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.
39
,
No
.
3
,
Sep
tem
b
er
2
0
2
5
,
p
p
.
1
7
2
4
~
1
735
I
SS
N:
2
5
0
2
-
4
7
5
2
,
DOI
: 1
0
.
1
1
5
9
1
/ijeecs.v
39
.i
3
.
p
p
1
7
2
4
-
1
7
3
5
1724
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ee
cs.ia
esco
r
e.
co
m
M
a
chine learning
appro
a
ch f
o
r cos
t
estima
tion
in
so
f
tware
pro
ject pla
nning
Aj
a
y
J
a
is
wa
l
1
,
J
a
g
dis
h Ra
ikwa
l
2
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
a
n
d
E
n
g
i
n
e
e
r
i
n
g
,
P
r
e
st
i
g
e
I
n
st
i
t
u
t
e
of
E
n
g
i
n
e
e
r
i
n
g
,
M
a
n
a
g
e
me
n
t
a
n
d
S
c
i
e
n
c
e
,
I
n
d
o
r
e
,
I
n
d
i
a
2
D
e
p
a
r
t
me
n
t
o
f
I
n
f
o
r
mat
i
o
n
Te
c
h
n
o
l
o
g
y
,
I
n
st
i
t
u
t
e
o
f
E
n
g
i
n
e
e
r
i
n
g
a
n
d
T
e
c
h
n
o
l
o
g
y
,
D
e
v
i
A
h
i
l
y
a
V
i
s
h
w
a
v
i
d
y
a
l
a
y
a
,
I
n
d
o
r
e
,
I
n
d
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Dec
13
,
2
0
2
4
R
ev
is
ed
Mar
27
,
2
0
2
5
Acc
ep
ted
J
u
l
2
,
2
0
2
5
S
u
c
c
e
ss
fu
l
o
rg
a
n
izin
g
a
n
d
h
a
n
d
li
n
g
o
f
so
ftwa
re
p
r
o
jec
ts
d
e
p
e
n
d
s
e
x
ten
siv
e
l
y
o
n
a
c
c
u
ra
te co
st es
ti
m
a
ti
o
n
.
T
h
is
stu
d
y
e
x
p
l
o
re
s th
e
e
ffe
c
ti
v
e
n
e
ss
o
f
m
a
c
h
in
e
lea
rn
in
g
m
o
d
e
ls
in
e
stim
a
ti
n
g
s
o
ftwa
re
p
r
o
jec
t
c
o
sts
u
sin
g
d
a
t
a
se
ts
li
k
e
De
sh
a
rn
a
is,
M
a
x
we
ll
,
a
n
d
Kitc
h
e
n
h
a
m
,
a
im
in
g
t
o
p
re
v
e
n
t
p
ro
jec
t
d
e
lay
s
a
n
d
re
so
u
rc
e
m
isa
ll
o
c
a
ti
o
n
.
It
sh
o
ws
h
o
w
m
o
d
e
l
se
lec
ti
o
n
h
a
s
a
m
a
jo
r
imp
a
c
t
o
n
fo
re
c
a
st
a
c
c
u
ra
c
y
th
ro
u
g
h
a
t
h
o
r
o
u
g
h
a
ss
e
ss
m
e
n
t.
An
R
-
sq
u
a
re
d
v
a
lu
e
(R2
)
o
f
0
.
8
0
4
i
n
d
ica
tes
th
a
t
th
e
s
u
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
(
S
VM)
m
o
d
e
l
p
e
rfo
rm
s
e
x
c
e
p
ti
o
n
a
ll
y
we
ll
in
th
e
De
sh
a
rn
a
is
d
a
tas
e
t.
On
t
h
e
M
a
x
we
ll
d
a
ta
se
t,
li
n
e
a
r
re
g
re
ss
io
n
(LR)
sta
n
d
s
o
u
t
wi
th
a
m
in
imu
m
m
e
a
n
a
b
so
lu
te
e
rr
o
r
(M
AE)
o
f
0
.
4
8
3
a
n
d
t
h
e
g
re
a
tes
t
R2
v
a
lu
e
o
f
0
.
6
0
7
,
wh
i
le
S
VM
h
a
s
th
e
l
o
we
st
ro
o
t
m
e
a
n
sq
u
a
re
d
e
rro
r
(RM
S
E)
o
f
0
.
5
3
7
.
S
imi
larly
,
o
n
t
h
e
Kitch
e
n
h
a
m
d
a
tas
e
t,
LR
a
n
d
S
VM
a
re
th
e
to
p
p
e
rf
o
r
m
e
rs,
with
M
AE
o
f
0
.
2
0
1
a
n
d
RM
S
E
o
f
0
.
2
7
4
,
re
sp
e
c
ti
v
e
ly
,
a
n
d
R
2
v
a
l
u
e
s
o
f
a
ro
u
n
d
0
.
9
2
9
.
T
h
e
se
fin
d
i
n
g
s
h
ig
h
li
g
h
t
th
e
imp
o
rta
n
c
e
o
f
tail
o
re
d
m
o
d
e
l
se
lec
ti
o
n
fo
r
a
c
c
u
ra
te
c
o
st
p
re
d
ictio
n
,
a
s
LR
a
n
d
S
VM
c
o
n
ti
n
u
o
u
sly
d
e
m
o
n
stra
te
re
li
a
b
il
it
y
a
c
ro
ss
v
a
ried
d
a
tas
e
ts.
M
L
tec
h
n
i
q
u
e
s
l
ik
e
LR
a
n
d
S
VM
c
a
n
e
n
h
a
n
c
e
so
ftwa
re
p
r
o
jec
t
p
lan
n
i
n
g
a
n
d
m
a
n
a
g
e
m
e
n
t
b
y
p
ro
v
id
i
n
g
a
c
c
u
ra
te
c
o
st
e
stim
a
ti
o
n
,
wit
h
fu
t
u
r
e
re
se
a
rc
h
e
x
p
lo
r
in
g
e
n
se
m
b
le l
e
a
rn
i
n
g
a
n
d
d
e
e
p
lea
rn
in
g
m
e
th
o
d
o
lo
g
ies
.
K
ey
w
o
r
d
s
:
Acc
u
r
ac
y
Dee
p
lear
n
in
g
Ma
ch
in
e
lear
n
in
g
Pro
ject
p
lan
n
in
g
an
d
b
u
d
g
et
So
f
twar
e
co
s
t e
s
tim
atio
n
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Ajay
J
aiswal
Dep
ar
tm
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
an
d
E
n
g
in
ee
r
in
g
Pre
s
tig
e
I
n
s
titu
te
o
f
E
n
g
in
ee
r
in
g
,
Ma
n
a
g
em
en
t a
n
d
R
eseac
h
I
n
d
o
r
e
,
I
n
d
ia
E
m
ail: a
jay
.
jais
wal5
5
5
5
5
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
So
f
twar
e
m
ea
s
u
r
em
en
t
(
SM)
in
v
o
lv
es
m
ea
s
u
r
i
n
g
s
o
f
twar
e
ch
ar
ac
ter
is
tics
ty
p
ically
r
el
ated
to
th
e
p
r
o
d
u
ct,
m
eth
o
d
,
an
d
r
eso
u
r
c
es
u
tili
ze
d
in
s
o
f
twar
e
d
ev
elo
p
m
en
t.
T
h
ese
in
d
icato
r
s
can
be
u
tili
ze
d
in
p
r
o
ject
m
an
ag
em
en
t
s
y
s
tem
s
to
ass
is
t
s
o
f
twar
e
d
ev
elo
p
er
s
in
ef
f
e
ctiv
ely
m
a
n
ag
in
g
th
eir
p
r
o
jects,
h
en
ce
r
e
d
u
cin
g
is
s
u
es
s
u
ch
as
co
s
t
o
v
er
r
u
n
s
a
n
d
s
ch
ed
u
lin
g
d
ef
icien
cies
[
1
]
.
On
e
of
th
e
m
o
s
t
ch
allen
g
in
g
a
s
p
ec
ts
of
m
an
ag
in
g
a
p
r
o
ject
is
esti
m
atin
g
s
o
f
twar
e.
Acc
u
r
ately
esti
m
atin
g
th
e
tim
e,
m
o
n
ey
,
an
d
ef
f
o
r
t
n
ee
d
ed
to
co
m
p
lete
an
en
d
ea
v
o
r
h
as
b
ee
n
a
ch
allen
g
e
f
o
r
p
r
o
ject
m
an
ag
e
r
s
f
o
r
a
l
o
n
g
tim
e.
I
t
is
ch
allen
g
in
g
to
p
r
ed
ict
th
ese
f
ac
t
o
r
s
ea
r
ly
in
a
p
r
o
ject'
s
life
cy
cle
wh
en
th
er
e
is
a
lo
t
o
f
u
n
c
er
tain
ty
ab
o
u
t
th
e
p
r
o
d
u
ct'
s
f
ea
tu
r
es,
an
d
th
e
b
o
u
n
d
ar
ies
of
ea
c
h
in
itiativ
e
n
ee
d
to
b
e
d
e
f
in
ed
[
2
]
.
S
o
f
twar
e
p
r
o
ject
m
an
a
g
em
en
t
r
elies
h
ea
v
ily
o
n
esti
m
atio
n
,
p
ar
ticu
lar
ly
in
th
e
ea
r
ly
s
tag
es
of
s
o
f
twar
e
d
ev
e
lo
p
m
en
t.
T
h
e
f
ir
s
t
s
tep
is
to
e
s
tim
ate
how
m
u
c
h
tim
e,
m
o
n
ey
,
an
d
ef
f
o
r
t
ar
e
ex
p
ec
ted
to
b
e
r
eq
u
ir
ed
to
f
i
n
is
h
th
e
s
o
f
twar
e
p
r
o
ject
[
3
]
,
[
4
]
.
C
o
n
s
e
q
u
e
n
t
l
y
,
i
n
e
f
f
i
c
i
e
n
t
r
e
s
o
u
r
c
e
u
t
il
i
z
at
i
o
n
a
n
d
l
e
n
g
t
h
y
d
e
l
i
v
e
r
y
d
e
l
a
y
s
m
a
y
r
e
s
u
l
t
f
r
o
m
s
o
f
twar
e
c
o
s
t
o
v
er
esti
m
atio
n
.
Ho
wev
er
,
in
ad
eq
u
ate
wo
r
k
f
o
r
ce
n
u
m
b
er
s
,
g
o
in
g
o
v
er
b
u
d
g
et,
an
d
late
d
eliv
er
y
tim
e
m
ig
h
t
o
cc
u
r
f
r
o
m
u
n
d
er
esti
m
atin
g
s
o
f
twar
e
ex
p
en
s
es
[
5
]
.
Plan
n
in
g
is
th
e
m
o
s
t
cr
u
cial
p
h
ase
in
p
r
o
ject
m
an
ag
em
en
t
b
ec
a
u
s
e
it
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
Ma
ch
in
e
lea
r
n
in
g
a
p
p
r
o
a
ch
fo
r
co
s
t
esti
ma
tio
n
in
s
o
ftw
a
r
e
p
r
o
ject
p
la
n
n
in
g
(
A
ja
y
J
a
is
w
a
l
)
1725
esti
m
ates
th
e
tim
e
an
d
m
o
n
ey
n
ee
d
ed
to
f
in
is
h
a
p
r
o
ject
p
r
o
p
er
ly
[
6
]
.
T
h
e
m
o
s
t
well
k
n
o
wn
asp
ec
ts
o
f
s
o
f
twar
e
ev
alu
atio
n
ar
e
s
o
f
twar
e
co
s
t
an
d
ef
f
o
r
t
esti
m
atio
n
.
T
h
e
ter
m
"c
o
s
t
esti
m
atio
n
"
r
ef
er
s
to
th
e
p
r
o
ce
s
s
of
d
eter
m
in
in
g
an
ap
p
r
o
x
im
ate
t
o
tal
co
s
t
f
o
r
a
p
r
o
ject,
p
r
o
g
r
a
m
,
o
r
p
r
o
d
u
ct
u
s
in
g
cu
r
r
en
t
d
ata.
Acc
u
r
ate
co
s
t
esti
m
atio
n
is
cr
itical
f
o
r
ea
ch
p
r
o
ject
ty
p
e
to
av
o
i
d
u
n
ex
p
ec
ted
ly
ex
ce
s
s
iv
e
ex
p
en
d
itu
r
es.
Acc
u
r
ate
esti
m
ates
ar
e
ess
en
tial
f
o
r
d
ec
is
io
n
m
ak
er
s
to
m
an
ag
e
r
is
k
s
,
all
o
ca
te
r
eso
u
r
ce
s
,
an
d
g
en
er
ate
p
r
ec
is
e
p
r
o
ject
s
ch
ed
u
les
[
7
]
.
T
h
e
d
ev
elo
p
m
en
t
o
f
esti
m
atio
n
ap
p
r
o
ac
h
es,
wh
ich
h
a
v
e
g
o
n
e
f
r
o
m
b
asic
s
tatis
tical
m
o
d
els
t
o
co
m
p
lex
ML
alg
o
r
ith
m
s
,
r
ev
ea
ls
th
e
i
n
d
u
s
tr
y
'
s
d
ed
icatio
n
to
m
an
a
g
in
g
co
s
ts
an
d
p
r
o
ject
s
u
cc
ess
.
Plan
n
in
g
an
d
b
u
d
g
etin
g
f
o
r
a
p
r
o
ject
r
e
q
u
ir
es
an
ac
cu
r
ate
ass
es
s
m
en
t
o
f
s
o
f
twar
e
co
s
ts
.
E
f
f
icien
t,
h
ig
h
-
q
u
ality
,
a
n
d
p
r
ec
is
ely
es
tim
ated
d
ata
ar
e
ess
en
tial
f
o
r
s
u
cc
ess
f
u
l
r
eg
u
latio
n
a
n
d
o
v
e
r
s
ig
h
t.
I
m
p
r
o
v
e
th
e
p
r
ec
is
io
n
an
d
ef
f
icac
y
of
p
r
o
je
ct
p
lan
n
in
g
an
d
b
u
d
g
etin
g
with
cu
ttin
g
-
ed
g
e
co
s
t
esti
m
atio
n
s
o
f
twar
e
th
at
r
elies
o
n
ML
.
ML
alg
o
r
ith
m
s
g
en
er
ate
m
o
r
e
p
r
ec
is
e
an
d
ad
a
p
tab
le
f
o
r
ec
asts
b
y
a
n
aly
zin
g
p
ast
d
ata,
p
r
o
ject
attr
ib
u
tes,
an
d
o
th
er
v
a
r
iab
les.
ML
Mo
d
els,
lik
e
r
an
d
o
m
f
o
r
e
s
ts
,
d
ec
is
io
n
tr
ee
s
,
SVM,
an
d
l
o
g
is
tic
r
eg
r
ess
io
n
,
im
p
r
o
v
e
p
r
o
ject
p
lan
n
in
g
an
d
b
u
d
g
etin
g
[
8
]
-
[
1
0
]
.
Or
g
an
izati
o
n
s
can
o
p
tim
ize
r
eso
u
r
ce
allo
ca
tio
n
,
m
ak
e
b
etter
d
ec
is
io
n
s
,
an
d
c
o
n
f
i
d
en
tly
a
n
d
ef
f
icien
tly
n
a
v
ig
ate
s
o
f
t
war
e
d
ev
elo
p
m
en
t
c
o
m
p
lex
it
y
with
d
ata
-
d
r
iv
en
in
s
ig
h
ts
.
I
n
ad
d
itio
n
,
t
h
e
u
n
ce
r
tain
n
atu
r
e
o
f
s
o
f
twar
e
d
e
v
elo
p
m
en
t,
ch
ar
ac
ter
ized
b
y
ev
e
r
-
ch
an
g
in
g
r
eq
u
ir
em
e
n
ts
an
d
ev
o
lv
i
n
g
tec
h
n
o
lo
g
y
,
f
u
r
t
h
er
co
m
p
licates
t
h
e
esti
m
atio
n
p
r
o
ce
s
s
.
ML
p
r
o
v
id
es
a
p
r
o
m
is
in
g
s
o
lu
tio
n
as
it
en
ab
les
lea
r
n
in
g
f
r
o
m
p
ast
p
r
o
jects
an
d
th
e
id
en
tific
atio
n
of
p
atter
n
s
th
at
wo
u
ld
h
elp
p
r
ed
ict
f
u
tu
r
e
p
r
o
ject
co
s
ts
more
r
elia
b
ly
.
T
h
is
r
esear
ch
is
m
o
tiv
ate
d
b
y
th
e
o
b
jectiv
e
o
f
u
tili
zin
g
ML
to
en
h
an
ce
th
e
p
r
ec
is
io
n
o
f
s
o
f
twar
e
p
r
o
ject
co
s
t
esti
m
ates,
th
er
eb
y
en
ab
li
n
g
p
r
o
ject
m
an
a
g
er
s
to
m
a
k
e
in
f
o
r
m
e
d
d
ec
is
io
n
s
.
T
h
is
s
tu
d
y
p
r
o
p
o
s
es
a
n
ew
a
p
p
r
o
ac
h
f
o
r
esti
m
atin
g
s
o
f
tw
ar
e
co
s
ts
u
s
in
g
ML
tech
n
iq
u
es.
It
tar
g
ets
b
etter
p
r
ed
ictio
n
s
in
ter
m
s
of
ac
cu
r
ac
y
an
d
d
ep
en
d
ab
ilit
y
on
ex
p
en
d
itu
r
e
d
u
r
in
g
s
o
f
twar
e
d
e
v
elo
p
m
en
t
p
r
o
ce
d
u
r
es
so
as
to
en
ab
le
co
r
p
o
r
ate
ex
ec
u
tiv
es
to
m
a
k
e
u
p
co
r
r
ec
t
f
u
n
d
in
g
d
ec
is
io
n
s
an
d
m
an
a
g
e
o
th
er
ac
tiv
ities
r
elate
d
to
r
eso
u
r
cin
g
a
p
p
r
o
p
r
iately
.
T
h
e
g
o
al
of
th
is
r
esear
ch
is
to
p
r
o
v
id
e
a
n
ew
co
s
t
esti
m
atio
n
f
r
am
ewo
r
k
f
o
r
s
o
f
twar
e
p
r
o
ject
p
lan
n
in
g
u
s
in
g
m
ac
h
i
n
e
lear
n
in
g
(
ML
)
tech
n
i
q
u
es.
T
h
e
m
ain
co
n
tr
ib
u
tio
n
s
o
f
th
is
s
tu
d
y
a
r
e
:
−
Desig
n
in
g
an
in
n
o
v
ativ
e
ML
f
r
am
ewo
r
k
th
at
s
u
p
p
o
r
ts
th
e
ef
f
icien
t
an
d
ac
cu
r
ate
f
o
r
ec
asti
n
g
o
f
co
s
ts
in
s
o
f
twar
e
d
ev
elo
p
m
en
t.
−
T
h
e
u
s
e
of
d
atasets
s
u
ch
as
Desh
ar
n
ais,
Kitch
en
h
am
,
an
d
Ma
x
well
f
o
r
em
p
ir
ical
v
alid
atio
n
o
f
th
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
.
−
D
a
t
a
d
r
i
v
e
n
i
n
s
i
g
h
t
s
f
o
r
i
m
p
r
o
v
e
d
p
r
o
j
e
c
t
p
l
a
n
n
i
n
g
a
n
d
r
e
s
o
u
r
c
e
m
a
n
a
g
e
m
e
n
t
.
T
h
ese
co
n
tr
ib
u
tio
n
s
aim
to
cl
o
s
e
th
e
g
a
p
b
etwe
en
tr
ad
itio
n
al
esti
m
atio
n
ap
p
r
o
ac
h
es
an
d
m
o
d
er
n
d
y
n
am
ic
r
eq
u
ir
em
e
n
ts
f
o
r
s
o
f
twar
e
p
r
o
j
ec
ts
.
2.
RE
L
AT
E
D
W
O
RK
Acc
u
r
ate
s
o
f
twar
e
ef
f
o
r
t
esti
m
atio
n
is
cr
u
cial
f
o
r
s
o
f
twar
e
p
r
o
ject
m
an
ag
em
e
n
t
[
1
1
]
.
So
f
tw
ar
e
ef
f
o
r
t
esti
m
atio
n
r
ef
er
s
to
th
e
tech
n
iq
u
e
o
f
f
o
r
ec
asti
n
g
th
e
e
f
f
o
r
t
r
eq
u
ir
ed
t
o
b
u
ild
s
o
f
twar
e
p
r
o
d
u
cts
in
ter
m
s
of
ex
p
en
s
es
[
1
2
]
.
Pro
ject
p
lan
n
in
g
an
d
b
u
d
g
et
allo
ca
tio
n
ar
e
two
ar
ea
s
wh
er
e
p
r
io
r
r
esear
ch
in
SC
E
h
as
d
em
o
n
s
tr
ated
its
f
u
n
d
am
en
ta
l
im
p
o
r
tan
ce
.
E
f
f
e
c
t
i
v
e
m
o
n
i
t
o
r
i
n
g
a
n
d
r
e
g
u
l
a
t
i
o
n
of
s
o
f
t
w
a
r
e
d
e
v
e
l
o
p
m
e
n
t
p
r
o
j
e
c
t
s
r
e
q
u
i
r
e
s
p
r
e
ci
s
e
e
s
t
im
ates
of
co
s
t,
p
r
ec
is
io
n
,
a
n
d
q
u
ality
.
C
o
n
v
en
tio
n
al
m
o
d
els,
s
u
ch
as
t
h
e
co
n
s
tr
u
ctiv
e
co
s
t
m
o
d
el
II
(
C
OC
OM
O
)
[
1
3
]
,
[
1
4
]
,
d
ep
e
n
d
s
ig
n
if
ican
tly
on
r
eliab
le
an
d
ac
cu
r
ate
d
ata
f
r
o
m
th
e
p
ast.
Olu
-
Ajay
i
[
1
5
]
.
T
h
ese
f
i
n
d
in
g
s
a
r
e
u
n
iq
u
e
an
d
p
r
o
m
is
in
g
,
c
o
n
tr
ib
u
tin
g
to
ef
f
ec
tiv
e
b
u
s
in
ess
p
lan
n
in
g
a
n
d
r
is
k
r
ed
u
ctio
n
c
o
m
p
ar
ed
to
p
r
e
v
io
u
s
r
esear
ch
.
D
r
az
et
a
l.
[
1
6
]
em
p
h
asize
th
e
ess
en
tial
r
o
le
of
p
lan
n
in
g
an
d
b
u
d
g
etin
g
in
s
o
f
twar
e
p
r
o
jec
ts
.
A
h
y
b
r
id
ap
p
r
o
ac
h
was
p
r
o
p
o
s
ed
in
th
is
s
tu
d
y
by
in
te
g
r
atin
g
Gr
ay
W
o
lf
Op
tim
izatio
n
f
o
r
s
o
f
twar
e
ef
f
o
r
t
esti
m
atio
n
.
W
h
en
it
ca
m
e
to
SC
E
,
an
o
th
e
r
s
tu
d
y
[
1
7
]
u
s
ed
a
h
y
b
r
id
m
o
d
el
th
at
u
s
ed
th
e
tab
u
s
ea
r
ch
(
T
S)
m
eth
o
d
[
1
8
]
with
th
e
in
v
asiv
e
weed
o
p
tim
izatio
n
(
I
W
O)
alg
o
r
ith
m
[
1
9
]
.
T
h
e
T
S
alg
o
r
ith
m
wo
r
k
ed
b
etter
with
th
e
I
W
O
alg
o
r
ith
m
's
s
o
l
u
tio
n
s
[
2
0
]
.
Prio
r
to
th
at,
in
2
0
2
3
,
an
an
al
y
s
is
was
co
n
d
u
cte
d
to
co
m
p
ar
e
th
e
cu
r
r
en
t
tax
o
n
o
m
ies
an
d
m
eth
o
d
o
lo
g
ies
em
p
lo
y
e
d
in
th
e
esti
m
atio
n
of
s
o
f
twar
e
co
s
ts
u
s
in
g
n
e
u
r
al
n
etwo
r
k
s
[
2
1
]
.
A
r
e
v
iew
f
o
u
n
d
th
at
t
h
e
m
ea
n
m
ag
n
itu
d
e
o
f
r
elativ
e
er
r
o
r
(
MM
R
E
)
,
p
er
ce
n
tag
e
r
elativ
e
er
r
o
r
d
ev
i
atio
n
(
PR
E
D)
,
an
d
r
o
o
t
m
ea
n
s
q
u
ar
ed
er
r
o
r
(
R
MSE
)
ar
e
t
h
e
m
o
s
t
co
m
m
o
n
ly
u
tili
ze
d
m
etr
ics
f
o
r
ev
alu
atin
g
ML
-
SC
E
m
o
d
els
[
2
2
]
.
Alau
t
h
m
an
et
a
l.
[
2
3
]
d
is
cu
s
s
ed
s
o
f
twar
e
d
ev
elo
p
m
en
t
co
s
t
esti
m
atio
n
r
e
g
r
ess
io
n
m
o
d
el
s
elec
tio
n
.
I
t
em
p
h
asized
u
s
in
g
m
o
d
els
th
at
m
atch
t
h
e
s
o
f
twar
e
d
e
v
elo
p
m
e
n
t
m
eth
o
d
o
l
o
g
y
an
d
d
ataset
u
tili
ze
d
in
esti
m
atio
n
.
Go
v
in
d
a
et
a
l.
[
2
4
]
u
s
ed
ML
to
ca
lcu
late
s
o
f
twar
e
co
s
t
f
o
r
p
r
o
ject
m
an
a
g
er
s
u
s
in
g
s
tan
d
a
r
d
in
p
u
t.
A
k
h
b
ar
d
eh
et
a
l.
[
2
5
]
ex
am
in
ed
th
e
p
r
o
ce
s
s
es
f
o
r
co
m
p
u
tab
le
elem
en
ts
th
at
af
f
ec
t
s
o
f
twar
e
co
s
t
an
d
p
r
esen
ted
r
esear
ch
th
at
u
s
ed
ML
m
eth
o
d
o
l
o
g
ies
to
co
n
s
tr
u
ct
a
cr
ed
ib
le
esti
m
atio
n
m
eth
o
d
.
T
ab
le
1
d
escr
ib
es
th
e
esti
m
atin
g
ap
p
r
o
ac
h
an
d
th
e
co
n
tr
ib
u
tio
n
o
f
th
e
r
ec
o
g
n
is
ed
p
ap
er
s
.
W
e
h
av
e
tak
en
ac
cu
r
ac
y
v
alu
es
u
n
d
er
th
e
af
o
r
em
e
n
tio
n
ed
s
y
s
tem
f
r
o
m
a
v
a
r
iety
o
f
d
atasets
an
d
m
eth
o
d
o
lo
g
ies
in
o
r
d
er
t
o
in
v
esti
g
ate
ac
cu
r
ate
p
er
f
o
r
m
a
n
ce
an
aly
s
is
.
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.
39
,
No
.
3
,
Sep
tem
b
er
20
25
:
1
7
2
4
-
1
7
3
5
1726
T
ab
le
1
.
Pre
d
ictio
n
ac
cu
r
ac
y
o
f
p
r
im
ar
y
r
esear
ch
o
n
s
tan
d
alo
n
e
m
eth
o
d
s
S
t
u
d
y
A
u
t
h
o
r(
s)
D
a
t
a
s
e
t
E
s
t
i
m
a
t
i
o
n
t
e
c
h
n
i
q
u
e
/
c
o
n
t
r
i
b
u
t
i
o
n
M
M
R
E
P
R
ED
[26
]
M
al
h
o
t
ra
an
d
Jai
n
4
9
9
s
o
f
t
w
a
r
e
p
r
o
j
e
c
t
s
B
a
g
g
i
n
g
,
A
N
N
,
D
T,
S
V
M
,
a
n
d
l
i
n
e
a
r
r
e
g
r
e
ss
i
o
n
(
L
R
)
w
e
r
e
e
v
a
l
u
a
t
e
d
a
n
d
c
o
n
t
r
a
s
t
e
d
o
n
a
so
f
t
w
a
r
e
p
r
o
j
e
c
t
d
a
t
a
se
t
.
0
.
1
7
52
[27
]
S
h
a
r
ma
a
n
d
S
i
n
g
h
4
s
o
f
t
w
a
r
e
p
r
o
j
e
c
t
s
M
a
d
e
u
s
e
o
f
r
a
n
d
o
m f
o
r
e
s
t
s,
mu
l
t
i
l
a
y
e
r
p
e
r
c
e
p
t
r
o
n
s,
a
n
d
su
p
p
o
r
t
v
e
c
t
o
r
m
a
c
h
i
n
e
s
.
0
.
3
0
7
2
.
0
9
[
2
8
]
P
o
sp
i
e
s
z
n
y
et
a
l
.
1
1
v
a
r
i
a
b
l
e
s
o
f
t
w
a
r
e
p
r
o
j
e
c
t
s
Th
e
e
n
sem
b
l
e
o
f
s
u
p
p
o
r
t
v
e
c
t
o
r
m
a
c
h
i
n
e
s,
n
e
u
r
a
l
n
e
t
w
o
r
k
s,
a
n
d
g
e
n
e
r
a
l
l
i
n
e
a
r
m
o
d
e
l
s
w
a
s
a
v
e
r
a
g
e
d
.
0
.
1
3
7
6
.
9
1
[
2
9
]
P
an
d
ey
et
al
.
S
A
M
OA
P
r
o
v
i
d
e
d
a
u
sef
u
l
me
t
h
o
d
f
o
r
se
l
e
c
t
i
n
g
t
h
e
b
e
st
e
s
t
i
m
a
t
e
t
e
c
h
n
i
q
u
e
f
o
r
a
p
p
e
f
f
o
r
t
e
st
i
ma
t
e
s fr
o
m o
f
f
o
u
r
w
e
l
l
-
l
i
k
e
d
met
h
o
d
s
:
G
A
,
M
L
R
,
M
LP
-
N
N
,
a
n
d
n
a
v
e
.
0
.
9
94
[30
]
D
a
n
e
t
a
l
.
C
O
C
O
M
O
-
I,
N
A
S
A
P
art
i
cl
e
swarm
o
p
t
i
mi
sati
o
n
was
u
sed
t
o
i
mpro
v
e
a
C
OC
OMO
i
n
t
eg
rat
ed
(P
S
O)
art
i
fi
cial
n
eu
ral
n
et
work
(A
N
N)
mod
el
.
0
.
4
0
5
5
.
1
0
T
h
e
s
u
cc
ess
f
u
l
p
lan
n
in
g
an
d
ex
ec
u
tio
n
of
s
o
f
twar
e
d
e
v
elo
p
m
en
t
p
r
o
jects
is
d
ep
e
n
d
en
t
o
n
ac
c
u
r
ate
s
o
f
twar
e
p
r
o
ject
co
s
t
esti
m
atio
n
,
wh
ic
h
also
i
n
f
lu
en
ce
s
r
eso
u
r
ce
m
an
ag
em
e
n
t,
b
u
d
g
et
all
o
ca
tio
n
,
a
n
d
p
r
o
ject
s
ch
ed
u
les.
E
x
p
e
r
t
ju
d
g
m
en
t,
an
alo
g
o
u
s
esti
m
atio
n
,
an
d
p
ar
am
etr
ic
m
o
d
els
lik
e
C
O
C
OM
O
ar
e
ex
am
p
les
of
tr
ad
itio
n
al
co
s
t
esti
m
atio
n
te
ch
n
iq
u
es
th
at
f
r
eq
u
en
tly
s
tr
u
g
g
le
with
ad
a
p
tab
ilit
y
,
p
r
ec
is
io
n
,
an
d
ab
ilit
y
to
h
an
d
le
c
o
m
p
le
x
u
n
p
r
e
d
ictab
le
r
elatio
n
s
h
ip
s
t
h
at
ar
is
e
in
s
o
f
twar
e
d
ev
elo
p
m
en
t
p
r
o
ce
s
s
es.
No
v
el
ap
p
r
o
ac
h
es
h
av
e
b
ee
n
d
ev
el
o
p
ed
in
th
is
f
ield
o
f
s
tu
d
y
,
a
n
d
th
e
y
r
eq
u
ir
e
r
eg
u
lar
co
m
p
ar
ativ
e
ass
ess
m
en
ts
.
Acc
u
r
ate
s
o
f
twar
e
co
s
t
esti
m
at
io
n
is
cr
itical
to
th
e
s
u
cc
es
s
o
f
s
o
f
twar
e
p
r
o
jects
b
ec
au
s
e
it
g
iv
es
in
f
o
r
m
atio
n
ab
o
u
t
th
e
r
is
k
s
an
d
ch
allen
g
es
ass
o
ciat
ed
with
d
ev
elo
p
m
en
t.
C
o
m
p
ar
ativ
e
r
esu
lts
s
h
o
w
th
at
th
e
p
r
o
p
o
s
ed
m
o
d
el
o
u
tp
er
f
o
r
m
s
ex
is
tin
g
tech
n
i
q
u
es
ac
r
o
s
s
all
d
atase
ts
an
d
ev
alu
atio
n
cr
iter
ia.
T
h
e
f
i
n
d
in
g
s
w
er
e
q
u
ite
p
r
o
m
is
in
g
f
o
r
f
o
r
ec
asti
n
g
s
o
f
twar
e
co
s
t
p
r
ed
ictio
n
.
T
h
e
en
o
r
m
o
u
s
d
iv
er
s
ity
o
f
ML
ap
p
r
o
ac
h
es
h
as
led
to
co
m
p
ar
is
o
n
s
an
d
ev
en
tu
ally
,
th
e
in
teg
r
atio
n
o
f
v
ar
io
u
s
tech
n
iq
u
es.
Dete
r
m
in
in
g
th
e
m
o
s
t
ef
f
ec
tiv
e
esti
m
atin
g
m
eth
o
d
s
h
as
b
ec
o
m
e
ess
en
tial
f
o
r
im
p
r
o
v
in
g
th
e
p
r
o
ject
d
ev
elo
p
m
en
t
p
r
o
ce
s
s
due
to
th
eir
m
an
y
b
e
n
ef
its
.
W
h
en
wo
r
k
in
g
o
n
co
m
p
l
ex
p
r
o
jects
o
r
p
r
o
jects
with
ch
an
g
in
g
r
eq
u
ir
e
m
en
ts
,
ac
cu
r
ac
y
is
s
o
m
etim
es
an
is
s
u
e
with
ev
o
lv
i
n
g
ML
tech
n
iq
u
es.
Acc
u
r
ate
esti
m
atio
n
of
c
o
s
ts
is
ess
en
tial
to
ex
ec
u
tin
g
p
r
o
jects
o
n
tim
e
a
n
d
with
in
b
u
d
g
et,
an
d
n
u
m
er
o
u
s
co
m
p
a
n
ies
m
ak
e
s
ig
n
if
ican
t
in
v
estme
n
ts
in
th
is
a
r
ea
in
o
r
d
er
t
o
en
s
u
r
e
r
a
p
id
g
r
o
wth
an
d
s
atis
f
ied
c
u
s
t
o
m
e
r
s
.
A
p
a
r
t
f
r
o
m
t
h
i
s
f
a
ct,
t
h
e
s
e
c
h
a
l
l
e
n
g
e
s
a
r
e
m
a
d
e
e
v
e
n
m
o
r
e
c
h
a
l
l
e
n
g
i
n
g
by
t
h
e
d
y
n
am
ic
ch
an
g
es
th
at
can
o
cc
u
r
in
an
y
s
o
f
twar
e
p
r
o
j
ec
t,
s
u
ch
as
ev
o
lv
i
n
g
r
eq
u
ir
e
m
en
ts
,
im
p
r
o
v
ed
tech
n
o
lo
g
y
,
o
r
ev
en
s
h
if
ts
in
th
e
team
'
s
s
k
ill
s
.
T
h
e
g
o
al
o
f
th
is
r
esear
ch
is
to
d
ev
elo
p
a
ML
b
ased
ap
p
r
o
ac
h
f
o
r
ev
al
u
atin
g
p
lan
n
i
n
g
c
o
s
ts
f
o
r
s
o
f
twar
e
p
r
o
j
e
c
t
s
.
3.
P
RO
P
O
SE
D
M
E
T
H
O
D
T
h
e
p
r
o
p
o
s
ed
m
eth
o
d
f
o
r
esti
m
atin
g
s
o
f
twar
e
d
ev
elo
p
m
en
t
co
s
ts
in
v
o
lv
es
n
in
e
s
tep
s
f
r
am
ewo
r
k
.
Fig
u
r
e
1
d
e
p
icts
th
e
p
r
o
p
o
s
ed
m
eth
o
d
f
o
r
s
o
f
twar
e
d
ev
elo
p
m
en
t p
r
o
ject
c
o
s
t e
s
tim
atio
n
.
Data
co
llectio
n
an
d
p
r
e
-
p
r
o
ce
s
s
in
g
:
−
Gath
er
th
e
Desh
ar
n
ais,
Kitch
en
h
am
,
a
n
d
Ma
x
well
d
atasets
,
wh
ich
c
o
n
tain
h
is
to
r
ical
d
at
a
o
n
s
o
f
twar
e
p
r
o
jects,
in
clu
d
in
g
attr
ib
u
tes s
u
ch
as p
r
o
ject
s
ize,
ef
f
o
r
t,
a
n
d
d
u
r
atio
n
.
−
Pre
p
r
o
ce
s
s
th
e
d
atasets
b
y
h
an
d
lin
g
m
is
s
in
g
v
alu
es a
n
d
o
u
tli
er
s
an
d
s
tan
d
ar
d
izi
n
g
n
u
m
er
ic
al
f
ea
tu
r
es.
W
o
r
d
2
Vec
f
ea
tu
r
e
e
x
tr
ac
tio
n
:
−
C
o
n
v
er
t
tex
tu
al
d
ata
(
if
an
y
)
,
s
u
ch
as
p
r
o
ject
d
escr
ip
tio
n
s
,
an
d
r
eq
u
ir
em
e
n
ts
,
in
to
n
u
m
er
ical
v
ec
to
r
s
u
s
in
g
W
o
r
d
2
Vec
.
−
W
o
r
d
2
Vec
ca
n
ca
p
tu
r
e
th
e
s
em
an
tic
m
ea
n
in
g
o
f
wo
r
d
s
a
n
d
p
h
r
ases
with
in
th
e
tex
t,
p
r
o
v
id
in
g
d
e
n
s
e
v
ec
to
r
r
e
p
r
esen
tatio
n
s
f
o
r
ea
c
h
wo
r
d
.
Me
r
g
e
W
o
r
d
2
Vec
f
e
atu
r
es w
ith
n
u
m
e
r
ical
f
ea
tu
r
es
:
−
C
o
m
b
in
e
th
e
W
o
r
d
2
Vec
f
ea
t
u
r
es
with
th
e
ex
is
tin
g
n
u
m
e
r
ical
f
ea
tu
r
es
f
r
o
m
th
e
d
atasets
to
f
o
r
m
a
co
m
p
r
eh
e
n
s
iv
e
f
ea
tu
r
e
s
et.
Featu
r
e
s
elec
tio
n
u
s
in
g
r
ec
u
r
s
iv
e
f
ea
tu
r
e
elim
in
atio
n
(
R
FE)
:
−
I
m
p
lem
en
t RF
E
to
s
elec
t th
e
m
o
s
t
im
p
o
r
tan
t f
ea
t
u
r
es f
r
o
m
t
h
e
co
m
b
i
n
ed
f
ea
tu
r
e
s
et.
−
R
FE
r
ec
u
r
s
iv
ely
r
em
o
v
es
f
ea
tu
r
es,
f
itti
n
g
th
e
m
o
d
el
m
u
lti
p
le
tim
es
an
d
ass
es
s
in
g
f
ea
tu
r
e
im
p
o
r
tan
c
e
u
n
til th
e
o
p
tim
al
f
ea
tu
r
e
s
u
b
s
e
t is id
en
tifie
d
.
Data
s
p
litt
in
g
:
−
Sp
lit
th
e
d
atasets
in
to
tr
ain
in
g
an
d
test
i
n
g
s
ets,
en
s
u
r
in
g
t
h
at
ea
ch
d
ataset
is
d
iv
id
ed
a
p
p
r
o
p
r
iately
t
o
m
ain
tain
its
in
teg
r
ity
.
Mo
d
el
tr
ain
in
g
:
−
T
r
ain
d
if
f
e
r
en
t M
L
m
o
d
els,
in
clu
d
in
g
L
STM
,
L
R
,
SVM,
FNN,
R
NN,
an
d
DT
,
u
s
in
g
th
e
tr
ain
in
g
d
ataset.
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
Ma
ch
in
e
lea
r
n
in
g
a
p
p
r
o
a
ch
fo
r
co
s
t
esti
ma
tio
n
in
s
o
ftw
a
r
e
p
r
o
ject
p
la
n
n
in
g
(
A
ja
y
J
a
is
w
a
l
)
1727
F
i
g
u
r
e
1
.
P
r
o
p
o
s
e
d
m
e
t
h
o
d
Mo
d
el
e
v
alu
atio
n
:
−
E
v
alu
ate
th
e
p
er
f
o
r
m
an
ce
o
f
ea
ch
m
o
d
el
u
s
in
g
ap
p
r
o
p
r
iate
ev
alu
atio
n
m
etr
ics
s
u
ch
as
M
AE
,
MSE
an
d
R
2
er
r
o
r
.
−
C
o
m
p
ar
e
th
e
p
e
r
f
o
r
m
an
ce
o
f
e
ac
h
ML
m
o
d
el
to
id
en
tif
y
th
e
b
est p
er
f
o
r
m
in
g
a
p
p
r
o
ac
h
f
o
r
ea
ch
d
ataset.
Mo
d
el
o
p
tim
izatio
n
:
−
R
ef
in
e
th
e
h
y
p
er
p
a
r
am
eter
s
o
f
ch
o
s
en
m
o
d
els
u
s
in
g
m
eth
o
d
s
s
u
ch
as
g
r
i
d
s
ea
r
ch
o
r
r
an
d
o
m
s
ea
r
ch
to
en
h
an
ce
p
er
f
o
r
m
an
ce
.
−
E
n
s
u
r
e
th
at
th
e
m
o
d
els ar
e
o
p
t
im
ized
to
g
en
e
r
alize
well
o
n
u
n
s
ee
n
d
ata.
Fin
al
m
o
d
el
s
elec
tio
n
:
−
Select
th
e
b
est p
er
f
o
r
m
i
n
g
m
o
d
el
f
o
r
ea
c
h
d
ataset
b
as
ed
o
n
e
v
alu
atio
n
m
etr
ics an
d
o
p
tim
iz
atio
n
r
esu
lts
.
4.
M
E
T
H
O
D
T
h
e
two
p
r
im
a
r
y
c
o
m
p
o
n
en
t
s
th
at
m
ak
e
u
p
th
is
s
ec
tio
n
ar
e
s
o
f
twar
e
co
s
tin
g
tech
n
iq
u
es
an
d
a
p
r
o
p
o
s
ed
m
eth
o
d
th
at
i
n
tr
o
d
u
ce
s
a
n
o
v
el
an
d
in
v
en
tiv
e
f
r
am
ewo
r
k
with
th
e
in
te
n
tio
n
o
f
e
n
h
an
ci
n
g
th
e
s
o
f
twar
e
co
s
t e
s
tim
atin
g
p
r
o
ce
s
s
.
a)
T
e
c
h
n
i
q
u
e
s
u
s
e
d
:
T
h
e
s
u
b
s
eq
u
en
t
s
ec
tio
n
e
x
p
lain
s
th
e
tech
n
i
q
u
es
em
p
l
o
y
ed
in
th
e
p
r
o
p
o
s
ed
m
eth
o
d
f
o
r
co
s
t e
s
tim
atio
n
.
b)
W
o
r
d
2
V
e
c
:
W
o
r
d
2
v
e
c
is
s
i
m
p
l
e
r
a
n
d
f
a
s
t
e
r
to
l
e
a
r
n
t
h
a
n
o
t
h
e
r
m
e
t
h
o
d
s
.
S
e
n
te
n
c
e
wo
r
d
s
em
an
tics
can
be
d
etec
ted
v
ia
W
o
r
d
2
v
ec
.
T
h
e
te
x
t
p
r
o
ce
s
s
in
g
in
W
o
r
d
2
v
ec
is
done
by
a
two
-
lay
e
r
n
eu
r
al
n
et
wo
r
k
.
T
h
e
two
p
r
im
ar
y
tech
n
iq
u
es
f
o
r
lear
n
i
n
g
W
o
r
d
2
v
e
c
a
r
e
C
B
O
W
a
n
d
S
k
i
p
G
r
a
m
.
W
h
i
l
e
t
h
e
C
B
O
W
m
o
d
e
l
u
s
e
s
s
u
r
r
o
u
n
d
i
n
g
c
o
n
t
e
x
ts
to
f
o
r
e
c
as
t
t
h
e
w
o
r
d
,
t
h
e
S
k
i
p
G
r
a
m
u
s
e
s
t
h
o
s
e
s
a
m
e
c
o
n
te
x
t
s
t
o
f
o
r
ec
a
s
t
th
e
wo
r
d
its
elf
[
3
1
]
.
It
i
m
p
r
o
v
es
f
ea
tu
r
e
s
elec
tio
n
,
m
o
d
el
tr
ai
n
in
g
,
an
d
ev
alu
atio
n
b
y
b
r
id
g
in
g
u
n
s
tr
u
ctu
r
ed
te
x
t
an
d
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.
39
,
No
.
3
,
Sep
tem
b
er
20
25
:
1
7
2
4
-
1
7
3
5
1728
s
tr
u
ctu
r
ed
n
u
m
er
ical
f
ea
tu
r
es.
T
h
e
q
u
ality
o
f
W
o
r
d
2
Vec
c
h
ar
ac
ter
is
tics
ca
n
co
n
s
id
er
ab
l
y
af
f
ec
t
m
o
d
el
p
er
f
o
r
m
an
ce
in
f
o
r
ec
asti
n
g
s
o
f
twar
e
p
r
o
ject
r
esu
lts
.
c)
R
e
c
u
r
s
i
v
e
f
e
a
t
u
r
e
el
i
m
i
n
at
i
o
n
(
R
F
E
)
:
I
r
r
el
e
v
a
n
t
f
e
at
u
r
e
s
a
r
e
c
o
m
m
o
n
i
n
l
a
r
g
e
d
a
t
as
e
ts
[
3
2
]
.
T
h
e
i
n
e
f
f
i
c
i
e
n
c
y
o
f
t
h
e
c
l
as
s
i
f
i
c
at
io
n
a
l
g
o
r
i
t
h
m
is
i
m
p
a
ct
e
d
b
y
r
e
c
u
r
r
i
n
g
f
e
a
t
u
r
e
s
.
D
e
t
e
r
m
i
n
e
w
h
i
c
h
v
a
r
i
a
b
l
es
a
r
e
c
r
u
c
i
a
l
f
o
r
m
a
k
i
n
g
a
c
c
u
r
a
t
e
f
o
r
e
c
a
s
ts
u
s
i
n
g
R
F
E
.
A
m
e
t
h
o
d
t
h
a
t
i
te
r
a
t
i
v
el
y
s
ea
r
c
h
e
s
f
o
r
a
t
a
r
g
e
t
n
u
m
b
e
r
o
f
a
t
t
r
i
b
u
t
es
i
s
"
r
e
c
u
r
s
i
v
e
"
.
Fi
g
u
r
e
2
i
s
a
v
is
u
a
l
r
e
p
r
e
s
e
n
t
a
ti
o
n
o
f
t
h
e
R
F
E
W
o
r
k
f
l
o
w
.
N
e
x
t
,
t
h
e
m
o
d
e
l
i
s
r
e
t
r
a
i
n
e
d
w
i
t
h
t
h
e
u
p
d
a
t
e
d
f
e
a
t
u
r
e
s
e
t
t
o
i
m
p
r
o
v
e
c
l
a
s
s
i
f
i
c
a
t
i
o
n
a
c
c
u
r
a
c
y
a
n
d
r
e
m
o
v
e
l
e
s
s
i
m
p
o
r
t
a
n
t
f
e
a
t
u
r
e
s
.
T
h
e
l
o
o
p
c
o
n
t
i
n
u
e
s
a
s
l
o
n
g
a
s
t
h
e
r
e
a
r
e
a
d
d
i
t
i
o
n
a
l
f
e
a
t
u
r
e
s
t
o
b
e
i
n
c
l
u
d
e
d
.
R
F
E
i
s
a
n
e
s
s
e
n
t
i
a
l
p
a
r
t
o
f
t
h
e
s
u
g
g
e
s
t
e
d
m
e
t
h
o
d
f
o
r
s
o
f
t
w
a
r
e
d
e
v
e
l
o
p
m
e
n
t
c
o
s
t
e
s
t
i
m
a
t
i
o
n
s
i
n
c
e
i
t
i
m
p
r
o
v
e
s
t
h
e
e
f
f
i
c
i
e
n
c
y
a
n
d
a
c
c
u
r
a
c
y
o
f
M
L
m
o
d
e
l
s
'
p
r
e
d
i
c
t
i
o
n
s
.
U
s
i
n
g
R
F
E
,
t
h
e
m
o
s
t
r
e
l
e
v
a
n
t
f
e
a
t
u
r
e
s
a
r
e
s
e
l
e
c
t
e
d
f
r
o
m
t
h
e
s
e
t
o
f
f
e
a
t
u
r
e
s
t
h
a
t
i
n
c
l
u
d
e
s
b
o
t
h
t
h
e
o
r
i
g
i
n
a
l
n
u
m
e
r
i
c
a
l
f
e
a
t
u
r
e
s
a
n
d
t
h
e
t
e
x
t
u
a
l
c
h
a
r
a
c
t
e
r
i
s
t
i
c
s
g
e
n
e
r
a
t
e
d
b
y
W
o
r
d
2
V
e
c
.
Fig
u
r
e
2.
R
F
E
w
o
r
k
f
l
o
w
s
c
h
em
a
t
i
c
[
3
2
]
d)
L
in
ea
r
r
eg
r
ess
io
n
m
o
d
el
:
On
e
way
to
f
in
d
th
e
r
elatio
n
s
h
ip
s
b
etwe
en
th
e
two
s
e
ts
o
f
v
a
r
i
a
b
l
e
s
i
s
t
o
u
s
e
a
LR
m
o
d
e
l
[
3
3
]
.
S
o
f
t
w
a
r
e
c
o
s
t
e
s
t
i
m
at
i
o
n
in
v
o
lv
es
cr
ea
tin
g
p
r
ed
ictio
n
s
a
b
o
u
t
th
e
d
e
p
en
d
en
t
v
ar
ia
b
le,
wh
ich
is
s
o
f
twar
e
co
s
t,
u
s
in
g
m
ea
s
u
r
em
en
ts
f
o
r
th
e
in
d
ep
en
d
en
t
v
ar
iab
les,
wh
ich
ar
e
p
r
o
d
u
cts,
p
r
o
jects,
an
d
p
r
o
ce
d
u
r
es.
T
h
e
LR
m
o
d
e
ls
ar
e
u
tili
ze
d
in
th
e
s
tu
d
y
to
co
n
d
u
ct
an
an
aly
s
is
of
th
e
am
o
u
n
t
of
m
o
n
ey
r
eq
u
ir
ed
to
d
ev
elo
p
s
o
f
twar
e.
A
s
in
g
le
in
d
ep
e
n
d
en
t
v
ar
ia
b
le,
a
s
in
g
le
LR
,
is
u
s
ed
to
p
r
ed
ict
th
e
co
s
t
d
e
p
e
n
d
e
n
t
v
a
r
i
a
b
l
e
.
A
s
t
r
ai
g
h
t
l
i
n
e
is
c
o
m
p
u
t
e
d
to
l
es
s
e
n
t
h
e
d
i
s
c
r
e
p
a
n
c
y
b
e
tw
e
e
n
th
e
ac
tu
a
l
m
u
ltip
le
LR
,
wh
ich
em
p
lo
y
s
a
lar
g
e
n
u
m
b
er
o
f
in
d
ep
e
n
d
en
t
v
ar
iab
les,
an
d
p
r
o
v
id
es
a
f
o
r
m
u
la
f
o
r
c
alcu
latin
g
th
e
esti
m
ated
co
s
t
u
s
in
g
a
lin
ea
r
c
o
m
b
in
atio
n
o
f
th
e
m
etr
ics [
3
4
]
,
[
3
5
]
.
=
+
1
1
+
2
2
+
⋯
+
ℰ
(
1
)
T
h
e
v
alu
es
o
f
x
1
th
r
o
u
g
h
x
n
ar
e
th
e
i
n
d
ep
e
n
d
en
t
v
a
r
iab
le
s
,
wh
er
ea
s
y
i
is
th
e
d
ep
e
n
d
e
n
t
v
ar
iab
le,
wh
er
e
i=1
…m
.
T
h
e
v
ar
iab
les
b1
th
r
o
u
g
h
b
m
s
tan
d
f
o
r
th
e
r
eg
r
ess
io
n
co
ef
f
icien
ts
,
an
d
th
e
letter
a
r
ep
r
esen
ts
th
e
in
ter
ce
p
t
(
eq
.
1
)
.
L
in
ea
r
m
o
d
el
s
ar
e
an
e
x
ce
llen
t
s
tar
tin
g
p
o
in
t
b
u
t
ca
n
n
o
t
g
u
ar
a
n
tee
f
law
less
d
ata
lin
ea
r
ity
.
Valid
atin
g
h
is
to
r
ical
e
f
f
o
r
ts
h
elp
s
ass
ess
th
eir
ac
cu
r
ac
y
.
Ac
tio
n
ab
le
a
n
d
ex
p
licab
le
c
o
s
t
e
s
tim
atio
n
f
o
r
m
u
las
ca
n
b
e
c
o
n
s
tr
u
cted
f
r
o
m
cu
r
r
e
n
t m
ea
s
u
r
es u
s
in
g
LR
.
e)
R
ec
u
r
r
en
t
n
e
u
r
al
n
etwo
r
k
s
(
R
NN)
:
R
NNs
h
av
e
s
h
o
wn
p
r
o
m
is
in
g
p
o
ten
tial
f
o
r
s
o
f
twar
e
co
s
t
esti
m
atio
n
task
s
.
R
N
Ns
o
p
er
ate
b
y
p
r
o
ce
s
s
in
g
s
eq
u
en
tial
d
ata,
wh
er
e
th
e
o
u
tp
u
t
at
ea
ch
tim
e
s
tep
is
in
f
lu
en
ce
d
b
y
th
e
cu
r
r
en
t
in
p
u
t
an
d
th
e
p
r
e
v
io
u
s
h
id
d
en
s
tate.
R
NNs
ar
e
p
ar
ticu
lar
ly
wellsu
ited
f
o
r
h
a
n
d
lin
g
v
ar
i
o
u
s
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
Ma
ch
in
e
lea
r
n
in
g
a
p
p
r
o
a
ch
fo
r
co
s
t
esti
ma
tio
n
in
s
o
ftw
a
r
e
p
r
o
ject
p
la
n
n
in
g
(
A
ja
y
J
a
is
w
a
l
)
1729
f
ac
to
r
s
th
at
in
f
lu
en
ce
s
o
f
twar
e
co
s
ts
,
s
u
ch
as
p
r
o
ject
s
iz
e,
co
m
p
lex
ity
,
an
d
team
ex
p
er
ien
ce
.
T
h
is
r
ec
u
r
r
en
t
n
atu
r
e
allo
ws
R
NN
s
to
m
ain
tain
an
in
ter
n
al
m
em
o
r
y
,
m
ak
i
n
g
th
em
well
-
s
u
ited
f
o
r
p
r
o
b
lem
s
in
v
o
lv
in
g
s
eq
u
en
tial
in
f
o
r
m
ati
o
n
,
s
u
ch
as so
f
twar
e
p
r
o
ject
d
ata
[
3
6
]
.
f)
L
o
n
g
s
h
o
r
t
-
ter
m
m
em
o
r
y
(
L
STM
)
:
T
h
e
L
STM
n
etwo
r
k
is
a
ty
p
e
o
f
R
NN
th
at
was
cr
ea
ted
f
o
r
co
n
s
ec
u
tiv
e
d
ata
p
r
o
ce
s
s
in
g
in
d
ee
p
lear
n
i
n
g
.
T
h
e
p
r
o
b
lem
o
f
ex
p
lo
d
in
g
a
n
d
v
a
n
is
h
in
g
g
r
ad
ien
ts
was
s
u
cc
ess
f
u
lly
ad
d
r
ess
ed
b
y
a
n
L
STM
,
wh
ich
was
d
esig
n
ed
to
d
e
p
ict
s
eq
u
en
ce
s
wi
th
lo
n
g
-
r
an
g
e
d
ep
e
n
d
e
n
cies
[
3
7
]
ac
cu
r
ately
.
E
ac
h
g
ate
is
tau
g
h
t
to
e
n
d
th
e
in
p
u
t
v
alu
e
b
y
th
e
s
y
s
tem
,
wh
ich
d
o
es
t
h
i
s
by
c
o
n
t
i
n
u
o
u
s
l
y
p
r
o
v
i
d
i
n
g
e
r
r
o
r
s
i
g
n
a
l
s
to
t
h
e
m
[
3
8
]
.
g)
Dec
is
io
n
tr
ee
(
DT
)
:
DT
ar
e
h
ier
ar
ch
ical
tr
ee
-
s
tr
u
ct
u
r
ed
m
o
d
els
u
s
ed
f
o
r
co
s
t
an
d
e
f
f
o
r
t
esti
m
atio
n
in
s
o
f
twar
e
[
3
9
]
.
T
h
ey
r
ec
u
r
s
iv
el
y
s
p
lit
th
e
d
ata
b
ased
o
n
attr
ib
u
te
test
s
r
ep
r
esen
ted
b
y
in
ter
n
al
n
o
d
es,
with
b
r
an
ch
es
d
ep
ictin
g
test
o
u
tco
m
es
an
d
leaf
n
o
d
es
co
n
tain
in
g
th
e
p
r
ed
icted
esti
m
ates.
T
h
is
s
tr
u
ctu
r
e
allo
ws d
ec
is
io
n
tr
ee
s
to
m
o
d
el
th
e
im
p
ac
t o
f
f
ac
to
r
s
o
n
p
r
o
je
ct
co
s
t a
n
d
ef
f
o
r
t [
4
0
]
.
h)
Su
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM)
:
T
h
e
SVM
m
o
d
el
is
a
v
er
s
atile
to
o
l
in
s
o
f
twar
e
co
s
t
esti
m
atio
n
,
a
d
ep
t
a
t
h
an
d
lin
g
b
o
th
class
if
icatio
n
an
d
r
eg
r
ess
io
n
task
s
.
W
h
eth
er
f
ac
in
g
lin
ea
r
o
r
n
o
n
lin
ea
r
p
r
o
b
lem
s
,
SVM
ef
f
ec
tiv
ely
p
ar
titi
o
n
s
d
ata
b
y
c
o
n
s
tr
u
ctin
g
a
h
y
p
e
r
p
lan
e
th
at
s
ep
ar
ates c
lass
es [
4
1
]
.
i)
Feed
-
f
o
r
war
d
n
e
u
r
al
n
etwo
r
k
(
FF
NN)
:
Fo
r
s
o
f
twar
e
co
s
t
est
im
atio
n
is
t
h
e
FF
NN.
As
its
n
am
e
im
p
lies
,
d
ata
f
lo
ws
in
o
n
ly
o
n
e
d
ir
ec
ti
o
n
,
f
r
o
m
in
p
u
t
to
o
u
tp
u
t.
T
h
e
m
o
s
t
b
asic
ty
p
e
of
ar
tific
ial
n
eu
r
al
n
etwo
r
k
,
k
n
o
wn
as
FF
NN,
can
o
n
ly
go
f
o
r
war
d
an
d
ca
n
n
o
t
r
ev
er
s
e
its
d
ir
ec
tio
n
o
f
o
p
er
atio
n
.
FF
NN
h
as
3
lay
er
s
(
in
p
u
t,
h
id
d
en
,
a
n
d
o
u
t
p
u
t
l
a
y
e
r
)
.
F
F
N
N
a
ll
o
w
s
t
h
e
n
e
t
w
o
r
k
to
d
i
s
c
o
v
e
r
i
n
t
r
i
c
a
t
e
p
at
t
e
r
n
s
a
n
d
r
e
l
a
tio
n
s
h
ip
s
with
in
th
e
d
ata
b
y
in
co
r
p
o
r
ati
n
g
m
u
ltip
le
h
id
d
en
lay
er
s
.
j)
Featu
r
e
s
elec
tio
n
u
s
in
g
r
ec
u
r
s
iv
e
f
ea
tu
r
e
elim
in
atio
n
:
I
m
p
lem
en
t
R
FE
to
s
elec
t
th
e
m
o
s
t
im
p
o
r
tan
t
f
ea
tu
r
es
f
r
o
m
th
e
co
m
b
in
ed
f
ea
tu
r
e
s
et.
Mo
r
e
o
v
er
,
FF
NN
m
ay
f
ea
tu
r
e
d
ir
ec
t
(
lin
ea
r
)
co
n
n
ec
tio
n
s
b
etwe
en
th
e
i
n
p
u
t
an
d
o
u
t
p
u
t
lay
er
s
,
f
ac
ilit
atin
g
th
e
m
ap
p
i
n
g
of
in
p
u
t
v
ar
iab
les
to
o
u
tp
u
t
p
r
ed
ictio
n
s
with
o
u
t d
ir
ec
t c
o
n
n
ec
tio
n
s
b
et
wee
n
in
d
iv
id
u
al
in
p
u
t a
n
d
o
u
t
p
u
t u
n
its
[
4
2
]
.
5.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
R
esu
lts
:
T
h
e
o
u
tco
m
es
of
th
e
s
y
s
tem
atic
r
esear
ch
co
n
d
u
cted
f
o
r
s
o
f
twar
e
p
r
o
ject
co
s
t
esti
m
atio
n
ar
e
p
r
esen
ted
in
th
is
s
ec
tio
n
.
I
t
lo
o
k
s
at
p
er
tin
en
t
f
ea
tu
r
es
th
at
c
o
m
e
f
r
o
m
s
elec
tio
n
an
d
ex
tr
a
ctio
n
as
well
as
th
e
o
u
tco
m
es o
f
ML
m
o
d
el
ev
alu
a
tio
n
f
o
r
esti
m
atio
n
.
a)
E
x
p
e
r
i
m
e
n
t
a
l
s
e
t
u
p
:
T
h
e
ex
p
er
i
m
en
t
was
co
n
d
u
cted
on
a
lap
t
o
p
co
m
p
u
ter
with
an
I
n
tel
C
o
r
e
i7
p
r
o
ce
s
s
o
r
,
6
4
GB
o
f
R
AM
.
T
h
e
ex
p
er
im
en
t
was
ca
r
r
ied
o
u
t
u
s
in
g
a
v
ar
iety
o
f
to
o
ls
.
Go
o
g
le
Dr
iv
e
was
u
tili
ze
d
to
u
p
lo
ad
d
ata
s
ets
f
o
r
t
h
e
e
x
p
er
i
m
en
t,
wh
ich
wer
e
t
h
en
u
p
lo
ad
ed
to
G
o
o
g
le
C
o
lab
.
T
h
is
s
tu
d
y
u
s
es
Py
th
o
n
to
p
r
esen
t,
ex
p
lain
,
d
ep
ict,
a
n
d
an
aly
ze
th
e
d
ata
,
as
well
as tr
ain
an
d
test
th
e
alg
o
r
ith
m
.
b)
E
v
a
l
u
a
t
i
o
n
c
r
i
t
e
r
i
a
:
In
ad
d
itio
n
to
s
tan
d
ar
d
ized
er
r
o
r
m
ea
s
u
r
e
m
en
ts
s
u
ch
as,
MA
E
,
MSE
,
an
d
R
MSE
,
two
co
m
m
o
n
l
y
u
s
ed
s
o
f
twar
e
esti
m
atin
g
cr
iter
ia
m
ea
n
m
ag
n
itu
d
e
r
elativ
e
er
r
o
r
,
or
MM
R
E
,
an
d
p
er
ce
n
tag
e
r
elativ
e
er
r
o
r
d
ev
iatio
n
,
o
r
PR
E
D
wer
e
em
p
lo
y
e
d
to
ev
al
u
ate
th
e
p
r
o
d
u
ce
d
m
o
d
els.
A
b
o
v
e
all,
th
ey
allo
w
f
o
r
t
h
e
co
m
p
ar
is
o
n
o
f
r
e
s
u
lts
f
r
o
m
m
u
ltip
le
p
r
ed
ictio
n
m
o
d
els
an
d
d
at
asets
s
in
ce
th
ey
ar
e
s
ca
le
an
d
u
n
it
in
d
ep
en
d
en
t.
B
o
th
ar
e
b
ased
o
n
MRE,
wh
ich
is
ex
p
lain
e
d
b
elo
w
an
d
m
ea
s
u
r
es
th
e
d
is
p
ar
ity
b
etwe
en
ac
tu
als
an
d
esti
m
a
t
e
s
.
Me
an
ab
s
o
lu
te
er
r
o
r
(
MA
E
)
:
MA
E
is
a
wid
ely
u
s
ed
s
ta
tis
ti
c
th
at
f
in
d
s
th
e
m
ea
n
s
q
u
ar
ed
d
if
f
er
en
ce
b
etwe
en
ex
p
e
cted
an
d
ac
t
u
al
v
alu
es
by
av
er
a
g
in
g
th
e
s
q
u
ar
ed
d
is
p
ar
ities
.
I
t
ev
alu
ates
th
e
o
v
er
all
ac
cu
r
ac
y
o
f
th
e
p
r
ed
ictiv
e
m
o
d
el
in
(
2
)
.
=
(
1
)
∗
∑
=
1
…
.
(
pr
e
dic
te
d
i
−
a
c
tua
l
i
)
^
2
(
2
)
w
h
e
r
e
n
is
t
h
e
t
o
t
a
l
n
u
m
b
e
r
of
o
b
s
e
r
v
a
t
i
o
n
s
,
y
is
t
h
e
ac
t
u
a
l
v
a
l
u
e
of
s
a
m
p
l
e
i,
a
n
d
y^
is
th
e
p
r
ed
ictio
n
m
ad
e
by
th
e
m
o
d
el
f
o
r
s
am
p
le
i.
R
MSE
:
T
o
co
m
p
u
te
th
e
v
al
u
e
o
f
it,
y
o
u
will
r
e
q
u
ir
e
th
e
ac
tu
al
v
alu
es
an
d
th
eir
ex
p
e
c
t
e
d
v
a
l
u
e
s
,
s
h
o
w
n
in
(
3
)
.
W
h
e
r
e
n
is
t
h
e
to
t
a
l
n
u
m
b
e
r
of
d
at
a
p
o
i
n
t
s
,
^2
is
t
h
e
s
q
u
a
r
e
o
f
th
e
d
if
f
er
en
ce
,
an
d
is
th
e
s
u
m
o
f
s
q
u
ar
e
d
if
f
e
r
en
ce
s
in
th
e
d
atasets
.
=
∑
(
pr
e
d
ic
te
d
i
−
a
c
tua
l
i
)
^
2
/
(
3
)
Ma
g
n
itu
d
e
o
f
r
elativ
e
er
r
o
r
(
MRE
)
:
Dete
r
m
in
e
th
e
Ma
g
n
it
u
d
e
o
f
R
elativ
e
E
r
r
o
r
(
4
)
f
o
r
ea
ch
d
ata
p
o
in
t in
o
r
d
er
to
ass
ess
th
e
d
eg
r
ee
o
f
esti
m
atin
g
er
r
o
r
in
a
s
i
n
g
le
e
s
t
i
m
a
t
e
.
=
|
(
pr
e
dic
te
d
i
−
a
c
tua
l
i
|
/
a
c
tua
l
(
4)
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.
39
,
No
.
3
,
Sep
tem
b
er
20
25
:
1
7
2
4
-
1
7
3
5
1730
Me
an
m
ag
n
it
u
d
e
o
f
th
e
r
elati
v
e
er
r
o
r
(
MM
R
E
):
T
h
e
m
ea
n
m
ag
n
itu
d
e
o
f
th
e
r
elativ
e
e
r
r
o
r
(
5
)
is
th
e
av
er
ag
e
p
r
o
p
o
r
tio
n
of
th
e
a
b
s
o
lu
te
v
alu
es
of
th
e
r
elativ
e
er
r
o
r
s
ac
r
o
s
s
th
e
co
m
p
lete
d
ata
s
et.
=
(
100
/
N
)
∗
∑
/
|
pr
e
d
ic
te
d
i
−
a
c
tua
l
i
|
/
a
c
tua
l
i
(
5
)
wh
er
e
,
N
=
to
tal
n
u
m
b
er
o
f
est
im
ate
PR
E
D(
n
)
Pre
d
ictio
n
Acc
u
r
ac
y
:
T
o
d
eter
m
in
e
th
e
ac
c
u
r
ac
y
r
ate
PR
E
D(
n
)
,
d
iv
id
e
th
e
to
tal
n
u
m
b
er
of
d
ata
p
o
in
ts
with
an
MRE
of
0
.
2
5
o
r
less
(
r
ep
r
esen
ted
b
y
k)
by
th
e
to
tal
n
u
m
b
er
o
f
d
a
ta
p
o
in
ts
in
th
e
s
et
(
r
ep
r
esen
ted
b
y
n
)
.
(
x
)
=
(
1
0
0
/
N
)
*Σ
i
……
N
1
if
MR
E
i
<
=
n
/
1
0
0
,
0
o
t
h
e
r
w
is
e
(
6
)
wi
th
n
=
0
.
2
5
,
th
e
in
(
6
)
is
PR
E
D
(
n
)
=
k
/n
)
[
43]
,
[
4
4
]
.
c)
E
x
t
r
a
c
t
i
n
g
f
e
a
t
u
r
es
f
o
r
e
s
t
i
m
a
tin
g
s
o
f
t
w
a
r
e
c
o
s
t
s
:
T
h
e
W
o
r
d
2
Vec
ap
p
r
o
ac
h
is
u
s
ed
in
th
is
s
tu
d
y
to
e
x
tr
ac
t
p
er
tin
en
t
f
ea
tu
r
es
f
r
o
m
th
e
Desh
ar
n
ais,
Kitch
en
h
am
,
an
d
Ma
x
well
d
atasets
.
Af
ter
t
h
ey
h
av
e
b
ee
n
ex
tr
ac
ted
,
th
ese
f
ea
tu
r
es
ar
e
ad
d
ed
t
o
th
e
co
s
t
ca
lcu
lati
o
n
p
r
o
ce
d
u
r
e.
T
ab
le
2
,
wh
ic
h
p
r
esen
ts
th
e
ch
ar
ac
ter
is
tics
ex
tr
ac
ted
f
r
o
m
th
r
ee
d
atasets
.
T
a
b
l
e
2
.
L
i
s
t
of
f
e
a
t
u
r
e
s
e
x
t
r
a
c
t
e
d
u
s
i
n
g
W
o
r
d
2
V
e
c
D
a
t
a
s
e
t
E
x
t
r
a
c
t
e
d
f
e
a
t
u
r
e
s
D
e
s
h
a
r
n
a
i
s
P
r
o
j
e
c
t
,
P
o
i
n
t
s
N
o
n
A
d
j
u
st
,
M
a
n
a
g
e
r
E
x
p
,
A
d
j
u
s
t
me
n
t
,
Y
e
a
r
-
E
n
d
,
Le
n
g
t
h
Tr
a
n
sac
t
i
o
n
s,
P
o
i
n
t
sA
j
u
st
,
Ef
f
o
r
t
,
Te
a
mE
x
p
M
a
x
w
e
l
l
Ef
f
o
r
t
,
H
a
r
,
Y
e
a
r
,
D
u
r
a
t
i
o
n
,
A
p
p
,
T1
4
,
S
o
u
r
c
e
,
N
l
a
n
,
T
0
6
,
T
0
5
,
T
1
5
,
T0
9
,
S
i
z
e
,
T
i
me
K
i
t
c
h
e
n
h
a
m
A
d
j
f
p
,
Es
t
i
m
a
t
e
me
t
h
o
d
,
C
l
i
e
n
t
c
o
d
e
,
Es
t
i
m
a
t
e
,
Pro
j
e
c
t
t
y
p
e
,
D
u
r
a
t
i
o
n
,
E
f
f
o
r
t
d)
S
e
l
e
ct
i
n
g
f
e
a
t
u
r
es
f
o
r
e
s
t
i
m
at
in
g
s
o
f
t
w
a
r
e
c
o
s
t
s
:
To
en
s
u
r
e
t
h
at
th
e
p
r
ed
ictiv
e
m
o
d
els
ar
e
tr
ain
ed
o
n
th
e
m
o
s
t
u
s
ef
u
l
v
a
r
iab
les
wh
ile
r
ed
u
cin
g
th
e
d
an
g
er
of
o
v
e
r
f
i
ttin
g
,
th
is
s
tu
d
y
u
s
ed
th
e
R
FE
ap
p
r
o
ac
h
to
id
en
tify
th
e
m
o
s
t
im
p
o
r
tan
t
f
e
atu
r
es
f
r
o
m
th
e
th
r
ee
d
atasets
.
R
eg
r
ess
io
n
an
aly
s
is
m
ak
es
ex
ten
s
iv
e
u
s
e
o
f
R
FE
f
o
r
r
eg
u
lar
izatio
n
an
d
v
ar
iab
le
s
elec
tio
n
.
I
t
f
u
n
ctio
n
s
b
y
r
ep
ea
ted
l
y
r
em
o
v
in
g
asp
ec
ts
th
at,
ac
co
r
d
in
g
to
a
p
r
eset
cr
iter
io
n
,
ar
e
ju
d
g
ed
u
n
n
ec
ess
ar
y
o
r
r
ed
u
n
d
an
t.
T
h
is
iter
ativ
e
p
r
o
c
ed
u
r
e
y
ield
s
a
m
o
r
e
co
m
p
r
eh
en
s
ib
le
m
o
d
el
u
n
til
ju
s
t
a
s
elec
tio
n
o
f
f
ea
tu
r
es
with
th
e
h
ig
h
est
p
r
ed
ictiv
e
p
o
ten
tial
is
lef
t.
T
ab
le
3
d
is
p
lay
s
th
e
c
h
o
s
en
c
h
ar
ac
ter
is
tics
th
at
wer
e
g
ath
e
r
ed
f
r
o
m
th
r
ee
d
is
tin
ct
d
atasets
u
s
in
g
th
e
R
FE
tech
n
iq
u
e
[
4
0
]
.
T
a
b
l
e
3.
L
i
s
t
of
f
e
a
t
u
r
e
s
u
s
i
n
g
R
F
E
D
a
t
a
s
e
t
S
e
l
e
c
t
e
d
f
e
a
t
u
r
e
s
D
e
s
h
a
r
n
a
i
s
P
r
o
j
e
c
t
,
T
r
a
n
s
a
c
t
i
o
n
s
,
E
f
f
o
r
t
,
T
e
a
m
E
x
p
,
P
o
i
n
t
s
A
-
j
u
s
t
,
P
o
i
n
t
sN
o
n
A
d
j
u
s
t
M
a
x
w
e
l
l
A
p
p
,
Ef
f
o
r
t
,
H
a
r
,
S
o
u
r
c
e
,
N
l
a
n
,
T
0
5
,
T
0
9
,
T
1
5
,
Y
e
a
r
,
D
u
r
a
t
i
o
n
,
T
i
me,
S
i
z
e
K
i
t
c
h
e
n
h
a
m
Ef
f
o
r
t
,
P
r
o
j
e
c
t
t
y
p
e
,
C
l
i
e
n
t
c
o
d
e
,
D
u
r
a
t
i
o
n
,
E
s
t
i
ma
t
e
,
A
d
j
f
p
P
e
r
f
o
r
m
a
n
c
e
e
v
a
l
u
a
t
i
o
n
o
u
t
c
o
m
e
s
:
T
h
is
s
ec
tio
n
d
is
cu
s
s
e
s
th
e
p
e
r
f
o
r
m
an
ce
m
etr
ics
of
v
ar
io
u
s
ML
m
o
d
els
on
all
th
r
ee
d
atasets
.
I
t d
escr
ib
es
d
if
f
er
e
n
t k
in
d
s
o
f
e
r
r
o
r
s
ca
lcu
lated
f
o
r
ea
ch
m
o
d
el
[4
5
]
.
a)
On
D
e
s
h
a
r
n
a
i
s
d
a
t
a
s
e
t
:
Dif
f
er
en
t
ML
m
o
d
els,
in
clu
d
in
g
L
R
,
FNN,
L
STM
,
R
NN,
DT
,
an
d
SVM,
wer
e
ev
alu
ated
o
n
th
e
Desh
ar
n
ais
d
ataset
u
s
in
g
v
ar
io
u
s
p
er
f
o
r
m
a
n
ce
m
etr
ics,
as
s
h
o
wn
in
T
ab
le
4
.
T
h
e
SVM
m
o
d
el
d
e
m
o
n
s
tr
ated
t
h
e
h
i
g
h
est
R
2
of
0
.
8
0
4
,
in
d
icatin
g
th
e
b
est
p
e
r
f
o
r
m
an
ce
b
ased
o
n
th
is
m
etr
ic.
In
ter
m
s
of
R
MSE
,
FNN
ac
h
iev
ed
th
e
l
o
west
R
MSE
v
alu
e
of
0
.
2
9
3
,
wh
ile
L
R
h
a
d
th
e
lo
west
MA
E
o
f
0
.
2
6
3
,
s
u
g
g
esti
n
g
t
h
eir
ef
f
ec
ti
v
en
ess
in
m
in
im
izin
g
p
r
ed
icti
o
n
er
r
o
r
s
f
o
r
t
h
is
d
ataset.
T
a
b
l
e
4
.
E
r
r
o
r
m
e
t
r
i
x
o
b
t
ai
n
e
d
on
d
e
s
h
a
r
n
a
i
s
d
a
ta
s
et
S
.
N
o
.
E
r
r
o
r
m
e
t
r
i
x
M
A
E
R2
R
M
S
E
1
LR
0
.
2
6
3
0
.
7
7
8
0
.
3
5
3
2
F
N
N
0
.
3
8
4
0
.
7
3
7
0
.
2
9
3
3
L
S
T
M
0
.
5
4
1
0
.
5
7
1
0
.
3
9
7
4
R
N
N
0
.
4
2
9
0
.
6
7
2
0
.
3
3
1
5
DT
0
.
4
3
2
0
.
5
3
2
0
.
3
7
2
6
S
V
M
0
.
3
3
1
0
.
8
0
4
0
.
3
3
1
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
Ma
ch
in
e
lea
r
n
in
g
a
p
p
r
o
a
ch
fo
r
co
s
t
esti
ma
tio
n
in
s
o
ftw
a
r
e
p
r
o
ject
p
la
n
n
in
g
(
A
ja
y
J
a
is
w
a
l
)
1731
Fig
u
r
e
3
,
illu
s
tr
ates
a
co
m
p
ar
is
o
n
of
d
if
f
e
r
en
t
ML
m
o
d
els
b
ased
on
e
r
r
o
r
m
e
t
r
i
c
s
a
c
r
o
s
s
t
h
e
D
e
s
h
a
r
n
a
is
d
at
a
s
et
.
I
t
f
u
r
t
h
e
r
e
m
p
h
a
s
i
z
es
t
h
e
p
e
r
f
o
r
m
a
n
c
e
v
a
r
iatio
n
s
am
o
n
g
th
e
m
o
d
el
s
,
h
ig
h
lig
h
tin
g
th
eir
s
tr
en
g
th
s
an
d
wea
k
n
ess
es
in
p
r
ed
ictin
g
s
o
f
twar
e
p
r
o
ject
co
s
ts
.
b)
O
n
M
a
x
w
e
l
l
d
a
t
as
et
:
E
r
r
o
r
m
e
t
r
i
c
s
f
o
r
v
a
r
i
o
u
s
M
L
m
o
d
e
l
s
a
p
p
l
i
e
d
t
o
t
h
e
M
a
x
w
e
l
l
d
a
ta
s
et
a
r
e
s
u
m
m
a
r
i
z
e
d
i
n
T
a
b
l
e
5
.
A
m
o
n
g
t
h
e
m
o
d
e
l
s
e
v
a
l
u
at
e
d
,
L
R
d
e
m
o
n
s
t
r
a
t
es
t
h
e
l
o
w
es
t
MA
E
o
f
0
.
4
8
3
,
i
n
d
i
c
a
t
i
n
g
t
h
e
s
m
a
l
le
s
t
a
b
s
o
l
u
t
e
d
i
f
f
e
r
e
n
c
e
b
et
w
e
e
n
p
r
e
d
i
ct
e
d
a
n
d
a
c
t
u
a
l
c
o
s
ts
.
L
R
al
s
o
e
x
h
i
b
i
ts
t
h
e
h
i
g
h
es
t
R
2
v
a
l
u
e
o
f
0
.
6
0
7
,
s
u
g
g
e
s
t
i
n
g
a
g
o
o
d
f
i
t
t
o
t
h
e
d
a
ta
.
C
o
n
v
e
r
s
e
l
y
,
t
h
e
L
S
T
M
m
o
d
e
l
y
i
el
d
s
t
h
e
h
i
g
h
es
t
MA
E
o
f
0
.
9
3
3
,
i
m
p
l
y
i
n
g
l
a
r
g
e
r
d
e
v
i
a
t
i
o
n
s
b
e
tw
e
e
n
p
r
e
d
i
c
t
e
d
a
n
d
a
c
t
u
al
c
o
s
ts
.
F
u
r
t
h
e
r
m
o
r
e
,
SV
M
s
t
a
n
d
s
o
u
t
w
i
t
h
t
h
e
l
o
w
e
s
t
R
MS
E
o
f
0
.
5
3
7
,
i
n
d
ic
a
ti
n
g
o
v
e
r
a
l
l
a
c
c
u
r
a
c
y
i
n
p
r
e
d
i
c
tio
n
s
.
T
h
e
s
e
r
es
u
l
ts
i
n
d
i
c
a
t
e
t
h
a
t L
R
a
n
d
S
VM
p
e
r
f
o
r
m
r
e
l
a
t
i
v
e
l
y
b
e
tt
e
r
i
n
t
e
r
m
s
o
f
b
o
t
h
M
A
E
a
n
d
R
M
S
E
o
n
t
h
e
M
a
x
w
e
ll
.
F
i
g
u
r
e
3
.
P
e
r
f
o
r
m
a
n
c
e
m
et
r
i
cs
c
o
m
p
a
r
i
s
o
n
o
n
th
e
d
esh
ar
n
ais
Fig
u
r
e
4
d
e
p
icts
th
e
er
r
o
r
m
et
r
ics
of
s
ev
er
al
ML
m
o
d
els
ap
p
lied
to
th
e
Ma
x
well
d
ataset.
T
h
e
g
r
a
p
h
p
r
o
v
id
es
a
co
m
p
ar
ativ
e
an
aly
s
is
o
f
th
e
p
er
f
o
r
m
an
ce
o
f
ea
c
h
m
o
d
el,
s
h
e
d
d
i
n
g
l
i
g
h
t
on
t
h
e
i
r
e
f
f
e
c
t
i
v
e
n
e
s
s
in
s
o
f
t
w
a
r
e
c
o
s
t
es
t
i
m
at
i
o
n
o
n
t
h
e
M
a
x
w
e
l
l
D
a
t
as
e
t
.
c)
On
K
i
t
c
h
e
n
h
a
m
d
a
t
as
e
t:
T
ab
le
6
d
ep
icts
th
e
p
er
f
o
r
m
a
n
ce
o
f
v
ar
io
u
s
ML
m
o
d
els
e
v
alu
ated
o
n
th
e
Kitch
en
h
am
d
ataset
u
s
in
g
d
if
f
er
en
t
e
r
r
o
r
m
etr
ics.
T
h
e
L
R
an
d
SVM
m
o
d
els
d
em
o
n
s
tr
ated
th
e
b
es
t
p
er
f
o
r
m
an
ce
,
with
L
R
h
av
in
g
th
e
lo
west
MA
E
o
f
0
.
2
0
1
an
d
SVM
h
av
in
g
th
e
lo
west
R
MSE
of
0
.
2
7
4
.
B
o
th
LR
an
d
SVM
al
s
o
ac
h
iev
ed
a
h
i
g
h
R
2
of
ar
o
u
n
d
0
.
9
2
9
,
in
d
icatin
g
a
g
o
o
d
f
it
to
th
e
d
ata.
Fi
g
u
r
e
4
.
Per
f
o
r
m
an
c
e
m
etr
ics co
m
p
ar
is
o
n
o
n
t
h
e
Ma
x
well
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.
39
,
No
.
3
,
Sep
tem
b
er
20
25
:
1
7
2
4
-
1
7
3
5
1732
T
a
b
l
e
5
.
E
r
r
o
r
m
e
t
r
i
x
o
b
t
ai
n
e
d
on
m
a
x
w
e
l
l
d
a
t
as
e
t
S
.
N
o
.
E
r
r
o
r
m
e
t
r
i
x
M
A
E
R2
R
M
S
E
1
LR
0
.
4
8
3
0
.
6
0
7
0
.
5
8
8
2
F
N
N
0
.
8
1
8
0
.
2
3
9
0
.
6
6
9
3
L
S
T
M
0
.
9
3
3
0
.
0
2
1
0
.
6
8
9
4
R
N
N
0
.
9
2
7
0
.
0
2
4
0
.
6
8
6
5
DT
0
.
9
2
9
0
.
0
1
7
0
.
7
0
6
6
S
V
M
0
.
3
3
1
0
.
8
0
4
0
.
3
3
1
T
h
e
FNN
an
d
R
NN
ex
h
ib
ited
r
elativ
ely
h
ig
h
er
MA
E
of
0
.
2
4
9
an
d
0
.
3
0
7
,
r
esp
ec
tiv
ely
,
an
d
lo
wer
R
2
v
alu
es.
No
tab
ly
,
th
e
L
STM
an
d
DT
m
o
d
els
s
h
o
we
d
r
elativ
el
y
p
o
o
r
er
p
e
r
f
o
r
m
an
ce
an
d
ex
h
ib
ited
co
n
s
id
er
a
b
l
y
lo
w
R
-
s
q
u
ar
ed
v
alu
es
o
f
0
.
8
4
1
an
d
0
.
7
6
1
,
r
esp
ec
tiv
ely
,
s
u
g
g
esti
n
g
a
p
o
o
r
f
it
to
th
e
Kitch
en
h
am
d
ataset.
Fig
u
r
e
5
d
is
p
lay
s
th
e
e
r
r
o
r
m
etr
ics
o
f
m
u
ltip
le
ML
m
o
d
els
wh
en
ap
p
lied
to
th
e
Kitch
en
h
am
d
ataset.
T
h
e
g
r
ap
h
p
r
o
v
i
d
es a
co
m
p
ar
ativ
e
an
aly
s
is
o
f
th
e
p
er
f
o
r
m
an
ce
o
f
ea
ch
m
o
d
el,
d
e
m
o
n
s
tr
atin
g
th
eir
ef
f
ec
tiv
en
ess
in
s
o
f
twar
e
co
s
t e
s
tim
atio
n
o
n
th
e
Kitch
en
h
am
d
ataset.
T
ab
le
6
.
E
r
r
o
r
m
etr
ix
o
b
tain
ed
o
n
k
itch
e
n
h
am
d
ataset
S
.
N
o
.
E
r
r
o
r
m
e
t
r
i
x
M
A
E
R2
R
M
S
E
1
LR
0
.
2
0
1
0
.
9
2
9
0
.
2
7
5
2
F
N
N
0
.
2
4
9
0
.
9
1
0
0
.
3
0
9
3
L
S
T
M
0
.
4
3
1
0
.
8
4
1
0
.
3
8
9
4
R
N
N
0
.
3
0
7
0
.
8
5
8
0
.
3
8
7
5
DT
0
.
3
2
6
0
.
7
6
1
0
.
4
0
6
6
S
V
M
0
.
2
0
2
0
.
9
2
9
0
.
2
7
4
Fig
u
r
e
5
.
Per
f
o
r
m
an
c
e
m
etr
ics
co
m
p
ar
is
o
n
o
n
t
h
e
Kitch
en
h
a
m
Data
s
et
Dis
cu
s
s
io
n
: P
er
f
o
r
m
an
ce
ev
alu
atio
n
o
u
tco
m
es sh
o
wed
th
at
d
if
f
er
en
t M
L
m
o
d
els p
er
f
o
r
m
e
d
b
etter
o
n
th
e
Desh
ar
n
ais
d
ataset,
w
ith
th
e
SVM
m
o
d
el
s
h
o
win
g
th
e
h
ig
h
est
R
2
v
alu
e
o
f
0
.
8
0
4
,
in
d
icatin
g
th
e
b
est
p
er
f
o
r
m
an
ce
b
ased
o
n
th
is
m
etr
ic.
On
th
e
Ma
x
well
d
ataset,
L
R
d
em
o
n
s
tr
ated
th
e
lo
we
s
t
MA
E
o
f
0
.
4
8
3
,
in
d
icatin
g
th
e
s
m
allest
ab
s
o
lu
te
d
if
f
er
e
n
ce
b
etwe
en
p
r
ed
i
cted
an
d
ac
tu
al
c
o
s
ts
.
L
STM
m
o
d
el
y
ield
e
d
th
e
h
ig
h
est
MA
E
o
f
0
.
9
3
3
,
im
p
ly
in
g
lar
g
er
d
ev
iatio
n
s
b
etwe
en
p
r
e
d
icted
a
n
d
ac
tu
al
c
o
s
ts
.
SVM
s
to
o
d
o
u
t
with
th
e
lo
west
R
MSE
o
f
0
.
5
3
7
,
i
n
d
icatin
g
o
v
er
all
ac
cu
r
ac
y
in
p
r
ed
ictio
n
s
.
On
th
e
Kitch
en
h
am
d
ataset,
th
e
L
R
an
d
SVM
m
o
d
els
d
em
o
n
s
tr
at
ed
th
e
b
est
p
er
f
o
r
m
an
ce
,
with
L
R
h
av
in
g
th
e
lo
west
MA
E
o
f
0
.
2
0
1
an
d
SVM
h
av
in
g
th
e
lo
west
R
MSE
o
f
0
.
2
7
4
.
FNN
an
d
R
NN
ex
h
ib
i
ted
r
elativ
ely
h
ig
h
er
MA
E
an
d
lo
wer
R
2
v
alu
es,
wh
ile
L
STM
an
d
DT
m
o
d
el
s
s
h
o
wed
p
o
o
r
e
r
p
er
f
o
r
m
an
c
e
an
d
lo
w
R
-
s
q
u
ar
ed
v
alu
es.
Ho
wev
er
,
f
u
r
th
er
r
esear
ch
co
u
ld
ex
p
lo
r
e
e
n
s
em
b
le
l
e
a
r
n
i
n
g
tech
n
iq
u
es
a
n
d
d
e
ep
lear
n
i
n
g
ar
c
h
itectu
r
es
to
e
n
h
an
ce
th
e
ac
cu
r
ac
y
an
d
r
o
b
u
s
tn
ess
of
s
o
f
twar
e
c
o
s
t e
s
tim
atio
n
m
o
d
els [
4
6
]
,
[
4
7
]
.
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
Ma
ch
in
e
lea
r
n
in
g
a
p
p
r
o
a
ch
fo
r
co
s
t
esti
ma
tio
n
in
s
o
ftw
a
r
e
p
r
o
ject
p
la
n
n
in
g
(
A
ja
y
J
a
is
w
a
l
)
1733
6.
CO
NCLU
SI
O
N
T
h
e
co
m
p
ar
is
o
n
s
tu
d
y
r
esu
lt
s
d
em
o
n
s
tr
ate
s
ig
n
if
ican
tly
h
ig
h
er
ac
cu
r
ac
y
o
n
ly
i
n
th
r
ee
s
tag
es
o
f
ev
alu
atio
n
i
n
th
e
p
r
esen
ce
o
f
n
u
m
er
o
u
s
lear
n
in
g
m
eth
o
d
o
lo
g
ies.
B
ased
o
n
th
e
Per
f
o
r
m
a
n
ce
Me
tr
ics
C
o
m
p
ar
is
o
n
o
f
Kitch
e
n
h
am
d
ataset
r
esu
lts
o
b
tain
ed
f
r
o
m
th
e
e
v
alu
atio
n
o
f
v
a
r
io
u
s
ML
m
o
d
els
o
n
th
e
Desh
ar
n
ais,
Ma
x
well,
an
d
Kit
ch
en
h
am
d
atasets
,
it
is
ev
id
en
t
th
at
th
e
ch
o
ice
o
f
m
o
d
el
s
ig
n
if
ican
tly
im
p
ac
ts
th
e
ac
cu
r
ac
y
a
n
d
ef
f
ec
tiv
en
e
s
s
o
f
s
o
f
twar
e
co
s
t
esti
m
atio
n
.
I
n
th
e
Desh
ar
n
ais
d
ataset,
th
e
SVM
m
o
d
e
l
o
u
tp
er
f
o
r
m
ed
o
th
er
s
with
th
e
h
ig
h
est
R
2
v
alu
e
o
f
0
.
8
0
4
,
in
d
i
ca
tin
g
s
u
p
er
io
r
p
r
e
d
ictiv
e
ca
p
ab
ilit
y
.
C
o
n
v
er
s
ely
,
th
e
Ma
x
well
d
ataset
s
h
o
wca
s
ed
L
R
an
d
S
VM
as
th
e
to
p
p
er
f
o
r
m
er
s
,
with
L
R
d
em
o
n
s
tr
atin
g
th
e
lo
west
MA
E
o
f
0
.
4
8
3
an
d
th
e
h
ig
h
est
R
2
v
a
lu
e
o
f
0
.
9
2
9
a
n
d
SVM
ex
h
ib
itin
g
th
e
lo
west
R
MSE
of
0
.
5
3
7
.
On
th
e
o
t
h
er
h
a
n
d
,
th
e
Kitch
en
h
am
d
ataset
illu
s
tr
ated
LR
an
d
SVM
as
th
e
m
o
s
t
r
eliab
le
m
o
d
els,
d
is
p
lay
in
g
th
e
lo
west
MA
E
an
d
R
MSE
v
alu
es
an
d
h
ig
h
R
2
v
alu
es
o
f
0
.
2
0
1
,
0
.
2
7
5
,
0
.
9
2
9
,
an
d
0
.
2
0
2
,
0
.
2
7
4
,
an
d
0
.
9
2
9
r
esp
ec
tiv
ely
.
T
h
ese
f
in
d
in
g
s
em
p
h
asize
th
e
im
p
o
r
tan
ce
of
s
elec
tin
g
ap
p
r
o
p
r
ia
te
ML
m
o
d
els
tailo
r
ed
to
s
p
ec
if
ic
d
atasets
f
o
r
ac
cu
r
ate
s
o
f
twar
e
co
s
t
esti
m
a
tio
n
.
Ad
d
itio
n
ally
,
L
R
an
d
S
VM
co
n
s
is
ten
tly
em
er
g
ed
as
s
tr
o
n
g
p
er
f
o
r
m
e
r
s
ac
r
o
s
s
all
d
atasets
,
s
u
g
g
esti
n
g
th
eir
r
eliab
ilit
y
an
d
ef
f
e
ctiv
en
ess
in
th
is
d
o
m
ain
.
T
h
u
s
,
lev
er
ag
in
g
ML
tech
n
iq
u
es,
p
ar
ticu
lar
ly
LR
an
d
SVM,
h
o
ld
s
p
r
o
m
is
e
f
o
r
en
h
an
cin
g
th
e
p
lan
n
in
g
an
d
m
an
a
g
em
en
t
of
s
o
f
twar
e
p
r
o
jects
th
r
o
u
g
h
more
p
r
ec
is
e
co
s
t
esti
m
atio
n
m
eth
o
d
o
lo
g
ies
.
ACK
NO
WL
E
DG
E
M
E
NT
S
T
h
e
au
th
o
r
wo
u
ld
lik
e
to
ex
t
en
d
d
ee
p
est
ap
p
r
ec
iatio
n
t
o
“
Dr
.
Ma
n
o
jk
u
m
ar
Desh
p
an
d
e”
an
d
“Dr
.
Piy
u
s
h
C
h
au
d
h
a
r
y
”
o
f
t
h
e
P
I
E
MR
,
I
n
d
o
r
e,
f
o
r
h
is
c
o
n
tr
ib
u
tio
n
s
to
th
is
s
tu
d
y
.
T
o
ev
e
r
y
o
n
e
wh
o
m
ad
e
th
is
s
tu
d
y
p
o
s
s
ib
le,
th
e
au
th
o
r
wo
u
ld
lik
e
to
ex
ten
d
th
eir
s
i
n
ce
r
e
g
r
atitu
d
e.
No
s
p
ec
if
ic
g
r
an
t
f
r
o
m
a
p
u
b
lic,
p
r
iv
ate,
or
n
o
n
p
r
o
f
it
f
u
n
d
in
g
o
r
g
an
izatio
n
was
o
b
tain
ed
f
o
r
th
is
s
tu
d
y
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
T
h
e
au
th
o
r
s
s
tate
th
at
n
o
f
u
n
d
i
n
g
was in
v
o
lv
e
d
in
th
e
r
esear
c
h
p
r
esen
ted
i
n
th
is
m
an
u
s
cr
ip
t.
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
Ajay
J
aiswal
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
J
ag
d
is
h
R
aik
wal
✓
✓
✓
✓
✓
✓
✓
✓
✓
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
ma
l
a
n
a
l
y
si
s
I
:
I
n
v
e
st
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
s
u
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
si
o
n
P
:
P
r
o
j
e
c
t
a
d
m
i
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
s
i
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
d
ec
lar
ed
n
o
p
o
te
n
tial
co
n
f
licts
o
f
in
ter
est
co
n
c
er
n
in
g
th
is
ar
ticle’
s
r
esear
ch
,
au
th
o
r
s
h
ip
,
an
d
p
u
b
licatio
n
.
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
in
g
s
o
f
th
is
s
tu
d
y
ar
e
av
ailab
l
e
u
p
o
n
r
eq
u
est
/
p
u
b
licly
a
v
ai
lab
le
at
/
in
clu
d
ed
with
in
t
h
e
ar
ticle
o
r
i
ts
s
u
p
p
lem
en
tar
y
m
ater
ials
.
RE
F
E
R
E
NC
E
S
[
1
]
M
.
R
a
h
m
a
n
,
P
.
P
.
R
o
y
,
M
.
A
l
i
,
T
.
G
o
n
ç
a
l
v
e
s
,
a
n
d
H
.
S
a
r
w
a
r
,
“
S
o
f
t
w
a
r
e
e
f
f
o
r
t
e
s
t
i
m
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
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
A
d
v
a
n
c
e
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
1
4
,
n
o
.
4
,
p
p
.
8
2
2
–
8
2
7
,
2
0
2
3
,
d
o
i
:
1
0
.
1
4
5
6
9
/
I
J
A
C
S
A
.
2
0
2
3
.
0
1
4
0
4
9
1
.
[
2
]
A
.
Za
i
,
R
.
B
u
t
t
,
a
n
d
S
.
N
a
w
a
z
,
“
A
s
u
r
v
e
y
o
f
s
o
f
t
w
a
r
e
q
u
a
l
i
t
y
m
e
t
r
i
c
s
f
o
r
t
h
e
so
f
t
w
a
r
e
me
a
su
r
e
me
n
t
p
r
o
c
e
ss
,
”
J
o
u
rn
a
l
o
f
S
o
f
t
w
a
r
e
En
g
i
n
e
e
ri
n
g
a
n
d
I
n
t
e
l
l
i
g
e
n
t
S
y
st
e
m
s
,
v
o
l
.
2
,
n
o
.
1
,
p
p
.
4
9
–
5
6
,
2
0
1
7
.
[
3]
A
.
O
.
S
o
u
s
a
e
t
a
l
.
,
“
A
p
p
l
y
i
n
g
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
t
o
e
st
i
ma
t
e
t
h
e
e
f
f
o
r
t
a
n
d
d
u
r
a
t
i
o
n
o
f
i
n
d
i
v
i
d
u
a
l
t
a
s
k
s
i
n
s
o
f
t
w
a
r
e
p
r
o
j
e
c
t
s
,
”
i
n
I
EEE
A
c
c
e
ss,
v
o
l
.
1
1
,
p
p
.
8
9
9
3
3
-
8
9
9
4
6
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
3
.
3
3
0
7
3
1
0
.
Evaluation Warning : The document was created with Spire.PDF for Python.