I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
pu
t
er
E
ng
ineering
(
I
J
E
CE
)
Vo
l.
14
,
No
.
6
,
Dec
em
b
er
20
24
,
p
p
.
6
9
5
8
~
6
9
6
8
I
SS
N:
2088
-
8
7
0
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijece.
v
14
i
6
.
pp
6
9
5
8
-
6
9
6
8
6958
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m
A
sig
nificant
f
ea
t
ures v
ector
for in
terne
t
tra
ff
ic
cla
ss
ificatio
n
ba
sed o
n multi
-
fe
a
tures selec
tion t
e
chniques a
nd
ran
ker,
v
o
ting
filt
ers
Alha
m
za
M
un
t
her
1
,
M
o
s
le
h
M
.
Abua
lha
j
2
,
Ala
ba
s
s
A
la
lo
us
i
3
,
H
ila
l A
.
F
a
dh
il
4
1
I
n
f
o
r
mat
i
o
n
T
e
c
h
n
o
l
o
g
y
D
e
p
a
r
t
me
n
t
,
C
o
l
l
e
g
e
o
f
C
o
m
p
u
t
i
n
g
a
n
d
I
n
f
o
r
m
a
t
i
o
n
S
c
i
e
n
c
e
s
,
U
n
i
v
e
r
si
t
y
T
e
c
h
n
o
l
o
g
y
a
n
d
A
p
p
l
i
e
d
S
c
i
e
n
c
e
s,
S
u
r
,
O
ma
n
2
F
a
c
u
l
t
y
o
f
I
n
f
o
r
m
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
,
A
l
-
A
h
l
i
y
y
a
A
mm
a
n
U
n
i
v
e
r
s
i
t
y
,
A
m
m
a
n
,
Jo
r
d
a
n
3
I
n
f
o
r
mat
i
o
n
T
e
c
h
n
o
l
o
g
y
D
e
p
a
r
t
me
n
t
,
F
a
c
u
l
t
y
o
f
I
n
f
o
r
m
a
t
i
c
s
a
n
d
C
o
m
p
u
t
e
r
s,
T
h
a
m
a
r
U
n
i
v
e
r
si
t
y
,
T
h
a
m
a
r
,
Y
e
me
n
4
El
e
c
t
r
i
c
a
l
a
n
d
C
o
mp
u
t
e
r
E
n
g
i
n
e
e
r
i
n
g
P
r
o
g
r
a
m
,
En
g
i
n
e
e
r
i
n
g
F
a
c
u
l
t
y
,
S
o
h
a
r
U
n
i
v
e
r
s
i
t
y
,
S
o
h
a
r
,
O
ma
n
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Oct
3
0
,
2
0
2
3
R
ev
is
ed
J
u
l 1
6
,
2
0
2
4
Acc
ep
ted
Au
g
6
,
2
0
2
4
Th
e
p
u
rs
u
it
o
f
e
ffe
c
ti
v
e
m
o
d
e
ls
with
h
i
g
h
d
e
tec
ti
o
n
a
c
c
u
ra
c
y
h
a
s
sp
a
rk
e
d
g
re
a
t
in
tere
st
i
n
a
n
o
m
a
ly
d
e
tec
ti
o
n
o
f
i
n
tern
e
t
traffic.
T
h
e
issu
e
stil
l
li
e
s
i
n
c
re
a
ti
n
g
a
tru
stwo
rth
y
a
n
d
e
ffe
c
ti
v
e
a
n
o
m
a
ly
d
e
tec
ti
o
n
sy
ste
m
th
a
t
c
a
n
h
a
n
d
le
m
a
ss
iv
e
d
a
ta
v
o
l
u
m
e
s
a
n
d
p
a
tt
e
r
n
s
th
a
t
c
h
a
n
g
e
in
re
a
l
-
ti
m
e
.
Th
e
d
e
tec
ti
o
n
tec
h
n
iq
u
e
s
u
se
d
,
e
sp
e
c
ially
th
e
fe
a
tu
re
se
lec
ti
o
n
m
e
th
o
d
s
a
n
d
m
a
c
h
in
e
lea
rn
in
g
a
lg
o
rit
h
m
s,
a
re
c
ru
c
ial
to
th
e
d
e
sig
n
o
f
su
c
h
a
sy
ste
m
.
Th
e
fu
n
d
a
m
e
n
tal
d
iffi
c
u
l
ty
i
n
fe
a
tu
re
se
lec
ti
o
n
is
se
lec
ti
n
g
a
sm
a
ll
e
r
su
b
se
t
o
f
fe
a
tu
re
s
th
a
t
a
re
m
o
re
re
late
d
to
th
e
c
las
s
b
u
t
a
re
les
s
n
u
m
e
ro
u
s
.
To
re
d
u
c
e
th
e
d
ime
n
sio
n
a
li
ty
o
f
t
h
e
d
a
tas
e
t,
th
is
re
se
a
rc
h
o
ffe
re
d
a
m
u
l
ti
-
fe
a
tu
re
se
lec
ti
o
n
tec
h
n
i
q
u
e
(M
F
S
T)
u
sin
g
f
o
u
r
fil
ter
tec
h
n
i
q
u
e
s:
fa
st
c
o
rre
latio
n
-
b
a
se
d
fil
ter,
sig
n
ifi
c
a
n
c
e
fe
a
tu
re
e
v
a
l
u
a
to
r,
c
h
i
-
sq
u
a
re
,
a
n
d
g
a
in
r
a
ti
o
.
Eac
h
tec
h
n
iq
u
e
'
s
o
u
t
p
u
t
v
e
c
to
r
is
p
u
t
v
ia
ra
n
k
e
r
a
n
d
B
o
rd
a
v
o
ti
n
g
fi
lt
e
rs.
Th
e
fe
a
tu
re
with
t
h
e
h
ig
h
e
st
n
u
m
b
e
r
o
f
v
o
tes
a
n
d
ra
n
k
v
a
l
u
e
s
will
b
e
se
lec
ted
fro
m
th
e
d
a
tas
e
t.
T
h
e
p
e
rf
o
rm
a
n
c
e
o
f
t
h
e
g
i
v
e
n
M
F
S
T
fra
m
e
wo
r
k
wa
s
th
e
b
e
st
wh
e
n
c
o
m
p
a
re
d
t
o
t
h
e
fo
u
r
stra
teg
ies
li
ste
d
a
b
o
v
e
fu
n
c
ti
o
n
i
n
g
a
lo
n
e
;
th
re
e
d
iffere
n
t
c
las
sifiers
we
re
e
m
p
lo
y
e
d
t
o
tes
t
t
h
e
a
c
c
u
ra
c
y
.
C
4
.
5
,
n
a
v
e
Ba
y
e
s,
a
n
d
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
.
T
h
e
e
x
p
e
rime
n
t
o
u
tc
o
m
e
s
e
m
p
lo
y
e
d
ten
d
a
tas
e
ts
o
f
d
iffere
n
t
siz
e
s
wit
h
1
0
,
0
0
0
-
3
0
0
,
0
0
0
in
sta
n
c
e
s.
O
n
ly
8
o
u
t
o
f
2
4
8
c
h
a
ra
c
teristics
we
re
c
h
o
se
n
,
with
c
las
sifiers
p
e
rc
e
n
tag
e
s
a
v
e
ra
g
in
g
6
5
%
,
9
3
.
8
%
,
a
n
d
9
5
.
5
%
.
K
ey
w
o
r
d
s
:
Data
m
in
in
g
Data
r
ed
u
ctio
n
Featu
r
es selectio
n
Fil
ter
f
ea
tu
r
e
s
elec
tio
n
Ma
ch
in
e
lear
n
in
g
Su
p
er
v
is
ed
lear
n
in
g
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
:
Alh
am
za
Mu
n
th
er
I
T
Dep
ar
tm
en
t,
C
o
lleg
e
o
f
C
o
m
p
u
tin
g
a
n
d
I
n
f
o
r
m
atio
n
Scie
n
ce
s
,
Un
iv
er
s
ity
T
ec
h
n
o
lo
g
y
a
n
d
Ap
p
lied
Scien
ce
s
4
1
1
Alh
u
s
ain
y
a
Stre
et,
Su
r
,
O
m
an
E
m
ail:
alh
am
za
.
war
d
i@
u
tas.ed
u
.
o
m
1.
I
NT
RO
D
UCT
I
O
N
A
co
m
m
o
n
tec
h
n
iq
u
e
f
o
r
p
r
ep
ar
in
g
d
ata
in
m
ac
h
in
e
le
ar
n
in
g
is
f
ea
tu
r
e
s
elec
tio
n
o
r
v
ar
ia
b
le
s
elec
tio
n
.
I
t
is
a
d
im
en
s
io
n
al
ity
r
ed
u
ctio
n
tech
n
i
q
u
e
p
r
im
a
r
ily
u
s
ed
to
r
em
o
v
e
u
n
n
ec
ess
ar
y
an
d
u
n
d
esire
d
f
ea
tu
r
es
f
r
o
m
a
n
y
d
ataset
[
1
]
.
T
o
d
iag
n
o
s
e
an
o
m
alies
in
I
n
te
r
n
et
tr
af
f
ic,
th
e
f
ea
tu
r
e
s
elec
tio
n
p
r
o
ce
s
s
r
em
o
v
es
r
ed
u
n
d
an
t
an
d
p
o
in
tles
s
attr
ib
u
tes
f
r
o
m
th
e
d
ataset.
I
n
n
etw
o
r
k
s
ec
u
r
ity
,
it
ca
n
p
in
p
o
in
t
t
h
e
m
o
s
t
s
ig
n
if
ican
t
f
ac
to
r
s
co
n
n
ec
te
d
to
a
ce
r
tai
n
attac
k
.
An
o
m
al
y
d
etec
tio
n
s
g
r
ea
tly
b
en
e
f
it
th
e
n
etwo
r
k
s
ec
u
r
ity
s
ec
to
r
as
d
ec
is
io
n
-
m
ak
in
g
h
as
b
ec
o
m
e
ea
s
ier
an
d
f
aster
with
th
e
r
ap
id
ac
cu
m
u
latio
n
o
f
h
ig
h
-
th
r
o
u
g
h
p
u
t
tech
n
o
lo
g
y
an
d
cu
ttin
g
-
e
d
g
e
m
ac
h
i
n
e
lear
n
in
g
m
eth
o
d
o
lo
g
ies.
Ma
ch
in
e
lear
n
in
g
is
ess
en
tial in
h
elp
in
g
n
etwo
r
k
o
p
e
r
ato
r
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
A
s
ig
n
ifica
n
t fe
a
tu
r
es v
ec
to
r
fo
r
in
tern
et
tr
a
ffic c
la
s
s
i
fica
tio
n
b
a
s
ed
o
n
…
(
A
l
h
a
mza
Mu
n
t
h
er
)
6959
an
aly
ze
th
e
r
is
k
v
ar
iab
les
ass
o
ciate
d
with
a
s
p
ec
if
ic
ass
au
lt
q
u
ick
ly
[
2
]
.
Ad
d
itio
n
ally
,
t
h
ese
s
o
lu
tio
n
s
s
u
p
p
o
r
t
s
ec
u
r
ity
ad
m
in
is
tr
ato
r
s
i
n
m
ak
in
g
wis
e
d
ec
is
io
n
s
.
R
esear
ch
er
s
ar
e
b
ein
g
p
u
s
h
ed
to
d
ay
to
u
s
e
f
ea
tu
r
e
r
ed
u
ctio
n
a
n
d
class
if
icatio
n
alg
o
r
ith
m
s
f
o
r
e
f
f
icien
t
attac
k
d
iag
n
o
s
is
.
Ma
ch
in
e
lear
n
in
g
is
ess
en
tial
in
h
elp
in
g
n
etwo
r
k
o
p
er
ato
r
s
an
aly
ze
th
e
r
is
k
f
ea
tu
r
es
ass
o
ciate
d
with
a
s
p
ec
if
ic
ass
au
lt
q
u
ick
ly
.
Ad
d
itio
n
ally
,
t
h
ese
s
o
lu
tio
n
s
s
u
p
p
o
r
t
s
ec
u
r
ity
an
aly
s
ts
in
m
ak
in
g
wis
e
d
ec
i
s
i
o
n
s
.
R
esear
ch
er
s
ar
e
b
ein
g
p
u
s
h
ed
to
d
ay
to
u
s
e
f
ea
tu
r
e
r
ed
u
ctio
n
an
d
class
if
icatio
n
alg
o
r
ith
m
s
f
o
r
ef
f
icien
t
attac
k
d
iag
n
o
s
is
.
T
o
ef
f
ec
tiv
ely
class
if
y
I
n
ter
n
et
tr
af
f
ic,
class
if
icatio
n
is
a
d
ata
m
in
in
g
task
.
Fr
o
m
tr
ai
n
in
g
d
ata,
it
f
ir
s
t
cr
ea
tes
a
m
o
d
el.
N
ex
t,
it
is
ap
p
lied
to
an
ticip
ate
n
ew
in
s
tan
ce
s
f
o
r
wh
ich
th
e
class
v
alu
es
ar
e
u
n
k
n
o
wn
.
B
ef
o
r
e
class
if
y
in
g
,
u
n
n
ec
ess
ar
y
an
d
r
ed
u
n
d
an
t
attr
ib
u
tes
ca
n
b
e
r
em
o
v
ed
f
r
o
m
th
e
d
ataset
u
s
in
g
th
e
p
r
o
p
er
f
ea
t
u
r
e
s
elec
tio
n
tech
n
i
q
u
e.
As
a
r
esu
lt,
class
if
ier
ac
cu
r
ac
y
is
im
p
r
o
v
e
d
b
ec
au
s
e
it
ju
s
t
n
ee
d
s
to
ex
am
in
e
th
e
d
ataset'
s
k
ey
f
ea
tu
r
es.
T
h
e
class
if
ier
m
o
d
el
p
er
f
o
r
m
s
b
etter
wh
en
p
r
o
p
e
r
ly
ch
o
s
en
f
ea
tu
r
es,
d
ec
r
ea
s
in
g
co
m
p
u
tatio
n
al
co
m
p
lex
ity
an
d
ex
ec
u
tio
n
tim
e.
I
t a
ls
o
s
im
p
lifie
s
d
ata
v
is
u
aliza
tio
n
an
d
b
o
o
s
t
s
th
e
d
ata'
s
r
ea
d
ab
ilit
y
.
Gen
er
all
y
,
t
h
r
e
e
b
as
ic
ca
te
g
o
r
ies
o
f
f
ea
t
u
r
e
s
el
ec
t
io
n
s
t
r
at
eg
i
es
f
a
ll
u
n
d
e
r
t
h
e
f
ilt
er
,
wr
ap
p
e
r
,
a
n
d
h
y
b
r
i
d
m
o
d
els
.
T
h
e
f
ilt
e
r
t
ec
h
n
i
q
u
e
r
el
ies
o
n
th
e
c
o
m
m
o
n
t
r
a
its
o
f
t
h
e
t
r
ai
n
i
n
g
d
a
ta
to
c
h
o
o
s
e
f
ea
tu
r
es
in
d
e
p
e
n
d
e
n
t
ly
o
f
a
n
y
class
if
i
er
.
T
h
e
a
d
v
a
n
ta
g
e
o
f
t
h
is
t
y
p
e
is
h
i
g
h
p
r
o
c
ess
i
n
g
s
p
e
ed
.
Ho
w
ev
er
,
it
h
as
lo
w
ac
c
u
r
a
cy
[
1
]
,
[
3
]
,
[
4
]
.
E
x
am
p
l
es
o
f
t
h
is
t
ec
h
n
i
q
u
e
a
r
e
R
eli
ef
[
5
]
a
n
d
c
o
r
r
el
ati
o
n
f
ea
t
u
r
es
s
ele
cti
o
n
c
o
r
r
e
lat
io
n
f
e
at
u
r
es
s
ele
cti
o
n
(
C
FS
)
[
6
]
,
f
ast
co
r
r
el
ati
o
n
b
ase
d
f
ea
t
u
r
es
s
ele
cti
o
n
f
ast
co
r
r
el
ati
o
n
-
b
ase
d
f
e
at
u
r
e
s
el
ec
t
io
n
(
FC
B
F)
a
n
d
ch
i
-
s
q
u
a
r
e
(
C
h
i
2
)
[
7
]
,
a
n
d
i
n
f
o
r
m
at
io
n
g
ai
n
[
8
]
.
W
r
a
p
p
e
r
s
f
ea
t
u
r
e
s
el
ec
ti
o
n
tec
h
n
i
q
u
es
.
T
h
is
m
et
h
o
d
u
s
es
a
s
p
ec
if
i
c
class
i
f
i
er
to
m
ea
s
u
r
e
t
h
e
s
i
g
n
if
ica
n
ce
o
f
t
h
e
f
e
at
u
r
e
s
e
t.
He
n
c
e,
t
h
is
t
y
p
e
d
e
p
e
n
d
s
o
n
t
h
e
class
i
f
i
ca
t
io
n
m
et
h
o
d
.
T
h
is
m
eth
o
d
i
n
t
r
o
d
u
ce
s
a
b
e
tte
r
p
e
r
f
o
r
m
a
n
ce
t
h
a
n
t
h
e
f
i
lte
r
m
et
h
o
d
r
e
g
a
r
d
in
g
ac
c
u
r
a
cy
class
i
f
i
ca
t
io
n
.
H
o
we
v
er
,
w
r
a
p
p
e
r
m
et
h
o
d
s
ca
u
s
e
a
n
e
x
p
e
n
s
iv
e
c
o
m
p
u
t
ati
o
n
c
o
s
t
[
3
]
,
[
4
]
.
T
h
ese
tec
h
n
i
q
u
es
in
c
lu
d
e
g
en
eti
c
al
g
o
r
i
th
m
[
9
]
an
d
a
n
ts
c
o
l
o
n
y
[
1
0
]
.
A
h
y
b
r
i
d
m
e
th
o
d
o
l
o
g
y
t
h
a
t
c
o
m
b
i
n
es
t
h
e
b
e
n
e
f
its
o
f
th
e
two
p
r
e
ce
d
i
n
g
a
p
p
r
o
a
ch
es
is
s
u
g
g
este
d
.
W
e
m
u
s
t
lis
t
s
o
m
e
o
f
th
ese
m
e
th
o
d
s
f
o
r
c
lass
i
f
y
i
n
g
n
etw
o
r
k
tr
af
f
i
c
as
[
1
1
]
,
[
1
2
]
.
Ho
we
v
e
r
,
th
is
a
p
p
r
o
a
ch
w
ill
c
a
r
r
y
o
v
er
s
o
m
e
d
r
a
wb
a
c
k
s
al
o
n
g
w
it
h
th
e
b
e
n
e
f
it
s
o
f
b
o
t
h
.
T
h
is
p
a
p
er
ad
o
p
te
d
th
e
f
ilt
e
r
te
c
h
n
iq
u
e
d
u
e
t
o
t
h
e
f
i
lte
r
m
e
th
o
d
'
s
q
u
i
ck
p
r
o
ce
s
s
i
n
g
an
d
th
e
f
a
ct
t
h
a
t
ti
m
e
p
r
o
c
ess
i
n
g
is
t
h
e
m
o
s
t i
m
p
o
r
t
a
n
t
as
p
ec
t
o
f
I
n
te
r
n
et
t
r
a
f
f
ic
c
lass
i
f
i
ca
t
io
n
,
p
a
r
t
ic
u
la
r
l
y
i
n
o
n
li
n
e
a
n
d
r
ea
l
-
t
im
e
s
etti
n
g
s
.
T
h
is
s
tu
d
y
em
p
lo
y
s
f
o
u
r
f
ilter
in
g
m
eth
o
d
s
:
ch
i
-
s
q
u
ar
e
,
g
ai
n
r
atio
,
s
ig
n
if
ican
ce
f
ea
tu
r
e
ev
al
u
ato
r
,
an
d
f
ast
co
r
r
elatio
n
-
b
ased
f
ilter
.
R
an
k
er
an
d
B
o
r
d
a
'
s
v
o
tin
g
f
ilter
s
ar
e
ap
p
lied
to
th
e
o
u
tp
u
t
v
ec
to
r
o
f
ea
ch
m
eth
o
d
.
On
l
y
th
e
f
ea
tu
r
es
wit
h
th
e
m
o
s
t
v
o
tes
an
d
r
an
k
v
alu
es
will
b
e
ch
o
s
en
.
T
h
e
n
e
x
t
s
ec
tio
n
will
p
r
esen
t
th
e
s
tu
d
ies
r
elate
d
to
th
e
wo
r
k
,
an
d
th
en
th
e
m
eth
o
d
o
lo
g
y
will
b
e
p
r
esen
ted
.
Su
b
s
eq
u
en
tl
y
,
th
e
d
ataset
d
etails
will
b
e
ex
p
lo
r
ed
,
th
e
r
esu
lts
a
n
d
an
al
y
s
is
will
b
e
d
is
cu
s
s
ed
,
an
d
last
ly
,
th
e
co
n
cl
u
s
io
n
o
f
t
h
is
r
esear
ch
will
b
e
in
tr
o
d
u
ce
d
.
T
h
e
s
tu
d
ies
th
at
ar
e
r
elev
an
t
to
t
h
e
wo
r
k
will
b
e
p
r
esen
ted
in
th
e
f
o
llo
win
g
s
ec
tio
n
.
Af
ter
th
at,
th
e
m
eth
o
d
o
l
o
g
y
o
f
t
h
e
tech
n
iq
u
e
will
b
e
d
escr
ib
ed
.
T
h
e
d
atase
t
s
p
ec
if
ics
will
th
en
b
e
e
x
p
lo
r
ed
.
T
h
e
r
esu
lts
an
d
an
aly
s
is
will th
en
b
e
co
v
er
e
d
.
Fin
ally
,
th
e
r
esear
ch
'
s
co
n
clu
s
io
n
will b
e
in
tr
o
d
u
ce
d
.
2.
RE
L
AT
E
D
WO
RK
S
T
h
e
p
r
e
-
p
r
o
ce
s
s
in
g
s
tag
e
o
f
f
ea
tu
r
e
s
elec
tio
n
is
in
cr
ea
s
in
g
ly
im
p
o
r
tan
t
in
d
ev
el
o
p
i
n
g
v
ar
io
u
s
s
y
s
tem
s
f
o
r
m
o
n
ito
r
i
n
g
in
te
r
n
et
tr
af
f
ic
a
n
d
s
p
o
ttin
g
an
o
m
alies
an
d
attac
k
s
b
ec
a
u
s
e
o
f
th
e
o
n
g
o
in
g
d
ev
elo
p
m
e
n
t
o
f
d
ata
d
im
e
n
s
io
n
ality
.
T
h
er
e
f
o
r
e,
o
n
e
o
f
th
e
d
if
f
icu
lties
m
o
s
t
r
esear
ch
er
s
en
co
u
n
ter
is
ch
o
o
s
in
g
th
e
m
o
s
t
im
p
o
r
tan
t
f
ea
tu
r
e
s
et.
T
h
e
g
o
al
o
f
th
e
s
elec
tio
n
p
r
o
ce
s
s
s
h
o
u
ld
b
e
to
less
en
th
e
d
ata
d
im
en
s
io
n
ality
wh
ile
m
ain
tain
in
g
h
i
g
h
ac
cu
r
a
cy
.
B
ased
o
n
th
e
b
est
f
ea
tu
r
es
o
f
th
e
n
etwo
r
k
tr
an
s
ac
tio
n
d
ata
th
at
wer
e
m
ad
e
av
ailab
le
f
o
r
tr
ain
in
g
,
Aljawa
r
n
eh
et
a
l.
[
1
3
]
d
ev
el
o
p
ed
a
h
y
b
r
id
m
o
d
el
th
at
ca
n
b
e
u
tili
ze
d
to
d
eter
m
in
e
th
e
in
tr
u
s
io
n
s
co
p
e
th
r
esh
o
ld
d
eg
r
ee
.
T
h
e
v
o
te
alg
o
r
ith
m
with
in
f
o
r
m
atio
n
g
ai
n
was
u
s
ed
to
f
ilter
th
e
d
ata,
co
m
b
in
in
g
th
e
b
ase
lear
n
er
s
'
p
r
o
b
ab
ilit
y
d
is
tr
ib
u
ti
o
n
s
to
s
elec
t
th
e
im
p
o
r
tan
t
f
e
atu
r
es
th
at
im
p
r
o
v
e
th
e
ac
cu
r
a
cy
o
f
th
e
s
u
g
g
ested
m
o
d
el.
T
h
e
au
th
o
r
s
u
s
ed
th
e
KDD
d
ataset,
wh
ich
h
as 4
1
f
ea
tu
r
es.
Dif
f
er
en
t c
lass
if
ier
s
b
a
s
ed
o
n
d
ec
is
io
n
tr
ee
s
wer
e
em
p
lo
y
ed
f
o
r
class
if
icatio
n
.
Am
b
u
s
aid
i
et
a
l.
[
1
4
]
s
u
g
g
ested
a
m
u
tu
al
in
f
o
r
m
atio
n
-
b
ased
ap
p
r
o
ac
h
f
o
r
ch
o
o
s
in
g
th
e
b
est
f
ea
tu
r
e
f
o
r
class
if
icatio
n
th
r
o
u
g
h
a
n
aly
tical
s
elec
tio
n
.
T
h
is
m
u
tu
al
in
f
o
r
m
atio
n
-
b
ased
f
ea
tu
r
e
s
elec
tio
n
alg
o
r
ith
m
ca
n
h
an
d
le
f
ea
tu
r
es
with
lin
ea
r
an
d
n
o
n
lin
ea
r
d
e
p
en
d
e
n
cies
in
th
e
d
ata.
I
n
th
e
in
s
tan
ce
s
o
f
n
etwo
r
k
in
tr
u
s
io
n
d
etec
tio
n
,
its
ef
f
ec
tiv
en
ess
i
s
ass
es
s
ed
.
W
ith
th
e
h
elp
o
f
th
e
f
ea
tu
r
es
ch
o
s
en
b
y
th
eir
s
u
g
g
ested
f
ea
tu
r
e
s
elec
tio
n
tech
n
i
q
u
e,
a
n
in
t
r
u
s
io
n
d
etec
tio
n
s
y
s
tem
(
I
DS)
k
n
o
wn
as
least
s
q
u
ar
e
s
u
p
p
o
r
t
v
ec
t
o
r
m
ac
h
i
n
e
-
b
as
ed
I
DS
(
L
SS
VM
-
I
DS)
is
c
o
n
s
tr
u
cted
.
Utilizin
g
th
r
ee
i
n
tr
u
s
io
n
d
etec
tio
n
ass
es
s
m
en
t
d
atasets
—
KD
D
C
u
p
9
9
,
NSL
-
KDD,
an
d
Ky
o
to
2
0
0
6
+d
ataset
—
th
e
p
er
f
o
r
m
a
n
ce
o
f
L
SS
VM
-
I
DS
is
ass
e
s
s
ed
.
E
ac
h
d
ataset
co
n
tain
ed
4
1
,
4
1
,
a
n
d
2
4
f
ea
tu
r
es,
r
esp
ec
tiv
ely
.
T
h
e
au
th
o
r
s
s
u
cc
e
s
s
f
u
lly
r
ed
u
ce
d
th
e
n
u
m
b
er
o
f
f
ea
tu
r
es
in
ea
c
h
d
ataset
ab
o
v
e
to
1
9
,
1
8
,
a
n
d
4
f
ea
tu
r
es.
Z
h
o
u
et
a
l.
[
1
5
]
p
r
o
p
o
s
ed
an
in
tr
u
s
io
n
d
etec
tio
n
f
r
a
m
ewo
r
k
b
ased
o
n
f
ea
tu
r
e
s
elec
tio
n
an
d
en
s
e
m
b
le
lear
n
in
g
ap
p
r
o
ac
h
es.
T
h
e
au
th
o
r
s
u
tili
ze
d
a
h
eu
r
is
tic
tech
n
iq
u
e
f
o
r
d
im
en
s
io
n
ality
r
ed
u
ctio
n
,
wh
ich
s
el
ec
ts
th
e
b
est
s
u
b
s
et
b
ased
o
n
f
ea
tu
r
e
co
r
r
elatio
n
.
T
h
en
,
th
ey
p
r
esen
te
d
a
h
y
b
r
id
s
tr
ateg
y
in
co
r
p
o
r
atin
g
th
e
C
4
.
5
,
r
an
d
o
m
f
o
r
est
(
R
F),
an
d
f
o
r
est
b
y
p
en
alizin
g
attr
ib
u
tes
(
Fo
r
est
PA)
alg
o
r
ith
m
s
.
Fin
ally
,
f
o
r
attac
k
r
ec
o
g
n
itio
n
,
th
e
v
o
tin
g
m
ec
h
an
is
m
is
u
tili
ze
d
to
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
14
,
No
.
6
,
Dec
em
b
e
r
20
24
:
6
9
5
8
-
6
9
6
8
6960
ag
g
r
eg
ate
th
e
p
r
o
b
ab
ilit
y
d
is
tr
ib
u
tio
n
s
o
f
t
h
e
b
ase
lear
n
er
s
.
Fo
r
th
e
e
x
p
er
im
e
n
tal
r
esu
lt
s
,
th
ey
u
s
ed
th
r
ee
s
ep
ar
ate
d
atasets
.
A
n
ewly
c
r
ea
ted
an
o
m
aly
d
etec
tio
n
s
y
s
tem
b
u
ilt
o
n
th
e
n
atu
r
al
f
u
s
io
n
o
f
v
a
r
io
u
s
d
ee
p
lear
n
in
g
m
eth
o
d
s
was
p
r
esen
t
ed
b
y
Z
h
o
n
g
et
a
l.
[
1
6
]
.
T
h
e
f
ir
s
t
s
tep
was
t
o
e
x
tr
ac
t
f
ea
t
u
r
es
f
r
o
m
n
etwo
r
k
tr
af
f
ic
u
s
in
g
th
e
d
am
p
e
d
in
c
r
e
m
en
tal
s
tatis
tics
alg
o
r
ith
m
.
T
h
e
s
ec
o
n
d
s
tep
in
v
o
lv
ed
tr
ain
i
n
g
th
e
au
to
e
n
co
d
e
r
with
a
s
m
all
am
o
u
n
t o
f
lab
el
d
ata.
T
h
e
th
ir
d
s
tep
in
v
o
lv
ed
u
s
in
g
th
e
au
t
o
en
co
d
er
to
m
ar
k
n
etwo
r
k
tr
af
f
ic
with
an
ab
n
o
r
m
al
s
co
r
e.
Fin
ally
,
t
h
e
ab
n
o
r
m
al
s
co
r
e
lab
el
d
ata
was
u
s
ed
to
tr
ain
th
e
l
o
n
g
-
s
h
o
r
t
-
ter
m
m
em
o
r
y
.
A
ty
p
e
o
f
r
e
cu
r
r
e
n
t
n
e
u
r
al
n
etwo
r
k
c
alled
L
STM
o
u
tp
e
r
f
o
r
m
s
co
n
v
en
tio
n
al
r
ec
u
r
r
en
t
n
eu
r
al
n
etwo
r
k
s
in
ter
m
s
o
f
m
e
m
o
r
y
.
Salo
et
a
l.
[
1
7
]
p
r
o
p
o
s
ed
an
en
s
em
b
le
class
if
ier
b
ased
o
n
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
i
n
e
(
SVM)
,
in
s
tan
ce
-
b
ased
lear
n
in
g
al
g
o
r
i
th
m
s
(
I
B
K)
,
an
d
m
u
ltil
ay
er
p
er
ce
p
tr
o
n
(
ML
P),
with
an
in
f
o
r
m
atio
n
g
ai
n
(
I
G)
an
d
p
r
i
n
cip
al
co
m
p
o
n
e
n
t
an
al
y
s
is
(
PC
A)
-
b
ased
d
im
en
s
io
n
ality
r
ed
u
ctio
n
tech
n
iq
u
e
f
o
r
i
n
tr
u
s
io
n
d
etec
tio
n
.
I
SC
X
2
0
1
2
,
NSL
-
KDD,
an
d
Ky
o
to
2
0
0
6
+
wer
e
u
s
ed
as
t
est
d
atasets
to
ev
alu
ate
th
e
p
er
f
o
r
m
a
n
ce
o
f
th
is
IG
-
PC
A
-
E
n
s
em
b
le
ap
p
r
o
ac
h
.
T
h
e
ex
p
er
i
m
en
tal
f
in
d
i
n
g
s
s
h
o
w
th
at
th
e
s
u
g
g
ested
h
y
b
r
id
d
im
e
n
s
io
n
ality
r
ed
u
ctio
n
s
tr
ateg
y
with
th
e
e
n
s
em
b
le
o
f
b
ase
lea
r
n
er
s
o
f
f
er
s
m
o
r
e
k
ey
q
u
alities
an
d
p
er
f
o
r
m
s
i
n
d
iv
id
u
al
s
tr
ateg
ies
r
eg
ar
d
in
g
ac
c
u
r
ac
y
an
d
f
alse
alar
m
r
ates.
A
c
o
m
p
ar
is
o
n
o
f
th
e
p
r
o
p
o
s
ed
tec
h
n
iq
u
e
to
co
m
p
ar
ab
le
wo
r
k
is
u
n
d
e
r
tak
en
,
an
d
th
e
y
f
in
d
t
h
at
th
e
s
u
g
g
ested
I
G
-
PC
A
-
E
n
s
em
b
le
m
eth
o
d
p
er
f
o
r
m
s
b
etter
i
n
ter
m
s
o
f
class
if
icatio
n
ac
cu
r
ac
y
.
Ab
d
u
llah
et
a
l.
[
1
8
]
d
ev
elo
p
e
d
an
a
p
p
r
o
ac
h
to
p
a
r
titi
o
n
th
e
in
p
u
t
d
ataset
in
to
v
ar
io
u
s
s
u
b
g
r
o
u
p
s
b
ased
o
n
ea
c
h
att
ac
k
.
T
h
en
,
f
o
r
ea
c
h
s
u
b
s
et,
t
h
ey
u
s
ed
a
f
ea
t
u
r
e
s
elec
tio
n
tech
n
iq
u
e
with
an
in
f
o
r
m
atio
n
g
ain
f
ilter
.
T
h
e
b
e
s
t
f
ea
tu
r
e
s
et
is
th
en
c
o
n
s
tr
u
ct
ed
b
y
co
m
b
in
in
g
th
e
lis
t
o
f
f
e
atu
r
e
s
ets
co
llected
f
o
r
ea
ch
attac
k
.
T
h
e
f
in
d
in
g
s
o
f
ex
p
e
r
im
en
ts
o
n
t
h
e
NSL
-
KDD
d
ataset
s
h
o
w
th
at
th
e
p
r
o
p
o
s
ed
s
tr
ateg
y
f
o
r
f
ea
tu
r
e
s
elec
tio
n
with
f
ewe
r
ch
ar
ac
ter
is
tics
im
p
r
o
v
es
s
y
s
tem
ac
cu
r
ac
y
wh
ile
m
i
n
im
izin
g
co
m
p
le
x
ity
.
Mo
u
s
taf
a
et
a
l.
[
1
9
]
p
r
esen
ted
an
en
s
em
b
le
in
tr
u
s
io
n
d
etec
ti
o
n
tech
n
iq
u
e
to
r
ed
u
ce
h
ar
m
f
u
l e
v
en
ts
.
I
n
in
ter
n
et
o
f
th
in
g
s
(
I
o
T
)
n
etwo
r
k
tr
af
f
i
c,
s
p
ec
if
ically
b
o
tn
et
ass
au
lts
o
n
th
e
d
o
m
ai
n
n
am
e
s
y
s
tem
(
DNS)
,
h
y
p
e
r
tex
t
tr
an
s
f
er
p
r
o
to
co
l
(
HT
T
P)
,
an
d
m
ess
ag
e
q
u
eu
in
g
telem
etr
y
tr
an
s
p
o
r
t
(
MQ
T
T
)
p
r
o
to
c
o
ls
u
s
ed
in
I
o
T
n
etwo
r
k
s
.
T
h
e
m
eth
o
d
s
d
ev
el
o
p
n
ew
s
tatis
tical
f
lo
w
f
ea
tu
r
es
b
ased
o
n
ex
am
in
in
g
th
eir
p
r
o
s
p
ec
tiv
e
attr
ib
u
tes.
T
h
en
,
to
ef
f
ec
tiv
ely
d
etec
t
d
etr
im
en
tal
o
cc
u
r
r
e
n
ce
s
,
an
Ad
aBo
o
s
t
en
s
em
b
le
lear
n
in
g
s
y
s
tem
is
d
ev
elo
p
ed
u
s
in
g
th
r
ee
m
ac
h
in
e
lear
n
i
n
g
tech
n
iq
u
es:
d
ec
is
io
n
tr
ee
s
,
n
aiv
e
B
ay
es
(
NB
)
,
an
d
ar
tific
ial
n
eu
r
al
n
et
wo
r
k
s
.
T
h
e
p
r
o
p
o
s
ed
ch
ar
ac
ter
is
tics
ar
e
ex
tr
ac
ted
,
an
d
th
e
e
n
s
em
b
le
tech
n
i
q
u
e
i
s
ev
alu
ated
u
s
in
g
t
h
e
UNSW
-
NB
1
5
an
d
n
etwo
r
k
in
f
o
r
m
atio
n
m
an
ag
e
m
en
t
an
d
s
ec
u
r
ity
(
NI
MS)
b
o
tn
ets
d
at
asets
,
in
clu
d
in
g
s
im
u
lated
I
o
T
s
en
s
o
r
d
ata.
T
h
e
ex
p
er
im
en
tal
f
in
d
in
g
s
s
h
o
w
th
at
th
e
s
u
g
g
ested
attr
ib
u
tes
h
av
e
th
e
p
o
te
n
tial
to
ex
h
ib
it
b
o
th
n
o
r
m
al
an
d
m
alicio
u
s
b
eh
av
io
r
tr
aits
u
s
in
g
th
e
c
o
r
r
elatio
n
co
ef
f
icien
t
a
n
d
co
r
r
en
tr
o
p
y
m
etr
ics.
T
h
e
s
u
g
g
ested
en
s
em
b
le
tech
n
iq
u
e
also
h
as a
h
ig
h
er
d
e
tectio
n
r
ate.
T
an
g
et
a
l.
[
2
0
]
o
f
f
er
ed
an
u
p
g
r
ad
ed
Ad
ab
o
o
s
t a
lg
o
r
ith
m
(
MF
-
Ad
ab
o
o
s
t)
an
d
s
ev
er
al
n
et
wo
r
k
tr
af
f
ic
f
ea
tu
r
es a
s
th
e
f
o
u
n
d
atio
n
f
o
r
t
h
eir
L
Do
S a
ttack
d
etec
tio
n
ap
p
r
o
ac
h
.
Af
ter
a
n
aly
zin
g
th
e
tr
a
f
f
ic,
th
ey
cr
ea
ted
a
n
etwo
r
k
f
ea
tu
r
e
s
et
th
at
is
u
tili
ze
d
f
o
r
f
ea
tu
r
e
s
elec
tio
n
an
d
f
ea
tu
r
e
ca
lcu
latio
n
o
f
n
etw
o
r
k
tr
a
f
f
ic
d
ata
.
B
y
ca
lcu
latin
g
f
ea
tu
r
es,
o
n
e
ca
n
m
in
im
ize
th
e
am
o
u
n
t
o
f
n
e
two
r
k
d
ata
wh
ile
ex
tr
ac
tin
g
th
e
m
o
s
t
v
alu
ab
le
in
f
o
r
m
atio
n
f
r
o
m
n
etwo
r
k
tr
af
f
ic
d
ata.
Featu
r
e
s
elec
tio
n
is
u
t
ilized
to
ch
o
o
s
e
th
e
b
est cla
s
s
i
f
icatio
n
f
ea
tu
r
es to
g
u
ar
an
tee
th
at
th
e
d
etec
tio
n
a
lg
o
r
ith
m
ca
n
b
e
tr
ain
ed
s
u
cc
e
s
s
f
u
lly
.
T
ests
ar
e
r
u
n
o
n
a
test
-
b
ed
p
latf
o
r
m
an
d
th
e
NS2
s
im
u
latio
n
p
latf
o
r
m
t
o
ass
ess
th
e
ef
f
ec
tiv
en
ess
o
f
t
h
is
ap
p
r
o
ac
h
.
T
h
e
ex
p
e
r
im
en
t'
s
f
in
d
in
g
s
s
h
o
w
th
at
th
e
ap
p
r
o
ac
h
ca
n
ef
f
icien
tly
d
etec
t L
Do
S a
ttack
s
.
B
y
co
m
b
in
in
g
tem
p
o
r
al,
b
y
te,
an
d
s
tatis
tical
d
ata
a
s
p
ec
ts
,
L
i
n
et
a
l.
[
2
1
]
c
r
ea
ted
a
m
u
lti
-
le
v
el
f
ea
tu
r
e
f
u
s
io
n
m
o
d
el
(
MFFu
s
io
n
)
th
at
ex
tr
ac
ts
r
eliab
le
in
f
o
r
m
atio
n
f
r
o
m
m
a
n
y
v
iewp
o
in
ts
an
d
c
r
ea
tes
a
m
o
r
e
r
o
b
u
s
t
an
d
ef
f
icien
t
m
o
d
el.
T
r
ials
d
em
o
n
s
tr
ate
th
at
m
o
d
els
ca
n
cu
t
tr
ain
in
g
tim
e,
s
tab
ilize
th
e
tr
ain
in
g
p
r
o
ce
s
s
,
an
d
en
h
an
ce
m
o
d
el
p
er
f
o
r
m
an
ce
.
Atten
tio
n
lo
s
s
ad
ap
tiv
ely
m
o
d
if
ies
s
am
p
le
we
ig
h
ts
to
in
cr
ea
s
e
th
e
d
etec
tio
n
r
ate
o
f
weir
d
s
am
p
les.
W
ith
s
ev
er
al
r
ea
l
n
etwo
r
k
d
atasets
,
MFFu
s
io
n
h
as
d
em
o
n
s
tr
ated
e
x
ce
llen
t
p
er
f
o
r
m
an
ce
in
ter
m
s
o
f
d
etec
tio
n
r
ate
an
d
f
al
s
e
alar
m
r
ate.
Usi
n
g
th
e
n
ewe
s
t
I
o
T
m
alicio
u
s
tr
af
f
ic
d
atase
t,
I
o
T
2
3
,
th
e
y
also
u
s
e
MFFu
s
io
n
f
o
r
I
o
T
n
etwo
r
k
an
o
m
aly
d
etec
tio
n
.
Mu
ltif
u
n
ctio
n
ality
a
n
d
s
u
itab
ilit
y
f
o
r
n
etwo
r
k
a
n
o
m
al
y
d
etec
tio
n
in
th
e
I
o
T
a
r
e
d
em
o
n
s
tr
ated
b
y
e
x
p
er
im
en
ts
co
n
d
u
cted
u
s
in
g
MFFu
s
io
n
.
Usi
n
g
f
ea
tu
r
e
f
u
s
io
n
an
d
m
ac
h
in
e
lear
n
in
g
,
L
i
et
a
l.
[
2
2
]
s
u
g
g
ested
a
m
eth
o
d
f
o
r
d
etec
tin
g
m
alicio
u
s
m
in
in
g
co
d
es.
T
h
ey
b
e
g
in
b
y
ex
tr
ac
tin
g
m
u
lti
-
d
im
en
s
io
n
al
in
f
o
r
m
atio
n
th
r
o
u
g
h
s
tatic
an
d
s
tati
s
tical
an
aly
tic
tech
n
iq
u
es.
Nex
t,
u
s
in
g
th
e
n
-
g
r
am
m
o
d
el
an
d
T
F
-
I
DF
to
ex
tr
ac
t
f
ea
tu
r
e
v
ec
to
r
s
f
o
r
m
u
lti
-
d
im
en
s
io
n
al
tex
t
f
ea
tu
r
es,
th
e
class
if
ier
s
elec
ts
th
e
b
est
f
ea
tu
r
e
v
ec
to
r
s
f
u
s
e
d
with
ad
d
itio
n
al
s
tatis
tical
f
ea
tu
r
es
to
tr
ain
o
u
r
d
etec
tio
n
m
o
d
el.
At
last
,
th
e
m
ac
h
in
e
lear
n
in
g
f
r
am
ew
o
r
k
is
u
tili
ze
d
to
c
o
n
d
u
ct
a
u
to
m
atic
d
etec
tio
n
.
Acc
o
r
d
in
g
to
th
e
ex
p
er
im
en
tal
f
in
d
in
g
s
,
o
u
r
tech
n
iq
u
e
ca
n
ac
h
iev
e
9
8
.
0
%
id
en
tific
atio
n
ac
cu
r
ac
y
,
an
F1
-
s
co
r
e
o
f
0
.
9
6
9
,
an
d
an
AU
C
o
f
0
.
9
7
3
f
o
r
th
e
R
OC
.
3.
M
E
T
H
O
D
T
h
is
r
esear
ch
p
r
o
p
o
s
ed
a
m
u
lti
-
f
ea
tu
r
es
s
elec
tio
n
m
eth
o
d
MFST
th
at
u
s
e
s
f
o
u
r
tech
n
iq
u
es:
th
e
f
ast
co
r
r
elatio
n
-
b
ased
f
ilter
(
FC
B
F
)
,
s
ig
n
if
ican
ce
f
ea
tu
r
e
ev
alu
at
o
r
(
SF
E
)
,
ch
i
-
s
q
u
a
r
e
(
C
h
i²)
,
a
n
d
g
ain
r
atio
(
GR
)
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
A
s
ig
n
ifica
n
t fe
a
tu
r
es v
ec
to
r
fo
r
in
tern
et
tr
a
ffic c
la
s
s
i
fica
tio
n
b
a
s
ed
o
n
…
(
A
l
h
a
mza
Mu
n
t
h
er
)
6961
E
ac
h
tech
n
iq
u
e
p
r
o
d
u
ce
s
a
f
e
atu
r
es
v
ec
to
r
ex
p
o
s
ed
to
th
e
r
an
k
er
an
d
B
o
r
d
a
v
o
tin
g
f
ilter
s
.
Fro
m
th
e
d
ataset,
th
e
ch
ar
ac
ter
is
tics
with
th
e
m
o
s
t v
o
tes will b
e
ch
o
s
en
.
Fo
u
r
f
ea
tu
r
e
s
elec
tio
n
s
tr
ateg
ies ar
e
u
s
ed
in
s
tead
o
f
o
n
e
f
o
r
th
e
f
o
llo
win
g
r
ea
s
o
n
s
:
Use
f
o
u
r
tech
n
iq
u
es
to
b
e
n
ef
it
f
r
o
m
ea
c
h
tech
n
iq
u
e’
s
ad
v
a
n
tag
es
an
d
tr
y
to
o
v
er
co
m
e
th
e
d
is
ad
v
a
n
tag
es
o
f
o
n
e
tech
n
iq
u
e
b
y
ad
o
p
tin
g
o
th
er
tech
n
iq
u
es.
A
d
o
p
tin
g
f
o
u
r
tech
n
i
q
u
es,
ea
c
h
in
tr
o
d
u
ce
s
a
d
if
f
er
en
t
s
elec
tio
n
s
ch
em
a,
p
r
o
d
u
cin
g
a
r
o
b
u
s
tn
ess
f
ea
tu
r
es
v
ec
to
r
.
Usi
n
g
o
n
e
f
ea
tu
r
e
s
elec
tio
n
m
ay
af
f
ec
t
th
e
class
if
ier
's
p
er
f
o
r
m
an
ce
in
ter
m
s
o
f
ac
cu
r
ac
y
.
I
t
allo
ws
s
elec
tin
g
th
e
m
o
s
t
s
ig
n
if
ican
t
f
ea
tu
r
es
s
u
b
s
et
f
r
o
m
th
e
im
p
o
r
tan
t
f
e
atu
r
es
s
et.
A
s
illu
s
tr
ated
in
F
ig
u
r
e
1
,
th
e
MFST
f
r
am
ewo
r
k
co
n
s
is
ts
o
f
th
r
ee
m
ajo
r
p
h
ases
.
T
h
e
f
ir
s
t
s
tep
,
k
n
o
wn
as
th
e
d
ataset
p
h
ase,
c
o
m
p
r
is
es
o
f
th
e
r
aw
d
ataset'
s
en
tr
y
.
Seco
n
d
,
it
is
th
e
f
ea
tu
r
e
s
elec
tio
n
s
tep
,
wh
i
ch
is
d
iv
id
ed
in
t
o
two
p
a
r
ts
.
Fig
u
r
e
1
.
T
h
e
p
r
o
p
o
s
ed
MFST
f
r
am
ewo
r
k
T
h
ese
v
ec
to
r
s
ar
e
t
r
an
s
m
itted
to
th
e
r
a
n
k
er
f
ilter
with
in
th
e
f
iltra
tio
n
en
g
in
e,
wh
ic
h
is
d
i
v
id
ed
in
to
two
p
o
r
tio
n
s
(
r
a
n
k
in
g
an
d
v
o
t
in
g
)
.
T
h
e
f
ea
tu
r
es
ar
e
r
a
n
k
ed
ac
co
r
d
in
g
to
th
e
d
eg
r
ee
o
f
im
p
o
r
tan
ce
m
ea
s
u
r
ed
b
y
th
e
r
an
k
in
g
f
ilter
alg
o
r
ith
m
.
Fo
llo
win
g
th
at,
th
e
v
ec
to
r
s
o
f
r
an
k
ed
f
ea
tu
r
es
g
o
th
r
o
u
g
h
th
e
s
ec
o
n
d
f
iltra
tio
n
s
tep
,
wh
ich
co
u
n
ts
th
e
v
o
tes
f
o
r
ea
ch
f
ea
tu
r
e
.
I
n
o
t
h
er
wo
r
d
s
,
ch
o
o
s
e
th
e
f
ir
s
t
M
h
ig
h
ly
r
a
n
k
ed
f
ea
tu
r
es
f
r
o
m
ea
ch
p
r
ec
ed
in
g
f
ea
tu
r
e
s
elec
tio
n
a
n
d
th
en
co
u
n
t
th
e
v
o
tes
f
o
r
ea
c
h
f
ea
t
u
r
e.
T
h
e
ch
ar
ac
ter
is
tics
with
th
e
m
o
s
t
v
o
tes
an
d
th
e
h
ig
h
est
r
an
k
v
al
u
e
(
v
al
u
e)
will
b
e
c
h
o
s
en
as
th
e
m
o
s
t
d
is
ce
r
n
m
en
t
f
ea
tu
r
es.
As
a
r
esu
lt,
we
ca
n
ass
u
r
e
th
at
th
is
co
llectio
n
o
f
attr
ib
u
tes
is
s
ig
n
if
ican
t
wh
e
n
we
s
elec
t
f
o
u
r
s
elec
tio
n
ap
p
r
o
ac
h
es.
T
h
e
n
e
x
t
s
ec
tio
n
will d
is
cu
s
s
ea
ch
u
n
it o
f
th
e
p
r
o
p
o
s
ed
MFST
f
r
am
e
wo
r
k
.
3
.
1
.
F
ilte
r
f
ea
t
ures t
ec
hn
iqu
es
Featu
r
e
s
elec
tio
n
with
f
ilter
in
g
tech
n
iq
u
es
is
a
m
eth
o
d
o
f
r
an
k
in
g
r
elev
a
n
t
ch
ar
ac
ter
is
tics
ac
co
r
d
in
g
to
s
p
ec
if
ic
s
tatis
tical
m
etr
ic
s
i
n
o
r
d
e
r
to
ch
o
o
s
e
a
s
u
b
s
et
o
f
t
h
em
.
T
h
ese
tech
n
i
q
u
es
ju
s
t
co
n
s
id
er
th
e
in
tr
in
s
ic
q
u
alities
o
f
th
e
d
ata
a
n
d
o
p
e
r
ate
in
d
ep
e
n
d
en
tly
o
f
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
.
T
h
is
p
a
p
er
will
ad
o
p
t
f
o
u
r
d
if
f
er
en
t
f
ilter
tech
n
iq
u
es to
e
x
tr
ac
t th
e
s
ig
n
if
ican
t f
ea
t
u
r
es
v
ec
to
r
in
I
n
ter
n
et
tr
a
f
f
ic.
3
.
1
.
1
.
Sig
nifica
nce
f
ea
t
ure
ev
a
lua
t
o
r
Sig
n
if
ican
ce
f
ea
tu
r
e
ev
alu
at
o
r
(
SFE
)
is
o
n
e
o
f
th
e
im
p
o
r
t
an
t
f
ilter
tech
n
iq
u
es
u
s
ed
to
s
elec
t
th
e
s
ig
n
if
ican
t
f
ea
tu
r
es
f
o
r
ce
r
tain
d
atasets
.
SF
E
is
b
u
ilt
b
ased
o
n
th
e
co
n
d
itio
n
al
p
r
o
b
ab
ilit
y
t
ec
h
n
iq
u
e
w
h
er
e
th
e
f
ea
tu
r
es
ar
e
s
elec
ted
b
ased
o
n
h
y
p
o
th
esis
s
tate
f
ea
tu
r
es
with
d
if
f
e
r
en
t
v
alu
es
an
d
class
es,
wh
ile
th
is
m
ay
n
o
t
b
e
f
o
r
in
s
ig
n
if
ican
t
f
ea
tu
r
es.
T
wo
m
ajo
r
r
ea
s
o
n
s
to
s
elec
t
SF
E
ar
e
th
at
i
t
in
cr
ea
s
es
co
m
p
u
tatio
n
s
p
ee
d
.
I
t
en
h
an
ce
s
th
e
q
u
ality
o
f
k
n
o
wled
g
e
in
ter
m
s
o
f
class
if
icatio
n
b
y
u
s
in
g
th
e
lik
elih
o
o
d
-
b
ased
m
eth
o
d
f
o
r
class
if
icato
r
y
k
n
o
wled
g
e.
T
h
e
r
ef
o
r
e,
SF
E
is
in
teg
r
ated
with
th
e
class
if
icatio
n
m
eth
o
d
b
ased
o
n
lik
elih
o
o
d
,
an
d
th
e
s
elec
ted
s
ig
n
if
ican
t
f
e
atu
r
es a
r
e
u
s
ed
to
m
ak
e
th
e
cla
s
s
if
icatio
n
d
ec
is
io
n
.
3
.
1
.
2
.
F
a
s
t
co
rr
el
a
t
io
n
-
ba
s
ed
f
ilte
r
T
h
i
s
a
p
p
r
o
a
c
h
u
s
es
c
o
r
r
e
l
at
i
o
n
m
e
a
s
u
r
e
m
e
n
ts
t
o
f
i
n
d
r
e
d
u
n
d
a
n
t
a
n
d
r
e
l
e
v
a
n
t
f
e
a
t
u
r
es
t
o
t
h
e
c
l
a
s
s
.
T
h
e
g
e
n
e
r
a
l
g
u
i
d
e
l
i
n
e
f
o
r
c
o
r
r
e
l
a
tio
n
m
e
a
s
u
r
e
m
e
n
t
is
t
h
a
t
a
f
e
a
tu
r
e
i
s
g
o
o
d
i
f
i
t
is
r
e
le
v
a
n
t
to
t
h
e
c
l
a
s
s
b
u
t
n
o
t
r
e
d
u
n
d
a
n
t
w
i
t
h
o
t
h
e
r
t
i
m
e
-
r
e
la
t
e
d
f
e
a
t
u
r
e
s
.
A
f
e
a
t
u
r
e
is
c
o
n
s
i
d
e
r
e
d
g
o
o
d
i
f
i
t
i
s
s
u
b
s
ta
n
ti
a
l
l
y
l
i
n
k
e
d
wi
t
h
t
h
e
c
l
a
s
s
r
at
h
e
r
t
h
a
n
w
i
t
h
o
t
h
e
r
f
e
a
t
u
r
e
s
.
T
h
e
r
e
a
r
e
o
f
t
e
n
tw
o
m
e
t
h
o
d
s
.
T
o
o
v
e
r
c
o
m
e
t
h
e
l
im
i
t
a
t
i
o
n
s
o
f
l
i
n
e
a
r
c
o
r
r
e
l
a
t
i
o
n
,
w
h
i
c
h
c
a
n
b
e
s
u
m
m
e
d
u
p
a
s
n
o
t
al
w
a
y
s
as
s
u
m
i
n
g
a
li
n
e
a
r
c
o
r
r
e
l
at
i
o
n
b
et
w
e
e
n
f
e
a
t
u
r
e
s
i
n
t
h
e
r
e
a
l
w
o
r
l
d
,
b
e
i
n
g
u
n
a
b
l
e
t
o
c
a
p
t
u
r
e
c
o
r
r
e
l
a
t
i
o
n
s
t
h
at
a
r
e
n
o
t
l
i
n
ea
r
,
a
n
d
t
h
e
c
a
l
c
u
l
a
t
i
o
n
r
e
q
u
i
r
in
g
a
l
l
f
e
a
t
u
r
e
s
t
o
b
e
T
r
a
i
n
i
n
g
D
a
t
a
s
e
t
T
e
s
t
i
n
g
D
a
t
a
s
e
t
S
F
E
C
h
i
2
F
C
B
F
G
R
}
,..
..,
,
{
2
1
1
f
f
f
V
m
=
}
,..
..,
,
{
2
1
2
f
f
f
V
m
=
}
,..
..,
,
{
2
1
3
f
f
f
V
m
=
}
,..
..,
,
{
2
1
4
f
f
f
V
m
=
S
o
r
t
i
n
g
b
a
s
e
d
o
n
h
i
g
h
v
a
l
u
e
C
o
u
n
t
i
n
g
V
o
t
e
s
R
a
n
k
e
r
V
o
t
i
n
g
F
i
l
t
r
a
t
i
o
n
E
n
g
i
n
e
M
F
S
T
E
n
g
i
n
e
s
D
a
t
a
s
e
t
F
e
a
t
u
r
e
S
e
l
e
c
t
i
o
n
P
r
o
c
e
s
s
S
V
M
N
i
a
v
e
B
a
y
s
e
C
4
.
5
C
l
a
s
s
i
f
i
e
r
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
14
,
No
.
6
,
Dec
em
b
e
r
20
24
:
6
9
5
8
-
6
9
6
8
6962
c
o
n
g
r
u
e
n
t
,
F
C
B
F
a
d
o
p
t
e
d
a
s
ec
o
n
d
a
p
p
r
o
a
c
h
b
a
s
e
d
o
n
i
n
f
o
r
m
a
t
i
o
n
t
h
e
o
r
y
.
T
h
e
c
o
r
r
e
l
a
ti
o
n
m
e
a
s
u
r
e
is
b
a
s
e
d
o
n
t
h
e
e
n
t
r
o
p
y
n
o
t
i
o
n
f
r
o
m
i
n
f
o
r
m
a
t
i
o
n
t
h
e
o
r
y
t
o
a
s
s
es
s
t
h
e
r
a
n
d
o
m
o
b
j
e
c
t'
s
l
e
v
e
l
o
f
u
n
c
e
r
t
ai
n
t
y
[
7
]
,
[
2
3
]
,
[
2
4
]
.
3
.
1
.
3
.
Chi
-
s
qu
a
re
d
f
ea
t
ure
s
e
lect
io
n
C
h
i
-
s
q
u
ar
ed
o
r
C
h
i²
is
b
ased
o
n
th
e
x
2
s
tatis
tic
an
d
c
o
n
s
is
ts
o
f
two
m
ain
p
h
ases
.
Ph
ase
1
,
it
b
e
g
in
s
with
a
h
ig
h
s
ig
n
if
ica
n
ce
lev
el
ca
lled
s
ig
L
ev
el
f
o
r
all
n
u
m
e
r
ic
attr
ib
u
tes.
E
ac
h
attr
ib
u
te
is
s
o
r
ted
ac
co
r
d
i
n
g
to
its
v
alu
es.
T
h
en
th
e
f
o
llo
win
g
is
p
er
f
o
r
m
e
d
:
−
Fo
r
ea
ch
p
air
o
f
n
eig
h
b
o
r
in
g
i
n
ter
v
als,
ca
lcu
late
th
e
x
2
u
s
in
g
(
3
)
.
−
T
h
e
ad
jace
n
t in
ter
v
als with
th
e
s
m
allest x
2
v
alu
e
s
h
o
u
ld
b
e
co
m
b
in
ed
.
−
Me
r
g
in
g
k
e
ep
s
o
n
u
n
til all
in
t
er
v
al
p
air
s
h
av
e
x
2
v
alu
es g
r
e
ater
th
an
th
e
p
ar
am
eter
s
et
b
y
s
ig
L
ev
el.
T
h
e
p
r
o
ce
d
u
r
e
is
r
ep
ea
ted
wit
h
a
lo
wer
s
ig
L
ev
el
u
p
u
n
til
an
in
co
n
s
is
ten
cy
r
ate
in
th
e
d
is
cr
etize
d
d
ata
is
ex
ce
ed
ed
,
at
w
h
ich
p
o
in
t
C
h
i²
au
to
m
atica
lly
ch
o
o
s
es
an
ap
p
r
o
p
r
iate
x
2
th
r
esh
o
ld
th
at
p
r
eser
v
es
th
e
ac
cu
r
ac
y
o
f
th
e
o
r
ig
in
al
d
ata.
Ph
ase
2
is
a
m
o
r
e
r
ef
in
ed
p
r
o
c
ed
u
r
e
o
f
Ph
ase
1
,
b
eg
in
n
in
g
w
ith
th
e
s
ig
L
ev
el
estab
lis
h
ed
in
Ph
ase
1
.
E
ac
h
a
ttrib
u
te
is
co
n
n
ec
ted
to
a
ce
r
t
ain
s
ig
L
ev
el
[
]
,
an
d
ea
ch
attr
ib
u
te
is
m
er
g
ed
o
n
e
at
a
tim
e
[
2
5
]
.
−
Ver
if
y
in
g
c
o
n
s
is
ten
cy
f
o
llo
wi
n
g
ea
ch
attr
ib
u
te
m
er
g
e.
−
T
h
e
s
ig
L
ev
el
f
o
r
attr
i
b
u
te
is
l
o
wer
ed
f
o
r
th
e
s
u
b
s
eq
u
e
n
t
r
o
u
n
d
o
f
m
er
g
in
g
if
th
e
in
c
o
n
s
is
ten
cy
r
ate
is
n
o
t
ex
ce
ed
ed
.
T
h
is
p
r
o
ce
s
s
co
n
tin
u
ed
u
n
til
n
o
attr
ib
u
te'
s
v
alu
es c
o
u
ld
b
e
m
er
g
e
d
.
−
I
f
an
attr
ib
u
te
h
as
o
n
ly
o
n
e
v
alu
e
at
th
e
en
d
o
f
Ph
ase
2
,
it
m
er
ely
s
ig
n
if
ies
th
at
th
e
attr
ib
u
te
d
o
es
n
o
t
ac
cu
r
ately
r
ep
r
esen
t th
e
o
r
ig
in
al
d
ata
s
et.
−
At
th
e
co
n
clu
s
io
n
o
f
d
is
cr
etiza
tio
n
,
f
ea
tu
r
e
s
elec
tio
n
is
co
m
p
leted
.
3
.
1
.
4
.
G
a
in
ra
t
i
o
f
ea
t
ure
s
elec
t
io
n
T
h
e
g
a
i
n
r
a
t
i
o
i
m
p
r
o
v
e
s
i
n
f
o
r
m
a
t
i
o
n
g
a
i
n
b
y
p
r
o
v
i
d
i
n
g
a
n
o
r
m
a
l
i
z
e
d
s
c
o
r
e
o
f
a
f
e
a
t
u
r
e'
s
c
o
n
t
r
i
b
u
t
i
o
n
t
o
t
h
e
b
e
s
t
c
l
as
s
i
f
i
c
at
i
o
n
d
e
c
i
s
i
o
n
b
a
s
e
d
o
n
i
n
f
o
r
m
a
t
i
o
n
g
a
i
n
.
T
h
e
g
a
i
n
r
a
t
i
o
i
s
u
s
e
d
i
n
a
n
i
t
e
r
a
t
i
v
e
p
r
o
c
e
s
s
w
h
e
r
e
w
e
c
h
o
o
s
e
i
n
c
r
e
as
i
n
g
l
y
s
m
a
lle
r
g
r
o
u
p
s
o
f
f
e
a
t
u
r
es
[
2
6
]
.
T
h
e
s
e
i
t
e
r
a
ti
o
n
s
e
n
d
w
h
e
n
j
u
s
t
a
c
e
r
t
a
i
n
n
u
m
b
e
r
o
f
f
e
a
t
u
r
e
s
a
r
e
l
e
f
t
.
T
h
e
g
a
i
n
r
at
io
i
s
o
n
e
o
f
t
h
e
d
i
s
p
a
r
i
t
y
m
e
asu
r
e
m
e
n
t
s
e
m
p
l
o
y
e
d
,
a
n
d
a
f
e
a
tu
r
e
'
s
h
i
g
h
g
a
i
n
r
at
i
o
s
u
g
g
e
s
ts
t
h
at
i
t
w
i
ll
b
e
h
e
l
p
f
u
l
f
o
r
c
a
t
e
g
o
r
i
z
a
t
i
o
n
.
G
a
i
n
r
a
ti
o
ap
p
l
i
e
d
n
o
r
m
a
l
i
z
a
ti
o
n
t
o
t
h
e
i
n
f
o
r
m
a
t
i
o
n
g
a
i
n
s
c
o
r
e
u
s
i
n
g
a
s
p
li
t
i
n
f
o
r
m
a
ti
o
n
v
a
l
u
e
,
w
h
i
c
h
w
as
i
n
i
ti
a
ll
y
u
t
i
l
iz
e
d
i
n
t
h
e
d
e
c
i
s
i
o
n
t
r
e
e
(
C
4
.
5
)
[
2
7
]
.
3
.
2
.
F
ea
t
ures f
ilte
r
C
o
n
ce
r
n
in
g
th
e
p
r
o
p
o
s
ed
m
o
d
el
s
h
o
wn
in
Fig
u
r
e
1
,
ea
c
h
f
ea
tu
r
e
s
elec
tio
n
tech
n
iq
u
e
w
ill
ch
o
o
s
e
=
[
1
,
2
,
3
,
4
,
.
.
.
]
.
T
h
is
is
wh
at
t
h
e
MFST
en
g
in
e
will
ex
p
er
ien
ce
.
E
ac
h
f
ea
tu
r
e
v
ec
to
r
in
t
h
e
f
ilter
in
g
en
g
in
e
will
b
e
s
o
r
ted
u
s
in
g
a
r
an
k
e
r
f
ilter
d
ep
en
d
in
g
o
n
its
r
an
k
in
g
v
alu
e
to
f
in
d
th
e
m
o
s
t
im
p
o
r
tan
t
ch
ar
ac
ter
is
tics
.
T
h
e
m
o
s
t
im
p
o
r
tan
t
asp
ec
t
is
i
ts
h
ig
h
v
alu
e.
Af
ter
th
at,
a
v
o
te
will
b
e
co
n
d
u
cte
d
f
o
r
ea
ch
f
ea
tu
r
e
to
d
eter
m
in
e
h
o
w
m
a
n
y
o
f
th
e
f
o
u
r
tec
h
n
iq
u
es
a
r
e
s
elec
ted
f
o
r
th
at
f
ea
t
u
r
e.
T
h
e
f
ea
tu
r
e
th
at
r
ec
eiv
es
th
e
m
o
s
t
v
o
tes
an
d
r
an
k
v
alu
e
will
b
e
ch
o
s
en
.
T
h
e
two
f
ilter
s
will
b
e
ex
p
lain
ed
in
th
e
n
ex
t su
b
s
ec
tio
n
s
.
3
.
2
.
1
.
F
a
s
t
f
e
a
t
ure
ra
nk
ing
MFST
f
r
am
ewo
r
k
is
ad
o
p
tin
g
two
f
ilter
s
to
s
elec
t
th
e
m
o
s
t
s
ig
n
if
ican
t
f
ea
tu
r
es
s
u
b
s
et
am
o
n
g
th
e
im
p
o
r
tan
t
f
ea
tu
r
es
s
et
b
y
s
elec
tin
g
th
e
h
ig
h
ly
r
an
k
ed
f
ea
tu
r
es
(
b
ased
o
n
th
e
r
an
k
er
f
ilter
)
as
well
a
s
s
elec
tin
g
th
e
f
ea
tu
r
e
th
at
h
as
th
e
h
ig
h
est
n
u
m
b
er
o
f
v
o
tes
f
r
o
m
th
e
f
o
u
r
f
ea
t
u
r
e
s
elec
tio
n
tech
n
iq
u
e
s
(
b
ased
o
n
v
o
tin
g
f
ilter
)
.
Featu
r
e
r
an
k
in
g
d
eter
m
in
es
th
e
m
o
s
t
s
ig
n
if
ican
t
f
ea
tu
r
e
f
r
o
m
th
e
ex
tr
ac
te
d
f
ea
tu
r
es.
On
e
o
f
th
e
m
o
s
t
ef
f
ec
tiv
e
alg
o
r
ith
m
s
u
s
ed
in
f
ea
tu
r
e
r
a
n
k
in
g
is
b
ased
o
n
in
f
o
r
m
atio
n
th
eo
r
y
,
s
u
c
h
as
m
u
tu
al
in
f
o
r
m
atio
n
,
wh
ich
is
co
n
s
id
er
ed
o
n
e
o
f
th
e
f
astes
t r
an
k
in
g
tech
n
iq
u
es,
w
h
ich
is
ca
lcu
lated
b
ased
o
n
alg
o
r
ith
m
s
b
u
ilt in
th
e
p
r
o
p
o
s
ed
m
eth
o
d
,
wh
ich
was
ad
ap
ted
v
ia
d
if
f
e
r
en
t
d
ata
m
in
in
g
p
r
o
g
r
am
s
MA
T
L
AB
,
W
ek
a,
an
d
R
ap
id
.
R
an
k
in
g
p
r
o
ce
s
s
co
n
s
is
ts
o
f
f
o
u
r
s
tep
s
:
-
I
n
itialize
s
et
to
th
e
wh
o
le
s
et
o
f
f
ea
tu
r
es
.
is
an
em
p
ty
s
et.
-
Fo
r
all
f
ea
tu
r
es
∈
co
m
p
u
te
m
u
tu
al
in
f
o
r
m
atio
n
(
MI
)
.
-
Fin
d
f
ea
tu
r
e
th
at
m
ax
im
izes (
MI
)
an
d
m
o
v
e
it to
.
-
R
ep
ea
t u
n
it th
e
C
ar
d
in
al
o
f
is
.
3
.
2
.
2
.
F
ea
t
ures v
o
t
ing
ba
s
ed
o
n
B
o
rda
co
un
t
Af
ter
r
an
k
in
g
t
h
e
f
ea
tu
r
es
as
t
h
e
f
ir
s
t
s
tep
in
th
e
f
iltra
tio
n
z
o
n
e,
th
e
f
ea
tu
r
es
u
n
d
er
g
o
th
e
s
e
co
n
d
s
tep
o
f
f
iltra
tio
n
,
wh
ich
is
v
o
ti
n
g
,
wh
ich
wo
r
k
s
tig
h
tly
with
t
h
e
r
an
k
in
g
u
n
it
—
T
h
e
p
r
o
p
o
s
ed
MFST
s
y
s
tem
ad
o
p
ted
B
o
r
d
a
-
c
o
u
n
t
as
a
ty
p
e
o
f
v
o
tin
g
.
Gen
er
ally
,
th
er
e
ar
e
th
r
ee
m
ajo
r
v
o
tin
g
m
eth
o
d
s
:
Ma
jo
r
ity
,
Plu
r
al,
an
d
B
o
r
d
a
.
T
h
e
m
ain
r
ea
s
o
n
th
at
m
o
tiv
ated
u
s
to
ch
o
o
s
e
B
o
r
d
a
-
co
u
n
t
r
ath
er
th
an
o
t
h
er
s
is
th
at
B
o
r
d
a
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
A
s
ig
n
ifica
n
t fe
a
tu
r
es v
ec
to
r
fo
r
in
tern
et
tr
a
ffic c
la
s
s
i
fica
tio
n
b
a
s
ed
o
n
…
(
A
l
h
a
mza
Mu
n
t
h
er
)
6963
co
n
s
id
er
s
r
an
k
v
alu
es
f
o
r
f
ea
t
u
r
es
b
esid
es
th
e
d
ec
is
io
n
o
f
v
o
ter
s
(
i.e
.
,
f
o
r
ea
ch
v
o
ter
(
T
ec
h
n
iq
u
es)
r
an
k
s
th
e
ca
n
d
id
ates (
f
ea
tu
r
es))
th
at
m
a
d
e
wo
r
k
f
it with
r
an
k
i
n
g
u
n
it
[
2
8
]
.
3
.
3
.
I
nte
rnet
t
ra
f
f
ic
da
t
a
s
et
s
B
en
ch
m
ar
k
d
atasets
wer
e
em
p
lo
y
ed
i
n
th
is
in
v
esti
g
atio
n
wh
i
ch
ca
n
b
e
f
in
d
in
[
2
9
]
.
T
o
g
en
er
ate
d
ata
tr
af
f
ic,
a
wo
r
ld
wid
e
s
o
u
r
ce
d
ep
en
d
s
o
n
r
esear
ch
u
n
d
er
tak
en
at
Qu
ee
n
Ma
r
y
Un
iv
er
s
ity
o
f
L
o
n
d
o
n
'
s
Dep
ar
tm
en
t
o
f
C
o
m
p
u
ter
Scien
ce
an
d
m
en
tio
n
e
d
in
C
aid
a
d
atab
ases
.
T
h
is
in
f
o
r
m
atio
n
is
g
ath
er
ed
f
r
o
m
a
n
etwo
r
k
'
s
ed
g
e.
I
t
p
er
m
its
ac
c
ess
to
all
tr
an
s
m
is
s
io
n
co
n
tr
o
l
p
r
o
to
co
l
(
T
C
P
)
,
u
s
er
d
atag
r
a
m
p
r
o
to
co
l
(
UDP
)
,
an
d
in
te
r
n
et
p
r
o
to
c
o
l
(
I
P)
co
n
n
ec
tio
n
-
r
elate
d
p
ac
k
ets
tr
av
e
lin
g
in
b
o
t
h
d
i
r
ec
tio
n
s
(
f
r
o
m
s
en
d
er
to
r
ec
ip
ien
t
an
d
v
ice
v
er
s
a)
.
T
h
er
e
f
o
r
e,
it c
an
g
et
a
d
d
itio
n
al
f
ea
tu
r
es f
o
r
ea
ch
p
ac
k
et
as
a
r
esu
lt.
Dif
f
er
en
t
d
ataset
s
izes
ar
e
u
s
ed
to
p
r
ev
en
t
th
e
p
r
o
p
o
s
ed
n
o
tio
n
f
r
o
m
b
ein
g
o
v
er
f
itted
an
d
to
tr
ac
k
b
eh
a
v
io
r
u
s
in
g
v
ar
io
u
s
d
atasets
.
T
h
e
d
ataset
co
n
s
is
ts
o
f
2
4
8
f
ea
tu
r
e
s
b
ased
o
n
v
ar
io
u
s
tr
af
f
ic
b
e
h
av
io
r
s
ex
tr
ac
ted
f
r
o
m
p
ac
k
et
h
ea
d
er
s
o
f
th
e
th
r
ee
d
ata
tr
an
s
f
er
m
eth
o
d
s
lis
ted
ab
o
v
e
as
in
Fig
u
r
e
2
:
t
r
a
f
f
ic
d
u
r
a
tio
n
,
T
C
P,
UDP
p
o
r
t,
an
d
Pay
l
o
ad
s
ize
s
tatis
tics
.
E
ac
h
f
ea
tu
r
e'
s
s
p
ec
if
ics
ar
e
d
e
s
cr
ib
ed
in
[
3
0
]
.
As
i
n
d
icate
d
i
n
T
ab
le
1
,
th
e
class
es
p
r
esen
t
in
th
e
d
atasets
ar
e
d
iv
id
ed
in
to
ten
g
r
o
u
p
s
.
E
ac
h
d
ata
en
tr
y
will
allo
ca
te
8
0
%
f
o
r
tr
ain
in
g
an
d
b
u
ild
in
g
th
e
class
if
ier
m
o
d
el
an
d
2
0
%
f
o
r
test
in
g
.
T
en
s
u
b
s
ets
wer
e
all
g
ath
er
ed
u
s
in
g
d
if
f
e
r
en
t
p
r
o
to
co
l
c
o
n
n
ec
tio
n
s
;
th
e
v
ar
iety
o
f
in
s
tan
ce
s
allo
ws
f
o
r
test
in
g
th
e
s
u
g
g
est
ed
s
o
lu
tio
n
in
a
h
eter
o
g
en
eo
u
s
en
v
ir
o
n
m
en
t.
T
h
e
in
s
tan
ce
s
u
b
s
et
s
ize
f
o
r
ea
c
h
d
ataset
is
d
is
p
lay
ed
in
T
ab
le
2
.
Fig
u
r
e
2
.
Data
s
et
en
v
ir
o
n
m
en
t
T
ab
le
1
.
Data
s
ets ca
teg
o
r
izatio
n
#
C
a
t
e
g
o
r
y
A
p
p
l
i
c
a
t
i
o
n
s
d
e
scri
p
t
i
o
n
1
W
e
b
-
b
r
o
w
si
n
g
h
t
t
p
,
h
t
t
p
s
2
M
a
i
l
I
map
,
P
O
P
2
,
P
O
P
3
,
S
M
TP
3
B
u
l
k
F
TP
4
A
t
t
a
c
k
P
o
r
t
sc
a
n
,
w
o
r
ms,
v
i
r
u
ses
,
e
mai
l
a
t
t
a
c
k
s
5
P
2
P
N
a
p
st
e
r
,
K
A
ZA
A
,
e
M
u
l
e
,
G
n
u
t
e
l
l
a
,
e
D
o
n
k
e
y
6
D
a
t
a
b
a
s
e
M
y
S
Q
L
,
d
b
a
se
,
O
r
a
c
l
e
S
Q
LN
e
t
7
M
u
l
t
i
me
d
i
a
W
i
n
d
o
w
s me
d
i
a
p
l
a
y
e
r
,
r
e
a
l
me
d
i
a
8
S
e
r
v
i
c
e
X
1
1
,
D
N
S
,
i
D
e
n
t
,
I
d
a
p
,
N
TP
9
I
n
t
e
r
a
c
t
i
v
e
S
S
H
,
Te
l
n
e
t
,
K
l
o
g
i
n
,
r
l
o
g
i
n
10
G
a
mes
M
i
c
r
o
s
o
f
t
d
i
r
e
c
t
p
l
a
y
T
ab
le
2
.
Nu
m
b
er
o
f
in
s
tan
ce
s
in
ea
ch
d
ataset
D
a
t
a
s
e
t
N
a
me
#
I
n
st
a
n
c
e
s
S
u
b
s
e
t
0
1
2
4
8
6
3
S
u
b
s
e
t
0
2
2
3
8
0
1
S
u
b
s
e
t
0
3
2
2
9
3
2
S
u
b
s
e
t
0
4
2
2
2
8
5
S
u
b
s
e
t
0
5
2
1
6
4
8
S
u
b
s
e
t
0
6
1
9
3
8
4
S
u
b
s
e
t
0
7
5
5
8
3
5
S
u
b
s
e
t
0
8
5
5
4
9
4
S
u
b
s
e
t
0
9
6
6
2
4
8
S
u
b
s
e
t
1
0
6
5
0
3
6
3
.
4
.
I
m
plem
ent
a
t
io
n o
f
M
F
ST
T
h
is
s
ec
tio
n
ex
p
lo
r
es
th
e
im
p
lem
en
tatio
n
s
tep
s
f
o
r
MFST.
As
m
en
tio
n
ed
,
MFST
co
n
s
is
ts
o
f
f
o
u
r
f
ea
tu
r
e
s
elec
tio
n
tech
n
iq
u
es
a
n
d
two
f
ilter
s
.
Firstl
y
,
we
n
ee
d
to
r
u
n
all
th
e
a
b
o
v
e
ten
d
ata
s
ets
in
d
iv
id
u
ally
i
n
E
n
d
h
o
s
t
O
r
g
a
n
i
z
a
t
i
o
n
B
a
c
k
b
o
n
e
I
n
t
r
a
n
e
t
E
n
v
i
r
o
n
m
e
n
t
E
d
g
e
/
B
o
r
d
e
r
e
n
v
i
r
o
m
e
n
t
B
a
c
k
b
o
n
e
e
n
v
i
r
o
n
m
e
n
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
14
,
No
.
6
,
Dec
em
b
e
r
20
24
:
6
9
5
8
-
6
9
6
8
6964
ea
ch
o
f
th
e
f
o
u
r
f
ea
tu
r
e
s
elec
t
io
n
tech
n
iq
u
es
a
n
d
g
et
th
e
r
an
k
v
al
u
e
to
g
et
th
e
m
o
s
t
s
ig
n
if
i
ca
n
t
f
ea
tu
r
e
v
ec
to
r
.
MFST
co
n
s
id
er
s
o
n
ly
th
e
f
ir
s
t
2
+
1
b
ased
o
n
in
f
o
r
m
atio
n
g
ai
n
[
3
1
]
,
wh
er
e
is
th
e
to
tal
n
u
m
b
er
o
f
=
248
.
So
,
t
h
e
n
u
m
b
er
o
f
co
n
s
id
er
e
d
f
ea
tu
r
es
is
o
n
ly
th
e
f
ir
s
t
eig
h
t
h
ig
h
ly
r
a
n
k
ed
d
ataset
f
ea
t
u
r
es.
T
h
e
n
e
x
t
s
u
b
s
ec
tio
n
s
h
o
ws
th
e
g
en
er
ate
d
f
ea
tu
r
e
v
ec
to
r
s
f
o
r
th
e
ab
o
v
e
d
atasets
.
T
ab
les 3
a
n
d
4
lis
t
th
e
f
ea
t
u
r
e
v
ec
to
r
s
s
elec
ted
b
y
ea
ch
tech
n
iq
u
e
ac
co
r
d
in
g
t
o
th
e
p
r
o
p
o
s
ed
tech
n
iq
u
e
SF
E
,
FC
B
F,
C
h
i²
,
an
d
GR
.
All
th
e
s
elec
ted
f
ea
tu
r
es a
r
e
o
r
d
er
ed
b
ased
o
n
th
e
r
an
k
v
alu
e
f
o
r
ea
c
h
f
ea
tu
r
e.
T
ab
le
3
.
Selecte
d
f
ea
tu
r
es v
ec
t
o
r
b
y
SF
E
,
FC
B
F
D
a
t
a
s
e
t
S
e
l
e
c
t
e
d
f
e
a
t
u
r
e
s
v
e
c
t
o
r
b
y
e
a
c
h
Te
c
h
n
i
q
u
e
SFE
F
C
B
F
S
u
b
s
e
t
0
1
1,
9
0
,
1
7
0
,
9
6
,
9
5
,
9
4
,
1
8
7
,
180
1,
6
0
,
9
5
,
9
6
,
8
6
,
8
4
,
8
2
,
1
8
6
S
u
b
s
e
t
0
2
1,
4
5
,
1
7
0
,
1
8
7
,
9
0
,
1
7
7
,
1
1
3
,
59
1,
9
5
,
1
3
7
,
6
0
,
4
5
,
1
1
3
,
1
2
5
,
59
S
u
b
s
e
t
0
3
1,
9
0
,
9
6
,
9
5
,
8
6
,
4
5
,
9
4
,
1
7
7
1,
9
5
,
6
0
,
9
6
,
4
5
,
1
1
3
,
8
4
,
59
S
u
b
s
e
t
0
4
7
1
,
1,
9
6
,
9
0
,
1
8
0
,
1
8
7
,
9
4
,
184
1,
6
0
,
1
3
7
,
9
5
,
4
5
,
9
6
,
1
8
7
,
180
S
u
b
s
e
t
0
5
1,
9
0
,
6
5
,
7
1
,
9
4
,
8
8
,
1
8
7
,
1
8
0
1,
6
0
,
4
5
,
9
5
,
8
8
,
1
1
3
,
5
9
,
1
3
7
S
u
b
s
e
t
0
6
7
0
,
1,
1
8
7
,
1
8
0
,
9
0
,
9
6
,
9
5
,
177
1,
8
3
,
9
5
,
6
0
,
1
8
7
,
1
8
0
,
9
6
,
88
S
u
b
s
e
t
0
7
1,
9
5
,
9
6
,
1
8
7
,
1
6
2
,
1
2
5
,
1
8
0
,
1
7
3
1,
1
2
5
,
1
3
7
,
9
5
,
1
1
3
,
5
9
,
4
5
,
1
3
3
S
u
b
s
e
t
0
8
1,
7
1
,
9
6
,
9
5
,
1
6
2
,
1
8
0
,
4
5
,
187
1,
1
2
5
,
9
5
,
1
1
3
,
5
9
,
4
5
,
1
3
7
,
1
6
2
S
u
b
s
e
t
0
9
1,
9
5
,
1
8
0
,
1
8
7
,
1
8
4
,
1
6
5
,
1
5
8
,
96
1,
9
5
,
1
6
2
,
8
3
,
5
9
,
4
5
,
1
1
3
,
47
S
u
b
s
e
t
1
0
1,
1
6
2
,
1
8
7
,
1
8
0
,
1
8
4
,
1
7
0
,
9
6
,
95
1,
1
6
2
,
5
9
,
4
5
,
1
1
3
,
9
5
,
1
3
7
,
60
T
ab
le
4
.
Selecte
d
f
ea
tu
r
es v
ec
t
o
r
b
y
C
h
i²
,
GR
D
a
t
a
N
a
me
S
e
l
e
c
t
e
d
v
e
c
t
o
r
b
y
M
F
S
T
a
r
r
a
n
g
e
d
b
a
sed
o
n
r
a
n
k
v
a
l
u
e
V
o
t
e
s
S
u
b
s
e
t
01
1,
9
6
,
9
5
,
1
8
7
,
1
8
0
,
8
2
,
8
6
,
90
(
4
,
3
,
3
,
3
,
3
,
3
,
2
,
2
)
S
u
b
s
e
t
02
1,
9
5
,
1
8
0
,
4
5
,
1
1
3
,
5
9
,
1
6
2
,
1
3
7
(
4
,
4
,
3
,
2
,
2
,
2
,
2
,
2
)
S
u
b
s
e
t
03
1,
9
5
,
9
6
,
4
5
,
1
8
6
,
1
7
9
,
1
1
3
,
1
8
0
(
4
,
4
,
4
,
3
,
3
,
2
,
2
,
2
)
S
u
b
s
e
t
04
1,
9
5
,
9
6
,
1
8
7
,
1
8
0
,
9
4
,
1
8
4
,
90
(
4
,
3
,
3
,
3
,
3
,
2
,
2
,
2
)
S
u
b
s
e
t
05
1,
9
5
,
7
1
,
8
8
,
9
6
,
1
8
7
,
1
8
4
,
180
(
4
,
3
,
2
,
2
,
2
,
2
,
2
,
2
)
S
u
b
s
e
t
06
1,
9
5
,
9
6
,
1
8
7
,
1
8
0
,
8
8
,
8
3
,
186
(
4
,
3
,
3
,
3
,
3
,
2
,
2
,
2
)
S
u
b
s
e
t
07
1,
1
2
5
,
9
5
,
9
6
,
8
3
,
4
5
,
1
6
2
,
187
(
4
,
3
,
3
,
3
,
3
,
2
,
2
,
2
)
S
u
b
s
e
t
08
1,
9
5
,
9
6
,
7
1
,
1
6
2
,
4
5
,
1
8
7
,
180
(
4
,
3
,
3
,
2
,
2
,
2
,
2
,
2
)
S
u
b
s
e
t
09
1,
9
5
,
1
6
2
,
4
3
,
9
5
,
9
6
,
1
8
4
,
187
(
4
,
3
,
2
,
2
,
2
,
2
,
2
,
2
)
S
u
b
s
e
t
10
1,
9
5
,
9
6
,
8
3
,
1
6
2
,
1
3
7
,
1
8
7
,
1
8
0
(
4
,
3
,
3
,
3
,
2
,
2
,
2
,
2
)
Th
e
h
i
g
h
l
y
r
a
n
k
e
d
a
n
d
v
o
t
e
s fe
a
t
u
r
e
s
v
e
c
t
o
r
1,
9
5
,
9
6
,
1
8
0
,
1
8
7
,
1
6
2
,
4
5
,
83
3
.
4
.
1
.
M
F
ST
ba
s
ed
o
n r
a
nk
i
ng
a
nd
v
o
t
ing
T
h
is
s
ec
tio
n
p
r
esen
ts
th
e
s
ele
cted
v
ec
to
r
s
o
f
MFST,
wh
ich
ar
e
ar
r
an
g
ed
b
ased
o
n
r
an
k
v
alu
e.
I
n
ad
d
itio
n
,
th
e
v
o
tes
f
o
r
ea
ch
f
ea
tu
r
e
ar
e
ca
lcu
lated
.
Nex
t,
th
e
h
ig
h
ly
r
an
k
ed
an
d
v
o
tes
v
ec
to
r
o
f
f
ea
tu
r
es
is
co
n
clu
d
e
d
f
o
r
th
e
d
atasets
as
i
n
T
ab
le
4
.
T
h
e
tab
le
d
is
p
lay
s
th
e
to
p
8
f
ea
tu
r
es
f
o
r
ea
ch
d
ata
in
p
u
t,
wh
ich
wer
e
ch
o
s
en
u
s
in
g
f
o
u
r
tech
n
i
q
u
es
in
th
e
p
r
o
p
o
s
ed
MFST
f
r
am
e
wo
r
k
.
I
n
v
o
tin
g
,
t
h
e
n
u
m
b
e
r
4
d
e
n
o
tes
a
f
ea
tu
r
e
ch
o
s
en
b
y
f
o
u
r
tech
n
iq
u
es,
th
e
n
u
m
b
e
r
3
in
d
icate
s
a
f
ea
tu
r
e
ch
o
s
en
b
y
t
h
r
ee
tech
n
iq
u
es,
an
d
th
e
n
u
m
b
e
r
2
s
h
o
ws a
f
ea
tu
r
e
ch
o
s
en
b
y
j
u
s
t two
tech
n
iq
u
es.
B
ased
o
n
T
ab
le
5
r
ev
e
als
th
e
d
escr
ip
tio
n
o
f
f
ea
t
u
r
es
v
ec
to
r
th
at
wer
e
s
elec
ted
b
ased
o
n
p
r
o
p
o
s
ed
MFST
f
o
r
t
h
e
d
atasets
.
T
ab
le
5
also
d
is
p
lay
s
th
e
f
in
al
f
ea
tu
r
e
v
ec
to
r
ch
o
s
en
b
ased
o
n
MFST.
T
h
e
Mo
o
r
e
d
ataset
wh
ich
ca
r
r
ies
th
e
n
u
m
b
er
s
1
,
9
5
,
9
6
,
1
8
0
,
1
8
7
,
1
6
2
,
4
5
,
an
d
8
3
,
w
h
ich
r
ep
r
esen
t
th
e
f
ea
tu
r
e
,
s
er
v
e
r
p
o
r
t,
in
itial
win
d
o
w
-
b
y
tes
clie
n
t
-
s
er
v
er
,
in
itial
win
d
o
w
-
b
y
te
s
s
er
v
er
-
clien
t,
v
ar
d
ata
wir
e
b
a,
v
a
r
d
ata
ip
b
a,
m
ed
d
ata
I
P a
b
,
ac
tu
al
d
ata
p
k
ts
a
b
,
m
in
s
eg
m
s
ize
a
b
.
T
ab
le
5
.
Hig
h
l
y
r
an
k
ed
a
n
d
p
o
p
u
lar
f
ea
tu
r
es v
ec
to
r
s
ch
o
s
en
b
y
MFST
D
a
t
a
s
e
t
S
e
l
e
c
t
e
d
f
e
a
t
u
r
e
s
v
e
c
t
o
r
b
y
e
a
c
h
t
e
c
h
n
i
q
u
e
C
h
i
-
S
q
u
a
r
e
d
G
a
i
n
R
a
t
i
o
S
u
b
s
e
t
0
1
1,
9
5
,
1
8
7
,
9
6
,
9
3
,
9
0
,
1
8
0
,
186
1,
2
4
,
1
5
1
,
1
4
3
,
1
3
3
,
1
4
7
,
1
0
1
,
137
S
u
b
s
e
t
0
2
1,
1
8
0
,
9
5
,
9
6
,
8
5
,
9
3
,
1
8
7
,
186
1,
1
3
7
,
1
3
3
,
1
2
5
,
1
5
1
,
2
4
,
1
4
3
,
147
S
u
b
s
e
t
0
3
1,
9
5
,
1
8
7
,
1
8
4
,
1
8
0
,
1
7
7
,
9
3
,
96
1,
1
3
3
,
7
1
,
1
3
7
,
1
2
5
,
6
0
,
9
5
,
1
8
8
S
u
b
s
e
t
0
4
1,
9
5
,
9
6
,
9
4
,
9
0
,
8
3
,
1
8
6
,
1
8
7
1,
1
6
7
,
2
4
,
1
3
3
,
1
3
7
,
1
2
5
,
7
1
,
92
S
u
b
s
e
t
0
5
1,
1
8
7
,
1
8
4
,
9
5
,
1
0
0
,
1
8
0
,
4
4
,
96
1,
1
6
7
,
2
4
,
1
0
1
,
1
8
8
,
1
2
5
,
1
3
7
,
133
S
u
b
s
e
t
0
6
1,
9
5
,
8
3
,
9
6
,
1
8
7
,
1
8
6
,
1
8
4
,
1
8
0
1,
2
4
,
1
6
7
,
1
2
5
,
1
3
3
,
1
3
7
,
6
1
,
63
S
u
b
s
e
t
0
7
1,
9
5
,
9
6
,
8
3
,
9
0
,
1
8
7
,
1
8
0
,
184
1,
1
6
7
,
1
3
3
,
1
2
5
,
5
7
,
1
5
1
,
2
4
,
1
3
5
S
u
b
s
e
t
0
8
1,
9
5
,
8
3
,
9
6
,
1
8
7
,
1
8
4
,
1
8
6
,
1
7
9
1,
1
6
7
,
1
5
1
,
7
1
,
1
3
3
,
1
3
5
,
1
2
5
,
57
S
u
b
s
e
t
0
9
1,
9
6
,
9
5
,
9
3
,
1
8
4
,
1
7
7
,
8
3
,
187
1,
1
6
7
,
2
4
,
1
5
1
,
5
7
,
9
2
,
1
3
3
,
1
0
1
S
u
b
s
e
t
1
0
1,
9
3
,
1
8
7
,
9
5
,
9
6
,
1
5
9
,
1
8
0
,
1
8
4
1,
7
7
,
7
5
,
2
4
,
1
3
7
,
1
3
3
,
1
2
5
,
1
4
5
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
A
s
ig
n
ifica
n
t fe
a
tu
r
es v
ec
to
r
fo
r
in
tern
et
tr
a
ffic c
la
s
s
i
fica
tio
n
b
a
s
ed
o
n
…
(
A
l
h
a
mza
Mu
n
t
h
er
)
6965
4
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
C
las
s
if
icatio
n
ac
cu
r
ac
y
is
o
n
e
o
f
th
e
m
o
s
t
im
p
o
r
tan
t
m
et
r
ics
u
s
ed
to
m
ea
s
u
r
e
p
e
r
f
o
r
m
an
ce
an
d
ass
es
s
th
e
tech
n
iq
u
es
a
n
d
m
et
h
o
d
s
in
th
e
d
ata
m
in
in
g
ar
ea
.
Acc
u
r
ac
y
is
ev
al
u
ated
u
s
in
g
m
an
y
f
ac
to
r
s
.
B
u
t
th
e
s
tan
d
ar
d
f
ac
to
r
co
m
b
i
n
in
g
all
th
ese
f
ac
to
r
s
is
ca
lled
class
if
icatio
n
ac
cu
r
ac
y
,
wh
ich
is
d
ef
in
ed
an
d
f
o
r
m
u
lated
in
(
1
)
:
=
(
+
)
(
+
+
+
)
100%
(
1
)
MFST
is
ev
alu
ated
an
d
co
m
p
ar
ed
with
th
e
f
o
u
r
m
o
s
t
f
a
m
o
u
s
f
ilter
f
ea
tu
r
e
s
elec
tio
n
tech
n
iq
u
es
(
FC
B
F,
GR
,
C
h
i²
,
an
d
SF
E
)
.
MFST
an
d
o
th
er
f
ea
tu
r
e
s
elec
tio
n
tech
n
iq
u
es
ar
e
test
ed
o
v
e
r
th
r
ee
class
if
ier
s
,
n
am
ely
(
n
aï
v
e
B
ay
es,
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
i
n
e
,
an
d
C
4
.
5
)
.
T
h
o
s
e
class
if
ier
s
ac
h
iev
ed
h
ig
h
p
er
f
o
r
m
a
n
ce
in
ter
m
s
o
f
ac
cu
r
ac
y
,
p
r
o
ce
s
s
in
g
tim
e,
an
d
m
em
o
r
y
co
n
s
u
m
p
tio
n
.
T
h
er
e
f
o
r
e,
th
e
ex
p
er
im
en
ts
wer
e
r
ep
ea
ted
f
iv
e
tim
es f
o
r
ea
ch
class
if
ier
,
r
esu
ltin
g
in
1
5
v
alu
es f
o
r
ea
ch
ch
ar
t
.
Gen
er
ally
,
in
th
is
s
ec
tio
n
,
t
h
e
d
is
cu
s
s
io
n
o
f
p
r
o
p
o
s
ed
m
eth
o
d
s
will
b
e
d
iv
id
e
d
in
to
two
g
r
o
u
p
s
,
th
e
f
ir
s
t
r
ef
er
r
in
g
t
o
th
e
f
ir
s
t
s
ix
d
atasets
o
f
th
e
d
ataset
(
Mo
o
r
e
d
atasets
)
an
d
th
e
s
ec
o
n
d
r
ef
er
r
in
g
to
th
e
last
f
iv
e
d
atasets
.
I
n
all
ex
p
er
im
en
ts
,
w
e
allo
ca
ted
7
0
% o
f
ea
ch
d
atas
et
f
o
r
th
e
tr
ain
i
n
g
p
h
ase,
wh
ile
3
0
% wa
s
allo
ca
ted
to
th
e
test
in
g
p
h
ase.
No
r
m
ally
,
th
e
d
ata
f
o
r
tr
ai
n
in
g
is
m
o
r
e
th
an
th
at
o
f
th
e
test
in
g
p
h
a
s
e
b
ec
au
s
e
tr
ain
in
g
n
ee
d
s
ex
tr
a
d
ata
t
o
b
u
ild
th
e
cl
ass
if
ier
'
s
m
o
d
el.
Fig
u
r
e
3
s
h
o
ws
th
e
class
if
icatio
n
ac
cu
r
ac
y
r
esu
lts
f
o
r
th
e
f
i
r
s
t
s
ix
d
if
f
er
en
t
d
atasets
wh
o
s
e
s
izes
ar
e
co
n
v
er
g
en
t,
r
an
g
in
g
b
etwe
en
(
2
4
,
8
3
6
an
d
1
9
,
3
8
4
)
in
s
tan
ce
s
/in
p
u
t.
T
h
o
s
e
d
atasets
g
av
e
s
im
ilar
ac
cu
r
ac
y
,
p
r
esen
tin
g
ap
p
r
o
x
im
ately
n
aï
v
e
B
ay
es
(
NB
)
,
wh
ich
r
esu
lted
in
th
e
lo
west
v
alu
e
o
v
er
t
h
e
s
ix
d
atasets
an
d
u
s
in
g
th
e
f
o
u
r
tech
n
i
q
u
es
(
FC
B
F,
G
R
,
C
h
i²
,
an
d
SF
E
)
,
wh
ich
r
an
g
e
b
etwe
en
(
5
9
%
-
78
%
)
.
W
h
ile
th
e
p
r
o
p
o
s
ed
MFST
r
an
g
es
b
etwe
en
(
8
1
%
-
85
%
)
.
No
ta
b
ly
,
th
e
MFST
o
p
tim
ized
NB
'
s
p
er
f
o
r
m
an
ce
,
r
esu
ltin
g
in
m
o
r
e
ac
cu
r
ac
y
o
b
tain
ed
.
SVM
an
d
C
4
.
5
s
h
o
w
h
ig
h
p
e
r
f
o
r
m
an
ce
o
v
er
th
e
s
ix
d
atasets
wh
eth
er
u
s
in
g
MFST
o
r
an
y
o
f
th
e
f
o
u
r
c
o
m
p
etin
g
tech
n
i
q
u
es,
wh
ich
r
ef
lecte
d
th
e
p
o
wer
f
u
ln
ess
o
f
th
o
s
e
class
if
ier
s
r
an
g
in
g
b
etwe
en
(
9
6
.
4
%
-
9
9
.
7
%
)
;
s
till
,
o
n
e
ca
n
s
ee
a
s
tab
le
p
er
f
o
r
m
a
n
ce
wi
th
MFST
wh
er
e
th
e
p
er
ce
n
ta
g
e
r
an
g
es
b
etwe
en
(
9
9
.
2
%
-
9
9
.
7
%
)
.
Ov
er
all,
C
4
.
5
was sli
g
h
tly
b
etter
th
an
SVM.
Fig
u
r
e
3
.
T
h
e
class
if
icatio
n
ac
cu
r
ac
y
u
s
in
g
3
class
if
ier
s
an
d
4
f
ea
tu
r
e
s
elec
tio
n
s
tr
ateg
ies
was c
o
m
p
ar
ed
to
th
e
s
u
g
g
ested
MFST
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
14
,
No
.
6
,
Dec
em
b
e
r
20
24
:
6
9
5
8
-
6
9
6
8
6966
Als
o
,
Fig
u
r
e
3
h
i
g
h
lig
h
ts
th
e
class
if
icatio
n
ac
cu
r
ac
y
o
f
t
h
e
last
f
iv
e
Mo
o
r
e
d
atasets
.
T
h
e
s
ize
o
f
th
is
g
r
o
u
p
o
f
d
atasets
r
an
g
es
b
et
wee
n
(
5
5
,
4
9
4
an
d
6
6
,
2
4
8
)
in
s
tan
ce
s
.
Gen
er
ally
,
th
e
p
er
f
o
r
m
an
ce
o
f
class
if
ier
s
b
ased
o
n
th
e
f
ea
tu
r
e
s
elec
tio
n
tech
n
iq
u
es
is
d
r
o
p
p
e
d
b
ac
k
.
Naïv
e
B
ay
es
wa
s
th
e
wo
r
s
t
with
a
p
er
ce
n
tag
e
r
an
g
in
g
b
etwe
en
(
4
0
%
-
52
%
)
b
ased
o
n
t
h
e
f
o
u
r
tec
h
n
iq
u
es,
wh
ile
NB
with
MFST
r
aises
t
h
e
ac
cu
r
ac
y
r
an
g
in
g
b
etwe
en
(
5
5
%
-
66
%
)
.
SVM
an
d
C
4
.
5
o
f
f
er
h
i
g
h
er
ac
cu
r
ac
y
th
an
NB
.
Sti
ll,
th
e
r
an
g
e
is
d
ec
ay
ed
co
m
p
ar
e
d
with
th
e
f
ir
s
t
an
d
ab
o
v
em
e
n
tio
n
ed
g
r
o
u
p
,
wh
er
e
th
e
p
er
ce
n
tag
e
lies
b
etwe
en
(
7
6
-
9
0
)
u
s
in
g
th
e
f
o
u
r
tech
n
iq
u
es,
wh
ile
th
e
p
r
o
p
o
s
ed
MFST
ac
h
iev
ed
a
h
i
g
h
er
p
er
ce
n
tag
e
r
an
g
in
g
b
etwe
en
(
8
9
%
-
93
%
).
Als
o
,
it
is
n
o
te
d
th
at
th
e
GR
tech
n
iq
u
e
alwa
y
s
in
tr
o
d
u
ce
d
b
etter
r
esu
lts
in
co
n
tr
ast
wit
h
th
e
f
o
u
r
tech
n
iq
u
es
with
t
h
e
C
4
.
5
cla
s
s
if
ier
.
At
th
e
s
am
e
tim
e,
t
h
e
SF
E
an
d
C
h
i²
ar
e
c
o
m
p
eti
tiv
e
with
a
s
lig
h
t
d
if
f
er
en
ce
in
f
a
v
o
r
o
f
th
e
SF
E
b
y
r
an
g
in
g
th
e
p
er
ce
n
t
ag
e
b
etwe
en
(
8
2
%
-
85
%
)
wh
ile
C
h
i²
ac
h
iev
ed
(
7
7
%
-
87
%
)
.
FC
B
F
h
ad
g
iv
en
less
p
er
ce
n
tag
e
th
an
o
th
er
te
ch
n
iq
u
es
with
th
e
NB
class
if
ier
.
I
n
co
n
tr
ast,
th
e
p
er
ce
n
tag
e
was
in
cr
ea
s
ed
u
s
in
g
o
th
e
r
class
if
ier
s
s
u
ch
as
S
VM
an
d
C
4
.
5
,
wh
e
r
e
th
e
v
al
u
e
r
an
g
e
d
b
etwe
e
n
(
4
1
%
-
43
%
)
o
v
er
t
h
is
g
r
o
u
p
o
f
d
atasets
.
4
.
1
.
Resul
t
s
a
na
ly
s
i
s
o
f
cla
s
s
if
ica
t
io
n a
cc
ura
cy
T
h
is
s
ec
tio
n
in
itiates
a
cr
itical
d
is
cu
s
s
io
n
o
f
th
e
p
r
o
p
o
s
ed
f
e
atu
r
es
s
ec
tio
n
tech
n
iq
u
e
MFST
o
v
er
te
n
b
en
ch
m
ar
k
d
atasets
to
h
i
g
h
l
ig
h
t
its
d
if
f
er
en
ce
s
f
r
o
m
o
t
h
er
tech
n
iq
u
es
i
n
th
e
co
n
te
x
t
o
f
class
if
icatio
n
ac
cu
r
ac
y
.
T
h
e
r
esu
lts
o
f
th
e
f
ir
s
t
s
ix
d
atasets
ar
e
s
im
ilar
d
u
e
to
th
e
co
n
c
u
r
r
en
t
d
ata
s
iz
e.
NB
r
ep
o
r
ted
th
e
lo
west
ac
cu
r
ac
y
d
u
e
to
its
s
tr
o
n
g
ass
u
m
p
tio
n
(
i.e
.
,
an
y
two
f
ea
tu
r
es
ar
e
in
d
ep
en
d
en
t,
g
iv
en
th
e
o
u
tp
u
t
class
)
.
Ho
wev
er
,
th
e
ac
cu
r
ac
y
o
f
th
e
MFST
ca
n
b
e
in
cr
ea
s
ed
b
y
i
n
t
r
o
d
u
cin
g
f
ea
tu
r
e
s
ets th
at
ar
e
clo
s
ely
co
r
r
elate
d
to
th
e
class
.
T
h
e
o
th
er
class
if
ier
s
p
r
esen
ted
a
h
ig
h
p
er
ce
n
tag
e
with
co
n
v
er
g
e
n
t
p
er
f
o
r
m
a
n
ce
f
o
r
th
e
f
ea
t
u
r
es
tech
n
iq
u
e.
T
h
e
s
ec
o
n
d
g
r
o
u
p
o
f
d
atasets
s
h
o
wed
a
s
lig
h
t
d
if
f
er
en
ce
i
n
th
e
r
esu
lt
o
f
th
e
SV
M
an
d
C
4
.
5
d
u
e
to
in
cr
ea
s
ed
d
ata
s
ize
an
d
th
e
cl
ass
if
ier
s
’
y
ield
lo
wer
p
er
ce
n
tag
es.
T
h
e
MFST
r
ep
o
r
ted
th
e
h
ig
h
est
p
er
ce
n
ta
g
e
d
u
e
to
its
d
ep
e
n
d
en
ce
o
n
h
i
g
h
r
elativ
e
d
is
cr
im
in
ates
s
ets,
wh
ich
h
elp
s
o
p
tim
ize
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
class
if
ier
s
an
d
th
eir
co
r
r
esp
o
n
d
in
g
class
if
icatio
n
ac
cu
r
ac
y
,
wh
ich
m
atch
es with
th
e
s
u
g
g
e
s
ted
o
b
jectiv
es.
5
CO
NCLU
SI
O
N
Featu
r
es
p
lay
a
s
ig
n
if
ican
t
r
o
l
e
in
class
if
icatio
n
;
th
at'
s
wh
y
m
an
y
s
tu
d
ies
h
av
e
b
ee
n
p
r
o
p
o
s
ed
to
f
in
d
th
e
m
o
s
t
im
p
o
r
tan
t
f
ea
tu
r
e
v
e
cto
r
,
wh
ich
h
elp
s
class
if
y
th
e
tr
af
f
ic
ac
cu
r
ately
an
d
with
o
u
t
co
s
t.
T
h
is
p
ap
er
p
r
esen
ts
h
y
b
r
i
d
f
ea
tu
r
e
s
elec
tio
n
tech
n
iq
u
es,
wh
ich
a
d
o
p
t
f
o
u
r
tech
n
iq
u
es
an
d
two
f
il
ter
m
eth
o
d
s
.
T
h
e
s
elec
tio
n
f
ea
tu
r
es
v
ec
to
r
is
p
er
f
o
r
m
ed
o
n
ly
o
n
ce
at
th
e
b
eg
in
n
in
g
o
f
class
if
icatio
n
.
T
h
en
,
th
e
s
elec
ted
f
ea
tu
r
es
u
n
d
er
g
o
r
a
n
k
in
g
an
d
v
o
tin
g
f
i
lter
s
to
id
en
tify
th
e
m
o
s
t
s
ig
n
if
ican
t
f
ea
tu
r
es
am
o
n
g
th
e
im
p
o
r
tan
t
f
ea
tu
r
es
s
et.
T
h
e
ex
p
er
im
en
ts
u
s
ed
te
n
d
at
asets
with
v
ar
y
in
g
in
s
tan
ce
s
,
p
r
o
v
in
g
t
h
at
th
e
p
r
o
p
o
s
ed
tec
h
n
iq
u
e
in
cr
ea
s
es
th
e
ac
cu
r
ac
y
,
esp
ec
ially
f
o
r
h
ig
h
-
t
r
af
f
ic
d
atasets
.
RE
F
E
R
E
NC
E
S
[
1
]
H
.
A
l
a
z
z
a
m,
A
.
S
h
a
r
i
e
h
,
a
n
d
K
.
E.
S
a
b
r
i
,
“
A
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
a
l
g
o
r
i
t
h
m
f
o
r
i
n
t
r
u
s
i
o
n
d
e
t
e
c
t
i
o
n
sy
s
t
e
m
b
a
s
e
d
o
n
P
i
g
e
o
n
i
n
s
p
i
r
e
d
o
p
t
i
m
i
z
e
r
,
”
Ex
p
e
r
t
S
y
s
t
e
m
s
w
i
t
h
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
1
4
8
,
J
u
n
.
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
sw
a
.
2
0
2
0
.
1
1
3
2
4
9
.
[
2
]
M
.
M
.
A
b
u
a
l
h
a
j
,
A
.
A
.
A
b
u
-
S
h
a
r
e
h
a
,
M
.
O
.
H
i
a
r
i
,
Y
.
A
l
r
a
b
a
n
a
h
,
M
.
A
l
-
Zy
o
u
d
,
a
n
d
M
.
A
.
A
l
s
h
a
r
a
i
a
h
,
“
A
p
a
r
a
d
i
g
m
f
o
r
D
o
S
a
t
t
a
c
k
d
i
s
c
l
o
su
r
e
u
s
i
n
g
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
A
d
v
a
n
c
e
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
1
3
,
n
o
.
3
,
p
p
.
1
9
2
–
2
0
0
,
2
0
2
2
,
d
o
i
:
1
0
.
1
4
5
6
9
/
I
JA
C
S
A
.
2
0
2
2
.
0
1
3
0
3
2
5
.
[
3
]
C
.
W
.
C
h
e
n
,
Y
.
H
.
Ts
a
i
,
F
.
R
.
C
h
a
n
g
,
a
n
d
W
.
C
.
L
i
n
,
“
E
n
sem
b
l
e
f
e
a
t
u
r
e
se
l
e
c
t
i
o
n
i
n
me
d
i
c
a
l
d
a
t
a
set
s:
c
o
m
b
i
n
i
n
g
f
i
l
t
e
r
,
w
r
a
p
p
e
r
,
a
n
d
e
mb
e
d
d
e
d
f
e
a
t
u
r
e
se
l
e
c
t
i
o
n
r
e
s
u
l
t
s,”
E
x
p
e
rt
S
y
st
e
m
s
,
v
o
l
.
3
7
,
n
o
.
5
,
A
p
r
.
2
0
2
0
,
d
o
i
:
1
0
.
1
1
1
1
/
e
x
sy
.
1
2
5
5
3
.
[
4
]
M
.
H
a
mm
a
mi
,
S
.
B
e
c
h
i
k
h
,
C
.
C
.
H
u
n
g
,
a
n
d
L.
B
e
n
S
a
i
d
,
“
A
m
u
l
t
i
-
o
b
j
e
c
t
i
v
e
h
y
b
r
i
d
f
i
l
t
e
r
-
w
r
a
p
p
e
r
e
v
o
l
u
t
i
o
n
a
r
y
a
p
p
r
o
a
c
h
f
o
r
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
,
”
Me
m
e
t
i
c
C
o
m
p
u
t
i
n
g
,
v
o
l
.
1
1
,
n
o
.
2
,
p
p
.
1
9
3
–
2
0
8
,
J
u
l
.
2
0
1
9
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
2
2
9
3
-
018
-
0
2
6
9
-
2.
[
5
]
R
.
J.
U
r
b
a
n
o
w
i
c
z
,
M
.
M
e
e
k
e
r
,
W
.
La
C
a
v
a
,
R
.
S
.
O
l
s
o
n
,
a
n
d
J.
H
.
M
o
o
r
e
,
“
R
e
l
i
e
f
-
b
a
se
d
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
:
i
n
t
r
o
d
u
c
t
i
o
n
a
n
d
r
e
v
i
e
w
,
”
J
o
u
r
n
a
l
o
f
B
i
o
m
e
d
i
c
a
l
I
n
f
o
r
m
a
t
i
c
s
,
v
o
l
.
8
5
,
p
p
.
1
8
9
–
2
0
3
,
S
e
p
.
2
0
1
8
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
b
i
.
2
0
1
8
.
0
7
.
0
1
4
.
[
6
]
K
.
R
.
P
u
s
h
p
a
l
a
t
h
a
a
n
d
A
.
G
.
K
a
r
e
g
o
w
d
a
,
“
C
F
S
b
a
se
d
f
e
a
t
u
r
e
s
u
b
se
t
sel
e
c
t
i
o
n
f
o
r
e
n
h
a
n
c
i
n
g
c
l
a
ssi
f
i
c
a
t
i
o
n
o
f
si
mi
l
a
r
l
o
o
k
i
n
g
f
o
od
g
r
a
i
n
s
-
a
f
i
l
t
e
r
a
p
p
r
o
a
c
h
,
”
D
e
c
.
2
0
1
8
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
E
C
I
T.
2
0
1
7
.
8
4
5
3
4
0
3
.
[
7
]
N
.
G
o
p
i
k
a
a
n
d
A
.
M
.
k
o
w
s
h
a
l
a
y
a
M
.
E
.
,
“
C
o
r
r
e
l
a
t
i
o
n
b
a
se
d
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
a
l
g
o
r
i
t
h
m
f
o
r
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
,
”
i
n
2
0
1
8
3
rd
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
C
o
m
m
u
n
i
c
a
t
i
o
n
a
n
d
El
e
c
t
r
o
n
i
c
s
S
y
st
e
m
s
(
I
C
C
E
S
)
,
O
c
t
.
2
0
1
8
,
p
p
.
6
9
2
–
6
9
5
,
d
o
i
:
1
0
.
1
1
0
9
/
C
ESY
S
.
2
0
1
8
.
8
7
2
3
9
8
0
.
[
8
]
B
.
V
e
n
k
a
t
e
s
h
a
n
d
J.
A
n
u
r
a
d
h
a
,
“
A
r
e
v
i
e
w
o
f
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
a
n
d
i
t
s
met
h
o
d
s,
”
C
y
b
e
r
n
e
t
i
c
s
a
n
d
I
n
f
o
rm
a
t
i
o
n
T
e
c
h
n
o
l
o
g
i
e
s
,
v
o
l
.
1
9
,
n
o
.
1
,
p
p
.
3
–
2
6
,
M
a
r
.
2
0
1
9
,
d
o
i
:
1
0
.
2
4
7
8
/
C
A
I
T
-
2
0
1
9
-
0
0
0
1
.
[
9
]
S
.
Jad
h
a
v
,
H
.
H
e
,
a
n
d
K
.
Je
n
k
i
n
s,
“
I
n
f
o
r
mat
i
o
n
g
a
i
n
d
i
r
e
c
t
e
d
g
e
n
e
t
i
c
a
l
g
o
r
i
t
h
m
w
r
a
p
p
e
r
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
f
o
r
c
r
e
d
i
t
r
a
t
i
n
g
,
”
Ap
p
l
i
e
d
S
o
f
t
C
o
m
p
u
t
i
n
g
J
o
u
r
n
a
l
,
v
o
l
.
6
9
,
p
p
.
5
4
1
–
5
5
3
,
A
u
g
.
2
0
1
8
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
so
c
.
2
0
1
8
.
0
4
.
0
3
3
.
[
1
0
]
M
.
G
h
o
s
h
,
R
.
G
u
h
a
,
R
.
S
a
r
k
a
r
,
a
n
d
A
.
A
b
r
a
h
a
m
,
“
A
w
r
a
p
p
e
r
-
f
i
l
t
e
r
f
e
a
t
u
r
e
sel
e
c
t
i
o
n
t
e
c
h
n
i
q
u
e
b
a
se
d
o
n
a
n
t
c
o
l
o
n
y
o
p
t
i
mi
z
a
t
i
o
n
,
”
N
e
u
ra
l
C
o
m
p
u
t
i
n
g
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
3
2
,
n
o
.
1
2
,
p
p
.
7
8
3
9
–
7
8
5
7
,
A
p
r
.
2
0
2
0
,
d
o
i
:
1
0
.
1
0
0
7
/
s
0
0
5
2
1
-
0
1
9
-
0
4
1
7
1
-
3.
[
1
1
]
A
.
M
u
n
t
h
e
r
,
I
.
J.
M
o
h
a
mm
e
d
,
M
.
A
n
b
a
r
,
a
n
d
A
.
M
.
H
i
l
a
l
,
“
P
e
r
f
o
r
m
a
n
c
e
e
v
a
l
u
a
t
i
o
n
f
o
r
f
o
u
r
su
p
e
r
v
i
s
e
d
c
l
a
ss
i
f
i
e
r
s
i
n
i
n
t
e
r
n
e
t
t
r
a
f
f
i
c
c
l
a
ss
i
f
i
c
a
t
i
o
n
,
”
i
n
C
o
m
m
u
n
i
c
a
t
i
o
n
s
i
n
C
o
m
p
u
t
e
r
a
n
d
I
n
f
o
rm
a
t
i
o
n
S
c
i
e
n
c
e
,
v
o
l
.
1
1
3
2
C
C
I
S
,
S
p
r
i
n
g
e
r
S
i
n
g
a
p
o
r
e
,
2
0
2
0
,
p
p
.
1
6
8
–
1
8
1
.
[
1
2
]
A
.
M
u
n
t
h
e
r
,
R
.
R
.
O
t
h
ma
n
,
A
.
S
.
A
l
saa
d
i
,
a
n
d
M
.
A
n
b
a
r
,
“
A
p
e
r
f
o
r
ma
n
c
e
st
u
d
y
o
f
h
i
d
d
e
n
m
a
r
k
o
v
m
o
d
e
l
a
n
d
r
a
n
d
o
m
f
o
r
e
st
i
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
A
s
ig
n
ifica
n
t fe
a
tu
r
es v
ec
to
r
fo
r
in
tern
et
tr
a
ffic c
la
s
s
i
fica
tio
n
b
a
s
ed
o
n
…
(
A
l
h
a
mza
Mu
n
t
h
er
)
6967
i
n
t
e
r
n
e
t
t
r
a
f
f
i
c
c
l
a
ssi
f
i
c
a
t
i
o
n
,
”
i
n
L
e
c
t
u
re
N
o
t
e
s i
n
El
e
c
t
r
i
c
a
l
E
n
g
i
n
e
e
r
i
n
g
,
v
o
l
.
3
7
6
,
S
p
r
i
n
g
e
r
S
i
n
g
a
p
o
r
e
,
2
0
1
6
,
p
p
.
3
1
9
–
3
2
9
.
[
1
3
]
S
.
A
l
j
a
w
a
r
n
e
h
,
M
.
A
l
d
w
a
i
r
i
,
a
n
d
M
.
B
.
Y
a
sse
i
n
,
“
A
n
o
m
a
l
y
-
b
a
se
d
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
sy
s
t
e
m
t
h
r
o
u
g
h
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
a
n
a
l
y
s
i
s
a
n
d
b
u
i
l
d
i
n
g
h
y
b
r
i
d
e
f
f
i
c
i
e
n
t
m
o
d
e
l
,
”
J
o
u
r
n
a
l
o
f
C
o
m
p
u
t
a
t
i
o
n
a
l
S
c
i
e
n
c
e
,
v
o
l
.
2
5
,
p
p
.
1
5
2
–
1
6
0
,
M
a
r
.
2
0
1
8
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
o
c
s
.
2
0
1
7
.
0
3
.
0
0
6
.
[
1
4
]
M
.
A
.
A
m
b
u
sa
i
d
i
,
X
.
H
e
,
P
.
N
a
n
d
a
,
a
n
d
Z.
T
a
n
,
“
B
u
i
l
d
i
n
g
a
n
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
s
y
st
e
m
u
s
i
n
g
a
f
i
l
t
e
r
-
b
a
se
d
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
a
l
g
o
r
i
t
h
m,
”
I
EEE
T
ra
n
s
a
c
t
i
o
n
s
o
n
C
o
m
p
u
t
e
rs
,
v
o
l
.
6
5
,
n
o
.
1
0
,
p
p
.
2
9
8
6
–
2
9
9
8
,
O
c
t
.
2
0
1
6
,
d
o
i
:
1
0
.
1
1
0
9
/
T
C
.
2
0
1
6
.
2
5
1
9
9
1
4
.
[
1
5
]
Y
.
Z
h
o
u
,
G
.
C
h
e
n
g
,
S
.
J
i
a
n
g
,
a
n
d
M
.
D
a
i
,
“
B
u
i
l
d
i
n
g
a
n
e
f
f
i
c
i
e
n
t
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
sy
s
t
e
m
b
a
s
e
d
o
n
f
e
a
t
u
r
e
sel
e
c
t
i
o
n
a
n
d
e
n
s
e
mb
l
e
c
l
a
ssi
f
i
e
r
,
”
C
o
m
p
u
t
e
r
N
e
t
w
o
rks
,
v
o
l
.
1
7
4
,
J
u
n
.
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
o
m
n
e
t
.
2
0
2
0
.
1
0
7
2
4
7
.
[
1
6
]
Y
.
Zh
o
n
g
e
t
a
l
.
,
“
H
E
LA
D
:
A
n
o
v
e
l
n
e
t
w
o
r
k
a
n
o
ma
l
y
d
e
t
e
c
t
i
o
n
m
o
d
e
l
b
a
sed
o
n
h
e
t
e
r
o
g
e
n
e
o
u
s
e
n
s
e
mb
l
e
l
e
a
r
n
i
n
g
,
”
C
o
m
p
u
t
e
r
N
e
t
w
o
rks
,
v
o
l
.
1
6
9
,
M
a
r
.
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
o
m
n
e
t
.
2
0
1
9
.
1
0
7
0
4
9
.
[
1
7
]
F
.
S
a
l
o
,
A
.
B
.
N
a
ss
i
f
,
a
n
d
A
.
Esse
x
,
“
D
i
m
e
n
s
i
o
n
a
l
i
t
y
r
e
d
u
c
t
i
o
n
w
i
t
h
I
G
-
P
C
A
a
n
d
e
n
sem
b
l
e
c
l
a
ss
i
f
i
e
r
f
o
r
n
e
t
w
o
r
k
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
,
”
C
o
m
p
u
t
e
r
N
e
t
w
o
rks
,
v
o
l
.
1
4
8
,
p
p
.
1
6
4
–
1
7
5
,
Ja
n
.
2
0
1
9
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
o
m
n
e
t
.
2
0
1
8
.
1
1
.
0
1
0
.
[
1
8
]
M
.
A
b
d
u
l
l
a
h
,
A
.
S
.
A
l
-
S
h
a
n
n
a
q
,
S
.
A
l
ma
b
d
y
,
A
.
B
a
l
a
mas
h
,
a
n
d
A
.
A
l
s
h
a
n
n
a
q
,
“
En
h
a
n
c
e
d
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
sy
s
t
e
m
u
s
i
n
g
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
me
t
h
o
d
a
n
d
e
n
se
mb
l
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
ms,”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
I
n
f
o
rm
a
t
i
o
n
S
e
c
u
r
i
t
y
,
v
o
l
.
1
6
,
n
o
.
2
,
p
p
.
4
8
–
5
5
,
2
0
1
8
.
[
1
9
]
N
.
M
o
u
s
t
a
f
a
,
B
.
Tu
r
n
b
u
l
l
,
a
n
d
K
.
K
.
R
.
C
h
o
o
,
“
A
n
e
n
s
e
m
b
l
e
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
t
e
c
h
n
i
q
u
e
b
a
se
d
o
n
p
r
o
p
o
se
d
st
a
t
i
st
i
c
a
l
f
l
o
w
f
e
a
t
u
r
e
s
f
o
r
p
r
o
t
e
c
t
i
n
g
n
e
t
w
o
r
k
t
r
a
f
f
i
c
o
f
i
n
t
e
r
n
e
t
o
f
t
h
i
n
g
s,”
I
EEE
I
n
t
e
r
n
e
t
o
f
T
h
i
n
g
s
J
o
u
r
n
a
l
,
v
o
l
.
6
,
n
o
.
3
,
p
p
.
4
8
1
5
–
4
8
3
0
,
Ju
n
.
2
0
1
9
,
d
o
i
:
1
0
.
1
1
0
9
/
JI
O
T.
2
0
1
8
.
2
8
7
1
7
1
9
.
[
2
0
]
D
.
Ta
n
g
,
L.
T
a
n
g
,
R
.
D
a
i
,
J.
C
h
e
n
,
X
.
Li
,
a
n
d
J.
J.
P
.
C
.
R
o
d
r
i
g
u
e
s
,
“
M
F
-
A
d
a
b
o
o
s
t
:
LD
o
S
a
t
t
a
c
k
d
e
t
e
c
t
i
o
n
b
a
s
e
d
o
n
m
u
l
t
i
-
f
e
a
t
u
r
e
s
a
n
d
i
mp
r
o
v
e
d
A
d
a
b
o
o
st
,
”
Fu
t
u
re
G
e
n
e
ra
t
i
o
n
C
o
m
p
u
t
e
r
S
y
st
e
m
s
,
v
o
l
.
1
0
6
,
p
p
.
3
4
7
–
3
5
9
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
f
u
t
u
r
e
.
2
0
1
9
.
1
2
.
0
3
4
.
[
2
1
]
K
.
L
i
n
,
X
.
X
u
,
a
n
d
F
.
X
i
a
o
,
“
M
F
F
u
si
o
n
:
a
mu
l
t
i
-
l
e
v
e
l
f
e
a
t
u
r
e
s
f
u
s
i
o
n
mo
d
e
l
f
o
r
ma
l
i
c
i
o
u
s
t
r
a
f
f
i
c
d
e
t
e
c
t
i
o
n
b
a
se
d
o
n
d
e
e
p
l
e
a
r
n
i
n
g
,
”
C
o
m
p
u
t
e
r
N
e
t
w
o
rks
,
v
o
l
.
2
0
2
,
Ja
n
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
o
mn
e
t
.
2
0
2
1
.
1
0
8
6
5
8
.
[
2
2
]
S
.
Li
,
L
.
J
i
a
n
g
,
Q
.
Z
h
a
n
g
,
Z
.
W
a
n
g
,
Z
.
Ti
a
n
,
a
n
d
M
.
G
u
i
z
a
n
i
,
“
A
m
a
l
i
c
i
o
u
s mi
n
i
n
g
c
o
d
e
d
e
t
e
c
t
i
o
n
m
e
t
h
o
d
b
a
s
e
d
o
n
m
u
l
t
i
-
f
e
a
t
u
r
e
s
f
u
si
o
n
,
”
I
E
EE
T
r
a
n
sa
c
t
i
o
n
s
o
n
N
e
t
w
o
r
k
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
r
i
n
g
,
v
o
l
.
1
0
,
n
o
.
5
,
p
p
.
2
7
3
1
–
2
7
3
9
,
S
e
p
.
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
TN
S
E.
2
0
2
2
.
3
1
5
5
1
8
7
.
[
2
3
]
L.
Y
u
a
n
d
H
.
Li
u
,
“
F
e
a
t
u
r
e
se
l
e
c
t
i
o
n
f
o
r
h
i
g
h
-
d
i
me
n
s
i
o
n
a
l
d
a
t
a
:
a
f
a
st
c
o
r
r
e
l
a
t
i
o
n
-
b
a
s
e
d
f
i
l
t
e
r
s
o
l
u
t
i
o
n
,
”
Pro
c
e
e
d
i
n
g
s,
T
w
e
n
t
i
e
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
Ma
c
h
i
n
e
L
e
a
r
n
i
n
g
,
v
o
l
.
2
,
p
p
.
8
5
6
–
8
6
3
,
2
0
0
3
.
[
2
4
]
Q
.
S
o
n
g
,
J.
N
i
,
a
n
d
G
.
W
a
n
g
,
“
A
f
a
st
c
l
u
s
t
e
r
i
n
g
-
b
a
se
d
f
e
a
t
u
r
e
s
u
b
set
s
e
l
e
c
t
i
o
n
a
l
g
o
r
i
t
h
m
f
o
r
h
i
g
h
-
d
i
me
n
si
o
n
a
l
d
a
t
a
,
”
I
EEE
T
ra
n
s
a
c
t
i
o
n
s
o
n
K
n
o
w
l
e
d
g
e
a
n
d
D
a
t
a
En
g
i
n
e
e
ri
n
g
,
v
o
l
.
2
5
,
n
o
.
1
,
p
p
.
1
–
1
4
,
Jan
.
2
0
1
3
,
d
o
i
:
1
0
.
1
1
0
9
/
T
K
D
E.
2
0
1
1
.
1
8
1
.
[
2
5
]
A
.
T
h
a
k
k
a
r
a
n
d
R
.
L
o
h
i
y
a
,
“
A
t
t
a
c
k
c
l
a
ss
i
f
i
c
a
t
i
o
n
u
si
n
g
f
e
a
t
u
r
e
se
l
e
c
t
i
o
n
t
e
c
h
n
i
q
u
e
s:
a
c
o
m
p
a
r
a
t
i
v
e
s
t
u
d
y
,
”
J
o
u
rn
a
l
o
f
Am
b
i
e
n
t
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
H
u
m
a
n
i
z
e
d
C
o
m
p
u
t
i
n
g
,
v
o
l
.
1
2
,
n
o
.
1
,
p
p
.
1
2
4
9
–
1
2
6
6
,
J
u
n
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
2
6
5
2
-
020
-
0
2
1
6
7
-
9.
[
2
6
]
A
.
H
.
M
o
h
a
mm
a
d
,
“
C
o
m
p
a
r
i
n
g
t
w
o
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
s
m
e
t
h
o
d
s
(
I
n
f
o
r
ma
t
i
o
n
g
a
i
n
a
n
d
g
a
i
n
r
a
t
i
o
)
o
n
t
h
r
e
e
d
i
f
f
e
r
e
n
t
c
l
a
s
si
f
i
c
a
t
i
o
n
a
l
g
o
r
i
t
h
ms
u
s
i
n
g
a
r
a
b
i
c
d
a
t
a
s
e
t
,
”
J
o
u
rn
a
l
o
f
T
h
e
o
re
t
i
c
a
l
a
n
d
A
p
p
l
i
e
d
I
n
f
o
rm
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
9
6
,
n
o
.
6
,
p
p
.
1
5
6
1
–
1
5
6
9
,
2
0
1
8
.
[
2
7
]
A
.
S
h
a
r
ma
a
n
d
S
.
D
e
y
,
“
P
e
r
f
o
r
ma
n
c
e
i
n
v
e
s
t
i
g
a
t
i
o
n
o
f
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
m
e
t
h
o
d
s
a
n
d
se
n
t
i
me
n
t
l
e
x
i
c
o
n
s
f
o
r
se
n
t
i
me
n
t
a
n
a
l
y
s
i
s,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
C
o
m
p
u
t
e
r
Ap
p
l
i
c
a
t
i
o
n
s
,
n
o
.
Ju
n
e
,
p
p
.
1
5
–
2
0
,
2
0
1
2
.
[
2
8
]
P
.
Em
e
r
so
n
,
“
T
h
e
o
r
i
g
i
n
a
l
B
o
r
d
a
c
o
u
n
t
a
n
d
p
a
r
t
i
a
l
v
o
t
i
n
g
,
”
S
o
c
i
a
l
C
h
o
i
c
e
a
n
d
We
l
f
a
r
e
,
v
o
l
.
4
0
,
n
o
.
2
,
p
p
.
3
5
3
–
3
5
8
,
O
c
t
.
2
0
1
3
,
d
o
i
:
1
0
.
1
0
0
7
/
s
0
0
3
5
5
-
011
-
0
6
0
3
-
9.
[
2
9
]
A
.
W
.
M
o
o
r
e
a
n
d
D
.
Zu
e
v
,
“
I
n
t
e
r
n
e
t
t
r
a
f
f
i
c
c
l
a
ss
i
f
i
c
a
t
i
o
n
u
si
n
g
b
a
y
e
si
a
n
a
n
a
l
y
si
s
t
e
c
h
n
i
q
u
e
s,
”
i
n
P
e
rf
o
rm
a
n
c
e
E
v
a
l
u
a
t
i
o
n
R
e
v
i
e
w
,
2
0
0
5
,
v
o
l
.
3
3
,
n
o
.
1
,
p
p
.
5
0
–
6
0
,
d
o
i
:
1
0
.
1
1
4
5
/
1
0
7
1
6
9
0
.
1
0
6
4
2
2
0
.
[
3
0
]
A
.
W
.
M
o
o
r
e
a
n
d
K
.
P
a
p
a
g
i
a
n
n
a
k
i
,
“
To
w
a
r
d
t
h
e
a
c
c
u
r
a
t
e
i
d
e
n
t
i
f
i
c
a
t
i
o
n
o
f
n
e
t
w
o
r
k
a
p
p
l
i
c
a
t
i
o
n
s,
”
i
n
L
e
c
t
u
r
e
N
o
t
e
s
i
n
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
3
4
3
1
,
S
p
r
i
n
g
e
r
B
e
r
l
i
n
H
e
i
d
e
l
b
e
r
g
,
2
0
0
5
,
p
p
.
4
1
–
5
4
.
[
3
1
]
H
.
Li
u
a
n
d
H
.
M
o
t
o
d
a
,
“
F
e
a
t
u
r
e
sel
e
c
t
i
o
n
a
sp
e
c
t
s,”
i
n
Fe
a
t
u
re
S
e
l
e
c
t
i
o
n
f
o
r
K
n
o
w
l
e
d
g
e
D
i
s
c
o
v
e
r
y
a
n
d
D
a
t
a
Mi
n
i
n
g
,
S
p
r
i
n
g
e
r
U
S
,
1
9
9
8
,
p
p
.
4
3
–
7
2
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Alh
a
m
z
a
Mu
n
th
e
r
is
a
n
a
ss
is
tan
t
p
r
o
fe
ss
o
r
i
n
u
n
i
v
e
rsity
tec
h
n
o
lo
g
y
a
n
d
a
p
p
li
e
d
sc
ien
c
e
s
in
S
u
r,
S
u
lt
a
n
a
te
o
f
Om
a
n
.
He
re
c
e
iv
e
d
h
is
B
.
S
c
.
in
c
o
m
p
u
ter
a
n
d
so
ftwa
re
e
n
g
in
e
e
rin
g
fro
m
th
e
Un
i
v
e
rsity
o
f
Tec
h
n
o
l
o
g
y
in
Ba
g
h
d
a
d
i
n
2
0
0
3
.
He
re
c
e
iv
e
d
a
m
a
ste
r'
s
d
e
g
re
e
in
a
d
v
a
n
c
e
d
c
o
m
p
u
ter
n
e
t
wo
rk
s
fr
o
m
U
n
iv
e
rsiti
S
a
i
n
s
M
a
l
a
y
sia
(USM
)
in
2
0
1
2
,
a
n
d
in
M
a
rc
h
2
0
1
7
,
h
e
wa
s a
w
a
rd
e
d
a
P
h
.
D.
in
c
o
m
p
u
ter en
g
i
n
e
e
rin
g
fro
m
Un
iv
e
rsiti
M
a
lay
sia
P
e
rli
s,
M
a
lay
sia
.
His
re
se
a
rc
h
fo
c
u
se
s
o
n
o
v
e
rlay
n
e
two
rk
s,
m
u
lt
i
m
e
d
ia
d
istri
b
u
ti
o
n
,
traffic
e
n
g
in
e
e
rin
g
,
d
a
ta
m
in
i
n
g
,
a
n
d
m
a
c
h
in
e
lea
rn
in
g
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
a
lh
a
m
z
a
.
wa
rd
i@u
tas
.
e
d
u
.
o
m
.
Mo
sle
h
M.
A
b
u
a
l
h
a
j
is
a
p
ro
fe
ss
o
r
i
n
Al
-
Ah
l
iy
y
a
Am
m
a
n
Un
iv
e
rsit
y
.
He
re
c
e
iv
e
d
h
is
first
d
e
g
re
e
in
c
o
m
p
u
ter
sc
ien
c
e
fro
m
P
h
il
a
d
e
lp
h
ia
Un
iv
e
rsity
,
Jo
rd
a
n
,
i
n
Ju
l
y
2
0
0
4
,
m
a
ste
r’s
d
e
g
re
e
in
c
o
m
p
u
t
e
r
in
fo
rm
a
ti
o
n
sy
ste
m
fro
m
th
e
Ara
b
Ac
a
d
e
m
y
fo
r
Ba
n
k
i
n
g
a
n
d
F
i
n
a
n
c
ial
S
c
ien
c
e
s,
Jo
rd
a
n
i
n
Ju
l
y
2
0
0
7
,
a
n
d
d
o
c
to
ra
te
d
e
g
r
e
e
in
m
u
lt
ime
d
ia
n
e
two
r
k
s
p
ro
t
o
c
o
ls
fr
o
m
Un
iv
e
rsit
y
S
a
i
n
s
M
a
lay
sia
in
2
0
1
1
.
His
re
se
a
rc
h
a
r
e
a
o
f
in
tere
st
in
c
l
u
d
e
s
Vo
I
P
,
m
u
lt
ime
d
ia
n
e
two
rk
i
n
g
,
a
n
d
c
o
n
g
e
stio
n
c
o
n
tr
o
l
.
A
p
a
rt
fr
o
m
re
se
a
rc
h
,
Dr.
A
b
u
-
Al
h
a
j
a
lso
d
o
e
s
c
o
n
su
lt
a
n
c
y
se
rv
ice
s
in
th
e
a
b
o
v
e
re
se
a
rc
h
a
r
e
a
s
a
n
d
d
irec
ts
t
h
e
Cisc
o
a
c
a
d
e
m
y
tea
m
a
t
Al
-
Ah
li
y
y
a
Am
m
a
n
Un
iv
e
rsit
y
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
m
.
a
b
u
a
lh
a
j@a
m
m
a
n
u
.
e
d
u
.
j
o
.
Evaluation Warning : The document was created with Spire.PDF for Python.