I
AE
S In
t
er
na
t
io
na
l J
o
urna
l o
f
Art
if
icia
l In
t
ellig
ence
(
I
J
-
AI
)
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
26
,
p
p
.
2543
~
2
5
5
3
I
SS
N:
2
2
5
2
-
8
9
3
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijai.v
15
.i
3
.
p
p
2
5
4
3
-
2
5
5
3
2543
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
M
a
chine learning
appro
a
ches for a
no
ma
ly
det
ection
o
f
J
a
ka
rta
a
ir quality
inde
x
M
uh
a
m
m
a
d Riz
k
y
Nurha
mb
a
li,
Yenni Ang
ra
ini
,
A
nwa
r
F
it
ria
nto
S
t
a
t
i
st
i
c
s
a
n
d
D
a
t
a
S
c
i
e
n
c
e
S
t
u
d
y
P
r
o
g
r
a
m
,
S
c
h
o
o
l
o
f
D
a
t
a
S
c
i
e
n
c
e
,
M
a
t
h
e
mat
i
c
s
,
a
n
d
I
n
f
o
r
mat
i
c
s
,
I
P
B
U
n
i
v
e
r
si
t
y
,
B
o
g
o
r
,
I
n
d
o
n
e
si
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Feb
12
,
2
0
2
5
R
ev
is
ed
Feb
9
,
2
0
2
6
Acc
ep
ted
May
12
,
2
0
2
6
An
o
m
a
li
e
s
in
ti
m
e
se
ries
d
a
ta
a
r
e
o
b
se
rv
a
ti
o
n
s
th
a
t
d
e
v
iate
m
a
rk
e
d
ly
fr
o
m
su
rro
u
n
d
in
g
v
a
lu
e
s
o
r
o
v
e
ra
ll
p
a
t
tern
s.
Air
q
u
a
li
ty
i
n
d
e
x
(AQ
I)
d
a
ta,
wh
ich
v
a
ry
o
v
e
r
ti
m
e
,
p
ro
v
id
e
a
su
i
tab
le
c
o
n
te
x
t
f
o
r
a
n
o
m
a
ly
d
e
tec
ti
o
n
.
T
ime
se
ries
a
n
o
m
a
ly
d
e
tec
ti
o
n
c
a
n
b
e
d
o
n
e
with
m
a
c
h
in
e
lea
rn
in
g
a
p
p
ro
a
c
h
e
s
li
k
e
l
o
n
g
sh
o
rt
-
term
m
e
m
o
ry
(
LS
TM
)
a
n
d
e
x
trem
e
g
ra
d
ien
t
b
o
o
stin
g
(
XG
Bo
o
st).
Th
e
se
m
e
th
o
d
s
h
a
v
e
a
d
v
a
n
tag
e
s
o
v
e
r
c
o
n
v
e
n
ti
o
n
a
l
m
e
th
o
d
s
in
h
a
n
d
li
n
g
n
o
n
li
n
e
a
rit
y
a
n
d
lar
g
e
d
a
ta
d
im
e
n
sio
n
s.
Th
is
stu
d
y
c
o
m
p
a
re
s
L
S
TM
a
n
d
XG
Bo
o
st
m
e
th
o
d
s
fo
r
d
e
tec
ti
n
g
a
n
o
m
a
li
e
s
in
Ja
k
a
rta'
s
h
o
u
rl
y
AQ
I
d
a
ta.
Th
e
d
a
tas
e
t
wa
s
o
b
tain
e
d
fro
m
th
e
AirNo
w
we
b
site
a
n
d
c
o
v
e
rs
th
e
p
e
rio
d
fr
o
m
Ja
n
u
a
ry
1
,
2
0
1
8
,
to
De
c
e
m
b
e
r
3
1
,
2
0
2
3
.
A
n
o
m
a
li
e
s
in
th
e
o
b
s
e
rv
e
d
d
a
ta
we
re
lab
e
led
u
sin
g
m
o
v
i
n
g
ra
n
g
e
(M
R)
(2
)
a
n
d
(3
)
a
p
p
r
o
a
c
h
e
s
with
t
h
re
e
a
n
d
fo
u
r
-
sig
m
a
t
h
re
sh
o
l
d
s,
a
n
d
fe
a
tu
re
e
n
g
i
n
e
e
rin
g
(F
E)
wa
s
a
p
p
li
e
d
to
imp
ro
v
e
m
o
d
e
l
p
e
rfo
rm
a
n
c
e
.
T
h
e
re
su
lt
s
in
d
ica
te
th
a
t
LS
TM
is
m
o
re
su
it
a
b
le
th
a
n
XG
Bo
o
st
fo
r
f
o
re
c
a
stin
g
a
n
d
c
las
sifica
ti
o
n
tas
k
s
in
AQ
I
d
a
ta.
LS
TM
a
c
h
iev
e
d
a
n
a
v
e
ra
g
e
m
e
a
n
a
b
so
lu
te
p
e
rc
e
n
tag
e
e
rro
r
(M
APE
)
o
f
1
0
.
3
8
4
0
%
,
a
ro
o
t
m
e
a
n
s
q
u
a
re
e
rro
r
(RM
S
E)
o
f
1
0
.
5
9
1
3
,
a
n
d
a
b
a
lan
c
e
d
a
c
c
u
ra
c
y
(BACC)
o
f
0
.
9
4
2
4
u
n
d
e
r
MR
(2
)
lab
e
li
n
g
with
t
h
e
fo
u
r
-
si
g
m
a
ru
le.
Th
e
a
n
o
m
a
li
e
s
d
e
tec
ted
m
o
stly
o
c
c
u
rre
d
b
e
twe
e
n
2
1
:
0
0
a
n
d
0
9
:
0
0
a
n
d
d
u
ri
n
g
th
e
ra
in
y
se
a
so
n
.
K
ey
w
o
r
d
s
:
Air
q
u
ality
in
d
e
x
An
o
m
alies
Ma
ch
in
e
lear
n
in
g
Ou
tlier
s
T
im
e
s
er
ies
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Mu
h
am
m
ad
R
izk
y
Nu
r
h
am
b
al
i
Statis
t
ics an
d
Data
Scien
ce
St
u
d
y
Pro
g
r
am
,
Sch
o
o
l o
f
Data
Scien
ce
,
Ma
th
em
atics
,
an
d
I
n
f
o
r
m
atics
I
PB
Un
iv
er
s
ity
B
o
g
o
r
,
I
n
d
o
n
esia
E
m
ail:
r
izk
y
n
u
r
h
am
b
ali@
ap
p
s
.
ip
b
.
ac
.
id
1.
I
NT
RO
D
UCT
I
O
N
An
an
o
m
aly
is
a
co
n
d
itio
n
wh
er
e
a
s
y
s
tem
d
o
es
n
o
t
wo
r
k
as
u
s
u
al
an
d
is
s
ig
n
if
ican
tly
d
if
f
er
en
t
f
r
o
m
th
e
g
en
er
al
s
y
s
tem
[
1
]
.
R
elate
d
to
d
ata,
an
o
m
alies
ar
e
co
m
m
o
n
ly
r
ef
er
r
e
d
to
as
o
u
tlier
s
.
I
n
tim
e
s
er
ies
d
ata,
an
an
o
m
aly
is
d
ef
in
ed
as
th
e
p
r
esen
ce
o
f
u
n
e
x
p
ec
ted
b
eh
av
io
r
in
a
tim
e
s
er
ies
d
ata
w
ith
in
a
ce
r
tain
tim
e
in
ter
v
al
[
2
]
.
Ho
wev
er
,
i
n
its
ap
p
licatio
n
,
an
o
m
alies
in
tim
e
s
er
ies
d
ata
ar
e
d
if
f
icu
lt
to
d
etec
t
ev
en
th
o
u
g
h
an
o
m
aly
d
etec
tio
n
is
an
im
p
o
r
tan
t
th
in
g
to
d
o
co
n
s
id
er
in
g
th
at
an
o
m
aly
d
etec
tio
n
ca
n
b
e
an
in
d
icato
r
o
f
f
u
tu
r
e
p
r
o
b
lem
s
.
An
o
m
aly
d
etec
tio
n
in
tim
e
s
e
r
ies
d
ata
ev
o
lv
ed
f
r
o
m
th
e
cl
ass
ical
o
r
co
n
v
en
tio
n
al
ap
p
r
o
ac
h
,
wh
ich
u
s
es
th
e
d
if
f
er
e
n
ce
b
etwe
en
f
o
r
ec
asti
n
g
r
esu
lts
an
d
ac
tu
al
d
ata.
Ho
wev
er
,
th
e
wea
k
n
ess
es
f
o
u
n
d
h
a
v
e
led
to
th
e
d
ev
elo
p
m
e
n
t
o
f
v
ar
io
u
s
m
eth
o
d
s
,
in
clu
d
in
g
m
ac
h
in
e
l
ea
r
n
in
g
.
E
x
t
r
em
e
g
r
ad
ie
n
t
b
o
o
s
tin
g
(
XGBo
o
s
t)
as
an
ex
am
p
le
o
f
m
ac
h
i
n
e
lear
n
i
n
g
ca
n
h
an
d
le
n
o
n
-
lin
ea
r
ity
o
f
tim
e
s
er
ies
with
s
tr
o
n
g
lear
n
in
g
ca
p
ab
ilit
ies
[
3
]
.
Fu
r
th
er
m
o
r
e
,
lo
n
g
s
h
o
r
t
-
ter
m
m
em
o
r
y
(
L
STM
)
as
o
n
e
p
a
r
t
o
f
d
ee
p
lear
n
in
g
ca
n
ig
n
o
r
e
th
e
s
tatio
n
ar
ity
ass
u
m
p
tio
n
b
ec
au
s
e
it
h
as
th
e
ab
ilit
y
to
h
a
n
d
le
n
o
n
lin
ea
r
r
e
latio
n
s
h
ip
s
an
d
lar
g
e
d
im
en
s
i
o
n
s
[
4
]
.
I
n
ad
d
itio
n
,
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
26
:
2
5
4
3
-
2
5
5
3
2544
th
e
s
p
ec
if
ic
m
o
d
el
r
e
q
u
ir
ed
b
y
co
n
v
e
n
tio
n
al
m
eth
o
d
s
in
m
a
k
in
g
p
r
ed
ictio
n
s
is
n
o
t
n
ee
d
ed
i
n
th
e
u
s
e
o
f
L
STM
an
d
o
th
e
r
m
ac
h
in
e
lear
n
in
g
m
eth
o
d
s
s
o
th
at
th
ey
ca
n
ad
a
p
t b
etter
.
On
e
ex
am
p
le
o
f
an
an
o
m
aly
t
h
at
m
ay
o
cc
u
r
is
air
q
u
ality
.
A
ir
q
u
ality
is
co
m
m
o
n
ly
r
ep
r
esen
ted
b
y
a
n
a
ir
q
u
ality
in
d
ex
(
AQI
)
th
a
t
r
ef
lects
th
e
lev
el
o
f
air
p
o
llu
tio
n
.
Air
p
o
llu
tio
n
h
as
b
ec
o
m
e
a
m
ajo
r
en
v
ir
o
n
m
en
tal
is
s
u
e
th
at
is
r
e
ce
iv
in
g
in
cr
ea
s
in
g
p
u
b
lic
atten
tio
n
,
p
a
r
ticu
lar
ly
alo
n
g
s
id
e
co
n
ce
r
n
s
r
elate
d
to
clim
ate
ch
an
g
e.
Gh
o
s
h
et
a
l
.
[
5
]
esti
m
ate
th
at
5
.
8
m
illi
o
n
p
r
em
atu
r
e
b
i
r
th
s
in
2
0
1
9
wer
e
ca
u
s
ed
b
y
air
p
o
llu
tio
n
.
Mo
r
eo
v
er
,
air
p
o
llu
tio
n
is
k
n
o
wn
t
o
ca
u
s
e
v
ar
io
u
s
r
esp
ir
ato
r
y
d
is
ea
s
es,
in
c
lu
d
in
g
p
n
eu
m
o
n
ia,
asth
m
a,
an
d
ac
u
te
r
esp
ir
ato
r
y
in
f
ec
tio
n
s
(
AR
I
)
.
Data
in
d
icate
th
at
b
etwe
en
2
0
2
0
a
n
d
2
0
2
2
th
er
e
wer
e
7
3
,
6
9
4
ca
s
es
o
f
p
n
eu
m
o
n
ia
in
ch
ild
r
en
a
n
d
1
5
,
8
2
5
ca
s
es
o
f
asth
m
a
in
ch
ild
r
e
n
in
J
ak
ar
ta
[
6
]
.
I
n
ad
d
itio
n
,
i
n
th
e
f
ir
s
t
s
em
es
ter
o
f
2
0
2
3
alo
n
e,
6
3
8
,
2
9
1
ca
s
es
o
f
AR
I
wer
e
r
ep
o
r
ted
in
J
ak
ar
ta
[
7
]
.
Alla
an
d
Ad
ar
i
[
8
]
n
o
te
d
th
at
th
e
AQI
m
ea
s
u
r
ed
b
y
m
u
ltip
le
s
en
s
o
r
s
ac
r
o
s
s
d
if
f
er
en
t
lo
ca
tio
n
s
an
d
co
llected
p
er
io
d
ically
th
r
o
u
g
h
a
ce
n
tr
alize
d
s
y
s
tem
in
h
e
r
en
tly
f
o
r
m
tim
e
s
er
ies
d
ata.
Su
c
h
ti
m
e
s
er
ies
d
ata
ca
n
b
e
u
tili
ze
d
as
in
p
u
t
f
o
r
n
e
u
r
al
n
etwo
r
k
–
b
ased
m
o
d
els to
p
er
f
o
r
m
an
o
m
aly
d
etec
tio
n
u
s
in
g
a
tim
e
s
er
ies
-
b
ased
ap
p
r
o
ac
h
.
T
h
er
e
ar
e
m
an
y
ap
p
r
o
ac
h
es
to
d
etec
tin
g
an
o
m
alies
in
tim
e
s
er
ies
d
ata,
o
n
e
o
f
wh
ich
is
m
ac
h
in
e
lear
n
in
g
,
wh
ic
h
o
v
er
co
m
es
co
n
v
en
tio
n
al
m
eth
o
d
s
'
lim
itatio
n
s
.
W
an
g
et
a
l
.
[
9
]
u
s
ed
t
h
e
L
STM
m
eth
o
d
,
wh
ic
h
was
co
m
p
ar
ed
with
th
e
K
-
m
e
an
s
m
eth
o
d
o
n
elec
tr
ic
p
o
wer
co
n
s
u
m
p
tio
n
d
ata,
r
esu
ltin
g
in
th
e
L
STM
m
eth
o
d
b
ein
g
a
b
etter
m
eth
o
d
in
ter
m
s
o
f
r
ec
all.
Fu
r
th
er
m
o
r
e,
in
L
i
u
'
s
[
1
0
]
r
esear
ch
,
an
o
m
aly
d
et
ec
tio
n
was
ca
r
r
ied
o
u
t
o
n
wate
r
d
is
tr
ib
u
tio
n
s
y
s
tem
s
with
v
ar
io
u
s
e
n
s
em
b
le
m
eth
o
d
s
.
T
h
e
s
tu
d
y
s
h
o
w
ed
th
e
s
tab
ilit
y
o
f
XGBo
o
s
t
ev
en
th
o
u
g
h
it
wa
s
n
o
t
th
e
b
est
m
eth
o
d
.
B
o
th
m
eth
o
d
s
h
av
e
p
r
ev
io
u
s
ly
b
ee
n
co
m
p
ar
e
d
b
y
T
r
izo
g
lo
u
et
a
l
.
[
1
1
]
u
s
in
g
wi
n
d
tu
r
b
in
e
d
ata.
Ho
wev
e
r
,
th
e
u
s
e
o
f
b
o
th
m
eth
o
d
s
is
s
till
lim
ited
in
d
etec
tin
g
air
q
u
ality
an
o
m
alies,
esp
ec
ia
lly
in
la
b
eled
tim
e
s
er
ies
d
ata
,
s
o
th
e
ap
p
licatio
n
an
d
co
m
p
ar
is
o
n
o
f
m
et
h
o
d
s
u
n
d
er
th
ese
c
o
n
d
itio
n
s
is
a
n
o
v
elty
in
th
is
s
tu
d
y
.
T
h
er
ef
o
r
e,
th
is
s
tu
d
y
is
b
ased
o
n
s
ev
er
al
q
u
esti
o
n
s
f
r
o
m
p
r
ev
io
u
s
s
tu
d
ies,
n
am
el
y
h
o
w
L
STM
an
d
XGBo
o
s
t
p
er
f
o
r
m
in
d
etec
tin
g
an
o
m
alies
i
n
J
ak
ar
ta'
s
lab
eled
AQI
,
an
d
h
o
w
th
e
o
cc
u
r
r
e
n
ce
o
f
air
q
u
ality
an
o
m
alies
is
r
elate
d
to
m
eteo
r
o
lo
g
ical
f
ac
to
r
s
.
T
h
r
o
u
g
h
th
is
s
tu
d
y
,
th
is
r
esear
ch
aim
s
to
p
r
o
v
id
e
in
s
i
g
h
t
in
to
an
o
m
aly
p
atter
n
s
in
air
q
u
ality
d
ata
an
d
ev
alu
at
e
th
e
f
ea
s
ib
ilit
y
o
f
m
ac
h
in
e
lear
n
in
g
m
eth
o
d
s
in
s
u
p
p
o
r
tin
g
air
q
u
ality
m
o
n
ito
r
i
n
g
an
d
ea
r
ly
war
n
in
g
s
y
s
tem
s
.
2.
M
E
T
H
O
D
2
.
1
.
L
o
ng
s
ho
rt
-
t
er
m
m
emo
ry
L
STM
is
a
ty
p
e
o
f
n
eu
r
al
n
et
wo
r
k
d
ev
elo
p
e
d
o
f
r
ec
u
r
r
e
n
t
n
eu
r
al
n
etwo
r
k
(
R
NN)
b
y
Ho
ch
r
eiter
an
d
Sch
m
id
h
u
b
e
r
[
1
2
]
.
L
STM
h
as
in
ter
n
al
m
em
o
r
y
(
c)
an
d
m
u
lt
ip
licativ
e
g
ates,
n
am
ely
th
e
f
o
r
g
et
g
ate
(
f
)
,
in
p
u
t
g
ate
(
i)
,
an
d
o
u
tp
u
t
g
ate
(
o
)
,
to
o
v
er
co
m
e
v
a
n
is
h
in
g
o
r
e
x
p
lo
d
in
g
g
r
ad
ie
n
ts
in
u
p
d
atin
g
weig
h
ts
in
R
NN.
T
h
is
also
g
iv
es
L
STM
th
e
ab
ilit
y
to
p
r
o
ce
s
s
s
eq
u
en
tial
d
ata
an
d
s
to
r
e
in
f
o
r
m
atio
n
b
o
th
in
th
e
s
h
o
r
t
an
d
lo
n
g
ter
m
[
1
3
]
.
T
h
e
r
ef
o
r
e
,
L
STM
s
ar
e
wid
ely
u
s
ed
to
p
r
o
ce
s
s
tex
t a
n
d
tim
e
s
er
ies d
ata
[
1
4
]
.
E
ac
h
m
u
ltip
licativ
e
g
ate
o
f
th
e
L
STM
h
as
a
d
if
f
er
en
t
f
u
n
ctio
n
.
T
h
e
f
o
r
g
et
g
ate
s
o
r
ts
o
u
t
th
e
in
f
o
r
m
atio
n
to
b
e
s
to
r
ed
o
r
d
elete
d
,
th
e
in
p
u
t
g
ate
u
p
d
ates
th
e
in
f
o
r
m
atio
n
in
t
h
e
in
ter
n
al
m
em
o
r
y
,
an
d
th
e
o
u
tp
u
t g
ate
r
eg
u
lates th
e
o
u
tp
u
t o
f
in
f
o
r
m
atio
n
.
Ma
th
em
atic
ally
,
th
e
wh
o
le
p
r
o
ce
s
s
in
L
STM
wo
r
k
s
f
o
llo
win
g
(
1
)
to
(
6
)
,
wh
er
e
is
th
e
we
ig
h
t,
h
is
th
e
h
id
d
en
s
tate,
is
th
e
b
ias,
is
th
e
ac
tiv
atio
n
f
u
n
ctio
n
(
co
m
m
o
n
ly
s
ig
m
o
id
an
d
ta
n
h
)
,
an
d
t
h
e
'
⊙
'
o
p
er
ato
r
in
d
icate
s
th
e
m
u
ltip
li
ca
tio
n
o
f
two
v
ec
to
r
s
in
th
e
s
am
e
d
ir
ec
tio
n
[
1
5
]
.
=
(
ℎ
ℎ
−
1
+
+
)
(
1
)
=
(
ℎ
ℎ
−
1
+
+
)
(
2
)
̃
=
(
̃
ℎ
ℎ
−
1
+
̃
+
̃
)
(
3
)
=
⊙
−
1
+
⊙
̃
(
4
)
=
(
ℎ
ℎ
−
1
+
+
)
(
5
)
ℎ
=
⊙
(
)
(
6
)
2
.
2
.
E
x
t
re
m
e
g
r
a
dient
bo
o
s
t
ing
XGBo
o
s
t
is
a
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
d
ev
elo
p
ed
b
y
C
h
en
an
d
Gu
estrin
[
1
6
]
.
I
t
u
s
es
th
e
g
r
ad
ie
n
t
b
o
o
s
tin
g
m
eth
o
d
with
a
n
e
n
s
em
b
le
alg
o
r
ith
m
,
wh
ic
h
co
m
b
in
es
s
ev
er
al
al
g
o
r
ith
m
s
(
l
ea
r
n
er
s
)
to
im
p
r
o
v
e
ac
cu
r
ac
y
d
esig
n
ed
f
o
r
co
m
p
u
tatio
n
al
ef
f
icien
c
y
a
n
d
m
o
d
el
f
lex
ib
ilit
y
.
T
h
is
m
eth
o
d
o
p
tim
izes
g
r
ad
ien
t
b
o
o
s
tin
g
b
y
cr
ea
tin
g
a
s
eq
u
e
n
tial
d
ec
is
io
n
tr
ee
th
at
m
in
im
iz
es
er
r
o
r
.
At
ea
c
h
iter
atio
n
,
th
e
er
r
o
r
v
alu
e
will
b
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
Ma
ch
in
e
lea
r
n
in
g
a
p
p
r
o
a
ch
es fo
r
a
n
o
ma
ly
d
etec
tio
n
o
f J
a
ka
r
ta
…
(
Mu
h
a
mma
d
R
iz
ky
N
u
r
h
a
mb
a
li
)
2545
ca
lcu
lated
to
co
r
r
ec
t
th
e
b
asic
alg
o
r
ith
m
in
th
e
p
r
ev
io
u
s
iter
a
tio
n
[
1
7
]
,
[
1
8
]
.
T
h
e
p
r
ed
ictio
n
r
esu
lt
in
XGBo
o
s
t
is
o
b
tain
ed
b
ased
o
n
th
e
s
u
m
m
atio
n
o
f
all
r
esu
lts
as
s
h
o
wn
in
(
7
)
,
wh
er
e
is
th
e
s
p
ac
e
o
f
r
eg
r
ess
io
n
tr
ee
s
,
(
)
is
th
e
r
esu
lt
o
f
tr
ee
s
,
an
d
is
t
h
e
p
r
ed
ictio
n
v
alu
e
o
f
th
e
i
-
th
ex
am
p
le
f
o
r
.
T
h
e
v
alu
e
o
f
is
th
en
o
b
tain
ed
b
ased
o
n
(
7
)
.
̂
=
∑
(
)
=
1
,
∈
(
7
)
2
.
3
.
M
et
rics e
v
a
lua
t
io
n
E
v
alu
atio
n
m
etr
ics
ar
e
m
ea
s
u
r
es
u
s
ed
to
d
eter
m
in
e
th
e
g
o
o
d
n
ess
o
f
a
m
o
d
el,
a
n
d
f
o
r
ec
asti
n
g
an
d
class
if
icatio
n
ar
e
n
o
ex
ce
p
tio
n
.
Stan
d
ar
d
e
v
alu
atio
n
m
etr
ic
s
u
s
ed
in
f
o
r
ec
asti
n
g
ar
e
r
o
o
t
m
ea
n
s
q
u
ar
e
er
r
o
r
(
R
MSE
)
an
d
m
ea
n
ab
s
o
lu
te
p
er
ce
n
tag
e
er
r
o
r
(
MA
PE)
.
L
o
wer
R
MSE
an
d
MA
PE
v
alu
es
r
ef
lect
h
ig
h
er
f
o
r
ec
asti
n
g
ac
cu
r
ac
y
a
n
d
i
n
d
i
ca
te
r
eliab
le
p
r
e
d
icted
v
alu
es.
R
MSE
an
d
MA
PE
ar
e
ca
lcu
lated
b
ased
o
n
(
8
)
an
d
(
9
)
wh
er
e
is
t
-
tim
e
v
al
u
e,
is
t
-
tim
e
f
o
r
ec
ast
v
al
u
e,
an
d
is
n
u
m
b
er
o
f
f
o
r
ec
ast
in
g
p
e
r
io
d
s
.
Fu
r
th
er
m
o
r
e
,
th
e
ev
alu
atio
n
m
etr
ic
u
s
ed
f
o
r
ca
lcu
latin
g
t
h
e
ac
cu
r
ac
y
o
f
class
if
icatio
n
r
esu
lts
i
s
b
alan
ce
d
ac
cu
r
ac
y
(
B
AC
C
)
.
Acc
o
r
d
in
g
to
B
ej
et
a
l
.
[
1
9
]
,
B
AC
C
is
u
s
ed
b
ec
au
s
e
o
f
th
e
im
b
alan
c
e
o
f
d
ata
b
etwe
en
an
o
m
aly
an
d
n
o
n
-
an
o
m
aly
cl
ass
es,
s
o
its
u
s
e
will
b
e
m
o
r
e
m
ea
n
in
g
f
u
l
th
an
ac
cu
r
ac
y
.
T
h
e
B
AC
C
v
alu
e
is
ca
lcu
lated
u
s
in
g
(
1
0
)
w
h
er
e
tr
u
e
p
o
s
itiv
e
(
TP
)
is
a
p
o
s
itiv
e
c
lass
th
at
is
co
r
r
ec
tly
class
if
ied
as
a
p
o
s
itiv
e
class
,
tr
u
e
n
e
g
ativ
e
(
TN
)
is
a
n
eg
ati
v
e
class
th
at
is
co
r
r
ec
tly
class
if
ied
as
a
n
eg
ativ
e
class
,
f
alse
p
o
s
itiv
e
(
FP
)
is
a
n
eg
ativ
e
class
th
at
is
class
if
ie
d
as
a
p
o
s
itiv
e
class
,
an
d
f
alse
n
eg
ativ
e
(
FN
)
is
a
p
o
s
itiv
e
class
th
at
is
clas
s
if
ied
as a
n
eg
ativ
e
class
.
=
∑
(
(
−
̂
)
2
)
1
2
=
1
(
8
)
=
1
∑
|
−
̂
|
=
1
×
100%
(
9
)
=
1
2
(
+
+
+
)
(
1
0
)
2
.
4
.
Da
t
a
T
h
e
AQI
in
I
n
d
o
n
esia
is
m
e
asu
r
ed
u
s
in
g
two
a
p
p
r
o
ac
h
es,
n
am
ely
th
e
I
n
d
o
n
esian
a
n
d
U
.
S
.
AQI
.
Ho
wev
er
,
ac
c
o
r
d
in
g
t
o
Pra
m
a
n
a
et
a
l
.
[
2
0
]
,
th
e
U
.
S
.
AQI
h
as
b
ee
n
estab
lis
h
ed
g
lo
b
ally
a
n
d
is
alig
n
ed
with
th
e
W
o
r
ld
Hea
lth
O
r
g
an
izatio
n
(
W
HO)
air
q
u
ality
g
u
id
elin
e
s
.
T
h
er
ef
o
r
e,
th
is
s
tu
d
y
will
r
e
f
er
to
th
e
u
s
e
o
f
th
e
U
.
S
.
AQI
,
wh
ich
h
as
s
ix
le
v
e
ls
:
g
o
o
d
(0
-
5
0
)
,
m
o
d
er
ate
(
5
1
-
1
0
0
)
,
u
n
h
ea
lth
y
f
o
r
s
en
s
itiv
e
g
r
o
u
p
s
(
1
0
1
-
1
5
0
)
,
u
n
h
ea
lth
y
(
1
5
1
-
2
0
0
)
,
v
er
y
u
n
h
ea
lth
y
(
2
0
1
-
3
0
0
)
,
an
d
h
az
ar
d
o
u
s
(
>3
0
0
)
.
T
h
e
d
ata
u
s
ed
ar
e
s
o
u
r
ce
d
f
r
o
m
Air
No
w
(
www.
air
n
o
w.
g
o
v
)
[
2
1
]
an
d
v
e
r
s
io
n
2
.
3
.
6
o
f
th
e
Pre
d
ictio
n
o
f
W
o
r
ld
wid
e
E
n
er
g
y
R
eso
u
r
c
es
(
POW
E
R
)
p
r
o
ject
p
er
h
o
u
r
o
n
2
0
2
4
/
0
6
/1
6
(
h
ttp
s
://p
o
wer
.
l
ar
c.
n
asa.
g
o
v
/d
ata
-
ac
ce
s
s
-
v
iewe
r
/)
[
2
2
]
.
Data
s
o
u
r
ce
d
f
r
o
m
Air
No
w
is
th
e
r
es
u
lt
o
f
PM2
.
5
m
ea
s
u
r
em
en
ts
at
th
e
U.
S.
E
m
b
ass
y
Air
Qu
ality
Mo
n
ito
r
in
g
Statio
n
in
C
en
tr
al
J
ak
ar
ta
as
r
e
s
p
o
n
s
e
v
ar
iab
les.
Fu
r
th
er
m
o
r
e
,
d
ata
s
o
u
r
ce
d
f
r
o
m
NASA
PO
W
E
R
i
s
m
eteo
r
o
lo
g
ical
d
ata
in
th
e
f
o
r
m
o
f
tem
p
er
atu
r
e
,
r
ain
f
all,
h
u
m
id
ity
,
win
d
s
p
ee
d
,
win
d
d
ir
ec
tio
n
,
an
d
s
u
r
f
ac
e
p
r
ess
u
r
e
as
ex
p
lan
ato
r
y
v
ar
ia
b
les.
T
h
ese
v
ar
iab
les
ar
e
m
o
r
e
clea
r
ly
s
h
o
wn
in
T
ab
le
1
.
E
ac
h
d
ata
u
s
ed
h
as
a
tim
e
u
n
it
in
th
e
f
o
r
m
o
f
h
o
u
r
s
u
s
in
g
W
ester
n
I
n
d
o
n
esian
t
im
e
(
UT
C
+7
)
with
a
tim
e
r
an
g
e
o
f
J
an
u
ar
y
1
,
2
0
1
8
,
at
0
0
:
0
0
u
n
til
Dec
em
b
er
3
1
,
2
0
2
3
,
at
2
3
:
0
0
,
a
n
d
is
d
o
wn
l
o
ad
ed
m
a
n
u
ally
o
n
th
e
av
ailab
le
s
ite.
T
ab
le
1
.
L
is
t o
f
v
ar
iab
les u
s
ed
Ty
p
e
s
o
f
m
e
t
e
o
r
o
l
o
g
i
c
a
l
v
a
r
i
a
b
l
e
s
V
a
r
i
a
b
l
e
(
c
o
d
e
)
U
n
i
t
AQI
H
u
mi
d
i
t
y
/
p
r
e
c
i
p
i
t
a
t
i
o
n
S
p
e
c
i
f
i
c
h
u
m
i
d
i
t
y
a
t
2
me
t
e
r
s (Q
V
2
M
)
g
/
k
g
R
e
l
a
t
i
v
e
h
u
mi
d
i
t
y
a
t
2
me
t
e
r
s
(
R
H
2
M
)
%
P
r
e
c
i
p
i
t
a
t
i
o
n
c
o
r
r
e
c
t
e
d
(
P
R
EC
TO
T
C
O
R
R
)
mm
/
h
o
u
r
W
i
n
d
/
p
r
e
ss
u
r
e
W
i
n
d
s
p
e
e
d
a
t
1
0
me
t
e
r
s (W
S
1
0
M
)
m/
s
W
i
n
d
s
p
e
e
d
a
t
5
0
me
t
e
r
s (W
S
5
0
M
)
m/
s
W
i
n
d
d
i
r
e
c
t
i
o
n
a
t
1
0
met
e
r
s (W
D
1
0
M
)
D
e
g
r
e
e
s
W
i
n
d
d
i
r
e
c
t
i
o
n
a
t
5
0
met
e
r
s (W
D
5
0
M
)
D
e
g
r
e
e
s
S
u
r
f
a
c
e
p
r
e
ssu
r
e
(
P
S
)
k
P
a
Te
mp
e
r
a
t
u
r
e
D
e
w
/
f
r
o
st
p
o
i
n
t
a
t
2
m
e
t
e
r
s (T
2
M
D
E
W
)
°
C
Te
mp
e
r
a
t
u
r
e
a
t
2
m
e
t
e
r
s
(
T2
M
)
°C
W
e
t
b
u
l
b
t
e
m
p
e
r
a
t
u
r
e
a
t
2
m
e
t
e
r
s (T
2
M
W
ET)
°C
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
26
:
2
5
4
3
-
2
5
5
3
2546
2
.
5
.
Da
t
a
a
na
ly
s
is
pro
ce
du
r
es
T
h
e
R
an
d
Py
th
o
n
s
o
f
twar
es
wer
e
u
s
ed
in
d
ata
an
aly
s
is
ac
co
r
d
in
g
to
th
e
f
lo
w
in
Fig
u
r
e
1
.
T
h
e
an
aly
s
is
b
eg
in
s
with
d
ata
im
p
u
tatio
n
b
ec
au
s
e
th
e
AQI
d
ata
o
b
tain
ed
h
as
2
,
2
8
1
b
lan
k
v
alu
es.
T
h
e
i
m
p
u
tatio
n
m
eth
o
d
u
s
ed
is
ad
ju
s
ted
to
th
e
am
o
u
n
t
o
f
s
eq
u
e
n
tial
m
is
s
in
g
d
ata
with
th
e
cr
iter
ia:
i
)
if
m
is
s
in
g
v
alu
e
at
o
n
e
tim
e,
u
s
in
g
lin
ea
r
in
te
r
p
o
latio
n
;
ii
)
if
m
is
s
in
g
v
alu
e
≤
4
co
n
s
ec
u
tiv
e
h
o
u
r
s
,
u
s
in
g
s
ea
s
o
n
a
l
d
ec
o
m
p
o
s
itio
n
o
r
s
ea
s
o
n
al
s
p
litt
in
g
;
an
d
iii
)
if
m
is
s
in
g
v
alu
e
>4
co
n
s
ec
u
tiv
e
h
o
u
r
s
,
u
s
in
g
d
is
ag
g
r
e
g
atio
n
o
r
L
STM
.
T
h
e
co
m
p
lete
d
ata
was
th
e
n
ex
p
l
o
r
ed
u
s
in
g
tim
e
s
er
i
es
p
lo
ts
an
d
co
r
r
elatio
n
p
lo
ts
to
o
b
tain
d
ata
ch
ar
ac
ter
is
tics
an
d
p
atter
n
s
.
T
h
en
,
th
e
u
n
lab
eled
AQI
d
ata
is
lab
eled
with
m
o
v
in
g
r
an
g
es o
f
len
g
th
2
(
MR
(
2
)
)
an
d
3
(
MR
(
3
)
)
,
wh
ic
h
ar
e
c
o
m
b
in
ed
with
th
e
3
-
s
ig
m
a
an
d
4
-
s
ig
m
a
r
u
les.
Af
ter
lab
elin
g
t
h
e
d
ata,
two
ty
p
es o
f
class
es
wi
ll
b
e
o
b
tain
ed
:
a
n
o
m
alies
an
d
n
o
n
-
an
o
m
alies.
T
h
e
an
o
m
aly
v
alu
es
wer
e
r
ep
lac
ed
with
th
e
a
v
er
ag
e
o
f
two
n
ea
r
b
y
d
ata,
s
u
ch
as p
e
r
f
o
r
m
in
g
lin
ea
r
i
n
ter
p
o
latio
n
to
f
ill in
m
is
s
in
g
v
alu
es.
Nex
t,
th
e
d
ata
is
s
p
lit in
to
an
n
u
al
d
ata
a
n
d
u
s
ed
f
o
r
m
o
d
e
lin
g
.
Fig
u
r
e
1
.
Flo
wch
ar
t
o
f
d
ata
an
aly
s
is
p
r
o
ce
d
u
r
e
L
STM
ar
ch
itectu
r
e
is
f
o
r
m
ed
with
two
m
ain
lay
er
s
a
n
d
a
d
en
s
e
lay
er
.
T
h
e
h
y
p
er
p
ar
am
eter
v
alu
es
u
s
ed
in
L
STM
wer
e
d
eter
m
in
ed
b
y
b
atc
h
s
ize
(
7
2
)
,
ep
o
c
h
(
5
0
)
,
o
p
tim
izer
(
Ad
am
)
,
a
n
d
le
ar
n
in
g
r
ate
(
0
.
0
0
1
)
.
T
h
e
b
atch
s
ize
an
d
ep
o
ch
v
al
u
es
ar
e
d
eter
m
in
ed
s
u
b
jectiv
ely
b
y
co
n
s
id
er
in
g
c
o
m
p
u
tatio
n
tim
e.
Ho
wev
er
,
th
e
o
p
tim
izer
an
d
lear
n
in
g
r
ate
v
alu
es
ar
e
d
eter
m
in
ed
u
s
in
g
s
lid
in
g
win
d
o
w
an
d
e
x
p
an
d
i
n
g
win
d
o
w
cr
o
s
s
-
v
alid
atio
n
.
I
n
th
e
s
am
e
way
,
th
e
b
est
XGBo
o
s
t
h
y
p
er
p
ar
am
eter
s
u
s
ed
wer
e
p
r
ev
io
u
s
ly
s
ea
r
ch
ed
u
s
in
g
s
lid
in
g
win
d
o
w
an
d
ex
p
an
d
i
n
g
win
d
o
w
c
r
o
s
s
-
v
alid
atio
n
,
b
u
t
n
o
r
esu
lts
wer
e
f
o
u
n
d
th
at
o
p
tim
ized
m
o
d
el
p
er
f
o
r
m
an
ce
.
T
h
er
ef
o
r
e,
h
y
p
e
r
p
ar
am
eter
s
wer
e
u
s
ed
in
Attaallah
an
d
Kh
a
n
[
2
3
]
r
esear
c
h
th
at
m
o
d
eled
ai
r
q
u
ality
with
v
ar
io
u
s
m
ac
h
in
e
l
ea
r
n
in
g
a
p
p
r
o
ac
h
es.
Valid
atio
n
is
p
er
f
o
r
m
e
d
f
o
r
e
ac
h
an
n
u
al
d
ataset
u
s
in
g
th
e
L
STM
an
d
XGBo
o
s
t
m
o
d
els.
T
h
e
er
r
o
r
(
d
if
f
er
e
n
ce
b
etwe
en
th
e
ac
t
u
al
an
d
f
o
r
ec
ast
v
alu
es)
was
ca
lc
u
lated
to
o
b
tain
th
e
m
et
h
o
d
'
s
ac
cu
r
ac
y
a
n
d
cr
ea
te
an
er
r
o
r
th
r
esh
o
l
d
.
C
lass
if
icati
o
n
is
p
er
f
o
r
m
ed
o
n
all
th
e
d
at
a
b
ased
o
n
th
e
th
r
esh
o
ld
o
b
tain
ed
.
T
h
e
d
ata
will
b
e
class
if
ied
as a
n
an
o
m
aly
if
it e
x
ce
ed
s
th
e
th
r
esh
o
ld
.
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
3
.
1
.
Da
t
a
e
x
plo
ra
t
io
n
T
h
e
co
m
p
lete
AQI
d
ata
was
th
en
ex
p
lo
r
ed
.
Descr
ip
tiv
e
s
tatis
tics
s
h
o
w
th
at
th
e
av
er
ag
e
AQI
f
o
r
2018
-
2
0
2
3
is
9
8
.
1
5
,
wh
ic
h
is
ca
teg
o
r
ized
as
m
o
d
er
ate
with
a
s
p
r
ea
d
o
f
3
9
.
1
3
8
8
f
r
o
m
th
e
m
ea
n
.
T
h
e
J
ak
ar
ta
AQI
r
ea
ch
ed
th
e
h
ig
h
est
v
alu
e
o
f
2
1
9
o
n
Feb
r
u
ar
y
1
4
,
2
0
1
9
,
at
0
4
:0
0
,
wh
ich
in
d
icate
s
u
n
h
ea
lth
y
co
n
d
itio
n
in
J
ak
ar
ta.
Me
an
wh
ile,
th
e
lo
west
AQI
v
alu
e
o
cc
u
r
r
ed
o
n
Ma
r
ch
2
7
,
2
0
1
8
,
at
0
2
:0
0
with
a
v
alu
e
o
f
1
(
g
o
o
d
)
,
in
d
icatin
g
clea
n
er
an
d
h
ea
lth
ie
r
air
.
Fig
u
r
e
2
s
h
o
ws
th
at
th
e
AQI
h
as
a
s
im
ilar
p
atter
n
ev
er
y
y
e
ar
.
T
h
e
AQI
v
alu
e
ten
d
s
to
r
e
ac
h
its
p
ea
k
in
th
e
m
o
r
n
i
n
g
a
r
o
u
n
d
0
9
:0
0
an
d
lo
west
in
th
e
e
v
en
in
g
ar
o
u
n
d
1
9
:0
0
.
T
h
e
h
ig
h
m
o
r
n
in
g
AQI
m
ay
b
e
d
u
e
t
o
th
e
h
ig
h
m
o
b
ilit
y
o
f
p
eo
p
le
in
th
e
m
o
r
n
in
g
wh
o
u
s
e
v
eh
icle
s
to
wo
r
k
o
r
g
o
to
s
ch
o
o
l.
Af
t
er
th
e
p
ea
k
p
h
ase,
th
e
AQI
d
ec
r
ea
s
es
p
er
io
d
icall
y
an
d
is
h
ig
h
at
n
ig
h
t.
Acc
o
r
d
in
g
t
o
Z
h
e
n
g
et
a
l.
[
2
4
]
,
th
e
r
e
is
a
r
elatio
n
s
h
ip
b
etwe
en
em
is
s
io
n
s
an
d
tr
af
f
ic
in
th
e
v
ar
iatio
n
o
f
AQI
wh
e
r
e
AQI
at
n
ig
h
t
is
in
f
lu
e
n
ce
d
b
y
th
e
in
c
r
ea
s
e
in
em
is
s
io
n
s
f
r
o
m
ea
r
th
h
ea
tin
g
,
v
eh
icles,
an
d
co
r
r
esp
o
n
d
in
g
p
ar
ticles ac
cu
m
u
latio
n
.
Fig
u
r
e
3
s
h
o
ws
th
e
co
ef
f
icie
n
t
o
f
c
o
r
r
elatio
n
v
alu
es
b
etw
ee
n
two
v
ar
ia
b
les
u
s
in
g
th
e
“
Sp
ea
r
m
an
”
m
eth
o
d
.
T
h
e
AQI
h
as a
n
eg
ati
v
e
co
r
r
elatio
n
with
alm
o
s
t a
ll v
ar
iab
les,
ex
ce
p
t PS a
n
d
T
2
M
h
eig
h
t.
I
n
a
d
d
itio
n
,
th
er
e
ar
e
r
elate
d
ex
p
la
n
ato
r
y
v
ar
iab
les,
n
am
ely
r
ain
f
all
(
PR
E
C
T
OT
C
OR
R
)
,
h
u
m
id
ity
(
QV2
M
an
d
R
H2
M)
,
d
ew
p
o
in
t
(
T
2
MD
E
W
)
,
a
n
d
w
in
d
d
ir
ec
tio
n
(
W
D1
0
M
an
d
W
D5
0
M)
.
T
h
ese
r
esu
lts
ar
e
in
a
cc
o
r
d
an
ce
with
th
e
r
esear
ch
b
y
T
ian
et
a
l
.
[
2
5
]
a
n
d
Han
d
h
ay
an
i'
s
[
2
6
]
,
wh
ich
d
is
cu
s
s
es
th
e
r
elatio
n
s
h
ip
b
etwe
en
m
eteo
r
o
l
o
g
ical
f
ac
to
r
s
an
d
air
q
u
ality
.
E
v
e
n
f
u
r
th
er
,
Han
d
h
ay
an
i'
s
[
2
6
]
m
e
n
tio
n
ed
th
at
th
e
r
elatio
n
s
h
ip
b
etwe
en
r
ain
f
all
an
d
air
q
u
ality
ca
n
b
e
u
tili
ze
d
to
m
ak
e
ar
tific
ial
r
ain
to
r
ed
u
ce
t
h
e
AQI
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
Ma
ch
in
e
lea
r
n
in
g
a
p
p
r
o
a
ch
es fo
r
a
n
o
ma
ly
d
etec
tio
n
o
f J
a
ka
r
ta
…
(
Mu
h
a
mma
d
R
iz
ky
N
u
r
h
a
mb
a
li
)
2547
Fig
u
r
e
2
.
Av
e
r
ag
e
AQI
f
o
r
2
4
h
o
u
r
s
p
e
r
y
ea
r
Fig
u
r
e
3
.
C
o
r
r
elatio
n
co
ef
f
icien
t
p
lo
t o
f
v
ar
iab
les
3
.
2
.
Va
lid
a
t
i
o
n m
o
del
Fig
u
r
e
4
s
h
o
ws
th
e
v
alid
atio
n
ac
cu
r
ac
y
o
f
L
STM
an
d
XGBo
o
s
t
b
ased
o
n
th
e
co
r
r
esp
o
n
d
i
n
g
MA
PE
an
d
R
MSE
v
alu
es.
T
h
e
av
er
ag
e
MA
PE
an
d
R
MSE
f
o
r
L
ST
M
v
alid
atio
n
r
esu
lts
ar
e
1
0
.
3
8
4
0
% a
n
d
1
0
.
5
9
1
3
.
I
t
is
b
etter
th
an
XGBo
o
s
t,
wh
ich
h
as
an
av
er
ag
e
MA
PE
o
f
2
3
.
4
2
6
7
%
an
d
a
R
MSE
o
f
2
1
.
3
0
8
2
.
T
h
e
l
o
w
MA
PE
an
d
R
MSE
v
alu
es
in
d
icate
th
at
L
STM
is
ab
le
an
d
b
etter
at
ca
p
tu
r
in
g
p
atter
n
s
in
th
e
d
ata
t
h
an
XGBo
o
s
t.
T
h
is
co
n
d
itio
n
m
a
y
also
in
d
icate
th
at
in
th
e
ca
s
e
o
f
th
e
AQI
,
XG
B
o
o
s
t
is
u
n
ab
le
to
ca
p
tu
r
e
n
o
n
-
lin
ea
r
p
atter
n
s
as
in
th
e
r
esear
c
h
o
f
R
ah
m
an
et
a
l
.
[
2
7
]
o
n
d
ata
o
n
C
OVI
D
-
1
9
ca
s
es
in
B
an
g
lad
esh
wh
ich
s
h
o
ws
XGBo
o
s
t
is
n
o
b
etter
th
an
th
e
au
to
r
eg
r
ess
iv
e
in
teg
r
ated
m
o
v
i
n
g
av
er
a
g
e
(
AR
I
MA
)
m
eth
o
d
.
I
n
ad
d
itio
n
,
b
ased
o
n
r
esear
ch
b
y
L
v
et
a
l
.
[
2
8
]
,
XGBo
o
s
t is m
o
r
e
s
tab
le
f
o
r
s
tatio
n
ar
y
d
ata
wi
th
o
u
t o
u
tlier
s
.
T
h
e
ad
d
itio
n
o
f
n
ew
ex
p
lan
at
o
r
y
v
ar
ia
b
les
was
tr
ied
to
im
p
r
o
v
e
th
e
ac
cu
r
ac
y
o
f
XGBo
o
s
t.
T
h
e
n
ew
ex
p
lan
ato
r
y
v
ar
iab
les
ar
e
tim
e
v
ar
iab
les
(
h
o
u
r
,
d
ay
,
an
d
m
o
n
th
)
an
d
n
u
m
er
ical
d
escr
ip
tiv
e
s
tatis
t
ics,
s
u
ch
a
s
s
u
m
,
m
ea
n
,
an
d
q
u
ar
tiles
o
f
r
esp
o
n
s
e
v
ar
iab
les
u
s
in
g
a
r
o
l
lin
g
win
d
o
w
f
o
r
2
4
h
o
u
r
s
.
T
h
e
r
esu
lts
s
h
o
w
a
n
im
p
r
o
v
em
e
n
t
in
t
h
e
MA
PE
an
d
R
MSE
v
alu
es
o
f
XGBo
o
s
t,
d
esp
ite
a
d
ec
r
ea
s
e
in
L
STM
.
T
h
e
d
ec
r
ea
s
e
in
th
e
ac
cu
r
ac
y
o
f
th
e
L
STM
co
u
ld
b
e
d
u
e
to
th
e
lo
s
s
o
f
s
o
m
e
in
f
o
r
m
atio
n
d
u
e
t
o
th
e
u
s
e
o
f
a
r
o
llin
g
win
d
o
w
f
o
r
t
h
e
d
escr
ip
tiv
e
s
tatis
tic
s
o
f
th
e
AQI
.
I
n
ad
d
itio
n
,
th
e
d
ec
r
ea
s
e
in
MA
PE
an
d
R
MSE
v
alu
es
o
f
L
STM
m
ay
b
e
an
in
d
icatio
n
o
f
an
o
m
alies in
th
e
d
ata.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
26
:
2
5
4
3
-
2
5
5
3
2548
Fig
u
r
e
4
.
Valid
atio
n
ac
cu
r
ac
y
o
f
L
STM
an
d
XGBo
o
s
t
3
.
3
.
Ano
m
a
ly
det
ec
t
io
n
3
.
3
.
1
.
I
nfluence
s
o
f
m
o
v
ing
r
a
ng
e
I
n
an
o
m
aly
d
etec
tio
n
,
MR
p
la
y
s
m
o
r
e
o
f
a
r
o
le
i
n
th
e
ac
tu
al
d
ata
lab
elin
g
p
r
o
ce
s
s
.
T
h
e
AQI
d
o
es
n
o
t
h
av
e
a
d
e
f
in
ite
lab
el,
s
o
t
h
e
MR
o
f
len
g
th
2
(
MR
(
2
)
)
a
n
d
3
(
MR
(
3
)
)
a
p
p
r
o
ac
h
es
a
r
e
u
s
ed
.
MR
(
2
)
clea
r
ly
ca
p
tu
r
es
two
co
n
s
ec
u
tiv
e
o
b
s
er
v
atio
n
s
as
m
in
im
u
m
an
d
m
ax
im
u
m
,
b
u
t
MR
(
3
)
is
n
o
t
n
ec
ess
ar
ily
o
b
tain
ed
f
r
o
m
two
co
n
s
ec
u
tiv
e
o
b
s
er
v
a
tio
n
s
.
Fu
r
th
er
m
o
r
e
,
ac
co
r
d
i
n
g
to
W
o
o
d
all
an
d
Mo
n
tg
o
m
er
y
[
2
9
]
,
u
s
in
g
MR
o
f
len
g
th
m
o
r
e
th
a
n
2
will
in
cr
ea
s
e
b
ias.
T
h
er
ef
o
r
e,
lab
elin
g
th
e
ac
tu
al
d
ata
with
MR
(
2
)
ca
n
ca
p
tu
r
e
h
ig
h
in
ter
tem
p
o
r
al
v
alu
e
ch
a
n
g
es c
o
m
p
ar
ed
to
MR (
3
)
.
T
ab
le
2
g
en
er
ally
s
h
o
ws
th
at
lab
elin
g
with
MR
(
2
)
i
s
m
o
r
e
s
u
itab
le
th
an
MR
(
3
)
f
o
r
th
e
L
STM
m
eth
o
d
.
Ho
wev
er
,
f
o
r
th
e
XG
B
o
o
s
t
m
eth
o
d
(
as
s
h
o
w
n
in
T
a
b
le
3
)
,
th
e
MR
(
3
)
lab
eli
n
g
m
e
th
o
d
l
o
o
k
s
s
lig
h
tly
b
etter
th
an
MR
(
2
)
o
n
av
er
ag
e.
T
h
is
is
in
d
icate
d
b
y
th
e
h
ig
h
er
class
if
icatio
n
ac
cu
r
ac
y
.
T
h
e
a
v
er
ag
e
class
if
icatio
n
ac
cu
r
ac
y
o
f
L
STM
with
MR
(
2
)
is
0
.
9
9
2
4
,
w
h
ile
with
MR
(
3
)
is
0
.
6
9
3
3
.
T
h
en
,
with
XGBo
o
s
t,
th
e
av
er
ag
e
class
if
icatio
n
ac
cu
r
ac
y
with
MR
(
2
)
is
0
.
5
1
0
1
an
d
MR
(
3
)
is
0
.
6
3
4
9
.
Alth
o
u
g
h
th
e
r
e
is
a
class
if
icatio
n
ac
cu
r
ac
y
v
alu
e
o
f
1
.
0
0
,
it
is
d
u
e
to
th
e
a
b
s
en
ce
o
f
o
b
s
er
v
atio
n
s
lab
eled
as
an
o
m
aly
class
es
in
th
e
ac
tu
al
d
ata
an
d
th
e
d
etec
tio
n
r
esu
lts
also
s
h
o
w
s
im
i
lar
r
esu
lts
.
B
a
s
ed
o
n
th
ese
r
esu
lts
,
L
STM
o
u
tp
er
f
o
r
m
s
XGBo
o
s
t to
ca
p
tu
r
e
f
lu
ctu
atin
g
d
ata.
3
.
3
.
2
.
I
nfluence
s
o
f
s
ig
m
a
rules
T
h
e
s
ig
m
a
r
u
le
in
a
n
o
m
aly
d
e
tectio
n
will
af
f
ec
t
th
e
an
o
m
al
y
lim
it
r
an
g
e
as
a
th
r
esh
o
ld
.
As
a
r
esu
lt,
th
e
n
u
m
b
er
o
f
an
o
m
alies
in
th
e
d
ata
also
ch
a
n
g
es.
T
h
e
s
m
aller
th
e
s
ig
m
a
v
alu
e
u
s
ed
,
th
e
n
ar
r
o
wer
t
h
e
non
-
a
n
o
m
aly
r
eg
i
o
n
will
b
e,
ca
u
s
in
g
a
l
o
t
o
f
d
ata
to
f
all
o
u
ts
id
e
th
e
in
ter
v
al
an
d
v
ice
v
er
s
a.
Ho
wev
er
,
n
o
t
o
n
ly
th
at,
b
u
t
th
e
u
s
e
o
f
th
e
s
ig
m
a
r
u
le
ca
n
also
af
f
ec
t
t
h
e
class
if
icatio
n
ac
cu
r
ac
y
in
an
o
m
aly
d
etec
tio
n
.
T
ab
les
2
an
d
3
s
h
o
ws
th
e
cl
ass
if
icatio
n
ac
cu
r
ac
y
o
f
th
e
L
STM
an
d
XGBo
o
s
t
m
eth
o
d
is
b
etter
u
s
in
g
th
e
4
-
s
ig
m
a
r
u
le.
T
h
e
av
er
a
g
e
c
lass
if
icatio
n
ac
cu
r
ac
y
o
f
L
S
T
M
with
3
an
d
4
-
s
ig
m
a
is
0
.
7
4
8
5
an
d
0
.
8
6
7
2
r
esp
ec
tiv
ely
,
wh
ile
XGBo
o
s
t
h
as
class
if
icatio
n
ac
cu
r
ac
y
v
a
lu
es
with
3
an
d
4
-
s
ig
m
a
o
f
0
.
5
0
9
6
a
n
d
0
.
6
4
0
8
r
esp
ec
tiv
ely
.
T
h
ese
r
esu
lts
ar
e
s
till
in
lin
e
w
ith
th
e
r
esear
ch
o
f
Z
h
en
g
et
a
l
.
[
3
0
]
,
wh
e
r
e
th
e
u
s
e
o
f
3
an
d
4
-
s
ig
m
a
is
th
e
b
est v
alu
e
am
o
n
g
o
th
e
r
v
alu
es (
1
,
2
,
5
,
6
,
an
d
7
)
.
T
ab
le
2
.
C
lass
if
icatio
n
ev
alu
atio
n
f
o
r
L
STM
m
eth
o
d
s
Y
e
a
r
M
R
(
2
)
M
R
(
3
)
3
-
S
i
g
ma
4
-
S
i
g
ma
3
-
S
i
g
ma
4
-
S
i
g
ma
FE
WFE
FE
WFE
FE
WFE
FE
WFE
2
0
1
8
0
.
9
0
9
7
0
.
9
3
1
3
0
.
9
3
0
3
0
.
9
7
1
8
0
.
5
4
4
1
0
.
5
9
1
6
0
.
9
9
9
3
0
.
9
9
9
7
2
0
1
9
0
.
9
2
9
2
0
.
9
4
2
6
0
.
8
6
1
7
0
.
8
8
6
7
0
.
6
0
0
4
0
.
5
6
6
3
0
.
9
9
9
5
0
.
9
9
9
3
2
0
2
0
0
.
8
8
2
8
0
.
9
2
7
5
0
.
9
0
4
5
0
.
9
6
4
1
0
.
5
6
1
2
0
.
5
7
7
6
0
.
6
6
6
2
0
.
9
9
9
7
2
0
2
1
0
.
8
6
3
3
0
.
9
4
9
9
0
.
9
3
0
2
0
.
9
7
2
0
0
.
5
2
4
5
0
.
5
3
9
1
0
.
4
9
9
6
0
.
4
9
9
5
2
0
2
2
0
.
8
8
0
0
0
.
9
1
7
8
0
.
9
4
4
0
0
.
9
4
5
4
0
.
6
2
9
0
0
.
6
4
3
2
0
.
4
9
8
9
0
.
4
9
3
2
2
0
2
3
0
.
9
2
0
7
0
.
9
2
0
5
0
.
9
3
8
6
0
.
9
1
4
1
0
.
6
0
0
8
0
.
6
0
9
8
0
.
9
9
7
8
0
.
9
9
7
7
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
Ma
ch
in
e
lea
r
n
in
g
a
p
p
r
o
a
ch
es fo
r
a
n
o
ma
ly
d
etec
tio
n
o
f J
a
ka
r
ta
…
(
Mu
h
a
mma
d
R
iz
ky
N
u
r
h
a
mb
a
li
)
2549
T
ab
le
3
.
C
lass
if
icatio
n
ev
alu
atio
n
f
o
r
XGBo
o
s
t m
eth
o
d
s
Y
e
a
r
M
R
(
2
)
M
R
(
3
)
3
-
S
i
g
ma
4
-
S
i
g
ma
3
-
S
i
g
ma
4
-
S
i
g
ma
FE
WFE
FE
WFE
FE
WFE
FE
WFE
2
0
1
8
0
.
5
1
4
5
0
.
5
0
2
3
0
.
4
9
9
5
0
.
4
9
9
9
0
.
4
9
8
0
0
.
4
9
8
2
0
.
9
9
9
5
0
.
9
9
9
8
2
0
1
9
0
.
5
2
1
8
0
.
5
0
2
3
0
.
4
9
9
7
0
.
4
9
9
9
0
.
4
9
7
3
0
.
4
9
8
2
0
.
9
9
9
4
0
.
9
9
9
8
2
0
2
0
0
.
5
2
7
5
0
.
5
0
5
4
0
.
5
3
5
7
0
.
5
1
1
7
0
.
5
3
1
2
0
.
5
1
3
9
0
.
6
6
6
4
0
.
4
9
9
8
2
0
2
1
0
.
5
0
5
9
0
.
5
0
1
6
0
.
5
0
0
0
0
.
5
0
0
0
0
.
4
9
7
4
0
.
4
9
7
9
0
.
5
0
0
0
0
.
5
0
0
0
2
0
2
2
0
.
5
4
5
2
0
.
5
0
8
8
0
.
5
2
7
4
0
.
4
9
9
9
0
.
5
2
9
4
0
.
4
9
6
7
0
.
4
9
9
6
0
.
4
9
9
9
2
0
2
3
0
.
5
1
9
5
0
.
5
0
1
8
0
.
5
1
2
2
0
.
5
0
0
0
0
.
5
1
7
6
0
.
4
9
8
7
0
.
9
9
9
9
1
.
0
0
0
0
3
.
3
.
3
.
I
nfluence
s
o
f
f
ea
t
ure
eng
ineering
F
e
a
t
u
r
e
e
n
g
i
n
e
e
r
i
n
g
(
F
E
)
i
n
f
l
u
e
n
c
e
s
b
o
t
h
m
e
t
h
o
d
s
i
n
v
a
l
i
d
a
t
i
o
n
r
e
s
u
lt
s
,
s
o
i
ts
i
n
f
lu
e
n
c
e
o
n
c
l
a
s
s
i
f
ic
a
t
i
o
n
a
c
c
u
r
a
c
y
i
s
a
ls
o
s
t
u
d
i
e
d
.
T
a
b
l
e
s
2
a
n
d
3
s
h
o
w
s
t
h
e
c
o
m
p
l
e
t
e
c
la
s
s
i
f
i
ca
t
i
o
n
r
e
s
u
l
t
s
o
f
t
h
e
m
e
t
h
o
d
s
.
W
h
e
n
a
v
e
r
a
g
e
d
,
t
h
e
L
S
T
M
m
et
h
o
d
w
i
t
h
t
h
e
a
d
d
i
t
i
o
n
o
f
FE
h
a
s
a
n
a
c
c
u
r
a
c
y
o
f
0
.
7
9
2
3
,
b
u
t
w
i
t
h
o
u
t
t
h
e
a
d
d
i
t
i
o
n
o
f
f
e
a
t
u
r
e
e
n
g
i
n
e
e
r
i
n
g
(
W
F
E
)
h
a
s
a
n
a
c
c
u
r
a
c
y
o
f
0
.
8
2
3
4
.
T
h
e
n
,
w
i
t
h
FE
,
t
h
e
a
v
e
r
a
g
e
v
a
l
u
e
o
f
XG
b
o
o
s
t
c
l
a
s
s
i
f
ic
a
t
i
o
n
a
c
c
u
r
a
c
y
i
s
0
.
5
8
1
0
,
w
h
i
l
e
W
F
E
,
i
t
i
s
0
.
5
6
4
0
.
B
o
t
h
c
o
n
d
i
ti
o
n
s
s
h
o
w
r
e
s
u
lt
s
t
h
a
t
al
i
g
n
wi
t
h
t
h
e
v
a
l
i
d
a
t
i
o
n
p
r
o
c
e
s
s
w
h
e
r
e
a
d
d
i
n
g
FE
m
a
k
e
s
t
h
e
a
c
c
u
r
a
c
y
o
f
t
h
e
L
S
T
M
m
e
t
h
o
d
d
e
c
r
e
as
e
a
n
d
XG
B
o
o
s
t
i
n
c
r
e
as
e
.
3
.
3
.
4
.
Air
qu
a
lity
ind
ex
a
no
m
a
ly
det
ec
t
i
o
n r
esu
lt
s
T
h
e
AQI
h
as
ca
teg
o
r
ies
b
ased
o
n
th
e
PM2
.
5
in
d
icato
r
v
alu
e
.
B
ased
o
n
Fig
u
r
e
5
,
an
o
m
aly
d
etec
tio
n
with
L
STM
(
Fig
u
r
e
5
(
a)
)
s
h
o
ws
th
at
m
an
y
AQI
an
o
m
al
ies
o
cc
u
r
in
th
e
u
n
h
ea
lth
y
f
o
r
s
en
s
itiv
e
g
r
o
u
p
ca
teg
o
r
y
.
Ho
wev
er
,
XGBo
o
s
t
ca
p
tu
r
es
m
an
y
an
o
m
alies
in
th
e
g
o
o
d
an
d
u
n
h
ea
lth
y
ca
teg
o
r
ies
as
s
h
o
wn
in
Fig
u
r
e
5
(
b
)
.
As
an
ex
am
p
le,
an
d
co
m
p
a
r
is
o
n
,
L
STM
lab
els
th
e
AQI
o
n
Ma
y
1
3
,
2
0
2
3
,
at
0
2
:0
0
,
wh
ile
XGBo
o
s
t
d
o
es
n
o
t.
B
ased
o
n
th
e
v
alu
e
d
etails,
th
e
AQI
ch
a
n
g
ed
b
y
3
3
p
o
in
ts
f
r
o
m
8
6
(
m
o
d
er
ate)
at
0
1
:
0
0
to
1
1
9
(
u
n
h
ea
lth
y
f
o
r
s
en
s
itiv
e
g
r
o
u
p
)
at
0
2
:0
0
.
T
h
en
,
o
n
Ma
y
1
0
,
2
0
1
9
,
at
0
7
:
0
0
,
XGBo
o
s
t
ca
p
tu
r
ed
t
h
e
AQI
v
alu
e
as
an
a
n
o
m
aly
,
wh
ile
L
STM
d
id
n
o
t.
At
th
at
tim
e,
t
h
e
AQI
was
8
(
g
o
o
d
)
,
an
d
an
h
o
u
r
ea
r
lier
,
it
was
2
3
(
g
o
o
d
)
.
B
ased
o
n
th
ese
co
n
d
itio
n
s
,
L
STM
is
b
etter
at
c
ap
tu
r
in
g
ex
tr
em
e
c
h
an
g
es
in
d
ata
v
alu
es,
wh
ile
XGBo
o
s
t is b
etter
at
ca
p
tu
r
in
g
ex
tr
em
e
d
ata
v
alu
es.
Fig
u
r
e
6
s
h
o
ws
th
e
r
esu
lts
o
f
AQI
an
o
m
alies
d
etec
ted
b
y
L
STM
as
th
e
b
est
m
eth
o
d
b
ased
o
n
v
alid
atio
n
an
d
class
if
icatio
n
ev
alu
atio
n
,
wh
ic
h
o
cc
u
r
r
ed
b
etwe
en
2
1
:0
0
to
0
9
:0
0
.
W
h
en
r
elate
d
to
th
e
ex
p
lan
ato
r
y
v
ar
iab
les
u
s
ed
,
th
e
win
d
s
p
ee
d
(
W
S1
0
M)
,
r
ain
f
all
(
PR
E
C
T
OT
C
O
R
R
)
,
an
d
air
h
u
m
id
ity
(
R
H2
M)
v
ar
iab
les
s
h
o
w
co
n
s
id
er
a
b
le
v
ar
iab
ilit
y
co
m
p
ar
ed
t
o
o
th
e
r
v
ar
iab
les.
B
ased
o
n
th
e
ex
p
lo
r
atio
n
,
th
e
r
e
is
a
n
o
ticea
b
le
d
if
f
er
en
ce
in
v
alu
e
s
b
etwe
en
d
ay
tim
e
an
d
n
ig
h
ttime
.
At
n
ig
h
t,
th
e
a
v
er
ag
e
win
d
s
p
ee
d
is
u
n
d
e
r
3
m
/s
,
r
an
g
in
g
f
r
o
m
2
.
4
9
m
/
s
(
2
3
:0
0
)
to
2
.
9
5
m
/s
(
0
7
:0
0
)
.
I
n
co
m
p
a
r
is
o
n
,
at
o
th
er
ti
m
es,
it
r
an
g
es
f
r
o
m
3
.
0
7
m
/s
(
1
7
:0
0
)
to
3
.
6
4
m
/s
(
1
5
:
0
0
)
.
Fu
r
th
e
r
m
o
r
e,
th
e
c
o
r
r
ec
ted
r
ain
f
all
at
n
ig
h
t
r
a
n
g
ed
f
r
o
m
0
.
2
5
m
m
(
2
2
:0
0
)
to
0
.
3
4
m
m
(
0
3
:0
0
)
.
Als
o
,
it
af
f
ec
ted
th
e
air
h
u
m
id
ity
,
wh
ich
av
e
r
ag
ed
ab
o
v
e
8
5
%
f
r
o
m
1
9
:0
0
t
o
0
6
:0
0
,
wh
ile
at
o
th
e
r
tim
es
ten
d
ed
to
b
e
u
n
d
e
r
th
is
v
alu
e.
Du
r
in
g
th
e
h
o
u
r
s
in
f
lu
e
n
ce
d
b
y
th
e
th
r
ee
wea
th
er
v
ar
iab
les,
ar
o
u
n
d
2
0
:0
0
to
0
4
:
0
0
,
th
e
AQI
d
etec
ted
as
an
an
o
m
aly
ten
d
s
to
in
cr
ea
s
e
(
f
o
r
e
x
am
p
le,
a
ch
an
g
e
in
th
e
m
o
d
er
ate
ca
teg
o
r
y
to
u
n
h
ea
lth
y
f
o
r
s
en
s
itiv
e
g
r
o
u
p
s
)
.
T
h
is
r
esu
lt
is
co
n
s
is
ten
t
w
ith
th
e
d
ata
ex
p
lo
r
atio
n
.
I
n
lin
e
with
Ku
s
u
m
an
in
g
t
y
as
et
a
l
.
[
3
1
]
,
w
h
o
m
en
tio
n
e
d
th
at
at
n
ig
h
t,
t
h
e
win
d
s
p
ee
d
te
n
d
s
to
b
e
lo
w
a
n
d
ca
lm
(
<3
m
/s
)
s
o
th
at
s
u
r
f
ac
e
p
o
llu
tan
ts
ca
n
n
o
t
m
o
v
e
v
er
tically
in
th
e
in
v
er
s
io
n
lay
er
th
at
s
u
r
r
o
u
n
d
s
th
e
s
u
r
f
ac
e
air
at
n
ig
h
t
lik
e
a
“ser
v
in
g
h
o
o
d
”.
I
n
ad
d
itio
n
,
a
h
ig
h
AQI
is
in
lin
e
with
th
e
h
ig
h
am
o
u
n
t
o
f
h
u
m
a
n
ac
tiv
ity
d
u
r
in
g
p
ea
k
h
o
u
r
s
,
m
a
in
ly
d
u
e
t
o
tr
af
f
ic
a
n
d
v
e
h
icle
p
o
llu
tio
n
[
3
2
]
.
T
h
en
,
wh
e
n
v
iewe
d
o
n
a
m
o
n
th
ly
tim
e
f
r
am
e,
a
n
o
m
ali
es
ten
d
to
o
cc
u
r
in
N
o
v
em
b
er
-
Ma
r
ch
.
E
x
p
lo
r
atio
n
o
f
ex
p
la
n
ato
r
y
v
a
r
iab
les
s
h
o
ws
th
at
win
d
s
p
ee
d
,
r
ain
f
all,
an
d
h
u
m
id
ity
ar
e
t
h
r
ee
v
ar
iab
les
with
h
ig
h
v
ar
ia
b
ilit
y
,
s
u
ch
as
h
o
u
r
ly
tim
ef
r
am
es.
Ho
wev
er
,
o
v
er
a
m
o
r
e
ex
ten
d
ed
p
e
r
io
d
,
clim
a
te
v
ar
iab
les
s
u
ch
as
r
ain
f
all
will
b
e
m
o
r
e
v
is
ib
le
t
h
an
wea
th
er
co
n
d
itio
n
s
s
u
ch
as
win
d
s
p
ee
d
d
u
e
to
th
e
in
f
lu
en
ce
o
f
s
ea
s
o
n
ality
.
Du
r
in
g
th
e
r
ain
y
s
ea
s
o
n
,
a
r
o
u
n
d
No
v
em
b
er
-
Ma
r
ch
,
th
e
r
ai
n
f
all
co
r
r
ec
tio
n
r
a
n
g
es
f
r
o
m
0
.
2
9
m
m
(
No
v
em
b
er
an
d
Ap
r
il)
to
0
.
6
5
m
m
(
Feb
r
u
ar
y
)
.
T
h
is
v
alu
e
is
h
ig
h
er
th
an
d
u
r
in
g
th
e
d
r
y
s
ea
s
o
n
(
M
ay
-
Octo
b
er
)
,
wh
ich
r
an
g
es
f
r
o
m
0
.
0
6
m
m
(
Au
g
u
s
t)
to
0
.
1
9
m
m
(
Ma
y
)
.
Hig
h
r
ain
f
all
will
ca
u
s
e
an
in
cr
ea
s
e
in
air
h
u
m
id
ity
,
s
u
ch
as
in
Feb
r
u
ar
y
,
with
an
av
er
a
g
e
h
u
m
id
ity
o
f
8
5
.
6
3
%.
T
h
is
c
o
n
d
itio
n
f
o
llo
ws
th
e
r
esear
ch
o
f
C
h
o
lian
awa
ti
et
a
l
.
[
3
3
]
,
with
d
ata
f
r
o
m
2
0
2
1
s
h
o
win
g
th
a
t
th
er
e
is
a
m
o
n
s
o
o
n
al
r
ain
f
all
p
atter
n
with
o
n
e
p
ea
k
i
n
Dec
em
b
er
-
J
an
u
a
r
y
-
Fe
b
r
u
ar
y
.
T
h
en
,
ac
co
r
d
in
g
to
th
e
I
n
d
o
n
esia
Me
teo
r
o
lo
g
ical,
C
lim
ato
lo
g
ical,
an
d
Geo
p
h
y
s
ical
Ag
en
c
y
(
B
MK
G)
in
Yu
lin
awa
ti
et
a
l
.
[
3
4
]
,
J
ak
ar
ta
en
ter
e
d
th
e
r
ain
y
s
ea
s
o
n
in
No
v
em
b
er
,
ca
u
s
in
g
wet
p
r
ec
ip
itatio
n
,
wh
ich
d
ec
r
ea
s
ed
th
e
AQI
v
alu
e.
I
n
ad
d
itio
n
,
ac
c
o
r
d
in
g
to
I
s
tian
a
et
a
l
.
[
3
2
]
,
th
e
L
a
Nin
a
p
h
e
n
o
m
e
n
o
n
r
es
u
lts
in
in
cr
ea
s
ed
r
ain
f
all
a
n
d
lo
wer
tem
p
er
atu
r
es,
as
well
as
win
d
s
p
ee
d
d
u
r
in
g
th
e
d
r
y
s
ea
s
o
n
in
J
u
ly
-
A
u
g
u
s
t
,
wh
ich
ca
n
af
f
ec
t
th
e
AQI
.
L
o
w
tem
p
er
atu
r
e
a
n
d
win
d
ca
u
s
e
p
o
llu
tan
ts
to
b
e
tr
ap
p
ed
an
d
f
o
r
m
p
ar
ticles,
th
u
s
in
cr
ea
s
in
g
th
e
AQI
.
T
h
is
co
n
d
itio
n
is
in
lin
e
with
th
e
av
er
ag
e
o
f
th
e
d
ata
u
s
ed
,
wh
ich
is
2
7
.
0
2
°C
(
J
u
ly
)
an
d
2
7
.
0
5
°C
(
Au
g
u
s
t)
,
lo
wer
th
an
m
o
s
t
m
o
n
th
s
in
th
e
r
ain
y
s
ea
s
o
n
,
s
u
ch
as
i
n
No
v
em
b
er
(
2
8
.
1
1
°C
)
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
26
:
2
5
4
3
-
2
5
5
3
2550
(
a)
(
b
)
Fig
u
r
e
5
.
Nu
m
b
er
o
f
an
o
m
alie
s
b
y
(
a)
L
STM
an
d
(
b
)
XGBo
o
s
t m
eth
o
d
Fig
u
r
e
6
.
Nu
m
b
er
o
f
an
o
m
alie
s
d
etec
ted
b
y
th
e
b
est m
eth
o
d
(
L
STM
)
b
ased
o
n
h
o
u
r
an
d
m
o
n
th
LSTM
Ho
url
y
M
o
nthl
y
X
GBo
o
s
t
Ho
url
y
M
o
nthl
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
Ma
ch
in
e
lea
r
n
in
g
a
p
p
r
o
a
ch
es fo
r
a
n
o
ma
ly
d
etec
tio
n
o
f J
a
ka
r
ta
…
(
Mu
h
a
mma
d
R
iz
ky
N
u
r
h
a
mb
a
li
)
2551
4.
CO
NCLU
SI
O
N
T
h
e
r
esu
lts
o
f
v
alid
atio
n
an
d
c
lass
if
icatio
n
o
f
th
e
AQI
s
h
o
w
s
im
ilar
r
esu
lts
,
i.e
.
,
th
e
L
STM
m
eth
o
d
is
b
etter
th
an
th
e
XGBo
o
s
t
m
eth
o
d
.
L
STM
p
er
f
o
r
m
s
b
est
with
th
e
co
m
b
in
atio
n
o
f
MR
(
2
)
,
4
-
s
ig
m
a,
an
d
W
FE
to
p
er
f
o
r
m
an
o
m
aly
d
etec
tio
n
esp
ec
ially
in
ex
tr
em
e
v
alu
e
c
h
an
g
es,
wh
ile
in
XGBo
o
s
t,
th
e
b
est
co
m
b
in
atio
n
is
MR
(
3
)
,
4
-
s
ig
m
a,
an
d
FE
to
ca
p
tu
r
e
ex
tr
e
m
e
v
alu
es.
Ho
wev
er
,
th
ese
r
esu
lts
n
ee
d
to
b
e
s
tu
d
ied
f
u
r
th
er
as
v
ar
io
u
s
f
ac
to
r
s
af
f
ec
t
th
e
m
o
d
el
ev
alu
atio
n
,
esp
ec
ially
lab
eli
n
g
th
e
ac
tu
al
d
ata.
T
h
e
a
n
o
m
a
ly
d
etec
tio
n
r
esu
lts
in
th
e
AQI
s
h
o
w
th
at
m
a
n
y
an
o
m
alies
o
cc
u
r
b
etwe
en
2
1
:0
0
an
d
0
9
:0
0
a
n
d
in
t
h
e
r
ain
y
s
ea
s
o
n
.
AQI
an
o
m
alies
ar
e
in
s
ep
ar
ab
le
f
r
o
m
h
u
m
an
ac
tiv
ities
,
in
d
u
s
tr
ial
ac
tiv
ities
,
an
d
wea
th
er
c
o
n
d
itio
n
s
(
r
ain
f
all,
h
u
m
id
ity
,
an
d
win
d
s
p
ee
d
)
.
A
ll
th
r
ee
in
f
lu
en
ce
ea
ch
o
th
er
,
s
o
im
p
r
o
v
in
g
air
q
u
ality
r
eq
u
i
r
es
th
e
co
o
p
er
atio
n
o
f
v
ar
io
u
s
p
ar
ties
an
d
th
e
f
o
r
m
u
latio
n
o
f
ap
p
r
o
p
r
iate
p
o
licies.
T
h
is
s
tu
d
y
h
as
lim
itatio
n
s
r
elate
d
to
t
h
e
lab
elin
g
p
r
o
ce
s
s
an
d
m
o
d
elin
g
ap
p
r
o
ac
h
u
s
ed
.
Fu
tu
r
e
r
esear
ch
co
u
ld
ex
p
l
o
r
e
alter
n
ativ
e
lab
elin
g
s
ch
em
es,
s
u
ch
as
th
e
u
s
e
o
f
d
if
f
er
en
t
MR
r
an
g
e
th
r
esh
o
ld
s
(
e.
g
.
,
MR
>3
)
,
wh
ich
m
ay
p
r
o
v
i
d
e
h
ig
h
er
s
en
s
itiv
ity
in
d
etec
tin
g
ch
an
g
es
in
d
iv
er
s
it
y
.
Oth
er
r
esear
ch
d
ir
ec
tio
n
s
th
at
co
u
ld
b
e
d
ev
elo
p
ed
f
r
o
m
th
e
lab
elin
g
s
id
e
in
clu
d
e
r
an
k
i
n
g
-
b
ased
lab
eli
n
g
,
d
is
tan
ce
-
b
ased
m
ea
s
u
r
es
,
o
r
p
r
o
b
ab
ilit
y
-
b
ased
lab
el
in
g
.
Ad
d
itio
n
ally
,
p
r
ed
ictiv
e
p
er
f
o
r
m
a
n
ce
co
u
l
d
b
e
im
p
r
o
v
ed
b
y
ex
p
lo
r
i
n
g
h
y
b
r
id
m
o
d
elin
g
s
tr
ateg
ies,
wh
er
e
L
STM
an
d
XGBo
o
s
t
ar
e
co
m
b
in
e
d
with
ea
ch
o
th
er
o
r
with
ad
v
an
ce
d
ar
ch
itectu
r
es
s
u
ch
as
v
a
r
iatio
n
al
au
to
en
c
o
d
er
s
(
VAE
–
L
STM
)
to
ca
p
tu
r
e
s
h
o
r
t
-
ter
m
in
f
o
r
m
atio
n
an
d
co
m
p
lex
an
o
m
aly
ch
ar
a
cter
is
tics
,
e
s
p
ec
ially
in
AQI
d
ata.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
T
h
is
r
esear
ch
was
s
u
p
p
o
r
ted
b
y
g
r
an
ts
f
r
o
m
th
e
B
I
MA
Pro
g
r
am
o
f
th
e
Min
is
tr
y
o
f
E
d
u
ca
tio
n
,
C
u
ltu
r
e,
R
esear
ch
,
an
d
T
ec
h
n
o
lo
g
y
o
f
th
e
R
ep
u
b
lic
o
f
I
n
d
o
n
esia
(
cu
r
r
en
tly
th
e
Min
is
tr
y
o
f
Hig
h
er
E
d
u
ca
tio
n
,
Scien
ce
,
an
d
T
ec
h
n
o
lo
g
y
)
u
n
d
er
co
n
tr
ac
t
n
u
m
b
e
r
0
2
7
/E5
/PG.0
2
.
0
0
.
PL/2
0
2
4
.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
T
h
is
jo
u
r
n
al
u
s
es
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT)
to
r
ec
o
g
n
ize
in
d
iv
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
llab
o
r
atio
n
.
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
Mu
h
am
m
ad
R
izk
y
Nu
r
h
am
b
ali
✓
✓
✓
✓
✓
✓
✓
✓
Yen
n
i A
n
g
r
ain
i
✓
✓
✓
✓
✓
✓
✓
✓
An
war
Fit
r
ian
to
✓
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
Au
th
o
r
s
s
tate
n
o
co
n
f
lict o
f
in
t
er
est.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
d
ata
th
at
s
u
p
p
o
r
t
th
e
f
in
d
in
g
s
o
f
th
e
r
esear
ch
ar
e
av
ailab
le
in
Air
No
w
an
d
NASA
PO
W
E
R
s
ite
at
h
ttp
s
://www.
air
n
o
w.
g
o
v
/
a
n
d
h
tt
p
s
://p
o
wer
.
lar
c.
n
asa.
g
o
v
/,
r
ef
er
e
n
ce
[
2
1
]
,
[
2
2
]
.
T
h
ese
d
ata
wer
e
d
er
i
v
ed
f
r
o
m
th
e
f
o
llo
win
g
r
eso
u
r
ce
s
av
ailab
le
in
th
e
p
u
b
lic
d
o
m
ain
:
h
ttp
s
://p
o
wer
.
lar
c.
n
asa.
g
o
v
/d
ata
-
ac
ce
s
s
-
v
iewe
r
/
.
B
u
t
u
n
f
o
r
tu
n
ately
,
th
e
s
p
ec
if
i
c
UR
L
f
o
r
r
ef
e
r
en
ce
[
2
1
]
is
n
o
lo
n
g
e
r
av
ailab
le.
T
o
s
u
p
p
o
r
t
d
ata
ac
ce
s
s
ib
ilit
y
an
d
r
e
p
r
o
d
u
cib
ilit
y
,
th
e
d
ataset
u
s
ed
in
t
h
is
s
tu
d
y
h
as
also
b
ee
n
m
ad
e
p
u
b
licly
av
ailab
le
th
r
o
u
g
h
at
h
ttp
s
://d
r
iv
e.
g
o
o
g
le.
co
m
/d
r
iv
e
/f
o
ld
er
s
/1
7
-
e
q
Z
R
UwL
Qv
k
k
7
i
HPC
Z
d
h
s
Qo
L
n
NVNQ
VQ.
RE
F
E
R
E
NC
E
S
[
1
]
H
.
B
o
r
g
e
s
,
R
.
A
k
b
a
r
i
n
i
a
,
a
n
d
F
.
M
a
sse
g
l
i
a
,
“
A
n
o
ma
l
y
d
e
t
e
c
t
i
o
n
i
n
t
i
m
e
seri
e
s,”
i
n
T
r
a
n
sa
c
t
i
o
n
s
o
n
L
a
rg
e
-
S
c
a
l
e
D
a
t
a
-
a
n
d
K
n
o
w
l
e
d
g
e
-
C
e
n
t
e
r
e
d
S
y
st
e
m
s L
,
S
p
r
i
n
g
e
r
B
e
r
l
i
n
H
e
i
d
e
l
b
e
r
g
,
2
0
2
1
,
p
p
.
4
6
–
62
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
662
-
6
4
5
5
3
-
6
_
3
.
[
2
]
K
.
C
h
o
i
,
J.
Y
i
,
C
.
P
a
r
k
,
a
n
d
S
.
Y
o
o
n
,
“
D
e
e
p
l
e
a
r
n
i
n
g
f
o
r
a
n
o
mal
y
d
e
t
e
c
t
i
o
n
i
n
t
i
me
-
s
e
r
i
e
s
d
a
t
a
:
r
e
v
i
e
w
,
a
n
a
l
y
si
s,
a
n
d
g
u
i
d
e
l
i
n
e
s,
”
I
EEE
A
c
c
e
ss
,
v
o
l
.
9
,
p
p
.
1
2
0
0
4
3
–
1
2
0
0
6
5
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
1
.
3
1
0
7
9
7
5
.
[
3
]
Z.
F
a
n
g
,
S
.
Y
a
n
g
,
C
.
L
v
,
S
.
A
n
,
a
n
d
W
.
W
u
,
“
A
p
p
l
i
c
a
t
i
o
n
o
f
a
d
a
t
a
-
d
r
i
v
e
n
X
G
B
o
o
s
t
m
o
d
e
l
f
o
r
t
h
e
p
r
e
d
i
c
t
i
o
n
o
f
C
O
V
I
D
-
1
9
i
n
t
h
e
U
S
A
:
a
t
i
m
e
-
seri
e
s s
t
u
d
y
,
”
BM
J
O
p
e
n
,
v
o
l
.
1
2
,
n
o
.
7
,
J
u
l
.
2
0
2
2
,
d
o
i
:
1
0
.
1
1
3
6
/
b
mj
o
p
e
n
-
2
0
2
1
-
0
5
6
6
8
5
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
26
:
2
5
4
3
-
2
5
5
3
2552
[
4
]
S
.
L
i
n
a
n
d
Y
.
F
e
n
g
,
“
R
e
se
a
r
c
h
o
n
s
t
o
c
k
p
r
i
c
e
p
r
e
d
i
c
t
i
o
n
b
a
se
d
o
n
o
r
t
h
o
g
o
n
a
l
G
a
u
ssi
a
n
b
a
s
i
s
f
u
n
c
t
i
o
n
e
x
p
a
n
s
i
o
n
a
n
d
P
e
a
r
so
n
c
o
r
r
e
l
a
t
i
o
n
c
o
e
f
f
i
c
i
e
n
t
w
e
i
g
h
t
e
d
LS
TM
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
Ad
v
a
n
c
e
s
i
n
C
o
m
p
u
t
e
r
,
S
i
g
n
a
l
s
a
n
d
S
y
s
t
e
m
s
,
v
o
l
.
6
,
n
o
.
5
,
2
0
2
2
,
d
o
i
:
1
0
.
2
3
9
7
7
/
a
c
ss.
2
0
2
2
.
0
6
0
5
0
4
.
[
5
]
R
.
G
h
o
s
h
,
K
.
C
a
u
se
y
,
K
.
B
u
r
k
a
r
t
,
S
.
W
o
z
n
i
a
k
,
A
.
C
o
h
e
n
,
a
n
d
M
.
B
r
a
u
e
r
,
“
A
mb
i
e
n
t
a
n
d
h
o
u
se
h
o
l
d
P
M
2
.
5
p
o
l
l
u
t
i
o
n
a
n
d
a
d
v
e
r
se
p
e
r
i
n
a
t
a
l
o
u
t
c
o
m
e
s:
a
me
t
a
-
r
e
g
r
e
ss
i
o
n
a
n
d
a
n
a
l
y
si
s
o
f
a
t
t
r
i
b
u
t
a
b
l
e
g
l
o
b
a
l
b
u
r
d
e
n
f
o
r
2
0
4
c
o
u
n
t
r
i
e
s
a
n
d
t
e
r
r
i
t
o
r
i
e
s
,
”
PLO
S
Me
d
i
c
i
n
e
,
v
o
l
.
1
8
,
n
o
.
9
,
S
e
p
.
2
0
2
1
,
d
o
i
:
1
0
.
1
3
7
1
/
j
o
u
r
n
a
l
.
p
me
d
.
1
0
0
3
7
1
8
.
[
6
]
B
.
H
a
r
y
a
n
t
o
,
B
.
Ja
l
a
l
u
d
i
n
,
A
.
A
s
y
a
r
y
,
N
.
R
o
e
s
t
a
n
d
y
,
a
n
d
F
.
N
u
g
r
a
h
a
,
“
A
ss
o
c
i
a
t
i
o
n
s
b
e
t
w
e
e
n
a
m
b
i
e
n
t
P
M
2
.
5
l
e
v
e
l
s a
n
d
c
h
i
l
d
r
e
n
’
s
p
n
e
u
mo
n
i
a
a
n
d
a
s
t
h
m
a
d
u
r
i
n
g
t
h
e
C
O
V
I
D
-
1
9
p
a
n
d
e
mi
c
i
n
G
r
e
a
t
e
r
Ja
k
a
r
t
a
(
J
a
b
o
d
e
t
a
b
e
k
)
,
”
A
n
n
a
l
s
o
f
G
l
o
b
a
l
H
e
a
l
t
h
,
v
o
l
.
9
1
,
n
o
.
1
,
F
e
b
.
2
0
2
5
,
d
o
i
:
1
0
.
5
3
3
4
/
a
o
g
h
.
4
6
2
3
.
[
7
]
T.
R
o
s
a
l
i
e
,
Y
.
R
i
a
n
i
,
a
n
d
M
e
i
r
y
a
n
i
,
“
Th
e
u
r
g
e
n
c
y
o
f
i
mp
l
e
m
e
n
t
i
n
g
c
a
r
b
o
n
t
a
x
o
n
J
a
k
a
r
t
a
’
s
a
i
r
q
u
a
l
i
t
y
(
t
a
x
i
n
c
e
n
t
i
v
e
s
a
s
mo
d
e
r
a
t
i
o
n
v
a
r
i
a
b
l
e
)
,
”
E
3
S
W
e
b
o
f
C
o
n
f
e
r
e
n
c
e
s
,
v
o
l
.
5
5
9
,
A
u
g
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
5
1
/
e
3
sc
o
n
f
/
2
0
2
4
5
5
9
0
4
0
2
5
.
[
8
]
S
.
A
l
l
a
a
n
d
S
.
K
.
A
d
a
r
i
,
“
P
r
a
c
t
i
c
a
l
u
s
e
c
a
ses
o
f
a
n
o
mal
y
d
e
t
e
c
t
i
o
n
,
”
i
n
B
e
g
i
n
n
i
n
g
A
n
o
m
a
l
y
D
e
t
e
c
t
i
o
n
U
si
n
g
P
y
t
h
o
n
-
Ba
s
e
d
D
e
e
p
L
e
a
r
n
i
n
g
,
B
e
r
k
e
l
e
y
,
U
n
i
t
e
d
S
t
a
t
e
s
:
A
p
r
e
ss,
2
0
1
9
,
p
p
.
2
9
7
–
318
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
1
-
4
8
4
2
-
5
1
7
7
-
5
_
8
.
[
9
]
X
.
W
a
n
g
,
T.
Z
h
a
o
,
H
.
L
i
u
,
a
n
d
R
.
H
e
,
“
P
o
w
e
r
c
o
n
su
m
p
t
i
o
n
p
r
e
d
i
c
t
i
n
g
a
n
d
a
n
o
mal
y
d
e
t
e
c
t
i
o
n
b
a
s
e
d
o
n
l
o
n
g
s
h
o
r
t
-
t
e
r
m
m
e
m
o
r
y
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
i
n
2
0
1
9
I
E
EE
4
t
h
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
C
l
o
u
d
C
o
m
p
u
t
i
n
g
a
n
d
B
i
g
D
a
t
a
A
n
a
l
y
s
i
s (I
C
C
C
BD
A)
,
A
p
r
.
2
0
1
9
,
p
p
.
4
8
7
–
491
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
C
C
B
D
A
.
2
0
1
9
.
8
7
2
5
7
0
4
.
[
1
0
]
Y
.
Li
u
,
“
A
n
o
ma
l
y
d
e
t
e
c
t
i
o
n
i
n
m
u
l
t
i
v
a
r
i
a
t
e
t
i
m
e
seri
e
s
u
s
i
n
g
e
n
s
e
m
b
l
e
me
t
h
o
d
.
”
M
.
E
n
g
.
t
h
e
s
i
s
,
S
c
h
o
o
l
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
En
g
i
n
e
e
r
i
n
g
,
N
a
n
y
a
n
g
Te
c
h
n
o
l
o
g
i
c
a
l
U
n
i
v
e
r
si
t
y
,
S
i
n
g
a
p
o
r
e
,
2
0
2
1
,
d
o
i
:
1
0
.
3
2
6
5
7
/
1
0
3
5
6
/
1
5
5
7
3
1
.
[
1
1
]
P
.
Tr
i
z
o
g
l
o
u
,
X
.
L
i
u
,
a
n
d
Z.
L
i
n
,
“
F
a
u
l
t
d
e
t
e
c
t
i
o
n
b
y
a
n
e
n
sem
b
l
e
f
r
a
me
w
o
r
k
o
f
e
x
t
r
e
me
g
r
a
d
i
e
n
t
b
o
o
st
i
n
g
(
X
G
B
o
o
st
)
i
n
t
h
e
o
p
e
r
a
t
i
o
n
o
f
o
f
f
sh
o
r
e
w
i
n
d
t
u
r
b
i
n
e
s,”
Re
n
e
w
a
b
l
e
E
n
e
r
g
y
,
v
o
l
.
1
7
9
,
p
p
.
9
4
5
–
9
6
2
,
D
e
c
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
r
e
n
e
n
e
.
2
0
2
1
.
0
7
.
0
8
5
.
[
1
2
]
S
.
H
o
c
h
r
e
i
t
e
r
a
n
d
J.
S
c
h
mi
d
h
u
b
e
r
,
“
Lo
n
g
sh
o
r
t
-
t
e
r
m
m
e
m
o
r
y
,
”
N
e
u
r
a
l
C
o
m
p
u
t
a
t
i
o
n
,
v
o
l
.
9
,
n
o
.
8
,
p
p
.
1
7
3
5
–
1
7
8
0
,
N
o
v
.
1
9
9
7
,
d
o
i
:
1
0
.
1
1
6
2
/
n
e
c
o
.
1
9
9
7
.
9
.
8
.
1
7
3
5
.
[
1
3
]
J.
Z
h
a
n
g
,
Y
.
Ze
n
g
,
a
n
d
B
.
S
t
a
r
l
y
,
“
R
e
c
u
r
r
e
n
t
n
e
u
r
a
l
n
e
t
w
o
r
k
s
w
i
t
h
l
o
n
g
t
e
r
m
t
e
m
p
o
r
a
l
d
e
p
e
n
d
e
n
c
i
e
s
i
n
m
a
c
h
i
n
e
t
o
o
l
w
e
a
r
d
i
a
g
n
o
si
s
a
n
d
p
r
o
g
n
o
si
s
,
”
S
N
A
p
p
l
i
e
d
S
c
i
e
n
c
e
s
,
v
o
l
.
3
,
n
o
.
4
,
A
p
r
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
s
4
2
4
5
2
-
0
2
1
-
0
4
4
2
7
-
5.
[
1
4
]
K
.
L.
T
a
n
,
C
.
P
.
Le
e
,
K
.
S
.
M
.
A
n
b
a
n
a
n
t
h
e
n
,
a
n
d
K
.
M
.
L
i
m,
“
R
o
B
E
R
T
a
-
LSTM
:
a
h
y
b
r
i
d
m
o
d
e
l
f
o
r
se
n
t
i
m
e
n
t
a
n
a
l
y
si
s
w
i
t
h
t
r
a
n
sf
o
r
mer
a
n
d
r
e
c
u
r
r
e
n
t
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
I
E
EE
A
c
c
e
ss
,
v
o
l
.
1
0
,
p
p
.
2
1
5
1
7
–
2
1
5
2
5
,
2
0
2
2
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
2
.
3
1
5
2
8
2
8
.
[
1
5
]
X
.
W
a
n
g
,
Y
.
Y
a
n
g
,
X
.
Z
h
a
o
,
M
.
H
u
a
n
g
,
a
n
d
Q
.
Zh
u
,
“
I
n
t
e
g
r
a
t
i
n
g
f
i
e
l
d
i
m
a
g
e
s
a
n
d
mi
c
r
o
c
l
i
m
a
t
e
d
a
t
a
t
o
r
e
a
l
i
z
e
mu
l
t
i
-
d
a
y
a
h
e
a
d
f
o
r
e
c
a
st
i
n
g
o
f
ma
i
z
e
c
r
o
p
c
o
v
e
r
a
g
e
u
si
n
g
C
N
N
-
LSTM
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
Ag
r
i
c
u
l
t
u
ra
l
a
n
d
Bi
o
l
o
g
i
c
a
l
En
g
i
n
e
e
ri
n
g
,
v
o
l
.
1
6
,
n
o
.
2
,
p
p
.
1
9
9
–
2
0
6
,
2
0
2
3
,
d
o
i
:
1
0
.
2
5
1
6
5
/
j
.
i
j
a
b
e
.
2
0
2
3
1
6
0
2
.
7
0
2
0
.
[
1
6
]
T.
C
h
e
n
a
n
d
C
.
G
u
e
s
t
r
i
n
,
“
X
G
B
o
o
s
t
:
a
s
c
a
l
a
b
l
e
t
r
e
e
b
o
o
s
t
i
n
g
s
y
st
e
m,
”
i
n
Pro
c
e
e
d
i
n
g
s
o
f
t
h
e
2
2
n
d
AC
M
S
I
G
K
D
D
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
K
n
o
w
l
e
d
g
e
D
i
s
c
o
v
e
r
y
a
n
d
D
a
t
a
Mi
n
i
n
g
,
A
u
g
.
2
0
1
6
,
p
p
.
7
8
5
–
794
,
d
o
i
:
1
0
.
1
1
4
5
/
2
9
3
9
6
7
2
.
2
9
3
9
7
8
5
.
[
1
7
]
Z.
A
r
i
f
A
l
i
,
Z.
H
.
A
b
d
u
l
j
a
b
b
a
r
,
H
.
A
.
Ta
h
i
r
,
A
.
B
.
S
a
l
l
o
w
,
a
n
d
S
.
M
.
A
l
mu
f
t
i
,
“
Ex
t
r
e
m
e
g
r
a
d
i
e
n
t
b
o
o
st
i
n
g
a
l
g
o
r
i
t
h
m
w
i
t
h
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
:
a
r
e
v
i
e
w
,
”
Ac
a
d
e
m
i
c
J
o
u
r
n
a
l
o
f
N
a
w
r
o
z
U
n
i
v
e
rs
i
t
y
,
v
o
l
.
1
2
,
n
o
.
2
,
p
p
.
3
2
0
–
3
3
4
,
M
a
y
2
0
2
3
,
d
o
i
:
1
0
.
2
5
0
0
7
/
a
j
n
u
.
v
1
2
n
2
a
1
6
1
2
.
[
1
8
]
C
.
B
e
n
t
é
j
a
c
,
A
.
C
s
ö
r
g
ő
,
a
n
d
G
.
M
.
-
M
u
ñ
o
z
,
“
A
c
o
mp
a
r
a
t
i
v
e
a
n
a
l
y
s
i
s
o
f
g
r
a
d
i
e
n
t
b
o
o
s
t
i
n
g
a
l
g
o
r
i
t
h
ms
,
”
Ar
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
Re
v
i
e
w
,
v
o
l
.
5
4
,
n
o
.
3
,
p
p
.
1
9
3
7
–
1
9
6
7
,
M
a
r
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
0
4
6
2
-
0
2
0
-
0
9
8
9
6
-
5.
[
1
9
]
S
.
B
e
j
,
N
.
D
a
v
t
y
a
n
,
M
.
W
o
l
f
i
e
n
,
M
.
N
a
ssar,
a
n
d
O
.
W
o
l
k
e
n
h
a
u
e
r
,
“
Lo
R
A
S
:
a
n
o
v
e
r
sam
p
l
i
n
g
a
p
p
r
o
a
c
h
f
o
r
i
mb
a
l
a
n
c
e
d
d
a
t
a
s
e
t
s,
”
Ma
c
h
i
n
e
L
e
a
r
n
i
n
g
,
v
o
l
.
1
1
0
,
n
o
.
2
,
p
p
.
2
7
9
–
3
0
1
,
F
e
b
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
0
9
9
4
-
0
2
0
-
0
5
9
1
3
-
4.
[
2
0
]
S
.
P
r
a
m
a
n
a
,
D
.
Y
.
P
a
r
a
mart
h
a
,
Y
.
A
d
h
i
n
u
g
r
o
h
o
,
a
n
d
M
.
N
u
r
m
a
l
a
sari
,
“
A
i
r
p
o
l
l
u
t
i
o
n
c
h
a
n
g
e
s
o
f
J
a
k
a
r
t
a
,
B
a
n
t
e
n
,
a
n
d
W
e
st
Ja
v
a
,
I
n
d
o
n
e
si
a
d
u
r
i
n
g
t
h
e
f
i
r
st
m
o
n
t
h
o
f
C
O
V
I
D
-
1
9
p
a
n
d
e
m
i
c
,
”
J
o
u
r
n
a
l
o
f
B
u
s
i
n
e
ss
,
Ec
o
n
o
m
i
c
s
a
n
d
En
v
i
r
o
n
m
e
n
t
a
l
S
t
u
d
i
e
s
,
v
o
l
.
1
0
,
n
o
.
4
,
p
p
.
1
5
–
1
9
,
2
0
2
0
,
d
o
i
:
1
0
.
1
3
1
0
6
/
j
b
e
e
s.
2
0
2
0
.
v
o
l
1
0
.
n
o
4
.
1
5
.
[
2
1
]
U
n
i
t
e
d
S
t
a
t
e
s
E
n
v
i
r
o
n
me
n
t
a
l
P
r
o
t
e
c
t
i
o
n
A
g
e
n
c
y
,
“
M
a
p
s
a
n
d
d
a
t
a
U
.
S
.
e
mb
a
ssi
e
s
a
n
d
c
o
n
s
u
l
a
t
e
s,
”
a
i
rn
o
w
.
g
o
v
.
A
c
c
e
ss
e
d
:
Ju
n
.
1
6
,
2
0
2
4
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s
:
/
/
d
r
i
v
e
.
g
o
o
g
l
e
.
c
o
m/
d
r
i
v
e
/
f
o
l
d
e
r
s/
1
7
-
e
q
ZR
U
w
LQ
v
k
k
7
i
H
P
C
Zd
h
sQ
o
L
n
N
V
N
Q
V
Q
[
2
2
]
N
A
S
A
P
r
e
d
i
c
t
i
o
n
o
f
W
o
r
l
d
w
i
d
e
En
e
r
g
y
R
e
s
o
u
r
c
e
s
(
P
O
W
E
R
)
,
“
D
a
t
a
a
c
c
e
ss
v
i
e
w
e
r
(
D
A
V
)
,
”
p
o
w
e
r
.
l
a
rc
.
n
a
s
a
.
g
o
v
.
A
c
c
e
ss
e
d
:
Ju
n
.
1
6
,
2
0
2
4
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s:
/
/
p
o
w
e
r
.
l
a
r
c
.
n
a
s
a
.
g
o
v
/
d
a
t
a
-
a
c
c
e
ss
-
v
i
e
w
e
r
[
2
3
]
A
.
A
t
t
a
a
l
l
a
h
a
n
d
R
.
A
.
K
h
a
n
,
“
S
M
O
TED
N
N
:
a
n
o
v
e
l
m
o
d
e
l
f
o
r
a
i
r
p
o
l
l
u
t
i
o
n
f
o
r
e
c
a
st
i
n
g
a
n
d
A
Q
I
c
l
a
ss
i
f
i
c
a
t
i
o
n
,
”
C
o
m
p
u
t
e
rs,
Ma
t
e
r
i
a
l
s &
C
o
n
t
i
n
u
a
,
v
o
l
.
7
1
,
n
o
.
1
,
p
p
.
1
4
0
3
–
1
4
2
5
,
2
0
2
2
,
d
o
i
:
1
0
.
3
2
6
0
4
/
c
mc.
2
0
2
2
.
0
2
1
9
6
8
.
[
2
4
]
S
.
Zh
e
n
g
,
Y
.
F
u
,
Y
.
S
u
n
,
C
.
Zh
a
n
g
,
Y
.
W
a
n
g
,
a
n
d
E
.
Li
c
h
t
f
o
u
se
,
“
H
i
g
h
r
e
so
l
u
t
i
o
n
ma
p
p
i
n
g
o
f
n
i
g
h
t
t
i
m
e
l
i
g
h
t
a
n
d
a
i
r
p
o
l
l
u
t
a
n
t
s
d
u
r
i
n
g
t
h
e
C
O
V
I
D
-
1
9
l
o
c
k
d
o
w
n
i
n
W
u
h
a
n
,
”
E
n
v
i
ro
n
m
e
n
t
a
l
C
h
e
m
i
s
t
ry
L
e
t
t
e
rs
,
v
o
l
.
19
,
n
o
.
4
,
M
a
r
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
0
3
1
1
-
021
-
0
1
2
2
2
-
x
.
[
2
5
]
Y
.
T
i
a
n
,
X
.
A
.
Y
a
o
,
L.
M
u
,
Q
.
F
a
n
,
a
n
d
Y
.
L
i
u
,
“
I
n
t
e
g
r
a
t
i
n
g
m
e
t
e
o
r
o
l
o
g
i
c
a
l
f
a
c
t
o
r
s
f
o
r
b
e
t
t
e
r
u
n
d
e
r
st
a
n
d
i
n
g
o
f
t
h
e
u
r
b
a
n
f
o
r
m
-
a
i
r
q
u
a
l
i
t
y
r
e
l
a
t
i
o
n
sh
i
p
,
”
L
a
n
d
s
c
a
p
e
E
c
o
l
o
g
y
,
v
o
l
.
3
5
,
n
o
.
1
0
,
p
p
.
2
3
5
7
–
2
3
7
3
,
O
c
t
.
2
0
2
0
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
0
9
8
0
-
0
2
0
-
0
1
0
9
4
-
6.
[
2
6
]
T.
H
a
n
d
h
a
y
a
n
i
,
“
A
n
i
n
t
e
g
r
a
t
e
d
a
n
a
l
y
si
s
o
f
a
i
r
p
o
l
l
u
t
i
o
n
a
n
d
m
e
t
e
o
r
o
l
o
g
i
c
a
l
c
o
n
d
i
t
i
o
n
s
i
n
J
a
k
a
r
t
a
,
”
S
c
i
e
n
t
i
f
i
c
Re
p
o
r
t
s
,
v
o
l
.
1
3
,
n
o
.
1
,
A
p
r
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
3
8
/
s
4
1
5
9
8
-
0
2
3
-
3
2
8
1
7
-
9.
[
2
7
]
M
.
S
.
R
a
h
ma
n
,
A
.
H
.
C
h
o
w
d
h
u
r
y
,
a
n
d
M
.
A
mr
i
n
,
“
A
c
c
u
r
a
c
y
c
o
mp
a
r
i
s
o
n
o
f
A
R
I
M
A
a
n
d
X
G
B
o
o
s
t
f
o
r
e
c
a
st
i
n
g
m
o
d
e
l
s
i
n
p
r
e
d
i
c
t
i
n
g
t
h
e
i
n
c
i
d
e
n
c
e
o
f
C
O
V
I
D
-
1
9
i
n
B
a
n
g
l
a
d
e
s
h
,
”
PLO
S
G
l
o
b
a
l
P
u
b
l
i
c
H
e
a
l
t
h
,
v
o
l
.
2
,
n
o
.
5
,
M
a
y
2
0
2
2
,
d
o
i
:
1
0
.
1
3
7
1
/
j
o
u
r
n
a
l
.
p
g
p
h
.
0
0
0
0
4
9
5
.
[
2
8
]
C.
-
X
.
Lv
,
S
.
-
Y
.
A
n
,
B
.
-
J.
Q
i
a
o
,
a
n
d
W
.
W
u
,
“
Ti
m
e
s
e
r
i
e
s
a
n
a
l
y
si
s
o
f
h
e
m
o
r
r
h
a
g
i
c
f
e
v
e
r
w
i
t
h
r
e
n
a
l
s
y
n
d
r
o
me
i
n
mai
n
l
a
n
d
C
h
i
n
a
b
y
u
s
i
n
g
a
n
X
G
B
o
o
st
f
o
r
e
c
a
st
i
n
g
mo
d
e
l
,
”
BM
C
I
n
f
e
c
t
i
o
u
s
D
i
se
a
s
e
s
,
v
o
l
.
2
1
,
n
o
.
1
,
D
e
c
.
2
0
2
1
,
d
o
i
:
1
0
.
1
1
8
6
/
s1
2
8
7
9
-
0
2
1
-
0
6
5
0
3
-
y.
[
2
9
]
W
.
H
.
W
o
o
d
a
l
l
a
n
d
D
.
C
.
M
o
n
t
g
o
m
e
r
y
,
“
U
si
n
g
r
a
n
g
e
s
t
o
e
st
i
ma
t
e
v
a
r
i
a
b
i
l
i
t
y
,
”
Q
u
a
l
i
t
y
E
n
g
i
n
e
e
r
i
n
g
,
v
o
l
.
1
3
,
n
o
.
2
,
p
p
.
2
1
1
–
2
1
7
,
D
e
c
.
2
0
0
0
,
d
o
i
:
1
0
.
1
0
8
0
/
0
8
9
8
2
1
1
0
1
0
8
9
1
8
6
4
3
.
[
3
0
]
J.
Z
h
e
n
g
,
D
.
F
e
n
g
,
Z.
Y
a
n
g
,
Y
.
X
i
a
n
g
,
H
.
Zh
a
n
g
,
a
n
d
S
.
L
i
,
“
Tr
a
n
s
K
S
:
a
n
a
n
o
m
a
l
y
d
e
t
e
c
t
i
o
n
m
e
t
h
o
d
f
o
r
t
e
l
e
c
o
mm
u
n
i
c
a
t
i
o
n
n
e
t
w
o
r
k
s
b
a
se
d
o
n
d
e
e
p
l
e
a
r
n
i
n
g
,
”
I
E
EE
A
c
c
e
ss
,
v
o
l
.
1
1
,
p
p
.
1
1
8
0
4
8
–
1
1
8
0
6
0
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
3
.
3
3
2
6
8
1
5
.
[
3
1
]
S
.
D
.
A
.
K
u
su
m
a
n
i
n
g
t
y
a
s,
A
.
N
.
K
h
o
i
r
,
E
.
F
i
b
r
i
a
n
t
i
k
a
,
a
n
d
E.
H
e
r
i
y
a
n
t
o
,
“
Ef
f
e
c
t
o
f
met
e
o
r
o
l
o
g
i
c
a
l
p
a
r
a
me
t
e
r
t
o
v
a
r
i
a
b
i
l
i
t
y
o
f
p
a
r
t
i
c
u
l
a
t
e
ma
t
t
e
r
(
P
M
)
c
o
n
c
e
n
t
r
a
t
i
o
n
i
n
u
r
b
a
n
Ja
k
a
r
t
a
c
i
t
y
,
I
n
d
o
n
e
si
a
,
”
I
O
P
C
o
n
f
e
re
n
c
e
S
e
r
i
e
s:
E
a
r
t
h
a
n
d
En
v
i
ro
n
m
e
n
t
a
l
S
c
i
e
n
c
e
,
v
o
l
.
7
2
4
,
n
o
.
1
,
A
p
r
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
8
8
/
1
7
5
5
-
1
3
1
5
/
7
2
4
/
1
/
0
1
2
0
5
0
.
[
3
2
]
T.
I
st
i
a
n
a
e
t
a
l
.
,
“
C
a
u
s
a
l
i
t
y
a
n
a
l
y
s
i
s
o
f
a
i
r
q
u
a
l
i
t
y
a
n
d
met
e
o
r
o
l
o
g
i
c
a
l
p
a
r
a
met
e
r
s
f
o
r
P
M
2
.
5
c
h
a
r
a
c
t
e
r
i
st
i
c
s
d
e
t
e
r
mi
n
a
t
i
o
n
:
e
v
i
d
e
n
c
e
f
r
o
m J
a
k
a
r
t
a
,
”
Ae
r
o
s
o
l
a
n
d
Ai
r Q
u
a
l
i
t
y
Re
s
e
a
r
c
h
,
v
o
l
.
2
3
,
n
o
.
9
,
2
0
2
3
,
d
o
i
:
1
0
.
4
2
0
9
/
a
a
q
r
.
2
3
0
0
1
4
.
[
3
3
]
N
.
C
h
o
l
i
a
n
a
w
a
t
i
e
t
a
l
.
,
“
D
i
u
r
n
a
l
a
n
d
d
a
i
l
y
v
a
r
i
a
t
i
o
n
s
o
f
P
M
2
.
5
a
n
d
i
t
s
mu
l
t
i
p
l
e
-
w
a
v
e
l
e
t
c
o
h
e
r
e
n
c
e
w
i
t
h
met
e
o
r
o
l
o
g
i
c
a
l
v
a
r
i
a
b
l
e
s
i
n
I
n
d
o
n
e
s
i
a
,
”
Ae
r
o
so
l
a
n
d
Ai
r
Q
u
a
l
i
t
y
Re
s
e
a
rc
h
,
v
o
l
.
2
4
,
n
o
.
3
,
2
0
2
4
,
d
o
i
:
1
0
.
4
2
0
9
/
a
a
q
r
.
2
3
0
1
5
8
.
[
3
4
]
H
.
Y
u
l
i
n
a
w
a
t
i
,
T.
K
h
a
i
r
a
n
i
,
a
n
d
L.
S
i
a
mi
,
“
A
n
a
l
y
si
s
o
f
i
n
d
o
o
r
a
n
d
o
u
t
d
o
o
r
p
a
r
t
i
c
u
l
a
t
e
(
P
M
2
.
5
)
a
t
a
w
o
m
e
n
a
n
d
c
h
i
l
d
r
e
n
’
s
h
o
s
p
i
t
a
l
i
n
W
e
s
t
J
a
k
a
r
t
a
,
”
I
O
P
C
o
n
f
e
re
n
c
e
S
e
ri
e
s:
E
a
r
t
h
a
n
d
E
n
v
i
r
o
n
m
e
n
t
a
l
S
c
i
e
n
c
e
,
v
o
l
.
7
3
7
,
n
o
.
1
,
A
p
r
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
8
8
/
1
7
5
5
-
1
3
1
5
/
7
3
7
/
1
/
0
1
2
0
6
7
.
Evaluation Warning : The document was created with Spire.PDF for Python.