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
.
6
,
Dec
em
b
er
201
7
,
p
p
.
3
7
45
~
3
7
52
I
SS
N:
2
0
8
8
-
8
7
0
8
,
DOI
: 1
0
.
1
1
5
9
1
/i
j
ec
e.
v
7
i6
.
p
p
37
45
-
37
5
2
3745
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
Ev
a
lua
ting Ag
g
rega
te
Func
tions
o
f
I
ceberg
Q
uery
Us
ing
P
rio
rity Ba
sed
B
i
t
m
a
p
I
ndex
ing
St
ra
tegy
K
a
le
Sa
rik
a
P
ra
k
a
s
h
1
,
P.
M.
J
o
e
P
ra
t
ha
p
2
1
,2
De
p
a
rtem
e
n
t
o
f
Co
m
p
u
ter
sc
ien
c
e
a
n
d
En
g
in
e
e
ri
n
g
,
S
t.
P
e
ter’s
In
stit
u
te o
f
Hig
h
e
r
Ed
u
c
a
ti
o
n
a
n
d
R
e
se
a
rc
h
,
S
t.
P
e
ter’s
Un
iv
e
rsity
,
Av
a
d
i
Ch
e
n
n
a
i,
I
n
d
ia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
A
p
r
10
,
2
0
1
7
R
ev
i
s
ed
Sep
8
,
2
0
1
7
A
cc
ep
ted
Sep
29
,
2
0
1
7
Ag
g
re
g
a
te
f
u
n
c
ti
o
n
a
n
d
ice
b
e
rg
q
u
e
ries
a
re
i
m
p
o
rtan
t
a
n
d
c
o
m
m
o
n
in
m
a
n
y
a
p
p
li
c
a
ti
o
n
s
o
f
d
a
ta
w
a
re
h
o
u
se
b
e
c
a
u
se
u
se
rs
a
re
g
e
n
e
ra
ll
y
in
tere
ste
d
in
lo
o
k
in
g
f
o
r
v
a
rian
c
e
o
r
u
n
u
s
u
a
l
p
a
tt
e
rn
s.
No
rm
a
ll
y
,
th
e
n
a
tu
re
o
f
th
e
q
u
e
ries
to
b
e
e
x
e
c
u
ted
o
n
d
a
ta
w
a
re
h
o
u
se
a
re
th
e
q
u
e
ries
w
it
h
a
g
g
re
g
a
t
e
f
u
n
c
ti
o
n
f
o
ll
o
w
e
d
b
y
h
a
v
in
g
c
lau
se
,
th
e
se
ty
p
e
o
f
q
u
e
ries
a
re
k
n
o
w
n
a
s
ice
b
e
rg
q
u
e
ry
.
Esp
e
c
ially
to
h
a
v
e
e
f
f
i
c
ien
t
tec
h
n
iq
u
e
s
f
o
r
p
ro
c
e
ss
in
g
a
g
g
r
e
g
a
te
f
u
n
c
ti
o
n
o
f
ice
b
e
rg
q
u
e
ry
is
v
e
r
y
i
m
p
o
rtan
t
b
e
c
a
u
se
th
e
ir
p
ro
c
e
ss
in
g
c
o
st
is
m
u
c
h
h
ig
h
e
r
th
a
n
t
h
a
t
o
f
th
e
o
th
e
r
b
a
sic
re
latio
n
a
l
o
p
e
ra
ti
o
n
s
su
c
h
a
s
S
E
L
ECT
a
n
d
P
ROJ
ECT
.
P
re
se
n
tl
y
a
v
a
il
a
b
le
ice
b
e
rg
q
u
e
ry
p
ro
c
e
ss
in
g
tec
h
n
iq
u
e
s
f
a
c
e
s
th
e
p
ro
b
lem
o
f
e
m
p
t
y
b
it
w
ise
A
ND
,
OR
X
OR
o
p
e
ra
ti
o
n
a
n
d
re
q
u
ires
m
o
re
I/O
a
c
c
e
s
s
a
n
d
ti
m
e
.
T
o
o
v
e
rc
o
m
e
th
e
se
p
ro
b
lem
s
p
ro
p
o
se
d
re
se
a
r
c
h
p
r
o
v
id
e
s
e
ff
ici
e
n
t
a
lg
o
rit
h
m
to
e
x
e
c
u
te
ic
e
b
e
rg
q
u
e
ries
us
in
g
p
rio
rit
y
b
a
s
e
d
b
it
m
a
p
in
d
e
x
in
g
stra
teg
y
.
P
rio
r
it
y
b
a
se
d
a
p
p
r
o
a
c
h
c
o
n
si
d
e
r
b
i
tm
a
p
v
e
c
to
r
to
b
e
e
x
e
c
u
ted
a
s
p
e
r
th
e
p
ri
o
rit
y
.
In
term
e
d
iate
re
su
lt
s
a
re
e
v
a
lu
a
te
d
to
f
in
d
p
ro
b
a
b
il
it
y
o
f
re
su
lt
.
F
r
u
it
les
s
o
p
e
ra
ti
o
n
s
a
re
id
e
n
ti
f
ied
a
n
d
sk
ip
p
e
d
i
n
a
d
v
a
n
c
e
w
h
ich
h
e
lp
to
re
d
u
c
e
I/O
a
c
c
e
ss
a
n
d
ti
m
e
.
T
i
m
e
a
n
d
it
e
ra
ti
o
n
re
q
u
ired
t
o
p
ro
c
e
ss
q
u
e
ry
is
re
d
u
c
e
d
[4
5
-
5
0
]
%
c
o
m
p
a
re
to
p
re
v
io
u
s
stra
teg
y
.
Ex
p
e
rime
n
tal
re
su
lt
p
ro
v
e
s th
e
su
p
e
rio
rit
y
o
f
p
rio
rt
y
b
a
se
d
a
p
p
ro
a
c
h
c
o
m
p
a
r
e
to
p
re
v
io
u
s b
it
m
a
p
p
ro
c
e
ss
in
g
a
p
p
ro
a
c
h
.
K
ey
w
o
r
d
:
I
ce
b
er
g
q
u
er
y
(
I
B
Q)
B
it
m
ap
i
n
d
ex
(
B
I
)
Ag
g
r
eg
ate
f
u
n
ctio
n
s
L
o
g
ica
l
o
p
er
atio
n
s
Data
w
ar
e
h
o
u
s
e
(
DW
)
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
:
Kale
Sar
ik
a
P
r
ak
as
h
,
Dep
ar
te
m
en
t o
f
C
o
m
p
u
ter
s
ci
en
ce
an
d
en
g
i
n
ee
r
in
g
,
St.P
eter
’
s
I
n
s
tit
u
te
o
f
Hi
g
h
er
E
d
u
ca
tio
n
an
d
R
e
s
ea
r
ch
,
St.
P
eter
’
s
Un
i
v
er
s
i
t
y
,
Av
ad
i,
C
h
e
n
n
ai,
I
n
d
ia.
E
m
ail:
k
alesar
i
k
ap
r
ak
as
h
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
Data
w
ar
eh
o
u
s
e
(
DW
)
is
co
llectio
n
o
f
s
u
b
j
ec
t
o
r
ien
ted
,
i
n
teg
r
ated
,
n
o
n
-
v
o
latile
an
d
ti
m
e
v
ar
ia
n
t
d
ataset
[
1
]
.
A
n
al
y
s
is
o
f
d
ata
f
r
o
m
d
ata
w
ar
e
h
o
u
s
e
is
v
er
y
i
m
p
o
r
tan
t
f
ac
to
r
f
o
r
th
e
d
ec
is
io
n
m
a
k
in
g
i
n
an
y
b
u
s
i
n
ess
o
r
g
a
n
izat
io
n
.
A
s
d
at
a
w
ar
e
h
o
u
s
e
i
s
h
u
g
e
s
o
a
n
al
y
s
i
s
is
al
s
o
co
m
p
lex
b
e
ca
u
s
e
f
o
r
an
al
y
s
i
s
m
u
ltid
ien
s
io
n
l
ap
p
r
o
ac
h
is
r
eq
u
ir
ed
[
2
]
.
A
n
al
y
s
is
o
f
s
u
ch
h
u
g
e
d
atab
ase
is
d
o
n
e
b
y
e
x
ec
u
t
in
g
co
m
p
le
x
q
u
er
ie
s
s
u
c
h
a
s
iceb
er
g
q
u
er
y
(
I
B
Q)
an
d
o
n
li
n
e
An
y
l
y
tical
p
r
o
ce
s
s
i
n
g
f
u
n
ctio
n
s
.
T
h
e
b
asic
o
p
er
ati
o
n
r
eq
u
ir
ed
i
n
D
W
an
al
y
s
is
is
ag
g
r
eg
a
te
f
u
n
ctio
n
s
s
u
c
h
a
s
MI
N,
M
A
X,
SUM,
A
V
G
a
n
d
C
OU
NT
.
Gen
er
all
y
t
h
e
q
u
er
ie
s
to
b
e
ex
ec
u
ted
o
n
DW
ar
e
t
h
e
q
u
er
ies
w
it
h
ag
g
r
eg
ate
f
u
n
ctio
n
f
o
l
lo
w
ed
b
y
H
AVI
NG
a
n
d
G
R
O
UP
B
y
cla
u
s
e,
s
u
c
h
a
q
u
er
y
i
s
k
n
o
w
n
a
s
I
B
Q.
I
t
c
o
n
s
is
ts
o
f
t
h
r
ee
m
ai
n
p
ar
a
m
et
e
r
s
s
u
c
h
as
ag
g
r
eg
ate
f
u
n
ct
io
n
,
H
A
VI
NG
clau
s
e
an
d
GR
OUP
B
Y
clau
s
e
w
h
ic
h
m
ak
e
s
th
e
q
u
er
y
m
o
r
e
co
m
p
l
ex
.
I
n
ad
d
itio
n
to
th
e
co
m
p
lex
it
y
o
f
I
B
Q,
th
e
lar
g
e
v
o
lu
m
e
o
f
d
ata
s
to
r
ed
in
DW
len
g
th
e
n
s
th
e
ti
m
e
n
ee
d
ed
to
r
u
n
q
u
er
ie
s
.
Hen
ce
p
er
f
o
r
m
a
n
ce
o
f
q
u
er
y
in
ter
m
s
o
f
ti
m
e
i
s
m
o
s
t
i
m
p
o
r
ta
n
t
r
eq
u
ir
e
m
e
n
t
o
f
a
n
y
lar
g
e
d
atab
ase
s
y
s
te
m
.
T
h
is
r
esear
ch
f
o
c
u
s
o
n
e
f
f
icie
n
t
ex
e
cu
tio
n
o
f
ag
g
r
eg
ate
f
u
n
ct
io
n
as
it
is
m
ai
n
p
ar
t
o
f
I
B
Q.
A
g
g
r
eg
a
tes
f
u
n
ct
io
n
ca
n
h
av
e
a
s
i
g
n
i
f
ican
t
i
m
p
ac
t
o
n
p
er
f
o
r
m
an
ce
o
f
q
u
er
y
i
n
ter
m
o
f
ti
m
e.
Fo
r
e
f
f
ic
ien
t
ex
ec
u
t
io
n
o
f
a
g
g
r
e
g
ate
f
u
n
ctio
n
w
e
r
eq
u
ir
e
ef
f
icie
n
t a
n
d
f
ast
p
r
o
ce
s
s
in
g
o
f
h
u
g
e
d
ata
[
3
]
.
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
.
6
,
Dec
em
b
er
2
0
1
7
:
3
7
45
–
3
7
52
3746
T
h
e
m
ai
n
o
b
j
ec
tiv
e
o
f
th
is
r
es
ea
r
ch
is
ef
f
icie
n
t
ex
ec
u
tio
n
o
f
ag
g
r
eg
ate
f
u
n
ctio
n
o
f
I
B
Q.
T
o
ac
h
iev
e
th
is
w
e
ar
e
m
a
k
in
g
u
s
e
o
f
p
r
io
r
ity
b
ased
b
it
m
ap
in
d
e
x
in
g
(
B
I
)
s
tr
ateg
y
.
Nu
m
b
er
o
f
r
esea
r
ch
er
s
[
4
]
-
[
8
]
w
o
r
k
to
i
m
p
r
o
v
e
p
er
f
o
r
m
a
n
ce
o
f
I
B
Q.
B
u
t
all
o
f
th
e
m
f
ac
es
t
h
e
p
r
o
b
lem
o
f
e
m
p
t
y
b
it
w
i
s
e
A
N
D
,
OR
XO
R
o
p
er
atio
n
as
w
e
ll
as
f
u
tile
q
u
e
u
e
p
u
s
h
i
n
g
p
r
o
b
lem
.
T
h
is
r
e
s
e
ar
ch
o
v
er
co
m
es
th
e
s
e
p
r
o
b
lem
s
b
y
u
s
i
n
g
p
r
io
r
it
y
b
as
ed
B
I
s
tr
ateg
y
.
T
h
is
s
tr
ate
g
y
f
ir
s
t
an
al
y
s
e
w
h
ich
o
p
er
ati
o
n
to
b
e
p
er
f
o
r
m
a
s
p
er
th
e
e
v
alu
a
tio
n
o
f
q
u
er
y
.
A
cc
o
r
d
in
g
to
e
v
alu
a
tio
n
p
r
o
ce
s
s
it
ar
r
an
g
e
t
h
e
s
eq
u
en
ce
o
f
o
p
er
atio
n
s
to
b
e
p
er
f
o
r
m
.
B
ased
o
n
r
esu
lts
o
f
cu
r
r
en
t
o
p
er
atio
n
it
ch
an
g
e
t
h
e
p
r
io
r
ity
a
n
d
p
er
f
o
r
m
t
h
e
r
e
m
ai
n
in
g
o
p
er
atio
n
.
I
n
th
is
w
a
y
in
b
et
w
ee
n
th
i
s
tech
n
iq
u
e
id
en
ti
f
y
u
s
eles
s
o
p
er
atio
n
in
ad
v
an
ce
a
n
d
s
k
ip
s
u
ch
u
s
ele
s
s
o
p
er
atio
n
s
.
T
h
u
s
b
y
p
er
f
o
r
m
in
g
o
n
l
y
r
eq
u
ir
ed
o
p
e
r
atio
n
s
it
r
ed
u
ce
s
I
/O
ac
ce
s
s
as
w
ell
as
ti
m
e
r
eq
u
ir
ed
to
ex
ec
u
te
I
B
Q.
T
h
is
s
tr
ateg
y
w
o
r
k
o
n
b
it
m
ap
v
ec
to
r
o
f
attr
ib
u
te
as
p
er
q
u
er
y
r
eq
u
ir
e
m
e
n
t.
T
h
e
B
it
m
ap
v
ec
to
r
s
ar
e
in
t
h
e
f
o
r
m
o
f
0
’
s
an
d
1
’
s
a
n
d
p
r
o
p
o
s
ed
s
tr
ateg
y
p
er
f
o
r
m
lo
g
ical
o
p
er
atio
n
s
s
u
c
h
a
s
O
R
,
AND
an
d
XOR
o
n
t
h
i
s
b
it
m
ap
v
ec
to
r
s
.
E
x
ec
u
ti
n
g
b
it
w
i
s
e
o
p
er
atio
n
s
o
n
0
’
s
a
n
d
1
’
s
ar
e
v
er
y
m
u
ch
co
s
t e
f
f
ec
ti
v
e
i
n
ter
m
o
f
I
/O
ac
ce
s
s
a
n
d
ti
m
e.
I
t d
ir
ec
tl
y
h
e
lp
s
to
im
p
r
o
v
e
I
B
Q
p
er
f
o
r
m
a
n
ce
.
Ou
r
ex
p
er
i
m
en
tal
r
esu
lt
p
r
o
v
es
th
at
p
er
f
o
r
m
a
n
ce
o
f
o
u
r
s
tr
ateg
y
i
s
b
etter
th
an
p
r
ev
io
u
s
a
lg
o
r
it
h
m
s
.
I
n
f
u
t
u
r
e
b
y
ex
ten
d
i
n
g
t
h
is
co
n
ce
p
t
o
n
u
n
s
tr
u
c
t
u
r
ed
d
ata
i
t
w
i
ll
b
e
ap
p
licab
le
f
o
r
b
ig
d
ata
an
al
y
s
is
[
9
]
.
2.
RE
VI
E
W
O
F
B
I
,
AG
G
R
E
G
AT
E
F
UNC
T
I
O
N
AND
I
B
Q
P
RO
CE
SS
I
NG
M
E
T
H
O
D
B
it
m
ap
in
d
ex
i
n
g
tech
n
iq
u
e
is
m
o
s
t
s
u
itab
le
an
d
ef
f
icie
n
t
f
o
r
r
ea
d
m
o
s
tl
y
,
ap
p
en
d
o
n
l
y
d
at
a
an
d
lar
g
e
s
ize
d
ataset.
B
I
is
co
m
m
o
n
l
y
u
s
ed
in
th
e
DW
ap
p
licatio
n
.
B
I
s
tr
ateg
y
p
er
f
o
r
m
s
b
etter
th
a
n
t
r
ee
b
ased
in
d
ex
in
g
m
et
h
o
d
s
lik
e
d
if
f
er
en
t
t
y
p
e
o
f
B
T
r
ee
an
d
R
T
r
ee
[
1
0
]
.
B
I
h
as
t
w
o
ad
v
a
n
tag
e
s
f
o
r
u
s
in
g
it
i
n
DW
ar
e
it
av
o
id
s
co
m
p
lete
tab
le
s
ca
n
an
d
s
av
e
s
d
is
k
ac
ce
s
s
[
1
1
]
,
[
1
2
]
.
T
h
is
r
esear
ch
m
ak
e
s
u
s
e
o
f
co
m
p
r
ess
ed
B
I
c
o
n
ce
p
t
w
h
ic
h
s
av
e
s
t
h
e
m
e
m
o
r
y
a
n
d
s
h
o
w
s
t
h
e
e
f
f
ec
ti
v
en
e
s
s
o
f
B
I
f
o
r
I
B
Q
ev
al
u
atio
n
[
4
]
.
B
I
p
er
f
o
r
m
s
e
f
f
ec
t
iv
el
y
as
it
w
o
r
k
s
o
n
i
n
d
ex
lev
el
r
ath
er
o
n
o
r
ig
i
n
al
tab
le.
T
h
is
f
ea
t
u
r
e
h
elp
to
i
m
p
r
o
v
e
p
er
f
o
r
m
a
n
c
e
in
ter
m
s
o
f
t
i
m
e
r
eq
u
ir
ed
to
ex
ec
u
te
q
u
er
y
,
m
e
m
o
r
y
r
eq
u
ir
ed
to
s
to
r
e
d
atab
ase
an
d
I
/O
ac
ce
s
s
co
s
t.
B
y
co
n
s
id
er
in
g
all
ab
o
v
e
f
ea
t
u
r
es o
f
B
I
w
e
ar
e
u
s
in
g
it
in
o
u
r
r
esear
ch
.
Ag
g
r
eg
atio
n
f
u
n
ct
io
n
s
ac
r
o
s
s
m
a
n
y
a
ttrib
u
te
s
ar
e
co
m
m
o
n
l
y
u
s
ed
in
q
u
er
ies
o
f
d
ata
m
i
n
in
g
,
DW
an
d
O
L
A
P
[
1
3
]
,
[
1
4
]
.
T
h
e
co
m
m
o
n
l
y
u
s
ed
q
u
er
ies
in
d
at
a
m
i
n
i
n
g
a
n
d
DW
ar
e
I
B
Q,
w
h
ic
h
p
er
f
o
r
m
a
n
ag
g
r
e
g
ate
f
u
n
ctio
n
ac
r
o
s
s
at
tr
ib
u
tes
a
n
d
th
e
n
r
e
m
o
v
e
ag
g
r
eg
ate
v
al
u
es
t
h
at
ar
e
b
elo
w
s
o
m
e
s
p
ec
i
f
ied
th
r
es
h
o
ld
v
al
u
e.
Ge
n
er
all
y
u
s
ed
ag
g
r
e
g
atio
n
f
u
n
ctio
n
s
ar
e
MI
N,
M
A
X,
SUM,
AVG
a
n
d
C
OUNT
.
E
f
f
icie
n
t
co
m
p
u
tatio
n
o
f
all
th
e
s
e
ag
g
r
eg
ate
f
u
n
ctio
n
s
is
r
eq
u
ir
e
d
in
m
o
s
t
lar
g
e
d
atab
ase
a
p
p
licatio
n
s
b
ec
au
s
e
p
r
o
ce
s
s
in
g
co
s
t
o
f
ag
g
r
eg
a
te
f
u
n
ctio
n
is
m
u
ch
h
i
g
h
er
t
h
an
th
at
o
f
t
h
e
o
th
er
b
asic
r
elati
o
n
al
o
p
er
atio
n
s
lik
e
SEL
E
C
T
an
d
P
R
OJ
E
C
T
.
I
B
Q
r
ef
er
to
a
cla
s
s
o
f
q
u
er
ies
w
h
ic
h
co
m
p
u
te
a
g
g
r
e
g
ate
f
u
n
ctio
n
s
ac
r
o
s
s
attr
ib
u
tes
to
f
i
n
d
ag
g
r
eg
at
e
v
alu
e
s
ab
o
v
e
s
o
m
e
s
p
ec
if
ied
th
r
es
h
o
ld
v
alu
e.
T
h
e
n
u
m
b
er
o
f
tu
p
les,
th
at
s
at
is
f
y
t
h
e
th
r
e
s
h
o
ld
in
th
e
h
a
v
i
n
g
clau
s
e,
i
s
r
elativ
el
y
s
m
al
l
co
m
p
ar
ed
to
th
e
lar
g
e
a
m
o
u
n
t
o
f
i
n
p
u
t
d
ata.
A
s
o
u
tp
u
t
r
es
u
lt
i
s
v
er
y
s
m
al
l
s
o
ti
m
e
r
eq
u
ir
ed
f
o
r
ex
tr
ac
ti
n
g
it
m
u
s
t
b
e
less
.
S
y
n
tax
o
f
I
B
Q
is
as
b
elo
w
.
Gi
v
e
n
a
r
elatio
n
R
w
it
h
attr
ib
u
tes
a
1
,
a
2
…
a
n
,
an
ag
g
r
e
g
ate
f
u
n
ctio
n
Ag
g
F
u
n
(
)
,
an
d
a
th
r
es
h
o
ld
T
.
SEL
E
C
T
a
1
,
a
2
…
a
n
,
A
g
g
F
u
n
(
*
)
FR
OM
r
elatio
n
R
GR
OUP
B
Y
a
1
, a
2
…
a
n
HAVI
NG
Ag
g
F
u
n
(
*
)
>=
T
I
B
Q
co
n
ce
p
t
is
f
ir
s
t
s
tu
d
ie
d
b
y
Min
Fa
n
g
[
1
0
]
in
1
9
9
8
.
I
n
th
is
r
esear
ch
r
esear
ch
e
r
s
ex
ten
d
p
r
o
b
a
b
ilis
tic
tec
h
n
iq
u
e
u
s
ed
i
n
[
1
5
]
an
d
p
r
o
p
o
s
es
h
y
b
r
id
an
d
m
u
l
ti
b
u
c
k
et
al
g
o
r
ith
m
.
T
h
is
r
e
s
ea
r
ch
co
m
b
in
e
s
a
m
p
li
n
g
an
d
m
u
lti
h
as
h
f
u
n
c
t
io
n
s
to
i
m
p
r
o
v
e
t
h
e
p
er
f
o
r
m
a
n
ce
o
f
I
B
Q
a
n
d
r
ed
u
ce
m
e
m
o
r
y
r
eq
u
ir
e
m
en
t.
B
u
t
th
ese
al
g
o
r
ith
m
s
ar
e
n
o
t
s
u
ita
b
le
f
o
r
lar
g
e
d
ata
s
ets.
T
o
s
o
lv
e
ab
o
v
e
p
r
o
b
lem
[
1
0
]
p
r
o
p
o
s
es
al
g
o
r
ith
m
s
b
ased
o
n
s
a
m
p
li
n
g
a
n
d
b
u
ck
et
co
u
n
ti
n
g
m
et
h
o
d
s
.
T
h
ese
m
et
h
o
d
s
r
ed
u
ce
s
n
u
m
b
er
o
f
f
al
s
e
p
o
s
iti
v
e
v
al
u
es
b
u
t
it
tak
es
m
o
r
e
ti
m
e
to
ex
ec
u
te
q
u
er
y
a
s
it r
eq
u
ir
e
m
u
ltip
le
s
ca
n
o
f
r
elatio
n
.
I
B
Q
p
r
o
ce
s
s
in
g
is
al
s
o
p
r
o
p
o
s
ed
b
y
[
1
6
]
,
f
o
cu
s
o
f
t
h
is
s
t
u
d
y
is
to
r
ed
u
ce
n
u
m
b
er
o
f
tab
le
s
ca
n
s
s
o
t
h
at
t
i
m
e
r
eq
u
ir
ed
to
e
x
ec
u
te
th
e
q
u
er
y
w
ill
g
et
r
ed
u
ce
d
.
I
t
in
tr
o
d
u
ce
s
m
eth
o
d
s
to
s
ele
ct
ca
n
d
id
ate
v
al
u
es
u
s
i
n
g
p
ar
titi
o
n
i
n
g
a
n
d
p
o
s
tp
o
n
e
p
ar
titi
o
n
in
g
alg
o
r
it
h
m
s
.
C
o
llecti
v
e
I
B
Q
E
v
alu
at
io
n
is
p
r
o
p
o
s
ed
b
y
[
1
8
]
w
h
ic
h
p
r
ese
n
t
co
m
p
ar
i
s
o
n
u
s
in
g
th
r
ee
m
e
th
o
d
s
s
o
r
t
m
er
g
e
ag
g
r
eg
a
te,
h
y
b
r
id
h
as
h
ag
g
r
eg
ate
an
d
OR
AC
L
E
.
T
h
is
s
tu
d
y
p
r
o
v
es
t
h
at
p
er
f
o
r
m
a
n
ce
o
f
s
o
r
t
m
er
g
e
ag
g
r
e
g
ate
i
s
b
etter
o
n
d
ata
s
e
ts
w
it
h
lo
w
to
m
o
d
er
ate
n
u
m
b
er
.
Hy
b
r
id
h
as
h
a
g
g
r
e
g
ate
p
er
f
o
r
m
an
ce
w
as
n
o
t
g
o
o
d
w
h
e
n
d
ata
s
et
i
s
lar
g
e.
All
ab
o
v
e
m
e
n
tio
n
ed
m
et
h
o
d
s
c
o
m
e
s
u
n
d
er
th
e
g
r
o
u
p
o
f
tu
p
le
s
ca
n
b
ased
,
w
h
ic
h
r
eq
u
ir
es
o
n
e
p
h
y
s
ical
tab
le
s
c
an
to
r
ea
d
d
ata
f
r
o
m
d
is
k
.
Ho
w
e
v
er
[
1
8
]
tr
ies
to
m
ak
e
u
s
e
o
f
t
h
is
p
r
o
p
er
ty
o
f
I
B
Q
an
d
u
s
es
B
I
b
u
t
it
s
u
f
f
e
r
s
f
r
o
m
e
m
p
t
y
b
it
w
is
e
A
ND
r
esu
lt
p
r
o
b
le
m
.
R
esear
ch
er
s
[
4
]
tr
ies
to
m
in
i
m
ize
th
is
p
r
o
b
le
m
u
s
in
g
d
y
n
a
m
ic
p
r
u
n
i
n
g
a
n
d
v
ec
to
r
alig
n
m
e
n
t
alg
o
r
ith
m
s
.
Ho
w
ev
er
th
e
y
n
o
tice
th
at
t
h
er
e
is
p
r
o
b
lem
o
f
m
a
s
s
i
v
el
y
e
m
p
t
y
b
it
w
is
e
AND
r
es
u
lts
an
d
e
x
tr
a
XOR
o
p
er
atio
n
.
T
o
o
v
er
co
m
e
th
is
ch
alle
n
g
e
t
h
e
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
E
va
lu
a
tin
g
A
g
g
r
eg
a
te
F
u
n
ctio
n
s
o
f I
ce
b
erg
Qu
ery
Usi
n
g
P
r
i
o
r
it
y
B
a
s
ed
B
itma
p
....
(
K
a
le
S
a
r
ika
P
r
a
ka
s
h
)
3747
d
ev
elo
p
v
ec
to
r
alig
n
m
en
t a
lg
o
r
ith
m
w
h
ich
h
elp
to
s
o
l
v
e
e
m
p
t
y
b
it
w
i
s
e
A
ND
o
p
er
atio
n
p
r
o
b
le
m
.
T
h
e
p
r
o
b
lem
w
it
h
t
h
i
s
alg
o
r
it
h
m
is
t
h
at
all
v
ec
to
r
s
m
a
y
n
o
t
h
a
v
e
1
b
it
at
s
a
m
e
p
o
s
itio
n
an
d
i
f
it
i
s
n
o
t
a
t
s
a
m
e
p
o
s
it
io
n
t
h
e
n
all
t
h
e
A
N
D
as
w
ell
as
XO
R
o
p
er
atio
n
s
ar
e
f
r
u
it
less
an
d
ti
m
e
co
n
s
u
m
i
n
g
.
I
n
t
h
i
s
wa
y
b
o
th
t
h
e
ab
o
v
e
ap
p
r
o
ac
h
es[4
]
s
u
f
f
er
f
r
o
m
f
r
u
i
tles
s
A
ND
as
w
ell
as
XOR
o
p
er
atio
n
s
.
R
esear
c
h
[
5
]
tr
y
to
h
an
d
le
e
m
p
t
y
XO
R
o
p
er
atio
n
p
r
o
b
lem
b
u
t
d
id
n
o
t
ab
le
to
s
o
lv
e
f
r
u
itles
s
b
it
w
is
e
AND
o
p
er
atio
n
p
r
o
b
le
m
.
B
o
th
t
h
e
r
esear
ch
[
4
]
an
d
[
5
]
f
ac
es th
e
p
r
o
b
le
m
o
f
f
u
tile q
u
e
u
e
p
u
s
h
i
n
g
.
I
n
t
h
is
p
ap
er
,
w
e
h
a
v
e
s
o
lv
ed
th
e
p
r
o
b
lem
o
f
f
r
u
itle
s
s
b
it
wis
e
A
N
D,
O
R
a
n
d
XO
R
o
p
er
atio
n
a
n
d
f
u
tile
q
u
e
u
e
p
u
s
h
i
n
g
b
y
u
s
in
g
p
r
io
r
ity
b
ased
B
I
s
tr
ateg
y
.
T
h
i
s
ap
p
r
o
ac
h
im
p
r
o
v
e
s
ef
f
icie
n
c
y
b
y
p
r
u
n
in
g
m
a
n
y
g
r
o
u
p
s
b
e
f
o
r
eh
an
d
.
T
h
is
r
es
ea
r
ch
u
s
ed
t
h
e
s
i
m
ilar
it
y
m
atc
h
i
n
g
co
n
ce
p
t b
e
f
o
r
e
ass
i
g
n
i
n
g
p
r
io
r
ity
to
v
ec
to
r
an
d
f
o
r
m
s
t
h
e
cl
u
s
ter
o
f
t
h
e
s
a
m
e
[
1
9
]
.
3.
P
RIOR
I
T
Y
B
ASE
D
B
I
ST
R
AT
E
G
Y
F
O
R
I
B
Q
E
V
AL
U
AT
I
O
N
3
.
1
.
Wo
rking
m
o
del o
f
P
rio
rit
y
ba
s
ed
B
I
s
t
ra
t
eg
y
f
o
r
I
B
Q
ev
a
lua
t
io
n
T
h
is
s
ec
tio
n
d
escr
ib
es
th
e
w
o
r
k
f
lo
w
o
f
p
r
io
r
it
y
b
ased
B
I
s
tr
ateg
y
f
o
r
I
B
Q
ev
alu
atio
n
.
As
s
h
o
w
n
i
n
Fig
u
r
e
1
p
r
io
r
ity
b
ased
ap
p
r
o
a
ch
is
w
o
r
k
alo
n
g
w
ith
tr
ac
k
i
n
g
p
o
in
ter
s
tr
ate
g
y
a
s
w
ell
as
l
o
o
k
ah
ea
d
m
atc
h
i
n
g
m
et
h
o
d
.
On
ce
th
e
b
it
m
ap
v
ec
t
o
r
is
g
e
n
er
ated
t
h
en
p
r
io
r
it
y
b
ased
ap
p
r
o
ac
h
w
i
ll
m
a
k
e
u
s
e
o
f
tr
ac
k
in
g
p
o
in
ter
co
n
ce
p
t
to
ass
ig
n
p
r
io
r
it
y
to
v
ec
to
r
s
as
p
er
th
e
p
o
s
itio
n
o
f
1
’
s
o
cc
u
r
in
v
ec
to
r
.
Af
ter
f
i
n
aliz
atio
n
o
f
v
ec
to
r
s
f
o
r
p
er
f
o
r
m
in
g
b
it
w
is
e
A
ND
o
p
er
atio
n
th
en
t
h
e
lo
o
k
ah
ea
d
m
atc
h
in
g
s
tr
ate
g
y
w
i
ll
g
et
a
ctiv
ate
to
f
i
n
d
o
u
t
p
r
o
b
a
b
i
lit
y
o
f
t
h
at
o
p
er
atio
n
w
h
et
h
er
it
w
ill
s
ati
s
f
y
t
h
r
es
h
o
ld
co
n
d
itio
n
o
r
n
o
t.
I
f
it
r
ec
o
g
n
ize
t
h
at
p
o
s
s
ib
ilit
y
o
f
s
u
cc
es
s
i
s
le
s
s
th
e
n
i
t
w
il
l
s
k
ip
f
u
r
t
h
er
A
N
D
o
p
er
atio
n
.
I
n
th
i
s
w
a
y
it
h
elp
to
r
ed
u
ce
u
n
n
ec
ess
ar
y
b
u
r
d
en
o
f
p
er
f
o
r
m
in
g
f
r
u
it
less
b
it
w
is
e
A
N
D
o
p
e
r
atio
n
.
Fin
al
l
y
o
u
r
m
o
d
u
le
5
w
ill
e
x
ec
u
te
co
m
p
ar
e
o
p
er
atio
n
to
co
m
b
i
n
e
all
t
h
e
r
esu
lt
w
h
ich
s
atis
f
ies t
h
r
es
h
o
ld
co
n
d
itio
n
.
Fig
u
r
e
1
.
W
o
r
k
f
lo
w
Dia
g
r
a
m
3
.
2
.
P
s
eudo
co
de
f
o
r
P
rio
rit
y
ba
s
ed
B
I
s
t
ra
t
eg
y
f
o
r
I
B
Q
ev
a
l
ua
t
io
n
T
h
is
s
u
b
s
ec
tio
n
r
ep
r
esen
t
s
t
h
e
p
r
o
ce
s
s
in
g
f
lo
w
o
f
p
r
io
r
it
y
b
ased
B
I
s
tr
ateg
y
f
o
r
I
B
Q
e
v
al
u
atio
n
.
T
h
is
s
tr
ateg
y
i
s
m
ai
n
l
y
w
o
r
k
i
n
t
h
r
ee
p
h
ase
s
li
k
e
g
e
n
er
ati
n
g
B
I
,
t
r
ac
k
in
g
p
o
i
n
ter
s
tr
ate
g
y
a
n
d
l
o
o
k
ah
ea
d
m
atc
h
i
n
g
m
et
h
o
d
.
T
h
e
w
o
r
k
f
lo
w
o
f
al
g
o
r
ith
m
is
as b
elo
w
:
I
np
ut:
(
I
ce
b
er
g
Qu
er
y
(
Attr
ib
u
te
X,
A
ttrib
u
te
Y,
th
r
es
h
o
ld
T
)
,
T
a
b
le
P,
B
itm
ap
Vec
to
r
tab
le
o
f
P
)
P
ro
ce
s
s
ing
:
P
r
o
ce
s
s
in
g
o
f
al
g
o
r
ith
m
is
b
ased
o
n
n
u
m
b
er
o
f
d
is
ti
n
ct
v
a
lu
e
s
o
f
I
B
Q
attr
i
b
u
te
an
d
T
h
r
esh
o
ld
,
O
utput
:
(
I
B
Q
R
E
SUL
T
)
M
o
d
u
l
e
1
:
G
e
n
e
r
a
t
e
B
i
t
m
a
p
M
o
d
u
l
e
2
:
P
r
i
o
r
i
t
y
B
a
s
e
d
A
p
p
r
o
a
c
h
M
o
d
u
l
e
3
:
T
r
a
c
k
i
n
g
P
o
i
n
t
e
r
S
t
r
a
t
e
g
y
M
o
d
u
l
e
4
:
L
o
o
k
A
h
e
a
d
M
a
t
c
h
i
n
g
M
e
t
h
o
d
M
o
d
u
l
e
5
:
C
o
m
b
i
n
e
a
l
l
R
E
S
U
L
T
S
w
h
i
c
h
S
A
T
I
S
F
Y
T
H
R
E
S
H
O
L
D
c
o
n
d
i
t
i
o
n
O
u
t
p
u
t
:
I
B
Q
R
E
S
U
L
T
I
n
p
u
t
:
D
a
t
a
b
a
s
e
I
B
Q
A
t
t
r
i
b
u
t
e
s
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
.
6
,
Dec
em
b
er
2
0
1
7
:
3
7
45
–
3
7
52
3748
P
ha
s
e
1
:
Cre
a
t
e
B
it
m
a
p Ve
c
t
o
r
G
ener
a
t
io
n F
un
ct
io
n
I
t
co
n
tain
s
m
ai
n
f
u
n
ctio
n
s
w
h
i
ch
ar
e
u
s
ed
to
co
n
v
er
t
I
NP
UT
in
to
OUT
P
U
T
.
First
Fu
n
ctio
n
is
C
r
ea
t
e
B
T
MA
P
VE
C
T
OR
o
n
I
B
Q
attr
ib
u
te.
I
t
w
o
r
k
s
o
n
f
o
llo
w
i
n
g
f
o
r
m
u
la:
B
ITMA
P
VEC
TO
R
=
[
(
c
a
r
din
a
l
ity
of
C
l
oumn
A
+
c
a
r
din
a
l
ity
of
C
l
oumn
B
+
⋯
…
.
+
c
a
r
din
a
l
ity
of
C
l
oumn
N
)
]
×
No
.
o
f
R
ow
s
pr
e
s
e
n
t
in
Da
ta
b
a
s
e
]
T
h
is
f
o
r
m
u
la
also
u
s
ed
to
f
i
n
d
th
e
Sp
ac
e
C
o
m
p
le
x
it
y
o
f
Alg
o
r
it
h
m
.
R
ela
tio
n
s
h
ip
b
et
w
ee
n
ea
c
h
ca
r
d
in
alit
y
is
o
n
e
to
o
n
e
m
ea
n
s
o
n
e
v
ec
to
r
r
elate
d
to
o
n
e
o
n
l
y
.
T
h
e
attr
ib
u
te
w
h
ich
h
as
th
i
s
r
elat
io
n
s
h
ip
i
s
SET
to
1
o
th
er
w
is
e
0
.
I
n
th
is
w
a
y
co
m
p
lete
B
I
T
MA
P
VE
C
T
OR
is
cr
ea
ted
.
P
ha
s
e
2
:
T
ra
ck
i
ng
P
o
inte
r
Str
a
t
eg
y
1
.
Fo
r
ea
ch
b
itm
ap
v
ec
to
r
a
o
f
Attr
ib
u
te
X
C
OUNT
(
Nu
m
b
er
o
f
1
’
s
in
ea
ch
B
it
m
ap
v
ec
to
r
)
if
it
is
>
T
th
en
o
n
l
y
k
ee
p
s
u
ch
v
ec
to
r
i
n
B
I
.
Oth
er
w
i
s
e
d
is
ca
r
d
it
f
r
o
m
t
h
e
lis
t.
Fo
r
ea
ch
b
it
m
ap
v
ec
to
r
a
o
f
A
ttrib
u
te
X.
Fin
d
f
ir
s
t
1
b
it
p
o
s
itio
n
an
d
ac
co
r
d
in
g
l
y
allo
ca
te
P
r
io
r
it
y
P
r
i
o
r
ity
Q
u
eu
e
X.
clea
r
,
P
r
io
r
it
y
Qu
e
u
e
Y.
c
lear
.
Fo
r
ea
ch
v
ec
to
r
x
o
f
attr
ib
u
te
X
d
o
I
f
(
x
.
co
u
n
t>=
T
)
th
en
x
.
n
ex
t1
=Fir
s
tO
n
eB
itP
o
s
itio
n
(
x
,
0
)
2
.
Fo
r
ea
ch
b
it
m
ap
v
ec
to
r
a
o
f
A
t
tr
ib
u
te
Y
C
OUNT
(
Nu
m
b
er
o
f
1
’
s
i
n
ea
ch
B
it
m
ap
v
ec
to
r
)
if
it
is
>
T
th
en
o
n
l
y
k
ee
p
s
u
ch
v
ec
to
r
in
B
I
o
th
er
w
is
e
d
is
ca
r
d
it f
r
o
m
th
e
li
s
t.
P
r
io
r
ity
Qu
e
u
e
X.
clea
r
,
P
r
io
r
i
t
y
Q
u
eu
e
Y.
clea
r
.
Fo
r
ea
ch
v
e
cto
r
y
o
f
attr
ib
u
te
Y
do
I
f
(
y
.
co
u
n
t>=
T
)
th
en
y
.
n
e
x
t1
=Fir
s
tO
n
eB
itP
o
s
itio
n
(
y
,
0
)
3
.
Fin
d
f
ir
s
t 1
b
it p
o
s
itio
n
o
f
v
ec
t
o
r
X
an
d
Y
an
d
ac
co
r
d
in
g
l
y
al
lo
ca
te
P
r
io
r
ity
.
I
f
(
X.
P
o
s
itio
n
o
f
1
B
it >
Y.
P
o
s
itio
n
o
f
1
B
it)
T
h
en
(
FirstP
r
io
r
ity
==
X.
v
ec
t
o
r
)
E
ls
e
(
FirstP
r
io
r
ity
==
Y.
v
ec
to
r
)
4
.
I
f
(
X.
P
o
s
itio
n
o
f
1
B
it =
=
Y.
Po
s
itio
n
o
f
1
B
it)
T
h
en
(
FirstP
r
io
r
ity
==
X.
v
ec
t
o
r
)
as X
v
ec
to
r
ap
p
ea
r
s
f
ir
s
t i
n
s
eq
u
en
ce
a
n
d
Y
co
m
es late
r
.
5.
P
r
io
r
ity
Q
u
eu
eX.
P
u
s
h
(
x
)
6
.
P
r
io
r
ity
Q
u
eu
eY.
P
u
s
h
(
y
)
7
.
x
,
y
=N
e
x
tMa
tc
h
Vec
to
r
(
P
r
io
r
ity
Q
u
eu
eX.
clea
r
,
P
r
io
r
ity
Qu
e
u
e
Y,
T
)
8.
W
h
ile
x
!
=N
U
L
L
&
y
!
=N
UL
L
d
o
9.
P
r
io
r
ity
Q
u
eu
eX.
P
o
p
10.
P
r
io
r
ity
Q
u
eu
eY.
P
o
p
11.
C
u
r
r
en
tR
e
s
u
l
t=B
it
w
is
e
AND(
x
,
y
)
12.
I
f
(
C
u
r
r
en
tR
e
s
u
lt.c
o
u
n
t>=
T
)
th
en
13.
A
d
d
I
B
Q
R
esu
lt
in
R
E
SU
L
T
(
x
.
v
al
u
e,
y
.
v
al
u
e,
C
u
r
r
en
t
R
es
u
lt.
co
u
n
t)
14.
x
.
co
u
n
t=x
.
co
u
n
t
-
C
u
r
r
en
tR
e
s
u
l
t.c
o
u
n
t
15.
y
.
co
u
n
t=
y
.
co
u
n
t
-
C
u
r
r
en
tR
e
s
u
l
t.c
o
u
n
t
16.
I
f
x
.
co
u
n
t>=
T
th
en
17.
x
.
n
e
x
t1
=Fir
s
tOn
eB
itP
o
s
itio
n
(
x
,
x
.
n
ex
t+1
)
18.
I
f
x
.
n
ex
t1
!
=N
U
L
L
t
h
en
19.
P
r
io
r
ity
Q
u
eu
eX.
P
u
s
h
(
x
)
20.
I
f
y
.
co
u
n
t>=
T
th
en
21.
y
.
n
e
x
t1
=Fir
s
tOn
eB
itP
o
s
itio
n
(
y
,
y
.
n
ex
t+1
)
22.
I
f
y
.
n
ex
t1
!
=N
U
L
L
t
h
en
23.
P
r
io
r
ity
Q
u
eu
eY.
P
u
s
h
(
y
)
24.
R
ep
at
s
tep
7
-
2
3
f
o
r
n
ex
t
v
ec
to
r
x
,
y
=N
e
x
tMa
tc
h
Vec
to
r
(
P
r
io
r
ity
Q
u
eu
eX,
P
r
io
r
it
y
Q
u
eu
eY,
T
)
P
ha
s
e
3
:
L
o
o
k
a
hea
d
m
a
t
chi
ng
m
et
ho
d
I
f
R
E
SU
L
T
s
atis
f
ies
T
HR
E
S
HOL
D
co
n
d
it
io
n
th
e
n
to
p
r
ed
ict
th
e
p
o
s
s
ib
ilit
y
o
f
f
r
u
it
f
u
l
r
esu
lt
lo
o
k
ah
ea
d
m
atc
h
i
n
g
s
tr
ateg
y
is
u
s
ed
.
T
h
is
h
elp
to
r
ed
u
ce
f
r
u
it
less
A
ND,
O
R
a
n
d
XOR
o
p
er
atio
n
.
I
t
p
r
u
n
e
t
h
e
v
ec
to
r
as
it
id
e
n
ti
f
y
t
h
at
th
i
s
v
ec
to
r
w
ill
n
o
t
ab
le
to
p
r
o
d
u
ce
p
o
s
itiv
e
r
es
u
lt
.
I
n
t
h
i
s
w
a
y
t
h
i
s
m
o
d
u
le
s
k
ip
f
u
r
t
h
er
o
p
er
atio
n
al
o
v
er
h
ea
d
o
f
I
B
Q
p
r
o
ce
s
s
in
g
.
2
5
.
GE
NE
R
A
T
E
n
e
w
v
ec
to
r
s
b
y
p
er
f
o
r
m
in
g
OR
o
p
er
atio
n
b
etw
ee
n
R
E
SU
L
T
an
d
th
e
n
e
w
v
ec
to
r
w
h
ic
h
is
alr
ea
d
y
p
ar
t o
f
R
E
SU
L
T
.
Ne
w
X
Vec
to
r
=
Old
X
v
ec
to
r
-
C
u
r
r
en
t
R
es
u
lt Ve
cto
r
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
E
va
lu
a
tin
g
A
g
g
r
eg
a
te
F
u
n
ctio
n
s
o
f I
ce
b
erg
Qu
ery
Usi
n
g
P
r
i
o
r
it
y
B
a
s
ed
B
itma
p
....
(
K
a
le
S
a
r
ika
P
r
a
ka
s
h
)
3749
Ne
w
Y
Vec
to
r
=
Old
Y
v
ec
to
r
-
C
u
r
r
en
t
R
es
u
lt Ve
cto
r
26.
I
f
(
Ne
w
X/Y
Vec
to
r
)
s
atis
f
y
T
h
r
esh
o
ld
co
n
d
itio
n
th
e
n
p
er
f
o
r
m
s
tep
7
-
2
3
o
n
n
e
w
l
y
g
en
er
ated
v
ec
to
r
o
th
er
w
is
e
s
k
ip
th
e
r
esp
ec
ti
v
e
a
ttrib
u
te
f
r
o
m
th
e
v
ec
to
r
lis
t.
T
h
is
s
tep
h
elp
s
to
id
en
ti
f
y
th
e
p
o
s
s
ib
ilit
y
o
f
v
ec
to
r
to
b
e
p
ar
t
o
f
R
E
SU
L
T
f
u
r
t
h
er
.
27
.
R
ep
ea
t step
7
-
2
6
till
th
e
v
ec
to
r
lis
t
w
ill b
e
e
m
p
t
y
.
28
.
R
etu
r
n
I
B
Q
R
E
SU
L
T
4.
RE
SU
L
T
S
AND
D
I
SCU
SS
I
O
N
T
h
e
p
r
o
p
o
s
ed
m
et
h
o
d
is
i
m
p
le
m
en
ted
u
s
i
n
g
J
A
V
A
p
lat
f
o
r
m
o
n
th
e
I
B
M
-
co
m
p
a
tib
le
P
C
w
it
h
I
n
tel(
R
)
C
o
r
e
i3
p
r
o
ce
s
s
o
r
@
3
.
4
0
GHz
an
d
2
GB
R
A
M.
T
h
e
ex
p
er
i
m
en
t
is
p
er
f
o
r
m
ed
o
n
s
y
n
t
h
etic
d
ataset
o
f
s
ize
5
K,
1
0
K,
20K
,
4
0
K
an
d
8
0
K.
P
ar
am
eter
s
co
n
s
id
er
f
o
r
co
m
p
ar
is
o
n
an
d
to
m
ea
s
u
r
e
th
e
p
er
f
o
r
m
a
n
ce
ar
e
d
atab
ase
s
ize,
th
r
es
h
o
ld
v
al
u
e,
n
u
m
b
er
o
f
iter
atio
n
s
r
eq
u
ir
ed
to
ex
ec
u
te
q
u
er
y
,
ti
m
e
a
n
d
a
g
g
r
e
g
ate
f
u
n
ctio
n
s
.
T
h
e
g
r
ap
h
ical
illu
s
tr
atio
n
i
s
s
h
o
w
n
f
o
r
C
OU
N
T
an
d
SUM
ag
g
r
eg
ate
f
u
n
ctio
n
s
i
n
Fi
g
u
r
e
2
,
3,
4,
5,
6
an
d
7
.
W
e
o
b
s
er
v
ed
s
ig
n
if
ican
t
i
m
p
r
o
v
e
m
en
t
in
I
B
Q
p
er
f
o
r
m
an
ce
i
n
t
er
m
s
o
f
n
u
m
b
er
o
f
iter
atio
n
s
an
d
ti
m
e
r
eq
u
ir
ed
to
ex
ec
u
te
I
B
Q
u
s
i
n
g
o
u
r
p
r
io
r
it
y
b
ased
ap
p
r
o
ac
h
(
PB
A
)
.
W
e
h
av
e
co
m
p
ar
e
t
h
e
p
er
f
o
r
m
an
ce
o
f
p
r
io
r
it
y
b
ased
ap
p
r
o
ac
h
(
P
B
A
)
w
it
h
th
e
b
it
m
ap
in
d
e
x
in
g
ap
p
r
o
ac
h
(
B
I
A
)
s
u
g
g
ested
i
n
p
r
ev
io
u
s
w
o
r
k
[
4
]
-
[
8
]
.
W
e
o
b
s
er
v
e
t
h
at
as
w
e
g
o
o
n
in
cr
ea
s
in
g
s
ize
o
f
d
ata
s
e
t
an
d
th
r
es
h
o
ld
v
al
u
e
th
e
n
also
q
u
er
y
p
er
f
o
r
m
a
n
ce
is
g
o
es
o
n
in
cr
ea
s
i
n
g
w
h
ic
h
is
s
h
o
w
n
i
n
Fig
u
r
e
2
,
3
,
4
an
d
5
.
W
ith
p
r
ev
io
u
s
ap
p
r
o
ac
h
e
we
n
o
ticed
th
at
a
s
d
ata
s
ize
in
cr
ea
s
es
th
e
ti
m
e
r
eq
u
ir
ed
to
ex
tr
ac
t
d
ata
is
al
s
o
in
cr
ea
s
es.
B
ased
o
n
o
u
r
ex
p
er
i
m
e
n
tal
r
es
u
lt
w
e
h
a
v
e
p
r
o
v
ed
th
at
th
r
o
u
g
h
o
u
r
ap
p
r
o
ac
h
ev
en
t
h
o
u
g
h
d
ata
s
ize
in
cr
ea
s
es
t
h
e
n
also
I
B
Q
r
esp
o
n
s
e
ti
m
e
g
et
r
ed
u
ce
d
.
W
e
ar
e
u
s
i
n
g
B
I
s
tr
ateg
y
w
h
ic
h
h
e
lp
to
h
an
d
le
h
u
g
e
d
ata
ef
f
ec
tiv
e
l
y
[
1
3
]
,
[
1
4
]
.
T
h
is
i
s
also
n
o
ticed
th
r
o
u
g
h
o
u
r
ex
p
er
i
m
en
tatio
n
as
d
ata
s
ize
is
g
o
o
n
i
n
cr
ea
s
i
n
g
t
h
e
p
er
ce
n
tag
e
o
f
r
esp
o
n
s
e
t
i
m
e
is
r
ed
u
ce
d
.
A
s
s
h
o
w
n
i
n
Fig
u
r
e
3
an
d
5
th
r
o
u
g
h
ti
m
e
an
al
y
s
is
w
e
o
b
s
er
v
e
th
a
t f
o
r
s
m
al
l
d
ata
s
et
s
ize
i.e
.
5
k
,
1
0
k
an
d
u
p
to
2
0
k
d
if
f
er
en
ce
i
n
ti
m
e
r
eq
u
ir
ed
is
o
n
l
y
1
0
-
2
0
%
b
u
t
as
w
e
g
o
o
n
i
n
cr
ea
s
i
n
g
d
ataset
s
ize
f
r
o
m
2
0
k
,
4
0
k
to
8
0
k
d
i
f
f
er
e
n
ce
i
n
t
i
m
e
r
eq
u
ir
ed
i
s
r
ea
ch
e
s
to
4
5
-
5
0
%.
Fig
u
r
e
6
an
d
Fig
u
r
e
7
s
h
o
w
s
th
e
co
m
p
ar
ati
v
e
an
al
y
s
i
s
f
o
r
iter
atio
n
an
d
ti
m
e
f
o
r
SUM
f
u
n
ctio
n
.
T
h
e
n
u
m
b
er
o
f
iter
atio
n
s
r
eq
u
ir
ed
ar
e
d
r
asti
ca
ll
y
d
ec
r
e
ases
a
s
s
h
o
w
n
i
n
F
ig
u
r
e
7
.
B
u
t
d
u
e
to
lar
g
e
d
at
ab
ase
ac
ce
s
s
t
i
m
e
r
eq
u
ied
to
ex
ec
u
te
is
r
ed
u
ce
d
to
4
5
-
5
0
%
o
n
l
y
w
h
ic
h
i
s
a
s
s
h
o
w
n
i
n
Fi
g
u
r
e
6
.
T
h
is
i
n
d
icate
s
t
h
at
o
u
r
s
tr
ate
g
y
i
s
w
el
l
s
u
itab
le
f
o
r
lar
g
e
d
ata
s
et.
T
h
r
o
u
g
h
o
u
r
ex
p
er
i
m
e
n
tal
r
e
s
u
l
t
w
e
h
av
e
p
r
o
v
ed
t
h
at
p
r
io
r
it
y
b
ased
ap
p
r
o
ac
h
f
o
r
I
B
Q
p
r
o
ce
s
s
i
n
g
is
s
u
p
er
io
r
to
th
e
p
r
ev
io
u
s
b
it
m
ap
i
n
d
ex
i
n
g
ap
p
r
o
ac
h
.
I
n
th
is
w
a
y
w
e
h
av
e
d
ev
elo
p
ed
th
e
f
r
a
m
e
w
o
r
k
f
o
r
C
OUNT
,
SU
M,
MI
N
an
d
MA
X
a
g
g
r
e
g
ate
f
u
n
ctio
n
u
s
ed
in
I
B
Q.
Fig
u
r
e
2
.
I
ter
atio
n
An
al
y
s
i
s
o
f
C
OUNT
f
u
n
ctio
n
0
50
00
10
000
15
000
20
000
25
000
5K
10
K
20
K
40
K
80
K
5K
10
K
20
K
40
K
80
K
5K
10
K
20
K
40
K
80
K
T
h
e
re
s
h
o
ld
:3
00
T
h
e
re
s
h
o
ld
:3
10
T
h
e
re
s
h
o
ld
:3
20
It
e
ra
tio
n
s
in
H
u
n
d
re
d
s
BIA
PBA
Dat
ase
t
s
ize
an
d
T
h
re
s
h
o
ld
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
.
6
,
Dec
em
b
er
2
0
1
7
:
3
7
45
–
3
7
52
3750
Fig
u
r
e
3
.
T
im
e
A
n
al
y
s
is
o
f
C
OUNT
f
u
n
c
tio
n
Fig
u
r
e
4
.
I
ter
atio
n
An
al
y
s
is
o
f
SUM
f
u
n
ctio
n
Fig
u
r
e
5
.
T
im
e
An
al
y
s
is
o
f
S
UM
f
u
n
ct
io
n
0
50
10
0
15
0
20
0
25
0
5K
10
K
20
K
40
K
80
K
5K
10
K
20
K
40
K
80
K
5K
10
K
20
K
40
K
80
K
T
h
re
s
h
o
ld
:3
00
T
h
re
s
h
o
ld
:3
10
T
h
re
s
h
o
ld
:3
20
BIA
PBA
Dat
ase
t
s
ize
an
d
T
h
re
s
h
o
ld
T
im
e
in
Ms
e
cs
0
0.2
0.4
0.6
0.8
1
1.2
1.4
5K
10
K
20
K
40
K
80
K
5K
10
K
20
K
40
K
80
K
5K
10
K
20
K
40
K
80
K
5K
10
K
20
K
40
K
80
K
15
0000
16
0000
17
0000
18
0000
It
e
ra
tio
n
s
in
Milli
o
n
s
B
IA
PB
A
Dat
ase
t
s
ize
an
d
T
h
re
s
h
o
ld
0
10
0
20
0
30
0
40
0
50
0
60
0
70
0
80
0
90
0
10
K
20
K
50
K
10
K
20
K
50
K
10
K
20
K
50
K
2
3
4
BIA
PBA
Dat
ase
t
s
ize
an
d
T
h
re
s
h
o
ld
T
im
e
in
M
s
e
cs
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
E
va
lu
a
tin
g
A
g
g
r
eg
a
te
F
u
n
ctio
n
s
o
f I
ce
b
erg
Qu
ery
Usi
n
g
P
r
i
o
r
it
y
B
a
s
ed
B
itma
p
....
(
K
a
le
S
a
r
ika
P
r
a
ka
s
h
)
3751
Fig
u
r
e
6
.
C
o
m
b
in
e
T
i
m
e
An
al
y
s
i
s
o
f
SUM
f
u
n
ct
io
n
Fig
u
r
e
7
.
C
o
m
b
in
e
I
ter
atio
n
A
n
al
y
s
i
s
o
f
SUM
f
u
n
ctio
n
5.
CO
NCLU
SI
O
N
Ag
g
r
eg
ate
f
u
n
ctio
n
s
ar
e
th
e
m
ai
n
p
ar
t
o
f
an
y
d
ata
an
al
y
s
i
s
task
.
T
o
an
aly
ze
t
h
e
h
u
g
e
d
ataset
lik
e
DW
w
e
n
ee
d
to
e
x
ec
u
te
q
u
er
ies
w
h
ich
co
n
s
is
t
o
f
a
g
g
r
e
g
at
e
f
u
n
ct
io
n
.
I
n
s
u
ch
a
s
it
u
atio
n
if
q
u
er
y
i
s
ab
le
to
ex
ec
u
te
a
g
g
r
eg
ate
f
u
n
c
tio
n
e
f
f
icie
n
tl
y
t
h
en
i
t
w
ill
d
ir
ec
tl
y
r
ef
lect
o
n
th
e
p
er
f
o
r
m
an
ce
o
f
q
u
er
y
.
I
n
te
n
tio
n
o
f
th
is
r
esear
c
h
i
s
to
i
m
p
r
o
v
e
e
f
f
icien
c
y
o
f
ag
g
r
eg
a
te
f
u
n
c
tio
n
s
an
d
I
B
Q
w
h
ic
h
g
e
n
er
all
y
ex
e
cu
te
o
n
h
u
g
e
d
ata
s
et.
I
n
o
u
r
e
x
p
er
i
m
e
n
tal
a
n
al
y
s
is
w
e
co
m
p
ar
e
t
h
e
p
er
f
o
r
m
a
n
ce
o
f
o
u
r
ap
p
r
o
ac
h
w
it
h
p
r
ev
io
u
s
w
o
r
k
an
d
we
n
o
tice
s
i
g
n
i
f
ica
n
t
i
m
p
r
o
v
e
m
e
n
t
i
n
I
B
Q
p
er
f
o
r
m
a
n
ce
b
y
u
s
i
n
g
o
u
r
p
r
io
r
it
y
b
ased
B
I
s
tr
ate
g
y
.
W
e
n
o
ticed
t
h
at
ev
en
t
h
o
u
g
h
th
e
d
ata
s
et
s
ize
a
n
d
th
r
es
h
o
ld
v
al
u
e
in
cr
ea
s
es
t
h
en
al
s
o
th
e
d
ata
ex
tr
ac
tio
n
ti
m
e
g
e
t
r
ed
u
ce
d
.
On
th
e
b
asi
s
o
f
e
x
p
er
i
m
e
n
tal
r
es
u
l
t
w
e
h
a
v
e
p
r
o
v
ed
t
h
e
s
u
p
er
io
r
it
y
o
f
o
u
r
r
esear
ch
.
T
h
e
r
es
u
lt
o
f
t
h
i
s
r
esear
c
h
w
i
ll
h
elp
to
ex
ec
u
te
q
u
er
i
e
s
w
ith
ag
g
r
e
g
ate
f
u
n
ctio
n
as
w
e
ll
as
I
B
Q
w
h
ic
h
i
m
p
r
o
v
e
t
h
e
p
er
f
o
r
m
a
n
ce
o
f
O
L
A
P
q
u
er
ies
o
n
DW
.
T
h
e
f
o
cu
s
o
f
th
i
s
r
esear
ch
is
o
n
l
y
s
tr
u
ctu
r
ed
d
atab
ase
b
u
t
in
f
u
t
u
r
e
w
e
ca
n
ap
p
l
y
t
h
e
s
a
m
e
lo
g
ic
f
o
r
q
u
er
y
p
r
o
ce
s
s
i
n
g
o
n
u
n
s
tr
u
ct
u
r
ed
d
ata
an
d
it
w
ill
h
elp
f
u
l
f
o
r
b
ig
d
ata
a
n
al
y
s
i
s
.
RE
F
E
R
E
NC
E
S
[1
]
W
.
H.
In
m
o
n
,
“
Bu
il
d
in
g
t
h
e
d
a
ta
w
a
r
e
h
o
u
se
,
”
W
il
e
y
.
c
o
m
,
2
0
0
5
.
0
10
0
20
0
30
0
40
0
5K
10
K
20
K
40
K
80
K
5K
10
K
20
K
40
K
80
K
5K
10
K
20
K
40
K
80
K
T
h
re
s
h
o
ld
:3
00
T
h
re
s
h
o
ld
:3
10
T
h
re
s
h
o
ld
:3
20
PB
A
BIA
Dat
ase
t
s
ize
an
d
T
h
re
s
h
o
ld
T
im
e
in
Ms
e
cs
0
20
0
40
0
60
0
80
0
10
00
12
00
14
00
16
00
5K
10
K
20
K
40
K
80
K
5K
10
K
20
K
40
K
80
K
5K
10
K
20
K
40
K
80
K
5K
10
K
20
K
40
K
80
K
15
0000
16
0000
17
0000
18
0000
Ite
r
ation
s i
n
Th
o
u
san
d
s
PBA
B
IA
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
.
6
,
Dec
em
b
er
2
0
1
7
:
3
7
45
–
3
7
52
3752
[2
]
S.
S
u
sa
n
a
,
“
Qu
e
ry
o
p
ti
m
iza
ti
o
n
u
sin
g
f
u
z
z
y
lo
g
ic
in
in
teg
ra
t
e
d
d
a
tab
a
se
,
”
In
d
o
n
e
sia
n
J
o
u
rn
a
l
o
f
El
e
c
trica
l
En
g
i
n
e
e
rin
g
a
n
d
c
o
m
p
u
ter
sc
ien
c
e
,
v
ol
/i
ss
u
e
:
4
(
3
)
,
p
p
.
6
3
7
~
6
4
2
,
2
0
1
6
.
[3
]
A
.
Du
b
e
y
,
e
t
a
l.
,
“
Ef
f
e
c
ts
o
f
A
g
g
re
g
a
ti
o
n
a
n
d
Da
ta
S
ize
o
n
Qu
e
r
y
P
e
rf
o
r
m
a
n
c
e
a
n
d
M
e
m
o
r
y
Re
q
u
irem
e
n
ts
o
f
a
Da
ta W
a
r
e
h
o
u
se
,
”
ICROIT
,
2
0
1
4
.
[4
]
B
.
He
,
e
t
a
l.
,
“
Ef
f
icie
n
t
Ic
e
b
e
r
g
Qu
e
r
y
Ev
a
lu
a
ti
o
n
Us
in
g
Co
m
p
re
ss
e
d
Bit
m
a
p
In
d
e
x
,
”
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
Kn
o
wled
g
e
a
n
d
D
a
ta
En
g
i
n
e
e
rin
g
,
v
o
l
/i
s
u
e
:
24
(
9
)
,
p
p
.
1
5
7
0
-
1
5
8
9
,
2
0
1
1
.
[5
]
C.
V.
G
.
Ra
o
a
n
d
V
.
S
h
a
n
k
a
r,
“
Eff
icie
n
t
Ic
e
b
e
rg
Qu
e
r
y
Ev
a
lu
a
ti
o
n
Us
in
g
Co
m
p
re
ss
e
d
Bit
m
a
p
In
d
e
x
b
y
De
fe
rrin
g
Bit
w
ise
-
X
OR Op
e
ra
ti
o
n
s,
”
IEE
E
,
2
0
1
2
.
[6
]
C.
V.
G
.
Ra
o
a
n
d
V
.
S
h
a
n
k
a
r,
“
Co
m
p
u
ti
n
g
Ic
e
b
e
rg
Qu
e
rie
s
E
ff
i
c
ien
tl
y
Us
in
g
Bit
m
a
p
In
d
e
x
P
o
si
ti
o
n
s,
”
ICHCI
-
IEE
E
,
2
0
1
3
.
[7
]
S.
V
u
p
p
u
a
n
d
C.
V.
G
.
Ra
o
,
“
Ca
c
h
e
Ba
se
d
Ev
a
lu
a
ti
o
n
o
f
Ic
e
b
e
rg
Qu
e
ries
,
”
IEE
E
In
ter
n
a
ti
o
n
a
l
c
o
n
fer
e
n
c
e
o
n
Co
mp
u
ter
a
n
d
c
o
mm
u
n
ica
t
io
n
T
e
c
h
n
o
l
o
g
ies
(
ICCCT
),
2
0
1
4
.
[8
]
V.
C
.
S.
Ra
o
,
“
Eff
icie
n
t
ice
b
e
rg
q
u
e
ry
e
v
a
lu
a
ti
o
n
u
sin
g
se
t
re
p
r
e
se
n
tatio
n
,
”
IEE
E
INDICO
N,
p
p
.
1
-
5
,
2
0
1
4
.
[9
]
M.
J
.
Ba
sh
a
l
a
n
d
K.
P.
Ka
li
y
a
m
u
rth
ie,
“
A
n
im
p
ro
v
e
d
sim
il
a
rit
y
m
a
tch
in
g
b
a
se
d
c
lu
ste
ri
n
g
f
ra
m
e
w
o
rk
f
o
r
sh
o
rt
a
n
d
se
n
ten
c
e
lev
e
l
tex
t,
”
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
Co
mp
u
ter
En
g
in
e
e
rin
g
,
v
o
l
.
7,
p
p
.
5
5
1
~
5
5
8
,
2
0
1
7
.
[1
0
]
M
.
F
a
n
g
,
e
t
a
l.
,
“
Co
m
p
u
ti
n
g
ice
b
e
rg
q
u
e
ries
e
ff
icie
n
tl
y
,
”
VL
DB
Co
n
fer
e
n
c
e
., p
p.
2
9
9
-
3
1
0
,
1
9
9
8
.
[1
1
]
J.
G
ra
y
,
e
t
a
l.
,
“
Da
ta
Cu
b
e
:
A
r
e
latio
n
a
l
a
g
g
re
g
a
ti
o
n
o
p
e
ra
to
r
g
e
n
e
ra
li
z
in
g
g
ro
u
p
-
b
y
,
c
ro
ss
-
tab
,
a
n
d
su
b
-
to
tals.,
”
Da
ta
M
in
in
g
a
n
d
Kn
o
wled
g
e
Dis
c
o
v
e
ry
, p
p.
29
-
53,
1
9
9
7
.
[1
2
]
M
.
Jrg
e
n
s,
“
T
re
e
Ba
s
e
d
In
d
e
x
e
s
v
e
rsu
s
Bit
m
a
p
In
d
e
x
e
s:
A
P
e
rf
o
r
m
a
n
c
e
S
tu
d
y
,
”
Pro
c
.
In
t’l
W
o
rk
s
h
o
p
De
sig
n
a
n
d
M
a
n
a
g
e
me
n
to
f
D
a
ta
W
a
re
h
o
u
se
s (
D
M
DW
)
,
1
9
9
9
.
[1
3
]
A
n
Ora
c
le
W
h
it
e
P
a
p
e
r,
“
Ora
c
le Da
tab
a
se
1
1
g
f
o
r
Da
ta W
a
re
h
o
u
si
n
g
a
n
d
B
u
sin
e
ss
In
tell
ig
e
n
c
e
,
”
Or
a
c
le
,
2
0
1
1
.
[1
4
]
A
n
Ora
c
le
W
h
it
e
P
a
p
e
r,
“
Ora
c
le Da
tab
a
se
1
2
c
-
Bu
il
t
f
o
r
Da
ta w
a
re
h
o
u
se
,
”
Or
a
c
le,
2
0
1
4
.
[1
5
]
K.
Y.
W
h
a
n
g
,
e
t
a
l.
,
“
A
L
in
e
a
r
-
T
i
m
e
P
ro
b
a
b
il
isti
c
Co
u
n
ti
n
g
A
l
g
o
rit
h
m
f
o
r
Da
tab
a
s
e
A
p
p
li
c
a
ti
o
n
s,
”
ACM
T
ra
n
s.
Da
ta
b
a
se
S
y
ste
ms
,
v
o
l
/i
ss
u
e
:
15
(
2
)
,
p
p
.
2
0
8
-
2
2
9
,
1
9
9
0
.
[1
6
]
J.
Ba
e
a
n
d
S
.
L
e
e
,
“
P
a
rti
ti
o
n
i
n
g
A
l
g
o
rit
h
m
s
f
o
r
th
e
Co
m
p
u
tatio
n
o
f
A
v
e
r
a
g
e
I
c
e
b
e
rg
Qu
e
ries
,
”
Pr
o
c
.
S
e
c
o
n
d
In
t’l
Co
n
f.
Da
t
a
W
a
re
h
o
u
sin
g
a
n
d
Kn
o
wled
g
e
Dis
c
o
v
e
ry
(
Da
W
a
K)
,
2
0
0
0
.
[1
7
]
K.
P
.
L
e
e
la,
e
t
a
l.
,
“
On
I
n
c
o
rp
o
ra
ti
n
g
Ic
e
b
e
rg
Qu
e
ries
in
Qu
e
ry
P
r
o
c
e
ss
o
rs,
”
Pro
c
.
In
t’l
Co
n
f.
D
a
ta
b
a
se
S
y
ste
ms
fo
r
Ad
v
a
n
c
e
s A
p
p
li
c
a
ti
o
n
s (
DAS
FA
A
),
p
p
.
4
3
1
-
4
4
2
,
2
0
0
4
.
[1
8
]
A
.
F
e
rro
,
e
t
a
l.
,
“
Bit
Cu
b
e
:
A
Bo
tt
o
m
-
Up
Cu
b
in
g
E
n
g
in
e
e
rin
g
,
”
Pro
c
.
In
t’l
C
o
n
f
.
Da
t
a
W
a
re
h
o
u
sin
g
a
n
d
Kn
o
wled
g
e
Disc
o
v
e
ry
(
Da
W
a
K)
,
p
p
.
1
8
9
-
2
0
3
,
2
0
0
9
.
[1
9
]
M
.
Err
it
a
li
,
e
t
a
l
.
,
“
A
n
a
p
p
ro
a
c
h
o
f
se
m
a
n
ti
c
si
m
il
a
rit
y
m
e
a
s
u
re
b
e
tw
e
e
n
d
o
c
u
m
e
n
ts
b
a
se
d
o
n
b
ig
d
a
ta
,
”
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
E
lec
trica
l
a
n
d
C
o
mp
u
ter
En
g
in
e
e
rin
g
,
v
ol
/i
ss
u
e
:
6
(
5
)
,
pp.
2
4
5
4
~
2
4
6
1
,
2
0
1
6
.
B
I
O
G
RAP
H
I
E
S
O
F
AUTH
O
RS
M
s.
Ka
le
S
a
rik
a
P
ra
k
a
sh
is
t
h
e
re
se
a
rc
h
sc
h
o
lo
r
i
n
t
h
e
d
e
p
a
rtm
e
n
t
o
f
c
o
m
p
u
ter
sc
ien
c
e
a
n
d
e
n
g
in
e
e
rin
g
a
t
S
t.
P
e
ters
U
n
iv
e
rsity
Ch
e
n
n
a
i.
S
h
e
o
b
tain
e
d
h
e
r
B.
E.
(Co
m
p
u
ter
En
g
in
e
e
rin
g
)
f
ro
m
Un
iv
e
r
sit
y
o
f
P
u
n
e
,
M
a
h
a
ra
sh
tra
in
th
e
y
e
a
r
2
0
0
0
a
n
d
M
.
E.
(Co
m
p
u
ter
sc
ien
c
e
a
n
d
En
g
in
e
e
rin
g
)
f
ro
m
S
R
T
M
Un
iv
e
r
sity
,
Na
n
d
e
d
,
M
a
h
a
ra
sh
tra
in
th
e
y
e
a
r
2
0
0
5
.
S
h
e
h
a
s
b
e
e
n
in
th
e
tea
c
h
in
g
p
ro
f
e
ss
io
n
f
ro
m
th
e
p
a
st
1
7
y
e
a
rs.
H
e
r
a
re
a
o
f
in
tere
st
in
c
lu
d
e
d
a
ta
m
in
in
g
,
d
a
ta
w
a
r
e
h
o
u
sin
g
,
b
ig
d
a
ta,
b
u
sin
e
ss
a
n
a
l
y
ti
c
s,
m
a
c
h
in
e
lea
rn
i
n
g
,
o
p
e
ra
ti
n
g
sy
ste
m
,
s
y
ste
m
p
ro
g
ra
m
m
in
g
,
so
f
t
w
a
re
e
n
g
in
e
e
ri
n
g
a
n
d
so
f
twa
re
tes
ti
n
g
.
S
h
e
h
a
s
p
u
b
li
sh
e
d
1
4
p
a
p
e
rs
in
v
a
rio
u
s
In
tern
a
ti
o
n
a
l
Jo
u
r
n
a
ls
a
n
d
Co
n
f
e
r
e
n
c
e
s.
S
h
e
h
a
s
a
tt
e
n
d
e
d
m
a
n
y
wo
rk
sh
o
p
s
,
se
m
in
a
rs
a
n
d
F
DP
s
sp
o
n
s
o
re
d
b
y
IS
T
E,
A
IC
T
E
a
n
d
P
u
n
e
u
n
iv
e
rsity
re
late
d
to
h
e
r
a
re
a
o
f
in
tere
st.
S
h
e
is
a
li
f
e
m
e
m
b
e
r
o
f
CS
I,
IS
T
E
&
IA
EN
G
.
Dr.
Jo
e
P
ra
t
h
a
p
P
M
,
is
a
n
A
ss
o
c
iate
P
ro
f
e
ss
o
r
in
t
h
e
De
p
a
rtm
e
n
t
o
f
In
f
o
r
m
a
ti
o
n
T
e
c
h
n
o
lo
g
y
,
sin
c
e
Ju
n
e
2
0
1
1
.
He
o
b
tai
n
e
d
h
is
B.
E
(CS
E)
f
ro
m
S
t
.
X
a
v
ier’s
Ca
th
o
li
c
C
o
ll
e
g
e
o
f
En
g
in
e
e
rin
g
,
Ch
u
n
k
a
n
k
a
d
a
i,
M
.
E
(CS
E)
f
ro
m
Ka
ru
n
y
a
In
stit
u
te
o
f
T
e
c
h
n
o
lo
g
y
,
Co
im
b
a
to
re
a
n
d
P
h
.
D.
d
e
g
re
e
f
ro
m
A
n
n
a
Un
iv
e
rsit
y
,
Ch
e
n
n
a
i.
He
h
a
s
b
e
e
n
in
th
e
tea
c
h
in
g
p
ro
f
e
ss
io
n
f
o
r
th
e
p
a
st
1
0
y
e
a
r
s
a
n
d
h
a
s
h
a
n
d
led
b
o
th
UG
a
n
d
P
G
p
r
o
g
ra
m
m
e
s.
His
a
re
a
s
o
f
in
tere
st
in
c
lu
d
e
d
a
ta
m
in
in
g
,
m
a
c
h
in
e
lea
rn
in
g
,
Co
m
p
u
ter
Ne
tw
o
r
k
s,
Ne
t
w
o
rk
S
e
c
u
rit
y
,
Op
e
ra
ti
n
g
S
y
ste
m
s,
M
o
b
il
e
Co
m
m
u
n
ica
ti
o
n
a
n
d
Ob
jec
t
Orie
n
ted
A
n
a
ly
sis
a
n
d
De
sig
n
.
He
h
a
s
p
u
b
li
sh
e
d
2
3
p
a
p
e
rs
in
va
rio
u
s
In
tern
a
ti
o
n
a
l
Jo
u
rn
a
ls
a
n
d
Co
n
f
e
re
n
c
e
s.
He
h
a
s
a
tt
e
n
d
e
d
m
a
n
y
w
o
rk
sh
o
p
s
&
F
DP
s
sp
o
n
s
o
re
d
b
y
A
IC
T
E,
DST
&
IE
EE
re
late
d
to
h
is
a
re
a
o
f
in
tere
st.
He
is
a
li
fe
m
e
m
b
e
r
o
f
IS
T
E
&
IA
EN
G
.
Evaluation Warning : The document was created with Spire.PDF for Python.