T
E
L
K
O
M
N
I
K
A
T
elec
o
m
m
un
ica
t
io
n,
Co
m
pu
t
ing
,
E
lect
ro
nics
a
nd
Co
ntr
o
l
Vo
l.
1
8
,
No
.
6
,
Dec
em
b
er
2
0
2
0
,
p
p
.
2
9
7
7
~
2
9
8
2
I
SS
N:
1
6
9
3
-
6
9
3
0
,
ac
cr
ed
ited
First Gr
ad
e
b
y
Kem
en
r
is
tek
d
i
k
ti,
Dec
r
ee
No
: 2
1
/E/KPT
/2
0
1
8
DOI
: 1
0
.
1
2
9
2
8
/TE
L
KOM
NI
K
A.
v
1
8
i
6
.
1
6
7
4
3
2977
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//jo
u
r
n
a
l.u
a
d
.
a
c.
id
/in
d
ex
.
p
h
p
/TELK
OM
N
I
K
A
H
a
rness
ing
dee
p l
ea
rning
alg
o
rithms to predi
ct
so
f
t
wa
re
refactoring
M
a
m
do
uh
Alenez
i
1
,
M
o
ha
m
m
ed
Ak
o
ur
2
,
O
s
a
m
a
Al
Q
a
s
em
3
1
Co
m
p
u
ter
S
c
ien
c
e
De
p
a
rtme
n
t,
P
rin
c
e
S
u
lt
a
n
U
n
iv
e
rsit
y
,
S
a
u
d
i
A
ra
b
ia
2,
3
In
f
o
rm
a
ti
o
n
S
y
ste
m
s De
p
a
rtme
n
t,
Ya
rm
o
u
k
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
Ma
y
1
9
,
2
0
2
0
R
ev
is
ed
J
u
n
2
7
,
2
0
2
0
Acc
ep
ted
J
u
l 6
,
2
0
2
0
Du
rin
g
so
ftwa
re
m
a
in
ten
a
n
c
e
,
s
o
f
twa
re
sy
ste
m
s
n
e
e
d
to
b
e
m
o
d
ifi
e
d
b
y
a
d
d
i
n
g
o
r
m
o
d
ify
in
g
so
u
rc
e
c
o
d
e
.
T
h
e
se
c
h
a
n
g
e
s
a
re
re
q
u
ired
t
o
fix
e
rro
r
s
o
r
a
d
o
p
t
n
e
w
re
q
u
irem
e
n
ts
ra
ise
d
b
y
sta
k
e
h
o
l
d
e
rs
o
r
m
a
rk
e
t
p
lac
e
.
Id
e
n
ti
fy
in
g
t
h
e
targ
e
ted
p
iec
e
o
f
c
o
d
e
f
o
r
re
fa
c
to
rin
g
p
u
rp
o
se
s
is
c
o
n
sid
e
re
d
a
re
a
l
c
h
a
ll
e
n
g
e
fo
r
so
ftwa
re
d
e
v
e
lo
p
e
rs.
Th
e
wh
o
le
p
ro
c
e
ss
o
f
re
fa
c
to
rin
g
m
a
in
l
y
re
li
e
s
o
n
so
ftwa
re
d
e
v
e
lo
p
e
rs’
sk
il
ls
a
n
d
in
t
u
it
i
o
n
.
In
th
is
p
a
p
e
r,
a
d
e
e
p
lea
rn
i
n
g
a
lg
o
rit
h
m
is
u
se
d
t
o
d
e
v
e
lo
p
a
re
fa
c
to
rin
g
p
re
d
icti
o
n
m
o
d
e
l
f
o
r
h
i
g
h
li
g
h
t
in
g
th
e
c
las
se
s th
a
t
re
q
u
ire refa
c
to
rin
g
.
M
o
re
sp
e
c
ifi
c
a
ll
y
,
th
e
g
a
ted
re
c
u
rre
n
t
u
n
it
a
lg
o
rit
h
m
is
u
se
d
with
p
r
o
p
o
se
d
p
r
e
-
p
ro
c
e
ss
in
g
ste
p
s
fo
r
re
fa
c
to
ri
n
g
p
re
d
ictio
n
a
t
th
e
c
las
s
lev
e
l.
Th
e
e
ffe
c
ti
v
e
n
e
ss
o
f
th
e
p
r
o
p
o
se
d
m
o
d
e
l
is
e
v
a
lu
a
ted
u
si
n
g
a
v
e
ry
c
o
m
m
o
n
d
a
tas
e
t
o
f
7
o
p
e
n
so
u
rc
e
jav
a
p
r
o
jec
ts.
Th
e
e
x
p
e
r
ime
n
ts
a
re
c
o
n
d
u
c
ted
b
e
fo
re
a
n
d
a
fter
b
a
lan
c
in
g
t
h
e
d
a
tas
e
t
to
in
v
e
sti
g
a
te
th
e
in
flu
e
n
c
e
o
f
d
a
ta
sa
m
p
li
n
g
o
n
t
h
e
p
e
rf
o
rm
a
n
c
e
o
f
th
e
p
re
d
icti
o
n
m
o
d
e
l.
Th
e
e
x
p
e
rime
n
tal
a
n
a
ly
sis re
v
e
a
ls a p
ro
m
isi
n
g
re
s
u
l
t
in
t
h
e
field
o
f
c
o
d
e
re
fa
c
to
ri
n
g
p
re
d
ictio
n
.
K
ey
w
o
r
d
s
:
Dee
p
lear
n
in
g
al
g
o
r
ith
m
s
Me
asu
r
em
en
t
So
f
twar
e
m
ain
ten
an
ce
So
f
twar
e
r
ef
ac
to
r
in
g
So
u
r
ce
co
d
e
an
aly
s
is
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
:
Ma
m
d
o
u
h
Alen
ez
i,
C
o
m
p
u
ter
Scien
ce
Dep
ar
tm
e
n
t,
Prin
ce
Su
ltan
Un
iv
er
s
ity
,
R
iy
ad
h
1
1
5
8
6
,
Sau
d
i A
r
ab
ia
.
E
m
ail:
m
alen
ez
i@
p
s
u
.
ed
u
.
s
a
1.
I
NT
RO
D
UCT
I
O
N
So
f
twar
e
ap
p
licatio
n
s
a
r
e
en
d
less
ly
m
ain
tain
ed
an
d
m
o
d
if
y
to
ad
d
n
ew
r
eq
u
ir
em
en
ts
,
f
ix
er
r
o
r
s
,
o
r
ad
ap
t
n
ew
m
o
d
u
les.
R
eq
u
ir
e
m
en
ts
co
n
tin
u
o
u
s
ly
c
h
an
g
e
a
s
th
e
m
ar
k
et
p
lace
is
s
u
s
ce
p
tib
le
to
s
tak
eh
o
ld
er
s
’
d
em
an
d
s
.
T
h
er
ef
o
r
e,
s
o
f
twar
e
ap
p
licatio
n
s
s
h
o
u
ld
b
e
ev
o
lv
e
d
co
n
tin
u
o
u
s
ly
to
m
a
k
e
s
u
r
e
t
h
e
s
tak
eh
o
ld
e
r
s
ar
e
s
atis
f
ied
.
Du
r
in
g
s
o
f
twar
e
m
ai
n
ten
an
ce
,
p
r
o
g
r
am
m
er
s
a
r
e
ask
ed
to
a
d
d
a
n
ew
f
ea
t
u
r
e,
r
em
o
v
e
an
d
/o
r
m
o
d
if
y
ex
is
tin
g
o
n
es.
T
h
e
p
r
o
ce
s
s
o
f
a
d
ap
tin
g
t
h
ese
f
ea
tu
r
es
n
ee
d
s
to
m
o
d
if
y
th
e
s
o
f
twar
e
s
y
s
tem
s
to
m
ee
t
th
e
r
eq
u
ir
ed
r
eq
u
ir
em
e
n
t,
th
is
p
r
o
ce
s
s
ca
lle
d
r
ef
ac
to
r
in
g
[
1
]
.
C
o
d
e
r
ef
ac
to
r
in
g
p
lay
s
an
im
p
o
r
tan
t
r
o
le
in
en
h
an
cin
g
s
o
f
twar
e
q
u
ality
b
y
ev
o
lv
i
n
g
th
e
i
n
ter
n
a
l stru
ctu
r
e
with
o
u
t a
f
f
ec
tin
g
th
e
in
ten
d
ed
b
eh
av
i
o
r
[
2
]
.
E
x
ten
s
iv
e
r
esear
ch
h
as
b
ee
n
c
o
n
d
u
cte
d
f
o
r
ad
d
r
ess
in
g
th
e
r
e
latio
n
s
h
ip
b
etwe
en
r
ef
ac
to
r
in
g
s
o
f
twar
e
s
y
s
tem
s
an
d
th
e
q
u
ality
m
ea
s
u
r
em
en
ts
[
3
-
10
]
.
All
ex
p
er
im
en
tal
r
esu
lts
s
h
o
w
h
o
w
r
ef
ac
t
o
r
in
g
h
as
a
d
i
r
ec
t
in
f
lu
en
ce
o
n
im
p
r
o
v
in
g
s
o
f
tw
ar
e
q
u
ality
.
T
h
er
e
f
o
r
e,
p
r
e
d
ictin
g
r
ef
ac
t
o
r
in
g
p
r
o
m
p
t
ly
s
h
o
u
ld
b
e
in
v
esti
g
atin
g
,
an
d
b
u
ild
in
g
a
n
ac
cu
r
ate
m
o
d
el
b
ec
o
m
es
m
an
d
ato
r
y
.
So
f
tw
ar
e
d
ev
elo
p
er
s
s
till
f
ac
e
a
r
ea
l
ch
allen
g
e
to
p
ick
th
e
r
ig
h
t
tim
e
an
d
s
o
f
twar
e
co
d
e
f
o
r
r
ef
ac
to
r
i
n
g
p
u
r
p
o
s
e
s
as
th
e
o
p
e
r
atio
n
n
ee
d
s
tim
e
an
d
b
u
d
g
et
[
11
]
.
T
h
er
ef
o
r
e,
d
ev
elo
p
er
s
s
h
o
u
ld
b
e
s
u
r
e
ab
o
u
t w
h
ich
p
iece
o
f
c
o
d
e
s
h
o
u
ld
b
e
ev
o
l
v
ed
b
ef
o
r
e
s
tar
tin
g
th
e
p
r
o
ce
s
s
o
f
r
ef
ac
to
r
i
n
g
to
ad
o
p
t
th
e
n
ew
r
eq
u
ir
em
en
ts
.
Alth
o
u
g
h
d
e
v
el
o
p
er
ex
p
er
ien
ce
s
h
ap
es
th
e
m
o
s
t
s
u
cc
es
s
f
u
l
f
ac
to
r
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
1
8
,
No
.
6
,
Dec
em
b
e
r
2
0
2
0
:
29
7
7
-
298
2
2978
in
th
is
p
r
o
ce
s
s
,
s
till
p
r
ed
ictio
n
alg
o
r
ith
m
s
m
ig
h
t b
e
a
h
elp
f
u
l
to
o
l
in
th
is
m
atter
.
T
h
ese
alg
o
r
ith
m
s
ca
n
p
r
o
v
id
e
th
e
d
ev
elo
p
er
s
with
s
o
m
e
in
s
ig
h
ts
ab
o
u
t w
h
ic
h
p
ar
t
o
f
co
d
e
s
h
o
u
ld
b
e
r
ef
ac
to
r
e
d
an
d
wh
en
.
D
i
f
f
e
r
e
n
t
m
e
t
h
o
d
o
l
o
g
i
e
s
a
r
e
d
e
s
i
g
n
e
d
a
n
d
b
u
i
l
t
t
o
h
e
l
p
d
e
v
e
l
o
p
e
r
s
i
n
t
h
e
r
e
f
a
c
t
o
r
i
n
g
p
r
o
c
e
s
s
s
u
c
h
a
s
c
o
d
e
s
m
e
l
l
s
d
e
t
e
c
t
i
o
n
s
t
r
a
t
e
g
i
e
s
[
12
]
,
l
o
g
i
c
m
e
t
a
-
p
r
o
g
r
a
m
m
i
n
g
[
1
3
]
,
i
n
v
a
r
i
a
n
t
m
i
n
i
n
g
[
1
4
]
a
n
d
s
e
a
r
c
h
-
b
a
s
e
d
[
1
5
,
1
6
]
.
M
o
r
e
o
v
e
r
,
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
i
s
h
a
r
n
e
s
s
e
d
i
n
t
h
e
a
r
e
a
o
f
p
r
e
d
i
c
t
i
o
n
a
n
d
s
h
o
w
s
n
o
t
i
c
e
a
b
l
e
p
e
r
f
o
r
m
a
n
c
e
i
n
t
e
r
m
s
o
f
p
r
e
d
i
c
t
i
o
n
i
n
v
a
r
i
o
u
s
f
i
e
l
d
s
a
s
c
o
m
p
u
t
e
r
v
i
s
i
o
n
,
d
e
f
e
c
t
p
r
e
d
i
c
t
i
o
n
,
n
a
t
u
r
a
l
l
a
n
g
u
a
g
e
p
r
o
c
e
s
s
i
n
g
,
c
o
d
e
c
o
m
p
r
e
h
e
n
s
i
o
n
,
b
i
o
i
n
f
o
r
m
a
t
i
c
s
,
s
p
e
e
c
h
r
e
c
o
g
n
i
t
i
o
n
,
a
n
d
fi
n
a
n
c
e
[
1
7
-
24
]
.
S
e
v
e
r
a
l
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
m
s
a
r
e
u
t
i
l
i
z
e
d
i
n
c
o
d
e
r
e
f
a
c
t
o
r
i
n
g
p
r
e
d
i
c
t
i
o
n
a
t
c
l
a
s
s
a
n
d
m
e
t
h
o
d
l
e
v
e
l
a
s
w
e
l
l
[
2
5
,
2
6
].
T
h
e
m
ain
co
n
tr
ib
u
tio
n
o
f
th
is
wo
r
k
is
in
v
esti
g
atin
g
th
e
ef
f
ec
tiv
en
ess
o
f
d
ee
p
lear
n
in
g
alg
o
r
ith
m
s
in
b
u
ild
in
g
r
e
f
ac
to
r
in
g
p
r
e
d
ictio
n
m
o
d
els
at
th
e
class
-
lev
el.
T
h
e
im
p
lem
en
ted
d
ee
p
lear
n
in
g
alg
o
r
ith
m
is
g
ated
r
ec
u
r
r
en
t
u
n
it
(
GR
U)
.
T
o
th
e
b
est
o
f
o
u
r
k
n
o
wled
g
e,
th
is
a
lg
o
r
ith
m
is
u
s
ed
f
o
r
th
e
f
ir
s
t
tim
e
f
o
r
r
ef
ac
to
r
i
n
g
p
r
ed
ictio
n
at
th
e
class
lev
el.
I
n
th
is
wo
r
k
7
o
p
e
n
-
s
o
u
r
ce
J
av
a
-
b
ased
p
r
o
jects
ar
e
u
s
ed
to
ass
ess
th
e
ef
f
ec
tiv
en
ess
o
f
th
e
s
tu
d
ied
alg
o
r
ith
m
.
2.
L
I
T
E
R
AT
U
RE
R
E
VI
E
W
T
h
er
e
ar
e
s
ev
e
r
al
atte
m
p
ts
in
t
h
e
liter
atu
r
e
to
u
s
e
m
ac
h
in
e
le
ar
n
in
g
t
o
p
r
e
d
ict
an
d
s
u
g
g
est
r
ef
ac
to
r
in
g
.
Am
al
et
a
l.
[2
7
]
u
s
ed
s
ea
r
ch
-
b
ased
s
o
f
twar
e
en
g
in
e
er
in
g
f
o
r
s
o
f
twar
e
r
ef
ac
to
r
i
n
g
.
T
h
ey
u
s
ed
an
a
r
tific
ial
n
eu
r
al
n
etwo
r
k
(
ANN)
a
n
d
g
en
etic
al
g
o
r
ith
m
s
(
GA)
t
o
ch
o
o
s
e
b
etw
ee
n
r
ef
ac
to
r
in
g
s
o
lu
ti
o
n
s
.
T
h
e
y
u
s
ed
th
e
o
p
i
n
io
n
o
f
1
6
d
if
f
e
r
en
t
s
o
f
twar
e
en
g
in
ee
r
s
to
m
an
u
ally
ev
alu
a
te
t
h
e
r
ef
ac
to
r
in
g
s
o
lu
tio
n
s
f
o
r
tr
ain
i
n
g
.
T
h
ey
d
e
v
elo
p
e
d
a
p
r
ed
ictiv
e
m
o
d
el
to
ev
al
u
ate
th
e
r
ef
ac
to
r
in
g
s
o
lu
tio
n
s
f
o
r
th
e
r
em
ain
in
g
iter
ati
o
n
s
.
T
h
e
ap
p
r
o
ac
h
o
u
tp
er
f
o
r
m
ed
t
h
e
m
an
u
al
r
e
f
a
cto
r
in
g
ap
p
r
o
ac
h
.
Ku
m
ar
et
a
l
.
[2
8
]
u
s
ed
2
5
s
o
u
r
ce
c
o
d
e
m
e
tr
ics
at
th
e
m
eth
o
d
lev
el
to
p
r
ed
ict
t
h
e
n
ee
d
f
o
r
r
e
f
ac
to
r
in
g
.
T
h
e
y
u
s
ed
a
p
u
b
licl
y
av
ailab
le
d
ataset
o
f
f
iv
e
o
p
e
n
-
s
o
u
r
ce
d
s
o
f
twar
e
s
y
s
tem
s
to
in
v
esti
g
ate
th
e
p
er
f
o
r
m
an
ce
o
f
ten
d
if
f
e
r
en
t
m
ac
h
in
e
lear
n
in
g
class
if
ier
s
.
T
h
ey
u
s
ed
th
r
ee
d
if
f
er
en
t
d
ata
s
am
p
lin
g
m
eth
o
d
s
to
tack
le
th
e
u
n
b
ala
n
ce
d
d
ata
is
s
u
es.
Al
Dallal
[
2
9
]
d
is
cu
s
s
ed
a
m
e
asu
r
e
an
d
a
m
o
d
el
t
o
p
r
e
d
ict
wh
eth
er
th
e
m
et
h
o
d
(
s
)
i
n
a
cl
ass
n
ee
d
in
g
m
o
v
e
m
eth
o
d
r
e
f
ac
to
r
in
g
an
d
ac
h
iev
ed
a
p
r
ed
ictio
n
ac
c
u
r
ac
y
o
f
m
o
r
e
th
an
9
0
%.
T
h
e
a
u
th
o
r
b
u
ilt
th
e
p
r
e
d
ictiv
e
m
o
d
el
o
v
er
s
ev
e
n
o
b
ject
-
o
r
ien
ted
s
y
s
tem
s
.
An
ich
e
et
a
l.
[
30
]
in
v
esti
g
ated
th
e
ef
f
ec
tiv
e
n
ess
o
f
s
ix
d
if
f
er
en
t
m
ac
h
in
e
lear
n
i
n
g
alg
o
r
ith
m
s
(
l
o
g
is
tic
r
eg
r
ess
io
n
,
n
aiv
e
B
ay
e
s
,
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
,
d
ec
is
io
n
tr
ee
s
,
r
an
d
o
m
f
o
r
est,
an
d
n
eu
r
al
n
etwo
r
k
)
in
p
r
ed
ictin
g
s
o
f
twar
e
r
ef
ac
t
o
r
in
g
.
T
h
ey
u
s
ed
a
d
ataset
co
n
s
is
ts
o
f
o
v
er
two
m
illi
o
n
r
ef
ac
to
r
in
g
s
f
r
o
m
1
1
,
1
4
9
r
ea
l
-
wo
r
ld
p
r
o
jects.
Pan
tiu
ch
in
a
et
a
l.
[
31
]
p
r
o
p
o
s
ed
an
ap
p
r
o
ac
h
to
p
r
ev
e
n
t
in
s
tead
o
f
f
i
x
co
d
e
q
u
ality
is
s
u
es
th
at
p
r
ed
ict
c
o
d
e
s
m
ells
.
T
h
e
a
p
p
r
o
ac
h
u
s
es
s
o
u
r
ce
c
o
d
e
q
u
ality
to
p
r
ed
ict
wh
eth
er
a
m
o
d
u
le
is
lik
ely
to
b
e
af
f
ec
t
ed
b
y
c
o
d
e
s
m
ells
in
th
e
f
u
tu
r
e.
T
h
e
to
p
ic
at
h
an
d
is
s
till
in
its
in
f
an
cy
.
Sev
er
al
ad
v
an
ce
s
ca
n
b
e
m
ad
e
to
p
r
e
d
ict
co
d
e
s
m
ells
an
d
r
e
f
ac
to
r
i
n
g
o
p
p
o
r
t
u
n
ities
th
at
wo
u
ld
e
v
en
tu
ally
im
p
r
o
v
e
s
o
f
twar
e
q
u
ality
an
d
m
ain
ten
a
n
ce
.
3.
RE
S
E
ARCH
M
E
T
H
O
DO
L
O
G
Y
T
h
is
s
ec
tio
n
p
r
esen
ts
th
e
p
r
o
p
o
s
ed
a
p
p
r
o
ac
h
f
o
r
So
f
twar
e
r
ef
ac
to
r
i
n
g
p
r
ed
ictio
n
.
T
h
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
is
d
iv
id
ed
in
to
two
m
ain
s
tag
es.
T
h
e
f
ir
s
t
s
tag
e
p
er
f
o
r
m
s
a
s
et
o
f
n
ec
ess
ar
y
p
r
e
-
p
r
o
ce
s
s
in
g
p
r
o
ce
d
u
r
es
o
n
d
atasets
.
I
n
th
e
s
ec
o
n
d
s
tag
e,
th
e
d
ee
p
lear
n
in
g
alg
o
r
ith
m
is
ap
p
lied
to
th
e
d
atasets
to
p
r
ed
ict
th
e
n
ee
d
f
o
r
r
ef
ac
to
r
in
g
o
r
n
o
t
b
y
u
s
in
g
g
a
ted
r
ec
u
r
r
en
t
u
n
it
(
GR
U)
alg
o
r
ith
m
.
T
h
e
s
tr
u
ctu
r
e
o
f
t
h
e
p
r
o
p
o
s
ed
a
p
p
r
o
a
ch
is
o
u
tlin
ed
in
Fig
u
r
e
1
.
Mo
r
e
d
etails o
f
th
e
ap
p
r
o
ac
h
a
r
e
g
iv
e
n
in
th
e
n
ex
t su
b
s
ec
tio
n
s
.
Fig
u
r
e
1
.
Pro
p
o
s
ed
m
et
h
o
d
o
lo
g
y
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
Ha
r
n
ess
in
g
d
ee
p
lea
r
n
in
g
a
lg
o
r
ith
ms to
p
r
ed
ict
s
o
ftw
a
r
e
r
e
f
a
cto
r
in
g
(
Ma
md
o
u
h
A
len
ezi
)
2979
3
.
1
.
Da
t
a
s
et
s
I
n
th
is
p
ap
er
,
we
u
s
ed
a
p
u
b
lic
em
p
ir
ical
d
ataset
co
n
tain
in
g
r
e
f
ac
to
r
in
g
d
ata
f
o
r
7
o
p
e
n
-
s
o
u
r
c
e
s
y
s
tem
s
(
an
tlr4
,
ju
n
it,
m
ap
d
b
,
m
cM
MO
,
m
ct,
o
r
y
x
,
titan
)
[
32
]
.
T
h
e
e
x
p
er
im
en
tal
d
ataset
u
s
ed
in
o
u
r
s
tu
d
y
is
f
r
ee
ly
an
d
p
u
b
licly
av
ailab
le
at
th
e
PR
O
MI
SE
R
ep
o
s
ito
r
y
.
T
h
is
m
ak
es
o
u
r
wo
r
k
ea
s
ily
r
ep
r
o
d
u
cib
le.
T
h
e
d
ataset
u
s
ed
in
o
u
r
ex
p
er
i
m
en
ts
is
m
a
n
u
ally
v
alid
ated
b
y
cr
ea
tin
g
t
h
e
s
o
u
r
c
e
co
d
e
m
etr
ics
a
n
d
th
e
r
ef
ac
to
r
in
g
d
ataset
f
o
r
two
s
u
b
s
eq
u
en
t
r
elea
s
es
o
f
7
well
-
k
n
o
wn
o
p
en
-
s
o
u
r
ce
s
o
f
twar
e
J
av
a
ap
p
licatio
n
s
.
T
h
e
to
o
ls
u
s
ed
to
cr
ea
te
th
is
d
ataset
ar
e
th
e
R
ef
Fin
d
er
to
o
l
f
o
r
id
e
n
tify
in
g
r
e
f
ac
to
r
in
g
in
t
h
e
s
o
u
r
ce
co
d
e
b
etwe
en
two
s
u
b
s
eq
u
en
t
r
elea
s
es
an
d
th
e
So
u
r
ce
Me
ter
to
o
l to
c
o
m
p
u
te
s
o
u
r
ce
co
d
e
m
etr
ics.
T
h
ese
d
atasets
u
s
ed
f
o
r
em
p
ir
i
ca
l in
v
esti
g
atio
n
s
o
n
s
o
u
r
ce
co
d
e
r
ef
ac
t
o
r
in
g
.
T
h
e
f
ea
tu
r
es
in
clu
d
e
s
o
u
r
ce
co
d
e
m
etr
ics,
th
e
r
e
f
ac
to
r
in
g
ty
p
e
s
,
an
d
th
e
r
elativ
e
m
ain
tain
ab
ilit
y
in
d
e
x
(
R
MI
)
.
T
h
er
e
ar
e
2
3
r
ef
ac
to
r
in
g
ty
p
es
at
th
e
class
lev
el.
T
h
e
d
ataset'
s
ch
ar
ac
ter
is
tics
ar
e
p
r
esen
ted
in
T
a
b
le
1.
T
ab
le
1
.
Data
s
ets ch
ar
ac
ter
is
tics
D
a
t
a
s
e
t
N
o
.
o
f
A
t
t
r
i
b
u
t
e
I
n
st
a
n
c
e
s
N
o
.
r
e
f
a
c
t
o
r
i
n
g
P
e
r
c
e
n
t
a
g
e
a
n
t
l
r
4
1
3
4
4
3
6
23
5
.
2
%
j
u
n
i
t
1
3
4
6
5
7
9
1
.
3
%
map
d
b
1
3
4
4
3
9
4
0
.
9
%
mcM
M
O
1
3
4
3
0
1
3
0
.
9
9
%
mct
1
3
4
2
1
6
2
15
0
.
6
9
%
o
r
y
x
1
3
4
5
3
6
15
2
.
7
%
t
i
t
a
n
1
3
4
1
4
8
6
13
0
.
8
7
%
I
n
th
e
f
ir
s
t stag
e,
d
ata
p
r
ep
r
o
c
ess
in
g
in
th
is
s
tu
d
y
ca
n
b
e
s
u
m
m
ar
ized
as f
o
llo
ws:
−
D
e
l
e
t
i
o
n
o
f
u
n
n
e
c
e
s
s
a
r
y
f
e
a
t
u
r
e
s
:
r
e
m
o
v
e
t
h
e
f
o
l
l
o
w
i
n
g
f
e
a
t
u
r
e
(
n
a
m
e
,
p
a
t
h
,
L
o
n
g
N
a
m
e
,
P
a
r
e
n
t
,
C
o
m
p
o
n
e
n
t
)
−
Dele
te
r
ef
ac
to
r
in
g
ty
p
e
f
ea
tu
r
e
: 2
3
r
ef
ac
t
o
r
in
g
t
y
p
es a
t a
class
lev
el
h
av
e
b
ee
n
d
elete
d
−
R
ep
lace
m
en
t
th
e
s
u
m
m
atio
n
o
f
r
ef
ac
to
r
in
g
f
ea
tu
r
e:
W
e
h
av
e
m
o
d
if
ie
d
th
e
v
alu
e
o
f
(
s
u
m
m
atio
n
o
f
r
ef
ac
to
r
in
g
ty
p
es)
in
ea
ch
in
s
ta
n
ce
to
b
in
a
r
y
v
alu
es
a
n
d
en
co
d
ed
th
e
s
u
m
m
atio
n
o
f
r
ef
ac
to
r
in
g
ty
p
es
"m
o
r
e
th
an
o
n
e
"
as (
1
)
a
n
d
"z
er
o
v
al
u
e"
as (
0
)
.
t
h
en
u
s
e
th
is
f
ea
tu
r
e
as a
class
lab
el.
−
Sam
p
lin
g
d
atasets
:
T
o
im
p
r
o
v
e
th
e
p
r
ed
ictio
n
o
f
th
e
m
i
n
o
r
ity
class
s
h
o
u
ld
b
e
c
o
r
r
ec
t
th
e
im
b
alan
c
e
p
r
o
b
lem
.
Fo
r
th
at,
we
u
s
ed
th
e
s
y
n
th
etic
m
in
o
r
ity
o
v
er
-
s
am
p
lin
g
tech
n
iq
u
e
(
SMOT
E
)
.
SM
OT
E
d
ea
ls
with
th
e
p
r
o
b
lem
o
f
im
b
alan
ce
d
d
i
s
tr
ib
u
tio
n
,
p
r
o
d
u
cin
g
n
ew
in
s
tan
ce
s
b
ased
o
n
K
-
n
ea
r
est
n
ei
g
h
b
o
r
(
KNN)
.
C
o
m
p
u
tin
g
th
e
KNN
v
al
u
e
f
o
r
m
s
b
ased
o
n
th
e
s
im
ilar
ity
(
we
co
n
s
id
er
in
t
h
is
p
ap
er
K
=
5
)
.
3
.
2
.
G
a
t
ed
re
curr
ent
u
nit
a
lg
o
rit
hm
G
R
U
m
o
d
e
l
i
s
a
p
o
w
e
r
f
u
l
d
e
e
p
l
e
a
r
n
i
n
g
m
o
d
e
l
p
r
o
p
o
s
e
d
b
y
[
33
]
a
l
s
o
i
n
t
r
o
d
u
c
e
d
i
n
[
34
]
,
G
R
U
i
s
o
n
e
k
i
n
d
o
f
t
h
e
g
a
t
e
d
R
N
N
s
w
h
i
c
h
a
r
e
u
s
e
d
t
o
s
o
l
v
e
t
h
e
c
o
m
m
o
n
p
r
o
b
l
e
m
s
o
f
g
r
a
d
i
e
n
t
v
a
n
i
s
h
i
n
g
o
f
t
r
a
d
i
t
i
o
n
a
l
R
N
N
[
35
]
.
G
R
U
c
o
n
t
a
i
n
s
t
w
o
g
a
t
e
s
t
h
a
t
u
s
e
i
t
t
o
c
o
n
t
r
o
l
t
h
e
i
n
f
o
r
m
a
t
i
o
n
f
l
o
w
f
r
o
m
t
h
e
t
h
r
o
u
g
h
t
h
e
n
e
t
w
o
r
k
.
F
i
r
s
t
,
t
h
e
g
a
t
e
t
o
c
o
n
t
r
o
l
t
h
e
i
n
f
o
r
m
a
t
i
o
n
f
l
o
w
s
i
n
t
o
m
e
m
o
r
y
k
n
o
w
n
a
s
a
n
u
p
d
a
t
e
g
a
t
e
.
S
e
c
o
n
d
,
i
s
t
o
c
o
n
t
r
o
l
t
h
e
i
n
f
o
r
m
a
t
i
o
n
t
h
a
t
f
l
o
w
s
o
u
t
o
f
m
e
m
o
r
y
k
n
o
w
n
a
s
r
e
s
e
t
g
a
t
e
u
n
l
i
k
e
l
o
n
g
s
h
o
r
t
-
t
e
r
m
m
e
m
o
r
y
(
L
S
T
M
)
,
G
R
U
h
a
s
n
'
t
h
a
d
s
e
p
a
r
a
t
e
m
e
m
o
r
y
c
e
l
l
,
i
n
s
t
e
a
d
o
f
t
h
a
t
g
a
t
i
n
g
u
n
i
t
t
h
a
t
c
o
n
t
r
o
l
s
t
h
e
f
l
o
w
o
f
i
n
f
o
r
m
a
t
i
o
n
i
n
s
i
d
e
t
h
e
u
n
i
t
[
36
].
3.
4
.
Str
uct
ura
l pa
ra
m
et
er
s
e
lect
io
n f
o
r
g
a
t
ed
re
curr
ent
u
nit
T
o
g
et
an
ef
f
ec
tiv
e
GR
U
m
o
d
el,
we
n
ee
d
to
s
et
k
e
y
s
tr
u
ctu
r
al
p
ar
am
eter
s
,
wh
ich
a
r
e:
1
)
t
h
e
n
u
m
b
e
r
o
f
h
id
d
en
lay
e
r
s
,
2
)
t
h
e
n
u
m
b
er
o
f
ep
o
ch
,
an
d
3
)
b
atch
s
ize.
T
o
s
im
p
lify
th
e
GR
U
m
o
d
el,
we
ch
o
o
s
e
th
r
ee
h
id
d
en
lay
er
s
,
wh
ich
is
a
g
en
e
r
al
co
n
f
ig
u
r
atio
n
f
o
llo
wed
b
y
[
37
]
,
an
d
we
u
tili
ze
f
o
r
th
e
n
u
m
b
er
o
f
th
e
ep
o
ch
f
r
o
m
1
0
u
n
til 2
5
0
0
.
Fo
r
t
h
e
b
a
tch
s
ize
u
s
e
it b
etwe
en
2
to
1
5
.
3.
5
.
Co
m
pa
riso
ns
T
h
e
last
s
tep
in
th
e
p
r
o
p
o
s
ed
m
eth
o
d
o
l
o
g
y
is
th
e
co
m
p
a
r
is
o
n
s
tu
d
y
,
wh
er
e
th
e
b
ala
n
ce
d
d
ata
s
et
an
d
u
n
b
alan
ce
d
d
ataset
r
esu
lt a
r
e
m
ea
s
u
r
e
d
an
d
co
m
p
ar
ed
.
4.
E
XP
E
R
I
M
E
N
T
A
L
RE
SUL
T
S
I
n
th
is
s
ec
tio
n
,
we
em
p
ir
ically
ev
alu
ate
th
e
p
e
r
f
o
r
m
an
ce
o
f
th
e
GR
U
m
o
d
el
b
y
u
s
in
g
th
e
7
o
p
en
-
s
o
u
r
ce
s
o
f
twar
e
s
y
s
tem
s
.
W
e
o
p
tim
ized
GR
U
with
th
r
ee
h
id
d
en
lay
er
s
,
with
a
s
ig
m
o
id
ac
tiv
atio
n
f
u
n
ctio
n
f
o
r
th
e
o
u
tp
u
t
lay
er
.
T
h
is
s
tu
d
y
is
ca
r
r
ied
o
u
t
o
n
7
d
atas
ets
th
at
b
elo
n
g
to
d
if
f
er
en
t
d
o
m
ain
s
.
T
h
e
p
er
f
o
r
m
an
ce
with
th
e
u
n
b
alan
ce
d
d
ataset
(
ex
p
er
im
e
n
tal
s
ce
n
ar
io
1
)
an
d
with
b
alan
ce
d
d
atasets
(
ex
p
e
r
im
en
t
al
s
ce
n
ar
io
s
2
)
ar
e
co
llected
an
d
co
m
p
a
r
ed
in
th
e
n
ex
t
s
u
b
s
ec
tio
n
s
.
T
h
e
p
e
r
f
o
r
m
an
ce
o
f
th
e
GR
U
m
o
d
el
is
ev
alu
a
ted
b
ased
o
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
1
8
,
No
.
6
,
Dec
em
b
e
r
2
0
2
0
:
29
7
7
-
298
2
2980
ac
cu
r
ac
y
,
F
-
m
ea
s
u
r
e,
r
ec
all,
a
n
d
p
r
ec
is
io
n
.
First,
we
co
n
d
u
c
ted
ex
p
e
r
im
en
ts
wh
er
e
th
e
d
at
aset
is
u
n
b
alan
ce
d
;
we
h
av
e
e
v
alu
ated
t
h
e
p
er
f
o
r
m
an
ce
o
f
th
e
s
tu
d
ied
alg
o
r
ith
m
o
n
7
d
atasets
.
T
h
en
we
r
ep
ea
t
th
e
s
am
e
ex
p
er
im
e
n
t
af
ter
ap
p
ly
in
g
SMOT
E
to
p
r
o
d
u
ce
a
b
alan
ce
d
d
ataset.
T
a
b
le
t
wo
s
u
m
m
ar
izes
th
e
o
b
tain
ed
Per
f
o
r
m
a
n
ce
R
esu
lts
b
ef
o
r
e
a
n
d
af
te
r
ap
p
l
y
in
g
SM
OT
E
.
I
n
th
is
s
ec
tio
n
,
we
r
ep
o
r
t
t
h
e
co
m
p
ar
is
o
n
r
esu
lts
o
f
p
r
e
d
ic
tio
n
p
e
r
f
o
r
m
an
ce
o
n
th
e
b
ala
n
ce
d
a
n
d
u
n
b
alan
ce
d
d
atasets
.
T
h
e
r
esu
l
ts
o
f
s
ce
n
ar
io
2
ex
p
er
im
en
t th
at
u
s
es th
e
b
alan
ce
d
d
ataset
s
h
o
w
b
etter
r
esu
lts
in
p
r
ed
ictio
n
.
Ou
r
g
o
al
h
e
r
e
is
to
in
v
esti
g
ate
wh
eth
er
th
e
b
alan
c
ed
d
atasets
ca
n
en
h
a
n
ce
th
e
r
ef
ac
to
r
in
g
p
r
ed
ictio
n
p
er
f
o
r
m
an
ce
u
s
in
g
th
e
s
tu
d
ied
d
ee
p
lear
n
in
g
alg
o
r
ith
m
o
n
th
ese
d
atasets
.
T
ab
l
e
2
s
h
o
ws
th
e
p
r
ed
ictio
n
r
esu
lts
with
u
n
b
ala
n
ce
d
d
ata.
T
a
b
le
3
r
ev
ea
ls
h
o
w
b
alan
cin
g
th
e
d
at
a
ca
n
im
p
r
o
v
e
th
e
p
r
ed
ictio
n
p
er
f
o
r
m
a
n
ce
r
esu
lts
.
Mo
s
t
m
ea
s
u
r
em
en
t
r
esu
lts
ar
e
in
cr
ea
s
ed
n
o
ticea
b
ly
,
esp
e
cially
f
o
r
r
ec
all,
p
r
ec
is
io
n
,
a
n
d
F
-
m
ea
s
u
r
es.
I
n
co
m
p
ar
is
o
n
b
etwe
en
u
n
b
alan
ce
d
an
d
b
alan
ce
d
d
ata,
th
e
F
-
m
ea
s
u
r
e
im
p
r
o
v
es
b
y
at
least
d
o
u
b
le
wh
ich
is
v
er
y
s
ig
n
if
ican
t.
T
ab
le
2
.
T
h
e
r
esu
lts
with
u
n
b
a
lan
ce
d
d
ata
D
a
t
a
s
e
t
A
c
c
u
r
a
c
y
R
e
c
a
l
l
P
r
e
c
i
s
i
o
n
F
-
mea
su
r
e
A
n
t
l
r
4
8
4
.
1
4
0
.
0
1
9
.
0
4
2
5
.
8
Ju
n
i
t
9
8
.
6
1
3
3
.
3
3
5
0
.
0
4
0
.
0
M
a
p
D
B
9
7
.
8
5
0
.
0
3
3
.
3
4
0
.
0
M
c
M
M
O
9
6
.
0
5
0
.
0
2
5
.
0
3
3
.
3
M
C
T
9
8
.
4
2
0
.
0
1
2
.
5
1
5
.
3
Ti
t
a
n
9
9
.
3
3
3
.
3
5
0
.
0
4
0
.
0
O
r
y
x
9
7
.
2
2
0
.
0
5
0
.
0
2
8
.
5
T
ab
le
3
.
T
h
e
r
esu
lts
with
b
alan
ce
d
d
ata
D
a
t
a
s
e
t
A
c
c
u
r
a
c
y
R
e
c
a
l
l
P
r
e
c
i
s
i
o
n
F
-
mea
su
r
e
A
n
t
l
r
4
9
1
.
9
1
0
0
8
6
.
1
9
2
.
5
Ju
n
i
t
9
8
.
6
1
0
0
9
7
.
2
9
8
.
6
M
a
p
D
B
9
9
.
3
1
0
0
9
8
.
6
9
9
.
3
M
c
M
M
O
99
1
0
0
98
99
M
C
T
9
9
.
8
1
0
0
9
7
.
8
9
8
.
8
Ti
t
a
n
9
9
.
3
1
0
0
9
8
.
7
9
9
.
3
O
r
y
x
9
9
.
3
1
0
0
9
8
.
7
9
9
.
3
5.
CO
NCLU
SI
O
N
AND
F
U
T
U
RE
WO
RK
T
h
is
wo
r
k
in
v
esti
g
ates
th
e
ef
f
ec
tiv
en
ess
o
f
u
s
in
g
d
ee
p
lear
n
in
g
alg
o
r
ith
m
s
in
r
ef
ac
to
r
in
g
p
r
ed
ictio
n
.
G
ated
r
ec
u
r
r
e
n
t u
n
it
(
GR
U)
alg
o
r
ith
m
is
u
s
ed
in
th
is
s
tu
d
y
a
n
d
th
e
p
er
f
o
r
m
an
ce
is
ev
alu
ate
d
o
n
7
o
p
en
-
s
o
u
r
ce
s
o
f
twar
e
p
r
o
d
u
cts
d
ataset.
Mo
r
eo
v
er
,
b
ala
n
cin
g
th
e
d
ata
s
et
as
an
en
h
an
ce
m
en
t
p
r
e
p
r
o
ce
s
s
in
g
s
tag
e
is
ad
d
r
ess
ed
in
th
is
s
tu
d
y
as
well.
T
h
e
s
y
n
th
etic
m
in
o
r
ity
o
v
e
r
-
s
am
p
l
in
g
tech
n
iq
u
e
(
SMOT
E
)
is
u
s
ed
f
o
r
b
alan
cin
g
th
e
d
ataset.
T
o
th
e
b
est
o
f
o
u
r
k
n
o
wled
g
e,
g
ated
r
ec
u
r
r
en
t
u
n
it
(
GR
U)
alg
o
r
ith
m
is
u
s
ed
f
o
r
th
e
f
ir
s
t
tim
e
f
o
r
r
ef
ac
to
r
in
g
p
r
ed
ictio
n
at
th
e
class
lev
el.
T
h
e
alg
o
r
ith
m
s
h
o
ws
p
r
o
m
is
in
g
p
e
r
f
o
r
m
an
ce
r
esu
lts
.
T
h
e
ex
p
er
im
en
ta
l
r
esu
lts
s
h
o
w
h
o
w
a
b
alan
ce
d
d
ataset
en
h
an
ce
s
th
e
p
r
e
d
ictio
n
p
er
f
o
r
m
an
ce
n
o
ticea
b
ly
,
w
h
er
e
all
u
s
ed
m
ea
s
u
r
es
in
th
is
s
tu
d
y
a
r
e
in
cr
ea
s
ed
af
te
r
u
s
in
g
a
b
alan
ce
d
d
ata
s
et
in
t
h
e
ex
p
er
im
e
n
ts
.
As
f
u
tu
r
e
wo
r
k
,
au
th
o
r
s
will
tr
y
to
p
r
ed
ict
th
e
r
ef
ac
to
r
i
n
g
ty
p
e
at
class
o
r
m
eth
o
d
lev
el
u
s
in
g
a
d
ee
p
lear
n
i
n
g
alg
o
r
ith
m
.
RE
F
E
R
E
NC
E
S
[1
]
A.
S
.
Ny
a
m
a
we
,
H.
Li
u
,
N.
Niu
,
Q.
Um
e
r
a
n
d
Z.
Niu
,
"
Au
t
o
m
a
ted
Re
c
o
m
m
e
n
d
a
ti
o
n
o
f
S
o
f
twa
re
Re
fa
c
to
rin
g
s
Ba
se
d
o
n
F
e
a
tu
re
Re
q
u
e
sts,"
2
0
1
9
IEE
E
2
7
t
h
I
n
ter
n
a
ti
o
n
a
l
Req
u
ire
me
n
ts
En
g
i
n
e
e
rin
g
Co
n
fer
e
n
c
e
(RE
),
Je
ju
Isla
n
d
,
K
o
re
a
(S
o
u
t
h
),
p
p
.
1
8
7
-
1
9
8
,
2
0
1
9
.
[2
]
A.
Qu
se
f,
M
.
O.
E
li
sh
a
n
d
D.
Bin
k
ley
,
"
An
Ex
p
lo
ra
t
o
ry
S
t
u
d
y
o
f
th
e
Re
latio
n
sh
i
p
Be
twe
e
n
S
o
ftwa
re
Tes
t
S
m
e
ll
s
a
n
d
F
a
u
lt
-
P
r
o
n
e
n
e
ss
,
"
IEE
E
Acc
e
ss
,
v
o
l.
7
,
p
p
.
1
3
9
5
2
6
-
1
3
9
5
3
6
,
2
0
1
9
.
[3
]
E.
A.
AlOm
a
r,
M
.
W.
M
k
a
o
u
e
r,
A.
Ou
n
i,
a
n
d
M
.
Ke
ss
e
n
ti
n
i
,
“
Do
d
e
sig
n
m
e
tri
c
s
c
a
p
t
u
re
d
e
v
e
lo
p
e
rs
p
e
rc
e
p
ti
o
n
o
f
q
u
a
li
t
y
?
a
n
e
m
p
iri
c
a
l
st
u
d
y
o
n
se
lf
-
a
ffir
m
e
d
re
fa
c
to
rin
g
a
c
ti
v
it
ies
,
”
Pro
c
e
e
d
in
g
s
o
f
t
h
e
1
3
t
h
ACM
/IE
E
E
In
ter
n
a
ti
o
n
a
l
S
y
mp
o
si
u
m o
n
Emp
irica
l
S
o
ft
w
a
r
e
En
g
i
n
e
e
rin
g
a
n
d
M
e
a
su
re
me
n
t
(ES
EM
2
0
1
9
),
p
p
.
3
0
0
-
3
1
1
,
2
0
1
9
.
[4
]
Y.
Ka
tao
k
a
,
T.
Im
a
i,
H.
A
n
d
o
u
,
a
n
d
T.
F
u
k
a
y
a
,
“
A
q
u
a
n
ti
tati
v
e
e
v
a
lu
a
ti
o
n
o
f
m
a
in
tain
a
b
il
it
y
e
n
h
a
n
c
e
m
e
n
t
b
y
re
fa
c
to
rin
g
,
”
I
n
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
S
o
ft
w
a
re
M
a
in
ten
a
n
c
e
,
P
ro
c
e
e
d
in
g
s
,
p
p
.
5
7
6
-
58
5
,
2
0
0
2
.
[5
]
E.
A.
AlOm
a
r,
M
.
W.
M
k
a
o
u
e
r,
A.
Ou
n
i
a
n
d
M
.
Ke
ss
e
n
ti
n
i,
"
O
n
th
e
Im
p
a
c
t
o
f
Re
fa
c
to
ri
n
g
o
n
t
h
e
Re
latio
n
s
h
ip
b
e
twe
e
n
Qu
a
li
ty
Attr
ib
u
tes
a
n
d
D
e
sig
n
M
e
tri
c
s,"
2
0
1
9
AC
M
/IE
EE
In
ter
n
a
t
io
n
a
l
S
y
mp
o
si
u
m
o
n
Emp
irica
l
S
o
ft
wa
re
En
g
i
n
e
e
rin
g
a
n
d
M
e
a
su
re
me
n
t
(E
S
EM
),
P
o
rto
d
e
G
a
li
n
h
a
s,
Re
c
ife,
Bra
z
il
,
p
p
.
1
-
11
,
2
0
1
9
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
Ha
r
n
ess
in
g
d
ee
p
lea
r
n
in
g
a
lg
o
r
ith
ms to
p
r
ed
ict
s
o
ftw
a
r
e
r
e
f
a
cto
r
in
g
(
Ma
md
o
u
h
A
len
ezi
)
2981
[6
]
M
.
Ia
m
m
a
rin
o
,
F
.
Zam
p
e
tt
i
,
L
.
A
v
e
rsa
n
o
a
n
d
M
.
Di
P
e
n
ta,
"
S
e
lf
-
A
d
m
it
ted
Tec
h
n
ica
l
De
b
t
Re
m
o
v
a
l
a
n
d
Re
fa
c
to
ri
n
g
Ac
ti
o
n
s:
C
o
-
Oc
c
u
rre
n
c
e
o
r
M
o
re
?
,
"
2
0
1
9
IEE
E
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
S
o
ft
w
a
re
M
a
i
n
ten
a
n
c
e
a
n
d
Evo
lu
ti
o
n
(ICS
M
E)
,
Clev
e
lan
d
,
OH
,
USA
,
p
p
.
1
8
6
-
1
9
0
,
2
0
1
9
.
[7
]
T
.
M
e
n
s
a
n
d
T
.
To
u
rwé
,
“
A
su
rv
e
y
o
f
so
ftwa
re
re
fa
c
to
rin
g
,”
IEE
E
T
ra
n
s
a
c
ti
o
n
s
o
n
so
ft
w
a
re
e
n
g
i
n
e
e
rin
g
,
vol
.
3
0
,
n
o
.
2
,
p
p
.
1
2
6
-
1
3
9
,
2
0
0
4
.
[8
]
M
.
Ak
o
u
r
,
a
n
d
M
.
Ale
n
e
z
i.
"
Tes
t
su
it
e
s
e
ffe
c
ti
v
e
n
e
ss
e
v
o
l
u
ti
o
n
in
o
p
e
n
so
u
rc
e
sy
ste
m
s:
e
m
p
iri
c
a
l
st
u
d
y
,
"
In
d
o
n
e
sia
n
J
o
u
rn
a
l
o
f
El
e
c
trica
l
En
g
in
e
e
rin
g
a
n
d
Co
m
p
u
ter
S
c
ien
c
e
,
v
o
l.
1
9
,
n
o
.
2
,
p
p
.
1
0
8
5
-
1
0
9
2
,
2
0
2
0
.
[9
]
M.
Ale
n
e
z
i,
,
M
.
Ak
o
u
r,
&
H.
Al
S
g
h
a
ier
,
“
T
h
e
Im
p
a
c
t
o
f
Co
-
e
v
o
l
u
ti
o
n
o
f
Co
d
e
P
ro
d
u
c
ti
o
n
a
n
d
Tes
t
S
u
it
e
s
t
h
ro
u
g
h
S
o
ftwa
re
Re
lea
se
s
in
Op
e
n
S
o
u
rc
e
S
o
ftwa
re
S
y
ste
m
s
,”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
I
n
n
o
v
a
ti
v
e
T
e
c
h
n
o
l
o
g
y
a
n
d
Exp
lo
ri
n
g
E
n
g
in
e
e
rin
g
,
v
o
l.
9
,
n
o
.
1
,
p
p
.
2
7
3
7
-
2
7
3
9
,
2
0
1
9
.
[1
0
]
M
.
Ale
n
e
z
i,
,
M
.
A
k
o
u
r,
A
.
Hu
ss
i
e
n
,
a
n
d
M
.
Z.
Al
-
S
a
a
d
,
"
Tes
t
s
u
it
e
e
ffe
c
ti
v
e
n
e
ss
:
a
n
in
d
ica
to
r
f
o
r
o
p
e
n
so
u
rc
e
so
ftwa
re
q
u
a
li
t
y
,
"
2
0
1
6
2
n
d
In
ter
n
a
ti
o
n
a
l
C
o
n
fer
e
n
c
e
o
n
Op
e
n
S
o
u
rc
e
S
o
ft
wa
re
Co
m
p
u
t
in
g
,
p
p
.
1
-
5
,
2
0
1
6
.
[1
1
]
M
.
Kim
,
T.
Zi
m
m
e
rm
a
n
n
,
a
n
d
N.
Na
g
a
p
p
a
n
,
“
A
field
st
u
d
y
o
f
re
fa
c
to
rin
g
c
h
a
ll
e
n
g
e
s
a
n
d
b
e
n
e
fit
s,”
Pro
c
e
e
d
in
g
s
o
f
th
e
ACM
S
IGS
OF
T
2
0
t
h
In
ter
n
a
ti
o
n
a
l
S
y
mp
o
siu
m
o
n
th
e
Fo
u
n
d
a
ti
o
n
s
o
f
S
o
ft
w
a
re
En
g
in
e
e
rin
g
,
p
p
.
5
0
,
2
0
1
2
.
[1
2
]
R.
M
a
rin
e
sc
u
,
“
De
tec
ti
o
n
stra
te
g
ies
:
M
e
tri
c
s
-
b
a
se
d
r
u
les
fo
r
d
e
tec
ti
n
g
d
e
sig
n
flaw
s,”
2
0
th
I
EE
E
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
S
o
ft
w
a
re
M
a
i
n
ten
a
n
c
e
,
Pro
c
e
e
d
i
n
g
s IE
EE
,
p
p
.
3
5
0
-
359
,
2
0
0
4
.
[1
3
]
T.
T
o
u
rw´e
a
n
d
T.
M
e
n
s,
“
Id
e
n
ti
f
y
in
g
re
fa
c
to
ri
n
g
o
p
p
o
rt
u
n
it
ies
u
si
n
g
lo
g
ic
m
e
ta
p
ro
g
ra
m
m
in
g
,
”
S
e
v
e
n
th
Eu
r
o
p
e
a
n
Co
n
fer
e
n
c
e
o
n
S
o
ft
wa
re
M
a
i
n
ten
a
n
c
e
a
n
d
Ree
n
g
in
e
e
rin
g
,
p
p
.
9
1
-
100
,
2
0
0
3
.
[1
4
]
Y.
Ka
tao
k
a
,
T.
Im
a
i,
H.
A
n
d
o
u
,
a
n
d
T.
F
u
k
a
y
a
,
“
A
q
u
a
n
ti
tati
v
e
e
v
a
lu
a
ti
o
n
o
f
m
a
in
tain
a
b
il
it
y
e
n
h
a
n
c
e
m
e
n
t
b
y
re
fa
c
to
rin
g
,
”
I
n
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
S
o
ft
w
a
re
M
a
in
ten
a
n
c
e
,
P
ro
c
e
e
d
in
g
s
,
p
p
.
5
7
6
-
5
8
5
,
2
0
0
2
.
[1
5
]
T.
M
a
rian
i
a
n
d
S
.
R.
Ve
r
g
il
i
o
,
“
A
sy
ste
m
a
ti
c
re
v
iew
o
n
se
a
rc
h
-
b
a
se
d
re
fa
c
to
rin
g
,
”
In
f
o
rm
a
ti
o
n
a
n
d
S
o
ft
w
a
re
T
e
c
h
n
o
l
o
g
y
,
v
o
l.
8
3
,
p
p
.
1
4
-
3
4
,
2
0
1
7
.
[1
6
]
M
.
O’K
e
e
ffe
a
n
d
M
.
O
.
Ci
n
n
´e
i
d
e
,
“
S
e
a
rc
h
-
b
a
se
d
re
fa
c
to
rin
g
f
o
r
so
ftwa
re
m
a
in
te
n
a
n
c
e
,
”
J
o
u
rn
a
l
o
f
S
y
ste
ms
a
n
d
So
ft
w
a
re
,
v
o
l.
8
1
,
n
o
.
4
,
p
p
.
5
0
2
-
5
1
6
,
2
0
0
8
.
[1
7
]
O
.
Al
Qa
se
m
,
M
.
Ak
o
u
r
,
"
S
o
ftw
a
re
F
a
u
lt
P
re
d
ictio
n
Us
in
g
De
e
p
Lea
rn
in
g
Alg
o
rit
h
m
s
,
"
I
n
ter
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
Op
e
n
S
o
u
rc
e
S
o
ft
w
a
re
a
n
d
Pr
o
c
e
ss
e
s
,
vol
.
10
,
pp
.
1
-
19
,
2
0
1
9
.
[1
8
]
J.
Jia
rp
a
k
d
e
e
,
C.
Ta
n
ti
t
h
a
m
th
a
v
o
r
n
,
H.
K.
Da
m
a
n
d
J.
G
ru
n
d
y
,
"
An
Emp
iri
c
a
l
S
tu
d
y
o
f
M
o
d
e
l
-
A
g
n
o
stic T
e
c
h
n
iq
u
e
s
fo
r
De
fe
c
t
P
re
d
ictio
n
M
o
d
e
ls,
"
IE
EE
T
ra
n
sa
c
ti
o
n
s
o
n
S
o
f
twa
re
E
n
g
in
e
e
rin
g
,
p
p
.
1
-
1
,
2
0
2
0
.
[
1
9
]
K
.
L
i
u
,
D
.
K
i
m
,
T
.
F
.
B
i
s
s
y
a
n
d
´
e
,
T
.
K
i
m
,
K
.
K
i
m
,
A
.
K
o
y
u
n
c
u
,
S
.
K
i
m
,
a
n
d
Y
.
L
.
T
r
a
o
n
,
“
L
e
a
r
n
i
n
g
t
o
s
p
o
t
a
n
d
r
e
f
a
c
t
o
r
i
n
c
o
n
s
i
s
t
e
n
t
m
e
t
h
o
d
n
a
m
e
s
,
”
P
r
o
c
e
e
d
i
n
g
s
o
f
t
h
e
4
1
s
t
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
S
o
f
t
w
a
r
e
E
n
g
i
n
e
e
r
i
n
g
,
p
p
.
1
-
1
2
,
2
0
1
9
.
[2
0
]
H
.
Alsg
h
a
ier
,
M
.
Ak
o
u
r
,
“
S
o
ftw
a
re
fa
u
lt
p
re
d
icti
o
n
u
si
n
g
p
a
rti
c
l
e
sw
a
rm
a
lg
o
rit
h
m
with
g
e
n
e
ti
c
a
lg
o
rit
h
m
a
n
d
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
c
las
sifier
,”
S
o
ft
w
a
re
:
Pra
c
ti
c
e
a
n
d
Exp
e
rie
n
c
e
,
v
o
l.
50
,
n
o
.
4
,
p
p
.
4
0
7
-
27.
2
0
2
0
.
[2
1
]
M.
Ak
o
u
r,
H.
Al
sg
h
a
ier
,
a
n
d
O.
Al
Qa
se
m
,
“
Th
e
e
ff
e
c
ti
v
e
n
e
ss
o
f
u
sin
g
d
e
e
p
lea
rn
in
g
a
lg
o
ri
th
m
s
in
p
re
d
ictin
g
stu
d
e
n
t
s
a
c
h
iev
e
m
e
n
ts
,”
In
d
o
n
e
sia
n
J
.
E
lec
t.
En
g
.
C
o
mp
u
t
.
S
c
i
.
,
v
o
l
.
19
,
n
o
.
1
,
p
p
.
3
8
7
-
3
9
3
.
2
0
2
0
.
[
2
2
]
M
.
A
k
o
u
r
,
&
O
.
A
l
Q
a
s
e
m
,
H
.
A
l
S
g
h
a
i
e
r
,
a
n
d
K
.
A
l
-
R
a
d
a
i
d
e
h
,
"
T
h
e
e
f
f
e
c
t
i
v
e
n
e
s
s
o
f
u
s
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
m
s
i
n
p
r
e
d
i
c
t
i
n
g
d
a
i
l
y
a
c
t
i
v
i
t
i
e
s
,
"
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
A
d
v
a
n
c
e
d
T
r
e
n
d
s
i
n
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
r
i
n
g
,
v
o
l
.
8
,
n
o
.
5
,
p
p
.
2
2
3
1
-
3
5
,
2019.
[2
3
]
M.
Ak
o
u
r
,
a
n
d
W
.
Ya
h
y
a
M
e
lh
e
m
,
"
S
o
ftwa
re
De
fe
c
t
P
re
d
ict
io
n
Us
in
g
G
e
n
e
ti
c
P
ro
g
ra
m
m
in
g
a
n
d
Ne
u
ra
l
Ne
two
rk
s
,
"
I
n
ter
n
a
ti
o
n
a
l
J
o
u
r
n
a
l
o
f
Op
e
n
S
o
u
rc
e
S
o
ft
w
a
re
a
n
d
Pr
o
c
e
ss
e
s
,
v
o
l.
8
,
n
o
.
4
,
p
p
.
32
-
51
,
2
0
1
7
.
[2
4
]
M.
Ak
o
u
r,
S
.
Ba
n
it
a
a
n
,
H.
Al
S
g
h
a
ier,
a
n
d
K
.
Al
-
Ra
d
a
id
e
h
,
"
P
re
d
ictin
g
Da
il
y
Ac
ti
v
i
ti
e
s
Eff
e
c
ti
v
e
n
e
ss
Us
in
g
Ba
se
-
lev
e
l
a
n
d
M
e
ta
lev
e
l
Clas
sifiers
,
"
2
0
1
9
7
th
I
n
ter
n
a
ti
o
n
a
l
S
y
mp
o
si
u
m
o
n
Dig
it
a
l
Fo
re
n
sic
s
a
n
d
S
e
c
u
ri
ty
,
p
p
.
1
-
7
,
2
0
1
9
.
[2
5
]
M
.
An
ic
h
e
,
E.
M
a
z
iero
,
R.
Du
re
ll
i,
a
n
d
V.
Du
re
ll
i,
"
Th
e
Eff
e
c
ti
v
e
n
e
ss
o
f
S
u
p
e
rv
ise
d
M
a
c
h
in
e
Lea
rn
in
g
Al
g
o
r
it
h
m
s
in
P
re
d
ictin
g
S
o
ftwa
re
Re
fa
c
to
rin
g
,
"
a
rXiv p
re
p
ri
n
t
a
rXiv:2
0
0
1
.
0
3
3
3
8
,
2
0
2
0
.
[2
6
]
L.
Ku
m
a
r
a
n
d
A.
S
u
re
k
a
,
“
Ap
p
l
ica
ti
o
n
o
f
L
S
S
VM
a
n
d
S
M
OTE
o
n
S
e
v
e
n
Op
e
n
S
o
u
rc
e
P
r
o
jec
ts
fo
r
P
re
d
ictin
g
Re
fa
c
to
rin
g
a
t
Clas
s L
e
v
e
l
,”
2
0
1
7
2
4
t
h
Asi
a
-
P
a
c
if
i
c
S
o
ft
wa
re
E
n
g
i
n
e
e
rin
g
C
o
n
fer
e
n
c
e
(APS
EC)
,
p
p
.
90
-
99
,
2
0
1
7
.
[2
7
]
A
.
Bo
u
k
h
d
h
ir,
M
.
Ke
ss
e
n
ti
n
i
,
S
.
Be
c
h
ik
h
,
J
.
De
a
,
a
n
d
L
.
Be
n
S
a
i
d
,
"
On
t
h
e
u
se
o
f
m
a
c
h
in
e
lea
rn
i
n
g
a
n
d
se
a
rc
h
-
b
a
se
d
so
ftwa
re
e
n
g
in
e
e
rin
g
f
o
r
il
l
-
d
e
fin
e
d
fit
n
e
ss
fu
n
c
ti
o
n
:
a
c
a
se
stu
d
y
o
n
so
ftwa
re
r
e
fa
c
to
rin
g
,
"
I
n
ter
n
a
ti
o
n
a
l
S
y
mp
o
siu
m
o
n
S
e
a
rc
h
B
a
se
d
S
o
ft
w
a
re
En
g
i
n
e
e
rin
g
,
p
p
.
3
1
-
45
,
2
0
1
4
.
[2
8
]
K
.
Lo
v
,
S
.
M
o
u
li
S
a
tap
a
t
h
y
,
a
n
d
L
.
Bh
a
n
u
M
u
rth
y
,
"
M
e
th
o
d
Lev
e
l
Re
fa
c
to
rin
g
P
re
d
icti
o
n
o
n
F
i
v
e
Op
e
n
S
o
u
rc
e
Ja
v
a
P
ro
jec
ts
u
si
n
g
M
a
c
h
i
n
e
Lea
rn
in
g
Tec
h
n
iq
u
e
s
,
"
Pr
o
c
e
e
d
in
g
s
o
f
th
e
1
2
th
In
n
o
v
a
ti
o
n
s
o
n
S
o
ft
wa
re
En
g
i
n
e
e
rin
g
Co
n
fer
e
n
c
e
(fo
rm
e
rly
k
n
o
w
n
a
s I
n
d
i
a
S
o
ft
w
a
re
En
g
in
e
e
rin
g
Co
n
fe
re
n
c
e
),
p
p
.
1
-
1
0
.
2
0
1
9
.
[2
9
]
J
.
Al
Da
ll
a
l,
"
P
re
d
ictin
g
m
o
v
e
m
e
th
o
d
re
fa
c
to
rin
g
o
p
p
o
rt
u
n
i
ti
e
s
in
o
b
jec
t
-
o
rie
n
ted
c
o
d
e
,
"
In
f
o
rm
a
ti
o
n
a
n
d
S
o
ft
wa
re
T
e
c
h
n
o
l
o
g
y
,
v
o
l.
92
,
p
p
.
1
0
5
-
1
2
0
,
2
0
1
7
.
[3
0
]
M
.
An
ic
h
e
,
E.
M
a
z
iero
,
R.
Du
re
ll
i,
a
n
d
V.
Du
re
ll
i,
"
Th
e
Eff
e
c
ti
v
e
n
e
ss
o
f
S
u
p
e
rv
ise
d
M
a
c
h
in
e
Lea
rn
in
g
Al
g
o
r
it
h
m
s
in
P
re
d
ictin
g
S
o
ftwa
re
Re
fa
c
to
rin
g
,
"
a
rXiv p
re
p
ri
n
t
a
rXiv:2
0
0
1
.
0
3
3
3
8
,
2
0
2
0
.
[3
1
]
P
.
Je
v
g
e
n
i
ja,
G
.
Ba
v
o
ta,
M
.
T
u
fa
n
o
,
a
n
d
D
.
P
o
sh
y
v
a
n
y
k
,
"
T
o
wa
rd
s
ju
st
-
in
-
ti
m
e
re
fa
c
to
rin
g
re
c
o
m
m
e
n
d
e
rs
,
"
Pro
c
e
e
d
in
g
s
o
f
t
h
e
2
6
t
h
Co
n
fer
e
n
c
e
o
n
Pro
g
ra
m C
o
mp
re
h
e
n
s
i
o
n
,
p
p
.
3
1
2
-
3
1
5
,
2
0
1
8
.
[3
2
]
P.
He
g
e
d
ű
s,
I.
Ká
d
á
r,
R.
F
e
re
n
c
,
a
n
d
T
.
G
y
imó
th
y
,
“
Emp
iri
c
a
l
e
v
a
lu
a
ti
o
n
o
f
s
o
ftwa
re
m
a
in
tain
a
b
il
it
y
b
a
se
d
o
n
a
m
a
n
u
a
ll
y
v
a
li
d
a
ted
re
fa
c
to
ri
n
g
d
a
tas
e
t
,”
In
fo
rm
a
t
io
n
a
n
d
S
o
ft
wa
r
e
T
e
c
h
n
o
l
o
g
y
,
v
o
l.
9
5
,
p
p
.
3
1
3
-
327
,
2
0
1
8
.
[3
3
]
B.
c
h
o
,
D.
v
a
n
M
e
rrie
n
b
o
e
r,
a
n
d
Y.
Be
n
g
io
,
“
On
th
e
p
r
o
p
e
rti
e
s
o
f
n
e
u
ra
l
m
a
c
h
in
e
tran
sla
ti
o
n
:
E
n
c
o
d
e
r
-
d
e
c
o
d
e
r
a
p
p
ro
a
c
h
e
s
,”
a
rX
iv p
re
p
rin
t
a
rX
i
v
:1
4
0
9
.
1
2
5
9
,
2
0
1
4
.
[3
4
]
J.
Ch
u
n
g
,
C
.
G
u
lce
h
re
,
K.
Ch
o
,
Y.
Be
n
g
i
o
,
Emp
ir
ica
l
e
v
a
lu
a
ti
o
n
o
f
g
a
ted
re
c
u
rre
n
t
n
e
u
ra
l
n
e
two
r
k
s
o
n
se
q
u
e
n
c
e
m
o
d
e
li
n
g
,”
a
rX
iv p
re
p
rin
t
a
rX
iv:1
4
1
2
.
3
5
5
5
,
2
0
1
4
.
[3
5
]
G.
S
h
e
n
,
Q.
Tan
,
H.
Zh
a
n
g
,
P.
Z
e
n
g
,
J.
Xu
,
“
De
e
p
Lea
rn
in
g
with
Ga
ted
Re
c
u
r
re
n
t
Un
it
Ne
two
rk
s
fo
r
F
i
n
a
n
c
ia
l
Se
qu
e
n
c
e
P
re
d
i
,
”
Pro
c
e
d
i
a
C
o
mp
u
ter
S
c
ien
c
e
,
v
o
l.
1
3
1
,
p
p
.
8
9
5
-
903
,
2
0
1
8
.
[3
6
]
Y.
G
a
o
,
a
n
d
D.
G
lo
wa
c
k
a
,
“
De
e
p
G
a
te Re
c
u
rre
n
t
Ne
u
ra
l
Ne
two
rk
,”
ArXi
v
,
a
b
s/
1
6
0
4
.
0
2
9
1
0
,
2
0
1
6
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
1
8
,
No
.
6
,
Dec
em
b
e
r
2
0
2
0
:
29
7
7
-
298
2
2982
[3
7
]
O.
Alq
a
se
m
,
,
M
.
A
k
o
u
r,
a
n
d
M
.
Ale
n
e
z
i,
“
Th
e
In
fl
u
e
n
c
e
o
f
De
e
p
Lea
rn
in
g
Alg
o
rit
h
m
s
F
a
c
to
rs
i
n
S
o
ftwa
re
F
a
u
lt
P
re
d
ictio
n
,”
IEE
E
Acc
e
ss
,
v
o
l
.
8
,
p
p
.
6
3
9
4
5
-
6
3
9
6
0
,
2
0
2
0
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Ma
m
d
o
u
h
Ale
n
e
z
i
is
c
u
rre
n
tl
y
t
h
e
De
a
n
o
f
Ed
u
c
a
ti
o
n
a
l
S
e
r
v
ice
s
a
t
P
rin
c
e
S
u
lt
a
n
U
n
iv
e
rsit
y
.
Dr.
Ale
n
e
z
i
re
c
e
iv
e
d
h
is
M
S
a
n
d
P
h
.
D.
d
e
g
re
e
s
fr
o
m
De
P
a
u
l
U
n
iv
e
rsity
a
n
d
N
o
rth
Da
k
o
ta
S
tate
Un
iv
e
rsity
in
2
0
1
1
a
n
d
2
0
1
4
,
re
sp
e
c
ti
v
e
ly
.
He
h
a
s
e
x
ten
siv
e
e
x
p
e
rien
c
e
in
d
a
ta
m
in
i
n
g
a
n
d
m
a
c
h
in
e
lea
rn
in
g
wh
e
re
h
e
a
p
p
li
e
d
se
v
e
ra
l
d
a
ta
m
in
i
n
g
tec
h
n
i
q
u
e
s
to
so
lv
e
se
v
e
ra
l
S
o
ftwa
re
En
g
i
n
e
e
rin
g
p
r
o
b
lem
s.
He
c
o
n
d
u
c
ted
se
v
e
ra
l
re
se
a
rc
h
a
re
a
s
a
n
d
d
e
v
e
lo
p
m
e
n
t
o
f
p
re
d
icti
v
e
m
o
d
e
ls
u
si
n
g
m
a
c
h
in
e
lea
rn
in
g
t
o
p
re
d
ict
fa
u
lt
-
p
ro
n
e
c
las
se
s,
c
o
m
p
re
h
e
n
d
so
u
rc
e
c
o
d
e
,
a
n
d
p
re
d
ict
t
h
e
a
p
p
r
o
p
r
iate
d
e
v
e
lo
p
e
r
to
b
e
a
ss
ig
n
e
d
t
o
a
n
e
w b
u
g
.
Mo
h
a
m
m
e
d
Ak
o
u
r
is
a
n
a
ss
o
c
iate
P
ro
fe
ss
o
r
o
f
S
o
ftwa
re
E
n
g
in
e
e
rin
g
a
t
Al
Ya
m
a
m
a
h
Un
iv
e
rsity
(YU
).
He
g
o
t
h
is
Ba
c
h
e
lo
r’s
(
2
0
0
6
)
a
n
d
M
a
ste
r’s
(
2
0
0
8
)
d
e
g
re
e
fro
m
Ya
rm
o
u
k
Un
iv
e
rsity
i
n
C
o
m
p
u
ter
I
n
fo
rm
a
t
io
n
S
y
ste
m
s
with
Ho
n
o
r.
He
jo
i
n
e
d
Ya
rm
o
u
k
Un
i
v
e
rsity
a
s
a
Lec
tu
re
r
in
Au
g
u
st
2
0
0
8
a
fter
g
r
a
d
u
a
ti
n
g
with
h
is
m
a
ste
r’s
in
Co
m
p
u
ter
In
f
o
rm
a
t
io
n
S
y
ste
m
s.
In
Au
g
u
st
2
0
0
9
,
He
left
Ya
rm
o
u
k
Un
iv
e
rsit
y
to
p
u
rsu
e
h
is
P
h
.
D.
in
S
o
ftwa
re
En
g
in
e
e
ri
n
g
a
t
No
rth
Da
k
o
ta
S
tate
Un
i
v
e
rsity
(
ND
S
U).
He
jo
i
n
e
d
Ya
rm
o
u
k
Un
iv
e
rsity
a
g
a
in
in
Ap
ril
2
0
1
3
a
fter
g
ra
d
u
a
ti
n
g
wit
h
h
is
P
h
.
D.
i
n
S
o
ftwa
re
En
g
i
n
e
e
rin
g
fr
o
m
ND
S
U
with
Ho
n
o
r
.
He
se
rv
e
s
a
s
Ke
y
n
o
te S
p
e
a
k
e
r,
Or
g
a
n
ize
r,
a
C
o
-
c
h
a
ir
a
n
d
p
u
b
li
c
it
y
Ch
a
ir
f
o
r
se
v
e
ra
l
IEE
E
c
o
n
fe
re
n
c
e
s,
a
n
d
a
s
ERB
fo
r
m
o
re
t
h
a
n
1
0
I
S
I
i
n
d
e
x
e
d
p
re
stig
i
o
u
s
jo
u
rn
a
ls.
He
is
a
m
e
m
b
e
r
o
f
t
h
e
In
tern
a
ti
o
n
a
l
As
so
c
iatio
n
o
f
En
g
in
e
e
rs
(IAE
NG
).
Dr.
Ak
o
u
r
a
t
Ya
rm
o
u
k
U
n
iv
e
rsit
y
se
rv
e
d
a
s
He
a
d
o
f
a
c
c
re
d
it
a
ti
o
n
a
n
d
Qu
a
li
t
y
a
ss
u
ra
n
c
e
a
n
d
th
e
n
wa
s
h
ired
a
s
d
irec
t
o
r
o
f
c
o
m
p
u
ter
a
n
d
I
n
fo
rm
a
ti
o
n
Ce
n
ter.
I
n
2
0
1
8
,
Dr.
Ak
o
u
r
h
a
s
b
e
e
n
h
ire
d
a
s
Vic
e
De
a
n
o
f
S
t
u
d
e
n
t
Affa
irs
a
t
Ya
rm
o
u
k
Un
iv
e
rsity
.
O
sa
m
a
Al
Q
a
se
m
go
t
h
is
M
a
ste
r’s
d
e
g
re
e
in
C
o
m
p
u
ter
I
n
fo
rm
a
ti
o
n
S
y
ste
m
s
fro
m
t
h
e
F
a
c
u
lt
y
o
f
In
f
o
rm
a
ti
o
n
Tec
h
n
o
l
o
g
y
a
n
d
Co
m
p
u
ter
S
c
ien
c
e
s,
Ya
rm
o
u
k
U
n
iv
e
rsit
y
.
Al
Qa
se
m
h
a
s
fe
w
p
u
b
li
c
a
ti
o
n
s
in
t
h
e
field
s
o
f
so
ftwa
re
e
n
g
in
e
e
rin
g
,
b
ig
d
a
ta
a
n
a
ly
ti
c
s,
a
n
d
s
o
ftwa
re
fa
u
lt
p
re
d
icti
o
n
a
n
d
c
u
rre
n
tl
y
h
e
is w
o
r
k
in
g
o
n
b
u
il
d
i
n
g
str
o
n
g
re
se
a
rc
h
i
n
th
e
a
re
a
o
f
b
i
g
d
a
ta.
Evaluation Warning : The document was created with Spire.PDF for Python.