I
A
E
S
I
n
t
e
r
n
at
io
n
al
Jou
r
n
al
of
A
r
t
if
ic
ia
l
I
n
t
e
ll
ig
e
n
c
e
(
I
J
-
A
I
)
V
ol
.
10
, N
o.
4
,
D
e
c
e
m
be
r
202
1
, pp.
1091
~
1102
I
S
S
N
:
2252
-
8938
,
D
O
I
:
10.11591/
ij
a
i.
v
10
.i
4
.pp
1091
-
1102
1091
Jou
r
n
al
h
om
e
page
:
ht
tp
:
//
ij
ai
.
ia
e
s
c
or
e
.c
om
A
sys
t
e
m
at
i
c
l
i
t
e
r
a
t
u
r
e
r
e
vi
e
w
o
f
m
ac
h
i
n
e
l
e
ar
n
i
n
g
m
e
t
h
od
s i
n
p
r
e
d
i
c
t
i
n
g c
ou
r
t
d
e
c
i
si
on
s
N
u
r
A
q
il
ah
K
h
ad
i
j
ah
R
os
il
i
1
, N
oor
H
id
ayah
Z
ak
ar
ia
2
, R
oh
a
yan
t
i
H
as
s
an
3
, S
h
ah
r
e
e
n
K
as
im
4
,
F
ar
id
Z
am
an
i
C
h
e
R
os
e
5
, T
ol
e
S
u
t
ik
n
o
6
1,
2,
3
School o
f Comput
ing, Fac
ulty of
Engine
ering,
Unive
rsiti Te
knologi
Malays
ia
,
Johor, Malaysia
4
Faculty
of Comp
uter Sc
ience
and I
nforma
tion Syste
m, Unive
rsiti Tu
n Husse
in Onn
,
Johor, Malaysia
5
School o
f Math
ematic
al Scie
nces,
Unive
rs
i
ti Sains
Malaysia
,
Penang
,
Malaysia
6
Department of Electical En
gineering, Universit
as Ahmad Dahlan, Yogyakarta,
Indonesia
A
r
t
ic
le
I
n
f
o
A
B
S
T
R
A
C
T
A
r
ti
c
le
h
is
to
r
y
:
R
e
c
e
iv
e
d
J
un 12, 20
21
R
e
vi
s
e
d
S
e
p 30
, 20
21
A
c
c
e
pt
e
d
O
c
t
13
, 20
21
Envisaging
legal
cases’
outcomes
can
assist
the
judicial
decision
-
making
process.
Prediction
is
possible
in
various
cases,
such
as
predicting
the
o
utcome
of
construction
litigation,
crime
-
related
cases,
parental
rights,
worker
types,
divorces,
and
tax
law.
the
machine
learning
me
thods
can
function
as
support
decision
tools
in
the
legal
system
with
artificial
intelligence’s
advancement.
This
study
aimed
to
impart
a
systematic
literature
review
(SLR)
of
studies
concerning
the
predicti
on
of
court
decision
s
via
machine
learning
m
ethod
s.
T
he
review
determines
and
analyses
the
machine
learning
methods
used
in
predicting
court
decisions.
Th
is
review
utilised
RepOrting
Standard
s
for
Systema
tic Evi
dence
Synthes
es
(
ROSES
)
publication standard.
Subseq
uently,
22
relevant
studies
that
most
comm
only
predicted
the
judgement
results
involving
binary
classification
were
chosen
from
significant
databases:
Scopus
and
Web
of
Sciences.
According
to
the
SLR’s
outcomes,
various
machine
learning
methods
can
be
used
in
predicting
court
decisions.
Additiona
l
ly,
the
performance
is
acceptable
since
most
methods
achieved
more
tha
n
70%
accuracy.
Nevertheles
s,
improvem
ents
can
be
m
ade
on
the
types
of
j
udicial
decisions predicted using the existing machine learning methods
.
K
e
y
w
o
r
d
s
:
J
udi
c
ia
l
c
a
s
e
s
L
e
ga
l
s
ys
te
m
M
a
c
hi
ne
l
e
a
r
ni
ng
P
r
e
di
c
ti
ng c
our
t
de
c
is
io
n
P
r
e
di
c
ti
ve
m
ode
l
ROSES
This is an
open
acce
ss artic
le unde
r the
CC BY
-
SA
license.
C
or
r
e
s
pon
di
n
g A
u
th
or
:
N
ur
A
qi
la
h K
ha
di
ja
h R
os
il
i
S
c
hool
of
C
om
put
in
g, F
a
c
ul
ty
of
E
ngi
ne
e
r
in
g
,
U
ni
ve
r
s
it
i
T
e
knol
ogi
M
a
la
ys
ia
S
ul
ta
n I
br
a
hi
m
C
ha
nc
e
ll
e
r
y B
ui
ld
in
g, J
a
la
n I
m
a
n, 81310 S
kuda
i,
J
ohor
, M
a
la
ys
ia
E
m
a
il
:
a
qi
la
hr
os
il
i@gma
i
l
.c
om
1.
I
N
T
R
O
D
U
C
T
I
O
N
T
he
gl
oba
li
s
e
d w
or
ld
to
da
y de
m
a
nds
s
pe
e
dy a
nd e
f
f
ic
ie
nt
ha
nd
li
ng of
e
ve
r
y a
c
ti
on
[
1]
–
[
3
]
.
T
he
f
a
s
t
-
m
ovi
ng
a
c
ti
ons
a
r
e
e
s
s
e
nt
ia
l
in
e
ns
ur
in
g
th
a
t
th
e
s
e
r
vi
c
e
s
c
a
n
be
im
pl
e
m
e
nt
e
d
in
li
ne
w
it
h
th
e
r
a
pi
d
de
ve
lo
pm
e
nt
of
te
c
hnol
ogy
a
nd
in
f
or
m
a
ti
on,
in
c
lu
di
ng
in
th
e
le
ga
l
s
ys
te
m
[
4]
–
[
20]
.
J
udge
s
a
nd
la
w
ye
r
s
ge
ne
r
a
ll
y
ha
ndl
e
le
ga
l
c
a
s
e
s
,
but
th
e
he
lp
of
te
c
hnol
ogy
is
c
r
it
ic
a
ll
y
e
s
s
e
nt
ia
l
due
to
th
e
m
a
s
s
iv
e
num
be
r
s
of
c
a
s
e
s
da
il
y.
T
he
e
f
f
e
c
t
of
‘
de
la
y
in
ju
s
ti
c
e
’
m
a
y
le
a
d
to
va
r
i
ous
c
ons
e
que
nc
e
s
,
s
uc
h
a
s
w
it
ne
s
s
hos
ti
li
ty
,
unf
it
ne
s
s
of
t
he
pl
a
in
ti
f
f
or
a
c
c
us
e
d a
nd othe
r
a
dve
r
s
e
i
m
pa
c
ts
[2
1]
.
L
e
ga
l
pr
of
e
s
s
io
na
ls
c
ur
r
e
nt
ly
f
oc
us
on
a
r
ti
f
ic
ia
l
in
te
ll
ig
e
nc
e
[
22]
.
A
c
c
or
di
ng
to
hi
s
to
r
ic
a
l
d
a
ta
s
e
ts
in
th
e
le
ga
l
c
ont
e
xt
,
ju
di
c
ia
l
de
c
is
io
n
s
’
pr
e
di
c
ti
on
is
s
ta
nda
r
d
a
nd
w
id
e
ly
pr
a
c
ti
s
e
d
in
th
e
w
or
ld
w
id
e
le
ga
l
s
ys
t
e
m
.
M
a
c
hi
ne
le
a
r
ni
ng
is
a
budding
s
c
ie
nt
if
ic
a
lg
or
it
hm
s
s
tu
dy,
a
nd
s
ta
ti
s
ti
c
a
l
m
ode
ls
a
r
e
a
r
ti
f
ic
ia
l
in
te
ll
ig
e
nc
e
’
s
pa
r
ts
t
ha
t
e
na
bl
e
s
y
s
te
m
s
t
o a
ut
om
a
ti
c
a
ll
y l
e
a
r
n a
nd i
m
pr
ovi
s
e
e
xpe
r
ie
nc
e
f
r
om
t
he
t
e
s
t
da
ta
[
23]
–
[
30]
.
T
he
le
ga
l
s
y
s
te
m
’
s
a
dva
nc
e
m
e
nt
vi
a
th
e
us
a
ge
of
th
e
m
a
c
hi
ne
le
a
r
ni
ng
a
lg
or
it
hm
is
c
r
uc
ia
l
in
r
e
duc
in
g
th
e
w
or
kl
oa
d
of
le
ga
l
pr
of
e
s
s
io
ns
a
nd
s
a
ve
s
th
e
ti
m
e
to
s
e
tt
le
pe
ndi
ng
c
a
s
e
s
dur
in
g
th
e
C
ovi
d
-
19
pa
nde
m
ic
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
10
, N
o.
4
,
D
e
c
e
m
be
r
20
2
1
:
1091
–
1102
1092
[
21]
, [
31
]
–
[
33]
.
T
he
r
e
f
or
e
, t
hi
s
s
tu
dy a
im
e
d t
o
i
nve
s
ti
ga
te
t
he
e
xi
s
ti
ng ma
c
hi
ne
l
e
a
r
n
in
g m
e
th
od de
ve
lo
pe
d t
o
pr
e
di
c
t
ju
di
c
ia
l
de
c
is
io
ns
.
th
e
c
a
s
e
s
th
a
t
us
e
d
th
is
a
ppr
oa
c
h
w
e
r
e
id
e
nt
if
ie
d,
a
nd
th
e
m
e
th
ods
’
pe
r
f
or
m
a
nc
e
w
a
s
m
oni
to
r
e
d t
o s
tu
dy t
he
m
e
th
ods
’
e
f
f
e
c
ti
ve
ne
s
s
.
2.
M
E
T
H
O
D
2
.1.
T
h
e
R
e
vi
e
w
P
r
ot
oc
ol
-
R
O
S
E
S
T
he
R
O
S
E
S
r
e
vi
e
w
pr
ot
oc
ol
le
a
d
th
e
c
ur
r
e
nt
r
e
s
e
a
r
c
h.
T
he
R
O
S
E
S
pr
ot
oc
ol
is
de
v
e
lo
pe
d
f
or
s
ys
te
m
a
ti
c
r
e
vi
e
w
a
nd
e
nvi
r
onm
e
nt
m
a
na
ge
m
e
nt
f
ie
ld
m
a
ps
[
34]
–
[
45]
.
A
ddi
ti
ona
ll
y,
th
e
R
O
S
E
S
p
r
ot
oc
ol
a
ls
o
e
nc
our
a
ge
s
r
e
s
e
a
r
c
he
r
s
to
gua
r
a
nt
e
e
th
a
t
th
e
y
of
f
e
r
th
e
c
or
r
e
c
t
in
f
or
m
a
ti
on
w
it
h
e
xpl
ic
it
de
ta
il
s
.
T
h
e
r
e
s
e
a
r
c
he
r
s
be
ga
n t
he
S
L
R
by f
or
m
ul
a
ti
ng r
e
s
e
a
r
c
h que
s
ti
ons
a
c
c
or
di
ng t
o t
he
r
e
vi
e
w
’
s
pr
ot
oc
ol
[
46]
–
[
48]
. S
ubs
e
que
nt
ly
,
th
e
r
e
s
e
a
r
c
he
r
s
w
e
r
e
r
e
qui
r
e
d
to
d
e
s
c
r
ib
e
th
e
s
y
s
te
m
a
ti
c
s
e
a
r
c
hi
ng
s
tr
a
te
gy
th
a
t
c
on
s
is
ts
of
th
r
e
e
pr
oc
e
s
s
e
s
,
s
uc
h
a
s
id
e
nt
if
ic
a
ti
on,
s
c
r
e
e
ni
ng
a
nd
e
li
gi
bi
li
ty
.
th
e
r
e
s
e
a
r
c
h
e
r
s
w
e
r
e
a
ls
o
r
e
qui
r
e
d
to
pe
r
f
or
m
a
qu
a
li
ty
a
ppr
a
is
a
l
of
th
e
s
e
le
c
te
d
a
r
ti
c
le
s
.
L
a
s
tl
y,
th
e
a
ut
hor
s
e
la
bor
a
te
d
on
th
e
o
ut
c
om
e
s
ge
ne
r
a
te
d
f
r
om
th
e
c
hos
e
n
pr
in
c
ip
a
l
a
r
ti
c
le
s
.
2
.
2
.
F
or
m
u
la
t
io
n
of
R
e
s
e
ar
c
h
Q
u
e
s
t
io
n
s
T
he
r
e
s
e
a
r
c
h
que
s
ti
ons
f
or
th
is
s
tu
dy
w
e
r
e
f
or
m
ul
a
te
d
a
c
c
or
di
ng
to
th
e
e
le
m
e
nt
s
of
P
opul
a
ti
on
o
r
P
r
obl
e
m
(
P
)
,
I
n
te
r
e
s
t
(
I
)
a
nd
C
ont
e
xt
(
C
o)
,
or
P
I
C
o.
T
he
P
I
C
o
is
a
to
ol
to
he
lp
r
e
s
e
a
r
c
he
r
s
to
c
on
s
tr
uc
t
r
e
s
e
a
r
c
h
que
s
ti
ons
f
or
th
e
r
e
vi
e
w
.
T
he
P
I
C
o c
ont
e
xt
e
n
c
om
pa
s
s
e
s
th
e
f
o
ll
ow
in
g
a
s
pe
c
t
s
in
th
i
s
r
e
s
e
a
r
c
h:
i)
P
opul
a
ti
on:
M
a
c
hi
ne
L
e
a
r
ni
ng
,
ii
)
I
nt
e
r
e
s
t:
P
r
e
di
c
ti
on
,
a
nd
i
ii
)
C
ont
e
xt
:
J
udi
c
ia
l
D
e
c
is
io
n
.
T
he
f
or
m
ul
a
te
d
r
e
s
e
a
r
c
h
que
s
ti
ons
w
e
r
e
:
1)
.
W
ha
t
ty
pe
s
of
j
udi
c
ia
l
de
c
is
io
n
s
ha
ve
b
e
e
n pr
e
di
c
te
d us
in
g t
he
m
a
c
hi
ne
l
e
a
r
ni
ng me
th
od?
2)
.
W
ha
t
a
r
e
t
he
m
a
c
hi
ne
l
e
a
r
ni
ng me
th
ods
us
e
d t
o pr
e
di
c
t
ju
di
c
ia
l
de
c
is
io
ns
?
3)
.
H
ow
w
a
s
t
he
pe
r
f
or
m
a
nc
e
of
t
he
m
a
c
hi
ne
l
e
a
r
ni
ng me
th
od us
e
d
t
o pr
e
di
c
t
ju
di
c
ia
l
de
c
is
io
ns
?
2
.
3
.
S
ys
t
e
m
at
ic
S
e
ar
c
h
in
g S
t
r
at
e
gi
e
s
T
he
s
e
a
r
c
hi
ng
pr
oc
e
s
s
in
S
L
R
c
om
pr
is
e
s
th
r
e
e
m
a
in
s
te
ps
:
i)
id
e
nt
if
ic
a
ti
on,
ii
)
s
c
r
e
e
ni
ng,
a
nd
ii
i)
e
li
gi
bi
li
ty
[
7
]
.
T
he
w
hol
e
pr
oc
e
s
s
w
a
s
s
um
m
a
r
is
e
d
in
th
e
f
lo
w
di
a
gr
a
m
de
pi
c
te
d
in
F
ig
ur
e
1
,
a
nd
e
xpl
a
in
e
d
in
th
e
be
lo
w
s
e
c
ti
on
s
.
2
.
3
.
1.
I
d
e
n
t
if
ic
at
io
n
T
he
pur
pos
e
of
th
e
id
e
nt
if
ic
a
ti
on
pr
oc
e
s
s
is
to
m
a
xi
m
is
e
th
e
n
um
be
r
of
ke
yw
or
ds
to
be
s
e
a
r
c
he
d
in
da
ta
ba
s
e
s
.
T
he
k
e
yw
or
ds
a
r
e
de
ve
lo
pe
d
ba
s
e
d
on
th
e
r
e
s
e
a
r
c
h
q
ue
s
ti
ons
f
or
m
ul
a
te
d.
T
he
v
a
r
ia
ti
on
of
ke
yw
or
ds
r
e
li
e
s
on
a
n
onl
in
e
th
e
s
a
ur
us
to
id
e
nt
if
y
s
ynonym
s
a
nd
r
e
la
te
d
te
r
m
s
,
ke
yw
or
ds
us
e
d
in
pr
e
vi
ous
s
tu
di
e
s
a
nd
s
ugge
s
te
d
by
da
ta
ba
s
e
s
a
nd
e
xpe
r
ts
.
N
e
ve
r
th
e
le
s
s
,
th
e
m
a
in
ke
y
w
or
ds
us
e
d
in
th
is
s
tu
dy
a
r
e
pr
e
di
c
ti
on,
ju
di
c
ia
l
de
c
is
io
n a
nd m
a
c
hi
ne
l
e
a
r
ni
ng. T
hi
s
s
tu
dy r
e
f
e
r
s
t
o t
w
o
m
a
jo
r
i
nde
xe
d da
ta
ba
s
e
s
, na
m
e
ly
S
c
opus
a
nd W
e
b of
S
c
ie
nc
e
. T
he
s
e
da
ta
b
a
s
e
s
w
e
r
e
c
ho
s
e
n due
t
o
s
e
ve
r
a
l
a
dva
nt
a
ge
s
.
F
ir
s
t,
th
e
da
ta
ba
s
e
s
c
ont
r
ol
th
e
a
r
ti
c
le
’
s
qua
li
ty
a
nd
c
ons
i
s
t
of
a
r
ti
c
le
s
f
r
om
va
r
io
us
m
ul
ti
di
s
c
ip
li
na
r
y
f
ie
ld
s
.
S
e
c
ond,
th
e
da
ta
ba
s
e
s
pr
ovi
de
c
om
pr
e
he
ns
iv
e
a
nd
a
dva
nc
e
s
e
a
r
c
hi
ng
f
unc
ti
ons
.
T
he
r
e
s
e
a
r
c
he
r
s
c
ons
tr
uc
te
d
a
f
ul
l
s
e
a
r
c
h
s
tr
in
g
us
in
g
th
e
B
ool
e
a
n
ope
r
a
to
r
“
A
N
D
”
a
nd
“
O
R
”
,
phr
a
s
e
s
e
a
r
c
hi
ng,
tr
unc
a
ti
on
a
nd
w
il
d c
a
r
d pr
ovi
de
d i
n both
da
ta
ba
s
e
s
,
a
s
T
a
bl
e
1
.
F
ur
th
e
r
m
or
e
, t
he
i
de
nt
if
ic
a
ti
on pr
oc
e
s
s
a
ls
o i
nc
lu
d
e
d m
a
nua
l
s
e
a
r
c
hi
ng
to
id
e
nt
if
y
r
e
le
va
nt
a
r
ti
c
le
s
in
pr
e
di
c
ti
ng
ju
di
c
ia
l
de
c
is
io
ns
us
in
g
m
a
c
hi
ne
le
a
r
ni
ng.
T
hi
s
pr
oc
e
s
s
m
a
na
ge
d t
o r
e
tr
ie
ve
94 a
r
ti
c
le
s
f
r
om
S
c
opus
a
nd 32 a
r
ti
c
le
s
f
r
o
m
W
e
b of
S
c
ie
nc
e
.
2
.
3
.
2.
S
c
r
e
e
n
in
g
T
he
s
c
r
e
e
ni
ng
pr
oc
e
s
s
w
a
s
unde
r
ta
ke
n
f
or
a
ll
th
e
s
e
le
c
te
d
a
r
ti
c
le
s
in
th
e
id
e
nt
if
ic
a
ti
on
pr
oc
e
s
s
.
T
he
pur
pos
e
of
t
he
s
c
r
e
e
ni
ng pur
pos
e
i
s
t
o i
nc
lu
d
e
a
nd e
xc
lu
de
a
r
ti
c
le
s
ba
s
e
d on th
e
c
r
it
e
r
ia
de
te
r
m
in
e
d.
th
e
in
it
ia
l
s
c
r
e
e
ni
ng pr
oc
e
s
s
r
e
s
tr
ic
ts
t
he
t
im
e
li
ne
t
o b
e
i
n a
s
pe
c
if
ic
i
nt
e
r
va
l
r
e
c
om
m
e
nde
d by
O
kol
i
[
49]
.
T
he
s
e
a
r
c
hi
ng
pr
oc
e
s
s
w
a
s
l
im
it
e
d t
o a
r
ti
c
le
s
publi
s
he
d
f
r
om
t
he
ye
a
r
2000 to
2021 only. Ne
ve
r
th
e
le
s
s
,
t
he
s
e
a
r
c
hi
ng pr
oc
e
s
s
w
a
s
s
ta
r
te
d
in
M
a
r
c
h
2021,
a
nd
th
e
ye
a
r
ha
s
not
c
om
e
to
a
n
e
nd.
T
hus
,
th
e
f
in
di
ngs
w
e
r
e
li
m
it
e
d
to
M
a
r
c
h
2021.
th
e
s
e
c
ond
in
c
lu
s
io
n
c
r
it
e
r
io
n
w
a
s
th
e
la
ngu
a
ge
us
e
d
in
th
e
publ
is
he
d
a
r
ti
c
le
s
or
jo
ur
na
ls
.
A
ll
non
-
E
ngl
is
h
la
ngua
ge
a
r
ti
c
le
s
w
e
r
e
e
xc
lu
de
d
du
e
to
po
s
s
ib
le
tr
a
n
s
la
ti
on
di
f
f
ic
ul
ti
e
s
.
T
h
e
in
c
lu
s
io
n
a
nd
e
x
c
lu
s
io
n
c
r
it
e
r
ia
a
r
e
e
nl
is
te
d i
n T
a
bl
e
2.
2
.
3
.
3.
E
li
gi
b
il
it
y
T
he
f
in
a
l
pr
oc
e
s
s
in
th
e
s
ys
te
m
a
ti
c
s
e
a
r
c
hi
ng
pr
oc
e
dur
e
is
e
li
gi
bi
li
ty
.
T
hi
s
pr
oc
e
s
s
w
a
s
unde
r
ta
ke
n
m
a
nua
ll
y
to
r
e
vi
e
w
th
e
a
r
ti
c
le
s
by
r
e
a
di
ng
a
ll
th
e
a
r
t
ic
le
s
’
ti
tl
e
s
a
nd
a
bs
tr
a
c
ts
th
or
oughly.
T
he
e
li
gi
bi
li
ty
s
te
p
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
A
s
y
s
te
m
at
ic
l
it
e
r
at
u
r
e
r
e
v
ie
w
of
m
ac
hi
n
e
l
e
ar
ni
ng m
e
th
od
s
i
n
pr
e
di
c
ti
ng
…
(
N
ur
A
qi
la
h K
hadi
ja
h R
os
il
i)
1093
e
ns
ur
e
s
t
ha
t
a
ll
t
he
s
e
le
c
te
d a
r
ti
c
le
s
c
om
pl
ie
d w
it
h t
he
pr
e
-
de
te
r
m
in
e
d c
r
it
e
r
ia
.
th
e
e
li
gi
b
il
it
y p
r
oc
e
s
s
i
nc
lu
de
d
20 a
r
ti
c
le
s
r
e
tr
ie
ve
d f
r
om
S
c
opus
a
nd 14 a
r
ti
c
le
s
f
r
om
W
e
b of
S
c
ie
nc
e
a
f
te
r
m
a
nua
ll
y r
e
vi
e
w
e
d.
T
a
bl
e
1.
T
he
s
e
a
r
c
h
s
tr
in
gs
D
a
t
a
ba
s
e
S
e
a
r
c
h S
t
r
i
ng
S
c
opus
T
I
T
L
E
-
A
B
S
-
K
E
Y
(
(
"pr
e
di
c
t
*" O
R
"pr
e
di
c
t
i
on*" O
R
"pr
e
di
c
t
i
ng*" O
R
"f
or
e
c
a
s
t
*")
A
N
D
(
"c
our
t
de
c
i
s
i
on*" O
R
"le
g
a
l
de
c
i
s
i
on*" O
R
"la
w
de
c
i
s
i
on*" O
R
"judi
c
i
a
l
c
a
s
e
*")
A
N
D
(
"ma
c
hi
ne
l
e
a
r
ni
ng*" O
R
"a
r
t
i
f
i
c
i
a
l
i
nt
e
l
l
i
ge
nc
e
*" O
R
"A
I
*" O
R
"s
upe
r
vi
s
e
* m
a
c
hi
ne
l
e
a
r
ni
ng*")
)
W
e
b of
S
c
i
e
nc
e
(
T
S
= (
(
"pr
e
di
c
t
i
on*" O
R
"pr
e
di
c
t
*" O
R
"pr
e
di
c
t
i
ng")
A
N
D
(
"c
our
t
de
c
i
s
i
on*" O
R
"judi
c
i
a
l
de
c
i
s
i
on*" O
R
"le
ga
l
de
c
i
s
i
on*")
A
N
D
(
"ma
c
hi
ne
l
e
a
r
ni
ng" O
R
"
A
I
")
)
)
T
a
bl
e
2
.
T
he
in
c
lu
s
io
n a
nd
e
xc
lu
s
io
n c
r
it
e
r
ia
C
r
i
t
e
r
i
a
I
nc
l
us
i
on
E
xc
l
us
i
on
T
i
m
e
l
i
ne
2000
-
2021
B
e
f
or
e
2000
L
a
ngua
ge
E
ngl
i
s
h
N
on
-
E
ngl
i
s
h
M
e
t
hods
M
a
c
hi
ne
l
e
a
r
ni
ng
O
t
he
r
t
ha
n m
a
c
hi
ne
l
e
a
r
ni
ng
F
ig
ur
e
1.
T
he
f
lo
w
di
a
gr
a
m
[
50]
2
.
4
.
Q
u
al
it
y A
p
p
r
ai
s
al
T
he
pur
pos
e
of
c
on
s
tr
uc
ti
ng
Q
ua
li
ty
A
s
s
e
s
s
m
e
nt
(
Q
A
)
is
to
d
e
c
id
e
c
on
c
e
r
ni
ng
th
e
c
hos
e
n
s
tu
di
e
s
’
ove
r
a
ll
qua
li
ty
[
22
]
.
T
hus
,
th
e
f
ol
lo
w
in
g
qua
li
ty
c
r
it
e
r
ia
w
e
r
e
u
ti
li
s
e
d
to
e
va
lu
a
te
th
e
c
hos
e
n
s
tu
di
e
s
to
f
ig
ur
e
out
t
he
s
tr
e
ngt
h of
t
he
s
tu
di
e
s
’
f
in
di
ngs
:
Q
A
1.
D
oe
s
t
he
s
tu
dy r
e
la
te
t
o t
he
r
e
s
e
a
r
c
h obje
c
ti
ve
s
?
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
10
, N
o.
4
,
D
e
c
e
m
be
r
20
2
1
:
1091
–
1102
1094
Q
A
2.
D
oe
s
t
he
s
tu
dy me
nt
io
n t
he
m
e
th
od or
a
ppr
oa
c
h u
s
e
d i
n p
r
e
di
c
ti
on?
Q
A
3.
I
s
t
he
r
e
s
e
a
r
c
h m
e
th
odol
ogy c
le
a
r
ly
e
xpl
a
in
e
d?
Q
A
4.
I
s
t
he
da
ta
c
ol
le
c
ti
on me
th
od de
s
c
r
ib
e
d?
Q
A
5.
D
oe
s
t
he
pe
r
f
or
m
a
nc
e
of
t
he
m
e
th
od us
e
d h
a
ve
be
e
n di
s
c
us
s
e
d?
T
he
26
s
e
le
c
te
d
s
tu
di
e
s
w
e
r
e
e
xa
m
in
e
d
th
r
ough
th
e
f
iv
e
Q
A
qu
e
s
ti
ons
to
de
te
r
m
in
e
th
e
r
e
s
e
a
r
c
he
r
s
’
c
onf
id
e
nc
e
in
th
e
c
hos
e
n
s
tu
di
e
s
’
c
r
e
di
bi
li
ty
.
T
w
o
e
xp
e
r
ts
w
e
r
e
in
vi
te
d
to
a
ppr
a
is
e
th
e
Q
A
to
de
te
r
m
in
e
th
e
a
r
ti
c
le
s
’
c
ont
e
nt
qua
li
ty
.
th
e
r
e
vi
e
w
e
r
r
a
nke
d
th
e
a
r
ti
c
le
s
in
to
th
r
e
e
le
ve
ls
:
lo
w
,
m
ode
r
a
te
,
a
nd
hi
gh,
a
s
s
ugge
s
te
d
b
y
[
51]
.
T
he
a
r
ti
c
le
s
r
a
nke
d
a
s
m
ode
r
a
te
a
nd
hi
gh
w
e
r
e
e
li
gi
bl
e
f
or
r
e
vi
e
w
in
th
e
f
ol
lo
w
in
g
pr
oc
e
s
s
.
T
he
r
e
s
e
a
r
c
he
r
s
a
da
pt
e
d
th
e
s
c
or
in
g
s
tr
a
te
gy
e
m
pl
oye
d
by
[
52]
to
a
s
s
e
s
s
th
e
a
r
ti
c
le
s
’
qua
li
ty
.
T
he
s
c
or
in
g
of
th
e
qua
li
ty
e
va
lu
a
ti
on
w
a
s
s
tr
uc
tu
r
e
d
a
s
:
i)
1
poi
nt
r
e
pr
e
s
e
nt
s
‘
Y
e
s
’
,
ii
)
0.5
poi
nt
r
e
pr
e
s
e
nt
s
‘
P
a
r
tl
y’
,
a
nd
ii
i)
0
poi
nt
r
e
pr
e
s
e
nt
s
‘
N
o’
.
T
he
s
c
or
in
g
poi
nt
r
a
nke
d
th
e
a
r
ti
c
le
s
in
to
th
r
e
e
c
a
te
gor
ie
s
:
i)
z
e
r
o
(
0
)
to
two
(
2
)
poi
nt
s
w
e
r
e
c
ons
id
e
r
e
d
a
s
lo
w
,
ii
)
two
-
poi
nt
-
f
iv
e
(
2.5)
to
th
r
e
e
(
3)
poi
nt
s
w
e
r
e
c
ons
id
e
r
e
d
a
s
m
ode
r
a
te
,
a
nd
ii
i)
th
r
e
e
-
poi
nt
-
f
iv
e
(
3.5)
to
f
iv
e
(
5)
poi
nt
s
w
e
r
e
c
ons
id
e
r
e
d
a
s
hi
gh.
F
in
a
ll
y,
onl
y
22
a
r
ti
c
le
s
w
e
r
e
e
li
gi
bl
e
f
or
Q
A
a
f
te
r
s
c
or
in
g
w
a
s
unde
r
ta
ke
n.
3.
R
E
S
U
L
T
S
A
N
D
D
I
S
C
U
S
S
I
O
N
T
he
out
c
om
e
s
of
th
e
c
hos
e
n
s
ig
ni
f
ic
a
nt
s
tu
di
e
s
,
vi
s
ua
li
s
a
ti
on
of
publ
ic
a
ti
on
ye
a
r
a
nd
th
e
out
li
ne
of
th
e
Q
A
f
in
di
ngs
a
r
e
s
um
m
a
r
is
e
d i
n t
he
f
ol
lo
w
in
g s
e
c
ti
on
s
.
3.1 S
e
le
c
t
e
d
P
r
im
ar
y S
t
u
d
ie
s
T
w
e
nt
y
-
two
s
tu
di
e
s
w
e
r
e
c
ho
s
e
n
th
r
ough
th
e
S
L
R
to
id
e
nt
if
y
t
he
ty
pe
s
of
le
ga
l
ju
dge
m
e
nt
c
a
s
e
s
th
a
t
e
m
pl
oy
th
e
m
a
c
hi
ne
le
a
r
ni
ng
m
e
th
od
to
e
nvi
s
a
g
e
th
e
f
in
di
ngs
.
S
ubs
e
que
nt
ly
,
th
e
m
a
c
hi
ne
l
e
a
r
ni
ng
m
e
th
ods
us
e
d a
r
e
l
is
t
e
d, a
nd t
he
p
e
r
f
or
m
a
nc
e
of
e
a
c
h m
e
th
od i
s
di
s
c
us
s
e
d. T
a
bl
e
3
s
um
m
a
r
is
e
s
th
e
s
e
le
c
te
d
s
tu
di
e
s
a
nd
c
ons
is
ts
of
th
e
s
tu
di
e
s
’
id
e
nt
it
y
(
I
D
)
,
th
e
publ
ic
a
ti
ons
’
ti
t
le
s
,
th
e
a
r
ti
c
le
s
’
a
ut
hor
s
a
nd
th
e
a
r
ti
c
le
s
’
publ
ic
a
ti
on
ye
a
r
.
T
a
bl
e
3
.
S
um
m
a
r
y of
s
e
le
c
te
d pr
im
a
r
y s
tu
di
e
s
I
D
T
i
t
l
e
A
ut
hor
s
Y
e
a
r
S1
P
r
e
di
c
t
i
ng
t
he
O
ut
c
om
e
of
C
ons
t
r
uc
t
i
on L
i
t
i
ga
t
i
on U
s
i
ng B
oos
t
e
d
D
e
c
i
s
i
on T
r
e
e
s
D
a
vi
d A
r
di
t
i
&
T
ha
ve
e
por
n P
ul
ke
t
2005
S2
P
r
e
di
c
t
i
ng
t
he
O
ut
c
om
e
of
C
ons
t
r
uc
t
i
on L
i
t
i
ga
t
i
on U
s
i
ng P
a
r
t
i
c
l
e
S
w
a
r
m
O
pt
i
m
i
s
a
t
i
on
K
W
C
ha
u
2005
S3
P
r
e
di
c
t
i
on of
C
ons
t
r
uc
t
i
on L
i
t
i
ga
t
i
on O
ut
c
om
e
-
A
C
B
R
A
ppr
oa
c
h
K
W
C
ha
u
2006
S4
P
r
e
di
c
t
i
ng
t
he
O
ut
c
om
e
of
C
ons
t
r
uc
t
i
on L
i
t
i
ga
t
i
on U
s
i
ng a
n I
nt
e
gr
a
t
e
d A
I
M
od
e
l
D
a
vi
d A
r
di
t
i
&
T
ha
ve
e
por
n P
ul
ke
t
2010
S5
L
i
t
i
ga
t
i
on O
ut
c
om
e
P
r
e
di
c
t
i
on of
D
i
f
f
e
r
i
ng S
i
t
e
C
ondi
t
i
on D
i
s
put
e
s
T
hr
ough M
a
c
hi
ne
L
e
a
r
ni
ng M
ode
l
s
T
a
r
e
k M
a
hf
ouz
&
A
m
r
K
a
ndi
l
2012
S6
S
t
udy
of
T
e
r
m
i
na
t
i
on
of
P
a
r
e
nt
a
l
R
i
ght
s
:
A
n
A
na
l
ys
i
s
of
I
s
r
a
e
l
i
C
our
t
D
e
c
i
s
i
on
s
i
n
F
a
vour
or
A
ga
i
ns
t
T
e
r
m
i
na
t
i
on
of
P
a
r
e
nt
a
l
R
i
ght
s
V
e
r
e
d B
e
n
-
D
a
vi
d
201
6
S7
P
r
e
di
c
t
i
ng J
udi
c
i
a
l
D
e
c
i
s
i
ons
of
t
he
E
ur
ope
a
n C
our
t
of
H
um
a
n R
i
ght
s
:
A
N
a
t
ur
a
l
L
a
ngua
ge
P
r
oc
e
s
s
i
ng P
e
r
s
pe
c
t
i
ve
N
i
kol
a
os
A
l
e
t
r
a
s
e
t
al
2016
S8
L
e
a
r
ni
ng
t
o
P
r
e
di
c
t
C
ha
r
ge
s
f
or
C
r
i
m
i
na
l
C
a
s
e
s
w
i
t
h
L
e
ga
l
B
a
s
i
s
B
i
ngf
e
ng L
uo
e
t
al
2017
S9
A
G
e
ne
r
a
l
A
ppr
oa
c
h
f
or
P
r
e
di
c
t
i
ng
t
he
B
e
ha
vi
our
of
t
he
S
upr
e
m
e
C
our
t
of
t
he
U
ni
t
e
d
S
t
a
t
e
s
D
a
ni
e
l
M
a
r
t
i
n K
a
t
z
e
t
al
2017
S
10
P
r
e
di
c
t
i
ng
t
he
O
ut
c
om
e
of
A
ppe
a
l
D
e
c
i
s
i
ons
i
n
G
e
r
m
a
ny’
s
T
a
x
L
a
w
B
e
r
nha
r
d W
a
l
t
l
e
t
al
2017
S
11
L
e
ga
l
j
udge
m
e
nt
pr
e
di
c
t
i
on vi
a
t
opol
ogi
c
a
l
l
e
a
r
ni
ng
H
a
oxi
z
hong
e
t
al
2018
S
12
R
e
s
e
a
r
c
h a
nd
D
e
s
i
gn on C
ogni
t
i
ve
C
om
put
i
ng F
r
a
m
e
w
or
k f
or
P
r
e
di
c
t
i
ng
J
udi
c
i
a
l
D
e
c
i
s
i
on
J
i
a
j
i
ng L
i
e
t
al
2018
S
13
U
s
i
ng M
a
c
hi
ne
L
e
a
r
ni
ng
t
o
P
r
e
di
c
t
D
e
c
i
s
i
on
s
of
t
he
E
C
H
R
M
a
s
ha
M
e
dve
de
v
a
e
t
al
2019
S
14
P
r
e
di
c
t
i
ng
O
ut
c
om
e
s
of
L
e
ga
l
C
a
s
e
s
B
a
s
e
d
on
L
e
g
a
l
F
a
c
t
or
s
U
s
i
ng C
l
a
s
s
i
f
i
e
r
R
a
f
e
A
t
ha
r
S
he
i
kh
e
t
al
2019
S
15
P
r
e
di
c
t
i
ng D
i
s
e
nga
ge
m
e
nt
f
r
om
J
udi
c
i
a
l
P
r
oc
e
e
di
ngs
by
F
e
m
a
l
e
V
i
c
t
i
m
s
of
I
nt
i
m
a
t
e
P
a
r
t
ne
r
V
i
ol
e
nc
e
i
n
S
pa
i
n:
A
S
ys
t
e
m
a
t
i
c
R
e
pl
i
c
a
t
i
on
w
i
t
h
P
r
os
pe
c
t
i
ve
D
a
t
a
M
a
r
i
a
G
a
r
c
í
a
-
J
i
m
é
ne
z
e
t
al
2019
S
16
P
r
e
di
c
t
i
ng O
ut
c
om
e
s
of
J
udi
c
i
a
l
C
a
s
e
s
a
nd
A
n
a
l
ys
i
s
U
s
i
ng M
a
c
hi
ne
L
e
a
r
ni
ng
P
r
of
P
r
i
ya
nka
B
hi
l
a
r
e
e
t
al
2019
S
17
A
D
e
e
p L
e
a
r
ni
ng M
e
t
hod
f
or
J
udi
c
i
a
l
D
e
c
i
s
i
on S
uppor
t
B
a
ogui
C
he
n
e
t
al
2019
S
18
D
e
t
e
r
m
i
ni
ng W
or
ke
r
T
ype
f
r
o
m
L
e
ga
l
T
e
xt
D
a
t
a
U
s
i
ng M
a
c
hi
ne
L
e
a
r
ni
ng
Y
i
f
e
i
Y
i
n
e
t
al
2020
S
19
D
e
e
p L
e
a
r
ni
ng A
l
gor
i
t
hm
f
or
J
udi
c
i
a
l
J
udge
m
e
nt
P
r
e
di
c
t
i
on B
a
s
e
d
on
B
E
R
T
Y
ongj
un W
a
ng
e
t
al
2020
S
20
P
r
e
di
c
t
i
ng
t
he
L
i
t
i
ga
t
i
on O
ut
c
om
e
of
PPP
P
r
oj
e
c
t
D
i
s
put
e
s
B
e
t
w
e
e
n P
ubl
i
c
A
ut
hor
i
t
y
a
nd
P
r
i
va
t
e
P
a
r
t
ne
r
U
s
i
ng
a
n
E
ns
e
m
bl
e
M
ode
l
X
i
a
oxi
a
o Z
he
ng
e
t
al
2020
S
21
A
N
ove
l
A
ppr
oa
c
h
on
A
r
gum
e
nt
B
a
s
e
d
L
e
ga
l
P
r
e
di
c
t
i
on M
ode
l
U
s
i
ng M
a
c
hi
n
e
L
e
a
r
ni
ng
R
i
ya
S
i
l
e
t
al
2020
S
22
T
w
o
-
L
a
ye
r
e
d F
uz
z
y L
ogi
c
B
a
s
e
d
M
ode
l
f
or
P
r
e
di
c
t
i
ng C
our
t
D
e
c
i
s
i
ons
i
n
C
on
s
t
r
uc
t
i
on
C
ont
r
a
c
t
D
i
s
put
e
s
N
a
vi
d B
a
ghe
r
i
a
n
-
M
a
r
a
ndi
e
t
al
2021
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
A
s
y
s
te
m
at
ic
l
it
e
r
at
u
r
e
r
e
v
ie
w
of
m
ac
hi
n
e
l
e
ar
ni
ng m
e
th
od
s
i
n
pr
e
di
c
ti
ng
…
(
N
ur
A
qi
la
h K
hadi
ja
h R
os
il
i)
1095
3.
2
P
u
b
li
c
at
io
n
Y
e
ar
s
T
he
c
ho
s
e
n
s
tu
di
e
s
w
e
r
e
publ
is
he
d
b
e
twe
e
n
2000
a
nd
2021.
N
e
v
e
r
th
e
le
s
s
,
th
e
e
a
r
li
e
s
t
s
tu
dy
publ
is
he
d
on
th
is
to
pi
c
w
a
s
f
r
om
2005.
F
ig
ur
e
2
di
s
pl
a
y
s
th
e
num
be
r
of
s
tu
di
e
s
publ
is
he
d
w
it
hi
n
th
e
s
e
le
c
te
d
ti
m
e
li
ne
.
N
e
ve
r
th
e
le
s
s
, t
he
gr
a
ph i
s
not
pl
ot
te
d f
or
t
he
ye
a
r
2021, a
s
t
he
r
e
s
e
a
r
c
h f
or
t
he
pa
r
ti
c
ul
a
r
ye
a
r
i
s
s
ti
ll
ongoing.
O
ve
r
a
ll
, t
he
onl
y l
a
te
s
t
s
tu
dy w
a
s
publi
s
he
d
in
J
a
nua
r
y 2021, w
hi
le
f
our
a
r
ti
c
le
s
w
e
r
e
publi
s
he
d i
n 2020.
F
iv
e
a
r
ti
c
le
s
w
e
r
e
publi
s
he
d
i
n 2019,
t
w
o i
n 2018,
t
hr
e
e
i
n 2017
a
nd
two in 2016. Only
one
a
r
ti
c
le
w
a
s
publi
s
he
d i
n
2012,
2010,
a
nd
2006,
w
he
r
e
a
s
two
a
r
ti
c
le
s
w
e
r
e
publ
is
he
d
in
2005.
B
a
s
e
d
on
th
e
r
e
s
ul
ts
,
m
a
ny
s
tu
di
e
s
w
e
r
e
obs
e
r
ve
d
to
ha
ve
be
e
n
publ
is
he
d
in
th
e
la
s
t
f
iv
e
ye
a
r
s
.
T
he
r
e
f
or
e
,
th
e
m
a
c
hi
ne
le
a
r
ni
ng
m
e
th
od
c
a
n
f
unc
ti
on
a
s
one
of
t
he
a
ppr
oa
c
he
s
i
n i
m
pr
ovi
ng t
he
l
e
ga
l
s
y
s
te
m
by pr
e
di
c
ti
ng outc
om
e
s
.
F
ig
ur
e
2
.
N
um
be
r
of
s
e
le
c
te
d
s
tu
di
e
s
ove
r
t
he
y
e
a
r
s
3.3 QA Re
s
u
lt
T
he
c
ho
s
e
n
s
tu
di
e
s
w
e
r
e
a
s
s
e
s
s
e
d
b
a
s
e
d
on
th
e
Q
A
qu
e
s
ti
ons
e
xp
la
in
e
d
in
S
e
c
ti
on
2.4,
a
nd
th
e
a
na
ly
s
i
s
is
pr
e
s
e
nt
e
d i
n T
a
bl
e
4.
T
he
ta
bl
e
de
m
ons
tr
a
te
s
t
ha
t
17 s
tu
di
e
s
r
e
c
e
iv
e
d hi
gh s
c
or
e
s
be
twe
e
n t
he
t
ot
a
l
s
c
or
e
of
th
r
ee
-
poi
nt
-
f
iv
e
(
3.5)
t
o f
iv
e
(
5)
, w
he
r
e
a
s
f
iv
e
s
tu
di
e
s
obt
a
in
e
d
a
m
ode
r
a
te
s
c
or
e
of
3.
C
onve
r
s
e
ly
, f
our
s
tu
di
e
s
th
a
t
obt
a
in
e
d l
ow
s
c
or
e
s
w
e
r
e
e
xc
lu
de
d f
r
om
t
he
r
e
vi
e
w
.
T
a
bl
e
4
.
Q
A
r
e
s
ul
ts
S
t
udy I
D
QA1
Q
A
2
QA3
QA4
QA5
S
c
or
e
R
a
t
i
ng
S1
1
0.5
0.5
0.5
0.5
3
M
ode
r
a
t
e
S2
1
0.5
0.5
0.5
0.5
3
M
ode
r
a
t
e
S3
1
0.5
0.5
0.5
0.5
3
M
ode
r
a
t
e
S4
1
1
1
1
1
5
H
i
gh
S5
1
1
0.5
1
0.5
4
H
i
gh
S6
1
0.5
0.5
0.5
0.5
3
M
ode
r
a
t
e
S7
1
0.5
0.5
0.5
0.5
3
M
ode
r
a
t
e
S8
1
1
1
1
1
5
H
i
gh
S9
1
1
1
1
1
5
H
i
gh
S
10
1
1
0.5
0.5
1
4
H
i
gh
S
11
1
1
0.5
1
1
4.5
H
i
gh
S
12
1
1
1
1
1
5
H
i
gh
S
13
1
1
1
1
1
5
H
i
gh
S
14
1
1
0.5
0.5
1
4
H
i
gh
S
15
1
1
0.5
0.5
0.5
3.5
H
i
gh
S
16
1
1
0.5
1
1
4.5
H
i
gh
S
17
1
1
1
1
0.5
4.5
H
i
gh
S
18
1
1
1
1
1
5
H
i
gh
S
19
1
1
1
0.5
1
4.5
H
i
gh
S
20
1
1
0.5
1
1
4.5
H
i
gh
S
21
1
1
0.5
1
1
4.5
H
i
gh
S
22
1
1
1
1
1
5
H
i
gh
0
0
0
0
0
2
1
0
0
0
1
0
1
0
0
0
2
3
2
5
4
0
1
2
3
4
5
6
2
0
0
0
2
0
0
1
2
0
0
2
2
0
0
3
2
0
0
4
2
0
0
5
2
0
0
6
2
0
0
7
2
0
0
8
2
0
0
9
2
0
1
0
2
0
1
1
2
0
1
2
2
0
1
3
2
0
1
4
2
0
1
5
2
0
1
6
2
0
1
7
2
0
1
8
2
0
1
9
2
0
2
0
P
ubl
i
cat
i
on
y
ear
Y
e
a
r
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
10
, N
o.
4
,
D
e
c
e
m
be
r
20
2
1
:
1091
–
1102
1096
Th
e
s
e
b
e
lo
w
s
e
c
ti
on
s
pr
ovi
de
a
br
e
a
kdown
of
th
e
r
e
s
ul
t
s
a
c
c
or
di
ng
to
th
e
r
e
s
e
a
r
c
h
que
s
ti
ons
id
e
nt
if
ie
d
in
S
e
c
ti
on
2.2.
T
he
d
e
s
c
r
ip
ti
on
of
th
e
f
in
di
ngs
f
r
om
e
a
c
h
r
e
s
e
a
r
c
h
que
s
ti
on
i
s
pr
e
s
e
nt
e
d
in
s
e
p
a
r
a
te
s
ub
s
e
c
ti
on
s
.
th
e
r
e
s
e
a
r
c
h que
s
ti
ons
a
r
e
a
bbr
e
vi
a
te
d
a
s
R
Q
he
r
e
a
f
te
r
.
3.
4
.
T
yp
e
s
of
Ju
d
ic
ia
l
D
e
c
i
s
io
n
T
he
r
e
s
e
a
r
c
h
que
s
ti
ons
a
r
e
di
s
c
u
s
s
e
d
in
th
is
s
e
c
ti
on.
T
h
e
f
ir
s
t
r
e
s
e
a
r
c
h
que
s
ti
on
th
a
t
w
a
s
a
ddr
e
s
s
e
d:
(
R
Q
1
)
W
ha
t
ty
pe
s
of
j
ud
ic
ia
l
de
c
is
io
ns
ha
ve
be
e
n pr
e
di
c
te
d us
in
g t
he
m
a
c
hi
ne
l
e
a
r
ni
ng me
th
od?
I
n t
he
w
o
r
ld
of
th
e
le
ga
l
s
ys
te
m
,
ju
dge
m
e
nt
c
ons
is
ts
of
va
r
io
us
s
ubt
a
s
ks
th
a
t
ha
ve
to
be
c
ons
id
e
r
e
d.
T
he
le
ga
l
s
ys
te
m
is
di
f
f
ic
ul
t
to
be
unde
r
s
to
od
by
th
e
c
iv
il
ia
ns
a
s
th
e
le
g
a
l
pr
oc
e
s
s
e
s
in
c
lu
de
in
te
r
a
c
ti
ng
w
it
h
a
la
w
ye
r
,
hi
r
in
g
th
e
la
w
ye
r
,
pr
oc
e
e
di
ng
de
c
i
s
io
ns
a
nd
th
e
le
g
a
l
de
c
is
io
n
s
’
c
ons
e
que
nc
e
s
a
nd
th
e
im
pl
ic
a
ti
ons
of
w
or
ds
in
th
e
c
a
s
e
f
il
e
s
[
53]
.
T
hi
s
s
tu
dy
in
ve
s
ti
ga
te
d
how
m
a
c
hi
ne
le
a
r
ni
ng
c
a
n
b
e
us
e
d
in
c
our
t
pr
oc
e
e
di
ng
s
to
pr
e
di
c
t
ju
di
c
ia
l
de
c
is
io
ns
.
th
e
pr
e
di
c
ti
on
c
a
n
be
of
va
r
io
us
ty
pe
s
,
s
uc
h
a
s
pr
e
di
c
ti
ng
th
e
le
ga
l
ju
dge
m
e
nt
’
s
out
c
om
e
or
th
e
c
ha
r
ge
s
th
a
t
r
e
qui
r
e
m
ul
ti
la
be
l
te
xt
c
la
s
s
if
ic
a
ti
on.
M
ul
ti
pl
e
s
u
bt
a
s
ks
in
le
ga
l
ju
dge
m
e
nt
ty
pi
c
a
ll
y
c
om
pr
is
e
c
om
pr
e
he
ns
iv
e
a
nd
c
om
pl
e
x
s
ub
-
c
la
u
s
e
s
,
s
uc
h
a
s
c
ha
r
ge
s
,
pe
n
a
lt
y
te
r
m
s
,
a
nd
f
in
e
s
[
52]
.
N
e
v
e
r
th
e
le
s
s
,
m
os
t
r
e
s
e
a
r
c
h
e
xpe
r
im
e
nt
e
d
w
it
h
a
bi
na
r
y
ta
s
k
th
a
t
c
la
s
s
if
ie
s
onl
y
t
w
o
pos
s
i
bl
e
out
c
om
e
s
.
B
e
s
id
e
s
pr
e
di
c
ti
ng
th
e
out
c
om
e
of
ju
di
c
ia
l
de
c
is
io
n,
s
e
ve
r
a
l
c
ount
r
ie
s
th
a
t
ut
il
is
e
th
e
c
iv
il
la
w
s
ys
te
m
,
s
u
c
h
a
s
G
e
r
m
a
ny,
F
r
a
nc
e
a
nd
C
hi
na
,
de
e
m
e
d
th
a
t
th
e
pr
e
di
c
ti
on
of
r
e
le
va
nt
a
r
ti
c
le
s
i
s
a
f
un
da
m
e
nt
a
l
s
ubt
a
s
k
th
a
t
gui
de
s
a
nd
s
uppor
ts
th
e
pr
e
di
c
ti
on
[
52]
.
I
n t
hi
s
S
L
R
, s
e
ve
n r
e
s
e
a
r
c
h pa
pe
r
s
w
e
r
e
f
ound to have
di
s
c
us
s
e
d e
nvi
s
a
gi
n
g c
ons
tr
uc
ti
on l
it
ig
a
ti
on’
s
out
c
om
e
.
A
r
di
ti
a
nd
P
hu
lk
e
t
[
54]
m
e
n
ti
one
d
th
a
t
c
ons
tr
uc
ti
on
l
it
ig
a
ti
on
is
or
di
na
r
y
in
num
e
r
ous
c
ons
tr
uc
ti
on
pr
oj
e
c
ts
,
e
xpl
ic
it
ly
in
vol
vi
ng
la
r
ge
c
ont
r
a
c
ts
.
M
is
c
om
m
uni
c
a
ti
on,
in
s
uf
f
ic
ie
nt
s
pe
c
if
ic
a
ti
ons
a
nd
pl
a
ns
,
r
ig
id
c
ont
r
a
c
ts
,
c
ha
nge
s
in
s
it
e
c
ondi
ti
ons
,
non
-
pa
ym
e
nt
,
c
a
tc
h
up
pr
of
it
s
,
li
m
it
e
d
w
or
kf
o
r
c
e
,
in
s
uf
f
ic
ie
nt
to
ol
s
a
nd
e
qui
pm
e
nt
,
in
e
f
f
e
c
ti
ve
s
upe
r
vi
s
io
n,
not
ic
e
r
e
qui
r
e
m
e
nt
s
,
c
on
s
t
r
uc
ti
ve
c
ha
nge
s
not
a
c
knowl
e
dge
d
by
ow
ne
r
,
de
la
ys
,
a
nd
a
c
c
e
le
r
a
ti
on
m
e
a
s
ur
e
s
pr
ovoking
c
la
im
s
a
nd
c
a
u
s
in
g
di
s
put
e
s
.
T
he
r
e
f
or
e
,
A
r
di
ti
a
nd
P
hul
ke
t
[
54
]
pr
opos
e
d
a
to
ol
to
pr
e
di
c
t
th
e
out
c
om
e
of
l
it
ig
a
ti
on
to
m
in
im
is
e
c
o
ns
tr
uc
ti
on
di
s
put
e
s
c
a
u
s
e
d
by
di
s
a
gr
e
e
m
e
nt
s
th
a
t
a
r
e
c
om
pl
ic
a
te
d t
o be
s
e
tt
le
d w
it
hout
e
nga
gi
ng i
n l
e
ga
l
a
c
ti
ons
[
54]
, [
55]
.
L
e
ga
l
a
c
ti
on
r
e
qui
r
e
s
a
hi
ghe
r
s
e
tt
le
m
e
nt
c
os
t
be
c
a
us
e
th
e
li
ti
ga
ti
on
pr
oc
e
s
s
is
c
os
tl
y
a
s
th
e
pr
oc
e
s
s
in
vol
ve
s
c
om
pl
e
x i
s
s
ue
s
. A
ddi
ti
ona
ll
y, t
he
di
s
a
gr
e
e
m
e
nt
be
twe
e
n c
li
e
nt
a
nd c
ont
r
a
c
to
r
m
a
y l
e
a
d t
o r
e
put
a
ti
o
n
da
m
a
ge
on
bot
h
s
id
e
s
[
54]
.
I
n
a
ddi
ti
on,
le
ga
l
a
c
ti
on
i
s
ti
m
e
-
c
on
s
um
in
g
f
or
c
om
pl
e
x
c
on
s
tr
uc
ti
on
di
s
put
e
s
a
nd
m
a
y
ta
ke
two
to
s
ix
ye
a
r
s
be
f
or
e
tr
ia
l,
de
pe
ndi
ng
on
th
e
ju
r
is
di
c
ti
on
[
56]
.
T
he
r
e
f
or
e
,
th
e
r
e
s
e
a
r
c
he
r
s
r
e
c
om
m
e
nd
s
e
ve
r
a
l
m
a
c
hi
ne
le
a
r
ni
ng
m
e
th
ods
to
e
n
s
ur
e
th
e
a
c
c
ur
a
c
y
of
pr
e
d
ic
ti
ng
a
di
s
put
e
r
e
s
ol
ut
io
n’
s
out
c
om
e
in
c
our
ts
.
th
e
m
e
th
ods
c
a
n
e
f
f
ic
ie
nt
ly
de
c
r
e
a
s
e
th
e
num
b
e
r
of
di
s
put
e
s
t
ha
t
r
e
qui
r
e
hi
ghe
r
s
pe
ndi
ng
c
o
s
ts
th
r
ough
th
e
li
ti
ga
ti
on pr
oc
e
s
s
[
51]
.
A
c
c
or
di
ng
to
th
e
c
ur
r
e
nt
s
tu
dy’
s
f
in
di
ngs
,
ni
ne
r
e
s
e
a
r
c
h
pa
pe
r
s
pr
e
di
c
te
d
th
e
out
c
om
e
f
or
c
r
im
e
-
r
e
la
te
d
c
a
s
e
s
.
N
e
ve
r
th
e
le
s
s
,
c
r
im
e
-
r
e
la
te
d
c
a
s
e
s
c
a
n
be
di
vi
de
d i
nt
o
f
e
w
c
a
te
gor
ie
s
. A
le
tr
a
s
pr
e
s
e
nt
e
d
th
e
f
ir
s
t
s
ys
te
m
a
ti
c
s
tu
dy t
ha
t
pr
e
di
c
te
d t
he
out
c
om
e
of
c
a
s
e
s
i
n t
he
E
ur
ope
a
n C
our
t
of
H
um
a
n R
ig
ht
s
ba
s
e
d on te
xt
ua
l
a
na
ly
s
is
[
57]
.
T
he
a
ut
hor
s
c
la
s
s
if
ie
d
th
e
pr
e
di
c
ti
on
out
put
s
in
to
‘
vi
o
la
ti
on’
a
nd
‘
non
-
vi
ol
a
ti
on’
ba
s
e
d
on
te
xt
e
xt
r
a
c
te
d
f
r
om
pr
e
vi
ous
c
a
s
e
s
.
F
ur
th
e
r
s
tu
di
e
s
w
e
r
e
c
onduc
t
e
d
by
im
pr
ovi
ng
th
e
num
b
e
r
of
a
r
ti
c
le
s
a
nd
di
f
f
e
r
e
nt
va
r
ia
bl
e
s
us
in
g
th
e
s
a
m
e
da
ta
s
e
t
[
58]
.
T
hi
s
pr
opo
s
a
l
c
a
n
be
ne
f
it
la
w
ye
r
s
a
nd
ju
dge
s
a
s
a
s
uppor
ti
ng
to
ol
t
o i
de
nt
if
y c
a
s
e
s
a
nd e
xt
r
a
c
t
te
xt
t
ha
t
gui
de
s
de
c
i
s
io
n
-
m
a
ki
ng
[
57]
.
L
uo
[
59]
a
s
s
e
r
te
d
th
a
t
th
e
te
c
hni
que
of
a
na
ly
s
in
g
te
xt
ua
l
f
a
c
t
is
c
r
uc
ia
l
f
or
le
ga
l
a
s
s
is
ta
nt
s
y
s
te
m
s
w
he
r
e
c
iv
il
ia
n
s
unf
a
m
il
ia
r
w
it
h
le
ga
l
te
r
m
s
c
a
n
f
in
d
s
im
il
a
r
c
a
s
e
s
or
po
s
s
ib
le
p
e
na
lt
ie
s
by
de
s
c
r
ib
in
g
a
c
a
s
e
w
it
h t
he
ir
ow
n
w
or
ds
a
nd unde
r
s
ta
nd t
he
l
e
ga
l
ba
s
is
of
t
he
ir
s
e
a
r
c
h c
a
s
e
s
. F
ur
th
e
r
m
or
e
, L
uo
[
59]
pr
opos
e
d a
n
a
tt
e
nt
io
n
-
ba
s
e
d ne
ur
a
l
ne
twor
k m
e
th
od a
s
a
s
ta
nd
a
r
d m
e
th
od t
o pr
e
di
c
t
c
ha
r
ge
s
a
nd e
xt
r
a
c
t
r
e
le
v
a
nt
a
r
ti
c
le
s
i
n
a
uni
f
ie
d
f
r
a
m
e
w
or
k.
T
he
f
in
di
ngs
de
m
ons
tr
a
te
d
th
a
t
pr
ov
i
di
ng
r
e
la
te
d
a
r
ti
c
le
s
c
a
n
e
nha
nc
e
th
e
c
ha
r
ge
pr
e
di
c
ti
on r
e
s
ul
ts
a
nd e
nvi
s
a
g
e
c
ha
r
ge
s
f
or
c
a
s
e
s
w
it
h di
ve
r
s
e
e
xpr
e
s
s
io
n s
ty
le
s
e
f
f
e
c
ti
ve
ly
.
Z
hong
e
t.
al
.
[
60]
pr
opos
e
d
a
di
f
f
e
r
e
nt
a
ppr
oa
c
h
in
m
ode
ll
in
g
th
e
ju
dge
m
e
nt
pr
e
di
c
ti
on
f
r
a
m
e
w
or
k
th
a
t
ut
il
is
e
s
m
ul
ti
pl
e
s
ubt
a
s
k
s
by
c
la
im
in
g
th
a
t
pr
e
vi
ous
s
tu
di
e
s
onl
y
de
s
ig
ne
d
a
ppr
oa
c
he
s
f
or
pa
r
ti
c
ul
a
r
s
ubt
a
s
ks
s
e
t
a
nd
di
f
f
ic
ul
t
to
s
c
a
le
to
ot
he
r
s
ubt
a
s
ks
a
lt
hou
gh
de
ve
lo
pe
d
to
pr
e
di
c
t
la
w
a
r
ti
c
le
s
a
nd
c
ha
r
ge
s
s
im
ul
ta
ne
ous
ly
.
A
ddi
ti
ona
ll
y,
it
f
oc
us
e
d
on
m
ur
de
r
r
e
la
te
d
c
a
s
e
s
by
unde
r
ta
ki
ng
s
u
c
h
a
na
ly
s
is
.
E
xt
r
a
c
ti
on
of
le
ga
l
ju
dge
m
e
nt
c
a
n
be
ut
il
is
e
d
to
id
e
nt
if
y
th
e
de
ta
il
s
of
c
a
s
e
-
s
pe
c
if
ic
le
ga
l
f
a
c
to
r
s
but
doe
s
no
t
in
vol
ve
e
a
s
y
w
or
k
a
nd
is
ti
m
e
-
c
ons
um
in
g.
T
he
r
e
f
or
e
,
e
s
s
e
nt
ia
l
f
a
c
to
r
s
th
a
t
w
i
ll
a
f
f
e
c
t
th
e
pr
e
di
c
ti
on
f
or
m
u
r
de
r
r
e
la
te
d
c
a
s
e
s
a
r
e
e
va
lu
a
te
d
by
pr
e
pa
r
in
g
a
da
ta
s
e
t
to
de
te
r
m
in
e
th
e
f
a
c
to
r
s
a
s
de
s
c
r
ip
to
r
s
f
or
pr
e
di
c
ti
on
out
c
om
e
s
.
T
he
out
c
om
e
pr
e
di
c
ti
on
is
vi
e
w
e
d
a
s
a
bi
na
r
y
c
la
s
s
if
ic
a
ti
on
f
or
c
la
s
s
e
s
a
s
‘
a
c
qui
tt
a
l’
a
nd
‘
c
onvi
c
ti
on’
of
th
e
a
c
c
us
e
d
pe
r
s
on.
T
he
c
ur
r
e
nt
s
tu
dy’
s
f
in
di
ng
i
s
f
ur
th
e
r
di
s
c
u
s
s
e
d
w
it
h
c
a
s
e
s
th
a
t
d
o
not
in
vol
ve
c
iv
il
la
w
a
nd
s
p
e
c
if
ic
a
ll
y
f
oc
us
on
f
a
m
il
y
la
w
c
a
s
e
s
.
A
m
ong
th
e
hi
ghl
ig
ht
e
d
c
a
s
e
s
a
r
e
di
s
e
nga
ge
m
e
nt
,
di
vor
c
e
,
pa
r
e
nt
a
l
r
ig
ht
s
a
nd
dow
r
y.
B
e
n
-
D
a
vi
d
[
61]
c
onduc
te
d
a
c
r
uc
ia
l
s
tu
dy r
e
ga
r
di
ng c
our
t
de
c
is
io
ns
i
n ‘
f
a
vour
’
or
‘
a
ga
in
s
t’
t
he
t
e
r
m
in
a
ti
on o
f
pa
r
e
nt
a
l
r
ig
ht
s
th
a
t
f
ound
th
e
ba
la
nc
e
be
twe
e
n
th
e
c
hi
ld
’
s
be
s
t
in
te
r
e
s
t,
th
e
p
a
r
e
nt
’
s
r
ig
ht
a
nd
th
e
pr
iv
a
c
y
of
th
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
A
s
y
s
te
m
at
ic
l
it
e
r
at
u
r
e
r
e
v
ie
w
of
m
ac
hi
n
e
l
e
ar
ni
ng m
e
th
od
s
i
n
pr
e
di
c
ti
ng
…
(
N
ur
A
qi
la
h K
hadi
ja
h R
os
il
i)
1097
f
a
m
il
y
uni
t
[
62]
.
L
i
e
t.
al
.
[
63]
pr
opos
e
d
a
pr
e
di
c
ti
on
m
ode
l
f
or
di
vor
c
e
.
th
e
r
e
s
e
a
r
c
h
obj
e
c
ti
ve
s
w
e
r
e
to
pr
e
di
c
t
th
e
de
c
is
io
ns
f
or
di
vor
c
e
c
a
s
e
s
w
it
h di
ve
r
s
e
e
xpr
e
s
s
io
n s
ty
le
s
a
nd pr
ovi
de
a
n e
a
s
y unde
r
s
ta
ndi
ng t
o t
he
publi
c
r
e
ga
r
di
ng t
he
r
e
s
ul
ts
[
60]
.
I
n
a
ddi
ti
on,
G
a
r
c
ía
-
J
im
é
ne
z
e
t.
al
.
[
34]
s
tu
di
e
d
di
s
e
ng
a
ge
m
e
nt
pr
e
di
c
ti
on
w
he
r
e
th
e
r
e
s
e
a
r
c
he
r
s
e
xa
m
in
e
d
th
e
v
a
r
ia
bl
e
ne
e
de
d
by
vi
c
ti
m
s
f
r
om
le
ga
l
pr
oc
e
e
di
ng
s
be
f
or
e
m
ode
ll
in
g
th
e
pr
e
di
c
ti
on
m
ode
l.
T
hi
s
s
tu
dy
de
ve
lo
pe
d
a
bi
na
r
y
lo
gi
s
ti
c
r
e
gr
e
s
s
io
n
m
ode
l
th
a
t
pr
e
di
c
ts
di
s
e
ng
a
ge
m
e
nt
w
it
h
two
va
r
ia
bl
e
s
th
a
t
a
r
e
d
if
f
e
r
e
nt
f
r
om
p
r
e
vi
ous
a
ppr
oa
c
he
s
.
th
e
f
ir
s
t
va
r
ia
bl
e
is
th
e
c
ont
a
c
t
w
it
h
th
e
a
bus
e
r
,
w
he
r
e
a
s
th
e
s
e
c
ond
va
r
ia
bl
e
is
th
e
in
te
r
a
c
ti
on
be
twe
e
n
th
e
c
ont
a
c
t
a
nd
th
ought
of
r
e
uni
ti
ng
w
it
h
th
e
a
bus
e
r
.
T
he
pa
pe
r
a
im
e
d
to
pr
e
di
c
t
di
s
e
nga
ge
m
e
nt
by
pr
ot
e
c
ti
ng
w
om
e
n
f
r
om
be
in
g
oppr
e
s
s
e
d
by
c
our
t
de
c
is
io
ns
.
T
he
y
b
e
li
e
ve
d
th
a
t
ot
he
r
f
a
c
to
r
s
s
houl
d
not
in
f
lu
e
nc
e
c
our
t
de
c
is
io
ns
in
di
s
e
nga
g
e
m
e
nt
c
a
s
e
s
,
s
u
c
h
a
s
not
gr
a
nt
e
d
a
pr
ot
e
c
ti
on
or
de
r
,
not
f
e
e
li
ng
s
uppor
te
d by la
w
ye
r
s
or
unc
onvi
nc
in
g r
e
s
pon
s
e
s
f
r
om
pr
of
e
s
s
io
na
ls
dur
in
g pr
oc
e
e
di
ngs
[
64]
.
B
e
ne
f
ic
ia
r
ie
s
in
I
ndi
a
s
pe
nt
a
lo
ng
ti
m
e
w
a
it
in
g
to
g
e
t
de
c
i
s
io
ns
f
r
om
th
e
c
our
t
due
to
th
e
s
c
a
r
c
it
y
of
s
ki
ll
e
d
w
or
kf
or
c
e
a
nd
in
f
r
a
s
tr
uc
tu
r
e
[
21]
.
T
he
pr
ol
onge
d
le
ga
l
pr
oc
e
e
di
ng
m
a
y
le
a
d
to
va
r
io
us
c
on
s
e
que
nc
e
s
.
S
il
e
t
al
.
pr
opos
e
d a
m
ode
l
th
a
t
w
il
l
a
s
s
is
t
le
g
a
l
pr
of
e
s
s
io
n
a
ls
i
n
a
na
ly
s
in
g a
nd
pe
r
f
or
m
in
g
pr
e
di
c
ti
ons
to
gi
ve
a
n
out
c
om
e
a
s
‘
gui
lt
y’
or
‘
not
gui
lt
y’
de
pe
ndi
ng
on
th
e
p
a
r
a
m
e
te
r
s
of
de
a
th
-
r
e
la
te
d
dow
r
y
c
a
s
e
s
[
21]
.
A
w
or
ke
r
ty
pe
a
ppr
oa
c
h
ha
s
a
l
s
o
be
e
n
pr
opos
e
d
in
pr
e
di
c
ti
ng
c
our
t
de
c
is
io
ns
f
or
e
m
pl
oym
e
nt
r
ig
ht
s
a
nd
pr
ot
e
c
ti
on
pur
pos
e
s
[
65]
.
T
he
out
c
om
e
of
va
r
io
us
ty
pe
s
of
c
a
s
e
s
ha
s
b
e
e
n
e
xpl
or
e
d
in
pr
e
di
c
ti
ng
th
e
out
c
om
e
of
c
our
t
de
c
is
io
ns
u
s
in
g
m
a
c
hi
ne
le
a
r
ni
ng,
le
a
di
ng
to
a
c
onc
lu
s
io
n
th
a
t
t
he
r
e
a
r
e
s
ti
ll
oppor
tu
ni
ti
e
s
a
nd
r
oom
f
or
ot
he
r
c
a
s
e
s
to
a
d
a
pt
th
e
m
a
c
hi
ne
le
a
r
ni
ng
m
e
th
od
a
s
a
s
upp
or
ti
ng
to
ol
in
de
c
is
io
n
-
m
a
ki
ng.
F
ut
ur
e
s
tu
di
e
s
c
a
n
in
c
lu
de
a
n
e
xt
e
ns
iv
e
s
tu
dy
on
c
a
s
e
s
th
a
t
r
e
qui
r
e
m
a
c
hi
ne
le
a
r
ni
ng
a
s
a
pr
e
di
c
ti
on
m
ode
l
to
le
s
s
e
n
d
e
c
is
io
n
-
m
a
ki
ng
ti
m
e
.
3.5
.
M
e
t
h
od
s
of
M
ac
h
in
e
L
e
a
r
n
in
g
I
n
th
is
s
e
c
ti
on,
th
e
f
ol
lo
w
in
g
r
e
s
e
a
r
c
h
que
s
ti
on
is
di
s
c
u
s
s
e
d:
(
R
Q
2
)
W
ha
t
a
r
e
th
e
m
a
c
hi
ne
le
a
r
ni
ng
m
e
th
ods
us
e
d
to
pr
e
di
c
t
ju
di
c
i
a
l
de
c
i
s
io
ns
?
L
e
ga
l
pr
of
e
s
s
io
n
a
ls
a
r
e
c
ur
r
e
nt
ly
f
oc
us
e
d
on
a
r
ti
f
ic
ia
l
in
te
ll
ig
e
nc
e
[
66]
. E
nvi
s
a
gi
ng
ju
di
c
ia
l
de
c
is
io
n
s
ba
s
e
d on
hi
s
to
r
ic
a
l
d
a
ta
s
e
t
s
i
n t
he
l
e
ga
l
doma
in
i
s
not
ne
w
a
nd w
id
e
ly
us
e
d
in
th
e
le
ga
l
s
ys
te
m
gl
oba
ll
y.
M
a
c
hi
ne
le
a
r
ni
ng
is
a
n
e
m
e
r
gi
ng
s
c
ie
nt
if
ic
s
tu
dy
of
a
lg
or
it
hm
s
a
nd
s
ta
ti
s
ti
c
a
l
m
ode
ls
th
a
t
a
r
e
pa
r
t
of
a
r
ti
f
ic
ia
l
in
te
ll
ig
e
nc
e
,
e
na
bl
in
g
th
e
s
y
s
te
m
to
le
a
r
n
a
ut
om
a
ti
c
a
ll
y
a
nd
im
pr
ove
th
e
e
xpe
r
ie
nc
e
f
r
om
te
s
t
da
ta
.
T
h
e
c
or
e
r
e
s
e
a
r
c
h
a
s
pe
c
ts
in
a
ppl
yi
ng
m
a
c
hi
ne
le
a
r
ni
ng
in
ju
r
is
pr
ude
nc
e
a
r
e
th
e
e
xt
r
a
c
ti
on
of
in
f
or
m
a
ti
on
a
n
d
a
na
ly
s
is
on
e
xi
s
ti
ng
le
ga
l
doc
um
e
nt
s
.
in
pr
e
vi
ous
pr
a
c
ti
c
e
s
,
la
w
ye
r
s
a
nd
ju
dge
s
ha
ve
to
do
a
ll
th
e
w
or
ks
m
a
nua
ll
y.
H
ow
e
ve
r
,
m
a
c
hi
n
e
le
a
r
ni
ng
ha
s
ta
ke
n
th
e
s
tr
e
a
m
of
s
oc
ie
ty
to
be
c
om
e
m
or
e
in
te
ll
ig
e
nt
by i
nt
e
r
pr
e
ti
ng t
he
t
e
xt
doc
um
e
nt
s
a
nd e
xt
r
a
c
ti
ng t
he
doc
um
e
nt
s
’
c
ont
e
nt
[
53]
.
T
he
r
e
s
e
a
r
c
he
r
s
ob
s
e
r
ve
d
th
e
pr
opos
e
d
m
a
c
hi
ne
le
a
r
ni
ng
in
th
is
S
L
R
by
de
te
r
m
in
in
g
th
e
ty
pe
s
a
nd
na
m
e
s
of
th
e
c
la
s
s
if
ie
r
us
e
d
in
pr
e
di
c
ti
ng
ju
di
c
ia
l
de
c
is
io
ns
.
t
he
m
a
jo
r
it
y
of
s
tu
di
e
s
a
tt
e
m
pt
e
d
to
e
xt
r
ic
a
te
e
f
f
ic
ie
nt
f
e
a
tu
r
e
s
f
r
om
te
xt
c
ont
e
nt
or
c
a
s
e
a
nnot
a
ti
ons
(
da
te
s
,
te
r
m
s
,
lo
c
a
ti
ons
,
a
nd
ty
pe
s
)
[
1]
.
N
e
ve
r
th
e
le
s
s
,
Z
hong
e
t
al
.
[
60]
a
s
s
e
r
te
d
th
a
t
th
e
c
onve
nt
io
na
l
m
e
th
od
s
c
ou
ld
onl
y
e
m
pl
oy
s
ha
ll
ow
te
xt
ua
l
f
e
a
tu
r
e
s
a
nd
m
a
nua
ll
y
de
s
ig
ne
d
f
a
c
to
r
s
.
th
e
f
e
a
tu
r
e
s
a
nd
f
a
c
to
r
s
ne
e
d
e
nor
m
ous
hum
a
n
e
f
f
or
ts
a
nd
r
e
gul
a
r
ly
unde
r
g
o
ge
ne
r
a
li
s
a
ti
on
pr
obl
e
m
s
w
he
n
a
ppl
ie
d
in
ot
he
r
s
c
e
na
r
io
s
.
th
e
a
c
hi
e
ve
m
e
nt
of
ne
ur
a
l
ne
twor
ks
on
na
tu
r
a
l
la
ngua
ge
pr
oc
e
s
s
in
g
(
N
L
P
)
ta
s
k
s
in
s
pi
r
e
d
th
e
r
e
s
e
a
r
c
he
r
s
to
s
ta
r
t
ha
ndl
in
g
le
ga
l
ju
dg
e
pr
e
di
c
ti
on
by
in
te
gr
a
ti
ng
ne
ur
a
l
m
ode
ls
w
it
h
le
ga
l
knowle
dge
[
59]
.
L
uo
[
59]
la
id
out
a
n
a
tt
e
nt
io
n
-
ba
s
e
d
n
e
ur
a
l
ne
twor
k
th
a
t
jo
in
tl
y
m
ode
ls
c
ha
r
ge
pr
e
di
c
ti
on
a
nd
r
e
le
v
a
nt
a
r
ti
c
le
e
xt
r
a
c
ti
on
.
N
one
t
he
le
s
s
,
th
e
s
e
m
ode
ls
a
r
e
de
s
ig
ne
d
f
or
s
p
e
c
if
ic
s
ubt
a
s
ks
.
T
he
r
e
f
or
e
,
non
-
tr
iv
ia
l
e
le
m
e
nt
s
s
houl
d
be
w
id
e
ne
d
to
ot
he
r
s
ubt
a
s
ks
of
le
ga
l
ju
dge
pr
e
di
c
ti
on
w
it
h
c
om
pl
e
x de
pe
nde
nc
ie
s
.
T
he
c
ur
r
e
nt
s
tu
dy
r
e
s
e
a
r
c
he
r
s
c
la
s
s
if
ie
d
th
e
m
e
th
od
s
us
in
g
two typ
e
s
:
s
in
gl
e
c
la
s
s
if
ie
r
a
nd
c
om
bi
ne
d
c
la
s
s
if
ie
r
.
S
ubs
e
que
nt
ly
,
th
e
r
e
s
e
a
r
c
he
r
s
id
e
nt
if
ie
d
th
e
na
m
e
of
th
e
c
la
s
s
if
ie
r
(
s
)
in
vol
ve
d
a
s
th
e
pr
e
di
c
ti
on
m
ode
l.
th
e
s
in
gl
e
c
la
s
s
if
ie
r
r
e
f
e
r
s
to
a
n
in
di
vi
dua
l
m
ode
l
of
m
a
c
hi
ne
le
a
r
ni
ng
th
a
t
i
s
u
s
e
d
in
th
e
pr
e
di
c
ti
on.
I
n
c
ont
r
a
s
t,
c
om
bi
ne
d
c
la
s
s
if
ie
r
s
r
e
f
e
r
to
a
n
e
ns
e
m
bl
e
m
ode
l
t
ha
t
us
e
d
m
or
e
th
a
n
one
c
la
s
s
if
ie
r
in
m
a
ki
ng
pr
e
di
c
ti
ons
.
A
s
s
how
n
in
T
a
bl
e
5,
th
e
m
os
t
c
om
m
on
c
la
s
s
if
ie
r
is
th
e
s
uppor
t
ve
c
to
r
m
a
c
hi
ne
(
S
V
M
)
.
A
c
c
or
di
ng
to
t
he
c
ur
r
e
nt
S
L
R
, s
ix
pa
pe
r
s
pr
opos
e
d S
V
M
a
s
t
he
pr
e
di
c
ti
on
m
ode
l
in
va
r
io
us
c
a
s
e
s
.
N
e
ve
r
th
e
le
s
s
, t
hi
s
f
in
di
ng c
a
nnot
be
c
on
c
lu
de
d a
s
t
he
pr
e
f
e
r
r
e
d m
e
th
od i
n pr
e
di
c
ti
on a
s
ot
he
r
m
ode
ls
a
ls
o di
s
pl
a
ye
d a
good pe
r
f
or
m
a
nc
e
i
n pr
e
di
c
ti
ng j
udi
c
ia
l
de
c
is
io
ns
de
pe
ndi
ng on c
a
s
e
s
.
T
he
e
n
s
e
m
bl
e
m
e
th
od
pr
o
vi
de
s
a
n
e
nha
nc
e
d
a
ppr
oa
c
h
w
he
n
c
om
pa
r
e
d
w
it
h
a
not
he
r
a
ppr
oa
c
h.
T
hus
,
th
e
r
e
s
e
a
r
c
he
r
s
c
onc
lu
de
d
th
a
t
th
is
r
e
s
e
a
r
c
h
a
r
e
a
i
s
s
ti
ll
n
e
w
a
nd
ope
n f
or
e
xpl
or
a
ti
on.
T
hi
s
r
e
s
e
a
r
c
h i
s
s
ti
ll
a
c
ti
ve
ly
ongoing
i
n t
he
r
e
c
e
nt
f
iv
e
ye
a
r
s
,
a
s
obs
e
r
ve
d
in
F
ig
ur
e
2.
T
he
r
e
f
or
e
,
a
gr
e
a
t
oppor
tu
ni
ty
is
pr
e
s
e
nt
f
or
f
u
r
th
e
r
r
e
s
e
a
r
c
h
c
onc
e
r
ni
ng
im
pl
e
m
e
nt
in
g m
a
c
hi
ne
l
e
a
r
ni
ng me
th
ods
i
n pr
e
di
c
ti
ng c
our
t
de
c
is
io
ns
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
10
, N
o.
4
,
D
e
c
e
m
be
r
20
2
1
:
1091
–
1102
1098
T
a
bl
e
5
.
S
um
m
a
r
y of
m
e
th
ods
us
e
d
T
ype
s
of
c
a
s
e
s
M
e
t
hod
C
l
a
s
s
i
f
i
e
r
(
s
)
i
nvol
ve
d
S
t
udy I
D
C
ons
t
r
uc
t
i
on
L
i
t
i
ga
t
i
on
S
i
ngl
e
C
l
a
s
s
i
f
i
e
r
B
oos
t
e
d D
e
c
i
s
i
on T
r
e
e
(
B
D
T
)
S1
P
a
r
t
i
c
l
e
S
w
a
r
m
O
pt
i
m
i
s
a
t
i
on (
P
S
O
)
S2
C
a
s
e
B
a
s
e
R
e
a
s
oni
ng (
C
B
R
)
S3
I
nt
e
gr
a
t
e
d P
r
e
di
c
t
i
on M
ode
l
(
I
P
M
)
S4
S
uppor
t
V
e
c
t
or
M
a
c
hi
ne
(
S
V
M
)
S5
T
w
o
-
L
a
ye
r
e
d F
uz
z
y L
ogi
c
S
22
C
om
bi
ne
d
C
l
a
s
s
i
f
i
e
r
G
r
a
di
e
nt
B
oos
t
i
ng D
e
c
i
s
i
on T
r
e
e
(
G
B
D
T
)
, k
-
ne
a
r
e
s
t
ne
i
ghbour
(
K
N
N
)
, M
ul
t
i
l
a
ye
r
P
e
r
c
e
pt
r
on (
M
L
P
)
S
20
C
r
i
m
e
S
i
ngl
e
C
l
a
s
s
i
f
i
e
r
S
uppor
t
V
e
c
t
or
M
a
c
hi
ne
(
S
V
M
)
S
7, S
8, S
13, S
16
R
a
ndom
F
or
e
s
t
(
R
F
)
S9
M
ul
t
i
T
a
s
k L
e
a
r
ni
ng (
M
T
L
)
S
11
C
l
a
s
s
i
f
i
c
a
t
i
on a
nd
R
e
gr
e
s
s
i
on T
r
e
e
s
(
C
A
R
T
)
S
14
C
onvol
ut
i
ona
l
N
e
ur
a
l
N
e
t
w
or
k (
C
N
N
)
S
17
C
om
bi
ne
d
C
l
a
s
s
i
f
i
e
r
B
i
di
r
e
c
t
i
ona
l
E
nc
ode
r
R
e
pr
e
s
e
nt
a
t
i
on f
r
om
T
r
a
ns
f
or
m
e
r
(
B
E
R
T
)
+
C
onvol
ut
i
ona
l
N
e
ur
a
l
N
e
t
w
or
k (
C
N
N
)
S
19
W
or
ke
r
t
ype
S
i
ngl
e
C
l
a
s
s
i
f
i
e
r
e
xt
e
nde
d M
ul
t
i
l
a
ye
r
P
e
r
c
e
pt
r
on (
e
M
L
P
)
S
18
D
i
s
e
nga
ge
m
e
nt
L
ogi
s
t
i
c
R
e
gr
e
s
s
i
on (
L
R
)
S
15
T
a
x L
a
w
N
a
ï
ve
B
a
ye
s
(
N
B
)
S
10
P
a
r
e
nt
a
l
R
i
ght
L
ogi
s
t
i
c
R
e
gr
e
s
s
i
on (
L
R
)
S6
D
i
vor
c
e
C
ogni
t
i
ve
C
om
put
i
ng F
r
a
m
e
w
or
k (
C
C
F
)
S
12
D
ow
r
y
S
uppor
t
V
e
c
t
or
M
a
c
hi
ne
(
S
V
M
)
S
21
3.6
.
P
e
r
f
or
m
an
c
e
o
f
t
h
e
M
ac
h
in
e
L
e
ar
n
in
g M
e
t
h
od
s
T
he
f
ol
lo
w
in
g
r
e
s
e
a
r
c
h
que
s
ti
on
is
a
ddr
e
s
s
e
d
in
th
i
s
s
e
c
ti
on:
(
RQ
3
)
H
ow
w
a
s
th
e
p
e
r
f
or
m
a
nc
e
of
m
a
c
hi
ne
le
a
r
ni
ng
m
e
th
ods
u
s
e
d
to
pr
e
di
c
t
ju
di
c
i
a
l
de
c
is
io
n
s
?
T
he
p
e
r
f
or
m
a
nc
e
of
th
e
pr
e
di
c
ti
on
m
ode
l
pr
opos
e
d s
houl
d be
a
s
s
e
s
s
e
d pr
io
r
t
o unde
r
s
ta
ndi
ng t
he
a
ppr
oa
c
h us
e
d.
th
e
e
f
f
ic
ie
nc
y of
a
ny ma
c
hi
ne
l
e
a
r
ni
ng
m
ode
l
c
a
n
be
m
e
a
s
ur
e
d
th
r
ough
k
-
f
ol
d
c
r
os
s
-
va
li
da
ti
on,
a
c
c
ur
a
c
y,
s
e
ns
it
iv
it
y,
s
pe
c
if
ic
it
y,
r
e
c
a
ll
,
pr
e
c
is
io
n,
a
nd
F
-
m
e
a
s
ur
e
[
63]
.
B
a
s
e
d
on
th
e
obs
e
r
va
ti
ons
f
r
om
th
e
22
r
e
vi
e
w
e
d
pa
pe
r
s
,
m
os
t
r
e
s
e
a
r
c
h
e
r
s
us
e
d
a
c
c
ur
a
c
y,
pr
e
c
is
io
n,
r
e
c
a
ll
a
nd
F
-
m
e
a
s
ur
e
in
e
va
lu
a
ti
ng
t
he
pe
r
f
o
r
m
a
nc
e
o
f
th
e
ir
m
ode
ls
.
F
-
m
e
a
s
ur
e
,
pr
e
c
is
io
n
a
nd r
e
c
a
ll
a
r
e
f
r
e
que
nt
ly
ut
il
is
e
d
in
e
xt
r
a
c
ti
ng
in
f
or
m
a
ti
on
a
s
pe
r
f
or
m
a
nc
e
m
e
a
s
ur
e
m
e
nt
s
in
c
e
m
a
c
hi
ne
le
a
r
ni
ng
pe
r
f
or
m
a
nc
e
a
s
s
e
s
s
m
e
nt
s
i
nc
lu
de
s
p
e
c
if
ic
t
r
a
de
-
of
f
l
e
ve
ls
be
tw
e
e
n t
r
ue
pos
it
iv
e
a
nd t
r
ue
ne
g
a
ti
ve
r
a
te
s
[
63]
.
T
a
bl
e
6 s
um
m
a
r
is
e
s
t
he
i
nf
or
m
a
ti
on r
e
ga
r
di
ng
a
c
c
ur
a
c
y, pr
e
c
is
io
n, r
e
c
a
ll
or
s
e
ns
it
iv
it
y a
da
pt
e
d f
r
om
[
21]
.
T
he
r
e
a
r
e
f
our
im
por
ta
nt
te
r
m
s
us
e
d
in
m
e
a
s
ur
in
g
th
e
pe
r
f
or
m
a
nc
e
m
e
tr
ic
s
,
na
m
e
ly
tr
ue
po
s
it
iv
e
(
tp
)
,
tr
ue
ne
ga
ti
ve
(
tn
)
,
f
a
ls
e
pos
it
iv
e
(
f
p)
a
nd
f
a
ls
e
ne
ga
ti
ve
(
f
n)
[
21]
.
E
a
r
li
e
r
r
e
s
e
a
r
c
h
(
S
1,
S
2,
S
3
a
nd
S
4)
us
e
d
di
f
f
e
r
e
nt
a
ppr
oa
c
he
s
in
e
va
lu
a
ti
ng
th
e
pe
r
f
or
m
a
nc
e
of
th
e
m
e
th
ods
us
e
d.
th
e
a
ve
r
a
ge
pr
e
di
c
ti
on
r
a
te
ge
ne
r
a
te
d
in
th
e
r
e
por
te
d
s
tu
dy
is
w
it
hi
n
th
e
r
a
nge
o
f
80
%
to
91%
.
N
e
ve
r
th
e
le
s
s
,
th
e
s
tu
dy
w
a
s
e
xpa
nde
d
in
to
th
e
ne
xt
s
ta
ge
by
a
dj
us
ti
ng t
he
numbe
r
a
nd f
or
m
a
t
of
a
tt
r
ib
ut
e
s
a
nd t
he
numbe
r
of
c
a
s
e
s
u
s
e
d t
o be
tt
e
r
pr
e
di
c
t
r
a
te
s
[
67]
.
T
a
bl
e
6
.
P
e
r
f
or
m
a
nc
e
m
e
tr
ic
s
f
or
m
ul
a
M
e
a
s
ur
e
of
P
e
r
f
or
m
a
nc
e
D
e
s
c
r
i
pt
i
on
F
or
m
ul
a
A
c
c
ur
a
c
y
T
he
r
a
t
i
o of
a
c
or
r
e
c
t
l
y pr
e
di
c
t
e
d
r
e
s
ul
t
t
o t
he
t
ot
a
l
a
c
t
ua
l
r
e
s
ul
t
tp
+
tn
tp
+
tn
+
fp
+
fn
P
r
e
c
i
s
i
on
T
he
r
a
t
i
o of
a
c
or
r
e
c
t
l
y pr
e
di
c
t
e
d pos
i
t
i
ve
r
e
s
ul
t
t
o t
he
t
ot
a
l
pos
i
t
i
ve
pr
e
di
c
t
e
d r
e
s
ul
t
tp
tp
+
fp
R
e
c
a
l
l
of
S
e
ns
i
t
i
vi
t
y
T
he
r
a
t
i
o of
a
c
or
r
e
c
t
l
y pr
e
di
c
t
e
d pos
i
t
i
ve
r
e
s
ul
t
t
o t
he
t
ot
a
l
r
e
s
ul
t
tp
tp
+
fn
F
1 S
c
or
e
T
he
w
e
i
ght
e
d a
ve
r
a
ge
of
pr
e
c
i
s
i
on a
nd r
e
c
a
l
l
i
f
t
he
c
l
a
s
s
di
s
t
r
i
but
i
on i
s
une
ve
n
2
(
r
e
c
a
l
l
∗
p
r
e
c
i
s
i
on
)
r
e
c
a
l
l
+
p
r
e
c
i
s
i
on
T
he
m
o
s
t
in
tr
ig
ui
ng f
in
di
ng of
t
he
S
L
R
f
ound
i
s
t
ha
t
16 out
of
t
he
22
s
e
le
c
te
d
r
e
vi
e
w
p
a
pe
r
s
obt
a
in
e
d
m
or
e
th
a
n 80%
of
a
c
c
ur
a
c
y, pr
e
c
is
io
n or
pr
e
di
c
ti
on r
a
te
t
hr
ough the
e
va
lu
a
ti
on pr
oc
e
s
s
. O
nl
y f
our
pa
pe
r
s
(
S
7,
S
10
S
15
a
nd S
22)
obt
a
in
e
d t
he
r
a
nge
of
a
c
c
ur
a
c
y
or
pr
e
c
is
io
n
o
f
50%
to
70%
.
C
onve
r
s
e
ly
,
two
pa
pe
r
s
(
S
6
a
nd
S
13)
di
d
not
di
s
c
u
s
s
th
e
pe
r
f
or
m
a
nc
e
of
th
e
i
r
pr
e
di
c
ti
on
m
od
e
ls
in
d
e
ta
il
.
th
e
s
um
m
a
r
y
of
th
e
pe
r
f
or
m
a
nc
e
r
e
s
ul
ts
of
t
he
22
r
e
vi
e
w
e
d
pa
pe
r
s
i
s
pr
e
s
e
nt
e
d i
n
T
a
bl
e
7.
T
hi
s
a
ppr
oa
c
h e
xpl
ic
it
ly
obs
e
r
ve
d t
ha
t
th
e
pr
e
di
c
ti
on
m
ode
l
c
oul
d be
a
r
e
li
a
bl
e
s
uppor
ti
ng t
ool
i
n de
te
r
m
in
in
g c
our
t
de
c
is
io
ns
a
s
t
he
m
ode
l
s
’
pe
r
f
or
m
a
nc
e
a
c
hi
e
ve
d
m
or
e
t
ha
n 70%
ove
r
a
ll
a
c
c
ur
a
c
y.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
A
s
y
s
te
m
at
ic
l
it
e
r
at
u
r
e
r
e
v
ie
w
of
m
ac
hi
n
e
l
e
ar
ni
ng m
e
th
od
s
i
n
pr
e
di
c
ti
ng
…
(
N
ur
A
qi
la
h K
hadi
ja
h R
os
il
i)
1099
T
a
bl
e
7
.
R
e
s
ul
ts
of
pe
r
f
or
m
a
nc
e
S
t
udy I
D
M
ode
l
R
e
s
ul
t
s
A
c
c
ur
a
c
y
P
r
e
c
i
s
i
on
R
e
c
a
l
l
F
1 S
c
or
e
P
r
e
di
c
t
i
on r
a
t
e
S1
B
D
T
90
S2
PSO
80
S3
CBR
84
S4
I
P
M
91
S5
S
V
M
98
98
98
98
S6
LR
S7
S
V
M
79
S8
S
V
M
98
95
97
S9
RF
70
70
69
S
10
NB
57
57
57
S
11
M
T
L
95.6
75.9
69.6
70.9
S
12
CCF
71.22
74.17
72.65
S
13
S
V
M
S
14
C
A
R
T
91.86
92.86
90.7
91.76
S
15
LR
74.7
74.4
76.2
S
16
S
V
M
92
91
91
S
17
C
N
N
88.75
86.27
S
18
e
M
L
P
91.7
89.4
90.6
90
S
19
B
E
R
T
, C
N
N
89.7
89.7
89.6
S
20
G
B
D
T
, K
N
N
, M
L
P
96.42
96.66
96.38
96.03
S
21
S
V
M
93
93
93
92
S
22
T
w
o
-
l
a
ye
r
e
d F
uz
z
y L
ogi
c
73.9
4.
C
O
N
C
L
U
S
I
O
N
T
hi
s
s
tu
dy
ha
s
pr
e
s
e
nt
e
d
a
n
in
ve
s
ti
ga
ti
on
r
e
ga
r
di
ng
pr
e
di
c
ti
ng
c
our
t
de
c
is
io
ns
u
s
in
g
m
a
c
hi
ne
l
e
a
r
ni
ng
m
e
th
ods
.
T
he
im
por
ta
nc
e
of
pr
e
di
c
ti
ng
ju
di
c
ia
l
de
c
is
io
ns
c
a
n
be
id
e
nt
if
ie
d
in
va
r
io
us
c
a
s
e
s
a
nd
f
r
om
th
e
r
e
s
e
a
r
c
h
out
c
om
e
obt
a
in
e
d.
T
hi
s
a
ppr
oa
c
h
c
a
n
im
pr
ovi
s
e
th
e
l
e
ga
l
s
ys
t
e
m
by
m
a
ki
ng
it
m
or
e
s
y
s
te
m
a
ti
c
a
nd
r
e
li
a
bl
e
.
th
e
m
e
th
ods
a
nd
f
e
a
tu
r
e
s
de
r
iv
e
d
f
r
om
th
e
f
in
di
ngs
c
oul
d
f
il
l
th
e
e
xi
s
ti
ng
ga
ps
in
th
e
s
tu
dy
a
r
e
a
f
or
f
ut
ur
e
s
c
hol
a
r
ly
w
or
k.
T
hi
s
s
ys
te
m
a
ti
c
r
e
vi
e
w
s
tu
dy
i
s
e
xp
e
c
t
e
d
to
c
ont
r
ib
ut
e
to
th
e
body
of
knowle
dge
by
pr
ovi
di
ng a
n ove
r
vi
e
w
r
e
ga
r
di
ng e
x
is
ti
ng mode
ls
us
e
d i
n pr
e
di
c
ti
ng j
udi
c
ia
l
de
c
is
io
ns
, t
he
pe
r
f
or
m
a
nc
e
of
t
he
pr
e
di
c
ti
ng
m
ode
l
a
nd
di
s
c
us
s
io
n
on
s
e
v
e
r
a
l
ty
pe
s
of
c
a
s
e
s
in
th
e
le
ga
l
s
y
s
te
m
th
a
t
a
da
pt
e
d
th
i
s
a
ppr
oa
c
h.
T
he
r
e
vi
e
w
a
ls
o
of
f
e
r
s
s
e
ve
r
a
l
r
e
c
om
m
e
nda
ti
ons
f
or
f
u
tu
r
e
s
tu
di
e
s
,
in
c
lu
di
ng
ne
w
ty
pe
s
of
c
a
s
e
s
f
or
p
r
e
di
c
ti
ng
ju
di
c
ia
l
de
c
is
io
ns
a
nd
a
ne
w
m
a
c
hi
ne
le
a
r
ni
ng
m
e
th
od
th
a
t
r
e
qui
r
e
s
a
c
om
bi
ne
d
c
la
s
s
if
ie
r
to
im
pr
ove
th
e
pr
e
di
c
ti
ng t
ool
s
’
pe
r
f
or
m
a
nc
e
.
A
C
K
N
O
WL
E
D
G
E
M
E
N
T
S
T
he
a
ut
hor
s
w
oul
d
li
ke
to
th
a
nk
U
ni
ve
r
s
it
i
T
e
knol
ogi
M
a
la
ys
ia
,
U
ni
ve
r
s
it
i
T
un
H
us
s
e
in
O
nn,
U
ni
ve
r
s
it
i
S
a
in
s
M
a
la
y
s
ia
a
nd U
ni
ve
r
s
it
a
s
A
hm
a
d D
a
hl
a
n t
o s
u
ppor
t
th
is
c
ol
la
bor
a
ti
ve
r
e
s
e
a
r
c
h
.
R
E
F
E
R
E
N
C
E
S
[1]
D.
M.
Katz,
M.
J.
Bommarito,
and
J.
Blackman,
“A
general
appr
oac
h
for
predicting
the
behavior
of
the
Supreme
Court of
the Unite
d States,”
PLoS One
, vol. 12, no. 4, pp. e0174698
–
e0174698
, 2017.
[2]
N.
Kamaruddin,
R.
D.
Safiyah,
and
A.
Wahab,
“Small
and
medi
um
enterprise
business
solutions
using
dat
a
visualization,”
B
ulletin
of
Electrical
Engineering
and
Informatics
,
vol.
9,
no.
6.
pp.
2562
–
2568,
2020,
doi:
10.11591/eei.v9i6.2463.
[3]
J.
Shetty,
B.
Sathish
Babu,
and
G.
Shobha,
“Proactive
cloud
service
assurance
framework
for
fault
remediation
i
n
cloud
environment,”
Int
ernatio
nal
Journal
of
Electrica
l
and
Comput
er
Engineeri
ng
,
vol.
10,
no.
1.
pp.
987
–
996,
2020, doi: 10.11591/ijece.v10i1.pp987
-
996.
[4]
A.
V
Zadgaonkar
and
A.
J.
Agrawal,
“An
overview
of
information
extr
action
techniques
for
legal
document
analysi
s
and
proc
essing,
”
Internati
onal
Journal
of
Electrical
and
Computer
Eng
ineering
,
vol.
11,
no.
6.
pp.
5450
–
5457,
2021,
doi: 10.11591/ijece.v11i6.pp5450
-
5457.
[5]
I. E. Olufem
i, A.
A. Adebiyi
,
F. A. Ibi
kunle,
M. O.
Adebiyi,
and
O. O.
Oludayo, “R
esearch trends on
CAPTC
HA: A
systematic literatur
e,”
Internati
onal Journ
al
of Electr
ical
and Comp
ute
r Engineeri
ng
, vol. 11,
no. 5. pp.
4300
–
4312,
2021, doi: 10.11591/ijece.v11i5.pp4300
-
4312.
[6]
A.
Murugan,
T.
Chechare,
B.
Muruganantham,
and
S.
Ganesh
Kum
ar,
“Healthcare
informat
ion
exchange
usin
g
blockchain technology,”
Internati
onal Journ
al
of Electr
ical and
Comp
uter Engi
neering
, vol. 10,
no. 1. pp.
421
–
426,
2020, doi: 10.11591/ijece.v10i1.pp421
-
426.
[7]
V. Siddesh P
adala, K. Gandhi, and D.
V Pushpalatha, “Machi
ne
learni
ng: The
new language for applica
tions,”
IAES
Internati
onal Journ
al of Ar
tificial
Intellig
ence
, vol. 8, no. 4. pp. 411
–
4
21, 2019, doi: 10.11591/ijai.v8.i4.pp411
-
421.
[8]
A.
M.
Muad,
N.
S.
M.
Bahaman,
A.
Hussain,
and
M.
Y.
P.
M.
Yusof,
“Tooth
segmentation
using
dyna
mic
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
10
, N
o.
4
,
D
e
c
e
m
be
r
20
2
1
:
1091
–
1102
1100
programming
-
gradient
inverse
coefficient
of
variation,”
Bulletin
of
Electrical
Engineering
and
Informatics
,
vol.
8,
no. 1. pp. 253
–
260, 2019, doi: 10.11591/eei.v8i1.1446.
[9]
G.
Hima
Bindu,
C.
Anuradha,
and
P.
S.
R.
Chandra
Murty,
“A
survey
on
multimedia
content
protection
mechanisms,”
Internati
onal
Journal
of
Electrical
and
Computer
Engineering
,
vol.
8,
no.
6.
pp.
4204
–
4211,
2018,
doi:
10.11591/ijece.v8i6.pp.4204
-
4211.
[10]
S.
K.
Srivastava,
“Artifi
cial
intelligence:
Way
forward
for
India,
”
IAES
International
Journal
of
Artificial
Intelligen
ce
, vol. 7, no. 1. pp. 19
–
32, 2018, doi: 10.11591/ijai.v7.i1.pp19
-
32.
[11]
N. Madhusudan and L.
Manjunatha Rao, “Ins
ights
on extent of effe
cti
veness, trend, and gap in existing framew
ork
s
for
e
-
procurement
system,”
Internati
onal
Journal
of
Electrical
and
C
omputer
Engineering
,
vol.
6,
no.
2.
pp.
751
–
758, 2016, doi: 10.11591/ijece.v6i2.9028.
[12]
O.
J.
Ayangbekun,
O.
F.
Bankole,
and
B.
A.
Saka,
“Analysi
s
of
security
mechanism
s
in
Nigeria
E
-
banking
platform,”
Internati
onal
Journal
of
Electrical
and
Computer
Engineering
,
vol.
4,
no.
6.
pp.
837
–
847,
2014,
[Online].
Available:
https://www.scopus.com/inward/record.uri?eid=
2
-
s2.0
-
84928530364&partnerID
=
40&md5=
fa9dcf0ac0f9e
8933c4121ec469d9adf.
[13]
Z.
Anna
and
E.
Vladimir,
“State
regulation
of
the
IoT
in
the
Russia
n
Federation:
Fundamentals
and
challenges,”
Internati
onal
Journal
of
Electrical
and
Computer
Engineering
,
v
ol.
11,
no.
5.
pp.
4542
–
4549,
2021,
doi:
10.11591/ijece.v11i5.pp4542
-
4549.
[14]
P.
Anki
and
A.
Bustamam,
“Measuring
the
accur
acy
of
LSTM
and
B
iLSTM
models
in
the
application
of
artificial
intelligence
by applying c
hatbot progr
amme,”
Indonesian
Journal of
Electrical
Engineerin
g and Com
puter
Science
,
vol. 23, no. 1. pp. 197
–
205, 2021, doi: 10.11591/ijeecs.v23.i1.pp197
-
205.
[15]
I.
Odun
-
Ayo,
O.
Alagbe,
and
J.
Yahaya,
“A
systematic
mapping
stu
dy
of
security,
trust
and
privacy
in
clouds,
”
Bulletin
of
Electrical
Engineering
and
Informatics
,
vol.
1
0,
no.
3.
pp.
1598
–
1610,
2021,
doi:
10.11591/eei.v10i3.1887.
[16]
N.
Nizamuddin
and
A.
A
bugabah,
“Blockchain
for
automotive:
An
in
sight
towards
the
IPFS
blockchain
-
based
auto
insurance
sector,”
Internati
onal
Journal
of
Electrical
and
Computer
Engineering
,
vol.
11,
no.
3.
pp.
2443
–
2456,
2021, doi: 10.11591/ijece.v11i3.pp2443
-
2456.
[17]
I.
Veik
sa,
“Protection
of
computer
-
generated
works
in
the
era
of
new
technologies,”
IAES
International
Journal
o
f
Artificial Intellig
ence
, vol. 10, no. 1. pp. 234
–
243, 2021, doi: 10.11591/ijai.v10.i1.pp234
-
243.
[18]
S.
Nasir,
S.
M.
Al
-
Qaraawi,
and
M.
S.
Croock,
“
QR
based
mana
gement
system
for
plants
shopping
center,”
Indonesian
Journal
of
Electrical
Engineering
and
Computer
Scien
ce
,
vol.
19,
no.
2.
pp.
931
–
939,
2020,
doi
:
10.11591/ijeecs.v19.i2.pp931
-
939.
[19]
V.
Kasinathan,
A.
Mustapha,
M.
F.
C.
Abdul
Rani,
and
S
.
A.
Mostafa, “The
role
of
chatterbots
in
enhancing
tourism
:
A
case
study
of
Penang
tourism
spots,”
IAES
International
Journal
of
Artificial
Intelligence
,
vol.
9,
no.
4.
pp.
569
–
575, 2020, doi: 10.11591/ijai.v9.i4.pp569
-
575.
[20]
A.
Zharova,
“The
protect
m
obile
user
data
in
Russia,”
Internati
onal
Jou
rnal
of
Electrical
and
Computer
Engineering
,
vol. 10, no. 3. pp. 3184
–
3192, 2020, doi: 10.11591/ijece.v10i3.pp3184
-
3192.
[21]
P.
Bhilare,
N.
Parab,
N.
Soni,
and
B.
Thakur,
“Predicting
outcome
of
judicial
cases
a
nd
analysis
using
machin
e
learning,”
Int. Res. J.
Eng. Techno
l.
, vol. 6, no. 3, pp. 326
–
330, 2019.
[22]
B.
Kitchenham
et
al.
,
“Systematic
literature
reviews
in
software
engin
eering
--
a
tertiary
study,”
Inf.
Softw.
Technol.
,
vol. 52, no. 8, pp. 792
–
805, 2010
.
[23]
K.
J.
Danjuma,
“Performance
evaluation
of
machine
learning
algorithms
in
post
-
operative
life
expectancy
in
the
lung
cancer patien
ts,”
arXiv Prepr. arXiv1504.04646
, 2015.
[24]
Hertina
et al.
, “Data mining applied about polygamy using sentiment
a
nalys
is on twitters
in indonesian
perce
ption,”
Bulletin
of
Electrical
Engineering
and
Informatics
,
vol.
1
0,
no.
4.
pp.
2231
–
2236,
2021,
doi:
10.11591/EEI.V10I4.2325.
[25]
A.
Al
Mamun,
P.
P.
Em,
and
J.
Hossen,
“Lane
marking
detection
using
simple
encode
decode
d
eep
learning
technique:
SegNet,”
Internati
onal
Journal
of
Electrical
and
Compute
r
Engineering
,
vol.
11,
no.
4.
pp.
3032
–
3039
,
2021, doi: 10.11591/ijece.v11i4.pp3032
-
3039.
[26]
R.
Usha
and
K.
Perumal,
“A
modified
fractal
texture
image
analysis
ba
sed
on
gray
scale
morphology
for
multi
-
mode
l
views
in
MR
Brain,”
Indonesian
Journal
of
Electrical
Engineering
an
d
Computer
Science
,
vol.
21,
no.
1.
pp.
154
–
163, 2021, doi: 10.11591/ijeecs.v21.i1.pp154
-
163.
[27]
J.
Y.
W.
Jien,
A.
Baharum,
S.
H.
A.
Wahab,
N.
Saad,
M.
Om
ar,
and
N.
A.
M.
Noor,
“Age
-
based
facial
recognition
using
convoluted
neural
network
deep
learning
algorithm,”
IAES
Inter
national
Journal
of
Artificial
Intelligence
,
vol
.
9, no. 3. pp. 424
–
428, 2020, doi: 10.11591/ijai.v9.i3.pp424
-
428.
[28]
A.
Al
-
Imam,
M.
A.
Motyka,
and
M.
Z.
Jędrzejko,
“Conflicting
opinions
in
connection
with
digital
superintelligence,
”
IAES
International
Journal
of
Artificial
Intelligence
,
vol.
9,
no.
2.
pp.
3
36
–
348,
2020,
doi:
10.11591/ijai.v9.i2.pp336
-
348.
[29]
L.
M.
Shi,
A.
Mustapha,
and
Y.
M.
M.
Hassim,
“Predicting
fatalities
am
ong
shark
attacks:
Comparison
of
classifiers,
”
IAES
International
Journal
of
Artificial
Intelligence
,
vol.
8,
no.
4.
pp.
3
60
–
366,
2019,
doi:
10.11591/ijai.v8.i4.pp360
-
366.
[30]
S.
A.
Diw
an,
“Proposed
study
on
evaluating
and
forecasting
the
resident
property
value
based
on
specific
determinants
by
case
base
reasoning
and
artificial
neur
al
network
approach,”
Indonesian
Journal
of
Electrical
Engineering and Computer Science
, vol. 17, no. 3.
pp. 1467
–
1473, 2019, doi: 10.11591/ijeecs.v17.i3.pp1467
-
1473.
[31]
H.
A.
Razak,
M.
A.
M.
Saleh,
and
N.
M.
Tahir,
“Review
on
anomalous
gait
behavior
detection
using
machine
learning
algorit
hms,”
Bulletin
of
Electrical
Engineering
and
Informatics
,
vol.
9,
no
.
5.
pp.
2090
–
2096,
2020,
doi:
Evaluation Warning : The document was created with Spire.PDF for Python.