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.
38
,
No
.
3
,
J
u
n
e
2
0
2
5
,
p
p
.
1
840
~
1
8
5
0
I
SS
N:
2
5
0
2
-
4
7
5
2
,
DOI
: 1
0
.
1
1
5
9
1
/ijeecs.v
38
.i
3
.
pp
1
8
4
0
-
1
8
5
0
1840
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ee
cs.ia
esco
r
e.
co
m
C
r
o
p p
red
i
ct
i
o
n u
s
i
ng
an e
nha
n
ced s
ta
c
ked ensemb
l
e m
a
chi
ne le
a
rn
i
ng
m
o
del
D
.
M
a
dh
u
S
ud
ha
n
Reddy
,
N
.
Ush
a
Ra
ni
D
e
p
a
r
t
me
n
t
o
f
C
o
mp
u
t
e
r
S
c
i
e
n
c
e
a
n
d
En
g
i
n
e
e
r
i
n
g
,
S
r
i
V
e
n
k
a
t
e
sw
a
r
a
U
n
i
v
e
r
si
t
y
C
o
l
l
e
g
e
o
f
E
n
g
i
n
e
e
r
i
n
g
,
T
i
r
u
p
a
t
i
,
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
Ap
r
3
,
2
0
2
4
R
ev
is
ed
No
v
28
,
2
0
2
4
Acc
ep
ted
Feb
27
,
2
0
2
5
In
In
d
ia,
a
g
ricu
lt
u
re
is
a
m
a
jo
r
se
c
to
r
th
a
t
fu
l
fil
s
t
h
e
p
o
p
u
lati
o
n
'
s
fo
o
d
re
q
u
irem
e
n
ts
a
n
d
sig
n
ifi
c
a
n
t
ly
c
o
n
tri
b
u
tes
to
th
e
g
ro
ss
d
o
m
e
sti
c
p
ro
d
u
c
t
(G
DP)
.
Th
e
c
a
re
fu
l
se
lec
ti
o
n
o
f
c
ro
p
s
is
f
u
n
d
a
m
e
n
tal
to
m
a
x
imiz
in
g
a
g
ricu
lt
u
ra
l
y
ield
,
th
e
re
b
y
e
lev
a
ti
n
g
t
h
e
e
c
o
n
o
m
ic
v
it
a
li
ty
o
f
t
h
e
fa
rm
in
g
c
o
m
m
u
n
it
y
.
P
re
c
isio
n
a
g
ricu
lt
u
r
e
(P
A)
lev
e
ra
g
e
s
we
a
th
e
r
a
n
d
s
o
il
d
a
ta
to
in
fo
rm
c
ro
p
se
lec
ti
o
n
stra
teg
ies
.
Co
n
v
e
n
ti
o
n
a
l
m
a
c
h
in
e
lea
rn
in
g
(M
L)
m
o
d
e
ls
su
c
h
a
s
d
e
c
isio
n
tree
s
(DT)
,
su
p
p
o
rt
v
e
c
to
r
c
las
sifier,
K
-
n
e
a
re
st
n
e
ig
h
b
o
r
s
(KN
N)
,
a
n
d
e
x
trem
e
g
ra
d
ien
t
b
o
o
st
(XG
Bo
o
st)
h
a
v
e
b
e
e
n
d
e
p
lo
y
e
d
to
p
re
d
ict
th
e
b
e
st
c
r
o
p
.
Ho
we
v
e
r,
t
h
e
se
m
o
d
e
l'
s
e
ff
icie
n
c
y
is
su
b
o
p
ti
m
a
l
i
n
t
h
e
c
u
rre
n
t
c
irc
u
m
sta
n
c
e
s.
Th
e
e
n
h
a
n
c
e
d
sta
c
k
e
d
e
n
se
m
b
le
ML
m
o
d
e
l
is
a
so
p
h
ist
ica
ted
m
e
ta
-
m
o
d
e
l
th
a
t
a
d
d
re
ss
e
s
th
e
se
li
m
i
tatio
n
s.
It
h
a
rn
e
ss
e
s
th
e
p
re
d
icti
v
e
p
o
we
r
o
f
in
d
i
v
id
u
a
l
ML
m
o
d
e
ls,
stra
ti
fied
in
a
lay
e
re
d
a
rc
h
it
e
c
tu
re
to
imp
r
o
v
e
th
e
p
re
d
icti
o
n
a
c
c
u
ra
c
y
.
T
h
is
a
d
v
a
n
c
e
d
m
o
d
e
l
h
a
s
d
e
m
o
n
stra
ted
a
c
o
m
m
e
n
d
a
b
le
a
c
c
u
ra
c
y
ra
te
o
f
9
3
.
1
%
p
re
d
ictio
n
b
y
tak
i
n
g
i
n
p
u
t
o
f
1
2
s
o
il
p
a
ra
m
e
ters
su
c
h
a
s
Nitro
g
e
n
,
P
h
o
sp
h
o
ru
s,
P
o
tas
siu
m
,
a
n
d
we
a
th
e
r
p
a
ra
m
e
ters
su
c
h
a
s
tem
p
e
ra
tu
re
a
n
d
ra
in
fa
ll
,
su
b
sta
n
ti
a
ll
y
o
u
t
p
e
rfo
rm
in
g
th
e
a
c
c
u
ra
c
ies
a
c
h
iev
e
d
b
y
th
e
in
d
i
v
id
u
a
l
c
o
n
tri
b
u
ti
n
g
m
o
d
e
ls.
T
h
e
e
ffica
c
y
o
f
th
e
p
ro
p
o
se
d
m
e
ta
-
m
o
d
e
l
in
c
ro
p
se
lec
ti
o
n
b
a
se
d
o
n
a
g
ro
n
o
m
ic
p
a
ra
m
e
ters
sig
n
ifi
e
s
a
s
u
b
sta
n
ti
a
l
a
d
v
a
n
c
e
m
e
n
t,
fo
r
ti
fy
i
n
g
th
e
e
c
o
n
o
m
ic res
il
ien
c
e
o
f
In
d
ia'
s a
g
ricu
lt
u
re
.
K
ey
w
o
r
d
s
:
C
r
o
p
p
r
ed
ictio
n
Dec
is
io
n
tr
ee
Ma
ch
in
e
lear
n
in
g
Mu
lti
-
lay
er
p
er
ce
p
tr
o
n
R
an
d
o
m
f
o
r
est
Stack
ed
en
s
em
b
le
XGBo
o
s
t
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
:
D
.
Ma
d
h
u
S
u
d
h
an
R
ed
d
y
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
,
Sri
Ven
k
ateswar
a
Un
iv
er
s
ity
C
o
lleg
e
o
f
E
n
g
i
n
ee
r
in
g
T
ir
u
p
ati,
An
d
h
r
a
Pra
d
esh
,
I
n
d
i
a
E
m
ail: m
ad
h
u
.
d
ag
ad
a@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
T
h
e
ag
r
icu
ltu
r
e
is
lead
in
g
s
ec
to
r
o
f
I
n
d
ia
as 6
0
% o
f
th
e
p
o
p
u
latio
n
is
d
ir
ec
tly
an
d
in
d
ir
ec
tly
in
v
o
lv
ed
in
th
e
d
ev
elo
p
m
e
n
t
b
y
co
n
tr
ib
u
tin
g
co
n
s
id
er
a
b
ly
to
th
e
n
ati
o
n
'
s
g
r
o
s
s
d
o
m
esti
c
p
r
o
d
u
ct
(
GDP)
,
em
p
lo
y
m
en
t,
an
d
f
o
o
d
s
ec
u
r
ity
.
Ag
r
icu
ltu
r
e
h
as
b
ee
n
a
lead
in
g
a
r
ea
o
f
t
h
e
I
n
d
ian
ec
o
n
o
m
y
f
o
r
ce
n
tu
r
ies,
s
u
p
p
o
r
tin
g
th
e
liv
elih
o
o
d
s
o
f
th
e
v
ast
p
o
p
u
latio
n
o
f
ap
p
r
o
x
im
ately
1
.
4
2
b
illi
o
n
.
I
n
ter
m
s
o
f
ec
o
n
o
m
ic
im
p
o
r
tan
ce
,
ag
r
icu
ltu
r
e
c
o
n
tr
ib
u
tes
ar
o
u
n
d
1
6
-
1
7
%
to
I
n
d
ia'
s
GDP
[
1
]
.
Far
m
in
g
is
a
s
tep
-
by
-
s
tep
p
r
o
ce
s
s
th
at
s
tar
ts
f
r
o
m
p
r
ep
ar
atio
n
o
f
s
o
il,
s
elec
tio
n
o
f
a
c
r
o
p
,
s
o
win
g
,
ad
d
i
n
g
m
an
u
r
e
an
d
f
er
tili
ze
r
s
,
ir
r
ig
at
io
n
,
h
ar
v
esti
n
g
,
an
d
s
to
r
ag
e.
Far
m
er
s
u
s
u
ally
f
o
llo
w
tr
ad
itio
n
al
m
eth
o
d
s
to
s
elec
t
cr
o
p
s
b
ased
o
n
p
r
ev
io
u
s
ex
p
er
ien
ce
an
d
s
im
ilar
cr
o
p
s
o
f
s
u
r
r
o
u
n
d
in
g
f
ar
m
e
r
s
.
T
h
ese
m
eth
o
d
s
ar
e
u
n
ab
le
t
o
p
r
o
d
u
ce
b
etter
r
esu
lts
ev
er
y
tim
e.
Ar
tific
ial
in
tellig
en
ce
(
AI
)
en
ab
led
f
ar
m
cu
ltiv
atio
n
,
wh
ich
h
elp
s
f
ar
m
er
s
to
m
ak
e
p
e
r
f
ec
t
d
ec
is
io
n
s
ab
o
u
t
cr
o
p
s
elec
tio
n
,
d
is
ea
s
e
p
r
ed
ictio
n
,
an
d
p
est
d
etec
tio
n
[
2
]
.
R
ec
en
tly
,
f
a
r
m
er
s
-
in
i
tiated
d
ata
-
d
r
iv
e
n
s
tr
ateg
ies
s
u
ch
as
p
r
ec
is
io
n
ag
r
icu
ltu
r
e
(
PA)
,
wh
ich
u
s
es
AI
-
d
r
iv
en
m
eth
o
d
s
to
in
cr
e
ase
cr
o
p
y
ield
s
b
y
s
elec
tin
g
s
u
itab
le
cr
o
p
s
an
d
s
u
p
p
o
r
tin
g
th
e
n
atio
n
'
s
ec
o
lo
g
ical
f
ar
m
in
g
g
r
o
wth
.
Ma
c
h
in
e
lear
n
in
g
(
ML
)
is
a
s
u
b
-
ar
ea
o
f
AI
.
T
h
e
u
n
d
e
r
l
y
i
n
g
ap
p
licatio
n
o
f
ML
[
3
]
in
th
e
p
r
esen
t
s
tu
d
y
is
th
e
p
r
ed
ictio
n
o
f
th
e
m
o
s
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
C
r
o
p
p
r
ed
ictio
n
u
s
in
g
a
n
en
h
a
n
ce
d
s
ta
ck
ed
e
n
s
emb
le
ma
ch
in
e
lea
r
n
in
g
mo
d
el
(
D.
Ma
d
h
u
S
u
d
h
a
n
R
e
d
d
y
)
1841
s
u
itab
le
cr
o
p
s
f
o
r
cu
ltiv
atio
n
.
T
h
e
co
r
e
co
n
ce
p
t o
f
ML
is
,
to
d
ev
elo
p
a
m
o
d
el
in
s
u
ch
a
way
th
at
it lea
r
n
s
f
r
o
m
ex
p
er
ien
ce
s
a
n
d
im
p
r
o
v
es
p
er
f
o
r
m
an
ce
.
Var
io
u
s
ML
a
p
p
lic
atio
n
s
h
av
e
b
ee
n
in
tr
o
d
u
ce
d
,
en
co
m
p
ass
in
g
a
u
to
-
i
r
r
i
g
a
t
i
o
n
s
y
s
t
e
m
s
f
o
r
a
g
r
i
-
f
a
r
m
s
[
4
]
,
d
r
o
n
e
s
t
o
a
n
a
l
y
z
e
a
g
r
ic
u
l
t
u
r
a
l
l
a
n
d
[
5
]
,
m
o
n
i
t
o
r
i
n
g
s
y
s
t
e
m
s
f
o
r
c
r
o
p
s
[
6
]
,
PA
[
7
]
,
an
d
an
im
al
id
e
n
tific
atio
n
,
am
o
n
g
o
th
er
s
.
C
o
n
s
eq
u
en
t
ly
,
th
is
ap
p
r
o
ac
h
p
r
o
v
es h
ig
h
l
y
ad
v
a
n
tag
eo
u
s
f
o
r
ag
r
icu
ltu
r
al
p
r
ac
titi
o
n
er
s
.
A
f
r
am
ewo
r
k
f
o
r
cr
o
p
r
ec
o
m
m
en
d
atio
n
s
[
8
]
was
cr
ea
ted
u
s
in
g
th
e
en
s
em
b
le
m
eth
o
d
o
f
ML
.
T
h
e
o
p
tim
al
cr
o
p
is
r
ec
o
m
m
en
d
e
d
b
y
th
e
en
s
em
b
le
tech
n
iq
u
e
b
ased
o
n
th
e
p
r
o
p
er
ties
an
d
n
at
u
r
e
o
f
t
h
e
s
o
il
with
an
ac
cu
r
ac
y
o
f
9
9
%.
T
h
e
en
s
e
m
b
le
m
o
d
el
u
s
es
r
an
d
o
m
f
o
r
est
(
R
F),
Naïv
e
B
ay
es
(
NB
)
,
an
d
l
in
ea
r
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
es
(
SVM
)
as
b
ase
class
if
ier
s
,
wh
ich
ar
e
c
o
n
v
en
tio
n
al
ML
tech
n
iq
u
es.
T
h
e
d
ataset
co
n
tai
n
ed
s
am
p
les
o
f
s
u
r
f
ac
e
tem
p
er
at
u
r
e
an
d
an
n
u
al
r
ai
n
f
all
as
well
as
ch
em
ical
an
d
p
h
y
s
ical
ch
ar
ac
ter
is
tics
o
f
th
e
s
o
il.
T
h
e
b
est
alg
o
r
ith
m
s
f
o
r
cr
o
p
ca
teg
o
r
izatio
n
wer
e
f
o
u
n
d
b
y
e
v
a
lu
atin
g
a
n
d
c
o
m
p
ar
in
g
th
e
o
u
tp
u
t
o
f
s
ev
er
al
class
if
icatio
n
alg
o
r
ith
m
s
[
9
]
.
A
d
d
itio
n
ally
,
th
ey
e
x
am
in
ed
th
e
im
p
ac
t
o
f
s
u
ch
a
lg
o
r
ith
m
s
o
n
cr
o
p
p
r
ed
ictio
n
a
n
d
o
f
f
er
ed
im
p
r
o
v
ed
cr
o
p
-
r
elate
d
tactics.
L
astl
y
,
th
ey
r
ec
o
m
m
en
d
e
d
en
h
a
n
cin
g
th
e
esti
m
atio
n
an
d
r
esp
o
n
s
e
tim
e
o
f
th
e
ex
is
tin
g
m
eth
o
d
s
.
Desp
ite
o
f
h
ig
h
ac
cu
r
ac
y
o
f
t
h
e
s
y
s
tem
,
v
er
y
f
ew
m
o
d
els
wer
e
co
n
s
id
er
ed
f
o
r
th
e
en
s
em
b
le
a
p
p
r
o
ac
h
.
A
m
ajo
r
ity
v
o
tin
g
p
r
o
ce
d
u
r
e
f
o
llo
wed
b
y
an
en
s
em
b
le
ap
p
r
o
ac
h
,
NB
,
K
-
n
ea
r
est
n
eig
h
b
o
r
s
(
KNN)
,
an
d
RF
as
b
ase
lear
n
er
s
,
was
s
u
g
g
ested
b
y
a
r
ec
o
m
m
e
n
d
ed
s
y
s
tem
[
1
0
]
to
c
r
o
p
f
o
r
s
ite
-
s
p
e
cif
ic
p
ar
am
eter
s
to
r
ec
o
m
m
en
d
a
cr
o
p
with
h
ig
h
ef
f
icien
cy
.
T
h
e
r
ec
o
m
m
en
d
e
d
cr
o
p
is
b
ased
o
n
th
e
cr
o
p
y
iel
d
esti
m
atio
n
m
o
d
el
[
1
1
]
,
wh
ich
ass
ess
ed
th
e
ANN
-
GW
O
(
ar
tific
ial
n
eu
r
al
n
etwo
r
k
s
with
g
r
ey
wo
lf
o
p
tim
i
ze
r
)
'
s
ef
f
icien
cy
f
o
r
cr
o
p
y
ield
with
a
r
o
o
t
m
ea
n
s
q
u
ar
e
er
r
o
r
(
R
MSE
)
o
f
3
.
1
9
,
an
d
m
ea
n
ab
s
o
lu
te
er
r
o
r
(
M
AE
)
o
f
2
6
.
6
5
.
T
h
is
r
esear
ch
'
s
m
ain
o
b
jectiv
e
was
to
d
ev
el
o
p
a
m
o
d
u
le
th
at
w
o
u
ld
h
el
p
f
ar
m
er
s
ch
o
o
s
e
th
e
b
est
cr
o
p
f
o
r
t
h
eir
r
eg
io
n
.
H
o
wev
er
,
it
i
s
a
co
m
p
lex
m
eth
o
d
f
o
r
f
ar
m
e
r
s
to
u
s
e
f
o
r
cr
o
p
r
ec
o
m
m
e
n
d
atio
n
.
T
h
ese
m
eth
o
d
s
wer
e
r
ec
o
m
m
en
d
ed
b
ased
o
n
y
ield
p
r
ed
ictio
n
o
f
in
d
iv
id
u
al
cr
o
p
s
.
An
au
to
m
ated
c
r
o
p
r
ec
o
m
m
en
d
atio
n
we
b
s
ite
[
1
2
]
wa
s
cr
ea
ted
,
u
tili
zin
g
d
atasets
th
at
o
f
f
er
co
m
p
r
eh
e
n
s
iv
e
r
ec
o
r
d
s
o
f
v
ar
io
u
s
ar
ea
ch
ar
ac
ter
is
tics
,
d
ev
elo
p
m
e
n
t
s
p
ec
if
ics,
an
d
s
o
il
p
ar
am
eter
s
.
Dep
en
d
in
g
o
n
th
e
p
ar
a
m
eter
s
in
t
h
e
d
ataset,
t
h
eir
s
y
s
tem
m
ig
h
t
s
u
g
g
est
cr
o
p
s
.
C
r
o
p
p
r
o
jectio
n
s
co
v
er
e
d
ev
er
y
t
y
p
e
o
f
cr
o
p
g
r
o
wn
in
t
h
e
US
an
d
wer
e
n
o
t r
estricte
d
t
o
an
y
o
n
e
c
r
o
p
s
p
ec
i
es.
T
h
e
d
ataset
in
clu
d
ed
d
ata
o
n
all
cr
o
p
s
in
e
v
er
y
p
r
o
v
in
ce
at
th
e
d
is
tr
ict
lev
el,
to
talin
g
o
v
er
2
.
5
lak
h
d
o
c
u
m
en
ts
.
T
h
e
r
e
s
u
lts
d
em
o
n
s
tr
ated
th
e
ef
f
ec
tiv
en
ess
o
f
ML
tech
n
iq
u
es.
W
ith
an
ac
cu
r
ac
y
o
f
9
3
.
2
%,
RF
o
u
tp
er
f
o
r
m
ed
th
e
o
th
er
class
if
ier
s
.
T
h
e
cr
o
p
is
s
u
g
g
ested
b
y
a
s
im
p
le
an
d
b
etter
m
o
b
ile
a
p
p
licatio
n
with
a
g
r
ap
h
ic
u
s
er
in
ter
f
ac
e
(
GUI
)
in
te
g
r
ated
with
th
e
m
o
d
el,
wh
ich
h
elp
s
to
s
u
g
g
est
cr
o
p
s
[
1
3
]
b
ased
o
n
in
p
u
t
p
ar
am
eter
s
o
f
wea
th
er
a
n
d
s
o
il
d
ata.
I
t
also
to
o
k
in
to
ac
co
u
n
t c
r
o
p
cu
ltiv
at
io
n
ex
p
e
n
s
es a
n
d
th
e
lo
ca
tio
n
,
tim
e,
an
d
s
o
u
r
ce
o
f
i
r
r
ig
atio
n
.
Sev
er
al
ML
m
eth
o
d
s
[
1
4
]
u
s
e
s
o
il
d
ata
ab
o
u
t
th
e
ar
ea
to
f
o
r
ec
ast
a
s
u
itab
le
c
r
o
p
f
o
r
a
r
ec
o
m
m
en
d
atio
n
.
Sev
e
r
al
M
L
ap
p
r
o
ac
h
es
ar
e
u
s
ed
f
o
r
s
o
il
class
if
icatio
n
,
s
u
ch
as
b
ag
g
ed
tr
ee
s
,
weig
h
ted
KNN,
an
d
SVM
m
o
d
el
s
wit
h
g
au
s
s
ian
k
er
n
el
ass
is
tan
ce
.
Acc
u
r
ac
y
ca
n
b
e
in
cr
ea
s
ed
s
in
ce
th
e
cr
o
p
was
ch
o
s
en
b
y
an
aly
zi
n
g
th
e
q
u
a
n
titi
es
o
f
s
o
il
r
ath
er
th
a
n
s
o
il
t
y
p
es.
Per
f
o
r
m
an
ce
m
et
r
ics
s
u
ch
as
ac
cu
r
ac
y
an
d
F1
-
s
co
r
es
o
f
a
f
ew
ML
alg
o
r
ith
m
s
[
1
5
]
,
in
clu
d
in
g
d
ec
is
io
n
tr
ee
(
DT
)
,
SVM,
NB
,
RF
,
lo
g
is
tic
r
eg
r
ess
io
n
(
L
R
)
,
an
d
ex
tr
em
e
g
r
ad
ie
n
t
b
o
o
s
tin
g
(
XGBo
o
s
t
)
ev
alu
ate
d
,
wh
ich
u
s
e
s
o
il
d
ata
to
p
r
o
p
o
s
e
cr
o
p
s
,
XGBo
o
s
t
p
er
f
o
r
m
ed
b
etter
th
an
th
e
o
th
e
r
m
o
d
els.
Far
m
er
s
ca
n
ch
o
o
s
e
cr
o
p
s
b
y
tak
in
g
in
to
co
n
s
id
er
atio
n
s
ev
e
r
al
f
ac
to
r
s
s
u
ch
as
g
eo
g
r
ap
h
ic
lo
ca
tio
n
,
s
o
il
ty
p
e,
an
d
p
lan
tin
g
s
ea
s
o
n
b
y
u
s
in
g
a
cr
o
p
-
s
u
g
g
ested
s
y
s
tem
[
1
6
]
.
I
n
ad
d
itio
n
,
m
o
d
els
lik
e
L
R
,
NB
,
KN
N
with
cr
o
s
s
-
v
alid
atio
n
,
KNN,
DT
,
a
n
d
n
eu
r
al
n
etwo
r
k
(
NN)
ar
e
ta
k
en
in
to
co
n
s
id
er
atio
n
in
PA,
wh
ich
co
n
ce
n
tr
ates
o
n
s
ite
-
s
p
ec
if
ic
c
r
o
p
m
a
n
ag
em
en
t
.
At
8
9
.
8
8
%,
th
e
NN
h
ad
p
r
o
v
id
e
d
a
m
o
r
e
ac
cu
r
ate
r
esu
lt.
Nev
er
th
eless
,
NN
im
p
lem
en
tatio
n
is
a
ch
allen
g
in
g
p
r
o
ce
s
s
.
T
h
r
ee
s
tep
s
weig
h
t
ca
lcu
latio
n
,
ca
teg
o
r
izatio
n
,
an
d
p
r
ed
ict
io
n
m
a
k
e
u
p
th
e
c
r
o
p
s
elec
tio
n
m
et
h
o
d
[
1
7
]
,
w
h
ich
was
d
ev
elo
p
e
d
.
T
h
e
r
e
wer
e
2
7
in
p
u
t
c
r
iter
ia
in
to
tal,
wh
ich
wer
e
b
r
o
k
en
d
o
wn
in
to
7
m
ajo
r
ca
teg
o
r
ies:
f
ac
ilit
ies,
s
o
il
r
is
k
,
in
p
u
t,
s
ea
s
o
n
,
wate
r
,
an
d
s
u
p
p
o
r
t.
T
h
e
in
itial
s
tag
e
in
v
o
lv
e
d
u
tili
zin
g
th
e
r
o
u
g
h
s
et
m
eth
o
d
o
lo
g
y
t
o
ass
ess
t
h
e
r
elativ
e
weig
h
ts
o
f
ea
ch
m
ain
cr
iter
i
o
n
'
s
s
u
b
-
cr
iter
ia,
an
d
th
en
ap
p
ly
in
g
Sh
an
n
o
n
'
s
e
n
tr
o
p
y
to
d
ete
r
m
in
e
th
e
r
elativ
e
weig
h
ts
o
f
th
e
m
ain
cr
iter
ia
th
em
s
elv
es.
VI
KOR
(
v
is
ek
r
iter
iju
m
s
k
a
o
p
tim
izac
ija
i
k
o
m
p
r
o
m
is
n
o
r
esen
je)
w
as
u
s
ed
to
d
eter
m
in
e
th
e
r
an
k
in
g
i
n
d
ex
o
f
th
e
p
r
im
ar
y
c
r
iter
ia
b
ec
au
s
e
i
t
is
an
ef
f
ec
tiv
e
m
eth
o
d
f
o
r
s
o
r
tin
g
th
e
alter
n
ativ
es
an
d
a
m
u
lticr
iter
ia
o
p
tim
izatio
n
an
d
co
m
p
r
o
m
is
e
s
o
lu
tio
n
.
Un
d
er
s
tan
d
in
g
t
h
is
m
o
d
el
will tak
e
m
o
r
e
e
x
p
er
tis
e
d
u
e
to
its
co
m
p
lex
ity
.
A
cr
o
p
is
r
ec
o
m
m
e
n
d
ed
b
y
t
h
e
s
u
g
g
esti
o
n
s
y
s
tem
[
1
8
]
u
s
es
p
atter
n
m
atch
i
n
g
tech
n
i
q
u
es
to
en
ab
le
f
ar
m
er
s
to
ch
o
o
s
e
th
e
b
est
cr
o
p
f
o
r
th
e
s
o
win
g
a
r
ea
an
d
s
ea
s
o
n
.
Far
m
er
s
g
et
b
en
e
f
its
f
r
o
m
it
as
a
r
esu
lt
s
in
ce
th
eir
n
et
p
r
o
f
it
will
in
cr
ea
s
e.
T
h
e
s
y
s
tem
i
s
ca
p
ab
le
o
f
r
ec
o
m
m
en
d
i
n
g
a
v
ar
iety
o
f
c
r
o
p
s
th
at
ar
e
m
o
s
t
ad
v
an
tag
e
o
u
s
to
p
r
o
d
u
ce
r
s
in
t
h
eir
d
ec
is
io
n
-
m
ak
i
n
g
p
r
o
ce
s
s
.
T
h
is
is
ac
co
m
p
lis
h
ed
b
y
an
aly
z
in
g
a
d
ataset
th
at
p
r
im
ar
ily
c
o
m
p
r
is
es
f
iv
e
cr
iter
ia:
r
ain
f
all,
s
o
il
m
o
is
tu
r
e,
tem
p
er
atu
r
e,
s
lo
p
e,
a
n
d
h
u
m
id
it
y
d
ata
v
alu
es
th
at
ar
e
ass
o
ciate
d
with
h
o
r
ticu
ltu
r
e
.
W
h
en
s
o
il
p
ar
am
eter
s
d
if
f
er
f
r
o
m
o
n
e
f
a
r
m
to
an
o
t
h
er
,
t
h
en
t
h
e
p
atter
n
-
m
atch
in
g
tech
n
iq
u
e
m
a
y
n
o
t b
e
ap
p
r
o
p
r
i
ate
to
co
n
s
id
er
in
th
e
m
o
d
el.
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.
38
,
No
.
3
,
J
u
n
e
20
25
:
1
8
4
0
-
1
8
5
0
1842
T
h
e
cr
o
p
s
elec
tio
n
m
eth
o
d
(
C
SM)
[
1
9
]
p
r
o
p
o
s
ed
to
s
o
lv
e
t
h
e
cr
o
p
s
elec
tio
n
p
r
o
b
lem
to
m
ax
im
ize
th
e
y
ield
o
f
cr
o
p
s
in
a
s
ea
s
o
n
.
I
t
led
t
o
m
a
x
im
u
m
ec
o
n
o
m
ic
im
p
r
o
v
em
en
t
f
o
r
th
e
n
atio
n
.
So
il
ch
ar
ac
ter
is
tics
ar
e
ig
n
o
r
ed
b
y
th
e
m
eth
o
d
in
th
e
cr
o
p
s
elec
tio
n
p
r
o
ce
s
s
,
th
o
u
g
h
it
is
an
im
p
o
r
tan
t
p
ar
am
eter
.
T
o
p
r
ed
ict
th
e
b
est
cr
o
p
(
s
)
f
o
r
th
e
ar
ea
,
a
c
o
m
p
ar
ativ
e
[
2
0
]
b
ased
an
aly
s
is
o
f
s
ev
er
al
wr
ap
p
er
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
s
alo
n
g
with
ML
class
if
icatio
n
t
ec
h
n
iq
u
es
was
c
o
n
d
u
cted
.
B
ased
o
n
th
e
r
esu
lts
o
f
th
e
tr
ial,
th
e
r
ec
u
r
s
iv
e
f
ea
tu
r
e
elim
in
atio
n
tech
n
iq
u
e
in
co
n
j
u
n
ctio
n
with
th
e
a
d
ap
tiv
e
b
a
g
g
in
g
class
if
ier
o
u
tp
e
r
f
o
r
m
s
th
e
o
th
er
a
n
aly
tical
ap
p
r
o
ac
h
es.
T
h
e
ac
cu
r
ac
y
o
f
th
is
ap
p
r
o
ac
h
ca
n
b
e
in
cr
ea
s
e
d
b
y
ad
ju
s
tin
g
th
e
h
y
p
er
-
p
ar
a
m
eter
s
o
f
th
e
ML
m
o
d
els.
T
h
e
d
ee
p
lear
n
in
g
tech
n
iq
u
e
(
DL
T
)
b
ased
cr
o
p
-
s
p
ec
if
ic
r
e
co
m
m
en
d
e
r
s
y
s
tem
[
2
1
]
b
y
c
o
n
s
id
er
in
g
h
is
to
r
ical
cr
o
p
an
d
clim
ate
d
ata.
AC
O
-
I
DC
NN
-
L
STM
,
a
h
y
b
r
id
tech
n
iq
u
e
co
m
b
in
in
g
a
n
t c
o
lo
n
y
o
p
tim
izatio
n
(
AC
O)
with
d
ee
p
co
n
v
o
lu
tio
n
n
e
u
r
al
n
etwo
r
k
s
(
DC
NN)
an
d
lo
n
g
s
h
o
r
t
-
ter
m
m
em
o
r
y
(
L
STM
)
,
h
as
b
ee
n
p
r
o
p
o
s
ed
f
o
r
c
r
o
p
p
r
e
d
ictio
n
in
DL
T
,
L
STM
,
an
d
DC
C
N
n
etwo
r
k
s
.
Hig
h
ac
cu
r
ac
y
le
v
e
ls
,
ty
p
ically
9
5
.
1
%,
a
r
e
o
f
te
n
a
c
h
ie
v
ed
b
y
D
C
NN
s
.
T
h
e
r
e
a
r
e
ad
d
i
t
io
n
a
l
l
ay
e
r
s
an
d
N
N
o
p
e
r
a
t
io
n
s
i
n
v
o
lv
e
d
i
n
t
h
i
s
p
r
o
ce
s
s
.
I
m
p
l
em
e
n
t
in
g
C
N
N
a
n
d
L
S
T
M
i
s
t
h
er
e
f
o
r
e
h
ig
h
ly
co
m
p
le
x
.
A
d
d
i
t
i
o
n
a
l
l
y
,
th
e
r
e
i
s
a
h
ig
h
A
C
O
c
o
n
v
e
r
g
e
n
c
e
r
a
t
e
.
A
s
tu
d
y
m
eth
o
d
o
lo
g
y
co
m
b
i
n
in
g
m
ac
h
in
e
lear
n
i
n
g
an
d
d
ata
b
alan
ce
was
p
u
t
o
u
t
[
2
2
]
f
o
r
cr
o
p
r
ec
o
m
m
en
d
atio
n
.
1
4
ML
m
o
d
els
ar
e
test
ed
u
s
in
g
Kag
g
le
d
ata,
an
d
b
o
o
s
tin
g
(
C
b
o
o
s
t)
o
b
tain
s
th
e
h
ig
h
est
ac
cu
r
ac
y
(
9
9
.
1
5
%),
F
-
m
ea
s
u
r
e
(
0
.
9
9
1
6
)
,
an
d
p
r
ec
is
io
n
(
0
.
9
9
1
8
)
.
Gau
s
s
ian
Naïv
e
B
ay
es
(
GNB
)
d
o
es
well
in
M
atth
ews
co
r
r
elatio
n
co
ef
f
i
cien
t
(
MCC
)
an
d
r
ec
eiv
er
o
p
er
atin
g
ch
ar
ac
te
r
is
tic
(
R
OC
)
(
0
.
9
5
6
9
)
.
M
o
s
t
class
if
ier
s
to
o
k
in
to
ac
co
u
n
t
a
s
m
all
n
u
m
b
er
o
f
f
ac
to
r
s
w
h
en
s
u
g
g
esti
n
g
c
r
o
p
s
.
A
cr
o
p
r
ec
o
m
m
e
n
d
atio
n
s
y
s
tem
(
C
R
S)
[
2
3
]
f
o
r
Ma
h
ar
ash
tr
a
th
at
im
p
r
o
v
es
f
ar
m
er
p
r
o
d
u
ctio
n
b
y
u
tili
zin
g
d
ata
f
r
o
m
2
0
0
1
–
2
0
2
2
.
B
y
u
s
in
g
DL
an
d
ML
,
s
u
ch
as
RF
f
o
r
9
2
%
ac
cu
r
ac
y
an
d
L
STM
f
o
r
wea
th
er
f
o
r
ec
asti
n
g
,
th
e
C
R
S
en
h
an
ce
s
ag
r
icu
ltu
r
al
e
f
f
icien
cy
b
y
r
ec
o
m
m
en
d
in
g
th
e
b
est
cr
o
p
s
b
ase
d
o
n
lo
ca
l
co
n
d
itio
n
s
.
T
h
is
m
o
d
el
is
lim
ited
t
o
a
r
elativ
ely
s
m
all
n
u
m
b
er
o
f
cr
o
p
s
.
B
y
m
ak
in
g
i
n
f
o
r
m
ed
ju
d
g
em
en
ts
r
eg
a
r
d
in
g
i
r
r
ig
a
tio
n
,
p
lan
tin
g
,
an
d
h
ar
v
esti
n
g
,
th
e
ML
p
r
ed
ictio
n
m
o
d
el
[
2
4
]
in
a
g
r
icu
ltu
r
e
im
p
r
o
v
es c
r
o
p
p
r
o
d
u
ctio
n
.
T
h
e
m
o
d
el
em
p
h
asizes th
e
p
o
ten
tial
o
f
in
te
g
r
atin
g
in
ter
n
et
o
f
th
in
g
s
(
I
o
T
)
d
ata
an
d
o
n
lin
e
r
eso
u
r
ce
s
to
en
h
a
n
ce
ac
cu
r
ac
y
,
attain
in
g
a
class
if
icatio
n
ac
cu
r
ac
y
o
f
u
p
t
o
9
9
.
5
9
% u
s
in
g
alg
o
r
ith
m
s
s
u
ch
as B
ay
es Ne
t
c
lass
if
ier
.
A
cr
o
p
r
ec
o
m
m
en
d
atio
n
s
y
s
te
m
[
2
5
]
was
p
r
o
p
o
s
ed
to
ass
is
t
p
r
o
d
u
ce
r
s
in
m
ak
i
n
g
in
f
o
r
m
e
d
d
ec
is
io
n
s
b
y
u
tili
zin
g
ML
.
T
h
e
s
y
s
tem
h
as
th
e
p
o
ten
tial
to
i
n
cr
ea
s
e
cr
o
p
o
u
tp
u
t
an
d
r
e
d
u
ce
co
s
ts
in
th
e
f
ac
e
o
f
ch
allen
g
es
s
u
ch
as
p
o
p
u
lati
o
n
g
r
o
wth
b
y
p
r
ed
ictin
g
ag
r
icu
ltu
r
al
y
ield
s
an
d
s
u
g
g
est
in
g
o
p
tim
u
m
c
r
o
p
m
an
ag
em
en
t
p
r
ac
tices
in
th
e
c
o
n
tex
t o
f
al
g
o
r
ith
m
s
s
u
ch
as
DT
,
NB
,
an
d
RF
.
T
h
e
ap
p
licat
io
n
o
f
d
ata
an
al
y
tics
tech
n
iq
u
es,
s
u
ch
as
LR
w
ith
NN
,
was
em
p
lo
y
ed
to
f
o
r
ec
ast
cr
o
p
p
r
ices
[
2
6
]
,
tak
in
g
in
to
ac
co
u
n
t
f
ac
to
r
s
s
u
ch
as
th
e
ar
ea
h
ar
v
ested
an
d
p
lan
ted
.
T
h
e
s
tu
d
y
d
eter
m
in
ed
th
a
t
XGBo
o
s
t
wa
s
th
e
m
o
s
t
ef
f
ec
tiv
e
tech
n
iq
u
e
f
o
r
p
r
ice
p
r
e
d
ictio
n
.
Utilizin
g
a
v
a
r
iety
o
f
v
is
u
aliza
tio
n
to
o
ls
,
a
m
o
d
el
[
2
7
]
i
n
co
r
p
o
r
ated
m
o
b
ile
ap
p
licatio
n
s
an
d
ML
to
ass
is
t
f
ar
m
er
s
in
id
en
tify
in
g
t
h
e
m
o
s
t
ef
f
ec
tiv
e
co
n
d
itio
n
s
f
o
r
p
lan
tin
g
,
h
ar
v
esti
n
g
,
an
d
f
e
r
tili
zin
g
cr
o
p
s
.
T
h
is
m
o
d
el
ca
n
also
b
e
m
o
d
if
ied
t
o
p
r
o
v
id
e
f
er
tili
ze
r
r
ec
o
m
m
en
d
atio
n
s
.
A
r
eg
r
ess
io
n
-
b
ased
ML
s
y
s
tem
[
2
8
]
th
at
em
p
lo
y
s
NB
class
if
ier
s
to
f
o
r
ec
ast
f
er
tili
ze
r
u
s
ag
e
an
d
cr
o
p
y
ield
f
o
r
cr
o
p
s
in
My
s
o
r
e
b
ased
o
n
s
o
il
n
u
tr
ie
n
ts
h
as
d
em
o
n
s
tr
ated
h
ig
h
ac
cu
r
a
cy
f
o
r
w
h
ea
t,
r
ag
i,
a
n
d
p
ad
d
y
.
T
h
ese
m
o
d
el
s
ca
n
also
b
e
im
p
r
o
v
e
d
to
cr
ea
te
u
s
er
-
f
r
ien
d
l
y
ap
p
licatio
n
s
s
p
ec
if
ically
d
esig
n
ed
to
m
ee
t th
e
r
eq
u
ir
em
en
ts
o
f
p
r
o
d
u
ce
r
s
.
E
x
is
tin
g
s
o
lu
tio
n
s
ar
e
d
e
v
elo
p
ed
u
s
in
g
s
y
n
t
h
etic
d
ata
t
o
m
o
d
el
d
esig
n
f
o
r
c
r
o
p
r
ec
o
m
m
en
d
atio
n
.
T
h
ese
m
o
d
els
ar
e
u
n
ab
le
to
co
n
s
id
er
o
th
e
r
s
o
il
p
ar
am
eter
s
lik
e
C
o
p
p
er
an
d
Su
lp
h
u
r
.
Few
s
o
lu
tio
n
s
u
tili
ze
d
DL
m
eth
o
d
s
f
o
r
cr
o
p
r
ec
o
m
m
en
d
atio
n
.
Par
ticu
lar
ly
in
d
e
v
elo
p
in
g
n
atio
n
s
,
it
is
im
p
er
ativ
e
to
cu
s
to
m
ize
r
ec
o
m
m
en
d
atio
n
s
f
o
r
s
m
all
-
s
c
ale
an
d
s
u
b
s
is
ten
ce
f
ar
m
er
s
.
S
ig
n
if
ican
t
co
m
p
u
tatio
n
al
r
eso
u
r
ce
s
ar
e
n
ec
ess
ar
y
f
o
r
n
u
m
er
o
u
s
s
o
p
h
is
ticated
m
o
d
els,
p
ar
ticu
lar
ly
th
o
s
e
th
at
i
n
v
o
lv
e
d
ee
p
lear
n
in
g
.
R
esear
ch
co
u
ld
in
v
esti
g
ate
m
eth
o
d
s
to
d
ec
r
ea
s
e
th
e
e
n
e
r
g
y
co
n
s
u
m
p
tio
n
o
f
th
ese
m
o
d
els,
p
ar
ticu
lar
ly
in
d
ev
elo
p
in
g
co
u
n
tr
ies
with
r
estricte
d
ac
ce
s
s
to
h
ig
h
-
p
er
f
o
r
m
an
ce
co
m
p
u
tatio
n
.
C
o
n
s
id
er
in
g
all
g
ap
s
in
th
e
e
x
is
tin
g
liter
atu
r
e,
a
p
r
o
p
o
s
ed
m
o
d
el
n
ee
d
s
to
b
e
less
co
m
p
lex
,
m
o
r
e
ac
cu
r
ate
an
d
co
n
s
id
er
all
s
o
il
an
d
wea
th
er
p
ar
am
eter
s
.
T
h
e
k
ey
co
n
tr
ib
u
tio
n
s
o
f
th
is
s
tu
d
y
h
av
e
b
ee
n
lis
ted
b
elo
w:
−
Pro
p
o
s
e
an
en
h
an
ce
d
s
tack
ed
en
s
em
b
le
m
o
d
el
f
o
r
cr
o
p
p
r
e
d
ictio
n
with
h
ig
h
ac
cu
r
ac
y
b
y
co
m
p
ar
in
g
s
ix
ML
m
o
d
els,
m
u
lti
-
lay
er
p
er
ce
p
tr
o
n
(
ML
P)
,
XGBo
o
s
t,
KNN
,
DT
,
an
d
SVM
.
−
Sev
en
d
if
f
e
r
en
t
c
r
o
p
s
h
av
e
b
ee
n
class
if
ied
f
o
r
th
e
p
r
ed
ict
io
n
,
b
ased
o
n
in
p
u
t
s
o
il
p
ar
a
m
eter
s
s
u
ch
as
Nitr
o
g
en
,
Ph
o
s
p
h
o
r
u
s
,
Po
tass
iu
m
,
p
H,
Ma
n
g
an
ese,
Or
g
an
ic
C
ar
b
o
n
,
Z
in
c,
E
lectr
ical
C
o
n
d
u
ctiv
ity
,
I
r
o
n
,
B
o
r
o
n
,
C
o
p
p
e
r
,
Su
lp
h
u
r
,
a
n
d
wea
th
er
p
ar
am
eter
s
s
u
ch
as r
ai
n
f
all,
an
d
tem
p
er
atu
r
e.
T
h
e
p
ap
er
h
as
b
ee
n
s
tr
u
ct
u
r
ed
as
f
o
llo
ws:
s
ec
tio
n
2
an
aly
s
es
th
e
liter
atu
r
e
r
ev
iew
o
f
cr
o
p
p
r
ed
ictio
n
o
r
r
ec
o
m
m
e
n
d
atio
n
s
y
s
tem
s
.
Sectio
n
3
ex
p
lain
s
th
e
d
etails
o
f
th
e
p
r
o
p
o
s
ed
m
o
d
el
alo
n
g
with
o
th
er
ML
m
o
d
e
l
s
w
it
h
a
n
a
l
y
s
is
.
Se
c
t
i
o
n
4
d
i
s
c
u
s
s
e
s
a
b
o
u
t
m
o
d
e
l'
s
p
e
r
f
o
r
m
a
n
c
e
e
v
a
l
u
a
t
i
o
n
r
e
s
u
lt
s
a
n
d
a
n
a
l
y
s
is
.
S
e
ct
i
o
n
5
co
n
clu
d
es th
e
r
esear
ch
p
a
p
er
.
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
C
r
o
p
p
r
ed
ictio
n
u
s
in
g
a
n
en
h
a
n
ce
d
s
ta
ck
ed
e
n
s
emb
le
ma
ch
in
e
lea
r
n
in
g
mo
d
el
(
D.
Ma
d
h
u
S
u
d
h
a
n
R
e
d
d
y
)
1843
2.
M
E
T
H
O
D
T
h
e
b
asic
ap
p
r
o
ac
h
o
f
ML
is
ca
teg
o
r
ized
in
to
th
r
ee
b
r
o
ad
t
y
p
es,
b
ased
o
n
th
e
n
atu
r
e
o
f
th
e
lear
n
in
g
p
ar
ad
ig
m
.
T
h
ese
ar
e
s
u
p
er
v
is
ed
lear
n
i
n
g
,
u
n
s
u
p
er
v
is
ed
lear
n
i
ng
,
an
d
r
ei
n
f
o
r
ce
m
en
t
lea
r
n
in
g
.
I
n
s
u
p
er
v
is
ed
lear
n
in
g
,
th
e
p
r
ed
ictiv
e
m
ac
h
in
e
lear
n
s
f
r
o
m
p
atter
n
s
an
d
r
elatio
n
s
h
ip
s
am
o
n
g
f
ea
tu
r
es
o
f
d
ata.
I
t
f
u
r
t
h
er
class
if
ies
in
to
two
ty
p
es
b
ased
o
n
th
e
tar
g
et/o
u
tp
u
t
v
alu
e
o
f
th
e
m
o
d
el,
i.e
.
class
if
icatio
n
an
d
r
e
g
r
es
s
io
n
.
E
x
am
p
les o
f
s
u
p
er
v
is
ed
lear
n
i
n
g
ar
e
DT
class
if
ier
,
RF
class
i
f
ier
,
KNN
c
lass
if
ier
,
an
d
MLP
.
ML
m
o
d
els
ar
e
lear
n
e
d
o
r
ex
p
er
ien
ce
d
b
y
tak
in
g
tr
ain
in
g
a
n
d
test
in
g
o
n
a
g
iv
e
n
d
ataset.
T
h
er
e
is
a
lim
itatio
n
in
in
d
iv
id
u
al
ML
m
o
d
els
th
at
te
n
d
to
p
er
f
o
r
m
p
o
o
r
ly
,
d
u
e
t
o
th
e
o
cc
u
r
r
en
c
e
o
f
h
ig
h
b
ias.
A
n
alter
n
ativ
e
s
o
lu
tio
n
is
to
co
m
b
in
e
in
d
iv
id
u
al
m
o
d
els
in
to
eith
er
p
ar
allel
o
r
s
eq
u
en
tial.
C
o
m
b
in
in
g
m
u
ltip
le
m
o
d
els
ca
n
h
ap
p
e
n
in
t
h
r
ee
way
s
,
n
am
ely
b
ag
g
in
g
,
b
o
o
s
ti
n
g
,
an
d
s
tack
ed
.
I
n
b
ag
g
in
g
,
t
h
e
s
am
e
ML
m
o
d
el
ca
n
b
e
co
n
s
id
er
e
d
p
ar
allelly
f
o
r
in
ter
m
ed
iate
p
r
ed
ictio
n
,
an
d
th
e
f
in
al
p
r
ed
ictio
n
is
ev
alu
ated
b
ased
o
n
th
e
m
ajo
r
v
o
tin
g
o
f
th
ese
in
ter
m
e
d
iate
p
r
ed
ictio
n
s
.
I
n
b
o
o
s
tin
g
,
th
e
s
am
e
ML
m
o
d
el
tak
es
in
s
eq
u
en
ce
s
o
th
at
in
co
r
r
ec
t
p
r
ed
ictio
n
o
f
t
h
e
f
ir
s
t
tr
ain
in
g
m
o
d
el
f
o
r
war
d
ed
t
o
th
e
n
e
x
t
tr
ain
in
g
m
o
d
el
to
m
ak
e
th
e
c
o
m
b
in
e
d
m
o
d
el
to
b
e
s
tr
o
n
g
in
p
r
e
d
ictio
n
.
Stack
in
g
m
ea
n
s
co
m
b
in
in
g
m
u
ltip
le
b
ase
m
o
d
els
an
d
m
ak
in
g
a
m
eta
-
m
o
d
el.
I
t
co
m
b
in
es
th
e
p
r
ed
ictio
n
o
f
m
u
ltip
le
ML
m
o
d
els
to
cr
ea
te
a
m
o
r
e
r
o
b
u
s
t
p
r
ed
ictiv
e
m
o
d
el.
I
t
lev
er
ag
es
t
h
e
d
iv
er
s
ity
am
o
n
g
in
d
iv
id
u
al
m
o
d
els
to
im
p
r
o
v
e
o
v
er
all
p
er
f
o
r
m
an
ce
,
ac
cu
r
ac
y
,
an
d
g
e
n
er
aliza
tio
n
.
E
n
s
em
b
le
m
eth
o
d
s
ar
e
wid
ely
u
s
ed
in
v
ar
io
u
s
ML
task
s
,
in
clu
d
in
g
r
eg
r
ess
io
n
,
c
lass
if
icatio
n
,
an
d
a
n
o
m
aly
d
e
tectio
n
.
T
h
e
b
asic
s
tr
u
ctu
r
e
o
f
s
tack
ed
en
s
em
b
le
lear
n
in
g
is
r
ep
r
esen
ted
in
Fig
u
r
e
1
.
I
t
is
a
f
o
u
r
-
s
tep
p
r
o
ce
s
s
,
in
itially
m
u
ltip
le
b
ase
lear
n
er
s
(
C
lass
if
ier
1
,
C
l
ass
if
ier
2
,
.
.
.
,
C
lass
if
ier
n
)
tr
ain
o
n
th
e
d
ataset.
Nex
t,
b
y
u
s
in
g
th
e
p
r
e
d
ictio
n
o
u
tp
u
ts
o
f
ea
ch
lear
n
er
to
f
o
r
m
a
n
ew
d
atas
et.
L
ater
a
m
eta
-
m
o
d
el
tr
ain
o
n
th
e
n
ewly
f
o
r
m
ed
d
ataset.
At
la
s
t,
th
e
m
eta
-
m
o
d
el
p
r
o
d
u
ce
s
th
e
f
in
al
p
r
ed
ictio
n
v
alu
e.
Fig
u
r
e
1
.
Simp
lifie
d
s
tr
u
ctu
r
e
o
f
s
tack
ed
en
s
em
b
le
lea
r
n
in
g
2
.
1
.
Da
t
a
s
et
d
escript
io
n
I
n
th
is
wo
r
k
,
th
e
d
ataset
th
at
h
as
b
ee
n
u
tili
ze
d
f
o
r
p
r
ed
ictio
n
is
tak
en
f
r
o
m
a
web
s
ite
[
2
9
]
.
T
h
is
d
ata
is
r
elate
d
to
f
iv
e
d
is
tr
icts
n
am
ed
An
an
tap
u
r
,
C
h
itto
o
r
,
K
ad
ap
a,
Ku
r
n
o
o
l,
an
d
SP
SR
Nello
r
e
o
f
An
d
h
r
a
Pra
d
esh
,
wh
ich
is
a
s
tate
in
s
o
u
th
I
n
d
ia,
wh
ich
c
o
n
tain
s
in
s
tan
ce
s
o
f
1
4
in
p
u
t
p
a
r
am
et
er
s
,
1
2
o
u
t
o
f
1
4
p
ar
am
eter
s
ar
e
s
o
il
p
ar
am
et
er
s
.
I
t
is
ca
teg
o
r
ized
in
to
m
ac
r
o
n
u
t
r
ien
ts
s
u
ch
as
Nitr
o
g
en
,
Ph
o
s
p
h
o
r
o
u
s
,
Po
tass
iu
m
,
Su
lp
h
u
r
,
C
alciu
m
,
an
d
Ma
g
n
esiu
m
an
d
m
icr
o
n
u
tr
ien
ts
s
u
ch
as
I
r
o
n
,
B
o
r
o
n
,
Z
i
n
c,
an
d
C
o
p
p
er
a
n
d
2
ar
e
clim
ate
p
ar
a
m
eter
s
s
u
c
h
as
r
ain
f
a
ll
an
d
tem
p
e
r
atu
r
e
.
T
h
ese
n
u
tr
ien
ts
ar
e
v
er
y
i
m
p
o
r
tan
t
to
g
r
o
w
a
h
ea
lth
y
cr
o
p
.
2
.
2
.
P
re
-
pro
ce
s
s
ing
T
h
e
d
ataset
co
n
tain
s
a
f
ew
o
u
t
lier
s
an
d
m
is
s
in
g
v
alu
es,
wh
ic
h
m
ay
b
e
s
en
s
itiv
e
to
a
f
ew
M
L
m
o
d
els
lik
e
SVM
,
LR
,
KNN
,
an
d
DT
.
As
it
is
r
elate
d
to
class
if
ica
ti
o
n
,
th
e
m
ea
n
im
p
u
tatio
n
m
eth
o
d
is
u
s
ed
to
m
ak
e
m
is
s
in
g
v
alu
es
in
to
s
u
itab
le
v
alu
es.
Featu
r
es
o
f
t
h
e
d
ataset
ar
e
i
n
d
if
f
er
en
t
r
a
n
g
es,
s
o
ea
ch
d
ata
p
o
in
t
o
f
f
ea
tu
r
es
n
ee
d
s
to
s
ca
le
in
th
e
s
am
e
r
an
g
e.
T
o
m
ak
e
s
ca
le
d
ata
p
o
in
ts
,
th
e
f
ir
s
t
m
ea
n
o
f
th
e
co
lu
m
n
v
ec
to
r
X
is
ca
lcu
lated
,
n
ex
t th
e
s
tan
d
ar
d
d
ev
iatio
n
o
f
X,
an
d
ca
lc
u
late
n
e
w
s
ca
le
v
alu
e
b
y
u
s
in
g
(
1
)
.
T
rai
n
i
ng
D
at
a
C
l
ass
i
f
i
er
1
C
l
ass
i
f
i
er
2
C
l
ass
i
f
i
er
n
N
ew
T
rai
n
i
ng
data
Me
t
a
Model
Fi
nal
Pr
edi
c
t
i
on
1
2
3
4
B
ase
L
e
an
e
r
s
.
.
.
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.
38
,
No
.
3
,
J
u
n
e
20
25
:
1
8
4
0
-
1
8
5
0
1844
=
−
(
1
)
2
.
3
.
Da
t
a
s
pli
t
t
ing
I
n
th
is
w
o
r
k
,
th
e
d
ataset
is
s
p
lit
in
to
“t
r
ain
-
test
s
”
with
d
if
f
er
en
t
s
izes
to
f
in
d
b
etter
tr
a
in
in
g
a
n
d
test
in
g
o
f
th
e
m
o
d
el
an
d
g
et
b
etter
ac
cu
r
ac
y
.
Her
e,
7
0
%
-
3
0
%
m
ea
n
s
7
0
%
o
f
t
o
tal
d
ata
is
u
s
ed
f
o
r
tr
ai
n
in
g
m
o
d
els
an
d
th
e
r
em
ain
i
n
g
3
0
%
is
k
ep
t
f
o
r
test
in
g
th
e
m
o
d
els.
T
h
e
to
tal
n
u
m
b
er
o
f
in
s
ta
n
ce
s
is
3
1
5
,
3
4
4
,
th
e
tr
ain
in
g
7
0
% si
ze
is
2
2
0
,
7
4
0
i
n
s
tan
ce
s
,
an
d
th
e
test
3
0
% si
ze
is
9
4
,
6
0
4
in
s
tan
ce
s
.
2
.
4
.
Cla
s
s
if
ica
t
io
n
a
lg
o
rit
hm
s
ML
m
o
d
els ar
e
u
s
ed
f
o
r
eith
er
class
if
icat
io
n
o
r
r
eg
r
ess
io
n
.
T
h
e
m
ajo
r
ity
o
f
ML
m
o
d
els p
er
f
o
r
m
b
o
t
h
p
r
ed
ictio
n
,
s
u
ch
as
class
if
icat
io
n
an
d
r
e
g
r
ess
io
n
.
T
h
is
p
ap
e
r
m
ain
ly
f
o
c
u
s
es
o
n
th
e
class
if
icatio
n
wo
r
k
an
d
m
o
d
els to
m
ak
e
a
clea
r
an
aly
s
is
o
f
th
e
s
tu
d
y
.
T
h
is
r
esear
ch
p
ap
er
is
co
n
ce
r
n
ed
with
class
if
icatio
n
m
o
d
els.
2
.
4
.
1
.
Dec
is
io
n
t
ree
DT
[
3
0
]
ar
e
a
p
o
p
u
lar
class
if
i
ca
tio
n
tech
n
iq
u
e
th
at
u
s
es
to
p
-
d
o
wn
p
r
o
ce
d
u
r
e
to
cr
ea
te
tr
ee
s
tr
u
ctu
r
e
class
if
ier
s
u
s
in
g
g
iv
en
d
ata.
T
h
e
I
D3
alg
o
r
ith
m
,
b
ased
o
n
en
tr
o
p
y
,
is
u
s
ed
to
ca
lcu
late
in
f
o
r
m
atio
n
g
ain
,
d
eter
m
in
in
g
wh
ich
attr
ib
u
te
to
b
e
a
s
r
o
o
t
an
d
in
ter
n
al
n
o
d
e
in
th
e
tr
ee
to
s
p
lit
f
u
r
th
er
.
An
ex
p
an
s
io
n
,
th
e
C
4
.
5
alg
o
r
ith
m
,
is
b
ased
o
n
I
D3
a
n
d
in
clu
d
es
f
ea
tu
r
es
lik
e
p
r
e
d
ictin
g
co
n
tin
u
o
u
s
v
alu
es
an
d
h
an
d
lin
g
m
is
s
in
g
v
alu
es.
T
h
e
DT
is
cr
ea
ted
b
y
s
elec
tin
g
th
e
h
ig
h
est
I
G
f
ea
tu
r
e
f
r
o
m
th
e
d
at
aset
an
d
s
p
litt
in
g
it
in
to
s
u
b
-
tr
ee
s
.
T
h
is
p
r
o
ce
s
s
is
r
ep
ea
ted
u
n
til
all
f
ea
tu
r
es a
r
e
co
v
e
r
ed
in
th
e
DT
.
2
.
4
.
2
.
M
ulti
-
la
y
er
perc
ept
ro
n
A
MLP
[
3
1
]
is
a
ty
p
e
o
f
AN
N
-
b
ased
n
o
n
-
lin
ea
r
m
o
d
el
th
a
t
f
alls
u
n
d
er
th
e
ca
te
g
o
r
y
o
f
f
ee
d
f
o
r
war
d
NN
s
.
I
t
co
n
s
is
ts
o
f
m
u
lt
ip
le
l
ay
er
s
o
f
n
o
d
es
(
n
eu
r
o
n
s
)
ar
r
a
n
g
ed
la
y
er
b
y
lay
er
s
tr
u
ctu
r
e,
in
clu
d
in
g
an
i
n
p
u
t
lay
er
,
o
n
e
o
r
m
o
r
e
h
id
d
e
n
lay
er
s
,
an
d
an
o
u
t
p
u
t
lay
er
.
ML
Ps
ar
e
wid
ely
u
s
ed
f
o
r
s
u
p
er
v
is
ed
lear
n
in
g
task
s
s
u
ch
as
r
eg
r
ess
io
n
an
d
class
if
icatio
n
.
T
h
e
p
r
ed
ictio
n
ca
p
ab
ili
ty
o
f
ML
P
co
m
es
f
r
o
m
b
y
m
ai
n
ten
an
ce
o
f
m
u
lti
-
lev
el
lay
er
s
o
f
n
e
u
r
o
n
n
etwo
r
k
s
.
T
h
e
b
asic
s
tr
u
ctu
r
e
o
f
ML
P
i
s
r
ep
r
esen
ted
in
Fig
u
r
e
2
,
X
1
,
X
2
,
X
3
,
an
d
X
4
ar
e
in
p
u
t
d
a
ta
v
alu
es,
wh
ich
ar
e
ass
ig
n
ed
to
n
e
u
r
o
n
s
o
f
th
e
in
p
u
t
lay
e
r
alo
n
g
th
e
b
ias.
I
n
ea
ch
lay
er
,
n
eu
r
o
n
s
p
er
f
o
r
m
o
p
er
atio
n
s
s
u
ch
as
th
e
s
u
m
m
atio
n
o
f
weig
h
ts
an
d
ac
tiv
atio
n
f
u
n
ctio
n
s
.
I
n
th
e
o
u
t
p
u
t
lay
er
,
th
e
So
f
tMa
x
f
u
n
ctio
n
g
en
er
ates
th
e
f
in
al
p
r
ed
icted
r
esu
lt
Y
b
ased
o
n
p
r
o
b
ab
ilit
y
.
ML
Ps
ca
n
m
ak
e
f
lex
ib
ilit
y
an
d
co
m
p
lex
m
o
d
e
l
r
elatio
n
s
h
ip
s
in
d
ata,
m
a
k
in
g
p
o
wer
f
u
l to
o
ls
f
o
r
v
ar
io
u
s
a
p
p
licatio
n
s
.
Fig
u
r
e
2
.
Mu
lti
-
lay
e
r
p
er
ce
p
tr
o
n
2
.
4
.
3
.
Su
pp
o
rt
v
ec
t
o
r
m
a
chi
nes
A
SVM
[
3
2
]
is
a
s
u
p
e
r
v
is
ed
ML
alg
o
r
ith
m
th
at
is
g
en
e
r
ally
ap
p
lied
in
h
i
g
h
-
d
im
en
s
io
n
al
s
p
ac
es
f
o
r
ap
p
licatio
n
s
o
f
class
if
icatio
n
a
n
d
r
eg
r
ess
io
n
.
I
t
is
a
b
in
a
r
y
cl
ass
if
ier
th
at
ca
teg
o
r
izes
d
ata
v
ar
iab
les
in
to
eit
h
er
I
n
p
u
t
L
ay
er
Hid
d
en
L
ay
er
Ou
tp
u
t
L
ay
er
B
ias
X
1
X
4
X
3
X
2
Y
B
ias
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
C
r
o
p
p
r
ed
ictio
n
u
s
in
g
a
n
en
h
a
n
ce
d
s
ta
ck
ed
e
n
s
emb
le
ma
ch
in
e
lea
r
n
in
g
mo
d
el
(
D.
Ma
d
h
u
S
u
d
h
a
n
R
e
d
d
y
)
1845
class
0
o
r
class
1
.
T
h
e
h
y
p
er
p
lan
e
o
f
t
h
e
SVM
is
s
elec
ted
to
o
p
tim
ize
lin
ea
r
s
ep
a
r
atio
n
b
etwe
en
two
-
class
d
ata
s
ets
o
f
two
-
d
im
en
s
io
n
al
s
p
ac
e
p
o
in
ts
.
T
h
e
o
b
jectiv
e
o
f
g
en
er
ali
z
atio
n
is
to
id
e
n
tify
an
n
-
d
im
e
n
s
io
n
al
h
y
p
er
p
lan
e
th
at
o
p
tim
i
z
es
th
e
s
ep
ar
atio
n
o
f
d
ata
p
o
in
ts
f
r
o
m
th
eir
p
o
ten
tial
class
es.
Data
p
o
in
ts
th
at
ar
e
clo
s
est
to
th
e
h
y
p
er
p
lan
e
an
d
h
av
e
th
e
m
in
im
u
m
d
is
tan
ce
a
r
e
r
ef
er
r
ed
to
as
s
u
p
p
o
r
t
v
ec
to
r
s
.
T
h
e
f
o
u
n
d
atio
n
f
o
r
d
ata
p
o
in
t
s
ep
ar
atio
n
ca
l
cu
latio
n
s
is
a
k
er
n
el
f
u
n
ctio
n
,
wh
ich
in
clu
d
es
lin
ea
r
,
p
o
l
y
n
o
m
ial,
g
au
s
s
ian
,
s
ig
m
o
id
,
an
d
r
ad
ial
b
asis
f
u
n
ctio
n
(
R
B
F)
f
u
n
ctio
n
s
.
T
h
e
ef
f
icien
cy
a
n
d
f
lu
id
ity
o
f
c
lass
s
ep
ar
atio
n
ar
e
r
eg
u
lated
b
y
th
ese
f
u
n
ctio
n
s
,
a
n
d
th
e
h
y
p
er
p
ar
am
eter
s
m
a
y
b
e
ad
ju
s
ted
to
in
d
u
ce
o
v
er
f
itti
n
g
o
r
u
n
d
er
f
itti
n
g
.
2
.
4
.
4
.
K
-
nea
re
s
t
neig
hb
o
urs
KNN
[
3
3
]
wo
r
k
s
b
ased
o
n
th
e
p
r
in
cip
le
o
f
‘
k
’
n
ea
r
est
lab
els
o
r
v
alu
es
to
d
atap
o
i
n
t.
I
t
is
u
s
ef
u
l
eith
er
f
o
r
class
if
icatio
n
o
r
r
e
g
r
ess
io
n
.
Fo
r
class
i
f
icatio
n
,
it
c
o
n
s
id
er
s
th
e
n
ea
r
est
‘
k
’
v
alu
es
tak
es
t
h
e
m
ajo
r
ity
v
o
tin
g
lab
el
an
d
g
iv
es
it
as
o
u
tp
u
t.
F
o
r
r
eg
r
ess
io
n
,
it
co
n
s
id
er
s
th
e
n
ea
r
est
‘
k
’
v
alu
es
av
e
r
ag
e
as
o
u
tp
u
t.
T
o
f
in
d
t
h
e
n
ea
r
est
‘
k
’
v
alu
es,
E
u
clid
ian
o
r
Ma
n
h
attan
d
is
tan
ce
m
ea
s
u
r
e
will
b
e
u
s
e
d
.
T
h
e
f
o
r
m
u
la
f
o
r
E
u
clid
ian
d
is
tan
ce
‘
’
o
f
two
p
o
in
ts
(
1
,
2
)
(
1
,
2
)
is
:
=
√
(
2
−
1
)
2
+
(
2
−
1
)
2
(
)
2
.
4
.
5
.
E
x
t
re
m
e
g
ra
dient
bo
o
s
t
cla
s
s
if
ier
T
h
e
XGBo
o
s
t
[
3
4
]
,
[
3
5
]
e
x
ten
s
io
n
f
o
r
g
r
a
d
ien
t
-
b
o
o
s
ted
DT
.
I
t
is
a
p
o
p
u
lar
an
d
s
k
illfu
lly
ex
ec
u
te
d
ap
p
r
o
ac
h
,
r
e
p
r
esen
ted
as
a
D
T
,
in
g
r
a
d
ien
t
-
b
o
o
s
ted
tr
ee
s
.
I
t
em
p
lo
y
s
a
tech
n
iq
u
e
th
at
b
u
i
ld
s
o
n
th
e
s
eq
u
e
n
ce
o
f
wea
k
lear
n
er
s
.
T
o
co
n
s
tr
u
ct
th
e
XGBo
o
s
t
tr
ee
,
s
tar
t w
i
th
f
in
d
in
g
th
e
r
esid
u
al
o
f
DT
-
1
,
a
wea
k
tr
ee
,
is
g
iv
en
to
DT
-
2
,
an
o
th
er
wea
k
tr
ee
,
t
o
r
ed
u
ce
th
e
o
v
e
r
all
r
esid
u
e.
T
h
is
p
r
o
ce
s
s
is
co
n
tin
u
ed
u
n
til
th
e
f
in
al
tr
ee
,
n
.
E
v
er
y
XGBo
o
s
t
tr
ee
m
o
d
el
l
o
wer
s
th
e
r
esid
u
al
f
r
o
m
th
e
t
r
ee
m
o
d
el
th
at
ca
m
e
b
ef
o
r
e
it
,
in
co
n
tr
ast
to
R
F
.
Similar
to
th
e
d
er
iv
atio
n
o
f
f
ir
s
t
-
o
r
d
er
f
o
r
er
r
o
r
in
f
o
r
m
ati
o
n
th
at
th
e
tr
ad
itio
n
al
g
r
ad
ien
t
b
o
o
s
ted
DT
(
GB
DT
)
em
p
lo
y
ed
.
C
o
s
t
f
u
n
ctio
n
s
ar
e
p
er
f
o
r
m
ed
b
y
XGBo
o
s
t
u
s
in
g
b
o
th
f
ir
s
t
-
an
d
s
ec
o
n
d
-
o
r
d
e
r
d
er
iv
ativ
es.
T
h
e
co
n
f
ig
u
r
ab
le
co
s
t
f
u
n
ctio
n
is
ad
d
itio
n
al
ly
en
a
b
led
b
y
th
e
XGBo
o
s
t
to
o
l.
T
o
ac
cu
r
atel
y
p
r
ed
ict
a
tar
g
et
v
ar
iab
le,
it c
o
m
b
in
es th
e
p
r
ed
i
ctio
n
s
o
f
s
im
p
ler
m
o
d
els an
d
m
u
ltip
le
wea
k
tr
ee
s
.
2
.
5
.
P
r
o
po
s
ed
enha
nced
s
t
a
c
k
ed
ens
em
ble le
a
rning
E
n
h
an
ce
d
s
tack
ed
e
n
s
em
b
le
lear
n
in
g
is
an
ad
v
an
ce
d
f
o
r
m
o
f
en
s
em
b
le
lear
n
in
g
b
ased
o
n
s
tack
g
en
er
aliza
tio
n
[
3
6
]
th
at
co
m
b
in
es
th
e
s
tr
en
g
th
s
o
f
in
d
iv
i
d
u
al
m
o
d
els
with
ad
d
itio
n
al
en
h
an
ce
m
e
n
ts
to
im
p
r
o
v
e
s
tack
e
d
m
o
d
el
p
er
f
o
r
m
an
ce
an
d
r
o
b
u
s
tn
ess
.
Alg
o
r
ith
m
1
.
Alg
o
r
ith
m
f
o
r
s
tack
ed
en
s
em
b
le
m
ac
h
in
e
lear
n
in
g
Input
:
Trai
ni
ng
da
ta
se
t
D
=
(
,
)
,
wh
er
e
X
∈
se
t
of
in
p
ut
fe
at
ur
es
an
d
Y
∈
ou
tp
ut
la
be
ls
.
A
set of base learners
‘B’
= {DT, XGBoost, KNN, SVM, MLP}, Meta learner: Random Forest model.
Output:
Predict a crop for a given input.
1:
The dataset
‘D’
is divided into
‘k’
(Example
‘k’=5
) fold partitions, denoted as
D={D
1
, D
2
, D
3
,…, D
k
}
. It helps in obtaining predictions for the training set without
overfitting.
2:
for
b=1 to B
do
3:
Train base classifier using
D
i
-
1
folds as the training set.
4:
end for
5:
Create a new training set for the meta
-
learner.
6:
for
b=1 to B
do
7:
Use fold
D
i
as the test set for prediction by the base classifier
.
8:
end for
9:
Aggregate the predictions from all
k
folds to form a new dataset
D′={(X′, Y)}
, where
X
i
′
is a vector of predictions from each base learner for the
j
th
vector sample.
Y
j
is the true output label for the
j
th
sample.
10:
Train the meta
-
learner RF on the new dataset
D′
with true output labels
Y
i
as the
target.
11:
To make a final prediction on a new, unseen test sample
X
test
:
12:
Obtain predictions from each base learner
B
i
on
X
test
.
13:
Combine these predictions to form a new feature vector
X
test
′
for the meta
-
learner.
14:
Predict the final output using the meta
-
learner model with
X
test
′
.
Stack
in
g
,
in
g
en
er
al,
in
v
o
lv
es
tr
ain
in
g
m
u
ltip
le
in
d
iv
i
d
u
al
b
ase
lear
n
er
s
p
ar
allelly
,
co
m
b
in
in
g
th
eir
p
r
ed
ictio
n
s
an
d
u
s
e
to
tr
ain
m
eta
lear
n
er
,
o
f
ten
r
ef
e
r
r
ed
to
as
a
m
eta
-
m
o
d
el
o
r
ag
g
r
eg
ato
r
.
I
n
Fig
u
r
e
3
,
DT
,
g
r
ad
ien
t
b
o
o
s
tin
g
,
KNN
,
s
u
p
p
o
r
t
v
ec
to
r
class
if
ier
,
an
d
MLP
ar
e
b
ase
lear
n
er
s
an
d
RF
is
a
m
eta
lear
n
er
.
T
h
e
f
in
al
p
r
ed
ictio
n
will
b
e
g
i
v
e
n
b
y
t
h
e
m
eta
-
lear
n
er
.
Fo
r
t
h
is
r
esear
ch
s
tu
d
y
,
en
h
a
n
ce
d
s
tack
ed
en
s
em
b
le
lea
r
n
in
g
h
as
d
if
f
er
e
n
t
ML
alg
o
r
ith
m
s
s
o
m
e
m
o
d
els
ar
e
b
ase
lear
n
er
s
at
th
e
in
itial
lev
el,
a
n
d
p
r
ed
ictio
n
r
esu
lts
o
f
th
ese
b
ase
lear
n
er
s
ar
e
co
n
s
id
er
ed
as
in
p
u
t
p
ar
am
eter
s
f
o
r
th
e
n
ex
t
lev
el
m
eta
lear
n
er
f
o
r
tr
ain
in
g
an
d
cr
o
s
s
-
v
alid
atio
n
.
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.
38
,
No
.
3
,
J
u
n
e
20
25
:
1
8
4
0
-
1
8
5
0
1846
Fig
u
r
e
3
.
Stru
ctu
r
e
o
f
th
e
p
r
o
p
o
s
ed
en
h
an
ce
d
s
tack
ed
e
n
s
em
b
le
m
o
d
el
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
is
r
esear
ch
s
tu
d
y
in
v
esti
g
a
ted
ex
is
tin
g
an
d
tr
ad
itio
n
al
ML
m
o
d
els
s
u
ch
as
DT
,
X
G
B
o
o
s
tin
g
,
KNN,
an
d
SVM,
wh
ich
h
av
e
n
o
t
co
m
p
r
eh
e
n
s
iv
ely
in
c
o
r
p
o
r
ate
d
with
s
o
il
an
d
we
ath
er
d
ata
f
o
r
c
r
o
p
r
ec
o
m
m
en
d
atio
n
.
A
d
d
itio
n
all
y
,
en
s
em
b
le
m
o
d
els
h
av
e
s
co
p
e
to
im
p
r
o
v
e
p
er
f
o
r
m
an
ce
in
ter
m
s
o
f
ac
c
u
r
ac
y
an
d
F1
-
s
co
r
e.
Acc
u
r
ac
y
m
ea
s
u
r
es
o
v
er
all
co
r
r
ec
tn
ess
,
p
r
ec
i
s
io
n
ev
alu
ates
th
e
q
u
ality
o
f
p
o
s
itiv
e
p
r
ed
ictio
n
s
,
r
ec
all
ev
alu
ates
s
en
s
itiv
ity
to
p
o
s
itiv
e
in
s
tan
ce
s
,
an
d
F1
-
s
co
r
e
b
alan
ce
s
p
r
ec
is
io
n
an
d
r
ec
a
ll.
I
t
is
ess
en
tial
to
co
m
p
r
eh
e
n
d
th
ese
m
etr
ics to
c
o
n
d
u
ct
a
th
o
r
o
u
g
h
ev
alu
atio
n
an
d
o
p
tim
izatio
n
o
f
m
o
d
els in
ML
ap
p
licatio
n
s
.
A
m
o
d
el
g
en
e
r
ates
th
e
a
p
p
r
o
p
r
iate
n
u
m
b
er
o
f
p
r
e
d
ictio
n
s
b
y
an
aly
zin
g
th
e
o
b
s
er
v
ed
v
alu
e
s
,
wh
ich
is
th
e
ess
en
ce
o
f
ac
cu
r
ac
y
.
T
h
e
d
ef
in
e
d
v
alu
es
a
r
e
e
v
alu
at
ed
to
d
eter
m
in
e
wh
eth
er
th
e
y
ar
e
t
r
u
e
o
r
f
alse.
A
m
ea
s
u
r
em
en
t
o
f
ac
cu
r
ac
y
is
illu
s
tr
at
ed
in
(
3
)
.
I
t
is
ass
ess
e
d
b
ased
o
n
tr
u
e
p
o
s
itiv
e
(
T
P),
tr
u
e
n
eg
ativ
e
(
T
N)
,
f
alse
p
o
s
itiv
e
(
FP
)
,
an
d
f
alse
n
eg
ativ
e
(
FN)
v
alu
es.
Her
e
T
P
m
ea
n
s
a
co
r
r
ec
t
p
r
e
d
ictio
n
th
at
an
o
u
tco
m
e
is
p
o
s
itiv
e,
T
N
m
ea
n
s
a
co
r
r
ec
t
p
r
ed
ictio
n
th
at
an
o
u
tco
m
e
is
n
eg
ativ
e,
FP
m
e
an
s
an
in
co
r
r
e
ct
p
r
ed
ictio
n
th
at
a
n
o
u
tco
m
e
is
p
o
s
itiv
e,
a
n
d
FN m
ea
n
s
an
in
co
r
r
ec
t p
r
e
d
ictio
n
th
at
an
o
u
tco
m
e
is
n
eg
ativ
e.
=
(
+
)
(
+
+
+
)
(
3
)
Pre
cisi
o
n
is
a
ter
m
th
at
is
u
s
ed
to
ass
es
s
th
e
s
en
s
itiv
ity
an
d
e
f
f
icac
y
o
f
a
class
if
icatio
n
m
o
d
el.
TP
an
d
FP
s
ta
tem
en
ts
ar
e
em
p
lo
y
ed
to
q
u
an
tify
it.
T
h
is
class
if
ier
g
en
er
ates
a
p
o
s
itiv
e
p
r
o
b
a
b
ilit
y
r
esu
lt,
wh
ich
is
co
m
p
u
ted
b
y
t
h
e
v
alu
es sp
ec
if
ied
in
(
4
)
.
=
(
+
)
(
4
)
R
ec
all
r
ef
er
s
to
th
e
s
ce
n
ar
io
in
wh
ich
class
if
icatio
n
o
u
tc
o
m
es
ar
e
d
ee
m
ed
b
ad
b
ased
o
n
th
e
class
if
ier
's
p
r
o
b
a
b
ilit
y
ass
es
s
m
en
t.
I
t is
as
s
ess
ed
b
y
g
en
u
in
e
p
o
s
itiv
e
an
d
f
alse n
eg
ativ
e
s
tatem
en
ts
.
T
h
e
(
5
)
illu
s
tr
ates t
h
e
co
m
p
u
tatio
n
o
f
r
ec
all.
=
(
+
)
(
5
)
T
h
e
F1
-
s
co
r
e
is
a
v
alu
e
th
at
i
s
u
tili
z
ed
in
th
e
p
r
o
ce
s
s
o
f
ca
l
cu
latin
g
p
r
ed
ictio
n
p
er
f
o
r
m
an
ce
.
R
ec
all
an
d
ac
cu
r
ac
y
ar
e
b
o
th
weig
h
te
d
an
d
av
e
r
ag
ed
t
o
g
eth
er
t
o
d
et
er
m
in
e
th
e
F1
-
s
co
r
e.
T
h
e
ac
cu
r
ac
y
an
d
r
ec
all
ar
e
th
e
m
etr
ics th
at
ar
e
u
s
ed
to
ev
alu
ate
it.
T
h
e
co
m
p
u
tatio
n
o
f
t
h
e
F1
-
s
co
r
e
is
d
is
p
lay
ed
i
n
th
e
(
6
)
.
1
=
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
C
r
o
p
p
r
ed
ictio
n
u
s
in
g
a
n
en
h
a
n
ce
d
s
ta
ck
ed
e
n
s
emb
le
ma
ch
in
e
lea
r
n
in
g
mo
d
el
(
D.
Ma
d
h
u
S
u
d
h
a
n
R
e
d
d
y
)
1847
I
n
Fig
u
r
e
4
,
d
if
f
e
r
en
t
ML
m
o
d
els
an
d
th
e
p
r
o
p
o
s
ed
e
n
h
an
ce
d
s
tack
e
d
en
s
em
b
le
m
o
d
els
ar
e
co
m
p
ar
ed
with
r
esp
ec
tiv
e
p
er
f
o
r
m
a
n
ce
u
s
in
g
ac
cu
r
ac
y
.
T
h
is
clea
r
ly
s
tates
th
at
th
e
en
h
an
ce
d
s
tack
ed
en
s
em
b
le
m
o
d
el
o
u
t
p
er
f
o
r
m
s
r
em
ain
in
g
ML
m
o
d
els
s
u
ch
as
DT
,
XGBo
o
s
t,
KNN,
ML
P
,
an
d
SVC
.
W
h
en
th
er
e
ar
e
im
b
alan
ce
s
in
th
e
c
lass
es
o
f
a
d
ataset,
th
e
F1
-
s
co
r
e
is
a
m
o
r
e
u
s
ef
u
l
m
etr
ic
th
an
ac
cu
r
ac
y
.
An
im
p
r
o
v
e
d
m
etr
ic
t
o
ass
ess
d
if
f
er
e
n
t
ML
m
o
d
els
alo
n
g
with
s
tack
e
d
en
s
em
b
l
e
lear
n
in
g
is
th
e
F1
-
s
co
r
e.
I
n
Fig
u
r
e
5
,
F1
-
s
c
o
r
e
co
m
p
ar
is
o
n
o
f
d
if
f
er
en
t
ML
m
o
d
els.
I
t
is
co
v
ey
th
at
s
tack
ed
e
n
s
em
b
le
lear
n
in
g
o
u
tp
er
f
o
r
m
s
th
an
r
e
m
ain
in
g
ML
m
o
d
els
SVC
,
ML
P,
KNN,
XB
,
an
d
DT
.
Fu
tu
r
e
r
esear
ch
ca
n
in
teg
r
ate
th
e
m
eth
o
d
with
web
o
r
m
o
b
ile
ap
p
licatio
n
s
ef
f
ec
tiv
ely
u
s
ed
b
y
f
ar
m
er
s
.
Fig
u
r
e
4
.
Acc
u
r
ac
y
c
o
m
p
ar
is
o
n
o
f
d
if
f
er
e
n
t M
L
m
o
d
els
Fig
u
r
e
5
.
F1
s
co
r
e
co
m
p
ar
is
o
n
o
f
d
if
f
er
en
t M
L
m
o
d
els
I
n
th
is
r
esear
ch
wo
r
k
,
ML
m
o
d
els
s
u
ch
as
DT
,
XB
,
ML
P,
a
n
d
SVC
p
er
f
o
r
m
b
etter
th
an
KNN
with
ab
o
v
e
8
4
%
ac
cu
r
ac
y
.
As
th
e
d
ataset
h
as
m
o
r
e
d
im
en
s
io
n
s
,
KNN
i
s
u
n
ab
le
to
h
an
d
le
it
p
r
o
p
er
ly
.
SVC
p
r
o
d
u
ce
s
8
4
.
3
0
%
ac
c
u
r
ac
y
b
y
f
o
r
m
in
g
m
u
ltip
le
p
lan
es
in
s
u
ch
a
wa
y
as
to
class
if
y
d
ata
ef
f
icien
tly
.
ML
P
h
as
g
iv
en
th
e
ac
cu
r
ac
y
at
8
6
.
3
%
b
y
tr
ain
in
g
d
ata
n
o
n
-
lin
ea
r
ly
with
ad
ju
s
ted
weig
h
ts
an
d
b
ia
s
.
T
h
e
ac
cu
r
ac
y
o
f
DT
is
8
6
.
3
%
g
iv
en
b
y
tak
in
g
th
e
h
ig
h
I
G
f
ea
tu
r
e
as
r
o
o
t
to
s
p
lit
tr
ee
.
A
b
o
o
s
tin
g
te
ch
n
iq
u
e,
XGBo
o
s
t
m
ain
tain
s
an
ac
cu
r
ac
y
o
f
8
8
.
5
%
wh
ich
is
b
etter
th
an
t
h
e
ac
c
u
r
ac
y
o
f
DT
.
T
h
e
XGBo
o
s
t
tr
ee
was
co
n
s
tr
u
cted
in
s
u
ch
a
wa
y
th
at
r
esid
u
als
o
f
a
DT
wer
e
r
ed
u
ce
d
lev
el
b
y
lev
el.
T
h
e
p
r
o
p
o
s
ed
en
h
a
n
ce
d
s
tack
ed
en
s
em
b
le
m
o
d
el
g
iv
es b
etter
p
er
f
o
r
m
an
c
e
at
9
0
-
1
0
tr
ain
test
s
ize
with
9
3
.
1
0
% a
cc
u
r
ac
y
.
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.
38
,
No
.
3
,
J
u
n
e
20
25
:
1
8
4
0
-
1
8
5
0
1848
T
h
e
in
d
iv
id
u
al
ML
m
o
d
el
j
u
s
t
cr
ea
tes
a
r
elatio
n
s
h
ip
f
u
n
cti
o
n
in
an
attem
p
t
to
m
ap
in
p
u
t
to
war
d
s
o
u
tp
u
t.
B
u
t
s
tack
ed
en
s
em
b
led
lear
n
in
g
tak
es
th
in
g
s
a
s
tep
f
u
r
th
er
b
y
d
eter
m
in
in
g
th
e
c
o
n
n
ec
tio
n
b
etwe
en
ea
ch
en
s
em
b
led
m
o
d
el'
s
p
r
ed
ictio
n
r
esu
lt
o
n
o
u
t
-
of
-
s
am
p
le
d
ata
an
d
th
e
ac
tu
al
v
alu
e.
As
s
tack
ed
en
s
em
b
led
ar
e
d
esig
n
ed
to
b
e
m
o
r
e
r
o
b
u
s
t
th
an
av
er
ag
e
b
o
o
s
tin
g
m
o
d
el
s
o
r
in
d
iv
id
u
al
m
o
d
els,
th
ey
ty
p
ically
p
r
o
d
u
ce
b
etter
p
r
e
d
ictiv
e
p
er
f
o
r
m
a
n
ce
.
T
h
er
e
ar
e
in
s
tan
ce
s
wh
er
e
ev
en
s
m
all
g
ain
s
in
p
r
e
d
ictio
n
ac
cu
r
ac
y
h
a
v
e
a
s
ig
n
if
ican
t im
p
ac
t o
n
th
e
b
u
s
in
ess
s
itu
atio
n
.
T
h
is
s
tu
d
y
co
n
s
id
er
e
d
an
d
ap
p
lied
to
f
iv
e
d
is
tr
icts
o
f
An
d
h
r
a
Pra
d
esh
,
f
u
tu
r
e
r
esear
c
h
ca
n
ap
p
ly
to
th
e
en
tire
s
tate
as
d
iv
er
s
e
s
o
il
an
d
wea
th
er
c
o
n
d
itio
n
s
.
Fu
tu
r
e
s
tu
d
ies
m
ay
ex
p
l
o
r
e
i
n
teg
r
atin
g
m
o
d
els
with
I
o
T
d
e
v
ices
to
g
et
r
ea
l
-
tim
e
s
o
il
an
d
wea
th
er
d
ata
f
o
r
cr
o
p
r
e
co
m
m
en
d
atio
n
.
B
y
u
tili
zin
g
a
r
ich
s
et
o
f
s
o
il
an
d
clim
ate
d
ata,
th
e
m
o
d
el
s
ig
n
i
f
ican
tly
im
p
r
o
v
es
th
e
p
r
ec
is
io
n
o
f
cr
o
p
r
ec
o
m
m
en
d
atio
n
s
,
th
er
eb
y
p
r
o
m
is
in
g
s
u
b
s
tan
tial b
en
ef
its
f
o
r
th
e
a
g
r
icu
ltu
r
al
s
ec
to
r
in
ter
m
s
o
f
y
ie
ld
o
p
tim
izatio
n
a
n
d
ec
o
n
o
m
ic
s
tab
ilit
y
.
4.
CO
NCLU
SI
O
N
E
n
h
an
cin
g
p
r
o
d
u
ctiv
ity
in
PA
h
as
r
e
q
u
ir
ed
th
e
r
ig
h
t
cr
o
p
-
s
elec
tio
n
m
eth
o
d
b
y
co
n
s
id
er
i
n
g
s
o
il
a
n
d
wea
th
er
d
ata.
T
r
ad
itio
n
al
ML
m
eth
o
d
s
lik
e
DT
an
d
XGBo
o
s
t
h
av
e
m
ad
e
s
tr
id
es
in
p
r
ed
ic
tin
g
cr
o
p
v
iab
ilit
y
,
b
u
t
s
till
,
th
ese
f
all
s
h
o
r
t
u
n
d
er
co
m
p
le
x
,
r
ea
l
-
wo
r
ld
co
n
d
itio
n
s
.
T
h
e
en
h
a
n
ce
d
s
tac
k
ed
en
s
em
b
le
ML
m
o
d
el
is
an
in
n
o
v
ativ
e
ap
p
r
o
ac
h
th
at
u
n
ites
th
e
s
tr
en
g
th
s
o
f
v
ar
io
u
s
m
o
d
els
in
to
a
s
in
g
u
lar
,
p
o
ten
ti
al
m
eta
-
m
o
d
el.
T
h
is
m
o
d
el'
s
lay
er
ed
ar
ch
itectu
r
e
s
ig
n
if
ican
tly
elev
ates
p
r
ed
ictio
n
ac
cu
r
ac
y
to
a
n
im
p
r
ess
iv
e
9
3
.
1
%,
b
y
an
al
y
z
in
g
1
2
s
o
il
p
a
r
am
eter
s
,
in
clu
d
in
g
ess
en
tial
n
u
tr
ien
ts
N,
P,
an
d
K,
alo
n
g
s
id
e
clim
atic
elem
e
n
ts
s
u
ch
as
tem
p
er
atu
r
e
an
d
r
ain
f
all.
T
h
e
o
u
ts
tan
d
in
g
p
er
f
o
r
m
an
ce
o
f
th
e
p
r
o
p
o
s
ed
en
s
em
b
le
m
o
d
el
m
ar
k
s
a
tr
an
s
f
o
r
m
ativ
e
leap
in
PA
,
p
r
o
m
is
in
g
to
r
e
v
o
lu
tio
n
iz
e
cr
o
p
s
elec
tio
n
p
r
o
ce
s
s
es.
B
y
d
o
in
g
s
o
,
it
em
p
o
wer
s
I
n
d
ia'
s
ag
r
icu
ltu
r
al
s
ec
to
r
with
th
e
to
o
ls
f
o
r
n
o
t
ju
s
t
s
u
r
v
iv
al,
b
u
t
also
f
o
r
th
r
iv
in
g
ec
o
n
o
m
ically
,
g
u
id
in
g
i
n
a
n
ew
er
a
o
f
ag
r
o
n
o
m
i
c
in
tellig
en
ce
an
d
ec
o
n
o
m
ic
s
u
s
tain
ab
ilit
y
.
ACK
NO
WL
E
DG
M
E
N
T
S
T
h
is
wo
r
k
was
s
u
p
p
o
r
ted
b
y
th
e
Dep
a
r
tm
en
t
o
f
C
o
m
p
u
ter
Scien
ce
an
d
E
n
g
in
ee
r
i
n
g
at
Sri
Ven
k
ateswar
a
Un
iv
er
s
ity
.
W
e
s
in
ce
r
ely
ap
p
r
ec
iate
th
eir
in
v
a
lu
ab
le
ass
is
tan
ce
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
T
h
e
au
th
o
r
s
d
ec
lar
e
th
at
th
e
y
h
av
e
n
o
k
n
o
wn
co
m
p
etin
g
f
in
an
cial
in
ter
ests
.
AUTHO
R
C
O
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
D
.
M
a
d
h
u
S
u
d
h
a
n
R
e
d
d
y
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
N.
Ush
a
R
an
i
✓
✓
✓
✓
✓
✓
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
Au
th
o
r
s
s
tate
n
o
co
n
f
lict o
f
in
t
er
est.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
d
ata
s
u
p
p
o
r
tin
g
th
is
s
tu
d
y
'
s
f
in
d
in
g
s
ar
e
av
ailab
le
f
r
o
m
a
g
o
v
er
n
m
en
t
web
s
ite.
R
estrictio
n
s
ap
p
ly
to
th
e
a
v
ailab
ilit
y
o
f
th
ese
d
ata,
wh
ich
wer
e
u
s
ed
u
n
d
er
licen
s
e
f
o
r
th
is
s
tu
d
y
.
Data
ar
e
av
ailab
le
at
h
ttp
s
://s
o
ilh
ea
lth
.
d
ac
.
g
o
v
.
in
w
ith
th
e
p
er
m
is
s
io
n
o
f
th
e
co
n
c
er
n
team
.
Data
s
et
was
o
b
tain
e
d
u
p
o
n
r
eq
u
est
f
o
r
r
esear
ch
ex
p
e
r
im
en
ts
f
r
o
m
th
e
g
o
v
er
n
m
en
t w
eb
s
ite
h
ttp
s
://s
o
ilh
ea
lth
.
d
ac
.
g
o
v
.
in
.
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
C
r
o
p
p
r
ed
ictio
n
u
s
in
g
a
n
en
h
a
n
ce
d
s
ta
ck
ed
e
n
s
emb
le
ma
ch
in
e
lea
r
n
in
g
mo
d
el
(
D.
Ma
d
h
u
S
u
d
h
a
n
R
e
d
d
y
)
1849
RE
F
E
R
E
NC
E
S
[
1
]
A
.
G
u
l
a
t
i
a
n
d
R
.
Ju
n
e
j
a
,
“
Tr
a
n
sf
o
r
mi
n
g
I
n
d
i
a
n
a
g
r
i
c
u
l
t
u
r
e
,
”
i
n
I
n
d
i
a
n
a
g
r
i
c
u
l
t
u
re
t
o
w
a
rd
s
,
2
0
2
2
,
p
p
.
9
–
3
7
.
[
2
]
D
.
C
r
e
v
i
e
r
,
AI
:
t
h
e
t
u
m
u
l
t
u
o
u
s
h
i
st
o
ry
o
f
t
h
e
se
a
rc
h
f
o
r
a
rt
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
.
U
n
i
t
e
d
S
t
a
t
e
s:
B
a
s
i
c
B
o
o
k
s
,
I
n
c
.
,
1
9
9
3
.
[
3
]
S
.
S
h
a
l
e
v
-
S
h
w
a
r
t
z
a
n
d
S
.
B
e
n
-
D
a
v
i
d
,
U
n
d
e
rs
t
a
n
d
i
n
g
Ma
c
h
i
n
e
L
e
a
r
n
i
n
g
:
Fro
m
T
h
e
o
r
y
t
o
A
l
g
o
r
i
t
h
m
s.
C
a
m
b
r
i
d
g
e
U
n
i
v
e
r
s
i
t
y
P
r
e
ss,
2
0
1
4
.
[
4
]
A
.
V
i
j
,
S
.
V
i
j
e
n
d
r
a
,
A
.
J
a
i
n
,
S
.
B
a
j
a
j
,
A
.
B
a
ssi
,
a
n
d
A
.
S
h
a
r
ma,
“
I
o
T
a
n
d
mac
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
e
s
f
o
r
a
u
t
o
ma
t
i
o
n
o
f
f
a
r
m
i
r
r
i
g
a
t
i
o
n
s
y
st
e
m,
”
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
6
7
,
p
p
.
1
2
5
0
–
1
2
5
7
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
r
o
c
s
.
2
0
2
0
.
0
3
.
4
4
0
.
[
5
]
L.
El
H
o
u
mm
a
i
d
i
,
A
.
L
a
r
a
b
i
,
a
n
d
K
.
A
l
a
m,
“
U
si
n
g
u
n
ma
n
n
e
d
a
e
r
i
a
l
s
y
s
t
e
ms a
n
d
d
e
e
p
l
e
a
r
n
i
n
g
f
o
r
a
g
r
i
c
u
l
t
u
r
e
map
p
i
n
g
i
n
D
u
b
a
i
,
”
H
e
l
i
y
o
n
,
v
o
l
.
7
,
n
o
.
1
0
,
p
.
e
0
8
1
5
4
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
h
e
l
i
y
o
n
.
2
0
2
1
.
e
0
8
1
5
4
.
[
6
]
L.
B
e
n
o
s
,
A
.
C
.
T
a
g
a
r
a
k
i
s,
G
.
D
o
l
i
a
s
,
R
.
B
e
r
r
u
t
o
,
D
.
K
a
t
e
r
i
s,
a
n
d
D
.
B
o
c
h
t
i
s,
“
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
i
n
a
g
r
i
c
u
l
t
u
r
e
:
a
c
o
m
p
r
e
h
e
n
si
v
e
u
p
d
a
t
e
d
r
e
v
i
e
w
,
”
S
e
n
so
r
s
,
v
o
l
.
2
1
,
n
o
.
1
1
,
p
.
3
7
5
8
,
M
a
y
2
0
2
1
,
d
o
i
:
1
0
.
3
3
9
0
/
s2
1
1
1
3
7
5
8
.
[
7
]
V
.
H
a
k
k
i
m,
E.
Jo
s
e
p
h
,
A
.
G
o
k
u
l
,
a
n
d
K
.
M
u
f
e
e
d
h
a
,
“
P
r
e
c
i
si
o
n
f
a
r
mi
n
g
:
t
h
e
f
u
t
u
r
e
o
f
I
n
d
i
a
n
a
g
r
i
c
u
l
t
u
r
e
,
”
J
o
u
rn
a
l
o
f
Ap
p
l
i
e
d
Bi
o
l
o
g
y
a
n
d
Bi
o
t
e
c
h
n
o
l
o
g
y
,
v
o
l
.
4
,
n
o
.
6
,
p
p
.
0
6
8
–
0
7
2
,
2
0
1
6
,
d
o
i
:
1
0
.
7
3
2
4
/
j
a
b
b
.
2
0
1
6
.
4
0
6
0
9
.
[
8
]
N
.
H
.
K
u
l
k
a
r
n
i
,
G
.
N
.
S
r
i
n
i
v
a
sa
n
,
B
.
M
.
S
a
g
a
r
,
a
n
d
N
.
K
.
C
a
u
v
e
r
y
,
“
I
mp
r
o
v
i
n
g
c
r
o
p
p
r
o
d
u
c
t
i
v
i
t
y
t
h
r
o
u
g
h
a
c
r
o
p
r
e
c
o
m
me
n
d
a
t
i
o
n
sy
st
e
m
u
si
n
g
e
n
s
e
m
b
l
i
n
g
t
e
c
h
n
i
q
u
e
,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
2
0
1
8
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
C
o
m
p
u
t
a
t
i
o
n
a
l
S
y
st
e
m
s
a
n
d
I
n
f
o
rm
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
f
o
r
S
u
st
a
i
n
a
b
l
e
S
o
l
u
t
i
o
n
s,
C
S
I
T
S
S
2
0
1
8
,
2
0
1
8
,
p
p
.
1
1
4
–
1
1
9
,
d
o
i
:
1
0
.
1
1
0
9
/
C
S
I
TSS
.
2
0
1
8
.
8
7
6
8
7
9
0
.
[
9
]
A
.
A
.
M
a
h
u
l
e
,
“
H
y
b
r
i
d
m
e
t
h
o
d
f
o
r
i
mp
r
o
v
i
n
g
a
c
c
u
r
a
c
y
o
f
c
r
o
p
-
t
y
p
e
d
e
t
e
c
t
i
o
n
u
s
i
n
g
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
Ad
v
a
n
c
e
d
T
re
n
d
s
i
n
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
En
g
i
n
e
e
r
i
n
g
,
v
o
l
.
9
,
n
o
.
2
,
p
p
.
2
2
8
4
–
2
2
8
8
,
2
0
2
0
,
d
o
i
:
1
0
.
3
0
5
3
4
/
i
j
a
t
c
se
/
2
0
2
0
/
2
0
9
9
2
2
0
2
0
.
[
1
0
]
K
.
P
a
l
a
n
i
v
e
l
a
n
d
C
.
S
u
r
i
a
n
a
r
a
y
a
n
a
n
,
“
A
n
A
p
p
r
o
a
c
h
f
o
r
p
r
e
d
i
c
t
i
o
n
o
f
c
r
o
p
y
i
e
l
d
u
s
i
n
g
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
n
d
b
i
g
d
a
t
a
t
e
c
h
n
i
q
u
e
s,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
C
o
m
p
u
t
e
r
En
g
i
n
e
e
ri
n
g
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
1
0
,
n
o
.
3
,
2
0
1
9
,
d
o
i
:
1
0
.
3
4
2
1
8
/
i
j
c
e
t
.
1
0
.
3
.
2
0
1
9
.
0
1
3
.
[
1
1
]
S
.
N
o
sr
a
t
a
b
a
d
i
,
K
.
S
z
e
l
l
,
B
.
B
e
sz
e
d
e
s
,
F
.
I
mr
e
,
S
.
A
r
d
a
b
i
l
i
,
a
n
d
A
.
M
o
sa
v
i
,
“
C
o
m
p
a
r
a
t
i
v
e
a
n
a
l
y
s
i
s
o
f
A
N
N
-
I
C
A
a
n
d
A
N
N
-
G
W
O
f
o
r
c
r
o
p
y
i
e
l
d
p
r
e
d
i
c
t
i
o
n
,
”
2
0
2
0
,
d
o
i
:
1
0
.
1
1
0
9
/
R
I
V
F
4
8
6
8
5
.
2
0
2
0
.
9
1
4
0
7
8
6
.
[
1
2
]
P
.
P
a
t
i
l
,
V
.
P
a
n
p
a
t
i
l
,
a
n
d
S
.
K
o
k
a
t
e
,
“
C
r
o
p
p
r
e
d
i
c
t
i
o
n
sy
s
t
e
m u
s
i
n
g
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
ms,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
Re
se
a
rc
h
J
o
u
rn
a
l
o
f
En
g
i
n
e
e
r
i
n
g
a
n
d
T
e
c
h
n
o
l
o
g
y
(
I
RJ
E
T
)
,
v
o
l
.
7
,
n
o
.
2
,
p
p
.
7
4
8
–
7
5
3
,
2
0
2
0
.
[
1
3
]
N
.
U
sh
a
R
a
n
i
a
n
d
G
.
G
o
w
t
h
a
m
i
,
“
S
m
a
r
t
c
r
o
p
s
u
g
g
e
s
t
e
r
,
”
i
n
A
d
v
a
n
c
e
s
i
n
C
o
m
p
u
t
a
t
i
o
n
a
l
a
n
d
B
i
o
-
En
g
i
n
e
e
ri
n
g
:
P
ro
c
e
e
d
i
n
g
o
f
t
h
e
In
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
a
t
i
o
n
a
l
a
n
d
B
i
o
E
n
g
i
n
e
e
r
i
n
g
,
2
0
2
0
,
p
p
.
4
0
1
–
4
1
3
.
[
1
4
]
S
.
A
.
Z.
R
a
h
ma
n
,
K
.
C
.
M
i
t
r
a
,
a
n
d
S
.
M
.
M
.
I
sl
a
m,
“
S
o
i
l
c
l
a
ssi
f
i
c
a
t
i
o
n
u
s
i
n
g
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
met
h
o
d
s
a
n
d
c
r
o
p
su
g
g
e
st
i
o
n
b
a
se
d
o
n
s
o
i
l
seri
e
s,”
i
n
2
0
1
8
2
1
st
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
f
C
o
m
p
u
t
e
r
a
n
d
I
n
f
o
rm
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
(
I
C
C
I
T
)
,
D
e
c
.
2
0
1
8
,
p
p
.
1
–
4
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
C
I
TEC
H
N
.
2
0
1
8
.
8
6
3
1
9
4
3
.
[
1
5
]
D
.
M
.
S
.
R
e
d
d
y
a
n
d
U
.
R
.
N
e
e
r
u
g
a
t
t
i
,
“
A
c
o
mp
a
r
a
t
i
v
e
a
n
a
l
y
s
i
s
o
f
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
mo
d
e
l
s
f
o
r
c
r
o
p
r
e
c
o
mm
e
n
d
a
t
i
o
n
i
n
I
n
d
i
a
,
”
Re
v
u
e
d
’
I
n
t
e
l
l
i
g
e
n
c
e
Art
i
f
i
c
i
e
l
l
e
,
v
o
l
.
3
7
,
n
o
.
4
,
p
p
.
1
1
8
1
–
1
1
9
0
,
A
u
g
.
2
0
2
3
,
d
o
i
:
1
0
.
1
8
2
8
0
/
r
i
a
.
3
7
0
4
3
0
.
[
1
6
]
A
.
P
r
i
y
a
d
h
a
r
sh
i
n
i
,
S
.
C
h
a
k
r
a
b
o
r
t
y
,
A
.
K
u
m
a
r
,
a
n
d
O
.
R
.
P
o
o
n
i
w
a
l
a
,
“
I
n
t
e
l
l
i
g
e
n
t
c
r
o
p
r
e
c
o
mm
e
n
d
a
t
i
o
n
s
y
st
e
m
u
si
n
g
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
-
5
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
C
o
m
p
u
t
i
n
g
M
e
t
h
o
d
o
l
o
g
i
e
s
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
,
I
C
C
MC
2
0
2
1
,
2
0
2
1
,
p
p
.
8
4
3
–
8
4
8
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
C
M
C
5
1
0
1
9
.
2
0
2
1
.
9
4
1
8
3
7
5
.
[
1
7
]
N
.
D
e
e
p
a
a
n
d
K
.
G
a
n
e
s
a
n
,
“
M
u
l
t
i
-
c
l
a
ss
c
l
a
ssi
f
i
c
a
t
i
o
n
u
s
i
n
g
h
y
b
r
i
d
s
o
f
t
d
e
c
i
s
i
o
n
m
o
d
e
l
f
o
r
a
g
r
i
c
u
l
t
u
r
e
c
r
o
p
se
l
e
c
t
i
o
n
,
”
N
e
u
r
a
l
C
o
m
p
u
t
i
n
g
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
3
0
,
n
o
.
4
,
p
p
.
1
0
2
5
–
1
0
3
8
,
A
u
g
.
2
0
1
8
,
d
o
i
:
1
0
.
1
0
0
7
/
s
0
0
5
2
1
-
0
1
6
-
2
7
4
9
-
y.
[
1
8
]
A
n
j
a
n
a
,
A
.
K
.
K
,
A
.
S
a
n
a
,
B
.
A
.
B
h
a
t
,
S
.
K
u
m
a
r
,
a
n
d
N
.
B
h
a
t
,
“
A
n
e
f
f
i
c
i
e
n
t
a
l
g
o
r
i
t
h
m
f
o
r
p
r
e
d
i
c
t
i
n
g
c
r
o
p
u
si
n
g
h
i
st
o
r
i
c
a
l
d
a
t
a
a
n
d
p
a
t
t
e
r
n
m
a
t
c
h
i
n
g
t
e
c
h
n
i
q
u
e
,
”
G
l
o
b
a
l
T
r
a
n
si
t
i
o
n
s
Pr
o
c
e
e
d
i
n
g
s
,
v
o
l
.
2
,
n
o
.
2
,
p
p
.
2
9
4
–
2
9
8
,
N
o
v
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
g
l
t
p
.
2
0
2
1
.
0
8
.
0
6
0
.
[
1
9
]
R
.
K
u
m
a
r
,
M
.
P
.
S
i
n
g
h
,
P
.
K
u
m
a
r
,
a
n
d
J
.
P
.
S
i
n
g
h
,
“
C
r
o
p
s
e
l
e
c
t
i
o
n
me
t
h
o
d
t
o
m
a
x
i
mi
z
e
c
r
o
p
y
i
e
l
d
r
a
t
e
u
si
n
g
mac
h
i
n
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
,
”
i
n
2
0
1
5
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
S
m
a
r
t
T
e
c
h
n
o
l
o
g
i
e
s
a
n
d
M
a
n
a
g
e
m
e
n
t
f
o
r C
o
m
p
u
t
i
n
g
,
C
o
m
m
u
n
i
c
a
t
i
o
n
,
C
o
n
t
ro
l
s,
En
e
r
g
y
a
n
d
Ma
t
e
r
i
a
l
s (I
C
S
T
M)
,
M
a
y
2
0
1
5
,
p
p
.
1
3
8
–
1
4
5
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
S
TM
.
2
0
1
5
.
7
2
2
5
4
0
3
.
[
2
0
]
A
.
S
u
r
u
l
i
a
n
d
i
,
G
.
M
a
r
i
a
m
mal
,
a
n
d
S
.
P
.
R
a
j
a
,
“
C
r
o
p
p
r
e
d
i
c
t
i
o
n
b
a
s
e
d
o
n
so
i
l
a
n
d
e
n
v
i
r
o
n
me
n
t
a
l
c
h
a
r
a
c
t
e
r
i
st
i
c
s
u
si
n
g
f
e
a
t
u
r
e
sel
e
c
t
i
o
n
t
e
c
h
n
i
q
u
e
s,
”
M
a
t
h
e
m
a
t
i
c
a
l
a
n
d
C
o
m
p
u
t
e
r
M
o
d
e
l
l
i
n
g
o
f
D
y
n
a
m
i
c
a
l
S
y
s
t
e
m
s
,
v
o
l
.
2
7
,
n
o
.
1
,
p
p
.
1
1
7
–
1
4
0
,
J
a
n
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
8
0
/
1
3
8
7
3
9
5
4
.
2
0
2
1
.
1
8
8
2
5
0
5
.
[
2
1
]
R
.
R
.
M
y
t
h
i
l
i
K
.
,
“
C
r
o
p
R
e
c
o
mm
e
n
d
a
t
i
o
n
f
o
r
b
e
t
t
e
r
c
r
o
p
y
i
e
l
d
f
o
r
p
r
e
c
i
s
i
o
n
a
g
r
i
c
u
l
t
u
r
e
u
si
n
g
a
n
t
c
o
l
o
n
y
o
p
t
i
m
i
z
a
t
i
o
n
w
i
t
h
d
e
e
p
l
e
a
r
n
i
n
g
m
e
t
h
o
d
,
”
An
n
a
l
s
o
f
t
h
e
R
o
m
a
n
i
a
n
S
o
c
i
e
t
y
f
o
r
C
e
l
l
Bi
o
l
o
g
y
,
v
o
l
.
2
5
,
p
p
.
4
7
8
3
–
4
7
9
4
,
2
0
2
1
,
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s
:
/
/
w
w
w
.
a
n
n
a
l
s
o
f
r
sc
b
.
r
o
/
i
n
d
e
x
.
p
h
p
/
j
o
u
r
n
a
l
/
a
r
t
i
c
l
e
/
v
i
e
w
/
3
0
2
4
.
[
2
2
]
S
.
K
.
A
p
a
t
,
J
.
M
i
s
h
r
a
,
K
.
S
.
R
a
j
u
,
a
n
d
N
.
P
a
d
h
y
,
“
A
n
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
-
b
a
s
e
d
c
r
o
p
r
e
c
o
m
m
e
n
d
a
t
i
o
n
s
y
s
t
e
m
u
s
i
n
g
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
,
”
J
o
u
r
n
a
l
o
f
S
c
i
e
n
t
i
f
i
c
a
n
d
I
n
d
u
s
t
r
i
a
l
R
e
s
e
a
r
c
h
,
v
o
l
.
8
2
,
n
o
.
5
,
p
p
.
5
5
8
–
5
6
7
,
M
a
y
2
0
2
3
,
d
o
i
:
1
0
.
5
6
0
4
2
/
j
s
i
r
.
v
8
2
i
0
5
.
1
0
9
2
.
[
2
3
]
Y
.
M
a
h
a
l
e
e
t
a
l
.
,
“
C
r
o
p
r
e
c
o
mm
e
n
d
a
t
i
o
n
a
n
d
f
o
r
e
c
a
s
t
i
n
g
sy
s
t
e
m
f
o
r
M
a
h
a
r
a
s
h
t
r
a
u
si
n
g
mac
h
i
n
e
l
e
a
r
n
i
n
g
w
i
t
h
LST
M
:
a
n
o
v
e
l
e
x
p
e
c
t
a
t
i
o
n
-
ma
x
i
m
i
z
a
t
i
o
n
t
e
c
h
n
i
q
u
e
,
”
D
i
sc
o
v
e
r
S
u
s
t
a
i
n
a
b
i
l
i
t
y
,
v
o
l
.
5
,
n
o
.
1
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
0
7
/
s
4
3
6
2
1
-
0
2
4
-
0
0
2
9
2
-
5.
[
2
4
]
E.
E
l
b
a
si
e
t
a
l
.
,
“
C
r
o
p
p
r
e
d
i
c
t
i
o
n
mo
d
e
l
u
s
i
n
g
mac
h
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
m
s,”
A
p
p
l
i
e
d
S
c
i
e
n
c
e
s
(
S
w
i
t
zerl
a
n
d
)
,
v
o
l
.
1
3
,
n
o
.
1
6
,
p
.
9
2
8
8
,
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
a
p
p
1
3
1
6
9
2
8
8
.
[
2
5
]
P
.
R
a
w
a
t
,
M
.
B
a
j
a
j
,
S
.
V
a
t
s
,
a
n
d
V
.
S
h
a
r
ma
,
“
A
n
a
n
a
l
y
si
s
o
f
c
r
o
p
r
e
c
o
mm
e
n
d
a
t
i
o
n
sy
s
t
e
m
s
e
m
p
l
o
y
i
n
g
d
i
v
e
r
se
mac
h
i
n
e
l
e
a
r
n
i
n
g
met
h
o
d
o
l
o
g
i
e
s
,
”
P
ro
c
e
e
d
i
n
g
s
-
I
E
EE
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
D
e
v
i
c
e
I
n
t
e
l
l
i
g
e
n
c
e
,
C
o
m
p
u
t
i
n
g
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
T
e
c
h
n
o
l
o
g
i
e
s
,
D
I
C
C
T
2
0
2
3
,
p
p
.
6
1
9
–
6
2
4
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
D
I
C
C
T
5
6
2
4
4
.
2
0
2
3
.
1
0
1
1
0
0
8
5
.
[
2
6
]
P
.
S
a
m
u
e
l
,
B
.
S
a
h
i
t
h
i
,
T.
S
a
h
e
l
i
,
D
.
R
a
ma
n
i
k
a
,
a
n
d
N
.
A
.
K
u
m
a
r
,
“
C
r
o
p
p
r
i
c
e
p
r
e
d
i
c
t
i
o
n
s
y
st
e
m
u
si
n
g
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
ms,
”
Q
u
e
st
J
o
u
r
n
a
l
s
J
o
u
r
n
a
l
o
f
S
o
f
t
w
a
re
E
n
g
i
n
e
e
r
i
n
g
a
n
d
S
i
m
u
l
a
t
i
o
n
,
v
o
l
.
6
,
n
o
.
1
,
p
p
.
1
4
–
2
0
,
2
0
2
0
.
[
2
7
]
T.
G
u
p
t
a
,
S
.
M
a
g
g
u
,
a
n
d
B
.
K
a
p
o
o
r
,
“
C
r
o
p
p
r
e
d
i
c
t
i
o
n
u
s
i
n
g
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
,
”
I
c
o
n
i
c
Re
s
e
a
rc
h
An
d
E
n
g
i
n
e
e
ri
n
g
J
o
u
r
n
a
l
s
,
v
o
l
.
6
,
n
o
.
9
,
p
p
.
2
7
9
–
2
8
4
,
2
0
2
3
.
[
2
8
]
C
.
C
h
a
n
d
a
n
a
a
n
d
G
.
P
a
r
t
h
a
s
a
r
a
t
h
y
,
“
Ef
f
i
c
i
e
n
t
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
r
e
g
r
e
ss
i
o
n
a
l
g
o
r
i
t
h
m
u
s
i
n
g
N
a
ï
v
e
B
a
y
e
s
c
l
a
ssi
f
i
e
r
f
o
r
c
r
o
p
y
i
e
l
d
p
r
e
d
i
c
t
i
o
n
a
n
d
o
p
t
i
ma
l
u
t
i
l
i
z
a
t
i
o
n
o
f
f
e
r
t
i
l
i
z
e
r
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
P
e
rf
o
rm
a
b
i
l
i
t
y
E
n
g
i
n
e
e
r
i
n
g
,
v
o
l
.
1
8
,
n
o
.
1
,
p
p
.
4
7
–
5
5
,
2
0
2
2
,
d
o
i
:
1
0
.
2
3
9
4
0
/
i
j
p
e
.
2
2
.
0
1
.
p
6
.
4
7
5
5
.
[
2
9
]
“
S
o
i
l
h
e
a
l
t
h
c
a
r
d
s
c
h
e
me
.
”
h
t
t
p
s:
/
/
s
o
i
l
h
e
a
l
t
h
.
d
a
c
.
g
o
v
.
i
n
(
a
c
c
e
sse
d
S
e
p
.
0
8
,
2
0
2
4
)
.
[
3
0
]
B
.
H
ss
i
n
a
,
A
.
M
e
r
b
o
u
h
a
,
H
.
Ez
z
i
k
o
u
r
i
,
a
n
d
M
.
Er
r
i
t
a
l
i
,
“
A
c
o
m
p
a
r
a
t
i
v
e
st
u
d
y
o
f
d
e
c
i
s
i
o
n
t
r
e
e
I
D
3
a
n
d
C
4
.
5
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
Ad
v
a
n
c
e
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
4
,
n
o
.
2
,
2
0
1
4
,
d
o
i
:
1
0
.
1
4
5
6
9
/
s
p
e
c
i
a
l
i
ss
u
e
.
2
0
1
4
.
0
4
0
2
0
3
.
[
3
1
]
F
.
R
o
s
e
n
b
l
a
t
t
,
“
T
h
e
p
e
r
c
e
p
t
r
o
n
:
a
p
r
o
b
a
b
i
l
i
st
i
c
m
o
d
e
l
f
o
r
i
n
f
o
r
mat
i
o
n
s
t
o
r
a
g
e
a
n
d
o
r
g
a
n
i
z
a
t
i
o
n
i
n
t
h
e
b
r
a
i
n
,
”
Psy
c
h
o
l
o
g
i
c
a
l
Re
v
i
e
w
,
v
o
l
.
6
5
,
n
o
.
6
,
p
p
.
1
9
–
2
7
,
1
9
5
8
.
[
3
2
]
G
.
G
u
o
,
H
.
W
a
n
g
,
D
.
B
e
l
l
,
Y
.
B
i
,
a
n
d
K
.
G
r
e
e
r
,
“
K
N
N
m
o
d
e
l
-
b
a
se
d
a
p
p
r
o
a
c
h
i
n
c
l
a
ssi
f
i
c
a
t
i
o
n
,
”
i
n
O
n
T
h
e
M
o
v
e
t
o
M
e
a
n
i
n
g
f
u
l
I
n
t
e
r
n
e
t
S
y
s
t
e
m
s
2
0
0
3
:
C
o
o
p
I
S
,
D
O
A,
a
n
d
O
D
BA
S
E.
O
T
M
2
0
0
3
,
2
0
0
3
,
p
p
.
9
8
6
–
9
9
6
.
Evaluation Warning : The document was created with Spire.PDF for Python.