I
nte
rna
t
io
na
l J
o
urna
l o
f
I
nfo
rm
a
t
ics a
nd
Co
m
m
un
ica
t
io
n T
ec
hn
o
lo
g
y
(
I
J
-
I
CT
)
Vo
l.
9
,
No
.
3
,
Dec
em
b
er
2
0
2
0
,
p
p
.
15
7
~
1
70
I
SS
N:
2252
-
8
7
7
6
,
DOI
: 1
0
.
1
1
5
9
1
/iji
ct.
v
9
i3
.
p
p
1
5
7
-
1
70
157
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ict.
ia
esco
r
e.
co
m
Im
pro
v
ed ICH
I
s
qua
re f
ea
t
ure sel
e
ction met
ho
d f
o
r
Ara
bic
cla
ss
ifiers
H
a
deel
N.
Als
ha
er
,
M
o
ha
m
m
ed
A.
O
t
a
ir,
L
a
it
h Abua
lig
a
h
F
a
c
u
lt
y
o
f
C
o
m
p
u
ter
S
c
ien
c
e
s a
n
d
In
f
o
rm
a
ti
c
s,
Am
m
a
n
Ara
b
Un
iv
e
rsity
,
Jo
r
d
a
n
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
an
23
,
2
0
20
R
ev
is
ed
Mar
3
,
2
0
20
Acc
ep
ted
Mar
2
0
,
2
0
20
F
e
a
tu
re
se
lec
ti
o
n
p
ro
b
lem
is
o
n
e
o
f
t
h
e
m
a
in
im
p
o
rtan
t
p
ro
b
lem
s
in
t
h
e
tex
t
a
n
d
d
a
ta
m
in
i
n
g
d
o
m
a
in
.
T
h
is
p
a
p
e
r
p
re
se
n
ts
a
c
o
m
p
a
ra
ti
v
e
st
u
d
y
o
f
fe
a
tu
re
se
lec
ti
o
n
m
e
th
o
d
s
f
o
r
Ara
b
ic
tex
t
c
las
sifica
ti
o
n
.
F
iv
e
o
f
th
e
fe
a
tu
r
e
se
lec
ti
o
n
m
e
th
o
d
s
we
re
se
lec
ted
:
ICHI
sq
u
a
re
,
CHI
sq
u
a
re
,
In
fo
rm
a
ti
o
n
G
a
in
,
M
u
tu
a
l
In
fo
rm
a
ti
o
n
a
n
d
Wr
a
p
p
e
r.
It
wa
s
tes
ted
with
fi
v
e
c
las
sifica
ti
o
n
a
lg
o
rit
h
m
s:
Ba
y
e
s Ne
t,
Na
iv
e
Ba
y
e
s,
Ra
n
d
o
m
F
o
re
st
,
De
c
isio
n
Tree
a
n
d
Artifi
c
ial
Ne
u
ra
l
Ne
two
rk
s.
In
a
d
d
it
io
n
,
Da
ta Co
ll
e
c
ti
o
n
wa
s u
se
d
in
Ara
b
ic c
o
n
sisti
n
g
o
f
9
0
5
5
d
o
c
u
m
e
n
ts,
wh
ich
we
re
c
o
m
p
a
r
e
d
b
y
f
o
u
r
c
rit
e
ria:
P
re
c
isio
n
,
Re
c
a
ll
,
F
-
m
e
a
su
re
a
n
d
Ti
m
e
to
b
u
il
d
m
o
d
e
l.
Th
e
re
su
lt
s
sh
o
we
d
t
h
a
t
th
e
imp
ro
v
e
d
I
CHI fea
tu
re
se
lec
ti
o
n
g
o
t
a
lmo
st
a
ll
th
e
b
e
st r
e
su
lt
s in
c
o
m
p
a
riso
n
with
o
t
h
e
r
m
e
th
o
d
s.
K
ey
w
o
r
d
s
:
C
las
s
if
icatio
n
alg
o
r
ith
m
s
Featu
r
e
s
elec
tio
n
m
eth
o
d
s
T
ex
t c
lass
if
icatio
n
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
:
L
aith
Ab
u
alig
ah
,
Facu
lty
o
f
C
o
m
p
u
ter
Scien
ce
s
an
d
I
n
f
o
r
m
atics,
Am
m
an
Ar
ab
Un
i
v
er
s
ity
,
J
o
r
d
an
Stre
et
–
Mu
b
is
,
Am
m
a
n
-
J
o
r
d
an
.
E
m
ail:
lx
.
8
9
@
y
ah
o
o
.
co
m
1.
I
NT
RO
D
UCT
I
O
N
R
esear
ch
an
d
d
ev
elo
p
m
en
t
in
th
e
s
cien
ce
o
f
in
f
o
r
m
ati
o
n
r
etr
iev
al
an
d
d
ev
el
o
p
m
e
n
t
in
th
e
tech
n
o
lo
g
y
u
s
ed
h
as
in
cr
ea
s
ed
in
m
an
y
ap
p
licatio
n
s
in
v
a
r
io
u
s
d
ata
s
u
ch
as
tex
ts
,
im
ag
es,
s
o
u
n
d
.
.
.
etc.
T
h
ese
d
ata
ar
e
tex
tu
al
in
d
if
f
er
e
n
t
lan
g
u
ag
es,
an
d
h
er
e
we
s
p
ea
k
th
e
Ar
ab
ic
lan
g
u
a
g
e
as
th
e
Ar
ab
ic
lan
g
u
ag
e
f
lo
u
r
is
h
ed
in
t
h
e
f
ield
o
f
in
f
o
r
m
atio
n
r
etr
ie
v
al
an
d
s
p
ec
if
ical
ly
in
th
e
f
ield
o
f
class
if
icatio
n
[
1
]
.
T
h
is
b
o
o
m
h
as
led
r
esear
ch
er
s
to
r
esear
ch
an
d
d
e
v
elo
p
p
r
o
p
er
ties
th
at
im
p
r
o
v
e
r
esu
lts
an
d
im
p
r
o
v
e
ef
f
ici
en
cy
.
I
n
f
o
r
m
atio
n
R
etr
iev
al
(
I
R
)
is
a
f
ie
ld
o
f
co
m
p
u
ter
s
cien
ce
o
f
g
r
ea
t im
p
o
r
tan
ce
in
o
u
r
tim
e
b
ec
au
s
e
o
f
t
h
e
i
n
cr
ea
s
in
g
v
o
lu
m
e
o
f
in
f
o
r
m
atio
n
[
2
]
.
T
h
is
in
f
o
r
m
atio
n
m
ay
n
ee
d
to
b
e
a
r
r
an
g
ed
an
d
class
if
ied
s
o
th
at
it
ca
n
b
e
ea
s
ily
r
etr
iev
ed
.
T
ex
t
class
if
icatio
n
(
T
C
)
a
p
r
o
ce
s
s
th
at
h
as
b
ee
n
e
m
er
g
ed
in
an
im
p
o
r
tan
t
m
a
n
n
er
in
v
ar
i
o
u
s
f
ield
s
,
esp
ec
ially
in
ar
ea
s
o
n
th
e
I
n
ter
n
et
[
3
]
.
T
ex
t
m
in
in
g
is
a
tex
tu
al
an
aly
s
is
o
f
d
ata
in
n
atu
r
al
lan
g
u
ag
e
tex
t
an
d
s
ee
k
s
to
ex
t
r
ac
t
u
s
ef
u
l
in
f
o
r
m
atio
n
f
r
o
m
tex
tu
al
d
ata
[
4
]
.
I
n
ad
d
itio
n
,
te
x
t
m
in
in
g
h
elp
s
o
r
g
a
n
izatio
n
s
ex
tr
ac
t
v
alu
ab
le
id
ea
s
f
r
o
m
d
o
cu
m
e
n
t
co
n
ten
t
.
T
ex
t
m
in
i
n
g
p
r
o
ce
s
s
u
s
ed
to
in
c
r
ea
s
e
th
e
ef
f
icien
cy
o
f
t
h
e
tex
t
r
etr
iev
al
p
r
o
ce
s
s
b
y
d
is
co
v
er
in
g
p
atter
n
s
in
th
e
te
x
t
an
d
th
e
r
elatio
n
s
h
ip
s
b
etwe
en
th
em
to
h
elp
th
e
r
etr
iev
al
o
f
tex
ts
co
r
r
ec
tly
[
5
]
.
T
h
e
m
o
s
t
im
p
o
r
tan
t
a
p
p
licatio
n
s
o
n
T
ex
t
Min
in
g
a
r
e:
I
R
,
I
n
f
o
r
m
atio
n
E
x
tr
ac
tio
n
,
C
lass
if
icatio
n
an
d
Natu
r
al
L
an
g
u
ag
e
Pro
ce
s
s
in
g
(
NL
P).
Me
an
in
g
it
ex
tr
ac
ts
u
s
ef
u
l
in
f
o
r
m
atio
n
f
r
o
m
a
lar
g
e
am
o
u
n
t
o
f
d
ata
to
r
ep
lace
d
ata
an
d
in
f
o
r
m
atio
n
s
ea
r
ch
p
r
o
b
lem
s
[
6
]
.
T
h
er
e
ar
e
m
an
y
p
r
o
ce
s
s
es
th
at
ca
n
b
e
s
ea
r
ch
ed
f
r
o
m
th
e
d
ata
co
llectio
n
u
s
ed
th
r
o
u
g
h
th
e
p
r
e
-
p
r
o
ce
s
s
in
g
to
th
e
class
if
icatio
n
p
r
o
ce
s
s
as
p
r
e
-
p
r
o
ce
s
s
in
g
co
n
tain
s
s
ev
er
al
s
tep
s
[
7
,
8
]
,
i
n
clu
d
in
g
:
T
o
k
en
izatio
n
,
No
r
m
aliza
tio
n
,
Sto
p
wo
r
d
r
e
m
o
v
al
a
n
d
Stem
m
in
g
.
T
h
er
e
ar
e
s
ev
e
r
al
alg
o
r
ith
m
s
in
th
es
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
7
6
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
,
Vo
l.
9
,
No
.
3,
Dec
em
b
e
r
20
20
:
15
7
–
1
70
158
p
r
o
ce
s
s
es.
I
n
ad
d
itio
n
to
th
is
t
h
er
e
is
a
s
tag
e
ca
lled
f
ea
tu
r
e
s
elec
tio
n
s
tep
an
d
th
er
e
ar
e
s
ev
er
al
alg
o
r
ith
m
s
th
at
d
o
th
is
p
r
o
ce
s
s
an
d
,
in
th
is
r
esear
ch
,
will
talk
ab
o
u
t
s
o
m
e
alg
o
r
ith
m
s
,
n
am
ely
:
C
HI
s
q
u
ar
e,
I
C
HI
s
q
u
ar
e,
I
n
f
o
r
m
atio
n
Gain
,
Mu
tu
al
I
n
f
o
r
m
atio
n
an
d
W
r
ap
p
er
.
W
h
er
e
th
ey
will
b
e
co
m
p
ar
ed
in
th
eir
u
s
e
with
th
e
class
if
icatio
n
p
r
o
ce
s
s
an
d
u
s
in
g
s
ev
er
al
alg
o
r
ith
m
s
class
if
i
ca
tio
n
:
B
ay
es
Net,
Naiv
e
B
ay
es,
R
an
d
o
m
Fo
r
est,
Dec
is
io
n
T
r
ee
an
d
Ar
tific
ial
Neu
r
al
Netwo
r
k
s
.
T
h
e
r
esu
lts
wil
l
b
e
co
m
p
ar
ed
b
ased
o
n
s
ev
er
al
cr
iter
ia:
Pre
cisi
o
n
,
R
ec
all,
F
-
m
ea
s
u
r
e
an
d
T
im
e
to
b
u
ild
m
o
d
el
.
T
h
ese
alg
o
r
ith
m
s
will
b
e
test
ed
u
s
in
g
t
h
e
p
r
e
-
p
r
o
ce
s
s
in
g
s
tep
,
T
o
k
e
n
izatio
n
,
No
r
m
aliza
tio
n
,
Sto
p
w
o
r
d
r
e
m
o
v
al
an
d
Stem
m
in
g
with
f
e
atu
r
e
s
elec
tio
n
s
tep
an
d
with
o
u
t
u
s
in
g
th
e
p
r
e
-
p
r
o
ce
s
s
in
g
s
tep
.
T
h
ese
p
r
o
ce
s
s
es
ar
e
u
s
ed
in
d
ata
m
in
i
n
g
ap
p
licatio
n
s
a
n
d
ex
am
p
les
o
f
im
p
o
r
tan
t
ap
p
lic
atio
n
s
in
d
ata
m
in
in
g
:
Natu
r
a
l
L
an
g
u
ag
e
Pro
c
ess
in
g
(
NL
P),
C
lass
if
icatio
n
an
d
I
n
f
o
r
m
atio
n
E
x
tr
ac
tio
n
[
9
,
1
0
]
.
Featu
r
e
s
elec
tio
n
p
r
o
b
lem
is
o
n
e
o
f
th
e
m
ain
im
p
o
r
tan
t
p
r
o
b
lem
s
in
th
e
tex
t
an
d
d
ata
m
in
in
g
d
o
m
ain
.
Featu
r
e
s
elec
tio
n
is
a
co
m
m
o
n
p
r
o
b
lem
u
s
ed
in
m
a
n
y
d
ata
m
in
in
g
ar
ea
s
;
it
is
u
s
ed
to
elim
in
ate
t
h
e
ir
r
elev
an
t
f
ea
tu
r
e
f
o
r
r
e
d
u
cin
g
th
e
n
u
m
b
er
o
f
s
elec
ted
f
ea
t
u
r
es
an
d
in
cr
ea
s
in
g
th
e
c
o
m
p
u
tatio
n
al
tim
e.
T
h
is
p
ap
er
p
r
esen
ts
a
co
m
p
a
r
ativ
e
s
tu
d
y
o
f
f
ea
tu
r
e
s
el
ec
tio
n
m
eth
o
d
s
f
o
r
Ar
ab
ic
tex
t
class
if
icatio
n
.
Fiv
e
o
f
th
e
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
s
wer
e
s
elec
ted
:
I
C
HI
s
q
u
ar
e,
C
H
I
s
q
u
ar
e,
I
n
f
o
r
m
atio
n
Gain
,
Mu
tu
al
I
n
f
o
r
m
atio
n
an
d
W
r
ap
p
er
.
I
t
was
test
ed
with
f
iv
e
class
if
icatio
n
alg
o
r
ith
m
s
:
B
ay
es
Net,
Naiv
e
B
ay
es,
R
an
d
o
m
Fo
r
est,
Dec
is
io
n
T
r
ee
an
d
Ar
tific
ial
Neu
r
al
Netwo
r
k
s
.
B
esid
es,
D
ata
C
o
llectio
n
was
u
s
ed
in
A
r
ab
ic
co
n
s
is
tin
g
o
f
9
0
5
5
d
o
cu
m
en
ts
,
wh
ich
wer
e
co
m
p
ar
e
d
b
y
f
o
u
r
cr
iter
ia:
Pre
cisi
o
n
,
R
ec
all,
F
-
m
ea
s
u
r
e
an
d
T
im
e
to
b
u
ild
m
o
d
el.
T
h
e
r
esu
lts
s
h
o
wed
t
h
at
th
e
im
p
r
o
v
e
d
I
C
HI
Sq
u
ar
e
Featu
r
e
Selectio
n
Me
th
o
d
f
o
r
Ar
ab
ic
C
lass
if
ier
s
g
o
t
b
etter
r
esu
lts
in
alm
o
s
t
a
ll
test
ca
s
es
in
th
is
p
ap
er
.
F
u
r
th
er
m
o
r
e,
t
h
e
im
p
r
o
v
e
d
I
C
HI
Sq
u
ar
e
Featu
r
e
Selectio
n
Me
th
o
d
f
o
r
Ar
a
b
ic
C
las
s
if
ier
s
is
co
m
p
ar
ed
to
o
th
er
well
-
k
n
o
wn
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
s
an
d
class
if
icatio
n
alg
o
r
ith
m
an
d
it
s
h
o
ws
its
s
u
p
er
io
r
ab
ilit
y
in
s
o
lv
in
g
th
e
f
ea
t
u
r
e
s
elec
tio
n
p
r
o
b
lem
s
.
T
h
e
r
est
o
f
th
is
p
ap
er
is
o
r
g
an
ized
as
Sectio
n
2
p
r
esen
ts
th
e
liter
atu
r
e
r
ev
iew
f
o
r
m
o
r
e
r
elat
ed
wo
r
k
s
,
t
h
e
tex
t
p
r
ep
r
o
ce
s
s
in
g
,
th
e
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
s
,
an
d
th
e
ev
al
u
atio
n
cr
iter
ia.
Sectio
n
3
p
r
e
s
en
ts
ex
p
er
im
en
ts
r
esu
lts
an
d
an
aly
s
is
.
Fin
ally
,
Sectio
n
4
p
r
esen
ts
co
n
clu
s
io
n
s
an
d
f
u
t
u
r
e
r
esear
ch
.
2.
L
I
T
E
R
AT
U
RE
R
E
VI
E
W
2
.
1
.
Da
t
a
c
o
llect
io
n
Data
co
llectio
n
h
as
b
ee
n
co
m
p
iled
f
r
o
m
s
ev
e
r
al
Ar
ab
n
ews
p
ap
er
s
an
d
m
a
g
az
in
es.
I
t
co
n
s
is
ts
o
f
9
0
5
5
d
o
cu
m
e
n
ts
class
if
ied
in
to
1
2
ca
teg
o
r
ies:
Ar
t
&
C
u
ltu
r
e,
E
co
n
o
m
y
,
E
d
u
ca
tio
n
,
E
n
g
in
e
er
in
g
,
L
aw,
L
o
ca
l
,
Me
d
icin
e,
Miscellan
eo
u
s
,
Po
li
tics
,
R
elig
io
n
,
Scien
ce
an
d
T
e
ch
n
o
lo
g
ical,
T
ab
le
1
s
h
o
ws
th
e
d
ata
co
llectio
n
in
d
etail.
E
ac
h
d
o
c
u
m
en
t c
o
n
s
is
ts
o
f
7
5
-
1
3
0
wo
r
d
s
o
r
m
o
r
e
[
1
0
]
.
T
ab
le
1
.
Ar
a
b
ic
d
ata
co
llectio
n
N
o
.
C
l
a
s
s Nam
e
N
o
.
o
f
D
o
c
u
me
n
t
1
A
r
t
& C
u
l
t
u
r
e
7
0
0
2
Ec
o
n
o
my
1
0
8
2
3
Ed
u
c
a
t
i
o
n
5
5
0
4
En
g
i
n
e
e
r
i
n
g
5
7
0
5
La
w
8
0
0
6
Lo
c
a
l
4
5
0
7
M
e
d
i
c
i
n
e
6
0
0
8
M
i
s
c
e
l
l
a
n
e
o
u
s
4
6
0
9
P
o
l
i
t
i
c
s
1
1
4
5
10
R
e
l
i
g
i
o
n
1
0
7
8
11
S
c
i
e
n
c
e
7
7
0
12
Te
c
h
n
o
l
o
g
i
c
a
l
8
5
0
To
t
a
l
9
0
5
5
2
.
2
.
P
re
-
pro
ce
s
s
ing
2
.
2
.
1
.
O
v
er
v
iew
Pre
-
p
r
o
ce
s
s
in
g
is
a
v
er
y
im
p
o
r
tan
t
s
tep
in
d
ata
m
in
in
g
o
p
er
atio
n
s
as
it
s
u
p
p
o
r
ts
th
e
r
esu
lts
an
d
in
cr
ea
s
es
th
eir
ef
f
icien
cy
s
o
it
is
v
er
y
im
p
o
r
tan
t
to
g
o
th
r
o
u
g
h
th
is
s
tep
b
ef
o
r
e
an
y
p
r
o
ce
s
s
.
I
n
th
is
r
esear
ch
,
we
will
u
s
e
f
o
u
r
p
r
o
ce
s
s
es
o
f
p
r
e
-
p
r
o
ce
s
s
in
g
t
o
im
p
r
o
v
e
th
e
class
if
icatio
n
p
r
o
ce
s
s
an
d
th
ese
p
r
o
ce
s
s
es
ar
e:
T
o
k
en
izatio
n
,
No
r
m
aliza
tio
n
,
Sto
p
wo
r
d
r
em
o
v
al
an
d
Stem
m
in
g
.
Fig
u
r
e
1
s
h
o
ws
th
e
s
tep
s
o
f
th
is
p
h
ase
f
r
o
m
d
ata
co
llectio
n
to
class
if
icatio
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
I
SS
N:
2252
-
8
7
7
6
I
mp
r
o
ve
d
I
C
HI
s
q
u
a
r
e
fea
tu
r
e
s
elec
tio
n
meth
o
d
fo
r
A
r
a
b
ic
c
la
s
s
ifie
r
s
(
Ha
d
ee
l N
)
159
Fig
u
r
e
1
.
Pre
-
p
r
o
ce
s
s
in
g
life
c
y
cle
2
.
2
.
2
.
T
o
k
eniza
t
io
n
At
th
is
p
h
ase,
th
e
s
en
ten
ce
s
ar
e
cu
t
in
to
a
s
er
ies
o
f
wo
r
d
s
,
d
e
p
en
d
in
g
o
n
th
e
wh
ite
s
p
ac
e
b
e
twee
n
th
e
wo
r
d
s
,
to
f
ac
ilit
ate
th
e
co
m
p
ilatio
n
o
f
wo
r
d
s
b
elo
n
g
in
g
to
th
e
s
am
e
ca
teg
o
r
y
.
T
h
ey
ar
e
s
eg
m
en
ted
in
to
a
s
er
ies
o
f
k
ey
wo
r
d
s
,
p
h
r
ases
,
etc.
,
w
h
e
r
e
s
o
m
e
c
h
ar
ac
t
er
s
an
d
s
y
m
b
o
ls
s
u
ch
as
p
u
n
ctu
atio
n
in
t
h
e
T
o
k
en
izatio
n
p
r
o
ce
s
s
ar
e
ig
n
o
r
ed
.
T
h
is
s
er
ies
b
ec
o
m
es
in
p
u
t
in
to
an
o
th
er
p
r
o
ce
s
s
s
u
ch
as
d
ata
an
aly
s
is
an
d
d
ata
m
in
in
g
.
T
o
k
e
n
izatio
n
is
a
ls
o
a
lan
g
u
ag
e
t
y
p
e
an
d
th
is
p
r
o
ce
s
s
is
v
er
y
im
p
o
r
tan
t i
n
d
ata
m
in
in
g
to
o
ls
[
1
1
]
.
2
.
2
.
3
.
No
rma
liza
t
io
n
T
h
is
p
h
ase
co
n
v
er
ts
th
e
s
et
o
f
wo
r
d
s
an
d
s
en
ten
ce
s
in
to
m
o
r
e
co
m
p
lex
an
d
m
o
r
e
p
r
ec
is
e
s
en
ten
ce
s
,
s
o
th
at
th
e
p
r
o
ce
s
s
o
f
p
r
o
ce
s
s
in
g
is
ea
s
y
an
d
d
ata
h
an
d
lin
g
e
asier
.
No
r
m
aliza
tio
n
im
p
r
o
v
es
tex
t
m
atch
in
g
an
d
im
p
r
o
v
es
r
etr
iev
al
b
y
tak
in
g
s
y
n
o
n
y
m
s
f
o
r
m
ea
n
in
g
f
u
l
wo
r
d
s
to
p
r
o
d
u
ce
th
e
b
est
r
esu
lts
an
d
h
ig
h
er
ef
f
icien
cy
.
Fo
r
ex
am
p
le,
in
th
i
s
p
r
o
ce
s
s
th
e
u
n
-
d
o
tted
Ar
ab
ic
letter
(
ى
)
r
ep
lace
to
th
e
d
o
tted
letter
(
ي
)
[
1
2
]
.
2
.
2
.
4
.
Sto
pwo
rd
re
m
o
v
a
l
T
h
is
p
h
ase
r
em
o
v
es
th
e
s
to
p
wo
r
d
f
r
o
m
th
e
d
o
cu
m
en
t
wh
er
e
th
e
s
to
p
wo
r
d
h
as
n
o
m
e
an
in
g
lik
e
p
r
o
n
o
u
n
s
,
v
o
wels
an
d
th
e
lik
e
.
T
h
at
is
,
at
th
is
p
o
in
t
s
to
p
wo
r
d
r
em
o
v
al,
wo
r
d
s
an
d
s
y
m
b
o
l
s
th
at
d
o
n
o
t
af
f
ec
t
th
e
m
ea
n
in
g
ar
e
r
em
o
v
e
d
an
d
th
u
s
im
p
r
o
v
e
th
e
r
esu
lts
o
f
th
e
r
etr
iev
al
an
d
class
if
icatio
n
p
r
o
ce
s
s
es.
E
x
am
p
les
ar
e
ico
n
s
(
@
,
#
,
!
,
$
,
%,
&
,
*
,
^,
+,
-
,
.
.
.
.
)
.
T
h
e
p
r
ep
o
s
itio
n
(
ن
ع
،
يف
،
ىلع
....
)
,
an
d
t
h
e
ad
v
e
r
b
s
(
تح
ت
،
ق
وف
....
)
[
1
3
]
.
2
.
2
.
5
.
Ste
mm
ing
T
h
e
s
tem
m
in
g
p
r
o
ce
s
s
is
im
p
o
r
tan
t
to
f
in
d
th
e
r
o
o
t
o
f
th
e
wo
r
d
,
t
o
r
e
d
u
ce
th
e
s
ize
o
f
th
e
d
o
c
u
m
en
t
b
y
r
em
o
v
in
g
th
e
Pre
f
ix
a
n
d
S
u
f
f
ix
es
in
ad
d
itio
n
to
r
em
o
v
i
n
g
th
e
in
cr
ea
s
e
ch
a
r
ac
ter
s
s
u
ch
as
(
ى
،
و
،
أ
)
.
T
h
e
p
r
o
ce
s
s
o
f
s
tem
m
in
g
is
b
y
p
r
e
-
p
r
o
ce
s
s
in
g
th
e
in
f
o
r
m
atio
n
r
etr
iev
al
an
d
also
f
ac
ilit
ates
s
ea
r
ch
es
in
s
ea
r
ch
en
g
in
es th
u
s
in
cr
ea
s
in
g
th
e
r
es
u
lts
[
1
4
]
.
2
.
3
.
F
e
a
t
ure
s
elec
t
io
n m
et
ho
ds
2
.
3
.
1
.
O
v
er
v
iew
T
h
e
f
ea
tu
r
e
s
elec
tio
n
p
r
o
ce
s
s
is
im
p
o
r
tan
t
t
o
r
e
d
u
ce
t
h
e
am
o
u
n
t
o
f
u
n
wan
ted
d
ata
an
d
to
m
ain
tain
th
e
m
o
s
t
im
p
o
r
tan
t
d
ata
.
T
h
e
f
ea
tu
r
es
ar
e
d
ef
in
e
d
ac
co
r
d
in
g
to
p
r
e
d
ef
in
ed
c
r
iter
ia
an
d
a
r
e
a
p
r
e
-
p
r
o
ce
s
s
in
g
s
tag
e
an
d
ar
e
im
p
o
r
tan
t in
im
p
r
o
v
in
g
th
e
class
if
icatio
n
p
r
o
ce
s
s
.
F
ig
u
r
e
2
s
h
o
ws th
e
s
tag
es
i
n
th
e
class
if
icatio
n
p
r
o
ce
s
s
u
s
in
g
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
s
.
Fig
u
r
e
1
.
Featu
r
e
s
elec
tio
n
life
cy
cle
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
7
6
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
,
Vo
l.
9
,
No
.
3,
Dec
em
b
e
r
20
20
:
15
7
–
1
70
160
2
.
3
.
2
.
CH
I
s
qu
a
re
I
s
a
s
tati
s
tical
m
eth
o
d
b
y
s
elec
tin
g
r
an
d
o
m
d
ata
an
d
d
e
p
en
d
s
o
n
two
in
d
ep
e
n
d
en
t
v
a
r
iab
les
o
r
v
ar
iab
les
f
r
o
m
th
e
d
ata
s
am
p
l
e
an
d
is
co
n
s
id
er
ed
as
a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
C
HI
s
q
u
ar
e
is
an
im
p
o
r
tan
t
p
r
e
-
p
r
o
ce
s
s
in
g
in
th
e
class
if
icatio
n
p
r
o
ce
s
s
an
d
C
HI
s
q
u
ar
e
is
u
s
ed
in
th
e
class
if
icatio
n
s
y
s
tem
[
1
5
]
.
2
.
3
.
3
.
I
m
pro
v
ed
CH
I
s
qu
a
re
(
I
CH
I
)
I
C
HI
is
a
d
ev
elo
p
m
en
t
o
f
th
e
C
HI
s
q
u
ar
e
m
eth
o
d
a
n
d
is
d
o
n
e
b
y
im
p
r
o
v
in
g
th
e
b
asic
p
r
o
p
e
r
ties
wh
er
e
I
C
HI
s
q
u
ar
e
is
an
ex
ten
s
io
n
o
f
th
e
o
r
ig
in
al
C
HI
s
q
u
ar
e
m
eth
o
d
.
I
C
HI
s
q
u
ar
e
h
as
b
ee
n
u
s
ed
with
C
h
in
ese
lan
g
u
ag
e.
T
h
e
r
esu
lts
h
av
e
p
r
o
v
ed
ef
f
ec
tiv
e
in
u
s
in
g
I
C
HI
s
q
u
ar
e
with
C
h
in
ese
lan
g
u
ag
e.
[
1
6
]
I
n
ad
d
itio
n
,
I
C
HI
s
q
u
ar
e
was
u
s
ed
in
Ar
ab
ic
u
s
in
g
s
ev
er
al
alg
o
r
ith
m
s
:
B
ay
es
Net,
Naïv
e
B
ay
es,
Naïv
e
B
ay
e
s
Mu
ltin
o
m
ial,
R
an
d
o
m
Fo
r
est,
Dec
is
io
n
T
r
ee
an
d
Ar
tific
ial
Neu
r
al
Netwo
r
k
s
.
T
h
e
r
esu
lts
p
r
o
v
e
d
ef
f
ec
tiv
e
in
u
s
in
g
Ar
ab
ic
[
1
7
]
.
2
.
3
.
4
.
I
nfo
rm
a
t
io
n
g
a
in
(
I
G
)
I
n
f
o
r
m
atio
n
Gain
is
b
ased
o
n
th
e
m
ea
s
u
r
e
m
en
t
o
f
th
e
n
u
m
b
er
o
f
b
its
g
ain
ed
b
y
th
e
i
n
f
o
r
m
atio
n
o
b
tain
ed
b
y
p
r
ed
ictin
g
th
e
ca
t
eg
o
r
y
b
y
th
e
p
r
esen
ce
o
f
th
e
w
o
r
d
o
r
n
o
t i
n
th
e
d
o
cu
m
e
n
t [
1
8
]
.
2
.
3
.
5
.
M
utua
l
info
rm
a
t
io
n
(
M
I
)
Mu
tu
al
in
f
o
r
m
atio
n
is
b
ased
o
n
th
e
f
ac
t
th
at
in
f
o
r
m
atio
n
ex
ch
an
g
e
d
is
a
m
ea
s
u
r
e
o
f
d
ep
en
d
en
ce
b
etwe
en
v
ar
iab
les,
s
in
ce
th
e
in
f
o
r
m
atio
n
ex
ch
a
n
g
ed
is
a
m
ea
s
u
r
e
o
f
in
f
o
r
m
atio
n
th
at
is
(
p
o
s
s
ib
ly
m
u
lti
-
d
im
en
s
io
n
al)
.
W
h
er
e
th
e
a
m
o
u
n
t
o
f
in
f
o
r
m
atio
n
o
b
tai
n
ed
is
d
eter
m
in
ed
b
y
o
n
e
o
r
m
o
r
e
r
an
d
o
m
v
ar
iab
les [
1
9
]
.
2
.
3
.
6
.
Wra
pp
er
W
r
ap
p
er
Alg
o
r
ith
m
is
b
ased
o
n
th
e
u
s
e
o
f
lear
n
in
g
alg
o
r
i
th
m
s
wh
er
e
it
s
elec
t
s
r
elev
an
t
f
ea
tu
r
es
b
ased
o
n
th
e
p
er
f
o
r
m
an
ce
o
f
lear
n
in
g
alg
o
r
ith
m
s
.
I
n
a
d
d
iti
o
n
to
th
at
th
e
p
r
o
ce
s
s
o
f
tr
ain
in
g
th
e
m
o
d
el
with
d
if
f
er
en
t
f
ea
tu
r
es,
o
r
s
elec
tin
g
th
e
f
ea
tu
r
es
th
at
lead
to
th
e
b
est
r
esu
lts
o
u
t
o
f
th
e
m
o
d
el
an
d
b
e
an
ex
am
p
le
o
f
a
co
m
p
lex
m
o
d
el
o
f
s
ev
er
al
f
e
atu
r
es r
elate
d
to
ea
ch
o
th
er
[
2
0
]
.
2
.
4
.
Ara
bic
t
e
x
t
cla
s
s
if
iers
2
.
4
.
1
.
O
v
er
v
iew
I
n
o
r
d
er
to
ca
r
r
y
o
u
t
th
e
clas
s
if
icatio
n
p
r
o
ce
s
s
,
we
n
ee
d
al
g
o
r
ith
m
s
th
at
p
er
f
o
r
m
th
e
class
if
icatio
n
p
r
o
ce
s
s
th
r
o
u
g
h
th
e
class
if
icatio
n
p
r
o
ce
s
s
th
r
o
u
g
h
th
e
tr
ain
in
g
p
r
o
ce
s
s
b
ased
o
n
a
d
ata
s
et
c
ateg
o
r
ized
in
to
th
e
test
in
g
p
r
o
ce
s
s
o
f
a
n
ew
tex
t
an
d
d
is
tin
g
u
is
h
it
in
to
wh
ich
ca
teg
o
r
y
b
elo
n
g
s
.
As
th
e
clas
s
if
icatio
n
p
r
o
ce
s
s
i
s
s
u
p
er
v
is
ed
lear
n
in
g
an
d
is
s
et
to
in
p
u
t.
Fig
u
r
e
3
s
h
o
ws T
ex
t
C
las
s
if
icatio
n
s
y
s
tem
.
Fig
u
r
e
2
.
C
lass
if
icatio
n
s
y
s
te
m
2
.
4
.
2
.
B
a
y
es
net
(
B
N)
B
ay
es
Net
clas
s
if
ier
is
a
m
o
d
el
th
at
r
ef
lects
s
itu
atio
n
s
th
at
ar
e
p
ar
t
o
f
th
e
B
ay
es
Net
class
if
ier
an
d
h
o
w
th
ey
r
elate
t
o
ea
ch
o
t
h
er
.
T
h
is
m
o
d
el
m
ay
b
e
r
elate
d
to
an
y
en
tity
in
t
h
is
wo
r
ld
an
d
ca
n
b
e
r
e
p
r
esen
ted
b
y
B
N
cla
s
s
if
ier
an
d
all
en
titi
e
s
o
cc
u
r
f
r
eq
u
en
tly
wh
en
an
o
t
h
er
en
tity
ex
is
ts
.
T
h
is
m
o
d
el
is
v
er
y
im
p
o
r
tan
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
I
SS
N:
2252
-
8
7
7
6
I
mp
r
o
ve
d
I
C
HI
s
q
u
a
r
e
fea
tu
r
e
s
elec
tio
n
meth
o
d
fo
r
A
r
a
b
ic
c
la
s
s
ifie
r
s
(
Ha
d
ee
l N
)
161
b
ec
au
s
e
it
h
elp
s
u
s
to
p
r
ed
ic
t
th
e
r
esu
lts
o
f
th
e
en
tity
in
th
is
wo
r
ld
an
d
it
is
ea
s
y
to
r
ep
r
esen
t
th
e
en
tity
th
r
o
u
g
h
th
is
m
o
d
el
[
2
1
]
.
2
.
4
.
3
.
Na
ïv
e
ba
y
es
(
NB
)
Naiv
e
B
ay
es
clas
s
if
ier
is
b
ase
d
o
n
B
ay
es
T
h
e
o
r
e
m
,
wh
er
e
NB
is
ea
s
y
to
b
u
ild
an
d
v
e
r
y
u
s
ef
u
l
with
lar
g
e
d
ata.
Desp
ite
th
e
ea
s
e
o
f
b
u
ild
in
g
th
e
NB
m
o
d
el
it
o
u
tp
er
f
o
r
m
s
m
an
y
o
f
th
e
wo
r
k
b
o
o
k
s
in
th
e
class
if
icatio
n
p
r
o
ce
s
s
[
1
1
]
.
I
n
th
e
NB
an
aly
s
is
p
r
o
ce
s
s
,
th
e
f
in
al
class
if
icatio
n
is
p
r
o
d
u
ce
d
b
y
c
o
m
b
in
in
g
all
s
o
u
r
ce
s
o
f
in
f
o
r
m
atio
n
.
T
h
e
m
o
d
el
is
cr
ea
ted
b
ased
o
n
tr
ain
in
g
d
ata
to
class
if
y
th
e
d
o
cu
m
en
t
in
to
a
n
y
ca
teg
o
r
y
[
2
1
]
.
2
.
4
.
4
.
Ra
nd
o
m
f
o
re
s
t
(
RF
)
R
an
d
o
m
Fo
r
est
C
lass
if
ier
is
a
s
et
o
f
class
if
icat
io
n
alg
o
r
ith
m
s
th
at
ar
e
wid
ely
u
s
ed
in
m
an
y
ap
p
licatio
n
s
,
esp
ec
ially
with
th
e
lar
g
e
d
ata
s
et
d
u
e
to
its
ch
ar
ac
ter
is
tic
p
r
o
p
er
ties
.
T
h
e
R
an
d
o
m
Fo
r
est
alg
o
r
ith
m
is
u
s
ed
in
m
an
y
ap
p
licatio
n
s
,
in
clu
d
in
g
:
Netwo
r
k
in
tr
u
s
io
n
d
etec
tio
n
,
E
m
ail
s
p
am
d
etec
tio
n
,
g
e
n
e
class
if
icatio
n
,
C
r
ed
it c
ar
d
f
r
au
d
d
etec
tio
n
,
a
n
d
T
e
x
t c
las
s
if
icatio
n
[
2
2
]
.
2
.
4
.
5
.
Dec
is
io
n
t
re
e
(
DT
)
Dec
is
io
n
T
r
ee
is
an
alg
o
r
ith
m
th
at
co
m
p
iles
an
d
is
also
u
s
ed
in
th
e
f
ield
o
f
e
x
tr
ac
tin
g
d
ata
in
v
ar
io
u
s
f
ield
s
o
n
th
e
I
n
ter
n
et.
Als
o
u
s
ed
in
d
ata
m
in
in
g
alg
o
r
ith
m
s
,
in
ad
d
itio
n
,
th
e
Dec
is
io
n
T
r
ee
alg
o
r
ith
m
is
u
s
e
d
in
Ar
ab
ic
T
ex
t
C
lass
if
icatio
n
[
2
3
]
.
T
h
e
Dec
is
io
n
T
r
ee
with
t
h
e
C
HI
s
q
u
ar
e
as
a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
was
u
s
ed
with
d
ata
co
llectio
n
in
Ar
ab
ic
lan
g
u
ag
e,
an
d
Dec
is
io
n
T
r
ee
was
u
s
ed
with
th
e
I
C
HI
s
q
u
ar
e
as
a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
R
esu
lts
s
h
o
wed
th
at
t
h
e
I
C
HI
s
q
u
ar
e
s
u
r
p
ass
ed
th
e
r
eg
u
lar
C
HI
s
q
u
ar
e
[
2
4
]
.
2
.
4
.
6
.
Art
if
icia
l
neura
l net
wo
rk
s
(
ANNs)
Ar
tific
ial
Neu
r
al
Netwo
r
k
s
i
s
an
alg
o
r
ith
m
th
at
is
u
s
ed
f
o
r
class
if
icatio
n
.
I
n
ad
d
itio
n
,
its
b
r
an
ch
o
f
ar
tific
ial
in
tellig
en
ce
,
wh
er
e
d
if
f
er
e
n
t
s
ets
o
f
f
u
n
ctio
n
s
h
av
e
b
ee
n
s
tu
d
ied
f
r
o
m
tr
ai
n
in
g
an
d
lear
n
in
g
to
th
e
test
in
g
p
r
o
ce
s
s
.
[
2
5
]
ANNs
h
as
b
ee
n
s
u
cc
ess
f
u
lly
ap
p
lied
to
p
r
o
b
lem
s
in
:
p
atter
n
class
if
icatio
n
,
f
u
n
ctio
n
ap
p
r
o
x
im
atio
n
,
o
p
tim
izatio
n
,
an
d
p
atter
n
m
atch
in
g
[
2
6
]
.
2
.
5
.
E
v
a
lua
t
i
o
n
m
ea
s
ure
2
.
5
.
1
.
O
v
er
v
iew
Af
ter
th
e
class
if
icatio
n
p
r
o
ce
s
s
s
h
o
ws
u
s
m
an
y
r
esu
lts
an
d
u
n
til
th
e
r
esu
lts
ar
e
r
ea
d
an
d
co
m
p
ar
ed
,
th
er
e
m
u
s
t
b
e
cr
iter
ia
f
o
r
th
is
co
m
p
ar
is
o
n
.
T
h
ese
r
esu
lts
ar
e
co
m
p
ar
e
d
to
f
o
u
r
cr
iter
ia:
P
r
ec
is
io
n
,
R
ec
all,
F
-
m
ea
s
u
r
e
an
d
T
im
e
to
b
u
ild
m
o
d
el.
2
.
5
.
2
.
P
re
cisi
o
n
Pre
cisi
o
n
is
a
p
o
s
itiv
e
p
r
ed
ictiv
e
v
alu
e
wh
ich
is
a
p
ar
t
o
f
t
h
e
ca
s
es
r
etr
iev
ed
r
elev
an
t
f
r
o
m
all
ca
s
es
r
etr
iev
ed
f
r
o
m
th
e
r
esu
lts
o
f
th
e
p
r
o
ce
s
s
[
2
6
,
27
,
2
8
].
W
h
er
e
p
r
ec
is
io
n
is
r
ep
r
esen
ted
b
y
th
e
f
o
llo
win
g
eq
u
atio
n
:
(1
)
2
.
5
.
3
.
Rec
a
ll
T
h
e
r
ec
all
is
d
escr
ib
ed
as
th
e
s
en
s
itiv
ity
m
ea
s
u
r
e,
wh
ich
is
t
h
e
p
ar
t
o
f
th
e
r
elev
a
n
t
ca
s
es
o
f
th
e
to
tal
n
u
m
b
er
o
f
r
ec
o
v
e
r
ed
ca
s
es.
(2
)
2
.
5
.
4
.
F
-
m
ea
s
ure
T
h
e
F
-
m
ea
s
u
r
e
is
a
m
ea
s
u
r
e
o
f
th
e
ac
cu
r
ac
y
o
f
th
e
class
if
ier
b
ein
g
test
ed
,
wh
ich
is
a
p
r
ec
is
io
n
an
d
r
ec
all
r
atio
.
(3
)
2
.
5
.
5
.
T
im
e
I
t
is
th
e
tim
e
it
tak
es
to
b
u
ild
th
e
d
ata
m
o
d
el,
wh
ich
is
also
t
h
e
tim
e
u
s
ed
in
th
e
p
r
o
ce
s
s
o
f
an
aly
zin
g
th
e
d
ata
an
d
ca
lcu
latin
g
t
h
e
p
r
o
ce
s
s
es p
er
f
o
r
m
ed
b
y
th
e
m
o
d
el.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
7
6
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
,
Vo
l.
9
,
No
.
3,
Dec
em
b
e
r
20
20
:
15
7
–
1
70
162
3.
E
XP
E
R
I
M
E
N
T
A
L
RE
SUL
T
S AN
D
A
NALY
SI
S
3
.
1
.
Resul
t
s
wit
ho
ut
na
t
ura
l la
ng
ua
g
e
pro
ce
s
s
ing
3
.
1
.
1
.
B
a
y
es
net
(
B
N)
T
ab
le
2
s
h
o
ws
b
ay
es
n
et
cla
s
s
if
ier
r
esu
lts
with
o
u
t
p
r
e
-
p
r
o
ce
s
s
in
g
u
s
in
g
s
ev
er
al
f
ea
tu
r
e
s
elec
tio
n
alg
o
r
ith
m
s
.
T
h
e
r
esu
lts
s
h
o
wed
th
at
in
th
e
ca
s
e
o
f
Pre
cisi
o
n
was
th
e
h
ig
h
est
r
esu
lt
o
f
class
if
ier
wh
en
u
s
ed
with
I
C
HI
s
q
u
ar
e
as
a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
Mo
r
eo
v
er
,
th
at
in
th
e
ca
s
e
o
f
R
ec
all
was
th
e
h
ig
h
est
r
esu
lts
o
f
th
e
class
if
ier
wh
en
u
s
ed
wi
th
I
C
HI
s
q
u
ar
e
as
a
f
ea
t
u
r
e
s
e
lectio
n
m
eth
o
d
,
b
u
t
in
th
e
ca
s
e
o
f
F
-
m
ea
s
u
r
e
was
th
e
h
ig
h
est
r
esu
lts
o
f
th
e
class
if
ier
wh
en
u
s
ed
with
I
C
HI
s
q
u
ar
e
as
a
f
ea
tu
r
e
s
elec
tio
n
m
et
h
o
d
an
d
in
th
e
ca
s
e
o
f
T
im
e
was
th
e
less
tim
e
to
b
u
ild
m
o
d
el
o
f
th
e
class
if
ier
wh
en
u
s
ed
with
I
n
f
o
r
m
atio
n
Gain
as
a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
Fig
u
r
e
4
s
h
o
ws d
iag
r
am
B
ay
es
Net
r
esu
lts
with
o
u
t p
r
e
-
p
r
o
ce
s
s
in
g
.
T
ab
le
2
.
B
ay
es
n
et
r
esu
lts
with
o
u
t p
r
e
-
p
r
o
ce
s
s
in
g
A
l
g
o
r
i
t
h
m
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F
-
mea
su
r
e
Ti
me
I
C
H
I
0
.
9
7
6
0
.
9
4
0
0
.
9
5
7
2
5
.
6
C
H
I
0
.
8
5
8
0
.
9
1
0
0
.
8
8
3
1
8
.
2
0
IG
0
.
8
5
1
0
.
8
9
9
0
.
8
7
4
1
3
.
3
MI
0
.
8
1
9
0
.
8
7
3
0
.
8
4
5
2
2
.
5
W
r
a
p
p
e
r
0
.
8
3
2
0
.
8
8
8
0
.
8
5
9
2
0
.
7
A
v
e
r
a
g
e
0
.
8
6
7
2
0
.
9
0
2
0
.
8
8
3
2
0
.
0
6
W
i
t
h
o
u
t
F
S
0
.
8
0
5
0
.
8
7
1
0
.
8
3
6
1
6
.
7
Fig
u
r
e
4
.
B
ay
es
n
et
r
esu
lts
with
o
u
t p
r
e
-
p
r
o
ce
s
s
in
g
3
.
1
.
2
.
Na
ïv
e
ba
y
es
(
NB
)
T
ab
le
3
s
h
o
ws
n
aïv
e
b
a
y
es
class
if
ier
r
esu
lts
with
o
u
t
p
r
e
-
p
r
o
ce
s
s
in
g
u
s
in
g
s
ev
er
al
f
ea
tu
r
e
s
elec
tio
n
alg
o
r
ith
m
s
.
T
h
e
r
esu
lts
s
h
o
wed
th
at
in
th
e
ca
s
e
o
f
Pre
cisi
o
n
was
th
e
h
ig
h
est
r
esu
lt
o
f
class
if
ier
wh
en
u
s
ed
with
I
C
HI
s
q
u
ar
e
as
a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
Mo
r
eo
v
er
,
th
at
in
th
e
ca
s
e
o
f
R
ec
all
was
th
e
h
ig
h
est
r
esu
lts
o
f
th
e
class
if
ier
wh
en
u
s
ed
wi
th
I
C
HI
s
q
u
ar
e
as
a
f
ea
t
u
r
e
s
e
lectio
n
m
eth
o
d
,
b
u
t
in
th
e
ca
s
e
o
f
F
-
m
ea
s
u
r
e
was
th
e
h
ig
h
est
r
esu
lts
o
f
th
e
class
if
ier
wh
en
u
s
ed
with
I
C
HI
s
q
u
ar
e
as
a
f
ea
tu
r
e
s
elec
tio
n
m
et
h
o
d
an
d
in
th
e
ca
s
e
o
f
T
im
e
was
th
e
less
tim
e
to
b
u
ild
m
o
d
el
o
f
th
e
class
if
ier
wh
en
u
s
ed
with
W
ith
o
u
t
FS
a
s
a
f
ea
tu
r
e
elec
tio
n
m
eth
o
d
.
Fig
u
r
e
5
s
h
o
ws d
iag
r
am
Naïv
e
B
ay
es r
esu
lts
with
o
u
t p
r
e
-
p
r
o
ce
s
s
in
g
.
T
ab
le
3
.
Naiv
e
b
ay
es r
esu
lts
with
o
u
t p
r
e
-
p
r
o
ce
s
s
in
g
A
l
g
o
r
i
t
h
m
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F
-
mea
su
r
e
Ti
me
I
C
H
I
0
.
9
7
6
0
.
9
5
4
0
.
9
6
4
1
8
.
2
0
C
H
I
0
.
9
2
6
0
.
9
5
0
0
.
9
3
7
4
.
1
1
IG
0
.
9
2
2
0
.
9
5
2
0
.
9
3
8
9
.
5
MI
0
.
9
1
3
0
.
9
4
5
0
.
9
2
8
8
.
4
W
r
a
p
p
e
r
0
.
9
1
9
0
.
9
5
0
0
.
9
3
4
9
.
2
A
v
e
r
a
g
e
0
.
9
3
1
0
.
9
5
0
0
.
9
4
0
9
.
8
8
W
i
t
h
o
u
t
F
S
0
.
9
1
3
0
.
9
5
1
0
.
9
3
1
3
.
1
6
Fig
u
r
e
5.
Naiv
e
b
ay
es r
esu
lts
with
o
u
t
p
r
e
-
p
r
o
ce
s
s
in
g
3
.
1
.
3
.
Ra
nd
o
m
f
o
re
s
t
(
RF
)
T
ab
le
4
s
h
o
ws
r
an
d
o
m
f
o
r
est
class
if
ier
r
esu
lts
with
o
u
t
p
r
e
-
p
r
o
ce
s
s
in
g
u
s
in
g
s
ev
er
al
f
ea
t
u
r
e
s
elec
tio
n
alg
o
r
ith
m
s
.
T
h
e
r
esu
lts
s
h
o
wed
th
at
in
th
e
ca
s
e
o
f
Pre
cisi
o
n
was
th
e
h
ig
h
est
r
esu
lt
o
f
class
if
ier
wh
en
u
s
ed
with
I
C
HI
s
q
u
ar
e
as
a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
Mo
r
eo
v
er
,
th
at
in
th
e
ca
s
e
o
f
R
ec
all
was
th
e
h
ig
h
est
r
esu
lts
o
f
th
e
class
if
ier
wh
en
u
s
ed
wi
th
I
C
HI
s
q
u
ar
e
as
a
f
ea
t
u
r
e
s
e
lectio
n
m
eth
o
d
,
b
u
t
in
th
e
ca
s
e
o
f
F
-
m
ea
s
u
r
e
was
th
e
h
ig
h
est
r
esu
lts
o
f
th
e
class
i
f
ier
wh
en
u
s
ed
with
I
C
HI
s
q
u
ar
e
as
a
f
ea
tu
r
e
s
elec
tio
n
m
et
h
o
d
an
d
in
th
e
ca
s
e
o
f
T
im
e
was
th
e
less
tim
e
to
b
u
ild
m
o
d
el
o
f
th
e
class
if
ier
wh
en
u
s
ed
with
W
r
ap
p
er
as
a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
Fig
u
r
e
6
s
h
o
ws d
iag
r
am
R
an
d
o
m
Fo
r
est r
esu
lts
with
o
u
t p
r
e
-
p
r
o
ce
s
s
i
ng
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
I
SS
N:
2252
-
8
7
7
6
I
mp
r
o
ve
d
I
C
HI
s
q
u
a
r
e
fea
tu
r
e
s
elec
tio
n
meth
o
d
fo
r
A
r
a
b
ic
c
la
s
s
ifie
r
s
(
Ha
d
ee
l N
)
163
T
ab
le
4
.
R
an
d
o
m
f
o
r
est r
esu
lts
with
o
u
t p
r
e
-
p
r
o
ce
s
s
in
g
A
l
g
o
r
i
t
h
m
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F
-
mea
su
r
e
Ti
me
I
C
H
I
0
.
9
4
7
0
.
9
4
2
0
.
9
4
4
9
4
.
3
C
H
I
0
.
9
3
8
0
.
8
0
5
0
.
8
6
6
6
6
.
5
IG
0
.
9
3
1
0
.
8
9
9
0
.
9
1
4
6
9
.
1
MI
0
.
9
1
7
0
.
8
4
9
0
.
8
8
1
7
0
.
3
W
r
a
p
p
e
r
0
.
9
2
9
0
.
8
5
1
0
.
8
8
8
6
6
.
4
A
v
e
r
a
g
e
0
.
9
3
2
0
.
8
6
9
0
.
8
9
9
7
3
.
3
W
i
t
h
o
u
t
F
S
0
.
9
1
5
0
.
8
4
4
0
.
8
7
8
7
1
.
4
8
Fig
u
r
e
6
.
R
an
d
o
m
f
o
r
est r
esu
lts
with
o
u
t p
r
e
-
p
r
o
ce
s
s
in
g
3
.
1
.
4
.
Dec
is
io
n
t
re
e
(
DT
)
T
ab
le
5
s
h
o
ws
d
ec
is
io
n
tr
ee
c
lass
if
ier
r
esu
lts
with
o
u
t
p
r
e
-
p
r
o
ce
s
s
in
g
u
s
in
g
s
ev
er
al
f
ea
tu
r
e
s
elec
tio
n
alg
o
r
ith
m
s
.
T
h
e
r
esu
lts
s
h
o
wed
th
at
in
th
e
ca
s
e
o
f
Pre
cisi
o
n
was
th
e
h
ig
h
est
r
esu
lt
o
f
class
if
ier
wh
en
u
s
ed
with
I
C
HI
s
q
u
ar
e
as
a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
Mo
r
eo
v
er
,
th
at
in
th
e
ca
s
e
o
f
R
ec
all
was
th
e
h
ig
h
est
r
esu
lts
o
f
th
e
class
if
ier
wh
en
u
s
ed
w
ith
C
HI
s
q
u
ar
e
as
a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
,
b
u
t
in
th
e
ca
s
e
o
f
F
-
m
ea
s
u
r
e
was
th
e
h
ig
h
est
r
esu
lts
o
f
th
e
class
if
ier
wh
en
u
s
ed
with
I
C
HI
s
q
u
ar
e
as
a
f
ea
tu
r
e
s
elec
tio
n
m
et
h
o
d
an
d
in
th
e
ca
s
e
o
f
T
im
e
was
th
e
less
tim
e
to
b
u
ild
m
o
d
el
o
f
th
e
class
if
ier
wh
en
u
s
ed
with
W
r
ap
p
er
as
a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
Fig
u
r
e
7
s
h
o
ws d
iag
r
am
Dec
is
io
n
T
r
ee
r
esu
lts
with
o
u
t p
r
e
-
p
r
o
ce
s
s
in
g
.
T
ab
le
5
.
Dec
is
io
n
tr
ee
r
esu
lts
with
o
u
t p
r
e
-
p
r
o
ce
s
s
in
g
A
l
g
o
r
i
t
h
m
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F
-
mea
su
r
e
Ti
me
I
C
H
I
0
.
6
0
4
0
.
5
1
8
0
.
5
5
7
4
.
4
0
C
H
I
0
.
5
5
2
0
.
5
3
7
0
.
5
4
4
2
.
5
1
IG
0
.
5
4
7
0
.
5
3
3
0
.
5
3
9
3
.
7
MI
0
.
5
2
9
0
.
5
0
0
0
.
5
1
4
8
.
9
W
r
a
p
p
e
r
0
.
5
3
9
0
.
5
3
0
0
.
5
3
4
6
.
3
A
v
e
r
a
g
e
0
.
5
5
4
0
.
5
2
3
0
.
5
3
8
5
.
1
6
W
i
t
h
o
u
t
F
S
0
.
5
2
8
0
.
5
2
3
0
.
5
2
5
1
.
0
1
Fig
u
r
e
7
.
Dec
is
io
n
tr
ee
r
esu
lts
with
o
u
t p
r
e
-
p
r
o
ce
s
s
in
g
3
.
1
.
5
.
Art
if
icia
l
neura
l ne
t
wo
rk
s
(
ANNs)
T
ab
le
6
s
h
o
ws
ar
tific
ial
n
eu
r
al
n
etwo
r
k
s
class
if
ier
r
esu
lts
with
o
u
t
p
r
e
-
p
r
o
ce
s
s
in
g
u
s
in
g
s
ev
er
a
l
f
ea
tu
r
e
s
elec
tio
n
alg
o
r
it
h
m
s
.
T
h
e
r
esu
lts
s
h
o
wed
th
at
in
th
e
ca
s
e
o
f
Pre
cisi
o
n
was
th
e
h
ig
h
est
r
esu
lt
o
f
class
if
ier
wh
en
u
s
ed
with
I
C
H
I
s
q
u
ar
e
as
a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
Mo
r
eo
v
er
,
t
h
at
in
t
h
e
ca
s
e
o
f
R
ec
all
was
th
e
h
ig
h
est
r
esu
lts
o
f
th
e
class
if
ier
wh
en
u
s
ed
with
C
HI
s
q
u
ar
e
as
a
f
ea
tu
r
e
s
elec
tio
n
m
et
h
o
d
,
b
u
t
in
th
e
ca
s
e
o
f
F
-
m
ea
s
u
r
e
was
th
e
h
ig
h
es
t
r
esu
lts
o
f
t
h
e
class
if
ier
wh
en
u
s
ed
with
C
HI
s
q
u
a
r
e
as
a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
an
d
in
th
e
ca
s
e
o
f
T
im
e
was
th
e
le
s
s
tim
e
to
b
u
ild
m
o
d
el
o
f
th
e
class
if
ier
wh
en
u
s
e
d
with
W
ith
o
u
t
F
S
as a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
Fig
u
r
e
8
s
h
o
ws d
iag
r
am
ANNs
r
esu
lts
with
o
u
t p
r
e
-
p
r
o
ce
s
s
in
g
.
T
ab
le
6
.
ANNs
r
esu
lts
with
o
u
t p
r
e
-
p
r
o
ce
s
s
in
g
A
l
g
o
r
i
t
h
m
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F
-
mea
su
r
e
Ti
me
I
C
H
I
0
.
7
1
3
0
.
4
9
2
0
.
5
8
2
4
4
.
1
7
C
H
I
0
.
6
9
9
0
.
5
4
2
0
.
6
1
1
4
3
.
7
IG
0
.
6
3
5
0
.
5
2
3
0
.
5
7
3
5
0
.
1
MI
0
.
5
9
0
0
.
4
6
0
0
.
5
1
6
5
9
.
9
W
r
a
p
p
e
r
0
.
6
0
1
0
.
4
9
9
0
.
5
4
5
4
9
.
1
A
v
e
r
a
g
e
0
.
6
4
7
0
.
5
0
3
0
.
5
6
5
4
9
.
4
W
i
t
h
o
u
t
F
S
0
.
5
8
6
0
.
4
5
2
0
.
5
1
0
3
6
.
3
4
Fig
u
r
e
8
.
ANNs
r
esu
lts
with
o
u
t p
r
e
-
p
r
o
ce
s
s
in
g
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
7
6
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
,
Vo
l.
9
,
No
.
3,
Dec
em
b
e
r
20
20
:
15
7
–
1
70
164
3
.
2
.
Resul
t
s
wit
h
na
t
ura
l la
ng
ua
g
e
pro
ce
s
s
ing
3
.
2
.
1
.
B
a
y
es
net
(
B
N)
T
ab
le
7
s
h
o
ws
b
ay
es
n
et
class
if
ier
r
esu
lts
with
p
r
e
-
p
r
o
c
ess
in
g
u
s
in
g
s
ev
er
al
f
ea
tu
r
e
s
elec
tio
n
alg
o
r
ith
m
s
.
T
h
e
r
esu
lts
s
h
o
wed
th
at
in
th
e
ca
s
e
o
f
Pre
cisi
o
n
was
th
e
h
ig
h
est
r
esu
lt
o
f
class
if
ier
wh
en
u
s
ed
with
I
C
HI
s
q
u
ar
e
as
a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
Mo
r
eo
v
er
,
th
at
in
th
e
ca
s
e
o
f
R
ec
all
was
th
e
h
ig
h
est
r
esu
lts
o
f
th
e
class
if
ier
wh
en
u
s
ed
w
i
th
I
C
HI
s
q
u
ar
e
as
a
f
ea
t
u
r
e
s
e
lectio
n
m
eth
o
d
,
b
u
t
in
th
e
ca
s
e
o
f
F
-
m
ea
s
u
r
e
was
th
e
h
ig
h
est
r
esu
lts
o
f
th
e
class
if
ier
wh
en
u
s
ed
with
I
C
HI
s
q
u
ar
e
as
a
f
ea
tu
r
e
s
elec
tio
n
m
et
h
o
d
an
d
in
th
e
ca
s
e
o
f
T
im
e
was
th
e
less
tim
e
to
b
u
ild
m
o
d
el
o
f
th
e
cla
s
s
if
ier
wh
en
u
s
ed
with
W
r
ap
p
er
as
a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
Fig
u
r
e
9
s
h
o
ws d
iag
r
am
B
ay
es Ne
t r
esu
lts
with
p
r
e
-
p
r
o
ce
s
s
in
g
.
T
ab
le
7
.
B
ay
es
n
et
r
esu
lts
with
p
r
e
-
p
r
o
ce
s
s
in
g
A
l
g
o
r
i
t
h
m
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F
-
mea
su
r
e
Ti
me
I
C
H
I
0
.
8
8
3
0
.
9
2
2
0
.
9
0
2
2
0
.
3
C
H
I
0
.
8
5
1
0
.
8
8
8
0
.
8
6
9
2
2
.
5
IG
0
.
8
4
4
0
.
8
9
1
0
.
8
6
6
2
7
.
1
MI
0
.
7
9
0
0
.
8
0
1
0
.
7
9
5
2
1
.
1
W
r
a
p
p
e
r
0
.
8
3
9
0
.
8
7
0
0
.
8
5
4
1
7
.
2
A
v
e
r
a
g
e
0
.
8
4
1
0
.
8
7
4
0
.
8
5
7
2
1
.
6
W
i
t
h
o
u
t
F
S
0
.
8
2
8
0
.
8
9
3
0
.
8
5
9
1
6
.
7
Fig
u
r
e
9
.
B
ay
es
n
et
r
esu
lts
with
p
r
e
-
p
r
o
ce
s
s
in
g
3
.
2
.
2
.
Na
ïv
e
ba
y
es
(
NB
)
T
ab
le
8
s
h
o
ws
n
aïv
e
b
ay
es
class
if
ier
r
esu
lts
with
p
r
e
-
p
r
o
ce
s
s
in
g
u
s
in
g
s
ev
er
al
f
ea
tu
r
e
s
elec
tio
n
alg
o
r
ith
m
s
.
T
h
e
r
esu
lts
s
h
o
wed
th
at
in
th
e
ca
s
e
o
f
Pre
cisi
o
n
was
th
e
h
ig
h
est
r
esu
lt
o
f
class
if
ier
wh
en
u
s
ed
with
C
HI
s
q
u
ar
e
as
a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
Mo
r
e
o
v
er
,
th
at
in
th
e
ca
s
e
o
f
R
ec
al
l
was
th
e
h
ig
h
est
r
esu
lts
o
f
th
e
class
if
ier
wh
en
u
s
ed
with
I
C
HI
s
q
u
ar
e
as
a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
,
b
u
t
in
th
e
ca
s
e
o
f
F
-
m
ea
s
u
r
e
was
th
e
h
ig
h
est
r
esu
lts
o
f
th
e
class
if
ie
r
wh
en
u
s
ed
with
I
C
HI
s
q
u
ar
e
as
a
f
ea
tu
r
e
s
elec
tio
n
m
e
th
o
d
an
d
in
th
e
ca
s
e
o
f
T
im
e
was
th
e
less
tim
e
to
b
u
ild
m
o
d
el
o
f
th
e
class
if
ier
wh
en
u
s
ed
with
W
ith
o
u
t
FS
as
a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
Fig
u
r
e
1
0
s
h
o
ws d
iag
r
am
Naïv
e
B
ay
es r
esu
lts
with
p
r
e
-
p
r
o
ce
s
s
in
g
.
T
ab
le
8
.
Naiv
e
b
ay
es r
esu
lts
with
p
r
e
-
p
r
o
ce
s
s
in
g
A
l
g
o
r
i
t
h
m
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F
-
mea
su
r
e
Ti
me
C
H
I
0
.
9
4
4
0
.
9
6
6
0
.
9
5
4
5
.
5
I
C
H
I
0
.
9
5
3
0
.
9
5
6
0
.
9
5
4
9
.
1
1
IG
0
.
9
2
9
0
.
9
5
0
0
.
9
3
9
1
1
.
2
MI
0
.
9
1
4
0
.
9
4
5
0
.
9
2
9
1
3
.
7
W
r
a
p
p
e
r
0
.
9
2
0
0
.
9
4
9
0
.
9
3
4
1
7
.
1
A
v
e
r
a
g
e
0
.
9
3
2
0
.
9
5
3
0
.
9
4
2
1
1
.
3
W
i
t
h
o
u
t
F
S
0
.
9
0
5
0
.
9
4
4
0
.
9
2
4
3
.
1
1
Fig
u
r
e
1
0
.
Naiv
e
b
ay
es r
esu
lts
with
p
r
e
-
p
r
o
ce
s
s
in
g
3
.
2
.
3
.
Ra
nd
o
m
f
o
re
s
t
(
RF
)
T
ab
le
9
s
h
o
ws
r
a
n
d
o
m
f
o
r
est
class
if
ier
r
esu
lts
with
p
r
e
-
p
r
o
ce
s
s
in
g
u
s
in
g
s
ev
er
al
f
ea
tu
r
e
s
elec
tio
n
alg
o
r
ith
m
s
.
T
h
e
r
esu
lts
s
h
o
wed
th
at
in
th
e
ca
s
e
o
f
Pre
cisi
o
n
was
th
e
h
ig
h
est
r
esu
lt
o
f
class
if
ier
wh
en
u
s
ed
with
I
C
HI
s
q
u
ar
e
as
a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
Mo
r
eo
v
er
,
th
at
in
th
e
ca
s
e
o
f
R
ec
all
was
th
e
h
ig
h
est
r
esu
lts
o
f
th
e
class
if
ier
wh
en
u
s
ed
wi
th
I
C
HI
s
q
u
ar
e
as
a
f
ea
t
u
r
e
s
e
lectio
n
m
eth
o
d
,
b
u
t
in
th
e
ca
s
e
o
f
F
-
m
ea
s
u
r
e
was
th
e
h
ig
h
est
r
esu
lts
o
f
th
e
class
if
ier
wh
en
u
s
ed
with
I
C
HI
s
q
u
ar
e
as
a
f
ea
tu
r
e
s
elec
tio
n
m
et
h
o
d
an
d
in
th
e
ca
s
e
o
f
T
im
e
was
th
e
less
tim
e
to
b
u
ild
m
o
d
el
o
f
t
h
e
class
if
ier
wh
en
u
s
ed
with
W
ith
o
u
t
FS
as
a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
Fig
u
r
e
1
1
s
h
o
ws d
iag
r
am
R
an
d
o
m
Fo
r
est
r
esu
lts
with
p
r
e
-
p
r
o
ce
s
s
in
g
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
I
SS
N:
2252
-
8
7
7
6
I
mp
r
o
ve
d
I
C
HI
s
q
u
a
r
e
fea
tu
r
e
s
elec
tio
n
meth
o
d
fo
r
A
r
a
b
ic
c
la
s
s
ifie
r
s
(
Ha
d
ee
l N
)
165
T
ab
le
9
.
R
an
d
o
m
f
o
r
est r
esu
lts
with
p
r
e
-
p
r
o
ce
s
s
in
g
A
l
g
o
r
i
t
h
m
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F
-
mea
su
r
e
Ti
me
I
C
H
I
0
.
9
5
8
0
.
8
6
8
0
.
9
1
1
7
0
.
1
C
H
I
0
.
9
1
7
0
.
7
9
6
0
.
8
5
2
9
1
.
2
IG
0
.
9
1
0
0
.
7
8
9
0
.
8
4
5
9
1
.
3
MI
0
.
8
5
1
0
.
7
7
1
0
.
8
0
9
8
4
.
1
W
r
a
p
p
e
r
0
.
8
9
1
0
.
7
8
1
0
.
8
3
2
7
1
.
9
A
v
e
r
a
g
e
0
.
9
0
5
0
.
8
0
1
0
.
8
4
9
8
1
.
7
W
i
t
h
o
u
t
F
S
0
.
8
4
9
0
.
7
6
5
0
.
8
0
4
6
1
.
3
Fig
u
r
e
1
1
.
R
an
d
o
m
f
o
r
est r
esu
lts
with
p
r
e
-
p
r
o
ce
s
s
in
g
3
.
2
.
4
.
Dec
is
io
n
t
re
e
(
DT
)
T
ab
le
1
0
s
h
o
ws
d
ec
is
io
n
tr
ee
class
if
ier
r
esu
lts
with
p
r
e
-
p
r
o
ce
s
s
in
g
u
s
in
g
s
ev
er
al
f
ea
tu
r
e
s
elec
tio
n
alg
o
r
ith
m
s
.
T
h
e
r
esu
lts
s
h
o
wed
th
at
in
th
e
ca
s
e
o
f
Pre
cisi
o
n
was
th
e
h
ig
h
est
r
esu
lt
o
f
class
if
ier
wh
en
u
s
ed
with
I
C
HI
s
q
u
ar
e
as
a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
Mo
r
eo
v
er
,
th
at
in
th
e
ca
s
e
o
f
R
ec
all
was
th
e
h
ig
h
est
r
esu
lts
o
f
th
e
class
if
ier
wh
en
u
s
ed
wi
th
I
C
HI
s
q
u
ar
e
as
a
f
ea
t
u
r
e
s
e
lectio
n
m
eth
o
d
,
b
u
t
in
th
e
ca
s
e
o
f
F
-
m
ea
s
u
r
e
was
th
e
h
ig
h
est
r
esu
lts
o
f
th
e
class
if
ier
wh
en
u
s
ed
with
I
C
HI
s
q
u
ar
e
as
a
f
ea
tu
r
e
s
elec
tio
n
m
et
h
o
d
an
d
in
th
e
ca
s
e
o
f
T
im
e
was
th
e
less
tim
e
to
b
u
ild
m
o
d
el
o
f
th
e
class
if
ier
wh
en
u
s
ed
with
W
ith
o
u
t
FS
as
a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
Fig
u
r
e
1
2
s
h
o
ws d
iag
r
am
Dec
is
io
n
T
r
ee
r
esu
lts
with
p
r
e
-
p
r
o
ce
s
s
in
g
.
T
ab
le
10
.
Dec
is
io
n
tr
ee
r
esu
lts
with
p
r
e
-
p
r
o
ce
s
s
in
g
A
l
g
o
r
i
t
h
m
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F
-
mea
su
r
e
Ti
me
I
C
H
I
0
.
5
7
7
0
.
5
3
9
0
.
5
5
7
3
.
1
C
H
I
0
.
5
2
3
0
.
5
0
2
0
.
5
1
2
3
.
3
4
IG
0
.
5
2
1
0
.
5
0
3
0
.
5
1
2
4
.
1
MI
0
.
5
1
5
0
.
5
0
0
0
.
5
0
7
5
.
4
W
r
a
p
p
e
r
0
.
5
1
9
0
.
5
0
2
0
.
5
1
0
8
.
2
A
v
e
r
a
g
e
0
.
5
3
1
0
.
5
0
9
0
.
5
1
9
4
.
8
2
W
i
t
h
o
u
t
F
S
0
.
5
1
6
0
.
5
0
3
0
.
5
0
9
1
.
0
5
Fig
u
r
e
1
2
.
Dec
is
io
n
tr
ee
r
esu
lt
s
with
p
r
e
-
p
r
o
ce
s
s
in
g
3
.
2
.
5
.
Art
if
icia
l
neura
l n
et
wo
rk
s
(
ANNs)
T
ab
le
1
1
s
h
o
ws
ar
tific
ial
n
eu
r
al
n
etwo
r
k
s
class
if
ier
r
e
s
u
lts
with
p
r
e
-
p
r
o
ce
s
s
in
g
u
s
in
g
s
ev
er
al
f
ea
tu
r
e
s
elec
tio
n
alg
o
r
ith
m
s
.
T
h
e
r
esu
lts
s
h
o
wed
th
at
i
n
th
e
ca
s
e
o
f
Pre
cisi
o
n
was
th
e
h
ig
h
est
r
esu
lt
o
f
class
if
ier
wh
e
n
u
s
ed
with
I
C
HI
s
q
u
ar
e
as
a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
M
o
r
eo
v
er
,
th
at
i
n
th
e
ca
s
e
o
f
R
ec
all
was
th
e
h
ig
h
est
r
esu
lts
o
f
th
e
class
if
ier
wh
en
u
s
ed
with
I
C
HI
s
q
u
ar
e
as
a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
,
b
u
t
in
th
e
ca
s
e
o
f
F
-
m
ea
s
u
r
e
was
th
e
h
ig
h
est
r
esu
l
ts
o
f
th
e
class
if
ier
wh
en
u
s
ed
with
I
C
HI
s
q
u
ar
e
as
a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
an
d
in
th
e
ca
s
e
o
f
T
im
e
was
t
h
e
less
tim
e
to
b
u
ild
m
o
d
el
o
f
th
e
class
if
ier
wh
en
u
s
ed
with
Mu
tu
al
I
n
f
o
r
m
atio
n
as a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
Fig
u
r
e
1
3
s
h
o
ws d
iag
r
am
AN
N
s
r
esu
lts
with
p
r
e
-
p
r
o
ce
s
s
in
g
.
T
ab
le
1
1
.
ANNs
r
esu
lts
with
p
r
e
-
p
r
o
ce
s
s
in
g
A
l
g
o
r
i
t
h
m
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F
-
mea
su
r
e
Ti
me
I
C
H
I
0
.
7
1
9
0
.
5
6
0
0
.
6
2
9
5
1
.
2
C
H
I
0
.
6
7
2
0
.
4
6
7
0
.
5
5
1
3
9
.
1
IG
0
.
6
6
6
0
.
5
5
9
0
.
6
0
7
2
9
.
1
MI
0
.
5
9
9
0
.
5
0
1
0
.
5
4
5
2
1
.
4
W
r
a
p
p
e
r
0
.
6
5
1
0
.
5
4
1
0
.
5
9
1
3
9
.
2
A
v
e
r
a
g
e
0
.
6
6
1
0
.
5
2
5
0
.
5
8
5
3
5
.
9
W
i
t
h
o
u
t
F
S
0
.
5
8
0
0
.
5
0
6
0
.
5
4
0
3
9
.
3
Fig
u
r
e
13
.
ANNs
r
esu
lts
with
p
r
e
-
p
r
o
ce
s
s
in
g
3
.
3
.
Dis
cus
s
io
n t
he
re
s
u
lt
s
w
it
ho
ut
pre
-
pro
ce
s
s
ing
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
7
6
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
,
Vo
l.
9
,
No
.
3,
Dec
em
b
e
r
20
20
:
15
7
–
1
70
166
3
.
3
.
1
.
Resul
t
s
ba
s
ed
o
n
avg
.
prec
is
io
n
T
ab
le
1
2
a
n
d
Fig
u
r
e
1
4
s
h
o
ws
r
esu
lts
b
ased
o
n
av
g
.
p
r
ec
is
io
n
with
o
u
t
p
r
e
-
p
r
o
ce
s
s
in
g
.
W
e
f
o
u
n
d
th
a
t
wh
en
we
co
m
p
a
r
e
th
e
s
tate
o
f
Pre
cisi
o
n
an
d
also
with
o
u
t
p
r
e
-
p
r
o
ce
s
s
in
g
,
th
e
b
est r
esu
lts
ar
e
as f
o
llo
ws:
•
B
ay
es Ne
t Cl
a
s
s
if
ier
wh
en
u
s
ed
with
I
C
HI
s
q
u
ar
e
as a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
•
Naïv
e
B
ay
es C
las
s
if
ier
wh
en
u
s
ed
with
I
C
HI
s
q
u
ar
e
as a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
•
R
an
d
o
m
Fo
r
est C
lass
if
ier
wh
e
n
u
s
ed
with
I
C
HI
s
q
u
ar
e
as a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
•
Dec
is
io
n
T
r
ee
C
lass
if
ier
wh
en
u
s
ed
with
I
C
HI
s
q
u
ar
e
as a
f
e
atu
r
e
s
elec
tio
n
m
eth
o
d
.
•
Ar
tific
ial
Neu
r
al
Netwo
r
k
s
C
lass
if
ier
wh
en
u
s
ed
with
I
C
HI
s
q
u
ar
e
as a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
T
ab
le
1
2
.
R
esu
lts
b
ased
o
n
a
v
g
.
p
r
ec
is
io
n
with
o
u
t p
r
e
-
p
r
o
ce
s
s
in
g
A
l
g
o
r
i
t
h
m
I
C
H
I
C
H
I
IG
MI
W
r
a
p
p
e
r
W
i
t
h
o
u
t
FS
BN
0
.
9
7
6
0
.
8
5
8
0
.
8
5
1
0
.
8
1
9
0
.
8
3
2
0
.
8
0
5
NB
0
.
9
7
6
0
.
9
2
6
0
.
9
2
2
0
.
9
1
3
0
.
9
1
9
0
.
9
1
3
RF
0
.
9
4
7
0
.
9
3
8
0
.
9
3
1
0
.
9
1
7
0
.
9
2
9
0
.
9
1
5
DT
0
.
6
0
4
0
.
5
5
2
0
.
5
4
7
0
.
5
2
9
0
.
5
3
9
0
.
5
2
8
ANNs
0
.
7
1
3
0
.
6
9
9
0
.
6
3
5
0
.
5
9
0
0
.
6
0
1
0
.
5
8
6
Fig
u
r
e
1
4
.
R
esu
lts
b
ased
o
n
av
g
.
p
r
ec
is
io
n
with
o
u
t p
r
e
-
p
r
o
ce
s
s
in
g
3
.
3
.
2
.
Resul
t
s
ba
s
ed
o
n a
v
g
.
re
ca
ll
T
ab
le
1
3
an
d
Fig
u
r
e
1
5
s
h
o
w
s
r
esu
lts
b
ased
o
n
a
v
g
.
r
ec
all
with
o
u
t
p
r
e
-
p
r
o
ce
s
s
in
g
.
W
e
f
o
u
n
d
th
at
wh
en
we
co
m
p
a
r
e
th
e
s
tate
o
f
R
ec
all
an
d
also
with
o
u
t p
r
e
-
p
r
o
ce
s
s
in
g
,
th
e
b
est r
esu
lts
ar
e
a
s
f
o
llo
ws:
•
B
ay
es Ne
t Cl
a
s
s
if
ier
wh
en
u
s
e
d
with
I
C
HI
s
q
u
ar
e
as a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
•
Naïv
e
B
ay
es C
las
s
if
ier
wh
en
u
s
ed
with
I
C
HI
s
q
u
ar
e
as a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
•
R
an
d
o
m
Fo
r
est C
lass
if
ier
wh
e
n
u
s
ed
with
I
C
HI
s
q
u
ar
e
as a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
•
Dec
is
io
n
T
r
ee
C
lass
if
ier
wh
en
u
s
ed
with
C
HI
s
q
u
ar
e
as a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
•
Ar
tific
ial
Neu
r
al
Netwo
r
k
s
C
lass
if
ier
wh
en
u
s
ed
with
C
HI
s
q
u
ar
e
as a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
T
ab
le
1
3
.
R
esu
lts
b
ased
o
n
a
v
g
.
r
ec
all
with
o
u
t
p
r
e
-
p
r
o
ce
s
s
in
g
A
l
g
o
r
i
t
h
m
I
C
H
I
C
H
I
IG
MI
W
r
a
p
p
e
r
W
i
t
h
o
u
t
FS
BN
0
.
9
4
0
0
.
9
1
0
0
.
8
9
9
0
.
8
7
3
0
.
8
8
8
0
.
8
7
1
NB
0
.
9
5
4
0
.
9
5
0
0
.
9
5
2
0
.
9
4
5
0
.
9
5
0
0
.
9
5
1
RF
0
.
9
4
2
0
.
8
0
5
0
.
8
9
9
0
.
8
4
9
0
.
8
5
1
0
.
8
4
1
DT
0
.
5
1
8
0
.
5
3
7
0
.
5
3
3
0
.
5
0
0
0
.
5
3
0
0
.
5
2
3
ANNs
0
.
4
9
2
0
.
5
4
2
0
.
5
2
3
0
.
4
6
0
0
.
4
9
9
0
.
5
2
3
Fig
u
r
e
15
.
R
esu
lts
b
ased
o
n
av
g
.
r
ec
all
with
o
u
t
p
r
e
-
p
r
o
ce
s
s
in
g
3
.
3
.
3
.
Resul
t
s
ba
s
ed
o
n a
v
g
.
f
-
m
ea
s
ure
T
ab
le
1
4
a
n
d
Fig
u
r
e
1
6
s
h
o
ws
r
esu
lts
b
ased
on
av
g
.
f
-
m
ea
s
u
r
e
with
o
u
t
p
r
e
-
p
r
o
ce
s
s
in
g
.
W
e
f
o
u
n
d
th
at
wh
en
we
co
m
p
ar
e
th
e
s
tat
e
o
f
F
-
m
ea
s
u
r
e
a
n
d
also
with
o
u
t p
r
e
-
p
r
o
ce
s
s
in
g
,
th
e
b
est
r
esu
lts
ar
e
as f
o
llo
ws:
•
B
ay
es Ne
t Cl
a
s
s
if
ier
wh
en
u
s
ed
with
I
C
HI
s
q
u
ar
e
as a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
•
Naïv
e
B
ay
es C
las
s
if
ier
wh
en
u
s
ed
with
I
C
HI
s
q
u
ar
e
as a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
.
•
R
an
d
o
m
Fo
r
est C
lass
if
ier
wh
e
n
u
s
ed
with
I
C
HI
s
q
u
ar
e
as a
f
ea
tu
r
e
s
elec
ti
o
n
m
eth
o
d
.
•
Dec
is
io
n
T
r
ee
C
lass
if
ier
wh
en
u
s
ed
with
I
C
HI
s
q
u
ar
e
as a
f
e
atu
r
e
s
elec
tio
n
m
eth
o
d
.
Evaluation Warning : The document was created with Spire.PDF for Python.