I
nd
o
ne
s
ia
n J
o
urna
l o
f
E
lect
rica
l En
g
ineering
a
nd
Co
m
p
u
t
er
Science
Vo
l.
11
,
No
.
1
,
J
u
l
y
201
8
,
p
p
.
2
48
~
2
55
I
SS
N:
2502
-
4752
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ee
cs
.
v
1
1
.
i1
.
p
p
248
-
2
55
248
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
co
r
e.
co
m/jo
u
r
n
a
ls
/in
d
ex
.
p
h
p
/
ijeec
s
Predic
tion
O
utco
m
e
f
o
r Ma
ss
iv
e
M
ultiplay
er Onli
ne G
a
m
es
Using
Data
Minin
g
Sh
a
zw
a
ni
Sa
m
s
uri
m
,
No
r
As
hi
k
in M
o
ha
m
a
d K
a
m
a
l,
M
a
rina
I
s
m
a
il,
No
riza
n M
a
t
Di
a
h
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter S
c
ien
c
e
,
F
a
c
u
lt
y
o
f
Co
m
p
u
ter an
d
M
a
th
e
m
a
ti
c
a
l
S
c
ien
c
e
s
Un
iv
e
rsiti
T
e
k
n
o
lo
g
i
M
A
R
A
S
h
a
h
A
lam
,
S
e
lan
g
o
r,
M
a
la
y
sia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
an
8
,
2
0
1
8
R
ev
i
s
ed
Ma
r
1
5
,
2
0
1
8
A
cc
ep
ted
A
p
r
2
1
,
2
0
1
8
M
a
ss
iv
e
M
u
lt
ip
lay
e
r
On
li
n
e
(M
M
O)
g
a
m
e
i
s
o
n
e
o
f
th
e
fa
m
o
u
s
g
a
m
e
g
e
n
re
s
a
m
o
n
g
tee
n
a
g
e
r
s
n
o
w
a
d
a
y
s.
M
M
O
g
a
m
e
s
a
ll
o
w
g
a
m
e
rs
to
in
tera
c
t
a
n
d
p
lay
w
it
h
u
p
to
t
h
o
u
sa
n
d
p
lay
e
rs.
Ra
in
b
o
w
S
ix
S
ieg
e
(RS
S
)
b
e
lo
n
g
s to
M
M
O t
y
p
e
o
f
g
a
m
e
.
Ho
w
e
v
e
r,
d
u
e
t
o
m
a
n
y
o
p
e
ra
to
rs
t
h
a
t
a
re
a
v
a
il
a
b
le
in
th
i
s
g
a
m
e
,
th
e
p
lay
e
r
n
e
e
d
s
to
c
h
o
o
se
t
h
e
rig
h
t
o
p
e
ra
to
r
t
o
c
o
u
n
ter
th
e
e
n
e
m
y
o
p
e
ra
t
o
r.
T
h
e
re
f
o
re
,
b
a
se
d
o
n
th
e
c
h
a
ra
c
teristic
o
f
th
e
se
le
c
ted
o
p
e
ra
to
r,
th
is
p
a
p
e
r
a
tt
e
m
p
ted
to
p
r
e
d
ict
th
e
o
u
tco
m
e
s
o
f
th
e
g
a
m
e
.
In
o
u
r
p
re
d
icti
o
n
m
o
d
e
l,
c
h
a
ra
c
teristics
f
o
r
th
e
se
o
p
e
ra
to
rs
we
re
e
x
trac
t
e
d
f
ro
m
1
2
0
li
v
e
stre
a
m
re
p
lay
s.
T
h
re
e
c
las
si
f
ic
a
ti
o
n
a
lg
o
rit
h
m
s
w
e
r
e
u
ti
li
z
e
d
to
p
re
d
ict
t
h
e
o
u
tco
m
e
o
f
th
e
g
a
m
e
.
Am
o
n
g
th
e
se
a
l
g
o
rit
h
m
s,
IBK
h
a
d
o
b
tai
n
e
d
o
u
tstan
d
i
n
g
p
e
rf
o
r
m
a
n
c
e
in
th
e
d
a
tas
e
t.
T
h
e
a
c
c
u
ra
c
y
o
f
th
e
m
o
d
e
l
is 9
3
.
7
5
%
,
a
p
p
ly
in
g
5
-
f
o
ld
c
ro
ss
-
v
a
li
d
a
ti
o
n
tes
t.
T
h
e
su
c
c
e
ss
ra
te
re
v
e
a
ls
th
a
t
o
u
r
p
r
o
p
o
se
d
m
o
d
e
l
is
su
it
a
b
le t
o
p
re
d
ict
t
h
e
o
u
tco
m
e
o
f
th
e
g
a
m
e
.
K
ey
w
o
r
d
s
:
C
las
s
i
f
icatio
n
Ga
m
es
I
B
K
J
4
8
MM
OG
Naïv
e
B
a
y
es
Co
p
y
rig
h
t
©
2
0
1
8
I
n
stit
u
te o
f
Ad
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
:
Sh
az
w
a
n
i
Sa
m
s
u
r
i
m
,
Dep
ar
t
m
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
,
Facu
l
t
y
o
f
C
o
m
p
u
ter
an
d
Ma
th
e
m
atica
l Scie
n
ce
s
Un
i
v
er
s
iti
T
ek
n
o
lo
g
i M
A
R
A
Sh
a
h
A
la
m
,
Selan
g
o
r
,
Ma
la
y
s
i
a
.
E
m
ail:
s
h
az
w
a
n
i
s
a
m
s
u
r
i
m
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
T
h
e
g
a
m
e
i
n
d
u
s
tr
y
is
f
ac
i
n
g
a
s
u
r
g
e
o
f
d
ata
w
h
ic
h
r
es
u
lt
s
f
r
o
m
i
n
cr
ea
s
i
n
g
l
y
a
v
ailab
le
h
i
g
h
l
y
d
etaile
d
in
f
o
r
m
atio
n
ab
o
u
t
th
e
b
e
h
av
i
o
r
o
f
s
o
f
t
w
ar
e
an
d
u
s
er
s
.
T
ee
n
ag
er
s
e
n
j
o
y
p
la
y
i
n
g
o
n
li
n
e
g
a
m
e
s
w
h
ich
ar
e
ch
alle
n
g
i
n
g
b
u
t
f
air
to
all
p
la
y
er
s
.
B
ef
o
r
e
o
n
lin
e
g
a
m
e
s
b
ec
o
m
e
p
o
p
u
lar
,
g
a
m
er
s
u
s
u
a
ll
y
p
lay
a
s
in
g
le
p
la
y
er
g
a
m
e
s
u
ch
a
s
Ma
r
io
an
d
Hu
g
o
.
A
f
ter
s
o
m
e
ti
m
e
t
h
e
g
a
m
e
s
ev
o
lv
e
a
n
d
b
ec
o
m
e
m
a
s
s
i
v
e
m
u
ltip
la
y
er
o
n
lin
e.
Ma
s
s
i
v
e
M
u
ltip
la
y
er
O
n
li
n
e
(
MM
O)
allo
w
s
g
a
m
er
s
to
i
n
te
r
ac
t
an
d
p
la
y
w
it
h
o
th
er
p
la
y
er
s
u
p
to
t
h
o
u
s
a
n
d
p
lay
er
s
.
Ga
m
er
s
ca
n
s
ta
y
at
h
o
m
e
a
n
d
p
la
y
w
it
h
th
e
ir
f
r
ien
d
s
as
lo
n
g
t
h
e
y
h
a
v
e
an
i
n
ter
n
et
co
n
n
ec
tio
n
.
T
h
e
r
is
e
o
f
o
n
lin
e
g
a
m
e
s
b
ec
o
m
e
m
o
r
e
p
o
p
u
lar
w
h
en
m
a
n
y
p
lat
f
o
r
m
s
w
h
ic
h
p
r
o
v
id
es
v
ar
io
u
s
co
llectio
n
o
f
o
n
lin
e
g
a
m
e
s
a
n
d
allo
w
s
g
a
m
er
s
to
co
m
m
u
n
icate
w
i
t
h
ea
ch
o
th
er
an
d
cr
ea
te
a
co
m
m
u
n
it
y
t
o
d
is
cu
s
s
b
a
s
ed
th
e
s
elec
ted
g
a
m
es s
u
c
h
as ST
E
AM
,
GAREN
A
a
n
d
OR
I
GI
N.
G
a
m
e
’
s
g
en
r
e
li
k
e
s
tr
ate
g
y
a
n
d
f
ir
s
t
-
p
er
s
o
n
s
h
o
o
ter
ar
e
th
e
m
o
s
t
w
el
l
-
k
n
o
w
n
g
en
r
es a
m
o
n
g
g
a
m
er
s
.
Data
m
i
n
in
g
h
a
s
b
ee
n
u
s
ed
to
i
m
p
r
o
v
e
th
e
s
tr
ate
g
y
o
f
t
h
e
g
a
m
e
[
1
]
.
On
e
o
f
th
e
m
ai
n
ch
al
len
g
e
s
f
o
r
g
a
m
e
d
ata
m
in
i
n
g
is
to
p
r
ed
ic
t
s
tr
ateg
ie
s
to
m
ak
e
t
h
e
g
a
m
e
b
ec
o
m
e
m
o
r
e
ch
alle
n
g
i
n
g
a
n
d
f
air
to
all
p
lay
er
.
Fo
r
m
ass
iv
e
m
u
lt
ip
la
y
er
g
a
m
e
s
u
c
h
a
s
R
ain
b
o
w
Six
Sie
g
e
(
R
SS
)
,
d
ata
m
i
n
in
g
tech
n
iq
u
e
s
ca
n
b
e
u
s
e
to
p
r
ed
i
ct
th
e
ch
o
s
e
n
o
p
er
ato
r
f
o
r
th
e
u
s
er
to
co
u
n
ter
en
e
m
y
o
p
er
ato
r
ab
ilit
y
.
R
S
S
co
m
es
w
i
th
m
u
lti
p
le
o
p
er
at
o
r
s
w
h
ic
h
h
av
e
a
n
in
d
i
v
id
u
a
l
ab
ilit
y
to
c
o
u
n
ter
i
n
th
e
g
a
m
es
t
h
u
s
,
it
is
b
est
to
p
r
ed
ict
w
h
ich
o
p
er
ato
r
is
s
u
itab
le
to
u
s
e
f
o
r
a
u
s
er
to
ch
a
llen
g
e
th
e
o
p
p
o
n
en
t
o
p
er
ato
r
’
s
ab
ilit
y
.
A
l
ar
g
e
n
u
m
b
er
o
f
r
ep
la
y
s
w
er
e
co
llected
f
r
o
m
l
iv
e
s
tr
ea
m
s
f
r
o
m
a
n
in
ter
n
atio
n
al
to
u
r
n
a
m
en
t.
E
ac
h
r
ep
lay
i
s
d
iv
id
ed
in
to
m
a
n
y
r
o
u
n
d
s
an
d
i
s
lab
elled
b
ased
o
n
p
lay
er
c
h
o
s
en
o
p
er
ato
r
an
d
class
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esia
n
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4752
P
r
ed
ictio
n
Ou
tco
me
fo
r
Ma
s
s
i
ve
Mu
ltip
la
ye
r
On
lin
e
Ga
mes U
s
in
g
Da
ta
Min
in
g
(
S
h
a
z
w
a
n
i
S
a
msu
r
im
)
249
T
h
is
p
ap
er
test
s
th
e
ac
cu
r
ac
y
o
f
th
r
ee
class
if
icatio
n
al
g
o
r
ith
m
s
to
p
r
ed
ict
th
e
o
u
tp
u
t
o
f
m
as
s
iv
e
m
u
ltip
la
y
er
o
n
lin
e
g
a
m
e.
T
h
e
r
e
m
ain
d
er
o
f
t
h
is
p
ap
er
is
o
r
g
a
n
ized
as f
o
llo
w
s
.
Sectio
n
2
d
is
cu
s
s
es p
ast s
tu
d
ies
o
f
MM
O
a
n
d
d
ata
m
i
n
i
n
g
i
n
g
a
m
e
s
.
Sectio
n
3
ex
p
r
es
s
es
th
e
m
et
h
o
d
s
i
n
v
o
lv
ed
i
n
th
i
s
r
es
ea
r
ch
.
T
h
e
p
er
f
o
r
m
a
n
ce
m
ea
s
u
r
e
m
e
n
ts
u
s
ed
ar
e
d
is
cu
s
s
ed
i
n
Sectio
n
4
.
Sectio
n
5
p
r
esen
ts
t
h
e
r
esu
lt
s
,
an
d
th
e
co
n
cl
u
s
io
n
an
d
f
u
tu
r
e
w
o
r
k
s
ar
e
p
r
o
v
id
ed
in
Sectio
n
6
.
2.
B
ACK
G
RO
UND
O
F
T
H
E
S
T
UDY
T
h
is
r
esear
ch
f
o
cu
s
e
s
o
n
th
e
o
p
er
ato
r
p
r
ed
ictio
n
b
ased
o
n
th
e
o
p
p
o
n
en
t
o
p
er
ato
r
th
at
w
i
ll
b
e
u
s
ed
b
y
th
e
u
s
er
in
R
SS
.
T
h
er
e
ar
e
s
ev
er
al
ap
p
r
o
ac
h
es
in
r
ec
o
g
n
izi
n
g
t
h
e
p
atter
n
i
n
o
p
er
ato
r
s
elec
tio
n
.
T
h
is
s
ec
tio
n
d
is
cu
s
s
es t
h
eo
v
er
v
ie
w
o
f
m
as
s
iv
e
m
u
lt
ip
la
y
er
o
n
li
n
e
g
a
m
ea
n
d
p
r
ev
io
u
s
s
t
u
d
y
o
n
g
a
m
e
s
.
2
.
1
.
M
a
s
s
iv
e
M
ultipla
y
er
O
nli
ne
G
a
m
e
(
M
M
O
G
)
An
o
n
lin
e
g
a
m
e
is
a
g
a
m
e
c
o
n
tr
o
lled
b
y
t
h
e
co
m
p
u
ter
wh
er
e
t
h
e
p
la
y
er
i
n
ter
ac
t
s
u
s
i
n
g
co
m
p
u
ter
n
et
w
o
r
k
[
2
]
.
Ma
s
s
i
v
e
m
u
ltip
la
y
er
o
n
li
n
e
g
a
m
e
s
(
MM
OG)
h
av
e
ca
p
tu
r
ed
atte
n
tio
n
i
n
t
h
e
g
a
m
e
i
n
d
u
s
tr
y
s
in
ce
m
illi
o
n
s
o
f
p
eo
p
le
p
ar
ticip
ate
in
t
h
e
b
attle
u
s
i
n
g
th
e
s
a
m
e
s
er
v
er
[
3
]
.
T
ab
le
1
s
h
o
w
s
th
e
v
ar
io
u
s
MM
OG
s
av
ailab
le
o
n
t
h
e
i
n
ter
n
e
t.
M
MO
G
r
ep
r
esen
t
co
m
p
le
x
d
at
a
th
at
ca
n
b
e
d
iv
id
ed
i
n
to
s
u
b
p
r
o
b
lem
t
h
u
s
it
i
s
n
ec
es
s
ar
y
to
d
o
d
ata
m
i
n
i
n
g
in
th
i
s
g
a
m
e
to
s
tu
d
y
t
h
e
p
atter
n
o
r
s
tr
ateg
y
o
f
t
h
e
g
a
m
e
to
in
c
r
e
ase
w
in
n
i
n
g
r
atio
o
f
t
h
e
g
a
m
e
p
la
y
.
T
h
er
e
ar
e
s
e
v
er
al
p
ap
er
s
r
eg
ar
d
in
g
g
a
m
e
d
ata
m
i
n
i
n
g
.
An
al
y
zin
g
b
eh
av
io
r
d
ata
f
r
o
m
m
as
s
i
v
e
m
u
lt
ip
la
y
er
g
a
m
e
d
ata
ca
n
b
e
ch
a
llen
g
i
n
g
s
i
n
ce
i
t i
n
v
o
l
v
es
h
u
g
e
n
u
m
b
er
s
o
f
ac
ti
v
e
p
la
y
er
s
ac
r
o
s
s
t
h
e
w
o
r
ld
.
T
h
is
r
esear
ch
p
ap
er
s
u
g
g
es
ted
d
ata
m
i
n
i
n
g
to
ex
p
lo
r
e
d
ata
an
d
d
is
co
v
er
p
atter
n
s
t
h
at
ca
n
r
ed
u
ce
o
v
er
all
co
m
p
lex
it
y
o
f
t
h
e
d
ata
[
4
]
.
Data
m
i
n
in
g
alg
o
r
it
h
m
w
a
s
u
s
ed
to
i
d
en
tify
t
h
e
d
if
f
er
en
ce
b
et
w
ee
n
h
u
m
an
a
n
d
A
I
b
o
ts
in
[
9
]
.
I
n
th
e
s
tu
d
y
,
1
4
2
r
ep
lay
s
w
er
e
co
lle
cted
an
d
tr
ain
ed
u
s
in
g
m
ac
h
i
n
e
lear
n
in
g
.
T
h
e
r
esu
lt
s
h
o
w
s
t
h
at
it
is
p
o
s
s
ib
le
to
in
cr
ea
s
e
t
h
e
w
i
n
r
atio
o
f
t
h
e
g
a
m
e
i
f
t
h
e
b
o
ts
ab
le
to
r
ec
o
g
n
ize
th
e
o
p
p
o
n
en
t
’
s
s
tr
ate
g
y
a
n
d
s
u
cc
es
s
f
u
l
b
u
ild
th
e
o
r
d
er
.
Sev
er
al
ap
p
r
o
ac
h
es
h
a
v
e
b
ee
n
u
s
ed
to
m
o
d
el
t
h
e
o
p
p
o
n
en
t
[
1
0
,
1
1
,
18
].
Th
e
ap
p
r
o
ac
h
es
w
er
e
ap
p
lied
o
n
th
e
R
ea
l
T
im
e
S
tr
ateg
y
g
a
m
e
to
p
r
ed
ict
th
e
s
tr
ateg
y
o
f
th
e
o
p
p
o
n
en
t
s
s
o
th
at
p
lay
er
ca
n
r
esp
o
n
d
ac
co
r
d
in
g
l
y
to
th
e
g
a
m
e.
T
h
e
p
r
ed
ictio
n
s
h
ad
ap
p
lied
KNN,
d
ec
is
io
n
tr
ee
an
d
lo
g
is
tic
r
eg
r
ess
io
n
cla
s
s
i
f
ier
s
to
p
r
ed
ict
th
e
ac
cu
r
ac
y
o
f
t
h
e
s
t
r
ateg
y
[
1
0
]
.
A
B
a
y
e
s
ia
n
m
o
d
el
w
a
s
u
s
ed
to
p
r
ed
ict
th
e
r
es
u
lt
o
r
o
u
tco
m
e
s
o
f
is
o
lates
b
attle
an
d
also
p
r
ed
ic
t
w
h
at
u
n
it
w
ill
b
e
u
s
ed
to
d
ef
ea
t
a
g
i
v
en
ar
m
y
[
1
1
]
.
T
h
e
g
o
al
o
f
th
e
o
p
p
o
n
en
t
m
o
d
el
ap
p
r
o
ac
h
is
to
lear
n
a
n
d
s
elec
t
w
h
ic
h
co
m
b
i
n
atio
n
o
f
u
n
it i
s
m
o
r
e
e
f
f
ec
t
iv
e
a
g
ai
n
s
t
o
th
er
s
.
T
ab
le
1
.
Ma
s
s
iv
e
Mu
l
tip
la
y
er
On
li
n
e
Ga
m
e
T
i
t
l
e
G
r
a
p
h
i
c
s
S
e
t
t
i
n
g
S
u
b
s
c
r
i
p
t
i
o
n
M
o
d
e
l
R
e
l
e
a
se
D
a
t
e
S
t
a
r
W
a
r
s G
a
l
a
x
i
e
s
3D
S
c
i
e
n
c
e
F
i
c
t
i
o
n
P
a
y
t
o
P
l
a
y
2
0
0
3
W
o
r
l
d
o
f
W
a
r
c
r
a
f
t
3D
F
a
n
t
a
sy
P
a
y
t
o
P
l
a
y
2
0
0
4
R
a
i
n
b
o
w
S
i
x
S
i
e
g
e
(
R
S
S
)
3D
S
c
i
e
n
c
e
F
i
c
t
i
o
n
B
u
y
t
o
P
l
a
y
2
0
1
5
T
i
b
i
a
2D
F
a
n
t
a
sy
F
r
e
e
t
o
P
l
a
y
1
9
9
7
X
y
s
o
n
3D
A
p
o
c
a
l
y
p
t
i
c
F
a
n
t
a
sy
B
u
y
t
o
P
l
a
y
2
0
1
1
R
ain
b
o
w
Si
x
Sie
g
e
(
R
SS
)
h
a
s
b
ee
n
o
n
e
o
f
t
h
e
s
u
cc
ess
f
u
l
M
MO
G
w
it
h
h
u
g
e
n
u
m
b
er
o
f
p
l
a
y
er
s
o
v
er
th
e
w
o
r
ld
.
Ma
n
y
w
eb
s
i
tes
p
r
o
v
id
e
th
e
g
a
m
ep
la
y
t
h
at
s
to
r
es
all
th
e
g
a
m
i
n
g
ev
e
n
ts
an
d
to
r
ep
lay
th
e
g
a
m
e
o
f
g
o
o
d
p
lay
er
s
f
r
o
m
t
h
at.
A
s
a
r
esu
lt,
th
e
R
SS
co
m
m
u
n
i
t
y
h
as
ac
cu
m
u
lated
th
eir
ex
p
er
i
en
ce
in
t
h
e
f
o
r
m
o
f
r
ep
lay
s
f
o
r
s
e
v
er
al
y
ea
r
s
.
R
e
ce
n
tl
y
,
R
SS
g
a
m
e
h
a
s
g
ai
n
ed
a
lo
t
o
f
i
n
ter
e
s
t
i
n
th
e
g
a
m
i
n
g
i
n
d
u
s
tr
y
.
B
esid
e
Do
ta2
an
d
C
o
u
n
ter
Strik
e,
R
S
S
p
lace
d
th
ir
d
f
o
r
th
e
m
o
s
t
p
o
p
u
lar
g
a
m
e
in
th
e
w
o
r
ld
.
Fig
u
r
e
1
s
h
o
w
s
th
e
m
o
s
t
p
o
p
u
lar
MM
OG
in
th
e
w
o
r
ld
i
n
Dec
e
m
b
er
2
0
1
7
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2502
-
4752
I
n
d
o
n
esia
n
J
E
lec
E
n
g
&
C
o
m
p
Sci,
Vo
l.
11
,
No
.
1
,
J
u
ly
2
0
1
8
:
2
4
8
–
2
5
5
250
Fig
u
r
e
1
.
E
x
a
m
p
le
o
f
Ma
s
s
i
v
e
Mu
ltip
la
y
er
Ga
m
e
O
n
li
n
e
2
.
2
.
O
v
er
v
ie
w
o
f
R
a
inb
o
w
Six
Sieg
e
R
ain
b
o
w
Si
x
Sie
g
e
(
R
SS
)
is
o
n
e
o
f
th
e
p
o
p
u
lar
First
-
P
er
s
o
n
S
h
o
o
ter
w
it
h
r
ea
l
-
t
i
m
e
s
tr
at
eg
y
w
h
ic
h
u
s
e
s
c
u
r
r
en
t
tech
n
o
lo
g
y
a
n
d
i
s
d
iv
id
ed
in
to
t
w
o
tea
m
s
w
h
i
ch
i
s
at
tack
an
d
d
e
f
e
n
d
.
P
la
y
er
s
ca
n
ch
o
o
s
e
3
8
o
p
er
ato
r
w
h
ic
h
is
1
8
p
er
team
w
it
h
d
i
f
f
er
en
t
ab
ilit
ie
s
to
w
in
th
e
g
a
m
e
s
.
T
h
e
o
p
e
r
ato
r
w
ill
b
e
class
if
ied
in
to
4
d
if
f
er
e
n
t
class
e
s
in
attac
k
an
d
d
ef
en
d
.
T
h
er
e
a
r
e
1
0
p
lay
er
s
co
n
n
ec
ted
eith
er
r
an
d
o
m
o
r
in
a
p
a
r
ty
w
h
ic
h
is
5
p
er
s
o
n
s
p
er
team
.
B
o
th
tea
m
s
h
av
e
d
if
f
er
e
n
t
tas
k
at
th
e
ea
r
l
y
s
ta
g
e
o
f
th
e
g
a
m
e
(
f
ir
s
t
3
0
s
ec
o
n
d
s
)
w
h
ic
h
o
n
attac
k
th
e
y
n
ee
d
to
f
i
n
d
t
h
e
l
o
ca
tio
n
o
f
th
e
o
b
j
ec
tiv
e
an
d
id
en
ti
f
y
th
e
o
p
er
ato
r
tak
e
n
b
y
t
h
e
d
e
f
en
d
u
s
i
n
g
d
r
o
n
e.
On
th
e
d
ef
e
n
d
s
id
e
th
e
y
n
ee
d
to
p
r
o
tect
th
e
o
b
j
ec
ti
v
e
f
r
o
m
at
tack
er
t
h
u
s
th
e
y
n
ee
d
to
b
ar
r
icad
e
o
r
r
ein
f
o
r
ce
t
h
e
w
all
i
n
o
r
d
er
to
s
lo
w
d
o
w
n
o
r
p
r
ev
en
t
t
h
e
att
ac
k
er
to
s
ec
u
r
e
t
h
e
o
b
j
ec
tiv
e.
T
h
er
e
ar
e
s
ev
er
al
g
a
m
ep
la
y
m
o
d
es
w
h
ic
h
is
p
la
n
t
th
e
b
o
m
b
,
r
escu
e
h
o
s
ta
g
e
a
n
d
s
ec
u
r
e
t
h
e
ar
ea
w
it
h
s
e
v
e
n
t
ee
n
d
if
f
er
en
t
m
ap
s
.
R
SS
m
a
k
es
t
h
e
tea
m
w
i
n
b
y
k
illi
n
g
all
e
n
e
m
y
a
n
d
s
ec
u
r
e
t
h
e
o
b
j
ec
tiv
e
f
o
r
attac
k
er
s
w
h
i
le
o
n
d
ef
e
n
d
eith
er
k
ill
al
l
en
e
m
y
o
r
d
ef
e
n
d
t
h
e
o
b
j
ec
tiv
e
w
it
h
i
n
a
ti
m
e
l
i
m
it.
A
s
t
u
d
y
h
as
b
ee
n
co
n
d
u
cted
t
o
p
r
ed
ict
o
p
p
o
n
en
t
p
o
s
itio
n
in
g
u
s
i
n
g
o
p
p
o
n
en
t
m
o
d
elin
g
i
n
f
ir
s
t
p
er
s
o
n
s
h
o
o
ter
[
5
,
6
]
.
T
h
is
r
esear
ch
f
o
cu
s
es
o
n
o
n
e
m
ap
w
h
ic
h
is
p
r
esid
en
tial p
la
n
e
s
i
n
ce
th
e
m
ap
is
p
o
p
u
lar
an
d
h
a
v
e
m
a
n
y
r
ep
lay
s
o
n
th
e
w
eb
.
3.
M
E
T
H
O
DO
L
O
G
Y
Me
th
o
d
o
lo
g
y
i
s
t
h
e
s
p
ec
i
fi
c
s
tep
s
tak
e
n
d
u
r
i
n
g
t
h
e
p
r
o
j
ec
t
in
v
e
s
ti
g
atio
n
Fi
g
u
r
e
2
s
h
o
w
s
th
e
p
h
a
s
e
s
in
v
o
l
v
ed
in
th
i
s
r
esear
ch
.
I
t
s
tar
ts
w
it
h
d
ataset
d
ev
elo
p
m
en
t,
ex
tr
ac
tio
n
o
f
f
ea
t
u
r
es,
cla
s
s
if
icatio
n
s
b
y
th
r
ee
class
i
f
ier
s
Naï
v
e
b
a
y
e
s
,
I
B
K
an
d
J
4
8
.
I
t is th
en
f
o
llo
w
ed
b
y
r
esu
lt
s
an
d
ev
a
lu
at
io
n
.
Fig
u
r
e
2
.
R
esear
ch
F
lo
w
c
h
ar
t
Data
s
et
De
v
el
op
m
en
t
S
tart
F
ea
ture
E
x
tr
ac
ti
o
n
Cl
as
s
i
f
i
c
ati
on
Naïv
e B
a
y
es
J
48
I
B
K
Res
ul
ts
E
v
al
u
ati
on
E
nd
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esia
n
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4752
P
r
ed
ictio
n
Ou
tco
me
fo
r
Ma
s
s
i
ve
Mu
ltip
la
ye
r
On
lin
e
Ga
mes U
s
in
g
Da
ta
Min
in
g
(
S
h
a
z
w
a
n
i
S
a
msu
r
im
)
251
3
.
1
.
Da
t
a
s
et
Dev
elo
p
m
ent
I
n
th
e
g
a
m
i
n
g
i
n
d
u
s
tr
y
,
s
ev
er
al
w
eb
s
ite
s
ar
e
d
ed
icate
d
to
co
llect
an
d
s
h
ar
in
g
R
S
S
r
ep
la
y
s
.
T
h
ese
w
eb
s
i
tes
allo
w
s
h
ar
i
n
g
a
lar
g
e
p
o
r
tio
n
o
f
r
ep
la
y
s
o
f
g
a
m
e
p
lay
f
r
o
m
p
r
o
f
ess
io
n
al
a
n
d
h
i
g
h
-
r
a
n
k
ed
a
m
ate
u
r
m
atc
h
es.
T
h
er
ef
o
r
e,
u
s
in
g
all
t
h
ese
d
ata
co
llected
f
r
o
m
th
o
s
e
w
eb
s
ite
s
,
it
is
p
o
s
s
ib
le
to
m
i
n
e
th
ese
w
eb
s
ite
s
in
o
r
d
er
to
b
u
ild
a
co
llectio
n
o
f
a
ll
p
o
s
s
ib
ilit
ies
o
f
s
tr
ateg
ie
s
t
h
a
t
u
s
er
ca
n
u
s
e
to
w
i
n
t
h
e
g
a
m
e
p
la
y
i
n
R
S
S.
D
u
e
to
a
lar
g
e
n
u
m
b
er
o
f
r
ep
la
y
s
av
ailab
le,
it
is
p
o
s
s
ib
le
to
lear
n
a
v
ar
iet
y
o
f
o
p
er
at
o
r
a
b
ilit
y
an
d
also
th
e
s
tr
ateg
ie
s
o
n
s
ev
er
al
m
ap
s
ag
ain
s
t
d
i
f
f
er
e
n
t
p
la
y
s
t
y
les
to
h
elp
im
p
r
o
v
e
u
s
er
s
k
ills
i
n
o
r
d
er
to
w
i
n
th
e
g
a
m
e
.
T
h
e
d
ata
ca
n
h
elp
i
m
p
r
o
v
e
o
n
lin
e
m
atc
h
m
ak
in
g
,
m
o
d
elli
n
g
p
lay
er
b
eh
a
v
io
r
an
d
p
r
ed
icti
n
g
s
tr
ateg
y
[
1
0
-
1
2
]
.
T
h
e
d
ata
o
f
th
e
r
ep
la
y
s
w
er
e
co
llected
f
r
o
m
www
.
r
ed
d
it.c
o
m
a
n
d
h
t
tp
://s
tea
m
co
m
m
u
n
it
y
.
co
m
,
t
w
o
p
o
p
u
lar
R
SS
w
eb
s
ite.
T
h
e
co
llectio
n
s
o
f
t
h
e
r
ep
la
y
s
f
r
o
m
th
e
p
r
o
f
e
s
s
io
n
a
l
to
u
r
n
a
m
en
t
w
er
e
d
o
wn
lo
ad
ed
f
r
o
m
th
e
s
e
w
eb
s
i
tes.
T
h
e
s
co
p
e
o
f
t
h
is
r
e
s
ea
r
ch
i
s
li
m
ited
to
o
n
e
m
ap
o
n
l
y
.
Ta
b
le
2
s
h
o
w
s
th
e
lis
t
o
f
o
p
er
ato
r
s
b
ased
o
n
class
i
n
th
e
R
SS
.
T
ab
le
2
.
L
is
t o
f
R
SS
Op
er
ato
r
s
b
ased
o
n
C
las
s
A
t
t
a
c
k
D
e
f
e
n
d
e
r
C
l
a
ss
O
p
e
r
a
t
o
r
C
l
a
ss
O
p
e
r
a
t
o
r
B
r
e
a
c
h
e
r
A
sh
,
T
h
e
r
mi
t
e
,
H
i
b
a
n
a
,
S
l
e
d
g
e
,
Z
o
f
i
a
T
r
a
p
F
r
o
st
,
K
a
p
k
a
n
,
El
a
,
L
e
si
o
n
S
h
i
e
l
d
M
o
n
t
a
g
e
,
B
l
i
t
z
C
o
r
e
D
e
f
e
n
se
C
a
st
l
e
,
M
u
t
e
,
R
o
o
k
,
D
o
c
,
B
a
n
d
i
t
,
Jae
g
e
r
D
a
mag
e
G
l
a
z
,
F
u
z
e
,
B
l
a
c
k
b
e
a
r
d
,
C
a
p
i
t
a
o
,
B
u
c
k
D
a
mag
e
V
i
g
i
l
,
T
a
c
h
a
n
k
a
n
,
C
a
v
e
i
r
a
,
S
mo
k
e
U
t
i
l
i
t
y
Tw
i
t
c
h
,
Y
i
n
g
,
T
h
a
t
c
h
e
r
,
I
Q
,
D
o
k
k
a
e
b
i
,
Jac
k
a
l
I
n
f
o
r
mat
i
o
n
V
a
l
k
y
r
i
e
,
M
i
r
a
,
Ec
h
o
,
P
u
l
se
T
h
e
r
em
a
in
d
er
o
f
t
h
is
s
ec
tio
n
w
ill
d
escr
ib
e
d
ata
m
i
n
i
n
g
ap
p
r
o
ac
h
to
au
to
m
atica
ll
y
lear
n
s
tr
ateg
ies
f
r
o
m
co
llectio
n
o
f
r
ep
la
y
s
a
n
d
s
u
b
s
eq
u
e
n
tl
y
w
ill p
r
ed
ict
w
i
n
o
r
lo
s
e
o
f
th
e
R
SS
g
a
m
e.
3
.
2
.
F
ea
t
ure
E
x
t
ra
ct
io
n
R
SS
r
ep
la
y
s
ar
e
co
llected
b
ased
o
n
s
ec
u
r
e
ar
ea
g
a
m
e
m
o
d
es
an
d
p
r
esid
en
tial
p
lan
e
m
ap
.
A
d
atase
t
co
n
tain
i
n
g
3
2
g
a
m
e
s
ar
e
cr
ea
t
ed
b
ased
o
n
th
e
o
p
er
ato
r
’
s
cla
s
s
ta
k
e
n
a
n
d
r
esu
l
t
o
f
th
e
o
p
er
ato
r
ab
ilit
y
co
u
n
ter
.
T
ab
le
3
r
ep
r
esen
ts
t
h
e
ch
ar
ac
t
er
is
tic
o
f
o
p
er
ato
r
’
s
class
f
o
r
th
e
attac
k
er
a
n
d
th
e
d
ef
e
n
d
er
tea
m
in
R
S
S
w
h
er
e
t
h
e
in
f
o
is
e
x
tr
ac
ted
f
r
o
m
r
a
in
b
o
w
s
ixsi
eg
e.
fa
n
d
o
m.
T
ab
le
3
.
C
h
ar
ac
ter
is
tic
o
f
Op
e
r
ato
r
C
lass
T
h
e
g
o
al
o
f
th
is
r
esear
c
h
is
t
o
b
u
ild
a
g
en
er
al
m
o
d
el
o
f
R
SS
t
h
at
ca
n
p
r
ed
ict
t
h
e
o
u
tco
m
e
o
f
t
h
e
g
a
m
e
p
la
y
b
ased
o
n
th
e
o
p
er
ato
r
s
elec
tio
n
.
T
h
is
d
if
f
er
s
f
r
o
m
t
h
e
p
r
ev
io
u
s
s
tu
d
y
w
h
ic
h
f
o
cu
s
ed
o
n
p
lay
er
’
s
ac
tio
n
.
B
y
ap
p
l
y
i
n
g
d
ata
m
i
n
i
n
g
o
n
g
a
m
e
r
ep
la
y
s
,
th
e
al
g
o
r
ith
m
ca
n
d
e
v
elo
p
a
s
e
t o
f
r
u
les
th
at
ca
n
p
r
ed
ict
th
e
r
esu
lt
o
f
t
h
e
g
a
m
e
p
la
y
h
en
ce
th
e
u
s
er
ca
n
lear
n
th
e
o
p
er
ato
r
’
s
ab
ilit
y
t
h
r
o
u
g
h
t
h
e
alg
o
r
i
th
m
s
u
g
g
ested
t
h
e
n
th
e
y
w
ill
i
m
p
r
o
v
e
th
e
s
tr
ate
g
y
to
w
in
t
h
e
g
a
m
e.
T
ab
le
4
s
h
o
w
s
t
h
e
d
ata
u
s
ed
in
t
h
i
s
s
tu
d
y
b
ased
o
n
t
h
e
m
atc
h
m
a
k
i
n
g
th
at
h
ad
b
ee
n
r
ec
o
r
d
ed
f
r
o
m
t
h
e
r
ep
lay
.
E
v
er
y
attac
k
o
p
er
ato
r
w
i
ll
b
e
m
atc
h
ed
o
n
ev
er
y
d
ef
e
n
d
o
p
er
ato
r
an
d
th
e
r
e
s
u
lt o
f
w
i
n
an
d
lo
s
s
w
ill b
ased
o
n
t
h
e
r
ep
la
y
.
C
l
a
ss
A
r
m
o
r
(
L
o
w
,
M
e
d
i
u
m
,
H
i
g
h
)
M
o
v
e
me
n
t
(
S
p
e
e
d
1
,
2
,
3
)
F
l
e
x
i
b
i
l
i
t
y
(
L
o
w
,
M
e
d
i
u
m
,
H
i
g
h
)
B
r
e
a
c
h
e
r
L
o
w
2
M
e
d
i
u
m
S
h
i
e
l
d
H
i
g
h
1
M
e
d
i
u
m
D
a
mag
e
(
A
t
t
a
c
k
)
L
o
w
3
M
e
d
i
u
m
U
t
i
l
i
t
y
H
i
g
h
2
H
i
g
h
T
r
a
p
M
e
d
i
u
m
2
H
i
g
h
C
o
r
e
H
i
g
h
1
L
o
w
D
a
mag
e
(
D
e
f
e
n
d
)
L
o
w
3
H
i
g
h
I
n
f
o
r
mat
i
o
n
M
e
d
i
u
m
3
H
i
g
h
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2502
-
4752
I
n
d
o
n
esia
n
J
E
lec
E
n
g
&
C
o
m
p
Sci,
Vo
l.
11
,
No
.
1
,
J
u
ly
2
0
1
8
:
2
4
8
–
2
5
5
252
T
ab
le
4
.
C
h
ar
ac
ter
is
tics
o
f
s
el
ec
ted
R
SS
clas
s
o
p
er
ato
r
C
l
a
ss
1
C
l
a
ss
2
A
r
m
o
r
M
o
v
e
me
n
t
F
l
e
x
i
b
i
l
i
t
y
R
e
su
l
t
C
l
a
ss
1
C
l
a
ss
2
C
l
a
ss
1
C
l
a
ss
2
C
l
a
ss
1
C
l
a
ss
2
B
r
e
a
c
h
e
r
T
r
a
p
L
o
w
M
e
d
i
u
m
S
p
e
e
d
2
S
p
e
e
d
2
M
e
d
i
u
m
H
i
g
h
L
o
s
e
D
a
mag
e
T
r
a
p
L
o
w
M
e
d
i
u
m
S
p
e
e
d
3
S
p
e
e
d
2
M
e
d
i
u
m
H
i
g
h
L
o
s
e
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
T
r
a
p
U
t
i
l
i
t
y
M
e
d
i
u
m
H
i
g
h
S
p
e
e
d
2
S
p
e
e
d
3
H
i
g
h
H
i
g
h
L
o
s
e
C
o
r
e
D
e
f
e
n
se
B
r
e
a
c
h
e
r
H
i
g
h
L
o
w
S
p
e
e
d
1
S
p
e
e
d
2
L
o
w
M
e
d
i
u
m
L
o
s
e
3
.
3
.
Cla
s
s
if
ica
t
io
n
Alg
o
rit
h
m
s
T
h
r
ee
class
if
ier
s
ar
e
u
s
ed
i
n
th
is
s
t
u
d
y
.
T
h
er
e
ar
e
Naïv
e
B
ay
e
s
,
(
K
-
Nea
r
e
s
t
Nei
g
h
b
o
r
)
,
I
B
K
an
d
d
ec
is
io
n
tr
ee
(
J
4
8
)
.
3
.
3
.
1
.
Na
ïv
e
B
a
y
es
Naïv
e
B
a
y
es
i
s
u
s
ed
to
p
r
ed
ic
t
th
e
class
d
ep
en
d
in
g
o
n
th
e
p
r
o
b
ab
ilit
y
o
f
b
elo
n
g
in
g
to
th
e
class
[
1
3
]
.
T
h
e
p
r
o
b
ab
ilit
y
i
s
ca
lcu
lated
b
ased
o
n
E
q
u
atio
n
1
.
(
|
)
=
(
|
)
(
)
(
)
(
1
)
w
h
er
e
P
(
c|
x
)
i
s
th
e
p
o
s
ter
i
o
r
p
r
o
b
ab
ilit
y
o
f
clas
s
(
tar
g
et)
g
iv
e
n
p
r
ed
icto
r
(
attr
ib
u
te)
,
P
(
c)
i
s
th
e
p
r
io
r
p
r
o
b
a
b
ilit
y
o
f
clas
s
.
P
(
x|
c)
is
th
e
l
ik
el
ih
o
o
d
w
h
ich
is
th
e
p
r
o
b
ab
ilit
y
o
f
p
r
ed
icto
r
g
i
v
en
cl
ass
a
n
d
P
(
x
)
is
t
h
e
p
r
io
r
p
r
o
b
a
b
ilit
y
o
f
p
r
ed
icto
r
.
3
.
3
.
2
.
I
B
K
(
k
-
Nea
re
s
t
Neig
hb
o
r)
L
az
y
lear
n
i
n
g
is
a
lear
n
i
n
g
m
eth
o
d
in
w
h
ic
h
th
e
s
y
s
te
m
tr
i
es
to
g
en
er
alize
th
e
tr
ai
n
in
g
d
ata
b
ef
o
r
e
r
ec
eiv
in
g
q
u
er
ie
s
is
d
ela
y
ed
u
n
til
a
q
u
er
y
i
s
m
ad
e
to
th
e
s
y
s
te
m
[
1
4
]
.
L
az
y
ca
n
d
ea
l
w
i
th
ch
an
g
es
o
f
p
r
o
b
le
m
ar
ea
an
d
s
o
lv
e
m
u
ltip
le
p
r
o
b
le
m
[
1
5
]
.
I
B
k
class
i
f
ier
is
l
ik
e
k
-
Nea
r
est
Nei
g
h
b
o
r
(
KNN)
class
i
f
ier
w
h
er
e
b
y
k
is
a
u
s
ed
d
ef
i
n
ed
co
n
s
ta
n
t
[
1
9
]
.
T
h
e
s
i
m
ilar
it
ies
o
f
d
i
f
f
er
e
n
t
in
s
ta
n
ce
s
ar
e
ca
lc
u
lated
u
s
i
n
g
E
u
clid
ea
n
d
is
ta
n
ce
as in
E
q
u
atio
n
2.
2
1
,
)
(
s
j
W
s
s
i
j
i
V
V
d
(
2
)
w
h
er
e
i
V
is
th
e
v
ec
to
r
o
f
th
e
i
-
t
h
in
s
ta
n
ce
,
w
h
ile
j
V
is
th
e
s
-
th
ele
m
en
ts
o
f
v
ec
to
r
i
V
,
an
d
ij
d
is
th
e
d
is
tan
ce
b
et
w
ee
n
i
V
an
d
j
V
.
3
.
3
.
3
.
J
4
8
(
Dec
is
io
n T
re
e)
T
h
e
J
4
8
class
if
ier
(
W
ek
a
[
1
7
]
im
p
le
m
e
n
tatio
n
o
f
th
e
w
ell
-
k
n
o
w
n
C
4
.
5
d
ec
is
io
n
tr
ee
alg
o
r
ith
m
)
w
i
ll
g
en
er
ate
s
o
m
e
r
u
les
i
n
o
r
d
er
to
p
r
ed
ict
th
e
o
u
tp
u
t
v
a
r
iab
le
an
d
also
h
elp
in
d
escr
ib
in
g
t
h
e
cr
itical
co
n
tr
ib
u
tio
n
s
to
b
ec
o
m
e
ea
s
i
l
y
u
n
d
er
s
ta
n
d
ab
le
[
1
6
]
.
T
h
e
ad
d
itio
n
al
f
ea
t
u
r
es
o
f
J
4
8
w
er
eu
s
ed
to
f
in
d
m
i
s
s
i
n
g
v
alu
e
s
,
d
ec
is
io
n
tr
ee
s
p
r
u
n
i
n
g
,
co
n
tin
u
o
u
s
attr
ib
u
te
v
alu
e
r
a
n
g
es
a
n
d
d
er
iv
atio
n
o
f
r
u
les.
4.
P
E
RF
O
RM
ANCE
M
E
ASURE
M
E
NT
T
h
e
r
esu
lts
o
f
t
h
e
clas
s
if
ier
s
ar
e
ev
alu
ated
b
y
u
s
in
g
ac
c
u
r
a
c
y
,
tr
u
e
p
o
s
iti
v
e
(
T
P
)
r
ate,
p
r
ec
is
io
n
,
F
-
m
ea
s
u
r
e
an
d
m
ea
n
ab
s
o
lu
te
e
r
r
o
r
(
MA
E
)
.
A
cc
u
r
ac
y
ass
e
s
s
es
th
e
o
v
er
all
ef
f
ec
ti
v
e
n
ess
o
f
th
e
alg
o
r
ith
m
.
I
t
is
g
iv
e
n
b
y
E
q
u
a
tio
n
3
.
100
*
TN
FN
FP
TP
TN
TP
A
c
c
u
r
a
c
y
(
3
)
w
h
er
e
T
P
(
tr
u
e
p
o
s
itiv
e)
an
d
T
N
(
tr
u
e
n
eg
ati
v
e)
ar
e
t
h
e
n
u
m
b
er
s
o
f
co
r
r
ec
tl
y
p
r
ed
icted
p
o
s
itiv
e
a
n
d
n
eg
at
iv
e
s
a
m
p
le
s
r
esp
ec
ti
v
el
y
.
FP
(
f
alse
p
o
s
itiv
e)
a
n
d
FN
(
f
alse
n
eg
a
tiv
e)
ar
e
t
h
e
n
u
m
b
er
s
o
f
in
co
r
r
ec
tl
y
p
r
ed
icted
p
o
s
itiv
e
an
d
n
e
g
ati
v
e
s
a
m
p
les,
r
esp
ec
ti
v
el
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esia
n
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4752
P
r
ed
ictio
n
Ou
tco
me
fo
r
Ma
s
s
i
ve
Mu
ltip
la
ye
r
On
lin
e
Ga
mes U
s
in
g
Da
ta
Min
in
g
(
S
h
a
z
w
a
n
i
S
a
msu
r
im
)
253
T
h
e
T
P
R
ate
d
eter
m
i
n
es
t
h
e
p
lay
ca
s
e
r
atio
f
o
r
p
r
ed
icted
co
r
r
ec
tly
ca
s
es
to
th
e
to
tal
o
f
p
o
s
iti
v
e
ca
s
es.
I
t
is
a
p
r
o
b
a
b
ilit
y
co
r
r
e
cted
m
ea
s
u
r
e
o
f
a
g
r
ee
m
e
n
t
b
et
w
ee
n
t
h
e
class
i
f
icat
io
n
s
a
n
d
th
e
tr
u
e
clas
s
es.
I
n
th
is
T
P
ev
alu
atio
n
,
T
P
r
ate
w
ill
d
eter
m
i
n
e
n
u
m
b
er
o
f
ex
a
m
p
les
p
r
ed
icted
p
o
s
itiv
e
t
h
at
ar
e
ac
tu
all
y
p
o
s
it
iv
e
f
o
r
th
e
r
es
u
lt o
f
t
h
e
g
a
m
e.
P
r
ec
is
io
n
d
ef
in
e
as
p
o
s
iti
v
e
p
r
ed
ictiv
e
v
alu
e
t
h
at
w
ill
ca
lc
u
late
h
o
w
m
an
y
p
o
s
iti
v
e
p
r
ed
ictio
n
s
ar
e
co
r
r
ec
t f
o
r
th
e
r
esu
lt o
f
t
h
e
g
a
m
e.
E
q
u
atio
n
4
s
h
o
w
s
t
h
e
f
o
r
m
u
la
to
ca
lcu
late
p
r
ec
is
io
n
.
P
r
ec
is
io
n
=
TP
/ (
TP+F
P
)
(
4
)
w
h
er
e
TP
r
e
f
er
to
tr
u
e
p
o
s
iti
v
e
an
d
FP
r
e
f
er
to
f
al
s
e
p
o
s
itiv
e.
F
-
Me
a
s
u
r
e
i
s
a
co
m
b
i
n
atio
n
o
f
r
ec
all
a
n
d
p
r
ec
is
io
n
w
h
ic
h
is
i
n
a
s
i
n
g
le
p
er
f
o
r
m
a
n
ce
.
E
q
u
atio
n
5
s
h
o
ws th
e
f
o
r
m
u
la
to
ca
lcu
late
F
-
m
ea
s
u
r
e.
F
-
Mea
s
u
r
e=
2
*
P
r
ec
is
io
n
*
R
ec
a
ll /
(
P
r
ec
is
io
n
+R
ec
a
ll)
(
5
)
T
h
e
m
ea
n
ab
s
o
lu
te
er
r
o
r
(
MA
E
)
is
u
s
ed
to
m
ea
s
u
r
e
h
o
w
f
ar
th
e
p
r
ed
ictio
n
s
d
if
f
er
f
r
o
m
t
h
e
ac
tu
al
v
al
u
es
[
2
0
]
.
T
h
e
f
o
r
m
u
la
f
o
r
MA
E
i
s
g
i
v
e
n
in
E
q
u
atio
n
6
.
n
i
i
i
y
f
n
M
A
E
1
1
(
6
)
w
h
er
e
n
=t
h
e
n
u
m
b
er
o
f
er
r
o
r
s
an
d
|
−
|
=
th
e
ab
s
o
lu
te
er
r
o
r
s
.
5.
RE
SU
L
T
S AN
D
E
VA
L
UA
T
I
O
N
T
h
is
s
ec
tio
n
f
o
c
u
s
e
s
o
n
th
e
e
v
alu
a
tio
n
o
f
t
h
e
s
elec
ted
clas
s
if
ica
tio
n
al
g
o
r
ith
m
.
T
ab
le
5
s
h
o
w
s
t
h
e
ac
cu
r
ac
y
f
o
r
e
v
er
y
cla
s
s
i
f
ica
ti
o
n
alg
o
r
it
h
m
s
,
Naïv
e
B
a
y
es,
I
B
K
an
d
J
4
8
.
T
h
e
r
esu
lts
ar
e
o
b
tain
ed
b
y
u
s
in
g
5
-
f
o
ld
cr
o
s
s
v
alid
atio
n
.
T
h
ese
m
ea
s
u
r
e
m
en
t
s
ar
e
o
b
tain
ed
b
y
u
s
in
g
W
ek
a
[
1
7
]
to
o
lk
it.
T
ab
le
5
.
C
lass
if
icatio
n
al
g
o
r
it
h
m
a
n
d
o
v
er
all
ac
cu
r
ac
y
C
l
a
ssi
f
i
e
r
A
c
c
u
r
a
c
y
(
%)
N
a
ï
v
e
B
a
y
e
s
0
.
7
5
I
B
K
0
.
9
4
J4
8
0
.
6
6
A
ll
t
h
e
class
if
ier
p
r
ed
icts
th
e
o
u
tco
m
e
o
f
th
e
g
a
m
e
ac
co
r
d
in
g
to
th
e
o
p
er
ato
r
class
s
elec
ted
.
Fro
m
T
ab
le
5
,
it
ca
n
b
e
co
n
cl
u
d
ed
th
at
I
B
K
alg
o
r
it
h
m
h
a
s
t
h
e
h
ig
h
est
ac
c
u
r
ac
y
co
m
p
ar
ed
to
Naïv
e
B
a
y
e
s
an
d
J
4
8
w
h
ic
h
r
ea
ch
es
ac
c
u
r
ac
y
o
f
9
3
.
7
5
%.
W
h
ile
Naïv
e
B
a
y
es
r
ea
ch
es
7
5
%
o
f
ac
c
u
r
ac
y
an
d
J
4
8
h
av
e
t
h
e
lo
w
e
s
t
ac
cu
r
ac
y
v
al
u
e
6
5
%.
Fi
g
u
r
e
3
s
h
o
w
s
th
e
p
er
f
o
r
m
a
n
ce
ev
a
lu
atio
n
o
f
Naïv
e
B
a
y
es,
I
B
K
an
d
J
4
8
class
if
icatio
n
alg
o
r
ith
m
s
.
I
t
co
m
p
ar
es
th
r
ee
ty
p
es
o
f
p
er
f
o
r
m
a
n
ce
ev
al
u
at
io
n
w
h
ic
h
ar
e
T
r
u
e
Po
s
itiv
e
(
T
P
)
r
ate,
p
r
ec
is
io
n
an
d
F
-
Me
a
s
u
r
e.
Fig
u
r
e
3
.
P
er
f
o
r
m
a
n
ce
m
ea
s
u
r
en
t f
o
r
clas
s
i
f
icatio
n
al
g
o
r
ith
m
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2502
-
4752
I
n
d
o
n
esia
n
J
E
lec
E
n
g
&
C
o
m
p
Sci,
Vo
l.
11
,
No
.
1
,
J
u
ly
2
0
1
8
:
2
4
8
–
2
5
5
254
Fro
m
Fi
g
u
r
e
4
,
it
is
o
b
s
er
v
ed
th
at
J
4
8
alg
o
r
ith
m
o
b
tai
n
th
e
h
ig
h
es
t
er
r
o
r
r
ate
w
h
ile
I
B
K
o
b
tain
th
e
lo
w
es
t e
r
r
o
r
r
ate.
T
h
er
ef
o
r
e,
I
B
K
class
i
f
icatio
n
al
g
o
r
ith
m
p
e
r
f
o
r
m
w
el
l si
n
ce
it
h
as t
h
e
h
ig
h
est ac
c
u
r
ac
y
r
ate.
Fig
u
r
e
4
.
E
r
r
o
r
r
ate
f
o
r
class
if
i
ca
tio
n
alg
o
r
it
h
m
6.
CO
NCLU
SI
O
N
AND
F
U
T
U
RE
WO
RK
S
T
h
is
p
ap
er
d
eter
m
i
n
es
t
h
e
b
es
t
alg
o
r
it
h
m
b
y
co
m
p
ar
in
g
t
h
r
e
e
class
i
f
icat
io
n
tec
h
n
iq
u
e
s
to
p
r
ed
ict
th
e
o
u
tp
u
t
o
f
t
h
e
R
SS
.
T
h
is
s
t
u
d
y
m
ai
n
l
y
f
o
cu
s
o
n
t
h
r
ee
clas
s
if
icatio
n
tech
n
iq
u
e
s
,
Naï
v
e
B
a
y
es,
J
4
8
an
d
I
B
K
.
E
x
p
er
i
m
e
n
ts
h
a
v
e
b
ee
n
co
n
d
u
cted
to
d
eter
m
i
n
e
t
h
e
p
e
r
f
o
r
m
ac
e
f
o
r
ea
ch
o
f
t
h
e
al
g
o
r
ith
m
.
Am
o
n
g
t
h
e
th
r
ee
class
i
f
icatio
n
alg
o
r
it
h
m
s
,
it
ca
n
b
e
co
n
clu
d
ed
th
a
t
I
B
K
is
th
e
b
est
alg
o
r
ith
m
to
p
r
ed
ict
th
e
o
u
tco
m
e
o
f
R
S
S
g
a
m
e
co
m
p
ar
ed
to
Naïv
e
B
ay
es
an
d
J
4
8
.
T
h
er
ef
o
r
e,
I
B
K
m
o
d
el
is
s
u
itab
le
to
b
e
ap
p
lie
d
f
o
r
p
r
ed
ict
io
n
o
f
MM
O
g
a
m
e
s
s
i
n
ce
I
B
K
al
g
o
r
ith
m
p
er
f
o
r
m
b
etter
t
h
an
Naïv
e
B
a
y
es
a
n
d
J
4
8
.
A
s
f
o
r
f
u
tu
r
e
w
o
r
k
s
,
t
h
e
a
u
t
h
o
r
p
lan
s
to
i
n
v
e
s
ti
g
ate
t
h
e
ef
f
ec
t
o
f
f
ea
t
u
r
e
s
elec
t
io
n
to
t
h
e
g
a
m
e
o
u
tco
m
e
p
r
ed
ictio
n
.
Fu
r
t
h
er
m
o
r
e,
u
s
i
n
g
o
t
h
er
class
i
f
ier
s
s
h
o
u
ld
also
b
e
u
s
ed
in
f
u
t
u
r
e
ex
p
er
i
m
e
n
t
s
.
ACK
NO
WL
E
D
G
M
E
NT
S
T
h
is
w
o
r
k
is
s
u
p
p
o
r
ted
b
y
Ui
T
M
I
n
ter
n
al
R
e
s
ea
r
ch
Gr
a
n
ts
Sch
e
m
e
(
L
E
ST
AR
I
)
(
6
0
0
-
I
R
MI
/DAN
A
5
/3
/L
E
ST
A
R
I
(
1
1
8
/2
0
1
6
)
)
f
r
o
m
th
e
r
esear
c
h
Ma
n
a
g
e
m
en
t
C
en
tr
e
(
R
MC),
U
n
iv
er
s
itiT
ek
n
o
lo
g
i
M
A
R
A
,
S
h
a
h
A
la
m
,
Selan
g
o
r
.
T
h
e
ap
p
r
ec
iatio
n
also
g
o
es
to
Fac
u
lt
y
o
f
C
o
m
p
u
ter
a
n
d
Ma
th
e
m
atica
l
Scien
ce
s
(
F
SKM)
,
Un
i
v
er
s
iti
T
ek
n
o
lo
g
i
M
A
R
A
,
Sh
ah
Ala
m
,
Se
lan
g
o
r
f
o
r
g
i
v
in
g
a
m
o
r
al
s
u
p
p
o
r
t
in
th
e
p
r
o
d
u
ctio
n
o
f
th
i
s
p
ap
er
.
RE
F
E
R
E
NC
E
S
[1
]
M
il
jk
o
v
ic
D,
Ga
ji
c
L
,
Ko
v
a
c
e
v
ic
A
,
Ko
n
jo
v
ic
Z.
T
h
e
u
se
o
f
d
a
t
a
min
i
n
g
f
o
r
b
a
sk
e
tb
a
ll
ma
tch
e
s
o
u
tco
me
s
p
re
d
ictio
n
.
IE
EE
8
th
I
n
tern
a
ti
o
n
a
l
S
y
m
p
o
siu
m
o
n
in
telli
g
e
n
t
a
n
d
in
f
o
r
m
a
ti
c
s.
S
u
b
o
ti
c
a
,
S
e
rb
ia.
2
0
1
0
:
3
0
9
-
3
1
2
.
[2
]
Hla
d
k
y
S
,
Bu
li
tk
o
V
.
A
n
e
v
a
l
u
a
ti
o
n
o
f
mo
d
e
ls
fo
r
p
re
d
ictin
g
o
p
p
o
n
e
n
t
p
o
si
ti
o
n
s
in
fi
rs
t
-
p
e
rs
o
n
sh
o
o
ter
v
id
e
o
gam
es
.
P
r
o
c
e
e
d
in
g
s
o
f
th
e
IE
EE
S
y
m
p
o
siu
m
o
n
Co
m
p
u
tatio
n
a
l
In
telli
g
e
n
c
e
in
G
a
m
e
s,
P
e
rth
,
A
u
str
a
li
a
.
2
0
0
8
:
3
9
-
46.
[3
]
Da
n
e
v
a
,
M
.
Ho
w
p
ra
c
ti
ti
o
n
e
rs
a
p
p
r
o
a
c
h
g
a
me
p
l
a
y
re
q
u
ire
me
n
ts?
An
e
x
p
o
lo
ra
t
io
n
i
n
to
t
h
e
c
o
n
tex
t
o
f
m
a
ss
ive
mu
lt
ip
l
a
y
e
r o
n
li
n
e
r
o
le
-
p
l
a
y
in
g
g
a
me
s
.
2
2
n
d
IEE
E
I
n
t.
Re
q
u
ir
.
E
n
g
.
Co
n
f
.
2
0
1
4
:
3
-
1
2
.
[4
]
Ba
u
c
k
h
a
g
e
C,
Dra
c
h
e
n
A
a
n
d
S
if
a
R.
Clu
ste
rin
g
g
a
me
b
e
h
a
v
i
o
r
d
a
ta
.
IEE
E
T
ra
n
s.
Co
m
p
u
t.
In
t
e
ll
.
A
I
Ga
m
e
s
.
2
0
1
5
;
7
(3
)
:
2
6
6
-
2
7
8
.
[5
]
Ja
n
sz
J
a
n
d
T
a
n
is M
.
A
p
p
e
a
l
o
f
P
la
y
in
g
On
li
n
e
F
irst
P
e
rso
n
S
h
o
o
te
r
Ga
m
e
s.
Cy
b
e
r
Ps
y
c
h
o
lo
gy
Beh
a
v
.
2
0
0
7
;
1
0
(
1
):
133
–
1
3
6
.
[6
]
W
e
b
e
r
B
G
,
M
a
tea
s
M
.
A
d
a
t
a
mi
n
in
g
a
p
p
r
o
a
c
h
to
stra
teg
y
p
re
d
ic
ti
o
n
.
IEE
E
S
y
m
p
.
Co
m
p
u
t.
In
tell.
G
a
m
e
s.
2
0
0
9
:
140
-
1
4
7
.
[7
]
De
re
sz
y
n
s
k
i
E,
Ho
ste
tl
e
r
J,
F
e
rn
A
,
Die
tt
e
rich
T
,
Ho
a
n
g
T
,
Ud
a
rb
e
M
.
L
e
a
rn
i
n
g
p
r
o
b
a
b
i
li
stic
b
e
h
a
v
io
r
mo
d
e
ls
in
re
a
l
-
ti
me
stra
teg
y
g
a
me
s
. 7
th
A
rti
fi
c
ial
In
telli
g
e
n
c
e
a
n
d
In
tera
c
ti
v
e
Dig
it
a
l
En
terta
in
m
e
n
t
Co
n
f
e
re
n
c
e
.
2
0
1
1
:
2
0
-
2
5
.
[8
]
P
a
rk
HS,
Ch
o
HC,
L
e
e
KY
,
Ki
m
KJ
.
Pre
d
ictio
n
o
f
e
a
rly
sta
g
e
o
p
p
o
n
e
n
ts
stra
teg
y
f
o
r
S
t
a
rCr
a
ft
AI
u
sin
g
sc
o
u
ti
n
g
a
n
d
ma
c
h
i
n
e
lea
r
n
in
g
.
I
n
P
r
o
c
e
e
d
in
g
s o
f
th
e
W
o
rk
sh
o
p
a
t
S
IGG
R
AP
H A
sia
(W
A
S
A
2
0
1
2
).
2
0
1
2
:
7
-
12.
[9
]
Ch
iu
KSY,
Ch
a
n
KCC.
Ga
me
e
n
g
i
n
e
d
e
sig
n
u
si
n
g
d
a
ta
min
in
g
.
P
ro
c
e
e
d
i
n
g
s
o
f
th
e
2
6
th
I
A
S
T
E
D
In
tern
a
ti
o
n
a
l
Co
n
f
e
re
n
c
e
o
n
A
rti
f
icia
l
In
telli
g
e
n
c
e
a
n
d
A
p
p
li
c
a
ti
o
n
s.
2
0
0
8
:
3
5
2
-
3
5
7
.
27,62
15,27
35,49
0
10
20
30
40
N
aï
v
e
B
a
ye
s
IB
K
J48
MAE
Cl
ass
f
ie
r
Mean
Ab
so
lut
e
Err
or
(
MAE)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esia
n
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4752
P
r
ed
ictio
n
Ou
tco
me
fo
r
Ma
s
s
i
ve
Mu
ltip
la
ye
r
On
lin
e
Ga
mes U
s
in
g
Da
ta
Min
in
g
(
S
h
a
z
w
a
n
i
S
a
msu
r
im
)
255
[1
0
]
Ho
ste
tl
e
r
J,
De
re
sz
y
n
sk
i
E,
Die
tt
re
rich
T
.
F
e
rn
A
.
In
fer
rin
g
str
a
teg
ies
fro
m
li
mi
ted
re
c
o
n
n
a
issa
n
c
e
in
re
a
l
-
ti
me
stra
teg
y
g
a
me
s
.
P
ro
c
e
e
d
i
n
g
s o
f
th
e
2
8
th
Co
n
f
e
re
n
c
e
o
n
Un
c
e
rtain
ty
in
A
rti
f
icia
l
In
telli
g
e
n
c
e
.
2
0
1
2
:
3
6
7
-
3
7
6
.
[1
1
]
Hs
ieh
JL
a
n
d
S
u
n
CT
.
Bu
il
d
i
n
g
a
p
l
a
y
e
r
stra
te
g
y
m
o
d
e
l
b
y
a
n
a
lyz
in
g
re
p
la
y
s
o
f
re
a
l
ti
me
stra
teg
y
g
a
me
s.
In
ter
n
a
t
io
n
a
l
J
o
in
t
C
o
n
fer
e
n
c
e
o
n
Ne
u
r
a
l
Ne
two
rk
s
.
p
a
rt
o
f
th
e
IEE
E
W
o
rl
d
Co
n
g
re
ss
o
n
Co
m
p
u
tatio
n
a
l
In
telli
g
e
n
c
e
,
Ho
n
g
Ko
n
g
,
Ch
in
a
.
2
0
0
8
:
3
1
0
6
-
3
1
1
1
.
[1
2
]
S
y
n
n
a
e
v
e
G
,
Be
ss
i
`
e
r
e
P
.
A
d
a
t
a
se
t
fo
r
S
ta
rc
ra
f
t
AI
&
a
n
e
x
a
mp
l
e
o
f
a
rm
ies
c
lu
ste
rin
g
.
A
rti
f
ici
a
l
In
telli
g
e
n
c
e
in
A
d
v
e
r
sa
rial
Re
a
l
-
T
i
m
e
Ga
m
e
s.
2
0
1
2
:
2
5
-
3
0
.
[1
3
]
T
v
e
it
A
.
Ga
me
u
sa
g
e
min
in
g
:
In
fo
rm
a
t
io
n
g
a
th
e
rin
g
f
o
r
k
n
o
w
led
g
e
d
isc
o
v
e
ry
in
ma
ss
ive
mu
lt
i
p
la
y
e
r
g
a
me
s.
P
r
o
c
e
e
d
in
g
s o
f
th
e
I
n
tern
a
ti
o
n
a
l
Co
n
f
e
re
n
c
e
o
n
In
tern
e
t
Co
m
p
u
ti
n
g
,
2
0
0
2
:
6
3
6
-
6
4
2
.
[1
4
]
Esm
it
a
A
P
,
G
u
p
ta M
E.
P
ro
c
e
ss
M
in
i
n
g
:
A
Co
m
p
a
ra
ti
v
e
S
tu
d
y
.
In
t.
J
.
A
d
v
.
2
0
1
4
;
3
(1
1
):
17
-
23
.
[1
5
]
Oth
m
a
n
M
F
,
Ya
u
T
M
S
.
Co
mp
a
ri
so
n
o
f
Di
ff
e
re
n
t
Cla
ss
if
ic
a
ti
o
n
T
e
c
h
n
iq
u
e
s
Us
in
g
W
EKA
f
o
r
Bre
a
st
Ca
n
c
e
r
.
3
rd
Ku
a
la L
u
m
p
u
r
In
ter
n
a
ti
o
n
a
l
Co
n
f
e
re
n
c
e
o
n
Bio
m
e
d
ica
l
En
g
in
e
e
rin
g
.
2
0
0
6
:
5
2
0
-
5
2
3
.
[1
6
]
Ka
u
r
G
,
Ch
h
a
b
ra
A
.
Im
p
ro
v
e
d
J
4
8
Clas
sif
ica
ti
o
n
A
lg
o
rit
h
m
f
o
r
th
e
P
re
d
icti
o
n
o
f
Dia
b
e
tes
.
In
t.
J
.
Co
mp
u
t.
A
p
p
l.
2
0
1
4
;
9
8
(
2
2
)
:
13
–
1
7
.
[1
7
]
W
it
ten
IH
a
n
d
F
ra
n
k
E.
Da
ta
M
in
in
g
:
P
ra
c
ti
c
a
l
m
a
c
h
in
e
lea
rn
in
g
to
o
ls
a
n
d
tec
h
n
i
q
u
e
s.S
a
n
F
ra
n
c
isc
o
,
Ca
li
f
o
rn
ia:
M
o
rg
a
n
Ka
u
fm
a
n
.
2
0
0
5
.
[1
8
]
Jia
ji
a
Z,
Xu
a
n
W
,
L
in
Y,
S
o
n
g
W
.
Bu
il
d
i
n
g
Op
p
o
n
e
n
t
M
o
d
e
l
in
Im
p
e
r
f
e
c
t
In
f
o
r
m
a
ti
o
n
Bo
a
rd
G
a
m
e
s.
T
EL
KOM
NIKA
In
d
o
n
e
si
a
n
J
o
u
rn
a
l
o
f
E
lec
trica
l
En
g
in
e
e
rin
g
.
2
0
1
4
;
1
2
(3
)
:
1
9
7
5
-
1
9
8
6
.
[1
9
]
Ra
sh
e
d
M
,
Din
g
ju
Z.
A
n
Eff
ici
e
n
t
L
ip
-
re
a
d
in
g
M
e
th
o
d
Us
in
g
K
-
n
e
a
re
st
Ne
i
g
h
b
o
r
A
lg
o
rit
h
m
.
T
EL
KOM
NIKA
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
.
2
0
1
5
;
1
3
(
1
):
1
8
0
-
1
8
6
.
[2
0
]
S
a
d
iq
H,
Ne
a
m
a
A
D,
F
a
d
l
M
B
-
A
,
Na
jo
u
a
R.
Ed
u
c
a
ti
o
n
a
l
Da
t
a
M
in
in
g
a
n
d
A
n
a
ly
sis
o
f
S
tu
d
e
n
ts’
A
c
a
d
e
m
ic
P
e
rf
o
rm
a
n
c
e
u
sin
g
W
EK
A
.
T
EL
KOM
NIKA
In
d
o
n
e
sia
n
J
o
u
r
n
a
l
o
f
El
e
c
trica
l
En
g
i
n
e
e
rin
g
.
2
0
1
5
;
1
3
(1
):
1
8
0
-
1
8
6
.
Evaluation Warning : The document was created with Spire.PDF for Python.