”
T
he
or
e
t
i
c
al
A
ppl
i
e
d
C
l
i
m
at
ol
og
y
,
vol
.
137,
no.
3,
2019,
doi
:
10.1007/
s
00704
-
019
-
02778
-
z.
[
4]
Y
. G
uo
e
t
al
.
, “
A
s
s
e
s
s
i
ng s
oc
i
oe
c
onom
i
c
dr
ought
ba
s
e
d on a
n i
m
pr
ove
d m
ul
t
i
va
r
i
a
t
e
s
t
a
nda
r
di
z
e
d r
e
l
i
a
bi
l
i
t
y a
nd r
e
s
i
l
i
e
nc
e
i
nde
x,
”
J
our
nal
of
H
y
dr
ol
ogy
, vol
. 568, 2019, doi
:
10.1016/
j
.j
hydr
ol
.2018.11.055.
[
5]
A
.
H
.
P
a
ya
b
a
nd
U
.
T
ü
r
ke
r
,
“
C
om
pa
r
i
s
on
of
s
t
a
nda
r
di
z
e
d
m
e
t
e
or
ol
ogi
c
a
l
i
ndi
c
e
s
f
or
dr
ought
m
oni
t
or
i
ng
a
t
t
he
nor
t
he
r
n
pa
r
t
o
f
C
ypr
us
,”
E
nv
i
r
onm
e
nt
al
E
ar
t
h Sc
i
e
nc
e
s
, vol
. 78, no. 10, 2019, doi
:
10.1007/
s
1
2665
-
019
-
8309
-
x.
[
6]
J
.
Z
hou,
X
.
C
he
n,
C
.
X
u,
a
nd
P
.
W
u,
“
A
s
s
e
s
s
i
ng
s
oc
i
oe
c
onom
i
c
dr
ought
ba
s
e
d
on
a
s
t
a
nda
r
di
z
e
d
s
uppl
y
a
nd
de
m
a
nd
w
a
t
e
r
i
nde
x,”
W
at
e
r
R
e
s
our
c
e
s
M
anage
m
e
nt
, vol
. 36, no. 6, 2022, doi
:
10.1007/
s
1126
9
-
022
-
03117
-
0.
[
7]
E
.
O
m
i
dva
r
a
nd
Z
.
N
.
T
a
hr
ood
i
,
“
E
va
l
ua
t
i
on
a
nd
p
r
e
di
c
t
i
on
of
m
e
t
e
or
o
l
ogi
c
a
l
dr
ought
c
ondi
t
i
ons
us
i
ng
t
i
m
e
-
s
e
r
i
e
s
a
nd
ge
ne
t
i
c
pr
ogr
a
m
m
i
ng m
ode
l
s
,”
J
our
nal
of
E
ar
t
h Sy
s
t
e
m
Sc
i
e
n
c
e
, vol
. 128, no. 3, 2019,
doi
:
10.1007/
s
12040
-
019
-
1103
-
z.
[
8]
M
.
M
.
M
oghi
m
i
a
nd
A
.
R
.
Z
a
r
e
i
,
“
E
va
l
ua
t
i
ng
pe
r
f
or
m
a
nc
e
a
nd
a
ppl
i
c
a
bi
l
i
t
y
of
s
e
ve
r
a
l
dr
ought
i
ndi
c
e
s
i
n
a
r
i
d
r
e
gi
ons
,”
A
s
i
a
P
ac
i
f
i
c
J
our
nal
of
A
t
m
os
phe
r
i
c
Sc
i
e
nc
e
s
, vol
. 57, no. 3, 2021, doi
:
10.1007/
s
13
143
-
019
-
00122
-
z.
[
9]
P
. M
a
hm
oudi
, A
. R
i
gi
, a
nd M
.
M
. K
a
m
a
k,
“
E
va
l
ua
t
i
ng t
he
s
e
n
s
i
t
i
vi
t
y of
pr
e
c
i
pi
t
a
t
i
on
-
ba
s
e
d dr
ought
i
ndi
c
e
s
t
o di
f
f
e
r
e
nt
l
e
ngt
hs
of
r
e
c
or
d,”
J
our
nal
of
H
y
dr
ol
ogy
, vol
. 579, 2019, doi
:
10.1016/
j
.j
hydr
ol
.2019.124181.
[
10]
P
.
B
huni
a
,
P
.
D
a
s
,
a
nd
R
.
M
a
i
t
i
,
“
M
e
t
e
or
ol
ogi
c
a
l
dr
ought
s
t
udy
t
hr
ough
S
P
I
i
n
t
hr
e
e
dr
ought
pr
one
di
s
t
r
i
c
t
s
of
W
e
s
t
B
e
nga
l
,
I
ndi
a
,”
E
ar
t
h Sy
s
t
e
m
s
and E
nv
i
r
onm
e
nt
, vol
. 4, no. 1, 2020, doi
:
10.1007/
s
41748
-
019
-
00137
-
6.
[
11]
B
.
S
.
S
obr
a
l
e
t
al
.
,
“
D
r
ought
c
ha
r
a
c
t
e
r
i
z
a
t
i
on f
or
t
he
s
t
a
t
e
of
R
i
o
de
J
a
ne
i
r
o
ba
s
e
d
on
t
he
a
nnua
l
S
P
I
i
nde
x:
t
r
e
nds
,
s
t
a
t
i
s
t
i
c
a
l
t
e
s
t
s
a
nd i
t
s
r
e
l
a
t
i
on w
i
t
h E
N
S
O
,”
A
t
m
os
phe
r
i
c
R
e
s
e
ar
c
h
, vol
. 220, 2019, doi
:
10.1016/
j
.a
t
m
os
r
e
s
.2019.01.003.
[
12]
O
.
M
.
K
a
t
i
pogl
u,
R
.
A
c
a
r
,
a
nd
S
.
Ş
e
ngül
,
“
C
om
pa
r
i
s
on
of
m
e
t
e
or
ol
ogi
c
a
l
i
ndi
c
e
s
f
or
dr
ought
m
oni
t
or
i
ng
a
nd
e
va
l
ua
t
i
ng:
a
c
a
s
e
s
t
udy f
r
om
E
uphr
a
t
e
s
B
a
s
i
n, T
ur
ke
y,”
J
our
nal
of
W
at
e
r
and C
l
i
m
at
e
C
hang
e
, v
ol
. 11, no. 1S
, 2020, doi
:
10.2166/
w
c
c
.2020.171.
[
13]
K
.
D
i
a
ni
,
I
.
K
a
c
i
m
i
,
M
.
Z
e
m
z
a
m
i
,
H
.
T
a
bya
oui
,
a
nd
A
.
T
.
H
a
ghi
ghi
,
“
E
va
l
ua
t
i
on
of
m
e
t
e
or
ol
ogi
c
a
l
dr
ought
us
i
ng
t
he
s
t
a
nda
r
di
z
e
d
pr
e
c
i
pi
t
a
t
i
on
i
nde
x
(
S
P
I
)
i
n
t
he
H
i
gh
Z
i
z
R
i
ve
r
ba
s
i
n,
M
or
oc
c
o,”
L
i
m
nol
ogi
c
al
R
e
v
i
e
w
,
vol
.
19,
no.
3,
2019
,
doi
:
10.2478/
l
i
m
r
e
-
2019
-
0011.
[
14]
V
.
L
.
S
i
va
kum
a
r
e
t
al
.
,
“
A
n
i
nt
e
gr
a
t
i
on
of
ge
os
pa
t
i
a
l
t
e
c
hnol
ogy
a
nd
s
t
a
nda
r
d
pr
e
c
i
pi
t
a
t
i
on
i
nde
x
(
S
P
I
)
f
or
d
r
ought
vul
ne
r
a
bi
l
i
t
y
a
s
s
e
s
s
m
e
nt
f
or
a
pa
r
t
of
N
a
m
a
kka
l
di
s
t
r
i
c
t
,
S
out
h
I
ndi
a
,”
i
n
M
at
e
r
i
al
s
T
oday
:
P
r
oc
e
e
di
ngs
,
2020.
doi
:
10.1016/
j
.m
a
t
pr
.2020.08.157.
[
15]
Y
.
K
hour
di
f
i
a
nd
M
.
B
a
ha
j
,
“
H
e
a
r
t
di
s
e
a
s
e
pr
e
di
c
t
i
on
a
nd
c
l
a
s
s
i
f
i
c
a
t
i
on
us
i
ng
m
a
c
hi
ne
l
e
a
r
ni
ng
a
l
gor
i
t
hm
s
opt
i
m
i
z
e
d
by
pa
r
t
i
c
l
e
s
w
a
r
m
opt
i
m
i
z
a
t
i
on
a
nd
a
nt
c
ol
ony
opt
i
m
i
z
a
t
i
on,”
I
nt
e
r
nat
i
onal
J
our
nal
of
I
nt
e
l
l
i
ge
nt
E
ngi
ne
e
r
i
ng
and
Sy
s
t
e
m
s
,
vol
.
12,
no.
1
,
2019, doi
:
10.22266/
i
j
i
e
s
2019.0228.24.
[
16]
Y
a
dda
r
a
bul
l
a
h
e
t
al
.
,
“
O
pt
i
m
i
z
e
d
pr
e
di
c
t
i
on
of
a
i
r
f
l
ow
vol
um
e
i
n
unde
r
-
a
c
t
u
a
t
e
d
z
one
s
t
hr
ough
m
ul
t
i
l
a
ye
r
pe
r
c
e
pt
r
on
a
r
t
i
f
i
c
i
a
l
ne
ur
a
l
ne
t
w
or
k,”
I
nt
e
r
nat
i
onal
J
our
nal
of
I
nt
e
l
l
i
ge
nt
E
ng
i
ne
e
r
i
ng
and
Sy
s
t
e
m
s
,
vol
.
18,
no.
1,
pp.
391
–
408,
2025,
doi
:
10.22266/
i
j
i
e
s
2025.0229.29.
[
17]
S
.
H
.
M
uhi
,
H
.
N
.
A
bdul
l
a
h,
a
nd
B
.
H
.
A
bd,
“
M
od
e
l
i
ng
f
or
pr
e
di
c
t
i
ng
t
he
s
e
ve
r
i
t
y
of
he
pa
t
i
t
i
s
ba
s
e
d
on
a
r
t
i
f
i
c
i
a
l
ne
ur
a
l
ne
t
w
or
ks
,”
I
nt
e
r
nat
i
onal
J
our
nal
of
I
nt
e
l
l
i
ge
nt
E
ngi
ne
e
r
i
ng and Sy
s
t
e
m
s
, vol
. 1
3, no. 3, 2020, doi
:
10.22266/
I
J
I
E
S
2020.0630.15.
[
18]
S
.
S
r
i
dha
r
a
,
G
.
M
.
C
ha
i
t
hr
a
,
a
nd P
.
G
opa
kka
l
i
,
“
A
s
s
e
s
s
m
e
nt
a
nd
m
oni
t
or
i
ng
of
dr
ought
i
n
C
hi
t
r
a
dur
ga
di
s
t
r
i
c
t
of
K
a
r
na
t
a
ka
us
i
ng
di
f
f
e
r
e
nt
dr
ought
i
nd
i
c
e
s
,”
J
our
nal
of
A
gr
om
e
t
e
o
r
ol
ogy
, vol
. 23, no. 2, 2021, d
oi
:
10.54386/
j
a
m
.v23i
2.72.
[
19]
A
.
E
l
hous
s
a
oui
,
M
.
Z
a
a
ga
ne
,
a
nd
L
.
B
e
na
a
bi
d
a
t
e
,
“
C
om
p
a
r
i
s
on
of
va
r
i
ous
d
r
ought
i
ndi
c
e
s
f
or
a
s
s
e
s
s
i
ng
dr
ought
s
t
a
t
us
of
t
h
e
N
or
t
he
r
n M
e
ke
r
r
a
w
a
t
e
r
s
he
d, N
or
t
hw
e
s
t
of
A
l
ge
r
i
a
,”
A
r
abi
an J
ou
r
nal
of
G
e
os
c
i
e
nc
e
s
, vol
. 14, no. 10, 2021, doi
:
10.1007/
s
12517
-
021
-
07269
-
y.
[
20]
Ö
.
C
oş
kun
a
nd
H
.
C
i
t
a
kogl
u,
“
P
r
e
di
c
t
i
on
of
t
he
s
t
a
nda
r
di
z
e
d
pr
e
c
i
pi
t
a
t
i
on
i
nde
x
ba
s
e
d
on
t
he
l
ong
s
hor
t
-
t
e
r
m
m
e
m
or
y
a
nd
e
m
pi
r
i
c
a
l
m
ode
de
c
om
pos
i
t
i
on
-
e
xt
r
e
m
e
l
e
a
r
ni
ng
m
a
c
hi
ne
m
ode
l
s
:
t
he
c
a
s
e
of
S
a
ka
r
ya
,
T
ür
ki
ye
,”
P
hy
s
i
c
s
and
C
h
e
m
i
s
t
r
y
of
t
he
E
ar
t
h
, vol
. 131, 2023, doi
:
10.1016/
j
.pc
e
.2023.103418.
[
21]
H
.
N
,
A
.
M
,
a
nd
S
.
A
.
A
hm
e
d,
“
S
pa
t
i
o
-
t
e
m
por
a
l
c
ha
r
a
c
t
e
r
i
s
t
i
c
s
of
r
a
i
nf
a
l
l
a
nd
dr
ought
c
ondi
t
i
ons
a
r
e
us
i
ng
t
he
di
f
f
e
r
e
nt
dr
ought
i
ndi
c
e
s
w
i
t
h
g
e
os
pa
t
i
a
l
a
ppr
oa
c
he
s
i
n
K
a
r
na
t
a
ka
s
t
a
t
e
,”
J
our
nal
of
A
t
m
os
phe
r
i
c
and
Sol
ar
T
e
r
r
e
s
t
r
i
al
P
hy
s
i
c
s
,
vol
.
265,
no. N
ove
m
be
r
2023, p. 106372, 2024, doi
:
10.1016/
j
.j
a
s
t
p.2024.106372.
[
22]
Z
.
P
e
i
,
S
. F
a
ng,
L
.
W
a
ng,
a
nd
W
.
Y
a
ng,
“
C
om
pa
r
a
t
i
ve
a
na
l
ys
i
s
of
dr
ought
i
ndi
c
a
t
e
d
by
t
he
S
P
I
a
nd
S
P
E
I
a
t
va
r
i
ous
t
i
m
e
s
c
a
l
e
s
i
n
I
nne
r
M
ongol
i
a
, C
hi
na
,”
W
at
e
r
(
Sw
i
t
z
e
r
l
and)
, vol
. 12, no. 7, 2020, doi
:
10.3390/
w
12071925.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8814
I
nt
J
A
dv A
ppl
S
c
i
,
V
ol
.
14
, N
o.
4
,
D
e
c
e
m
be
r
20
25
:
1146
-
1154
1154
[
23]
S
.
D
e
hgha
n,
N
.
S
a
l
e
hni
a
,
N
.
S
a
ya
r
i
,
a
nd
B
.
B
a
kht
i
a
r
i
,
“
P
r
e
di
c
t
i
on
of
m
e
t
e
or
ol
ogi
c
a
l
dr
ought
i
n
a
r
i
d
a
nd
s
e
m
i
-
a
r
i
d
r
e
gi
ons
us
i
ng
P
D
S
I
a
nd
S
D
S
M
:
a
c
a
s
e
s
t
udy
i
n
F
a
r
s
P
r
ovi
nc
e
,
I
r
a
n,”
J
ou
r
nal
of
A
r
i
d
L
and
,
vol
.
12,
no.
2,
2020,
doi
:
10.1007/
s
40333
-
020
-
0095
-
5.
[
24]
A
. A
di
b a
nd S
. S
. M
a
r
a
s
hi
, “
M
e
t
e
or
ol
ogi
c
a
l
dr
ought
m
oni
t
or
i
ng a
nd pr
e
pa
r
a
t
i
on of
l
ong
-
t
e
r
m
a
nd s
hor
t
-
t
e
r
m
d
r
ought
z
oni
ng m
a
ps
us
i
ng
r
e
gi
ona
l
f
r
e
que
nc
y
a
na
l
ys
i
s
a
nd
L
-
m
om
e
nt
i
n
t
he
K
huz
e
s
t
a
n
pr
ovi
nc
e
of
I
r
a
n,”
T
he
or
e
t
i
c
al
A
ppl
i
e
d
C
l
i
m
at
ol
ogy
,
vol
.
137,
no. 1
--
2, 2019, doi
:
10.1007/
s
00704
-
018
-
2572
-
8.
[
25]
Y
.
D
i
ng,
G
.
Y
u,
R
.
T
i
a
n,
a
nd
Y
.
S
un,
“
A
ppl
i
c
a
t
i
on
of
a
hybr
i
d
C
E
E
M
D
-
L
S
T
M
m
ode
l
ba
s
e
d
on
t
he
s
t
a
nda
r
di
z
e
d
pr
e
c
i
pi
t
a
t
i
on
i
nde
x
f
or
dr
ought
f
or
e
c
a
s
t
i
ng:
t
he
c
a
s
e
of
t
he
X
i
nj
i
a
ng
U
ygu
r
A
ut
onom
ous
R
e
gi
on,
C
hi
na
,”
A
t
m
os
phe
r
e
(
B
as
e
l
)
,
vol
.
13,
no.
9,
2022, doi
:
10.3390/
a
t
m
os
13091504.
[
26]
H
.
N
,
S
.
A
.
A
hm
e
d,
S
.
K
um
a
r
,
a
nd
A
.
M
,
“
C
om
put
a
t
i
on
of
t
he
s
pa
t
i
o
-
t
e
m
por
a
l
e
xt
e
nt
of
r
a
i
nf
a
l
l
a
nd
l
ong
-
t
e
r
m
m
e
t
e
or
ol
ogi
c
a
l
dr
ought
a
s
s
e
s
s
m
e
nt
us
i
ng
s
t
a
nda
r
di
z
e
d
pr
e
c
i
pi
t
a
t
i
on
i
nde
x
ov
e
r
K
ol
a
r
a
nd
C
hi
kka
ba
l
l
a
pur
a
di
s
t
r
i
c
t
s
,
K
a
r
na
t
a
ka
dur
i
ng
1951
–
2019,”
R
e
m
ot
e
Se
ns
i
ng A
ppl
i
c
at
i
ons
:
Soc
i
e
t
y
and E
nv
i
r
onm
e
nt
, vol
. 27, 2022,
doi
:
10.1016/
j
.r
s
a
s
e
.2022.100768.
[
27]
U
.
W
.
H
um
phr
i
e
s
,
M
.
W
a
q
a
s
,
P
.
T
.
H
l
i
a
ng,
P
.
D
e
c
hpi
c
ha
i
,
a
nd
A
.
W
a
ngw
ongc
ha
i
,
“
A
d
e
e
p
l
e
a
r
ni
ng
pe
r
s
p
e
c
t
i
ve
o
n
m
e
t
e
or
ol
ogi
c
a
l
dr
ought
s
pr
e
di
c
t
i
on
i
n
t
he
M
un
R
i
ve
r
B
a
s
i
n,
T
ha
i
l
a
nd,”
A
I
P
A
dv
anc
e
s
,
vol
.
14,
no.
8,
pp.
1
–
26,
2024,
doi
:
10.1063/
5.0209709.
[
28]
A
.
D
i
ks
hi
t
,
B
.
P
r
a
dha
n,
a
nd
A
.
M
.
A
l
a
m
r
i
,
“
L
ong
l
e
a
d
t
i
m
e
dr
ought
f
or
e
c
a
s
t
i
ng
us
i
ng
l
a
gge
d
c
l
i
m
a
t
e
va
r
i
a
bl
e
s
a
nd
a
s
t
a
c
ke
d
l
o
ng
s
hor
t
-
t
e
r
m
m
e
m
or
y m
ode
l
,”
Sc
i
e
nc
e
of
t
he
T
ot
al
E
nv
i
r
onm
e
nt
, vol
. 755, 2021, doi
:
10.1016/
j
.s
c
i
t
ot
e
nv.2020.142638.
[
29]
S
.
A
z
i
m
i
a
nd
M
.
A
.
M
ogha
dda
m
,
“
M
od
e
l
i
ng
s
hor
t
t
e
r
m
r
a
i
nf
a
l
l
f
or
e
c
a
s
t
us
i
ng
ne
ur
a
l
n
e
t
w
or
ks
,
a
nd
G
a
us
s
i
a
n
pr
oc
e
s
s
c
l
a
s
s
i
f
i
c
a
t
i
on
ba
s
e
d
on
t
he
S
P
I
dr
ought
i
nde
x,”
W
at
e
r
R
e
s
our
c
e
s
M
anage
m
e
nt
,
vol
.
34,
no.
4,
2020,
doi
:
10.1007/
s
11269
-
020
-
02507
-
6.
[
30]
M
.
S
.
O
youna
l
s
oud,
M
.
A
bda
l
l
a
h,
A
.
G
.
Y
i
l
m
a
z
,
M
.
S
i
ddi
que
,
a
nd
S
.
A
t
a
b
a
y,
“
A
ne
w
m
e
t
e
or
ol
ogi
c
a
l
dr
ought
i
nde
x
ba
s
e
d
on
f
uz
z
y
l
ogi
c
:
de
ve
l
opm
e
nt
a
nd
c
om
pa
r
a
t
i
ve
a
s
s
e
s
s
m
e
nt
w
i
t
h
c
onv
e
nt
i
ona
l
dr
ought
i
ndi
c
e
s
,”
J
ou
r
nal
of
H
y
dr
ol
ogy
,
vol
.
619,
2023
,
doi
:
10.1016/
j
.j
hydr
ol
.2023.129306.
[
31]
A
. G
hoz
a
t
,
A
. S
ha
r
a
f
a
t
i
, S
.
B
. H
.
S
. A
s
a
dol
l
a
h, a
nd
D
.
M
ot
t
a
,
“
A
nove
l
i
nt
e
l
l
i
ge
nt
a
ppr
oa
c
h f
or
pr
e
di
c
t
i
ng m
e
t
e
or
ol
ogi
c
a
l
dr
o
ught
ba
s
e
d
on
s
a
t
e
l
l
i
t
e
-
ba
s
e
d
pr
e
c
i
pi
t
a
t
i
on
pr
oduc
t
:
a
ppl
i
c
a
t
i
on
of
a
n
E
M
D
-
D
F
A
-
D
B
N
hybr
i
d
m
ode
l
,”
C
om
put
e
r
s
and
E
l
e
c
t
r
oni
c
s
i
n
A
gr
i
c
ul
t
ur
e
, vol
. 211, 2023, doi
:
10.1016/
j
.c
om
pa
g.2023.107946.
[
32]
Z
.
S
a
’
a
di
,
Z
.
Y
us
op,
N
.
E
.
A
l
i
a
s
,
M
.
S
.
S
hi
r
u,
M
.
K
.
I
.
M
uha
m
m
a
d,
a
nd
M
.
W
.
A
.
R
a
m
l
i
,
“
A
ppl
i
c
a
t
i
on
of
C
H
I
R
P
S
da
t
a
s
e
t
i
n
t
he
s
e
l
e
c
t
i
on
of
r
a
i
n
-
ba
s
e
d
i
ndi
c
e
s
f
or
dr
ought
a
s
s
e
s
s
m
e
nt
s
i
n
J
ohor
R
i
ve
r
B
a
s
i
n,
M
a
l
a
ys
i
a
,
”
Sc
i
e
nc
e
of
t
he
T
ot
al
E
nv
i
r
onm
e
nt
,
vol
. 892, 2023, doi
:
10.1016/
j
.s
c
i
t
ot
e
nv.2023.164471.
[
33]
E
.
H
.
A
l
a
m
da
r
l
oo,
H
.
K
hos
r
a
vi
,
S
.
N
a
s
a
bpour
,
a
nd
A
.
G
hol
a
m
i
,
“
A
s
s
e
s
s
m
e
n
t
of
d
r
ought
ha
z
a
r
d,
vu
l
ne
r
a
bi
l
i
t
y
a
nd
r
i
s
k
i
n
I
r
a
n
us
i
ng G
I
S
t
e
c
hni
que
s
,”
J
ou
r
nal
of
A
r
i
d L
and
, vol
. 12, no. 6, 2020, doi
:
10.1007/
s
40333
-
020
-
0096
-
4.
[
3
4
]
F
.
A
h
m
a
d
i
,
S
.
M
e
h
d
i
z
a
de
h
,
a
n
d
B
.
M
oh
a
m
m
a
d
i
,
“
D
e
v
e
l
o
pm
e
n
t
o
f
b
i
o
-
i
ns
p
i
r
e
d
-
a
n
d
w
a
v
e
l
e
t
-
b
a
s
e
d
h
yb
r
i
d
m
o
de
l
s
f
o
r
r
e
c
o
n
na
i
s
s
a
n
c
e
d
r
o
u
g
h
t
i
nd
e
x
m
o
d
e
l
i
n
g
,”
W
a
t
e
r
R
e
s
o
ur
c
e
s
M
a
n
a
ge
m
e
n
t
,
vo
l
.
3
5,
n
o
.
1
2
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
s
11
2
6
9
-
0
21
-
02934
-
z.
[
35]
M
.
S
.
O
youna
l
s
oud,
A
.
G
.
Y
i
l
m
a
z
,
M
.
A
bda
l
l
a
h,
a
nd
A
.
A
bde
l
j
a
be
r
,
“
D
r
ou
ght
pr
e
di
c
t
i
on
us
i
ng
a
r
t
i
f
i
c
i
a
l
i
nt
e
l
l
i
ge
nc
e
m
ode
l
s
ba
s
e
d on c
l
i
m
a
t
e
da
t
a
a
nd s
oi
l
m
oi
s
t
ur
e
,”
Sc
i
e
nt
i
f
i
c
R
e
por
t
s
, vol
. 14, no. 1, D
e
c
.
2024, doi
:
10.1038/
s
41598
-
024
-
70406
-
6.
[
36]
M
.
G
.
G
üm
üş
,
H
.
Ç
.
Ç
i
f
t
ç
i
,
a
nd
K
.
G
üm
ü
ş
,
“
D
e
t
e
r
m
i
na
t
i
on
of
t
he
p
e
r
f
or
m
a
nc
e
of
t
r
a
i
ni
ng
a
l
gor
i
t
hm
s
a
nd
a
c
t
i
va
t
i
on
f
unc
t
i
ons
i
n
m
e
t
e
or
ol
ogi
c
a
l
dr
ought
i
nde
x
pr
e
di
c
t
i
on
w
i
t
h
nonl
i
ne
a
r
a
ut
o
r
e
gr
e
s
s
i
ve
ne
ur
a
l
ne
t
w
or
k,”
E
ar
t
h
Sc
i
e
nc
e
I
nf
or
m
at
i
c
s
,
vol
.
18,
no.
2,
F
e
b. 2025, doi
:
10.1007/
s
12145
-
025
-
01711
-
5.
[
37]
N
.
N
a
f
i
i
ya
h,
S
.
N
a
bi
l
a
h,
N
.
A
.
A
f
f
a
ndy,
T
.
Z
.
N
i
s
a
,
a
nd
R
.
A
.
F
a
r
oh,
“
P
e
r
f
or
m
a
nc
e
of
t
i
m
e
s
e
r
i
e
s
a
nd
de
e
p
l
e
a
r
ni
ng
m
ode
l
s
f
or
pr
e
di
c
t
i
ng
s
e
ve
r
e
dr
ought
a
r
e
a
s
i
n
L
a
m
onga
n
R
e
ge
nc
y,”
D
i
qi
u
K
e
x
ue
-
Z
hong
guo
D
i
z
hi
D
ax
ue
X
ue
bao/
E
ar
t
h
Sc
i
e
nc
e
J
our
nal
of
C
hi
na U
ni
v
e
r
s
i
t
y
of
G
e
os
c
i
e
n
c
e
s
, vol
. 50, no. 1, A
pr
. 2025, doi
:
10.5281/
z
e
nod
o.
B
I
O
G
R
A
P
H
I
E
S
O
F
A
U
T
H
O
R
S
Nur
Nafiiyah
received
her
Bachelor
of
Informatics
Engineering
f
ro
m
Universitas
Islam
Lamongan,
Indonesia
(2005
-
2009)
,
and
M
aster
of
Informatio
n
Technology
from
the
Sekolah
Tinggi
Teknik
Suraba
ya,
Indone
sia
(2011
-
2013).
She
ho
ld
s
a
Ph.D.
degree
in
Computer
Science
from
the
Department
of
Informatics,
Institut
Tekno
logi
Sepuluh
Nopember
(2019
-
2023).
She
is
currently
interested
in
artificial
intelligence,
deep
learning,
and
computer
vision.
She
has
also
been
teaching
artificial
intelligence
and
image
processing.
She
can
be
contacted
at email
: mynaff@
unisla.
ac.id.
Ali
Mokhtar
received
her
Bachelor
's
1991
in
Mechanical
Engineering
from
Universitas
Muhammadiyah
Malang,
Indonesia
and
her
Master
of
Mechanical
Engineering
from
Universitas
Indonesia,
Indonesia
,
in
2003.
He
is
the
Engineer
Pr
ofession
the
Universitas
Muhammadiyah
Malang
2019.
He
is
currently
inter
ested
in
energy
conversion
.
He
can
be
contacted
at email
: mokh
tar@
umm.ac.i
d.
Evaluation Warning : The document was created with Spire.PDF for Python.