I
AE
S
I
nte
rna
t
io
na
l J
o
urna
l o
f
Art
if
icia
l In
t
ellig
ence
(
I
J
-
AI
)
Vo
l.
9
,
No
.
2
,
J
u
n
e
2020
,
p
p
.
327
~
3
3
5
I
SS
N:
2252
-
8938
,
DOI
: 1
0
.
1
1
5
9
1
/i
j
ai.
v
9
.i
2
.
p
p
327
-
3
3
5
327
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
Bio
g
eo
g
ra
phy
opt
i
m
i
z
a
tion
alg
o
rith
m
ba
sed nex
t
w
eb
pag
e
prediction
using
w
eblo
g
and
w
eb content
featur
es
Ro
s
ha
n A.
G
a
ng
ur
de
1
,
B
ino
d K
u
m
a
r
2
1
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter S
c
ien
c
e
,
S
a
v
it
rib
a
i
P
h
u
le
P
u
n
e
U
n
iv
e
rsity
,
M
a
h
a
ra
sh
tra,
In
d
ia
2
Ja
y
a
wa
n
t
In
stit
u
te o
f
Co
m
p
u
t
e
r
A
p
p
li
c
a
ti
o
n
s,
P
u
n
e
,
M
a
h
a
ra
sh
tra,
In
d
ia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Dec
2
5
,
2
0
19
R
ev
i
s
ed
F
eb
2
9
,
2
0
20
A
cc
ep
ted
M
ar
1
,
2
0
20
Re
c
o
m
m
e
n
d
a
ti
o
n
o
f
we
b
p
a
g
e
a
s
p
e
r
u
se
rs’
in
tere
st
is
a
b
ro
a
d
a
n
d
im
p
o
rtan
t
a
re
a
o
f
re
se
a
rc
h
.
Re
se
a
rc
h
e
r
a
d
o
p
ts
u
se
r
b
e
h
a
v
io
r
f
ro
m
a
c
ti
o
n
s
p
re
se
n
t
i
n
c
o
o
k
ies
,
lo
g
s
a
n
d
se
a
rc
h
q
u
e
ries
.
T
h
is
p
a
p
e
r
h
a
s
u
ti
li
z
e
d
a
p
ri
o
r
we
b
p
a
g
e
f
e
tch
in
g
m
o
d
e
l
u
sin
g
w
e
b
p
a
g
e
p
re
d
ictio
n
.
F
o
r
t
h
is
p
u
rp
o
se
,
w
e
b
c
o
n
ten
t
in
f
o
r
m
o
f
te
x
t
a
n
d
w
e
b
lo
g
f
e
a
t
u
re
s
a
re
a
n
a
l
y
z
e
d
.
A
s
p
e
r
d
y
n
a
m
ic
u
se
r
b
e
h
a
v
io
r,
p
ro
p
o
se
d
m
o
d
e
l
LW
P
P
-
BOA
(L
o
g
isti
c
Web
P
a
g
e
P
re
d
ictio
n
By
Bio
g
e
o
g
ra
p
h
y
Op
ti
m
i
z
a
ti
o
n
A
lg
o
rit
h
m
)
p
re
d
ict
p
a
g
e
b
y
u
sin
g
g
e
n
e
ti
c
a
lg
o
rit
h
m
.
Ba
s
e
d
o
n
u
se
r
a
c
ti
o
n
s
,
we
b
lo
g
f
e
a
tu
re
a
re
d
e
v
e
lo
p
e
d
in
f
o
r
m
o
f
a
ss
o
c
iatio
n
ru
les
,
w
h
il
e
w
e
b
c
o
n
ten
t
g
iv
e
s
a
se
t
o
f
r
e
lev
a
n
t
tex
t
p
a
tt
e
rn
s.
P
a
g
e
p
re
d
ictio
n
a
s
p
e
r
ra
n
d
o
m
u
se
r
b
e
h
a
v
io
r
is
e
n
h
a
n
c
e
d
b
y
m
e
a
n
s
o
f
Bio
g
e
o
g
ra
p
h
y
Op
ti
m
i
z
a
ti
o
n
A
lg
o
rit
h
m
w
h
e
re
c
ro
ss
o
v
e
r
o
p
e
ra
ti
o
n
is
p
e
rf
o
r
m
e
d
a
s
p
e
r
imm
ig
ra
ti
o
n
a
n
d
e
m
ig
r
a
ti
o
n
v
a
lu
e
s.
He
re
p
o
p
u
latio
n
u
p
d
a
t
io
n
d
e
p
e
n
d
s
o
n
o
t
h
e
r
p
a
ra
m
e
ters
o
f
c
h
ro
m
o
so
m
e
e
x
c
e
p
t
f
it
n
e
ss
v
a
lu
e
.
Ex
p
e
rime
n
ts
a
re
c
o
n
d
u
c
ted
o
n
re
a
l
d
a
tas
e
t
h
a
v
in
g
w
e
b
c
o
n
ten
t
a
n
d
w
e
b
lo
g
s.
Re
su
lt
s
a
re
c
o
m
p
a
r
e
d
u
sin
g
p
re
c
i
sio
n
,
c
o
v
e
ra
g
e
,
M
-
M
e
tri
c
,
M
A
E
a
n
d
RM
S
E
p
a
ra
m
e
ters
a
n
d
it
in
d
ica
tes
th
a
t
th
e
p
r
o
p
o
se
d
w
o
rk
is
b
e
tt
e
r
t
h
a
n
o
th
e
r
a
p
p
ro
a
c
h
e
s a
lrea
d
y
in
u
se
.
K
ey
w
o
r
d
s
:
Ass
o
ciatio
n
r
u
le
m
in
in
g
B
io
g
eo
g
r
ap
h
y
o
p
ti
m
izatio
n
On
to
lo
g
y
P
r
ed
ictio
n
m
o
d
el
W
eb
p
ag
e
r
ec
o
m
m
e
n
d
atio
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
:
R
o
s
h
a
n
A
.
Ga
n
g
u
r
d
e,
R
esear
ch
Sch
o
lar
,
Dep
ar
t
m
e
n
t
o
f
C
o
m
p
u
ter
Scien
ce
,
Sav
itrib
ai
P
h
u
le
P
u
n
e
U
n
i
v
er
s
it
y
,
Ma
h
ar
as
h
tr
a,
I
n
d
ia
.
E
m
ail: r
o
s
h
an
a
n
t
@
g
m
ai
l.c
o
m
1.
I
NT
RO
D
UCT
I
O
N
T
h
e
n
et
w
o
r
k
is
v
er
y
b
ig
,
d
i
s
s
i
m
ilar
an
d
e
v
er
ch
a
n
g
i
n
g
.
E
x
citi
n
g
i
n
f
o
r
m
atio
n
ex
tr
ac
tio
n
f
r
o
m
W
eb
in
f
o
r
m
atio
n
h
as
b
ec
o
m
e
v
er
y
tr
e
n
d
y
a
n
d
as
a
n
o
u
tc
o
m
e
o
f
t
h
at,
w
eb
m
in
in
g
h
as
f
asci
n
ated
lo
ts
o
f
a
w
ar
e
n
ess
i
n
r
ec
en
t
ti
m
e.
W
eb
m
i
n
i
n
g
is
a
f
u
n
ctio
n
o
f
w
eb
in
f
o
r
m
at
io
n
m
i
n
i
n
g
f
r
o
m
h
u
g
e
w
eb
in
f
o
r
m
atio
n
r
ep
o
s
ito
r
ies
[
1
]
.
A
t
th
is
t
i
m
e,
t
h
e
s
etb
ac
k
o
f
m
o
d
eli
n
g
a
n
d
f
o
r
ec
asti
n
g
a
u
s
er
’
s
w
eb
b
r
o
w
s
i
n
g
p
er
f
o
r
m
an
ce
o
n
th
e
in
ter
n
et
h
as
f
asci
n
ated
lo
t
o
f
r
esear
ch
er
s
as
it
ca
n
b
e
u
s
e
f
u
l
in
d
ev
elo
p
i
n
g
w
eb
c
ac
h
e
p
er
f
o
r
m
a
n
ce
,
W
eb
P
ag
e
R
ec
o
m
m
e
n
d
atio
n
s
,
Sear
ch
E
n
g
i
n
e
Op
ti
m
iza
tio
n
s
,
ac
ce
p
tin
g
an
d
in
f
l
u
en
ci
n
g
b
u
y
i
n
g
s
a
m
p
le
s
,
an
d
p
r
ac
ticin
g
t
h
e
w
eb
s
ea
r
ch
ex
p
er
ien
ce
o
f
t
h
e
u
s
er
s
[
2
]
.
I
n
e
-
co
m
m
er
ce
,
th
e
u
p
co
m
in
g
p
ag
e
f
o
r
ec
asti
n
g
i
s
ex
tr
e
m
e
l
y
ess
e
n
tia
l
a
n
d
s
er
io
u
s
.
T
h
e
ca
lc
u
lat
io
n
s
u
p
p
o
r
ts
co
r
p
o
r
atio
n
s
to
h
a
n
d
le
t
h
e
i
s
s
u
e
s
c
o
n
ce
r
n
in
g
to
u
s
er
s
s
u
c
h
as t
h
eir
m
o
v
e
m
e
n
t
s
in
p
u
r
ch
a
s
in
g
a
n
d
in
ter
est
s
o
f
th
e
u
s
er
s
i
n
an
y
s
p
ec
i
f
ic
p
r
o
d
u
cts.
W
eb
u
s
ag
e
m
i
n
in
g
ca
n
b
e
u
til
i
ze
d
to
v
er
y
ca
u
tio
u
s
l
y
e
x
a
m
in
e
w
eb
lo
g
r
ec
o
r
d
s
w
h
ic
h
ar
e
c
o
llected
in
w
eb
s
er
v
er
s
f
o
r
p
atter
n
d
e
tectio
n
.
W
eb
m
in
in
g
c
h
ief
l
y
co
n
ce
n
tr
ates
o
n
t
h
e
p
r
ep
r
o
ce
s
s
in
g
s
ec
tio
n
an
d
g
r
o
u
p
in
g
s
ec
tio
n
.
De
f
i
n
i
n
g
a
cl
u
s
ter
i
s
h
ar
d
an
d
t
h
at
is
w
h
y
,
w
e
f
in
d
d
is
s
i
m
ilar
k
in
d
s
o
f
al
g
o
r
ith
m
s
co
n
n
ec
ted
to
clu
s
ter
i
n
g
.
T
h
e
ex
is
te
n
ce
o
f
co
llectio
n
o
f
r
ec
o
r
d
s
is
th
e
n
o
r
m
al
th
i
n
g
th
at
co
n
n
ec
ts
all
t
h
ese
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
I
n
t J
A
r
ti
f
I
n
tell
,
Vo
l.
9
,
No
.
2
,
J
u
n
e
20
20
:
327
–
3
3
5
328
alg
o
r
ith
m
s
j
o
in
tl
y
.
I
t
co
m
p
r
i
s
es
o
f
th
r
ee
p
ar
ts
:
p
r
ep
r
o
ce
s
s
in
g
,
p
atter
n
d
etec
tio
n
an
d
p
atter
n
ex
a
m
i
n
atio
n
.
P
r
ep
r
o
ce
s
s
in
g
is
m
an
d
ato
r
y
to
ch
an
g
e
t
h
e
r
a
w
i
n
f
o
r
m
atio
n
i
n
to
a
s
ig
n
i
f
ica
n
t
f
o
r
m
h
e
lp
f
u
l
f
o
r
w
el
l
-
o
r
g
a
n
ized
p
r
o
ce
s
s
in
g
.
P
atter
n
d
etec
tio
n
co
m
p
r
is
es
m
et
h
o
d
s
to
tak
e
o
u
t
t
h
e
p
atter
n
an
d
i
n
cl
u
d
es
n
u
m
er
ical
a
n
al
y
s
is
,
s
eq
u
en
tial p
atter
n
m
i
n
in
g
,
p
at
h
an
al
y
s
is
,
r
elatio
n
s
h
ip
r
u
le
m
in
i
n
g
,
ca
te
g
o
r
izatio
n
,
an
d
b
u
n
ch
in
g
[
3
-
4
]
.
T
h
is
p
ap
er
h
as
attem
p
ted
to
im
p
r
o
v
e
th
e
ac
cu
r
ac
y
o
f
th
e
n
ex
t
w
eb
p
ag
e
p
r
ed
ictio
n
w
h
er
e
ac
cu
r
ac
y
o
f
p
r
ed
ictio
n
d
ep
en
d
s
o
n
th
e
u
tili
za
t
io
n
o
f
w
eb
m
in
in
g
f
ea
tu
r
es.
I
n
p
r
ev
io
u
s
r
esear
ch
,
Ma
r
k
o
v
m
o
d
el
w
a
s
u
tili
ze
d
f
o
r
w
eb
lo
g
f
ea
tu
r
e
[
5
]
.
T
h
is
n
ee
d
s
to
b
e
r
ep
lace
d
b
y
o
th
er
m
ath
e
m
at
ical
m
o
d
el
w
h
er
e
w
eb
co
n
ten
t
o
r
w
eb
s
tr
u
ct
u
r
e
f
ea
tu
r
e
ar
e
als
o
u
s
ed
.
A
s
w
eb
p
ag
e
p
r
ed
ictio
n
is
h
i
g
h
l
y
d
y
n
a
m
ic
in
n
at
u
r
e
,
a
s
u
itab
le
g
e
n
etic
alg
o
r
ith
m
n
ee
d
to
b
e
p
r
o
p
o
s
e
d
in
co
llab
o
r
atio
n
w
it
h
p
r
o
p
er
u
tili
za
tio
n
o
f
ass
o
ciatio
n
r
u
les
to
i
n
cr
ea
s
e
t
h
e
ac
cu
r
ac
y
o
f
p
r
ed
ictio
n
m
o
d
el.
2.
RE
L
AT
E
D
WO
RK
I
n
[
6
]
,
au
th
o
r
p
r
o
p
o
s
ed
a
n
ap
p
r
o
ac
h
h
y
b
r
id
u
ti
lizi
n
g
Su
p
p
o
r
t
Vec
to
r
Ma
ch
in
es,
an
d
th
e
A
ll
-
K
th
Ma
r
k
o
v
m
o
d
el,
to
d
eter
m
i
n
e
ca
lc
u
latio
n
u
tili
zin
g
De
m
p
s
ter
’
s
r
u
le.
T
o
b
o
o
s
t
th
e
p
o
w
er
o
f
d
is
cr
i
m
i
n
atio
n
,
t
h
e
y
ap
p
l
y
a
ttr
ib
u
te
ex
tr
ac
tio
n
o
f
SVM.
A
d
d
in
g
to
g
et
h
er
,
d
u
r
in
g
f
o
r
ec
ast
i
t
o
cc
u
p
ies
d
o
m
ain
in
f
o
r
m
atio
n
to
d
ec
r
ea
s
e
th
e
n
u
m
b
er
o
f
clas
s
i
f
ier
s
f
o
r
th
e
d
ev
elo
p
m
en
t
o
f
co
r
r
ec
tn
ess
a
n
d
th
e
d
ec
r
ea
s
e
o
f
ca
lcu
latio
n
ti
m
e.
I
n
[
7
]
,
au
th
o
r
ex
p
lain
ed
in
te
r
n
et
m
i
n
i
n
g
al
g
o
r
ith
m
th
a
t
tar
g
ets
at
m
o
d
if
y
i
n
g
th
e
an
a
l
y
s
i
s
o
f
th
e
d
r
af
t’
s
p
r
o
d
u
ctio
n
o
f
co
n
n
ec
ti
o
n
r
u
le
m
in
i
n
g
.
T
h
is
alg
o
r
it
h
m
is
b
ei
n
g
e
x
tr
e
m
e
l
y
u
tili
ze
d
in
in
ter
n
et
m
in
in
g
.
T
h
e
en
d
r
esu
lts
ac
h
iev
ed
es
tab
lis
h
ed
t
h
e
s
tr
e
n
g
t
h
o
f
t
h
e
al
g
o
r
ith
m
p
r
o
j
ec
ted
in
th
is
p
ap
er
.
I
n
[
8
]
,
au
t
h
o
r
co
n
ce
n
tr
ated
o
n
W
eb
Usa
g
e
M
in
i
n
g
w
h
er
ei
n
t
h
e
i
n
ter
n
et
u
s
er
r
o
u
ti
n
g
p
a
tter
n
s
an
d
th
eir
u
til
izatio
n
o
f
w
eb
r
eso
u
r
ce
s
ar
e
d
is
co
v
er
ed
.
T
h
e
d
is
s
im
ilar
p
h
ase
s
o
cc
u
p
ied
in
th
i
s
m
in
i
n
g
p
r
o
ce
d
u
r
e
an
d
w
ith
t
h
e
p
r
o
p
o
r
tio
n
al
s
tu
d
y
b
et
w
ee
n
t
h
e
p
atter
n
d
etec
tio
n
alg
o
r
it
h
m
s
:
A
p
r
io
r
i a
n
d
FP
-
g
r
o
w
t
h
al
g
o
r
ith
m
.
I
n
[
9
]
,
au
th
o
r
ex
ec
u
ted
t
h
e
p
r
ep
r
o
ce
s
s
in
g
m
et
h
o
d
s
to
c
h
an
g
e
th
e
lo
g
f
ile
i
n
to
clien
t
s
e
s
s
i
o
n
s
w
h
ic
h
ar
e
ap
p
r
o
p
r
iate
f
o
r
m
in
i
n
g
a
n
d
d
ec
r
ea
s
e
th
e
r
an
g
e
o
f
t
h
e
s
ess
io
n
f
i
le
b
y
s
o
r
ti
n
g
th
e
leas
t
d
e
m
an
d
ed
p
ag
e
s
u
tili
zi
n
g
t
h
e
p
r
ep
r
o
ce
s
s
in
g
m
et
h
o
d
.
I
n
f
o
r
m
at
io
n
P
r
ep
r
o
c
ess
i
n
g
i
s
o
n
e
o
f
th
e
s
i
g
n
i
f
ic
an
t
m
is
s
io
n
s
b
ef
o
r
e
in
s
er
ti
n
g
m
i
n
in
g
a
lg
o
r
it
h
m
s
.
I
t
ch
a
n
g
e
s
t
h
e
r
a
w
r
ec
o
r
d
f
ile
i
n
to
clie
n
t
s
es
s
io
n
.
I
n
t
h
is
v
o
c
atio
n
,
t
h
e
y
h
a
v
e
i
n
b
r
ief
estab
lis
h
ed
r
ec
o
r
d
f
ile
p
r
ep
r
o
ce
s
s
in
g
a
n
d
ap
p
lied
it
in
a
C
T
I
r
ec
o
r
d
f
ile.
A
ls
o
,
t
h
e
y
cr
ea
ted
th
e
r
ev
ie
w
o
f
th
e
clie
n
t ses
s
io
n
f
ile.
T
h
e
y
h
a
v
e
u
ti
lized
f
ilter
i
n
g
m
e
th
o
d
to
er
ad
icate
s
lig
h
test
d
e
m
an
d
ed
r
eso
u
r
ce
s
.
I
n
[
1
0
]
,
au
th
o
r
p
r
o
j
ec
ted
u
tili
zin
g
GR
U
s
to
s
t
u
d
y
f
u
r
th
er
m
e
an
in
g
f
u
l
ag
g
r
eg
a
tio
n
f
o
r
clien
t
b
r
o
w
s
i
n
g
h
is
to
r
y
(
b
r
o
w
s
ed
n
e
w
s
)
,
a
n
d
s
u
g
g
ested
n
e
w
s
ar
ticles
w
it
h
h
id
d
en
f
ea
t
u
r
e
m
o
d
el.
Œe
en
d
r
es
u
lt
s
h
o
w
a
co
n
s
id
er
ab
le
d
ev
elo
p
m
e
n
t
co
m
p
ar
ed
w
it
h
t
h
e
co
n
v
e
n
tio
n
al
w
o
r
d
-
b
ased
ap
p
r
o
ac
h
.
Œe
s
y
s
te
m
h
as
b
ee
n
f
u
ll
y
d
ep
lo
y
ed
to
o
n
lin
e
co
n
s
tr
u
c
tio
n
s
er
v
ices a
n
d
h
elp
in
g
o
v
er
te
n
m
i
llio
n
d
is
tin
c
tiv
e
u
s
er
’
s
e
v
e
r
y
d
a
y
.
I
n
[
1
1
]
,
au
th
o
r
p
lan
n
ed
a
s
eq
u
en
tial
D
SS
M
m
o
d
el
w
h
i
ch
in
co
r
p
o
r
ates
R
NNs
i
n
to
DSSM
f
o
r
s
u
g
g
e
s
tio
n
s
.
B
ased
o
n
co
n
v
en
t
io
n
al
DSSM,
T
DSSM
p
u
t
b
ac
k
t
h
e
le‰
n
et
w
o
r
k
w
ith
p
o
in
t
s
tatic
ch
ar
ac
ter
is
tic
s
,
an
d
th
e
r
ig
h
t
n
et
w
o
r
k
w
it
h
t
w
o
s
u
b
-
n
et
w
o
r
k
s
to
m
o
d
elin
g
clien
t
s
t
atic
ch
ar
ac
ter
is
tic
s
(
w
it
h
M
L
P
)
an
d
clien
t seq
u
en
t
ial
f
ea
t
u
r
es (
w
it
h
R
NN
s
)
.
I
n
[
1
2
]
,
au
th
o
r
p
r
o
j
ec
te
d
an
in
co
r
p
o
r
ated
s
tr
u
ctu
r
e
w
i
th
C
N
Ns
an
d
R
NN
s
f
o
r
m
o
d
i
f
ied
k
e
y
s
tr
u
ct
u
r
e
(
in
v
id
eo
s
)
s
u
g
g
esti
o
n
,
in
w
h
ich
C
NN
s
ar
e
u
tili
ze
d
to
s
tu
d
y
c
h
ar
ac
ter
is
tic
d
e
m
o
n
s
tr
ati
o
n
s
f
r
o
m
k
e
y
f
r
a
m
e
i
m
a
g
es a
n
d
R
NNs a
r
e
u
tili
ze
d
to
p
r
ac
tice
th
e
tex
tu
a
l c
h
ar
ac
te
r
is
tics
.
3.
P
RO
P
O
SE
D
M
E
T
H
O
DO
L
O
G
Y
T
h
e
co
m
p
lete
w
o
r
k
i
s
b
r
o
ad
l
y
d
i
v
id
ed
in
t
w
o
p
h
a
s
es.
T
h
e
f
ir
s
t
p
h
ase
is
d
ev
elo
p
in
g
o
n
to
lo
g
y
b
y
u
s
i
n
g
W
eb
C
o
n
te
n
t
(
W
C
)
an
d
W
eb
lo
g
(
W
L
)
f
ea
tu
r
e
s
as
s
h
o
w
n
u
s
i
n
g
Fi
g
u
r
e
1
.
Seco
n
d
p
h
ase
is
tes
tin
g
w
h
er
e
in
p
u
t
is
u
s
er
s
’
r
ec
e
n
t
v
is
ited
w
eb
p
a
g
es
a
n
d
o
n
to
lo
g
y
.
So
o
u
tp
u
t
o
f
s
ec
o
n
d
m
o
d
el
is
p
r
ed
icted
w
eb
p
a
g
e.
T
h
e
b
lo
ck
d
iag
r
am
o
f
p
r
o
p
o
s
ed
L
W
P
P
-
B
OA
m
o
d
el
is
as
s
h
o
w
n
i
n
Fi
g
u
r
e
2
.
T
ab
le
1
d
e
s
cir
b
es
th
e
s
y
m
b
o
ls
u
s
ed
in
t
h
i
s
r
esear
ch
p
ap
er
.
Data
s
et
p
r
e
-
p
r
o
ce
s
s
i
n
g
T
h
er
e
ar
e
m
an
y
d
i
g
ital
f
ea
t
u
r
es
o
f
t
h
e
w
eb
p
o
r
tal.
I
n
o
r
d
er
to
p
r
ed
ict
th
e
n
ex
t
w
eb
p
a
g
e,
th
e
p
r
o
p
o
s
ed
m
o
d
el
w
o
r
k
s
o
n
w
eb
co
n
te
n
t
w
h
ich
is
i
n
t
h
e
f
o
r
m
w
r
itte
n
te
x
t
an
d
w
eb
lo
g
is
in
f
o
r
m
o
f
u
s
er
p
r
ev
io
u
s
s
eq
u
en
ce
o
f
ac
tio
n
s
.
So
r
a
w
d
ataset
n
ee
d
to
b
e
p
r
e
-
p
r
o
ce
s
s
ed
f
ir
s
t,
b
y
r
e
m
o
v
in
g
t
h
e
n
o
is
y
d
ata.
W
eb
co
n
ten
t p
r
e
-
p
r
o
ce
s
s
in
g
As
te
x
t
co
n
te
n
t
o
n
w
eb
p
a
g
e
h
a
v
e
w
o
r
d
s
w
h
ic
h
n
ee
d
p
r
e
-
p
r
o
ce
s
s
in
g
b
y
r
e
m
o
v
i
n
g
s
to
p
-
w
o
r
d
s
.
So
s
et
o
f
s
to
p
-
w
o
r
d
s
ar
e
r
em
o
v
ed
an
d
f
il
ter
ed
w
o
r
d
s
ar
e
f
u
r
th
er
p
r
o
ce
s
s
ed
to
co
llect
w
eb
k
e
y
w
o
r
d
s
.
Fu
t
h
er
,
ea
ch
w
eb
p
ag
e
h
a
s
its
o
w
n
s
et
o
f
k
e
y
w
o
r
d
s
th
at
d
ep
en
d
s
o
n
t
y
p
e
o
f
co
n
ten
t
p
r
esen
t
in
t
h
e
w
eb
p
ag
e.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
A
r
ti
f
I
n
tell
I
SS
N:
2252
-
8938
Bi
o
g
eo
g
r
a
p
h
y
o
p
timiz
a
tio
n
a
l
g
o
r
ith
m
b
a
s
ed
n
ex
t web
p
a
g
e
p
r
ed
ictio
n
u
s
in
g
.
.
.
(
R
o
s
h
a
n
A
.
Ga
n
g
u
r
d
e
)
329
T
h
er
e
m
a
y
b
e
ch
a
n
ce
s
o
f
co
m
m
o
n
k
e
y
w
o
r
d
s
t
h
at
m
a
y
e
x
is
t
b
et
w
ee
n
p
a
g
es.
So
let
p
ag
es
b
e
P
1
,
P
2
,
P
m
,
w
h
er
e
m
is
to
tal
n
u
m
b
er
o
f
p
a
g
es i
n
th
e
s
ite.
No
w
P
m
p
ag
e
h
as
w
eb
co
n
te
n
t {
w
1
,
w
2
,
s
1
,
w
3
,
s
2
,
s
1
…….
w
n
}
n
is
to
tal
n
u
m
b
er
o
f
w
o
r
d
s
i
n
P
m
p
a
g
e.
A
f
ter
r
e
m
o
v
al
o
f
s
t
o
p
-
w
o
r
d
s
{s1
,
s
2
,
….
.
},
i
m
p
o
r
tan
t
w
o
r
d
s
w
ill
b
e
{
w
1
,
w
2
,
w
3
,
…….
w
n
}
[
1
3
]
.
Fin
all
y
to
g
et
k
e
y
w
o
r
d
s
,
T
er
m
Fre
q
u
en
c
y
i
s
ev
a
lu
ated
f
r
o
m
th
e
av
ai
lab
le
w
o
r
d
s
an
d
th
e
w
o
r
d
s
w
h
ic
h
cr
o
s
s
m
i
n
i
m
u
m
f
r
eq
u
en
c
y
ac
t
as
k
e
y
wo
r
d
s
o
f
th
at
p
ag
e.
Her
e,
s
eq
u
en
ce
o
f
w
o
r
d
s
o
f
t
h
e
w
eb
p
ag
e
d
o
es n
o
t a
f
f
ec
t t
h
e
p
r
e
-
p
r
o
ce
s
s
in
g
o
u
tco
m
es.
Fig
u
r
e
1
.
B
lo
ck
Diag
r
a
m
o
f
D
ev
elo
p
in
g
On
to
lo
g
y
T
ab
le
1
.
Sy
m
b
o
l
a
n
n
o
tat
io
n
ta
b
le
S
y
mb
o
l
u
se
d
M
e
a
n
i
n
g
o
f
S
y
mb
o
l
U
i
i
th
W
e
b
U
se
r
P
m
m
th
W
e
b
P
a
g
e
K
t
t
th
W
e
b
P
a
g
e
K
e
y
w
o
r
d
WL
W
e
b
l
o
g
AR
r
r
th
A
sso
c
i
a
t
i
o
n
R
u
l
e
F
F
C
F
e
e
d
F
o
r
w
a
r
d
C
o
u
n
t
e
r
SC
S
e
t
C
o
u
n
t
e
r
L
o
g
i
st
i
c
P
r
o
b
a
b
i
l
i
t
y
H
H
a
b
i
t
a
t
λ
Emi
g
r
a
t
i
o
n
R
a
t
e
α
I
mm
i
g
r
a
t
i
o
n
R
a
t
e
M
u
t
a
t
i
o
n
P
r
o
b
a
b
i
l
i
t
y
H
P
o
p
u
l
a
t
i
o
n
S
i
z
e
P
r
e
-
p
r
o
ce
s
s
in
g
o
f
w
eb
lo
g
I
n
th
is
s
tep
,
u
n
n
ec
es
s
ar
y
co
l
u
m
n
s
f
r
o
m
t
h
e
W
eb
lo
g
d
ataset
ar
e
r
em
o
v
ed
ex
ce
p
t
I
P
a
d
d
r
ess
an
d
ac
ce
s
s
UR
L
r
eq
u
es
t.
Her
e
ea
ch
v
i
s
ited
p
ag
e
is
id
en
ti
f
ied
b
y
a
u
n
iq
u
e
p
ag
e
n
u
m
b
er
.
So
W
eb
lo
g
W
L
h
as
a
s
et
o
f
v
i
s
ited
p
ag
es (
P
1
, P
2
,
P
3
,
…
P
m
)
o
f
a
s
in
lg
e
u
s
er
U
i
.
Featu
r
e
g
e
n
er
atio
n
f
r
o
m
w
eb
l
o
g
an
d
w
eb
co
n
ten
t
R
u
le
s
o
b
tain
ed
f
r
o
m
w
eb
lo
g
ar
e
g
en
er
ated
b
y
s
etti
n
g
th
e
co
u
n
ter
f
o
r
all
s
et
o
f
p
atter
n
s
f
r
o
m
th
e
p
ag
es [
1
4
]
.
So
f
o
r
n
p
ag
es,
s
et
o
f
co
u
n
ter
s
SC
i
s
r
ep
r
esen
ted
b
y
(
1
)
.
=
(
+
1
)
2
(
1
)
Hen
ce
,
f
o
r
n
=3
,
SC
=6
m
ea
n
s
SC
={
(
1
)
,
(
2
)
,
(
3
)
,
(
1
,
2
)
,
(
1
,
3
)
,
(
1
,
2
,
3
)
}
s
o
ca
r
d
in
alit
y
o
f
|
S
C
|
is
6
.
Ass
o
ciatio
n
r
u
le
g
en
er
atio
n
b
y
ab
o
v
e
co
u
n
ter
tech
n
iq
u
e
r
ed
u
ce
d
n
u
m
b
er
o
f
d
atase
t
s
ca
n
t
o
o
n
e
p
ass
[
1
5
]
.
Hen
ce
th
is
r
u
le
g
e
n
er
atio
n
is
ter
m
ed
as Fee
d
Fo
r
w
ar
d
C
o
u
n
ter
(
FF
C
)
.
AR
FF
C
(
W
L
,
S
C
)
(
2
)
Si
m
i
lar
it
y
m
atr
i
x
(
SM)
Ke
y
w
o
r
d
s
o
b
tain
ed
f
r
o
m
th
e
p
ag
es
af
ter
p
r
e
-
p
r
o
ce
s
s
in
g
ar
e
ar
r
an
g
ed
in
m
a
tr
ix
w
h
er
e
n
u
m
b
er
o
f
r
o
w
an
d
co
lu
m
n
ar
e
eq
u
al
to
n
u
m
b
er
o
f
p
ag
es.
So
m
atr
i
x
SM
m
x
m
h
a
v
e
ce
ll
v
al
u
es
w
h
i
c
h
i
s
t
h
e
co
u
n
t
o
f
s
i
m
ila
r
w
o
r
d
s
b
et
w
ee
n
2
p
a
g
es
in
r
o
w
an
d
co
l
u
m
n
.
Fo
r
ex
a
m
p
le,
p
ag
e
P
1
{
K
1
,
K
2
,
K
3
}
a
n
d
P
2
{
K
4
,
K
1
,
K
5
}
h
av
e
s
i
m
ilar
it
y
v
al
u
e
1
as k
e
y
w
o
r
d
K
1
is
co
m
m
o
n
i
n
b
o
th
th
e
p
a
g
es.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
I
n
t J
A
r
ti
f
I
n
tell
,
Vo
l.
9
,
No
.
2
,
J
u
n
e
20
20
:
327
–
3
3
5
330
L
o
g
i
s
tic
r
eg
r
e
s
s
io
n
L
HS
p
ag
e
p
atter
n
s
o
f
t
h
e
As
s
o
ciatio
n
r
u
les
ar
e
co
llected
to
g
en
er
ate
t
h
e
r
eg
r
e
s
s
io
n
v
a
lu
e
as
th
e
p
r
o
b
a
b
ilit
y
o
f
R
HS
p
ag
e
to
b
e
f
o
u
n
d
.
T
h
is
ca
n
b
e
u
n
d
er
s
ta
n
d
as
f
o
llo
w
s
:
C
o
n
s
id
er
th
e
r
u
l
es,
P
1
,
P
5
P
10
an
d
P
1
,
P
5
P
12
.
B
o
th
th
ese
r
u
le
s
h
a
v
e
co
m
m
o
n
s
et
o
f
L
HS
p
ag
e
p
atter
n
(
P
1
,
P
5
)
an
d
d
if
f
er
en
t
R
H
S
p
atter
n
.
So
r
eg
r
ess
io
n
w
i
ll
g
e
n
er
ate
p
r
o
b
a
b
ilit
y
o
f
n
e
x
t
p
ag
e
b
y
u
s
i
n
g
co
n
f
id
en
ce
v
alu
e
o
f
th
e
r
u
le
a
n
d
s
i
m
ilar
it
y
m
atr
i
x
.
So
in
p
u
t
to
r
eg
r
ess
io
n
is
o
n
e
f
ea
t
u
r
e
w
h
i
ch
is
th
e
m
u
ltip
licatio
n
o
f
s
i
m
ilar
it
y
m
a
tr
ix
an
d
co
n
f
id
e
n
ce
v
al
u
e.
I
n
ab
o
v
e
ex
a
m
p
le,
t
w
o
c
ate
g
o
r
ies
w
er
e
f
o
u
n
d
,
p
ag
e
P
10
an
d
P
12.
So
in
p
u
t to
th
e
lo
g
is
tic
f
u
n
ctio
n
is
C
1
=Co
n
f
(
P
1
,P
5
P
10
)
,
C
2
=Co
n
f
(
P
1
,P
5
P
12
)
T
h
e
in
p
u
t p
ar
a
m
eter
f
o
r
r
eg
r
es
s
io
n
ca
lc
u
latio
n
i
s
s
h
o
w
n
w
i
th
th
e
h
elp
o
f
e
x
a
m
p
le
a
s
s
h
o
w
n
in
T
ab
le
2
.
T
ab
le
2
.
I
n
p
u
t p
ar
am
e
ter
s
f
o
r
r
eg
r
e
s
s
io
n
ca
lcu
latio
n
C
a
t
e
g
o
r
y
1
C
a
t
e
g
o
r
y
2
(
X
)
P
10
(
X
)
P
12
S
M
[
P
1,
P
10
]
x
C
1
S
M
[
P
1,
P
12
]
x
C
2
S
M
[
P
5
P
10
]
x
C
1
S
M
[
P
5
P
12
]
x
C
2
S
M
[
P
1,
P
5
]
x
C
1
S
M
[
P
1,
P
5
]
x
C
2
Ou
tp
u
t
o
f
th
i
s
lo
g
i
s
tic
f
u
n
ctio
n
g
i
v
es
a
n
in
ter
ce
p
t
β
o
a
n
d
β
1
.
Fin
all
y
b
y
p
u
tti
n
g
v
al
u
es
o
f
p
r
ed
icto
r
s
,
o
n
e
ca
n
g
et
p
r
o
b
ab
ilit
y
v
al
u
es a
s
s
h
o
wn
in
(
3
)
:
=
1
1
+
−
(
+
1
×
1
+
2
×
2
)
(
3
)
Fil
ter
r
u
le
T
h
o
s
e
r
u
les
w
h
ich
h
a
v
e
v
er
y
l
o
w
s
u
p
p
o
r
t
v
alu
e
s
s
h
o
u
ld
b
e
f
ilter
ed
f
r
o
m
t
h
e
g
e
n
er
ated
r
u
le
s
,
as
th
e
s
e
r
u
les
ac
t
as
n
o
is
e.
Hen
ce
o
n
e
th
r
es
h
o
ld
v
alu
e
n
ee
d
s
to
b
e
s
et
as
p
er
d
ataset
r
u
les.
Ass
o
cia
tio
n
r
u
les
cr
o
s
s
in
g
m
i
n
i
m
u
m
s
u
p
p
o
r
t v
alu
e
ar
e
u
s
ed
f
o
r
p
ag
e
p
r
ed
ictio
n
in
g
en
e
t
ic
alg
o
r
ith
m
.
T
esti
n
g
p
h
a
s
e
Her
e,
th
e
d
ataset
is
a
g
ai
n
p
r
ep
r
o
ce
s
s
ed
f
o
r
th
e
w
eb
lo
g
p
o
r
tio
n
in
o
r
d
er
to
g
et
th
e
lo
g
s
t
h
at
ar
e
u
s
ed
f
o
r
test
i
n
g
th
e
b
u
ilt
m
o
d
el.
P
r
e
-
P
r
o
ce
s
s
in
g
s
tep
s
ar
e
s
i
m
ilar
as
d
o
n
e
in
p
r
ev
io
u
s
s
tep
s
o
f
m
o
d
el.
T
h
e
o
n
l
y
d
i
f
f
er
en
ce
h
er
e
is
th
at
p
r
e
-
p
r
o
ce
s
s
ed
lo
g
s
ar
e
b
r
ea
k
s
u
c
h
t
h
at
L
H
S
p
ag
e
s
o
f
th
e
ass
o
ciatio
n
r
u
le
ar
e
in
th
e
te
s
ti
n
g
p
ar
t a
n
d
th
e
R
H
S p
ag
e
o
f
th
e
a
s
s
o
ciatio
n
r
u
le
i
s
u
s
ed
f
o
r
th
e
ev
al
u
atio
n
o
f
r
es
u
lt.
B
io
g
eo
g
r
ap
h
y
o
p
ti
m
izatio
n
al
g
o
r
ith
m
(
B
OA
)
Sp
ec
ies
in
n
a
tu
r
e
ad
o
p
t
ch
an
g
es
as
p
er
s
u
itab
le
en
v
ir
o
n
m
en
tal
co
n
d
itio
n
s
.
So
ch
an
g
e
o
f
h
ab
it
is
o
n
e
t
y
p
e
o
f
c
h
a
n
g
e
ad
o
p
t
b
y
s
p
ec
ies
ti
m
e
to
t
i
m
e.
B
ased
o
n
th
is
,
a
u
th
o
r
p
r
o
p
o
s
ed
a
m
ath
e
m
at
ical
alg
o
r
ith
m
in
ea
r
l
y
1
9
6
0
,
w
h
e
r
ein
t
h
e
m
ai
n
co
n
ce
r
n
o
f
t
h
is
m
o
d
el
i
s
to
u
n
d
er
s
ta
n
d
t
h
e
m
i
g
r
atio
n
o
f
s
p
ec
ies
f
r
o
m
o
n
e
h
ab
itat
to
a
n
o
th
er
[
1
6
]
.
B
i
o
g
eo
g
r
ap
h
y
w
as
a
tr
en
d
in
g
r
esear
c
h
ar
ea
at
th
at
ti
m
e.
So
in
2
0
0
8
,
au
th
o
r
p
r
o
p
o
s
ed
a
g
en
er
alize
d
g
en
e
tic
al
g
o
r
ith
m
to
r
eso
l
v
e
s
i
m
ilar
t
y
p
e
o
f
is
s
u
e
s
[
1
7
]
.
So
m
e
o
f
b
asic
ter
m
s
r
elate
d
to
th
is
w
o
r
k
ar
e:
Hab
itat su
i
tab
ilit
y
i
n
d
ex
(
H
SI)
T
h
is
is
ter
m
ed
a
s
f
it
n
ess
v
a
lu
e
o
f
t
h
e
h
ab
itat,
m
ea
n
s
h
i
g
h
er
v
alu
e
s
h
o
w
s
t
h
at
p
o
o
r
p
lace
to
l
iv
e
w
h
il
e
lo
w
v
alu
e
m
ea
n
s
g
o
o
d
p
lace
t
o
liv
e
in
ter
m
s
o
f
r
eso
u
r
ce
s
,
li
f
e,
etc.
I
m
m
ig
r
atio
n
an
d
e
m
ig
r
atio
n
r
ate
So
m
e
o
f
b
asic te
r
m
s
o
f
I
m
m
ig
r
atio
n
λ
an
d
E
m
ig
r
atio
n
α
w
a
s
d
o
n
e
b
y
:
=
(
1
−
)
(
4
)
=
ℎ
(
5
)
w
h
er
e
R
is
r
a
n
k
o
f
h
ab
itat i
n
t
er
m
s
o
f
H
SI
v
al
u
e,
w
h
i
le
h
is
t
o
tal
n
u
m
b
er
o
f
h
ab
itat
s
.
Gen
er
ate
h
ab
itat
s
P
o
s
s
ib
le
s
et
o
f
s
o
l
u
tio
n
s
w
h
ic
h
ar
e
t
er
m
ed
as
h
ab
itat
in
th
is
alg
o
r
ith
m
ar
e
g
e
n
er
ated
in
t
h
i
s
s
tep
.
E
ac
h
h
ab
itat
is
s
et
o
f
p
o
s
s
ib
le
p
a
g
es
o
b
tai
n
ed
f
r
o
m
as
s
o
ciatio
n
r
u
le
w
h
er
e
lef
t
s
id
e
p
ag
es
o
f
t
h
e
r
u
le
ar
e
th
e
v
is
i
ted
p
ag
es
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
A
r
ti
f
I
n
tell
I
SS
N:
2252
-
8938
Bi
o
g
eo
g
r
a
p
h
y
o
p
timiz
a
tio
n
a
l
g
o
r
ith
m
b
a
s
ed
n
ex
t web
p
a
g
e
p
r
ed
ictio
n
u
s
in
g
.
.
.
(
R
o
s
h
a
n
A
.
Ga
n
g
u
r
d
e
)
331
o
f
th
e
u
s
er
.
Hen
ce
h
ab
itat
is
c
o
m
b
i
n
atio
n
o
f
H=
{P
1
,
P
5
,
P
m
},
w
h
er
e
p
o
p
u
latio
n
h
a
s
to
tal
h
n
u
m
b
er
o
f
h
ab
itat
s
.
Hen
ce
p
o
p
u
latio
n
g
en
er
atio
n
f
u
n
ct
io
n
i
n
th
i
s
al
g
o
r
ith
m
is
s
h
o
w
n
b
y
(
6
)
.
H
Hab
itat(
m
,
h
,
A
R
)
(
6
)
T
h
e
b
lo
ck
d
iag
r
a
m
o
f
t
h
e
p
r
o
p
o
s
ed
L
W
P
P
-
B
OA
(
L
o
g
is
t
ic
W
eb
P
ag
e
P
r
ed
ictio
n
b
y
B
io
g
eo
g
r
ap
h
y
Op
ti
m
izatio
n
A
l
g
o
r
ith
m
)
m
o
d
el
is
s
h
o
w
n
i
n
F
i
g
u
r
e
2
.
Fig
u
r
e
2
.
B
lo
ck
d
iag
r
a
m
o
f
p
r
o
p
o
s
ed
L
W
PP
-
B
OA
m
o
d
el.
Fit
n
e
s
s
f
u
n
ctio
n
Hab
itat
Su
i
tab
ilit
y
I
n
d
ex
(
HSI
)
o
f
an
y
h
ab
itat
d
ep
en
d
s
o
n
t
h
e
s
u
m
m
atio
n
o
f
lo
g
i
s
tic
v
a
lu
e
s
o
b
tain
ed
f
r
o
m
t
h
e
o
n
to
lo
g
y
s
to
r
ed
d
u
r
i
n
g
tr
ai
n
in
g
o
f
t
h
e
m
o
d
el.
Hen
ce
r
an
k
i
o
f
t
h
e
h
ab
itat
d
ep
en
d
o
n
t
h
e
l
o
g
is
tic
p
r
o
b
ab
ilit
y
v
alu
e
L
p
o
f
ea
ch
p
r
ed
icted
p
ag
e
as p
er
v
is
ited
p
ag
es
f
o
r
test
i
n
g
lo
g
.
HSI
=
R
an
k
(
,
,
H)
C
r
o
s
s
o
v
er
E
m
i
g
r
atio
n
o
f
p
a
g
es
i
n
t
h
e
f
o
r
m
o
f
s
p
ec
ies
f
r
o
m
o
n
e
h
ab
itat
to
o
th
er
d
ep
en
d
s
o
n
e
m
i
g
r
atio
n
r
ate.
W
h
ile
p
er
m
i
tti
n
g
th
e
s
p
ec
ies to
e
n
ter
in
a
h
ab
itat d
ep
en
d
s
o
n
i
m
m
i
g
r
atio
n
r
ate.
He
n
ce
f
o
r
cr
o
s
s
o
v
er
f
r
o
m
o
n
e
h
ab
itat
to
o
th
er
,
b
o
th
ty
p
e
o
f
r
ate
n
ee
d
to
f
in
d
.
So
cr
o
s
s
o
v
er
d
ep
en
d
s
o
n
f
o
llo
w
i
n
g
co
n
d
itio
n
.
L
o
o
p
x
=1
:h
I
f
C
r
o
s
s
o
v
er
_
L
i
m
it>
L
o
o
p
y
=1
:
h
I
f
C
r
o
s
s
o
v
er
_
L
i
m
it>
m
R
a
n
d
(
)
H[
x
,
m
]
H[
y
,
m
]
E
n
d
I
f
E
n
d
L
o
o
p
E
n
d
I
f
E
n
d
L
o
o
p
W
h
er
e
C
r
o
s
s
o
v
er
_
L
i
m
it
is
r
an
d
o
m
n
u
m
b
er
r
an
g
e
b
et
wee
n
0
-
1
,
x
an
d
y
is
h
ab
ita
t
p
o
s
itio
n
s
p
ec
i
f
y
i
m
m
i
g
r
atio
n
,
e
m
ig
r
atio
n
o
p
er
atio
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
I
n
t J
A
r
ti
f
I
n
tell
,
Vo
l.
9
,
No
.
2
,
J
u
n
e
20
20
:
327
–
3
3
5
332
Mu
tatio
n
I
n
th
i
s
w
o
r
k
,
af
ter
cr
o
s
s
o
v
er
,
m
u
tatio
n
i
s
also
p
er
f
o
r
m
ed
.
So
ch
an
ce
s
o
f
n
e
w
s
o
l
u
tio
n
s
g
e
t
in
cr
ea
s
es.
Fo
r
th
i
s
,
au
th
o
r
h
a
s
in
v
o
l
v
ed
m
u
tatio
n
p
r
o
b
a
b
ilit
y
w
h
er
e
m
u
tatio
n
i
s
p
er
f
o
r
m
ed
i
n
s
elec
ted
h
ab
ita
ts
as p
er
HSI
v
alu
e.
=
(
ℎ
)
(
7
)
=
(
)
(
8
)
Hen
ce
h
ab
itat
w
h
ic
h
cr
o
s
s
e
s
a
co
n
s
ta
n
t
m
u
tatio
n
_
cr
o
s
s
_
li
m
it
w
h
ic
h
h
as
r
an
g
e
i
n
0
-
1
,
g
iv
e
s
a
m
u
tatio
n
r
an
k
f
o
r
th
e
h
ab
itat
as
p
er
HS
I
v
alu
e.
Hi
g
h
er
t
h
e
v
al
u
e
o
f
HSI
,
h
ig
h
er
w
ill
b
e
th
e
m
u
tat
io
n
r
an
k
.
He
n
ce
th
e
h
ab
itat
w
h
ich
h
as
h
ig
h
er
m
u
t
atio
n
r
an
k
w
ill
h
a
v
e
h
ig
h
er
Mu
tatio
n
P
r
o
b
a
b
ilit
y
.
So
h
ab
i
tat
w
h
ic
h
h
a
s
lo
w
er
Mu
tatio
n
P
r
o
b
ab
ilit
y
as c
o
m
p
ar
ed
to
Mu
tatio
n
_
C
r
o
s
s
_
T
h
r
esh
o
ld
u
n
d
er
g
o
e
s
to
m
u
tatio
n
.
Fin
al
So
lu
tio
n
I
n
th
i
s
w
o
r
k
,
a
f
ter
s
u
f
f
icie
n
t
n
u
m
b
er
o
f
iter
atio
n
s
,
b
est
p
o
s
s
ib
le
h
ab
itats
ar
e
o
b
tain
ed
an
d
s
et
o
f
th
o
s
e
p
ag
es
ar
e
r
ec
o
m
m
en
d
ed
p
ag
es b
y
t
h
e
p
r
o
p
o
s
ed
m
o
d
el
o
f
B
io
g
eo
g
r
ap
h
y
Op
ti
m
izatio
n
A
lg
o
r
it
h
m
.
P
r
o
p
o
se
d
A
l
g
o
r
i
t
h
m
I
n
p
u
t
:
W
F
/
/
W
F
:
W
e
b
F
e
a
t
u
r
e
s
O
u
t
p
u
t
:
N
P
P
/
/
N
P
P
:
N
e
x
t
P
r
e
d
i
c
t
e
d
P
a
g
e
1.
[
W
L
W
C
]
Pre
-
P
r
o
c
e
ssi
n
g
(
W
F
)
2.
=
(
+
1
)
2
/
/
n
n
u
m
b
e
r
o
f
p
a
g
e
s i
n
W
L
3.
AR
F
F
C
(
W
L
,
S
C
)
4.
SM
S
i
mi
l
a
r
i
t
y
_
M
a
t
r
i
x
(
W
C
)
5.
L
o
o
p
1
:
r
/
/
r
:
n
u
mb
e
r
o
f
r
u
l
e
s
6.
L
p
[
r
]
L
o
g
i
st
i
c
_
R
e
g
r
e
ssi
o
n
(
A
R
,
S
M
)
7.
En
d
L
o
o
p
8.
T
l
U
se
r
_
V
i
si
t
e
d
_
P
a
g
e
/
/
T
e
st
i
n
g
o
f
M
o
d
e
l
9.
I
n
i
t
i
a
l
i
z
e
B
O
A
P
a
r
a
me
t
e
r
s m,
h
,
λ
,
α
1
0
.
H
G
e
n
e
r
a
t
e
_
H
a
b
i
t
a
t
e
(
m,
h
,
A
R
)
1
1
.
L
o
o
p
1
:
i
t
e
r
a
t
i
o
n
1
2
.
H
S
I
=
R
a
n
k
(
,
,
H
)
1
3
.
H
C
r
o
sso
v
e
r
(
H
,
λ
,
α)
1
4
.
H
M
u
t
a
t
i
o
n
(
H
,
H
S
I
,
h
)
1
5
.
En
d
L
o
o
p
1
6
.
H
S
I
=
R
a
n
k
(
,
,
H
)
1
7
.
N
P
P
M
a
x
(
H
S
I
)
4.
E
XP
E
R
I
M
E
NT
AND
R
E
SU
L
T
4.
1
.
Da
t
a
s
et
s
I
n
th
i
s
w
o
r
k
,
au
t
h
o
r
s
h
a
v
e
u
s
ed
r
ea
l
ti
m
e
d
ataset
f
r
o
m
“
P
r
o
j
ec
t
T
u
n
n
e
l”
w
e
b
s
ite
[
1
8
]
.
T
h
e
d
ataset
co
n
tain
s
w
eb
lo
g
f
o
r
th
e
m
o
n
t
h
of
A
p
r
il
2
0
1
9
.
T
h
is
w
eb
lo
g
h
a
s
2
0
,
0
0
0
s
ess
io
n
s
o
f
6
,
2
4
0
u
s
er
s
.
Nu
m
b
er
o
f
p
a
g
es i
n
th
e
w
eb
s
it
e
is
2
7
8
.
4.
2
.
E
x
peri
m
ent
s
et
u
p
E
x
p
er
i
m
e
n
tal
s
e
t
u
p
w
a
s
d
ev
elo
p
ed
o
n
MA
T
L
A
B
s
o
f
t
w
a
r
e
w
h
er
e
n
u
m
b
er
o
f
in
b
u
i
lt
f
u
n
ctio
n
in
cr
ea
s
es
th
e
ea
s
i
n
ess
o
f
i
m
p
le
m
e
n
tatio
n
.
C
o
m
p
ar
is
o
n
o
f
p
r
o
p
o
s
ed
m
o
d
el
i
s
d
o
n
e
o
n
t
h
e
b
asis
o
f
P
ASO
alg
o
r
ith
m
p
r
o
p
o
s
ed
in
[
1
9
]
,
T
er
m
NetW
P
alg
o
r
ith
m
p
r
o
p
o
s
ed
in
[
1
2
]
,
W
P
P
M
(
W
eb
P
ag
e
P
r
ed
i
ctio
n
Mo
d
el)
m
o
d
el
w
h
er
e
o
n
l
y
w
eb
lo
g
f
ea
tu
r
e
is
u
s
ed
w
it
h
P
A
P
SO
alg
o
r
ith
m
[
1
3
]
,
L
W
PP
M
(
L
o
g
i
s
tic
W
eb
P
ag
e
P
r
ed
ictio
n
Mo
d
el)
m
o
d
el
w
i
th
w
eb
lo
g
an
d
w
eb
co
n
te
n
t
f
ea
t
u
r
es
u
s
in
g
lo
g
is
tic
r
eg
r
ess
io
n
an
d
P
A
SO
alg
o
r
ith
m
.
E
v
alu
a
tio
n
P
ar
a
m
eter
s
:
P
r
ec
is
io
n
P
r
ec
is
io
n
=
A
p
p
r
o
x
i
m
a
te_
C
o
r
r
ec
t_
p
ag
es /
A
l
l_
p
r
ed
ictio
n
s
C
o
v
er
ag
e
C
o
v
er
ag
e
=
A
p
p
r
o
x
i
m
ate_
C
o
r
r
ec
t_
p
ag
es /
A
l
l_
Vis
ited
_
P
ag
e
s
M
-
m
etr
ic
M
-
m
etr
ic
=
(
2
x
P
r
ec
is
o
n
x
C
o
v
er
ag
e)
/ (
P
r
ec
is
io
n
+
C
o
v
er
ag
e)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
A
r
ti
f
I
n
tell
I
SS
N:
2252
-
8938
Bi
o
g
eo
g
r
a
p
h
y
o
p
timiz
a
tio
n
a
l
g
o
r
ith
m
b
a
s
ed
n
ex
t web
p
a
g
e
p
r
ed
ictio
n
u
s
in
g
.
.
.
(
R
o
s
h
a
n
A
.
Ga
n
g
u
r
d
e
)
333
E
x
ec
u
t
io
n
T
im
e:
T
o
tal
ti
m
e
r
eq
u
ir
ed
f
o
r
th
e
ex
ec
u
tio
n
o
f
th
e
alg
o
r
ith
m
f
o
r
p
r
ed
ictin
g
o
f
th
e
p
ag
e
b
ased
o
n
th
e
d
if
f
er
en
t sizes o
f
d
ataset.
Me
an
A
b
s
o
l
u
te
E
r
r
o
r
(
MA
E
)
is
av
er
a
g
e
d
i
f
f
er
e
n
ce
b
et
w
ee
n
th
e
p
r
ed
icted
p
ag
e
c
h
r
o
m
o
s
o
m
e
v
alu
e
to
ac
t
u
al
p
ag
e
ch
r
o
m
o
s
o
m
e
f
i
tn
e
s
s
v
alu
e.
=
∑
_
−
_
(
9
)
I
n
(
9
)
,
n
is
to
tal
n
u
m
b
er
o
f
p
ag
es
p
r
ed
icted
,
F_
P
r
ed
ict
j
i
s
th
e
j
th
p
r
ed
icted
p
ag
e
ch
r
o
m
o
s
o
m
e
v
al
u
e
an
d
F_
R
ea
l
j
is
t
h
e
f
itn
e
s
s
v
al
u
e
o
f
ch
r
o
m
o
s
o
m
e
o
f
p
ag
e
ac
t
u
all
y
o
p
en
ed
b
y
u
s
er
f
o
r
j
th
p
r
ed
icti
o
n
.
R
o
o
t M
ea
n
Sq
u
ar
e
E
r
r
o
r
(
R
MSE
)
h
ig
h
li
g
h
ts
t
h
e
s
u
p
er
io
r
d
e
v
iatio
n
s
.
=
√
∑
_
−
_
(
10
)
4
.
3
.
Resul
t
T
ab
le
3
s
h
o
w
s
t
h
at
t
h
e
p
r
o
p
o
s
ed
L
W
P
P
-
B
OA
m
o
d
el
h
as
p
er
f
ec
tl
y
u
t
ilized
w
eb
lo
g
an
d
w
eb
co
n
ten
t
f
ea
t
u
r
es
w
i
th
lo
g
i
s
tic
r
eg
r
ess
io
n
r
elatio
n
,
h
e
n
ce
p
r
ec
i
s
io
n
v
al
u
e
g
e
t
i
n
cr
ea
s
es.
I
t
is
al
s
o
o
b
s
er
v
ed
th
at
u
s
e
o
f
B
io
g
eo
g
r
ap
h
y
O
p
ti
m
izatio
n
A
l
g
o
r
ith
m
f
o
r
p
ag
e
p
r
ed
ictio
n
g
i
v
es
b
etter
r
esu
lt
s
a
s
co
m
p
ar
ed
to
T
e
r
m
NetW
P
[
1
2
]
an
d
P
A
SO [
1
9
]
m
o
d
els o
n
d
if
f
er
en
t d
at
aset sizes.
T
ab
le
3
.
P
r
ec
is
io
n
b
ased
co
m
p
ar
is
o
n
o
f
p
ag
e
r
ec
o
m
m
e
n
d
atio
n
m
o
d
els
T
e
st
i
n
g
D
a
t
a
se
t
S
i
z
e
P
e
r
c
e
n
t
a
g
e
L
W
P
P
-
B
O
A
L
W
P
P
M
W
P
P
M
(
W
i
t
h
o
u
t
C
o
n
t
e
n
t
F
e
a
t
u
r
e
)
[
1
3
]
P
A
S
O
[
1
9
]
T
e
r
mN
e
t
W
P
[
1
2
]
20
0
.
8
5
4
2
0
.
8
3
3
3
0
.
6
3
5
4
0
.
5
6
2
5
0
.
5
1
6
9
30
0
.
8
3
9
2
0
.
8
3
2
2
0
.
5
3
8
5
0
.
5
2
4
5
0
.
4
7
2
9
40
0
.
7
5
9
2
0
.
7
2
7
7
0
.
4
8
5
0
0
.
4
4
6
9
0
.
4
6
5
5
50
0
.
7
1
8
5
0
.
6
8
0
7
0
.
4
7
2
6
0
.
4
3
2
8
0
.
4
3
1
8
T
ab
le
4
s
h
o
w
s
t
h
at
u
s
e
o
f
B
O
A
al
g
o
r
ith
m
i
n
L
W
P
P
-
B
OA
r
ed
u
ce
s
th
e
e
x
ec
u
tio
n
ti
m
e
v
a
l
u
e,
as
cr
o
s
s
o
v
er
an
d
m
u
tatio
n
o
p
er
atio
n
s
d
ep
en
d
s
o
n
i
m
m
i
g
r
atio
n
,
e
m
ig
r
atio
n
a
n
d
m
u
tatio
n
r
ate
v
al
u
es.
So
it
r
ed
u
ce
d
th
e
n
u
m
b
er
o
f
o
p
er
a
tio
n
s
a
m
o
n
g
h
ab
itat
s
(
ch
r
o
m
o
s
o
m
es).
W
h
ile
i
n
S
w
ar
m
Op
ti
m
izatio
n
,
ea
ch
ite
r
atio
n
p
er
f
o
r
m
s
all
o
p
er
atio
n
in
ea
ch
s
et
o
f
ch
r
o
m
o
s
o
m
e.
T
ab
le
4
.
E
x
ec
u
tio
n
ti
m
e
(
s
ec
o
n
d
s
)
b
ased
co
m
p
ar
is
o
n
o
f
p
ag
e
r
ec
o
m
m
e
n
d
atio
n
m
o
d
els
T
e
st
i
n
g
D
a
t
a
se
t
S
i
z
e
P
e
r
c
e
n
t
a
g
e
L
W
P
P
-
B
O
A
LW
PPM
W
P
P
M
(
W
i
t
h
o
u
t
C
o
n
t
e
n
t
F
e
a
t
u
r
e
)
[
1
3
]
P
A
S
O
[
1
9
]
T
e
r
mN
e
t
W
P
[
1
2
]
20
3
.
2
1
1
0
.
3
4
2
0
.
3
5
2
2
.
3
6
4
.
4
0
6
6
30
3
.
5
6
1
5
.
6
3
2
7
.
6
8
3
0
.
5
2
4
.
2
0
3
40
4
.
6
7
2
1
.
0
6
3
3
.
2
3
3
7
.
2
1
5
.
2
1
4
5
50
6
.
9
8
2
5
.
7
1
4
2
.
5
6
4
7
.
4
5
6
.
2
4
6
5
Fro
m
T
ab
le
s
5
an
d
6
,
it
is
o
b
s
er
v
ed
th
at
p
r
o
p
o
s
ed
L
W
P
P
-
B
OA
m
o
d
el
h
as
p
er
f
ec
tl
y
u
til
iz
ed
w
eb
lo
g
an
d
w
eb
co
n
ten
t
f
ea
t
u
r
es
w
ith
lo
g
i
s
tic
r
eg
r
ess
io
n
r
elatio
n
h
e
n
ce
co
v
er
ag
e
v
al
u
e
g
et
i
n
cr
ea
s
e
s
.
I
t
i
s
also
o
b
s
er
v
ed
t
h
at
u
s
e
o
f
B
io
g
eo
g
r
ap
h
y
Op
ti
m
iz
atio
n
A
lg
o
r
it
h
m
f
o
r
w
eb
p
ag
e
p
r
ed
ictio
n
g
iv
es
b
etter
r
esu
lts
as
co
m
p
ar
ed
to
T
er
m
NetW
P
[
1
2
]
an
d
P
A
SO
[
1
9
]
m
o
d
els
o
n
d
if
f
er
en
t
d
ataset
s
izes.
T
h
u
s
in
cr
ea
s
e
i
n
p
r
ec
is
io
n
an
d
co
v
er
ag
e
u
lti
m
atel
y
i
n
cr
ea
s
es t
h
e
M
-
m
e
tr
ic
v
alu
e
a
s
w
ell.
T
ab
le
5
.
C
o
v
er
ag
e
b
ased
co
m
p
ar
is
o
n
o
f
p
ag
e
r
ec
o
m
m
en
d
ati
o
n
m
o
d
els
T
e
st
i
n
g
D
a
t
a
se
t
S
i
z
e
P
e
r
c
e
n
t
a
g
e
L
W
P
P
-
B
O
A
L
W
P
P
M
W
P
P
M
(
W
i
t
h
o
u
t
C
o
n
t
e
n
t
F
e
a
t
u
r
e
)
[
1
3
]
P
A
S
O
[
1
9
]
T
e
r
mN
e
t
W
P
[
1
2
]
20
0
.
4
3
1
6
0
.
4
2
1
1
0
.
1
3
4
7
0
.
1
1
9
2
0
.
2
4
2
1
30
0
.
4
2
2
5
0
.
4
1
9
0
0
.
1
0
9
5
0
.
1
0
6
7
0
.
2
1
4
8
40
0
.
3
8
1
6
0
.
3
6
5
8
0
.
0
8
5
8
0
.
0
9
3
8
0
.
2
1
3
2
50
0
.
3
6
0
8
0
.
3
4
1
8
0
.
0
9
5
8
0
.
0
8
0
9
0
.
2
0
0
4
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
I
n
t J
A
r
ti
f
I
n
tell
,
Vo
l.
9
,
No
.
2
,
J
u
n
e
20
20
:
327
–
3
3
5
334
T
ab
le
6
.
M
-
Me
tr
ic
b
ased
co
m
p
ar
is
o
n
o
f
p
ag
e
r
ec
o
m
m
en
d
ati
o
n
m
o
d
els
T
e
st
i
n
g
D
a
t
a
se
t
S
i
z
e
P
e
r
c
e
n
t
a
g
e
L
W
P
P
-
B
O
A
L
W
P
P
M
W
P
P
M
(
W
i
t
h
o
u
t
C
o
n
t
e
n
t
F
e
a
t
u
r
e
)
[
1
3
]
P
A
S
O
[
1
9
]
T
e
r
mN
e
t
W
P
[
1
2
]
20
0
.
5
7
3
4
0
.
5
5
9
4
0
.
2
2
2
2
0
.
1
9
6
7
0
.
3
2
9
7
30
0
.
5
6
2
1
0
.
5
5
7
4
0
.
1
8
2
0
0
.
1
7
7
3
0
.
2
9
5
4
40
0
.
5
0
7
9
0
.
4
8
6
9
0.
1
4
3
8
0
.
1
5
7
4
0
.
2
9
2
4
50
0
.
4
8
0
3
0
.
4
5
5
1
0
.
1
6
1
5
0
.
1
3
6
3
0
.
2
7
4
2
T
ab
le
7
an
d
T
ab
le
8
s
h
o
w
s
t
h
at
u
s
e
o
f
B
O
A
al
g
o
r
ith
m
i
n
L
W
PP
-
B
OA
r
ed
u
ce
d
M
A
E
a
n
d
R
M
A
E
v
al
u
e,
as
cr
o
s
s
o
v
er
an
d
m
u
tatio
n
o
p
er
atio
n
d
ep
en
d
s
o
n
i
m
m
i
g
r
atio
n
,
e
m
i
g
r
atio
n
a
n
d
m
u
ta
tio
n
r
ate
v
alu
e
s
.
He
n
ce
m
o
r
e
r
elev
an
t
p
ag
es
p
er
ch
r
o
m
o
s
o
m
es
ar
e
id
en
ti
f
ied
.
W
h
i
le
i
n
S
w
ar
m
Op
ti
m
izatio
n
,
ea
c
h
it
er
atio
n
p
er
f
o
r
m
al
l
o
p
er
atio
n
in
ea
ch
s
et
o
f
c
h
r
o
m
o
s
o
m
e
s
o
ch
a
n
ce
o
f
g
ett
in
g
r
elev
a
n
t
p
ag
e
g
et
r
ed
u
ce
s
,
as
ele
m
e
n
t
s
i
n
th
e
cr
o
s
s
o
v
er
ar
e
d
ep
en
d
o
n
f
itn
e
s
s
v
al
u
e.
T
ab
l
e
7
.
MA
E
b
ased
co
m
p
ar
is
o
n
o
f
p
ag
e
r
ec
o
m
m
en
d
atio
n
m
o
d
els
T
e
st
i
n
g
D
a
t
a
se
t
S
i
z
e
P
e
r
c
e
n
t
a
g
e
L
W
P
P
-
B
O
A
L
W
P
P
M
W
P
P
M
(
W
i
t
h
o
u
t
C
o
n
t
e
n
t
F
e
a
t
u
r
e
)
[
1
3
]
P
A
S
O
[
1
9
]
T
e
r
mN
e
t
W
P
[
1
2
]
20
0
.
1
1
1
6
0
.
2
3
4
8
0
.
2
3
3
7
0
.
2
8
8
4
1
.
2
30
0
.
1
1
6
9
0
.
1
8
6
4
0
.
2
7
2
8
0
.
2
8
8
3
1
.
1
8
3
1
40
0
.
1
1
7
1
0
.
3
0
9
4
0
.
3
0
0
5
0
.
2
5
1
9
0
.
9
6
8
4
50
0
.
2
1
2
5
0
.
2
7
8
2
0
.
2
6
8
1
0
.
3
8
3
6
0
.
7
9
7
5
T
ab
le
8
.
R
MA
E
b
ased
co
m
p
ar
is
o
n
o
f
p
ag
e
r
ec
o
m
m
e
n
d
atio
n
m
o
d
el
s
T
e
st
i
n
g
D
a
t
a
se
t
S
i
z
e
P
e
r
c
e
n
t
a
g
e
L
W
P
P
-
B
O
A
L
W
P
P
M
W
P
P
M
(
W
i
t
h
o
u
t
C
o
n
t
e
n
t
F
e
a
t
u
r
e
)
[
1
3
]
P
A
S
O
[
1
9
]
T
e
r
mN
e
t
W
P
[
1
2
]
20
0
.
3
3
4
1
0
.
4
5
9
5
0
.
4
8
3
4
0
.
5
3
7
0
1
.
0
9
5
4
30
0
.
3
4
2
0
0
.
4
3
1
7
0
.
5
2
2
3
0
.
5
3
7
0
1
.
0
8
7
7
40
0
.
3
4
2
1
0
.
5
5
6
3
0
.
5
4
8
2
0
.
5
0
1
9
0
.
9
8
4
50
0
.
4
6
1
0
0
.
5
2
7
4
0
.
5
1
7
8
0
.
6
1
9
4
0
.
8
9
3
T
h
e
r
esu
lt
s
ec
tio
n
h
as
s
h
o
w
n
t
h
at
p
r
o
p
o
s
ed
m
o
d
el
h
as
r
ed
u
ce
d
th
e
ex
ec
u
tio
n
ti
m
e
as
w
ell
a
s
i
m
p
r
o
v
ed
th
e
p
r
ec
is
io
n
o
f
n
e
x
t
w
eb
p
ag
e
p
r
ed
ictio
n
.
I
n
t
h
i
s
r
esear
ch
,
a
u
t
h
o
r
s
tr
ied
to
ef
f
ec
ti
v
el
y
u
t
ilize
t
h
e
w
eb
lo
g
f
ea
t
u
r
e
a
s
w
ell
as
w
e
b
co
n
ten
t
f
ea
t
u
r
es
w
h
ich
r
e
s
u
lted
in
to
r
ed
u
ce
d
M
A
E
an
d
R
MA
E
v
al
u
es.
Her
e
p
r
o
p
o
s
ed
m
o
d
el
h
as
o
v
er
co
m
e
li
m
itatio
n
s
o
f
p
r
ev
io
u
s
w
o
r
k
w
h
er
e
ass
o
ciatio
n
r
u
les
ta
k
e
lar
g
e
ti
m
e.
W
eb
co
n
ten
t
f
ea
t
u
r
e
u
tili
za
t
io
n
h
el
p
ed
to
i
m
p
r
o
v
e
th
e
r
e
g
r
es
s
io
n
m
o
d
el
as
p
r
ev
io
u
s
ap
p
r
o
ac
h
es
h
a
s
u
s
ed
s
e
m
a
n
tic
co
llectio
n
o
f
w
o
r
d
s
w
h
ic
h
r
ed
u
ce
d
th
e
w
o
r
k
e
f
f
ien
c
y
.
5.
CO
NCLU
SI
O
N
I
n
o
r
d
er
to
im
p
r
o
v
e
u
s
er
f
a
m
il
iar
it
y
o
f
w
eb
s
ite
s
,
n
u
m
b
e
r
o
f
p
ar
am
eter
s
is
a
v
ailab
le
b
u
t
ea
ch
o
f
t
h
e
m
h
as
its
o
w
n
li
m
ita
tio
n
,
s
o
r
esear
ch
er
f
o
cu
s
o
n
u
s
er
b
eh
av
io
r
.
Hen
ce
n
u
m
b
er
o
f
u
s
er
f
ea
t
u
r
es
is
p
r
o
p
o
s
ed
b
y
r
esear
ch
er
o
f
t
h
is
f
ield
w
h
ic
h
en
h
a
n
ce
d
th
e
w
eb
p
ag
e
p
r
ed
ictio
n
w
o
r
k
.
T
h
is
p
ap
er
h
as
also
in
v
o
l
v
ed
a
Mu
lti
m
o
d
al
L
o
g
is
t
ic
r
eg
r
es
s
i
o
n
m
et
h
o
d
f
o
r
d
ev
elo
p
in
g
a
co
n
cr
ete
f
ea
t
u
r
e
u
s
in
g
w
eb
c
o
n
ten
t
a
n
d
w
eb
lo
g
d
ataset.
A
s
t
h
is
f
in
d
a
p
r
o
b
ab
ilit
y
o
f
th
e
s
eq
u
en
ce
o
f
p
ag
es
as
p
er
u
s
er
s
’
p
ast
s
er
ies
o
f
v
is
its
i
n
w
eb
lo
g
a
n
d
t
y
p
e
o
f
tex
t
co
n
te
n
ts
p
r
esen
t
o
n
p
ag
es.
On
e
m
o
r
e
b
en
e
f
it
o
f
t
h
is
co
n
cr
ete
f
ea
t
u
r
e
o
f
lo
g
is
tic
r
eg
r
ess
io
n
is
h
ig
h
r
ed
u
ctio
n
in
e
x
ec
u
t
io
n
ti
m
e
o
f
p
r
o
p
o
s
ed
m
o
d
el.
Fu
r
th
er
r
an
d
o
m
ad
o
p
tio
n
o
f
p
ag
e
p
r
ed
ictio
n
is
m
ain
ta
in
ed
b
y
B
io
g
eo
g
r
ap
h
ic
Op
ti
m
izatio
n
A
l
g
o
r
ith
m
.
E
x
p
er
i
m
e
n
t
i
s
d
o
n
e
o
n
r
ea
l
ti
m
e
d
ata
s
et
o
f
p
r
o
j
ec
ttu
n
n
el
w
eb
s
ite.
C
o
m
p
ar
is
o
n
o
f
p
r
o
p
o
s
ed
m
o
d
el
is
d
o
n
e
w
it
h
e
x
is
ti
n
g
P
A
SO
m
et
h
o
d
an
d
it
is
o
b
s
er
v
ed
th
a
t p
r
o
p
o
s
ed
m
o
d
el
i
s
b
etter
as
it
en
h
an
ce
d
t
h
e
p
r
ed
ictio
n
ac
cu
r
ac
y
b
y
6
1
.
2
4
%,
r
e
d
u
ce
d
Me
an
A
v
er
a
g
e
E
r
r
o
r
b
y
5
3
.
9
5
%.
I
n
f
u
t
u
r
e,
p
ag
e
p
r
ed
ictio
n
alg
o
r
it
h
m
m
a
y
i
m
p
r
o
v
e
t
h
is
w
o
r
k
b
y
in
tr
o
d
u
cin
g
co
o
k
ie
s
d
ata
as
w
e
ll.
RE
F
E
R
E
NC
E
S
[1
]
Ko
sa
la,
Ra
y
m
o
n
d
,
a
n
d
He
n
d
rik
Blo
c
k
e
e
l.
“
W
e
b
m
in
in
g
re
s
e
a
rc
h
:
A su
rv
e
y
.
”
A
CM
S
ig
k
d
d
Ex
p
lo
ra
ti
o
n
s Ne
w
sle
tt
e
r
2
,
n
o
.
1
,
p
p
.
1
-
15
,
2
0
0
0
.
[2
]
S
h
a
rm
a
,
Ka
v
it
a
,
G
u
lsh
a
n
S
h
riv
a
sta
v
a
,
a
n
d
V
ik
a
s
Ku
m
a
r.
“
W
e
b
min
in
g
:
T
o
d
a
y
a
n
d
to
mo
rr
o
w
.
”
In
El
e
c
tro
n
ics
Co
m
p
u
ter T
e
c
h
n
o
lo
g
y
(ICECT
),
2
0
1
1
3
rd
In
tern
a
ti
o
n
a
l
Co
n
f
e
re
n
c
e
o
n
,
v
o
l.
1
,
p
p
.
3
9
9
-
4
0
3
.
IEE
E,
2
0
1
1
.
[3
]
S
riv
a
sta
v
a
,
J.,
R.
Co
o
ley
,
M
.
De
s
h
p
a
n
d
e
M
u
k
u
n
d
,
a
n
d
P
.
N.
T
a
n
.
“
W
e
b
u
sa
g
e
min
in
g
:
d
isc
o
v
e
ry
a
n
d
a
p
p
li
c
a
ti
o
n
o
f
u
sa
g
e
p
a
t
ter
n
s fro
m web
d
a
ta
.
”
P
ro
c
e
e
d
in
g
s o
f
S
IG
KD
D e
x
p
lo
ra
ti
o
n
s
1
,
n
o
.
2
,
2
0
0
2
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
A
r
ti
f
I
n
tell
I
SS
N:
2252
-
8938
Bi
o
g
eo
g
r
a
p
h
y
o
p
timiz
a
tio
n
a
l
g
o
r
ith
m
b
a
s
ed
n
ex
t web
p
a
g
e
p
r
ed
ictio
n
u
s
in
g
.
.
.
(
R
o
s
h
a
n
A
.
Ga
n
g
u
r
d
e
)
335
[4
]
E.
T
u
b
a
,
R.
Jo
v
a
n
o
v
ic,
R.
C
.
Hro
si
k
,
A
.
A
li
h
o
d
z
ic
a
n
d
M
.
T
u
b
a
,
“
W
e
b
In
telli
g
e
n
c
e
Da
t
a
Cl
u
ste
rin
g
b
y
B
a
re
Bo
n
e
Fi
re
wo
rk
s
Al
g
o
rith
m
C
o
mb
i
n
e
d
wit
h
K
-
M
e
a
n
s
”
.
In
P
r
o
c
e
e
d
in
g
s
o
f
th
e
8
th
In
tern
a
ti
o
n
a
l
Co
n
f
e
re
n
c
e
o
n
W
e
b
In
telli
g
e
n
c
e
,
M
in
in
g
a
n
d
S
e
m
a
n
ti
c
s,
(a
rti
c
le 7
).
A
CM
.
2
0
1
8
.
[5
]
De
e
p
ti
Ra
z
d
a
n
,
“
T
h
e
Ne
x
t
P
a
g
e
A
c
c
e
ss
P
re
d
icti
o
n
Us
in
g
M
a
k
o
v
M
o
d
e
l”,
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
tro
n
ic
s
Co
mm
u
n
ica
ti
o
n
a
n
d
Co
m
p
u
ter
T
e
c
h
n
o
l
o
g
y
(
IJ
ECCT
)
,
V
o
l
u
m
e
1
,
Iss
u
e
1
,
S
e
p
tem
b
e
r
2
0
1
1
,
I
S
S
N:
2
2
4
9
-
7
8
3
8
.
[6
]
M
.
Aw
a
d
,
L
.
Kh
a
n
,
a
n
d
B.
T
h
u
ra
isin
g
h
a
m
.
“
P
re
d
ictin
g
W
WW
su
rf
in
g
u
sin
g
m
u
lt
ip
le
e
v
id
e
n
c
e
c
o
m
b
in
a
ti
o
n
,
”
VL
DB
J
.
,
v
o
l
.
1
7
,
n
o
.
3
,
p
p
.
4
0
1
-
4
1
7
,
2
0
0
8
.
[7
]
A
.
L
a
d
e
k
a
r,
P
.
P
a
w
a
r,
D.
Ra
ik
a
r
a
n
d
J.
Ch
a
u
d
h
a
ri
,
“
W
e
b
L
o
g
Ba
se
d
A
n
a
l
y
sis
o
f
Us
e
r’s
Bro
ws
in
g
Be
h
a
v
io
r”
,
IJ
CS
IT
-
I
n
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
Co
mp
u
ter
S
c
ien
c
e
a
n
d
I
n
fo
rm
a
ti
o
n
T
e
c
h
n
o
l
o
g
ies
,
Vo
l.
6
(
2
),
2
0
1
5
.
[8
]
S
.
P
a
rv
a
ti
k
a
r
a
n
d
B.
Jo
sh
i,
“
An
a
lys
is
o
f
Us
e
r
Beh
a
v
io
r
th
ro
u
g
h
W
e
b
Us
a
g
e
M
in
in
g
,
”
ICA
S
T
-
In
tern
a
ti
o
n
a
l
Co
n
f
e
re
n
c
e
o
n
A
d
v
a
n
c
e
s in
S
c
ien
c
e
a
n
d
T
e
c
h
n
o
lo
g
y
,
2
0
1
4
.
[9
]
A
.
De
e
p
a
,
a
n
d
P
.
Ra
a
jan
,
“
An
e
ff
i
c
ien
t
p
re
p
ro
c
e
ss
in
g
me
th
o
d
o
l
o
g
y
o
f
l
o
g
f
il
e
fo
r
W
e
b
u
sa
g
e
mi
n
i
n
g
,
”
NCRIIA
M
I
-
Na
ti
o
n
a
l
Co
n
f
e
re
n
c
e
o
n
Re
se
a
rc
h
Iss
u
e
s in
Im
a
g
e
A
n
a
l
y
sis a
n
d
M
i
n
in
g
In
telli
g
e
n
c
e
,
2
0
1
5
.
[1
0
]
S
h
u
m
p
e
i
Ok
u
ra
,
Yu
k
ih
iro
T
a
g
a
m
i,
S
h
in
g
o
On
o
,
a
n
d
A
k
ira
Tajima
.
E
m
b
e
d
d
in
g
-
b
a
se
d
Ne
ws
Re
c
o
m
m
e
n
d
a
ti
o
n
f
o
r
M
il
li
o
n
s
o
f
Us
e
rs.
In
S
IG
KD
D.
2
0
1
7
.
[1
1
]
Ya
n
g
S
o
n
g
,
A
li
M
a
m
d
o
u
h
El
k
a
h
k
y
,
a
n
d
X
ia
o
d
o
n
g
He
.
M
u
lt
i
-
ra
te
d
e
e
p
lea
rn
i
n
g
f
o
r
tem
p
o
ra
l
re
c
o
m
m
e
n
d
a
ti
o
n
.
In
S
IG
IR.
9
0
9
-
9
1
2
,
2
0
1
6.
[1
2
]
X
u
C
h
e
n
,
Y
o
n
g
f
e
n
g
Zh
a
n
g
,
Qin
g
y
a
o
A
i,
Ho
n
g
ten
g
X
u
,
Ju
n
c
h
i
Ya
n
,
a
n
d
Z
h
e
n
g
Qin
.
P
e
rso
n
a
li
z
e
d
Ke
y
F
ra
m
e
Re
c
o
m
m
e
n
d
a
ti
o
n
.
I
n
S
IG
IR,
2
0
1
7
.
[1
3
]
T
h
iT
h
a
n
h
S
a
n
g
Ng
u
y
e
n
,
Ha
i
Ya
n
L
u
,
Jie
L
u
.
“
Web
-
p
a
g
e
Re
c
o
m
m
e
n
d
a
ti
o
n
b
a
se
d
o
n
W
e
b
Us
a
g
e
a
n
d
Do
m
a
in
Kn
o
w
led
g
e
,
”
1
0
4
1
-
4
3
4
7
/1
3
/
$
3
1
.
0
0
©
2
0
1
3
IEE
E
.
[1
4
]
Ro
sh
a
n
A
.
G
a
n
g
u
rd
e
a
n
d
Bin
o
d
Ku
m
a
r.
“
Ne
x
t
W
e
b
P
a
g
e
P
re
d
ictio
n
u
si
n
g
Ge
n
e
ti
c
A
lg
o
rit
h
m
a
n
d
F
e
e
d
F
o
rw
a
rd
A
s
so
c
iatio
n
Ru
le
b
a
se
d
o
n
W
e
b
-
L
o
g
F
e
a
tu
re
s”
.
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Per
fo
rm
a
b
il
it
y
En
g
in
e
e
rin
g
,
1
6
(1
):
1
0
-
18
,
2
0
2
0
.
[1
5
]
Am
it
Bh
a
g
a
t,
S
a
n
jay
S
h
a
r
m
a
,
K.
R.
P
a
rd
a
sa
n
i.
“
On
t
o
lo
g
ica
l
F
re
q
u
e
n
t
P
a
tt
e
rn
s
M
in
i
n
g
b
y
p
o
ten
ti
a
l
u
se
o
f
Ne
u
ra
l
Ne
tw
o
rk
”
.
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Co
m
p
u
ter
A
p
p
l
ica
ti
o
n
s
,
Vo
l
3
6
,
No
.
1
0
,
p
p
.
0
9
7
5
-
8
8
8
7
,
2
0
1
1
.
[1
6
]
M
a
c
A
rth
u
r
R.
,
W
il
so
n
E
.
T
h
e
T
h
e
o
ry
o
f
B
io
g
e
o
g
r
a
p
h
y
.
P
rin
c
e
t
o
n
,
NJ
,
US
A
:
P
ri
n
c
e
to
n
U
n
iv
e
rsity
P
re
ss
;
1
9
6
7
.
[1
7
]
S
im
o
n
D.
Bio
g
e
o
g
ra
p
h
y
-
b
a
se
d
o
p
ti
m
iza
ti
o
n
.
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
Evo
l
u
ti
o
n
a
ry
Co
mp
u
t
a
ti
o
n
.
1
2
(6
):
7
0
2
-
7
1
3
,
2
0
0
8
.
d
o
i:
1
0
.
1
1
0
9
/
tev
c
.
2
0
0
8
.
9
1
9
0
0
4
.
[1
8
]
Da
tse
t
so
u
rc
e
:
h
tt
p
s://
p
r
o
jec
tt
u
n
n
e
l.
c
o
m
/p
a
g
e
s/S
u
p
p
o
rt
[1
9
]
R.
M
a
n
ik
a
n
d
a
n
.
“
A
n
o
v
e
l
a
p
p
ro
a
c
h
o
n
P
a
rti
c
le
A
g
e
n
t
S
w
a
r
m
O
p
ti
m
iza
ti
o
n
(
P
A
S
O)
in
se
m
a
n
ti
c
m
in
in
g
f
o
r
we
b
p
a
g
e
re
c
o
m
m
e
n
d
e
r
s
y
ste
m
o
f
m
u
lt
im
e
d
ia
d
a
ta:
a
h
e
a
lt
h
c
a
re
p
e
rsp
e
c
ti
v
e
”
.
S
p
rin
g
e
r
S
c
ien
c
e
+
Bu
sin
e
ss
M
e
d
ia,
LL
C,
p
a
rt
o
f
S
p
ri
n
g
e
r
Na
tu
re
1
0
Ja
n
u
a
r
y
2
0
1
9
.
B
I
O
G
RAP
H
I
E
S
O
F
AUTH
O
R
S
M
r.
Ro
sh
a
n
A
.
Ga
n
g
u
rd
e
is
a
Re
se
a
rc
h
S
c
h
o
lar
a
t
De
p
a
rt
m
e
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
,
S
a
v
it
rib
a
i
P
h
u
le
P
u
n
e
Un
iv
e
rsiit
y
,
P
u
n
e
,
M
a
h
a
ra
sh
tra,
In
d
ia.
His
a
re
a
s
o
f
in
tere
st
a
re
Da
ta
M
in
in
g
,
Bu
sin
e
ss
A
n
a
l
y
ti
c
s a
n
d
M
a
c
h
in
e
L
e
a
rn
in
g
.
Dr.
Bin
o
d
K
u
m
a
r
is
P
ro
f
e
ss
o
r
&
Dire
c
to
r
a
t
Ja
y
a
w
a
n
t
In
stit
u
t
e
o
f
Co
m
p
u
ter
A
p
p
li
c
a
ti
o
n
s
(JSP
M
'
s
G
ro
u
p
),
a
ff
il
iate
d
to
S
a
v
it
rib
a
iP
h
u
le
P
u
n
e
Un
iv
e
rsity
,
In
d
ia.
He
h
a
s
o
b
tain
e
d
P
h
D
(Co
m
p
u
ter
S
c
.
)
in
2
0
1
0
,
M
.
P
h
i
l.
(Co
m
p
u
ter
S
c
.
)
in
2
0
0
6
,
M
CA
(NIT
Ja
m
sh
e
d
p
u
r)
i
n
1
9
9
8
a
n
d
M
.
S
c
.
(BHU
)
&
B
.
S
c
.
(BHU
)
i
n
1
9
9
5
&
1
9
9
3
re
sp
e
c
ti
v
e
ly
.
He
is
h
a
v
in
g
m
o
re
th
a
n
2
1
y
e
a
rs
o
f
e
x
p
e
rien
c
e
in
v
a
rio
u
s
c
a
p
a
c
it
ies
in
re
se
a
rc
h
,
tea
c
h
i
n
g
a
n
d
a
c
a
d
e
m
ic
a
d
m
in
istratio
n
.
He
g
o
t
e
x
p
o
su
re
a
s
A
ss
o
c
iate
P
ro
f
e
ss
o
r
a
t
S
c
h
o
o
l
o
f
En
g
in
e
e
rin
g
a
n
d
Co
m
p
u
ter
T
e
c
h
n
o
lo
g
y
,
Qu
e
st
In
ter
n
a
ti
o
n
a
l
Un
iv
e
rsity
,
M
a
la
y
si
a
.
He
h
a
s
b
e
e
n
a
g
u
id
e
a
n
d
a
n
e
v
a
lu
a
to
r
f
o
r
M
CA
/
M
.
P
h
il
.
d
isse
rtatio
n
s
a
n
d
P
h
.
D
th
e
sis
o
f
s
tu
d
e
n
ts
o
f
v
a
rio
u
s
Un
iv
e
rsiti
e
s.
H
e
is
re
c
o
g
n
ize
d
P
h
.
D
g
u
id
e
i
n
Co
m
p
u
ter
S
c
ien
c
e
&
Co
m
p
u
ter
M
a
n
a
g
e
m
e
n
t
u
n
d
e
r
S
a
v
it
ri
b
a
i
P
h
u
le
P
u
n
e
U
n
iv
e
r
sity
.
A
t
p
re
se
n
t
f
iv
e
P
h
.
D.
c
a
n
d
id
a
tes
a
re
re
g
ister
e
d
u
n
d
e
r
h
is su
p
e
rv
isio
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.