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.
4
,
No
.
2
,
J
u
n
e
201
5
,
p
p
.
5
3
~6
1
I
SS
N:
2252
-
8938
53
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
jo
u
r
n
a
l.c
o
m/o
n
lin
e/in
d
ex
.
p
h
p
/I
J
AI
Desig
n an Alg
o
rit
h
m
f
o
r Sof
tw
a
re
Dev
elo
p
m
e
nt
i
n C
bse
E
nv
iro
n
m
en
t
u
si
ng
F
eed
Forw
a
rd
Neura
l Ne
tw
o
rk
A
m
it
Ver
m
a
,
E
r.
P
a
rdee
p
ka
ur
De
p
a
rtme
n
t
C
o
m
p
u
ter
S
c
ien
c
e
E
n
g
in
e
e
rin
g
,
Ch
a
n
d
ig
a
rh
Un
iv
e
rsity
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Feb
5
,
2
0
1
6
R
ev
i
s
ed
A
p
r
il
8
,
2
0
1
6
A
cc
ep
ted
Ma
y
9
,
2
0
1
6
A
In
so
f
t
w
a
re
d
e
v
e
lo
p
m
e
n
t
o
rg
a
n
iza
ti
o
n
s,
Co
m
p
o
n
e
n
t
b
a
se
d
S
o
f
tw
a
re
e
n
g
in
e
e
rin
g
(CBS
E)
is
e
m
e
r
g
in
g
p
a
ra
d
ig
m
f
o
r
so
f
t
w
a
re
d
e
v
e
lo
p
m
e
n
t
a
n
d
g
a
in
e
d
w
id
e
a
c
c
e
p
tan
c
e
a
s
it
o
f
ten
re
su
lt
s
i
n
i
n
c
re
a
se
q
u
a
li
ty
o
f
so
f
t
wa
r
e
p
ro
d
u
c
t
w
it
h
in
d
e
v
e
lo
p
m
e
n
t
ti
m
e
a
n
d
b
u
d
g
e
t.
I
n
c
o
m
p
o
n
e
n
t
re
u
sa
b
i
li
ty
,
m
a
in
c
h
a
ll
e
n
g
e
s
a
re
th
e
rig
h
t
c
o
m
p
o
n
e
n
t
id
e
n
ti
f
ica
ti
o
n
f
ro
m
larg
e
re
p
o
sito
ries
a
t
rig
h
t
ti
m
e
.
T
h
e
m
a
jo
r
o
b
jec
ti
v
e
o
f
th
is
w
o
rk
is
to
p
ro
v
id
e
e
ff
icie
n
t
a
lg
o
rit
h
m
f
o
r
sto
ra
g
e
a
n
d
e
ff
e
c
ti
v
e
re
tri
e
v
a
l
o
f
c
o
m
p
o
n
e
n
ts
u
si
n
g
n
e
u
ra
l
n
e
tw
o
rk
a
n
d
p
a
ra
m
e
ters
b
a
se
d
o
n
u
se
r
c
h
o
ice
t
h
ro
u
g
h
c
lu
ste
ri
n
g
.
T
h
is
re
se
a
rc
h
p
a
p
e
r
a
im
s
to
p
ro
p
o
se
a
n
a
lg
o
rit
h
m
th
a
t
p
ro
v
id
e
s
e
rro
r
f
r
e
e
a
n
d
a
u
to
m
a
ti
c
p
ro
c
e
ss
(
f
o
r
re
tri
e
v
a
l
o
f
th
e
c
o
m
p
o
n
e
n
ts)
w
h
il
e
re
u
se
o
f
th
e
c
o
m
p
o
n
e
n
t.
In
th
is
a
lg
o
rit
h
m
,
k
e
y
w
o
rd
s
(o
r
c
o
m
p
o
n
e
n
ts)
a
re
e
x
trac
ted
f
ro
m
so
f
t
w
a
re
d
o
c
u
m
e
n
t,
a
f
ter
b
y
a
p
p
ly
in
g
k
m
e
a
n
c
lu
ste
rin
g
a
l
g
o
rit
h
m
.
T
h
e
n
w
e
ig
h
ts
a
ss
ig
n
e
d
t
o
t
h
o
se
k
e
y
w
o
rd
s
b
a
se
d
o
n
th
e
ir
f
re
q
u
e
n
c
y
a
n
d
a
f
ter
a
ss
ig
n
in
g
w
e
ig
h
ts,
A
NN
p
re
d
icts
w
h
e
th
e
r
c
o
rr
e
c
t
w
e
ig
h
t
i
s
a
ss
i
g
n
e
d
to
k
e
y
w
o
rd
s
(o
r
c
o
m
p
o
n
e
n
ts)
o
r
n
o
t,
o
th
e
rw
ise
it
b
a
c
k
p
ro
p
a
g
a
tes
in
to
in
it
ial
ste
p
(re
-
a
ss
ig
n
th
e
we
ig
h
ts).
In
las
t,
sto
re
t
h
o
se
a
ll
k
e
y
w
o
rd
s
in
t
o
Re
p
o
sito
r
ies
f
o
r
e
ff
e
c
ti
v
e
re
tri
e
v
a
l.
P
r
o
p
o
se
d
a
lg
o
rit
h
m
is
v
e
r
y
e
ffe
c
ti
v
e
in
th
e
e
rr
o
r
c
o
rre
c
ti
o
n
a
n
d
d
e
tec
ti
o
n
w
it
h
u
se
r
b
a
se
c
h
o
ice
w
h
il
e
c
h
o
i
c
e
o
f
c
o
m
p
o
n
e
n
t
f
o
r
re
u
sa
b
il
it
y
fo
r
e
ff
icie
n
t
re
tri
e
v
a
l
is
th
e
re
.
T
o
c
h
e
c
k
th
e
re
su
lt
s
o
f
o
u
r
a
lg
o
rit
h
m
b
a
se
d
o
n
f
a
c
to
rs
li
k
e
a
c
c
u
ra
c
y
,
p
re
c
isio
n
a
n
d
re
c
a
ll
c
o
m
p
a
re
w
it
h
e
x
isti
n
g
tec
h
n
iq
u
e
i.
e
.
in
teg
ra
ted
c
las
si
f
ica
ti
o
n
sc
h
e
m
e
f
o
r
re
tri
e
v
a
l
o
f
c
o
m
p
o
n
e
n
ts
b
a
se
d
o
n
k
e
y
wo
rd
se
a
rc
h
a
n
d
re
su
lt
s are
so
e
n
c
o
u
ra
g
in
g
.
K
ey
w
o
r
d
:
B
ac
k
p
r
o
p
ag
atio
n
alg
o
r
ith
m
K
e
y
w
o
r
d
b
ased
r
etr
iev
al
C
o
m
p
o
n
e
n
t
b
ased
d
ev
elo
p
m
e
n
t
c
lu
s
ter
i
n
g
Co
p
y
rig
h
t
©
2
0
1
5
In
stit
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
Al
l
rig
h
ts
re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Am
it Ve
r
m
a
,
Dep
ar
t
m
en
t
C
o
m
p
u
ter
Scie
n
c
e
E
n
g
i
n
ee
r
in
g
,
C
h
a
n
d
ig
ar
h
U
n
iv
er
s
it
y
.
1.
I
NT
RO
D
UCT
I
O
N
I
n
2
1
th
ce
n
t
u
r
y
c
u
r
r
en
t
tr
en
d
o
f
au
to
m
o
ti
v
e
tec
h
n
o
lo
g
ies
e
m
er
g
i
n
g
at
v
er
y
h
ig
h
p
ac
e.
Du
r
in
g
th
e
p
ast
f
o
r
t
y
y
ea
r
s
,
v
ar
io
u
s
ap
p
r
o
ac
h
es
f
o
r
s
o
f
t
w
ar
e
d
ev
elo
p
m
en
t
co
m
e
i
n
to
ex
i
s
te
n
ce
.
Fro
m
th
e
last
f
e
w
y
ea
r
s
,
th
e
b
asic a
p
p
r
o
ac
h
o
r
m
et
h
o
d
u
s
ed
b
y
d
e
v
elo
p
er
is
to
s
ep
ar
a
te
th
e
s
o
f
t
w
ar
e
i
n
to
p
h
a
s
es a
n
d
w
o
r
k
ac
co
r
d
in
g
to
th
o
s
e
p
h
a
s
es
[
1
]
,
s
o
t
h
at
t
h
e
y
ca
n
co
n
ce
n
tr
ate
o
n
l
y
o
n
e
p
h
ase
at
a
ti
m
e.
T
h
e
f
ir
s
t
s
o
f
t
w
ar
e
d
ev
elo
p
m
en
t
ap
p
r
o
ac
h
af
ter
s
o
f
t
w
ar
e
cr
is
is
co
m
e
s
i
n
to
e
x
is
te
n
ce
i
n
1
9
7
0
s
.
On
t
h
e
b
asi
s
o
f
d
if
f
er
en
t
n
ee
d
s
o
f
c
u
s
to
m
er
s
a
n
d
o
r
g
an
izatio
n
’
s
tar
g
ets,
d
ev
elo
p
er
s
r
ef
er
d
if
f
er
en
t
ap
p
r
o
ac
h
es
f
o
r
d
ev
elo
p
m
e
n
t.
I
n
s
o
f
t
w
ar
e
d
ev
elo
p
m
e
n
t,
m
o
s
tl
y
ch
al
len
g
es
f
ac
ed
b
y
p
r
o
j
ec
t
m
a
n
a
g
e
m
e
n
t
an
d
p
r
o
j
e
ct
lead
lik
e
d
ea
d
lin
es
o
f
p
r
o
ject,
ex
tr
a
r
eso
u
r
ce
s
n
ee
d
ed
,
o
v
er
b
u
d
g
et
etc.
T
h
ese
all
is
s
u
e
s
s
o
m
e
ti
m
e
s
o
cc
u
r
f
r
o
m
p
r
o
j
ec
t
co
m
p
lex
it
y
in
cl
u
d
e
s
ize,
les
s
k
n
o
w
led
g
e
etc.
T
o
o
v
er
co
m
e
th
ese
i
s
s
u
es,
co
m
p
o
n
e
n
t
b
ase
d
d
ev
elo
p
m
e
n
t
co
m
e
in
to
ex
is
ten
ce
in
1
9
9
0
’
s
[
8
]
.
T
h
r
o
u
g
h
t
h
i
s
m
eth
o
d
,
o
r
g
a
n
iz
atio
n
s
ca
n
d
ev
e
lo
p
th
eir
s
o
f
t
war
e
b
y
s
elec
ti
n
g
ap
p
r
o
p
r
iate
C
o
m
m
er
cial
-
O
f
f
-
th
e
-
s
h
el
f
(
C
OT
S)
co
m
p
o
n
en
t
s
an
d
ass
e
m
b
le
t
h
e
m
.
T
h
ese
C
OT
S
co
m
p
o
n
en
t
s
d
ev
elo
p
b
y
d
if
f
er
en
t
d
ev
elo
p
er
s
b
y
u
s
i
n
g
d
if
f
er
e
n
t
p
latf
o
r
m
s
an
d
lan
g
u
a
g
es.
C
o
m
p
o
n
en
t
ca
n
b
e
s
o
m
e
co
d
e,
u
tili
t
y
f
u
n
ctio
n
s
o
r
p
r
o
g
r
am
m
i
n
g
u
n
i
t
an
d
co
m
p
o
n
e
n
t
ca
n
b
e
p
r
o
d
u
ct
s
p
ec
if
ic,
d
o
m
ain
s
p
ec
if
ic
a
n
d
d
o
m
ain
in
d
ep
en
d
e
n
t.
T
h
e
ai
m
o
f
C
B
SE
i
s
to
ac
h
iev
e
m
u
lt
ip
le
q
u
alit
y
o
b
j
ec
tiv
es
s
u
ch
as
r
e
u
s
ab
ilit
y
,
i
n
ter
o
p
er
ab
ilit
y
,
i
m
p
le
m
e
n
tati
o
n
tr
an
s
p
ar
en
c
y
.
I
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
4
,
No
.
2
,
J
u
n
e
201
5
:
5
3
–
61
54
w
h
o
le
p
r
o
ce
s
s
,
m
ai
n
e
m
p
h
a
s
i
s
is
o
n
r
e
u
s
e
a
co
m
p
o
n
e
n
t,
c
lu
s
ter
in
g
[
2
2
]
an
d
r
etr
ie
v
al
ar
e
t
w
o
i
m
p
o
r
tan
t
p
ar
ts
o
f
co
m
p
o
n
en
t
b
ased
d
ev
elo
p
m
en
t.
C
l
u
s
ter
i
n
g
[
3
2
]
is
f
o
r
m
o
f
u
n
s
u
p
er
v
i
s
ed
lear
n
in
g
,
is
th
e
p
r
o
ce
s
s
o
f
p
ar
titi
o
n
in
g
a
s
et
o
f
d
ata
in
to
a
s
et
o
f
m
ea
n
i
n
g
f
u
l
s
u
b
class
e
s
ca
lled
clu
s
ter
.
I
t
is
b
asicall
y
o
r
g
an
iz
in
g
d
ata
in
to
g
r
o
u
p
s
b
ased
o
n
t
h
eir
s
i
m
ilar
i
t
y
.
C
l
u
s
ter
in
g
is
w
id
el
y
u
s
ed
in
ec
o
n
o
m
ic
s
cie
n
ce
,
P
atter
n
r
ec
o
g
n
itio
n
,
s
p
atial
d
ata
an
al
y
s
i
s
,
i
m
ag
e
p
r
o
ce
s
s
i
n
g
etc.
I
n
s
o
f
t
w
ar
e
r
eu
s
ab
ilit
y
,
t
h
e
f
ir
s
t
a
n
d
f
o
r
e
m
o
s
t
f
u
n
d
a
m
en
tal
p
r
o
b
le
m
i
s
lo
ca
tin
g
an
d
r
etr
ie
v
i
n
g
r
i
g
h
t
co
m
p
o
n
en
t
f
r
o
m
lar
g
e
r
ep
o
s
ito
r
y
[
3
0
]
.
R
etr
iev
al
o
f
co
m
p
o
n
en
t
s
h
o
u
ld
b
e
ef
f
icien
t
a
n
d
ti
m
e
co
n
s
u
m
in
g
.
Ma
i
n
l
y
f
o
r
r
eu
s
e
a
co
m
p
o
n
en
t,
d
e
v
elo
p
er
s
h
a
v
e
to
s
t
o
r
e
th
eir
r
elev
a
n
t
co
m
p
o
n
e
n
t
in
to
r
ep
o
s
ito
r
y
.
Va
r
io
u
s
cl
u
s
ter
i
n
g
al
g
o
r
ith
m
s
li
k
e
K
m
ea
n
[
1
4
]
,
K
-
m
o
d
e
[
1
3
]
,
Gen
etic
Alg
o
r
it
h
m
[
2
1
]
etc.
ap
p
lied
o
n
r
ep
o
s
ito
r
y
to
m
a
k
e
clu
s
ter
o
n
b
ased
o
f
b
eh
av
io
r
al,
s
tr
u
c
t
u
r
al
o
r
f
u
n
ct
io
n
al
attr
ib
u
te
s
.
T
h
en
n
e
x
t
s
tep
is
to
r
etr
ie
v
e
a
co
m
p
o
n
e
n
t
b
y
ap
p
l
y
i
n
g
v
ar
io
u
s
co
m
p
o
n
en
t
r
etr
iev
al
tech
n
i
q
u
es
li
k
e
Ke
y
w
o
r
d
b
ased
m
e
th
o
d
,
S
y
n
ta
x
b
ased
,
Se
m
a
n
tic
b
ased
an
d
Gen
e
tic
b
ased
o
p
tim
izatio
n
m
e
th
o
d
[
3
8
]
etc.
I
n
last
,
th
e
y
m
a
y
b
e
ap
p
l
y
m
etr
ic
s
li
k
e
c
y
clo
m
etr
ic
co
m
p
le
x
it
y
,
Hal
s
t
ea
d
m
etr
ics,
r
eg
u
lar
it
y
m
etr
i
cs
etc
o
n
s
tr
u
ct
u
r
al
attr
ib
u
tes
a
n
d
r
es
u
lt
s
f
ee
d
u
p
i
n
to
n
e
u
r
al
n
et
w
o
r
k
’
s
al
g
o
r
ith
m
s
[
4
]
.
T
h
e
f
ield
o
f
n
e
u
r
al
n
et
w
o
r
k
s
h
as
a
h
is
to
r
y
o
f
s
o
m
e
f
iv
e
d
ec
ad
es
b
u
t
h
as
f
o
u
n
d
ap
p
licatio
n
o
n
l
y
i
n
t
h
e
p
ast
f
i
f
tee
n
y
ea
r
s
,
an
d
t
h
e
f
ie
l
d
is
s
till
d
ev
elo
p
in
g
co
n
tin
u
o
u
s
l
y
.
Ne
u
r
al
n
et
w
o
r
k
is
ab
le
to
u
s
e
s
o
m
e
h
id
d
en
u
n
k
n
o
w
n
i
n
f
o
r
m
atio
n
in
th
e
d
ata.
T
h
is
p
r
o
ce
s
s
o
f
ca
p
tu
r
in
g
h
id
d
en
i
n
f
o
r
m
atio
n
i
s
ca
lled
lear
n
in
g
o
r
tr
ai
n
i
n
g
n
e
t
w
o
r
k
.
W
e
ca
n
tr
ain
a
n
eu
r
al
n
et
w
o
r
k
to
p
er
f
o
r
m
a
p
ar
ticu
lar
f
u
n
ctio
n
b
y
ad
j
u
s
ti
n
g
th
e
v
al
u
es
o
f
t
h
e
co
n
n
ec
t
io
n
s
(
w
ei
g
h
ts
)
b
et
w
ee
n
ele
m
en
ts
.
Var
io
u
s
alg
o
r
ith
m
s
ar
e
B
ac
k
P
r
o
p
ag
atio
n
alg
o
r
it
h
m
[
3
6
]
,
s
elf
o
r
g
an
iz
in
g
m
ap
al
g
o
r
ith
m
an
d
v
ar
io
u
s
co
n
j
u
g
ate
g
r
ad
ien
t
al
g
o
r
it
h
m
s
[
4
]
lik
e
Flet
ch
er
-
r
ee
v
e
s
,
P
o
lak
R
ib
ier
e,
P
o
w
el
l
B
ea
le
an
d
Scaled
co
n
j
u
g
ate
g
r
ad
ien
t
alg
o
r
ith
m
.
I
n
b
ac
k
p
r
o
p
ag
atio
n
al
g
o
r
ith
m
ad
j
u
s
t
s
t
h
e
w
ei
g
h
t
s
in
th
e
s
teep
est
d
escen
t d
ir
ec
tio
n
(
n
e
g
ati
v
e
o
f
th
e
g
r
ad
ien
t)
a
n
d
ca
lcu
late
s
m
ea
n
s
q
u
ar
e
er
r
o
r
(
MSE
)
th
at
s
h
o
w
s
th
e
d
i
f
f
er
en
c
e
b
et
w
ee
n
r
esu
lted
o
u
tp
u
t
a
n
d
tar
g
et
o
u
tp
u
t.
T
h
is
r
esear
ch
p
a
p
er
aim
s
to
p
r
o
p
o
s
e
an
alg
o
r
it
h
m
th
at
p
r
o
v
id
es e
r
r
o
r
f
r
ee
an
d
au
to
m
atic
p
r
o
ce
s
s
(
f
o
r
r
etr
iev
al
o
f
th
e
co
m
p
o
n
e
n
ts
)
w
h
ile
r
eu
s
e
o
f
t
h
e
co
m
p
o
n
en
t.
T
h
e
m
aj
o
r
o
b
j
ec
tiv
e
o
f
th
is
w
o
r
k
i
s
to
p
r
o
v
id
e
ef
f
icien
t
al
g
o
r
ith
m
f
o
r
s
to
r
ag
e
an
d
e
f
f
ec
ti
v
e
r
etr
i
ev
al
o
f
co
m
p
o
n
e
n
t
s
u
s
i
n
g
n
eu
r
al
n
et
w
o
r
k
a
n
d
p
ar
am
eter
s
b
ased
o
n
u
s
er
ch
o
i
ce
th
r
o
u
g
h
cl
u
s
ter
i
n
g
.
Ma
n
y
tech
n
iq
u
es
a
n
d
tec
h
n
o
lo
g
ies
h
av
e
b
ee
n
p
r
o
p
o
s
ed
an
d
ev
al
u
ated
b
u
t
s
o
f
a
r
n
o
n
e
o
f
m
et
h
o
d
o
lo
g
y
h
as
f
o
cu
s
ed
o
n
e
r
r
o
r
co
r
r
ec
tio
n
an
d
d
etec
tio
n
w
it
h
u
s
er
b
a
s
e
ch
o
ice
w
h
i
le
c
h
o
ice
o
f
co
m
p
o
n
en
t
f
o
r
r
eu
s
ab
ilit
y
f
o
r
ef
f
icie
n
t
r
etr
iev
al,
r
ep
o
s
ito
r
ies
m
u
s
t
b
e
w
el
l
s
ep
ar
ated
(
b
o
u
n
d
ar
ies
s
h
o
u
ld
b
e
d
ef
in
ed
th
r
o
u
g
h
p
ar
a
m
eter
lis
t
in
g
)
r
elate
d
to
th
e
co
m
p
o
n
e
n
t
s
.
T
h
is
s
er
v
e
s
as
a
p
i
v
o
t
p
o
in
t
f
o
r
th
e
p
r
o
p
o
s
ed
m
et
h
o
d
o
lo
g
y
.
T
h
e
p
ap
e
r
’
s
o
r
g
an
iza
tio
n
is
a
s
f
o
llo
w
s
:
1.
Sectio
n
2
d
escr
ib
es P
r
o
b
lem
s
t
ate
m
e
n
t
2.
Sectio
n
3
d
escr
ib
es th
e
p
r
o
p
o
s
ed
alg
o
r
ith
m
f
o
r
co
m
p
o
n
e
n
t r
eu
s
ab
ilit
y
3.
Sectio
n
4
d
escr
ib
es i
m
p
le
m
e
n
t
atio
n
w
o
r
k
4.
Sectio
n
5
d
escr
ib
es th
e
E
x
p
er
i
m
en
tal
r
esu
lts
5.
Sectio
n
6
co
v
er
s
th
e
co
n
clu
s
io
n
an
d
F
u
t
u
r
e
s
co
p
e
2.
CURR
E
NT
AR
E
AS
T
O
B
E
WO
RK
UP
O
N:
T
he
pro
ble
m
a
d
dres
s
ed
in
t
his
pa
per
ca
n
be
f
o
r
m
a
l
ly
s
t
a
t
ed
a
s
:
T
h
er
e
is
s
et
o
f
co
m
p
o
n
en
t
s
S
C
an
d
co
m
p
o
n
en
t
s
h
a
v
e
th
e
ir
o
w
n
d
if
f
er
en
t
n
at
u
r
e
r
ep
r
esen
ti
n
g
F
R
(
F
u
n
ct
io
n
al
R
eq
u
ir
e
m
e
n
ts
)
an
d
NF
R
(
No
n
Fu
n
ctio
n
al
R
eq
u
ir
e
m
en
ts
)
.
E
ac
h
co
m
p
o
n
en
t
co
v
er
s
s
o
m
e
r
e
q
u
ir
e
m
e
n
ts
R
.
W
e
h
av
e
to
s
to
r
e
co
m
p
o
n
e
n
ts
i
n
to
th
eir
r
esp
ec
ti
v
e
R
ep
o
s
ito
r
ies
b
y
p
as
s
in
g
th
r
o
u
g
h
M
u
lti
la
y
er
Feed
Fo
r
w
ar
d
Ne
u
r
al
N
et
w
o
r
k
(
M
L
FF
N
N)
.
T
h
er
e
is
d
ir
e
n
e
ed
to
m
a
n
a
g
e
r
ep
o
s
ito
r
y
t
h
r
o
u
g
h
cl
u
s
ter
in
g
a
n
d
m
a
k
i
n
g
th
e
p
r
o
ce
s
s
a
u
to
m
at
ic.
A
n
er
r
o
r
co
r
r
ec
tio
n
an
d
d
etec
tio
n
ar
e
th
e
k
e
y
i
s
s
u
es
w
it
h
a
u
to
m
atio
n
t
h
at
ca
n
o
v
er
co
m
e
u
s
i
n
g
n
e
u
r
al
n
et
w
o
r
k
tech
n
iq
u
es.
Nee
d
a
nd
Sig
nifica
nce
:
Fo
r
r
eu
s
ab
ili
t
y
o
f
co
m
p
o
n
e
n
ts
,
C
o
m
p
o
n
e
n
t
id
e
n
ti
f
icatio
n
o
r
r
etr
iev
al
o
f
ex
ac
t
co
m
p
o
n
e
n
t
is
n
o
t
an
ea
s
y
ta
s
k
i
n
C
B
SE.
Des
ig
n
i
n
g
a
s
y
s
te
m
b
y
r
e
u
s
i
n
g
ex
i
s
ti
n
g
c
o
m
p
o
n
en
t
s
lead
s
to
f
aster
ti
m
e
to
m
ar
k
et.
Ho
w
e
v
er
,
f
in
d
in
g
t
h
e
ap
p
r
o
p
r
iate
co
m
p
o
n
en
t
th
at
s
a
tis
f
ies
t
h
e
s
et
o
f
r
eq
u
ir
e
m
en
t
s
is
b
ec
o
m
i
n
g
th
e
m
o
s
t
d
i
f
f
ic
u
lt
a
n
d
ch
alle
n
g
in
g
tas
k
.
T
h
er
e
is
d
ir
e
n
ee
d
to
m
a
n
ag
e
r
ep
o
s
ito
r
ies,
So
th
at
it
co
u
l
d
b
e
ea
s
y
to
s
ea
r
ch
co
m
p
o
n
e
n
t t
h
at
co
v
er
s
all
r
eq
u
ir
e
m
e
n
ts
i
n
a
m
an
n
er
th
a
t
m
a
x
i
m
izes t
h
e
q
u
alit
y
o
f
p
r
o
d
u
ct.
3.
AL
G
O
RI
T
H
M
I
n
s
o
f
t
w
ar
e
d
ev
elo
p
m
e
n
t
o
r
g
an
izati
o
n
s
,
co
n
ce
p
t
o
f
r
eu
s
e
a
co
m
p
o
n
en
t
is
i
m
p
o
r
tan
t
to
m
ai
n
tai
n
r
eliab
ilit
y
o
r
to
in
cr
ea
s
e
a
q
u
alit
y
a
n
d
a
lev
el
o
f
p
r
o
d
u
ctiv
i
t
y
.
R
e
u
s
e
a
p
ar
t
o
f
s
o
f
t
w
ar
e
is
p
r
o
v
ed
to
b
e
v
er
y
b
en
ef
icia
l
to
o
v
er
co
m
e
t
h
e
c
h
alle
n
g
e
s
o
cc
u
r
in
s
o
f
t
w
ar
e
d
ev
elo
p
m
en
t,
i
s
o
v
er
b
u
d
g
et,
ti
m
e
co
m
p
le
x
it
y
,
d
ea
d
lin
es
o
f
co
m
p
lex
p
r
o
j
ec
t
etc.
I
t
tak
es
lar
g
e
ti
m
e
to
im
p
le
m
e
n
t
s
o
f
t
w
ar
e
f
r
o
m
s
cr
atch
.
On
t
h
at
ti
m
e
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
Desig
n
a
n
A
lg
o
r
ith
m
fo
r
S
o
ftw
a
r
e
Dev
elo
p
men
t in
C
b
s
e
E
n
vi
r
o
n
men
t u
s
in
g
…
(
A
mit V
erma
)
55
co
m
p
o
n
e
n
t
b
ased
d
ev
elo
p
m
e
n
t
m
u
s
t
b
e
v
er
y
h
elp
f
u
l.
C
o
m
p
o
n
en
t
is
in
d
ep
en
d
e
n
t
p
ar
t
p
er
f
o
r
m
co
m
p
lete
f
u
n
ctio
n
alitie
s
.
C
o
m
p
o
n
e
n
t
B
ased
s
o
f
t
w
a
r
e
e
n
g
in
ee
r
i
n
g
is
u
s
ed
to
d
ev
elo
p
o
r
ass
e
m
b
le
s
o
f
t
w
ar
e
f
r
o
m
ex
is
t
in
g
co
m
p
o
n
en
t
s
.
Fo
r
t
h
e
s
u
cc
e
s
s
es
o
f
s
o
f
t
w
ar
e
p
r
o
j
ec
t,
th
er
e
i
s
d
ir
e
n
ee
d
to
m
a
k
e
t
h
e
p
r
o
ce
s
s
b
etter
an
d
co
m
p
o
n
e
n
t
r
eu
s
ab
ilit
y
is
o
p
ti
m
al
s
o
l
u
tio
n
.
Fo
r
th
i
s
p
u
r
p
o
s
e,
C
o
m
p
o
n
en
t
s
to
r
ag
e
an
d
r
et
r
iev
al
ar
e
t
w
o
k
e
y
is
s
u
es.
I
n
t
h
e
co
n
ce
p
t
o
f
r
e
u
s
ab
ilit
y
,
e
f
f
icie
n
t
co
m
p
o
n
e
n
t
s
to
r
ag
e
an
d
r
etr
iev
al
is
ch
al
le
n
g
i
n
g
ta
s
k
b
ec
au
s
e
w
h
o
le
s
u
cc
es
s
es
o
f
s
o
f
t
w
ar
e
d
ep
en
d
u
p
o
n
t
h
ese
k
e
y
asp
ec
t
s
.
Ou
r
p
r
o
p
o
s
ed
alg
o
r
ith
m
m
u
s
t
p
r
o
v
e
to
b
e
v
er
y
b
en
ef
icia
l
f
o
r
th
is
p
u
r
p
o
s
e
t
h
a
t
s
h
o
w
,
h
o
w
w
e
ca
n
e
f
f
ec
ti
v
el
y
m
a
n
a
g
e
t
h
e
r
ep
o
s
ito
r
ies
th
r
o
u
g
h
cl
u
s
ter
i
n
g
an
d
m
ak
e
t
h
e
p
r
o
ce
s
s
au
to
m
atic
(
f
o
r
r
etr
iev
al
o
f
co
m
p
o
n
e
n
t
s
)
.
T
ab
le
1
.
P
s
eu
d
o
C
o
d
e
o
f
A
l
g
o
r
ith
m
P
se
u
d
o
c
o
d
e
o
f
a
l
g
o
r
i
t
h
m
I
n
i
t
i
a
l
i
z
e
T
i
F
o
r
(
T
i
=
1
,
T
i
< T
n
,
i
+
+
)
{
P
e
r
f
o
r
m C
R
}
S
R
F
R
a
n
d
N
S
R
NF
R
C
1
F
R
C
2
NF
R
F
R
W
C
(F
R
NF
R
W
C
(N
F
R
C
r
e
a
t
e
D
D
;
F
o
r
e
a
c
h
D
D
i
(
i
=
0
,
……,
n
)
DD
0
W
C(F
R
)
DD
1
W
C(N
F
R
)
w
h
e
r
e
W
c
(F
R
)
≠
Wc
(N
F
R
)
W
i
W
H
C(F
R
W
i
--
;
N
e
t
w
o
r
k
w
e
i
g
h
t
i
n
i
t
i
a
l
i
z
e
W
e
i
g
h
t
s(
w
o
r
d
,
v
a
l
u
e
)
F
o
r
(
i
=
1
t
o
n
)
If
{
W
i
= W
i
(W
H
C(F
R
)
)
{
El
se
{
B
a
c
k
P
r
o
p
a
g
a
t
e
Er
r
o
r
(
R
e
-
i
n
i
t
i
a
l
i
z
e
w
e
i
g
h
t
s,
w
o
r
d
,
v
a
l
u
e
)
}
}
}
C
r
e
a
t
e
R
R
K
(F
R
)
R
K
(N
F
R
)
F
K
R
/
/
T
e
st
d
o
c
u
me
n
t
/
/
R
e
t
r
i
e
v
a
l
o
f
c
o
mp
o
n
e
n
t
/
/
F
R
-
F
u
n
c
t
i
o
n
a
l
r
e
q
u
i
r
e
me
n
t
a
n
d
N
F
R
-
N
o
n
f
u
n
c
t
i
o
n
a
l
R
e
q
u
i
r
e
me
n
t
/
/
C
1
a
n
d
C
2
t
w
o
c
l
u
st
e
r
s
)
/
/
W
o
r
d
c
o
u
n
t
o
f
F
u
n
c
t
i
o
n
a
l
R
e
q
u
i
r
e
me
n
t
)
/
/
W
o
r
d
c
o
u
n
t
o
f
n
o
n
f
u
n
c
t
i
o
n
a
l
R
e
q
u
i
r
e
me
n
t
/
/
D
a
t
a
d
i
c
t
i
o
n
a
r
y
/
/
S
t
o
r
e
w
o
r
d
c
o
u
n
t
s (
k
e
y
w
o
r
d
s)
i
n
t
o
d
a
t
a
d
i
c
t
i
o
n
a
r
y
/
/
f
o
r
w
o
r
d
c
o
u
n
t
)
/
/
A
ssi
g
n
w
e
i
g
h
t
s a
c
c
o
r
d
i
n
g
t
o
h
i
g
h
e
st
v
a
l
u
e
o
f
k
e
y
w
o
r
d
a
n
d
so
o
n
/
/
w
e
i
g
h
t
s
d
e
c
r
e
a
se
a
s w
o
r
d
c
o
u
n
t
d
e
c
r
e
a
se
/
/
a
ssi
g
n
w
e
i
g
h
t
s t
o
a
l
l
w
o
r
d
c
o
u
n
t
(
k
e
y
w
o
r
d
s)
/
/
c
h
e
c
k
N
e
u
r
a
l
n
e
t
w
o
r
k
A
l
g
o
r
i
t
h
m
,
A
ssi
g
n
f
i
r
st
w
e
i
g
h
t
o
r
p
r
i
o
r
i
t
y
t
o
h
i
g
h
e
st
w
o
r
d
c
o
u
n
t
/
/
c
r
e
a
t
e
R
e
p
o
si
t
o
r
i
e
s
/
/
S
t
o
r
e
k
e
y
w
o
r
d
s i
n
t
o
r
e
p
o
si
t
o
r
i
e
s
/
/
F
e
t
c
h
k
e
y
w
o
r
d
s fr
o
m re
p
o
si
t
o
r
y
t
o
r
e
t
r
i
e
v
e
d
o
c
u
me
n
t
4.
I
M
P
L
E
M
E
NT
AT
I
O
N
O
F
A
L
G
O
R
I
T
H
M
I
n
p
u
t: So
f
t
w
ar
e
d
o
cu
m
e
n
t,
(
S
o
f
t
w
a
r
e
R
eq
u
ir
e
m
e
n
t Sp
ec
i
f
ic
atio
n
)
,
e
.
g
.
Ho
tel
Ma
n
a
g
e
m
e
n
t
S
y
s
te
m
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
4
,
No
.
2
,
J
u
n
e
201
5
:
5
3
–
61
56
Fig
u
r
e
1
.
C
r
ea
te
t
w
o
C
lu
s
ter
s
o
f
Fu
n
ctio
n
al
an
d
No
n
F
u
n
cti
o
n
al
R
eq
u
ir
e
m
e
n
ts
t
h
r
o
u
g
h
k
m
ea
n
Alg
o
r
it
h
m
an
d
S
h
o
w
W
o
r
d
C
o
u
n
t
s
o
f
K
e
y
w
o
r
d
s
Fig
u
r
e
2
Ass
i
g
n
W
ei
g
h
ts
to
K
e
y
w
o
r
d
s
,
A
cc
o
r
d
in
g
th
e
ir
W
o
r
d
C
o
u
n
t
V
al
u
e
a
n
d
S
h
o
w
W
o
r
k
in
g
o
f
B
ac
k
P
r
o
p
ag
atio
n
A
l
g
o
r
ith
m
Fig
u
r
e
3
.
R
ep
o
s
it
o
r
y
o
f
F
u
n
ct
i
o
n
al
R
eq
u
ir
e
m
e
n
t
Fig
u
r
e
4
.
Select
K
e
y
w
o
r
d
f
r
o
m
D
r
o
p
D
o
w
n
l
is
t
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
Desig
n
a
n
A
lg
o
r
ith
m
fo
r
S
o
ftw
a
r
e
Dev
elo
p
men
t in
C
b
s
e
E
n
vi
r
o
n
men
t u
s
in
g
…
(
A
mit V
erma
)
57
Fig
u
r
e
5
.
s
h
o
w
th
e
Selecte
d
D
o
cu
m
e
n
t B
ased
Fig
u
r
e
6
.
W
h
en
o
n
e
clic
k
s
o
n
an
y
f
ile
to
s
h
o
w
,
it o
n
en
ter
i
n
g
k
e
y
w
o
r
d
5.
E
XP
E
R
I
M
E
NT
A
L
RE
SUL
T
S
T
h
is
s
ec
tio
n
d
escr
ib
es
th
e
d
e
tail
o
f
ex
p
er
i
m
en
tal
r
es
u
lt
s
o
f
o
u
r
d
o
cu
m
en
t
r
etr
iev
al
p
r
o
ce
s
s
w
i
th
f
ac
to
r
s
,
ac
co
r
d
in
g
to
p
r
o
p
o
s
ed
alg
o
r
ith
m
.
I
t
also
p
r
o
v
id
es
u
s
t
h
e
co
m
p
ar
is
o
n
o
f
o
u
r
p
r
o
p
o
s
ed
id
ea
w
it
h
th
e
ex
is
t
in
g
tech
n
iq
u
e
o
f
i
n
te
g
r
ated
class
if
ica
tio
n
s
c
h
e
m
e
f
o
r
ef
f
ec
tiv
e
r
etr
iev
a
l
o
f
co
m
p
o
n
en
ts
b
ased
o
n
k
e
y
w
o
r
d
s
.
T
ab
le
2
.
Sh
o
w
C
o
m
p
ar
i
s
o
n
s
o
f
P
r
o
p
o
s
ed
T
ec
h
n
iq
u
e
w
ith
E
x
is
ti
n
g
R
etr
iev
a
l T
ec
h
n
iq
u
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
4
,
No
.
2
,
J
u
n
e
201
5
:
5
3
–
61
58
Fig
u
r
e
7
.
Sh
o
w
s
v
alu
e
s
o
f
all
f
ac
to
r
s
an
d
click
o
n
g
r
ap
h
b
u
tt
o
n
to
s
h
o
w
g
r
ap
h
s
I
n
th
i
s
F
i
g
u
r
e
7
,
en
ter
t
h
e
v
a
l
u
es
o
f
al
l
f
ac
to
r
s
s
h
o
w
n
a
b
o
v
e.
W
h
en
s
o
m
eo
n
e
f
etch
k
e
y
wo
r
d
f
r
o
m
r
ep
o
s
ito
r
y
to
s
ea
r
ch
t
h
e
d
o
cu
m
en
t,
t
h
en
t
h
ese
f
ac
to
r
s
v
a
lu
es
b
ased
o
n
k
e
y
w
o
r
d
s
h
o
w
n
o
n
co
n
s
o
le
an
d
i
n
s
ec
o
n
d
co
lu
m
n
e
n
ter
th
e
f
ac
to
r
s
v
al
u
es o
f
p
r
ev
io
u
s
tec
h
n
iq
u
e
w
it
h
w
h
ich
w
e
co
m
p
ar
e
o
u
r
r
esu
lt
s
.
5
.
1
.
Co
m
pa
riso
n o
f
Resul
t
s
ba
s
e
d o
n
Acc
ura
cy
I
n
th
i
s
Fi
g
u
r
e
8
,
it
s
h
o
w
s
t
h
e
ac
cu
r
ac
y
o
f
o
u
r
r
etr
iev
al
p
r
o
ce
s
s
o
f
co
m
p
o
n
en
t
(
o
r
k
e
y
w
o
r
d
)
f
r
o
m
r
ep
o
s
ito
r
y
.
W
e
co
m
p
ar
e
o
u
r
a
lg
o
r
ith
m
’
s
e
f
f
icie
n
c
y
w
it
h
th
e
ex
is
t
in
g
tec
h
n
iq
u
e
e.
g
.
I
n
te
g
r
ated
class
i
f
icatio
n
s
ch
e
m
e
.
I
n
th
i
s
s
ch
e
m
e,
t
w
o
k
e
y
w
o
r
d
s
w
er
e
in
te
g
r
ated
in
p
ar
ticu
lar
iter
atio
n
f
o
r
s
ea
r
ch
i
n
g
co
m
p
o
n
e
n
ts
f
r
o
m
r
ep
o
s
ito
r
y
.
T
h
e
ac
c
u
r
ac
y
o
f
t
h
is
co
m
p
o
n
e
n
t
r
etr
ie
v
al
m
et
h
o
d
is
n
o
t
b
etter
t
h
an
o
u
r
p
r
o
p
o
s
ed
m
eth
o
d
f
o
r
ef
f
ec
tiv
e
r
etr
ie
v
al
o
f
co
m
p
o
n
e
n
ts
.
Fig
u
r
e
8
.
T
h
e
A
cc
u
r
ac
y
o
f
o
u
r
R
etr
ie
v
al
P
r
o
ce
s
s
o
f
C
o
m
p
o
n
en
t (
o
r
k
e
y
w
o
r
d
)
f
r
o
m
F
ep
o
s
it
o
r
y
5
.
2
.
Co
m
pa
r
is
o
n o
f
Resul
t
s
ba
s
e
d o
n Re
ca
ll
I
n
t
h
is
Fi
g
u
r
e
9
,
it
s
h
o
w
s
th
e
p
r
ec
is
io
n
f
ac
to
r
o
f
o
u
r
r
etr
iev
al
p
r
o
ce
s
s
o
f
co
m
p
o
n
e
n
t
(
o
r
k
e
y
w
o
r
d
)
f
r
o
m
r
ep
o
s
ito
r
y
.
W
e
co
m
p
a
r
e
o
u
r
alg
o
r
ith
m
’
s
ef
f
icie
n
c
y
w
it
h
t
h
e
e
x
is
tin
g
tec
h
n
iq
u
e
e.
g
.
I
n
te
g
r
ated
class
i
f
icatio
n
s
c
h
e
m
e
.
I
n
p
r
ec
is
io
n
f
ac
to
r
,
it
tells
u
s
t
h
e
r
atio
o
f
r
etr
iev
al
o
f
r
elev
a
n
t
co
m
p
o
n
e
n
t
s
f
r
o
m
ir
r
elev
an
t
co
m
p
o
n
e
n
t
s
an
d
to
tal
n
o
.
o
f
co
m
p
o
n
e
n
ts
.
T
h
e
ac
cu
r
ate
v
al
u
e
o
f
p
r
ec
is
io
n
is
b
et
w
ee
n
0
an
d
1
.
T
h
e
p
r
ec
i
s
io
n
o
f
o
u
r
r
etr
ie
v
al
p
r
o
ce
s
s
o
f
co
m
p
o
n
en
t
s
co
m
es
v
er
y
ac
c
u
r
ate
b
et
w
ee
n
o
a
n
d
1
an
d
less
th
a
n
p
r
ev
io
u
s
r
esu
lt
s
.
Fig
u
r
e
9
.
T
h
e
Pre
cisi
o
n
Facto
r
o
f
o
u
r
R
etr
iev
al
P
r
o
ce
s
s
o
f
C
o
m
p
o
n
en
t (
o
r
k
e
y
w
o
r
d
)
f
r
o
m
R
ep
o
s
ito
r
y
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
Desig
n
a
n
A
lg
o
r
ith
m
fo
r
S
o
ftw
a
r
e
Dev
elo
p
men
t in
C
b
s
e
E
n
vi
r
o
n
men
t u
s
in
g
…
(
A
mit V
erma
)
59
5
.
3
.
Co
m
pa
riso
n o
f
Resul
t
s
ba
s
e
d o
n P
re
cisi
o
n
I
n
th
i
s
F
i
g
u
r
e
10
,
it
s
h
o
w
s
t
h
e
r
ec
all
f
ac
to
r
o
f
o
u
r
r
etr
iev
al
p
r
o
ce
s
s
o
f
co
m
p
o
n
e
n
t
(
o
r
k
e
y
wo
r
d
)
f
r
o
m
r
ep
o
s
ito
r
y
.
W
e
co
m
p
ar
e
o
u
r
a
lg
o
r
ith
m
’
s
e
f
f
icie
n
c
y
w
it
h
th
e
ex
is
t
in
g
tec
h
n
iq
u
e
e.
g
.
I
n
te
g
r
ated
class
i
f
icatio
n
s
ch
e
m
e
.
I
n
r
ec
all
f
ac
to
r
,
it
tell
s
u
s
th
e
r
at
io
o
f
r
etr
ie
v
al
o
f
r
e
lev
an
t
co
m
p
o
n
e
n
ts
f
r
o
m
to
tal
n
o
.
o
f
co
m
p
o
n
e
n
ts
.
T
h
e
ac
cu
r
ate
v
alu
e
o
f
r
ec
all
i
s
b
et
w
ee
n
0
a
n
d
1
.
T
h
e
r
ec
all
o
f
o
u
r
r
etr
iev
al
p
r
o
ce
s
s
o
f
co
m
p
o
n
en
ts
co
m
e
s
v
er
y
ac
cu
r
ate
b
et
w
ee
n
o
an
d
1
an
d
g
r
ea
ter
th
a
n
p
r
ev
io
u
s
r
es
u
lt
s
.
Fig
u
r
e
1
0
.
R
ec
all
Facto
r
o
f
Ou
r
R
etr
iev
al
P
r
o
ce
s
s
o
f
C
o
m
p
o
n
en
t (
o
r
k
e
y
w
o
r
d
)
f
r
o
m
R
ep
o
s
ito
r
y
6.
CO
NCLU
SI
O
N
AND
F
U
T
U
RE
SCO
P
E
T
h
is
s
t
u
d
y
e
m
p
ir
icall
y
v
alid
a
tes
t
h
e
co
n
ce
p
t
o
f
r
e
u
s
e
in
s
o
f
t
w
ar
e
d
ev
elo
p
m
e
n
t.
R
e
u
s
e
a
p
ar
t
o
r
co
m
p
o
n
e
n
t
o
f
s
o
f
t
w
ar
e
m
a
y
o
v
er
co
m
e
th
e
v
ar
io
u
s
c
h
al
l
en
g
e
s
ex
i
s
t
i
n
d
ev
elo
p
m
e
n
t
o
f
s
o
f
t
w
ar
e.
I
n
th
i
s
r
esear
ch
p
ap
er
,
p
r
o
p
o
s
e
an
alg
o
r
ith
m
t
h
at
p
r
o
v
id
es
er
r
o
r
f
r
ee
an
d
au
to
m
atic
p
r
o
ce
s
s
(
f
o
r
r
etr
iev
al
o
f
th
e
co
m
p
o
n
e
n
t
s
)
w
h
ile
r
e
u
s
e
o
f
t
h
e
co
m
p
o
n
en
t.
I
t
also
p
r
o
v
id
e
s
t
h
e
e
f
f
icien
t
s
to
r
ag
e
m
et
h
o
d
o
f
co
m
p
o
n
en
t
s
b
y
p
ass
in
g
t
h
r
o
u
g
h
n
e
u
r
al
n
et
w
o
r
k
an
d
p
ar
a
m
eter
s
b
ased
o
n
u
s
er
c
h
o
ice
th
r
o
u
g
h
cl
u
s
ter
i
n
g
.
I
n
la
s
t,
r
etr
iev
al
m
u
s
t
b
e
d
o
n
e
b
y
u
s
i
n
g
k
e
y
wo
r
d
b
ased
r
etr
iev
al
m
e
th
o
d
.
P
r
o
p
o
s
ed
alg
o
r
ith
m
is
v
er
y
e
f
f
ec
tiv
e
in
t
h
e
er
r
o
r
co
r
r
ec
tio
n
an
d
d
etec
tio
n
w
it
h
u
s
er
b
ase
ch
o
ice
w
h
ile
c
h
o
ice
o
f
co
m
p
o
n
en
t
f
o
r
r
eu
s
a
b
ilit
y
f
o
r
e
f
f
icien
t
r
etr
iev
al
i
s
t
h
er
e.
T
h
is
al
g
o
r
it
h
m
also
s
er
v
e
s
b
etter
d
ec
is
io
n
m
ak
in
g
ap
p
r
o
ac
h
a
n
d
b
est
o
p
tim
a
l
co
m
p
o
n
e
n
t
w
il
l b
e
s
elec
ted
.
6
.
1
.
F
uture
s
co
pe
I
n
th
e
Fu
t
u
r
e
Sco
p
e,
tex
t
d
o
c
u
m
e
n
t
ca
n
b
e
u
s
ed
f
o
r
th
e
e
f
f
ec
tiv
e
co
m
p
o
n
e
n
t
(
o
r
k
ey
w
o
r
d
s
)
s
to
r
ag
e
an
d
r
etr
iev
al
m
e
th
o
d
f
o
r
r
eu
s
a
b
ilit
y
i
n
d
ev
elo
p
m
e
n
t
o
f
s
o
f
t
w
ar
e.
Ot
h
er
ca
n
ta
k
e
co
s
t
a
n
d
ti
m
e
attr
ib
u
tes
in
t
o
co
n
s
id
er
atio
n
f
o
r
o
p
ti
m
al
co
m
p
o
n
en
t
s
elec
t
io
n
a
n
d
f
o
r
e
f
f
ec
tiv
e
r
etr
ie
v
al
p
r
o
ce
s
s
o
f
co
m
p
o
n
en
ts
alo
n
g
w
it
h
th
e
test
o
r
tex
t
d
o
cu
m
en
ts
.
I
n
f
u
t
u
r
e,
th
e
ap
p
r
o
ac
h
o
f
ef
f
ec
tiv
e
s
to
r
ag
e
a
n
d
r
etr
iev
al
o
f
c
o
m
p
o
n
en
t
s
ca
n
b
e
e
m
b
ed
d
ed
w
i
th
o
th
er
allied
te
ch
n
o
lo
g
ies
s
u
ch
a
s
Ma
tlab
,
PHP
an
d
P
y
th
o
n
.
I
m
p
le
m
e
n
tati
o
n
ca
n
b
e
ca
r
r
ied
o
u
t
w
it
h
t
h
e
h
elp
o
f
v
ar
io
u
s
s
i
m
u
lato
r
s
s
u
c
h
as
NS2
,
NS3
an
d
w
ith
t
h
e
h
elp
o
f
m
icr
o
p
r
o
ce
s
s
o
r
s
f
o
r
in
ter
r
u
p
t
g
en
er
atio
n
in
ca
s
e
w
r
o
n
g
k
e
y
w
o
r
d
s
elec
ted
o
r
p
ath
o
f
d
o
cu
m
en
t f
i
le
w
ill
n
o
t b
e
co
r
r
ec
t.
RE
F
E
R
E
NC
E
S
[
1
]
X
W
in
sto
n
W
Ro
y
c
e
.
M
a
n
a
g
in
g
th
e
d
e
v
e
lo
p
m
e
n
t
o
f
larg
e
so
f
t
wa
re
s
y
ste
m
s.
IEE
E
,
1
9
7
0
:
1
-
9
.
[
2
]
Wi
ll
ia
m
HE
Da
y
,
He
rb
e
rt
Ed
e
lsb
ru
n
n
e
r.
Ef
f
icie
n
t
A
l
g
o
rit
h
m
s
f
o
r
Ag
g
lo
m
e
ra
ti
v
e
Hie
ra
rc
h
ica
l
Clu
ste
rin
g
M
e
th
o
d
.
J
o
u
rn
a
l
o
f
Cla
ss
if
ica
ti
o
n
.
1
9
8
4
:
7
-
24
.
[
3
]
Ja
m
e
s
C
B
e
z
d
e
k
,
Ro
b
e
rt
Eh
rli
c
h
,
W
il
li
a
m
F
u
ll
.
F
CM
:
T
h
e
F
u
z
z
y
C
-
M
e
a
n
s
Clu
ste
rin
g
A
l
g
o
rit
h
m
.
Co
mp
u
ter
s
&
Ge
o
sc
ie
n
c
e
s
.
1
9
8
4
;
10
(
2
-
3
):
1
9
1
-
2
0
3
.
[
4
]
M
a
rti
n
F
o
d
sle
tt
e
M
o
ll
a
r.
A
sc
a
led
c
o
n
ju
g
a
te
g
ra
d
ien
t
a
lg
o
ri
th
m
f
o
r
f
a
st
su
p
e
rv
i
se
d
lea
rn
in
g
.
Un
iv
e
rsit
y
o
f
A
a
rh
u
s,
1
9
9
3
;
6
:
5
2
5
-
5
3
3
.
[
5
]
Da
n
iel
S
v
o
z
il
,
V
lad
im
ir
Kv
a
sn
iek
a
,
Jie
P
o
sp
ich
a
l
.
In
tr
o
d
u
c
ti
o
n
to
m
u
lt
i
-
lay
e
r
fe
e
d
-
f
o
r
w
a
rd
n
e
u
ra
l
n
e
tw
o
rk
s.
EL
S
EV
I
ER
,
1
9
9
7
:
43
-
62
.
[
6
]
Jo
h
a
n
n
e
s S
a
m
e
ti
n
g
e
r.
S
o
f
t
w
a
r
e
En
g
in
e
e
rin
g
w
it
h
Re
u
sa
b
le Co
m
p
o
n
e
n
ts
.
S
p
ri
n
g
e
r
-
Ver
la
g
,
1
9
9
7
:
1
-
2
7
2
.
[
7
]
V
u
T
ra
n
'
Da
r
-
Biau
L
iu
.
Co
m
p
o
n
e
n
t
-
b
a
se
d
S
y
ste
m
s
De
v
e
lo
p
m
e
n
t:
Ch
a
ll
e
n
g
e
s
a
n
d
L
e
ss
o
n
s
L
e
a
rn
e
d
.
IEE
E
,
1
9
9
7
:
452
-
4
6
2
.
[
8
]
Bro
w
n
,
A
W
,
W
a
ll
n
a
u
,
KC.
T
h
e
c
u
rre
n
t
sta
te o
f
CBS
E.
IEE
E
s
o
ft
w
a
re
.
1
9
9
8
;
1
5
(5
)
:
3
7
-
46
.
[
9
]
M
a
rie Ch
a
v
e
n
t.
A
m
o
n
o
t
h
e
ti
c
c
lu
ste
rin
g
m
e
th
o
d
.
P
a
tt
e
rn
Re
c
o
g
n
i
ti
o
n
L
e
tt
e
rs 1
9
,
1
9
9
8
:
9
8
9
-
9
9
6
.
[
1
0
]
A
.
k
J
a
in
,
M
N M
u
rth
y
,
P
J F
ly
n
n
.
Da
ta Cl
u
ste
rin
g
:
A
Re
v
ie
w
.
A
CM
Co
m
p
u
ti
n
g
S
u
rv
e
y
s,
19
9
9
;
31
(
3
)
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
4
,
No
.
2
,
J
u
n
e
201
5
:
5
3
–
61
60
[
1
1
]
T
a
k
a
sh
i
Ko
b
a
y
a
sh
i
a
n
d
M
o
t
o
sh
i
S
a
e
k
i.
S
o
f
t
w
a
re
D
e
v
e
lo
p
m
e
n
t
Ba
se
d
o
n
S
o
f
twa
re
P
a
tt
e
rn
Ev
o
lu
ti
o
n
.
IEE
E
,
1
9
9
9
:
18
-
25
.
[
1
2
]
Bo
g
d
a
n
G
a
b
r
y
s
,
A
n
d
rz
e
j
Ba
r
g
ie
la
.
Ge
n
e
ra
l
F
u
z
z
y
M
in
-
M
a
x
N
e
u
ra
l
Ne
tw
o
rk
f
o
r
Clu
ste
rin
g
a
n
d
Clas
si
f
ica
ti
o
n
.
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
Ne
u
ra
l
Ne
t
wo
rk
s.
2
0
0
0
;
11
(
3
):
7
6
9
-
7
8
3
.
[
1
3
]
A
n
il
c
h
u
terv
e
d
i,
Kra
f
t
f
o
o
d
s,
P
a
u
l
E.
G
re
e
n
.
K m
o
d
e
s clu
ste
rin
g
.
J
o
u
rn
a
l
o
f
c
la
ss
if
ica
t
io
n
.
2
0
0
1
:
35
-
5
5
.
[
1
4
]
T
a
p
a
s
Ka
n
u
n
g
o
,
Da
v
id
M
.
M
o
u
n
t,
Na
th
a
n
S
Ne
tan
y
a
h
u
,
Ch
rist
i
n
e
D
P
iatk
o
,
Ru
t
h
S
il
v
e
rm
a
n
,
An
g
e
la
Y
W
u
.
A
n
Eff
icie
n
t
k
-
Me
a
n
s Clu
ste
rin
g
A
lg
o
rit
h
m
:
A
n
a
l
y
sis
a
n
d
Im
p
le
m
e
n
ta
ti
o
n
.
IEE
E
T
r
a
n
sa
c
ti
o
n
s
o
n
P
a
tt
e
rn
An
a
lys
is
a
n
d
M
a
c
h
in
e
In
telli
g
e
n
c
e
.
2
0
0
2
;
24
(
7
):
8
8
1
-
8
9
2
.
[
1
5
]
M
a
rk
Cro
w
n
e
.
W
h
y
so
f
t
wa
re
p
ro
d
u
c
t
sta
rt
-
u
p
s f
a
il
a
n
d
w
h
a
t
to
d
o
a
b
o
u
t
it
.
I
EE
E
,
2
0
0
2
:
3
3
8
-
3
4
3
.
[
1
6
]
V
it
h
a
ra
n
a
,
P
.
Risk
s
a
n
d
c
h
a
ll
e
n
g
e
s
o
f
c
o
m
p
o
n
e
n
t
-
b
a
se
d
s
o
f
twa
re
d
e
v
e
lo
p
m
e
n
t
.
Co
mm
u
n
ica
ti
o
n
s
o
f
t
h
e
ACM
.
2
0
0
3
;
46
(
8
):
67
-
7
2
.
[
1
7
]
M
in
g
-
Ch
u
a
n
Hu
n
g
,
Ju
n
g
p
in
W
u
,
Jin
-
Hu
a
C
h
a
n
g
,
Do
n
-
L
in
Ya
n
g
.
A
n
Eff
ici
e
n
t
k
-
M
e
a
n
s
Clu
ste
rin
g
A
lg
o
rit
h
m
Us
in
g
S
im
p
le P
a
rti
ti
o
n
in
g
.
J
o
u
rn
a
l
o
f
In
f
o
rm
a
ti
o
n
S
c
i
e
n
c
e
a
n
d
En
g
in
e
e
rin
g
.
2
0
0
5
;
2
1
:
1
1
5
7
-
1
1
7
7
.
[
1
8
]
Rich
a
rd
W
S
e
lb
y
.
En
a
b
li
n
g
Re
u
se
-
Ba
se
d
S
o
f
t
w
a
re
De
v
e
lo
p
m
e
n
t
o
f
L
a
rg
e
-
S
c
a
le S
y
ste
m
s.
IEE
E
.
2
0
0
5
:
4
9
5
-
5
1
0
.
[
1
9
]
Ha
n
s
-
He
r
m
a
n
n
Bo
c
k
.
Orig
in
s
a
n
d
e
x
ten
sio
n
s
o
f
th
e
k
-
m
e
a
n
s
a
lg
o
rit
h
m
in
c
lu
ste
r
a
n
a
ly
sis.
El
e
c
tro
n
ic
J
o
u
r
n
a
l
fo
r
Histo
ry
o
f
Pr
o
b
a
b
il
it
y
a
n
d
S
t
a
ti
sti
c
s
,
2
0
0
8
;
4
(
2
):
1
-
1
8
.
[
2
0
]
A
b
b
a
s
Y
A
l
Ba
y
a
ti
,
N
a
j
m
a
d
d
in
A
S
u
laim
a
n
,
G
u
ln
a
r
W
.
S
a
d
iq
.
M
o
d
if
ied
Co
n
ju
g
a
te
G
r
a
d
ien
t
F
o
rm
u
la
f
o
r
B
a
c
k
P
r
o
p
a
g
a
ti
o
n
Ne
u
ra
l
Ne
tw
o
rk
A
l
g
o
rit
h
m
.
J
o
u
r
n
a
l
o
f
C
o
mp
u
ter
S
c
ie
n
c
e
s
.
2
0
0
9
;
5
(
1
1
):
8
4
9
-
8
5
6
.
[
2
1
]
Dix
i
t
A
,
S
a
x
e
n
a
P
C.
S
o
ft
wa
re
Co
mp
o
n
e
n
t
Retrie
v
a
l
Us
in
g
Ge
n
e
ti
c
Al
g
o
rit
h
ms
.
I
n
tern
a
ti
o
n
a
l
Co
n
f
e
re
n
c
e
o
n
Co
m
p
u
ter an
d
A
u
to
m
a
ti
o
n
E
n
g
in
e
e
rin
g
,
IEE
E.
2
0
0
9
;
1
5
1
-
15
.
[
2
2
]
G
h
o
la
m
Re
z
a
S
h
a
h
m
o
h
a
m
m
a
d
ia,
S
a
e
e
d
Ja
li
li
,
S
e
y
e
d
M
o
h
a
m
m
a
d
Ho
ss
e
in
Ha
sh
e
m
in
e
jad
.
Id
e
n
ti
f
ica
ti
o
n
o
f
S
y
s
te
m
S
o
f
tw
a
r
e
Co
m
p
o
n
e
n
ts
Us
in
g
Clu
s
terin
g
A
p
p
ro
a
c
h
.
J
o
u
rn
a
l
o
f
o
b
jec
t
T
e
c
h
n
o
lo
g
y
,
2
0
1
0
;
9
(
6
)
:
77
-
9
8
.
[
2
3
]
S
A
ji
th
a
,
T
V
S
u
re
sh
Ku
m
a
r,
D
Ev
a
n
g
e
li
n
G
e
e
th
a
,
K
Ra
jn
ik
a
n
th
.
Per
fo
rm
a
n
c
e
Pre
d
icti
o
n
I
n
E
a
rly
S
t
a
g
e
s
O
f
S
o
ft
w
a
re
S
y
ste
ms
:
A
rtif
icia
l
Ne
u
r
a
l
Ne
two
rk
M
o
d
e
l
.
IC
CCCT
,
2
0
1
0
:
7
4
3
-
7
4
7
.
[
2
4
]
A
r
v
in
d
e
r
Ka
u
r,
Ku
lv
in
d
e
r
S
i
n
g
h
M
a
n
n
.
Co
m
p
o
n
e
n
t
S
e
lec
ti
o
n
f
o
r
Co
m
p
o
n
e
n
t
b
a
se
d
S
o
f
tw
a
r
e
En
g
in
e
e
rin
g
.
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
C
o
mp
u
ter
Ap
p
l
ica
ti
o
n
s
,
2
0
1
0
;
2
(
1
):
1
0
9
-
1
1
4
.
[
2
5
]
S
o
n
ia
M
a
n
h
a
s,
P
a
rv
in
d
e
r
S
S
a
n
d
h
u
,
V
i
n
a
y
Ch
o
p
ra
,
Nirv
a
ir
Ne
e
ru
.
Id
e
n
ti
f
ica
ti
o
n
o
f
Re
u
sa
b
le
S
o
f
twa
re
M
o
d
u
les
i
n
F
u
n
c
ti
o
n
Orie
n
ted
S
o
f
tw
a
re
S
y
ste
m
s
u
sin
g
Ne
u
ra
l
Ne
t
w
o
rk
B
a
se
d
T
e
c
h
n
iq
u
e
.
W
o
rl
d
A
c
a
d
e
m
y
o
f
S
c
ien
c
e
,
En
g
in
e
e
rin
g
a
n
d
T
e
c
h
n
o
lo
g
y
.
2
0
1
0
;
4
:
0
7
-
23
.
[
2
6
]
F
u
y
u
a
n
Ca
o
,
Jiy
e
L
ian
g
,
De
y
u
L
i
,
L
ian
g
Ba
i,
Ch
u
a
n
g
y
in
Da
n
g
.
A
d
issim
il
a
rit
y
m
e
a
su
re
f
o
r
th
e
k
-
M
o
d
e
s
c
lu
ste
ri
n
g
a
lg
o
rit
h
m
.
El
se
v
ier
.
2
0
1
1
:
1
2
0
-
127
.
[
2
7
]
Da
n
iel
M
ü
ll
n
e
r.
M
o
d
e
r
n
h
iera
rc
h
ica
l,
a
g
g
lo
m
e
ra
ti
v
e
c
lu
ste
rin
g
a
lg
o
rit
h
m
s
.
2
0
1
1
:
1
-
2
9
.
[
2
8
]
Ha
rp
re
e
t
S
in
g
h
,
V
is
h
a
l
Ku
m
a
r
To
o
ra
.
Ne
u
ro
F
u
z
z
y
L
o
g
ic
M
o
d
e
l
f
o
r
Co
m
p
o
n
e
n
t
Ba
se
d
S
o
f
tw
a
re
En
g
i
n
e
e
rin
g
.
An
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
E
n
g
in
e
e
rin
g
S
c
ien
c
e
s
.
2
0
1
1
;
1
:
303
-
3
1
4
.
[
2
9
]
Zh
e
n
-
G
u
o
Ch
e
,
T
z
u
-
A
n
Ch
ian
g
,
Zh
e
n
-
Hu
a
C
h
e
.
F
e
e
d
F
o
rw
a
rd
Ne
u
ra
l
Ne
tw
o
rk
s
T
ra
in
in
g
:
A
c
o
m
p
a
riso
n
b
e
tw
e
e
n
G
e
n
e
ti
c
a
l
g
o
rit
h
m
a
n
d
Ba
c
k
p
ro
p
a
g
a
ti
o
n
lea
rn
in
g
A
lg
o
rit
h
m
.
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
In
n
o
v
a
ti
v
e
Co
mp
u
ti
n
g
,
In
fo
rm
a
t
io
n
a
n
d
Co
n
tro
l
ICIC
In
t
e
rn
a
ti
o
n
a
l
.
2
0
1
1
;
7
(
10
):
5
8
3
9
-
5
8
5
0
.
[
3
0
]
V
a
n
e
e
t
Ka
u
r
,
S
h
iv
a
n
i
G
o
e
l.
F
a
c
e
t
s
o
f
S
o
f
tw
a
re
Co
m
p
o
n
e
n
t
Re
p
o
sit
o
ry
.
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
n
C
o
mp
u
ter
S
c
ie
n
c
e
a
n
d
En
g
i
n
e
e
rin
g
.
2
0
1
1
;
3
(
6
):
2
4
7
3
-
2
4
7
6
.
[
3
1
]
Ra
c
h
n
a
R
a
tra,
Na
v
n
e
e
t
S
in
g
h
Ra
n
d
h
a
w
a
,
P
a
rn
e
e
t
Ka
u
r,
G
u
rd
e
v
S
in
g
h
.
Early
P
re
d
ictio
n
o
f
F
a
u
lt
P
ro
n
e
M
o
d
u
les
u
sin
g
Clu
ste
ri
n
g
Ba
se
d
v
s.
Ne
u
ra
l
Ne
tw
o
r
k
A
p
p
ro
a
c
h
in
S
o
f
twa
re
S
y
st
e
m
s.
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
tro
n
ics
&
Co
mm
u
n
ica
ti
o
n
T
e
c
h
n
o
l
o
g
y
,
2
0
1
1
;
2
(
4
):
4
7
-
50
.
[
3
2
]
N
M
d
J
u
b
a
ir
Ba
sh
a
,
C
h
a
n
d
ra
M
o
h
a
n
.
A
stra
teg
y
to
id
e
n
ti
f
y
c
o
m
p
o
n
e
n
ts
u
sin
g
c
l
u
ste
rin
g
a
p
p
r
o
a
c
h
f
o
r
c
o
m
p
o
n
e
n
t
re
u
sa
b
il
it
y
.
Co
mp
u
ter
S
c
ien
c
e
&
I
n
fo
rm
a
t
io
n
T
e
c
h
n
o
l
o
g
y
.
2
0
1
2
;
397
–
4
0
6
.
[
3
3
]
A
n
u
p
a
m
a
Ka
u
r.
Im
p
a
c
t
o
f
T
r
a
in
in
g
F
u
n
c
ti
o
n
Ba
se
d
Ne
u
ra
l
Ne
tw
o
rk
o
n
Re
u
sa
b
le
S
o
f
t
wa
re
M
o
d
u
le
s
.
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
In
fo
rm
a
ti
o
n
T
e
c
h
n
o
l
o
g
ies
,
2
0
1
2
;
3
:
4
0
2
4
–
4
0
2
7
.
[
3
4
]
N
Ko
tes
w
a
r
a
Ra
o
,
G
S
rid
h
a
r
Re
d
d
y
.
Disc
o
v
e
r
y
o
f
P
re
li
m
in
a
r
y
C
e
n
tro
i
d
s
Us
in
g
Im
p
ro
v
e
d
K
-
M
e
a
n
s
Clu
ste
rin
g
A
l
g
o
rit
h
m
.
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
I
n
fo
rm
a
ti
o
n
T
e
c
h
n
o
lo
g
ies
.
2
0
1
2
;
3
:
4
5
5
8
-
1
5
6
1
.
[
3
5
]
S
u
re
sh
C
h
a
n
d
G
u
p
ta,
A
sh
o
k
Ku
m
a
r.
A
Ne
u
ra
l
Ne
tw
o
rk
b
a
se
d
M
e
th
o
d
to
O
p
ti
m
ize
th
e
S
o
f
twa
re
Co
m
p
o
n
e
n
t
S
e
a
rc
h
in
g
Re
su
lt
s in
K
-
M
o
d
e
l
.
I
n
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Co
m
p
u
ter
Ap
p
li
c
a
ti
o
n
s
.
2
0
1
3
;
72
(
7
):
20
-
2
7
.
[
3
6
]
K
u
ld
i
p
V
o
ra
,
S
h
ru
t
i
Ya
g
n
ik
.
A
S
u
rv
e
y
o
n
Ba
c
k
p
ro
p
a
g
a
ti
o
n
A
l
g
o
rit
h
m
s
f
o
r
F
e
e
d
f
o
r
w
a
rd
Ne
u
ra
l
Ne
tw
o
rk
s
.
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
E
n
g
in
e
e
rin
g
De
v
e
lo
p
me
n
t
a
n
d
Res
e
a
rc
h
.
2
0
1
3
:
1
9
3
-
1
9
7
.
[
3
7
]
S
a
n
d
e
e
p
Ku
m
a
r
Ja
in
,
M
a
n
u
P
ra
tap
S
in
g
h
.
Esti
m
a
ti
o
n
f
o
r
F
a
u
lt
s
P
re
d
ict
io
n
f
ro
m
Co
m
p
o
n
e
n
t
Ba
se
d
S
o
f
t
w
a
re
De
sig
n
u
sin
g
F
e
e
d
F
o
rw
a
rd
Ne
u
ra
l
Ne
tw
o
rk
s
.
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
A
d
v
a
n
c
e
d
Res
e
a
rc
h
i
n
Co
mp
u
ter
a
n
d
Co
mm
u
n
ica
ti
o
n
E
n
g
i
n
e
e
rin
g
,
2
0
1
3
;
2
(
7
):
2
6
3
1
-
2
6
6
0
.
[
3
8
]
Ka
m
m
n
a
M
a
h
a
jan
,
M
a
n
d
e
e
p
k
a
u
r.
Co
m
p
o
n
e
n
t
Re
tri
e
v
a
l
Us
in
g
G
e
n
e
ti
c
A
lg
o
rit
h
m
Ba
se
d
Op
ti
m
iza
t
io
n
T
e
c
h
n
iq
u
e
.
In
ter
n
a
t
io
n
a
l
jo
u
rn
a
l
o
f
c
o
mp
u
ter
sc
ien
c
e
a
n
d
T
e
c
h
n
o
lo
g
y
.
2
0
1
3
;
4
(
2
)
58
5
-
5
8
7
.
[
3
9
]
M
a
h
sa
Ha
sa
n
i
S
a
d
i,
Er
ic
Yu
.
A
n
a
ly
z
in
g
th
e
Ev
o
lu
ti
o
n
o
f
S
o
f
tw
a
r
e
De
v
e
lo
p
m
e
n
t:
F
ro
m
Cre
a
ti
v
e
Ch
a
o
s
to
S
o
f
tw
a
re
Eco
sy
st
e
m
s.
IEE
E
,
2
0
1
4
:.
1
-
11
.
[
4
0
]
X
iao
h
u
i
C
u
i,
T
h
o
m
a
s
E
P
o
t
o
k
.
Do
c
u
m
e
n
t
Clu
ste
rin
g
A
n
a
l
y
sis
Ba
s
e
d
o
n
Hy
b
rid
P
S
O+
K
-
m
e
a
n
s
A
l
g
o
rit
h
m
.
Ap
p
li
e
d
S
o
ft
w
a
re
En
g
in
e
e
rin
g
Res
e
a
rc
h
Gr
o
u
p
,
Co
mp
u
t
a
ti
o
n
a
l
S
c
ien
c
e
s a
n
d
E
n
g
in
e
e
rin
g
Div
isio
n
.
1
-
8.
[
4
1
]
T
o
n
g
G
a
o
,
Hu
i
M
a
,
I
L
in
g
Ye
n
,
L
a
ti
f
u
r
Kh
a
n
,
F
a
ro
k
h
Ba
sta
n
i
.
A
Re
p
o
sito
ry
f
o
r
Co
m
p
o
n
e
n
t
-
Ba
se
d
Em
b
e
d
d
e
d
S
o
f
tw
a
r
e
De
v
e
lo
p
m
e
n
t.
1
-
3
7
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
Desig
n
a
n
A
lg
o
r
ith
m
fo
r
S
o
ftw
a
r
e
Dev
elo
p
men
t in
C
b
s
e
E
n
vi
r
o
n
men
t u
s
in
g
…
(
A
mit V
erma
)
61
[
4
2
]
P
a
rm
in
d
e
r
Ka
u
r,
Ja
rn
a
il
S
in
g
h
,
Ha
rd
e
e
p
S
in
g
h
.
Co
m
p
o
n
e
n
t
S
e
le
c
ti
o
n
Re
p
o
s
it
o
ry
w
it
h
Risk
Id
e
n
ti
f
ica
ti
o
n
.
IEE
E
,
2
0
1
4
:
5
2
4
-
5
3
1
.
[
4
3
]
S
w
a
ti
T
h
a
k
ra
l,
S
h
ra
d
d
h
a
S
a
g
a
r,
V
in
a
y
.
Re
u
sa
b
il
it
y
in
Co
m
p
o
n
e
n
t
Ba
se
d
S
o
f
twa
r
e
De
v
e
lo
p
m
e
n
t
-
A
Re
v
ie
w
.
W
o
rld
Ap
p
li
e
d
S
c
ie
n
c
e
s Jo
u
rn
a
l
.
2
0
1
4
;
3
1
(
1
2
):
2
0
6
8
-
2
0
7
2
.
[
4
4
]
A
d
n
a
n
Kh
a
n
,
Kh
a
li
d
K
h
a
n
,
M
u
h
a
m
m
a
d
Am
ir,
M
N
A
Kh
a
n
.
A
Co
m
p
o
n
e
n
t
-
Ba
se
d
F
ra
m
e
w
o
r
k
f
o
r
S
o
f
twa
re
Re
u
sa
b
il
it
y
.
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
S
o
ft
wa
re
En
g
in
e
e
rin
g
a
n
d
Its
Ap
p
li
c
a
ti
o
n
s
.
D
o
i.
1
0
.
1
4
2
5
7
/
ij
s
e
ia.2
0
1
4
.
8
.
1
0
.
0
2
.
2
0
1
4
;
8
(1
0
):
1
3
-
24
.
[
4
5
]
In
d
e
rji
t
sin
g
h
,
A
sh
ima
S
in
g
h
.
A
S
u
rv
e
y
o
n
v
a
rio
u
s
Co
m
p
o
n
e
n
t
R
e
p
o
sito
rie
s
wit
h
d
e
ta
il
stu
d
y
o
f
d
i
ff
e
re
n
t
M
e
th
o
d
s
o
f
S
t
o
ra
g
e
a
n
d
Extra
c
ti
o
n
o
f
Co
mp
o
n
e
n
ts
.
IEE
E,
I
n
tern
a
ti
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
A
d
v
a
n
c
e
s
in
El
e
c
tro
n
ics
,
Co
m
p
u
ters
a
n
d
Co
m
m
u
n
ica
ti
o
n
s.
2
0
1
4
:
1
-
6
.
Evaluation Warning : The document was created with Spire.PDF for Python.