I
n
d
on
e
s
i
a
n
Jo
u
r
n
al
of
E
l
e
c
t
r
i
c
al
E
n
g
i
n
e
e
r
i
n
g a
n
d
C
om
p
u
t
e
r
S
c
i
e
n
c
e
V
ol
.
40
, N
o.
1
,
O
c
t
obe
r
2025
, p
p.
490
~
498
I
S
S
N
:
2502
-
4752, D
O
I
:
10.11591/
i
j
e
e
c
s
.
v
40
.i
1
.
pp
490
-
498
490
Jou
r
n
al
h
om
e
page
:
ht
t
p:
/
/
i
j
e
e
c
s
.i
ae
s
c
or
e
.c
om
A
n
i
m
p
r
ove
d
c
o
n
ve
r
sat
i
o
n
e
m
ot
i
o
n
d
e
t
e
c
t
i
o
n
u
si
n
g
h
y
b
r
i
d
f
-
n
n
c
l
ass
i
f
i
e
r
A
b
h
i
s
h
e
k
A
. V
i
c
h
ar
e
1
,2
,
S
at
i
s
h
k
u
m
ar
L
. V
ar
m
a
3
1
D
e
pa
r
t
m
e
n
t
of
C
o
m
put
e
r
,
M
uke
s
h
P
a
t
e
l
S
c
h
ool
of
T
e
c
hnol
o
g
y
M
a
n
a
g
e
m
e
nt
a
n
d
E
ngi
n
e
e
r
i
ng,
S
V
K
M
's
N
M
I
M
S
,
M
u
m
ba
i
,
I
n
di
a
2
D
e
pa
r
t
m
e
n
t
of
C
o
m
put
e
r
,
P
i
l
l
a
i
C
ol
l
e
g
e
of
E
ngi
n
e
e
r
i
ng
,
N
e
w
P
a
n
ve
l
,
I
n
di
a
3
D
e
pa
r
t
m
e
n
t
of
I
n
f
o
r
m
a
t
i
o
n
T
e
c
hn
ol
o
g
y
,
D
w
a
r
ka
da
s
J
.
S
a
n
gh
vi
C
ol
l
e
g
e
of
E
ngi
n
e
e
r
i
ng
,
V
i
l
e
P
a
r
l
e
,
M
u
m
ba
i
,
I
n
di
a
A
r
t
i
c
l
e
I
n
f
o
A
B
S
T
R
A
C
T
A
r
t
i
c
l
e
h
i
s
t
o
r
y
:
R
e
c
e
i
v
e
d
O
c
t
4
,
2024
R
e
v
i
s
e
d
M
a
r
19
,
2025
A
c
c
e
pt
e
d
J
ul
4
,
2025
E
m
o
ti
o
n
r
e
c
o
g
n
i
ti
o
n
f
r
o
m
tex
t
i
s
a
c
r
u
c
i
a
l
t
a
s
k
i
n
n
a
tu
r
a
l
l
a
n
g
u
a
g
e
p
r
o
c
e
s
s
i
n
g
(
NL
P
)
wi
th
a
p
p
li
c
a
t
i
o
n
s
i
n
s
e
n
ti
m
e
n
t
a
n
a
l
y
s
i
s
,
h
u
m
a
n
-
c
o
m
p
u
ter
i
n
ter
a
c
t
i
o
n
,
a
n
d
p
s
y
c
h
o
l
o
g
i
c
a
l
r
e
s
e
a
r
c
h
.
I
n
th
i
s
s
tu
d
y
,
we
p
r
e
s
e
n
t
a
n
o
v
e
l
a
p
p
r
o
a
c
h
f
o
r
tex
t
-
b
a
s
e
d
e
m
o
ti
o
n
r
e
c
o
g
n
i
t
i
o
n
u
s
i
n
g
a
m
o
d
i
f
i
e
d
f
i
r
e
f
l
y
a
lg
o
r
i
th
m
(
M
F
A
)
.
T
h
e
f
ir
e
f
l
y
a
l
g
o
r
i
th
m
i
s
a
s
wa
r
m
i
n
telli
g
e
n
c
e
m
e
th
o
d
i
n
s
p
i
r
e
d
b
y
th
e
b
i
o
l
u
m
i
n
e
s
c
e
n
t
c
o
m
m
u
n
i
c
a
ti
o
n
o
f
f
i
r
e
f
li
e
s
,
a
n
d
i
t
i
s
k
n
o
wn
f
o
r
i
ts
s
i
m
p
li
c
i
ty
a
n
d
e
f
f
i
c
i
e
n
c
y
i
n
o
p
ti
m
iza
t
i
o
n
tas
k
s
.
I
n
th
i
s
p
a
p
e
r
M
F
A
-
b
a
s
e
d
m
o
d
e
l
i
s
e
v
a
l
u
a
ted
o
n
t
h
e
i
n
ter
n
a
ti
o
n
a
l
s
u
r
v
e
y
o
n
e
m
o
ti
o
n
a
n
tec
e
d
e
n
ts
a
n
d
r
e
a
c
ti
o
n
s
(
I
S
E
AR
)
d
a
tas
e
t,
wh
ich
i
n
c
l
u
d
e
s
tex
t
e
n
tr
i
e
s
c
a
teg
o
r
i
z
e
d
b
y
v
a
r
i
o
u
s
e
m
o
ti
o
n
s
.
E
x
p
e
r
i
m
e
n
tal
r
e
s
u
l
ts
i
n
d
i
c
a
te
th
a
t
o
u
r
a
p
p
r
o
a
c
h
a
c
h
i
e
v
e
d
p
r
o
m
i
s
i
n
g
o
u
tco
m
e
s
.
S
p
e
c
i
f
i
c
a
ll
y
,
th
e
p
r
o
p
o
s
e
d
m
e
th
o
d
,
wh
i
c
h
c
o
m
b
i
n
e
s
th
e
f
ir
e
f
l
y
a
l
g
o
r
i
th
m
wi
th
a
m
u
l
ti
l
a
y
e
r
p
e
r
c
e
p
tr
o
n
(
M
L
P
)
,
a
tt
a
i
n
e
d
a
n
a
c
c
u
r
a
c
y
o
f
9
2
.
0
7
%
,
s
u
r
p
a
s
s
i
n
g
m
o
s
t
o
t
h
e
r
a
p
p
r
o
a
c
h
e
s
r
e
p
o
r
ted
i
n
th
e
l
i
te
r
a
tu
r
e
.
K
e
y
w
or
d
s
:
E
m
ot
i
on r
e
c
og
ni
t
i
on
F
i
r
e
f
l
y
a
l
g
o
r
i
t
h
m
M
a
c
hi
ne
L
e
a
r
ni
ng
N
L
P
S
w
a
r
m
i
nt
e
l
l
i
g
e
nc
e
T
h
i
s
i
s
a
n
o
p
e
n
a
c
c
e
s
s
a
r
ti
c
l
e
u
n
d
e
r
th
e
C
C
B
Y
-
SA
l
i
c
e
n
s
e
.
C
or
r
e
s
pon
di
n
g A
u
t
h
or
:
A
bhi
s
he
k A
. V
i
c
h
a
r
e
D
e
pa
r
t
m
e
nt
of
C
o
m
put
e
r
, M
uke
s
h P
a
t
e
l
S
c
hool
of
T
e
c
hnol
ogy
M
a
na
g
e
m
e
nt
a
nd E
ng
i
ne
e
r
i
ng
S
V
K
M
'
s
N
M
I
M
S
M
um
ba
i
,
I
ndi
a
E
m
a
i
l
:
v
i
c
ha
r
e
1@
gm
a
i
l
.c
o
m
1.
I
N
T
R
O
D
U
C
T
I
O
N
E
m
ot
i
on
r
e
c
o
g
ni
t
i
on
f
r
o
m
t
e
xt
i
s
i
m
por
t
a
nt
r
e
s
e
a
r
c
h
do
m
a
i
n
i
n
na
t
ur
a
l
l
a
n
g
ua
g
e
pr
o
c
e
s
s
i
ng
(
N
L
P
)
due
t
o
i
t
s
w
i
de
r
a
n
g
e
of
a
ppl
i
c
a
t
i
ons
[
1]
-
[
3]
.
U
nde
r
s
t
a
ndi
ng
hu
m
a
n
e
m
ot
i
ons
e
m
b
e
dde
d
i
n
t
e
xt
ua
l
da
t
a
c
a
n
e
nha
nc
e
t
he
p
e
r
f
or
m
a
nc
e
of
v
a
r
i
ous
s
y
s
t
e
m
s
s
uc
h
a
s
c
h
a
t
bot
s
,
s
oc
i
a
l
m
e
di
a
a
n
a
l
y
z
i
n
g
t
ool
s
,
a
nd
m
e
nt
a
l
he
a
l
t
h
a
ppl
i
c
a
t
i
ons
.
U
nde
r
s
t
a
ndi
ng
e
m
ot
i
on
i
s
c
ons
i
de
r
e
d
a
s
c
ha
l
l
e
ng
i
ng
t
a
s
k
[
1]
,
[
2]
.
E
m
o
t
i
on
r
e
c
og
ni
t
i
on
i
s
i
m
por
t
a
nt
be
c
a
us
e
e
m
ot
i
ons
c
a
n
a
ppe
a
r
i
n
di
f
f
e
r
e
nt
w
a
y
s
,
l
i
ke
s
t
r
e
s
s
.
H
e
a
l
t
h
ps
y
c
hol
ogi
s
t
s
s
t
udy
how
t
o
i
de
nt
i
f
y
e
m
ot
i
ons
t
o
he
l
p
pa
t
i
e
nt
s
by
unde
r
s
t
a
ndi
ng
t
he
c
onne
c
t
i
on
be
t
w
e
e
n
phy
s
i
c
a
l
w
e
l
l
-
be
i
ng
,
s
t
r
e
s
s
,
a
nd
e
m
ot
i
ona
l
s
t
a
t
e
[
3]
.
A
m
on
g
t
he
nu
m
e
r
ous
da
t
a
s
e
t
s
a
v
a
i
l
a
bl
e
f
or
e
m
ot
i
on r
e
c
o
g
ni
t
i
on, t
he
I
n
t
e
r
na
t
i
ona
l
S
ur
v
e
y
on
E
m
ot
i
on
A
nt
e
c
e
de
nt
s
a
nd
R
e
a
c
t
i
ons
(
I
S
E
A
R
)
d
a
t
a
s
e
t
s
t
a
nds
out
due
t
o
i
t
s
c
om
pr
e
he
n
s
i
v
e
c
ol
l
e
c
t
i
on
of
e
m
ot
i
ona
l
r
e
s
pons
e
s
a
c
r
os
s
di
v
e
r
s
e
s
c
e
na
r
i
os
[
4
]
-
[
6
].
T
r
a
di
t
i
ona
l
ML
a
nd
DL
a
ppr
o
a
c
he
s
h
a
v
e
be
e
n
e
xt
e
ns
i
v
e
l
y
us
e
d
f
or
e
m
ot
i
on
r
e
c
o
g
ni
t
i
on
t
a
s
ks
.
H
o
w
e
v
e
r
,
t
he
s
e
a
ppr
oa
c
he
s
of
t
e
n
r
e
qui
r
e
l
a
r
g
e
a
m
ount
s
of
l
a
be
l
l
e
d
da
t
a
a
nd
i
m
por
t
a
nt
c
o
m
put
a
t
i
ona
l
r
e
s
our
c
e
s
,
w
hi
c
h
m
a
y
not
a
l
w
a
y
s
be
f
e
a
s
i
bl
e
.
T
r
a
di
t
i
ona
l
a
l
g
or
i
t
h
m
s
ha
v
e
not
pr
o
v
i
de
d
g
ood
r
e
s
ul
t
s
o
n
I
S
E
A
R
da
t
a
s
e
t
.
A
s
a
n
a
l
t
e
r
na
t
i
v
e
,
s
w
a
r
m
i
nt
e
l
l
i
g
e
nc
e
a
l
g
or
i
t
h
m
s
,
i
nf
l
ue
nc
e
d
by
t
he
c
ol
l
e
c
t
i
v
e
be
ha
v
i
our
of
s
oc
i
a
l
or
g
a
ni
s
m
s
,
of
f
e
r
a
pr
o
m
i
s
i
ng
s
ol
ut
i
on
[
7
]
,
[
8
]
.
T
he
s
e
a
l
g
or
i
t
h
m
s
a
r
e
i
de
nt
i
f
i
e
d
f
o
r
t
he
i
r
s
i
m
pl
i
c
i
t
y
,
f
l
e
xi
bi
l
i
t
y
,
a
nd
a
bi
l
i
t
y
t
o
f
i
nd
f
e
a
s
i
bl
e
s
ol
ut
i
ons
i
n
c
ha
l
l
e
ng
i
ng
s
e
a
r
c
h
s
pa
c
e
s
w
i
t
h
r
e
l
a
t
i
v
e
l
y
l
o
w
c
o
m
put
a
t
i
ona
l
c
os
t
[
9
]
. O
ur
s
t
udy
a
i
m
s
t
o i
n
v
ol
v
e
t
he
F
i
r
e
f
l
y
A
l
g
or
i
t
h
m
t
o i
m
p
r
o
v
e
ho
w
f
e
a
t
ur
e
s
a
r
e
s
e
l
e
c
t
e
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
ndone
s
i
a
n J
E
l
e
c
E
n
g
&
C
om
p S
c
i
I
S
S
N
:
2502
-
4752
A
n i
m
pr
ov
e
d
c
onv
e
r
s
at
i
on e
m
ot
i
o
n
d
e
t
e
c
t
i
on
us
i
ng hy
b
r
i
d
f
-
nn c
l
as
s
i
f
i
e
r
(
A
b
hi
s
he
k
A
. V
i
c
har
e
)
491
a
nd
c
l
a
s
s
i
f
i
e
d
f
or
e
m
ot
i
on
r
e
c
o
g
ni
t
i
on
us
i
ng
t
he
I
S
E
A
R
da
t
a
s
e
t
.
T
he
I
S
E
A
R
da
t
a
s
e
t
c
o
ns
i
s
t
s
e
m
ot
i
ona
l
r
e
s
pons
e
s
c
a
t
e
g
or
i
z
e
d
i
nt
o
s
e
v
e
n
e
m
ot
i
ons
(
j
oy
,
f
e
a
r
,
a
n
g
e
r
,
s
a
dne
s
s
,
di
s
g
us
t
,
s
ha
m
e
,
a
nd
g
ui
l
t
)
,
pr
ov
i
de
s
a
r
obus
t
f
ounda
t
i
on f
or
e
v
a
l
ua
t
i
ng
t
he
pr
opos
e
d a
ppr
oa
c
h.
I
n
t
hi
s
s
t
udy
,
w
e
i
n
v
e
s
t
i
ga
t
e
ho
w
w
e
l
l
t
he
f
i
r
e
f
l
y
a
l
g
or
i
t
h
m
(
F
A
)
,
a
t
y
pe
of
s
w
a
r
m
i
nt
e
l
l
i
g
e
nc
e
m
e
t
hod,
w
or
ks
f
or
i
de
nt
i
f
y
i
ng
e
m
ot
i
ons
i
n
t
e
xt
.
T
he
F
A
i
s
i
n
s
pi
r
e
d
by
how
f
i
r
e
f
l
i
e
s
us
e
l
i
g
ht
t
o
c
om
m
uni
c
a
t
e
a
nd
a
t
t
r
a
c
t
e
a
c
h
ot
he
r
.
I
t
h
a
s
be
e
n
e
f
f
e
c
t
i
v
e
l
y
us
e
d
i
n
m
a
n
y
opt
i
m
i
z
a
t
i
on
t
a
s
ks
,
s
how
i
ng
f
a
s
t
a
nd
a
c
c
u
r
a
t
e
r
e
s
ul
t
s
.
W
e
be
g
i
n
by
p
r
e
-
pr
o
c
e
s
s
i
ng
t
he
t
e
xt
da
t
a
t
o
e
xt
r
a
c
t
m
e
a
ni
n
g
f
ul
f
e
a
t
ur
e
s
t
ha
t
c
a
n
a
c
c
ur
a
t
e
l
y
r
e
pr
e
s
e
nt
t
he
e
m
ot
i
ona
l
c
ont
e
nt
.
N
e
xt
,
w
e
e
m
pl
o
y
t
he
f
i
r
e
f
l
y
a
l
g
o
r
i
t
h
m
t
o
opt
i
m
i
z
e
t
he
s
e
l
e
c
t
i
on
of
t
he
s
e
f
e
a
t
ur
e
s
,
a
i
m
i
ng
t
o
i
nc
r
e
a
s
e
t
he
c
l
a
s
s
i
f
i
c
a
t
i
on
pe
r
f
or
m
a
nc
e
.
F
i
na
l
l
y
,
w
e
a
s
s
e
s
s
t
he
r
e
s
ul
t
s
of
o
ur
pr
op
os
e
d
m
e
t
hod
w
i
t
h
t
r
a
di
t
i
ona
l
t
e
c
hni
que
s
t
o
de
m
ons
t
r
a
t
e
i
t
s
e
f
f
e
c
t
i
v
e
ne
s
s
.
I
n
t
hi
s
w
or
k,
w
e
g
i
v
e
a
t
hor
ough
i
nv
e
s
t
i
g
a
t
i
on
of
e
m
ot
i
on
r
e
c
o
g
ni
t
i
on
i
n
c
onv
e
r
s
a
t
i
on
t
e
xt
us
i
ng
f
e
a
t
ur
e
s
e
l
e
c
t
i
on
a
nd
t
he
f
i
r
e
f
l
y
a
l
g
or
i
t
hm
.
T
hr
o
u
g
h
t
hi
s
i
nt
e
r
di
s
c
i
pl
i
na
r
y
r
e
s
e
a
r
c
h,
w
e
a
i
m
t
o pr
o
v
i
de
g
r
ound f
o
r
f
ut
ur
e
s
t
udi
e
s
a
t
t
he
i
nt
e
r
s
e
c
t
i
on of
c
o
m
put
e
r
s
c
i
e
nc
e
a
nd bi
ol
og
y
, e
nc
our
a
g
i
ng
ne
w
i
de
a
s
a
nd di
s
c
o
v
e
r
i
e
s
i
n e
m
ot
i
on r
e
c
og
ni
t
i
on a
nd ot
he
r
a
r
e
a
s
.
W
hi
l
e
s
e
nt
i
m
e
nt
a
na
l
y
s
i
s
a
nd
e
m
ot
i
on
r
e
c
o
g
ni
t
i
on
a
r
e
bot
h
c
onc
e
r
ne
d
w
i
t
h
unde
r
s
t
a
ndi
ng
t
he
f
e
e
l
i
ng
s
e
xpr
e
s
s
e
d
i
n
t
e
xt
,
t
he
y
s
e
r
v
e
di
f
f
e
r
e
nt
pur
pos
e
s
a
nd
r
e
qui
r
e
di
f
f
e
r
e
nt
a
ppr
o
a
c
he
s
[
10]
.
S
e
nt
i
m
e
nt
a
na
l
y
s
i
s
g
i
v
e
s
a
s
u
m
m
a
r
y of
t
he
o
v
e
r
a
l
l
s
e
nt
i
m
e
nt
,
w
he
r
e
a
s
e
m
ot
i
on r
e
c
o
g
ni
t
i
on di
g
s
i
nt
o i
de
nt
i
f
y
i
ng
s
pe
c
i
f
i
c
e
m
ot
i
ona
l
s
t
a
t
e
s
.
B
ot
h
a
r
e
v
a
l
ua
bl
e
i
n
v
a
r
i
ous
a
ppl
i
c
a
t
i
ons
,
f
r
om
m
a
r
ke
t
r
e
s
e
a
r
c
h
a
nd
c
us
t
om
e
r
s
e
r
v
i
c
e
t
o
ps
y
c
hol
og
i
c
a
l
s
t
udi
e
s
a
nd
hu
m
a
n
-
c
o
m
put
e
r
i
nt
e
r
a
c
t
i
on
[
11]
.
E
m
ot
i
on
r
e
c
o
g
ni
t
i
on,
a
l
s
o
know
n
a
s
e
m
ot
i
on
de
t
e
c
t
i
on,
i
s
t
he
m
e
t
hod
of
i
de
nt
i
f
y
i
ng
t
he
e
m
ot
i
ons
c
onv
e
ye
d
i
n
da
t
a
.
E
m
ot
i
on
r
e
c
og
ni
t
i
on
i
s
pr
om
i
s
i
ng
ne
w
f
i
e
l
d
[
12]
.
I
t
g
oe
s
be
y
ond
t
he
g
e
ne
r
a
l
s
e
nt
i
m
e
nt
t
o
pi
npoi
nt
pa
r
t
i
c
ul
a
r
e
m
ot
i
ons
s
uc
h
a
s
h
a
ppi
ne
s
s
,
s
a
dne
s
s
,
a
ng
e
r
, f
e
a
r
, s
ur
pr
i
s
e
, or
di
s
g
us
t
.
T
he
c
ont
r
i
but
i
ons
of
t
hi
s
r
e
s
e
a
r
c
h
a
r
e
t
hr
e
e
f
ol
d:
i
)
i
nt
r
oduc
i
ng
t
he
f
i
r
e
f
l
y
a
l
g
or
i
t
h
m
a
s
a
no
v
e
l
a
ppr
oa
c
h
f
or
e
m
ot
i
on
r
e
c
o
g
ni
t
i
on
f
r
o
m
t
e
xt
,
ii
)
opt
i
m
i
z
i
ng
f
e
a
t
ur
e
e
xt
r
a
c
t
i
on
us
i
ng
t
he
f
i
r
e
f
l
y
a
l
g
or
i
t
h
m
t
o
i
nc
r
e
a
s
e
c
l
a
s
s
i
f
i
c
a
t
i
on
a
c
c
ur
a
c
y
,
a
nd
iii
)
of
f
e
r
i
ng
a
t
hor
oug
h
a
s
s
e
s
s
m
e
nt
of
t
he
s
ug
g
e
s
t
e
d
a
ppr
oa
c
h
on
t
he
I
S
E
A
R
da
t
a
s
e
t
.
I
n
t
he
f
ol
l
ow
i
ng
s
e
c
t
i
ons
,
w
e
pr
e
s
e
nt
a
de
t
a
i
l
e
d
o
v
e
r
v
i
e
w
of
r
e
l
a
t
e
d
w
or
k,
de
s
c
r
i
be
t
he
m
e
t
hodol
ogy
, a
nd di
s
c
us
s
t
he
e
xpe
r
i
m
e
nt
a
l
r
e
s
ul
t
s
.
2.
L
I
T
E
R
A
T
U
R
E
R
E
V
I
E
W
S
w
a
r
m
i
nt
e
l
l
i
g
e
n
c
e
a
l
g
o
r
i
t
hm
s
a
r
e
c
o
m
m
onl
y
a
ppl
i
e
d
f
or
ha
ndl
i
ng
c
o
m
pl
e
x
pr
obl
e
m
s
by
m
i
m
i
c
ki
n
g
how
s
oc
i
a
l
c
r
e
a
t
ur
e
s
l
i
ke
a
nt
s
,
bi
r
ds
,
a
nd
be
e
s
w
or
k
t
og
e
t
he
r
[
1
3
]
,
[1
4
]
.
T
he
s
e
a
l
g
or
i
t
h
m
s
a
r
e
di
f
f
e
r
e
nt
f
r
o
m
t
r
a
di
t
i
ona
l
m
e
t
hods
be
c
a
us
e
t
he
y
c
a
n
qui
c
kl
y
e
xpl
or
e
pos
s
i
bl
e
s
ol
u
t
i
ons
a
nd
f
i
nd
v
e
r
y
good
a
ns
w
e
r
s
.
T
hi
s
r
e
v
i
e
w
l
ooks
a
t
i
m
por
t
a
nt
r
e
s
e
a
r
c
h
a
nd
us
e
s
of
s
w
a
r
m
i
nt
e
l
l
i
g
e
nc
e
a
l
g
or
i
t
h
m
s
,
f
oc
us
i
ng
on
how
t
he
y'
r
e
us
e
d
i
n
di
f
f
e
r
e
nt
a
r
e
a
s
a
nd
ho
w
w
e
l
l
t
he
y
s
ol
v
e
r
e
a
l
pr
obl
e
m
s
.
B
y
s
t
udy
i
ng
t
he
s
e
a
l
g
or
i
t
h
m
s
,
a
unde
r
s
t
a
ndi
ng
t
o
s
e
e
how
t
he
f
i
r
e
f
l
y
a
l
g
or
i
t
h
m
c
a
n
be
us
e
f
ul
f
or
r
e
c
o
g
ni
z
i
ng
e
m
ot
i
ons
f
r
om
t
e
xt
i
n
t
he
I
S
E
A
R
da
t
a
s
e
t
i
s
pr
opos
e
d.
A
r
un
e
t
al
.
[
15]
i
nt
r
oduc
e
d
a
ppr
o
a
c
h
t
o
f
a
c
i
a
l
m
i
c
r
o
e
xpr
e
s
s
i
on
e
m
ot
i
on
r
e
c
o
g
ni
t
i
on
us
i
ng
s
w
a
r
m
b
a
s
e
d
m
o
di
f
i
e
d
C
N
N
.
T
h
i
s
a
p
pr
oa
c
h
a
c
hi
e
v
e
d
99
.
4%
a
c
c
ur
a
c
y
i
n
i
de
n
t
i
f
y
i
ng
f
a
c
i
a
l
e
m
ot
i
o
ns
.
O
l
m
e
z
e
t
al
.
[
16]
pr
opos
e
d
m
odi
f
i
e
d
s
w
a
r
m
i
nt
e
l
l
i
g
e
nc
e
ba
s
e
d
P
S
O
a
l
g
or
i
t
h
m
t
o
i
m
pr
o
v
e
e
f
f
e
c
t
i
v
e
n
e
s
s
f
or
E
E
G
-
ba
s
e
d
hu
m
a
n
e
m
ot
i
on
r
e
c
o
g
ni
t
i
on.
H
a
m
di
e
t
al
.
[
17
]
pr
e
s
e
nt
e
d
a
n
a
f
f
i
r
m
a
t
i
v
e
a
nt
c
ol
ony
opt
i
m
i
z
a
t
i
on
(
A
C
O
)
a
p
pr
oa
c
h
c
o
m
bi
ne
d
w
i
t
h a
S
V
M
t
o i
de
nt
i
f
y
t
he
s
e
nt
i
m
e
nt
s
a
nd e
m
ot
i
ons
hi
dde
n i
n t
he
t
e
xt
ua
l
w
or
ds
. S
V
M
w
a
s
us
e
d a
s
t
he
c
l
a
s
s
i
f
i
e
r
a
nd
t
he
A
C
O
a
l
g
or
i
t
hm
w
a
s
us
e
d
f
or
f
e
a
t
ur
e
s
e
l
e
c
t
i
on.
S
i
g
ni
f
i
c
a
nt
g
a
i
ns
i
n
c
a
t
e
g
or
i
z
a
t
i
on
a
c
c
ur
a
c
y
w
e
r
e
not
e
d by
t
he
s
t
udy
.
H
a
dn
i
e
t
al
.
[
18
]
s
u
gg
e
s
t
e
d
a
ne
w
m
e
t
hod
f
or
A
r
a
bi
c
f
e
a
t
ur
e
s
e
l
e
c
t
i
on
us
i
ng
t
he
f
i
r
e
f
l
y
a
l
g
or
i
t
h
m
(
C
F
A
)
a
nd
t
he
c
ha
ot
i
c
a
ppr
oa
c
h.
T
he
r
e
s
ul
t
s
f
r
o
m
t
he
c
ha
ot
i
c
a
ppl
i
c
a
t
i
on
i
s
us
e
d
t
o
r
e
pl
a
c
e
t
he
f
i
r
e
f
l
y
a
l
g
or
i
t
h
m
'
s
a
t
t
r
a
c
t
i
on
c
oe
f
f
i
c
i
e
nt
.
T
he
i
m
p
r
o
v
e
m
e
nt
a
l
s
o
br
oug
ht
a
ne
w
s
e
a
r
c
h
m
e
t
hod.
U
t
i
l
i
s
i
ng
c
l
a
s
s
i
f
i
e
r
s
l
i
ke
N
a
i
v
e
B
a
y
e
s
(
N
B
)
,
S
V
M
,
a
nd
K
-
ne
a
r
e
s
t
ne
i
g
hbour
s
(
K
N
N
)
,
pe
r
f
or
m
a
n
c
e
e
v
a
l
ua
t
i
on
i
s
c
a
r
r
i
e
d
out
.
T
he
e
xpe
r
i
m
e
nt
s
de
m
ons
t
r
a
t
e
d
t
ha
t
c
om
bi
ni
ng
C
F
A
w
i
t
h
S
V
M
c
l
a
s
s
i
f
i
e
r
s
out
pe
r
f
or
m
e
d
ot
he
r
c
o
m
bi
na
t
i
ons
,
pa
r
t
i
c
ul
a
r
l
y
e
xc
e
l
l
i
ng
i
n
pr
e
c
i
s
i
on.
T
he
pa
pe
r
hi
g
hl
i
g
ht
e
d
t
he
c
ha
l
l
e
ng
e
of
di
m
e
ns
i
on
a
l
i
t
y
r
e
duc
t
i
on
i
n
c
l
a
s
s
i
f
i
c
a
t
i
on
pr
oc
e
s
s
e
s
,
pa
r
t
i
c
ul
a
r
l
y
w
he
n
de
a
l
i
ng
w
i
t
h
da
t
a
s
e
t
s
c
ont
a
i
ni
ng
num
e
r
ous
f
e
a
t
ur
e
s
,
s
o
m
e
of
w
hi
c
h
m
a
y
be
unr
e
l
i
a
bl
e
.
H
o
w
e
v
e
r
,
i
n
t
he
A
r
a
bi
c
l
a
n
g
ua
g
e
,
t
he
a
ppl
i
c
a
t
i
on
of
m
e
t
a
-
he
ur
i
s
t
i
c
a
l
g
or
i
t
hm
s
f
o
r
f
e
a
t
ur
e
e
xt
r
a
c
t
i
on i
s
l
i
m
i
t
e
d due
t
o t
he
l
a
ng
ua
ge
'
s
r
ul
e
s
a
nd r
i
c
h
m
or
phol
o
g
y
.
I
s
m
a
i
l
e
t
al
.
[
19
]
pr
e
s
e
nt
e
d
t
he
G
r
e
y
W
ol
f
O
pt
i
m
i
z
e
r
,
a
g
r
ound
-
br
e
a
ki
n
g
bi
o
-
i
ns
pi
r
e
d
opt
i
m
i
s
a
t
i
on
a
l
g
or
i
t
h
m
t
ha
t
m
i
m
i
c
s
w
o
l
v
e
s
'
na
t
ur
a
l
hunt
i
ng
t
e
c
hni
que
s
,
a
l
ong
w
i
t
h
t
he
K
N
N
c
l
a
s
s
i
f
i
e
r
,
t
o
c
r
e
a
t
e
a
n
i
nt
e
l
l
i
g
e
nt
f
e
a
t
ur
e
s
e
l
e
c
t
i
on
a
ppr
oa
c
h.
U
s
i
n
g
t
hr
e
e
di
f
f
e
r
e
nt
da
t
a
ba
s
e
s
t
he
A
r
a
bi
c
E
m
i
r
a
t
i
-
a
c
c
e
nt
e
d
s
pe
e
c
h
da
t
a
ba
s
e
,
R
A
V
D
E
S
S
a
nd
t
he
s
ur
r
e
y
a
udi
o
-
v
i
s
ua
l
e
xpr
e
s
s
e
d
e
m
ot
i
on
da
t
a
s
e
t
(
S
A
V
E
E
)
e
m
o
t
i
on
c
l
a
s
s
i
f
i
c
a
t
i
on
t
a
s
ks
a
r
e
us
e
d
t
o
de
m
ons
t
r
a
t
e
t
he
e
f
f
e
c
t
i
v
e
ne
s
s
of
t
hi
s
m
e
t
hod.
B
ot
h
c
om
bi
ne
d
a
nd
s
i
ng
l
e
f
e
a
t
ur
e
e
xt
r
a
c
t
i
on
m
e
t
hodol
og
i
e
s
a
r
e
us
e
d
a
c
r
os
s
t
he
s
e
da
t
a
s
e
t
s
.
I
n
s
pe
e
c
h
e
m
ot
i
on
r
e
c
o
g
ni
t
i
on
s
y
s
t
e
m
s
,
t
he
ne
w
G
W
O
-
K
N
N
t
e
c
hni
que
out
pe
r
f
or
m
e
d
m
or
e
e
s
t
a
bl
i
s
he
d a
ppr
oa
c
h
e
s
i
nc
l
udi
ng
t
he
ba
t
a
l
g
or
i
t
h
m
(
B
A
T
)
, c
u
c
koo s
e
a
r
c
h (
C
S
)
,
w
hi
t
e
s
ha
r
k opt
i
m
i
z
e
r
(
W
S
H
)
,
a
nd a
r
i
t
h
m
e
t
i
c
opt
i
m
i
s
a
t
i
on a
l
g
or
i
t
h
m
(
A
O
A
)
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2502
-
4752
I
ndone
s
i
a
n J
E
l
e
c
E
n
g
&
C
om
p S
c
i
,
V
ol
.
40
, N
o.
1
,
O
c
t
obe
r
20
25
:
4
90
-
498
492
E
l
a
ng
o
v
a
n
e
t
al
.
[
2
0
]
pr
opos
e
d
f
e
a
t
ur
e
s
e
l
e
c
t
i
on
(
F
S
)
a
nd
c
a
t
e
g
or
i
z
a
t
i
on
m
e
t
hods
t
o
de
v
e
l
o
p
a
n
e
f
f
i
c
i
e
nt
S
A
a
ppr
oa
c
h
f
or
i
nt
e
r
ne
t
r
e
v
i
e
w
s
.
I
n
pa
r
t
i
c
ul
a
r
,
f
e
a
t
ur
e
e
xt
r
a
c
t
i
on
f
r
o
m
w
e
b
-
ba
s
e
d
e
v
a
l
ua
t
i
ons
i
s
done
us
i
ng
t
he
F
i
r
e
F
l
y
(
F
F
)
a
nd l
e
v
y
f
l
i
g
ht
s
(
F
F
L
)
a
l
g
or
i
t
h
m
s
,
w
hi
l
e
e
m
ot
i
on c
a
t
e
g
or
i
z
a
t
i
on i
s
done
us
i
ng
t
he
M
ul
t
i
l
a
y
e
r
P
e
r
c
e
pt
r
on
(
M
L
P
)
f
r
a
m
e
w
or
k.
T
he
s
t
udy
of
t
he
F
F
-
M
L
P
m
ode
l
on
a
c
o
m
m
on
D
V
D
da
t
a
ba
s
e
s
how
e
d
i
t
s
e
f
f
e
c
t
i
v
e
ne
s
s
.
W
i
t
h
a
c
c
ur
a
c
y
of
97.97%
,
F
-
s
c
or
e
of
98.75
t
he
f
i
ndi
ng
s
s
how
e
d
r
e
m
a
r
ka
bl
e
pe
r
f
or
m
a
nc
e
g
a
i
ns
.
T
hi
s
pa
pe
r
f
oc
us
e
d on
s
e
nt
i
m
e
nt
a
na
l
y
s
i
s
(
S
A
)
a
ppl
i
e
d t
o w
e
b
-
ba
s
e
d c
u
s
t
om
e
r
r
e
v
i
e
w
s
.
T
he
pe
r
f
o
r
m
a
n
c
e
of
di
f
f
e
r
e
nt
A
I
a
l
g
or
i
t
h
m
s
i
n
t
e
xt
-
ba
s
e
d
e
m
ot
i
on
i
de
nt
i
f
i
c
a
t
i
on
f
or
c
onv
e
r
s
a
t
i
ona
l
a
ge
nt
s
w
a
s
i
n
v
e
s
t
i
g
a
t
e
d b
y
K
us
a
l
e
t
al
.
[2
1
]
i
n t
he
i
r
s
t
udy
.
T
he
y
t
e
s
t
e
d B
E
R
T
a
nd ot
he
r
m
o
de
l
s
i
n i
de
nt
i
f
y
i
ng
e
m
ot
i
ons
f
r
o
m
t
e
xt
ua
l
i
nput
.
T
he
r
e
s
e
a
r
c
h
e
r
s
us
e
d
m
e
t
hods
i
nc
l
udi
ng
a
dj
us
t
i
ng
B
E
R
T
on
da
t
a
s
e
t
s
l
a
be
l
l
e
d
w
i
t
h
e
m
ot
i
ons
a
nd
a
s
s
e
s
s
i
ng
i
t
s
c
l
a
s
s
i
f
i
c
a
t
i
on
pr
e
c
i
s
i
on
a
c
r
os
s
v
a
r
i
ous
e
m
ot
i
ona
l
c
a
t
e
g
or
i
e
s
.
T
he
out
c
o
m
e
s
s
how
e
d
t
ha
t
B
E
R
T
pe
r
f
or
m
e
d
c
o
m
p
e
t
i
t
i
v
e
l
y
i
n
e
m
ot
i
on
i
de
nt
i
f
i
c
a
t
i
on
t
a
s
ks
w
he
n
c
om
p
a
r
e
d
t
o
c
onv
e
nt
i
ona
l
m
a
c
hi
ne
l
e
a
r
ni
ng
t
e
c
hni
que
s
b
y
ut
i
l
i
z
i
ng
i
t
s
c
ont
e
xt
ua
l
unde
r
s
t
a
ndi
ng
of
l
a
ng
ua
g
e
.
H
o
w
e
v
e
r
,
i
n
or
de
r
t
o
a
c
hi
e
v
e
t
he
b
e
s
t
f
i
ne
-
t
uni
ng
,
s
i
g
ni
f
i
c
a
nt
c
o
m
put
a
t
i
ona
l
r
e
s
our
c
e
s
a
nd
a
hu
g
e
v
ol
u
m
e
of
a
nnot
a
t
e
d
da
t
a
a
r
e
r
e
qui
r
e
d.
T
he
r
e
s
e
a
r
c
h
r
e
v
i
e
w
e
d
s
how
s
t
ha
t
m
a
n
y
di
f
f
e
r
e
nt
m
e
t
hods
ha
v
e
be
e
n
us
e
d
t
o
c
hoos
e
t
he
be
s
t
f
e
a
t
ur
e
s
a
nd
c
l
a
s
s
i
f
y
da
t
a
i
n
v
a
r
i
ous
da
t
a
s
e
t
s
.
S
om
e
ne
w
m
e
t
hods
l
i
ke
t
he
R
e
t
ur
n
-
c
os
t
-
b
a
s
e
d
bi
na
r
y
F
F
A
,
m
odi
f
i
e
d
f
i
r
e
f
l
y
a
l
g
or
i
t
h
m
, c
h
a
ot
i
c
f
i
r
e
f
l
y
a
l
g
or
i
t
h
m
, a
nd
g
r
e
y
w
ol
f
opt
i
m
i
z
e
r
ha
v
e
be
e
n us
e
d. R
e
v
i
e
w
s
t
udi
e
s
ha
v
e
a
l
s
o
l
ooke
d
a
t
ho
w
t
he
s
e
m
e
t
hods
c
a
n
i
m
pr
o
v
e
t
e
xt
c
l
a
s
s
i
f
i
c
a
t
i
on
us
i
ng
s
w
a
r
m
i
nt
e
l
l
i
g
e
nc
e
.
S
w
a
r
m
i
nt
e
l
l
i
g
e
nc
e
a
l
g
or
i
t
h
m
pr
o
v
i
de
g
ood
r
e
s
ul
t
s
i
n
t
a
s
ks
l
i
ke
s
e
nt
i
m
e
nt
a
na
l
y
s
i
s
,
f
e
a
t
ur
e
s
e
l
e
c
t
i
on
i
n
A
r
a
bi
c
l
a
ng
ua
g
e
,
r
e
c
og
ni
z
i
n
g
a
nd
c
l
a
s
s
i
f
y
i
n
g
e
m
ot
i
ons
f
r
o
m
t
e
xt
.
T
hi
s
s
how
s
t
ha
t
u
s
i
ng
opt
i
m
i
z
a
t
i
on
t
e
c
hni
que
s
c
a
n
r
e
a
l
l
y
he
l
p i
n
m
a
n
y
di
f
f
e
r
e
nt
s
i
t
ua
t
i
ons
.
3.
M
E
T
H
O
D
F
i
g
ur
e
1
out
l
i
ne
s
a
s
y
s
t
e
m
a
t
i
c
pr
oc
e
s
s
f
or
e
m
ot
i
on
r
e
c
og
ni
t
i
on
f
r
o
m
t
e
xt
us
i
ng
a
m
odi
f
i
e
d
f
i
r
e
f
l
y
a
l
g
or
i
t
h
m
(
M
F
A
)
.
T
he
pr
oc
e
s
s
be
g
i
ns
by
l
oa
di
n
g
t
he
t
e
xt
e
m
ot
i
on
da
t
a
,
w
hi
c
h
c
ont
a
i
ns
l
a
be
l
l
e
d
e
m
ot
i
ons
.
T
hi
s
da
t
a
i
s
t
he
n
a
r
r
a
nge
d
a
nd
pr
e
pa
r
e
d
f
o
r
pr
oc
e
s
s
i
ng
,
i
n
v
ol
v
i
n
g
s
t
e
ps
s
uc
h
a
s
t
e
xt
c
l
e
a
ni
ng
,
nor
m
a
l
i
z
a
t
i
on,
a
nd f
or
m
a
t
t
i
ng
.
3.1
.
Wor
k
f
l
ow
d
i
agr
am
A
s
s
ho
w
n
i
n
F
i
g
ur
e
1,
a
n
obj
e
c
t
i
v
e
f
unc
t
i
on
i
s
de
f
i
ne
d
,
t
y
pi
c
a
l
l
y
a
i
m
e
d
a
t
m
i
ni
m
i
z
i
ng
t
he
m
e
a
n
a
bs
ol
ut
e
e
r
r
or
(
M
A
E
)
.
E
a
c
h
f
e
a
t
ur
e
of
t
he
t
e
xt
da
t
a
i
s
c
ons
i
de
r
e
d
a
s
one
f
i
r
e
f
l
y
,
a
nd
i
ni
t
i
a
l
pos
i
t
i
ons
a
nd
pr
ope
r
t
i
e
s
of
t
he
f
i
r
e
f
l
i
e
s
a
r
e
s
e
t
up.
T
h
e
a
l
g
or
i
t
h
m
i
t
e
r
a
t
e
s
o
v
e
r
a
pr
e
de
f
i
ne
d nu
m
be
r
o
f
g
e
ne
r
a
t
i
ons
or
unt
i
l
a
c
on
v
e
r
ge
nc
e
c
r
i
t
e
r
i
on
i
s
m
e
t
,
upda
t
i
ng
t
he
pos
i
t
i
ons
of
t
he
f
i
r
e
f
l
i
e
s
ba
s
e
d
on
t
he
i
r
a
t
t
r
a
c
t
i
v
e
ne
s
s
a
nd
t
he
di
s
t
a
nc
e
t
o
ot
he
r
f
i
r
e
f
l
i
e
s
us
i
ng
a
s
pe
c
i
f
i
c
f
or
m
ul
a
.
T
he
da
t
a
i
s
di
v
i
de
d
i
nt
o
t
r
a
i
ni
ng
a
nd
t
e
s
t
i
ng
s
e
t
s
t
o
m
e
a
s
ur
e
t
he
e
f
f
e
c
t
i
v
e
n
e
s
s
of
t
he
m
e
t
hod.
T
he
s
e
l
e
c
t
i
on
pr
oba
bi
l
i
t
y
of
m
o
r
e
s
uc
c
e
s
s
f
ul
f
i
r
e
f
l
i
e
s
i
nc
r
e
a
s
e
s
,
e
ns
ur
i
ng
t
ha
t
t
he
a
l
g
or
i
t
h
m
f
oc
us
e
s
on pr
o
m
i
s
i
n
g
s
ol
ut
i
ons
.
S
t
a
r
t
L
o
a
d
T
e
x
t
E
m
o
t
i
o
n
D
a
t
a
A
r
r
a
n
g
e
D
a
t
a
o
f
P
r
o
c
e
s
s
P
e
r
f
o
r
m
S
t
o
p
w
o
r
d
R
e
m
o
v
a
l
S
t
e
m
m
i
n
g
P
r
e
p
r
o
c
e
s
s
i
n
g
F
o
r
E
a
c
h
C
a
t
e
g
o
r
y
i
n
E
m
o
t
i
o
n
L
i
s
t
E
x
t
r
a
c
t
T
F
-
I
D
F
O
b
j
-
F
u
n
O
F
(
A
)
,
M
i
n
(
M
A
E
)
I
n
i
t
i
a
t
e
F
i
r
e
f
l
y
C
o
n
s
i
d
e
r
E
a
c
h
F
e
a
t
u
r
e
a
s
1
F
i
r
e
f
l
y
F
o
r
M
a
x
-
G
e
n
U
p
d
a
t
e
A
I
o
f
E
a
c
h
F
i
r
e
f
l
y
U
s
i
n
g
E
q
(
8
)
If
(
if
j
)
>
(
if
i
)
S
e
l
e
c
t
i
o
n
P
r
o
b
a
b
i
l
i
t
y
o
f
j
I
n
c
r
e
a
s
e
s
S
e
l
e
c
t
i
o
n
P
r
o
b
a
b
i
l
i
t
y
o
f
i
I
n
c
r
e
a
s
e
s
Y
e
s
No
S
p
l
i
t
D
a
t
a
i
n
t
o
T
r
a
i
n
a
n
d
T
e
s
t
S
t
o
p
T
r
a
i
n
M
L
E
v
a
l
u
a
t
e
Q
u
a
n
t
i
t
a
t
i
v
e
P
a
r
a
m
e
t
e
r
s
1
2
3
4
5
6
(
a
)
6
(
b
)
7
8
(
a
)
8
(
b
)
9
10
11
F
i
g
ur
e
1. W
or
kf
l
ow
di
a
g
r
a
m
f
or
e
m
ot
i
on r
e
c
o
g
ni
t
i
on
Evaluation Warning : The document was created with Spire.PDF for Python.
I
ndone
s
i
a
n J
E
l
e
c
E
n
g
&
C
om
p S
c
i
I
S
S
N
:
2502
-
4752
A
n i
m
pr
ov
e
d
c
onv
e
r
s
at
i
on e
m
ot
i
o
n
d
e
t
e
c
t
i
on
us
i
ng hy
b
r
i
d
f
-
nn c
l
as
s
i
f
i
e
r
(
A
b
hi
s
he
k
A
. V
i
c
har
e
)
493
S
ubs
e
que
nt
l
y
,
a
m
a
c
hi
ne
l
e
a
r
ni
ng
m
ode
l
,
s
uc
h
a
s
a
n
M
L
P
i
s
t
r
a
i
ne
d
us
i
ng
t
he
opt
i
m
i
z
e
d
f
e
a
t
ur
e
s
.
T
hi
s
m
e
t
hod
i
s
t
he
n
e
v
a
l
ua
t
e
d
on
t
he
ba
s
i
s
of
qua
nt
i
t
a
t
i
v
e
pa
r
a
m
e
t
e
r
s
l
i
ke
a
c
c
ur
a
c
y
,
pr
e
c
i
s
i
on,
r
e
c
a
l
l
,
F
-
m
e
a
s
ur
e
,
a
nd
c
o
m
put
a
t
i
ona
l
e
f
f
i
c
i
e
nc
y
t
o
a
s
s
e
s
s
i
t
s
e
f
f
e
c
t
i
v
e
ne
s
s
.
T
hi
s
c
o
m
pr
e
h
e
ns
i
v
e
w
or
kf
l
o
w
de
m
ons
t
r
a
t
e
s
t
he
us
e
of
M
F
A
t
o
i
nc
r
e
a
s
e
t
he
r
e
s
ul
t
s
a
nd
e
f
f
e
c
t
i
v
e
ns
s
of
e
m
ot
i
on
de
t
e
c
t
i
on
i
n N
L
P
t
a
s
ks
.
3.2
.
D
at
as
e
t
u
s
e
d
i
n
r
e
s
e
ar
c
h
T
he
I
S
E
A
R
da
t
a
s
e
t
,
know
n
a
s
t
he
I
nt
e
r
na
t
i
ona
l
S
ur
v
e
y
on
E
m
ot
i
on
A
nt
e
c
e
de
n
t
s
a
nd
R
e
a
c
t
i
ons
da
t
a
s
e
t
,
i
s
w
i
de
l
y
r
e
c
o
g
ni
z
e
d
i
n
r
e
s
e
a
r
c
h
f
or
i
t
s
c
om
pr
e
he
ns
i
v
e
c
ol
l
e
c
t
i
on
of
hum
a
n
e
m
ot
i
ona
l
r
e
s
pons
e
s
a
c
r
os
s
v
a
r
i
ous
s
c
e
na
r
i
os
. I
t
c
ont
a
i
ns
a
s
t
r
uc
t
ur
e
d c
o
m
pi
l
a
t
i
on of
e
m
ot
i
ona
l
e
xpe
r
i
e
nc
e
s
c
a
t
e
g
or
i
z
e
d i
nt
o s
e
v
e
n
pr
i
m
a
r
y
e
m
ot
i
ons
:
j
oy
,
f
e
a
r
,
a
n
g
e
r
,
s
a
dne
s
s
,
di
s
g
us
t
,
s
ha
m
e
,
a
nd
g
ui
l
t
.
E
a
c
h
e
m
ot
i
ona
l
r
e
s
p
ons
e
i
s
a
s
s
oc
i
a
t
e
d
w
i
t
h
a
de
t
a
i
l
e
d
de
s
c
r
i
pt
i
on
of
t
he
t
r
i
gg
e
r
i
ng
e
v
e
nt
,
pr
o
v
i
di
ng
r
e
s
e
a
r
c
he
r
s
w
i
t
h
v
a
l
ua
bl
e
i
ns
i
g
ht
s
i
nt
o
t
he
c
ont
e
xt
a
nd
f
a
c
t
or
s
i
nf
l
ue
nc
i
n
g
e
m
ot
i
ona
l
r
e
a
c
t
i
ons
.
D
ue
t
o
i
t
s
r
i
c
hne
s
s
a
nd
di
v
e
r
s
i
t
y
,
t
he
I
S
E
A
R
da
t
a
s
e
t
s
e
r
v
e
s
a
s
a
r
obus
t
be
nc
h
m
a
r
k
f
or
e
v
a
l
u
a
t
i
ng
e
m
ot
i
on
r
e
c
o
g
ni
t
i
on
s
y
s
t
e
m
s
a
nd
ha
s
b
e
e
n
ut
i
l
i
z
e
d
i
n
nu
m
e
r
ous
s
t
udi
e
s
t
o
e
xpl
or
e
t
he
e
f
f
i
c
i
e
nc
y
of
di
f
f
e
r
e
nt
c
o
m
put
a
t
i
ona
l
a
ppr
oa
c
he
s
i
n
a
na
l
y
z
i
n
g
a
nd
unde
r
s
t
a
ndi
ng
hum
a
n e
m
ot
i
ons
f
r
o
m
t
e
xt
ua
l
da
t
a
[
22
]
,
[
23
].
3.3
.
T
e
xt
p
r
e
p
r
oc
e
s
s
i
n
g
T
he
a
r
c
hi
t
e
c
t
ur
e
c
a
n
be
i
l
l
us
t
r
a
t
e
d
us
i
ng
A
l
g
or
i
t
h
m
1
f
or
pr
e
-
p
r
oc
e
s
s
i
ng
a
nd
s
t
e
m
m
i
n
g
a
nd
t
he
o
v
e
r
a
l
l
w
or
k a
r
c
hi
t
e
c
t
ur
e
c
a
n be
i
l
l
us
t
r
a
t
e
d us
i
ng
w
or
k f
l
o
w
di
a
g
r
a
m
s
how
n i
n F
i
g
ur
e
1
.
A
l
g
or
i
t
h
m
1
.
T
e
xt
p
re
-
pr
oc
e
s
s
i
ng
w
i
t
h
s
t
opw
or
d r
e
m
o
v
a
l
a
nd s
t
e
m
m
i
n
g
1.
F
u
n
c
t
i
o
n
\
P
r
e
p
r
o
c
e
s
s
T
e
x
t
\
l
e
f
t
(
i
n
p
u
t
_{
t
e
x
t
}
\
r
i
g
h
t
)
:
2.
t
o
k
e
n
i
z
e
d
_{
t
e
x
t
}
\
l
o
n
g
l
e
f
t
a
r
r
o
w
T
o
k
e
n
i
z
e
(
i
n
p
u
t
_{
t
e
x
t
}
)
3.
stop
_{
w
o
r
d
s
}
\
g
e
t
s
L
o
a
d
S
t
o
p
w
o
r
d
s
(
)
4.
f
i
l
t
e
r
e
d
_{
w
o
r
d
s
}
\
g
e
t
s
R
e
m
o
v
e
S
t
o
p
w
o
r
d
s
(
t
o
k
e
n
i
z
d
_{
t
e
x
t
}
,
\
s
t
o
p
_{
w
o
r
d
s
}
)
5.
s
t
e
m
m
e
d
_{
w
o
r
d
s
}
\
g
e
t
s
A
p
p
l
y
S
t
e
m
m
i
n
g
(
f
i
l
t
e
r
e
d
_{
w
o
r
d
s
}
)
6.
c
l
e
a
n
_{
t
e
x
t
}
\
g
e
t
s
J
o
i
n
W
o
r
d
s
(
s
t
e
m
m
e
d
_{
w
o
r
d
s
}
)
7.
r
e
t
u
r
n
\
c
l
e
a
n
_{
t
e
x
t
}
\
8.
F
u
n
c
t
i
o
n
\
T
o
k
e
n
i
z
e
\
l
e
f
t
(
t
x
t
\
r
i
g
h
t
)
:
9.
r
e
t
u
r
n
\
L
i
s
t
\
of
\
w
o
r
d
s
\
o
b
t
a
i
n
e
d
\
by
\
s
p
l
i
t
t
i
n
g
\
t
e
x
t
10.
F
u
n
c
t
i
o
n
\
L
o
a
d
S
t
o
p
w
o
r
d
s
(
)
:
11.
r
e
t
u
r
n
\
L
i
s
t
\
of
\
c
o
m
m
o
n
\
s
t
o
p
w
o
r
d
s
12.
F
u
n
c
t
i
o
n
\
R
e
m
o
v
e
S
t
o
p
w
o
r
d
s
\
l
e
f
t
(
t
e
x
t
,
\
S
t
o
p
_{
w
o
r
d
s
}
\
r
i
g
h
t
)
13.
f
i
l
t
e
r
e
d
_{
t
e
x
t
}
\
g
e
t
s
[
]
14.
for
\
w
o
r
d
\
in
\
t
e
x
t
\
do
15.
\
if
\
w
o
r
d
\
n
o
t
\
in
\
s
t
o
p
_{
w
o
r
d
s
}
\
t
h
e
n
16.
f
i
l
t
e
r
e
d
_{
t
e
x
t
}
.
a
p
p
e
n
d
(
w
o
r
d
)
17.
end
\
if
18.
end
\
f
o
r
19.
r
e
t
u
r
n
\
f
i
l
t
e
r
e
d
_{
t
e
x
t
}
20.
F
u
n
c
t
i
o
n
\
A
p
p
l
y
S
t
e
m
m
i
n
g
\
l
e
f
t
(
t
e
x
t
\
r
i
g
h
t
)
:
21.
s
t
e
m
m
e
d
_{
t
e
x
t
}
\
l
o
n
g
l
e
f
t
a
r
r
o
w
[
]
22.
for
\
w
o
r
d
\
in
\
t
e
x
t
\
do
23.
S
t
e
m
m
e
d
_{
w
o
r
d
}
\
\
g
e
t
s
A
p
p
l
y
\
s
t
e
m
m
i
n
g
\
a
l
g
o
r
i
t
h
m
\
to
\
w
o
r
d
24.
s
t
e
m
m
e
d
_{
t
e
x
t
}
.
a
p
p
w
n
d
\
l
e
f
t
(
s
t
e
m
m
e
d
_{
w
o
r
d
}
\
r
i
g
h
t
)
25.
end
\
f
o
r
26.
r
e
t
u
r
n
\
S
t
e
m
m
e
d
_{
t
e
x
t
}
27.
F
u
n
c
t
i
o
n
\
J
o
i
n
W
o
r
d
s
\
l
e
f
t
(
w
o
r
d
s
\
r
i
g
h
t
)
:
28.
r
e
t
u
r
n
\
C
o
n
c
a
t
e
n
a
t
e
\
w
o
r
d
s
\
i
n
t
o
\
s
i
n
g
l
e
\
s
t
r
i
n
g
T
e
r
m
f
r
e
q
ue
nc
y
(
T
F
)
:
A
t
e
r
m
's
(
w
or
d
'
s
)
T
F
v
a
l
ue
i
ndi
c
a
t
e
s
how
f
r
e
que
nt
l
y
i
t
a
ppe
a
r
s
i
n
a
g
i
v
e
n
doc
u
m
e
nt
.
I
t
m
e
a
s
ur
e
s
a
t
e
r
m
's r
e
l
a
t
i
v
e
w
e
i
g
ht
i
ns
i
de
a
g
i
v
e
n doc
u
m
e
nt
.
T
he
T
F
f
or
m
ul
a
i
s
:
(
,
)
=
(
1)
W
he
r
e
t
i
s
t
he
c
ur
r
e
nt
t
e
r
m
, O
t
i
s
t
he
oc
c
ur
r
e
nc
e
of
t
he
t
e
r
m
s
i
n a
t
ot
a
l
c
ount
of
w
or
ds
C
.
I
n
v
e
r
s
e
D
oc
u
m
e
nt
F
r
e
que
nc
y
(
I
D
F
)
:
I
D
F
a
na
l
y
s
e
s
a
t
e
r
m
'
s
g
l
oba
l
s
i
g
ni
f
i
c
a
nc
e
b
y
a
s
s
e
s
s
i
ng
i
t
s
r
a
r
i
t
y
t
hr
oug
hout
t
he
w
hol
e
c
or
pus
o
f
doc
um
e
nt
s
.
I
t
g
i
v
e
s
l
e
s
s
w
e
i
g
ht
t
o
c
om
m
on
t
e
r
m
s
a
nd
m
or
e
w
e
i
g
ht
t
o
r
a
r
e
t
e
r
m
s
.
T
he
f
or
m
ul
a
f
or
I
D
F
i
s
:
(
)
=
l
o
g
(
)
(
2)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2502
-
4752
I
ndone
s
i
a
n J
E
l
e
c
E
n
g
&
C
om
p S
c
i
,
V
ol
.
40
, N
o.
1
,
O
c
t
obe
r
20
25
:
4
90
-
498
494
W
he
r
e
de
not
e
s
t
he
t
ot
a
l
num
be
r
of
doc
u
m
e
nt
s
c
ont
a
i
ni
ng
t
a
nd
de
not
e
s
t
he
t
ot
a
l
num
be
r
of
doc
u
m
e
nt
s
i
n t
he
l
i
s
t
.T
he
o
v
e
r
a
l
l
T
F
-
I
D
F
c
a
n be
v
i
c
t
or
i
e
d a
s
f
ol
l
o
w
s
:
−
(
,
)
=
(
,
)
∗
(
)
(
3)
I
n
f
e
a
t
ur
e
s
e
l
e
c
t
i
on,
i
t
he
l
ps
i
de
nt
i
f
y
t
he
m
os
t
r
e
l
e
v
a
nt
t
e
r
m
s
(
f
e
a
t
ur
e
s
)
f
or
e
m
ot
i
on
a
na
l
y
s
i
s
.
F
i
r
e
f
l
y
a
l
g
or
i
t
h
m
s
c
ons
i
de
r
t
he
T
F
-
I
D
F
v
a
l
ue
s
a
s
t
he
a
t
t
r
a
c
t
i
v
e
n
e
s
s
of
f
e
a
t
ur
e
s
a
nd i
t
e
r
a
t
i
v
e
l
y
r
e
f
i
n
e
t
he
f
e
a
t
ur
e
s
e
t
t
o
m
a
xi
m
i
z
e
t
he
c
l
a
s
s
i
f
i
c
a
t
i
on
of
e
m
ot
i
ons
.
T
he
s
e
t
e
c
hni
que
s
ha
v
e
e
v
ol
v
e
d
t
o
ha
ndl
e
l
a
r
ge
a
nd
hi
g
h
-
di
m
e
ns
i
ona
l
t
e
xt
da
t
a
s
e
t
s
e
f
f
i
c
i
e
nt
l
y
,
e
ns
ur
i
ng
t
ha
t
t
he
m
os
t
i
nf
or
m
a
t
i
v
e
f
e
a
t
ur
e
s
a
r
e
r
e
t
a
i
ne
d
f
or
a
c
c
ur
a
t
e
e
m
ot
i
on
r
e
c
o
g
ni
t
i
on.
B
y
s
i
m
ul
a
t
i
ng
t
he
s
w
a
r
m
i
nt
e
l
l
i
g
e
nc
e
of
f
i
r
e
f
l
i
e
s
,
t
he
F
i
r
e
f
l
y
a
l
g
or
i
t
hm
a
nd
ot
he
r
S
I
t
e
c
hni
que
s
ha
v
e
be
c
o
m
e
c
r
u
c
i
a
l
t
ool
i
n
t
he
a
r
e
a
of
t
e
xt
-
ba
s
e
d
e
m
ot
i
on
a
na
l
y
s
i
s
,
e
n
a
bl
i
ng
i
m
pr
o
v
e
d
m
ode
l
pe
r
f
or
m
a
nc
e
a
nd
m
or
e
d
e
t
a
i
l
e
d e
m
ot
i
on unde
r
s
t
a
ndi
ng
.
3.4
.
M
L
P
c
l
as
s
i
f
i
er
T
a
bl
e
1
pr
e
s
e
nt
s
h
y
pe
r
p
a
r
a
m
e
t
e
r
s
us
e
d
i
n e
xpe
r
i
m
e
nt
a
t
i
on f
or
M
L
P
a
l
g
or
i
t
h
m
.
T
ot
a
l
10 n
e
ur
ons
a
r
e
us
e
d i
n hi
dde
n
l
a
y
e
r
. L
e
a
r
ni
n
g
r
a
t
e
a
ppl
i
e
d f
or
m
ode
l
i
s
0.01.
T
a
bl
e
1. H
y
pe
r
pa
r
a
m
e
t
e
r
s
of
M
L
P
H
ype
r
pa
r
a
m
e
t
e
r
D
e
s
c
r
i
pt
i
o
n
V
a
l
ue
s
H
i
dde
n
L
a
ye
r
S
i
z
e
T
ot
a
l
n
e
ur
o
n
s
i
n
t
h
e
h
i
dde
n
l
a
ye
r
.
10
A
c
t
i
va
t
i
o
n
F
u
n
c
t
i
o
n
A
c
t
i
va
t
i
o
n
f
u
n
c
t
i
o
n
i
n
t
h
e
h
i
dde
n
l
a
ye
r
.
S
i
gm
oi
d
L
e
a
r
n
i
ng
R
a
t
e
T
h
e
r
a
t
e
a
t
w
h
i
c
h
t
h
e
m
o
de
l
l
e
a
r
n
s
.
0.
01
N
um
be
r
of
E
poc
h
s
T
h
e
n
u
m
be
r
of
t
r
a
i
n
i
ng
e
poc
h
s
.
100
4.
R
E
S
U
S
L
T
S
A
N
D
D
I
S
C
U
S
S
I
O
N
I
ni
t
i
a
l
l
y
,
a
c
om
pa
r
a
t
i
v
e
a
na
l
y
s
i
s
i
s
pe
r
f
or
m
e
d
us
i
ng
I
S
E
A
R
da
t
a
s
e
t
f
or
pe
r
f
or
m
a
nc
e
e
v
a
l
ua
t
i
on
f
a
c
t
or
s
na
m
e
l
y
,
pr
e
c
i
s
i
o
n,
r
e
c
a
l
l
,
f
-
m
e
a
s
ur
e
a
nd
a
c
c
ur
a
c
y
us
i
ng
num
be
r
of
s
a
m
pl
e
s
r
a
ng
i
ng
f
r
om
1
,
500
t
o
7
,
50
0.
T
he
s
t
u
dy
a
s
s
e
s
s
e
d how
pe
r
f
or
m
a
nc
e
t
r
e
nd
s
c
ha
ng
e
a
s
t
he
n
um
be
r
of
s
a
m
pl
e
s
i
nc
r
e
a
s
e
s
. R
e
s
u
l
t
s
a
l
s
o
g
i
v
e
s
i
ns
i
g
ht
s
i
nt
o how
da
t
a
s
e
t
s
i
z
e
i
nf
l
ue
nc
e
s
c
l
a
s
s
i
f
i
c
a
t
i
on e
f
f
e
c
t
i
v
e
ne
s
s
.
4.1.
P
r
e
c
i
s
i
o
n
, r
e
c
al
l
, f
m
e
as
u
r
e
an
d
ac
c
u
r
ac
y
4.1.1.
P
r
e
c
i
s
i
on
T
he
pr
e
c
i
s
i
on
a
na
l
y
s
i
s
s
um
m
a
r
i
z
e
d
i
n
T
a
bl
e
2
s
how
s
t
ha
t
"
P
r
opos
e
d
F
i
r
e
f
l
y
+
M
L
P
"
m
ode
l
e
xhi
bi
t
s
a
n
e
xc
e
pt
i
ona
l
pr
e
c
i
s
i
on
s
c
or
e
of
0.9647,
i
ndi
c
a
t
i
ng
i
t
s
a
bi
l
i
t
y
t
o
m
a
ke
hi
g
hl
y
a
c
c
ur
a
t
e
po
s
i
t
i
v
e
pr
e
di
c
t
i
ons
.
O
t
he
r
m
ode
l
s
s
uc
h
a
s
"
M
L
P
O
nl
y
,
"
"
P
S
O
+
M
L
P
,"
a
nd
"
A
C
O
+
M
L
P
"
a
l
s
o
de
m
ons
t
r
a
t
e
hi
gh
pr
e
c
i
s
i
on
v
a
l
ue
s
a
r
ound
0.945,
s
ug
g
e
s
t
i
ng
t
he
i
r
c
a
pa
c
i
t
y
t
o
m
a
ke
a
c
c
ur
a
t
e
pos
i
t
i
v
e
pr
e
di
c
t
i
ons
.
"
P
S
O
+
K
N
N
"
m
a
i
nt
a
i
ns
a
c
o
m
pe
t
i
t
i
v
e
pr
e
c
i
s
i
on
v
a
l
ue
of
0.9455, i
m
pl
y
i
n
g
t
ha
t
i
t
a
l
s
o e
xc
e
l
s
i
n r
e
duc
i
ng
f
a
l
s
e
pos
i
t
i
v
e
pr
e
di
c
t
i
on
s
.
T
a
bl
e
2. P
e
r
f
or
m
a
nc
e
a
na
l
y
s
i
s
of
e
m
ot
i
on r
e
c
og
ni
t
i
o
n us
i
ng
pr
e
c
i
s
i
on
ov
e
r
I
S
E
A
R
T
ot
a
l
n
u
m
be
r
of
S
a
m
pl
e
s
P
r
e
c
i
s
i
o
n
p
r
opos
e
d
f
i
r
e
f
l
y+
M
L
P
P
r
e
c
i
s
i
o
n
M
L
P
O
n
l
y
P
r
e
c
i
s
i
o
n
P
S
O
+
M
L
P
P
r
e
c
i
s
i
o
n
P
S
O
+
K
N
N
P
r
e
c
i
s
i
o
n
A
C
O
+
M
L
P
1500
0.
96558256
0.
94176373
0.
94024276
0.
93837535
0.
94625284
2500
0.
96505334
0.
94933712
0.
93816156
0.
95086321
0.
94457948
3500
0.
96248461
0.
94582393
0.
94967532
0.
94935022
0.
94088099
4500
0.
95958759
0.
94725335
0.
94962093
0.
94818927
0.
94763471
5500
0.
96036966
0.
94791209
0.
94963876
0.
93654638
0.
94437086
6500
0.
96499099
0.
94107673
0.
94832995
0.
94595061
0.
93607492
7500
0.
96470851
0.
94248766
0.
94125998
0.
9491576
0.
93706165
4.1.
2
.
R
e
c
al
l
T
he
"
P
r
opos
e
d
F
i
r
e
f
l
y
+
M
L
P
"
m
ode
l
s
c
or
e
a
r
e
c
a
l
l
v
a
l
ue
of
0.9895,
w
hi
c
h
de
m
ons
t
r
a
t
e
s
i
t
s
e
xc
e
l
l
e
nt
a
bi
l
i
t
y
t
o pr
ope
r
l
y
i
de
nt
i
f
y
a
l
m
os
t
a
l
l
pos
i
t
i
v
e
c
a
s
e
s
i
n t
he
da
t
a
s
e
t
. M
ode
l
s
l
i
ke
"
M
L
P
O
nl
y
,
"
"
P
S
O
+
M
L
P
,"
a
nd
"
A
C
O
+
M
L
P
"
a
l
s
o
di
s
pl
a
y
hi
g
h
r
e
c
a
l
l
v
a
l
ue
s
,
i
n
di
c
a
t
i
ng
t
he
i
r
pr
of
i
c
i
e
nc
y
i
n
c
a
pt
ur
i
ng
pos
i
t
i
v
e
i
ns
t
a
nc
e
s
.
T
a
bl
e
3
s
ho
w
s
t
ha
t
"
P
S
O
+
K
N
N
,
"
w
hi
l
e
s
l
i
g
ht
l
y
l
ow
e
r
i
n
r
e
c
a
l
l
a
t
0.8359,
s
t
i
l
l
m
a
i
nt
a
i
ns
a
r
e
s
pe
c
t
a
bl
e
pe
r
f
or
m
a
nc
e
, c
o
r
r
e
c
t
l
y
i
de
nt
i
f
y
i
n
g
a
s
i
g
ni
f
i
c
a
nt
por
t
i
on of
pos
i
t
i
v
e
i
ns
t
a
nc
e
s
.
T
he
"
R
e
c
a
l
l
P
r
opos
e
d
F
i
r
e
f
l
y
+
M
L
P
"
a
l
g
or
i
t
h
m
ha
s
t
he
hi
g
he
s
t
a
v
e
r
a
g
e
r
e
c
a
l
l
,
a
r
ound
0.98943,
i
ndi
c
a
t
i
ng
i
t
s
c
om
pe
t
e
n
c
e
i
n pr
ope
r
l
y
de
t
e
c
t
i
ng
pos
i
t
i
v
e
e
v
e
nt
s
. T
he
s
e
d
a
t
a
i
ndi
c
a
t
e
c
ha
n
g
e
s
i
n t
he
a
l
g
or
i
t
hm
s
'
Evaluation Warning : The document was created with Spire.PDF for Python.
I
ndone
s
i
a
n J
E
l
e
c
E
n
g
&
C
om
p S
c
i
I
S
S
N
:
2502
-
4752
A
n i
m
pr
ov
e
d
c
onv
e
r
s
at
i
on e
m
ot
i
o
n
d
e
t
e
c
t
i
on
us
i
ng hy
b
r
i
d
f
-
nn c
l
as
s
i
f
i
e
r
(
A
b
hi
s
he
k
A
. V
i
c
har
e
)
495
c
a
pa
c
i
t
i
e
s
t
o a
c
c
ur
a
t
e
l
y
de
t
e
c
t
pos
i
t
i
v
e
c
a
s
e
s
a
c
r
os
s
v
a
r
y
i
n
g
s
a
m
pl
e
s
i
z
e
s
,
w
i
t
h t
he
"
R
e
c
a
l
l
P
r
opos
e
d F
i
r
e
f
l
y
+
M
L
P
"
m
e
t
hod c
ons
i
s
t
e
nt
l
y
di
s
pl
a
y
i
ng
g
r
e
a
t
e
r
pe
r
f
or
m
a
nc
e
.
T
a
bl
e
3. P
e
r
f
or
m
a
nc
e
a
na
l
y
s
i
s
of
e
m
ot
i
on r
e
c
og
ni
t
i
o
n us
i
ng
r
e
c
a
l
l
ov
e
r
I
S
E
A
R
T
ot
a
l
n
u
m
be
r
of
S
a
m
pl
e
s
R
e
c
a
l
l
P
r
opos
e
d
F
i
r
e
f
l
y
+
M
L
P
R
e
c
a
l
l
M
L
P
O
n
l
y
R
e
c
a
l
l
P
S
O
+
M
L
P
R
e
c
a
l
l
P
S
O
+
K
N
N
R
e
c
a
l
l
A
C
O
+
M
L
P
1500
0.
98925501
0.
98606272
0.
98435973
0.
80143541
0.
76499388
2500
0.
99011603
0.
98719842
0.
98594848
0.
87352582
0.
75691469
3500
0.
99047912
0.
98854061
0.
98850963
0.
90761389
0.
88868613
4500
0.
98949128
0.
98662112
0.
98753815
0.
83691087
0.
93160905
5500
0.
98944488
0.
98718242
0.
98784262
0.
83430421
0.
87395301
6500
0.
98974529
0.
98513424
0.
98713023
0.
77592068
0.
83966399
7500
0.
98953555
0.
98558719
0.
98550725
0.
82173745
0.
75697033
4.1.
3
.
F
-
m
e
as
u
r
e
F
-
m
e
a
s
ur
e
a
n
a
l
y
s
i
s
us
i
ng
I
S
E
A
R
da
t
a
s
e
t
g
i
v
e
n
i
n
T
a
bl
e
4
s
ho
w
s
t
ha
t
t
he
"
P
r
opos
e
d
F
i
r
e
f
l
y
+
M
L
P
"
m
od
e
l
a
t
t
a
i
ns
t
he
hi
g
h
e
s
t
F
-
m
e
a
s
ur
e
of
0.9769,
r
e
f
l
e
c
t
i
n
g
i
t
s
e
xc
e
pt
i
ona
l
ba
l
a
nc
e
be
t
w
e
e
n
pr
e
c
i
s
i
on
a
nd
r
e
c
a
l
l
.
"
M
L
P
O
nl
y
,
"
"
P
S
O
+
M
L
P
,"
a
nd
"
A
C
O
+
M
L
P
"
m
ode
l
s
a
l
s
o
d
e
m
ons
t
r
a
t
e
c
o
m
p
e
t
i
t
i
v
e
F
-
m
e
a
s
ur
e
v
a
l
ue
s
, i
ndi
c
a
t
i
ng
t
he
i
r
e
f
f
e
c
t
i
v
e
ne
s
s
i
n s
e
nt
i
m
e
nt
a
na
l
y
s
i
s
.
"
P
S
O
+
K
N
N
"
a
c
hi
e
v
e
s
a
r
e
s
pe
c
t
a
bl
e
F
-
m
e
a
s
ur
e
o
f
0.8869.
T
he
"
F
-
m
e
a
s
ur
e
P
r
opos
e
d
F
i
r
e
f
l
y
+
M
L
P
"
a
l
g
o
r
i
t
h
m
e
xhi
bi
t
s
t
he
hi
g
he
s
t
a
v
e
r
a
ge
F
-
m
e
a
s
ur
e
a
t
a
ppr
oxi
m
a
t
e
l
y
0.9
7647,
i
ndi
c
a
t
i
ng
i
t
s
e
f
f
e
c
t
i
v
e
n
e
s
s
i
n
a
c
hi
e
v
i
n
g
a
ba
l
a
nc
e
be
t
w
e
e
n
pr
e
c
i
s
i
on
a
nd
r
e
c
a
l
l
.
F
ol
l
ow
i
ng
c
l
os
e
l
y
i
s
t
he
"
F
-
m
e
a
s
ur
e
M
L
P
O
nl
y"
a
l
g
or
i
t
h
m
w
i
t
h
a
n
a
v
e
r
a
ge
F
-
m
e
a
s
ur
e
of
a
ppr
oxi
m
a
t
e
l
y
0.96558, s
how
c
a
s
i
ng
i
t
s
c
o
m
pe
t
i
t
i
v
e
pe
r
f
or
m
a
nc
e
.
T
a
bl
e
4. P
e
r
f
or
m
a
nc
e
a
na
l
y
s
i
s
of
e
m
ot
i
on r
e
c
og
ni
t
i
o
n u
s
i
ng
F
-
m
e
a
s
ur
e
o
v
e
r
I
S
E
A
R
T
ot
a
l
n
u
m
be
r
of
S
a
m
pl
e
s
F
m
e
a
s
ur
e
P
r
op
os
e
d
F
i
r
e
f
l
y
+
M
L
P
F
m
e
a
s
ur
e
M
L
P
O
n
l
y
F
m
e
a
s
ur
e
P
S
O
+
M
L
P
F
m
e
a
s
ur
e
P
S
O
+
K
N
N
F
m
e
a
s
ur
e
A
C
O
+
M
L
P
1500
0.
97727545
0.
96340426
0.
96179561
0.
86451613
0.
84602369
2500
0.
97742405
0.
96789766
0.
9614616
0.
91055532
0.
84039792
3500
0.
97628122
0.
96671061
0.
96870343
0.
92801303
0.
91403904
4500
0.
97431004
0.
96653653
0.
96820845
0.
88908167
0.
93955355
5500
0.
97469049
0.
96714878
0.
96836403
0.
88247376
0.
90779841
6500
0.
9772114
0.
96260163
0.
96734118
0.
85254066
0.
88525221
7500
0.
97696432
0.
96355571
0.
96287556
0.
88086343
0.
8374433
4.1.
4
.
A
c
c
u
r
ac
y
A
na
l
y
z
i
n
g
t
he
a
c
c
u
r
a
c
y t
r
e
nd a
c
r
os
s
t
he
v
a
r
i
ous
a
l
g
or
i
t
h
m
s
us
e
d on t
he
I
S
E
A
R
da
t
a
s
e
t
r
e
v
e
a
l
s
s
o
m
e
not
a
bl
e
pa
t
t
e
r
ns
.
T
he
"
A
c
c
ur
a
c
y
P
r
opos
e
d
F
i
r
e
f
l
y
+
M
L
P
"
a
l
g
or
i
t
h
m
c
ons
i
s
t
e
nt
l
y
a
c
hi
e
v
e
s
t
he
be
s
t
l
e
v
e
l
s
o
f
a
c
c
ur
a
c
y
,
w
i
t
h
a
s
t
e
a
dy
pe
r
f
or
m
a
nc
e
t
r
e
nd
a
c
r
os
s
v
a
r
i
ous
s
a
m
pl
e
s
i
z
e
s
a
s
s
how
n
i
n
T
a
bl
e
5
.
I
n
c
ont
r
a
s
t
,
t
he
"
A
c
c
ur
a
c
y
P
S
O
+
K
N
N
"
a
nd
"
A
c
c
ur
a
c
y
A
C
O
+
M
L
P
"
a
l
g
or
i
t
h
m
s
s
how
a
de
c
r
e
a
s
i
n
g
t
r
e
nd
i
n
a
c
c
ur
a
c
y
a
s
t
he
s
a
m
pl
e
s
i
z
e
i
nc
r
e
a
s
e
s
,
i
ndi
c
a
t
i
ng
pot
e
nt
i
a
l
c
ons
t
r
a
i
nt
s
or
i
ne
f
f
i
c
i
e
nc
i
e
s
i
n
ha
ndl
i
ng
l
a
r
g
e
r
da
t
a
s
e
t
s
.
W
he
n
w
e
a
na
l
y
z
e
t
he
a
v
e
r
a
ge
v
a
l
ue
us
i
ng
t
he
I
S
E
A
R
da
t
a
s
e
t
,
w
e
s
e
e
t
ha
t
t
he
"
A
c
c
u
r
a
c
y
P
r
opos
e
d
F
i
r
e
f
l
y
+
M
L
P
"
t
e
c
hni
que
ha
s
t
he
hi
g
he
s
t
a
v
e
r
a
g
e
a
c
c
u
r
a
c
y
,
a
r
ound
92.07%
.
T
he
"
A
c
c
ur
a
c
y
M
L
P
O
nl
y"
m
e
t
hod
f
ol
l
ow
s
c
l
os
e
l
y
be
hi
nd,
w
i
t
h
a
n
a
v
e
r
a
g
e
a
c
c
ur
a
c
y
of
a
ppr
oxi
m
a
t
e
l
y
89.95%
.
T
he
s
e
f
i
ndi
ng
s
i
ndi
c
a
t
e
t
ha
t
t
he
"
A
c
c
ur
a
c
y
P
r
opos
e
d
F
i
r
e
f
l
y
+
M
L
P
"
t
e
c
hni
que
s
ur
pa
s
s
e
s
o
t
he
r
s
on
a
v
e
r
a
g
e
i
n
t
e
r
m
s
of
a
c
c
ur
a
c
y
w
h
e
n
a
ppl
i
e
d t
o t
he
I
S
E
A
R
da
t
a
s
e
t
, de
m
ons
t
r
a
t
i
ng
i
t
s
us
e
f
ul
ne
s
s
i
n obt
a
i
ni
ng
hi
g
h a
c
c
ur
a
c
y
l
e
v
e
l
s
.
T
a
bl
e
5. P
e
r
f
or
m
a
nc
e
a
na
l
y
s
i
s
of
e
m
ot
i
on r
e
c
og
ni
t
i
o
n us
i
ng
a
c
c
ur
a
c
y
o
v
e
r
I
S
E
A
R
T
ot
a
l
n
u
m
be
r
of
S
a
m
pl
e
s
A
c
c
ur
a
c
y
P
r
op
os
e
d
F
i
r
e
f
l
y
+
M
L
P
A
c
c
ur
a
c
y
M
L
P
O
n
l
y
A
c
c
ur
a
c
y
P
S
O
+
M
L
P
A
c
c
ur
a
c
y
P
S
O
+
K
N
N
A
c
c
ur
a
c
y
A
C
O
+
M
L
P
1500
92.
0666667
89.
3449092
88.
9575972
73.
3576642
71.
2250712
2500
92.
1600001
90.
5191874
88.
6315789
81.
0260279
70.
4751542
3500
92.
1428571
90.
4130703
90.
5853205
83.
7202468
81.
031614
4500
92.
0666667
90.
0727651
90.
8516612
77.
6238909
85.
5641846
5500
92.
0363636
90.
0793651
90.
5022276
76.
2872361
80.
3833145
6500
92.
0615385
89.
0672016
90.
3322201
72.
2024516
76.
8253968
7500
92.
0700001
89.
9487097
89.
1728312
76.
3767247
69.
6720187
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2502
-
4752
I
ndone
s
i
a
n J
E
l
e
c
E
n
g
&
C
om
p S
c
i
,
V
ol
.
40
, N
o.
1
,
O
c
t
obe
r
20
25
:
4
90
-
498
496
4.
2
.
C
om
p
ar
i
s
on
p
r
o
p
os
e
d
m
e
t
h
od
r
e
s
u
l
t
w
i
t
h
I
S
E
A
R
d
at
as
e
t
T
a
bl
e
6
pr
e
s
e
nt
s
c
o
m
pa
r
i
s
on
of
r
e
s
ul
t
w
i
t
h
I
S
E
A
R
D
a
t
a
s
e
t
.
T
he
c
o
m
bi
na
t
i
on
of
t
h
e
F
i
r
e
f
l
y
A
l
g
or
i
t
h
m
,
a
s
w
a
r
m
i
nt
e
l
l
i
g
e
n
c
e
opt
i
m
i
z
a
t
i
on
t
e
c
hni
que
,
w
i
t
h
M
L
P
,
a
t
y
pe
of
f
e
e
df
or
w
a
r
d
ne
ur
a
l
ne
t
w
or
k,
a
c
hi
e
v
e
d
a
s
i
g
ni
f
i
c
a
nt
l
y
hi
g
he
r
a
c
c
ur
a
c
y
of
92.07%
c
om
pa
r
e
d
t
o
r
e
s
ul
t
s
m
e
nt
i
one
d
i
n
e
xi
s
t
i
ng
l
i
t
e
r
a
t
ur
e
.
T
hi
s
i
ndi
c
a
t
e
s
a
s
ubs
t
a
nt
i
a
l
i
m
pr
o
v
e
m
e
nt
o
v
e
r
t
r
a
di
t
i
ona
l
de
e
p
l
e
a
r
ni
ng
m
e
t
hods
.
T
he
f
i
r
e
f
l
y
a
l
g
o
r
i
t
hm
l
i
ke
l
y
opt
i
m
i
z
e
d
t
he
f
e
a
t
ur
e
s
e
l
e
c
t
i
on
pr
oc
e
s
s
e
f
f
e
c
t
i
v
e
l
y
,
e
na
bl
i
ng
t
he
M
L
P
t
o
f
oc
us
on
t
he
m
os
t
r
e
l
e
v
a
nt
f
e
a
t
ur
e
s
f
or
e
m
ot
i
on
c
l
a
s
s
i
f
i
c
a
t
i
on.
T
hi
s
s
upe
r
i
or
pe
r
f
or
m
a
n
c
e
hi
g
hl
i
g
ht
s
t
he
pot
e
nt
i
a
l
of
i
nt
e
g
r
a
t
i
n
g
s
w
a
r
m
i
nt
e
l
l
i
g
e
nc
e
w
i
t
h ne
ur
a
l
ne
t
w
or
ks
t
o e
nha
nc
e
e
m
ot
i
on
r
e
c
o
g
ni
t
i
on a
c
c
ur
a
c
y
.
T
a
bl
e
6. C
om
pa
r
a
t
i
v
e
pe
r
f
or
m
a
nc
e
a
na
l
y
s
i
s
o
v
e
r
I
S
E
A
R
da
t
a
s
e
t
R
e
f
e
r
e
n
c
e
D
a
t
a
s
e
t
M
e
t
h
od
R
e
s
ul
t
[
2]
C
om
bi
n
e
d
da
t
a
s
e
t
(
I
S
E
A
R
,
W
A
S
S
A
,
a
n
d
E
m
ot
i
o
n
-
s
t
i
m
u
l
us
)
S
V
M
78.
97
A
c
c
ur
a
c
y
R
a
n
do
m
F
o
r
e
s
t
76.
25
A
c
c
ur
a
c
y
NB
68.
94
A
c
c
ur
a
c
y
Bi
-
G
R
U
79.
46
A
c
c
ur
a
c
y
H
yb
r
i
d
(
C
N
N
+
B
i
-
G
R
U
+
S
V
M
)
M
ode
l
80.
11
A
c
c
ur
a
c
y
[
24]
I
S
E
A
R
L
S
T
M
57.
65
A
c
c
ur
a
c
y
B
i
L
S
T
M
59.
30
A
c
c
ur
a
c
y
G
R
U
60.
26
A
c
c
ur
a
c
y
[
25]
I
S
E
A
R
U
n
i
gr
a
m
M
i
xt
ur
e
M
ode
l
(
U
M
M
)
39.
48
F
s
c
or
e
[
26]
I
S
E
A
R
A
t
w
o
-
s
t
a
g
e
t
e
x
t
f
e
a
t
ur
e
e
x
t
r
a
c
t
i
o
n
m
e
t
hod
72.
43
A
c
c
ur
a
c
y
[
27]
I
S
E
A
R
S
V
M
80
P
r
e
c
i
s
i
on,
78
F
M
e
a
s
ur
e
,
79
R
e
c
a
l
l
K
N
N
75
P
r
e
c
i
s
i
on,
78
F
M
e
a
s
ur
e
,
76
R
e
c
a
l
l
XG
-
B
o
os
t
67
P
r
e
c
i
s
i
on,
69
F
M
e
a
s
ur
e
,
68
R
e
c
a
l
l
L
o
g
i
s
t
i
c
R
e
gr
e
s
s
i
on
86
P
r
e
c
i
s
i
on,
84
F
M
e
a
s
ur
e
,
85
R
e
c
a
l
l
[
28]
I
S
E
A
R
R
a
n
do
m
F
o
r
e
s
t
47.
0
A
c
c
ur
a
c
y
S
V
M
54.
3
A
c
c
ur
a
c
y
[
29]
I
S
E
A
R
R
ul
e
B
a
s
e
d
65
A
c
c
ur
a
c
y
[
30]
I
S
E
A
R
+
O
A
N
C
da
t
a
s
e
t
M
L
P
92.
16
A
c
c
ur
a
c
y
D
T
91.
82
A
c
c
ur
a
c
y
K
N
N
78.
16
A
c
c
ur
a
c
y
R
F
89.
41
A
c
c
ur
a
c
y
A
da
boos
t
63.
82
A
c
c
ur
a
c
y
G
B
90.
32
A
c
c
ur
a
c
y
C
N
N
99.
14
A
c
c
ur
a
c
y
[
31]
I
S
E
A
R
N
a
ï
ve
B
a
ye
s
43.
24
A
c
c
ur
a
c
y
R
a
n
do
m
F
o
r
e
s
t
37.
26
A
c
c
ur
a
c
y
S
V
M
14.
48
A
c
c
ur
a
c
y
[
32]
I
S
E
A
R
B
i
G
R
U
54
A
c
c
ur
a
c
y
B
i
L
S
T
M
+
W
or
d
2
V
e
c
57
A
c
c
ur
a
c
y
C
N
N
+
W
or
d2
V
e
c
36
A
c
c
ur
a
c
y
B
i
G
R
U
+
G
l
o
V
e
53
A
c
c
ur
a
c
y
P
r
opos
e
d
M
e
t
h
od
I
S
E
A
R
F
i
r
e
f
l
y
w
i
t
h
M
L
P
92.
07
A
c
c
ur
a
c
y
5.
C
O
N
C
L
U
S
I
O
N
T
he
"
I
m
pr
o
v
e
d
F
i
r
e
f
l
y
+
M
L
P
"
pr
o
v
i
de
a
c
c
ur
a
c
y
of
92.07%
o
v
e
r
I
S
E
A
R
da
t
a
s
e
t
.
T
hi
s
i
n
di
c
a
t
e
s
a
s
i
g
ni
f
i
c
a
nt
e
nha
nc
e
m
e
nt
c
o
m
pa
r
e
d
t
o
c
on
v
e
nt
i
ona
l
a
ppr
oa
c
he
s
.
R
e
s
ul
t
s
a
c
hi
e
v
e
d
i
n
t
hi
s
w
or
k
v
a
l
i
da
t
e
t
he
m
od
e
l
'
s
c
ons
i
s
t
e
nc
y
a
nd s
uc
c
e
s
s
i
n e
m
ot
i
on r
e
c
o
g
ni
t
i
on. T
he
I
m
p
r
o
v
e
d F
i
r
e
f
l
y
A
l
g
or
i
t
h
m
,
a
s
pr
opos
e
d i
n t
hi
s
w
or
k,
r
e
pr
e
s
e
nt
s
a
po
w
e
r
f
ul
t
ool
f
or
f
e
a
t
ur
e
s
e
l
e
c
t
i
on,
e
nha
nc
i
n
g
e
m
ot
i
on
r
e
c
og
ni
t
i
on
i
n
r
e
a
l
-
w
or
l
d
a
ppl
i
c
a
t
i
ons
.
B
y
c
om
bi
ni
ng
s
w
a
r
m
i
nt
e
l
l
i
g
e
n
c
e
w
i
t
h
M
L
P
'
s
l
e
a
r
ni
ng
a
bi
l
i
t
y
,
t
hi
s
a
ppr
oa
c
h
of
f
e
r
s
s
ubs
t
a
nt
i
a
l
i
m
pr
o
v
e
m
e
nt
s
i
n
e
m
ot
i
on
r
e
c
o
g
ni
t
i
on
a
c
c
ur
a
c
y
.
T
he
s
e
di
s
c
o
v
e
r
i
e
s
br
i
ng
up
oppor
t
uni
t
i
e
s
f
or
e
nha
nc
e
d
e
m
ot
i
on r
e
c
o
g
ni
t
i
on a
na
l
y
s
i
s
t
o t
he
na
t
ur
a
l
l
a
ng
ua
ge
pr
oc
e
s
s
i
ng
a
nd ot
he
r
do
m
a
i
ns
.
A
C
K
N
O
WL
E
D
G
M
E
N
T
S
A
ut
hor
s
w
oul
d
l
i
ke
t
o
t
ha
nk
t
he
m
a
n
a
g
e
m
e
nt
of
P
i
l
l
a
i
C
ol
l
e
g
e
of
E
n
g
i
ne
e
r
i
ng
,
P
a
n
v
e
l
;
S
V
K
M
’
s
N
M
I
M
S
, M
um
ba
i
;
a
nd D
. J
. S
a
ng
ha
v
i
C
ol
l
e
g
e
of
E
n
g
i
ne
e
r
i
n
g
, M
u
m
ba
i
.
F
U
N
D
I
N
G
I
N
F
O
R
M
A
T
I
O
N
A
ut
hor
s
s
t
a
t
e
no f
undi
ng
i
n
v
ol
v
e
d.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
ndone
s
i
a
n J
E
l
e
c
E
n
g
&
C
om
p S
c
i
I
S
S
N
:
2502
-
4752
A
n i
m
pr
ov
e
d
c
onv
e
r
s
at
i
on e
m
ot
i
o
n
d
e
t
e
c
t
i
on
us
i
ng hy
b
r
i
d
f
-
nn c
l
as
s
i
f
i
e
r
(
A
b
hi
s
he
k
A
. V
i
c
har
e
)
497
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
j
our
na
l
us
e
s
t
he
C
ont
r
i
but
or
R
ol
e
s
T
a
xono
m
y
(
C
R
e
di
T
)
t
o
r
e
c
og
ni
z
e
i
ndi
v
i
du
a
l
a
ut
hor
c
ont
r
i
but
i
ons
,
r
e
duc
e
a
ut
hor
s
hi
p di
s
pu
t
e
s
, a
nd f
a
c
i
l
i
t
a
t
e
c
ol
l
a
bor
a
t
i
on.
N
am
e
of
A
u
t
h
or
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
A
bhi
s
he
k A
. V
i
c
h
a
r
e
S
a
t
i
s
hkum
a
r
L
. V
a
r
m
a
C
:
C
on
c
e
pt
ua
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
dol
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
da
t
i
o
n
Fo
:
Fo
r
m
a
l
a
n
a
l
ys
i
s
I
:
I
n
ve
s
t
i
g
a
t
i
o
n
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
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
ng
-
R
e
v
i
e
w
&
E
di
t
i
ng
Vi
:
Vi
s
ua
l
i
z
a
t
i
o
n
Su
:
Su
pe
r
vi
s
i
on
P
:
P
r
oj
e
c
t
a
dm
i
n
i
s
t
r
a
t
i
o
n
Fu
:
Fu
n
di
ng
a
c
qui
s
i
t
i
o
n
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
t
a
t
e
no c
onf
l
i
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
t
ha
t
s
uppor
t
t
he
f
i
ndi
ng
s
of
t
hi
s
s
t
udy
a
r
e
a
v
a
i
l
a
bl
e
f
r
o
m
t
he
c
or
r
e
s
pondi
ng
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
.
M
a
c
h
ová
,
M
.
S
z
a
bóova
,
J
.
P
a
r
a
l
i
č
,
a
n
d
J
.
M
i
č
ko,
“
D
e
t
e
c
t
i
o
n
of
e
m
ot
i
o
n
by
t
e
x
t
a
n
a
l
ys
i
s
us
i
n
g
m
a
c
hi
n
e
l
e
a
r
n
i
ng,
”
F
r
ont
i
e
r
s
i
n
P
s
y
c
hol
o
gy
,
vol
.
14,
p.
1190326,
2023,
doi
:
10.
3389/
f
ps
y
g
.
2023.
1190326.
[
2
]
S
.
K
.
B
h
a
r
t
i
,
S
.
V
a
r
a
d
h
a
g
a
n
a
pa
t
h
y
,
R
.
K
.
G
upt
a
,
P
.
K
.
S
h
u
kl
a
,
M
.
B
ouye
,
S
.
K
.
H
i
ng
a
a
,
a
n
d
A
.
M
a
hm
oud
,
“
T
e
xt
-
ba
s
e
d
e
m
ot
i
o
n
r
e
c
ogni
t
i
o
n
us
i
ng
de
e
p
l
e
a
r
ni
ng
a
ppr
oa
c
h
,
”
C
om
put
at
i
onal
I
n
t
e
l
l
i
ge
n
c
e
and
N
e
ur
os
c
i
e
n
c
e
,
vol
.
2022,
p.
2645381,
20
22
,
doi
:
10.
1155/
2022/
264
5381.
[
3
]
N
.
S
h
e
l
ke
,
S
.
C
h
a
ud
h
ur
y
,
S
.
C
h
a
kr
a
ba
r
t
i
,
S
.
L
.
B
a
ng
a
r
e
,
G
.
Y
o
g
a
pr
i
ya
,
a
n
d
P
.
P
a
n
de
y,
“
A
n
e
f
f
i
c
i
e
n
t
w
a
y
of
t
e
x
t
-
ba
s
e
d
e
m
ot
i
o
n
a
n
a
l
ys
i
s
f
r
o
m
s
oc
i
a
l
m
e
di
a
us
i
ng
L
R
A
-
D
N
N
,
”
N
e
ur
os
c
i
e
n
c
e
I
nf
or
m
at
i
c
s
,
vol
.
2,
p.
100048
,
2022,
d
oi
:
10.
1016/
j
.
n
e
u
r
i
.
2022.
100048.
[
4
]
F
.
A
c
h
e
a
m
po
ng,
W
.
C
h
e
n,
a
n
d
H
.
N
u
n
oo
-
M
e
n
s
a
h,
“
T
e
x
t
-
ba
s
e
d
e
m
ot
i
o
n
de
t
e
c
t
i
o
n
:
a
dva
n
c
e
s
,
c
h
a
l
l
e
ng
e
s
a
n
d
oppor
t
u
n
i
t
i
e
s
,
”
E
ngi
ne
e
r
i
ng
R
e
por
t
s
,
2020
,
doi
:
10.
1002/
e
ng
2.
1218
9
.
[
5
]
M
.
R
os
h
a
n
,
M
.
R
a
w
a
t
,
K
.
A
r
ya
n,
E
.
L
ya
ks
o,
A
.
M
.
M
e
ka
l
a
,
a
n
d
N
.
R
uba
n,
“
L
i
ng
ui
s
t
i
c
ba
s
e
d
e
m
ot
i
o
n
a
n
a
l
ys
i
s
us
i
ng
s
of
t
m
a
x
ove
r
t
i
m
e
a
t
t
e
n
t
i
o
n
m
e
c
h
a
ni
s
m
,
”
P
L
oS
O
ne
,
vol
.
19,
2024,
doi
:
10.
1371/
j
o
ur
n
a
l
.
po
n
e
.
0301336.
[
6
]
N
.
A
l
s
w
a
i
da
n
a
n
d
M
.
E
.
B
.
M
e
n
a
i
,
“
A
s
ur
ve
y
of
s
t
a
t
e
-
of
-
t
h
e
-
a
r
t
a
ppr
oa
c
h
e
s
f
or
e
m
ot
i
o
n
r
e
c
o
gn
i
t
i
o
n
i
n
t
e
x
t
,
”
K
now
l
e
dge
and
I
nf
or
m
at
i
on
Sy
s
t
e
m
s
,
vol
.
62,
pp.
2937
–
2987,
2
020,
doi
:
10.
1007/
s
10115
-
020
-
01449
-
0.
[
7
]
E
.
C
ue
va
s
,
F
.
F
a
us
t
o,
a
nd
A
.
G
onz
a
l
e
z
,
“
A
S
w
a
r
m
A
l
g
o
r
i
t
hm
I
ns
pi
r
e
d
by
t
h
e
C
ol
l
e
c
t
i
ve
A
n
i
m
a
l
B
e
h
a
vi
or
,
”
I
n
b
ook
:
N
e
w
A
dv
an
c
e
m
e
nt
s
i
n
Sw
ar
m
A
l
gor
i
t
hm
s
:
O
pe
r
at
or
s
and
A
ppl
i
c
at
i
ons
,
2020,
doi
:
10.
1007/
978
-
3
-
03
0
-
16339
-
6_6.
[
8
]
E
.
C
ue
va
s
,
M
.
C
i
e
n
f
ue
g
os
,
D
.
Z
a
l
dí
va
r
,
a
n
d
M
.
P
é
r
e
z
-
C
i
s
n
e
r
os
,
“
A
s
w
a
r
m
opt
i
m
i
z
a
t
i
o
n
a
l
g
or
i
t
hm
i
n
s
pi
r
e
d
i
n
t
h
e
be
h
a
vi
or
of
t
he
s
oc
i
a
l
-
s
pi
de
r
,
”
E
x
pe
r
t
Sy
s
t
e
m
s
w
i
t
h
A
ppl
i
c
at
i
ons
,
vol
.
40,
pp.
6374
–
6384,
2013
,
doi
:
10.
1016/
j
.
e
s
w
a
.
2013.
05.
041.
[
9
]
G
.
B
e
n
i
,
“
S
w
a
r
m
I
n
t
e
l
l
i
g
e
n
c
e
,
”
I
n
:
Sot
om
ay
or
,
M
.
,
P
é
r
e
z
-
C
as
t
r
i
l
l
o,
D
.
,
C
as
t
i
gl
i
one
,
F
.
(
e
ds
)
C
om
pl
e
x
So
c
i
al
and
B
e
hav
i
or
a
l
Sy
s
t
e
m
s
.
E
nc
y
c
l
ope
di
a
of
C
om
pl
e
x
i
t
y
and
Sy
s
t
e
m
s
Sc
i
e
nc
e
Se
r
i
e
s
,
S
pr
i
ng
e
r
,
2020,
doi
:
10.
1007/
9
78
-
1
-
0716
-
0
368
-
0_530.
[
1
0
]
V
i
c
h
a
r
e
a
n
d
S
.
L
.
V
a
r
m
a
,
“
S
e
n
t
i
m
e
nt
a
n
a
l
ys
i
s
:
i
n
di
a
n
l
a
ng
ua
g
e
s
pe
r
s
pe
c
t
i
ve
,
”
i
n
A
dv
anc
e
d
C
om
put
i
ng,
M
a
c
hi
ne
L
e
ar
ni
ng,
R
obot
i
c
s
and
I
nt
e
r
ne
t
T
e
c
hnol
ogi
e
s
.
A
M
R
I
T
2023.
C
om
m
uni
c
at
i
ons
i
n
C
om
put
e
r
and
I
n
f
or
m
at
i
on
Sc
i
e
nc
e
,
S
pr
i
ng
e
r
,
2024
,
vol
.
1953,
doi
:
10.
1007/
978
-
3
-
0
31
-
4722
4
-
4_23.
[
1
1
]
J
.
G
uo,
“
D
e
e
p
l
e
a
r
n
i
ng
a
ppr
oa
c
h
t
o
t
e
x
t
a
n
a
l
ys
i
s
f
or
h
u
m
a
n
e
m
ot
i
o
n
de
t
e
c
t
i
o
n
f
r
o
m
bi
g
da
t
a
,
”
J
our
nal
o
f
I
nt
e
l
l
i
ge
nt
Sy
s
t
e
m
s
,
vol
.
31,
pp.
113
–
126,
2022
,
doi
:
10.
1515/
j
i
s
ys
-
2022
-
0001.
[
1
2
]
R
.
G
um
m
ul
a
,
V
.
A
r
um
u
g
a
m
,
a
n
d
A
.
A
r
a
ng
a
n
a
t
h
a
n
,
“
D
e
e
p
-
S
F
E
R
:
de
e
p
c
on
vol
u
t
i
o
n
a
l
n
e
ur
a
l
n
e
t
w
or
k
a
n
d
M
F
C
C
a
n
e
f
f
e
c
t
i
ve
s
pe
e
c
h
a
n
d
f
a
c
e
e
m
ot
i
o
n
r
e
c
o
gn
i
t
i
o
n,
”
I
n
done
s
i
an
J
our
nal
of
E
l
e
c
t
r
i
c
al
E
ng
i
ne
e
r
i
ng
and
C
om
put
e
r
Sc
i
e
n
c
e
(
I
J
E
E
C
S)
,
vol
.
36
,
n
o.
3,
pp.
1448
–
1459,
2024
,
doi
:
10.
11591
/
i
j
e
e
c
s
.
v36.
i
3.
pp1448
-
1459.
[
1
3
]
M
.
O
l
i
ve
i
r
a
,
D
.
P
i
n
h
e
i
r
o
,
M
.
M
a
c
e
do,
R
.
B
a
s
t
os
-
F
i
l
h
o
,
a
n
d
R
.
M
e
n
e
z
e
s
,
“
U
n
c
ove
r
i
ng
t
h
e
s
oc
i
a
l
i
n
t
e
r
a
c
t
i
o
n
n
e
t
w
o
r
k
i
n
s
w
a
r
m
i
n
t
e
l
l
i
g
e
n
c
e
a
l
g
o
r
i
t
hm
s
,
”
A
ppl
i
e
d
N
e
t
w
or
k
Sc
i
e
n
c
e
,
vol
.
5,
p.
24,
2020,
doi
:
10.
1007/
s
41109
-
020
-
0
0260
-
8.
[
1
4
]
J
.
T
a
n
g
,
G
.
L
i
u
a
n
d
Q
.
P
a
n
,
"A
R
e
v
i
e
w
o
n
R
e
pr
e
s
e
nt
a
t
i
ve
s
w
a
r
m
i
n
t
e
l
l
i
g
e
n
c
e
a
l
g
or
i
t
hm
s
f
o
r
s
ol
vi
ng
opt
i
m
i
z
a
t
i
on
pr
obl
e
m
s
:
a
ppl
i
c
a
t
i
o
n
s
a
n
d
t
r
e
n
ds
,
"
i
n
I
E
E
E
/
C
A
A
J
our
nal
of
A
ut
om
at
i
c
a
Si
ni
c
a
,
vol
.
8,
n
o.
10,
pp.
1627
-
16
43,
O
c
t
obe
r
2021,
doi
:
10.
1109/
J
A
S
.
2021.
1004129.
[
1
5
]
N
.
A
r
un,
P
.
M
a
h
e
s
w
a
r
a
ve
n
ka
t
e
s
h
,
a
n
d
T
.
J
a
ya
s
a
n
ka
r
,
“
F
a
c
i
a
l
m
i
c
r
o
e
m
ot
i
o
n
de
t
e
c
t
i
o
n
a
n
d
c
l
a
s
s
i
f
i
c
a
t
i
o
n
us
i
ng
s
w
a
r
m
i
n
t
e
l
l
i
g
e
n
c
e
ba
s
e
d
m
odi
f
i
e
d
c
on
vo
l
ut
i
o
n
a
l
n
e
t
w
or
k
,
”
E
x
pe
r
t
Sy
s
t
e
m
s
w
i
t
h
A
ppl
i
c
at
i
ons
,
vol
.
233,
p.
120947,
2023,
doi
:
10.
1016/
j
.
e
s
w
a
.
2023.
120947.
[
1
6
]
Y
.
O
l
m
e
z
,
G
.
O
.
K
oc
a
,
A
.
S
e
ng
ur
,
e
t
a
l
.
,
“
P
S
-
V
T
S
:
P
a
r
t
i
c
l
e
s
w
a
r
m
w
i
t
h
vi
s
i
t
t
a
bl
e
s
t
r
a
t
e
g
y
f
o
r
a
ut
o
m
a
t
e
d
e
m
o
t
i
o
n
r
e
c
o
gn
i
t
i
o
n
w
i
t
h
E
E
G
s
i
gn
a
l
s
,
”
H
e
al
t
h
I
nf
or
m
at
i
on
Sc
i
e
n
c
e
and
Sy
s
t
e
m
s
,
vol
.
11,
n
o.
22,
2023,
doi
:
10.
1007/
s
13755
-
023
-
0
02
24
-
z.
[
1
7
]
M
.
H
a
m
di
,
“
A
f
f
i
r
m
a
t
i
ve
a
n
t
c
ol
o
n
y
opt
i
m
i
z
a
t
i
o
n
ba
s
e
d
s
uppor
t
ve
c
t
or
m
a
c
h
i
n
e
f
or
s
e
n
t
i
m
e
n
t
c
l
a
s
s
i
f
i
c
a
t
i
o
n
,
”
E
l
e
c
t
r
oni
c
s
,
vol
.
11,
p.
1051,
2022,
doi
:
10.
3390/
e
l
e
c
t
r
o
ni
c
s
11071051.
[
1
8
]
M
.
H
a
d
n
i
a
n
d
H
.
H
j
i
a
j
,
“
N
e
w
m
ode
l
of
f
e
a
t
ur
e
s
e
l
e
c
t
i
on
ba
s
e
d
c
h
a
ot
i
c
f
i
r
e
f
l
y
a
l
g
or
i
t
hm
f
or
a
r
a
bi
c
t
e
x
t
c
a
t
e
g
o
r
i
z
a
t
i
o
n
,
”
T
he
I
nt
e
r
nat
i
onal
A
r
ab
J
our
nal
of
I
nf
or
m
at
i
on
T
e
c
hnol
ogy
,
vol
.
20,
2023
,
doi
:
10.
34028/
i
a
j
i
t
/
20/
3A
/
3.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2502
-
4752
I
ndone
s
i
a
n J
E
l
e
c
E
n
g
&
C
om
p S
c
i
,
V
ol
.
40
, N
o.
1
,
O
c
t
obe
r
20
25
:
4
90
-
498
498
[
1
9
]
S
.
I
s
m
a
i
l
,
A
.
O
.
N
a
s
s
i
f
,
A
.
B
.
N
a
s
s
i
f
,
I
.
A
f
youni
,
I
.
A
.
H
a
s
h
e
m
,
a
nd
A
.
E
l
n
a
g
a
r
,
“
A
n
e
f
f
i
c
i
e
n
t
f
e
a
t
ur
e
s
e
l
e
c
t
i
on
m
e
t
h
o
d
f
or
A
ra
bi
c
a
n
d
E
ng
l
i
s
h
s
pe
e
c
h
e
m
ot
i
o
n
r
e
c
o
gn
i
t
i
o
n
us
i
ng
G
r
e
y
W
ol
f
O
pt
i
m
i
z
e
r
,
”
A
ppl
i
e
d
A
c
ous
t
i
c
s
,
vol
.
205,
p.
109279,
2023,
doi
:
10.
1016/
j
.
a
pa
c
ous
t
.
2023.
109279.
[
2
0
]
D
.
E
l
a
ngova
n
a
n
d
V
.
S
ube
d
h
a
,
“
F
i
r
e
f
l
y
w
i
t
h
l
e
vy
ba
s
e
d
f
e
a
t
ur
e
s
e
l
e
c
t
i
o
n
w
i
t
h
m
ul
t
i
l
a
ye
r
pe
r
c
e
pt
r
o
n
f
o
r
s
e
nt
i
m
e
n
t
a
n
a
l
ys
i
s
,
”
J
our
nal
of
A
dv
anc
e
s
i
n
I
nf
or
m
at
i
on
T
e
c
hnol
o
gy
,
vol
.
14,
no.
2,
pp.
342
–
349,
2023,
doi
:
10.
12720/
j
a
i
t
.
14.
2.
342
-
34
9.
[
2
1
]
S
.
D
.
K
us
a
l
,
S
.
G
.
P
a
t
i
l
,
J
.
C
h
oudr
i
e
,
a
n
d
K
.
V
.
K
ot
e
c
h
a
,
“
U
n
de
r
s
t
a
n
d
i
ng
t
h
e
pe
r
f
or
m
a
n
c
e
of
a
i
a
l
g
o
r
i
t
hm
s
i
n
t
e
xt
-
ba
s
e
d
e
m
ot
i
o
n
de
t
e
c
t
i
on
f
or
c
o
n
ve
r
s
a
t
i
o
n
a
l
a
g
e
n
t
s
,
”
A
C
M
T
r
ans
ac
t
i
ons
on
A
s
i
an
and
L
ow
-
R
e
s
our
c
e
L
anguage
I
nf
or
m
at
i
on
P
r
oc
e
s
s
i
ng
,
2023
,
doi
:
10.
1145/
3643133.
[
2
2
]
V
i
c
h
a
r
e
a
n
d
S
.
L
.
V
a
r
m
a
,
“
A
n
a
l
ys
i
s
of
m
a
c
h
i
n
e
l
e
a
r
n
i
ng
a
n
d
de
e
p
l
e
a
r
n
i
ng
t
e
c
hn
i
que
s
f
or
e
m
ot
i
o
n
r
e
c
o
gni
t
i
o
n
us
i
ng
E
ng
l
i
s
h
t
e
x
t
,
”
I
nt
e
r
nat
i
ona
l
J
our
nal
of
A
ppl
i
e
d
E
ngi
ne
e
r
i
ng
and
T
e
c
hnol
ogy
(
L
ondon)
,
vol
.
5,
p.
427,
2023.
[
2
3
]
M
.
M
.
S
.
Z
i
da
n,
I
.
E
l
h
e
n
a
w
y,
A
.
R
.
A
ba
s
,
a
n
d
M
.
O
t
hm
a
n,
“
t
e
xt
ua
l
e
m
ot
i
o
n
de
t
e
c
t
i
o
n
a
ppr
oa
c
h
e
s
:
a
s
ur
ve
y
,
”
F
u
t
ur
e
C
om
put
i
ng
and
I
nf
or
m
at
i
c
s
J
our
nal
,
vol
.
7,
p.
3,
2022,
doi
:
10.
54623/
f
ue
.
f
c
i
j
.
7.
1.
3.
[
2
4
]
D
.
Y
oh
a
n
e
s
,
J
.
S
.
P
ut
r
a
,
K
.
F
i
l
be
r
t
,
K
.
M
.
S
ur
ya
ni
ng
r
u
m
,
a
nd
H
.
A
.
S
a
put
r
i
,
“
E
m
ot
i
o
n
de
t
e
c
t
i
o
n
i
n
t
e
x
t
ua
l
da
t
a
us
i
ng
de
e
p
l
e
a
r
ni
ng
,
”
P
r
o
c
e
di
a
C
om
put
e
r
Sc
i
e
n
c
e
,
vol
.
227,
pp.
464
–
473,
2023,
doi
:
10.
1016/
j
.
pr
oc
s
.
2023.
10.
547.
[
2
5
]
B
a
n
d
h
a
ka
vi
,
N
.
W
i
r
a
t
u
ng
a
,
D
.
P
a
dm
a
n
a
b
h
a
n,
a
n
d
S
.
M
a
s
s
i
e
,
“
L
e
x
i
c
o
n
ba
s
e
d
f
e
a
t
ur
e
e
x
t
r
a
c
t
i
o
n
f
or
e
m
ot
i
o
n
t
e
x
t
c
l
a
s
s
i
f
i
c
a
t
i
on
,
”
P
at
t
e
r
n
R
e
c
ogni
t
i
o
n
L
e
t
t
e
r
s
,
vol
.
93,
pp.
133
–
142,
2017,
doi
:
10.
1016/
j
.
pa
t
r
e
c
.
2016.
12.
009.
[
2
6
]
L
.
J
h
a
j
j
,
S
.
S
i
ngh
,
a
n
d
N
.
A
gg
a
r
w
a
l
,
“
T
w
o
-
s
t
a
g
e
t
e
x
t
f
e
a
t
u
r
e
s
e
l
e
c
t
i
on
m
e
t
hod
f
or
h
u
m
a
n
e
m
ot
i
o
n
r
e
c
ogni
t
i
o
n
,
”
in
P
r
oc
e
e
di
ngs
of
I
C
C
C
N
2018,
N
I
T
T
T
R
C
handi
gar
h,
I
ndi
a
,
2019
,
doi
:
10.
1007/
978
-
981
-
13
-
1217
-
5_51.
[
2
7
]
F
.
A
l
ot
a
i
bi
,
“
C
l
a
s
s
i
f
yi
ng
t
e
x
t
-
ba
s
e
d
e
m
ot
i
o
n
s
us
i
ng
l
o
g
i
s
t
i
c
r
e
g
r
e
s
s
i
o
n
,
”
V
A
W
K
U
M
T
r
ans
ac
t
i
ons
on
C
om
put
e
r
Sc
i
e
nc
e
s
(
V
T
C
S)
,
vol
.
16,
n
o.
2,
pp.
31
–
37,
2019,
doi
:
10.
21015
/
vt
c
s
.
v16i
2.
551.
[
2
8
]
B
.
K
r
a
t
z
w
a
l
d
,
S
.
I
l
i
ć
,
M
.
K
r
a
us
,
S
.
F
e
ue
r
r
i
e
g
e
l
,
a
n
d
H
.
P
r
e
n
di
ng
e
r
,
“
D
e
e
p
l
e
a
r
n
i
ng
f
o
r
a
f
f
e
c
t
i
ve
c
om
p
ut
i
ng
:
t
e
x
t
-
ba
s
e
d
e
m
ot
i
o
n
r
e
c
ogni
t
i
o
n
i
n
d
e
c
i
s
i
o
n
s
uppor
t
,
”
D
e
c
i
s
i
on
Suppor
t
Sy
s
t
e
m
s
,
vol
.
115,
pp.
24
–
35,
2018,
doi
:
10.
1016/
j
.
ds
s
.
2018.
09.
002.
[
2
9
]
D
.
S
e
a
l
,
U
.
K
.
R
oy,
a
n
d
R
.
B
a
s
a
k,
“
S
e
n
t
e
n
c
e
-
l
e
ve
l
e
m
ot
i
o
n
de
t
e
c
t
i
o
n
f
r
o
m
t
e
x
t
ba
s
e
d
on
s
e
m
a
n
t
i
c
r
ul
e
s
,
”
i
n
P
r
oc
e
e
di
ngs
of
t
he
I
nf
or
m
at
i
on
and
C
om
m
uni
c
at
i
o
n
T
e
c
hn
ol
ogy
f
or
Sus
t
ai
nabl
e
D
e
v
e
l
opm
e
nt
,
S
pr
i
ng
e
r
,
2020
,
pp.
423
–
430.
[
3
0
]
F
.
G
h
a
n
ba
r
i
-
A
d
i
vi
a
n
d
M
.
M
os
l
e
h,
“
T
e
x
t
e
m
ot
i
o
n
de
t
e
c
t
i
on
i
n
s
oc
i
a
l
n
e
t
w
or
ks
us
i
ng
a
n
ove
l
e
n
s
e
m
bl
e
c
l
a
s
s
i
f
i
e
r
ba
s
e
d
on
pa
r
z
e
n
t
r
e
e
e
s
t
i
m
a
t
o
r
(
T
P
E
)
,
”
N
e
ur
al
C
om
put
i
ng
and
A
ppl
i
c
at
i
ons
,
vol
.
31,
pp.
8971
–
8983,
2019,
doi
:
10.
1007/
s
00521
-
01
9
-
04230
-
9.
[
3
1
]
K
.
S
a
i
l
u
n
a
z
a
n
d
R
.
A
l
h
a
j
j
,
“
E
m
ot
i
o
n
a
n
d
s
e
n
t
i
m
e
nt
a
n
a
l
ys
i
s
f
r
o
m
T
w
i
t
t
e
r
t
e
x
t
,
”
J
our
nal
of
C
om
put
at
i
onal
S
c
i
e
nc
e
,
vol
.
36,
p
.
101003,
2019
,
doi
:
10.
1016/
j
.
j
oc
s
.
2019.
05.
009.
[
3
2
]
D
.
N
a
z
a
r
e
n
ko
,
I
.
A
f
a
n
a
s
i
e
va
,
N
.
G
ol
i
a
n
,
a
n
d
V
.
G
ol
i
a
n
,
“
I
nve
s
t
i
g
a
t
i
o
n
of
t
h
e
D
e
e
p
L
e
a
r
n
i
ng
A
pp
r
oa
c
h
e
s
t
o
C
l
a
s
s
i
f
y
E
m
ot
i
o
n
s
i
n
T
e
x
t
s
,
”
i
n
P
r
oc
e
e
di
n
gs
of
t
he
I
nt
e
r
nat
i
onal
C
onf
e
r
e
nc
e
on
C
om
put
at
i
onal
L
i
ngui
s
t
i
c
s
and
I
nt
e
l
l
i
ge
nt
Sy
s
t
e
m
s
C
O
L
I
N
S
,
2021.
B
I
O
G
R
A
P
H
I
E
S
O
F
A
U
T
H
O
R
S
Abh
is
h
e
k
A.
Vic
h
a
r
e
r
e
c
e
i
v
e
d
M
.
E
.
d
e
g
r
e
e
i
n
C
o
m
p
u
ter
E
n
g
i
n
e
e
r
i
n
g
f
r
o
m
T
h
a
d
o
m
a
l
S
h
a
h
a
n
i
E
n
g
i
n
e
e
r
i
n
g
C
o
ll
e
g
e
,
B
a
n
d
r
a
,
M
u
m
b
a
i
,
I
n
d
i
a
,
a
n
d
i
s
c
u
r
r
e
n
tl
y
p
u
r
s
u
i
n
g
a
P
h
.
D.
d
e
g
r
e
e
i
n
C
o
m
p
u
ter
E
n
g
i
n
e
e
r
i
n
g
a
t
th
e
M
u
m
b
a
i
U
n
i
v
e
r
s
i
ty
R
e
s
e
a
r
c
h
C
e
n
tr
e
,
P
il
l
a
i
C
o
l
l
e
g
e
o
f
E
n
g
i
n
e
e
r
i
n
g
,
Ne
w
P
a
n
v
e
l.
C
u
r
r
e
n
tl
y
,
h
e
i
s
a
n
As
s
i
s
tan
t
P
r
o
f
e
s
s
o
r
i
n
th
e
De
p
a
r
tm
e
n
t
o
f
C
o
m
p
u
ter
E
n
g
i
n
e
e
r
i
n
g
a
t
S
VK
M
’
s
NM
I
M
S
M
u
k
e
s
h
P
a
te
l
S
c
h
o
o
l
o
f
T
e
c
h
n
o
l
o
g
y
M
a
n
a
g
e
m
e
n
t
&
E
n
g
i
n
e
e
r
i
n
g
,
M
u
m
b
a
i
,
I
n
d
i
a
.
H
i
s
r
e
s
e
a
r
c
h
i
n
ter
e
s
ts
i
n
c
l
u
d
e
a
r
ti
f
ic
i
a
l
i
n
telli
g
e
n
c
e
,
n
a
tu
r
a
l
l
a
n
g
u
a
g
e
p
r
o
c
e
s
s
i
n
g
a
n
d
d
a
ta
m
i
n
i
n
g
.
He
c
a
n
b
e
c
o
n
tac
te
d
a
t
e
m
a
i
l:
v
ich
a
r
e
1
@
g
m
a
il
.
c
o
m
.
S
a
t
is
h
k
u
m
a
r
L
.
Va
r
m
a
h
a
s
b
e
e
n
a
wa
r
d
e
d
P
h
.
D
d
e
g
r
e
e
i
n
C
o
m
p
u
ter
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
r
i
n
g
u
n
d
e
r
th
e
g
u
i
d
a
n
c
e
o
f
Dr
.
T
a
l
b
a
r
S
.
N
.
f
r
o
m
S
GG
S
I
E
&
T
,
S
R
T
M
U,
Na
n
d
e
d
,
I
n
d
i
a
.
He
r
e
c
e
iv
e
d
B
.
T
e
c
h
.
a
n
d
M
.
T
e
c
h
.
d
e
g
r
e
e
s
i
n
C
o
m
p
u
ter
E
n
g
i
n
e
e
r
i
n
g
f
r
o
m
DB
A
T
U
n
i
v
e
r
s
i
ty
a
t
L
o
n
e
r
e
,
M
a
h
a
r
a
s
h
tr
a
,
I
n
d
i
a
i
n
J
u
n
e
2
0
0
0
a
n
d
J
a
n
u
a
r
y
2
0
0
4
r
e
s
p
e
c
ti
v
e
l
y
.
He
i
s
c
u
r
r
e
n
tl
y
e
m
p
l
o
y
e
d
a
s
As
s
o
c
i
a
te
P
r
o
f
e
s
s
o
r
,
De
p
a
r
tm
e
n
t
o
f
I
n
f
o
r
m
a
ti
o
n
T
e
c
h
n
o
l
o
g
y
a
t
S
VK
M
'
s
Dw
a
r
k
a
d
a
s
J
.
S
a
n
g
h
v
i
C
o
ll
e
g
e
o
f
E
n
g
i
n
e
e
r
i
n
g
,
M
u
m
b
a
i
,
I
n
d
i
a
.
H
i
s
r
e
s
e
a
r
c
h
i
n
ter
e
s
ts
i
n
c
l
u
d
e
I
m
a
g
e
a
n
d
Vi
d
e
o
P
r
o
c
e
s
s
i
n
g
,
M
e
d
i
c
a
l
I
m
a
g
e
P
r
o
c
e
s
s
i
n
g
,
Da
ta
M
i
n
i
n
g
a
n
d
I
n
f
o
r
m
a
ti
o
n
R
e
tr
i
e
v
a
l.
H
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
v
s
a
t
2
k
@
g
m
a
il
.
c
o
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.