I
nd
o
ne
s
ia
n J
o
urna
l o
f
E
lect
rica
l En
g
ineering
a
nd
Co
m
pu
t
er
Science
Vo
l.
42
,
No
.
3
,
J
u
n
e
2
0
2
6
,
p
p
.
818
~
82
6
I
SS
N:
2
5
0
2
-
4
7
5
2
,
DOI
: 1
0
.
1
1
5
9
1
/ijeecs.v
42
.i
3
.
pp
818
-
82
6
818
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ee
cs.ia
esco
r
e.
co
m
Statistica
l com
pa
riso
n of ML
P
a
nd
LSTM
f
o
r
mo
bil
e health
sentimen
t
a
na
ly
sis
G
ha
nim
K
a
nu
g
ra
ha
n
1
,
Win Ce
2
,
Vit
o
H
a
f
izh C
a
ha
y
a
P
ut
ra
1
,
Yudi
Ra
m
dh
a
ni
3
,
F
ebri
y
a
nti
P
a
nja
it
a
n
4
1
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
F
a
c
u
l
t
y
o
f
C
r
e
a
t
i
v
e
Te
c
h
n
o
l
o
g
y
,
S
a
t
u
U
n
i
v
e
r
s
i
t
y
,
B
a
n
d
u
n
g
,
I
n
d
o
n
e
s
i
a
2
I
n
f
o
r
mat
i
o
n
S
y
st
e
ms D
e
p
a
r
t
me
n
t
,
S
c
h
o
o
l
o
f
I
n
f
o
r
mat
i
o
n
S
y
s
t
e
ms
,
B
i
n
a
N
u
san
t
a
r
a
U
n
i
v
e
r
si
t
y
,
Ja
k
a
r
t
a
,
I
n
d
o
n
e
si
a
3
D
e
p
a
r
t
me
n
t
o
f
I
n
f
o
r
mat
i
o
n
S
y
s
t
e
ms
,
F
a
c
u
l
t
y
o
f
C
r
e
a
t
i
v
e
Te
c
h
n
o
l
o
g
y
,
S
a
t
u
U
n
i
v
e
r
si
t
y
,
B
a
n
d
u
n
g
,
I
n
d
o
n
e
s
i
a
4
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
F
a
c
u
l
t
y
o
f
C
r
e
a
t
i
v
e
Te
c
h
n
o
l
o
g
y
,
S
a
t
u
U
n
i
v
e
r
s
i
t
y
,
P
a
l
e
mb
a
n
g
,
I
n
d
o
n
e
si
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
u
n
3
0
,
2
0
2
5
R
ev
is
ed
Ma
r
7
,
2
0
2
6
Acc
ep
ted
Ma
y
2
6
,
2
0
2
6
Th
is
stu
d
y
i
n
v
e
stig
a
tes
u
se
r
se
n
ti
m
e
n
t
to
wa
rd
s
t
h
e
M
o
b
il
e
JK
N
p
u
b
li
c
h
e
a
lt
h
a
p
p
li
c
a
ti
o
n
b
y
a
p
p
ly
i
n
g
tex
t
c
las
sifica
ti
o
n
m
o
d
e
ls
b
a
se
d
o
n
d
e
e
p
lea
rn
in
g
.
Two
a
p
p
r
o
a
c
h
e
s
we
re
c
o
m
p
a
re
d
:
a
m
u
lt
i
-
lay
e
r
p
e
rc
e
p
tro
n
(M
LP
)
with
TF
-
IDF
fe
a
tu
re
s
a
nd
lo
n
g
sh
o
rt
-
t
e
rm
m
e
m
o
ry
(LS
TM
)
with
Wo
rd
2
Ve
c
e
m
b
e
d
d
in
g
s.
Th
e
d
a
tas
e
t
c
o
n
si
sts
o
f
1
1
4
,
3
6
4
In
d
o
n
e
sia
n
-
lan
g
u
a
g
e
u
se
r
re
v
iew
s
c
o
ll
e
c
ted
fr
o
m
t
h
e
G
o
o
g
le
P
lay
S
t
o
re
.
To
a
d
d
re
ss
c
las
s
i
m
b
a
lan
c
e
,
we
a
p
p
li
e
d
ra
n
d
o
m
o
v
e
rsa
m
p
li
n
g
.
Eac
h
m
o
d
e
l
wa
s
e
v
a
lu
a
ted
u
sin
g
5
-
f
o
ld
stra
ti
fied
sh
u
ffle
sp
l
it
c
ro
ss
-
v
a
li
d
a
ti
o
n
.
T
h
e
re
su
lt
s
sh
o
we
d
t
h
a
t
M
LP
m
o
d
e
ls
a
c
h
iev
e
d
h
i
g
h
e
r
a
c
c
u
ra
c
y
(u
p
t
o
8
3
.
9
0
%
)
,
wh
il
e
LS
T
M
m
o
d
e
ls d
e
m
o
n
stra
ted
b
e
tt
e
r
re
c
a
ll
a
n
d
p
re
c
isio
n
o
n
m
in
o
rit
y
c
las
se
s
su
c
h
a
s
n
e
u
tral
s
e
n
ti
m
e
n
t.
Ho
we
v
e
r,
sta
ti
stica
l
v
a
li
d
a
ti
o
n
u
s
in
g
t
h
e
Wi
lco
x
o
n
si
g
n
e
d
-
ra
n
k
tes
t
re
v
e
a
led
th
a
t
th
e
p
e
rfo
rm
a
n
c
e
d
iffere
n
c
e
s
b
e
twe
e
n
m
o
d
e
ls
we
re
n
o
t
st
a
ti
stica
ll
y
sig
n
ifi
c
a
n
t
(p
>
0
.
0
5
).
Th
e
se
fi
n
d
i
n
g
s
s
u
g
g
e
st
th
a
t
b
o
t
h
m
o
d
e
ls
a
re
v
iab
le
f
o
r
se
n
ti
m
e
n
t
a
n
a
ly
sis,
with
trad
e
-
o
f
fs
d
e
p
e
n
d
in
g
o
n
t
h
e
e
v
a
lu
a
t
io
n
m
e
tri
c
o
f
in
tere
st.
F
u
tu
re
wo
rk
m
a
y
e
x
p
l
o
r
e
h
y
b
rid
a
rc
h
it
e
c
tu
re
a
n
d
larg
e
r
d
a
tas
e
ts
fo
r
imp
ro
v
e
d
p
e
rf
o
rm
a
n
c
e
a
n
d
sta
ti
st
ica
l
c
o
n
fid
e
n
c
e
.
K
ey
w
o
r
d
s
:
L
o
n
g
s
h
o
r
t
-
ter
m
m
e
m
o
r
y
Mo
b
ile
ap
p
Mu
lti
-
lay
er
p
er
ce
p
tr
o
n
Sen
tim
en
t a
n
aly
s
is
W
ilco
x
o
n
test
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
W
in
C
e
I
n
f
o
r
m
atio
n
Sy
s
tem
s
Dep
ar
tm
en
t,
Sch
o
o
l
o
f
I
n
f
o
r
m
atio
n
Sy
s
tem
s
B
in
a
Nu
s
an
tar
a
Un
iv
er
s
ity
1
1
4
8
0
W
est J
ak
ar
ta,
DKI
J
ak
ar
ta,
I
n
d
o
n
esia
E
m
ail:
wn
.
@
b
in
u
s
.
ed
u
1.
I
NT
RO
D
UCT
I
O
N
T
h
e
r
ap
id
g
r
o
wth
o
f
m
o
b
ile
a
p
p
licatio
n
s
h
as
s
ig
n
if
ican
tly
ch
an
g
ed
h
o
w
p
eo
p
le
in
ter
ac
t
with
d
ig
ital
s
er
v
ices.
W
ith
s
m
ar
tp
h
o
n
es
a
n
d
th
e
in
ter
n
et
b
ec
o
m
in
g
m
o
r
e
af
f
o
r
d
ab
le,
m
o
b
ile
ap
p
s
h
av
e
b
ec
o
m
e
ess
en
tial
to
o
ls
in
m
o
d
er
n
life
[
1
]
.
T
h
is
tr
en
d
h
as
g
r
o
wn
r
a
p
id
ly
o
v
er
th
e
p
ast
d
ec
a
d
e
ar
o
u
n
d
th
e
w
o
r
ld
[
2
]
,
[
3
]
.
Peo
p
le
n
o
w
u
s
e
th
ese
m
o
b
ile
ap
p
lic
atio
n
s
f
o
r
m
a
n
y
ac
tiv
ities
,
f
r
o
m
s
o
cial
n
etwo
r
k
in
g
to
m
a
n
ag
in
g
d
aily
task
s
.
Mo
b
ile
ap
p
licatio
n
s
h
a
v
e
ex
p
an
d
ed
ac
r
o
s
s
v
ar
io
u
s
s
ec
to
r
s
,
in
clu
d
in
g
ed
u
ca
tio
n
,
f
i
n
an
ce
,
an
d
tr
a
n
s
p
o
r
tatio
n
[
4
]
–
[
6
]
.
O
n
e
o
f
th
e
s
ec
to
r
s
th
at
h
as
b
ee
n
g
r
ea
tly
af
f
ec
ted
b
y
m
o
b
ile
a
p
p
d
ev
elo
p
m
en
t
is
h
ea
lth
ca
r
e
[
7
]
,
[
8
]
.
Hea
lth
-
r
elate
d
m
o
b
ile
ap
p
licatio
n
s
n
o
w
p
r
o
v
id
e
s
er
v
ices
s
u
ch
as
telem
ed
icin
e,
ap
p
o
i
n
tm
en
t
s
ch
ed
u
lin
g
,
a
n
d
m
ed
ical
r
ec
o
r
d
m
an
a
g
em
en
t
[
9
]
,
[
1
0
]
.
I
n
I
n
d
o
n
esia,
th
e
Mo
b
ile
J
KN
ap
p
d
ev
elo
p
ed
b
y
B
PJ
S
Kes
eh
atan
s
er
v
es
as
a
g
o
o
d
e
x
am
p
le
o
f
th
is
h
ea
lth
ca
r
e
tr
an
s
f
o
r
m
atio
n
[
1
1
]
.
T
h
e
ap
p
is
d
esig
n
ed
to
h
e
lp
p
ar
ticip
an
ts
o
f
th
e
n
atio
n
a
l
h
ea
lth
in
s
u
r
an
ce
(
J
KN
-
KI
S)
p
r
o
g
r
am
a
n
d
o
f
f
er
s
v
ar
io
u
s
d
ig
ital
s
er
v
ices,
in
clu
d
in
g
r
eg
is
tr
atio
n
,
f
ac
ilit
y
s
ea
r
ch
,
p
r
e
m
iu
m
p
ay
m
en
t,
an
d
m
ed
ical
h
is
to
r
y
tr
ac
k
in
g
[
1
2
]
.
Desp
ite
b
ein
g
u
s
ef
u
l
an
d
wid
ely
u
s
ed
,
Mo
b
i
le
J
KN
h
as
m
ix
ed
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4
7
5
2
S
ta
tis
tica
l c
o
mp
a
r
is
o
n
o
f MLP
a
n
d
LS
TM
fo
r
mo
b
ile
h
e
a
lth
s
en
timen
t a
n
a
lysi
s
(
Gh
a
n
im
K
a
n
u
g
r
a
h
a
n
)
819
u
s
er
s
atis
f
ac
tio
n
lev
els.
T
h
is
ca
n
b
e
s
ee
n
in
th
e
m
an
y
r
ev
iew
s
p
o
s
ted
o
n
th
e
G
o
o
g
le
Play
St
o
r
e,
wh
ich
p
r
o
v
id
e
v
alu
ab
le
f
ee
d
b
ac
k
a
b
o
u
t
u
s
er
ex
p
er
ien
ce
s
,
co
m
p
lain
ts
,
an
d
e
x
p
ec
tatio
n
s
.
Ho
wev
er
,
th
is
f
ee
d
b
ac
k
is
ty
p
ically
u
n
s
tr
u
ctu
r
e
d
an
d
d
if
f
ic
u
lt
to
in
ter
p
r
et
m
an
u
ally
d
u
e
t
o
its
lar
g
e
v
o
lu
m
e
a
n
d
v
ar
ied
n
atu
r
e
[
1
3
]
.
W
ith
s
o
m
an
y
u
s
er
r
ev
iews
av
ailab
le,
B
PJ
S
Keseh
atan
,
as
th
e
o
wn
er
o
f
t
h
e
Mo
b
ile
J
KN
ap
p
licatio
n
,
ca
n
tak
e
ad
v
an
tag
e
o
f
th
is
d
ata
to
u
n
d
er
s
tan
d
wh
at
u
s
er
s
ex
p
er
i
en
ce
an
d
im
p
r
o
v
e
th
e
q
u
ality
o
f
th
e
ap
p
licatio
n
[
9
]
.
Sev
er
al
s
tu
d
ies
h
a
v
e
ex
p
lo
r
ed
th
e
u
s
e
o
f
s
en
tim
en
t
a
n
a
ly
s
is
an
d
m
ac
h
in
e
lear
n
in
g
to
u
n
d
er
s
tan
d
u
s
er
o
p
in
io
n
s
f
r
o
m
a
p
p
r
e
v
iews
[
1
4
]
–
[
1
6
]
.
T
h
ese
r
ev
iews
p
r
o
v
id
e
u
s
ef
u
l
in
s
ig
h
ts
in
to
th
e
m
eth
o
d
s
an
d
m
o
d
els th
at
c
an
b
e
ap
p
lied
to
a
n
aly
ze
u
n
s
tr
u
ctu
r
ed
tex
t
d
ata
f
r
o
m
m
o
b
ile
ap
p
licatio
n
s
[
1
7
]
.
Nu
r
f
ik
r
i
[
1
8
]
co
n
d
u
cted
a
s
tu
d
y
o
n
th
e
Halo
d
o
c
telem
ed
i
cin
e
ap
p
.
T
h
e
aim
o
f
th
e
s
tu
d
y
was
to
an
aly
ze
u
s
er
s
en
tim
en
t
to
war
d
s
Halo
d
o
c
in
I
n
d
o
n
esia
a
f
ter
th
e
C
OVI
D
-
1
9
p
a
n
d
em
ic
b
eg
an
to
d
ec
r
ea
s
e.
T
h
e
d
ataset
co
n
s
is
ted
o
f
1
,
1
2
9
u
s
er
r
ev
iews
co
llected
f
r
o
m
th
e
G
o
o
g
le
Play
Sto
r
e
b
etwe
en
J
u
n
e
an
d
Au
g
u
s
t
2
0
2
2
.
He
u
s
ed
q
u
a
n
titativ
e
an
aly
s
is
to
ca
teg
o
r
ize
u
s
er
r
atin
g
s
in
t
o
f
iv
e
s
en
tim
en
t
lev
els.
Af
ter
th
at,
h
e
also
u
s
ed
q
u
alitativ
e
an
aly
s
is
u
s
in
g
NVI
VO
s
o
f
twar
e
to
id
en
tify
th
e
m
o
s
t
co
m
m
o
n
w
o
r
d
s
in
p
o
s
itiv
e
an
d
n
eg
ativ
e
r
ev
iews.
T
h
e
s
tu
d
y
f
o
u
n
d
th
at
7
4
.
8
%
o
f
t
h
e
r
ev
iews
wer
e
p
o
s
itiv
e,
wh
ile
1
5
.
5
%
wer
e
n
e
g
ativ
e.
T
h
e
r
esu
lts
s
u
g
g
est
th
at
alth
o
u
g
h
u
s
er
s
en
tim
en
t
was
m
o
s
tly
p
o
s
itiv
e,
th
er
e
is
s
til
l
r
o
o
m
f
o
r
im
p
r
o
v
em
en
t,
esp
ec
ially
in
s
y
s
tem
p
er
f
o
r
m
a
n
ce
an
d
s
er
v
i
ce
q
u
ality
.
Ho
u
an
d
Z
h
u
[
1
9
]
co
n
d
u
cte
d
a
s
tu
d
y
to
id
en
tify
th
e
u
s
ef
u
ln
ess
o
f
o
n
lin
e
p
r
o
d
u
ct
r
ev
iews
b
y
co
m
b
in
in
g
g
r
o
u
n
d
e
d
th
eo
r
y
a
n
d
a
m
u
lti
-
lay
er
p
e
r
ce
p
tr
o
n
(
ML
P)
n
eu
r
al
n
etwo
r
k
.
T
h
e
r
esear
ch
u
s
ed
6
,
2
1
5
u
s
er
r
ev
iews
co
llected
f
r
o
m
J
D.
co
m
,
f
o
cu
s
in
g
o
n
m
o
b
ile
p
h
o
n
e
p
r
o
d
u
cts.
T
h
e
au
th
o
r
s
f
i
r
s
t
co
n
d
u
cte
d
s
em
i
-
s
tr
u
ctu
r
ed
in
ter
v
iews
with
3
5
co
n
s
u
m
er
s
to
ex
tr
ac
t
f
ea
tu
r
es
r
elate
d
to
d
if
f
er
en
t
s
tag
es
o
f
th
e
p
u
r
c
h
asin
g
p
r
o
ce
s
s
.
T
h
e
p
r
o
ce
s
s
es
ar
e
d
em
an
d
g
en
e
r
atio
n
,
in
f
o
r
m
atio
n
co
llectio
n
,
p
r
o
d
u
ct
ev
alu
atio
n
,
p
u
r
ch
ase
d
ec
is
io
n
,
an
d
p
o
s
t
-
p
u
r
ch
ase
b
e
h
a
v
io
r
.
T
h
ese
f
ea
tu
r
es
wer
e
th
e
n
u
s
ed
to
tr
ain
an
ML
P
m
o
d
el.
T
h
e
r
esu
lts
s
h
o
wed
th
at
th
e
ML
P
m
o
d
el
tr
ain
e
d
with
f
ea
tu
r
es
b
ased
o
n
co
n
s
u
m
e
r
d
ec
is
io
n
-
m
ak
in
g
ac
h
iev
e
d
an
F
1
-
s
co
r
e
o
f
8
9
.
3
%,
s
ig
n
if
ican
tly
o
u
tp
er
f
o
r
m
in
g
m
o
d
els
u
s
in
g
o
n
l
y
T
F
-
I
DF
f
ea
tu
r
es
(
F1
-
s
co
r
e
o
f
5
9
.
2
%)
an
d
tr
ad
itio
n
al
class
if
ier
s
lik
e
SVM
an
d
Naï
v
e
B
ay
es.
T
h
is
r
esear
ch
s
u
g
g
ests
th
at
in
clu
d
in
g
b
eh
av
i
o
r
al
co
n
tex
t
in
f
ea
tu
r
e
d
esig
n
ca
n
im
p
r
o
v
e
th
e
p
er
f
o
r
m
an
ce
o
f
m
ac
h
in
e
lear
n
in
g
m
o
d
els in
id
en
tify
i
n
g
u
s
ef
u
l o
n
li
n
e
r
ev
iews.
W
h
ile
th
ese
s
t
u
d
ies
s
h
o
w
g
r
ea
t
r
esu
lts
,
d
eter
m
in
i
n
g
th
e
tr
u
e
ef
f
ec
tiv
en
ess
o
f
d
if
f
er
e
n
t
m
o
d
els
r
eq
u
ir
es
m
o
r
e
th
a
n
ju
s
t
co
m
p
ar
in
g
p
e
r
f
o
r
m
an
ce
m
etr
ic
s
.
W
h
en
ev
alu
atin
g
m
ac
h
in
e
lear
n
in
g
m
o
d
els,
r
esear
ch
er
s
m
u
s
t
en
s
u
r
e
th
at
o
b
s
er
v
ed
p
er
f
o
r
m
a
n
ce
d
if
f
e
r
en
ce
s
ar
e
s
tatis
tically
s
ig
n
i
f
ican
t
r
ath
er
th
an
o
cc
u
r
r
in
g
b
y
c
h
an
ce
.
R
ec
o
g
n
izin
g
th
is
c
r
itical
n
ee
d
f
o
r
s
tatis
tical
v
alid
atio
n
,
Sh
ar
m
a
an
d
Kau
r
[
2
0
]
co
n
d
u
cte
d
a
b
en
c
h
m
ar
k
in
g
s
tu
d
y
in
v
o
l
v
in
g
3
5
d
ee
p
lear
n
in
g
m
o
d
els
f
o
r
asp
ec
t
-
le
v
el
s
en
t
im
en
t
class
if
icatio
n
(
AL
SC
)
.
T
h
ey
u
s
ed
eig
h
t
b
en
ch
m
ar
k
d
atasets
an
d
ev
alu
ate
d
m
o
d
el
p
er
f
o
r
m
a
n
ce
b
ased
o
n
ac
cu
r
ac
y
,
m
ac
r
o
-
F1
s
co
r
e,
an
d
tr
ain
in
g
tim
e.
T
o
ch
ec
k
t
h
e
s
tatis
tical
v
alid
ity
o
f
th
e
p
er
f
o
r
m
an
ce
d
if
f
er
e
n
c
es,
th
ey
ap
p
lied
th
e
Frie
d
m
an
test
,
f
o
llo
wed
b
y
p
o
s
t
-
h
o
c
test
s
s
u
ch
as
th
e
Nem
en
y
i
a
n
d
W
ilco
x
o
n
s
ig
n
ed
-
r
a
n
k
test
s
.
T
h
e
W
ilco
x
o
n
test
was
u
s
ed
to
p
er
f
o
r
m
p
air
wis
e
co
m
p
a
r
is
o
n
s
b
etwe
en
to
p
-
p
er
f
o
r
m
i
n
g
m
o
d
els.
T
h
eir
r
esu
lts
s
h
o
wed
th
at
m
o
d
els
lik
e
GAT
-
B
E
R
T
an
d
ASGC
N
ac
h
iev
ed
th
e
b
est
p
er
f
o
r
m
an
ce
,
alth
o
u
g
h
s
o
m
e
tr
ad
e
-
o
f
f
s
ex
is
ted
b
etwe
en
ac
cu
r
ac
y
an
d
ef
f
icien
cy
.
I
n
s
en
tim
en
t
an
aly
s
is
r
esear
ch
,
th
e
s
elec
tio
n
o
f
f
ea
tu
r
e
ex
tr
ac
tio
n
m
eth
o
d
a
n
d
m
o
d
el
ar
ch
itectu
r
e
p
lay
s
an
im
p
o
r
ta
n
t
r
o
le
in
d
e
ter
m
in
in
g
class
if
icatio
n
p
er
f
o
r
m
an
ce
.
O
n
e
co
m
m
o
n
l
y
u
s
ed
tech
n
iq
u
e
is
T
er
m
Fre
q
u
en
cy
-
I
n
v
e
r
s
e
Do
cu
m
e
n
t
Fre
q
u
en
c
y
(
T
F
-
I
DF)
,
wh
ich
co
n
v
e
r
ts
tex
t
in
to
n
u
m
e
r
ical
v
ec
to
r
s
b
y
g
i
v
in
g
weig
h
t
to
ter
m
s
b
ased
o
n
th
ei
r
f
r
eq
u
en
cy
in
a
d
o
cu
m
en
t
an
d
th
eir
r
ar
ity
ac
r
o
s
s
th
e
d
atase
t
[
2
1
]
.
T
h
is
m
et
h
o
d
h
as
b
ee
n
p
r
o
v
en
e
f
f
ec
tiv
e
in
tr
ad
itio
n
al
m
ac
h
i
n
e
lear
n
i
n
g
t
ask
s
,
as
it
h
elp
s
ca
p
tu
r
e
im
p
o
r
tan
t
ter
m
s
wh
ile
m
in
im
izin
g
th
e
in
f
lu
en
ce
o
f
c
o
m
m
o
n
l
y
u
s
ed
wo
r
d
s
[
2
2
]
,
[
2
3
]
.
MLP
is
a
b
asic
d
ee
p
lear
n
in
g
m
o
d
el
co
n
s
is
tin
g
o
f
s
ev
er
a
l
f
u
lly
co
n
n
ec
ted
lay
er
s
th
at
ca
n
lear
n
co
m
p
lex
p
atter
n
s
in
d
ata
u
s
in
g
b
ac
k
p
r
o
p
ag
atio
n
[
2
4
]
.
W
h
en
co
m
b
in
ed
with
TF
-
I
DF
f
ea
tu
r
es,
ML
P
m
o
d
els
ca
n
ef
f
ec
tiv
ely
r
ec
o
g
n
ize
r
elat
io
n
s
h
ip
s
b
etwe
en
wo
r
d
s
in
tex
t
class
if
icat
io
n
task
s
,
esp
ec
iall
y
b
ec
au
s
e
th
ey
ar
e
ca
p
ab
le
o
f
m
o
d
elin
g
n
o
n
-
lin
ea
r
d
ep
en
d
en
cies b
etwe
en
f
ea
t
u
r
es
[
2
5
]
,
[
2
6
]
.
On
th
e
o
th
er
h
a
n
d
,
W
o
r
d
2
Ve
c
is
a
f
ea
tu
r
e
ex
tr
ac
tio
n
tech
n
iq
u
e
t
h
at
tr
a
n
s
f
o
r
m
s
w
o
r
d
s
i
n
to
d
e
n
s
e
v
ec
to
r
r
ep
r
esen
tatio
n
s
b
ased
o
n
th
eir
co
n
tex
t
in
s
en
ten
ce
s
[
2
7
]
.
Un
lik
e
th
e
s
p
ar
s
e
v
ec
to
r
s
g
en
er
ated
b
y
TF
-
I
DF,
W
o
r
d
2
Vec
p
r
eser
v
e
th
e
s
em
an
tic
m
ea
n
in
g
o
f
wo
r
d
s
,
allo
win
g
th
e
m
o
d
el
to
u
n
d
er
s
tan
d
s
i
m
ilar
ity
b
ased
o
n
u
s
ag
e
[
2
8
]
.
T
h
ese
d
en
s
e
em
b
ed
d
in
g
s
ar
e
p
ar
tic
u
lar
ly
well
-
s
u
ited
f
o
r
lo
n
g
s
h
o
r
t
-
ter
m
m
e
m
o
r
y
(
L
STM
)
n
etwo
r
k
s
,
wh
ich
ar
e
a
ty
p
e
o
f
r
ec
u
r
r
en
t
n
eu
r
al
n
etwo
r
k
d
esig
n
e
d
to
h
a
n
d
le
s
eq
u
en
ce
d
ata
[
2
9
]
.
L
STM
s
h
av
e
s
p
ec
ial
g
atin
g
m
ec
h
an
is
m
s
th
at
allo
w
th
em
to
r
etain
o
r
f
o
r
g
et
in
f
o
r
m
atio
n
as
n
ee
d
e
d
,
m
ak
in
g
th
em
ef
f
ec
tiv
e
f
o
r
u
n
d
er
s
tan
d
in
g
th
e
f
l
o
w
an
d
c
o
n
tex
t
o
f
n
at
u
r
al
lan
g
u
a
g
e
tex
t
[
3
0
]
,
[
3
1
]
.
Sev
er
al
p
r
ev
i
o
u
s
s
tu
d
ies
h
av
e
ap
p
lied
m
ac
h
in
e
lear
n
in
g
an
d
d
ee
p
lea
r
n
in
g
tech
n
i
q
u
es
f
o
r
s
en
tim
en
t
an
aly
s
is
o
n
m
o
b
ile
ap
p
licati
o
n
s
,
in
clu
d
i
n
g
h
ea
lth
ca
r
e
p
l
atf
o
r
m
s
s
u
ch
as
Mo
b
ile
J
KN.
E
x
is
tin
g
wo
r
k
s
g
en
er
ally
r
e
p
o
r
t
th
at
m
o
d
els
s
u
ch
as
L
STM
an
d
tr
ad
itio
n
al
class
if
ier
s
ca
n
ef
f
ec
tiv
ely
clas
s
if
y
u
s
er
s
en
tim
en
t,
with
p
er
f
o
r
m
an
ce
ev
alu
atio
n
p
r
im
ar
ily
b
ased
o
n
a
cc
u
r
ac
y
-
b
ased
m
etr
ics
[
1
1
]
,
[
1
2
]
.
Ho
wev
er
,
th
ese
s
tu
d
ies
r
ely
o
n
d
ir
ec
t
m
etr
ic
co
m
p
ar
is
o
n
s
with
o
u
t
s
tati
s
tical
v
alid
atio
n
,
m
ak
in
g
it
u
n
clea
r
wh
eth
er
o
b
s
er
v
e
d
p
er
f
o
r
m
an
ce
d
if
f
e
r
en
ce
s
b
etwe
en
m
o
d
els
ar
e
s
ig
n
if
ican
t.
I
n
ad
d
itio
n
,
co
m
p
ar
ativ
e
an
aly
s
e
s
b
etwe
en
d
if
f
er
en
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
5
2
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
42
,
No
.
3
,
J
u
n
e
20
2
6
:
8
1
8
-
82
6
820
d
ee
p
lear
n
in
g
ar
ch
itectu
r
es
an
d
f
ea
tu
r
e
ex
tr
ac
tio
n
m
eth
o
d
s
,
p
ar
ticu
lar
ly
ML
P
with
T
F
-
I
DF
an
d
L
STM
with
W
o
r
d
2
Vec
,
r
em
ain
lim
ited
f
o
r
lar
g
e
-
s
ca
le
I
n
d
o
n
esian
m
o
b
ile
h
ea
lth
r
ev
iew
d
ata.
T
h
er
ef
o
r
e,
th
is
s
tu
d
y
aim
s
to
s
y
s
tem
atica
lly
co
m
p
ar
e
th
e
s
e
two
m
o
d
elin
g
ap
p
r
o
ac
h
es
with
in
a
u
n
if
ie
d
e
x
p
er
im
e
n
tal
f
r
am
ewo
r
k
an
d
t
o
em
p
lo
y
th
e
W
ilco
x
o
n
s
ig
n
ed
-
r
an
k
test
to
s
tatis
tically
ev
alu
ate
th
e
s
ig
n
if
ican
ce
o
f
p
er
f
o
r
m
a
n
ce
d
if
f
er
e
n
ce
s
.
2.
M
E
T
H
O
D
T
h
is
s
t
u
d
y
a
p
p
li
es
a
m
ac
h
i
n
e
l
ea
r
n
in
g
a
p
p
r
o
a
ch
t
o
an
al
y
z
e
u
s
er
s
e
n
ti
m
e
n
t
f
r
o
m
M
o
b
ile
J
K
N
r
ev
iews
.
T
h
e
p
r
o
c
e
s
s
c
o
n
s
is
ts
o
f
f
o
u
r
m
ai
n
s
t
ag
es.
T
h
e
s
t
ag
es
a
r
e
d
ata
co
lle
cti
o
n
,
d
a
ta
p
r
e
p
r
o
c
ess
in
g
,
m
o
d
el
b
u
il
d
i
n
g
,
an
d
m
o
d
el
e
v
a
lu
ati
o
n
.
Fi
g
u
r
e
1
ill
u
s
t
r
at
es
th
e
c
o
m
p
l
ete
e
x
p
e
r
i
m
e
n
ta
l s
et
u
p
a
n
d
wo
r
k
f
l
o
w
u
s
ed
i
n
th
is
s
t
u
d
y
.
Fig
u
r
e
1
.
R
esear
ch
f
r
am
ewo
r
k
2
.1
.
Da
t
a
c
o
llect
io
n
User
r
ev
iew
d
ata
was
co
llected
f
r
o
m
th
e
Go
o
g
le
Play
Sto
r
e
u
s
in
g
th
e
g
o
o
g
le_
p
lay
_
s
cr
ap
er
Py
th
o
n
lib
r
ar
y
.
Data
s
cr
ap
in
g
was
co
n
d
u
cted
o
n
Ap
r
il
3
0
,
2
0
2
5
,
an
d
it
in
clu
d
es
r
ev
iews
f
r
o
m
v
er
s
io
n
1
.
1
8
(
J
u
ly
2
1
,
2
0
2
1
)
u
p
to
v
e
r
s
io
n
4
.
1
2
(
Ap
r
il
2
9
,
2
0
2
5
)
.
W
e
f
o
cu
s
ed
o
n
ly
o
n
r
ev
iews
wr
itten
in
I
n
d
o
n
e
s
ian
lan
g
ag
e.
E
ac
h
r
ev
iew
was
ex
tr
ac
ted
alo
n
g
with
o
th
er
d
ata
s
u
ch
as
r
ev
ie
w
co
n
ten
t,
r
atin
g
s
co
r
e
,
an
d
t
im
estam
p
.
Fro
m
th
e
to
tal
n
u
m
b
e
r
o
f
Mo
b
ile
J
KN
r
ev
iews
(
m
o
r
e
t
h
an
2
m
illi
o
n
r
ev
iews),
we
o
n
ly
s
elec
ted
2
0
0
,
0
0
0
r
ev
iews
as
a
s
am
p
le
o
r
ar
o
u
n
d
1
0
%
to
tal
r
e
v
iews
,
f
o
r
th
is
s
tu
d
y
.
T
h
e
ex
tr
ac
ted
d
ata
ca
n
b
e
d
escr
ib
ed
in
T
ab
le
1
.
T
ab
le
1
.
Sam
p
le
o
f
co
llected
u
s
er
r
ev
iews
R
e
v
i
e
w
Id
U
ser
n
a
me
C
o
n
t
e
n
t
S
c
o
r
e
A
p
p
V
e
r
s
i
o
n
b
e
7
7
e
4
d
3
…
Ju
n
a
e
d
i
S
e
t
e
l
a
h
d
i
u
p
d
a
t
e
b
a
ru
.
.
.
m
a
n
t
a
a
p
.
.
.
b
i
sa
p
a
k
e
…
5
3
.
4
.
0
7
6
e
0
c
b
e
7
…
Ek
o
t
r
i
J
a
g
a
k
e
se
h
a
t
a
n
t
e
t
a
p
sema
n
g
a
t
t
e
r
a
p
k
a
n
h
i
d
u
p
…
5
3
.
4
.
0
6
d
6
e
3
f
e
d
…
A
n
d
r
i
L
u
k
ma
n
a
t
i
d
a
k
b
i
s
a
m
a
su
k
1
1
.
1
8
2
.
2
.
Da
t
a
p
re
pro
ce
s
s
ing
Af
ter
th
e
r
ev
iews
wer
e
co
lle
cted
,
we
ad
d
ed
a
n
ew
co
lu
m
n
ca
lled
s
en
tim
en
t.
I
n
th
is
c
o
lu
m
n
,
we
tr
an
s
f
o
r
m
ed
th
e
r
atin
g
s
co
r
es
in
to
s
en
tim
en
t
lab
els:
r
ev
iews
with
a
s
co
r
e
o
f
1
o
r
2
wer
e
l
ab
eled
as
n
eg
ativ
e
(
lab
el
0
)
,
s
co
r
es
o
f
3
as
n
eu
tr
al
(
lab
el
1
)
,
an
d
s
co
r
es
o
f
4
o
r
5
as
p
o
s
itiv
e
(
lab
el
2
)
.
T
h
e
tex
t
p
r
ep
r
o
ce
s
s
in
g
p
ip
eli
n
e
co
n
s
is
ted
o
f
t
h
e
f
o
llo
win
g
s
tep
s
:
-
E
m
o
ji
c
o
n
v
e
r
s
i
o
n
:
e
m
o
j
is
w
er
e
c
o
n
v
e
r
te
d
i
n
t
o
te
x
t
u
a
l
r
e
p
r
es
en
t
ati
o
n
u
s
i
n
g
t
h
e
e
m
o
ji
li
b
r
ar
y
.
-
T
e
x
t
n
o
r
m
ali
za
t
io
n
:
t
ex
t w
as
c
o
n
v
er
te
d
to
l
o
w
er
ca
s
e.
-
No
is
e
r
e
m
o
v
al
:
p
u
n
ct
u
at
io
n
,
el
o
n
g
ate
d
c
h
a
r
a
cte
r
s
,
an
d
UR
L
s
wer
e
r
e
m
o
v
e
d
.
-
Sto
p
w
o
r
d
r
em
o
v
a
l
a
n
d
s
te
m
m
i
n
g
:
w
e
u
s
ed
t
h
e
I
n
d
o
n
esia
n
s
t
o
p
w
o
r
d
lis
t
f
r
o
m
N
L
T
K
a
n
d
a
p
p
li
ed
s
t
em
m
i
n
g
u
s
i
n
g
t
h
e
S
astr
awi
s
t
em
m
e
r
.
T
h
e
r
ev
iew
tex
t
was
av
ailab
le
in
th
e
co
n
te
n
t
co
lu
m
n
.
T
h
e
c
lean
ed
tex
t
was
s
to
r
e
d
in
a
n
e
w
co
lu
m
n
ca
lled
clea
n
ed
_
c
o
n
ten
t.
Du
r
in
g
p
r
e
p
r
o
ce
s
s
in
g
,
we
f
o
u
n
d
th
at
m
an
y
u
s
er
r
e
v
iews
wer
e
ex
tr
em
ely
s
h
o
r
t,
s
u
ch
as
“
ok
”
o
r
“
to
p
”
,
wh
ich
ca
u
s
ed
th
e
clea
n
e
d
_
co
n
ten
t
to
b
ec
o
m
e
em
p
ty
a
f
ter
f
ilter
in
g
.
T
o
a
d
d
r
ess
th
is
is
s
u
e,
we
r
etain
ed
th
ese
s
h
o
r
t
tex
ts
b
y
c
o
p
y
in
g
th
em
d
i
r
ec
tly
in
t
o
th
e
cle
an
ed
_
co
n
ten
t
c
o
lu
m
n
wh
en
th
e
clea
n
ed
r
esu
lt
was e
m
p
ty
.
T
h
is
en
s
u
r
ed
th
at
all
r
ev
iews st
ill r
ef
lecte
d
th
eir
o
r
ig
in
al
m
ea
n
i
n
g
an
d
s
en
tim
e
n
t.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4
7
5
2
S
ta
tis
tica
l c
o
mp
a
r
is
o
n
o
f MLP
a
n
d
LS
TM
fo
r
mo
b
ile
h
e
a
lth
s
en
timen
t a
n
a
lysi
s
(
Gh
a
n
im
K
a
n
u
g
r
a
h
a
n
)
821
Nex
t,
d
u
p
licate
r
ev
iews
wer
e
r
em
o
v
ed
to
av
o
i
d
b
iased
lear
n
in
g
f
r
o
m
r
e
p
ea
ted
s
am
p
les.
Af
ter
th
at,
o
n
ly
two
co
lu
m
n
s
wer
e
r
etain
ed
f
o
r
m
o
d
el
tr
ain
in
g
an
d
e
v
alu
atio
n
,
wh
ich
ar
e
clea
n
ed
_
co
n
ten
t
an
d
s
en
tim
en
t.
T
h
e
f
in
al
d
ata
t
h
at
will b
e
u
s
e
d
ca
n
b
e
d
escr
ib
ed
i
n
T
ab
le
2
.
T
ab
le
2
.
Sam
p
le
o
f
ex
tr
ac
te
d
d
ata
f
r
o
m
d
ata
p
r
ep
r
o
ce
s
s
in
g
c
l
e
a
n
e
d
_
c
o
n
t
e
n
t
sen
t
i
m
e
n
t
U
p
d
a
t
e
m
a
n
t
a
a
p
p
a
k
e
n
i
k
2
J
a
g
a
se
h
a
t
s
e
m
a
n
g
a
t
t
e
r
a
p
h
i
d
u
p
s
e
h
a
t
2
M
a
s
u
k
0
2
.
3
.
M
o
del
b
uilding
B
ef
o
r
e
tr
ain
in
g
th
e
m
o
d
els,
r
a
n
d
o
m
o
v
e
r
s
am
p
lin
g
was
ap
p
li
ed
to
h
an
d
le
th
e
class
im
b
ala
n
ce
in
th
e
d
ataset.
Af
ter
p
r
ep
r
o
ce
s
s
in
g
,
th
e
n
eg
ativ
e
s
en
tim
en
t
class
h
ad
th
e
lar
g
est
n
u
m
b
er
o
f
s
am
p
les,
wh
ile
th
e
n
eu
tr
al
an
d
p
o
s
itiv
e
class
e
s
w
er
e
m
u
ch
s
m
aller
.
T
o
r
ed
u
ce
t
h
is
im
b
alan
ce
,
th
e
R
an
d
o
m
Ov
er
Sam
p
ler
f
r
o
m
th
e
im
b
lear
n
lib
r
ar
y
was u
s
ed
to
o
v
er
s
am
p
le
th
e
n
eu
tr
al
an
d
p
o
s
i
tiv
e
class
e
s
b
y
d
u
p
licatin
g
ex
is
tin
g
s
am
p
les u
n
til
th
eir
n
u
m
b
e
r
s
wer
e
eq
u
al
to
t
h
e
n
eg
ativ
e
class
.
T
h
is
o
v
er
s
a
m
p
lin
g
p
r
o
ce
s
s
was
ap
p
lied
o
n
ly
to
t
h
e
tr
ain
in
g
d
ata
in
ea
ch
f
o
ld
o
f
th
e
cr
o
s
s
-
v
alid
atio
n
,
wh
ile
th
e
test
d
ata
was
k
ep
t
u
n
ch
an
g
e
d
.
T
h
is
s
tep
was
in
ten
d
ed
to
h
elp
th
e
m
o
d
els lea
r
n
m
o
r
e
ev
en
ly
f
r
o
m
all
s
en
tim
en
t c
lass
es.
W
e
u
s
e
a
d
ee
p
lear
n
in
g
ap
p
r
o
ac
h
to
co
m
p
ar
e
th
e
p
e
r
f
o
r
m
an
ce
o
f
two
ty
p
es
o
f
m
o
d
els:
ML
P
with
TF
-
I
DF
f
ea
tu
r
es,
a
n
d
L
STM
with
W
o
r
d
2
Vec
em
b
ed
d
i
n
g
s
.
W
e
ap
p
lied
th
e
TF
-
I
DF
v
ec
t
o
r
izer
with
u
n
ig
r
a
m
(
n
g
r
am
_
r
an
g
e=
(
1
,
1
)
)
an
d
a
m
ax
im
u
m
o
f
5
,
0
0
0
f
ea
tu
r
es
t
o
c
o
n
v
er
t
th
e
clea
n
ed
te
x
t
in
to
n
u
m
er
ical
v
ec
to
r
s
.
O
n
th
e
o
th
er
h
an
d
,
we
tr
ain
ed
a
W
o
r
d
2
Vec
m
o
d
el
o
n
th
e
to
k
e
n
ized
r
ev
iews
u
s
in
g
v
ec
to
r
_
s
i
ze
=1
0
0
,
win
d
o
w=
5
,
an
d
m
in
_
co
u
n
t=1
.
T
h
e
em
b
e
d
d
in
g
f
o
r
ea
ch
r
ev
iew
was
o
b
tain
ed
b
y
av
e
r
ag
in
g
th
e
v
ec
to
r
s
o
f
th
e
wo
r
d
s
it
co
n
tain
ed
.
W
e
also
u
s
e
th
e
Ad
am
o
p
tim
izer
with
a
le
ar
n
in
g
r
ate
o
f
0
.
0
0
1
an
d
wer
e
tr
ai
n
ed
u
s
in
g
C
r
o
s
s
E
n
tr
o
p
y
L
o
s
s
.
T
h
e
m
o
d
el
s
wer
e
tr
ain
e
d
f
o
r
2
0
ep
o
ch
s
with
a
b
atc
h
s
ize
o
f
6
4
.
Fo
u
r
m
o
d
el
a
r
ch
itectu
r
es
wer
e
b
u
ilt:
-
ML
P
A
:
o
n
e
h
i
d
d
e
n
la
y
e
r
wit
h
1
2
8
u
n
its
a
n
d
d
r
o
p
o
u
t
.
-
ML
P B
:
t
wo
h
i
d
d
e
n
l
a
y
e
r
s
wit
h
1
2
8
an
d
6
4
u
n
i
ts
r
es
p
ec
t
iv
el
y
,
b
o
th
wi
t
h
d
r
o
p
o
u
t.
-
L
STM
A
:
o
n
e
L
S
T
M
la
y
e
r
wit
h
hi
dd
e
n_
di
m
=1
28
wit
h
d
r
o
p
o
u
t
a
n
d
a
d
en
s
e
la
y
e
r
.
-
L
STM
B
:
t
w
o
s
ta
ck
e
d
L
STM
l
ay
er
s
wi
th
d
i
m
e
n
s
i
o
n
s
1
2
8
a
n
d
6
4
wit
h
d
r
o
p
o
u
t
a
n
d
a
d
en
s
e
l
ay
er
.
All
ex
p
er
im
en
ts
wer
e
co
n
d
u
cted
in
Py
th
o
n
u
s
in
g
Py
T
o
r
ch
,
with
T
F
-
I
DF
an
d
W
o
r
d
2
Vec
im
p
le
m
en
ted
v
ia
Scik
it
-
lear
n
an
d
Gen
s
im
.
T
r
ain
i
n
g
was a
cc
eler
ated
u
s
in
g
an
NVI
DI
A
3
0
5
0
T
i L
ap
to
p
GPU
.
2
.
4
.
M
o
del
e
v
a
lua
t
i
o
n
T
h
e
d
ataset
was
s
p
lit
u
s
in
g
Stra
tifie
d
Sh
u
f
f
leSp
lit
with
5
s
p
lits
with
8
0
%
tr
ain
in
g
an
d
2
0
%
test
in
g
s
et
.
T
h
is
m
eth
o
d
en
s
u
r
ed
th
at
th
e
p
r
o
p
o
r
tio
n
o
f
s
en
tim
en
t
lab
els
r
em
ain
ed
co
n
s
is
ten
t
ac
r
o
s
s
th
e
tr
ain
in
g
an
d
test
s
et
s
.
I
n
ea
ch
f
o
ld
o
f
th
e
cr
o
s
s
-
v
alid
atio
n
,
b
o
t
h
ML
P
m
o
d
els
wer
e
tr
ain
ed
u
s
in
g
t
h
e
T
F
-
I
DF
f
ea
tu
r
e
ex
tr
ac
tio
n
,
wh
ile
b
o
th
L
STM
m
o
d
els
wer
e
tr
ain
ed
u
s
in
g
W
o
r
d
2
V
ec
f
ea
tu
r
e
ex
tr
ac
tio
n
.
Af
ter
tr
ain
in
g
,
ea
ch
m
o
d
el
was
ev
alu
ate
d
o
n
th
e
test
s
et,
an
d
th
e
class
if
icatio
n
ac
cu
r
ac
y
alo
n
g
with
th
e
co
n
f
u
s
io
n
m
atr
i
x
was
r
ec
o
r
d
e
d
.
On
ce
all
f
o
ld
s
wer
e
co
m
p
leted
,
th
e
av
er
ag
e
ac
cu
r
ac
y
o
f
ea
ch
m
o
d
el
was
ca
lcu
lated
an
d
co
m
p
ar
ed
to
ass
ess
o
v
er
all
p
er
f
o
r
m
a
n
ce
ac
r
o
s
s
th
e
d
ataset.
Af
ter
all
f
o
ld
s
wer
e
co
m
p
leted
,
t
h
e
a
v
er
ag
e
ac
cu
r
ac
y
o
f
ea
c
h
m
o
d
el
was
co
m
p
a
r
ed
.
T
h
is
ev
alu
atio
n
aim
ed
to
d
eter
m
in
e
wh
eth
er
o
n
e
m
o
d
el
ty
p
e
s
ig
n
if
ican
tly
o
u
tp
er
f
o
r
m
ed
th
e
o
th
er
s
in
ter
m
s
o
f
class
if
icatio
n
ac
cu
r
ac
y
.
T
h
e
W
ilco
x
o
n
Sig
n
ed
-
R
an
k
T
est wa
s
th
en
u
s
ed
to
ex
am
in
e
th
e
s
tatis
tical
s
ig
n
if
ican
ce
b
etwe
en
:
-
ML
P
A
v
s
ML
P
B
-
L
STM
A
v
s
L
ST
M
B
-
T
h
e
b
es
t
ML
P
v
s
t
h
e
b
est
L
ST
M
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
is
s
ec
tio
n
p
r
esen
ts
an
d
d
is
cu
s
s
es
th
e
r
esu
lt
s
o
f
s
en
tim
en
t
class
if
ica
tio
n
ex
p
er
im
en
ts
u
s
in
g
ML
P
an
d
L
STM
m
o
d
els.
T
h
e
ev
a
lu
atio
n
f
o
cu
s
es
o
n
class
if
icatio
n
p
er
f
o
r
m
an
ce
b
ased
o
n
a
cc
u
r
ac
y
,
co
n
f
u
s
io
n
m
atr
ix
,
an
d
a
d
d
itio
n
al
m
etr
ics
s
u
ch
as
p
r
ec
is
io
n
,
r
ec
all,
an
d
F1
-
s
co
r
e.
I
n
ad
d
itio
n
,
we
c
o
n
d
u
cted
a
s
tatis
tical
test
to
d
eter
m
in
e
wh
eth
er
p
er
f
o
r
m
an
ce
d
if
f
er
en
ce
s
b
etwe
en
m
o
d
els
ar
e
s
ig
n
if
ican
t.
T
h
e
o
b
jectiv
e
is
to
ass
e
s
s
h
o
w
well
ea
ch
m
o
d
el
class
if
ies u
s
er
s
en
tim
en
t in
th
e
Mo
b
ile
J
KN
ap
p
r
ev
iews.
3
.
1
.
Da
t
a
prepro
ce
s
s
ing
re
s
ult
I
n
itially
,
th
e
d
ataset
co
n
tain
ed
2
0
0
,
0
0
0
u
s
er
r
ev
iews.
Af
ter
r
em
o
v
in
g
d
u
p
licate
en
tr
ies an
d
r
o
ws with
m
is
s
in
g
v
alu
es (
NaN
)
,
th
e
n
u
m
b
er
o
f
u
s
ab
le
d
ata
p
o
in
ts
d
ec
r
ea
s
ed
to
1
1
4
,
3
6
4
.
T
h
is
d
ata
r
ed
u
ctio
n
r
ef
lects th
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
5
2
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
42
,
No
.
3
,
J
u
n
e
20
2
6
:
8
1
8
-
82
6
822
im
p
o
r
tan
ce
o
f
th
e
clea
n
in
g
p
r
o
ce
s
s
in
im
p
r
o
v
i
n
g
d
ata
q
u
alit
y
an
d
e
n
s
u
r
in
g
th
at
o
n
ly
v
alid
r
ev
iews
wer
e
u
s
ed
in
th
e
a
n
aly
s
is
.
T
h
e
d
ata
clea
n
in
g
p
r
o
ce
s
s
also
af
f
ec
te
d
th
e
d
is
tr
ib
u
tio
n
o
f
s
en
tim
en
t
cla
s
s
es
in
th
e
d
ataset.
Sev
er
al
r
ev
iews
wer
e
r
em
o
v
ed
u
n
ev
e
n
ly
ac
r
o
s
s
s
en
tim
en
t
ca
teg
o
r
ies,
wh
ich
alter
ed
th
e
o
r
ig
in
al
b
alan
ce
b
etwe
en
p
o
s
itiv
e,
n
eg
ati
v
e
,
a
n
d
n
eu
tr
al
r
ev
iews.
As
a
r
esu
lt,
th
e
clea
n
ed
d
ataset
ex
h
i
b
ited
a
s
h
if
t
i
n
class
p
r
o
p
o
r
tio
n
s
co
m
p
ar
ed
to
th
e
o
r
ig
in
al
d
ata.
As
illu
s
tr
ated
in
Fig
u
r
e
2
,
p
o
s
itiv
e
r
ev
iews
ex
p
er
ien
ce
d
th
e
lar
g
est
r
ed
u
ctio
n
,
d
ec
r
ea
s
in
g
f
r
o
m
1
2
1
,
8
1
3
t
o
4
2
,
8
0
0
s
am
p
les.
Neg
at
iv
e
r
ev
iews
wer
e
r
ed
u
c
ed
f
r
o
m
7
0
,
9
0
2
to
6
5
,
4
1
1
s
a
m
p
les,
wh
ile
n
eu
tr
al
r
ev
iews
s
h
o
wed
th
e
s
m
allest
ch
an
g
e,
d
ec
r
ea
s
in
g
f
r
o
m
7
,
2
1
5
to
6
,
1
5
3
s
am
p
les.
T
h
is
u
n
ev
en
r
ed
u
ctio
n
ac
r
o
s
s
s
en
tim
en
t c
lass
es c
au
s
ed
n
eg
ativ
e
r
ev
iews to
b
ec
o
m
e
th
e
d
o
m
in
an
t c
lass
in
th
e
cle
an
ed
d
ataset.
T
h
e
r
esu
ltin
g
class
im
b
alan
ce
was
ad
d
r
ess
ed
d
u
r
in
g
th
e
m
o
d
el
tr
ain
in
g
s
tag
e
b
y
ap
p
ly
in
g
a
r
an
d
o
m
o
v
er
s
am
p
lin
g
tech
n
i
q
u
e.
I
n
th
is
p
r
o
ce
s
s
,
th
e
n
eu
tr
al
a
n
d
p
o
s
itiv
e
s
en
tim
en
t
class
es
wer
e
tr
ea
ted
as
m
in
o
r
ity
class
es
an
d
wer
e
o
v
er
s
am
p
led
b
y
d
u
p
l
icatin
g
ex
is
tin
g
s
am
p
les
u
n
til
th
eir
s
izes
m
at
ch
ed
th
e
n
e
g
ativ
e
s
en
tim
en
t
class
.
T
h
is
s
tr
ateg
y
was
ad
o
p
ted
to
r
ed
u
ce
b
ias
to
war
d
th
e
m
ajo
r
ity
class
an
d
t
o
allo
w
th
e
m
o
d
els
to
lear
n
s
en
tim
en
t p
atter
n
s
m
o
r
e
ev
en
ly
.
Fig
u
r
e
2
.
Data
d
is
tr
ib
u
tio
n
b
e
f
o
r
e
an
d
af
ter
3.
2
.
M
o
del
t
ra
ini
ng
a
nd
ev
a
l
ua
t
io
n set
up
Af
ter
b
alan
cin
g
th
e
d
ataset,
we
tr
ain
ed
f
o
u
r
d
if
f
er
e
n
t
m
o
d
els:
ML
P
A,
ML
P
B
,
L
STM
A,
an
d
L
STM
B
.
E
ac
h
m
o
d
el
was
e
v
alu
ated
u
s
in
g
5
-
f
o
ld
cr
o
s
s
-
v
ali
d
atio
n
to
en
s
u
r
e
r
o
b
u
s
t
an
d
g
e
n
er
ali
za
b
le
r
esu
lts
.
T
h
e
5
-
f
o
ld
cr
o
s
s
-
v
alid
atio
n
a
p
p
r
o
ac
h
d
i
v
id
es
th
e
d
ata
in
to
f
iv
e
eq
u
al
p
a
r
ts
,
wh
er
e
ea
ch
f
o
ld
s
er
v
es
as
a
test
s
et
wh
ile
th
e
r
em
ai
n
in
g
f
o
u
r
f
o
ld
s
ar
e
u
s
ed
f
o
r
tr
ai
n
in
g
.
T
h
e
av
er
ag
e
co
n
f
u
s
io
n
m
atr
ix
r
e
s
u
lts
ac
r
o
s
s
all
f
iv
e
f
o
ld
s
,
s
h
o
win
g
h
o
w
ea
ch
m
o
d
el
class
if
ied
s
am
p
les in
ea
ch
s
en
tim
en
t c
ateg
o
r
y
ca
n
b
e
d
esc
r
ib
ed
in
T
a
b
le
3
.
T
ab
le
3
s
h
o
ws
th
e
class
if
icat
io
n
p
e
r
f
o
r
m
an
ce
o
f
ea
ch
m
o
d
el
ac
r
o
s
s
th
e
th
r
ee
s
en
tim
en
t
class
es:
n
eg
ativ
e,
n
eu
tr
al,
an
d
p
o
s
itiv
e
.
W
h
en
lo
o
k
in
g
at
n
eg
ativ
e
s
en
tim
en
t,
ML
P
m
o
d
els
wo
r
k
ed
b
etter
th
a
n
L
STM
m
o
d
els.
ML
P
B
co
r
r
ec
tly
id
en
tifie
d
1
1
,
9
7
0
n
eg
ativ
e
r
e
v
i
ew
s
,
wh
ich
was
th
e
b
est
r
es
u
lt,
wh
ile
ML
P
A
id
en
tifie
d
1
1
,
7
5
4
n
eg
ativ
e
r
ev
iews.
T
h
e
L
STM
m
o
d
els
d
id
n
o
t
p
er
f
o
r
m
as
well,
esp
ec
ially
L
STM
A,
wh
ich
in
co
r
r
ec
tly
la
b
eled
m
a
n
y
n
eg
a
tiv
e
r
ev
iews
as
n
eu
tr
al
(
3
,
2
9
6
r
ev
iews
)
.
T
h
is
s
h
o
ws
th
at
ML
P
m
o
d
els
ar
e
b
etter
at
r
ec
o
g
n
izin
g
n
eg
ati
v
e
s
en
tim
en
ts
.
Fo
r
n
eu
tr
al
s
en
tim
en
t,
all
m
o
d
els
h
ad
d
if
f
icu
lty
,
wh
ich
m
ak
es
s
en
s
e
b
ec
au
s
e
n
e
u
tr
al
f
e
elin
g
s
ar
e
h
ar
d
er
to
d
etec
t.
Ho
wev
er
,
L
STM
m
o
d
els
p
er
f
o
r
m
e
d
s
lig
h
tly
b
etter
h
e
r
e.
L
STM
A
c
o
r
r
ec
tly
id
en
tifie
d
6
2
2
n
eu
tr
al
r
e
v
iews,
wh
ich
w
as
m
u
ch
b
etter
th
a
n
ML
P
A
(
8
1
)
a
n
d
ML
P
B
(
7
1
)
.
T
h
is
s
u
g
g
ests
th
at
L
STM
m
o
d
els
m
ig
h
t b
e
b
etter
at
u
n
d
er
s
tan
d
i
n
g
s
u
b
tle
n
e
u
tr
al
s
en
tim
en
ts
b
ec
au
s
e
th
ey
ca
n
p
r
o
ce
s
s
tex
t i
n
s
eq
u
en
ce
.
T
ab
le
3
.
Av
e
r
ag
e
co
n
f
u
s
s
io
n
m
atr
ix
P
r
e
d
i
c
t
e
d
l
a
b
e
l
N
e
g
a
t
i
v
e
N
e
u
t
r
a
l
P
o
si
t
i
v
e
Tr
u
e
l
a
b
e
l
:
n
e
g
a
t
i
v
e
M
LP A
1
1
7
5
4
4
0
3
9
2
3
M
LP
B
1
1
9
7
0
3
5
9
7
5
2
LSTM
A
9
1
9
3
3
2
9
6
5
9
2
LSTM
B
9
3
6
0
2
9
5
8
7
6
3
Tr
u
e
l
a
b
e
l
:
n
e
u
t
r
a
l
M
LP A
8
7
7
81
2
7
2
M
LP
B
9
1
6
71
2
4
2
LSTM
A
4
5
1
6
2
2
1
5
7
LSTM
B
5
6
4
4
5
2
2
1
4
Tr
u
e
l
a
b
e
l
:
p
o
s
i
t
i
v
e
M
LP A
1
1
1
7
1
5
0
7
2
9
2
M
LP
B
1
2
7
0
1
4
0
7
1
4
8
LSTM
A
5
6
8
1
0
5
1
6
9
4
0
LSTM
B
7
8
6
7
7
9
6
9
9
4
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4
7
5
2
S
ta
tis
tica
l c
o
mp
a
r
is
o
n
o
f MLP
a
n
d
LS
TM
fo
r
mo
b
ile
h
e
a
lth
s
en
timen
t a
n
a
lysi
s
(
Gh
a
n
im
K
a
n
u
g
r
a
h
a
n
)
823
W
h
en
class
if
y
in
g
p
o
s
itiv
e
s
e
n
tim
en
t,
ML
P
m
o
d
els
ag
ain
p
er
f
o
r
m
ed
b
etter
th
a
n
L
ST
M
m
o
d
els.
MLP
A
h
ad
th
e
h
ig
h
est
n
u
m
b
er
o
f
c
o
r
r
ec
t
p
r
ed
ictio
n
s
(
7
,
2
9
2
)
,
f
o
llo
wed
b
y
ML
P
B
(
7
,
1
4
8
)
.
L
STM
A
o
f
ten
co
n
f
u
s
ed
p
o
s
itiv
e
r
ev
iews
with
n
eu
tr
al
o
n
es
(
1
,
0
5
1
tim
es),
wh
ich
m
ad
e
its
o
v
er
all
p
e
r
f
o
r
m
an
ce
lo
we
r
f
o
r
p
o
s
itiv
e
s
en
tim
en
t c
lass
if
icati
o
n
.
To
b
etter
u
n
d
e
r
s
tan
d
th
e
p
er
f
o
r
m
an
ce
o
f
ea
ch
m
o
d
el,
f
o
u
r
ev
alu
atio
n
m
etr
ics
wer
e
c
alcu
lated
,
n
am
ely
ac
c
u
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
an
d
F1
-
s
co
r
e
u
s
in
g
m
ac
r
o
av
e
r
ag
in
g
.
Ma
cr
o
a
v
er
a
g
in
g
ass
ig
n
s
eq
u
al
im
p
o
r
tan
ce
to
ea
c
h
class
an
d
is
th
er
ef
o
r
e
s
u
itab
le
f
o
r
im
b
al
an
c
ed
d
atasets
.
T
h
e
av
er
ag
e
p
er
f
o
r
m
a
n
ce
r
esu
lts
ac
r
o
s
s
f
iv
e
-
f
o
ld
cr
o
s
s
-
v
alid
atio
n
,
alo
n
g
with
th
e
s
tan
d
ar
d
d
ev
iatio
n
o
f
ac
c
u
r
ac
y
to
r
e
f
lect
p
er
f
o
r
m
a
n
ce
s
tab
ilit
y
ac
r
o
s
s
f
o
ld
s
,
ar
e
p
r
es
en
ted
in
T
ab
le
4
.
B
ased
o
n
th
e
ac
cu
r
ac
y
r
esu
lts
,
th
e
ML
P
-
b
ased
m
o
d
els
ac
h
iev
ed
h
ig
h
e
r
o
v
er
all
ac
cu
r
ac
y
th
an
b
o
th
L
STM
m
o
d
els,
with
ML
P
B
s
lig
h
tly
o
u
tp
er
f
o
r
m
i
n
g
M
L
P
A.
T
h
e
ac
cu
r
ac
y
d
if
f
er
en
ce
b
etwe
en
th
e
b
est
-
p
er
f
o
r
m
in
g
ML
P
m
o
d
el
(
ML
P
B
)
an
d
th
e
b
est
-
p
er
f
o
r
m
in
g
L
STM
m
o
d
el
(
L
STM
B
)
was
ap
p
r
o
x
im
ately
1
0
%.
Ho
wev
er
,
a
d
if
f
e
r
en
t p
at
ter
n
ca
n
b
e
o
b
s
er
v
e
d
wh
en
ex
am
in
in
g
p
r
ec
is
io
n
,
r
ec
all,
a
n
d
F1
-
s
co
r
e.
L
STM
A
ac
h
iev
ed
th
e
h
ig
h
est
r
ec
all
v
alu
e
o
f
6
7
.
3
1
%
an
d
th
e
h
ig
h
est
p
r
ec
is
io
n
o
f
6
4
.
2
7
%
,
in
d
icatin
g
th
at
th
is
m
o
d
el
w
as
m
o
r
e
ef
f
ec
tiv
e
at
co
r
r
ec
tl
y
id
en
tify
in
g
r
elev
a
n
t
s
am
p
l
es
ac
r
o
s
s
s
en
tim
en
t
class
es.
Desp
ite
h
av
in
g
lo
wer
o
v
er
all
ac
cu
r
ac
y
,
th
is
r
esu
lt
s
u
g
g
ests
th
at
th
e
L
STM
-
b
ase
d
m
o
d
el
m
ay
b
etter
ca
p
tu
r
e
p
atter
n
s
f
r
o
m
m
in
o
r
ity
class
es,
p
ar
ticu
lar
ly
th
e
n
eu
tr
al
s
en
tim
en
t c
ateg
o
r
y
.
T
h
e
o
b
s
er
v
e
d
p
e
r
f
o
r
m
an
ce
d
if
f
er
en
ce
s
b
etwe
en
th
e
ML
P
an
d
L
STM
m
o
d
els
ca
n
b
e
f
u
r
t
h
er
ex
p
lain
ed
b
y
th
e
n
atu
r
e
o
f
th
e
r
ev
iew
tex
ts
in
th
e
d
ataset.
Mo
s
t
u
s
er
r
ev
iews
in
th
e
Mo
b
ile
J
KN
ap
p
licatio
n
ar
e
r
elativ
ely
s
h
o
r
t
an
d
co
n
tain
lim
ited
co
n
tex
tu
al
in
f
o
r
m
a
tio
n
.
I
n
s
u
ch
ca
s
es,
T
F
-
I
DF
r
ep
r
esen
tatio
n
s
ar
e
ef
f
ec
tiv
e
in
h
ig
h
lig
h
tin
g
d
is
cr
im
in
ativ
e
k
e
y
wo
r
d
s
ass
o
ciat
ed
with
s
en
tim
en
t,
wh
ich
b
e
n
ef
its
ML
P
-
b
ased
class
if
ier
s
d
e
s
p
ite
th
eir
s
im
p
le
r
ar
ch
itectu
r
e.
On
th
e
o
th
er
h
a
n
d
,
L
STM
m
o
d
els
r
ely
o
n
s
eq
u
en
tial
in
f
o
r
m
atio
n
an
d
d
is
tr
ib
u
ted
wo
r
d
r
ep
r
esen
tatio
n
s
.
Alth
o
u
g
h
th
is
ap
p
r
o
a
ch
m
ay
n
o
t
m
ax
im
ize
o
v
er
all
ac
cu
r
ac
y
o
n
s
h
o
r
t
tex
ts
,
it
en
ab
les
th
e
m
o
d
el
to
b
etter
ca
p
tu
r
e
s
u
b
tle
s
em
an
tic
p
atter
n
s
,
p
ar
ticu
lar
ly
f
o
r
m
in
o
r
ity
class
es
s
u
ch
as
n
eu
tr
al
s
en
tim
en
t.
T
h
is
ex
p
lain
s
wh
y
L
STM
A
ac
h
iev
e
d
h
ig
h
er
r
ec
all
an
d
p
r
ec
is
io
n
v
al
u
es,
ev
en
t
h
o
u
g
h
its
ac
cu
r
ac
y
was lo
wer
th
a
n
th
at
o
f
th
e
ML
P m
o
d
els.
T
h
ese
f
in
d
in
g
s
h
ig
h
lig
h
t
a
t
r
ad
e
-
o
f
f
b
etwe
en
ac
h
iev
i
n
g
h
ig
h
o
v
e
r
all
ac
cu
r
ac
y
a
n
d
e
f
f
ec
tiv
ely
id
en
tify
in
g
u
n
d
er
r
ep
r
esen
ted
s
en
tim
en
t
class
e
s
.
T
h
er
e
f
o
r
e,
th
e
ch
o
ice
o
f
m
o
d
el
s
h
o
u
l
d
b
e
alig
n
ed
with
th
e
ap
p
licatio
n
o
b
jectiv
e.
Fo
r
lar
g
e
-
s
ca
le
s
en
tim
en
t
m
o
n
ito
r
in
g
wh
er
e
ac
c
u
r
ac
y
is
p
r
io
r
itize
d
,
ML
P
m
o
d
els
ar
e
m
o
r
e
s
u
itab
le.
Ho
wev
er
,
f
o
r
a
p
p
licatio
n
s
th
at
r
eq
u
ir
e
d
ee
p
er
an
aly
s
is
o
f
n
eu
tr
al
o
r
am
b
ig
u
o
u
s
u
s
e
r
f
ee
d
b
ac
k
,
L
STM
-
b
ased
m
o
d
els m
ay
o
f
f
e
r
ad
d
itio
n
al
a
d
v
an
tag
es.
T
ab
le
4
.
Av
e
r
ag
e
s
co
r
e
M
LP A
M
LP
B
LSTM
A
LSTM
B
A
c
c
u
r
a
c
y
83
.
6
3
±
0
.
2
8
%
83
.
9
0
±
0
.
1
8
%
73
.
26
±
0
.
6
9
%
73
.
4
8
±
0
.
9
0
%
P
r
e
c
i
s
i
o
n
61
.
4
0
%
61
.
6
0
%
64
.
2
7
%
61
.
9
8
%
R
e
c
a
l
l
60
.
5
5
%
60
.
2
7
%
67
.
3
1
%
63
.
3
4
%
F1
60
.
6
2
%
60
.
4
6
%
61
.
4
8
%
60
.
0
0
%
3.
3
.
St
a
t
is
t
ica
l
s
ig
nifica
nce
t
esting
T
o
d
eter
m
in
e
wh
et
h
er
t
h
e
o
b
s
er
v
ed
p
er
f
o
r
m
an
ce
d
if
f
e
r
e
n
ce
s
wer
e
s
tatis
tically
m
ea
n
in
g
f
u
l,
we
co
n
d
u
cte
d
W
ilco
x
o
n
s
ig
n
ed
-
r
an
k
test
s
co
m
p
ar
i
n
g
th
e
m
o
d
els.
T
h
is
test
is
s
u
itab
le
f
o
r
co
m
p
ar
i
n
g
p
air
ed
s
am
p
les,
s
u
ch
as e
v
alu
atio
n
s
co
r
es f
r
o
m
c
r
o
s
s
-
v
alid
atio
n
r
esu
l
ts
.
T
h
e
f
ir
s
t set o
f
test
s
was
co
n
d
u
cte
d
u
s
in
g
th
e
ac
cu
r
ac
y
s
co
r
es
o
b
tain
e
d
f
r
o
m
ea
ch
o
f
t
h
e
f
iv
e
f
o
l
d
s
.
T
h
e
ev
alu
atio
n
was
co
n
d
u
cted
i
n
th
r
ee
s
tag
es.
First,
we
co
m
p
ar
ed
th
e
two
ML
P
m
o
d
els
(
ML
P
A
v
s
ML
P
B
)
to
ass
es
s
wh
eth
er
th
e
ad
d
itio
n
al
h
id
d
en
lay
er
in
ML
P_
B
led
to
a
s
ig
n
if
ican
t
p
er
f
o
r
m
a
n
ce
g
ain
.
Seco
n
d
,
we
co
m
p
a
r
ed
th
e
two
L
STM
m
o
d
els
(
L
STM
A
v
s
L
STM
B
)
to
ev
alu
ate
wh
eth
er
th
e
d
ee
p
er
L
STM
B
o
u
tp
er
f
o
r
m
ed
th
e
s
im
p
ler
L
STM
A
.
Fin
ally
,
th
e
b
est
-
p
er
f
o
r
m
in
g
m
o
d
el
f
r
o
m
ea
ch
g
r
o
u
p
,
MLP
B
an
d
L
STM
B
,
was
co
m
p
ar
e
d
to
d
eter
m
in
in
g
wh
ich
ar
ch
itectu
r
e
o
f
f
er
ed
b
etter
o
v
er
all
p
er
f
o
r
m
an
ce
in
s
en
tim
en
t c
lass
if
icatio
n
.
T
h
e
r
esu
lts
ar
e
p
r
esen
ted
in
T
ab
le
5
.
As
s
h
o
wn
in
T
ab
le
5
,
th
e
d
if
f
er
en
ce
s
b
etwe
en
m
o
d
el
A
an
d
m
o
d
el
B
with
in
th
e
s
am
e
a
r
ch
itectu
r
e
f
am
ily
wer
e
n
o
t
s
tatis
tically
s
ig
n
if
ican
t,
as
b
o
th
p
-
v
alu
es
(
0
.
4
3
7
5
a
n
d
0
.
8
1
2
5
)
wer
e
ab
o
v
e
th
e
s
tan
d
a
r
d
th
r
esh
o
ld
o
f
0
.
0
5
.
T
h
is
in
d
ic
ates
th
at
ad
d
in
g
ad
d
itio
n
al
lay
er
s
o
r
co
m
p
lex
ity
in
th
e
B
v
ar
ian
ts
d
id
n
o
t
co
n
s
is
ten
tly
im
p
r
o
v
e
m
o
d
el
p
er
f
o
r
m
a
n
ce
ac
r
o
s
s
th
e
f
o
ld
s
.
T
h
e
co
m
p
ar
is
o
n
b
etwe
en
t
h
e
b
est
ML
P
m
o
d
el
(
ML
P
B
)
an
d
th
e
b
est
L
STM
m
o
d
el
(
L
STM
B
)
r
es
u
lted
in
a
p
-
v
alu
e
o
f
0
.
0
6
2
5
,
w
h
ich
is
s
lig
h
tly
ab
o
v
e
th
e
0
.
0
5
th
r
esh
o
ld
.
Alth
o
u
g
h
th
is
r
esu
lt
is
n
o
t
s
tatis
t
ically
s
ig
n
if
ican
t,
it
i
s
cl
o
s
e
to
th
e
b
o
u
n
d
a
r
y
,
s
u
g
g
esti
n
g
th
at
ML
P
B
m
ay
h
av
e
b
etter
p
er
f
o
r
m
a
n
ce
th
an
L
STM
B
.
W
ith
a
lar
g
er
d
atase
t
o
r
m
o
r
e
v
alid
atio
n
f
o
ld
s
,
th
is
d
if
f
er
en
ce
m
ig
h
t
r
ea
ch
s
tatis
ti
ca
l
s
ig
n
if
ican
ce
in
f
u
tu
r
e
r
esear
c
h
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
5
2
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
42
,
No
.
3
,
J
u
n
e
20
2
6
:
8
1
8
-
82
6
824
T
ab
le
5
.
T
h
e
r
esu
lt u
s
in
g
w
ilc
o
x
o
n
si
g
n
ed
-
r
a
n
k
t
est
C
o
m
p
a
r
i
so
n
p
-
v
a
l
u
e
S
i
g
n
i
f
i
c
a
n
t
M
LP A
v
s
M
LP
B
0
.
4
3
7
5
No
LSTM
A
v
s
LST
M
B
0
.
8
1
2
5
No
M
LP
B
v
s L
S
TM
B
0
.
0
6
2
5
No
4.
CO
NCLU
SI
O
N
T
h
is
s
tu
d
y
co
m
p
a
r
ed
th
e
p
er
f
o
r
m
an
ce
o
f
ML
P
an
d
L
STM
m
o
d
els
f
o
r
s
en
tim
en
t
class
if
icatio
n
o
n
u
s
er
r
ev
iews
o
f
th
e
Mo
b
ile
J
KN
ap
p
.
T
h
e
r
esu
lts
s
h
o
wed
t
h
at
ML
P
m
o
d
els
u
s
in
g
T
F
-
I
DF
f
ea
tu
r
es
ac
h
iev
ed
h
ig
h
er
o
v
e
r
all
class
if
icatio
n
a
cc
u
r
ac
y
,
wh
ile
L
STM
m
o
d
els
with
W
o
r
d
2
Vec
e
m
b
ed
d
in
g
s
d
em
o
n
s
tr
ated
b
etter
p
r
ec
is
io
n
an
d
r
ec
all,
p
ar
ticu
l
ar
ly
f
o
r
n
eu
tr
al
s
en
tim
en
t.
Ho
wev
er
,
s
tatis
tical
test
in
g
u
s
in
g
th
e
W
ilco
x
o
n
s
ig
n
ed
-
r
an
k
test
in
d
icate
d
th
at
th
e
o
b
s
er
v
ed
p
er
f
o
r
m
a
n
ce
d
if
f
er
en
ce
s
wer
e
n
o
t
s
tatis
t
ically
s
ig
n
if
ican
t.
T
h
ese
f
in
d
in
g
s
s
u
g
g
est th
at
b
o
th
ML
P a
n
d
L
STM
ap
p
r
o
ac
h
e
s
ar
e
co
m
p
ar
ab
ly
v
iab
le
f
o
r
s
e
n
tim
en
t a
n
aly
s
is
o
f
s
h
o
r
t u
s
er
-
g
e
n
er
ated
tex
t i
n
th
i
s
ap
p
licatio
n
co
n
tex
t.
F
o
r
f
u
tu
r
e
wo
r
k
,
m
o
r
e
ad
v
a
n
ce
d
f
ea
t
u
r
e
r
ep
r
esen
tatio
n
s
s
u
ch
as
co
n
tex
tu
al
w
o
r
d
e
m
b
ed
d
in
g
s
(
e.
g
.
,
I
n
d
o
B
E
R
T
)
ca
n
b
e
e
x
p
l
o
r
ed
t
o
b
etter
ca
p
tu
r
e
s
em
an
t
ic
in
f
o
r
m
atio
n
i
n
s
h
o
r
t
an
d
a
m
b
ig
u
o
u
s
r
ev
iews.
I
n
ad
d
itio
n
,
alter
n
ativ
e
tech
n
i
q
u
es
f
o
r
h
a
n
d
lin
g
class
im
b
a
la
n
ce
,
s
u
ch
as
s
y
n
th
etic
o
v
er
s
am
p
lin
g
m
eth
o
d
s
,
m
ay
b
e
in
v
esti
g
ated
to
im
p
r
o
v
e
m
in
o
r
ity
class
r
ec
o
g
n
itio
n
.
Fin
ally
,
ex
p
an
d
i
n
g
th
e
d
ataset
o
r
in
co
r
p
o
r
atin
g
m
o
d
el
in
ter
p
r
etab
ilit
y
an
aly
s
is
co
u
ld
p
r
o
v
id
e
d
ee
p
er
i
n
s
ig
h
ts
an
d
p
o
ten
tially
lead
to
m
o
r
e
s
tatis
tically
s
ig
n
if
ican
t p
er
f
o
r
m
an
ce
d
if
f
er
en
ce
s
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
T
h
is
r
esear
ch
was
s
u
p
p
o
r
ted
b
y
Satu
Un
iv
er
s
ity
an
d
B
in
u
s
Un
iv
er
s
ity
,
b
o
th
o
f
wh
ich
ar
e
p
a
r
t
o
f
B
in
u
s
Hig
h
er
E
d
u
ca
tio
n
.
T
h
e
au
th
o
r
s
ac
k
n
o
wled
g
e
t
h
e
in
s
titu
tio
n
al
s
u
p
p
o
r
t p
r
o
v
id
e
d
d
u
r
in
g
th
is
s
tu
d
y
.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
T
h
is
jo
u
r
n
al
u
s
es
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT)
to
r
ec
o
g
n
ize
in
d
iv
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
llab
o
r
atio
n
.
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
Gh
an
im
Kan
u
g
r
ah
an
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
W
in
C
e
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
Vito
Haf
izh
C
ah
ay
a
Pu
tr
a
✓
✓
✓
✓
✓
✓
✓
✓
Yu
d
i Ram
d
h
an
i
✓
✓
✓
✓
✓
✓
✓
✓
✓
Feb
r
iy
an
ti Pan
jaitan
✓
✓
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
T
h
e
au
th
o
r
s
s
tate
n
o
co
n
f
lict o
f
in
ter
est.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
d
ata
th
at
s
u
p
p
o
r
t
th
e
f
in
d
i
n
g
s
o
f
th
is
s
tu
d
y
ar
e
av
ailab
le
f
r
o
m
t
h
e
c
o
r
r
esp
o
n
d
in
g
au
th
o
r
,
W
in
C
e,
u
p
o
n
r
ea
s
o
n
ab
le
r
eq
u
est.
RE
F
E
R
E
NC
E
S
[
1
]
S
.
G
u
n
a
t
h
i
l
a
k
a
a
n
d
N
.
D
e
S
i
l
v
a
,
“
A
sp
e
c
t
-
b
a
se
d
se
n
t
i
m
e
n
t
a
n
a
l
y
si
s
o
n
m
o
b
i
l
e
a
p
p
l
i
c
a
t
i
o
n
r
e
v
i
e
w
s,
”
i
n
2
0
2
2
2
2
n
d
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
Ad
v
a
n
c
e
s
i
n
I
C
T
f
o
r
Em
e
r
g
i
n
g
Re
g
i
o
n
s
(
I
C
T
e
r)
,
I
EEE,
N
o
v
.
2
0
2
2
,
p
p
.
1
8
3
-
1
8
8
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
Te
r
5
8
0
6
3
.
2
0
2
2
.
1
0
0
2
4
0
7
0
.
[
2
]
A
.
S
u
b
b
a
r
a
o
,
A
.
S
i
d
d
i
k
a
,
M
.
A
.
F
a
t
h
u
l
l
a
h
,
a
n
d
M
.
A
.
B
i
n
S
a
n
w
a
n
i
,
“
Th
e
r
o
l
e
o
f
mo
b
i
l
e
a
p
p
l
i
c
a
t
i
o
n
s
i
n
sh
a
p
i
n
g
d
i
g
i
t
a
l
t
r
a
n
sf
o
r
ma
t
i
o
n
i
n
h
i
g
h
e
r
e
d
u
c
a
t
i
o
n
a
mo
n
g
g
e
n
e
r
a
t
i
o
n
i
:
a
b
i
b
l
i
o
g
r
a
p
h
i
c
st
u
d
y
,
”
I
n
f
o
rm
a
t
i
o
n
,
v
o
l
.
1
6
,
n
o
.
1
2
,
p
.
1
0
2
6
,
N
o
v
.
2
0
2
5
,
d
o
i
:
1
0
.
3
3
9
0
/
i
n
f
o
1
6
1
2
1
0
2
6
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4
7
5
2
S
ta
tis
tica
l c
o
mp
a
r
is
o
n
o
f MLP
a
n
d
LS
TM
fo
r
mo
b
ile
h
e
a
lth
s
en
timen
t a
n
a
lysi
s
(
Gh
a
n
im
K
a
n
u
g
r
a
h
a
n
)
825
[
3
]
J.
Jo
s
e
p
h
,
S
.
V
i
n
e
e
t
h
a
,
a
n
d
N
.
V
.
S
o
b
h
a
n
a
,
“
A
su
r
v
e
y
o
n
d
e
e
p
l
e
a
r
n
i
n
g
b
a
s
e
d
s
e
n
t
i
m
e
n
t
a
n
a
l
y
si
s,
”
M
a
t
e
ri
a
l
s
T
o
d
a
y
:
Pr
o
c
e
e
d
i
n
g
s
,
v
o
l
.
5
8
,
p
p
.
4
5
6
–
4
6
0
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
ma
t
p
r
.
2
0
2
2
.
0
2
.
4
8
3
.
[
4
]
K
.
T
.
S
h
a
n
d
a
n
a
,
A
.
A
m
i
n
u
d
d
i
n
,
E
.
H
.
S
a
p
u
t
r
a
,
F
.
F
.
A
b
d
u
l
l
o
h
,
M
.
R
a
h
a
r
d
i
,
a
n
d
B
.
P
.
A
s
a
d
d
u
l
l
o
h
,
“
S
e
n
t
i
m
e
n
t
a
n
a
l
y
s
i
s
o
f
g
o
o
g
l
e
c
l
a
s
s
r
o
o
m
a
p
p
l
i
c
a
t
i
o
n
u
s
i
n
g
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s
,
”
i
n
2
0
2
3
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
A
d
v
a
n
c
e
d
M
e
c
h
a
t
r
o
n
i
c
s
,
I
n
t
e
l
l
i
g
e
n
t
M
a
n
u
f
a
c
t
u
r
e
a
n
d
I
n
d
u
s
t
r
i
a
l
A
u
t
o
m
a
t
i
o
n
(
I
C
A
M
I
M
I
A
)
,
I
E
E
E
,
N
o
v
.
2
0
2
3
,
p
p
.
9
5
4
-
9
5
9
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
A
M
I
M
I
A
6
0
8
8
1
.
2
0
2
3
.
1
0
4
2
7
7
0
6
.
[
5
]
N
.
I
.
Le
st
a
r
i
,
S
.
M
.
T
a
i
b
,
W
.
W
i
b
o
w
o
,
I
.
A
.
A
z
i
z
,
a
n
d
M
.
R
.
H
a
b
i
b
i
,
“
A
s
p
e
c
t
-
b
a
se
d
se
n
t
i
m
e
n
t
a
n
a
l
y
s
i
s
f
o
r
m
o
b
i
l
e
a
p
p
r
e
v
i
e
w
u
si
n
g
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
(
C
N
N
)
a
n
d
W
o
r
d
2
V
e
c
,
”
i
n
2
0
2
4
I
EEE
7
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
El
e
c
t
ri
c
a
l
,
El
e
c
t
r
o
n
i
c
s
a
n
d
S
y
s
t
e
m
E
n
g
i
n
e
e
ri
n
g
(
I
C
EE
S
E)
,
I
EEE,
N
o
v
.
2
0
2
4
,
p
p
.
1
–
6
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
EESE
6
2
3
1
5
.
2
0
2
4
.
1
0
8
2
8
5
4
1
.
[
6
]
A
.
A
ma
l
i
a
,
D
.
G
u
n
a
w
a
n
,
a
n
d
K
.
N
a
su
t
i
o
n
,
“
S
e
n
t
i
m
e
n
t
a
n
a
l
y
si
s
o
f
G
O
-
J
EK
ser
v
i
c
e
s
q
u
a
l
i
t
y
u
s
i
n
g
mu
l
t
i
-
l
a
b
e
l
c
l
a
ss
i
f
i
c
a
t
i
o
n
,
”
J
o
u
rn
a
l
o
f
Ph
y
si
c
s:
C
o
n
f
e
re
n
c
e
S
e
ri
e
s
,
v
o
l
.
1
8
3
0
,
n
o
.
1
,
p
.
0
1
2
0
0
3
,
A
p
r
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
8
8
/
1
7
4
2
-
6
5
9
6
/
1
8
3
0
/
1
/
0
1
2
0
0
3
.
[
7
]
T.
M
e
s
c
h
e
r
,
R
.
L.
H
a
c
k
e
r
,
L
.
A
.
M
a
r
t
i
n
e
z
,
C
.
D
.
M
o
r
r
i
s
,
M
.
C
.
M
i
s
h
k
i
n
d
,
a
n
d
C
.
E
.
G
a
r
v
e
r
-
A
p
g
a
r
,
“
M
o
b
i
l
e
h
e
a
l
t
h
a
p
p
s:
g
u
i
d
a
n
c
e
f
o
r
e
v
a
l
u
a
t
i
o
n
a
n
d
i
mp
l
e
m
e
n
t
a
t
i
o
n
b
y
h
e
a
l
t
h
c
a
r
e
w
o
r
k
e
r
s
,
”
J
o
u
r
n
a
l
o
f
T
e
c
h
n
o
l
o
g
y
i
n
B
e
h
a
v
i
o
ra
l
S
c
i
e
n
c
e
,
v
o
l
.
1
0
,
n
o
.
2
,
p
p
.
2
2
4
-
2
3
5
,
S
e
p
.
2
0
2
5
,
d
o
i
:
1
0
.
1
0
0
7
/
s4
1
3
4
7
-
0
2
4
-
0
0
4
4
1
-
7.
[
8
]
K
a
r
t
i
n
i
,
A
.
K
.
D
a
r
maw
a
n
,
R
.
I
.
S
y
a
h
,
a
n
d
M
.
M
a
k
r
u
f
,
“
S
e
n
t
i
me
n
t
a
n
a
l
y
si
s
o
f
so
c
i
a
l
m
e
d
i
a
f
o
r
i
n
d
o
n
e
si
a
n
m
-
he
a
l
t
h
P
e
d
u
l
i
Li
n
d
u
n
g
i
mo
b
i
l
e
-
a
p
p
s(
P
L
M
A
)
w
i
t
h
l
e
x
i
c
o
n
-
b
a
s
e
d
a
n
d
su
p
p
o
r
t
v
e
c
t
o
r
ma
c
h
i
n
e
a
p
p
r
o
a
c
h
,
”
i
n
2
0
2
3
I
EE
E
9
t
h
I
n
f
o
rm
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
I
n
t
e
rn
a
t
i
o
n
a
l
S
e
m
i
n
a
r (I
T
I
S
)
,
I
EEE,
O
c
t
.
2
0
2
3
,
p
p
.
1
-
7
.
d
o
i
:
1
0
.
1
1
0
9
/
I
TI
S
5
9
6
5
1
.
2
0
2
3
.
1
0
4
1
9
9
9
4
.
[
9
]
M
.
H
a
o
u
e
s,
R
.
M
o
k
n
i
,
a
n
d
A
.
S
e
l
l
a
mi
,
“
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
f
o
r
mH
e
a
l
t
h
a
p
p
s
q
u
a
l
i
t
y
e
v
a
l
u
a
t
i
o
n
:
a
n
a
p
p
r
o
a
c
h
b
a
se
d
o
n
u
s
e
r
f
e
e
d
b
a
c
k
a
n
a
l
y
s
i
s,
”
S
o
f
t
w
a
re
Q
u
a
l
i
t
y
J
o
u
rn
a
l
,
v
o
l
.
3
1
,
n
o
.
4
,
p
p
.
1
1
7
9
–
1
2
0
9
,
D
e
c
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
2
1
9
-
023
-
0
9
6
3
0
-
8.
[
1
0
]
F
.
A
l
k
h
u
z
a
i
mi
,
D
.
R
a
i
n
e
y
,
C
.
B
r
o
w
n
W
i
l
so
n
,
a
n
d
J.
B
l
o
o
mf
i
e
l
d
,
“
T
h
e
i
m
p
a
c
t
o
f
m
o
b
i
l
e
h
e
a
l
t
h
i
n
t
e
r
v
e
n
t
i
o
n
s
o
n
s
e
r
v
i
c
e
u
sers
’
h
e
a
l
t
h
o
u
t
c
o
m
e
s
a
n
d
t
h
e
r
o
l
e
o
f
h
e
a
l
t
h
p
r
o
f
e
ss
i
o
n
s
:
a
sy
s
t
e
m
a
t
i
c
r
e
v
i
e
w
o
f
sy
s
t
e
mat
i
c
r
e
v
i
e
w
s
,
”
B
MC
D
i
g
i
t
a
l
H
e
a
l
t
h
,
v
o
l
.
3
,
n
o
.
1
,
p
.
3
,
F
e
b
.
2
0
2
5
,
d
o
i
:
1
0
.
1
1
8
6
/
s
4
4
2
4
7
-
0
2
4
-
0
0
1
4
3
-
3.
[
1
1
]
G
.
Ta
mam
i
,
W
.
A
.
Tr
i
y
a
n
t
o
,
a
n
d
S
.
M
u
z
i
d
,
“
S
e
n
t
i
me
n
t
a
n
a
l
y
si
s
m
o
b
i
l
e
J
K
N
r
e
v
i
e
w
s
u
si
n
g
S
M
O
TE
ba
s
e
d
LST
M
,
”
I
J
C
C
S
(
I
n
d
o
n
e
si
a
n
J
o
u
rn
a
l
o
f
C
o
m
p
u
t
i
n
g
a
n
d
C
y
b
e
r
n
e
t
i
c
s
S
y
st
e
m
s)
,
v
o
l
.
1
9
,
n
o
.
1
,
p
.
1
3
,
Ja
n
.
2
0
2
5
,
d
o
i
:
1
0
.
2
2
1
4
6
/
i
j
c
c
s.
1
0
1
9
1
0
.
[
1
2
]
E.
S
e
p
t
i
a
n
i
,
T.
M
.
A
k
h
r
i
z
a
,
a
n
d
M
.
H
u
s
n
i
,
“
C
o
m
p
a
r
i
so
n
o
f
t
h
e
a
c
c
u
r
a
c
y
b
e
t
w
e
e
n
n
a
i
v
e
b
a
y
e
s
c
l
a
ss
i
f
i
e
r
a
n
d
s
u
p
p
o
r
t
v
e
c
t
o
r
mac
h
i
n
e
a
l
g
o
r
i
t
h
m
s
f
o
r
se
n
t
i
m
e
n
t
a
n
a
l
y
s
i
s
i
n
mo
b
i
l
e
J
K
N
a
p
p
l
i
c
a
t
i
o
n
r
e
v
i
e
w
s,
”
T
ra
n
s
a
c
t
i
o
n
s
o
n
I
n
f
o
rm
a
t
i
c
s
a
n
d
D
a
t
a
S
c
i
e
n
c
e
,
v
o
l
.
1
,
n
o
.
1
,
p
p
.
2
1
–
3
2
,
A
p
r
.
2
0
2
4
,
d
o
i
:
1
0
.
2
4
0
9
0
/
t
i
d
s
.
v
1
i
1
.
1
2
2
3
2
.
[
1
3
]
A
.
A
.
Q
u
r
e
s
h
i
,
M
.
A
h
m
a
d
,
S
.
U
l
l
a
h
,
M
.
N
.
Y
a
si
r
,
F
.
R
u
st
a
m,
a
n
d
I
.
A
sh
r
a
f
,
“
P
e
r
f
o
r
ma
n
c
e
e
v
a
l
u
a
t
i
o
n
o
f
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
mo
d
e
l
s
o
n
l
a
r
g
e
d
a
t
a
se
t
o
f
a
n
d
r
o
i
d
a
p
p
l
i
c
a
t
i
o
n
s
r
e
v
i
e
w
s,
”
Mu
l
t
i
m
e
d
i
a
T
o
o
l
s
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
8
2
,
n
o
.
2
4
,
p
p
.
3
7
1
9
7
–
3
7
2
1
9
,
O
c
t
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
0
4
2
-
0
2
3
-
1
4
7
1
3
-
6.
[
1
4
]
M
.
I
r
sad
a
n
d
A
.
K
h
a
r
e
,
“
S
e
n
t
i
me
n
t
c
l
a
ssi
f
i
c
a
t
i
o
n
o
n
m
o
b
i
l
e
r
e
v
i
e
w
u
si
n
g
e
x
t
r
a
c
t
i
o
n
o
f
se
n
t
i
m
e
n
t
c
o
n
v
e
y
i
n
g
se
n
t
e
n
c
e
s
,
”
J
o
u
rn
a
l
o
f
I
n
f
o
rm
a
t
i
c
s E
l
e
c
t
ri
c
a
l
a
n
d
El
e
c
t
ro
n
i
c
s E
n
g
i
n
e
e
ri
n
g
(
J
I
E
EE)
,
v
o
l
.
5
,
n
o
.
2
,
p
p
.
1
–
8
,
2
0
2
4
,
d
o
i
:
1
0
.
5
4
0
6
0
/
a
2
z
j
o
u
r
n
a
l
s.
j
i
e
e
e
.
1
1
6
.
[
1
5
]
S
.
A
h
a
m
ma
d
,
S
.
A
.
S
i
n
t
h
i
a
,
M
.
A
h
m
e
d
,
M
.
H
o
ss
a
i
n
,
N
.
A
.
A
.
A
s
i
f
,
a
n
d
N
.
A
.
I
k
r
a
m,
“
D
e
e
p
l
e
a
r
n
i
n
g
-
b
a
s
e
d
se
n
t
i
me
n
t
a
n
a
l
y
si
s
o
f
u
ser
g
e
n
e
r
a
t
e
d
r
e
v
i
e
w
s
o
f
v
a
r
i
o
u
s
A
I
p
o
w
e
r
e
d
mo
b
i
l
e
a
p
p
l
i
c
a
t
i
o
n
s
,
”
i
n
2
0
2
4
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
I
n
v
e
n
t
i
v
e
C
o
m
p
u
t
a
t
i
o
n
T
e
c
h
n
o
l
o
g
i
e
s
(
I
C
I
C
T
)
,
I
EEE,
A
p
r
.
2
0
2
4
,
p
p
.
5
0
5
–
5
1
2
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
I
C
T6
0
1
5
5
.
2
0
2
4
.
1
0
5
4
4
6
9
9
.
[
1
6
]
P
.
R
.
H
e
n
a
o
,
J.
F
i
s
c
h
b
a
c
h
,
D
.
S
p
i
e
s,
J
.
F
r
a
t
t
i
n
i
,
a
n
d
A
.
V
o
g
e
l
sa
n
g
,
“
Tr
a
n
sf
e
r
l
e
a
r
n
i
n
g
f
o
r
mi
n
i
n
g
f
e
a
t
u
r
e
r
e
q
u
e
st
s
a
n
d
b
u
g
r
e
p
o
r
t
s
f
r
o
m
t
w
e
e
t
s
a
n
d
a
p
p
st
o
r
e
r
e
v
i
e
w
s
,
”
i
n
2
0
2
1
I
EEE
2
9
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
R
e
q
u
i
reme
n
t
s
E
n
g
i
n
e
e
r
i
n
g
C
o
n
f
e
r
e
n
c
e
Wo
rksh
o
p
s
(
RE
W)
,
I
EEE,
S
e
p
.
2
0
2
1
,
p
p
.
8
0
–
8
6
.
d
o
i
:
1
0
.
1
1
0
9
/
R
EW
5
3
9
5
5
.
2
0
2
1
.
0
0
0
1
9
.
[
1
7
]
K
.
S
.
N
u
g
r
o
h
o
,
A
.
Y
.
S
u
k
ma
d
e
w
a
,
H
.
W
u
sw
i
l
a
h
a
k
e
n
D
w
,
F
.
A
.
B
a
c
h
t
i
a
r
,
a
n
d
N
.
Y
u
d
i
s
t
i
r
a
,
“
B
E
R
T
f
i
n
e
-
t
u
n
i
n
g
f
o
r
s
e
n
t
i
m
e
n
t
a
n
a
l
y
si
s
o
n
I
n
d
o
n
e
s
i
a
n
m
o
b
i
l
e
a
p
p
s
r
e
v
i
e
w
s
,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
6
t
h
i
n
t
e
r
n
a
t
i
o
n
a
l
c
o
n
f
e
re
n
c
e
o
n
su
st
a
i
n
a
b
l
e
i
n
f
o
rm
a
t
i
o
n
e
n
g
i
n
e
e
ri
n
g
a
n
d
t
e
c
h
n
o
l
o
g
y
,
N
e
w
Y
o
r
k
,
N
Y
,
U
S
A
:
A
C
M
,
S
e
p
.
2
0
2
1
,
p
p
.
2
5
8
–
2
6
4
.
d
o
i
:
1
0
.
1
1
4
5
/
3
4
7
9
6
4
5
.
3
4
7
9
6
7
9
.
[
1
8
]
A
.
N
u
r
f
i
k
r
i
,
“
S
e
n
t
i
m
e
n
t
a
n
a
l
y
s
i
s
t
e
l
e
me
d
i
c
i
n
e
a
p
p
s
r
e
v
i
e
w
s
u
s
i
n
g
N
V
I
V
O
,
”
i
n
T
h
e
5
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
V
o
c
a
t
i
o
n
a
l
E
d
u
c
a
t
i
o
n
A
p
p
l
i
e
d
S
c
i
e
n
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
2
0
2
2
,
B
a
s
e
l
S
w
i
t
z
e
r
l
a
n
d
:
M
D
P
I
,
D
e
c
.
2
0
2
2
,
p
.
4
.
d
o
i
:
1
0
.
3
3
9
0
/
p
r
o
c
e
e
d
i
n
g
s
2
0
2
2
0
8
3
0
0
4
.
[
1
9
]
J.
H
o
u
a
n
d
A
.
Zh
u
,
“
I
d
e
n
t
i
f
i
c
a
t
i
o
n
o
f
u
s
e
f
u
l
n
e
ss
f
o
r
o
n
l
i
n
e
r
e
v
i
e
w
s
b
a
se
d
o
n
g
r
o
u
n
d
e
d
t
h
e
o
r
y
a
n
d
m
u
l
t
i
l
a
y
e
r
p
e
r
c
e
p
t
r
o
n
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
A
p
p
l
i
e
d
S
c
i
e
n
c
e
s
,
v
o
l
.
1
3
,
n
o
.
9
,
p
.
5
3
2
1
,
A
p
r
.
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
a
p
p
1
3
0
9
5
3
2
1
.
[
2
0
]
T.
S
h
a
r
ma
a
n
d
K
.
K
a
u
r
,
“
B
e
n
c
h
mar
k
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
m
e
t
h
o
d
s
f
o
r
a
s
p
e
c
t
l
e
v
e
l
se
n
t
i
me
n
t
c
l
a
ssi
f
i
c
a
t
i
o
n
,
”
A
p
p
l
i
e
d
S
c
i
e
n
c
e
s
,
v
o
l
.
1
1
,
n
o
.
2
2
,
p
.
1
0
5
4
2
,
N
o
v
.
2
0
2
1
,
d
o
i
:
1
0
.
3
3
9
0
/
a
p
p
1
1
2
2
1
0
5
4
2
.
[
2
1
]
M
.
Q
.
H
.
O
c
t
a
v
a
,
D
.
G
.
P
r
a
se
t
y
o
P
u
t
r
i
,
F
.
M
.
H
i
l
m
y
,
U
.
F
a
r
o
o
q
,
R
.
A
.
N
u
r
h
a
l
i
z
a
,
a
n
d
A
.
G
a
n
j
a
r
,
“
W
e
b
-
b
a
se
d
s
e
n
t
i
me
n
t
a
n
a
l
y
s
i
s
sy
st
e
m
U
si
n
g
S
V
M
a
n
d
TF
-
I
D
F
w
i
t
h
st
a
t
i
st
i
c
a
l
f
e
a
t
u
r
e
,
”
i
n
2
0
2
3
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
I
n
n
o
v
a
t
i
o
n
a
n
d
I
n
t
e
l
l
i
g
e
n
c
e
f
o
r
I
n
f
o
rm
a
t
i
c
s,
C
o
m
p
u
t
i
n
g
,
a
n
d
T
e
c
h
n
o
l
o
g
i
e
s
,
3
I
C
T
2
0
2
3
,
I
EEE,
N
o
v
.
2
0
2
3
,
p
p
.
9
–
1
4
.
d
o
i
:
1
0
.
1
1
0
9
/
3
I
C
T6
0
1
0
4
.
2
0
2
3
.
1
0
3
9
1
7
3
4
.
[
2
2
]
M
.
B
i
r
j
a
l
i
,
M
.
K
a
sr
i
,
a
n
d
A
.
B
e
n
i
-
H
s
san
e
,
“
A
c
o
m
p
r
e
h
e
n
si
v
e
s
u
r
v
e
y
o
n
s
e
n
t
i
m
e
n
t
a
n
a
l
y
si
s
:
a
p
p
r
o
a
c
h
e
s
,
c
h
a
l
l
e
n
g
e
s
a
n
d
t
r
e
n
d
s,
”
K
n
o
w
l
e
d
g
e
-
B
a
se
d
S
y
s
t
e
m
s
,
v
o
l
.
2
2
6
,
p
.
1
0
7
1
3
4
,
A
u
g
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
k
n
o
s
y
s.
2
0
2
1
.
1
0
7
1
3
4
.
[
2
3
]
S
.
S
i
n
g
h
,
K
.
K
u
mar
,
a
n
d
B
.
K
u
m
a
r
,
“
S
e
n
t
i
me
n
t
a
n
a
l
y
si
s
o
f
Tw
i
t
t
e
r
d
a
t
a
u
si
n
g
TF
-
I
D
F
a
n
d
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s,
”
i
n
2
0
2
2
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
M
a
c
h
i
n
e
L
e
a
rn
i
n
g
,
Bi
g
D
a
t
a
,
C
l
o
u
d
a
n
d
P
a
r
a
l
l
e
l
C
o
m
p
u
t
i
n
g
(
C
O
M
-
IT
-
C
O
N
)
,
I
EEE,
M
a
y
2
0
2
2
,
p
p
.
2
5
2
–
2
5
5
.
d
o
i
:
1
0
.
1
1
0
9
/
C
O
M
-
IT
-
C
O
N
5
4
6
0
1
.
2
0
2
2
.
9
8
5
0
4
7
7
.
[
2
4
]
A
.
A
l
R
y
a
n
,
H
.
D
.
A
r
p
i
t
a
,
A
.
T
a
b
a
s
s
u
m
,
M
.
S
.
A
h
a
m
m
a
d
,
a
n
d
M
.
A
.
A
k
r
a
m
,
“
A
c
o
m
p
a
r
a
t
i
v
e
a
n
a
l
y
s
i
s
b
e
t
w
e
e
n
d
e
e
p
l
e
a
r
n
i
n
g
a
n
d
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
m
s
b
a
s
e
d
o
n
u
s
e
r
r
e
v
i
e
w
s
e
n
t
i
m
e
n
t
a
n
a
l
y
s
i
s
f
r
o
m
v
a
r
i
o
u
s
O
T
T
ap
p
l
i
c
a
t
i
o
n
s
,
”
i
n
2
0
2
4
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
C
o
m
p
u
t
e
r
,
E
l
e
c
t
r
i
c
a
l
&
C
o
m
m
u
n
i
c
a
t
i
o
n
E
n
g
i
n
e
e
r
i
n
g
(
I
C
C
E
C
E
)
,
I
E
E
E
,
F
e
b
.
2
0
2
4
,
p
p
.
1
–
7
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
C
E
C
E
5
8
6
4
5
.
2
0
2
4
.
1
0
4
9
7
2
2
9
.
[
2
5
]
G
.
M
.
R
a
z
a
,
Z.
S
.
B
u
t
t
,
S
.
La
t
i
f
,
a
n
d
A
.
W
a
h
i
d
,
“
S
e
n
t
i
me
n
t
an
a
l
y
si
s
o
n
C
O
V
I
D
t
w
e
e
t
s:
a
n
e
x
p
e
r
i
m
e
n
t
a
l
a
n
a
l
y
s
i
s
o
n
t
h
e
i
m
p
a
c
t
o
f
c
o
u
n
t
v
e
c
t
o
r
i
z
e
r
a
n
d
TF
-
I
D
F
o
n
se
n
t
i
men
t
p
r
e
d
i
c
t
i
o
n
s
u
si
n
g
d
e
e
p
l
e
a
r
n
i
n
g
m
o
d
e
l
s,
”
i
n
2
0
2
1
I
n
t
e
rn
a
t
i
o
n
a
l
c
o
n
f
e
re
n
c
e
o
n
d
i
g
i
t
a
l
f
u
t
u
r
e
s
a
n
d
t
r
a
n
s
f
o
rm
a
t
i
v
e
t
e
c
h
n
o
l
o
g
i
e
s (I
C
o
D
T
2
)
,
I
EEE,
M
a
y
2
0
2
1
,
p
p
.
1
-
6
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
o
D
T
2
5
2
2
8
8
.
2
0
2
1
.
9
4
4
1
5
0
8
.
[
2
6
]
K
.
L.
Ta
n
,
C
.
P
.
L
e
e
,
a
n
d
K
.
M
.
Li
m
,
“
A
su
r
v
e
y
o
f
se
n
t
i
me
n
t
a
n
a
l
y
si
s
:
a
p
p
r
o
a
c
h
e
s,
d
a
t
a
set
s,
a
n
d
f
u
t
u
r
e
r
e
sea
r
c
h
,
”
Ap
p
l
i
e
d
S
c
i
e
n
c
e
s
,
v
o
l
.
1
3
,
n
o
.
7
,
p
.
4
5
5
0
,
A
p
r
.
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
a
p
p
1
3
0
7
4
5
5
0
.
[
2
7
]
R
.
A
.
M
a
n
g
n
g
a
l
l
e
,
M
.
D
.
P
u
r
b
o
l
a
k
so
n
o
,
a
n
d
W
.
A
st
u
t
i
,
“
S
e
n
t
i
m
e
n
t
a
n
a
l
y
s
i
s
o
f
L
a
z
a
d
a
a
p
p
r
e
v
i
e
w
u
si
n
g
W
o
r
d
2
V
e
c
a
n
d
s
u
p
p
o
r
t
v
e
c
t
o
r
ma
c
h
i
n
e
,
”
i
n
2
0
2
3
3
r
d
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
I
n
t
e
l
l
i
g
e
n
t
C
y
b
e
rn
e
t
i
c
s
T
e
c
h
n
o
l
o
g
y
&
A
p
p
l
i
c
a
t
i
o
n
s
(
I
C
I
C
y
T
A)
,
I
EEE
,
D
e
c
.
2
0
2
3
,
p
p
.
1
8
2
–
1
8
7
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
I
C
y
TA
6
0
1
7
3
.
2
0
2
3
.
1
0
4
2
8
7
7
1
.
[
2
8
]
H
.
J
i
a
n
g
,
C
.
H
u
,
a
n
d
F
.
J
i
a
n
g
,
“
T
e
x
t
s
e
n
t
i
m
e
n
t
a
n
a
l
y
s
i
s
o
f
m
o
v
i
e
r
e
v
i
e
w
s
b
a
s
e
d
o
n
W
o
r
d
2
V
e
c
-
L
S
T
M
,
”
i
n
2
0
2
2
1
4
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
c
o
n
f
e
r
e
n
c
e
o
n
a
d
v
a
n
c
e
d
c
o
m
p
u
t
a
t
i
o
n
a
l
i
n
t
e
l
l
i
g
e
n
c
e
(
I
C
A
C
I
)
,
I
E
E
E
,
J
u
l
.
2
0
2
2
,
p
p
.
1
2
9
–
1
3
4
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
A
C
I
5
5
5
2
9
.
2
0
2
2
.
9
8
3
7
5
0
5
.
[
2
9
]
P
.
F
.
M
u
h
a
mm
a
d
,
R
.
K
u
s
u
m
a
n
i
n
g
r
u
m
,
a
n
d
A
.
W
i
b
o
w
o
,
“
S
e
n
t
i
me
n
t
a
n
a
l
y
s
i
s
u
s
i
ng
W
o
r
d
2
v
e
c
a
n
d
l
o
n
g
s
h
o
r
t
-
t
e
r
m me
m
o
r
y
(
LS
T
M
)
f
o
r
I
n
d
o
n
e
s
i
a
n
H
o
t
e
l
r
e
v
i
e
w
s
,
”
Pr
o
c
e
d
i
a
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
1
7
9
,
p
p
.
7
2
8
–
7
3
5
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
r
o
c
s
.
2
0
2
1
.
0
1
.
0
6
1
.
[
3
0
]
J.
S
h
i
n
e
t
a
l
.
,
“
E
x
p
l
o
r
i
n
g
t
h
e
e
f
f
e
c
t
i
v
e
n
e
ss
o
f
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
a
n
d
d
e
e
p
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
ms
f
o
r
s
e
n
t
i
m
e
n
t
a
n
a
l
y
si
s
:
a
s
y
st
e
ma
t
i
c
l
i
t
e
r
a
t
u
r
e
r
e
v
i
e
w
,
”
C
o
m
p
u
t
e
rs,
M
a
t
e
r
i
a
l
s
a
n
d
C
o
n
t
i
n
u
a
,
v
o
l
.
8
4
,
n
o
.
3
,
p
p
.
4
1
0
5
–
4
1
5
3
,
2
0
2
5
,
d
o
i
:
1
0
.
3
2
6
0
4
/
c
mc
.
2
0
2
5
.
0
6
6
9
1
0
.
[
3
1
]
A
.
Li
g
t
h
a
r
t
,
C
.
C
a
t
a
l
,
a
n
d
B
.
Te
k
i
n
e
r
d
o
g
a
n
,
“
S
y
s
t
e
ma
t
i
c
r
e
v
i
e
w
s
i
n
s
e
n
t
i
me
n
t
a
n
a
l
y
si
s
:
a
t
e
r
t
i
a
r
y
st
u
d
y
,
”
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
Re
v
i
e
w
,
v
o
l
.
5
4
,
n
o
.
7
,
p
p
.
4
9
9
7
–
5
0
5
3
,
O
c
t
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
0
4
6
2
-
0
2
1
-
0
9
9
7
3
-
3.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
5
2
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
42
,
No
.
3
,
J
u
n
e
20
2
6
:
8
1
8
-
82
6
826
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
G
h
a
n
im
K
a
n
u
g
r
a
h
a
n
G
h
a
n
im
Ka
n
u
g
ra
h
a
n
re
c
e
iv
e
d
h
is
b
a
c
h
e
lo
r
’
s
d
e
g
re
e
in
c
o
m
p
u
ter
sc
ien
c
e
fr
o
m
Un
iv
e
rsitas
Dia
n
Nu
sw
a
n
t
o
ro
in
2
0
1
9
,
a
n
d
h
is
m
a
ste
r
’
s
d
e
g
re
e
in
i
n
fo
rm
a
ti
o
n
tec
h
n
o
l
o
g
y
fro
m
Un
iv
e
rsitas
In
d
o
n
e
sia
in
2
0
2
1
.
His
m
a
in
a
re
a
s
o
f
e
x
p
e
rti
se
i
n
c
lu
d
e
d
a
ta
sc
ien
c
e
,
s
e
n
ti
m
e
n
t
a
n
a
l
y
sis,
a
n
d
m
a
c
h
in
e
lea
rn
in
g
.
He
is
c
u
rre
n
tl
y
a
l
e
c
tu
re
r
in
th
e
in
f
o
rm
a
ti
c
s
p
ro
g
ra
m
a
t
S
a
tu
Un
iv
e
rsity
Ba
n
d
u
n
g
,
wh
e
re
h
e
is
a
c
ti
v
e
ly
in
v
o
lv
e
d
in
tea
c
h
in
g
,
re
se
a
rc
h
,
a
n
d
c
o
m
m
u
n
it
y
se
rv
ice
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
g
h
a
n
im.
k
a
n
u
g
ra
h
a
n
@
u
n
i
v
.
sa
tu
.
a
c
.
id
.
Win
Ce
is
a
n
a
c
a
d
e
m
ic
a
n
d
p
ro
fe
ss
io
n
a
l
t
h
a
t
h
a
v
e
b
e
e
n
a
c
ti
v
e
a
s
a
lec
tu
re
r
a
t
BINU
S
Un
iv
e
rsity
a
n
d
S
ATU
U
n
iv
e
rsit
y
,
sp
e
c
ifi
c
a
ll
y
with
in
t
h
e
In
fo
rm
a
ti
o
n
S
y
ste
m
s
stu
d
y
p
ro
g
ra
m
.
His
e
x
p
e
rien
c
e
is
n
o
t
l
imited
to
a
c
a
d
e
m
ia;
h
e
fre
q
u
e
n
t
l
y
se
rv
e
s
a
s
a
sp
e
a
k
e
r
a
n
d
e
x
p
e
rt
f
o
r
v
a
rio
u
s
c
o
m
p
a
n
i
e
s,
f
u
lfi
ll
i
n
g
a
n
e
x
p
e
rt
ro
le
in
t
h
e
i
n
d
u
str
y
.
Wi
n
Ce
,
S
.
Ko
m
.
,
M.M.
’
s res
e
a
rc
h
in
tere
sts sp
a
n
se
v
e
ra
l
k
e
y
a
re
a
s in
in
f
o
rm
a
ti
o
n
tec
h
n
o
lo
g
y
,
in
c
lu
d
i
n
g
so
f
twa
re
d
e
v
e
lo
p
m
e
n
t,
m
a
c
h
in
e
lea
rn
in
g
,
d
e
c
isio
n
su
p
p
o
rt
sy
ste
m
s
(DSS
),
a
n
d
in
fo
rm
a
ti
o
n
sy
ste
m
s
.
H
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
w
n
@b
in
u
s.e
d
u
a
n
d
win
.
c
e
@u
n
iv
.
sa
tu
.
a
c
.
id
.
Vito
H
a
fizh
C
a
h
a
y
a
Pu
tr
a
d
e
v
e
l
o
p
e
d
a
n
i
n
tere
st
in
I
n
fo
rm
a
ti
c
s
th
ro
u
g
h
th
e
Co
m
p
u
ter
a
n
d
Ro
b
o
ti
c
s
e
x
tr
a
c
u
rricu
lar
a
c
ti
v
it
y
i
n
2
0
1
1
.
He
th
e
n
p
u
rs
u
e
d
a
b
a
c
h
e
lo
r
’
s
d
e
g
re
e
in
in
f
o
rm
a
ti
c
s
e
n
g
i
n
e
e
rin
g
a
t
Un
i
v
e
rsitas
Wi
d
y
a
tam
a
a
n
d
late
r
c
o
m
p
lete
d
h
is
p
o
stg
ra
d
u
a
te
st
u
d
ies
a
t
Un
i
v
e
rsitas
La
n
g
lan
g
b
u
a
n
a
.
He
is
c
u
rre
n
tl
y
tea
c
h
i
n
g
a
t
Un
i
v
e
rsitas
S
a
tu
i
n
t
h
e
I
n
fo
rm
a
ti
c
s
p
r
o
g
ra
m
,
c
o
v
e
rin
g
s
u
b
jec
ts
s
u
c
h
a
s
o
b
jec
t
-
o
rien
te
d
p
r
o
g
ra
m
m
in
g
,
a
lg
o
rit
h
m
s,
Io
T,
a
n
d
we
b
p
ro
g
ra
m
m
in
g
.
His
re
se
a
rc
h
i
n
tere
sts
a
re
fo
c
u
se
d
o
n
a
rti
ficia
l
in
telli
g
e
n
c
e
(AI)
,
m
a
c
h
in
e
l
e
a
rn
in
g
,
n
a
t
u
ra
l
lan
g
u
a
g
e
p
ro
c
e
ss
in
g
,
a
n
d
th
e
I
o
T
.
H
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
v
i
to
.
p
u
tra@u
n
iv
.
sa
tu
.
a
c
.
i
d
.
Yud
i
Ra
m
d
h
a
n
i
is
a
lec
tu
re
r
a
n
d
re
se
a
rc
h
e
r
a
t
S
a
t
u
Un
iv
e
rsity
i
n
t
h
e
In
f
o
rm
a
ti
o
n
S
y
ste
m
s
S
tu
d
y
P
r
o
g
ra
m
.
He
h
o
ld
s
a
m
a
ste
r
’
s
d
e
g
re
e
in
c
o
m
p
u
ter
sc
ien
c
e
a
n
d
h
a
s
fo
c
u
se
d
h
is
re
se
a
rc
h
o
n
d
a
ta
m
in
i
n
g
,
d
a
ta
sc
ien
c
e
,
a
n
d
d
e
c
isio
n
su
p
p
o
rt
sy
st
e
m
s
.
Wi
th
se
v
e
ra
l
y
e
a
rs
o
f
e
x
p
e
rien
c
e
in
b
o
th
tea
c
h
i
n
g
a
n
d
re
se
a
rc
h
,
h
e
h
a
s
b
e
e
n
a
c
ti
v
e
ly
i
n
v
o
lv
e
d
in
d
e
v
e
l
o
p
i
n
g
d
a
ta
-
d
riv
e
n
a
c
a
d
e
m
ic
p
r
o
jec
ts
a
n
d
i
n
n
o
v
a
ti
v
e
c
u
rricu
l
u
m
d
e
si
g
n
.
He
h
a
s
p
u
b
li
sh
e
d
re
se
a
rc
h
p
a
p
e
rs
in
re
p
u
tab
le
j
o
u
r
n
a
ls
a
n
d
re
g
u
larl
y
c
o
n
tri
b
u
tes
to
a
c
a
d
e
m
ic
fo
ru
m
s
a
n
d
se
m
in
a
rs
.
H
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
y
u
d
i.
ra
m
d
h
a
n
i@
u
n
i
v
.
sa
tu
.
a
c
.
i
d
.
Febriy
a
n
t
i
Pa
n
ja
it
a
n
e
a
rn
e
d
h
e
r
d
o
c
t
o
ra
l
d
e
g
re
e
fro
m
th
e
F
a
c
u
lt
y
o
f
En
g
i
n
e
e
rin
g
,
Un
i
v
e
rsitas
S
riwij
a
y
a
,
with
a
c
o
n
c
e
n
trati
o
n
in
I
n
fo
r
m
a
ti
c
s
En
g
in
e
e
rin
g
,
i
n
2
0
2
3
.
He
r
m
a
in
a
re
a
s
o
f
e
x
p
e
rti
se
in
c
l
u
d
e
d
a
ta
sc
ien
c
e
,
m
a
c
h
in
e
lea
rn
i
n
g
,
a
n
d
d
e
e
p
lea
rn
in
g
,
with
a
re
se
a
rc
h
fo
c
u
s
o
n
d
e
v
e
l
o
p
i
n
g
AI
m
o
d
e
ls
f
o
r
c
o
m
p
le
x
d
a
ta
a
n
a
ly
s
is.
S
h
e
is
a
lec
tu
re
r
i
n
t
h
e
In
fo
rm
a
ti
c
s
En
g
in
e
e
rin
g
P
r
o
g
ra
m
F
a
c
u
lt
y
o
f
Cre
a
ti
v
e
Tec
h
n
o
lo
g
y
a
t
S
a
tu
Un
i
v
e
rsity
,
a
n
d
sh
e
is
a
c
ti
v
e
ly
in
v
o
lv
e
d
in
tea
c
h
in
g
,
re
se
a
r
c
h
,
a
n
d
c
o
m
m
u
n
it
y
se
rv
ic
e
.
Sh
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
p
e
b
rian
ti
.
p
a
n
jaita
n
@u
n
iv
.
s
a
tu
.
a
c
.
id
.
Evaluation Warning : The document was created with Spire.PDF for Python.