T
E
L
K
O
M
NIKA
T
elec
o
mm
un
ica
t
io
n Co
m
pu
t
i
ng
E
lect
ro
nics
a
nd
Co
ntr
o
l
Vo
l.
2
3
,
No
.
2
,
A
p
r
il 2
0
2
5
,
p
p
.
3
8
2
~
3
9
2
I
SS
N:
1
6
9
3
-
6
9
3
0
,
DOI
: 1
0
.
1
2
9
2
8
/
T
E
L
KOM
NI
K
A
.
v
2
3
i2
.
2
6
5
1
0
382
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//telko
mn
ika
.
u
a
d
.
a
c.
i
d
O
v
ersa
m
pling
v
s.
undersa
m
pling
in
TF
-
ID
F
v
a
ria
ti
o
ns
for
i
m
ba
la
nced
Indo
nesia
n
sho
rt
t
ex
ts
cla
ss
ificatio
n
I
Ny
o
m
a
n P
ra
y
a
na
T
risn
a
,
Ni
Wa
y
a
n E
mm
y
Ro
s
ia
na
D
ew
i,
M
uh
a
mm
a
d Ala
m
P
a
s
ir
ull
o
h
D
e
p
a
r
t
me
n
t
o
f
I
n
f
o
r
mat
i
o
n
T
e
c
h
n
o
l
o
g
y
,
F
a
c
u
l
t
y
o
f
En
g
i
n
e
e
r
i
n
g
,
U
d
a
y
a
n
a
U
n
i
v
e
r
si
t
y
,
B
a
l
i
,
I
n
d
o
n
e
si
a
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
u
l 2
9
,
2024
R
ev
i
s
ed
Oct
25,
2
0
2
4
A
cc
ep
ted
Dec
26
,
2
0
2
4
Ev
e
n
th
o
u
g
h
it
is
c
o
n
si
d
e
re
d
a
m
o
re
trad
it
io
n
a
l
m
e
th
o
d
c
o
m
p
a
re
d
to
m
o
re
m
o
d
e
rn
a
lg
o
rit
h
m
s,
term
f
re
q
u
e
n
c
y
in
v
e
rse
d
d
o
c
u
m
e
n
t
f
re
q
u
e
n
c
y
(T
F
-
IDF)
n
e
v
e
rth
e
les
s
p
ro
d
u
c
e
s
g
o
o
d
re
su
lt
s
in
a
ra
n
g
e
o
f
tex
t
m
in
in
g
tas
k
s.
T
h
is
stu
d
y
a
ss
e
s
se
s
th
e
e
ff
e
c
ti
v
e
n
e
ss
o
f
se
v
e
r
a
l
T
F
-
IDF
m
o
d
if
ica
ti
o
n
s
f
o
r
sh
o
r
t
tex
t
c
las
si
f
ica
ti
o
n
.
Im
b
a
lan
c
e
d
d
a
tas
e
ts
a
re
a
n
o
th
e
r
issu
e
th
a
t
is
a
d
d
re
ss
e
d
in
th
is
re
se
a
rc
h
.
T
o
re
c
ti
fy
th
e
i
m
b
a
lan
c
e
d
issu
e
,
w
e
in
teg
r
a
te
sta
n
d
a
rd
,
l
o
g
-
sc
a
led
,
a
n
d
b
o
o
lea
n
T
F
-
IDF
in
sh
o
rt
tex
t
c
las
sif
ic
a
ti
o
n
w
it
h
u
n
d
e
r
sa
m
p
li
n
g
a
n
d
o
v
e
rsa
m
p
li
n
g
m
e
th
o
d
s.
P
re
c
isio
n
,
re
c
a
ll
,
a
n
d
f
-
m
e
a
su
re
m
e
tri
c
s
a
re
u
se
d
to
e
v
a
lu
a
te
e
a
c
h
e
x
p
e
ri
m
e
n
t.
T
h
e
b
e
st
re
su
lt
is
o
b
tain
e
d
w
h
e
n
a
p
p
ly
in
g
b
o
o
lea
n
T
F
-
IDF
w
it
h
th
e
o
v
e
r
sa
m
p
li
n
g
m
e
th
o
d
.
Ov
e
rsa
m
p
li
n
g
m
e
th
o
d
s
o
u
t
p
e
rf
o
rm
th
e
u
n
d
e
rsa
m
p
li
n
g
m
e
th
o
d
s
i
n
e
v
e
r
y
e
x
p
e
ri
m
e
n
t,
a
lt
h
o
u
g
h
th
e
re
a
re
so
m
e
c
a
se
s
w
h
e
re
e
x
p
e
ri
m
e
n
ts
w
it
h
u
n
d
e
rsa
m
p
li
n
g
m
e
th
o
d
s
a
re
c
o
n
sid
e
ra
b
le.
A
d
d
it
i
o
n
a
ll
y
,
o
u
r
c
o
n
d
u
c
ted
stu
d
y
re
v
e
a
ls
th
a
t
e
m
p
lo
y
in
g
m
o
d
if
ied
T
F
-
IDF,
su
c
h
a
s
b
o
o
lea
n
o
r
l
o
g
-
sc
a
led
v
e
rsio
n
s,
p
r
o
v
id
e
s
g
re
a
ter
a
d
v
a
n
tag
e
s
to
c
las
sif
ic
a
ti
o
n
p
e
rf
o
r
m
a
n
c
e
,
p
a
rti
c
u
larly
in
h
a
n
d
li
n
g
im
b
a
lan
c
e
d
d
a
tas
e
ts,
w
h
e
n
c
o
m
p
a
re
d
to
so
lely
re
ly
in
g
o
n
th
e
sta
n
d
a
rd
T
F
-
IDF
a
p
p
r
o
a
c
h
.
K
ey
w
o
r
d
s
:
B
ah
asa
I
n
d
o
n
es
ia
I
m
b
alan
ce
d
d
ataset
Ov
er
s
a
m
p
lin
g
m
et
h
o
d
Sh
o
r
t
-
tex
t c
la
s
s
i
f
icat
io
n
T
er
m
f
r
eq
u
e
n
c
y
i
n
v
er
s
ed
d
o
cu
m
en
t f
r
eq
u
e
n
c
y
Un
d
er
s
a
m
p
li
n
g
m
e
th
o
d
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
:
I
Ny
o
m
a
n
P
r
ay
a
n
a
T
r
is
n
a
Dep
ar
t
m
en
t o
f
I
n
f
o
r
m
atio
n
T
e
ch
n
o
lo
g
y
,
Fac
u
lt
y
o
f
E
n
g
i
n
ee
r
in
g
,
Ud
a
y
an
a
U
n
i
v
er
s
it
y
St
.
Ka
m
p
u
s
B
u
k
it J
i
m
b
ar
an
,
K
u
ta
Selata
n
,
B
ad
u
n
g
,
B
ali,
I
n
d
o
n
esia
E
m
ail: p
r
a
y
a
n
a.
tr
is
n
a@
u
n
u
d
.
ac
.
id
1.
I
NT
RO
D
UCT
I
O
N
Fo
r
y
ea
r
s
,
m
a
n
y
m
et
h
o
d
o
lo
g
ies
w
er
e
co
n
d
u
cted
f
o
r
th
e
b
es
t
r
esu
lt
in
te
x
t
m
i
n
i
n
g
f
o
r
tex
t
d
o
cu
m
en
t,
s
p
ec
if
icall
y
in
m
ac
h
in
e
lear
n
i
n
g
ap
p
r
o
ac
h
.
A
s
u
n
s
tr
u
ct
u
r
ed
d
ata,
d
o
cu
m
e
n
t
n
ee
d
s
to
b
e
v
ec
to
r
ized
to
u
n
d
er
g
o
in
to
m
ac
h
i
n
e
lear
n
i
n
g
m
et
h
o
d
o
lo
g
ies.
T
h
e
co
m
b
i
n
at
io
n
o
f
t
er
m
f
r
eq
u
e
n
c
y
(
T
F)
alo
n
g
w
it
h
in
v
er
s
e
d
o
cu
m
en
t
f
r
eq
u
en
c
y
(
I
DF)
is
s
till
p
r
o
m
i
n
en
t
v
ec
to
r
izatio
n
f
o
r
tex
t
m
i
n
i
n
g
.
T
h
is
m
et
h
o
d
is
al
s
o
k
n
o
w
n
as
ter
m
f
r
eq
u
en
c
y
i
n
v
er
s
ed
d
o
cu
m
e
n
t
f
r
eq
u
en
c
y
(
TF
-
I
DF
)
.
Ko
w
s
ar
i
et
a
l.
[
1
]
s
tates
alth
o
u
g
h
TF
-
I
DF
–
o
r
T
F
al
o
n
e
f
ailed
to
ca
p
tu
r
e
s
y
n
tactic
a
n
d
s
e
m
an
tic
ch
ar
ac
ter
is
tic
o
f
te
x
t
,
th
e
co
m
p
u
tatio
n
o
f
v
ec
to
r
izat
io
n
is
q
u
ite
s
i
m
p
le
an
d
m
an
a
g
ea
b
le
w
it
h
lo
w
-
r
es
o
u
r
ce
co
m
p
u
t
in
g
.
C
o
m
p
ar
ed
to
s
tate
-
of
-
th
e
-
ar
t
d
o
cu
m
e
n
t
v
e
cto
r
izatio
n
s
u
ch
as
w
o
r
d
-
e
m
b
ed
d
in
g
[
2
]
,
f
astT
ex
t
b
y
Fac
eb
o
o
k
[
3
]
,
o
r
ev
en
tr
an
s
f
o
r
m
er
-
b
ased
lan
g
u
a
g
e
m
o
d
el
lik
e
B
E
R
T
[
4
]
TF
-
I
DF
s
till
y
ield
s
d
ec
e
n
t
p
er
f
o
r
m
a
n
ce
f
o
r
m
u
ltip
le
tex
t
m
in
in
g
tas
k
.
T
h
is
is
al
s
o
p
r
o
v
en
b
y
p
r
io
r
r
esear
ch
b
y
Ma
r
cin
cz
u
k
et
a
l.
[
5
]
,
w
h
o
at
te
m
p
ts
to
co
m
p
ar
e
m
o
d
er
n
ap
p
r
o
ac
h
es
s
u
c
h
a
s
w
o
r
d
2
v
ec
a
n
d
B
E
R
T
w
it
h
T
F
-
I
DF
as
“
clas
s
ical
o
n
e”
in
f
o
u
r
ty
p
es
o
f
d
atasets
.
T
h
e
ex
p
er
i
m
e
n
t
s
h
o
w
s
t
h
at
T
F
-
I
DF
r
an
k
s
as
1
st
p
lace
in
2
d
atasets
an
d
2
nd
in
1
d
ataset
with
s
i
g
n
i
f
ican
tl
y
f
as
ter
co
m
p
u
t
atio
n
ti
m
e.
Do
cu
m
e
n
t
cla
s
s
i
f
icat
io
n
i
s
o
n
e
o
f
p
o
p
u
lar
tex
t
m
i
n
i
n
g
t
a
s
k
.
T
h
is
is
d
o
n
e
b
y
g
r
o
u
p
i
n
g
t
h
e
d
o
cu
m
en
ts
in
to
m
u
ltip
le
clas
s
w
ith
lab
el.
Ko
w
s
ar
i
et
a
l.
[
1
]
s
p
ec
if
ies
th
er
e
ar
e
f
o
u
r
t
y
p
es
o
f
d
o
cu
m
en
t
s
co
p
es
in
tex
t
class
i
f
icatio
n
:
d
o
cu
m
en
t
lev
e
l,
p
ar
ag
r
ap
h
lev
el,
s
en
te
n
ce
l
ev
el,
an
d
s
u
b
-
s
e
n
te
n
ce
lev
e
l,
alth
o
u
g
h
m
o
s
t
o
f
d
o
cu
m
en
t
class
i
f
icatio
n
is
f
o
c
u
s
ed
o
n
s
en
te
n
ce
an
d
/o
r
d
o
cu
m
en
t
lev
el
[
6
]
.
T
h
is
lev
el
o
f
s
co
p
e
also
d
ef
in
es
Evaluation Warning : The document was created with Spire.PDF for Python.
383
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
Ove
r
s
a
mp
lin
g
vs.
u
n
d
ers
a
mp
lin
g
in
TF
-
I
DF
v
a
r
ia
tio
n
s
fo
r
imb
a
la
n
ce
d
…
(
I
N
yo
ma
n
P
r
a
y
a
n
a
Tr
is
n
a
)
len
g
th
o
f
t
h
e
cla
s
s
i
f
ied
te
x
t.
C
o
m
p
ar
ed
to
d
o
cu
m
e
n
t
le
v
el
s
co
p
e
lik
e
n
e
w
s
ar
ticles
[
7
]
o
r
p
ar
ag
r
ap
h
lev
el
d
o
cu
m
en
t
lik
e
ab
s
tr
ac
t
[
8
]
,
s
e
n
ten
ce
le
v
el
d
o
cu
m
e
n
t
s
h
av
e
s
ig
n
i
f
ica
n
t
d
if
f
er
en
ce
s
.
T
h
e
m
ain
d
if
f
er
en
ce
is
t
h
e
n
u
m
b
er
o
f
te
x
ts
in
t
h
e
d
o
cu
m
en
t,
w
h
er
e
d
o
cu
m
e
n
t
o
r
p
ar
ag
r
ap
h
lev
el
d
o
cu
m
e
n
t
h
av
e
lo
n
g
er
te
x
t
t
h
an
s
en
te
n
ce
lev
el
d
o
cu
m
e
n
t.
Sh
o
r
ter
d
o
cu
m
e
n
t
m
ea
n
s
t
h
e
r
eso
u
r
ce
f
o
r
d
o
cu
m
en
t
ta
s
k
s
ar
e
s
ca
r
ce
r
an
d
n
o
is
ier
co
m
p
ar
ed
to
lo
n
g
er
tex
t.
T
h
e
s
ca
r
cit
y
o
f
tex
t
r
eso
u
r
ce
in
s
e
n
ten
ce
le
v
el
d
o
cu
m
e
n
t
class
i
f
i
ca
tio
n
w
il
l
b
e
s
ee
n
as sp
ar
s
e
v
ec
to
r
[
9
]
.
A
lo
n
g
s
id
e
th
e
s
p
ar
s
e
v
ec
to
r
w
h
ich
is
r
es
u
lted
f
r
o
m
s
h
o
r
ter
t
ex
t,
th
e
co
m
m
o
n
p
r
o
b
lem
f
o
r
d
o
cu
m
en
t
class
i
f
icat
io
n
is
t
h
e
i
m
b
alan
ce
d
d
is
tr
ib
u
tio
n
o
f
cla
s
s
e
s
.
I
t
is
u
s
u
all
y
r
ep
h
r
ased
as
i
m
b
a
lan
ce
d
d
ataset.
I
m
b
alan
ce
d
d
ataset
co
u
ld
lead
to
in
ad
eq
u
ate
p
er
f
o
r
m
a
n
ce
o
f
th
e
class
if
ier
m
o
d
el.
T
h
is
is
b
ec
au
s
e
m
aj
o
r
it
y
o
f
class
i
f
icatio
n
m
o
d
els
r
eq
u
ir
e
b
alan
ce
d
class
es
to
o
b
t
ain
o
p
tim
al
p
er
f
o
r
m
a
n
ce
[
1
0
]
.
T
h
e
p
r
o
b
lem
o
f
i
m
b
alan
ce
d
d
atase
t
b
ec
o
m
es
m
o
r
e
u
n
ce
r
tain
a
n
d
p
r
o
b
lem
a
t
ic
if
t
h
e
d
ataset
is
i
m
b
ala
n
ce
d
to
th
e
ex
tr
e
m
e.
T
h
e
ex
tr
e
m
e
i
m
b
ala
n
ce
d
d
ataset
m
ea
n
s
t
h
er
e
ar
e
m
aj
o
r
ity
a
n
d
m
in
o
r
it
y
c
lass
e
s
,
w
h
er
e
t
h
e
p
r
esen
ce
o
f
th
e
m
i
n
o
r
it
y
clas
s
es i
s
o
n
l
y
r
ep
r
esen
ted
w
ith
l
ittl
e
to
n
ea
r
l
y
n
o
n
e
in
s
ta
n
ce
s
co
m
p
ar
ed
to
m
aj
o
r
ity
clas
s
es [
1
1
]
.
Ho
w
e
v
er
,
f
o
r
th
e
p
ast
y
ea
r
s
m
eth
o
d
o
lo
g
ie
s
f
o
r
s
o
lv
i
n
g
i
m
b
ala
n
ce
d
d
ataset
ar
e
d
ev
elo
p
ed
.
E
x
p
er
i
m
e
n
tal
r
ev
ie
w
d
o
n
e
b
y
T
an
h
a
et
a
l.
[
1
2
]
r
ev
ea
ls
f
o
u
r
m
et
h
o
d
s
to
h
an
d
le
i
m
b
alan
ce
d
d
ataset:
d
ata
-
lev
el
m
et
h
o
d
,
alg
o
r
ith
m
-
lev
el
m
et
h
o
d
,
h
y
b
r
id
-
m
et
h
o
d
,
an
d
b
o
o
s
tin
g
-
m
et
h
o
d
.
T
h
e
d
ata
-
lev
e
l
m
et
h
o
d
s
w
o
r
k
b
y
r
esa
m
p
li
n
g
t
h
e
n
u
m
b
er
o
f
t
h
e
in
s
ta
n
ce
s
i
n
th
e
d
ataset
[
1
3
]
.
T
h
e
r
esam
p
li
n
g
p
r
o
ce
s
s
is
ca
lled
u
n
d
er
s
a
m
p
l
in
g
w
h
e
n
t
h
e
i
n
s
ta
n
ce
s
w
it
h
m
aj
o
r
it
y
clas
s
ar
e
s
a
m
p
led
d
o
w
n
s
o
th
at
t
h
o
s
e
i
n
s
tan
ce
s
ar
e
b
ala
n
ce
d
w
it
h
m
i
n
o
r
it
y
in
s
ta
n
ce
s
.
O
n
th
e
co
n
tr
ar
y
,
w
h
en
th
e
m
i
n
o
r
it
y
i
n
s
ta
n
ce
s
ar
e
s
y
n
t
h
e
s
ized
s
o
th
e
m
in
o
r
it
y
cla
s
s
es
h
a
v
e
th
e
s
a
m
e
d
is
tr
ib
u
tio
n
as
m
aj
o
r
ity
cla
s
s
es,
it
is
ca
ll
ed
o
v
er
s
a
m
p
li
n
g
[
1
3
]
.
B
o
t
h
o
f
u
n
d
e
r
s
am
p
l
in
g
a
n
d
o
v
e
r
s
am
p
l
in
g
t
e
c
h
n
i
q
u
e
a
r
e
v
a
s
t
ly
a
p
p
l
ie
d
in
r
e
c
en
t
r
e
s
ea
r
c
h
e
r
s
.
Fu
r
th
e
r
m
o
r
e
,
t
h
es
e
te
ch
n
i
q
u
e
s
a
r
e
f
u
r
t
h
e
r
ly
d
ev
el
o
p
e
d
i
n
t
o
s
ev
e
r
a
l
alg
o
r
i
th
m
s
f
o
r
m
u
l
t
i
p
le
c
a
s
es
,
lik
e
a
d
a
p
t
iv
e
s
y
n
th
et
i
c
(
A
DA
S
Y
N
)
[
1
4
]
,
s
y
n
t
h
e
t
i
c
m
in
o
r
i
ty
o
v
e
r
s
am
p
l
in
g
t
e
c
h
n
i
q
u
e
(
S
MO
T
E
)
[
1
5
]
,
r
a
n
d
o
m
-
b
a
s
e
d
u
n
d
e
r
s
am
p
l
in
g
[
1
6
]
,
a
n
d
n
e
ig
h
b
o
r
-
b
as
e
d
u
n
d
e
r
s
am
p
l
i
n
g
[
1
7
]
.
Ou
r
r
esear
ch
ex
p
er
i
m
en
ts
t
h
e
d
ata
-
lev
el
i
m
b
alan
ce
d
h
an
d
li
n
g
m
et
h
o
d
b
y
co
m
p
ar
i
n
g
o
v
e
r
s
a
m
p
li
n
g
tech
n
iq
u
e
a
n
d
u
n
d
er
s
a
m
p
li
n
g
tech
n
iq
u
e
f
o
r
s
h
o
r
t
te
x
t
clas
s
i
f
icatio
n
.
T
h
e
ex
p
er
i
m
e
n
ts
ar
e
d
o
n
e
s
p
ec
if
ica
ll
y
in
B
ah
asa
I
n
d
o
n
esia
’
s
d
ataset.
Alth
o
u
g
h
it
is
s
p
o
k
en
r
o
u
g
h
l
y
b
y
h
u
n
d
r
ed
s
o
f
m
illi
o
n
s
p
ea
k
er
s
w
o
r
ld
w
id
e
[
1
8
]
,
th
e
r
eso
u
r
ce
s
f
o
r
B
ah
asa
I
n
d
o
n
esia
tex
t
m
in
in
g
tas
k
ar
e
l
i
m
ited
[
1
9
]
.
Fu
r
th
er
m
o
r
e
,
o
u
r
r
esear
ch
em
p
lo
y
s
tr
ad
itio
n
al
T
F
-
I
DF v
ec
to
r
izer
ev
en
s
o
b
ec
au
s
e
its
s
i
m
p
l
icit
y
an
d
th
e
b
en
e
f
icial
to
t
h
e
co
m
p
u
tatio
n
a
l ti
m
e.
2.
RE
L
AT
E
D
WO
RK
T
h
e
f
o
llo
w
i
n
g
p
ar
ag
r
ap
h
s
w
ill
d
is
cu
s
s
ab
o
u
t
th
e
r
elate
d
w
o
r
k
s
t
h
at
in
s
p
ir
e
th
e
r
esear
ch
.
T
h
e
r
elate
d
w
o
r
k
s
f
o
c
u
s
o
n
th
e
v
ec
to
r
izer
m
et
h
o
d
,
i
m
b
alan
ce
d
d
ataset
h
an
d
lin
g
s
,
m
ac
h
i
n
e
lear
n
i
n
g
m
eth
o
d
s
,
an
d
t
h
e
tex
t
class
i
f
icatio
n
p
r
o
b
lem
s
.
Z
h
u
e
t
a
l.
[
2
0
]
u
tili
ze
s
T
F
-
I
DF
m
et
h
o
d
f
o
r
h
o
t
to
p
ic
d
etec
tio
n
i
n
n
e
w
s
ar
tic
les.
T
h
is
r
esear
ch
r
ef
i
n
es
T
F
-
I
DF
v
ec
t
o
r
izer
to
ad
ap
t
to
t
im
e
-
d
is
tr
ib
u
ted
in
f
o
r
m
a
tio
n
a
n
d
u
s
er
att
en
tio
n
.
T
h
e
r
ef
in
ed
v
ec
to
r
is
t
h
e
n
clu
s
ter
ed
w
it
h
c
lu
s
ter
i
n
g
m
eth
o
d
to
ex
tr
ac
t
th
e
h
o
t
to
p
ics
o
f
th
e
n
e
w
s
n
e
t
wo
r
k
.
Si
m
ilar
s
u
b
j
ec
t
o
f
h
o
t
to
p
ic
d
etec
tio
n
is
also
co
n
d
u
cted
b
y
B
o
k
et
a
l.
[
2
1
]
w
h
o
m
o
d
if
ie
s
T
F
-
I
DF
to
ca
r
r
y
o
u
t
th
e
te
m
p
o
r
al
in
f
o
r
m
atio
n
o
f
d
o
cu
m
e
n
t
f
r
eq
u
en
cie
s
.
I
n
ad
d
itio
n
to
m
o
d
if
i
ed
d
o
cu
m
en
t
f
r
eq
u
en
c
y
,
B
o
k
et
a
l.
[
2
1
]
s
ca
les
t
h
e
ter
m
f
r
eq
u
en
c
y
o
f
t
h
e
w
o
r
d
s
i
n
to
lo
g
ar
it
h
m
ic
s
ca
le.
T
h
e
lo
g
ar
ith
m
ic
s
ca
l
in
g
o
f
ter
m
f
r
eq
u
en
c
y
is
al
s
o
d
o
n
e
i
n
th
e
co
m
p
ar
ativ
e
r
esear
ch
b
y
P
is
k
o
r
s
k
i
an
d
J
ac
q
u
et
[
2
2
]
.
T
h
e
co
m
p
ar
is
o
n
is
co
n
d
u
c
ted
b
et
w
ee
n
lo
g
-
s
ca
led
TF
-
I
DF
ch
ar
ac
ter
N
-
g
r
a
m
s
an
d
w
o
r
d
em
b
ed
d
in
g
f
o
r
f
in
e
-
g
r
ain
ed
class
i
f
ica
tio
n
tas
k
s
h
o
w
s
t
h
at
lo
g
-
s
ca
led
TF
-
I
DF a
p
p
r
o
ac
h
o
u
tp
er
f
o
r
m
w
o
r
d
e
m
b
ed
d
in
g
ap
p
r
o
ac
h
in
m
o
s
t ta
s
k
s
.
I
m
b
alan
ce
d
d
ataset
h
a
n
d
li
n
g
s
ar
e
d
o
n
e
in
s
e
v
er
al
p
r
ev
io
u
s
r
esear
ch
es.
I
s
h
aq
et
a
l.
[
1
5
]
co
m
b
in
e
o
v
er
s
a
m
p
li
n
g
tec
h
n
iq
u
e
w
i
th
s
ev
er
al
d
ata
m
i
n
i
n
g
tec
h
n
iq
u
e
s
to
i
m
p
r
o
v
e
t
h
e
p
r
ed
ictio
n
o
f
h
ea
r
t
f
ail
u
r
e
ca
s
e.
T
h
is
r
esear
ch
e
m
p
lo
y
s
SMOT
E
to
o
v
er
s
a
m
p
le
th
e
m
i
n
o
r
it
y
class
w
h
ich
is
t
h
e
m
o
r
talit
y
c
ase.
T
h
e
co
n
d
u
cted
r
esear
ch
also
s
h
o
w
s
t
h
at
r
a
n
d
o
m
f
o
r
est
clas
s
if
ier
y
ield
s
th
e
m
o
s
t
p
r
o
m
is
i
n
g
r
es
u
lts
b
ased
o
n
s
ev
er
al
ev
alu
a
tio
n
s
.
I
n
n
et
w
o
r
k
attac
k
,
Z
u
ec
h
et
a
l.
[
1
6
]
ex
p
lo
r
e
s
th
e
s
a
m
p
l
in
g
m
et
h
o
d
s
b
y
u
n
d
er
s
a
m
p
li
n
g
t
h
e
m
aj
o
r
ity
c
lass
.
T
h
i
s
r
esear
ch
also
s
h
o
w
s
t
h
at
r
an
d
o
m
f
o
r
es
t
class
i
f
ier
o
u
tp
er
f
o
r
m
s
m
o
s
t
o
f
th
e
c
lass
if
ier
s
.
An
o
th
er
r
esear
ch
b
y
Os
k
o
u
ei
an
d
B
ig
h
a
m
[
2
3
]
ex
p
er
im
e
n
t
s
o
v
er
s
a
m
p
lin
g
an
d
u
n
d
er
s
a
m
p
lin
g
tech
n
iq
u
es
i
n
ex
tr
e
m
e
l
y
i
m
b
ala
n
ce
d
d
ataset
.
T
h
e
ex
p
lo
r
ed
d
ataset
s
co
n
s
is
t
o
f
1
3
s
tan
d
ar
d
r
ea
l
d
atasets
f
r
o
m
o
p
en
-
s
o
u
r
ce
r
ep
o
s
ito
r
y
.
T
h
e
r
esear
ch
s
h
o
w
s
t
h
at
in
i
m
b
ala
n
ce
d
p
r
o
b
l
e
m
r
esa
m
p
li
n
g
m
eth
o
d
is
cr
u
cial,
an
d
is
m
o
r
e
p
r
ef
er
r
ed
th
an
e
x
p
lo
r
in
g
t
h
e
in
f
l
u
en
ce
o
f
t
h
e
cla
s
s
i
f
ier
.
T
h
e
r
esear
ch
also
co
n
cl
u
d
es
th
at
o
v
er
s
a
m
p
l
in
g
m
et
h
o
d
s
o
u
tp
er
f
o
r
m
i
n
all
ca
s
es c
o
m
p
ar
ed
to
u
n
d
er
s
a
m
p
li
n
g
m
eth
o
d
s
.
As
s
tated
in
p
r
ev
io
u
s
p
ar
ag
r
ap
h
,
r
an
d
o
m
f
o
r
est
class
if
ier
y
ie
ld
s
s
u
f
f
icie
n
tl
y
w
ell
p
er
f
o
r
m
a
n
ce
in
class
i
f
icatio
n
,
in
cl
u
d
in
g
t
h
e
p
r
o
b
lem
w
it
h
i
m
b
ala
n
ce
d
d
ataset
[
1
5
]
,
[
1
6
]
.
T
r
iay
u
d
i
a
n
d
Fit
r
i
[
2
4
]
ex
p
lo
r
es
v
ar
io
u
s
m
eth
o
d
o
f
class
i
f
icati
o
n
s
in
ed
u
ca
tio
n
a
l
d
ata
m
i
n
i
n
g
.
E
v
en
th
o
u
g
h
it
is
n
o
t
th
e
p
er
f
ec
t
r
esu
lt,
r
an
d
o
m
f
o
r
est
clas
s
i
f
ier
p
er
f
o
r
m
s
w
e
ll
in
m
aj
o
r
ity
tas
k
,
esp
ec
iall
y
in
m
o
d
elli
n
g
w
it
h
o
u
t
a
n
y
f
ea
t
u
r
e
s
elec
tio
n
s
.
I
n
an
o
th
er
i
m
b
ala
n
ce
d
ca
s
e,
Mo
h
a
m
m
ed
et
a
l
.
[
2
5
]
ex
p
er
im
e
n
t
s
s
e
v
er
al
m
o
d
els
o
f
clas
s
i
f
icatio
n
i
n
tr
an
s
ac
tio
n
s
d
ata.
T
h
e
d
ata
co
n
tain
s
i
m
m
en
s
e
n
u
m
b
er
o
f
co
lu
m
n
s
an
d
r
o
w
s
.
R
an
d
o
m
f
o
r
est
clas
s
i
f
ier
o
u
tp
er
f
o
r
m
s
all
o
th
er
m
o
d
els
in
o
v
er
s
a
m
p
li
n
g
tech
n
iq
u
e
,
an
d
f
alls
in
to
2
nd
p
o
s
i
tio
n
in
u
n
d
er
s
a
m
p
lin
g
Evaluation Warning : The document was created with Spire.PDF for Python.
384
I
SS
N:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
2
3
,
No
.
2
,
A
p
r
il 2
0
2
5
: 3
8
2
-
392
tech
n
iq
u
e.
T
h
ese
p
r
io
r
r
esear
ch
es
[
1
5
]
,
[
1
6
]
,
[
2
5
]
c
o
n
clu
d
es
th
at
r
an
d
o
m
f
o
r
est
cla
s
s
i
f
i
er
is
b
ef
itti
n
g
f
o
r
class
i
f
icatio
n
p
r
o
b
lem
w
i
th
i
m
b
alan
ce
d
p
r
o
b
lem
.
T
h
er
e
h
av
e
b
ee
n
s
ev
er
al
s
tu
d
ies
d
o
n
e
o
n
s
h
o
r
t
tex
t
m
i
n
i
n
g
.
B
er
n
ar
d
et
a
l.
[
2
6
]
ex
p
lo
r
e
th
e
clu
s
ter
in
g
m
et
h
o
d
f
o
r
tr
ac
k
in
g
n
e
w
s
s
to
r
ies
in
s
h
o
r
t
m
es
s
ag
in
g
i
n
C
o
v
id
-
1
9
ar
ea
.
T
h
is
r
esear
ch
u
til
izes
th
e
s
p
ar
s
e
T
F
-
I
DF
co
m
b
in
ed
w
it
h
T
r
an
s
f
o
r
m
er
as
t
h
e
v
ec
to
r
izatio
n
m
et
h
o
d
s
.
P
r
ev
io
u
s
m
et
h
o
d
b
y
Mir
an
d
a
et
a
l.
[
2
7
]
w
as
u
s
ed
i
n
t
h
is
r
esear
c
h
[
2
6
]
,
w
h
ich
u
s
es
s
u
p
er
v
is
ed
cl
u
s
ter
i
n
g
f
r
o
m
m
o
n
o
li
n
g
u
al
an
d
cr
o
s
s
l
in
g
u
al
ap
p
r
o
ac
h
es.
B
esid
es
th
at,
u
n
s
u
p
er
v
is
ed
K
-
m
ea
n
s
w
as
a
ls
o
u
tili
ze
d
an
d
co
m
b
in
ed
w
it
h
p
r
io
r
r
esear
ch
[
2
7
]
.
T
h
e
r
esu
lt
s
h
o
w
s
th
a
t T
F
-
I
DF i
s
s
till
r
o
b
u
s
t
f
o
r
d
o
in
g
m
u
ltip
le
s
h
o
r
t te
x
t
m
i
n
i
n
g
,
e
v
e
n
w
h
en
is
co
m
b
in
ed
w
it
h
o
t
h
er
t
y
p
e
o
f
v
ec
to
r
izatio
n
.
I
n
an
o
t
h
er
r
esear
ch
b
y
Ma
r
i
v
ate
a
n
d
Sef
ar
a
[
2
8
]
co
n
d
u
cted
tex
t
clas
s
i
f
icatio
n
in
m
u
ltip
le
task
s
.
T
h
e
r
esear
ch
u
tili
ze
s
g
lo
b
al
au
g
m
e
n
tatio
n
m
et
h
o
d
w
h
ich
u
s
es
s
y
n
o
n
y
m
a
u
g
m
en
tatio
n
,
s
e
m
a
n
tic
s
i
m
ilar
i
t
y
a
u
g
m
e
n
tatio
n
,
an
d
r
o
u
n
d
-
tr
ip
tr
an
s
latio
n
.
A
l
th
o
u
g
h
th
e
lo
s
s
i
s
r
ed
u
ce
d
w
h
en
t
h
e
g
lo
b
al
au
g
m
e
n
tatio
n
is
e
m
p
lo
y
ed
,
t
h
e
r
esu
lt
s
h
o
w
s
t
h
at
t
h
e
r
ed
u
ctio
n
o
f
lo
s
s
w
it
h
g
lo
b
al
au
g
m
e
n
tatio
n
is
n
o
t
s
ig
n
i
f
ica
n
t.
Mo
r
eo
v
er
,
th
is
r
e
s
ea
r
ch
e
x
p
lo
r
es
t
h
e
m
et
h
o
d
in
E
n
g
l
is
h
w
h
ic
h
h
as
lar
g
e
r
es
o
u
r
ce
s
an
d
co
r
p
o
r
a.
T
h
e
p
r
o
p
o
s
ed
m
eth
o
d
o
f
t
h
is
r
esear
ch
m
a
y
n
o
t n
ec
es
s
ar
il
y
b
e
ap
p
licab
le
in
o
th
er
lan
g
u
a
g
e
s
.
I
n
B
ah
asa
I
n
d
o
n
esia
te
x
t,
S
etiab
u
d
i
et
a
l.
[
29
]
ex
p
lo
r
es
th
e
e
f
f
ec
t
m
is
s
p
elled
w
o
r
d
in
B
ah
a
s
a
I
n
d
o
n
esia
’
s
te
x
t
class
if
icatio
n
.
L
ev
e
n
s
h
tei
n
d
is
ta
n
ce
is
e
m
p
l
o
y
ed
to
f
i
x
th
e
m
is
s
p
elled
w
o
r
d
.
T
h
e
m
i
s
s
p
elled
co
r
r
ec
tio
n
its
el
f
i
s
co
n
d
u
cted
b
ef
o
r
e
th
e
m
o
d
el
p
er
f
o
r
m
s
a
s
p
r
ep
r
o
ce
s
s
in
g
.
T
h
e
r
esu
l
t
s
h
o
w
s
t
h
at
w
it
h
t
h
e
m
is
s
p
elled
co
r
r
ec
tio
n
w
it
h
t
h
e
Naïv
e
B
a
y
es
m
o
d
el
o
u
tp
er
f
o
r
m
th
e
b
aseli
n
e
m
o
d
el
b
y
8
.
2
%.
Ho
w
e
v
er
,
t
h
is
r
esear
ch
also
s
h
o
w
s
t
h
at
th
e
ad
d
itio
n
o
f
th
is
p
r
ep
r
o
ce
s
s
in
g
ad
d
s
th
e
co
m
p
le
x
it
y
a
n
d
elap
s
ed
ti
m
e
o
f
th
e
m
o
d
el.
San
to
s
o
et
a
l.
[
30
]
wo
r
k
w
it
h
s
e
n
ti
m
e
n
t
a
n
al
y
s
i
s
a
n
d
h
o
a
x
cla
s
s
i
f
icatio
n
i
n
B
ah
asa
I
n
d
o
n
e
s
ia.
T
h
e
s
tu
d
y
s
u
g
g
est
s
u
s
i
n
g
p
ar
ticle
s
w
ar
m
o
p
ti
m
izatio
n
(
P
SO)
to
in
cr
ea
s
e
Naï
v
e
B
a
y
es
’
ac
cu
r
a
c
y
.
B
o
th
r
esear
ch
es
b
y
Se
tiab
u
d
i
et
a
l.
[
2
9
]
an
d
San
to
s
o
et
a
l.
[
3
0
]
f
u
r
th
er
p
r
o
v
e
th
at
B
ah
asa
I
n
d
o
n
e
s
ia
’
s
r
eso
u
r
ce
s
f
o
r
tex
t
m
i
n
in
g
a
n
d
class
i
f
icatio
n
ar
e
lack
in
g
.
3.
M
E
T
H
O
D
T
h
is
s
ec
tio
n
i
s
d
iv
id
ed
in
t
o
s
ev
er
al
s
u
b
s
ec
tio
n
s
:
r
esea
r
ch
m
et
h
o
d
o
lo
g
y
,
d
ataset,
s
ce
n
ar
io
o
f
ex
p
er
i
m
e
n
t,
an
d
ev
a
lu
atio
n
m
etr
ics.
E
ac
h
s
u
b
s
ec
tio
n
w
i
ll b
e
ex
p
lain
ed
f
u
r
t
h
er
.
3
.
1
.
Resea
rc
h
m
et
ho
do
lo
g
y
T
h
e
f
lo
w
ch
ar
t
in
Fi
g
u
r
e
1
b
r
i
ef
l
y
d
escr
ib
es
h
o
w
th
i
s
r
esear
ch
is
ac
co
m
p
li
s
h
ed
.
T
h
e
r
esear
ch
b
eg
in
s
w
it
h
th
e
co
llected
d
ataset.
T
h
e
d
ataset
w
il
l
b
e
ex
p
lain
ed
f
u
r
th
er
in
s
u
b
s
ec
tio
n
3
.
2
.
B
ef
o
r
e
u
n
d
er
g
o
an
y
p
r
o
ce
s
s
,
th
e
d
ataset
is
th
e
n
p
r
ep
r
o
ce
s
s
ed
u
s
i
n
g
u
s
u
al
s
tan
d
ar
d
p
r
ep
o
ce
s
s
f
o
r
tex
t
m
i
n
in
g
,
w
h
ic
h
ar
e
t
o
k
en
izat
io
n
,
ca
s
e
-
f
o
ld
in
g
,
an
d
s
tem
m
i
n
g
[
3
1
]
.
T
h
e
d
ata
t
h
en
is
s
p
litt
ed
in
to
t
w
o
p
ar
ts
;
tr
ain
in
g
d
ata
an
d
test
i
n
g
d
ata.
T
h
e
tr
ain
i
n
g
d
ata
w
ill
b
e
t
h
e
b
ase
o
f
t
h
e
T
F
-
I
DF
v
ec
to
r
izatio
n
,
a
n
d
r
esu
lt
s
in
b
ag
-
of
-
w
o
r
d
s
.
T
h
e
b
ag
-
of
-
w
o
r
d
s
is
u
s
ed
as
v
ec
t
o
r
izer
to
t
r
an
s
f
o
r
m
b
o
th
tr
ai
n
in
g
d
ata
an
d
tes
tin
g
d
ata.
On
ce
all
t
h
e
d
ata
is
tr
an
s
f
o
r
m
ed
in
to
v
ec
to
r
u
s
i
n
g
b
ag
-
of
-
w
o
r
d
s
,
th
e
tr
ai
n
in
g
d
ata
is
ap
p
lied
in
to
th
e
m
o
d
el.
I
n
s
p
ir
ed
b
y
p
r
io
r
r
esear
ch
es
[
1
5
]
,
[
1
6
]
,
[
2
5
]
,
th
e
r
an
d
o
m
f
o
r
est
al
g
o
r
ith
m
is
e
m
p
lo
y
ed
to
y
ield
t
h
e
b
etter
r
esu
lt.
T
h
e
te
s
ti
n
g
p
r
o
ce
s
s
is
d
o
n
e
af
ter
th
e
r
an
d
o
m
f
o
r
est
m
o
d
el
is
b
u
ild
,
en
g
ag
in
g
th
e
tes
tin
g
d
ata
as
th
e
b
en
ch
m
ar
k
f
o
r
th
e
p
r
ed
ictio
n
r
esu
lt.
T
h
e
ev
alu
atio
n
m
etr
ic
s
w
ill b
e
u
s
ed
to
co
m
p
ar
e
th
e
p
er
f
o
r
m
a
n
ce
o
f
ea
c
h
ex
p
er
i
m
en
t.
I
n
Fi
g
u
r
e
1
,
th
e
bol
d
ed
b
lo
ck
s
ar
e
th
e
p
r
o
ce
s
s
es
t
h
at
w
il
l
b
e
ex
p
er
i
m
en
ted
w
it
h
m
u
ltip
le
s
ce
n
ar
io
.
T
h
e
d
etail
o
f
e
x
p
er
i
m
e
n
t
s
ce
n
ar
io
s
a
n
d
th
e
ev
a
lu
at
io
n
m
etr
ics ar
e
r
esp
ec
ti
v
el
y
el
u
cid
ated
in
s
u
b
s
ec
tio
n
s
3
.
3
an
d
3
.
4
.
3
.
2
.
Da
t
a
s
et
T
h
e
d
ataset
u
s
ed
is
th
e
titl
e
o
f
th
e
f
in
al
as
s
i
g
n
m
en
t
o
f
I
n
f
o
r
m
atio
n
T
ec
h
n
o
lo
g
y
s
t
u
d
en
t
s
at
Un
i
v
er
s
ita
s
Ud
a
y
a
n
a
.
T
h
e
class
i
f
ied
class
o
f
ea
ch
as
s
i
g
n
m
e
n
t
is
t
h
e
to
p
ic
o
f
co
r
r
esp
o
n
d
in
g
ass
i
g
n
m
en
t.
T
h
is
d
ataset
is
d
escr
ib
ed
s
i
m
p
l
y
in
T
ab
le
1
.
T
h
is
d
ataset
is
u
s
ed
b
ec
au
s
e
o
f
th
e
f
o
llo
w
i
n
g
r
ea
s
o
n
s
:
−
T
h
e
d
ataset
is
p
r
esen
ted
in
f
o
r
m
al
I
n
d
o
n
e
s
ian
la
n
g
u
a
g
e
.
−
T
h
e
d
ataset
titl
e
o
f
ea
c
h
as
s
ig
n
m
en
t,
co
n
tai
n
s
a
r
elati
v
el
y
s
h
o
r
t
n
u
m
b
er
o
f
w
o
r
d
s
co
m
p
ar
ed
to
p
ar
ag
r
ap
h
s
o
r
ab
s
tr
ac
ts
o
f
th
e
ass
i
g
n
m
e
n
ts
.
T
h
is
s
u
p
p
o
r
ts
th
e
s
en
te
n
ce
lev
e
l c
lass
if
ica
tio
n
.
−
T
h
e
d
ataset
h
a
s
m
an
y
c
lass
e
s
,
b
u
t
t
h
er
e
ar
e
s
o
m
e
cla
s
s
es
t
h
at
h
av
e
a
v
er
y
s
m
all
n
u
m
b
e
r
o
f
i
n
s
ta
n
ce
s
co
m
p
ar
ed
to
o
th
er
class
e
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
385
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
Ove
r
s
a
mp
lin
g
vs.
u
n
d
ers
a
mp
lin
g
in
TF
-
I
DF
v
a
r
ia
tio
n
s
fo
r
imb
a
la
n
ce
d
…
(
I
N
yo
ma
n
P
r
a
y
a
n
a
Tr
is
n
a
)
Fig
u
r
e
1
.
T
h
e
r
esear
ch
m
e
th
o
d
ex
p
lain
ed
in
f
lo
w
c
h
ar
t
T
ab
le
1
.
Su
m
m
ar
y
o
f
t
h
e
d
atas
et
T
o
p
i
c
C
o
d
e
N
u
mb
e
r
o
f
a
ss
i
g
n
me
n
t
s
D
i
g
i
t
a
l
B
a
l
i
t
o
u
r
i
sm
D
B
T
10
D
i
g
i
t
a
l
i
mag
i
n
g
sy
st
e
m
D
I
S
12
D
a
t
a
sci
e
n
c
e
DS
52
D
i
g
i
t
a
l
e
c
o
n
o
my
ED
14
I
n
t
e
r
n
e
t
o
f
t
h
i
n
g
I
O
T
18
I
n
f
o
r
mat
i
o
n
sy
st
e
m
IS
74
N
e
t
w
o
r
k
a
n
d
c
l
o
u
d
c
o
mp
u
t
i
n
g
N
C
C
11
I
T
g
o
v
e
r
n
a
n
c
e
T
K
T
I
9
T
o
t
a
l
2
0
0
3
.
3
.
Scena
rio
o
f
ex
peri
m
e
nt
As
p
ictu
r
ed
in
Fig
u
r
e
1
,
th
e
b
o
ld
e
d
b
l
o
ck
s
ar
e
th
e
p
r
o
ce
s
s
w
h
ic
h
w
ill
b
e
ex
p
er
i
m
e
n
t
ed
in
th
is
r
esear
ch
.
T
h
u
s
,
t
h
e
T
F
-
I
DF
v
ec
to
r
izatio
n
m
et
h
o
d
s
an
d
t
h
e
s
a
m
p
li
n
g
m
e
th
o
d
s
ar
e
t
h
e
p
ar
am
eter
s
o
f
t
h
e
ex
p
er
i
m
e
n
t.
T
ab
le
2
ex
p
lain
s
th
e
p
ar
a
m
eter
s
a
n
d
th
e
h
o
w
t
h
o
s
e
p
ar
am
e
ter
s
co
n
tr
ib
u
te
in
ea
ch
ex
p
er
i
m
en
t.
Evaluation Warning : The document was created with Spire.PDF for Python.
386
I
SS
N:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
2
3
,
No
.
2
,
A
p
r
il 2
0
2
5
: 3
8
2
-
392
T
ab
le
2
.
E
x
p
er
im
e
n
ted
s
ce
n
ar
i
o
s
S
c
e
n
a
r
i
o
TF
-
I
D
F
u
se
d
S
a
mp
l
i
n
g
me
t
h
o
d
S
c
e
n
a
r
i
o
0
S
t
a
n
d
a
r
d
T
F
-
I
D
F
N
o
samp
l
i
n
g
S
c
e
n
a
r
i
o
1
S
t
a
n
d
a
r
d
T
F
-
I
D
F
O
v
e
r
samp
l
i
n
g
S
c
e
n
a
r
i
o
2
L
o
g
-
sca
l
e
d
TF
-
I
D
F
O
v
e
r
samp
l
i
n
g
S
c
e
n
a
r
i
o
3
B
o
o
l
e
a
n
TF
-
I
D
F
O
v
e
r
samp
l
i
n
g
S
c
e
n
a
r
i
o
4
S
t
a
n
d
a
r
d
T
F
-
I
D
F
U
n
d
e
r
sam
p
l
i
n
g
S
c
e
n
a
r
i
o
5
L
o
g
-
s
c
a
l
e
d
TF
-
I
D
F
U
n
d
e
r
sam
p
l
i
n
g
S
c
e
n
a
r
i
o
6
B
o
o
l
e
a
n
TF
-
I
D
F
U
n
d
e
r
sam
p
l
i
n
g
Scen
ar
io
0
in
T
ab
le
2
is
u
s
ed
as
b
aselin
e
f
o
r
th
e
m
o
d
el.
I
n
th
ese
s
ce
n
ar
io
s
,
lo
g
-
s
ca
led
a
n
d
b
o
o
lean
m
o
d
i
f
icat
io
n
o
f
T
F
-
I
DF
ar
e
in
tr
o
d
u
ce
d
[
3
2
]
.
T
h
e
s
tan
d
ar
d
T
F
-
I
DF
is
d
ef
i
n
ed
i
n
(
1
)
,
w
h
er
e
th
e
lo
g
-
s
ca
led
T
F
-
I
DF is f
o
r
m
u
la
ted
in
(
2
)
,
an
d
b
o
o
lean
T
F
-
I
DF is in
(
3
)
.
,
=
,
×
(
)
(
1
)
−
,
=
(
1
+
(
,
)
)
×
(
)
(
2
)
−
,
=
{
(
)
,
,
>
0
0
,
ℎ
(
3
)
tf
t,
d
its
elf
r
ep
r
esen
t
s
th
e
f
r
eq
u
en
c
y
o
f
ter
m
t
i
n
d
o
cu
m
e
n
t
d
,
an
d
df
t
r
ep
r
esen
ts
n
u
m
b
er
o
f
d
o
cu
m
en
ts
th
a
t
co
n
tain
ter
m
t
.
T
h
e
to
tal
d
o
cu
m
en
t
i
n
t
h
e
co
llect
io
n
i
s
s
y
m
b
o
lized
w
it
h
N
.
I
n
s
u
ch
,
tfid
f
t,
d
r
ep
r
esen
ts
th
e
T
F
-
I
DF v
al
u
e
o
f
ter
m
t
i
n
d
o
cu
m
e
n
t
d
.
C
o
m
b
i
n
ed
w
i
th
T
F
-
I
DF
m
o
d
if
icatio
n
s
,
s
a
m
p
li
n
g
m
et
h
o
d
s
ar
e
also
ex
p
er
im
e
n
ted
.
B
o
th
o
f
o
v
er
s
a
m
p
li
n
g
an
d
u
n
d
er
s
a
m
p
l
in
g
ar
e
co
n
d
u
c
ted
in
d
if
f
er
en
t
s
ce
n
ar
io
s
.
T
h
e
o
v
er
s
a
m
p
li
n
g
m
eth
o
d
is
ca
r
r
ied
o
u
t
b
y
ad
d
i
n
g
n
e
w
i
n
s
ta
n
ce
s
to
m
i
n
o
r
it
y
class
e
s
s
o
th
at
th
o
s
e
class
es
h
a
v
e
th
e
s
a
m
e
n
u
m
b
er
o
f
in
s
tan
ce
s
w
it
h
th
e
m
aj
o
r
it
y
clas
s
.
I
n
co
n
tr
a
s
t,
u
n
d
er
s
a
m
p
li
n
g
m
et
h
o
d
cu
ts
th
e
n
u
m
b
er
o
f
i
n
s
ta
n
ce
s
i
n
m
aj
o
r
it
y
clas
s
es,
r
esu
lti
n
g
th
e
m
aj
o
r
it
y
an
d
m
i
n
o
r
ity
cla
s
s
e
s
h
a
v
e
th
e
s
a
m
e
to
tal
o
f
in
s
tan
ce
s
[
3
3
].
3
.
4
.
E
v
a
lua
t
i
o
n
m
et
rics
T
h
e
ev
alu
atio
n
w
il
l
b
e
co
n
clu
d
ed
in
ea
ch
s
ce
n
ar
io
in
T
ab
le
2
.
P
r
ec
is
io
n
,
r
ec
all,
an
d
f
-
m
e
asu
r
e
ar
e
u
s
ed
to
ev
alu
ate
t
h
e
ex
p
er
i
m
en
ts
.
P
r
ec
is
io
n
an
d
r
ec
all
ar
e
m
o
r
e
f
a
v
o
r
ab
le
in
i
m
b
alan
c
ed
d
ataset
f
o
r
th
eir
ab
ilit
ies
to
elab
o
r
ate
t
h
e
m
o
d
el
p
er
f
o
r
m
a
n
ce
i
n
s
p
ec
i
f
ic
c
lass
,
r
ath
er
t
h
an
o
v
er
all
d
ata
s
et
w
i
th
all
clas
s
es.
T
h
i
s
is
th
e
o
p
p
o
s
ite
o
f
ac
c
u
r
ac
y
m
ea
s
u
r
e,
w
h
ic
h
e
v
alu
a
te
th
e
o
v
er
all
p
er
f
o
r
m
an
ce
o
f
m
o
d
el
.
A
cc
u
r
ac
y
te
n
d
s
to
m
ea
s
u
r
e
th
e
p
er
f
o
r
m
an
ce
o
f
t
h
e
m
o
d
el
b
y
t
h
e
m
aj
o
r
ity
cla
s
s
[
3
4
].
P
r
ec
is
io
n
is
d
ef
i
n
ed
as
r
atio
o
f
co
r
r
ec
t
p
r
ed
ictio
n
w
it
h
to
t
al
p
r
ed
ictio
n
,
w
h
er
ea
s
r
ec
all
is
r
atio
o
f
co
r
r
ec
t
p
r
ed
ictio
n
w
it
h
to
tal
o
f
ac
tu
al
clas
s
es.
B
o
th
p
r
ec
is
io
n
an
d
r
ec
all
ar
e
m
ea
s
u
r
ed
in
s
p
ec
if
ic
class
e
s
,
f
-
m
ea
s
u
r
e
co
m
b
i
n
es
b
o
th
o
f
p
r
ec
is
io
n
an
d
r
ec
all,
an
d
is
u
s
e
d
to
m
ea
s
u
r
e
in
s
p
ec
i
f
ic
clas
s
es
as
w
e
ll.
T
h
ey
ar
e
d
if
f
er
e
n
t
w
i
th
ac
cu
r
ac
y
w
h
ic
h
m
ea
s
u
r
es
i
n
o
v
er
all
clas
s
es.
In
(
4
)
to
(
6
)
s
h
o
w
th
e
f
o
r
m
u
la
o
f
p
r
ec
is
io
n
,
r
ec
all,
an
d
f
-
m
ea
s
u
r
e
r
esp
ec
tiv
el
y
,
wh
er
e
c
r
ep
r
esen
ts
s
p
ec
i
f
ic
clas
s
in
th
e
ca
s
e.
=
(
4
)
=
(
5
)
1
−
=
2
×
×
+
(
6
)
A
ll
o
f
th
e
m
etr
ics
i
n
(
4
)
to
(
6
)
w
i
ll
b
e
s
u
m
m
ar
ized
in
w
e
ig
h
ted
av
er
ag
e.
T
h
e
w
ei
g
h
te
d
av
er
ag
e
a
cc
o
u
n
t
s
th
e
n
u
m
b
er
o
f
clas
s
e
s
in
t
h
e
test
i
n
g
d
ata.
In
(
7
)
ex
p
lain
s
t
h
e
ca
lc
u
latio
n
o
f
w
ei
g
h
t
ed
av
er
ag
e
f
u
r
th
er
.
(
)
=
∑
×
=
1
∑
=
1
(
7
)
W
A
(
m)
r
e
p
r
es
en
ts
th
e
w
e
ig
h
t
ed
a
v
e
r
ag
e
o
f
m
et
r
i
c
m
.
M
et
r
i
c
m
c
an
b
e
ei
th
e
r
p
r
e
c
is
i
o
n
,
r
e
ca
l
l
,
o
f
f
-
m
e
a
s
u
r
e
.
m
c
r
e
p
r
e
s
en
t
s
th
e
m
e
asu
r
em
en
t
o
f
m
e
t
r
i
c
m
i
n
c
l
as
s
c
,
a
n
d
s
c
r
e
p
r
e
s
en
ts
t
o
t
al
m
em
b
e
r
o
f
c
l
as
s
c
in
t
es
ti
n
g
d
at
a
.
Evaluation Warning : The document was created with Spire.PDF for Python.
387
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
Ove
r
s
a
mp
lin
g
vs.
u
n
d
ers
a
mp
lin
g
in
TF
-
I
DF
v
a
r
ia
tio
n
s
fo
r
imb
a
la
n
ce
d
…
(
I
N
yo
ma
n
P
r
a
y
a
n
a
Tr
is
n
a
)
4.
RE
SU
L
T
AND
DI
SCUS
SI
O
N
Ou
r
ex
p
er
i
m
e
n
t
s
w
ith
v
ar
io
u
s
p
ar
am
eter
s
as
i
n
T
ab
le
2
y
iel
d
co
m
p
li
y
in
g
r
es
u
lts
.
T
ab
le
3
illu
s
tr
ates
in
b
r
ief
ab
o
u
t
th
e
r
esu
lts
o
f
o
u
r
ex
p
er
i
m
en
t
s
.
T
h
e
n
u
m
b
er
s
in
T
ab
le
3
ar
e
th
e
w
e
ig
h
ted
a
v
er
ag
e
a
s
ex
p
lai
n
ed
in
(
7
)
.
T
h
e
d
etailed
r
esu
lt
o
f
ea
ch
ex
p
er
i
m
en
t
p
r
o
j
ec
ted
in
th
e
co
n
f
u
s
io
n
m
atr
i
x
.
T
ab
l
e
s
4
to
6
s
h
o
w
th
e
co
n
f
u
s
io
n
m
atr
i
x
f
o
r
ea
ch
s
ce
n
ar
io
.
Fro
m
t
h
e
T
ab
le
3
,
it
ca
n
b
e
co
n
clu
d
ed
th
a
t
s
ce
n
ar
io
3
w
i
th
o
v
er
s
a
m
p
led
lo
g
-
s
ca
led
T
F
-
I
DF
r
esu
lts
th
e
b
est
s
co
r
es
o
f
all
m
e
tr
ics.
T
h
e
b
aselin
e
m
o
d
el
i
n
s
ce
n
a
r
io
0
–
u
s
in
g
s
ta
n
d
ar
d
T
F
-
I
DF
w
ith
n
o
s
a
m
p
lin
g
,
alr
ea
d
y
s
h
o
w
s
e
x
ce
lle
n
t
p
er
f
o
r
m
an
ce
.
I
ts
a
v
er
ag
e
p
r
ec
is
io
n
,
r
ec
all,
an
d
f
-
m
ea
s
u
r
e
ar
e
ex
c
ee
d
in
g
8
0
%.
A
ll
o
f
th
e
s
ce
n
ar
io
s
w
ith
t
h
e
o
v
er
s
a
m
p
lin
g
m
eth
o
d
(
s
ce
n
ar
io
1
-
3
)
p
r
o
v
id
e
b
etter
o
u
tco
m
e
s
th
a
n
b
aselin
e
s
ce
n
ar
io
.
On
th
e
o
t
h
er
h
an
d
,
s
ce
n
ar
io
4
-
6
,
w
h
ic
h
e
m
p
lo
y
t
h
e
u
n
d
er
s
a
m
p
lin
g
s
tr
ate
g
y
,
co
n
s
id
er
ab
l
y
u
n
d
er
p
er
f
o
r
m
th
e
b
aselin
e
m
o
d
el
i
n
all
cr
iter
ia.
T
ab
le
3
.
R
esu
lt o
f
ea
c
h
s
ce
n
ar
io
s
in
s
u
m
m
ar
y
S
c
e
n
a
r
i
o
A
v
g
.
p
r
e
c
i
si
o
n
(
%)
A
v
g
.
r
e
c
a
l
l
(
%)
A
v
g
.
f
-
me
a
su
r
e
(
%)
S
c
e
n
a
r
i
o
0
8
0
.
9
6
5
8
5
.
0
0
0
8
1
.
9
0
0
S
c
e
n
a
r
i
o
1
8
1
.
9
4
3
8
7
.
5
0
0
8
3
.
9
5
5
S
c
e
n
a
r
i
o
2
8
2
.
5
0
0
8
7
.
5
0
0
8
4
.
1
5
2
S
c
e
n
a
r
i
o
3
9
1
.
3
7
3
9
0
.
0
0
0
8
8
.
7
6
4
S
c
e
n
a
r
i
o
4
6
7
.
9
0
5
5
7
.
5
0
0
5
3
.
5
8
3
S
c
e
n
a
r
i
o
5
6
9
.
7
5
0
6
0
.
0
0
0
5
6
.
4
5
5
S
c
e
n
a
r
i
o
6
6
6
.
7
6
6
5
7
.
5
0
0
5
3
.
0
2
1
T
ab
le
4
.
C
o
n
f
u
s
io
n
m
atr
i
x
f
o
r
b
aselin
e
s
ce
n
ar
io
P
r
e
d
i
c
t
e
d
l
a
b
e
l
D
B
T
D
I
S
DS
ED
I
O
T
IS
N
C
C
TK
TI
T
r
u
e
l
a
b
e
l
(
b
a
se
l
i
n
e
sce
n
a
r
i
o
)
D
B
T
2
0
0
0
0
0
0
0
D
I
S
0
2
0
0
0
0
0
0
DS
0
0
9
0
0
1
0
0
ED
0
0
0
0
0
3
0
0
I
O
T
0
0
0
0
4
0
0
0
IS
1
0
0
0
0
14
0
0
N
C
C
0
0
0
0
0
0
2
0
TK
TI
0
0
0
0
0
1
0
1
T
ab
le
5
.
C
o
n
f
u
s
io
n
m
atr
i
x
es
f
o
r
s
ce
n
ar
io
s
w
ith
o
v
er
s
a
m
p
li
n
g
P
r
e
d
i
c
t
e
d
l
a
b
e
l
D
B
T
D
I
S
DS
ED
I
O
T
IS
N
C
C
TK
TI
T
r
u
e
l
a
b
e
l
(
sce
n
a
r
i
o
1
)
D
B
T
2
0
0
0
0
0
0
0
D
I
S
0
2
0
0
0
0
0
0
DS
0
0
10
0
0
0
0
0
ED
0
0
1
0
0
2
0
0
I
O
T
0
0
0
0
4
0
0
0
IS
1
0
0
0
0
14
0
0
N
C
C
0
0
0
0
0
0
2
0
TK
TI
0
0
0
0
0
1
0
1
T
r
u
e
l
a
b
e
l
(
sce
n
a
r
i
o
2
)
D
B
T
2
0
0
0
0
0
0
0
D
I
S
0
2
0
0
0
0
0
0
DS
0
0
10
0
0
0
0
0
ED
0
0
0
0
0
3
0
0
I
O
T
0
0
0
0
4
0
0
0
IS
1
0
0
0
0
14
0
0
N
C
C
0
0
0
0
0
0
2
0
TK
TI
0
0
0
0
0
1
0
1
T
r
u
e
l
a
b
e
l
(
sce
n
a
r
i
o
3
)
D
B
T
2
0
0
0
0
0
0
0
D
I
S
0
2
0
0
0
0
0
0
DS
0
0
10
0
0
0
0
0
ED
0
0
1
1
0
1
0
0
I
O
T
0
0
0
0
4
0
0
0
IS
1
0
0
0
0
14
0
0
N
C
C
0
0
0
0
0
0
2
0
TK
TI
0
0
0
0
0
1
0
1
Evaluation Warning : The document was created with Spire.PDF for Python.
388
I
SS
N:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
2
3
,
No
.
2
,
A
p
r
il 2
0
2
5
: 3
8
2
-
392
T
ab
le
6
.
C
o
n
f
u
s
io
n
m
atr
i
x
es
f
o
r
s
ce
n
ar
io
s
w
ith
u
n
d
er
s
a
m
p
li
n
g
P
r
e
d
i
c
t
e
d
l
a
b
e
l
D
B
T
D
I
S
DS
ED
I
O
T
IS
N
C
C
TK
TI
T
r
u
e
l
a
b
e
l
(
sce
n
a
r
i
o
4
)
D
B
T
2
0
0
0
0
0
0
0
D
I
S
0
2
0
0
0
0
0
0
DS
0
0
9
0
0
1
0
0
ED
1
0
0
2
0
0
0
0
I
O
T
0
0
0
0
4
0
0
0
IS
6
1
4
1
0
2
0
1
N
C
C
0
0
1
0
0
0
1
0
TK
TI
1
0
0
0
0
1
0
1
T
r
u
e
l
a
b
e
l
(
sce
n
a
r
i
o
5
)
D
B
T
2
0
0
0
0
0
0
0
D
I
S
0
2
0
0
0
0
0
0
DS
0
0
9
0
0
1
0
0
ED
1
0
0
2
0
0
0
0
I
O
T
0
0
0
0
4
0
0
0
IS
6
1
3
1
0
2
0
2
N
C
C
0
0
0
0
0
0
2
0
TK
TI
1
0
0
0
0
1
0
1
T
r
u
e
l
a
b
e
l
(
sce
n
a
r
i
o
6
)
D
B
T
2
0
0
0
0
0
0
0
D
I
S
0
2
0
0
0
0
0
0
DS
0
0
9
0
0
1
0
0
ED
1
0
0
2
0
0
0
0
I
O
T
0
0
0
0
4
0
0
0
IS
6
1
4
1
0
2
0
1
N
C
C
0
0
1
0
0
0
1
0
TK
TI
1
0
0
1
0
1
0
1
T
h
e
o
u
tco
m
e
o
f
r
eso
l
v
i
n
g
t
h
e
u
n
b
ala
n
ce
d
d
ataset
is
s
u
e
i
s
d
is
p
la
y
ed
in
T
ab
le
3
.
B
ased
o
n
o
u
r
co
m
p
leted
s
it
u
atio
n
s
,
th
e
o
v
er
s
a
m
p
li
n
g
m
et
h
o
d
s
g
r
ea
tl
y
o
u
t
p
er
f
o
r
m
t
h
e
u
n
d
er
s
a
m
p
li
n
g
m
eth
o
d
s
in
s
h
o
r
t
tex
t
class
i
f
icatio
n
.
C
o
m
p
ar
i
n
g
s
h
o
r
t
tex
t
clas
s
i
f
icatio
n
to
lo
n
g
te
x
t
clas
s
i
f
icatio
n
,
t
h
e
co
m
p
lete
d
b
ag
-
of
-
w
o
r
d
s
h
a
s
s
ig
n
i
f
ica
n
tl
y
f
e
w
er
te
x
t r
eso
u
r
ce
s
.
A
cc
o
r
d
in
g
to
De
B
o
o
m
et
a
l.
[
9
]
,
s
p
ar
s
e
v
ec
to
r
s
ar
e
p
r
o
d
u
ce
d
w
h
e
n
te
x
t
r
eso
u
r
ce
s
f
o
r
s
h
o
r
t
tex
t
class
i
f
icatio
n
ar
e
lim
ited
.
Un
d
er
s
a
m
p
li
n
g
m
ak
e
s
th
e
alr
ea
d
y
li
m
ited
r
eso
u
r
ce
s
ev
e
n
m
o
r
e
s
o
.
Scar
ce
r
eso
u
r
ce
s
p
r
o
d
u
ce
a
j
u
m
b
le
o
f
w
o
r
d
s
t
h
at
ar
e
u
n
ab
le
to
d
is
tin
g
u
is
h
b
et
w
ee
n
clas
s
i
f
icatio
n
cla
s
s
e
s
.
On
th
e
o
t
h
er
h
a
n
d
,
th
e
class
if
ier
m
o
d
el
m
a
y
m
o
r
e
ea
s
il
y
id
en
ti
f
y
th
e
cla
s
s
es
b
y
cr
ea
ti
n
g
a
s
y
n
t
h
esi
s
v
ec
to
r
w
it
h
o
v
er
s
a
m
p
li
n
g
s
in
ce
th
e
b
ag
-
of
-
w
o
r
d
s
v
ec
t
o
r
h
as
lar
g
er
d
i
m
e
n
s
io
n
s
.
W
ith
t
h
e
li
m
ited
d
ata
s
et
i
n
th
i
s
r
esear
c
h
,
th
e
u
n
d
er
s
a
m
p
li
n
g
ap
p
r
o
ac
h
is
m
o
r
e
lik
el
y
to
p
r
o
d
u
ce
s
p
ar
s
e
v
ec
to
r
,
th
u
s
r
es
u
lti
n
g
p
o
o
r
r
esu
lt
co
m
p
ar
ed
to
t
h
e
o
v
er
s
a
m
p
li
n
g
ap
p
r
o
ac
h
w
it
h
s
y
n
t
h
etic
v
ec
to
r
.
Desp
ite
p
r
o
d
u
cin
g
p
o
o
r
p
er
f
o
r
m
a
n
ce
,
t
h
is
r
esear
ch
d
e
m
o
n
s
tr
ates
th
at
u
n
d
er
s
a
m
p
l
in
g
ca
n
b
ea
t
b
aselin
e
m
o
d
els,
e
v
en
o
v
er
s
a
m
p
li
n
g
m
eth
o
d
s
,
esp
ec
iall
y
i
n
m
i
n
o
r
it
y
cla
s
s
es.
C
lass
E
D
w
ill
b
e
u
s
ed
as
an
illu
s
tr
atio
n
.
C
la
s
s
E
D
is
co
n
s
is
ten
tl
y
m
is
s
p
r
ed
icted
in
th
e
b
aselin
e
m
o
d
el
f
r
o
m
T
ab
le
4
,
an
d
n
o
ad
d
itio
n
al
class
es
ar
e
p
r
o
j
ec
ted
to
b
e
E
D,
th
u
s
m
a
k
i
n
g
t
h
e
p
r
ec
is
io
n
an
d
r
ec
all
r
es
u
lt
s
f
o
r
E
D
ar
e
ze
r
o
.
T
h
is
is
s
u
e
p
er
s
is
ts
ev
e
n
in
ca
s
es
o
f
o
v
er
s
a
m
p
li
n
g
,
w
it
h
o
n
l
y
o
n
e
ca
s
e
o
f
class
E
D
is
ac
cu
r
atel
y
p
r
ed
icted
b
y
s
ce
n
ar
io
3
(
b
est
s
ce
n
ar
io
)
.
T
h
e
u
n
d
er
s
a
m
p
lin
g
ap
p
r
o
ac
h
p
r
ev
en
ts
t
h
i
s
.
B
ased
o
n
T
ab
le
6
,
tw
o
o
f
t
h
e
th
r
ee
in
s
ta
n
ce
s
o
f
class
E
D
i
n
t
h
e
te
s
ti
n
g
d
ata
a
r
e
co
r
r
ec
tly
cla
s
s
i
f
ied
b
y
t
h
e
s
ce
n
ar
io
u
s
i
n
g
th
e
u
n
d
er
s
a
m
p
lin
g
m
eth
o
d
.
B
etter
r
ec
all
f
o
r
class
E
D
in
t
h
e
u
n
d
e
r
s
a
m
p
li
n
g
ap
p
r
o
ac
h
is
t
h
e
o
u
tc
o
m
e
o
f
th
is
.
C
las
s
E
D
h
as
th
e
f
i
f
t
h
-
lo
w
est
n
u
m
b
er
o
f
in
s
tan
ce
s
;
th
e
o
th
er
class
es
w
i
th
les
s
in
s
tan
ce
s
th
an
cla
s
s
E
D
ar
e
DB
T
,
DI
S,
NC
C
,
an
d
T
K
T
I
.
R
ec
alcu
lati
n
g
th
e
f
1
-
m
ea
s
u
r
e,
p
r
ec
is
io
n
,
an
d
r
ec
all
f
o
r
th
ese
f
i
v
e
clas
s
es
y
ield
s
t
h
e
f
o
llo
w
i
n
g
r
esu
lt
s
f
o
r
ea
ch
s
ce
n
ar
io
:
T
ab
le
7
.
B
as
ed
o
n
T
ab
le
7
,
s
ce
n
ar
io
s
4
-
6
w
it
h
u
n
d
er
s
a
m
p
li
n
g
d
em
o
n
s
tr
ate
s
t
h
e
s
a
m
e,
i
f
n
o
t
s
u
p
er
io
r
,
r
ec
all
f
o
r
m
in
o
r
it
y
class
cla
s
s
i
f
icatio
n
,
ev
e
n
t
h
o
u
g
h
s
ce
n
ar
io
3
s
till
p
er
f
o
r
m
s
b
est i
n
av
er
ag
e
p
r
ec
i
s
io
n
an
d
f
-
m
ea
s
u
r
e
f
o
r
m
i
n
o
r
ity
clas
s
es.
T
ab
le
7
.
R
esu
lt o
f
ea
c
h
s
ce
n
ar
io
s
(
DB
T
,
DI
S,
E
D,
NC
C
,
an
d
T
K
T
I
o
n
l
y
)
S
c
e
n
a
r
i
o
A
v
g
.
p
r
e
c
i
si
o
n
(
%)
A
v
g
.
r
e
c
a
l
l
(
%)
A
v
g
.
f
-
me
a
su
r
e
(
%)
S
c
e
n
a
r
i
o
0
6
6
.
6
6
7
6
3
.
6
3
6
6
3
.
0
3
0
S
c
e
n
a
r
i
o
1
6
6
.
6
6
7
6
3
.
6
3
6
6
3
.
0
3
0
S
c
e
n
a
r
i
o
2
6
6
.
6
6
7
6
3
.
6
3
6
6
3
.
0
3
0
S
c
e
n
a
r
i
o
3
9
3
.
9
3
9
7
2
.
7
2
7
7
6
.
6
6
7
S
c
e
n
a
r
i
o
4
6
1
.
2
1
2
7
2
.
7
2
7
6
0
.
0
0
0
S
c
e
n
a
r
i
o
5
5
8
.
1
8
2
8
1
.
8
1
8
6
4
.
2
4
2
S
c
e
n
a
r
i
o
6
5
7
.
0
7
1
7
2
.
7
2
7
5
7
.
9
5
4
Evaluation Warning : The document was created with Spire.PDF for Python.
389
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
Ove
r
s
a
mp
lin
g
vs.
u
n
d
ers
a
mp
lin
g
in
TF
-
I
DF
v
a
r
ia
tio
n
s
fo
r
imb
a
la
n
ce
d
…
(
I
N
yo
ma
n
P
r
a
y
a
n
a
Tr
is
n
a
)
A
cc
o
u
n
ti
n
g
o
n
l
y
m
i
n
o
r
it
y
cla
s
s
es
a
s
s
h
o
w
n
i
n
T
ab
le
7
in
d
icate
s
t
h
at
t
h
e
u
n
d
er
s
a
m
p
li
n
g
m
et
h
o
d
ca
n
s
till
p
er
f
o
r
m
w
ell
i
n
m
in
o
r
it
y
class
es
d
e
s
p
ite
o
v
er
all
p
er
f
o
r
m
an
ce
r
e
s
u
l
ts
t
h
at
ar
e
d
r
o
p
p
in
g
,
p
ar
ticu
lar
l
y
w
h
e
n
w
e
co
n
s
id
er
co
v
er
ag
e
o
f
tr
u
e
p
r
ed
ictio
n
as
th
e
p
r
i
m
ar
y
f
ac
t
o
r
.
Ho
w
e
v
er
,
th
e
tr
ad
e
-
o
f
f
o
f
th
is
ap
p
r
o
ac
h
w
it
h
u
n
d
er
s
a
m
p
li
n
g
m
et
h
o
d
is
th
e
d
ec
lin
e
p
er
f
o
r
m
a
n
ce
o
f
m
aj
o
r
ity
class
e
s
.
As
s
tated
in
T
a
b
le
6
,
w
i
th
t
h
e
escalatio
n
i
n
th
e
p
er
f
o
r
m
a
n
c
e
o
f
m
in
o
r
it
y
cla
s
s
e
s
s
u
c
h
a
s
class
E
D,
th
e
u
n
d
er
s
a
m
p
li
n
g
m
et
h
o
d
ten
d
s
to
n
eg
lec
t
t
h
e
m
aj
o
r
ity
s
u
ch
a
s
cl
ass
I
S.
As
i
n
d
icate
d
i
n
T
ab
le
3
,
th
is
r
e
s
u
l
ts
i
n
a
d
ec
li
n
e
i
n
t
h
e
o
v
er
all
o
u
tco
m
e
s
f
o
r
u
n
d
er
s
a
m
p
li
n
g
m
et
h
o
d
.
T
h
e
ex
p
er
i
m
en
t
r
es
u
lt
s
i
n
T
a
b
le
s
3
an
d
7
y
ie
ld
f
u
r
th
er
q
u
e
s
tio
n
:
w
h
ic
h
T
F
-
I
DF
m
o
d
i
f
ic
atio
n
is
t
h
e
b
est
f
o
r
eit
h
er
o
v
er
s
a
m
p
li
n
g
a
n
d
u
n
d
er
s
a
m
p
li
n
g
m
et
h
o
d
?
W
e
n
o
w
s
i
m
p
l
y
p
a
y
at
ten
t
io
n
to
th
e
o
v
er
s
a
m
p
li
n
g
m
et
h
o
d
’
s
o
u
tp
u
t,
w
h
ich
is
s
h
o
w
n
i
n
T
ab
le
5
.
Scen
ar
io
s
1
-
3
’
s
co
n
f
u
s
io
n
m
atr
ices
ar
e
es
s
en
tia
ll
y
t
h
e
s
a
m
e.
Scen
ar
io
s
1
-
3
a
r
e
id
en
tical
b
u
t
f
o
r
th
e
clas
s
E
D.
C
lass
E
D
i
s
o
n
l
y
ac
c
u
r
atel
y
clas
s
i
f
ied
i
n
S
ce
n
ar
io
3
,
an
d
th
a
t
is
o
n
l
y
in
o
n
e
o
u
t
o
f
th
r
ee
in
s
tan
ce
s
.
Fo
r
th
i
s
r
ea
s
o
n
,
ev
e
n
t
h
o
u
g
h
th
er
e
is
n
’
t
m
u
c
h
o
f
a
d
if
f
er
en
ce
b
et
w
ee
n
s
ce
n
ar
io
s
1
-
3
,
s
ce
n
ar
io
3
w
ith
B
o
o
lean
T
F
-
I
DF
is
th
e
b
e
s
t
s
c
en
ar
io
o
u
t
o
f
all
th
e
T
F
-
I
DF
m
o
d
i
f
icat
io
n
s
in
th
e
o
v
er
s
a
m
p
li
n
g
m
et
h
o
d
.
I
n
s
h
o
r
t
tex
t
d
o
cu
m
e
n
t
li
k
e
titl
e,
o
n
e
w
o
r
d
t
y
p
icall
y
o
cc
u
r
s
i
n
f
e
w
o
cc
u
r
e
n
ce
,
s
o
m
et
i
m
es
ev
e
n
al
m
o
s
t
o
n
ce
.
As
a
r
es
u
lt,
t
h
e
r
eg
u
lar
T
F
-
I
DF
(
as
i
n
s
ce
n
ar
io
1
)
w
ill
p
r
o
d
u
ce
a
v
ec
to
r
th
at
is
al
m
o
s
t
ex
clu
s
i
v
el
y
1
an
d
0
if
th
e
ter
m
s
d
o
n
o
t
o
cc
u
r
.
A
v
ec
to
r
t
h
at
co
n
tain
s
o
n
l
y
1
a
n
d
0
is
r
ef
er
r
ed
to
b
e
b
o
o
lean
v
ec
to
r
.
Scen
ar
io
3
is
ex
ec
u
te
d
u
s
i
n
g
th
e
w
o
r
d
’
s
b
o
o
lean
f
ea
tu
r
e;
i
f
t
h
e
ter
m
ex
is
t
s
,
its
o
cc
u
r
r
en
ce
w
ill
b
e
co
n
s
id
er
ed
as
T
r
u
e
(
o
r
1
)
an
d
v
ice
v
er
s
a
.
T
h
is
i
s
w
h
y
th
e
s
c
en
a
r
i
o
s
1
-
3
h
av
e
s
im
il
a
r
r
esu
l
t
as
s
t
a
te
d
in
T
a
b
l
e
5
,
w
it
h
s
ce
n
ar
io
3
h
a
s
s
li
g
h
t
s
u
p
er
io
r
ity
.
Desp
ite
B
o
o
lean
T
F
-
I
DF
v
ec
to
r
izatio
n
w
it
h
u
n
d
er
s
a
m
p
li
n
g
m
et
h
o
d
s
h
o
w
s
d
o
m
i
n
an
ce
i
n
p
er
f
o
r
m
an
ce
,
th
e
b
est
u
n
d
er
s
a
m
p
lin
g
m
e
t
h
o
d
d
o
esn
’
t
s
h
ar
e
th
e
s
i
m
ilar
v
ec
to
r
izatio
n
.
A
cc
o
r
d
in
g
to
T
ab
le
s
3
an
d
7
,
s
ce
n
ar
io
5
w
h
ic
h
m
ak
e
s
u
s
e
o
f
lo
g
-
s
ca
led
T
F
-
I
DF,
p
r
o
d
u
ce
s
th
e
b
est
r
esu
lts
w
h
e
n
u
s
in
g
t
h
e
u
n
d
er
s
a
m
p
li
n
g
m
eth
o
d
.
I
n
u
n
d
er
s
a
m
p
li
n
g
m
et
h
o
d
,
th
e
b
ag
-
of
-
w
o
r
d
d
i
m
en
s
io
n
i
s
m
u
ch
le
s
s
t
h
a
n
in
t
h
e
n
o
n
-
s
a
m
p
l
in
g
m
et
h
o
d
.
I
n
co
m
p
ar
is
o
n
to
s
t
an
d
ar
d
b
ag
-
of
-
w
o
r
d
s
f
r
o
m
n
o
n
-
s
a
m
p
li
n
g
ap
p
r
o
ac
h
,
th
e
s
m
aller
b
ag
-
of
-
w
o
r
d
s
f
r
o
m
u
n
d
er
s
a
m
p
li
n
g
m
eth
o
d
r
etu
r
n
s
e
v
en
m
o
r
e
s
p
ar
s
e
v
ec
t
o
r
.
L
o
g
-
s
ca
led
T
F
-
I
DF
ca
n
b
e
u
s
ed
to
s
o
lv
e
t
h
e
s
p
ar
s
e
v
ec
to
r
p
r
o
b
lem
.
T
h
is
is
d
u
e
to
th
e
f
ac
t
th
at
lo
g
-
s
ca
led
T
F
-
I
DF
al
w
a
y
s
y
ie
ld
s
n
o
n
-
ze
r
o
T
F
b
ased
o
n
(
2
)
.
T
h
e
o
cc
u
r
in
g
ter
m
s
w
ill
b
e
v
alu
ed
g
r
ea
ter
th
a
n
1
–
d
ep
en
d
i
n
g
o
n
t
h
e
lo
g
v
alu
e
o
f
th
e
f
r
eq
u
en
c
y
,
w
h
ile
t
h
e
n
o
n
-
o
cc
u
r
r
in
g
ter
m
s
w
il
l
b
e
v
alu
ed
at
1
(
n
o
t
0
)
.
T
h
e
ab
s
en
c
e
o
f
0
i
n
t
h
e
v
e
cto
r
r
es
u
lts
t
h
e
less
s
p
ar
s
e
v
ec
to
r
.
C
o
m
p
ar
ed
to
s
tan
d
ar
d
T
F
-
I
DF
o
r
b
o
o
lean
T
F
-
I
DF,
w
h
e
r
e
n
o
n
-
o
cc
u
r
r
in
g
ter
m
s
ar
e
v
alu
ed
at
0
,
th
is
i
s
s
ig
n
i
f
ica
n
tl
y
d
if
f
er
en
t
b
ec
au
s
e
th
e
y
w
o
u
ld
cr
ea
te
t
h
e
m
o
r
e
s
p
ar
s
e
v
ec
to
r
.
A
p
p
ar
en
tl
y
,
b
y
m
a
k
i
n
g
l
i
m
ited
r
eso
u
r
ce
s
ev
e
n
le
s
s
w
it
h
u
n
d
er
s
a
m
p
li
n
g
m
et
h
o
d
,
th
e
les
s
s
p
a
r
s
ed
v
ec
to
r
is
n
ee
d
ed
.
Fro
m
th
e
p
r
io
r
d
is
cu
s
s
io
n
,
th
e
ex
p
er
i
m
en
t
also
f
in
d
th
at
cla
s
s
E
D
is
th
e
m
o
s
t
m
is
c
lass
if
ie
d
class
in
m
o
s
t
s
ce
n
ar
io
s
,
w
h
er
e
cla
s
s
E
D
is
p
r
i
m
ar
i
l
y
m
i
s
clas
s
i
f
ied
as
class
I
S.
Me
an
w
h
ile
in
u
n
d
er
s
a
m
p
li
n
g
ap
p
r
o
ac
h
m
o
s
t
clas
s
es
ar
e
m
i
s
clas
s
i
f
ie
d
as
I
S.
T
h
is
ca
n
b
e
s
h
o
w
n
f
r
o
m
s
ce
n
ar
io
0
as
in
T
ab
le
4
,
s
ce
n
ar
io
s
w
ith
a
n
o
v
er
s
a
m
p
li
n
g
m
et
h
o
d
as in
T
ab
le
5
,
an
d
s
ce
n
ar
io
s
w
i
th
u
n
d
e
r
s
a
m
p
li
n
g
a
s
in
T
ab
le
6
.
T
h
e
last
ass
i
g
n
m
e
n
t
to
p
ic,
d
i
g
ital
ec
o
n
o
m
y
(
E
D)
in
T
ab
le
1
h
as
n
u
m
er
o
u
s
in
ter
s
ec
tio
n
s
w
it
h
o
t
h
er
to
p
ics
p
ar
ticu
lar
l
y
in
f
o
r
m
a
tio
n
s
y
s
te
m
(
I
S).
T
h
is
is
d
u
e
to
t
h
e
f
ac
t
t
h
at
titl
es
p
er
tain
i
n
g
to
t
h
e
d
ig
ital
ec
o
n
o
m
y
t
y
p
icall
y
u
s
e
th
e
w
o
r
d
s
“
bangun
”
(
b
u
ild
in
g
)
,
“
imp
leme
n
ta
s
i
”
(
i
m
p
le
m
e
n
ti
n
g
)
,
o
r
“
r
a
n
ca
n
g
"
(
d
esig
n
i
n
g
)
,
e
v
en
“
s
is
tem
in
fo
r
ma
s
i
”
(
i
n
f
o
r
m
ati
o
n
s
y
s
te
m
)
,
w
h
ich
ar
e
ter
m
s
t
h
at
ca
n
b
e
u
s
ed
to
r
ef
er
to
in
f
o
r
m
atio
n
s
y
s
te
m
s
.
As
s
h
o
w
n
i
n
T
ab
le
8
(
in
A
p
p
en
d
i
x
)
,
w
e
ca
n
o
b
s
er
v
e
t
h
at
s
e
v
er
al
p
h
r
ases
f
r
o
m
clas
s
E
D
ar
e
also
w
id
el
y
u
s
ed
in
class
I
S b
y
s
e
lectin
g
f
i
v
e
e
x
a
m
p
les
f
r
o
m
t
h
e
en
tire
d
ataset
f
o
r
ea
ch
E
D
an
d
I
S.
5.
CO
NCLU
SI
O
N
TF
-
I
DF,
h
o
w
e
v
er
r
eg
ar
d
ed
a
s
a
tr
ad
itio
n
al
ap
p
r
o
ac
h
in
c
o
m
p
ar
is
o
n
to
co
n
te
m
p
o
r
ar
y
a
lg
o
r
ith
m
s
,
co
n
tin
u
es
to
y
ield
ex
ce
lle
n
t
r
e
s
u
lt
s
i
n
a
v
ar
iet
y
o
f
te
x
t
m
in
i
n
g
ta
s
k
s
.
I
n
t
h
is
s
tu
d
y
,
th
e
u
s
e
o
f
s
ev
er
al
T
F
-
I
DF
m
o
d
i
f
icat
io
n
f
o
r
s
h
o
r
t
tex
t
ca
teg
o
r
is
atio
n
is
ev
al
u
ated
.
An
o
t
h
er
p
r
o
b
lem
is
i
m
b
alan
ce
d
d
atasets
ar
e
a
c
o
m
m
o
n
is
s
u
e
i
n
tex
t
m
in
i
n
g
j
o
b
s
.
I
n
o
r
d
er
to
ad
d
r
ess
th
e
i
m
b
alan
ce
d
p
r
o
b
lem
,
w
e
co
m
b
i
n
e
eit
h
er
o
v
er
s
a
m
p
li
n
g
an
d
u
n
d
er
s
a
m
p
li
n
g
m
et
h
o
d
s
w
it
h
s
tan
d
ar
d
,
lo
g
-
s
ca
led
,
an
d
b
o
o
lean
T
F
-
I
DF
in
s
h
o
r
t
tex
t
class
i
f
icatio
n
.
E
ac
h
ex
p
er
i
m
e
n
t is a
s
s
es
s
ed
u
s
in
g
m
ea
s
u
r
e
m
e
n
t
s
o
f
p
r
ec
is
io
n
,
r
e
ca
ll,
an
d
f
-
m
ea
s
u
r
e.
A
cc
o
r
d
in
g
to
th
e
r
esu
lts
,
w
e
f
in
d
th
at
th
e
u
n
d
er
s
a
m
p
li
n
g
m
eth
o
d
p
er
f
o
r
m
s
b
ad
l
y
w
h
e
n
co
m
p
ar
ed
to
t
h
e
s
ta
n
d
ar
d
ap
p
r
o
ac
h
,
w
h
er
ea
s
th
e
o
v
er
s
a
m
p
l
in
g
m
et
h
o
d
p
er
f
o
r
m
s
s
i
g
n
i
f
ica
n
tl
y
b
etter
th
an
t
h
e
s
ta
n
d
ar
d
ap
p
r
o
ac
h
in
s
e
v
er
al
T
F
-
I
DF
m
o
d
if
ica
tio
n
.
O
n
t
h
e
o
t
h
er
h
an
d
,
t
h
e
u
n
d
er
s
a
m
p
li
n
g
tec
h
n
iq
u
e
co
v
er
s
tr
u
e
p
r
ed
ictio
n
b
etter
th
an
th
e
s
ta
n
d
ar
d
an
d
o
v
er
s
am
p
li
n
g
m
e
th
o
d
if
o
n
l
y
m
i
n
o
r
it
y
cla
s
s
e
s
ar
e
m
ea
s
u
r
ed
,
lead
in
g
to
a
b
etter
r
ec
all
m
ea
s
u
r
e
m
e
n
t.
Ou
r
e
x
p
er
i
m
e
n
t
al
s
o
f
in
d
t
h
at
b
o
o
lean
T
F
-
I
DF
is
s
li
g
h
tl
y
b
e
tter
u
til
ized
th
a
n
th
e
s
tan
d
ar
d
T
F
-
I
DF
if
co
m
b
i
n
ed
w
it
h
o
v
er
s
a
m
p
li
n
g
m
et
h
o
d
.
Desp
ite
o
f
p
o
o
r
p
er
f
o
r
m
an
ce
f
o
r
u
n
d
er
s
a
m
p
li
n
g
m
et
h
o
d
,
o
u
r
ex
p
er
i
m
e
n
t
s
h
o
w
s
t
h
at
lo
g
-
s
ca
led
T
F
-
I
DF
i
s
b
etter
s
u
ited
,
b
ec
au
s
e
its
ab
il
it
y
to
h
an
d
le
s
p
ar
s
e
v
ec
to
r
.
W
ith
t
h
ese
f
i
n
d
in
g
s
,
we
b
eliev
e
t
h
at
u
tili
zi
n
g
o
v
er
s
am
p
lin
g
ap
p
r
o
ac
h
co
m
b
i
n
ed
w
i
th
b
o
o
lean
T
F
-
I
DF
v
ec
to
r
izatio
n
is
b
es
t
s
u
ited
f
o
r
i
m
b
alan
ce
d
s
h
o
r
t
tex
t
clas
s
i
f
icatio
n
,
esp
ec
iall
y
in
I
n
d
o
n
esi
an
lan
g
u
ag
e
w
h
ic
h
Evaluation Warning : The document was created with Spire.PDF for Python.
390
I
SS
N:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
2
3
,
No
.
2
,
A
p
r
il 2
0
2
5
: 3
8
2
-
392
h
as
l
i
m
ited
r
eso
u
r
ce
s
.
A
d
d
it
io
n
al
r
esear
ch
o
n
T
F
-
I
D
F
m
o
d
i
f
icat
io
n
ca
n
b
e
co
n
d
u
c
ted
in
t
h
e
f
u
tu
r
e,
p
ar
ticu
lar
l
y
w
h
en
co
m
p
ar
ed
to
th
e
m
o
s
t
ad
v
an
ce
d
w
o
r
d
2
v
ec
an
d
lar
g
e
-
lan
g
u
ag
e
m
o
d
ell
in
g
(
L
L
M)
tech
n
iq
u
es.
I
t
is
also
f
ea
s
ib
l
e
to
ex
p
lo
r
e
in
f
u
tu
r
e
r
esear
ch
u
s
i
n
g
th
e
h
y
b
r
id
tech
n
iq
u
e
b
y
ad
j
u
s
tin
g
th
e
o
v
er
s
a
m
p
li
n
g
an
d
u
n
d
er
s
a
m
p
l
in
g
.
AP
P
E
NDI
X
T
ab
le
8
.
Sam
p
le
titl
e
s
f
o
r
clas
s
E
D
an
d
I
S
ACK
NO
WL
E
D
G
E
M
E
NT
S
T
h
e
au
th
o
r
s
w
o
u
ld
lik
e
to
ex
p
r
ess
th
eir
s
in
ce
r
e
g
r
atit
u
d
e
to
Ud
ay
an
a
U
n
i
v
er
s
it
y
f
o
r
th
e
g
en
er
o
u
s
f
i
n
an
cia
l
s
u
p
p
o
r
t
p
r
o
v
id
ed
t
h
r
o
u
g
h
t
h
e
S
tu
d
y
P
r
o
g
r
a
m
P
r
ef
er
r
ed
R
esear
ch
Gr
an
t
w
i
th
co
n
tr
ac
t
n
u
m
b
er
B
/2
5
5
.
3
0
/UN1
4
.
4
.
A
/P
T
.
0
1
.
0
3
/2
0
2
4
.
T
h
is
r
esear
ch
w
o
u
ld
n
o
t h
av
e
b
ee
n
p
o
s
s
ib
le
w
ith
o
u
t t
h
eir
co
m
m
it
m
e
n
t.
RE
F
E
R
E
NC
E
S
[
1
]
K
.
K
o
w
sari
,
K
.
J.
M
e
i
ma
n
d
i
,
M
.
H
e
i
d
a
r
y
safa,
S
.
M
e
n
d
u
,
L
.
B
a
r
n
e
s,
a
n
d
D
.
B
r
o
w
n
,
“
T
e
x
t
c
l
a
ssi
f
i
c
a
t
i
o
n
a
l
g
o
r
i
t
h
ms:
A
su
r
v
e
y
,
”
I
n
f
o
rm
a
t
i
o
n
,
v
o
l
.
1
0
,
n
o
.
4
,
p
p
.
1
–
6
8
,
A
p
r
.
2
0
1
9
,
d
o
i
:
1
0
.
3
3
9
0
/
i
n
f
o
1
0
0
4
0
1
5
0
.
[
2
]
T
.
M
i
k
o
l
o
v
,
K
.
C
h
e
n
,
G
.
C
o
r
r
a
d
o
,
a
n
d
J
.
D
e
a
n
,
“
D
i
st
r
i
b
u
t
e
d
R
e
p
r
e
se
n
t
a
t
i
o
n
s
o
f
W
o
r
d
s
a
n
d
P
h
r
a
se
s
a
n
d
t
h
e
i
r
C
o
mp
o
si
t
i
o
n
a
l
i
t
y
,
”
Ad
v
a
n
c
e
s i
n
N
e
u
r
a
l
I
n
f
o
rm
a
t
i
o
n
Pro
c
e
ssi
n
g
S
y
st
e
m
s
,
v
o
l
.
2
6
,
p
p
.
3
1
1
1
–
3
1
1
9
,
2
0
1
3
.
[
3
]
A
.
Jo
u
l
i
n
,
E
.
G
r
a
v
e
,
P
.
B
o
j
a
n
o
w
sk
i
,
a
n
d
T
.
M
i
k
o
l
o
v
,
“
B
a
g
o
f
t
r
i
c
k
s
f
o
r
e
f
f
i
c
i
e
n
t
t
e
x
t
c
l
a
ssi
f
i
c
a
t
i
o
n
,
”
a
rX
i
v
p
r
e
p
r
i
n
t
,
a
r
X
i
v
:
1
6
0
7
.
0
1
7
5
9
,
2
0
1
6
.
[
4
]
J.
D
e
v
l
i
n
,
M
.
-
W
.
C
h
a
n
g
,
K
.
L
e
e
,
a
n
d
K
.
T
o
u
t
a
n
o
v
a
,
“
B
E
R
T
:
P
r
e
-
t
r
a
i
n
i
n
g
o
f
d
e
e
p
b
i
d
i
r
e
c
t
i
o
n
a
l
t
r
a
n
sf
o
r
me
r
s
f
o
r
l
a
n
g
u
a
g
e
u
n
d
e
r
st
a
n
d
i
n
g
,
”
a
rX
i
v
p
r
e
p
r
i
n
t
,
a
r
X
i
v
:
1
8
1
0
.
0
4
8
0
5
,
2
0
1
8
.
[
5
]
M
.
M
a
r
c
i
n
c
z
u
k
,
M
.
G
n
i
e
w
k
o
w
sk
i
,
T
.
W
a
l
k
o
w
i
a
k
,
a
n
d
M
.
B
e
d
k
o
w
sk
i
,
“
T
e
x
t
d
o
c
u
me
n
t
c
l
u
st
e
r
i
n
g
:
W
o
r
d
n
e
t
v
s.
TF
-
I
D
F
v
s.
w
o
r
d
e
mb
e
d
d
i
n
g
s,
”
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
1
1
t
h
G
l
o
b
a
l
W
o
rd
n
e
t
C
o
n
f
e
re
n
c
e
,
p
p
.
2
0
7
–
2
1
4
,
2
0
2
1
.
[
6
]
O
.
A
r
a
n
d
j
e
l
o
v
i
ć
,
“
T
a
r
g
e
t
e
d
A
d
a
p
t
a
b
l
e
S
a
mp
l
e
f
o
r
A
c
c
u
r
a
t
e
a
n
d
Ef
f
i
c
i
e
n
t
Q
u
a
n
t
i
l
e
Est
i
mat
i
o
n
i
n
N
o
n
-
S
t
a
t
i
o
n
a
r
y
D
a
t
a
S
t
r
e
a
ms,”
Ma
c
h
i
n
e
L
e
a
r
n
i
n
g
a
n
d
K
n
o
w
l
e
d
g
e
Ex
t
ra
c
t
i
o
n
,
v
o
l
.
1
,
n
o
.
3
,
p
p
.
8
4
8
–
8
7
0
,
J
u
l
.
2
0
1
9
,
d
o
i
:
1
0
.
3
3
9
0
/
ma
k
e
1
0
3
0
0
4
9
.
[
7
]
G
.
X
i
a
o
n
i
n
g
,
T
.
D
e
Z
h
e
r
n
,
S
.
W
.
K
i
n
g
,
T
.
Y
.
F
e
i
,
a
n
d
L
.
H
.
S
h
u
a
n
,
“
N
e
w
s
r
e
l
i
a
b
i
l
i
t
y
e
v
a
l
u
a
t
i
o
n
u
s
i
n
g
L
a
t
e
n
t
S
e
man
t
i
c
A
n
a
l
y
si
s
,”
T
e
l
k
o
m
n
i
k
a
(
T
e
l
e
c
o
m
m
u
n
i
c
a
t
i
o
n
C
o
m
p
u
t
i
n
g
E
l
e
c
t
r
o
n
i
c
s
a
n
d
C
o
n
t
r
o
l
)
,
v
o
l
.
1
6
,
n
o
.
4
,
p
p
.
1
7
0
4
–
1
7
1
1
,
2
0
1
8
,
d
o
i
:
1
0
.
1
2
9
2
8
/
T
EL
K
O
M
N
I
K
A
.
v
1
6
i
4
.
9
0
6
2
.
[
8
]
I
.
N
.
P
.
T
r
i
sn
a
a
n
d
A
.
N
u
r
w
i
d
y
a
n
t
o
r
o
,
“
S
i
n
g
l
e
d
o
c
u
me
n
t
k
e
y
w
o
r
d
s
e
x
t
r
a
c
t
i
o
n
i
n
B
a
h
a
sa
I
n
d
o
n
e
si
a
u
si
n
g
p
h
r
a
se
c
h
u
n
k
i
n
g
,
”
T
e
l
k
o
m
n
i
k
a
(
T
e
l
e
c
o
m
m
u
n
i
c
a
t
i
o
n
C
o
m
p
u
t
i
n
g
E
l
e
c
t
r
o
n
i
c
s
a
n
d
C
o
n
t
r
o
l
)
,
v
o
l
.
1
8
,
n
o
.
4
,
p
p
.
1
9
1
7
–
1
9
2
5
,
2
0
2
0
,
d
o
i
:
1
0
.
1
2
9
2
8
/
T
EL
K
O
M
N
I
K
A
.
V
1
8
I
4
.
1
4
3
8
9
.
[
9
]
C
.
D
e
B
o
o
m,
S
.
V
a
n
C
a
n
n
e
y
t
,
T
.
D
e
me
e
st
e
r
,
a
n
d
B
.
D
h
o
e
d
t
,
“
R
e
p
r
e
se
n
t
a
t
i
o
n
l
e
a
r
n
i
n
g
f
o
r
v
e
r
y
sh
o
r
t
t
e
x
t
s
u
s
i
n
g
w
e
i
g
h
t
e
d
w
o
r
d
e
mb
e
d
d
i
n
g
a
g
g
r
e
g
a
t
i
o
n
,
”
Pa
t
t
e
rn
Re
c
o
g
n
i
t
i
o
n
L
e
t
t
e
rs
,
v
o
l
.
8
0
,
p
p
.
1
5
0
–
1
5
6
,
2
0
1
6
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
a
t
r
e
c
.
2
0
1
6
.
0
6
.
0
1
2
.
[
1
0
]
H
.
S
.
A
l
-
A
sh
,
M
.
F
.
P
u
t
r
i
,
P
.
M
u
r
san
t
o
,
a
n
d
A
.
B
u
st
a
mam
,
“
E
n
se
mb
l
e
L
e
a
r
n
i
n
g
A
p
p
r
o
a
c
h
o
n
I
n
d
o
n
e
si
a
n
F
a
k
e
N
e
w
s
C
l
a
ssi
f
i
c
a
t
i
o
n
,
”
i
n
2
0
1
9
3
r
d
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
I
n
f
o
rm
a
t
i
c
s
a
n
d
C
o
m
p
u
t
a
t
i
o
n
a
l
S
c
i
e
n
c
e
s
(
I
C
I
C
o
S
)
,
2
0
1
9
,
p
p
.
1
–
6
,
d
o
i
:
S
a
mp
l
e
t
i
t
l
e
s l
a
b
e
l
l
e
d
a
s E
D
S
a
mp
l
e
t
i
t
l
e
s l
a
b
e
l
l
e
d
a
s I
S
An
a
l
i
si
s
b
i
sn
i
s
p
ro
s
e
s
d
a
n
i
m
p
l
e
m
e
n
t
a
si
e
n
t
e
r
p
r
i
se
r
e
so
u
r
c
e
p
l
a
n
n
i
n
g
(
E
RP)
p
a
d
a
B
a
t
i
k
S
r
i
k
a
n
d
i
Ba
n
y
u
w
a
n
g
i
(
En
g
l
i
sh
:
B
u
s
i
n
e
ss
P
r
o
c
e
ss
A
n
a
l
y
si
s
a
n
d
I
mp
l
e
m
e
n
t
a
t
i
o
n
o
f
ER
P
i
n
B
a
t
i
k
S
r
i
k
a
n
d
i
B
a
n
y
u
w
a
n
g
i
)
Pe
ra
n
c
a
n
g
a
n
s
i
st
e
m
i
n
f
o
rm
a
si
p
e
n
g
e
m
b
a
n
g
a
n
k
a
ri
r
u
n
t
u
k
m
a
h
a
si
s
w
a
t
e
k
n
o
l
o
g
i
i
n
f
o
rm
a
s
i
b
e
r
b
a
si
s w
e
b
si
t
e
(
En
g
l
i
sh
:
W
e
b
-
b
a
se
d
I
n
f
o
r
mat
i
o
n
S
y
st
e
m
o
f
C
a
r
e
e
r
D
e
v
e
l
o
p
me
n
t
f
o
r
S
t
u
d
e
n
t
o
f
I
n
f
o
r
mat
i
o
n
T
e
c
h
n
o
l
o
g
y
D
e
p
a
r
t
me
n
t
)
Ra
n
c
a
n
g
b
a
n
g
u
n
c
u
s
t
o
m
e
r
rel
a
t
i
o
n
s
h
i
p
m
a
n
a
j
e
m
e
n
t
p
a
d
a
si
st
e
m
i
n
f
o
rm
a
s
i
p
e
n
j
u
a
l
a
n
j
a
s
a
w
a
h
a
n
a
w
a
t
e
rs
p
o
rt
b
e
r
b
a
si
s
w
e
b
si
t
e
(
En
g
l
i
sh
:
D
e
si
g
n
i
n
g
a
n
d
I
mp
l
e
men
t
a
t
i
o
n
o
f
W
e
b
-
b
a
se
d
C
R
M
o
n
W
a
h
a
n
a
W
a
t
e
r
sp
o
r
t
S
e
r
v
i
c
e
)
Ra
n
c
a
n
g
b
a
n
g
u
n
a
p
l
i
k
a
si
p
e
n
g
e
n
a
l
a
n
b
i
o
t
a
l
a
u
t
m
e
l
a
l
u
i
b
u
k
u
b
e
r
g
a
m
b
a
r
d
e
n
g
a
n
a
u
g
m
e
n
t
e
d
re
a
l
i
t
y
(
En
g
l
i
sh
:
D
e
si
g
n
i
n
g
a
n
d
I
mp
l
e
men
t
a
t
i
o
n
o
f
M
a
r
i
n
e
B
i
o
t
a
I
d
e
n
t
i
f
i
c
a
t
i
o
n
t
h
r
o
u
g
h
A
u
g
me
n
t
e
d
R
e
a
l
i
t
y
-
b
a
se
d
D
r
a
w
i
n
g
B
o
o
k
)
Ra
n
c
a
n
g
k
n
o
w
l
e
d
g
e
b
a
se
s
y
st
e
m
d
a
n
C
RM
p
a
d
a
t
o
k
o
b
a
t
i
k
n
y
o
m
a
n
b
e
r
b
a
si
s
w
e
b
s
i
t
e
(
En
g
l
i
sh
:
D
e
si
g
n
i
n
g
W
e
b
-
b
a
se
d
K
n
o
w
l
e
d
g
e
B
a
se
S
y
st
e
m
a
n
d
C
R
M
o
f
N
y
o
man
B
a
t
i
k
S
h
o
p
)
An
a
l
i
si
s
s
i
st
e
m
m
e
n
g
g
u
n
a
k
a
n
m
e
t
o
d
e
syst
e
m
u
s
a
b
i
l
i
t
y
s
c
a
l
e
(
s
u
s)
d
a
n
c
o
n
c
u
rr
e
n
t
t
h
i
n
k
a
l
o
u
d
(
c
t
a
)
t
e
r
h
a
d
a
p
k
e
p
u
a
s
a
n
p
e
n
g
g
u
n
a
(
En
g
l
i
sh
:
S
y
st
e
m
A
n
a
l
y
si
s
o
n
C
u
s
t
o
me
r
S
a
t
i
sf
a
c
t
i
o
n
s
U
si
n
g
S
y
st
e
m U
sab
i
l
i
t
y
S
c
a
l
e
a
n
d
C
o
n
c
u
r
r
e
n
t
T
h
i
n
k
A
l
o
u
d
)
Pe
n
e
ra
p
a
n
e
l
e
c
t
r
o
n
i
c
-
c
u
s
t
o
m
e
r
re
l
a
t
i
o
n
s
h
i
p
m
a
n
a
g
e
m
e
n
t
(E
-
C
RM
)
b
e
r
b
a
si
s
w
e
b
s
i
t
e
p
a
d
a
PT
P
a
n
c
a
N
i
a
g
a
B
a
l
i
(
En
g
l
i
sh
:
I
mp
l
e
m
e
n
t
a
t
i
o
n
o
f
W
e
b
-
b
a
se
d
El
e
c
t
r
o
n
i
c
C
R
M
i
n
P
a
n
c
a
N
i
a
g
a
B
a
l
i
L
t
d
.
)
Pe
n
e
ra
p
a
n
b
u
si
n
e
ss
i
n
t
e
l
l
i
g
e
n
c
e
u
n
t
u
k
m
e
n
e
n
t
u
k
a
n
st
ra
t
e
g
i
m
a
rke
t
i
n
g
p
a
d
a
k
r
i
sn
a
o
l
e
h
–
o
l
e
h
b
a
l
i
m
e
n
g
g
u
n
a
k
a
n
m
i
c
r
o
so
f
t
p
o
w
e
r
BI
(
En
g
l
i
sh
:
I
mp
l
e
me
n
t
a
t
i
o
n
o
f
B
u
s
i
n
e
ss
I
n
t
e
l
l
i
g
e
n
c
e
f
o
r
M
a
r
k
e
t
i
n
g
S
t
r
a
t
e
g
y
o
n
K
r
i
sn
a
O
l
e
h
-
O
l
e
h
B
a
l
i
U
si
n
g
M
i
c
r
o
so
f
t
P
o
w
e
r
B
I
)
I
m
p
l
e
m
e
n
t
a
s
i
e
n
t
e
rp
r
i
se
r
e
so
u
rc
e
p
l
a
n
n
i
n
g
(
ER
P)
p
a
d
a
C
V
.
C
i
p
t
a
A
n
u
g
e
r
a
h
Ba
k
t
i
Ma
n
d
i
r
i
(
W
e
b
-
b
a
se
d
E
R
P
I
mp
l
e
me
n
t
a
t
i
o
n
o
n
C
V
.
C
i
p
t
a
A
n
u
g
e
r
a
h
B
a
k
t
i
M
a
n
d
i
r
i
)
Ra
n
c
a
n
g
b
a
n
g
u
n
s
i
st
e
m
i
n
f
o
rm
a
si
m
a
n
a
j
e
m
e
n
d
a
n
p
e
n
g
a
w
a
s
a
n
p
ro
y
e
k
k
o
n
t
r
u
k
si
k
o
l
a
m
r
e
n
a
n
g
b
e
r
b
a
s
i
s
w
e
b
s
i
t
e
p
a
d
a
m
i
m
b
a
p
o
o
l
(
En
g
l
i
sh
:
D
e
si
g
n
i
n
g
a
n
d
I
mp
l
e
m
e
n
t
a
t
i
o
n
o
f
I
n
f
o
r
mat
i
o
n
S
y
st
e
m
f
o
r
W
e
b
-
b
a
se
d
M
a
n
a
g
e
me
n
t
a
n
d
M
o
n
i
t
o
r
i
n
g
P
o
o
l
C
o
n
st
r
u
c
t
i
o
n
i
n
M
i
mb
a
P
o
o
l
C
o
mp
a
n
y
)
Evaluation Warning : The document was created with Spire.PDF for Python.
391
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
Ove
r
s
a
mp
lin
g
vs.
u
n
d
ers
a
mp
lin
g
in
TF
-
I
DF
v
a
r
ia
tio
n
s
fo
r
imb
a
la
n
ce
d
…
(
I
N
yo
ma
n
P
r
a
y
a
n
a
Tr
is
n
a
)
1
0
.
1
1
0
9
/
I
C
I
C
o
S
4
8
1
1
9
.
2
0
1
9
.
8
9
8
2
4
0
9
.
[
1
1
]
H
.
H
e
a
n
d
E.
A
.
G
a
r
c
i
a
,
“
L
e
a
r
n
i
n
g
f
r
o
m
i
mb
a
l
a
n
c
e
d
d
a
t
a
,
”
I
EEE
T
ra
n
s
a
c
t
i
o
n
s
o
n
K
n
o
w
l
e
d
g
e
a
n
d
D
a
t
a
E
n
g
i
n
e
e
ri
n
g
,
v
o
l
.
2
1
,
n
o
.
9
,
p
p
.
1
2
6
3
–
1
2
8
4
,
S
e
p
.
2
0
0
9
,
d
o
i
:
1
0
.
1
1
0
9
/
T
K
D
E.
2
0
0
8
.
2
3
9
.
[
1
2
]
J.
T
a
n
h
a
,
Y
.
A
b
d
i
,
N
.
S
a
m
a
d
i
,
N
.
R
a
z
z
a
g
h
i
,
a
n
d
M
.
A
sad
p
o
u
r
,
“
B
o
o
s
t
i
n
g
me
t
h
o
d
s
f
o
r
mu
l
t
i
-
c
l
a
ss
i
m
b
a
l
a
n
c
e
d
d
a
t
a
c
l
a
ss
i
f
i
c
a
t
i
o
n
:
a
n
e
x
p
e
r
i
me
n
t
a
l
r
e
v
i
e
w
,
”
J
o
u
r
n
a
l
o
f
B
i
g
D
a
t
a
,
v
o
l
.
7
,
n
o
.
1
,
p
p
.
1
–
4
7
,
2
0
2
0
,
d
o
i
:
1
0
.
1
1
8
6
/
s
4
0
5
3
7
-
020
-
0
0
3
4
9
-
y.
[
1
3
]
G
.
L
e
ma,
F
.
N
o
g
u
e
i
r
a
,
a
n
d
C
.
K
.
A
r
i
d
a
s,
“
I
mb
a
l
a
n
c
e
d
-
l
e
a
r
n
:
A
P
y
t
h
o
n
T
o
o
l
b
o
x
t
o
T
a
c
k
l
e
t
h
e
C
u
r
se
o
f
I
mb
a
l
a
n
c
e
d
D
a
t
a
se
t
s
i
n
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
,
”
J
o
u
r
n
a
l
o
f
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
Re
se
a
r
c
h
,
v
o
l
.
1
8
,
n
o
.
1
7
,
p
p
.
1
–
5
,
2
0
1
7
.
[
1
4
]
A
.
A
l
h
u
d
h
a
i
f
,
“
A
N
o
v
e
l
M
u
l
t
i
-
c
l
a
s
s
I
mb
a
l
a
n
c
e
d
E
EG
S
i
g
n
a
l
s
C
l
a
ssi
f
i
c
a
t
i
o
n
B
a
se
d
o
n
t
h
e
A
d
a
p
t
i
v
e
S
y
n
t
h
e
t
i
c
S
a
m
p
l
i
n
g
(
A
D
A
S
Y
N
)
a
p
p
r
o
a
c
h
,
”
P
e
e
rJ
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
7
,
p
p
.
1
–
1
5
,
2
0
2
1
,
d
o
i
:
1
0
.
7
7
1
7
/
P
E
ER
J
-
C
S
.
5
2
3
.
[
1
5
]
A
.
I
sh
a
q
e
t
a
l
.
,
“
I
mp
r
o
v
i
n
g
t
h
e
P
r
e
d
i
c
t
i
o
n
o
f
H
e
a
r
t
F
a
i
l
u
r
e
P
a
t
i
e
n
t
s
’
S
u
r
v
i
v
a
l
U
si
n
g
S
M
O
T
E
a
n
d
Ef
f
e
c
t
i
v
e
D
a
t
a
M
i
n
i
n
g
T
e
c
h
n
i
q
u
e
s,”
I
EEE
Ac
c
e
ss
,
v
o
l
.
9
,
p
p
.
3
9
7
0
7
–
3
9
7
1
6
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
1
.
3
0
6
4
0
8
4
.
[
1
6
]
R
.
Z
u
e
c
h
,
J.
H
a
n
c
o
c
k
,
a
n
d
T
.
M
.
K
h
o
sh
g
o
f
t
a
a
r
,
“
D
e
t
e
c
t
i
n
g
w
e
b
a
t
t
a
c
k
s
u
si
n
g
r
a
n
d
o
m
u
n
d
e
r
samp
l
i
n
g
a
n
d
e
n
se
mb
l
e
l
e
a
r
n
e
r
s,”
J
o
u
rn
a
l
o
f
Bi
g
D
a
t
a
,
v
o
l
.
8
,
n
o
.
1
,
p
p
.
1
–
2
0
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
8
6
/
s
4
0
5
3
7
-
0
2
1
-
0
0
4
6
0
-
8.
[
1
7
]
P
.
V
u
t
t
i
p
i
t
t
a
y
a
mo
n
g
k
o
l
a
n
d
E.
El
y
a
n
,
“
N
e
i
g
h
b
o
u
r
h
o
o
d
-
b
a
se
d
u
n
d
e
r
sam
p
l
i
n
g
a
p
p
r
o
a
c
h
f
o
r
h
a
n
d
l
i
n
g
i
mb
a
l
a
n
c
e
d
a
n
d
o
v
e
r
l
a
p
p
e
d
d
a
t
a
,
”
I
n
f
o
rm
a
t
i
o
n
S
c
i
e
n
c
e
s
,
v
o
l
.
5
0
9
,
p
p
.
4
7
–
7
0
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
n
s.
2
0
1
9
.
0
8
.
0
6
2
.
[
1
8
]
A
.
A
.
N
u
g
r
a
h
a
,
A
.
A
r
i
f
i
a
n
t
o
,
a
n
d
S
u
y
a
n
t
o
,
“
G
e
n
e
r
a
t
i
n
g
i
mag
e
d
e
scri
p
t
i
o
n
o
n
I
n
d
o
n
e
si
a
n
l
a
n
g
u
a
g
e
u
si
n
g
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
a
n
d
g
a
t
e
d
r
e
c
u
r
r
e
n
t
u
n
i
t
,
”
i
n
2
0
1
9
7
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
I
n
f
o
rm
a
t
i
o
n
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
,
I
C
o
I
C
T
2
0
1
9
,
2
0
1
9
,
p
p
.
1
–
6
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
o
I
C
T
.
2
0
1
9
.
8
8
3
5
3
7
0
.
[
1
9
]
D
.
M
u
n
a
n
d
a
r
,
A
.
F
.
R
o
z
i
e
,
a
n
d
A
.
A
r
i
sal
,
“
A
mu
l
t
i
d
o
mai
n
s
s
h
o
r
t
me
ssag
e
se
n
t
i
me
n
t
c
l
a
ss
i
f
i
c
a
t
i
o
n
u
si
n
g
h
y
b
r
i
d
n
e
u
r
a
l
n
e
t
w
o
r
k
a
r
c
h
i
t
e
c
t
u
r
e
,
”
B
u
l
l
e
t
i
n
o
f
El
e
c
t
ri
c
a
l
E
n
g
i
n
e
e
ri
n
g
a
n
d
I
n
f
o
rm
a
t
i
c
s
,
v
o
l
.
1
0
,
n
o
.
4
,
p
p
.
2
1
8
1
–
2
1
9
1
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
5
9
1
/
E
EI
.
V
1
0
I
4
.
2
7
9
0
.
[
2
0
]
Z
.
Z
h
u
,
J
.
L
i
a
n
g
,
D
.
L
i
,
H
.
Y
u
,
a
n
d
G
.
L
i
u
,
“
H
o
t
T
o
p
i
c
D
e
t
e
c
t
i
o
n
B
a
se
d
o
n
a
R
e
f
i
n
e
d
T
F
-
I
D
F
A
l
g
o
r
i
t
h
m,
”
I
E
EE
A
c
c
e
ss
,
v
o
l
.
7
,
p
p
.
2
6
9
9
6
–
2
7
0
0
7
,
2
0
1
9
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
1
9
.
2
8
9
3
9
8
0
.
[
2
1
]
K
.
B
o
k
,
Y
.
N
o
h
,
J
.
L
i
m,
a
n
d
J.
Y
o
o
,
“
H
o
t
t
o
p
i
c
p
r
e
d
i
c
t
i
o
n
c
o
n
si
d
e
r
i
n
g
i
n
f
l
u
e
n
c
e
a
n
d
e
x
p
e
r
t
i
se
i
n
so
c
i
a
l
m
e
d
i
a
,
”
E
l
e
c
t
r
o
n
i
c
C
o
m
m
e
r
c
e
Re
s
e
a
r
c
h
,
v
o
l
.
2
1
,
n
o
.
3
,
p
p
.
6
7
1
–
6
8
7
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
0
6
6
0
-
0
1
8
-
0
9
3
2
7
-
2.
[
2
2
]
J.
P
i
sk
o
r
sk
i
a
n
d
G
.
Ja
c
q
u
e
t
,
“
TF
-
I
D
F
C
h
a
r
a
c
t
e
r
N
-
g
r
a
ms
v
e
r
su
s
W
o
r
d
Emb
e
d
d
i
n
g
-
b
a
se
d
M
o
d
e
l
s
f
o
r
F
i
n
e
-
g
r
a
i
n
e
d
Ev
e
n
t
C
l
a
ssi
f
i
c
a
t
i
o
n
:
A
P
r
e
l
i
mi
n
a
r
y
S
t
u
d
y
,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
W
o
rks
h
o
p
o
n
Au
t
o
m
a
t
e
d
E
x
t
r
a
c
t
i
o
n
o
f
S
o
c
i
o
-
p
o
l
i
t
i
c
a
l
E
v
e
n
t
s
f
ro
m
N
e
w
s
2
0
2
0
,
2
0
2
0
,
p
p
.
2
6
–
3
4
.
[
2
3
]
R
.
J
.
O
sk
o
u
e
i
a
n
d
B
.
S
.
B
i
g
h
a
m
,
“
O
v
e
r
-
samp
l
i
n
g
v
i
a
u
n
d
e
r
-
samp
l
i
n
g
i
n
st
r
o
n
g
l
y
i
mb
a
l
a
n
c
e
d
d
a
t
a
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
Ad
v
a
n
c
e
d
I
n
t
e
l
l
i
g
e
n
c
e
P
a
r
a
d
i
g
m
s
,
v
o
l
.
9
,
n
o
.
1
,
p
p
.
5
8
–
6
6
,
2
0
1
7
,
d
o
i
:
1
0
.
1
5
0
4
/
i
j
a
i
p
.
2
0
1
7
.
1
0
0
0
2
0
2
6
.
[
2
4
]
A
.
Tr
i
a
y
u
d
i
a
n
d
I
.
F
i
t
r
i
,
“
C
o
mp
a
r
i
so
n
O
f
T
h
e
F
e
a
t
u
r
e
S
e
l
e
c
t
i
o
n
A
l
g
o
r
i
t
h
m
I
n
Ed
u
c
a
t
i
o
n
a
l
D
a
t
a
M
i
n
i
n
g
,
”
T
e
l
k
o
m
n
i
k
a
(
T
e
l
e
c
o
m
m
u
n
i
c
a
t
i
o
n
C
o
m
p
u
t
i
n
g
El
e
c
t
r
o
n
i
c
s
a
n
d
C
o
n
t
r
o
l
)
,
v
o
l
.
1
9
,
n
o
.
6
,
p
p
.
1
8
6
5
–
1
8
7
1
,
2
0
2
1
,
d
o
i
:
1
0
.
1
2
9
2
8
/
T
EL
K
O
M
N
I
K
A
.
v
1
9
i
6
.
2
1
5
9
4
.
[
2
5
]
R
.
M
o
h
a
mm
e
d
,
J
.
R
a
w
a
sh
d
e
h
,
a
n
d
M
.
A
b
d
u
l
l
a
h
,
“
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
w
i
t
h
O
v
e
r
samp
l
i
n
g
a
n
d
U
n
d
e
r
sam
p
l
i
n
g
T
e
c
h
n
i
q
u
e
s:
O
v
e
r
v
i
e
w
S
t
u
d
y
a
n
d
Ex
p
e
r
i
me
n
t
a
l
R
e
su
l
t
s,”
i
n
2
0
2
0
1
1
t
h
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
I
n
f
o
rm
a
t
i
o
n
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
S
y
s
t
e
m
s,
I
C
I
C
S
2
0
2
0
,
2
0
2
0
,
p
p
.
2
4
3
–
2
4
8
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
I
C
S
4
9
4
6
9
.
2
0
2
0
.
2
3
9
5
5
6
.
[
2
6
]
G
.
B
e
r
n
a
r
d
,
C
.
S
u
i
r
e
,
C
.
F
a
u
c
h
e
r
,
A
.
D
o
u
c
e
t
,
a
n
d
P
.
R
o
sso
,
“
T
r
a
c
k
i
n
g
N
e
w
s
S
t
o
r
i
e
s
i
n
S
h
o
r
t
M
e
ssag
e
s
i
n
t
h
e
Er
a
o
f
I
n
f
o
d
e
mi
c
,
”
L
e
c
t
u
re
N
o
t
e
s
i
n
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
(
i
n
c
l
u
d
i
n
g
su
b
ser
i
e
s
L
e
c
t
u
r
e
N
o
t
e
s
i
n
Ar
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
L
e
c
t
u
re
N
o
t
e
s
i
n
Bi
o
i
n
f
o
rm
a
t
i
c
s)
,
p
p
.
1
8
–
3
2
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
0
3
1
-
1
3
6
4
3
-
6
_
2
.
[
2
7
]
S
.
M
i
r
a
n
d
a
,
A
.
Z
n
o
t
i
n
š
,
S
.
B
.
C
o
h
e
n
,
a
n
d
G
.
B
a
r
z
d
i
n
s,
“
M
u
l
t
i
l
i
n
g
u
a
l
c
l
u
s
t
e
r
i
n
g
o
f
st
r
e
a
mi
n
g
n
e
w
s,”
a
rX
i
v
p
re
p
ri
n
t
,
a
r
X
i
v
:
1
8
0
9
.
0
0
5
4
0
,
2
0
1
8
.
[
2
8
]
V
.
M
a
r
i
v
a
t
e
a
n
d
T
.
S
e
f
a
r
a
,
"
I
mp
r
o
v
i
n
g
S
h
o
r
t
T
e
x
t
C
l
a
ss
i
f
i
c
a
t
i
o
n
T
h
r
o
u
g
h
G
l
o
b
a
l
A
u
g
me
n
t
a
t
i
o
n
M
e
t
h
o
d
s
,"
I
n
C
D
-
MA
K
E
2
0
2
0
:
Ma
c
h
i
n
e
L
e
a
r
n
i
n
g
a
n
d
K
n
o
w
l
e
d
g
e
Ex
t
ra
c
t
i
o
n
,
p
p
.
3
8
5
-
3
9
9
,
2
0
2
0
,
d
o
i
:
1
0
.
4
8
5
5
0
/
a
r
X
i
v
.
1
9
0
7
.
0
3
7
5
2
.
[2
9
]
R
.
S
e
t
i
a
b
u
d
i
,
N
.
M
.
S
.
I
s
w
a
r
i
,
a
n
d
A
.
R
u
sl
i
,
“
En
h
a
n
c
i
n
g
t
e
x
t
c
l
a
ss
i
f
i
c
a
t
i
o
n
p
e
r
f
o
r
man
c
e
b
y
p
r
e
p
r
o
c
e
ssi
n
g
m
i
ssp
e
l
l
e
d
w
o
r
d
s
i
n
I
n
d
o
n
e
si
a
n
l
a
n
g
u
a
g
e
,
”
T
e
l
k
o
m
n
i
k
a
(
T
e
l
e
c
o
m
m
u
n
i
c
a
t
i
o
n
C
o
m
p
u
t
i
n
g
El
e
c
t
ro
n
i
c
s
a
n
d
C
o
n
t
r
o
l
)
,
v
o
l
.
1
9
,
n
o
.
4
,
p
p
.
1
2
3
4
–
1
2
4
1
,
2
0
2
1
,
d
o
i
:
1
0
.
1
2
9
2
8
/
T
EL
K
O
M
N
I
K
A
.
v
1
9
i
4
.
2
0
3
6
9
.
[
30
]
H
.
A
.
S
a
n
t
o
so
,
E.
H
.
R
a
c
h
maw
a
n
t
o
,
A
.
N
u
g
r
a
h
a
,
A
.
A
.
N
u
g
r
o
h
o
,
D
.
R
.
I
.
M
.
S
e
t
i
a
d
i
,
a
n
d
R
.
S
.
B
a
s
u
k
i
,
“
H
o
a
x
c
l
a
ssi
f
i
c
a
t
i
o
n
a
n
d
se
n
t
i
me
n
t
a
n
a
l
y
si
s
o
f
I
n
d
o
n
e
si
a
n
n
e
w
s
u
si
n
g
N
a
i
v
e
B
a
y
e
s
o
p
t
i
mi
z
a
t
i
o
n
,
”
T
e
l
k
o
m
n
i
k
a
(
T
e
l
e
c
o
m
m
u
n
i
c
a
t
i
o
n
C
o
m
p
u
t
i
n
g
E
l
e
c
t
r
o
n
i
c
s
a
n
d
C
o
n
t
ro
l
)
,
v
o
l
.
1
8
,
n
o
.
2
,
p
p
.
7
9
9
–
8
0
6
,
2
0
2
0
,
d
o
i
:
1
0
.
1
2
9
2
8
/
T
EL
K
O
M
N
I
K
A
.
V
1
8
I
2
.
1
4
7
4
4
.
[3
1
]
R
.
A
.
R
a
ma
d
a
n
i
,
I
.
K
.
G
.
D
.
P
u
t
r
a
,
M
.
S
u
d
a
r
ma
,
a
n
d
I
.
A
.
D
.
G
i
r
i
a
n
t
a
r
i
,
“
A
n
e
w
t
e
c
h
n
o
l
o
g
y
o
n
t
r
a
n
sl
a
t
i
n
g
I
n
d
o
n
e
si
a
n
s
p
o
k
e
n
l
a
n
g
u
a
g
e
i
n
t
o
I
n
d
o
n
e
si
a
n
si
g
n
l
a
n
g
u
a
g
e
s
y
st
e
m,”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
El
e
c
t
r
i
c
a
l
a
n
d
C
o
m
p
u
t
e
r
En
g
i
n
e
e
ri
n
g
,
v
o
l
.
1
1
,
n
o
.
4
,
p
p
.
3
3
3
8
–
3
3
4
6
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
c
e
.
v
1
1
i
4
.
p
p
3
3
3
8
-
3
3
4
6
.
[3
2
]
C
.
D
.
M
a
n
n
i
n
g
,
I
n
t
ro
d
u
c
t
i
o
n
t
o
I
n
f
o
r
m
a
t
i
o
n
R
e
t
r
i
e
v
a
l
.
C
a
mb
r
i
d
g
e
U
n
i
v
e
r
si
t
y
P
r
e
ss,
2
0
0
8
.
d
o
i
:
1
0
.
1
0
1
7
/
c
b
o
9
7
8
0
5
1
1
8
0
9
0
7
1
.
[3
3
]
T
.
W
o
n
g
v
o
r
a
c
h
a
n
,
S
.
H
e
,
a
n
d
O
.
B
u
l
u
t
,
“
A
C
o
mp
a
r
i
so
n
o
f
U
n
d
e
r
samp
l
i
n
g
,
O
v
e
r
samp
l
i
n
g
,
a
n
d
S
M
O
T
E
M
e
t
h
o
d
s
f
o
r
D
e
a
l
i
n
g
w
i
t
h
I
mb
a
l
a
n
c
e
d
C
l
a
ssi
f
i
c
a
t
i
o
n
i
n
Ed
u
c
a
t
i
o
n
a
l
D
a
t
a
M
i
n
i
n
g
,
”
I
n
f
o
rm
a
t
i
o
n
,
v
o
l
.
1
4
,
n
o
.
1
,
p
p
.
1
–
1
5
,
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
i
n
f
o
1
4
0
1
0
0
5
4
.
[3
4
]
R
.
S
o
l
e
y
man
i
,
E.
G
r
a
n
g
e
r
,
a
n
d
G
.
F
u
me
r
a
,
“
F
-
me
a
su
r
e
c
u
r
v
e
s:
A
t
o
o
l
t
o
v
i
s
u
a
l
i
z
e
c
l
a
ssi
f
i
e
r
p
e
r
f
o
r
man
c
e
u
n
d
e
r
i
mb
a
l
a
n
c
e
,
”
Pa
t
t
e
r
n
Re
c
o
g
n
i
t
i
o
n
,
v
o
l
.
1
0
0
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
a
t
c
o
g
.
2
0
1
9
.
1
0
7
1
4
6
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
I
Ny
o
m
a
n
Pra
y
a
n
a
Tr
is
n
a
re
c
e
iv
e
d
th
e
B
a
c
h
e
lo
r
’
s
a
n
d
M
a
ste
r
’
s
d
e
g
r
e
e
in
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
El
e
c
tro
n
ics
f
ro
m
th
e
Un
iv
e
rsitas
Ga
d
jah
M
a
d
a
,
Yo
g
y
a
k
a
rta,
In
d
o
n
e
sia
,
in
2
0
1
7
a
n
d
2
0
2
0
re
sp
e
c
ti
v
e
ly
.
He
is
c
u
rre
n
tl
y
a
n
A
ss
istan
t
P
ro
f
e
ss
o
r
in
th
e
In
f
o
rm
a
ti
o
n
T
e
c
h
n
o
lo
g
y
S
tu
d
y
P
ro
g
ra
m
,
F
a
c
u
lt
y
o
f
En
g
in
e
e
rin
g
,
Ud
a
y
a
n
a
Un
i
v
e
rs
it
y
,
Ba
li
,
In
d
o
n
e
sia
.
H
i
s
c
u
r
r
e
n
t
r
e
se
a
rc
h
i
n
t
e
r
e
s
ts
i
n
c
l
u
d
e
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
,
e
v
o
l
u
t
i
o
n
a
ry
c
o
m
p
u
t
a
t
i
o
n
,
n
a
t
u
r
a
l
l
a
n
g
u
a
g
e
m
o
d
e
l
l
i
n
g
,
a
n
d
t
e
x
t
m
i
n
i
n
g
.
H
e
c
a
n
b
e
c
o
n
t
a
c
t
e
d
a
t
e
m
a
i
l
:
p
r
a
y
a
n
a
.
t
ri
s
n
a
@
u
n
u
d
.
a
c
.
i
d
.
Evaluation Warning : The document was created with Spire.PDF for Python.