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
.
14
, N
o.
2
,
A
pr
il
2025
, pp.
1401
~
1409
I
S
S
N
:
2252
-
8938
,
D
O
I
:
10.11591/
ij
a
i.
v
14
.i
2
.pp
1401
-
1409
1401
Jou
r
n
al
h
om
e
page
:
ht
tp
:
//
ij
ai
.
ia
e
s
c
or
e
.c
om
A
c
om
p
ar
at
i
ve
an
al
ysi
s of
e
xp
on
e
n
t
i
a
l
sm
oot
h
i
n
g m
e
t
h
od
a
n
d
d
e
e
p
l
e
ar
n
i
n
g m
o
d
e
l
s f
or
b
i
t
c
oi
n
p
r
i
c
e
p
r
e
d
i
c
t
i
on
N
r
u
s
in
gh
a T
r
ip
at
h
y
1
,
D
e
b
ah
u
t
i
M
is
h
r
a
1
, S
ar
b
e
s
w
ar
a H
ot
a
2
, M
an
d
ak
in
i
P
r
iy
a
d
ar
s
h
an
i
B
e
h
e
r
a
3
,
G
ob
in
d
a C
h
an
d
r
a D
as
4
, S
as
an
k
a
S
e
k
h
ar
D
al
ai
1
,
S
u
b
r
at
K
u
m
ar
N
ayak
1
1
D
e
pa
r
t
m
e
nt
of
C
om
put
e
r
S
c
i
e
nc
e
a
nd E
ngi
ne
e
r
i
ng, S
i
ks
ha
‘
O
’
A
nus
a
ndha
n (
D
e
e
m
e
d t
o be
U
ni
ve
r
s
i
t
y)
, B
huba
ne
s
w
a
r
, I
ndi
a
2
D
e
pa
r
t
m
e
nt
of
C
om
put
e
r
A
ppl
i
c
a
t
i
on, S
i
ks
ha
‘
O
’
A
nus
a
ndha
n
(
D
e
e
m
e
d t
o be
U
ni
ve
r
s
i
t
y
)
,
B
huba
ne
s
w
a
r
, I
ndi
a
3
S
c
hool
of
C
om
put
e
r
E
ngi
ne
e
r
i
ng
,
K
a
l
i
nga
I
ns
t
i
t
ut
e
of
I
ndus
t
r
i
a
l
T
e
c
hnol
ogy
(
D
e
e
m
e
d t
o be
U
ni
ve
r
s
i
t
y
)
, B
huba
ne
s
w
a
r
,
I
ndi
a
4
D
e
pa
r
t
m
e
nt
of
C
om
put
e
r
S
c
i
e
nc
e
a
nd A
ppl
i
c
a
t
i
ons
, K
L
D
e
e
m
e
d t
o be
U
ni
ve
r
s
i
t
y,
G
unt
ur
,
I
ndi
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
M
a
r
15
,
2024
R
e
vi
s
e
d
O
c
t
23
,
2024
A
c
c
e
pt
e
d
N
ov
14
,
2024
Blockcha
in
technolo
gy
is
the
foundati
on
of
cryptoc
urrenc
ies,
whi
ch
are
virtual
currencies.
The
decentralized
nature
of
cryptocurrencies
has
r
esulted
in
a
significant
reduction
of
central
authority
over
them,
whic
h
has
implications
for
global
trade
and
relations.
The
need
for
an
effec
tive
model
to
anticipate
the
price
of
cryptocurrencies
is
essential
due
to
thei
r
wide
variations
in
value.
Due
to
the
shortcomings
of
conventional
prod
uction
forecasting,
in
this
work,
four
distinct
models
were
used.
The
deep
le
arning
models
are
the
l
ong
short
-
t
erm
memory
(
LSTM
)
and
b
idirectiona
l
l
ong
short
-
term
memory
(
Bi
-
LSTM
)
,
and
both
the
Faceb
ook
-
P
rophe
t
and
Silverk
ite
support
the
expone
ntial
smoothing
techni
que.
Silver
kite
is
designed to handle a wide range
of time series forecasting tasks.
Consi
dering
past
b
itcoin
information
from
January
2012
to
March
2021,
a
p
eriod
of
nine
years,
we
looked
at
the
models.
The
Bi
-
LSTM
model
yields
a
7.073
mean
absolut
e
error
(MAE)
and
a
3.639
root
mean
squared
error
(RMSE
).
The
Bi
-
LSTM model identifies th
e deviations
that might
draw attention an
d avert
any probl
ems.
K
e
y
w
o
r
d
s
:
Bi
-
L
S
T
M
C
r
ypt
oc
ur
r
e
nc
y
F
a
c
e
book
-
P
r
ophe
t
F
in
a
nc
ia
l
da
ta
a
na
ly
s
i
s
L
ong s
hor
t
-
te
r
m
m
e
m
or
y
S
il
ve
r
ki
te
This is an
open
acce
ss artic
le unde
r the
CC BY
-
SA
license.
C
or
r
e
s
pon
di
n
g A
u
th
or
:
N
r
us
in
gha
T
r
ip
a
th
y
D
e
pa
r
tm
e
nt
of
C
om
put
e
r
S
c
ie
nc
e
a
nd E
ngi
ne
e
r
in
g, S
ik
s
ha
‘
O
’
A
nus
a
ndha
n (
D
e
e
m
e
d t
o b
e
U
ni
ve
r
s
it
y)
B
huba
ne
s
w
a
r
751030, I
ndi
a
E
m
a
il
:
nr
us
in
gha
tr
ip
a
th
y654@
gm
a
il
.c
om
1.
I
N
T
R
O
D
U
C
T
I
O
N
T
he
de
c
e
nt
r
a
li
z
e
d
ne
twor
k
of
c
om
put
e
r
s
th
a
t
pow
e
r
c
r
ypt
oc
ur
r
e
nc
ie
s
c
ol
la
bor
a
te
to
va
li
da
te
th
e
lo
g
tr
a
ns
a
c
ti
ons
.
T
h
e
r
e
f
or
e
,
th
e
m
one
y
is
no
lo
nge
r
s
ubj
e
c
t
to
th
e
r
he
os
ta
t
of
a
ut
hor
it
y
li
ke
a
ba
nk
o
r
th
e
gove
r
nm
e
nt
.
C
r
ypt
oc
ur
r
e
nc
y
tr
a
ns
a
c
ti
ons
a
r
e
doc
um
e
nt
e
d
o
n
a
bl
oc
kc
ha
in
,
a
n
im
m
ut
a
bl
e
,
de
c
e
nt
r
a
li
z
e
d
le
dge
r
[
1]
,
[
2]
.
T
o
e
na
bl
e
pr
iv
a
te
a
nd
s
e
c
ur
e
tr
a
ns
a
c
ti
on
s
,
publ
ic
a
nd
pr
iv
a
te
ke
ys
a
r
e
e
m
pl
oye
d.
L
a
hm
ir
i
a
nd
B
e
ki
r
os
[
3]
te
s
ti
ng
th
e
nonl
in
e
a
r
it
y,
pr
oduc
e
d
d
a
ta
s
how
in
g
f
r
a
c
ta
l
dyna
m
ic
s
,
e
xt
e
nde
d
m
e
m
or
y,
a
nd
s
e
lf
-
s
im
il
a
r
it
y
in
a
ll
di
gi
ta
l
c
ur
r
e
nc
y
ti
m
e
s
e
r
ie
s
.
T
he
r
e
is
a
c
a
p
on
th
e
to
ta
l
qua
nt
it
y
of
c
oi
ns
or
to
ke
ns
th
a
t
m
a
y
e
ve
r
be
is
s
ue
d
f
or
s
e
ve
r
a
l
c
r
ypt
oc
ur
r
e
nc
ie
s
.
A
s
a
n
i
ll
us
tr
a
ti
on,
th
e
m
a
xi
m
um
qua
nt
it
y
of
B
it
c
oi
n
c
oi
ns
is
s
e
t
a
t
21
m
il
li
on,
a
m
e
a
s
ur
e
in
te
nde
d
to
pr
oduc
e
s
c
a
r
c
it
y
a
nd
pe
r
ha
ps
boos
t
va
lu
e
ove
r
ti
m
e
.
T
r
a
ns
a
c
ti
on
s
w
it
h
c
r
ypt
oc
ur
r
e
nc
ie
s
a
r
e
of
te
n
f
a
s
t
e
r
a
nd
m
a
y
be
le
s
s
e
xpe
n
s
iv
e
th
a
n
th
o
s
e
us
in
g
tr
a
di
ti
ona
l
ba
nki
ng s
ys
te
m
s
,
a
nd i
t
c
a
n be
s
e
nt
a
nd r
e
c
e
iv
e
d a
nyw
he
r
e
on t
he
gl
obe
.
C
r
ypt
o
c
ur
r
e
n
c
ie
s
a
r
e
a
popu
la
r
in
v
e
s
t
m
e
nt
o
pt
io
n
f
or
in
v
e
s
to
r
s
.
I
n
a
ddi
ti
o
n
to
be
i
ng
u
s
e
d
f
or
onl
in
e
pa
ym
e
nt
s
a
nd
r
e
m
it
t
a
n
c
e
s
,
di
gi
t
a
l
a
s
s
e
t
s
a
r
e
a
ls
o
u
ti
li
z
e
d
f
or
ot
he
r
pr
a
c
t
ic
e
s
.
A
c
c
or
di
ng
to
S
un
e
t
al
.
[
4]
,
th
e
pr
e
di
c
ti
on'
s
pe
r
f
or
m
a
nc
e
i
s
im
pa
c
te
d
by
t
he
c
r
ypt
o
c
ur
r
e
n
c
y'
s
ove
r
a
ll
s
tr
e
ngt
h
.
T
hi
s
c
a
n
r
e
du
c
e
r
i
s
k
s
a
nd
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
.
14
, N
o.
2
,
A
pr
il
20
25
:
1401
-
1409
1402
pr
ovi
de
in
v
e
s
to
r
s
w
it
h
e
xc
e
ll
e
nt
gui
da
nc
e
w
he
n
bu
il
di
ng
a
b
it
c
oi
n
por
tf
ol
i
o.
N
on
e
th
e
l
e
s
s
,
th
e
m
a
r
ke
t
f
or
c
r
ypt
oc
ur
r
e
n
c
ie
s
is
r
e
now
n
e
d
f
or
it
s
u
npr
e
di
c
ta
bi
li
ty
,
di
f
f
i
c
ul
ti
e
s
w
i
th
r
e
gul
a
ti
on
s
,
a
nd
c
o
ns
t
a
nt
te
c
hni
c
a
l
a
dva
n
c
e
m
e
nt
s
[
5]
.
P
r
e
d
ic
ti
o
ns
m
a
y
b
e
us
e
d
by
tr
a
de
r
s
a
nd
in
v
e
s
to
r
s
to
c
r
e
a
te
he
dg
in
g
s
tr
a
t
e
gi
e
s
,
w
hi
c
h
tr
y
to
r
e
duc
e
to
ta
l
r
i
s
k
by
of
f
s
e
tt
in
g
p
os
s
ib
l
e
lo
s
s
e
s
in
on
e
a
r
e
a
of
th
e
por
tf
ol
io
w
it
h
pr
of
it
s
in
a
not
h
e
r
.
P
r
e
di
c
ti
on
s
a
bout
c
r
y
pt
oc
ur
r
e
nc
i
e
s
m
ig
ht
be
u
s
e
f
ul
to
tr
a
de
r
s
w
he
n
de
c
id
i
ng
w
he
th
e
r
to
e
nt
e
r
or
le
a
v
e
th
e
m
a
r
k
e
t.
T
hi
s
m
ig
ht
b
e
e
s
s
e
nt
i
a
l
f
or
o
pt
im
iz
i
ng g
a
in
s
or
r
e
duc
i
ng l
o
s
s
e
s
und
e
r
qui
c
kl
y
s
h
if
ti
ng m
a
r
ke
t
c
ir
c
um
s
t
a
n
c
e
s
[
6]
‒
[
8]
.
I
n
th
is
w
or
k,
lo
ng
s
hor
t
-
te
r
m
m
e
m
or
y
(
L
S
T
M
)
a
nd
bi
d
ir
e
c
ti
ona
l
-
lo
ng
s
hor
t
-
te
r
m
m
e
m
or
y
(
B
i
-
L
S
T
M
)
a
r
e
e
m
pl
oye
d
th
a
t
a
ut
om
a
ti
c
a
ll
y
e
xt
r
a
c
t
pe
r
ti
ne
nt
c
ha
r
a
c
te
r
is
ti
c
s
.
T
hi
s
i
s
he
lp
f
ul
be
c
a
u
s
e
it
e
na
bl
e
s
th
e
m
ode
l
to
de
r
iv
e
a
c
c
ur
a
te
de
pi
c
ti
ons
f
r
om
th
e
in
pu
t
da
ta
,
e
s
pe
c
ia
ll
y
w
he
n
w
or
ki
ng
w
it
h
bi
g
a
nd
c
om
pl
ic
a
te
d
da
ta
s
e
ts
[
9]
.
S
im
il
a
r
ly
,
th
e
e
xpone
nt
ia
l
s
m
oot
hi
ng
m
e
th
ods
F
a
c
e
book
-
P
r
ophe
t
(
FB
-
P
r
ophe
t
)
a
nd
S
il
ve
r
ki
te
a
r
e
c
a
s
to
f
f
;
th
e
s
e
m
ode
l
s
a
r
e
s
tr
a
ig
ht
f
or
w
a
r
d
a
nd
obvi
ous
m
e
th
ods
th
a
t
gi
v
e
hi
s
to
r
ic
a
l
ob
s
e
r
va
ti
ons
e
xpone
nt
ia
ll
y di
m
in
is
hi
ng w
e
ig
ht
s
.
2.
R
E
S
E
A
R
C
H
M
E
T
H
O
D
O
L
O
G
Y
W
e
a
r
e
us
in
g
th
e
r
e
a
l
ti
m
e
upda
te
pr
ovi
de
d
by
th
e
B
it
c
oi
n
e
x
c
ha
nge
da
ta
s
e
t.
T
he
da
ta
s
e
t
c
on
s
is
ts
of
th
e
f
ol
lo
w
in
g
e
ig
ht
pa
r
a
m
e
te
r
s
:
ti
m
e
s
ta
m
p,
ope
n,
hi
gh,
lo
w
,
c
lo
s
e
,
vol
um
e
in
c
ur
r
e
nc
y,
vol
um
e
in
b
it
c
oi
n,
a
nd
w
e
ig
ht
e
d
pr
ic
e
.
T
he
c
or
r
e
la
ti
on
he
a
tm
a
p
of
B
it
c
oi
n
da
ta
s
e
t
is
gi
ve
n
in
F
ig
ur
e
1.
T
he
ti
m
e
s
ta
m
p
da
ta
c
ol
um
ns
in
c
lu
de
no
tr
a
ns
a
c
ti
on
s
or
a
c
ti
vi
ty
a
r
e
oc
c
upi
e
d
w
it
h
n
ot
a
num
be
r
(
N
a
N
s
)
.
I
f
a
ti
m
e
s
ta
m
p
is
a
bs
e
nt
,
it
m
a
y
be
due
to
a
n
un
a
nt
ic
ip
a
te
d
te
c
hni
c
a
l
f
a
ul
t
in
d
a
ta
r
e
por
ti
ng
or
c
ol
le
c
ti
on
[
10]
.
F
r
om
K
a
ggl
e
c
ont
e
s
ts
,
w
e
a
m
a
s
s
e
d
b
it
c
oi
n hi
s
to
r
ic
a
l
da
ta
. H
e
r
e
, t
he
hi
s
to
r
ic
a
l
b
it
c
oi
n
da
ta
i
s
a
t
1
-
m
in
ut
e
i
nt
e
r
lu
de
s
.
F
ig
ur
e
1. C
or
r
e
la
ti
on he
a
tm
a
p of
b
it
c
oi
n da
ta
2
.1.
L
on
g s
h
or
t
-
t
e
r
m
m
e
m
o
r
y
P
r
ic
e
s
of
c
r
ypt
oc
ur
r
e
nc
ie
s
f
r
e
que
nt
ly
s
ho
w
lo
ng
-
te
r
m
pa
tt
e
r
ns
a
nd
in
te
r
de
pe
nd
e
nc
e
.
I
n
or
de
r
to
ove
r
c
om
e
th
e
va
ni
s
hi
ng
gr
a
di
e
nt
is
s
ue
th
a
t
pl
a
gue
s
c
onve
nt
io
n
a
l
r
e
c
ur
r
e
nt
ne
ur
a
l
ne
twor
k
s
(
R
N
N
s
)
,
L
S
T
M
s
a
r
e
m
a
de
to
be
a
bl
e
to
le
a
r
n
a
nd
r
e
ta
in
d
a
ta
a
c
r
o
s
s
lo
nge
r
s
e
que
nc
e
s
[
11]
.
L
S
T
M
s
do
not
r
e
qui
r
e
hum
a
n
f
e
a
tu
r
e
e
ngi
ne
e
r
in
g
s
in
c
e
th
e
y
c
a
n
a
ut
om
a
ti
c
a
ll
y
e
xt
r
a
c
t
r
e
le
va
nt
c
ha
r
a
c
te
r
is
ti
c
s
f
r
om
th
e
in
put
da
ta
[
12]
.
F
o
r
b
it
c
oi
n
pr
e
di
c
ti
on
jo
bs
in
vol
vi
ng
c
om
pl
ic
a
te
d
a
nd
hi
gh
-
di
m
e
n
s
io
na
l
da
ta
,
th
is
is
a
dva
nt
a
ge
ou
s
.
In
(
1)
to
(
6
)
s
how
s
th
e
s
im
pl
e
L
S
T
M
f
or
m
ul
a
ti
ons
.
S
in
c
e
L
S
T
M
s
c
a
n
to
le
r
a
te
va
r
yi
ng
in
te
r
va
ls
of
ti
m
e
be
twe
e
n
obs
e
r
va
ti
ons
,
th
e
y
a
r
e
a
good
f
it
f
or
da
ta
on
c
r
ypt
oc
ur
r
e
nc
ie
s
,
s
in
c
e
pr
ic
e
f
lu
c
tu
a
ti
ons
m
a
y
not
ha
ppe
n
a
t
r
e
gul
a
r
in
te
r
va
ls
.
W
he
r
e
is
th
e
f
or
ge
t
ga
te
,
is
th
e
in
put
ga
te
,
is
th
e
out
put
,
̃
is
th
e
ne
w
c
a
ndi
da
te
va
lu
e
,
ℎ
is
th
e
hi
dde
n
s
ta
te
,
is
th
e
s
ig
m
oi
d
in
s
ti
ga
ti
on
f
unc
t
io
n
,
a
nd
ta
nh
is
th
e
hype
r
bol
ic
ta
ng
e
nt
in
s
ti
ga
ti
on f
unc
ti
on.
=
(
.
[
ℎ
−
1
,
]
+
)
(
1)
=
(
.
[
ℎ
−
1
,
]
+
)
(
2)
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
A
c
om
par
at
iv
e
analy
s
i
s
of
e
x
pone
nt
ia
l
s
m
oot
hi
ng m
e
th
od and d
e
e
p l
e
ar
ni
ng
…
(
N
r
us
in
gha T
r
ip
at
hy
)
1403
=
(
.
[
ℎ
−
1
,
]
+
)
(
3)
̃
=
ta
nh
(
.
[
ℎ
−
1
,
]
+
)
(
4)
=
.
−
1
+
.
̃
(
5)
ℎ
=
.
t
a
n
h
(
)
(
6)
2
.
2
.
F
ac
e
b
ook
-
P
r
op
h
e
t
P
r
ophe
t
c
a
n
ha
ndl
e
h
ol
id
a
ys
a
nd
s
e
a
s
on
a
li
ty
o
n
a
w
e
e
kl
y
a
n
d
a
nnu
a
l
ba
s
is
.
T
hi
s
h
e
lp
s
to
c
a
pt
ur
e
r
e
c
ur
r
e
nt
tr
e
nd
s
a
nd
th
e
po
s
s
ib
le
in
f
l
ue
n
c
e
of
hol
i
da
y
s
or
uni
qu
e
oc
c
a
s
io
n
s
on
pr
ic
e
m
ove
m
e
nt
s
in
th
e
c
ont
e
xt
of
b
it
c
oi
n
pr
e
di
c
ti
on
[
13]
.
P
ot
e
nt
i
a
l
c
h
a
ng
e
poi
n
ts
,
or
lo
c
a
ti
on
s
w
he
r
e
th
e
ti
m
e
s
e
r
i
e
s
s
ho
w
s
a
no
ta
bl
e
s
hi
f
t
in
tr
e
nd,
a
r
e
a
ut
om
a
t
ic
a
l
ly
de
t
e
c
te
d
b
y
P
r
op
he
t
.
I
n
(
7)
to
(
9)
s
ho
w
s
th
e
F
B
-
P
r
op
he
t
f
or
m
ul
a
ti
on
s
.
T
hi
s
he
l
ps
to
r
e
c
or
d a
br
upt
f
lu
c
t
ua
ti
o
ns
in
th
e
pr
i
c
e
of
c
r
ypt
o
c
ur
r
e
n
c
ie
s
. B
e
c
a
us
e
P
r
ophe
t'
s
s
e
tt
in
g
s
a
r
e
s
im
pl
e
to
und
e
r
s
t
a
nd
a
nd a
dj
u
s
t,
e
ve
n
i
ndi
vi
dua
l
s
w
it
hout
a
l
ot
of
e
xpe
r
i
e
nc
e
w
i
th
t
im
e
s
e
r
ie
s
m
od
e
ll
in
g m
a
y u
s
e
i
t
[
14]
, [
15]
.
(
)
=
∑
(
.
s
in
(
2
)
+
.
c
o
s
(
2
)
)
=
1
(
7)
W
he
r
e
(
)
is
th
e
s
e
a
s
ona
li
ty
a
t
ti
m
e
,
is
th
e
num
be
r
of
s
e
a
s
ona
l
c
om
pone
nt
s
a
nd
a
r
e
th
e
c
oe
f
f
ic
ie
nt
s
f
or
i
th
c
om
pone
nt
,
is
t
he
pe
r
io
d of
s
e
a
s
ona
li
ty
.
(
)
=
(
)
+
∑
(
.
ℎ
(
−
)
)
−
1
(
8)
W
he
r
e
(
)
is
th
e
tr
e
nd
a
t
ti
m
e
,
(
)
is
th
e
ba
s
e
li
ne
gr
ow
th
,
is
th
e
num
e
r
a
l
of
c
ha
nge
poi
nt
s
,
is
th
e
m
a
gni
tu
de
of
n
-
th
c
ha
nge
poi
nt
.
D
ur
in
g
m
ode
l
t
r
a
in
in
g,
m
a
n
a
gi
ng
hol
id
a
ys
e
nt
a
il
s
de
f
in
in
g
hol
id
a
ys
a
nd
how
th
e
y
a
f
f
e
c
t
th
e
ti
m
e
s
e
r
ie
s
da
ta
.
P
r
ophe
t'
s
a
r
c
hi
te
c
tu
r
e
a
l
lo
w
s
f
or
th
e
in
c
lu
s
io
n
of
hol
id
a
ys
a
nd
ot
he
r
s
pe
c
ia
l
oc
c
a
s
io
ns
in
th
e
f
or
e
c
a
s
ti
ng
pr
oc
e
s
s
.
(
)
is
th
e
obs
e
r
ve
d
va
lu
e
a
t
ti
m
e
,
(
)
is
th
e
le
a
ni
ng,
(
)
is
th
e
s
e
a
s
ona
li
ty
,
ℎ
(
)
is
th
e
d
a
y
of
f
e
f
f
e
c
t
,
a
nd
∈
is
th
e
f
a
ul
t
te
r
m
.
T
he
f
or
m
ul
a
ti
on
f
or
ha
ndl
in
g
hol
id
a
ys
is
gi
ve
n
in
(
9)
.
(
)
=
(
)
+
(
)
+
ℎ
(
)
+
∈
(
9)
2
.
3
.
S
il
ve
r
k
it
e
L
in
ke
dI
n
r
e
le
a
s
e
s
th
e
ti
m
e
-
s
e
r
ie
s
pr
e
di
c
ti
on
li
br
a
r
y
G
r
e
yki
te
to
s
im
pl
if
y
f
or
e
c
a
s
ti
ng
f
or
da
ta
s
c
ie
nt
is
ts
.
T
he
pr
im
a
r
y f
or
e
c
a
s
ti
ng a
lg
or
it
hm
i
n t
hi
s
pa
c
ka
ge
i
s
S
il
ve
r
ki
te
, a
n a
ut
om
a
te
d pr
e
di
c
ti
on
t
e
c
hni
que
.
L
in
ke
dI
n
de
ve
lo
pe
d
G
r
e
yki
te
to
he
lp
e
m
pl
oye
e
s
m
a
ke
in
f
or
m
e
d
de
c
is
io
ns
us
in
g
ti
m
e
-
s
e
r
ie
s
f
or
e
c
a
s
ti
ng
m
ode
ls
.
A
m
ul
ti
tu
de
of
f
e
a
tu
r
e
s
a
r
e
a
ut
om
a
ti
c
a
ll
y
ge
n
e
r
a
te
d
by
S
il
ve
r
ki
te
,
c
a
pt
ur
in
g
va
r
io
us
pr
ope
r
ti
e
s
of
th
e
ti
m
e
s
e
que
nc
e
.
S
il
ve
r
ki
te
'
s
in
c
or
por
a
ti
on
of
va
r
io
us
pe
r
io
di
c
pa
tt
e
r
ns
,
tr
e
nds
,
a
nd
uni
que
oc
c
ur
r
e
nc
e
s
e
na
bl
e
s
f
le
xi
bl
e
f
or
e
c
a
s
ti
ng.
T
hi
s
a
da
pt
a
bi
li
ty
he
lp
s
to
c
a
pt
ur
e
th
e
dyna
m
ic
a
nd
di
ve
r
s
if
ie
d
c
ha
r
a
c
te
r
of
th
e
b
it
c
oi
n
m
a
r
ke
ts
.
B
e
c
a
u
s
e
S
il
ve
r
ki
te
is
bui
lt
to
w
i
th
s
ta
nd
out
li
e
r
s
a
nd
m
is
s
in
g
da
ta
,
it
c
a
n
pr
oduc
e
a
c
c
ur
a
te
f
or
e
c
a
s
ts
e
ve
n
in
s
it
ua
ti
ons
w
he
r
e
th
e
s
uppl
ie
d
da
ta
is
in
c
om
pl
e
te
.
In
(
10)
a
nd
(
11)
s
how
s
th
e
S
il
ve
r
ki
te
f
or
e
c
a
s
t
f
or
m
ul
a
ti
on.
B
y
opt
im
iz
in
g
f
or
f
or
e
c
a
s
t
a
c
c
ur
a
c
y,
S
il
ve
r
ki
te
a
ut
om
a
ti
c
a
ll
y
c
hoos
e
s
th
e
opt
im
a
l
f
or
e
c
a
s
ti
ng
m
ode
l
f
r
om
a
g
r
oup
of
c
a
ndi
da
te
m
ode
ls
.
T
hi
s
c
a
n
be
he
lp
f
ul
in
th
e
pr
e
di
c
ti
on
o
f
c
r
ypt
oc
ur
r
e
nc
ie
s
,
w
he
r
e
s
e
le
c
ti
ng
a
s
ui
ta
bl
e
f
or
e
c
a
s
ti
ng
m
ode
l
m
ig
ht
be
di
f
f
ic
ul
t.
W
he
r
e
̂
is
th
e
c
a
lc
ul
a
te
d
v
a
lu
e
a
t
ti
m
e
,
0
is
th
e
in
te
r
c
e
pt
,
,
,
a
r
e
c
oe
f
f
ic
ie
nt
s
,
∈
is
t
he
e
r
r
or
t
e
r
m
.
̂
=
0
+
∑
=
1
.
,
+
∑
.
ℎ
,
=
1
+
∑
.
,
=
1
+
∈
(
10)
=
(
∑
(
−
̂
,
)
2
=
1
)
(
11)
2
.
4
.
B
id
ir
e
c
t
io
n
al
-
lo
n
g s
h
or
t
-
t
e
r
m
m
e
m
o
r
y
T
he
pr
ic
e
f
lu
c
tu
a
ti
on
s
of
c
r
ypt
oc
ur
r
e
nc
ie
s
f
r
e
que
nt
ly
s
how
lo
ng
-
te
r
m
te
nde
nc
ie
s
a
nd
r
e
la
ti
ons
hi
ps
.
L
ong
-
te
r
m
tr
e
nds
in
th
e
hi
s
to
r
ic
a
l
pr
ic
in
g
da
ta
m
a
y
be
f
ound
a
nd
us
e
d
w
it
h
B
i
-
L
S
T
M
'
s
c
a
pa
c
it
y
to
r
e
c
or
d
knowle
dge
ga
th
e
r
e
d
f
r
om
c
ur
r
e
nt
a
nd
f
ut
ur
e
ti
m
e
s
te
p
s
[
16]
.
N
um
e
r
ous
v
a
r
ia
bl
e
s
im
pa
c
t
c
r
ypt
oc
ur
r
e
nc
y
m
a
r
ke
ts
,
a
nd
th
e
r
e
m
a
y
be
nonl
in
e
a
r
in
te
r
a
c
ti
ons
be
twe
e
n
th
e
s
e
e
le
m
e
nt
s
.
B
e
in
g
a
ki
nd
of
ne
ur
a
l
ne
twor
k,
Bi
-
L
S
T
M
m
a
y
a
dj
u
s
t
to
th
e
dyna
m
ic
a
nd
f
r
e
que
nt
ly
unpr
e
di
c
ta
bl
e
c
ha
r
a
c
te
r
of
c
r
ypt
oc
ur
r
e
nc
y
m
a
r
ke
ts
by
le
a
r
ni
ng
in
tr
ic
a
te
nonl
in
e
a
r
c
or
r
e
la
ti
ons
in
th
e
da
ta
[
17]
,
[
18]
.
In
(
12)
to
(
18)
s
how
s
th
e
Bi
-
L
S
T
M
f
or
e
c
a
s
t
f
or
m
ul
a
ti
on.
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
.
14
, N
o.
2
,
A
pr
il
20
25
:
1401
-
1409
1404
⃖
=
(
⃖
[
ℎ
−
1
,
ℎ
−
1
]
+
⃖
)
(
12)
⃖
=
(
⃖
[
ℎ
−
1
,
ℎ
−
1
]
+
⃖
)
(
13)
⃖
=
(
⃖
[
ℎ
−
1
,
ℎ
−
1
]
+
⃖
)
(
14)
⃖
=
⃖
.
⃖
−
1
+
⃖
.
⃖
⃖
−
1
(
15)
⃖
=
t
a
n
h
(
⃖
[
ℎ
−
1
,
ℎ
−
1
]
+
⃖
)
(
16)
ℎ
⃖
=
⃖
.
t
a
n
h
(
⃖
)
(
17)
=
[
ℎ
⃖
,
ℎ
⃖
]
+
(
18)
3.
P
R
O
P
O
S
E
D
M
O
D
E
L
T
he
dyna
m
ic
s
of
c
r
ypt
oc
ur
r
e
nc
y
m
a
r
ke
tp
la
c
e
s
r
e
s
ul
t
in
f
lu
c
tu
a
ti
ng
va
lu
e
s
due
to
m
a
ny
va
r
ia
bl
e
s
.
A
s
a
ddi
ti
ona
l
da
ta
be
c
om
e
s
a
va
il
a
bl
e
,
de
e
p
le
a
r
ni
ng
m
ode
ls
m
a
y
upda
te
th
e
ir
in
te
r
na
l
r
e
pr
e
s
e
nt
a
ti
ons
to
r
e
f
le
c
t
th
e
s
hi
f
ti
ng
m
a
r
ke
t
c
ondi
ti
ons
[
19]
.
E
xpe
r
ts
a
t
a
ut
om
a
ti
c
a
ll
y
e
xt
r
a
c
ti
ng
pe
r
ti
ne
nt
c
ha
r
a
c
te
r
is
ti
c
s
f
r
om
unpr
oc
e
s
s
e
d
da
ta
a
r
e
d
e
e
p
le
a
r
ni
ng
a
lg
or
it
hm
s
.
I
n
th
e
dom
a
in
of
c
r
ypt
oc
ur
r
e
nc
y
f
or
e
c
a
s
ti
ng,
w
h
e
r
e
in
tr
ic
a
te
c
or
r
e
la
ti
ons
e
xi
s
t
be
twe
e
n
a
tt
r
ib
ut
e
s
a
nd
pr
ic
e
f
lu
c
tu
a
ti
on
s
,
th
is
is
p
a
r
ti
c
ul
a
r
ly
a
dva
nt
a
ge
ou
s
[
20]
,
[
21]
.
T
he
pr
opos
e
d
w
or
kf
lo
w
di
a
gr
a
m
is
gi
ve
n
in
F
ig
ur
e
2.
E
ve
r
y
da
ta
poi
nt
in
th
e
c
ol
le
c
ti
on
ha
s
a
ti
m
e
s
ta
m
p
or
pe
r
io
d
a
tt
a
c
he
d
to
it
th
a
t
in
di
c
a
te
s
th
e
e
xa
c
t
m
om
e
nt
th
e
obs
e
r
va
ti
on
w
a
s
m
a
de
.
T
he
c
on
s
e
que
nc
e
s
of
th
e
D
ic
ke
y
-
F
ul
le
r
t
e
s
t
(
D
F
T
)
a
r
e
s
how
n i
n T
a
bl
e
1.
T
im
e
s
e
r
ie
s
da
ta
m
a
y
b
e
s
t
udi
e
d
to
u
nc
ov
e
r
und
e
r
ly
in
g
dyn
a
m
i
c
s
a
nd
f
or
e
c
a
s
t
f
u
tu
r
e
v
a
lu
e
s
s
in
c
e
it
f
r
e
que
nt
l
y
s
h
ow
s
pa
tt
e
r
n
s
,
tr
e
n
ds
,
or
s
e
a
s
on
a
li
ty
.
C
r
ypt
o
c
ur
r
e
nc
y
m
a
r
ke
t
s
s
how
in
tr
ic
a
te
a
nd
no
n
-
li
ne
a
r
in
te
r
a
c
ti
on
s
be
twe
e
n
s
e
v
e
r
a
l
pr
i
c
e
-
in
f
lu
e
n
c
in
g
e
le
m
e
nt
s
[
22]
.
D
e
e
p
l
e
a
r
ni
n
g
m
od
e
ls
c
a
n
i
de
nt
if
y
c
om
p
li
c
a
te
d
pa
tt
e
r
n
s
a
nd
c
or
r
e
la
ti
on
s
i
n
t
he
da
t
a
be
c
a
u
s
e
th
e
y c
a
n
r
e
pr
e
s
e
nt
c
om
pl
e
x
f
u
nc
ti
o
ns
.
H
um
a
n
f
e
a
tu
r
e
e
ngi
n
e
e
r
in
g
no
lo
ng
e
r
b
e
n
e
c
e
s
s
a
r
y
a
s
d
e
e
p
le
a
r
ni
ng
a
lg
or
it
hm
s
a
ut
om
a
ti
c
a
ll
y
e
xt
r
a
c
t
r
e
le
va
nt
f
e
a
tu
r
e
s
f
r
om
r
a
w
da
t
a
.
I
n t
he
B
it
c
oi
n m
a
r
ke
t,
w
h
e
r
e
p
e
r
ti
ne
nt
a
s
p
e
c
t
s
c
oul
d
be
d
yna
m
i
c
a
nd
c
om
pl
i
c
a
t
e
d, t
hi
s
i
s
us
e
f
ul
[
23]
‒
[
25]
.
F
ig
ur
e
2. P
r
opos
e
d w
or
kf
lo
w
di
a
gr
a
m
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
A
c
om
par
at
iv
e
analy
s
i
s
of
e
x
pone
nt
ia
l
s
m
oot
hi
ng m
e
th
od and d
e
e
p l
e
ar
ni
ng
…
(
N
r
us
in
gha T
r
ip
at
hy
)
1405
T
a
bl
e
1. F
a
ll
out
s
of
D
F
T
P
a
r
a
m
e
t
e
r
s
S
pe
e
d
(
r
pm
)
#L
a
gs
u
s
e
d
28.000000
F
r
e
que
nc
y of
obs
e
r
va
t
i
ons
3151.000000
C
r
i
t
i
c
a
l
va
l
ue
(
1%
)
-
3.531219
C
r
i
t
i
c
a
l
va
l
ue
(
5%
)
-
2.751343
C
r
i
t
i
c
a
l
va
l
ue
(
10%
)
-
2.657257
D
a
t
a
t
ype
f
l
oa
t
64
T
e
s
t
s
t
a
t
i
s
t
i
c
-
1.342317
p
-
va
l
ue
0.584528
4.
R
E
S
U
L
T
S
A
N
A
L
Y
S
I
S
C
r
ypt
oc
ur
r
e
nc
y
f
or
e
c
a
s
ti
ng
c
a
n
a
s
s
is
t
pe
opl
e
a
nd
or
ga
ni
z
a
ti
o
ns
in
e
v
a
lu
a
ti
ng
a
nd
c
ont
r
ol
li
ng
th
e
r
is
ks
br
ought
on
by
pr
ic
e
s
w
in
gs
.
D
e
ve
lo
pi
ng
m
e
th
ods
to
m
it
i
ga
te
r
is
ks
i
s
m
a
de
pos
s
ib
le
by
unde
r
s
ta
ndi
ng
pos
s
ib
le
pr
ic
e
f
lu
c
tu
a
ti
ons
.
P
r
e
di
c
ti
ons
a
r
e
a
c
om
m
on
to
ol
us
e
d
by
in
ve
s
to
r
s
to
m
a
xi
m
iz
e
th
e
ir
bi
tc
oi
n
hol
di
ngs
.
T
he
y
c
a
n
m
odi
f
y
th
e
ir
a
s
s
e
t
por
tf
ol
io
s
to
e
nha
nc
e
p
r
of
it
s
or
r
e
duc
e
lo
s
s
e
s
by
f
or
e
c
a
s
ti
ng
m
a
r
ke
t
f
lu
c
tu
a
ti
ons
.
I
n
th
is
w
or
k
w
e
a
r
e
ta
ki
ng
f
our
di
f
f
e
r
e
nt
m
ode
ls
f
r
om
two
s
e
pa
r
a
te
f
ie
ld
s
a
nd
m
a
ki
ng
th
e
c
om
pa
r
is
on.
W
he
n
c
om
pa
r
e
d
to
ot
he
r
m
ode
l
s
,
th
e
B
i
-
L
S
T
M
m
ode
l
pe
r
f
or
m
s
w
e
ll
.
F
ig
ur
e
3
di
s
pl
a
y
s
th
e
L
S
T
M
a
nt
ic
ip
a
te
d
pr
ic
e
of
b
it
c
oi
n,
w
he
r
e
a
s
F
ig
ur
e
4
di
s
pl
a
ys
th
e
F
B
-
P
r
ophe
t
e
xpe
c
te
d
pr
ic
e
of
b
it
c
oi
n.
I
n
a
s
im
il
a
r
ve
in
,
F
ig
ur
e
s
5
a
nd
6
il
lu
s
tr
a
te
th
e
pr
e
di
c
te
d
pr
ic
e
s
f
or
b
it
c
oi
n
by
S
il
ve
r
ki
te
a
nd
B
i
-
L
S
T
M
,
r
e
s
pe
c
ti
ve
ly
.
F
ig
ur
e
3. L
S
T
M
pr
e
di
c
te
d B
T
C
pr
ic
e
F
ig
ur
e
4. F
B
-
P
r
ophe
t
pr
e
di
c
te
d B
T
C
pr
ic
e
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
.
14
, N
o.
2
,
A
pr
il
20
25
:
1401
-
1409
1406
F
ig
ur
e
5. S
il
ve
r
ki
te
pr
e
di
c
te
d B
T
C
pr
ic
e
F
ig
ur
e
6. B
i
-
L
S
T
M
pr
e
di
c
te
d B
T
C
pr
ic
e
T
a
bl
e
2
di
s
pl
a
y
s
th
e
m
e
a
n a
bs
ol
ut
e
e
r
r
or
(
M
A
E
)
,
m
e
a
n
s
qu
a
r
e
d
e
r
r
or
(
M
S
E
)
,
a
nd
r
oot
m
e
a
n s
qua
r
e
d
e
r
r
or
(
R
M
S
E
)
s
c
or
e
s
of
e
a
c
h
m
ode
l.
T
he
B
i
-
L
S
T
M
m
ode
l
ha
s
a
n
R
M
S
E
s
c
or
e
of
3.639
a
nd
a
m
in
im
a
l
M
A
E
s
c
or
e
of
7.073
w
he
n
c
om
pa
r
e
d
to
th
e
L
S
T
M
,
F
B
-
P
r
ophe
t,
a
nd
S
il
ve
r
ki
te
m
ode
ls
.
B
y
ut
il
iz
in
g
th
e
c
a
pa
c
it
y
to
r
e
c
ogni
z
e
in
tr
ic
a
te
tr
e
nds
in
s
e
que
nt
ia
l
da
ta
a
nd
to
r
e
c
or
d
l
ong
-
te
r
m
de
pe
nde
nc
ie
s
,
th
e
B
i
-
L
S
T
M
m
ode
l
pr
ovi
de
s
va
r
io
us
pot
e
nt
ia
l
a
dva
nt
a
ge
s
f
or
b
it
c
oi
n
pr
ic
e
pr
e
di
c
ti
on.
P
r
ic
e
s
of
c
r
ypt
oc
ur
r
e
nc
ie
s
f
r
e
que
nt
ly
s
how
lo
ng
-
te
r
m
t
r
e
nds
a
nd i
nt
e
r
de
pe
nde
nc
e
. A
s
a
va
r
ia
nt
of
L
S
T
M
w
it
h bi
di
r
e
c
ti
ona
l
pr
oc
e
s
s
in
g,
Bi
-
L
S
T
M
m
ode
ls
a
r
e
a
bl
e
to
m
or
e
a
c
c
ur
a
te
ly
r
e
pr
e
s
e
nt
lo
ng
-
te
r
m
de
pe
nde
nc
ie
s
in
th
e
da
ta
by
c
a
pt
ur
in
g
da
ta
f
r
om
bot
h
pa
s
t
a
nd
f
ut
ur
e
ti
m
e
s
te
ps
.
M
ode
ls
'
M
A
E
,
M
S
E
,
a
nd
R
M
S
E
s
c
or
e
s
a
r
e
pl
ot
te
d
in
a
ba
r
di
a
gr
a
m
c
ons
pi
r
a
c
y
in
F
ig
ur
e
7.
P
r
ognos
e
s
f
or
th
e
f
ut
ur
e
c
a
n
be
us
e
d
by
c
r
ypt
oc
ur
r
e
n
c
y
-
r
e
la
te
d
bus
in
e
s
s
e
s
to
put
h
e
dgi
ng
pl
a
ns
in
to
pl
a
c
e
.
I
n
or
de
r
to
pr
ovi
de
s
om
e
a
m
ount
of
r
is
k
m
a
na
ge
m
e
nt
,
th
is
e
nt
a
il
s
e
s
ta
bl
is
hi
ng
pos
it
io
ns
to
c
ount
e
r
pos
s
ib
le
l
os
s
e
s
f
r
om
unf
a
vor
a
bl
e
m
a
r
ke
t
f
lu
c
tu
a
ti
ons
. F
ut
ur
e
pr
oj
e
c
ti
ons
c
a
n he
lp
bus
in
e
s
s
e
s
t
ha
t
a
dopt
or
us
e
c
r
ypt
oc
ur
r
e
nc
ie
s
m
a
ke
w
is
e
de
c
is
io
n
s
a
bout
f
in
a
nc
i
a
l
pl
a
nni
n
g,
budge
ti
ng,
a
nd
ge
n
e
r
a
l
c
om
pa
ny
s
tr
a
te
gi
e
s
.
F
or
e
c
a
s
ts
f
or
th
e
f
ut
ur
e
he
lp
r
a
i
s
e
a
w
a
r
e
ne
s
s
a
nd
e
duc
a
t
e
pe
o
pl
e
a
bout
th
e
b
it
c
oi
n
in
du
s
tr
y.
E
nt
r
e
pr
e
ne
ur
s
ke
e
p
up
w
it
h
th
e
dyna
m
ic
s
of
th
e
e
ve
r
-
c
ha
ngi
ng
c
r
ypt
oc
ur
r
e
n
c
y
in
dus
tr
y
by
be
in
g
a
w
a
r
e
of
pos
s
ib
le
f
ut
ur
e
de
ve
lo
pm
e
nt
s
.
T
a
bl
e
2.
E
r
r
or
s
c
or
e
of
e
a
c
h m
ode
l
M
ode
l
na
m
e
M
A
E
M
S
E
R
M
S
E
L
S
T
M
14.728
26.525
5.471
FB
-
P
r
ophe
t
8.357
15.354
3.917
S
i
l
ve
r
ki
t
e
7.662
13.948
3.834
Bi
-
L
S
T
M
7.073
11.929
3.639
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
A
c
om
par
at
iv
e
analy
s
i
s
of
e
x
pone
nt
ia
l
s
m
oot
hi
ng m
e
th
od and d
e
e
p l
e
ar
ni
ng
…
(
N
r
us
in
gha T
r
ip
at
hy
)
1407
F
ig
ur
e
7. M
A
E
, M
S
E
,
a
nd R
M
S
E
r
e
s
ul
t
of
m
ode
ls
5.
C
O
N
C
L
U
S
I
O
N
R
e
s
e
a
r
c
h
a
nd
a
n
a
ly
s
is
of
th
e
m
a
r
ke
t
be
ne
f
it
f
r
om
p
r
e
di
c
ti
ons
.
T
he
y
of
f
e
r
pe
r
c
e
pt
io
ns
of
m
a
r
ke
t
m
ood,
pos
s
ib
le
tr
e
nds
,
a
nd
va
r
ia
bl
e
s
a
f
f
e
c
ti
ng
b
it
c
oi
n
pr
ic
in
g.
U
s
in
g
th
is
knowle
dge
to
in
f
or
m
s
tr
a
te
gi
c
de
c
is
io
n
-
m
a
ki
ng
is
be
n
e
f
ic
ia
l.
P
r
e
di
c
ti
ons
c
a
n
he
lp
or
ga
ni
z
a
ti
ons
th
a
t
de
a
l
w
it
h
c
r
ypt
oc
ur
r
e
nc
ie
s
,
in
c
lu
di
ng
e
xc
ha
nge
s
a
nd
pa
ym
e
nt
pr
oc
e
s
s
or
s
,
m
a
ke
c
hoi
c
e
s
.
I
n
th
is
w
or
k
a
ga
in
s
t
th
e
L
S
T
M
,
F
B
-
P
r
ophe
t
,
a
nd
S
il
ve
r
ki
te
m
ode
ls
,
th
e
B
i
-
L
S
T
M
m
ode
l
ha
s
a
m
in
im
a
l
M
A
E
s
c
or
e
of
7.073
a
nd
a
n
R
M
S
E
s
c
or
e
of
3.639.
T
hi
s
m
ig
ht
in
vol
ve
r
is
k
m
a
na
ge
m
e
nt
,
pr
ic
e
ta
c
ti
c
s
,
a
nd
m
a
r
ke
t
f
lu
c
tu
a
ti
on
pr
e
pa
r
a
ti
on.
P
r
e
di
c
ti
ons
c
a
n
be
us
e
d
by
or
ga
ni
z
a
ti
ons
th
a
t
de
a
l
w
it
h
c
r
ypt
oc
ur
r
e
nc
ie
s
,
s
uc
h
a
s
c
or
por
a
ti
ons
a
nd
in
s
ti
tu
ti
ona
l
in
ve
s
to
r
s
,
to
c
r
e
a
te
he
dgi
ng
s
tr
a
te
gi
e
s
th
a
t
w
oul
d
le
s
s
e
n
th
e
r
is
k
of
lo
s
s
e
s
br
ought
on
by
m
a
r
ke
t
vol
a
ti
li
ty
.
F
in
a
nc
ia
l
m
a
r
ke
ts
a
r
e
s
o
dyna
m
ic
a
nd
c
om
pl
ic
a
te
d
th
a
t
it
is
a
lw
a
ys
di
f
f
ic
ul
t
to
f
or
e
c
a
s
t
th
e
ir
m
ove
m
e
nt
s
,
e
s
pe
c
ia
ll
y
th
os
e
of
b
it
c
oi
n
va
lu
e
s
.
F
ut
ur
e
de
v
e
lo
pm
e
nt
s
in
d
a
ta
s
c
ie
nc
e
a
nd
a
th
or
ough
c
om
pr
e
he
ns
io
n
of
th
e
dyna
m
ic
c
r
ypt
oc
ur
r
e
nc
y
e
nvi
r
onm
e
nt
w
il
l
pr
oba
bl
y
be
us
e
d
to
a
nt
ic
ip
a
te
b
it
c
oi
n
pr
ic
e
s
.
S
c
hol
a
r
s
a
nd
e
xpe
r
ts
in
th
is
dom
a
in
w
il
l
pe
r
s
is
t
in
in
ve
s
ti
ga
ti
ng
in
ve
nt
iv
e
m
e
th
ods
to
a
ugm
e
nt
th
e
pr
e
c
is
io
n
a
nd
de
pe
nda
bi
li
ty
of
pr
e
di
c
ti
ons
f
or
t
he
b
it
c
oi
n.
R
E
F
E
R
E
N
C
E
S
[
1]
O
. K
r
a
a
i
j
e
ve
l
d a
nd J
. D
.
S
m
e
dt
, “
T
he
pr
e
di
c
t
i
ve
pow
e
r
of
publ
i
c
T
w
i
t
t
e
r
s
e
nt
i
m
e
nt
f
or
f
or
e
c
a
s
t
i
ng c
r
ypt
oc
ur
r
e
nc
y pr
i
c
e
s
,”
J
our
nal
of
I
nt
e
r
nat
i
onal
F
i
nanc
i
al
M
ar
k
e
t
s
, I
ns
t
i
t
ut
i
ons
and M
one
y
, vol
. 65, 2020, doi
:
10.1016/
j
.i
nt
f
i
n.2020.101188.
[
2]
B
.
C
he
n,
F
.
W
e
i
,
a
nd
C
.
G
u,
“
B
i
t
c
oi
n
t
he
f
t
de
t
e
c
t
i
on
b
a
s
e
d
on
s
upe
r
vi
s
e
d
m
a
c
hi
ne
l
e
a
r
ni
ng
a
l
gor
i
t
hm
s
,”
Se
c
ur
i
t
y
an
d
C
om
m
uni
c
at
i
on N
e
t
w
o
r
k
s
, vol
. 2021, 2021, doi
:
10.1155/
2021/
6643763.
[
3]
S
.
L
a
hm
i
r
i
a
nd
S
.
B
e
ki
r
os
,
“
C
r
ypt
oc
ur
r
e
nc
y
f
or
e
c
a
s
t
i
ng
w
i
t
h
de
e
p
l
e
a
r
ni
n
g
c
ha
ot
i
c
n
e
ur
a
l
ne
t
w
or
ks
,”
C
haos
,
Sol
i
t
ons
and
F
r
ac
t
al
s
, vol
. 118, pp. 35
–
40, 2019, doi
:
10.1016/
j
.c
ha
os
.2018.11.014.
[
4]
X
.
S
un,
M
.
L
i
u,
a
nd
Z
.
S
i
m
a
,
“
A
nove
l
c
r
ypt
oc
ur
r
e
nc
y
pr
i
c
e
t
r
e
nd
f
or
e
c
a
s
t
i
n
g
m
ode
l
ba
s
e
d
on
L
i
ght
G
B
M
,”
F
i
nanc
e
R
e
s
e
a
r
c
h
L
e
t
t
e
r
s
, vol
. 32, 2020, doi
:
10.1016/
j
.f
r
l
.2018.12.032.
[
5]
S
. B
or
odo, S
. M
.
S
ha
m
s
uddi
n, S
. H
a
s
a
n, a
nd S
. M
.
B
or
odo, “
B
i
g da
t
a
pl
a
t
f
or
m
s
a
nd t
e
c
hni
que
s
,”
I
ndone
s
i
an J
ou
r
nal
of
E
l
e
c
t
r
i
c
a
l
E
ngi
ne
e
r
i
ng and C
om
put
e
r
Sc
i
e
n
c
e
, vol
. 1, no. 1, pp. 191
–
200, 2017, doi
:
10.1
1591/
i
j
e
e
c
s
.v1.i
1.pp191
-
200.
[
6]
H
. I
. F
a
w
a
z
, G
. F
or
e
s
t
i
e
r
, J
. W
e
be
r
,
L
. I
doum
gha
r
, a
nd P
. A
. M
ul
l
e
r
, “
D
e
e
p l
e
a
r
ni
ng f
or
t
i
m
e
s
e
r
i
e
s
c
l
a
s
s
i
f
i
c
a
t
i
on:
a
r
e
vi
e
w
,
”
D
at
a
M
i
ni
ng and K
now
l
e
dge
D
i
s
c
ov
e
r
y
, vol
. 33, no. 4, pp. 917
–
963, 2019, doi
:
10.1
007/
s
10618
-
019
-
00619
-
1.
[
7]
M
.
R
.
C
he
r
a
t
i
,
A
.
H
a
e
r
i
,
a
nd
S
.
F
.
G
ha
nn
a
dpour
,
“
C
r
ypt
oc
ur
r
e
nc
y
di
r
e
c
t
i
on
f
o
r
e
c
a
s
t
i
ng
us
i
ng
de
e
p
l
e
a
r
ni
ng
a
l
gor
i
t
hm
s
,”
J
ou
r
nal
of
St
at
i
s
t
i
c
al
C
om
put
at
i
on and Si
m
ul
at
i
on
, vol
. 91, no. 12, pp. 2475
–
2489, 2021, doi
:
10.1080/
00949655.2021.1899179.
[
8]
S
.
K
.
N
a
ya
k,
A
.
K
.
N
a
ya
k,
S
.
M
i
s
hr
a
,
P
.
M
oha
nt
y,
N
.
T
r
i
pa
t
hy,
a
nd S
.
P
r
us
t
y, “
I
m
pr
ovi
ng
K
ui
di
gi
t
r
e
c
ogni
t
i
on
t
h
r
ough
m
a
c
hi
ne
l
e
a
r
ni
ng
a
nd
da
t
a
a
ugm
e
nt
a
t
i
on
t
e
c
hni
que
s
,”
I
ndone
s
i
an
J
our
nal
of
E
l
e
c
t
r
i
c
al
E
ngi
ne
e
r
i
ng
and
C
om
put
e
r
Sc
i
e
n
c
e
,
vol
.
35,
no.
2
,
pp. 867
–
877, 2024, doi
:
10.11591/
i
j
e
e
c
s
.v35.i
2.pp867
-
877.
[
9]
H
.
S
e
ba
s
t
i
ã
o
a
nd
P
.
G
odi
nho,
“
F
or
e
c
a
s
t
i
ng
a
nd
t
r
a
di
ng
c
r
ypt
oc
ur
r
e
nc
i
e
s
w
i
t
h
m
a
c
hi
ne
l
e
a
r
ni
ng
unde
r
c
ha
ngi
ng
m
a
r
ke
t
c
ondi
t
i
ons
,”
F
i
nanc
i
al
I
nnov
at
i
on
, vol
. 7, no. 1, 2021, doi
:
10.1186/
s
40854
-
020
-
00217
-
x.
[
10]
M
.
I
qba
l
,
M
.
S.
I
qb
a
l
,
F
.
H
.
J
a
s
ka
ni
,
K
.
I
q
ba
l
,
a
nd
A
.
H
a
s
s
a
n,
“
T
i
m
e
-
s
e
r
i
e
s
pr
e
di
c
t
i
on
o
f
c
r
yp
t
oc
u
r
r
e
nc
y
m
a
r
ke
t
us
i
ng
m
a
c
h
i
ne
l
e
a
r
n
i
ng
t
e
c
hn
i
q
ue
s
,
”
E
A
I
E
nd
or
s
e
d
T
r
ans
a
c
t
i
ons
o
n C
r
e
at
i
v
e
T
e
c
hn
ol
ogi
e
s
,
vo
l
.
8,
no.
2
8,
20
21,
d
oi
:
1
0.
410
8/
e
a
i
.7
-
7
-
2
02
1.1
70
286
.
[
11]
N
. T
r
i
pa
t
h
y
, S
.
H
o
t
a
,
a
n
d
D
.
M
i
s
h
r
a
, “
P
e
r
f
o
r
m
a
nc
e
a
n
a
l
ys
i
s
o
f
bi
t
c
o
i
n
f
o
r
e
c
a
s
t
i
ng
u
s
i
n
g de
e
p l
e
a
r
n
i
ng
t
e
c
hn
i
qu
e
s
,
”
I
nd
o
ne
s
i
a
n J
o
ur
nal
o
f
E
l
e
c
t
r
i
c
a
l
E
n
gi
ne
e
r
i
n
g
a
nd
C
om
pu
t
e
r
Sc
i
e
nc
e
,
v
o
l
.
31
,
n
o.
3
,
p
p.
15
15
–
15
22
,
2
02
3
,
d
oi
:
1
0.
1
15
91
/
i
j
e
e
c
s
.
v
31
.
i
3
.p
p
15
15
-
1
52
2.
[
12]
T
.
A
s
hf
a
q
e
t
al
.
,
“
A
m
a
c
hi
ne
l
e
a
r
ni
ng
a
nd
bl
oc
kc
ha
i
n
ba
s
e
d
e
f
f
i
c
i
e
nt
f
r
a
ud
de
t
e
c
t
i
on
m
e
c
ha
ni
s
m
,”
Se
n
s
or
s
,
vol
.
22,
no.
19,
2022,
doi
:
10.3390/
s
22197162.
[
13]
H
.
K
a
ne
z
a
s
hi
,
T
.
S
uz
um
ur
a
,
X
.
L
i
u,
a
nd
T
.
H
i
r
of
uc
hi
,
“
E
t
he
r
e
um
f
r
a
ud
de
t
e
c
t
i
on
w
i
t
h
he
t
e
r
oge
ne
ous
gr
a
ph
ne
ur
a
l
ne
t
w
or
ks
,”
i
n
I
nt
e
r
nat
i
onal
W
or
k
s
hop on M
i
ni
ng and L
e
ar
ni
ng w
i
t
h G
r
aph
s
, 2018, pp. 1
–
8.
[
14]
A
. A
.
A
m
po
ns
a
h
, A
.
F
. A
de
ko
ya
,
a
n
d B
.
A
.
W
e
yo
r
i
,
“
A
no
ve
l
f
r
a
u
d de
t
e
c
t
i
o
n a
n
d
p
r
e
ve
n
t
i
on
m
e
t
hod
f
o
r
he
a
l
t
hc
a
r
e
c
l
a
i
m
p
r
oc
e
s
s
i
n
g
us
i
n
g
m
a
c
h
i
ne
l
e
a
r
ni
ng
a
n
d b
l
o
c
kc
ha
i
n
t
e
c
h
no
l
og
y,”
D
e
c
i
s
i
on
A
na
l
y
t
i
c
s
J
o
ur
na
l
,
vol
.
4,
20
22,
d
oi
:
1
0.
101
6/
j
.
da
j
ou
r
.
202
2.
100
12
2.
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
.
14
, N
o.
2
,
A
pr
il
20
25
:
1401
-
1409
1408
[
15]
N
.
T
r
i
pa
t
hy,
S
.
K
.
B
a
l
a
ba
nt
a
r
a
y,
S
.
P
a
r
i
da
,
a
nd
S
.
K
.
N
a
ya
k,
“
C
r
ypt
oc
ur
r
e
nc
y
f
r
a
ud
de
t
e
c
t
i
on
t
hr
ough
c
l
a
s
s
i
f
i
c
a
t
i
on
t
e
c
hni
qu
e
s
,
”
I
nt
e
r
nat
i
onal
J
our
nal
of
E
l
e
c
t
r
i
c
al
and
C
om
put
e
r
E
ngi
ne
e
r
i
ng
,
vol
.
14,
no.
3,
pp.
2918
–
2926,
2024,
doi
:
10.11591/
i
j
e
c
e
.v14i
3.pp2918
-
2926.
[
16]
Y
.
K
um
a
r
,
“
A
I
t
e
c
hni
que
s
i
n
bl
oc
kc
h
a
i
n
t
e
c
hnol
ogy
f
or
f
r
a
ud
de
t
e
c
t
i
on
a
nd
p
r
e
ve
nt
i
on,”
i
n
Se
c
ur
i
t
y
E
ngi
ne
e
r
i
ng
f
or
E
m
be
dde
d
and C
y
be
r
-
P
hy
s
i
c
al
Sy
s
t
e
m
s
,
C
R
C
P
r
e
s
s
, 2023.
[
17]
V
. P
a
t
e
l
, L
.
P
a
n, a
nd
S
. R
a
j
a
s
e
ga
r
a
r
, “
G
r
a
ph
de
e
p l
e
a
r
ni
ng ba
s
e
d
a
nom
a
l
y d
e
t
e
c
t
i
on i
n e
t
he
r
e
um
bl
oc
kc
ha
i
n
ne
t
w
or
k,”
i
n
N
e
t
w
o
r
k
and Sy
s
t
e
m
Se
c
u
r
i
t
y
, C
ha
m
, S
w
i
t
z
e
r
l
a
nd:
S
pr
i
nge
r
, 2020, pp. 132
–
148, doi
:
10.1007/
978
-
3
-
030
-
65745
-
1_8.
[
18]
K
.
A
r
i
ya
,
S
.
C
ha
na
i
m
,
a
nd
A
.
Y
.
D
a
w
od,
“
C
or
r
e
l
a
t
i
on
be
t
w
e
e
n
c
a
pi
t
a
l
m
a
r
ke
t
s
a
nd
c
r
ypt
oc
ur
r
e
nc
y:
i
m
pa
c
t
of
t
he
c
or
ona
vi
r
us
,”
I
nt
e
r
nat
i
onal
J
our
nal
of
E
l
e
c
t
r
i
c
al
and
C
om
put
e
r
E
ngi
ne
e
r
i
ng
,
vol
.
13,
no.
6,
pp.
6637
–
6645,
2023,
doi
:
10.11591/
i
j
e
c
e
.v13i
6.pp6637
-
6645.
[
19]
I
.
N
a
s
i
r
t
a
f
r
e
s
hi
,
“
F
or
e
c
a
s
t
i
ng
c
r
ypt
oc
ur
r
e
nc
y
pr
i
c
e
s
us
i
ng
r
e
c
ur
r
e
nt
ne
ur
a
l
n
e
t
w
or
k
a
nd
l
ong
s
hor
t
-
t
e
r
m
m
e
m
or
y,”
D
at
a
and
K
now
l
e
dge
E
ngi
ne
e
r
i
ng
, vol
. 139, 2022, doi
:
10.1016/
j
.da
t
a
k.2022.102009.
[
20]
N
.
T
r
i
pa
t
hy,
S
.
H
ot
a
,
S
.
P
r
us
t
y,
a
nd
S
.
K
.
N
a
ya
k,
“
P
e
r
f
or
m
a
nc
e
a
na
l
y
s
i
s
of
d
e
e
p
l
e
a
r
ni
ng
t
e
c
hni
que
s
f
or
t
i
m
e
s
e
r
i
e
s
f
or
e
c
a
s
t
i
ng,”
i
n
2023
I
nt
e
r
nat
i
onal
C
onf
e
r
e
nc
e
i
n
A
dv
anc
e
s
i
n
P
ow
e
r
,
Si
gnal
,
and
I
nf
or
m
at
i
on
T
e
c
hnol
ogy
(
A
P
S
I
T
)
, I
E
E
E
,
2023,
pp.
639
–
644,
doi
:
10.1109/
A
P
S
I
T
58554.2023.10201734.
[
21]
R
.
F
.
I
br
a
hi
m
,
A
.
M
.
E
l
i
a
n,
a
nd
M
.
A
ba
bne
h,
“
I
l
l
i
c
i
t
a
c
c
ount
de
t
e
c
t
i
on
i
n
t
he
e
t
he
r
e
um
bl
oc
kc
ha
i
n
us
i
ng
m
a
c
hi
ne
l
e
a
r
ni
ng,”
i
n
2021
I
nt
e
r
nat
i
onal
C
onf
e
r
e
nc
e
on
I
nf
or
m
at
i
on
T
e
c
hnol
ogy
(
I
C
I
T
)
,
I
E
E
E
,
2021,
pp.
488
–
493,
doi
:
10.1109/
I
C
I
T
52682.2021.9491653.
[
22]
O
.
I
.
J
a
c
i
nt
a
,
A
.
E
.
O
m
ol
a
r
a
,
M
.
A
l
a
w
i
da
,
O
.
I
.
A
bi
odun,
a
nd
A
.
A
l
a
bdul
t
i
f
,
“
D
e
t
e
c
t
i
on
of
P
onz
i
s
c
h
e
m
e
on
E
t
he
r
e
um
u
s
i
ng
m
a
c
hi
ne
l
e
a
r
ni
ng a
l
gor
i
t
hm
s
,”
Sc
i
e
nt
i
f
i
c
R
e
por
t
s
, vol
. 13, no. 1, 2023, doi
:
10.1
038/
s
41598
-
023
-
45275
-
0.
[
23]
T
.
H
u
e
t
al
.
,
“
T
r
a
ns
a
c
t
i
on
-
ba
s
e
d
c
l
a
s
s
i
f
i
c
a
t
i
on
a
nd
de
t
e
c
t
i
on
a
ppr
oa
c
h
f
or
E
t
he
r
e
um
s
m
a
r
t
c
ont
r
a
c
t
,”
I
nf
or
m
at
i
on
P
r
oc
e
s
s
i
ng
and
M
anage
m
e
nt
, vol
. 58, no. 2, 2021, doi
:
10.1016/
j
.i
pm
.2020.102462.
[
24]
N
.
T
r
i
pa
t
hy,
P
.
S
a
t
a
pa
t
hy,
S
.
H
ot
a
,
S
.
K
.
N
a
ya
k,
a
nd
D
.
M
i
s
hr
a
,
“
E
m
pi
r
i
c
a
l
f
o
r
e
c
a
s
t
i
ng
a
na
l
ys
i
s
of
bi
t
c
oi
n
pr
i
c
e
s
:
A
c
om
pa
r
i
s
on
of
m
a
c
hi
ne
l
e
a
r
ni
ng,
de
e
p
l
e
a
r
ni
ng,
a
nd
e
ns
e
m
bl
e
l
e
a
r
ni
ng
m
ode
l
s
,
”
I
nt
e
r
nat
i
onal
J
our
nal
of
E
l
e
c
t
r
i
c
al
and
C
o
m
put
e
r
E
ngi
ne
e
r
i
ng Sy
s
t
e
m
s
, vol
. 15, no. 1, pp. 21
–
29, 2024, doi
:
10.32985/
i
j
e
c
e
s
.15.1
.3.
[
25]
R
.
T
a
n,
Q
.
T
a
n,
P
.
Z
ha
ng,
a
nd
Z
.
L
i
,
“
G
r
a
ph
ne
ur
a
l
ne
t
w
or
k
f
o
r
e
t
he
r
e
u
m
f
r
a
ud
de
t
e
c
t
i
on,”
i
n
2021
I
E
E
E
I
nt
e
r
nat
i
onal
C
onf
e
r
e
nc
e
on B
i
g K
now
l
e
dge
(
I
C
B
K
)
, I
E
E
E
, 2021, pp. 78
–
85, doi
:
10.1109/
I
C
K
G
52313.2021.00020.
B
I
O
G
R
A
P
H
I
E
S
O
F
A
U
T
H
O
R
S
Nrusingha
Tripathy
received
an
MCA
degree
in
computer
science
from
Ravensha
w
Universi
ty,
Cuttack,
Odisha,
India,
in
2018,
and
an
M.T
ech.
degree
in
computer
science
from Utkal U
niversity
, Bhubaneswar,
Odisha,
India, in
2020.
Currently,
he is purs
uing
a
Ph.D.
in
computer
science
and
engineerin
g
at
the
Institute
of
T
echnical
Education
and
Resear
ch
(ITER)
at
Siksha
‘O’
Anusandh
an
(Deeme
d
to
be)
Universi
ty,
Bhubanes
war,
India.
His research interests in
clude
machine learning,
deep learning and
fin
ancial market prediction
.
He
has
published
twelve
con
ference
papers
and
ten
journal
pap
ers.
With
over
five
years
of
teaching e
xperienc
e
.
He can be contacted
at
email
:
nrusinghatri
pathy654@
gmail.com.
Dr.
Debahuti
Mishra
received
an
M.Tech
.
in
computer
science
a
nd
engineerin
g
from
KIIT
Deemed
to
be
University,
Bhubaneswar,
India,
in
2006,
a
nd
a
Ph.D.
from
Siksha
‘O’
Anusandhan
(Deemed
to
be)
University,
Bhubaneswar,
India,
in
2011.
She
is
currently
Profes
sor
and
head
of
the
Department
of
Computer
Science
and
Engineering
at
ITER,
Siksha
‘O’
Anusandhan
(Deemed
to
be)
University.
Her
research
interest
s
include
data
mining,
financial
market
prediction,
and
image
processing.
She
has
authored
e
ight
books
and
over
260
research papers. S
he can be contacted
at
email:
debahutimishra@
soa.a
c.in.
Dr.
Sarbeswara
Hota
received
his
MCA
from
the
Nation
al
Institut
e
of
Technology
(NIT),
Rourkela,
India,
in
2002,
and
his
Ph.D.
in
computer
science
and
engineerin
g
from
Siksha
‘O’
Anusandhan
(Deemed
to
be)
University
in
2019.
He
is
currently
an
Associate
Professor
in
the
Department
of
Computer
Application
at
the
same
University.
His
research
interests
include
data
mining,
machine
learning,
and
deep
learning.
He
has
published
30
papers
in
various
international
journals
and
conferences.
He
can
be
contacted
at
sarbeswara
hota@
soa.ac.in.
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
A
c
om
par
at
iv
e
analy
s
i
s
of
e
x
pone
nt
ia
l
s
m
oot
hi
ng m
e
th
od and d
e
e
p l
e
ar
ni
ng
…
(
N
r
us
in
gha T
r
ip
at
hy
)
1409
Dr.
Mandakini
Priyadarshani
Behera
received
an
M.Tech
.
in
co
mputer
science
and engin
eering
from Gandhi
Institut
e of Technolog
ical Advancement
, Bhubaneswar,
India, in
2018,
and
a
Ph.D.
from
Siksha
‘O’
Anusandhan
(Deemed
to
be)
U
niversity,
Bhubaneswar,
India,
in
2024.
She
is
currently
a
Faculty
Associate
of
School
of
C
omputer
Engineering
at
KIIT,
(Deemed
to
be)
University,
Bhubaneswar,
Odisha.
Her
res
earch
interests
include
machine
learning
and
deep
learning
.
She
has
authored
ov
er
10
rese
arch
papers.
She
can
be
contacted
at
email:
mandakini.be
herafc
s@
kiit.ac.in.
Gobinda
Chandra
Das
received
his
MCA
degree
from
IPSAR
C
ollege
Cuttack
in
2016,
O
disha
and
an
M.Tech
.
(
computer
science
and
engineerin
g
)
form
CIME
C
ollege
i
n
2019
Bhubaneswar,
Odiaha.
Currently
working
as
Assistant
Prof
essor
in
D
epartment
of
Computer
Science
and
Enginee
ring
KL
(Deemed
to
be)
University
Vijaywada
and
p
ursuing
(Ph
.
D
.
)
in
KL
(Deemed
to)
University
Vijayawada
(Andhrapradesh)
India.
He
has
published
2
patents,
2
journal
publications
,
and
also
have
6yrs+
teaching
e
xperiences.
He
can
be
contacted
at
e
mail
:
govindachandrad@gmail.com.
Sasanka Se
khar Dal
ai
received the M.
Tech
.
degree in
computer
science
from the
College
of
Enginee
ring
&
Technol
ogy,
BPUT
Bhubanes
war,
Odisha,
India.
He
is
curre
ntly
pursuing his Ph.D. in
computer
science and
engineeri
ng
at the In
stitut
e
of Techni
cal Educati
on
and
Research
(I.T.E.R.)
in
Siksha
O
Anusandhan
Deemed
to
be
U
niversity,
Bhubaneswar,
India,
and
has
published
one
conference
and
two
journal
papers.
Alth
ough,
he
has
9+
years
o
f
teaching e
xperienc
e. His re
search
interests inc
lude
machine le
arning, d
eep lea
rning, and ima
ge
processing
. He can be contac
ted at email: sasanka.sekhar.d
alai@
gmail.com.
Subrat
Kumar
Nayak
received
an
MCA
from
Biju
Patnaik
University
o
f
Technology,
Odisha,
in
2010
and
an
M.Tech
.
in
computer
scien
ce
from
Utkal
University,
Bhubanes
war,
in
2012.
He
is
curre
ntly
pursuing
a
Ph.D.
in
computer
science
and
engineerin
g
at
Siksha
‘O’
Anusandhan
(Deemed
to
be)
University,
Bhubaneswar,
India.
He
has
published
10
papers
in
international
journals
and
conferences
and
qualified
U
GC
NET
in
2012.
With
over
six
years
of
academic
experience
and
three
years
of
governmen
t
experience
.
H
e
can
be
contacted
at
email:
subratsilicon28@gmail.com.
Evaluation Warning : The document was created with Spire.PDF for Python.