I
AE
S In
t
er
na
t
io
na
l J
o
urna
l o
f
Art
if
icia
l In
t
ellig
ence
(
I
J
-
AI
)
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
26
,
p
p
.
2459
~
2
4
7
0
I
SS
N:
2
2
5
2
-
8
9
3
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijai.v
15
.i
3
.
p
p
2
4
5
9
-
2
4
7
0
2459
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
O
ptimi
zed
cla
ss
ifi
ca
tion o
f
stu
dent
performa
nce ou
tc
o
mes
using
LE
E feature
selec
tion in
the
co
n
te
x
t
o
f
ed
uca
tiona
l d
a
ta mining
K
is
h
o
r
e
K
u
m
a
r
K
a
ma
r
a
j
u
g
a
dd
a
1
,
M
o
v
v
a
P
a
v
a
n
i
2
,
R
a
ni
Va
n
a
t
hi
G
ur
us
a
my
3
,
Na
g
a
r
a
j
a
n
K
a
rt
hi
k
e
y
a
n
4
,
P
a
v
a
n K
um
a
r
Nidu
m
o
lu
5
,
Desid
i N
a
rsim
ha
Red
dy
6
,
M
u
nia
pp
a
n Ra
m
a
ra
j
7
,
Ra
j
a
s
eka
ra
n Nit
hy
a
8
1
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
o
n
i
c
s a
n
d
C
o
m
mu
n
i
c
a
t
i
o
n
En
g
i
n
e
e
r
i
n
g
,
I
C
F
A
I
F
o
u
n
d
a
t
i
o
n
f
o
r
H
i
g
h
e
r
Ed
u
c
a
t
i
o
n
,
H
y
d
e
r
a
b
a
d
,
I
n
d
i
a
2
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
o
n
i
c
s a
n
d
C
o
m
mu
n
i
c
a
t
i
o
n
En
g
i
n
e
e
r
i
n
g
,
N
a
l
l
a
M
a
l
l
a
R
e
d
d
y
En
g
i
n
e
e
r
i
n
g
C
o
l
l
e
g
e
,
H
y
d
e
r
a
b
a
d
,
I
n
d
i
a
3
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
R
a
mak
r
i
sh
n
a
C
o
l
l
e
g
e
o
f
A
r
t
s
a
n
d
S
c
i
e
n
c
e
f
o
r
W
o
m
e
n
,
C
o
i
mb
a
t
o
r
e
,
I
n
d
i
a
4
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
r
i
n
g
,
K
r
i
st
u
Ja
y
a
n
t
i
C
o
l
l
e
g
e
,
B
e
n
g
a
l
u
r
u
,
I
n
d
i
a
5
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
r
i
n
g
,
K
L
U
n
i
v
e
r
s
i
t
y
,
A
n
d
h
r
a
P
r
a
d
e
s
h
,
I
n
d
i
a
6
D
a
t
a
C
o
n
su
l
t
a
n
t
,
S
o
n
i
k
s
C
o
n
s
u
l
t
i
n
g
LLC
,
D
a
l
l
a
s
,
U
n
i
t
e
d
S
t
a
t
e
s
7
S
c
h
o
o
l
o
f
Q
u
a
n
t
u
m S
c
i
e
n
c
e
,
C
o
m
p
u
t
i
n
g
a
n
d
AI
,
R
a
t
h
i
n
a
m
G
l
o
b
a
l
D
e
e
m
e
d
t
o
b
e
U
n
i
v
e
r
s
i
t
y
,
C
o
i
m
b
a
t
o
r
e
,
I
n
d
i
a
8
S
c
h
o
o
l
o
f
C
o
m
p
u
t
i
n
g
S
c
i
e
n
c
e
,
K
P
R
C
o
l
l
e
g
e
o
f
A
r
t
s S
c
i
e
n
c
e
a
n
d
R
e
s
e
a
r
c
h
,
C
o
i
m
b
a
t
o
r
e
,
I
n
d
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Oct
23
,
2
0
2
5
R
ev
is
ed
Ap
r
24
,
2
0
2
6
Acc
ep
ted
May
11
,
2
0
2
6
S
tu
d
e
n
t
sp
e
c
u
lati
v
e
v
icto
r
y
is
a
v
it
a
l
a
re
a
th
a
t
n
e
e
d
s
to
b
e
p
r
e
d
icte
d
to
imp
ro
v
e
t
h
e
q
u
a
li
ty
o
f
e
d
u
c
a
ti
o
n
a
n
d
a
id
th
e
i
n
stit
u
ti
o
n
a
l
d
e
c
isio
n
m
a
k
in
g
.
Th
is
re
se
a
rc
h
wo
r
k
h
a
s
t
o
p
lan
n
e
d
t
o
u
se
lea
rn
i
n
g
b
a
se
d
e
n
h
a
n
c
e
d
e
v
a
lu
a
ti
o
n
(LE
E)
fe
a
tu
re
se
lec
ti
o
n
m
e
th
o
d
with
re
a
l
wo
rl
d
e
d
u
c
a
ti
o
n
a
l
d
a
tas
e
ts
fo
r
o
p
ti
m
ize
d
d
a
ta
m
in
i
n
g
a
p
p
r
o
a
c
h
t
o
p
re
d
ict
stu
d
e
n
t
p
e
rfo
rm
a
n
c
e
.
Hig
h
d
ime
n
sio
n
a
l
it
y
a
n
d
irrele
v
a
n
t
fe
a
tu
re
s
a
re
c
o
m
m
o
n
p
r
o
b
lem
s
wit
h
e
n
h
a
n
c
e
d
m
o
d
e
ls,
a
ffe
c
ti
n
g
c
las
sifica
ti
o
n
a
c
c
u
ra
c
y
a
n
d
e
fficie
n
c
y
.
LE
E
f
e
a
tu
re
a
lg
o
rit
h
m
is
u
se
d
t
o
e
x
t
ra
c
t
imp
o
r
tan
t
fe
a
tu
re
s,
th
a
t
e
n
h
a
n
c
e
t
h
e
p
e
r
fo
rm
a
n
c
e
o
f
th
e
m
o
d
e
l,
re
d
u
c
e
t
h
e
c
a
lcu
lati
o
n
q
u
a
n
ti
ty
o
f
th
e
m
o
d
e
l.
T
h
e
m
e
th
o
d
o
l
o
g
y
c
o
n
sists
o
f
p
re
-
p
ro
c
e
ss
in
g
o
f
t
h
e
d
a
tas
e
t,
fe
a
tu
re
se
lec
ti
o
n
u
sin
g
LE
E
a
lg
o
rit
h
m
,
a
n
d
tes
ti
n
g
fo
u
r
c
las
sifiers
n
a
m
e
ly
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
(S
VM),
k
-
n
e
a
re
st
n
e
ig
h
b
o
r
(KN
N),
a
d
a
p
ti
v
e
lea
rn
in
g
,
a
n
d
n
a
ïv
e
Ba
y
e
s.
Th
e
in
c
o
rp
o
ra
ti
o
n
o
f
LE
E
imp
r
o
v
e
s
th
e
m
o
d
e
l’s
a
b
il
it
y
b
y
re
d
u
c
in
g
n
o
ise
a
n
d
h
ig
h
li
g
h
ti
n
g
t
h
e
in
f
lu
e
n
t
ial
fe
a
tu
re
s.
Ex
p
e
rime
n
tal
re
su
l
ts
s
h
o
w
th
a
t
o
p
ti
m
ize
d
tec
h
n
i
q
u
e
s
a
re
b
e
tt
e
r
in
term
s
o
f
a
c
c
u
ra
c
y
a
n
d
r
o
b
u
st
n
e
ss
th
a
n
o
th
e
rs.
Th
e
m
o
d
e
ls
a
re
e
v
a
lu
a
ted
b
a
se
d
o
n
imp
o
rta
n
t
p
e
rfo
rm
a
n
c
e
m
e
tri
c
s
su
c
h
a
s
a
c
c
u
ra
c
y
,
p
re
c
isio
n
,
re
c
a
ll
,
F
1
-
sc
o
re
,
a
n
d
train
i
n
g
t
ime
.
Th
e
e
n
h
a
n
c
e
d
a
p
p
ro
a
c
h
will
h
e
lp
t
o
a
d
d
t
o
t
h
e
li
tera
tu
re
o
f
th
e
field
o
f
e
d
u
c
a
ti
o
n
a
l
d
a
ta
m
in
in
g
(EDM
)
,
p
ro
v
id
i
n
g
a
p
ra
c
ti
c
a
l
a
n
d
e
ffe
c
ti
v
e
wa
y
o
f
p
re
d
ictin
g
st
u
d
e
n
t
p
e
rfo
rm
a
n
c
e
in
re
a
l
a
c
a
d
e
m
ic se
tt
in
g
s.
K
ey
w
o
r
d
s
:
C
las
s
if
icatio
n
m
eth
o
d
s
E
d
u
ca
tio
n
al
d
ata
m
in
in
g
Op
tim
izatio
n
alg
o
r
ith
m
s
Per
f
o
r
m
an
ce
m
ea
s
u
r
es
Stu
d
en
t d
ataset
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Mu
n
iap
p
an
R
am
ar
aj
Sch
o
o
l o
f
Q
u
an
tu
m
Scien
ce
,
C
o
m
p
u
tin
g
an
d
AI
,
R
ath
in
am
Glo
b
al
Dee
m
ed
to
b
e
Un
iv
er
s
i
ty
C
o
im
b
ato
r
e,
T
am
il
N
ad
u
,
I
n
d
i
a
E
m
ail: r
am
ar
aj.
p
h
d
cs@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
A
n
ew
em
p
h
asis
o
f
ed
u
ca
tio
n
al
d
ata
m
in
in
g
(
E
DM
)
h
as
b
ee
n
th
e
a
b
ilit
y
to
en
v
is
ag
e
ac
ad
em
ic
p
er
f
o
r
m
an
ce
o
f
s
tu
d
e
n
ts
.
T
o
f
r
u
s
tr
ate
th
is
,
f
ea
tu
r
e
s
elec
tio
n
tech
n
iq
u
es
p
lay
a
n
im
p
o
r
tan
t
r
o
le
in
f
in
d
in
g
t
h
e
attr
ib
u
tes
o
f
u
tm
o
s
t
im
p
o
r
tan
ce
to
af
f
ec
t
s
tu
d
en
ts
’
p
er
f
o
r
m
an
ce
[
1
]
.
T
h
is
s
tu
d
y
p
r
o
p
o
s
es
an
o
p
tim
ized
d
ata
m
in
in
g
ap
p
r
o
ac
h
,
wh
ich
u
s
es
th
e
lear
n
in
g
b
ased
en
h
an
ce
d
ev
alu
atio
n
(
L
E
E
)
f
ea
tu
r
e
s
elec
tio
n
alg
o
r
ith
m
to
en
h
an
ce
th
e
p
r
e
d
ictiv
e
m
o
d
el
ac
cu
r
ac
y
an
d
ef
f
icien
cy
.
T
h
e
p
r
o
p
o
s
ed
f
r
am
ewo
r
k
,
wh
ic
h
u
s
es
L
E
E
to
r
ea
l
wo
r
ld
e
d
u
ca
tio
n
al
d
ata,
d
is
ca
r
d
s
r
ed
u
n
d
an
t
an
d
u
n
in
f
o
r
m
ati
v
e
f
ea
tu
r
es
an
d
allo
ws
m
o
r
e
a
cc
u
r
ate
p
r
ed
ictio
n
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
26
:
2
4
5
9
-
2
4
7
0
2460
[
2
]
.
T
h
ese
s
elec
ted
f
ea
tu
r
es
ar
e
th
en
ap
p
lied
in
tr
ai
n
in
g
a
n
d
test
in
g
s
ev
er
al
class
if
icatio
n
alg
o
r
ith
m
s
s
u
ch
as
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM
)
,
k
-
n
ea
r
est
n
eig
h
b
o
r
(
KNN)
,
ad
ap
tiv
e
lear
n
i
n
g
,
a
n
d
n
aï
v
e
B
ay
es
class
if
icatio
n
wh
ich
ar
e
well
estab
lis
h
ed
i
n
class
if
icatio
n
p
r
o
b
lem
s
[
3
]
.
T
h
e
tr
ad
itio
n
al
m
o
d
els
ar
e
o
f
ten
f
o
u
n
d
to
b
e
s
u
b
-
o
p
tim
al
in
ca
p
tu
r
i
n
g
th
e
c
o
m
p
lex
n
o
n
lin
ea
r
r
elatio
n
s
h
ip
in
en
v
ir
o
n
m
e
n
tal
d
ata
an
d
th
u
s
in
p
r
ed
ictin
g
th
e
d
ata
with
an
ac
ce
p
tab
le
ac
c
u
r
a
cy
[
4
]
.
W
h
ile
th
e
p
o
ten
tial o
f
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
to
p
r
ed
ict
tu
n
n
el
g
r
o
u
n
d
b
eh
a
v
io
r
h
as b
ee
n
s
h
o
wn
in
r
ec
en
t
s
tu
d
ies,
th
er
e
ar
e
s
till
s
ev
er
al
k
ey
ch
allen
g
es
[
5
]
.
T
h
er
e
ar
e
v
er
y
f
ew
p
u
b
lic
ly
av
ailab
le
lar
g
e
d
atasets
o
f
lab
elled
,
lab
eled
d
ata
wh
ich
s
u
p
p
o
r
t
lo
n
g
-
ter
m
t
u
n
n
el
p
e
r
f
o
r
m
an
ce
an
d
o
n
ly
a
lim
ited
n
u
m
b
e
r
o
f
em
p
ir
ical
m
o
d
els
tr
ai
n
ed
o
n
s
m
all
s
ite
-
s
p
ec
if
ic
d
ata
s
ets
wh
ich
d
o
n
o
t
h
av
e
t
h
e
g
e
n
er
ality
n
ec
ess
ar
y
f
o
r
u
s
e
ac
r
o
s
s
a
r
an
g
e
o
f
g
eo
lo
g
ical
an
d
ex
ca
v
atio
n
co
n
d
itio
n
s
.
T
h
is
v
ar
iety
o
f
s
tu
d
en
t
d
if
f
er
e
n
ce
in
lear
n
in
g
s
ty
les
is
an
im
p
o
r
ta
n
t
ch
allen
g
e
wh
en
d
esig
n
in
g
ef
f
ec
tiv
e
in
s
tr
u
ctio
n
al
s
tr
ateg
ies
[
6
]
.
T
h
e
v
is
u
al,
a
u
d
ito
r
y
,
k
in
esth
etic
(
VAK)
m
o
d
el
g
iv
es
a
th
eo
r
etica
l
b
asis
f
o
r
class
if
icatio
n
o
f
l
ea
r
n
er
s
,
b
u
t
co
n
v
en
tio
n
al
ass
ess
m
en
t
m
eth
o
d
s
ar
e
b
ased
o
n
s
u
b
jectiv
e
q
u
esti
o
n
n
air
es,
m
an
u
al
ass
ess
m
en
t
an
d
r
eq
u
ir
e
m
u
ch
tim
e
an
d
a
r
e
n
o
t
ac
cu
r
ate
en
o
u
g
h
[
7
]
.
B
u
t
th
e
p
r
ev
io
u
s
s
tu
d
y
lack
s
o
f
s
tr
o
n
g
p
r
ed
ictiv
e
m
o
d
els
wh
ich
ca
n
class
if
y
lear
n
in
g
ac
tiv
ities
in
to
lear
n
in
g
s
ty
le
ca
teg
o
r
ies
with
h
ig
h
ac
cu
r
ac
y
with
o
u
t
f
ac
in
g
th
e
p
r
o
b
lem
o
f
“f
ea
tu
r
e
s
elec
tio
n
,
”
“c
lass
im
b
alan
ce
,
”
an
d
“o
v
er
f
itti
n
g
”
.
Aca
d
em
ic
p
r
ess
u
r
e,
ir
r
e
g
u
lar
life
an
d
p
s
y
ch
o
lo
g
ical
p
r
o
b
lem
s
ar
e
o
f
ten
en
co
u
n
ter
e
d
b
y
u
n
i
v
er
s
ity
s
tu
d
en
ts
th
at
h
av
e
a
g
r
ea
t im
p
ac
t
o
n
th
eir
s
u
b
jectiv
e
well
-
b
ein
g
(
SW
B
)
[
8
]
.
An
im
p
o
r
tan
t
c
h
allen
g
e
in
c
r
e
atin
g
ac
cu
r
ate
p
r
ed
ictiv
e
m
o
d
els
is
th
e
f
ac
t
th
at
s
tu
d
en
t
p
er
f
o
r
m
an
ce
d
ata
s
ets
ar
e
o
f
ten
im
b
alan
ce
d
,
with
a
m
ajo
r
ity
o
f
s
tu
d
en
ts
t
y
p
ically
in
th
e
“a
v
er
ag
e”
o
r
“
p
ass
”
class
es,
an
d
a
m
in
o
r
ity
o
f
s
tu
d
en
ts
in
class
es
s
u
ch
as
“e
x
ce
llen
t”
o
r
“f
ail”
[
9
]
.
T
h
is
im
b
alan
ce
c
r
ea
tes
an
o
p
p
o
r
tu
n
ity
f
o
r
b
iased
m
o
d
el
tr
ain
in
g
,
ca
u
s
in
g
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
to
b
e
m
o
r
e
ef
f
ec
tiv
e
at
d
etec
tin
g
th
e
m
ajo
r
ity
class
an
d
lead
in
g
to
in
ac
c
u
r
at
e
d
etec
tio
n
an
d
m
is
class
if
icat
io
n
o
f
s
tu
d
e
n
ts
at
r
is
k
o
f
ac
a
d
em
ic
f
ailu
r
e.
As
a
r
esu
lt,
th
e
m
o
d
els’
p
r
ed
ictiv
e
r
eliab
ilit
y
,
f
air
n
ess
,
an
d
g
e
n
er
a
lizab
ilit
y
is
co
n
s
id
er
ab
ly
h
am
p
er
ed
[
1
0
]
.
2.
L
I
T
E
R
AT
U
RE
R
E
VI
E
W
T
h
e
r
ec
e
n
t
r
esear
ch
ar
ea
is
a
g
r
o
win
g
ar
ea
o
f
r
esear
ch
to
g
ain
in
s
ig
h
ts
f
r
o
m
la
r
g
e
-
s
ca
le
e
d
u
ca
tio
n
al
d
ata,
wh
ich
ca
n
b
e
u
s
ed
to
im
p
r
o
v
e
s
tu
d
en
t
o
u
tco
m
es
an
d
t
o
g
u
id
e
d
ec
is
io
n
m
a
k
in
g
.
Ma
n
y
m
ac
h
in
e
lear
n
in
g
an
d
d
ata
m
in
in
g
alg
o
r
ith
m
s
h
av
e
b
ee
n
u
s
ed
t
o
p
r
e
d
ict
s
tu
d
en
t
p
er
f
o
r
m
an
ce
in
cl
u
d
in
g
S
VM
,
d
ec
is
io
n
tr
ee
s
(
DT
)
,
KNN,
n
aïv
e
B
ay
es,
an
d
en
s
em
b
le
m
o
d
els.
C
h
en
et
a
l.
[
1
1
]
h
as
s
u
b
s
id
ized
to
th
e
ex
is
tin
g
b
o
d
y
o
f
k
n
o
wled
g
e
b
y
r
ev
iewin
g
th
e
a
p
p
licatio
n
o
f
m
ac
h
in
e
lear
n
in
g
tech
n
iq
u
es
in
h
ig
h
er
ed
u
ca
tio
n
with
a
f
o
cu
s
o
n
p
r
ed
ictin
g
s
tu
d
en
ts
’
p
er
f
o
r
m
a
n
ce
,
lear
n
in
g
en
g
a
g
em
en
t,
an
d
s
elf
-
ef
f
icac
y
.
B
ilal
et
a
l.
[
1
2
]
h
ad
illu
s
tr
ated
with
a
m
eth
o
d
ical
p
r
ep
r
o
ce
s
s
in
g
m
eth
o
d
to
o
v
e
r
co
m
e
s
p
ellin
g
v
ar
iatio
n
s
,
tex
tu
al
n
o
is
e
an
d
lack
o
f
g
r
am
m
atica
l
u
n
if
o
r
m
ity
in
R
o
m
a
n
-
Ur
d
u
o
p
in
io
n
s
.
Gu
o
et
a
l.
[
1
3
]
h
as
in
d
is
tin
g
u
is
h
ab
le
m
o
m
en
to
u
s
o
f
th
is
s
tu
d
y
th
at
p
r
o
v
id
es
lar
g
e
-
s
ca
le,
p
r
o
s
p
ec
ti
v
e
an
d
m
u
lticen
ter
co
h
o
r
t
ev
i
d
en
ce
o
n
t
h
e
h
ea
lth
,
b
e
h
av
io
r
al
an
d
p
s
y
ch
o
s
o
cial
d
ev
elo
p
m
e
n
t
o
f
C
h
in
ese
ad
o
l
escen
ts
[
1
3
]
.
Sid
d
iq
u
a
et
a
l.
[
1
4
]
h
as
in
tr
o
d
u
ce
d
with
n
e
w
en
h
an
ce
s
to
th
e
ar
tific
ial
in
tellig
en
ce
(
AI
)
f
i
eld
o
f
m
en
tal
h
ea
lth
,
p
r
o
p
o
s
in
g
AI
DA
—
a
n
o
v
el
AI
-
b
as
ed
f
r
am
ew
o
r
k
f
o
r
ass
es
s
in
g
d
ep
r
ess
io
n
in
th
e
co
n
tex
t
o
f
B
an
g
lad
esh
i
s
tu
d
en
ts
.
T
h
e
p
r
o
p
o
s
ed
m
eth
o
d
s
with
ass
o
ciatio
n
s
o
f
s
o
cially
r
elev
an
t f
ac
to
r
s
a
n
d
v
alid
ated
p
s
y
ch
o
lo
g
ical
m
etr
ics to
im
p
r
o
v
e
co
n
tex
tu
al
ac
cu
r
a
cy
.
Po
u
r
h
o
s
s
ein
i
et
a
l.
[
1
5
]
h
as
s
ig
n
if
ican
t
n
o
v
elty
o
f
th
is
wo
r
k
is
th
e
n
o
v
el
h
y
b
r
id
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
NHSVM
)
f
r
am
ewo
r
k
h
av
e
b
ee
n
estab
lis
h
ed
an
d
o
p
tim
ized
f
o
r
ac
cu
r
ate
p
r
ed
icti
o
n
o
f
to
tal
d
is
s
o
lv
ed
s
o
lid
s
(
T
DS)
in
wate
r
s
y
s
tem
s
.
T
h
e
an
ticip
ated
h
y
b
r
id
SVM
ap
p
r
o
ac
h
c
o
m
b
in
es
t
h
e
s
tr
en
g
th
s
o
f
co
n
v
en
tio
n
a
l
p
r
ed
ictiv
e
m
o
d
els
with
o
p
tim
i
za
tio
n
tech
n
iq
u
es
a
n
d
i
n
tellig
en
t
lear
n
in
g
m
ec
h
an
is
m
s
,
b
o
o
s
tin
g
th
e
r
o
b
u
s
tn
ess
an
d
p
r
e
d
ictiv
e
ac
cu
r
ac
y
o
f
t
h
e
m
o
d
els.
Dash
tg
o
li
et
a
l.
[
1
6
]
h
ad
illu
s
tr
ate
o
f
th
is
r
e
s
ea
r
ch
is
b
ased
o
n
d
ata
-
d
r
iv
e
n
m
eth
o
d
s
to
ac
cu
r
a
tely
an
d
ef
f
icien
tly
p
r
ed
ict
u
s
i
n
g
co
n
v
e
n
tio
n
al
n
u
m
er
ical
o
r
em
p
ir
ical
m
eth
o
d
s
,
wh
ich
d
em
an
d
s
co
m
p
lex
ass
u
m
p
tio
n
s
,
h
ig
h
c
o
m
p
u
tatio
n
al
c
o
s
t,
an
d
lim
ited
ad
ap
tab
ilit
y
to
th
e
h
eter
o
g
e
n
eo
u
s
s
o
il
co
n
d
itio
n
s
.
So
n
g
et
a
l.
[
1
7
]
h
as
a
co
n
tex
t
f
o
r
m
u
ltid
im
en
s
io
n
al
in
v
esti
g
atio
n
is
o
b
tain
ab
le,
co
alescin
g
id
io
s
y
n
cr
atic
r
ea
ctio
n
an
d
o
b
j
ec
tiv
e
en
v
ir
o
n
m
e
n
tal
p
ar
am
et
er
s
,
to
p
r
ed
ict
th
er
m
al
co
m
f
o
r
t
in
lib
r
ar
y
s
p
ac
es
wh
er
e
u
s
er
s
atis
f
ac
tio
n
an
d
co
n
ce
n
tr
atio
n
a
r
e
h
ig
h
ly
s
en
s
itiv
e
to
m
icr
o
clim
atic
co
n
d
itio
n
s
.
T
h
e
p
r
im
ar
y
in
n
o
v
atio
n
o
f
th
is
wo
r
k
is
th
e
d
esig
n
o
f
a
m
o
d
i
f
ied
a
n
d
ex
p
licab
le
ass
ess
m
en
t
an
d
en
h
a
n
ce
m
en
t
o
f
m
eth
o
d
o
l
o
g
y
f
o
r
n
eu
r
o
s
u
r
g
ic
al
s
k
ills
b
ased
o
n
v
ir
t
u
al
r
ea
lity
(
VR
)
s
im
u
latio
n
s
[
1
8
]
.
It
s
s
u
g
g
es
ti
o
n
s
w
er
e
th
o
r
o
u
g
h
r
e
v
i
ew
o
f
t
h
e
c
u
r
r
en
t
m
et
h
o
d
o
lo
g
i
es
i
n
E
E
G
-
b
ase
d
s
tr
ess
ass
ess
m
e
n
t
,
wit
h
a
f
o
cu
s
o
n
tr
en
d
s
,
p
r
o
g
r
ess
,
an
d
t
h
e
e
f
f
ic
ac
y
o
f
d
i
f
f
er
e
n
t s
ig
n
al
p
r
o
ce
s
s
in
g
a
n
d
m
ac
h
i
n
e
l
ea
r
n
i
n
g
[
1
9
]
.
Mu
m
en
in
et
a
l.
[
2
0
]
h
a
d
illu
s
tr
ate
o
n
th
e
ex
is
tin
g
m
eth
o
d
s
u
b
s
id
izes
to
th
e
f
ield
o
f
m
e
n
tal
h
ea
lth
ass
es
s
m
en
t
in
h
ig
h
er
ed
u
ca
tio
n
b
y
d
ev
el
o
p
in
g
a
n
ef
f
ec
tiv
e
an
d
d
ata
-
d
r
iv
en
ap
p
r
o
ac
h
f
o
r
ea
r
ly
d
etec
tio
n
o
f
d
ep
r
ess
io
n
am
o
n
g
u
n
iv
er
s
ity
s
tu
d
en
ts
.
Sh
o
aib
et
a
l.
[
2
1
]
h
as
illu
s
tr
ate
o
n
th
e
k
ey
r
o
le
is
in
u
n
itin
g
m
u
lti
-
attr
ib
u
te
an
al
y
s
is
with
th
e
d
ev
elo
p
m
en
t
o
f
ad
v
a
n
ce
d
p
r
e
d
ictiv
e
m
o
d
ellin
g
t
o
f
a
cilitate
th
e
co
r
r
ec
t
id
en
tific
atio
n
o
f
g
r
a
d
u
ates’
ca
r
ee
r
ch
an
g
e,
jo
b
s
tab
ilit
y
an
d
s
ec
to
r
al
m
o
b
ilit
y
.
T
h
e
s
tu
d
y
p
r
esen
ts
a
s
y
s
tem
atic
co
m
b
in
atio
n
o
f
s
ev
er
al
p
r
ed
ic
tiv
e
alg
o
r
ith
m
s
s
u
ch
as
SVM,
KNN,
n
aïv
e
B
ay
es
an
d
e
n
s
em
b
le
m
o
d
els,
wh
ich
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
Op
timiz
ed
cla
s
s
if
ica
tio
n
o
f stu
d
en
t p
erfo
r
ma
n
ce
o
u
tco
mes u
s
in
g
LEE
…
(
K
a
ma
r
a
ju
g
a
d
d
a
K
is
h
o
r
e
K
u
ma
r
)
2461
ac
co
u
n
ts
f
o
r
c
o
m
p
lex
p
atter
n
s
in
e
d
u
ca
tio
n
al
d
ata
[
2
2
]
.
T
h
e
alg
o
r
ith
m
’
s
ap
p
lica
b
ilit
y
in
r
ea
l
-
wo
r
ld
ac
a
d
em
ic
en
v
ir
o
n
m
en
ts
is
illu
s
tr
ated
b
y
th
e
wo
r
k
o
f
Kh
an
et
a
l.
[
2
3
]
wh
ich
s
h
o
ws
h
o
w
th
e
alg
o
r
ith
m
ca
n
b
e
ap
p
lied
in
a
v
ar
iety
o
f
ac
ad
em
ic
co
n
tex
ts
,
m
ak
in
g
it
a
p
o
we
r
f
u
l
a
n
d
s
ca
lab
le
to
o
l
f
o
r
ed
u
ca
tio
n
al
in
s
titu
tio
n
s
to
im
p
r
o
v
e
s
tu
d
en
t
s
u
cc
ess
an
d
r
eten
tio
n
.
W
ah
ee
d
et
a
l.
[
2
4
]
h
ad
s
h
o
w
n
h
o
w
well
n
eu
r
al
n
etwo
r
k
a
r
ch
itectu
r
es
h
an
d
le
co
m
p
lex
,
n
o
n
-
lin
ea
r
r
elatio
n
s
h
ip
s
in
s
id
e
ed
u
ca
tio
n
al
d
ata,
p
r
o
v
id
i
n
g
a
s
ca
lab
le
an
d
f
lex
ib
le
ap
p
r
o
ac
h
f
o
r
a
v
ar
iety
o
f
e
-
lear
n
in
g
p
latf
o
r
m
s
.
T
h
e
wo
r
k
p
r
esen
te
d
in
t
h
is
co
n
tr
ib
u
tio
n
h
ig
h
lig
h
ts
th
e
p
r
o
m
is
e
o
f
in
tellig
en
t
m
o
d
els
th
at
ca
n
y
ield
h
ig
h
e
r
q
u
ality
p
er
s
o
n
alize
d
lear
n
in
g
a
n
d
b
etter
en
ab
le
ed
u
ca
ti
o
n
al
s
u
p
p
o
r
t
in
s
elf
-
p
ac
ed
lear
n
in
g
en
v
ir
o
n
m
en
t.
I
s
m
ail
an
d
Yu
s
o
f
[
2
5
]
h
a
d
class
if
ies
d
if
f
e
r
en
t
d
ata
p
r
ep
r
o
ce
s
s
in
g
tech
n
iq
u
es,
f
ea
tu
r
e
s
elec
tio
n
tech
n
iq
u
es,
an
d
m
o
d
el
o
p
tim
izatio
n
tech
n
iq
u
es
u
s
ed
in
s
cien
ce
,
tech
n
o
lo
g
y
,
en
g
in
ee
r
in
g
,
a
n
d
m
ath
em
atics
(
STE
M
)
ed
u
ca
tio
n
d
atasets
an
d
p
r
o
v
id
es
in
s
i
g
h
tf
u
l
i
n
f
o
r
m
atio
n
a
b
o
u
t
f
ac
t
o
r
s
th
at
af
f
ec
t
th
e
p
r
ed
ictio
n
ac
c
u
r
ac
y
.
On
e
o
f
th
e
m
ain
d
r
awb
ac
k
s
id
en
tifie
d
i
n
p
r
e
v
io
u
s
r
esear
ch
is
in
clu
s
io
n
o
f
ir
r
elev
an
t
an
d
r
ed
u
n
d
an
t f
ea
tu
r
es in
ed
u
ca
tio
n
d
atasets
,
af
f
ec
tin
g
th
e
e
f
f
icie
n
cy
o
f
th
e
m
o
d
els an
d
th
e
co
m
p
u
tatio
n
al
co
s
t.
3.
M
E
T
H
O
DS
T
h
e
an
ticip
ated
ap
p
r
o
ac
h
s
y
n
d
icate
s
with
en
h
a
n
ce
d
d
ata
m
in
in
g
f
r
am
ewo
r
k
with
f
ea
tu
r
e
s
elec
tio
n
f
o
r
im
p
r
o
v
in
g
th
e
p
r
e
d
ictio
n
o
f
s
tu
d
en
t
p
er
f
o
r
m
an
ce
.
First,
th
e
r
aw
teac
h
in
g
d
at
a
is
co
llected
an
d
p
r
e
-
p
r
o
ce
s
s
ed
in
o
r
d
er
t
o
g
u
a
r
an
tee
th
e
q
u
ality
an
d
co
n
s
is
ten
cy
o
f
th
e
d
ata.
I
t
co
n
s
id
er
s
with
cu
t
d
o
wn
th
e
d
im
en
s
io
n
ality
an
d
to
r
e
m
o
v
e
ir
r
elev
an
t
attr
ib
u
tes,
L
E
E
f
ea
tu
r
e
s
elec
tio
n
alg
o
r
ith
m
is
u
s
ed
wh
ich
is
u
s
ed
to
id
en
tify
th
e
m
o
s
t
im
p
o
r
ta
n
t
p
r
ed
icto
r
s
lead
i
n
g
to
t
h
e
s
tu
d
en
t’
s
ac
ad
em
ic
o
u
tco
m
es
s
y
s
tem
atica
lly
an
d
s
y
s
tem
atica
lly
.
E
n
h
an
ce
d
L
E
E
f
ea
tu
r
e
s
elec
tio
n
alg
o
r
ith
m
is
o
n
e
o
f
th
e
en
h
a
n
ce
d
v
e
r
s
io
n
s
o
f
th
e
tr
ad
itio
n
a
l
L
E
E
m
eth
o
d
th
at
c
o
n
s
id
er
ed
to
r
ec
o
g
n
ize
th
e
m
o
s
t
p
er
tin
e
n
t
s
tr
u
ctu
r
es
wh
ile
p
lu
m
m
etin
g
r
ed
u
n
d
a
n
cy
an
d
co
m
p
u
tatio
n
al
c
o
s
t.
I
t
in
v
o
lv
e
s
an
iter
ativ
e
p
r
o
ce
s
s
o
f
lear
n
i
n
g
,
ev
alu
atio
n
,
an
d
elim
in
atio
n
an
d
is
well
s
u
ited
to
ed
u
ca
tio
n
al
d
ata
s
ets
with
a
n
u
m
b
e
r
o
f
ac
a
d
em
ic,
d
e
m
o
g
r
ap
h
ic,
an
d
b
eh
a
v
io
r
al
at
tr
ib
u
tes.
I
t
co
n
tain
s
s
o
cio
-
ec
o
n
o
m
ic,
g
en
d
er
an
d
d
em
o
g
r
a
p
h
ic
f
ac
to
r
s
th
at
aid
in
d
is
ce
r
n
in
g
tr
e
n
d
s
an
d
d
is
p
ar
ities
at
th
e
p
o
p
u
latio
n
lev
el.
T
h
e
b
asic
d
ata
in
th
e
d
ataset
ar
e
u
s
u
ally
ac
ad
em
ic
attr
ib
u
tes,
s
u
ch
as
m
ar
k
s
in
th
e
in
te
r
n
al
ass
es
s
m
en
ts
,
m
ar
k
s
in
ass
ig
n
m
en
ts
,
atten
d
a
n
ce
p
er
ce
n
ta
g
e,
p
r
ev
io
u
s
s
em
ester
g
r
a
d
e
s
,
an
d
ex
am
in
atio
n
g
r
ad
es.
T
h
e
tar
g
et
v
ar
iab
le
is
ty
p
ically
a
n
u
m
er
ic
s
co
r
e
o
r
a
p
ass
/f
ail
m
ar
k
,
o
r
a
g
r
ad
e
c
ateg
o
r
y
(
ex
ce
llen
t,
g
o
o
d
,
a
v
er
ag
e
,
an
d
p
o
o
r
)
.
B
ef
o
r
e
th
e
m
o
d
el
is
tr
ain
ed
,
th
er
e
ar
e
p
r
ep
r
o
ce
s
s
in
g
s
tep
s
to
b
e
tak
en
f
o
r
th
e
d
ata
s
et,
in
clu
d
in
g
f
illi
n
g
in
m
is
s
i
n
g
d
ata,
n
o
r
m
alizin
g
th
e
d
ata,
en
co
d
in
g
ca
teg
o
r
ical
d
ata
an
d
f
ea
tu
r
e
s
elec
tio
n
(
s
u
ch
as m
o
d
if
ie
d
L
E
E
f
ea
t
u
r
e
s
elec
tio
n
)
.
3
.
1
.
O
pti
m
ized
lea
rning
ba
s
ed
enha
nced
ev
a
lua
t
io
n
f
ea
t
ure
s
elec
t
io
n wit
h SV
M
cla
s
s
if
ier
T
h
e
p
r
o
p
o
s
ed
wo
r
k
s
y
n
d
icate
s
th
e
o
p
tim
ized
L
E
E
f
ea
tu
r
e
s
elec
tio
n
with
SVM
class
if
ier
,
to
ac
h
iev
e
b
etter
ac
cu
r
ac
y
an
d
e
f
f
icien
cy
in
p
r
ed
ictin
g
s
tu
d
en
t
p
e
r
f
o
r
m
an
ce
.
T
h
e
o
p
tim
ized
f
ea
t
u
r
e
s
p
ac
e
is
th
en
f
ed
t
o
an
SVM
class
if
ier
wh
ich
f
o
r
m
s
a
d
ec
is
io
n
b
o
u
n
d
ar
y
th
at
s
ep
ar
ates
th
e
ca
teg
o
r
ies
o
f
s
tu
d
en
ts
’
p
er
f
o
r
m
an
ce
with
a
m
ax
im
u
m
m
ar
g
in
.
T
h
e
p
r
o
ce
s
s
o
f
o
p
tim
ized
L
E
E
f
ea
tu
r
e
s
elec
tio
n
with
SVM
cl
ass
if
ier
f
o
r
s
tu
d
en
t
p
er
f
o
r
m
an
ce
p
r
ed
ictio
n
(
SP
P)
is
s
h
o
wn
in
Fig
u
r
e
1
.
I
t star
ts
with
th
e
p
r
ep
r
o
ce
s
s
in
g
p
r
o
ce
d
u
r
es,
wh
ich
in
clu
d
e
th
e
im
p
u
tatio
n
o
f
m
is
s
in
g
v
al
u
es,
n
o
r
m
aliza
tio
n
,
an
d
en
c
o
d
in
g
.
T
h
e
ev
alu
atio
n
o
f
ea
ch
f
ea
tu
r
e
is
p
er
f
o
r
m
ed
b
y
L
E
E
f
u
n
ctio
n
th
at
co
m
b
in
es
f
ea
tu
r
e
r
elev
an
ce
with
f
ea
t
u
r
e
r
ed
u
n
d
a
n
cy
with
weig
h
t
p
ar
am
eter
s
α
an
d
β.
T
h
e
b
est f
ea
tu
r
e
s
u
b
s
et
is
th
en
ch
o
s
en
to
m
a
x
im
ize
th
e
d
is
cr
i
m
in
ativ
e
p
o
wer
.
T
h
is
s
im
p
lifie
d
r
ep
r
esen
tatio
n
is
g
iv
en
to
t
h
e
SVM
class
if
ier
,
an
d
k
er
n
el
f
u
n
ctio
n
s
ar
e
u
s
ed
to
ca
lcu
late
th
e
d
ec
is
io
n
b
o
u
n
d
ar
ies.
3
.
1
.
1
.
L
ea
rning
ba
s
ed
enha
n
ce
d e
v
a
lua
t
io
n
f
ea
t
ure
s
elec
t
io
n
L
E
E
f
ea
tu
r
e
s
elec
tio
n
tech
n
iq
u
e
is
u
s
ed
to
s
elec
t
im
p
o
r
tan
t
f
ea
tu
r
es
o
f
s
tu
d
en
t
d
ata
to
r
eta
in
o
n
ly
th
e
m
o
s
t in
f
o
r
m
ativ
e
p
r
ed
icto
r
s
f
o
r
p
er
f
o
r
m
an
ce
m
o
d
ellin
g
.
L
EE
(
)
=
×
(
,
)
−
×
(
,
)
(
1
)
∗
=
∑
(
)
∈
(
2
)
W
h
er
e
(
,
)
r
elev
an
ce
o
f
f
ea
tu
r
e
with
r
esp
ec
t
to
tar
g
et
(
m
ea
s
u
r
ed
u
s
in
g
m
u
tu
al
in
f
o
r
m
ati
o
n
o
r
co
r
r
elatio
n
)
.
(
,
)
r
ed
u
n
d
an
cy
o
f
w
ith
alr
ea
d
y
s
elec
ted
f
ea
tu
r
e
s
et
F.
,
weig
h
t
p
ar
am
eter
s
to
b
alan
ce
r
elev
an
ce
an
d
r
ed
u
n
d
an
cy
.
∗
r
ep
r
esen
ts
th
e
o
p
tim
ized
s
u
b
s
et
o
f
f
ea
tu
r
es.
3
.
1
.
2
.
O
ptim
ized
lea
rning
ba
s
ed
enha
nce
d e
v
a
lua
t
io
n ba
s
ed
SVM
cla
s
s
if
ier
Af
ter
f
ea
tu
r
e
s
elec
tio
n
,
th
e
r
e
f
in
ed
d
ataset
is
clas
s
if
ied
u
s
i
n
g
SVM.
SVM
aim
s
to
f
in
d
th
e
o
p
tim
al
h
y
p
er
p
lan
e
H
th
at
m
a
x
im
izes th
e
m
ar
g
in
b
etwe
en
class
es.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
26
:
2
4
5
9
-
2
4
7
0
2462
(
)
=
(
∑
,
,
(
,
)
+
)
(
3
)
Her
e,
ar
e
lag
r
a
n
g
e
m
u
ltip
lier
s
,
is
a
class
lab
el
v
alu
es,
(
,
)
is
th
e
k
er
n
el
f
u
n
ctio
n
,
b
is
th
e
b
ias t
er
m
.
Fig
u
r
e
1
.
Op
tim
ized
L
E
E
b
ase
d
s
u
p
p
o
r
t v
ec
to
r
m
ac
h
in
e
(
L
SVM)
alg
o
r
ith
m
u
s
ed
to
SPP
3
.
2
.
E
nh
a
nced
lea
rning
ba
s
e
d e
nh
a
nced
ev
a
lua
t
io
n
f
ea
t
ure
s
elec
t
io
n wit
h K
NN
T
h
e
f
ea
tu
r
es
ar
e
s
elec
ted
u
s
in
g
th
e
e
n
h
an
ce
d
L
E
E
f
ea
t
u
r
e
s
elec
tio
n
m
eth
o
d
an
d
class
if
ied
u
s
in
g
th
e
KNN
clas
s
if
ier
with
an
aim
o
f
en
h
an
ci
n
g
class
if
icatio
n
ac
cu
r
ac
y
b
y
elim
in
atin
g
ir
r
elev
an
t
an
d
r
ed
u
n
d
a
n
t
f
ea
tu
r
es.
T
h
e
o
p
tim
ized
L
E
E
f
ea
tu
r
e
s
elec
tio
n
an
d
th
e
KNN
r
u
le
f
o
r
p
r
ed
ictin
g
s
tu
d
e
n
t
p
er
f
o
r
m
an
ce
is
illu
s
tr
ated
as
in
Fig
u
r
e
2
.
Dat
a
p
r
ep
r
o
ce
s
s
in
g
s
tep
s
s
tar
t
wi
th
th
e
tr
ea
tm
en
t
o
f
m
is
s
in
g
d
ata,
n
o
r
m
aliza
tio
n
,
an
d
en
co
d
in
g
.
T
h
e
L
E
E
f
u
n
ctio
n
is
u
s
ed
f
o
r
ea
ch
f
ea
tu
r
e
to
d
eter
m
in
e
its
r
elev
an
ce
to
class
lab
els
an
d
m
in
im
ize
r
ed
u
n
d
an
c
y
.
T
h
e
o
p
tim
al
f
ea
tu
r
e
s
u
b
s
et
F*
is
th
e
o
n
e
t
h
at
m
ax
im
izes
th
e
v
alu
e
o
f
L
E
E
.
T
h
is
s
im
p
lifie
d
f
ea
tu
r
e
s
et
is
th
e
n
f
ed
in
to
th
e
KNN
class
if
ier
,
an
d
th
e
E
u
clid
ea
n
d
is
tan
ce
is
ca
lcu
lated
to
f
in
d
th
e
K
n
ea
r
est
n
eig
h
b
o
r
s
.
T
h
e
p
r
ed
icted
lab
el
is
g
iv
en
b
y
th
e
m
aj
o
r
ity
v
o
tin
g
with
n
eig
h
b
o
r
.
3
.
2
.
1
.
O
ptim
ized
lea
rning
ba
s
ed
enha
nce
d e
v
a
lua
t
io
n
ba
s
ed
K
NN
cla
s
s
if
ier
rule
T
h
e
o
p
tim
ized
L
E
E
b
ased
k
-
n
ea
r
est
n
eig
h
b
o
r
(
L
KNN)
cla
s
s
if
ier
r
u
le
clas
s
if
ies
a
n
ew
d
ata
p
o
in
t
ac
co
r
d
in
g
to
th
e
m
a
x
im
u
m
cla
s
s
o
f
its
n
ea
r
est
n
eig
h
b
o
r
s
.
I
t
i
s
b
ased
o
n
th
e
d
is
tan
ce
b
etwe
en
th
e
n
ew
in
s
tan
ce
an
d
all
t
h
e
tr
ain
i
n
g
i
n
s
tan
ce
s
an
d
c
h
o
o
s
e
th
e
k
n
ea
r
est
o
n
es
an
d
u
s
e
m
aj
o
r
ity
v
o
tin
g
(
o
r
weig
h
ted
v
o
tin
g
)
f
o
r
d
ec
id
in
g
wh
ic
h
lab
el
to
ass
ig
n
to
th
e
n
ew
in
s
tan
ce
.
(
,
)
=
√
∑
(
−
)
2
=
1
(
4
)
̂
=
∑
1
(
=
)
∈
∈
(
5
)
W
h
er
e
c
is
c
lass
lab
el
v
alu
es
an
d
is
d
en
o
ted
b
y
class
o
f
n
eig
h
b
o
r
i,
th
en
1
is
d
en
o
ted
b
y
i
n
d
icato
r
f
u
n
ctio
n
(
1
if
tr
u
e,
0
o
t
h
er
wis
e)
.
T
h
e
e
n
h
an
ce
d
L
E
E
m
eth
o
d
s
y
s
te
m
atica
lly
s
elec
ts
f
ea
tu
r
es
th
at
ar
e
s
im
u
ltan
eo
u
s
ly
in
f
o
r
m
ativ
e,
n
o
n
-
r
ed
u
n
d
a
n
t,
a
n
d
s
tab
le
ac
r
o
s
s
v
alid
atio
n
f
o
l
d
s
.
3
.
3
.
O
pti
m
ized
lea
rning
ba
s
ed
enha
nced
ev
a
lua
t
io
n
f
ea
t
ure
s
elec
t
io
n wit
h a
da
ptiv
e
lea
rning
a
lg
o
rit
hm
T
h
e
o
p
tim
ized
L
E
E
f
ea
tu
r
e
s
e
lectio
n
with
ad
ap
tiv
e
lear
n
i
n
g
alg
o
r
ith
m
h
as
in
teg
r
ated
th
e
a
d
v
an
tag
es
o
f
d
im
en
s
io
n
ality
r
e
d
u
ctio
n
a
n
d
d
y
n
am
ic
lear
n
in
g
to
im
p
r
o
v
e
p
r
ed
ictio
n
ac
cu
r
ac
y
.
Ho
wev
er
,
th
e
LEE
m
eth
o
d
s
elec
ts
m
o
s
t
in
f
o
r
m
ativ
e
an
d
r
elev
an
t
f
ea
tu
r
es
f
r
o
m
th
e
d
at
aset,
r
em
o
v
in
g
th
e
r
ed
u
n
d
an
t
an
d
n
o
is
y
f
ea
tu
r
es
an
d
h
en
ce
r
e
d
u
cin
g
th
e
co
m
p
u
tatio
n
al
co
m
p
le
x
ity
.
T
h
e
o
p
tim
ized
L
E
E
f
ea
tu
r
e
s
elec
tio
n
in
teg
r
ated
wit
h
ad
ap
tiv
e
lear
n
in
g
al
g
o
r
ith
m
f
o
r
SPP
is
s
h
o
wn
in
Fig
u
r
e
3
.
T
h
e
f
ir
s
t
s
tep
in
th
e
p
r
o
ce
d
u
r
e
is
p
r
ep
r
o
ce
s
s
in
g
o
f
th
e
ed
u
ca
tio
n
al
d
ata,
s
u
ch
as
d
ea
lin
g
with
m
is
s
in
g
v
alu
es,
n
o
r
m
alizin
g
th
e
d
ata,
an
d
en
co
d
in
g
th
e
d
ata.
T
h
ese
f
ea
tu
r
es
ar
e
s
co
r
e
d
with
th
e
L
E
E
f
u
n
ctio
n
wh
ich
is
a
b
al
an
ce
m
ea
s
u
r
e
o
f
r
elev
a
n
ce
to
th
e
tar
g
et
(
s
tu
d
en
t
p
er
f
o
r
m
an
ce
)
an
d
r
ed
u
n
d
a
n
cy
with
o
th
er
f
ea
tu
r
es.
T
h
is
ad
a
p
tiv
e
m
ec
h
an
is
m
will
s
tab
iliz
e
tr
ain
in
g
,
im
p
r
o
v
e
co
n
v
er
g
en
ce
an
d
av
o
id
o
v
er
f
it
tin
g
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
Op
timiz
ed
cla
s
s
if
ica
tio
n
o
f stu
d
en
t p
erfo
r
ma
n
ce
o
u
tco
mes u
s
in
g
LEE
…
(
K
a
ma
r
a
ju
g
a
d
d
a
K
is
h
o
r
e
K
u
ma
r
)
2463
Fig
u
r
e
2
.
E
n
h
an
ce
d
L
E
E
f
ea
tu
r
e
s
elec
tio
n
with
KNN
ap
p
lied
o
n
SP
P
Fig
u
r
e
3
.
Op
tim
ized
L
E
E
f
ea
t
u
r
e
s
elec
tio
n
with
ad
ap
tiv
e
lear
n
in
g
alg
o
r
ith
m
u
s
in
g
SP
P
3
.
3
.
1
.
O
ptim
ized
a
da
ptiv
e
le
a
rning
cla
s
s
if
ier
T
h
e
o
p
tim
ized
a
d
ap
tiv
e
lear
n
in
g
class
if
ier
(
Ad
aBo
o
s
t
class
i
f
ier
alg
o
r
ith
m
s
ar
e
a
p
p
lied
in
t
o
s
tu
d
en
t
d
ataset
to
p
r
ed
ict
t
h
e
s
tu
d
e
n
t
ac
ad
em
ic
p
e
r
f
o
r
m
an
ce
th
en
i
n
teg
r
atin
g
with
f
ea
tu
r
e
s
elec
tio
n
tec
h
n
iq
u
es
th
at
in
co
r
p
o
r
ates
with
L
E
E
tec
h
n
i
q
u
es
to
r
e
d
u
ce
t
h
e
n
o
is
e
an
d
im
p
r
o
v
e
th
e
ef
f
icien
cy
)
is
a
m
ac
h
in
e
lea
r
n
in
g
ap
p
r
o
ac
h
wh
e
r
e
th
e
m
o
d
el
d
y
n
am
ically
a
d
ju
s
ts
its
lear
n
i
n
g
p
r
o
ce
s
s
b
ased
o
n
th
e
d
at
a,
en
v
i
r
o
n
m
e
n
t,
o
r
tr
ain
in
g
p
r
o
g
r
ess
.
Ad
ap
tiv
e
lear
n
in
g
class
if
ier
s
ch
an
g
e
th
eir
lea
r
n
in
g
r
ate,
weig
h
t
s
,
an
d
/o
r
d
ec
is
io
n
b
o
u
n
d
ar
ies
d
u
r
in
g
lear
n
in
g
to
m
ax
im
ize
p
er
f
o
r
m
an
ce
an
d
s
tab
ilize
lear
n
in
g
,
wh
er
ea
s
s
tatic
class
if
ier
s
d
o
n
o
t
ad
ju
s
t th
eir
p
ar
am
eter
s
d
u
r
in
g
lear
n
in
g
an
d
u
s
e
a
f
i
x
ed
lear
n
in
g
r
ate.
(
)
=
1
∑
=
1
(
(
;
)
.
)
(
6
)
+
1
=
.
1
1
+
.
(
7
)
Her
e,
is
a
d
ec
ay
f
ac
t
o
r
an
d
is
th
e
iter
atio
n
.
T
h
is
p
r
e
v
e
n
ts
o
v
er
s
h
o
o
tin
g
in
ea
r
l
y
s
tag
es
an
d
allo
ws
f
in
e
-
tu
n
in
g
later
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
26
:
2
4
5
9
-
2
4
7
0
2464
3
.
4
.
O
pti
m
ized
lea
rning
ba
s
ed
enha
nced
ev
a
lua
t
io
n
f
ea
t
ure
s
elec
t
io
n wit
h na
ïv
e
B
a
y
es a
lg
o
rit
hm
Op
tim
ized
L
E
E
f
ea
tu
r
e
s
elec
tio
n
with
n
aïv
e
B
ay
es
alg
o
r
ith
m
im
p
r
o
v
es
th
e
SPP
b
y
s
elec
tin
g
m
o
s
t
r
elev
an
t
f
ea
tu
r
es
an
d
elim
in
atin
g
n
o
is
e
an
d
r
ed
u
n
d
an
c
y
in
t
h
e
d
ata.
T
h
e
L
E
E
a
p
p
r
o
ac
h
d
eter
m
in
es
th
e
s
u
b
s
et
o
f
attr
ib
u
tes
th
at
is
m
o
s
t
u
s
ef
u
l
in
m
a
x
im
izin
g
class
if
icatio
n
ac
cu
r
ac
y
an
d
n
aïv
e
B
ay
es
is
u
s
ed
o
n
th
is
s
tr
ea
m
lin
ed
f
ea
tu
r
e
s
p
ac
e
b
ec
au
s
e
o
f
its
s
im
p
licity
an
d
e
f
f
i
cien
cy
.
T
h
e
o
p
tim
ized
L
E
E
f
e
atu
r
e
s
elec
tio
n
with
n
aïv
e
B
ay
es
alg
o
r
ith
m
f
o
r
SP
P
is
s
h
o
wn
in
Fig
u
r
e
4
.
I
t
s
tar
ts
with
p
r
ep
r
o
ce
s
s
in
g
,
m
is
s
in
g
v
alu
e
tr
ea
tm
en
t,
n
o
r
m
aliza
tio
n
an
d
en
co
d
i
n
g
to
m
ak
e
s
u
r
e
th
e
d
ata
is
co
n
s
i
s
ten
t.
E
ac
h
f
ea
tu
r
e
is
th
en
ex
am
in
ed
b
y
th
e
L
E
E
f
u
n
ctio
n
,
an
d
th
e
s
ig
n
if
ican
ce
is
ca
lcu
lated
b
ased
o
n
th
e
c
o
r
r
elatio
n
b
etwe
en
th
e
f
ea
tu
r
e
an
d
th
e
tar
g
et
class
an
d
o
n
th
e
r
ed
u
n
d
an
c
y
o
f
th
e
f
ea
tu
r
es.
T
h
is
lo
wer
ed
d
im
en
s
io
n
ality
is
th
en
p
ass
ed
o
n
to
th
e
n
aïv
e
B
ay
es
class
if
ier
,
wh
er
e
th
e
p
o
s
ter
io
r
p
r
o
b
a
b
ilit
ies o
f
ea
ch
class
is
co
m
p
u
ted
b
ased
o
n
B
ay
es’
t
h
eo
r
em
.
3
.
4
.
1
.
Na
ïv
e
B
a
y
es c
la
s
s
if
ier
a
lg
o
rit
hm
Naïv
e
B
ay
es
clas
s
if
ier
is
a
s
i
m
p
le,
y
et
p
o
wer
f
u
l
p
r
o
b
ab
ilis
tic
m
ac
h
in
e
lear
n
in
g
al
g
o
r
ith
m
b
ased
o
n
B
ay
es’
th
eo
r
em
ass
u
m
in
g
c
o
n
d
itio
n
al
in
d
ep
e
n
d
en
ce
o
f
f
ea
tu
r
es.
I
t
ap
p
r
o
x
im
ates
th
e
p
r
o
b
a
b
ilit
y
o
f
ea
ch
class
,
g
iv
en
a
s
et
o
f
f
ea
tu
r
es a
n
d
ass
ig
n
s
th
e
m
o
s
t lik
ely
class
to
th
e
s
et
o
f
in
p
u
t
f
ea
tu
r
es.
(
|
)
=
(
)
∏
(
|
)
=
1
(
)
(
8
)
=
(
)
∏
(
|
)
=
1
(
9
)
Her
e,
is
d
en
o
ted
with
class
lab
el
v
alu
es
an
d
p
r
ed
ict
th
e
s
tu
d
en
t
in
f
o
r
m
atio
n
,
x
is
an
iter
a
tio
n
v
ec
to
r
o
f
th
e
g
iv
en
d
ata,
(
)
is
d
en
o
ted
with
th
e
p
r
io
r
p
r
o
b
ab
ilit
y
o
f
class
o
f
,
th
en
lik
elih
o
o
d
f
u
n
ctio
n
is
an
d
g
iv
e
class
o
f
.
Fig
u
r
e
4
.
Op
tim
ized
L
E
E
f
ea
t
u
r
e
s
elec
tio
n
with
n
aïv
e
B
ay
es
alg
o
r
ith
m
u
s
in
g
SP
P
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
e
d
ata
s
et
co
n
tain
s
m
o
r
e
th
an
2
,
7
6
5
s
tu
d
en
ts
’
in
f
o
r
m
atio
n
an
d
is
u
s
ed
to
ass
es
s
th
e
im
p
ac
t
o
f
LEE
-
b
ased
f
ea
tu
r
e
s
elec
tio
n
o
n
ac
ad
em
ic
p
er
f
o
r
m
a
n
ce
p
r
ed
ictio
n
.
T
h
e
u
s
e
o
f
L
E
E
g
r
ea
tly
im
p
r
o
v
ed
th
e
ac
cu
r
ac
y
o
f
th
e
m
o
d
els
b
y
ch
o
o
s
in
g
th
e
m
o
s
t
p
er
tin
en
t
at
tr
ib
u
tes.
L
E
E
-
b
ased
SVM
an
d
ad
ap
tiv
e
lear
n
in
g
m
o
d
els
ac
h
iev
ed
h
ig
h
ac
cu
r
a
cy
,
p
r
ec
is
io
n
an
d
F1
-
s
co
r
e
a
m
o
n
g
th
e
class
if
ier
s
u
s
ed
:
S
VM
,
KNN,
ad
ap
tiv
e
lear
n
in
g
,
a
n
d
n
aï
v
e
B
ay
es.
T
h
e
r
esu
lts
s
h
o
w
th
at
it
is
p
o
s
s
ib
le
to
im
p
r
o
v
e
th
e
p
r
ed
ictio
n
ac
cu
r
ac
y
an
d
d
ec
r
ea
s
e
th
e
tr
ain
in
g
tim
e
if
th
e
ir
r
elev
an
t a
n
d
n
o
is
y
f
ea
tu
r
es
ar
e
r
em
o
v
ed
.
4
.
1
.
Acc
ura
cy
I
n
o
p
tim
ize
d
L
E
E
f
ea
t
u
r
e
s
elec
tio
n
f
r
am
ewo
r
k
,
ac
c
u
r
ac
y
is
ca
lcu
lated
p
o
s
t
f
ea
tu
r
e
s
elec
tio
n
(
m
o
s
t
s
u
itab
le
f
ea
tu
r
es)
an
d
af
ter
class
if
icatio
n
(
KNN,
S
V
M,
n
aïv
e
B
ay
es
,
an
d
ad
ap
tiv
e
lear
n
in
g
)
.
T
h
e
L
E
E
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
Op
timiz
ed
cla
s
s
if
ica
tio
n
o
f stu
d
en
t p
erfo
r
ma
n
ce
o
u
tco
mes u
s
in
g
LEE
…
(
K
a
ma
r
a
ju
g
a
d
d
a
K
is
h
o
r
e
K
u
ma
r
)
2465
f
u
n
ctio
n
elim
in
ates
r
ed
u
n
d
an
t
an
d
ir
r
elev
an
t
f
ea
tu
r
es,
wh
ic
h
ca
n
h
elp
to
r
ed
u
ce
n
o
is
e
a
n
d
im
p
r
o
v
e
m
o
d
el
ef
f
icien
cy
.
Hen
ce
,
a
cc
u
r
ac
y
i
s
u
s
ed
as
a
b
aselin
e
cr
iter
io
n
to
ass
es
s
th
e
p
er
f
o
r
m
an
ce
o
f
ea
ch
class
if
ier
in
p
r
ed
ictin
g
s
tu
d
e
n
t p
er
f
o
r
m
a
n
c
e.
̅
̅
̅
̅
̅
=
1
∑
(
)
=
1
(
1
0
)
=
√
1
−
1
∑
(
(
)
−
̅
̅
̅
̅
̅
̅
)
2
=
1
(
1
1
)
Fo
r
co
m
p
ar
is
o
n
ac
r
o
s
s
alg
o
r
ith
m
s
,
p
r
esen
t
a
tab
le
o
f
̅
̅
̅
̅
̅
,
.
Op
tio
n
ally
p
er
f
o
r
m
a
p
air
ed
s
tatis
ti
ca
l
test
o
n
f
o
ld
-
wis
e
ac
cu
r
ac
ies
to
ch
ec
k
wh
eth
er
t
h
e
o
b
s
er
v
e
d
̅
̅
̅
̅
̅
is
s
tati
s
tically
s
ig
n
if
ican
t.
T
h
is
p
r
o
ce
d
u
r
e
en
s
u
r
es
a
f
air
,
r
ep
r
o
d
u
cib
le
co
m
p
ar
is
o
n
o
f
h
o
w
o
p
tim
ized
L
E
E
af
f
ec
ts
ea
ch
class
if
ier
’
s
p
r
ed
ictiv
e
p
er
f
o
r
m
an
ce
o
n
th
e
s
tu
d
e
n
t
-
p
e
r
f
o
r
m
a
n
ce
d
ataset.
I
n
th
e
T
a
b
le
1
s
h
o
ws
th
at
th
e
o
v
er
all
co
m
p
ar
is
o
n
o
f
th
e
SPP
,
th
e
p
r
ed
ictio
n
with
L
K
NN
h
as
th
e
h
ig
h
est
p
r
ec
is
io
n
(
0
.
9
3
5
6
)
a
n
d
r
ec
all
(
0
.
9
2
5
)
,
wh
ich
m
ak
e
s
it
h
av
e
th
e
b
est
F
m
ea
s
u
r
e
(
0
.
9
2
3
6
)
an
d
ar
ea
u
n
d
er
th
e
cu
r
v
e
o
r
AUC
(
0
.
9
5
2
)
.
T
h
e
o
v
er
all
b
alan
ce
o
f
L
E
E
b
ased
n
aïv
e
B
ay
es
(
L
NB
)
class
if
ier
i
s
b
etter
with
AUC
=0
.
9
6
5
an
d
it
is
p
r
e
cisi
o
n
an
d
r
ec
all
wer
e
c
o
n
s
is
ten
tly
h
ig
h
with
0
.
9
2
4
4
an
d
0
.
9
0
5
r
esp
ec
tiv
ely
.
I
n
th
e
m
ea
n
tim
e,
L
SVM
ac
h
iev
e
d
r
eliab
le
r
esu
lts
with
a
p
r
ec
is
io
n
o
f
0
.
9
0
5
7
a
n
d
a
n
AUC
o
f
0
.
9
5
6
;
lik
ewise,
th
e
L
E
E
b
ased
ad
ap
tiv
e
lear
n
i
n
g
(
L
AL
)
m
o
d
el
s
till
ac
h
iev
e
d
c
o
m
p
etitiv
e
r
esu
lts
with
a
p
r
ec
is
io
n
o
f
0
.
9
0
8
1
an
d
an
AUC
o
f
0
.
9
6
0
.
T
h
e
r
esu
lts
s
u
g
g
est
th
at
all
alg
o
r
ith
m
s
ar
e
s
u
cc
ess
f
u
l,
with
th
e
L
N
B
an
d
L
KNN
alg
o
r
ith
m
s
p
r
o
v
id
i
n
g
th
e
h
ig
h
est lev
el
o
f
p
r
ed
ictiv
e
s
tu
d
en
t
p
er
f
o
r
m
an
c
e
ac
cu
r
ac
y
.
T
h
e
co
m
p
a
r
ativ
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
d
if
f
er
en
t a
lg
o
r
ith
m
s
(
L
SVM,
L
KNN,
L
NB
,
an
d
L
AL
)
u
s
in
g
th
e
p
r
ec
is
io
n
,
r
ec
all
an
d
F
-
m
ea
s
u
r
e
f
o
r
p
r
ed
ictin
g
s
tu
d
en
t
p
er
f
o
r
m
an
ce
is
s
h
o
wn
in
Fig
u
r
e
5
.
T
h
e
L
KNN
m
o
d
el
y
ield
ed
th
e
b
est
r
esu
lts
with
a
p
r
ec
is
io
n
o
f
n
ea
r
ly
0
.
9
4
,
a
r
ec
all
r
ate
ex
ce
ed
in
g
0
.
9
2
,
an
d
an
F
-
m
ea
s
u
r
e
o
f
0
.
9
2
3
,
d
em
o
n
s
tr
atin
g
b
ala
n
ce
d
an
d
ac
c
u
r
ate
class
if
icatio
n
.
T
h
e
L
NB
class
if
ier
also
s
h
o
wed
g
o
o
d
p
er
f
o
r
m
an
ce
with
a
p
r
ec
is
io
n
o
f
0
.
9
2
3
,
t
h
e
r
ec
all
o
f
0
.
9
0
5
a
n
d
a
co
m
p
etitiv
e
F
-
m
ea
s
u
r
e
o
f
0
.
9
0
8
.
Ho
wev
e
r
,
L
SVM
an
d
L
AL
ac
h
iev
ed
s
lig
h
tly
lo
wer
r
esu
lts
with
a
p
r
ec
is
io
n
o
f
ab
o
u
t
0
.
9
0
an
d
a
r
ec
all
o
f
less
th
an
0
.
8
8
an
d
F
-
m
ea
s
u
r
e
b
elo
w
0
.
8
8
,
w
h
ich
s
u
g
g
ests
th
at
th
ey
lack
th
e
s
en
s
itiv
ity
o
f
id
en
tify
in
g
r
ea
l
s
tu
d
en
t o
u
tco
m
es.
T
ab
le
1
.
Ov
e
r
all
co
m
p
ar
is
o
n
o
f
L
E
E
p
er
f
o
r
m
a
n
ce
m
ea
s
u
r
es
A
l
g
o
r
i
t
h
ms
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F
-
m
e
a
su
r
e
s
R
O
C
/
A
U
C
LSV
M
0
.
9
0
5
6
8
4
0
.
8
7
5
0
.
8
5
8
7
6
2
0
.
9
5
6
LK
N
N
0
.
9
3
5
6
8
6
0
.
9
2
5
0
.
9
2
3
6
2
6
0
.
9
5
2
LN
B
0
.
9
2
4
3
9
7
0
.
9
0
5
0
.
9
0
7
9
8
6
0
.
9
6
5
LA
L
0
.
9
0
8
0
8
2
0
.
8
7
5
0
.
8
5
9
5
6
0
.
9
6
0
Fig
u
r
e
5
.
Per
f
o
r
m
an
c
e
co
m
p
ar
is
o
n
o
f
o
p
tim
izatio
n
ap
p
r
o
ac
h
4
.
2
.
P
re
cisi
o
n,
r
ec
a
ll,
a
nd
F
-
m
ea
s
ure
T
h
e
o
p
tim
ized
L
E
E
f
ea
tu
r
e
s
elec
tio
n
is
d
o
n
e
o
n
t
h
e
s
tu
d
en
t
p
er
f
o
r
m
a
n
ce
d
ataset
to
s
elec
t
th
e
m
o
s
t
d
is
cr
im
in
ativ
e
an
d
least
r
e
d
u
n
d
an
t
attr
ib
u
tes,
an
d
th
e
n
f
e
d
in
to
v
ar
io
u
s
class
if
ier
s
in
clu
d
in
g
KNN,
SVM,
n
aïv
e
B
ay
es,
an
d
ad
ap
tiv
e
lea
r
n
in
g
.
E
ac
h
alg
o
r
ith
m
is
test
e
d
with
p
r
ec
is
io
n
,
r
ec
all
,
an
d
F
-
m
ea
s
u
r
e,
p
r
o
v
id
in
g
0.90
5684
0.93
5686
0.92
4397
0.90
8082
0.87
5
0.92
5
0.90
5
0.87
5
0.85
8762
0.92
3626
0.90
7986
0.85
956
0.8
2
0.8
4
0.8
6
0.8
8
0.9
0.9
2
0.9
4
0.9
6
LSVM
LKN
N
LN
B
LAL
A
c
c
r
ac
y
A
l
g
o
r
i
t
hm
s
P
r
e
c
i
si
o
n
R
e
c
al
l
F
-
M
e
as
ur
e
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
26
:
2
4
5
9
-
2
4
7
0
2466
m
o
r
e
in
s
ig
h
t
in
to
t
h
e
q
u
ality
o
f
class
if
icatio
n
th
an
ac
cu
r
a
cy
.
Fo
r
ea
ch
alg
o
r
ith
m
,
af
ter
th
e
L
E
E
s
elec
tio
n
,
co
m
p
u
te
th
ese
m
etr
ics
o
n
th
e
L
E
E
o
p
tim
ized
f
ea
tu
r
e
s
et
(
o
r
o
n
th
e
L
E
E
o
p
tim
ized
alg
o
r
it
h
m
-
s
p
ec
if
ic
f
ea
t
u
r
e
s
et
if
y
o
u
o
p
tim
ize
p
er
alg
o
r
ith
m
)
.
T
y
p
ically
,
b
etter
im
p
r
o
v
em
en
ts
f
r
o
m
L
E
E
will
r
esu
lt
in
an
in
cr
ea
s
e
in
tr
u
e
p
o
s
itiv
e
(
TP
)
an
d
d
ec
r
ea
s
e
i
n
f
alse
p
o
s
itiv
e
(
FP
)
an
d
/o
r
f
alse
n
eg
ativ
e
(
FN
)
,
lead
in
g
to
an
in
cr
ea
s
e
in
p
r
ec
is
io
n
an
d
/o
r
r
ec
all
an
d
a
l
ar
g
er
F
-
m
ea
s
u
r
e.
T
h
e
r
ec
eiv
er
o
p
er
atin
g
ch
a
r
ac
ter
is
tic
(
R
OC
)
cu
r
v
es o
f
s
elec
ted
f
ea
tu
r
es
u
s
in
g
L
E
E
f
o
r
v
ar
io
u
s
class
if
ier
s
(
K
NN,
SVM,
n
aïv
e
B
ay
es,
an
d
ad
ap
tiv
e
lear
n
in
g
)
ar
e
s
h
o
w
n
in
Fig
u
r
e
6
.
T
h
e
m
o
d
els
h
a
v
e
g
o
o
d
d
is
cr
im
in
ativ
e
p
o
wer
,
wit
h
an
AUC
ab
o
v
e
0
.
9
5
in
all
m
o
d
el.
T
h
e
h
i
g
h
est
AUC
(
0
.
9
6
5
)
is
f
o
r
t
h
e
n
aïv
e
B
ay
es
clas
s
ier
wh
ich
h
as
th
e
h
ig
h
est
class
if
icatio
n
p
o
wer
f
o
r
d
is
tin
g
u
is
h
in
g
s
u
cc
ess
f
u
l
an
d
u
n
s
u
cc
ess
f
u
l
s
tu
d
en
ts
.
Ad
ap
tiv
e
l
ea
r
n
in
g
i
s
n
ex
t
with
an
AU
C
o
f
0
.
9
6
0
,
in
d
icatin
g
g
o
o
d
ad
ap
tab
ilit
y
an
d
p
r
ed
ictio
n
.
SVM
r
ec
o
r
d
s
AUC
o
f
0
.
9
5
6
,
wh
ich
in
d
icate
s
g
o
o
d
d
ec
is
io
n
b
o
u
n
d
ar
ies
wh
ile
KNN
r
ec
o
r
d
s
AUC o
f
0
.
9
5
2
,
wh
ich
is
s
lig
h
tly
less
b
u
t sti
ll
u
s
ef
u
l f
o
r
p
atter
n
r
ec
o
g
n
itio
n
ap
p
licatio
n
s
.
=
Σ
Σ
(
+
)
(
1
2
)
=
Σ
Σ
(
+
)
(
1
3
)
−
=
1
∑
,
.
=
1
(
1
4
)
Fig
u
r
e
6
.
R
OC
cu
r
v
es f
o
r
o
p
ti
m
ize
f
u
n
ctio
n
ap
p
lied
o
n
SP
P
4
.
3
.
T
im
e
a
nd
RO
C
wit
h AU
C
T
y
p
ically
,
two
m
ea
s
u
r
es
ar
e
r
ep
o
r
ted
f
o
r
o
p
tim
ized
L
E
E
:
i)
tim
e
(
wall
clo
c
k
o
r
C
PU
tim
e)
th
at
r
ef
lects
th
e
co
s
t
o
f
f
ea
t
u
r
e
s
el
ec
tio
n
an
d
m
o
d
el
tr
ain
i
n
g
/in
f
er
en
ce
an
d
ii)
th
e
d
is
cr
im
in
atio
n
p
e
r
f
o
r
m
an
ce
i
n
ter
m
s
o
f
t
h
e
R
OC
cu
r
v
e
a
n
d
its
s
u
m
m
ar
y
m
ea
s
u
r
e
AUC.
I
n
p
r
ac
tice,
y
o
u
r
ep
o
r
t
th
e
to
tal
r
u
n
n
in
g
tim
e,
w
h
ich
in
clu
d
es
th
e
tim
e
s
p
en
t
in
f
ea
tu
r
e
s
elec
tio
n
an
d
th
e
tim
e
s
p
en
t
co
m
p
u
tin
g
th
e
class
if
ier
o
u
tp
u
t
(
it
is
co
n
tin
u
o
u
s
s
co
r
es,
wh
ich
ar
e
p
r
o
b
a
b
ilit
ies
o
r
m
ar
g
in
s
,
o
n
th
e
test
s
et)
,
an
d
y
o
u
r
e
p
o
r
t
th
e
R
OC
/AU
C
ca
lcu
lated
o
n
th
e
p
r
o
b
ab
ilit
ies/
m
ar
g
in
s
o
u
tp
u
t b
y
th
e
class
if
ier
o
n
th
e
test
s
et.
=
(
.
.
)
(
1
5
)
=
∑
+
1
+
2
−
1
=
1
.
(
+
1
−
)
(
1
6
)
A
lar
g
er
AUC
m
ea
n
s
b
etter
class
if
icatio
n
ab
ilit
y
to
d
is
tin
g
u
is
h
s
tu
d
en
ts
’
s
u
cc
ess
v
s
f
ail
u
r
e.
T
h
e
o
p
tim
ized
LEE
-
b
ased
KNN,
SVM,
n
aïv
e
B
ay
es,
an
d
ad
a
p
tiv
e
lear
n
i
n
g
m
o
d
els
ar
e
th
er
ef
o
r
e
c
o
m
p
a
r
ed
an
d
co
n
t
r
asted
b
ased
o
n
b
o
th
e
x
ec
u
tio
n
tim
e
an
d
R
OC
/AU
C
in
o
r
d
er
to
m
a
k
e
a
b
alan
ce
b
etwe
en
co
s
t a
n
d
ac
cu
r
ac
y
o
f
SPP
.
T
h
e
p
e
r
f
o
r
m
an
ce
o
f
v
ar
io
u
s
al
g
o
r
ith
m
s
with
L
E
E
f
ea
tu
r
e
s
elec
tio
n
ap
p
r
o
ac
h
is
s
h
o
w
n
in
T
ab
le
2
,
in
wh
ich
L
KNN
alg
o
r
ith
m
g
iv
es
h
ig
h
est
ac
cu
r
ac
y
(
0
.
9
2
5
)
with
th
e
lo
west
co
m
p
u
tatio
n
tim
e
(
0
.
0
0
1
3
5
4
s
)
f
o
llo
wed
b
y
L
NB
with
ac
c
u
r
a
cy
(
0
.
9
0
5
)
a
n
d
c
o
m
p
u
tatio
n
ti
m
e
(
0
.
0
0
1
5
3
5
s
)
.
T
h
e
m
o
d
er
at
e
ac
cu
r
ac
y
o
f
0
.
8
7
5
0.95
2
0.95
6
0.96
0.96
5
0.9
45
0.9
5
0.9
55
0.9
6
0.9
65
0.9
7
L
K
N
N
L
S
V
M
L
A
P
L
N
B
T
R
UE
P
O
S
IT
IVE
R
A
T
E
F
A
LSE
P
O
S
IT
IVE
R
A
T
E
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
Op
timiz
ed
cla
s
s
if
ica
tio
n
o
f stu
d
en
t p
erfo
r
ma
n
ce
o
u
tco
mes u
s
in
g
LEE
…
(
K
a
ma
r
a
ju
g
a
d
d
a
K
is
h
o
r
e
K
u
ma
r
)
2467
was
ac
h
iev
ed
b
y
L
SVM,
wh
e
r
ea
s
th
e
co
m
p
ar
ativ
ely
h
ig
h
er
ex
ec
u
tio
n
tim
es
wer
e
0
.
0
0
9
0
4
7
s
an
d
0
.
0
0
7
8
8
9
s
f
o
r
L
SVM
an
d
L
AL
,
r
esp
ec
tiv
ely
.
T
h
e
r
esu
lts
u
n
d
er
lin
e
th
e
ef
f
icien
cy
o
f
L
KNN
an
d
L
N
B
to
ac
h
iev
e
co
r
r
ec
t
p
r
ed
ictio
n
o
f
s
tu
d
en
ts
’
p
e
r
f
o
r
m
an
ce
wh
ile
tr
ain
i
n
g
tim
e
is
k
ep
t
to
a
m
i
n
im
u
m
.
As
s
h
o
wn
in
th
e
Fig
u
r
e
7
,
L
KNN
alg
o
r
ith
m
h
as
th
e
h
ig
h
est
ac
cu
r
ac
y
(
0
.
9
2
5
)
a
n
d
th
e
l
o
west
co
m
p
u
tatio
n
tim
e
(
0
.
0
0
1
3
5
4
s
)
,
f
o
llo
wed
b
y
L
NB
with
th
e
ac
cu
r
ac
y
o
f
0
.
9
0
5
an
d
tim
e
o
f
0
.
0
0
1
5
3
5
s
.
Ho
wev
er
,
L
SVM
an
d
L
AL
g
av
e
less
er
ac
cu
r
ac
y
,
i.e
.
,
0
.
8
7
5
,
a
n
d
h
ig
h
er
ex
ec
u
tio
n
tim
e
was
0
.
0
0
9
0
4
7
s
a
n
d
0
.
0
0
7
8
8
9
s
,
r
esp
ec
tiv
ely
.
T
h
e
r
esu
lts
s
h
o
w
th
at
L
KNN
an
d
L
NB
ca
n
ac
h
iev
e
b
etter
p
r
ed
ictio
n
ac
cu
r
ac
y
wh
ile
m
ain
tain
in
g
th
e
d
e
m
an
d
f
o
r
c
o
m
p
u
tatio
n
al
r
eso
u
r
ce
s
,
an
d
ac
h
iev
e
b
etter
r
esu
lts
th
an
L
SVM
an
d
L
AL
.
T
ab
le
2
.
SPP
b
ased
o
n
L
E
E
s
elec
tio
n
ap
p
r
o
ac
h
A
l
g
o
r
i
t
h
ms
Ti
me
A
c
c
u
r
a
c
y
LSV
M
0
.
0
0
9
0
4
7
0
.
8
7
5
LK
N
N
0
.
0
0
1
3
5
4
0
.
9
2
5
LN
B
0
.
0
0
1
5
3
5
0
.
9
0
5
LA
L
0
.
0
0
7
8
8
9
0
.
8
7
5
Fig
u
r
e
7
.
C
o
m
p
a
r
is
o
n
o
f
ac
cu
r
ac
y
an
d
tim
e
b
ased
o
n
L
E
E
m
eth
o
d
5.
CO
NCLU
SI
O
N
T
h
e
r
esear
ch
p
ap
er
will b
e
p
r
o
v
id
in
g
an
im
p
r
o
v
ed
d
ata
m
in
i
n
g
m
o
d
el
to
p
r
e
d
ict
s
tu
d
en
t p
er
f
o
r
m
a
n
ce
u
s
in
g
lear
n
i
n
g
-
b
ased
en
h
an
ce
d
ev
alu
atio
n
f
ea
t
u
r
e
s
elec
tio
n
m
eth
o
d
an
d
tr
a
d
itio
n
al
lear
n
in
g
class
if
ier
s
.
T
h
is
ar
ticle
s
h
o
wed
th
at
p
r
u
n
in
g
ir
r
elev
an
t
an
d
r
ed
u
n
d
a
n
t
f
ea
tu
r
es
n
o
t
o
n
ly
i
n
cr
ea
s
ed
th
e
clas
s
if
icatio
n
r
ates
b
u
t
also
r
ed
u
ce
d
th
e
co
m
p
u
tatio
n
al
co
m
p
lex
ity
an
d
was
an
ef
f
ec
tiv
e
an
d
f
ea
s
ib
le
m
eth
o
d
in
an
ed
u
ca
tio
n
al
en
v
ir
o
n
m
en
t.
E
v
alu
atio
n
r
es
u
lts
th
u
s
co
n
f
i
r
m
ed
th
at
th
e
u
s
e
o
f
f
ea
tu
r
e
o
p
tim
izatio
n
is
o
f
v
er
y
cr
itical
im
p
o
r
tan
ce
in
im
p
r
o
v
in
g
th
e
p
r
ed
ictiv
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
class
if
ier
s
u
s
ed
in
L
E
E
b
ase
d
on
SVM,
KNN,
ad
ap
tiv
e
lear
n
in
g
a
n
d
n
aïv
e
B
ay
es.
T
h
e
ex
p
e
r
im
en
tal
h
as
ac
m
e
th
e
n
ee
d
t
o
ass
im
ilate
in
tellectu
al
f
ea
tu
r
e
s
elec
tio
n
an
d
p
r
ed
ictiv
e
m
o
d
e
ls
f
o
r
b
etter
d
ec
is
io
n
-
m
ak
in
g
in
ed
u
ca
tio
n
.
Op
tim
ized
tech
n
iq
u
es,
h
o
wev
er
,
is
r
o
b
u
s
t
b
u
t
th
e
r
e
is
r
o
o
m
f
o
r
r
ef
in
em
en
t.
E
x
p
er
im
e
n
tal
r
esu
lts
s
h
o
w
th
at
th
e
ac
cu
r
ac
y
o
f
th
e
f
r
am
ewo
r
k
is
9
2
.
5
%
an
d
ex
ec
u
tio
n
tim
e
is
0
.
0
0
1
3
5
4
s
ec
o
n
d
s
,
an
d
th
e
lo
west
ac
cu
r
ac
y
is
8
7
.
5
%
wi
th
ex
ec
u
tio
n
tim
e
0
.
0
0
9
0
4
7
s
ec
o
n
d
s
r
esp
ec
tiv
ely
.
I
n
f
u
tu
r
e
s
tu
d
ies,
en
s
em
b
l
e
lear
n
in
g
tech
n
iq
u
es,
d
ee
p
l
ea
r
n
in
g
m
o
d
els
o
r
h
y
b
r
id
o
p
tim
izatio
n
m
eth
o
d
s
c
o
u
ld
b
e
in
co
r
p
o
r
ated
to
b
etter
ca
p
tu
r
e
m
o
r
e
c
o
m
p
lex
s
tu
d
en
t
b
eh
av
i
o
r
p
atter
n
s
.
Mo
r
eo
v
er
,
h
av
in
g
th
e
d
ata
co
llected
f
r
o
m
a
v
ar
iety
o
f
a
ca
d
em
ic
in
s
titu
tio
n
s
an
d
wit
h
th
e
in
teg
r
atio
n
o
f
r
ea
l
-
tim
e
lear
n
in
g
an
aly
tics
,
g
en
er
aliza
b
ilit
y
an
d
ad
ap
ta
b
ilit
y
wo
u
ld
b
e
en
h
an
ce
d
f
u
r
th
e
r
.
T
h
is
wo
u
ld
h
av
e
ev
en
g
r
ea
ter
im
p
licatio
n
s
f
o
r
th
e
u
s
e
o
f
th
e
a
p
p
r
o
ac
h
a
n
d
wo
u
ld
g
i
v
e
teac
h
er
s
e
v
en
b
etter
to
o
ls
f
o
r
b
o
o
s
tin
g
s
tu
d
en
t su
cc
ess
.
ACK
NO
WL
E
DG
M
E
N
T
S
T
h
e
au
th
o
r
g
r
atef
u
lly
ac
k
n
o
w
led
g
es
th
e
s
u
p
p
o
r
t
to
r
ec
ei
v
e
f
r
o
m
th
e
Dep
ar
tm
e
n
t
of
B
io
te
ch
n
o
lo
g
y
(
DB
T
)
,
Go
v
er
n
m
en
t
of
I
n
d
ia
u
n
d
er
th
e
DB
T
Star
C
o
lleg
e
Sch
em
e
(
HR
D
-
1
1
0
1
1
/1
0
/2
0
2
4
-
HR
D
-
DB
T
)
f
o
r
0
.
0
0
9
0
4
7
0
.
0
0
1
3
5
4
0
.
0
0
1
5
3
5
0
.
0
0
7
8
8
9
0
.
8
7
5
0
.
9
2
5
0
.
9
0
5
0
.
8
7
5
0
0
.
1
0
.
2
0
.3
0
.
4
0
.
5
0
.
6
0
.
7
0
.
8
0
.
9
1
LSV
M
LK
N
N
LN
B
LA
L
A
c
c
u
r
a
c
y
A
l
g
o
r
i
t
h
m
s
S
t
u
d
e
n
t
p
e
r
f
o
r
man
c
e
p
r
e
d
i
c
t
i
o
n
Ti
me
A
c
c
u
r
a
c
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
26
:
2
4
5
9
-
2
4
7
0
2468
p
r
o
v
id
i
n
g
th
e
n
ec
ess
ar
y
in
f
r
a
s
tr
u
ctu
r
e
an
d
r
eso
u
r
ce
s
to
ca
r
r
y
o
u
t
th
is
r
esear
ch
wo
r
k
in
th
e
lab
o
r
ato
r
y
of
Dep
ar
tm
en
t
of
C
o
m
p
u
ter
Scie
n
ce
at
R
ath
in
am
C
o
lleg
e
of
Ar
ts
an
d
Scien
ce
,
C
o
im
b
ato
r
e.
AUTHO
R
CO
NT
RI
B
UT
I
O
N
ST
A
T
E
M
E
NT
T
h
is
jo
u
r
n
al
u
s
es
th
e
C
o
n
t
r
ib
u
to
r
R
o
les
T
a
x
o
n
o
m
y
(
C
R
ed
iT)
to
r
ec
o
g
n
ize
in
d
iv
i
d
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
llab
o
r
atio
n
.
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
K
i
s
h
o
r
e
K
u
m
a
r
K
a
m
a
r
a
j
u
g
a
d
d
a
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
Mo
v
v
a
Pav
an
i
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
R
a
n
i
Va
n
a
t
h
i
G
u
r
u
s
a
m
y
✓
✓
✓
✓
✓
✓
✓
✓
✓
Nag
ar
ajan
Kar
th
ik
e
y
an
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
P
a
v
a
n
K
u
m
a
r
N
i
d
u
m
o
l
u
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
D
e
s
i
d
i
N
a
r
s
i
m
h
a
R
e
d
d
y
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
Mu
n
iap
p
an
R
am
ar
aj
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
R
ajasek
ar
an
Nith
y
a
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
T
h
e
au
th
o
r
s
d
ec
lar
e
th
at
th
e
y
h
av
e
n
o
co
n
f
licts
o
f
in
ter
est r
e
lated
to
th
is
wo
r
k
.
DATA AV
AI
L
AB
I
L
I
T
Y
Data
a
v
aila
b
i
lit
y
is
n
o
t
a
p
p
li
ca
b
le
t
o
t
h
is
p
ap
er
as
n
o
n
e
w
d
a
t
a
w
er
e
cr
ea
te
d
o
r
an
al
y
z
e
d
i
n
t
h
is
s
t
u
d
y
.
RE
F
E
R
E
NC
E
S
[
1
]
J.
C
.
M
.
B
u
s
t
i
l
l
o
,
“
O
p
t
i
mi
z
a
t
i
o
n
-
b
a
s
e
d
t
e
c
h
n
i
q
u
e
s
p
r
e
d
i
c
t
i
o
n
m
o
d
e
l
i
n
d
e
t
e
r
m
i
n
i
n
g
e
m
p
l
o
y
e
e
t
u
r
n
o
v
e
r
,
”
Pro
c
e
d
i
a
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
2
5
2
,
p
p
.
4
4
0
–
4
4
9
,
2
0
2
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
r
o
c
s.
2
0
2
5
.
0
1
.
0
0
3
.
[
2
]
J.
P
a
n
e
t
a
l
.
,
“
P
r
e
d
i
c
t
i
o
n
o
f
p
l
a
q
u
e
p
r
o
g
r
e
ssi
o
n
u
s
i
n
g
d
i
f
f
e
r
e
n
t
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
mo
d
e
l
s
o
f
p
e
r
i
c
o
r
o
n
a
r
y
a
d
i
p
o
se
t
i
ss
u
e
r
a
d
i
o
mi
c
s
b
a
s
e
d
o
n
c
o
r
o
n
a
r
y
c
o
mp
u
t
e
d
t
o
m
o
g
r
a
p
h
y
a
n
g
i
o
g
r
a
p
h
y
,
”
Eu
r
o
p
e
a
n
J
o
u
rn
a
l
o
f
R
a
d
i
o
l
o
g
y
O
p
e
n
,
v
o
l
.
1
4
,
J
u
n
.
2
0
2
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
j
r
o
.
2
0
2
5
.
1
0
0
6
3
8
.
[
3
]
A
.
K
a
n
w
a
r
,
R
.
M
.
S
i
n
g
h
,
K
.
L
a
t
a
,
A
.
D
h
i
ma
n
,
a
n
d
P
.
S
i
n
g
h
,
“
A
n
e
f
f
e
c
t
i
v
e
me
t
h
o
d
o
l
o
g
y
f
o
r
i
d
e
n
t
i
f
y
i
n
g
a
d
v
e
r
se
d
r
u
g
r
e
a
c
t
i
o
n
s
u
si
n
g
f
i
r
e
f
l
y
a
l
g
o
r
i
t
h
m
,
”
Pro
c
e
d
i
a
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
2
5
8
,
p
p
.
4
0
6
0
–
4
0
6
9
,
2
0
2
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
r
o
c
s.
2
0
2
5
.
0
4
.
6
5
7
.
[
4
]
B
.
A
l
n
a
s
y
a
n
,
M
.
B
a
s
h
e
r
i
,
a
n
d
M
.
A
l
a
ssafi,
“
Th
e
p
o
w
e
r
o
f
d
e
e
p
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s
f
o
r
p
r
e
d
i
c
t
i
n
g
st
u
d
e
n
t
p
e
r
f
o
r
man
c
e
i
n
v
i
r
t
u
a
l
l
e
a
r
n
i
n
g
e
n
v
i
r
o
n
m
e
n
t
s:
a
sy
s
t
e
mat
i
c
l
i
t
e
r
a
t
u
r
e
r
e
v
i
e
w
,
”
C
o
m
p
u
t
e
rs
a
n
d
Ed
u
c
a
t
i
o
n
:
Ar
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
6
,
Ju
n
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
a
e
a
i
.
2
0
2
4
.
1
0
0
2
3
1
.
[
5
]
J
.
Z
h
o
u
,
Y
.
Z
h
a
n
g
,
Y
.
Q
i
u
,
K
.
P
e
n
g
,
a
n
d
M
.
K
h
a
n
d
e
l
w
a
l
,
“
E
n
h
a
n
c
i
n
g
t
u
n
n
e
l
s
a
f
e
t
y
w
i
t
h
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
m
o
d
e
l
s
f
o
r
g
r
o
u
n
d
b
e
h
a
v
i
o
r
p
r
e
d
i
c
t
i
o
n
,
”
T
u
n
n
e
l
l
i
n
g
a
n
d
U
n
d
e
r
g
r
o
u
n
d
S
p
a
c
e
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
1
6
5
,
N
o
v
.
2
0
2
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
t
u
s
t
.
2
0
2
5
.
1
0
6
8
8
8
.
[
6
]
R
.
B
o
e
h
m
e
,
S
.
C
o
o
r
s,
P
.
O
st
e
r
,
M
.
M
.
-
K
i
e
f
e
r
,
a
n
d
S
.
H
i
l
b
e
r
t
,
“
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
f
o
r
s
p
e
l
l
i
n
g
a
c
q
u
i
si
t
i
o
n
:
h
o
w
a
c
c
u
r
a
t
e
i
s
t
h
e
p
r
e
d
i
c
t
i
o
n
o
f
s
p
e
c
i
f
i
c
sp
e
l
l
i
n
g
e
r
r
o
r
s
i
n
g
e
r
ma
n
p
r
i
m
a
r
y
s
c
h
o
o
l
st
u
d
e
n
t
s
?
,
”
C
o
m
p
u
t
e
rs
a
n
d
Ed
u
c
a
t
i
o
n
:
Ar
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
6
,
J
u
n
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
a
e
a
i
.
2
0
2
4
.
1
0
0
2
3
3
.
[
7
]
D
.
L.
S
h
i
v
a
p
r
a
sa
d
,
D
.
S
.
G
u
r
u
,
a
n
d
R
.
K
a
v
i
t
h
a
,
“
A
n
o
v
e
l
a
p
p
r
o
a
c
h
t
o
w
a
r
d
s
k
e
y
-
p
o
i
n
t
b
a
se
d
r
e
a
l
-
t
i
me
c
h
i
l
d
r
e
n
e
mo
t
i
o
n
p
r
e
d
i
c
t
i
o
n
,
”
Pr
o
c
e
d
i
a
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
2
5
8
,
p
p
.
5
8
8
–
5
9
8
,
2
0
2
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
r
o
c
s
.
2
0
2
5
.
0
4
.
2
9
3
.
[
8
]
A
.
R
.
S
a
y
e
d
,
M
.
H
.
K
h
a
f
a
g
y
,
M
.
A
l
i
,
a
n
d
M
.
H
.
M
o
h
a
m
e
d
,
“
Ex
p
l
o
r
i
n
g
t
h
e
v
a
k
mo
d
e
l
t
o
p
r
e
d
i
c
t
s
t
u
d
e
n
t
l
e
a
r
n
i
n
g
s
t
y
l
e
s
b
a
se
d
o
n
l
e
a
r
n
i
n
g
a
c
t
i
v
i
t
y
,
”
I
n
t
e
l
l
i
g
e
n
t
S
y
s
t
e
m
s
w
i
t
h
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
2
5
,
M
a
r
.
2
0
2
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
sw
a
.
2
0
2
5
.
2
0
0
4
8
3
.
[
9
]
A
.
C
.
K
ı
l
ı
ç
,
A
.
K
a
r
a
k
u
ş
,
a
n
d
E
.
A
l
p
t
e
k
i
n
,
“
P
r
e
d
i
c
t
i
o
n
o
f
u
n
i
v
e
r
s
i
t
y
s
t
u
d
e
n
t
s
’
s
u
b
j
e
c
t
i
v
e
w
e
l
l
-
b
e
i
n
g
w
i
t
h
s
l
e
e
p
a
n
d
p
h
y
s
i
c
a
l
a
c
t
i
v
i
t
y
d
a
t
a
u
s
i
n
g
c
l
a
s
s
i
f
i
c
a
t
i
o
n
a
l
g
o
r
i
t
h
m
s
,
”
P
r
o
c
e
d
i
a
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
2
0
7
,
p
p
.
2
6
4
8
–
2
6
5
7
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
r
o
c
s
.
2
0
2
2
.
0
9
.
3
2
3
.
[
1
0
]
O
.
M
.
El
-
D
e
e
b
,
W
.
E
l
b
a
d
a
w
y
,
a
n
d
D
.
S
.
El
z
a
n
f
a
l
y
,
“
T
h
e
e
f
f
e
c
t
o
f
i
m
b
a
l
a
n
c
e
d
c
l
a
s
ses
o
n
s
t
u
d
e
n
t
s’
a
c
a
d
e
mi
c
p
e
r
f
o
r
m
a
n
c
e
p
r
e
d
i
c
t
i
o
n
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
e
-
C
o
l
l
a
b
o
r
a
t
i
o
n
,
v
o
l
.
1
8
,
n
o
.
1
,
p
p
.
1
–
1
7
,
J
u
l
.
2
0
2
2
,
d
o
i
:
1
0
.
4
0
1
8
/
I
JeC
.
3
0
4
3
7
3
.
[
1
1
]
J.
C
h
e
n
,
X
.
Z
h
o
u
,
J
.
Y
a
o
,
a
n
d
S
.
-
K
.
T
a
n
g
,
“
A
p
p
l
i
c
a
t
i
o
n
o
f
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
i
n
h
i
g
h
e
r
e
d
u
c
a
t
i
o
n
t
o
p
r
e
d
i
c
t
st
u
d
e
n
t
s’
p
e
r
f
o
r
m
a
n
c
e
,
l
e
a
r
n
i
n
g
e
n
g
a
g
e
m
e
n
t
a
n
d
se
l
f
-
e
f
f
i
c
a
c
y
:
a
sy
s
t
e
m
a
t
i
c
l
i
t
e
r
a
t
u
r
e
r
e
v
i
e
w
,
”
Asi
a
n
E
d
u
c
a
t
i
o
n
a
n
d
D
e
v
e
l
o
p
m
e
n
t
S
t
u
d
i
e
s
,
v
o
l
.
1
4
,
n
o
.
2
,
p
p
.
2
0
5
–
2
4
0
,
M
a
r
.
2
0
2
5
,
d
o
i
:
1
0
.
1
1
0
8
/
A
ED
S
-
08
-
2
0
2
4
-
0
1
6
6
.
[
1
2
]
M
.
B
i
l
a
l
,
H
.
I
sr
a
r
,
M
.
S
h
a
h
i
d
,
a
n
d
A
.
K
h
a
n
,
“
S
e
n
t
i
m
e
n
t
c
l
a
ssi
f
i
c
a
t
i
o
n
o
f
r
o
man
-
u
r
d
u
o
p
i
n
i
o
n
s
u
s
i
n
g
n
a
ï
v
e
b
a
y
e
s
i
a
n
,
d
e
c
i
si
o
n
t
r
e
e
a
n
d
k
n
n
c
l
a
ss
i
f
i
c
a
t
i
o
n
t
e
c
h
n
i
q
u
e
s,
”
J
o
u
rn
a
l
o
f
K
i
n
g
S
a
u
d
U
n
i
v
e
rs
i
t
y
-
C
o
m
p
u
t
e
r
a
n
d
I
n
f
o
rm
a
t
i
o
n
S
c
i
e
n
c
e
s
,
v
o
l
.
2
8
,
n
o
.
3
,
p
p
.
3
3
0
–
3
4
4
,
J
u
l
.
2
0
1
6
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
k
su
c
i
.
2
0
1
5
.
1
1
.
0
0
3
.
[
1
3
]
X
.
G
u
o
e
t
a
l
.
,
“
L
o
n
g
i
t
u
d
i
n
a
l
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
p
r
e
d
i
c
t
i
o
n
o
f
n
o
n
-
su
i
c
i
d
a
l
sel
f
-
i
n
j
u
r
y
a
mo
n
g
c
h
i
n
e
s
e
a
d
o
l
e
s
c
e
n
t
s:
a
p
r
o
s
p
e
c
t
i
v
e
mu
l
t
i
c
e
n
t
e
r
c
o
h
o
r
t
st
u
d
y
,
”
J
o
u
r
n
a
l
o
f
Af
f
e
c
t
i
v
e
D
i
so
r
d
e
rs
,
v
o
l
.
3
9
2
,
J
a
n
.
2
0
2
6
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
a
d
.
2
0
2
5
.
1
2
0
1
1
0
.
Evaluation Warning : The document was created with Spire.PDF for Python.