I
n
t
ern
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
A
d
v
a
n
ces
i
n
A
p
p
l
i
ed
S
ci
en
ces
(
I
J
A
A
S
)
V
o
l
.
7
,
N
o
.
1
,
M
ar
ch
20
1
8
,
p
p
.
7~
12
I
S
S
N
:
225
2
-
88
14
,
D
O
I
:
10.
115
91
/ij
a
a
s
.
v7
.
i
1
.
p
p7
-
12
7
Jo
u
r
n
al
h
om
e
pa
ge
:
h
t
t
p
:
/
/
i
a
e
s
co
r
e
.
c
o
m/
o
n
l
i
n
e
/
i
n
d
e
x
.
p
h
p
/
I
J
A
A
S
D
a
t
a
M
i
ni
ng
T
echni
que
s
f
or
P
r
ovi
di
ng
N
et
w
or
k Sec
ur
i
t
y
t
h
rou
g
h
I
nt
r
us
i
on D
e
t
ec
t
i
o
n Sy
s
t
ems
:
a
Sur
vey
P
r
a
b
hu
K
a
v
i
n
B
,
G
a
na
pa
t
hy
S
S
c
hool
of Com
p
ut
i
ng S
c
i
e
nc
e
a
n
d E
ngi
n
e
e
ri
ng
,
V
IT
U
ni
ve
rs
i
t
y
-
Che
nna
i
Ca
m
pus
,
Che
nna
i
,
Indi
a
A
rt
i
cl
e
I
n
f
o
A
BS
TR
A
C
T
Ar
t
i
c
l
e
h
i
s
t
o
r
y
:
R
ecei
v
ed
Ju
n
1
,
201
7
Re
v
i
s
e
d
J
a
n
5
,
2
01
8
A
ccep
t
e
d
F
eb
11
,
2
01
8
Int
rus
i
on D
e
t
e
c
t
i
on S
y
s
t
e
m
s
a
re
pl
a
y
i
ng m
a
j
or rol
e
i
n ne
t
w
ork
s
e
c
uri
t
y
i
n
t
hi
s
i
nt
e
rn
e
t
w
orl
d.
M
a
n
y
r
e
s
e
a
rc
h
e
rs
ha
ve
be
e
n i
nt
rodu
c
e
d
num
be
r of
i
nt
rus
i
on de
t
e
c
t
i
on s
y
s
t
e
m
s
i
n t
he
pa
s
t
.
E
v
e
n t
ho
ugh,
no s
y
s
t
e
m
w
a
s
de
t
e
c
t
e
d
a
l
l
k
i
nd of
a
t
t
a
c
ks
a
nd a
c
hi
e
v
e
d be
t
t
e
r d
e
t
e
c
t
i
on a
c
c
ura
c
y
.
M
os
t
of t
he
i
nt
rus
i
on de
t
e
c
t
i
on s
y
s
t
e
m
s
a
re
us
e
d da
t
a
m
i
ni
ng t
e
c
hn
i
que
s
s
uc
h
as
c
lu
s
te
r
in
g
,
o
u
tli
e
r
d
e
te
c
tio
n
,
c
l
a
s
s
if
ic
a
tio
n
,
c
la
s
s
if
ic
a
tio
n
th
r
o
u
g
h
le
a
r
n
in
g
t
e
c
hn
i
que
s
.
M
o
s
t
of t
he
re
s
e
a
rc
he
rs
ha
ve
b
e
e
n a
ppl
i
e
d s
oft
c
om
put
i
ng
t
e
c
hn
i
que
s
for m
a
ki
ng e
ffe
c
t
i
v
e
de
c
i
s
i
on ove
r
t
he
ne
t
w
ork da
t
a
s
e
t
for
e
nha
nc
i
ng t
he
de
t
e
c
t
i
on a
c
c
ur
a
c
y
i
n Int
rus
i
o
n D
e
t
e
c
t
i
on S
y
s
t
e
m
.
F
e
w
r
es
ear
ch
er
s
al
s
o
ap
p
l
i
ed
ar
t
i
f
i
ci
al
i
n
t
el
l
i
g
en
ce
t
ech
n
i
q
u
es
al
o
n
g
w
i
t
h
d
at
a
m
i
ni
ng a
l
gori
t
h
m
s
for
m
a
ki
ng d
y
na
m
i
c
de
c
i
s
i
on
.
T
hi
s
pa
pe
r di
s
c
us
s
e
s
a
bout
t
he
num
be
r of i
nt
rus
i
on de
t
e
c
t
i
on s
y
s
t
e
m
s
t
ha
t
a
re
propos
e
d for provi
di
n
g
n
et
w
o
r
k
s
ecu
r
i
t
y
. F
i
n
al
l
y
,
co
m
p
ar
at
i
v
e an
al
y
s
i
s
m
ad
e b
et
w
een
t
h
e ex
i
s
t
i
n
g
s
y
s
t
e
m
s
a
nd s
ugge
s
t
e
d s
om
e
ne
w
i
de
a
s
for e
nha
nc
i
ng t
he
pe
r
for
m
a
nc
e
of t
he
e
x
is
tin
g
s
y
s
t
e
m
s
.
Ke
y
wo
r
d
s
:
A
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
Cl
a
s
s
i
f
i
c
a
t
i
o
n
Cl
u
s
t
e
r
i
n
g
D
a
t
a
m
i
n
i
n
g
I
nt
r
us
i
on
de
t
e
c
t
i
on
s
y
s
t
e
m
S
of
t
c
om
put
i
n
g
Copy
r
i
ght
©
201
8
Ins
t
i
t
ut
e
o
f
A
d
v
anc
e
d
E
ngi
n
e
e
r
i
ng and S
c
i
e
nc
e
.
A
l
l
ri
g
h
t
s re
se
rv
e
d
.
C
or
r
e
s
po
n
di
n
g
A
u
t
h
or
:
P
r
a
bh
u
K
a
vi
n
B
,
S
c
h
ool
o
f
C
om
put
i
n
g
S
c
i
e
nc
e
a
n
d
E
n
gi
ne
e
r
i
ng
,
V
I
T
U
n
i
v
e
r
s
i
t
y
-
C
he
nna
i
C
a
m
pus
,
Ch
e
n
n
a
i
,
60
0
12
7
-
In
d
i
a
.
E
m
a
i
l
:
l
s
n
t
l
@
c
c
u
.
e
d
u
.
t
w
1.
I
N
T
R
O
D
U
C
T
I
O
N
N
o
w a
da
y
’
s
i
nt
e
r
ne
t
ha
s
be
c
om
e
a
pa
r
t
o
f
ou
r
l
i
f
e
.
T
he
r
e
a
r
e
m
a
ny
s
i
gni
f
i
c
a
nt
l
o
s
s
e
s
i
n
pr
e
s
e
nt
i
n
t
e
r
n
e
t
-
ba
s
e
d
i
nf
or
m
a
t
i
on p
r
oc
e
s
s
i
n
g s
y
s
t
e
m
s
.
S
o,
t
he
i
m
por
t
a
nc
e
of
t
he
i
nf
o
r
m
a
ti
on s
e
c
ur
i
t
y
h
as
b
ee
n
i
nc
r
e
a
s
e
d.
T
he
one
a
nd
o
nl
y
ba
s
i
c
m
ot
t
o
of
t
he
i
nf
or
m
a
t
i
on s
e
c
u
r
i
t
y
s
y
s
t
e
m
i
s
t
o de
ve
l
o
pe
d i
nf
or
m
a
t
i
on
de
f
e
ns
i
ve
s
y
s
t
e
m
t
o whi
c
h a
r
e
s
e
c
ur
e
d f
r
o
m
a
n u
nj
us
t
i
f
i
e
d a
c
c
e
s
s
,
r
e
v
e
l
a
t
i
on,
i
nt
e
r
f
e
r
e
nc
e
a
n
d a
l
t
e
r
a
t
i
on
.
M
o
re
o
v
e
r,
t
h
e
ri
s
k
s
w
e
re
re
l
a
t
e
d
t
o
t
h
e
c
o
n
f
i
d
e
n
t
i
a
l
i
t
y
,
p
r
o
b
i
t
y
a
n
d
a
v
a
i
l
a
b
i
l
i
t
y
w
i
l
l
h
a
v
e
b
e
e
n
m
i
n
i
m
i
z
e
d
.
T
h
e i
n
t
er
n
et
-
b
a
s
e
d a
t
t
a
c
ks
w
e
r
e
i
d
e
nt
i
f
i
e
d a
n
d
bl
oc
ke
d
us
i
n
g di
f
f
e
r
e
nt
s
y
s
t
e
m
s
t
ha
t
ha
ve
be
e
n
de
s
i
g
ne
d i
n
t
he
pa
s
t
.
T
he
I
nt
r
us
i
o
n
de
t
e
c
t
i
on
s
y
s
t
e
m
(
I
D
S
)
i
s
o
ne
of
t
he
m
os
t
i
m
por
t
a
nt
s
y
s
t
e
m
s
a
m
ong t
he
m
b
ecau
s
e t
h
ey
r
es
i
s
t
ex
t
er
n
al
a
t
t
ack
s
ef
f
ect
i
v
e
l
y
.
M
o
r
eo
v
e
r
,
t
h
e I
D
S
act
as
t
h
e w
al
l
o
f
d
ef
e
n
s
e t
o
t
h
e co
m
p
u
t
e
r
s
y
s
t
e
m
s
o
v
e
r
t
h
e
a
t
t
a
c
k
o
n
t
h
e
i
n
t
e
r
n
e
t
.
T
h
e
t
r
a
d
i
t
i
o
n
a
l
f
i
r
e
w
a
l
l
d
e
t
e
c
t
s
t
h
e
i
n
t
r
u
s
i
o
n
o
n
t
h
e
s
y
s
t
e
m
b
u
t
t
h
e
ID
S
p
e
r
fo
rm
a
n
c
e
i
s
m
uc
h
b
e
t
t
e
r
t
ha
n
t
he
f
i
r
e
w
a
l
l
s
pe
r
f
o
r
m
a
nc
e
.
Us
ua
l
l
y
t
he
be
ha
vi
o
r
of
t
he
i
nt
r
ud
e
r
s
i
s
di
f
f
e
r
e
nt
f
r
om
t
he
no
r
m
a
l
be
ha
vi
or
of
t
he
l
e
ga
l
us
e
r
,
de
p
e
ndi
ng
u
po
n t
he
be
ha
vi
or
t
h
e
a
s
s
um
pt
i
on i
s
m
a
de
a
nd t
he
i
nt
r
us
i
on
de
t
e
c
t
i
on i
s
do
ne
[
1]
.
T
he
c
om
put
e
r
s
y
s
t
e
m
f
i
l
e
s
,
c
a
l
l
s
,
l
ogs
,
a
n
d t
he
ne
t
w
or
k e
ve
nt
s
a
r
e
m
oni
t
or
e
d
by
t
he
I
D
S
t
o i
d
e
nt
i
f
y
t
he
t
h
r
e
a
t
s
on t
he
h
os
t
s
of
t
he
c
om
put
e
r
.
B
y
m
o
ni
t
or
i
ng t
he
n
e
t
w
o
r
k
poc
ke
t
s
,
t
he
a
bn
o
r
m
a
l
be
ha
vi
o
r
i
s
de
t
e
c
t
e
d.
T
he
a
t
t
a
c
k
pa
t
t
e
r
n i
s
k
no
wn
by
f
i
ndi
n
g
t
he
p
os
s
i
bl
e
a
t
t
a
c
k
s
i
gna
t
ur
e
s
a
n
d
co
m
p
ar
i
n
g
t
h
e
m
.
B
y
t
h
e k
n
o
w
n
at
t
ack
s
i
g
n
at
u
r
e,
t
h
e t
h
r
ea
t
s
ar
e d
et
ect
ed
eas
i
l
y
b
y
t
h
e s
y
s
t
e
m
w
h
e
r
e
as
i
t
can
n
o
t
d
et
ect
t
h
e
u
n
k
n
o
w
n
at
t
ack
s
[
2
]
.
A
n
i
n
t
e
l
l
i
g
e
n
t
I
D
S
a
c
t
s
f
l
e
x
i
b
l
e
t
o
i
n
c
r
e
a
s
e
t
h
e
a
c
c
u
r
a
c
y
r
a
t
e
o
f
t
h
e
d
e
t
e
c
t
i
o
n
.
I
n
t
e
l
l
i
g
e
n
t
I
D
S
a
r
e
not
hi
n
g
b
ut
i
nt
e
l
l
i
ge
nt
c
om
pu
t
e
r
pr
o
gr
a
m
s
t
ha
t
a
r
e
l
oc
a
t
e
d
i
n
h
os
t
or
ne
t
w
or
k.
B
y
f
i
r
i
n
g
t
he
r
ul
e
s
of
i
nf
e
r
e
nc
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
252
-
88
1
4
IJ
A
A
S
V
o
l
.
7
,
N
o
.
1
,
M
a
r
ch
2
018
:
7
–
12
8
a
nd
a
l
s
o
by
l
e
a
r
ni
ng
t
he
e
n
vi
r
onm
e
nt
,
t
he
a
c
t
i
ons
t
o
be
pe
r
f
or
m
e
d
a
r
e
c
om
put
e
d
by
t
he
i
n
t
e
l
l
i
ge
nt
I
D
S
o
n
t
ha
t
e
nvi
r
o
nm
e
nt
[
1]
.
T
he
r
e
g
ul
a
r
ne
t
w
or
k
s
e
r
vi
c
e
i
s
di
s
r
u
pt
e
d
by
t
r
a
n
s
m
it
t
i
ng
t
he
l
a
r
ge
a
m
ount
of
da
t
a
t
o
e
xe
c
ut
e
a l
o
w
er
l
ev
el
d
en
i
al
o
f
s
er
v
i
c
e at
t
ack
s
.
T
o
cau
s
e a d
en
i
al
o
f
s
er
v
i
ce at
t
ack
t
o
t
h
e u
s
er
,
t
h
e r
ecei
v
er
’
s
n
e
t
w
o
r
k
co
n
n
ect
i
v
i
t
y
w
as
o
v
er
w
h
el
m
ed
b
y
c
r
eat
i
n
g
a s
p
eci
f
i
c s
e
r
v
i
ce r
eq
u
es
t
o
r
b
y
s
en
d
i
n
g
a l
ar
g
e am
o
u
n
t
o
f
d
at
a.
T
he
i
ni
t
i
a
t
i
on
of
a
t
t
a
c
k
w
a
s
do
ne
by
a
s
i
ng
l
e
s
e
nde
r
o
r
t
h
e
c
om
pr
om
i
s
e
d h
os
t
s
by
t
he
a
t
t
a
c
ke
r
a
n
d f
r
om
t
h
e
l
a
t
t
e
r
v
a
r
i
a
n
t
w
i
l
l
i
d
e
n
t
i
f
y
t
h
e
D
i
s
t
r
i
b
u
t
e
d
D
e
n
i
a
l
o
f
S
e
r
v
i
c
e
(
D
D
o
S
)
[
3
]
.
T
h
e
I
D
S
w
o
r
k
s
a
m
e
a
s
t
h
e
T
r
a
ns
p
a
r
e
n
t I
ntr
us
i
o
n De
t
e
c
t
i
on S
y
s
t
e
m
(
T
I
D
S
)
a
n
d f
o
r
t
he
no
n
-
d
i
s
t
r
i
b
u
t
e
d
a
t
t
a
c
k
s
t
h
e
f
u
n
c
t
i
o
n
a
l
i
t
y
t
o
p
r
e
v
e
n
t
t
h
e at
t
ack
ar
e p
r
o
v
i
d
ed
.
T
h
e s
cal
ab
i
l
i
t
y
o
f
t
h
e t
r
a
f
f
i
c p
r
o
ce
s
s
o
r
i
s
ach
i
ev
ed
b
y
t
h
e l
o
ad
b
al
a
n
ci
n
g
al
g
o
r
i
t
h
m
an
d
t
h
e s
y
s
t
e
m
s
ecu
r
i
t
y
i
s
ach
i
ev
ed
b
y
t
h
e t
r
an
s
p
ar
e
nc
y
o
f
no
de
s
.
T
he
m
e
t
hod
ol
o
gy
o
f
a
n
om
a
ly
-
ba
s
e
d a
t
t
a
c
k d
e
t
e
c
t
i
on i
s
us
e
d i
n hi
g
h s
pe
e
d ne
t
w
o
r
k t
o
de
t
e
c
t
DD
oS
a
t
t
a
c
ks
,
i
n t
h
i
s
m
e
t
hod t
he
S
D
N
c
om
pone
nt
s
a
r
e
c
o
upl
e
d
w
i
t
h
t
r
a
f
f
i
c
pr
oc
e
s
s
or
[
3]
.
A
m
o
n
g
m
an
y
cy
b
er
t
h
r
eat
s
B
o
t
n
et
at
t
ack
w
as
o
n
e t
h
e m
o
s
t
s
ev
er
e
cy
b
er
t
h
r
eat
.
I
n
t
h
i
s
at
t
ack
bot
m
a
s
t
e
r
i
s
a
c
ont
r
ol
l
i
n
g c
o
m
put
e
r
t
ha
t
c
o
m
pr
om
i
s
e
d a
n
d r
e
m
ot
e
c
ont
r
ol
l
e
d.
Hu
ge
n
um
be
r
s
o
f
b
ot
s
we
r
e
s
pr
e
a
d o
ve
r
t
h
e
i
nt
e
r
ne
t
a
n
d
t
he
bot
m
a
s
t
e
r
us
e
s
t
he
bot
ne
t
by
m
a
i
nt
a
i
ni
ng u
nde
r
i
t
s
c
o
nt
r
ol
.
T
he
bot
n
e
t
w
a
s
us
e
d
f
o
r
va
r
i
o
us
p
ur
po
s
e
s
b
y
t
he
bot
m
a
s
t
e
r
,
i
n t
ha
t
f
e
w
a
r
e
l
a
u
nc
hi
n
g
a
nd
p
e
r
f
o
r
m
i
n
g
of
di
s
t
r
i
but
e
d c
y
b
e
r
a
t
t
a
c
ks
a
nd c
o
m
put
a
t
i
ona
l
t
a
s
ks
.
T
he
I
D
S
bui
l
t
f
o
r
b
ot
ne
t
s
a
r
e
r
ul
e
ba
s
e
d a
n
d pe
r
f
or
m
a
nc
e
de
pe
nd
a
nt
.
B
y
e
xa
m
i
ni
ng t
he
ne
t
w
or
k t
r
a
f
f
i
c
a
nd c
om
pa
r
i
ng
w
i
t
h
kn
ow
n b
ot
ne
t
s
i
g
na
t
ur
e
t
he
b
ot
ne
t
w
a
s
f
o
un
d i
n
a
r
ul
e
-
ba
s
e
d
bot
ne
t
I
D
S
.
H
o
we
ve
r
,
ke
e
pi
n
g t
h
e
s
e
r
ul
e
s
u
p
da
t
e
d
i
n t
he
i
nc
r
e
a
s
i
ng
ne
t
w
o
r
k
t
r
a
f
f
i
c
i
s
m
or
e
t
e
di
o
us
,
d
i
f
f
i
c
u
l
t
a
n
d
t
i
m
e
-
c
ons
um
i
ng [
4]
.
M
a
c
hi
ne
-
l
e
a
r
ni
n
g
(
M
L
)
t
e
c
hni
q
ue
i
s
a
t
e
c
hni
q
ue
us
e
d
t
o a
ut
om
a
t
e
bot
ne
t
d
et
ec
t
i
on
pr
oc
e
s
s
.
F
r
om
pr
e
v
i
ous
l
y
k
no
w
n a
t
t
a
c
k s
i
gna
t
u
r
e
s
a
m
ode
l
w
a
s
bui
l
t
by
t
he
l
e
a
r
ni
ng s
y
s
t
e
m
.
T
he
f
e
a
t
u
r
e
s
l
i
k
e
f
l
e
x
i
b
i
l
i
t
y
,
a
d
a
p
t
a
b
i
l
i
t
y
a
n
d
a
u
t
o
m
a
t
e
d
-
l
e
a
r
n
i
n
g
a
b
i
l
i
t
y
o
f
M
L
i
s
s
i
g
n
i
f
i
c
a
n
t
l
y
b
e
t
t
e
r
t
h
a
n
t
h
e
r
u
l
e
-
ba
s
e
d
I
D
S
s
.
Hi
gh
c
om
put
a
t
i
ona
l
c
os
t
i
s
ne
ed
e
d
f
o
r
t
h
e
m
ach
i
n
e
l
ear
n
i
n
g
b
a
s
ed
ap
p
r
o
ach
es
[
4
]
.
I
n t
hi
s
pa
pe
r
,
w
e
ha
ve
di
s
c
u
s
s
e
d a
b
o
ut
t
he
va
r
i
ous
t
y
pe
s
of
I
nt
r
us
i
on
D
e
t
e
c
t
i
on S
y
s
t
e
m
s
whi
c
h a
r
e
us
e
d
da
t
a
m
i
ning
t
e
c
hni
que
s
.
R
e
s
t
of
t
hi
s
pa
pe
r
i
s
or
ga
ni
z
e
d
a
s
f
ol
l
o
ws
:
S
e
c
t
i
on
2
p
r
ovi
de
s
t
he
r
e
l
a
t
e
d
w
or
ks
i
n t
hi
s
di
r
e
c
t
i
o
n.
S
e
c
t
i
o
n
3 s
ho
w
s
t
h
e
c
om
pa
r
a
t
i
ve
a
na
l
y
s
i
s
.
S
e
c
t
i
on
4
s
ug
ge
s
t
s
ne
w
i
de
a
s
t
o i
m
pr
o
ve
t
he
pe
r
f
o
r
m
a
nc
e
of
t
he
e
xi
s
t
i
ng
s
y
s
t
e
m
s
.
S
e
c
t
i
on
5
c
onc
l
ude
s
t
h
e
pa
pe
r
.
2.
R
ELA
TED
W
O
R
K
S
T
hi
s
s
e
c
t
i
on i
s
c
l
a
s
s
i
f
i
e
d i
n
t
o t
w
o m
a
jor
s
ubs
e
c
t
i
ons
f
or
f
e
a
t
u
r
e s
el
ect
i
o
n
an
d
cl
as
s
i
f
i
cat
i
o
n
t
e
c
hni
q
ue
s
w
hi
c
h
a
r
e
pr
op
os
e
d
i
n
t
hi
s
di
r
e
c
t
i
on
i
n
t
he
pa
s
t
.
2
.
1
.
R
e
l
a
te
d
W
o
r
ks
o
n
F
e
a
t
u
re
S
el
ect
i
o
n
M
e
t
ho
ds
F
e
a
t
ur
e
s
e
l
e
c
t
i
on
w
a
s
t
he
m
os
t
f
a
m
ous
t
e
c
hni
q
ue
f
or
di
m
e
ns
i
o
na
l
i
t
y
r
e
du
c
t
i
on.
I
n t
hi
s
t
he
r
e
l
e
va
n
t
f
e
a
t
ur
e
s
i
s
o
f
de
t
e
c
t
e
d a
nd t
he
i
r
r
e
l
e
va
nt
one
s
a
r
e
di
s
c
a
r
de
d
[5
]
.
F
r
o
m
t
h
e en
t
i
r
e d
at
as
et
t
h
e p
r
o
ces
s
o
f
s
el
ect
i
n
g
a
f
ea
t
u
r
e s
u
b
s
et
f
o
r
f
u
r
t
h
e
r
p
r
o
ce
s
s
i
n
g
w
a
s
p
r
o
ceed
ed
i
n
f
eat
u
r
e
s
el
ect
i
o
n
[6
]
.
F
eat
u
r
e
s
el
ect
i
o
n
m
e
t
hods
a
r
e
c
l
a
s
s
i
f
i
e
d
i
nt
o
t
w
o
t
y
pe
s
,
i
n
di
vi
dua
l
e
va
l
ua
t
i
o
n
a
nd
s
u
bs
e
t
e
va
l
ua
t
i
o
n.
A
c
c
or
di
n
g
t
o
t
he
i
r
de
g
r
e
e
s
of
i
m
por
t
a
nc
e
f
e
a
t
ur
e
r
a
n
ki
n
g m
e
t
hods
e
s
t
i
m
a
t
e
f
e
a
t
ur
e
s
a
nd a
l
l
ot
w
e
i
g
ht
s
f
o
r
t
he
m
.
I
n c
o
nt
r
a
s
t
,
bui
l
d on
a
s
o
m
e
s
ear
ch
m
et
h
o
d
s
u
b
s
et
ev
al
u
at
i
o
n
m
et
h
o
d
s
s
el
ect
can
d
i
d
at
e f
eat
u
r
e
[4
]
.
F
eat
u
r
e
s
el
ect
i
o
n
m
et
h
o
d
s
i
s
di
vi
de
d i
nt
o t
hr
e
e
m
e
t
ho
ds
t
he
y
a
r
e
w
r
a
p
pe
r
s
m
e
t
hods
,
f
i
l
t
e
r
s
m
e
t
hods
a
nd
e
m
be
dde
d
m
e
t
hods
[5
]
.
A
n
i
n
t
e
l
l
i
g
e
n
t
c
o
n
d
i
t
i
o
n
a
l
r
a
n
d
o
m
f
i
e
l
d
-
b
as
ed
f
eat
u
r
e
s
el
ect
i
o
n
al
g
o
r
i
t
h
m
h
as
b
ee
n
pr
opo
s
e
d
i
n
[
7
] f
o
r
e
f
f
ect
i
v
e
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
.
T
h
i
s
w
i
l
l
b
e
h
e
l
p
f
u
l
f
o
r
i
m
p
r
o
v
i
n
g
t
h
e
c
l
a
s
s
i
f
i
c
a
t
i
o
n
a
c
c
u
r
a
c
y
.
I
n
w
r
a
p
pe
r
m
e
t
ho
d
opt
i
m
i
z
a
ti
on
of
a
p
r
e
di
c
t
or
i
s
i
n
v
ol
ve
d a
s
a
s
e
gm
e
nt
of
t
he
s
e
l
e
c
t
i
on
p
r
oc
e
s
s
,
w
he
r
e
a
s
i
n
f
i
l
t
e
r
m
e
t
hod s
e
l
e
c
t
i
on t
he
f
e
a
t
ur
e
s
w
i
t
h
s
e
l
f
de
t
e
r
m
i
na
t
i
on
of
a
ny
pr
e
di
c
t
or
by
r
e
l
y
i
ng
on
t
he
g
en
e
r
al
ch
ar
act
er
i
s
t
i
cs
o
f
t
h
e t
r
ai
n
i
n
g
d
at
a i
s
d
o
n
e
.
I
n e
m
be
dde
d m
e
t
hods
f
o
r
c
l
a
s
s
i
f
i
c
a
t
i
on m
a
c
hi
ne
l
e
a
r
ni
ng
m
ode
l
s
wa
s
ge
ne
r
a
l
l
y
us
e
d,
a
n
d t
he
n t
he
c
l
a
s
s
i
f
i
e
r
a
l
go
r
i
t
hm
bui
l
ds
a
n o
pt
i
m
a
l
s
ubs
e
t
or
r
a
nki
ng
f
eat
u
r
e
s
[5
]
.
W
r
a
p
pe
r
s
m
e
tho
d
a
n
d e
m
be
dde
d
m
e
t
hod t
r
i
e
d t
o
pe
r
f
or
m
be
t
t
e
r
but
h
a
vi
n
g t
he
r
i
s
k
o
f
ov
e
r
f
i
t
t
i
ng w
he
n t
h
e
s
a
m
pl
e
s
i
z
e
is
s
m
a
l
l
a
nd be
i
ng
ve
r
y
t
i
m
e
c
on
s
um
i
ng.
O
n t
he
ot
he
r
ha
nd
,
f
i
l
t
e
r
m
e
t
h
od
wa
s
m
o
r
e s
u
i
t
ab
l
e
f
o
r
l
ar
g
e
d
at
as
et
s
an
d
m
uc
h f
a
s
t
e
r
.
C
om
pa
r
i
ng w
i
t
h
w
r
a
p
pe
r
s
a
n
d e
m
be
dde
d
m
e
t
ho
ds
f
i
l
t
e
r
s
w
er
e i
m
p
l
e
m
e
n
t
ed
ea
s
i
l
y
an
d
h
as
b
et
t
er
s
cal
e u
p
t
h
a
n
t
h
o
s
e m
et
h
o
d
s
.
F
i
l
t
er
can
b
e ab
l
e t
o
u
s
e
as
a
pr
e
pr
oc
e
s
s
i
ng
s
t
e
p p
r
i
o
r
t
r
y
i
ng t
o
ot
he
r
c
o
m
pl
e
x f
e
a
t
ur
e
s
e
l
e
c
t
i
on m
e
t
hods
.
T
he
t
w
o
m
e
t
r
i
c
s
o
f
t
h
e
f
i
l
t
e
r
m
e
t
hods
i
n
c
l
a
s
s
i
f
i
c
a
t
i
on
p
r
o
bl
e
m
s
a
r
e
c
or
r
e
l
a
t
i
on
a
nd
m
ut
ua
l
i
nf
o
r
m
a
ti
on,
a
l
o
ng
w
i
t
h
s
om
e
ot
he
r
m
e
tr
i
c
s
o
f
t
h
e
f
i
l
t
e
r
m
e
t
h
o
d
l
i
k
e
e
r
r
o
r
p
r
o
b
a
b
i
l
i
t
y
,
p
r
o
b
a
b
i
l
i
s
t
i
c
d
i
s
t
a
n
c
e
,
e
n
t
r
o
p
y
o
r
c
o
n
s
i
s
t
e
n
c
y
[5
]
.
In
wr
a
p
p
e
r
ap
p
r
o
ach
b
as
e
d
o
n
s
p
eci
f
i
e
d
l
ear
n
i
n
g
al
g
o
r
i
t
h
m
i
t
s
el
ect
s
a f
eat
u
r
e
s
u
b
s
et
w
i
t
h
a h
i
g
h
e
r
p
r
e
d
i
ct
i
o
n
pe
r
f
o
r
m
a
nc
e
.
I
n
e
m
be
dde
d
a
s
s
im
i
l
a
r
a
s
wr
a
ppe
r
a
pp
r
oa
c
h
du
r
i
n
g
t
he
l
e
a
r
ni
n
g
p
r
o
ces
s
o
f
a
s
p
eci
f
i
ed
l
e
ar
n
i
n
g
a
l
g
o
r
i
t
h
m
i
t
s
e
l
e
c
t
s
t
h
e
b
e
s
t
f
e
a
t
u
r
e
s
u
b
s
e
t
.
I
n
t
h
e
f
i
l
t
e
r
a
p
p
r
o
a
c
h
t
h
e
f
e
a
t
u
r
e
s
u
b
s
e
t
i
s
c
h
o
s
e
n
f
r
o
m
t
h
e
o
r
i
g
i
n
a
l
f
eat
u
r
e
s
p
a
ce acco
r
d
i
n
g
t
o
p
r
e
-
s
p
e
c
i
f
i
e
d
e
v
a
l
u
a
t
i
o
n
c
r
i
t
e
r
i
o
n
s
s
u
b
s
e
t
u
s
i
n
g
o
n
l
y
t
h
e
d
a
t
a
s
e
t
.
I
n
h
y
b
r
i
d
ap
p
r
oa
c
h
c
om
bi
ni
n
g
t
he
a
dv
a
nt
a
ge
s
o
f
t
he
w
r
a
p
pe
r
a
p
pr
o
a
c
h
a
nd
t
he
f
i
l
t
e
r
a
pp
r
oa
c
h
i
t
us
e
s
t
he
i
n
di
vi
dua
l
i
s
t
i
c
cr
i
t
er
i
o
n
a
n
d
a
l
ear
n
i
n
g
al
g
o
r
i
t
h
m
t
o
r
at
e
t
h
e
can
d
i
d
at
e
f
eat
u
r
e
s
u
b
s
et
s
[8
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
A
A
S
I
S
S
N
:
225
2
-
88
14
D
at
a
M
i
ni
n
g
T
e
c
hni
que
s
f
or
P
r
ov
i
di
ng
N
e
t
w
or
k
Se
c
u
r
i
t
y
t
hr
o
u
gh
…
(
P
r
ab
hu
K
a
v
in
B
)
9
I
n
hi
g
h
di
m
e
ns
i
ona
l
a
ppl
i
c
a
t
i
ons
f
e
a
t
u
r
e
s
e
l
e
c
t
i
on i
s
ve
r
y
m
uc
h im
por
t
a
nt
.
F
r
om
t
he
n
um
be
r
o
f
o
r
i
g
i
n
al
f
eat
u
r
es
,
t
h
e
f
eat
u
r
e
s
el
ect
i
o
n
w
as
t
h
e c
o
m
b
i
n
at
o
r
i
al
p
r
o
b
l
em
a
nd
f
o
un
d
t
he
o
pt
i
m
a
l
s
ubs
e
t
w
a
s
N
P
-
h
a
r
d
.
W
h
i
l
e f
aci
n
g
i
m
b
al
an
ced
d
at
a s
et
s
f
eat
u
r
e s
el
ec
t
i
o
n
i
s
v
er
y
m
u
ch
h
el
p
f
u
l
[9
]
.
R
o
ug
h s
e
t
-
b
as
ed
ap
p
r
o
ach
u
s
es
at
t
r
i
b
u
t
e
d
e
p
e
n
d
en
cy
t
o
t
a
k
e
aw
ay
t
h
e
f
eat
u
r
e
s
el
ect
i
o
n
,
w
h
i
ch
w
as
i
m
p
o
r
t
an
t
.
T
h
e
d
e
p
en
d
en
cy
m
eas
u
r
e t
h
at
w
as
n
ece
s
s
a
r
y
f
o
r
t
h
e
c
a
l
c
u
l
a
t
i
o
n
o
f
t
h
e
p
o
s
i
t
i
v
e
r
e
g
i
o
n
b
u
t
w
h
i
l
e
c
a
l
c
u
l
a
t
i
n
g
i
t
w
a
s
a
n
e
xt
r
a
va
ga
nt
t
a
s
k
[6
]
.
De
pe
n
d on t
he
pa
r
t
i
c
l
e
s
wa
r
m
opt
im
iz
a
t
i
on (
P
S
O
)
a
nd r
o
ug
h s
e
t
s
,
t
he
pos
i
t
i
ve
r
e
gi
o
n
-
ba
s
e
d a
p
pr
oa
c
h ha
s
be
e
n p
r
e
s
e
nt
e
d
.
I
t
i
s
a
s
upe
r
i
nt
e
nde
d c
om
bi
ne
d f
e
a
t
ur
e
s
e
l
e
c
t
i
on a
l
g
or
i
t
hm
a
nd
by
us
i
n
g
t
h
e co
n
v
e
n
t
i
o
n
al
d
ep
e
n
d
e
n
cy
,
f
i
t
n
es
s
f
u
n
ct
i
o
n
w
as
m
eas
u
r
ed
f
o
r
each
p
ar
t
i
cl
e i
s
ev
al
u
at
ed
.
T
h
e al
g
o
r
i
t
h
m
s
f
i
g
u
re
-
o
u
t
t
h
e
s
t
r
e
n
g
t
h
o
f
t
h
e
s
e
l
e
c
t
e
d
f
e
a
t
u
r
e
w
i
t
h
v
a
r
i
o
u
s
c
o
n
s
o
l
i
d
a
t
i
o
n
s
b
y
s
e
l
e
c
t
i
n
g
a
n
a
t
t
r
i
b
u
t
e
w
i
t
h
a
hi
g
he
r
de
pe
n
d
e
nc
y
va
l
ue
.
I
f
t
he
pa
r
t
i
c
l
e
'
s
f
i
t
ne
s
s
va
l
ue
i
s
hi
g
he
r
t
ha
n
t
he
p
r
e
vi
ou
s
b
e
s
t
va
l
ue
wi
t
h
i
n t
h
e
c
ur
r
e
nt
s
wa
r
m
(
p
be
s
t
)
,
t
he
n t
h
e
pa
r
t
i
c
l
e
va
l
u
e
i
s
t
h
e
c
u
r
r
e
n
t
b
e
s
t
(
g
b
e
s
t
)
.
T
h
e
n
i
t
s
f
i
t
n
e
s
s
w
a
s
c
o
m
p
a
r
e
d
w
i
t
h
t
h
e
p
o
p
u
l
a
t
i
o
n
'
s
o
v
e
r
a
l
l
p
r
e
v
i
o
u
s
b
e
s
t
f
i
t
n
e
s
s
.
T
h
e
a
r
t
i
c
l
e
f
i
t
n
e
s
s
w
h
i
c
h
i
s
b
e
t
t
e
r
w
i
l
l
b
e
a
t
t
h
e
p
o
s
i
t
i
o
n
o
f
b
e
s
t
f
eat
u
r
e
s
u
b
s
et
.
T
h
e
p
a
r
t
i
cl
e
v
el
o
ci
t
i
es
w
e
r
e u
p
d
at
ed
at
t
h
e l
as
t
.
T
h
e
d
e
pe
n
de
nc
y
of
t
he
de
c
i
s
i
on
a
t
t
r
i
but
e
s
w
hi
c
h wa
s
o
n
t
he
c
o
ndi
t
i
o
n
a
l
a
t
t
r
i
but
e
s
w
a
s
c
a
l
c
ul
a
t
e
d
by
p
os
i
t
i
ve
r
e
gi
o
n ba
s
e
d de
pe
n
de
nc
y
m
e
a
s
ur
e
a
n
d
o
n
l
y
b
e
c
a
u
s
e
o
f
b
o
t
t
l
e
n
e
c
k
f
o
r
l
a
r
g
e
d
a
t
a
s
e
t
s
i
t
i
s
s
u
i
t
a
b
l
e
o
n
l
y
f
o
r
s
m
a
l
l
e
r
o
n
e
s
[6
]
.
I
nc
r
e
m
e
nt
a
l
f
e
a
t
ur
e
s
e
l
e
c
t
i
on
a
l
gor
i
t
hm
(
I
F
S
A)
i
s
m
a
i
nl
y
de
s
i
g
ne
d f
o
r
t
h
e
pur
p
os
e
o
f
s
ubs
e
t
f
e
a
t
ur
e
s
e
l
e
c
t
i
o
n
.
T
h
e
s
t
a
r
t
i
n
g
p
o
i
n
t
i
s
t
h
e
o
r
i
g
i
n
a
l
f
e
a
t
u
r
e
s
u
b
s
e
t
P
,
i
n
a
n
i
n
c
r
e
m
e
nt
a
l
m
a
nne
r
t
he
ne
w
de
pe
nde
nc
y
f
u
n
ct
i
o
n
w
as
cal
cu
l
at
ed
a
n
d
r
e
q
u
i
r
ed
f
eat
u
r
e
s
u
b
s
et
s
ar
e ch
ec
k
e
d
.
P
i
s
t
h
e
n
e
w
f
e
at
u
r
e
s
u
b
s
et
i
f
t
h
e
de
pe
n
de
nc
y
f
u
nc
t
i
on
P
i
s
e
q
ua
l
t
o t
he
f
e
a
t
ur
e
s
u
bs
e
t
i
f
n
ot
i
t
c
om
put
e
s
a
ne
w
f
e
a
t
ur
e
s
u
bs
e
t
.
T
he
gr
a
dua
l
l
y
s
el
ect
ed
s
i
g
n
i
f
i
can
t
f
eat
u
r
es
w
er
e
ad
d
e
d
t
o
t
h
e
f
eat
u
r
e
s
u
b
s
e
t
.
F
i
n
al
l
y
,
b
y
r
e
m
o
v
i
n
g
t
h
e
r
e
d
u
n
d
an
t
f
eat
u
r
es
t
h
e
opt
i
m
a
l
out
put
i
s
e
ns
u
r
e
d
.
T
he
n
a
ga
i
n,
t
he
a
l
g
or
i
t
hm
us
e
d
t
h
e
pos
i
t
i
ve
r
e
gi
on
-
ba
s
e
d
de
pe
nde
nc
y
m
e
a
s
ur
e
,
a
n
d
t
o
m
a
k
e
i
t
u
n
s
u
i
t
a
b
l
e
f
o
r
l
a
r
g
e
d
a
t
a
s
e
t
s
[6
]
.
F
i
s
h
S
w
a
r
m
a
l
g
o
r
i
t
h
m
w
a
s
s
t
a
r
t
e
d
w
i
t
h
a
n
i
n
i
t
i
a
l
p
o
p
u
l
a
t
i
o
n
(s
wa
rm
) o
f fi
s
h
fo
r s
e
a
rc
h
i
n
g
t
h
e
fo
o
d
.
He
re
e
v
e
ry
c
a
n
d
i
d
a
t
e
s
o
l
u
t
i
o
n
i
s
re
p
re
s
e
n
t
e
d
b
y
a
fi
s
h
.
T
h
e
s
w
a
rm
c
ha
n
ge
s
t
he
i
r
pos
i
t
i
o
n a
n
d c
om
m
uni
c
a
t
e
s
wi
t
h e
a
c
h
ot
he
r
i
n
s
e
a
r
c
hi
n
g
of
t
he
be
s
t
l
oc
a
l
po
s
i
t
i
on a
n
d
t
he
be
s
t
g
l
o
b
a
l
p
o
s
i
t
i
o
n
s
.
W
h
e
n
a
f
i
s
h
a
c
h
i
e
v
e
d
m
a
x
i
m
u
m
s
t
r
e
n
g
t
h
,
i
t
l
o
s
e
s
i
t
s
n
o
r
m
a
l
q
u
a
l
i
t
y
a
f
t
e
r
o
b
t
a
i
n
i
n
g
t
h
e
Re
d
u
c
t
r
o
u
g
h
s
e
t
.
A
f
t
e
r
a
l
l
o
f
t
h
e
f
i
s
h
e
s
h
a
v
e
l
o
s
t
i
t
n
o
r
m
a
l
q
u
a
l
i
t
y
t
h
e
n
e
x
t
i
t
e
r
a
t
i
o
n
s
t
a
r
t
s
.
A
f
t
e
r
t
h
e
s
i
m
i
l
a
r
f
eat
u
r
e
r
e
d
u
ct
w
as
o
b
t
ai
n
e
d
u
n
d
e
r
t
h
r
ee co
n
s
ecu
t
i
v
e i
t
er
at
i
o
n
s
o
r
t
h
e l
ar
g
es
t
i
t
er
at
i
o
n
co
n
d
i
t
i
o
n
w
as
r
e
ach
ed
,
t
he
n t
he
a
l
go
r
i
t
hm
ha
l
t
s
.
T
he
n e
qui
va
l
e
nt
r
ou
g
h s
e
t
-
ba
s
e
d
de
pe
n
de
nc
y
m
e
a
s
ur
e
wa
s
u
s
e
d i
n t
hi
s
a
l
g
or
i
t
hm
a
nd
i
t
s
u
f
f
e
r
s
f
r
om
t
he
s
a
m
e
pr
obl
e
m
of
t
he
l
a
r
ge
da
t
a
s
e
t
s
pe
r
f
or
m
a
nc
e
de
gr
a
da
t
i
on
[6
]
.
C
o
rr
e
l
a
t
i
o
n
-
b
a
s
e
d
F
e
a
t
u
r
e
S
e
l
e
c
t
i
o
n
i
s
a
m
u
l
t
i
v
a
r
i
a
t
e
s
u
b
s
e
t
f
i
l
t
e
r
a
l
g
o
r
i
t
h
m
.
A
s
e
a
r
c
h
a
l
g
o
r
i
t
h
m
u
n
i
t
e
d
w
i
t
h
a
n
e
s
t
i
m
a
t
i
o
n
f
u
n
c
t
i
o
n
t
h
a
t
w
a
s
u
s
e
d
t
o
e
v
a
l
u
a
t
e
t
h
e b
en
ef
i
t
o
f
f
eat
u
r
e
s
u
b
s
et
s
.
T
h
e
i
m
p
l
e
m
e
n
t
a
t
i
o
n
o
f
CF
S
u
s
e
d
t
h
e
f
o
r
w
a
r
d
b
e
s
t
f
i
r
s
t
s
e
a
r
c
h
a
s
i
t
s
s
e
a
r
c
h
i
n
g
a
l
g
o
r
i
t
h
m
.
Be
s
t
f
i
r
s
t
s
e
a
r
c
h
i
s
o
n
e
o
f
t
h
e ar
t
i
f
i
ci
al
i
n
t
el
l
i
g
en
ce s
ear
c
h
s
cen
ar
i
o
i
n
w
h
i
c
h
b
ack
t
r
ac
k
i
n
g
w
as
al
l
o
w
e
d
al
o
n
g
w
i
t
h
t
h
e s
ear
c
h
pa
t
h.
B
y
m
a
king
s
om
e
l
im
i
te
d a
dj
us
t
m
e
nt
t
o t
he
c
ur
r
e
nt
f
e
a
t
ur
e
s
u
bs
e
t
i
t
m
ove
s
t
hr
o
u
gh
t
he
s
e
a
r
c
h
s
pa
c
e
.
Th
i
s
a
l
g
or
i
t
hm
c
a
n ba
c
kt
r
a
c
k
t
o t
he
e
a
r
l
i
e
r
s
ub
s
e
t
w
h
e
n t
h
e
e
xpl
or
e
d pa
t
h l
o
oks
un
e
xc
i
t
i
ng a
n
d a
d
va
n
c
e
t
he
s
ear
ch
f
r
o
m
t
h
er
e
o
n
.
T
h
e
n
t
h
e s
ear
c
h
h
al
t
ed
,
i
f
f
i
v
e
s
u
cces
s
i
v
e f
u
l
l
y
ex
p
a
n
d
ed
t
h
e s
u
b
s
et
s
s
h
o
w
s
n
o
de
ve
l
opm
e
nt
o
ve
r
t
he
pr
e
s
e
nt
be
s
t
s
ubs
e
t
[5
]
.
T
he
o
bje
c
t
i
ve
o
f
S
R
F
S
i
s
t
o f
i
nd
t
he
f
e
a
t
ur
e
s
u
bs
e
t
S
wi
t
h
t
he
s
i
z
e
d
,
w
hi
c
h
c
ont
a
i
n
s
t
he
r
e
p
r
e
s
e
nt
a
t
i
ve
f
e
a
t
ur
e
s
,
i
n w
hi
c
h b
ot
h
t
he
l
a
be
l
e
d
a
n
d un
l
a
be
l
e
d
d
a
t
a
s
e
t
a
r
e
e
x
p
l
o
i
t
i
n
g
.
I
n
t
h
i
s
t
h
e
f
e
a
t
u
r
e
r
e
l
e
v
a
n
c
e
i
s
c
l
a
s
s
i
f
i
e
d
i
n
t
o
t
h
r
e
e
d
i
s
j
o
i
n
t
c
a
t
e
g
o
r
i
e
s
:
s
t
r
o
n
g
l
y
r
e
l
e
v
a
n
t
,
w
e
a
k
l
y
r
e
l
e
v
an
t
an
d
i
r
r
el
ev
an
t
f
eat
u
r
e
s
[
1
0
-
12
]
.
A
s
t
r
o
n
g
r
el
ev
an
t
f
eat
u
r
e
w
as
al
w
ay
s
b
as
i
c f
o
r
a
n
o
p
t
i
m
a
l
or
s
ub
opt
im
a
l
f
e
a
t
ur
e
s
u
b
se
t
.
I
f
t
h
e s
t
r
o
n
g
r
el
ev
a
n
t
f
eat
u
r
e i
s
ev
acu
at
e
d
,
u
s
i
n
g
t
h
e
f
e
at
u
r
e s
u
b
s
et
t
h
e cl
as
s
i
f
i
cat
i
o
n
a
b
i
l
i
t
y
i
s
d
i
r
e
c
t
l
y
i
nf
l
ue
nc
e
d.
E
x
cep
t
f
o
r
an
o
p
t
i
m
al
o
r
s
u
b
o
p
t
i
m
al
f
eat
u
r
e s
u
b
s
et
at
cer
t
ai
n
co
n
d
i
t
i
o
n
s
,
a
w
eak
r
el
e
v
an
t
f
eat
u
r
e
i
s
n
o
t
al
w
ay
s
n
eces
s
a
r
y
.
I
r
r
el
e
v
an
t
f
eat
u
r
e i
t
o
n
l
y
en
l
a
r
g
es
s
ear
c
h
s
p
ace an
d
m
ak
es
t
h
e p
r
o
b
l
em
m
o
r
e
c
om
pl
e
x,
a
n
d i
t
doe
s
n'
t
pr
o
vi
de
a
ny
i
nf
or
m
a
t
i
on t
o i
m
pr
o
ve
t
he
pr
e
di
c
t
i
on a
c
c
u
r
a
c
y
s
o
i
t
i
s
not
ne
c
e
s
s
a
r
y
a
t
an
y
t
i
m
e.
H
en
ce al
l
f
eat
u
r
es
o
f
s
t
r
o
n
g
l
y
r
el
ev
an
t
an
d
s
u
b
s
et
f
eat
u
r
es
o
f
w
eak
l
y
r
e
l
e
va
nt
a
nd n
o i
r
r
e
l
e
va
nt
f
e
a
t
ur
e
s
s
ho
ul
d
be
i
nc
l
ude
d
b
y
t
he
o
pt
im
a
l
f
e
a
t
ur
e
s
u
bs
e
t
.
A
n
i
n
a
ddi
t
i
o
n
s
up
e
r
vi
s
e
d
f
e
a
t
ur
e
s
e
l
e
c
t
i
on
m
e
t
hod
t
h
a
t
u
s
e
s
t
h
e
b
i
l
a
t
e
r
a
l
i
n
f
o
r
m
a
t
i
o
n
b
e
t
w
e
e
n
f
e
a
t
u
r
e
a
n
d
c
l
a
s
s
t
h
a
t
t
e
n
d
t
o
f
i
n
d
t
h
e
o
p
t
i
m
a
l
o
r
s
u
b
o
p
t
i
m
a
l
f
eat
u
r
e
s
o
v
e
r
f
i
t
t
ed
t
o
t
h
e
l
ab
el
ed
d
at
a,
w
h
e
n
a
s
m
al
l
n
u
m
b
er
o
f
l
a
b
el
ed
d
at
a ar
e
a
v
ai
l
ab
l
e.
I
n
t
h
i
s
cas
e,
d
at
a
m
i
ti
ga
t
i
on m
a
y
be
a
bl
e
t
o
oc
c
u
r
i
n t
hi
s
pr
o
bl
e
m
on u
s
i
ng u
nl
a
be
l
e
d
da
t
a
.
T
he
r
e
f
or
e
,
r
e
l
e
va
nc
e
ga
i
n
co
n
s
i
d
er
i
n
g
f
e
at
u
r
e r
el
e
v
a
n
c
e i
n
u
n
l
ab
el
e
d
d
at
as
et
,
an
d
p
r
o
p
o
s
e a n
e
w
f
r
am
ew
o
r
k
f
o
r
f
eat
u
r
e
s
el
ect
i
o
n
o
n
r
em
o
v
i
n
g
t
h
e i
r
r
el
ev
a
n
t
an
d
r
ed
u
n
d
an
t
f
eat
u
r
es
cal
l
ed
as
S
e
m
i
-
s
u
p
e
r
v
i
s
e
d
R
ep
r
es
e
n
t
at
i
v
es
F
eat
u
r
e S
el
ect
i
o
n
a
l
g
o
r
i
t
h
m
i
s
d
e
f
i
n
e
d
.
S
RF
S
i
s
a
s
e
m
i
-
s
upe
r
vi
s
e
d
f
i
l
t
e
r
f
e
a
t
ur
e
s
e
l
e
c
t
i
on
ba
s
e
d
on
t
he
M
a
r
k
ov
bl
a
nke
t
[8
]
2
.
2
.
R
e
l
a
te
d
W
o
r
k
s
o
n
C
l
a
s
s
i
fi
c
a
ti
o
n
Al
g
o
r
i
th
m
s
T
he
c
om
bi
ne
d
r
e
s
po
ns
e
c
o
m
pos
e
d
by
t
h
e
m
ul
t
i
pl
e
c
l
a
s
s
i
f
i
e
r
s
i
nt
o a
s
i
n
gl
e
r
e
s
p
o
ns
e
w
a
s
t
he
en
s
em
b
l
e cl
as
s
i
f
i
er
.
E
v
e
n
t
h
o
u
g
h
m
an
y
en
s
e
m
b
l
e t
ech
n
i
q
u
es
e
x
i
s
t
,
f
o
r
a p
a
r
t
i
cu
l
ar
d
a
t
as
et
i
t
w
as
h
a
r
d
t
o
f
o
u
n
d
s
u
i
t
a
b
l
e
e
n
s
e
m
b
l
e
c
o
n
f
i
g
u
r
a
t
i
o
n
.
E
n
s
e
m
b
l
e
c
l
a
s
s
i
f
i
e
r
s
a
r
e
u
s
e
d
t
o
m
ax
i
m
i
ze t
h
e cer
t
ai
n
t
y
o
f
s
e
v
er
al
c
l
a
s
s
i
f
i
c
a
t
i
on t
a
s
ks
.
M
a
ny
m
e
t
ho
ds
h
a
ve
b
e
e
n p
r
op
os
e
d,
w
i
t
h m
e
a
n c
om
bi
ne
r
,
m
a
x c
om
bi
ne
r
,
m
e
di
a
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
252
-
88
1
4
IJ
A
A
S
V
o
l
.
7
,
N
o
.
1
,
M
a
r
ch
2
018
:
7
–
12
10
c
om
bi
ne
r
,
m
a
jor
i
t
y
vot
i
n
g a
nd
we
i
g
he
d m
a
jo
r
i
t
y
vot
i
n
g
(
W
M
V
)
w
he
r
e
a
s
t
he
i
ndi
vi
d
ua
l
c
l
a
s
s
i
f
i
e
r
s
c
a
n be
co
n
n
ect
ed
u
s
i
n
g a
ny
o
ne
of
t
he
s
e
m
e
t
hods
[
13
]
.
T
o s
ol
v
e
c
l
a
s
s
i
f
i
c
a
t
i
on
a
n
d r
e
gr
e
s
s
i
o
n
pr
obl
e
m
s
s
u
pp
o
r
t
ve
c
t
o
r
m
a
c
hi
ne
s
(
S
VM
)
i
s
a
n e
f
f
e
c
t
i
ve
t
e
c
hni
que
.
S
VM
w
a
s
t
he
i
m
pl
e
m
e
nt
a
t
i
on
of
V
a
pni
k’
s
S
t
r
uc
t
u
r
a
l
R
i
s
k
M
i
nim
i
z
a
ti
on (
S
R
M
)
p
r
i
nc
i
pl
e
w
hi
c
h ha
s
c
om
pa
r
a
t
i
ve
l
y
low
ge
ne
r
a
l
i
z
a
t
i
on e
r
r
or
a
nd
doe
s
n
ot
s
u
f
f
e
r
m
uc
h
f
r
om
ove
r
f
i
t
t
i
ng
t
o t
he
t
r
a
i
ni
ng
da
t
a
s
e
t
.
W
h
e
n a
m
ode
l
pe
r
f
o
r
m
s
po
or
a
n
d
not
l
oc
a
t
e
d
i
n t
he
t
r
a
i
ni
ng
s
e
t
t
he
n
i
t
w
a
s
s
a
i
d
t
o
b
e
o
v
e
r
f
i
t
a
n
d
h
a
s
h
i
g
h
g
e
n
e
r
a
l
i
z
a
t
i
o
n
e
r
r
o
r
[
13
]
.
Re
c
e
n
t
l
y
a
s
i
g
n
i
f
i
c
a
n
t
a
t
t
e
n
t
i
o
n
w
a
s
a
t
t
r
a
c
t
e
d
b
y
t
h
e
m
u
l
t
i
-
l
a
be
l
c
l
a
s
s
i
f
i
c
a
t
i
on
,
w
hi
c
h
wa
s
m
oti
va
t
e
d
by
m
or
e
num
be
r
of
a
ppl
i
c
a
t
i
o
ns
.
E
xa
m
pl
e
i
nc
l
ude
t
e
xt
cat
eg
o
r
i
zat
i
o
n
,
i
m
ag
e cl
as
s
i
f
i
cat
i
o
n
,
v
i
d
eo
cl
as
s
i
f
i
cat
i
o
n
,
m
u
s
i
c
c
a
t
e
gor
i
z
a
t
i
on,
ge
ne
a
nd pr
ot
e
i
n
f
u
n
c
t
i
on
pr
e
di
c
t
i
on
,
m
e
di
c
a
l
di
a
g
n
os
i
s
,
c
he
m
i
c
a
l
a
na
ly
s
i
s
,
s
oc
i
a
l
ne
t
w
o
r
k
m
i
ni
ng
a
nd
di
r
e
c
t
m
a
r
k
e
t
i
ng
a
nd
m
a
ny
m
or
e
e
xa
m
pl
e
s
f
ou
n
d.
T
o i
m
pr
ove
t
he
c
l
a
s
s
i
f
i
c
a
t
i
on
pe
r
f
or
m
a
nc
e
by
t
he
ut
i
l
i
z
a
t
i
on o
f
l
a
be
l
d
e
pe
n
de
nc
i
e
s
w
a
s
t
he
ke
y
pr
obl
e
m
in m
ul
ti
-
l
a
be
l
le
a
r
ni
ng a
nd h
o
w
i
t
i
s
m
o
t
i
va
te
d by
w
hi
c
h n
um
be
r
of
m
ul
ti
-
l
a
b
e
l
a
l
g
o
r
i
t
h
m
t
h
a
t
ha
ve
be
e
n
p
r
o
pos
e
d i
n
r
e
c
e
nt
y
e
a
r
s
(
f
or
e
xt
e
ns
i
ve
c
om
pa
r
i
s
on
o
f
s
e
ve
r
a
l
m
e
t
hods
)
.
T
he
pr
o
gr
e
s
s
i
n t
he
M
L
C
i
n
r
ece
n
t
t
i
m
e w
as
s
u
m
m
ar
i
z
ed
.
F
eat
u
r
e
s
p
ace
D
i
m
e
n
s
i
o
n
a
l
i
t
y
r
e
d
u
c
t
i
o
n
,
i
.
e
.
r
e
d
u
c
i
n
g
t
h
e
d
i
m
e
n
s
i
o
n
a
l
i
t
y
o
f
t
he
ve
c
t
o
r
x i
s
one
of
t
he
t
r
e
n
di
n
g c
ha
l
l
e
n
ge
s
i
n M
L
C
.
T
he
di
m
e
ns
i
ona
l
i
t
y
of
f
e
a
t
ur
e
s
p
a
c
e
c
a
n be
ve
r
y
l
a
r
ge
a
n
d
t
h
i
s
i
s
s
u
e
i
n
p
r
a
c
t
i
c
a
l
a
p
p
l
i
c
a
t
i
o
n
s
i
s
v
e
r
y
i
m
p
o
r
t
a
n
t
[
14
]
.
M
a
n
y
i
n
t
e
l
l
i
g
e
n
t
i
n
t
r
u
s
i
o
n
d
e
t
e
c
t
i
o
n
s
y
s
t
e
m
s
ha
ve
be
e
n di
s
c
us
s
e
d i
n [
1]
a
n
d a
l
s
o br
i
e
f
l
y
de
s
c
r
i
be
d t
he
us
a
ge
of
a
r
t
i
f
i
c
i
a
l
i
nt
e
ll
i
ge
nc
e
a
nd s
o
f
t
c
om
put
i
n
g
t
e
c
hni
q
ue
s
f
o
r
pr
ovi
di
n
g ne
t
w
o
r
k s
e
c
u
r
i
t
y
.
M
or
e
o
ve
r
,
a
ne
w
i
nt
e
l
l
i
ge
n
t
a
g
e
n
t
b
a
s
e
d
M
u
l
t
i
c
l
a
s
s
S
u
p
p
o
r
t
V
e
c
t
o
r
M
a
c
h
i
n
e
a
l
g
o
r
i
t
h
m
w
h
i
c
h
i
s
t
h
e
c
o
m
b
i
n
a
t
i
o
n
o
f
i
n
t
e
l
l
i
g
e
n
t
a
g
e
n
t
,
d
e
c
i
s
i
o
n
t
r
e
e
a
n
d
c
l
u
s
t
e
r
i
n
g
i
s
a
l
s
o
pr
o
pos
e
d
a
n
d
i
m
pl
e
m
e
nt
e
d.
T
he
y
pr
ov
e
d
t
he
i
r
s
y
s
t
e
m
wa
s
be
t
t
e
r
w
he
n
c
o
m
pa
r
e
d
wi
t
h
ot
he
r
e
xi
s
t
i
n
g
s
y
s
t
e
m
s
.
Re
c
e
n
t
l
y
,
t
e
m
p
o
r
al
f
eat
u
r
es
ar
e al
s
o
i
n
co
r
p
o
r
at
ed
w
i
t
h
f
u
zzy
l
o
g
i
c f
o
r
m
a
ki
ng de
c
i
s
i
on
dy
na
m
i
c
a
l
ly
[
15
].
T
h
ey
ac
h
i
ev
e
d
b
et
t
er
cl
as
s
i
f
i
c
at
i
o
n
acc
u
r
acy
o
v
e
r
t
h
e
r
eal
t
i
m
e
d
at
a
s
et
s
.
2
.
3
.
R
el
a
t
e
d
w
o
rk
s
o
n
C
l
u
s
t
eri
n
g
a
n
d
O
u
t
l
i
er
D
et
e
ct
i
o
n
C
l
us
t
e
r
i
n
g t
e
c
hni
que
s
a
r
e
ve
r
y
u
s
e
f
u
l
f
o
r
en
h
a
n
ci
n
g
t
h
e
cl
as
s
i
f
i
cat
i
o
n
accu
r
acy
.
M
an
y
cl
u
s
t
er
i
n
g
a
l
go
r
i
t
hm
s
ha
ve
be
e
n u
s
e
d
i
n va
r
i
ous
i
nt
r
us
i
o
n de
t
e
c
t
i
on s
y
s
t
e
m
s
i
n t
he
pa
s
t
f
or
a
c
hi
e
vi
n
g
be
t
t
e
r
pe
r
f
o
r
m
a
nc
e
.
C
l
us
t
e
r
i
n
g t
e
c
hni
que
s
a
r
e
u
s
e
f
ul
i
n
bot
h
da
t
a
s
e
t
s
s
uc
h a
s
ne
t
w
or
k t
r
a
c
e
da
t
a
a
n
d
be
nc
h m
a
r
k
d
at
as
et
f
o
r
m
a
ki
n
g e
f
f
e
c
t
i
ve
gr
o
upi
ng [
1
6
]
,
[
17
]
.
O
u
t
l
i
e
r
d
e
t
e
c
t
i
o
n
i
s
a
l
s
o
u
s
e
f
u
l
f
o
r
i
d
e
n
t
i
f
y
i
n
g
t
h
e
u
n
r
e
l
a
t
e
d
us
e
r
s
i
n a
ne
t
w
or
k.
T
hi
s
out
l
i
e
r
de
t
e
c
t
i
o
n t
e
c
hni
que
i
s
us
e
d f
or
i
de
nt
i
f
y
i
n
g t
he
o
ut
l
i
e
r
s
i
n a
ne
t
w
or
k.
I
t
c
a
n
be
a
p
p
l
i
e
d
i
n
r
e
a
l
n
et
w
o
r
k
s
cen
a
r
i
o
an
d
b
o
t
h
d
at
as
et
s
s
u
ch
as
n
et
w
o
r
k
t
r
ace d
at
as
et
an
d
t
h
e b
en
c
h
m
ar
k
d
at
as
et
.
M
or
e
ove
r
,
s
o
f
t
c
om
put
i
n
g t
e
c
hni
que
s
a
r
e
us
e
d i
n
t
he
s
e
t
w
o a
pp
r
oa
c
he
s
f
or
m
a
ki
ng
f
i
na
l
de
c
i
s
i
o
ns
o
ve
r
t
he
da
t
a
s
e
t
s
.
T
he
e
xi
s
t
i
ng
w
or
ks
[1
8
]
,
[1
9
]
ac
h
i
e
v
ed
b
et
t
er
d
et
e
ct
i
o
n
acc
u
r
acy
.
3.
C
O
MP
A
R
A
T
I
VE
A
N
A
L
Y
S
I
S
M
os
t
o
f
t
he
I
nt
r
us
i
on
D
e
t
e
c
t
i
on
S
y
s
t
e
m
s
ha
ve
be
e
n
us
e
d
d
a
t
a
m
i
ni
ng t
e
c
hni
que
s
s
uc
h
a
s
C
l
us
t
e
r
i
n
g,
O
ut
l
i
e
r
de
t
e
c
t
i
on
,
C
l
a
s
s
i
f
i
c
a
t
i
on a
nd
da
t
a
pr
e
pr
oc
e
s
s
i
ng
.
He
r
e
,
da
t
a
p
r
e
pr
oc
e
s
s
i
ng t
e
c
hni
que
s
a
r
e
us
e
d t
o
en
h
a
n
ce t
h
e c
l
as
s
i
f
i
cat
i
o
n
ac
cu
r
acy
.
F
eat
u
r
e s
el
ect
i
o
n
m
e
t
h
o
d
s
ar
e
u
s
ed
t
o
r
ed
u
ce t
h
e
cl
as
s
i
f
i
cat
i
o
n
t
i
m
e.
T
hi
s
pa
pe
r
de
s
c
r
i
be
s
va
r
i
o
u
s
t
y
pe
s
of
f
e
a
t
ur
e
s
e
l
e
c
t
i
on
w
hi
c
h a
r
e
p
r
o
pos
e
d
i
n
t
h
i
s
d
i
r
e
c
t
i
o
n
i
n
t
h
e
p
a
s
t
.
T
he
a
ve
r
a
ge
pe
r
f
or
m
a
nc
e
of
t
he
e
xi
s
t
i
n
g
c
l
a
s
s
i
f
i
c
a
t
i
on
a
l
g
o
r
i
t
hm
s
i
s
94%
a
nd
i
t
ha
s
i
m
pr
ove
d
i
nt
o
9
6%
w
he
n
ap
p
l
i
ed
d
at
a
p
r
e
p
r
o
ces
s
i
n
g
.
I
n
a
d
d
i
t
i
o
n
,
t
h
e av
e
r
ag
e
d
et
ect
i
o
n
accu
r
acy
i
s
r
each
e
d
t
o
9
9
%
w
h
en
u
s
e
d
cl
u
s
t
er
i
n
g
o
r
o
u
t
l
i
er
d
et
ect
i
o
n
t
ech
n
i
q
u
e
s
.
T
a
b
l
e
1
s
h
o
w
s
t
h
e
p
e
r
f
o
r
m
an
ce
co
m
p
ar
at
i
v
e
an
al
y
s
i
s
.
T
ab
l
e
1
.
C
o
m
p
ar
at
i
v
e
A
n
al
y
s
i
s
No
.
A
ut
hor
na
m
e
M
eth
o
d
Ov
er
all
A
ccu
r
ac
y
(
%
)
1
Sr
in
iv
as
M
u
k
k
a
m
a
la et al
[
2
0
]
SVM
99.
63%
2
Gan
ap
ath
y
et
al [
2
1
]
IA
E
MS
V
M
91.
13%
3
Gan
ap
ath
y
et
al [
1
]
IR
EMS
V
M
91.
26%
4
So
o
-
Yeo
n
J
i et al [
2
]
MLI
D
S
96%
5
Om
ar
Y.
Al
-
J
ar
r
ah
et al [
4
]
R
DPL
M
99.
98%
6
Ab
d
u
lla A
m
in
Ab
u
r
o
m
m
an
e
t al [
5
]
KNN
91.
68%
7
Ab
d
u
lla A
m
in
Ab
u
r
o
m
m
an
e
t al [
5
]
E
ns
e
m
bl
e
92.
74%
8
Vin
o
d
k
u
m
ar
Deh
a
r
iy
a et
al [
1
6
]
FKM
83.
16%
9
UjjwalM
au
lik
et al
[
1
6
]
GA
-
FK
M
88.
46%
10
C
h
en
jieGu
et al [
1
6
]
IG
A
-
FKK
M
93.
01%
11
Gan
ap
ath
y
et
al [
1
1
]
IG
A
-
NW
FC
M
94.
86%
12
J.
R
o
s
s
Qu
in
lan
et
al [
1
2
]
ID
3
95.
58%
13
E
r
n
s
t Kr
ets
ch
m
an
n
et al [
1
6
]
C
4
.5
96.
19%
14
Gu
o
liJ
i et al [
1
8
]
MS
V
M
98.
38%
15
Gan
ap
ath
y
et
al [
1
5
]
EMS
V
M
99.
10%
16
Gan
ap
ath
y
et
al [
1
9
]
WD
B
O
D
99.
52%
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
A
A
S
I
S
S
N
:
225
2
-
88
14
D
at
a
M
i
ni
n
g
T
e
c
hni
que
s
f
or
P
r
ov
i
di
ng
N
e
t
w
or
k
Se
c
u
r
i
t
y
t
hr
o
u
gh
…
(
P
r
ab
hu
K
a
v
in
B
)
11
F
r
om
T
a
bl
e
1,
i
t
c
a
n be
s
e
e
n t
ha
t
t
he
pe
r
f
or
m
a
nc
e
of
t
h
e
m
e
t
hod R
D
P
L
M
pe
r
f
o
r
m
w
e
l
l
t
ha
n t
he
ex
i
s
t
i
n
g
m
et
h
o
d
s
an
d
t
h
e e
x
i
s
t
i
n
g
cl
as
s
i
f
i
er
S
V
M
ach
i
e
v
ed
v
er
y
l
es
s
d
et
e
ct
i
o
n
accu
r
acy
t
h
an
o
t
h
er
s
.
T
h
i
s
i
s
due
t
o
t
he
us
e
of
va
r
i
o
us
c
om
bi
na
t
i
o
ns
of
m
e
t
ho
ds
a
n
d
t
h
e
u
s
e
o
f
i
n
t
e
l
l
i
g
e
n
t
a
g
e
n
t
s
.
F
i
gu
r
e
1
de
m
ons
t
r
a
t
e
s
t
he
pe
r
f
or
m
a
nc
e
a
na
l
y
s
i
s
i
n gr
a
ph
be
t
w
e
e
n t
he
t
o
p f
i
ve
m
e
t
hod
s
w
hi
c
h a
r
e
pr
o
pos
e
d
i
n
t
h
e
pa
s
t
by
va
r
i
o
us
r
e
s
e
a
r
c
he
r
s
.
H
e
r
e
,
w
e
ha
ve
c
ons
i
de
r
e
d
t
he
s
a
m
e
s
e
t
of
r
e
c
or
ds
f
or
c
on
d
uc
t
i
ng
e
xpe
r
i
m
e
nt
s
f
or
f
i
n
di
n
g
t
h
e cl
as
s
i
f
i
cat
i
o
n
ac
cu
r
acy
.
C
l
as
s
i
f
i
cat
i
o
n
accu
r
ac
y
o
f
v
ar
i
o
u
s
m
et
h
o
d
s
i
s
co
n
s
i
d
er
e
d
f
o
r
c
o
m
p
ar
at
i
v
e
an
al
y
s
i
s
.
F
i
g
u
r
e
1
.
P
er
f
o
r
m
an
ce
a
n
a
l
y
s
i
s
F
r
om
f
i
gur
e
1,
i
t c
a
n be
obs
e
r
ve
d t
ha
t
t
he
pe
r
f
o
r
m
a
nc
e
of
t
he
m
e
t
hod
R
DP
L
M
i
s
pe
r
f
o
r
m
e
d w
e
l
l
w
h
e
n
i
t
i
s
c
o
m
pa
r
e
d w
i
t
h
e
xi
s
t
i
ng m
e
t
hods
.
M
or
e
ove
r
,
t
he
I
G
A
-
N
W
F
C
M
m
e
t
hod a
c
hi
e
ve
s
ve
r
y
l
e
s
s
de
t
e
c
t
i
on
a
c
c
u
r
a
c
y
t
ha
n
t
he
ot
he
r
e
xi
s
t
i
n
g
a
l
go
r
i
t
hm
s
w
hi
c
h
a
r
e
c
ons
i
de
r
e
d
f
o
r
c
om
pa
r
a
t
i
ve
a
na
l
y
s
i
s
4.
SU
G
G
E
ST
I
O
N
P
R
O
P
O
SE
D
T
he
pe
r
f
o
r
m
a
nc
e
of
t
he
e
xi
s
t
i
ng s
y
s
t
e
m
s
c
a
n be
i
m
pr
o
ve
d
by
t
he
i
nt
r
od
u
c
t
i
on o
f
i
nt
e
l
l
i
ge
nt
a
ge
nt
s
a
nd
s
o
f
t
c
o
m
put
i
ng t
e
c
h
ni
q
ue
s
l
i
ke
f
uz
z
y
l
ogi
c
,
ne
ur
a
l
ne
t
wo
r
k
a
n
d
g
e
ne
t
i
c
a
l
g
or
i
t
hm
s
f
or
e
f
f
e
c
t
i
ve
de
c
i
s
i
on
ove
r
t
he
da
t
a
s
e
t
.
I
n
t
hi
s
f
a
s
t
w
or
l
d
,
t
im
e
a
nd
s
p
ace
ar
e
al
s
o
v
e
r
y
i
m
p
o
r
t
an
t
t
o
t
ak
e
ef
f
e
ct
i
v
e
de
c
i
s
i
o
n.
F
i
na
l
l
y
,
c
a
n i
nt
r
o
d
uc
e
a
ne
w s
y
s
t
e
m
whi
c
h c
o
nt
a
i
ns
ne
w
i
nt
e
l
l
i
ge
nt
a
ge
nt
s
,
ne
ur
a
l
ne
t
wo
r
k f
o
r
t
r
a
i
n
i
n
g
,
e
f
f
e
c
t
i
v
e
s
p
a
t
i
o
-
f
uz
z
y
t
e
m
por
a
l
ba
s
e
d da
t
a
pr
e
p
r
o
c
e
s
s
i
ng m
e
t
ho
d a
n
d
f
uz
z
y
t
e
m
por
a
l
r
ul
e
s
c
a
n b
e
u
s
e
d
f
o
r
m
ak
i
n
g
e
f
f
ect
i
v
e
d
eci
s
i
o
n
an
d
al
s
o
can
d
et
ect
at
t
ack
er
s
ef
f
ec
t
i
v
el
y
.
T
h
i
s
co
m
b
i
n
at
i
o
n
i
s
ab
l
e t
o
p
r
o
v
i
d
e b
et
t
er
p
er
f
o
r
m
an
ce
.
5.
C
O
N
CL
US
I
O
N
A
n e
f
f
e
c
t
i
ve
s
ur
ve
y
m
a
de
i
n t
he
di
r
e
c
t
i
o
n o
f
da
t
a
m
i
n
i
ng t
e
c
hni
que
-
b
a
s
e
d
i
n
t
r
u
s
i
o
n
d
e
t
e
c
t
i
o
n
s
ys
te
m
s
.
Ma
n
y
f
e
a
t
ur
e
s
e
l
e
c
t
i
on m
e
t
hod
s
ha
ve
be
e
n d
i
s
c
us
s
e
d i
n t
hi
s
pa
pe
r
a
nd
t
he
i
r
im
por
t
a
n
c
e
a
r
e
h
i
g
h
l
i
g
h
t
e
d
.
Cl
a
s
s
i
f
i
c
a
t
i
o
n
,
Cl
u
s
t
e
r
i
n
g
a
n
d
o
u
t
l
i
e
r
d
e
t
e
c
t
i
o
n
t
e
c
h
n
i
q
u
e
s
a
r
e
e
x
p
l
a
i
n
e
d
i
n
t
h
i
s
p
a
p
e
r
a
n
d
a
l
s
o
e
xpl
a
i
ne
d
h
ow
m
uc
h i
t
i
s
he
l
pf
ul
f
or
e
n
ha
n
c
i
ng t
he
pe
r
f
or
m
a
nc
e
.
F
i
na
l
l
y
,
s
u
g
ge
s
t
i
on
a
l
s
o
pr
op
os
e
d i
n
t
hi
s
pa
pe
r
ba
s
e
d
on
t
he
c
om
pa
r
a
t
i
ve
a
na
l
y
s
i
s
o
f
t
he
e
xi
s
t
i
ng
s
y
s
t
e
m
s
.
R
EF
ER
E
N
C
ES
[1]
.
S
.
G
a
na
pa
t
h
y
,
K
.
K
ul
ot
hunga
n
,
S
.
M
ut
hu
ra
j
ku
m
a
r,
M
.
V
i
j
a
y
a
l
a
ks
hm
i
,
P
.
Y
oge
s
h,
A
.
K
a
nn
a
n,
“
Int
e
l
l
i
ge
nt
fe
a
t
ur
e
s
e
le
c
tio
n
a
n
d
c
l
a
s
s
if
ic
a
ti
on
t
e
c
hn
i
que
s
for
i
nt
rus
i
on
de
t
e
c
t
i
on
i
n
ne
t
w
orks
:
a
s
ur
ve
y
”
,
E
U
R
A
S
I
P
W
i
r
el
es
s
Jo
u
r
n
a
l
of
Com
m
uni
c
at
i
ons
and Ne
t
wor
k
i
ng
,
vo
l
.
2013,
p
p.
1
–
16
,
2013
.
[2]
.
S
.
Y
.
J
i
,
B.
K
.
J
e
ong,
S
.
Choi
,
a
nd D
.
H
.
J
e
ong,
“
A
m
ul
t
i
-
l
e
ve
l
i
nt
rus
i
on de
t
e
c
t
i
on m
e
t
hod for a
bnorm
a
l
ne
t
w
or
k
be
ha
vi
ors
,
”
J
.
N
e
t
w.
Com
put
.
A
p
pl
.
,
vol
.
62
,
pp
.
9
–
17,
2016
.
[3]
.
O
. J
o
l
d
zi
c
, Z
. D
j
u
r
i
c,
an
d
P
. V
u
l
et
i
c, “
A
t
r
an
s
p
ar
en
t
an
d
s
ca
l
ab
l
e
an
o
m
al
y
-
b
a
se
d
Do
S
de
t
e
c
t
i
on m
e
t
hod,
”
Com
put
.
Ne
t
wor
k
s
,
vo
l
.
1
04,
pp
.
27
–
42
,
2
016.
[4]
.
O.
Y.
Al
-
J
ar
r
ah
,
O
. A
l
h
u
s
s
ei
n
, P
. D
. Y
o
o
, S
. M
u
h
ai
d
a
t
, K
.
T
ah
a,
an
d
K
. K
i
m
, “
D
at
a R
an
d
o
m
i
za
t
i
o
n
an
d
C
l
u
s
t
er
-
Ba
s
e
d P
a
rt
i
t
i
on
i
ng for Bot
n
e
t
Int
rus
i
on D
e
t
e
c
t
i
on
,
”
I
E
E
E
T
r
ans
.
Cy
be
r
n.
,
vol
.
46
,
no.
8,
pp.
1796
–
1806,
2016
.
[5]
.
A
.
A
.
A
burom
m
a
n a
nd M
.
Bi
n Ibne
Re
a
z
,
“
A
nove
l
S
V
M
-
k
NN
-
P
S
O
e
ns
e
m
bl
e
m
e
t
hod for i
nt
rus
i
on de
t
e
c
t
i
on
s
y
s
t
e
m
,
”
A
ppl
.
S
of
t
Com
put
.
J
.
,
v
ol
.
38
,
pp
.
360
–
3
72,
2016
.
[6]
.
P
.
T
e
i
s
s
e
y
re
,
“
N
e
uroc
om
put
i
ng
F
e
a
t
ure
r
a
nki
ng
for m
ul
t
i
-
la
b
e
l c
la
s
s
i
fi
c
a
t
i
on us
i
ng M
a
rkov ne
t
w
orks
,
”
vol
.
205
,
pp.
439
–
454
,
20
16.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
252
-
88
1
4
IJ
A
A
S
V
o
l
.
7
,
N
o
.
1
,
M
a
r
ch
2
018
:
7
–
12
12
[7]
.
S
G
an
ap
at
h
y
, P
V
i
j
a
y
ak
u
m
ar
, P
Y
o
g
es
h
, A
K
an
n
an
, “
A
n Int
e
l
l
i
ge
nt
CRF
Ba
s
e
d F
e
a
t
ure
S
e
l
e
c
t
i
on for E
ffe
c
t
i
v
e
Int
rus
i
on D
e
t
e
c
t
i
on
”,
In
t
e
r
nat
i
on
al
A
r
ab
J
our
nal
of
Inf
or
m
at
i
on
T
e
c
hnol
og
y
,
vol
.
16,
no
.
2
,
2016
.
[8]
.
V
.
Bol
ón
-
Ca
ne
d
o n,
I.
P
ort
o
-
D
í
a
z, N
. S
án
ch
e
z
-
M
a
roño,
A
.
A
l
ons
o
-
Be
t
a
n
z
os
,
“
A
fra
m
e
w
ork for c
os
t
-
ba
s
e
d fe
a
t
ur
e
s
e
le
c
tio
n
,
”
P
a
t
t
e
r
n
R
eco
g
n
i
t
i
o
n
, E
l
s
ev
i
er
,
vol
.
47,
pp.
2481
–
726
,
2
014.
[9]
.
M
. S
. R
aza
an
d
U
. Q
am
ar
, “
A
n
i
n
cr
em
en
t
al
d
e
p
en
d
en
c
y
cal
cu
l
at
i
o
n
t
ech
n
i
q
u
e
f
o
r fe
a
t
ur
e
s
e
l
e
c
t
i
on us
i
ng rough
s
et
s
,”
In
f
.
Sc
i
.
(
Ny
)
.
,
vol
.
343
–
344
,
pp
.
41
–
65
,
201
6.
[10]
.
L
. Y
u
, H
. L
i
u
,
“
E
f
f
i
ci
en
t
f
ea
t
u
r
e s
el
ec
t
i
o
n
v
i
a
an
al
y
s
i
s
o
f
r
el
e
v
an
ce an
d
r
ed
u
n
d
an
c
y
”
,
T
he
J
our
nal
of
Mac
hi
ne
L
e
ar
ni
ng R
e
s
e
ar
c
h,
v
ol
.
5,
p
p.
1
20
5
–
1224,
2004
.
[11]
.
G.
H.
Jo
h
n
,
R.
K
oha
vi
,
K
.
P
fl
e
ge
r,
e
t
a
l
.
,
“
Irre
l
e
va
n
t
fe
a
t
ure
s
a
nd t
he
s
ub
-
s
e
t
s
e
l
e
c
t
i
on
probl
e
m
”
,
i
n: Mac
hi
n
e
L
e
ar
ni
ng: P
r
oc
e
e
di
ngs
of
t
he
E
l
e
v
e
nt
h
Int
e
r
nat
i
o
nal
Conf
e
r
e
nc
e
,
pp.
121
–
129
,
19
94.
[12]
.
B.
G
re
c
huk,
A
.
M
ol
y
boha
,
M
.
Z
a
ba
ra
nk
i
n,
“
M
a
xi
m
um
e
nt
rop
y
pri
n
c
i
pl
e
w
i
t
h g
e
ne
ra
l
de
vi
a
t
i
on m
e
a
s
ure
s
”
,
Mat
he
m
at
i
c
s
of
O
pe
r
at
i
ons
R
e
s
e
ar
c
h
,
vol
.
34,
no.
2,
pp.
445
–
467,
2009.
[13]
.
Q
.
L
i
,
Z
.
S
un,
Z
.
L
i
n
,
a
nd R.
H
e
,
“
A
ut
hor ’ s
A
c
c
e
pt
e
d M
a
nus
c
ri
pt
T
ra
ns
form
a
t
i
o
n Inva
ri
a
n
t
S
ubs
pa
c
e
C
l
us
t
e
r
i
ng
R
ef
er
en
ce
:
T
o
a
p
p
ear
i
n
:
P
at
t
er
n
R
eco
g
n
i
t
i
o
n
,
”
2
016.
[14]
.
S
. M
al
d
o
n
ad
o
,
R
. W
eb
er
,
an
d
F
. F
am
i
l
i
, “
F
eat
u
r
e s
el
ect
i
o
n
f
o
r
h
i
g
h
-
di
m
e
ns
i
ona
l
c
l
a
s
s
-
i
m
ba
l
a
n
c
e
d da
t
a
s
e
t
s
us
i
ng
S
upport
V
e
c
t
or
M
a
c
hi
ne
s
,
”
In
f
.
Sc
i
.
(
Ny
)
.
,
vol
.
2
86,
pp
.
228
–
246
,
2014.
[15]
.
S
G
a
na
pa
t
h
y
,
R
S
e
t
hukka
ra
s
i
,
P
Y
oge
s
h,
P
V
i
j
a
y
a
kum
a
r,
A
K
a
nna
n,
“
A
n i
nt
e
l
l
i
ge
nt
t
e
m
pora
l
pa
t
t
e
rn c
l
a
s
s
i
fi
c
a
t
i
o
n
s
y
s
t
e
m
us
i
ng fu
z
z
y
t
e
m
pora
l
ru
l
e
s
a
nd pa
rt
i
c
l
e
s
w
a
rm
opt
i
m
i
z
a
t
i
on
”,
Sadhana
,
v
ol
.
3
9,
n
o.
2,
p
p.
28
3
-
302,
2014
.
[16]
.
S
G
a
na
pa
t
h
y
,
K
K
ul
ot
hunga
n,
P
Y
oge
s
h,
A
K
a
nna
n,
“
A No
v
e
l
W
e
i
ght
e
d F
uz
z
y
C
–
M
e
a
ns
Cl
u
s
t
e
ri
ng Ba
s
e
d o
n
Im
m
une
G
e
ne
t
i
c
A
l
gori
t
hm
for
I
nt
rus
i
on D
e
t
e
c
t
i
on
”,
P
r
oc
e
d
i
a E
ngi
ne
e
r
i
ng
,
vol
.
38,
pp
.
1750
-
17
57,
2012
.
[17]
.
K
K
ul
ot
hunga
n,
S
G
a
na
pa
t
h
y
,
S
Indra
G
a
ndhi
,
P
Y
oge
s
h,
A
K
a
nna
n,
“
Int
e
l
l
i
ge
n
t
s
e
c
ure
d fa
ul
t
t
ol
e
ra
n
t
rout
i
ng i
n
w
i
re
l
e
s
s
s
e
ns
or
ne
t
w
orks
us
i
ng
c
l
us
t
e
r
i
ng a
ppro
a
c
h
”,
Int
e
r
nat
i
o
nal
J
our
nal
of
Sof
t
Com
put
i
ng
,
vol
.
6,
no.
5,
p
p
.
210
-
215,
2011
.
[18]
.
S
.
G
a
na
pa
t
h
y
,
N
.
J
a
i
s
a
nka
r,
P
.
Y
o
ge
s
h,
A
.
K
a
nna
n
,
“
A
n Int
e
l
l
i
ge
nt
S
y
s
t
e
m
for Int
rus
i
on D
e
t
e
c
t
i
on us
i
ng O
ut
l
i
e
r
D
e
te
c
tio
n
”
,
201
1 Int
e
r
nat
i
ona
l
Conf
e
r
e
nc
e
on
R
e
c
e
nt
T
r
e
nds
i
n Inf
or
m
at
i
on T
e
c
hnol
og
y
(
ICR
T
IT
)
,
pp.
119
-
1
23,
2011.
[19]
.
N
J
ai
s
an
k
ar
, S
G
a
na
pa
t
h
y
,
P
Y
oge
s
h,
A
K
a
nna
n,
K
A
na
nd,
“
A
n i
nt
e
l
l
i
g
e
nt
a
ge
nt
ba
s
e
d
i
nt
rus
i
o
n de
t
e
c
t
i
on
s
y
s
t
e
m
us
i
ng
fuz
z
y
roug
h
s
e
t b
a
s
e
d
o
u
tli
e
r
d
e
te
c
tio
n
”
,
So
f
t
Com
put
i
ng
T
e
c
hni
que
s
i
n
V
i
s
i
on Sc
i
e
nc
e
,
pp.
1
47
-
153,
2012
.
[20]
.
A
.
H
.
S
ung
a
nd
S
.
M
ukka
m
a
l
a
,
“
Ide
nt
i
f
y
i
ng
I
m
port
a
nt
F
e
a
t
ur
e
s
for Int
rus
i
on
D
e
t
e
c
t
i
on
U
s
i
ng S
upport
V
e
c
t
or
M
a
c
hi
ne
s
a
nd
N
e
ura
l
N
e
t
w
or
ks
"
,
D
e
par
t
m
e
nt
of
Com
put
e
r
Sc
i
e
nc
e
Ne
w Me
x
i
c
o Ins
t
i
t
ut
e
of
Mi
ni
ng and
T
e
c
hnol
ogy
,
p
p
. 3
–
10,
2003
.
[21]
.
S
.
G
a
na
p
a
t
h
y
,
P
.
Y
oge
s
h,
a
nd
A
.
K
a
nn
a
n,
“
Int
e
l
l
i
ge
nt
A
ge
nt
-
B
a
s
e
d Int
rus
i
on D
e
t
e
c
t
i
on S
y
s
t
e
m
U
s
i
ng E
nha
nc
e
d
M
ul
t
i
c
l
a
s
s
S
V
M
,
”
vo
l
.
2012,
201
2.
Evaluation Warning : The document was created with Spire.PDF for Python.