T
E
L
KO
M
NIK
A
, V
ol
.
16
,
No.
6,
Dec
em
be
r
20
1
8,
p
p.
2
80
9
~
2
81
6
IS
S
N: 1
69
3
-
6
93
0
,
accr
ed
ited
F
irst
Gr
ad
e b
y K
em
en
r
istekdikti,
Decr
ee
No: 2
1/E/
K
P
T
/20
18
DOI:
10.12928/TE
LK
OM
N
IK
A
.v
1
6
i
6
.
10568
28
09
Rec
ei
v
ed
J
ul
y
15
, 2
01
8
; Re
v
i
s
ed
S
ep
te
mb
er 7
,
20
1
8
;
A
c
c
ep
ted
O
c
tob
er
8
, 2
01
8
A
n
o
m
al
y
D
etec
t
ion
b
ased
o
n
C
on
tr
ol
-
f
lo
w
Pattern
o
f
P
arall
el
B
usi
ne
ss
P
rocesse
s
Hend
r
a D
ar
m
aw
an
*
, Ri
ya
n
ar
t
o
S
a
r
n
o
,
A
d
h
atu
s
S
o
li
chah
A
h
mad
i
ya
h
,
Kell
y
Ro
ss
a
S
u
n
g
kon
o
,
C
ah
y
anin
g
t
y
a
s
S
ek
a
r
W
ah
y
u
n
i
Dep
a
rtm
e
n
t
o
f
In
f
o
rm
a
ti
c
s
,
In
s
ti
tu
t
T
e
k
n
o
l
o
g
i
Se
p
u
l
u
h
No
p
e
m
b
e
r
Ka
m
p
u
s
I
T
S Su
k
o
l
i
l
o
,
J
a
l
a
n
R
a
y
a
I
T
S,
Su
ra
b
a
y
a
,
J
a
w
a
T
i
m
u
r 6
0
1
1
1
,
(
0
3
1
)
5
9
9
4
2
5
1
,
I
n
d
o
n
e
s
i
a
*C
o
rre
s
p
o
n
d
i
n
g
a
u
th
o
r
,
e
-
m
a
i
l
:
h
e
n
d
r
a
d
n
9
6
@gm
a
i
l
.c
o
m
Ab
strac
t
Th
e
p
u
rp
o
s
e
o
f
th
i
s
p
a
p
e
r
was
to
d
i
s
c
o
v
e
r
a
n
a
n
o
m
a
l
o
u
s
-
fre
e
b
u
s
i
n
e
s
s
p
ro
c
e
s
s
m
o
d
e
l
fro
m
e
v
e
n
t
l
o
g
s
.
Th
e
p
ro
c
e
s
s
d
i
s
c
o
v
e
ry
was
c
o
n
d
u
c
te
d
u
s
i
n
g
a
g
ra
p
h
d
a
t
a
b
a
s
e
,
s
p
e
c
i
f
i
c
a
l
l
y
u
s
i
n
g
Neo
4
J
t
o
o
l
i
n
v
o
l
v
i
n
g
tr
a
c
e
c
l
u
s
te
r
i
n
g
a
n
d
d
a
ta
f
i
l
te
r
i
n
g
p
ro
c
e
s
s
e
s
.
W
e
a
l
s
o
d
e
v
e
l
o
p
e
d
a
c
o
n
tro
l
-
fl
o
w
p
a
tt
e
rn
t
o
a
d
d
re
s
s
,
AND
re
l
a
ti
o
n
b
e
twe
e
n
a
c
t
i
v
i
ti
e
s
n
a
m
e
d
p
a
ra
l
l
e
l
b
u
s
i
n
e
s
s
p
ro
c
e
s
s
.
Th
e
re
s
u
l
t
s
h
o
wed
th
a
t
t
h
e
p
ro
p
o
s
e
d
m
e
th
o
d
i
m
p
ro
v
e
d
t
h
e
p
re
c
i
s
i
o
n
v
a
l
u
e
o
f
th
e
g
e
n
e
r
a
te
d
b
u
s
i
n
e
s
s
p
ro
c
e
s
s
m
o
d
e
l
f
ro
m
0
.6
4
t
o
0
.
8
1
c
o
m
p
a
re
d
t
o
th
e
e
x
i
s
t
i
n
g
a
l
g
o
ri
th
m
.
Th
e
b
e
tt
e
r
o
u
tc
o
m
e
i
s
c
o
n
s
tru
c
te
d
b
y
a
p
p
l
y
i
n
g
tra
c
e
c
l
u
s
t
e
ri
n
g
a
n
d
d
a
ta
f
i
l
te
r
i
n
g
to
re
m
o
v
e
th
e
a
n
o
m
a
l
y
o
n
t
h
e
e
v
e
n
t
l
o
g
a
s
wel
l
a
s
d
i
s
c
o
v
e
ri
n
g
p
a
ra
l
l
e
l
(AND
)
re
l
a
ti
o
n
b
e
twe
e
n
a
c
ti
v
i
ti
e
s
.
Key
w
ords
:
a
n
o
m
a
l
y
d
a
ta
f
i
l
te
r
i
n
g
,
c
o
n
tr
o
l
-
f
l
o
w
p
a
tt
e
rn
,
g
ra
p
h
d
a
ta
b
a
s
e
,
p
ro
c
e
s
s
d
i
s
c
o
v
e
r
y
Copy
righ
t
©
2
0
1
8
Uni
v
e
rsi
t
a
s
Ahm
a
d
D
a
hl
a
n.
All
rig
ht
s
r
e
s
e
rve
d
.
1.
Int
r
o
d
u
ctio
n
P
r
oc
es
s
di
s
c
ov
er
y
ha
s
be
en
i
nte
ns
i
v
el
y
us
e
d
to
ge
n
erate
b
us
i
n
es
s
proc
es
s
m
od
e
l
[1
]
as
i
t
i
s
b
en
ef
i
c
i
al
i
n
s
up
p
orti
ng
the
c
om
pa
ni
es
to
u
nd
ers
tan
d
th
e
bu
s
i
ne
s
s
pr
oc
es
s
,
i
de
n
ti
f
y
po
s
s
i
bl
e
an
om
al
i
es
,
m
on
i
tor
an
d
c
on
tr
o
l
the
ex
ec
ute
d
op
erat
i
o
n
[2]
.
T
he
ge
n
er
ate
d
b
us
i
n
es
s
proc
es
s
m
od
el
r
e
pres
en
ts
t
he
b
eh
a
v
i
or of
pe
r
f
or
m
ed
a
c
ti
v
i
ti
es
m
i
ne
d f
r
o
m
ev
en
t
l
og
s
[3
].
T
he
r
e
are
s
ev
eral
al
go
r
i
t
hm
s
i
ntrodu
c
ed
to
m
i
ne
t
he
ev
en
t
l
og
s
,
s
uc
h
as
A
l
ph
a
Mi
n
er
[4]
,
He
uris
ti
c
Mi
ne
r
[
5
,
6],
A
l
p
ha
-
T
[7]
,
Co
up
l
ed
Hi
dd
en
Ma
r
k
ov
M
od
e
l
of
Non
-
f
r
ee
Cho
i
c
e
an
d
In
v
i
s
i
bl
e
P
r
i
m
e
tas
k
(
C
HMM
-
NCIT
)
[8]
,
Hi
dd
en
M
ark
ov
Mo
de
l
of
P
ara
l
l
el
B
u
s
i
ne
s
s
P
r
oc
es
s
(
HMM
-
P
ar
al
l
el
)
[9]
,
an
d
Co
up
l
ed
H
i
dd
en
M
ark
ov
Mo
de
l
of
In
v
i
s
i
bl
e
T
as
k
(
CHMM
-
Inv
i
s
i
b
l
e) [1
].
T
he
r
ec
en
t
r
es
e
arc
he
s
o
n
bu
s
i
ne
s
s
proc
es
s
di
s
c
ov
e
r
y
are
o
n
a
no
m
al
y
de
t
ec
ti
on
.
A
n
A
l
p
ha
a
l
go
r
i
thm
i
s
a
ba
s
i
c
al
go
r
i
t
hm
to
ge
ne
r
at
e
a
bu
s
i
n
es
s
proc
es
s
m
od
el
f
r
om
the
e
v
en
t
l
og
s
[10
]
,
bu
t
i
t
i
s
l
i
m
i
ted
to
m
i
ni
ng
s
ho
r
t
l
oo
ps
[11
].
A
l
p
ha
+
al
go
r
i
t
hm
w
as
the
n
i
ntrodu
c
e
d
to
s
ol
v
e
A
l
ph
a
al
g
orit
hm
pro
bl
em
,
bu
t
i
t
f
ai
l
ed
to
m
i
ne
no
n
-
f
r
ee
c
ho
i
c
e
or
i
m
pl
i
c
i
t
de
pe
nd
e
nc
i
es
an
om
al
y
[
12
].
A
l
ph
a+
+
al
g
orit
hm
an
s
wer
ed
s
h
ort
l
oo
p
m
i
ni
ng
an
d
no
n
-
f
r
ee
c
ho
i
c
e
pr
ob
l
em
s
,
ho
w
e
v
er,
i
t
was
u
na
b
l
e
t
o
m
i
ne
i
n
v
i
s
i
b
l
e
tas
k
an
o
m
al
y
whi
c
h
the
n
a
dd
r
es
s
ed
b
y
A
l
ph
a
#
al
g
orit
hm
[13
].
A
l
p
ha
$
a
l
g
orit
hm
[14
]
w
as
ab
l
e
to
di
s
c
ov
er
bo
th
i
nv
i
s
i
b
l
e
tas
k
an
d
no
n
-
f
r
ee
c
ho
i
c
e
an
om
al
i
es
s
i
m
ul
tan
eo
us
l
y
.
Ho
w
e
v
er,
A
l
ph
a
m
i
ne
r
al
g
orit
hm
s
as
the
ab
o
v
e
-
m
en
ti
on
ed
proc
es
s
ed
ev
en
t
l
og
s
d
i
r
e
c
tl
y
wi
tho
ut
da
ta
f
i
l
t
erin
g
,
an
om
al
y
bu
s
i
ne
s
s
proc
e
s
s
du
e
to
the
ex
i
s
ten
c
e
of
l
o
w
-
f
r
eq
ue
nc
y
v
a
l
ue
s
o
n s
e
v
era
l
trac
e c
l
u
s
ters
.
T
he
r
ef
ore,
thi
s
pa
pe
r
pro
po
s
ed
a
m
eth
od
to
tac
k
l
e
thi
s
prob
l
em
b
y
us
i
ng
a
graph
da
ta
ba
s
e
i
nc
orpor
at
i
ng
tr
ac
e
c
l
us
teri
n
g
an
d
da
ta
f
i
l
t
eri
ng
.
G
r
ap
h
d
ata
b
as
e
w
as
u
s
ed
s
i
nc
e
i
t
i
s
f
l
ex
i
bl
e
to
s
tore
i
nte
r
c
on
n
ec
ted
ac
ti
v
i
t
i
es
of
bu
s
i
ne
s
s
[15
],
to
s
ho
w
th
e
r
el
ati
o
ns
am
on
g
ac
ti
v
i
t
i
es
[16
]
,
an
d
t
o
ad
j
u
s
t
bu
s
i
ne
s
s
ne
e
ds
[17
].
N
eo
4J
w
as
em
pl
o
y
e
d
a
l
on
g
wi
th
C
y
ph
er
Q
ue
r
y
L
an
g
ua
ge
(
CQ
L)
[
18
]
.
2.
Re
se
a
r
ch M
eth
o
d
F
i
gu
r
e
1
s
h
o
w
s
o
ur
prop
os
ed
m
eth
od
to
ge
n
erate
an
an
om
al
ou
s
-
f
r
ee
bu
s
i
ne
s
s
p
r
oc
es
s
m
od
el
.
It
c
o
ns
i
s
ts
of
t
w
o
m
ai
n
s
tag
es
.
T
he
f
i
r
s
t
o
ne
i
s
t
he
an
om
al
y
de
l
e
ti
o
n
s
tag
e,
w
h
i
l
e
t
he
s
ec
on
d s
tag
e
d
ea
l
s
wi
t
h b
u
s
i
ne
s
s
proc
es
s
di
s
c
o
v
er
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N: 16
93
-
6
93
0
T
E
L
KO
M
NIK
A
V
ol
.
16
,
No
.
6,
D
ec
em
be
r
20
18
:
280
9
-
281
6
2810
F
i
gu
r
e
1
.
F
r
om
ev
en
t
l
og
s
t
o g
e
ne
r
at
ed
a
no
m
al
ou
s
-
f
r
ee
bu
s
i
ne
s
s
proc
es
s
m
od
el
i
n
the
f
orm
o
f
a
graph
m
od
el
2.1.
A
n
o
m
al
y
D
ele
t
ion
S
t
a
g
e
T
hi
s
s
tag
e
i
s
de
di
c
at
ed
to
r
em
ov
i
ng
a
l
l
p
os
s
i
b
l
e
b
u
s
i
ne
s
s
proc
es
s
an
om
al
i
es
i
n
th
e
ev
e
nt
l
og
.
F
i
r
s
t,
a
grap
h
m
od
e
l
i
s
c
on
s
tr
uc
te
d
f
r
o
m
the
ev
en
t
l
og
s
b
y
l
oa
di
n
g
th
e
ev
e
nt
l
og
s
i
nto
NE
O
4J
.
T
he
r
e
are
tw
o
f
un
c
ti
on
s
to
l
o
ad
the
e
v
e
nt
l
o
gs
,
i
.e.
,
c
r
ea
te
an
d
m
er
ge
f
un
c
ti
on
s
.
T
he
c
r
ea
te
f
un
c
ti
on
i
s
us
ed
s
i
n
c
e
i
t
c
r
ea
te
s
n
od
es
b
as
ed
on
a
l
l
ex
i
s
ti
n
g
da
t
a
i
n
the
e
v
en
t
l
o
g
s
.
T
he
ev
e
nt
l
og
s
are
l
oa
d
ed
i
nto
two
l
a
be
l
s
,
i
.e.
,
A
c
ti
v
i
t
y
a
nd
Cas
e
A
c
ti
v
i
t
y
.
A
c
ti
v
i
t
y
l
a
be
l
c
on
tai
n
s
al
l
c
ol
um
n
s
i
ns
i
de
of
ev
en
t
l
o
g
s
w
h
i
l
e
Cas
e
A
c
ti
v
i
t
y
l
ab
el
o
nl
y
c
on
ta
i
ns
Cas
e
Id
an
d
ac
ti
v
i
t
y
na
m
e.
T
he
A
c
ti
v
i
t
y
l
a
be
l
i
s
us
ed
f
or
da
ta
f
i
l
teri
ng
s
tep
,
w
hi
l
e
Cas
e
A
c
ti
v
i
t
y
l
ab
el
i
s
us
ed
f
or
c
as
e
s
eq
ue
nc
e
d
i
s
c
ov
er
y
an
d
tr
ac
e
c
l
us
teri
n
g
s
tep
s
.
F
i
gu
r
e
2
s
ho
w
s
the
q
ue
r
y
s
tr
uc
t
ure
of
l
oa
di
ng
ev
e
nt
l
og
s
i
nto
Neo
4J
.
F
i
gu
r
e
2
. L
oa
d
e
v
en
t
l
o
g
i
nt
o
n
e
o4
j
us
i
n
g
c
r
ea
t
e
f
un
c
ti
o
n
q
u
er
y
Hav
i
ng
l
o
ad
e
d
the
ev
en
t
l
o
gs
i
nto
no
d
es
,
the
y
are
c
o
nn
ec
te
d
b
y
c
as
e
ID
an
d
th
e
order
of
ac
ti
v
i
t
y
i
n
the
as
s
oc
i
ate
d
ev
e
nt
l
og
s
i
ns
te
ad
of
att
ac
he
d
b
y
c
as
e.
T
he
r
ef
ore,
i
n
the
s
ec
on
d
s
tep
,
th
e
c
as
e
s
eq
u
en
c
e
i
s
di
s
c
ov
ered.
If
a
no
de
i
s
the
s
uc
c
es
s
or
of
the
oth
er
no
de
,
a
N
E
X
T
l
ab
el
i
s
pu
t
on
th
e
r
e
l
at
i
o
n
b
et
w
e
en
the
m
.
F
i
gu
r
e
3
de
pi
c
ts
t
he
c
as
e
s
eq
ue
n
c
e
c
on
s
tr
uc
ti
o
n
qu
er
y
.
F
i
gu
r
e
3
.
C
as
e
s
eq
u
en
c
e
d
i
s
c
ov
er
y
q
ue
r
y
A
f
ter
c
on
ne
c
ti
n
g
the
no
de
s
,
the
ne
x
t
s
tep
i
s
pe
r
f
or
m
i
ng
tr
ac
e
c
l
us
teri
ng
.
T
he
pu
r
p
os
e
o
f
tr
ac
e
c
l
us
ter
i
ng
i
s
to
gr
ou
p
no
d
e
s
eq
ue
nc
es
ha
v
i
ng
t
he
s
am
e
tr
ac
e
a
n
d
c
al
c
u
l
at
e
i
ts
oc
c
urr
en
c
es
wi
th
i
n
ea
c
h
c
l
us
ter. T
he
qu
er
y
of
trac
e c
l
us
teri
ng
i
n Ne
o4
J
i
s
s
ho
wn
i
n Fi
gu
r
e
4.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NIK
A
IS
S
N: 1
69
3
-
6
93
0
A
no
ma
l
y
Dete
c
t
i
on
B
as
ed
on
Co
ntro
l
-
fl
ow
P
att
ern
of
P
ara
l
l
e
l
B
us
i
ne
s
s
...
(
He
nd
r
a Da
r
m
awan
)
2811
F
i
gu
r
e
4.
T
r
ac
e
c
l
us
teri
ng
q
ue
r
y
A
f
ter
tr
ac
es
are
c
l
us
tered
,
da
ta
f
i
l
t
erin
g
i
s
c
on
du
c
ted
.
T
he
ob
j
ec
ti
v
e
of
thi
s
s
te
p
i
s
to
r
em
ov
e
the
an
om
al
y
d
ata
.
I
n
thi
s
c
as
e,
i
t
i
s
the
t
wo
l
o
w
es
t
f
r
eq
ue
nc
i
es
of
oc
c
urr
en
c
e
of
the
tr
ac
e
.
B
ec
au
s
e
the
Ne
o4
J
c
an
no
t
s
av
e
t
he
r
es
u
l
t
of
a
qu
er
y
,
a
uto
m
ati
c
de
l
et
i
o
n
c
an
no
t
b
e
d
on
e
us
i
ng
Ne
o4
J
.
T
o
ov
erc
om
e
thi
s
probl
em
da
ta
de
l
et
i
on
i
s
pe
r
f
or
m
ed
b
y
r
ea
d
i
ng
th
e
tr
ac
e
l
en
gt
h
i
n
ea
c
h
c
as
e
I
D
a
nd
th
en
t
he
t
w
o
l
o
w
e
s
t
f
r
eq
ue
nc
i
es
are
de
l
ete
d.
T
he
qu
er
y
r
e
m
ov
e
s
the
da
ta
ba
s
ed
on
th
e
r
es
ul
ts
of
the
tr
ac
e
c
l
us
teri
ng
b
y
c
h
an
g
i
ng
F
r
eq
ue
nc
i
es
v
ar
i
ab
l
e
o
n
w
he
r
e
c
l
au
s
e
as
de
p
i
c
ted
i
n F
i
g
ure 5
.
F
i
gu
r
e
5
. D
ata
f
i
l
teri
ng
c
i
ph
er
q
ue
r
y
T
o
ob
tai
n
a
c
orr
ec
t
bu
s
i
n
es
s
proc
es
s
m
od
el
i
n
t
he
ne
x
t
s
tag
e
,
a
l
l
no
de
s
h
av
i
ng
th
e
s
a
m
e
ac
ti
v
i
t
y
na
m
e
m
us
t
be
m
erged,
an
d
t
he
ex
i
s
ti
ng
c
as
e
s
eq
ue
nc
e
r
e
l
a
t
i
on
s
m
us
t
be
r
em
ov
ed
.
A
f
ter
bo
t
h
proc
e
s
s
es
as
m
en
ti
on
e
d
ab
ov
e,
i
t
f
ol
l
o
w
s
w
i
t
h
d
el
e
ti
n
g
C
as
e
ID
i
ns
i
d
e
Cas
eA
c
ti
v
i
t
y
l
ab
el
s
i
nc
e
i
t
l
ea
ds
to
pro
du
c
i
n
g
w
r
on
g
ev
en
t
l
og
s
e
qu
e
nc
e
r
el
at
i
on
s
.
F
i
gu
r
e
6
di
s
pl
a
y
s
a Ne
o4
J
q
ue
r
y
to
c
on
du
c
t t
he
m
ergi
ng
an
d
d
el
et
i
o
n p
r
oc
es
s
es
.
F
i
gu
r
e
6
. N
od
e
m
ergi
n
g a
n
d
r
el
a
ti
o
n
d
el
ete
q
ue
r
y
2.2.
Bu
sines
s P
r
o
ce
ss
Di
sc
o
v
er
y
S
t
age
T
he
s
ec
on
d
s
t
ag
e
of
th
e
p
r
op
os
ed
m
eth
od
i
s
d
i
s
c
ov
e
r
i
ng
the
b
us
i
ne
s
s
proc
es
s
as
the
c
on
ti
n
ua
t
i
on
of
the
pr
ev
i
o
us
s
tag
e.
F
i
r
s
t,
an
ev
en
t
l
og
s
e
qu
e
nc
e
n
ee
ds
t
o
be
c
r
ea
te
d
b
y
c
on
ne
c
ti
ng
al
l
no
de
s
ba
s
e
d
on
t
he
ord
er
of
ac
ti
v
i
t
i
es
.
T
he
Neo4
J
qu
er
y
of
ev
en
t
l
og
s
e
qu
e
nc
e
di
s
c
ov
er
y
pres
en
ted
i
n
F
i
g
ure
7
i
s
s
l
i
gh
t
l
y
s
i
m
i
l
ar
to
the
on
e
us
e
d
f
or
c
as
e
s
eq
ue
nc
e
di
s
c
ov
er
y
us
ed
i
n t
h
e p
r
e
v
i
ou
s
s
ta
ge
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N: 16
93
-
6
93
0
T
E
L
KO
M
NIK
A
V
ol
.
16
,
No
.
6,
D
ec
em
be
r
20
18
:
280
9
-
281
6
2812
F
i
gu
r
e
7
.
E
v
en
t
l
o
g
d
i
s
c
ov
e
r
y
s
eq
u
en
c
e
q
ue
r
y
T
he
s
ec
on
d
s
tep
of
th
i
s
s
tag
e
i
s
d
i
s
c
ov
erin
g
the
c
o
ntro
l
-
f
l
ow
pa
t
tern
.
It
c
on
s
i
s
t
s
of
A
ND,
X
O
R,
O
R
S
P
LIT
,
an
d
O
R
J
O
IN
r
el
at
i
on
s
.
T
he
i
nc
om
i
ng
a
nd
ou
t
go
i
ng
ed
g
es
de
term
i
ne
the
s
e
r
el
at
i
on
s
.
F
i
gu
r
e
8
s
ho
w
s
t
he
ad
de
d
A
ND
a
nd
X
O
R
c
on
ne
c
t
i
on
qu
er
i
es
al
on
g
wi
th
the
ex
i
s
ti
n
g
O
R r
el
at
i
o
n q
u
erie
s
. T
hi
s
s
t
ep
pro
du
c
es
th
e f
i
na
l
gra
ph
m
od
el
.
F
i
gu
r
e
8
. C
on
tr
o
l
-
f
l
o
w
p
att
e
r
n
d
i
s
c
ov
er
y
q
ue
r
y
3.
Re
sult
s
a
n
d
A
n
al
y
s
is
T
he
ev
en
t
l
o
g
m
us
t
c
on
tai
n
Cas
e
_ID
as
a
proc
es
s
i
ns
tan
c
e
,
ac
ti
v
i
ti
es
na
m
e,
s
tart
ti
m
es
ta
m
p
an
d
en
d
t
i
m
es
t
am
p.
T
ab
l
e
1
de
s
c
r
i
b
es
th
e
c
on
ten
t
of
ev
e
nt
l
og
s
of
i
nte
r
l
i
brar
y
l
o
an
ap
p
l
i
c
at
i
on
us
e
d
i
n
th
i
s
r
es
ea
r
c
h.
In
tot
a
l
,
i
t
c
on
t
ai
ns
74
c
as
es
an
d
12
93
ac
t
i
v
i
ti
es
whi
c
h
are
di
v
i
d
ed
i
nt
o 3
1
t
y
p
es
of
ac
ti
v
i
t
i
es
.
T
ab
l
e
1
.
T
he
Fr
ag
m
en
t o
f
I
nte
r
l
i
brar
y
Lo
an
E
v
en
t
Lo
g
C
a
s
e
_
I
D
A
c
t
iv
it
y
S
t
a
r
t
_
T
i
m
e
s
t
a
m
p
E
n
d
_
Ti
m
e
s
t
a
mp
PP1
R
e
q
u
e
s
t
mat
e
r
ial
02
-
05
-
1
7
0
7
:
0
5
02
-
05
-
1
7
0
7
:
1
0
PP1
C
h
e
c
k
a
v
a
il
a
b
i
li
t
y
02
-
05
-
1
7
0
7
:
1
0
02
-
05
-
1
7
1
0
:
1
5
PP1
C
o
n
f
ir
m
m
a
t
e
r
ial
a
v
a
il
a
b
i
li
t
y
02
-
05
-
1
7
1
0
:
1
5
02
-
05
-
1
7
1
0
:
2
0
PP1
S
e
n
d
m
a
t
e
r
ial
02
-
05
-
1
7
1
0
:
2
0
02
-
05
-
1
7
1
4
:
1
5
PP1
C
o
n
f
ir
m
m
a
t
e
r
ial
a
r
r
iv
a
l
02
-
05
-
1
7
1
4
:
1
5
02
-
05
-
1
7
1
4
:
2
0
PP1
Tak
e
mat
e
r
ial
02
-
05
-
1
7
1
4
:
2
0
03
-
05
-
1
7
1
4
:
3
5
PP1
U
s
e
m
a
t
e
r
ial
03
-
05
-
1
7
1
4
:
3
5
15
-
05
-
1
7
0
7
:
1
5
PP1
R
e
t
u
r
n
m
a
t
e
r
ial
b
e
f
o
r
e
d
u
e
t
i
m
e
15
-
05
-
1
7
0
7
:
1
5
15
-
05
-
1
7
0
7
:
3
0
PP1
S
e
n
d
b
a
c
k
mat
e
r
ial
15
-
05
-
1
7
0
7
:
3
0
15
-
05
-
1
7
1
0
:
0
7
PP1
C
o
n
f
ir
m
m
a
t
e
r
ial
s
e
n
d
b
a
c
k
t
h
e
a
r
r
iv
a
l
15
-
05
-
1
7
1
0
:
0
7
15
-
05
-
1
7
1
0
:
1
2
PP1
C
h
e
c
k
mat
e
r
ial
c
o
n
d
it
ion
15
-
05
-
1
7
1
0
:
1
2
15
-
05
-
1
7
1
3
:
1
2
PP1
C
o
n
f
ir
m
m
a
t
e
r
ial
in
g
o
o
d
c
o
n
d
it
ion
15
-
05
-
1
7
1
3
:
1
2
15
-
05
-
1
7
1
3
:
1
7
PP1
C
o
n
f
ir
m
loa
n
f
ini
s
h
15
-
05
-
1
7
1
3
:
1
7
15
-
05
-
17
3
:
2
5
F
i
gu
r
e
9
de
pi
c
ts
th
e
i
m
pl
e
m
en
tat
i
on
r
es
u
l
t
of
l
o
ad
da
ta
an
d
c
as
e
s
eq
ue
nc
e
di
s
c
ov
e
r
y
.
T
hi
s
r
es
ea
r
c
h
l
i
m
i
ts
the
c
as
e
s
eq
ue
nc
e
d
i
s
c
o
v
er
y
on
tw
o
Cas
e
I
Ds
,
i
.
e.,
P
P
1
an
d
P
P
2.
T
he
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NIK
A
IS
S
N: 1
69
3
-
6
93
0
A
no
ma
l
y
Dete
c
t
i
on
B
as
ed
on
Co
ntro
l
-
fl
ow
P
att
ern
of
P
ara
l
l
e
l
B
us
i
ne
s
s
...
(
He
nd
r
a Da
r
m
awan
)
2813
r
es
ul
t
of
tr
ac
e
c
l
us
ter
i
ng
whi
c
h
i
s
th
e
i
s
th
e
f
r
eq
ue
nc
y
of
oc
c
urr
en
c
es
of
tr
ac
es
is
s
ho
wn
i
n
T
ab
l
e 2
.
I
n
F
i
gu
r
e
9
, th
e n
a
m
e
s
of
ac
ti
v
i
t
i
es
are
al
i
as
es
of
th
e rea
l
n
am
es
.
F
i
gu
r
e
9
.
D
ata
l
oa
di
ng
an
d
c
as
e s
eq
ue
nc
e d
i
s
c
ov
er
y
l
i
m
i
ted
on
P
P
1
an
d P
P
2
c
as
e
I
Ds
T
ab
l
e
2
. T
r
ac
e Cl
us
teri
n
g I
m
pl
e
m
en
tat
i
o
n
u
s
i
ng
N
eo
4
J
Fr
e
q
u
e
n
c
ies
Tr
a
c
e
_
Ty
p
e
C
a
s
e
I
d
Tr
a
c
e
_
L
e
n
g
t
h
12
R
M
,
C
A
,
C
M
A
,
S
M
,
M
A
,
TM
,
U
M
,
U
M
,
S
B
M
,
M
S
B
,
C
M
C
,
GC
,
L
F
P
P
3
4
,
P
P
1
6
,
P
P
1
,
P
P
4
8
,
P
P
1
5
,
P
P
1
8
,
P
P
6
8
,
P
P
5
0
,
P
P
4
7
,
P
P
2
0
,
P
P
6
7
,
P
P
5
2
13
8
R
M
,
C
A
,
C
M
A
,
S
M
,
M
A
,
TM
,
U
M
,
R
M
A
,
S
B
M
,
M
S
B
,
C
M
C
,
GC
,
D
F,
S
FD
,
A
P
F,
P
F,
P
A
Y
,
L
F
P
P
4
9
,
P
P
5
8
,
P
P
7
,
P
P
1
9
,
P
P
3
9
,
P
P
5
1
,
P
P
2
6
,
P
P
1
7
18
6
R
M
,
C
A
,
C
M
A
,
S
M
,
M
A
,
TM
,
U
M
,
R
L
E
,
L
E
A
,
A
L
E
,
U
M
,
R
L
E
,
L
E
A
,
D
L
E
,
U
M
,
S
B
M
,
M
S
B
,
C
M
C
,
GC
,
L
F
P
P
2
5
,
P
P
5
7
,
P
P
6
9
,
P
P
6
,
P
P
3
8
,
P
P
7
0
20
6
R
M
,
C
A
,
C
M
A
,
S
M
,
M
A
,
TM
,
U
M
,
R
L
E
,
L
E
A
,
D
L
E
,
U
M
,
S
B
M
,
M
S
B
,
C
M
C
,
B
C
,
D
F,
S
FD
,
A
P
F
,
P
F
,
P
A
Y
,
L
F
P
P
3
1
,
P
P
1
3
,
P
P
4
5
,
P
P
1
2
,
P
P
4
4
,
P
P
3
2
21
5
R
M
,
C
A
,
C
M
A
,
S
M
,
M
A
,
TM
,
U
M
,
R
L
E
,
L
E
A
,
A
L
E
,
U
M
,
R
M
A
,
S
B
M
,
M
S
B
,
C
M
C
,
GC
,
D
F,
S
FD
,
A
P
F,
P
F,
P
A
Y
,
L
F
P
P
1
0
,
P
P
6
3
,
P
P
6
6
,
P
P
4
2
,
P
P
2
9
22
5
R
M
,
C
A
,
C
M
A
,
S
M
,
M
A
,
TM
,
U
M
,
R
L
E
,
L
E
A
,
A
L
E
,
U
M
,
R
L
E
,
L
E
A
,
A
L
E
,
U
M
,
U
M
,
S
B
M
,
M
S
B
,
C
M
C
,
GC
,
L
F
P
P
4
6
,
P
P
3
3
,
P
P
6
2
,
P
P
6
5
,
P
P
1
4
21
5
R
M
,
C
A
,
C
M
A
,
S
M
,
M
A
,
TM
,
U
M
,
R
L
E
,
L
E
A
,
A
L
E
,
U
M
,
U
M
,
S
B
M
,
M
S
B
,
C
M
C
,
B
C
,
D
F
,
S
FD
,
A
P
F,
P
F,
P
A
Y
,
L
F
P
P
6
1
,
P
P
4
3
,
P
P
6
4
,
P
P
3
0
,
P
P
1
1
22
4
R
M
,
C
A
,
C
M
A
,
S
M
,
M
A
,
E
X
T,
S
B
M
,
M
S
B
,
L
C
P
P
3
7
,
P
P
2
4
,
P
P
5
6
,
P
P
5
16
4
R
M
,
C
A
,
C
M
A
,
S
M
,
M
A
,
TM
,
U
M
,
R
L
E
,
L
E
A
,
D
L
E
,
U
M
,
S
B
M
,
M
S
B
,
C
M
C
,
GC
,
L
F
P
P
5
5
,
P
P
4
,
P
P
3
6
,
P
P
2
3
18
4
R
M
,
C
A
,
C
M
A
,
S
M
,
M
A
,
TM
,
U
M
,
R
L
E
,
L
E
A
,
A
L
E
,
U
M
,
U
M
,
S
B
M
,
M
S
B
,
C
M
C
,
GC
,
L
F
P
P
2
2
,
P
P
3
,
P
P
5
4
,
P
P
3
5
17
4
R
M
,
C
A
,
C
M
A
,
S
M
,
M
A
,
TM
,
U
M
,
R
M
A
,
S
B
M
,
M
S
B
,
C
M
C
,
B
C
,
D
F
,
S
FD
,
A
P
F,
P
F,
P
A
Y
,
L
F
P
P
2
8
,
P
P
9
,
P
P
6
0
,
P
P
4
1
9
4
R
M
,
C
A
,
C
M
A
,
S
M
,
M
A
,
TM
,
U
M
,
U
M
,
S
B
M
,
M
S
B
,
C
M
C
,
B
C
,
D
F
,
S
FD
,
A
P
F,
P
F,
P
A
Y
,
L
F
P
P
4
0
,
P
P
2
7
,
P
P
5
9
,
P
P
8
18
3
R
M
,
C
A
,
C
M
U
,
D
R
,
L
C
P
P
2
1
,
P
P
5
3
,
P
P
2
5
2
R
M
,
P
M
R
,
B
S
,
U
B
L
,
DR
,
LC
P
P
7
2
,
P
P
7
1
6
2
R
M
,
B
S
,
P
M
R
,
U
B
L
,
DR
,
LC
P
P
7
3
,
P
P
7
4
6
Ma
pp
i
n
g
of
al
i
as
es
an
d
r
ea
l
na
m
es
c
an
be
s
ee
n
on
T
ab
l
e
3
.
T
ab
l
e
3
c
on
ta
i
ns
ac
ti
v
i
t
y
na
m
es
as
the
r
ea
l
n
am
es
of
ac
ti
v
i
ti
es
i
n
th
e
e
v
en
t
l
o
g,
an
d
al
i
as
es
are
us
ed
t
o
s
i
m
pl
i
f
y
th
os
e
na
m
es
of
ac
ti
v
i
ti
es
.
B
as
ed
on
t
he
r
es
u
l
ts
of
tr
ac
e
c
l
us
teri
n
g
qu
er
y
i
n
T
ab
l
e
2
,
th
e
t
w
o
s
m
al
l
es
t
v
a
l
u
e
s
are
two
a
nd
t
hree
whi
c
h
ap
p
ea
r
s
i
n
((
P
P
72
,
P
P
71
)
,
(
P
P
73
,
P
P
74
)
)
a
nd
(
P
P
2
,
P
P
21
,
P
P
53
)
,
r
es
pe
c
ti
v
el
y
.
B
y
do
i
n
g
No
d
e
Me
r
g
i
n
g
,
th
e
r
el
at
i
on
s
h
i
p
be
t
w
e
en
no
d
es
is
s
tac
k
ed
b
y
t
he
n
um
be
r
of
oc
c
urr
en
c
es
of
the
r
el
a
ti
on
.
In
th
i
s
pa
p
er,
the
n
u
m
be
r
o
f
NE
X
T
r
el
ati
on
be
twee
n
ac
ti
v
i
t
y
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N: 16
93
-
6
93
0
T
E
L
KO
M
NIK
A
V
ol
.
16
,
No
.
6,
D
ec
em
be
r
20
18
:
280
9
-
281
6
2814
Requ
es
t
m
ate
r
i
al
an
d
Ch
e
c
k
av
a
i
l
a
bi
l
i
t
y
ac
ti
v
i
t
y
i
s
7
0
r
el
ati
on
s
.
T
he
r
ef
ore,
i
t
i
s
ne
c
es
s
ar
y
t
o
de
l
ete
N
E
X
T
r
el
ati
on
an
d
t
he
n
r
ed
o
s
eq
u
en
c
e
d
i
s
c
ov
er
y
to
f
orm
the
c
orr
ec
t
bu
s
i
ne
s
s
proc
es
s
m
od
el
.
A
f
ter
de
l
eti
ng
t
he
N
E
X
T
r
el
ati
on
,
Cas
e
I
D
on
t
he
Cas
e
A
c
ti
v
i
t
y
l
ab
el
m
us
t
be
r
em
ov
ed
to
ge
ne
r
ate
t
he
c
orr
ec
t b
us
i
n
e
s
s
proc
es
s
m
od
el
i
n
th
e
e
v
en
t
l
og
s
e
qu
e
nc
e d
i
s
c
ov
er
y
s
tep
.
T
ab
l
e
3
.
A
l
i
as
of
In
t
erl
i
bra
r
y
Lo
a
n E
v
e
nt
L
og
A
c
ti
v
i
t
i
es
A
li
a
s
A
c
t
iv
it
y
N
a
m
e
A
li
a
s
A
c
t
iv
it
y
N
a
m
e
RM
R
e
q
u
e
s
t
mat
e
r
ial
DR
D
e
n
y
r
e
q
u
e
s
t
CA
C
h
e
c
k
a
v
a
il
a
b
i
li
t
y
LC
C
o
n
f
ir
m
loa
n
c
a
n
c
e
l
C
M
A
C
o
n
f
ir
m
m
a
t
e
r
ial
a
v
a
il
a
b
i
li
t
y
R
L
E
R
e
q
u
e
s
t
len
d
ing
e
x
t
e
n
s
ion
SM
S
e
n
d
m
a
t
e
r
ial
L
E
A
A
s
k
len
d
ing
e
x
t
e
n
s
ion
a
v
a
il
a
b
il
i
t
y
MA
C
o
n
f
ir
m
m
a
t
e
r
ial
a
r
r
iv
a
l
A
L
E
A
ll
o
w
len
d
ing
e
x
t
e
n
s
ion
TM
Tak
e
mat
e
r
ial
D
L
E
D
e
n
y
len
d
ing
e
x
t
e
n
s
ion
UM
U
s
e
m
a
t
e
r
ial
E
X
T
R
e
q
u
e
s
t
e
d
m
a
t
e
r
ial
p
i
c
k
u
p
e
x
c
e
e
d
t
h
e
t
i
m
e
li
m
i
t
R
M
B
R
e
t
u
r
n
m
a
t
e
r
ial
b
e
f
o
r
e
d
u
e
t
i
m
e
R
M
A
R
e
t
u
r
n
m
a
t
e
r
ial
a
f
t
e
r
d
u
e
t
i
m
e
SBM
S
e
n
d
b
a
c
k
mat
e
r
ial
DF
D
e
t
e
r
mi
n
e
t
h
e
a
m
o
u
n
t
s
o
f
f
ine
s
t
o
b
e
p
a
id
M
S
B
C
o
n
f
ir
m
m
a
t
e
r
ial
s
e
n
d
b
a
c
k
t
h
e
a
r
r
iv
a
l
S
FD
S
e
n
d
d
e
t
a
il
o
f
t
h
e
f
ine
s
C
M
C
C
h
e
c
k
mat
e
r
ial
c
o
n
d
it
ion
APF
A
s
k
t
h
e
u
s
e
r
t
o
p
a
y
t
h
e
f
ine
GC
C
o
n
f
ir
m
m
a
t
e
r
ial
in
g
o
o
d
c
o
n
d
it
ion
PF
P
a
y
t
h
e
f
ine
s
a
n
d
c
o
n
f
ir
m
p
a
y
m
e
n
t
LF
C
o
n
f
ir
m
loa
n
f
ini
s
h
P
A
Y
C
o
n
f
ir
m
p
a
y
m
e
n
t
C
M
U
C
o
n
f
ir
m
m
a
t
e
r
ial
u
n
a
v
a
il
a
b
le
BC
C
o
n
f
ir
m
m
a
t
e
r
ial
in
a
b
a
d
c
o
n
d
i
t
io
n
BS
C
h
e
c
k
u
s
e
r
b
la
c
k
li
s
t
s
t
a
t
u
s
P
M
R
C
h
e
c
k
u
s
e
r
p
a
s
t
m
a
t
e
r
ial
r
e
q
u
e
s
t
U
B
L
A
u
s
e
r
i
s
b
la
c
k
li
s
t
e
d
,
o
r
u
s
e
r
h
a
s
n
o
t
r
e
t
u
r
n
e
d
m
a
t
e
r
ial
a
f
t
e
r
t
h
e
t
i
m
e
li
m
it
T
he
r
es
ul
t
of
ev
en
t
l
og
s
eq
ue
nc
e
di
s
c
o
v
er
y
i
s
s
ho
wn
i
n
F
i
gu
r
e
10
.
T
he
ev
en
t
l
og
s
eq
ue
nc
e
d
i
s
c
ov
er
y
c
o
nn
e
c
ts
en
ti
r
e
no
de
s
t
o
a
n
e
v
e
nt
l
og
i
n
th
e
ord
er
of
e
v
en
t
s
.
T
he
r
el
at
i
on
c
r
ea
ted
at
t
hi
s
s
tag
e
de
s
c
r
i
be
s
the
o
v
era
l
l
ac
ti
v
i
t
i
es
f
r
om
s
tart
to
f
i
ni
s
h
an
d
bran
c
hi
ng
ac
t
i
v
i
t
y
i
f
the
r
e
are
branc
h
es
i
n
the
ev
e
nt
l
o
g.
In
F
i
gu
r
e
10
,
th
e
ac
ti
v
i
ti
es
tha
t
on
l
y
ex
i
s
t
on
P
P
2,
P
P
21
,
and
P
P
5
3
are
De
n
y
R
eq
u
es
t
an
d
Co
nf
i
r
m
U
na
v
a
i
l
ab
l
e
M
ate
r
i
a
l
N
ot
F
ou
nd
.
T
hi
s
i
nd
i
c
ate
s
tha
t
an
om
al
y
da
ta
de
l
et
i
o
n
h
as
be
en
s
uc
c
es
s
f
ul
l
y
pe
r
f
or
m
ed
.
T
he
no
de
s
f
orm
ed
i
n
F
i
g
ure
11
are
2
6
no
de
s
,
i
n c
o
ntras
t t
o b
ef
ore
th
e
de
l
eti
on
of
an
om
al
ou
s
da
ta
the
nu
m
be
r
of
no
de
s
i
s
31
.
F
i
gu
r
e
1
. Fi
l
ter
ed
g
r
ap
h
m
od
el
wi
t
h
s
eq
u
en
c
e
r
e
l
at
i
o
n
T
he
r
es
ul
t
of
di
s
c
ov
eri
ng
c
on
tr
ol
-
f
l
o
w
p
att
erns
i
s
s
h
o
w
n
i
n
F
i
g
ure
11
.
I
n
the
f
i
l
t
ered
graph
m
od
el
of
an
i
nte
r
l
i
br
ar
y
l
o
an
,
on
l
y
X
O
R
pa
tt
ern
ex
i
s
t
s
.
F
i
gu
r
e
11
i
s
a
m
od
i
f
i
ed
F
i
gu
r
e
1
0
.
S
om
e
s
eq
ue
nc
e
r
el
a
ti
o
ns
i
s
de
l
ete
d
an
d
r
ep
l
ac
e
d
b
y
X
O
RS
P
LIT
an
d
X
O
RJ
O
IN
as
the
r
es
ul
ts
o
f
c
on
tr
ol
-
f
l
o
w
pa
tt
ern d
i
s
c
ov
e
r
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NIK
A
IS
S
N: 1
69
3
-
6
93
0
A
no
ma
l
y
Dete
c
t
i
on
B
as
ed
on
Co
ntro
l
-
fl
ow
P
att
ern
of
P
ara
l
l
e
l
B
us
i
ne
s
s
...
(
He
nd
r
a Da
r
m
awan
)
2815
F
i
gu
r
e
2
. Fi
l
ter
ed
g
r
ap
h
m
od
el
wi
t
h
c
on
tr
o
l
-
f
l
o
w
p
att
ern
T
he
r
es
ul
t
of
m
od
el
i
n
g
th
e
ev
e
nt
l
og
wi
tho
ut
d
ata
f
i
l
t
erin
g
i
s
s
ho
w
n
i
n
F
i
gu
r
e
1
2
.
T
he
r
e
are f
i
v
e
m
ore no
de
s
or
ac
ti
v
i
t
i
es
i
n
F
i
gu
r
e
12
c
om
pa
r
ed
to
F
i
gu
r
e 1
1
.
S
om
e o
f
th
e
de
l
ete
d n
od
es
c
on
tai
n
A
ND
S
P
LIT
an
d
A
NDJ
O
IN
r
el
at
i
o
n,
an
d
s
om
e
no
de
s
c
on
tai
ne
d
X
O
RS
P
LIT
an
d
X
O
RJ
O
IN.
F
i
gu
r
e
3
. U
nf
i
l
tere
d g
r
a
ph
m
od
el
wi
th
a
c
on
tr
ol
-
f
l
o
w
p
att
ern
A
f
ter
the
graph
m
od
el
h
as
be
en
f
or
m
ed
,
the
l
as
t
s
te
p
i
s
to
c
om
pa
r
e
bo
t
h
gra
ph
f
i
tne
s
s
an
d
prec
i
s
i
on
.
F
i
tn
es
s
an
d
prec
i
s
i
o
n
are
o
ne
of
th
e
m
ea
s
ure
m
en
t
as
pe
c
ts
to
m
ea
s
ure
the
qu
a
l
i
t
y
of
the
bu
s
i
ne
s
s
pr
oc
es
s
m
od
el
.
F
i
tn
es
s
m
ea
s
ures
ho
w
m
an
y
proc
es
s
e
s
tha
t
c
a
n
b
e
de
p
i
c
ted
i
n
th
e
m
od
el
.
Me
a
n
w
hi
l
e,
prec
i
s
i
on
es
t
i
m
ate
s
ho
w
m
an
y
tr
ac
es
i
n
th
e
e
v
en
t
l
o
g
c
an
be
de
p
i
c
ted
i
n
th
e
m
od
el
.
In
(
1),
the
nu
m
be
r
o
f
c
ap
tured
c
as
es
i
n
t
he
m
od
el
i
s
s
tored
i
n
v
ari
ab
l
e
(
_
)
,
whi
l
e
t
he
n
um
be
r
of
c
as
es
i
n
the
e
v
e
nt
l
o
g
i
s
s
tore
d
i
n
v
ari
ab
l
e
(
_
_
_
)
.
In
(
2),
the
n
um
be
r
of
c
ap
ture
d
tr
ac
es
i
n
the
m
od
el
i
s
s
tore
d
i
n
v
ari
ab
l
e
(
)
,
w
h
i
l
e
the
nu
m
be
r
of
tr
ac
es
i
n
the
e
v
en
t
l
og
i
s
s
tored
i
n
v
ari
ab
l
e
(
_
_
_
)
.
(
)
=
(
_
)
(
_
_
_
)
⁄
(
1)
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N: 16
93
-
6
93
0
T
E
L
KO
M
NIK
A
V
ol
.
16
,
No
.
6,
D
ec
em
be
r
20
18
:
280
9
-
281
6
2816
(
)
=
(
_
)
(
_
_
_
)
⁄
(
2)
B
oth
f
i
l
tere
d
an
d
un
f
i
l
tered
m
od
el
ha
v
e
th
e
s
am
e
f
i
tne
s
s
v
al
u
es
of
1.
B
ut
t
he
pre
c
i
s
i
on
v
a
l
ue
of
the
f
i
l
tere
d
m
od
el
i
s
hi
gh
er
tha
n
t
he
prec
i
s
i
o
n
v
a
l
ue
of
the
un
f
i
l
tere
d
m
od
e
l
.
P
r
ec
i
s
i
on
v
a
l
ue
of
the
f
i
l
t
ered
m
od
el
i
s
0.81
.
Me
an
w
h
i
l
e,
th
e
prec
i
s
i
o
n
v
a
l
u
e
of
the
un
f
i
l
t
ered
m
od
el
i
s
0.
6
4
.
4.
Co
n
clus
ion
T
hi
s
r
es
ea
r
c
h
prop
os
ed
a
m
eth
od
ol
o
g
y
to
d
i
s
c
ov
er
a
be
tte
r
bu
s
i
ne
s
s
proc
es
s
m
od
e
l
b
y
c
on
du
c
ti
ng
tr
ac
e
c
l
us
ter
i
ng
.
T
he
ev
en
t
l
og
i
s
i
m
po
r
ted
an
d
m
od
el
ed
to
f
orm
a
bu
s
i
ne
s
s
proc
es
s
m
od
el
.
T
he
r
e
are
s
ev
eral
tr
ac
es
c
ap
tured
i
n
th
e
m
od
el
.
S
om
e
o
f
the
tr
ac
es
ap
pe
a
r
ed
wi
th
a
l
o
w
-
f
r
eq
ue
nc
y
v
al
ue
.
T
he
l
o
w
-
f
r
eq
ue
nc
y
v
al
u
e
i
s
c
on
s
i
d
ered
as
an
om
al
ou
s
.
T
he
n,
d
ata
f
i
l
ter
i
ng
i
s
do
ne
t
o
r
em
ov
e
the
an
om
al
o
us
tr
ac
es
.
A
f
ter
tha
t,
pr
oc
es
s
di
s
c
ov
er
y
i
s
m
ad
e
to
f
or
m
a
ne
w
bu
s
i
n
es
s
proc
es
s
m
od
el
.
T
he
r
es
ul
t
f
ou
nd
i
m
prov
em
e
nt
of
th
e
prec
i
s
i
o
n
v
a
l
ue
of
the
propos
e
d
m
eth
od
.
T
he
r
ef
ore, trac
e c
l
us
teri
ng
an
d d
ata
f
i
l
teri
ng
c
an
f
orm
a
be
tte
r
bu
s
i
ne
s
s
p
r
oc
es
s
m
od
el
.
Ref
er
en
ce
s
[1
]
R
Sa
rn
o
,
KR
S
u
n
g
k
o
n
o
.
Cou
p
l
e
d
Hi
d
d
e
n
M
a
rk
o
v
M
o
d
e
l
fo
r
Pro
c
e
s
s
M
i
n
i
n
g
o
f
I
n
v
i
s
i
b
l
e
Pr
i
m
e
T
a
s
k
s
.
In
te
rn
a
ti
o
n
a
l
Re
v
i
e
w Com
p
u
te
rs
a
n
d
So
ft
war
e
(I
REC
O
S)
.
2
0
1
6
;
1
1
(
6
):
5
3
9
-
5
4
7
.
[2
]
NY
Se
ti
a
w
a
n
,
R
Sa
rn
o
.
M
u
l
ti
-
c
ri
te
r
i
a
Dec
i
s
i
o
n
M
a
k
i
n
g
fo
r
Se
l
e
c
t
i
n
g
Se
m
a
n
t
i
c
W
e
b
Se
rv
i
c
e
Con
s
i
d
e
ri
n
g
Va
ri
a
b
i
l
i
ty
a
n
d
C
o
m
p
l
e
x
i
ty
T
ra
d
e
-
o
ff
.
J
o
u
rn
a
l
o
f
Th
e
o
r
e
ti
c
a
l
a
n
d
Ap
p
l
i
e
d
I
n
fo
rm
a
ti
o
n
Te
c
h
n
o
l
o
gy
.
2
0
1
6
;
8
6
(2
)
:
316
-
326.
[3
]
M
e
k
h
a
l
a
.
A
Rev
i
e
w
Pa
p
e
r
o
n
Pro
c
e
s
s
M
i
n
i
n
g
.
In
te
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
En
g
i
n
e
e
ri
n
g
a
n
d
T
e
c
h
n
i
q
u
e
s
.
2
0
1
5
;
1
(4
)
:
11
-
1
7
.
[4
]
W
M
P
v
a
n
d
e
r
Aa
l
s
t,
T
W
e
i
j
t
e
rs
,
L
M
a
ru
s
te
r
.
W
o
r
k
fl
o
w
M
i
n
i
n
g
:
D
i
s
c
o
v
e
ri
n
g
Pro
c
e
s
s
M
o
d
e
l
s
fr
o
m
Ev
e
n
t
L
o
g
s
.
IEEE
Tr
a
n
s
a
c
t
i
o
n
s
o
n
K
n
o
wle
d
g
e
a
n
d
D
a
ta
En
g
i
n
e
e
r
i
n
g
.
2
0
0
4
;
1
6
(
9
):
1
1
2
8
-
1
1
4
2
.
[5
]
AJ
M
M
W
e
i
j
t
e
rs
,
W
M
P
v
a
n
d
e
r
Aa
l
s
t
,
AAK
M
e
d
e
i
ro
s
.
Pro
c
e
s
s
M
i
n
i
n
g
w
i
th
th
e
Heu
ri
s
ti
c
s
M
i
n
e
r
-
A
l
g
o
ri
th
m
.
Ei
n
d
h
o
v
e
n
:
E
i
n
d
h
o
v
e
n
Un
i
v
e
rs
i
ty
o
f
T
e
c
h
n
o
l
o
g
y
.
2
0
0
6
.
[6
]
R
Sa
rn
o
,
F
Hary
a
d
i
t
a
,
D
Su
n
a
ry
o
n
o
,
A
M
u
n
i
f.
M
o
d
e
l
Dis
c
o
v
e
ry
o
f
Pa
ra
l
l
e
l
Bu
s
i
n
e
s
s
Pro
c
e
s
s
e
s
u
s
i
n
g
M
o
d
i
fi
e
d
Heu
ri
s
ti
c
M
i
n
e
r
.
In
t
e
r
n
a
ti
o
n
a
l
Co
n
fe
r
e
n
c
e
o
n
S
c
i
e
n
c
e
i
n
In
f
o
rm
a
ti
o
n
T
e
c
h
n
o
l
o
g
y
(
ICSIT
e
c
h
).
Y
o
g
y
a
k
a
rta
.
2
0
1
5
:
30
-
3
5
.
[7
]
Herm
a
w
a
n
,
R
Sa
rn
o
.
A
M
o
re
Eff
i
c
i
e
n
t
Det
e
rm
i
n
i
s
ti
c
Al
g
o
ri
th
m
f
o
r
Pro
c
e
s
s
M
o
d
e
l
Dis
c
o
v
e
ry
.
In
te
rn
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
In
n
o
v
a
ti
v
e
Com
p
u
ti
n
g
,
In
fo
rm
a
t
i
o
n
,
a
n
d
Co
n
tro
l
(I
J
ICIC)
.
2
0
1
8
;
1
4
(3
)
:
971
-
9
9
5
.
[8
]
R
Sa
rn
o
,
KR
Su
n
g
k
o
n
o
.
Cou
p
l
e
d
Hid
d
e
n
M
a
rk
o
v
M
o
d
e
l
fo
r
Pro
c
e
s
s
M
i
n
i
n
g
o
f
Non
-
f
re
e
C
h
o
i
c
e
a
n
d
In
v
i
s
i
b
l
e
Pr
i
m
e
T
a
s
k
s
.
Pro
c
e
d
i
a
Co
m
p
u
t
e
r S
c
i
e
n
c
e
.
2
0
1
8
;
1
2
4
:
1
3
4
-
1
4
1
.
[9
]
R
Sa
rn
o
,
KR
Su
n
g
k
o
n
o
.
Hid
d
e
n
M
a
rk
o
v
M
o
d
e
l
fo
r
Pro
c
e
s
s
M
i
n
i
n
g
o
f
Pa
r
a
l
l
e
l
Bu
s
i
n
e
s
s
Pro
c
e
s
s
e
s
.
In
te
rn
a
ti
o
n
a
l
Re
v
i
e
w o
n
Co
m
p
u
te
r a
n
d
So
f
twa
re
(I
REC
O
S)
.
2
0
1
6
;
1
1
(4
)
:
2
9
0
-
3
0
0
.
[1
0
]
AAK
M
e
d
e
i
ro
s
,
W
M
P
v
a
n
d
e
r
Aa
l
s
t,
A
J
M
M
W
e
i
j
te
r
s
.
Wo
r
k
fl
o
w
M
i
n
i
n
g
:
Curr
e
n
t
St
a
tu
s
a
n
d
Fu
t
u
r
e
Dir
e
c
ti
o
n
s
.
O
n
T
h
e
M
o
v
e
to
M
e
a
n
i
n
g
fu
l
I
n
te
rn
e
t
Sy
s
te
m
s
2
0
0
3
:
C
o
o
p
IS,
DO
A,
a
n
d
O
D
BASE.
O
TM
2
0
0
3
.
C
a
ta
n
i
a
.
2
0
0
3
:
3
8
9
-
4
0
6
.
[1
1
]
AAK
M
e
d
e
i
ro
s
,
BF
v
a
n
Don
g
e
n
,
W
M
P
v
a
n
d
e
r
Aa
l
s
t,
A
J
M
M
W
e
i
j
t
e
rs
.
Pr
o
c
e
s
s
M
i
n
i
n
g
:
Ex
te
n
d
i
n
g
th
e
A
l
pha
-
a
l
g
o
r
i
th
m
t
o
M
i
n
e
Sh
o
rt
L
o
o
p
s
.
T
e
c
h
n
i
s
c
h
e
Un
i
v
e
rs
i
te
i
t
Ei
n
d
h
o
v
e
n
.
Ei
n
d
h
o
v
e
n
.
2
0
0
4
.
[1
2
]
L
W
e
n
,
W
M
P
v
a
n
d
e
r
Aa
l
s
t,
J
W
a
n
g
,
J
Su
n
.
M
i
n
i
n
g
Pro
c
e
s
s
M
o
d
e
l
s
w
i
th
Non
-
Fre
e
-
Cho
i
c
e
Con
s
tru
c
t
s
.
Dat
a
M
i
n
i
n
g
a
n
d
K
n
o
wle
d
g
e
Di
s
c
o
v
e
r
y
.
2
0
0
7
;
1
5
(2
):
1
4
5
-
1
8
0
.
[1
3
]
L
W
e
n
,
W
M
P
v
a
n
d
e
r
A
a
l
s
t,
B
Hua
n
g
,
J
Su
n
.
M
i
n
i
n
g
Pro
c
e
s
s
M
o
d
e
l
s
w
i
th
Pri
m
e
I
n
v
i
s
i
b
l
e
T
a
s
k
s
.
Dat
a
&
Kn
o
w
l
e
d
g
e
En
g
i
n
e
e
ri
n
g
.
2
0
1
0
;
6
9
(1
0
)
:
999
-
1
0
2
1
.
[1
4
]
Q
G
u
o
,
L
W
e
n
,
J
W
a
n
g
,
Z
Y
a
n
,
PS
Y
u
.
M
i
n
i
n
g
In
v
i
s
i
b
l
e
Ta
s
k
s
i
n
Non
-
fre
e
-
c
h
o
i
c
e
C
o
n
s
tr
u
c
t
s
.
Bu
s
i
n
e
s
s
Pro
c
e
s
s
M
a
n
a
g
e
m
e
n
t.
BPM
2
0
1
6
.
Rio
d
e
J
a
n
e
i
ro
.
2
0
1
5
:
109
-
1
2
5
.
[1
5
]
I
Rob
i
n
s
o
n
,
J
W
e
b
b
e
r,
E
Ei
fr
e
m
.
G
ra
p
h
D
a
ta
b
a
s
e
:
New
O
p
p
o
rtu
n
i
ti
e
s
fo
r
Con
n
e
c
t
e
d
Dat
a
.
Am
e
r
i
k
a
Se
ri
k
a
t:
O'
Rei
l
l
y
M
e
d
i
a
,
In
c
.
2
0
1
5
.
[1
6
]
R
Sa
rn
o
,
KR
Su
n
g
k
o
n
o
,
R
Se
p
ti
a
r
a
k
h
m
a
n
.
G
ra
p
h
-
Ba
s
e
d
Ap
p
ro
a
c
h
fo
r
M
o
d
e
l
i
n
g
a
n
d
M
a
tc
h
i
n
g
P
a
ra
l
l
e
l
Bu
s
i
n
e
s
s
Pro
c
e
s
s
e
s
.
I
n
te
rn
a
t
i
o
n
a
l
I
n
fo
rm
a
ti
o
n
I
n
s
t
i
tu
te
.
2
0
1
8
;
2
1
(5
):
1
6
0
3
-
1
6
1
4
.
[1
7
]
I
Rob
i
n
s
o
n
,
J
W
e
b
b
e
r
,
E
Ei
fre
m
.
G
ra
p
h
Dat
a
b
a
s
e
s
:
New
O
p
p
o
rtu
n
i
ti
e
s
f
o
r
Con
n
e
c
te
d
Da
ta
,
2
nd
e
d
.
,
O
'
Rei
l
l
y
M
e
d
i
a
,
In
c
.
2
0
1
5
.
[1
8
]
N
Fra
n
c
i
s
,
A
T
a
y
l
o
r,
A
G
re
e
n
,
P
G
u
a
g
l
i
a
rd
o
,
L
L
i
b
k
i
n
,
T
L
i
n
d
a
a
k
e
r,
V
M
a
rs
a
u
l
t,
S
Pl
a
n
ti
k
o
w
,
M
Ry
d
b
e
rg
,
P
Se
l
m
e
r.
C
y
p
h
e
r
:
An
Ev
o
l
v
i
n
g
Q
u
e
r
y
L
a
n
g
u
a
g
e
fo
r
Pro
p
e
rt
y
G
ra
p
h
s
.
SI
G
M
O
D
'
1
8
Pro
c
e
e
d
i
n
g
s
o
f
th
e
2
0
1
8
In
te
rn
a
t
i
o
n
a
l
Con
fe
re
n
c
e
o
n
M
a
n
a
g
e
m
e
n
t
o
f
Dat
a
.
Hou
s
to
n
.
2
0
1
8
:
1433
-
1
4
4
5
.
Evaluation Warning : The document was created with Spire.PDF for Python.