I
n
t
ern
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
E
l
ect
ri
ca
l
a
n
d
C
o
m
p
u
t
er E
n
g
i
n
eeri
n
g
(
I
J
E
C
E
)
V
o
l.
8
,
No
.
5
,
O
c
t
obe
r
20
1
8,
p
p.
29
94~
3
002
I
S
S
N
:
2088
-
8708
,
D
O
I
:
10.
11
591/
i
j
ece
.
v8
i
5
.
pp
299
4
-
3002
2994
Jou
r
n
al
h
om
e
p
age
:
h
ttp
:
//ia
e
s
c
o
r
e
.
c
o
m/
j
our
nal
s
/
i
nde
x
.
php/
I
J
E
C
E
An
A
rc
hi
t
ec
t
u
re
f
o
r
S
im
pli
f
i
e
d a
nd
A
u
t
om
at
ed
M
a
c
hi
ne
L
ea
rn
ing
Ji
t
t
ap
oo P
ool
w
an
,
Suc
ha
S
m
a
nc
ha
t
F
a
c
ul
t
y
of
I
nf
or
m
a
t
i
on T
e
c
hnol
og
y
,
K
i
ng
M
ong
k
ut
’
s
U
ni
v
e
r
s
i
t
y
o
f
T
e
c
hnol
og
y
N
or
t
h B
a
ng
k
ok
,
T
ha
i
l
a
nd
A
rt
i
cl
e I
n
f
o
AB
S
T
RAC
T
A
r
tic
le
h
is
to
r
y
:
R
ecei
v
ed
O
c
t 3
0
,
201
7
Re
v
i
se
d
J
a
n 1
2
,
2
01
8
A
ccep
t
ed
Au
g
5
,
201
8
R
ecen
t
l
y
,
m
ach
i
n
e l
ear
n
i
n
g
h
as
b
een
ad
o
p
t
ed
b
y
b
u
s
i
n
es
s
es
t
o
an
al
y
ze t
h
ei
r
v
as
t
d
at
a
i
n
o
r
d
er
t
o
m
ak
e s
t
r
at
eg
i
c
d
eci
s
i
o
n
.
H
o
w
ev
er
,
k
now
l
e
dge
i
n
m
ach
i
n
e l
ear
n
i
n
g
an
d
t
ech
n
i
cal
s
k
i
l
l
ar
e u
s
u
al
l
y
r
eq
u
i
r
ed
t
o
p
r
ep
ar
e d
at
a an
d
p
er
f
o
r
m
m
ach
i
n
e l
ear
n
i
n
g
t
as
k
s
.
T
h
i
s
o
b
s
t
acl
e p
r
e
v
en
t
s
s
m
al
l
er
b
u
s
i
n
es
s
es
w
ith
n
o
te
c
h
n
ic
a
l k
n
o
w
le
d
g
e
to
u
tiliz
e
m
a
c
h
in
e
le
a
r
n
in
g.
I
n t
h
i
s
p
ap
er
,
w
e
pr
o
pos
e
a
n
ar
ch
i
t
ect
u
r
e f
o
r
s
i
m
p
l
i
f
i
ed
an
d
au
t
o
m
at
ed
m
ach
i
n
e l
ear
n
i
n
g
p
r
o
ces
s
cu
r
r
en
t
l
y
s
u
p
p
o
r
t
i
n
g
t
h
e
d
at
a cl
as
s
i
f
i
cat
i
o
n
t
as
k
.
T
h
e ar
ch
i
t
ect
u
r
e
i
nc
l
u
de
s
a
m
e
t
hod f
or
c
ha
r
a
c
t
e
r
i
z
i
ng
da
t
a
s
e
t
s
,
w
hi
c
h a
l
l
ow
s
f
or
s
i
m
pl
i
f
y
i
ng
a
nd a
ut
om
a
t
i
ng
m
a
c
hi
ne
l
e
a
r
ni
ng
m
ode
l
a
nd hy
pe
r
pa
r
am
et
er
s
el
e
c
t
i
o
n
b
as
ed
on
hi
s
t
or
i
c
a
l
e
x
e
c
ut
i
on
c
onf
i
g
ur
a
t
i
ons
.
U
s
e
r
s
c
a
n s
i
m
pl
y
upl
oa
d t
he
i
r
da
t
a
s
e
t
s
v
i
a
a
w
e
b br
ow
s
e
r
,
a
nd t
he
s
y
s
t
e
m
w
i
l
l
de
t
e
r
m
i
ne
t
he
pos
s
i
bl
e
m
ode
l
s
a
nd
t
he
i
r
hy
pe
r
pa
r
a
m
e
t
e
r
c
onf
i
g
ur
a
t
i
ons
f
or
t
he
us
e
r
s
t
o
c
h
oos
e
f
r
om
.
T
he
pr
ot
ot
y
pe
s
how
s
t
he
f
e
a
s
i
bi
l
i
t
y
of
t
he
pr
op
os
e
d a
r
c
hi
t
e
c
t
ur
e
.
A
l
t
ho
ug
h t
he
a
c
c
u
r
a
c
y
is
s
till lim
ite
d
b
y
th
e
s
m
a
ll e
x
e
c
u
tio
n
h
is
to
r
y
a
n
d
th
e
c
l
e
a
n
lin
e
s
s
o
f
t
h
e i
n
p
u
t
d
at
as
et
s
,
t
h
e ar
ch
i
t
ect
u
r
e can
m
i
n
i
m
i
z
e u
s
er
i
n
v
o
l
v
em
en
t
i
n
t
h
e
m
ach
i
n
e l
ear
n
i
n
g
p
r
o
ces
s
s
o
t
h
at
n
o
n
-
t
ech
n
i
cal
u
s
er
s
can
p
e
r
f
o
r
m
d
at
a
c
l
a
s
s
i
f
i
c
a
t
i
on t
hr
oug
h a
w
e
b br
ow
s
e
r
w
i
t
hout
i
ns
t
a
l
l
i
ng
a
ny
s
of
t
w
a
r
e
.
Ke
y
wo
rd
:
A
ut
o
ML
D
at
a cl
as
s
i
f
i
cat
i
o
n
D
a
t
a
m
i
ni
ng
M
ach
i
n
e l
ear
n
i
n
g
C
opy
r
i
g
ht
©
201
8
I
ns
t
i
t
ut
e
o
f
A
d
v
anc
e
d E
ngi
ne
e
r
i
ng
an
d
Sc
i
e
nc
e
.
All r
ig
h
ts
r
es
er
ved
.
Co
rre
sp
o
n
d
i
n
g
Au
t
h
o
r
:
J
i
t
t
a
poo P
o
ol
w
a
n
,
F
a
c
u
l
t
y
of
I
nf
or
m
a
t
i
on
T
e
c
h
nol
ogy
,
K
i
ng
M
ong
ku
t
’
s
U
n
i
v
e
r
s
i
t
y
of
T
e
c
h
n
ol
ogy
N
or
t
h
B
a
ngk
ok
,
1518
P
r
a
c
h
a
r
a
t
1
R
d,
W
on
g
s
a
w
a
ng
,
B
a
n
gs
u
e
,
B
a
ng
k
o
k
,
110
00,
T
h
a
i
l
a
n
d.
E
m
a
il: j
.
p
o
o
lw
a
n
@
g
m
a
i
l.
c
o
m
1.
I
NT
RO
D
UCT
I
O
N
I
n
th
is
m
o
d
e
r
n
w
o
r
ld
,
b
u
s
i
n
e
s
s
e
s
u
s
u
a
ll
y
u
til
iz
e
d
a
ta
m
in
i
n
g
to
h
e
lp
m
a
k
in
g
s
e
n
s
e
o
f
th
e
ir
d
a
ta
l
ead
i
n
g
t
o
s
t
r
at
e
g
i
c d
eci
s
i
o
n
.
A
m
o
n
g
d
i
f
f
er
en
t
m
et
h
o
d
s
i
n
d
at
a
m
i
n
i
n
g
an
d
an
al
y
t
i
c
s
,
m
ac
h
i
n
e l
ear
n
i
n
g
ca
n
b
e
e
m
p
l
o
y
ed
t
o
g
e
n
er
at
e v
a
l
u
e
s
f
r
o
m
i
m
m
e
n
s
e d
at
a
s
u
c
h
as
b
y
t
h
e d
at
a cl
as
s
i
f
i
cat
i
o
n
.
H
o
w
e
v
er
,
i
t
o
f
t
en
r
eq
u
i
r
es
t
ech
n
i
cal
s
k
i
l
l
an
d
k
n
o
w
l
ed
g
e t
o
i
m
p
l
e
m
en
t
an
d
u
t
i
l
i
ze
m
ach
i
n
e
l
ear
n
i
n
g
.
T
h
i
s
o
b
s
t
a
cl
e p
r
ev
en
t
s
s
m
al
l
er
b
u
s
i
n
es
s
e
s
w
i
t
h
l
i
m
i
t
ed
t
ech
n
i
cal
k
n
o
w
l
ed
g
e t
o
e
m
p
l
o
y
m
ac
h
i
n
e l
ear
n
i
n
g
i
n
o
r
d
er
t
o
co
m
p
et
e i
n
t
h
e
m
ar
k
et
s
.
M
an
y
m
ach
i
n
e l
ear
n
i
n
g
s
er
v
i
ces
an
d
s
o
f
t
w
ar
e
l
i
b
r
ar
i
es
h
av
e b
eco
m
e
av
ai
l
ab
l
e d
u
r
i
n
g
t
h
e p
as
t
f
e
w
y
ear
s
s
u
c
h
a
s
A
m
azo
n
Mac
h
i
n
e
L
ear
n
i
n
g
a
n
d
T
e
ns
o
r
Fl
o
w
[
1
]
.
T
h
es
e s
er
v
i
ce
s
at
t
e
m
p
t
t
o
eas
e t
h
e
m
ac
h
i
n
e
l
ear
n
i
n
g
p
r
o
ces
s
f
o
r
u
s
er
s
b
y
r
ed
u
ci
n
g
h
u
m
a
n
e
f
f
o
r
t
.
H
o
w
ev
er
,
t
as
k
s
s
u
c
h
a
s
d
at
a p
r
ep
r
o
ces
s
i
n
g
,
s
el
ect
i
n
g
a
n
ap
p
r
o
p
r
i
at
e m
ach
i
n
e
l
ear
n
i
ng
m
ode
l
,
a
n
d hy
pe
r
pa
r
a
m
e
t
e
r
t
u
r
n
i
ng
a
r
e
n
on
-
tr
i
v
ia
l a
n
d
s
til
l r
e
q
u
ir
e
te
c
h
n
ic
a
l
s
k
il
l to
p
e
r
f
o
r
m
[
2
-
4
]
.
T
h
i
s
p
ap
er
p
r
es
en
t
s
an
ar
ch
i
t
e
ct
u
r
e d
es
i
g
n
f
o
r
a s
i
m
p
l
i
f
i
ed
a
n
d
au
t
o
m
a
t
ed
m
ach
i
n
e l
ear
n
i
n
g
p
r
o
ces
s
f
o
r
d
at
a cl
as
s
i
f
i
cat
i
o
n,
w
hi
c
h
i
s
o
ne
o
f
t
he
m
o
s
t
c
o
m
m
o
n
d
a
t
a
m
i
ni
n
g t
a
s
ks
.
O
ur
a
i
m
i
s
t
o
m
i
ni
m
i
z
e
us
e
r
i
nvo
l
ve
m
e
n
t
s
o
t
ha
t
e
nd
us
e
r
s
,
w
i
t
h l
i
m
i
t
e
d
k
no
w
l
e
d
ge
a
nd
t
e
c
h
ni
c
a
l
s
ki
l
l
i
n
m
a
c
hi
ne
l
e
a
r
ni
n
g,
c
a
n
s
i
m
p
l
y
u
p
l
o
ad
t
h
ei
r
t
r
ai
n
i
n
g
d
at
a t
o
t
h
e s
y
s
t
e
m
v
i
a
a
w
eb
b
r
o
w
s
er
an
d
u
s
e t
he
a
ut
o
-
g
en
er
at
ed
cl
a
s
s
i
f
i
cat
i
o
n
m
o
d
el
t
o
c
l
a
s
s
i
f
y t
he
i
r
f
ut
ur
e
d
a
t
a
w
i
t
ho
ut
ha
vi
n
g t
o
i
n
s
t
a
l
l
a
n
y
s
o
f
t
w
a
r
e
.
T
he
s
ys
t
e
m
c
ur
r
e
nt
l
y
s
up
p
o
r
t
s
f
i
ve
c
la
s
s
i
f
ic
a
tio
n
m
o
d
e
ls
i
n
c
l
u
d
in
g
K
-
n
ear
es
t
n
ei
g
h
b
o
r
s
,
d
eci
s
i
o
n
t
r
ees
,
ar
t
i
f
i
ci
al
n
eu
r
al
n
et
w
o
r
k
,
s
u
p
p
o
r
t
v
ect
o
r
m
ach
i
n
e,
a
n
d
N
aï
v
e B
a
y
e
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
E
l
e
c
&
C
o
m
p
E
ng
I
S
S
N
:
2088
-
8708
A
n
A
r
c
h
i
te
c
tu
r
e
fo
r
S
i
m
pl
i
f
i
e
d a
nd A
ut
om
a
t
e
d
M
a
c
hi
ne
…
(
J
i
t
t
a
poo P
ool
w
an
)
2995
T
o
ach
i
ev
e o
u
r
o
b
j
ect
i
v
e,
s
ev
e
r
al
co
m
p
o
n
e
n
t
s
i
n
m
ac
h
i
n
e l
ear
n
i
n
g
p
r
o
ces
s
n
eed
t
o
b
e au
t
o
m
at
ed
[
5
]
.
F
i
r
s
t
l
y
,
t
h
e s
y
s
t
e
m
n
eed
s
t
o
k
n
o
w
(
o
r
at
l
eas
t
ap
p
r
o
x
i
m
at
e)
t
h
e t
y
p
e o
f
each
d
at
a f
i
e
l
d
i
n
a d
at
as
et
u
p
l
o
ad
ed
b
y
t
h
e u
s
er
,
an
d
t
h
e
n
p
r
ep
ar
es
t
h
e d
at
as
et
f
o
r
m
a
ch
i
n
e l
ear
n
i
n
g
p
r
o
ces
s
.
A
t
t
h
i
s
s
t
a
g
e,
t
h
e
d
at
as
et
i
s
as
s
u
m
ed
t
o
b
e i
n
t
ab
u
l
ar
f
o
r
m
at
an
d
t
h
e l
as
t
co
l
u
m
n
i
s
t
h
e cl
a
s
s
.
S
ec
o
n
d
l
y
,
t
h
e s
y
s
t
e
m
n
eed
s
t
o
s
el
ect
an
ap
p
r
o
p
r
i
at
e
m
ach
i
n
e
l
ear
n
i
n
g
m
o
d
el
a
n
d
i
t
s
h
y
p
er
p
ar
a
m
et
er
o
r
o
f
f
er
c
h
o
i
ces
t
o
t
h
e
u
s
er
b
as
ed
o
n
cer
t
ai
n
cr
i
t
er
i
a
s
u
ch
as
t
h
e cl
as
s
i
f
i
cat
i
o
n
accu
r
ac
y
.
T
h
e s
el
ect
i
o
n
ap
p
r
o
ach
s
h
o
u
l
d
al
s
o
co
n
s
i
d
er
r
es
p
o
n
s
e t
i
m
e b
ec
au
s
e e
n
d
u
s
er
s
m
a
y
n
o
t
b
e f
a
m
i
l
i
ar
w
i
t
h
l
o
n
g
p
r
o
ces
s
i
n
g
t
i
m
e co
n
s
u
m
ed
b
y
e
x
i
s
t
i
n
g
h
y
p
er
p
ar
a
m
et
er
s
el
ect
i
o
n
ap
p
r
o
ach
es
[
5
-
7
]
.
F
i
na
l
l
y,
t
he
s
ys
t
e
m
s
ho
ul
d
a
ut
o
m
a
t
i
c
a
l
l
y b
ui
l
d
a
nd
t
r
a
i
n
t
he
s
e
l
e
c
t
e
d
m
a
c
hi
ne
l
e
a
r
ni
n
g
m
o
d
e
l
,
a
nd
t
he
n
p
r
o
v
i
d
es
an
i
n
t
er
f
ace f
o
r
u
s
er
s
t
o
p
er
f
o
r
m
cl
as
s
i
f
i
cat
i
o
n
o
n
t
h
ei
r
d
at
a.
2.
RE
L
AT
E
D W
O
RK
W
i
t
h t
he
a
d
va
nc
e
m
e
nt
i
n c
l
o
u
d
c
o
m
p
ut
i
n
g,
s
e
ve
r
a
l
m
a
c
hi
ne
l
e
a
r
ni
n
g s
e
r
vi
c
e
s
ha
ve
b
e
e
n l
a
unc
he
d
a
s
cl
o
u
d
s
er
v
i
ces
o
v
er
t
h
e p
as
t
f
e
w
y
ear
s
.
F
o
r
ex
a
m
p
l
e,
A
m
azo
n
Mach
i
n
e L
ear
n
i
n
g
o
f
f
er
s
cl
o
u
d
-
b
as
ed
m
ac
h
i
n
e
le
a
r
n
in
g
s
e
r
v
ic
e
th
a
t
s
u
p
p
o
r
ts
m
o
d
e
l b
u
ild
in
g
a
n
d
d
a
ta
p
r
e
d
i
c
tio
n
.
I
t f
a
c
ilita
te
s
m
a
c
h
i
n
e
le
a
r
n
in
g
b
y
p
r
ov
i
di
ng
to
o
ls
f
o
r
d
a
ta
tr
a
n
s
f
o
r
m
a
tio
n
,
m
o
d
e
l e
v
a
l
u
a
tio
n
a
n
d
v
i
s
u
a
liz
a
tio
n
.
A
p
ar
t
f
r
o
m
co
m
m
er
ci
a
l
s
er
v
i
c
es
,
m
ac
h
i
n
e l
ear
n
i
n
g
l
i
b
r
ar
i
es
h
av
e
al
s
o
b
een
p
u
b
l
i
s
h
ed
.
T
en
s
o
r
F
l
o
w
[
1
]
i
s
a g
r
ap
h
-
b
as
ed
d
at
af
l
o
w
f
o
r
n
u
m
er
i
cal
co
m
p
u
t
at
i
o
n
.
I
t
f
aci
l
i
t
at
e
s
m
ach
i
n
e l
ear
n
i
n
g
,
e
s
p
eci
al
l
y
i
n
n
e
u
r
al
n
e
t
w
or
k
a
n
d de
e
p l
e
a
r
n
i
ng
,
by
pr
ov
i
di
n
g pr
og
r
a
m
m
i
n
g
l
i
br
a
r
y
a
n
d
m
a
c
h
i
n
e
l
e
a
r
ni
ng da
t
a
f
l
o
w
pr
oc
e
s
s
e
xe
c
ut
i
o
n.
S
c
i
ki
t
-
le
a
r
n
[
8
]
,
a
va
i
l
a
b
l
e
i
n P
yt
ho
n p
r
o
gr
a
m
m
i
n
g l
a
n
gua
ge
,
i
s
a
m
a
c
h
i
ne
l
e
a
r
ni
n
g l
i
b
r
a
r
y
f
aci
l
i
t
at
i
n
g
d
at
a
m
i
n
i
n
g
an
d
a
n
al
y
t
i
cs
t
as
k
s
s
u
c
h
as
d
at
a
p
r
ep
r
o
ces
s
i
n
g
,
d
at
a cl
as
s
i
f
i
cat
i
o
n
,
d
at
a
cl
u
s
t
er
i
n
g
,
an
d
m
o
d
el
s
el
ect
i
o
n
.
A
l
t
ho
ug
h
t
h
e
s
e s
er
v
i
ces
a
n
d
s
o
f
t
w
ar
e l
i
b
r
ar
i
e ar
e a
v
ai
l
a
b
l
e,
t
h
e
y
m
er
el
y
f
aci
l
i
t
at
e
t
h
e
m
ac
h
i
n
e
le
a
r
n
in
g
p
r
o
c
e
s
s
a
n
d
p
r
o
g
r
a
m
m
i
n
g
.
U
s
e
r
s
s
till
n
e
e
d
s
te
c
h
n
i
c
a
l k
n
o
w
le
d
g
e
a
n
d
s
k
ill to
e
f
f
e
c
tiv
e
l
y
u
tiliz
e
t
h
e
s
e
s
er
v
i
ces
an
d
l
i
b
r
ar
i
es
.
T
o
ad
d
r
es
s
t
h
i
s
i
s
s
u
e,
f
e
w
at
t
e
m
p
t
s
h
a
v
e b
een
m
ad
e t
o
a
u
t
o
m
at
e
m
ac
h
i
n
e l
ear
n
i
n
g
p
r
o
c
e
s
s.
A
u
t
o
-
W
E
K
A
[
6
,
7
]
p
r
opos
e
s
a
m
e
t
h
od f
or
m
a
c
h
i
n
e
l
e
a
r
n
i
ng
m
ode
l
an
d
h
y
p
er
p
ar
a
m
et
er
s
el
ect
i
o
n
.
I
t
ev
al
u
a
t
es
d
i
f
f
er
e
n
t
m
ac
h
i
n
e l
ear
n
i
n
g
m
o
d
el
s
f
o
r
cl
as
s
i
f
i
cat
i
o
n
p
ai
r
ed
w
i
t
h
v
ar
y
i
n
g
as
s
o
ci
at
ed
hy
pe
r
pa
r
a
m
e
t
e
r
s
ba
s
e
d on
s
e
qu
e
nt
i
a
l
m
ode
-
b
a
s
e
d
a
lg
o
r
ith
m
c
o
n
f
ig
u
r
a
tio
n
(
S
M
A
C
)
to
m
i
n
i
m
iz
e
th
e
m
is
c
la
s
s
if
ic
a
tio
n
r
a
te
.
T
h
e
d
r
a
w
b
a
c
k
o
f
t
hi
s
m
e
t
ho
d
i
s
t
he
t
i
m
e
t
a
ke
n
t
o
m
a
ke
t
he
s
e
l
e
c
t
i
o
n
o
f
a
n
o
p
t
i
m
a
l
m
o
d
el
al
o
n
g
w
i
t
h
i
t
s
h
y
p
er
p
ar
am
et
er
.
W
h
i
l
e accep
t
ab
l
e
f
o
r
ad
v
an
ced
u
s
er
s
,
t
h
i
s
d
u
r
at
i
o
n
m
a
y
n
o
t
b
e
s
at
i
s
f
act
o
r
y
f
o
r
en
d
u
s
er
s
.
A
ut
o
-
s
kl
e
a
r
n
[
5
]
,
b
as
ed
o
n
S
ci
k
i
t
-
le
a
r
n
lib
r
a
r
y
,
is
a
to
o
l f
o
r
a
n
a
u
to
m
a
tio
n
o
f
m
a
c
h
i
ne
l
e
a
r
ni
n
g
pr
oc
e
s
s
e
m
ph
a
s
i
z
i
n
g
o
n
m
ode
l
s
e
l
e
c
t
i
on
a
n
d
hy
pe
r
pa
r
a
m
e
t
e
r
t
un
i
n
g
.
T
h
i
s
t
ool
e
m
pl
o
y
s
B
a
y
e
s
i
a
n
o
p
tim
iz
a
tio
n
to
f
in
d
a
n
o
p
ti
m
a
l
m
a
c
h
i
n
e
le
a
r
n
in
g
m
o
d
e
l a
n
d
its
h
y
p
e
r
p
a
r
a
m
e
te
r
s
b
y
ite
r
a
tin
g
a
c
r
o
s
s
cl
as
s
i
f
i
cat
i
o
n
al
g
o
r
i
t
h
m
s
a
n
d
p
r
ep
r
o
ces
s
o
r
s
av
ai
l
ab
l
e
i
n t
he
s
ys
t
e
m
.
A
l
t
ho
u
gh e
f
f
i
c
i
e
nt
i
n
t
e
r
m
s
o
f
a
c
c
ur
a
c
y,
t
h
i
s
m
et
h
o
d
i
s
co
m
p
u
t
at
i
o
n
al
l
y
ex
p
e
n
s
i
v
e as
i
t
h
a
s
t
o
t
r
ai
n
a
n
d
t
es
t
a
v
ai
l
ab
l
e
m
o
d
el
s
an
d
p
r
ep
r
o
ces
s
o
r
s
acr
o
s
s
dom
a
i
n
s
of
hy
pe
r
pa
r
a
m
e
t
e
r
s
(
a
l
t
h
oug
h
n
ot
e
nt
i
r
e
l
y
)
f
or
e
a
c
h
n
e
w
da
t
a
s
e
t
.
T
hi
s
m
a
y
l
e
a
d
to
la
te
n
c
y
in
t
h
e
a
ut
o
m
a
t
i
o
n,
w
h
i
c
h
m
a
y
no
t
b
e
f
r
i
e
nd
l
y
t
o
e
nd
u
s
e
r
s
.
I
n
ad
d
i
t
i
o
n
,
u
s
er
s
n
eed
t
o
b
e
ab
l
e
t
o
cl
ean
t
h
ei
r
d
at
a
b
ef
o
r
e f
eed
i
n
g
t
h
e
m
t
o
t
h
e p
r
o
ces
s
as
t
h
e i
n
p
u
t
d
at
a ar
e u
s
ed
t
o
g
en
er
at
e m
et
a
-
f
eat
u
r
es
t
h
at
ar
e em
p
l
o
y
ed
i
n
m
o
d
e
l s
e
le
c
tio
n
.
D
e
sp
i
t
e t
h
es
e e
f
f
o
r
t
s
,
t
h
er
e i
s
s
t
i
l
l
a s
h
o
r
t
co
m
i
n
g
i
n
t
h
e au
t
o
m
at
i
o
n
o
f
m
ach
i
n
e l
ear
n
i
n
g
p
r
o
ces
s
f
o
r
en
d
u
s
er
s
.
T
ech
n
i
cal
s
k
i
l
l
s
ar
e s
t
i
l
l
r
eq
u
i
r
ed
o
f
u
s
er
s
w
h
e
n
d
et
er
m
i
n
i
n
g
an
d
p
r
ep
r
o
ces
s
i
n
g
d
i
f
f
er
en
t
t
y
p
es
o
f
d
a
t
a
.
E
ve
nt
ho
u
g
h
A
ut
o
-
s
kl
e
a
r
n o
f
f
e
r
s
a
hi
gh
p
e
r
f
o
r
m
i
n
g
h
y
p
e
r
p
a
r
a
m
e
te
r
t
u
n
in
g
te
c
h
n
i
q
u
e
,
it is
s
til
l n
o
t
s
i
m
pl
i
f
i
e
d
f
or
n
on
-
t
ec
h
n
i
ca
l
u
s
er
s
.
T
h
u
s
,
w
e ai
m
t
o
g
o
f
u
r
t
h
er
b
y
au
t
o
m
at
i
n
g
t
h
e
s
e t
ec
h
n
i
cal
co
m
p
o
n
en
t
s
s
o
t
ha
t
e
nd
us
e
r
s
w
i
t
h no
n
-
t
ec
h
n
i
ca
l
b
ack
g
r
o
u
n
d
ca
n
eas
i
l
y
ap
p
l
y
m
ach
i
n
e l
ear
n
i
n
g
,
s
p
eci
f
i
cal
l
y
f
or
da
t
a
c
la
s
s
i
f
ic
a
tio
n
in
t
h
i
s
w
o
r
k
,
to
th
e
ir
b
u
s
i
n
e
s
s
e
s
.
3.
R
ES
EA
R
C
H
M
ETH
O
D
A
s
m
en
t
i
o
n
ed
ear
l
i
er
,
t
o
a
u
t
o
m
at
e t
h
e
m
ac
h
i
n
e l
ear
n
i
n
g
p
r
o
ces
s
,
i
t
i
s
n
eces
s
ar
y
t
o
a
u
t
o
m
at
e
s
ev
er
al
s
t
ep
s
i
n
cl
u
d
i
n
g
t
h
e ap
p
r
o
x
i
m
a
t
i
o
n
o
f
d
at
a t
y
p
e,
d
at
as
et
p
r
ep
r
o
ces
s
i
n
g
,
m
ach
i
n
e
l
ear
n
i
n
g
m
o
d
el
s
el
ect
i
o
n
,
a
n
d
m
ode
l
bu
i
l
di
n
g a
n
d e
x
e
c
ut
i
on
.
I
n
t
h
i
s
w
or
k
,
f
i
v
e cl
a
s
s
i
f
i
cat
i
o
n
m
o
d
el
s
i
n
cl
u
d
i
n
g
K
-
ne
a
r
e
s
t
ne
i
ghb
o
r
s
,
d
eci
s
i
o
n
t
r
ees
(
C
4
.
5
)
,
ar
t
i
f
i
ci
al
n
eu
r
al
n
e
t
w
o
r
k
,
s
u
p
p
o
r
t
v
ect
o
r
m
ac
h
i
n
e,
an
d
N
aï
v
e B
ay
es
ar
e co
n
s
i
d
er
ed
.
T
h
es
e f
i
v
e
m
o
d
el
s
r
ep
r
es
en
t
d
i
f
f
er
e
n
t
cat
eg
o
r
i
es
o
f
m
ac
h
i
n
e l
ear
n
i
n
g
t
ech
n
i
q
u
e
s
[
9
,
10
]
.
W
e
pr
o
pos
e
a
n
ar
ch
i
t
ect
u
r
e
w
i
t
h
f
i
v
e co
m
p
o
n
en
t
s
t
o
ad
d
r
es
s
t
h
e
s
e i
s
s
u
e
s
as
d
ep
i
ct
ed
i
n
F
i
g
u
r
e 1
.
3
.1
.
D
at
a
Sc
a
nne
r
T
he
f
i
r
s
t
t
a
s
k
i
n
m
a
c
hi
ne
l
e
a
r
ni
n
g i
s
t
he
p
r
e
p
a
r
a
t
i
o
n
o
f
d
at
a
s
et
.
W
e as
s
u
m
e t
h
at
en
d
u
s
er
h
as
l
i
m
i
t
ed
t
ech
n
i
cal
k
n
o
w
l
ed
g
e
s
o
t
h
e s
y
s
t
e
m
h
as
t
o
ap
p
r
o
x
i
m
at
e t
h
e t
y
p
e
s
o
f
d
at
a i
n
t
h
e i
n
p
u
t
d
at
as
et
.
D
at
a S
can
n
er
i
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
20
88
-
8708
I
nt
J
E
l
e
c
&
C
o
m
p
E
ng,
V
o
l.
8
, N
o
.
5
,
O
c
t
o
be
r
201
8
:
2
994
-
3002
2996
th
e
p
r
o
p
o
s
e
d
a
r
c
h
ite
c
tu
r
e
h
a
n
d
le
s
th
is
ta
s
k
b
y
s
c
a
n
n
in
g
t
h
e
in
p
u
t
d
a
ta
s
e
t
[
11
,
12
]
.
F
o
r
s
m
al
l
d
at
as
et
s
,
it
is
p
o
s
s
i
b
l
e t
o
s
can
t
h
e
w
h
o
l
e d
at
as
et
s
.
H
o
w
ev
er
,
i
t
i
s
n
o
t
p
r
act
i
cal
f
o
r
d
at
as
et
s
o
f
v
er
y
l
ar
g
e s
i
ze l
ead
i
n
g
t
o
l
o
n
g
l
at
en
c
y
.
T
h
u
s
,
s
a
m
p
l
es
ar
e r
a
n
d
o
m
l
y
s
e
l
ect
ed
f
r
o
m
t
h
e d
at
as
et
f
o
r
s
can
n
i
n
g
.
T
h
e s
i
ze o
f
t
h
e s
a
m
p
l
e i
s
d
ef
i
n
ed
b
y
S=
1000l
n(
N
)
w
h
er
e
N
i
s
t
h
e s
i
ze o
f
a
g
i
v
en
d
at
as
et
.
F
o
r
ex
a
m
p
l
e,
i
f
a d
at
as
et
co
n
t
ai
n
s
o
n
e
m
i
l
l
i
o
n
r
eco
r
d
s
,
t
h
e
s
a
m
pl
e
s
i
z
e
w
oul
d be
138
16 r
e
c
o
r
ds
.
WEB INTERFACE
MACHINE LEARNING ORCHESTRATOR
DATA SCANNER
PROFILE MANAGER
MODEL BUILDER
AND EXECUTOR
RESOURCE
MANAGER
Input
Dataset
New Execution
Profile
Model and
Parameters
Request
Compute
Resource
Dataset
Profile
Execution
Profiles
New Execution
Profile
Compute
Resource
F
i
gu
r
e
1.
A
r
ch
i
t
ect
u
r
e o
v
er
v
i
e
w
W
e ch
ar
act
er
i
ze each
d
at
a f
i
el
d
i
n
t
o
t
h
r
ee as
p
ect
s
,
n
a
m
el
y
,
d
at
a
t
y
p
e,
d
at
a s
cal
e,
an
d
d
a
t
a f
o
r
m
at
.
D
a
t
a
t
y
pe
s
c
ons
i
de
r
e
d i
n t
h
e
pr
opos
e
d a
r
c
h
i
t
e
c
t
u
r
e
a
r
e
ba
s
i
c
da
t
a
t
y
pe
s
i
n
c
l
u
di
n
g i
n
t
e
g
e
r
,
f
l
oa
t
i
ng
poi
n
t
,
s
t
r
i
n
g
an
d
b
o
o
l
ean
.
D
at
a s
cal
es
ar
e t
h
e d
at
a
m
ea
s
u
r
e
m
e
n
t
s
cal
es
:
n
o
m
i
n
al
,
i
n
t
er
v
al
a
n
d
r
at
i
o
s
cal
es
.
F
i
n
al
l
y
,
d
a
ta
f
o
r
m
at
s
i
n
cl
u
d
e
bi
nar
y
,
day
of
m
ont
h
,
d
a
y o
f
w
eek
,
day
_s
t
r
i
ng
,
m
ont
h
,
m
ont
h_s
t
r
i
ng
,
ge
nde
r
,
ho
ur
24
,
hour
12
,
m
in
u
te
,
yea
r
,
a
nd
qu
ar
t
e
r
.
W
e ch
o
o
s
e
t
h
e
s
e d
at
a
f
o
r
m
at
s
as
t
h
e
y
ar
e
co
m
m
o
n
l
y
f
o
u
n
d
i
n
m
a
n
y
d
a
ta
s
e
ts
in
t
h
e
U
C
I
r
e
p
o
s
ito
r
y
[
13
]
i
n
o
r
d
er
t
o
c
r
eat
e d
at
as
et
p
r
o
f
i
l
e.
W
h
en
s
ca
n
n
i
n
g
,
each
d
at
a f
i
el
d
i
s
as
s
i
g
n
ed
a t
y
p
e
w
h
er
e
ap
p
l
i
cab
l
e i
n
t
h
e r
es
p
ect
i
v
el
y
o
r
d
er
o
f
i
n
t
e
g
e
r
,
f
l
oa
t
i
ng
poi
n
t
,
a
n
d s
t
r
i
n
g
.
F
or
e
x
a
m
p
l
e
,
“
m
a
c
h
i
n
e
101”
w
ou
l
d be
a
s
s
i
g
n
e
d a
s
s
t
r
i
ng
be
c
a
us
e
i
t
c
a
nno
t
b
e co
n
v
er
t
ed
i
n
t
o
i
n
t
eg
er
o
r
f
l
o
at
i
n
g
p
o
i
n
t
.
B
o
o
l
ean
t
y
p
e i
s
as
s
i
g
n
ed
w
h
e
n
ev
er
y
i
n
s
t
an
ce o
f
a g
i
v
e
n
d
at
a f
i
el
d
t
a
ke
s
t
he
va
l
ue
i
m
p
l
yi
n
g b
o
o
l
e
a
n s
uc
h a
s
0
,
1
,
t
r
ue
,
F
A
L
S
E
,
t
,
a
nd
F
.
R
eg
ar
d
i
n
g
d
at
a s
cal
e,
if
a
d
a
ta
f
ie
ld
is
p
r
e
v
io
u
s
l
y
id
e
n
tif
ie
d
a
s
b
o
o
l
e
a
n o
r
s
t
r
i
ng,
t
he
n i
t
i
s
a
s
s
i
gne
d
t
o
t
h
e n
o
m
i
n
al
s
cal
e.
F
o
r
an
i
n
t
eg
er
o
r
a f
l
o
at
i
n
g
p
o
i
n
t
,
i
f
t
h
er
e ex
i
s
t
a n
e
g
at
i
v
e v
al
u
e,
t
h
en
i
t
i
s
as
s
i
g
n
ed
t
o
t
h
e i
n
t
er
v
al
s
cal
e;
o
t
h
er
w
i
s
e
i
t
i
s
as
s
i
g
n
ed
t
o
t
h
e r
at
i
o
s
cal
e.
T
h
e r
eas
o
n
f
o
r
ex
cl
u
d
i
n
g
t
h
e
o
r
d
i
n
al
s
ca
le
is
th
a
t
d
et
er
m
i
n
i
n
g
w
h
e
t
h
er
a d
at
a f
i
e
l
d
r
ep
r
es
en
t
s
an
o
r
d
er
r
eq
u
i
r
es
v
er
y
s
p
eci
f
i
c
m
et
h
o
d
s
f
o
r
d
i
f
f
er
en
t
d
at
a t
y
p
es
.
A
d
at
a
f
i
el
d
i
s
as
s
i
g
n
ed
t
o
o
n
e o
r
m
o
r
e d
at
a f
o
r
m
at
s
i
f
al
l
o
f
i
t
s
v
a
l
u
e
s
m
eet
t
h
e cr
i
t
er
i
a a
s
s
p
eci
f
i
ed
in
T
a
b
le
1
.
T
h
is
is
in
te
n
d
e
d
a
s
to
c
o
v
e
r
th
e
d
if
f
e
r
e
n
t
p
o
s
s
ib
ilitie
s
o
f
t
h
e
d
a
ta
s
e
t
f
o
r
d
e
te
r
m
in
in
g
s
i
m
i
la
r
it
y
b
et
w
ee
n
d
at
as
et
s
l
at
er
o
n
.
F
r
o
m
t
h
e t
h
r
ee as
p
ect
s
,
t
h
er
e ar
e 1
9
d
at
a ch
ar
act
er
i
s
t
i
cs
i
n
t
o
t
al
t
h
at
ar
e u
s
ed
t
o
c
o
n
s
tr
u
c
t d
a
ta
s
e
t p
r
o
f
ile
.
T
ab
l
e
1
.
D
at
a
F
o
r
m
a
t C
r
ite
r
ia
D
at
a
Fo
r
m
a
t
C
r
ite
r
ia
b
i
n
a
r
y
A
n
y
d
a
ta
ty
p
e
w
it
h
o
n
l
y
tw
o
d
i
f
f
e
r
e
n
t v
a
l
u
e
s
d
a
y
_
o
f
_
m
o
n
t
h
I
n
t
eg
er
v
a
l
u
e i
n
t
h
e r
an
g
e o
f
1
-
3
1
d
a
y
_
o
f
_
w
e
e
k
I
n
t
eg
er
v
a
l
u
e i
n
t
h
e r
an
g
e o
f
1
-
7
d
a
y
_
s
t
r
S
t
r
i
n
g
r
ep
r
es
en
t
i
n
g
d
a
y
s
o
f
w
eek
s
u
ch
a
s
“S
u
n
d
a
y
”
a
n
d
“s
at
”
M
on
t
h
I
n
t
eg
er
v
a
l
u
e
i
n
t
h
e r
an
g
e 1
-
1
2
m
o
n
t
h
_
s
t
r
S
t
r
i
n
g
r
e
p
r
e
s
e
n
t
i
n
g
m
o
n
t
h
s
o
f
y
e
a
r
s
u
c
h
a
s
“
A
p
r
i
l
”
a
n
d
“
f
e
b
”
g
en
d
er
S
t
r
i
n
g
r
ep
r
es
en
t
i
n
g
g
en
d
er
s
u
ch
a
s
“m
a
l
e”
,
“m
”
,
“F
E
M
A
L
E
”
,
a
n
d
“F
”
h
o
u
r
24
I
n
t
eg
er
v
a
l
u
e i
n
t
h
e r
an
g
e 0
-
2
3
h
o
u
r
12
I
n
t
eg
er
v
a
l
u
e i
n
t
h
e r
an
g
e 0
-
1
1
m
in
u
te
I
n
t
eg
er
v
a
l
u
e i
n
t
h
e r
an
g
e 0
-
5
9
y
ea
r
I
n
t
e
g
e
r
v
a
l
u
e
i
n
t
h
e
r
a
n
g
e
1
8
0
0
-
2
10
0
q
u
a
r
t
er
I
n
t
eg
er
v
a
l
u
e i
n
t
h
e r
an
g
e 1
-
4
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
E
l
e
c
&
C
o
m
p
E
ng
I
S
S
N
:
2088
-
8708
A
n
A
r
c
h
i
te
c
tu
r
e
fo
r
S
i
m
pl
i
f
i
e
d a
nd A
ut
om
a
t
e
d
M
a
c
hi
ne
…
(
J
i
t
t
a
poo P
ool
w
an
)
2997
F
r
o
m
o
u
r
ear
l
y
t
es
t
b
y
r
ep
eat
i
n
g
t
h
e s
ca
n
o
f
t
h
e s
a
m
e d
at
as
et
s
o
n
f
i
v
e d
i
f
f
er
e
n
t
d
at
as
et
s
i
n
t
h
e
U
CI
r
e
pos
i
t
or
y
,
s
h
o
w
n
i
n
T
a
bl
e
2
,
i
t is
f
o
u
n
d
t
h
at
t
h
i
s
ap
p
r
o
ach
w
i
t
h
t
h
e d
ef
i
n
ed
s
a
m
p
l
e
s
i
ze
m
e
n
t
i
o
n
ed
ear
l
i
er
ca
n
co
n
s
i
s
t
en
t
l
y
ap
p
r
o
x
i
m
a
t
e t
h
e t
y
p
e
s
o
f
d
at
a.
3
.
2
.
P
r
o
f
ile
M
an
a
ge
r
I
n
t
h
e p
r
o
p
o
s
ed
ar
ch
i
t
ect
u
r
e,
t
h
e
s
el
ect
i
o
n
o
f
m
ac
h
i
n
e l
e
ar
n
i
n
g
m
o
d
el
a
n
d
i
t
s
h
y
p
er
p
ar
a
m
et
er
s
u
tiliz
e
s
t
h
e
h
is
to
r
y
o
f
m
o
d
el
t
r
ai
n
i
n
g
ex
ec
u
t
i
o
n
o
n
p
r
e
v
i
o
u
s
d
at
as
et
s
.
T
h
e P
r
o
f
i
l
e M
an
ag
er
co
m
p
o
n
en
t
m
an
a
g
e
s
t
h
e
s
e ex
ec
u
t
i
o
n
p
r
o
f
i
l
es
,
w
h
i
c
h
ar
e g
r
o
u
p
ed
acco
r
d
i
n
g
t
o
eac
h
m
ac
h
i
n
e l
ear
n
i
n
g
m
o
d
el
(
i
.
e.
t
h
er
e ar
e
f
i
v
e
g
r
ou
ps
o
f
pr
of
i
l
e
s
f
or
t
h
e
f
i
v
e
s
u
ppor
t
e
d c
l
a
s
s
i
f
i
c
a
t
i
o
n
m
ode
l
s
)
.
E
ach
ex
ec
u
t
i
o
n
p
r
o
f
i
l
e i
s
co
m
p
o
s
ed
o
f
a
d
a
ta
s
e
t p
r
o
file
,
hy
pe
r
par
am
e
t
e
r
s
us
e
d,
r
e
s
o
u
r
c
e
u
tiliz
a
tio
n
,
a
nd
execu
t
i
o
n
r
es
u
l
t
a
s
s
ho
w
n
i
n F
i
g
ur
e
2
.
Dataset Profile
Hyperparameters
Resource Utilization
Execution Result
F
ig
u
r
e
2
.
E
x
e
c
u
tio
n
p
r
o
f
ile
B
ecau
s
e
w
e
w
a
n
t
t
o
al
l
o
w
e
n
d
u
s
er
s
t
o
s
i
m
p
l
y
u
p
l
o
ad
t
h
ei
r
d
at
as
et
s
,
t
h
e d
at
as
e
t
s
m
a
y
n
o
t
b
e cl
ean
an
d
ap
p
r
o
p
r
i
at
e f
o
r
m
ach
i
n
e l
ear
n
i
n
g
p
r
o
ces
s
.
T
h
u
s
,
w
e c
an
n
o
t
ap
p
l
y
s
o
p
h
i
s
t
i
cat
ed
m
e
t
a
-
f
e
a
t
ur
e
s
s
uc
h a
s
t
ho
s
e
us
e
d
i
n A
ut
o
-
s
kl
e
a
r
n
[
5
]
t
o
ch
ar
act
er
i
ze d
at
as
et
s
.
I
n
o
u
r
ap
p
r
o
ach
,
t
he
dat
a
s
e
t p
r
o
file
c
o
nt
a
i
n
s
t
he
na
m
e
o
f
t
h
e
d
at
a
f
i
l
e
a
n
d
a
s
et
o
f
1
9
i
n
t
eg
er
s
t
h
at
ar
e
t
h
e
n
u
m
b
er
s
o
f
o
cc
u
r
en
ces
o
f
t
h
e 1
9
d
at
a
ch
ar
act
er
i
s
t
i
c
s
d
es
cr
i
b
ed
i
n
t
h
e p
r
ev
i
o
u
s
s
e
ct
i
o
n
.
F
o
r
ex
a
m
p
l
e,
t
h
e
f
a
m
o
u
s
“
I
r
i
s
” d
at
as
et
co
n
t
ai
n
s
f
i
v
e d
at
a f
i
el
d
s
:
f
o
u
r
f
l
o
a
t
i
ng p
o
i
nt
s
a
nd
o
ne
s
t
r
i
ng.
S
ub
s
e
q
ue
nt
l
y,
f
o
ur
r
a
t
i
o
-
s
cal
e d
at
a an
d
o
n
e n
o
m
i
al
-
s
cal
e d
at
a ar
e p
o
s
s
i
b
l
e.
A
s
f
o
r
t
h
e d
at
a
f
o
r
m
at
s
,
t
h
es
e
f
i
v
e d
at
a f
i
el
d
s
i
n
t
h
e I
r
i
s
d
at
as
et
d
o
n
o
t
m
at
ch
a
n
y
d
ef
i
n
ed
d
at
a f
o
r
m
at
.
T
h
u
s
,
t
h
e
p
r
o
f
i
l
e o
f
t
h
e
“
I
r
i
s
” d
at
as
et
i
s
as
s
h
o
w
n
i
n T
a
b
l
e
2,
a
l
o
ng
w
i
t
h t
he
e
xa
m
p
l
e
s
o
f
f
o
ur
o
t
he
r
d
a
t
a
s
e
t
s
.
T
ab
l
e
2
.
E
x
a
m
p
le
s
o
f
d
a
ta
s
e
t
p
r
o
f
ile
s
F
ile
N
a
m
e
in
te
g
e
r
f
l
o
at
s
tr
in
g
b
o
o
l
e
a
n
n
o
m
i
n
a
l
i
n
t
e
r
v
a
l
r
a
t
io
b
i
n
a
r
y
d
a
y
_
o
f
_
m
o
n
t
h
d
a
y
_
o
f
_
w
e
e
k
d
a
y
_
s
t
r
m
o
n
th
m
o
n
t
h
_
s
t
r
g
e
n
d
e
r
h
o
u
r
24
h
o
u
r
12
m
i
n
u
te
y
ear
q
u
ar
t
er
A
b
a
l
o
n
e
1
7
1
0
1
0
8
0
1
0
0
0
0
1
1
0
1
0
0
A
d
u
l
t
6
0
9
0
9
0
6
2
1
0
0
0
0
1
1
0
1
0
0
Ir
i
s
0
4
1
0
1
0
4
0
0
0
0
0
0
0
0
0
0
0
0
W
i
n
e
-
q
u
a
lit
y
-
w
h
i
te
1
1
1
0
0
0
0
1
2
0
1
0
0
1
0
0
1
1
1
0
0
Y
ea
s
t
0
8
2
0
2
0
8
1
0
0
0
0
0
0
0
0
0
0
0
H
yp
er
p
a
r
a
m
et
er
s
i
n
t
h
e e
x
e
cu
t
i
o
n
p
r
o
f
i
l
e ar
e s
p
eci
f
i
c t
o
each
m
ac
h
i
n
e l
ear
n
i
n
g
m
o
d
el
o
f
each
ex
ecu
t
i
o
n
.
O
n
l
y
i
m
p
o
r
t
an
t
h
y
p
er
p
ar
am
et
er
s
,
a
s
l
i
s
t
ed
i
n
T
ab
l
e 3
,
ar
e cu
r
r
en
t
l
y
av
a
i
l
ab
l
e
f
o
r
ad
j
u
s
t
m
e
n
t
d
u
r
i
n
g
hy
pe
r
pa
r
a
m
e
t
e
r
s
e
l
e
c
t
i
on i
n
t
h
i
s
w
or
k
t
o a
v
oi
d c
o
m
pl
e
x
i
t
y
.
A
ddi
t
i
on
a
l
hy
pe
r
pa
r
a
m
e
t
e
r
s
c
a
n
s
i
m
pl
y
be
i
nc
l
ud
e
d
i
n f
ut
ur
e
w
o
r
k b
y
a
d
d
i
ng t
o
t
he
e
xe
c
u
t
i
o
n p
r
o
f
i
l
e
.
R
e
s
o
u
r
c
e
u
tiliz
a
tio
n
i
s
c
om
pos
e
d of
C
P
U
u
tiliz
a
t
io
n
,
m
e
m
o
r
y
u
t
iliz
a
tio
n
,
a
n
d
p
r
o
c
e
s
s
in
g
ti
m
e
d
u
r
i
n
g
m
o
d
e
l b
u
ild
in
g
.
E
xecu
t
i
o
n
r
es
u
l
t
c
o
nt
a
i
ns
t
he
va
l
u
e
s
o
f
acc
u
r
ac
y
,
r
ecal
l
,
p
r
eci
s
i
o
n
,
a
n
d
F
-
m
ea
s
u
r
e f
r
o
m
t
h
e m
o
d
el
v
al
i
d
at
i
o
n
.
A
n
e
x
a
m
p
l
e
o
f
e
x
ecu
t
i
o
n
pr
of
i
l
e
of
K
N
N
m
ode
l
i
s
s
h
o
w
n
i
n
F
i
gu
r
e
3.
T
ab
l
e
3
.
H
y
p
er
p
ar
am
e
t
er
s
f
o
r
t
h
e
S
u
ppor
t
M
ode
l
s
M
o
d
el
P
a
r
a
m
et
er
s
K
-
N
ea
r
es
t
N
ei
g
h
b
o
r
s
K
-
v
a
l
u
e,
w
ei
g
h
t
t
y
p
e
(
u
ni
f
or
m
o
r
di
s
t
a
n
c
e
)
D
ec
i
s
i
o
n
T
r
ees
m
e
t
r
ic
(
e
n
t
r
o
p
y
o
r
g
in
i
)
,
s
p
l
it
te
r
A
r
t
if
ic
ia
l
N
e
u
r
a
l
N
e
tw
o
r
k
n
u
m
b
e
r
o
f
h
i
d
d
e
n
n
o
d
e
s
,
n
u
m
b
e
r
o
f
h
i
d
d
e
n
l
a
y
e
r
s
S
u
pp
or
t
V
e
c
t
o
r
M
a
c
h
i
n
e
n
u
m
b
e
r
o
f
c
l
a
s
s
v
a
l
u
e
,
k
e
r
n
e
l
(
r
b
f
,
l
i
n
e
a
r
,
p
o
l
y
,
o
r
s
i
g
m
o
i
d
)
N
a
ï
v
e B
a
y
es
n
o
n
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
20
88
-
8708
I
nt
J
E
l
e
c
&
C
o
m
p
E
ng,
V
o
l.
8
, N
o
.
5
,
O
c
t
o
be
r
201
8
:
2
994
-
3002
2998
F
i
gu
r
e
3.
A
n
e
x
a
m
pl
e
of
e
x
e
c
u
t
i
on
pr
of
i
l
e
3
.
3
.
M
a
c
hi
ne
L
ea
rn
i
n
g
O
rch
es
t
ra
t
o
r
T
h
e M
ach
i
n
e
L
ear
n
i
n
g
O
r
ch
es
t
r
at
o
r
i
s
t
h
e
m
ai
n
co
m
p
o
n
en
t
t
h
at
m
an
a
g
es
t
h
e
m
ach
i
n
e l
ear
n
i
n
g
p
r
o
ces
s
b
y
i
n
t
er
act
i
n
g
w
i
t
h
o
t
h
er
co
m
p
o
n
en
t
s
.
W
h
e
n
a u
s
er
u
p
l
o
ad
s
a n
e
w
d
at
as
e
t
i
n
t
o
t
h
e s
y
s
t
e
m
,
M
ac
h
i
n
e
L
ear
n
i
n
g
O
r
ch
e
s
t
r
at
o
r
i
n
v
o
k
e
s
t
h
e D
a
t
a S
ca
n
n
er
t
o
s
ca
n
t
h
e d
at
as
et
an
d
cr
eat
e i
t
s
d
at
a p
r
o
f
i
l
e as
d
es
cr
i
b
ed
ear
l
i
er
.
T
h
e n
ex
t
s
t
ep
i
s
t
o
d
e
t
er
m
i
n
e ap
p
r
o
p
r
i
at
e
m
o
d
el
s
a
n
d
t
h
e as
s
o
ci
at
ed
h
y
p
er
p
ar
a
m
et
er
s
.
T
o
av
o
i
d
t
h
e
l
at
en
c
y
cau
s
ed
b
y
f
i
n
e
-
t
un
i
ng a
ppr
oa
c
h
e
s
a
s
i
n
e
x
i
s
t
i
ng
w
or
k
[
4
,
5
,
7
]
,
w
e e
m
p
l
o
y
a
s
i
m
p
l
er
ap
p
r
o
ach
o
f
s
el
ect
i
n
g
t
h
e
m
o
d
el
an
d
i
t
s
p
ar
a
m
et
er
f
r
o
m
h
i
s
t
o
r
i
cal
ex
ec
u
t
i
o
n
p
r
o
f
i
l
es
t
o
ac
h
i
ev
e
f
a
s
t
er
r
es
p
o
n
s
e,
al
t
h
o
u
g
h
p
o
s
s
i
b
l
y
s
acr
i
f
i
ci
n
g
o
p
t
i
m
al
ac
cu
r
ac
y
.
T
h
e ex
ecu
t
i
o
n
p
r
o
f
i
l
es
o
f
eac
h
o
f
t
h
e f
i
v
e cl
as
s
i
f
i
c
a
tio
n
m
o
d
e
ls
a
r
e
f
ir
s
t c
lu
s
te
r
e
d
u
s
i
n
g
K
-
M
ean
c
l
us
t
e
r
i
n
g
[
14
]
. K
-
M
ean
cl
u
s
t
er
i
n
g
i
s
e
m
p
l
o
y
ed
b
ecau
s
e o
f
i
t
s
f
a
s
t
p
r
o
c
e
ssi
n
g
t
i
m
e
i
n o
r
d
e
r
t
o
a
r
c
hi
ve
f
a
s
t
r
es
p
o
n
s
e t
i
m
e
f
o
r
g
en
er
al
u
s
er
s
[
15
]
.
K
i
s
s
et
t
o
t
h
e n
u
m
b
er
o
f
u
n
i
q
u
e d
at
as
et
p
r
o
f
i
l
es
t
h
at
h
av
e b
een
t
r
ai
n
ed
p
r
ev
i
o
u
s
l
y
w
i
t
h
each
m
ac
h
i
n
e
l
ear
n
i
n
g
m
o
d
el
(
r
eg
ar
d
l
es
s
o
f
t
h
e n
a
m
e
s
o
f
t
h
e d
at
as
et
s
)
s
o
t
h
at
t
h
e r
es
u
l
t
i
n
g
cl
u
s
t
er
s
ar
e
u
n
i
q
u
e b
as
ed
o
n
t
h
e 1
9
d
at
a ch
ar
act
er
i
s
t
i
cs
.
S
u
b
s
eq
u
en
t
l
y
,
t
h
e
p
r
o
f
ile
c
l
u
s
te
r
w
ith
t
h
e
m
in
i
m
u
m
E
u
cl
i
d
ean
d
i
s
t
a
n
ce f
r
o
m
t
h
e
n
e
w
d
at
as
et
p
r
o
f
i
l
e i
s
s
el
ec
t
e
d
.
T
h
e E
u
cl
i
d
ean
d
i
s
t
an
ce
d
i
st
i
s
cal
cu
l
at
ed
as
s
ho
w
n
i
n E
q
ua
t
i
o
n 1
w
he
r
e
do
i
ar
e t
h
e d
at
as
et
ch
ar
act
er
i
s
t
i
c
s
o
f
a cl
u
s
t
er
a
n
d
dn
i
ar
e t
h
o
s
e o
f
t
h
e
n
e
w
d
at
as
et
.
2
19
1
)
(
i
i
i
dn
do
dis
t
−
=
∑
=
(1
)
H
y
p
er
p
ar
am
e
t
er
s
ar
e s
el
ect
ed
f
r
o
m
t
h
e e
x
ecu
t
i
o
n
p
r
o
f
i
l
e
s
i
n
t
h
e p
r
o
f
i
l
e cl
u
s
t
er
w
i
t
h
t
h
e
m
i
n
i
m
u
m
d
i
s
t
.
F
o
r
each
p
r
o
f
i
l
e
cl
u
s
t
er
o
f
t
h
e
f
i
v
e
m
ac
h
i
n
e
l
ear
n
i
n
g
m
o
d
el
s
,
t
h
e p
r
o
f
i
l
e
w
i
t
h
t
h
e
h
i
g
h
e
s
t
accu
r
ac
y
i
s
s
el
ect
ed
.
F
i
n
al
l
y
,
t
h
e
s
e
f
i
v
e
h
y
p
er
p
ar
a
m
et
er
co
n
f
i
g
u
r
at
i
o
n
s
ar
e p
r
es
en
t
ed
t
o
t
h
e
u
s
er
al
o
n
g
w
i
t
h
t
h
e acc
u
r
ac
y
an
d
p
r
o
ces
s
i
n
g
t
i
m
e.
T
h
e u
s
er
can
t
h
e
n
s
i
m
p
l
y
s
e
l
ect
o
n
e o
f
t
h
e
m
o
d
el
s
o
r
t
h
e s
y
s
t
e
m
ca
n
a
u
t
o
m
at
i
cal
l
y
s
el
ect
t
h
e o
n
e
w
i
t
h
t
h
e
h
i
g
h
es
t
acc
u
r
ac
y
.
T
h
e s
el
ec
t
ed
co
n
f
i
gur
a
t
i
o
n
w
i
l
l
t
he
n b
e
e
xe
c
u
t
e
d
b
y t
he
M
o
d
e
l
B
ui
l
d
e
r
a
nd
E
x
ecu
t
o
r
co
m
p
o
n
e
n
t
.
T
h
e
m
o
d
el
s
el
ect
i
o
n
p
r
o
ces
s
i
s
d
ep
i
ct
ed
i
n
F
i
g
u
r
e 4
.
I
n
t
h
i
s
ap
p
r
o
ach
,
h
y
p
er
p
ar
m
e
t
er
s
el
ect
i
o
n
r
el
i
es
h
ea
v
i
l
y
o
n
e
x
ecu
t
i
o
n
h
i
s
t
o
r
y
.
T
h
u
s
,
i
t
i
s
n
e
ces
s
ar
y
t
o
h
av
e a l
ar
g
e n
u
m
b
er
o
f
h
is
to
r
i
c
a
l e
x
e
c
u
tio
n
p
r
o
f
ile
s
w
it
h
d
if
f
e
r
e
n
t
h
y
p
e
r
p
a
r
a
m
e
te
r
s
i
n
o
r
d
e
r
f
o
r
th
e
s
y
s
te
m
to
o
f
f
er
b
et
t
er
ch
o
i
ces
f
o
r
t
h
e
u
s
er
.
I
n
i
t
i
al
l
y
,
t
h
e e
x
ecu
t
i
o
n
p
r
o
f
i
l
es
h
a
v
e b
een
cr
eat
ed
b
y
t
r
ai
n
i
n
g
t
h
e f
i
v
e
cl
as
s
i
f
i
cat
i
o
n
m
o
d
el
s
w
i
t
h
d
i
f
f
er
en
t
h
y
p
er
p
ar
a
m
et
er
s
o
n
a
num
be
r
of
da
t
a
s
e
t
s
f
r
o
m
t
h
e
U
C
I
r
e
pos
i
t
or
y
.
T
h
e
p
o
o
l
o
f
ex
ecu
t
i
o
n
p
r
o
f
i
l
es
w
i
l
l
co
n
t
i
n
u
e t
o
g
r
o
w
a
s
u
s
er
s
u
s
e t
h
e s
y
s
t
e
m
.
A
l
s
o
,
t
h
e
M
ach
i
n
e L
ear
n
i
n
g
O
r
c
h
e
s
tr
a
to
r
w
ill b
e
m
o
d
if
ie
d
to
a
u
to
m
a
tic
a
ll
y
b
u
ild
m
o
r
e
p
r
o
f
ile
s
o
f
d
if
f
e
r
e
n
t
m
o
d
e
ls
a
n
d
h
y
p
e
r
p
a
r
a
m
e
te
r
s
o
n
t
h
e av
a
i
l
ab
l
e d
at
as
et
s
w
h
en
t
h
e s
y
s
t
e
m
b
eco
m
e
s
i
d
l
e.
O
n
ce t
h
e
m
o
d
el
t
r
ai
n
i
n
g
i
s
co
m
p
l
et
ed
,
t
h
i
s
co
m
p
o
n
e
n
t
g
e
n
er
at
es
a
g
r
ap
h
i
cal
u
s
er
i
n
t
er
f
ace o
n
a
w
eb
b
r
o
w
s
er
w
i
t
h
i
n
p
u
t
f
i
el
d
s
co
r
r
es
p
o
n
d
i
n
g
t
o
t
h
e i
n
p
u
t
d
at
as
et
f
o
r
d
at
a cl
as
s
i
f
i
cat
i
o
n
b
as
ed
o
n
t
h
e f
i
n
i
s
h
ed
m
o
d
el
r
et
u
r
n
ed
b
y
t
h
e Mo
d
el
B
u
i
l
d
er
an
d
E
x
ecu
t
o
r
co
m
p
o
n
e
n
t
.
U
s
er
s
can
t
h
e
n
p
er
f
o
r
m
d
at
a cl
as
s
i
f
i
cat
i
o
n
o
n
t
h
ei
r
o
w
n
t
h
r
o
u
g
h
t
h
i
s
i
n
t
er
f
ace.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
E
l
e
c
&
C
o
m
p
E
ng
I
S
S
N
:
2088
-
8708
A
n
A
r
c
h
i
te
c
tu
r
e
fo
r
S
i
m
pl
i
f
i
e
d a
nd A
ut
om
a
t
e
d
M
a
c
hi
ne
…
(
J
i
t
t
a
poo P
ool
w
an
)
2999
KNN Profiles
K
-
Mean
Clustering
Decision Tree
Profiles
K
-
Mean
Clustering
Neural Network
Profiles
K
-
Mean
Clustering
Input Data
Profile
SVM Profiles
K
-
Mean
Clustering
Naïve Bayes
Profile
K
-
Mean
Clustering
Select profile
with highest
accuracy
Select profile
with highest
accuracy
Select profile
with highest
accuracy
Select profile
with highest
accuracy
Select profile
with highest
accuracy
User’s
selection
or
Best accuracy
selection
Nearest
cluster
Nearest
cluster
Nearest
cluster
Nearest
cluster
Nearest
cluster
F
i
g
u
r
e 4
.
M
o
d
el
s
el
ect
i
o
n
p
r
o
ces
s
3
.
4
.
M
o
de
l
B
ui
l
de
r
a
nd
E
xe
c
u
t
o
r
T
hi
s
co
m
p
o
n
e
n
t
t
a
k
es
t
h
e i
n
p
u
t
d
at
as
et
an
d
t
h
e
s
el
ect
ed
h
y
p
er
p
ar
am
et
er
co
n
f
i
g
u
r
at
i
o
n
t
o
t
r
ai
n
t
h
e
s
el
ect
ed
m
ac
h
i
n
e l
ear
n
i
n
g
m
o
d
el
.
T
h
e i
n
p
u
t
d
at
as
et
i
s
f
i
r
s
t
p
r
ep
r
o
ces
s
ed
:
an
y
r
eco
r
d
w
i
t
h
m
i
s
s
i
n
g
v
al
u
e i
s
e
x
c
l
u
de
d
f
r
o
m
t
h
e
m
ode
l
bui
l
di
n
g
;
bool
e
a
n
v
a
l
u
e
s
,
bi
n
a
r
y
va
l
u
e
s
a
n
d g
e
n
de
r
s
a
r
e
c
onv
e
r
t
e
d t
o 0 or
1;
s
t
r
i
n
gs
r
ep
r
es
en
t
i
n
g
d
a
y
s
o
f
w
ee
k
ar
e co
n
v
er
t
ed
t
o
1
-
7
;
s
tr
in
g
s
r
e
p
r
e
s
e
n
tin
g
m
o
n
t
h
s
a
r
e
c
o
n
v
e
r
te
d
to
1
-
12;
ot
h
e
r
s
t
r
i
n
g va
l
ue
s
a
r
e
c
o
nve
r
t
e
d
i
nt
o
i
nt
e
ge
r
s
.
T
h
e p
r
ep
r
o
ces
s
ed
d
at
as
et
i
s
t
h
en
u
s
ed
t
o
t
r
ai
n
t
h
e s
el
ect
ed
m
o
d
el
.
T
h
e S
ci
k
i
t
-
le
a
r
n
lib
r
a
r
y
[
8
]
i
s u
se
d
f
or
m
ode
l
bu
i
l
di
n
g
u
s
i
ng
10
-
f
o
ld
c
r
o
s
s
v
a
lid
a
tio
n
.
O
n
c
e
th
e
tr
a
in
in
g
is
c
o
m
p
le
te
d
,
th
e
p
e
r
f
o
r
m
a
n
c
e
i
nf
or
m
a
t
i
on i
n
c
l
u
di
n
g
t
h
e cl
a
s
s
i
f
i
cat
i
o
n
acc
u
r
ac
y
,
r
ecal
l
,
p
r
eci
s
i
o
n
,
F
-
m
eas
u
r
e,
p
r
o
ces
s
i
n
g
d
u
r
at
i
o
n
,
C
P
U
u
tiliz
a
t
io
n
,
a
n
d
m
e
m
o
r
y
u
tiliz
a
tio
n
is
r
e
c
o
r
d
e
d
(
a
lo
n
g
w
it
h
th
e
m
o
d
e
l a
n
d
its
h
y
p
e
r
p
a
r
a
m
e
te
r
s
)
a
s
execu
t
i
o
n
p
r
o
file
i
n t
he
P
r
o
f
i
l
e
M
a
na
ge
r
.
T
he
r
e
s
ul
t
i
n
g
m
o
d
e
l
i
s
t
he
n r
e
t
ur
ne
d
t
o
t
he
M
a
c
hi
ne
L
e
a
r
ni
ng O
r
c
he
s
t
r
a
t
o
r
t
o
g
en
er
at
e a G
U
I
f
o
r
t
h
e u
s
er
s
.
3
.
5
.
R
es
o
u
rce
M
an
a
ge
r
R
es
o
u
r
ce Ma
n
a
g
er
co
m
p
o
n
e
n
t
m
a
n
ag
e
s
t
h
e co
m
p
u
t
at
i
o
n
r
e
s
o
u
r
ces
t
h
at
ar
e
u
s
ed
i
n
t
h
e s
y
s
t
e
m
.
T
h
i
s
c
om
pon
e
nt
a
i
m
s
t
o
s
up
p
o
r
t
m
a
c
hi
ne
l
e
a
r
ni
ng e
xe
c
ut
i
o
n i
n t
he
c
l
o
ud
.
C
ur
r
e
nt
l
y,
t
he
p
r
o
t
o
t
y
p
e
i
s
b
e
i
n
g t
e
s
t
o
n
A
m
azo
n
E
C
2
.
W
h
en
a
u
s
er
u
p
l
o
ad
s
a n
e
w
d
at
as
et
,
i
t
i
s
d
i
r
ect
ed
t
o
A
m
azo
n
S
3
f
o
r
s
t
o
r
a
g
e.
O
n
ce t
h
e u
s
er
s
ta
r
ts
w
o
r
k
in
g
o
n
t
h
e
d
a
ta
s
e
t,
it is
s
ta
g
e
d
o
n
to
t
h
e
E
C
2
i
ns
t
a
n
c
e
a
n
d t
h
e
n t
h
e
m
ode
l
bu
i
l
di
n
g i
s
pe
r
f
or
m
e
d.
A
t
t
h
i
s
s
t
a
g
e,
t
h
e u
s
er
can
r
eq
u
es
t
t
o
r
u
n
m
ach
i
n
e l
ear
n
i
n
g
p
r
o
ces
s
o
n
a s
ep
ar
at
e E
C
2
i
n
s
t
a
n
ce.
H
o
w
e
v
er
,
ad
v
an
ced
cl
o
u
d
f
eat
u
r
es
s
u
c
h
as
au
t
o
s
cal
i
n
g
an
d
r
es
o
u
r
ce al
l
o
cat
i
o
n
h
av
e
n
o
t
b
een
i
n
cl
u
d
e
d
i
n
ou
r
s
c
ope
.
4.
RE
S
U
L
T
S
AND D
I
S
CU
S
S
I
O
N
T
h
e
pr
ot
ot
y
pe
of
t
h
e
pr
opos
e
d a
r
c
h
i
t
e
c
t
u
r
e
i
s
i
m
pl
e
m
e
n
t
e
d i
n
P
y
t
h
on l
a
ngu
a
g
e
a
n
d e
m
p
l
o
y
s
t
h
e
S
c
ik
it
-
le
a
r
n
lib
r
a
r
y
[
8
]
f
o
r
m
a
c
hi
ne
l
e
a
r
ni
n
g p
r
o
gr
a
m
m
i
ng.
F
o
r
t
he
e
va
l
ua
t
i
o
n o
f
o
ur
a
p
p
r
o
a
c
h
to
a
u
to
m
a
te
m
a
c
h
in
e
le
a
r
n
i
n
g
p
r
o
c
e
s
s
,
w
e
te
s
t t
h
e
p
r
o
to
t
y
p
e
u
s
i
n
g
th
e
d
a
ta
s
e
ts
f
r
o
m
th
e
U
C
I
r
e
p
o
s
ito
r
y
.
T
o
in
it
ia
liz
e
t
h
e
pr
ot
ot
y
pe
f
or
t
h
e
t
e
s
t
,
e
x
e
c
u
t
i
on
pr
of
i
l
e
s
i
n
t
h
e
P
r
of
i
l
e
M
a
na
g
e
r
m
u
s
t
f
i
r
s
t
be
popu
l
a
t
e
d.
F
or
t
h
i
s
p
u
r
pos
e
,
24
d
at
as
et
s
f
o
r
cl
as
s
i
f
i
cat
i
o
n
f
r
o
m
t
h
e U
C
I
r
ep
o
s
i
t
o
r
y
ar
e
u
s
ed
t
o
t
r
ai
n
t
h
e
f
i
v
e s
u
p
p
o
r
t
ed
m
ach
i
n
e
l
ear
n
i
n
g
m
ode
l
s
w
i
t
h
v
a
r
y
i
ng
c
o
m
bi
n
a
t
i
on
s
of
hy
pe
r
pa
r
a
m
e
t
e
r
s
,
r
e
s
u
l
t
i
ng
i
n
a
ppr
ox
i
m
a
t
e
l
y
4000 pr
of
i
l
e
s
.
A
n
o
t
h
er
3
1
d
at
as
et
s
ar
e t
h
en
u
s
ed
t
o
t
es
t
t
h
e p
r
o
t
o
t
y
p
e.
T
h
es
e 3
1
d
at
as
et
s
s
i
m
u
la
te
u
n
s
e
e
n
d
a
ta
s
e
ts
f
r
o
m
u
s
e
r
s
,
e
a
c
h o
f
w
h
i
c
h
w
i
l
l
go
t
hr
o
ug
h
K
-
M
e
a
n
c
lu
s
te
r
in
g
to
f
i
n
d
t
h
e
p
r
o
f
ile
c
l
u
s
te
r
w
it
h
th
e
m
i
n
i
m
u
m
E
u
cl
i
d
ean
d
i
s
t
a
n
ce.
T
h
en
,
t
h
e
p
r
o
f
i
l
e
w
i
t
h
t
h
e
h
i
g
h
es
t
accu
r
ac
y
i
n
t
h
e cl
u
s
t
er
o
f
eac
h
cl
as
s
i
f
i
ca
t
i
o
n
m
o
d
el
a
s
m
e
n
tio
n
e
d
e
a
r
lie
r
w
i
l
l
b
e s
el
ec
t
ed
.
T
h
e f
i
n
al
s
el
ect
i
o
n
i
s
t
h
e c
l
as
s
i
f
i
cat
i
o
n
m
o
d
el
an
d
i
t
s
h
y
p
er
p
ar
am
et
er
s
w
i
t
h
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
20
88
-
8708
I
nt
J
E
l
e
c
&
C
o
m
p
E
ng,
V
o
l.
8
, N
o
.
5
,
O
c
t
o
be
r
201
8
:
2
994
-
3002
3000
th
e
h
i
g
h
e
s
t a
c
c
u
r
a
c
y
a
m
o
n
g
t
h
e
f
i
v
e
m
o
d
e
ls
.
D
u
r
in
g
th
is
te
s
t,
th
e
ti
m
e
ta
k
e
n
f
o
r
th
e
s
y
s
te
m
to
s
e
le
c
t a
m
o
d
e
l
a
n
d hy
pe
r
pa
r
a
m
e
t
e
r
s
t
h
r
ough
K
-
M
ea
n
cl
u
s
t
er
i
n
g
o
f
e
x
e
c
u
tio
n
p
r
o
f
ile
s
i
s
n
e
g
li
g
ib
le
.
F
o
r
each
o
f
t
h
e
3
1
d
at
as
et
s
,
w
e r
ep
o
r
t
i
n
T
ab
l
e 4
t
h
e s
el
ect
e
d
m
o
d
el
,
i
t
s
acc
u
r
ac
y
f
r
o
m
t
h
e
ex
ec
u
t
i
o
n
p
r
o
f
i
l
e
,
a
nd
t
he
a
c
c
ur
a
c
y
o
f
t
he
m
o
d
e
l
w
he
n t
a
ki
n
g t
he
d
a
t
a
s
e
t
i
n q
ue
s
t
i
o
n a
s
t
he
ne
w
i
n
p
ut
.
I
n a
d
d
i
t
i
o
n,
w
e
a
l
s
o
r
un e
a
c
h
o
f
t
h
e 3
1
d
at
as
et
s
t
h
r
o
u
g
h
t
h
e o
t
h
er
f
o
u
r
cl
as
s
i
f
i
cat
i
o
n
m
o
d
el
s
f
o
r
co
m
p
ar
i
s
o
n
.
T
h
e l
as
t
co
l
u
m
n
i
n
T
ab
l
e 4
i
n
d
i
cat
es
w
h
et
h
er
t
h
e s
el
ect
ed
m
o
d
el
act
u
al
l
y
p
er
f
o
r
m
s
b
e
s
t
a
m
o
n
g
t
h
e
f
i
v
e
m
o
d
el
s
;
t
h
e b
es
t
m
o
d
el
an
d
accu
r
ac
y
ar
e o
t
h
er
w
i
s
e s
u
p
p
l
i
ed
.
F
r
o
m
T
ab
l
e
4,
t
h
e
pr
ot
ot
y
pe
c
a
n
s
e
l
e
c
t
t
h
e
be
s
t
c
l
a
s
s
f
i
c
i
a
t
i
on
m
ode
l
f
or
20 ou
t
of
3
1 da
t
a
s
e
t
s
.
A
m
o
n
g
t
h
es
e 2
0
d
at
as
et
s
,
t
h
e act
u
al
acc
u
r
ac
y
i
s
at
l
eas
t
5
p
er
cen
t
s
w
o
r
s
e
t
h
a
n
p
r
ev
i
o
u
s
e
x
ecu
t
i
o
n
s
i
n
8
d
at
as
et
s
(
h
i
g
h
l
i
g
h
t
ed
i
n
b
o
l
d
t
y
p
e)
.
T
h
er
e i
s
o
n
e
d
at
as
et
,
s
e
ns
or
_r
e
adi
ngs
_24
,
w
h
er
e t
h
e act
u
al
accu
r
ac
y
i
s
h
i
g
h
er
.
T
h
i
s
s
acr
i
f
i
ce o
f
acc
u
r
ac
y
i
s
e
x
p
ect
ed
b
ecau
s
e o
u
r
ap
p
r
o
ach
d
o
es
n
o
t
p
er
f
o
r
m
f
i
n
e
-
t
u
ni
n
g o
f
t
he
h
y
p
er
p
ar
a
m
et
er
s
a
n
d
f
eat
u
r
e s
el
ect
i
o
n
t
o
ad
j
u
s
t
t
h
e
m
o
d
el
t
o
n
e
w
d
at
a
s
et
s
i
n
o
r
d
er
t
o
av
o
i
d
l
at
en
c
y
i
n
au
t
o
m
at
i
n
g
m
o
d
e
l s
e
le
c
tio
n
.
A
s
f
o
r
t
h
e o
t
h
er
1
1
d
at
as
et
s
f
o
r
w
h
i
ch
t
h
e p
r
o
t
o
t
y
p
e d
o
es
n
o
t
s
el
ect
t
h
e b
es
t
m
o
d
el
,
t
h
e act
u
al
accu
r
ac
y
v
al
u
es
ar
e
w
o
r
s
e t
o
h
i
g
h
er
d
eg
r
ees
.
W
e at
t
r
i
b
u
t
e t
h
i
s
m
ai
n
l
y
t
o
t
h
e
s
m
al
l
s
i
ze o
f
ex
ecu
t
i
o
n
p
r
o
f
i
l
e
s
,
w
h
ic
h
li
m
its
t
h
e
p
o
s
s
ib
le
c
o
m
b
i
n
at
i
o
n
s
o
f
h
y
p
er
p
ar
a
m
et
er
s
.
A
n
o
t
h
er
p
o
s
s
i
b
l
e r
eas
o
n
i
s
t
h
at
t
h
e acc
u
r
ac
y
i
s
af
f
ec
t
ed
b
y
t
h
e d
at
as
et
s
t
h
e
m
s
el
v
es
b
eca
u
s
e
t
h
e
s
y
s
t
e
m
al
l
o
w
s
n
o
n
-
t
ech
n
i
cal
u
s
er
s
t
o
u
p
l
o
ad
d
at
as
et
s
w
i
t
h
o
u
t
cl
ean
i
n
g
t
h
e
m
.
A
s
w
e
p
r
o
b
e t
h
e d
at
as
et
s
w
i
t
h
l
o
w
accu
r
ac
y
,
i
t
i
s
f
o
u
n
d
t
ha
t
t
he
W
hol
e
s
al
e
c
us
t
om
e
r
s
dat
a
-
R
e
gi
on
d
at
as
et
co
n
t
ai
n
s
o
n
l
y
4
3
9
r
eco
r
d
s
,
w
h
i
c
h
ar
e i
n
s
u
f
f
i
ci
en
t
t
o
t
r
ai
n
an
acc
u
r
at
e
m
o
d
el
.
T
h
e
ta
e
d
at
as
et
c
on
t
a
i
ns
on
l
y
150 r
e
c
or
ds
a
n
d t
h
e
S
P
E
C
T
F
H
e
a
r
t
-
tr
a
in
d
a
t
a
s
e
t
c
o
nt
a
i
ns
o
nl
y 7
9
r
e
c
o
r
d
s
o
f
b
i
na
r
y va
l
ue
s
.
T
he
M
O
N
K’
s
P
ro
b
l
e
m
s
da
t
a
s
e
t
c
on
t
a
i
n
s
on
l
y
414 r
e
c
or
ds
w
i
t
h a
c
ol
um
n of
de
s
c
r
i
pt
i
v
e
s
t
r
i
ng
s
.
T
h
e
tu
r
k
iy
e
-
s
t
ude
nt
-
e
v
al
uat
i
o
n_ge
ne
r
i
c
d
at
as
et
co
n
t
ai
n
s
5
-
p
o
i
n
t
L
i
k
er
t
S
cal
e o
r
d
i
n
al
d
at
a,
w
h
i
ch
a
r
e n
o
t
y
et
h
a
n
d
l
ed
p
r
o
p
er
l
y
b
y
t
h
e D
at
a S
ca
n
n
er
.
T
ab
l
e
4
.
E
va
l
ua
t
i
o
n
Re
su
l
t
D
at
as
e
t
N
am
e
S
el
ec
t
ed
M
o
d
el
P
ro
fi
l
e
A
ccu
r
acy
A
ccu
r
acy
W
i
t
h
N
ew
D
at
as
e
t
D
i
ffe
re
n
c
e
B
e
s
t
o
f
F
i
ve
B
r
e
a
st
T
i
ss
u
e
N
a
ï
v
e B
a
y
es
0
.
7
6
3
0
.
6
5
6
-
0
.
1
07
Y
es
E
c
o
li
N
a
ï
v
e B
a
y
es
0
.
8
4
2
0
.
8
3
0
-
0
.
0
12
Y
es
L
u
n
g C
a
nc
e
r
N
a
ï
v
e B
a
y
es
0
.
6
4
4
0
.
5
8
0
-
0
.
0
64
Y
es
M
ONK
’
s
P
r
o
b
l
e
m
s
N
a
ï
v
e B
a
y
es
0
.
6
3
2
0
.
6
2
6
-
0
.
0
06
Y
es
P
a
r
k
i
n
s
o
n
s
D
ec
i
s
i
o
n
T
r
ee
0
.
8
7
7
0
.
8
2
0
-
0
.
0
57
Y
es
SPE
C
T
F
H
e
a
r
t
2
-
t
r
a
in
N
a
ï
v
e B
a
y
es
0
.
8
4
0
0
.
7
5
8
-
0
.
0
82
Y
es
S
t
at
l
o
g
(
G
e
r
m
an
C
r
e
d
i
t
D
at
a)
-
n
u
m
er
i
c
N
a
ï
v
e B
a
y
es
0
.
7
4
4
0
.
7
2
3
-
0
.
0
21
Y
es
S
t
a
t
l
og
(
G
e
r
m
a
n
C
r
e
d
i
t
D
at
a)
N
a
ï
v
e B
a
y
es
0
.
7
1
6
0
.
7
3
1
+
0
.
0
1
5
Y
es
S
ta
tl
o
g
(
H
e
a
r
t)
N
a
ï
v
e B
a
y
es
0
.
8
4
6
0
.
8
3
6
-
0
.
0
1
0
Y
es
S
ta
tl
o
g
(
V
e
h
ic
l
e
S
il
h
o
u
e
tte
s
)
D
ec
i
s
i
o
n
T
r
ee
0
.
7
1
4
0
.
6
8
0
-
0
.
0
34
Y
es
U
r
b
a
n
L
a
n
d
C
ov
e
r
-
tr
a
in
i
n
g
N
a
ï
v
e B
a
y
es
0
.
8
2
1
0
.
7
6
5
-
0
.
0
56
Y
es
W
h
o
l
es
a
l
e c
u
s
t
o
m
er
s
d
at
a
-
C
ha
n
a
l
N
a
ï
v
e B
a
y
es
0
.
9
0
5
0
.
9
0
2
-
0
.
0
03
Y
es
W
il
t
-
t
r
a
in
in
g
KN
N
0
.
9
8
5
0
.
9
9
0
+
0
.
0
0
5
Y
es
Y
ea
s
t
N
a
ï
v
e B
a
y
es
0
.
5
7
7
0
.
5
1
1
-
0
.
0
66
Y
es
Zo
o
N
a
ï
v
e B
a
y
es
0
.
9
6
8
0
.
9
5
0
-
0
.
0
18
Y
es
l
ea
f
N
a
ï
v
e B
a
y
es
0
.
7
7
5
0
.
7
1
6
-
0
.
0
59
Y
es
p
e
n
d
i
g
i
t
s
-
t
ra
i
n
KN
N
0
.
9
8
6
0
.
9
9
0
+
0
.
0
0
4
Y
es
s
e
n
s
o
r
_
r
e
a
d
i
n
g
s
_
2
4
D
ec
i
s
i
o
n
T
r
ee
0
.
8
7
7
0
.
9
6
0
+
0
.
0
8
3
Y
es
t
ae
D
ec
i
s
i
o
n
T
r
ee
0
.
6
4
0
0
.
6
6
0
+
0
.0
2
0
Y
es
v
e
r
t
e
b
r
a
l
_
c
o
l
u
m
n
_
da
t
a
_
3
C
KN
N
0
.
9
8
5
0
.
8
4
0
-
0
.
1
45
Y
es
Ir
i
s
D
ec
i
s
i
o
n
T
r
ee
0
.
9
9
3
0
.
9
5
0
-
0
.
0
43
Ne
u
r
a
l
Ne
t
wo
r
k
(
0
.
9
70
,
-
0
.
0
2
3
)
L
en
s
es
N
a
ï
v
e B
a
y
es
0
.
9
5
0
0
.
8
5
0
-
0
.1
0
0
D
ec
i
s
i
o
n
T
r
ee
(
0
.
8
8
0
,
-
0
.
0
7
)
M
A
G
I
C
G
am
m
a T
e
l
e
s
co
p
e
N
a
ï
v
e B
a
y
es
0
.
7
6
3
0
.
7
2
7
-
0
.
0
36
Ne
u
r
a
l
Ne
t
wo
r
k
(
0
.
8
22
,
+
0.
0
59
)
SPE
C
T
F
H
e
a
r
t
-
tr
a
in
N
a
ï
v
e B
a
y
es
0
.
8
0
5
0
.
6
8
5
-
0
.
1
2
0
S
VM
(
0
.
74
0
,
-
0
.
0
6
5
)
W
a
ve
f
o
r
m
D
ec
i
s
i
o
n
T
r
ee
0
.
8
7
7
0
.
7
5
-
0
.
1
27
K
NN
&
NN
(
0
.
8
5
0
,
-
0
.
0
27
)
W
h
o
l
es
a
l
e c
u
s
t
o
m
er
s
d
at
a
-
R
e
g
i
on
N
a
ï
v
e B
a
y
es
0
.
9
0
5
0
.
4
7
8
-
0
.
4
27
K
NN
&
S
VM
(
0
.
7
2
0
,
-
0
.
1
8
5
)
s
e
n
s
o
r
_
r
e
a
d
i
n
g
s
_
4
KN
N
0
.
9
8
5
0
.
9
6
-
0
.
0
25
D
ec
i
s
i
o
n
T
r
ee
(
0
.
9
9
0
,
+
0
.
0
0
5
)
s
t
u
d
e
n
t
-
ma
t
S
VM
0
.
9
3
9
0
.
9
4
+
0
.
0
0
1
KN
N
(
0
.
9
50
,
+
0.
01
1
)
s
t
u
d
e
n
t
-
p
o
r
S
VM
0
.
9
3
9
0
.
8
8
-
0
.
0
59
KN
N
(
0
.
8
9
0
,
-
0
.
0
4
9
)
tu
r
k
iy
e
-
s
t
u
d
e
n
t
-
ev
a
l
u
at
i
o
n
_
g
en
er
i
c
KN
N
1
.
0
0
0
.
3
2
-
0
.
6
8
0
S
VM
(
0
.
37
0
,
-
0
.
6
3
0
)
w
i
n
e
q
u
a
l
i
t
y
-
w
h
i
te
KN
N
0
.
6
3
2
0
.
4
5
-
0
.
1
82
D
ec
i
s
i
o
n
T
r
ee
(
0
.
6
3
0
,
-
0
.
0
02
)
5.
CO
NCL
U
S
I
O
N
T
h
i
s
p
ap
er
p
r
o
p
o
s
es
an
ar
ch
i
t
ect
u
r
e f
o
r
t
h
e a
u
t
o
m
at
i
o
n
o
f
m
ach
i
n
e l
ear
n
i
n
g
p
r
o
ces
s
s
o
t
h
at
n
o
n
-
t
ech
n
i
cal
u
s
er
s
ca
n
p
er
f
o
r
m
d
at
a cl
as
s
i
f
i
cat
i
o
n
t
a
s
k
.
U
n
l
i
k
e
ex
i
s
t
i
n
g
s
y
s
t
e
m
s
,
w
h
i
c
h
f
o
cu
s
m
o
s
t
l
y
o
n
m
o
d
el
s
el
ect
i
o
n
a
n
d
h
y
p
er
p
ar
a
m
et
er
t
u
n
i
n
g
,
w
e al
s
o
al
l
o
w
u
s
e
r
s
t
o
u
pl
oa
d da
t
a
w
i
t
h
ou
t
c
l
e
a
n
i
ng a
n
d pr
e
pr
oc
e
s
s
i
n
g
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
E
l
e
c
&
C
o
m
p
E
ng
I
S
S
N
:
2088
-
8708
A
n
A
r
c
h
i
te
c
tu
r
e
fo
r
S
i
m
pl
i
f
i
e
d a
nd A
ut
om
a
t
e
d
M
a
c
hi
ne
…
(
J
i
t
t
a
poo P
ool
w
an
)
3001
th
e
m
f
ir
s
t.
F
o
r
th
is
r
e
q
u
ir
e
m
e
n
t,
a
d
a
ta
s
e
t p
r
o
f
ili
n
g
m
e
t
h
o
d
is
p
r
o
p
o
s
e
d
s
o
th
a
t it is
p
o
s
s
ib
le
to
f
in
d
s
i
m
ila
r
d
a
t
a
s
e
t
s
t
hr
o
ug
h K
-
M
ea
n
cl
u
s
t
er
i
n
g
f
r
o
m
p
r
ev
i
o
u
s
e
x
ecu
t
i
o
n
s
a
n
d
s
el
ect
t
h
e cl
as
s
i
f
i
ca
t
i
o
n
m
o
d
e
l a
n
d
its
h
y
p
er
p
ar
a
m
et
er
s
acco
r
d
i
n
g
l
y
.
W
i
t
h
o
u
r
ap
p
r
o
ach
,
a u
s
er
can
s
i
m
p
l
y
u
p
l
o
ad
t
r
ai
n
i
n
g
d
at
a,
ch
o
o
s
es
a
m
o
n
g
t
h
e
cl
as
s
i
f
i
cat
i
o
n
m
o
d
el
s
t
h
at
h
a
v
e b
een
d
et
er
m
i
n
ed
b
y
t
h
e
s
y
s
t
e
m
,
a
n
d
t
h
en
p
er
f
o
r
m
s
t
h
e d
at
a cl
as
s
i
f
i
ca
t
i
o
n
t
as
k
o
n t
he
i
r
f
ut
ur
e
d
a
t
a
t
hr
o
u
g
h
w
e
b br
ow
s
e
r
.
T
h
e
e
v
a
l
u
a
t
i
on ba
s
e
d on
o
ur
pr
ot
ot
y
pe
r
unn
i
n
g on
A
m
a
z
on
E
C
2
s
h
o
w
s
t
h
e
f
e
a
s
i
bi
l
i
t
y
o
f
t
h
e
ap
p
r
o
ach
an
d
t
h
at
i
t
can
s
el
e
ct
cl
as
s
i
f
i
cat
i
o
n
m
o
d
el
s
b
as
e
d
o
n
s
i
m
i
l
i
ar
d
at
as
et
s
p
r
ev
i
o
u
s
l
y
p
r
o
ces
s
ed
w
i
t
h
n
eg
l
i
g
i
b
l
e
r
es
p
o
n
s
e
t
i
m
e.
H
o
w
e
v
er
,
at
t
h
i
s
ea
r
l
y
s
ta
g
e
,
t
h
e
a
c
c
u
r
a
c
y
o
f
t
h
e
c
la
s
s
i
f
ic
a
tio
n
is
li
m
ite
d
b
y
p
r
ev
i
o
u
s
e
x
ec
u
t
i
o
n
co
n
f
i
g
u
r
at
i
o
n
s
.
A
l
s
o
,
f
o
r
u
n
cl
ean
ed
d
at
a,
t
h
e
acc
u
r
ac
y
m
a
y
b
eco
m
es
u
n
p
r
ed
i
ct
ab
l
e.
A
s
f
u
t
ur
e
w
o
r
k,
t
he
s
ys
t
e
m
w
i
l
l
p
e
r
f
o
r
m
a
ut
o
m
a
t
i
c
m
o
d
e
l
t
r
a
i
ni
n
g
us
i
ng d
i
f
f
e
r
e
nt
m
a
c
hi
ne
l
e
a
r
ni
ng
m
ode
l
s
a
n
d
v
ar
y
i
n
g
h
y
p
er
p
ar
a
m
et
er
co
m
b
i
n
at
i
o
n
s
o
n
av
ai
l
ab
l
e d
at
as
et
s
w
h
e
n
t
h
e s
y
s
t
e
m
i
s
i
d
l
e
i
n
o
r
d
er
t
o
i
m
p
r
o
v
e t
h
e
m
o
d
e
l a
n
d
h
y
p
e
r
p
a
r
a
m
e
te
r
s
e
le
c
tio
n
.
A
ls
o
,
it is
p
o
s
s
ib
le
to
in
c
lu
d
e
a
f
e
a
tu
r
e
s
e
le
c
tio
n
s
t
e
p
a
f
te
r
g
e
tti
n
g
t
h
e
m
ode
l
a
n
d hy
pe
r
pa
r
a
m
e
t
e
r
s
f
r
o
m
t
h
e p
r
ev
i
o
u
s
e
x
ecu
t
i
o
n
s
t
o
i
m
p
r
o
v
e accu
r
ac
y
.
F
i
n
a
l
l
y
,
f
o
r
t
h
e s
cal
ab
i
l
i
t
y
o
f
t
h
e s
y
s
t
e
m
,
t
h
e e
x
ecu
t
i
o
n
o
f
m
ach
i
n
e
l
ear
n
i
n
g
ca
n
b
e i
m
p
r
o
v
ed
w
i
t
h
ap
p
r
o
p
r
i
at
e r
es
o
u
r
ce al
l
o
cat
i
o
n
s
o
t
h
at
th
e
y
c
a
n
r
u
n
o
n
m
u
ltip
le
v
ir
t
u
a
l
m
a
c
h
in
e
s
in
t
h
e
c
lo
u
d
.
R
EF
ER
EN
C
ES
[
1]
M
.
A
ba
di
,
e
t a
l.
(
20
15)
.
T
e
ns
or
F
l
ow
:
L
ar
ge
-
Sc
al
e
M
ac
hi
ne
L
e
ar
ni
ng on
H
e
t
e
r
o
ge
ne
o
us
D
i
s
t
r
i
b
ut
e
d Sy
s
t
e
m
s
.
[
2]
F
.
H
ut
t
e
r
,
H
.
H
.
H
o
os
a
nd T
.
S
t
ü
t
z
l
e
,
"
A
ut
om
a
t
i
c
a
l
g
or
i
t
hm
c
onf
i
g
ur
a
t
i
on ba
s
e
d
on l
oc
a
l
s
e
a
r
c
h"
,
in
A
A
A
I
,
2
00
7,
pp.
11
52
-
1
15
7.
[
3]
J
.
S
.
Be
rg
s
t
ra
,
e
t a
l.
,
"
A
l
g
or
i
t
hm
s
f
or
hy
pe
r
-
p
ar
a
m
et
er
o
p
t
i
m
i
zat
i
o
n
"
,
A
dv
a
nc
e
s
i
n N
e
ur
al
I
nf
or
m
a
t
i
on P
r
oc
e
s
s
i
ng
S
y
st
e
ms,
v
ol
.
2
4,
pp.
2
54
6
-
2
55
4,
201
1.
[
4]
B.
K
o
m
e
r,
J
.
Be
rg
s
t
ra
a
n
d
C.
E
l
i
a
s
m
i
t
h
,
"
H
y
p
e
ro
p
t
-
s
k
l
ear
n
:
au
t
o
m
at
i
c h
y
p
er
p
ar
a
m
et
er
c
onf
i
g
ur
a
t
i
on f
or
s
c
ik
it
-
l
ear
n
"
,
in
P
r
oc
e
e
di
ngs
of
t
he
1
3t
h
P
y
t
ho
n i
n Sc
i
e
nc
e
C
onf
e
r
e
nc
e
(
Sc
i
P
y
20
14)
,
20
14
.
[
5]
M
.
F
eu
r
er
,
e
t a
l.
,
"
E
f
f
i
c
i
e
nt
a
nd r
obus
t
a
u
t
om
a
t
e
d
m
a
c
hi
n
e l
ear
n
i
n
g
"
,
A
dv
anc
e
s
i
n N
e
ur
al
I
nf
or
m
at
i
on P
r
oc
e
s
s
i
ng
Sy
s
t
e
m
s
(
N
I
P
S 201
5)
,
v
ol
.
2
8,
pp
.
29
62
-
29
70,
2
01
5.
[
6]
L
.
K
ot
t
hof
f
,
e
t a
l.
,
"
Au
t
o
-
W
E
K
A
2.
0:
A
ut
om
a
t
i
c
m
ode
l
s
e
l
e
c
t
i
on a
nd
hy
pe
r
p
ar
am
et
er
o
p
t
i
m
i
zat
i
o
n
i
n
W
E
K
A
"
,
J
our
n
al
o
f
M
ac
hi
ne
L
e
ar
n
i
n
g R
e
s
e
ar
c
h,
vo
l
.
1
8,
pp
.
1
-
5,
20
17
.
[
7]
C
.
T
hor
nt
on
,
e
t a
l.
,
"
Au
t
o
-
W
E
K
A
:
co
m
b
i
n
ed
s
el
ect
i
o
n
an
d
h
y
p
er
p
ar
a
m
et
er
o
p
t
i
m
i
zat
i
o
n
o
f
cl
as
s
i
f
i
cat
i
o
n
a
lg
o
r
it
h
ms
"
,
in
P
r
oc
e
e
di
ngs
of
t
h
e
19t
h A
C
M
SI
G
K
D
D
i
nt
e
r
nat
i
on
al
c
onf
e
r
e
nc
e
on K
n
ow
l
e
d
ge
di
s
c
ov
e
r
y
and dat
a
m
in
in
g
,
C
hi
c
a
g
o,
I
l
l
i
noi
s
,
U
S
A
,
2
013
,
p
p.
84
7
-
8
55
,
d
oi
:
1
0.
1
14
5/
2
487
57
5.
24
87
62
9.
[
8]
F
.
P
ed
r
eg
o
s
a
,
e
t a
l.
,
"
S
c
ik
it
-
le
ar
n
:
M
ach
i
n
e L
ear
n
i
n
g
i
n
P
y
t
h
o
n
"
,
J
our
n
al
o
f
M
ac
hi
ne
L
e
ar
ni
ng
R
e
s
e
ar
c
h,
v
ol
.
12,
pp.
28
25
-
2
83
0,
20
11
.
[
9]
M
.
A
bda
r
,
e
t a
l.
,
"
C
o
m
p
ar
i
n
g
P
er
f
o
r
m
an
ce o
f
D
at
a M
i
n
i
n
g
A
l
g
o
r
i
t
h
m
s
i
n
P
r
ed
i
ct
i
o
n
H
ear
t
D
i
s
eas
es
"
,
I
nt
e
r
nat
i
o
nal
J
our
n
al
o
f
E
l
e
c
t
r
i
c
al
an
d C
om
pu
t
e
r
E
ngi
ne
e
r
i
ng (
I
J
E
C
E
)
,
vo
l
.
5
,
pp
.
15
69
-
157
6,
20
15
.
[
1
0]
S
.
B
.
K
o
t
s
i
a
n
t
i
s
,
I
.
D
.
Z
ah
ar
ak
i
s
an
d
P
.
E
.
P
i
n
t
el
as
,
"
M
ach
i
n
e l
ear
n
i
n
g
:
a r
ev
i
ew
o
f
cl
as
s
i
f
i
c
at
i
o
n
an
d
co
m
bi
ni
ng
t
ech
n
i
q
u
es
"
,
Ar
tific
ia
l I
n
te
lli
g
e
n
c
e
Re
v
ie
w
,
v
ol
.
26,
pp.
15
9
-
1
90,
N
ov
e
m
be
r
01 200
6,
d
oi
:
1
0.
10
0
7/
s
10
46
2
-
00
7
-
905
2
-
3.
[
1
1]
S
.
F
.
Q
i
a
ng
G
ua
n,
"
a
ut
o
-
AI
D
:
A
da
t
a
m
i
ni
ng
f
r
a
m
e
w
or
k
f
or
a
ut
on
om
i
c
a
nom
a
l
y
i
de
nt
i
f
i
c
a
t
i
o
n i
n
ne
t
w
or
k
e
d
co
m
p
u
t
er
s
y
s
t
e
m
s
"
,
P
e
r
f
or
m
a
nc
e
C
om
put
i
n
g a
nd C
om
m
u
ni
c
at
i
ons
C
o
nf
e
r
e
nc
e
(
I
P
C
C
C
)
,
2
01
0,
d
oi
:
10.
11
09/
P
C
C
C
.
2
01
0.
56
82
33
4.
[
1
2]
S
.
W
a
n
g
,
e
t a
l.
,
"
E
a
r
l
y
de
t
e
c
t
i
on of
num
e
r
i
c
a
l
t
y
pi
ng
e
r
r
or
s
us
i
ng
da
t
a
m
i
ni
ng
t
e
c
hni
q
ue
s
"
,
I
E
E
E
T
r
ans
ac
t
i
ons
on
Sy
s
t
e
m
s
,
M
an,
an
d C
y
be
r
ne
t
i
c
s
-
P
a
rt
A
:
S
y
st
e
ms a
n
d
Hu
ma
n
s,
v
ol
.
41,
pp
.
1
19
9
-
12
12
,
2
01
1.
[
1
3]
T
he
U
C
I
M
ac
hi
ne
L
e
ar
n
i
n
g R
e
p
o
s
i
t
or
y
.
A
v
a
ila
b
le
: h
ttp
s
://a
r
c
h
iv
e
.
i
c
s
.
u
c
i.
e
d
u
/m
l/d
a
ta
s
e
ts
.
h
tm
l A
c
c
e
s
s
e
d
2
0
1
6
.
[
1
4]
R
.N
.V
.J
. M
o
h
a
n
, R
.
S
. R
a
o
a
n
d
K
.R
.S
. R
a
o
, "
E
f
f
i
c
i
e
n
t
K
-
M
ean
s
C
l
u
s
t
er
R
el
i
ab
i
l
i
t
y
o
n
T
er
n
ar
y
F
ace R
e
co
g
n
i
t
i
o
n
U
s
i
ng
A
ng
l
e
O
r
i
e
nt
e
d A
ppr
oa
c
h
"
,
I
nt
e
r
n
at
i
on
al
J
o
ur
na
l
of
E
l
e
c
t
r
i
c
al
a
nd C
om
p
ut
e
r
E
n
gi
ne
e
r
i
n
g (
I
J
E
C
E
)
,
v
ol
.
2,
201
3,
do
i
:
ht
t
p
:
/
/
dx
.
doi
.
or
g
/
1
0.
1
1
591
/
i
j
-
ic
t.
v
2
i
1.
17
79.
[
1
5]
B
.
S
.
C
ha
n
da
na
,
K
.
S
r
i
ni
v
a
s
a
nd R
.
K
.
K
um
a
r
,
"
C
l
us
t
e
r
i
ng
A
l
g
or
i
t
hm
C
o
m
bi
ne
d w
i
t
h H
i
l
l
C
l
i
m
bi
ng
f
or
C
l
a
s
s
i
f
i
c
a
t
i
on of
R
e
m
ot
e
S
e
ns
i
ng I
m
a
ge
"
,
I
nt
e
r
nat
i
on
al
J
our
nal
of
E
l
e
c
t
r
i
c
al
and C
om
put
e
r
E
n
gi
ne
e
r
i
ng (
I
J
E
C
E
)
,
v
ol
.
4,
p
p.
92
3
-
9
30,
2
01
4,
doi
:
ht
t
p:
/
/
d
x
.
doi
.
or
g
/
1
0.
11
59
1/
i
j
e
c
e
.
v
4i
6.
6
60
8.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
20
88
-
8708
I
nt
J
E
l
e
c
&
C
o
m
p
E
ng,
V
o
l.
8
, N
o
.
5
,
O
c
t
o
be
r
201
8
:
2
994
-
3002
3002
B
I
O
G
RAP
H
I
ES
O
F
AUT
H
O
RS
J
i
t
t
a
poo
P
ool
w
a
n r
e
c
e
i
v
e
d hi
s
m
a
s
t
e
r
’
s
de
g
r
e
e
i
n I
nf
or
m
a
t
i
on T
e
c
hnol
og
y
f
r
o
m
K
i
ng
M
ong
k
ut
'
s
U
ni
v
e
r
s
i
t
y
of
T
e
c
hnol
og
y
N
or
t
h B
a
ng
k
ok
,
i
n 200
8.
H
e
i
s
c
ur
r
e
nt
l
y
pur
s
ui
ng
P
h.
D
.
de
g
r
e
e
i
n
I
n
f
o
r
m
at
i
o
n
T
ech
n
o
l
o
g
y
at
t
h
e s
a
m
e i
n
s
t
i
t
u
t
e.
H
i
s
r
es
ear
ch
i
n
t
er
e
s
t
s
ar
e d
at
a
m
i
n
i
n
g
,
m
a
ch
i
n
e
le
a
r
n
in
g
a
n
d
o
n
lin
e
i
n
te
llig
e
n
t f
o
r
e
c
a
s
tin
g
s
y
s
t
e
m
.
S
uc
ha
S
m
a
nc
ha
t
i
s
c
ur
r
e
nt
l
y
a
l
e
c
t
ur
e
r
a
t
t
he
F
a
c
ul
t
y
of
I
n
f
or
m
a
t
i
on T
e
c
hnol
og
y
,
K
i
ng
Mo
n
g
k
ut
’
s
U
n
i
v
e
r
s
i
t
y
of
T
e
c
hno
l
og
y
N
or
t
h B
a
ng
k
ok
,
T
ha
i
l
a
nd.
H
e
obt
a
i
ne
d h
i
s
P
hD
de
g
r
e
e
a
t
M
ona
s
h U
n
i
v
e
r
s
i
t
y
i
n M
e
l
bour
ne
,
A
us
t
r
a
l
i
a
.
D
ur
i
ng
hi
s
s
t
udy
,
he
w
a
s
i
nv
ol
v
e
d
w
i
t
h t
he
de
v
e
l
opm
e
nt
of
a
pr
ot
ot
y
pe
s
c
he
dul
e
r
f
or
N
i
m
r
od/
K
s
y
s
t
e
m
.
H
i
s
c
ur
r
e
nt
r
e
s
e
a
r
c
h i
nt
e
r
es
t
s
ar
e i
n
c
l
oud
w
or
kf
l
ow
s
c
he
dul
i
ng
t
e
c
hni
que
s
,
a
n
d m
a
c
hi
ne
l
e
a
r
ni
ng
a
nd
a
l
g
or
i
t
hm
s
f
or
bi
oi
nf
or
m
a
t
i
c
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.