T
E
L
K
O
MN
I
K
A
T
el
eco
m
m
u
n
i
ca
t
i
o
n
,
C
o
m
p
u
t
i
n
g
,
E
l
ect
ro
n
i
cs
a
n
d
C
o
n
t
ro
l
Vo
l
.
19
, N
o
.
5
,
O
ct
o
b
er
2021
, p
p
.
1735
~
1746
I
S
S
N
:
1693
-
6930,
a
c
c
r
e
di
t
e
d F
i
r
s
t
G
r
a
de
by K
e
m
e
nr
i
s
t
e
kdi
kt
i
,
D
e
c
r
e
e
N
o:
21/
E
/
K
P
T
/
2018
D
O
I
:
10.
12928/
T
E
L
K
O
M
N
I
K
A
.
v19i
5.
21059
1735
Jou
r
n
al
h
om
e
page
:
ht
t
p:
/
/
j
our
nal
.
uad
.
ac
.
i
d/
i
nde
x
.
php/
T
E
L
K
O
M
N
I
K
A
L
ar
ge
s
c
al
e
d
at
a a
n
al
ys
i
s
u
s
i
n
g M
L
l
i
b
A
hm
e
d H
us
s
e
i
n A
l
i
1
,
M
aan
N
aw
af
A
b
b
od
2
,
Mo
h
a
m
m
ed
K
h
a
m
ees
K
h
a
l
eel
3
,
M
os
t
af
a A
b
d
u
l
gh
a
f
oor
M
oh
am
m
e
d
4
,
To
l
e
Sut
i
k
no
5
1
I
C
C
I
,
I
nf
or
m
a
t
ic
s I
n
st
it
ute
f
or
P
os
tgr
a
dua
te
S
t
ud
ie
s,
B
a
ghd
a
d,
I
r
a
q
1
C
om
pu
te
r
S
c
ie
nc
e
De
pa
r
tm
e
nt,
AL
S
a
la
m
U
ni
ve
r
s
it
y C
oll
e
ge
,
B
a
g
hda
d,
Iraq
2,
4
I
m
a
m
Aa
d
ha
m
U
ni
ve
r
s
it
y
C
ol
le
ge
,
I
r
a
q
1,
2,
3
De
pa
r
tm
e
n
t of
C
om
pu
te
r
,
C
ol
le
ge
of
Ed
uc
a
t
io
n,
AL
-
I
r
a
qia
U
ni
ve
r
s
it
y,
Bag
h
da
d,
Iraq
5
De
pa
r
tm
e
n
t of
E
le
c
t
ic
a
l E
ng
in
e
e
r
i
ng,
U
ni
ve
r
s
ita
s Ahm
a
d Da
h
la
n,
Y
og
ya
ka
r
ta
,
I
nd
one
sia
A
rt
i
cl
e I
n
f
o
AB
S
T
RACT
A
r
tic
le
h
is
to
r
y
:
R
ecei
v
ed
J
ul
9,
2021
R
ev
i
s
ed
S
e
p 4,
2021
A
ccep
t
ed
S
e
p 12,
2021
R
e
c
e
nt a
d
va
nc
e
m
e
nt
s in t
he
in
te
r
ne
t,
soc
ia
l m
e
d
ia
,
a
nd i
nte
r
ne
t of
t
hi
ng
s
(
I
oT)
de
vic
e
s ha
ve
si
gn
if
ic
a
nt
ly
inc
r
e
a
se
d t
he
a
m
o
un
t o
f
da
ta
ge
ne
r
a
te
d in a
va
r
ie
ty
of
f
or
m
a
t
s.
The
da
ta
m
u
st
be
c
on
ve
r
te
d
in
to f
or
m
a
t
s t
ha
t
is
e
a
s
il
y
ha
n
dle
d b
y t
he
da
ta
a
na
l
ys
is
t
e
c
hn
iq
ue
s.
I
t i
s m
a
t
he
m
a
t
ic
a
l
ly a
nd
ph
ys
ic
a
l
ly
e
xpe
ns
ive
to
a
pp
ly
m
a
c
hi
ne
le
a
r
ni
ng
a
l
gor
it
hm
s to bi
g a
nd
c
om
p
lic
a
te
d da
ta
se
ts.
I
t
is a
r
e
so
ur
c
e
-
in
te
n
si
ve
pr
oc
e
s
s t
ha
t
ne
c
e
ss
ita
te
s a
h
uge
a
m
o
un
t of
log
ic
a
l a
n
d p
hy
sic
a
l r
e
s
our
c
e
s.
M
a
c
h
ine
le
a
r
n
in
g i
s
a
so
ph
is
tic
a
te
d da
ta
a
na
l
yt
ic
s te
c
hn
ol
og
y t
ha
t ha
s ga
ine
d in im
por
ta
nc
e
a
s a
r
e
su
lt of
t
he
m
a
s
si
ve
a
m
ou
nt of
da
ta
ge
ne
r
a
te
d
da
i
ly t
ha
t
ne
e
d
s t
o be
e
xa
m
ine
d.
Apa
c
he
S
pa
r
k
m
a
c
hi
ne
le
a
r
ni
ng li
br
a
r
y (
M
L
li
b)
i
s o
ne
of
the
b
ig
da
ta
a
na
ly
si
s p
la
tf
or
m
s
tha
t pr
o
vi
de
s a
v
a
r
ie
ty of
ou
ts
ta
n
di
ng f
unc
ti
on
s f
or
va
r
i
ous m
a
c
hi
ne
le
a
r
ni
ng
ta
s
ks,
spa
nn
in
g f
r
om
c
la
s
sif
ic
a
t
io
n to r
e
gr
e
s
si
on a
n
d dim
e
ns
io
n r
e
duc
t
io
n.
F
r
om
a
c
om
pu
ta
t
io
na
l
s
ta
n
dp
oi
nt,
th
is r
e
se
a
r
c
h
in
ve
sti
ga
te
d A
pa
c
he
S
pa
r
k
M
L
lib
2.
0 a
s a
n
ope
n s
our
c
e
,
a
ut
on
om
o
us,
sc
a
la
b
le
,
a
n
d d
is
tr
i
bu
te
d
le
a
r
n
in
g
l
i
b
r
a
r
y
.
S
e
v
e
r
a
l
r
e
a
l
-
wor
ld m
a
c
hi
ne
le
a
r
ni
ng e
xpe
r
im
e
nts a
r
e
c
a
r
r
ie
d o
ut
in
or
de
r
to e
va
l
ua
t
e
t
he
pr
ope
r
tie
s
of
t
he
pla
tf
or
m
o
n a
q
ua
l
ita
ti
ve
a
nd
qua
nt
ita
ti
ve
le
ve
l.
S
om
e
of
the
f
u
nda
m
e
n
ta
l c
o
nc
e
p
ts
a
nd a
p
pr
oa
c
he
s f
or
de
ve
lo
pi
n
g a
sc
a
la
ble
da
ta
m
ode
l i
n a
dis
tr
i
bu
te
d e
nvi
r
o
nm
e
n
t a
r
e
a
lso
dis
c
us
se
d.
Ke
y
wo
r
d
s
:
B
i
g
d
at
a
D
at
a an
al
y
s
i
s
M
ach
i
n
e l
ear
n
i
n
g
O
pe
n s
our
c
e
P
ar
al
l
el
p
r
o
ces
s
i
n
g
S
pa
r
k M
L
l
i
b
T
his
is
a
n
o
pe
n
ac
c
e
s
s
ar
tic
le
u
nde
r
the
CC
B
Y
-
SA
lic
e
n
se
.
C
or
r
e
s
pon
di
n
g A
u
t
h
or
:
A
hm
e
d H
us
s
e
i
n A
l
i
D
e
pa
r
t
m
e
nt
of
C
om
put
e
r
AL
-
I
r
a
q
ia
U
n
iv
e
r
s
ity
A
'dha
m
i
a
H
a
i
ba
t
K
ha
t
on
S
t
r
e
e
t
,
22
M
a
ha
l
l
a
308,
B
a
ghda
d,
I
r
a
q
E
ma
il:
ms
c
.
a
h
me
d
.
h
.
a
li@
g
ma
il.
c
o
m
1.
I
NT
RO
DUC
T
I
O
N
T
he
a
ppl
i
c
a
t
i
on
of
i
n
f
or
m
a
t
i
on
t
e
c
hnol
ogy
i
n
s
e
ve
r
a
l
f
i
e
l
ds
,
s
uc
h
a
s
c
om
put
i
ng,
ne
t
w
o
r
ki
ng,
a
nd
s
t
o
r
ag
e cap
aci
t
y
,
h
as
s
een
s
i
g
n
i
f
i
can
t
ad
v
an
cem
en
t
s
i
n
t
h
e
p
r
e
vi
ous
de
c
a
de
[
1]
–
[
6]
.
T
he
r
e
s
ul
t
of
t
hi
s
ad
v
an
cem
en
t
i
s
t
h
e em
er
g
en
ce o
f
a n
ew
s
ci
en
t
i
f
i
c p
ar
ad
i
g
m
:
t
h
e er
a o
f
h
u
g
e d
at
a co
l
l
ect
i
o
n
an
d
ex
p
l
o
r
at
i
o
n
,
w
hi
c
h ha
s
e
vol
ve
d i
nt
o a
s
c
i
e
nt
i
f
i
c
di
s
c
ov
e
r
y
a
ppr
oa
c
h t
ha
t
i
s
on a
n e
qua
l
f
oot
i
ng
w
i
t
h c
on
ve
nt
i
ona
l
t
he
or
e
t
i
c
a
l
a
na
l
ys
i
s
,
e
xpe
r
i
m
e
nt
a
l
de
s
i
gns
,
a
nd c
o
m
put
e
r
s
i
m
ul
a
t
i
on.
T
he
a
m
ount
of
da
t
a
ge
ne
r
a
t
e
d a
nd s
t
or
e
d
ha
s
e
xpa
nde
d dr
a
m
a
t
i
c
a
l
l
y ove
r
t
he
l
a
s
t
t
w
o de
c
a
de
s
a
s
a
r
e
s
ul
t
of
t
he
de
ve
l
opm
e
nt
of
t
h
e
i
nt
e
r
ne
t
of
t
hi
ngs
(
I
oT
)
,
a
r
t
i
f
i
c
i
a
l
i
nt
e
l
l
i
ge
nc
e
,
c
l
oud c
om
put
i
ng,
a
nd ot
he
r
c
ut
t
i
ng
-
e
dge
c
om
put
e
r
t
e
c
hnol
ogi
e
s
[
7]
–
[
19]
.
B
e
c
a
us
e
m
or
e
t
ha
n 6000 t
w
e
e
t
s
a
r
e
s
e
nt
out
pe
r
s
e
c
ond on T
w
i
t
t
e
r
,
a
nd t
he
s
i
m
i
l
a
r
t
r
e
nd c
a
n be
f
ound on
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
1693
-
6930
T
E
L
KOM
NI
KA
T
el
eco
m
m
u
n
C
om
put
E
l
C
ont
r
o
l
,
Vo
l
.
19
, N
o
.
5
,
O
c
t
obe
r
2021:
1735
-
1746
1736
F
a
c
e
book,
W
ha
t
s
A
pp,
a
nd ot
h
e
r
s
o
c
ia
l me
d
ia
p
la
tf
o
r
ms
.
S
o
ci
al
m
ed
i
a an
d
t
h
e I
n
t
e
r
n
et
h
av
e m
ad
e s
u
b
s
t
an
t
i
al
c
ont
r
i
but
i
ons
t
o
t
hi
s
r
a
t
e
of
da
t
a
ge
ne
r
a
t
i
on
[
2
0]
–
[
24]
.
L
a
r
ge
a
m
ount
s
of
da
t
a
a
r
e
ge
ne
r
a
t
e
d by
a
ppl
i
c
a
t
i
on/
w
e
b s
e
r
ve
r
s
f
r
om
t
he
i
r
l
ogs
a
t
t
he
o
r
ga
ni
z
a
t
i
ona
l
l
e
ve
l
.
O
t
he
r
s
ys
t
e
m
s
a
l
s
o c
ont
r
i
but
e
t
o t
he
i
n
cr
eas
ed
r
at
e o
f
d
at
a g
en
er
at
i
o
n
.
A
s
a
r
es
u
l
t
,
d
at
a h
as
ev
o
l
v
ed
i
nt
o a
n e
s
s
e
nt
i
a
l
c
om
pone
nt
of
bus
i
ne
s
s
ex
i
s
t
en
ce.
B
ecau
s
e o
f
t
h
e i
n
cr
eas
ed
l
ev
el
o
f
d
i
g
i
t
al
d
at
a g
en
er
at
i
o
n
,
co
m
b
i
n
ed
w
i
t
h
t
h
e i
n
cr
eas
i
n
g
co
m
p
l
ex
i
t
y
of
t
he
da
t
a
.
I
t
ha
s
be
c
om
e
i
m
pos
s
i
bl
e
t
o pr
oc
e
s
s
t
he
m
us
i
ng c
onve
nt
i
ona
l
da
t
a
p
r
oc
e
s
s
i
ng m
e
t
hods
;
a
s
a
r
e
s
ul
t
,
e
f
f
or
t
s
a
r
e
now
be
i
ng di
r
e
c
t
e
d t
ow
a
r
d de
ve
l
opi
ng a
dva
nc
e
d c
om
put
i
ng i
nf
r
a
s
t
r
uc
t
ur
e
s
t
ha
t
c
a
n ha
ndl
e
da
t
a
vol
um
e
s
a
nd c
om
pl
e
xi
t
y of
t
hi
s
m
a
gni
t
ude
(
r
e
f
e
r
r
e
d t
o
a
s
pa
r
a
l
l
e
l
a
nd
di
s
t
r
i
but
e
d p
r
oc
e
s
s
i
ng)
[
2
5]
–
[
42]
.
M
a
na
gi
ng va
s
t
a
m
ount
s
of
da
t
a
i
s
a
di
f
f
i
c
ul
t
t
a
s
k t
ha
t
de
m
a
nds
t
he
de
ve
l
opm
e
nt
o
f
m
or
e
c
om
pl
e
x
s
y
s
t
em
s
i
n
o
r
d
er
t
o
ach
i
ev
e accu
r
at
e an
d
t
i
m
el
y
en
o
r
m
o
u
s
d
at
a an
al
y
s
i
s
[
4
3]
–
[
45]
.
I
n
or
de
r
t
o
pr
oc
e
s
s
bi
g
d
at
a an
al
y
t
i
cs
p
r
o
b
l
em
s
i
n
a t
i
m
el
y
an
d
r
el
i
ab
l
e m
an
n
er
,
i
n
f
r
as
t
r
uc
t
ur
e
f
or
bi
g da
t
a
ha
s
be
e
n de
ve
l
ope
d,
a
l
l
ow
i
ng f
or
hi
gh
-
q
u
al
i
t
y
p
er
f
o
r
m
an
ce an
d
r
es
o
u
r
ce av
ai
l
ab
i
l
i
t
y
f
o
r
s
el
f
-
s
e
r
vi
c
e
a
nd c
onve
ni
e
nc
e
o
f
us
e
on
d
em
an
d
.
T
h
er
e ar
e
n
u
m
er
o
u
s
m
ach
i
n
e l
ear
n
i
n
g
f
r
a
m
ew
o
r
k
s
f
o
r
l
ar
g
e d
at
a an
al
y
s
i
s
p
r
es
en
t
l
y
acces
s
i
b
l
e;
t
h
ey
a
r
e
r
e
l
e
va
nt
t
o di
f
f
e
r
e
nt
s
c
i
e
nt
i
f
i
c
dom
a
i
ns
a
nd
ha
ve
be
e
n s
how
n t
o be
us
e
f
ul
i
n he
a
l
t
hc
a
r
e
i
n
f
or
m
a
t
i
c
s
ge
ne
t
i
c
da
t
a
a
na
l
ys
i
s
,
t
e
xt
e
xpl
or
a
t
i
on
,
a
nd
r
a
nd
om
pi
c
t
ur
e
m
ode
l
i
ng
,
a
m
ong
ot
he
r
a
ppl
i
c
a
t
i
ons
.
A
pa
c
he
S
pa
r
k
ma
c
h
in
e
le
a
r
n
in
g
lib
r
a
r
y
(
M
L
lib
)
i
s
a
n ope
n
-
s
o
u
r
ce,
i
n
-
de
m
a
nd,
a
nd i
nde
pe
nde
nt
l
i
br
a
r
y f
or
bi
g da
t
a
a
na
l
ys
i
s
us
i
ng m
a
c
hi
ne
l
e
a
r
ni
ng t
e
c
hni
que
s
[
46]
.
I
t
ha
s
t
he
a
dva
nt
a
ge
of
ha
vi
ng a
n a
ut
om
a
t
i
c
da
t
a
b
a
l
a
nc
i
ng
a
nd a
di
s
t
r
i
but
e
d de
s
i
gn,
m
a
ki
ng i
t
a
good
c
hoi
c
e
f
or
bi
g
da
t
a
a
na
l
ys
i
s
.
A
c
ol
l
e
c
t
i
on
of
dom
i
na
t
i
ng p
e
opl
e
i
n
oc
c
upa
t
i
ons
f
or
num
e
r
ous
m
ach
i
n
e
l
ear
n
i
n
g
t
as
k
s
,
s
u
ch
as
cl
as
s
i
f
i
cat
i
o
n
,
r
eg
r
es
s
i
o
n
,
b
as
e co
m
p
i
l
at
i
o
n
an
d
e
xt
r
a
c
t
i
on (
a
nd di
m
e
ns
i
ona
l
r
e
duc
t
i
on)
,
i
s
i
nt
r
o
duc
e
d by A
pa
c
he
S
pa
r
k
[
47]
–
[
51]
.
D
e
s
p
ite
th
e
f
a
c
t th
a
t
num
e
r
ous
r
e
s
e
a
r
c
h ha
ve
be
e
n c
onduc
t
e
d
on
m
a
c
hi
ne
l
e
a
r
ni
ng
a
nd
i
t
s
us
e
f
ul
ne
s
s
,
M
L
l
i
br
a
r
i
e
s
f
or
bi
g
da
t
a
a
na
l
ys
i
s
,
s
uc
h a
s
A
pa
c
he
S
pa
r
k M
L
l
i
b,
ha
ve
r
e
c
e
i
ve
d l
i
t
t
l
e
a
t
t
e
nt
i
on
.
P
e
r
ha
ps
t
hi
s
i
s
t
he
f
i
r
s
t
s
t
udy t
o
l
ook a
t
l
i
b
r
ar
i
es
f
o
r
l
ar
g
e d
at
a an
al
y
s
i
s
t
h
at
ar
e b
as
ed
o
n
m
ach
i
n
e l
ear
n
i
n
g
t
ech
n
i
q
u
es
.
B
i
g
d
at
a an
al
y
t
i
cs
i
s
p
r
i
m
ar
i
l
y
c
onc
e
r
ne
d w
i
t
h t
he
a
dva
nc
e
m
e
nt
of
c
om
put
e
r
i
nf
r
a
s
t
r
uc
t
ur
e
s
i
n s
uc
h a
w
a
y t
ha
t
da
t
a
m
i
ni
ng a
nd a
na
l
ys
i
s
c
a
n
be
c
om
pl
e
t
e
d qui
c
kl
y a
nd e
f
f
ic
ie
n
tly
[
45]
,
[
52
]
–
[
56]
.
I
t
i
s
t
he
pr
i
m
a
r
y
dr
i
vi
ng
f
or
c
e
be
hi
nd t
he
e
xi
s
t
i
ng
bus
i
ne
s
s
.
B
e
c
a
us
e
l
a
r
ge
da
t
a
a
na
l
yt
i
c
s
i
s
a
c
om
p
ut
a
t
i
ona
l
l
y i
nt
e
ns
i
ve
ope
r
a
t
i
on,
t
he
us
e
r
e
xpe
r
i
e
nc
e
dur
i
ng
l
a
r
ge
da
t
a
a
na
l
yt
i
c
s
i
s
i
nf
l
ue
nc
e
d by
t
he
s
e
t
up of
di
f
f
e
r
e
nt
s
of
t
w
a
r
e
a
nd de
vi
c
e
s
[
57]
–
[
66]
.
S
ev
er
al
b
i
g
d
at
a
p
r
o
ces
s
i
n
g
t
ech
n
i
q
u
es
h
av
e b
een
r
eco
m
m
en
d
ed
s
i
n
ce t
h
e l
as
t
d
ecad
e d
u
e t
h
e
f
ai
l
u
r
e
of
t
he
c
onve
nt
i
ona
l
p
r
oc
e
s
s
i
ng m
e
t
hods
t
o
ha
ndl
e
t
he
l
a
r
ge
vo
l
um
e
o
f
da
t
a
ge
ne
r
a
t
e
d da
i
l
y
f
r
om
bus
i
ne
s
s
a
nd
i
ndus
t
r
i
a
l
pr
oc
e
s
s
e
s
[
67]
.
A
s a
r
e
su
l
t
,
r
e
s
e
a
r
c
he
r
s
ha
ve
be
e
n c
onc
e
nt
r
a
t
i
ng t
he
i
r
e
f
f
or
t
s
on de
ve
l
opi
ng m
or
e
e
f
f
e
c
t
i
ve
m
e
t
hods
of
obt
a
i
ni
ng
va
l
ue
-
a
dde
d i
nf
or
m
a
t
i
on f
r
om
l
a
r
ge
a
m
ount
s
of
da
t
a
.
T
he
r
e
a
r
e
m
a
ny
di
f
f
e
r
e
nt
t
ype
s
of
s
t
udi
e
s
i
n
t
he
a
r
e
a
of
bi
g
da
t
a
p
r
oc
e
s
s
i
ng m
ode
l
s
.
F
o
r
ex
am
p
l
e,
d
at
a
f
l
ow
m
ode
l
s
s
uc
h a
s
M
a
pR
e
duc
e
,
w
hi
c
h
f
a
c
i
l
i
t
a
t
e
da
t
a
p
r
oc
e
s
s
i
ng ut
i
l
i
z
i
ng a
va
r
i
e
t
y
of
ope
r
a
t
or
s
w
hi
l
e
s
ha
r
i
ng s
t
a
bl
e
s
t
or
a
ge
s
ys
t
e
m
s
,
a
r
e
one
t
ype
of
s
t
udy
[
68]
–
[
7
4]
.
R
e
s
ilie
n
t d
is
tr
ib
u
te
d
d
a
ta
s
e
ts
(
R
DDs
)
ar
e a m
o
r
e e
f
f
i
c
i
en
t
d
at
a
s
ha
r
i
ng a
bs
t
r
a
c
t
i
on f
r
o
m
s
t
a
bl
e
s
t
or
a
ge
s
i
nc
e
t
h
e
y do not
r
e
qui
r
e
da
t
a
c
opyi
ng
,
w
hi
c
h
s
a
ve
s
m
one
y.
I
n
m
os
t
hi
gh
-
l
ev
el
a
ppl
i
c
a
t
i
on pr
ogr
a
m
m
i
ng
i
nt
e
r
f
a
c
e
s
(
A
P
I
s
)
f
or
da
t
a
f
l
ow
s
ys
t
e
m
s
,
i
nt
e
gr
a
t
e
d
l
a
ngua
ge
A
P
I
s
[
75]
–
[
81]
a
r
e
pr
ovi
de
d,
w
hi
c
h a
l
l
ow
t
he
us
e
r
t
o i
nt
e
r
a
c
t
w
i
t
h "
pa
r
a
l
l
e
l
gr
oups
"
t
hr
ough
ope
r
a
t
or
s
s
uc
h
a
s
m
a
p a
nd j
oi
n.
P
a
r
a
l
l
e
l
gr
oups
on
t
he
s
e
s
ys
t
e
m
s
,
on t
he
o
t
he
r
ha
nd,
e
i
t
he
r
r
e
pr
e
s
e
nt
f
i
l
e
s
on
di
s
k or
t
he
t
e
m
por
a
r
y da
t
a
s
e
t
s
t
ha
t
w
e
r
e
us
e
d f
or
que
r
y pl
a
n
e
xpr
e
s
s
i
on on t
he
s
e
s
y
s
t
e
m
s
.
D
e
s
pi
t
e
t
he
f
a
c
t
t
ha
t
s
ys
t
e
m
s
ha
ve
t
he
a
bi
l
i
t
y
t
o
c
onv
e
y da
t
a
vi
a
t
he
ope
r
a
t
or
s
i
n
t
he
s
a
m
e
que
r
y
,
da
t
a
e
xc
ha
nge
t
hr
ough
i
nqui
r
i
e
s
ha
s
pr
ove
n t
o
be
i
ne
f
f
i
c
i
e
nt
.
A
s
a
r
e
s
ul
t
,
S
pa
r
k's
A
P
I
i
s
bui
l
t
on
t
he
pa
r
a
l
l
e
l
s
um
m
a
t
i
on
m
ode
l
,
w
hi
c
h
i
s
c
onve
ni
e
nt
t
o i
m
pl
e
m
e
nt
.
I
t d
o
e
s
n
o
t
c
la
im to
b
e
th
e
f
ir
s
t to
u
s
e
a
n
in
te
gr
a
t
e
d i
nt
e
r
f
a
c
e
l
a
ngua
ge
,
but
by
i
nc
l
udi
ng R
D
D
s
a
s
a
s
t
or
a
ge
l
a
ye
r
be
hi
nd t
h
i
s
i
nt
e
r
f
a
c
e
,
i
t
c
a
n
s
uppor
t
a
l
a
r
ge
r
r
a
nge
of
a
ppl
i
c
a
t
i
ons
.
T
he
s
ys
t
e
m
s
i
n t
he
f
ol
l
ow
i
ng c
a
t
e
gor
y a
r
e
t
hos
e
t
ha
t
pr
ovi
de
hi
gh
-
l
ev
el
i
n
t
er
f
aces
f
o
r
s
p
eci
al
i
zed
a
p
p
lic
a
tio
n
s
th
a
t r
eq
u
i
r
e d
at
a s
h
ar
i
n
g
,
as
d
es
cr
i
b
ed
ab
o
v
e.
P
r
eg
el
[
82]
pr
ovi
de
s
s
uppor
t
f
or
r
e
dunda
nc
y
d
ia
g
r
a
mmin
g
a
p
p
lic
a
tio
n
s
,
w
h
ils
t
T
w
is
te
r
[
8
3]
a
nd
H
a
L
oop
[
84]
ar
e
i
t
e
r
at
i
v
e M
ap
R
ed
u
ce p
r
o
g
r
am
s
,
r
es
p
ect
i
v
el
y
.
T
h
es
e f
r
am
ew
o
r
k
s
o
n
l
y
pr
ovi
de
da
t
a
s
ha
r
i
ng f
o
r
t
he
c
a
l
c
ul
a
t
i
on s
t
yl
e
s
t
ha
t
a
r
e
s
uppor
t
e
d,
a
nd
t
he
y do not
p
r
ovi
de
a
n uni
ve
r
s
a
l
a
bs
t
r
a
c
t
i
ng
f
r
a
m
e
w
or
k.
T
he
y c
a
n onl
y be
us
e
d by
t
he
us
e
r
t
o
s
ha
r
e
s
e
l
e
c
t
e
d
da
t
a
f
r
om
s
pe
c
i
f
i
e
d ope
r
a
t
i
ons
.
F
o
r
ex
am
p
l
e,
a u
s
er
can
n
o
t
l
o
ad
d
at
a i
n
t
o
m
e
m
or
y
us
i
ng P
r
e
ge
l
or
T
w
i
s
t
e
r
a
nd t
he
n s
e
l
e
c
t
t
he
que
r
y
t
o
r
un on
i
t
a
f
t
e
r
i
t
ha
s
be
e
n l
oa
de
d.
T
he
f
a
c
t
t
ha
t
R
D
D
s
e
xpr
e
s
s
l
y
pr
ovi
de
di
s
t
r
i
but
e
d s
t
or
a
ge
m
e
a
ns
t
ha
t
i
t
c
a
n
be
us
e
d t
o
e
n
a
bl
e
a
ppl
i
c
a
t
i
ons
t
ha
t
a
r
e
not
c
ur
r
e
nt
l
y s
uppor
t
e
d
by t
he
s
e
s
p
eci
al
i
zed
s
ys
t
e
m
s
,
s
uc
h a
s
i
nt
e
r
a
c
t
i
ve
da
t
a
m
i
ni
ng.
A
c
c
or
di
ng
t
o
[
85]
,
it
is
p
r
opos
e
d
a
m
e
t
hodol
ogy t
ha
t
de
m
ons
t
r
a
t
e
s
t
o
be
a
n upgr
a
de
ove
r
s
t
a
nda
r
d bi
g da
t
a
a
na
l
yt
i
c
s
m
e
t
hods
t
ha
t
us
e
e
i
t
he
r
H
a
doop/
S
pa
r
k or
de
e
p
l
e
a
r
ni
ng a
s
di
s
t
i
nc
t
c
om
pone
nt
s
.
L
unga
e
t
a
l.
[
85]
p
r
opos
e
d
a
f
r
a
m
e
w
or
k t
ha
t
m
a
ke
s
us
e
of
S
pa
r
k's
d
is
tr
ib
u
te
d
c
o
mp
u
tin
g
c
a
p
a
b
ilitie
s
a
s
w
e
ll
a
s
d
e
e
p
le
a
r
n
in
g
a
r
c
h
ite
c
tu
r
e
f
o
r
mu
ltip
le
la
y
e
r
s
p
e
r
c
e
p
tr
o
n
(
M
L
P
)
u
s
i
n
g
cas
cad
e l
ear
n
i
n
g
t
o
t
r
ai
n
m
u
l
t
i
p
l
e l
ay
er
s
p
er
cep
t
r
o
n
s
i
s
pr
o
p
o
se
d
.
A
f
r
a
m
e
w
or
k
f
or
i
n
-
de
pt
h t
r
a
i
ni
ng
l
ear
n
i
n
g
m
o
d
el
s
w
i
t
h
A
p
ach
e S
p
ar
k
h
as
b
een
cr
eat
ed
an
d
d
ev
el
o
p
ed
in
[4
7
],
[4
8
],
[5
0
],
[5
1
],
[5
7
],
[6
9
],
[8
6
]
–
[
92]
.
T
hi
s
f
r
a
m
e
w
or
k s
hor
t
e
ns
t
he
t
r
a
i
ni
ng
t
i
m
e
by
t
a
ki
ng a
dva
nt
a
ge
of
t
he
a
dva
nt
a
ge
s
of
bot
h da
t
a
a
nd pa
r
i
t
y
m
ode
l
i
ng a
t
t
he
s
a
m
e
time
.
I
t
is
p
o
s
s
ib
le
to
c
r
e
a
te
d
a
ta
p
a
r
a
lle
lis
m
b
y
d
is
tr
ib
u
tin
g
tr
a
in
in
g
d
a
ta
a
c
r
o
s
s
ma
n
y
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
el
eco
m
m
u
n
C
om
put
E
l
C
ont
r
o
l
L
ar
ge
s
c
al
e
dat
a anal
y
s
i
s
us
i
ng M
L
l
i
b
(
A
hm
e
d H
u
s
s
e
i
n A
l
i
)
1737
S
pa
r
k bl
oc
k m
a
c
hi
ne
s
a
nd r
e
pl
i
c
a
t
i
ng m
ode
l
s
on
e
a
c
h de
vi
c
e
[
86]
.
E
a
c
h m
ode
l
goe
s
t
hr
ough
i
t
s
t
r
a
i
ni
ng i
n
p
a
r
a
lle
l w
ith
th
e
d
a
ta
p
a
r
t
.
T
h
e
p
a
r
a
lle
lis
m mo
d
e
l is
imp
le
me
n
te
d
b
y
d
is
tr
ib
u
tin
g
e
a
c
h
r
e
p
lic
a
o
f
t
h
e
d
e
e
p
ne
ur
a
l
ne
t
w
or
k m
ode
l
ove
r
t
he
s
pa
r
k
gr
oup
i
n a
l
a
y
e
r
-
by
-
l
a
ye
r
f
a
s
hi
on.
T
he
i
m
pa
c
t
of
va
r
i
ous
s
of
t
w
a
r
e
a
nd
ha
r
dw
a
r
e
c
onf
i
gur
a
t
i
ons
on
t
he
p
r
obl
e
m
o
f
bi
g
da
t
a
p
r
oc
e
s
s
i
ng
is
e
xpl
or
e
d i
n t
hi
s
r
e
s
e
a
r
c
h.
T
he
f
oc
us
of
t
he
pr
e
s
e
nt
a
t
i
on i
s
on t
he
c
a
pa
bi
l
i
t
i
e
s
a
nd
a
dva
nt
a
ge
s
of
A
pa
c
he
S
pa
r
k M
L
l
i
b 2.
0
a
s
a
l
a
r
ge
da
t
a
a
na
l
yt
i
c
s
t
ool
,
pa
r
t
i
c
ul
a
r
l
y i
n r
e
l
a
t
i
on t
o
H
a
doop.
T
h
i
s
s
t
udy
is
de
ve
l
ope
d a
s
a
m
e
a
ns
of
pr
ovi
di
ng
i
ns
i
ght
i
nt
o
t
he
us
a
ge
o
f
m
a
c
hi
ne
l
e
a
r
ni
ng
l
i
br
a
r
i
e
s
i
n
bi
g
da
t
a
a
na
l
ys
i
s
f
r
om
t
he
s
t
a
ndpoi
nt
of
i
ndus
t
r
y
.
T
hi
s
w
or
k
ope
n
s
t
he
door
t
o
ot
he
r
e
le
me
n
ts
o
f
b
ig
d
a
ta
a
n
a
ly
s
is
u
tiliz
in
g
ma
c
h
in
e
l
e
a
r
ni
ng m
e
t
hods
,
w
hi
c
h i
s
r
e
ga
r
de
d t
o be
a
r
a
pi
dl
y e
xpa
ndi
ng s
t
udy t
opi
c
.
S
e
ve
r
a
l
r
e
a
l
-
w
o
r
ld
te
s
ts
is
car
r
i
ed
out
t
o
i
nve
s
t
i
ga
t
e
t
he
qua
l
i
t
a
t
i
ve
a
nd
qua
nt
i
t
a
t
i
ve
a
s
pe
c
t
s
of
A
pa
c
he
S
pa
r
k
M
L
l
i
b 2
.
0
.
M
or
e
ove
r
,
a
co
m
p
ar
at
i
v
e r
es
ear
ch
is
c
a
r
r
i
e
d ou
t
us
i
ng t
he
ma
s
s
iv
e
o
n
lin
e
a
n
a
ly
tic
s
(
MO
A
)
l
i
b
r
ar
y
,
w
h
i
ch
i
s
a w
el
l
-
know
n
Ja
v
a
-
b
a
s
e
d m
a
c
hi
ne
l
e
a
r
ni
ng l
i
b
r
a
r
y
t
ha
t
i
s
w
i
de
l
y us
e
d i
n
t
he
i
ndus
t
r
y
.
F
ur
t
he
r
m
or
e
,
t
he
pe
r
f
or
m
a
nc
e
of
s
e
ve
r
a
l
c
om
m
onl
y us
e
d m
a
c
hi
ne
l
e
a
r
ni
ng m
ode
l
s
f
or
b
i
g da
t
a
a
na
l
ys
i
s
is
ex
am
i
n
ed
,
an
d
co
m
p
ar
ed
acr
o
s
s
a
va
r
i
e
t
y of
s
of
t
w
a
r
e
a
nd ha
r
dw
a
r
e
s
e
t
t
i
ngs
.
T
he
r
e
m
a
i
ni
ng pa
r
t
of
t
hi
s
a
r
t
i
c
l
e
i
s
a
r
r
a
nge
d t
hus
:
s
e
c
t
i
on
2
i
nt
r
oduc
e
s
A
pa
c
he
S
pa
r
k M
L
l
i
b.
T
he
m
e
t
hod a
nd
c
om
pone
nt
s
of
t
he
i
nve
s
t
i
ga
t
e
d A
pa
c
he
S
pa
r
k M
L
l
i
b 2.
0
ar
e
pr
e
s
e
nt
e
d i
n
s
ect
i
o
n
3,
w
h
ile
th
e
r
e
s
ul
t
s
a
nd
di
s
c
us
s
i
on of
t
he
f
e
a
t
ur
e
s
a
nd be
nc
hm
a
r
ki
ng
ar
e
p
r
e
s
en
t
ed
i
n
s
e
c
t
i
on 4,
a
nd c
onc
l
us
i
on
is
p
r
e
s
e
nt
e
d i
n s
e
c
t
i
on 5.
2.
AP
ACH
E
S
P
ARK
M
L
L
I
B
2
.
0
T
h
i
s
i
s
a s
cal
ab
l
e an
d
f
as
t
b
i
g
d
at
a p
r
o
ces
s
i
n
g
en
g
i
n
e
t
ha
t
w
a
s
f
i
r
s
t
de
ve
l
ope
d by t
he
A
M
P
L
a
b a
t
t
he
U
ni
ve
r
s
i
t
y of
C
a
l
i
f
or
n
i
a
,
B
e
r
ke
l
e
y
[
93]
–
[
9
5]
.
I
t
m
a
y be
us
e
d
t
o
c
ons
t
r
uc
t
di
s
t
r
i
but
e
d a
ppl
i
c
a
t
i
ons
i
n
a
v
ar
i
et
y
of
c
om
put
e
r
l
a
ngua
ge
s
,
i
nc
l
udi
ng J
a
va
,
P
yt
hon,
a
nd ot
he
r
pr
og
r
a
m
m
i
ng
l
a
ngua
ge
s
[
96]
–
[
105
]
.
W
h
e
n
it is
in
s
ta
lle
d
,
it in
c
lu
d
e
s
f
o
u
r
ma
jo
r
lib
r
a
r
ie
s
: A
p
a
c
h
e
S
p
a
r
k
s
t
r
uc
t
ur
e
d que
r
y l
a
ngua
ge
(
S
QL
)
,
A
pa
c
he
S
pa
r
k
S
t
r
eam
i
n
g
,
A
p
ach
e S
p
ar
k
M
L
l
i
b
,
an
d
A
p
ach
e S
p
ar
k
G
r
ap
h
X
.
T
h
es
e l
i
b
r
ar
i
es
ar
e d
es
cr
i
b
ed
i
n
m
o
r
e d
et
ai
l
be
l
ow
[
106]
–
[
108]
.
H
ow
e
ve
r
,
de
s
pi
t
e
t
he
f
a
c
t
t
h
a
t
t
he
m
os
t
ba
s
i
c
s
c
he
dul
i
ng S
pa
r
k
m
odul
e
s
a
r
e
A
pa
c
he
S
pa
r
k S
t
r
e
a
m
i
ng
,
w
hi
c
h i
s
f
a
ul
t
t
ol
e
r
a
nt
a
nd pe
r
f
or
m
s
hi
gh l
e
ve
l
a
na
l
yt
i
c
s
,
A
pa
ch
e S
p
ar
k
S
Q
L
p
er
f
o
r
m
s
r
el
at
i
o
n
al
q
u
er
i
es
f
o
r
a
v
ar
i
et
y
o
f
m
i
n
i
n
g
d
at
ab
as
es
b
ecau
s
e i
t
i
n
co
r
p
o
r
at
es
a d
at
a
ab
s
t
r
act
i
o
n
m
o
d
el
k
n
o
w
n
as
d
at
a f
r
am
es
[
109]
.
I
t
i
s
i
m
po
r
t
a
nt
t
o
not
e
t
ha
t
A
pa
c
he
S
pa
r
k G
r
a
phX
[
110
]
i
s
a
hi
gh
-
l
ev
el
A
p
ach
e S
p
ar
k
pr
oc
e
s
s
i
ng l
i
br
a
r
y t
ha
t
c
a
n ha
ndl
e
t
w
o
c
om
m
onl
y
us
e
d da
t
a
s
t
r
uc
t
ur
e
s
ut
i
l
i
z
i
ng
di
s
t
r
i
but
e
d a
r
i
t
hm
e
t
i
c
m
ode
l
s
.
A
pa
c
he
S
pa
r
k M
L
l
i
b pr
ovi
de
s
>
55 s
c
a
l
a
bl
e
m
a
c
hi
ne
l
e
a
r
ni
ng a
l
gor
i
t
hm
s
f
o
r
bi
g da
t
a
a
na
l
yt
i
c
s
,
t
a
ki
ng
ad
v
a
nt
a
ge
of
t
he
a
dva
nt
a
ge
s
of
bot
h da
t
a
a
nd
t
he
da
t
a
c
ol
l
e
c
t
i
on m
e
t
hod.
A
s
w
e
l
l
a
s
e
na
b
l
i
ng t
he
i
m
pl
e
m
e
nt
a
t
i
on of
num
e
r
ous
m
a
c
hi
ne
l
e
a
r
ni
ng s
t
r
a
t
e
gi
e
s
,
s
uc
h a
s
gr
oupi
ng a
nd r
e
gr
e
s
s
i
on;
c
l
a
s
s
i
f
i
c
a
t
i
on;
r
ul
e
e
xt
r
a
c
t
i
on;
a
nd di
m
e
ns
i
ona
l
r
e
duc
t
i
on
.
I
t
al
s
o
en
ab
l
e
s
t
he
r
a
pi
d a
nd s
i
m
pl
e
c
r
e
a
t
i
on of
m
a
c
hi
ne
l
e
a
r
ni
ng
a
ppr
oa
c
he
s
f
or
l
a
r
ge
-
s
cal
e ap
p
l
i
cat
i
o
n
s
[
67]
,
[
111]
–
[
116]
.
A
s
u
ite
o
f
mu
ltip
le
-
l
a
ngua
ge
A
P
I
s
i
s
a
l
s
o a
va
i
l
a
bl
e
on
t
he
A
pa
c
he
S
pa
r
k
M
L
l
i
b
[
117]
p
l
a
t
fo
rm
fo
r
t
he
e
va
l
ua
t
i
on a
nd de
pl
oym
e
nt
o
f
a
w
i
de
r
a
nge
of
m
a
c
hi
ne
l
e
a
r
ni
ng
t
e
c
hni
que
s
.
I
n
r
e
c
e
nt
ye
a
r
s
,
s
e
ve
r
a
l
ch
an
g
es
h
av
e b
een
m
ad
e
t
o
m
u
l
t
i
p
l
e
ar
eas
o
f
d
at
a
s
ci
en
ce s
o
l
u
t
i
o
n
s
[
118]
,
[
119]
,
an
d
a
n
u
m
b
er
o
f
ac
ad
em
i
cs
ha
ve
c
om
m
i
t
t
e
d
a
t
t
e
nt
i
on
t
o
t
he
c
r
e
a
t
i
on
of
t
he
c
om
pone
nt
s
of
A
pa
c
he
S
pa
r
k
M
L
l
i
b
f
or
bi
g
da
t
a
a
na
l
yt
i
c
s
.
F
i
gur
e
1
de
pi
c
t
s
t
he
de
v
e
l
opm
e
nt
s
i
de
of
A
pa
c
he
S
pa
r
k
M
L
l
i
b
t
r
a
c
k
2.
0
,
w
i
t
h
a
un
i
que
num
be
r
of
a
nc
hor
s
as
s
i
g
n
ed
t
o
each
r
el
eas
e o
f
t
h
e l
i
b
r
a
r
y
[
120
]
–
[
122
]
.
T
hi
s
s
e
c
t
i
on di
s
c
us
s
e
s
s
om
e
of
t
he
r
e
c
e
nt
i
m
pr
o
ve
m
e
nt
s
i
n A
pa
c
he
S
pa
r
k
M
L
l
i
b a
ppl
i
c
a
t
i
ons
,
i
nc
l
udi
ng
s
om
e
of
t
he
ne
w
f
e
a
t
ur
e
s
i
nt
r
oduc
e
d.
I
n o
r
de
r
t
o a
i
d i
n
t
he
de
ve
l
opm
e
nt
of
s
m
a
r
t
t
r
a
ns
por
t
a
t
i
on a
p
p
l
i
cat
i
o
n
s
,
a s
cal
ab
l
e an
d
o
p
en
-
s
our
c
e
pl
a
t
f
or
m
know
n a
s
co
n
n
ect
ed
ve
hi
c
l
e
s
a
nd s
m
a
r
t
t
r
a
ns
por
t
a
t
i
on
(
C
V
ST
)
ha
s
b
e
e
n pr
opos
e
d by a
num
be
r
of
r
e
s
e
a
r
c
he
r
s
.
T
he
pr
opos
e
d
C
V
S
T
i
s
bui
l
t
of
f
our
e
s
s
e
nt
i
a
l
c
om
pone
nt
s
:
da
t
a
di
s
t
r
i
but
i
on,
r
e
s
our
c
e
m
a
na
ge
m
e
nt
,
bus
i
n
es
s
i
n
t
el
l
i
g
en
ce,
a
nd a
ppl
i
c
a
t
i
on.
T
he
bus
i
ne
s
s
i
nt
e
l
l
i
ge
nc
e
c
om
pone
nt
i
s
i
n c
ha
r
ge
of
da
t
a
a
na
l
yt
i
c
s
,
a
nd i
t
m
a
k
e
s
us
e
of
M
L
l
i
b t
o pr
oc
e
s
s
a
nd t
r
a
ns
m
i
t
da
t
a
t
o
t
he
f
r
on
t
e
nd.
A
c
c
or
di
ng t
o
t
he
f
i
ndi
ngs
of
t
he
s
t
udy
[
107
]
,
[
123]
,
[
124]
,
an
ar
ch
i
t
ect
u
r
al
d
es
i
g
n
f
o
r
acad
em
i
c
i
nf
o
r
m
a
t
i
on s
ys
t
e
m
s
e
r
vi
c
e
s
f
or
s
t
ude
nt
s
e
nr
ol
l
m
e
nt
pa
t
t
e
r
n
a
na
l
ys
i
s
s
houl
d be
c
ons
i
de
r
e
d.
T
hi
s
s
ys
t
e
m
m
a
ke
s
us
e
of
M
L
l
i
b t
o
a
nt
i
c
i
pa
t
e
t
he
s
ugge
s
t
e
d c
our
s
e
s
f
or
t
he
f
or
t
hc
om
i
ng s
e
m
e
s
t
e
r
,
w
hi
c
h
i
s
a
pow
e
r
f
ul
p
r
e
di
c
t
i
on t
ool
.
S
pa
r
kt
e
xt
i
s
a
t
e
xt
m
i
ni
ng
f
r
a
m
e
w
or
k
de
ve
l
ope
d by Y
e
e
t a
l
.
[
125]
f
o
r
u
s
e w
i
t
h
A
p
ach
e S
p
ar
k
l
e
a
r
ni
ng a
nd f
l
ow
a
l
gor
i
t
hm
s
i
n c
onj
unc
t
i
on w
i
t
h t
he
C
a
s
s
a
ndr
a
N
o
S
Q
L
d
at
ab
as
e
[
90]
,
[
126]
–
[
1
29]
.
T
h
e
da
t
a
ba
s
e
w
a
s
bui
l
t
us
i
ng a
bi
g c
ol
l
e
c
t
i
on of
m
e
di
c
a
l
publ
i
c
a
t
i
ons
f
or
t
he
pur
pos
e
of
c
a
nc
e
r
t
ype
c
l
a
s
s
i
f
i
c
a
t
i
on.
A
ur
or
a
[
130
]
d
em
o
n
s
t
r
at
ed
h
o
w
t
o
an
al
y
ze w
eb
-
s
our
c
e
d m
obi
l
e
da
t
a
us
i
ng A
pa
c
he
's
K
-
a
l
gor
i
t
hm
S
pa
r
k
M
L
lib
,
w
hi
c
h i
s
ba
s
e
d on t
he
S
pa
r
k a
l
go
r
i
t
hm
.
T
he
s
t
udy ga
ve
a
n e
f
f
e
c
t
i
ve
t
e
c
hni
que
o
f
de
t
e
r
m
i
ni
ng t
he
num
be
r
of
gr
i
d us
e
r
s
ba
s
e
d on t
he
gr
oupi
ng of
l
a
t
i
t
ude
a
nd l
ongi
t
ude
i
nf
or
m
a
t
i
on
,
w
hi
c
h w
a
s
ba
s
e
d on t
he
r
e
s
ul
t
s
of
t
he
i
nve
s
t
i
ga
t
i
on.
W
he
n l
e
a
r
ni
ng hum
a
n be
ha
vi
or
s
,
t
he
s
t
udy by
[
122]
pr
ovi
de
d A
L
M
D
,
w
hi
c
h
pe
r
f
or
m
s
f
e
a
t
ur
e
de
s
c
r
i
pt
i
on
by m
on
i
t
or
i
ng
t
he
a
ppe
a
r
a
nc
e
a
nd m
ove
m
e
nt
r
a
ndom
l
y ba
s
e
d on
t
he
us
a
ge
of
t
he
A
pa
c
he
S
pa
r
k M
L
l
i
br
a
r
y a
nd t
he
us
a
ge
of
A
pa
c
he
S
pa
r
k M
L
l
i
br
a
r
y.
A
sse
f
i
et
a
l
.
[
131]
ha
ve
d
es
cr
i
b
ed
t
he
c
ons
t
r
uc
t
i
on of
a
f
r
a
m
e
w
or
k
f
or
de
m
og
r
a
phi
c
s
a
na
l
ys
i
s
ut
i
l
i
z
i
ng ne
xt
-
ge
ne
r
a
t
i
on
d
at
a s
eq
u
en
ci
n
g
as
a
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
1693
-
6930
T
E
L
KOM
NI
KA
T
el
eco
m
m
u
n
C
om
put
E
l
C
ont
r
o
l
,
Vo
l
.
19
, N
o
.
5
,
O
c
t
obe
r
2021:
1735
-
1746
1738
cas
e s
t
u
d
y
.
I
n
o
r
d
er
t
o
o
p
t
i
m
i
ze t
h
e s
y
s
t
em
,
i
t
i
s
n
eces
s
ar
y
t
o
u
p
d
at
e t
h
e r
es
o
u
r
ce es
t
i
m
at
o
r
an
d
o
p
t
i
m
i
ze t
h
e
c
om
pone
nt
s
.
T
he
s
ys
t
e
m
w
a
s
de
ve
l
ope
d e
nt
i
r
e
l
y
on A
pa
c
he
S
pa
r
k,
a
nd a
s
a
r
e
s
ul
t
,
i
t
t
a
ke
s
a
dva
nt
a
ge
of
t
he
f
av
o
r
ab
l
e as
p
ect
s
of
t
he
M
L
l
i
b a
nd ot
he
r
S
pa
r
k
c
om
pone
nt
s
.
B
i
gN
N
w
a
s
de
ve
l
ope
d by
A
sse
f
i
e
t a
l
.
[
131]
a
s
an
o
t
h
er
f
as
ci
n
at
i
n
g
f
eat
u
r
e o
f
b
i
g
d
at
a an
al
y
t
i
cs
o
n
A
p
ach
e S
p
ar
k
.
I
t
i
s
c
a
pa
bl
e
of
ha
ndl
i
ng bi
om
e
di
c
a
l
s
t
r
i
ngs
o
n
a v
er
y
l
ar
g
e
s
cal
e,
w
h
i
ch
i
s
v
er
y
u
s
ef
u
l
i
n
t
he
he
a
l
t
hc
a
r
e
i
ndus
t
r
y.
M
L
l
i
b c
a
n
be
i
m
pl
e
m
e
nt
e
d us
i
ng
pr
ogr
a
m
s
w
r
i
t
t
e
n
i
n
R
,
S
c
a
l
a
,
P
yt
hon
,
a
nd J
a
va
,
a
m
ong ot
he
r
p
r
ogr
a
m
m
i
ng
l
a
ngua
ge
s
.
V
e
c
t
or
,
L
a
be
l
e
dP
oi
nt
,
a
nd r
a
t
i
ng
a
r
e
t
he
c
o
r
e
da
t
a
a
bs
t
r
a
c
t
i
ons
us
e
d b
y M
L
l
i
b;
a
s
a
r
e
s
ul
t
,
t
he
pe
de
s
t
r
i
a
n a
nd
ot
he
r
s
ta
tis
tic
a
l
c
om
pone
nt
s
of
M
L
l
i
b w
or
k on da
t
a
r
e
pr
e
s
e
nt
e
d by t
he
s
e
a
bs
t
r
a
c
t
i
ons
.
O
bs
e
r
va
t
i
ona
l
da
t
a
f
e
a
t
ur
e
s
a
r
e
c
a
pt
ur
e
d us
i
ng t
he
ve
c
t
or
t
ype
,
w
hi
c
h
r
e
pr
e
s
e
nt
s
a
n i
nde
x s
e
t
of
doubl
e
t
ype
va
l
ue
s
w
i
t
h
a
z
e
r
o
-
i
nde
x of
t
he
i
nt
t
ype
.
T
he
ve
c
t
or
t
ype
i
s
us
e
d t
o
r
eco
r
d
t
h
e o
b
s
e
r
v
at
i
o
n
al
d
at
a f
eat
u
r
es
.
A
ve
c
t
or
of
l
e
ngt
h n
m
i
ght
t
he
or
e
t
i
c
a
l
l
y r
e
pr
e
s
e
nt
a
not
e
w
i
t
h n pr
ope
r
t
i
e
s
,
w
hi
c
h w
oul
d i
m
p
l
y t
ha
t
i
t
r
e
pr
e
s
e
nt
s
a
n obj
e
c
t
i
n
a
f
i
l
e
w
i
t
h
N
di
m
e
ns
i
ons
.
T
he
ve
c
t
or
t
ype
of
f
e
r
e
d
by
M
L
l
i
b di
f
f
e
r
s
f
r
om
t
h
e
ve
c
t
or
t
ype
s
u
p
p
lie
d
b
y
th
e
S
c
a
la
s
e
t lib
r
a
r
y
in
th
a
t th
e
v
e
c
to
r
ty
p
e
in
M
L
lib
imp
le
me
n
ts
th
e
d
ig
i
ta
l v
e
c
to
r
c
o
n
c
e
p
t
f
r
om
l
i
ne
a
r
a
l
ge
br
a
,
but
t
he
ve
c
t
or
t
ype
i
n
t
he
S
c
a
l
a
s
e
t
l
i
b
r
a
r
y
doe
s
not
.
M
L
l
i
b
i
s
c
a
pa
bl
e
o
f
ha
nd
l
i
ng bot
h
de
ns
e
a
nd s
pa
r
s
e
ve
c
t
or
t
ype
s
.
I
n
a
ddi
t
i
on,
be
c
a
us
e
t
he
M
L
l
i
b V
e
c
t
or
t
ype
i
s
c
ons
i
de
r
e
d
a
n a
dj
e
c
t
i
ve
,
i
t
c
a
nnot
be
i
ns
t
a
nt
i
a
t
e
d di
r
e
c
t
l
y by t
he
a
ppl
i
c
a
t
i
on;
i
ns
t
e
a
d,
t
he
f
a
c
t
or
y m
e
t
hods
gi
ve
n by M
L
l
i
b
m
us
t
be
u
t
i
l
i
zed
t
o
co
n
s
t
r
u
ct
an
i
n
s
t
an
ce o
f
ei
t
h
er
t
h
e s
p
ar
s
e v
ect
o
r
cl
as
s
o
r
t
h
e d
en
s
e v
ect
o
r
c
l
a
s
s
.
I
t
s
houl
d
be
not
e
d
t
h
at
t
h
e f
act
o
r
y
m
et
h
o
d
s
f
o
r
cr
eat
i
n
g
i
n
s
t
an
ces
o
f
t
h
e d
en
s
e v
ect
o
r
o
r
s
p
ar
s
e v
ect
o
r
cl
as
s
e
s
ar
e
al
r
ead
y
s
pe
c
i
f
i
e
d i
n t
he
ve
c
t
or
s
obj
e
c
t
,
w
hi
c
h i
s
c
onve
ni
e
nt
.
F
i
gur
e
1.
M
L
l
i
b i
n s
pa
r
k e
c
os
ys
t
e
m
3.
M
ETH
O
D
S
p
ar
k
'
s
m
ach
i
n
e l
ear
ni
ng
l
i
br
a
r
y
,
M
L
l
i
b,
ha
s
be
e
n
unde
r
he
a
vy
de
ve
l
opm
e
nt
s
i
nc
e
i
t
s
i
nc
e
pt
i
on,
a
nd
unl
i
ke
t
he
S
pa
r
k c
or
e
. I
t
is
s
till n
o
t in
a
f
u
lly
s
ta
b
le
s
ta
te
w
ith
r
e
g
a
r
d
to
its
o
v
e
r
a
ll A
P
I
a
n
d
d
e
s
ig
n
.
A
s
o
f
S
pa
r
k ve
r
s
i
on 1.
2
.
0,
a
ne
w
,
e
xpe
r
i
m
e
nt
a
l
A
P
I
f
or
M
L
l
i
b ha
s
be
e
n r
el
ea
s
ed
u
n
d
er
t
h
e m
l
p
ack
ag
e (
w
h
er
eas
t
h
e
c
ur
r
e
nt
l
i
br
a
r
y
r
e
s
i
de
s
unde
r
t
he
M
L
l
i
b pa
c
ka
ge
)
.
F
i
gur
e
1 s
how
s
t
he
S
pa
r
k e
c
os
ys
t
e
m
w
i
t
h M
L
l
i
b.
T
hi
s
ne
w
A
P
I
ai
m
s
t
o
en
h
an
ce t
h
e A
P
I
s
an
d
i
n
t
e
r
f
aces
f
o
r
m
o
d
el
s
as
w
el
l
as
f
eat
u
r
e ex
t
r
act
i
o
n
an
d
t
r
an
s
f
o
r
m
at
i
o
n
s
o
a
s
t
o
m
ak
e
i
t
eas
i
er
t
o
b
u
i
l
d
p
i
p
el
i
n
es
t
h
at
ch
ai
n
t
o
g
et
h
er
s
t
ep
s
t
h
at
i
n
cl
u
d
e
f
eat
u
r
e ex
t
r
act
i
o
n
,
n
o
r
m
al
i
zat
i
o
n
,
da
t
a
s
e
t
t
r
a
ns
f
or
m
a
t
i
ons
,
m
ode
l
t
r
a
i
ni
ng,
a
nd c
r
os
s
-
v
a
lid
a
tio
n
.
S
in
c
e
th
e
n
e
w
A
P
I
is
s
till
e
x
p
e
r
ime
n
ta
l
,
it
ma
y
be
s
ubj
e
c
t
t
o
m
a
j
or
c
ha
nge
s
i
n
t
h
e
n
ex
t
f
ew
S
p
ar
k
r
el
eas
es
.
O
v
er
t
i
m
e,
t
h
e
v
ar
i
o
u
s
f
eat
u
r
e
-
pr
oc
e
s
s
i
ng
t
e
c
hni
que
s
a
nd m
ode
l
s
t
ha
t
w
e
w
i
l
l
c
ove
r
w
i
l
l
s
i
m
pl
y be
por
t
e
d
t
o
t
he
ne
w
A
P
I
;
how
e
ve
r
,
t
he
c
or
e
c
onc
e
pt
s
a
nd m
os
t
unde
r
l
yi
ng c
ode
w
i
l
l
r
e
m
a
i
n l
a
r
ge
l
y
unc
ha
nge
d.
T
h
is
s
e
c
tio
n
s
u
mma
r
iz
e
d
t
he
t
e
s
t
s
c
a
r
r
i
e
d out
on
t
h
e
s
i
x da
t
a
s
e
t
s
l
i
s
t
e
d i
n
T
a
bl
e
1.
T
he
f
i
nd
i
ngs
w
e
r
e
pr
ovi
de
d i
n
t
e
r
m
s
of
t
he
p
r
oc
e
s
s
i
ng t
i
m
e
f
or
M
L
l
i
b a
nd
M
O
A
w
he
n
bot
h
pr
og
r
a
m
s
w
e
r
e
r
un
on
t
he
s
a
m
e
h
ar
d
w
ar
e.
T
h
e p
e
r
f
o
r
m
an
ce
o
f
A
p
ach
e S
p
ar
k
M
L
l
i
b
2
.
0
w
as
co
m
p
ar
ed
an
d
ev
al
u
at
ed
o
n
s
ix
d
is
tin
c
t la
r
g
e
da
t
a
s
e
t
s
obt
a
i
ne
d f
r
om
t
he
U
ni
ve
r
s
i
t
y
of
C
a
l
i
f
or
n
i
a
,
I
r
vi
ne
's
m
ach
i
n
e
l
ear
n
i
n
g
r
e
pos
i
t
or
y
.
T
he
e
xpe
r
i
m
e
nt
a
l
s
e
t
up us
e
d i
n t
hi
s
w
or
k c
ons
i
s
t
e
d of
a
s
t
a
nda
l
one
S
pa
r
k c
l
us
t
e
r
t
ha
t
m
a
ke
s
us
e
of
a
n H
D
F
S
s
t
or
a
ge
s
ys
t
e
m
an
d
A
p
ach
e Z
ep
p
el
i
n
0
.
7.
1
a
s
a
n
e
di
t
o
r
,
bot
h
of
w
hi
c
h w
e
r
e
de
ve
l
ope
d by
t
he
a
ut
hor
s
.
T
he
S
pa
r
k
c
l
us
t
e
r
i
s
m
a
de
up of
t
he
f
ol
l
ow
i
ng c
om
pone
nt
s
:
a
m
a
s
t
e
r
n
ode
t
ha
t
r
uns
a
dr
i
ve
r
s
of
t
w
a
r
e
;
t
h
r
e
e
w
or
ke
r
node
s
;
a
nd a
da
t
a
node
(
i
nc
l
ude
s
1 w
or
ke
r
node
t
ha
t
e
xe
c
ut
e
s
on t
he
m
a
s
t
e
r
nod
e
)
.
S
imila
r
to
th
e
d
e
s
ig
n
illu
s
tr
a
te
d
in
T
a
bl
e
2,
t
he
t
hr
e
e
node
s
ha
d a
s
i
m
i
l
a
r
c
onf
i
gu
r
a
t
i
o
n.
T
he
t
h
r
e
e
w
or
ke
r
node
s
e
a
c
h
ha
d a
m
e
m
or
y
c
a
p
a
c
i
t
y of
48 G
B
,
a
nd e
a
c
h w
or
ke
r
node
w
a
s
c
onf
i
gur
e
d w
i
t
h f
our
e
xe
c
ut
or
s
(
e
a
c
h w
i
t
h a
m
e
m
o
r
y c
a
pa
c
i
t
y
of
4 G
B
)
a
nd t
w
o C
P
U
s
.
E
a
c
h w
or
ke
r
i
n
t
he
m
a
s
t
e
r
node
w
a
s
c
onf
i
gur
e
d w
i
t
h t
hr
e
e
e
xe
c
ut
or
s
(
e
a
c
h w
i
t
h
a
s
i
z
e
of
5 G
B
)
a
nd
t
w
o
c
or
e
s
,
a
s
s
how
n i
n
t
he
di
a
gr
a
m
.
A
t
ot
a
l
of
16
G
B
of
R
A
M
w
a
s
a
l
l
oc
a
t
e
d
t
o
t
he
dr
i
ve
r
pr
oc
e
s
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
el
eco
m
m
u
n
C
om
put
E
l
C
ont
r
o
l
L
ar
ge
s
c
al
e
dat
a anal
y
s
i
s
us
i
ng M
L
l
i
b
(
A
hm
e
d H
u
s
s
e
i
n A
l
i
)
1739
T
he
M
L
l
i
b w
a
s
r
un
on a
S
c
a
l
a
2
.
11.
8
P
L
i
n
a
S
pa
r
k
2.
2
.
1 c
l
us
t
e
r
,
w
i
t
h
H
a
doop 2
.
7.
3
s
e
r
vi
n
g a
s
t
he
di
s
t
r
i
but
e
d s
t
or
a
ge
de
vi
c
e
,
a
nd t
he
r
e
s
ul
t
s
w
e
r
e
pu
bl
i
s
he
d.
T
he
a
m
ount
of
R
A
M
a
va
i
l
a
bl
e
t
o t
he
e
xe
c
ut
or
s
i
n
e
a
c
h w
or
ke
r
node
w
a
s
c
ha
nge
d by
e
m
pl
oyi
ng
t
h
e
opt
i
m
a
l
num
be
r
of
da
t
a
pa
r
t
i
t
i
ons
i
n
or
de
r
t
o
o
bt
a
i
n t
he
f
as
t
es
t
p
o
s
s
i
b
l
e ex
ecu
t
i
o
n
t
i
m
e.
T
ab
l
e 1
d
es
cr
i
b
e
s
t
h
e ch
ar
act
er
i
s
t
i
cs
o
f
t
h
e d
at
as
et
s
t
h
at
w
er
e u
s
e
d
i
n
t
h
i
s
i
nve
s
t
i
ga
t
i
on i
n t
e
r
m
s
o
f
t
he
a
m
oun
t
of
a
t
t
r
i
but
e
s
,
r
e
c
or
ds
,
a
nd c
l
a
s
s
e
s
t
ha
t
t
he
y c
ont
a
i
ne
d.
T
ab
l
e 1
.
D
at
as
et
d
es
cr
i
p
t
i
o
n
D
a
ta
N
o of
re
c
o
rd
N
o
o
f
a
ttr
ib
u
te
s
N
o of
c
l
a
s
s
e
s
C
ovt
ype
581012
54
7
C
ovt
ype
-
2
581012
54
2
H
i
ggs
11,000,000
28
2
B
ot
ne
t
A
tta
c
k
s
7,062,606
115
10
D
o
ta
2
102944
116
2
SU
SY
5,000,000
18
2
T
a
bl
e
2.
S
ys
t
e
m
de
s
c
r
i
pt
i
on
P
ar
am
et
er
S
p
e
c
if
ic
a
tio
n
O
p
e
r
a
tin
g
s
y
s
te
m
W
i
ndow
s
10
C
PU
I
n
te
l®
C
o
r
e
™
i7
-
6700 C
P
U
@
3.40 G
H
z
w
i
t
h 8 l
ogi
c
a
l
c
or
e
s
M
e
m
or
y
16 G
B
N
o. of
w
or
ke
r
s
3
C
om
put
a
t
i
ona
l
f
r
a
m
e
w
or
k
A
pa
c
he
S
pa
r
k 2.2.1
C
o
mp
a
tib
le
f
r
a
me
w
o
r
k
R
a
doop
D
SS
H
D
F
S
(
H
a
doop 2.7.3)
C
ode
de
ve
l
opm
e
nt
e
di
t
or
A
pa
c
he
Z
e
ppe
l
i
n 0.7.1
C
odi
ng l
a
ngua
ge
S
c
a
l
a
2.11.8
4.
RE
S
UL
T
S
AND DI
S
CUS
S
I
O
N
T
he
i
m
pl
e
m
e
nt
a
t
i
on pr
oc
e
s
s
w
a
s
ki
c
ke
d of
f
by f
i
r
s
t
de
f
i
ni
ng t
he
S
pa
r
k c
ont
e
xt
f
or
t
he
pr
ogr
a
m
t
ha
t
w
a
s
s
e
le
c
te
d
.
A
s
p
r
e
v
io
u
s
ly
s
ta
te
d
,
th
is
is
th
e
p
r
i
m
a
r
y poi
nt
o
f
e
nt
r
y
f
or
S
pa
r
k
f
unc
t
i
ona
l
i
t
y,
a
nd
i
t
m
us
t
be
g
i
v
en
b
ef
o
r
e at
t
em
p
t
i
n
g
t
o
c
r
eat
e t
h
e R
D
D
s
.
T
h
e
t
h
r
ee S
p
ar
k
C
o
n
t
ex
t
p
ar
am
et
er
s
,
w
h
i
ch
ar
e
t
h
e ap
p
l
i
cat
i
o
n
na
m
e
,
t
he
num
be
r
of
c
or
e
s
,
a
nd t
he
U
R
L
of
t
he
c
l
us
t
e
r
,
w
e
r
e
a
l
s
o s
uppl
i
e
d i
n
t
he
c
onf
i
g
ur
a
t
i
on.
I
n
a
ddi
t
i
on,
t
he
na
m
e
of
t
he
a
ppl
i
c
a
t
i
on
s
houl
d be
s
i
gni
f
i
c
a
nt
i
n o
r
de
r
t
o c
l
e
a
r
l
y
i
de
nt
i
f
y
t
he
pr
ogr
a
m
's
obj
e
c
t
i
ve
.
T
o
s
p
eci
f
y
t
h
e n
am
e o
f
an
ap
p
l
i
cat
i
o
n
f
o
r
a l
o
cal
cl
u
s
t
er
,
t
h
e k
ey
w
o
r
d
“l
o
cal
” i
s
u
s
ed
.
Wo
r
k
er
n
o
d
es
ar
e
r
e
s
pons
i
bl
e
f
or
pr
oc
e
s
s
i
n
g w
or
k
i
n
S
pa
r
k a
nd
,
a
s
p
r
e
vi
ous
l
y s
t
a
t
e
d,
t
he
nu
m
be
r
o
f
w
or
ke
r
node
s
t
o
be
f
or
m
e
d
i
s
di
c
t
a
t
e
d by
t
he
num
be
r
of
c
or
e
s
a
va
i
l
a
bl
e
.
T
he
f
ol
l
ow
i
ng s
t
e
p
i
s
t
o
t
r
a
i
n
t
he
m
ode
l
us
i
ng t
he
t
r
a
i
ni
ng da
t
a
an
d
t
o
p
r
o
v
i
d
e t
h
e p
a
r
am
et
er
s
t
h
at
a
r
e acces
s
i
b
l
e f
o
r
t
h
e s
u
p
e
r
vi
s
e
d m
a
c
hi
ne
l
e
a
r
ni
ng m
e
t
hods
t
ha
t
h
a
ve
be
e
n
s
el
ect
ed
(
s
uppor
t
ve
c
t
or
m
a
c
hi
ne
s
(
S
V
M
)
,
de
c
i
s
i
on t
r
e
e
,
a
nd
l
ogi
s
t
i
c
r
e
gr
e
s
s
i
on
)
.
T
h
e
p
ar
am
et
er
s
f
o
r
t
h
e
de
c
i
s
i
on t
r
e
e
,
S
V
M
,
a
nd l
ogi
s
t
i
c
r
e
gr
e
s
s
i
on m
e
t
hods
w
e
r
e
s
how
n i
n T
a
bl
e
s
3
,
4
,
a
nd 5
,
r
e
s
pe
c
t
i
ve
l
y.
T
h
e
te
s
tin
g
o
f
th
e
tr
a
in
e
d
mo
d
e
l o
n
th
e
te
s
tin
g
s
e
t is
th
e
n
e
x
t
s
te
p
; th
is
w
a
s
a
c
c
o
mp
lis
h
e
d
u
s
in
g
th
e
“
pr
e
di
c
t
”
m
e
t
hod w
hi
c
h w
a
s
i
m
pl
e
m
e
nt
e
d us
i
ng t
he
“
m
a
p”
t
r
a
ns
f
or
m
a
t
i
on of
S
pa
r
k
f
or
e
a
c
h
r
ow
o
f
t
he
t
e
s
t
s
e
t
.
T
he
c
om
pa
r
i
s
on
of
t
he
c
om
put
a
t
i
ona
l
t
i
m
e
of
A
pa
c
h
e
S
pa
r
k
M
L
l
i
b a
nd M
O
A
unde
r
di
f
f
e
r
e
nt
e
xpe
r
i
m
e
nt
a
l
c
ondi
t
i
ons
i
s
s
how
n i
n
F
i
gur
e
2
.
T
ab
l
e 3
.
T
h
e
d
eci
s
i
o
n
t
r
ee
cl
as
s
i
f
i
cat
i
o
n
t
ech
n
i
q
u
e
r
el
i
es
o
n
a n
u
m
b
er
o
f
p
ar
am
et
er
s
P
ar
am
et
er
E
xpl
a
na
t
i
on
V
al
u
e u
s
ed
ma
x
B
in
s
T
he
r
e
qui
r
e
d num
be
r
of
bi
ns
f
or
f
i
ndi
ng t
he
s
pl
i
t
s
a
t
e
a
c
h node
;
t
he
de
f
a
ul
t
va
l
ue
i
s
32.
32
min
I
n
f
o
G
a
in
T
he
m
i
ni
m
um
i
nf
o ga
i
n ne
e
de
d f
or
t
he
c
r
e
a
t
i
on of
a
s
pl
i
t
;
t
he
de
f
a
ul
t
va
l
ue
i
s
0.0
0.15
n
u
mC
la
s
s
e
s
T
h
e r
eq
u
i
r
ed
n
u
m
b
er
o
f
cl
as
s
es
t
o
ex
e
cu
t
e cl
a
s
s
i
f
i
cat
i
o
n
t
as
k
s
2
ma
x
D
e
p
th
T
h
e
ma
x
imu
m tr
e
e
d
e
p
th
; th
e
d
e
f
a
u
lt v
a
lu
e
is
5
.
6
imp
u
r
ity
T
he
r
e
qui
r
e
d c
r
i
t
e
r
i
on f
or
t
he
s
e
l
e
c
t
i
on of
i
nf
or
m
a
t
i
on ga
i
n (
gi
n
i
or
e
nt
r
opy)
.
e
nt
r
opy
T
ab
l
e 4
.
T
h
e
p
a
r
a
me
te
r
s
th
a
t w
e
r
e
u
s
e
d
in
th
e
S
V
M
c
la
s
s
if
ic
a
tio
n
a
lg
o
r
ith
m
P
ar
am
et
er
E
xpl
a
na
t
i
on
V
al
u
e u
s
ed
v
a
lid
a
te
D
a
ta
D
a
t
a
m
us
t
be
va
l
i
da
t
e
d by t
he
a
l
gor
i
t
hm
be
f
or
e
t
r
a
i
ni
ng
T
R
UE
ite
r
a
tio
n
s
T
he
num
be
r
of
c
ons
i
de
r
e
d i
t
e
r
a
t
i
ons
1000
n
u
mC
la
s
s
e
s
T
he
num
be
r
of
c
ons
i
de
r
e
d c
l
a
s
s
e
s
;
t
he
de
f
a
ul
t
va
l
ue
i
s
2.
2
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
1693
-
6930
T
E
L
KOM
NI
KA
T
el
eco
m
m
u
n
C
om
put
E
l
C
ont
r
o
l
,
Vo
l
.
19
, N
o
.
5
,
O
c
t
obe
r
2021:
1735
-
1746
1740
T
ab
l
e 5
.
T
h
e p
a
r
am
et
er
s
t
h
at
w
er
e
u
s
ed
i
n
t
h
e
lo
g
is
tic
r
e
g
r
e
s
s
io
n
a
l
gor
i
t
hm
P
ar
am
et
er
E
xpl
a
na
t
i
on
V
al
u
e u
s
ed
v
a
lid
a
te
D
a
ta
D
a
t
a
m
us
t
be
va
l
i
da
t
e
d by t
he
a
l
gor
i
t
hm
be
f
or
e
t
r
a
i
ni
ng
T
R
UE
ite
r
a
tio
n
s
T
he
num
be
r
of
c
ons
i
de
r
e
d i
t
e
r
a
t
i
ons
1000
n
u
mC
la
s
s
e
s
T
he
num
be
r
of
c
ons
i
de
r
e
d c
l
a
s
s
e
s
;
t
he
de
f
a
ul
t
va
l
ue
i
s
2.
2
(
a)
(
b)
(
c)
(
d)
(
e)
(f)
F
i
gur
e
2.
C
om
put
a
t
i
ona
l
t
i
m
e
of
A
pa
c
he
S
pa
r
k M
L
l
i
b a
nd M
O
A
unde
r
di
f
f
e
r
e
nt
d
at
as
et
s
:
(
a
)
C
ove
t
ype
,
(
b
)
C
ove
t
ype
2,
(
c
)
H
i
ggs
,
(
d
)
B
o
t
n
et
at
t
ack
s
,
(
e
)
D
o
t
a2
a
nd (
f
)
SU
SY
d
at
ab
as
e
T
he
r
e
w
a
s
a
c
l
os
e
s
i
m
i
l
a
r
i
t
y
i
n
t
he
a
r
e
a
unde
r
t
he
R
O
C
f
or
bot
h A
pa
c
he
S
pa
r
k
M
L
l
i
b a
nd M
O
A
a
s
t
h
e d
i
f
f
er
en
ce b
et
w
een
t
h
em
w
as
not
s
ta
tis
tic
a
lly
s
ig
n
if
ic
a
n
t.
H
o
w
e
v
e
r
,
th
e
little
d
if
f
e
r
e
n
c
e
b
e
tw
e
e
n
th
e
m
co
u
l
d
b
e d
u
e t
o
t
h
e d
et
ai
l
ed
p
ar
am
et
r
i
c s
et
t
i
n
g
s
o
f
eac
h c
l
a
s
s
i
f
i
e
r
dur
i
ng t
he
r
a
ndom
s
e
l
e
c
t
i
on of
t
he
t
e
s
t
a
nd
t
r
a
i
n da
t
a
s
e
t
s
.
O
bvi
ous
l
y,
A
pa
c
he
S
pa
r
k M
L
l
i
b w
a
s
f
a
s
t
e
r
t
ha
n M
O
A
ba
s
e
d on t
he
obs
e
r
ve
d c
om
put
a
t
i
ona
l
time
s
o
f
th
e
c
la
s
s
if
ie
r
s
; h
o
w
e
v
e
r
,
th
e
c
lu
s
te
r
in
g
m
e
th
o
d
s
h
o
w
e
d
s
ta
tis
tic
a
lly
s
ig
n
if
ic
a
n
t d
if
f
e
r
e
n
ces
b
et
w
een
t
he
A
pa
c
he
S
pa
r
k M
L
l
i
b a
nd M
O
A
.
5.
CO
NCL
US
I
O
N
D
at
a g
en
er
at
i
o
n
h
as
i
n
cr
eas
ed
at
an
al
ar
m
i
n
g
r
at
e i
n
r
ecen
t
y
ear
s
,
n
eces
s
i
t
at
i
n
g
ad
v
an
c
em
en
t
s
i
n
d
at
a
a
na
l
yt
i
c
s
a
nd pr
oc
e
s
s
i
ng t
ool
s
i
n or
de
r
t
o e
na
bl
e
t
he
e
xt
r
a
c
t
i
on of
r
e
l
e
va
nt
i
nf
o
r
m
a
t
i
on f
r
om
va
s
t
a
m
ount
s
of
or
ga
ni
z
e
d a
nd uns
t
r
uc
t
ur
e
d da
t
a
.
B
i
g da
t
a
m
a
c
hi
n
e
l
e
a
r
ni
ng t
e
c
hni
que
s
w
hi
c
h a
r
e
be
l
i
e
ve
d t
o be
e
f
f
i
c
i
e
nt
i
n
pa
t
t
e
r
n f
i
ndi
ng,
c
a
n
be
u
se
d
t
o
m
or
e
e
f
f
i
c
i
e
nt
l
y ha
ndl
e
t
hi
s
c
ha
l
l
e
nge
.
A
pa
c
he
S
pa
r
k
M
L
l
i
b i
s
a
w
i
d
e
l
y us
e
d
m
a
c
hi
ne
l
e
a
r
ni
ng l
i
br
a
r
y f
o
r
bi
g da
t
a
,
a
nd i
t
i
s
a
p
ow
e
r
f
ul
t
ool
f
o
r
bi
g da
t
a
a
na
l
yt
i
c
s
.
A
s
pr
ove
d i
n t
hi
s
s
t
udy,
it p
r
o
v
id
e
s
e
x
c
e
lle
n
t
p
e
r
f
o
r
ma
n
c
e
in
te
r
ms
o
f
c
o
mp
u
ta
tio
n
a
l time
.
M
a
s
s
iv
e
o
n
lin
e
a
n
a
ly
tic
s
(
M
O
A
)
,
on
t
he
ot
he
r
ha
nd,
i
s
s
l
i
ght
l
y s
l
ow
e
r
t
ha
n A
pa
c
he
S
pa
r
k M
L
l
i
b dur
i
ng b
i
g
d
at
a an
al
y
s
i
s
;
h
o
w
ev
er
,
b
ec
au
s
e t
h
e
c
l
a
s
s
i
f
i
e
r
s
us
e
di
f
f
e
r
e
nt
c
onf
i
gur
a
t
i
ons
a
nd f
i
l
e
s
ys
t
e
m
s
,
t
he
c
om
pa
r
i
s
on m
a
y not
be
a
ppr
opr
i
a
t
e
.
M
L
l
i
b w
a
s
imp
le
me
n
te
d
o
n
th
e
S
p
a
r
k
d
is
tr
ib
u
te
d
f
ile
s
y
s
te
m,
w
h
e
r
e
a
s
th
e
M
O
A
c
la
s
s
if
ie
r
w
a
s
imp
le
me
n
te
d
o
n
th
e
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
el
eco
m
m
u
n
C
om
put
E
l
C
ont
r
o
l
L
ar
ge
s
c
al
e
dat
a anal
y
s
i
s
us
i
ng M
L
l
i
b
(
A
hm
e
d H
u
s
s
e
i
n A
l
i
)
1741
H
a
doop di
s
t
r
i
but
e
d
f
i
l
e
s
ys
t
e
m
,
t
he
c
om
pa
r
i
s
on
m
a
y not
be
a
ppr
op
r
i
a
t
e
.
B
e
c
a
us
e
w
e
w
a
nt
t
o
de
m
ons
t
r
a
t
e
h
o
w
w
el
l
S
p
ar
k
p
er
f
o
r
m
s
o
n
l
ar
g
e d
at
a s
et
s
u
s
i
n
g
M
O
A
as
a b
en
ch
m
ar
k
,
i
t
i
s
as
s
u
m
ed
t
h
at
t
h
er
e ar
e m
an
y
M
O
A
f
eat
u
r
es
t
h
at
S
p
ar
k
can
n
o
t
co
m
p
et
e w
i
t
h
,
s
u
ch
as
t
h
e av
a
i
l
a
bi
l
i
t
y of
a
l
a
r
ge
pool
o
f
r
e
s
ou
r
c
e
s
a
nd
doc
um
e
nt
s
f
or
M
O
A
us
e
r
s
,
t
he
e
a
s
e
w
i
t
h w
hi
c
h
non
-
ex
p
er
t
s
can
i
m
p
l
em
en
t
M
O
A
,
an
d
t
h
e p
r
es
en
ce o
f
a
good gr
a
phi
c
a
l
us
e
r
i
nt
e
r
f
a
c
e
i
n M
O
A
,
a
m
ong ot
he
r
t
hi
ngs
.
T
he
s
e
c
ha
r
a
c
t
e
r
i
s
t
i
c
s
a
r
e
t
he
r
e
a
s
on w
hy M
O
A
s
uppor
t
s
a
v
ar
i
et
y
o
f
m
ach
i
n
e l
ear
n
i
n
g
t
ech
n
i
q
u
es
.
A
C
K
N
O
W
LED
G
EM
EN
TS
T
he
a
ut
hor
s
w
oul
d l
i
ke
t
o
t
ha
nk
I
C
C
I
,
I
nf
or
m
a
t
i
c
s
I
ns
t
i
t
ut
e
f
or
P
os
t
gr
a
dua
t
e
S
t
udi
e
s
,
I
r
a
qi
a
U
ni
ve
r
s
i
t
y,
a
nd
A
l
S
a
la
m
U
n
iv
e
r
s
ity
C
o
lle
g
e
f
or
t
he
i
r
f
a
c
i
l
i
t
i
e
s
a
nd s
uppor
t
;
a
nd U
ni
ve
r
s
i
t
a
s
A
hm
a
d
D
a
hl
an
t
o s
uppor
t
t
hi
s
c
ol
l
a
bo
r
a
t
i
ve
r
e
s
e
a
r
c
h
.
R
EF
ER
EN
C
ES
[1
]
V.
C
ha
ng,
“
A
n e
th
ic
a
l f
r
a
m
e
w
or
k f
or
b
ig da
ta
a
nd s
m
a
r
t c
it
ie
s,
”
T
e
c
hn
ol.
F
ore
c
as
t.
Soc
.
C
ha
nge
,
vo
l.
16
5,
p.
120
55
9,
20
21
,
do
i:
1
0.
10
16
/j.
te
c
hf
or
e
.
2
02
0.
12
05
59
.
[2
]
S
.
A.
Alom
a
r
i,
M
.
S
.
Alz
b
oo
n,
M
.
S
.
Al
-
B
a
ta
h,
a
n
d
B
.
Z
a
qa
ibe
h,
“
A no
ve
l a
da
p
ti
ve
sc
h
e
m
a
to f
a
c
il
ita
te
s
pla
yba
c
k s
wi
tc
h
in
g te
c
hn
iq
ue
f
or
v
ide
o de
li
ve
r
y in
de
nse
L
TE c
e
ll
ul
a
r
he
t
e
r
o
ge
ne
ou
s ne
tw
or
k e
nv
ir
o
nm
e
n
ts
,
”
I
nte
rn
at
io
na
l J
our
na
l of E
le
c
t
ric
al a
nd C
o
mp
ute
r En
gi
ne
e
r
in
g
,
vol.
10,
no.
5
,
pp.
53
47
–
53
67
,
2
02
0,
doi
: 1
0.
11
59
1/
i
je
c
e
.
v
10
i5
.
pp
53
47
-
53
67
.
[3
]
S
.
R
e
shm
a
a
nd C
he
ta
na
pr
a
ka
s
h,
“
Ad
va
n
c
e
m
e
n
t in
inf
ot
a
inm
e
nt s
ys
te
m
i
n a
u
tom
ot
ive
se
c
tor
wi
th
ve
h
ic
u
la
r
c
l
o
ud
ne
t
wor
k a
nd c
ur
r
e
nt s
ta
te
of
a
r
t,
”
I
nte
rn
at
io
na
l J
ou
rn
al
of Ele
c
tr
ic
a
l an
d C
om
pu
te
r E
ng
in
e
e
r
in
g
,
vol.
1
0,
no.
2
,
pp.
20
77
–
2
08
7,
20
20,
d
oi
: 1
0.
11
59
1/
ij
e
c
e
.
v1
0i
2.
p
p2
07
7
-
20
87.
[4
]
W
.
N.
W
.
Ab
d M
a
na
n a
nd M
.
A.
S
a
la
m
a
t,
“
C
onc
e
pt of
m
inim
iz
in
g the
r
e
sp
on
se
tim
e
f
or
r
e
du
c
i
ng d
y
n
a
m
i
c
d
a
t
a
r
e
du
nda
nc
y in c
l
ou
d c
om
p
ut
in
g,
”
I
ndo
ne
s
ia
n J
ou
rn
al o
f Ele
c
t
ric
al E
ng
ine
e
ri
ng a
nd C
om
pu
te
r Sc
ie
nc
e
,
vo
l.
1
5,
no.
3
,
pp.
1
59
7
–
16
02,
2
01
9,
do
i: 10.
1
15
91
/i
je
e
c
s.
v1
5.
i
3.
pp1
59
7
-
16
02.
[5
]
M
.
El G
hm
a
r
y,
T.
C
ha
ny
our
,
Y.
Hm
im
z
,
a
nd
M
.
O.
C
.
M
a
lk
i,
“
Ef
f
ic
i
e
nt m
ul
ti
-
ta
s
k of
f
loa
di
ng
wi
th
e
ne
r
gy a
n
d
c
om
p
uta
ti
ona
l r
e
s
our
c
e
s o
pt
im
iz
a
t
io
n i
n a
m
o
bi
le
e
dge
c
om
p
ut
in
g n
ode
,
”
I
nte
rn
at
io
na
l J
o
ur
na
l o
f
Ele
c
tr
ic
a
l a
n
d
C
omp
ut
e
r E
ng
ine
e
ri
ng
, v
o
l
. 9
, n
o
. 6
,
pp.
4
90
8
–
49
19,
2
0
19,
do
i:
10.
1
15
91
/i
je
c
e
.
v
9i
6.
p
p4
90
8
-
49
19
.
[6
]
K
.
K.
C
ha
kr
a
va
r
t
hi a
nd V.
Vi
ja
ya
kum
a
r
,
“
W
or
kf
lo
w
sc
he
du
li
ng te
c
hn
iq
ue
s a
nd a
lg
or
i
thm
s in I
a
a
S
c
lo
ud
:
A
sur
ve
y,
”
I
n
te
r
na
ti
on
al J
o
ur
na
l o
f Ele
c
tr
ic
a
l a
nd C
om
put
e
r E
ng
ine
e
ri
ng
,
v
ol.
8,
n
o.
2
,
pp.
1
25
6
–
12
68,
2
01
8,
doi
: 1
0.
11
59
1/
ij
e
c
e
.
v8
i2.
pp
12
56
-
12
68.
[7
]
M.
R.
B
e
lga
um
,
Z
.
Ala
nsa
r
i,
S
.
M
usa
,
M
.
M
.
Ala
m
,
a
nd M
.
S
.
M
a
z
liha
m
,
“
R
ole
of
a
r
tif
ic
ia
l i
nte
ll
ige
nc
e
in c
l
ou
d
c
om
p
ut
in
g,
I
oT a
nd S
DN
: R
e
lia
bi
li
ty a
n
d sc
a
la
bi
li
ty i
s
sue
s,
”
I
nte
r
na
ti
on
al J
o
ur
na
l of Ele
c
tr
ic
a
l an
d C
om
pu
te
r
E
ng
in
e
e
ri
ng
,
vo
l.
11,
n
o.
5
,
pp.
4
45
8
–
4
47
0,
2
02
1,
do
i: 1
0.
11
59
1/
ije
c
e
.
v1
1i
5.
pp
44
58
-
44
70.
[8
]
J.
Ka
r
a
nde
a
n
d S
.
Joshi,
“
D
ED
A: An a
l
gor
it
hm
f
or
e
a
r
l
y de
te
c
t
io
n of
to
po
lo
gy a
tt
a
c
k
s in the
i
nte
r
ne
t of
th
in
gs,
”
I
nte
rn
at
io
na
l J
our
na
l of E
le
c
t
ric
al a
nd C
o
mp
ute
r En
gi
ne
e
r
in
g
,
vol.
11,
no.
2
,
pp.
17
61
–
17
70
,
2
02
1,
doi
: 1
0.
11
59
1/
ij
e
c
e
.
v1
1i
2.
p
p1
76
1
-
17
70
.
[9
]
F
.
Ar
e
na
a
nd G
.
P
a
u,
“
An
ove
r
vie
w
of
b
ig
da
ta
a
na
ly
sis,
”
Bu
lle
ti
n
of Ele
c
tr
ic
a
l En
gin
e
e
r
in
g
a
nd
I
nf
or
ma
tic
s
,
vol.
9,
n
o.
4
,
pp.
1
64
6
–
16
53,
2
02
0,
do
i: 10.
1
15
91
/e
e
i.
v9
i4.
2
35
9.
[
10]
K.
F
.
M
a
hm
m
o
d,
M
.
M
.
Az
e
e
z
,
a
n
d Z
.
H.
I
sm
a
e
l,
“
De
s
ig
n a
n a
c
ti
ve
ve
r
if
ic
a
t
io
n
m
e
c
ha
ni
sm
f
or
c
e
r
t
if
ic
a
t
e
s
r
e
voc
a
ti
on i
n OC
S
P
f
or
in
te
r
ne
t a
u
the
nt
ic
a
t
io
n,
”
I
nte
rn
ati
on
al
J
ou
rn
al of
E
le
c
t
ric
al an
d
C
om
pu
te
r En
gi
ne
e
r
in
g
,
vol.
1
0,
no.
4
,
p
p.
42
08
–
4
21
6,
20
20,
d
oi
: 1
0.
11
59
1/
ij
e
c
e
.
v10
i4.
p
p4
20
8
-
42
16.
[
11]
H.
Hos
se
i
nia
n,
H.
S
ha
hinz
a
de
h,
G
.
B
.
G
ha
r
e
hpe
t
ia
n
,
Z.
Az
a
ni,
a
nd M
.
S
ha
ne
h,
“
B
loc
kc
ha
i
n ou
tl
oo
k f
or
de
p
lo
ym
e
n
t of
I
oT
in
di
str
ib
ut
io
n ne
tw
or
k
s a
n
d sm
a
r
t h
om
e
s,
”
I
n
te
r
na
ti
on
al J
ou
rn
al
of
Ele
c
tr
ic
a
l a
nd C
om
pu
t
e
r
E
ng
in
e
e
ri
ng
,
vo
l.
10,
n
o.
3
,
pp.
2
78
7
–
27
96,
2
02
0,
do
i:
1
0.
11
59
1/
ije
c
e
.
v1
0i
3.
pp
27
87
-
27
96.
[
12]
A.
Oua
c
ha
a
n
d M
.
El G
hm
a
r
y,
“
V
ir
t
ua
l m
a
c
hi
ne
m
i
gr
a
tio
n i
n m
e
c
ba
se
d a
r
tif
ic
ia
l i
nte
ll
ige
nc
e
te
c
hn
iq
ue
,
”
I
AES
I
nte
rn
at
io
na
l
J
o
urn
al of Ar
ti
fic
ia
l
I
n
te
l
li
ge
n
c
e
,
vol.
1
0,
n
o.
1
,
pp
.
2
44
–
25
2,
2
02
1,
doi
: 1
0.
11
59
1/
ij
a
i.
v
10.
i1.
p
p2
44
-
25
2.
[
13]
A
. H
. A
li a
nd M
.
Z
.
Abdul
la
h,
“
A pa
r
a
l
le
l gr
id o
pt
im
iz
a
ti
on of
S
VM
hype
r
pa
r
a
m
e
te
r
f
or
big da
ta
c
la
s
sif
ic
a
t
io
n
usi
ng spa
r
k
R
a
d
oo
p,
”
K
ar
ba
la I
n
t.
J
.
Mod.
Sc
i.
,
v
o
l
.
6
,
n
o
.
1
,
p
p
.
5
-
18
,
20
20
,
do
i:
10
.
33
64
0/
24
05
-
6
09
X.
12
70
.
[
14]
P
r
oni
ka
a
n
d S
.
S
.
Tya
g
i,
“
P
e
r
f
or
m
a
nc
e
a
na
ly
sis
of
e
nc
r
ypt
io
n a
n
d de
c
r
yp
ti
on a
lg
or
i
thm
,
”
I
nd
on
e
si
an
J
ou
rn
al
of
Ele
c
tr
ic
a
l En
gi
ne
e
r
in
g a
nd
C
om
pu
te
r Sc
ie
nc
e
,
vol.
2
3,
no.
2
,
p
p.
10
30
–
10
38
,
20
21
,
doi
: 1
0.
11
59
1/
ij
e
e
c
s.
v
23.
i2.
p
p1
03
0
-
10
38
.
[
15]
Y.
A.
G
ha
ni A
ly
ouz
ba
k
i a
nd M
.
F
.
Al
-
R
a
w
i,
“
No
ve
l l
o
a
d ba
la
nc
i
ng a
p
pr
oa
c
h ba
se
d o
n a
n
t c
ol
on
y op
tim
iz
a
t
io
n
te
c
h
ni
que
i
n c
lo
ud c
om
pu
ti
ng,
”
Bu
lle
ti
n of E
le
c
t
ric
al E
n
gin
e
e
r
in
g and I
n
fo
rm
at
ic
s
,
v
ol.
10,
n
o.
4
,
pp.
232
0
–
2
326
,
202
1,
do
i:
10.
1
15
91
/e
e
i.
v1
0i
4.
2
94
7
.
[
16]
A.
M
.
N.
G
.
M
olk,
M
.
R
.
Ar
e
f
,
a
nd R
.
R
.
Khor
s
hi
dd
ou
st,
“
A
na
l
ys
is of
d
e
s
ig
n goa
ls
of
c
r
yp
to
gr
a
p
hy a
lg
or
i
thm
s
ba
se
d
o
n d
if
f
e
r
e
n
t c
om
po
ne
n
ts,
”
I
n
do
ne
s
ia
n J
o
ur
na
l
o
f
Ele
c
tr
ic
a
l
E
ng
ine
e
ri
ng
an
d C
o
mp
ute
r Sc
ie
nc
e
,
v
o
l
.
2
3
,
no.
1
,
pp.
5
40
–
5
48,
2
02
1,
do
i:
10.
1
15
91
/
i
je
e
c
s.
v2
3.
i
1.
pp
540
-
5
48.
[
17]
Y.
N.
S
e
itk
ul
ov,
S
.
N.
B
or
a
nba
ye
v,
G
.
B
.
Ulyuk
ova
,
B
.
B
.
Ye
r
ga
li
ye
va
,
a
n
d D.
S
a
ty
ba
l
di
na
,
“
M
e
th
od
s f
or
se
c
ur
e
c
lo
ud pr
oc
e
ss
in
g
of
bi
g
da
ta
,
”
I
n
do
ne
s
ia
n J
o
ur
na
l o
f
Ele
c
tr
ic
a
l E
ng
ine
e
ri
ng
an
d C
o
mp
ute
r Sc
ie
nc
e
,
v
o
l
.
2
2
,
no.
3
,
p
p
.
16
50
–
16
58,
2
02
1,
do
i: 10.
1
15
91
/i
je
e
c
s.
v2
2.
i
3.
pp1
65
0
-
16
58.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
1693
-
6930
T
E
L
KOM
NI
KA
T
el
eco
m
m
u
n
C
om
put
E
l
C
ont
r
o
l
,
Vo
l
.
19
, N
o
.
5
,
O
c
t
obe
r
2021:
1735
-
1746
1742
[
18]
M
. F
. F
a
l
a
h
e
t al.
,
“
C
om
pa
r
i
so
n of
c
lo
ud c
om
p
ut
in
g
pr
ovi
de
r
s f
or
de
ve
lo
pm
e
n
t of
bi
g da
ta
a
nd
in
te
r
ne
t of
th
in
gs
a
pp
lic
a
ti
on,
”
I
n
do
ne
s
ia
n J
o
ur
na
l o
f E
le
c
t
ric
al
En
gi
ne
e
r
i
ng a
nd C
om
pu
te
r
Sc
ie
nc
e
,
v
ol.
22,
n
o.
3
,
pp
.
17
23
–
1
73
0,
202
1,
do
i: 10.
1
15
91
/i
je
e
c
s.
v2
2.
i
3.
pp
17
23
-
17
30.
[
19]
H.
Kim
,
J.
P
a
r
k,
a
nd H.
Jung,
“
Au
tom
a
ti
c
c
ontr
ol s
ys
te
m
ba
se
d on I
o
T da
ta
i
de
n
tif
ic
a
t
io
n,
”
I
ndo
ne
si
an J
ou
rn
al
of
Ele
c
tr
ic
a
l En
gi
ne
e
r
in
g a
nd
C
om
pu
te
r Sc
ie
nc
e
,
vol.
1
9,
no.
3
,
p
p.
15
25
–
15
32
,
20
20
,
doi
: 1
0.
11
59
1/
ij
e
e
c
s.
v
19.
i3.
p
p1
52
5
-
15
32
.
[
20]
W
.
B
udi
ha
r
t
o
a
nd
M
.
M
e
ilia
na
,
“
P
r
e
dic
ti
on
a
nd
a
na
ly
s
is of
I
nd
one
sia
P
r
e
si
de
n
tia
l e
le
c
t
io
n
f
r
om
Twi
tte
r
us
in
g
se
n
tim
e
nt a
na
ly
si
s,
”
J
.
Big da
ta
,
v
ol.
5,
n
o.
1,
pp.
1
–
1
0,
201
8
,
do
i:
10.
1
1
86
/s
40
53
7
-
0
18
-
01
64
-
1
.
[
21]
T.
Yi
gi
tc
a
n
la
r
,
N.
Ka
n
ka
na
m
ge
,
a
nd
K.
Ve
l
la
,
“
Ho
w a
r
e
sm
a
r
t c
i
ty c
o
nc
e
p
ts a
nd
te
c
h
no
lo
gie
s pe
r
c
e
i
ve
d a
n
d
uti
li
z
e
d?
A sy
ste
m
a
t
ic
ge
o
-
Tw
it
te
r
a
na
l
ys
is of
sm
a
r
t c
iti
e
s in A
us
tr
a
l
ia
,
”
J
.
Urba
n T
e
c
hn
ol.
,
vo
l.
28,
no.
1
–
2
,
pp.
13
5
–
1
54
,
20
21
,
do
i:
1
0.
10
80
/1
06
30
73
2.
2
02
0.
17
53
4
83
.
[
22]
H.
S
hir
da
st
ia
n,
M
.
L
a
r
oc
he
,
a
nd M
.
-
O.
R
ic
ha
r
d,
“
U
si
ng
big
da
ta
a
na
ly
tic
s to s
tu
dy br
a
nd a
u
the
nt
ic
i
ty se
nt
im
e
nts
:
The
c
a
se
of
S
ta
r
b
uc
k
s on Tw
it
te
r
,
”
I
nt.
J
.
I
nf.
Mana
ge
.
,
vol.
48,
pp.
2
91
–
30
7
,
2
019
,
doi
:
1
0.
10
16
/j.
ij
inf
om
g
t.
20
17.
09.
0
07
.
[
23]
Y.
Ale
m
a
m
i,
M
.
A.
M
oha
m
e
d,
S
.
Atie
wi,
a
n
d M
.
M
a
m
a
t,
“
S
pe
e
c
h e
nc
r
y
pt
io
n b
y m
ul
ti
ple
c
ha
ot
ic
m
a
ps w
it
h f
a
s
t
fo
u
r
i
er t
ran
s
fo
rm,
”
I
nte
rn
at
io
na
l J
ou
rn
al
of
Ele
c
tr
ic
a
l a
nd C
o
mp
ute
r E
ng
ine
e
ri
ng
,
v
ol.
1
0,
no.
6
,
pp.
5
65
8
–
5
664
,
202
0,
do
i: 10.
1
15
91
/i
je
c
e
.
v
10
i6.
pp
56
58
-
56
64.
[
24]
A.
Alr
um
a
ih,
A.
Al
-
S
a
bba
gh,
R
.
Alsa
ba
h,
H.
Kha
r
r
uf
a
,
a
nd J.
B
a
ldwi
n,
“
S
e
nt
im
e
n
t a
na
l
ys
is of
c
om
m
e
nt
s i
n
soc
ia
l m
e
d
ia
,
”
I
n
te
r
na
ti
on
al J
ou
rn
al
of
Ele
c
tr
ic
a
l a
nd
C
omp
ute
r E
ng
ine
e
ri
ng
,
vo
l.
10,
n
o.
6
,
p
p.
59
17
–
59
22
,
202
0,
do
i: 10.
1
15
91
/i
je
c
e
.
v
10
i6.
pp
59
17
-
59
22.
[
25]
P
.
Anita
,
M
.
De
vi,
a
nd P
.
Ani
ta
,
“
Hi
gh pe
r
f
or
m
a
nc
e
m
odi
f
ie
d bit
-
ve
c
tor
ba
se
d pa
c
ke
t c
la
ss
if
ic
a
ti
on m
o
du
le
o
n
low
-
c
o
st F
P
G
A,
”
I
nt
e
r
na
ti
on
al J
o
ur
na
l o
f E
le
c
t
ric
al
an
d C
om
pu
te
r
En
gi
ne
e
r
in
g
,
vo
l.
1
1,
no.
5
,
p
p.
38
55
–
38
63,
202
1,
do
i: 10.
1
15
91
/i
je
c
e
.
v
11
i5.
pp
38
55
-
38
63.
[
26]
N.
S
r
ide
vi
a
n
d M
.
M
e
e
na
ks
hi,
“
Ef
f
ic
ie
n
t r
e
c
o
nf
i
gur
a
b
le
a
r
c
h
ite
c
tur
e
f
or
m
ov
in
g o
bje
c
t de
te
c
ti
on
wi
th m
ot
io
n
c
om
pe
nsa
ti
on,
”
I
nd
one
si
an J
o
ur
na
l of Ele
c
tr
ic
a
l En
gi
ne
e
ri
ng an
d C
om
pu
te
r Sc
ie
nc
e
,
vo
l.
23,
no.
2
,
pp.
802
–
810
,
202
1,
do
i: 10.
1
15
91
/i
je
e
c
s.
v2
3.
i
2.
pp
8
02
-
8
10.
[
27]
P
.
Ve
e
r
e
sh,
R
.
P
r
a
ve
e
n S
a
m
,
a
nd C
.
S
hoba
B
in
du,
“
R
e
lia
ble
f
a
u
lt t
ole
r
a
nc
e
sy
ste
m
f
or
s
e
r
v
ic
e
c
om
p
os
it
io
n i
n
m
obi
le
Ad H
oc
ne
t
wor
k,
”
I
nte
rn
at
io
na
l J
ou
rn
al of
Ele
c
tr
ic
al
an
d C
om
pu
te
r E
ng
ine
e
ri
ng
,
v
ol.
9,
no.
4.
pp.
25
23
–
2
53
3,
20
19,
d
oi
: 1
0.
11
59
1/
ij
e
c
e
.
v9
i4.
pp
25
23
-
253
3.
[
28]
E.
A.
Ab
de
s
sa
m
a
d,
N.
B
a
hr
i,
A.
M
a
n
so
ur
i,
N.
M
a
sm
ou
d,
a
n
d A.
A
li,
“
Ar
e
a
a
nd
po
we
r
e
f
f
ic
ie
n
t
VL
S
I
a
r
c
hi
te
c
t
ur
e
of
m
o
de
de
c
is
io
n i
n in
te
g
e
r
m
ot
io
n e
s
ti
m
a
ti
on f
or
HE
VC
v
ide
o c
od
in
g s
ta
n
da
r
d,
”
I
nte
rn
at
io
na
l
J
ou
rn
al of Ele
c
tr
ic
a
l a
n
d C
o
mp
ute
r E
ng
in
e
e
rin
g
,
v
o
l
.
9
,
n
o
.
4
,
pp.
24
69
–
24
80
,
2
01
9,
doi
: 1
0.
11
59
1/
ij
e
c
e
.
v9
i4.
pp
24
69
-
24
80.
[
29]
L
.
E.
Aik,
T.
W
.
Ho
ng,
a
n
d A.
K.
Ju
no
h,
“
An
im
pr
ove
d r
a
dia
l ba
si
s f
unc
ti
on
ne
t
wor
ks ba
se
d
on
qua
nt
u
m
e
vo
lu
ti
ona
r
y a
lg
or
i
thm
f
or
tr
a
in
in
g no
nl
ine
a
r
da
ta
se
ts,
”
I
AES I
n
te
r
na
ti
on
al J
o
ur
na
l of Ar
ti
fi
c
i
al I
nt
e
l
li
ge
nc
e
,
vol.
8,
n
o.
2
,
pp.
1
20
–
1
31,
2
01
9,
do
i:
10.
1
15
91
/i
ja
i.
v8.
i2
.
pp1
20
-
13
1.
[
30]
C
.
Jitta
wir
iya
nu
ko
on
,
“
G
r
a
nula
r
it
y a
na
l
ys
is of
c
la
s
sif
ic
a
ti
on a
nd e
s
tim
a
ti
on f
or
c
om
p
le
x da
ta
se
ts w
it
h M
OA,
”
I
nte
rn
at
io
na
l J
o
ur
na
l o
f
Ele
c
tr
ic
a
l a
nd
C
o
mp
ute
r
En
gi
ne
e
r
in
g
,
v
ol.
9,
n
o.
1
,
pp
.
40
9
–
4
16,
20
1
9,
doi
: 1
0.
11
59
1/
ij
e
c
e
.
v9
i1.
pp
40
9
-
4
16.
[
31]
S
.
Alz
a
e
e
m
i,
M
.
A.
M
a
ns
or
,
M
.
S
.
M
oh
d
Ka
si
hm
u
dd
i
n,
S
.
S
a
tha
si
va
m
,
a
nd
M
.
M
a
m
a
t,
“
R
a
d
ia
l
ba
s
is f
unc
ti
on
ne
ur
a
l ne
tw
or
k
f
or
2 sa
ti
sf
ia
bi
li
ty pr
ogr
a
m
m
in
g,
”
I
ndo
ne
s
ia
n J
our
na
l of
E
le
c
t
ric
al
E
ng
ine
e
ri
ng
a
nd C
om
pu
te
r
Sc
ie
nc
e
,
vo
l.
18,
n
o.
1
,
pp.
4
59
–
4
69,
2
01
9,
do
i: 1
0.
1
15
9
1/i
je
e
c
s.
v1
8.
i
1.
pp
45
9
-
4
69.
[
32]
K.
S
ur
e
sh a
n
d R
.
J
.
Ka
nna
n,
“
R
e
vie
w of
a
dva
nc
e
m
e
nt
s in m
u
lt
i
-
te
na
n
t f
r
a
m
e
wo
r
k in c
lo
ud c
om
pu
ti
n
g,
”
I
nd
one
si
an J
o
ur
na
l of
Ele
c
tr
ic
a
l En
gi
ne
e
ri
ng a
nd C
o
mp
ute
r Sc
ie
nc
e
,
vo
l.
11,
n
o.
3
,
pp.
1
10
2
–
1
10
8
,
20
18,
doi
: 1
0.
11
59
1/
ij
e
e
c
s.
v
11.
i3.
p
p1
10
2
-
11
08
.
[
33]
N.
Ed
wa
r
d
a
n
d J.
Elc
oc
k,
“
Ta
s
k sc
he
d
ul
in
g i
n he
te
r
o
ge
ne
ou
s m
u
lt
ipr
oc
e
s
sor
e
nv
ir
o
n
m
e
nt
s
-
a
n
e
f
f
ic
ie
nt AC
O
-
ba
se
d a
ppr
oa
c
h,
”
I
nd
one
si
an J
o
ur
na
l of E
le
c
t
ric
al
En
gi
ne
e
r
in
g an
d C
om
pu
te
r Sc
ie
nc
e
,
vo
l.
10,
no.
1
,
pp.
32
0
–
32
9,
20
18,
d
oi
: 1
0.
11
59
1/
ij
e
e
c
s.
v
10.
i1.
p
p3
20
-
3
29.
[
34]
B
.
S
.
S
a
br
ina
,
K.
Ha
m
ou
di,
a
n
d K.
S
a
l
im
,
“
B
i
-
o
bje
c
ti
ve
sc
he
du
li
ng wi
t
h c
o
ope
r
a
ti
ng
he
ur
is
tic
s f
or
e
m
be
dde
d
r
e
a
l
-
t
im
e
sy
ste
m
s,
”
I
nd
one
si
an J
o
ur
na
l of
Ele
c
tr
ic
a
l En
gi
ne
e
r
in
g an
d C
om
pu
te
r
Sc
ie
nc
e
,
vo
l.
9,
no.
3
,
pp.
78
9
–
79
8,
20
18,
d
oi
: 1
0.
11
59
1/
ij
e
e
c
s.
v
9.
i
3.
pp
78
9
-
798
.
[
35]
A.
B
.
I
br
a
him
,
C
.
Z
.
Z
ulkif
li,
S
.
A.
Ar
if
f
in,
a
nd N.
H.
K
a
ha
r
,
“
Hi
gh f
r
e
que
nc
y
of
lo
w n
oi
se
a
m
p
lif
ie
r
a
r
c
h
ite
c
tur
e
f
or
W
iM
AX a
pp
lic
a
ti
on
: A r
e
vie
w,
”
I
n
te
r
na
ti
on
al
J
o
urn
al of Ele
c
tr
ic
a
l
a
nd
C
om
pu
te
r En
gi
ne
e
r
in
g
,
vo
l.
1
1,
no.
3
,
pp.
2
15
3
–
21
64,
2
02
1,
do
i: 10.
1
15
91
/i
je
c
e
.
v
11
i3.
p
p21
53
-
21
64
.
[
36]
B
.
U.
V P
r
a
sha
nt
h,
M
.
R
.
Ahm
e
d,
a
nd M
.
R
.
Kou
nte
,
“
De
s
ig
n a
nd im
ple
m
e
n
ta
t
io
n of
DA F
I
R
f
il
te
r
f
or
bio
-
i
nsp
ir
e
d c
om
pu
ti
ng a
r
c
h
ite
c
tur
e
,
”
I
nte
rn
at
io
na
l J
o
urn
al o
f Ele
c
tr
ic
a
l and C
o
mp
ute
r En
gi
ne
e
r
in
g
,
vol.
1
1,
no.
2
,
pp.
1
70
9
–
17
18,
2
02
1,
do
i: 10.
1
15
91
/i
je
c
e
.
v
11
i2.
p
p17
09
-
17
18
.
[
37]
L
.
K.
R
a
m
a
sa
m
y,
S
.
Ka
dr
y,
a
nd S
.
L
im
,
“
S
e
le
c
tion of
o
ptim
a
l hy
pe
r
-
pa
r
a
m
e
te
r
va
l
ue
s of
s
up
por
t ve
c
t
or
m
a
c
hi
n
e
f
or
se
nt
im
e
n
t a
na
ly
si
s ta
sk
s u
si
ng
na
tur
e
-
in
sp
ir
e
d op
tim
iz
a
t
io
n
m
e
t
ho
ds,
”
Bu
lle
ti
n
o
f
E
le
c
t
ric
al En
gi
ne
e
r
in
g an
d
I
nf
or
mat
ic
s
,
vo
l.
10,
n
o.
1
,
pp.
2
90
–
2
98,
2
02
1,
do
i: 1
0.
1
159
1/
e
e
i.
v
10
i1.
20
98.
[
38]
H.
F
.
S
.
S
a
ipol a
nd
N.
Al
ia
s,
“
Num
e
r
ic
a
l
sim
ula
ti
on
of
DI
C
dr
yi
ng
pr
oc
e
ss
on m
a
tla
b d
is
tr
i
bu
te
d c
om
p
ut
i
ng
s
e
r
v
e
r
,
”
I
nd
one
si
an J
o
ur
na
l o
f Ele
c
tr
ic
a
l En
gi
ne
e
r
in
g a
nd C
o
mp
ute
r Sc
ie
nc
e
,
vol.
2
0,
no.
1
,
pp.
33
8
–
34
6,
20
2
0,
doi
: 1
0.
11
59
1/
ij
e
e
c
s.
v
20.
i1.
p
p3
38
-
34
6.
[
39]
G
.
B
.
P
a
lla
vi a
n
d P
.
Ja
ya
r
e
kha
,
“
A
n e
f
f
ic
ie
n
t r
e
s
our
c
e
s
ha
r
i
ng
te
c
h
ni
que
f
or
m
u
lt
i
-
te
na
n
t da
ta
ba
s
e
s,
”
I
nte
rn
at
io
na
l J
our
na
l of E
le
c
t
ric
al a
nd C
o
mp
ute
r En
gi
ne
e
r
in
g
,
vol.
10,
no.
3
,
pp.
32
16
–
32
26
,
2
02
0,
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
el
eco
m
m
u
n
C
om
put
E
l
C
ont
r
o
l
L
ar
ge
s
c
al
e
dat
a anal
y
s
i
s
us
i
ng M
L
l
i
b
(
A
hm
e
d H
u
s
s
e
i
n A
l
i
)
1743
doi
: 1
0.
11
59
1/
ij
e
c
e
.
v1
0i
3.
p
p3
21
6
-
32
26
.
[
40]
N
. R
.
L
a
tha
a
nd G
.
R
.
P
r
a
sa
d,
“
M
e
m
or
y a
nd I
/
O o
pt
im
iz
e
d r
e
c
t
il
ine
a
r
S
te
i
ne
r
m
i
nim
um
tr
e
e
r
o
ut
in
g f
or
VL
S
I
,
”
I
nte
rn
at
io
na
l J
our
na
l of E
le
c
t
ric
al a
nd C
o
mp
ute
r En
gi
ne
e
r
in
g
,
vol.
10,
no.
3
,
pp.
29
59
–
29
68
,
2
02
0,
doi
: 1
0.
11
59
1/
ij
e
c
e
.
v1
0i
3.
p
p2
95
9
-
29
68
.
[
41]
N.
H.
Ja
’
a
f
a
r
a
nd
A.
Ahm
a
d,
“
Alg
or
i
thm
de
ve
l
op
m
e
nt
a
nd
ha
r
dwa
r
e
im
ple
m
e
nta
ti
on
f
or
m
e
d
ic
a
l im
a
ge
c
om
pr
e
ss
io
n
s
ys
te
m
:
A r
e
v
ie
w,
”
I
n
do
ne
s
ia
n J
o
ur
na
l
o
f
Ele
c
tr
ic
a
l
E
ng
ine
e
ri
ng
an
d C
o
mp
ute
r Sc
ie
nc
e
,
v
o
l
.
1
8
,
no.
3
,
pp.
1
33
1
–
13
41,
2
02
0,
do
i: 10.
1
15
91
/i
je
e
c
s.
v1
8.
i
3.
pp1
33
1
-
13
41.
[
42]
M
.
Alja
r
a
h,
M
.
S
hur
m
a
n,
a
nd S
.
H.
Alna
be
l
si,
“
C
oo
p
e
r
a
ti
ve
hie
r
a
r
c
hic
a
l ba
se
d e
d
ge
-
c
om
pu
ti
ng a
p
pr
oa
c
h f
or
r
e
so
ur
c
e
s a
l
loc
a
ti
on of
d
is
tr
i
bu
te
d m
ob
ile
a
n
d I
oT
a
pp
lic
a
ti
on
s,
”
I
nte
r
na
ti
on
al J
ou
rn
al o
f Ele
c
tr
ic
a
l an
d
C
omp
ut
e
r E
ng
ine
e
ri
ng
,
v
ol.
1
0,
n
o
. 1
,
pp
.
2
96
–
30
7,
20
2
0,
doi
: 1
0.
11
59
1/
ij
e
c
e
.
v1
0i
1.
p
p2
96
-
30
7.
[
43]
N.
R
a
z
a
li,
S
.
I
sm
a
il,
a
nd A.
M
us
ta
p
ha
,
“
M
a
c
hine
le
a
r
ni
ng a
p
pr
oa
c
h f
or
f
lo
od r
i
sk
s pr
e
d
ic
t
io
n,
”
I
A
ES
I
nte
rn
at
io
na
l J
ou
rn
al
of A
rt
if
ic
i
al I
nte
ll
ige
nc
e
,
v
o
l
.
9
,
n
o
.
1
,
pp.
73
–
8
0,
20
20,
d
oi
:
1
0
.
11
59
1/
ij
a
i.
v
9.
i
1.
pp
73
-
80.
[
44]
H.
B
a
ghe
r
i a
nd
A.
A.
S
ha
l
to
ok
i,
“
B
ig
da
ta
: C
ha
l
le
n
ge
s
,
opp
or
t
un
it
ie
s a
n
d c
l
ou
d ba
se
d s
ol
ut
io
ns,
”
I
nt
e
rn
at
io
nal
J
ou
rn
al of E
le
c
t
ric
al a
nd C
o
mp
ute
r En
gi
ne
e
ri
ng
,
vo
l.
5,
no.
2
,
pp.
340
–
3
43,
20
1
5,
doi
: 1
0.
11
59
1/
ij
e
c
e
.
v5
i2.
pp
34
0
-
3
43.
[
45]
N.
Niz
a
m
ud
di
n
a
nd
A.
A
bu
ga
ba
h,
“
B
loc
kc
ha
in
f
or
a
ut
om
ot
iv
e
: A
n
in
si
gh
t t
owa
r
ds
t
he
I
P
F
S
bl
oc
kc
ha
i
n
-
b
as
ed
a
ut
o in
sur
a
nc
e
se
c
tor
,
”
I
n
te
r
na
ti
on
al J
o
ur
na
l o
f El
e
c
tr
ic
a
l an
d C
om
pu
te
r En
gi
ne
e
r
in
g
,
vo
l.
11,
n
o.
3
,
pp.
24
43
–
2
45
6,
20
21,
d
oi
: 1
0.
11
59
1/
ij
e
c
e
.
v
11
i3
.
pp
24
43
-
24
56.
[
46]
R
.
B
a
ndi,
J.
Am
ud
ha
ve
l,
a
n
d R
.
Ka
r
t
hi
k,
“
M
a
c
h
ine
le
a
r
ni
ng
wi
th P
yS
pa
r
k
–
R
e
v
i
e
w
,
”
I
nd
one
si
an
J
o
ur
na
l of
Ele
c
tr
ic
a
l En
gi
ne
e
r
in
g a
nd C
o
mp
ute
r Sc
ie
n
c
e
,
vol.
1
2,
no.
1
,
pp.
1
02
–
1
06,
2
01
8,
doi
: 1
0.
11
59
1/
ij
e
e
c
s.
v
12.
i1.
p
p1
02
-
10
6.
[
47]
M
.
Ar
m
a
nur
R
a
hm
a
n,
J.
Hos
se
n,
A.
S
ul
ta
na
,
A.
A.
M
a
m
un,
a
nd N.
A.
A.
Az
iz
,
“
A sm
a
r
t m
e
th
od f
or
s
pa
r
k us
i
ng
ne
ur
a
l ne
t
wor
k f
or
bi
g da
ta
,
”
I
nte
rn
at
io
na
l J
ou
rn
al o
f Ele
c
tr
ic
a
l and C
o
mp
ute
r En
gi
ne
e
r
in
g
,
vo
l.
11,
no.
3
,
pp.
25
25
–
2
53
4,
20
21,
d
oi
: 1
0.
11
59
1/
ij
e
c
e
.
v
11
i3
.
pp
25
25
-
25
34.
[
48]
S
.
Ha
r
ti
ni,
Z
.
R
u
sta
m
,
G
.
S
.
S
a
r
a
g
ih,
a
nd
M
.
J.
S
.
Va
r
ga
s,
“
E
stim
a
t
in
g pr
oba
bi
li
ty
of
ba
n
ki
ng
c
r
i
se
s
us
in
g r
a
nd
o
m
fo
res
t
,
”
I
AES I
n
te
r
na
ti
on
al J
ou
rn
al o
f Ar
ti
fic
ia
l
I
nte
ll
ige
nc
e
,
vo
l.
10,
no.
2
,
pp.
407
–
4
13,
20
21,
doi
: 1
0.
11
59
1/
I
JA
I
.
V1
0.
I
2.
P
P
4
07
-
41
3.
[
49]
M
. M
. A
k
a
w
e
e
, M
. A
. M
. A
l
-
Oba
id
i,
H.
M
.
T.
A
l
-
H
ilf
i,
S
.
I
.
Ja
ssim
,
a
nd
T.
S
ut
ik
no,
“
An e
f
f
ic
ie
nt
hy
br
i
d m
o
de
l
f
or
se
c
ur
e
tr
a
nsm
is
si
on of
da
ta
by u
si
ng e
f
f
ic
ie
nt da
ta
c
olle
c
ti
on a
n
d di
sse
m
ina
ti
o
n (
E
DC
D)
a
l
gor
it
hm
ba
se
d
W
S
N
,”
I
n
do
ne
s
ia
n J
o
ur
na
l of
Ele
c
tr
ic
a
l E
ng
ine
e
ri
ng
a
nd C
o
mp
ute
r Sc
ie
nc
e
,
v
ol.
20,
n
o.
1
,
pp
.
5
45
–
55
1,
20
2
0,
doi
: 1
0.
11
59
1/
ij
e
e
c
s.
v
20.
i1.
p
p5
45
-
55
1.
[
50]
N.
P
.
S
he
tty,
J.
S
he
t
ty,
R
.
Na
r
u
la
,
a
nd K.
Ta
nd
ona
,
“
C
om
pa
r
is
on s
tu
dy of
m
a
c
h
in
e
le
a
r
n
in
g c
la
s
sif
ie
r
s t
o de
te
c
t
a
nom
a
lie
s,
”
I
nte
rn
at
io
na
l J
o
ur
na
l of
Ele
c
tr
ic
a
l a
nd C
o
mp
ute
r En
gi
ne
e
r
in
g
,
vo
l.
1
0,
no.
5
,
p
p.
54
45
–
54
52
,
20
20
,
doi
: 1
0.
11
59
1/
I
J
EC
E.
V
10I
5.
P
P
54
45
-
54
52.
[
51]
R
.
A.
I
br
a
he
m
A
lha
ya
l
i,
M
.
A.
A
hm
e
d,
Y.
M
.
M
oh
ia
l
de
n,
a
n
d
A.
H.
A
li,
“
Ef
f
ic
ie
n
t m
e
th
od f
or
br
e
a
st
c
a
nc
e
r
c
la
s
sif
ic
a
t
io
n ba
se
d
on
e
n
se
m
b
le
h
of
f
e
d
in
g
tr
e
e
a
n
d
na
ï
ve
B
a
ye
s,
”
I
n
do
ne
s
ia
n
J
o
ur
na
l of Ele
c
tr
ic
a
l En
gi
ne
e
r
in
g
and C
om
pu
te
r S
c
ie
nc
e
,
vo
l.
18,
n
o.
2
,
pp.
1
07
4
–
10
80,
2
020,
d
oi
:
1
0.
11
59
1/
ij
e
e
c
s.
v
18.
i2.
p
p1
07
4
-
10
80
.
[
52]
A
.
V Vi
tia
ni
ng
si
h,
N.
S
ur
ya
na
,
a
nd Z
.
Ot
hm
a
n,
“
S
pa
t
ia
l
a
na
ly
si
s
m
o
de
l f
or
tr
a
f
f
ic
a
c
c
ide
nt
-
pr
o
ne
r
oa
d
s
c
la
s
sif
ic
a
t
io
n:
A pr
op
ose
d f
r
a
m
e
w
or
k,
”
I
A
ES
I
nte
rn
at
ion
al J
o
urn
al of Ar
ti
fic
ia
l
I
n
te
l
li
ge
nc
e
,
vo
l.
1
0,
no.
2
,
pp.
36
5
–
37
3,
20
21,
d
oi
: 1
0.
11
59
1/
ij
a
i.
v
10.
i2.
pp
36
5
-
37
3.
[
53]
K.
A.
S
ha
ki
l,
M
.
Ala
m
,
a
nd S
.
Kha
n,
“
A la
te
nc
y
-
a
w
a
r
e
m
a
x
-
m
in a
l
gor
it
hm
f
or
r
e
s
our
c
e
a
l
loc
a
ti
on i
n c
l
ou
d,
”
I
nte
rn
at
io
na
l
J
ou
rn
al
of
Ele
c
tr
ic
a
l
a
nd C
om
pu
te
r
Eng
ine
e
ri
ng
,
vo
l.
11,
no.
1
,
pp
.
67
1
–
6
85,
2
02
1,
doi
:
1
0.
11
59
1/
ij
e
c
e
.
v1
1i
1.
p
p6
71
-
68
5.
[
54]
N
.
S
e
m
a
n
a
n
d
N
.
A
.
R
a
z
m
i
,
“
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
-
ba
se
d te
c
h
ni
que
f
or
b
ig
da
ta
se
n
tim
e
nt
s e
xt
r
a
c
t
io
n,
”
I
A
ES
I
nte
rn
at
io
na
l J
ou
rn
al o
f Ar
ti
fic
ia
l I
nte
ll
ige
nc
e
,
vo
l.
9,
no.
3
,
pp.
47
3
–
47
9,
20
20,
do
i:
10.
1
15
91
/i
ja
i.
v9.
i
3.
p
p4
7
3
-
47
9.
[
55]
B
.
C
.
M
a
nuja
k
sh
i a
n
d K.
B
.
R
a
m
e
sh,
“
F
r
a
m
e
w
or
k f
or
c
os
t
-
e
f
f
e
c
t
ive
a
na
l
yt
ic
a
l m
o
de
lli
ng f
or
se
ns
or
y
da
ta
ove
r
c
lo
ud e
nv
ir
o
nm
e
n
t,
”
I
nte
rn
at
io
na
l
J
our
na
l
of
E
le
c
t
ric
al
and
C
o
mp
ute
r
En
gi
ne
e
r
in
g
,
vol.
9,
no.
5
,
pp.
382
2
–
3
832
,
201
9,
do
i: 10.
1
15
91
/i
je
c
e
.
v
9
i5
.
pp
38
22
-
38
32
.
[
56]
H.
As
hr
a
f
,
M
.
Ale
ne
z
i,
M
.
Na
de
e
m
,
a
n
d Y.
Ja
vi
d,
“
S
e
c
ur
it
y a
s
se
s
sm
e
n
t f
r
a
m
e
w
or
k f
or
e
duc
a
ti
on
a
l E
R
P
sys
te
m
s,
”
I
n
te
r
na
ti
on
al J
o
ur
na
l of Ele
c
tr
ic
a
l an
d C
om
put
e
r En
gi
ne
e
r
in
g
,
vo
l.
9,
no.
6
,
pp.
5570
–
55
85
,
2
0
19,
doi
: 1
0.
11
59
1/
ij
e
c
e
.
v9
i6.
pp
55
70
-
55
85.
[
57]
A.
B
ou
kha
lf
a
,
N.
Hm
i
na
,
a
nd H.
C
ha
ou
i,
“
P
a
r
a
l
le
l pr
oc
e
ssi
ng
us
in
g b
ig da
ta
a
n
d m
a
c
hi
ne
le
a
r
ni
ng
te
c
h
ni
que
s f
or
int
r
us
io
n de
te
c
ti
on,
”
I
A
ES I
nte
rn
at
io
na
l J
ou
rn
al o
f Arti
fi
c
i
al I
nte
ll
ige
nc
e
,
vo
l.
9,
no.
3,
pp.
553
–
5
60,
2
02
0,
doi
: 1
0.
11
59
1/
ij
a
i.
v
9.
i
3.
pp
55
3
-
5
60.
[
58]
U.
Na
r
a
ya
na
n,
V.
P
a
ul,
a
nd S
.
J
ose
ph,
“
A n
ove
l a
p
pr
oa
c
h t
o b
ig
da
ta
a
na
l
ys
is
us
in
g de
e
p be
lie
f
ne
t
wor
k f
or
th
e
de
te
c
ti
on
of
a
n
dr
o
id m
a
lwa
r
e
,
”
I
nd
one
si
an J
ou
rn
al
of
Ele
c
tr
ic
a
l E
ng
ine
e
ri
ng
an
d C
o
mp
ute
r Sc
ie
nc
e
,
vo
l.
1
6,
no.
3
,
pp.
1
44
7
–
14
54,
2
01
9,
do
i: 10.
1
15
91
/i
je
e
c
s.
v1
6.
i
3.
pp1
44
7
-
14
54.
[
59]
N.
H.
M
.
Ar
if
f
in,
F
.
Ahm
a
d,
a
nd U.
M
.
Ha
ne
e
f
,
“
Ac
c
e
pta
nc
e
of
m
ob
ile
pa
ym
e
nt
s by r
e
ta
i
le
r
s u
si
ng U
TA
U
T
m
ode
l,
”
I
nd
one
si
an J
o
ur
na
l of
Ele
c
tr
ic
a
l En
gi
ne
e
ri
ng a
nd C
o
mp
ute
r Sc
ie
nc
e
,
vo
l.
19,
n
o
.
1
,
pp.
1
49
–
15
5
,
20
20,
doi
: 1
0.
11
59
1/
ij
e
e
c
s.
v
19.
i1.
p
p1
49
-
15
5.
[
60]
J.
I
.
N
a
se
r
a
nd
A
.
J
.
Ka
dhi
m
,
“
M
ulti
c
a
st
r
outing
st
r
a
t
e
gy
f
or
S
DN
-
c
lust
e
r
ba
s
e
d
M
ANET
,
”
I
nt
e
rnational
J
ournal
of
Ele
c
tric
al
and
C
omput
e
r
Engine
e
ring
,
vol
.
10,
no
.
5
,
pp
.
4447
–
4457,
2020
,
d
oi:
10.
11591/ij
e
c
e
.
v10i5.
pp4447
-
4457
.
[
61]
K.
N
a
ga
r
a
thn
a
,
“
Ene
r
gy
-
a
wa
r
e
str
a
te
gy
f
or
da
t
a
f
o
r
wa
r
di
ng in
I
oT
e
c
osyste
m
,
”
I
nt
e
rnational
J
ournal
of El
e
c
tri
c
a
l
and C
omput
e
r Engin
e
e
ring
,
vol
.
10
,
no
.
5
,
pp
.
4863
–
4871
,
2020
,
doi:
10.
11591/i
je
c
e
.
v10i5.
pp4863
-
4871
.
[
62]
A.
B
ha
w
iy
uga
,
S
.
A.
K
ha
r
i
sm
a
,
B
.
J.
S
a
nto
so,
D.
P
.
Ka
r
ti
ka
sa
r
i,
a
nd
A.
P
.
Kir
a
na
,
“
C
lo
ud
-
ba
se
d m
id
dle
wa
r
e
f
or
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
1693
-
6930
T
E
L
KOM
NI
KA
T
el
eco
m
m
u
n
C
om
put
E
l
C
ont
r
o
l
,
Vo
l
.
19
, N
o
.
5
,
O
c
t
obe
r
2021:
1735
-
1746
1744
sup
po
r
ti
ng
ba
tc
h a
nd
str
e
a
m
a
c
c
e
s
s o
ve
r
sm
a
r
t he
a
l
thc
a
r
e
we
a
r
a
ble
de
v
ic
e
,
”
Bu
lle
ti
n
of E
le
c
t
ric
al
En
gi
ne
e
r
in
g
and I
nf
or
ma
tic
s
,
vo
l.
9,
no.
5
,
p
p.
19
90
–
1
99
7,
20
20,
d
oi
:
10.
11
59
1/
e
e
i.
v
9i
5.
1
97
8.
[
63]
C
.
C
.
Uc
he
nna
,
N.
Ja
m
i
l,
R
.
I
sm
a
il,
L
.
K.
Ya
n,
a
nd M
.
A.
M
oha
m
e
d,
“
M
a
lwa
r
e
t
hr
e
a
t a
na
l
ys
is te
c
hn
iq
ue
s a
n
d
a
ppr
oa
c
he
s f
or
i
ot a
pp
lic
a
ti
on
s: A r
e
vie
w,
”
Bu
lle
ti
n o
f
Ele
c
t
ric
al
En
gi
ne
e
r
in
g an
d I
n
fo
rm
at
ic
s
,
v
ol.
1
0,
no.
3
,
pp.
15
58
–
1
57
1,
20
21,
d
oi
: 1
0.
11
59
1/
e
e
i.
v
10
i3.
24
23.
[
64]
N.
S
.
M
.
P
a
khr
ud
in,
M
.
Ka
ss
im
,
a
n
d
A.
I
dr
i
s,
“
A
r
e
vie
w o
n or
c
he
s
tr
a
t
io
n
di
str
ib
ute
d s
ys
te
m
s f
or
I
oT
sm
a
r
t
se
r
v
ic
e
s
in f
og
c
om
pu
ti
ng,
”
I
nte
rn
at
io
na
l J
o
ur
na
l o
f
Ele
c
tr
ic
a
l a
nd
C
o
mp
ute
r E
ng
ine
e
ri
ng
,
vo
l.
1
1,
n
o.
2
,
pp.
18
12
–
1
82
2,
20
21,
d
oi
: 1
0.
11
59
1/
ij
e
c
e
.
v1
1i
2.
p
p1
81
2
-
18
22.
[
65]
S
.
-
T
.
N
a
m
,
C
.
-
Y.
Jin,
a
nd S
.
-
Y.
S
hin,
“
A f
or
e
c
a
s
ti
ng of
sto
c
k tr
a
di
ng pr
ic
e
us
in
g tim
e
se
r
ie
s i
nf
or
m
a
ti
on ba
se
d o
n
big da
ta
,
”
I
nte
rn
at
io
na
l J
ou
rn
al of E
le
c
t
ric
al a
nd C
o
m
put
e
r En
gi
ne
e
r
in
g
,
vo
l.
11,
no.
3
,
pp.
254
8
–
25
54,
20
2
1,
doi
: 1
0.
11
59
1/
ij
e
c
e
.
v1
1i
3.
p
p2
54
8
-
25
54
.
[
66]
Y.
I
.
Alz
ou
bi,
A.
A
l
-
A
hm
a
d,
a
nd
A.
Ja
r
a
da
t,
“
F
og
c
om
put
in
g
se
c
ur
it
y a
n
d
pr
i
va
c
y
is
sue
s,
o
pe
n
c
ha
lle
nge
s,
a
n
d
blo
c
kc
ha
i
n so
lu
ti
on
: An o
ve
r
v
ie
w,
”
I
n
te
r
na
ti
on
al
J
o
ur
n
al
o
f
Ele
c
tr
ic
a
l
an
d
C
om
pu
te
r
E
ng
ine
e
ri
ng
,
v
ol.
11,
n
o.
6
,
pp.
50
81
–
5
08
8,
20
21,
d
oi
: 1
0.
11
59
1/
ij
e
c
e
.
v1
1i
6.
p
p5
08
1
-
50
88.
[
67]
K.
R
a
dha
,
B
.
T.
R
a
o,
S
.
M
.
B
a
bu,
K.
T.
R
a
o,
V.
K.
R
e
ddy,
a
n
d P
.
S
a
ik
ir
a
n,
“
S
e
r
v
ic
e
le
ve
l a
gr
e
e
m
e
nt
s in c
lo
u
d
c
om
p
ut
in
g a
nd b
ig da
ta
,
”
I
nte
rn
at
io
na
l J
ou
rn
al o
f Ele
c
tr
ic
a
l a
n
d C
om
pu
te
r E
ng
ine
e
ri
ng
,
v
ol.
5,
no.
1.
pp.
15
8
–
16
5,
20
15,
d
oi
: 1
0.
11
59
1/
ij
e
c
e
.
v5
i1.
pp
15
8
-
1
65.
[
68]
A.
H.
Ka
tr
a
wi,
R
.
A
bd
ul
la
h,
M
.
An
ba
r
,
a
n
d A.
K.
A
ba
s
i
,
“
Ea
r
lie
r
sta
ge
f
or
s
tr
a
g
gle
r
de
te
c
ti
on a
nd
ha
n
dl
in
g u
si
n
g
c
om
b
ine
d C
P
U te
s
t a
n
d L
ATE m
e
t
ho
do
lo
gy,
”
I
n
te
r
na
tio
na
l J
o
ur
na
l of
Ele
c
tr
ic
a
l an
d C
om
pu
te
r
En
gi
ne
e
r
in
g
,
vol.
1
0,
no.
5
,
p
p.
49
10
–
4
91
7,
20
20,
d
oi
: 1
0.
11
59
1/
ij
e
c
e
.
v10
i5.
p
p4
91
0
-
49
17.
[
69]
J.
S
a
nto
sh K
um
a
r
,
B
.
K.
R
a
gha
ve
ndr
a
,
S
.
R
a
gha
ve
ndr
a
,
a
nd M
e
e
na
ks
hi,
“
P
e
r
f
or
m
a
nc
e
e
va
l
ua
t
io
n of
M
a
p
-
r
e
d
u
c
e
ja
r
pi
g
hi
ve
a
nd s
pa
r
k w
it
h m
a
c
hi
ne
le
a
r
n
in
g us
in
g bi
g
da
ta
,
”
I
nte
r
na
ti
on
al J
o
ur
na
l of E
le
c
t
ric
al a
nd C
o
mp
ut
e
r
E
ng
in
e
e
ri
ng
,
vo
l.
10,
n
o.
4
,
pp.
3
81
1
–
38
18,
2
02
0,
do
i:
1
0.
11
59
1/
ije
c
e
.
v1
0i
4.
pp
38
11
-
38
18.
[
70]
M
.
B
.
M
a
sa
de
h,
M
.
S
.
Az
m
i,
a
nd S
.
S
.
S
.
A
hm
a
d,
“
Ava
ila
bl
e
t
e
c
hn
iq
ue
s
in
ha
do
op sm
a
l
l
f
i
le
is
sue
,
”
I
nte
rn
at
io
na
l J
our
na
l of E
le
c
t
ric
al a
nd C
o
mp
ute
r En
gi
ne
e
r
in
g
,
vol.
10,
no.
2
,
pp.
20
97
–
21
01
,
2
02
0,
doi
: 1
0.
11
59
1/
ij
e
c
e
.
v1
0i
2.
p
p2
09
7
-
21
01
.
[
71]
A
.
B
h
as
k
ar an
d
R
.
R
an
j
an
,
“O
p
t
i
m
i
zed
me
mo
ry
m
o
d
el
fo
r
h
ad
o
o
p
m
ap
r
ed
u
c
e f
r
am
ew
o
rk
,
”
I
nt
e
rnational J
ournal
of
Ele
c
tric
al
and
C
omput
e
r
Engine
e
ring
,
vol
.
9,
no
.
5
,
pp
.
4
396
–
4407,
2019
,
doi:
10.
11591/ij
e
c
e
.
v9i5.
pp4396
-
4407
.
[
72]
U.
Na
r
a
ya
na
n,
V.
P
a
ul,
a
nd S
.
Jose
ph,
“
A
l
ig
ht we
ig
h
t e
nc
r
y
pt
io
n ove
r
b
ig da
ta
in
i
nf
or
m
a
ti
on s
toc
kp
il
in
g o
n
c
lo
ud,
”
I
n
do
ne
s
ia
n J
o
ur
na
l of
Ele
c
tr
ic
a
l E
ng
ine
e
ri
ng
a
nd C
o
mp
ute
r Sc
ie
nc
e
,
v
ol.
17,
n
o.
1
,
pp
.
3
89
–
39
7,
20
1
9,
doi
: 1
0.
11
59
1/
ij
e
e
c
s.
v
17.
i1.
p
p3
89
-
3
97
.
[
73]
J.
P
a
r
sola
,
D.
G
a
ng
od
ka
r
,
a
nd
A.
M
i
tta
l,
“
P
ost
e
ve
n
t
i
nve
st
iga
ti
on
of
m
u
lt
i
-
s
tr
e
a
m
vi
de
o
da
ta
u
ti
liz
in
g
ha
d
oo
p
c
l
u
s
t
e
r
,
”
I
n
te
r
na
ti
on
al J
o
ur
na
l o
f Ele
c
tr
ic
al a
nd C
om
put
e
r E
ng
ine
e
ri
ng
,
v
ol.
8,
n
o.
6
,
pp.
5
08
9
–
50
97,
2
01
8,
doi
: 1
0.
11
59
1/
ij
e
c
e
.
v8
i6.
pp.
5
0
89
-
50
97
.
[
74]
D.
C
.
Vi
nu
tha
a
nd G
.
T.
R
a
ju,
“
An a
c
c
ur
a
te
a
nd
e
f
f
ic
ie
nt
sc
he
du
le
r
f
or
ha
do
op m
a
pr
e
d
uc
e
f
r
a
m
e
w
or
k,
”
I
nd
one
si
an J
o
ur
na
l of
Ele
c
tr
ic
a
l En
gi
ne
e
ri
ng a
nd C
o
mp
ute
r Sc
ie
nc
e
,
vo
l.
12,
n
o.
3
,
pp.
1
13
2
–
1
14
2
,
20
18,
doi
: 1
0.
11
59
1/
ij
e
e
c
s.
v
12.
i3.
p
p1
13
2
-
1
14
2.
[
75]
M
.
I
sa
r
d a
n
d
Y.
Y
u,
“
D
is
tr
i
bu
te
d
da
ta
-
pa
r
a
l
le
l c
om
put
in
g u
si
ng
a
hi
gh
-
le
ve
l pr
ogr
a
m
m
in
g la
ng
ua
ge
,
”
i
n
Pro
c
e
e
d
in
gs o
f t
he
20
09
AC
M SI
GM
OD I
nte
rn
at
io
na
l C
onf
e
re
nc
e
on
Ma
na
ge
me
nt o
f d
at
a
,
20
09,
p
p.
98
7
–
99
4.
[
76]
C.
Ch
amb
ers
e
t a
l.
,
“
F
lum
e
Ja
va
: e
a
sy,
e
f
f
ic
ie
n
t da
ta
-
pa
r
a
l
le
l p
ipe
li
ne
s,
”
AC
M S
ig
pl
an N
o
t.
,
vol.
4
5,
no.
6
,
pp.
36
3
–
37
5,
20
10.
[
77]
J.
F
a
r
oo
q,
P
.
S
ha
r
m
a
,
a
nd R
.
S
.
Kum
a
r
,
“
A B
I
M
-
ba
s
e
d de
ta
ile
d e
le
c
tr
ic
a
l l
oa
d e
st
im
a
t
io
n,
c
o
st
in
g a
nd c
od
e
c
h
e
c
k
i
n
g
,
”
I
nte
rn
at
io
na
l
J
o
urn
al
of
Ele
c
tr
ic
a
l an
d
C
o
m
p
ute
r En
gi
ne
e
r
in
g
,
v
o
l
.
8
,
n
o
.
5
,
pp.
3
48
4
–
34
95,
2
01
8,
doi
: 1
0.
11
59
1/
ij
e
c
e
.
v8
i5.
pp
34
84
-
34
95.
[
78]
C.
-
Y.
S
ong a
n
d E.
-
S
.
C
ho,
“
A se
r
vic
e
-
or
ie
n
te
d c
l
ou
d
m
ode
li
ng m
e
t
ho
d
a
nd
pr
oc
e
s
s,
”
I
nte
rn
at
io
na
l
J
ou
rn
al of
Ele
c
tr
ic
a
l an
d C
o
mp
ute
r En
gi
ne
e
r
in
g
,
vo
l.
10,
n
o.
1
.
pp.
962
–
9
77,
2
02
0,
do
i: 1
0.
1
15
91
/i
je
c
e
.
v
10
i1.
p
p9
62
-
97
7.
[
79]
R
.
Kr
a
le
va
,
M
.
S
a
ba
n
i,
V.
Kr
a
le
v,
a
n
d
D.
K
os
ta
d
in
o
va
,
“
A
n a
p
pr
oa
c
h
to
de
s
ig
ni
ng a
nd
de
ve
l
op
in
g a
n
L
M
S
f
r
a
m
e
wor
k a
ppr
opr
ia
te
f
or
yo
un
g pu
pi
ls,
”
I
n
te
r
na
ti
on
al J
ou
rn
al of E
le
c
t
ric
al a
nd C
o
mp
ute
r E
ng
in
e
e
ri
ng
,
vol.
1
0,
no.
2
,
pp.
1
57
7
–
15
91,
2
02
0,
do
i: 10.
1
15
91
/i
je
c
e
.
v
10
i2.
p
p15
77
-
15
91
.
[
80]
S
.
I
sla
m
,
Y.
S
.
Nugr
oh
o,
a
nd M
.
Ja
ve
d Ho
ssa
in,
“
W
ha
t
ne
twor
k sim
ula
tor
q
ue
s
ti
on
s do use
r
s a
sk?
A la
r
ge
-
s
c
a
l
e
stu
dy of
s
ta
c
k o
ve
r
f
l
ow p
os
ts,
”
I
n
do
ne
s
ia
n J
ou
rn
al o
f
Ele
c
t
ric
al En
gi
ne
e
ri
ng an
d C
om
pu
te
r Sc
ie
nc
e
,
vol.
2
1,
no.
3
,
pp.
1
62
2
–
16
33,
2
02
1,
do
i: 10.
1
15
91
/i
je
e
c
s.
v2
1.
i
3.
pp1
62
2
-
16
33.
[
81]
F
.
R
e
na
ldi,
I
.
S
a
ntika
r
a
m
a
,
E.
C
.
Dja
m
a
l,
a
nd A.
J.
M
a
ul
id
in,
“
S
e
r
vic
e
la
n
dsc
a
pe
f
or
pr
i
va
te
un
ive
r
si
tie
s in
I
nd
one
sia
ba
se
d on se
r
v
ic
e
or
ie
nte
d
a
r
c
hit
e
c
t
ur
e
a
n
d
c
lo
ud
te
c
hn
ol
og
y,
”
I
nd
on
e
s
ia
n
J
o
ur
na
l of Ele
c
tr
ic
a
l
En
gi
ne
e
r
in
g an
d C
o
mp
ute
r Sc
ie
nc
e
,
vo
l.
22,
n
o.
1
,
pp.
4
97
–
50
6,
20
21,
d
oi
:
1
0.
11
59
1/
ij
e
e
c
s.
v
22.
i1.
p
p4
97
-
50
6.
[
82]
G
.
M
a
l
e
w
i
c
z
e
t al.
,
“
P
r
e
ge
l
: a
sys
te
m
f
or
la
r
ge
-
sc
a
le
gr
a
ph pr
oc
e
s
si
ng,
”
in
Pr
oc
e
e
d
in
gs o
f the
2
01
0 AC
M
SI
G
MO
D I
nte
rn
at
io
na
l C
o
nfe
re
nc
e
o
n M
an
age
me
n
t of
d
ata
,
2
01
0,
pp.
1
35
–
1
46.
[
83]
J.
Eka
na
y
a
ke
e
t a
l.
,
“
T
wi
ste
r
: a
r
un
tim
e
f
or
ite
r
a
ti
ve
m
a
pr
e
d
uc
e
,
”
i
n
P
roc
e
e
di
ng
s of the
19
th
AC
M
i
nte
rn
at
io
n
al
sy
m
po
si
um o
n h
ig
h pe
r
fo
rm
anc
e
di
st
ri
bu
te
d c
o
mp
ut
in
g
,
201
0,
pp.
8
10
–
8
18.
[
84]
Y
.
B
u,
B
.
Howe
,
M
.
B
a
la
z
in
ska
,
a
n
d M
.
D.
Er
ns
t,
“
Ha
L
oop
: Ef
f
ic
ie
nt
ite
r
a
ti
ve
da
ta
pr
oc
e
s
si
ng
on
la
r
ge
c
lu
ste
r
s,
”
Pro
c
.
VL
D
B En
dow
.
,
v
ol.
3,
n
o.
1
–
2,
pp.
2
85
–
29
6
,
20
10.
[
85]
D.
L
unga
,
J.
G
e
r
r
a
nd,
L
.
Ya
n
g,
C
.
L
a
yto
n,
a
nd R
.
S
te
wa
r
t,
“
Apa
c
he
s
pa
r
k a
c
c
e
le
r
a
te
d
de
e
p le
a
r
ni
ng i
nf
e
r
e
nc
e
f
o
r
la
r
ge
sc
a
le
sa
te
l
li
te
im
a
ge
a
na
l
yt
ic
s,
”
I
EE
E J
.
Se
l.
T
o
p
.
App
l.
E
ar
th
O
bs.
Re
m
ote
Se
n
s.
,
vo
l.
1
3,
p
p.
2
71
–
283
,
202
0
,
do
i: 10.
1
10
9/
JS
T
AR
S
.
2
01
9.
29
59
70
7
.
[
86]
P
.
G
a
ne
sh,
K.
S
a
ila
ja
Kum
a
r
,
D.
Eva
n
ge
l
in G
e
e
tha
,
a
n
d T.
V S
ur
e
sh Kum
a
r
,
“
P
e
r
f
o
r
m
a
n
c
e
e
v
a
l
u
a
t
i
o
n
o
f
c
l
o
u
d
Evaluation Warning : The document was created with Spire.PDF for Python.