I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
p
ute
r
E
ng
in
ee
ring
(
I
J
E
CE
)
Vo
l.
9
,
No
.
2
,
A
p
r
il
201
9
,
p
p
.
1304
~
1
3
1
2
I
SS
N:
2088
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v
9
i
2
.
pp
1
3
0
4
-
1312
1304
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
co
r
e
.
co
m/
jo
u
r
n
a
ls
/in
d
ex
.
p
h
p
/
I
JE
C
E
Ea
rthqua
k
e t
ren
d predic
tion
usin
g
lo
ng
sho
rt
-
ter
m
m
e
m
o
ry
R
NN
Va
rda
a
n K
.
1
,
T
a
nv
i B
ha
nd
a
rk
a
r
2
,
Ni
kh
il Sa
t
is
h
3
,
S.
Srid
ha
r
4
,
R.
Siv
a
k
u
m
a
r
5
,
S
neha
s
is
h
G
ho
s
h
6
1,
2,
3,
4
De
p
a
rtm
e
n
t
o
f
Co
m
p
u
ter S
c
ien
c
e
a
n
d
E
n
g
in
e
e
rin
g
,
S
RM
In
stit
u
te o
f
S
c
ien
c
e
a
n
d
T
e
c
h
n
o
lo
g
y
,
In
d
ia
5,
6
Earth
q
u
a
k
e
Re
se
a
rc
h
Ce
ll
,
De
p
a
rtme
n
t
o
f
Civ
il
En
g
in
e
e
rin
g
,
S
R
M
In
stit
u
te o
f
S
c
ien
c
e
a
n
d
T
e
c
h
n
o
lo
g
y
,
In
d
ia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Sep
2
5
,
2
0
1
7
R
ev
i
s
ed
Oct
1
5
,
2
0
1
8
A
cc
ep
ted
No
v
3
0
,
2
0
1
8
T
h
e
p
re
d
ictio
n
o
f
a
n
a
tu
ra
l
c
a
la
m
it
y
su
c
h
a
s
e
a
rth
q
u
a
k
e
s
h
a
s
b
e
e
n
a
n
a
re
a
o
f
in
tere
st
f
o
r
a
lo
n
g
ti
m
e
b
u
t
a
c
c
u
ra
te
re
su
lt
s
in
e
a
rth
q
u
a
k
e
f
o
re
c
a
stin
g
h
a
v
e
e
v
a
d
e
d
sc
ien
ti
sts,
e
v
e
n
lea
d
in
g
s
o
m
e
to
d
e
e
m
it
in
tri
n
sic
a
ll
y
i
m
p
o
ss
ib
le
to
f
o
re
c
a
st
th
e
m
a
c
c
u
ra
tel
y
.
In
th
is
p
a
p
e
r
a
n
a
tt
e
m
p
t
to
f
o
re
c
a
st
e
a
rth
q
u
a
k
e
s
a
n
d
tren
d
s
u
sin
g
a
d
a
ta
o
f
a
s
e
ries
o
f
p
a
st
e
a
rth
q
u
a
k
e
s.
A
t
y
p
e
o
f
re
c
u
r
re
n
t
n
e
u
ra
l
n
e
tw
o
rk
c
a
ll
e
d
L
o
n
g
S
h
o
r
t
-
T
e
rm
M
e
m
o
r
y
(
L
S
T
M
)
is
u
se
d
t
o
m
o
d
e
l
th
e
se
q
u
e
n
c
e
o
f
e
a
rth
q
u
a
k
e
s.
T
h
e
trai
n
e
d
m
o
d
e
l
is
th
e
n
u
se
d
t
o
p
re
d
ict
th
e
f
u
tu
re
tren
d
o
f
e
a
rth
q
u
a
k
e
s.
A
n
o
rd
in
a
ry
F
e
e
d
F
o
rw
a
rd
Ne
u
ra
l
Ne
tw
o
r
k
(F
F
NN
)
so
lu
ti
o
n
f
o
r
t
h
e
sa
m
e
p
ro
b
lem
wa
s
d
o
n
e
f
o
r
c
o
m
p
a
riso
n
.
T
h
e
L
S
T
M
n
e
u
ra
l
n
e
tw
o
rk
w
a
s
f
o
u
n
d
to
o
u
tp
e
rf
o
rm
th
e
F
F
NN
.
T
h
e
R^2
sc
o
re
o
f
t
h
e
L
S
T
M
is
b
e
tt
e
r
th
a
n
th
e
F
F
NN
’s b
y
5
9
%
.
K
ey
w
o
r
d
s
:
A
r
ti
f
icial
n
e
u
r
al
n
et
w
o
r
k
E
ar
th
q
u
ak
e
Feed
f
o
r
w
ar
d
n
e
u
r
al
n
et
w
o
r
k
L
o
n
g
s
h
o
r
t
-
ter
m
m
e
m
o
r
y
R
ec
u
r
r
en
t n
e
u
r
al
n
et
w
o
r
k
Co
p
y
rig
h
t
©
2
0
1
9
In
stit
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
Al
l
rig
h
ts
re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Var
d
aa
n
Kis
h
o
r
e,
Dep
ar
t
m
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
E
n
g
i
n
ee
r
in
g
,
SR
M
I
n
s
tit
u
te
o
f
Sc
ien
ce
a
n
d
T
ec
h
n
o
lo
g
y
,
4
0
4
6
,
estan
cia
to
w
er
-
4
,
Gu
d
v
a
n
ch
er
r
y
,
C
h
e
n
n
ai,
P
in
-
603203
,
I
n
d
ia
.
E
m
ail:
v
ar
d
aa
n
_
k
i
s
h
o
r
e
@
s
r
m
u
n
i
v
.
ed
u
.
i
n
1.
I
NT
RO
D
UCT
I
O
N
E
ar
th
q
u
ak
e
s
ar
e
a
n
at
u
r
al
h
az
a
r
d
w
h
ic
h
ca
n
ca
u
s
e
a
lo
t
o
f
d
am
ag
e
a
n
d
lo
s
s
to
h
u
m
a
n
li
f
e.
I
t
d
o
es
n
o
t
f
o
llo
w
a
n
y
s
et
o
f
p
atter
n
s
i
n
o
cc
u
r
en
ce
an
d
th
u
s
p
r
ed
ictin
g
th
e
tr
en
d
h
as
al
w
a
y
s
b
ee
n
an
i
m
p
o
r
tan
t
ar
ea
o
f
r
esear
ch
.
A
f
a
u
lt
i
n
g
eo
lo
g
y
is
a
f
r
ac
tu
r
e
in
r
o
ck
ac
r
o
s
s
w
h
ic
h
t
h
er
e
h
as
b
ee
n
s
i
g
n
i
f
ican
t
r
o
ck
m
a
s
s
m
o
v
e
m
e
n
t.
Fa
u
lt
s
ar
e
cr
ea
te
d
b
y
t
h
e
ac
tio
n
o
f
p
late
tect
o
n
ic
f
o
r
ce
s
.
T
h
e
en
er
g
y
r
ele
ase
ca
u
s
ed
b
y
r
ap
id
m
o
v
e
m
e
n
t
o
f
p
late
s
ac
r
o
s
s
c
er
tain
f
a
u
lt
s
ca
n
b
e
ca
lled
ac
tiv
e
f
au
lts
a
n
d
is
t
h
e
m
o
s
t
c
o
m
m
o
n
ca
u
s
e
o
f
ea
r
th
q
u
a
k
es.
T
h
is
e
n
er
g
y
tr
av
els
to
th
e
s
u
r
f
ac
e
o
f
th
e
E
ar
t
h
as
w
av
e
s
.
T
h
er
e
ar
e
th
r
ee
k
i
n
d
s
o
f
s
e
is
m
ic
w
a
v
es.
S
an
d
P
w
a
v
es
p
en
etr
ate
t
h
e
in
ter
io
r
o
f
th
e
ea
r
t
h
an
d
ar
e
h
en
ce
ca
lled
b
o
d
y
w
av
e
s
[
1
]
.
T
h
e
th
ir
d
k
in
d
o
f
w
a
v
es
a
n
d
t
h
e
m
o
s
t
d
estr
u
ctiv
e
ar
e
s
u
r
f
ac
e
w
av
e
s
,
w
h
ich
ar
e
s
i
m
ilar
to
w
ater
an
d
tr
a
v
el
a
cr
o
s
s
th
e
s
u
r
f
ac
e
o
f
th
e
ea
r
th
.
Du
e
to
its
d
estru
c
tiv
e
p
o
ten
tial,
h
u
m
a
n
k
i
n
d
h
as
lo
n
g
b
e
en
s
ea
r
ch
i
n
g
f
o
r
an
ea
r
th
q
u
ak
e
tr
e
n
d
p
r
ed
ictio
n
m
et
h
o
d
.
P
r
ed
ictin
g
an
ea
r
th
q
u
ak
e
i
m
p
lie
s
s
tati
n
g
t
h
e
ex
a
ct
ti
m
e,
m
a
g
n
it
u
d
e
an
d
lo
ca
tio
n
o
f
a
co
m
in
g
ea
r
t
h
q
u
a
k
e.
P
r
ed
ictio
n
m
o
d
el
s
co
m
e
u
n
d
er
eith
er
s
h
o
r
t
-
ter
m
p
r
ed
ictio
n
(
<
1
-
y
ea
r
ti
m
e
s
ca
le)
,
lo
n
g
ter
m
p
r
ed
ictio
n
(
1
0
to
1
0
0
y
ea
r
s
ti
m
e
s
ca
le)
o
r
i
n
ter
m
ed
iate
t
er
m
p
r
ed
ictio
n
(
1
to
1
0
y
ea
r
s
t
i
m
e
s
ca
le)
[2
]
,
[
3
]
.
G
r
ea
t
ef
f
o
r
t
h
as
b
ee
n
m
ad
e
b
y
t
h
e
s
cien
ti
f
ic
co
m
m
u
n
it
y
b
u
t
d
u
e
to
th
e
in
tr
in
s
ic
r
an
d
o
m
n
a
tu
r
e
o
f
th
e
p
h
en
o
m
e
n
o
n
i
ts
el
f
,
n
o
v
alid
a
n
d
r
eliab
le
m
e
th
o
d
h
as
y
et
b
ee
n
f
o
u
n
d
.
Nev
er
t
h
eles
s
,
ea
r
th
q
u
ak
es
g
en
er
atio
n
is
n
o
t
a
cy
cl
ical
p
r
o
ce
s
s
d
u
e
to
th
e
v
ar
iatio
n
o
f
r
u
p
tu
r
e
ar
ea
an
d
ea
r
th
q
u
ak
e
-
m
ed
iated
in
ter
ac
tio
n
s
alo
n
g
o
th
er
f
au
lts
.
T
h
is
m
ea
n
s
t
h
at
t
h
e
ti
m
e
b
et
w
ee
n
ev
e
n
ts
ca
n
b
e
e
x
tr
e
m
el
y
ir
r
eg
u
lar
.
C
o
n
s
eq
u
e
n
t
l
y
,
t
h
e
p
r
ed
ictio
n
o
f
th
e
ti
m
e,
o
r
a
r
elativ
el
y
clo
s
e
t
i
m
e
in
ter
v
al,
o
f
an
o
n
co
m
in
g
l
ar
g
e
ea
r
th
q
u
a
k
e
is
s
till
a
d
if
f
ic
u
lt
tas
k
.
A
l
th
o
u
g
h
co
n
s
id
er
ab
le
r
esear
ch
is
d
e
v
o
ted
to
th
e
s
cie
n
ce
o
f
s
h
o
r
t
-
ter
m
ea
r
t
h
q
u
a
k
e
f
o
r
ec
asti
n
g
,
s
ta
n
d
ar
d
is
atio
n
o
f
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8708
E
a
r
th
q
u
a
ke
tr
en
d
p
r
ed
ictio
n
u
s
in
g
lo
n
g
s
h
o
r
t
-
term me
mo
r
y
R
N
N
(
Var
d
aa
n
Ki
s
h
o
r
e
)
1305
o
p
er
atio
n
al
p
r
o
ce
d
u
r
es
is
in
a
n
ascen
t
s
tag
e
o
f
d
ev
elo
p
m
en
t.
T
h
e
p
r
o
b
lem
is
c
h
alle
n
g
in
g
b
ec
au
s
e
lar
g
e
ea
r
th
q
u
a
k
es c
an
n
o
t b
e
r
eliab
l
y
p
r
ed
icted
f
o
r
s
p
ec
if
ic
r
eg
io
n
s
o
v
er
ti
m
e
s
ca
le
s
t
h
at
s
p
an
le
s
s
th
an
d
ec
ad
es.
T
h
er
e
ar
e
t
w
o
g
e
n
er
al
ap
p
r
o
ac
h
es
to
p
r
ed
ict
ea
r
th
q
u
ak
es,
p
r
ec
u
r
s
o
r
s
b
ased
an
d
tr
en
d
b
ased
.
P
r
ec
u
r
s
o
r
s
ar
e
an
o
m
alo
u
s
p
h
en
o
m
e
n
a
t
h
at
m
ig
h
t
s
i
g
n
a
l
a
n
i
m
p
en
d
i
n
g
ea
r
t
h
q
u
a
k
e
[
4
]
s
u
c
h
a
s
r
ad
o
n
g
as
e
m
is
s
io
n
s
,
u
n
u
s
u
al
a
n
i
m
al
b
eh
av
io
u
r
,
e
lectr
o
m
a
g
n
et
ic
an
o
m
alie
s
etc.
T
r
en
d
b
ased
m
et
h
o
d
s
i
n
v
o
lv
e
id
en
ti
f
y
i
n
g
p
atter
n
s
o
f
s
eis
m
icit
y
t
h
at
p
r
ec
ed
e
an
ea
r
t
h
q
u
ak
e.
I
n
t
h
is
p
ap
er
,
a
tr
en
d
s
-
b
ased
ap
p
r
o
ac
h
is
ad
o
p
ted
an
d
th
e
L
ST
M
n
eu
r
al
n
et
w
o
r
k
i
s
u
s
ed
to
ca
p
tu
r
e
th
e
tr
en
d
in
v
o
lv
i
n
g
s
tati
s
tical
tec
h
n
iq
u
e
s
.
T
h
e
r
elatio
n
s
h
ip
b
et
w
ee
n
th
e
m
ax
i
m
u
m
o
f
ea
r
th
q
u
ak
e
af
f
e
ctin
g
co
ef
f
icie
n
t
a
n
d
s
ite
a
n
d
b
ase
m
e
n
t
co
n
d
itio
n
w
as
s
tu
d
ied
,
also
p
r
o
p
o
s
ed
a
m
o
d
el
b
ased
o
n
ea
r
th
q
u
a
k
e
m
a
g
n
it
u
d
e
p
r
ed
ictio
n
u
s
in
g
ar
t
i
f
icial
n
eu
r
al
n
et
w
o
r
k
i
n
th
e
n
o
r
th
er
n
r
ea
d
s
ea
ar
ea
[
5
]
.
A
m
u
lti
la
y
e
r
u
s
in
g
co
m
p
r
es
s
io
n
d
ata
f
o
r
p
r
ec
u
r
s
o
r
d
etec
tio
n
in
elec
tr
o
m
a
g
n
e
tic
w
av
e
o
b
s
e
r
v
atio
n
w
as
p
r
o
p
o
s
ed
[
6
]
.
A
ti
m
e
s
er
ies
ap
p
r
o
ac
h
co
m
p
o
s
ed
o
f
s
eis
m
ic
e
v
en
ts
o
cc
u
r
r
ed
in
Gr
ee
ce
w
as
ap
p
lied
[
7
]
.
A
s
t
u
d
y
b
et
w
ee
n
r
ad
o
n
an
d
ea
r
th
q
u
ak
e
u
s
in
g
an
ar
ti
f
icial
n
e
u
r
al
n
et
w
o
r
k
s
m
o
d
el
w
a
s
d
o
n
e
[
8
]
.
A
r
elatio
n
s
h
ip
b
et
w
ee
n
r
ad
o
n
co
n
ce
n
tr
atio
n
an
d
e
n
v
ir
o
n
m
e
n
tal
p
ar
a
m
eter
s
f
o
r
ea
r
th
q
u
a
k
e
p
r
ed
ictio
n
w
as
m
o
d
elled
u
s
in
g
a
n
A
NN
in
t
h
e
r
eg
io
n
o
f
T
h
ailan
d
[
9
]
.
A
n
eu
r
al
n
et
w
o
r
k
f
o
r
class
i
f
icatio
n
af
ter
a
n
al
y
s
i
n
g
th
e
elec
tr
ic
f
ield
d
ata
an
d
th
e
s
eis
m
icit
y
co
llected
f
r
o
m
d
i
f
f
er
en
t
s
tatio
n
s
w
as
s
tu
d
ied
a
n
d
r
es
u
lt
s
w
er
e
p
r
ett
y
ac
c
u
r
ate
[
1
0
]
.
I
n
v
e
s
ti
g
ated
t
h
e
s
eis
m
ic
d
a
m
a
g
e
id
e
n
ti
f
icati
o
n
b
y
u
s
in
g
a
P
C
A
-
co
m
p
r
es
s
ed
r
esp
o
n
s
e
f
u
n
c
tio
n
an
d
ar
ti
f
icia
l
n
e
u
r
al
n
et
w
o
r
k
s
[
1
1
]
.
P
r
ed
ictio
n
o
f
ea
r
th
q
u
ak
e
d
a
m
ag
e
s
a
n
d
r
eliab
ilit
y
a
n
al
y
s
is
u
s
i
n
g
f
u
zz
y
s
et
s
[
1
2
]
.
T
h
e
v
ar
iatio
n
o
f
T
o
tal
E
lectr
o
n
C
o
n
te
n
t
(
T
E
C
)
as
an
an
o
m
al
y
as
a
n
in
d
icatio
n
o
f
ea
r
th
q
u
a
k
e
a
f
e
w
d
ay
s
o
r
h
o
u
r
s
b
ef
o
r
e
it,
th
is
wa
s
u
s
ed
b
y
t
h
e
m
to
b
u
ild
a
m
o
d
el
[
1
3
]
.
R
ec
u
r
s
iv
e
s
a
m
p
le
-
en
tr
o
p
y
tec
h
n
iq
u
e
f
o
r
ea
r
th
q
u
a
k
e
f
o
r
ec
asti
n
g
,
w
h
er
e
th
e
ea
r
th
d
ata
b
ased
o
n
V
A
N
m
et
h
o
d
w
as
u
s
ed
f
o
r
th
e
m
o
d
elli
n
g
[
1
4
]
.
Mo
d
els
b
ased
o
n
m
ea
s
u
r
e
m
e
n
t
o
f
elast
ic
an
d
elec
tr
o
m
a
g
n
etic
w
a
v
es
to
p
r
ed
ict
ea
r
th
q
u
a
k
es
an
d
ts
u
n
a
m
i
w
a
s
d
o
n
e
[
1
5
]
.
E
ar
th
q
u
ak
e
h
az
ar
d
ass
ess
m
e
n
t
w
as
d
o
n
e
u
s
i
n
g
E
aHa
AsT
o
to
o
l
f
o
r
v
is
u
aliza
t
io
n
[
1
6
]
.
Dete
r
m
i
n
ed
th
e
th
r
es
h
o
ld
en
er
g
y
lead
i
n
g
t
o
s
eis
m
ic
ac
t
iv
i
t
y
[
1
7
]
.
2.
RE
S
E
ARCH
M
E
T
H
O
D
An
ar
ti
f
icial
n
e
u
r
al
n
et
w
o
r
k
i
s
a
m
at
h
e
m
atica
l
m
o
d
el
t
h
at
m
i
m
ics
t
h
e
b
io
lo
g
ica
l
n
e
u
r
o
n
s
in
b
r
ai
n
.
A
n
e
u
r
al
n
et
w
o
r
k
is
a
s
et
o
f
i
n
p
u
t,
o
u
tp
u
t
an
d
h
id
d
en
la
y
er
s
.
T
h
ese
la
y
er
s
h
av
e
n
o
d
es
w
h
ich
ar
e
in
ter
co
n
n
ec
ted
th
r
o
u
g
h
li
n
k
s
.
T
h
ese
lin
k
s
h
a
v
e
s
o
m
e
as
s
o
ci
ated
n
u
m
er
ic
w
e
ig
h
t
w
h
ich
d
e
ter
m
i
n
es
h
o
w
m
u
c
h
th
e
in
p
u
t
co
n
tr
ib
u
tes
to
an
d
a
f
f
ec
ts
t
h
e
r
esu
lts
.
T
h
e
w
e
ig
h
t
s
an
d
ac
tiv
a
tio
n
f
u
n
ctio
n
s
ca
n
b
e
m
o
d
if
ied
b
y
a
p
r
o
ce
s
s
ca
lled
lear
n
in
g
w
h
ic
h
is
g
o
v
er
n
ed
b
y
a
lear
n
i
n
g
r
u
le
[
1
8
]
.
I
n
th
is
p
ap
er
w
e
h
av
e
co
m
p
ar
ed
t
h
e
s
tr
u
ct
u
r
es
o
f
Feed
Fo
r
w
ar
d
N
eu
r
al
Net
w
o
r
k
(
F
FNN)
an
d
R
ec
u
r
r
en
t
Ne
u
r
al
Net
w
o
r
k
(
R
NN)
o
n
ti
m
e
-
s
er
ies
b
ased
d
ata.
2
.
1
.
F
ee
d
f
o
rwa
rd
neura
l net
w
o
r
k
Feed
f
o
r
w
ar
d
n
et
w
o
r
k
s
ar
e
ac
y
clic
n
et
w
o
r
k
u
s
u
all
y
ar
r
a
n
g
e
d
in
lay
er
s
,
w
h
er
e
ea
ch
n
e
u
r
o
n
r
ec
eiv
es
in
p
u
t
s
o
n
l
y
f
r
o
m
th
e
i
m
m
ed
i
atel
y
p
r
ec
ed
in
g
la
y
er
.
T
h
e
ar
ch
itect
u
r
e
o
f
an
FF
NN
w
i
th
2
h
id
d
en
la
y
er
s
i
s
s
h
o
w
n
in
F
ig
u
r
e
1
.
Fig
u
r
e
1
.
Feed
Fo
r
w
ar
d
Neu
r
al
Net
w
o
r
k
w
it
h
in
p
u
t la
y
er
,
o
u
tp
u
t la
y
er
a
n
d
2
h
id
d
en
la
y
er
s
FF
NN
s
ar
e
s
till
s
u
cc
ess
f
u
ll
y
ap
p
lied
to
m
a
n
y
p
r
o
b
le
m
s
b
u
t
s
till
ca
n
n
o
t
ca
p
tu
r
e
l
o
n
g
ter
m
d
ep
en
d
en
cies.
Ma
n
y
m
o
d
e
ls
h
av
e
i
m
p
l
icitl
y
ca
p
tu
r
ed
ti
m
e
b
y
co
n
ca
te
n
ati
n
g
ea
ch
i
n
p
u
t
w
i
th
s
o
m
e
n
u
m
b
er
o
f
its
i
m
m
ed
iate
p
r
ed
ec
ess
o
r
s
an
d
s
u
cc
es
s
o
r
s
,
p
r
esen
ti
n
g
th
e
m
ac
h
in
e
lear
n
i
n
g
m
o
d
el
w
it
h
a
s
lid
in
g
w
i
n
d
o
w
o
f
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
9
,
No
.
2
,
A
p
r
il
201
9
:
1
3
0
4
-
1312
1306
co
n
tex
t
ab
o
u
t
ea
c
h
p
o
in
t
o
f
in
ter
es
t
[
1
9
]
.
T
h
is
ap
p
r
o
ac
h
i
s
f
o
llo
w
ed
b
y
co
n
ca
te
n
ati
n
g
a
f
ix
ed
n
u
m
b
er
(
f
o
u
r
teen
)
o
f
p
as
t
ea
r
th
q
u
a
k
e
d
ata
an
d
g
iv
i
n
g
t
h
e
m
a
s
in
p
u
t
to
th
e
FF
NM
w
i
th
t
h
e
n
e
x
t
ea
r
th
q
u
a
k
e
as
t
h
e
tar
g
et.
T
h
e
m
o
d
el
u
s
es
t
w
o
h
i
d
d
en
la
y
er
s
w
it
h
2
0
n
o
d
es
a
n
d
6
0
n
o
d
es
r
esp
ec
ti
v
el
y
.
A
ll
t
h
e
n
o
d
es
h
av
i
n
g
t
h
e
s
ig
m
o
id
f
u
n
ct
io
n
a
s
t
h
eir
ac
t
i
v
atio
n
.
T
h
e
lear
n
i
n
g
r
u
le
u
s
ed
it
t
h
e
’
r
m
s
p
r
o
p
’
.
T
h
is
m
o
d
el
is
tr
ai
n
ed
f
o
r
1
0
0
0
ep
o
ch
s
o
n
t
h
e
d
ataset.
T
h
ese
a
ttrib
u
tes
w
er
e
s
elec
ted
af
ter
e
m
p
lo
y
i
n
g
a
g
r
id
s
ea
r
c
h
m
eth
o
d
w
h
ich
s
elec
ted
t
h
e
b
est ar
ch
itect
u
r
e
b
ased
o
n
th
e
er
r
o
r
r
ate.
2
.
2
.
L
o
ng
s
ho
rt
-
t
er
m
m
e
m
o
ry
L
o
n
g
S
h
o
r
t
-
T
er
m
Me
m
o
r
y
(
L
ST
M)
is
a
R
ec
u
r
r
en
t
Neu
r
al
Net
w
o
r
k
(
R
NN)
ar
ch
itect
u
r
e
(
an
ar
tif
icia
l
n
eu
r
al
n
et
w
o
r
k
)
p
r
o
p
o
s
ed
b
y
Sep
p
Ho
ch
r
eiter
an
d
J
ü
r
g
e
n
S
ch
m
id
h
u
b
er
in
1
9
9
7
[
2
0
]
.
R
NNs
ca
n
ca
p
tu
r
e
t
h
e
d
y
n
a
m
ics
o
f
s
eq
u
en
ce
s
v
ia
c
y
cles
in
th
e
n
e
t
w
o
r
k
.
B
u
t
s
o
m
e
R
NNs
s
u
f
f
er
f
r
o
m
t
h
e
v
an
is
h
in
g
a
n
d
ex
p
lo
d
in
g
g
r
ad
ien
ts
p
r
o
b
le
m
in
w
h
ic
h
g
r
ad
ien
ts
ar
e
eith
er
s
q
u
as
h
ed
to
ze
r
o
o
r
in
cr
ea
s
e
w
it
h
o
u
t
b
o
u
n
d
d
u
r
in
g
b
ac
k
p
r
o
p
ag
atio
n
th
r
o
u
g
h
a
lar
g
e
n
u
m
b
er
o
f
ti
m
e
s
tep
s
.
L
ST
M
is
in
tr
o
d
u
ce
d
p
r
i
m
ar
il
y
to
o
v
e
r
c
o
m
e
th
e
p
r
o
b
lem
o
f
v
an
i
s
h
in
g
g
r
ad
ien
t
s
.
I
t
h
as
ch
a
in
li
k
e
s
tr
u
c
tu
r
e,
h
av
in
g
t
h
r
ee
o
r
f
o
u
r
n
eu
r
al
n
et
w
o
r
k
la
y
er
o
r
“g
ates”
w
h
ich
ar
e
i
m
p
le
m
en
ted
u
s
i
n
g
lo
g
is
tic
f
u
n
ctio
n
.
T
h
e
in
f
o
r
m
a
tio
n
g
i
v
en
in
[
1
9
]
d
ep
icts
ab
o
u
t th
e
f
o
r
w
ar
d
p
ass
an
d
b
ac
k
w
ar
d
p
ass
i
n
L
S
T
Ms
.
I
n
ter
m
s
o
f
th
e
f
o
r
w
ar
d
p
ass
,
th
e
L
ST
M
ca
n
lear
n
w
h
en
to
let
ac
ti
v
atio
n
in
to
t
h
e
in
ter
n
al
s
ta
te.
As
lo
n
g
as
th
e
i
n
p
u
t
g
ate
ta
k
es
v
al
u
e
ze
r
o
,
n
o
ac
tiv
atio
n
ca
n
g
et
i
n
.
Si
m
ilar
l
y
,
t
h
e
o
u
tp
u
t
g
ate
lear
n
s
w
h
e
n
t
o
let
th
e
v
a
lu
e
o
u
t.
W
h
en
b
o
th
g
ate
s
ar
e
clo
s
ed
,
t
h
e
ac
ti
v
atio
n
is
tr
ap
p
ed
in
m
e
m
o
r
y
ce
ll,
n
eit
h
er
g
r
o
w
in
g
n
o
r
s
h
r
i
n
k
i
n
g
.
I
n
ter
m
s
o
f
b
ac
k
w
ar
d
s
p
ass
,
th
e
co
n
s
ta
n
t
er
r
o
r
ca
r
o
u
s
el
en
ab
les
th
e
g
r
ad
ien
t
to
p
r
o
p
ag
ate
b
ac
k
ac
r
o
s
s
m
a
n
y
ti
m
e
s
tep
s
,
n
eit
h
er
ex
p
lo
d
in
g
n
o
r
v
a
n
i
s
h
i
n
g
.
Fig
u
r
e
2
s
h
o
w
s
a
ch
u
n
k
o
f
n
e
u
r
al
n
et
w
o
r
k
,
A
,
w
h
ic
h
lo
o
k
s
at
s
o
m
e
i
n
p
u
t
x
t
an
d
o
u
tp
u
t
s
a
v
al
u
e
h
t
.
A
lo
o
p
allo
w
s
in
f
o
r
m
at
io
n
to
b
e
p
ass
ed
f
r
o
m
o
n
e
s
tep
o
f
t
h
e
n
et
w
o
r
k
to
th
e
n
e
x
t.
Fig
u
r
e
2
.
R
NN
-
L
ST
M
ar
ch
itectu
r
e.
T
h
e
in
p
u
t o
f
x
t
is
f
ed
in
t
o
th
e
m
e
m
o
r
y
ce
ll
A
w
h
ich
g
i
v
es o
u
tp
u
t
h
t
T
h
e
L
ST
M
R
NN
m
o
d
el
p
r
o
p
o
s
ed
in
t
h
is
s
t
u
d
y
i
n
cl
u
d
es
t
w
o
h
id
d
en
la
y
er
s
w
it
h
4
0
h
id
d
en
u
n
i
ts
ea
c
h
th
at
ar
e
L
ST
M
ce
lls
.
T
h
e
b
ac
k
p
r
o
p
ag
atio
n
th
r
o
u
g
h
ti
m
e
is
li
m
ited
to
1
5
s
tep
s
.
A
d
r
o
p
o
u
t
la
y
er
is
i
n
cl
u
d
ed
b
et
w
ee
n
th
e
2
h
id
d
en
la
y
er
s
f
o
r
r
eg
u
lar
is
atio
n
[
2
1
]
.
I
t
w
ill
r
an
d
o
m
l
y
e
x
clu
d
e
3
0
%
o
f
th
e
ac
tiv
atio
n
s
o
f
th
e
p
r
ev
io
u
s
la
y
er
f
r
o
m
p
r
o
p
ag
ati
n
g
to
p
r
ev
e
n
t
o
v
er
f
i
tti
n
g
.
T
h
e
R
o
o
t
Me
an
Sq
u
ar
e
(
R
MS)
lo
s
s
is
r
ed
u
ce
d
u
s
i
n
g
th
e
A
d
ag
r
ad
alg
o
r
ith
m
w
h
ic
h
in
cr
ea
s
e
s
th
e
lear
n
i
n
g
r
ate
f
o
r
m
o
r
e
s
p
ar
s
e
p
ar
am
eter
s
an
d
d
ec
r
ea
s
es
th
e
lear
n
in
g
r
ate
f
o
r
less
s
p
ar
s
e
o
n
es.
T
h
is
s
tr
ate
g
y
o
f
te
n
i
m
p
r
o
v
es
co
n
v
er
g
en
ce
p
er
f
o
r
m
an
ce
o
v
er
s
ta
n
d
ar
d
s
to
ch
ast
ic
g
r
ad
ien
t
d
esce
n
t
in
s
etti
n
g
s
w
h
er
e
d
ata
i
s
s
p
ar
s
e
[
2
2
]
.
T
h
e
in
itia
l
lear
n
in
g
r
ate
i
s
ta
k
en
to
b
e
7
an
d
is
e
x
p
o
n
en
tiall
y
d
ec
r
ea
s
ed
w
h
en
t
h
e
R
MS
lo
s
s
d
o
es
n
o
t
i
m
p
r
o
v
e
f
o
r
m
o
r
e
th
a
n
1
0
ep
o
ch
s
.
T
h
e
tr
ai
n
in
g
w
a
s
s
to
p
p
ed
af
ter
th
e
lo
s
s
s
tar
ted
to
f
l
u
ctu
a
te
d
esp
ite
v
er
y
lo
w
lear
n
in
g
r
ate.
T
h
e
n
u
m
b
er
o
f
ep
o
ch
s
ca
m
e
to
b
e
1
6
0
0
.
T
h
ese
p
ar
am
e
ter
s
w
er
e
s
elec
ted
af
ter
tr
y
i
n
g
o
u
t o
th
er
ar
ch
itect
u
r
es.
3.
RE
SU
L
T
S
A
ND
AN
AL
Y
SI
S
3
.
1
.
Da
t
a
e
x
plo
ra
t
io
n
Nu
m
b
er
o
f
o
cc
u
r
r
en
ce
s
o
f
ea
r
th
q
u
a
k
es
r
ec
o
r
d
ed
o
v
er
v
ar
io
u
s
r
eg
io
n
s
is
s
h
o
w
n
in
Fig
u
r
e
3
.
T
h
e
Af
g
h
a
n
is
ta
n
-
T
aj
ik
is
tan
r
eg
io
n
r
ec
o
r
d
ed
clo
s
e
to
5
0
0
0
ea
r
t
h
q
u
a
k
es
b
ei
n
g
t
h
e
h
i
g
h
e
s
t.
T
h
e
L
ak
s
h
ad
w
ee
p
r
eg
io
n
r
ec
o
r
d
ed
th
e
lo
w
e
s
t
n
u
m
b
er
o
f
ea
r
t
h
q
u
a
k
es.
T
h
e
d
is
tr
ib
u
tio
n
o
f
co
u
n
t
o
f
ea
r
th
q
u
ak
e
s
b
ased
o
n
t
h
e
m
ag
n
it
u
d
e
r
ec
o
r
d
ed
o
n
ea
ch
ea
r
th
q
u
a
k
e
is
s
h
o
w
n
in
Fi
g
u
r
e
4
.
E
ar
th
q
u
ak
es
o
f
m
a
g
n
it
u
d
e
3
.
4
o
n
th
e
R
ic
h
ter
s
ca
le
w
er
e
h
i
g
h
e
s
t
i
n
a
m
o
u
n
t
b
ein
g
c
l
o
s
e
to
4
5
o
cc
u
r
r
en
ce
s
w
h
er
ea
s
t
h
at
o
f
h
i
g
h
er
r
an
g
e
c
lo
s
e
to
7
.
9
ar
e
lo
w
in
o
cc
u
r
r
en
ce
s
.
Fi
g
u
r
e
5
s
h
o
w
s
t
h
e
a
v
er
ag
e
m
a
g
n
it
u
d
e
o
f
ea
r
th
q
u
a
k
es
in
ea
ch
r
eg
io
n
.
T
h
e
w
e
s
ter
n
x
iza
n
g
-
in
d
ia
b
o
r
d
er
h
as
th
e
h
i
g
h
est
av
er
ag
e
m
ag
n
it
u
d
e
w
h
er
ea
s
T
h
ailan
d
an
d
No
r
th
-
E
as
ter
n
I
n
d
ia
h
as
t
h
e
lo
w
e
s
t.
Fig
u
r
e
6
s
h
o
w
s
th
e
av
er
a
g
e
d
ep
th
o
f
ea
ch
ea
r
th
q
u
a
k
e
r
ec
o
r
d
ed
in
k
ilo
m
etr
es.
An
ea
r
th
q
u
a
k
e
o
f
1
4
0
k
m
is
t
h
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8708
E
a
r
th
q
u
a
ke
tr
en
d
p
r
ed
ictio
n
u
s
in
g
lo
n
g
s
h
o
r
t
-
term me
mo
r
y
R
N
N
(
Var
d
aa
n
Ki
s
h
o
r
e
)
1307
d
ee
p
est
r
ec
o
r
d
e
d
ea
r
th
q
u
ak
e,
o
cc
u
r
r
ed
in
th
e
T
a
j
ik
is
ta
n
r
eg
io
n
.
T
h
e
ea
r
th
q
u
ak
e
s
r
ec
o
r
d
ed
in
th
e
th
e
L
a
k
s
h
ad
w
ee
p
an
d
L
ao
s
r
eg
io
n
w
er
e
th
e
lo
w
e
s
t in
d
ep
th
b
ei
n
g
v
er
y
c
lo
s
e
to
th
e
s
u
r
f
ac
e.
Fig
u
r
e
3
.
Nu
m
b
er
o
f
o
cc
u
r
r
en
ce
s
o
f
ea
r
th
q
u
a
k
e
b
ased
o
n
th
e
r
eg
io
n
s
Fig
u
r
e
4
.
Dis
tr
ib
u
tio
n
o
f
ea
r
th
q
u
ak
e
m
ag
n
it
u
d
es
f
r
o
m
lo
w
to
h
ig
h
Fig
u
r
e
5
.
Av
er
ag
e
m
ag
n
it
u
d
e
o
f
ea
r
th
q
u
a
k
e
in
ea
c
h
r
eg
io
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
9
,
No
.
2
,
A
p
r
il
201
9
:
1
3
0
4
-
1312
1308
Fig
u
r
e
6
.
Av
er
ag
e
d
ep
th
o
f
o
c
cu
r
r
in
g
ea
r
th
q
u
a
k
e
3
.
2
.
E
x
peri
m
e
nta
l R
esu
lt
s
3
.
2
.
1.
F
ee
d
f
o
r
w
a
rd
neura
l
net
w
o
rk
Fig
u
r
e
7
s
h
o
w
s
t
h
e
Me
an
Sq
u
ar
e
E
r
r
o
r
af
ter
ev
er
y
ep
o
ch
f
o
r
FF
NN.
As
t
h
e
m
o
d
el
co
n
v
er
g
e
s
,
th
e
er
r
o
r
b
ec
o
m
es
s
tatic
an
d
ac
h
ie
v
es
h
i
g
h
est
v
a
lu
e
s
h
o
w
in
g
its
in
ab
ilit
y
to
m
o
d
el
s
eq
u
e
n
ce
d
ata.
Fi
g
u
r
e
8
s
h
o
w
s
th
e
R
^2
s
co
r
e
f
o
r
ea
ch
v
ar
iab
le
u
s
ed
f
o
r
p
r
ed
ictio
n
an
d
ar
e
n
eg
at
iv
e
w
h
ic
h
i
n
d
icate
s
t
h
at
th
e
FF
NM
is
n
o
t
ab
le
to
ca
p
tu
r
e
th
e
ch
ao
tic
n
atu
r
e
o
f
th
e
attr
ib
u
tes.
Fi
g
u
r
e
9
s
h
o
w
s
th
e
o
r
ig
i
n
al
d
ata
s
et
p
lo
tted
o
n
m
ap
.
Fig
u
r
e
1
0
s
h
o
w
s
t
h
e
p
r
ed
icted
6
0
0
0
ea
r
th
q
u
ak
es
b
y
t
h
e
F
FNN
p
lo
tted
o
n
th
e
m
ap
,
as
s
ee
n
t
h
e
r
eg
io
n
i
s
clu
s
ter
ed
s
h
o
w
i
n
g
a
w
ea
k
p
r
e
d
ictio
n
tr
en
d
.
Fi
g
u
r
e
1
1
s
h
o
w
s
th
e
r
eg
io
n
w
i
s
e
d
is
tr
ib
u
ti
o
n
o
f
ea
r
th
q
u
ak
e
s
o
f
t
h
e
ea
r
th
q
u
a
k
es
p
r
ed
icted
b
y
t
h
e
FF
NN,
a
s
o
b
s
er
v
ed
t
h
e
d
ata
i
s
s
k
e
w
ed
an
d
it
s
h
o
w
s
th
e
Xiza
n
g
r
eg
io
n
w
ill
h
a
v
e
m
o
r
e
t
h
a
n
7
0
0
0
ea
r
th
q
u
ak
es i
n
th
e
p
r
ed
ictio
n
tr
en
d
.
Fig
u
r
e
7
.
MSE
af
ter
ev
er
y
ep
o
ch
f
o
r
FF
N
N
Fig
u
r
e
8
.
R
^2
Sco
r
e
f
o
r
ea
ch
v
ar
iab
le
at
th
e
en
d
o
f
tr
ain
i
n
g
o
f
FF
NN
Fig
u
r
e
9
.
Giv
en
d
ata
p
lo
tted
o
n
a
m
ap
Fig
u
r
e
1
0
.
Nex
t 8
9
8
8
FF
NN
p
r
ed
icted
ea
r
th
q
u
ak
es
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8708
E
a
r
th
q
u
a
ke
tr
en
d
p
r
ed
ictio
n
u
s
in
g
lo
n
g
s
h
o
r
t
-
term me
mo
r
y
R
N
N
(
Var
d
aa
n
Ki
s
h
o
r
e
)
1309
Fig
u
r
e
1
1
.
R
eg
io
n
w
i
s
e
d
is
tr
ib
u
tio
n
o
f
p
r
ed
icted
ea
r
th
q
u
ak
e
b
y
FF
NN
3
.
2
.
2.
L
o
ng
s
ho
rt
-
t
er
m
m
e
mo
ry
Fig
u
r
e
1
2
s
h
o
w
s
th
e
MSE
af
te
r
ev
er
y
ep
o
ch
f
o
r
L
ST
M.
A
s
w
e
ca
n
s
ee
th
e
MSE
co
n
v
er
g
e
s
at
a
p
o
in
t
lo
w
er
t
h
an
t
h
at
o
f
t
h
e
FF
NN
m
o
d
el,
t
h
i
s
s
h
o
w
th
a
t
t
h
e
m
o
d
el
is
co
n
v
er
g
i
n
g
an
d
t
h
e
er
r
o
r
r
ate
is
lo
w
as
w
el
l.
Fig
u
r
e
1
3
s
h
o
w
s
th
e
R
^2
s
co
r
e
o
f
ev
er
y
v
ar
iab
le
o
n
t
h
e
test
s
et.
T
h
e
s
co
r
es
ar
e
all
p
o
s
iti
v
e
.
T
h
er
e
is
a
d
r
asti
c
i
m
p
r
o
v
e
m
en
t
in
th
e
s
co
r
e
f
o
r
T
im
esta
m
p
a
n
d
th
i
s
p
u
l
ls
u
p
t
h
e
o
v
er
all
s
co
r
e
to
-
0
.
2
5
2
w
h
i
ch
is
5
9
%
m
o
r
e
th
a
n
th
e
s
co
r
e
f
o
r
FF
NN
b
u
t
is
s
t
ill
n
eg
ati
v
e.
T
h
is
in
d
icate
s
t
h
at
t
h
o
u
g
h
e
x
p
licitl
y
ca
p
t
u
r
i
n
g
s
eq
u
en
ce
i
n
f
o
r
m
atio
n
b
y
u
s
i
n
g
L
ST
Ms le
ad
s
to
b
etter
r
esu
lts
t
h
a
n
tr
ad
itio
n
al
n
eu
r
al
n
et
w
o
r
k
s
.
T
h
e
R
^2
s
co
r
e
is
u
s
ed
to
ev
alu
ate
o
u
r
m
o
d
el
s
.
R
^2
(
co
ef
f
icie
n
t
o
f
d
eter
m
i
n
atio
n
)
r
eg
r
e
s
s
io
n
s
co
r
e
f
u
n
ctio
n
.
B
est
p
o
s
s
ib
le
s
co
r
e
is
1
.
0
an
d
it
ca
n
b
e
n
eg
at
iv
e
(
b
ec
au
s
e
t
h
e
m
o
d
el
ca
n
b
e
ar
b
itra
r
il
y
w
o
r
s
e)
.
A
c
o
n
s
ta
n
t
m
o
d
el
th
at
al
w
a
y
s
p
r
ed
icts
th
e
ex
p
ec
ted
v
alu
e
o
f
y
,
d
is
r
eg
ar
d
in
g
t
h
e
i
n
p
u
t
f
ea
t
u
r
es,
w
o
u
ld
g
et
a
R
^2
s
co
r
e
o
f
0
.
0
.
R
^2
s
co
r
es
ar
e
co
n
s
id
er
ed
to
b
e
b
etter
th
a
n
t
h
e
MSE
b
ec
au
s
e
it
is
s
ca
led
b
et
w
ee
n
0
-
1,
w
h
er
e
as
M
SE
is
n
o
t
s
ca
led
to
a
n
y
p
ar
ticu
lar
v
al
u
e
s
.
R
^2
ca
n
b
e
in
ter
p
r
eted
m
o
r
e
ea
s
il
y
.
T
ab
le
1
g
iv
es
th
e
c
o
m
p
ar
is
o
n
o
f
t
h
e
R
^2
s
co
r
es
b
et
w
ee
n
th
e
Feed
Fo
r
w
ar
d
Neu
r
al
Net
w
o
r
k
a
n
d
L
o
n
g
S
h
o
r
t
-
T
er
m
Me
m
o
r
y
R
NN.
Fig
u
r
e
1
2
.
MSE
af
ter
ev
er
y
ep
o
ch
f
o
r
L
ST
M
Fig
u
r
e
1
3
.
R
^2
s
co
r
e
f
o
r
ea
ch
v
ar
iab
le
at
th
e
en
d
o
f
tr
ain
i
n
g
o
f
L
ST
M
T
ab
le
1
.
C
o
m
p
ar
is
o
n
o
f
R
^2
S
co
r
e
A
t
t
r
i
b
u
t
e
F
F
N
N
L
S
T
M
O
v
e
r
a
l
l
T
i
me
S
t
a
mp
D
e
p
t
h
M
a
g
n
i
t
u
d
e
L
o
n
g
i
t
u
d
e
L
a
t
i
t
u
d
e
-
0
.
6
1
8
8
-
4
.
6
7
7
-
0
.
2
0
1
-
0
.
0
0
4
-
0
.
0
4
0
-
0
.
0
2
9
-
0
.
2
5
2
-
0
.
5
2
4
-
0
.
0
3
2
-
0
.
2
4
6
-
0
.
3
1
-
0
.
0
8
3
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
9
,
No
.
2
,
A
p
r
il
201
9
:
1
3
0
4
-
1312
1310
B
y
p
lo
tti
n
g
t
h
e
ea
r
th
q
u
a
k
e
s
p
r
ed
icted
f
r
o
m
t
h
e
d
ata
s
et
o
n
a
m
ap
ac
r
o
s
s
ep
o
ch
s
w
e
ca
n
g
et
a
f
ee
l
o
f
ho
w
th
e
d
ata
i
s
s
p
r
ea
d
.
Fig
u
r
e
1
4
s
h
o
w
s
th
e
Fu
t
u
r
e
6
0
0
0
p
r
ed
icted
ea
r
th
q
u
ak
es
o
n
t
h
e
m
ap
;
th
e
s
p
r
ea
d
-
o
u
t
p
o
in
ts
s
h
o
w
th
at
th
e
L
ST
M
is
b
etter
th
an
th
e
F
FNN
i
n
p
r
ed
ictio
n
.
Fig
u
r
e
1
5
s
h
o
w
s
th
e
r
e
g
io
n
w
i
s
e
d
is
tr
ib
u
tio
n
o
f
t
h
e
ea
r
t
h
q
u
a
k
es
p
r
ed
icted
b
y
t
h
e
L
ST
M
n
et
wo
r
k
.
Nep
al
r
eg
io
n
is
p
r
ed
icted
to
h
av
e
th
e
h
i
g
h
es
t
n
u
m
b
er
o
f
ea
r
th
q
u
ak
e
s
in
t
h
e
f
u
tu
r
e.
Fig
u
r
e
1
4
.
Ma
p
o
f
Nex
t 6
0
0
1
p
r
ed
icted
ea
r
th
q
u
ak
e
o
f
L
ST
M
Fig
u
r
e
1
5
.
R
eg
io
n
w
i
s
e
d
is
tr
ib
u
tio
n
o
f
ea
r
th
q
u
a
k
e
f
r
o
m
L
ST
M
p
r
ed
ictio
n
T
ab
le
2
s
h
o
w
s
co
m
p
ar
is
o
n
o
f
f
u
tu
r
e
p
r
ed
icted
ea
r
th
q
u
ak
es
o
f
b
o
th
m
o
d
els
ag
ai
n
s
t
t
h
e
o
r
ig
in
al
d
ataset,
it
ca
n
b
e
o
b
s
er
v
ed
t
h
at
th
e
F
FNN
s
h
o
w
s
a
lar
g
e
n
u
m
b
er
o
f
ea
r
t
h
q
u
a
k
es
i
n
Xiza
n
g
an
d
to
tall
y
ze
r
o
ea
r
th
q
u
a
k
es
f
o
r
So
u
t
h
er
n
I
n
d
i
a,
th
is
i
n
d
icate
s
th
a
t
th
e
m
o
d
el
is
w
ea
k
.
L
ST
M
s
h
o
w
s
a
ch
an
g
i
n
g
ea
r
t
h
q
u
a
k
e
tr
en
d
in
t
h
e
f
u
t
u
r
e
w
ith
m
u
ch
m
o
r
e
ea
r
t
h
q
u
a
k
es a
lo
n
g
Nep
al
an
d
So
u
t
h
I
n
d
ia.
T
ab
le
2
.
C
o
m
p
ar
is
o
n
o
f
Fu
tu
r
e
P
r
ed
icted
E
ar
th
q
u
ak
es o
f
B
o
th
Mo
d
els
ag
ai
n
s
t
th
e
Or
i
g
i
n
al
Data
s
et
A
r
e
a
O
r
i
g
i
n
a
l
D
a
t
a
F
F
N
N
L
S
T
M
X
i
z
a
n
g
4
8
0
0
7
4
0
0
1
4
0
0
N
e
p
a
l
3
0
0
5
0
0
2
4
0
0
S
o
u
t
h
e
r
n
I
n
d
i
a
9
0
0
0
1
7
5
0
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8708
E
a
r
th
q
u
a
ke
tr
en
d
p
r
ed
ictio
n
u
s
in
g
lo
n
g
s
h
o
r
t
-
term me
mo
r
y
R
N
N
(
Var
d
aa
n
Ki
s
h
o
r
e
)
1311
4.
CO
NCLU
SI
O
N
L
ST
M
m
o
d
el
p
r
ed
icts
ea
r
t
h
q
u
ak
e
m
ag
n
it
u
d
e,
d
ep
th
,
ti
m
e,
la
titu
d
e
a
n
d
lo
n
g
it
u
d
e
an
d
g
o
t
s
ig
n
if
ican
t
r
esu
lt
s
th
a
t
in
d
icate
t
h
at
t
h
e
p
r
esen
t
m
o
d
el
d
o
es
in
d
ee
d
ca
p
tu
r
e
s
o
m
e
tr
en
d
s
i
n
t
h
e
g
iv
e
n
ea
r
th
q
u
a
k
e
d
ata.
T
h
e
r
esu
lt
s
clea
r
l
y
s
h
o
w
th
e
s
u
p
er
io
r
it
y
o
f
L
ST
Ms
o
v
er
r
eg
u
lar
FF
NM
s
in
th
e
ta
s
k
o
f
m
o
d
elli
n
g
th
e
s
eq
u
e
n
ce
o
f
ea
r
th
q
u
a
k
es.
T
h
e
R
^2
s
co
r
e
y
ield
ed
b
y
L
ST
M
o
n
th
e
test
d
ata
is
5
9
%
h
ig
h
er
th
an
t
h
e
FF
NN’
s
.
Fu
t
u
r
e
w
o
r
k
ca
n
b
e
d
o
n
e
b
y
co
llecti
n
g
d
ata
o
f
ea
r
th
q
u
a
k
e
s
in
r
e
g
io
n
s
li
k
e
Si
k
k
i
m
in
I
n
d
ia
w
h
ich
ar
e
ea
r
th
q
u
a
k
e
p
r
o
n
e.
T
h
e
p
r
esen
t
m
o
d
el
ca
n
b
e
i
m
p
r
o
v
ed
b
y
co
m
b
i
n
i
n
g
d
i
f
f
er
e
n
t
d
ee
p
lear
n
in
g
m
o
d
els f
o
r
r
ea
l t
i
m
e
u
s
e.
RE
F
E
R
E
NC
E
S
[1
]
P
.
S
h
e
a
re
r,
“
In
tro
d
u
c
ti
o
n
to
S
e
ism
o
lo
g
y
,
”
Ca
m
b
rid
g
e
Un
iv
e
rsit
y
P
re
ss
,
1
9
9
9
.
[2
]
M
a
sa
sh
i
Ha
y
a
k
a
w
a
,
“
Earth
q
u
a
k
e
s an
d
EQ
P
re
d
ictio
n
,
”
Jo
h
n
W
il
e
y
S
o
n
s,
S
i
n
g
a
p
o
re
P
te.
L
td
,
p
p
.
1
–
1
7
,
2
0
1
5
.
[3
]
S
e
iy
a
U
y
e
d
a
,
T
o
sh
i
y
a
su
Na
g
a
o
,
a
n
d
M
a
sa
sh
i
Ka
m
o
-
g
a
w
a
,
“
S
h
o
rt
-
T
e
r
m
Earth
q
u
a
k
e
P
re
d
icti
o
n
:
Cu
rre
n
t
S
tat
u
s
of
S
e
ism
o
-
El
e
c
tro
m
a
g
n
e
ti
c
s
,”
T
e
c
to
n
o
p
h
y
sic
s,
v
o
l.
4
7
0
(3
)
,
p
p
.
2
0
5
–
213
,
2
0
0
9
.
[4
]
R.
J.
G
e
ll
e
r,
“
Earth
q
u
a
k
e
P
re
d
ictio
n
:
A
Crit
ica
l
R
e
v
ie
w
,
”
Ge
o
p
h
y
sic
a
l
J
o
u
r
n
a
l
In
ter
n
a
ti
o
n
a
l,
v
o
l.
1
3
1
,
p
p
.
4
2
5
–
4
5
0
,
De
c
e
m
b
e
r
1
9
9
7
.
[5
]
Zh
u
Qin
g
-
Jie
S
u
,
Y
o
u
-
P
o
,
“
A
p
p
li
c
a
ti
o
n
o
f
A
n
n
to
P
re
d
icti
o
n
o
f
Earth
q
u
a
k
e
In
f
lu
e
n
c
e
,
”
S
e
c
o
n
d
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
In
f
o
rm
a
ti
o
n
a
n
d
C
o
mp
u
t
in
g
S
c
i
e
n
c
e
,
v
o
l.
2
,
p
p
.
2
3
4
–
2
3
7
,
2
0
0
9
.
[6
]
A
.
Itai,
H.
Ya
su
k
a
wa
,
I.
T
a
k
u
m
i,
a
n
d
M
.
Ha
ta,
“
M
u
lt
i
-
L
a
y
e
r
Ne
u
ra
l
Ne
tw
o
rk
F
o
r
P
re
c
u
rs
o
r
S
ig
n
a
l
De
tec
ti
o
n
I
n
El
e
c
tro
m
a
g
n
e
ti
c
Wav
e
Ob
se
rv
a
t
io
n
A
p
p
li
e
d
t
o
G
re
a
t
Earth
q
u
a
k
e
P
re
d
icti
o
n
,
”
In
N
S
IP
2
0
0
5
A
b
stra
c
ts.
IEE
E
-
Eu
ra
sip
No
n
li
n
e
a
r S
i
g
n
a
l
a
n
d
Im
a
g
e
Pro
c
e
ss
in
g
,
2
0
0
5
,
p
p
.
3
1
–
,
M
a
y
2
0
0
5
.
[7
]
V
.
P
.
P
lag
ian
a
k
o
s
a
n
d
E.
T
z
a
n
a
k
i,
“
Ch
a
o
ti
c
A
n
a
l
y
sis
of
S
e
is
m
i
c
T
i
m
e
S
e
ries
a
n
d
S
h
o
rt
-
T
e
rm
F
o
re
c
a
stin
g
U
sin
g
Ne
u
ra
l
Ne
tw
o
rk
s
,
”
In
Ne
u
ra
l
Ne
two
rk
s,
2
0
0
1
.
Pro
c
e
e
d
in
g
s.
IJ
CNN
’0
1
.
I
n
ter
n
a
ti
o
n
a
l
J
o
in
t
Co
n
fer
e
n
c
e
o
n
,
v
o
l
u
m
e
3
,
v
o
l.
3
,
p
p
.
1
5
9
8
–
1
6
0
2
,
2
0
0
1
.
[8
]
K
lah
c
i
F
.
,
In
c
e
o
z
M
.
,
o
g
ru
,
E
.
A
k
so
y
,
O.
a
y
k
a
r
a
,
“
A
n
n
F
o
r
Earth
q
u
a
k
e
P
re
d
icti
o
n
W
it
h
Ra
d
o
n
M
o
n
it
o
ri
n
g
,
”
Ap
p
li
e
d
R
a
d
i
a
ti
o
n
a
n
d
Is
o
to
p
e
s,
v
o
l.
6
7
,
p
p
.
2
1
2
–
2
1
9
,
2
0
0
9
.
[9
]
A
.
Ne
g
a
r
e
sta
n
i,
S
.
S
e
tay
e
sh
i,
M
.
G
h
a
n
n
a
d
i
-
M
a
ra
g
h
e
h
,
a
n
d
.
A
k
a
sh
e
,
“
L
a
y
e
re
d
Ne
u
ra
l
Ne
t
w
o
rk
s
-
Ba
se
d
A
n
a
l
y
sis
of
Ra
d
o
n
Co
n
c
e
n
tratio
n
a
n
d
En
v
i
ro
n
m
e
n
tal
P
a
ra
m
e
ters
in
Earth
q
u
a
k
e
P
re
d
ictio
n
,
”
J
En
v
iro
n
Ra
d
i
o
a
c
t,
v
o
l.
6
2
(
3
)
,
p
p
.
2
2
5
–
2
3
3
,
2
0
0
2
.
[1
0
]
M
e
h
m
e
t
S
irac
Oz
e
rd
e
m
,
e
rk
Us
t
u
n
d
a
g
,
a
n
d
R.
M
u
ra
t
e
m
irer,
“
S
e
lf
-
Or
g
a
n
ize
d
M
a
p
s Ba
se
d
Ne
u
ra
l
Ne
tw
o
rk
s F
o
r
De
tec
ti
o
n
Of
P
o
ss
ib
le
Earth
q
u
a
k
e
P
re
c
u
rso
ry
El
e
c
tri
c
F
ield
P
a
t
tern
s
,
”
Ad
v
a
n
c
e
s
in
En
g
in
e
e
rin
g
S
o
ft
wa
re
,
v
o
l.
3
7
(
4
),
p
p
.
2
0
7
–
2
1
7
,
2
0
0
6
.
[1
1
]
Y.Q.
Ni,
X
.
T
.
Zh
o
u
,
a
n
d
J.M
.
Ko
,
“
Ex
p
e
rime
n
tal
In
v
e
stig
a
ti
o
n
o
f
S
e
is
m
ic
Da
m
a
g
e
Id
e
n
ti
f
ica
ti
o
n
Us
in
g
P
c
a
-
Co
m
p
re
ss
e
d
F
re
q
u
e
n
c
y
Re
sp
o
n
se
F
u
n
c
ti
o
n
s
A
n
d
Ne
u
ra
l
Ne
tw
o
rk
s
,
”
J
o
u
rn
a
l
o
f
S
o
u
n
d
a
n
d
V
ib
ra
t
io
n
,
v
o
l.
2
9
0
(
1
)
,
p
p
.
2
4
2
–
2
6
3
,
2
0
0
6
.
[1
2
]
Zh
o
n
g
Zh
i
h
u
a
n
a
n
d
Yu
J
u
n
ji
n
g
,
“
P
re
d
ictio
n
of
Earth
q
u
a
k
e
Da
m
a
g
e
s
a
n
d
Re
li
a
b
il
it
y
A
n
a
l
y
sis Us
in
g
F
u
z
z
y
S
e
ts
,
”
In
[
1
9
9
0
]
Pro
c
e
e
d
in
g
s
.
Fi
rs
t
In
ter
n
a
ti
o
n
a
l
S
y
mp
o
siu
m
o
n
Un
c
e
rta
in
ty
M
o
d
e
li
n
g
a
n
d
A
n
a
lys
is
,
p
p
.
1
7
3
–
1
7
6
,
De
c
1
9
9
0
.
[1
3
]
M
.
H.
Ju
s
o
h
,
N
.
Ya
’a
c
o
b
,
H.
S
a
a
d
,
A
.
A
.
S
u
laim
a
n
,
N.
H.
a
b
a
,
R.
A
.
Aw
a
n
g
,
a
n
d
Z.
I.
Kh
a
n
,
“
Earth
q
u
a
k
e
P
re
d
ictio
n
T
e
c
h
n
iq
u
e
Ba
se
d
o
n
G
p
s
Du
a
l
F
re
q
u
e
n
c
y
S
y
ste
m
i
n
Eq
u
a
to
rial
Re
g
io
n
,
”
In
2
0
0
8
IEE
E
I
n
ter
n
a
ti
o
n
a
l
RF
a
n
d
M
icr
o
wa
v
e
Co
n
fer
e
n
c
e
,
p
p
.
3
7
2
–
3
7
6
,
De
c
2
0
0
8
.
[1
4
]
S
h
o
h
e
i
S
h
im
izu
,
Ko
ich
i
S
u
g
isa
k
i,
a
n
d
Hiro
m
it
su
Oh
m
o
ri,
“
Re
c
u
rs
iv
e
S
a
m
p
le
-
En
tro
p
y
M
e
th
o
d
a
n
d
Its
A
p
p
li
c
a
ti
o
n
F
o
r
Co
m
p
lex
it
y
Ob
se
rv
a
ti
o
n
Of
Earth
Cu
rre
n
t
,
”
In
2
0
0
8
In
ter
n
a
ti
o
n
a
l
C
o
n
fer
e
n
c
e
o
n
C
o
n
tro
l,
Au
to
m
a
ti
o
n
a
n
d
S
y
ste
ms
,
p
p
.
1
2
5
0
–
1
2
5
3
,
Oc
t
2
0
0
8
.
[1
5
]
G
.
P
.
T
u
rm
o
v
,
V
.
I.
K
o
ro
c
h
e
n
tse
v
,
E.
V
.
G
o
ro
d
e
tsk
a
y
a
,
A
.
M
.
M
i
ro
n
e
n
k
o
,
D.
V
.
Kislit
sin
,
a
n
d
O.
A
.
S
taro
d
u
b
tse
v
,
“
F
o
re
c
a
st
O
f
Un
d
e
rwa
ter
Earth
q
u
a
k
e
s
W
it
h
A
G
re
a
t
De
g
re
e
Of
P
ro
b
a
b
i
li
ty
,
”
In
Pro
c
e
e
d
in
g
s
o
f
th
e
2
0
0
0
In
ter
n
a
t
io
n
a
l
S
y
mp
o
si
u
m o
n
U
n
d
e
rwa
ter
T
e
c
h
n
o
l
o
g
y
(Ca
t.
No
.
0
0
E
X
4
1
8
)
,
p
p
.
1
1
0
–
1
1
5
,
2
0
0
0
.
[1
6
]
R.
S
iv
a
k
u
m
a
r
a
n
d
S
n
e
h
a
sish
G
h
o
sh
,
“
Earth
q
u
a
k
e
Ha
z
a
rd
A
ss
e
ss
m
e
n
t
th
r
o
u
g
h
G
e
o
sp
a
ti
a
l
M
o
d
e
l
a
n
d
De
v
e
lo
p
m
e
n
t
of
Eah
a
a
sto
T
o
o
l
f
o
r
Visu
a
li
z
a
ti
o
n
:
A
n
In
teg
ra
ted
G
e
o
lo
g
ica
l
a
n
d
G
e
o
in
f
o
r
m
a
ti
c
s
A
p
p
ro
a
c
h
,
”
En
v
iro
n
me
n
t
a
l
E
a
rt
h
S
c
ien
c
e
s
,
v
o
l.
7
6
(
1
2
),
p
p
.
4
4
2
,
J
u
n
2
0
1
7
.
[1
7
]
R.
S
iv
a
k
u
m
a
r
a
n
d
S
n
e
h
a
sish
G
h
o
sh
,
“
e
term
in
a
ti
o
n
o
f
T
h
re
sh
o
ld
En
e
rg
y
f
o
r
th
e
De
v
e
lo
p
m
e
n
t
o
f
S
e
is
m
ic
En
e
rg
y
A
n
o
m
a
l
y
M
o
d
e
l
th
ro
u
g
h
In
teg
ra
ted
Ge
o
tec
to
n
ic
a
n
d
G
e
o
in
f
o
rm
a
ti
c
s
A
p
p
ro
a
c
h
,
”
Na
tu
r
a
l
Ha
za
rd
s:
J
o
u
rn
a
l
o
f
th
e
In
ter
n
a
t
io
n
a
l
S
o
c
iety
f
o
r th
e
Pre
v
e
n
ti
o
n
a
n
d
M
it
ig
a
ti
o
n
o
f
Na
tu
ra
l
Ha
za
rd
s
,
v
o
l
.
8
6
(
2
),
p
p
.
7
1
1
–
7
4
0
,
2
0
1
7
.
[1
8
]
A
.
Zell,
S
imu
l
a
ti
o
n
n
e
u
ro
n
a
ler
N
e
tze
.
Old
e
n
b
o
u
rg
,
1
9
9
7
.
[1
9
]
Zac
h
a
r
y
Ch
a
se
L
ip
to
n
,
“
A
Crit
ica
l
Re
v
ie
w
of
Re
c
u
rre
n
t
Ne
u
ra
l
Ne
t
w
o
rk
s
f
o
r
S
e
q
u
e
n
c
e
L
e
a
rn
in
g
,”
Co
RR
,
a
b
s/1
5
0
6
.
0
0
0
1
9
,
2
0
1
5
.
[2
0
]
S
e
p
p
Ho
c
h
re
it
e
r
a
n
d
J
rg
e
n
S
c
h
m
id
h
u
b
e
r.
“
L
o
n
g
S
h
o
rt
-
T
e
r
m
M
e
m
o
r
y
,
”
Ne
u
ra
l
Co
mp
u
t.
,
v
o
l.
9
(8
),
p
p
.
1
7
3
5
–
1
7
8
0
,
N
o
v
e
m
b
e
r
1
9
9
7
.
[2
1
]
W
o
jcie
c
h
Zare
m
b
a
,
Il
y
a
S
u
tsk
e
v
e
r,
a
n
d
Orio
l
V
i
n
y
a
ls,
“
R
e
c
u
rre
n
t
Ne
u
ra
l
Ne
tw
o
rk
Re
g
u
lariz
a
ti
o
n
,
”
Co
R
R,
a
b
s
/
1
4
0
9
.
2
3
2
9
,
2
0
1
4
.
[2
2
]
Jo
h
n
u
c
h
i,
E
lad
Ha
z
a
n
,
a
n
d
Y
o
ra
m
S
in
g
e
r,
“
A
d
a
p
ti
v
e
S
u
b
g
ra
d
ie
n
t
M
e
th
o
d
s
f
o
r
On
li
n
e
L
e
a
rn
in
g
a
n
d
S
t
o
c
h
a
stic
Op
ti
m
iza
ti
o
n
,
”
J
.
M
a
c
h
.
L
e
a
rn
.
R
e
s.,
v
o
l.
1
2
,
p
p
.
2
1
2
1
–
2
1
5
9
,
J
u
ly
2
0
1
1
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
9
,
No
.
2
,
A
p
r
il
201
9
:
1
3
0
4
-
1312
1312
B
I
O
G
RAP
H
I
E
S
O
F
AUTH
O
RS
Ta
n
v
i
B
h
a
n
d
a
r
k
a
r
is
p
u
rsu
i
n
g
h
e
r
b
a
c
h
e
lo
r
d
e
g
re
e
in
C
o
m
p
u
ter
S
c
ien
c
e
a
n
d
E
n
g
in
e
e
rin
g
f
ro
m
S
RM
In
stit
u
te
o
f
S
c
ien
c
e
a
n
d
tec
h
n
o
l
o
g
y
.
H
e
r
re
s
e
a
rc
h
in
tere
sts
in
c
lu
d
e
a
p
p
li
c
a
ti
o
n
s
o
f
in
telli
g
e
n
t
sy
ste
m
s,
d
e
e
p
lea
rn
in
g
a
n
d
p
a
tt
e
rn
re
c
o
g
n
i
ti
o
n
.
Va
r
d
a
a
n
K
h
a
n
te
d
is
a
n
En
g
in
e
e
rin
g
S
tu
d
e
n
t
a
t
S
RM
Un
iv
e
rsity
,
p
u
rsu
in
g
Co
m
p
u
ter
S
c
ien
c
e
En
g
in
e
e
rin
g
a
n
d
is
w
o
rk
in
g
in
th
e
a
re
a
o
f
A
rti
f
icia
l
In
telli
g
e
n
c
e
a
n
d
M
a
c
h
in
e
L
e
a
rn
in
g
m
a
i
n
ly
f
o
c
u
sin
g
o
n
c
ro
ss
-
d
o
m
a
in
s ap
p
li
c
a
ti
o
n
s
o
f
it
.
Ni
k
h
il
is
a
n
En
g
in
e
e
rin
g
S
tu
d
e
n
t
a
t
S
RM
Un
iv
e
rsit
y
,
p
u
rsu
in
g
Co
m
p
u
ter
S
c
ien
c
e
En
g
in
e
e
rin
g
a
n
d
is
w
o
rk
in
g
in
th
e
a
re
a
o
f
A
rti
f
ici
a
l
In
telli
g
e
n
c
e
a
n
d
M
a
c
h
in
e
L
e
a
rn
in
g
m
a
in
ly
f
o
c
u
sin
g
o
n
c
r
o
ss
-
d
o
m
a
in
s
a
p
p
li
c
a
ti
o
n
s o
f
it
.
S
h
e
h
a
sis
h
G
h
o
s
h
is
c
u
rre
n
tl
y
th
e
S
e
n
i
o
r
Re
se
a
rc
h
F
e
ll
o
w
a
t
Earth
q
u
a
k
e
Re
se
a
rc
h
Ce
ll
a
t
S
RM
Un
iv
e
rsit
y
.
His
re
se
a
rc
h
a
re
a
s
a
re
Re
m
o
te
se
n
sin
g
&
G
IS
,
G
e
o
d
y
n
a
m
ic
b
e
h
a
v
io
r
a
n
d
e
a
rth
q
u
a
k
e
e
n
e
rg
y
stu
d
y
,
A
c
ti
v
e
f
a
u
lt
m
a
p
p
in
g
&
c
las
si
f
ica
ti
o
n
,
De
v
e
lo
p
m
e
n
t
o
f
G
IS
c
u
sto
m
iza
ti
o
n
to
o
l.
Dr
.
R.
S
iv
a
k
u
m
a
r
is
a
n
A
s
so
c
iate
P
ro
f
e
ss
o
r
in
th
e
De
p
a
rtm
e
n
t
o
f
Civ
il
En
g
in
e
e
rin
g
a
n
d
h
a
s
g
o
t
rich
re
se
a
rc
h
e
x
p
e
rien
c
e
in
Re
m
o
te
se
n
sin
g
&
G
IS
.
His
o
t
h
e
r
re
se
a
rc
h
in
tere
sts
a
re
G
e
o
lo
g
ica
l
re
m
o
t
e
se
n
sin
g
,
Ge
o
in
f
o
rm
a
ti
c
s
in
u
rb
a
n
h
y
d
ro
lo
g
y
,
E
a
rth
q
u
a
k
e
stu
d
i
e
s,
Disa
ste
r
stu
d
ies
&
m
it
ig
a
ti
o
n
,
H
y
p
e
rsp
e
c
tral
re
m
o
te se
n
sin
g
,
Dig
it
a
l
Im
a
g
e
p
ro
c
e
ss
in
g
.
S.
S.
S
r
id
h
a
r
re
c
e
iv
e
d
h
is
Ba
c
h
e
lo
r
o
f
En
g
in
e
e
ri
n
g
,
B.
E.
(C
o
m
p
u
ter
En
g
in
e
e
rin
g
)
f
ro
m
M
a
d
u
ra
i
Ka
m
a
ra
j
Un
iv
e
r
sit
y
,
In
d
ia
in
1
9
9
0
a
n
d
M
a
ste
r
o
f
S
c
ien
c
e
,
M
.
S
(S
o
f
tw
a
re
S
y
ste
m
s)
f
ro
m
Bi
rla
In
stit
u
te
o
f
T
e
c
h
n
o
l
o
g
y
a
n
d
S
c
ien
c
e
,
P
il
a
n
i,
In
d
ia
in
1
9
9
5
a
n
d
P
h
.
D
in
Co
n
stru
c
ti
v
e
Ne
u
ra
l
Ne
t
w
o
r
k
s
f
ro
m
S
RM
Un
iv
e
rsit
y
.
His
re
s
e
a
rc
h
in
tere
st
in
c
lu
d
e
d
P
a
tt
e
r
n
Re
c
o
g
n
it
io
n
,
Earth
q
u
a
k
e
stu
d
i
e
s,
A
rti
f
icia
l
In
telli
g
e
n
c
e
,
Im
a
g
e
P
ro
c
e
ss
in
g
a
n
d
Bra
in
Co
m
p
u
ter I
n
ter
f
a
c
in
g
.
Evaluation Warning : The document was created with Spire.PDF for Python.