T
E
L
KO
M
N
I
KA
T
e
lec
om
m
u
n
icat
ion
,
Com
p
u
t
i
n
g,
E
lec
t
r
on
ics
an
d
Cont
r
ol
Vol.
18
,
No.
1
,
F
e
br
ua
r
y
202
0
,
pp.
106
~
113
I
S
S
N:
1693
-
6930,
a
c
c
r
e
dit
e
d
F
ir
s
t
G
r
a
de
by
Ke
me
nr
is
tekdikti
,
De
c
r
e
e
No:
21/E
/KP
T
/2018
DO
I
:
10.
12928/
T
E
L
KO
M
NI
KA
.
v18i1.
12982
106
Jou
r
n
al
h
omepage
:
ht
tp:
//
jour
nal.
uad
.
ac
.
id/
index
.
php/T
E
L
K
OM
N
I
K
A
G
raph
-
base
d
al
gori
t
hm
f
or
checki
ng
w
rong
i
ndi
rec
t
rel
at
i
on
s
hi
ps
i
n
non
-
f
ree
choi
ce
Agun
g
Wirat
m
o
,
Kel
ly
Ros
s
a
S
u
n
gk
on
o
,
Riya
n
ar
t
o
S
ar
n
o
D
ep
ar
t
men
t
o
f
In
f
o
rmat
i
cs
,
Facu
l
t
y
o
f
In
f
o
rmat
i
o
n
T
ec
h
n
o
l
o
g
y
an
d
Co
mm
u
n
i
cat
i
o
n
,
In
s
t
i
t
u
t
T
ek
n
o
l
o
g
i
Se
p
u
l
u
h
N
o
p
emb
er,
S
u
ra
b
ay
a,
In
d
o
n
e
s
i
a
Ar
t
icle
I
n
f
o
AB
S
T
RA
CT
A
r
ti
c
le
h
is
tor
y
:
R
e
c
e
ived
M
a
y
15
,
2019
R
e
vis
e
d
J
un
14
,
20
19
Ac
c
e
pted
Ju
l
1
,
20
19
In
t
h
i
s
co
n
t
e
x
t
,
t
h
i
s
p
a
p
er
p
ro
p
o
s
es
a
co
m
b
i
n
at
i
o
n
o
f
p
aramet
eri
s
ed
d
ec
i
s
i
o
n
mi
n
i
n
g
a
n
d
re
l
at
i
o
n
s
e
q
u
e
n
ces
t
o
d
e
t
ect
w
ro
n
g
i
n
d
i
rect
rel
a
t
i
o
n
s
h
i
p
i
n
t
h
e
n
o
n
-
free
c
h
o
i
ce.
T
h
e
ex
i
s
t
i
n
g
d
ec
i
s
i
o
n
mi
n
i
n
g
w
i
t
h
o
u
t
p
arame
t
er
can
o
n
l
y
d
et
ec
t
t
h
e
d
i
rect
i
o
n
,
b
u
t
n
o
t
t
h
e
co
rrec
t
n
e
s
s
.
T
h
i
s
p
a
p
er
ai
m
s
t
o
i
d
en
t
i
f
y
t
h
e
d
i
rec
t
i
o
n
a
n
d
co
rrec
t
n
e
s
s
w
i
t
h
d
ec
i
s
i
o
n
mi
n
i
n
g
w
i
t
h
p
aramet
er.
T
h
i
s
p
ap
er
d
i
s
co
v
ers
a
g
rap
h
p
ro
ce
s
s
mo
d
el
b
a
s
ed
o
n
t
h
e
ev
en
t
l
o
g
.
T
h
e
n
,
i
t
an
a
l
y
s
es
t
h
e
g
ra
p
h
p
ro
ces
s
mo
d
el
fo
r
o
b
t
a
i
n
i
n
g
d
eci
s
i
o
n
p
o
i
n
t
s
.
E
a
ch
d
ec
i
s
i
o
n
p
o
i
n
t
i
s
p
ro
ce
s
s
e
d
b
y
u
s
i
n
g
p
arame
t
eri
s
ed
d
ec
i
s
i
o
n
mi
n
i
n
g
,
s
o
t
h
at
d
eci
s
i
o
n
r
u
l
e
s
are
fo
rmed
.
T
h
e
d
er
i
v
e
d
d
eci
s
i
o
n
ru
l
e
s
are
u
s
ed
as
p
aramet
e
rs
o
f
ch
eck
i
n
g
w
r
o
n
g
i
n
d
i
rec
t
rel
at
i
o
n
s
h
i
p
i
n
t
h
e
n
o
n
-
free
ch
o
i
ce.
T
h
e
ev
al
u
a
t
i
o
n
s
h
o
w
s
t
h
a
t
t
h
e
ch
ec
k
i
n
g
w
ro
n
g
i
n
d
i
rect
re
l
at
i
o
n
s
h
i
p
s
i
n
n
o
n
-
free
ch
o
i
ce
w
i
t
h
p
aramet
er
i
s
e
d
d
eci
s
i
o
n
mi
n
i
n
g
h
a
v
e
1
0
0
%
accu
racy
,
w
h
erea
s
t
h
e
ex
i
s
t
i
n
g
d
eci
s
i
o
n
mi
n
i
n
g
h
as
9
0
.
7
%
accu
racy
.
K
e
y
w
o
r
d
s
:
De
c
is
ion
mi
ning
P
a
r
a
mete
r
ize
de
c
is
ion
mi
ning
P
r
oc
e
s
s
model
W
r
ong
indi
r
e
c
t
r
e
lati
ons
hips
Th
i
s
i
s
a
n
o
p
en
a
c
ces
s
a
r
t
i
c
l
e
u
n
d
e
r
t
h
e
CC
B
Y
-
SA
l
i
ce
n
s
e
.
C
or
r
e
s
pon
din
g
A
u
th
or
:
R
iyana
r
to
S
a
r
no,
De
pa
r
tm
e
nt
of
I
nf
or
mat
ics
,
F
a
c
ult
y
o
f
I
nf
or
mation
T
e
c
hnology
a
nd
C
omm
unica
ti
on
,
I
ns
ti
tut
T
e
knologi
S
e
puluh
Nope
mber
,
S
ur
a
ba
ya
,
I
ndone
s
ia.
E
mail:
r
iyana
r
to@i
f
.
it
s
.
a
c
.
id
1.
I
NT
RODU
C
T
I
ON
E
a
c
h
c
ompany
r
e
c
or
ds
e
ve
nts
c
a
r
r
ied
out
in
the
e
ve
nt
log.
T
he
a
na
lys
a
ti
on
of
inf
or
mation
f
r
om
the
e
ve
nt
log
obtains
the
obtain
knowle
dge
[
1]
.
T
h
e
pr
oc
e
s
s
of
ga
ini
ng
knowle
dge
f
r
om
e
ve
nt
log
e
xtr
a
c
ti
on
is
c
a
ll
e
d
the
pr
oc
e
s
s
mi
ning,
whic
h
a
im
s
to
f
ind
out,
moni
tor
ing
a
nd
im
p
r
oving
the
pr
oc
e
s
s
e
s
that
oc
c
ur
[
2
]
.
I
n
the
pr
oc
e
s
s
mi
ning,
ther
e
a
r
e
two
mos
t
p
r
omi
ne
nt
pr
oc
e
s
s
e
s
,
na
mely:
1)
c
on
f
or
manc
e
c
he
c
king
[
3]
a
nd
2)
p
r
oc
e
s
s
dis
c
ove
r
y
[
4]
.
T
he
pr
oc
e
s
s
mi
ning
f
oun
ds
the
w
r
ong
p
r
oc
e
s
s
in
the
e
ve
nt
log.
T
his
pa
pe
r
dis
c
ove
r
s
a
gr
a
ph
pr
oc
e
s
s
model
ba
s
e
d
on
the
e
ve
nt
log.
T
h
e
a
na
lys
a
ti
on
of
the
gr
a
ph
pr
oc
e
s
s
model
obtains
de
c
is
ion
point
s
.
E
a
c
h
de
c
is
ion
point
is
p
r
oc
e
s
s
e
d
by
us
ing
pa
r
a
mete
r
is
e
d
de
c
is
ion
mi
n
ing,
s
o
that
de
c
is
ion
r
ules
a
r
e
f
or
med.
T
he
de
r
ived
de
c
is
ion
r
ules
a
r
e
us
e
d
a
s
pa
r
a
mete
r
s
of
c
he
c
king
w
r
ong
indi
r
e
c
t
r
e
latio
ns
hip
in
the
non
-
f
r
e
e
c
hoice
.
S
e
ve
r
a
l
pr
e
vious
s
tudi
e
s
dis
c
us
s
de
c
is
ion
mi
ning
i
n
r
e
c
e
nt
ye
a
r
s
.
A
s
tudy
c
onduc
ted
by
R
oz
inat
[
5]
e
xplains
de
c
is
ion
mi
ning
on
bus
ines
s
pr
oc
e
s
s
e
s
.
H
owe
ve
r
,
de
c
is
io
n
mi
ning
is
not
i
mpl
e
mente
d
a
s
a
de
c
is
ion
r
ule
f
or
c
he
c
king
e
r
r
or
s
in
e
ve
nt
logs
.
Hor
i
ta
[
6]
made
de
c
is
ions
on
e
ve
nt
logs
that
r
e
s
ult
in
li
ne
a
r
tempor
a
l
logi
c
,
but
the
tempor
a
l
logi
c
is
not
a
ppli
e
d
f
or
e
r
r
o
r
s
e
a
r
c
he
s
in
e
ve
nt
logs
.
T
he
e
xis
ti
ng
methods
in
c
he
c
king
indi
r
e
c
t
r
e
lations
hips
[
7,
8
]
in
non
-
f
r
e
e
c
hoice
on
ly
us
ing
dir
e
c
ti
on
s
o
the
e
r
r
or
c
a
n
be
de
tec
ted
o
nly
f
r
om
dir
e
c
ti
ona
l
e
r
r
or
,
but
c
or
r
e
c
tnes
s
f
r
om
c
hoos
ing
di
r
e
c
ti
ona
l
c
a
nnot
be
obtaine
d.
T
he
p
r
opos
e
d
method,
na
mely
pa
r
a
mete
r
ize
d
de
c
is
ion
mi
n
ing
is
to
us
e
the
de
c
is
ion
r
u
le
in
c
he
c
king
e
ve
nt
logs
with
no
ti
c
e
not
o
nly
f
r
om
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
Gr
aph
-
bas
e
d
algor
it
hm
for
c
he
c
k
ing
w
r
ong
indi
r
e
c
t
r
e
lat
ions
hips
in
non
-
fr
e
e
c
hoice
(
A
gung
W
ir
atm
o
)
107
the
dir
e
c
ti
on
but
a
ls
o
the
pa
r
a
mete
r
s
in
the
e
ve
nt
log.
I
n
thi
s
r
e
s
e
a
r
c
h,
the
de
c
is
ion
r
ule
is
us
e
d
to
f
ind
e
r
r
or
s
in
the
e
ve
nt
log
,
s
o
the
f
a
il
ur
e
of
e
ve
nt
logs
c
a
n
be
f
in
ding
mor
e
a
c
c
ur
a
te
in
ter
ms
of
the
dir
e
c
ti
on
a
nd
c
o
r
r
e
c
tnes
s
of
c
hoice
.
2.
RE
S
E
AR
CH
M
E
T
HO
D
I
n
th
is
s
e
c
ti
on,
pa
r
a
mete
r
ize
d
de
c
is
ion
mi
ning
will
be
pr
e
s
e
nted
to
f
ind
a
wr
ong
indi
r
e
c
t
r
e
lations
hip
in
non
-
f
r
e
e
c
hoice
us
ing
the
gr
a
ph
da
taba
s
e
.
T
he
non
-
f
r
e
e
c
hoice
is
a
c
ondit
ion
that
is
not
f
r
e
e
to
make
c
hoice
s
,
but
c
hoice
s
de
pe
nd
on
the
r
e
s
ult
s
of
the
pr
e
vious
e
l
e
c
ti
on
[
9]
.
C
he
c
king
pr
oc
e
s
s
models
on
the
no
n
-
f
r
e
e
c
hoice
pa
r
t
of
the
e
ve
nt
log
c
a
n
be
done
c
or
r
e
c
tl
y
mus
t
c
ons
ider
a
ll
de
pe
nde
nc
ies
[
10]
.
T
he
r
e
a
r
e
two
types
of
de
pe
nde
nc
e
on
the
pr
oc
e
s
s
of
the
model,
dir
e
c
t
d
e
pe
nde
nc
e
or
r
e
f
e
r
r
e
d
to
a
s
dir
e
c
t
r
e
lations
hip
a
nd
indi
r
e
c
t
de
pe
nde
n
c
e
or
r
e
f
e
r
r
e
d
to
a
n
indi
r
e
c
t
r
e
lations
hip
[
7,
11
,
12]
.
A
dir
e
c
t
r
e
lations
hip
is
a
r
e
latio
ns
hip
or
de
pe
nde
nc
y
that
is
dir
e
c
tl
y
be
twe
e
n
tas
ks
.
C
onve
r
s
e
ly,
a
n
indi
r
e
c
t
r
e
lations
hip
is
a
r
e
lations
hip
or
a
de
pe
nde
nc
e
that
is
indi
r
e
c
tl
y
be
twe
e
n
tas
ks
[
13]
.
A
gr
a
ph
da
taba
s
e
is
a
NoSQ
L
da
taba
s
e
whe
r
e
is
d
e
picte
d
in
the
f
or
m
of
gr
a
ph
[
14
-
16]
.
G
r
a
ph
da
taba
s
e
s
will
f
or
m
da
ta
a
s
node
s
a
nd
r
e
lations
be
twe
e
n
node
s
[
15
,
17]
.
T
he
pr
oc
e
s
s
mi
ning
is
us
e
d
to
e
xtr
a
c
t
inf
or
mat
ion
f
r
om
the
e
v
e
nt
log
to
s
e
e
bus
ines
s
pr
oc
e
s
s
e
s
[
10,
18
,
19]
.
T
h
e
pr
oc
e
s
s
mi
ning
c
a
n
be
us
e
d
to
bu
il
d
a
p
r
oc
e
s
s
model
[
16,
20
–
23]
.
De
c
is
ion
mi
ning
is
us
e
d
to
s
tudy
pa
r
a
mete
r
s
that
c
a
n
inf
luenc
e
the
s
e
lec
ti
on
of
gr
oove
s
[
24
,
25]
.
De
c
is
ion
mi
ning
is
us
e
d
to
f
ind
r
ules
f
o
r
br
a
nc
hing
f
r
om
e
a
c
h
de
c
is
ion
point
.
B
y
u
s
ing
a
g
r
a
ph
da
taba
s
e
,
a
de
c
is
ion
point
c
a
n
be
known
f
r
om
a
nod
e
that
ha
s
a
xor
s
pli
t
r
e
lation.
T
he
a
lgo
r
it
hm
us
e
d
in
de
c
is
ion
mi
ning
r
e
s
e
a
r
c
h
is
us
ing
the
C
4.
5
de
c
is
ion
tr
e
e
a
lgor
it
hm
[
5
,
26]
.
A
de
c
is
ion
tr
e
e
is
us
e
d
to
pr
e
dict
a
n
a
c
ti
vit
y
s
e
e
n
f
r
om
the
pa
r
a
mete
r
s
of
a
da
ta.
T
he
de
c
is
ion
t
r
e
e
ha
s
s
e
ve
r
a
l
ter
ms
.
T
hos
e
ter
ms
a
r
e
a
r
oot
a
s
th
e
ini
t
ial
node
,
a
lea
f
node
a
s
the
c
hil
d
of
a
node
,
a
nd
the
de
pth
of
a
node
a
s
the
length
of
the
pa
th
be
twe
e
n
the
node
s
to
the
lea
f
node
[
27
,
28
]
.
T
he
f
ir
s
t
s
tep
is
to
dis
c
ove
r
gr
a
ph
pr
oc
e
s
s
model
of
the
e
ve
nt
log
ba
s
e
d
on
t
he
gr
a
ph
da
taba
s
e
.
T
he
n,
gr
a
ph
pr
oc
e
s
s
model
be
a
na
lyze
d
to
f
ind
the
de
c
is
ion
po
int
.
T
he
s
e
c
ond
s
tep
is
dis
c
ove
r
ing
de
c
is
ion
r
ule
us
ing
de
c
is
ion
mi
ning
f
r
om
e
a
c
h
de
c
i
s
ion
point
with
noti
c
e
pa
r
a
mete
r
in
e
ve
nt
log.
T
his
de
c
is
ion
r
ule
will
be
us
e
d
a
s
a
pa
r
a
mete
r
in
de
ter
mi
ning
th
e
wr
ong
de
c
is
ion
in
non
-
f
r
e
e
c
hoice
.
T
he
las
t
s
tep
s
e
a
r
c
he
s
e
a
c
h
c
a
s
e
in
the
e
ve
nt
log
with
the
pa
r
a
mete
r
s
s
tate
d
e
a
r
li
e
r
.
2
.
1
.
Dis
c
ove
r
y
p
r
oc
e
s
s
m
od
e
l
b
as
e
d
o
n
gr
ap
h
d
a
t
ab
as
e
T
he
f
ir
s
t
s
tep
f
or
the
dis
c
ove
r
y
of
the
g
r
a
ph
model
pr
oc
e
s
s
is
to
e
nter
e
ve
nt
logs
li
ke
in
T
a
ble
1
int
o
the
gr
a
ph
da
taba
s
e
us
ing
a
que
r
y
li
ke
in
T
a
ble
2.
T
he
pa
r
a
mete
r
s
in
the
e
ve
nt
log
us
e
d
in
the
gr
a
ph
da
taba
s
e
a
r
e
C
a
s
e
_I
D,
Ac
ti
vi
ty,
a
nd
T
im
e
.
I
n
the
T
a
ble
2
s
h
own
que
r
ies
in
t
he
g
r
a
ph
da
taba
s
e
f
or
:
(
1
)
I
mpo
r
t
a
ll
da
ta
in
e
ve
nt
log,
(
2
)
I
mpo
r
t
only
unique
a
c
ti
vit
y,
(
3
)
c
r
e
a
te
r
e
lation:
s
e
que
nc
e
,
xor
s
pli
t,
xor
joi
n
,
a
nds
pli
t,
a
ndjoi
n,
non
-
f
r
e
e
c
hoice
,
a
nd
(
4
)
ge
t
the
de
c
is
ion
point
.
I
n
the
model
p
r
oc
e
s
s
,
ther
e
will
be
s
e
ve
r
a
l
r
e
l
a
ti
ons
hips
s
uc
h
a
s
xor
s
pli
t
[
29
]
,
xor
joi
n
[
29
]
,
a
nds
pli
t
[
30]
,
a
ndjoi
n
[
30]
a
nd
non
-
f
r
e
e
c
hoice
s
h
own
in
F
igu
r
e
1
.
J
oint
r
e
lat
ions
a
r
e
a
r
e
lation
of
t
he
union
f
r
om
b
r
a
nc
hing
to
s
pli
t
r
e
lations
a
t
the
ba
s
e
of
b
r
a
nc
hing.
T
he
xor
r
e
lation
is
a
br
a
nc
hing
r
e
lation
whic
h
mea
ns
that
the
f
low
of
the
e
ve
nt
log
c
a
n
only
c
hoos
e
one
of
the
e
nti
r
e
br
a
nc
he
s
of
the
e
ve
nt
log.
T
he
a
nd
r
e
lation
is
a
f
low
r
e
lations
hip
that
will
do
a
ll
e
ve
nts
e
ve
n
tho
ugh
the
d
if
f
e
r
e
nt
or
de
r
.
T
he
r
e
s
ult
s
o
f
wo
r
king
on
que
r
ies
in
T
a
ble
2
p
r
oduc
e
the
model
pr
oc
e
s
s
s
hown
in
F
igu
r
e
1.
A
f
ter
e
a
c
h
a
c
ti
vit
y
is
r
e
p
r
e
s
e
nted
in
the
node
,
the
ne
xt
s
tep
is
to
make
r
e
lations
b
e
twe
e
n
node
s
s
u
c
h
a
s
s
e
que
nc
e
,
xor
s
pli
t,
xor
joi
n
,
a
ndjoi
n
,
a
nds
pli
t
r
e
lati
ons
,
a
nd
non
-
f
r
e
e
c
hoice
.
F
igur
e
1.
E
xa
mpl
e
s
o
f
the
pr
oc
e
s
s
model
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
1693
-
6930
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
,
Vol.
18
,
No
.
1
,
F
e
br
ua
r
y
2020
:
106
-
113
108
Af
ter
the
gr
a
ph
pr
oc
e
s
s
model
is
f
or
med
.
T
he
ne
x
t
s
tep
is
to
de
ter
mi
ne
the
de
c
is
ion
point
by
us
ing
the
que
r
y
i
n
T
a
ble
2
.
De
c
is
ion
point
is
the
node
whe
r
e
br
a
nc
hing
of
the
pr
oc
e
s
s
be
gins
.
I
n
F
igur
e
1
s
hows
gr
a
ph
pr
oc
e
s
s
model
whe
r
e
c
ontaining
the
de
c
is
ion
point
s
in
node
A
a
nd
node
E
.
Node
A
is
the
ba
s
e
of
membe
r
node
B
or
node
C
.
Node
E
is
the
ba
s
e
of
br
a
nc
hin
g
nod
e
G
a
nd
node
F
.
f
ur
ther
mo
r
e
,
de
c
is
ion
r
u
le
would
be
dis
c
ove
r
e
d
by
de
c
is
ion
mi
ning
to
f
ind
pa
r
a
mete
r
s
e
a
c
h
br
a
nc
hing
of
e
a
c
h
de
c
is
ion
point
.
2
.
2.
E
xt
r
ac
t
in
g
a
p
ar
am
e
t
e
r
i
n
d
e
c
is
ion
p
oin
t
s
T
he
de
c
is
ion
point
s
hown
in
F
igu
r
e
1,
ther
e
a
r
e
tw
o
de
c
is
ion
point
p
oint
s
.
E
a
c
h
de
c
is
ion
point
will
be
a
na
lyze
d
by
c
ons
ider
ing
the
pa
r
a
mete
r
s
in
the
e
ve
nt
log
to
ge
t
the
de
c
is
ion
r
ule.
T
he
pr
oc
e
s
s
f
or
e
xtr
a
c
ti
on
de
c
is
ion
r
ule
is
c
a
ll
e
d
de
c
is
ion
mi
ning.
T
he
a
lgo
r
it
hm
of
de
c
is
ion
mi
ning
is
us
e
d
C
4.
5
de
c
is
ion
tr
e
e
a
l
gor
it
hm.
T
he
f
ir
s
t
s
tep
is
ge
tt
ing
lea
f
node
s
f
r
om
e
a
c
h
de
c
is
ion
point
.
T
he
n,
the
ne
xt
s
tep
is
ge
tt
ing
a
n
e
ve
nt
log
that
ha
s
a
c
ti
vit
ies
s
uc
h
a
s
lea
f
node
s
with
a
lgor
it
hm
in
T
a
bl
e
3.
T
he
da
ta
ne
e
de
d
in
a
lgor
it
h
m
in
T
a
ble
3
is
,
,
d
a
n
.
E
a
c
h
de
c
is
ion
point
of
lea
f
node
s
obtaine
d
f
r
om
a
lgo
r
it
hm
in
T
a
ble
3
is
us
e
d
in
the
de
c
is
ion
mi
ning
p
r
oc
e
s
s
.
T
he
de
c
is
ion
mi
ning
a
lgor
it
hm
is
s
e
e
n
in
T
a
ble
4.
Algor
it
hm
in
T
a
ble
4
ha
s
f
ive
da
ta
va
r
iable
s
:
X
is
the
e
ve
nt
log
o
n
node
lea
f
.
Y
is
a
n
a
tt
r
ibu
te
owne
d
by
X
.
T
he
is
a
n
a
tt
r
ibut
e
us
e
d
a
s
a
s
olvi
ng
pa
r
a
mete
r
.
T
he
ℎ
is
the
method
us
e
d
to
f
ind
the
be
s
t
f
r
a
c
ti
on
v
a
lue.
T
he
method
us
e
d
to
f
ind
the
be
s
t
s
pli
tt
ing
c
r
it
e
r
ion
is
C
4.
5
with
the
gini
index
pa
r
a
mete
r
in
(
2)
a
nd
(
3)
.
T
he
f
unc
ti
on
of
(
1
)
is
to
e
va
luate
the
s
e
pa
r
a
ti
on
in
e
ve
nt
log
e
a
c
h
a
tt
r
ibut
e
.
T
he
f
unc
ti
on
of
(
2)
is
to
e
va
luate
the
s
e
pa
r
a
ti
on
in
e
ve
nt
log
e
a
c
h
pa
r
a
mete
r
.
(
)
=
1
−
∑
[
(
)
]
2
(
2)
whe
r
e
(
)
r
e
pr
e
s
e
nts
the
f
r
e
que
nc
y
of
the
j
a
tt
r
ibut
e
in
a
c
ti
vit
y
t
.
=
∑
(
)
=
1
(
3)
whe
r
e
is
the
number
of
pa
r
ti
ti
ons
,
is
the
a
mount
of
da
ta
in
pa
r
ti
ti
on,
is
the
a
mount
of
da
ta
in
the
node
.
T
he
be
s
t
s
pli
t
va
lue
is
indi
c
a
ted
by
the
s
malles
t
.
T
he
is
the
va
lue
o
f
the
pa
r
a
mete
r
that
is
us
e
d
a
s
a
s
olver
.
is
a
nod
e
.
T
he
a
lgor
it
hm
i
n
T
a
ble
4
will
c
onti
nue
to
be
r
e
pe
a
ted
unt
il
the
da
ta
is
e
mpt
y.
T
he
r
e
s
ult
s
of
a
lgo
r
it
hm
in
T
a
ble
4
a
r
e
a
d
e
c
is
ion
tr
e
e
by
s
howing
the
pa
r
a
mete
r
s
a
nd
the
le
a
f
va
lues
c
a
n
be
s
e
e
n
in
F
igur
e
2.
T
a
ble
1.
P
r
oc
e
s
s
mi
ning
will
be
c
a
r
r
ied
ou
t
by
the
e
ve
nt
log
C
a
s
e
_I
D
a
mount
s
ta
c
kT
yp
e
S
ta
tu
s
T
im
e
A
c
ti
vi
ty
P
P
10
3
nonr
e
e
f
e
r
c
ompl
e
te
3/
11/
2016 0:52
A
P
P
10
3
nonr
e
e
f
e
r
c
ompl
e
te
3/
11/
2016 2:00
C
P
P
10
3
nonr
e
e
f
e
r
c
ompl
e
te
3/
11/
2016 3:08
D
P
P
10
3
nonr
e
e
f
e
r
c
ompl
e
te
3/
11/
2016 4:16
E
P
P
10
3
nonr
e
e
f
e
r
c
ompl
e
te
3/
11/
2016 5:24
F
P
P
10
3
nonr
e
e
f
e
r
c
ompl
e
te
3/
11/
2016 6:32
H
P
P
412
17
r
e
e
f
e
r
in
c
ompl
e
te
7/
2/
2016 22:28
A
P
P
412
17
r
e
e
f
e
r
in
c
ompl
e
te
7/
2/
2016 23:36
C
P
P
412
17
r
e
e
f
e
r
in
c
ompl
e
te
7/
3/
2016 0:44
D
P
P
412
17
r
e
e
f
e
r
in
c
ompl
e
te
7/
3/
2016 1:52
E
P
P
412
17
r
e
e
f
e
r
in
c
ompl
e
te
7/
3/
2016 3:00
G
P
P
412
17
r
e
e
f
e
r
in
c
ompl
e
te
7/
3/
2016 4:08
H
P
P
735
13
nonr
e
e
f
e
r
in
c
ompl
e
te
10/
2/
2016 10:52
A
P
P
735
13
nonr
e
e
f
e
r
in
c
ompl
e
te
10/
2/
2016 12:00
B
P
P
735
13
nonr
e
e
f
e
r
in
c
ompl
e
te
10/
2/
2016 13:08
D
P
P
735
13
nonr
e
e
f
e
r
in
c
ompl
e
te
10/
2/
2016 14:16
E
P
P
735
13
nonr
e
e
f
e
r
in
c
ompl
e
te
10/
2/
2016 15:24
G
P
P
735
13
nonr
e
e
f
e
r
in
c
ompl
e
te
10/
2/
2016 16:32
H
P
P
1050
10
nonr
e
e
f
e
r
c
ompl
e
te
12/
30/
2016 16:52
A
P
P
1050
10
nonr
e
e
f
e
r
c
ompl
e
te
12/
30/
2016 18:00
B
P
P
1050
10
nonr
e
e
f
e
r
c
ompl
e
te
12/
30/
2016 19:08
D
P
P
1050
10
nonr
e
e
f
e
r
c
ompl
e
te
12/
30/
2016 20:16
E
P
P
1050
10
nonr
e
e
f
e
r
c
ompl
e
te
12/
30/
2016 21:24
F
P
P
1050
10
nonr
e
e
f
e
r
c
ompl
e
te
12/
30/
2016 22:32
H
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
Gr
aph
-
bas
e
d
algor
it
hm
for
c
he
c
k
ing
w
r
ong
indi
r
e
c
t
r
e
lat
ions
hips
in
non
-
fr
e
e
c
hoice
(
A
gung
W
ir
atm
o
)
109
T
a
ble
2.
Que
r
ies
in
the
gr
a
ph
da
taba
s
e
No
Q
ue
r
ie
s
1
de
f
i
mpor
tAc
ti
vi
ty
T
(
tx
, f
il
e
N
a
me
)
:
tx
.r
un (
"
L
O
A
D
C
S
V
w
it
h he
a
de
r
s
F
R
O
M
'
f
il
e
:/
//
"
+
f
il
e
N
a
me
+
"
'
A
S
l
in
e
"
"
M
e
r
ge
(
:
A
c
ti
vi
ty
{
C
a
s
e
I
d:
l
in
e
. C
a
s
e
_I
D
, N
a
me
:
li
ne
. A
c
ti
vi
t
y, A
mount
:
to
I
nt
(
li
ne
.
a
mount
)
,
S
ta
c
kT
ype
:
li
ne
. s
ta
c
kT
yp
e
, S
ta
tu
s
:
li
ne
. s
ta
tu
s
, T
im
e
:
li
ne
.
T
im
e
}
)
"
)
2
de
f
i
mpor
tC
a
s
e
A
c
ti
vi
ty
(
tx
, f
il
e
N
a
me
)
:
tx
.r
un (
"
L
O
A
D
C
S
V
w
it
h he
a
de
r
s
F
R
O
M
'
f
il
e
:/
//
"
+
f
il
e
N
a
me
+
"
'
A
S
l
in
e
M
e
r
ge
(
:
C
a
s
e
A
c
ti
vi
ty
{
N
a
me
:
li
ne
.
A
c
ti
vi
ty
}
)
“
)
3
de
f
c
r
e
a
te
R
e
la
ti
ons
hi
p(
tx
)
:
# c
r
e
a
te
s
e
que
n
c
e
r
e
la
ti
on
tx
.r
un (
"
M
A
T
C
H
(
c
:
A
c
ti
vi
ty
)
"
"
W
I
T
H
C
O
L
L
E
C
T
(
c
)
A
S
C
a
s
e
li
s
t
"
"
U
N
W
I
N
D
R
A
N
G
E
(
0, S
iz
e
(
C
a
s
e
li
s
t)
-
2)
a
s
i
dx "
"
W
I
T
H
C
a
s
e
li
s
t[
id
x]
A
S
s
1, C
a
s
e
li
s
t[
id
x+
1]
A
S
s
2 "
"
M
A
T
C
H
(
b:
C
a
s
e
A
c
ti
vi
ty
)
, (
a
:
C
a
s
e
A
c
ti
vi
ty
)
"
"
W
H
E
R
E
s
1. C
a
s
e
I
d =
s
2. C
a
s
e
I
d A
N
D
s
1.
N
a
me
=
a
. N
a
me
A
N
D
s
2. N
a
me
=
b. N
a
me
"
"
M
E
R
G
E
(
a
)
-
[
r
:
S
E
Q
U
E
N
C
E
]
-
>
(
b
)
"
)
# c
r
e
a
te
xor
s
pl
it
r
e
la
ti
on
tx
.r
un (
"
M
A
T
C
H
(
be
f
)
-
[
r
]
-
>
(
a
f
t)
"
"
W
H
E
R
E
s
iz
e
(
(
be
f
)
--
>
(
)
)
>
1 A
N
D
s
iz
e
(
(
a
f
t)
<
--
(
)
)
=
1 A
N
D
(
s
iz
e
(
(
a
f
t)
--
>
(
)
)
=
1 O
R
s
iz
e
(
(
a
f
t)
--
>
(
)
)
>
1)
"
"
C
R
E
A
T
E
(
be
f
)
-
[
:
X
O
R
S
P
L
I
T
]
-
>
(
a
f
t
)
"
"
D
E
L
E
T
E
r
"
)
# c
r
e
a
te
xor
jo
in
r
e
la
ti
on
tx
.r
un (
"
M
A
T
C
H
(
be
f
)
-
[
r
]
-
>
(
a
f
t)
"
"
W
H
E
R
E
(
s
iz
e
(
(
be
f
)
--
>
(
)
)
=
1 O
R
s
iz
e
(
(
be
f
)
--
>
(
)
)
>
1)
A
N
D
s
iz
e
(
(
a
f
t)
<
--
(
)
)
>
1 "
"
C
R
E
A
T
E
(
be
f
)
-
[
:
X
O
R
J
O
I
N
]
-
>
(
a
f
t)
"
"
D
E
L
E
T
E
r
"
)
# c
r
e
a
te
a
nds
pl
it
r
e
la
ti
on
tx
.r
un (
"
M
A
T
C
H
(
a
f
t1
)
<
-
[r]
-
(
be
f
)
-
[
s
]
-
>
(
a
f
t2
)
"
"
W
H
E
R
E
s
iz
e
(
(
be
f
)
--
>
(
)
)
>
1 "
"
A
N
D
s
iz
e
(
(
a
f
t2
)
--
>
(
)
)
=
s
iz
e
(
(
be
f
)
--
>
(
)
)
A
N
D
s
iz
e
(
(
a
f
t1
)
--
>
(
)
)
=
s
iz
e
(
(
be
f
)
--
>
(
)
)
"
"
A
N
D
not
(
a
f
t1
)
-
[
:
S
E
Q
U
E
N
C
E
]
-
>
(
be
f
)
A
N
D
not
(
a
f
t2
)
-
[
:
S
E
Q
U
E
N
C
E
]
-
>
(
be
f
)
"
"
M
E
R
G
E
(
a
f
t1
)
<
-
[
:
A
N
D
S
P
L
I
T
]
-
(
be
f
)
-
[
:
A
N
D
S
P
L
I
T
]
-
>
(
a
f
t
2)
"
"
D
E
L
E
T
E
r
, s
"
)
# c
r
e
a
te
a
ndj
oi
n r
e
la
ti
on
tx
.r
un (
"
M
A
T
C
H
(
a
f
t1
)
-
[
r
]
-
>
(
be
f
)
<
-
[
s
]
-
(
a
f
t2
)
"
"
W
H
E
R
E
s
iz
e
(
(
be
f
)
<
--
(
)
)
>
1 "
"
A
N
D
s
iz
e
(
(
a
f
t2
)
--
>
(
)
)
=
s
iz
e
(
(
be
f
)
<
--
(
)
)
A
N
D
s
iz
e
(
(
a
f
t1
)
--
>
(
)
)
=
s
iz
e
(
(
be
f
)
<
--
(
)
)
"
"
A
N
D
not
(
)
-
[
:
A
N
D
S
P
L
I
T
]
-
>
(
be
f
)
"
"
M
E
R
G
E
(
a
f
t1
)
-
[
:
A
N
D
J
O
I
N
]
-
>
(
be
f
)
<
-
[
:
A
N
D
J
O
I
N
]
-
(
a
f
t2
)
"
"
D
E
L
E
T
E
r
, s
"
)
# c
r
e
a
te
N
on
-
F
r
e
e
C
hoi
c
e
tx
.r
un (
"
ma
tc
h (
)
-
[
c
:
X
O
R
S
P
L
I
T
]
-
>
(
n)
"
"
ma
tc
h (
a
)
-
[
b:
X
O
R
J
O
I
N
]
-
>
(
)
"
"
ma
tc
h (
k:
A
c
ti
vi
ty
)
, (
l:
A
c
ti
vi
ty
)
"
"
w
he
r
e
a
. N
a
me
<
>
n.N
a
me
a
nd k. N
a
me
=
a
.
N
a
me
a
nd l
. N
a
m
e
=
n.N
a
me
a
nd k. C
a
s
e
I
d=
l.
C
a
s
e
I
d a
nd k.
T
im
e
<
l.
T
im
e
"
"
me
r
ge
(
a
)
-
[
:
N
O
N
F
R
E
E
C
H
O
I
C
E
]
-
>
(
n
)
"
)
4
de
f
pr
in
tS
ta
r
ti
ngN
ode
N
onF
r
e
e
C
hoi
c
e
(
tx
)
:
node
s
=
[
]
gl
oba
l
node
S
ta
r
te
dN
onF
r
e
e
C
hoi
c
e
f
or
r
e
c
or
d i
n t
x.r
un
(
"
M
A
T
C
H
(
p)
-
[
r
:
X
O
R
S
P
L
I
T
]
-
>
(
)
R
E
T
U
R
N
p. N
a
me
O
R
D
E
R
B
Y
p. N
a
me
"
)
:
node
s
.
a
ppe
nd (
r
e
c
or
d [
"
p. N
a
me
"
]
)
node
S
ta
r
te
dN
onF
r
e
e
C
hoi
c
e
=
np. unique (
np. a
r
r
a
y(
node
s
)
)
r
e
tu
r
n node
S
ta
r
te
dN
onF
r
e
e
C
hoi
c
e
T
a
ble
3.
Algor
it
hm
f
o
r
ge
tt
ing
e
ve
nt
log
of
lea
f
no
de
s
e
a
c
h
de
c
is
ion
point
D
a
ta
:
,
,
R
e
s
ul
t:
No
P
s
e
udoc
ode
1
2
ℎ
3
ℎ
ℎ
4
5
ℎ
ℎ
ℎ
6
ℎ
7
8
9
10
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
1693
-
6930
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
,
Vol.
18
,
No
.
1
,
F
e
br
ua
r
y
2020
:
106
-
113
110
T
a
ble
4.
Algor
it
hm
f
o
r
e
xtr
a
c
ti
ng
a
pa
r
a
mete
r
e
a
c
h
de
c
is
ion
point
us
ing
de
c
is
ion
mi
ning
D
a
ta
:
X
,
Y
,
s
p
l
it
in
g
At
r
ibut
e
,
a
t
r
ibut
e
S
e
l
e
c
t
io
n
M
e
t
ho
d
,
s
p
l
it
in
g
C
r
it
e
r
ia
,
N
R
e
s
ul
t:
P
a
r
a
m
e
t
e
r
d
e
c
is
io
n
in
d
e
c
is
io
n
p
o
in
t
s
ho
we
d
by
d
e
c
is
io
n
t
r
e
e
No
P
s
e
udoc
ode
1
2
ℎ
ℎ
3
ℎ
4
5
ℎ
6
ℎ
7
8
ℎ
(
,
)
9
(
−
)
ℎ
10
←
−
11
12
ℎ
13
ℎ
ℎ
14
ℎ
15
ℎ
ℎ
16
17
ℎ
(
,
)
18
19
20
F
igur
e
2.
De
c
is
ion
tr
e
e
in
:
(
a
)
node
A
,
a
nd
(
b)
nod
e
E
F
r
om
F
ig
ur
e
2
(
a
)
it
c
a
n
be
c
onc
luded
that
the
c
o
ndit
ion
f
o
r
do
ing
a
c
ti
vit
y
B
is
(
(
s
tatus
=
c
ompl
e
te
AN
D
a
mount
>
8.
5)
OR
(
s
tatus
=
incomplete
A
ND
a
mount
≤
9
)
)
.
W
he
r
e
a
s
to
do
a
c
ti
vit
y
C
it
mus
t
be
c
ondit
ioned
(
(
s
tatus
=
c
ompl
e
te
AN
D
a
mount
≤
8.
5)
OR
(
s
tatus
=
incompl
e
te
AN
D
a
mount
>
9
)
)
.
F
r
om
F
ig
ur
e
2
(
b)
it
c
a
n
be
c
onc
luded
that
the
c
on
dit
ion
f
o
r
doing
a
c
ti
vit
y
F
is
(
(
s
tatus
=
c
ompl
e
te
AN
D
a
mount
>
8.
5)
OR
(
s
tatus
=
incomplete
AN
D
a
mount
≤
9)
)
.
W
he
r
e
a
s
to
do
a
c
ti
vit
y
G
mus
t
be
c
ondit
ioned
(
(
s
tatus
=
c
ompl
e
te
AN
D
a
mount
≤
8.
5)
OR
(
s
tatus
=
incomplete
AN
D
a
mount
t>
9)
)
.
T
he
ne
xt
s
tep
is
to
c
he
c
k
the
e
ve
nt
log
with
pa
r
a
mete
r
s
that
ha
ve
be
e
n
obtai
ne
d
pr
e
vious
ly.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
Gr
aph
-
bas
e
d
algor
it
hm
for
c
he
c
k
ing
w
r
ong
indi
r
e
c
t
r
e
lat
ions
hips
in
non
-
fr
e
e
c
hoice
(
A
gung
W
ir
atm
o
)
111
2
.
3
.
C
h
e
c
k
in
g
wr
on
g
in
d
ire
c
t
r
e
la
t
ion
s
h
ip
s
Af
ter
f
indi
ng
thes
e
pa
r
a
mete
r
s
,
then
looki
ng
f
o
r
w
r
ong
indi
r
e
c
t
r
e
lations
hip
in
non
-
f
r
e
e
c
hoice
.
C
he
c
king
is
done
in
e
a
c
h
c
a
s
e
.
T
his
is
be
c
a
us
e
e
a
c
h
c
a
s
e
ha
s
dif
f
e
r
e
nt
pa
r
a
mete
r
s
.
T
he
goa
l
is
to
f
i
nd
f
a
ult
s
with
pr
e
c
e
s
s
ion
f
a
r
be
tt
e
r
than
jus
t
us
ing
the
dir
e
c
ti
on
of
e
a
c
h
c
a
s
e
.
C
he
c
king
s
c
he
me
f
or
non
-
f
r
e
e
c
hoice
s
c
ontaining
indi
r
e
c
t
r
e
lations
a
s
in
a
lgor
it
hm
in
T
a
bl
e
5.
F
r
om
the
a
lgor
it
h
m
in
T
a
ble
5,
the
r
e
a
r
e
ne
e
de
d
s
e
ve
r
a
l
pa
r
a
mete
r
s
li
ke
d
e
cis
io
nP
a
r
a
m
e
te
r
a
nd
.
T
he
pr
oc
e
s
s
will
r
e
pe
a
t
a
s
many
c
a
s
e
s
a
s
in
E
v
e
nt
L
o
g
a
nd
c
he
c
king
in
c
a
s
e
i
w
it
h
.
I
f
the
p
r
oc
e
s
s
is
not
s
a
me
with
the
r
ule,
then
the
c
a
s
e
i
will
be
a
dde
d
in
ℎ
.
T
a
ble
5.
Algor
it
hm
f
o
r
c
he
c
king
indi
r
e
c
t
r
e
lations
h
ip
in
non
-
f
r
e
e
c
hoice
D
a
ta
:
d
e
c
is
io
n
P
a
r
a
m
e
t
e
r
,
Ev
e
n
t
L
o
g
R
e
s
ul
t:
In
d
ir
e
c
t
R
e
l
a
t
io
n
s
hi
p
No
P
s
e
udoc
ode
1
In
d
ir
e
c
t
R
e
l
a
t
io
n
s
hi
p
2
fo
r
e
a
c
h
c
a
s
e
i
of
Ev
e
n
t
L
o
g
do
3
if
c
a
s
e
i
p
a
r
a
m
e
t
e
r
n
o
t
e
q
ua
l
wit
h
d
e
c
is
io
n
P
a
r
a
m
e
t
e
r
t
he
n
4
a
t
t
a
c
h
a
c
a
s
e
i
to
n
o
d
e
In
d
ir
e
c
t
R
e
l
a
t
io
n
s
hi
p
5
e
n
d
6
e
n
d
3.
RE
S
UL
T
S
A
ND
AN
AL
YSI
S
T
he
pr
opos
e
d
method
is
im
p
leme
nted
in
1199
c
a
s
e
s
in
e
ve
nt
log
.
T
he
e
ve
nt
log
ha
s
many
va
r
ious
o
f
a
tt
r
ibut
e
in
pa
r
a
mete
r
l
ike
a
mount
,
s
tac
kT
ype
,
s
tatus
,
a
nd
ti
mes
a
s
in
T
a
ble
1.
F
r
om
the
1199
c
a
s
e
a
c
ti
vit
y,
the
r
e
s
ult
s
of
c
he
c
king
us
ing
the
pr
opos
e
d
met
hod
c
a
n
be
s
e
e
n
in
F
igur
e
3.
F
r
om
F
igu
r
e
3
(
a
)
ha
ving
the
s
e
que
nc
e
of
e
ve
nts
A
→
C
→
D
→
E
→
F
→
H
is
the
s
e
que
nc
e
of
e
ve
nts
that
a
r
e
wr
ong
ba
s
e
d
on
the
pa
r
a
mete
r
s
a
nd
ba
s
e
d
on
the
or
de
r
of
non
-
f
r
e
e
c
hoice
.
C
a
s
e
_I
D
P
P
10
ha
s
s
e
ve
r
a
l
pa
r
a
mete
r
s
,
whi
c
h
va
lue
of
pa
r
a
mete
r
s
tatus
is
c
ompl
e
te
a
nd
in
pa
r
a
mete
r
a
mount
ha
s
3.
I
n
F
igur
e
3
(
a
)
s
hows
a
f
ter
a
c
ti
vi
ty
A
goe
s
to
a
c
ti
vit
y
C
is
a
c
or
r
e
c
t
but
w
r
ong
de
c
is
ion
a
f
te
r
a
c
ti
vit
y
E
goe
s
to
a
c
ti
vit
y
F
.
F
igur
e
3
(
b)
ha
s
the
o
r
de
r
o
f
e
ve
nts
A
→
C
→
D
→
E
→
G
→
H
is
the
s
e
que
nc
e
of
e
ve
nts
incor
r
e
c
tl
y
ba
s
e
d
on
pa
r
a
mete
r
s
due
to
incomplete
a
nd
a
mount
7
s
tatus
pa
r
a
mete
r
s
.
I
n
de
c
is
ion
pa
r
a
mete
r
s
obtai
ne
d
f
r
om
de
c
is
ion
mi
ning
r
e
qui
r
e
s
that
it
c
a
n
pa
s
s
a
c
ti
vit
y
C
then
c
ondit
ions
f
ulf
i
ll
e
d
(
(
s
tatus
=
c
ompl
e
te
AN
D
a
mount
8.
5)
OR
(
s
tatus
=
inc
ompl
e
te
AN
D
a
mount
>
9)
)
a
nd
c
ondi
ti
ons
f
o
r
pa
s
s
ing
a
c
ti
vit
y
G
m
us
t
mee
t
the
r
e
quir
e
ments
(
(
s
tatus
=
c
ompl
e
te
AN
D
a
mount
8.
5)
OR
(
s
tatus
=
incomplete
AN
D
a
mount
>
9)
)
.
How
e
ve
r
,
whe
n
view
e
d
ba
s
e
d
on
the
or
de
r
non
-
f
r
e
e
c
hoice
is
c
or
r
e
c
t.
F
igur
e
3
(
c
)
ha
s
th
e
s
e
que
nc
e
of
e
ve
nts
A
→
B
→
D
→
E
→
G
→
H
is
the
s
e
que
nc
e
of
e
ve
nts
that
a
r
e
wr
ong
ba
s
e
d
on
the
pa
r
a
mete
r
s
a
nd
ba
s
e
d
on
the
o
r
de
r
of
non
-
f
r
e
e
c
hoice
.
C
a
s
e
_I
D
P
P
735
ha
s
s
e
ve
r
a
l
pa
r
a
mete
r
s
,
whic
h
va
lue
of
pa
r
a
mete
r
s
tatus
is
incomplete
a
nd
in
pa
r
a
mete
r
a
mou
nt
ha
s
13.
I
n
F
ig
ur
e
3
(
a
)
s
hows
a
f
ter
a
c
ti
vit
y
A
goe
s
to
a
c
ti
vit
y
B
is
a
c
or
r
e
c
t
but
wr
ong
de
c
is
ion
a
f
ter
a
c
ti
vit
y
E
goe
s
to
a
c
ti
vit
y
G.
F
igur
e
3
(
d
)
ha
s
the
or
de
r
of
e
ve
nts
A
→
B
→
D
→
E
→
F
→
H
is
the
s
e
que
nc
e
of
e
ve
nts
wr
ong
ba
s
e
d
on
pa
r
a
mete
r
s
be
c
a
us
e
c
ompl
e
te
s
tatu
s
pa
r
a
mete
r
s
a
nd
a
mount
3.
I
n
the
de
c
is
ion
pa
r
a
mete
r
s
obtaine
d
f
r
om
de
c
is
ion
mi
ning
r
e
quir
e
s
that
it
c
a
n
pa
s
s
a
c
ti
vit
y
B
then
the
c
ondit
ions
f
ul
f
il
led
(
(
s
tatus
=
c
ompl
e
te
AN
D
a
mount
>
8.
5
)
OR
(
s
tatus
=
incomplete
AN
D
a
mo
unt
9)
)
a
nd
the
c
ondi
ti
on
f
or
pa
s
s
ing
a
c
ti
vit
y
F
m
us
t
mee
t
the
r
e
quir
e
ments
(
(
s
tatus
=
c
ompl
e
te
AN
D
a
mount
>
8.
5)
OR
(
s
tatus
=
incomplete
AN
D
a
mount
9
)
)
.
How
e
ve
r
,
whe
n
view
e
d
ba
s
e
d
on
the
or
de
r
non
-
f
r
e
e
c
hoice
is
c
or
r
e
c
t.
Ac
c
ur
a
c
y
of
the
e
xis
ti
ng
method
s
hows
900
c
a
s
e
a
c
ti
v
it
y
in
,
188
c
a
s
e
a
c
ti
vit
y
in
,
111
c
a
s
e
a
c
ti
vit
y
in
da
n
z
e
r
o
a
c
ti
vit
y
in
.
=
900
+
188
900
+
188
+
111
+
0
100%
=
0
.
907
100%
=
90
.
7%
T
he
a
c
c
ur
a
c
y
of
the
pa
r
a
mete
r
ize
de
c
is
ion
mi
ning
s
hows
900
c
a
s
e
a
c
ti
vit
y
in
,
299
c
a
s
e
a
c
ti
vit
y
in
,
z
e
r
o
c
a
s
e
a
c
ti
vit
y
in
da
n
z
e
r
o
a
c
ti
vit
y
in
.
=
900
+
299
900
+
299
+
0
+
0
100%
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
1693
-
6930
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
,
Vol.
18
,
No
.
1
,
F
e
br
ua
r
y
2020
:
106
-
113
112
=
1
100%
=
100%
F
igur
e
3.
T
he
r
e
s
ult
of
the
p
r
oc
e
s
s
whic
h
c
ontains
wr
ong
indi
r
e
c
t
r
e
lations
hip
in
C
a
s
e
_I
D:
(
a
)
P
P
10
,
(
b
)
P
P
412,
(
c
)
P
P
735
,
a
nd
(
d)
P
P
1050
4.
CONC
L
USI
ON
T
he
de
c
is
ion
mi
ning
ove
r
c
omes
the
c
or
r
e
c
t
f
low
i
n
a
dir
e
c
ti
on
but
not
in
a
pa
r
a
mete
r
ize
d
ir
e
c
ti
on.
T
he
pa
r
a
mete
r
ize
d
de
c
is
ion
m
ini
ng
c
ons
ider
s
pa
r
a
mete
r
s
in
the
s
e
lec
ti
on
o
f
g
r
oove
s
.
T
h
is
pa
pe
r
p
r
opos
e
s
a
c
ombi
na
ti
on
of
pa
r
a
mete
r
ize
d
de
c
is
ion
mi
ning
a
n
d
r
e
lation
s
e
que
nc
e
s
to
de
tec
t
the
dir
e
c
ti
on
a
nd
c
or
r
e
c
tnes
s
.
F
ir
s
tl
y,
dis
c
ove
r
ing
a
gr
a
ph
p
r
oc
e
s
s
model
ba
s
e
d
on
the
e
ve
nt
log.
T
he
n,
a
n
a
na
lys
is
of
the
g
r
a
ph
p
r
oc
e
s
s
model
obtains
de
c
is
ion
point
s
.
T
he
p
r
oc
e
s
s
of
e
a
c
h
de
c
i
s
ion
point
is
us
ing
pa
r
a
mete
r
ize
d
de
c
is
ion
mi
ning
,
s
o
that
de
c
is
ion
r
ules
a
r
e
f
or
med
.
T
he
de
r
ived
de
c
is
ion
r
ules
a
r
e
us
e
d
a
s
pa
r
a
mete
r
s
of
c
he
c
king
wr
ong
indi
r
e
c
t
r
e
lations
hip
in
the
non
-
f
r
e
e
c
hoice
.
T
he
a
c
c
ur
a
c
y
of
the
pa
r
a
mete
r
ize
d
de
c
is
ion
mi
ning
r
e
a
c
he
s
100%
.
I
t
mea
ns
the
pr
opos
e
d
method
c
a
n
de
tec
t
e
r
r
or
s
f
a
r
mo
r
e
pr
e
c
is
e
ly
than
the
e
xis
ti
ng
method
on
ly
ge
t
90
.
7%
a
c
c
ur
a
c
y.
RE
F
E
RE
NC
E
S
[1
]
T
ax
N
.
,
Si
d
o
r
o
v
a
N
.
,
V
a
n
D
er
A
a
l
s
t
W
.
M
.
P.
,
"
D
i
s
c
o
v
er
i
n
g
m
o
re
p
reci
s
e
p
r
o
ces
s
m
o
d
e
l
s
fr
o
m
e
v
en
t
l
o
g
s
b
y
f
i
l
t
eri
n
g
o
u
t
c
h
ao
t
i
c
ac
t
i
v
i
t
i
e
s
,"
Jo
u
r
n
a
l
o
f
In
t
e
l
l
i
g
e
n
t
I
n
f
o
r
m
a
t
i
o
n
S
ys
t
e
m
s
,
v
o
l
.
52
,
n
o
.
1
,
p
p
.
1
0
7
–
1
3
9
,
2
0
1
9
.
[2
]
V
an
D
er
A
a
l
s
t
W
.
M
.
P.
,
"
D
eco
m
p
o
s
i
n
g
Pe
t
r
i
n
e
t
s
f
o
r
p
r
o
ces
s
mi
n
i
n
g
:
A
g
en
er
i
c
a
p
p
r
o
ach
,"
D
i
s
t
r
i
b
u
t
e
d
a
n
d
P
a
r
a
l
l
el
D
a
t
a
b
a
s
es
,
v
o
l
.
31
,
n
o
.
4
,
p
p
.
4
7
1
–
5
0
7
,
2
0
1
3
.
[3
]
A
l
eem
S
.
,
Fern
an
d
o
Cap
ret
z
L
.
,
A
h
med
F.
,
"
Bu
s
i
n
es
s
Pro
ces
s
Mi
n
i
n
g
A
p
p
r
o
ach
e
s
:
A
Rel
a
t
i
v
e
Co
mp
ar
i
s
o
n
,"
In
t
e
r
n
a
t
i
o
n
a
l
Jo
u
r
n
a
l
o
f
S
c
i
en
ce,
Tech
n
o
l
o
g
y
&
M
a
n
a
g
e
m
en
t
,
v
o
l
.
4
,
n
o
.
1
,
p
p
.
1
5
5
7
–
1
5
6
4
,
2
0
1
5
.
[4
]
V
an
d
er
A
al
s
t
W
.
,
"
Pro
ces
s
D
i
s
c
o
v
er
y
:
A
n
In
t
ro
d
u
c
t
i
o
n
,"
P
r
o
ces
s
M
i
n
i
n
g
-
S
p
r
i
n
g
er
,
p
p
.
1
2
5
–
1
5
6
,
2
0
1
1
.
[5
]
Ro
zi
n
at
A
,
V
a
n
d
er
A
al
s
t
W
.
M
.
P.
,
"
D
eci
s
i
o
n
M
i
n
i
n
g
i
n
Pro
M
,"
In
t
e
r
n
a
t
i
o
n
a
l
Co
n
f
e
r
en
ce
o
n
B
u
s
i
n
e
s
s
P
r
o
c
es
s
M
a
n
a
g
em
e
n
t
,
p
p
.
4
2
0
–
4
2
5
,
2
0
0
6
.
[6
]
H
o
r
i
t
a
H
.
,
H
i
ray
ama
H
.
,
H
ay
as
e
T
.
,
T
a
h
ara
Y
,
O
h
s
u
g
a
A
.
,
"
Pro
ces
s
m
i
n
i
n
g
ap
p
ro
ac
h
b
a
s
ed
o
n
p
art
i
al
s
t
r
u
ct
u
re
s
o
f
ev
en
t
l
o
g
s
a
n
d
d
eci
s
i
o
n
t
ree
l
ear
n
i
n
g
,"
P
r
o
cee
d
i
n
g
s
-
2
0
1
6
5
th
IIA
I
I
n
t
e
r
n
a
t
i
o
n
a
l
Co
n
g
r
es
s
o
n
A
d
v
a
n
ce
d
A
p
p
l
i
ed
In
f
o
r
m
a
t
i
c
s
,
IIA
I
-
A
A
I
,
2
0
1
6
.
[7
]
W
en
L
.
,
W
an
g
J
.
,
Su
n
J
.
,
"
D
et
ect
i
n
g
i
m
p
l
i
ci
t
d
ep
e
n
d
en
ci
e
s
b
et
w
een
t
a
s
k
s
fro
m
ev
e
n
t
l
o
g
s
,"
A
s
i
a
-
P
a
ci
f
i
c
W
eb
Co
n
f
er
e
n
ce
,
p
p
.
5
9
1
–
6
0
3
,
2
0
0
6
.
[8
]
V
an
D
er
A
al
s
t
W
.
M
.
P
.
,
D
e
Med
ei
ro
s
A
.
K
.
A.
,
"
Pro
ces
s
mi
n
i
n
g
an
d
s
ecu
r
i
t
y
:
D
et
ec
t
i
n
g
an
o
mal
o
u
s
p
r
o
ces
s
ex
ecu
t
i
o
n
s
an
d
ch
ec
k
i
n
g
p
ro
c
es
s
co
n
fo
rma
n
ce
,"
E
l
ec
t
r
o
n
i
c
N
o
t
e
s
i
n
Th
e
o
r
e
t
i
c
a
l
C
o
m
p
u
t
er
S
c
i
en
ce
,
v
o
l
.
1
2
1
,
p
p
.
3
-
2
1
,
2
0
0
5
.
[9
]
Su
n
g
k
o
n
o
K
.
R
.
,
Sarn
o
R.
,
"
Co
n
s
t
r
u
c
t
i
n
g
c
o
n
t
ro
l
-
fl
o
w
p
at
t
ern
s
co
n
t
a
i
n
i
n
g
i
n
v
i
s
i
b
l
e
t
a
s
k
a
n
d
n
o
n
-
free
ch
o
i
c
e
b
as
ed
o
n
d
ec
l
arat
i
v
e
m
o
d
e
l
,"
In
t
er
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
In
n
o
v
a
t
i
ve
Co
m
p
u
t
i
n
g
,
In
f
o
r
m
a
t
i
o
n
a
n
d
Co
n
t
r
o
l
,
v
o
l
.
1
4
,
n
o
.
4
,
p
p
.
1
2
8
5
-
1
2
9
9
,
2
0
1
8
.
[1
0
]
W
en
L
.
,
V
an
D
er
A
al
s
t
W
.
M
.
P
.
,
W
an
g
J
.
,
Su
n
J
.
,
"
Mi
n
i
n
g
p
r
o
ces
s
mo
d
e
l
s
w
i
t
h
n
o
n
-
free
-
ch
o
i
ce
co
n
s
t
ru
ct
s
,"
D
a
t
a
M
i
n
i
n
g
a
n
d
Kn
o
wl
e
d
g
e
D
i
s
co
ve
r
y
,
v
o
l
.
15
,
n
o
.
2
,
p
p
.
145
–
1
8
0
,
2
0
0
7
.
[1
1
]
K
al
y
n
y
ch
e
n
k
o
O
.
,
Ch
al
y
i
S
.
,
Bo
d
y
an
s
k
i
y
Y
.
,
G
o
l
i
an
V
.
,
G
o
l
i
a
n
N
.
,
"
Imp
l
emen
t
a
t
i
o
n
o
f
s
earc
h
mech
an
i
s
m
fo
r
i
m
p
l
i
c
i
t
d
ep
e
n
d
e
n
ces
i
n
p
r
o
ces
s
mi
n
i
n
g
,"
P
r
o
c
eed
i
n
g
s
o
f
t
h
e
2
0
1
3
IE
E
E
7
th
In
t
er
n
a
t
i
o
n
a
l
Co
n
f
e
r
en
ce
o
n
In
t
el
l
i
g
en
t
D
a
t
a
A
cq
u
i
s
i
t
i
o
n
a
n
d
A
d
v
a
n
ce
d
Co
m
p
u
t
i
n
g
S
ys
t
em
s
,
2
0
1
3
.
[1
2
]
V
an
D
er
A
a
l
s
t
W
.
M
.
P
.
,
D
e
Beer
H
.
T
.
,
V
an
D
o
n
g
en
B
.
F.
,
"
Pro
ces
s
M
i
n
i
n
g
an
d
V
er
i
fi
ca
t
i
o
n
o
f
Pro
p
er
t
i
e
s
:
A
n
A
p
p
ro
ac
h
b
as
e
d
o
n
T
emp
o
ral
L
o
g
i
c
,"
P
r
o
ceed
i
n
g
s
o
f
t
h
e
2
0
0
5
Co
n
f
e
d
er
a
t
e
d
I
n
t
e
r
n
a
t
i
o
n
a
l
Co
n
f
e
r
en
ce
o
n
O
n
t
h
e
M
o
ve
t
o
M
ea
n
i
n
g
f
u
l
In
t
er
n
et
S
y
s
t
e
m
s
,
p
p
.
1
3
0
–
1
4
7
,
2
0
0
5
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
Gr
aph
-
bas
e
d
algor
it
hm
for
c
he
c
k
ing
w
r
ong
indi
r
e
c
t
r
e
lat
ions
hips
in
non
-
fr
e
e
c
hoice
(
A
gung
W
ir
atm
o
)
113
[1
3
]
Ch
ab
r
o
l
M
.
,
D
al
mas
B
.
,
N
o
rre
S
.
,
Ro
d
i
er
S.
,
"
A
p
ro
ces
s
t
ree
-
b
a
s
ed
al
g
o
r
i
t
h
m
fo
r
t
h
e
d
e
t
ec
t
i
o
n
o
f
i
m
p
l
i
ci
t
d
ep
e
n
d
e
n
ci
e
s
,"
2
0
1
6
IE
E
E
Te
n
t
h
In
t
er
n
a
t
i
o
n
a
l
Co
n
f
er
e
n
ce
o
n
R
es
e
a
r
c
h
Ch
a
l
l
en
g
es
i
n
I
n
f
o
r
m
a
t
i
o
n
S
c
i
e
n
ce
(R
C
IS
)
,
2
0
1
6
.
[1
4
]
J
o
i
s
h
i
J
.
,
Su
rek
a
A
.
,
"
G
rap
h
o
r
Rel
at
i
o
n
a
l
D
at
a
b
as
e
s
:
A
Sp
eed
Co
mp
ar
i
s
o
n
fo
r
Pro
ce
s
s
Mi
n
i
n
g
A
l
g
o
ri
t
h
m
,"
A
r
X
i
v
,
2
0
1
6
.
[1
5
]
K
u
s
h
w
ah
a
A
.
,
Pan
d
e
y
R
.
S.
,
"
A
G
rap
h
Bas
ed
A
p
p
ro
ac
h
T
o
Id
e
n
t
i
fy
O
b
j
ec
t
s
U
s
i
n
g
Id
en
t
i
f
y
i
n
g
A
t
t
ri
b
u
t
e
,"
In
d
o
n
e
s
i
a
n
Jo
u
r
n
a
l
o
f
E
l
ect
r
i
c
a
l
E
n
g
i
n
ee
r
i
n
g
a
n
d
Co
m
p
u
t
er
S
ci
e
n
c
e
,
v
o
l
.
6
,
n
o
.
2
,
p
p
.
4
3
8
-
4
4
6
,
2
0
1
7
.
[1
6
]
Pat
i
l
N
.
S
.
,
K
i
ran
P
.
,
K
i
ran
N
.
P
.
,
N
ares
h
P
.
K
.
M.
,
"
A
Su
rv
ey
o
n
G
rap
h
D
at
a
b
as
e
Man
a
g
emen
t
T
ech
n
i
q
u
e
s
fo
r
H
u
g
e
U
n
s
t
r
u
ct
u
red
D
a
t
a
,"
In
t
er
n
a
t
i
o
n
a
l
Jo
u
r
n
a
l
o
f
E
l
ec
t
r
i
ca
l
a
n
d
Co
m
p
u
t
e
r
E
n
g
i
n
eer
i
n
g
,
v
o
l
.
8
,
n
o
.
2
,
p
p
.
1
1
4
0
-
1
1
4
9
,
2
0
1
8
.
[1
7
]
Sarn
o
R
.
,
Su
n
g
k
o
n
o
K
.
,
J
o
h
a
n
es
R
.
,
Su
n
ar
y
o
n
o
D
.
,
"
G
rap
h
-
Bas
e
d
A
l
g
o
r
i
t
h
ms
fo
r
D
i
s
co
v
eri
n
g
a
Pro
ce
s
s
M
o
d
e
l
Co
n
t
ai
n
i
n
g
In
v
i
s
i
b
l
e
T
a
s
k
s
,"
In
t
er
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
In
t
el
l
i
g
en
t
E
n
g
i
n
eer
i
n
g
a
n
d
S
y
s
t
e
m
s
,
v
o
l
.
1
2
,
n
o
.
2
,
p
p
.
8
5
-
9
4
,
2
0
1
9
.
[1
8
]
Iw
as
a
k
i
K
.
,
K
u
ri
y
ama
Y
.
,
K
o
n
d
o
h
S
.
,
Sh
i
ra
y
o
r
i
A
.
,
"
St
r
u
ct
u
ri
n
g
e
n
g
i
n
eer
s
’
i
m
p
l
i
ci
t
k
n
o
w
l
e
d
g
e
o
f
f
o
rmi
n
g
p
ro
c
es
s
d
es
i
g
n
b
y
u
s
i
n
g
a
g
rap
h
mo
d
e
l
,"
P
r
o
c
ed
i
a
CIR
P
,
v
o
l
.
6
7
,
p
p
.
563
–
5
6
8
,
2
0
1
8
.
[1
9
]
Ch
ap
e
l
a
-
Camp
a
D
.
,
Mu
ci
e
n
t
e
s
M
.
,
L
ama
M.
,
"
Mi
n
i
n
g
F
req
u
e
n
t
Pat
t
er
n
s
i
n
Pro
ce
s
s
M
o
d
e
l
s
,"
In
f
o
r
m
a
t
i
o
n
S
c
i
en
c
es
,
v
o
l
.
47
2
,
p
p
.
2
3
5
–
2
5
7
,
2
0
1
9
.
[2
0
]
Caes
ari
t
a
Y
.
,
Sarn
o
R
.
,
Su
n
g
k
o
n
o
K
.
R.
,
"
Id
en
t
i
f
y
i
n
g
b
o
t
t
l
en
ec
k
s
an
d
frau
d
o
f
b
u
s
i
n
es
s
p
ro
ce
s
s
u
s
i
n
g
al
p
h
a
+
+
an
d
h
eu
r
i
s
t
i
c
mi
n
er
al
g
o
r
i
t
h
ms
(Ca
s
e
s
t
u
d
y
:
CV
.
W
i
cak
s
an
a
A
rt
h
a)
,
"
2
0
1
7
1
1
th
In
t
e
r
n
a
t
i
o
n
a
l
Co
n
f
e
r
en
ce
o
n
I
n
f
o
r
m
a
t
i
o
n
&
Co
m
m
u
n
i
ca
t
i
o
n
Tec
h
n
o
l
o
g
y
a
n
d
S
ys
t
em
,
2
0
1
7
.
[2
1
]
Bo
l
t
A
.
,
d
e
L
eo
n
i
M
.
,
v
an
d
er
A
a
l
s
t
W
.
M
.
P.
,
"
Pro
ces
s
v
ari
an
t
co
m
p
ari
s
o
n
:
U
s
i
n
g
ev
e
n
t
l
o
g
s
t
o
d
e
t
ect
d
i
ff
eren
ce
s
i
n
b
eh
a
v
i
o
r
an
d
b
u
s
i
n
es
s
ru
l
e
s
,"
In
f
o
r
m
a
t
i
o
n
S
y
s
t
e
m
s
,
vo
l
.
7
4
,
p
p
.
53
–
66
,
2
0
1
8
.
[2
2
]
D
ak
i
c
D
.
,
St
efan
o
v
i
c
D
.
,
"
Bu
s
i
n
e
s
s
Pro
ces
s
Mi
n
i
n
g
A
p
p
l
i
ca
t
i
o
n
:
A
L
i
t
erat
u
re
Rev
i
ew
,"
P
r
o
cee
d
i
n
g
s
o
f
t
h
e
2
9
th
In
t
e
r
n
a
t
i
o
n
a
l
D
A
A
A
M
S
y
m
p
o
s
i
u
m
,
pp.
0
8
6
6
–
0
8
7
5
,
2
0
1
8
.
[2
3
]
Rezai
ee
A
.
M
.
,
K
ar
i
mi
A
.
,
"
A
N
ew
D
y
n
ami
c
In
t
el
l
i
g
en
t
Mo
d
e
l
t
o
D
e
t
ermi
n
e
Rel
i
ab
i
l
i
t
y
an
d
T
r
u
s
t
o
f
O
n
l
i
n
e
Ba
n
k
i
n
g
b
y
U
s
i
n
g
F
u
zzy
C
-
Mea
n
,"
In
d
o
n
es
i
a
n
Jo
u
r
n
a
l
o
f
E
l
e
ct
r
i
ca
l
E
n
g
i
n
eer
i
n
g
a
n
d
C
o
m
p
u
t
er
S
c
i
en
ce
,
v
o
l
.
4
,
n
o
.
3
,
p
p
.
6
0
5
-
6
1
0
,
2
0
1
6
.
[2
4
]
Ait
-
Ml
o
u
k
A
.
,
A
g
o
u
t
i
T
.
,
G
h
ar
n
at
i
F.
,
"
Mi
n
i
n
g
an
d
p
r
i
o
r
i
t
i
zat
i
o
n
o
f
a
s
s
o
ci
a
t
i
o
n
ru
l
es
f
o
r
b
i
g
d
a
t
a:
m
u
l
t
i
-
cr
i
t
eri
a
d
eci
s
i
o
n
an
a
l
y
s
i
s
ap
p
ro
ac
h
,"
Jo
u
r
n
a
l
o
f
B
i
g
D
a
t
a
,
v
o
l
.
4
,
n
o
.
4
2
,
p
p
.
1
-
2
1
,
2
0
1
7
.
[2
5
]
Man
n
h
ard
t
F
.
,
D
e
L
eo
n
i
M
.
,
Rei
j
ers
H
.
A
.
,
V
an
D
er
A
al
s
t
W
.
M
.
P.
,
"
D
eci
s
i
o
n
Mi
n
i
n
g
Rev
i
s
i
t
ed
-
D
i
s
co
v
er
i
n
g
O
v
er
l
ap
p
i
n
g
Ru
l
es
,
"
In
t
e
r
n
a
t
i
o
n
a
l
Co
n
f
e
r
en
ce
o
n
A
d
va
n
ced
In
f
o
r
m
a
t
i
o
n
S
ys
t
em
s
E
n
g
i
n
eer
i
n
g
,
p
p
.
3
7
7
-
3
9
2
,
2
0
1
6
.
[2
6
]
Si
n
g
h
D
.
,
Ch
o
u
d
h
ary
N
.
,
Samo
t
a
J
.
,
"
A
n
al
y
s
i
s
o
f
D
a
t
a
Mi
n
i
n
g
C
l
as
s
i
f
i
cat
i
o
n
w
i
t
h
D
ec
i
s
i
o
n
t
ree
T
ec
h
n
i
q
u
e
,"
G
l
o
b
a
l
Jo
u
r
n
a
l
s
I
n
c.
(U
S
A
)
,
v
o
l
.
1
3
,
p
p
.
1
-
5
,
2
0
1
3
.
[2
7
]
Bo
mb
ara
G
.
,
V
as
i
l
e
C
-
I,
Pen
e
d
o
F
.
,
Y
as
u
o
k
a
H
.
,
Bel
t
a
C.
,
"
A
D
eci
s
i
o
n
T
ree
A
p
p
ro
ac
h
t
o
D
at
a
Cl
a
s
s
i
fi
ca
t
i
o
n
u
s
i
n
g
Si
g
n
a
l
T
emp
o
ral
L
o
g
i
c
,"
P
r
o
ceed
i
n
g
s
o
f
t
h
e
1
9
th
In
t
er
n
a
t
i
o
n
a
l
C
o
n
f
er
e
n
ce
o
n
H
yb
r
i
d
S
y
s
t
e
m
s
:
Co
m
p
u
t
a
t
i
o
n
a
n
d
Co
n
t
r
o
l
-
H
S
CC
’
1
6
,
2
0
1
6
.
[2
8
]
Saet
t
l
er
A
.
,
L
ab
er
E
,
d
e
A
.
,
Mel
l
o
Perei
ra
F.
,
"
D
eci
s
i
o
n
t
ree
cl
as
s
i
f
i
cat
i
o
n
w
i
t
h
b
o
u
n
d
ed
n
u
mb
er
o
f
erro
rs
,"
In
f
o
r
m
a
t
i
o
n
P
r
o
ce
s
s
i
n
g
Let
t
er
s
,
v
o
l
.
1
2
7
,
p
p
.
2
7
-
3
1
,
2
0
1
7
.
[2
9
]
Card
o
s
o
J
.
,
"
Bu
s
i
n
es
s
Pro
ces
s
Co
n
t
ro
l
-
Fl
o
w
Co
m
p
l
e
x
i
t
y
:
Met
ri
c,
E
v
a
l
u
a
t
i
o
n
,
an
d
V
al
i
d
a
t
i
o
n
,"
In
t
e
r
n
a
t
i
o
n
a
l
Jo
u
r
n
a
l
o
f
W
e
b
S
e
r
vi
ce
s
R
es
e
a
r
c
h
,
v
o
l
.
5
,
n
o
.
2
,
p
p
.
4
9
-
7
6
,
2
0
0
8
.
[3
0
]
L
i
m
H
.
W
.
,
K
ers
ch
b
au
m
F
.
,
W
an
g
H
.
,
"
W
o
r
k
fl
o
w
Si
g
n
a
t
u
res
fo
r
Bu
s
i
n
es
s
Pro
ces
s
Co
mp
l
i
a
n
ce
,"
IE
E
E
Tr
a
n
s
a
c
t
i
o
n
s
o
n
D
e
p
en
d
a
b
l
e
a
n
d
S
ecu
r
e
Co
m
p
u
t
i
n
g
,
v
o
l
.
9
,
n
o
.
5
,
p
p
.
7
5
6
-
7
6
9
,
2
0
1
2
.
Evaluation Warning : The document was created with Spire.PDF for Python.