Co
m
pu
ter Sci
ence a
nd Inf
or
mat
i
on
Tec
h
no
lo
gies
Vo
l.
1
, No
.
2
,
J
ul
y
2020
,
pp.
47
~
53
IS
S
N:
27
22
-
3221
,
DOI: 10
.11
591
/
csi
t.v
1i
2
.p
47
-
53
47
Journ
al h
om
e
page
:
http:
//
ia
esprime
.com/i
ndex.
php/csit
Buil
din
g a
mu
ltili
ngual
ontology fo
r educati
on dom
ain
using
mo
nto m
ethod
Merli
n
Florre
nce
Depa
rt
m
ent
o
f
C
om
pute
r
Applica
ti
ons
,
Sa
cre
d
He
art
Co
ll
eg
e, T
i
ru
pat
tur
,
Ind
ia
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
Sep
24
, 20
19
Re
vised
Ma
y
1
, 20
2
0
Accepte
d
Ma
y
21
, 20
2
1
Ontologi
es
ar
e
e
m
erg
ing
t
ec
hno
l
og
y
in
bu
il
ding
k
nowledge
base
d
informati
on
ret
ri
eva
l
s
y
st
ems
.
It
is
used
to
concept
ua
li
z
e
the
informa
ti
on
in
hum
an
under
standa
b
le
m
anne
r.
Know
l
edge
base
d
info
rm
at
ion
r
et
ri
eval
a
re
wid
e
l
y
used
in
the
dom
ai
n
li
k
e
Educ
a
tion,
Ar
t
ifi
c
ia
l
Int
el
li
g
ence,
Hea
lthcar
e
and
s
o
on.
I
t
is
importa
nt
to
prov
ide
m
ult
ilingual
infor
m
at
ion
of
those
dom
ai
ns
to
fac
i
li
t
at
e
m
ult
i
-
la
nguag
e
users.
In
thi
s
p
ape
r
,
we
propose
a
MO
nt
o
(Multi
li
ngu
al
Ontolog
y
)
m
et
ho
dolog
y
to
dev
e
lop
m
ult
ilingua
l
ontolo
g
y
app
lications
for
educ
a
ti
on
dom
ain.
N
ew
al
gori
th
m
s
are
proposed
for
m
erg
ing
and
m
appi
ng
m
ult
ilingual onto
lo
gie
s.
Ke
yw
or
d
s
:
Me
thodo
l
og
y,
Mult
il
ing
ual
,
On
t
ology
This
is an
open
acc
ess arti
cl
e
un
der
the
CC
B
Y
-
SA
l
ic
ense
.
Corres
pond
in
g
Aut
h
or
:
Me
rlin Florre
nc
e,
Dep
a
rtm
ent o
f C
om
pu
te
r
A
pp
li
cat
ion
s,
Sacre
d Heart C
ollege,
Tir
up
at
tur, I
nd
ia
.
Em
a
il
:
m
erli
nf
lor
ren
ce
@
gm
ail.co
m
1.
INTROD
U
CTION
On
t
ology
e
na
bles
the
nat
ural
la
ngua
ge
proces
sin
g
of
the
data
in
an
eff
ic
ie
nt
way
.
It
retriev
es
the
in
form
at
io
n
base
d
on
th
e
knowle
dge
and
co
nce
ptu
a
li
zes
the
in
for
m
at
ion
in
for
m
al
way.
En
orm
ou
s
inf
or
m
at
ion
is
avail
able
over
the
i
ntern
et
in
a
s
pecific
la
ng
uag
e
.
It
is
es
se
ntial
to
pr
ov
i
de
the
in
f
or
m
at
i
on
i
n
diff
e
re
nt
na
tur
al
la
ngua
ges
t
o
be
nef
it
m
ulti
-
la
ngua
ge
us
e
rs.
O
nto
l
og
ie
s
pla
y
vital
r
ole
in
pro
vid
i
ng
kn
owle
dge
base
d
in
form
ation
syst
em
s.
O
nto
lo
gy
is
a
“
f
or
m
al
,
exp
li
ci
t
sp
eci
ficat
io
n
of
a
s
har
e
d
c
onc
eptuali
zat
ion
”
[
1]
.
It
is
a
c
ollec
ti
on
of
set
of
c
on
ce
pts,
pro
per
ti
es,
relat
io
ns
,
inst
ances,
axi
om
s
and
r
ules
w
hic
h
ca
n
be
re
pr
e
sente
d
as,
(O
ntolo
gy)
O
=
{C,
P,
R
,
I,
A
}.
‘C’
re
pr
ese
nts
the
cl
asses
or
co
nce
pts
of
the
do
m
ai
n.
‘
P’
si
gn
ifi
es
the
pro
per
ti
es
of
the
c
on
ce
pt.
‘
R’
de
note
s
th
e
bin
a
ry
r
el
at
ion
s
betwee
n
t
he
c
on
ce
pts
(
1
-
1,
1
-
M,
M
-
M).
‘A’
represe
nts
axi
om
s
and
r
ules
w
hich
a
re
us
e
d
a
s
a
basis
f
or
r
ea
so
ni
ng
[
2]
.
In
o
nto
lo
gy
a
set
of
te
rm
s
fo
r
desc
r
ibing
a
dom
ai
n
is
a
r
range
d
hiera
rc
hical
ly
that
ca
n
be
use
d
as
a
s
kelet
al
f
ound
at
ion
f
or
a
kn
owle
dgeba
se
[
1]
.
This
natu
re
of
ontol
og
y e
na
bles th
e d
e
velo
per to
i
m
ple
m
ent se
m
antic
b
ase
d pe
r
so
na
li
zed lear
ni
ng
a
ppli
cat
ions.
The
on
t
ology
de
velo
ped
f
or
t
he
edu
cat
i
on
al
dom
ai
n
con
ta
ins
the
knowle
dge
for
de
velo
pi
ng
intel
li
gen
t
le
arn
in
g
syst
em
.
Mon
olin
gual
ontolo
gy
a
pp
li
cat
io
ns
for
le
arn
i
ng
syst
e
m
are
be
de
velo
ped
by
ad
op
ti
ng
the
m
e
tho
dol
ogy
[
3]
.
O
nto
lo
gies
are
us
e
d
to
re
pr
e
sent
know
le
dg
e
w
hich
re
flect
s
the
r
el
evan
t
in
form
at
ion
of
the
co
nce
pts
and
relat
ion
s
.
T
her
e
wer
e
m
an
y
m
et
ho
do
l
og
i
es
pro
posed
t
o
bu
il
d
ontol
ogy
app
li
cat
ion
s
wh
ic
h
hav
e
their
ow
n
pitfal
ls.
Mo
deling,
e
valuat
ing
a
nd
m
ai
nt
ai
nin
g
ontol
ogie
s
are
a
c
omplex
ta
s
ks
i
n
m
os
t
app
li
cat
io
ns
s
uc
h
as
healt
hcare
,
bu
si
ness,
c
om
m
erce
and
m
any
oth
er
.
T
he
r
e
are
m
any
dom
ai
ns
tha
t
nec
essit
at
e
sat
isfyi
ng
t
he
di
ff
ere
nt
la
ngua
ge
us
e
rs.
F
or
e
xam
ple
the
us
e
rs
of
governm
e
nt
se
rv
ic
es,
le
a
rn
i
ng
sit
es,
e
du
cat
ion
do
m
ai
ns
,
healt
hcar
e
dom
ai
ns
dem
and
s
to
acc
ess
in
form
at
io
n
i
n
thei
r
l
ocal
la
nguag
e
.
In
s
uc
h
sce
na
rio,
on
tolog
y
play
s
a
vital
ro
le
to
pro
vid
e
knowle
dge
ba
sed
in
form
at
io
n.
Nu
m
ero
us
m
et
ho
ds
a
nd
too
ls
a
re
pro
po
sed
f
or
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2722
-
3221
Com
pu
t. Sci.
I
nf. Tec
hnol.
,
V
ol.
1
, N
o.
2
,
J
ul
y
20
20
:
47
–
53
48
bu
il
di
ng
m
onol
ing
ual
ont
ologies.
Ve
ry
fe
w
m
et
ho
ds
li
ke
C
ollab
or
at
i
ve
platf
or
m
are
pro
posed
t
o
buil
d
m
ul
ti
li
ng
ual
ontolo
gies
bu
t
t
hey
are
li
m
i
ted
to
s
om
e
la
nguag
e
s.
T
his
ch
a
pter
pro
po
ses
new
m
et
hodo
l
og
y
t
o
bu
il
d
m
ulti
l
ingual
ontol
og
ie
s
.
Ra
pid
de
velo
pm
ent
of
i
nter
net
use
rs
dem
and
s
on
i
nfo
rm
at
ion
in
their
na
tural
la
nguag
e
s
w
hich
le
a
ds
t
o
t
he
dev
el
op
m
ent
of
m
ulti
li
ng
ual
a
pp
li
cat
io
ns
.
Th
e
ai
m
of
this
pa
per
is
t
o
giv
e
a
n
i
de
a
to
d
e
velo
p
m
ulti
li
ng
ual
on
t
ologies
f
or
ed
ucati
on
dom
ain
us
in
g
the
pr
opos
e
d
M
On
t
o
m
et
ho
dolo
gy
.
Ne
w
al
gorithm
s
are
pro
posed
f
or
m
erg
ing
a
nd
m
app
ing
ontolo
gies
dev
el
oped
in
dif
fer
e
nt
nat
ur
al
la
ngua
ges
.
The
pa
pe
r
org
anized
as
fo
ll
ows:
an
ov
e
r
view
of
ontol
og
y
base
d
le
arn
i
ng
syst
e
m
s
are
nar
rated
i
n
sect
ion
2.
Sect
ion
3
pr
opos
es
a
new
m
et
hodolo
gy
to
bu
il
d
m
ulti
li
ng
ua
l
on
tol
og
ie
s
C
oncl
us
i
on
s
a
re
pro
posed
i
n
sect
ion
4.
2.
STATE
-
OF
-
T
HE
-
A
RT
OF
ONTOLO
GY
-
BASE
D
LE
A
RN
I
NG
Learn
i
ng
O
ntol
og
ie
s
are
us
e
d
in
s
of
t
war
e
agen
ts
,
la
ngu
a
ge
in
dep
e
nde
nt
app
li
cat
ion
s
and
pro
blem
so
lvi
ng
m
et
ho
ds
.
On
t
ology
app
li
cat
io
ns
are
be
de
velo
pe
d
usi
ng
on
t
ology
dev
el
op
m
ent
la
ngua
ges
(
O
WL
,
RDF
,
TURTLE
,
Tri
pl
e
and
so
on)
an
d
on
t
ology
de
ve
lop
m
ent
too
ls
(P
r
otégé
,
On
t
oE
dit,
Chim
aera
an
d
s
o
on
).
Le
arn
i
ng
on
t
ol
ogy
ap
plica
ti
on
are
be
i
m
ple
m
ented
in
two
dif
fer
e
nt
strat
egies:
i)
ontolo
gy
of
le
ar
ning
re
source
s
and
ii
)
on
t
ology
of
te
achin
g
st
rategy
[
4]
.
The
on
t
ology
of
le
ar
ning
r
eso
ur
ces
is
us
e
d
for
te
ac
hing
knowle
dge
m
od
el
in
g
in e
-
le
a
rn
i
ng s
yst
e
m
. Th
e
ontolo
gy
of te
achi
ng strate
gies e
xh
i
bits a s
eries
of m
acro
teac
hing
desig
n an
d
m
ic
ro
te
aching
act
ivit
ie
s.
O
nto
l
og
y
f
or
le
ar
ning
m
a
y
have
per
s
ona
li
zed
le
arn
i
ng
paths
[
5]
w
hic
h a
re
us
e
d t
o
i
m
pr
ov
e
the
ef
fecti
ven
e
ss
of
le
ar
ning
s
yst
e
m
.
Persona
li
zat
ion
of
e
-
le
arn
i
ng
proces
s
for
the
c
hose
n
ta
rg
et
gr
oup
w
il
l
be
achieve
d
by
s
et
ti
ng
up
the
l
earn
i
ng
path
f
or
eac
h
us
er
a
ccordin
g
to
t
he
ir
pro
file
.
Som
e
m
od
el
s
ha
ve
bee
n
pro
po
se
d
to
de
velo
p
we
b
bas
ed
e
-
le
ar
ning
s
yst
e
m
s
[
6]
.
Th
ese
m
od
el
ha
ve
been
de
velo
pe
d
base
d
on
se
m
antic
web
te
c
hno
l
ogie
s
and
e
-
le
ar
ni
ng
sta
nd
a
r
ds
.
These
m
od
el
s
pro
vid
e
tw
o
ki
nd
s
of
c
on
te
nts
to
the
le
ar
ne
rs
,
they
are:
i)
Lea
rn
i
ng
c
onte
nt
an
d
ii
)
As
sessm
ent
co
nte
nt
a
nd
pro
vid
es
le
ar
ning
ser
vice
a
nd
assessm
ent
ser
vice
resp
ect
ively
.
T
hese
m
od
el
s
use
the
kn
ow
le
dg
e
base
d
i
n
f
orm
ation
retrie
va
l
appro
ac
h
t
o
reposses
s
le
ar
ning
resou
rces. The
le
arn
in
g res
our
ces are
desc
rib
ed by m
eans o
f
m
e
ta
data to i
m
ple
m
ent the
knowle
dge
bas
e.
So
m
e
on
tolo
gy
based
le
a
rn
i
ng
syst
e
m
s
hav
e
been
de
velo
pe
d
to
sto
re
an
d
r
et
rieve
sem
antic
m
et
adata
to pr
ovid
e b
et
t
er r
es
ults to t
he
learne
r
al
on
g wit
h perso
naliz
ed
le
ar
ning
[
7]
. A syst
em
atic appr
oach
is
propo
s
ed
towa
rd
s
the
de
velo
pm
ent
of
s
e
m
antic
web
s
erv
ic
es
for
e
–
le
arn
i
ng
dom
ain
.
T
he
fo
ll
owi
ng
ste
ps
[8]
a
re
us
e
d
to
dev
el
op
ontol
og
y
for
e
-
le
ar
ning:
i)
deter
m
ining
the
sc
op
e
of
do
m
ain
,
ii
)
reusi
ng
existi
ng
ontol
og
ie
s
,
ii
i)
enu
m
erati
ng
im
po
rtant
te
rm
s
in
the
ontol
ogy,
iv)
def
i
ning
the
cl
asses
an
d
it
s
hie
rar
c
hy,
v)
de
fini
ng
the
cl
as
s
pro
per
ti
es,
vi
)
def
i
ning
t
he
fa
cet
s,
vii)
c
reati
ng
in
sta
nces
a
nd
viii
)
c
heck
i
ng
a
no
m
al
y.
The
ontolo
gies
can
be
evaluate
d
us
i
ng
SRI
O
NTO
(
So
ft
war
e
Ri
s
k
Id
e
ntific
at
ion
ON
T
Ology)
t
o
identify
the
pro
blem
and
risk
i
n
it
[
9]
.
The
re
quire
d
c
on
ce
pts,
t
he
se
m
antic
descr
ip
ti
on
of
the
c
oncepts
an
d
the
i
nterr
el
at
io
nship
am
on
g
the
c
on
ce
pts
al
ong
with
al
l
oth
e
r
ontol
og
ic
al
com
po
ne
nts
hav
e
bee
n
c
ollec
te
d
from
var
iou
s
li
te
ratur
es
.
E
-
le
ar
ning
res
ource
s
can
be
c
ollec
te
d
us
in
g
s
ome
fr
am
ew
orks
[
10]
.
T
hese
fra
m
ewo
r
ks
us
e
d
to
colle
ct
e
-
le
arn
in
g
m
ultim
edia
resou
rces
from
the inte
rn
et
a
nd a
uto
m
at
ic
all
y l
ink
them
w
it
h
to
pics.
On
t
ology
-
base
d
a
ppro
a
ch
ca
n
be
us
e
d
to
de
ve
lop
pe
rsonali
zed
e
-
le
a
rn
i
ng
[
11]
.
It
is
us
ed
to
create
a
n
adap
ti
ve
c
on
te
nt
ba
sed
on
le
arn
e
r’
s
a
bili
ti
e
s,
le
ar
ning
sty
le
,
le
vel
of
kn
ow
le
dg
e
a
nd
pr
e
fer
e
nces.
I
n
this
appr
oach,
ont
ol
og
y
is
us
ed
to
represe
nt
the
c
on
te
nt
m
od
el
,
le
arn
e
r
m
od
el
a
nd
dom
ai
n
m
od
el
.
T
he
c
on
te
nt
m
od
el
descr
i
bes
the
s
tructu
re
of
c
ou
rses
an
d
their
com
po
ne
nts.
T
h
e
le
a
rn
e
r
m
odel
descr
i
bes
the
c
ha
racteri
sti
cs
of
le
arn
er
’s
that
a
re
require
d
to
de
li
ver
ta
il
ored
con
te
nt.
The
dom
ai
n
m
od
el
c
on
sist
s
of
so
m
e
cl
asses
a
nd
pr
op
e
rtie
s
to
def
i
ne
dom
a
in
to
pics
an
d
se
m
antic
relat
ionships
betwee
n
them
.
It
is
us
e
d
to
asse
ss
t
he
le
a
rn
e
r’
s
perfor
m
ance
by
c
onduct
in
g
the
te
sts
an
d
th
e
res
ults
a
re
ev
al
uated.
T
he
s
yst
e
m
recogn
iz
es
c
hanges
in
t
he
le
a
rn
e
r’s
le
vel
of
knowle
dge
as
they
pro
gr
ess
and
t
he
le
ar
ne
r
m
od
el
is
upda
te
d
base
d
on
the
le
arn
e
r’s
pro
gr
e
ss
acco
r
di
ng
ly
.
Howe
ver,
m
os
t
of
t
he
le
arn
i
ng
app
li
cat
io
ns
ar
e
dev
el
op
e
d
ei
t
her
i
n
En
glish
or
i
n
the
dev
el
op
e
r
la
ng
uag
e
wh
ic
h
beco
m
e
the
hu
rd
le
s
of
dif
fere
nt
la
ngua
ge
use
rs
to
le
a
rn.
Nowa
days
us
e
rs
of
inter
net
pr
e
fer
t
o
sha
re
their
knowle
dge
in
t
heir
natu
ral
la
ngua
ges
w
hich
em
erg
es
the
t
echno
lo
gies
t
o
sup
port
di
ff
e
r
ent
nat
ur
al
la
ngua
ges
.
In
a
c
urre
nt
sc
enar
i
o,
e
norm
ou
s
le
ar
ning
m
a
te
rial
s
are
avail
able
over
the w
eb
w
hich
al
lo
ws
the u
se
r
to b
ene
fit
from
anywh
e
re
in
the
w
or
l
d.
Th
ough
the
use
r
gets
la
r
ge
a
m
ou
nt
of
in
for
m
at
ion
sti
ll
they
are
lo
ng
i
n
g
f
or
t
he
inf
or
m
at
ion
in
their
own
la
ng
uag
e
s.
This
m
otivate
s
us
to
de
velo
p
m
ulti
li
ng
ual
on
t
ology
a
pp
li
cat
io
ns
t
o
be
nef
it
diff
e
re
nt
nat
ural
la
ngua
ges
.
I
n
orde
r
to
do
that,
MO
nto
m
et
hodo
l
og
y
is
pro
posed
t
o
bu
il
d
m
ul
ti
li
ng
ual on
tolog
ie
s
.
3.
MO
NTO
ME
THOD
OLOG
Y
TO
D
E
VELOP
MU
LT
ILI
NGUAL
ONT
OLOGIE
S
A
m
et
ho
dolo
gy
is
a
“com
pr
e
hensi
ve,
i
nteg
r
at
ed
series
of
t
echn
i
qu
e
s
or
m
et
ho
ds
creati
ng
a
gen
e
ral
syst
e
m
s
theor
y
of
how
a
cl
ass
of
t
hought
-
int
ensive
w
ork
ought
to
be
perf
or
m
ed”
[
12
]
.
Me
thodo
l
og
y
c
on
sist
s
of
m
et
ho
ds
a
nd
te
c
hn
i
qu
es
w
her
e
m
et
ho
d
is
a
process
of
pe
rfor
m
ing
s
ome
ta
sk
a
nd
te
c
hniq
ue
is
a
proc
edure
us
e
d
to
ac
hiev
e
giv
e
n
obj
ect
ive.
T
his
rese
arch
w
ork
pro
po
s
es
MO
nto
m
et
ho
dolo
gy
to
buil
d
m
ulti
lin
gual
Evaluation Warning : The document was created with Spire.PDF for Python.
Com
pu
t. Sci.
I
nf. Tec
hnol.
Buildin
g a m
ulti
li
ng
ua
l
ontol
og
y
for
educ
ation d
omai
n usi
ng MOnt
o method
… (
Merl
in
Florre
nce
)
49
on
t
ology
ap
plica
ti
on
s.
This
m
et
hodo
l
og
y
c
onsist
s
of
fi
ve
phases
as
giv
e
n
in
Fig
ur
e
1.
viz
.
I
nput,
B
uildin
g
MO
,
On
t
ology m
edi
at
ion
, Retrie
va
l and Vis
ualiz
at
ion
of ontol
ogy.
Figure
1
.
Mo
nto
m
et
ho
dolo
gy for
buil
ding m
ulti
li
ng
ual
onto
log
y
3.1.
Pha
se
1: I
npu
t
This
phase
i
niti
al
iz
es
th
e
con
te
nt
to
be
co
ns
ide
red
f
or
buil
ding
ontol
ogie
s.
A
set
of
m
et
ho
ds
a
nd
te
chn
iq
ues
ar
e
us
e
d
f
or
bu
il
ding
ontolo
gy
from
distribu
te
d
a
nd
heter
og
eneous
knowle
dg
e
a
nd
in
f
orm
at
ion
so
urces
.
I
nform
at
ion
can
be
retrieve
d
f
r
om
diff
ere
nt
s
ources
li
ke
,
op
e
n
co
r
pu
s
,
cl
ose
d
co
r
pu
s
an
d
existi
ng
on
t
ologies.
Al
l
the
sou
rces
are
un
der
thre
e
cat
egories:
Un
st
ru
ct
ur
e
d
s
ources,
sem
i
-
structu
red
s
our
ce
an
d
structu
re
d
s
our
ce.
Un
st
ru
ct
ured
s
ources
in
volve
NL
P
te
c
hniq
ues,
m
or
ph
ologica
l
an
d
s
ynta
ct
ic
analy
sis,
et
c.
Sem
i
-
structu
re
d
s
ource
el
ic
it
s
ontol
ogy
f
rom
so
ur
ces
tha
t
ha
ve
s
om
e
prede
fine
d
struc
ture,
s
uch
as
XM
L
Schem
a.
Stru
ct
ur
e
d
data
e
xtra
ct
s
con
ce
pts
an
d
relat
io
ns
f
rom
kn
owle
dge
c
on
ta
ine
d
in
st
r
uctu
red
d
at
a,
s
uch
as
databases
.
Cl
ose
d
c
orp
us
is
a
t
ext
f
ro
m
the
te
xt
bo
ok
s
,
st
ud
y
m
a
te
rial
s
et
c.
Op
e
n
c
orp
us
re
fer
s
t
o
t
he
in
for
m
at
ion
avail
able
on
t
he
we
b.
C
orpu
s
is
us
ed
to
repre
sent
the
repres
ents
ontol
og
y
by
us
in
g
a
set
of
te
chn
i
qu
es
to
extract
the
knowle
dge
f
ro
m
the
te
xt.
I
n
t
his
phase,
the
sc
ope
a
nd
do
m
ai
n
f
or
bu
i
lding
MO
is
i
de
nt
ifie
d.
I
n
or
de
r
to
bu
il
d
a
ne
w
on
tolog
y
f
or
the
sp
eci
fied
dom
a
in,
it
is
i
m
po
rtant
to
m
ake
su
re
that
there
is
any
ontolo
gy
a
lready
avail
able
t
o
t
he
pa
rtic
ular
do
m
ai
n.
I
n
t
hat
case
,
the
on
t
ology
ha
s
to
be
c
onsid
ered
f
or
reusi
ng
a
nd
re
-
e
ngine
erin
g
for
buil
din
g
M
O.
The
s
ources
f
or
bu
il
di
ng
MO
is
colle
ct
ed
as
giv
e
n
i
n
T
able
1.
Th
e
de
velo
per
has
t
o
identify
the
dom
ai
n
to
dev
el
op
M
O
and
has
to
c
ol
le
ct
the
infor
m
at
ion
from
var
io
us
sou
rces
in
diff
e
re
nt
na
tural
la
nguag
e
s.
T
he
co
ll
ect
ed res
ources
are
an
al
y
zed a
nd clas
sif
ie
d
in t
his initi
al
p
ha
se.
Table
1
.
D
ocum
ent
m
at
rix
f
or c
ollec
ti
ng
res
ources i
n diff
e
r
ent n
at
ur
al
la
ngua
ges
So
u
rce/L
an
g
u
ag
e
L1
L2
…
Ln
Op
en
corp
u
s
√
√
√
Clo
sed
corp
u
s
√
√
Exis
tin
g
on
to
lo
g
y
√
√
3.2.
Pha
se
2: Buil
ding
onto
l
ogie
s
On
ce
the
do
m
ai
n
is
i
den
ti
fie
d,
the
te
xt
e
xtr
act
ed
from
cl
os
ed
c
orp
us
an
d
open
c
orpus
in
dif
fer
e
nt
natu
ral
la
ng
ua
ges
is
ar
ra
ng
e
d
hiera
rch
ic
al
l
y
with
the
pr
op
e
r
cl
assi
ficat
ion
s.
T
he
te
r
m
s
req
uire
d
t
o
buil
d
m
ul
ti
li
ng
ual on
tolog
y a
re c
ollec
te
d
in
dif
fer
e
nt n
at
ur
al
la
nguag
e
s.
L
1
= L
1
t
1
, L
1
t
2
, L
1
t
3
,…………
…..L
1
t
m
L
2
= L
2
t
1
, L
2
t
2
, L
1
t
3
,…………
…..L
2
t
m
L
n
= L
n
t
1
, L
n
t
2
, L
1
t
3
,…………
…..L
n
t
m
This ca
n be
represente
d
as
,
∀
=
1
=
1
,
2
,
3
…
…
…
whe
re
≠
Coll
ect
ed
te
rm
s
are
a
naly
zed
and
ir
releva
nt
t
erm
s
are
filt
ere
d.
The
te
rm
s
ar
e
cl
assifi
ed
hie
rar
c
hical
ly
and the
relat
io
ns
betwee
n
t
he
term
s ar
e esta
blishe
d
as,
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2722
-
3221
Com
pu
t. Sci.
I
nf. Tec
hnol.
,
V
ol.
1
, N
o.
2
,
J
ul
y
20
20
:
47
–
53
50
→
The
relat
io
ns
be
tween
the
te
r
m
s
are
est
ablis
hed
a
nd
vocab
ularies
of
t
he
te
rm
s
are
form
ulate
d.
Usi
ng
the
la
dde
rin
g
struct
ur
e
ontol
og
ie
s
are
de
vel
oped
i
n
dif
fer
e
nt
natu
ral
la
ng
ua
ges
(
OL
1
,
O
L
2,
…,
OL
n
w
he
re
OL
is
On
t
ology
Lan
gu
a
ge
)
.
‘
N
’
O
nto
lo
gies
(
1
,
2
…
)
are
dev
el
op
e
d
for
natural
la
ngua
ges
usi
ng
the
te
rm
s th
at
are
hierar
c
hical
ly
stru
ct
ured
as s
how
n
i
n
Fi
gure
2.
Figure
2
.
Illust
rati
on of
bu
il
din
g o
nto
l
og
ie
s
for t
w
o natur
al
l
angua
ges
(Tam
il
an
d En
glish
)
3.3.
Pha
se
3:
Onto
logy
me
diat
i
on met
hods
On
t
ology
m
ediat
ion
ena
bles
r
eusin
g
of
data
acro
s
s
a
ppli
cat
ion
s
on
Sem
antic
W
e
b,
a
nd
s
ha
rin
g
of
data
betwee
n
heter
og
e
ne
ous
knowle
dge
bases
.
Ma
jor
kinds
of
ontol
ogy
m
ediat
ion
are
m
app
i
ng
a
nd
m
erg
i
ng
.
On
t
ology
m
app
in
g
is
to
i
den
t
ify
the
co
rr
es
ponde
nce
betwe
en
the
te
rm
s
and
ontol
ogy
m
er
ging
is
creati
ng
ne
w
on
t
ology
w
hich
is
the
un
i
on
of
existi
ng
tw
o
or
m
or
e
ontol
ogie
s.
I
n
this
ph
ase,
ontol
og
ie
s
dev
el
op
e
d
i
n
di
ff
e
ren
t
natu
ral
la
ng
ua
ges
are
m
erg
ed
i
nto
sin
gle
ontolo
gy
a
nd
t
he
co
rr
e
spo
nd
e
nces
bet
ween
t
he
te
rm
s
of
di
ff
e
ren
t
natu
ral
la
ngua
ges
a
re
est
a
blishe
d.
For
e
xa
m
ple,
OL
1
,
O
L
2
…
OL
n
are
the
ont
ologies
dev
el
op
e
d
in
di
ff
ere
nt
natu
ral la
ngua
ges for t
he
sel
e
ct
ed
dom
ai
n.
where
,
OL
1
= { L
1
t
1
,
L
1
t
2 ,
L
1
t
3 ,
…
L
1
t
i
}
OL
2
= { L
2
t
1
,
L
1
t
2 ,
L
1
t
3 ,
…
L
2
t
i
}
OL
n
= { L
n
t
1
,
L
n
t
2 ,
L
n
t
3 ,
…
L
n
t
i
}
On
t
ologies
devel
op
e
d
in
d
i
ff
e
ren
t
natu
ral la
ngua
ges
are
m
erg
ed
into
a si
ng
l
e ontolo
gy.
ML = {
OL
1
U
O
L
2
U
…
U O
L
n
}
Corresp
onde
nc
es b
et
ween t
he t
erm
s in
diff
e
r
ent
nat
ur
al
la
ngua
ges
are
c
re
at
ed
L
n
t
i
L
n
t
k
wh
e
re i a
nd
k vary
from
1
to i
term
s in
diff
e
ren
t l
a
ngua
ges.
Evaluation Warning : The document was created with Spire.PDF for Python.
Com
pu
t. Sci.
I
nf. Tec
hnol.
Buildin
g a m
ulti
li
ng
ua
l
ontol
og
y
for
educ
ation d
omai
n usi
ng MOnt
o method
… (
Merl
in
Florre
nce
)
51
On
t
ologies
t
ha
t
are
dev
el
op
e
d
in
di
ff
e
ren
t
natu
ral
la
ngua
ges
are
m
erg
e
d
into
si
ng
le
on
t
ology
t
o
structu
re m
ulti
l
ingual o
nto
l
og
y app
li
cat
io
n.
I
n form
al
, it can b
e
re
pr
ese
nted
as,
=
{
:
1
1
∪
2
2
∪
…
.
.
∪
}
ℎ
≥
2
˄
1
≠
1
∩
Her
e
, MO
–
Mult
il
ing
ual
O
nto
lo
gy
L
–
Lan
guage
X
–
Set
of ele
m
ents
X
is
a
c
ollec
ti
on
of
el
em
ents
or
te
rm
s
wh
ic
h
a
re
i
nteg
rated
th
e
s
ources
of
th
e
sam
e
dom
ai
n
in
diff
e
re
nt
natu
ral
la
ngua
ges.
Ma
ny
to
ols
li
ke
O
nto
Cl
e
an,
FCAMer
ge
,
an
d
Obser
ver
are
a
vaila
ble
to
m
erg
e
ontol
og
ie
s
.
The
m
erg
ed
on
tolog
y
c
om
po
sed
of
set
of
te
r
m
s
in
dif
fer
e
nt
na
tural
la
ng
ua
ges.
On
t
ology
m
erg
in
g
ca
n
be
do
ne
by
us
in
g SM
A
RT
al
gorithm
[
13]
. T
his
al
gor
it
h
m
deals
with
m
erg
ing
a
nd
al
ign
in
g o
f m
on
olin
gual
ontol
og
y
of
the
do
m
ai
n.
I
n
order
to
ove
rcom
e
this,
the
al
gorithm
s
for
ont
ology
m
ediat
ion
m
et
ho
ds
a
re
pro
po
se
d
f
or
m
erg
i
ng
and
m
app
in
g
ontol
ogy
[
14
-
2
1
]
.
T
he
r
esearch
a
da
pted
t
hose
al
gorithm
s
f
or
m
erg
in
g
a
nd
m
app
in
g
m
ul
ti
li
ng
ual on
tolog
ie
s
.
3.4.
Pha
se
4:
Mult
il
ingua
l
in
fo
r
mat
i
on
re
triev
al
usin
g
S
PA
RQL
Inform
at
ion
re
trie
val
is
t
he
proces
s
of
ret
rievin
g
or
e
xtra
ct
ing
t
he
i
nform
at
ion
f
ro
m
the
re
po
sit
or
y
base
d
on
t
he
use
r’
s
nee
d
a
nd
query.
Re
t
rievin
g
in
form
at
ion
in
var
i
ou
s
la
ng
uag
e
s
ca
n
be
nam
ed
as
m
ulti
l
ingual
inf
or
m
at
i
on
re
trie
val.
I
n
ontolo
gies,
S
PA
R
QL
qu
e
ry
is
use
d
to
extr
act
the
kn
ow
le
dg
e
from
the
on
t
olog
y
reposit
ory
.
RD
F
ta
gs
are
us
e
d
in
S
PA
R
QL
query
t
o
filt
er
t
he
res
ults
by
m
eans
of
la
ngua
ge
.
T
his
phase
e
nab
le
s
the u
ser
s
t
o
e
xt
ract kno
wled
ge
in
their
o
w
n
la
ngua
ges
us
i
ng SP
AR
QL.
S
P
ARQL
p
r
ovide
s
the
f
unct
io
na
li
ty
to
retrieve t
he
i
nfor
m
at
ion
in dif
fer
e
nt n
at
ur
al
l
angua
ges. The
sam
ple SPA
RQL
query is
gi
ven as
fo
ll
ows:
PREFI
X
scs: <
http:/
/ww
w.
s
hc
tptcs.o
rg#>
SELECT
?
Sub
j
ect
?
O
bject
WHER
E
{
?
s
ub
j
ect
sc
s
: v
erse
?
ob
j
ect
.
FILT
ER (
L
ang
(? o
b
jec
t)
=
"ta")
}
The
giv
e
n
SP
ARQL
use
d
‘
FI
LTER
’
t
o
s
or
t
t
he
res
ult
and
giv
e
t
he
r
esults
of
inf
or
m
at
ion
in
a
sp
eci
fied
lan
gu
age.
3.5.
Pha
se
5:
Visu
aliz
ing
multil
ingu
al
on
t
ol
ogy
Visu
al
iz
at
ion
is
a
re
presentat
i
on
of
te
xt
or obj
ect
i
n
t
he
f
or
m
of
i
m
age
or char
t.
It
e
na
bles
the r
ead
ers
to
ca
pture
t
he
kn
ow
le
dg
e
e
f
fecti
vely
.
On
t
ology
is
a
hie
rar
c
hical
ly
str
uctu
red
m
od
el
w
hich
has
nu
m
ero
us
visu
al
iz
at
ion
too
ls
(
O
WLGr
Ed,
Na
vigO
W
L,
I
sA
Viz
et
c)
an
d
plug
-
i
ns
(
On
t
oGraf,
O
WLv
iz
,
C
ropCirc
le
s
an
d
so
on
)
.
All
the
existi
ng
on
t
ology
visu
al
iz
at
io
n
to
ols
are
la
ck
ing
in
visu
al
iz
ing
non
-
E
ng
li
sh
la
nguag
es
.
S
om
e
of
them
req
ui
re
a
dd
it
io
nal
c
onfi
gurati
on
t
o
suppo
rt
diff
e
re
nt
nat
ur
al
la
ngua
ges.
I
n
this
phase,
the
new
plug
-
i
n
known
ML
Gr
a
fV
iz
is
pro
pose
d
to v
isuali
ze
ontolo
gy
in d
iff
eren
t natur
al
la
ngua
ges.
F
or
e
xam
ple
,
the
pa
ssage
giv
e
n
i
n
Fig
ure
3
is
re
pr
ese
nt
ed
diag
ram
m
a
t
ic
al
ly
in
Fig
ure
3
this
de
picts
that
t
he
grap
hi
cal
re
pr
ese
ntati
on
of
the text is cl
ea
r
er th
a
n
t
he pas
sage
w
her
e t
he
u
se
r
m
ay
f
eel
v
ag
ue w
hile re
adi
ng a
passa
ge
.
MLGra
fV
iz
is
dev
el
op
e
d
us
in
g
Java
a
nd
Gr
a
phviz
al
gorith
m
s.
In
it
ia
ll
y,
it
al
lows
t
he
us
e
r
t
o
c
reate
a
new
on
t
ology
or
to
im
po
rt
a
n
e
xisti
ng
ontolo
gy
int
o
Pro
té
gé
wor
ks
pac
e.
T
he
im
po
rted
ontolo
gy
w
il
l
be
disp
la
ye
d
in
a
cl
ass
br
ow
se
r.
MLGra
fV
iz
en
ables
t
he
us
e
r
t
o
sel
ect
the
la
ngua
ge
to
vis
ual
iz
e
the
ontol
ogy.
T
he
request
is
s
ubm
itted
to
Goo
gl
e
translat
e
A
P
I w
hich
pe
rform
s
statistical
m
achine
tra
ns
la
ti
on
a
nd
the
n
the
te
rm
s
are
tra
ns
la
te
d
i
nto
t
he
de
sire
d
natu
ral
la
ngua
ges.
G
oogle
t
ran
sla
te
AP
I
i
s
an
op
e
n
s
ourc
e
translat
or
use
d
to
translat
e
te
xt,
s
peech,
im
ages
and
vid
e
os
f
r
om
so
ur
ce
la
ng
uag
e
t
o
ta
rg
et
l
angua
ge.
It
pro
vid
es
a
n
A
PI
wh
ic
h
al
lows
t
he
de
ve
lop
e
r
t
o
bu
il
d
an
e
xten
sio
n
a
nd
s
of
t
war
e
to
translat
e
the
s
ource
.
Goo
gle
tr
anslat
e
us
es
sta
ti
sti
c
al
analy
ses
i
ns
te
ad
of
ru
le
based
analy
ses.
Sinc
e
ontol
ogy
is
hi
erarch
ic
al
ly
st
ru
ct
ur
e
d
te
rm
s,
sta
ti
sti
cal
m
ac
hine
translat
or
provi
des
bette
r
res
ult
than
the
rul
e
base
d
tra
ns
la
tor.
R
ule
bas
ed
m
achine
tr
anslat
ion
is
use
d
i
n
translat
ing
t
he
pa
ssage
gr
a
m
m
a
ti
c
al
l
y.
Finall
y,
the
tra
nsl
at
ed
te
rm
s
are
disp
la
ye
d
i
n
ML
G
rafViz
pa
nel
.
MLGra
fV
iz
fa
ci
li
ta
te
s
the
use
r
to
vis
ualiz
e
the
ontol
og
y
i
n
diff
e
re
nt
nat
ur
al
la
ngua
ges
with
ou
t
c
ha
nging
t
he
cor
e
ontol
og
y
structu
re as
d
e
picte
d
in
Fi
gur
e
4
(a),
(b).
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2722
-
3221
Com
pu
t. Sci.
I
nf. Tec
hnol.
,
V
ol.
1
, N
o.
2
,
J
ul
y
20
20
:
47
–
53
52
(a)
(b)
Figure
3.
G
raphical
r
e
pr
ese
nt
at
ion
,
(a) Ste
ps
involve
d
i
n program
m
ing
–
t
ext
,
(b) visuali
zat
ion
of steps
involve
d
in
pr
ogram
m
ing
–
di
agr
am
m
atic rep
rese
ntati
on
(a)
(b)
Figure
4.
MLG
rafViz
p
a
nel,
(
a)
Visu
al
iz
at
io
n
in
Tam
il
lang
uag
e
,
(b)
visu
a
li
zat
ion
in
Z
ul
u
la
ng
uag
e
4.
CONCL
US
I
O
N
We
hav
e
pr
opos
e
d
MO
nto
(
Mult
il
ing
ual
O
nto
lo
gy)
m
et
h
odology
to
de
velo
p
m
ulti
li
n
gu
al
ontolo
gy
app
li
cat
io
n
f
or
e
ducat
ion
dom
ai
n.
Ne
w
al
gorithm
s
are
pr
opos
e
d
to
perform
m
erg
ing
a
nd
m
app
i
ng
of
m
ul
ti
li
ng
ual
ontolog
ie
s
.
T
his
m
et
ho
d
al
lows
the
us
er
to
le
ar
n
t
he
sub
j
ect
f
r
om
their
own
na
tural
la
ngua
ge
w
hic
h
giv
es
bette
r
unde
rstan
ding
of
the
s
ubj
ect
.
This
re
searc
h
work
ide
ntifie
s
the
ne
ed
of
bu
il
di
ng
m
ultil
ingual
app
li
cat
io
n
w
hi
ch
play
s
vital
ro
le
i
n
ed
ucat
ion
al
do
m
ai
n.
If
t
he
le
ar
ning
m
a
te
rial
s
are
in
dif
fer
e
nt
na
tural
la
nguag
e
s,
t
he l
earn
er
will
f
ee
l com
fo
rtable
in
le
ar
ning.
Lea
rn
i
ng th
rou
gh the
natu
ral la
ng
uag
e
s is a
n ess
entia
l
thing
w
hich
e
ncou
rag
es
t
he
le
arn
er
t
o
le
arn
m
any
thing
s
.
I
n
f
uture
,
m
ult
il
ing
ual
app
li
cat
io
ns
can
be
i
m
ple
m
ented
f
or
di
ff
e
ren
t
do
m
ai
n
li
ke
healt
hcar
e
.
It
is
im
p
or
ta
nt
to
prov
i
de
the
eval
uatio
n
m
et
rics
an
d
m
et
ho
ds
to v
al
idate
m
ul
ti
li
ng
ual
on
t
ologies.
REFERE
NCE
S
[1]
As
unci
ón
Góm
ez
-
Pére
z
,
Mar
ia
n
o
Ferná
nde
z
-
Ló
pez
and
Os
ca
r
Corcho.
“
Ontolo
gic
a
l
Eng
ine
er
in
g
with
ex
ample
s
from
the
areas
of
Know
le
dg
e
Mana
gement,
e
-
Com
m
erc
e
and
the
Sem
ant
i
c
W
eb.
”
London
Be
rli
n
Heid
el
ber
g
,
Springer
,
2004
.
[2]
Nave
en
Kum
ar.
“
Ontolog
y
base
d
Books Inform
at
ion
Ret
r
ie
va
l
usi
ng
SP
ARQ
L.
”
Inte
rnational
Jou
rnal
of
Compute
r
Appl
ic
a
ti
ons
,
vo
l.
67
,
no
.
13
,
pp
.
24
-
27,
April
201
3
.
Evaluation Warning : The document was created with Spire.PDF for Python.
Com
pu
t. Sci.
I
nf. Tec
hnol.
Buildin
g a m
ulti
li
ng
ua
l
ontol
og
y
for
educ
ation d
omai
n usi
ng MOnt
o method
… (
Merl
in
Florre
nce
)
53
[3]
Kane
ll
opou
los,
Dim
it
ris,
Sotiri
s
B.
Kotsian
ti
s
an
d
Pana
y
iotis
E
.
P
int
elas.
“
Ontolog
y
-
bas
ed
L
ea
rn
in
g
Applicati
ons:
A
Deve
lopment
M
et
hodolog
y
.
”
IA
STED Conf.
on
Soft
ware
Engi
ne
ering
,
2006
.
[4]
W
ang,
Quan
y
u
,
Xingen
Yu
,
Gui
l
ong
L
i
and
Guob
in
Lv.
“
Ontolog
y
-
Based
Ec
o
logical
S
y
s
te
m
Mod
el
of
e
-
L
ea
rn
ing.”
Inte
rnational
Jo
urnal
of
In
formation
and
Educ
a
tion Tec
hnolog
y
,
pp.
595
-
599
,
20
12
.
[5]
Baj
en
aru
,
Li
di
a,
Ana
-
Maria
Boro
za
n
and
Ion
Sm
eur
ea
nu.
“
Us
ing
Ontologi
es
for
th
e
E
-
l
ea
rn
ing
S
y
s
te
m
in
Hea
lt
hc
ar
e
Hum
an
Resourc
es
Mana
gemen
t.”
Informati
că econom
ic
ă
,
vo
l. 19
,
pp
.
15
-
24
,
201
5
.
[6]
Nass
im
Khozooy
i
,
Na
fise
s
e
y
edi,
Ra
zie
Ma
le
kho
seini
.
“
Ontolog
y
-
base
d
e
-
learni
n
g
.”
IRA
CST
-
Int
ernati
onal
Journ
al
of
Computer
S
cienc
e
and
In
formation
Te
chn
olog
y
&
Sec
urity
(
IJCSI
TS)
,
Vol.
2,
No.
4,
Augus
t
2
012
.
[7]
Yus
of,
Noraz
ah
and
Andi
Bess
e
Fird
ausia
h
M
ansur.
“
Ontolog
y
Deve
lopment
of
e
-
Learni
ng
Moodle
for
Soc
ia
l
Le
arn
ing
Netwo
rk
Anal
y
s
is.”
W
orld
Ac
ademy
o
f
Scienc
e,
Engi
n
ee
ring
and
Te
ch
nology
,
In
te
rnat
ional
J
ournal
o
f
Computer,
E
lect
rical
,
Aut
omat
io
n,
Control
and
I
nformation
Eng
i
nee
ring
,
vol
.
7
,
pp.
853
-
858
,
20
13
.
[8]
Robin,
C.
R
.
Re
ne
and
As
sistant.
“
Design
and
Deve
lopment
of
S
RION
TO:
An
Educ
ational
Onto
l
og
y
Rep
rese
nt
in
g
Software
R
isk
I
dent
ifica
ti
on
Kn
owledge
.
”
2
nd
I
nte
rnational
Co
nfe
renc
e
and
w
orkshop
on
Eme
rging
Tr
ends
i
n
Technol
ogy
(
ICWET)
,
2011
.
[9]
Alhawit
i
,
Moha
m
m
ed
Mfari
j
F.
and
Yass
er
Abde
lha
m
id.
“
An
Ont
olog
y
-
B
ase
d
Fra
m
ework
for
Coll
ec
t
ing
E
-
Learni
n
g
Resourc
es.
”
Int
e
rnational
Journal
of
Learning,
Te
achi
ng
and
Edu
c
ati
onal Re
search
,
vol
.
9
,
2014
.
[10]
Yara
ndi,
Ma
r
y
a
m
,
Hos
sein
Jaha
nkhani
and
Abd
el
-
Rahman
H.
T
awil
.
“
A
per
sona
li
z
ed
ada
p
ti
ve
e
-
le
arn
ing
appr
oa
c
h
base
d
on
s
em
antic
web
t
ec
hnolo
g
y
.
”
W
ebol
og
y
,
vol.
10
,
2013
.
[11]
Ferná
ndez
-
López,
Mar
ia
no
,
As
un
Gom
ez
-
Pere
z
,
Jérôm
e
Euz
en
at,
Aldo
Gange
m
i,
Yannis
Kalf
oglo
u,
Dom
eni
co
M.
Pis
ane
ll
i
,
Ma
rco
Schorle
m
m
er,
Geri
Stev
e,
L
ji
l
j
ana
Sto
ja
novi
ć,
Gerd
Stum
m
e
a
nd
York
Sure
-
V
et
t
er.
“
A
surve
y
o
n
m
et
hodologi
es
f
or
developing
,
m
ai
nt
ai
ning
,
i
nte
g
rat
ing
,
eva
lu
at
in
g
and
r
ee
ng
ineer
ing
ontol
o
gi
es.
”
We
bology
,
2002
.
[12]
Fridm
an,
Nat
alya
and
Mark
A
.
M
usen.
“
SM
ART:
Autom
at
ed
Supp
ort
for
Onto
log
y
Merging
and
Ali
gnm
ent
.
”
(1999).
Proce
ed
ings o
f
t
he
12th
Worksho
p
on
Knowl
edge
Ac
quisit
ion, Mo
del
li
ng
,
and
Ma
nageme
nt
(
KAW'
99)
,
1999.
[13]
Jos
eph,
M.F.,
Lo
urdusam
y
,
R
.
“
Ontolog
y
m
edi
a
ti
o
n
m
et
hod
for
bui
ldi
ng
m
ult
ilingu
al
onto
logi
es.
”
I
nt.
j
.
in
f.
tecnol
.
Vol.
10
,
pp
.
11
–
19
,
2018
.
[14]
Ravi
Lourdus
a
m
y
,
Mer
li
n
Flo
rre
nce
Jos
eph
.
“
A
Survey
on
Multi
li
ngu
al
On
tol
ogie
s
.
”
In
te
rn
ati
onal
Journal
of
Mode
rn Comput
er
Scienc
e
(
IJM
CS)
,
pp
.
71
-
76
.
2015.
[15]
Troj
ahn
,
Cássia
& Q
ua
resm
a,
Pa
ulo & V
ieira
,
Re
nat
a
.
“
An
AP
I fo
r M
ult
i
-
li
ngu
al
Ontolog
y
M
at
ch
ing.
”
Proceedi
n
gs
of
7th
conference
on
Language
R
esource
s and Ev
aluat
ion
C
on
fe
re
nce
(
LREC)
.
pp.
3830
-
3835,
201
0.
[16]
Gana
pat
h
y
,
G
.
,
&
Lou
rdusam
y
,
R
.
“
Ontolog
y
Merging
and
Mat
chi
ng
Us
i
ng
Ontolog
y
A
bstrac
t
Ma
chi
ne
.
”
Proce
ed
ings o
f
Knowle
dge
Man
ageme
nt
5
th
In
ter
nati
onal
Conf
ere
nce
,
2010
.
[17]
Dos
Santos,
C.
T
.
,
Quare
sm
a,
P
.
,
&
Vie
ira,
R
. “
A
Fram
ework
fo
r M
ult
il
ingu
al Ont
olog
y
Mapping
.
”
LREC
,
2008
.
.
[18]
Schwall
,
J.
Cr
eat
ing
an
Ontolog
y
for
a
Multi
l
ingu
al
E
-
Com
m
erc
e Dict
ion
ar
y
.
Univ
ersity
of
Müns
ter:
s.n
.
,
2007.
[19]
Mejí
a
,
M
.
E.,
M
onti
el
-
Pons
oda
,
E.
,
d
e
C
ea,
G
.
A
.
,
&
Góm
ez
-
Pér
ez
,
A
.
“
Ontolog
y
lo
ca
l
iz
a
ti
on
.
”
Berl
in
Heid
el
be
r
g,
Springer,
Onto
lo
gy
E
ng
ine
ering
i
n
a
Ne
tworke
d
World
,
2012
.
[20]
De
Melo, G.
,
& S
ie
rsdorfer, S
. “
Multi
li
ngu
al T
ex
t
Cl
assific
a
ti
on
u
sing Ont
ologies.
”
Springer
,
2011
.
[21]
Dave
Be
ckett.
W
hat
does
SP
ARQ
L
stand
fo
r
?
SP
AR
QL.
[Onli
ne]
Octobe
r
6
,
2
011.
[Ci
te
d:
Jul
y
21,
2016.
]
Evaluation Warning : The document was created with Spire.PDF for Python.