I
A
E
S
I
n
t
e
r
n
at
io
n
al
Jou
r
n
al
of
A
r
t
if
ic
ia
l
I
n
t
e
ll
ig
e
n
c
e
(
I
J
-
AI
)
V
ol
.
13
, N
o.
4
,
D
e
c
e
m
be
r
2024
, pp.
4572
~
4582
I
S
S
N
:
2252
-
8938
,
D
O
I
:
10.11591/
ij
a
i.
v
13
.i
4
.pp
4572
-
4582
4572
Jou
r
n
al
h
om
e
page
:
ht
tp
:
//
ij
ai
.
ia
e
s
c
or
e
.c
om
Gr
i
n
d
u
l
u
f
au
l
t
c
l
ou
d
r
ad
o
n
d
at
a f
or
e
ar
t
h
q
u
ak
e
m
agn
i
t
u
d
e
p
r
e
d
i
c
t
i
on
u
si
n
g
m
ac
h
i
n
e
l
e
ar
n
i
n
g
T
h
om
as
O
k
a P
r
at
am
a
1,
2
, S
u
n
ar
n
o
2
, A
gu
s
B
u
d
h
ie
Wi
j
at
n
a
2
, E
k
o H
ar
yon
o
1,
3
1
D
oc
t
or
a
t
e
P
r
ogr
a
m
e
i
n E
nvi
r
on
m
e
nt
a
l
S
c
i
e
nc
e
, G
r
a
dua
t
e
S
c
hool
, U
ni
ve
r
s
i
t
a
s
G
a
dj
a
h M
a
da
, Y
ogy
a
ka
r
t
a
, I
ndone
s
i
a
2
D
e
pa
r
t
m
e
nt
of
N
uc
l
e
a
r
E
ngi
ne
e
r
i
ng a
nd E
ngi
ne
e
r
i
ng P
hys
i
c
s
, F
a
c
ul
t
y of
E
ngi
ne
e
r
i
ng, U
ni
ve
r
s
i
t
a
s
G
a
dj
a
h
M
a
da
, Y
ogya
ka
r
t
a
,
I
ndone
s
i
a
3
D
e
pa
r
t
m
e
nt
of
E
nvi
r
onm
e
nt
a
l
G
e
ogr
a
phy, F
a
c
ul
t
y of
G
e
ogr
a
phy, U
ni
ve
r
s
i
t
a
s
G
a
dj
a
h M
a
da
, Y
ogy
a
ka
r
t
a
, 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
F
e
b
13
,
2024
R
e
vi
s
e
d
A
pr
24,
2024
A
c
c
e
pt
e
d
J
un
5
,
2024
The
study
investigates
the
potential
of
integrating
radon
gas
concentration
telemonitoring
systems
with
machine
learning
techniques
to
enhance
earthquake
magnitu
de
predicti
on.
Conduct
ed
in
Pacitan,
East
Java,
Indonesi
a,
where
the
stations
are
near
the
acti
ve
Grindulu
fault,
the
research
employs
random forest
(RF),
extreme grad
ient bo
osting
(XGB),
neural network
(NN),
AdaBoost
(AB),
and
support
vector
machine
(SVM)
methods.
The
stud
y
aims
to
refine
earthquake
magnitude
prediction,
utilizing
real
-
time
rad
on
g
as
concentrati
on
measurement
s,
crucial
for
disaster
preparedness
.
The
eval
uation
involves
multiple
metrics
like
mean
absolute
error
(MAE),
mean
a
bsolute
percentage
error
(MAPE),
root
mean
square
error
(RMSE),
mean
s
quared
error
(MSE),
symmetric
mean
absolute
percentage
error
(SMAPE
),
and
conformal
normali
zed
mean
absolut
e
percentage
error
(cnSMAP
E)
.
X
GB
and
SVM
emerge
as
top
performers,
showcasing
superior
predictive
accuracy
with
minimal
errors
across
various
metrics.
XGB
achieve
d
MAE
(0.33),
MAPE
(6.03%),
RMSE
(0.51),
MSE
(0.26),
SMAPE
(0.06),
and
cnMAPE
(0.97),
while
SVM
recorded
MAE
(0.34),
MAPE
(6.20%),
RMSE
(0.51),
MSE
(0.26),
SMAPE
(0.06),
and
cn
S
MAPE
(0.97).
The
analysis
reveals
X
GB
as
the
most
effective
method,
boasting
the
lowe
st
error
values.
The
study
underscores
the
importance
of
expanding
data
availability
to
e
nhance
predictive
models,
ultimately
contributing
to
more
precise
eart
hquake
magnitude pr
edictions an
d effec
tive mitigation str
ategies.
K
e
y
w
o
r
d
s
:
E
a
r
th
qua
ke
M
a
c
hi
ne
l
e
a
r
ni
ng
M
a
gni
tu
de
P
r
e
di
c
ti
on
R
a
don
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
:
S
una
r
no
D
e
p
a
r
t
m
e
nt
of
N
u
c
l
e
a
r
E
n
g
in
e
e
r
in
g
a
n
d
E
ng
i
n
e
e
r
i
n
g
P
h
y
s
i
c
s
,
F
a
c
u
l
t
y
of
E
n
g
in
e
e
r
in
g
,
U
n
i
v
e
r
s
i
t
a
s
G
a
d
j
a
h
M
a
d
a
St
. G
r
a
f
ik
a
N
o.
2, S
in
dua
di
, M
la
ti
, S
le
m
a
n, Y
ogya
ka
r
ta
, I
ndone
s
ia
E
m
a
il
:
s
una
r
no@
ugm
.a
c
.i
d
1.
I
N
T
R
O
D
U
C
T
I
O
N
T
he
pr
e
di
c
ti
on
of
e
a
r
th
qua
ke
m
a
gni
tu
de
is
a
c
r
it
ic
a
l
a
r
e
a
of
r
e
s
e
a
r
c
h
due
to
it
s
pot
e
nt
ia
l
im
pa
c
t
on
publ
ic
s
a
f
e
ty
a
nd
in
f
r
a
s
tr
uc
tu
r
e
.
R
a
don
ga
s
h
a
s
b
e
e
n
r
e
g
a
r
de
d
a
s
a
po
s
s
ib
le
pr
e
c
ur
s
or
to
e
a
r
th
qua
ke
s
f
or
a
c
ons
id
e
r
a
bl
e
t
im
e
, a
nd c
r
e
a
ti
ng a
c
om
pr
e
he
n
s
iv
e
f
or
e
c
a
s
ti
ng s
y
s
te
m
t
ha
t
c
a
n a
c
c
ur
a
te
ly
pr
e
di
c
t
th
e
da
te
, t
im
e
,
m
a
gni
tu
de
,
a
nd
e
x
a
c
t
e
pi
c
e
nt
e
r
of
f
ut
ur
e
e
a
r
th
qua
ke
s
c
ont
in
ue
s
to
be
a
c
ha
ll
e
nge
[
1]
–
[
11]
.
O
ne
a
ppr
oa
c
h
to
e
a
r
th
qua
ke
m
a
gni
tu
de
pr
e
di
c
ti
on
in
vol
ve
s
us
in
g
a
te
le
m
oni
to
r
in
g
s
ys
te
m
ba
s
e
d
on
r
a
don
ga
s
c
onc
e
nt
r
a
ti
on
c
oupl
e
d w
it
h m
a
c
hi
ne
l
e
a
r
ni
ng t
e
c
hni
que
s
.
P
r
e
s
e
nt
ly
,
a
r
ti
f
ic
i
a
l
in
te
l
li
ge
n
c
e
(
A
I
)
m
e
th
od
ol
ogi
e
s
p
la
y
a
c
r
uc
ia
l
r
ol
e
in
e
a
r
th
qu
a
ke
pr
e
di
c
ti
on,
hi
ghl
ig
ht
i
ng
th
e
ir
c
a
pa
b
il
it
y
t
o
e
nh
a
nc
e
r
e
a
di
ne
s
s
a
nd
r
e
s
pon
s
e
t
a
c
ti
c
s
i
n
vul
n
e
r
a
bl
e
r
e
gi
on
s
[
5]
, [
11]
–
[
17]
.
T
hi
s
a
ppr
oa
c
h
l
e
ve
r
a
ge
s
a
d
va
n
c
e
d
t
e
c
h
nol
ogi
e
s
to
a
n
a
ly
z
e
a
nd
in
t
e
r
pr
e
t
d
a
ta
f
or
a
c
c
ur
a
te
pr
e
di
c
ti
on
s
.
M
a
c
hi
ne
le
a
r
ni
n
g
a
l
gor
it
hm
s
, s
u
c
h a
s
s
uppor
t
ve
c
to
r
m
a
c
hi
ne
s
(
S
V
M
)
, e
xt
r
e
m
e
le
a
r
ni
ng ma
c
hi
ne
s
(
E
L
M
)
,
a
nd a
r
ti
f
ic
i
a
l
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
G
r
in
dul
u
fa
ul
t
c
lo
ud r
adon data f
or
e
ar
th
qua
k
e
m
agni
tu
de
pr
e
di
c
ti
on
…
(
T
hom
as
O
k
a P
r
at
am
a
)
4573
ne
ur
a
l
n
e
tw
or
ks
(
A
N
N
s
)
,
ha
ve
b
e
e
n
a
ppl
i
e
d
in
e
a
r
th
qua
k
e
m
a
gni
tu
de
pr
e
di
c
ti
on
m
od
e
ls
[
14]
,
[
18]
–
[2
1
]
.
A
ddi
ti
on
a
ll
y,
in
t
e
gr
a
ti
ng
a
da
pt
iv
e
n
e
ur
o
-
f
uz
z
y
in
f
e
r
e
nc
e
s
y
s
te
m
s
(
A
N
F
I
S
)
ha
s
s
h
ow
n
pr
om
i
s
e
i
n
e
nh
a
n
c
in
g
t
he
a
c
c
ur
a
c
y
of
e
a
r
th
q
ua
k
e
m
a
gni
tu
d
e
pr
e
di
c
ti
ons
[
22]
.
F
ur
t
he
r
m
or
e
,
th
e
us
e
of
e
ns
e
m
bl
e
l
e
a
r
ni
n
g
m
e
t
hod
s
,
s
uc
h
a
s
A
da
B
oo
s
t,
h
a
s
be
e
n e
xpl
or
e
d t
o i
m
pr
ove
t
he
r
obu
s
tn
e
s
s
of
e
a
r
th
qua
ke
m
a
g
ni
tu
d
e
pr
e
di
c
ti
o
n m
od
e
l
s
[
23]
.
V
a
r
io
us
m
a
c
hi
ne
le
a
r
ni
ng
m
e
th
od
s
ha
ve
b
e
e
n
e
m
pl
oye
d
in
di
f
f
e
r
e
nt
s
tu
di
e
s
to
pr
e
di
c
t
e
a
r
th
qua
ke
m
a
gni
tu
de
s
.
F
or
in
s
ta
nc
e
,
th
e
m
in
im
um
a
r
e
a
o
f
th
e
a
la
r
m
m
e
th
od
w
a
s
ut
il
iz
e
d
to
pr
e
di
c
t
e
a
r
th
qua
ke
s
w
it
h
m
a
gni
tu
de
s
gr
e
a
te
r
t
ha
n 6.0
a
nd 5.5
i
n
J
a
pa
n
a
nd
C
a
li
f
or
ni
a
[
24]
.
S
im
il
a
r
ly
,
m
a
c
hi
ne
l
e
a
r
ni
ng t
e
c
hni
que
s
w
e
r
e
a
ppl
ie
d
f
or
e
a
r
th
qua
ke
m
a
gni
tu
de
pr
e
di
c
ti
on i
n t
h
e
H
in
dukus
h
r
e
gi
on
[
25]
.
F
ur
th
e
r
m
or
e
,
a
pr
oba
bi
li
s
ti
c
ne
ur
a
l
ne
twor
k
(
N
N
)
w
a
s
de
ve
lo
pe
d
f
or
e
a
r
th
qua
ke
m
a
gni
tu
de
pr
e
di
c
ti
on,
a
c
hi
e
vi
ng
good
e
a
r
th
qua
ke
r
e
s
ul
ts
w
it
h
m
a
gni
tu
de
s
r
a
ngi
ng
f
r
om
4.0
to
6.0
[
26]
.
A
ddi
t
io
na
ll
y,
it
pr
op
os
e
d
a
de
e
p
-
le
a
r
ni
ng
NN
f
or
la
r
ge
e
a
r
th
qua
ke
m
a
gni
tu
de
pr
e
di
c
ti
on
in
T
a
iwa
n,
obt
a
in
in
g
s
uc
c
e
s
s
f
ul
pr
e
di
c
ti
ons
f
or
e
a
r
th
qua
ke
s
w
it
h
m
a
gni
tu
de
s
be
twe
e
n
4.0
a
nd
6.0
[
27]
.
T
he
s
e
s
tu
di
e
s
de
m
on
s
tr
a
te
th
e
e
f
f
e
c
ti
ve
ne
s
s
of
m
a
c
hi
ne
le
a
r
ni
ng
m
e
th
ods
in
e
a
r
th
qua
ke
m
a
gni
tu
de
pr
e
di
c
ti
on.
E
a
r
th
qua
ke
m
a
gni
tu
de
pr
e
di
c
ti
on
ha
s
be
e
n
a
ddr
e
s
s
e
d
th
r
ough
va
r
io
us
m
a
c
hi
ne
le
a
r
ni
ng
m
e
th
ods
,
in
c
lu
di
ng
pr
oba
bi
li
s
ti
c
NN
a
nd
de
e
p
le
a
r
ni
ng
NN
.
T
he
in
c
or
por
a
ti
on
of
r
a
don
ga
s
c
onc
e
nt
r
a
ti
on
da
ta
in
to
th
e
pr
e
di
c
ti
on
m
ode
ls
a
li
gns
w
it
h
th
e
f
in
di
ngs
of
th
os
e
w
ho
in
ve
s
ti
ga
te
d
th
e
c
ha
r
a
c
te
r
is
ti
c
s
of
ge
oe
l
e
c
tr
ic
f
ie
ld
s
ig
na
ls
pr
io
r
to
e
a
r
th
qua
ke
s
[
28]
.
T
hi
s
s
ugge
s
ts
th
a
t
di
ve
r
s
e
ge
ophys
ic
a
l
d
a
ta
,
in
c
lu
di
ng
r
a
don
ga
s
c
onc
e
nt
r
a
ti
on,
c
ont
r
ib
ut
e
s
to
a
c
om
pr
e
he
ns
iv
e
unde
r
s
ta
n
di
ng
of
e
a
r
th
qua
ke
pr
e
c
ur
s
or
s
.
M
or
e
ove
r
,
G
óm
e
z
e
t
al
.
[
29]
pr
e
s
e
nt
e
d
a
nove
l
m
e
th
odol
ogy
f
or
pr
e
di
c
ti
ng
la
r
ge
-
m
a
gni
tu
de
e
a
r
th
qua
ke
s
in
c
hi
le
us
in
g
e
ns
e
m
bl
e
le
a
r
ni
ng,
e
m
pha
s
iz
in
g
th
e
pot
e
nt
ia
l
of
m
a
c
hi
ne
le
a
r
ni
ng
in
f
or
e
c
a
s
ti
ng
s
e
is
m
ic
e
ve
nt
s
w
it
h
s
ig
ni
f
ic
a
nt
m
a
gni
tu
de
s
.
A
ddi
ti
ona
ll
y,
G
it
is
a
nd
D
e
r
e
ndya
e
v
[
30]
di
s
c
u
s
s
e
d
m
a
c
hi
ne
le
a
r
ni
ng
m
e
th
ods
f
or
s
pa
ti
a
l
f
or
e
c
a
s
ti
ng
of
m
a
xi
m
um
pos
s
ib
le
e
a
r
th
qua
k
e
m
a
gni
tu
de
s
,
f
u
r
th
e
r
hi
ghl
ig
ht
in
g
th
e
r
e
le
va
nc
e
of
m
a
c
hi
ne
le
a
r
ni
ng i
n s
e
is
m
ic
ha
z
a
r
d
a
s
s
e
s
s
m
e
nt
s
.
C
ons
id
e
r
in
g t
he
a
ppl
ic
a
ti
on of
AI
in
e
a
r
th
qua
ke
pr
e
di
c
ti
on, B
a
n
na
e
t
al
.
[
11]
a
na
ly
s
is
unde
r
s
c
or
e
s
i
t
s
s
uc
c
e
s
s
in
f
or
e
c
a
s
ti
ng
e
a
r
th
qua
k
e
s
w
it
hi
n
c
e
r
ta
in
m
a
gni
tu
de
th
r
e
s
hol
ds
(
M
3
to
M
5)
.
H
ow
e
v
e
r
,
c
ha
ll
e
nge
s
a
r
is
e
in
pr
e
di
c
ti
ng
hi
gh
-
m
a
gni
tu
de
e
ve
nt
s
due
to
th
e
ir
in
f
r
e
que
nc
y
a
nd
unpr
e
di
c
ta
bl
e
oc
c
ur
r
e
nc
e
pa
tt
e
r
ns
.
S
ig
ni
f
ic
a
nt
ly
,
not
a
bl
e
di
s
c
r
e
pa
nc
ie
s
ha
ve
be
e
n
not
e
d
in
pr
e
di
c
ti
ng
th
e
ti
m
in
g
a
nd
lo
c
a
ti
on
of
e
a
r
th
qua
ke
s
,
w
it
h
de
vi
a
ti
ons
of
up
to
70
m
il
e
s
a
nd
c
ons
id
e
r
a
bl
e
va
r
ia
ti
ons
in
pr
e
di
c
ti
on
ti
m
e
f
r
a
m
e
s
r
a
ngi
n
g
f
r
om
20
da
ys
to
5
m
ont
hs
[
11]
.
A
c
c
or
di
ng
to
T
e
hs
e
e
n
e
t
al
.
[3
1
]
,
T
a
bl
e
1
pr
e
s
e
nt
s
th
e
a
c
c
ur
a
c
y
of
th
e
e
xpe
r
t
s
y
s
te
m
s
ugge
s
te
d
f
or
f
or
e
c
a
s
ti
ng
e
a
r
th
qua
ke
s
th
r
ough
a
n
in
de
pe
nd
e
nt
te
s
t
da
ta
s
e
t.
T
h
e
s
e
r
e
s
ul
ts
hi
ghl
ig
ht
th
e
in
tr
ic
a
c
y
a
nd
in
tr
in
s
ic
unc
e
r
ta
in
ti
e
s
li
nke
d
w
it
h
e
a
r
th
qua
ke
f
or
e
c
a
s
ti
ng,
pr
om
pt
in
g
c
ont
in
uous
e
xpl
or
a
ti
on
a
nd
im
pr
ove
m
e
nt
of
m
e
th
od
ol
ogi
e
s
to
boo
s
t
pr
e
di
c
ti
ve
a
c
c
ur
a
c
y
a
nd
de
pe
nda
b
il
it
y.
T
he
r
e
s
e
a
r
c
h
f
oc
u
s
h
a
s
s
hi
f
te
d
to
w
a
r
d
m
a
c
hi
ne
le
a
r
ni
ng
a
nd
d
e
e
p
l
e
a
r
n
i
n
g
t
e
c
h
n
i
qu
e
s
s
i
n
c
e
2
0
1
8
,
s
ig
n
i
f
y
i
n
g
a
n
ot
e
w
o
r
t
h
y
e
v
ol
u
t
io
n
i
n
e
a
r
t
h
q
u
a
k
e
p
r
e
d
i
c
ti
o
n
a
p
p
r
o
a
c
h
e
s
.
T
a
bl
e
1
. A
c
c
ur
a
c
y i
s
c
la
im
e
d i
n
a
n e
xpe
r
t
s
ys
t
e
m
us
in
g a
n i
nde
pe
nde
nt
t
e
s
t
s
e
t
[3
1
]
R
e
f
e
r
e
nc
e
s
N
um
be
r
of
e
a
r
t
hqua
ke
r
e
c
or
ds
A
c
c
ur
a
c
y
(%)
M
a
gni
t
ude
r
a
nge
[3
2
]
9
,
531
69.8
≥
2.0
[
33
]
12
,
690
50.14
≥
3.0
[
34
]
337
63
≥
3.0
[
35
]
10
,
567
40
0.1
–
5.9
I
nt
e
gr
a
ti
ng
r
a
don
ga
s
c
onc
e
nt
r
a
ti
on
te
l
e
m
oni
to
r
in
g
s
ys
te
m
s
w
it
h
m
a
c
hi
ne
le
a
r
ni
ng
te
c
hni
que
s
pr
e
s
e
nt
s
a
pr
om
is
in
g
a
ppr
oa
c
h
to
e
a
r
th
qua
ke
m
a
gni
tu
de
pr
e
di
c
ti
on.
B
y
le
ve
r
a
gi
ng
a
dva
nc
e
d
a
lg
or
it
hm
s
a
nd
di
ve
r
s
e
ge
ophys
ic
a
l
da
ta
,
r
e
s
e
a
r
c
he
r
s
a
im
to
im
pr
ove
th
e
a
c
c
ur
a
c
y
a
nd
r
e
li
a
bi
li
ty
of
e
a
r
th
qua
ke
m
a
gni
tu
d
e
pr
e
di
c
ti
ons
,
ul
ti
m
a
te
ly
c
ont
r
ib
ut
in
g
to
e
nha
nc
e
d
di
s
a
s
te
r
pr
e
pa
r
e
dne
s
s
a
nd
r
is
k
m
it
ig
a
ti
on.
I
n
I
ndone
s
ia
,
th
e
r
e
is
a
m
e
th
od
f
or
pr
e
di
c
ti
ng
th
e
ti
m
e
o
f
a
n
e
a
r
th
qua
ke
1
-
4
da
ys
a
f
t
e
r
th
e
a
la
r
m
goe
s
o
f
f
w
it
h
a
m
a
gn
it
ude
g
r
e
a
te
r
th
a
n
4.5M
ba
s
e
d
on
th
e
r
a
don
ga
s
c
onc
e
nt
r
a
ti
on
te
le
m
oni
to
r
in
g,
but
it
doe
s
not
s
pe
c
if
ic
a
ll
y
pr
e
di
c
t
th
e
m
a
gni
tu
de
of
th
e
e
a
r
th
qua
ke
[
5]
,
[
3
6
]
.
I
n
th
is
s
tu
dy,
e
a
r
th
qua
ke
m
a
gni
tu
de
pr
e
di
c
ti
on
is
c
onduc
te
d
ba
s
e
d
on
r
e
a
l
-
ti
m
e
te
le
m
oni
to
r
in
g
of
r
a
don
ga
s
c
onc
e
nt
r
a
ti
on
a
t
s
ta
ti
ons
i
n
P
a
c
it
a
n,
E
a
s
t
J
a
va
,
I
ndone
s
ia
ne
a
r
G
r
in
dul
u
f
a
ul
t
us
in
g
m
a
c
hi
ne
le
a
r
ni
ng
m
e
th
ods
r
a
ndom
f
or
e
s
t
(
R
F
)
,
e
xt
r
e
m
e
gr
a
di
e
nt
boos
ti
ng
(
X
G
B
)
,
N
N
,
a
da
boos
t
(
A
B
)
, S
V
M
. T
he
s
e
s
ta
ti
ons
a
lr
e
a
dy
ha
ve
a
n e
a
r
th
qua
ke
t
im
e
pr
e
di
c
ti
on a
lg
or
it
hm
ba
s
e
d
on
P
r
a
ta
m
a
e
t
al
.
[3
6
]
w
it
h
s
e
ns
it
iv
it
y
a
nd
pr
e
c
is
io
n
of
78.79%
a
nd
70.27%
,
r
e
s
pe
c
ti
ve
ly
.
T
he
de
ve
lo
pm
e
nt
of
th
e
e
a
r
th
qua
ke
m
a
gni
tu
de
pr
e
di
c
ti
on
s
ys
te
m
c
a
n
he
lp
m
it
ig
a
te
e
a
r
th
qua
ke
na
tu
r
a
l
di
s
a
s
te
r
s
s
o
th
a
t
it
c
a
n
pr
ovi
de
w
a
r
ni
ngs
to
th
e
publi
c
, vol
unt
e
e
r
s
, or
t
he
gove
r
nm
e
nt
t
o m
in
im
iz
e
c
a
s
ua
lt
ie
s
.
2.
M
E
T
H
O
D
T
he
r
e
a
l
-
ti
m
e
m
oni
to
r
in
g
s
ys
te
m
f
or
r
a
don
ga
s
c
onc
e
nt
r
a
ti
on
i
s
c
lo
s
e
to
th
e
Gr
i
ndul
u
a
c
ti
ve
f
a
ul
t
in
P
a
c
it
a
n,
E
a
s
t
J
a
va
,
I
ndone
s
ia
,
m
a
ki
ng
it
pr
one
to
s
e
is
m
ic
a
c
ti
vi
ty
.
T
he
r
a
don
ga
s
tr
a
ns
duc
e
r
i
s
s
tr
a
te
gi
c
a
ll
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
13
, N
o.
4
,
D
e
c
e
m
be
r
2024
:
4572
-
4582
4574
pos
it
io
ne
d
w
it
hi
n
th
e
c
ha
m
be
r
r
oom
to
e
ns
ur
e
pr
e
c
is
e
m
e
a
s
ur
e
m
e
nt
of
r
a
don
ga
s
e
m
is
s
io
ns
,
m
a
in
ta
in
in
g
a
m
a
xi
m
um
of
4
.
142
c
m
a
bove
gr
ound
le
ve
l.
R
a
don
ga
s
m
e
a
s
ur
e
m
e
nt
s
a
r
e
ta
ke
n
e
v
e
r
y
10
m
in
ut
e
s
to
m
it
ig
a
te
th
e
in
f
lu
e
nc
e
of
r
a
di
a
ti
on
e
m
i
s
s
io
ns
[3
7
]
.
T
he
e
a
r
th
qua
ke
pr
e
di
c
ti
on
s
ys
te
m
'
s
de
s
ig
n
is
il
lu
s
tr
a
te
d
in
F
ig
ur
e
1.
S
ubs
e
que
nt
ly
,
da
ta
c
ol
le
c
te
d
by
th
e
r
a
don
ga
s
tr
a
ns
du
c
e
r
is
s
e
nt
to
th
e
m
ic
r
opr
oc
e
s
s
or
a
nd
f
or
w
a
r
de
d
to
th
e
c
lo
ud
s
e
r
ve
r
f
or
r
e
a
l
-
ti
m
e
m
on
it
or
in
g,
c
ont
in
ge
nt
upon
a
n
a
c
ti
ve
in
te
r
ne
t
c
onne
c
ti
on.
N
ot
a
bl
y,
th
e
r
e
c
or
de
d
r
a
don ga
s
da
t
a
i
s
s
e
c
ur
e
ly
s
to
r
e
d i
n a
de
s
ig
na
t
e
d da
ta
s
to
r
a
ge
s
e
r
ve
r
a
nd a
c
c
e
s
s
ib
le
vi
a
a
d
e
di
c
a
te
d
w
e
b
s
e
r
ve
r
.
M
or
e
ove
r
,
e
a
r
th
qua
ke
-
r
e
la
te
d
d
a
ta
is
a
c
qui
r
e
d
f
r
om
G
e
of
on
P
ot
s
da
m
a
nd
th
e
I
ndone
s
ia
n
A
ge
nc
y
f
or
m
e
te
or
ol
ogy, c
l
im
a
to
lo
gy, a
nd ge
ophys
ic
s
, e
nr
ic
hi
ng t
he
und
e
r
s
ta
ndi
ng of
s
e
is
m
ic
e
v
e
nt
s
.
F
ig
ur
e
1. E
a
r
th
qua
ke
pr
e
di
c
ti
on s
ys
te
m
bl
oc
k di
a
gr
a
m
[3
8
]
T
he
a
lg
or
it
hm
f
or
pr
e
di
c
ti
ng
e
a
r
th
qua
ke
m
a
gni
tu
de
is
c
r
e
a
te
d
th
r
ough
s
upe
r
vi
s
e
d
m
a
c
hi
ne
le
a
r
ni
ng
m
e
th
ods
,
ut
il
iz
in
g
da
ta
e
xt
r
a
c
te
d
f
r
om
r
a
don
c
lo
ud
da
ta
a
nd
hi
s
to
r
ic
a
l
E
a
r
th
qua
ke
in
c
id
e
nt
s
.
T
he
m
ode
l'
s
pe
r
f
or
m
a
nc
e
a
s
s
e
s
s
m
e
nt
in
vol
ve
s
va
r
io
us
m
e
tr
ic
s
,
in
c
lu
di
ng
a
c
c
ur
a
c
y
,
m
e
a
n
s
qua
r
e
d
e
r
r
or
(
M
S
E
)
,
r
oot
m
e
a
n
s
qua
r
e
e
r
r
or
(
R
M
S
E
)
, m
e
a
n a
bs
ol
ut
e
pe
r
c
e
nt
a
ge
e
r
r
or
(
M
A
P
E
)
,
m
e
a
n a
bs
ol
ut
e
e
r
r
or
(
M
A
E
)
, s
ym
m
e
tr
ic
m
e
a
n
a
bs
ol
ut
e
pe
r
c
e
nt
a
ge
e
r
r
or
(
S
M
A
P
E
)
,
c
onf
or
m
a
l
no
r
m
a
li
z
e
d
m
e
a
n
a
bs
ol
ut
e
pe
r
c
e
nt
a
ge
e
r
r
or
(
c
nS
M
A
P
E
)
,
a
nd
a
c
c
ur
a
c
y
.
F
ol
lo
w
in
g
e
va
lu
a
ti
on,
th
e
m
os
t
e
f
f
e
c
ti
ve
m
ode
l
w
il
l
be
im
pl
e
m
e
nt
e
d
on
a
c
lo
ud
s
e
r
ve
r
to
de
li
ve
r
e
a
r
th
qua
ke
pr
e
di
c
ti
on a
le
r
ts
.
T
a
bl
e
2
il
lu
s
tr
a
te
s
th
e
c
om
pos
it
io
n
of
th
e
r
a
don
da
ta
s
e
t
[
5]
.
S
ubs
e
que
nt
ly
,
T
a
bl
e
3
pr
e
s
e
nt
s
th
e
s
ys
te
m
a
ti
c
or
ga
ni
z
a
ti
on
of
in
f
or
m
a
ti
on
r
e
ga
r
di
ng
r
a
don
ga
s
c
onc
e
nt
r
a
ti
ons
a
nd
e
a
r
th
qua
ke
oc
c
ur
r
e
nc
e
s
,
a
dhe
r
in
g
to
P
r
a
ta
m
a
a
ppr
oa
c
h
[
5]
.
T
he
tr
a
in
in
g
da
ta
s
e
t
c
ons
is
te
d
of
r
a
don
ga
s
c
onc
e
nt
r
a
ti
on
da
ta
c
or
r
e
s
ponding
to
t
he
da
y of
t
he
e
a
r
th
qua
ke
pr
e
di
c
ti
on
. I
n
c
ont
r
a
s
t,
th
e
t
e
s
t
da
ta
s
e
t
in
c
lu
de
d e
a
r
th
qua
ke
s
w
it
hi
n 1
-
4 da
ys
a
f
te
r
th
e
in
it
ia
l
e
a
r
th
qua
ke
d
a
te
pr
e
di
c
ti
on,
w
it
h a
m
in
im
um
m
a
gni
tu
de
th
r
e
s
hol
d
of
M
4.5
be
twe
e
n
th
e
E
ur
a
s
ia
a
nd
I
ndo
-
A
us
tr
a
li
a
P
la
te
s
.
T
h
e
da
ta
c
ol
le
c
ti
o
n
pe
r
io
d
r
a
nge
d
f
r
om
2/
2/
2022
to
22/
1/
2024,
c
om
pr
is
in
g
117
da
t
a
poi
nt
s
. S
e
ve
nt
y
pe
r
c
e
nt
of
t
he
d
a
ta
s
e
t
w
a
s
a
ll
oc
a
te
d f
or
t
r
a
in
in
g
pur
pos
e
s
, w
it
h t
he
r
e
m
a
in
in
g por
ti
on r
e
s
e
r
ve
d
f
or
e
va
lu
a
ti
ng a
nd pr
e
di
c
ti
ng e
a
r
th
qua
ke
m
a
gni
tu
de
s
.
T
a
bl
e
2
. C
om
po
s
it
io
n of
t
he
da
ta
s
e
t
[
5]
V
a
r
i
a
bl
e
D
e
s
c
r
i
pt
i
on
d
Rd
R
(
d
-
1)
R
(
d
-
2)
.
.
R
(
d
-
6)
R
(
d
-
7)
.
.
X
R
(
d
-
3)
X
R
(
d
-
7)
X
R
(
d
-
14)
T
he
da
y on w
hi
c
h t
he
a
l
gor
i
t
hm
f
or
e
c
a
s
t
w
a
s
f
i
na
l
i
z
e
d ut
i
l
i
z
i
ng
t
he
P
r
a
t
a
m
a
m
e
t
hodol
ogy
[
5
]
R
a
don da
t
a
a
ve
r
a
ge
d
a
y d
R
a
don da
t
a
a
ve
r
a
ge
d
a
y d
-
1
R
a
don da
t
a
a
ve
r
a
ge
d
a
y d
-
2
.
.
R
a
don da
t
a
a
ve
r
a
ge
d
a
y d
-
6
R
a
don da
t
a
a
ve
r
a
ge
d
a
y d
-
7
.
.
R
a
don da
t
a
a
ve
r
a
ge
3 d
a
ys
be
f
or
e
R
(
d
-
2)
= a
ve
r
a
ge
R
(
d
-
3)
t
o R
(
d
-
5)
R
a
don da
t
a
a
ve
r
a
ge
7 d
a
ys
be
f
or
e
R
(
d
-
2)
= a
ve
r
a
ge
R
(
d
-
3)
t
o R
(
d
-
9)
R
a
don da
t
a
a
ve
r
a
ge
14 d
a
ys
be
f
or
e
R
(
d
-
2)
= a
ve
r
a
ge
R
(
d
-
3)
t
o R
(
d
-
17)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
G
r
in
dul
u
fa
ul
t
c
lo
ud r
adon data f
or
e
ar
th
qua
k
e
m
agni
tu
de
pr
e
di
c
ti
on
…
(
T
hom
as
O
k
a P
r
at
am
a
)
4575
T
a
bl
e
3
. A
n i
ns
t
a
nc
e
of
a
da
ta
s
e
t
E
a
r
t
h
q
u
a
ke
d
a
t
e
p
r
e
d
i
c
t
i
o
n
XR
(d
-
14)
XR
(d
-
7)
X
R
(d
-
3)
R
(d
-
7)
R
(d
-
6)
R
(d
-
5)
R
(d
-
4)
R
(d
-
3)
R
(d
-
2)
R
(d
-
1)
E
a
r
t
h
q
u
a
ke
d
a
t
e
D
i
s
t
a
n
c
e
(
k
m
)
A
c
t
u
a
l
M
a
g
ni
t
u
d
e
2
/
2
/
2
0
2
2
58.02
58.81
68.17
45.24
64.41
77.03
71.48
56.00
33.43
49.93
2
/
4
/
2
0
2
2
577.73
5.5
2
/
9
/
2
0
2
2
53.88
48.96
57.33
40.83
46.57
43.88
55.47
72.62
74.74
47.90
2
/
1
0
/
2
0
22
1
,
0
1
0
.
32
4.6
2
/
1
5
/
2
0
22
56.54
66.49
70.63
47.90
58.30
63.77
60.03
88.08
78.26
47.48
2
/
1
7
/
2
0
22
468.23
5.1
T
he
m
a
c
hi
ne
le
a
r
ni
ng
pr
oc
e
s
s
e
m
pl
oye
d
in
th
is
s
tu
dy
f
ol
lo
w
s
s
u
pe
r
vi
s
e
d
le
a
r
ni
ng
pr
in
c
ip
le
s
,
ut
il
iz
in
g
a
r
e
gr
e
s
s
io
n t
e
c
hni
que
a
s
de
pi
c
te
d i
n F
ig
ur
e
2.
T
he
obj
e
c
ti
ve
i
s
f
or
t
he
m
ode
l
to
di
s
c
e
r
n und
e
r
ly
in
g pa
tt
e
r
ns
or
r
e
la
ti
ons
hi
ps
w
it
hi
n t
he
da
ta
s
e
t,
e
na
bl
in
g a
c
c
ur
a
te
pr
e
di
c
ti
on
s
o
f
e
a
r
th
qua
ke
m
a
gni
tu
de
s
on uns
e
e
n da
ta
ba
s
e
d
on
th
e
r
a
don
g
a
s
c
onc
e
nt
r
a
ti
on
f
r
om
a
te
le
m
oni
to
r
in
g
s
ta
ti
on
ne
a
r
th
e
G
r
in
dul
u
f
a
ul
t.
R
F
,
X
G
B
,
N
N
, A
B
, a
nd
S
V
M
a
r
e
m
a
c
hi
ne
le
a
r
ni
ng
te
c
hni
que
s
us
e
d
in
de
r
iv
in
g
e
a
r
th
qua
ke
m
a
gni
tu
de
pr
e
di
c
ti
on
a
lg
or
it
hm
s
[2
7
]
–
[3
4
]
.
T
he
t
r
a
in
in
g da
ta
s
e
t
is
ut
il
iz
e
d t
o c
on
s
tr
uc
t
th
e
e
a
r
th
qua
ke
m
a
gn
it
ude
pr
e
di
c
ti
on mode
l,
w
hi
c
h i
s
s
ubs
e
que
nt
ly
e
va
lu
a
te
d us
in
g t
he
t
e
s
t
da
ta
s
e
t
to
a
s
s
e
s
s
i
ts
e
f
f
ic
a
c
y.
T
hi
s
r
e
s
e
a
r
c
h
im
pl
e
m
e
nt
e
d
R
F
,
X
G
B
,
N
N
,
A
B
,
a
nd
S
V
M
m
e
th
odol
ogi
e
s
us
in
g
or
a
nge
da
ta
m
in
in
g
ve
r
s
io
n
3.36.2
s
of
twa
r
e
.
T
he
tu
ni
ng
is
done
on
th
e
f
e
a
tu
r
e
s
of
e
a
c
h
m
a
c
hi
ne
le
a
r
ni
ng
m
e
th
od
to
obt
a
in
th
e
be
s
t
m
ode
l.
B
y
in
te
gr
a
ti
ng
th
e
a
ppl
ic
a
ti
on
of
m
a
c
hi
ne
le
a
r
ni
ng
m
od
e
ls
w
it
h
e
va
lu
a
ti
ons
ba
s
e
d
on
va
r
io
us
m
e
tr
ic
s
,
th
is
s
tu
dy
a
im
s
to
of
f
e
r
a
c
om
pr
e
he
ns
iv
e
in
s
ig
ht
in
to
th
e
m
ode
l'
s
pe
r
f
or
m
a
nc
e
in
pr
e
di
c
ti
ng
or
a
na
ly
z
in
g
th
e
da
ta
unde
r
c
ons
id
e
r
a
ti
on.
F
ig
ur
e
2
. S
tr
uc
tu
r
e
of
a
s
upe
r
vi
s
e
d m
a
c
hi
ne
l
e
a
r
ni
ng mode
l
[
38
]
3.
R
E
S
U
L
T
S
A
N
D
D
I
S
C
U
S
S
I
O
N
T
hi
s
s
tu
dy'
s
e
a
r
th
qua
ke
m
a
gni
tu
de
pr
e
di
c
ti
on
is
ba
s
e
d
on
r
a
don
ga
s
c
onc
e
nt
r
a
ti
on
m
e
a
s
ur
e
m
e
nt
s
ne
a
r
th
e
a
c
ti
ve
G
r
in
du
lu
f
a
ul
t.
T
he
r
a
don
ga
s
c
onc
e
nt
r
a
ti
on
da
ta
s
to
r
a
ge
is
c
onf
ig
ur
e
d
a
c
c
or
di
ng
to
th
e
pr
e
de
te
r
m
in
e
d da
ta
s
e
t.
T
he
da
ta
s
e
t
f
or
t
hi
s
s
tu
dy
c
ons
is
t
s
of
82
tr
a
in
in
g da
ta
a
nd 35 te
s
t
da
t
a
poi
nt
s
. F
r
om
t
hi
s
tr
a
in
in
g
da
ta
s
e
t,
m
ode
li
ng
is
p
e
r
f
or
m
e
d
us
in
g
R
F
,
X
G
B
,
N
N
,
A
B
,
a
nd
S
V
M
m
e
th
ods
,
e
a
c
h
w
it
h
di
f
f
e
r
e
nt
f
e
a
tu
r
e
s
e
tt
in
gs
,
unt
il
th
e
be
s
t
r
e
s
ul
ts
a
r
e
a
c
hi
e
ve
d.
T
he
out
c
om
e
de
r
iv
e
d
f
r
om
th
is
m
a
c
hi
ne
le
a
r
ni
ng
p
r
oc
e
dur
e
yi
e
ld
s
t
he
f
or
e
c
a
s
te
d m
a
gni
tu
de
of
f
or
th
c
om
in
g e
a
r
th
qua
ke
s
b
a
s
e
d on the
i
nput
t
e
s
t
da
ta
.
T
a
bl
e
4
il
lu
s
tr
a
te
s
t
he
f
or
e
c
a
s
t
out
c
om
e
s
de
r
iv
e
d
f
r
om
th
e
tr
a
in
in
g
da
ta
s
e
t
us
in
g
a
c
onf
us
io
n
m
a
tr
ix
a
nd
m
e
tr
ic
s
s
uc
h a
s
M
A
E
, M
S
E
, R
M
S
E
, a
nd s
ta
nda
r
d de
vi
a
ti
on r
e
pr
e
s
e
nt
in
g t
he
va
r
ia
nc
e
be
twe
e
n
a
c
tu
a
l
a
nd
pr
e
di
c
te
d
m
a
gni
tu
de
.
A
tr
ue
pos
it
iv
e
s
c
e
na
r
io
e
m
e
r
ge
s
w
he
n
t
he
a
c
tu
a
l
m
a
gni
tu
de
li
e
s
w
it
hi
n
th
e
pr
e
di
c
te
d
m
a
gni
tu
de
r
a
nge
pl
us
or
m
in
us
th
e
s
ta
nda
r
d
de
vi
a
ti
on
e
r
r
or
.
C
o
nve
r
s
e
ly
,
a
f
a
ls
e
pos
it
iv
e
a
r
is
e
s
w
he
n
th
e
a
c
tu
a
l
m
a
gni
tu
de
f
a
ll
s
out
s
id
e
th
e
pr
e
di
c
te
d
m
a
gni
tu
de
r
a
nge
pl
us
or
m
in
us
th
e
s
ta
nda
r
d
de
vi
a
ti
o
n
e
r
r
or
.
T
he
e
va
lu
a
ti
on
r
e
s
ul
ts
of
th
e
tr
a
in
in
g
da
ta
s
how
e
d
th
a
t
th
e
X
G
B
m
e
t
hod
ha
s
th
e
hi
ghe
s
t
a
c
c
ur
a
c
y
c
om
pa
r
e
d
to
ot
he
r
m
e
th
ods
, w
it
h a
t
r
ue
pos
it
iv
e
o
f
68,
f
ol
lo
w
e
d by
t
he
A
B
, N
N
, S
V
M
, a
nd R
F
w
it
h t
r
ue
pos
it
iv
e
s
c
ons
e
c
ut
iv
e
ly
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
13
, N
o.
4
,
D
e
c
e
m
be
r
2024
:
4572
-
4582
4576
58, 53, 51, a
nd 51.
T
he
X
G
B
m
e
th
od ha
s
t
he
hi
ghe
s
t
a
c
c
ur
a
c
y
w
it
h a
va
lu
e
of
82.93%
a
nd
t
he
l
ow
e
s
t
s
ta
nd
a
r
d
de
vi
a
ti
on,
M
A
E
,
M
S
E
,
a
nd
R
M
S
E
va
lu
e
s
c
om
pa
r
e
d
to
ot
he
r
m
e
th
ods
.
I
t
in
di
c
a
te
s
be
tt
e
r
a
c
c
ur
a
c
y
c
om
pa
r
e
d
to
ot
he
r
m
ode
ls
.
T
h
e
m
ode
ls
va
r
y
in
a
bi
li
ty
to
c
or
r
e
c
tl
y
id
e
nt
if
y
pos
it
iv
e
c
a
s
e
s
(
tr
ue
pos
it
iv
e
s
)
a
nd
m
in
im
iz
e
f
a
ls
e
pos
it
iv
e
s
.
T
a
bl
e
4
. M
a
c
hi
ne
le
a
r
ni
ng da
ta
t
r
a
in
in
g t
e
s
t
r
e
s
ul
t
P
a
r
a
m
e
t
e
r
L
e
a
r
ni
ng M
e
t
hod
s
RF
XGB
NN
AB
S
V
M
S
t
a
nda
r
d D
e
vi
a
t
i
on
0.29
0.15
0.5
0.2
0.44
M
A
E
0.21
0.05
0.37
0.08
0.34
M
S
E
0.08
0.02
0.27
0.04
0.19
R
M
S
E
0.29
0.15
0.52
0.2
0.43
T
r
ue
P
os
i
t
i
ve
50
68
53
58
51
F
a
l
s
e
P
os
i
t
i
ve
32
14
29
24
31
A
c
c
ur
a
c
y (
%
)
60.98
82.93
64.63
70.73
62.20
B
a
s
e
d
on
th
e
e
va
lu
a
ti
on
r
e
s
ul
ts
obt
a
in
e
d
f
or
th
e
pr
e
di
c
ti
ve
a
lg
o
r
it
hm
m
ode
ls
,
e
a
r
th
qua
ke
m
a
gni
tu
de
pr
e
di
c
ti
ons
w
e
r
e
m
a
de
u
s
in
g
te
s
t
da
t
a
ba
s
e
d
on
e
a
r
th
qua
ke
da
t
e
pr
e
di
c
ti
ons
f
r
om
P
r
a
ta
m
a
[
5]
m
e
th
od,
w
he
r
e
e
a
r
th
qua
ke
pr
e
di
c
ti
ons
a
r
e
v
a
li
d
1
-
4
da
y
s
a
f
te
r
th
e
pr
e
di
c
te
d
e
a
r
th
qua
ke
da
te
.
T
a
bl
e
5
il
lu
s
tr
a
te
s
th
e
e
r
r
or
e
va
lu
a
ti
on
of
th
e
m
a
c
hi
ne
le
a
r
ni
ng
-
ba
s
e
d
e
a
r
th
qua
ke
m
a
gni
tu
de
pr
e
di
c
ti
on
m
e
th
od
s
ubs
e
que
nt
to
te
s
ti
ng
th
e
tr
a
in
in
g
da
ta
s
e
t
a
nd
c
onf
ig
ur
in
g
f
e
a
tu
r
e
s
to
ge
ne
r
a
te
opt
im
a
l
pr
e
di
c
ti
ons
.
T
he
RF
a
nd
S
V
M
me
th
od
ha
s
th
e
lo
w
e
s
t
s
ta
nda
r
d de
vi
a
ti
on
(
0.
38
)
. T
he
s
ta
nda
r
d de
vi
a
ti
on of
a
bs
ol
ut
e
e
r
r
or
e
va
lu
a
ti
on ha
s
a
va
lu
e
t
ha
t
doe
s
not
di
f
f
e
r
m
uc
h
be
twe
e
n
m
a
c
hi
ne
le
a
r
ni
ng
m
e
th
ods
.
T
he
X
G
B
ha
s
th
e
lo
w
e
s
t
va
lu
e
of
M
A
E
(
0.33)
,
M
A
P
E
(
6.03%
)
,
R
M
S
E
(
0.51
)
,
M
S
E
(
0.26)
,
S
M
A
P
E
(
0.06)
a
nd
c
n
M
A
P
E
(
0.97)
.
S
m
a
ll
e
r
va
lu
e
s
of
th
e
s
e
m
e
tr
ic
s
s
ugge
s
t
s
upe
r
io
r
pe
r
f
or
m
a
nc
e
of
t
he
a
lg
or
it
hm
.
S
V
M
a
ls
o pe
r
f
or
m
s
c
ons
is
te
nt
ly
w
e
ll
, w
it
h c
om
p
e
ti
ti
ve
va
lu
e
s
a
c
r
os
s
m
os
t
e
r
r
or
i
ndi
c
e
s
.
R
F
, N
N
, a
nd A
B
s
how
s
li
ght
ly
hi
ghe
r
e
r
r
or
in
di
c
e
s
th
a
n
X
G
B
a
nd S
V
M
, i
ndi
c
a
ti
ng
s
li
ght
ly
lo
w
e
r
pr
e
di
c
ti
ve
a
c
c
ur
a
c
y.
S
V
M
ha
s
th
e
hi
gh
e
s
t
m
a
gni
tu
de
pr
e
di
c
ti
on
a
c
c
ur
a
c
y
w
it
h
a
va
lu
e
of
77.14%
,
f
ol
lo
w
e
d
by
A
B
(
71.43%
)
,
N
N
(
68.57
%
)
,
X
G
B
,
a
nd
R
F
(
65.71%
)
.
X
G
B
a
nd
S
V
M
a
ppe
a
r
to
be
th
e
to
p
-
pe
r
f
or
m
in
g l
e
a
r
ni
ng
m
e
th
ods
ba
s
e
d on
t
he
pr
ovi
de
d
e
r
r
or
i
ndi
c
e
s
.
A
c
c
or
di
ng t
o our
a
na
ly
s
is
,
X
G
B
ha
s
t
h
e
lo
w
e
s
t
M
A
E
a
nd
R
M
S
E
va
lu
e
s
,
in
di
c
a
ti
ng
good
pe
r
f
or
m
a
nc
e
in
m
in
im
iz
in
g
th
e
a
bs
ol
ut
e
e
r
r
o
r
in
r
e
gr
e
s
s
io
n
ta
s
ks
.
H
ow
e
ve
r
,
S
V
M
a
c
hi
e
ve
d
th
e
hi
ghe
s
t
a
c
c
ur
a
c
y
a
m
ong
th
e
c
la
s
s
if
ic
a
ti
on
ta
s
k
a
lg
or
it
hm
s
.
T
he
s
e
r
e
s
ul
ts
s
ugge
s
t
th
a
t
X
G
B
pe
r
f
or
m
s
be
tt
e
r
in
ta
s
ks
w
he
r
e
a
bs
ol
ut
e
e
r
r
or
m
in
im
iz
a
ti
on
is
im
por
ta
nt
,
w
he
r
e
a
s
S
V
M
pe
r
f
or
m
s
be
tt
e
r
in
ta
s
ks
w
he
r
e
c
l
a
s
s
if
ic
a
ti
on
a
c
c
ur
a
c
y
i
s
pa
r
a
m
o
unt
.
R
e
s
e
a
r
c
h
e
r
s
a
nd
pr
a
c
ti
ti
one
r
s
c
a
n
us
e
th
e
s
e
in
s
i
ght
s
to
s
e
le
c
t
th
e
m
os
t
a
ppr
opr
ia
te
m
a
c
hi
ne
le
a
r
ni
ng
a
lg
or
it
hm
s
ba
s
e
d
on
th
e
ir
s
pe
c
if
ic
goa
l
s
a
nd
pe
r
f
or
m
a
nc
e
c
r
it
e
r
ia
.
T
a
bl
e
5
. E
a
r
th
qua
ke
m
a
gni
tu
de
pr
e
di
c
ti
on e
r
r
or
e
va
lu
a
ti
on
E
r
r
or
i
nde
x
L
e
a
r
ni
ng
m
e
t
hods
RF
XGB
NN
AB
S
V
M
S
t
D
e
v of
A
bs
ol
ut
e
E
r
r
or
0.38
0.40
0.39
0.40
0.38
M
A
E
0.37
0.33
0.38
0.36
0.34
M
A
P
E
(
%
)
6.91
6.03
6.90
6.55
6.20
R
M
S
E
0.53
0.51
0.54
0.53
0.51
M
S
E
0.28
0.26
0.29
0.29
0.26
S
M
A
P
E
0.07
0.06
0.07
0.07
0.06
c
nS
M
A
P
E
0.96
0.97
0.96
0.97
0.97
A
c
c
ur
a
c
y
(%)
65.71
65.71
68.57
71.43
77.14
F
ig
ur
e
3
di
s
pl
a
ys
th
e
di
s
pe
r
s
io
n
e
r
r
or
s
ut
il
iz
in
g
boxplot
r
e
pr
e
s
e
nt
a
ti
on
f
or
e
a
c
h
m
e
th
od,
e
lu
c
id
a
ti
ng
th
e
ir
e
r
r
or
c
ha
r
a
c
te
r
is
ti
c
s
. T
h
e
N
N
m
e
th
od
e
xhi
bi
ts
t
he
gr
e
a
t
e
s
t
e
r
r
or
di
s
pe
r
s
io
n, t
r
a
il
e
d by R
F
, A
B
,
a
nd X
G
B
,
w
it
h
S
V
M
de
m
ons
tr
a
ti
ng
th
e
lo
w
e
s
t
e
r
r
or
di
s
pe
r
s
io
n
ba
s
e
d
on
th
e
da
ta
s
e
t.
T
he
X
G
B
m
e
th
od
h
a
s
a
m
e
di
a
n
va
lu
e
c
lo
s
e
st
to
0
c
om
p
a
r
e
d
to
th
e
ot
he
r
m
e
th
od
.
T
he
r
e
a
r
e
2
out
li
e
r
da
ta
poi
nt
s
f
or
a
ll
m
e
th
ods
f
r
om
th
e
e
a
r
th
qua
ke
pr
e
di
c
ti
on
e
va
lu
a
ti
on
ba
s
e
d
on
r
a
don
ga
s
c
onc
e
nt
r
a
ti
on
da
ta
.
T
he
y
r
e
pr
e
s
e
nt
da
ta
poi
nt
s
th
a
t
a
r
e
unus
ua
ll
y hi
gh c
om
pa
r
e
d t
o t
he
r
e
s
t
of
t
he
da
ta
s
e
t.
F
ig
ur
e
4 s
how
s
t
he
hi
s
to
gr
a
m
of
e
r
r
or
s
f
or
X
G
B
in
F
ig
ur
e
4
(
a
)
, R
F
in
F
ig
ur
e
4
(
b)
, N
N
in
F
ig
ur
e
4
(
c
)
,
A
B
in
F
ig
ur
e
4
(
d)
,
a
nd
S
V
M
in
F
ig
ur
e
4
(
e
)
a
r
e
s
c
r
ut
in
iz
e
d
m
e
ti
c
ul
ous
ly
to
e
xa
m
in
e
th
e
s
ig
n
of
de
vi
a
ti
on.
A
ll
m
a
c
hi
ne
le
a
r
ni
ng
m
e
th
ods
us
e
d
in
th
e
s
tu
dy
ha
ve
th
e
hi
ghe
s
t
f
r
e
que
nc
y
of
va
lu
e
s
a
t
0
to
-
0.25
M
e
r
r
or
.
T
he
X
G
B
ha
s
th
e
hi
ghe
s
t
e
r
r
or
f
r
e
que
nc
y
w
it
h
a
qua
nt
it
y
of
14
a
t
t
ha
t
f
r
e
que
nc
y,
f
ol
lo
w
e
d
by
N
N
a
nd
S
V
M
w
it
h
12 a
nd R
F
a
nd
A
B
m
e
th
ods
w
it
h 10.
T
he
y
pr
oduc
e
d a
hi
ghe
r
q
ua
nt
it
y of
ne
ga
ti
ve
e
r
r
or
s
t
ha
n n
e
ga
ti
ve
bi
a
s
, a
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
G
r
in
dul
u
fa
ul
t
c
lo
ud r
adon data f
or
e
ar
th
qua
k
e
m
agni
tu
de
pr
e
di
c
ti
on
…
(
T
hom
as
O
k
a P
r
at
am
a
)
4577
c
a
n
be
not
ic
e
d
in
th
e
hi
s
to
gr
a
m
s
of
F
ig
ur
e
4.
N
e
ga
ti
ve
bi
a
s
r
e
f
e
r
s
to
a
s
ys
te
m
a
ti
c
te
nd
e
nc
y
f
or
e
s
ti
m
a
te
s
to
c
ons
is
te
nt
ly
und
e
r
e
s
ti
m
a
te
th
e
tr
ue
v
a
lu
e
or
e
xhi
bi
t
a
t
e
nde
nc
y
to
w
a
r
d
lo
w
e
r
va
lu
e
s
.
I
n
s
ta
ti
s
ti
c
a
l
te
r
m
s
,
it
m
e
a
ns
t
ha
t
th
e
a
ve
r
a
g
e
of
t
he
e
s
ti
m
a
t
e
s
t
e
nds
t
o be
l
ow
e
r
t
ha
n t
he
r
e
a
l
va
lu
e
of
m
a
gni
tu
de
.
F
i
g
ur
e
3
.
B
o
x
p
l
ot
g
e
n
e
r
a
t
e
d
b
y
m
a
c
hi
n
e
l
e
a
r
n
in
g
a
l
go
r
it
h
m
s
du
r
in
g
t
h
e
e
a
r
t
h
q
u
a
k
e
p
r
e
d
i
c
t
io
n
of
th
e
3
5
d
a
t
a
t
e
s
t
s
e
t
(a)
(b)
(c)
(d)
(e)
F
ig
ur
e
4
. H
is
to
gr
a
m
s
of
t
he
e
r
r
or
s
pr
oduc
e
d by (
a
)
X
G
B
, (
b
)
R
F
, (
c
)
N
N
, (
d)
A
B
, a
nd (
e
)
S
V
M
a
lg
or
it
hm
s
w
he
n pr
e
di
c
ti
ng t
he
35 da
ta
t
e
s
t
s
e
t
M
a
c
h
i
n
e
L
ea
r
n
i
n
g
M
et
h
o
d
s
E
r
r
or
(
M
)
.
0
0
0
.
0
0
.
0
0
0
.
0
.
0
0
.
0
2
.
0
0
2
.
0
E
a
r
t
h
qu
a
ke
M
a
g
n
i
t
u
de
P
r
e
di
c
t
i
on
E
r
r
or
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
13
, N
o.
4
,
D
e
c
e
m
be
r
2024
:
4572
-
4582
4578
I
n
th
is
r
e
s
e
a
r
c
h,
w
e
a
na
ly
z
e
d
e
r
r
or
s
a
s
de
pi
c
te
d
in
F
ig
ur
e
5
(
a
)
f
or
m
e
a
ns
of
a
bs
ol
ut
e
e
r
r
or
a
nd
F
ig
ur
e
5(
b)
f
or
th
e
s
ta
nda
r
d
de
vi
a
ti
on
of
a
bs
ol
ut
e
e
r
r
or
.
T
he
r
e
s
ul
ts
r
e
ve
a
l
th
a
t
th
e
X
G
B
m
e
th
od
e
xhi
bi
ts
th
e
lo
w
e
s
t
M
A
E
f
or
th
e
M
4.8
-
M
5
e
a
r
th
qua
ke
,
r
e
gi
s
te
r
in
g
a
t
0.11
a
nd
1.58
f
or
>
M
6.5.
A
ddi
ti
ona
ll
y,
th
e
X
G
B
m
e
th
od
de
m
ons
tr
a
te
s
a
lo
w
s
ta
nda
r
d
de
vi
a
ti
on
of
a
bs
ol
ut
e
e
r
r
or
a
c
r
os
s
m
os
t
m
a
gni
tu
de
r
a
ng
e
s
,
w
it
h
th
e
lo
w
e
s
t
r
e
c
or
de
d a
t
0.07 f
or
t
he
M
4.8
-
M
5 m
a
gni
tu
de
r
a
nge
.
T
he
N
N
m
e
th
od dis
pl
a
ys
t
he
l
ow
e
s
t
M
A
E
f
or
M
4.
5
-
M
4.7
a
nd
M
5.7
-
M
5.9,
r
e
c
or
di
ng
va
lu
e
s
of
0.35.
M
e
a
nw
hi
le
,
th
e
S
V
M
m
e
th
od
s
how
c
a
s
e
s
th
e
lo
w
e
s
t
M
A
E
f
or
M
5.1
-
M
5.3
a
nd
M
5.4
-
M
5.6,
w
it
h
va
lu
e
s
of
0.13
a
nd
0.32,
r
e
s
pe
c
ti
ve
ly
.
C
onve
r
s
e
ly
,
th
e
R
F
m
e
th
od
e
xhi
bi
ts
th
e
hi
ghe
s
t
M
A
E
f
or
M
4.5
-
M
4.7
a
nd
M
4.8
-
M
5,
w
he
r
e
a
s
t
he
A
B
m
e
th
od
s
how
s
th
e
hi
ghe
s
t
M
A
E
f
or
m
a
gni
tu
de
s
a
bove
6.5.
(
a
)
(
b)
F
ig
ur
e
5.
A
na
ly
s
is
of
t
he
(
a
)
m
e
a
n a
nd
(
b)
s
ta
nda
r
d de
vi
a
ti
on of
t
he
a
bs
ol
ut
e
e
r
r
or
of
e
a
r
th
qua
ke
m
a
gni
tu
de
pr
e
di
c
ti
on
e
r
r
or
E
a
r
th
qua
ke
s
a
bove
M
5.5
a
r
e
in
f
r
e
que
nt
oc
c
ur
r
e
nc
e
s
.
I
nc
r
e
a
s
in
g
th
e
a
va
il
a
bl
e
da
ta
w
il
l
e
nha
nc
e
th
e
s
ys
te
m
'
s
le
a
r
ni
ng
c
a
pa
bi
li
ti
e
s
,
e
na
bl
in
g
m
or
e
pr
e
c
i
s
e
a
nd
a
c
c
ur
a
te
pr
e
di
c
ti
ons
of
e
a
r
th
qua
ke
m
a
gni
tu
de
s
.
C
ont
in
uous
m
e
a
s
ur
e
m
e
nt
da
ta
of
r
a
don
ga
s
c
on
c
e
nt
r
a
ti
on
c
a
n
a
s
s
i
s
t
r
e
s
e
a
r
c
he
r
s
in
de
s
ig
ni
ng
e
a
r
th
qua
ke
pr
e
di
c
ti
on
a
lg
or
it
hm
s
,
th
us
a
id
in
g
i
n
e
a
r
th
qua
ke
m
it
ig
a
ti
on
e
f
f
or
ts
,
pa
r
ti
c
ul
a
r
ly
in
I
ndone
s
ia
.
F
e
a
tu
r
e
s
s
tu
di
e
s
m
a
y
e
xpl
or
e
e
a
r
th
qua
ke
pr
e
di
c
ti
on
ba
s
e
d
on
r
a
don
ga
s
c
on
c
e
nt
r
a
ti
on
a
nd
to
ta
l
e
le
c
tr
on
c
ont
e
nt
(
T
E
C
)
m
e
a
s
ur
e
m
e
nt
s
m
e
a
s
ur
e
d a
t
th
e
e
x
a
c
t
lo
c
a
ti
on s
o
a
s
t
o de
s
ig
n a
r
e
li
a
bl
e
e
a
r
th
qua
ke
pr
e
di
c
ti
on s
y
s
te
m
.
A
uni
que
m
e
th
odol
ogy
f
or
e
a
r
th
qua
ke
m
a
gni
tu
de
pr
e
di
c
ti
on,
w
hi
c
h
us
e
s
r
a
don
ga
s
c
on
c
e
nt
r
a
ti
on
m
e
a
s
ur
e
m
e
nt
s
a
lo
ng
th
e
a
c
ti
ve
f
a
ul
t
of
G
r
in
dul
u
in
P
a
c
it
a
n,
E
a
s
t
J
a
va
,
ha
s
s
how
n
be
tt
e
r
r
e
s
ul
ts
th
a
n
e
a
r
li
e
r
s
tu
dy
a
ppr
oa
c
he
s
in
T
a
bl
e
6
(
s
e
e
a
ppe
ndi
x)
.
M
A
E
,
M
A
P
E
,
R
M
S
E
,
M
S
E
,
a
nd
S
M
A
P
E
w
e
r
e
a
ll
lo
w
e
r
th
a
n
pr
e
vi
ous
ly
r
e
por
te
d
le
ve
ls
.
O
th
e
r
s
tu
di
e
s
ha
ve
f
ound
li
m
it
s
in
e
a
r
th
qua
ke
m
a
gni
tu
de
e
s
ti
m
a
te
s
,
w
it
h
num
be
r
s
th
a
t
a
r
e
e
it
he
r
to
o
w
id
e
or
to
o
na
r
r
ow
.
L
ong
f
or
e
c
a
s
t
dur
a
ti
ons
pr
ovi
de
is
s
ue
s
,
e
it
he
r
r
e
s
ul
ti
ng
in
le
ngt
h
y
p
e
r
io
ds
of
a
le
r
t
or
in
s
uf
f
ic
ie
nt
ti
m
e
f
or
e
s
c
a
pe
.
P
r
e
vi
ous
r
e
s
e
a
r
c
h a
ls
o
l
a
c
ke
d
in
f
or
m
a
ti
on
a
bout
th
e
e
xpe
c
te
d
lo
c
a
ti
on
(
s
uc
h
a
s
w
hi
c
h
pl
a
te
or
f
a
ul
t
s
e
c
ti
on)
.
F
ur
th
e
r
m
or
e
,
I
ndone
s
ia
ha
d
f
e
w
e
a
r
th
qua
ke
m
a
gni
tu
de
f
or
e
c
a
s
ts
ba
s
e
d
on r
a
don ga
s
c
onc
e
nt
r
a
ti
ons
.
T
hi
s
r
e
s
e
a
r
c
h ha
s
a
c
nS
M
A
P
E
of
0.97 a
nd a
n a
c
c
ur
a
c
y of
77.14%
. T
he
a
c
c
ur
a
c
y
of
th
e
pr
e
di
c
ti
on
a
ppr
oa
c
h
us
e
d
he
r
e
in
out
pe
r
f
or
m
s
pr
io
r
r
e
s
e
a
r
c
h,
w
hi
c
h
f
a
il
e
d
to
m
e
e
t
a
75%
a
c
c
ur
a
c
y
ba
r
f
or
e
a
r
th
qua
ke
m
a
gni
tu
de
f
or
e
c
a
s
ts
w
hi
le
ke
e
pi
ng
pr
e
c
i
s
e
pr
e
di
c
ti
on
va
lu
e
s
a
nd
a
ppr
opr
ia
te
pr
e
di
c
ti
on
dur
a
ti
ons
.
4.
C
O
N
C
L
U
S
I
O
N
T
he
e
a
r
th
qua
ke
m
a
gni
tu
de
pr
e
di
c
ti
on w
a
s
c
onduc
te
d us
in
g r
a
d
on ga
s
c
onc
e
nt
r
a
ti
on me
a
s
ur
e
m
e
nt
s
a
t
th
e
P
a
c
it
a
n
s
ta
ti
on,
E
a
s
t
J
a
v
a
,
I
ndone
s
ia
,
s
it
ua
te
d
ne
a
r
th
e
a
c
ti
ve
G
r
in
dul
u
f
a
ul
t
ut
il
iz
e
d
f
or
m
ode
li
ng
us
in
g
R
F
,
X
G
B
,
N
N
,
A
B
,
a
nd
S
V
M
m
e
th
ods
w
it
h
va
r
io
us
f
e
a
tu
r
e
s
e
tt
in
gs
f
or
1
-
4
da
ys
e
a
r
th
qua
ke
pr
e
di
c
ti
on
be
twe
e
n
E
ur
a
s
ia
a
nd
I
ndo
-
A
us
tr
a
li
a
P
la
te
s
.
E
r
r
or
e
va
lu
a
ti
ons
of
m
a
c
hi
ne
le
a
r
ni
ng
m
e
th
ods
w
e
r
e
c
onduc
te
d,
in
c
lu
di
ng
R
e
la
ti
ve
e
r
r
or
,
M
A
E
,
M
A
P
E
,
R
M
S
E
,
M
S
E
,
S
M
A
P
E
,
a
nd
c
nS
M
A
P
E
.
X
G
B
a
nd
S
V
M
e
m
e
r
ge
d
a
s
th
e
be
s
t
-
pe
r
f
or
m
in
g
m
e
th
ods
,
pr
oduc
in
g
th
e
lo
w
e
s
t
e
r
r
or
va
lu
e
s
a
c
r
os
s
th
e
s
e
m
e
tr
ic
s
.
S
pe
c
if
ic
a
ll
y,
X
G
B
a
c
hi
e
ve
d
M
A
E
(
0.33)
, M
A
P
E
(
6.03%
)
, R
M
S
E
(
0.51)
, M
S
E
(
0.26)
, S
M
A
P
E
(
0.
06)
,
c
nM
A
P
E
(
0.97)
, a
nd a
c
c
ur
a
c
y (
65.71%
)
,
w
hi
le
S
V
M
r
e
c
or
de
d
M
A
E
(
0.34)
,
M
A
P
E
(
6.20%
)
,
R
M
S
E
(
0.51)
,
M
S
E
(
0.26)
,
S
M
A
P
E
(
0.06)
,
c
nM
A
P
E
(
0.97)
,
a
nd
a
c
c
ur
a
c
y
(
77.14%
)
.
A
uni
que
m
e
th
od
f
or
pr
e
di
c
ti
ng e
a
r
th
qua
ke
m
a
gni
tu
de
,
us
in
g
m
e
a
s
ur
e
m
e
nt
s
of
r
a
don
c
onc
e
nt
r
a
ti
ons
a
lo
ng
th
e
a
c
ti
ve
G
r
in
dul
u
f
a
ul
t
in
P
a
c
it
a
n,
E
a
s
t
J
a
va
,
pe
r
f
or
m
s
be
tt
e
r
th
a
n
pr
e
vi
ous
r
e
s
e
a
r
c
h
m
e
th
od
s
on
r
a
don.
N
e
ve
r
th
e
le
s
s
,
in
c
r
e
a
s
in
g
d
a
ta
a
v
a
il
a
bi
li
ty
c
a
n
e
nha
n
c
e
th
e
le
a
r
ni
ng
c
a
pa
bi
li
ti
e
s
of
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
G
r
in
dul
u
fa
ul
t
c
lo
ud r
adon data f
or
e
ar
th
qua
k
e
m
agni
tu
de
pr
e
di
c
ti
on
…
(
T
hom
as
O
k
a P
r
at
am
a
)
4579
pr
e
di
c
ti
ve
m
ode
ls
,
c
ont
r
ib
ut
in
g
to
m
or
e
pr
e
c
is
e
e
a
r
th
qua
ke
m
a
gni
tu
de
pr
e
di
c
ti
ons
a
nd
a
id
in
g
in
e
a
r
th
qua
ke
m
it
ig
a
ti
on e
f
f
or
ts
.
A
C
K
N
O
WL
E
D
G
E
M
E
N
T
S
T
he
a
ut
hor
th
a
nks
th
e
S
e
ns
or
a
nd T
e
le
-
C
ont
r
ol
S
ys
te
m
s
L
a
bor
a
to
r
y r
e
s
e
a
r
c
h t
e
a
m
a
t
th
e
D
e
pa
r
tm
e
nt
of
N
uc
le
a
r
E
ngi
ne
e
r
in
g
a
nd
E
ngi
ne
e
r
in
g
P
hys
ic
s
,
U
ni
v
e
r
s
it
a
s
G
a
dj
a
h
M
a
d
a
,
I
ndone
s
ia
,
f
or
f
ur
ni
s
hi
ng
th
e
r
a
don
ga
s
te
le
m
oni
to
r
in
g
s
ys
te
m
.
A
ppr
e
c
ia
ti
on
is
e
xt
e
nde
d
to
th
e
P
os
td
a
m
G
e
of
on
a
nd
th
e
B
a
da
n
M
e
te
or
ol
ogi
,
K
li
m
a
to
lo
gi
da
n
G
e
of
is
ik
a
I
ndone
s
ia
f
or
s
uppl
yi
ng
th
e
e
a
r
th
qu
a
ke
da
ta
.
W
e
a
ls
o
a
c
knowle
dge
th
e
s
uppor
t
of
th
e
R
T
A
P
r
ogr
a
m
a
t
U
ni
ve
r
s
it
a
s
G
a
dj
a
h M
a
d
a
w
it
h G
r
a
nt
N
um
be
r
5075/UN1.P.I
I
/Di
t
-
L
it
/P
T
.01.01/2023.
A
P
P
E
N
D
I
X
T
a
bl
e
6. O
th
e
r
r
e
s
e
a
r
c
h e
a
r
th
qua
ke
pr
e
di
c
ti
on r
e
s
ul
t
s
(
c
ont
in
ue
d…
)
R
e
f
.
V
a
r
i
a
bl
e
(
s
)
M
e
t
hod
L
i
m
i
t
a
t
i
on
R
e
s
ul
t
[
12]
A
ni
m
a
l
be
ha
vi
our
,
e
nvi
r
onm
e
nt
a
l
dyna
m
i
c
s
, a
nd
c
he
m
i
c
a
l
c
ha
nge
s
B
e
l
i
e
f
r
ul
e
-
ba
s
e
d e
xpe
r
t
s
ys
t
e
m
(
B
R
B
E
S
)
T
i
m
e
:
12
-
hour
t
i
m
e
f
r
a
m
e
L
oc
a
t
i
on:
-
M
a
gni
t
ude
:
M
>6.5
AUC
B
R
B
E
S
:
0.969, F
L
B
E
S
:
0.789,
A
N
N
:
0.862
[3
2
]
I
nde
pe
nde
nt
da
t
a
t
e
s
t
E
xpe
r
t
s
ys
t
e
m
T
i
m
e
:
-
L
oc
a
t
i
on:
-
M
a
gni
t
ude
:
M
0.1
–
M
5.9
A
c
c
ur
a
c
y be
l
ow
70%
[
5]
R
a
don a
nd
G
r
oundw
a
t
e
r
l
e
ve
l
S
t
a
t
i
s
t
i
c
T
i
m
e
:
1
-
4 da
ys
L
oc
a
t
i
on:
B
e
t
w
e
e
n E
ur
a
s
i
a
a
nd
I
ndo
-
A
us
t
r
a
l
i
a
P
l
a
t
e
s
M
a
gni
t
ude
:
>4.5
S
e
ns
i
t
i
vi
t
y a
nd pr
e
c
i
s
i
on >80%
(
t
i
m
e
pr
e
di
c
t
i
on)
[
13]
S
e
i
s
m
i
c
i
t
y
E
xpe
r
t
s
ys
t
e
m
T
i
m
e
:
12
-
hour
t
i
m
e
f
r
a
m
e
L
oc
a
t
i
on:
one
-
qua
r
t
e
r
of
t
he
E
a
r
t
h
M
a
gni
t
ude
:
M
3.6
-
M
9.1
A
c
c
ur
a
c
y 100%
[
14]
M
e
t
e
or
ol
ogi
c
a
l
a
nd
s
e
i
s
m
i
c
da
t
a
S
uppor
t
ve
c
t
or
r
e
gr
e
s
s
i
on
P
r
e
di
c
t
i
on num
be
r
of
e
a
r
t
hqua
ke
s
i
n
a
m
on
t
h
.
T
he
a
ve
r
a
ge
m
a
g
n
i
t
u
de
of
a
n e
a
r
t
hqua
ke
i
n a
m
ont
h
P
r
e
c
i
s
i
on 96%
pr
e
di
c
t
i
ng t
he
m
e
a
n m
a
gni
t
ude
.
A
c
c
ur
a
c
y
7
8
%
f
o
r
t
h
e
e
x
pe
c
t
e
d
e
a
r
t
h
q
ua
k
e
c
ount
i
n a
m
ont
h
[
15]
S
e
i
s
m
i
c
pa
r
a
m
e
t
e
r
s
P
a
t
t
e
r
n r
e
c
ogni
t
i
on
NN
,
r
e
c
ur
r
e
nt
ne
ur
a
l
ne
t
w
or
k,
RF
a
nd l
i
ne
a
r
pr
ogr
a
m
m
i
ng boos
t
e
ns
e
m
bl
e
T
i
m
e
:
one
m
ont
h
L
oc
a
t
i
on:
H
i
ndukus
h
a
gni
t
ude
:
≥
.
T
r
a
i
ni
ng:
A
c
c
ur
a
c
y 79%
T
e
s
t
:
A
c
c
ur
a
c
y 65%
[
16]
C
l
oud
-
ba
s
e
d bi
g
da
t
a
i
nf
r
a
s
t
r
uc
t
ur
e
R
e
gr
e
s
s
i
on a
l
gor
i
t
hm
T
i
m
e
:
7 da
ys
L
oc
a
t
i
on:
C
a
l
i
f
or
ni
a
M
a
gni
t
ude
:
M
3
-
M7
M
A
E
:
0:
59 ±
0:
66 (
M
3
-
4)
0:
25 ±
0:
52 (
M
4
-
5)
0:
27 ±
0:
60 (
M
5
-
6)
0:
28 ±
0:
75 (
M
6
-
7)
M
S
E
:
0:
79 ±
1:
53 (
M
3
-
4)
0:
34 ±
1:
42 (
M
4
-
5)
0:
43 ±
2:
06 (
M
5
-
6)
0:
63 ±
2:
73 (
M
6
-
7)
[
18]
S
e
i
s
m
i
c
da
t
a
M
ul
t
i
l
a
ye
r
pe
r
c
e
pt
r
on
NN
T
i
m
e
:
da
ys
L
oc
a
t
i
on:
-
M
a
g
ni
t
u
d
e
:
4
C
l
a
s
s
i
f
i
c
a
t
i
o
n
,
M
>
4
A
c
c
ur
a
c
y:
73.79%
[
19]
S
e
i
s
m
i
c
E
l
e
c
t
r
i
c
S
i
gna
l
s
(
S
E
S
)
ANN
T
i
m
e
:
da
ys
L
oc
a
t
i
on:
r
e
gi
on of
G
r
e
e
c
e
M
a
gni
t
ude
:
M
2.1
–
M
5.2
a
nd
M
>5.2
A
c
c
ur
a
c
y 84.01%
(
a
l
l
)
,
8.02%
,
≥
.2 on t
he
i
c
ht
e
r
s
c
a
l
e
)
.
[
20]
S
e
i
s
m
i
c
da
t
a
A
N
F
I
S
T
i
m
e
:
-
L
oc
a
t
i
on:
r
e
gi
on of
I
r
a
n
M
a
gni
t
ude
:
M
> 5.5
A
N
F
I
S
by G
r
i
d P
a
r
t
i
t
i
on
a
l
gor
i
t
hm
R
2
= 0.94, M
A
E
= 0.149, R
M
S
E
= 0.173
[
26]
S
e
i
s
m
i
c
da
t
a
P
N
N
T
i
m
e
:
m
ont
h
L
oc
a
t
i
on:
C
a
l
i
f
or
ni
a
M
a
gni
t
ude
:
7 c
l
a
s
s
i
f
i
c
a
t
i
on, M
<
4.5
–
M
> 7.5
R
S
c
or
e
:
0.62
-
0.78 (
M
4.5
–
M
6.0)
[
27]
H
i
s
t
or
i
c
a
l
s
e
i
s
m
i
c
e
ve
nt
s
NN
m
ode
l
T
i
m
e
:
30 da
ys
L
oc
a
t
i
on:
s
pe
c
i
f
i
c
l
oc
a
t
i
on or
a
r
e
a
i
n T
a
i
w
a
n
a
gni
t
ude
:
≥
6
R
S
c
or
e
i
s
0.303
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
13
, N
o.
4
,
D
e
c
e
m
be
r
2024
:
4572
-
4582
4580
T
a
bl
e
6. O
th
e
r
r
e
s
e
a
r
c
h e
a
r
th
qua
ke
pr
e
di
c
ti
on r
e
s
ul
t
s
R
E
F
E
R
E
N
C
E
S
[
1]
C
.
H
.
c
hol
z
,
L
.
.
yke
s
,
a
nd
Y
.
P
.
gga
r
w
a
l
,
“
E
a
r
t
hqua
ke
pr
e
di
c
t
i
on:
ph
ys
i
c
a
l
ba
s
i
s
,”
Sc
i
e
nc
e
,
vol
.
181,
no.
4102,
pp.
803
–
810, 1973.
[
2]
M
.
A
l
-
H
i
l
a
l
a
nd
.
out
y,
“
a
don
m
oni
t
or
i
ng f
or
e
a
r
t
hqua
ke
pr
e
di
c
t
i
on
on
l
-
ha
b
f
a
ul
t
o
f
yr
i
a
,”
N
uc
l
e
ar
G
e
ophy
s
i
c
s
,
vol
.
8,
no. 3,
pp. 291
-
299
,
1994.
[
3]
.
l
a
m
,
.
W
a
ng,
.
Z
ha
o,
a
nd
.
a
r
ka
t
,
“
I
m
pl
i
c
a
t
i
on
of
r
a
don
m
oni
t
or
i
ng
f
or
e
a
r
t
hqua
ke
s
ur
ve
i
l
l
a
nc
e
us
i
ng
s
t
a
t
i
s
t
i
c
a
l
t
e
c
hni
que
s
:
a
c
a
s
e
s
t
udy of
w
e
n
c
hua
n e
a
r
t
hqua
ke
,”
G
e
of
l
ui
ds
,
vol
. 2020, no. 1,
2020
, doi
:
10.1155/
2020/
2429165
.
[
4]
u
n
a
r
n
o
,
H
.
L
.
i
r
da
u
s
,
Y
.
.
L
uc
k
ya
r
n
o
,
.
.
W
a
r
u
w
u,
a
n
d
.
W
i
j
a
y
a
,
“
D
e
t
e
c
t
i
on
s
ys
t
e
m
f
o
r
d
e
t
e
r
m
i
n
i
s
t
i
c
e
a
r
t
hq
u
a
ke
p
r
e
d
i
c
t
i
o
n
b
a
s
e
d
o
n
r
a
d
o
n
c
o
nc
e
nt
r
a
t
i
o
n
c
ha
n
g
e
s
i
n
I
n
d
o
ne
s
i
a
,”
J
ou
r
na
l
o
f
E
ng
i
n
e
e
r
i
n
g
Sc
i
e
nc
e
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
1
5
,
n
o
.
3,
p
p
.
1
7
8
7
–
1
7
9
8,
2
0
20
.
[
5]
T
.
O
.
P
r
a
t
a
m
a
,
“
E
a
r
t
hqua
ke
e
a
r
l
y
w
a
r
ni
ng
s
ys
t
e
m
b
a
s
e
d
on
r
a
don
ga
s
c
on
c
e
nt
r
a
t
i
on
a
nd
gr
oundw
a
t
e
r
l
e
ve
l
f
l
uc
t
ua
t
i
on
a
t
Y
ogya
ka
r
t
a
R
e
gi
on
-
I
ndone
s
i
a
,”
M
.E
ng.
T
he
s
i
s
, U
ni
ve
r
s
i
t
a
s
G
a
dj
a
h M
a
da
, Y
ogya
ka
r
t
a
,
I
nd
one
s
i
a
, 2021.
[
6]
E
. D
a
z
é
, “
E
a
r
t
hqua
ke
pr
e
c
ur
s
or
s
:
r
a
don ga
s
,”
Sl
i
de
P
l
ay
e
r
. [
O
nl
i
ne
]
. A
va
i
l
a
bl
e
:
ht
t
ps
:
/
/
s
l
i
de
pl
a
ye
r
.c
om
/
s
l
i
de
/
15798487/
[
7]
.
.
a
f
t
i
,
“
T
he
r
e
l
a
t
i
on
of
c
onc
e
nt
r
a
t
i
on
r
a
don
ga
s
i
n
unde
r
gr
ound
w
a
t
e
r
w
i
t
h
t
e
c
t
oni
c
m
ove
m
e
nt
s
a
nd
S
e
i
s
m
i
c
i
t
y.
C
a
s
e
S
t
udy
:
J
or
j
a
f
k f
a
ul
t
, C
e
nt
r
a
l
I
r
a
n,”
Sc
i
e
nt
i
f
i
c
Q
uar
t
e
r
l
y
J
our
nal
of
G
e
o
s
c
i
e
nc
e
s
,
vol
. 30
, no. 117, doi
:
10.22071/
gs
j
.2018.91418.1186
.
[
8]
L
.
C
a
o
a
nd
.
a
o,
“
C
or
r
e
l
a
t
i
vi
t
y
be
t
w
e
e
n
s
hor
t
-
t
e
r
m
c
ha
nge
of
f
a
ul
t
ga
s
r
a
don
c
onc
e
nt
r
a
t
i
on
a
nd
a
i
r
t
e
m
pe
r
a
t
ur
e
&
a
t
m
os
phe
r
i
c
pr
e
s
s
ur
e
,”
A
c
t
a Se
i
s
m
ol
ogi
c
a Si
ni
c
a
, no. 4, pp. 719
–
729, 2014
, doi
:
10.3969/
j
.i
s
s
n.0253
-
3782.2014.04.017
.
[
9]
C
.
-
C
.
F
u
e
t
al
.,
“
P
r
e
s
e
i
s
m
i
c
a
nom
a
l
i
e
s
i
n
s
oi
l
-
ga
s
r
a
don
a
s
s
oc
i
a
t
e
d
w
i
t
h
20
6
6.6
e
i
nong
e
a
r
t
hqua
ke
,
out
he
r
n
T
a
i
w
a
n,
”
T
e
r
r
e
s
t
r
i
al
, A
t
m
os
phe
r
i
c
and
O
c
e
ani
c
Sc
i
e
n
c
e
s
,
vol
. 28, no. 5, pp. 787
-
798,
20
17
, doi
:
10.3319/
T
A
O
.2017.03.22.01
.
[
10]
una
r
no,
.
.
W
a
r
uw
u, a
nd
.
W
i
j
a
ya
, “
D
e
ve
l
opm
e
nt
of
t
he
r
e
a
l
t
i
m
e
t
e
l
e
m
oni
t
or
i
ng s
ys
t
e
m
f
or
e
a
r
t
hqua
ke
pr
e
di
c
t
i
on d
e
duc
e
d
f
r
om
f
l
uc
t
ua
t
i
ons
i
n
gr
oundw
a
t
e
r
l
e
ve
l
s
a
t
Y
ogya
ka
r
t
a
r
e
gi
on
-
I
ndone
s
i
a
,”
J
our
nal
of
T
he
o
r
e
t
i
c
al
and
A
ppl
i
e
d
I
nf
or
m
at
i
on
T
e
c
hnol
ogy
, vol
. 83, no. 1, pp. 95
–
99, 2016.
[
11]
d.
H
.
.
a
n
na
,
K
.
.
T
a
he
r
,
.
.
K
a
i
s
e
r
,
.
a
hm
ud,
d.
.
a
h
m
a
n
,
.
.
.
.
H
os
e
n,
a
nd
.
H
.
C
h
o,
“
p
pl
i
c
a
t
i
on
o
f
a
r
t
i
f
i
c
i
a
l
i
nt
e
l
l
i
ge
nc
e
i
n
pr
e
d
i
c
t
i
n
g e
a
r
t
hq
ua
k
e
s
:
s
t
a
t
e
-
of
-
t
h
e
-
a
r
t
a
nd
f
ut
ur
e
c
ha
l
l
e
n
ge
s
,”
I
E
E
E
A
c
c
e
s
s
,
vo
l
.
8,
p
p.
19
288
0
–
192
92
3,
20
20.
[
12]
.
.
H
os
s
a
i
n,
.
.
H
a
s
a
n,
.
uh
a
,
a
nd
K
.
nd
e
r
s
s
on,
“
be
l
i
e
f
r
ul
e
ba
s
e
d
e
xp
e
r
t
s
ys
t
e
m
t
o
pr
e
di
c
t
e
a
r
t
hqua
ke
und
e
r
unc
e
r
t
a
i
nt
y
,”
J
our
nal
of
W
i
r
e
l
e
s
s
M
obi
l
e
N
e
t
w
o
r
k
s
, U
bi
qui
t
ous
C
om
put
i
ng, and D
e
pe
ndabl
e
A
ppl
i
c
at
i
ons
, vol
. 9, no. 2, p
p.
26
-
41, 2018
.
[
13]
.
I
kr
a
m
a
nd
U
.
Q
a
m
a
r
,
“
D
e
ve
l
opi
ng
a
n
e
xp
e
r
t
s
ys
t
e
m
ba
s
e
d
on
a
s
s
o
c
i
a
t
i
on
r
ul
e
s
a
nd
pr
e
di
c
a
t
e
l
ogi
c
f
or
e
a
r
t
hqua
ke
pr
e
di
c
t
i
on,
”
K
now
l
e
dge
-
B
as
e
d Sy
s
t
e
m
s
, vol
. 75, pp. 87
–
103, F
e
b. 2015
, doi
:
10.1016/
j
.knos
ys
.2014.11.024
.
[
14]
P
.
H
a
j
i
khoda
ve
r
di
kha
n,
M
.
N
a
z
a
r
i
,
M
.
M
oh
s
e
ni
z
a
d
e
h,
S
.
S
ha
m
s
hi
r
ba
nd,
a
nd
K
.
-
W
.
C
ha
u,
“
E
a
r
t
hqua
ke
pr
e
di
c
t
i
on
w
i
t
h
m
e
t
e
or
ol
ogi
c
a
l
da
t
a
by
pa
r
t
i
c
l
e
f
i
l
t
e
r
-
ba
s
e
d
s
uppor
t
ve
c
t
or
r
e
gr
e
s
s
i
on,
”
E
ngi
ne
e
r
i
ng
A
ppl
i
c
at
i
ons
of
C
om
put
at
i
onal
F
l
ui
d
M
e
c
hani
c
s
, vol
. 12, no. 1, pp. 679
–
688, 2018
, doi
:
10.1080/
19942060.2018.1512010
.
R
e
f
V
a
r
i
a
bl
e
(
s
)
M
e
t
hod
L
i
m
i
t
a
t
i
on
R
e
s
ul
t
[
29]
S
e
i
s
m
i
c
da
t
a
C
l
a
s
s
i
f
i
e
r
s
a
nd
e
ns
e
m
bl
e
l
e
a
r
ni
ng
T
i
m
e
:
5 da
ys
L
oc
a
t
i
on:
C
hi
l
e
(
S
a
nt
i
a
go,
V
a
l
pa
r
a
í
s
o, T
a
l
c
a
a
nd P
i
c
hi
l
e
m
u)
M
a
gni
t
ude
:
M
4
–
M7
S
a
nt
i
a
go:
S
e
ns
i
t
i
vi
t
y 0.46
S
pe
c
i
f
i
c
i
t
y 0.98
P
os
i
t
i
ve
pr
e
di
c
t
i
ve
va
l
ue
0.65
F
s
c
or
e
0.75
A
uc
0.72
V
a
l
pa
r
a
i
s
o M
5
S
e
ns
i
t
i
vi
t
y 0.52
S
pe
c
i
f
i
c
i
t
y 0.89
P
os
i
t
i
ve
pr
e
di
c
t
i
ve
va
l
ue
0.36
F
s
c
or
e
0.67
A
uc
0.70
T
a
l
c
a
S
e
ns
i
t
i
vi
t
y 0.9
S
pe
c
i
f
i
c
i
t
y 1
P
os
i
t
i
ve
pr
e
di
c
t
i
ve
va
l
ue
1
F
s
c
or
e
0.98
A
uc
0.97
[
39]
S
e
i
s
m
i
c
da
t
a
S
V
M
a
nd N
a
ï
ve
B
a
ye
s
T
i
m
e
:
da
y
L
oc
a
t
i
on:
I
ndone
s
i
a
M
a
gni
t
ude
:
-
S
V
M
R
M
S
E
:
0.751008, M
A
P
E
:
0.156257, M
S
E
:
0.564013,
M
A
E
:
0.598473.
[
40]
C
l
i
m
a
t
e
da
t
a
,
s
e
i
s
m
i
c
da
t
a
L
ong s
hor
t
-
t
e
r
m
m
e
m
or
y
(
L
S
T
M
)
, bi
di
r
e
c
t
i
ona
l
l
ong
s
hor
t
-
t
e
r
m
m
e
m
or
y (
B
i
-
L
S
T
M
)
, a
nd t
r
a
ns
f
or
m
e
r
m
ode
l
s
T
i
m
e
:
da
y
L
oc
a
t
i
on:
J
a
pa
n, I
ndone
s
i
a
, a
nd
t
he
H
i
ndu
-
K
us
h K
a
r
a
kor
a
m
H
i
m
a
l
a
ya
n (
H
K
K
H
)
r
e
gi
on
M
a
gni
t
ude
:
M
3.8
–
M
5.8
I
ndone
s
i
a
M
A
E
= 0.066,
M
S
E
= 0.007,
l
og c
os
h l
os
s
= 0.039,
M
S
L
E
= 0.003.
J
e
pa
ng
M
A
E
= 0.083,
M
S
E
= 0.015,
l
og c
os
h l
os
s
= 0.054,
M
S
L
E
= 0.007.
HKKH
M
A
E
= 0.083,
M
S
E
= 0.011,
l
og c
os
h l
os
s
= 0.039,
M
S
L
E
= 0.005
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
G
r
in
dul
u
fa
ul
t
c
lo
ud r
adon data f
or
e
ar
th
qua
k
e
m
agni
tu
de
pr
e
di
c
ti
on
…
(
T
hom
as
O
k
a P
r
at
am
a
)
4581
[
15]
K
.
M
.
A
s
i
m
,
F
.
M
a
r
t
í
ne
z
-
Á
l
va
r
e
z
,
.
a
s
i
t
,
a
nd
T
.
I
qba
l
,
“
E
a
r
t
hqua
ke
m
a
gni
t
ude
pr
e
di
c
t
i
on
i
n
H
i
ndukus
h
r
e
gi
on
us
i
ng
m
a
c
hi
ne
l
e
a
r
ni
ng t
e
c
hni
que
s
,”
N
at
H
a
z
ar
d
s
, vol
. 85, no. 1, pp. 471
–
486, J
a
n. 2017.
[
16]
G
. A
.
-
C
or
t
é
s
, A
. M
.
-
E
s
t
e
ba
n, X
. S
ha
ng, a
nd F
.
M
.
-
Á
l
va
r
e
z
, “
E
a
r
t
hqua
ke
pr
e
d
i
c
t
i
on i
n C
a
l
i
f
or
ni
a
us
i
ng r
e
gr
e
s
s
i
on a
l
gor
i
t
hm
s
a
nd
c
l
oud
-
ba
s
e
d bi
g da
t
a
i
nf
r
a
s
t
r
uc
t
ur
e
,”
C
om
put
e
r
s
and G
e
o
s
c
i
e
nc
e
s
, vol
. 115, 20
17, pp. 198
–
210, 2018.
[
17]
T
.
K
ha
n,
M
.
R
a
bba
ni
,
S
.
M
.
T
.
i
ddi
que
e
,
a
nd
.
a
j
um
de
r
,
“
n
i
nnova
t
i
v
e
da
t
a
m
i
ni
ng
a
ppr
oa
c
h
f
or
de
t
e
r
m
i
ne
e
a
r
t
hqua
ke
pr
oba
bi
l
i
t
y
ba
s
e
d
on
l
i
ne
a
r
r
e
gr
e
s
s
i
on
a
l
gor
i
t
h
m
,”
i
n
2019
I
E
E
E
I
nt
e
r
na
t
i
onal
C
onf
e
r
e
nc
e
on
E
l
e
c
t
r
i
c
al
,
C
om
put
e
r
and
C
om
m
uni
c
at
i
on T
e
c
hnol
ogi
e
s
(
I
C
E
C
C
T
)
,
C
oi
m
ba
t
or
e
, I
ndi
a
, 2019, pp. 1
-
4, doi
:
10.1109/
I
C
E
C
C
T
.2019.8869286.
[
18]
J
.
a
hm
oudi
,
.
.
r
j
om
a
nd,
.
e
z
a
e
i
,
a
nd
.
H
.
oha
m
m
a
di
,
“
P
r
e
di
c
t
i
ng
t
he
E
a
r
t
hqua
ke
a
gni
t
ude
U
s
i
ng
t
he
ul
t
i
l
a
ye
r
P
e
r
c
e
pt
r
on
e
ur
a
l
e
t
w
or
k w
i
t
h T
w
o H
i
dde
n L
a
ye
r
s
,”
C
i
v
i
l
E
ngi
ne
e
r
i
ng J
ou
r
nal
, vol
. 2, no. 1, pp. 1
–
12, J
a
n. 2016.
[
19]
.
ous
t
r
a
,
.
vr
a
a
m
i
de
s
,
a
nd
C
.
C
hr
i
s
t
odoul
ou,
“
r
t
i
f
i
c
i
a
l
ne
ur
a
l
ne
t
w
or
ks
f
or
e
a
r
t
hqua
ke
pr
e
di
c
t
i
on
us
i
ng
t
i
m
e
s
e
r
i
e
s
m
a
gni
t
ude
da
t
a
or
e
i
s
m
i
c
E
l
e
c
t
r
i
c
i
gna
l
s
,”
E
x
pe
r
t
Sy
s
t
e
m
s
w
i
t
h A
ppl
i
c
at
i
ons
,
vol
. 38, no. 12, pp. 15032
–
15039, N
ov. 2011.
[
20]
.
i
r
r
a
s
hi
d,
“
E
a
r
t
hqua
ke
m
a
gni
t
ude
pr
e
di
c
t
i
on
by
a
da
pt
i
ve
ne
ur
o
-
f
uz
z
y
i
nf
e
r
e
nc
e
s
y
s
t
e
m
(
A
N
F
I
S
)
ba
s
e
d
on
f
uz
z
y
C
-
m
e
a
n
s
a
l
gor
i
t
hm
,”
N
at
ur
al
H
az
ar
d
s
, vol
. 74, pp. 1577
–
1593, D
e
c
. 2014.
[
21]
. Z
ha
ng, Z
.
hi
,
. W
a
ng,
.
Y
a
n, a
nd
Z
. Z
ha
ng, “
ppl
i
c
a
t
i
on of
t
he
e
xt
r
e
m
e
gr
a
di
e
nt
boos
t
i
ng m
e
t
hod t
o qua
nt
i
t
a
t
i
ve
l
y a
n
a
l
y
ze
t
he
m
e
c
ha
ni
s
m
of
r
a
don
a
nom
a
l
ous
c
ha
nge
i
n
a
ngl
a
z
ha
ng
hot
s
pr
i
ng
be
f
or
e
t
he
L
i
j
i
a
ng
w
7.0
e
a
r
t
hqua
ke
,”
J
our
nal
of
H
y
dr
ol
ogy
,
vol
. 612,
S
e
p.
2022
, doi
:
10.1016/
j
.j
hydr
ol
.2022.128249
.
[
22]
X
.
W
u
e
t
al
.
,
“
T
op
0
a
l
gor
i
t
hm
s
i
n
da
t
a
m
i
ni
ng,”
K
now
l
e
dge
and
I
nf
or
m
at
i
on
Sy
s
t
e
m
s
,
vol
.
14,
no.
1,
pp.
1
–
37,
J
a
n.
2008
,
doi
:
10.1007/
s
10115
-
007
-
0114
-
2
.
[
23]
W
.
s
t
ut
i
,
W
.
e
di
ono,
.
km
e
l
i
a
w
a
t
i
,
.
.
i
bi
nu,
a
nd
.
J
.
E
.
a
l
a
m
i
,
“
I
n
ve
s
t
i
ga
t
i
on
of
t
he
c
h
a
r
a
c
t
e
r
i
s
t
i
c
s
of
g
e
oe
l
e
c
t
r
i
c
f
i
e
l
d
s
i
gna
l
s
pr
i
or
t
o e
a
r
t
hqua
ke
s
us
i
ng a
d
a
pt
i
ve
S
T
F
T
t
e
c
hni
que
s
,”
N
at
ur
al
H
az
a
r
ds
and E
ar
t
h Sy
s
t
e
m
Sc
i
e
nc
e
, 2013.
[
24]
.
.
i
t
i
s
,
.
.
D
e
r
e
ndya
e
v,
a
nd
K
.
.
P
e
t
r
ov,
“
na
l
yz
i
ng
t
he
pe
r
f
o
r
m
a
nc
e
of
G
P
S
da
t
a
f
o
r
e
a
r
t
hqua
ke
pr
e
di
c
t
i
on,”
R
e
m
ot
e
Se
ns
i
ng
,
vol
. 13, no. 9,
2021
, doi
:
10.3390/
r
s
13091842
.
[
25]
K
. M
. A
s
i
m
, F
. M
.
-
Á
l
va
r
e
z
,
.
a
s
i
t
, a
nd T
.
I
qba
l
, “
E
a
r
t
hqua
ke
m
a
gni
t
ude
pr
e
di
c
t
i
on i
n hi
ndukus
h r
e
gi
on us
i
ng
m
a
c
hi
ne
l
e
a
r
ni
ng
t
e
c
hni
qu
e
s
,”
N
at
u
r
al
H
az
a
r
ds
,
vol
. 85, pp. 471
–
486,
2016
, doi
:
10.1007/
s
11069
-
016
-
2579
-
3
.
[
26]
H
.
de
l
i
a
nd
. P
a
na
kka
t
,
“
pr
oba
bi
l
i
s
t
i
c
n
e
ur
a
l
ne
t
w
or
k f
or
e
a
r
t
hqua
ke
m
a
g
ni
t
ude
pr
e
di
c
t
i
on,”
N
e
u
r
al
N
e
t
w
or
k
s
,
vol
. 22,
no. 7
,
pp. 1018
-
1024,
2009
, doi
:
10.1016/
j
.ne
une
t
.2009.05.003
.
[
27]
J.
P.
H
ua
ng,
X
.
A
.
W
a
ng,
Y
.
Z
ha
o,
C
.
X
i
n
, a
nd
H
.
X
i
a
ng
, “
L
a
r
ge
e
a
r
t
hqua
ke
m
a
gni
t
ude
pr
e
di
c
t
i
on
i
n
t
a
i
w
a
n
ba
s
e
d
on
de
e
p
l
e
a
r
ni
ng
ne
ur
a
l
ne
t
w
o
r
k,”
N
e
ur
al
N
e
t
w
or
k
W
or
l
d
,
pp. 149
–
160,
2018.
[
28]
D
.
M
a
r
c
he
t
t
i
e
t
al
.
,
“
W
or
l
dw
i
de
s
t
a
t
i
s
t
i
c
a
l
c
or
r
e
l
a
t
i
on
of
e
i
ght
ye
a
r
s
of
s
w
a
r
m
s
a
t
e
l
l
i
t
e
da
t
a
w
i
t
h
m
5.5+
e
a
r
t
hqua
ke
s
:
ne
w
hi
nt
s
a
bout
t
he
pr
e
s
e
i
s
m
i
c
phe
nom
e
n
a
f
r
om
s
pa
c
e
,”
R
e
m
ot
e
Se
n
s
i
ng
,
vol
. 14, no. 11,
2022
, doi
:
10.3390/
r
s
14112649
.
[
29]
M
. F
.
-
G
óm
e
z
,
G
. A
.
-
C
or
t
é
s
,
A
. T
r
onc
os
o,
a
nd F
.
M
.
-
Á
l
va
r
e
z
,
“
L
a
r
ge
e
a
r
t
hqu
a
ke
m
a
gni
t
ude
pr
e
di
c
t
i
on
i
n c
hi
l
e
w
i
t
h i
m
ba
l
a
nc
e
d
c
l
a
s
s
i
f
i
e
r
s
a
nd e
n
s
e
m
bl
e
l
e
a
r
ni
ng,”
A
ppl
i
e
d Sc
i
e
nc
e
s
, vol
. 7, no. 6, J
un. 2017
, doi
:
10.3390/
a
pp7060625
.
[
30]
.
i
t
i
s
a
nd
. D
e
r
e
ndya
e
v,
“
T
he
m
e
t
hod of
t
he
m
i
ni
m
um
a
r
e
a
of
a
l
a
r
m
f
or
e
a
r
t
hqua
ke
m
a
gni
t
ude
pr
e
di
c
t
i
on
,”
F
r
ont
i
e
r
s
i
n E
ar
t
h
Sc
i
e
nc
e
, vol
. 11, N
ov. 2020
, doi
:
10.3389/
f
e
a
r
t
.2020.585317
.
[
31
]
. T
e
hs
e
e
n,
.
.
a
r
ooq, a
nd
.
bi
d, “
E
a
r
t
hqua
ke
P
r
e
di
c
t
i
on
us
i
ng e
xpe
r
t
s
y
s
t
e
m
s
:
a
s
ys
t
e
m
a
t
i
c
m
a
ppi
ng s
t
udy,”
Sus
t
ai
nabi
l
i
t
y
,
vol
. 12, no. 6, M
a
r
. 2020, doi
:
10.3390/
s
u12062420.
[3
2
]
.
L
a
s
t
,
.
a
bi
now
i
t
z
,
a
nd
.
L
e
ona
r
d,
“
C
or
r
e
c
t
i
on:
P
r
e
di
c
t
i
ng
t
he
m
a
xi
m
u
m
e
a
r
t
hqua
ke
m
a
gni
t
ude
f
r
om
s
e
i
s
m
i
c
da
t
a
i
n
I
s
r
a
e
l
a
nd i
t
s
ne
i
ghbor
i
ng c
ount
r
i
e
s
,”
P
L
oS one
, 2016, doi
:
10.1371/
j
our
na
l
.pone
.0151751.
[
33
]
.
hor
ba
ni
,
.
a
r
a
r
i
,
a
nd
.
H
o
s
e
i
ni
,
“
P
r
e
s
e
nt
i
ng
a
ne
w
m
e
t
hod
t
o
i
m
pr
ove
t
he
de
t
e
c
t
i
on
of
m
i
c
r
o
-
s
e
i
s
m
i
c
e
ve
nt
s
,
”
E
nv
i
r
onm
e
nt
al
M
oni
t
or
i
ng and A
s
s
e
s
s
m
e
nt
, vol
. 190, 2018, doi
:
10.1007/
s
1066
1
-
018
-
6837
-
6.
[
34
]
.
T
a
he
r
ni
a
,
“
uz
z
y
-
l
ogi
c
t
r
e
e
a
ppr
oa
c
h
f
or
s
e
i
s
m
i
c
ha
z
a
r
d
a
n
a
l
ys
i
s
,”
I
nt
e
r
nat
i
onal
J
our
nal
of
E
ngi
ne
e
r
i
ng
and
T
e
c
hnol
ogy
,
vol
.
6, no. 3, pp. 182
-
185
, 2014
, doi
:
10.7763/
I
J
E
T
.2014.V
6.692
.
[
35
]
B
.
R
.
‐
L
e
duc
,
C
.
H
ul
be
r
t
,
.
L
ubbe
r
s
,
K
.
B
a
r
r
os
,
C
.
J
.
H
um
phr
e
ys
,
a
nd
P
.
A
.
J
ohns
on
,
“
a
c
hi
ne
l
e
a
r
ni
ng
pr
e
di
c
t
s
l
a
bor
a
t
or
y
e
a
r
t
hqua
ke
s
,”
G
e
oph
y
s
i
c
al
R
e
s
e
a
r
c
h L
e
t
t
e
r
s
,
vol
. 44, no. 18
,
pp. 9276
-
9282,
2017.
[
36
]
T
.
O
.
P
r
a
t
a
m
a
,
M
.
M
.
W
a
r
uw
u,
a
nd
.
W
i
j
a
ya
,
“
E
a
r
t
hqua
ke
da
t
e
pr
e
di
c
t
i
on
b
a
s
e
d
on
t
he
f
l
uc
t
ua
t
i
on
of
r
a
don
ga
s
c
onc
e
nt
r
a
t
i
on
ne
a
r
gr
undul
u f
a
ul
t
,”
J
ur
nal
L
i
ngk
ungan dan B
e
nc
ana G
e
ol
ogi
,
vol
. 14, no. 2
, pp. 104
-
113
.
[
37
]
H
. C
e
m
be
r
a
nd T
. E
. J
ohns
on,
I
nt
r
oduc
t
i
on t
o he
al
t
h phy
s
i
c
s
, 4t
h e
d. N
e
w
Y
or
k
:
M
c
G
r
a
w
-
H
i
l
l
M
e
di
c
a
l
, 2009.
[
38
]
T
.
O
.
P
r
a
t
a
m
a
,
.
una
r
no,
.
.
W
i
j
a
t
na
,
a
nd
E
.
H
a
r
yono,
“
E
a
r
t
hqua
ke
m
a
gni
t
ude
pr
e
di
c
t
i
on
i
n
I
ndone
s
i
a
us
i
ng
a
s
upe
r
vi
s
e
d
m
e
t
hod
ba
s
e
d
on
c
l
oud
r
a
don
da
t
a
,”
I
nt
e
r
nat
i
onal
J
our
nal
of
R
e
c
onf
i
gur
abl
e
and
E
m
be
dde
d
Sy
s
t
e
m
s
(
I
J
R
E
S)
,
vo
l
.
13,
no.
3,
p
p.
577
-
585
, N
ov. 2024
, doi
:
10.11591/
i
j
r
e
s
.v13.i
3.pp577
-
585
.
[
39]
I
.
.
ur
w
a
nt
a
r
a
, P
. Y
ugopus
pi
t
o, a
nd
. H
e
r
m
a
w
a
n, “
C
om
pa
r
i
s
on of
m
a
c
hi
ne
l
e
a
r
ni
ng pe
r
f
or
m
a
nc
e
f
or
e
a
r
t
hqua
ke
pr
e
di
c
t
i
on i
n
I
ndone
s
i
a
us
i
ng
30
ye
a
r
s
hi
s
t
or
i
c
a
l
da
t
a
,”
T
E
L
K
O
M
N
I
K
A
,
vol
.
18,
no.
3,
p
p
.
1331
-
1342
,
J
un.
2020
,
doi
:
10.12928/
t
e
l
kom
ni
ka
.v18i
3.14756
.
[
40]
.
a
dhukha
n,
.
C
ha
kr
a
bor
t
y,
.
ukh
e
r
j
e
e
,
a
nd
.
K
.
a
m
a
nt
a
,
“
C
l
i
m
a
t
i
c
a
nd
s
e
i
s
m
i
c
da
t
a
-
dr
i
ve
n
de
e
p
l
e
a
r
ni
ng
m
ode
l
f
or
e
a
r
t
hqua
ke
m
a
gni
t
ude
pr
e
di
c
t
i
on,”
F
r
ont
i
e
r
s
i
n E
ar
t
h Sc
i
e
n
c
e
, vol
. 11, F
e
b. 20
23
, doi
:
10.3389/
f
e
a
r
t
.2023.1082832
.
B
I
O
G
R
A
P
H
I
E
S
O
F
A
U
T
H
O
R
S
Thomas
Oka
Pratama
earned
his
Doctorate
in
Environmental
Science
from
Universitas
Gadjah
Mada
in
2021
and
holds
previous
degrees,
i
ncluding
a
Bachelor
of
Engineering
in
Engineering
Physics
and
a
Master
of
Engineering
in
the
same
field,
both
obtained
from
Universitas
Gadjah
Mada
in
2014
and
2021,
respectively.
Presently,
he
is
lecture
r
in
the
Engineering
Physics
Program
at
the
Department
of
Nuclear
Engineering
and
Engineering
Physics
at
Univer
sitas
Gadja
h
Mada,
Indone
sia.
His
resea
rch
intere
st
s
encomp
ass
developing
telemonitoring
systems
and
earthqua
ke
prediction
methods
utilizing
radon
gas
concentr
ation
and groun
dwater level
s. He can b
e
contacted
at
email
:
thomas.o.p@ugm.ac.id.
Evaluation Warning : The document was created with Spire.PDF for Python.