I
nte
rna
t
io
na
l J
o
urna
l o
f
I
nfo
rm
a
t
ics a
nd
Co
m
m
un
ica
t
io
n T
ec
hn
o
lo
g
y
(
I
J
-
I
CT
)
Vo
l.
1
4
,
No
.
1
,
A
p
r
il
20
2
5
,
p
p
.
50
~
58
I
SS
N:
2252
-
8
7
7
6
,
DOI
:
1
0
.
1
1
5
9
1
/iji
ct
.
v
1
4
i
1
.
pp
50
-
58
50
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ict.
ia
esco
r
e.
co
m
A
hy
brid ma
chi
n
e learning
appro
a
ch f
o
r im
pro
v
ed
po
nzi
schem
e de
tec
tion
using
adv
a
nced f
e
a
ture
eng
ine
ering
F
a
ha
d H
o
s
s
a
in
1
,
M
ehed
i H
a
s
a
n Shu
v
o
2
,
J
ia
Uddi
n
3
1
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
a
n
d
I
n
f
o
r
m
a
t
i
o
n
S
c
i
e
n
c
e
,
F
l
o
r
i
d
a
I
n
t
e
r
n
a
t
i
o
n
a
l
U
n
i
v
e
r
s
i
t
y
,
M
i
a
m
i
,
U
S
A
2
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
r
i
n
g
,
D
h
a
k
a
U
n
i
v
e
r
s
i
t
y
o
f
E
n
g
i
n
e
e
r
i
n
g
a
n
d
Te
c
h
n
o
l
o
g
y
(
D
U
ET),
G
a
z
i
p
u
r
,
B
a
n
g
l
a
d
e
s
h
3
D
e
p
a
r
t
me
n
t
o
f
A
I
a
n
d
B
i
g
D
a
t
a
,
En
d
i
c
o
t
t
C
o
l
l
e
g
e
,
W
o
o
s
o
n
g
U
n
i
v
e
r
s
i
t
y
,
D
a
e
j
e
o
n
,
S
o
u
t
h
K
o
r
e
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Sep
6
,
2
0
2
4
R
ev
is
ed
Oct
1
8
,
2
0
2
4
Acc
ep
ted
No
v
1
9
,
2
0
2
4
P
o
n
z
i
sc
h
e
m
e
s
d
e
c
e
iv
e
i
n
v
e
sto
rs
with
p
ro
m
ise
s
o
f
h
i
g
h
re
tu
r
n
s,
r
e
ly
in
g
o
n
fu
n
d
s
fro
m
n
e
w
in
v
e
sto
rs
t
o
p
a
y
e
a
rli
e
r
o
n
e
s,
c
re
a
ti
n
g
a
m
islea
d
in
g
a
p
p
e
a
ra
n
c
e
o
f
p
ro
f
it
a
b
il
i
ty
.
Th
e
se
sc
h
e
m
e
s
a
re
in
h
e
re
n
tl
y
u
n
su
sta
in
a
b
le,
c
o
ll
a
p
sin
g
wh
e
n
n
e
w
in
v
e
stm
e
n
ts
wa
n
e
,
lea
d
in
g
t
o
sig
n
ifi
c
a
n
t
fin
a
n
c
ial
lo
ss
e
s.
M
a
n
y
re
se
a
rc
h
e
rs
h
a
v
e
fo
c
u
se
d
o
n
d
e
tec
ti
n
g
su
c
h
sc
h
e
m
e
s,
b
u
t
c
h
a
ll
e
n
g
e
s
re
m
a
in
d
u
e
t
o
th
e
ir
e
v
o
lv
i
n
g
n
a
tu
re
.
Th
is
stu
d
y
p
ro
p
o
s
e
s
a
n
o
v
e
l
h
y
b
rid
m
a
c
h
i
n
e
-
lea
rn
in
g
a
p
p
ro
a
c
h
to
e
n
h
a
n
c
e
P
o
n
z
i
sc
h
e
m
e
d
e
tec
ti
o
n
.
In
it
iall
y
,
we
train
a
n
XG
Bo
o
st
c
l
a
ss
ifi
e
r
a
n
d
e
x
trac
t
it
s
fe
a
t
u
re
s.
M
e
a
n
wh
il
e
,
we
to
k
e
n
ize
o
p
c
o
d
e
se
q
u
e
n
c
e
s,
tr
a
in
a
g
a
ted
re
c
u
rre
n
t
u
n
i
t
(G
RU)
m
o
d
e
l
o
n
th
e
se
se
q
u
e
n
c
e
s,
a
n
d
e
x
trac
t
fe
a
t
u
re
s
fro
m
th
e
G
RU.
By
c
o
n
c
a
ten
a
ti
n
g
th
e
fe
a
tu
re
s
fro
m
t
h
e
XG
Bo
o
st
c
las
sifier
a
n
d
t
h
e
G
RU,
w
e
train
a
fin
a
l
X
G
Bo
o
st
m
o
d
e
l
o
n
t
h
is
c
o
m
b
i
n
e
d
fe
a
tu
re
se
t.
Ou
r
m
e
th
o
d
o
lo
g
y
,
lev
e
ra
g
in
g
a
d
v
a
n
c
e
d
fe
a
tu
re
e
n
g
in
e
e
rin
g
a
n
d
h
y
b
ri
d
m
o
d
e
li
n
g
,
a
c
h
iev
e
s
a
d
e
tec
ti
o
n
a
c
c
u
ra
c
y
o
f
9
6
.
5
7
%
.
T
h
is
a
p
p
ro
a
c
h
d
e
m
o
n
st
ra
tes
th
e
e
ffica
c
y
o
f
c
o
m
b
i
n
in
g
XG
Bo
o
st
a
n
d
G
RU
m
o
d
e
ls,
a
lo
n
g
wit
h
so
p
h
isti
c
a
ted
fe
a
tu
re
e
n
g
i
n
e
e
rin
g
,
in
id
e
n
ti
f
y
in
g
fra
u
d
u
len
t
a
c
ti
v
it
ies
in
Et
h
e
re
u
m
sm
a
rt
c
o
n
trac
ts.
Th
e
re
su
lt
s
h
ig
h
li
g
h
t
th
e
p
o
ten
ti
a
l
o
f
t
h
is h
y
b
rid
m
o
d
e
l
to
o
ffe
r
m
o
re
ro
b
u
st
a
n
d
a
c
c
u
ra
te
P
o
n
z
i
sc
h
e
m
e
d
e
tec
ti
o
n
,
a
d
d
re
ss
i
n
g
t
h
e
li
m
it
a
ti
o
n
s
o
f
p
re
v
io
u
s m
e
th
o
d
s
.
K
ey
w
o
r
d
s
:
C
r
y
p
to
cu
r
r
e
n
cy
f
r
au
d
E
th
er
eu
m
s
m
ar
t c
o
n
tr
ac
ts
Featu
r
e
e
n
g
in
ee
r
i
n
g
Op
co
d
e
t
o
k
en
izatio
n
Po
n
zi
s
ch
em
e
d
etec
tio
n
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
J
ia
Ud
d
in
Dep
ar
tm
en
t o
f
AI
an
d
B
ig
Data
,
E
n
d
ico
tt C
o
lleg
e,
W
o
o
s
o
n
g
Un
iv
er
s
ity
Dae
jeo
n
,
So
u
th
Ko
r
ea
E
m
ail: jia.
u
d
d
in
@
wsu
.
ac
.
k
r
1.
I
NT
RO
D
UCT
I
O
N
A
Po
n
zi
s
ch
em
e
is
a
ty
p
e
o
f
f
r
au
d
w
h
er
e
m
o
n
e
y
f
r
o
m
n
ew
i
n
v
esto
r
s
is
u
s
ed
t
o
p
ay
r
etu
r
n
s
to
ea
r
lier
in
v
esto
r
s
,
cr
ea
tin
g
an
illu
s
io
n
o
f
a
s
u
cc
ess
f
u
l
in
v
estme
n
t.
Ho
wev
er
,
th
er
e
is
n
o
ac
tu
al
p
r
o
f
it
b
ein
g
g
e
n
er
ated
;
th
e
s
ch
em
e
d
ep
en
d
s
o
n
co
n
tin
u
o
u
s
ly
r
ec
r
u
itin
g
n
ew
in
v
es
to
r
s
to
s
u
s
tain
it
s
o
p
er
atio
n
s
.
W
h
en
it
b
ec
o
m
es
im
p
o
s
s
ib
le
to
attr
ac
t
n
ew
in
v
esto
r
s
o
r
wh
en
to
o
m
an
y
p
ar
ticip
an
ts
attem
p
t
to
with
d
r
aw
th
eir
m
o
n
ey
s
im
u
ltan
eo
u
s
ly
,
th
e
s
ch
em
e
in
ev
itab
ly
co
llap
s
es,
leav
in
g
m
a
n
y
in
v
esto
r
s
with
s
ig
n
if
ican
t
f
i
n
an
cial
lo
s
s
es.
Po
n
zi
s
ch
em
es
d
ec
eiv
e
in
d
i
v
id
u
als
b
y
p
r
o
m
is
in
g
h
ig
h
r
etu
r
n
s
with
m
in
im
al
r
is
k
,
s
im
ilar
t
o
p
y
r
am
i
d
s
ch
em
es
wh
er
e
f
u
n
d
s
f
r
o
m
n
ew
in
v
esto
r
s
ar
e
u
s
ed
to
p
ay
ea
r
lier
p
ar
ticip
an
ts
[
1
]
.
Fig
u
r
e
1
illu
s
tr
ates
th
e
m
ec
h
an
ics
o
f
Po
n
zi
s
ch
em
es,
s
h
o
w
in
g
h
o
w
th
ese
f
r
au
d
u
le
n
t
o
p
er
atio
n
s
r
ely
o
n
a
co
n
s
t
an
t
in
f
lu
x
o
f
n
ew
in
v
estme
n
ts
to
m
ain
tain
th
e
f
a
ca
d
e
o
f
p
r
o
f
itab
ilit
y
.
On
ce
t
h
e
f
lo
w
o
f
n
ew
in
v
esto
r
s
ce
ases
an
d
f
u
n
d
s
b
ec
o
m
e
in
s
u
f
f
icien
t,
th
e
s
ch
em
e
u
n
r
a
v
els.
C
r
itics
lik
e
R
o
u
b
in
i
an
d
Qu
in
n
ar
g
u
e
th
at
cr
y
p
to
cu
r
r
en
cies
s
u
ch
as
B
itco
in
ex
h
ib
it
s
im
ilar
ch
ar
ac
ter
is
tics
to
Po
n
zi
s
ch
em
es,
with
ea
r
ly
in
v
esto
r
s
p
r
o
f
itin
g
f
r
o
m
t
h
e
in
f
lu
x
o
f
n
ew
p
ar
ticip
an
ts
with
o
u
t
th
e
g
e
n
er
atio
n
o
f
r
ea
l
v
alu
e.
I
n
o
r
d
er
to
id
en
tify
Sm
ar
t
Po
n
zi
s
ch
em
e
s
in
s
id
e
th
e
B
itco
in
n
etwo
r
k
,
m
an
y
m
eth
o
d
o
lo
g
ie
s
wer
e
u
s
ed
,
in
clu
d
in
g
o
n
es
f
o
cu
s
ed
o
n
d
ata
m
in
i
n
g
[
2
]
.
T
h
e
o
b
jectiv
e
is
to
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
I
SS
N:
2252
-
8
7
7
6
A
h
yb
r
id
ma
ch
in
e
lea
r
n
in
g
a
p
p
r
o
a
ch
fo
r
imp
r
o
ve
d
p
o
n
z
i sch
eme
d
etec
tio
n
…
(
F
a
h
a
d
Ho
s
s
a
in
)
51
u
tili
ze
d
ata
m
in
in
g
m
eth
o
d
o
l
o
g
ies
to
id
en
tify
B
itco
in
ad
d
r
ess
es
as
s
o
ciate
d
with
Po
n
zi
s
ch
em
es.
T
h
is
in
v
o
lv
es
an
aly
zin
g
v
ar
io
u
s
f
ea
tu
r
es
s
u
c
h
as
th
e
ad
d
r
ess
’
s
life
s
p
an
,
ac
tiv
ity
d
u
r
atio
n
,
an
d
to
tal
v
alu
e
tr
an
s
f
er
r
ed
to
th
e
ad
d
r
ess
.
T
h
e
task
is
f
r
am
ed
as
a
b
in
ar
y
class
if
icatio
n
p
r
o
b
lem
,
wh
er
e
a
class
if
ier
is
tr
ain
ed
to
d
is
tin
g
u
is
h
b
etwe
en
ad
d
r
ess
es
lab
eled
as
‘
Po
n
zi
’
an
d
th
o
s
e
la
b
eled
as
‘
n
o
n
-
Po
n
zi
’
.
Su
r
v
i
v
al
an
aly
s
is
h
as
b
ee
n
u
s
ed
in
th
e
r
ea
lm
o
f
B
itco
in
to
id
en
tify
t
h
e
elem
en
ts
th
at
co
n
tr
ib
u
te
to
th
e
p
er
s
is
ten
ce
o
f
s
ca
m
s
[
3
]
.
Vasek
an
d
Mo
o
r
e
d
is
co
v
er
ed
th
at
in
c
r
ea
s
in
g
in
t
er
ac
tio
n
b
etwe
en
f
r
au
d
s
ter
s
a
n
d
th
eir
v
ictim
s
p
r
o
lo
n
g
s
th
e
life
s
p
an
o
f
s
ca
m
s
,
wh
er
ea
s
s
ch
em
es
ten
d
to
h
av
e
s
h
o
r
ter
d
u
r
atio
n
s
wh
en
cr
o
o
k
s
cr
ea
te
th
eir
ac
co
u
n
ts
o
n
th
e
s
am
e
d
ay
th
ey
in
itiate
th
e
s
ca
m
.
Ad
d
itio
n
al
r
esear
ch
h
as
p
r
o
v
id
ed
m
o
r
e
d
et
ailed
in
f
o
r
m
atio
n
o
n
th
ese
f
r
a
u
d
u
len
t
s
ch
em
es b
y
co
llectin
g
an
d
ex
a
m
in
in
g
d
o
c
u
m
en
ted
in
s
tan
ce
s
o
f
s
ca
m
s
[
3
]
,
[
4
]
.
T
h
e
u
s
e
o
f
s
m
ar
t c
o
n
tr
a
cts h
as si
g
n
if
ican
tly
en
ab
led
th
e
s
p
r
ea
d
o
f
Po
n
zi
s
ch
em
es.
B
y
u
s
in
g
s
m
ar
t
co
n
tr
ac
ts
,
s
ch
em
e
in
itiato
r
s
m
ay
o
p
er
ate
an
o
n
y
m
o
u
s
ly
with
o
u
t
an
y
n
ee
d
t
o
r
ev
ea
l
th
e
co
n
tr
ac
t
’
s
n
am
e
o
r
th
e
ca
s
h
with
d
r
awn
f
r
o
m
it
o
n
c
e
it
is
e
s
tab
lis
h
ed
.
Mo
r
eo
v
er
,
th
e
in
h
er
en
t
d
ec
en
tr
aliza
tio
n
o
f
s
m
ar
t
co
n
tr
ac
ts
im
p
lies
th
at
o
n
ce
th
ey
ar
e
ac
t
iv
ated
,
th
er
e
is
n
o
m
ec
h
an
is
m
in
p
lace
t
o
ter
m
in
a
te
th
em
o
r
p
r
o
v
id
e
co
m
p
en
s
atio
n
to
th
o
s
e
wh
o
h
a
v
e
s
u
f
f
e
r
ed
f
in
an
cial
lo
s
s
es.
Fig
u
r
e
1
.
Po
n
zi
s
ch
em
e
co
n
tr
a
ct
ca
teg
o
r
iz
atio
n
Fu
r
th
er
m
o
r
e
,
Po
n
zi
s
ch
em
es
ar
e
in
cr
ea
s
in
g
ly
u
s
in
g
B
itco
in
as
a
p
ay
m
en
t
s
y
s
tem
alo
n
g
s
id
e
co
n
v
en
tio
n
al
cr
im
i
n
al
en
d
ea
v
o
r
s
lik
e
r
an
s
o
m
war
e
[5
]
-
[
7]
an
d
m
o
n
ey
lau
n
d
e
r
in
g
[8
]
,
[
9]
.
T
h
ese
s
ca
m
s
p
r
esen
t
th
em
s
elv
es
as
h
ig
h
-
y
ield
in
v
estme
n
t
p
r
o
g
r
am
s
,
b
u
t
in
r
e
ality
,
th
ey
o
n
ly
r
eim
b
u
r
s
e
in
v
esto
r
s
with
f
u
n
d
s
co
n
tr
ib
u
ted
b
y
n
ew
m
em
b
er
s
.
C
o
n
s
eq
u
en
tly
,
th
ese
s
ca
m
s
co
llap
s
e
wh
en
th
ey
f
ail
to
d
r
a
w
in
n
ew
in
v
esto
r
s
[
1
0
]
.
C
u
r
r
e
n
tly
,
ex
am
in
in
g
B
itco
in
f
r
au
d
s
ty
p
ically
n
ec
ess
itates
an
in
itia
l
p
h
ase
th
at
d
em
an
d
s
s
ig
n
if
ic
an
t
ef
f
o
r
t a
n
d
tim
e,
in
v
o
lv
i
n
g
m
an
u
al
o
r
p
ar
tially
au
to
m
ated
o
n
li
n
e
s
ea
r
ch
es
[
1
1
]
-
[
1
4
]
to
g
ath
e
r
B
itco
in
ad
d
r
ess
es
ass
o
ciate
d
with
th
e
f
r
au
d
.
Au
t
o
m
ated
e
x
am
in
atio
n
o
f
th
e
s
ca
m
’
s
ef
f
ec
t
ca
n
o
n
ly
o
cc
u
r
af
t
er
th
is
s
tep
,
n
am
ely
b
y
ev
al
u
atin
g
b
lo
ck
c
h
ain
tr
an
s
ac
tio
n
in
s
p
ec
tio
n
s
.
Ho
we
v
er
,
th
ese
m
et
h
o
d
s
f
all
s
h
o
r
t
wh
en
f
r
au
d
u
len
t
lo
ca
tio
n
s
r
em
ain
h
id
d
en
,
s
u
c
h
as
wh
en
th
ey
ar
e
o
n
ly
ac
c
ess
ib
le
th
r
o
u
g
h
th
e
d
ee
p
o
r
d
ar
k
web
o
r
p
r
iv
ately
s
h
ar
ed
with
au
th
o
r
ized
in
d
iv
i
d
u
als.
I
n
s
u
ch
s
itu
atio
n
s
,
u
s
in
g
tech
n
o
l
o
g
ies
th
at
ca
n
in
d
ep
en
d
en
tly
s
ea
r
ch
th
e
B
itco
in
b
lo
ck
ch
ain
f
o
r
s
u
s
p
ic
io
u
s
b
eh
av
io
r
s
an
d
d
etec
t
ad
d
r
ess
es
a
s
s
o
ciate
d
with
f
r
au
d
u
len
t
co
n
d
u
ct
wo
u
l
d
b
e
q
u
ite
b
en
e
f
icial.
T
h
e
co
r
e
ten
ets
o
f
a
Sm
ar
t
Po
n
zi
s
ch
em
e
ar
e:
p
ar
ticip
a
n
ts
ar
e
r
eq
u
ir
ed
to
m
ak
e
a
m
in
im
u
m
in
v
estme
n
t
in
o
r
d
e
r
to
jo
in
th
e
p
lan
,
p
ay
m
e
n
ts
to
in
v
esto
r
s
will
o
n
ly
b
e
g
in
i
f
th
er
e
ar
e
e
n
o
u
g
h
am
o
u
n
ts
o
f
ca
s
h
av
ailab
le
,
t
h
e
s
tr
ateg
y
f
ails
wh
en
it
n
o
lo
n
g
er
attr
ac
ts
n
ew
in
v
esto
r
s
,
an
d
i
n
s
u
f
f
icien
t
ca
s
h
to
co
m
p
en
s
ate
in
v
esto
r
s
will r
esu
lt in
th
e
s
ch
em
e
’
s
f
ailu
r
e
.
T
h
is
p
ap
er
in
tr
o
d
u
ce
s
a
n
o
v
el
s
em
an
tic
-
awa
r
e
m
eth
o
d
f
o
r
t
h
e
d
etec
tio
n
o
f
Po
n
zi
s
ch
em
es
th
r
o
u
g
h
th
e
u
s
e
o
f
ad
v
an
ce
d
f
ea
tu
r
e
e
n
g
in
ee
r
in
g
.
I
n
o
r
d
er
to
im
p
r
o
v
e
th
e
ac
cu
r
ac
y
o
f
Po
n
zi
s
ch
em
e
d
etec
tio
n
,
we
im
p
lem
en
t
a
two
-
p
r
o
n
g
e
d
a
p
p
r
o
ac
h
.
I
n
i
t
i
a
l
l
y
,
a
n
X
GB
o
o
s
t
c
l
a
s
s
i
f
ie
r
i
s
u
t
il
i
z
e
d
t
o
t
r
a
i
n
o
n
s
t
r
u
c
t
u
r
e
d
f
e
a
t
u
r
es
e
x
t
r
a
c
t
e
d
f
r
o
m
f
i
n
a
n
c
ia
l
t
r
a
n
s
a
c
ti
o
n
d
a
t
a
.
S
u
b
s
e
q
u
e
n
t
l
y
,
a
t
o
k
e
n
i
z
e
r
is
u
s
e
d
t
o
e
n
c
o
d
e
o
p
c
o
d
e
s
e
q
u
e
n
c
es
e
x
t
r
a
c
t
e
d
f
r
o
m
s
m
a
r
t
c
o
n
t
r
a
c
ts
,
t
r
a
i
n
i
n
g
a
G
R
U
o
n
t
h
e
s
e
s
e
q
u
e
n
c
e
s
t
o
c
a
p
t
u
r
e
t
e
m
p
o
r
a
l
p
a
t
t
e
r
n
s
.
F
e
a
t
u
r
es
d
e
r
i
v
e
d
f
r
o
m
b
o
t
h
t
h
e
X
GB
o
o
s
t
c
l
as
s
i
f
i
e
r
a
n
d
GR
U
[
1
5
]
a
r
e
c
o
n
c
a
t
e
n
a
t
e
d
t
o
f
o
r
m
a
c
o
m
p
r
eh
e
n
s
i
v
e
f
e
at
u
r
e
s
e
t
.
F
i
n
a
l
l
y
,
a
f
i
n
a
l
X
GB
o
o
s
t
m
o
d
e
l
i
s
t
r
a
i
n
e
d
o
n
t
h
e
s
e
c
o
n
c
at
e
n
a
t
e
d
f
e
a
t
u
r
es
t
o
l
e
v
e
r
a
g
e
b
o
t
h
t
h
e
s
t
r
u
c
t
u
r
e
d
f
i
n
a
n
c
i
a
l
d
at
a
a
n
d
t
e
m
p
o
r
a
l
p
a
t
t
e
r
n
s
e
n
c
o
d
e
d
i
n
o
p
c
o
d
e
s
eq
u
e
n
c
e
s
.
T
h
e
p
e
r
f
o
r
m
a
n
c
e
o
f
t
h
e
f
i
n
a
l
m
o
d
e
l
is
e
v
a
l
u
a
t
e
d
r
i
g
o
r
o
u
s
l
y
t
o
a
s
s
ess
i
t
s
e
f
f
e
c
ti
v
e
n
e
s
s
i
n
d
e
t
e
c
t
in
g
P
o
n
z
i
s
c
h
e
m
e
s
w
i
t
h
i
m
p
r
o
v
e
d
a
c
c
u
r
a
c
y
a
n
d
r
e
l
i
a
b
i
li
t
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
7
6
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
,
Vo
l.
1
4
,
No
.
1
,
A
p
r
il
20
2
5
:
50
-
58
52
T
h
e
c
o
n
t
r
i
b
u
t
i
o
n
s
o
f
o
u
r
p
a
p
e
r
a
r
e
:
(
i
)
h
a
v
e
p
r
o
p
o
s
e
d
a
n
d
i
m
p
l
e
m
e
n
t
e
d
a
n
o
v
e
l
h
y
b
r
i
d
m
o
d
e
l
f
o
r
r
e
c
o
g
n
i
z
i
n
g
s
m
a
r
t
P
o
n
z
i
s
c
h
em
e
s
i
n
E
t
h
e
r
e
u
m
c
o
n
t
r
a
c
t
s
,
(
ii
)
h
a
v
e
e
n
g
i
n
e
e
r
e
d
n
e
w
f
e
a
t
u
r
e
s
t
h
a
t
w
i
ll
e
n
h
a
n
c
e
t
h
e
p
e
r
f
o
r
m
a
n
c
e
o
f
t
h
e
m
o
d
e
l
,
(
ii
i
)
c
o
n
s
t
r
u
c
t
ed
a
GR
U
o
n
o
p
c
o
d
e
s
e
q
u
e
n
c
es
i
n
o
r
d
e
r
t
o
e
x
t
r
ac
t
t
e
m
p
o
r
a
l
f
e
a
t
u
r
es
,
a
n
d
(
iv
)
t
h
e
p
r
o
p
o
s
e
d
m
o
d
e
l
m
i
t
i
g
a
t
es
t
h
e
f
a
ls
e
p
o
s
i
ti
v
e
r
a
te
as
w
el
l
a
s
t
h
e
f
al
s
e
n
e
g
a
t
i
v
e
r
a
t
e
w
h
i
c
h
is
p
r
o
m
is
i
n
g
e
n
o
u
g
h
c
o
m
p
a
r
e
d
t
o
t
h
e
e
x
i
s
t
i
n
g
s
y
s
t
e
m
i
n
o
u
r
c
o
n
t
e
x
t
.
Ou
r
ev
alu
atio
n
is
d
r
iv
en
b
y
th
r
ee
r
esear
ch
q
u
esti
o
n
s
th
at
ad
d
r
ess
th
e
k
ey
is
s
u
e
o
f
w
h
et
h
e
r
an
d
h
o
w
th
e
n
o
tio
n
o
f
d
etec
tin
g
Po
n
zi
s
ch
em
es.
−
R
Q1
:
H
o
w
ca
n
ML
m
o
d
els ar
e
u
s
ed
to
d
etec
t Po
n
zi
s
ch
e
m
e
s
?
−
R
Q2
:
H
o
w
ef
f
ec
tiv
e
ar
e
h
y
b
r
id
m
ac
h
in
e
lear
n
in
g
ap
p
r
o
ac
h
es
co
m
b
in
in
g
s
tr
u
ctu
r
ed
f
i
n
an
cial
d
ata
with
s
m
ar
t c
o
n
tr
ac
t a
n
aly
s
is
?
−
R
Q3
:
H
o
w
ca
n
th
e
ac
cu
r
ac
y
o
f
Po
n
zi
s
ch
em
e
d
etec
tio
n
m
o
d
els ar
e
ev
alu
ated
?
2.
RE
L
AT
E
D
WO
RK
W
ith
th
e
ad
v
an
ce
m
en
t
o
f
b
l
o
c
k
ch
ain
tech
n
o
lo
g
y
n
ew
v
ar
ian
ts
o
f
th
e
Po
n
zi
s
ch
em
e
also
em
er
g
ed
.
I
n
2
0
1
8
,
it
was
f
o
u
n
d
th
at
t
h
er
e
ar
e
ar
o
u
n
d
f
o
u
r
h
u
n
d
r
ed
Po
n
zi
s
ch
em
es
in
E
th
e
r
eu
m
b
y
C
h
en
et
a
l.
[
1
6
]
an
d
t
h
ey
ex
tr
ac
ted
th
e
f
ea
t
u
r
es
f
r
o
m
o
p
er
atio
n
co
d
es
b
y
u
s
in
g
m
ac
h
in
e
lear
n
in
g
a
n
d
d
ata
m
in
in
g
.
W
an
g
an
d
Hu
an
g
u
tili
ze
d
th
e
n
-
g
r
am
al
g
o
r
ith
m
f
o
r
e
n
h
an
ce
d
o
p
c
o
d
e
f
ea
tu
r
e
ex
tr
ac
tio
n
an
d
in
teg
r
ated
it
with
co
n
tr
ac
t
ac
co
u
n
t
f
ea
tu
r
es
[
1
7
]
.
T
h
ey
also
in
tr
o
d
u
ce
d
ad
a
p
tiv
e
s
y
n
th
etic
s
am
p
lin
g
(
ADASYN)
to
h
an
d
le
class
im
b
alan
ce
in
th
e
d
ata
an
d
u
s
e
d
th
e
im
p
r
o
v
ed
Ad
aBo
o
s
t
class
if
ier
f
o
r
id
en
tify
in
g
Po
n
zi
s
ch
em
e
co
n
tr
ac
ts
.
I
n
2
0
2
2
,
Aljo
f
ey
et
a
l.
[
1
8
]
ta
ck
led
th
e
p
r
o
b
lem
o
f
d
etec
tin
g
s
m
ar
t
Po
n
zi
co
n
tr
ac
ts
o
v
e
r
th
e
E
th
er
e
u
m
b
lo
ck
ch
ain
b
y
co
n
s
tr
u
ctin
g
an
ef
f
ec
tiv
e
d
etec
tio
n
m
o
d
el
u
s
in
g
d
ata
m
in
in
g
tech
n
iq
u
es.
T
h
e
p
r
o
ce
s
s
th
ey
u
s
ed
in
clu
d
ed
ex
p
an
d
i
n
g
th
e
d
ataset
o
f
s
m
ar
t
Po
n
zi
co
n
tr
ac
ts
,
b
al
an
cin
g
th
e
d
ata
with
ad
ap
tiv
e
s
y
n
th
etic
s
am
p
lin
g
,
an
d
c
r
ea
tin
g
f
o
u
r
d
if
f
e
r
e
n
t
f
e
atu
r
e
s
ets
d
r
awn
f
r
o
m
th
e
o
p
e
r
atio
n
c
o
d
es
(
o
p
co
d
es)
o
f
s
m
ar
t
co
n
tr
ac
ts
.
T
h
ese
s
p
ec
if
ic
f
ea
tu
r
es,
s
u
ch
as
o
p
co
d
e
f
r
e
q
u
en
c
y
,
co
u
n
t
v
ec
to
r
,
n
-
g
r
am
ter
m
f
r
eq
u
en
cy
-
in
v
er
s
e
d
o
cu
m
e
n
t
f
r
eq
u
e
n
cy
(
TF
-
I
DF
)
,
an
d
o
p
c
o
d
e
s
eq
u
e
n
ce
attr
ib
u
tes,
s
tr
en
g
th
en
ed
th
e
m
o
d
el
’
s
d
ep
en
d
a
b
ilit
y
af
ter
t
h
e
s
m
ar
t
co
n
tr
ac
t
’
s
in
tr
o
d
u
ctio
n
t
o
th
e
E
th
er
eu
m
B
lo
ck
ch
ai
n
.
Aljo
f
ey
et
a
l.
[
1
9
]
p
r
o
v
id
es
i
m
p
o
r
tan
t
in
s
ig
h
ts
in
to
th
e
u
n
d
er
s
tan
d
in
g
o
f
Po
n
zi
s
ch
e
m
e
d
etec
tio
n
in
E
th
er
eu
m
,
h
ig
h
lig
h
tin
g
th
e
ef
f
icac
y
o
f
en
s
em
b
le
m
o
d
els
th
at
u
tili
ze
o
p
co
d
e
-
b
ased
f
ea
tu
r
es.
Xu
et
a
l.
[
2
0
]
d
iv
ed
d
ee
p
in
to
th
e
ch
allen
g
e
o
f
d
etec
tin
g
B
itco
in
m
ix
in
g
s
er
v
ices,
wh
ich
b
o
o
s
t
an
o
n
y
m
i
ty
b
y
u
n
clea
r
f
u
n
d
f
lo
w
b
u
t
ar
e
o
f
ten
e
x
p
lo
ited
f
o
r
illeg
al
ac
tiv
ities
lik
e
m
o
n
ey
lau
n
d
er
in
g
.
I
b
b
a
et
a
l.
[
2
1
]
,
E
th
e
r
eu
m
’
s
ca
p
ab
ilit
ies
f
o
r
p
ee
r
-
to
-
p
ee
r
p
r
o
g
r
a
m
m
in
g
an
d
s
m
ar
t
co
n
tr
ac
t
p
u
b
lis
h
in
g
ar
e
e
x
p
lo
r
e
d
,
f
o
cu
s
in
g
o
n
th
e
d
etec
tio
n
o
f
Po
n
zi
s
ch
em
es
u
s
in
g
m
ac
h
in
e
lea
r
n
in
g
.
Fu
r
t
h
er
m
o
r
e
,
Yu
et
a
l.
[
2
2
]
p
r
o
p
o
s
ed
a
GC
N
m
o
d
el
(
g
r
ap
h
co
n
v
o
lu
tio
n
al
n
etwo
r
k
)
to
id
e
n
tify
P
o
n
zi
co
n
tr
ac
ts
with
in
E
th
er
eu
m
.
T
h
e
s
tu
d
y
s
h
o
wca
s
es
th
at
th
e
p
r
o
p
o
s
ed
GC
N
-
b
ased
m
o
d
el
o
f
f
er
s
p
r
o
m
is
in
g
r
esu
lts
co
m
p
ar
ed
to
g
e
n
er
al
m
ac
h
in
e
lear
n
in
g
m
eth
o
d
s
,
co
n
tr
ib
u
tin
g
to
th
e
o
n
g
o
in
g
e
f
f
o
r
ts
to
m
ain
tain
th
e
s
u
s
tain
ab
le
d
ev
elo
p
m
en
t
a
n
d
s
ec
u
r
it
y
o
f
t
h
e
E
th
er
e
u
m
p
latf
o
r
m
.
B
ar
to
letti
et
a
l.
[
2
]
s
h
o
w
a
d
ata
m
i
n
in
g
a
p
p
r
o
ac
h
f
o
r
d
etec
tin
g
B
itco
in
a
d
d
r
ess
es
ass
o
ciate
d
with
Po
n
zi
s
ch
em
es.
T
h
ey
lev
er
ag
e
th
e
p
s
eu
d
o
n
y
m
ity
o
f
B
itco
in
to
tr
ac
e
f
r
au
d
u
len
t
in
v
estme
n
ts
th
at
r
ely
o
n
r
ec
r
u
itin
g
n
ew
u
s
er
s
to
r
ep
ay
ex
is
tin
g
o
n
es.
Z
h
an
g
et
a
l.
[
2
3
]
h
ig
h
li
g
h
ts
two
ex
is
tin
g
ch
a
llen
g
es
in
d
etec
tin
g
s
u
ch
s
ch
em
es
in
th
e
b
lo
ck
ch
ain
:
in
co
m
p
lete
f
ea
tu
r
es
f
o
r
d
etec
tio
n
an
d
in
ef
f
icien
t
alg
o
r
ith
m
s
.
T
h
e
au
th
o
r
s
p
r
o
p
o
s
e
an
in
n
o
v
ativ
e
ap
p
r
o
ac
h
th
at
co
m
b
in
es
b
y
teco
d
e
f
ea
tu
r
es
with
u
s
er
tr
an
s
a
ctio
n
an
d
o
p
co
d
e
f
r
eq
u
e
n
cies,
cr
ea
tin
g
m
o
r
e
c
o
m
p
r
eh
e
n
s
iv
e
f
ea
t
u
r
es.
C
h
en
et
a
l
.
[
1
6
]
p
r
o
p
o
s
e
SADPo
n
zi,
an
in
n
o
v
ativ
e
s
em
an
tic
-
awa
r
e
d
etec
tio
n
ap
p
r
o
ac
h
wh
er
e
th
e
m
o
d
el
u
tili
ze
s
h
eu
r
is
tic
-
g
u
id
ed
s
y
m
b
o
lic
ex
ec
u
tio
n
to
g
en
er
at
e
s
em
an
tic
in
f
o
r
m
atio
n
f
o
r
f
ea
s
ib
l
e
p
ath
s
in
s
m
ar
t
c
o
n
tr
ac
ts
,
id
en
tify
in
g
in
v
esto
r
-
r
elate
d
t
r
an
s
f
er
b
e
h
av
io
u
r
s
an
d
d
is
tr
ib
u
tio
n
s
tr
ateg
ies.
Fan
et
a
l.
[
2
4
]
p
r
o
p
o
s
e
a
n
o
v
el
d
etec
tio
n
m
eth
o
d
f
o
r
Po
n
z
i
s
ch
em
es
o
n
s
m
ar
t
co
n
tr
ac
t
p
latf
o
r
m
s
.
T
h
e
ap
p
r
o
ac
h
u
tili
ze
s
o
r
d
er
ed
tar
g
et
s
tati
s
tics
(
T
S)
to
p
r
o
ce
s
s
ca
teg
o
r
y
f
ea
tu
r
es,
em
p
lo
y
s
d
ata
au
g
m
en
tatio
n
to
ad
d
r
ess
d
ataset
im
b
alan
ce
,
an
d
ad
o
p
ts
th
e
o
r
d
e
r
ed
b
o
o
s
tin
g
alg
o
r
ith
m
to
co
m
b
at
p
r
ed
ictio
n
s
h
if
ts
.
L
ou
et
a
l.
[
2
5
]
p
r
o
p
o
s
ed
an
i
m
p
r
o
v
e
d
C
NN
m
o
d
el
f
o
r
Po
n
zi
s
ch
em
e
d
etec
tio
n
.
Z
h
en
g
et
a
l.
[
2
6
]
p
r
esen
ted
a
n
o
v
el
m
eth
o
d
th
at
u
s
es
a
lar
g
e
d
ataset
to
ex
tr
ac
t
f
ea
tu
r
es
f
r
o
m
th
e
p
er
s
p
ec
tiv
es
o
f
b
y
teco
d
e,
s
em
an
tics
,
an
d
d
ev
elo
p
er
s
to
a
d
d
r
ess
th
ese
d
if
f
icu
lties
.
T
h
ey
d
em
o
n
s
tr
at
e
h
ig
h
e
r
ac
c
u
r
ac
y
in
i
d
en
tif
y
in
g
cle
v
er
Po
n
zi
s
ch
em
es,
ev
en
at
t
h
e
b
e
g
in
n
in
g
o
f
th
eir
f
o
r
m
atio
n
,
u
s
in
g
a
m
ac
h
in
e
lear
n
i
n
g
-
b
ased
m
o
d
e
l
d
u
b
b
ed
th
e
m
u
lti
-
v
iew
ca
s
ca
d
e
en
s
em
b
le
.
Z
h
an
g
et
a
l.
[
2
7
]
,
a
PD
-
SECR
d
e
tectio
n
ap
p
r
o
ac
h
is
p
r
esen
ted
wh
ich
u
s
es
th
e
SMOT
E
E
NN
-
m
ix
ed
s
am
p
lin
g
alg
o
r
ith
m
to
im
p
r
o
v
e
th
e
c
o
m
b
in
ed
m
o
d
el
o
f
c
o
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
s
an
d
r
an
d
o
m
f
o
r
ests
.
Ho
wev
e
r
,
th
ese
s
tu
d
ies
ar
e
n
o
t
with
o
u
t
th
eir
lim
itatio
n
s
,
d
esp
ite
th
eir
co
n
tr
ib
u
tio
n
s
.
So
m
e
o
f
th
e
d
atasets
th
ey
r
es
ea
r
ch
ed
we
r
e
s
m
all,
f
o
r
e
x
am
p
le,
th
e
p
r
o
p
o
s
ed
SADPo
n
zi
m
o
d
el
p
r
o
p
o
s
ed
b
y
C
h
en
et
a
l.
[
1
6
]
ex
p
er
im
e
n
ted
with
o
n
ly
1
3
9
5
s
am
p
les.
Als
o
,
s
o
m
e
o
f
th
e
p
r
o
p
o
s
ed
m
o
d
e
ls
d
id
n
o
t
p
r
o
v
id
e
a
clea
r
p
ictu
r
e
o
f
f
alse
p
o
s
itiv
e
an
d
f
alse
n
eg
ativ
e
r
ate
s
[
2
6
]
.
L
ian
g
et
a
l.
[
2
8
]
p
r
o
p
o
s
ed
a
Po
n
ziGu
ar
d
Po
n
z
i
s
ch
em
e
u
s
in
g
a
co
n
tr
ac
t
r
u
n
tim
e
b
eh
av
io
r
g
r
ap
h
(
C
R
B
G)
.
T
h
e
lim
itatio
n
s
o
f
C
R
B
G
ar
e
co
m
p
u
tatio
n
al
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
I
SS
N:
2252
-
8
7
7
6
A
h
yb
r
id
ma
ch
in
e
lea
r
n
in
g
a
p
p
r
o
a
ch
fo
r
imp
r
o
ve
d
p
o
n
z
i sch
eme
d
etec
tio
n
…
(
F
a
h
a
d
Ho
s
s
a
in
)
53
co
m
p
lex
ity
,
s
ca
lab
ilit
y
,
o
p
e
r
at
e
o
n
ly
o
n
a
ce
r
tain
lev
el
o
f
ab
s
tr
ac
tio
n
.
On
u
et
a
l.
[
2
9
]
ap
p
li
ed
s
ev
er
al
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
f
o
r
Po
n
zi
d
etec
tio
n
to
ad
d
r
ess
th
e
n
eg
ati
v
e
im
p
ac
t
o
f
Po
n
zi
s
ch
em
es
u
s
in
g
th
e
E
th
er
eu
m
tr
an
s
ac
tio
n
s
d
ataset.
T
h
e
s
ize
o
f
th
e
d
ataset
u
s
ed
to
v
alid
ate
th
e
m
o
d
els is
s
m
all
in
s
ize.
3.
P
RO
P
O
SE
D
M
E
T
H
O
D
F
ig
u
r
e
2
esp
ec
ially
F
ig
u
r
e
2
(
a)
,
we
illu
s
tr
ate
th
e
s
tep
s
i
n
v
o
lv
ed
i
n
d
ev
elo
p
in
g
an
d
ev
al
u
atin
g
o
u
r
h
y
b
r
id
m
ac
h
in
e
-
lear
n
in
g
ap
p
r
o
ac
h
f
o
r
Po
n
zi
s
ch
em
e
d
etec
t
io
n
.
T
h
e
m
eth
o
d
o
lo
g
y
co
n
s
is
ts
o
f
s
ev
er
al
cr
u
cial
s
tag
es,
in
clu
d
in
g
d
ataset
s
elec
tio
n
,
d
ata
p
r
ep
r
o
ce
s
s
i
n
g
,
f
ea
tu
r
e
ex
tr
ac
tio
n
,
a
n
d
m
o
d
el
ev
al
u
a
tio
n
.
3
.
1
.
Da
t
a
s
et
W
e
co
llected
th
e
d
ataset
f
r
o
m
th
e
Kag
g
le
web
s
ite,
wh
ich
p
r
o
v
id
es
d
etailed
in
f
o
r
m
atio
n
o
n
E
th
er
eu
m
s
m
ar
t
co
n
tr
ac
ts
[
3
0
]
.
T
h
e
d
at
aset
co
m
p
r
is
es
3
7
8
6
en
tr
ies
a
n
d
in
clu
d
es
f
o
u
r
k
ey
f
ea
tu
r
es
:
ad
d
r
ess
,
o
p
co
d
e,
lab
el,
an
d
c
r
ea
to
r
.
T
h
e
a
d
d
r
es
s
f
ea
tu
r
e
lis
ts
th
e
u
n
iq
u
e
id
en
tifie
r
s
f
o
r
ea
ch
s
m
ar
t
co
n
t
r
ac
t,
wh
ile
th
e
o
p
co
d
e
f
ea
tu
r
e
co
n
tain
s
th
e
d
is
ass
e
m
b
led
b
y
teco
d
e
in
s
tr
u
ctio
n
s
o
f
th
ese
c
o
n
tr
ac
ts
.
T
h
e
lab
el
f
ea
tu
r
e
in
d
icate
s
wh
eth
er
a
s
m
ar
t
co
n
tr
ac
t
is
a
Po
n
zi
s
ch
em
e,
as
d
eter
m
in
e
d
th
r
o
u
g
h
m
an
u
al
i
n
s
p
ec
tio
n
.
Fin
ally
,
th
e
cr
ea
to
r
f
ea
tu
r
e
id
e
n
tifie
s
th
e
in
d
iv
id
u
als
o
r
en
titi
es
th
at
cr
ea
ted
th
ese
s
m
ar
t
co
n
tr
ac
ts
.
T
h
is
d
ataset
s
er
v
es
as
th
e
f
o
u
n
d
atio
n
f
o
r
o
u
r
an
aly
s
is
an
d
m
o
d
el
tr
ai
n
in
g
.
3
.
2
.
Da
t
a
prepro
ce
s
s
in
g
E
f
f
ec
tiv
e
d
ata
p
r
ep
r
o
ce
s
s
in
g
is
es
s
en
tial
f
o
r
b
u
ild
in
g
a
r
o
b
u
s
t
m
ac
h
in
e
-
lear
n
in
g
m
o
d
el.
T
h
is
s
tu
d
y
’
s
p
r
ep
r
o
ce
s
s
in
g
s
tep
s
in
clu
d
e
d
ata
clea
n
in
g
,
lab
el
en
co
d
i
n
g
,
an
d
s
ca
lin
g
th
e
d
ata
u
s
in
g
t
h
e
Min
Ma
x
Scaler
wh
ich
is
illu
s
tr
ated
in
Fig
u
r
e
2
(
b
)
.
(
a)
(
b
)
Fig
u
r
e
2
.
B
lo
ck
d
iag
r
am
o
f
(
a
)
p
r
o
p
o
s
ed
m
eth
o
d
o
lo
g
y
an
d
(
b
)
d
ata
p
r
e
p
r
o
ce
s
s
in
g
3
.
2
.
1
.
Da
t
a
c
lea
nin
g
Data
clea
n
in
g
is
a
cr
u
cial
s
t
ep
in
p
r
ep
r
o
ce
s
s
in
g
to
en
s
u
r
e
th
e
q
u
ality
o
f
th
e
d
ata.
T
h
is
in
v
o
lv
es
ad
d
r
ess
in
g
m
is
s
in
g
v
alu
es
b
y
eith
er
ig
n
o
r
in
g
th
e
m
o
r
f
illi
n
g
th
em
in
with
ap
p
r
o
p
r
iate
esti
m
ates.
Ad
d
itio
n
ally
,
n
o
is
y
d
ata,
wh
ich
m
ay
r
esu
lt
f
r
o
m
r
a
n
d
o
m
er
r
o
r
s
o
r
v
ar
ia
n
ce
s
,
is
s
m
o
o
th
e
d
u
s
in
g
tech
n
iq
u
es
lik
e
b
in
n
i
n
g
,
r
eg
r
ess
io
n
,
a
n
d
clu
s
ter
in
g
.
Fo
r
e
x
am
p
le,
b
in
n
in
g
o
r
g
an
iz
es
d
ata
in
to
eq
u
al
-
s
ized
b
in
s
,
allo
win
g
f
o
r
th
e
r
ep
lace
m
en
t
o
f
v
al
u
es
with
th
e
b
in
’
s
m
ea
n
o
r
m
ed
ia
n
.
Ou
tli
er
s
,
o
r
d
ata
p
o
in
ts
th
at
d
e
v
iate
s
ig
n
if
ican
tly
f
r
o
m
o
th
er
s
,
ar
e
id
e
n
tifie
d
an
d
r
em
o
v
ed
u
s
in
g
clu
s
ter
in
g
m
eth
o
d
s
,
wh
er
e
in
co
n
s
is
ten
t
d
ata
is
s
ep
ar
ated
f
r
o
m
th
e
m
ain
g
r
o
u
p
s
.
3
.
2
.
2
.
L
a
bel
e
nco
der
L
ab
el
en
co
d
in
g
[
3
1
]
is
u
s
ed
to
co
n
v
er
t
ca
teg
o
r
ical
lab
els
in
t
o
n
u
m
er
ical
v
alu
es,
m
a
k
in
g
th
e
m
s
u
itab
le
f
o
r
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
.
T
h
is
p
r
o
ce
s
s
ass
ig
n
s
a
u
n
iq
u
e
i
n
teg
er
to
ea
c
h
ca
teg
o
r
y
,
tr
an
s
f
o
r
m
in
g
th
e
d
ataset
in
to
a
f
o
r
m
at
t
h
at
th
e
m
o
d
el
ca
n
ea
s
ily
in
ter
p
r
et.
Fo
r
in
s
tan
ce
,
if
y
is
th
e
ca
teg
o
r
ic
al
v
ar
iab
le,
th
e
lab
el
en
co
d
er
m
ap
s
ea
ch
ca
teg
o
r
y
y
i to
a
n
u
m
e
r
ical
v
alu
e
y
ˆ
i a
s
:
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
7
6
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
,
Vo
l.
1
4
,
No
.
1
,
A
p
r
il
20
2
5
:
50
-
58
54
ˆ
=
(
)
t
h
is
tr
an
s
f
o
r
m
atio
n
is
ess
en
tial
wh
en
d
ea
lin
g
with
ca
teg
o
r
ical
d
ata
in
th
e
tr
ain
i
n
g
p
r
o
ce
s
s
.
3
.
2
.
3
.
M
inM
a
x
s
ca
ler
T
h
e
Min
Ma
x
s
ca
ler
is
ap
p
lie
d
to
n
o
r
m
alize
th
e
d
ata
b
y
s
ca
lin
g
all
th
e
f
ea
tu
r
e
v
alu
es
to
a
s
p
ec
if
ic
r
an
g
e,
ty
p
ically
b
etwe
en
0
a
n
d
1
.
T
h
is
en
s
u
r
es
th
at
n
o
f
ea
tu
r
e
d
o
m
i
n
ates
th
e
lear
n
in
g
p
r
o
ce
s
s
d
u
e
to
its
m
ag
n
itu
d
e.
T
h
e
s
ca
lin
g
is
d
o
n
e
u
s
in
g
th
e
f
o
r
m
u
la:
̂
=
−
−
wh
er
e
x
r
e
p
r
esen
ts
th
e
o
r
ig
in
a
l
f
ea
tu
r
e
v
al
u
e,
an
d
x
min
an
d
x
max
ar
e
th
e
m
i
n
im
u
m
an
d
m
ax
im
u
m
v
alu
es
of
th
a
t
f
ea
tu
r
e,
r
esp
ec
tiv
ely
.
No
r
m
alizin
g
th
e
d
ata
in
th
i
s
way
h
elp
s
im
p
r
o
v
e
th
e
p
er
f
o
r
m
an
ce
an
d
co
n
v
er
g
en
ce
s
p
ee
d
of
th
e
m
ac
h
in
e
lear
n
i
n
g
m
o
d
el.
Af
ter
p
r
ep
r
o
ce
s
s
in
g
of
th
e
d
ataset,
we
s
p
lit
it
in
to
tr
ain
in
g
an
d
test
in
g
s
ets.
A
clas
s
ic
80
-
20
d
iv
id
e
was
u
t
ilized
to
m
ak
e
s
u
r
e
th
e
m
o
d
e
l
h
ad
en
o
u
g
h
m
ater
ial
to
lear
n
f
r
o
m
wh
ile
s
till
h
av
in
g
s
u
f
f
icien
t d
ata
lef
t
f
o
r
an
u
n
b
iase
d
ev
alu
atio
n
.
3
.
3
.
F
e
a
t
ure
ex
t
r
a
ct
io
n/b
a
s
e
lea
rner
Featu
r
e
ex
tr
ac
tio
n
/
B
ase
L
ea
r
n
er
is
a
cr
u
cial
s
tep
in
th
e
p
r
o
p
o
s
ed
h
y
b
r
id
ap
p
r
o
ac
h
,
as
it
en
h
an
ce
s
th
e
m
o
d
el
’
s
ab
ilit
y
to
d
etec
t
Po
n
zi
s
ch
em
es
by
lev
er
ag
in
g
th
e
s
tr
en
g
th
s
of
b
o
th
s
tatic
an
d
s
eq
u
en
tial
d
ata
an
aly
s
is
.
In
th
is
s
tag
e,
we
u
s
e
two
d
if
f
er
en
t
m
o
d
els
-
XGBo
o
s
t
an
d
g
ate
d
r
ec
u
r
r
en
t
u
n
it
(
GR
U)
-
to
ex
tr
ac
t
m
ea
n
in
g
f
u
l
f
ea
tu
r
es
f
r
o
m
t
h
e
d
ataset.
3
.
3
.
1
.
XG
B
o
o
s
t
c
la
s
s
if
ier
T
h
e
f
ir
s
t
s
tep
in
f
ea
tu
r
e
ex
t
r
ac
tio
n
in
v
o
lv
es
tr
ai
n
in
g
an
XGBo
o
s
t
class
if
ier
o
n
th
e
d
ataset
[
3
2
]
.
XGBo
o
s
t
is
a
p
o
wer
f
u
l
g
r
a
d
ien
t
-
b
o
o
s
tin
g
alg
o
r
ith
m
k
n
o
wn
f
o
r
its
ef
f
icien
c
y
a
n
d
h
i
g
h
p
er
f
o
r
m
an
ce
in
class
if
icatio
n
task
s
.
On
ce
th
e
m
o
d
el
is
tr
ain
ed
,
we
ex
tr
ac
t
th
e
m
o
s
t
im
p
o
r
tan
t
f
ea
t
u
r
e
s
id
en
tifie
d
b
y
th
e
XGBo
o
s
t
clas
s
if
ier
.
T
h
ese
f
ea
tu
r
es
ca
p
tu
r
e
th
e
s
tatic
r
elatio
n
s
h
ip
s
with
in
th
e
d
ata
an
d
s
er
v
e
as
an
ess
en
tial
in
p
u
t
to
o
u
r
h
y
b
r
id
a
p
p
r
o
ac
h
.
Ma
th
em
atica
lly
,
XGBo
o
s
t
m
in
im
izes
th
e
f
o
llo
win
g
o
b
jectiv
e
f
u
n
ctio
n
d
u
r
in
g
tr
ain
in
g
:
Ob
j
(
)
=
∑
=
1
(
,
̂
)
+
∑
=
1
Ω
(
)
wh
er
e
L
(
y
i,
y
ˆ
i)
is
th
e
lo
s
s
f
u
n
ctio
n
,
an
d
Ω
(
f
k
)
is
th
e
r
eg
u
la
r
izatio
n
ter
m
to
p
r
ev
e
n
t
o
v
er
f
itt
in
g
.
T
h
e
im
p
o
r
tan
t
f
ea
tu
r
es
ex
tr
ac
ted
f
r
o
m
XGB
o
o
s
t
ar
e
th
o
s
e
th
at
co
n
tr
ib
u
te
m
o
s
t
s
ig
n
if
ican
tly
to
m
in
i
-
m
izin
g
th
is
o
b
jectiv
e
f
u
n
ctio
n
,
th
u
s
p
r
o
v
id
in
g
v
alu
a
b
le
in
s
ig
h
ts
in
to
th
e
d
ataset.
3
.
3
.
2
.
G
a
t
ed
re
curr
ent
un
it
I
n
p
ar
allel,
we
em
p
lo
y
a
GR
U
m
o
d
el
[
3
3
]
to
an
aly
ze
th
e
o
p
co
d
e
s
eq
u
en
ce
s
in
th
e
d
ataset.
GR
U
s
,
a
v
ar
ian
t
of
r
ec
u
r
r
en
t
n
eu
r
al
n
et
wo
r
k
s
(
R
NNs)
[
3
4
]
,
ar
e
well
-
s
u
ited
f
o
r
h
a
n
d
lin
g
s
eq
u
en
tial
d
ata,
s
u
ch
as
o
p
co
d
e
s
eq
u
en
ce
s
f
o
u
n
d
in
s
m
ar
t
co
n
tr
ac
ts
.
T
h
e
GR
U
m
o
d
el
lear
n
s
p
atter
n
s
o
v
er
tim
e
an
d
ex
tr
ac
ts
f
ea
tu
r
es
th
at
r
ef
lect
th
e
tem
p
o
r
al
d
ep
en
d
en
cies
in
t
h
e
d
ata.
T
h
e
GR
U
ce
ll
’
s
o
p
e
r
atio
n
s
ca
n
be
d
escr
ib
ed
m
ath
em
atica
lly
as
f
o
llo
ws:
=
(
⋅
[
ℎ
−
1
,
]
)
=
(
⋅
[
ℎ
−
1
,
]
)
ℎ
̃
=
ta
n
h
(
⋅
[
∗
ℎ
−
1
,
]
)
ℎ
=
(
1
−
)
∗
ℎ
−
1
+
∗
ℎ
̃
wh
er
e
is
th
e
u
p
d
ate
g
ate,
is
th
e
r
eset
g
ate,
h
t
is
th
e
h
id
d
en
s
tate,
a
n
d
x
t
is
th
e
in
p
u
t
at
tim
e
s
tep
t
.
T
h
e
GR
U
m
o
d
el
g
en
er
ates
f
ea
tu
r
es
th
at
ca
p
tu
r
e
th
e
s
eq
u
e
n
tial
d
y
n
am
ics
of
th
e
d
ata,
w
h
ich
ar
e
cr
u
cial
f
o
r
u
n
d
er
s
tan
d
i
n
g
c
o
m
p
lex
p
atter
n
s
in
s
m
ar
t
co
n
tr
ac
t
b
eh
av
io
r
.
3
.
3
.
3
.
H
y
brid
f
ea
t
ure
s
elec
t
io
n
Af
ter
ex
tr
ac
ti
ng
f
ea
tu
r
es
f
r
o
m
b
o
th
th
e
XGBo
o
s
t
class
if
ier
an
d
th
e
GR
U
m
o
d
el,
we
co
n
ca
te
n
ate
th
es
e
f
ea
tu
r
es
to
c
r
ea
te
a
c
o
m
p
r
e
h
e
n
s
iv
e
h
y
b
r
id
f
ea
tu
r
e
s
et.
T
h
is
co
m
b
in
e
d
f
ea
tu
r
e
s
et
lev
e
r
ag
es
th
e
s
tr
en
g
th
s
o
f
b
o
th
m
o
d
els,
i
n
teg
r
atin
g
s
tatic
an
d
s
eq
u
en
tial
i
n
f
o
r
m
atio
n
.
T
h
e
h
y
b
r
id
f
ea
tu
r
e
s
elec
tio
n
all
o
ws
th
e
s
u
b
s
eq
u
e
n
t
m
eta
-
lear
n
er
to
m
ak
e
m
o
r
e
in
f
o
r
m
ed
p
r
ed
ictio
n
s
,
im
p
r
o
v
in
g
th
e
o
v
er
all
d
etec
tio
n
ac
cu
r
ac
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
I
SS
N:
2252
-
8
7
7
6
A
h
yb
r
id
ma
ch
in
e
lea
r
n
in
g
a
p
p
r
o
a
ch
fo
r
imp
r
o
ve
d
p
o
n
z
i sch
eme
d
etec
tio
n
…
(
F
a
h
a
d
Ho
s
s
a
in
)
55
3
.
4
.
M
et
a
-
l
e
a
rner
(
XG
B
o
o
s
t
a
lg
o
rit
hm
)
T
h
e
m
eta
-
lear
n
er
in
o
u
r
ap
p
r
o
ac
h
u
tili
ze
s
th
e
XG
B
o
o
s
t
alg
o
r
ith
m
to
co
m
b
in
e
an
d
r
e
f
in
e
f
e
atu
r
es
ex
-
tr
ac
ted
f
r
o
m
th
e
b
ase
lear
n
er
s
.
T
h
is
s
ec
tio
n
elab
o
r
ates
o
n
th
e
r
o
le
o
f
XGBo
o
s
t
as
a
m
eta
-
lear
n
er
,
a
n
d
its
in
teg
r
atio
n
in
to
th
e
o
v
e
r
all
m
eth
o
d
o
lo
g
y
.
3
.
4
.
1
.
XG
B
o
o
s
t
a
s
a
m
et
a
-
lea
rner
I
n
o
u
r
h
y
b
r
id
m
ac
h
in
e
-
lear
n
in
g
f
r
a
m
ewo
r
k
,
th
e
XGBo
o
s
t
alg
o
r
ith
m
is
em
p
lo
y
ed
as
a
m
et
a
-
lear
n
er
to
lev
er
ag
e
t
h
e
f
ea
tu
r
es
e
x
tr
ac
ted
f
r
o
m
th
e
XGBo
o
s
t
class
if
ier
an
d
t
h
e
GR
U
m
o
d
el.
T
h
e
r
o
le
of
th
e
m
eta
-
lear
n
er
is
to
co
m
b
in
e
th
ese
f
ea
tu
r
es
an
d
m
ak
e
f
in
al
p
r
e
d
ictio
n
s
by
in
teg
r
atin
g
th
e
in
s
ig
h
ts
g
ain
ed
f
r
o
m
b
o
th
b
ase
m
o
d
els.
XGBo
o
s
t,
k
n
o
w
n
f
o
r
its
h
ig
h
p
e
r
f
o
r
m
an
ce
a
n
d
ac
cu
r
ac
y
in
class
if
icatio
n
task
s
,
en
h
an
ce
s
th
e
p
r
ed
ictiv
e
p
o
wer
of
o
u
r
ap
p
r
o
a
ch
by
ef
f
ec
tiv
ely
h
an
d
lin
g
c
o
m
p
lex
in
ter
ac
tio
n
s
b
etwe
en
f
e
atu
r
es.
3
.
4
.
2
.
F
ea
t
ure
c
o
m
bin
a
t
io
n a
nd
t
ra
ini
ng
T
h
e
co
m
b
in
ed
f
ea
tu
r
e
s
et,
co
n
s
is
tin
g
o
f
f
ea
t
u
r
es
f
r
o
m
b
o
th
th
e
XGBo
o
s
t
class
if
ier
an
d
t
h
e
GR
U
m
o
d
el,
is
u
s
ed
as
in
p
u
t
to
th
e
XGBo
o
s
t
m
eta
-
lear
n
er
.
T
h
is
in
teg
r
atio
n
allo
ws
th
e
m
eta
-
lear
n
er
to
ca
p
tu
r
e
an
d
ex
p
lo
it
th
e
co
m
p
lem
en
tar
y
s
tr
en
g
th
s
of
th
e
b
ase
m
o
d
el
s.
T
r
ain
in
g
th
e
XGBo
o
s
t
m
et
a
-
lear
n
er
in
v
o
lv
es
f
itti
n
g
th
e
m
o
d
el
on
th
is
h
y
b
r
i
d
f
ea
tu
r
e
s
et,
en
ab
lin
g
it
to
m
ak
e
in
f
o
r
m
ed
d
ec
is
io
n
s
b
ased
on
th
e
co
m
b
in
ed
in
s
ig
h
ts
.
T
h
e
alg
o
r
ith
m
’
s
g
r
ad
ien
t
b
o
o
s
tin
g
f
r
am
ewo
r
k
f
u
r
th
er
r
e
f
in
es
th
e
m
o
d
el
’
s
p
r
ed
ictio
n
s
b
y
m
in
im
izin
g
th
e
e
r
r
o
r
t
h
r
o
u
g
h
i
ter
ativ
e
b
o
o
s
tin
g
.
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
I
n
th
is
s
ec
tio
n
,
we
p
r
esen
t
t
h
e
r
esu
lts
o
f
o
u
r
h
y
b
r
id
m
ac
h
i
n
e
-
lear
n
in
g
ap
p
r
o
ac
h
f
o
r
Po
n
zi
s
ch
em
e
d
etec
tio
n
,
f
o
cu
s
in
g
on
k
ey
p
e
r
f
o
r
m
a
n
ce
m
etr
ics
an
d
v
is
u
al
r
ep
r
esen
tati
o
n
s
.
T
h
e
p
er
f
o
r
m
a
n
ce
o
f
o
u
r
m
o
d
el
is
ev
alu
ated
u
s
in
g
s
ev
er
al
k
e
y
m
etr
ics:
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all
,
an
d
F1
-
s
co
r
e
w
h
ich
is
illu
s
tr
ated
in
Fig
u
r
e
3
.
Acc
u
r
ac
y
r
e
f
lects
th
e
o
v
er
al
l
p
er
f
o
r
m
an
ce
o
f
th
e
m
o
d
el
b
y
ca
lcu
latin
g
th
e
r
atio
o
f
co
r
r
ec
tl
y
p
r
ed
icted
in
s
tan
ce
s
to
th
e
to
tal
in
s
tan
ce
s
.
Pre
cisi
o
n
in
d
icate
s
th
e
p
er
ce
n
tag
e
o
f
tr
u
e
p
o
s
itiv
e
p
r
ed
ictio
n
s
am
o
n
g
all
p
o
s
itiv
e
p
r
ed
ictio
n
s
g
en
er
ated
b
y
th
e
m
o
d
el.
R
ec
all
a
s
s
es
s
es
th
e
m
o
d
el
’
s
ab
ilit
y
to
co
r
r
ec
tly
id
en
tify
ac
tu
al
p
o
s
itiv
e
in
s
tan
ce
s
.
T
h
e
F1
-
s
co
r
e
,
wh
ich
is
th
e
h
ar
m
o
n
ic
m
ea
n
o
f
p
r
ec
is
io
n
an
d
r
ec
all,
o
f
f
er
s
a
b
alan
ce
d
m
ea
s
u
r
e
th
at
ac
co
u
n
ts
f
o
r
b
o
th
m
etr
ics.
Fo
r
o
u
r
m
o
d
el,
th
e
a
cc
u
r
ac
y
is
9
6
.
8
3
%,
with
a
p
r
ec
is
io
n
o
f
7
8
.
3
8
%,
a
r
ec
all
o
f
6
4
.
4
5
%,
a
n
d
an
F1
-
s
co
r
e
of
7
0
.
7
3
%.
T
h
e
co
n
f
u
s
io
n
m
atr
ix
p
r
o
v
i
d
e
s
a
d
etailed
b
r
ea
k
d
o
wn
o
f
th
e
m
o
d
el
’
s
class
if
icatio
n
p
er
f
o
r
m
an
ce
b
y
co
m
p
ar
in
g
p
r
ed
icted
lab
els
to
th
e
ac
tu
al
lab
els
(
Fig
u
r
e
4
)
.
I
t is
an
ess
en
tial
to
o
l
f
o
r
u
n
d
e
r
s
tan
d
in
g
th
e
t
y
p
es
o
f
er
r
o
r
s
m
ad
e
b
y
t
h
e
m
o
d
el
an
d
ass
ess
in
g
its
ef
f
e
ctiv
en
ess
.
Fig
u
r
e
4
(
a)
illu
s
tr
ated
th
e
co
n
f
u
s
io
n
m
atr
ix
f
o
r
th
e
h
y
b
r
id
m
ac
h
i
n
e
-
lear
n
in
g
m
o
d
el.
Her
e
’
s
wh
at
ea
c
h
v
alu
e
in
th
e
co
n
f
u
s
io
n
m
atr
ix
r
ep
r
esen
ts
:
tr
u
e
n
e
g
ativ
es
(
T
N)
=
7
0
5
: T
h
e
n
u
m
b
er
o
f
No
n
-
Po
n
zi
in
s
tan
ce
s
co
r
r
ec
tly
c
lass
if
ied
as
No
n
-
Po
n
zi
m
,
f
alse
p
o
s
itiv
es (
FP
)
=
8
:
t
h
e
n
u
m
b
e
r
o
f
No
n
-
Po
n
zi
in
s
t
an
ce
s
in
co
r
r
ec
tly
class
if
ied
as
Po
n
zi
,
f
alse
n
eg
ativ
es
(
FN)
=
1
6
:
t
h
e
n
u
m
b
er
o
f
Po
n
zi
in
s
tan
ce
s
in
co
r
r
ec
tly
class
if
ied
as
No
n
-
Po
n
zi
,
an
d
tr
u
e
p
o
s
itiv
es
(
T
P)
=
2
9
:
t
h
e
n
u
m
b
er
o
f
Po
n
zi
in
s
tan
ce
s
co
r
r
ec
tly
clas
s
if
ied
as
Po
n
zi.
B
y
an
al
y
zin
g
th
e
co
n
f
u
s
io
n
m
atr
ix
,
we
ca
n
s
ee
th
at
th
e
m
o
d
el
p
er
f
o
r
m
s
well
in
id
en
tify
in
g
No
n
-
Po
n
zi
in
s
tan
ce
s
,
with
a
h
ig
h
n
u
m
b
er
o
f
TN
.
Ho
we
v
er
,
th
er
e
is
a
tr
ad
e
-
o
f
f
b
etwe
en
FP
an
d
FN
,
wh
ich
r
e
f
lects a
r
ea
s
wh
er
e
th
e
m
o
d
el
c
o
u
ld
b
e
im
p
r
o
v
e
d.
Fig
u
r
e
3
.
Per
f
o
r
m
an
c
e
m
etr
ics
of
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
an
d
F1
-
s
co
r
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
7
6
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
,
Vo
l.
1
4
,
No
.
1
,
A
p
r
il
20
2
5
:
50
-
58
56
Un
d
er
s
tan
d
in
g
th
ese
m
etr
ics
h
elp
s
u
s
r
ef
in
e
o
u
r
ap
p
r
o
a
ch
to
ac
h
iev
e
b
etter
p
er
f
o
r
m
an
ce
an
d
ac
cu
r
ac
y
in
d
etec
tin
g
Po
n
zi
s
ch
em
es.
T
h
e
r
ec
eiv
er
o
p
e
r
atin
g
ch
a
r
ac
ter
is
tic
(
R
OC
)
cu
r
v
e
illu
s
tr
ates
th
e
m
o
d
el
’
s
p
e
r
f
o
r
m
an
ce
ac
r
o
s
s
d
if
f
er
en
t
th
r
esh
o
ld
s
.
T
h
e
a
r
ea
u
n
d
e
r
th
e
cu
r
v
e
(
AUC)
is
a
k
ey
in
d
icato
r
o
f
t
h
e
m
o
d
el
’
s
ab
ilit
y
to
d
is
cr
im
in
ate
b
etwe
en
Po
n
zi
an
d
No
n
-
Po
n
z
i in
s
tan
ce
s
.
Ou
r
m
o
d
el
ac
h
ie
v
es a
n
AUC o
f
0
.
9
3
,
in
d
icatin
g
a
h
i
g
h
lev
el
o
f
p
e
r
f
o
r
m
an
ce
.
Fig
u
r
e
4
(
b
)
d
is
p
lay
s
th
e
R
OC
cu
r
v
e
f
o
r
o
u
r
m
o
d
el.
(
a)
(
b)
Fig
u
r
e
4
.
Per
f
o
r
m
an
c
e
o
f
Po
n
zi
(
a)
c
o
n
f
u
s
io
n
m
atr
i
x
an
d
(
b
)
R
OC
c
u
r
v
e
T
h
e
r
esu
lts
o
f
o
u
r
h
y
b
r
id
m
a
ch
in
e
lear
n
in
g
ap
p
r
o
ac
h
s
h
o
w
s
tr
o
n
g
p
er
f
o
r
m
a
n
ce
in
d
ete
ctin
g
Po
n
zi
s
ch
em
es.
T
h
e
m
o
d
el
ac
h
iev
e
d
an
ac
cu
r
ac
y
o
f
9
6
.
8
3
%,
d
e
m
o
n
s
tr
atin
g
its
o
v
er
all
r
eliab
ilit
y
.
Pre
cisi
o
n
was
7
8
.
3
8
%,
in
d
icatin
g
a
g
o
o
d
p
r
o
p
o
r
tio
n
o
f
co
r
r
ec
tly
id
en
tifie
d
Po
n
zi
s
ch
em
es
am
o
n
g
th
o
s
e
p
r
ed
icted
.
R
ec
all
was
6
4
.
4
5
%,
r
ef
lectin
g
th
e
m
o
d
el
’
s
ab
ilit
y
to
co
r
r
ec
tly
i
d
en
tify
Po
n
zi
s
ch
em
es
f
r
o
m
all
ac
tu
al
in
s
tan
ce
s
.
T
h
e
F1
-
s
co
r
e
o
f
7
0
.
7
3
%
b
ala
n
ce
s
th
ese
two
m
etr
ics,
co
n
f
ir
m
in
g
th
e
m
o
d
el
’
s
ef
f
ec
tiv
en
es
s
.
Ad
d
itio
n
ally
,
th
e
R
OC
cu
r
v
e
with
an
AUC
o
f
0
.
9
3
f
u
r
th
e
r
h
ig
h
lig
h
ts
th
e
m
o
d
el
’
s
s
tr
o
n
g
d
is
cr
im
in
ato
r
y
p
o
wer
b
etwe
en
Po
n
zi
an
d
No
n
-
Po
n
zi
in
s
tan
ce
s
.
Ov
er
all,
th
ese
r
esu
lts
v
alid
ate
th
e
r
o
b
u
s
tn
ess
o
f
o
u
r
ap
p
r
o
ac
h
in
ac
cu
r
ately
id
en
tify
in
g
f
r
au
d
u
len
t
ac
tiv
iti
es
in
E
th
er
eu
m
s
m
ar
t
co
n
tr
a
cts.
I
n
T
ab
le
1
we
h
a
v
e
d
is
cu
s
s
ed
th
e
r
esear
ch
q
u
esti
o
n
an
d
an
s
wer
f
o
r
ex
p
lai
n
in
g
o
u
r
m
eth
o
d
s
.
T
ab
le
1
.
R
esear
ch
q
u
esti
o
n
an
d
an
s
wer
b
ased
o
n
o
u
r
r
esear
c
h
SL
Q
u
e
st
i
o
n
A
n
sw
e
r
R
Q
1
W
h
a
t
r
o
l
e
d
o
e
s
X
G
B
o
o
s
t
p
l
a
y
i
n
t
h
e
p
r
o
p
o
s
e
d
me
t
h
o
d
o
l
o
g
y
?
X
G
B
o
o
st
i
s
i
n
i
t
i
a
l
l
y
u
se
d
a
s
a
c
l
a
ss
i
f
i
e
r
t
o
i
d
e
n
t
i
f
y
p
a
t
t
e
r
n
s
a
n
d
e
x
t
r
a
c
t
f
e
a
t
u
r
e
s fr
o
m
t
h
e
d
a
t
a
se
t
,
w
h
i
c
h
a
r
e
t
h
e
n
u
se
d
i
n
c
o
n
j
u
n
c
t
i
o
n
w
i
t
h
G
R
U
-
e
x
t
r
a
c
t
e
d
f
e
a
t
u
r
e
s t
o
e
n
h
a
n
c
e
t
h
e
f
i
n
a
l
m
o
d
e
l
’
s
p
r
e
d
i
c
t
i
v
e
a
c
c
u
r
a
c
y
.
R
Q
2
H
o
w
d
o
e
s
t
h
e
G
R
U
m
o
d
e
l
c
o
n
t
r
i
b
u
t
e
t
o
t
h
e
d
e
t
e
c
t
i
o
n
p
r
o
c
e
ss?
Th
e
G
R
U
m
o
d
e
l
p
r
o
c
e
sses
a
n
d
t
o
k
e
n
i
z
e
s
o
p
c
o
d
e
se
q
u
e
n
c
e
s fr
o
m
E
t
h
e
r
e
u
m
smar
t
c
o
n
t
r
a
c
t
s
,
c
a
p
t
u
r
i
n
g
se
q
u
e
n
t
i
a
l
d
e
p
e
n
d
e
n
c
i
e
s
a
n
d
e
x
t
r
a
c
t
i
n
g
r
e
l
e
v
a
n
t
f
e
a
t
u
r
e
s
,
w
h
i
c
h
a
r
e
t
h
e
n
c
o
mb
i
n
e
d
w
i
t
h
X
G
B
o
o
s
t
f
e
a
t
u
r
e
s f
o
r
i
m
p
r
o
v
e
d
d
e
t
e
c
t
i
o
n
.
R
Q
3
H
o
w
d
o
e
s
a
d
v
a
n
c
e
d
f
e
a
t
u
r
e
e
n
g
i
n
e
e
r
i
n
g
i
m
p
r
o
v
e
P
o
n
z
i
sch
e
me
d
e
t
e
c
t
i
o
n
?
A
d
v
a
n
c
e
d
f
e
a
t
u
r
e
e
n
g
i
n
e
e
r
i
n
g
e
n
h
a
n
c
e
s d
e
t
e
c
t
i
o
n
b
y
e
x
t
r
a
c
t
i
n
g
a
n
d
c
o
m
b
i
n
i
n
g
r
e
l
e
v
a
n
t
f
e
a
t
u
r
e
s fr
o
m
d
i
f
f
e
r
e
n
t
mo
d
e
l
s,
a
l
l
o
w
i
n
g
t
h
e
h
y
b
r
i
d
m
o
d
e
l
t
o
b
e
t
t
e
r
c
a
p
t
u
r
e
t
h
e
c
h
a
r
a
c
t
e
r
i
s
t
i
c
s
o
f
P
o
n
z
i
s
c
h
e
mes
,
l
e
a
d
i
n
g
t
o
i
m
p
r
o
v
e
d
p
r
e
d
i
c
t
i
v
e
a
c
c
u
r
a
c
y
.
R
Q
4
W
h
a
t
i
s
t
h
e
p
o
t
e
n
t
i
a
l
i
m
p
a
c
t
o
f
t
h
i
s res
e
a
r
c
h
o
n
t
h
e
b
r
o
a
d
e
r
f
i
e
l
d
o
f
f
r
a
u
d
d
e
t
e
c
t
i
o
n
?
Th
i
s
r
e
se
a
r
c
h
h
a
s
t
h
e
p
o
t
e
n
t
i
a
l
t
o
si
g
n
i
f
i
c
a
n
t
l
y
i
mp
r
o
v
e
f
r
a
u
d
d
e
t
e
c
t
i
o
n
i
n
b
l
o
c
k
c
h
a
i
n
e
n
v
i
r
o
n
m
e
n
t
s,
o
f
f
e
r
i
n
g
a
r
o
b
u
st
,
a
c
c
u
r
a
t
e
,
a
n
d
sc
a
l
a
b
l
e
so
l
u
t
i
o
n
t
h
a
t
c
a
n
b
e
a
d
a
p
t
e
d
t
o
v
a
r
i
o
u
s
t
y
p
e
s
o
f
f
i
n
a
n
c
i
a
l
f
r
a
u
d
b
e
y
o
n
d
P
o
n
z
i
sc
h
e
m
e
s.
5.
CO
NCLU
SI
O
N
Po
n
zi
s
ch
em
es
ar
e
d
ec
ep
tiv
e
f
in
an
cial
o
p
er
atio
n
s
th
at
p
r
o
m
is
e
h
ig
h
r
etu
r
n
s
with
litt
le
r
is
k
,
o
f
ten
lead
in
g
to
s
ig
n
if
ican
t
f
i
n
an
ci
al
lo
s
s
es
wh
en
th
ey
co
llap
s
e.
Dete
ctin
g
s
u
ch
s
ch
em
es,
es
p
ec
ially
with
in
th
e
co
m
p
lex
an
d
ev
o
l
v
in
g
lan
d
s
ca
p
e
o
f
b
lo
c
k
ch
ain
tec
h
n
o
lo
g
y
,
r
em
ain
s
a
s
ig
n
if
ican
t
ch
alle
n
g
e.
I
n
r
esp
o
n
s
e
to
th
is
ch
allen
g
e,
o
u
r
r
esear
ch
in
tr
o
d
u
ce
d
a
h
y
b
r
i
d
m
ac
h
in
e
le
ar
n
in
g
ap
p
r
o
ac
h
th
at
co
m
b
in
es
th
e
s
tr
en
g
th
s
o
f
XGBo
o
s
t
an
d
GR
U
m
o
d
els,
co
u
p
led
with
ad
v
a
n
ce
d
f
ea
tu
r
e
en
g
in
ee
r
in
g
,
to
im
p
r
o
v
e
th
e
d
etec
tio
n
o
f
Po
n
zi
s
ch
em
es
in
E
th
e
r
eu
m
s
m
ar
t
co
n
tr
ac
ts
.
Ou
r
m
eth
o
d
o
lo
g
y
b
eg
an
b
y
lev
er
a
g
in
g
XGBo
o
s
t
to
id
en
tif
y
in
itial
p
atter
n
s
an
d
ex
tr
ac
t
r
elev
an
t
f
ea
tu
r
es.
Simu
ltan
eo
u
s
ly
,
we
u
s
ed
a
GR
U
m
o
d
el
to
p
r
o
ce
s
s
o
p
co
d
e
s
eq
u
e
n
ce
s
f
r
o
m
s
m
ar
t
c
o
n
tr
ac
ts
,
ex
t
r
ac
tin
g
s
eq
u
en
tial
f
ea
t
u
r
es
th
at
ca
p
tu
r
e
t
h
e
in
tr
icac
ies
o
f
t
r
an
s
ac
tio
n
p
atter
n
s
.
B
y
in
teg
r
atin
g
th
e
f
ea
t
u
r
es
f
r
o
m
b
o
th
m
o
d
els,
we
tr
ain
e
d
a
f
in
al
XGBo
o
s
t
class
if
ier
th
at
d
em
o
n
s
tr
ated
s
u
p
er
io
r
p
er
f
o
r
m
an
ce
co
m
p
a
r
ed
to
tr
a
d
itio
n
al
m
eth
o
d
s
.
T
h
e
h
y
b
r
id
m
o
d
el
ac
h
ie
v
ed
a
n
i
m
p
r
ess
iv
e
d
etec
tio
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
I
SS
N:
2252
-
8
7
7
6
A
h
yb
r
id
ma
ch
in
e
lea
r
n
in
g
a
p
p
r
o
a
ch
fo
r
imp
r
o
ve
d
p
o
n
z
i sch
eme
d
etec
tio
n
…
(
F
a
h
a
d
Ho
s
s
a
in
)
57
ac
cu
r
ac
y
o
f
9
6
.
8
3
%,
alo
n
g
with
s
tr
o
n
g
p
r
ec
is
io
n
,
r
ec
all,
an
d
F1
-
s
co
r
e
m
etr
ics,
s
h
o
wca
s
i
n
g
its
r
o
b
u
s
tn
ess
in
id
en
tify
in
g
f
r
au
d
u
len
t
ac
tiv
ities
.
T
h
e
r
esu
lts
o
f
o
u
r
s
tu
d
y
h
ig
h
lig
h
t
th
e
ef
f
ec
tiv
e
n
ess
o
f
co
m
b
in
in
g
m
ac
h
in
e
lear
n
in
g
m
o
d
els
with
s
o
p
h
is
ti
ca
ted
f
ea
tu
r
e
en
g
i
n
ee
r
in
g
to
ta
ck
le
th
e
co
m
p
lex
ities
o
f
Po
n
zi
s
ch
em
e
d
etec
tio
n
.
T
h
is
ap
p
r
o
ac
h
n
o
t
o
n
ly
ad
d
r
ess
es
th
e
lim
itatio
n
s
o
f
p
r
ev
io
u
s
m
eth
o
d
s
b
u
t
also
s
ets
a
n
ew
s
tan
d
ar
d
f
o
r
ac
cu
r
ac
y
an
d
r
eliab
ilit
y
in
f
r
a
u
d
d
etec
tio
n
with
in
b
lo
c
k
ch
ai
n
en
v
ir
o
n
m
en
ts
.
Fu
tu
r
e
wo
r
k
co
u
ld
f
u
r
t
h
er
r
ef
in
e
th
is
m
eth
o
d
o
lo
g
y
,
e
x
p
lo
r
in
g
ad
d
itio
n
al
m
o
d
el
s
an
d
ex
p
an
d
in
g
its
ap
p
licatio
n
to
o
th
er
f
o
r
m
s
o
f
f
in
an
cial
f
r
au
d
,
p
o
ten
tially
b
r
o
ad
e
n
in
g
t
h
e
im
p
ac
t o
f
o
u
r
r
esear
ch
o
n
t
h
e
b
r
o
a
d
er
f
ield
o
f
f
r
au
d
d
etec
tio
n
.
ACK
NO
WL
E
DG
E
M
E
NT
S
T
h
is
r
esear
ch
was f
u
n
d
e
d
b
y
W
o
o
s
o
n
g
Un
iv
er
s
ity
Aca
d
em
i
c
R
esear
ch
2
0
2
4
.
RE
F
E
R
E
NC
E
S
[
1
]
M
.
A
r
t
z
r
o
u
n
i
,
“
T
h
e
ma
t
h
e
ma
t
i
c
s
o
f
P
o
n
z
i
s
c
h
e
mes
,
”
M
a
t
h
e
m
a
t
i
c
a
l
S
o
c
i
a
l
S
c
i
e
n
c
e
s
,
v
o
l
.
5
8
,
n
o
.
2
,
p
p
.
1
9
0
–
2
0
1
,
2
0
0
9
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
m
a
t
h
s
o
c
sc
i
.
2
0
0
9
.
0
5
.
0
0
3
.
[
2
]
M
.
B
a
r
t
o
l
e
t
t
i
,
B
.
P
e
s,
a
n
d
S
.
S
e
r
u
si
,
“
D
a
t
a
m
i
n
i
n
g
f
o
r
d
e
t
e
c
t
i
n
g
b
i
t
c
o
i
n
p
o
n
z
i
sc
h
e
m
e
s,”
i
n
2
0
1
8
C
ry
p
t
o
Va
l
l
e
y
C
o
n
f
e
re
n
c
e
o
n
Bl
o
c
k
c
h
a
i
n
T
e
c
h
n
o
l
o
g
y
(
C
V
C
BT)
,
Ju
n
.
2
0
1
8
,
p
p
.
7
5
–
8
4
,
d
o
i
:
1
0
.
1
1
0
9
/
C
V
C
B
T.
2
0
1
8
.
0
0
0
1
4
.
[
3
]
M
.
V
a
s
e
k
a
n
d
T.
M
o
o
r
e
,
“
A
n
a
l
y
z
i
n
g
t
h
e
b
i
t
c
o
i
n
p
o
n
z
i
sc
h
e
me
e
c
o
s
y
st
e
m,”
i
n
L
e
c
t
u
re
N
o
t
e
s
i
n
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
(
i
n
c
l
u
d
i
n
g
su
b
s
e
ri
e
s Le
c
t
u
r
e
N
o
t
e
s
i
n
Ar
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
L
e
c
t
u
r
e
N
o
t
e
s
i
n
B
i
o
i
n
f
o
rm
a
t
i
c
s)
,
v
o
l
.
1
0
9
5
8
LN
C
S
,
2
0
1
9
,
p
p
.
1
0
1
–
1
1
2
.
[
4
]
Y
.
B
o
s
h
m
a
f
,
C
.
El
v
i
t
i
g
a
l
a
,
H
.
A
l
J
a
w
a
h
e
r
i
,
P
.
W
i
j
e
se
k
e
r
a
,
a
n
d
M
.
A
l
S
a
b
a
h
,
“
I
n
v
e
s
t
i
g
a
t
i
n
g
M
M
M
p
o
n
z
i
sc
h
e
m
e
o
n
b
i
t
c
o
i
n
,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
1
5
t
h
A
C
M
Asi
a
C
o
n
f
e
re
n
c
e
o
n
C
o
m
p
u
t
e
r
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
s
S
e
c
u
ri
t
y
,
O
c
t
.
2
0
2
0
,
p
p
.
5
1
9
–
5
3
0
,
d
o
i
:
1
0
.
1
1
4
5
/
3
3
2
0
2
6
9
.
3
3
8
4
7
1
9
.
[
5
]
M
.
S
p
a
g
n
u
o
l
o
,
F
.
M
a
g
g
i
,
a
n
d
S
.
Z
a
n
e
r
o
,
“
B
i
t
I
o
d
i
n
e
:
e
x
t
r
a
c
t
i
n
g
i
n
t
e
l
l
i
g
e
n
c
e
f
r
o
m
t
h
e
b
i
t
c
o
i
n
n
e
t
w
o
r
k
,
”
i
n
L
e
c
t
u
re
N
o
t
e
s
i
n
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
(
i
n
c
l
u
d
i
n
g
su
b
s
e
ri
e
s
L
e
c
t
u
re
N
o
t
e
s
i
n
Ar
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
L
e
c
t
u
r
e
N
o
t
e
s
i
n
B
i
o
i
n
f
o
rm
a
t
i
c
s)
,
v
o
l
.
8
4
3
7
,
2
0
1
4
,
p
p
.
4
5
7
–
4
6
8
.
[
6
]
K
.
L
i
a
o
,
Z
.
Z
h
a
o
,
A
.
D
o
u
p
e
,
a
n
d
G
.
-
J.
A
h
n
,
“
B
e
h
i
n
d
c
l
o
s
e
d
d
o
o
r
s:
m
e
a
su
r
e
m
e
n
t
a
n
d
a
n
a
l
y
s
i
s
o
f
C
r
y
p
t
o
L
o
c
k
e
r
r
a
n
s
o
ms
i
n
B
i
t
c
o
i
n
,
”
i
n
2
0
1
6
AP
WG
S
y
m
p
o
si
u
m
o
n
El
e
c
t
ro
n
i
c
C
r
i
m
e
R
e
se
a
rc
h
(
e
C
ri
m
e
)
,
J
u
n
.
2
0
1
6
,
v
o
l
.
2
0
1
6
-
J
u
n
e
,
p
p
.
1
–
1
3
,
d
o
i
:
1
0
.
1
1
0
9
/
EC
R
I
M
E.
2
0
1
6
.
7
4
8
7
9
3
8
.
[
7
]
S
.
B
i
s
t
a
r
e
l
l
i
,
M
.
P
a
r
r
o
c
c
i
n
i
,
a
n
d
F
.
S
a
n
t
i
n
i
,
“
V
i
s
u
a
l
i
s
i
n
g
b
i
t
c
o
i
n
f
l
o
w
s
o
f
r
a
n
s
o
mw
a
r
e
:
W
a
n
n
a
C
r
y
o
n
e
w
e
e
k
l
a
t
e
r
,
”
C
EU
R
Wo
r
k
s
h
o
p
Pro
c
e
e
d
i
n
g
s
,
v
o
l
.
2
0
5
8
,
2
0
1
8
.
[
8
]
C
.
B
r
e
n
i
g
,
R
.
A
c
c
o
r
si
,
a
n
d
G
.
M
ö
l
l
e
r
,
“
E
c
o
n
o
mi
c
a
n
a
l
y
si
s
o
f
c
r
y
p
t
o
c
u
r
r
e
n
c
y
b
a
c
k
e
d
m
o
n
e
y
l
a
u
n
d
e
r
i
n
g
,
”
2
3
r
d
E
u
r
o
p
e
a
n
C
o
n
f
e
re
n
c
e
o
n
I
n
f
o
rm
a
t
i
o
n
S
y
st
e
m
s
,
EC
I
S
2
0
1
5
,
v
o
l
.
2
0
1
5
-
M
a
y
,
2
0
1
5
.
[
9
]
M
.
M
o
ser,
R
.
B
o
h
me,
a
n
d
D
.
B
r
e
u
k
e
r
,
“
A
n
i
n
q
u
i
r
y
i
n
t
o
m
o
n
e
y
l
a
u
n
d
e
r
i
n
g
t
o
o
l
s
i
n
t
h
e
B
i
t
c
o
i
n
e
c
o
sy
st
e
m,”
i
n
2
0
1
3
AP
WG e
C
r
i
m
e
Re
se
a
rc
h
e
rs
S
u
m
m
i
t
,
S
e
p
.
2
0
1
3
,
p
p
.
1
–
1
4
,
d
o
i
:
1
0
.
1
1
0
9
/
e
C
R
S
.
2
0
1
3
.
6
8
0
5
7
8
0
.
[
1
0
]
T.
M
o
o
r
e
,
J.
H
a
n
,
a
n
d
R
.
C
l
a
y
t
o
n
,
“
Th
e
p
o
s
t
m
o
d
e
r
n
p
o
n
z
i
sc
h
e
m
e
:
e
m
p
i
r
i
c
a
l
a
n
a
l
y
si
s
o
f
h
i
g
h
-
y
i
e
l
d
i
n
v
e
st
men
t
p
r
o
g
r
a
ms,
”
L
e
c
t
u
re
N
o
t
e
s
i
n
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
(
i
n
c
l
u
d
i
n
g
s
u
b
ser
i
e
s
L
e
c
t
u
r
e
N
o
t
e
s
i
n
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
L
e
c
t
u
r
e
N
o
t
e
s
i
n
Bi
o
i
n
f
o
rm
a
t
i
c
s)
,
v
o
l
.
7
3
9
7
LN
C
S
,
p
p
.
4
1
–
5
6
,
2
0
1
2
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
6
4
2
-
3
2
9
4
6
-
3
_
4
.
[
1
1
]
M
.
V
a
se
k
a
n
d
T.
M
o
o
r
e
,
“
T
h
e
r
e
’
s
n
o
f
r
e
e
l
u
n
c
h
,
e
v
e
n
u
s
i
n
g
b
i
t
c
o
i
n
:
t
r
a
c
k
i
n
g
t
h
e
p
o
p
u
l
a
r
i
t
y
a
n
d
p
r
o
f
i
t
s
o
f
v
i
r
t
u
a
l
c
u
r
r
e
n
c
y
s
c
a
ms,
”
i
n
L
e
c
t
u
re
N
o
t
e
s
i
n
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
(
i
n
c
l
u
d
i
n
g
s
u
b
seri
e
s
L
e
c
t
u
re
N
o
t
e
s
i
n
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
L
e
c
t
u
re
N
o
t
e
s
i
n
Bi
o
i
n
f
o
rm
a
t
i
c
s)
,
v
o
l
.
8
9
7
5
,
2
0
1
5
,
p
p
.
4
4
–
6
1
.
[
1
2
]
T.
M
o
o
r
e
,
“
T
h
e
p
r
o
mi
se
a
n
d
p
e
r
i
l
s
o
f
d
i
g
i
t
a
l
c
u
r
r
e
n
c
i
e
s,”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
C
ri
t
i
c
a
l
I
n
f
ra
s
t
r
u
c
t
u
r
e
Pr
o
t
e
c
t
i
o
n
,
v
o
l
.
6
,
n
o
.
3
–
4
,
p
p
.
1
4
7
–
1
4
9
,
D
e
c
.
2
0
1
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
j
c
i
p
.
2
0
1
3
.
0
8
.
0
0
2
.
[
1
3
]
M
.
M
ö
ser,
R
.
B
ö
h
m
e
,
a
n
d
D
.
B
r
e
u
k
e
r
,
“
T
o
w
a
r
d
s
r
i
s
k
s
c
o
r
i
n
g
o
f
b
i
t
c
o
i
n
t
r
a
n
s
a
c
t
i
o
n
s,”
i
n
L
e
c
t
u
re
N
o
t
e
s
i
n
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
(
i
n
c
l
u
d
i
n
g
s
u
b
ser
i
e
s L
e
c
t
u
r
e
N
o
t
e
s
i
n
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
L
e
c
t
u
re
N
o
t
e
s i
n
Bi
o
i
n
f
o
rm
a
t
i
c
s)
,
v
o
l
.
8
4
3
8
,
2
0
1
4
,
p
p
.
1
6
–
3
2
.
[
1
4
]
L.
V
i
n
e
t
a
n
d
A
.
Z
h
e
d
a
n
o
v
,
“
A
‘
m
i
ss
i
n
g
’
f
a
m
i
l
y
o
f
c
l
a
ssi
c
a
l
o
r
t
h
o
g
o
n
a
l
p
o
l
y
n
o
m
i
a
l
s,”
Br
o
o
k
l
y
n
J
o
u
r
n
a
l
o
f
I
n
t
e
rn
a
t
i
o
n
a
l
L
a
w
,
v
o
l
.
3
9
,
p
.
8
2
9
,
N
o
v
.
2
0
1
0
,
d
o
i
:
1
0
.
1
0
8
8
/
1
7
5
1
-
8
1
1
3
/
4
4
/
8
/
0
8
5
2
0
1
.
[
1
5
]
H
.
D
a
r
maw
a
n
,
M
.
Y
u
l
i
a
n
a
,
a
n
d
M
.
Z.
S
a
ms
o
n
o
H
a
d
i
,
“
G
R
U
a
n
d
X
G
B
o
o
st
p
e
r
f
o
r
ma
n
c
e
w
i
t
h
h
y
p
e
r
p
a
r
a
m
e
t
e
r
t
u
n
i
n
g
u
s
i
n
g
G
r
i
d
S
e
a
r
c
h
C
V
a
n
d
b
a
y
e
s
i
a
n
o
p
t
i
mi
z
a
t
i
o
n
o
n
a
n
I
o
T
-
b
a
s
e
d
w
e
a
t
h
e
r
p
r
e
d
i
c
t
i
o
n
sy
s
t
e
m
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
n
A
d
v
a
n
c
e
d
S
c
i
e
n
c
e
,
E
n
g
i
n
e
e
ri
n
g
a
n
d
I
n
f
o
rm
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
1
3
,
n
o
.
3
,
p
p
.
8
5
1
–
8
6
2
,
J
u
n
.
2
0
2
3
,
d
o
i
:
1
0
.
1
8
5
1
7
/
i
j
a
se
i
t
.
1
3
.
3
.
1
8
3
7
7
.
[
1
6
]
W
.
C
h
e
n
e
t
a
l
.
,
“
S
A
D
P
o
n
z
i
:
d
e
t
e
c
t
i
n
g
a
n
d
c
h
a
r
a
c
t
e
r
i
z
i
n
g
p
o
n
z
i
s
c
h
e
mes
i
n
e
t
h
e
r
e
u
m
sm
a
r
t
c
o
n
t
r
a
c
t
s,
”
P
ro
c
e
e
d
i
n
g
s
o
f
t
h
e
A
C
M
o
n
M
e
a
s
u
rem
e
n
t
a
n
d
An
a
l
y
si
s o
f
C
o
m
p
u
t
i
n
g
S
y
s
t
e
m
s
,
v
o
l
.
5
,
n
o
.
2
,
p
p
.
1
–
3
0
,
Ju
n
.
2
0
2
1
,
d
o
i
:
1
0
.
1
1
4
5
/
3
4
6
0
0
9
3
.
[
1
7
]
M
.
W
a
n
g
a
n
d
J.
H
u
a
n
g
,
“
D
e
t
e
c
t
i
n
g
e
t
h
e
r
e
u
m
p
o
n
z
i
sc
h
e
m
e
s
t
h
r
o
u
g
h
o
p
c
o
d
e
c
o
n
t
e
x
t
a
n
a
l
y
si
s
a
n
d
o
v
e
r
s
a
mp
l
i
n
g
-
b
a
s
e
d
A
d
a
B
o
o
s
t
a
l
g
o
r
i
t
h
m,
”
C
o
m
p
u
t
e
r
S
y
st
e
m
s
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
ri
n
g
,
v
o
l
.
4
7
,
n
o
.
1
,
p
p
.
1
0
2
3
–
1
0
4
2
,
2
0
2
3
,
d
o
i
:
1
0
.
3
2
6
0
4
/
c
s
se.
2
0
2
3
.
0
3
9
5
6
9
.
[
1
8
]
A
.
A
l
j
o
f
e
y
,
Q
.
Ji
a
n
g
,
a
n
d
Q
.
Q
u
,
“
A
su
p
e
r
v
i
se
d
l
e
a
r
n
i
n
g
m
o
d
e
l
f
o
r
d
e
t
e
c
t
i
n
g
p
o
n
z
i
c
o
n
t
r
a
c
t
s
i
n
e
t
h
e
r
e
u
m
b
l
o
c
k
c
h
a
i
n
,
”
i
n
C
o
m
m
u
n
i
c
a
t
i
o
n
s i
n
C
o
m
p
u
t
e
r
a
n
d
I
n
f
o
rm
a
t
i
o
n
S
c
i
e
n
c
e
,
v
o
l
.
1
5
6
3
C
C
I
S
,
2
0
2
2
,
p
p
.
6
5
7
–
6
7
2
.
[
1
9
]
A
.
A
l
j
o
f
e
y
,
A
.
R
a
s
o
o
l
,
Q
.
J
i
a
n
g
,
a
n
d
Q
.
Q
u
,
“
A
f
e
a
t
u
r
e
-
b
a
s
e
d
r
o
b
u
st
met
h
o
d
f
o
r
a
b
n
o
r
m
a
l
c
o
n
t
r
a
c
t
s
d
e
t
e
c
t
i
o
n
i
n
e
t
h
e
r
e
u
m
b
l
o
c
k
c
h
a
i
n
,
”
El
e
c
t
ro
n
i
c
s
,
v
o
l
.
1
1
,
n
o
.
1
8
,
p
.
2
9
3
7
,
S
e
p
.
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
e
l
e
c
t
r
o
n
i
c
s1
1
1
8
2
9
3
7
.
[
2
0
]
C
.
X
u
,
R
.
X
i
o
n
g
,
X
.
S
h
e
n
,
L
.
Z
h
u
,
a
n
d
X
.
Zh
a
n
g
,
“
H
o
w
t
o
f
i
n
d
a
b
i
t
c
o
i
n
m
i
x
e
r
:
a
d
u
a
l
e
n
s
e
m
b
l
e
m
o
d
e
l
f
o
r
b
i
t
c
o
i
n
m
i
x
i
n
g
s
e
r
v
i
c
e
d
e
t
e
c
t
i
o
n
,
”
I
EEE
I
n
t
e
r
n
e
t
o
f
T
h
i
n
g
s
J
o
u
r
n
a
l
,
v
o
l
.
1
0
,
n
o
.
1
9
,
p
p
.
1
7
2
2
0
–
1
7
2
3
0
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
JI
O
T.
2
0
2
3
.
3
2
7
5
1
5
8
.
[
2
1
]
G
.
I
b
b
a
,
G
.
A
.
P
i
e
r
r
o
,
a
n
d
M
.
D
i
F
r
a
n
c
e
sc
o
,
“
E
v
a
l
u
a
t
i
n
g
m
a
c
h
i
n
e
-
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s
f
o
r
d
e
t
e
c
t
i
n
g
smar
t
p
o
n
z
i
s
c
h
e
m
e
s,
”
i
n
2
0
2
1
I
EEE/
A
C
M
4
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
W
o
rks
h
o
p
o
n
Em
e
r
g
i
n
g
T
re
n
d
s
i
n
S
o
f
t
w
a
r
e
E
n
g
i
n
e
e
ri
n
g
f
o
r
Bl
o
c
k
c
h
a
i
n
(
WE
T
S
EB)
,
M
a
y
2
0
2
1
,
p
p
.
3
4
–
4
0
,
d
o
i
:
1
0
.
1
1
0
9
/
W
ETSE
B
5
2
5
5
8
.
2
0
2
1
.
0
0
0
1
2
.
[
2
2
]
S
.
Y
u
,
J.
Ji
n
,
Y
.
X
i
e
,
J.
S
h
e
n
,
a
n
d
Q
.
X
u
a
n
,
“
P
o
n
z
i
s
c
h
e
me
d
e
t
e
c
t
i
o
n
i
n
e
t
h
e
r
e
u
m
t
r
a
n
s
a
c
t
i
o
n
n
e
t
w
o
r
k
,
”
i
n
C
o
m
m
u
n
i
c
a
t
i
o
n
s
i
n
C
o
m
p
u
t
e
r
a
n
d
I
n
f
o
rm
a
t
i
o
n
S
c
i
e
n
c
e
,
v
o
l
.
1
4
9
0
C
C
I
S
,
2
0
2
1
,
p
p
.
1
7
5
–
1
8
6
.
[
2
3
]
Y
.
Z
h
a
n
g
,
W
.
Y
u
,
Z.
Li
,
S
.
R
a
z
a
,
a
n
d
H
.
C
a
o
,
“
D
e
t
e
c
t
i
n
g
e
t
h
e
r
e
u
m
p
o
n
z
i
sc
h
e
mes
b
a
se
d
o
n
i
mp
r
o
v
e
d
L
i
g
h
t
G
B
M
a
l
g
o
r
i
t
h
m
,
”
I
EEE
T
r
a
n
s
a
c
t
i
o
n
s
o
n
C
o
m
p
u
t
a
t
i
o
n
a
l
S
o
c
i
a
l
S
y
s
t
e
m
s
,
v
o
l
.
9
,
n
o
.
2
,
p
p
.
6
2
4
–
6
3
7
,
A
p
r
.
2
0
2
2
,
d
o
i
:
1
0
.
1
1
0
9
/
T
C
S
S
.
2
0
2
1
.
3
0
8
8
1
4
5
.
[
2
4
]
S
.
F
a
n
,
S
.
F
u
,
H
.
X
u
,
a
n
d
C
.
Z
h
u
,
“
E
x
p
o
s
e
y
o
u
r
ma
s
k
:
sm
a
r
t
p
o
n
z
i
sc
h
e
m
e
s
d
e
t
e
c
t
i
o
n
o
n
b
l
o
c
k
c
h
a
i
n
,
”
i
n
2
0
2
0
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
i
n
t
C
o
n
f
e
re
n
c
e
o
n
N
e
u
r
a
l
N
e
t
w
o
r
k
s (I
J
C
N
N
)
,
Ju
l
.
2
0
2
0
,
p
p
.
1
–
7
,
d
o
i
:
1
0
.
1
1
0
9
/
I
JC
N
N
4
8
6
0
5
.
2
0
2
0
.
9
2
0
7
1
4
3
.
[
2
5
]
Y
.
Lo
u
,
Y
.
Zh
a
n
g
,
a
n
d
S
.
C
h
e
n
,
“
P
o
n
z
i
c
o
n
t
r
a
c
t
s
d
e
t
e
c
t
i
o
n
b
a
s
e
d
o
n
i
m
p
r
o
v
e
d
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
i
n
2
0
2
0
I
EE
E
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
S
e
r
v
i
c
e
s
C
o
m
p
u
t
i
n
g
(
S
C
C
)
,
N
o
v
.
2
0
2
0
,
p
p
.
3
5
3
–
3
6
0
,
d
o
i
:
1
0
.
1
1
0
9
/
S
C
C
4
9
8
3
2
.
2
0
2
0
.
0
0
0
5
3
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
7
6
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
,
Vo
l.
1
4
,
No
.
1
,
A
p
r
il
20
2
5
:
50
-
58
58
[
2
6
]
Z.
Z
h
e
n
g
,
W
.
C
h
e
n
,
Z.
Z
h
o
n
g
,
Z.
C
h
e
n
,
a
n
d
Y
.
L
u
,
“
S
e
c
u
r
i
n
g
t
h
e
E
t
h
e
r
e
u
m
f
r
o
m
smar
t
p
o
n
z
i
s
c
h
e
mes
:
i
d
e
n
t
i
f
i
c
a
t
i
o
n
u
s
i
n
g
st
a
t
i
c
f
e
a
t
u
r
e
s,
”
AC
M
T
r
a
n
s
a
c
t
i
o
n
s
o
n
S
o
f
t
w
a
re
E
n
g
i
n
e
e
r
i
n
g
a
n
d
Me
t
h
o
d
o
l
o
g
y
,
v
o
l
.
3
2
,
n
o
.
5
,
p
p
.
1
–
2
8
,
S
e
p
.
2
0
2
3
,
d
o
i
:
1
0
.
1
1
4
5
/
3
5
7
1
8
4
7
.
[
2
7
]
S
.
Z
h
a
n
g
,
T.
La
n
,
L.
W
a
n
g
,
S
.
X
u
,
a
n
d
W
.
S
h
a
o
,
“
E
t
h
e
r
e
u
m
p
o
n
z
i
s
c
h
e
me
d
e
t
e
c
t
i
o
n
b
a
se
d
o
n
P
D
-
S
E
C
R
,
”
S
e
c
u
r
i
t
y
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
N
e
t
w
o
rk
s
,
v
o
l
.
2
0
2
2
,
p
p
.
1
–
1
5
,
S
e
p
.
2
0
2
2
,
d
o
i
:
1
0
.
1
1
5
5
/
2
0
2
2
/
2
3
1
6
3
1
0
.
[
2
8
]
R
.
Li
a
n
g
e
t
a
l
.
,
“
P
o
n
z
i
G
u
a
r
d
:
d
e
t
e
c
t
i
n
g
p
o
n
z
i
s
c
h
e
mes
o
n
e
t
h
e
r
e
u
m
w
i
t
h
c
o
n
t
r
a
c
t
r
u
n
t
i
me
b
e
h
a
v
i
o
r
g
r
a
p
h
(
C
R
B
G
)
,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
I
EEE
/
A
C
M
4
6
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
S
o
f
t
w
a
re
E
n
g
i
n
e
e
r
i
n
g
,
F
e
b
.
2
0
2
4
,
p
p
.
1
–
1
2
,
d
o
i
:
1
0
.
1
1
4
5
/
3
5
9
7
5
0
3
.
3
6
2
3
3
1
8
.
[
2
9
]
I
.
J.
O
n
u
,
A
.
E.
O
mo
l
a
r
a
,
M
.
A
l
a
w
i
d
a
,
O
.
I
.
A
b
i
o
d
u
n
,
a
n
d
A
.
A
l
a
b
d
u
l
t
i
f
,
“
D
e
t
e
c
t
i
o
n
o
f
P
o
n
z
i
sc
h
e
m
e
o
n
Et
h
e
r
e
u
m
u
s
i
n
g
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
ms,”
S
c
i
e
n
t
i
f
i
c
R
e
p
o
r
t
s
,
v
o
l
.
1
3
,
n
o
.
1
,
p
.
1
8
4
0
3
,
O
c
t
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
3
8
/
s4
1
5
9
8
-
0
2
3
-
4
5
2
7
5
-
0.
[
3
0
]
P
o
l
a
r
w
o
l
f
,
“
P
o
n
z
i
sc
h
e
m
e
c
o
n
t
r
a
c
t
s
o
n
e
t
h
e
r
e
u
m,
”
2
0
2
0
,
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s
:
/
/
w
w
w
.
k
a
g
g
l
e
.
c
o
m
/
d
a
t
a
se
t
s
/
p
o
l
a
r
w
o
l
f
/
p
o
n
z
i
-
sch
e
me
-
c
o
n
t
r
a
c
t
s
-
on
-
e
t
h
e
r
e
u
m
?
sel
e
c
t
=
P
o
n
z
i
c
o
n
t
r
a
c
t
s
.
c
s
v
.
[
3
1
]
B
.
B
i
n
J
i
a
a
n
d
M
.
L.
Zh
a
n
g
,
“
M
u
l
t
i
-
d
i
me
n
s
i
o
n
a
l
c
l
a
ssi
f
i
c
a
t
i
o
n
v
i
a
sp
a
r
s
e
l
a
b
e
l
e
n
c
o
d
i
n
g
,
”
Pr
o
c
e
e
d
i
n
g
s
o
f
Ma
c
h
i
n
e
L
e
a
r
n
i
n
g
Re
se
a
rc
h
,
v
o
l
.
1
3
9
,
p
p
.
4
9
1
7
–
4
9
2
6
,
2
0
2
1
.
[
3
2
]
A
.
Q
.
A
b
d
u
l
g
h
a
n
i
,
O
.
N
.
U
c
a
n
,
a
n
d
K
.
M
.
A
.
A
l
h
e
e
t
i
,
“
C
r
e
d
i
t
c
a
r
d
f
r
a
u
d
d
e
t
e
c
t
i
o
n
u
s
i
n
g
X
G
B
o
o
s
t
a
l
g
o
r
i
t
h
m,”
i
n
2
0
2
1
1
4
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
D
e
v
e
l
o
p
m
e
n
t
s
i
n
e
S
y
s
t
e
m
s
En
g
i
n
e
e
r
i
n
g
(
D
e
S
E)
,
D
e
c
.
2
0
2
1
,
v
o
l
.
2
0
2
1
-
D
e
c
e
m,
p
p
.
4
8
7
–
4
9
2
,
d
o
i
:
1
0
.
1
1
0
9
/
D
e
S
E
5
4
2
8
5
.
2
0
2
1
.
9
7
1
9
5
8
0
.
[
3
3
]
B
.
C
u
i
a
n
d
G
.
W
a
n
g
,
“
P
o
n
z
i
s
c
h
e
me
d
e
t
e
c
t
i
o
n
b
a
se
d
o
n
C
N
N
a
n
d
B
i
G
R
U
c
o
m
b
i
n
e
d
w
i
t
h
a
t
t
e
n
t
i
o
n
m
e
c
h
a
n
i
s
m,”
i
n
2
0
2
4
2
7
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
C
o
m
p
u
t
e
r
S
u
p
p
o
r
t
e
d
C
o
o
p
e
r
a
t
i
v
e
W
o
rk
i
n
D
e
si
g
n
(
C
S
C
WD)
,
M
a
y
2
0
2
4
,
p
p
.
1
8
5
2
–
1
8
5
7
,
d
o
i
:
1
0
.
1
1
0
9
/
C
S
C
W
D
6
1
4
1
0
.
2
0
2
4
.
1
0
5
8
0
6
9
2
.
[
3
4
]
Z.
F
e
r
d
o
u
sh
,
B
.
N
.
M
a
h
mu
d
,
A
.
C
h
a
k
r
a
b
a
r
t
y
,
a
n
d
J
.
U
d
d
i
n
,
“
A
sh
o
r
t
-
t
e
r
m h
y
b
r
i
d
f
o
r
e
c
a
s
t
i
n
g
m
o
d
e
l
f
o
r
t
i
m
e
ser
i
e
s
e
l
e
c
t
r
i
c
a
l
-
l
o
a
d
d
a
t
a
u
s
i
n
g
r
a
n
d
o
m
f
o
r
e
st
a
n
d
b
i
d
i
r
e
c
t
i
o
n
a
l
l
o
n
g
s
h
o
r
t
-
t
e
r
m
me
mo
r
y
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
E
l
e
c
t
r
i
c
a
l
a
n
d
C
o
m
p
u
t
e
r
En
g
i
n
e
e
ri
n
g
(
I
J
E
C
E)
,
v
o
l
.
1
1
,
n
o
.
1
,
p
.
7
6
3
,
F
e
b
.
2
0
2
1
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
c
e
.
v
1
1
i
1
.
p
p
7
6
3
-
7
7
1
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Fa
h
a
d
H
o
ss
a
in
is
a
g
ra
d
u
a
te
stu
d
e
n
t
at
th
e
De
p
a
rtme
n
t
of
C
o
m
p
u
ter
a
n
d
In
fo
rm
a
ti
o
n
S
c
ien
c
e
in
t
h
e
F
lo
r
i
d
a
In
tern
a
ti
o
n
a
l
Un
i
v
e
rsity
(F
IU),
M
iam
i,
F
lo
r
id
a
,
Un
i
ted
S
tate
s
o
f
Am
e
rica
.
P
r
io
r
to
sta
rti
n
g
M
.
S
.
C
a
t
F
IU
,
Ho
ss
a
in
c
o
m
p
lete
d
h
is
Ba
c
h
e
lo
r
’
s
d
e
g
re
e
i
n
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
in
e
e
ri
n
g
fr
o
m
Bra
c
Un
iv
e
rsity
,
D
h
a
k
a
,
Ba
n
g
lad
e
sh
.
His
re
se
a
rc
h
in
tere
sts
in
c
l
u
d
e
fin
a
n
c
ial
fra
u
d
d
e
tec
ti
o
n
,
b
l
o
c
k
c
h
a
i
n
,
a
n
d
c
lo
u
d
n
e
tw
o
rk
s.
He
can
be
c
o
n
tac
ted
at
e
m
a
il
:
fh
o
ss
0
0
6
@fiu
.
e
d
u
.
Mehedi
H
a
sa
n
S
h
u
v
o
is
a
B.
S
c
.
En
g
in
e
e
r,
h
a
v
in
g
g
ra
d
u
a
ted
fro
m
th
e
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter
S
c
ie
n
c
e
a
n
d
En
g
in
e
e
rin
g
a
t
Dh
a
k
a
Un
i
v
e
rsity
o
f
En
g
in
e
e
rin
g
a
n
d
Tec
h
n
o
l
o
g
y
in
G
a
z
ip
u
r,
Ba
n
g
lad
e
sh
.
He
is
a
c
o
u
rteo
u
s
a
n
d
d
e
d
ica
ted
in
d
i
v
i
d
u
a
l
wit
h
stro
n
g
tec
h
n
ica
l
sk
il
ls,
p
a
ss
io
n
a
te
a
b
o
u
t
tac
k
l
in
g
c
h
a
ll
e
n
g
e
s
a
n
d
c
re
a
ti
n
g
e
n
g
a
g
in
g
,
in
f
o
rm
a
ti
v
e
c
o
n
ten
t.
He
h
a
s
th
re
e
y
e
a
rs
o
f
in
d
u
strial
e
x
p
e
rien
c
e
in
m
o
b
i
l
e
a
p
p
li
c
a
ti
o
n
d
e
v
e
lo
p
m
e
n
t
(An
d
ro
id
a
n
d
iOS
)
a
n
d
h
a
s
w
o
rk
e
d
fu
ll
-
ti
m
e
i
n
th
re
e
p
ro
fe
ss
o
rs
’
l
a
b
s
a
t
h
is
u
n
iv
e
rsit
y
,
wh
ich
fu
rth
e
r
re
fi
n
e
d
h
i
s
e
x
p
e
rti
se
.
H
is
re
se
a
rc
h
in
tere
sts
sp
a
n
fe
d
e
ra
ted
lea
rn
in
g
,
e
x
p
lain
a
b
le
a
rti
ficia
l
in
telli
g
e
n
c
e
,
c
o
m
p
u
ter
v
isio
n
,
e
d
g
e
c
o
m
p
u
ti
n
g
,
m
a
c
h
in
e
lea
rn
in
g
,
d
e
e
p
lea
rn
i
n
g
,
o
b
jec
t
d
e
tec
ti
o
n
,
a
n
d
wa
ter
a
n
d
e
n
v
iro
n
m
e
n
tal
su
sta
in
a
b
il
it
y
.
He
c
a
n
b
e
c
o
n
tac
ted
at
emai
l
:
m
e
h
e
d
ih
a
sa
n
sh
u
v
o
.
m
a
il
@g
m
a
il
.
c
o
m
.
J
ia
Ud
d
in
is
a
n
a
ss
istan
t
p
ro
fe
ss
o
r
in
t
h
e
AI
a
n
d
Big
Da
ta
De
p
a
rtme
n
t,
E
n
d
ic
o
tt
Co
ll
e
g
e
,
Wo
o
so
n
g
Un
iv
e
rsit
y
,
Ko
re
a
.
He
is
an
a
ss
o
c
iate
p
ro
fe
ss
o
r
(c
u
rre
n
tl
y
on
lea
v
e
)
in
th
e
CS
E
d
e
p
a
rtme
n
t
a
t
BRAC
Un
iv
e
rsity
,
Ba
n
g
lad
e
sh
.
He
re
c
e
iv
e
d
a
P
h
.
D.
d
e
g
re
e
(Co
m
p
u
ter
En
g
i
n
e
e
rin
g
)
fr
o
m
th
e
Un
i
v
e
rsi
t
y
of
U
lsa
n
,
S
o
u
th
Ko
re
a
in
Ja
n
u
a
ry
2
0
1
5
a
n
d
an
M
S
c
in
Tele
c
o
m
m
u
n
ica
ti
o
n
s
fr
o
m
Blek
in
g
e
In
stit
u
te
o
f
Tec
h
n
o
l
o
g
y
,
S
w
e
d
e
n
,
in
2
0
1
0
.
He
wa
s
a
m
e
m
b
e
r
o
f
th
e
S
e
lf
-
As
se
ss
m
e
n
t
Tea
m
(S
AC)
o
f
CS
E,
BRAC
U i
n
t
h
e
HEQEP
p
ro
jec
t
f
u
n
d
e
d
b
y
th
e
W
o
rld
Ba
n
k
a
n
d
t
h
e
Un
i
v
e
rsity
o
f
G
ra
n
t
Co
m
m
issio
n
Ba
n
g
lad
e
sh
i
n
2
0
1
6
–
2
0
1
7
.
His
re
se
a
rc
h
a
re
a
in
c
lu
d
e
s
fa
u
lt
d
iag
n
o
sis
u
sin
g
AI,
a
u
d
io
,
a
n
d
ima
g
e
p
r
o
c
e
ss
in
g
.
He
can
be
c
o
n
tac
ted
at
e
m
a
il
:
ji
a
.
u
d
d
in
@w
s
u
.
a
c
.
k
r
.
Evaluation Warning : The document was created with Spire.PDF for Python.