I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
p
ute
r
E
ng
in
ee
ring
(
I
J
E
CE
)
Vo
l.
7
,
No
.
1
,
Feb
r
u
ar
y
201
7
,
p
p
.
3
3
0
~
3
3
6
I
SS
N:
2088
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v
7
i
1
.
p
p
3
3
0
-
3
3
6
330
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
jo
u
r
n
a
l.c
o
m/o
n
lin
e/in
d
ex
.
p
h
p
/I
JE
C
E
Sensing
Trending
To
pics in Tw
itt
er
f
o
r
G
reater
J
a
k
a
rta
Area
Ang
g
a
P
ra
t
a
m
a
Sito
rus
,
H
endri
M
urf
i
,
Siti
Nurro
h
m
a
h,
Af
if
Ak
ba
r
De
p
a
rte
m
e
n
t
o
f
M
a
th
e
m
a
ti
c
s
,
Un
iv
e
rsit
as
In
d
o
n
e
sia
,
De
p
o
k
1
6
4
2
4
-
In
d
o
n
e
sia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Au
g
1
,
2
0
1
6
R
ev
i
s
ed
Oct
2
1
,
2
0
1
6
A
cc
ep
ted
Dec
5
,
2
0
1
6
In
f
o
rm
a
ti
o
n
a
n
d
c
o
m
m
u
n
ica
ti
o
n
tec
h
n
o
l
o
g
y
g
ro
w
s
so
f
a
st
n
o
w
a
d
a
y
s,
e
sp
e
c
iall
y
re
late
d
to
th
e
in
ter
n
e
t
.
Tw
it
ter
is
o
n
e
o
f
in
tern
e
t
a
p
p
li
c
a
ti
o
n
s
t
h
a
t
p
ro
d
u
c
e
a
larg
e
a
m
o
u
n
t
o
f
tex
tu
a
l
d
a
ta
c
a
ll
e
d
tw
e
e
ts.
T
h
e
t
w
e
e
ts
m
a
y
re
p
re
se
n
t
re
a
l
-
w
o
rld
situ
a
ti
o
n
d
is
c
u
ss
e
d
in
a
c
o
m
m
u
n
it
y
.
T
h
e
re
f
o
re
,
Tw
it
te
r
c
a
n
b
e
a
n
i
m
p
o
rtan
t
m
e
d
ia
f
o
r
u
rb
a
n
m
o
n
it
o
rin
g
.
T
h
e
a
b
il
it
y
to
m
o
n
it
o
r
th
e
situ
a
ti
o
n
s
m
a
y
g
u
id
e
lo
c
a
l
g
o
v
e
rn
m
e
n
t
to
re
sp
o
n
d
q
u
ick
ly
o
r
m
a
k
e
p
u
b
li
c
p
o
li
c
y
.
T
o
p
ic
d
e
tec
ti
o
n
is
a
n
im
p
o
rtan
t
a
u
to
m
a
ti
c
to
o
l
to
u
n
d
e
rsta
n
d
th
e
tw
e
e
ts
,
f
o
r
e
x
a
m
p
le,
u
sin
g
n
o
n
-
n
e
g
a
ti
v
e
m
a
tri
x
f
a
c
to
riza
ti
o
n
.
I
n
t
h
is
p
a
p
e
r,
w
e
c
o
n
d
u
c
ted
a
st
u
d
y
to
im
p
le
m
e
n
t
T
w
it
ter
a
s
a
m
e
d
ia
f
o
r
th
e
u
rb
a
n
m
o
n
it
o
rin
g
in
Ja
k
a
rta
a
n
d
it
s
su
rro
u
n
d
in
g
a
re
a
s
c
a
ll
e
d
Gr
e
a
ter
J
a
k
a
rta
.
F
irstl
y
,
we
a
n
a
l
y
z
e
th
e
a
c
c
u
ra
c
y
o
f
th
e
d
e
tec
ted
t
o
p
ics
i
n
ter
m
o
f
th
e
ir
in
terp
re
tab
il
it
y
le
v
e
l
.
Ne
x
t,
w
e
v
isu
a
li
z
e
th
e
tren
d
o
f
th
e
to
p
ics
to
id
e
n
ti
f
y
p
o
p
u
lar
t
o
p
ics
e
a
sily
.
Ou
r
sim
u
la
ti
o
n
s
sh
o
w
th
a
t
th
e
t
o
p
ic
d
e
tec
ti
o
n
m
e
th
o
d
s
c
a
n
e
x
trac
t
to
p
ics
in
a
c
e
rtain
lev
e
l
o
f
a
c
c
u
ra
c
y
a
n
d
d
ra
w
th
e
tren
d
s
su
c
h
th
a
t
th
e
to
p
ic m
o
n
it
o
r
in
g
c
a
n
b
e
c
o
n
d
u
c
ted
e
a
sily
.
K
ey
w
o
r
d
:
C
o
h
er
en
ce
NM
F
T
o
p
ic
d
etec
tio
n
T
o
p
ic
s
en
s
i
n
g
T
w
it
ter
Co
p
y
rig
h
t
©
2
0
1
7
In
stit
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
Al
l
rig
h
ts
re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
An
g
g
a
P
r
ata
m
a
Si
to
r
u
s
,
Dep
ar
te
m
en
t o
f
Ma
th
e
m
atic
s
,
Un
i
v
er
s
it
as
I
n
d
o
n
esia,
Dep
o
k
1
6
4
2
4
-
I
n
d
o
n
esia,
E
m
ail:
ap
s
ito
r
u
s
@
s
ci.
u
i.a
c.
id
1.
I
NT
RO
D
UCT
I
O
N
Dev
elo
p
m
e
n
t
o
f
I
n
ter
n
et
tec
h
n
o
lo
g
y
h
a
s
m
ad
e
th
e
s
p
r
ea
d
o
f
i
n
f
o
r
m
atio
n
h
as
in
cr
ea
s
ed
s
i
g
n
i
f
ica
n
tl
y
.
On
e
o
f
th
e
I
n
ter
n
et
ap
p
licatio
n
s
t
h
at
s
u
p
p
o
r
t
th
e
s
p
r
ea
d
o
f
s
u
ch
i
n
f
o
r
m
atio
n
i
s
s
o
cial
m
ed
ia
[1
]
,
[
2]
.
I
n
s
o
cial
m
ed
ia,
u
s
er
s
n
o
t
o
n
l
y
as
co
n
s
u
m
er
s
o
f
i
n
f
o
r
m
atio
n
,
b
u
t
t
h
e
y
ca
n
ac
t
as
p
r
o
d
u
ce
r
o
f
in
f
o
r
m
atio
n
al
s
o
.
On
e
o
f
th
e
p
o
p
u
lar
s
o
cial
m
ed
ia
f
o
r
th
e
s
p
r
ea
d
o
f
in
f
o
r
m
atio
n
is
T
w
it
ter
.
T
w
itter
f
ac
i
litates
u
s
er
s
to
s
en
d
an
d
r
ea
d
tex
t
-
b
ased
in
f
o
r
m
a
tio
n
k
n
o
w
n
as
tw
ee
t
s
.
A
lo
n
g
w
i
th
ac
tiv
e
u
s
er
s
p
r
o
d
u
cin
g
t
w
ee
t
s
,
th
e
n
T
w
itter
ca
n
b
e
a
s
en
s
o
r
o
f
r
ea
l
-
w
o
r
ld
ev
e
n
t
s
[
3
]
.
An
o
th
er
s
i
m
i
lar
u
s
e
o
f
T
w
i
tter
is
as
a
s
en
s
o
r
o
f
s
itu
at
io
n
s
in
a
n
u
r
b
an
ar
ea
.
T
h
e
ab
ilit
y
to
m
o
n
i
to
r
th
e
s
it
u
atio
n
s
m
a
y
d
ir
ec
t
t
h
e
lo
ca
l
g
o
v
er
n
m
e
n
t
to
r
esp
o
n
d
q
u
ick
l
y
o
r
m
a
k
e
p
u
b
lic
p
o
licy
.
T
h
er
ef
o
r
e,
T
w
i
tter
is
a
p
o
ten
tial
m
ed
ia
f
o
r
an
u
r
b
an
m
o
n
ito
r
i
n
g
to
s
u
p
p
o
r
t a
s
m
ar
t
cit
y
[
4
].
To
p
ic
d
etec
tio
n
is
th
e
p
r
o
ce
s
s
o
f
d
eter
m
i
n
i
n
g
th
e
to
p
ics
o
f
a
s
et
o
f
tex
t
u
al
d
ata
[
5
]
.
T
h
is
p
r
o
ce
s
s
is
o
n
e
i
m
p
o
r
ta
n
t
s
tep
to
u
n
d
er
s
ta
n
d
in
g
t
h
e
t
w
ee
ts
w
h
ich
ar
e
te
x
tu
a
l
d
ata.
T
o
p
ic
d
etec
tio
n
ca
n
b
e
d
o
n
e
m
a
n
u
all
y
b
y
r
ea
d
in
g
t
h
e
co
n
te
n
ts
o
f
t
wee
ts
.
Ho
w
e
v
er
,
th
e
m
a
n
u
al
wa
y
is
n
o
t
f
ea
s
ib
le
f
o
r
lar
g
e
d
at
a
o
r
a
f
ast
r
esp
o
n
s
e
ti
m
e.
T
h
er
ef
o
r
e,
w
e
n
ee
d
a
u
t
o
m
a
tic
to
p
ic
d
etec
tio
n
m
e
t
h
o
d
s
to
m
ee
t
t
h
o
s
e
r
eq
u
ir
e
m
e
n
t
s
.
On
e
o
f
au
to
m
atic
to
p
ic
d
etec
tio
n
m
eth
o
d
s
w
id
el
y
u
s
ed
is
non
-
n
eg
a
tive
ma
tr
ix
fa
cto
r
iz
a
tio
n
(
NM
F)
[
6
].
B
ased
o
n
Se
m
io
ca
s
t
’
s
r
ep
o
r
t,
J
ak
ar
ta
is
th
e
m
o
s
t
ac
ti
v
e
cit
y
in
p
r
o
d
u
cin
g
t
w
ee
t
s
1
.
Hen
ce
,
T
w
itter
m
a
y
b
e
a
n
i
m
p
o
r
tan
t
m
ed
ia
f
o
r
u
r
b
an
m
o
n
ito
r
i
n
g
i
n
J
ak
ar
t
a
ar
ea
.
I
n
th
is
p
ap
er
,
w
e
co
n
d
u
cted
a
s
tu
d
y
o
f
t
h
e
ap
p
licatio
n
o
f
T
w
itter
a
s
a
m
ed
ia
f
o
r
t
h
e
u
r
b
a
n
m
o
n
ito
r
i
n
g
in
J
ak
ar
ta
a
n
d
its
s
u
r
r
o
u
n
d
in
g
r
e
g
io
n
s
ca
lled
Grea
ter
Ja
ka
r
ta
.
Firstl
y
,
w
e
an
al
y
ze
t
h
e
ac
cu
r
ac
y
o
f
th
e
g
en
er
ated
to
p
ics
in
ter
m
o
f
t
h
eir
in
ter
p
r
etab
ilit
y
lev
el.
P
o
in
t
w
is
e
m
u
t
u
al
in
f
o
r
m
atio
n
(
P
MI
)
is
a
q
u
an
t
itati
v
e
m
ea
s
u
r
e
to
ca
lcu
late
t
h
e
i
n
ter
p
r
etab
ilit
y
lev
el
[
7
]
.
1
h
tt
p
:/
/se
m
io
c
a
st.co
m
/en
/p
u
b
l
ica
ti
o
n
s/2
0
1
2
_
0
7
_
3
0
_
T
w
it
ter_
re
a
c
h
e
s
_
h
a
lf
_
a
_
b
il
li
o
n
_
a
c
c
o
u
n
ts_
1
4
0
m
_
in
_
t
h
e
_
US
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
J
E
C
E
Vo
l.
7
,
No
.
1
,
Feb
r
u
ar
y
2
0
1
7
:
3
3
0
–
336
331
B
y
u
s
i
n
g
th
is
m
ea
s
u
r
e,
w
e
s
el
ec
t
th
e
o
p
ti
m
al
n
u
m
b
er
o
f
to
p
ics
g
e
n
er
ated
f
o
r
a
p
er
io
d
.
Ne
x
t,
w
e
v
i
s
u
a
li
ze
th
e
f
r
eq
u
en
c
y
o
f
g
e
n
er
ated
to
p
ics
in
ti
m
e
s
er
ie
s
to
s
h
o
w
th
e
tr
en
d
o
f
t
h
e
to
p
ics.
T
h
is
v
i
s
u
al
izatio
n
h
e
lp
s
u
s
to
id
en
ti
f
y
p
o
p
u
lar
to
p
ics o
r
ex
tr
ao
r
d
in
ar
y
to
p
ics i
f
t
h
e
y
e
x
is
t.
T
h
e
o
u
tlin
e
o
f
th
is
p
ap
er
is
a
s
f
o
llo
w
s
:
I
n
Sec
tio
n
2
,
w
e
d
is
cu
s
s
t
h
e
r
esear
c
h
m
e
th
o
d
u
s
ed
in
th
is
s
tu
d
y
.
Sectio
n
3
d
escr
ib
es
s
o
m
e
r
e
s
u
lt
s
f
r
o
m
o
u
r
s
i
m
u
l
atio
n
s
.
Fi
n
all
y
,
a
g
e
n
er
al
co
n
clu
s
io
n
ab
o
u
t
t
h
e
s
i
m
u
lat
io
n
s
is
p
r
esen
ted
i
n
Se
ctio
n
4
.
2.
RE
S
E
ARCH
M
E
T
H
O
D
T
h
e
f
ir
s
t
s
tep
i
n
s
en
s
i
n
g
tr
en
d
in
g
to
p
ics
i
n
T
w
it
ter
is
co
ll
ec
tin
g
d
ata
b
y
ac
q
u
i
s
itio
n
t
wee
t
s
f
r
o
m
T
w
it
ter
.
T
h
e
a
cq
u
is
it
io
n
p
r
o
ce
s
s
u
s
e
s
T
w
itte
r
A
P
I
b
y
s
tr
ea
m
i
n
g
all
t
w
ee
t
s
p
o
s
ted
i
n
a
p
er
io
d
.
I
n
th
i
s
p
ap
er
,
w
e
f
o
cu
s
ed
o
n
t
w
ee
ts
f
r
o
m
J
ak
ar
t
a,
B
o
g
o
r
,
Dep
o
k
,
T
an
g
er
an
g
,
B
ek
asi
t
h
at
co
m
m
o
n
l
y
k
n
o
w
n
as
Gr
ea
ter
J
ak
ar
ta.
T
w
it
ter
f
ac
ilit
a
tes
t
h
e
t
w
ee
t
ac
q
u
is
itio
n
to
a
s
p
ec
if
ic
lo
ca
tio
n
b
ased
o
n
g
eo
g
r
ap
h
ic
co
o
r
d
in
ates
o
f
th
e
lo
ca
tio
n
.
T
h
ese
co
o
r
d
in
ates
co
n
s
i
s
ted
o
f
t
w
o
p
o
in
t
s
th
at
r
ep
r
esen
t
t
h
e
r
ec
tan
g
u
lar
s
h
ap
e
o
f
th
e
c
o
v
er
ag
e
ar
ea
.
E
ac
h
p
o
in
t
r
ep
r
esen
ts
t
h
e
t
w
o
co
m
p
o
n
en
ts
,
n
a
m
e
l
y
lo
n
g
itu
d
e
an
d
la
titu
d
e
.
Fo
r
th
e
Gr
ea
ter
J
a
k
ar
ta
ar
ea
,
th
e
f
ir
s
t
p
o
in
t
h
as
lat
itu
d
e
6
°1
'
4
1
.
9
2
"
S
an
d
lo
n
g
itu
d
e
1
0
6
°2
0
'
1
9
.
3
4
"E
,
w
h
ile
t
h
e
s
ec
o
n
d
p
o
in
t
h
as
lati
tu
d
e
6
°4
1
'
4
2
.
5
0
"
S
an
d
L
o
n
g
it
u
d
e
0
7
°1
9
'
4
.
9
3
"
E
.
T
h
e
t
w
o
p
o
in
ts
a
r
e
s
h
o
w
n
as t
w
o
y
ello
w
m
ar
k
s
in
Fi
g
u
r
e
1
.
Fig
u
r
e
1
.
L
atit
u
d
e
an
d
L
o
n
g
it
u
d
e
o
f
th
e
Gr
ea
ter
J
ak
ar
ta
ar
ea
Af
ter
co
llectin
g
th
e
t
w
ee
t
s
,
th
e
n
ex
t
p
r
o
ce
s
s
is
p
r
ep
ar
in
g
th
e
m
b
y
d
o
in
g
f
ea
tu
r
e
ex
tr
a
ctio
n
.
T
h
e
f
ea
t
u
r
e
ex
tr
ac
tio
n
p
r
o
ce
s
s
c
o
n
s
is
ts
o
f
p
r
ep
r
o
ce
s
s
in
g
,
to
k
en
iz
in
g
,
w
o
r
d
f
ilter
i
n
g
,
a
n
d
w
ei
g
h
t
in
g
.
T
h
e
p
r
ep
r
o
ce
s
s
in
g
i
s
ch
a
n
g
in
g
t
h
e
tex
t
o
n
ev
er
y
t
w
ee
t
i
n
to
lo
w
er
ca
s
e
a
n
d
d
eleti
n
g
e
v
er
y
l
in
k
u
s
er
n
a
m
e.
T
h
e
to
k
en
izi
n
g
p
r
o
ce
s
s
is
s
ep
ar
atin
g
w
o
r
d
s
in
t
w
ee
t
s
an
d
s
av
e
t
h
e
m
i
n
to
a
w
o
r
d
d
ictio
n
ar
y
.
T
h
e
w
o
r
d
d
ictio
n
ar
y
th
at
o
b
tain
ed
f
r
o
m
t
h
e
p
r
ev
io
u
s
p
r
o
ce
s
s
w
ill
b
e
f
i
lter
ed
b
y
s
to
p
w
o
r
d
s
to
g
et
th
e
m
o
s
t
p
o
ten
tial
w
o
r
d
s
th
at
co
n
tr
ib
u
te
to
th
e
m
ea
n
in
g
o
f
to
p
ics
.
E
ac
h
t
w
ee
t
is
r
ep
r
esen
ted
as
a
v
ec
to
r
w
h
o
s
e
f
ea
tu
r
e
s
ar
e
w
o
r
d
s
in
t
h
e
d
ictio
n
ar
y
.
T
h
e
w
ei
g
h
t
s
o
f
th
e
v
ec
to
r
ar
e
term
f
r
eq
u
en
cy
in
ve
r
s
e
d
o
cu
men
t
fr
eq
u
en
cy
(
t
f
-
id
f
)
[
6
]
.
Tf
-
id
f
is
d
ef
in
ed
b
y
(
)
w
h
er
e
(
)
r
ep
r
esen
t
s
w
ei
g
h
t
o
f
w
o
r
d
o
n
d
o
cu
m
en
t
,
r
ep
r
ese
n
t
f
r
eq
u
e
n
c
y
o
f
w
o
r
d
o
n
d
o
c
u
m
e
n
t
,
an
d
r
ep
r
esen
t
th
e
n
u
m
b
er
o
f
d
o
cu
m
en
t
s
t
h
at
co
n
tai
n
w
o
r
d
.
T
h
e
o
u
tp
u
t
o
f
th
e
f
ea
t
u
r
e
ex
tr
ac
tio
n
p
r
o
ce
s
s
is
a
w
o
r
d
b
y
t
w
ee
t
m
atr
ix
w
h
o
s
e
co
lu
m
n
s
ar
e
t
w
ee
t
s
.
No
n
-
n
e
g
ati
v
e
m
atr
ix
f
ac
to
r
iza
tio
n
(
NM
F)
i
s
an
au
to
m
atic
t
o
p
ic
d
etec
tio
n
m
et
h
o
d
to
ex
tr
ac
ts
to
p
ics
f
r
o
m
t
h
e
w
o
r
d
b
y
t
w
ee
t
m
atr
ix
A
.
Gi
v
en
th
e
m
a
tr
ix
A
,
N
MF
m
e
th
o
d
is
a
m
atr
i
x
f
ac
to
r
izatio
n
m
e
th
o
d
t
h
at
f
ac
to
r
izes t
h
e
m
atr
i
x
A
in
to
t
wo
n
o
n
-
n
e
g
ati
v
e
m
atr
ice
s
,
i.e
.
a
m
x
k
m
a
tr
ix
W
a
n
d
a
k
x
n
m
a
tr
ix
H
,
s
u
ch
t
h
at,
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
S
en
s
in
g
Tr
en
d
in
g
To
p
ics in
Tw
itter
fo
r
Grea
ter Ja
ka
r
ta
A
r
e
a
(
A
n
g
g
a
P
r
a
ta
m
a
S
ito
r
u
s
)
332
w
h
er
e
r
ep
r
esen
ts
t
h
e
n
u
m
b
er
o
f
w
o
r
d
s
o
r
th
e
d
i
m
e
n
s
io
n
o
f
th
e
t
w
ee
ts
,
a
n
d
r
ep
r
esen
ts
t
h
e
n
u
m
b
er
o
f
t
h
e
t
w
ee
t
s
,
an
d
is
t
h
e
n
u
m
b
er
o
f
to
p
ics.
A
cc
o
r
d
in
g
to
th
is
f
o
r
m
u
latio
n
,
th
e
co
lu
m
n
s
o
f
W
r
ep
r
ese
n
t
to
p
ics
ex
is
t
in
g
in
t
w
ee
ts
.
I
n
o
th
er
wo
r
d
s
,
a
to
p
ic
is
a
v
e
cto
r
o
f
w
o
r
d
s
.
Ho
w
e
v
er
,
it
u
s
u
all
y
u
s
e
s
o
n
l
y
to
p
f
r
eq
u
e
n
t
w
o
r
d
s
to
d
escr
ib
e
a
to
p
ic
,
i.e
.
to
p
ten
f
r
eq
u
en
t
w
o
r
d
s
.
W
h
ile
th
e
r
o
w
s
o
f
H
r
ep
r
esen
t
t
h
e
a
s
s
o
ciatio
n
b
et
w
ee
n
to
p
ics
an
d
t
w
ee
t
s
.
I
f
w
e
o
r
d
er
th
e
t
w
ee
ts
b
y
ti
m
e
t
h
en
w
e
ca
n
d
escr
ib
e
th
e
tr
en
d
o
f
ea
ch
to
p
ic
u
s
in
g
ea
ch
r
o
w
o
f
H
.
T
h
e
co
m
m
o
n
m
et
h
o
d
to
s
o
lv
e
NM
F
p
r
o
b
lem
is
iter
ati
v
e
m
et
h
o
d
s
b
y
m
i
n
i
m
izi
n
g
th
e
d
if
f
er
en
ce
s
b
et
w
ee
n
an
d
.
T
h
e
NM
F
p
r
o
b
le
m
i
s
f
o
r
m
u
lized
i
n
f
o
r
m
o
f
bound
-
co
n
s
tr
a
in
ed
o
p
timiz
a
tio
n
p
r
o
b
lem
as
f
o
llo
w
:
(
)
‖
‖
w
h
er
e
‖
‖
is
Fro
b
en
iu
s
No
r
m
.
Alth
o
u
g
h
th
e
f
u
n
ctio
n
f
(
W
,
H
)
is
co
n
v
e
x
o
n
th
e
W
o
r
H
,
h
o
w
e
v
er
,
it
is
n
o
t
co
n
v
e
x
o
n
t
h
e
W
an
d
H
.
T
h
u
s
,
th
e
r
ea
lis
tic
s
o
l
u
tio
n
o
f
NM
F
p
r
o
b
lem
i
s
a
lo
ca
l
m
in
i
m
u
m
[
9
]
.
T
h
er
e
ar
e
s
o
m
e
p
r
o
p
o
s
ed
m
et
h
o
d
s
to
s
o
lv
e
th
e
NM
F
p
r
o
b
lem
,
th
at
is
,
p
r
o
je
cted
g
r
a
d
ien
t
d
escen
t
[
10
]
,
mu
ltip
lica
tive
u
p
d
a
te
r
u
le
[
11
]
,
an
d
a
lter
n
a
tin
g
n
o
n
-
n
eg
a
tive
lea
s
t
s
q
u
a
r
e
(
A
NL
S
)
.
Fro
m
co
n
v
er
g
e
n
ce
p
o
in
t
o
f
v
ie
w
,
A
N
L
S
g
iv
e
s
b
etter
co
n
v
er
g
e
n
ce
co
n
d
itio
n
.
T
h
e
m
et
h
o
d
g
u
ar
an
tee
s
t
h
at
th
e
s
o
lu
tio
n
s
a
l
w
a
y
s
co
n
v
er
g
e
to
th
e
s
tatio
n
ar
y
p
o
in
ts
[
9
]
.
I
n
th
is
m
et
h
o
d
,
t
w
o
n
o
n
-
n
e
g
ati
v
e
leas
t
s
q
u
ar
e
(
N
L
S)
p
r
o
b
le
m
s
ar
e
s
o
l
v
ed
in
alt
er
n
atin
g
p
r
o
ce
s
s
as
s
h
o
w
n
i
n
A
l
g
o
r
it
h
m
1
.
T
h
er
ef
o
r
e,
th
e
m
ain
p
r
o
b
lem
o
f
th
e
A
N
L
S
m
et
h
o
d
is
h
o
w
to
s
o
lv
e
th
e
N
L
S
p
r
o
b
le
m
.
T
h
er
e
ar
e
s
o
m
e
m
et
h
o
d
s
to
s
o
lv
e
t
h
e
N
L
S
p
r
o
b
le
m
,
i.e
.
p
r
o
jecte
d
g
r
a
d
ien
t
d
escen
t
[1
2
]
,
a
ctive
-
s
et
[1
3
]
,
an
d
b
lo
ck
p
r
in
cip
a
l
p
ivo
tin
g
[1
4
]
.
I
n
o
u
r
s
i
m
u
latio
n
,
w
e
s
o
l
v
e
th
e
NL
S
p
r
o
b
le
m
u
s
i
n
g
t
h
e
p
r
o
j
ec
te
d
g
r
ad
ien
t
d
escen
t
m
e
th
o
d
.
P
r
o
j
ec
ted
g
r
ad
ien
t
d
escen
t
al
g
o
r
it
h
m
p
r
o
j
e
cts
a
n
e
g
ati
v
e
v
al
u
e
i
n
to
its
n
ea
r
est
n
o
n
-
n
eg
at
iv
e
v
alu
e,
t
h
at
is
,
0
.
Mo
r
e
o
v
er
,
w
e
in
itialize
W
u
s
in
g
s
i
n
g
u
lar
v
alu
e
d
ec
o
m
p
o
s
i
tio
n
b
ased
in
it
ializatio
n
[
1
5
]
.
W
e
u
s
e
s
c
ikit
-
lea
r
n
,
a
P
y
t
h
o
n
-
b
ase
d
lib
r
ar
y
,
to
i
m
p
le
m
en
t t
h
e
m
e
th
o
d
s
[
1
6
]
.
A
l
g
o
r
i
t
h
m
1
.
A
l
t
e
r
n
a
t
i
n
g
n
o
n
-
n
e
g
a
t
i
v
e
l
e
a
st
sq
u
a
r
e
a
l
g
o
r
i
t
h
m
1
.
i
n
p
u
t
:
A
;
O
u
t
p
u
t
:
W
,
H
2
.
i
=
0
3
.
i
n
i
si
a
l
i
sas
i
4
.
w
h
i
l
e
st
o
p
p
i
n
g
_
c
r
i
t
e
r
i
a
=
f
a
l
se
5
.
i
=
i
+
1
6
.
(
)
(
(
)
)
7
.
(
)
(
(
)
)
8
.
e
n
d
w
h
i
l
e
B
ef
o
r
e
ex
ec
u
ti
n
g
an
NM
F
al
g
o
r
ith
m
,
w
e
m
u
s
t
s
et
t
h
e
v
al
u
e
o
f
t
h
e
p
ar
a
m
eter
k
.
T
h
e
p
ar
a
m
eter
k
r
ep
r
esen
ts
th
e
n
u
m
b
er
o
f
to
p
ics
to
b
e
ex
tr
ac
ted
.
Ho
w
to
f
i
n
d
th
e
o
p
tim
a
l
n
u
m
b
er
o
f
to
p
ics
is
th
e
p
r
o
b
le
m
i
n
mo
d
el
s
elec
tio
n
o
f
to
p
ic
d
etec
tio
n
u
s
i
n
g
NM
F.
Fo
r
th
is
ta
s
k
,
w
e
ca
lcu
late
t
h
e
ac
c
u
r
ac
ies
o
f
to
p
ics
f
o
r
s
o
m
e
n
u
m
b
er
s
,
i.e
.
1
0
,
2
0
,
…,
1
0
0
.
T
h
e
n
u
m
b
er
t
h
at
g
i
v
e
s
th
e
b
es
t a
cc
u
r
ac
y
is
c
h
o
s
e
n
as t
h
e
o
p
ti
m
al
n
u
m
b
er
.
Af
ter
ex
tr
ac
tin
g
to
p
ics,
w
e
c
alcu
late
t
h
e
a
v
er
ag
e
ac
c
u
r
ac
y
o
f
t
h
e
to
p
ics
u
s
i
n
g
p
o
in
tw
is
e
mu
tu
a
l
in
fo
r
ma
tio
n
(
P
MI
)
[
7
]
.
P
MI
s
h
o
w
s
t
h
e
co
h
er
en
ce
v
al
u
es
b
et
w
ee
n
w
o
r
d
s
in
a
to
p
ic
to
s
h
o
w
th
e
to
p
ic’
s
le
v
el
o
f
in
ter
p
r
etab
ilit
y
.
L
et
-
th
to
p
ic
o
f
th
e
m
atr
i
x
W
co
n
s
is
t
s
o
f
w
o
r
d
s
,
th
a
t
is
(
)
,
P
MI
s
co
r
e
o
f
th
e
t
-
t
h
to
p
ic
is
(
)
∑
∑
(
)
(
)
(
)
w
h
er
e
(
)
is
t
h
e
j
o
in
t p
r
o
b
ab
ilit
y
o
f
w
o
r
d
s
an
d
o
r
co
m
e
s
u
p
to
g
eth
er
w
it
h
w
o
r
d
,
(
)
an
d
(
)
ar
e
th
e
in
d
i
v
id
u
al
p
r
o
b
ab
ilit
y
o
f
w
o
r
d
an
d
.
T
h
e
h
ig
h
er
th
e
P
MI
s
co
r
e
o
n
a
to
p
ic
m
ea
n
s
m
o
r
e
co
h
er
en
t
w
o
r
d
s
o
n
t
h
e
to
p
ic.
I
n
o
th
er
w
o
r
d
s
,
t
h
e
i
n
ter
p
r
etatio
n
o
f
th
e
to
p
ic
is
ea
s
ier
.
W
e
n
ee
d
a
r
eferen
ce
co
r
p
u
s
to
ca
lcu
la
te
th
o
s
e
p
r
o
b
ab
ilit
ies.
I
n
o
u
r
s
i
m
u
la
tio
n
,
w
e
u
s
e
t
w
o
r
ef
er
e
n
ce
co
r
p
u
s
es,
th
a
t
is
,
WikiId
w
h
ich
is
I
n
d
o
n
e
s
ia
n
W
ik
ip
ed
ia
d
u
m
p
f
r
o
m
J
an
u
ar
y
2
0
1
6
an
d
B
erit
a
w
h
ic
h
i
s
a
co
llectio
n
o
f
I
n
d
o
n
esia
n
n
e
w
s
.
W
e
u
s
e
a
n
o
p
en
-
s
o
u
r
ce
to
o
lk
it to
ca
lcu
la
te
th
e
P
MI
s
co
r
es
2
.
2
h
tt
p
s://
g
it
h
u
b
.
c
o
m
/j
h
lau
/t
o
p
ic_
i
n
terp
re
tab
il
it
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
J
E
C
E
Vo
l.
7
,
No
.
1
,
Feb
r
u
ar
y
2
0
1
7
:
3
3
0
–
336
333
3.
RE
SU
L
T
S AN
D
AN
AL
Y
SI
S
Fo
r
th
e
s
i
m
u
la
tio
n
s
,
w
e
u
s
e
t
wo
ty
p
es
o
f
d
ata
b
ased
o
n
a
p
er
i
o
d
.
T
h
e
f
ir
s
t
d
ata
is
d
ail
y
d
ata
co
llected
o
n
Ma
r
ch
1
2
,
2016
an
d
Ma
r
ch
1
3
,
2
0
1
6
.
T
h
e
s
ec
o
n
d
o
n
e
i
s
m
o
n
th
l
y
d
ata
co
llec
ted
in
Ma
r
ch
2
0
1
6
an
d
A
p
r
il
2
0
1
6
.
T
h
e
n
u
m
b
er
o
f
t
w
ee
t
s
is
ap
p
r
o
x
i
m
atel
y
3
3400
0
tw
e
ets
p
er
m
o
n
th
a
n
d
1
6000
t
w
ee
ts
p
er
d
ay
.
A
f
ter
p
r
ep
r
o
ce
s
s
in
g
s
tep
,
th
e
n
u
m
b
er
o
f
w
o
r
d
s
i
s
ap
p
r
o
x
i
m
ate
l
y
9
5
0
0
w
o
r
d
s
p
er
m
o
n
t
h
a
n
d
8
5
0
0
w
o
r
d
s
p
er
d
a
y
.
Th
e
d
etail
s
tatis
tics
o
f
th
e
d
ata
is
g
i
v
en
i
n
Fi
g
u
r
e
2
.
Fig
u
r
e
2
.
T
h
e
St
atis
tics
o
f
Da
ta
T
o
f
in
d
t
h
e
ac
c
u
r
ac
y
o
f
e
x
tr
a
cted
to
p
ics
f
o
r
a
p
er
io
d
,
w
e
s
i
m
u
late
t
h
e
ac
c
u
r
ac
ies
f
o
r
s
o
m
e
n
u
m
b
er
s
o
f
to
p
ics,
i.e
.
1
0
,
2
0
,
…,
1
0
0
.
T
h
e
b
est
ac
cu
r
ac
y
of
t
h
e
s
e
n
u
m
b
er
s
o
f
t
h
e
to
p
ic
r
ep
r
esen
ts
t
h
e
ac
cu
r
ac
y
o
f
t
h
e
p
er
io
d
.
Fig
u
r
e
3
an
d
Fig
u
r
e
4
s
h
o
w
t
h
e
ac
cu
r
ac
y
o
f
e
x
tr
ac
ted
to
p
ics
in
Ma
r
ch
2
0
1
6
an
d
A
p
r
il
2
0
1
6
,
r
esp
ec
tiv
el
y
.
Fro
m
b
o
t
h
f
ig
u
r
e
s
,
w
e
s
ee
th
at
th
e
r
ef
er
e
n
ce
co
r
p
u
s
W
ik
iI
d
al
w
a
y
s
g
i
v
e
h
i
g
h
er
P
MI
s
co
r
es
th
a
n
th
e
r
e
f
er
en
ce
co
r
p
u
s
B
er
ita
f
o
r
all
n
u
m
b
er
s
o
f
t
h
e
to
p
ic
.
O
n
e
r
ea
s
o
n
f
o
r
th
e
s
e
r
es
u
lt
s
i
s
t
h
at
t
h
e
d
o
cu
m
e
n
t
s
o
f
W
ik
iI
d
ar
e
lo
n
g
er
t
h
an
o
n
es
o
f
B
er
ita.
T
h
er
ef
o
r
e,
th
e
p
r
o
b
ab
ilit
y
t
h
at
t
w
o
w
o
r
d
s
e
x
is
t
i
n
a
s
i
m
ilar
d
o
cu
m
e
n
t
o
f
W
ik
iI
d
is
h
ig
h
er
th
a
n
th
e
p
r
o
b
ab
ilit
y
o
f
t
h
e
w
o
r
d
s
in
a
s
i
m
ilar
d
o
cu
m
e
n
t
o
f
B
er
ita.
Fro
m
Fi
g
u
r
e
3
,
w
e
ca
n
s
ee
th
at
th
e
h
ig
h
er
ac
cu
r
ac
y
i
s
ac
h
iev
ed
w
h
e
n
w
e
ex
tr
ac
t
ten
to
p
ics
f
r
o
m
t
w
ee
t
s
in
Ma
r
ch
2
0
1
6
.
I
t
m
ea
n
s
th
at
th
e
o
p
ti
m
a
l
n
u
m
b
er
o
f
to
p
ics
f
o
r
th
at
p
er
io
d
is
te
n
w
it
h
a
P
MI
s
co
r
e
o
f
0
.
4
3
.
W
e
ca
n
also
s
ee
f
r
o
m
Fig
u
r
e
4
th
at
t
h
e
h
i
g
h
er
P
MI
s
co
r
e
i
s
0
.
5
1
in
A
p
r
il
2
0
1
6
.
T
h
is
h
i
g
h
er
s
co
r
e
is
ac
h
iev
ed
w
h
e
n
w
e
ex
tr
ac
t
te
n
to
p
ics
f
r
o
m
t
w
ee
ts
i
n
t
h
at
p
er
io
d
.
T
h
is
o
p
ti
m
al
n
u
m
b
er
o
f
to
p
ics
is
s
i
m
ilar
to
th
e
o
n
e
e
x
tr
ac
ted
f
r
o
m
th
e
p
r
ev
io
u
s
m
o
n
t
h
.
He
n
ce
,
ten
is
t
h
e
f
ir
s
t
ca
n
d
id
ate
o
f
t
h
e
n
u
m
b
er
o
f
to
p
ics
w
h
e
n
w
e
e
x
tr
ac
t
to
p
ics
f
o
r
n
e
x
t
m
o
n
th
s
,
esp
ec
iall
y
,
w
h
en
w
e
n
ee
d
a
f
a
s
t r
esp
o
n
d
ti
m
e
i
n
to
p
ic
d
etec
tio
n
f
o
r
th
e
m
o
n
t
h
l
y
t
w
ee
t
s
.
Nex
t,
w
e
co
n
d
u
ct
s
i
m
ilar
s
i
m
u
latio
n
s
f
o
r
d
ail
y
d
ata,
i.e
.
Ma
r
ch
1
2
,
2016
an
d
Ma
r
ch
1
3
,
2
0
1
6
.
Fig
u
r
e
5
an
d
Fig
u
r
e
6
v
is
u
ali
ze
th
e
ac
cu
r
ac
y
o
f
ex
tr
ac
ted
to
p
ics
o
n
Ma
r
ch
12
,
2
0
1
6
an
d
Ma
r
ch
1
3
,
2
0
1
6
,
r
esp
ec
tiv
el
y
.
Fro
m
b
o
th
f
i
g
u
r
e
s
,
w
e
s
ee
t
h
at
t
h
e
r
e
f
er
en
ce
co
r
p
u
s
W
ik
iI
d
s
t
ill
g
iv
e
h
i
g
h
er
P
MI
s
co
r
es
th
a
n
t
h
e
r
ef
er
en
ce
co
r
p
u
s
B
er
ita
f
o
r
all
n
u
m
b
er
s
o
f
t
h
e
to
p
ic
.
Fro
m
Fig
u
r
e
5
,
w
e
ca
n
s
ee
th
at
t
h
e
h
ig
h
er
ac
cu
r
ac
y
i
s
ac
h
iev
ed
w
h
en
w
e
ex
tr
ac
t
f
o
r
t
y
to
p
ics
f
r
o
m
t
w
ee
t
s
o
n
Ma
r
ch
12
,
2
0
1
6
.
T
h
ese
ex
tr
ac
ted
to
p
ics
h
av
e
a
P
MI
s
co
r
e
o
f
0
.
3
0
.
W
e
ca
n
also
s
ee
f
r
o
m
Fi
g
u
r
e
5
th
at
t
h
e
h
ig
h
er
P
MI
s
co
r
e
is
0
.
3
0
o
n
Ma
r
ch
1
3
,
2
0
1
6
.
T
h
is
h
ig
h
er
P
MI
s
co
r
e
is
ac
h
ie
v
ed
w
h
e
n
w
e
e
x
tr
ac
t
f
o
r
t
y
t
o
p
ics f
r
o
m
t
w
ee
ts
in
t
h
at
p
er
io
d
.
T
h
is
o
p
ti
m
al
n
u
m
b
er
o
f
to
p
ics
is
s
i
m
ilar
to
th
e
o
n
e
e
x
t
r
ac
ted
f
r
o
m
t
h
e
p
r
ev
io
u
s
d
a
y
.
T
h
er
ef
o
r
e
,
f
o
r
ty
ca
n
b
e
t
h
e
f
ir
s
t
ca
n
d
id
ate
o
f
t
h
e
o
p
tim
a
l
n
u
m
b
er
of
d
ail
y
to
p
ics
if
w
e
do
n
o
t
w
a
n
t
to
p
er
f
o
r
m
m
o
d
el
s
elec
tio
n
i
n
to
p
ic
d
etec
tio
n
f
o
r
th
e
d
ail
y
t
w
ee
t
s
.
Af
ter
esti
m
ati
n
g
th
e
ac
cu
r
ac
y
o
f
th
e
g
en
er
ated
to
p
ic
s
,
th
e
n
ex
t
s
tep
is
to
v
is
u
al
ize
th
e
tr
en
d
o
f
th
e
to
p
ic
s
.
T
h
is
v
is
u
aliza
tio
n
ca
n
u
s
e
t
h
e
to
p
ic
b
y
t
w
ee
t
m
a
tr
i
x
H
w
h
ich
r
ep
r
ese
n
ts
t
h
e
d
eg
r
ee
o
f
ass
o
ciat
io
n
b
et
w
ee
n
t
o
p
ics
an
d
t
w
ee
t
s
.
If
w
e
a
s
s
u
m
e
th
at
t
w
ee
t
s
h
a
v
e
b
ee
n
s
o
r
ted
b
y
ti
m
e,
th
e
n
w
e
c
an
d
r
a
w
t
h
e
tr
e
n
d
o
f
a
to
p
ic
b
ased
o
n
t
h
e
a
s
s
o
ciati
o
n
v
a
lu
e
s
b
et
w
ee
n
t
h
e
to
p
ic
a
n
d
ea
ch
t
w
ee
t
.
T
h
e
h
i
g
h
er
th
e
ass
o
ciatio
n
v
al
u
es
,
th
en
t
h
e
h
ig
h
er
th
e
p
o
p
u
lar
it
y
o
f
th
e
to
p
ic.
Vis
u
aliza
tio
n
o
f
th
o
s
e
tr
en
d
s
allo
w
s
u
s
to
m
o
n
ito
r
to
p
ics
f
o
r
a
ce
r
tain
p
er
io
d
.
M
a
rc
h
Ap
ri
l
1
2
-M
a
r
1
3
-M
a
r
Wo
rd
s
9
5
0
3
9
5
2
9
8
3
1
1
8
5
8
2
Twe
e
t
s
3
3
4
1
7
3
3
3
3
8
8
1
1
5
5
5
2
1
6
3
8
8
0
50
1
0
0
1
5
0
2
0
0
2
5
0
3
0
0
3
5
0
4
0
0
Th
o
u
san
d
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
S
en
s
in
g
Tr
en
d
in
g
To
p
ics in
Tw
itter
fo
r
Grea
ter Ja
ka
r
ta
A
r
e
a
(
A
n
g
g
a
P
r
a
ta
m
a
S
ito
r
u
s
)
334
Fig
u
r
e
3
.
T
h
e
A
cc
u
r
ac
y
o
f
T
o
p
ics E
x
tr
ac
ted
in
Ma
r
ch
2016
Fig
u
r
e
4
.
T
h
e
Ac
cu
r
ac
y
o
f
T
o
p
ics E
x
tr
ac
ted
in
A
p
r
il
2016
Fig
u
r
e
5
.
T
h
e
A
cc
u
r
ac
y
o
f
T
o
p
ics
E
x
tr
ac
ted
o
n
Ma
r
ch
12
,
2
0
1
6
Fig
u
r
e
6
.
T
h
e
ac
cu
r
ac
y
o
f
to
p
i
cs
ex
tr
ac
ted
o
n
Ma
r
ch
13
,
2
0
1
6
0.
00
0.
10
0.
20
0.
30
0.
40
0.
50
10
20
30
40
50
60
70
80
90
100
P
M
I
T
h
e
n
u
m
be
r
of
topi
c
s
W
iki
I
d
B
e
r
it
a
0.
00
0.
10
0.
20
0.
30
0.
40
0.
50
10
20
30
40
50
60
70
80
90
100
P
M
I
T
h
e
n
u
m
be
r
of
topi
c
s
W
iki
I
d
B
e
r
it
a
0
.
0
0
0
.
1
0
0
.
2
0
0
.
3
0
10
20
30
40
50
60
70
80
90
100
P
M
I
T
h
e
n
u
m
be
r
of
topi
c
s
W
iki
I
d
B
e
r
it
a
0.
00
0.
10
0.
20
0.
30
10
20
30
40
50
60
70
80
90
100
P
M
I
T
h
e
n
u
m
be
r
of
topi
c
s
W
iki
I
d
B
e
r
it
a
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
J
E
C
E
Vo
l.
7
,
No
.
1
,
Feb
r
u
ar
y
2
0
1
7
:
3
3
0
–
336
335
Fig
u
r
e
7
d
r
aw
s
th
e
tr
en
d
s
o
f
4
0
to
p
ics
ex
tr
ac
ted
o
n
Ma
r
ch
1
2
,
2
0
1
6
.
Fro
m
Fig
u
r
e
7
w
e
c
an
s
ee
th
e
p
o
p
u
lar
to
p
ics
an
d
th
eir
p
o
p
u
lar
it
y
ch
a
n
g
es
o
v
er
t
h
e
p
er
io
d
.
Du
e
to
its
p
o
p
u
lar
it
y
i
n
m
a
n
y
ap
p
licatio
n
s
,
w
e
also
v
is
u
alize
t
h
e
tr
en
d
s
o
f
1
0
0
to
p
ics
ex
tr
ac
ted
in
th
e
s
a
m
e
p
er
io
d
in
Fig
u
r
e
8
.
Fro
m
Fig
u
r
e
8
,
w
e
ca
n
id
en
ti
f
y
p
o
p
u
lar
to
p
ics
ea
s
i
l
y
b
ec
au
s
e
th
e
to
p
ics
b
ec
o
m
e
m
o
r
e
s
p
ec
i
f
ic
s
o
th
at
ea
ch
t
wee
t
is
ass
o
ciate
d
to
m
o
r
e
r
elev
a
n
t
to
p
ics.
Fo
r
ex
a
m
p
le,
th
er
e
is
a
to
p
ic
p
r
esen
ted
b
y
g
r
ee
n
c
u
r
v
e
s
u
d
d
en
l
y
i
n
cr
ea
s
e
at
a
ce
r
tain
p
er
io
d
.
T
h
e
in
cr
ea
s
in
g
c
u
r
v
e
i
m
p
lies
t
h
at
th
e
to
p
ic
is
s
u
b
m
i
t
ted
in
te
n
s
iv
e
l
y
i
n
th
at
ti
m
e
b
u
t
n
o
t
in
o
th
er
ti
m
e.
Ho
w
e
v
er
,
th
e
r
is
i
n
g
to
p
ics
ar
e
n
o
t
al
w
a
y
s
to
b
e
i
m
p
o
r
tan
t
to
p
ics,
i.e
.
s
p
am
t
w
ee
ts
.
He
n
c
e,
w
e
n
ee
d
f
u
r
t
h
er
an
al
y
s
is
to
k
n
o
w
w
h
at
th
e
to
p
ics
ar
e
ab
o
u
t
.
Mo
r
eo
v
er
,
t
h
e
i
n
ter
p
r
etab
ilit
y
s
co
r
e
o
f
1
0
0
to
p
ics
is
s
m
aller
t
h
a
n
o
n
e
o
f
4
0
to
p
ics
as
s
h
o
w
n
i
n
Fig
u
r
e
5
.
A
p
o
s
s
ib
le
ca
u
s
e
o
f
th
is
r
es
u
lt
i
s
th
at
t
h
e
u
n
p
o
p
u
l
ar
to
p
ics
h
av
e
lo
w
P
MI
s
co
r
es.
T
h
er
ef
o
r
e,
if
w
e
co
n
s
id
er
o
n
l
y
t
h
e
tr
en
d
s
o
f
p
o
p
u
lar
to
p
ics,
th
e
n
w
e
m
a
y
e
x
tr
ac
t
1
0
0
to
p
ics
to
m
o
n
ito
r
t
h
e
m
ea
s
il
y
.
Fig
u
r
e
7
.
T
r
en
d
s
o
f
4
0
to
p
ics
ex
tr
ac
ted
o
n
Ma
r
ch
1
2
,
2
0
1
6
Fig
u
r
e
8
.
T
r
en
d
s
o
f
1
0
0
to
p
ics
ex
tr
ac
ted
o
n
Ma
r
ch
1
2
,
2
0
1
6
W
e
co
n
d
u
ct
s
i
m
ilar
s
i
m
u
lat
io
n
s
f
o
r
t
w
ee
ts
co
llected
o
n
Ma
r
ch
1
3
,
2
0
1
6
.
Fig
u
r
e
9
s
h
o
w
s
th
e
tr
en
d
s
o
f
4
0
to
p
ics
ex
tr
ac
ted
o
n
Ma
r
ch
1
3
,
2
0
1
6
,
w
h
ile
Fi
g
u
r
e
10
g
iv
e
s
th
e
tr
e
n
d
s
o
f
1
0
0
to
p
ics
ex
tr
ac
ted
o
n
a
s
i
m
ilar
d
ate.
Si
m
ilar
to
t
h
e
p
r
ev
io
u
s
d
a
y
,
w
e
ca
n
id
e
n
ti
f
y
p
o
p
u
lar
to
p
ics
m
o
r
e
ea
s
il
y
i
n
1
0
0
to
p
ics
th
an
i
n
4
0
to
p
ics.
T
h
er
e
is
also
a
to
p
ic
t
h
at
s
u
d
d
en
l
y
i
n
cr
ea
s
es
at
a
ce
r
tain
p
er
io
d
.
T
h
e
to
p
ic
h
as
a
s
i
m
ilar
tr
en
d
to
th
e
to
p
ic
f
r
o
m
t
h
e
p
r
ev
io
u
s
d
a
y
.
Af
ter
f
u
r
t
h
er
a
n
al
y
s
is
,
w
e
r
ea
li
ze
th
at
t
h
e
to
p
ic
i
s
an
ad
v
er
tis
e
m
en
t.
T
h
ese
r
es
u
lt
s
s
u
p
p
o
r
t o
u
r
ea
r
lier
s
u
g
g
est
io
n
th
at
w
e
m
a
y
e
x
tr
ac
t 1
0
0
to
p
ics
to
m
o
n
ito
r
th
e
m
o
s
t
f
r
eq
u
en
t t
o
p
ics ea
s
il
y
.
Fig
u
r
e
9
.
T
r
en
d
s
o
f
4
0
to
p
ics
ex
tr
ac
ted
o
n
Ma
r
ch
1
3
,
2
0
1
6
Fig
u
r
e
10
.
T
r
en
d
s
o
f
1
0
0
to
p
ic
s
ex
tr
ac
ted
o
n
Ma
r
ch
1
3
,
2
0
1
6
4.
CO
NCLU
SI
O
N
T
o
p
ic
d
etec
tio
n
,
s
u
ch
as
n
o
n
-
n
eg
at
iv
e
m
atr
ix
f
ac
to
r
izatio
n
,
ca
n
r
ea
lize
T
w
itter
as
a
m
ed
ia
f
o
r
u
r
b
an
m
o
n
ito
r
i
n
g
i
n
Gr
ea
ter
J
ak
ar
ta
ar
ea
.
Usi
n
g
i
n
ter
p
r
etab
ilit
y
o
r
co
h
er
en
ce
s
co
r
es,
w
e
s
h
o
w
th
at
th
e
e
x
tr
ac
ted
to
p
ics
h
av
e
s
i
m
ilar
s
co
r
es
to
t
h
e
p
r
ev
io
u
s
s
i
m
u
l
atio
n
s
f
o
r
o
th
er
tex
t
u
al
d
ata.
Mo
r
eo
v
er
,
th
e
r
ef
er
en
ce
co
r
p
u
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
S
en
s
in
g
Tr
en
d
in
g
To
p
ics in
Tw
itter
fo
r
Grea
ter Ja
ka
r
ta
A
r
e
a
(
A
n
g
g
a
P
r
a
ta
m
a
S
ito
r
u
s
)
336
W
ik
iI
d
g
iv
e
s
h
i
g
h
er
s
co
r
es
th
an
th
e
r
ef
er
e
n
ce
co
r
p
u
s
B
er
ita
f
o
r
ev
er
y
s
i
m
u
latio
n
.
T
h
er
ef
o
r
e,
W
ik
iI
d
m
a
y
b
e
th
e
f
ir
s
t
ca
n
d
id
ate
in
ca
lcu
la
ti
n
g
t
h
e
co
h
er
en
ce
s
co
r
es
in
I
n
d
o
n
esian
.
I
n
th
e
v
is
u
aliza
t
io
n
o
f
tr
en
d
s
,
w
e
ca
n
id
en
ti
f
y
p
o
p
u
lar
to
p
ics
m
o
r
e
ea
s
il
y
i
n
1
0
0
to
p
ics
th
an
in
th
e
o
p
ti
m
al
n
u
m
b
er
o
f
to
p
ics.
T
h
er
ef
o
r
e,
if
w
e
co
n
s
id
er
o
n
l
y
t
h
e
tr
en
d
s
o
f
p
o
p
u
lar
to
p
ics
,
th
en
w
e
m
a
y
e
x
tr
ac
t 1
0
0
to
p
ics
to
m
o
n
ito
r
th
e
m
ea
s
il
y
.
RE
F
E
R
E
NC
E
S
[1
]
S
.
G
o
y
a
l,
"
F
a
c
e
b
o
o
k
,
Tw
it
ter,
G
o
o
g
le+
:
S
o
c
ial
N
e
t
w
o
rk
in
g
"
,
I
n
ter
n
a
ti
o
n
a
l
J
o
u
r
n
a
l
o
f
S
o
c
i
a
l
Ne
two
rk
in
g
a
n
d
Vi
rtu
a
l
Co
mm
i
n
it
ies
,
v
o
l.
1
,
n
o
.
1
,
2
0
1
2
.
[2
]
H.
Ba
g
h
e
ri,
"
Big
Da
t
a
:
Ch
a
ll
e
n
g
e
s,
Op
p
o
rt
u
n
i
ti
e
s
a
n
d
Cl
o
u
d
Ba
se
d
S
o
l
u
ti
o
n
s
"
,
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
trica
l
a
n
d
C
o
m
p
u
ter
E
n
g
i
n
e
e
rin
g
,
v
o
l.
5
,
n
o
.
2
,
p
p
.
3
4
0
-
3
4
3
,
2
0
1
5
.
[3
]
L
.
A
.
A
iello
,
e
t
a
l.
"
S
e
n
sin
g
T
re
n
d
in
g
T
o
p
ic
in
T
w
it
ter
"
,
IEE
E
T
ra
n
sa
c
ti
o
n
o
n
M
u
lt
ime
d
i
a
,
v
o
l
.
15
,
n
o
.
6
,
p
p
.
1
2
6
8
-
1
2
8
0
,
2
0
1
3
.
[4
]
D.
Qu
e
rc
ia
,
"
T
a
l
k
o
f
th
e
Cit
y
:
Ou
r
Tw
e
e
ts,
Ou
r
Co
m
m
u
n
it
y
H
a
p
p
in
e
ss
"
,
Pro
c
e
e
d
in
g
s
o
f
th
e
S
i
x
th
In
ter
n
a
t
io
n
a
l
AA
AI
Co
n
fer
e
n
c
e
o
n
W
e
b
lo
g
s
a
n
d
S
o
c
ia
l
M
e
d
ia
,
p
p
.
9
6
5
-
9
6
8
,
2
0
1
2
[5
]
J.
A
ll
a
n
,
T
o
p
ic De
tec
ti
o
n
a
n
d
T
r
a
c
k
in
g
:
Eve
n
t
-
b
a
se
d
I
n
f
o
rm
a
ti
o
n
Or
g
a
n
iz
a
ti
o
n
.
N
o
rw
e
ll
,
USA
,
2
0
0
2
.
[6
]
D.D.
L
e
e
a
n
d
H.S
.
S
e
u
n
g
,
"
L
e
a
rn
in
g
th
e
p
a
rts
o
f
o
b
jec
t
b
y
n
o
n
n
e
g
a
ti
v
e
m
a
tri
x
f
a
c
to
riza
ti
o
n
"
,
Na
tu
r
e
,
v
o
l.
402
,
p
p
.
788
-
7
9
1
,
1
9
9
9
.
[7
]
J.H.
L
a
u
,
e
t
a
l,
"
M
a
c
h
in
e
Re
a
d
in
g
T
e
a
L
e
a
v
e
s:
A
u
to
m
a
ti
c
a
ll
y
Ev
a
lu
a
ti
n
g
T
o
p
ic
Co
h
e
re
n
c
e
a
n
d
T
o
p
ic
M
o
d
e
l
Qu
a
li
ty
"
,
1
4
th
Co
n
fer
e
n
c
e
s
o
f
th
e
Eu
ro
p
e
a
n
Ch
a
p
ter
o
f
th
e
Asso
c
i
a
ti
o
n
f
o
r
Co
mp
u
t
a
ti
o
n
a
l
L
i
n
g
u
isti
c
,
p
p
.
5
3
0
-
5
3
9
,
2
0
1
4
.
[8
]
R.
B.
Ya
tes
a
n
d
B.
R
.
Ne
to
,
"
M
o
d
e
rn
In
f
o
rm
a
ti
o
n
Re
tri
e
v
a
l
"
,
Ad
d
iso
n
W
e
sle
y
L
o
n
g
m
a
n
,
1
9
9
9
[9
]
M
.
W
.
Be
rr
y
,
e
t
a
l
.
,
"
A
l
g
o
rit
h
m
s
a
n
d
a
p
p
li
c
a
ti
o
n
s
f
o
r
a
p
p
ro
x
im
a
te
n
o
n
n
e
g
a
ti
v
e
m
a
tri
x
fa
c
to
riza
ti
o
n
"
,
Co
mp
u
t
a
ti
o
n
a
l
S
ta
t
isti
c
&
Da
ta
A
n
a
lys
is
.
v
o
l
.
52
,
p
p
.
1
5
5
-
1
7
3
,
2
0
0
7
.
[1
0
]
P.
P
a
a
tero
a
n
d
U.
T
a
p
p
e
r,
"
P
o
sit
iv
e
m
a
tri
x
fa
c
to
riza
ti
o
n
:
A
n
o
n
-
n
e
g
a
ti
v
e
f
a
c
to
r
m
o
d
e
l
w
it
h
o
p
ti
m
a
l
u
ti
li
z
a
ti
o
n
o
f
e
rro
r
e
stim
a
te
s o
f
d
a
ta v
a
lu
e
s
"
,
E
n
v
iro
n
me
trics
,
v
o
l.
5
,
p
p
.
1
1
1
–
126
,
1
9
9
4
.
[1
1
]
D.D.
L
e
e
a
n
d
H.S
.
S
e
u
n
g
.
A
lg
o
rit
h
m
s
f
o
r
n
o
n
-
n
e
g
a
ti
v
e
m
a
tri
x
f
a
c
to
riza
ti
o
n
.
In
Ad
v
a
n
c
e
s
in
Ne
u
r
a
l
In
fo
rm
a
t
io
n
Pro
c
e
ss
in
g
S
y
ste
m
.
2
0
0
1
;
5
5
6
–
562
.
[1
2
]
C.
J.
L
in
,
"
P
r
o
jec
ted
G
ra
d
ien
t
M
e
th
o
d
s
f
o
r
No
n
-
n
e
g
a
ti
v
e
M
a
tri
x
F
a
c
to
riza
ti
o
n
"
,
J
o
u
rn
a
l
Ne
u
r
a
l
C
o
mp
u
t
a
ti
o
n
,
v
o
l
.
19
,
n
o
.
10
,
p
p
.
2
7
5
6
-
2
7
7
9
,
2
0
0
7
.
[1
3
]
H.
Kim
a
n
d
H.
P
a
rk
,
"
No
n
n
e
g
a
ti
v
e
m
a
tri
x
f
a
c
to
riza
ti
o
n
b
a
se
d
o
n
a
lt
e
rn
a
ti
n
g
n
o
n
n
e
g
a
ti
v
it
y
c
o
n
stra
in
e
d
lea
st sq
u
a
re
s
a
n
d
a
c
ti
v
e
se
t
m
e
th
o
d
"
,
S
I
AM
J
o
u
rn
a
l
o
n
M
a
trix A
n
a
lys
is
a
n
d
Ap
p
l
ica
ti
o
n
s
,
v
o
l.
30
,
n
o
.
2
,
p
p
.
7
1
3
–
7
3
0
,
2
0
0
8
[1
4
]
J.
Kim
a
n
d
H.
P
a
rk
,
"
F
a
st
n
o
n
n
e
g
a
ti
v
e
m
a
tri
x
fa
c
to
riza
ti
o
n
:
a
n
a
c
ti
v
e
-
s
e
t
-
li
k
e
m
e
th
o
d
a
n
d
c
o
m
p
a
riso
n
s
"
,
S
IAM
J
o
u
rn
a
l
o
n
S
c
ien
ti
fi
c
C
o
mp
u
ti
n
g
,
v
o
l.
33
,
n
o
.
6
,
p
p
.
3
2
6
1
-
3
2
8
1
,
2
0
1
1
.
[1
5
]
C.
Bo
u
tsid
is
a
n
d
E.
G
a
ll
o
p
o
u
l
o
s
,
"
S
V
D
b
a
se
d
in
it
ializa
ti
o
n
:
a
h
e
a
d
sta
rt
f
o
r
n
o
n
n
e
g
a
ti
v
e
m
a
tri
x
fa
c
to
riza
ti
o
n
"
,
Pa
tt
e
rn
Rec
o
g
n
it
io
n
,
v
o
l
.
41
,
n
o
.
4
,
p
p
.
1
3
5
0
-
1
3
6
2
,
2
0
0
8
[1
6
]
P
e
d
re
g
o
sa
,
e
t
a
l
.,
"
S
c
ik
it
-
lea
rn
:
M
a
c
h
in
e
L
e
a
rn
in
g
in
P
y
th
o
n
"
,
J
o
u
rn
a
l
o
f
M
a
c
h
i
n
e
L
e
a
r
n
in
g
Res
e
a
rc
h
,
v
o
l
.
12
,
p
p
.
2
8
2
5
-
2
8
3
0
,
2
0
1
1
.
Evaluation Warning : The document was created with Spire.PDF for Python.