I
nte
rna
t
io
na
l J
o
urna
l o
f
Adv
a
nces in Applie
d Science
s
(
I
J
AAS)
Vo
l.
9
,
No
.
2
,
J
u
n
e
2
0
2
0
,
p
p
.
93
~
1
0
0
I
SS
N:
2
2
5
2
-
8
8
1
4
,
DOI
: 1
0
.
1
1
5
9
1
/i
j
aa
s
.
v
9
.
i2
.
p
p
93
-
100
93
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
a
s
.
ia
esco
r
e.
co
m
A
c
hi
-
sq
ua
re
-
SVM
ba
sed peda
g
o
g
ica
l rule
ex
t
ra
ctio
n
m
ethod
for
m
icro
a
rray
d
a
ta a
na
ly
sis
M
uk
hta
r
Da
m
o
la
Sa
la
w
u
1
,
M
ichea
l O
l
a
o
lu Ar
o
w
o
lo
2
,
S
ula
i
m
a
n O
la
niy
i A
bd
uls
a
la
m
3
,
Ra
f
iu M
o
pe
I
s
ia
k
a
4
,
B
il
k
i
s
u
J
i
m
a
da
-
O
j
uo
la
pe
5
,
M
ud
a
s
hiru
L
a
t
ee
f
O
lu
m
ide
6
,
K
a
ze
e
m A.
G
bo
la
g
a
de
7
1
,
2,
3,
4
,7
De
p
a
rtm
e
n
t
o
f
Co
m
p
u
ter S
c
ien
c
e
,
Co
ll
e
g
e
o
f
In
f
o
rm
a
ti
o
n
a
n
d
Co
m
m
u
n
ica
ti
o
n
S
c
ie
n
c
e
,
Kw
a
ra
S
tate
Un
iv
e
rsit
y
,
Nig
e
ria
5
De
p
a
rtme
n
t
o
f
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
a
n
d
C
o
m
p
u
ter E
n
g
in
e
e
rin
g
,
Kw
a
ra
S
tate
Un
iv
e
rsit
y
,
Nig
e
ria
6
S
c
h
o
o
l
o
f
Co
m
p
u
ter S
c
ien
c
e
,
U
n
iv
e
rsiti
S
a
in
s M
a
lay
sia
,
M
a
la
y
sia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Dec
15
,
2
0
1
9
R
ev
i
s
ed
Feb
9
,
2
0
20
A
cc
ep
ted
Mar
1
4
,
2
0
20
S
u
p
p
o
rt
V
e
c
to
r
M
a
c
h
i
n
e
(S
VM)
is
c
u
rre
n
t
ly
a
n
e
ff
icie
n
t
c
las
sif
ic
a
ti
o
n
tec
h
n
iq
u
e
d
u
e
t
o
it
s a
b
il
it
y
to
c
a
p
t
u
re
n
o
n
li
n
e
a
rit
ies
in
d
iag
n
o
stic sy
ste
m
s,
b
u
t
it
d
o
e
s
n
o
t
re
v
e
a
l
th
e
k
n
o
w
led
g
e
lea
rn
t
d
u
ri
n
g
trai
n
in
g
.
It
is
im
p
o
rtan
t
t
o
u
n
d
e
rsta
n
d
o
f
h
o
w
a
d
e
c
isio
n
is
r
e
a
c
h
e
d
in
th
e
m
a
c
h
in
e
lea
rn
in
g
tec
h
n
o
l
o
g
y
,
su
c
h
a
s
b
io
i
n
f
o
rm
a
ti
c
s.
On
th
e
o
th
e
r
h
a
n
d
,
a
d
e
c
isio
n
tree
h
a
s
g
o
o
d
c
o
m
p
re
h
e
n
sib
il
it
y
;
th
e
p
ro
c
e
ss
o
f
c
o
n
v
e
rti
n
g
su
c
h
in
c
o
m
p
re
h
e
n
sib
le
m
o
d
e
ls
in
to
a
n
u
n
d
e
rsta
n
d
a
b
le
m
o
d
e
l
is
o
f
ten
re
g
a
rd
e
d
a
s
ru
le
e
x
trac
ti
o
n
.
In
th
is
p
a
p
e
r
w
e
p
ro
p
o
se
d
a
n
a
p
p
ro
a
c
h
f
o
r
e
x
trac
ti
n
g
ru
les
f
ro
m
S
V
M
f
o
r
m
icro
a
rra
y
d
a
tas
e
t
b
y
c
o
m
b
in
in
g
th
e
m
e
rit
s
o
f
b
o
th
th
e
S
V
M
a
n
d
d
e
c
isio
n
tree
.
T
h
e
p
ro
p
o
se
d
a
p
p
ro
a
c
h
c
o
n
sists
o
f
th
re
e
ste
p
s
;
th
e
S
VM
-
CHI
-
S
QU
A
RE
is
e
m
p
lo
y
e
d
to
re
d
u
c
e
th
e
f
e
a
tu
re
se
t.
D
a
tas
e
t
with
re
d
u
c
e
d
f
e
a
tu
re
s
is
u
se
d
to
o
b
tain
S
V
M
m
o
d
e
l
a
n
d
s
y
n
th
e
ti
c
d
a
ta
is
g
e
n
e
ra
ted
.
Clas
sif
ic
a
ti
o
n
a
n
d
Re
g
re
ss
io
n
T
re
e
(CA
R
T
)
is
u
se
d
to
g
e
n
e
ra
te
Ru
le
s
a
s
th
e
L
a
st
p
h
a
se
.
W
e
u
se
b
re
a
st
m
a
ss
e
s
d
a
tas
e
t
f
ro
m
UCI
re
p
o
sito
ry
w
h
e
re
c
o
m
p
re
h
e
n
sib
il
it
y
is
a
k
e
y
re
q
u
irem
e
n
t.
F
ro
m
th
e
re
su
lt
o
f
th
e
e
x
p
e
rim
e
n
t
a
s
th
e
re
d
u
c
e
d
f
e
a
tu
re
d
a
tas
e
t
is
u
se
d
,
th
e
p
ro
p
o
se
d
a
p
p
r
o
a
c
h
e
x
trac
ts
sm
a
ll
e
r
len
g
th
ru
les
,
th
e
re
b
y
im
p
ro
v
in
g
th
e
c
o
m
p
re
h
e
n
sib
il
it
y
o
f
th
e
sy
ste
m
.
W
e
o
b
tain
e
d
a
c
c
u
ra
c
y
o
f
9
3
.
5
3
%
,
se
n
siti
v
it
y
o
f
8
9
.
5
8
%
,
sp
e
c
if
icity
o
f
9
6
.
7
0
%
,
a
n
d
train
i
n
g
ti
m
e
o
f
3
.
1
9
5
se
c
o
n
d
s.
A
c
o
m
p
a
ra
ti
v
e
a
n
a
l
y
sis i
s c
a
rried
o
u
t
d
o
n
e
w
it
h
o
th
e
r
a
lg
o
rit
h
m
s.
K
ey
w
o
r
d
s
:
Ma
ch
i
n
e
lear
n
i
n
g
,
Me
d
ical
d
iag
n
o
s
i
s
,
R
u
le
-
e
x
tr
ac
tio
n
,
SVMs
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
:
Mic
h
ea
l O
lao
l
u
A
r
o
w
o
lo
,
Dep
ar
t
m
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
,
K
w
ar
a
Sta
te
Un
i
v
er
s
i
t
y
,
Ma
lete,
Nig
er
ia
.
E
m
ail:
ar
o
w
o
lo
.
o
lao
lu
@
g
m
a
il
.
co
m
1.
I
NT
RO
D
UCT
I
O
N
Ov
er
th
e
p
as
t
t
w
o
d
ec
ad
es,
th
e
w
o
r
ld
h
as
w
it
n
ess
ed
a
tr
u
e
e
x
p
lo
s
io
n
o
f
d
ata,
w
h
ic
h
h
a
s
m
ain
l
y
b
ee
n
d
r
iv
en
b
y
in
n
o
v
at
iv
e
s
to
r
ag
e
t
ec
h
n
o
lo
g
y
a
n
d
t
h
e
in
cr
ea
s
in
g
p
o
p
u
lar
it
y
o
f
t
h
e
I
n
ter
n
et.
T
o
d
ay
,
a
h
u
g
e
a
m
o
u
n
t
o
f
d
ata
is
g
en
er
ated
i
n
th
e
m
e
d
ical
d
o
m
ai
n
.
A
p
o
p
u
lar
s
o
u
r
ce
is
m
icr
o
ar
r
ay
d
ata.
Mic
r
o
ar
r
ay
i
s
a
b
io
lo
g
ica
l
p
latf
o
r
m
f
o
r
g
ath
er
i
n
g
g
en
e
ex
p
r
ess
io
n
s
[
1
]
.
T
h
e
tech
n
o
lo
g
y
h
elp
s
in
th
e
id
en
t
if
ica
tio
n
o
f
n
e
w
g
e
n
es.
T
h
e
y
h
elp
to
k
n
o
w
ab
o
u
t
th
e
f
u
n
c
tio
n
i
n
g
an
d
ex
p
r
es
s
io
n
le
v
els
u
n
d
er
d
if
f
er
e
n
t
co
n
d
itio
n
s
.
I
t
al
s
o
h
elp
s
to
lear
n
m
o
r
e
ab
o
u
t
d
if
f
er
e
n
t
d
is
ea
s
e
s
s
u
ch
a
s
h
ea
r
t
d
is
ea
s
e,
m
e
n
tal
ill
n
es
s
,
in
f
ec
tio
u
s
d
is
ea
s
e
a
n
d
m
o
s
t
i
m
p
o
r
tan
tl
y
t
h
e
s
t
u
d
y
o
f
ca
n
ce
r
.
Dif
f
er
e
n
t
t
y
p
es
o
f
ca
n
ce
r
h
av
e
b
ee
n
class
if
ied
o
n
th
e
p
r
e
m
is
e
o
f
th
e
o
r
g
a
n
s
i
n
w
h
ic
h
th
e
tu
m
o
r
s
d
ev
elo
p
.
W
it
h
th
e
h
elp
o
f
m
i
cr
o
ar
r
ay
tech
n
o
lo
g
y
,
it
w
i
ll
b
e
p
o
s
s
ib
le
f
o
r
th
e
r
esear
ch
er
s
to
f
u
r
th
er
clas
s
i
f
y
th
e
t
y
p
e
s
o
f
ca
n
ce
r
o
n
t
h
e
b
asi
s
o
f
th
e
p
atter
n
s
o
f
g
e
n
e
ac
ti
v
it
y
in
t
h
e
tu
m
o
r
ce
lls
[
2
,
3
]
.
Mic
r
o
ar
r
ay
is
a
p
o
p
u
lar
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
: 2
2
5
2
-
8814
I
n
t J
A
d
v
A
p
p
l Sci
,
Vo
l.
9
,
No
.
2
,
J
u
n
e
2
0
2
0
:
93
–
1
0
0
94
s
o
u
r
ce
o
f
d
ata.
T
h
e
in
tr
in
s
ic
p
r
o
b
lem
o
f
a
t
y
p
ical
d
ata
s
et
p
r
o
d
u
ce
d
b
y
m
icr
o
ar
r
ay
s
is
t
h
e
s
a
m
p
le
s
ize
an
d
th
e
h
i
g
h
d
i
m
en
s
io
n
a
lit
y
o
f
th
e
d
ata
s
et
th
u
s
m
a
k
i
n
g
an
al
y
s
is
d
if
f
icu
l
t.
I
n
ad
d
itio
n
th
e
co
m
p
licated
r
elatio
n
s
a
m
o
n
g
th
e
d
if
f
er
en
t
g
e
n
es
m
a
k
e
an
al
y
s
i
s
m
o
r
e
d
if
f
ic
u
lt
an
d
r
e
m
o
v
i
n
g
ex
ce
s
s
f
ea
t
u
r
es
ca
n
i
m
p
r
o
v
e
th
e
q
u
alit
y
o
f
th
e
r
es
u
lts
.
I
n
th
i
s
p
ap
er
,
Featu
r
e
s
elec
tio
n
u
s
i
n
g
S
VM
-
C
HI
-
SQ
U
A
R
E
al
g
o
r
ith
m
is
e
m
p
lo
y
ed
in
t
h
e
f
ir
s
t
p
h
ase
to
r
ed
u
ce
d
im
e
n
s
io
n
alit
y
o
f
t
h
e
d
ata
b
y
y
ie
ld
in
g
th
e
k
e
y
attr
i
b
u
te
in
t
h
e
d
ata.
T
h
u
s
,
f
e
w
er
n
u
m
b
er
an
d
s
m
aller
r
u
les
ar
e
o
b
tai
n
ed
r
es
u
lti
n
g
i
n
th
e
i
m
p
r
o
v
e
m
e
n
t
o
f
t
h
e
co
m
p
r
eh
en
s
ib
ilit
y
o
f
t
h
e
s
y
s
te
m
.
A
l
s
o
an
ap
p
r
o
ac
h
f
o
r
p
ed
ag
o
g
ical
r
u
le
-
ex
tr
ac
tio
n
was
p
r
o
p
o
s
ed
w
h
ich
tr
ea
t
s
th
e
cl
ass
i
f
ier
as
a
“
b
lack
b
o
x
”
an
d
d
ir
ec
tl
y
ex
tr
ac
t
r
u
les
w
h
ic
h
r
elate
t
h
e
i
n
p
u
t
an
d
t
h
e
o
u
tp
u
t
o
f
t
h
e
SV
M.
Dec
is
io
n
t
r
ee
s
(
C
AR
T
)
w
h
ic
h
h
a
v
e
a
s
ig
n
i
fi
ca
n
t
ad
v
an
ta
g
e
o
f
p
r
o
d
u
cin
g
in
ter
p
r
etab
le
r
u
les
w
er
e
u
s
ed
.
R
u
les
w
er
e
g
e
n
er
ated
b
y
in
te
g
r
ati
n
g
m
er
its
o
f
b
o
th
SVMs
an
d
d
ec
is
io
n
tr
ee
.
R
u
le
s
et
p
er
f
o
r
m
an
ce
is
th
e
n
esti
m
ated
u
s
i
n
g
th
e
m
ea
s
u
r
ed
r
ates
o
f
tr
u
e
p
o
s
itiv
es
(
T
P
s
)
an
d
f
alse
p
o
s
itiv
es
(
FP
s
)
.
Usi
n
g
t
h
is
ap
p
r
o
ac
h
,
w
e
w
il
l
s
h
o
w
b
o
t
h
a
n
i
m
p
r
o
v
ed
class
if
icatio
n
p
er
f
o
r
m
an
ce
an
d
co
m
p
r
e
h
en
s
ib
ilit
y
co
m
p
ar
ed
to
th
e
p
r
ev
io
u
s
l
y
p
r
o
p
o
s
ed
tech
n
iq
u
e
s
.
2.
M
E
T
H
O
DS
I
n
th
is
s
ec
tio
n
,
w
e
p
r
o
v
id
e
a
b
r
ief
in
tr
o
d
u
ctio
n
o
f
F
ea
tu
r
e
Sel
ec
tio
n
,
SVM,
R
u
le
ex
tr
ac
tio
n
an
d
C
A
R
T
2
.
1
.
F
ea
t
ure
s
elec
t
io
n
A
n
u
m
b
er
o
f
m
et
h
o
d
s
h
av
e
b
ee
n
p
r
o
p
o
s
ed
f
o
r
r
u
le
ex
tr
ac
tio
n
f
r
o
m
S
VM
s
.
B
r
o
ad
ly
s
p
ea
k
in
g
,
t
h
es
e
m
et
h
o
d
s
ca
n
b
e
ca
teg
o
r
ized
in
to
th
r
ee
m
ai
n
f
a
m
ilie
s
w
h
ic
h
ar
e:
p
ed
ag
o
g
ical,
d
ec
o
m
p
o
s
it
io
n
,
an
d
ec
lectic
[
4
]
.
So
m
e
o
f
t
h
e
s
e
m
e
th
o
d
s
to
d
ate
s
till
p
r
o
d
u
ce
r
elativ
el
y
lar
g
e
r
u
le
s
ets,
w
h
ic
h
li
m
it
s
th
eir
ex
p
la
n
atio
n
ca
p
ab
ilit
y
[
5
]
.
R
u
le
s
e
ts
ca
n
o
n
l
y
o
f
f
er
ex
p
la
n
atio
n
i
f
t
h
e
n
u
m
b
er
o
f
r
u
le
s
in
th
e
r
u
le
s
et
is
r
elativ
el
y
s
m
all
a
n
d
its
class
i
f
icat
io
n
ac
c
u
r
ac
y
i
s
h
i
g
h
.
Si
m
p
ler
r
u
les
also
o
f
f
er
b
e
tter
u
n
d
er
s
ta
n
d
in
g
an
d
ex
p
lan
a
tio
n
[
6
]
.
T
o
ex
tr
ac
t
m
o
r
e
co
m
p
r
e
h
e
n
s
ib
le
r
u
les,
ir
r
elev
an
t
f
ea
tu
r
es
w
h
ich
d
o
n
o
t
co
n
tr
ib
u
te
to
t
h
e
clas
s
i
f
icatio
n
d
ec
is
io
n
s
h
o
u
ld
n
o
t
b
e
in
th
e
r
u
le
an
tece
d
en
ts
.
T
h
i
s
h
ig
h
li
g
h
ts
a
r
eq
u
ir
e
m
e
n
t
to
co
n
s
id
er
f
ea
tu
r
e
s
elec
tio
n
as
an
in
te
g
r
al
p
ar
t
o
f
r
u
le
ex
tr
ac
tio
n
.
I
n
f
ea
tu
r
e
s
elec
tio
n
,
o
n
e
s
elec
ts
o
n
l
y
t
h
o
s
e
i
n
p
u
t
d
i
m
e
n
s
io
n
s
t
h
at
co
n
tai
n
t
h
e
r
elev
an
t
i
n
f
o
r
m
atio
n
f
o
r
s
o
lv
i
n
g
t
h
e
p
ar
ticu
lar
p
r
o
b
le
m
.
T
h
er
e
ar
e
th
r
ee
ca
te
g
o
r
ies
o
f
f
ea
t
u
r
e
s
ele
ctio
n
w
h
ic
h
ar
e:
f
il
ter
s
,
w
r
ap
p
er
s
,
an
d
e
m
b
ed
d
ed
tech
n
iq
u
e
s
.
T
h
is
w
o
r
k
f
o
cu
s
e
s
o
n
f
i
lte
r
-
b
ased
ap
p
r
o
ac
h
.
C
h
i
-
s
q
u
ar
e
to
b
e
s
p
ec
if
ic.
T
h
e
d
if
f
er
en
ce
b
et
w
ee
n
C
h
i
-
s
q
u
ar
e
an
d
o
th
er
m
et
h
o
d
s
an
d
th
e
r
ea
s
o
n
it
w
i
ll
b
e
u
s
ed
is
t
h
at
it
is
v
er
y
r
o
b
u
s
t
w
it
h
r
e
s
p
ec
t
t
o
d
is
tr
ib
u
tio
n
o
f
t
h
e
d
ata,
it
s
s
i
m
p
lic
it
y
o
f
c
o
m
p
u
tatio
n
,
t
h
e
d
etailed
i
n
f
o
r
m
atio
n
t
h
at
ca
n
b
e
d
er
iv
ed
f
r
o
m
th
e
te
s
t,
an
d
it
s
f
lex
ib
ilit
y
i
n
m
a
n
a
g
in
g
d
ata
f
r
o
m
b
o
th
t
w
o
g
r
o
u
p
an
d
m
u
ltip
l
e
g
r
o
u
p
s
tu
d
ie
s
.
2
.
2
.
SVM
s
T
h
e
SVM
alg
o
r
ith
m
[
7
]
is
a
class
i
fi
ca
tio
n
al
g
o
r
ith
m
t
h
at
p
r
o
d
u
ce
s
s
tate
-
of
-
t
h
e
-
ar
t
p
er
f
o
r
m
an
ce
in
a
v
ast
v
ar
iet
y
o
f
ap
p
licatio
n
d
o
m
a
in
s
,
in
cl
u
d
in
g
b
io
in
f
o
r
m
at
ics.
T
h
er
e
a
r
e
tw
o
k
e
y
r
ea
s
o
n
s
f
o
r
u
s
in
g
th
e
SVM
in
b
io
in
f
o
r
m
atics
[
8
]
.
First,
m
an
y
b
io
lo
g
ical
is
s
u
e
s
in
v
o
lv
e
h
ig
h
-
d
i
m
en
s
io
n
al,
n
o
i
s
y
d
ata.
T
h
e
SVM
is
k
n
o
w
n
to
b
eh
av
e
v
er
y
w
ell
w
it
h
t
h
ese
d
ata
co
m
p
ar
ed
to
o
th
er
s
tati
s
tic
al
o
r
m
ac
h
in
e
lear
n
i
n
g
m
et
h
o
d
s
.
Seco
n
d
,
co
n
tr
ar
y
to
m
o
s
t
m
ac
h
in
e
lear
n
i
n
g
tech
n
iq
u
e,
k
er
n
el
m
et
h
o
d
s
li
k
e
t
h
e
SVM
ca
n
ea
s
il
y
h
an
d
le
n
o
n
v
e
cto
r
in
p
u
ts
,
s
u
c
h
a
s
v
ar
iab
le
len
g
t
h
s
eq
u
e
n
ce
s
o
r
g
r
ap
h
s
[
9
]
.
Fo
r
class
i
f
icatio
n
p
r
o
b
le
m
s
SVM
f
in
d
s
a
m
a
x
i
m
al
m
ar
g
in
h
y
p
er
p
lan
e
th
a
t
d
iv
id
e
s
t
w
o
cla
s
s
e
s
.
T
h
e
m
ai
n
in
te
n
t
o
f
SV
M
is
t
o
f
in
d
an
o
p
ti
m
al
s
ep
ar
ati
n
g
h
y
p
er
p
lan
e
t
h
at
co
r
r
ec
tl
y
clas
s
if
ie
s
d
ata
p
o
in
ts
as
m
u
c
h
as
p
o
s
s
ib
le
b
y
r
ed
u
ci
n
g
th
e
r
is
k
o
f
m
is
cla
s
s
i
f
y
in
g
th
e
tr
ain
i
n
g
s
a
m
p
le
s
an
d
u
n
s
ee
n
test
s
a
m
p
le
s
.
T
o
ad
d
r
ess
w
i
th
n
o
n
-
li
n
ea
r
is
s
u
es,
SVM
f
ir
s
t
p
r
o
j
ec
ts
d
ata
i
n
to
h
i
g
h
er
d
i
m
en
s
io
n
al
f
ea
t
u
r
e
s
p
ac
e
an
d
tr
ies
to
f
i
n
d
th
e
li
n
ea
r
m
ar
g
in
i
n
t
h
e
n
e
w
f
ea
t
u
r
e
s
p
ac
e
[
10
]
.
Ass
u
m
in
g
{(
1
,
1
)
,
…
,
(
,
)
}
b
e
a
tr
ain
in
g
s
et
w
it
h
1
an
d
is
th
e
co
r
r
esp
o
n
d
in
g
tar
g
et
class
.
SVM
ca
n
b
e
r
ef
o
r
m
u
lat
ed
as
(
1
)
an
d
(
2
)
:
Ma
x
i
m
ize=
∑
−
1
2
∑
∑
(
,
)
=
1
=
1
=
1
(
1
)
Su
b
j
ec
t to
;
∑
=
0
≥
0
,
=
1
,
2
,
…
,
=
1
(
2
)
T
h
e
k
er
n
el
f
u
n
ct
io
n
i
s
u
s
ed
to
s
o
lv
e
t
h
e
p
r
o
b
le
m
.
T
h
e
Ker
n
e
l
f
u
n
ctio
n
a
n
al
y
s
es
th
e
r
elatio
n
s
h
ip
a
m
o
n
g
th
e
d
ata
an
d
it c
r
ea
tes a
co
m
p
l
ex
d
iv
i
s
io
n
i
n
th
e
s
p
ac
e
[
11
].
2
.
2
.
1
.
Rule
ex
t
ra
ct
io
n f
ro
m
s
v
m
T
h
e
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM)
m
eth
o
d
is
a
p
r
o
m
i
s
i
n
g
class
i
fi
ca
tio
n
an
d
r
eg
r
ess
io
n
tech
n
iq
u
e
p
r
o
p
o
s
ed
b
y
Vap
id
an
d
h
is
c
o
w
o
r
k
er
s
[
12
]
.
T
h
e
SVM
h
as
b
ee
n
s
u
cc
es
s
f
u
l
l
y
ap
p
lied
to
a
w
id
e
v
ar
iet
y
o
f
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
A
d
v
A
p
p
l Sci
I
SS
N:
2
2
5
2
-
8814
A
ch
i
-
s
q
u
a
r
e
-
S
V
M
b
a
s
ed
p
e
d
a
g
o
g
ica
l ru
le
ex
tr
a
ctio
n
met
h
o
d
fo
r
micro
a
r
r
a
y
…
(
Mu
kh
ta
r
Da
mo
la
S
a
la
w
u
)
95
ap
p
licatio
n
d
o
m
ai
n
s
[
13
]
in
clu
d
in
g
b
io
in
f
o
r
m
at
ics
[
14
]
.
I
t
is
esp
ec
iall
y
i
m
p
o
r
tan
t
f
o
r
th
e
f
ie
ld
o
f
co
m
p
u
tatio
n
a
l
b
io
lo
g
y
b
ec
au
s
e
it
is
u
s
ed
f
o
r
p
atter
n
r
ec
o
g
n
itio
n
p
r
o
b
lem
s
in
cl
u
d
in
g
p
r
o
tein
r
e
m
o
te
h
o
m
o
lo
g
y
d
etec
tio
n
,
m
icr
o
ar
r
a
y
g
e
n
e
ex
p
r
es
s
io
n
an
al
y
s
is
,
r
ec
o
g
n
itio
n
o
f
tr
a
n
s
latio
n
s
tar
t
s
ite
s
,
p
r
o
tein
s
t
r
u
ctu
r
e
p
r
ed
ictio
n
,
f
u
n
ctio
n
al
cla
s
s
i
ficatio
n
o
f
p
r
o
m
o
ter
r
eg
i
o
n
s
,
p
r
ed
ictio
n
o
f
p
r
o
tein
–
p
r
o
tein
in
ter
ac
ti
o
n
s
,
an
d
p
ep
tid
e
id
en
ti
fi
ca
tio
n
f
r
o
m
m
a
s
s
s
p
ec
t
r
o
m
etr
y
d
ata
[
15
].
T
h
e
SVM
h
a
s
s
h
o
w
n
a
s
u
p
er
io
r
p
er
f
o
r
m
a
n
ce
th
a
n
m
o
s
t tr
ad
itio
n
al
m
ac
h
in
e
lear
n
i
n
g
m
eth
o
d
s
s
u
c
h
a
s
n
eu
r
al
n
et
w
o
r
k
s
(
NN
s
)
in
m
a
n
y
ap
p
licatio
n
s
.
Ho
w
ev
er
,
t
h
es
e
tech
n
iq
u
e
s
till
p
r
o
d
u
ce
s
b
lack
b
o
x
m
o
d
els
w
it
h
litt
le
o
r
n
o
ex
p
lan
atio
n
ca
p
ab
ilit
y
.
I
n
ap
p
licatio
n
ar
ea
s
s
u
ch
as
m
ed
ical
d
iag
n
o
s
is
,
t
h
er
e
is
an
ev
id
en
t
n
ee
d
f
o
r
an
e
x
p
lan
atio
n
co
m
p
o
n
en
t
to
b
e
ass
o
ciate
d
w
it
h
cla
s
s
i
f
icat
i
o
n
d
ec
is
io
n
s
i
n
o
r
d
er
to
aid
t
h
e
ac
ce
p
tan
ce
o
f
t
h
ese
m
et
h
o
d
s
b
y
u
s
er
s
[
16
]
.
T
o
b
e
a
b
le
to
u
s
e
t
h
e
e
x
tr
a
ac
cu
r
ac
y
o
f
th
e
S
VM
,
w
h
ich
ca
n
lead
to
li
v
es
s
av
ed
o
r
m
o
n
e
y
g
ain
ed
,
as
w
ell
as
to
o
b
tai
n
a
u
s
ab
le,
r
ea
d
ab
le
m
o
d
el
,
r
u
les
ca
n
b
e
ex
tr
ac
ted
f
r
o
m
th
e
co
m
p
lex
,
in
co
m
p
r
eh
e
n
s
ib
le
SVM
m
o
d
el
s
.
T
h
ese
r
u
les ar
e
in
ter
p
r
etab
le
b
y
h
u
m
a
n
s
an
d
k
ee
p
as
m
u
ch
o
f
th
e
ac
cu
r
ac
y
o
f
th
e
b
lack
b
o
x
as p
o
s
s
ib
le
[
17
].
2
.
3
.
Rule
ex
t
ra
ct
io
n t
ec
hn
iq
ue
C
o
m
p
r
eh
e
n
s
ib
il
it
y
ca
n
b
e
ad
d
ed
to
SVMs
b
y
ex
tr
ac
tin
g
s
y
m
b
o
lic
r
u
les
f
r
o
m
t
h
e
tr
ai
n
ed
m
o
d
el.
R
u
le
ex
tr
ac
tio
n
tec
h
n
iq
u
e
s
att
e
m
p
t
to
o
p
en
u
p
t
h
e
SVM
b
l
ac
k
b
o
x
a
n
d
g
en
er
ate
s
y
m
b
o
l
ic,
co
m
p
r
eh
e
n
s
ib
le
d
escr
ip
tio
n
s
w
it
h
ap
p
r
o
x
i
m
a
tel
y
t
h
e
s
a
m
e
p
r
ed
ictiv
e
p
o
w
er
as
th
e
m
o
d
el
its
e
lf
.
An
ad
v
a
n
t
ag
e
o
f
u
s
i
n
g
SVM
s
as
a
s
tar
ti
n
g
p
o
in
t
f
o
r
r
u
le
e
x
tr
ac
tio
n
is
t
h
at
t
h
e
SVM
c
o
n
s
id
er
s
t
h
e
co
n
tr
ib
u
tio
n
o
f
th
e
i
n
p
u
t
s
to
w
ar
d
s
class
i
fi
ca
tio
n
a
s
a
g
r
o
u
p
,
w
h
il
e
d
ec
is
io
n
tr
ee
alg
o
r
it
h
m
s
li
k
e
[
18
]
C
A
R
T
m
ea
s
u
r
e
t
h
e
in
d
i
v
id
u
al
co
n
tr
ib
u
tio
n
o
f
th
e
i
n
p
u
t
s
o
n
e
at
a
ti
m
e
as t
h
e
tr
ee
is
g
r
o
w
n
.
I
n
g
e
n
er
al,
r
u
le
ex
tr
ac
tio
n
te
ch
n
iq
u
es
ar
e
d
iv
id
ed
i
n
to
t
wo
m
aj
o
r
g
r
o
u
p
s
i.e
.
d
ec
o
m
p
o
s
itio
n
a
n
d
p
ed
ag
o
g
ical.
Dec
o
m
p
o
s
it
io
n
tech
n
iq
u
es
v
ie
w
t
h
e
m
o
d
el
at
its
m
i
n
i
m
u
m
(
o
r
fin
est)
l
ev
el
o
f
g
r
a
n
u
lar
i
t
y
(
at
th
e
le
v
el
o
f
h
id
d
en
an
d
o
u
t
p
u
t
u
n
it
s
i
n
ca
s
e
o
f
A
N
N)
.
R
u
l
es
ar
e
fi
r
s
t
e
x
tr
ac
ted
at
i
n
d
i
v
id
u
al
u
n
i
t
le
v
el,
t
h
ese
s
u
b
s
et
s
o
f
r
u
le
s
ar
e
th
e
n
ag
g
r
eg
ated
to
f
o
r
m
g
lo
b
al
r
elatio
n
s
h
i
p
[
19
]
.
On
th
e
o
th
er
h
a
n
d
,
a
p
ed
ag
o
g
ical
al
g
o
r
ith
m
co
n
s
id
er
s
t
h
e
tr
ain
ed
m
o
d
el
as
a
b
lack
b
o
x
.
I
n
s
tead
o
f
lo
o
k
i
n
g
at
t
h
e
i
n
ter
n
al
s
tr
u
ctu
r
e,
t
h
es
e
alg
o
r
ith
m
s
d
o
n
o
t
m
ak
e
u
s
e
o
f
th
e
s
u
p
p
o
r
t
v
ec
to
r
s
o
r
SVM
d
ec
is
io
n
b
o
u
n
d
ar
y
,
b
u
t
d
ir
ec
tl
y
ex
tr
ac
t
r
u
les
u
s
i
n
g
t
h
e
i
n
p
u
t
–
o
u
tp
u
t
m
ap
p
in
g
d
efin
ed
b
y
t
h
e
SVM
m
o
d
el.
T
h
ese
tech
n
iq
u
es
t
y
p
i
ca
ll
y
u
s
e
th
e
tr
ain
ed
SVM
m
o
d
el
as
an
o
r
ac
le
to
lab
el
o
r
class
if
y
(
ar
ti
fi
ciall
y
g
e
n
er
ated
)
tr
ain
in
g
ex
a
m
p
les
w
h
i
ch
ar
e
th
en
u
s
ed
b
y
a
s
y
m
b
o
lic
lear
n
in
g
al
g
o
r
ith
m
.
T
h
e
id
ea
b
eh
in
d
th
ese
tec
h
n
iq
u
es
i
s
th
e
a
s
s
u
m
p
t
io
n
t
h
at
t
h
e
tr
ain
ed
m
o
d
el
ca
n
b
etter
r
ep
r
esen
t
t
h
e
d
ata
th
a
n
th
e
o
r
ig
i
n
al
d
ataset.
T
h
at
is
,
th
e
d
ata
is
clea
n
er
,
f
r
ee
o
f
ap
p
ar
en
t
co
n
fl
ic
ts
.
Si
n
ce
th
e
m
o
d
el
i
s
v
ie
w
ed
as
a
b
lack
b
o
x
,
m
o
s
t
p
ed
ag
o
g
ica
l
alg
o
r
it
h
m
s
le
n
d
th
e
m
s
elv
e
s
v
er
y
ea
s
i
l
y
to
r
u
le
e
x
tr
ac
tio
n
f
r
o
m
o
th
e
r
m
ac
h
i
n
e
lear
n
in
g
alg
o
r
ith
m
s
[
20
].
2
.
4
.
Dec
is
io
n t
re
e
T
h
e
co
m
p
r
eh
e
n
s
ib
ili
t
y
o
f
d
ec
is
io
n
tr
ee
s
i
s
o
n
e
t
h
eir
m
o
s
t u
s
ef
u
l c
h
ar
ac
ter
is
tics
,
s
i
n
ce
d
o
m
ain
ex
p
er
t
s
ca
n
ea
s
i
l
y
u
n
d
er
s
tan
d
th
e
p
r
i
n
cip
le
o
f
t
h
e
tr
ee
,
an
d
w
h
y
a
c
er
tain
o
b
j
ec
t
is
clas
s
if
ied
to
b
elo
n
g
to
a
s
p
ec
i
f
ic
class
.
Mo
r
eo
v
er
,
d
ec
is
io
n
tr
ee
s
ar
e
p
r
o
b
ab
ly
th
e
m
o
s
t
e
x
ten
s
iv
el
y
r
e
s
ea
r
ch
ed
m
ac
h
in
e
lea
r
n
in
g
m
et
h
o
d
,
ca
n
d
ea
l
w
ith
an
y
k
i
n
d
o
f
in
p
u
t
d
ata
(
d
is
cr
ete
,
c
o
n
tin
u
o
u
s
,
b
in
a
r
y
,
attr
ib
u
te
s
)
.
T
h
ey
ca
n
also
co
p
e
w
it
h
m
i
s
s
i
n
g
v
alu
e
s
,
s
i
n
ce
th
e
in
f
o
r
m
atio
n
th
at
attr
ib
u
te
v
a
lu
e
s
ar
e
m
i
s
s
i
n
g
f
o
r
s
p
ec
if
ic
o
b
j
ec
ts
ca
n
b
e
p
r
o
ce
s
s
ed
b
y
m
o
s
t
d
ec
is
io
n
tr
ee
alg
o
r
ith
m
s
.
T
h
e
l
ea
r
n
in
g
p
r
o
ce
s
s
o
f
d
ec
is
io
n
tr
ee
s
is
u
s
u
all
y
q
u
i
te
f
ast
co
m
p
ar
ed
to
o
th
er
m
et
h
o
d
s
lik
e
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
i
n
es
o
r
n
eu
r
al
n
et
w
o
r
k
s
,
an
d
s
i
n
ce
m
o
s
t
tr
ee
s
ar
e
p
r
u
n
ed
,
th
eir
class
i
f
icatio
n
p
r
o
ce
s
s
is
u
s
u
all
y
also
v
er
y
f
a
s
t.
Se
v
er
al
s
t
u
d
ies
[
21
]
h
av
e
s
h
o
w
n
t
h
at
t
h
e
c
lass
if
ica
tio
n
ac
cu
r
ac
y
i
s
g
en
er
al
l
y
co
m
p
ar
ab
le
to
th
e
q
u
alit
y
o
f
k
NN
an
d
r
u
le
-
b
ased
lear
n
er
s
,
b
u
t
ca
n
n
o
t
r
ea
ch
th
e
q
u
al
it
y
o
f
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
s
o
r
em
b
ed
d
ed
m
et
h
o
d
s
,
w
h
ic
h
,
o
n
th
e
co
n
tr
ar
y
,
ar
e
h
ar
d
l
y
co
m
p
r
eh
e
n
s
ib
le
(
d
if
f
ic
u
lt
to
u
n
d
er
s
ta
n
d
f
o
r
d
o
m
ai
n
e
x
p
er
ts
)
an
d
ar
e
n
o
t
g
o
o
d
in
h
a
n
d
lin
g
m
is
s
i
n
g
d
ata
(
s
in
ce
m
is
s
i
n
g
d
ata
h
as
t
o
b
e
r
ep
lace
d
w
it
h
alter
n
ati
v
e
v
al
u
es
s
u
c
h
as
m
ea
n
o
r
ze
r
o
v
alu
es b
ef
o
r
e
class
if
i
ca
tio
n
)
.
2
.
5
.
Co
m
bin
ed
s
v
m
a
nd
deci
s
io
n
t
re
e
T
h
e
in
s
p
ir
atio
n
o
f
co
m
b
i
n
in
g
t
h
e
SVM
an
d
d
ec
is
io
n
tr
ee
is
t
o
m
er
g
e
th
e
s
tr
o
n
g
g
e
n
er
aliza
t
io
n
ab
ilit
y
o
f
th
e
S
VM
an
d
t
h
e
s
tr
o
n
g
co
m
p
r
e
h
e
n
s
ib
ilit
y
o
f
r
u
le
i
n
d
u
cti
o
n
.
Sp
ec
i
fi
ca
ll
y
,
o
u
r
alg
o
r
it
h
m
e
m
p
lo
y
s
t
h
e
S
VM
as
a
p
r
ep
r
o
ce
s
s
o
f
d
ec
is
io
n
tr
ee
an
d
co
n
s
is
t
s
o
f
th
r
ee
m
aj
o
r
s
tep
s
.
First,
a
lab
elled
d
ata
s
et
is
u
s
ed
f
o
r
SVM
lear
n
in
g
p
u
r
p
o
s
es,
i.e
.
to
b
u
ild
a
m
o
d
el
w
it
h
ac
ce
p
tab
le
ac
cu
r
ac
y
.
A
s
ec
o
n
d
d
ata
s
et
is
g
en
er
ated
w
i
th
th
e
s
a
m
e
attr
ib
u
tes
b
u
t
d
i
f
f
er
e
n
t
v
al
u
es
to
ex
p
lo
r
e
th
e
g
en
er
aliza
tio
n
b
eh
av
io
r
o
f
th
e
S
VM
.
T
h
at
is
,
th
e
SVM
is
u
s
ed
to
g
et
t
h
e
cla
s
s
lab
el
s
f
o
r
th
i
s
d
at
a
s
et.
He
n
ce
a
s
y
n
t
h
etic
d
ata
s
e
t
is
o
b
tain
ed
.
Fin
a
ll
y
,
th
e
s
y
n
t
h
etic
s
et
i
s
t
h
e
n
u
s
ed
to
tr
ain
a
m
ac
h
i
n
e
lear
n
i
n
g
tec
h
n
iq
u
e
w
i
th
e
x
p
la
n
atio
n
ca
p
a
b
ilit
y
.
T
h
er
eb
y
,
r
u
le
s
ar
e
g
en
e
r
ated
th
at
r
ep
r
esen
t
th
e
g
e
n
er
aliza
tio
n
b
eh
av
io
r
o
f
th
e
SVM
[
22
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
: 2
2
5
2
-
8814
I
n
t J
A
d
v
A
p
p
l Sci
,
Vo
l.
9
,
No
.
2
,
J
u
n
e
2
0
2
0
:
93
–
1
0
0
96
3.
P
RO
P
O
SE
D
RU
L
E
E
XT
RA
CT
I
O
N
AP
P
RO
ACH
T
h
e
p
r
o
p
o
s
ed
s
y
s
te
m
is
a
p
ed
a
g
o
g
ical/
L
ea
r
n
i
n
g
-
B
ased
p
r
o
ce
d
u
r
e
f
o
r
ex
tr
ac
tin
g
r
u
le
s
f
r
o
m
SVM
an
d
is
co
m
p
o
s
ed
o
f
t
h
r
ee
p
h
ases
.
Firstl
y
,
f
ea
t
u
r
e
s
elec
t
io
n
u
s
i
n
g
C
h
i
-
s
q
u
ar
e
-
SVM
i
s
f
ir
s
t
e
m
p
l
o
y
ed
a
n
d
th
e
ac
t
u
a
l
tar
g
et
v
a
lu
e
s
o
f
tr
ain
i
n
g
i
n
s
ta
n
ce
s
ar
e
r
ep
lace
d
b
y
t
h
e
p
r
ed
icti
o
n
s
o
f
SV
M
m
o
d
els
a
n
d
C
ase
-
P
(
i.e
.
tr
ain
in
g
in
s
ta
n
ce
s
w
i
th
co
r
r
esp
o
n
d
in
g
p
r
ed
icted
tar
g
et
v
alu
es)
d
atasets
ar
e
g
en
er
ated
.
Fo
r
th
e
h
ig
h
d
i
m
en
s
io
n
al
it
y
d
at
a
s
et
p
r
o
d
u
ce
d
b
y
m
icr
o
ar
r
a
y
,
W
is
co
n
s
i
n
b
r
ea
s
t
ca
n
ce
r
d
atas
et
is
a
n
al
y
s
ed
.
I
t
is
o
b
s
er
v
ed
t
h
at
r
ed
u
ce
d
f
ea
tu
r
e
s
r
ed
u
ce
th
e
co
m
p
lex
it
y
o
f
th
e
s
y
s
te
m
a
n
d
in
cr
ea
s
e
s
th
e
co
m
p
r
eh
en
s
ib
ili
t
y
o
f
t
h
e
r
u
le
s
Sec
o
n
d
l
y
,
th
e
r
ed
u
ce
d
d
ataset
is
u
s
ed
to
tr
ain
t
h
e
S
VM
,
a
s
ec
o
n
d
d
ataset
i
s
g
e
n
e
r
ated
i.e
SVM
is
u
s
ed
to
g
et
th
e
clas
s
lab
els
f
o
r
th
e
d
ataset.
He
n
ce
a
s
y
n
th
etic
d
ata
is
o
b
tain
ed
.
Fin
all
y
,
R
u
le
s
ar
e
g
en
er
ated
u
s
i
n
g
NB
T
r
ee
.
Oth
er
ca
s
es
w
ill
b
e
d
is
cu
s
s
ed
is
s
u
b
s
eq
u
e
n
t
w
o
r
k
.
T
h
e
ar
ch
itectu
r
e
o
f
t
h
e
ap
p
r
o
a
ch
p
r
o
p
o
s
ed
is
s
h
o
w
n
i
n
Fi
g
u
r
e
1
.
4.
E
XP
E
R
I
M
E
NT
A
L
SE
T
UP
4
.
1
.
Da
t
a
s
et
des
cr
iptio
n
Fo
r
b
u
ild
in
g
a
n
d
test
i
n
g
t
h
e
ef
f
ec
tiv
e
n
e
s
s
o
f
o
u
r
al
g
o
r
ith
m
,
w
e
p
er
f
o
r
m
ed
ex
p
er
i
m
e
n
t
o
n
ca
n
ce
r
d
ataset.
T
h
e
b
r
ea
s
t
ca
n
ce
r
d
at
ab
ase
w
a
s
o
b
tain
ed
f
r
o
m
t
h
e
Un
i
v
er
s
it
y
o
f
W
is
co
n
s
i
n
Ho
s
p
itals
,
Ma
d
is
o
n
f
r
o
m
Dr
.
W
illi
a
m
H.
W
o
lb
er
g
.
T
h
is
d
ataset
is
ch
o
s
e
n
b
ec
a
u
s
e
o
f
i
ts
p
u
b
lic
ac
ce
s
s
ib
ilit
y
a
n
d
h
as
p
r
ev
io
u
s
l
y
b
ee
n
u
s
ed
f
o
r
s
e
v
er
al
Ma
c
h
in
e
L
ea
r
n
i
n
g
s
tu
d
ie
s
.
T
h
e
p
r
o
b
lem
is
to
clas
s
if
y
b
r
ea
s
t
m
a
s
s
e
s
a
s
eit
h
er
b
e
n
ig
n
o
r
m
ali
g
n
an
t,
u
s
i
n
g
n
i
n
e
attr
ib
u
tes,
all
w
i
th
i
n
teg
er
v
al
u
es
b
et
w
ee
n
1
an
d
1
0
.
Fo
r
th
e
class
lab
el
(
2
f
o
r
b
en
ig
n
,
4
f
o
r
m
al
ig
n
a
n
t)
.
Fig
u
r
e
1
.
A
r
ch
itectu
r
e
o
f
t
h
e
p
r
o
p
o
s
ed
ch
i
-
s
q
u
ar
e
-
SVM
p
ed
ag
o
g
ical
r
u
le
ex
tr
ac
tio
n
ap
p
r
o
a
ch
4
.
2
.
E
x
peri
m
e
nta
l set
up
T
h
is
d
ata
s
et
h
as
6
9
9
s
a
m
p
les
.
I
n
th
is
e
x
p
er
i
m
e
n
t
th
e
y
w
er
e
d
iv
id
ed
in
to
t
w
o
p
ar
ts
o
f
8
0
:2
0
r
atio
s
.
8
0
%
an
d
2
0
%
tr
ain
in
g
a
n
d
tes
t/v
alid
atio
n
s
et
s
w
i
th
5
6
0
an
d
1
3
9
s
a
m
p
les
r
esp
ec
ti
v
el
y
.
T
h
e
2
0
%
o
f
t
h
e
d
ata
i
s
k
ep
t
asid
e
f
o
r
later
u
s
e.
Usi
n
g
v
alid
atio
n
s
et,
ef
f
icie
n
c
y
o
f
th
e
r
u
les
g
e
n
er
ated
d
u
r
in
g
th
e
ex
p
er
i
m
e
n
t
is
ev
alu
a
ted
.
T
h
e
ex
p
er
i
m
en
tal
s
etu
p
w
er
e
ca
r
r
ied
o
u
t
an
d
d
ev
elo
p
ed
w
ith
th
e
Ma
t
lab
p
r
o
g
r
am
m
in
g
(
M
A
T
L
A
B
2
0
1
5
A
)
v
ar
io
u
s
f
u
n
ctio
n
s
w
e
r
e
d
ev
elo
p
ed
an
d
lin
k
ed
to
a
g
r
ap
h
ic
u
s
er
in
ter
f
ac
e
f
o
r
u
s
er
in
ter
ac
tiv
i
t
y
a
n
d
r
esp
o
n
s
iv
e
n
e
s
s
.
Du
r
i
n
g
t
h
e
tr
ain
i
n
g
o
f
th
e
SV
M
m
o
d
el
th
e
f
o
llo
w
i
n
g
ar
e
th
e
p
ar
am
e
ter
s
ettin
g
s
th
a
t
w
a
s
u
s
ed
C
o
s
t=0
.
7
6
2
9
4
2
7
7
9
2
9
1
5
5
3
,
Ker
n
el
f
u
n
ctio
n
=
R
B
f
Ker
n
el
th
e
d
ev
elo
p
ed
s
y
s
te
m
s
m
ad
e
u
s
e
o
f
v
ar
io
u
s
co
m
p
o
n
e
n
t e
n
v
ir
o
n
m
en
t
s
in
Ma
tlab
to
d
ev
elo
p
an
d
o
u
tp
u
t r
esu
lt o
f
t
h
e
d
ata
m
in
i
n
g
tas
k
.
5.
RE
SU
L
T
AND
DI
SCUS
SI
O
N
W
e
ev
alu
ate
an
d
d
i
s
cu
s
s
t
h
e
p
er
f
o
r
m
an
ce
o
f
o
u
r
ap
p
r
o
ac
h
S
VM
+N
B
T
r
ee
u
s
in
g
C
a
s
e
-
P
w
i
th
r
esp
ec
t
to
s
p
ec
if
icit
y
,
s
e
n
s
i
tiv
it
y
,
an
d
ac
cu
r
ac
y
.
Du
r
in
g
f
ir
s
t
f
ir
s
t
p
h
ase
o
f
th
e
p
r
o
p
o
s
ed
a
p
p
r
o
ac
h
,
SVM
-
C
h
i
-
s
q
u
ar
e
alg
o
r
ith
m
is
e
m
p
lo
y
ed
f
o
r
f
ea
tu
r
e
s
elec
tio
n
an
d
s
ix
at
tr
ib
u
t
es
ar
e
th
en
s
elec
ted
,
th
o
s
e
ar
e
,
C
lu
m
p
th
ic
k
n
es
s
,
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
A
d
v
A
p
p
l Sci
I
SS
N:
2
2
5
2
-
8814
A
ch
i
-
s
q
u
a
r
e
-
S
V
M
b
a
s
ed
p
e
d
a
g
o
g
ica
l ru
le
ex
tr
a
ctio
n
met
h
o
d
fo
r
micro
a
r
r
a
y
…
(
Mu
kh
ta
r
Da
mo
la
S
a
la
w
u
)
97
Un
i
f
o
r
m
it
y
o
f
ce
ll
s
ize,
Ma
r
g
i
n
al
ad
h
esio
n
,
Sin
g
le
ep
ith
e
lial c
ell
s
ize,
B
ar
e
n
u
clei,
an
d
B
lan
d
ch
r
o
m
ati
n
.
A
f
ter
s
elec
tio
n
o
f
o
p
ti
m
al
s
u
b
s
et
th
e
s
elec
ted
s
u
b
s
et
w
as
h
e
n
ce
d
iv
id
ed
in
to
tr
ain
in
g
s
et
an
d
test
s
et
in
th
e
r
atio
o
f
8
0
%: 2
0
%.
T
h
en
,
th
e
tr
ain
i
n
g
s
et
w
a
s
p
ass
ed
to
SVM
m
o
d
el
f
o
r
th
e
p
r
ed
ictio
n
o
f
n
e
w
clas
s
lab
el
w
h
ic
h
f
o
r
m
s
th
e
s
y
n
th
et
ic
class
lab
el.
A
tr
ain
i
n
g
ti
m
e
o
f
3
.
1
9
5
1
2
s
ec
o
n
d
s
w
a
s
o
b
tain
ed
.
T
h
e
s
y
n
t
h
eti
c
s
et
is
n
o
w
u
s
ed
to
tr
ain
t
h
e
d
ec
i
s
io
n
tr
ee
.
T
h
e
C
AR
T
d
ec
is
io
n
tr
ee
w
a
s
u
s
ed
to
g
en
er
alize
th
e
o
b
tai
n
ed
r
esu
l
ts
.
T
h
er
eb
y
,
a
to
tal
o
f
n
in
e
r
u
les
w
er
e
g
e
n
er
ated
th
at
r
ep
r
esen
ts
th
e
g
e
n
er
aliza
tio
n
b
eh
av
io
r
o
f
th
e
SVM.
Fi
g
u
r
e
2
s
h
o
w
s
th
e
C
AR
T
tr
ee
g
en
er
ated
.
Fig
u
r
e
2
.
C
A
R
T
tr
ee
g
en
er
ate
d
5
.
1
.
Co
nfusi
o
n
m
a
t
rix
f
o
r
o
ur
pr
o
po
s
ed
m
o
del
T
h
e
ef
f
ec
ti
v
en
e
s
s
o
f
o
u
r
ap
p
r
o
ac
h
is
ev
alu
ated
u
s
i
n
g
A
cc
u
r
ac
y
,
Se
n
s
i
tiv
it
y
,
a
n
d
Sp
ec
if
icit
y
.
T
h
e
m
ea
s
u
r
es
u
s
ed
to
ev
al
u
ate
th
e
p
er
f
o
r
m
a
n
ce
o
f
o
u
r
m
o
d
el
is
d
ef
in
ed
as
f
o
llo
w
s
:
Sen
s
iti
v
it
y
=
T
P
/
(
T
P
+FN)
,
Sp
ec
if
icit
y
=
T
N/
(
FP
+
T
N,
)
,
A
cc
u
r
ac
y
=
(
T
P
+
T
N)
/
(
T
P
+
T
N+
FP
+FN)
,
T
P
=
T
r
u
e
P
o
s
itiv
e,
T
N
=
T
r
u
e
Neg
ativ
e,
FP
=
Fals
e
P
o
s
iti
v
e,
F
N
=
Fal
s
e
Neg
a
tiv
e.
Fi
g
u
r
e
3
s
h
o
w
s
t
h
e
co
n
f
u
s
io
n
m
atr
i
x
f
o
r
o
u
r
p
r
o
p
o
s
ed
m
o
d
el
i.e
C
a
s
e
P
Fig
u
r
e
3
.
C
o
n
f
u
s
io
n
m
atr
ix
f
o
r
o
u
r
p
r
o
p
o
s
ed
m
o
d
el
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
: 2
2
5
2
-
8814
I
n
t J
A
d
v
A
p
p
l Sci
,
Vo
l.
9
,
No
.
2
,
J
u
n
e
2
0
2
0
:
93
–
1
0
0
98
Fro
m
th
e
F
ig
u
r
e
1
ab
o
v
e
o
u
r
v
alu
es c
a
n
b
e
o
b
tain
ed
:
T
P
=
4
7
,
T
N
=
8
3
,
F
P
=
8
,
FN =
1
Sen
s
iti
v
it
y
=
4
7
/4
7
+
1
=
0
.
9
7
9
1
=
9
7
.
9
1
6
6
%
Sp
ec
if
icit
y
=
8
3
/3
+
8
8
=
0
.
9
6
7
0
3
3
=
9
6
.
7
0
3
3
%
A
cc
u
r
ac
y
=
4
7
+
8
3
/ 4
7
+
8
3
+
8
+
1
=
0
.
9
3
5
2
=
9
3
.
5
2
5
1
%
5
.
2
.
Co
m
pa
riso
n o
f
o
ur
t
ec
hn
iqu
e
t
o
o
t
her
s
v
m
ru
le
ex
t
ra
ct
io
n t
ec
hn
iqu
e
T
ab
le
1
s
h
o
w
s
co
m
p
ar
ati
v
e
r
e
s
u
lt
s
o
f
SVM
+
C
A
R
T
+C
h
i
-
s
q
u
ar
e
ag
ain
s
t
a
n
u
m
b
er
o
f
o
t
h
er
p
r
ev
io
u
s
l
y
p
u
b
lis
h
ed
SV
M
r
u
le
ex
tr
ac
tio
n
m
et
h
o
d
s
.
T
ab
le
1
.
C
o
m
p
ar
is
o
n
w
i
th
o
t
h
er
r
u
le
ex
tr
ac
tio
n
m
et
h
o
d
T
e
c
h
n
i
q
u
e
A
c
c
u
r
a
c
y
(
%)
S
e
n
si
t
i
v
i
t
y
(
%)
S
p
e
c
i
f
i
c
i
t
y
(
%)
N
u
mb
e
r
o
f
R
u
l
e
s
T
i
me
(
se
c
)
S
V
M
+
N
B
T
r
e
e
-
F
a
r
q
u
a
d
(
2
0
0
9
)
7
7
.
0
7
6
8
.
5
2
78
N
/
A
N
/
A
S
V
M
+
C
4
.
5
-
M
a
r
t
e
n
s (
2
0
0
7
)
9
4
.
6
N
/
A
N
/
A
9
N
/
A
S
V
M
+
C
A
R
T
+
C
h
i
-
s
q
u
a
r
e
(
2
0
1
6
)
9
3
.
5
2
5
1
8
9
.
5
8
3
3
9
6
.
7
0
3
3
9
3
.
1
9
5
T
h
e
ch
u
r
n
p
r
ed
ictio
n
d
ataset
w
a
s
an
al
y
s
ed
b
y
Far
q
u
ad
,
w
h
ile
w
is
co
s
in
b
r
ea
s
t
ca
n
ca
er
d
ata
s
et
w
as
an
al
y
ze
d
b
y
Ma
r
ten
s
a
n
d
u
s
.
I
n
th
e
s
e
tab
les,
i
t
is
e
v
id
e
n
t
t
h
at
SV
M+
C
AR
T
+Ch
i
-
s
q
u
ar
e
h
as,
o
n
th
e
d
ata
s
et,
s
ig
n
i
f
ica
n
tl
y
s
m
a
ll
r
u
le
s
e
ts
an
d
,
th
er
ef
o
r
e,
p
o
ten
tia
l
l
y
i
m
p
r
o
v
ed
co
m
p
r
eh
e
n
s
ib
i
lit
y
.
I
n
ad
d
itio
n
,
SVM+
C
AR
T
+Ch
i
-
s
q
u
ar
e
h
as
g
o
o
d
p
er
f
o
r
m
an
ce
,
as
m
ea
s
u
r
ed
b
y
th
e
o
v
er
all
ac
c
u
r
ac
y
.
I
t
w
il
l
b
e
n
o
ted
th
at
C
4
.
5
o
u
tp
er
f
o
r
m
ed
o
u
r
alg
o
r
ith
m
i
n
ter
m
s
o
f
ac
cu
r
ac
y
alo
n
e
b
u
t
s
en
s
iti
v
it
y
,
s
p
ec
if
ici
t
y
,
an
d
ti
m
e
w
er
e
n
o
t
r
ep
o
r
ted
f
o
r
a
c
o
m
p
ar
is
o
n
.
I
t
is
co
n
cl
u
d
ed
th
at
t
h
er
e
is
n
o
o
n
e
alg
o
r
ith
m
t
h
at
ca
n
b
e
f
av
o
r
ed
in
g
e
n
er
al.
T
h
er
e
is
cu
r
r
en
tl
y
n
o
o
n
e
m
e
t
h
o
d
th
at
ca
n
f
u
lf
ill all
cr
iter
ia
s
i
m
u
lta
n
eo
u
s
l
y
.
T
h
er
e
is
al
w
a
y
s
tr
ad
eo
f
f
s
.
6.
CO
NCLU
SI
O
N
I
n
th
i
s
w
o
r
k
,
w
e
tr
ea
ted
Fea
tu
r
e
Selectio
n
as
a
n
i
n
te
g
r
al
p
ar
t
o
f
r
u
le
e
x
tr
ac
tio
n
.
E
m
p
lo
y
i
n
g
Feat
u
r
e
s
elec
tio
n
lead
to
th
e
r
em
o
v
al
o
f
ir
r
elev
an
t
f
ea
tu
r
e
s
w
h
ic
h
d
o
n
o
t
c
o
n
tr
ib
u
te
to
class
if
ic
atio
n
d
ec
is
io
n
an
d
ex
tr
ac
tio
n
o
f
m
o
r
e
co
m
p
r
eh
e
n
s
ib
le
r
u
les.
SVM
s
h
av
e
p
r
o
v
en
to
b
e
a
class
if
icatio
n
tech
n
iq
u
e
w
it
h
ex
ce
lle
n
t
p
r
ed
ictiv
e
p
er
f
o
r
m
a
n
ce
.
A
s
f
o
r
m
a
n
y
ap
p
licatio
n
s
,
th
e
o
p
aq
u
en
es
s
o
f
t
h
e
tr
ain
ed
n
o
n
l
in
ea
r
m
o
d
el
is
an
u
n
b
r
ea
ch
ab
le
b
ar
r
ier
;
m
o
r
e
co
m
p
r
e
h
e
n
s
ib
le
s
o
lu
tio
n
s
n
ee
d
to
b
e
f
o
u
n
d
.
T
h
e
m
o
s
t
co
m
p
r
eh
e
n
s
ib
le
clas
s
if
icatio
n
m
o
d
el
s
b
ein
g
r
u
le
s
ets,
SVM
r
u
le
ex
tr
ac
tio
n
tr
ies
to
co
m
b
i
n
e
th
e
p
r
ed
ictiv
e
ac
cu
r
ac
y
o
f
t
h
e
tr
ain
ed
SVM
m
o
d
el
w
it
h
t
h
e
co
m
p
r
eh
e
n
s
ib
il
it
y
o
f
th
e
r
u
le
s
e
t f
o
r
m
a
t
R
u
le
ex
tr
ac
tio
n
tech
n
iq
u
e
s
g
e
n
er
ate
class
i
fi
ca
tio
n
m
o
d
els
th
at
h
av
e
clea
r
ad
v
an
tag
e
s
.
First
o
f
all,
th
e
y
ar
e
co
m
p
r
eh
e
n
s
ib
le
an
d
th
er
e
f
o
r
e
ea
s
y
to
in
co
r
p
o
r
ate
in
r
ea
l
-
l
if
e
ap
p
licatio
n
s
w
h
er
e
clar
it
y
o
f
th
e
clas
s
i
fi
ca
t
io
n
s
m
ad
e
is
n
ee
d
ed
.
Seco
n
d
l
y
,
th
e
ex
tr
ac
ted
r
u
le
s
o
n
l
y
lo
s
e
a
s
m
all
p
er
ce
n
tag
e
in
ac
c
u
r
ac
y
o
f
t
h
e
b
lack
b
o
x
m
o
d
el
f
r
o
m
w
h
ich
t
h
e
y
ar
e
g
en
er
ated
.
Sin
ce
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
i
n
es
ar
e
a
m
o
n
g
th
e
b
est
p
er
f
o
r
m
in
g
cla
s
s
i
fier
s
,
r
u
le
s
ex
tr
ac
ted
f
r
o
m
SVMs
ac
h
ie
v
e
an
ac
cu
r
ac
y
th
a
t
o
f
te
n
s
u
r
p
as
s
es
t
h
at
o
f
t
h
e
clas
s
ical
m
et
h
o
d
s
,
s
u
c
h
as
C
A
R
T
an
d
C
4
.
5
.
Usi
n
g
th
e
SVM
m
o
d
el
in
s
tead
o
f
th
e
o
r
ig
in
al
d
ata
p
o
in
ts
eli
m
in
a
tes
th
e
ap
p
ar
en
t
co
n
fl
icts
a
n
d
cr
ea
tes
a
clea
n
er
d
ataset.
I
n
o
u
r
ex
p
er
i
m
en
ts
,
th
e
r
u
le
s
g
e
n
er
ated
b
y
C
AR
T
o
n
th
e
d
ata
w
it
h
lab
els
p
r
ed
icted
b
y
th
e
SVM
ev
e
n
o
u
tp
er
f
o
r
m
t
h
e
C
AR
T
r
u
les
th
at
r
esu
lt
f
r
o
m
th
e
d
ataset
w
i
th
t
h
e
ac
tu
al
clas
s
lab
els.
T
h
ese
ad
v
an
ta
g
es
m
ak
e
it
ap
p
r
o
p
r
iate
to
co
n
s
id
er
SVMs
a
n
d
th
eir
e
x
tr
ac
ted
r
u
le
s
f
o
r
ap
p
licatio
n
s
w
h
er
e
b
o
t
h
ac
cu
r
ac
y
an
d
co
m
p
r
eh
e
n
s
ib
ili
t
y
ar
e
r
eq
u
ir
ed
.
On
e
n
o
lo
n
g
er
n
ee
d
s
to
s
ettle
f
o
r
th
e
tr
ad
itio
n
al
co
m
p
r
eh
e
n
s
ib
le,
y
et
les
s
ac
cu
r
ate
clas
s
i
fi
ca
t
io
n
m
et
h
o
d
s
.
RE
F
E
R
E
NC
E
S
[1
]
Zen
a
,
M
.
H.,
a
n
d
G
il
li
e
s,
D.
F.
,
“
A
re
v
ie
w
o
f
f
e
a
tu
re
se
lec
ti
o
n
a
n
d
f
e
a
tu
re
e
x
trac
ti
o
n
m
e
th
o
d
s a
p
p
li
e
d
o
n
m
icro
a
rra
y
d
a
ta
”
,
De
p
a
rtme
n
t
o
f
C
o
mp
u
ti
n
g
,
Im
p
e
rial
Co
ll
e
g
e
,
2
0
1
5
.
[2
]
S
e
lv
a
ra
ja,
S
.
,
“
M
icro
a
rra
y
Da
ta
An
a
ly
sis
T
o
o
l
(M
AT
)
”,
.
Ak
ro
n
:
G
r
a
d
u
a
te
F
a
c
u
lt
y
o
f
T
h
e
Un
iv
e
rsit
y
o
f
A
k
ro
n
,
2
0
0
8
.
[3
]
A
ro
w
o
lo
M
.
O.,
A
b
d
u
lsa
lam
S
.
O.,
S
a
h
e
e
d
Y.K.,
a
n
d
S
a
law
u
M
.
D.,
“
A
f
e
a
tu
re
se
lec
ti
o
n
b
a
se
d
o
n
o
n
e
-
w
a
y
-
A
NO
V
A
f
o
r
m
icro
a
rra
y
d
a
ta cla
ss
i
f
ic
a
ti
o
n
,
”
Al
-
Hik
ma
h
J
o
u
rn
a
l
o
f
P
u
re
a
n
d
Ap
p
li
e
d
S
c
ien
c
e
s
,
v
o
l.
3
,
n
o
.
1
,
p
p
.
3
0
-
3
5
,
2
0
1
6
.
[4
]
A
n
d
re
w
s,
R.
,
Die
d
e
rich
,
J.,
a
n
d
T
ick
le,
A
.
B.
,
“
A
S
u
rv
e
y
a
n
d
c
rit
iq
u
e
o
f
tec
h
n
iq
u
e
s f
o
r
e
x
trac
ti
n
g
ru
les
f
ro
m
train
e
d
A
rti
f
icia
l
Ne
u
ra
l
Ne
t
w
o
rk
s”
,
Kn
o
wled
g
e
Ba
se
d
S
y
ste
ms
,
1
9
9
5
.
[5
]
Ba
ra
k
a
t,
N.
H.,
&
Bra
d
ley
,
A
.
P
.
,
“
Ru
le
e
x
trac
ti
o
n
f
ro
m
S
u
p
p
o
rt
V
e
c
to
r
M
a
c
h
i
n
e
s:
A
S
e
q
u
e
n
t
ial
Co
v
e
rin
g
A
p
p
ro
a
c
h
IEE
E
tran
sa
c
ti
o
n
s o
n
k
n
o
w
led
g
e
a
n
d
d
a
ta en
g
i
n
e
e
rin
g
”
,
2
0
0
7
.
[6
]
Du
c
h
,
W
.
,
S
e
ti
o
n
o
,
R.
,
&
Zu
ra
d
a
,
J.,
“
Co
m
p
u
tatio
n
a
l
in
tell
ig
e
n
c
e
m
e
th
o
d
s
f
o
r
r
u
le
-
b
a
se
d
d
a
ta
u
n
d
e
rsta
n
d
in
g
”
,
2
0
0
4
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
A
d
v
A
p
p
l Sci
I
SS
N:
2
2
5
2
-
8814
A
ch
i
-
s
q
u
a
r
e
-
S
V
M
b
a
s
ed
p
e
d
a
g
o
g
ica
l ru
le
ex
tr
a
ctio
n
met
h
o
d
fo
r
micro
a
r
r
a
y
…
(
Mu
kh
ta
r
Da
mo
la
S
a
la
w
u
)
99
[7
]
V
a
p
n
ik
,
V
.
,
“
S
tatist
ica
l
L
e
a
rn
in
g
T
h
e
o
r
y
”
Ne
w
Yo
rk
:
W
il
e
y
,
1
9
9
8
.
[8
]
S
c
h
o
e
lk
o
p
f
,
B.
,
T
su
d
a
,
K.,
a
n
d
V
e
rt,
J.
P
.
,
“
Ke
rn
e
l
M
e
th
o
d
s
in
Co
m
p
u
tati
o
n
a
l
Bio
l
o
g
y
”
,
M
A:
M
IT
Pre
ss
,
p
p
.
7
1
–
9
2
,
2
0
0
4
.
[9
]
A
ro
w
o
lo
M
.
O.,
Isia
k
a
R.
M
.
,
A
b
d
u
lsa
la
m
S
.
O.,
S
a
h
e
e
d
Y.K.,
a
n
d
G
b
o
lag
a
d
e
K.A
.
,
“
A
c
o
m
p
a
ra
ti
v
e
a
n
a
ly
sis
o
f
f
e
a
tu
re
e
x
trac
ti
o
n
m
e
th
o
d
s
f
o
r
c
las
sify
in
g
c
o
lo
n
c
a
n
c
e
r
m
icro
a
rra
y
d
a
ta,”
E
AI
En
d
o
rs
e
d
T
ra
n
s
a
c
ti
o
n
o
n
S
c
a
l
a
b
le
In
f
o
rm
a
ti
o
n
S
y
ste
m
,
v
o
l.
4
,
p
p
.
1
-
1
4
,
2
0
1
7
.
[1
0
]
F
a
rq
u
a
d
,
M
.
A
.
H.,
Ra
v
i,
V.,
S
r
iram
je
e
,
a
n
d
P
ra
v
e
e
n
G
.
,
“
Cre
d
it
S
c
o
rin
g
Us
in
g
P
CA
-
S
V
M
Hy
b
rid
M
o
d
e
l”,
S
p
rin
g
e
r
-
Ver
la
g
Be
rli
n
He
id
e
lb
e
rg
,
2
0
1
1
.
[1
1
]
Isa
b
e
ll
e
,
G
.
,
Ja
so
n
,
W
.
,
S
tep
h
e
n
B.
,
a
n
d
V
a
p
n
ik
,
V
.
,
“
G
e
n
e
se
lec
t
io
n
f
o
r
c
a
n
c
e
r
c
las
sif
ic
a
ti
o
n
u
si
n
g
su
p
p
o
r
t
v
e
c
to
r
m
a
c
h
in
e
s”
,
M
a
c
h
.
L
e
a
rn
,
p
p
.
3
8
9
-
4
2
2
,
2
0
0
2
).
[1
2
]
Co
rtes
,
C.
,
a
n
d
V
a
p
n
ik
,
V
.
,
“
S
u
p
p
o
rt
-
v
e
c
to
r
n
e
tw
o
rk
s”
,
M
a
c
h
.
L
e
a
r
n
.
,
v
o
l
.
2
0
,
p
p
.
2
3
7
–
2
9
7
,
1
9
9
5
.
[1
3
]
Cristi
a
n
in
i
,
N.,
a
n
d
S
h
a
w
e
-
Ta
y
lo
r,
J.,
“
A
n
I
n
tro
d
u
c
ti
o
n
t
o
S
u
p
p
o
r
t
Ve
c
to
r
M
a
c
h
i
n
e
s
a
n
d
Oth
e
r
Ke
rn
e
l
-
b
a
se
d
L
e
a
rn
in
g
M
e
th
o
d
s”
,
Ca
mb
ri
d
g
e
U
n
iv.
Pre
s
s
,
2
0
0
0
.
[1
4
]
S
c
h
o
e
lk
o
p
f
,
B.
,
T
su
d
a
,
K.,
a
n
d
V
e
rt,
J.
P
.
,
“
Ke
rn
e
l
M
e
th
o
d
s
in
Co
m
p
u
tati
o
n
a
l
Bio
l
o
g
y
”
,
M
A:
M
IT
Pre
ss
,
p
p
.
7
1
–
9
2
,
2
0
0
4
.
[1
5
]
No
b
le,
W
.
S
.
,
I
n
B.
S
c
h
o
e
lk
o
p
f
,
K
.
T
su
d
a
,
&
J.
-
P
.
V
e
rt
(E
d
s.),
“
Ke
r
n
e
l
m
e
th
o
d
s
i
n
c
o
m
p
u
tati
o
n
a
l
b
io
l
o
g
y
”
,
M
A:
M
IT
Pre
ss
,
”
,
p
p
.
7
1
–
9
2
,
2
0
0
4
.
[1
6
]
F
u
n
g
,
G
.
,
S
a
n
d
il
y
a
,
S
.
,
a
n
d
Ra
o
,
R.
,
“
Ru
le E
x
trac
ti
o
n
f
ro
m
L
in
e
a
r
S
u
p
p
o
rt
V
e
c
to
r
M
a
c
h
i
n
e
s”
,
Pro
c
.
1
1
t
h
I
n
t’l
C
o
n
f.
Kn
o
wled
g
e
Disc
o
v
e
ry
a
n
d
Da
t
a
M
in
in
g
,
2
0
0
5
.
[1
7
]
M
a
rten
s,
D.,
Ba
e
se
n
s,
B.
,
V
a
n
-
g
e
ste
l,
T
.
,
a
n
d
V
a
n
t
h
ien
e
n
,
J.,
“
C
o
m
p
re
h
e
n
sib
le
c
re
d
it
sc
o
ri
n
g
m
o
d
e
ls
u
si
n
g
ru
le
e
x
trac
ti
o
n
f
ro
m
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
s”
,
Eu
ro
p
e
a
n
J
o
u
rn
a
l
o
f
O
p
e
ra
ti
o
n
a
l
Res
e
a
rc
h
,
v
o
l
.
1
8
,
p
p
.
1
4
6
6
–
1
4
7
6
,
2
0
0
7
.
[1
8
]
M
a
rten
s,
D.,
Ba
e
se
n
s,
B.
,
V
a
n
-
g
e
ste
l,
T
.
,
a
n
d
V
a
n
t
h
ien
e
n
,
J.,
“
C
o
m
p
re
h
e
n
sib
le
c
re
d
it
sc
o
ri
n
g
m
o
d
e
ls
u
si
n
g
ru
le
e
x
trac
ti
o
n
f
r
o
m
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
s”
,
Eu
ro
p
e
a
n
J
o
u
rn
a
l
o
f
O
p
e
ra
ti
o
n
a
l
Res
e
a
rc
h
,
v
o
l
.
1
8
,
p
p
.
1
4
6
6
–
1
4
7
6
,
2
0
0
7
.
[1
9
]
F
a
rq
u
a
d
,
M
.
A
.
H.,
Ra
v
i,
D.,
a
n
d
Ra
ju
,
S
.
V
.
,
“
Ch
u
r
n
p
re
d
icti
o
n
u
s
in
g
c
o
m
p
re
h
e
n
sib
le
s
u
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
:
A
n
a
n
a
ly
ti
c
a
l
CRM
a
p
p
li
c
a
ti
o
n
”
,
2
0
1
4
.
[2
0
]
M
a
rten
s,
D.
,
Ba
e
se
n
s,
B.
,
V
a
n
-
g
e
ste
l,
T
.
,
a
n
d
V
a
n
t
h
ien
e
n
,
J.,
“
C
o
m
p
re
h
e
n
sib
le
c
re
d
it
sc
o
ri
n
g
m
o
d
e
ls
u
si
n
g
ru
le
e
x
trac
ti
o
n
f
ro
m
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
s”
,
Eu
ro
p
e
a
n
J
o
u
rn
a
l
o
f
O
p
e
ra
ti
o
n
a
l
Res
e
a
rc
h
,
v
o
l
.
1
8
,
p
p
.
1
4
6
6
–
1
4
7
6
,
2
0
0
7
.
[2
1
]
Ja
n
e
c
e
k
G
.
K.,
Ga
n
ste
re
r
W
.
N,
De
m
e
l
M
.
,
a
n
d
Eck
e
r
G
.
F
.
,
“
On
th
e
Re
latio
n
sh
i
p
Be
tw
e
e
n
F
e
a
tu
r
e
S
e
lec
ti
o
n
a
n
d
Clas
sif
ic
a
ti
o
n
A
c
c
u
ra
c
y
”
,
J
M
L
R:
W
o
rk
sh
o
p
a
n
d
Co
n
fer
e
n
c
e
Pro
c
e
e
d
in
g
s
,
2
0
0
8
.
[2
2
]
A
ro
w
o
lo
M
.
O.,
A
d
e
b
iy
i
M
.
O.,
a
n
d
A
d
e
b
iy
i
A
.
A
.
,
“
Di
m
e
n
sio
n
a
l
re
d
u
c
e
d
m
o
d
e
l
f
o
r
th
e
c
las
sif
ica
ti
o
n
o
f
RNA
-
S
e
q
A
n
o
p
h
e
le
s
g
a
m
b
iae
d
a
ta,”
J
o
u
rn
a
l
o
f
T
h
e
o
re
ti
c
a
l
a
n
d
Ap
p
li
e
d
In
fo
rm
a
t
io
n
T
e
c
h
n
o
lo
g
y
,
v
o
l.
9
7
,
n
o
.
2
3
,
p
p
.
1
-
1
0
,
2
0
1
9
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
M
u
k
h
tar
Da
m
o
la
S
a
law
u
re
c
e
iv
e
d
h
is
B
.
S
c
f
ro
m
th
e
d
e
p
a
rtm
e
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
,
A
l
-
Hik
m
a
h
Un
iv
e
rsit
y
.
He
is
a
CCN
A
,
CCN
P
a
n
d
M
icro
s
o
f
t
Ce
rti
f
ied
Ex
p
e
rt
,
w
it
h
h
is
re
se
a
rc
h
in
tere
sts
in
Da
ta
S
c
ien
c
e
,
Da
ta
M
i
n
i
n
g
,
Bi
o
-
in
f
o
rm
a
ti
c
s,
M
a
c
h
in
e
L
e
a
rn
in
e
,
Co
m
p
u
ter
A
rit
h
m
e
ti
c
s.
He
c
u
rre
n
tl
y
re
sid
e
s in
th
e
Un
it
e
d
S
tate
o
f
Am
e
rica
.
M
ich
e
a
l
Ola
o
l
u
A
ro
w
o
lo
re
c
e
iv
e
d
h
is
B.
S
c
a
n
d
M
.
S
c
f
ro
m
th
e
d
e
p
a
rtme
n
t
o
f
Co
m
p
u
ter
S
c
ie
n
c
e
,
Al
-
Hik
m
a
h
Un
iv
e
rsit
y
a
n
d
Kw
a
ra
S
tate
Un
iv
e
rsit
y
,
Nig
e
ria
,
in
2
0
1
2
a
n
d
2
0
1
7
re
sp
e
c
ti
v
e
l
y
.
He
is
p
re
se
n
tl
y
a
P
h
D
S
t
u
d
e
n
t
a
t
L
a
n
d
m
a
rk
Un
iv
e
rsit
y
,
O
m
u
-
A
ra
n
,
Ni
g
e
ria.
He
i
s
a
L
e
c
tu
re
r
a
t
th
e
In
stit
u
te
o
f
P
r
o
f
e
ss
io
n
a
l
S
t
u
d
ies
,
Kw
a
ra
S
tate
Un
iv
e
rsity
,
M
a
lete
,
h
e
is
a
n
Ora
c
le
Ce
rti
f
ied
Ex
p
e
rt,
a
m
e
m
b
e
r
o
f
th
e
I
A
EN
G
a
n
d
S
D
IW
C,
w
it
h
h
is
re
se
a
rc
h
in
tere
sts
in
Da
ta
S
c
ien
c
e
,
Da
ta
M
in
in
g
,
Bio
-
in
f
o
rm
a
ti
c
s,
M
a
c
h
in
e
L
e
a
rn
in
e
,
Co
m
p
u
ter A
rit
h
m
e
ti
c
s.
S
u
laim
a
n
O
lan
iy
i
A
b
d
u
lsa
la
m
Ho
ld
s
a
Ba
c
h
e
lo
r
d
e
g
re
e
a
n
d
M
a
ste
r
o
f
S
c
ien
c
e
De
g
re
e
in
Co
m
p
u
ter
S
c
ien
c
e
,
b
o
th
f
ro
m
U
n
iv
e
rsity
o
f
Ilo
rin
,
Nig
e
ria.
He
is
c
u
rre
n
tl
y
a
lec
tu
re
r
a
t
th
e
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter,
L
ib
ra
r
y
a
n
d
In
f
o
rm
a
ti
o
n
S
c
ien
c
e
,
Kw
a
ra
S
tate
Un
iv
e
rsit
y
,
M
a
lete
,
Nig
e
ria.
He
is
a
m
e
m
b
e
r
o
f
Nig
e
ria
Co
m
p
u
ter
sc
ien
c
e
.
His
r
e
se
a
r
c
h
a
re
a
s
in
c
lu
d
e
Da
ta
M
in
in
g
,
S
o
f
tw
a
r
e
En
g
in
e
e
rin
g
a
n
d
a
rti
f
icia
l
In
telli
g
e
n
c
e
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
: 2
2
5
2
-
8814
I
n
t J
A
d
v
A
p
p
l Sci
,
Vo
l.
9
,
No
.
2
,
J
u
n
e
2
0
2
0
:
93
–
1
0
0
100
Ra
f
iu
M
.
Isia
k
a
h
e
h
a
s
h
is
P
h
.
D
.
in
Co
m
p
u
ter
sc
ien
c
e
,
h
e
is
a
l
e
c
tu
re
r
in
th
e
De
p
a
rtm
e
n
t
o
f
Co
m
p
u
t
e
r
S
c
ien
c
e
,
Kw
a
ra
S
tate
Un
iv
e
rsit
y
M
a
lete
sin
c
e
2
0
0
9
.
Hi
s
re
se
a
rc
h
in
tere
st
in
c
l
u
d
e
s
s
o
f
t
c
o
m
p
u
ti
n
g
,
e
-
lea
rn
in
g
,
d
a
ta m
in
in
g
a
n
d
in
f
o
rm
a
ti
o
n
se
c
u
rit
y
.
Bil
k
isu
Jim
a
d
a
-
Oju
o
lap
e
re
c
e
iv
e
d
h
e
r
B
.
S
c
d
e
g
re
e
in
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
a
n
d
El
e
c
tro
n
ic
En
g
in
e
e
rin
g
f
ro
m
th
e
K
w
a
m
e
N
k
ru
m
a
h
Un
iv
e
rsit
y
o
f
S
c
ien
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
,
Ku
m
a
si,
G
h
a
n
a
in
2
0
1
1
,
a
n
d
h
e
r
M
.
S
c
De
g
re
e
in
S
y
ste
m
s
En
g
in
e
e
rin
g
,
L
o
u
g
h
b
o
r
o
h
U
n
iv
e
rsity
,
L
o
u
g
h
b
o
r
o
u
g
h
,
Un
it
e
d
Kin
g
d
o
m
in
2
0
1
3
.
S
h
e
i
s
c
u
rre
n
tl
y
p
u
rsu
in
g
h
e
r
P
h
D
d
e
g
re
e
w
it
h
th
e
S
c
h
o
o
l
o
f
El
e
c
tri
c
a
l
a
n
d
El
e
c
tro
n
ic E
n
g
in
e
e
rin
g
,
Un
iv
e
rsit
y
S
a
in
s M
a
lay
si
a
.
S
h
e
w
o
rk
e
d
w
it
h
th
e
Nig
e
rian
El
e
c
tri
c
it
y
Re
g
u
lato
ry
Co
m
issio
n
(NERC)
f
o
r
a
b
o
u
t
a
y
e
a
r,
w
h
e
re
sh
e
w
a
s
in
v
o
lv
e
d
in
li
c
e
n
se
a
p
p
li
c
a
ti
o
n
e
v
a
lu
a
ti
o
n
f
o
r
In
d
e
p
e
n
d
e
n
t
P
o
w
e
r
P
ro
d
u
c
e
rs
a
n
d
In
sp
e
c
ti
o
n
o
f
su
b
sta
ti
o
n
s
w
it
h
in
A
b
u
ja,
Nig
e
ria,
a
m
o
n
g
st
o
th
e
r
d
u
ti
e
s.
S
h
e
w
a
s
a
lso
a
Na
ti
o
n
a
l
S
y
st
e
m
En
g
in
e
e
r
w
it
h
t
h
e
NT
A
-
S
tar
T
V
Ne
tw
o
rk
(S
tarT
im
e
s),
w
h
e
re
sh
e
m
a
n
a
g
e
d
th
e
o
p
e
ra
ti
o
n
s
o
f
ten
c
it
ies
a
n
d
w
a
s
a
ls
o
p
a
rt
o
f
a
th
i
n
k
tan
k
c
o
m
m
it
tee
th
a
t
w
a
s
b
ra
in
sto
rm
in
g
n
e
w
p
ro
d
u
c
ts
f
o
r
th
e
c
o
m
p
a
n
y
.
S
h
e
jo
i
n
e
d
Kw
a
ra
S
tate
Un
iv
e
rs
it
y
(K
WA
S
U)
in
2
0
1
5
.
He
r
c
u
r
re
n
t
re
se
a
rc
h
in
tere
sts
a
re
in
re
n
e
wa
b
le
e
n
e
rg
y
,
p
o
w
e
r
sy
ste
m
s,
sm
a
rt
g
rid
re
li
a
b
il
it
y
,
ICT
s
in
sm
a
rt
g
rid
,
a
n
d
sm
a
ll
h
y
d
ro
p
o
w
e
r
g
e
n
e
ra
ti
o
n
,
M
a
c
h
i
n
e
L
e
a
rn
in
g
.
M
r
s.
Jim
a
d
a
-
Oju
o
lap
e
is
a
m
e
m
b
e
r
o
f
th
e
As
so
c
iatio
n
o
f
P
ra
c
ti
c
in
g
W
o
m
e
n
En
g
in
e
e
rs
in
Nig
e
ria
(
A
P
W
EN)
a
n
d
t
h
e
Nig
e
rian
S
o
c
i
e
t
y
o
f
En
g
i
-
n
e
e
rs
(NSE
)
a
n
d
c
e
rti
f
ied
b
y
th
e
Co
u
n
c
il
f
o
r
th
e
R
e
g
u
latio
n
o
f
En
g
in
e
e
rin
g
in
Nig
e
r
ia (COREN).
M
u
d
a
sh
i
ru
L
a
tee
f
Olu
m
id
e
B.
sc
h
o
l
d
e
r
a
n
d
a
n
in
tel
Re
tail
c
e
rti
f
ied
sp
e
c
ialist,
h
e
is
a
s
sista
n
t
M
a
n
a
g
e
r
a
t
Na
ti
o
n
a
l
i
d
e
n
ti
ty
m
a
n
a
g
e
m
e
n
t
c
o
m
m
issio
n
(NIM
C).
He
is
p
re
se
n
tl
y
a
n
M
.
S
c
S
t
u
d
e
n
t
o
f
Un
iv
e
rsiti
S
a
i
n
s
M
a
lay
sia
m
a
jo
rin
g
i
n
I
n
f
o
rm
a
ti
c
s
w
it
h
a
r
e
se
a
rc
h
in
tere
st
in
B
u
sin
e
ss
In
telli
g
e
n
c
e
a
n
d
Da
ta W
a
re
h
o
u
sin
g
.
Ka
z
e
e
m
A
.
G
b
o
lag
a
d
e
,
A
P
ro
f
e
ss
o
r
a
n
d
P
ro
v
o
st
a
t
th
e
C
o
ll
e
g
e
o
f
Co
m
p
u
ter
in
I
n
f
o
rm
a
ti
o
n
S
c
ien
c
e
,
Kw
a
ra
S
tate
Un
iv
e
rsit
y
,
M
a
lete
,
Nig
e
ria.
w
a
s
b
o
rn
in
Iw
o
(Os
u
n
S
tate
),
Nig
e
ria,
o
n
th
e
2
7
t
h
o
f
A
u
g
u
st,
1
9
7
4
.
He
re
c
e
iv
e
d
h
is
B.
S
c
d
e
g
re
e
in
2
0
0
0
i
n
Co
m
p
u
ter
S
c
ien
c
e
f
ro
m
th
e
Un
iv
e
rsit
y
o
f
Ilo
rin
,
Kw
a
ra
S
tat
e
,
Nig
e
ria.
In
2
0
0
4
,
h
e
o
b
tain
e
d
h
is
M
a
ste
rs
d
e
g
re
e
f
ro
m
th
e
Un
iv
e
rsit
y
o
f
Ib
a
d
a
n
,
Nig
e
ria.
In
A
p
ril
2
0
0
7
,
h
e
j
o
in
e
d
th
e
Co
m
p
u
ter
E
n
g
in
e
e
rin
g
L
a
b
o
ra
to
ry
g
ro
u
p
a
t
t
h
e
De
lf
t
Un
iv
e
rsity
o
f
Tec
h
n
o
l
o
g
y
(
T
U
De
lf
t),
T
h
e
Ne
th
e
rlan
d
s.
In
T
U
De
lf
t,
h
e
p
u
rs
u
e
d
a
P
h
D
d
e
g
re
e
u
n
d
e
r
t
h
e
su
p
e
rv
isio
n
o
f
P
r
o
f
.
S
o
rin
Co
t
o
f
a
n
a
.
He
is
a
m
e
m
b
e
r
o
f
th
e
IEE
E.
His
re
se
a
rc
h
in
tere
sts in
c
lu
d
e
Dig
it
a
l
L
o
g
ic De
si
g
n
,
Co
m
p
u
ter A
rit
h
m
e
ti
c
,
Re
sid
u
e
N
u
m
b
e
r
S
y
st
e
m
s,
V
L
S
I
De
sig
n
,
a
n
d
N
u
m
e
rica
l
Co
m
p
u
ti
n
g
.
His
re
se
a
rc
h
in
tere
sts
i
n
c
lu
d
e
Dig
it
a
l
L
o
g
ic
De
sig
n
,
Co
m
p
u
ter A
rit
h
m
e
ti
c
,
Re
sid
u
e
Nu
m
b
e
r
S
y
ste
m
s,
V
L
S
I
De
sig
n
,
a
n
d
Nu
m
e
rica
l
Co
m
p
u
ti
n
g
Evaluation Warning : The document was created with Spire.PDF for Python.