I
A
E
S
I
n
t
e
r
n
at
io
n
al
Jou
r
n
al
of
A
r
t
if
ic
ia
l
I
n
t
e
ll
ig
e
n
c
e
(
I
J
-
AI
)
V
ol
.
14
, N
o.
3
,
J
une
2025
, pp.
1781
~
1789
I
S
S
N
:
2252
-
8938
,
D
O
I
:
10.11591/
ij
a
i.
v
14
.i
3
.pp
1781
-
1789
1781
Jou
r
n
al
h
om
e
page
:
ht
tp
:
//
ij
ai
.
ia
e
s
c
or
e
.c
om
N
ove
l
p
r
e
e
m
p
t
i
ve
i
n
t
e
l
l
i
ge
n
t
ar
t
i
f
i
c
i
al
i
n
t
e
l
l
i
ge
n
c
e
-
m
od
e
l
f
or
d
e
t
e
c
t
i
n
g i
n
c
on
si
st
e
n
c
y d
u
r
i
n
g sof
t
w
ar
e
t
e
st
i
n
g
S
an
ge
e
t
h
a G
ovi
n
d
a
1
, B
.
G
.
P
r
as
an
t
h
i
2
, A
gn
e
s
N
al
in
i
V
in
c
e
n
t
3
1
D
e
pa
r
t
m
e
nt
o
f
C
om
put
e
r
S
c
i
e
nc
e
,
C
hr
i
s
t
U
ni
ve
r
s
i
t
y
,
B
a
nga
l
or
e
C
e
nt
r
a
l
C
a
m
p
us
,
B
a
nga
l
u
r
u
,
I
ndi
a
2
D
e
pa
r
t
m
e
nt
o
f
C
om
put
e
r
S
c
i
e
nc
e
a
nd
A
ppl
i
c
a
t
i
ons
,
St
.
J
os
e
phs
U
ni
ve
r
s
i
t
y
,
B
a
nga
l
u
r
u
,
I
ndi
a
.
3
F
a
c
ul
t
y
o
f
I
nf
or
m
a
t
i
on T
e
c
hnol
ogy
,
A
M
I
T
Y
I
ns
t
i
t
ut
e
o
f
H
i
ghe
r
E
duc
a
t
i
on
,
Q
u
a
t
r
e
B
or
ne
s
,
M
a
ur
i
t
i
us
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
A
pr
10, 2024
R
e
vi
s
e
d
F
e
b 26, 2025
A
c
c
e
pt
e
d
M
a
r
15, 2025
The
contribution
of
artificial
intell
igence
(AI)
-
based
modelling
is
highly
significant
in
automating
the
software
testing
process;
thereby
enhanci
ng
the
cost,
resources,
and
productivity
while
performing
testing.
Revi
ew
of
existin
g
AI
-
models
towards
software
testing
showcases
yet
an
open
-
scope
for
further
improvement
as
yet
the
conventional
AI
-
mod
el
suffers
from
various
challenges
especially
in
perspective
of
test
case
gene
ration.
Therefore,
the
proposed
scheme
presents
a
novel
preemptive
inte
lligent
computat
ional
framework
that
harnesses
a
unique
ensembled
AI
-
mo
del
for
generating
and
executing
highly
precise
and
optimized
test
-
cases
resul
ting
in
an
outcome
of
adversary
or
inconsistencies
associated
with
test
case
s.
The
ensembled
AI
-
model
uses
both
unsupervised
and
supervised
le
arning
approaches
on
publicl
y
available
outlier
dataset.
The
benchmarked
o
utcome
exhibit
s
supervis
ed
learning
-
based
AI
-
model
to
offer
21%
of
reduce
d
error
and
1.6%
of
reduced
processing
time
in
contrast
to
unsupervised
s
cheme
while performing so
ftware testing.
K
e
y
w
o
r
d
s
:
A
r
ti
f
ic
ia
l
in
te
ll
ig
e
nc
e
A
ut
om
a
ti
on
E
r
r
or
I
nc
ons
is
te
nc
y
S
of
twa
r
e
t
e
s
ti
ng
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
:
B
.
G
.
P
r
a
s
a
nt
hi
D
e
pa
r
tm
e
nt
of
C
om
put
e
r
S
c
ie
nc
e
a
nd
A
ppl
ic
a
ti
ons
, S
t.
J
o
s
e
ph
s
U
ni
ve
r
s
it
y
36, L
a
ngf
or
d R
d, L
a
ngf
or
d
G
a
r
de
ns
, B
e
nga
lu
r
u, K
a
r
na
ta
ka
560027
,
I
ndi
a
E
m
a
il
:
pr
a
s
a
nt
hi
.b.g@
s
ju
.e
du.i
n
1.
I
N
T
R
O
D
U
C
T
I
O
N
T
he
m
e
c
ha
ni
s
m
of
s
of
twa
r
e
te
s
ti
ng
pl
a
ys
a
c
r
uc
i
a
l
a
nd
in
te
gr
a
l
r
ol
e
in
s
of
twa
r
e
de
v
e
lo
pm
e
nt
th
a
t
c
ons
is
ts
of
a
c
r
it
ic
a
l
a
na
ly
s
is
of
ove
r
a
ll
qua
li
ty
to
e
n
s
ur
e
th
a
t
it
m
e
e
ts
de
m
a
nd
e
d
s
pe
c
if
ic
a
ti
on
[
1]
.
T
he
pr
im
a
r
y
a
ge
nda
of
di
f
f
e
r
e
nt
ty
pe
s
of
s
of
twa
r
e
te
s
ti
ng
is
to
pe
r
f
or
m
va
li
da
ti
on,
ve
r
if
ic
a
ti
on,
id
e
nt
i
f
ic
a
ti
on
of
de
f
e
c
ts
, a
nd e
nha
nc
in
g t
he
qua
li
ty
[
2]
.
H
ow
e
ve
r
, t
he
r
e
a
r
e
s
om
e
e
s
s
e
nt
ia
l
c
ha
ll
e
nge
s
a
s
s
oc
ia
te
d w
it
h s
of
twa
r
e
te
s
ti
ng
[
3]
.
S
o
m
e
of
th
e
not
a
bl
e
c
ha
ll
e
nge
s
a
r
e
c
om
pl
e
xi
ty
o
f
s
of
twa
r
e
,
c
ha
ngi
ng
r
e
qui
r
e
m
e
nt
s
,
ti
m
e
a
nd
r
e
s
our
c
e
,
la
c
k
of
te
s
t
da
ta
,
de
pe
nde
nc
y
m
a
na
ge
m
e
nt
,
a
ut
om
a
ti
on
c
ha
ll
e
nge
s
,
non
-
de
te
r
m
in
is
ti
c
be
ha
vi
or
,
pl
a
tf
or
m
a
nd
e
nvi
r
onm
e
nt
di
ve
r
s
it
y,
a
nd
pe
r
f
or
m
a
nc
e
a
nd
s
c
a
la
bi
li
ty
te
s
ti
ng
[
4]
–
[
7]
.
A
t
pr
e
s
e
nt
,
it
is
not
e
d
th
a
t
a
r
ti
f
ic
ia
l
in
te
ll
ig
e
nc
e
(
A
I
)
ha
s
a
s
ig
ni
f
ic
a
nt
c
ont
r
ib
ut
io
n
t
ow
a
r
ds
in
c
or
por
a
ti
ng
a
ut
om
a
ti
on
in
s
of
twa
r
e
te
s
ti
ng
th
e
r
e
by
le
ve
r
a
gi
ng
th
e
e
f
f
ic
ie
nc
y
of
te
s
ti
ng
pr
oc
e
dur
e
[
8]
–
[
10]
.
T
he
r
e
a
r
e
va
r
io
us
s
tu
di
e
s
to
s
how
c
a
s
e
th
a
t
a
ut
om
a
te
d
ge
ne
r
a
ti
on
of
te
s
t
-
c
a
s
e
s
c
a
n
be
don
e
by
A
I
a
lg
or
it
hm
s
f
or
gi
ve
n
c
ode
,
s
pe
c
if
ic
a
ti
on,
a
nd
s
of
twa
r
e
r
e
qui
r
e
m
e
nt
s
[
11]
.
S
uc
h
f
or
m
of
ge
ne
r
a
te
d
te
s
t
-
c
a
s
e
s
c
a
n
be
he
lp
f
ul
f
o
r
c
ove
r
in
g
va
r
ie
d
f
or
m
s
o
f
s
c
e
na
r
io
s
a
nd
c
om
pe
ti
ti
ve
c
a
s
e
s
in
or
dr
e
to
a
c
c
om
pl
is
h
a
br
oa
de
r
s
pe
c
tr
um
of
te
s
ti
ng.
A
I
a
ls
o
a
s
s
is
ts
in
of
f
e
r
in
g
te
s
t
pr
io
r
it
iz
a
ti
on
th
a
t
c
a
n
e
f
f
e
c
ti
ve
ly
a
na
ly
z
e
a
dve
r
s
a
r
ie
s
a
nd
a
ll
r
is
k
e
le
m
e
nt
s
c
onne
c
t
e
d
w
it
h
di
f
f
e
r
e
nt
c
om
pone
nt
s
of
s
of
twa
r
e
a
nd
a
c
c
or
di
ngl
y
pr
io
r
it
iz
e
s
th
e
te
s
t
c
a
s
e
s
[
12]
.
S
uc
h
f
e
a
tu
r
e
s
of
te
s
t
pr
io
r
it
iz
a
ti
on
c
a
n
be
us
e
d
f
or
m
or
e
e
m
pha
s
is
to
w
a
r
ds
r
ig
or
ou
s
te
s
ti
ng
on
pot
e
nt
ia
ll
y
e
xt
e
ns
iv
e
de
f
e
c
ts
a
nd
he
nc
e
r
e
s
our
c
e
s
of
te
s
ti
ng
a
r
e
hi
ghl
y
opt
im
iz
e
d
unl
ik
e
c
onv
e
nt
io
na
l
te
s
ti
ng.
A
I
-
pow
e
r
e
d
te
s
ti
ng
to
ol
s
c
a
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
. 14, No. 3, J
une
2025
:
1781
-
1789
1782
a
ut
om
a
te
th
e
e
x
e
c
u
ti
on
of
t
e
s
t
c
a
s
e
s
,
r
e
duc
i
ng
th
e
ne
e
d
f
or
m
a
nua
l
i
nt
e
r
v
e
nt
io
n
a
nd
s
p
e
e
di
ng
up
th
e
te
s
ti
n
g
pr
oc
e
s
s
.
T
hi
s
in
c
l
ude
s
bot
h
f
unc
t
io
na
l
a
nd
non
-
f
un
c
ti
on
a
l
t
e
s
ti
ng,
s
u
c
h
a
s
r
e
gr
e
s
s
io
n
t
e
s
ti
ng,
pe
r
f
or
m
a
nc
e
te
s
ti
ng,
a
nd
s
e
c
ur
it
y
t
e
s
ti
ng
[
13]
.
A
I
a
lg
or
it
hm
s
c
a
n
a
na
ly
z
e
hi
s
to
r
i
c
a
l
da
t
a
f
r
om
pr
e
vi
ou
s
t
e
s
t
in
g
c
y
c
le
s
t
o
pr
e
di
c
t
pot
e
n
ti
a
l
d
e
f
e
c
ts
or
a
r
e
a
s
of
t
he
s
of
tw
a
r
e
th
a
t
a
r
e
m
or
e
l
ik
e
ly
to
c
ont
a
in
bu
gs
.
T
hi
s
h
e
lp
s
te
s
te
r
s
f
oc
us
th
e
ir
e
f
f
or
t
s
on high
-
r
i
s
k
a
r
e
a
s
a
n
d
a
ll
oc
a
te
r
e
s
our
c
e
s
m
or
e
e
f
f
e
c
ti
ve
l
y
[
14]
.
A
I
c
a
n
a
c
t
a
s
dyn
a
m
i
c
te
s
t
or
a
c
le
s
by l
e
a
r
ni
n
g t
he
e
xpe
c
te
d b
e
ha
vi
or
of
t
he
s
of
twa
r
e
t
hr
ough tr
a
in
i
ng on his
t
or
ic
a
l
d
a
ta
or
u
s
e
r
i
nt
e
r
a
c
ti
on
s
. I
t
c
a
n
th
e
n
c
om
pa
r
e
th
e
a
c
tu
a
l
be
h
a
vi
or
of
th
e
s
of
tw
a
r
e
dur
in
g
te
s
t
in
g
w
it
h
t
he
e
xpe
c
te
d
be
h
a
vi
or
a
nd
i
de
nt
if
y
de
vi
a
t
io
n
s
or
a
nom
a
li
e
s
[
15]
.
A
I
t
e
c
hn
iq
ue
s
,
s
u
c
h
a
s
m
a
c
hi
ne
l
e
a
r
ni
ng
a
n
d
a
n
om
a
ly
de
t
e
c
t
io
n
a
lg
or
it
hm
s
,
c
a
n
id
e
nt
if
y
un
e
xp
e
c
t
e
d
pa
tt
e
r
n
s
or
de
vi
a
ti
on
s
f
r
om
nor
m
a
l
be
ha
v
io
r
in
th
e
s
of
t
w
a
r
e
un
de
r
te
s
t.
T
hi
s
c
a
n
h
e
lp
de
te
c
t
s
ubt
l
e
b
ugs
or
s
e
c
ur
it
y
vul
n
e
r
a
bi
li
ti
e
s
t
ha
t
m
a
y
not
be
a
p
pa
r
e
nt
th
r
oug
h
tr
a
di
ti
on
a
l
t
e
s
ti
ng me
th
o
ds
[
16]
.
N
a
tu
r
a
l
l
a
ngu
a
ge
pr
o
c
e
s
s
in
g
(
N
L
P
)
c
a
n
be
u
s
e
d
t
o a
na
l
yz
e
a
nd
unde
r
s
ta
nd
n
a
tu
r
a
l
l
a
ngu
a
ge
r
e
qui
r
e
m
e
nt
s
,
us
e
r
s
to
r
ie
s
,
a
nd
do
c
um
e
nt
a
ti
on.
A
I
-
p
ow
e
r
e
d
to
ol
s
c
a
n
e
xt
r
a
c
t
te
s
ta
bl
e
s
c
e
n
a
r
io
s
a
n
d
ge
n
e
r
a
t
e
te
s
t
c
a
s
e
s
di
r
e
c
tl
y
f
r
om
t
hi
s
te
xt
u
a
l
in
f
or
m
a
ti
o
n, i
m
pr
o
vi
ng t
e
s
t
c
o
ve
r
a
ge
a
nd
a
c
c
ur
a
c
y
[
1
7]
.
H
ow
e
ve
r
,
th
e
r
e
a
r
e
va
r
io
us
c
h
a
ll
e
nge
s
of
e
xi
s
ti
ng
s
ys
te
m
of
A
I
in
s
of
twa
r
e
te
s
ti
ng
a
s
f
ol
lo
w
in
g
:
i)
t
he
pr
im
a
r
y
c
ha
ll
e
nge
is
r
e
la
te
d
to
ov
e
r
f
it
ti
ng
by
ge
ne
r
a
ti
ng
te
s
t
c
a
s
e
s
th
a
t
a
r
e
onl
y
r
e
le
va
nt
to
to
th
e
s
pe
c
if
ic
tr
a
in
in
g
s
c
e
na
r
io
s
a
nd
m
a
y
not
ge
ne
r
a
li
z
e
w
e
ll
to
ne
w
,
uns
e
e
n
s
it
ua
ti
ons
;
ii
)
i
t
w
a
s
a
ls
o
not
e
d
th
a
t
AI
-
ge
ne
r
a
te
d
te
s
t
c
a
s
e
s
m
a
y
not
c
ove
r
a
ll
pos
s
ib
le
s
c
e
na
r
io
s
or
e
dge
c
a
s
e
s
,
le
a
di
ng
to
ga
ps
in
te
s
t
c
ove
r
a
ge
.
B
a
la
nc
in
g
be
twe
e
n
ge
ne
r
a
ti
ng
e
nough
di
ve
r
s
e
te
s
t
c
a
s
e
s
a
n
d
a
voi
di
ng
r
e
dunda
nt
o
r
ir
r
e
le
va
nt
one
s
is
a
c
ha
ll
e
nge
;
ii
i)
A
I
a
lg
or
it
hm
s
m
a
y
s
tr
uggl
e
to
unde
r
s
ta
nd
a
n
d
m
ode
l
th
e
in
tr
ic
a
c
ie
s
of
c
om
pl
e
x
s
of
twa
r
e
s
ys
te
m
s
a
c
c
ur
a
te
ly
.
T
hi
s
c
a
n
r
e
s
ul
t
in
th
e
ge
n
e
r
a
ti
on
of
te
s
t
c
a
s
e
s
th
a
t
ov
e
r
lo
ok
c
r
it
ic
a
l
in
te
r
a
c
ti
ons
or
de
pe
nde
nc
ie
s
w
it
hi
n
th
e
s
ys
te
m
,
le
a
di
ng
to
in
c
om
pl
e
te
te
s
ti
ng
;
iv
)
s
ys
te
m
s
th
a
t
e
xhi
bi
t
dyna
m
ic
be
ha
vi
or
o
r
ha
ve
f
r
e
que
nt
c
ha
nge
s
pos
e
c
ha
ll
e
nge
s
f
or
A
I
-
ba
s
e
d
te
s
t
c
a
s
e
ge
ne
r
a
ti
on.
A
I
m
ode
ls
m
a
y
s
tr
uggl
e
to
a
da
pt
qui
c
kl
y
to
c
ha
ng
e
s
in
th
e
s
of
twa
r
e
or
th
e
te
s
ti
ng
e
nvi
r
onm
e
nt
,
le
a
di
ng
to
out
d
a
te
d
or
in
e
f
f
e
c
ti
ve
te
s
t
c
a
s
e
s
;
a
nd
v)
A
I
m
a
y
s
tr
uggl
e
to
a
c
c
ur
a
te
ly
de
f
in
e
th
e
e
xpe
c
te
d
ou
tc
om
e
s
f
or
te
s
t
c
a
s
e
s
,
e
s
p
e
c
ia
ll
y
in
c
om
pl
e
x
s
ys
te
m
s
or
w
he
n
r
e
qui
r
e
m
e
nt
s
a
r
e
a
m
bi
guou
s
.
W
it
hout
c
le
a
r
or
a
c
le
s
,
it
'
s
c
ha
ll
e
ngi
ng
to
a
s
s
e
s
s
w
he
th
e
r
th
e
s
ys
te
m
be
ha
vi
or
i
s
c
or
r
e
c
t,
l
e
a
di
ng t
o
di
f
f
ic
ul
ti
e
s
i
n va
li
da
ti
ng t
he
ge
ne
r
a
te
d t
e
s
t
c
a
s
e
s
.
T
he
r
e
la
te
d
w
or
k
c
a
r
r
ie
d
out
to
w
a
r
ds
A
I
im
pl
e
m
e
nt
a
ti
on
in
s
of
twa
r
e
te
s
ti
ng
a
r
e
a
s
f
ol
lo
w
s
:
R
obi
s
c
o
a
nd
M
a
r
tí
ne
z
[
18]
ha
ve
de
ve
lo
p
e
d
a
r
is
k
e
va
lu
a
ti
on
s
c
he
m
e
us
in
g
m
a
c
hi
ne
le
a
r
ni
ng
a
nd
N
L
P
in
or
de
r
to
de
te
c
t
th
e
de
f
a
ul
te
r
s
.
E
va
lu
a
ti
on
of
th
e
e
xe
c
ut
a
bl
e
f
il
e
s
ha
s
be
e
n
c
a
r
r
ie
d
out
by
A
r
a
ke
ly
a
n
e
t
al
.
[
19]
w
he
r
e
c
onvolut
io
n
ne
twor
k
us
in
g
gr
a
phs
ha
s
be
e
n
im
pl
e
m
e
nt
e
d
to
w
a
r
ds
s
e
m
a
nt
ic
a
na
ly
s
is
of
th
e
vul
ne
r
a
bi
li
ty
of
da
ta
.
A
dopt
io
n
o
f
A
I
ha
s
a
ls
o
be
e
n
w
it
ne
s
s
e
d
in
in
ve
s
ti
ga
ti
ng
th
e
a
ndr
oi
d
-
ba
s
e
d
a
ppl
ic
a
ti
on
w
he
r
e
a
ge
ne
r
a
ti
on of
s
ta
te
f
ul
e
ve
nt
ha
s
be
e
n di
s
c
us
s
e
d i
n w
or
k of
Y
e
r
im
a
e
t
al
.
[
20]
us
in
g m
a
c
hi
ne
l
e
a
r
ni
ng
.
H
a
i
e
t
al
.
[
21]
ha
ve
us
e
d
de
e
p
le
a
r
ni
ng
a
ppr
oa
c
h
f
or
t
r
a
c
ki
ng
s
of
twa
r
e
e
r
r
or
s
pr
e
s
e
nt
in
c
lo
ud
a
ppl
ic
a
ti
on
f
o
r
de
te
c
ti
ng s
of
twa
r
e
bugs
us
in
g m
ul
ti
la
ye
r
e
d pe
r
c
e
pt
r
ons
. F
ur
th
e
r
a
dopt
io
n of
m
ul
ti
la
ye
r
e
d pe
r
c
e
pt
r
on w
a
s
a
ls
o
di
s
c
us
s
e
d
by
C
ui
e
t
al
.
[
22]
w
he
r
e
th
e
id
e
a
i
s
to
ge
ne
r
a
te
s
ui
ta
bl
e
te
s
t
c
a
s
e
s
us
in
g
hi
s
to
r
ic
a
l
da
ta
.
J
a
m
m
a
la
m
a
da
ka
a
nd
P
a
r
ve
e
n
[
23]
ha
ve
us
e
d
de
e
p
b
e
li
e
f
ne
twor
k
in
or
de
r
to
e
va
lu
a
te
c
ove
r
a
ge
of
s
of
twa
r
e
te
s
ti
ng
w
it
h
m
ul
ti
pl
e
da
ta
s
e
t
s
.
L
a
r
a
nj
e
ir
o
e
t
al
.
[
24]
ha
ve
de
v
e
lo
pe
d
a
n
in
te
ll
ig
e
nt
m
ode
l
to
w
a
r
ds
de
te
c
ti
ng
s
ub
-
opt
im
a
l
da
ta
qua
li
ty
a
s
s
oc
i
a
te
d
w
it
h
w
e
b
s
e
r
vi
c
e
s
a
nd
a
ppl
ic
a
ti
ons
.
M
a
r
ti
n
[
25]
ha
ve
pr
e
s
e
nt
e
d
a
c
om
pr
e
he
ns
iv
e
a
na
ly
s
is
of
m
ul
ti
pl
e
m
a
c
hi
ne
le
a
r
ni
ng
m
ode
ls
to
f
in
d
th
e
e
f
f
e
c
ti
ve
ne
s
s
to
w
a
r
ds
pr
e
di
c
ti
ve
s
of
twa
r
e
te
s
ti
ng.
A
dopt
io
n
of
uns
upe
r
vi
s
e
d
m
a
c
hi
ne
le
a
r
ni
ng
is
w
it
ne
s
s
e
d
in
w
or
k
pr
e
s
e
nt
e
d
by
S
e
ba
s
ti
a
n
e
t
al
.
[
26]
w
it
h a
t
a
r
ge
t
to
m
in
im
iz
e
t
he
t
e
s
t
c
a
s
e
s
t
he
r
e
by mi
ni
m
iz
in
g
th
e
c
os
t
a
nd t
im
e
i
nvol
ve
d i
n
pr
oc
e
s
s
in
g t
e
s
t
c
a
s
e
s
. M
ogha
d
a
m
e
t
al
.
[
27]
ha
ve
us
e
d bi
o
-
in
s
p
ir
e
d a
ppr
oa
c
he
s
i
n A
I
a
nd ma
c
hi
ne
l
e
a
r
ni
ng
i
n
or
de
r
to
ge
ne
r
a
te
s
ui
ta
bl
e
te
s
t
c
a
s
e
s
a
u
s
e
c
a
s
e
of
la
ne
di
s
c
ip
li
ne
in
r
oa
d
tr
a
ns
por
t.
T
he
w
or
k
c
a
r
r
ie
d
out
by
A
hm
e
d
e
t
al
.
[
28]
ha
ve
pr
e
s
e
nt
e
d
a
uni
que
s
c
he
m
e
th
a
t
c
a
n
pr
io
r
it
iz
e
te
s
t
c
a
s
e
s
w
hi
le
pe
r
f
or
m
in
g
r
e
gr
e
s
s
io
n
-
ba
s
e
d
a
s
s
e
s
s
m
e
nt
.
I
t
w
a
s
a
l
s
o
not
e
d
th
a
t
f
e
a
tu
r
e
e
xt
r
a
c
ti
on
p
la
ys
one
of
th
e
c
r
it
ic
a
l
r
ol
e
s
w
hi
le
a
ppl
yi
ng
m
a
c
hi
ne
le
a
r
ni
ng
in
s
of
twa
r
e
te
s
ti
ng.
S
tu
dy
in
s
uc
h
di
r
e
c
ti
on
w
a
s
c
a
r
r
ie
d
out
by
C
he
n
e
t
al
.
[
29]
w
he
r
e
th
e
a
ut
hor
s
c
ont
r
ib
ut
e
to
w
a
r
ds
pot
e
nt
ia
l
f
e
a
tu
r
e
s
e
le
c
ti
on
th
a
t
is
ne
c
e
s
s
a
r
y
f
or
pr
e
di
c
ti
ve
c
la
s
s
if
ic
a
ti
on.
T
he
c
or
e
r
e
s
e
a
r
c
h
ga
p
is
th
a
t
e
xi
s
ti
ng
a
ppr
oa
c
he
s
a
dopt
s
s
ophi
s
ti
c
a
t
e
d
A
I
s
c
he
m
e
w
he
r
e
th
e
s
ig
ni
f
ic
a
nt
lo
ophole
s
in
e
va
lu
a
ti
ng
s
of
twa
r
e
d
e
s
ig
n
qua
li
ty
i
s
of
te
n
unnoti
c
e
d
f
or
th
e
ir
ove
r
f
it
ti
ng.
F
ur
th
e
r
,
th
e
r
e
is
a
ls
o
a
g
a
p
to
w
a
r
ds
e
vol
vi
ng out a
ny l
ow
-
c
os
t
in
vol
ve
d a
lg
or
it
hm
ic
a
ppr
oa
c
h.
T
he
pr
im
e
c
ont
r
ib
ut
io
n
of
th
e
pr
opos
e
d
s
tu
dy
is
to
de
ve
lo
p
a
nove
l
in
te
ll
ig
e
nt
c
om
put
a
ti
ona
l
f
r
a
m
e
w
or
k
to
de
te
r
m
in
e
th
e
in
c
ons
is
te
nc
ie
s
w
it
hi
n
te
s
t
e
c
os
y
s
t
e
m
.
T
he
va
lu
e
-
a
dde
d c
ont
r
ib
ut
io
n
of
th
e
s
tu
dy
a
r
e
:
i)
to
de
s
ig
n
a
pr
e
e
m
pt
iv
e
s
of
t
-
c
om
put
in
g
m
ode
l
w
he
r
e
A
I
ha
s
be
e
n
im
pl
e
m
e
nt
e
d
to
de
te
r
m
in
e
a
dve
r
s
a
r
y
-
ba
s
e
d
out
c
om
e
s
w
it
h
f
a
ls
e
pos
it
iv
e
s
,
ii
)
to
de
ve
lo
p
a
n
in
vol
unt
a
r
y
ge
ne
r
a
ti
on
a
nd
e
xe
c
ut
io
n
of
th
e
opt
im
a
l
te
s
t
c
a
s
e
s
w
it
h
hi
ghe
r
de
gr
e
e
of
r
e
li
a
bi
l
it
y
m
a
ppi
ng
w
it
h
pr
a
c
ti
c
a
l
w
or
ld
da
ta
s
e
t,
ii
i)
th
e
pr
im
e
a
ge
nda
of
th
is
m
ode
l
is
t
o ge
ne
r
a
te
a
ne
ga
ti
ve
out
c
om
e
by p
r
oc
e
s
s
in
g va
r
ie
d t
e
s
t
c
a
s
e
s
i
n or
de
r
t
o
id
e
nt
if
y a
ll
pos
s
ib
le
f
or
m
s
of
in
c
ons
is
te
nc
ie
s
pr
e
s
e
nt
w
it
hi
n
th
e
gi
ve
n
s
of
twa
r
e
d
e
s
ig
n,
iv
)
th
e
pr
opos
e
d
A
I
-
m
ode
l
is
de
s
ig
n
e
d
c
ons
id
e
r
in
g
r
e
vi
s
e
d
ve
r
s
io
n
of
bot
h
uns
upe
r
vi
s
e
d
a
nd
s
up
e
r
vi
s
e
d
le
a
r
ni
ng
s
c
he
m
e
th
a
t
g
e
ne
r
a
te
s
a
c
om
put
e
d
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
N
ov
e
l
pr
e
e
m
pt
iv
e
i
nt
e
ll
ig
e
nt
ar
ti
fi
c
ia
l
in
te
ll
ig
e
nc
e
-
m
ode
l
fo
r
d
e
te
c
ti
ng i
nc
ons
is
te
nc
y
…
(
Sange
e
th
a G
o
v
in
da
)
1783
pr
e
di
c
ti
ve
ve
r
di
c
t
to
w
a
r
ds
tr
us
twor
th
y
te
s
tc
a
s
e
s
,
a
nd
v)
a
c
om
pr
e
he
ns
iv
e
be
nc
hm
a
r
ki
ng
is
c
a
r
r
ie
d
out
to
in
ve
s
ti
ga
te
t
he
e
r
r
or
a
nd a
lg
or
it
hm
pr
oc
e
s
s
in
g t
im
e
us
in
g unc
on
ve
nt
io
na
l
m
e
th
ods
of
l
e
a
r
ni
ng i
n A
I
-
m
ode
l.
2.
M
E
T
H
O
D
T
he
pr
im
e
a
im
of
th
e
pr
opos
e
d
s
tu
dy
is
to
de
s
ig
n
a
c
om
put
a
ti
ona
l
f
r
a
m
e
w
or
k
th
a
t
c
ont
r
ib
ut
e
s
to
w
a
r
ds
ge
ne
r
a
ti
on of
i
nt
e
ll
ig
e
nt
a
nd a
ppr
opr
ia
te
t
e
s
t
c
a
s
e
s
i
n o
r
de
r
t
o a
s
s
e
s
s
t
he
s
u
s
ta
in
a
bi
li
ty
of
t
he
s
of
twa
r
e
pr
ot
ot
ype
us
in
g
A
I
.
T
he
a
r
c
hi
te
c
tu
r
e
a
dopt
e
d
f
or
pr
opos
e
d
s
tu
dy
is
s
how
n
in
F
ig
u
r
e
1.
T
he
nove
lt
y
o
f
pr
opos
e
d
s
tu
dy
is
th
a
t
it
e
m
pha
s
iz
e
s
on
a
dv
e
r
s
e
te
s
ti
ng
out
c
om
e
a
nd
not
on
nor
m
a
l
te
s
ti
ng
out
c
om
e
in
or
de
r
to
f
ig
ur
e
out
p
r
e
s
e
nc
e
of
a
bnor
m
a
li
ti
e
s
in
a
na
ly
s
is
of
da
ta
.
T
he
c
or
e
id
e
a
of
th
is
m
ode
l
s
how
n
in
F
ig
ur
e
1
is
to
w
a
r
ds
ha
r
ne
s
s
in
g
th
e
pot
e
nt
ia
l
of
A
I
f
or
a
ut
he
nt
ic
a
ti
ng
th
e
g
e
nui
ne
ne
s
s
of
a
dve
r
s
e
te
s
t
out
c
om
e
,
de
te
r
m
in
e
pot
e
nt
ia
l
c
r
a
s
he
s
a
nd
f
a
il
ur
e
s
w
it
hi
n
th
e
a
s
s
e
s
s
m
e
nt
e
nvi
r
onm
e
nt
,
a
nd
a
na
ly
z
e
d
th
e
out
c
om
e
s
obt
a
in
e
d
f
r
om
th
e
t
e
s
ti
ng of
t
he
s
of
twa
r
e
. T
he
pr
opos
e
d s
tu
dy a
ls
o c
ont
r
ib
ut
e
s
t
ow
a
r
ds
de
ve
lo
pi
ng a
n a
ut
om
a
te
d m
e
c
ha
ni
s
m
th
a
t
is
c
a
pa
bl
e
of
s
ophi
s
ti
c
a
t
e
d c
ha
r
a
c
te
r
is
ti
c
of
s
of
twa
r
e
f
ol
lo
w
e
d by c
or
r
e
c
tl
y i
de
nt
if
yi
ng t
he
i
nc
ons
is
te
nc
ie
s
a
s
s
oc
ia
t
e
d w
it
h s
of
twa
r
e
de
s
ig
n.
T
he
pr
im
a
r
y ope
r
a
ti
on c
a
r
r
ie
d out by
t
he
pr
opos
e
d s
tu
dy mode
l
i
s
t
o pe
r
f
or
m
c
ol
le
c
ti
on
of
da
ta
f
r
o
m
th
e
a
s
s
e
s
s
m
e
nt
e
c
o
-
s
ys
te
m
w
hi
le
pe
r
f
or
m
in
g
th
e
e
v
a
lu
a
ti
on
of
s
of
twa
r
e
r
obus
tn
e
s
s
.
I
n
or
de
r
to
m
a
ke
th
e
a
s
s
e
s
s
m
e
nt
s
c
e
na
r
io
a
ppl
ic
a
bl
e
f
or
pr
a
c
ti
c
a
l
w
or
ld
,
th
e
s
tu
dy
c
ons
id
e
r
s
ha
r
dw
a
r
e
in
th
e
lo
op
e
nvi
r
onm
e
nt
f
o
r
a
s
s
e
s
s
in
g
th
e
e
xpe
r
im
e
nt
a
l
pr
ot
ot
ype
a
lo
ng
w
it
h
ha
r
dw
a
r
e
in
te
gr
a
te
d
w
it
h
s
of
twa
r
e
d
e
s
ig
n.
T
he
m
ode
l
is
a
na
ly
z
e
d
us
in
g
r
e
a
l
-
w
or
ld
s
e
ns
or
y
da
ta
of
a
m
obi
le
obj
e
c
ts
w
it
h
a
n
a
pr
io
r
i
r
a
te
o
f
s
a
m
pl
in
g.
T
he
da
ta
s
e
t
a
s
s
oc
ia
t
e
d
w
it
h
tr
a
in
in
g
ope
r
a
ti
on
c
ons
is
t
s
of
te
s
t
-
c
a
s
e
s
of
no
r
m
a
l
ty
pe
a
nd
a
bnor
m
a
l
ty
p
e
.
F
ur
th
e
r
s
e
t
of
pr
e
pr
oc
e
s
s
in
g
ope
r
a
ti
on
of
th
e
s
e
ns
or
y
da
ta
is
c
a
r
r
ie
d
out
f
ol
lo
w
e
d
by
th
e
tr
a
in
in
g
ope
r
a
ti
on
us
in
g
A
I
m
ode
l
.
T
he
ge
ne
r
a
te
d
a
n
a
ly
ti
c
a
l
m
ode
l
is
us
e
d
f
or
de
te
r
m
in
in
g
th
e
a
bnor
m
a
li
ti
e
s
lo
c
a
li
z
e
d
w
it
hi
n
th
e
c
ons
id
e
r
e
d
e
nvi
r
onm
e
nt
th
a
t
f
ur
th
e
r
ge
ne
r
a
te
s
th
e
a
dve
r
s
e
te
s
t
c
a
s
e
s
. T
he
g
e
ne
r
a
te
d
te
s
t
out
c
om
e
i
s
f
ur
th
e
r
r
e
por
te
d
ba
c
k
to
t
he
t
e
s
ti
ng pr
of
e
s
s
io
na
l
in
or
de
r
t
o de
te
r
m
in
e
t
he
r
e
li
a
bi
li
ty
of
a
dve
r
s
e
t
e
s
t
r
e
s
ul
ta
nt
.
F
ig
ur
e
1. P
r
opos
e
d a
r
c
hi
te
c
tu
r
e
2.1
.
P
r
e
p
r
oc
e
s
s
in
g op
e
r
at
io
n
T
hi
s
op
e
r
a
ti
on
is
c
a
r
r
ie
d
out
f
or
th
e
da
ta
th
a
t
is
a
ggr
e
ga
te
d
in
t
he
pr
oc
e
s
s
of
e
va
lu
a
ti
on
w
it
h
th
e
te
s
t
-
c
a
s
e
s
t
ha
t
in
vol
ve
s
a
ll
t
he
e
s
s
e
nt
ia
l
in
f
or
m
a
ti
on a
s
s
oc
ia
te
d w
it
h
t
he
t
e
s
t
e
nvi
r
onm
e
nt
. I
t
i
s
t
o be
not
e
d t
ha
t
t
e
s
t
out
c
om
e
is
ba
s
ic
a
ll
y
th
e
r
e
s
ul
ta
nt
of
th
e
e
xe
c
ut
io
n
of
in
vol
unt
a
r
y
c
a
s
e
of
te
s
ti
ng
a
nd
th
is
c
oul
d
be
e
it
he
r
ne
ga
ti
ve
or
pos
it
iv
e
,
de
pe
ndi
ng
on
it
s
n
a
tu
r
e
of
f
or
m
a
ti
on.
W
he
n
a
ne
ga
ti
ve
out
c
om
e
is
ge
ne
r
a
te
d
ow
in
g
to
th
e
a
s
s
e
r
ti
on
of
e
ve
n
one
c
ondi
ti
on,
th
e
s
c
he
m
e
c
ons
id
e
r
s
it
s
te
s
t
to
be
ne
ga
ti
ve
.
T
he
c
om
pl
e
te
in
f
or
m
a
ti
on
a
s
s
oc
ia
t
e
d
w
it
h
th
e
be
ha
vi
our
of
th
e
te
s
t
e
nvi
r
onm
e
nt
c
a
n
be
r
e
pr
e
s
e
nt
e
d
in
th
e
f
or
m
of
tr
a
c
e
lo
g
obj
e
c
ts
th
a
t
c
ons
is
ts
of
a
ll
i
nf
or
m
a
ti
on a
s
s
oc
ia
te
d w
it
h t
he
dyna
m
ic
c
ha
nge
s
of
a
n e
nvi
r
onm
e
nt
a
l
obj
e
c
ts
. T
he
c
or
e
i
de
a
of
th
is
pa
r
t
of
im
pl
e
m
e
nt
a
ti
on
of
pr
e
pr
oc
e
s
s
in
g
ope
r
a
ti
on
is
to
o
pt
f
or
e
xc
lu
s
iv
e
a
nd
uni
que
f
e
a
tu
r
e
e
xt
r
a
c
te
d
f
r
om
th
e
te
s
t
c
a
s
e
s
.
A
s
th
e
c
om
pl
e
te
out
c
om
e
s
of
th
e
t
e
s
ti
ng
c
a
n
b
e
s
to
r
e
d
in
e
a
s
il
y
a
c
c
e
s
s
ib
le
c
li
e
nt
a
ppl
ic
a
ti
on
in
te
r
f
a
c
e
,
th
e
pr
opos
e
d
s
c
he
m
e
c
a
n
now
pe
r
f
or
m
s
im
pl
e
r
a
nd
f
a
s
te
r
e
xt
r
a
c
ti
on
of
f
e
a
tu
r
e
s
f
r
om
th
is
in
te
r
f
a
c
e
.
T
he
pr
opos
e
d
s
c
h
e
m
e
c
a
te
gor
iz
e
s
th
e
da
ta
obt
a
in
e
d
f
r
om
th
e
tr
a
c
e
lo
g
obj
e
c
t
in
to
m
ul
ti
pl
e
c
la
s
s
e
s
w
hi
c
h
a
r
e
m
a
in
ly
r
e
la
te
d
to
th
e
e
nvi
r
onm
e
nt
a
l
a
tt
r
ib
ut
e
s
a
nd
f
in
a
ll
y
r
e
ta
in
s
a
ll
th
e
in
f
or
m
a
ti
on
i
n
pl
a
in
te
xt
f
il
e
s
f
or
e
a
s
ie
r
a
c
c
e
s
s
to
in
f
or
m
a
ti
on.
T
he
pr
oc
e
s
s
of
da
ta
c
le
a
ni
ng
i
s
c
a
r
r
ie
d
out
by
e
li
m
in
a
ti
ng
th
e
c
ol
um
ns
c
ons
is
ti
ng
of
unne
c
e
s
s
a
r
y
da
ta
w
hi
le
a
ny
e
r
r
or
s
a
s
s
oc
ia
te
d
w
it
h
f
or
m
a
tt
in
g
is
r
e
pa
ir
e
d.
T
he
s
c
he
m
e
th
e
n
c
a
r
r
ie
s
out
a
da
ta
p
a
r
s
in
g
f
or
in
ve
s
ti
ga
ti
ng
th
e
e
nt
r
ie
s
of
tr
a
c
e
lo
g
obj
e
c
ts
f
ol
lo
w
e
d
by
e
xt
r
a
c
ti
ng
th
e
pot
e
nt
ia
ll
y
s
ig
ni
f
ic
a
nt
in
f
or
m
a
ti
on
f
r
om
th
e
te
s
t
out
c
om
e
.
A
s
pe
c
if
ic
f
or
m
of
pa
tt
e
r
ns
on
th
e
ba
s
is
of
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
. 14, No. 3, J
une
2025
:
1781
-
1789
1784
e
nvi
r
onm
e
nt
a
l
a
tt
r
ib
ut
e
s
a
r
e
f
or
m
e
d
f
or
th
e
s
tr
uc
tu
r
e
d
da
ta
o
bt
a
in
e
d
f
r
om
t
r
a
c
e
lo
g
obj
e
c
ts
.
T
he
pr
opos
e
d
s
c
he
m
e
a
ls
o e
xt
r
a
c
t
s
t
he
i
nf
or
m
a
ti
on o
f
t
he
t
im
e
i
nvol
ve
d f
or
p
a
r
ti
c
ul
a
r
s
ta
te
s
of
a
dyna
m
ic
s
ys
te
m
dur
in
g t
he
pr
oc
e
s
s
of
f
e
a
tu
r
e
e
ngi
ne
e
r
in
g.
F
ur
th
e
r
,
la
be
l
c
odi
ng
a
ppr
oa
c
h
is
us
e
d
f
or
a
c
qui
r
in
g
num
e
r
ic
a
l
s
c
or
e
s
f
r
om
th
e
c
a
te
gor
ic
a
l
a
tt
r
ib
ut
e
s
f
ol
lo
w
e
d
by
pe
r
f
or
m
in
g
s
ta
ge
of
da
ta
t
r
a
ns
f
or
m
a
ti
on,
w
he
r
e
nor
m
a
li
z
a
ti
on
is
c
a
r
r
ie
d
out
f
or
th
e
r
e
c
e
nt
ly
a
c
qui
r
e
d
num
e
r
ic
a
l
a
tt
r
ib
ut
e
s
.
70
-
30
pr
opor
ti
on
r
ul
e
is
us
e
d
f
or
c
la
s
s
if
yi
ng
th
e
tr
a
in
in
g
a
nd
te
s
ti
ng
da
ta
.
W
hi
le
a
ppl
yi
ng
A
I
-
ba
s
e
d
a
ppr
oa
c
h,
it
i
s
e
s
s
e
nt
ia
l
to
of
f
e
r
m
or
e
im
por
ta
nc
e
to
w
a
r
ds
f
or
m
a
ti
on
of
th
e
ba
la
nc
e
d
da
ta
w
hi
le
c
a
r
r
yi
ng
out
th
e
tr
a
in
in
g
th
a
t
ha
s
a
pos
it
iv
e
e
f
f
e
c
t
on
pe
r
f
or
m
a
nc
e
.
T
he
ne
xt
pa
r
t
di
s
c
us
s
e
s
a
bout
t
he
pr
opos
e
d A
I
-
m
ode
l.
2.2
.
A
r
t
if
ic
ia
l
in
t
e
ll
ig
e
n
c
e
-
m
od
e
l
T
he
pr
op
os
e
d
s
tu
dy
t
ow
a
r
d
s
A
I
-
m
od
e
l
i
s
c
a
r
r
i
e
d
out
us
i
ng
py
th
on
o
w
in
g
to
it
s
r
obu
s
t
a
n
d
f
l
e
xi
bl
e
s
uppor
t
a
bi
li
t
y
to
w
a
r
d
s
d
a
ta
s
c
ie
n
c
e
u
s
in
g i
ts
r
ic
h
s
e
t
of
op
e
r
a
to
r
s
a
nd
li
br
a
r
ie
s
.
A
n
ove
l
f
un
c
ti
on i
s
c
on
s
tr
u
c
te
d
us
in
g
P
yt
h
on
t
ha
t
is
c
a
pa
bl
e
of
de
t
e
r
m
in
in
g
th
e
pr
e
s
e
nc
e
of
a
ny
f
or
m
of
a
bnor
m
a
li
ti
e
s
a
nd
in
c
on
s
i
s
te
n
c
ie
s
pr
e
s
e
nt
w
it
hi
n
th
e
s
of
tw
a
r
e
c
o
de
de
s
ig
n.
S
uc
h
f
or
m
of
f
un
c
ti
on c
a
n
b
e
a
l
s
o
us
e
d
f
or
bot
h
u
pc
om
i
ng
a
s
w
e
ll
a
s
c
ur
r
e
nt
s
of
twa
r
e
pr
o
je
c
t
d
e
ve
l
opm
e
nt
i
n
or
d
e
r
t
o
id
e
nt
if
y
i
s
s
ue
s
in
e
v
a
lu
a
ti
on
da
t
a
.
T
h
e
pr
opo
s
e
d
s
tu
dy
m
od
e
l
is
a
n
a
ly
z
e
d
u
s
in
g
in
te
gr
a
te
d
c
om
bi
na
ti
on
of
m
ul
ti
p
le
s
up
e
r
vi
s
e
d
a
nd
un
s
up
e
r
vi
s
e
d
m
a
c
hi
n
e
le
a
r
ni
ng
a
lg
or
it
hm
s
a
s
i
ts
A
I
-
m
ode
l
. F
ol
lo
w
in
g
a
r
e
t
he
br
ie
f
in
g
s
of
th
e
l
e
a
r
ni
ng
a
ppr
o
a
c
h
e
s
us
e
d i
n
pr
opo
s
e
d s
tu
dy:
−
S
up
e
r
vi
s
e
d
l
e
a
r
ni
ng
:
i)
S
L
1
:
t
h
e
f
ir
s
t
s
up
e
r
vi
s
e
d
m
e
th
od
i
s
a
c
om
bi
na
ti
o
n
of
tr
a
n
s
f
or
m
e
r
-
ba
s
e
d
a
tt
r
i
bu
te
to
k
e
ni
z
e
r
a
nd
r
e
vi
s
e
d
r
e
s
i
du
a
l
n
e
t
w
or
k
[
30]
,
ii
)
S
L
2
:
th
e
ne
xt
t
e
c
hni
qu
e
i
s
a
c
om
b
in
a
ti
on
of
m
ul
t
il
a
y
e
r
pe
r
c
e
pt
r
on
w
it
h
s
upp
or
t
v
e
c
to
r
m
a
c
h
in
e
th
a
t
i
s
m
e
a
nt
f
or
d
e
t
e
r
m
in
i
ng
t
he
de
f
e
c
t
s
[
3
1]
,
i
ii
)
S
L
3
:
t
he
th
ir
d
le
a
r
ni
ng
s
c
he
m
e
i
s
ba
s
e
d
on
b
un
dl
in
g
of
f
e
a
tu
r
e
s
in
t
e
g
r
a
t
e
d
w
it
h
s
a
m
pl
in
g
u
s
i
ng
gr
a
d
ie
nt
m
e
th
o
d
s
[
32]
.
T
h
e
s
a
m
pl
in
g
i
s
c
a
r
r
ie
d
ou
t
by
e
li
m
in
a
ti
ng
a
ll
i
n
s
t
a
n
c
e
s
c
h
a
r
a
c
t
e
r
i
z
e
d
by
m
in
i
s
c
a
l
e
v
a
l
ue
s
of
gr
a
di
e
nt
w
hi
l
e
th
e
r
e
m
n
a
n
t
in
s
ta
nc
e
s
a
r
e
ut
i
li
z
e
d
t
o
e
va
lu
a
t
e
t
he
ga
in
i
n
in
f
or
m
a
ti
on
.
T
h
e
m
e
c
h
a
ni
s
m
of
f
e
a
tu
r
e
bun
dl
i
ng
c
on
tr
ib
ut
e
s
to
w
a
r
d
s
f
e
a
t
ur
e
r
e
du
c
ti
on
,
a
nd
i
v)
S
L
4
:
th
e
f
our
t
h
s
up
e
r
v
i
s
e
d
a
ppr
o
a
c
h
i
m
pl
e
m
e
n
te
d
in
pr
op
o
s
e
d
s
c
h
e
m
e
i
s
a
un
iq
ue
m
e
th
od
of
e
x
e
c
ut
in
g
gr
a
d
ie
nt
bo
os
ti
n
g
[
33]
w
he
r
e
pe
r
m
ut
a
t
io
n
-
b
a
s
e
d
s
c
h
e
m
e
i
s
u
s
e
d
d
e
pl
o
yi
n
g
s
e
qu
e
n
ti
a
l
bo
os
ti
n
g.
T
hi
s
s
c
h
e
m
e
c
a
n
a
na
l
yz
e
f
e
a
tu
r
e
s
w
i
th
c
a
t
e
gor
ic
a
l
v
a
lu
e
s
.
−
U
n
s
u
pe
r
vi
s
e
d
le
a
r
ni
n
g
:
i)
U
S
L
1
:
th
e
a
s
s
e
s
s
m
e
nt
i
s
c
a
r
r
i
e
d
o
ut
c
o
ns
id
e
r
in
g
r
a
ndo
m
f
or
e
s
t
in
or
de
r
t
o
s
e
l
e
c
t
opt
i
m
a
l
f
e
a
t
ur
e
r
a
n
dom
ly
f
o
r
de
te
r
m
in
a
t
io
n
of
in
c
o
n
s
i
s
t
e
n
c
i
e
s
in
o
ut
c
om
e
.
T
he
a
p
pr
o
a
c
h
p
e
r
f
or
m
s
a
ll
oc
a
ti
on
of
a
bn
or
m
a
li
t
y
va
lu
e
of
ob
s
e
r
v
a
ti
o
n
ba
s
e
d
o
n
t
e
m
por
a
l
a
tt
r
i
but
e
f
or
f
a
s
t
e
r
op
e
r
a
ti
on
ov
e
r
l
a
r
g
e
r
a
nd
h
e
te
r
og
e
n
e
o
u
s
d
a
t
a
s
e
t
[
34
]
, i
i)
U
S
L
2
:
ge
ne
r
a
t
iv
e
a
dv
e
r
s
a
r
ia
l
ne
t
w
or
k h
a
s
b
e
e
n i
m
pl
e
m
e
nt
e
d
f
or
s
i
m
il
a
r
r
e
a
s
on
t
ow
a
r
d
s
r
e
du
c
i
ng
t
he
r
e
c
on
s
t
r
u
c
ti
on
e
r
r
or
[
3
5]
,
i
ii
)
U
S
L
3
:
a
ut
o
e
n
c
od
e
r
a
lg
or
it
hm
h
a
s
b
e
e
n
f
ur
t
he
r
us
e
d
f
or
m
i
ni
m
iz
in
g
t
he
r
e
c
o
n
s
tr
u
c
t
io
n
e
r
r
or
th
e
r
e
b
y
f
a
c
il
it
a
ti
n
g
g
e
ne
r
a
ti
on
of
e
xt
e
n
s
i
ve
lo
gi
c
a
l
r
e
pr
e
s
e
nt
a
ti
on
of
a
n
out
c
o
m
e
[
36]
,
i
v)
U
S
L
4
:
e
n
s
e
m
bl
e
ba
s
e
d
k
-
n
e
a
r
e
s
t
ne
ig
hbo
ur
a
lg
or
it
hm
ha
s
be
e
n
im
pl
e
m
e
n
te
d
f
or
a
c
c
o
m
pl
is
hi
n
g
f
a
s
te
r
de
te
c
ti
on of
in
c
o
n
s
i
s
t
e
n
c
i
e
s
i
n a
dv
e
r
s
e
ou
tc
om
e
s
[
37]
,
v)
U
S
L
5
:
th
e
pr
op
o
s
e
d
s
ys
te
m
a
l
s
o
u
s
e
s
r
e
v
is
e
d
k
-
n
e
a
r
e
s
t
ne
ig
h
bo
ur
a
l
go
r
it
hm
w
h
e
r
e
H
il
be
r
t
c
ur
v
e
i
s
u
s
e
d
f
or
c
om
put
in
g
th
e
w
e
i
ght
ta
r
ge
ti
n
g
to
w
a
r
d
s
b
e
tt
e
r
c
o
nv
e
r
g
e
n
c
e
r
a
t
e
a
nd
m
in
im
i
z
e
d
ti
m
e
c
o
m
pl
e
x
it
y
[
38]
,
vi
)
U
S
L
6
:
t
he
ne
xt
a
p
pr
o
a
c
h
u
s
e
d
is
b
a
s
e
d
on
di
m
e
ns
io
n
a
l
r
e
du
c
ti
o
n
in
or
d
e
r
t
o
m
i
ni
m
iz
e
t
he
c
om
p
ut
a
ti
on
a
l
bur
d
e
n
a
s
w
e
l
l
a
s
e
x
pl
o
r
e
th
e
l
a
t
e
nt
p
a
tt
e
r
ns
i
n da
ta
di
s
tr
ib
ut
i
on
[
39]
,
a
nd
v
ii
)
U
S
L
7
:
t
he
f
i
na
l
u
n
s
u
pe
r
vi
s
e
d
a
pp
r
o
a
c
h
us
e
d
i
n
pr
o
po
s
e
d
s
c
h
e
m
e
c
o
nt
r
i
bu
te
s
to
w
a
r
d
s
pr
e
s
e
n
ti
ng
a
no
n
-
p
a
r
a
m
e
tr
i
c
m
e
th
o
d
of
e
v
a
l
ua
ti
n
g
th
e
po
s
s
i
bl
e
in
put
da
ta
di
s
tr
ib
u
ti
o
n
[
4
0]
.
T
he
e
n
d
pr
ob
a
bi
li
t
y
a
s
s
oc
ia
te
d
w
it
h
a
ll
th
e
d
a
t
a
p
oi
n
t
s
a
r
e
e
v
a
lu
a
t
e
d
w
it
h
t
h
is
d
is
tr
ib
ut
i
on
.
T
h
e
ne
xt
s
e
c
ti
on
pr
e
s
e
n
t
s
d
i
s
c
us
s
io
n
of
r
e
s
u
lt
be
in
g
a
c
c
om
pl
i
s
he
d.
3.
R
E
S
U
L
T
S
T
he
s
c
r
ip
ti
ng
of
th
e
pr
opos
e
d
s
tu
dy
i
s
c
a
r
r
ie
d
out
in
pyt
ho
n
e
nvi
r
onm
e
nt
c
ons
id
e
r
in
g
publ
ic
ly
a
va
il
a
bl
e
ope
n
-
a
c
c
e
s
s
da
ta
s
e
t
[
41]
w
it
h a
t
a
r
ge
t
to
i
nve
s
ti
ga
te
t
he
e
f
f
e
c
ti
ve
ne
s
s
of
pr
opos
e
d A
I
m
ode
l
in
or
de
r
to
de
te
c
t
th
e
a
dve
r
s
e
out
c
om
e
.
T
h
e
da
ta
s
e
t
c
ons
is
t
s
of
in
f
or
m
a
ti
on
r
e
la
te
d
to
pr
opor
ti
on
of
in
c
ons
is
te
nc
ie
s
,
f
e
a
tu
r
e
s
,
a
nd
s
iz
e
w
hi
le
th
e
c
om
pl
e
te
da
ta
s
e
t
is
la
be
ll
e
d
w
hi
c
h
m
a
ke
s
th
e
ta
s
k
e
a
s
ie
r
f
or
a
na
ly
z
in
g
bot
h
uns
upe
r
vi
s
e
d
a
nd
s
upe
r
vi
s
e
d
le
a
r
ni
ng
a
ppr
oa
c
he
s
.
T
he
c
om
pl
e
te
e
va
lu
a
ti
on
of
th
e
out
c
om
e
is
c
a
r
r
ie
d
ou
t
c
ons
id
e
r
in
g
two
s
ta
nda
r
d
pe
r
f
or
m
a
nc
e
m
e
tr
ic
of
m
e
a
n
s
qua
r
e
d
e
r
r
or
(
M
S
E
)
a
nd
p
r
oc
e
s
s
in
g
ti
m
e
of
a
n
a
lg
or
it
hm
.
F
or
a
n
e
f
f
e
c
ti
ve
a
na
ly
s
is
,
th
e
s
tu
dy
c
on
s
id
e
r
in
g
e
va
lu
a
ti
on
pe
r
f
or
m
a
nc
e
f
or
M
S
E
in
to
two
f
ol
ds
vi
z
:
i)
e
va
lu
a
ti
on
of
M
S
E
f
or
tr
a
in
im
a
ge
s
(
M
S
E
_1)
a
nd
ii
)
e
va
lu
a
ti
on
of
M
S
E
f
or
te
s
t
im
a
ge
s
(
M
S
E
_2)
.
T
he
be
nc
hm
a
r
ke
d nume
r
ic
a
l
out
c
om
e
i
s
t
a
bul
a
t
e
d i
n T
a
bl
e
s
1
a
nd
2.
D
ur
in
g
th
e
pr
oc
e
s
s
of
in
ve
s
ti
ga
ti
ng
th
e
e
f
f
e
c
ti
ve
ne
s
s
of
va
r
i
e
d
f
or
m
of
A
I
-
m
ode
ls
pe
r
ta
in
in
g
to
de
te
r
m
in
a
ti
on
of
in
c
ons
is
te
nc
ie
s
on
bot
h
tr
a
in
in
g
da
t
a
s
e
t
a
s
w
e
ll
a
s
t
e
s
ti
ng
da
ta
s
e
t,
th
e
s
tu
dy
out
c
om
e
e
xhi
bi
te
d
in
a
bove
ta
bul
a
t
e
d
va
lu
e
s
s
how
s
s
om
e
in
te
r
e
s
ti
ng
ou
tc
om
e
s
ove
r
M
S
E
a
nd
pr
oc
e
s
s
in
g
ti
m
e
s
c
or
e
.
F
r
om
T
a
bl
e
1,
th
e
di
ve
r
s
if
ie
d
out
c
om
e
s
how
s
th
e
e
f
f
e
c
ti
ve
ne
s
s
of
s
om
e
of
th
e
pa
r
ti
c
ul
a
r
a
lg
or
it
hm
s
in
A
I
-
m
ode
l
on
th
e
ba
s
i
s
of
th
e
da
ta
be
in
g
u
s
e
d
a
s
a
n
in
put
.
I
t
w
a
s
f
ound
th
a
t
U
S
L
4
us
in
g
r
e
vi
s
e
d
ne
a
r
e
s
t
ne
ig
hbor
in
g i
s
f
ound with m
os
t
e
f
f
e
c
ti
ve
out
c
om
e
f
r
om
t
he
pe
r
s
pe
c
ti
ve
of
hi
ghl
y r
e
duc
e
d M
S
E
s
c
or
e
f
or
bot
h
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
N
ov
e
l
pr
e
e
m
pt
iv
e
i
nt
e
ll
ig
e
nt
ar
ti
fi
c
ia
l
in
te
ll
ig
e
nc
e
-
m
ode
l
fo
r
d
e
te
c
ti
ng i
nc
ons
is
te
nc
y
…
(
Sange
e
th
a G
o
v
in
da
)
1785
tr
a
in
in
g
(
M
S
E
=
0.02)
a
nd
te
s
ti
ng
da
ta
(
M
S
E
=
0.11)
.
N
e
a
r
ly
s
i
m
il
a
r
pe
r
f
or
m
a
nc
e
is
a
ls
o
obs
e
r
ve
d
f
or
U
S
L
3
us
in
g
a
ut
oe
nc
ode
r
(
M
S
E
_1
=
0.02
a
nd
M
S
E
_2
=
0.17)
,
U
S
L
7
us
in
g
non
-
pa
r
a
m
e
tr
ic
a
ppr
oa
c
h
to
w
a
r
ds
e
s
ti
m
a
ti
ng
in
c
ons
is
te
nc
ie
s
(
M
S
E
_1
=
002
a
nd
M
S
E
_2
=
0.16)
,
U
S
L
6
us
in
g
di
m
e
ns
io
na
l
r
e
duc
ti
on
a
ppr
oa
c
h
(
M
S
E
_1
=
0.024
a
nd
M
S
E
_2
=
0.16)
.
F
r
om
th
e
p
e
r
s
p
e
c
ti
v
e
of
s
u
pe
r
vi
s
e
d
le
a
r
n
in
g
-
b
a
s
e
d
A
I
-
m
o
de
ls
e
xh
ib
i
te
d
i
n
T
a
bl
e
2,
it
is
no
te
d
t
ha
t
opt
i
m
a
l
p
e
r
f
or
m
a
nc
e
i
s
s
h
ow
n
b
y
S
L
1
a
p
pr
o
a
c
h
t
ha
t
d
e
pl
oy
s
tr
a
ns
f
or
m
e
r
-
b
a
s
e
d a
ppr
oa
c
h
w
i
th
r
e
s
id
ua
l
n
e
t
w
or
k
(
M
S
E
_
1
=
0 a
n
d M
S
E
_
2
=
0.
03)
. T
h
e
s
e
c
ond
-
b
e
s
t
p
e
r
f
or
m
a
n
c
e
i
s
e
xh
ib
i
te
d
b
y
S
L
3
(
M
S
E
_1
=
0
a
nd M
S
E
_2
=
0
.05
)
a
nd
S
L
4
(
M
S
E
_
1
=
0
a
nd
M
S
E
_2
=
0.
05)
u
s
i
ng
f
e
a
t
ur
e
bun
dl
i
ng
w
it
h
gr
a
di
e
nt
s
a
m
pl
in
g
a
n
d
gr
a
di
e
n
t
boo
s
ti
n
g
r
e
s
p
e
c
ti
v
e
l
y.
H
o
w
e
ve
r
,
a
di
f
f
e
r
e
nt
f
or
m
of
p
e
r
f
o
r
m
a
n
c
e
tr
e
nd
i
s
e
xhi
bi
t
e
d
b
y
S
L
2
a
pp
r
o
a
c
h
t
ha
t
u
s
e
s
m
ul
t
i
-
l
a
y
e
r
e
d
p
e
r
c
e
pt
r
o
n
w
it
h
s
up
por
t
v
e
c
t
or
m
a
c
hi
ne
.
I
t
s
uni
qu
e
n
e
s
s
c
a
n
be
ju
s
t
if
i
e
d
f
r
om
r
e
d
uc
e
d
M
S
E
_
1
=
0.
00
1
a
nd
M
S
E
_
2
=
0.
01
d
ur
i
ng
t
he
te
s
t
in
g
w
h
ic
h
i
s
r
e
du
c
e
d
v
a
l
ue
in
c
on
tr
a
s
t
to
S
L
1
,
S
L
3
,
a
nd
S
L
4
a
pp
r
o
a
c
he
s
i
n
e
va
lu
a
ti
on.
T
a
bl
e
1. N
um
e
r
ic
a
l
out
c
om
e
f
or
uns
upe
r
vi
s
e
d l
e
a
r
ni
ng me
th
od
U
ns
upe
r
vi
s
e
d
l
e
a
r
ni
ng
M
S
E
_1
M
S
E
_2
P
r
oc
e
s
s
i
ng
time
U
S
L
₁
0.02
0.18
0.307
U
S
L
₂
0.05
0.13
0.399
U
S
L
₃
0.02
0.17
0.128
U
S
L
₄
0.02
0.11
0.256
U
S
L
₅
0.12
0.82
0.271
U
S
L
₆
0.024
0.16
0.602
U
S
L
₇
0.02
0.16
0.472
T
a
bl
e
2. N
um
e
r
ic
a
l
out
c
om
e
f
or
s
upe
r
vi
s
e
d l
e
a
r
ni
ng
m
e
th
od
S
upe
r
vi
s
e
d
l
e
a
r
ni
ng
M
S
E
_1
M
S
E
_2
P
r
oc
e
s
s
i
ng
time
S
L
₁
0
0.03
0.277
S
L
₂
0.001
0.01
0.493
S
L
₃
0
0.05
0.298
S
L
₄
0
0.05
0.257
A
c
lo
s
e
r
lo
ok
in
to
th
e
ke
y
f
in
di
ngs
of
out
c
om
e
s
e
ve
nt
ua
ll
y
s
h
ow
s
th
a
t
th
e
r
e
a
r
e
c
e
r
ta
in
A
I
-
m
ode
ls
th
a
t
e
xhi
bi
ts
m
uc
h
be
tt
e
r
pe
r
f
or
m
a
nc
e
w
hi
le
s
ubj
e
c
te
d
to
tr
a
in
in
g
da
ta
s
e
t
but
w
a
s
w
it
ne
s
s
e
d
w
it
h
ove
r
-
f
it
ti
n
g
pr
obl
e
m
s
t
oo w
he
n a
na
ly
z
e
d w
it
h t
e
s
ti
ng da
ta
(
e
.g., S
L
3
a
nd
S
L
4
)
. T
he
r
e
a
s
on f
or
pe
r
f
or
m
a
nc
e
de
gr
a
da
ti
on i
n
c
e
r
ta
in
A
I
-
m
ode
ls
is
due
to
th
e
c
ha
ll
e
ng
e
s
e
nc
ount
e
r
e
d
by
th
e
s
e
m
ode
ls
dur
in
g
th
e
pr
oc
e
s
s
of
ge
ne
r
a
li
z
in
g
unt
r
a
in
e
d
da
ta
.
T
h
e
e
va
lu
a
te
d
s
c
or
e
of
M
S
E
a
l
s
o
in
te
r
pr
e
ts
th
e
pos
s
ib
le
a
c
c
ur
a
c
y
in
a
bnor
m
a
li
ty
de
te
c
ti
on
w
hi
le
pe
r
f
or
m
in
g
pr
e
di
c
ti
ve
a
na
ly
s
i
s
.
F
ur
th
e
r
,
a
c
lo
s
e
r
lo
ok
in
to
S
L
2
a
lg
or
it
hm
s
how
s
th
a
t
it
of
f
e
r
s
be
tt
e
r
c
ons
is
te
nt
pe
r
f
or
m
a
nc
e
of
r
e
duc
e
d
e
r
r
or
r
a
te
in
bot
h
tr
a
in
in
g
da
ta
s
e
t
a
nd
t
e
s
ti
ng
da
ta
s
e
t
th
a
t
is
a
di
r
e
c
ti
on
r
e
pr
e
s
e
nt
a
ti
on
of
r
obus
tn
e
s
s
a
s
s
oc
ia
te
d
w
it
h
th
is
pa
r
ti
c
ul
a
r
A
I
-
m
ode
l.
T
he
tr
e
nd
of
c
ons
is
te
nt
lo
w
e
r
M
S
E
s
c
or
e
s
i
s
a
ls
o
a
di
r
e
c
t
in
di
c
a
ti
on
of
hi
ghe
r
a
c
c
ur
a
c
y
w
hi
le
pe
r
f
or
m
in
g
pr
e
di
c
ti
ve
ope
r
a
ti
on
in
pr
opos
e
d
A
I
-
m
ode
l.
A
not
he
r
in
tr
in
s
ic
ob
s
e
r
va
ti
on
is
th
a
t
S
L
2
m
ode
l
a
c
tu
a
ll
y
w
or
ks
e
x
c
e
pt
io
na
ll
y
r
e
li
a
bl
y
w
e
ll
w
he
n
us
e
d
w
it
h
onl
y
s
uppor
t
ve
c
to
r
m
a
c
hi
ne
;
a
s
in
c
lu
s
io
n
of
m
ul
ti
-
la
ye
r
e
d
pe
r
c
e
pt
r
on
w
a
s
not
e
d
w
it
h
r
e
duc
e
d
M
S
E
pe
r
f
or
m
a
nc
e
f
or
te
s
ti
ng
da
ta
a
nd
hi
gh
e
r
M
S
E
s
c
or
e
in
tr
a
in
i
ng
da
ta
.
T
hi
s
is
a
ls
o
a
n
in
di
c
a
ti
on
of
hi
ghe
r
s
e
ns
it
iv
it
y
of
m
ul
ti
-
la
ye
r
e
d
pe
r
c
e
pt
r
on
a
ppr
oa
c
h
a
s
a
s
ta
nda
lo
ne
to
w
a
r
ds
tr
a
in
in
g
da
ta
.
H
ow
e
ve
r
,
it
is
s
ti
ll
r
e
c
om
m
e
nde
d t
o i
nt
e
gr
a
te
a
nd us
e
bot
h m
ul
ti
la
ye
r
e
d pe
r
c
e
pt
r
on a
nd s
uppor
t
ve
c
to
r
m
a
c
hi
ne
t
oge
th
e
r
t
o r
e
a
c
h
th
e
opt
im
a
l
s
ta
te
of
out
c
om
e
i
n pr
opos
e
d A
I
-
m
ode
l.
F
ig
ur
e
s
2
a
nd
3
s
how
c
a
s
e
s
th
e
s
t
a
nda
lo
ne
o
ut
c
om
e
of
bot
h
u
ns
up
e
r
vi
s
e
d
a
nd
s
upe
r
vi
s
e
d
le
a
r
ni
ng
m
ode
l
w
it
h
r
e
s
pe
c
t
to
M
S
E
r
e
s
p
e
c
ti
ve
ly
.
A
s
t
he
num
be
r
of
l
e
a
r
ni
ng
a
p
pr
oa
c
he
s
u
s
e
d
in
un
s
up
e
r
vi
s
e
d
a
r
e
s
li
ght
l
y
m
or
e
in
c
ont
r
a
s
t
to
num
b
e
r
of
s
up
e
r
vi
s
e
d
a
ppr
oa
c
he
s
,
he
nc
e
,
t
he
pr
opo
s
e
d
s
c
he
m
e
e
xt
r
a
c
t
s
th
e
m
e
a
n
va
lu
e
of
M
S
E
_1
,
M
S
E
_2,
a
nd
pr
oc
e
s
s
in
g
ti
m
e
f
r
om
T
a
bl
e
s
1
a
nd
2
f
or
b
e
tt
e
r
in
f
e
r
e
nc
e
.
I
t
s
how
s
th
a
t
m
e
a
n
va
lu
e
of
M
S
E
_
1
f
or
un
s
up
e
r
vi
s
e
d
l
e
a
r
ni
ng
a
ppr
oa
c
he
s
i
s
0.
03
914
w
hi
l
e
m
e
a
n
va
lu
e
of
M
S
E
_2
i
s
0.
24714
,
w
hi
le
t
he
pr
o
c
e
s
s
in
g
ti
m
e
m
e
a
n
s
c
or
e
i
s
f
ound
to
be
0.34
785
s
.
S
im
il
a
r
ob
s
e
r
v
a
ti
on
i
s
c
a
r
r
ie
d
out
f
or
m
e
a
n
va
lu
e
M
S
E
_1
f
or
s
upe
r
vi
s
e
d
le
a
r
ni
ng
a
p
pr
oa
c
h
t
ha
t
i
s
f
ou
nd
t
o
be
0
.0002
5
w
hi
le
m
e
a
n
va
l
ue
of
M
S
E
_2
f
or
s
upe
r
vi
s
e
d
s
c
h
e
m
e
is
f
o
und
t
o
be
0.035
.
T
he
m
e
a
n
va
l
ue
of
pr
o
c
e
s
s
in
g
ti
m
e
f
or
s
upe
r
vi
s
e
d
le
a
r
ni
ng
a
ppr
oa
c
h
is
f
ound
to
be
0.331
25
s
.
T
he
ou
tc
om
e
to
w
a
r
d
s
c
om
p
a
r
i
s
io
n
s
how
c
a
s
e
s
t
ha
t
s
up
e
r
vi
s
e
d
l
e
a
r
ni
n
g
a
p
pr
oa
c
he
s
of
f
e
r
s
b
e
tt
e
r
pe
r
f
or
m
a
nc
e
in
c
ont
r
a
s
t
to
u
ns
u
pe
r
vi
s
e
d
le
a
r
ni
n
g
a
ppr
o
a
c
h
e
s
.
T
he
m
e
a
n
s
c
or
e
of
s
up
e
r
vi
s
e
d
le
a
r
ni
n
g
a
p
pr
oa
c
he
s
f
or
tr
a
in
in
g
(
M
S
E
_1)
i
s
f
oun
d
to
be
a
ppr
ox
im
a
t
e
ly
3
8%
r
e
du
c
e
d
c
om
pa
r
e
d
to
uns
up
e
r
vi
s
e
d
le
a
r
ni
ng
. T
he
k
e
y
f
in
di
ng
s
s
how
s
th
a
t
th
e
m
e
a
n
s
c
or
e
of
s
u
pe
r
vi
s
e
d
le
a
r
ni
ng
a
ppr
oa
c
h
f
or
te
s
ti
n
g
(
M
S
E
_2)
i
s
f
ou
nd
t
o
b
e
a
ppr
oxi
m
a
te
l
y
21%
r
e
d
uc
e
d
c
om
pa
r
e
d
to
t
ha
t
of
s
up
e
r
vi
s
e
d
le
a
r
ni
ng
a
ppr
o
a
c
h.
T
h
e
pr
oc
e
s
s
in
g
ti
m
e
f
or
s
up
e
r
vi
s
e
d
l
e
a
r
ni
n
g
is
s
li
gh
tl
y
in
c
r
e
a
s
e
d
c
o
m
pa
r
e
d
to
uns
upe
r
vi
s
e
d
le
a
r
ni
ng
a
ppr
oa
c
h
by
1.6%
, w
hi
c
h i
s
a
b
s
ol
ut
e
ly
l
e
s
s
s
ig
ni
f
i
c
a
nt
a
nd
doe
s
n’
t
ha
v
e
a
ny
pot
e
nt
i
a
l
im
p
a
c
t
on
c
om
put
a
ti
on
a
l
e
f
f
or
t
u
s
e
d.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
. 14, No. 3, J
une
2025
:
1781
-
1789
1786
F
ig
ur
e
2. B
e
nc
hm
a
r
ke
d outc
om
e
of
M
S
E
f
or
uns
upe
r
vi
s
e
d l
e
a
r
ni
ng
F
ig
ur
e
3. B
e
nc
hm
a
r
ke
d outc
om
e
of
M
S
E
f
or
s
upe
r
vi
s
e
d l
e
a
r
ni
ng
F
ig
ur
e
s
4
a
nd
5
s
how
c
a
s
e
s
th
e
be
nc
hm
a
r
ke
d
out
c
om
e
of
pr
oc
e
s
s
in
g
ti
m
e
f
or
uns
upe
r
vi
s
e
d
a
nd
s
upe
r
vi
s
e
d
le
a
r
ni
ng
a
ppr
oa
c
he
s
to
f
in
d
th
a
t
th
e
r
e
is
no
s
ig
ni
f
ic
a
nt
di
f
f
e
r
e
nc
e
be
twe
e
n
th
e
m
e
a
n
s
c
or
e
of
pr
oc
e
s
s
in
g
ti
m
e
be
twe
e
n
bot
h
th
e
A
I
a
ppr
oa
c
he
s
.
H
ow
e
ve
r
,
f
r
om
th
e
pe
r
s
pe
c
ti
ve
of
gr
a
nul
a
r
s
tu
dy
to
w
a
r
ds
in
di
vi
dua
l
out
c
om
e
s
of
e
a
c
h
A
I
m
ode
ls
,
it
is
not
e
d
th
a
t
uns
upe
r
vi
s
e
d
a
ppr
oa
c
h
U
S
L
3
us
in
g
a
ut
oe
nc
ode
r
of
f
e
r
s
th
e
m
os
t
r
e
duc
e
d
pr
oc
e
s
s
in
g
ti
m
e
(
t
=
0.128
s
)
.
T
h
e
pe
r
f
or
m
a
nc
e
of
pr
oc
e
s
s
in
g
ti
m
e
f
or
s
upe
r
vi
s
e
d
a
ppr
oa
c
h
S
L
4
(
t
=
0.257
s
)
a
nd
un
s
upe
r
vi
s
e
d
a
ppr
oa
c
h
U
S
L
4
us
in
g
r
e
vi
s
e
d
e
n
s
e
m
bl
e
d
n
e
a
r
e
s
t
n
e
ig
hbor
in
g
a
ppr
oa
c
h
(
t
=
0.256
s
)
a
r
e
ne
a
r
ly
s
im
il
a
r
.
S
uc
h
ne
a
r
s
im
il
a
r
p
r
oc
e
s
s
in
g
ti
m
e
is
a
ls
o
w
it
ne
s
s
e
d
by
s
up
e
r
vi
s
e
d
a
ppr
oa
c
h
of
S
L
1
th
a
t
us
e
s
r
e
s
id
ua
l
ne
twor
k
w
it
h
tr
a
ns
f
or
m
e
r
s
c
he
m
e
(
t
=
0.277
s
)
a
nd
uns
upe
r
vi
s
e
d
a
ppr
oa
c
h
of
U
S
L
5
th
a
t
us
e
s
r
e
vi
s
e
d
k
-
ne
a
r
e
s
t
ne
ig
hbor
in
g
a
ppr
oa
c
h
(
t
=
0.271
s
)
.
F
in
a
ll
y,
i
t
is
not
e
d
th
a
t
hi
ghe
r
c
ons
um
pt
io
n of
a
lg
or
it
hm
ic
p
r
oc
e
s
s
in
g t
im
e
i
s
e
xhi
bi
te
d by one
s
upe
r
vi
s
e
d s
c
he
m
e
of
S
L
2
us
in
g m
ul
ti
la
ye
r
e
d
pe
r
c
e
pt
r
on
w
it
h
s
uppor
t
ve
c
to
r
m
a
c
hi
n
e
(
t
=
0.493
s
)
a
nd
two
uns
upe
r
vi
s
e
d
s
c
he
m
e
s
of
U
S
L
6
us
in
g
di
m
e
ns
io
na
l
r
e
duc
ti
on (
t
=
0.602
s
)
a
nd U
S
L
7
us
in
g non
-
pa
r
a
m
e
tr
ic
pr
oba
bi
li
ty
e
s
ti
m
a
ti
on me
th
od (
t
=
0.472
s
)
.
F
ig
ur
e
4. B
e
nc
hm
a
r
ke
d outc
om
e
of
pr
oc
e
s
s
in
g t
im
e
f
or
uns
upe
r
vi
s
e
d l
e
a
r
ni
ng
F
ig
ur
e
5. B
e
nc
hm
a
r
ke
d outc
om
e
of
pr
oc
e
s
s
in
g t
im
e
f
or
s
upe
r
vi
s
e
d l
e
a
r
ni
ng
4.
C
O
N
C
L
U
S
I
O
N
T
he
pr
im
e
a
ge
n
da
of
th
i
s
m
a
nu
s
c
r
i
pt
i
s
t
o
pr
e
s
e
nt
a
nove
l
a
nd
i
n
te
ll
ig
e
nt
c
om
put
a
ti
on
a
l
a
ppr
o
a
c
h
th
a
t
c
a
n
h
a
r
ne
s
s
th
e
pot
e
nt
ia
l
of
A
I
in
th
e
pr
oc
e
s
s
of
v
a
li
d
a
ti
ng
th
e
out
c
om
e
dur
in
g
s
of
tw
a
r
e
de
s
ig
n
t
e
s
t
in
g.
A
t
pr
e
s
e
nt
,
s
uc
h f
or
m
s
of
de
c
is
io
n m
a
ki
ng
of
de
t
e
r
m
in
in
g
t
he
a
dv
e
r
s
a
r
y o
ut
c
om
e
s
i
s
done
by e
xpe
r
t
s
y
s
te
m
, w
hi
l
e
th
e
pr
opo
s
e
d
s
c
h
e
m
e
h
ypot
h
e
s
i
z
e
s
th
a
t
s
u
c
h
de
c
is
i
on
m
a
ki
ng
c
a
n
be
a
u
to
nom
o
us
l
y
done
b
y
A
I
-
m
ode
l
s
.
T
he
pr
opos
e
d
s
t
udy
ha
s
in
v
e
s
ti
ga
ti
on
bot
h
un
s
u
pe
r
vi
s
e
d
a
nd
s
upe
r
vi
s
e
d
A
I
-
m
ode
l
in
or
de
r
to
a
c
c
om
pl
is
h
m
uc
h
gr
a
nul
a
r
i
ty
i
n i
t
s
out
c
om
e
. U
nl
ik
e
t
he
m
yt
h bor
n
e
d by i
ndu
s
tr
ia
l
A
I
-
ba
s
e
d a
ppl
ic
a
ti
on t
e
s
ti
ng t
h
a
t
un
s
up
e
r
vi
s
e
d
is
a
be
tt
e
r
c
hoi
c
e
i
s
pr
ove
n
ot
h
e
r
w
i
s
e
in
pr
opo
s
e
d
in
ve
s
ti
g
a
ti
on
. T
he
s
ub
-
opt
im
a
l
p
e
r
f
or
m
a
n
c
e
of
un
s
up
e
r
vi
s
e
d
le
a
r
ni
n
g
a
ppr
o
a
c
h
i
s
w
it
n
e
s
s
e
d
in
pr
opo
s
e
d
i
nve
s
ti
g
a
ti
on
th
a
t
c
a
n
be
ju
s
ti
f
i
e
d
on
th
e
gr
ound
of
e
x
te
n
s
iv
e
pr
oc
e
s
s
in
g
ti
m
e
a
nd
hi
gh
e
r
e
r
r
or
r
a
te
s
.
T
he
pr
opo
s
e
d
s
t
udy
c
on
tr
ib
ut
e
s
to
w
a
r
d
s
in
c
or
por
a
ti
ng
f
ol
lo
w
i
ng
n
ove
l
c
ha
r
e
c
te
r
i
s
ti
c
s
a
s
f
ol
lo
w
s
:
i)
a
no
ve
l
e
va
l
ua
ti
o
n
pl
a
tf
or
m
i
s
pr
e
s
e
nt
e
d
th
a
t
is
c
a
pa
bl
e
of
de
t
e
r
m
in
in
g
th
e
f
a
ls
e
pos
it
iv
e
s
a
nd
in
c
ons
i
s
t
e
nc
i
e
s
w
hi
le
pe
r
f
or
m
i
ng
t
e
s
ti
ng
of
s
of
t
w
a
r
e
de
s
ig
n,
ii
)
th
e
pr
op
os
e
d
s
c
he
m
e
pr
e
s
e
nt
s
a
s
im
pl
if
i
e
d
m
e
c
ha
ni
s
m
of
ge
n
e
r
a
ti
ng
te
s
t
c
a
s
e
s
c
o
ns
i
de
r
in
g
e
xp
e
r
im
e
nt
a
l
pr
ot
ot
y
pi
ng
u
s
in
g
publ
ic
ly
a
va
il
a
bl
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
N
ov
e
l
pr
e
e
m
pt
iv
e
i
nt
e
ll
ig
e
nt
ar
ti
fi
c
ia
l
in
te
ll
ig
e
nc
e
-
m
ode
l
fo
r
d
e
te
c
ti
ng i
nc
ons
is
te
nc
y
…
(
Sange
e
th
a G
o
v
in
da
)
1787
da
ta
s
e
t
of
out
l
ie
r
s
,
ii
i)
t
he
m
ode
l
is
c
a
pa
bl
e
of
e
xe
c
ut
in
g
va
r
i
e
d f
or
m
of
te
s
t
c
a
s
e
s
a
l
ong with con
s
tr
uc
ti
ng
n
e
w
te
s
t
c
a
s
e
s
in
or
d
e
r
to
ge
ne
r
a
te
a
dv
e
r
s
a
r
y
out
c
om
e
w
it
h
ne
ga
ti
v
e
r
e
s
ul
t
s
th
a
t
a
s
f
ur
th
e
r
r
e
c
or
de
d
a
s
tr
a
c
e
l
og
obj
e
c
t
s
,
iv
)
th
e
m
ode
l
pr
e
pr
oc
e
s
s
e
s
th
e
da
ta
f
ol
l
ow
e
d
by
a
ppl
y
in
g
e
ns
e
m
bl
e
d
A
I
-
m
od
e
ls
in
or
d
e
r
to
pe
r
f
or
m
de
te
c
ti
on
of
in
c
ons
i
s
t
e
nc
i
e
s
th
a
t
a
r
e
f
ur
th
e
r
a
na
l
yz
e
d
to
ge
n
e
r
a
t
e
a
v
e
r
di
c
t
t
o
a
s
s
e
s
s
t
he
ir
r
e
li
a
bi
li
ty
,
a
nd
v)
th
e
s
tu
dy
f
i
ndi
ng
s
s
how
c
a
s
e
s
th
a
t
s
upe
r
vi
s
e
d
s
c
he
m
e
of
S
L
2
u
s
in
g
m
ul
ti
la
ye
r
e
d
pe
r
c
e
pt
r
o
n
a
nd
s
up
por
t
v
e
c
to
r
m
a
c
hi
n
e
a
r
e
hi
gh
e
r
r
e
c
om
m
e
nde
d
ow
in
g
t
o
r
e
duc
e
d e
r
r
or
s
c
or
e
a
nd
o
pt
im
a
l
pr
o
c
e
s
s
i
ng
ti
m
e
i
nvol
v
e
m
e
nt
.
T
h
e
li
m
it
a
ti
on
of
th
e
pr
o
pos
e
d
s
tu
dy
i
s
th
a
t
it
is
n
ot
a
s
s
e
s
s
e
d
w
it
h
r
e
s
pe
c
t
to
a
l
l
th
e
u
s
e
-
c
a
s
e
s
w
h
e
r
e
th
e
pe
r
f
or
m
a
n
c
e
c
a
n
be
im
pr
ove
d
u
s
in
g
s
e
m
i
-
s
up
e
r
vi
s
e
d
a
ppr
oa
c
he
s
.
T
he
f
ut
ur
e
w
or
k
w
il
l
be
c
a
r
r
ie
d
out
us
i
ng
a
nove
l
f
or
m
of
s
e
m
i
-
s
up
e
r
vi
s
e
d
a
lg
or
it
hm
i
n
or
d
e
r
t
o
in
ve
s
ti
g
a
te
th
i
s
pe
r
s
pe
c
iv
e
.
F
ur
th
e
r
,
th
e
m
o
de
l
c
a
n
b
e
s
ubj
e
c
t
e
d t
o
im
pr
ov
e
m
e
nt
c
on
s
id
e
r
in
g m
or
e
n
um
be
r
of
t
e
s
t
-
c
a
s
e
s
w
it
h t
e
m
por
a
l
f
e
a
tu
r
e
s
.
A
C
K
N
O
WL
E
D
G
M
E
N
T
S
We
w
oul
d
li
ke
to
e
xpr
e
s
s
s
in
c
e
r
e
gr
a
ti
tu
de
to
C
hr
is
t
U
ni
ve
r
s
it
y
a
nd
S
t.
J
os
e
ph'
s
uni
ve
r
s
it
y
f
or
pr
ovi
di
ng
th
e
r
e
s
our
c
e
s
a
nd
s
uppor
t
ne
c
e
s
s
a
r
y
f
or
th
e
c
o
m
pl
e
ti
on
of
th
is
r
e
s
e
a
r
c
h.
T
he
a
ut
ho
rs
a
ls
o
a
c
knowle
dge
t
he
c
ont
r
ib
ut
io
ns
of
L
a
b t
e
a
m
a
nd c
ol
le
a
gue
s
f
or
t
he
ir
a
s
s
is
ta
n
c
e
w
it
h t
he
r
e
s
e
a
r
c
h w
or
k.
F
U
N
D
I
N
G
I
N
F
O
R
M
A
T
I
O
N
A
ut
hor
s
s
ta
te
no f
undi
ng i
nvol
ve
d.
A
U
T
H
O
R
C
O
N
T
R
I
B
U
T
I
O
N
S
S
T
A
T
E
M
E
N
T
T
hi
s
jo
ur
na
l
us
e
s
th
e
C
ont
r
ib
ut
or
R
ol
e
s
T
a
xonomy
(
C
R
e
di
T
)
to
r
e
c
ogni
z
e
in
di
vi
dua
l
a
ut
hor
c
ont
r
ib
ut
io
ns
, r
e
duc
e
a
ut
hor
s
hi
p di
s
put
e
s
,
a
nd f
a
c
il
it
a
te
c
ol
la
bo
r
a
ti
on.
N
am
e
o
f
A
u
t
h
or
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
S
a
nge
e
th
a
G
ovi
nda
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
B
.
G
.
P
r
a
s
a
nt
hi
✓
✓
✓
✓
✓
✓
✓
✓
✓
A
gne
s
N
a
li
ni
V
in
c
e
nt
✓
✓
✓
✓
✓
✓
✓
✓
C
:
C
onc
e
pt
ua
l
i
z
a
t
i
on
M
:
M
e
t
hodol
ogy
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
da
t
i
on
Fo
:
Fo
r
m
a
l
a
na
l
ys
i
s
I
:
I
nve
s
t
i
ga
t
i
on
R
:
R
e
s
our
c
e
s
D
:
D
a
t
a
C
ur
a
t
i
on
O
:
W
r
i
t
i
ng
-
O
r
i
gi
na
l
D
r
a
f
t
E
:
W
r
i
t
i
ng
-
R
e
vi
e
w
&
E
di
t
i
ng
Vi
:
Vi
s
ua
l
i
z
a
t
i
on
Su
:
Su
pe
r
vi
s
i
on
P
:
P
r
oj
e
c
t
a
dm
i
ni
s
t
r
a
t
i
on
Fu
:
Fu
ndi
ng a
c
qui
s
i
t
i
on
C
O
N
F
L
I
C
T
O
F
I
N
T
E
R
E
S
T
S
T
A
T
E
M
E
N
T
A
ut
hor
s
s
ta
te
no c
onf
li
c
t
of
i
nt
e
r
e
s
t.
D
A
T
A
A
V
A
I
L
A
B
I
L
I
T
Y
T
he
da
t
a
th
a
t
s
uppor
t
th
e
f
in
di
ngs
of
th
is
s
tu
dy
a
r
e
a
va
il
a
bl
e
f
r
om
th
e
c
or
r
e
s
ponding
a
ut
hor
upon
r
e
a
s
ona
bl
e
r
e
que
s
t.
R
E
F
E
R
E
N
C
E
S
[
1]
K
. S
a
l
a
ko a
nd X
. Z
ha
o, “
T
he
unne
c
e
s
s
i
t
y of
a
s
s
um
i
ng s
t
a
t
i
s
t
i
c
a
l
l
y i
nde
pe
nde
nt
t
e
s
t
s
i
n
B
a
ye
s
i
a
n
s
of
t
w
a
r
e
r
e
l
i
a
bi
l
i
t
y a
s
s
e
s
s
m
e
nt
s
,
”
I
E
E
E
T
r
ans
ac
t
i
ons
on Sof
t
w
ar
e
E
ngi
ne
e
r
i
ng
, vol
. 49, no. 4, pp. 2829
–
2838, 2023, doi
:
10.1109/
T
S
E
.2022.3233802.
[
2]
M
.
T
a
r
om
i
r
a
d
a
nd
P
.
R
un
e
s
on,
“
A
l
i
t
e
r
a
t
ur
e
s
ur
ve
y
of
a
s
s
e
r
t
i
ons
i
n
s
of
t
w
a
r
e
t
e
s
t
i
ng,”
E
ngi
ne
e
r
i
ng
of
C
om
put
e
r
-
B
as
e
d
Sy
s
t
e
m
s
,
pp. 75
–
96, 2024, doi
:
10.1007/
978
-
3
-
031
-
49252
-
5_8.
[
3]
S
.
S
t
r
a
dow
s
ki
a
nd
L
.
M
a
de
y
s
ki
,
“
E
xpl
or
i
ng
t
he
c
h
a
l
l
e
nge
s
i
n
s
of
t
w
a
r
e
t
e
s
t
i
ng
of
t
he
5G
s
ys
t
e
m
a
t
N
oki
a
:
a
s
ur
ve
y,”
I
nf
or
m
at
i
o
n
and Sof
t
w
ar
e
T
e
c
hnol
ogy
, vol
. 153, 2023, doi
:
10.1016/
j
.i
nf
s
of
.2022.107067.
[
4]
O
. P
a
r
r
y, G
. M
.
K
a
pf
ha
m
m
e
r
,
M
. H
i
l
t
on,
a
nd P
.
M
c
M
i
nn, “
A
s
ur
ve
y
of
F
l
a
ky
t
e
s
t
s
,”
A
C
M
T
r
ans
a
c
t
i
ons
on
Sof
t
w
ar
e
E
ngi
ne
e
r
i
n
g
and M
e
t
hodol
ogy
, vol
. 31, no. 1, 2021, doi
:
10.1145/
3476105.
[
5]
S
.
M
a
r
t
í
ne
z
-
F
e
r
ná
nde
z
e
t
al
.
,
“
S
of
t
w
a
r
e
e
ngi
ne
e
r
i
ng
f
o
r
A
I
-
ba
s
e
d
s
ys
t
e
m
s
:
a
s
ur
ve
y,”
A
C
M
T
r
ans
ac
t
i
ons
on
Sof
t
w
ar
e
E
ngi
ne
e
r
i
ng and M
e
t
hodol
ogy
, vol
. 31, no. 2, 2022, doi
:
10.1145/
3487043.
[
6]
V
.
G
a
r
ous
i
,
M
.
F
e
l
de
r
e
r
,
M
.
K
uhr
m
a
nn,
K
.
H
e
r
ki
l
oğl
u,
a
nd
S
.
E
l
dh,
“
E
xpl
or
i
ng
t
he
i
ndus
t
r
y’
s
c
ha
l
l
e
nge
s
i
n
s
of
t
w
a
r
e
t
e
s
t
i
ng:
a
n
e
m
pi
r
i
c
a
l
s
t
udy,”
J
our
nal
of
Sof
t
w
ar
e
:
E
v
ol
ut
i
on and P
r
oc
e
s
s
, vol
. 32, no. 8, 2
020, doi
:
10.1002/
s
m
r
.2251.
[
7]
T
.
F
ul
c
i
ni
,
R
.
C
oppol
a
,
L
.
A
r
di
t
o,
a
nd
M
.
T
or
c
hi
a
no,
“
A
r
e
vi
e
w
on
t
ool
s
,
m
e
c
ha
ni
c
s
,
be
ne
f
i
t
s
,
a
nd
c
ha
l
l
e
ng
e
s
of
g
a
m
i
f
i
e
d
s
of
t
w
a
r
e
t
e
s
t
i
ng,”
A
C
M
C
om
put
i
ng Sur
v
e
y
s
, vol
. 55, no. 14 s
, 2023, doi
:
10.114
5/
3582273.
[
8]
D
.
A
m
a
l
f
i
t
a
no,
S
.
F
a
r
a
l
l
i
,
J
.
C
.
R
.
H
a
uc
k,
S
.
M
a
t
a
l
onga
,
a
nd
D
.
D
i
s
t
a
nt
e
,
“
A
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
ge
nc
e
a
ppl
i
e
d
t
o
s
of
t
w
a
r
e
t
e
s
t
i
n
g:
a
t
e
r
t
i
a
r
y s
t
udy,”
A
C
M
C
om
put
i
ng Sur
v
e
y
s
, vol
. 56, no. 3, 2023, doi
:
10.1145/
36
16372.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
. 14, No. 3, J
une
2025
:
1781
-
1789
1788
[
9]
M
.
B
oukhl
i
f
,
M
.
H
a
ni
ne
,
a
nd
N
.
K
ha
r
m
oum
,
“
A
de
c
a
de
of
i
nt
e
l
l
i
ge
nt
s
o
f
t
w
a
r
e
t
e
s
t
i
ng
r
e
s
e
a
r
c
h:
a
bi
bl
i
om
e
t
r
i
c
a
na
l
ys
i
s
,
”
E
l
e
c
t
r
oni
c
s
, vol
. 12, no. 9, 2023, doi
:
10.3390/
e
l
e
c
t
r
oni
c
s
12092109.
[
10]
F
.
A
.
B
a
t
a
r
s
e
h,
R
.
M
ohod,
A
.
K
um
a
r
,
a
nd
J
.
B
ui
,
“
T
he
a
ppl
i
c
a
t
i
on
of
a
r
t
i
f
i
c
i
a
l
i
nt
e
l
l
i
ge
nc
e
i
n
s
of
t
w
a
r
e
e
ngi
ne
e
r
i
ng:
a
r
e
vi
e
w
c
ha
l
l
e
ngi
ng
c
onve
nt
i
ona
l
w
i
s
dom
,”
D
at
a
D
e
m
oc
r
ac
y
:
A
t
t
he
N
e
x
us
of
A
r
t
i
f
i
c
i
al
I
nt
e
l
l
i
ge
nc
e
,
Sof
t
w
ar
e
D
e
v
e
l
opm
e
nt
,
and
K
now
l
e
dge
E
ngi
ne
e
r
i
ng
, pp. 179
–
232, 2020, doi
:
10.1016/
B
978
-
0
-
12
-
818366
-
3.00010
-
1.
[
11]
A
.
F
ont
e
s
a
nd
G
.
G
a
y,
“
T
he
i
nt
e
gr
a
t
i
on
of
m
a
c
hi
ne
l
e
a
r
ni
ng
i
nt
o
a
ut
om
a
t
e
d
t
e
s
t
ge
ne
r
a
t
i
on:
a
s
ys
t
e
m
a
t
i
c
m
a
ppi
ng
s
t
udy,”
Sof
t
w
ar
e
T
e
s
t
i
ng V
e
r
i
f
i
c
at
i
on and R
e
l
i
abi
l
i
t
y
, vol
. 33, no. 4, 2023, doi
:
10.1002/
s
t
vr
.1845.
[
12]
J
.
A
.
P
.
L
i
m
a
a
nd
S
.
R
.
V
e
r
gi
l
i
o,
“
T
e
s
t
c
a
s
e
pr
i
or
i
t
i
z
a
t
i
on
i
n
c
ont
i
nuous
i
nt
e
gr
a
t
i
on
e
nvi
r
onm
e
nt
s
:
a
s
ys
t
e
m
a
t
i
c
m
a
ppi
ng
s
t
udy,
”
I
nf
or
m
at
i
on and Sof
t
w
ar
e
T
e
c
hnol
ogy
, vol
. 121, 2020, doi
:
10.1016/
j
.i
nf
s
of
.2020.106268.
[
13]
J
. J
.
L
i
,
A
.
U
l
r
i
c
h,
X
.
B
a
i
,
a
nd
A
.
B
e
r
t
ol
i
no,
“
A
dva
nc
e
s
i
n
t
e
s
t
a
ut
om
a
t
i
on f
or
s
of
t
w
a
r
e
w
i
t
h
s
pe
c
i
a
l
f
oc
us
on
a
r
t
i
f
i
c
i
a
l
i
nt
e
l
l
i
ge
nc
e
a
nd m
a
c
hi
ne
l
e
a
r
ni
ng,”
Sof
t
w
ar
e
Q
ual
i
t
y
J
our
nal
, vol
. 28, no. 1, pp. 245
–
248,
2020, doi
:
10.1007/
s
11219
-
019
-
09472
-
3.
[
14]
J
.
P
a
c
houl
y,
S
.
A
hi
r
r
a
o,
K
.
K
ot
e
c
ha
,
G
.
S
e
l
va
c
ha
ndr
a
n,
a
nd
A
.
A
br
a
ha
m
,
“
A
s
ys
t
e
m
a
t
i
c
l
i
t
e
r
a
t
ur
e
r
e
vi
e
w
on
s
of
t
w
a
r
e
de
f
e
c
t
pr
e
di
c
t
i
on
us
i
ng
a
r
t
i
f
i
c
i
a
l
i
nt
e
l
l
i
ge
nc
e
:
D
a
t
a
s
e
t
s
,
da
t
a
va
l
i
da
t
i
on
m
e
t
hods
,
a
p
pr
oa
c
he
s
,
a
nd
t
ool
s
,”
E
ngi
ne
e
r
i
ng
A
ppl
i
c
at
i
ons
of
A
r
t
i
f
i
c
i
al
I
nt
e
l
l
i
ge
nc
e
, vol
. 111, 2022, doi
:
10.1016/
j
.e
nga
ppa
i
.2022.104773.
[
15]
A
. G
a
r
t
z
i
a
n
di
a
e
t
a
l
.
,
“
M
a
c
h
i
ne
l
e
a
r
ni
ng
-
b
a
s
e
d
t
e
s
t
o
r
a
c
l
e
s
f
o
r
pe
r
f
o
r
m
a
nc
e
t
e
s
t
i
n
g of
c
ybe
r
-
ph
ys
i
c
a
l
s
ys
t
e
m
s
:
a
n
i
nd
us
t
r
i
a
l
c
a
s
e
s
t
ud
y
on e
l
e
va
t
or
s
d
i
s
pa
t
c
h
i
n
g a
l
go
r
i
t
hm
s
,”
J
our
n
al
o
f
S
of
t
w
ar
e
:
E
v
o
l
u
t
i
o
n
and
P
r
oc
e
s
s
, v
ol
. 3
4,
no
.
11,
2
022
,
doi
:
10
.10
02
/
s
m
r
.
246
5.
[
16]
V
.
R
ous
opoul
ou
e
t
al
.
,
“
C
ogni
t
i
ve
a
na
l
yt
i
c
s
pl
a
t
f
or
m
w
i
t
h
A
I
s
ol
ut
i
ons
f
o
r
a
n
om
a
l
y
de
t
e
c
t
i
on,”
C
om
put
e
r
s
i
n
I
ndus
t
r
y
,
vol
.
134,
2022, doi
:
10.1016/
j
.c
om
pi
nd.2021.103555.
[
17]
V
.
G
a
r
ous
i
,
S
.
B
a
ue
r
,
a
nd
M
.
F
e
l
de
r
e
r
,
“
N
L
P
-
a
s
s
i
s
t
e
d
s
of
t
w
a
r
e
t
e
s
t
i
ng:
a
s
ys
t
e
m
a
t
i
c
m
a
ppi
ng
of
t
he
l
i
t
e
r
a
t
ur
e
,”
I
nf
or
m
at
i
on
and
Sof
t
w
ar
e
T
e
c
hnol
ogy
, vol
. 126, 2020, doi
:
10.1016/
j
.i
nf
s
of
.2020.106321.
[
18]
A
.
A
.
R
obi
s
c
o
a
nd
J
.
M
.
C
.
M
a
r
t
í
ne
z
,
“
M
e
a
s
ur
i
ng
t
he
m
ode
l
r
i
s
k
-
a
dj
us
t
e
d
pe
r
f
or
m
a
nc
e
of
m
a
c
hi
ne
l
e
a
r
ni
ng
a
l
gor
i
t
hm
s
i
n
c
r
e
di
t
de
f
a
ul
t
pr
e
di
c
t
i
on,”
F
i
nanc
i
al
I
nnov
at
i
on
, vol
. 8, no. 1, 2022, doi
:
10.1186
/
s
40854
-
022
-
00366
-
1.
[
19]
S
. A
r
a
ke
l
ya
n, S
. A
r
a
s
t
e
h,
C
. H
a
u
s
e
r
, E
. K
l
i
ne
,
a
nd
A
. G
a
l
s
t
ya
n, “
B
i
n2ve
c
:
l
e
a
r
ni
ng r
e
pr
e
s
e
nt
a
t
i
ons
of
bi
na
r
y e
xe
c
ut
a
bl
e
pr
ogr
a
m
s
f
or
s
e
c
ur
i
t
y t
a
s
ks
,”
C
y
b
e
r
s
e
c
ur
i
t
y
, vol
. 4, no. 1, 2021, doi
:
10.1186/
s
42400
-
021
-
00088
-
4.
[
20]
S
.
Y
.
Y
e
r
i
m
a
,
M
.
K
.
A
l
z
a
yl
a
e
e
,
a
nd
S
.
S
e
z
e
r
,
“
M
a
c
hi
ne
l
e
a
r
ni
ng
-
ba
s
e
d
dyna
m
i
c
a
na
l
ys
i
s
of
A
ndr
oi
d
a
pps
w
i
t
h
i
m
pr
ove
d
c
ode
c
ove
r
a
ge
,”
E
ur
as
i
p J
ou
r
nal
on I
nf
or
m
at
i
on Se
c
ur
i
t
y
, vol
. 2019, no. 1, 2019, doi
:
10.1186/
s
13635
-
019
-
0087
-
1.
[
21]
T
.
H
a
i
,
J
.
Z
hou,
N
.
L
i
,
S
.
K
.
J
a
i
n,
S
.
A
gr
a
w
a
l
,
a
nd
I
.
B
.
D
ha
ou,
“
C
l
oud
-
ba
s
e
d
bug
t
r
a
c
ki
ng
s
of
t
w
a
r
e
de
f
e
c
t
s
a
na
l
ys
i
s
us
i
ng
de
e
p
l
e
a
r
ni
ng,”
J
our
nal
of
C
l
oud C
om
put
i
ng
, vol
. 11, no. 1, 2022, doi
:
10.1186/
s
136
77
-
022
-
00311
-
8.
[
22]
M
. C
ui
, S
.
L
ong,
Y
. J
i
a
ng, a
nd
X
. N
a
,
“
R
e
s
e
a
r
c
h
of
s
of
t
w
a
r
e
de
f
e
c
t
pr
e
di
c
t
i
on
m
ode
l
ba
s
e
d
on c
om
pl
e
x
ne
t
w
or
k a
nd gr
a
ph
ne
ur
a
l
ne
t
w
or
k,”
E
nt
r
opy
, vol
. 24, no. 10, 2022, doi
:
10.3390/
e
24101373.
[
23]
K
.
J
a
m
m
a
l
a
m
a
da
ka
a
nd
N
.
P
a
r
ve
e
n,
“
T
e
s
t
i
ng
c
ove
r
a
ge
c
r
i
t
e
r
i
a
f
or
opt
i
m
i
z
e
d
de
e
p
be
l
i
e
f
ne
t
w
or
k
w
i
t
h
s
e
a
r
c
h
a
nd
r
e
s
c
ue
,
”
J
our
nal
of
B
i
g D
at
a
, vol
. 8, no. 1, 2021, doi
:
10.1186/
s
40537
-
021
-
00453
-
7.
[
24]
N
.
L
a
r
a
nj
e
i
r
o,
S
.
N
.
S
oyde
m
i
r
,
N
.
I
va
ki
,
a
nd
J
.
B
e
r
na
r
di
no,
“
T
e
s
t
i
ng
da
t
a
-
c
e
nt
r
i
c
s
e
r
vi
c
e
s
us
i
ng
poor
qua
l
i
t
y
da
t
a
:
f
r
om
r
e
l
a
t
i
ona
l
t
o
N
o
S
Q
L
doc
um
e
nt
da
t
a
ba
s
e
s
,”
J
o
ur
na
l
of
t
h
e
B
r
az
i
l
i
a
n C
o
m
pu
t
e
r
So
c
i
e
t
y
,
v
ol
.
23,
no
.
1,
20
17,
d
oi
:
1
0.
118
6/
s
131
73
-
0
17
-
0
06
3
-
x.
[
25]
C
.
L
.
-
M
a
r
t
í
n,
“
M
a
c
hi
ne
l
e
a
r
ni
ng
t
e
c
hni
que
s
f
or
s
of
t
w
a
r
e
t
e
s
t
i
ng
e
f
f
or
t
pr
e
di
c
t
i
on,”
Sof
t
w
ar
e
Q
ual
i
t
y
J
our
nal
,
vol
.
30,
no.
1,
pp.
65
–
100, 2022, doi
:
10.1007/
s
11219
-
020
-
09545
-
8.
[
26]
A
.
S
e
ba
s
t
i
a
n,
H
.
N
a
s
e
e
m
,
a
nd
C
.
C
a
t
a
l
,
“
U
ns
upe
r
vi
s
e
d
m
a
c
hi
ne
l
e
a
r
ni
ng
a
ppr
oa
c
he
s
f
or
t
e
s
t
s
ui
t
e
r
e
duc
t
i
on,”
A
ppl
i
e
d
A
r
t
i
f
i
c
i
al
I
nt
e
l
l
i
ge
nc
e
, vol
. 38, no. 1, 2024, doi
:
10.1080/
08839514.2024.2322336.
[
27]
M
.
H
.
M
ogha
da
m
,
M
.
B
or
g,
M
.
S
a
a
da
t
m
a
nd,
S
.
J
.
M
ous
a
vi
r
a
d,
M
.
B
ohl
i
n,
a
nd
B
.
L
i
s
pe
r
,
“
M
a
c
hi
ne
l
e
a
r
ni
ng
t
e
s
t
i
ng
i
n
a
n
A
D
A
S
c
a
s
e
s
t
udy
us
i
ng
s
i
m
ul
a
t
i
on
-
i
nt
e
gr
a
t
e
d
bi
o
-
i
ns
pi
r
e
d
s
e
a
r
c
h
-
ba
s
e
d
t
e
s
t
i
ng,”
J
ou
r
nal
of
Sof
t
w
ar
e
:
E
v
ol
ut
i
on
and
P
r
oc
e
s
s
,
vol
.
36
,
no. 5, 2024, doi
:
10.1002/
s
m
r
.2591.
[
28]
F
. S
. A
hm
e
d, A
. M
a
j
e
e
d,
T
. A
. K
ha
n, a
nd S
. N
.
B
ha
t
t
i
, “
V
a
l
ue
-
ba
s
e
d c
o
s
t
-
c
ogni
z
a
nt
t
e
s
t
c
a
s
e
pr
i
or
i
t
i
z
a
t
i
on f
or
r
e
gr
e
s
s
i
on t
e
s
t
i
ng,
”
P
L
oS O
N
E
, vol
. 17, no. 5 M
a
y, 2022, doi
:
10.1371/
j
our
na
l
.pone
.0264972.
[
29]
R
. C
. C
h
e
n, C
. D
e
w
i
, S
. W
. H
ua
ng,
a
nd R
. E
.
C
a
r
a
ka
, “
S
e
l
e
c
t
i
ng
c
r
i
t
i
c
a
l
f
e
a
t
ur
e
s
f
or
da
t
a
c
l
a
s
s
i
f
i
c
a
t
i
on ba
s
e
d on m
a
c
hi
ne
l
e
a
r
ni
ng
m
e
t
hods
,”
J
our
nal
of
B
i
g D
at
a
, vol
. 7, no. 1, 2020, doi
:
10.1186/
s
40537
-
020
-
00327
-
4.
[
30]
Y
. G
or
i
s
hni
y, I
. R
uba
c
he
v,
V
. K
hr
ul
kov, a
nd A
.
B
a
be
nko,
“
R
e
vi
s
i
t
i
ng de
e
p l
e
a
r
ni
ng m
ode
l
s
f
or
t
a
bul
a
r
da
t
a
,”
A
dv
anc
e
s
i
n N
e
u
r
al
I
nf
or
m
at
i
on P
r
oc
e
s
s
i
ng Sy
s
t
e
m
s
, vol
. 23, pp. 18932
–
18943, 2021.
[
31]
P
.
F
.
O
r
r
ù,
A
.
Z
oc
c
he
ddu,
L
.
S
a
s
s
u,
C
.
M
a
t
t
i
a
,
R
.
C
oz
z
a
,
a
nd
S
.
A
r
e
na
,
“
M
a
c
hi
ne
l
e
a
r
ni
ng
a
ppr
oa
c
h
u
s
i
ng
M
L
P
a
nd
S
V
M
a
l
gor
i
t
hm
s
f
or
t
he
f
a
ul
t
p
r
e
di
c
t
i
on
of
a
c
e
nt
r
i
f
uga
l
pum
p
i
n
t
he
oi
l
a
nd
ga
s
i
ndus
t
r
y,”
Sus
t
ai
nabi
l
i
t
y
,
vol
.
12,
no.
11,
2020,
doi
:
10.3390/
s
u12114776.
[
32]
S
.
I
.
K
a
m
pe
z
i
dou,
A
.
T
.
R
a
y,
A
.
P
.
B
ha
t
,
O
.
J
.
P
.
F
i
s
c
he
r
,
a
nd
D
.
N
.
M
a
vr
i
s
,
“
F
unda
m
e
nt
a
l
c
om
pone
nt
s
a
nd
pr
i
nc
i
pl
e
s
of
s
upe
r
vi
s
e
d
m
a
c
hi
ne
l
e
a
r
ni
ng
w
or
kf
l
ow
s
w
i
t
h
num
e
r
i
c
a
l
a
nd
c
a
t
e
gor
i
c
a
l
da
t
a
,
”
E
ngi
ne
e
r
i
ng
,
vol
.
5,
no.
1,
pp.
384
–
416,
2024,
doi
:
10.3390/
e
ng5010021.
[
33]
L
.
P
r
okhor
e
nkova
,
G
.
G
us
e
v,
A
.
V
or
obe
v,
A
.
V
.
D
or
ogus
h,
a
nd
A
.
G
ul
i
n,
“
C
a
t
boos
t
:
unbi
a
s
e
d
boos
t
i
ng
w
i
t
h
c
a
t
e
gor
i
c
a
l
f
e
a
t
ur
e
s
,”
A
dv
anc
e
s
i
n N
e
ur
al
I
nf
or
m
at
i
on P
r
oc
e
s
s
i
ng Sy
s
t
e
m
s
, vol
. 2018, pp. 6
638
–
6648, 2018.
[
34]
S
.
H
a
l
da
r
a
nd
L
.
F
.
C
a
pr
e
t
z
,
“
I
nt
e
r
p
r
e
t
a
bl
e
s
of
t
w
a
r
e
de
f
e
c
t
pr
e
di
c
t
i
on
f
r
om
p
r
oj
e
c
t
e
f
f
or
t
a
nd
s
t
a
t
i
c
c
ode
m
e
t
r
i
c
s
,”
C
om
put
e
r
s
,
vol
. 13, no. 2, 2024, doi
:
10.3390/
c
om
put
e
r
s
13020052.
[
35]
H
.
Z
e
na
t
i
,
M
.
R
om
a
i
n,
C
.
S
.
F
oo,
B
.
L
e
c
oua
t
,
a
nd
V
.
C
ha
ndr
a
s
e
kha
r
,
“
A
dve
r
s
a
r
i
a
l
l
y
l
e
a
r
ne
d
a
nom
a
l
y
de
t
e
c
t
i
on,”
I
E
E
E
I
nt
e
r
nat
i
onal
C
onf
e
r
e
nc
e
on D
at
a M
i
ni
ng, I
C
D
M
, pp. 727
–
736, 2018, doi
:
10.1109/
I
C
D
M
.2018.00088.
[
36]
Z
.
P
e
ng,
X
.
X
i
a
o,
G
.
H
u,
A
.
K
.
S
a
nga
i
a
h,
M
.
A
t
i
quz
z
a
m
a
n,
a
nd
S
.
X
i
a
,
“
A
B
F
L
:
a
n
a
ut
oe
nc
ode
r
ba
s
e
d
pr
a
c
t
i
c
a
l
a
ppr
oa
c
h
f
or
s
of
t
w
a
r
e
f
a
ul
t
l
oc
a
l
i
z
a
t
i
on,”
I
nf
or
m
at
i
on Sc
i
e
nc
e
s
, vol
. 510, pp. 108
–
121, 2020,
doi
:
10.1016/
j
.i
ns
.2019.08.077.
[
37]
M
. M
a
npr
e
e
t
a
nd
J
. K
. C
hha
br
a
, “
A
hybr
i
d a
ppr
oa
c
h ba
s
e
d on k
-
ne
a
r
e
s
t
ne
i
ghb
or
s
a
nd de
c
i
s
i
on t
r
e
e
f
or
s
of
t
w
a
r
e
f
a
ul
t
pr
e
di
c
t
i
on,”
K
uw
ai
t
J
our
nal
of
Sc
i
e
nc
e
, 2022, doi
:
10.48129/
kj
s
.18331.
[
38]
K
.
B
a
r
ka
l
ov,
A
.
S
ht
a
nyuk,
a
nd
A
.
S
ys
oye
v,
“
A
f
a
s
t
kN
N
a
l
gor
i
t
hm
us
i
ng
m
ul
t
i
pl
e
s
pa
c
e
-
f
i
l
l
i
ng
c
ur
ve
s
,”
E
nt
r
opy
,
vol
.
24,
no.
6,
2022, doi
:
10.3390/
e
24060767.
[
39]
M
.
M
us
t
a
q
e
e
m
a
nd
M
.
S
a
qi
b,
“
P
r
i
nc
i
pa
l
c
om
pone
nt
ba
s
e
d
s
uppor
t
ve
c
t
or
m
a
c
hi
ne
(
P
C
-
S
V
M
)
:
a
hybr
i
d
t
e
c
hni
que
f
or
s
of
t
w
a
r
e
de
f
e
c
t
de
t
e
c
t
i
on,”
C
l
us
t
e
r
C
om
put
i
ng
, vol
. 24, no. 3, pp. 2581
–
2595, 2021, doi
:
10.1007/
s
10586
-
021
-
03282
-
8.
[
40]
Z
.
L
i
,
Y
.
Z
ha
o,
X
.
H
u,
N
.
B
ot
t
a
,
C
.
I
one
s
c
u,
a
nd
G
.
H
.
C
he
n,
“
E
C
O
D
:
un
s
upe
r
vi
s
e
d
out
l
i
e
r
de
t
e
c
t
i
on
us
i
ng
e
m
pi
r
i
c
a
l
c
um
ul
a
t
i
v
e
di
s
t
r
i
but
i
on
f
unc
t
i
ons
,”
I
E
E
E
T
r
ans
ac
t
i
ons
on
K
now
l
e
dge
and
D
at
a
E
ngi
ne
e
r
i
ng
,
vol
.
35,
no.
12,
pp.
12181
–
12193,
2023
,
doi
:
10.1109/
T
K
D
E
.2022.3159580.
[
41]
X
.
D
u,
E
.
Z
uo,
Z
.
C
hu,
Z
.
H
e
,
a
nd
J
.
Y
u,
“
F
l
uc
t
ua
t
i
on
-
ba
s
e
d
out
l
i
e
r
de
t
e
c
t
i
on,”
Sc
i
e
nt
i
f
i
c
R
e
por
t
s
,
vol
.
13,
no.
1,
2023,
doi
:
10.1038/
s
41598
-
023
-
29549
-
1.
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
N
ov
e
l
pr
e
e
m
pt
iv
e
i
nt
e
ll
ig
e
nt
ar
ti
fi
c
ia
l
in
te
ll
ig
e
nc
e
-
m
ode
l
fo
r
d
e
te
c
ti
ng i
nc
ons
is
te
nc
y
…
(
Sange
e
th
a G
o
v
in
da
)
1789
B
I
O
G
R
A
P
H
I
E
S
O
F
A
U
T
H
O
R
S
Dr.
Sangeetha
Govinda
an
associate
professor
in
the
Department
of
Computer
Scienc
e
at
Christ
(Dee
med
to
be
Univer
sity),
Centra
l
Campus,
Bangal
ore,
India,
holds
a
Ph
.
D
.
degree
from
Bharathiar
University,
Coimbatore
.
With
an
extensive
c
areer
spanning
over
19+
years,
she
has
made
significant
contributions
to
teaching,
research,
and
administration,
shaping
educational
methodologies
across
undergradua
te
and
po
stgraduate
levels.
Her
scholarly
endeavors
are
underscored
by
the
publication
of
16
natio
nal
and
10
international
research
papers
in
prestigio
us
journals
indexed
in
IEEE,
WoS,
and
Scopus.
She
has
5
textbooks
and
2
reference
books
to
her
credit.
Beyond
academia,
she
actively
serves
as
a
board
of
examination
(BOE)
member
for
esteemed
institutions
such
as
Bengaluru
City
University,
Bangalore
University,
and
Mount
Carmel
College.
Addit
ionally,
she
contributes
her
expertise
as
a
member
of
the
review
committee
for
ASTES
journ
al.
Her
diverse
research
interests
encompass
data
mining,
IoT,
software
engineering
,
c
ryptograph
y,
computer
networks
,
R
basics
,
and
IT
for
business
.
Her
dedication
to
acade
mic
excellence
extends
beyond
research,
as
evidenced
by
her
numerous
invited
talks,
guest
le
ctures,
international
and
national
conference
participation,
and
organization
of
workshops,
seminars,
and
Faculty
Development
Programs
(FDPs).
Recogniz
ed
for
her
outstandin
g
contribut
ions,
she
was
honored
as
a
Microsoft
Research
Fellow
in
2014.
Further
more
,
she
has
received
acclaim
for
her
innovative
work,
including
an
Australian
Patent
(No.
2021103341
)
granted
for
eight
years
from
June
15,
2021,
on
August
4,
2021,
for
her
groundbreaking
project
titled
"
Artificial
intelligence
based
automatic
detection
of
infection
rate
of
COVID
-
19
."
She
can
be
contacted
at email
: sangeet
ha.g@
christun
iversit
y.in
.
Dr.
B
.
G
.
Prasanthi
works
as
associate
dean
and
HOD
in
the
Department
of
Computer
Science
and
Applications
at
St
Joseph’s
University,
Bang
alore.
She
holds
degrees
of
M.Sc.
,
M.Tech
.
,
M.Phil
.
,
M.B.A
.,
and
Ph.D.
She
has
been
engage
d
in
innovative
teaching
practices,
resulting
in
100%
distinctions
to
all
MCA
students
in
subj
ects
like
file
structures,
computer
networks,
cyber
security,
and
mobile
apps.
Helped
stude
nts
in
taking
up
online
courses.
Her
research
endeavor
resulted
in
developing
an
embedded
chip
for
CISCO
routers.
She
also
believes
in
continuously
enhancing
knowledge
and
works
to
wards
the
same.
She
ha
s
published
30+
international
journals
and
20+
national
and
i
nternational
conference
manuscript
s.
As
a
supervisor,
she
guided
research
scholars
for
Ph
.
D
.
from
Bharathiar
University,
M.Phil.
students
for
PRIST
University,
M.Tech
.
,
and
MC
A
students
for
Bangalore
University,
and
MCA,
M
.
S
c.
students
for
St.
Joseph’s
University.
She
has
also
been
appointed
as
Deputy
Custodian
for
MCA,
M.Sc.,
computer
science
for
Ban
galore
University,
BOS
member
Bangalore
University,
Garden
City
University,
National
Coll
ege,
Maharanis
College,
Surana
College,
Ramaiah
College,
Jyothi
Nivas
College,
BOE
Jain
University,
Nati
onal
College,
Mahara
ni
Ammani
College,
Indian
Acade
my.
She
was
a
valuation
and
external
examiner
for
many
universi
ties
and
a
reviewer
for
many
internation
al
journals.
She
can
be
contacted
at email
:
prasanthi.b.g@
sju.edu.in
.
Ms
.
Agnes Nalini
Vincent
is the Dea
n of
Faculty of
Informatio
n
T
echnology a
nd
Head
of
Teaching
and
L
earning
at
AMITY
Institute
of
Higher
Educa
tion
(AIHE),
Mauritius.
She
holds
a
Master’s
in
Engineering
degree
from
Ann
a
University
,
Chennai,
India
and
is
currently
pursuin
g
her
Ph
.
D
.
in
the
field
of
artificial
intell
igence
.
With
over
17+
years
of
experience
in
teaching,
research
and
adminis
tration,
she
has
been
inst
rumental
in
framing
the
pedagogies
from
undergraduate
to
post
graduate
studies.
She
has
bee
n
the
pioneer
to
develop
curriculum
in
big
data
analytics,
internet
of
things
and
data
science
s
and
launched
them
as
courses
in
Mauriti
us.
She
has
publis
hed
national
and
internati
onal
res
earch
papers
in
journals
indexed
in
IEEE
and
Scopus.
As
a
faculty,
she
has
been
offering
a
wealth
of
talent
in
the
development
and
implementation
of
educational
technology
tools
and
applications
in
the
classroom
.
Her
area
of
inter
ests
includes
artificial
intell
igence,
data
mi
ning,
big
data
analytics
,
internet
of
things,
compute
r
networks,
and
business
data
analy
tics.
Sh
e
has
deeply
inv
ested
in
achieving
her
tenure
through
adminis
trative
service
contribu
tions
a
nd
an
accomplis
hment
-
oriented
approach
to
teaching.
She
has
given
more
than
10+
invited
talks,
12+
guest
lectures,
has
conducted
1
internationa
l
confer
ence,
has
organized
5+
nat
ional
and
internationa
l
workshops.
Her
contribution
to
quality
stan
dards
formulation
and
t
o
development
of
credit
system
in
th
e
capacity
of
head
of
teaching
and
learning
at
AIHE
is
r
emarkable.
Her
sterling
track
record
of
academic
excellence
and
visionary
leadership
brings
a
wealth
of
knowledge,
experience,
and
insigh
t
to
the
forefront
of
the
ICT
field.
Through
her
unique
and
people
-
centric
approach
towards
adminis
tration
and
management
,
she
has
always
fostered
a
supportive
and
collaborative
environment
that
enables
each
member
o
f
the
team
to
reach
their
full pot
ential. S
h
e can be con
tacted at
email:
vanalini@
mauritius.amity.edu
.
Evaluation Warning : The document was created with Spire.PDF for Python.