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
.
426
~
434
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
i
2
.
25654
426
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//telko
mn
ika
.
u
a
d
.
a
c.
i
d
A co
m
pa
ra
tive a
n
a
ly
sis
of t
ra
nsfer l
ea
rning
m
o
d
els o
n suicide
a
nd no
n
-
suicide
t
ex
tual da
ta
M
er
ind
a
L
esta
nd
y
1
,
Abdu
rr
a
hi
m
2
,
A
m
rul F
a
ru
q
1
,
M
uh
a
mm
a
d Irf
a
n
1
1
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
i
c
a
l
En
g
i
n
e
e
r
i
n
g
,
F
a
c
u
l
t
y
o
f
En
g
i
n
e
e
r
i
n
g
,
U
n
i
v
e
r
si
t
a
s
M
u
h
a
mm
a
d
i
y
a
h
M
a
l
a
n
g
,
M
a
l
a
n
g
,
I
n
d
o
n
e
si
a
2
I
n
f
o
r
mat
i
c
s
M
a
s
t
e
r
P
r
o
g
r
a
m,
F
a
c
u
l
t
y
o
f
I
n
d
u
st
r
i
a
l
T
e
c
h
n
o
l
o
g
y
,
I
s
l
a
mi
c
U
n
i
v
e
r
si
t
y
o
f
I
n
d
o
n
e
si
a
,
D
a
e
r
a
h
I
st
i
me
w
a
Y
o
g
y
a
k
a
r
t
a
,
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
Au
g
2
7
,
2024
R
ev
i
s
ed
J
an
6
,
2
0
2
5
A
cc
ep
ted
J
an
2
2
,
2
0
2
5
T
h
e
rise
o
f
so
c
ial
m
e
d
ia
h
a
s
a
ll
o
w
e
d
i
n
d
iv
id
u
a
ls
t
o
e
x
p
re
ss
th
e
m
se
l
v
e
s
f
re
e
l
y
,
in
c
re
a
sin
g
th
e
v
isib
il
it
y
o
f
m
e
n
tal
h
e
a
lt
h
c
o
n
c
e
rn
s,
in
c
l
u
d
i
n
g
su
icid
a
l
ten
d
e
n
c
ies
.
T
h
is
issu
e
is
p
a
rti
c
u
larly
sig
n
i
f
ica
n
t,
a
s
su
icid
e
is
o
n
e
o
f
th
e
lea
d
in
g
c
a
u
se
s
o
f
d
e
a
th
g
lo
b
a
ll
y
.
T
h
e
o
b
jec
ti
v
e
o
f
th
is
stu
d
y
is
to
d
e
v
e
lo
p
a
m
o
d
e
l
c
a
p
a
b
le
o
f
a
c
c
u
ra
tel
y
d
e
tec
ti
n
g
su
icid
e
-
re
late
d
tex
tu
a
l
d
a
ta
u
sin
g
a
d
v
a
n
c
e
d
n
a
tu
ra
l
lan
g
u
a
g
e
p
ro
c
e
ss
in
g
tec
h
n
i
q
u
e
s.
T
o
a
c
h
iev
e
t
h
is,
w
e
a
p
p
li
e
d
tran
sf
e
r
le
a
rn
in
g
m
o
d
e
ls,
i
n
c
lu
d
in
g
b
id
irec
ti
o
n
a
l
e
n
c
o
d
e
r
re
p
re
se
n
tatio
n
s
f
ro
m
tran
sf
o
r
m
e
rs
(BER
T
)
,
ro
b
u
s
tl
y
o
p
ti
m
iz
e
d
b
id
irec
ti
o
n
a
l
e
n
c
o
d
e
r
re
p
re
se
n
tatio
n
s
f
ro
m
tran
s
f
o
rm
e
r
s
(Ro
BERT
)
,
a
li
t
e
BERT
(
AL
B
ERT
)
,
a
n
d
d
e
c
o
d
i
n
g
-
e
n
h
a
n
c
e
d
BERT
w
it
h
d
ise
n
tan
g
led
a
tt
e
n
ti
o
n
(De
BE
RT
a
)
.
th
e
d
a
tas
e
t
u
se
d
i
n
t
h
is
re
se
a
rc
h
in
c
l
u
d
e
s
2
3
2
,
0
7
4
p
o
sts
f
ro
m
Re
d
d
it
,
c
a
teg
o
rize
d
in
to
su
icid
e
a
n
d
n
o
n
-
su
ici
d
e
lab
e
ls.
P
re
p
ro
c
e
ss
in
g
ste
p
s
su
c
h
a
s
re
m
o
v
in
g
HT
M
L
tag
s,
sp
e
c
ial
c
h
a
ra
c
ter
s,
a
n
d
p
u
n
c
tu
a
ti
o
n
w
e
re
a
p
p
li
e
d
,
f
o
ll
o
w
e
d
b
y
sto
p
w
o
rd
re
m
o
v
a
l
a
n
d
le
m
m
a
ti
z
a
t
io
n
.
T
h
e
m
o
d
e
ls
w
e
re
train
e
d
a
n
d
e
v
a
lu
a
ted
u
sin
g
a
c
c
u
ra
c
y
,
p
re
c
isio
n
,
re
c
a
ll
,
a
n
d
F
1
-
sc
o
re
m
e
tri
c
s.
Am
o
n
g
th
e
m
o
d
e
ls
tes
ted
,
De
BERT
a
d
e
m
o
n
stra
ted
s
u
p
e
rio
r
p
e
rf
o
rm
a
n
c
e
,
a
c
h
iev
in
g
a
n
a
c
c
u
ra
c
y
o
f
9
8
.
7
0
%
a
n
d
a
n
F
1
-
sc
o
re
o
f
9
8
.
7
0
%
.
T
h
e
se
f
in
d
in
g
s
su
g
g
e
st
th
a
t
tran
sf
e
r
lea
rn
in
g
m
o
d
e
ls,
p
a
rti
c
u
larly
De
BERT
a
,
a
re
e
ffe
c
ti
v
e
in
id
e
n
ti
f
y
i
n
g
su
icid
a
l
id
e
a
ti
o
n
in
tex
tu
a
l
d
a
ta.
K
ey
w
o
r
d
s
:
Dee
p
lear
n
in
g
Me
d
ia
s
o
cial
Sen
ti
m
e
n
t
an
a
l
y
s
is
Su
icid
e
T
r
an
s
f
o
r
m
er
m
o
d
el
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
:
Me
r
in
d
a
L
es
tan
d
y
Dep
ar
t
m
en
t o
f
E
lectr
ical
E
n
g
i
n
ee
r
in
g
,
Fac
u
lt
y
o
f
E
n
g
i
n
ee
r
in
g
,
Un
i
v
er
s
i
tas M
u
h
a
m
m
ad
i
y
a
h
Ma
lan
g
Ma
lan
g
,
I
n
d
o
n
esia
E
m
ail:
m
er
in
d
ale
s
tan
d
y
@
u
m
m
.
ac
.
id
1.
I
NT
RO
D
UCT
I
O
N
T
h
e
is
s
u
e
o
f
m
en
tal
h
ea
lt
h
h
a
s
a
g
lo
b
al
im
p
ac
t,
ex
er
tin
g
s
u
b
s
tan
tial
e
f
f
ec
t
s
o
n
co
u
n
tr
ies
t
h
r
o
u
g
h
o
u
t
th
e
s
p
ec
tr
u
m
o
f
d
e
v
elo
p
m
e
n
t.
A
cc
o
r
d
in
g
to
th
e
m
e
n
tal
h
ea
lt
h
ac
tio
n
p
la
n
f
o
r
th
e
y
ea
r
s
2
0
1
3
-
2
0
2
0
,
as r
ep
o
r
te
d
b
y
th
e
W
o
r
ld
Hea
lth
Or
g
a
n
izat
io
n
(
W
HO)
,
ap
p
r
o
x
im
a
tel
y
o
n
e
o
u
t
o
f
e
v
er
y
f
o
u
r
i
n
d
i
v
id
u
al
s
g
lo
b
all
y
e
n
co
u
n
ter
s
m
en
tal
d
is
ea
s
es
o
f
v
ar
io
u
s
s
ev
er
it
y
[
1
]
.
U
n
f
o
r
tu
n
a
te
ly
,
a
s
ig
n
if
i
c
an
t
p
r
o
p
o
r
t
i
o
n
o
f
in
d
iv
i
d
u
a
ls
ex
p
e
r
ie
n
c
in
g
m
en
t
al
h
e
a
l
th
d
if
f
ic
u
l
t
ie
s
,
n
am
e
ly
7
5
%
,
d
o
n
o
t
o
b
t
a
in
s
u
f
f
i
ci
en
t
th
e
r
a
p
y
,
h
en
c
e
in
c
r
e
as
in
g
th
e
i
r
ex
is
t
i
n
g
c
h
a
ll
en
g
e
s
.
I
n
t
h
e
p
a
s
t
f
ew
y
e
a
r
s
,
t
h
e
r
e
h
as
b
e
e
n
a
g
r
o
w
in
g
r
e
co
g
n
i
t
i
o
n
o
f
th
e
h
e
ig
h
t
en
e
d
im
p
o
r
t
a
n
c
e
o
f
em
o
t
i
o
n
a
l
c
o
n
c
e
r
n
s
as
s
o
c
ia
t
e
d
w
i
th
c
o
r
o
n
av
i
r
u
s
d
is
e
a
s
e
2
0
1
9
(
C
O
V
I
D
-
19
)
an
d
th
e
ex
p
e
r
i
en
c
e
o
f
i
s
o
l
a
t
i
o
n
[
2
]
.
T
h
is
i
s
p
ar
ticu
lar
l
y
n
o
te
w
o
r
t
h
y
in
t
h
e
co
n
tex
t
o
f
i
n
d
iv
id
u
als w
h
o
alr
ea
d
y
h
a
v
e
p
r
e
-
ex
is
ti
n
g
m
e
n
tal
h
ea
lt
h
illn
e
s
s
e
s
[
3
]
.
T
h
e
af
o
r
e
m
en
t
io
n
ed
p
h
e
n
o
m
e
n
o
n
h
as
r
es
u
lted
i
n
ele
v
ated
le
v
els
o
f
an
x
iet
y
,
d
esp
air
,
s
u
b
s
t
an
ce
m
is
u
s
e,
s
o
cial
is
o
latio
n
,
i
n
ti
m
ate
p
ar
t
n
er
v
io
l
en
ce
,
an
d
,
i
n
s
e
v
er
e
i
n
s
ta
n
ce
s
,
s
elf
-
i
n
f
licted
m
o
r
t
al
it
y
[
4
]
.
I
t
is
w
o
r
t
h
m
e
n
tio
n
i
n
g
th
at
t
h
er
e
h
as b
ee
n
a
n
i
n
cr
ea
s
e
in
th
e
p
r
o
b
ab
ilit
y
o
f
atte
m
p
te
d
s
u
icid
e
w
it
h
in
t
h
e
g
en
er
al
p
o
p
u
latio
n
[
2
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
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
A
co
mp
a
r
a
tive
a
n
a
lysi
s
o
f tra
n
s
fer lea
r
n
in
g
mo
d
els o
n
s
u
ici
d
e
a
n
d
n
o
n
-
s
u
icid
e
textu
a
l
…
(
Me
r
in
d
a
Les
ta
n
d
y
)
427
B
a
s
e
d
o
n
d
a
t
a
f
r
o
m
th
e
W
HO
,
i
t
h
as
b
e
en
d
e
t
e
r
m
in
e
d
th
a
t
a
r
o
u
n
d
8
0
0
,
0
0
0
i
n
d
i
v
i
d
u
al
s
en
g
ag
e
i
n
a
c
ts
o
f
s
u
i
c
i
d
e
an
n
u
al
ly
,
w
i
th
th
e
a
g
e
c
o
h
o
r
t
r
an
g
in
g
f
r
o
m
1
5
t
o
2
9
y
e
a
r
s
ex
h
i
b
i
t
in
g
th
e
h
ig
h
es
t
in
c
id
e
n
c
e
r
a
t
es
[
5
]
.
T
h
i
s
ex
e
m
p
li
f
ie
s
th
e
f
ac
t
t
h
at
s
u
ici
d
e
r
an
k
s
as
th
e
s
ec
o
n
d
m
o
s
t
p
r
o
m
in
e
n
t
ca
u
s
e
o
f
m
o
r
talit
y
a
m
o
n
g
ad
o
lescen
t
s
o
n
a
g
lo
b
al
s
ca
le.
C
u
r
r
en
tl
y
,
s
o
ci
al
m
ed
ia
p
lat
f
o
r
m
s
e
n
ab
le
i
n
t
er
ac
tiv
e
co
m
m
u
n
icatio
n
,
e
n
ab
lin
g
u
s
er
s
to
s
h
ar
e
th
eir
v
ie
w
s
a
n
d
e
m
o
tio
n
s
th
r
o
u
g
h
t
h
e
m
ea
n
s
o
f
p
o
s
ts
a
n
d
co
m
m
en
t
s
.
So
cial
m
ed
ia
p
lat
f
o
r
m
s
ca
n
f
u
n
ct
io
n
as
a
m
ed
iu
m
f
o
r
d
is
s
e
m
in
a
tin
g
in
f
o
r
m
at
io
n
r
eg
ar
d
in
g
r
is
k
f
ac
to
r
s
ass
o
ciate
d
w
it
h
s
u
icid
e.
T
h
e
an
al
y
s
i
s
in
cl
u
d
es
en
v
ir
o
n
m
e
n
tal
f
ac
to
r
s
,
s
u
ch
a
s
i
n
s
ta
n
ce
s
o
f
ab
u
s
e
a
n
d
e
x
p
o
s
u
r
e
to
s
tr
ess
f
u
l
e
v
e
n
ts
;
h
ea
l
th
-
r
elate
d
ch
a
llen
g
es,
s
u
c
h
as
t
h
e
p
r
esen
ce
o
f
c
h
r
o
n
i
c
d
is
ea
s
es
an
d
m
e
n
tal
h
ea
lt
h
d
is
o
r
d
er
s
;
an
d
h
i
s
to
r
ical
f
ac
to
r
s
,
in
cl
u
d
in
g
f
a
m
il
y
h
is
to
r
y
a
n
d
p
r
ev
io
u
s
s
u
icid
e
atte
m
p
ts
.
Al
l
th
e
s
e
q
u
al
itie
s
ar
e
p
o
s
itiv
el
y
as
s
o
ciate
d
w
it
h
s
u
ic
id
al
in
te
n
t.
Mo
r
eo
v
er
,
th
er
e
ex
i
s
t
v
ar
io
u
s
f
ac
to
r
s
,
s
p
ec
if
icall
y
t
h
r
ee
p
r
im
ar
y
r
is
k
f
ac
to
r
s
,
t
h
at
ca
n
co
n
tr
ib
u
te
to
t
h
e
d
ev
elo
p
m
en
t
o
f
s
u
icid
al
in
te
n
t
o
r
id
ea
tio
n
:
en
v
ir
o
n
m
en
ta
l
f
ac
to
r
s
(
s
u
ch
as
ab
u
s
e
an
d
h
ig
h
l
y
s
tr
es
s
f
u
l
li
f
e
ev
en
t
s
)
,
h
ea
l
th
f
ac
to
r
s
(
i
n
cl
u
d
in
g
c
h
r
o
n
ic
p
ai
n
a
n
d
m
e
n
tal
h
e
alth
i
s
s
u
es),
a
n
d
h
i
s
to
r
ical
f
ac
t
o
r
s
(
s
u
ch
as
a
f
a
m
il
y
h
is
to
r
y
o
f
s
u
icid
e
o
r
p
r
ev
io
u
s
s
u
icid
e
atte
m
p
t
s
)
.
T
h
ese
r
is
k
f
ac
to
r
s
ca
n
p
o
ten
tiall
y
lead
to
th
e
d
ev
elo
p
m
en
t
o
f
s
u
ic
id
al
in
ten
t
io
n
s
o
r
th
o
u
g
h
t
s
[
6
]
.
T
h
er
ef
o
r
e,
th
e
f
ield
o
f
n
atu
r
al
lan
g
u
a
g
e
p
r
o
ce
s
s
in
g
(
NL
P
)
ass
u
m
e
s
a
s
ig
n
i
f
ica
n
t
r
o
le
i
n
t
h
e
d
etec
ti
o
n
an
d
id
e
n
ti
f
icatio
n
o
f
s
u
ici
d
e
in
te
n
tio
n
s
e
x
p
r
ess
ed
i
n
te
x
tu
a
l
co
n
te
n
t.
T
h
e
p
r
o
v
id
ed
d
ata
p
r
esen
ts
v
al
u
a
b
le
in
s
ig
h
ts
t
h
at
ca
n
b
e
u
tili
s
ed
in
t
h
e
d
e
v
elo
p
m
e
n
t
o
f
s
y
s
te
m
s
t
h
at
h
av
e
th
e
ca
p
ab
ilit
y
to
f
o
r
ec
ast
an
in
d
i
v
i
d
u
al
’
s
p
r
o
p
en
s
it
y
f
o
r
en
g
ag
i
n
g
in
s
u
ic
id
e
atte
m
p
t
s
[
7
]
.
R
esear
ch
er
s
u
tili
s
e
m
a
n
y
t
y
p
es
o
f
m
o
d
els,
in
cl
u
d
in
g
m
a
ch
in
e
lear
n
in
g
m
o
d
els,
d
ee
p
le
ar
n
in
g
m
o
d
els,
an
d
tr
an
s
f
o
r
m
e
r
-
b
ased
m
o
d
els,
i
n
o
r
d
er
to
i
d
en
tify
te
x
t
u
al
co
n
t
en
t
t
h
at
i
n
d
icate
s
p
o
ten
tia
l
s
u
icid
e
th
o
u
g
h
t
s
[
6
]
.
T
h
ese
ef
f
o
r
ts
co
n
tr
ib
u
te
to
p
r
o
v
id
in
g
p
r
o
tectio
n
an
d
r
ed
u
cin
g
t
h
e
i
n
cid
en
ce
o
f
s
u
icid
e.
I
n
r
ec
en
t
y
ea
r
s
,
th
e
r
e
h
as
b
e
en
a
s
ig
n
if
i
ca
n
t
f
o
cu
s
in
th
e
f
iel
d
o
f
NL
P
o
n
th
e
au
to
m
ate
d
id
e
n
tif
icat
io
n
o
f
m
en
t
al
h
e
alth
d
is
o
r
d
e
r
s
u
s
in
g
d
iv
e
r
s
e
d
at
a
s
o
u
r
c
es
s
u
ch
as
ele
ctr
o
n
i
c
h
ea
l
th
r
e
c
o
r
d
s
(
E
HR
s
)
,
clin
i
ca
l
r
e
co
r
d
s
,
b
i
o
m
ar
k
er
s
,
w
r
itten
t
ex
ts
,
an
d
o
n
lin
e
p
o
s
ts
[
8
]
-
[
1
0
]
.
NL
P
m
eth
o
d
o
l
o
g
ies
h
av
e
ex
h
i
b
it
ed
t
h
eir
ef
f
ica
cy
in
th
e
class
if
i
ca
t
io
n
o
f
in
i
tia
l
in
d
i
ca
ti
o
n
s
o
f
m
en
tal
i
lln
ess
,
ak
in
to
t
h
e
s
tu
d
y
o
f
s
en
t
im
en
t
[
1
1
]
-
[
1
3
]
.
T
h
e
tr
an
s
f
o
r
m
er
-
b
as
ed
l
an
g
u
ag
e
m
o
d
el
[
1
4
]
is
an
a
p
p
r
o
ac
h
th
a
t
h
as
d
em
o
n
s
tr
a
te
d
r
em
ar
k
ab
le
ef
f
e
ctiv
en
es
s
.
T
h
e
t
r
an
s
d
u
ct
io
n
m
o
d
el
u
n
d
er
d
is
cu
s
s
i
o
n
is
f
o
u
n
d
e
d
u
p
o
n
a
p
h
en
o
m
en
o
n
k
n
o
w
n
as
att
en
ti
o
n
,
w
h
ich
h
as
s
ig
n
if
ican
tly
tr
an
s
f
o
r
m
ed
b
r
ain
en
c
o
d
e
r
s
d
esig
n
ed
f
o
r
n
atu
r
a
l
l
an
g
u
ag
e
s
e
q
u
en
ce
s
.
T
h
e
tr
an
s
f
o
r
m
er
a
r
ch
i
te
ctu
r
e
o
m
its
an
y
n
o
is
e
o
r
co
n
v
o
lu
ti
o
n
al
s
t
r
u
ctu
r
es,
h
en
c
e
all
o
w
in
g
th
e
ac
q
u
is
i
ti
o
n
o
f
s
e
q
u
en
ce
in
f
o
r
m
ati
o
n
in
th
e
in
p
u
t
ex
clu
s
iv
ely
th
r
o
u
g
h
atten
t
io
n
m
ec
h
an
is
m
s
.
T
h
e
p
r
esen
ce
o
f
a
s
elf
-
att
en
ti
o
n
m
ec
h
an
is
m
w
ith
in
th
e
en
co
d
e
r
’
s
p
r
o
ce
s
s
in
g
b
l
o
ck
is
r
es
p
o
n
s
ib
le
f
o
r
th
is
o
u
tc
o
m
e.
T
h
e
b
id
ir
ec
t
io
n
al
c
o
n
tex
t
in
f
o
r
m
ati
o
n
p
r
o
ce
s
s
in
g
is
s
u
e,
w
h
ich
in
v
o
lv
es
p
r
o
ce
s
s
in
g
w
o
r
d
an
d
s
e
q
u
en
ce
in
p
u
ts
s
im
u
ltan
e
o
u
s
ly
,
c
an
b
e
ef
f
ec
tiv
ely
ad
d
r
ess
e
d
b
y
tr
an
s
f
o
r
m
er
s
[
1
4
]
-
[
1
6
]
.
S
o
p
h
is
tic
at
ed
co
n
t
ex
tu
al
l
an
g
u
ag
e
u
n
d
e
r
s
t
a
n
d
in
g
m
o
d
els
ca
n
b
e
ac
q
u
ir
e
d
,
w
h
ich
co
u
l
d
c
at
ch
s
u
b
tl
e
an
d
d
et
ail
e
d
lex
i
ca
l
p
at
te
r
n
s
.
T
h
is
r
esu
l
ts
in
th
e
d
e
v
elo
p
m
en
t
o
f
a
c
o
m
p
r
eh
en
s
iv
e
f
ea
tu
r
e
r
e
p
r
es
en
tat
io
n
o
f
a
g
iv
en
tex
t
[
1
7
]
.
S
ev
e
r
al
s
ch
o
l
ar
ly
r
ese
a
r
ch
[
1
8
]
-
[
2
1
]
h
av
e
in
v
esti
g
at
e
d
th
e
a
p
p
lic
ati
o
n
o
f
tr
an
s
f
o
r
m
er
s
in
th
e
f
iel
d
o
f
NL
P
,
s
p
e
cif
ic
ally
in
th
e
d
o
m
ain
o
f
s
u
ici
d
e
-
r
el
a
ted
tex
t
i
d
en
tif
i
ca
t
io
n
[
6
]
,
[
2
2
]
,
[
2
3
]
.
Hen
ce
,
th
is
r
esear
ch
m
ak
e
s
s
v
a
lu
ab
le
co
n
tr
ib
u
tio
n
to
th
e
s
u
b
s
eq
u
en
t
d
o
m
ain
s
,
in
cl
u
d
i
n
g
s
an
a
l
y
ze
th
e
s
u
icid
al
id
ea
tio
n
t
ex
t
d
atasets
,
it
is
r
ec
o
m
m
e
n
d
ed
to
u
tili
ze
ad
v
an
ce
d
d
ee
p
lear
n
in
g
m
o
d
els
li
k
e
b
id
ir
ec
tio
n
al
lo
n
g
s
h
o
r
t
-
ter
m
m
e
m
o
r
y
(
B
i
L
ST
M)
,
as
w
ell
as
v
ar
io
u
s
tr
an
s
f
er
lear
n
in
g
m
o
d
els
s
u
c
h
as
b
id
ir
ec
tio
n
al
en
co
d
er
r
ep
r
esen
tatio
n
s
f
r
o
m
tr
an
s
f
o
r
m
er
s
(
B
E
R
T
)
,
r
o
b
u
s
tl
y
o
p
ti
m
ized
B
E
R
T
ap
p
r
o
ac
h
(
R
o
B
E
R
T
a)
,
a
lite
B
E
R
T
(
A
L
B
E
R
T
)
,
d
ec
o
d
in
g
-
e
n
h
an
ce
d
B
E
R
T
w
it
h
d
is
e
n
ta
n
g
led
atte
n
tio
n
(
DeB
E
R
T
a)
.
T
h
is
s
tr
ateg
y
s
ee
k
s
to
in
v
esti
g
ate
t
h
e
ef
f
icac
y
o
f
th
ese
m
o
d
els
i
n
id
en
ti
f
y
i
n
g
s
u
ic
id
al
id
ea
tio
n
.
Fu
r
th
er
m
o
r
e,
th
i
s
r
esear
ch
en
h
a
n
ce
s
to
cr
ea
te
m
o
r
e
s
o
p
h
is
ticat
ed
tr
an
s
f
er
lear
n
in
g
m
o
d
el
s
in
co
n
tr
a
s
t
to
co
n
v
e
n
ti
o
n
al
d
ee
p
lear
n
in
g
m
o
d
el
s
.
T
h
is
b
r
ea
k
th
r
o
u
g
h
i
n
co
r
p
o
r
ates
p
r
e
-
tr
ain
in
g
an
d
f
in
e
-
tu
n
in
g
m
e
th
o
d
o
lo
g
ies
t
h
at
g
r
ea
tl
y
en
h
a
n
ce
th
e
ab
ilit
y
to
d
etec
t
s
u
icid
al
th
o
u
g
h
ts
i
n
wr
itten
lan
g
u
ag
e.
I
n
ad
d
itio
n
,
th
is
s
t
u
d
y
also
co
n
tr
i
b
u
tes
th
a
t
DeB
E
R
T
a
ex
h
ib
its
r
o
b
u
s
t
p
er
f
o
r
m
an
ce
a
n
d
is
o
n
p
ar
w
it
h
co
m
p
ar
ab
le
m
o
d
el
s
in
ter
m
s
o
f
co
m
p
etiti
v
en
es
s
.
2.
RE
L
AT
E
D
WO
R
K
S
R
ec
en
t
s
t
u
d
ies
h
av
e
i
n
v
est
ig
ated
th
e
co
r
r
elatio
n
b
et
w
ee
n
m
en
tal
w
ell
-
b
ei
n
g
a
n
d
th
e
wa
y
p
eo
p
le
ex
p
r
ess
t
h
e
m
s
el
v
e
s
li
n
g
u
i
s
tic
all
y
to
d
etec
t
s
i
g
n
s
o
f
s
u
icid
al
id
ea
tio
n
.
P
r
io
r
r
esear
ch
i
n
co
r
p
o
r
ated
lan
g
u
ag
e
co
m
p
o
n
e
n
t
s
d
er
iv
ed
f
r
o
m
p
s
y
ch
ia
tr
ic
liter
atu
r
e,
in
cl
u
d
in
g
lin
g
u
is
tic
in
q
u
ir
y
a
n
d
w
o
r
d
co
u
n
t
(
L
I
W
C
)
[
2
4
]
,
e
m
o
tio
n
al
c
h
ar
ac
ter
is
tic
s
[
2
5
]
,
an
d
s
u
icid
e
letter
s
[
2
6
]
.
Nev
er
th
eles
s
,
th
e
s
e
m
e
th
o
d
o
lo
g
ie
s
ar
e
co
n
s
tr
ai
n
ed
in
th
eir
ab
ilit
y
to
ass
e
s
s
in
d
i
v
id
u
al
p
o
s
ts
an
d
ar
e
in
ad
eq
u
ate
f
o
r
d
atasets
th
at
ar
e
d
iv
er
s
e
o
r
e
x
ten
s
i
v
e
[
2
7
]
.
T
h
e
u
tili
za
t
io
n
o
f
s
o
cial
m
ed
ia
an
d
NL
P
in
th
e
f
ield
o
f
m
en
ta
l
h
ea
lt
h
r
esear
ch
h
as
ex
p
er
ie
n
c
ed
a
n
o
tab
le
s
u
r
g
e
in
p
o
p
u
lar
it
y
.
T
h
e
s
t
u
d
y
o
f
s
e
n
ti
m
en
t
a
n
al
y
s
is
is
ex
p
a
n
d
in
g
to
in
cl
u
d
e
o
n
li
n
e
m
e
n
tal
h
ea
lt
h
f
o
r
u
m
s
an
d
s
o
cial
m
ed
ia
d
ata
as
n
e
w
ar
ea
s
o
f
in
v
e
s
ti
g
atio
n
.
T
ad
ess
e
et
a
l.
[
2
8
]
d
ev
is
ed
a
h
y
b
r
id
m
o
d
el
th
at
in
te
g
r
ates
laten
t
d
ir
ich
let
allo
ca
tio
n
(
L
D
A
)
,
lin
g
u
i
s
tic
in
q
u
ir
y
a
n
d
w
o
r
d
co
u
n
t
a
n
al
y
s
i
s
(
L
I
W
C
A
)
,
B
ig
r
a
m
,
an
d
m
u
lti
la
y
er
p
er
ce
p
tr
o
n
(
ML
P
)
,
ac
h
iev
i
n
g
a
co
m
m
e
n
d
ab
le
ac
cu
r
ac
y
r
ate
o
f
9
0
%.
A
d
d
itio
n
al
r
esear
ch
[
2
9
]
-
[
3
1
]
g
ath
er
ed
in
f
o
r
m
atio
n
f
r
o
m
T
w
itter
a
n
d
u
s
i
n
g
d
if
f
er
en
t
m
ac
h
in
e
lear
n
i
n
g
tec
h
n
iq
u
es
to
ca
teg
o
r
ize
s
u
i
cid
al
id
ea
tio
n
.
Dee
p
lear
n
in
g
tec
h
n
iq
u
es
s
u
c
h
as
lo
n
g
s
h
o
r
t
-
ter
m
m
e
m
o
r
y
(
L
ST
M)
an
d
co
n
v
o
lu
tio
n
al
n
e
u
r
al
n
et
w
o
r
k
(
C
NN)
h
a
v
e
Evaluation Warning : The document was created with Spire.PDF for Python.
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
20
2
5
:
4
2
6
-
434
428
m
ad
e
s
u
b
s
ta
n
tia
l
p
r
o
g
r
ess
in
NL
P
b
ec
au
s
e
o
f
th
e
w
id
esp
r
e
ad
u
s
e
o
f
w
o
r
d
e
m
b
ed
d
in
g
s
.
M
L
a
p
p
r
o
a
c
h
e
s
a
r
e
c
o
n
s
t
r
a
in
e
d
b
y
r
es
t
r
ic
t
i
o
n
s
s
u
ch
a
s
t
h
e
c
u
r
s
e
o
f
d
im
en
s
i
o
n
a
li
ty
,
d
a
t
a
s
p
a
r
s
i
ty
,
an
d
t
im
e
c
o
n
s
u
m
p
t
i
o
n
,
w
h
ic
h
r
en
d
e
r
t
h
em
i
n
a
p
p
r
o
p
r
i
a
t
e
f
o
r
s
o
m
e
a
p
p
l
i
c
a
t
i
o
n
s
.
D
e
e
p
le
a
r
n
in
g
im
p
r
o
v
e
s
u
p
o
n
c
o
n
v
e
n
t
i
o
n
a
l
m
a
ch
in
e
l
ea
r
n
in
g
m
eth
o
d
s
b
y
ex
t
r
a
ct
in
g
h
ig
h
e
r
-
l
ev
e
l
f
e
a
tu
r
es
f
r
o
m
in
p
u
t
d
a
t
a
u
s
in
g
a
g
r
e
a
t
e
r
n
u
m
b
e
r
o
f
l
ay
e
r
s
in
th
e
m
o
d
e
l
,
r
esu
l
tin
g
in
m
o
r
e
r
e
li
a
b
l
e
an
d
p
r
e
ci
s
e
c
l
ass
if
i
c
at
i
o
n
.
Stu
d
i
e
s
in
d
i
c
at
e
th
at
d
e
e
p
l
e
a
r
n
in
g
m
o
d
e
ls
h
av
e
s
u
p
e
r
i
o
r
p
r
e
d
i
c
t
i
o
n
a
c
c
u
r
ac
y
in
i
d
en
t
if
y
in
g
s
u
i
c
i
d
al
i
d
e
a
t
i
o
n
w
h
en
c
o
m
p
a
r
e
d
t
o
o
th
e
r
m
a
ch
in
e
l
e
a
r
n
in
g
c
l
ass
if
i
e
r
s
[
3
2
]
,
[
3
3
]
.
P
r
esen
tl
y
,
n
u
m
er
o
u
s
r
esear
ch
er
s
ar
e
u
tili
zi
n
g
B
E
R
T
-
b
ased
m
o
d
el
s
[
2
3
]
to
id
en
tify
s
u
icid
al
id
ea
tio
n
[
3
4
]
,
as
th
ese
m
o
d
els
h
a
v
e
t
h
e
ab
ilit
y
to
p
r
ec
is
el
y
ca
p
tu
r
e
s
e
m
an
t
ic
an
d
co
n
tex
t
u
al
ch
ar
ac
ter
is
tic
s
[
3
5
]
.
Nev
er
th
e
less
,
s
u
icid
e
d
etec
tio
n
r
esear
ch
is
co
n
s
tr
ai
n
ed
b
y
i
ts
e
m
p
h
asi
s
o
n
T
w
itter
,
a
p
latf
o
r
m
t
h
at
r
estrict
s
w
r
iti
n
g
to
a
m
ax
i
m
u
m
o
f
2
8
0
ch
ar
ac
ter
s
.
P
r
ev
io
u
s
r
esear
ch
e
m
p
lo
y
i
n
g
tr
an
s
f
o
r
m
er
m
o
d
els
[
6
]
em
p
lo
y
ed
s
u
icid
e
-
r
elate
d
d
ata
f
r
o
m
T
w
itter
as
th
e
d
ataset.
T
h
e
m
o
d
els
u
til
ized
in
th
e
s
e
i
n
v
e
s
ti
g
a
tio
n
s
w
er
e
B
E
R
T
,
Dis
tilB
E
R
T
,
AL
B
E
R
T
,
R
o
B
E
R
T
a,
an
d
Dis
tilR
o
B
E
R
T
a.
A
m
o
n
g
th
e
s
e
m
o
d
els,
R
o
B
E
R
T
a
d
em
o
n
s
tr
ated
th
e
h
ig
h
e
s
t
lev
el
o
f
ac
cu
r
ac
y
,
at
tain
i
n
g
a
s
co
r
e
o
f
9
5
.
3
9
%.
A
d
i
f
f
er
en
t
s
t
u
d
y
[
3
6
]
em
p
lo
y
ed
a
m
er
g
ed
R
o
B
E
R
T
a
an
d
C
NN
m
o
d
el,
w
it
h
a
d
atas
et
co
n
s
is
t
in
g
o
f
1
1
0
,
0
4
0
d
ata
p
o
in
ts
.
T
h
e
co
m
p
o
s
ite
m
o
d
el
attain
ed
an
F1
-
s
co
r
e
o
f
9
6
.
8
1
%.
T
h
is
w
o
r
k
tr
ies
to
cr
ea
te
a
m
o
d
if
ied
v
er
s
io
n
o
f
t
h
e
B
E
R
T
m
o
d
el
th
at
ad
d
r
ess
e
s
th
e
li
m
i
ts
o
f
h
u
g
e
d
atasets
an
d
f
o
c
u
s
e
s
o
n
co
m
b
in
i
n
g
s
ev
er
al
m
o
d
el
s
r
ath
er
th
an
m
o
d
i
f
y
i
n
g
th
e
B
E
R
T
m
o
d
el
its
el
f
.
T
h
is
s
tu
d
y
ai
m
s
to
co
n
s
tr
u
c
t
m
o
d
els
u
s
in
g
v
ar
io
u
s
B
E
R
T
v
ar
iatio
n
s
,
in
cl
u
d
in
g
B
E
R
T
,
R
o
B
E
R
T
a,
A
L
B
E
R
T
,
an
d
DeB
E
R
T
a.
No
ta
b
ly
,
t
h
e
DeB
E
R
T
a
m
o
d
el
h
as
n
o
t b
ee
n
ex
a
m
i
n
ed
in
p
r
ev
io
u
s
s
t
u
d
ies.
3.
M
E
T
H
O
D
T
h
e
in
v
esti
g
atio
n
co
m
m
en
ce
d
th
r
o
u
g
h
s
e
v
er
al
m
et
h
o
d
o
lo
g
ical
s
tag
es,
as
ill
u
s
tr
ated
in
Fi
g
u
r
e
1
.
T
h
e
d
ataset
in
cl
u
d
ed
of
tex
ts
p
er
tain
in
g
to
s
u
ic
id
e
an
d
n
o
n
-
s
u
icid
e.
T
h
e
in
itial st
ep
in
v
o
lv
ed
p
r
ep
r
o
ce
s
s
in
g
t
h
e
tex
t
b
y
e
li
m
in
a
tin
g
s
p
ec
ial
ch
ar
ac
ter
s
,
n
u
m
er
ical
v
al
u
e
s
,
an
d
p
u
n
ct
u
atio
n
m
ar
k
s
,
f
o
llo
w
ed
b
y
co
n
v
er
ti
n
g
i
t
to
lo
w
er
ca
s
e.
W
e
co
n
d
u
cted
le
m
m
a
tizatio
n
to
r
ed
u
ce
w
o
r
d
s
to
th
eir
b
ase
f
o
r
m
.
W
e
u
tili
ze
d
t
h
e
p
r
o
ce
s
s
ed
d
ata
in
p
r
e
-
tr
ain
ed
m
o
d
els
s
u
ch
a
s
B
E
R
T
,
R
o
B
E
R
T
a
,
AL
B
E
R
T
,
an
d
DeB
E
R
T
a,
ad
ap
tin
g
th
e
m
to
s
p
ec
if
ic
a
n
al
y
s
i
s
task
s
v
ia
tr
a
n
s
f
er
lear
n
i
n
g
.
W
e
ca
teg
o
r
ized
th
e
o
u
tco
m
e
s
to
an
ticip
ate
d
ata
cla
s
s
i
f
icat
io
n
s
.
W
e
ass
ess
ed
th
e
ac
cu
r
ac
y
o
f
t
h
e
m
o
d
el
b
y
u
ti
liz
in
g
a
co
n
f
u
s
io
n
m
atr
i
x
an
d
p
er
f
o
r
m
ed
m
o
d
el
co
m
p
ar
is
o
n
s
to
d
eter
m
in
e
t
h
e
m
o
s
t
id
ea
l
p
er
f
o
r
m
a
n
ce
.
T
h
is
ap
p
r
o
ac
h
in
te
g
r
ates
te
x
t
p
r
o
ce
s
s
in
g
m
eth
o
d
o
lo
g
ies
w
i
th
d
ee
p
lear
n
in
g
alg
o
r
it
h
m
s
to
ac
h
iev
e
e
f
f
icie
n
t a
n
a
l
y
s
is
.
Fig
u
r
e
1
.
S
y
s
te
m
ar
ch
itect
u
r
e
o
f
th
e
p
r
o
p
o
s
ed
w
o
r
k
3
.
1
.
P
re
pro
ce
s
s
ing
Pr
e
p
r
o
c
e
s
s
in
g
i
s
a
c
r
i
ti
c
a
l
s
t
ag
e
i
n
en
s
u
r
in
g
th
a
t
th
e
t
ex
t
u
a
l
d
a
t
a
i
s
in
o
p
t
im
a
l
f
o
r
m
f
o
r
a
n
a
ly
s
i
s
an
d
m
o
d
e
l
t
r
a
in
in
g
.
I
n
th
is
s
tu
d
y
,
w
e
b
e
g
a
n
b
y
r
em
o
v
in
g
u
n
n
e
c
ess
a
r
y
e
l
em
en
ts
s
u
ch
a
s
H
T
M
L
ta
g
s
,
s
p
e
c
i
a
l
ch
a
r
a
ct
e
r
s
,
n
u
m
e
r
a
ls
,
an
d
p
u
n
ct
u
a
t
i
o
n
m
ar
k
s
.
T
h
i
s
w
a
s
n
e
ce
s
s
a
r
y
t
o
e
lim
in
at
e
ex
t
r
an
e
o
u
s
s
y
m
b
o
l
s
th
a
t
d
o
n
o
t
c
o
n
t
r
i
b
u
t
e
m
e
an
in
g
f
u
lly
t
o
th
e
l
in
g
u
i
s
t
i
c
s
t
r
u
ct
u
r
e
,
en
s
u
r
i
n
g
th
a
t
th
e
m
o
d
e
l
s
c
o
u
l
d
f
o
cu
s
o
n
es
s
en
t
i
a
l
t
ex
tu
a
l
p
a
t
te
r
n
s
w
i
th
o
u
t
b
e
i
n
g
d
is
t
r
a
ct
e
d
b
y
i
r
r
e
l
ev
an
t
n
o
i
s
e
[
3
7
]
.
A
f
t
e
r
c
l
e
an
in
g
th
e
d
a
t
as
et
o
f
u
n
n
e
c
es
s
a
r
y
ch
a
r
a
ct
e
r
s
,
t
h
e
t
ex
t
w
as
s
t
an
d
a
r
d
i
z
e
d
b
y
c
o
n
v
e
r
tin
g
a
l
l
w
o
r
d
s
t
o
l
o
w
e
r
ca
s
e
.
T
h
is
s
t
e
p
p
l
ay
s
a
s
ig
n
if
i
c
an
t
r
o
l
e
i
n
r
e
d
u
cin
g
r
e
d
u
n
d
a
n
cy
c
au
s
e
d
b
y
c
a
s
e
d
i
f
f
e
r
en
c
es
,
e
s
p
e
c
ia
l
ly
f
o
r
l
a
n
g
u
ag
es
li
k
e
E
n
g
l
is
h
,
w
h
e
r
e
c
a
p
i
t
a
li
z
a
ti
o
n
d
o
e
s
n
o
t
ch
an
g
e
t
h
e
m
e
an
in
g
o
f
w
o
r
d
s
[
3
8
]
.
T
h
i
s
s
t
an
d
a
r
d
i
z
at
i
o
n
h
e
l
p
s
in
m
in
im
iz
in
g
d
i
s
c
r
e
p
an
ci
e
s
b
e
t
w
e
en
w
o
r
d
s
lik
e
“
Su
ic
i
d
e
”
a
n
d
“
s
u
i
c
i
d
e
,
”
t
r
e
a
ti
n
g
th
em
a
s
i
d
en
t
i
ca
l
d
u
r
in
g
m
o
d
e
l
t
r
a
in
in
g
[
3
9
]
.
S
u
b
s
e
q
u
en
t
ly
,
s
t
o
p
w
o
r
d
r
em
o
v
a
l
w
as
c
o
n
d
u
ct
e
d
.
St
o
p
w
o
r
d
s
a
r
e
c
o
m
m
o
n
w
o
r
d
s
,
s
u
ch
as
“
t
h
e
,
”
“
i
s
,
”
an
d
“
in
,
”
w
h
i
ch
,
w
h
i
l
e
es
s
e
n
t
i
al
f
o
r
g
r
am
m
a
t
ic
a
l
c
o
r
r
e
c
tn
es
s
,
d
o
n
o
t
c
a
r
r
y
s
i
g
n
i
f
i
c
an
t
m
e
an
in
g
f
o
r
t
e
x
t
c
la
s
s
if
ic
a
t
i
o
n
Sui
ci
d
e
a
n
d
No
n
-
Sui
ci
d
e
Da
t
a
se
t
Con
f
usi
on
Ma
t
r
i
x
M
od
el
Comparison
R
e
m
ov
e
Spe
ci
a
l
Ch
a
r
a
ct
e
r
a
n
d
Di
g
i
t
R
e
m
ov
e
Pun
ct
u
a
ti
on
L
ow
e
r
T
e
xt
L
e
m
m
a
ti
z
e
BER
T
R
oBER
T
A
Albert
DeB
ER
T
a
Clas
s
ifier
BER
T
R
oBER
T
A
Al
bert
DeB
ER
T
a
Pr
e
-
T
r
ai
ned
Model
Pr
e
-
P
r
ocess
i
ng
T
r
an
s
f
er
Learnin
g
Model
Evaluation Warning : The document was created with Spire.PDF for Python.
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
A
co
mp
a
r
a
tive
a
n
a
lysi
s
o
f tra
n
s
fer lea
r
n
in
g
mo
d
els o
n
s
u
ici
d
e
a
n
d
n
o
n
-
s
u
icid
e
textu
a
l
…
(
Me
r
in
d
a
Les
ta
n
d
y
)
429
t
a
s
k
s
.
B
y
r
em
o
v
in
g
t
h
e
s
e
s
t
o
p
w
o
r
d
s
,
w
e
e
n
s
u
r
e
th
at
th
e
m
o
d
e
l
c
o
n
c
en
t
r
a
t
es
o
n
th
e
m
o
r
e
c
r
it
i
c
a
l
a
n
d
in
f
o
r
m
at
iv
e
w
o
r
d
s
,
th
u
s
en
h
an
c
in
g
i
ts
p
e
r
f
o
r
m
an
c
e
[
4
0
]
.
A
d
d
i
t
i
o
n
al
ly
,
th
i
s
s
t
e
p
r
e
d
u
c
e
s
t
h
e
d
im
en
s
i
o
n
a
l
i
ty
o
f
th
e
d
a
t
as
e
t
,
a
l
l
o
w
in
g
f
o
r
m
o
r
e
ef
f
i
c
i
en
t
c
o
m
p
u
t
a
ti
o
n
[
4
1
]
.
T
h
e
f
in
a
l
s
t
e
p
in
p
r
e
p
r
o
c
es
s
i
n
g
i
n
v
o
lv
e
d
l
em
m
at
i
z
at
i
o
n
,
w
h
e
r
e
w
o
r
d
s
a
r
e
r
e
d
u
ce
d
t
o
th
e
i
r
b
a
s
e
o
r
r
o
o
t
fo
r
m
s
.
T
h
is
p
r
o
c
es
s
en
s
u
r
es
t
h
at
w
o
r
d
s
l
ik
e
“
r
u
n
n
in
g
,
”
“
r
u
n
s
,
”
a
n
d
“
r
a
n
”
a
r
e
t
r
e
at
e
d
as
“
r
u
n
,
”
p
r
e
v
en
ti
n
g
th
e
m
o
d
el
f
r
o
m
t
r
e
at
in
g
th
em
a
s
s
e
p
a
r
at
e
e
n
t
i
ti
e
s
.
L
em
m
a
ti
z
a
t
i
o
n
th
u
s
a
i
d
s
in
c
a
p
tu
r
in
g
th
e
c
o
r
e
m
e
an
in
g
o
f
w
o
r
d
s
an
d
im
p
r
o
v
e
s
th
e
m
o
d
e
l'
s
a
b
i
li
ty
t
o
g
e
n
e
r
a
li
z
e
p
a
tt
e
r
n
s
f
r
o
m
th
e
d
at
a
[
4
2
]
.
3
.
2
.
T
ra
nfo
rm
er
B
E
R
T
,
as
d
escr
ib
e
d
in
th
e
lite
r
atu
r
e
[
1
5
]
,
is
a
n
o
tab
le
b
r
ea
k
th
r
o
u
g
h
i
n
th
e
f
ield
o
f
N
L
P
,
o
w
i
n
g
to
its
u
tili
za
t
io
n
o
f
th
e
tr
a
n
s
f
er
lear
n
in
g
m
o
d
el
.
T
h
e
m
o
d
el
co
n
s
is
t
s
o
f
a
s
eq
u
e
n
ce
o
f
tr
an
s
f
o
r
m
er
en
co
d
er
la
y
er
s
,
an
d
it
o
f
f
er
s
s
e
v
er
al
ad
v
a
n
tag
e
s
,
s
u
c
h
as
a
co
h
esi
v
e
ar
ch
itec
tu
r
e
f
o
r
d
iv
er
s
e
ap
p
licatio
n
s
a
n
d
b
id
ir
ec
tio
n
al
p
r
e
-
tr
ain
i
n
g
.
T
h
e
p
r
i
m
ar
y
tas
k
o
f
s
elf
-
s
u
p
er
v
i
s
ed
p
r
e
-
tr
ain
i
n
g
,
r
ef
er
r
ed
to
as
m
a
s
k
ed
lan
g
u
a
g
e
m
o
d
elin
g
(
ML
M)
,
p
lay
s
a
cr
u
cial
r
o
le
in
e
n
ab
li
n
g
b
id
ir
ec
tio
n
al
it
y
i
n
s
id
e
th
e
m
o
d
el.
I
t
in
v
o
l
v
es
p
r
ed
ictin
g
m
a
s
k
ed
w
o
r
d
s
in
u
n
lab
eled
tex
t
b
y
co
n
s
id
er
in
g
b
o
th
th
e
p
r
ec
ed
in
g
an
d
s
u
cc
ee
d
in
g
co
n
te
x
t w
o
r
d
s
co
n
cu
r
r
en
t
l
y
.
I
n
ad
d
itio
n
to
its
p
r
im
ar
y
p
r
e
-
tr
ai
n
i
n
g
o
b
j
ec
tiv
e,
B
E
R
T
h
as
also
b
ee
n
eq
u
i
p
p
ed
w
i
th
t
h
e
ca
p
ab
ilit
y
to
ef
f
ec
tiv
e
l
y
p
r
o
ce
s
s
n
u
m
er
o
u
s
s
eq
u
e
n
ce
s
b
y
m
ea
n
s
o
f
th
e
n
e
x
t
s
e
n
te
n
ce
p
r
ed
ictio
n
(
NSP
)
w
o
r
k
.
T
h
e
NSP
task
is
d
esig
n
ed
to
d
is
ce
r
n
w
h
et
h
er
a
g
iv
en
s
eq
u
e
n
ce
s
er
v
es
as
a
co
n
tin
u
atio
n
o
f
an
o
t
h
er
s
eq
u
en
ce
.
Nu
m
er
o
u
s
en
h
an
ce
m
en
ts
h
a
v
e
b
ee
n
d
ev
is
ed
af
ter
th
e
in
tr
o
d
u
ctio
n
o
f
B
E
R
T
.
R
o
B
E
R
T
a
[
4
3
]
is
a
r
ef
in
ed
v
er
s
io
n
o
f
B
E
R
T
,
a
p
o
p
u
lar
lan
g
u
ag
e
m
o
d
el,
w
h
i
ch
i
n
co
r
p
o
r
ates
a
m
o
r
e
e
f
f
icien
t
tr
ai
n
i
n
g
m
et
h
o
d
o
lo
g
y
.
T
o
im
p
r
o
v
e
co
n
tex
t
u
al
u
n
d
er
s
ta
n
d
in
g
i
n
th
e
tex
t
g
e
n
er
ated
b
y
t
h
e
m
o
d
el,
R
o
B
E
R
T
a
eli
m
i
n
ates
th
e
N
SP
task
.
A
d
d
itio
n
all
y
,
s
h
e
e
n
h
an
ce
s
t
h
e
tr
ai
n
i
n
g
p
r
o
ce
s
s
b
y
au
g
m
en
tin
g
s
en
ten
ce
p
air
in
g
s
t
h
r
o
u
g
h
r
a
n
d
o
m
m
o
d
if
icatio
n
s
u
s
i
n
g
a
lar
g
e
co
r
p
u
s
o
f
tex
t
d
ata.
I
n
ad
d
itio
n
,
R
o
B
E
R
T
a
em
p
lo
y
s
p
r
o
lo
n
g
ed
tr
ain
in
g
d
u
r
atio
n
s
a
n
d
in
cr
ea
s
ed
b
atch
s
ize
s
to
i
m
p
r
o
v
e
th
e
p
er
f
o
r
m
a
n
ce
o
f
t
h
e
m
o
d
el.
I
n
th
e
y
ea
r
2
0
1
9
,
a
tr
an
s
f
o
r
m
e
r
-
b
ased
m
o
d
el
ca
lled
A
L
B
E
R
T
[
4
4
]
w
a
s
cr
ea
ted
,
f
ea
tu
r
in
g
a
r
ed
u
ctio
n
in
its
p
ar
a
m
eter
co
u
n
t.
T
h
e
o
b
j
ec
tiv
e
o
f
th
is
B
E
R
T
a
d
d
itio
n
is
to
en
h
a
n
ce
ef
f
icie
n
c
y
w
h
ile
m
ain
tain
in
g
th
e
q
u
alit
y
o
f
li
n
g
u
is
tic
co
n
tex
t
c
o
m
p
r
e
h
en
s
io
n
.
AL
B
E
R
T
e
m
p
lo
y
s
a
p
ar
a
m
eter
s
h
ar
i
n
g
tech
n
iq
u
e
a
m
o
n
g
la
y
er
s
in
th
e
tr
an
s
f
o
r
m
er
ar
ch
itect
u
r
e,
r
esu
ltin
g
i
n
a
r
ed
u
ctio
n
in
th
e
n
u
m
b
er
o
f
p
ar
a
m
eter
s
t
h
at
n
e
ed
to
b
e
tr
ain
ed
.
B
y
e
m
p
lo
y
i
n
g
t
h
i
s
m
e
th
o
d
o
lo
g
y
,
AL
B
E
R
T
w
a
s
ab
le
to
attain
s
i
m
ilar
,
if
n
o
t b
etter
,
r
esu
lt
s
co
m
p
ar
ed
to
B
E
R
T
,
all
w
h
ile
u
tili
zi
n
g
a
s
m
al
ler
n
u
m
b
er
o
f
p
ar
am
eter
s
.
T
h
e
y
ea
r
2
0
2
0
w
it
n
ess
ed
t
h
e
e
m
er
g
en
c
e
o
f
a
n
o
v
el
m
o
d
el
k
n
o
w
n
as
DeB
E
R
T
a
[
4
5
]
,
w
h
ich
i
n
tr
o
d
u
ce
d
m
o
d
if
icatio
n
s
t
o
th
e
atte
n
tio
n
m
ec
h
an
is
m
i
n
s
id
e
th
e
tr
an
s
f
o
r
m
er
ar
ch
itect
u
r
e,
h
en
ce
e
n
h
a
n
ci
n
g
th
e
ca
p
ab
ilit
ies
o
f
tr
a
n
s
f
o
r
m
e
r
-
b
ased
lan
g
u
a
g
e
m
o
d
el
s
.
T
h
e
DeB
E
R
T
a
m
o
d
el
e
m
p
lo
y
s
a
d
is
en
tan
g
led
atten
t
i
o
n
m
ec
h
a
n
i
s
m
t
h
at
ef
f
ec
ti
v
el
y
d
iv
id
es
atten
tio
n
in
to
t
w
o
co
m
p
o
n
en
ts
:
co
n
te
n
t
-
b
ased
atten
tio
n
an
d
p
o
s
itio
n
-
b
ased
atten
tio
n
.
T
h
is
m
et
h
o
d
o
lo
g
y
f
ac
ilit
a
tes
t
h
e
r
eso
l
u
tio
n
o
f
ce
r
tain
o
b
s
tacle
s
en
co
u
n
ter
ed
b
y
tr
an
s
f
o
r
m
er
-
b
ased
m
o
d
els,
i
n
clu
d
i
n
g
th
e
r
elian
ce
o
n
in
f
le
x
ib
le
w
o
r
d
s
eq
u
en
ci
n
g
a
n
d
th
e
in
ca
p
ac
it
y
to
co
m
p
r
eh
en
d
n
o
n
li
n
ea
r
ass
o
ciatio
n
s
a
m
o
n
g
wo
r
d
s
.
B
y
in
co
r
p
o
r
atin
g
th
e
d
is
en
ta
n
g
led
atten
tio
n
m
ec
h
a
n
i
s
m
,
DeB
E
R
T
a
d
em
o
n
s
tr
ates
e
n
h
a
n
ce
d
te
x
t
r
ep
r
esen
t
atio
n
s
a
n
d
i
m
p
r
o
v
ed
p
er
f
o
r
m
a
n
ce
i
n
m
an
y
n
at
u
r
al
lan
g
u
a
g
es p
r
o
ce
s
s
i
n
g
tas
k
s
,
i
n
clu
d
in
g
te
x
t c
o
m
p
r
eh
e
n
s
io
n
,
s
en
ti
m
en
t a
n
al
y
s
is
,
an
d
clas
s
i
f
i
ca
tio
n
tas
k
s
.
3
.
3
.
E
v
a
lua
t
i
o
n
ma
t
rix
T
h
e
ev
alu
atio
n
o
f
th
e
m
o
d
el
will
b
e
co
n
d
u
cted
u
s
in
g
a
s
ep
ar
ate
s
et
o
f
test
in
g
d
ata,
an
d
th
e
r
esu
lt
s
w
ill
b
e
g
iv
en
i
n
th
e
f
o
r
m
o
f
a
co
n
f
u
s
io
n
m
atr
i
x
.
T
h
e
p
er
f
o
r
m
a
n
ce
o
f
th
is
al
g
o
r
ith
m
w
ill
b
e
ev
alu
a
ted
b
ased
o
n
m
etr
ics
s
u
c
h
as
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
an
d
F
-
1
s
co
r
e.
T
h
e
m
ea
s
u
r
e
o
f
ac
cu
r
ac
y
i
s
d
eter
m
in
ed
b
y
th
e
d
eg
r
ee
o
f
p
r
o
x
i
m
i
t
y
b
et
w
ee
n
t
h
e
p
r
o
j
ec
ted
v
alu
e
an
d
t
h
e
ac
t
u
al
v
al
u
e,
as
s
h
o
w
n
i
n
(
1
)
.
T
h
e
ter
m
“
tr
u
e
p
o
s
iti
v
e
”
(
T
P
)
is
u
s
ed
to
d
escr
i
b
e
in
s
tan
ce
s
w
h
e
n
p
o
s
itiv
e
d
ata
is
ac
cu
r
atel
y
p
r
ed
icted
,
w
h
ile
“
tr
u
e
n
e
g
at
iv
e
”
(
T
N)
is
u
s
ed
to
d
escr
ib
e
in
s
ta
n
ce
s
w
h
er
e
n
e
g
at
iv
e
d
ata
is
ac
c
u
r
atel
y
p
r
ed
icted
.
T
h
e
ter
m
“
f
alse
p
o
s
it
iv
e
”
(
FP
)
r
ef
er
s
to
in
s
tan
ce
s
w
h
er
e
n
e
g
ati
v
e
d
ata
is
w
r
o
n
g
l
y
c
lass
if
ied
as
p
o
s
iti
v
e
d
ata,
w
h
ile
“
f
alse
n
eg
at
iv
e
”
(
FN)
d
en
o
tes
ca
s
e
s
w
h
er
e
p
o
s
itiv
e
d
ata
is
i
n
co
r
r
ec
tl
y
cl
ass
i
f
ied
as
n
eg
at
iv
e
d
ata.
In
(
2
)
r
ep
r
esen
ts
p
r
ec
is
io
n
,
w
h
ich
is
a
m
e
tr
ic
u
s
ed
to
ass
es
s
th
e
ac
c
u
r
ac
y
o
f
d
ata
b
y
ev
alu
a
tin
g
t
h
e
p
r
o
p
o
r
tio
n
o
f
co
r
r
ec
t p
r
e
d
ictio
n
s
.
=
+
+
+
+
(
1
)
P
r
ec
is
io
n
is
a
m
ea
s
u
r
e
th
at
q
u
an
ti
f
ies
t
h
e
d
eg
r
ee
o
f
p
r
ec
is
io
n
ex
h
ib
ited
b
y
d
ata,
co
n
s
id
er
in
g
b
o
th
th
e
v
er
ac
it
y
o
f
th
e
i
n
f
o
r
m
atio
n
an
d
its
ab
ilit
y
to
m
a
k
e
ac
cu
r
ate
p
r
e
d
ictio
n
s
.
In
(
2
)
is
u
tili
ze
d
to
ar
t
icu
late
th
e
co
n
ce
p
t
o
f
p
r
ec
is
io
n
.
=
+
(
2
)
Evaluation Warning : The document was created with Spire.PDF for Python.
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
20
2
5
:
4
2
6
-
434
430
T
h
e
ef
f
ec
ti
v
e
n
es
s
o
f
t
h
e
m
o
d
e
l
in
ac
c
u
r
atel
y
id
e
n
ti
f
y
in
g
a
c
er
tain
clas
s
ca
n
b
e
e
v
alu
a
ted
b
y
i
ts
r
ec
all
m
etr
ic,
w
h
ic
h
ca
n
b
e
co
m
p
u
ted
u
s
i
n
g
(
3
)
.
=
+
(
3
)
T
h
e
F1
-
s
co
r
e
is
a
m
etr
ic
t
h
at
i
n
teg
r
ate
s
t
w
o
f
u
n
d
a
m
en
ta
l
id
e
as
in
m
o
d
el
e
v
alu
a
tio
n
,
n
a
m
e
l
y
p
r
ec
is
io
n
an
d
r
ec
all.
P
r
ec
is
io
n
ass
e
s
s
es
th
e
d
eg
r
ee
o
f
ac
c
u
r
ac
y
i
n
t
h
e
d
ata
p
r
o
j
ec
te
d
b
y
th
e
m
o
d
el,
w
h
er
ea
s
r
ec
all
ev
alu
a
tes
th
e
ef
f
icac
y
o
f
th
e
m
o
d
el
in
co
r
r
ec
tl
y
r
ec
o
g
n
izi
n
g
a
s
p
ec
if
ic
class
.
T
h
e
F1
-
s
co
r
e
u
s
e
(
4
)
to
ev
alu
ate
th
e
o
v
er
all
e
f
f
ec
t
iv
e
n
es
s
o
f
t
h
e
m
o
d
el
in
ac
cu
r
atel
y
p
r
ed
ictin
g
o
u
tco
m
e
s
.
1
−
=
2
(
×
)
(
+
)
(
4
)
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
4
.
1
.
Da
t
a
s
et
T
ab
le
1
s
h
o
w
s
t
h
e
d
ataset
th
at
w
as
ac
q
u
ir
ed
f
r
o
m
th
e
s
o
cial
m
ed
ia
n
et
w
o
r
k
R
ed
d
it
an
d
h
as
b
ee
n
ca
teg
o
r
ized
in
to
t
w
o
d
is
t
in
ct
l
ab
els:
lab
el
1
d
en
o
tes
i
n
s
ta
n
ce
s
o
f
s
u
icid
e,
w
h
ile
lab
el
0
r
ep
r
esen
t
s
n
o
n
-
s
u
icid
al
o
cc
u
r
r
en
ce
s
.
T
h
e
d
ataset
u
tili
ze
d
in
th
is
s
t
u
d
y
co
n
s
i
s
ts
o
f
t
wo
d
is
tin
ct
s
u
b
r
ed
d
its
.
T
h
e
d
ata
co
llectio
n
p
r
o
ce
s
s
in
v
o
l
v
ed
u
til
izin
g
t
h
e
P
u
s
h
s
h
if
t
A
P
I
to
r
etr
iev
e
p
o
s
ts
f
r
o
m
th
e
Su
icid
eW
atch
s
u
b
r
ed
d
it,
s
p
an
n
i
n
g
th
e
ti
m
e
f
r
o
m
Dec
e
m
b
er
1
6
th
,
2
0
0
8
,
to
J
an
u
ar
y
2
nd
,
2
0
2
1
.
T
h
e
d
ata
co
llecti
o
n
p
r
o
ce
s
s
in
v
o
l
v
ed
u
tili
zi
n
g
th
e
P
u
s
h
s
h
if
t A
P
I
to
r
etr
iev
e
p
o
s
ts
f
r
o
m
t
h
e
S
u
icid
eW
atch
s
u
b
r
ed
d
it,
en
co
m
p
ass
i
n
g
th
e
t
i
m
e
s
p
an
n
i
n
g
f
r
o
m
De
ce
m
b
er
1
6
th
,
2
0
0
8
,
to
J
an
u
ar
y
2
nd
,
2
0
2
1
.
A
co
llec
t
io
n
o
f
p
o
s
t
s
p
er
tain
i
n
g
to
n
o
n
-
s
u
icid
al
s
u
b
j
ec
ts
w
a
s
co
m
p
ile
d
f
r
o
m
t
h
e
s
u
b
r
ed
d
it
r
/teen
ag
er
s
.
T
h
e
d
ataset
co
m
p
r
is
es
a
to
tal
o
f
2
3
2
,
0
7
4
en
tr
ies,
w
h
ic
h
ar
e
ca
teg
o
r
ized
in
t
o
t
w
o
g
r
o
u
p
s
:
n
o
n
-
s
u
icid
e
d
ata
lab
els
(
1
1
6
,
0
3
7
e
n
tr
ies)
a
n
d
s
u
icid
e
d
ata
lab
el
s
(
1
1
6
,
0
3
7
en
tr
ies).
T
h
e
f
o
llo
w
i
n
g
tab
le
p
r
o
v
id
es
a
n
illu
s
tr
atio
n
o
f
b
o
th
t
h
e
d
ataset
s
p
er
tain
in
g
to
s
u
icid
e
an
d
n
o
n
-
s
u
icid
e
ca
s
es.
T
ab
le
1
.
E
x
am
p
le
r
es
u
lt
s
o
f
n
o
n
-
s
u
icid
e
an
d
s
u
icid
e
d
atasets
T
e
x
t
L
a
b
e
l
M
y
a
p
o
l
o
g
i
e
s
t
o
t
h
e
r
a
n
d
o
m
d
u
d
e
o
n
A
mo
n
g
U
s
S
o
u
h
h
.
.
.
.
I
w
o
u
l
d
l
i
k
e
t
o
a
p
o
l
o
g
i
z
e
t
o
t
h
e
r
a
n
d
o
m
d
u
d
e
n
a
me
d
P
o
n
o
n
a
mo
n
g
u
s
.
I
h
a
d
h
o
st
e
d
a
g
a
me
f
o
r
my
g
f
a
n
d
h
e
r
f
r
i
e
n
d
t
o
l
e
a
r
n
h
o
w
t
o
p
l
a
y
a
n
d
w
h
a
t
e
v
e
r
a
n
d
.
.
.
w
e
ma
y
h
a
v
e
t
a
k
e
n
o
v
e
r
t
h
e
c
h
a
t
w
i
t
h
U
w
U
s
a
n
d
u
h
.
.
.
.
o
t
h
e
r
q
u
e
st
i
o
n
a
b
l
e
l
a
n
g
u
a
g
e
.
.
.
.
*
c
o
u
g
h
*
I
w
a
s
c
a
l
l
e
d
“
mo
mm
y
y
e
l
l
o
w
”
*
c
o
u
g
h
*
A
N
Y
W
A
Y
S
so
r
r
y
y
a
h
a
d
t
o
w
i
t
n
e
ss
t
h
a
t
d
u
d
e
.
0
A
s
my
i
s
o
l
a
t
i
o
n
g
r
o
w
s
l
o
n
g
e
r
,
t
h
o
u
g
h
t
s
o
f
m
u
r
d
e
r
a
n
d
r
a
p
e
e
n
t
e
r
m
y
m
i
n
d
.
T
h
e
h
u
m
a
n
s
I
s
e
e
e
v
e
r
y
d
a
y
s
t
a
r
t
t
o
l
o
o
k
l
e
s
s
a
n
d
l
e
s
s
l
i
k
e
t
h
e
s
a
me
s
p
e
c
i
e
s
a
s
me
.
T
h
e
y
s
t
a
r
t
t
o
f
e
e
l
l
i
k
e
j
u
s
t
a
n
o
t
h
e
r
a
n
i
m
a
l
.
I
st
a
r
t
t
o
f
a
n
t
a
s
i
z
e
a
b
o
u
t
h
o
w
I
c
o
u
l
d
c
o
r
n
e
r
o
n
e
o
f
t
h
e
m
a
n
d
a
t
t
a
c
k
t
h
e
m
.
S
a
t
i
s
f
y
so
m
e
o
f
my
c
a
r
n
a
l
n
e
e
d
s
.
I
d
o
n
’
t
w
a
n
t
t
o
b
e
t
h
i
s
w
a
y
,
o
r
h
a
v
e
t
h
e
s
e
t
h
o
u
g
h
t
s
.
B
u
t
t
h
e
l
e
s
s
r
e
a
l
a
f
f
e
c
t
i
o
n
I
r
e
c
e
i
v
e
,
t
h
e
m
o
r
e
a
t
t
r
a
c
t
i
v
e
a
n
d
c
o
m
f
o
r
t
i
n
g
t
h
e
s
e
t
h
o
u
g
h
t
s
s
t
a
r
t
t
o
b
e
.
W
h
y
w
o
u
l
d
s
o
me
b
o
d
y
w
a
n
t
m
e
a
l
i
v
e
?
I
’
m
l
i
k
e
a
d
a
n
g
e
r
o
u
s
a
n
i
m
a
l
.
I
w
i
s
h
s
o
m
e
b
o
d
y
w
o
u
l
d
c
o
m
e
a
n
d
k
i
l
l
m
e
.
I
w
o
u
l
d
b
e
b
e
t
t
e
r
o
f
f
d
e
a
d
I
’
m
s
u
r
e
.
A
n
d
a
n
y
b
o
d
y
I
’
m
a
r
o
u
n
d
w
o
u
l
d
b
e
b
e
t
t
e
r
o
f
f
t
o
o
.
M
y
u
r
g
e
s
t
o
k
i
l
l
my
s
e
l
f
h
a
v
e
a
l
w
a
y
s
b
e
e
n
s
t
r
o
n
g
e
r
t
h
a
n
m
y
u
r
g
e
s
t
o
h
u
r
t
o
t
h
e
r
s
.
I
’
v
e
c
o
m
e
c
l
o
se
t
o
k
i
l
l
i
n
g
my
s
e
l
f
.
S
t
a
n
d
i
n
g
a
t
t
h
e
e
d
g
e
o
f
t
a
l
l
s
t
r
u
c
t
u
r
e
s
.
I
g
o
w
i
t
h
t
h
e
i
n
t
e
n
t
t
o
j
u
m
p
,
a
n
d
I
c
o
u
l
d
i
f
I
h
a
d
j
u
s
t
m
a
d
e
o
n
e
s
h
a
r
p
a
n
d
s
u
d
d
e
n
m
o
v
e
m
e
n
t
.
I
’
v
e
a
l
s
o
t
h
r
o
w
n
m
y
se
l
f
i
n
t
r
a
f
f
i
c
,
b
u
t
i
t
w
a
s
p
r
e
t
t
y
l
o
w
s
p
e
e
d
a
n
d
I
g
o
t
u
p
w
i
t
h
o
u
t
a
s
c
r
a
t
c
h
.
T
h
e
c
l
o
se
s
t
I
’
v
e
g
o
t
t
o
h
u
r
t
i
n
g
a
n
y
b
o
d
y
w
a
s
w
h
e
n
I
s
a
w
a
p
r
e
t
t
y
g
i
r
l
w
a
l
k
i
n
g
a
t
n
i
g
h
t
a
n
d
I
s
t
a
r
t
e
d
t
o
f
o
l
l
o
w
h
e
r
o
n
my
b
i
k
e
.
I
c
o
u
l
d
t
e
l
l
s
h
e
n
o
t
i
c
e
d
m
e
a
n
d
w
a
s
s
c
a
r
e
d
.
B
u
t
I
f
e
e
l
l
i
k
e
i
t
w
a
s
a
f
a
r
c
r
y
f
r
o
m
a
c
t
u
a
l
l
y
i
n
t
e
r
a
c
t
i
n
g
w
i
t
h
h
e
r
,
l
e
t
a
l
o
n
e
h
u
r
t
i
n
g
h
e
r
.
I
a
s
s
u
m
e
a
s
I
my
p
e
r
i
o
d
o
f
i
so
l
a
t
i
o
n
a
n
d
l
o
n
e
l
i
n
e
s
s
c
o
n
t
i
n
u
e
s
,
I
w
i
l
l
s
t
a
r
t
t
o
g
e
t
mo
r
e
c
o
u
r
a
g
e
t
o
d
o
b
o
t
h
.
B
u
t
I
r
e
a
l
l
y
h
o
p
e
I
k
i
l
l
my
se
l
f
b
e
f
o
r
e
h
u
r
t
i
n
g
a
n
y
b
o
d
y
.
1
4
.
2
.
P
re
-
pro
ce
s
s
ing
T
h
e
p
r
e
-
p
r
o
ce
s
s
in
g
s
tag
e
o
cc
u
r
s
af
ter
th
e
g
ath
er
i
n
g
o
f
d
ata.
Du
r
in
g
th
i
s
s
tag
e,
s
e
v
er
al
clea
n
in
g
p
r
o
ce
d
u
r
es
ar
e
u
s
ed
to
th
e
d
ata.
T
h
ese
p
r
o
ce
d
u
r
es
in
v
o
l
v
e
th
e
eli
m
in
at
io
n
o
f
HT
M
L
ele
m
e
n
t
s
,
s
p
ec
ial
ch
ar
ac
ter
s
,
n
u
m
er
al
s
,
an
d
p
u
n
ctu
a
tio
n
.
A
d
d
it
io
n
all
y
,
th
e
d
ata
is
co
n
v
er
ted
to
lo
w
er
ca
s
e
an
d
s
to
p
w
o
r
d
s
ar
e
r
e
m
o
v
ed
.
T
h
e
r
esu
lts
o
f
t
h
e
p
r
e
-
p
r
o
ce
s
s
i
n
g
p
r
o
ce
d
u
r
e
ar
e
d
e
p
icted
in
Fig
u
r
e
2
.
4
.
3
.
T
est
re
s
ult
T
h
e
ev
alu
atio
n
r
esu
l
ts
o
f
d
ee
p
lear
n
in
g
an
d
tr
an
s
f
er
lear
n
i
n
g
m
o
d
els
ar
e
p
r
esen
ted
in
T
ab
le
2
.
W
e
e
m
p
lo
y
ed
t
h
e
B
iLST
M
m
o
d
el
f
o
r
d
ee
p
lear
n
in
g
a
n
d
ev
alu
ated
t
h
e
p
er
f
o
r
m
a
n
ce
o
f
B
E
R
T
,
R
o
B
E
R
T
a,
AL
B
E
R
T
,
an
d
DeB
E
R
T
a
as
t
r
an
s
f
er
lear
n
in
g
m
o
d
el
s
.
T
h
e
DeB
E
R
T
a
m
o
d
el
attain
ed
a
r
e
m
ar
k
ab
le
ac
cu
r
ac
y
s
co
r
e
o
f
0
.
9
8
7
0
%,
s
h
o
w
ca
s
i
n
g
co
n
s
is
te
n
t
p
er
f
o
r
m
a
n
ce
i
n
ter
m
s
o
f
p
r
ec
is
io
n
,
r
ec
all,
an
d
F1
-
s
co
r
e.
T
h
e
s
u
b
s
eq
u
en
t
m
o
d
el,
R
o
B
E
R
T
a,
attain
ed
an
ac
cu
r
ac
y
s
co
r
e
o
f
0
.
9
8
3
4
%,
w
h
il
s
t B
E
R
T
h
ad
a
litt
le
lo
w
er
s
co
r
e
o
f
0
.
9
8
0
0
%.
AL
B
E
R
T
attain
ed
a
p
r
ec
is
io
n
s
co
r
e
o
f
0
.
9
5
0
0
%.
T
h
ese
f
i
n
d
i
n
g
s
d
e
m
o
n
s
t
r
ate
th
at
DeB
E
R
T
a
p
o
s
s
ess
es
ex
ce
p
tio
n
al
s
k
il
ls
a
n
d
ca
n
r
iv
a
l
o
th
er
m
o
d
els,
as
ev
id
en
ce
d
b
y
t
h
e
r
es
u
lt
s
p
r
esen
te
d
in
T
ab
le
2
o
f
th
e
test
i
n
g
d
ata.
Evaluation Warning : The document was created with Spire.PDF for Python.
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
A
co
mp
a
r
a
tive
a
n
a
lysi
s
o
f tra
n
s
fer lea
r
n
in
g
mo
d
els o
n
s
u
ici
d
e
a
n
d
n
o
n
-
s
u
icid
e
textu
a
l
…
(
Me
r
in
d
a
Les
ta
n
d
y
)
431
I
n
co
n
tr
ast to
p
r
i
o
r
r
esear
ch
[
3
6
]
,
th
er
e
is
a
n
o
tab
le
en
h
an
ce
m
en
t.
T
h
e
p
r
io
r
s
tu
d
y
d
o
cu
m
e
n
ted
an
F1
-
s
co
r
e
o
f
9
6
.
8
1
%.
T
h
is
en
h
an
c
e
m
en
t
in
d
icate
s
t
h
at
th
e
f
u
n
d
am
en
tal
m
o
d
els
cr
ea
ted
in
th
is
r
esear
ch
ar
e
o
n
p
ar
w
it
h
co
m
p
o
s
ite
m
o
d
els
lik
e
R
o
B
E
R
T
a
-
C
NN.
T
h
e
in
clu
s
io
n
o
f
th
i
s
f
ea
tu
r
e
e
n
ab
les
t
h
e
tr
a
n
s
f
o
r
m
er
m
o
d
el
to
s
u
r
p
ass
th
e
co
n
s
tr
ain
t
s
i
n
s
p
ee
d
an
d
ef
f
icie
n
c
y
t
h
at
ar
e
s
ee
n
in
r
ec
u
r
r
e
n
t
m
o
d
el
s
li
k
e
B
i
L
S
T
M.
T
h
e
B
iL
ST
M
m
o
d
el
f
ac
es
ch
alle
n
g
es
i
n
co
llectin
g
n
o
n
-
li
n
ea
r
co
r
r
elatio
n
s
b
et
w
ee
n
w
o
r
d
s
th
at
ar
e
w
id
e
l
y
s
ep
ar
ated
in
th
e
tex
t,
h
en
ce
h
i
n
d
er
in
g
it
s
ab
il
it
y
to
ac
c
u
r
atel
y
ex
p
r
es
s
i
n
tr
icate
s
e
m
an
tic
co
n
n
ec
tio
n
s
.
T
h
e
tab
le
’
s
r
es
u
lt
s
d
em
o
n
s
tr
ate
th
a
t
t
h
e
u
tili
za
t
io
n
o
f
tr
an
s
f
er
lear
n
i
n
g
h
as
t
h
e
c
ap
ab
ilit
y
to
o
v
er
co
m
e
t
h
e
li
m
i
tatio
n
s
li
n
k
ed
to
t
h
e
B
iL
ST
M
m
o
d
el.
T
h
e
o
b
j
ec
tiv
e
o
f
th
is
w
o
r
k
is
t
o
ass
ess
t
h
e
ef
f
ec
ti
v
en
e
s
s
o
f
s
e
v
er
al
t
y
p
es
o
f
tr
a
n
s
f
o
r
m
er
m
o
d
els
in
tex
t
class
i
f
icatio
n
task
s
th
at
i
n
v
o
l
v
e
s
en
s
it
iv
e
to
p
ics,
s
u
c
h
as
id
en
tify
i
n
g
i
n
d
icatio
n
s
o
f
s
u
icid
e
o
n
s
o
cial
m
ed
ia.
T
h
e
r
esu
lt
s
u
n
d
er
s
co
r
e
th
e
s
i
g
n
if
ic
an
ce
o
f
ch
o
o
s
in
g
th
e
s
u
itab
le
m
o
d
el
to
en
h
an
ce
p
er
f
o
r
m
a
n
c
e
in
task
s
r
elate
d
t
o
NL
P
,
p
ar
ticu
lar
l
y
i
n
ex
tr
e
m
el
y
cr
u
cial
an
d
in
f
l
u
en
t
ial
cir
cu
m
s
tan
ce
s
s
u
c
h
as e
m
o
tio
n
al
d
ata
s
ets.
Nev
er
t
h
ele
s
s
,
th
er
e
ar
e
s
til
l
co
n
s
tr
ain
t
s
r
eg
ar
d
in
g
t
h
e
p
er
f
o
r
m
a
n
ce
o
f
th
ese
m
o
d
els
o
n
d
iv
er
s
e
d
atas
ets
o
r
in
c
h
an
g
i
n
g
cir
cu
m
s
ta
n
ce
s
.
S
u
b
s
eq
u
en
t in
v
esti
g
atio
n
s
co
u
ld
p
r
io
r
itize
a
d
d
itio
n
al
ex
p
lo
r
atio
n
an
d
m
o
r
e
ex
ten
s
iv
e
test
in
g
to
en
h
a
n
ce
t
h
e
o
v
er
all
ap
p
licab
ilit
y
o
f
t
h
ese
m
o
d
el
s
.
Fig
u
r
e
2
.
P
r
e
-
p
r
o
ce
s
s
ed
r
esu
lts
o
f
s
u
icid
e
an
d
n
o
n
-
s
u
icid
e
d
atasets
T
ab
le
2
.
T
esti
n
g
r
es
u
lts
of
tr
an
s
f
o
r
m
er
m
o
d
el
A
l
g
o
r
i
t
ma
mo
d
e
l
A
c
c
u
r
a
c
y
(
%)
P
r
e
c
i
si
o
n
(
%)
R
e
c
a
l
l
(
%)
F1
-
sco
r
e
(
%)
B
i
L
S
T
M
+
G
l
o
V
e
0
.
9
3
4
6
0
.
9
5
3
2
0
.
9
1
4
2
0
.
9
3
3
2
B
ER
T
0
.
9
8
0
.
9
8
0
.
9
8
0
.
9
8
R
o
B
E
R
T
a
0
.
9
8
3
4
0
.
9
8
3
4
0
.
9
8
3
4
0
.
9
8
3
4
A
L
B
ER
T
0
.
9
5
0
.
9
5
0
.
9
5
0
.
9
5
D
e
B
ER
T
a
0
.
9
8
7
0
0
.
9
8
7
0
0
.
9
8
7
0
0
.
9
8
7
0
R
o
B
E
R
T
a
-
C
N
N
[
3
6
]
0
.
9
8
0
.
9
6
9
8
0
.
9
6
4
0
.
9
6
8
1
My
ap
ol
ogies
to
th
e
r
an
d
om
dude
on
A
m
ong
U
s
So
uhh
…
.
I
w
ould
lik
e
to
a
p
ol
ogiz
e
to
th
e
r
an
d
om
du
d
e
n
am
ed
P
on
on
am
ong
us
.
I
h
ad
h
os
t
ed
a
g
ame
f
or
my
gf
an
d
h
er
fr
ien
d
to
learn
h
o
w
to
p
la
y
and
wha
te
v
er
a
n
d
…
we
m
a
y
h
a
ve
tak
e
n
ov
er
th
e
c
ha
t
wi
th
U
W
U
s
an
d
uh
…
.
othe
r
q
u
e
s
tionab
le
langu
ag
e
…
.
*
c
ough
*
I
w
as
c
alled
"m
om
m
y
y
ello
w
"
*
c
ough
*
AN
YW
A
Y
S
so
r
r
y
ya
h
a
d
to
w
itne
ss
th
a
t
dude
My
ap
ol
ogies
to
th
e
r
an
d
om
dude
on
A
m
ong
U
s
So
uhh
…
.
I
w
ould
lik
e
to
a
p
ol
ogiz
e
to
th
e
r
an
d
om
du
d
e
n
am
e
d
P
on
on
am
ong
us
.
I
h
ad
h
os
t
e
d
a
g
ame
f
or
my
gf
an
d
h
er
fr
ien
d
to
learn
h
o
w
to
p
la
y
and
wha
te
v
er
a
n
d
…
we
m
a
y
h
a
ve
tak
e
n
ov
er
th
e
c
ha
t
wi
th
U
W
U
s
an
d
uh
…
.
othe
r
q
u
e
s
tionab
le
langu
ag
e
…
.
*
c
ough
*
I
w
as
c
alled
momm
y
y
ello
w
c
o
ugh
AN
YW
A
Y
S
so
rr
y
ya
h
ad
to
wi
tn
ess
th
a
t
dude
My
ap
ol
ogies
to
th
e
r
an
d
om
dude
on
A
m
ong
U
s
So
uhh
I
w
ould
li
k
e
to
ap
olo
giz
e
to
th
e
r
an
d
om
du
d
e
n
am
e
d
P
on
on
am
ong
us
I
h
ad
h
os
t
ed
a
g
ame
f
or
my
gf
an
d
h
er
fr
ien
d
to
learn
h
o
w
to
p
la
y
and
wha
te
v
er
a
n
d
we
m
a
y
h
a
ve
tak
e
n
ov
er
th
e
c
ha
t
wi
t
h
U
W
U
s
an
d
uh
ot
h
er
q
u
estionab
l
e
langu
ag
e
c
ough
I
w
as
c
alled
m
omm
y
y
ello
w
c
ough
A
N
YW
A
Y
S
so
rr
y
ya
h
ad
to
wi
tn
ess
th
a
t
dude
my
ap
ol
ogies
to
th
e
r
an
d
om
dude
on
am
ong
us
so
uhh
i
w
ould
lik
e
to
ap
ol
ogiz
e
to
th
e
r
an
d
om
d
u
d
e
n
am
ed
p
on
on
am
ong
us
i
h
ad
h
os
t
ed
a
g
am
e
f
or
my
gf
an
d
h
er
fr
ien
d
to
learn
h
o
w
to
pla
y
an
d
wha
te
v
er
a
n
d
we
m
a
y
h
a
ve
tak
e
n
ov
er
the
c
h
a
t
wi
t
h
uwu
s
an
d
uh
oth
er
q
u
e
s
tionab
l
e
lan
gu
ag
e
c
oug
h
i
w
as
c
alled
m
omm
y
y
ello
w
c
ough
an
yw
a
y
s
so
rr
y
ya
h
ad
to
wi
tn
ess
th
a
t
dude
ap
olo
gie
s
r
an
d
om
d
u
d
e
am
ong
u
h
h
ap
olo
giz
e
r
an
d
om
dude
n
am
e
d
p
on
am
ong
h
os
t
ed
g
am
e
gf
fr
ien
d
learn
p
la
y
t
a
k
en
c
h
at
u
wu
q
u
es
tionab
l
e
lan
gu
ag
e
c
oug
h
c
al
led
m
omm
y
y
ello
w
c
ough
wi
tn
ess
dude
ap
ol
ogy
r
an
d
om
dude
am
ong
u
h
h
a
p
ol
ogiz
e
r
an
d
om
d
u
d
e
n
am
e
p
on
am
ong
h
os
t
g
am
e
gf
fr
ien
d
le
arn
p
la
y
tak
e
c
h
at
u
wu
q
u
e
s
tionab
le
lan
gu
ag
e
c
oug
h
c
all
m
om
m
y
y
ello
w
c
ough
wi
tn
ess
dude
Or
i
ginal
Doc
ume
n
t
Spe
c
i
al
Cha
r
act
er
and
Di
git
R
emo
v
e
P
unct
ua
ti
on
Lo
w
er
T
e
xt
St
opw
or
d
s
Le
mm
a
ti
z
e
Evaluation Warning : The document was created with Spire.PDF for Python.
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
20
2
5
:
4
2
6
-
434
432
T
h
e
class
d
iv
i
s
io
n
o
f
t
h
e
co
n
f
u
s
io
n
m
atr
i
x
o
u
tco
m
es
f
o
r
th
e
DeB
E
R
T
A
m
o
d
el
is
d
is
p
la
y
ed
i
n
T
ab
le
3
.
T
h
e
m
atr
ix
e
n
co
m
p
a
s
s
es
a
to
tal
o
f
4
6
,
4
1
5
test
in
g
d
ata
in
s
ta
n
ce
s
.
T
h
e
F1
-
s
co
r
e
f
o
r
th
e
n
o
n
-
s
u
icid
e
lab
el
is
0
.
9
8
7
1
%,
in
d
icatin
g
t
h
at
t
h
e
m
o
d
el
’
s
id
en
ti
f
ica
ti
o
n
o
f
n
o
n
-
s
u
icid
e
te
x
t
i
s
as
s
o
ciate
d
w
i
th
a
m
in
i
m
a
l
n
u
m
b
er
o
f
f
al
s
e
p
o
s
iti
v
es.
I
n
a
s
i
m
ilar
v
ei
n
,
t
h
e
F1
s
co
r
e
f
o
r
th
e
s
u
icid
e
lab
el
w
o
r
d
in
g
is
0
.
9
8
6
9
%.
T
h
e
DeB
E
R
T
a
m
o
d
el
p
r
o
v
id
es
th
e
ad
v
an
ta
g
e
o
f
e
m
p
lo
y
i
n
g
d
is
e
n
t
an
g
led
a
tten
tio
n
,
w
h
ic
h
f
ac
ili
ta
tes
a
m
o
r
e
f
o
cu
s
ed
an
d
s
tr
u
ct
u
r
ed
atten
tio
n
m
ec
h
an
i
s
m
.
T
h
is
ap
p
r
o
ac
h
h
elp
s
ad
d
r
ess
ch
alle
n
g
e
s
r
elate
d
to
u
n
d
er
s
ta
n
d
in
g
t
h
e
r
elatio
n
s
h
ip
b
et
w
ee
n
w
o
r
d
s
i
n
lo
n
g
er
tex
t
s
an
d
i
m
p
r
o
v
e
s
th
e
ab
ilit
y
to
p
er
ce
iv
e
co
n
tex
t
i
n
a
b
r
o
a
d
er
r
an
g
e
o
f
in
f
o
r
m
atio
n
.
T
h
e
s
u
cc
es
s
f
u
l
o
u
tco
m
e
o
f
th
is
h
as
b
ee
n
p
r
o
v
en
b
y
t
h
e
f
i
n
d
in
g
s
p
r
esen
ted
in
T
ab
le
2
th
at
r
esu
lt
f
r
o
m
ea
ch
tr
ia
l c
o
n
d
u
cted
o
n
t
h
e
tr
an
s
f
o
r
m
er
m
o
d
els.
T
ab
le
3
.
C
lass
d
iv
is
io
n
o
f
t
h
e
DeB
E
R
T
a
m
o
d
el
L
a
b
e
l
P
r
e
c
i
si
o
n
R
e
c
a
l
l
F1
-
s
c
o
r
e
N
on
-
su
i
c
i
d
e
0
.
9
8
3
5
0
.
9
9
0
7
0
.
9
8
7
1
S
u
i
c
i
d
e
0
.
9
9
0
6
0
.
9
8
3
2
0
.
9
8
6
9
5.
CO
NCLU
SI
O
N
T
h
is
s
tu
d
y
ai
m
s
to
ac
cu
r
atel
y
id
en
ti
f
y
s
u
icid
al
ten
d
en
c
ies
b
y
te
x
t
u
al
an
al
y
s
i
s
,
w
h
ic
h
is
cr
u
cial
s
i
n
ce
s
u
icid
e
is
t
h
e
s
ec
o
n
d
m
o
s
t
co
m
m
o
n
ca
u
s
e
o
f
d
ea
th
a
m
o
n
g
t
h
o
s
e
ag
ed
1
5
to
2
9
w
o
r
ld
w
id
e
,
as
r
ep
o
r
ted
b
y
th
e
W
HO
.
T
h
e
C
OVI
D
-
1
9
ep
id
em
ic
h
as
e
x
ac
er
b
ated
m
en
ta
l
h
ea
lt
h
d
if
f
ic
u
ltie
s
,
h
ei
g
h
ten
in
g
t
h
e
lik
e
lih
o
o
d
o
f
s
u
icid
e.
So
cial
m
ed
ia
h
as
e
m
e
r
g
ed
as
a
cr
u
cial
p
latf
o
r
m
f
o
r
th
o
s
e
ex
p
er
ien
ci
n
g
d
is
tr
ess
to
o
p
en
l
y
ex
p
r
ess
an
d
d
is
s
e
m
in
ate
t
h
eir
e
m
o
tio
n
s
,
th
er
eb
y
s
er
v
i
n
g
as
a
h
elp
f
u
l
to
o
l
f
o
r
p
r
o
m
p
tl
y
id
en
ti
f
y
i
n
g
an
d
in
ter
v
e
n
i
n
g
in
s
u
c
h
ca
s
es.
T
h
is
s
tu
d
y
u
til
ized
d
ee
p
lear
n
in
g
a
n
d
tr
an
s
f
er
lear
n
i
n
g
m
o
d
el
s
to
ac
cu
r
atel
y
clas
s
if
y
tex
ts
as
eit
h
er
s
u
icid
e
-
r
elate
d
o
r
n
o
n
-
s
u
icid
e
-
r
elate
d
.
W
e
im
p
le
m
e
n
ted
m
eti
cu
lo
u
s
d
ata
clea
n
s
i
n
g
p
r
o
ce
d
u
r
es,
w
h
ic
h
in
v
o
lv
ed
d
eletin
g
HT
ML
co
m
p
o
n
e
n
ts
,
s
p
ec
ial
ch
ar
ac
ter
s
,
d
ig
its
,
p
u
n
ctu
atio
n
,
a
n
d
e
m
o
j
is
.
A
d
d
itio
n
all
y
,
w
e
co
n
v
er
ted
th
e
te
x
t
to
lo
w
er
ca
s
e,
r
e
m
o
v
ed
s
to
p
w
o
r
d
s
,
an
d
p
er
f
o
r
m
ed
le
m
m
atiza
tio
n
o
n
t
h
e
w
o
r
d
s
.
W
e
ev
alu
ated
o
u
r
m
o
d
el
s
b
y
co
m
p
ar
in
g
t
h
e
m
to
o
th
er
s
o
p
h
i
s
ticated
tr
an
s
f
o
r
m
er
s
an
d
d
ee
p
lear
n
i
n
g
tech
n
iq
u
es,
s
u
c
h
as
B
E
R
T
,
R
o
B
E
R
T
a,
A
L
B
E
R
T
,
an
d
De
B
E
R
T
a,
in
ad
d
itio
n
to
th
e
B
i
L
ST
M
m
o
d
el
b
ased
o
n
Glo
Ve
em
b
ed
d
in
g
s
.
T
h
e
f
i
n
d
in
g
s
i
n
d
icate
d
t
h
at
t
h
e
De
B
E
R
T
a
m
o
d
el
h
ad
s
u
p
er
io
r
p
er
f
o
r
m
an
ce
co
m
p
ar
ed
to
o
th
er
m
o
d
els
i
n
ter
m
s
o
f
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
a
n
d
F1
-
s
co
r
e.
T
h
e
r
ea
s
o
n
b
eh
i
n
d
DeB
E
R
T
a
’
s
e
x
ce
p
tio
n
al
p
er
f
o
r
m
an
ce
ca
n
b
e
attr
ib
u
ted
to
its
d
is
e
n
ta
n
g
led
atten
tio
n
m
ec
h
a
n
is
m
.
T
h
is
tec
h
n
iq
u
e
e
n
ab
les
t
h
e
m
o
d
el
to
c
h
o
o
s
e
to
co
n
ce
n
tr
ate
o
n
p
er
tin
en
t
tex
tu
al
co
m
p
o
n
e
n
ts
w
h
ile
r
ej
ec
tin
g
e
x
tr
a
n
eo
u
s
i
n
f
o
r
m
ati
o
n
.
T
h
is
f
ea
t
u
r
e
i
m
p
r
o
v
es
it
s
ef
f
icac
y
i
n
ca
p
tu
r
in
g
th
e
i
n
tr
icate
a
n
d
in
tr
icate
asp
ec
ts
o
f
la
n
g
u
a
g
e
ass
o
ciate
d
w
it
h
s
u
icid
al
id
ea
tio
n
.
T
h
e
r
esu
lts
o
f
o
u
r
r
esear
ch
i
n
d
icate
th
at
ad
v
a
n
ce
d
tr
an
s
f
o
r
m
er
m
o
d
els,
s
p
ec
i
f
icall
y
DeB
E
R
T
a,
s
h
o
w
g
r
ea
t
p
o
ten
tial
i
n
i
m
p
r
o
v
in
g
t
h
e
id
en
tific
atio
n
o
f
s
u
icid
al
ten
d
en
cies
b
y
a
n
al
y
zi
n
g
te
x
t
f
r
o
m
s
o
cial
m
ed
ia.
T
h
ese
f
i
n
d
in
g
s
h
av
e
s
ig
n
i
f
ica
n
t
co
n
s
eq
u
en
ce
s
f
o
r
ea
r
l
y
i
n
ter
v
e
n
ti
o
n
an
d
p
r
ev
en
tio
n
m
et
h
o
d
s
,
p
o
ten
tiall
y
r
ed
u
ci
n
g
m
o
r
ta
lit
y
r
ates
b
y
id
e
n
ti
f
y
in
g
p
er
s
o
n
s
at
r
is
k
m
o
r
e
p
r
ec
is
el
y
an
d
s
w
i
f
tl
y
.
S
u
b
s
eq
u
e
n
t
s
tu
d
ie
s
s
h
o
u
ld
i
n
v
e
s
ti
g
ate
th
e
ex
te
n
t
to
w
h
ic
h
t
h
ese
m
o
d
els
m
a
y
b
e
ap
p
lied
to
d
iv
er
s
e
d
atasets
an
d
ex
a
m
i
n
e
h
o
w
d
if
f
er
en
t
d
ata
p
r
ep
r
o
ce
s
s
in
g
ap
p
r
o
ac
h
es
a
f
f
e
ct
th
e
p
er
f
o
r
m
an
ce
o
f
t
h
e
m
o
d
els.
Fu
r
t
h
er
m
o
r
e,
in
co
r
p
o
r
atin
g
a
w
id
e
r
an
g
e
o
f
lin
g
u
i
s
tic
a
n
d
c
u
lt
u
r
al
co
n
te
x
ts
co
u
ld
s
i
g
n
if
ican
t
l
y
s
tr
en
g
t
h
e
n
th
e
r
eliab
ili
t
y
a
n
d
p
r
ac
ticalit
y
o
f
th
e
s
e
m
o
d
els
i
n
r
ea
l
-
li
f
e
s
it
u
atio
n
s
.
ACK
NO
WL
E
D
G
E
M
E
NT
S
T
h
is
r
esear
ch
w
a
s
s
u
p
p
o
r
ted
b
y
t
h
e
Dep
ar
t
m
en
t
o
f
E
lectr
ical
E
n
g
in
ee
r
i
n
g
,
Un
i
v
er
s
it
y
o
f
Mu
h
a
m
m
ad
i
y
ah
Ma
la
n
g
an
d
t
h
e
C
o
m
p
u
ter
L
ab
o
r
ato
r
y
o
f
t
h
e
Dep
ar
tm
e
n
t
o
f
E
lectr
ical
E
n
g
i
n
ee
r
in
g
,
Un
iv
er
s
it
y
o
f
Mu
h
a
m
m
ad
i
y
ah
Ma
lan
g
,
wh
ich
p
r
o
v
id
ed
in
s
ig
h
t
s
an
d
ex
p
er
tis
e
th
a
t r
ea
ll
y
h
elp
ed
th
e
r
es
ea
r
ch
.
RE
F
E
R
E
NC
E
S
[
1
]
K
.
W
i
n
d
f
u
h
r
a
n
d
N
.
K
a
p
u
r
,
“
S
u
i
c
i
d
e
a
n
d
me
n
t
a
l
i
l
l
n
e
ss:
a
c
l
i
n
i
c
a
l
r
e
v
i
e
w
o
f
1
5
y
e
a
r
s
f
i
n
d
i
n
g
s
f
r
o
m
t
h
e
U
K
N
a
t
i
o
n
a
l
C
o
n
f
i
d
e
n
t
i
a
l
I
n
q
u
i
r
y
i
n
t
o
S
u
i
c
i
d
e
,
”
Br
i
t
i
sh
Me
d
i
c
a
l
Bu
l
l
e
t
i
n
,
v
o
l
.
1
0
0
,
p
p
.
1
0
1
–
1
2
1
,
2
0
1
1
,
d
o
i
:
1
0
.
1
0
9
3
/
b
m
b
/
l
d
r
0
4
2
.
[
2
]
M
.
A
.
R
e
g
e
r
,
I
.
H
.
S
t
a
n
l
e
y
,
a
n
d
T
.
E.
Jo
i
n
e
r
,
“
S
u
i
c
i
d
e
M
o
r
t
a
l
i
t
y
a
n
d
C
o
r
o
n
a
v
i
r
u
s
D
i
se
a
se
2
0
1
9
—
A
P
e
r
f
e
c
t
S
t
o
r
m?
,
”
J
A
M
A
Psy
c
h
i
a
t
r
y
,
v
o
l
.
7
7
,
n
o
.
1
1
,
p
p
.
1
0
9
3
–
1
0
9
4
,
N
o
v
.
2
0
2
0
,
d
o
i
:
1
0
.
1
0
0
1
/
j
a
ma
p
sy
c
h
i
a
t
r
y
.
2
0
2
0
.
1
0
6
0
.
[
3
]
I
.
B
u
n
e
v
i
c
i
e
n
e
,
R
.
B
u
n
e
v
i
c
i
u
s
,
S
.
B
a
g
d
o
n
a
s
,
a
n
d
A
.
B
u
n
e
v
i
c
i
u
s,
“
T
h
e
i
m
p
a
c
t
o
f
p
r
e
-
e
x
i
st
i
n
g
c
o
n
d
i
t
i
o
n
s
a
n
d
p
e
r
c
e
i
v
e
d
h
e
a
l
t
h
st
a
t
u
s
o
n
me
n
t
a
l
h
e
a
l
t
h
d
u
r
i
n
g
t
h
e
C
O
V
I
D
-
1
9
p
a
n
d
e
mi
c
,
”
J
o
u
r
n
a
l
o
f
Pu
b
l
i
c
H
e
a
l
t
h
,
v
o
l
.
4
4
,
n
o
.
1
,
p
p
.
e
8
8
–
e
9
5
,
M
a
r
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
9
3
/
p
u
b
me
d
/
f
d
a
b
2
4
8
.
[
4
]
S
.
G
a
l
e
a
,
R
.
M
.
M
e
r
c
h
a
n
t
,
a
n
d
N
.
L
u
r
i
e
,
“
T
h
e
M
e
n
t
a
l
H
e
a
l
t
h
C
o
n
se
q
u
e
n
c
e
s
o
f
C
O
V
I
D
-
1
9
a
n
d
P
h
y
si
c
a
l
D
i
s
t
a
n
c
i
n
g
:
T
h
e
N
e
e
d
f
o
r
P
r
e
v
e
n
t
i
o
n
a
n
d
Ea
r
l
y
I
n
t
e
r
v
e
n
t
i
o
n
,
”
J
AM
A
I
n
t
e
rn
a
l
Me
d
i
c
i
n
e
,
v
o
l
.
1
8
0
,
n
o
.
6
,
p
p
.
8
1
7
–
8
1
8
,
Ju
n
.
2
0
2
0
,
d
o
i
:
1
0
.
1
0
0
1
/
j
a
ma
i
n
t
e
r
n
me
d
.
2
0
2
0
.
1
5
6
2
.
Evaluation Warning : The document was created with Spire.PDF for Python.
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
A
co
mp
a
r
a
tive
a
n
a
lysi
s
o
f tra
n
s
fer lea
r
n
in
g
mo
d
els o
n
s
u
ici
d
e
a
n
d
n
o
n
-
s
u
icid
e
textu
a
l
…
(
Me
r
in
d
a
Les
ta
n
d
y
)
433
[
5
]
W
o
r
l
d
H
e
a
l
t
h
O
r
g
a
n
i
z
a
t
i
o
n
,
N
a
t
i
o
n
a
l
s
u
i
c
i
d
e
p
r
e
v
e
n
t
i
o
n
s
t
ra
t
e
g
i
e
s
.
,
v
o
l
.
3
0
.
2
0
1
8
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s:
/
/
a
p
p
s.w
h
o
.
i
n
t
/
i
r
i
s
/
b
i
t
s
t
r
e
a
m/
h
a
n
d
l
e
/
1
0
6
6
5
/
2
7
9
7
6
5
/
9
7
8
9
2
4
1
5
1
5
0
1
6
-
e
n
g
.
p
d
f
?
u
a
=
1
(
A
c
c
e
sse
d
:
Ju
l
.
0
1
,
2
0
2
3)
[
6
]
G
.
A
n
a
n
t
h
a
k
r
i
sh
n
a
n
,
A
.
K
.
Jay
a
r
a
man
,
T
.
T
r
u
e
man
,
S
.
M
i
t
r
a
,
A
.
A
. K
,
a
n
d
A
.
M
u
r
u
g
a
p
p
a
n
,
S
u
i
c
i
d
a
l
I
n
t
e
n
t
i
o
n
D
e
t
e
c
t
i
o
n
i
n
T
w
e
e
t
s
U
si
n
g
B
ERT
-
B
a
se
d
T
r
a
n
s
f
o
rm
e
rs
.
2
0
2
2
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
C
C
I
S
5
6
4
3
0
.
2
0
2
2
.
1
0
0
3
7
6
7
7
.
[
7
]
I
.
A
mee
r
,
M
.
A
r
i
f
,
G
.
S
i
d
o
r
o
v
,
H
.
G
ò
me
z
-
A
d
o
r
n
o
,
a
n
d
A
.
G
e
l
b
u
k
h
,
“
M
e
n
t
a
l
I
l
l
n
e
ss
C
l
a
ssi
f
i
c
a
t
i
o
n
o
n
S
o
c
i
a
l
M
e
d
i
a
T
e
x
t
s
u
s
i
n
g
D
e
e
p
L
e
a
r
n
i
n
g
a
n
d
T
r
a
n
sf
e
r
L
e
a
r
n
i
n
g
,
”
a
rX
i
v
p
re
p
r
i
n
t
,
2
0
2
2
,
d
o
i
:
1
0
.
4
8
5
5
0
/
a
r
X
i
v
.
2
2
0
7
.
0
1
0
1
2
.
[
8
]
U
.
A
h
me
d
,
J.
C
.
W
.
L
i
n
,
a
n
d
G
.
S
r
i
v
a
st
a
v
a
,
“
H
y
p
e
r
-
g
r
a
p
h
-
b
a
se
d
a
t
t
e
n
t
i
o
n
c
u
r
r
i
c
u
l
u
m l
e
a
r
n
i
n
g
u
s
i
n
g
a
l
e
x
i
c
a
l
a
l
g
o
r
i
t
h
m
f
o
r
me
n
t
a
l
h
e
a
l
t
h
,
”
P
a
t
t
e
rn
Re
c
o
g
n
i
t
i
o
n
L
e
t
t
e
rs
,
v
o
l
.
1
5
7
,
p
p
.
1
3
5
–
1
4
3
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
a
t
r
e
c
.
2
0
2
2
.
0
3
.
0
1
8
.
[
9
]
A
.
K
u
mar,
K
.
S
h
a
r
ma,
a
n
d
A
.
S
h
a
r
ma,
“
H
i
e
r
a
r
c
h
i
c
a
l
d
e
e
p
n
e
u
r
a
l
n
e
t
w
o
r
k
f
o
r
me
n
t
a
l
s
t
r
e
ss
st
a
t
e
d
e
t
e
c
t
i
o
n
u
s
i
n
g
I
o
T
b
a
se
d
b
i
o
mar
k
e
r
s,”
P
a
t
t
e
r
n
Re
c
o
g
n
i
t
i
o
n
L
e
t
t
e
rs
,
v
o
l
.
1
4
5
,
p
p
.
8
1
–
8
7
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
a
t
r
e
c
.
2
0
2
1
.
0
1
.
0
3
0
.
[
1
0
]
R
.
S
k
a
i
k
a
n
d
D
.
I
n
k
p
e
n
,
“
U
si
n
g
S
o
c
i
a
l
M
e
d
i
a
f
o
r
M
e
n
t
a
l
H
e
a
l
t
h
S
u
r
v
e
i
l
l
a
n
c
e
:
A
R
e
v
i
e
w
,
”
AC
M
C
o
m
p
u
t
i
n
g
S
u
rve
y
s
,
v
o
l
.
5
3
,
n
o
.
6
,
p
p
.
1
-
3
1
,
D
e
c
.
2
0
2
0
,
d
o
i
:
1
0
.
1
1
4
5
/
3
4
2
2
8
2
4
.
[
1
1
]
A
.
L
e
G
l
a
z
e
t
a
l
.
,
“
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
a
n
d
N
a
t
u
r
a
l
L
a
n
g
u
a
g
e
P
r
o
c
e
ssi
n
g
i
n
M
e
n
t
a
l
H
e
a
l
t
h
:
S
y
st
e
mat
i
c
R
e
v
i
e
w
,
”
J
o
u
r
n
a
l
o
f
m
e
d
i
c
a
l
I
n
t
e
r
n
e
t
r
e
se
a
rc
h
,
v
o
l
.
2
3
,
n
o
.
5
,
M
a
y
2
0
2
1
,
d
o
i
:
1
0
.
2
1
9
6
/
1
5
7
0
8
.
[
1
2
]
T
.
Z
h
a
n
g
,
A
.
M
.
S
c
h
o
e
n
e
,
S
.
Ji
,
a
n
d
S
.
A
n
a
n
i
a
d
o
u
,
“
N
a
t
u
r
a
l
l
a
n
g
u
a
g
e
p
r
o
c
e
ssi
n
g
a
p
p
l
i
e
d
t
o
me
n
t
a
l
i
l
l
n
e
ss
d
e
t
e
c
t
i
o
n
:
a
n
a
r
r
a
t
i
v
e
r
e
v
i
e
w
,
”
N
PJ
D
i
g
i
t
a
l
M
e
d
i
c
i
n
e
,
v
o
l
.
5
,
n
o
.
1
,
p
p
.
1
–
1
3
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
3
8
/
s
4
1
7
4
6
-
0
2
2
-
0
0
5
8
9
-
7.
[
1
3
]
R
.
S
a
l
a
s
-
Z
á
r
a
t
e
,
G
.
A
l
o
r
-
H
e
r
n
á
n
d
e
z
,
M
.
A
.
P
a
r
e
d
e
s
-
V
a
l
v
e
r
d
e
,
M
.
d
e
l
P
.
S
a
l
a
s
-
Z
á
r
a
t
e
,
M
.
B
u
st
o
s
-
L
ó
p
e
z
,
a
n
d
J.
L
.
S
á
n
c
h
e
z
-
C
e
r
v
a
n
t
e
s,
“
M
e
n
t
a
l
-
H
e
a
l
t
h
:
A
n
N
L
P
-
B
a
se
d
S
y
st
e
m
f
o
r
D
e
t
e
c
t
i
n
g
D
e
p
r
e
ssi
o
n
L
e
v
e
l
s
t
h
r
o
u
g
h
U
se
r
C
o
mm
e
n
t
s
o
n
Tw
i
t
t
e
r
(
X
)
,
”
Ma
t
h
e
m
a
t
i
c
s
,
v
o
l
.
1
2
,
n
o
.
1
3
,
p
p
.
1
–
3
0
,
2
0
2
4
,
d
o
i
:
1
0
.
3
3
9
0
/
m
a
t
h
1
2
1
3
1
9
2
6
.
[
1
4
]
A
.
V
a
sw
a
n
i
e
t
a
l
.
,
“
A
t
t
e
n
t
i
o
n
i
s
a
l
l
y
o
u
n
e
e
d
,
”
A
d
v
a
n
c
e
s
i
n
N
e
u
r
a
l
I
n
f
o
rm
a
t
i
o
n
Pr
o
c
e
ss
i
n
g
S
y
s
t
e
m
s
,
p
p
.
5
9
9
9
–
6
0
0
9
,
2
0
1
7
.
[
1
5
]
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
,
”
i
n
N
AAC
L
H
L
T
2
0
1
9
-
2
0
1
9
C
o
n
f
e
r
e
n
c
e
o
f
t
h
e
N
o
rt
h
A
m
e
ri
c
a
n
C
h
a
p
t
e
r
o
f
t
h
e
Ass
o
c
i
a
t
i
o
n
f
o
r
C
o
m
p
u
t
a
t
i
o
n
a
l
L
i
n
g
u
i
st
i
c
s:
H
u
m
a
n
L
a
n
g
u
a
g
e
T
e
c
h
n
o
l
o
g
i
e
s
-
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
C
o
n
f
e
re
n
c
e
,
2
0
1
9
,
p
p
.
4
1
7
1
–
4
1
8
6
.
[
1
6
]
T
.
W
o
l
f
e
t
a
l
.
,
“
T
r
a
n
sf
o
r
me
r
s
:
S
t
a
t
e
-
of
-
t
h
e
-
A
r
t
N
a
t
u
r
a
l
L
a
n
g
u
a
g
e
P
r
o
c
e
ssi
n
g
,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
2
0
2
0
C
o
n
f
e
r
e
n
c
e
o
n
Em
p
i
r
i
c
a
l
Me
t
h
o
d
s
i
n
N
a
t
u
ra
l
L
a
n
g
u
a
g
e
Pr
o
c
e
s
si
n
g
:
S
y
s
t
e
m
D
e
m
o
n
st
r
a
t
i
o
n
s
,
2
0
2
0
,
p
p
.
3
8
–
4
5
,
d
o
i
:
1
0
.
1
8
6
5
3
/
v
1
/
2
0
2
0
.
e
mn
l
p
-
d
e
mo
s.6
.
[
1
7
]
C
.
M
.
G
r
e
c
o
,
A
.
S
i
me
r
i
,
A
.
T
a
g
a
r
e
l
l
i
,
a
n
d
E
.
Z
u
m
p
a
n
o
,
“
T
r
a
n
sf
o
r
me
r
-
b
a
se
d
l
a
n
g
u
a
g
e
mo
d
e
l
s
f
o
r
me
n
t
a
l
h
e
a
l
t
h
i
s
su
e
s:
A
su
r
v
e
y
,
”
Pa
t
t
e
r
n
Re
c
o
g
n
i
t
i
o
n
L
e
t
t
e
rs
,
v
o
l
.
1
6
7
,
p
p
.
2
0
4
–
2
1
1
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
a
t
r
e
c
.
2
0
2
3
.
0
2
.
0
1
6
.
[
1
8
]
Q
.
u
n
N
i
sa
a
n
d
R
.
M
u
h
a
mm
a
d
,
“
T
o
w
a
r
d
s
t
r
a
n
sf
e
r
l
e
a
r
n
i
n
g
u
si
n
g
B
ER
T
f
o
r
e
a
r
l
y
d
e
t
e
c
t
i
o
n
o
f
se
l
f
-
h
a
r
m
o
f
so
c
i
a
l
me
d
i
a
u
se
r
s,”
C
EU
R
Wo
r
k
s
h
o
p
Pro
c
.
,
v
o
l
.
2
9
3
6
,
p
p
.
1
0
5
9
–
1
0
7
0
,
2
0
2
1
.
[
1
9
]
S
.
Ji
,
T
.
Z
h
a
n
g
,
L
.
A
n
sari
,
J.
F
u
,
P
.
T
i
w
a
r
i
,
a
n
d
E.
C
a
mb
r
i
a
,
“
M
e
n
t
a
l
B
E
R
T
:
P
u
b
l
i
c
l
y
A
v
a
i
l
a
b
l
e
P
r
e
t
r
a
i
n
e
d
L
a
n
g
u
a
g
e
M
o
d
e
l
s
f
o
r
M
e
n
t
a
l
H
e
a
l
t
h
c
a
r
e
,
”
2
0
2
2
L
a
n
g
u
a
g
e
Re
so
u
r
c
e
s
a
n
d
Ev
a
l
u
a
t
i
o
n
C
o
n
f
e
r
e
n
c
e
,
L
R
EC
2
0
2
2
,
p
p
.
7
1
8
4
–
7
1
9
0
,
2
0
2
2
.
[
2
0
]
H
.
W
a
n
g
,
X
.
H
u
,
a
n
d
H
.
Z
h
a
n
g
,
“
S
e
n
t
i
me
n
t
a
n
a
l
y
si
s
o
f
c
o
mm
o
d
i
t
y
r
e
v
i
e
w
s
b
a
se
d
o
n
A
L
B
ER
T
-
L
S
T
M
,
”
J
o
u
rn
a
l
o
f
P
h
y
s
i
c
s:
C
o
n
f
e
re
n
c
e
S
e
r
i
e
s
,
v
o
l
.
1
6
5
1
,
n
o
.
1
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
8
8
/
1
7
4
2
-
6
5
9
6
/
1
6
5
1
/
1
/
0
1
2
0
2
2
.
[
2
1
]
S
.
K
a
u
r
,
R
.
B
h
a
r
d
w
a
j
,
A
.
Jai
n
,
M
.
G
a
r
g
,
a
n
d
C
.
S
a
x
e
n
a
,
“
C
a
u
s
a
l
C
a
t
e
g
o
r
i
z
a
t
i
o
n
o
f
M
e
n
t
a
l
H
e
a
l
t
h
P
o
st
s
u
s
i
n
g
T
r
a
n
sf
o
r
me
r
s,”
AC
M
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
Pro
c
e
e
d
i
n
g
S
e
r
i
e
s
,
v
o
l
.
1
,
n
o
.
1
,
p
p
.
4
3
–
4
6
,
2
0
2
2
,
d
o
i
:
1
0
.
1
1
4
5
/
3
5
7
4
3
1
8
.
3
5
7
4
3
3
4
.
[
2
2
]
T
.
Z
h
a
n
g
,
A
.
M
.
S
c
h
o
e
n
e
,
a
n
d
S
.
A
n
a
n
i
a
d
o
u
,
“
A
u
t
o
mat
i
c
i
d
e
n
t
i
f
i
c
a
t
i
o
n
o
f
su
i
c
i
d
e
n
o
t
e
s
w
i
t
h
a
t
r
a
n
sf
o
r
me
r
-
b
a
s
e
d
d
e
e
p
l
e
a
r
n
i
n
g
mo
d
e
l
,
”
I
n
t
e
rn
e
t
I
n
t
e
r
v
e
n
t
i
o
n
s
,
v
o
l
.
2
5
,
p
p
.
1
–
8
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
n
v
e
n
t
.
2
0
2
1
.
1
0
0
4
2
2
.
[
2
3
]
F
.
H
a
q
u
e
,
R
.
U
.
N
u
r
,
S
.
A
.
J
a
h
a
n
,
Z
.
M
a
h
m
u
d
,
a
n
d
F
.
M
.
S
h
a
h
,
“
A
T
r
a
n
sf
o
r
me
r
B
a
se
d
A
p
p
r
o
a
c
h
T
o
D
e
t
e
c
t
S
u
i
c
i
d
a
l
I
d
e
a
t
i
o
n
U
si
n
g
Pre
-
T
r
a
i
n
e
d
L
a
n
g
u
a
g
e
M
o
d
e
l
s,
”
i
n
2
0
2
0
2
3
r
d
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
C
o
m
p
u
t
e
r
a
n
d
I
n
f
o
rm
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
(
I
C
C
I
T
)
,
2
0
2
0
,
p
p
.
1
–
5
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
C
I
T
5
1
7
8
3
.
2
0
2
0
.
9
3
9
2
6
9
2
.
[
2
4
]
R
.
Z
.
L
u
mo
n
t
o
d
I
I
I
,
“
S
e
e
i
n
g
t
h
e
i
n
v
i
s
i
b
l
e
:
Ex
t
r
a
c
t
i
n
g
si
g
n
s
o
f
d
e
p
r
e
ssi
o
n
a
n
d
su
i
c
i
d
a
l
i
d
e
a
t
i
o
n
f
r
o
m
c
o
l
l
e
g
e
s
t
u
d
e
n
t
s’
w
r
i
t
i
n
g
u
si
n
g
LI
W
C
a
c
o
mp
u
t
e
r
i
z
e
d
t
e
x
t
a
n
a
l
y
si
s,”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
R
e
se
a
rc
h
S
t
u
d
i
e
s
i
n
E
d
u
c
a
t
i
o
n
,
v
o
l
.
9
,
n
o
.
4
,
p
p
.
3
1
–
4
4
,
2
0
2
0
,
d
o
i
:
1
0
.
5
8
6
1
/
i
j
r
se
.
2
0
2
0
.
5
0
0
7
.
[
2
5
]
N
.
M
a
s
u
d
a
,
I
.
K
u
r
a
h
a
s
h
i
,
a
n
d
H
.
O
n
a
r
i
,
“
C
o
r
r
e
c
t
i
o
n
:
S
u
i
c
i
d
e
i
d
e
a
t
i
o
n
o
f
i
n
d
i
v
i
d
u
a
l
s
i
n
o
n
l
i
n
e
so
c
i
a
l
n
e
t
w
o
r
k
s (P
L
o
S
O
N
E)
,
”
PLo
S
O
N
E
,
v
o
l
.
9
,
n
o
.
1
,
A
p
r
.
2
0
1
4
,
d
o
i
:
1
0
.
1
3
7
1
/
a
n
n
o
t
a
t
i
o
n
/
d
5
8
9
8
5
7
d
-
b
3
c
6
-
4
a
1
6
-
a
c
f
e
-
4
2
3
f
9
b
f
5
2
9
f
1
.
[
2
6
]
J.
P
e
st
i
a
n
,
H
.
N
a
sr
a
l
l
a
h
,
P
.
M
a
t
y
k
i
e
w
i
c
z
,
A
.
B
e
n
n
e
t
t
,
a
n
d
A
.
L
e
e
n
a
a
r
s,
“
S
u
i
c
i
d
e
N
o
t
e
C
l
a
ssi
f
i
c
a
t
i
o
n
U
si
n
g
N
a
t
u
r
a
l
L
a
n
g
u
a
g
e
P
r
o
c
e
ssi
n
g
:
A
C
o
n
t
e
n
t
A
n
a
l
y
si
s,”
B
i
o
m
e
d
i
c
a
l
I
n
f
o
rm
a
t
i
c
s I
n
si
g
h
t
s
,
v
o
l
.
3
,
p
.
B
I
I
.
S
4
7
0
6
,
J
a
n
.
2
0
1
0
,
d
o
i
:
1
0
.
4
1
3
7
/
B
I
I
.
S
4
7
0
6
.
[
2
7
]
R
.
H
a
q
u
e
,
N
.
I
s
l
a
m,
M
.
I
sl
a
m,
a
n
d
M
.
M
.
A
h
san
,
“
A
C
o
mp
a
r
a
t
i
v
e
A
n
a
l
y
si
s
o
n
S
u
i
c
i
d
a
l
I
d
e
a
t
i
o
n
D
e
t
e
c
t
i
o
n
U
s
i
n
g
N
L
P
,
M
a
c
h
i
n
e
,
a
n
d
D
e
e
p
L
e
a
r
n
i
n
g
,
”
T
e
c
h
n
o
l
o
g
i
e
s
,
v
o
l
.
1
0
,
n
o
.
3
,
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
t
e
c
h
n
o
l
o
g
i
e
s1
0
0
3
0
0
5
7
.
[
2
8
]
M
.
M
.
T
a
d
e
sse
,
H
.
L
i
n
,
B
.
X
u
,
a
n
d
L
.
Y
a
n
g
,
“
D
e
t
e
c
t
i
o
n
o
f
D
e
p
r
e
ssi
o
n
-
R
e
l
a
t
e
d
P
o
st
s
i
n
R
e
d
d
i
t
S
o
c
i
a
l
M
e
d
i
a
F
o
r
u
m,”
I
EEE
Ac
c
e
ss
,
v
o
l
.
7
,
p
p
.
4
4
8
8
3
–
448
9
3
,
2
0
1
9
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
1
9
.
2
9
0
9
1
8
0
.
[
2
9
]
S
.
P
a
c
h
o
u
l
y
,
G
.
R
a
u
t
,
K
.
B
u
t
e
,
R
.
T
a
mb
e
,
a
n
d
S
.
B
h
a
v
sar,
“
D
e
p
r
e
ssi
o
n
D
e
t
e
c
t
i
o
n
o
n
S
o
c
i
a
l
M
e
d
i
a
N
e
t
w
o
r
k
(
T
w
i
t
t
e
r
)
u
si
n
g
S
e
n
t
i
me
n
t
A
n
a
l
y
si
s,”
I
n
t
e
r
n
a
t
i
o
n
a
l
R
e
sea
r
c
h
J
o
u
r
n
a
l
o
f
E
n
g
i
n
e
e
ri
n
g
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
8
,
n
o
.
1
,
p
p
.
1
8
3
4
–
1
8
3
9
,
2
0
2
1
[
3
0
]
M
.
S
t
a
n
k
e
v
i
c
h
,
A
.
L
a
t
y
sh
e
v
,
E.
K
u
mi
n
s
k
a
y
a
,
I
.
S
mi
r
n
o
v
,
a
n
d
O
.
G
r
i
g
o
r
i
e
v
,
“
D
e
p
r
e
ssi
o
n
d
e
t
e
c
t
i
o
n
f
r
o
m
so
c
i
a
l
me
d
i
a
t
e
x
t
s,”
i
n
S
e
l
e
c
t
e
d
P
a
p
e
rs
o
f
t
h
e
XXI
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
D
a
t
a
A
n
a
l
y
t
i
c
s
a
n
d
M
a
n
a
g
e
m
e
n
t
i
n
D
a
t
a
I
n
t
e
n
s
i
v
e
D
o
m
a
i
n
s
(
D
AM
D
I
D
/
RC
D
L
2
0
1
9
)
,
2
0
1
9
,
p
p
.
279
–
2
8
9
.
[
3
1
]
M
.
M
a
n
i
s
h
a
,
A
.
K
o
d
a
l
i
,
a
n
d
V
.
S
r
i
l
a
k
sh
mi
,
“
M
a
c
h
i
n
e
c
l
a
ss
i
f
i
c
a
t
i
o
n
f
o
r
s
u
i
c
i
d
e
i
d
e
a
t
i
o
n
d
e
t
e
c
t
i
o
n
o
n
t
w
i
t
t
e
r
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
I
n
n
o
v
a
t
i
v
e
T
e
c
h
n
o
l
o
g
y
a
n
d
E
x
p
l
o
ri
n
g
E
n
g
i
n
e
e
r
i
n
g
,
v
o
l
.
8
,
n
o
.
1
2
,
p
p
.
4
1
5
4
–
4
1
6
0
,
2
0
1
9
,
d
o
i
:
1
0
.
3
5
9
4
0
/
i
j
i
t
e
e
.
L
3
6
5
5
.
1
0
8
1
2
1
9
.
[
3
2
]
R
.
S
a
w
h
n
e
y
,
P
.
M
a
n
c
h
a
n
d
a
,
P
.
M
a
t
h
u
r
,
R
.
S
h
a
h
,
a
n
d
R
.
S
i
n
g
h
,
“
Ex
p
l
o
r
i
n
g
a
n
d
L
e
a
r
n
i
n
g
S
u
i
c
i
d
a
l
I
d
e
a
t
i
o
n
C
o
n
n
o
t
a
t
i
o
n
s
o
n
S
o
c
i
a
l
M
e
d
i
a
w
i
t
h
D
e
e
p
L
e
a
r
n
i
n
g
,
”
i
n
Pro
c
e
e
d
i
n
g
s
o
f
t
h
e
9
t
h
Wo
r
k
s
h
o
p
o
n
C
o
m
p
u
t
a
t
i
o
n
a
l
A
p
p
ro
a
c
h
e
s
t
o
S
u
b
j
e
c
t
i
v
i
t
y
,
S
e
n
t
i
m
e
n
t
a
n
d
S
o
c
i
a
l
M
e
d
i
a
A
n
a
l
y
s
i
s
,
O
c
t
.
2
0
1
8
,
p
p
.
1
6
7
–
1
7
5
,
d
o
i
:
1
0
.
1
8
6
5
3
/
v
1
/
W
1
8
-
6
2
2
3
.
[
3
3
]
S
.
Ji
,
C
.
P
.
Y
u
,
S
.
F
.
F
u
n
g
,
S
.
P
a
n
,
a
n
d
G
.
L
o
n
g
,
“
S
u
p
e
r
v
i
se
d
l
e
a
r
n
i
n
g
f
o
r
su
i
c
i
d
a
l
i
d
e
a
t
i
o
n
d
e
t
e
c
t
i
o
n
i
n
o
n
l
i
n
e
u
se
r
c
o
n
t
e
n
t
,
”
C
o
m
p
l
e
x
i
t
y
,
n
o
.
1
,
p
p
.
1
–
1
0
,
Ja
n
.
2
0
1
8
,
d
o
i
:
1
0
.
1
1
5
5
/
2
0
1
8
/
6
1
5
7
2
4
9
.
[
3
4
]
F
.
A
.
A
c
h
e
a
mp
o
n
g
,
H
.
N
u
n
o
o
-
M
e
n
s
a
h
,
a
n
d
W
.
C
h
e
n
,
“
T
r
a
n
sf
o
r
me
r
mo
d
e
l
s
f
o
r
t
e
x
t
-
b
a
se
d
e
mo
t
i
o
n
d
e
t
e
c
t
i
o
n
:
a
r
e
v
i
e
w
o
f
B
ER
T
-
b
a
se
d
a
p
p
r
o
a
c
h
e
s,”
Ar
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
Re
v
i
e
w
,
v
o
l
.
5
4
,
n
o
.
8
,
p
p
.
5
7
8
9
–
5
8
2
9
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
0
462
-
021
-
0
9
9
5
8
-
2.
[
3
5
]
V
.
V
e
n
e
k
,
S
.
S
c
h
e
r
e
r
,
L
.
P
.
M
o
r
e
n
c
y
,
A
.
S
.
R
i
z
z
o
,
a
n
d
J
.
P
e
st
i
a
n
,
“
A
d
o
l
e
sce
n
t
s
u
i
c
i
d
a
l
r
i
s
k
a
sse
ssm
e
n
t
i
n
c
l
i
n
i
c
i
a
n
-
p
a
t
i
e
n
t
i
n
t
e
r
a
c
t
i
o
n
:
A
st
u
d
y
o
f
v
e
r
b
a
l
a
n
d
a
c
o
u
st
i
c
b
e
h
a
v
i
o
r
s,”
i
n
2
0
1
4
I
EEE
W
o
rksh
o
p
o
n
S
p
o
k
e
n
L
a
n
g
u
a
g
e
T
e
c
h
n
o
l
o
g
y
,
S
L
T
2
0
1
4
-
Pro
c
e
e
d
i
n
g
s
,
A
p
r
.
2
0
1
4
,
p
p
.
2
7
7
–
2
8
2
,
d
o
i
:
1
0
.
1
1
0
9
/
S
L
T
.
2
0
1
4
.
7
0
7
8
5
8
7
.
[
3
6
]
E.
L
i
n
,
J.
S
u
n
,
H
.
C
h
e
n
,
a
n
d
M
.
H
.
M
a
h
o
o
r
,
“
D
a
t
a
Q
u
a
l
i
t
y
M
a
t
t
e
r
s:
S
u
i
c
i
d
e
I
n
t
e
n
t
i
o
n
D
e
t
e
c
t
i
o
n
o
n
S
o
c
i
a
l
M
e
d
i
a
P
o
st
s
U
s
i
n
g
a
R
o
B
E
R
T
a
-
C
N
N
M
o
d
e
l
,
”
a
r
Xi
v
p
re
p
r
i
n
t
,
p
p
.
1
3
–
1
7
,
2
0
24
,
d
o
i
:
1
0
.
1
1
0
9
/
E
M
B
C
5
3
1
0
8
.
2
0
2
4
.
1
0
7
8
2
6
4
7
.
[
3
7
]
S
.
B
i
r
d
,
E.
K
l
e
i
n
,
a
n
d
E.
L
o
p
e
r
,
N
a
t
u
r
a
l
L
a
n
g
u
a
g
e
Pr
o
c
e
ss
i
n
g
w
i
t
h
Py
t
h
o
n
.
2
0
0
9
.
[
3
8
]
C
.
D
.
M
a
n
n
i
n
g
,
P
.
R
a
g
h
a
v
a
n
,
a
n
d
H
.
S
c
h
ü
t
z
e
,
I
n
t
ro
d
u
c
t
i
o
n
t
o
I
n
f
o
rm
a
t
i
o
n
Re
t
r
i
e
v
a
l
.
U
S
A
:
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
.
[3
9]
J.
H
.
M
a
r
t
i
n
a
n
d
D
.
Ju
r
a
f
sk
y
,
S
p
e
e
c
h
a
n
d
L
a
n
g
u
a
g
e
Pr
o
c
e
ssi
n
g
.
S
t
a
n
f
o
r
d
,
20
2
5
.
[
4
0
]
R
.
F
e
l
d
man
a
n
d
J
.
S
a
n
g
e
r
,
T
h
e
T
e
x
t
Mi
n
i
n
g
H
a
n
d
b
o
o
k
:
A
d
v
a
n
c
e
d
A
p
p
ro
a
c
h
e
s
i
n
An
a
l
y
zi
n
g
U
n
st
r
u
c
t
u
re
d
D
a
t
a
.
C
a
mb
r
i
d
g
e
U
n
i
v
e
r
si
t
y
P
r
e
ss,
2
0
0
6
.
d
o
i
:
1
0
.
1
0
1
7
/
C
B
O
9
7
8
0
5
1
1
5
4
6
9
1
4
.
Evaluation Warning : The document was created with Spire.PDF for Python.
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
20
2
5
:
4
2
6
-
434
434
[
4
1
]
C
.
C
.
A
g
g
a
r
w
a
l
a
n
d
C
.
X
.
Z
h
a
i
,
Mi
n
i
n
g
t
e
x
t
d
a
t
a
,
v
o
l
.
9
7
8
1
4
6
1
4
3
2
2
3
4
.
2
0
1
3
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
1
-
4
6
1
4
-
3
2
2
3
-
4.
[
4
2
]
F
.
A
z
u
a
j
e
,
I
.
W
i
t
t
e
n
,
a
n
d
F
.
E,
“
D
a
t
a
M
i
n
i
n
g
:
P
r
a
c
t
i
c
a
l
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
T
o
o
l
s
a
n
d
T
e
c
h
n
i
q
u
e
s,
”
B
i
o
Me
d
i
c
a
l
En
g
i
n
e
e
ri
n
g
O
n
L
i
n
e
,
v
o
l
.
5
,
p
p
.
1
–
2
,
J
a
n
.
2
0
0
6
.
[
4
3
]
Y
.
L
i
u
e
t
a
l
.
,
“
R
o
B
ER
T
a
:
A
R
o
b
u
s
t
l
y
O
p
t
i
mi
z
e
d
B
E
R
T
P
r
e
t
r
a
i
n
i
n
g
A
p
p
r
o
a
c
h
,
”
a
rX
i
v
p
re
p
ri
n
t
n
o
.
1
,
2
0
1
9
,
d
o
i
:
1
0
.
4
8
5
5
0
/
a
r
X
i
v
.
1
9
0
7
.
1
1
6
9
2
.
[
4
4
]
Z
.
L
a
n
,
M
.
C
h
e
n
,
S
.
G
o
o
d
man
,
K
.
G
i
mp
e
l
,
P
.
S
h
a
r
ma,
a
n
d
R
.
S
o
r
i
c
u
t
,
“
A
L
B
ER
T:
A
L
i
t
e
B
ER
T
f
o
r
S
e
l
f
-
su
p
e
r
v
i
se
d
L
e
a
r
n
i
n
g
o
f
L
a
n
g
u
a
g
e
R
e
p
r
e
se
n
t
a
t
i
o
n
s,
”
a
r
Xi
v
p
re
p
ri
n
t
,
p
p
.
1
–
1
7
,
2
0
1
9
,
d
o
i
:
1
0
.
4
8
5
5
0
/
a
r
X
i
v
.
1
9
0
9
.
1
1
9
4
2
.
[
4
5
]
P
.
H
e
,
X
.
L
i
u
,
J.
G
a
o
,
a
n
d
W
.
C
h
e
n
,
“
D
e
B
ER
T
a
:
D
e
c
o
d
i
n
g
-
e
n
h
a
n
c
e
d
B
E
R
T
w
i
t
h
D
i
se
n
t
a
n
g
l
e
d
A
t
t
e
n
t
i
o
n
,
”
a
r
Xi
v
p
re
p
ri
n
t
,
2
0
2
0
,
d
o
i
:
1
0
.
4
8
5
5
0
/
a
r
X
i
v
.
2
0
0
6
.
0
3
6
5
4
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
M
e
r
in
d
a
Le
sta
n
d
y
is
a
re
se
a
rc
h
e
r
a
t
t
h
e
De
p
a
rtm
e
n
t
o
f
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
,
F
a
c
u
lt
y
o
f
En
g
in
e
e
rin
g
,
Un
iv
e
rsi
tas
M
u
h
a
m
m
a
d
i
y
a
h
M
a
lan
g
(U
M
M
),
M
a
lan
g
,
I
n
d
o
n
e
sia
.
S
h
e
re
c
e
iv
e
d
h
e
r
B
a
c
h
e
lo
r
’
s
a
n
d
M
a
ste
r
’
s
De
g
re
e
s
f
ro
m
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
a
n
d
In
f
o
rm
a
ti
c
s
De
p
a
rt
m
e
n
t,
Un
iv
e
rsitas
M
u
h
a
m
m
a
d
i
y
a
h
M
a
lan
g
,
a
n
d
Bra
w
ij
a
y
a
Un
iv
e
r
sit
y
,
In
d
o
n
e
sia
,
i
n
2
0
1
5
a
n
d
2
0
1
8
re
sp
e
c
ti
v
e
l
y
.
Cu
rre
n
tl
y
sh
e
is
a
se
n
io
r
l
e
c
tu
re
r
f
o
r
re
se
a
rc
h
f
ield
s
o
f
d
a
ta
m
in
in
g
,
se
n
ti
m
e
n
t
a
n
a
ly
si
s
,
tex
t
m
in
in
g
,
a
n
d
it
s
a
p
p
li
c
a
ti
o
n
s.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
m
e
rin
d
a
les
tan
d
y
@u
m
m
.
a
c
.
id
.
Abd
u
r
r
a
h
i
m
is
a
g
ra
d
u
a
te
stu
d
e
n
t
o
f
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
a
t
th
e
Un
iv
e
rsitas
M
u
h
a
m
m
a
d
i
y
a
h
M
a
lan
g
(UMM
)
w
it
h
th
e
stu
d
y
p
ro
g
ra
m
tak
e
n
is
E
lec
tri
c
a
l
E
n
g
in
e
e
rin
g
,
h
e
f
o
c
u
se
s
o
n
tele
m
a
ti
c
s
w
it
h
th
e
sc
o
p
e
o
f
se
n
ti
m
e
n
t
a
n
a
ly
sis.
Cu
rre
n
tl
y
h
e
is
p
u
rsu
in
g
a
M
a
ste
r
’
s d
e
g
re
e
a
t
th
e
Isla
m
ic Un
iv
e
rsit
y
o
f
In
d
o
n
e
sia
,
m
a
jo
rin
g
in
I
n
f
o
rm
a
ti
c
s,
f
o
c
u
sin
g
o
n
re
se
a
rc
h
in
th
e
f
ield
o
f
d
a
ta
sc
ien
c
e
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
2
2
9
1
7
0
0
2
@s
tu
d
e
n
ts.
u
ii
.
a
c
.
i
d
.
A
m
r
u
l
Fa
r
u
q
is
a
n
El
e
c
tri
c
a
l
En
g
in
e
e
r
a
n
d
Co
m
p
u
ter
S
c
ien
c
e
En
g
in
e
e
r
.
He
o
b
tai
n
e
d
B
a
c
h
e
lo
r
’
s
a
n
d
M
a
ste
r
’
s
d
e
g
re
e
in
E
lec
tri
c
a
l
E
n
g
in
e
e
rin
g
in
2
0
0
9
a
n
d
2
0
1
3
,
f
ro
m
Un
iv
e
rsitas
M
u
h
a
m
m
a
d
iy
a
h
M
a
lan
g
a
n
d
Un
iv
e
rsiti
T
e
k
n
o
lo
g
i
M
a
la
y
sia
,
re
sp
e
c
ti
v
e
l
y
.
His
P
h
.
D.
o
b
tain
e
d
f
ro
m
th
e
M
a
la
y
sia
-
Ja
p
a
n
In
tern
a
ti
o
n
a
l
In
stit
u
te
o
f
T
e
c
h
n
o
lo
g
y
(M
JII
T
),
Un
iv
e
rsiti
T
e
k
n
o
lo
g
i
M
a
la
y
s
ia,
Ku
a
la
L
u
m
p
u
r.
His
re
se
a
rc
h
in
te
re
sts
a
b
o
u
t
c
o
m
p
u
tati
o
n
a
l
d
a
ta sc
ien
c
e
a
n
d
o
p
ti
m
iza
ti
o
n
a
lg
o
rit
h
m
s
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
f
a
ru
q
@u
m
m
.
a
c
.
id
.
M
u
h
a
m
m
a
d
Ir
f
a
n
wa
s
b
o
rn
i
n
M
o
j
o
k
e
rto
,
In
d
o
n
e
sia
in
1
9
6
6
.
He
g
ra
d
u
a
ted
in
1
9
9
1
w
it
h
a
Ba
c
h
e
l
o
r
o
f
E
n
g
in
e
e
rin
g
d
e
g
re
e
,
f
ro
m
th
e
De
p
a
rtm
e
n
t
o
f
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
,
Bra
w
ij
a
y
a
Un
iv
e
r
sit
y
M
a
lan
g
,
a
n
d
a
M
a
ste
r
o
f
En
g
in
e
e
rin
g
in
2
0
0
0
f
ro
m
th
e
De
p
a
rt
m
e
n
t
o
f
In
f
o
rm
a
ti
c
s,
S
e
p
u
lu
h
No
p
e
m
b
e
r
I
n
stit
u
te
o
f
T
e
c
h
n
o
lo
g
y
(I
T
S
),
S
u
ra
b
a
y
a
.
Cu
rre
n
tl
y
,
h
e
is
a
se
n
io
r
lec
tu
re
r
a
t
th
e
Un
iv
e
rsit
y
o
f
M
u
h
a
m
m
a
d
i
y
a
h
M
a
lan
g
(UMM
)
a
n
d
is
a
c
ti
v
e
in
Re
se
a
rc
h
a
n
d
Co
m
m
u
n
it
y
S
e
rv
ic
e
.
He
is
c
u
rre
n
tl
y
p
u
rsu
in
g
h
is
d
o
c
to
ra
l
re
se
a
rc
h
p
ro
g
ra
m
a
t
th
e
Ra
z
a
k
F
a
c
u
l
ty
o
f
T
e
c
h
n
o
lo
g
y
a
n
d
In
f
o
rm
a
ti
c
s,
Un
iv
e
rsiti
T
e
k
n
o
lo
g
i
M
a
lay
sia
.
His
re
se
a
rc
h
in
tere
sts
a
re
in
re
n
e
w
a
b
le
e
n
e
rg
y
a
n
d
Co
m
p
u
ter
En
g
in
e
e
rin
g
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
irf
a
n
@u
m
m
.
a
c
.
id
.
Evaluation Warning : The document was created with Spire.PDF for Python.