Internati
o
nal
Journal of Ele
c
trical
and Computer
Engineering
(IJE
CE)
V
o
l.
5, N
o
. 2
,
A
p
r
il
201
5, p
p
.
34
4
~
35
4
I
S
SN
: 208
8-8
7
0
8
3
44
Jo
urn
a
l
h
o
me
pa
ge
: h
ttp
://iaesjo
u
r
na
l.com/
o
n
lin
e/ind
e
x.ph
p
/
IJECE
Towards a S
y
st
em of Gui
d
ance
, Assistance and Learning
Analytics Based on Multi Agen
t System Applied on Serious
Gam
e
s
Lotfi Elaac
hak,
Amine Belahbibe,
Mohammed B
o
uhor
ma
Com
puter s
c
ien
c
e, s
y
s
t
em
s
and
t
e
lecom
m
unicat
io
n Labor
ator
y
(Li
S
T)
Faculty
of Scien
ces an
d
Technologies, A
bdelmalek Essaad
i Univ
ersity
Tangier, Morocco
Article Info
A
B
STRAC
T
Article histo
r
y:
Received Ja
n
6, 2015
Rev
i
sed
Feb
12
, 20
15
Accepted
Feb 26, 2015
With the revolu
t
ion that
the
edu
cation field h
a
s
known concern
i
ng the new
tools and man
n
ers of learnin
g
and
especially
the integration of new
techno
log
y
as to
ols of teaching
,
seve
ral n
e
w tools have app
eared
and among
thes
e tools
ther
e
are s
e
rious
gam
e
s
,
s
e
rious games as new tool dedicated to
education have
occupied an important
place, and replaced oth
e
r tools often
used in
the
lear
ning process. B
u
t in
the order
that ser
i
ous games reach
the
intend
ed objectives and help
instruct
ors
to
achi
e
ve th
eir
pers
pect
ives
considered
,
it m
u
st be
that
this
k
i
nd of v
i
deo
ga
m
e
s will be
equ
i
pped with
a
guidanc
e and a
ssistance s
y
st
e
m
that will assist the le
arners
during the
progression of a sequence of the video
game, and in addition they
will be
equipped wi
th a
s
y
stem
of le
ar
ning ana
l
yti
c
s t
h
at will
help
in
structors to
improve the learning pro
cess and
te
aching
methods acco
rding to
th
e
perform
ances
an
d feedba
cks
of t
h
eir l
earne
rs
. In
this
pers
pec
tive
of res
ear
ch
and deve
lopm
ent we will est
a
bli
s
h in th
is paper
a new s
y
stem
of
assistance
,
guidanc
e and lea
r
ning anal
yt
ics based on a m
u
lti
agent s
y
st
em
that will work
in tandem with
a serious game.
Keyword:
Ed
ucat
i
onal
da
t
a
m
i
ni
ng
E-learning
Learni
ng analytics
Mu
lti ag
en
t sy
ste
m
R
u
l
e
base
d i
n
f
e
rence
en
gi
ne
Serious gam
e
s
Copyright ©
201
5 Institut
e
o
f
Ad
vanced
Engin
eer
ing and S
c
i
e
nce.
All rights re
se
rve
d
.
Co
rresp
ond
i
ng
Autho
r
:
Lo
tfi Elaach
a
k,
Com
puter scie
nce,
syste
m
s and te
lecomm
u
n
ication Laboratory
(LiST),
Faculty of Scie
nces a
n
d Tec
h
nol
ogies
,
Abde
lm
alek Essaadi
Unive
r
sity
Tangier, M
o
rocco.
Em
a
il: lo
tfi1
0
0
2
@
g
m
ail.co
m
1.
INTRODUCTION
Seri
o
u
s g
a
m
e
s or l
ear
ni
n
g
vi
deo
gam
e
s hav
e
fo
r m
i
ssi
on t
h
e t
r
an
sfe
r
o
f
t
h
e k
n
o
wl
e
dge
i
n
a fu
n an
d
i
n
t
e
ract
i
v
e wa
y
;
t
h
i
s
feat
ure
has m
a
de t
h
i
s
ki
n
d
of
vi
de
o gam
e
s one
of t
h
e m
o
st
powe
r
f
u
l
t
o
ol
s
use
d
f
o
r
teaching a
n
d learni
ng, and it is within the s
c
ope
of
usin
g the ne
w technologies in the e
ducational fiel
d. For
seriou
s g
a
m
e
s p
r
ov
e th
eir effectiv
en
ess; th
ey
m
u
st b
e
ab
le to
tran
smit
k
n
o
wledge p
r
op
erly an
d
allo
w
measuring a
n
d analyzing the
learning
outcomes. Unfortunately, it is a
challenge in any
vide
o
gam
e
to teach
a
pl
ay
er h
o
w
t
o
pl
ay
and t
o
g
u
i
d
e t
h
em
t
h
ro
ug
h t
h
e
gam
e
worl
d.
, especi
al
l
y
whe
n
deal
i
ng
wi
t
h
a seri
ous
gam
e
,
t
h
e chal
l
e
n
g
e i
n
t
e
nsi
f
i
e
s
due
t
o
t
h
e i
nhe
rent
vari
at
i
o
n
s
i
n
st
ude
nt
bac
k
g
r
o
u
n
d
s, m
a
ki
ng t
h
e ch
oi
ce of
h
o
w t
o
g
u
i
d
e
th
e st
u
d
en
t fro
m
th
e start to
th
e end o
f
t
h
e g
a
m
e
with
ou
t d
i
rect in
stru
ct
o
r
i
n
teractio
n
s
a com
p
lex
p
r
ob
lem
[1
]. In
ad
d
ition
,
t
h
e
d
i
fficu
lties in
measu
r
ing
learn
i
ng
o
u
t
co
m
e
s ach
iev
e
d
t
h
ro
ug
h seri
o
u
s
g
a
mes u
s
e
have
bee
n
a main
barrier for succes
sful depl
oym
e
nt
a
n
d adoption of s
u
ch
vi
de
o gam
e
s
within form
al
ed
u
cation
[2
]. Th
ere are sev
e
ral re
search works was
directed
b
y
d
i
ff
eren
t in
stitu
tio
n
s
and
labo
rato
ries,
conce
r
ni
n
g
t
h
e
i
n
cl
usi
o
n of a
ssi
st
ance an
d l
earni
ng a
n
al
y
t
ics sy
st
em
i
n
seri
o
u
s gam
e
s, am
ong t
h
e real
i
zat
i
on
th
at ex
ist there is a
v
i
rtu
a
l i
n
stru
ctor en
ab
led
m
o
b
ile augmen
ted
reality learn
i
n
g
“MARL” g
a
m
e
s h
a
v
e
the
pot
e
n
t
i
a
l
t
o
pr
ovi
de a fu
n an
d ed
ucat
i
onal
expe
ri
ence
. Th
i
s
ki
nd o
f
vi
de
o gam
e
s assi
st l
earni
n
g
conc
ept
u
al
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
To
ward
s a S
y
stem
o
f
Gu
idan
ce, Assistan
ce
an
d Lea
r
n
i
ng
An
a
l
ytics Ba
sed
o
n
Mu
lti Ag
en
t …
(Lo
tfi Ela
a
c
h
a
k
)
34
5
k
nowledg
e as well as
p
s
ych
o
m
o
t
o
r
task
in
real
world
en
v
i
ron
m
en
ts [3
].
In th
e literatu
re th
ere are also
m
e
t
hods
an
d st
udi
es
o
n
t
h
e i
n
t
e
ract
i
o
n
o
f
se
ri
ous
g
a
m
e
wi
t
h
l
earni
n
g
a
n
al
y
t
i
c
s sy
st
em
[4-
7
]
.
W
i
t
h
t
h
e aim
to
ov
erco
m
e
th
e
d
i
fficu
lties
cited
abov
e, t
h
erefore, we
will d
e
tail in
t
h
is
p
a
p
e
r t
h
e
diffe
re
nt steps of the esta
blishm
ent of a syste
m
that w
ill be able t
o
react
with th
e
learners by giving
them
in
fo
rm
atio
n
o
r
assistan
ce d
u
rin
g
th
eir
p
r
o
g
ressio
n
in
th
e
vid
e
o
g
a
m
e
, th
is relatio
n
s
h
i
p
will h
e
lp
learners to
i
m
p
r
ov
e th
ei
r learn
i
n
g
lev
e
l an
d also their p
e
rfo
r
m
a
n
ce. On
t
h
e
o
t
h
e
r h
a
nd
,
we will d
e
tail also
th
e
estab
lish
m
en
t
o
f
th
e system
capabl
e
o
f
savi
ng
dat
a
abo
u
t
l
earne
rs i
n
o
r
de
r t
o
anal
y
ze t
h
em
by
usi
ng s
e
vera
l
algorithm
s
, specific
m
e
thods a
nd techniques
, the results of suc
h
analy
zes will help the inst
ructors to im
prove
bot
h their strat
e
gy and their
teaching m
e
thods,
based
on the feedback
of
t
h
eir learners
, the inclusi
on
of suc
h
syste
m
will
mak
e
seriou
s g
a
mes a co
m
p
lete too
l
su
itab
l
e
fo
r th
e
pro
c
ess of learn
i
n
g
.
In
t
h
is persp
ect
iv
e
o
f
researc
h
a
n
d
d
e
vel
o
pm
ent
we
ai
m
i
n
t
h
i
s
pa
per
t
o
de
vel
o
p
a seri
ous
gam
e
s ab
o
u
t
a
wast
e
so
rt
i
n
g
pr
oces
s, t
h
i
s
vide
o
gam
e
will be equippe
d
with a m
u
lti agent system
com
posed of se
veral in
telligent
age
n
ts,
whic
h each
agent in the sy
ste
m
has a specific task
, there is an agent that assists pl
ayers accordi
ng t
o
their pe
rformances
du
ri
n
g
a
seq
u
e
nce
of
vi
deo
gam
e
s, t
h
ere
i
s
al
so a
n
a
g
e
n
t that e
x
tract
s and a
n
alyses
data c
o
ncerning the
learn
e
rs in order to b
e
in
terp
reted
b
y
t
h
e i
n
stru
ct
o
r
s.
2.
THEORETICAL BACKGROUND
The establishment of suc
h
syste
m
require
s a speci
fic architecture with the interaction of se
ve
ra
l
tech
no
log
i
es; th
erefore, we hav
e
u
s
ed
the co
m
b
in
ati
on
of
t
h
e
rul
e
-ba
s
e
d
i
n
fe
rence
en
gi
ne,
t
h
e
ed
uc
at
i
ona
l
d
a
ta min
i
n
g
too
l
with
th
e learn
i
n
g
an
alytics tech
n
i
qu
es
, al
l o
f
tho
s
e techn
o
l
o
g
i
es
will b
e
h
o
s
ted
in
d
i
fferen
t
ag
en
ts of t
h
e
m
u
l
ti ag
en
t syste
m
, in
th
is
sectio
n
w
e
w
ill d
e
tail th
e th
eoretical b
a
ck
ground
of t
h
e m
u
lti ag
en
t
syste
m
architecture a
n
d the
technologies
use
d
in
t
h
e e
s
t
a
bl
i
s
hm
ent
of t
h
e
pr
o
pose
d
sy
st
em
.
2.1.
Rule
-Base
d
Inference E
n
gine
The m
a
i
n
rol
e
of t
h
e i
n
fere
nc
e engi
ne i
s
de
ri
ved
new fact
s f
r
om
t
h
e kno
wl
edge
base as sho
w
n i
n
t
h
e
Fi
gu
re 1
,
ot
he
rwi
s
e t
h
e i
n
fe
rence e
ngi
ne appl
i
e
d l
o
gi
cal
rul
e
s t
o
t
h
e kn
o
w
l
e
d
g
e ba
se, t
h
en i
n
fer
r
e
d ne
w
k
nowledg
e, t
h
ere are
d
i
ffer
en
t typ
e
s
o
f
i
n
feren
ce eng
i
n
e
s, bu
t in
th
is
pap
e
r
we
will fo
cus on
ru
le-based
i
n
fere
nce e
ngi
ne w
ho a
p
pl
i
e
s several
r
u
l
e
s
wi
t
h
dat
a
t
o
de
ri
ve ne
w fact
s
.
Thi
s
ki
n
d
o
f
t
h
e i
n
fe
re
nce en
gi
ne i
s
al
way
s
com
posed o
f
t
h
ree co
m
ponent
s, t
h
e
i
n
t
e
rp
ret
e
r t
h
at
execut
e
s t
h
e chos
en age
n
da el
em
ent
s
by
appl
y
i
n
g
th
e co
rr
esp
ondin
g
b
a
se ru
les, th
e sch
e
d
u
l
er th
at
m
a
in
tain
s th
e con
t
r
o
l
o
v
e
r
th
e ag
end
a
b
y
estim
a
t
i
n
g
t
h
e
effects of applying infe
renc
e rules in crit
eria on t
h
e
a
g
enda, t
h
e consistency enforcer that attem
p
ts
to
main
tain
a consisten
t
rep
r
esen
ta
tio
n
of th
e emerg
i
ng
so
lu
ti
o
n
.
Fi
gu
re
1.
I
n
fe
r
e
nce e
ngi
ne a
r
chi
t
ect
ure
Gene
rally,
in every rule
-bas
ed
infere
nce engi
nes,
t
h
ere
are t
w
o
ki
nd
s o
f
i
n
fere
nce
,
bac
k
w
a
r
d
chaining and forward chai
ning. The fo
rward ch
ain
i
ng
, acco
r
d
i
ng
to
th
is alg
o
rith
m
th
e i
n
feren
ce is in
feren
ce
i
s
t
r
i
ggere
d
b
y
t
h
e arri
val
o
f
ne
w dat
a
i
n
wo
rki
ng
m
e
m
o
ry;
it’s als
o
called
d
a
ta-d
irected
i
n
feren
ce. Th
e
backwa
rd chai
ning
whe
r
e i
n
ferences
are
not
pe
rform
e
d by
th
e system
is mad
e
to
p
r
ov
e
a p
a
rticu
l
ar
goal; it’s
also
called
h
ypo
th
esis driv
en
,
or goal
directe
d
infere
nce.
2.
2. E
d
uc
ati
o
n
a
l
D
a
ta
Mi
ni
n
g
The Dat
a
M
i
ni
ng i
s
t
h
e
pr
oce
ss of a
n
al
y
z
i
n
g
dat
a
fr
om
di
fferent
pe
rs
pect
i
v
es an
d s
u
m
m
ari
z
i
ng t
h
e
resu
lts as v
a
luab
le in
fo
rm
ati
o
n. It h
a
s b
e
en d
e
fin
e
d
as th
e n
o
n
t
riv
i
al process o
f
id
en
tifyin
g
v
a
lid
,
po
ten
tially
usef
ul
, n
o
v
el
and
ul
t
i
m
a
t
e
l
y
u
nde
rst
a
n
d
a
b
l
e
pat
t
e
rns i
n
dat
a
[8, 9]
. Am
ong t
h
e m
o
st
fam
ous branc
h
es
of dat
a
mining there
are the educational data m
i
n
i
ng “E
DM” th
at
descri
bes a
researc
h
fi
el
d
conce
r
ne
d wi
t
h
t
h
e
appl
i
cat
i
o
n
of
dat
a
m
i
ni
ng,
m
achi
n
e l
ear
ni
ng
al
g
o
ri
t
h
m
s
and
st
at
i
s
t
i
c
s t
ool
s t
o
i
n
f
o
rm
at
i
on
gene
rat
e
d f
r
o
m
educat
i
o
nal
are
a
. Ot
her de
fi
n
i
t
i
on of ed
ucat
i
onal
dat
a
m
i
ni
ng as t
o
ol
of
M
i
ni
ng i
n
ed
u
cat
i
onal
envi
ro
nm
ent
,
conce
r
n
wi
t
h
devel
opi
ng
n
e
w m
e
t
hods
t
o
di
sc
ove
r
k
n
o
w
l
e
d
g
e fr
om
educat
i
o
nal
da
t
a
bases [1
0]
. I
t
‘s
an
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
IJEC
E V
o
l
.
5, No
. 2, A
p
ri
l
20
15
:
34
4 – 3
5
4
34
6
em
ergi
n
g
di
sci
p
l
i
n
e, co
ncer
n
e
d wi
t
h
de
vel
o
pi
n
g
m
e
t
hods
fo
r
e
xpl
ori
n
g
t
h
e uni
qu
e
t
y
pe
s
o
f
dat
a
t
h
at
com
e
fr
om
educat
i
o
n
a
l
param
e
t
e
rs, and
usi
ng t
hos
e
m
e
t
hods
a
n
d
t
echni
q
u
es t
o
bet
t
e
r u
n
d
erst
a
nd st
u
d
ent
s
, a
n
d t
h
e
settin
g
s
wh
ich th
ey learn
in
[11
]
. Th
ere are sev
e
ral
d
a
ta
m
i
ni
ng t
ech
ni
q
u
es u
s
ed t
o
extract useful dat
a
that
hel
p
use
r
s t
o
p
r
edi
c
t
un
k
n
o
w
n
or
fut
u
re
val
u
es
of t
h
e at
t
r
i
but
es
, an
d al
s
o
t
o
descri
be t
h
e dat
a
i
n
a m
a
nne
r
un
de
rst
a
n
d
abl
e
and i
n
t
e
r
p
ret
a
bl
e t
o
use
r
s.
Am
ong t
h
os
e t
echni
q
u
es t
h
ere i
s
C
l
assi
fi
cat
i
on, C
l
ust
e
ri
n
g
,
Ass
o
ci
at
i
on
r
u
l
e
, R
e
gre
ssi
o
n
,
an
d
ot
he
rs m
achi
n
e l
e
a
r
ni
ng
al
go
ri
t
h
m
s
l
i
k
e ne
ural
net
w
o
r
ks,
B
a
y
e
s net
w
o
r
k
,
etc.
2.
3. L
e
arni
n
g
An
al
yti
c
s
The learning a
n
alytics research trays t
o
a
n
s
w
er
i
n
crea
si
n
g
l
y
several
que
s
t
i
ons a
b
out
w
h
at
a l
earne
r
kn
o
w
s a
nd
w
h
et
her a l
ear
ner
i
s
enga
ge
d. T
h
e appl
i
cat
i
o
n fi
el
ds o
f
t
h
e l
ear
ni
n
g
anal
y
t
i
c
s conce
r
n m
odel
i
ng
o
f
user
k
n
o
w
l
e
d
g
e
, use
r
be
ha
vi
or
, an
d u
s
er e
xpe
ri
ence
, use
r
p
r
o
f
i
l
i
ng;
m
odel
i
ng
o
f
key
conce
p
t
s
i
n
a
dom
ai
n
and m
odeling
a dom
a
in’s knowle
dge
com
pone
nts, and tr
e
n
d analysis. Be
low
we
will de
tail those c
onc
epts.
Th
e
u
s
er
k
nowledg
e m
o
d
e
lin
g is th
e Co
ll
ectio
n
o
f
u
s
er’s sk
ills and
k
nowledg
e ex
t
r
acted
fro
m
di
ffe
re
nt
dat
a
sou
r
ces
, l
o
o
k
i
ng at
t
h
e regi
s
t
ered dat
a
th
at
rep
r
esen
ts th
e in
teractio
n
b
e
tween
u
s
er and
bo
th
learning system
or serious
gam
e
s.
A
m
ong the inform
ation e
x
tracted
th
ere are: c
o
rrectness of a s
t
ude
nt
resp
o
n
se al
o
n
e
or i
n
a seri
es, t
i
m
e
spent
on
pract
i
ce bef
o
re at
t
e
m
p
ti
ng t
o
a
n
swe
r
quest
i
o
n
or t
o
d
o
man
i
p
u
l
ation
s
, n
u
m
b
e
r an
d natu
re
o
f
h
i
n
t
s
requ
ested,
repartitio
n
of
wron
g
an
swers and
errors m
a
d
e
. Su
ch
infere
nces ca
n
be m
odel by a pre
d
ictive m
odel or by t
eache
r
looking at student
da
ta on dashboa
rd. A
popul
a
r
m
e
t
hod
fo
r est
i
m
a
ti
ng st
ude
n
t
s’ k
n
o
wl
e
dge
i
s
C
o
r
b
et
t
an
d
An
der
s
o
n
’
s
[1
2]
k
n
o
wl
e
d
g
e
t
r
aci
n
g
m
odel
,
a
n
ap
pro
ach
th
at
u
s
es a Bayesian
-n
etwo
rk-b
ased
m
o
d
e
l fo
r
esti
m
a
t
i
n
g
th
e p
r
o
b
a
b
ility th
at a stu
d
e
n
t
kn
ows a
sk
ill b
a
sed
o
n
o
b
s
erv
a
tio
ns
o
f
h
i
m
o
r
h
e
r attem
p
t
i
n
g
to p
e
rfo
r
m
th
e sk
ill.
Mo
re
recen
tly, a n
e
w stud
y
[13
]
h
a
s
pr
o
pose
d
a
ne
w m
e
t
hod
fo
r
k
n
o
w
l
e
d
g
e t
r
aci
ng
usi
n
g a
m
achi
n
e l
earni
n
g
a
p
p
r
oa
ch
t
o
m
a
ke cont
ext
u
a
l
esti
m
a
t
i
o
n
s
of th
e p
r
o
b
a
b
ility th
at a stu
d
e
n
t
h
a
s g
u
e
ssed
or slip
p
e
d
.
In
corpo
r
ating
m
o
d
e
ls o
f
g
u
e
ssing
and
slipping into
predictions of stude
nt
future pe
rform
a
nce was
shown to in
c
r
ease the accura
cy of the pre
d
ictions
by
u
p
t
o
4
8
pe
rcent
.
The
Use
r
be
ha
vi
o
r
m
odel
i
n
g
i
n
e
d
u
c
at
i
on
oft
e
n
cha
r
act
eri
zes st
u
d
e
nt
act
i
ons
as
on
-
or
of
f-t
as
k an
d ca
n be
use
d
as a
pr
o
x
y
fo
r st
u
d
e
nt
en
gagem
e
nt
. It
rel
i
e
s on t
h
e sam
e
ki
nds
of l
ear
ni
n
g
dat
a
use
d
i
n
p
r
e
d
i
c
t
i
n
g
user
k
n
o
wl
e
d
g
e
pl
u
s
ot
he
r
m
easures,
suc
h
as
h
o
w
m
u
ch t
i
m
e a st
udent
has
s
p
ent
o
n
l
i
n
e
,
whet
her a student
has
com
p
leted a c
o
urse, doc
u
m
e
nted
c
h
an
ges
i
n
t
h
e cl
ass
r
o
o
m
or sch
ool
cont
e
x
t
,
at
t
e
ndan
ce, t
a
r
d
i
n
ess
,
an
d s
o
m
e
t
i
m
e
s a st
udent
’s l
e
vel
of
kn
owledg
e as inferred
fro
m
h
i
s o
r
h
e
r
work
with
th
e
l
earni
n
g
sy
st
em
or fr
om
other s
u
ch
dat
a
sou
r
ces as st
a
nda
r
d
i
zed t
e
st
scores
. B
a
ker
and c
o
l
l
eagu
e
s ha
v
e
con
d
u
ct
ed a s
e
ri
es of st
udi
e
s
on
det
ect
i
n
g
and a
d
apt
i
ng
t
o
st
ude
nt
s’
of
f-t
as
k be
havi
o
r
s cal
l
e
d gam
i
ng t
h
e
syste
m
in adaptive learni
ng syste
m
s
that teach algebra
[14].T
he Use
r
expe
rience modelin
g asce
rtaining
whet
her a
st
u
d
e
nt
i
s
sat
i
s
fi
ed
wi
t
h
t
h
e l
ear
ni
ng e
x
peri
e
n
ce
can be
j
u
dge
d
by
st
u
d
ent
s
’ re
spo
n
ses
t
o
f
o
l
l
ow
-u
p
sur
v
ey
s
or
q
u
e
s
t
i
o
n
n
ai
res a
n
d
by
t
h
ei
r c
h
oi
ces, be
ha
vi
o
r
s,
per
f
o
r
m
a
nce, and
ret
e
nt
i
o
n i
n
su
bseq
ue
nt
l
earni
n
g
uni
t
s
or
co
u
r
se
s. U
s
er e
x
peri
e
n
ce t
h
r
o
u
g
h
m
e
t
h
o
d
s
ot
he
r t
h
an
dat
a
m
i
ni
ng
, col
l
ect
ed
t
i
m
e
spe
n
t
on
re
de
si
gne
d
co
ur
se co
m
p
on
en
ts,
p
e
r
i
od
i
c
sur
v
eys of stud
en
ts
’ mo
tiv
atio
n state du
ri
n
g
the
course, and l
earni
ng
perform
a
nce. The use
r
profile is a co
l
l
ecti
on of
pers
o
n
al
d
a
t
a
descri
bi
n
g
the essential charact
eristics of
a user.
User
p
r
o
f
i
l
i
ng
refe
rs t
o
t
h
e
pr
ocess
of
co
nst
r
uct
i
n
g
an
d a
p
p
l
y
i
ng st
u
d
ent
o
r
g
r
ou
p
pr
ofi
l
e
s usi
ng
dat
a
m
i
ni
n
g
and m
achine learni
ng al
gorithm
s
. In ed
ucat
i
o
nal
dat
a
m
i
ni
ng t
ech
ni
q
u
es
, s
u
ch
as cl
assi
fi
c
a
t
i
on a
n
d
cl
ust
e
ri
n
g
,
are
oft
e
n
u
s
ed
t
o
cat
eg
o
r
i
ze l
earne
rs
base
d
o
n
t
h
e
ki
nd
s
of
pers
o
n
al
l
ear
ni
ng
dat
a
.
2.
4.
Mul
t
i
A
g
e
n
t S
y
s
t
em
A m
u
l
t
i
agent
sy
st
em
“M
AS” i
s
a co
m
put
i
ng
di
st
ri
b
u
t
e
d sy
st
em
, com
posed of a num
ber o
f
in
teractin
g com
p
u
t
atio
n
a
l entities, b
u
t
t
h
e sin
g
l
e
d
i
ffer
en
ce b
e
tween classical d
i
stribu
ted
system
s an
d m
u
lti
agent system i
s
that the en
tit
ies that interact in the
syste
m
,
are intellig
ent. Thes
e e
n
tities that react
in the
syste
m
are cal
l
e
d age
n
ts, and
m
u
st be able to comm
uni
cate
each ot
her. The conce
p
t of m
u
lti agent system has
in
flu
e
n
c
ed
th
e
in
itial d
e
v
e
lopmen
ts in
areas lik
e co
gn
itiv
e m
o
d
e
lin
g
and in
stru
ctio
nal
d
e
sign
[15
,
16]. Un
ti
l
n
o
wad
a
ys, th
e
m
u
lti ag
en
t syste
m
s estab
l
i
s
h
a m
a
j
o
r res
earch
sub
j
ect i
n
d
i
stri
b
u
t
ed
artificial in
telli
g
e
n
c
e.
M
A
S t
ech
nol
o
g
y
aim
s
t
o
creat
e gene
ral
an
d
speci
al
i
zed be
havi
oral
a
nd i
n
t
e
ract
i
on m
ode
l
s
and t
o
i
m
pl
em
ent
t
h
ese m
odel
s
int
o
di
st
ri
b
u
t
e
d
and i
n
t
e
ract
i
n
g com
put
er p
r
og
ram
s
cal
l
e
d
agent
s
. The
de
si
gn
of s
u
ch
m
odel
s
follows ce
rtain gui
delines t
h
a
t
characterize a
g
ents
[17].
T
h
e MAS
have been
used in m
a
ny fields t
o
si
m
u
la
te
com
p
l
e
x sy
st
em
s, such as
de
ci
si
on s
u
pp
o
r
t
t
ool
s
fo
r
di
st
ri
but
e
d
deci
si
o
n
pr
o
b
l
e
m
s
.
3.
THE SERIOUS
GAME
Wast
e s
o
rt
i
n
g
has a
n
d s
h
oul
d
bec
o
m
e
part
o
f
ou
r
dai
l
y
l
i
f
e
t
o
i
m
prove
o
u
r
l
i
v
i
n
g e
nvi
r
o
n
m
ent
.
W
i
t
h
t
h
e i
m
port
a
nce
of t
h
e wa
st
e s
o
rt
i
n
g a
nd t
h
e
bene
fi
t
s
t
h
at
it p
r
esen
ts, t
o
th
i
s
effect m
a
n
y
in
stru
ctio
n
a
l exp
e
rts
have
found t
h
a
t
teaching the
basics of wast
e sorting fo
r
kids dice their
young a
g
e
ca
n be be
neficial for the
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
To
ward
s a S
y
stem
o
f
Gu
idan
ce, Assistan
ce
an
d Lea
r
n
i
ng
An
a
l
ytics Ba
sed
o
n
Mu
lti Ag
en
t …
(Lo
tfi Ela
a
c
h
a
k
)
34
7
envi
ronm
ent and the ec
onomy. For t
h
is
reason
th
ere is no
m
o
re rob
u
st way to
learn
th
e b
a
sics of waste
so
rting
b
e
tter th
an
t
h
e v
i
d
e
o
g
a
m
e
s, v
i
ew to
th
eir ad
v
a
n
t
ag
es su
ch
in
teractiv
ity an
d
p
l
ayab
ility
th
at at
tracts
t
h
e i
n
t
e
nt
i
o
n and t
h
e
desi
re i
n
t
o
t
h
e
pl
ay
ers
t
o
pl
ay
m
o
re.
In t
h
i
s
pe
rs
pe
ct
i
v
e of de
vel
opm
ent
ou
r re
search
tea
m
h
a
s d
e
v
e
l
o
p
e
d
an
in
teractiv
e web
b
a
sed
seri
o
u
s
g
a
m
e
for waste so
rti
n
g
d
e
d
i
cated
fo
r ch
ild
ren
th
at will
b
e
d
e
scri
b
e
d in th
is section
.
3.
1.
T
h
e Con
c
ept of
W
a
st
e
Sorting
Serious
Game
The m
a
i
n
ob
je
ct
i
v
e of
t
h
e
pr
op
ose
d
se
ri
o
u
s
gam
e
Fi
gure
2 i
s
t
o
t
eac
h
ki
ds a
b
o
u
t
re
cy
cl
i
ng
di
f
f
ere
n
t
wast
e. T
h
e
pl
ay
er sh
oul
d s
o
r
t
di
ffe
rent
was
t
e i
n
t
o
t
r
as
h,
p
a
per
,
pl
ast
i
c
,
m
e
t
a
l
,
gl
ass, and
o
r
ga
ni
c, et
c
.
Th
e
sort
i
n
g i
s
d
o
n
e
by
cat
chi
n
g
d
i
ffere
nt
ob
ject
s
ge
nerat
e
d
ran
dom
l
y
and
dr
o
ppi
ng
i
n
t
h
e a
p
p
r
op
ri
at
e co
n
t
ai
ner
according to t
h
eir types, this
m
echanism
w
ill be done
by
using a tool c
a
lled a l
eap
m
o
tion c
o
ntrolle
r. The
waste sorting
seriou
s
g
a
m
e
will b
e
equ
i
pped
b
y
t
h
e ti
mer, an
d
t
h
e assessm
en
t syste
m
th
at ev
alu
a
tes th
e
p
l
ayers acco
r
din
g
to th
eir
p
e
rfo
r
m
a
n
ces if
th
ey
m
a
k
e
a go
od
ch
o
i
ce t
h
e reward
will be th
e g
a
i
n
of
so
me
p
o
i
n
t
s, altho
ugh
if th
e
o
ppo
site case th
e p
u
n
i
sh
m
e
n
t
w
ill
b
e
th
e lo
ss of
so
m
e
p
o
i
n
t
s.
W
i
t
h
th
e assessm
e
n
t
sy
st
em
, t
h
e t
i
m
e
r, an
d t
h
e i
n
t
e
ract
i
v
i
t
y
ba
sed
on
ha
nd
m
ovem
e
nt
t
h
e pr
o
pose
d
se
ri
ous
gam
e
wi
l
l
be m
o
re
ch
allen
g
i
ng
and
attractiv
e esp
ecially for
k
i
d
s
, it
will
allo
w th
em
to
liv
e a
b
e
n
e
ficial an
d unforg
e
t
t
ab
le
expe
ri
ence
. Th
e pr
op
ose
d
ser
i
ous
gam
e
has
been
devel
o
pe
d by
Java
Scri
p
t
API, t
h
e
r
ef
or
e, i
t
need j
u
st
a we
b
br
ow
ser
t
o
be r
u
n
.
Fi
gu
re
2.
Scree
n
s
h
o
o
t
s
fr
om
wast
e s
o
rt
i
n
g
v
i
deo
gam
e
.
3.
2. I
n
ter
a
c
t
iv
ity w
i
th
Lea
p
Mo
tio
n
Co
ntr
o
ller
The Leap Moti
on c
ontroller is a s
m
all device that can
be connected to a
com
puter using a USB. It
uses infra
red (IR) im
aging to determ
ine the
position of
pr
edefi
n
ed
obj
e
c
t
s in a li
mi
ted space in real time. It
can t
h
en se
nse
han
d
an
d fi
n
g
er m
ovem
e
nts i
n
t
h
e ai
r above i
t
,
an
d t
h
e
s
e
m
ovem
e
nt
s are reco
gni
ze
d a
n
d
translated int
o
actions for the com
p
u
t
er to
p
e
rform
.
Acco
rd
ing
to
th
e
o
f
ficial in
formatio
n
fo
und
ed
in
th
e
of
fi
ci
al
web si
t
e
of l
eap m
o
t
i
on [
18]
, t
h
e
Lea
p
so
ft
wa
re ana
l
y
zes t
h
e ob
jec
t
s obse
r
ve
d i
n
t
h
e de
vi
ce’s fi
el
d o
f
vi
ew. It
rec
o
gni
zes h
a
n
d
s,
fi
nge
rs, an
d
t
ool
s, re
po
rt
i
ng
di
scret
e
p
o
si
t
i
ons
, gest
u
r
es, an
d m
o
t
i
on
. Th
e
cont
rol
l
e
r
’
s fi
e
l
d o
f
vi
ew i
s
a
n
i
n
vert
e
d
py
r
a
m
i
d cent
e
re
d
on
t
h
e
devi
ce
Fi
gu
re
3. T
h
e
effect
i
v
e
ra
nge
of
t
h
e
co
n
t
ro
ller ex
ten
d
s
fro
m
ap
pro
x
i
m
a
tely 2
5
to
600
m
i
l
l
i
m
eters abov
e the d
e
v
i
ce. Th
e con
t
ro
ller itself is
accessed and programm
ed through App
lication Programming Interfaces (APIs)
,
with support for a va
riety of
p
r
og
rammin
g
lan
g
u
a
g
e
s,
rangin
g
fro
m
C++
t
o
Pytho
n
an
d
Jav
a
Scrip
t
. Th
e p
o
s
ition
s
o
f
the recog
n
i
zed
ob
j
e
ct
s
are ac
qui
re
d t
h
ro
u
gh t
h
ese
A
P
Is.
The
C
a
rt
es
i
a
n an
d s
p
heri
c
a
l
coo
r
di
nat
e
s
y
st
em
s used t
o
descri
be
po
si
t
i
ons
i
n
the controller’s
sens
ory
space
.
Fi
gu
re 3.
Leap
m
o
ti
on'
s
fi
el
d of
vi
e
w
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
IJEC
E V
o
l
.
5, No
. 2, A
p
ri
l
20
15
:
34
4 – 3
5
4
34
8
W
i
t
h
the feat
ures that leap
m
o
tion controller o
ffe
rs and
wi
t
h
t
h
e
use Java
Scri
pt
appl
i
cat
i
o
n
pr
o
g
ram
m
i
ng i
n
t
e
rface
, we h
a
ve i
n
t
e
g
r
at
ed
i
t
i
n
t
h
e pro
p
o
s
ed vi
deo
gam
e
, t
h
eref
o
r
e, t
h
e pl
ay
er has t
o
m
ove
his
hand
and c
a
tch the
ra
ndom
generated objects
in orde
r
to
dra
g
a
n
d
drop t
h
em
, then
place them
in a
correc
t
co
n
t
ain
e
r. W
i
t
h
th
is po
ssib
ility
th
e
propo
sed
v
i
d
e
o
g
a
m
e
will b
eco
m
e
m
o
re in
teractiv
e and
so
close to
t
h
e real
case. Th
at
will create en
v
y
i
n
to
t
h
e
p
l
ayer
to
p
l
ay m
o
re
.
In add
itio
n, th
i
s
con
c
ep
t will
allo
w
u
s
t
o
sav
e
all
g
e
stures d
o
n
e
b
y
th
e p
l
ayers
du
ring
a sequen
ce o
f
v
i
d
e
o
g
a
m
e
; th
is d
a
ta will b
e
u
s
ed b
y
edu
cation
a
l d
a
ta
m
a
im
i
ng t
o
u
n
d
erst
a
n
d
t
h
e
pl
ay
er’s
be
havi
o
r
s a
n
d also to a
n
alyze their
pe
rform
a
nces.
4.
THE ESTAB
L
ISHME
N
T
OF THE
SYSTEM
Th
e syste
m
o
f
g
u
i
d
a
n
c
e, assistan
ce an
d
learnin
g
an
alytics p
r
opo
sed
in
th
is p
a
p
e
r
will b
e
estab
lish
e
d
v
i
a th
e u
s
e of
th
e m
u
lti ag
ent syste
m
; th
e ch
o
s
en system
i
s
co
m
p
o
s
ed
o
f
sev
e
ral en
titie
s called
in
tellig
ent
agents
,
whe
r
e
each a
g
e
n
t of the
system
is equippe
d
with a
rules e
ngine
,
data m
i
ning algorithm
s
or the
m
e
t
hods
o
f
t
h
e l
earni
n
g
a
n
a
l
y
t
i
c
s. As sh
o
w
i
n
t
h
e Fi
g
u
r
e 4 t
h
ere
are
di
f
f
ere
n
t
t
ech
nol
ogi
es
use
d
i
n
t
h
e
architecture of the propose
d
syste
m
,
in the first place we
have a we
b a
pplication de
ployed in a
pplic
ation
serve
r
, t
h
i
s
we
b ap
pl
i
cat
i
on i
s
com
posed o
f
a web base
d seri
o
u
s gam
e
that
i
n
t
e
ract
wi
t
h
t
h
e serve
r
si
de by
u
s
ing
Aj
ax
tech
no
log
y
, th
e use o
f
th
e in
teractiv
e in
te
rface allo
ws real t
i
m
e
in
teractio
n b
e
tween
th
e actio
n
s
and
t
h
e
gest
u
r
e
s
o
f
t
h
e l
e
a
r
ne
r
s
an
d t
h
e
ap
pe
arance
o
f
t
h
e
a
ssi
st
ance m
e
ssages t
h
at
t
h
e s
y
st
em
shoul
d
p
r
o
v
i
d
e
to
th
e learn
e
rs
to
gu
id
e th
em
in
o
r
d
e
r, to im
p
r
ov
e t
h
eir
p
e
rfo
rm
an
ces.
Fi
gu
re
4.
A
r
chi
t
ect
ure
of t
h
e s
y
st
em
based
o
n
M
A
S
fo
r
gui
dance
an
d
Eval
uat
i
o
n
C
o
nce
r
n
i
ng
Seri
ous
Ga
m
e
s
Players.
R
e
gar
d
i
n
g t
h
e
m
u
lt
i
agent
sy
s
t
em
part
, Ja
de
[1
9]
has
al
l
o
w
e
d
us a
n
easy
i
m
pl
em
ent
a
t
i
on of
t
h
e m
u
l
t
i
ag
en
t system “
M
AS”, as presen
ted
in
th
e sch
e
m
a
o
u
r
MAS is co
m
p
o
s
ed o
f
to
w in
tellig
en
t ag
en
ts; the first
one is the
gui
dance a
n
d assistance age
n
t; by against th
e
second a
g
ent i
s
for learni
ng
analytics. Each age
n
t
reacts in its
own way a
n
d im
ple
m
ents
its
own al
gorithm
s
to ens
u
re
th
e prope
r functioning of
t
h
e
whol
e
sy
st
em
. As m
e
nt
i
one
d
bef
o
re
t
h
e fi
rst
a
g
e
n
t
i
s
eq
ui
p
p
ed
wi
t
h
a r
u
l
e
s e
n
gi
n
e
, i
n
or
der
t
o
b
u
i
l
d
s
u
ch
en
gi
ne w
e
have
ch
ose
n
dr
ool
s
[
20]
a
B
u
s
i
ness R
u
l
e
s M
a
nagem
e
nt
Sy
st
em
“B
R
M
S” sol
u
t
i
o
n.
It
pr
ov
i
d
es a c
o
re
B
u
s
i
nes
s
R
u
l
e
s E
ngi
ne”
B
R
E
”.
W
e
ha
v
e
est
a
bl
i
s
he
d a
r
u
l
e
t
r
ee
Fi
g
u
r
e
5 that c
o
ntains all the
poss
ible cases
of fi
gures
that will guide
and assist learners
duri
ng the
i
r video
gam
e
sequence
an
d
according t
o
their pe
rform
a
nces.
In
th
e ru
les tree
th
ere are sev
e
ral ele
m
en
ts , in
th
e lev
e
l
one that re
pres
ents the ti
m
e
of t
h
e se
ri
o
u
s
gam
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
To
ward
s a S
y
stem
o
f
Gu
idan
ce, Assistan
ce
an
d Lea
r
n
i
ng
An
a
l
ytics Ba
sed
o
n
Mu
lti Ag
en
t …
(Lo
tfi Ela
a
c
h
a
k
)
34
9
sequence
, as presented in the t
r
ee there
are t
h
ree in
terv
als o
f
th
e ti
m
e
[1
20
s to
9
0
s
], [9
0s to
30
s ] an
d
[30
s
to
0s ].
In the le
vel tow re
pres
ents the
Num
b
er of
ha
nd
ges
t
ures “
N
G”
done
by th
e player
to place differe
n
t
ob
ject
s i
n
t
h
e
several
c
o
nt
ai
ners
. I
n
t
h
e l
e
vel
t
h
ree
“%N
F” re
prese
n
t
s
t
h
e n
u
m
b
er o
f
faul
t
s
di
vi
de
d
by
t
h
e
num
ber
of
ges
t
ures.
I
n
t
h
e l
e
vel
fi
ve “C
F
”
rep
r
ese
n
t
s
t
h
e cont
ai
ner t
h
at
has t
h
e l
a
r
g
est
num
ber o
f
faul
t
s
co
mmitted
b
y
learn
e
rs.
As last ele
m
en
ts there are sev
e
ral
m
e
ssag
e
s th
at will b
e
d
i
sp
layed
on
t
h
e screen
t
o
assist the learners
.
All the
m
e
ssages s
h
own by the
ru
l
e
s en
gi
ne a
r
e
det
a
i
l
e
d i
n
t
h
e t
a
bl
e
1.
T
h
e a
g
ent
in
tercep
ts all th
e actio
n
s
an
d g
e
stu
r
es o
f
t
h
e learn
e
rs,
and th
en
in
terp
ret
s
all th
e d
a
ta
to
g
i
v
e
th
e correct
message that will be displayed on t
h
e scree
n
,
and t
h
en
t
h
e learne
r will react
accordin
g to t
h
e m
e
ssage already
di
spl
a
y
e
d
o
n
t
h
e scree
n
,
wi
t
h
t
h
e ai
m
t
o
im
prove
hi
s
l
earni
n
g
pr
ocess
.
Tabl
e
1. M
e
ssa
ges s
h
ow
n
by
t
h
e sy
st
em
t
o
g
u
i
d
e l
e
a
r
ne
rs
d
u
ri
ng
t
h
ei
r
seq
u
ence
o
f
se
ri
o
u
s
gam
e
N° Message
M
S
G1
You need to focus
m
o
r
e
to have
a good r
e
sult,
y
ou ar
e in
the beginning y
o
u
can do better
.
M
S
G2
Good per
f
orm
a
nce up to now,
tr
y to keep the sa
m
e
level
of concentr
ation to
get the good r
e
sult.
M
S
G3
You have to focus
m
o
r
e
, y
our
r
e
sult
is too bad,
and y
our
choices ar
e r
a
ndom
.
M
S
G4
E
x
cellent per
f
orm
a
nce,
k
eep going in the sam
e
level.
M
S
G5
You have to do
m
o
r
e
effor
t
,
and
m
ove y
our
hands to collect
m
o
re objects.
MSG6
The Glass
waste is
co
m
posed of
bottl
es, cups, etc
.,
y
ou
have to put those o
b
jects in the Gr
een container
.
M
S
G7
T
h
e M
e
tal waste is
co
m
posed of
cans,
m
e
tallic objects,
you have to put
thos
e objects in the yellow container
.
M
S
G8
T
h
e Or
ganic
waste is co
m
posed of r
e
st of fo
od;
y
ou hav
e
to put those objects in the br
own co
ntainer
.
MSG9
The Tr
ash waste is
co
m
posed of
anything not recyclable
lik
e tir
es, plastic b
a
g,
etc. You have to put those objects
in the
black container
.
M
S
G10
T
h
e Paper waste is
co
m
posed of boo
k
s
,
news paper
,
et
c.
You have to put th
ose obj
ects in the blue container
.
MSG11
The Plastic
waste i
s
co
m
posed of
dif
f
erent of
pl
astic objects; y
ou have to p
u
t those objects in the r
e
d.
M
S
G12
You have done a lot of gestur
es a
nd y
our
per
centage of good choices is lo
w,
try
to focus
m
o
re dur
ing the choice of
objects.
M
S
G13
E
x
cellent per
f
orm
a
nce,
y
ou
m
a
ster
the subject well.
M
S
G14
Bad per
f
orm
a
nce,
y
ou should r
e
vise y
our
c
our
se to under
s
tand the basics
of waste sor
ting.
M
S
G15
You have a lot of pr
oblem
s; all of y
o
ur
c
hoices ar
e r
a
nd
o
m
,
y
ou should r
e
vise y
our
cour
se.
Fig
u
re
5
.
Sch
e
me rep
r
esen
ts
a ru
le tree t
h
at
will g
u
i
d
e
learn
e
rs.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
IJEC
E V
o
l
.
5, No
. 2, A
p
ri
l
20
15
:
34
4 – 3
5
4
35
0
Th
e second
agen
t will feed
a b
ack
en
d
web-b
a
sed
ap
p
licatio
n
;
th
is app
licatio
n
will b
e
u
s
ed
b
y
th
e
teachers a
n
d trainees.
Its m
a
in role is
to
give a global
view conce
r
ni
ng th
e learners
tha
t
play the vi
de
o gam
e
,
and s
h
ow
g
r
ap
hs an
d
das
h
b
o
a
r
d
of l
ear
ni
n
g
anal
y
t
i
c
s. The
schem
e
i
n
t
h
e Fi
gu
re 6
re
pres
ent
t
h
e dat
a
ba
se t
h
at
will sav
e
all t
h
e d
a
ta co
n
c
ern
i
n
g
learn
e
rs, th
is d
a
ta
will b
e
in
terp
reted
b
y
th
e ed
ucatio
n
a
l d
a
ta min
i
ng
alg
o
rith
m
s
, we h
a
v
e
u
s
ed
clu
s
tering
al
go
rith
m
e.g
.
“K-m
ean
s algorith
m
”
th
at will clu
s
ter th
e learn
e
rs
according t
o
t
h
eir
perform
a
n
ces a
n
d the
result
will be
presente
d in
pie fl
owc
h
ar
t. In a
ddition t
h
e
data
record
ed
in
t
h
e d
a
ta b
a
se
will b
e
in
terp
reted
b
y
lear
n
i
n
g
analytics alg
o
r
ith
m
s
b
a
sed
on
user profilin
g
m
e
th
od
an
d so
rt
p
l
ayers si
n
ce their sco
r
e,
n
u
m
b
e
r
o
f
g
e
st
u
r
es,
et
c. th
ere is
o
t
her info
rm
atio
n
will b
e
g
i
v
e
n b
y
t
h
e
agent
a
n
d l
ear
ni
n
g
a
n
al
y
t
i
c
s m
e
t
hods
l
i
k
e
g
r
o
u
p
i
n
g
di
ffe
re
nt
ki
nd
o
f
c
o
nt
ai
ners ac
co
rdi
n
g
t
o
t
h
e
num
ber
o
f
good and ba
d
choices
done by the learner.
All of this
inform
at
ion will be viewed
by
the teachers in orde
r to
m
a
ke a g
o
o
d
d
eci
si
on t
o
i
m
prove
t
h
ei
r
m
a
nner
of t
eac
hi
n
g
.
Fi
gu
re
6.
The
s
c
hem
e
of t
h
e
d
a
t
a
base
The c
o
m
m
uni
cat
i
on
bet
w
ee
n t
h
e
web
ap
pl
i
cat
i
ons t
h
at
h
o
st
s t
h
e
we
b-
base
d se
ri
o
u
s
gam
e
and
d
i
fferen
t
ag
en
ts are prov
id
ed b
y
th
e u
s
e
o
f
th
e j
a
de g
a
teway. Th
is layer will en
sure th
e co
mm
u
n
i
catio
n
b
e
tween
tho
s
e sep
a
rated
syst
e
m
s an
d
will allo
w th
em
to
fun
c
tion
in
h
a
rm
o
n
y
. In
t
h
e
n
e
x
t
section
we will
det
a
i
l
t
h
e dash
boa
r
d
o
f
t
h
e l
earni
ng a
n
al
y
t
i
c
s, an
d al
so t
h
e
ki
n
d
o
f
m
e
ssages gi
ven
by
t
h
e r
u
l
e
en
gi
ne,
t
h
at
will g
u
i
d
e
and
assist th
e learners.
5.
RESULTS
Th
e in
tend
ed
resu
lt is a syste
m
co
m
p
o
s
ed
of a web
-
b
a
sed
ap
p
lication
with
a m
u
lt
i ag
en
t syste
m
, th
e
web
-
based
ap
p
l
i
cat
i
on de
pl
oy
ed i
n
t
h
e se
rve
r
ap
pl
i
cat
i
on
was
devel
ope
d
by
u
s
i
n
g se
ve
ral
t
ech
nol
ogi
e
s
an
d
tools e
.
g. “Ja
v
aScript
APIs,
Aja
x
, Ja
va Te
chn
o
l
o
gi
es,
Ja
de,
D
r
o
o
l
s
,
an
d
Wek
a
[
2
1]
”.
The
com
b
i
n
at
i
ons
o
f
t
h
ese t
echn
o
l
o
gi
es ha
ve ens
u
red t
h
e p
r
ope
r fu
nct
i
o
ni
n
g
of
th
e syste
m
an
d
d
e
sp
ite co
m
p
lex
ity asso
ciated
with
t
h
e im
pl
em
entat
i
on o
f
suc
h
a sy
st
em
. The conce
r
ned
web
-
base
d ap
pl
i
cat
i
on i
s
com
pos
ed o
f
t
w
o
part
s;
t
h
e
fi
rst
pa
rt
i
s
a web
-
based se
ri
ous
gam
e
deve
l
ope
d by
u
s
ing sev
e
ral Jav
a
Scrip
t
APIs th
at
allo
w th
e creatio
n
of
2d
vi
de
o
gam
e
s on t
h
e we
b
;
t
h
eref
ore
,
i
t
need
s j
u
st
a w
e
b b
r
o
w
se
r t
o
be r
u
n
,
i
n
a
d
d
i
t
i
on, t
h
e i
n
t
e
r
act
i
on
bet
w
ee
n l
earne
rs an
d seri
o
u
s gam
e
i
s
done t
h
r
o
ug
h an i
n
t
e
ract
i
v
e i
n
t
e
rfac
e
“l
eap
m
o
t
i
on cont
r
o
l
l
e
r”
ba
sed
o
n
t
h
e ha
nd
s an
d
fi
n
g
er
gest
u
r
es
. The Fi
g
u
re
7
sho
w
s a sc
ree
n
sh
ot
f
r
om
t
h
e pr
op
ose
d
web
-
base
d se
ri
o
u
s
gam
e
abo
u
t
t
h
e
wast
e so
rt
i
n
g
pr
oc
ess;
wi
t
h
t
h
e
a
ssi
st
ance’s
m
e
ssages
di
s
p
l
a
y
e
d
o
n
t
h
e
sc
re
en t
o
gui
de l
e
arne
rs
according to their perform
a
nces and th
eir progressi
on in t
h
e vide
o gam
e
.
The second part is a backe
n
d we
b-
base
d
a
pplication
is
a dashboard use
d
by teachers
a
n
d
t
r
ainees i
n
order to ha
ve a
global vie
w
on thei
r
st
ude
nt
s;
t
h
ey
can m
easure a
n
d
anal
y
ze t
h
e
pe
rf
orm
a
nce
of
t
h
ei
r
st
u
d
en
t
s
by
usi
n
g a
c
o
m
b
i
n
at
i
on
of
several
educat
i
o
nal
dat
a
m
i
ni
ng al
g
o
ri
t
h
m
s
and l
ear
n
i
ng a
n
al
y
t
i
c
s m
e
t
h
o
d
s.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
To
ward
s a S
y
stem
o
f
Gu
idan
ce, Assistan
ce
an
d Lea
r
n
i
ng
An
a
l
ytics Ba
sed
o
n
Mu
lti Ag
en
t …
(Lo
tfi Ela
a
c
h
a
k
)
35
1
Fig
u
re 7
.
Screen
sho
t
o
f
th
e web
-
b
a
sed
serious
g
a
m
e
with
app
a
rition
o
f
assistan
ce m
e
ssag
e
The Fi
gu
re
8
sho
w
s
a pi
e c
h
art
a
b
o
u
t
t
h
e
use
r
p
r
ofi
l
i
n
g
t
h
at
cl
ust
e
rs
several
gr
o
u
p
s
of
l
ear
ners
according to their perform
a
nces. The pi
e chart was fed by
k-m
eans cluster al
gorithm
,
in this pie chart
there
are fo
ur
gr
ou
p
s
of t
h
e l
earne
rs gr
o
upe
d acc
or
di
n
g
t
o
t
h
ei
r
e
perf
orm
a
nce
s
, t
h
e ora
nge
part
o
f
t
h
e pi
e chart
represen
ts
th
e g
ood
learn
e
rs who
un
d
e
rstood
all
th
e
b
a
sics of
waste s
o
rting,
vie
w
to their sc
ore a
n
d the
num
ber
of
t
h
ei
r
han
d
gest
ure
s
. F
o
r
t
h
e
l
ear
ners
t
h
at
bel
o
n
g
t
o
t
h
e
bl
ac
k
part
of
t
h
e
pi
e
cha
r
t
,
a
r
e
gen
e
ral
l
y
go
o
d
, b
u
t
t
h
ey
need s
o
m
e
gui
dance a
nd e
xpl
anat
i
on a
b
o
u
t
the basics of the waste sor
ting. Tak
i
ng
th
e case o
f
the green
pa
rt of t
h
e pie c
h
art the l
ear
ner
s
t
h
at
bel
o
n
g
t
o
t
h
i
s
part
,
ha
ve an
ave
r
age
l
e
vel
,
t
h
ey
ha
ve f
e
w
p
r
ob
lem
s
r
e
g
a
rd
ing
th
ei
r
und
er
stand
i
ng
ab
out th
e pr
opo
se
d
topic a
n
d they
need an
explanation a
n
d assistance
t
o
u
nde
rst
a
nd t
h
e ba
si
cs o
f
w
a
st
e sort
i
n
g.
F
o
r t
h
e l
a
st
pa
rt
t
h
e bl
ue pa
rt
t
h
e m
o
st
of t
h
e
l
earner
s
t
h
at
be
l
o
n
g
t
o
th
is p
a
rt of th
e p
i
e ch
art
h
a
v
e
sev
e
ral prob
lem
s
an
d
d
i
fficulties, th
ey h
a
ve cho
s
en th
e
ob
j
ects ran
d
o
m
l
y
and
wi
t
h
o
u
t
t
h
i
nki
ng
. T
h
ey
nee
d
ex
pl
i
cat
i
on
o
n
basi
cs
of
waste sorti
n
g, a
n
d a s
p
ecial assistance, i
n
orde
r, t
o
increase their level of com
p
re
hension. In the sam
e
web interface there is
a gra
p
h about a num
ber
of
good
a
nd
bad c
h
oices groupe
d by t
h
e type of
the c
o
ntainer.
The
r
e is
also a we
b
inte
rface t
h
at shows a
gra
p
h about the
l
earner
s
’ i
n
f
o
r
m
at
i
on i
n
t
h
i
s
case we
ha
ve
prese
n
t
e
d
t
h
e
num
ber
of
gest
ures
an
d t
o
t
a
l
score
o
f
eac
h l
earne
r
,
as shown in the Figure 9. T
h
is repr
esentation give
s a globa
l
view on the a
c
tivity of each learner that
plays the
seri
o
u
s
gam
e
, and
wi
t
h
t
h
i
s
d
a
sh
boa
rd
t
h
e i
n
st
r
u
ct
o
r
can
m
a
ke deci
si
o
n
,
i
n
o
r
der t
o
i
m
pr
o
v
e an
d c
h
a
nge
hi
s
manner and st
rategy of teachi
n
g with the ai
m
of transm
it knowledge i
n
a
correct way.
Figure 8.
Scree
n
shot of
the da
shboa
r
d
conce
r
ning User
profi
li
ng a
n
d Num
b
er
of
fa
ults and succes
s
by
cont
ai
ne
rs
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
IJEC
E V
o
l
.
5, No
. 2, A
p
ri
l
20
15
:
34
4 – 3
5
4
35
2
Fi
gu
re
9.
Scree
n
sh
ot
of
t
h
e
da
shb
o
a
r
d
co
nce
r
ni
n
g
l
ear
ner
s
’s
pe
rf
orm
a
nces
The se
ri
o
u
s
ga
m
e
and t
h
e
ba
cken
d
web
-
bas
e
d ap
p
lication comm
unicate
with
guidance/
assistance
agent a
n
d learning analytics
agent
via a J
a
de Gate
wa
y,
th
is in
terface in
sure
s a
real-t
im
e co
mm
unication
b
e
tween
two
heterog
e
n
e
ou
s en
v
i
ron
m
en
ts web
and
m
u
lti a
g
en
t system
. with
th
is co
m
p
le
men
t
ary syste
m
th
at
will en
ab
le learn
e
rs to
i
n
crease th
eir lev
e
l
of learn
i
ng
and
th
eir
lev
e
l o
f
un
d
e
rstand
ing
based
o
n
t
h
e
m
e
ssag
e
s
and
g
u
i
d
a
n
ce of
fere
d by
t
h
e sy
st
em
duri
ng a
seq
u
e
n
ce of vi
de
o gam
e
s,
al
t
hou
g
h
t
h
e
ap
pl
i
cat
i
on Das
h
bo
a
r
d
will enable teachers t
o
ada
p
t their
teaching strategies
and m
e
thods
base
d on
perform
ance and t
h
e res
u
lts of
th
eir stud
en
ts.
Co
m
p
arin
g
t
h
e app
r
o
ach fo
llowed in
th
is p
a
p
e
r t
o
estab
lish th
e lean
i
n
g
analytics syste
m
,
with
o
t
h
e
r
app
r
oaches
i
n
ot
he
r
co
nt
ri
b
u
t
i
ons [2
2, 23]
, we
can
c
o
nc
lud
e
th
at th
e m
a
i
n
obj
ectiv
e of
th
e pr
opo
sed
wo
rk
and
othe
r contribut
i
on is to sim
p
lify teach
ers’ tas
k
when
using s
e
rious gam
e
s
by providi
ng re
al
-tim
e information
of the act
ual learne
r’
use of t
h
e gam
e
s while in the cla
ssroom
.
But the concept a
nd
t
h
e technologies
us
ed are
d
i
f
f
e
r
e
n
t
, in
the o
t
h
e
r
con
t
r
i
bu
tio
n
t
h
e au
t
h
or
s h
a
v
e
u
s
ed
XML tech
no
logy to
d
e
liv
er
assessm
en
t d
a
ta to
the
teachers,
by cons in the
proposed pa
pe
r we
have use
d
a c
o
m
b
ination of e
ducational data
m
i
ning and le
arni
ng
analytics to extract bene
fit inform
ati
on to t
h
e teache
r
s in
order, they ha
ve
a gl
obal vi
ew about the l
e
vel of
th
eir learn
e
rs,
an
d
will allo
w
th
em
to
p
r
o
p
e
rly id
en
tify
learn
e
rs
g
a
p
s
. Regard
i
n
g
the assistan
ce and
gu
i
d
an
ce
syste
m
th
e stud
y abou
t so
lari
s on
e seriou
s
ga
m
e
[1
] th
e the au
thors seek
to
first assess th
e st
u
d
e
n
t
’s skill to
ascertain the a
m
ount and type of hel
p
to provi
de. Be
fore
each gam
e
, a set of m
u
ltiple-choice
questions are
prese
n
ted to
fi
nd out
what t
h
e studen
t
k
n
o
w
s an
d
wh
at
t
h
e
y
m
a
y
need
hel
p
. B
y
c
o
ns
ou
r
sy
st
em
represe
n
t
s
a
n
adeve
n
tage be
cause it reacts with lear
ne
rs a
u
tom
a
tically a
nd acc
ording to
their progres
s in the video
gam
e
,
wh
ich
will allo
w learn
e
rs to liv
e
a b
e
n
e
ficial ex
p
e
rien
ce.
6.
CO
NCL
USI
O
N
In t
h
i
s
pa
per
we ha
ve
det
a
i
l
e
d t
h
e est
a
bl
i
s
hm
ent
of g
u
i
d
ance
, assi
st
a
n
ce an
d l
ear
ni
ng a
n
al
y
t
i
c
s
syste
m
b
a
sed
o
n
a m
u
lti ag
en
t system
, i
t
will wo
rk
in
co
l
l
ab
oratio
n
with
a seriou
s
g
a
me d
e
v
e
lop
e
d
b
y
ou
r
researc
h
team
.
The proposed syste
m
has s
e
veral adva
nt
a
g
es and be
ne
fits that concern both learners and
i
n
st
ruct
ors
.
O
n
e of t
h
e m
o
st
im
port
a
nt
ad
va
nt
ages
of
o
u
r
s
y
ste
m
is it efficiency. In one
hand it can as
sist and
gui
des the learners
, which wi
ll im
prove their level of
understanding and accelerates
their learning process
an
d th
eir an
aly
tical sk
ills, o
u
r syste
m
p
r
ov
ides an
env
i
ro
n
m
en
t bo
th in
teractiv
e and
en
tertain
i
n
g
t
h
at carefu
lly
target the knowledge that the learner
m
u
st acqui
re. In the
ot
her hand it will allow teach
ers and i
n
structors to
make their de
cisions
easily, in
orde
r to i
m
prove t
h
eir
teaching m
e
thodologies, stra
tegies and teaching
man
n
e
rs, th
e decisio
n
will b
e
m
a
d
e
with
the h
e
lp
of th
e
d
a
sh
bo
ard
t
h
at h
a
s sev
e
ral grap
hs and
flowch
arts
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
To
ward
s a S
y
stem
o
f
Gu
idan
ce, Assistan
ce
an
d Lea
r
n
i
ng
An
a
l
ytics Ba
sed
o
n
Mu
lti Ag
en
t …
(Lo
tfi Ela
a
c
h
a
k
)
35
3
wh
ich
are fed b
y
th
e learn
e
rs’ data in
terpreted
v
i
a
bo
th ed
u
cation
a
l data
min
i
n
g
algo
rith
m
s
an
d
lean
ing
anal
y
t
i
c
s
m
e
t
h
ods
. Am
ong t
h
e pe
rsp
ect
i
v
e
s
envi
sa
ge
d t
h
ere i
s
t
h
e evol
ut
i
on
of t
h
e cu
rre
nt
sy
st
em
by
addi
n
g
ot
he
r age
n
t
s
eq
ui
p
p
ed
wi
t
h
ot
her al
go
ri
t
h
m
s
of t
h
e e
d
ucat
i
o
nal
dat
a
m
i
ni
ng,
ot
her l
e
a
r
ni
n
g
anal
y
t
i
c
s m
e
tho
d
s
,
an
d
m
ach
in
e l
earn
i
n
g
al
g
o
rith
m
s
, th
e i
m
p
l
e
m
en
tatio
n
of su
ch
system
will o
p
e
n
a
n
e
w
trend
toward
s
sm
ar
t
serious gam
e
s.
ACKNOWLE
DGE
M
ENTS
Thi
s
resear
ch
pape
r i
s
m
a
de pos
si
bl
e t
h
r
o
u
gh t
h
e
hel
p
an
d su
pp
o
r
t
fro
m
t
h
e st
udent
s
of com
put
er
engi
neeri
n
g
.
We
grat
ef
ul
l
y
ack
no
wl
ed
ge
t
h
e s
u
pp
ort
of t
h
e
bac
h
el
or
, m
a
st
er st
ude
nt
s a
n
d al
l
ot
he
r
p
a
rticip
an
ts.
REFERE
NC
ES
[1]
Yang T, Kauser
J, Sachin
S, C
h
ristopher JF.
S
o
laris One – A
Serious Game for Thermodynamics
. 121st ASEE
Annual conf
eren
ce
&
Exposition
.
Indian
apolis
. 20
14.
[2]
J
.
Baals
r
ud Hau
g
e, F
.
Bel
l
ott
i
,
R.
Na
dolski,
M. Kic
k
meier-Rust, R. Berta,
M.
B. Carvalho
.
Deploying Seriou
s
Games for Mana
gement in Higher Educ
ation: les
s
ons learned and good practices
. P
r
oceedings
of
the 7th Europe
a
n
conferen
ce on g
a
mes based
lear
ning, 2013
, pp
. 2
25-234.
[3]
Jay
f
us D,
Ka
thle
e
n
H.
Exte
nding the
‘Se
r
ious Ga
me
’
Bounda
ry
: Virtua
l Instruc
t
ors
in Mobile
Mixe
d Re
a
lity
Learn
i
ng Games
.
Digital
Games Research Association
Conference
. 2007
.
[4]
Reese, D. (2014
).
Digital Knowledge Maps: The
Foundation for Le
arning Analytics Through
In
structional Games
.
In D. If
enth
aler
& R. Hanewald
(Eds.), Digital Know
ledge Maps
in Edu
cation (pp
.
299-327): Springer New York
.
[5]
Minovi
ć
, M., Milovanovi
ć
, M
.
Real-
tim
e le
ar
ning anal
yt
ics
i
n
educat
ional g
a
m
e
s
.
P
a
per pre
s
ented at th
e 1
s
t
International Co
nference on
Technological
Ecos
y
s
tem for
Enh
a
n
c
ing Multicu
ltur
a
lity
.
(2013
).
[6]
Fulantel
li
, G., T
a
ibi
,
D., & Arrig
o
, M. A sem
a
ntic approa
ch
to m
obile l
earning
an
al
yti
c
s
.
P
a
per pr
es
ented
at the 1s
t
International Co
nference on
Technological
Ecos
y
s
tem for
Enh
a
n
c
ing Multicu
ltur
a
lity
.
(2013
).
[7]
Greller
,
W.,
Ebn
e
r, M., & Schön
, M. Learn
i
ng A
n
aly
t
ics:
From Theor
y
to Practice–
Data Support
for Learn
i
ng and
Te
ac
hing Compute
r
Assiste
d
As
se
ssme
n
t.
Re
se
arc
h
into E-Asse
ssme
n
t
(pp. 79-87
): Springer
.
(201
4).
[8]
Fay
y
a
d
,
U. M.,
Pitatesk
y
-
Shapir
o, G.,
Sm
y
t
h
,
P., and
Uthurasamy
,
R. (1996).
Ad
vances
in Knowledge Disco
ver
y
and Data Min
i
n
g,
AAAI/MIT Press.
[9]
Frawley
W., Piatetski-Shapiro
G. and Matheus C. (1992)
. Knowledge discover
y
in
databases: an
overview
.
AI
Magazine
. 14
(3)
.
57-70
.
[10]
Erdogan & Timor. (2005). A Data Mining
Application in A Stu
d
ent Database.
Journal of Aeron
autic and Space
Technologies
. Ju
ly
2005 Vol. 2
N
u
mber 2 (53-57).
[11]
Margaret H. Dunham “Data Min
i
ng In
trodu
ctor
y
and
Advanced
Topics”.
[12]
Corbett, A., & Anderson, J. (1994).
Knowledge Tracing: Modeling the Ac
quisition of Procedural Knowledge.
User
Modeling
and U
s
er-Adapted In
teraction
, 4, 253-2
78. http://dx.doi.org/10.1007/BF
01099821.
[13]
Baker, R
.
and
Aleven, V
.
2008.
I
m
proving contextual models
of g
u
essing and slip
ping with a
truncated
training set.
In Educationa
l Data Mining 2008
. 1s
t Interna
t
i
onal Conferen
ce
on Education
a
l
Data M
i
ning, P
r
oceed
ings
. UNC-
Charlotte, Computer
Science D
e
pt., M
ontreal, Canada, 67–76.
[14]
Baker, R
.
Corbett, A.T Ke
oding
er, K., & Wagner, A.Z.
Off-tas
k
behavior in the
cogniti
ve tu
tor classroom: When
students game the system
. In: P
r
oceed
ing of the
S
I
GCHI Confe
r
ence In hum
an
F
actors
in Co
m
puting S
y
s
t
em
s
.
2004. (pp
.
383-3
90).
[15]
O.
G.
Selfridge.
Pandemonium:
a paradigm for learning
.
In P
r
oceed
ings
of th
e
S
y
m
pos
ium
on M
echanis
a
tion o
f
Thought Processes (pp. 511-529)
. Her
Maje
sty
'
s
Stationer
y
Office, London.
1959.
[16]
M. Minsky
. Th
e society
th
eor
y
of
thi
nking. In Artificial intelligen
ce: an
MIT persp
ective (pp. 423-
450). MIT Press.
1979.
[17]
Wooldridge, M. and Jennings,
N.R. (1995). In
telligent
Agen
ts: Theor
y
and P
r
actic
e. Knowledge Engineering
Review, 10(2), 1
15-152.
[18]
Leap
Motion
Controller. Availab
l
e on
line: https
://www.leapmotio
n.com (accessed
on 21 Octob
e
r 2
014).
[19]
JAVA Agent Development Fram
ework. Availab
l
e on
lin
e:
h
ttp://jade.tilab.c
om/ (accessed on
21 O
c
tober
2014).
[20]
Drools a Business Rules Mana
gement S
y
stem.
Available on line: http
://www.
drools.
org/ (acc
essed on 21 October
2014).
[21]
Weka : Data Mining Software
in Java Available on line http://www.cs
.waikato
.ac.nz/ml/weka// (accessed on 21
October 2014)
.
[22]
Ángel S, Jav
i
er
T, Pablo M, B
a
ltas
ar F (2014)
: Application of
Learning An
aly
t
ics in
Edu
cational Vid
e
ogames.
Enter
t
ainment C
o
mputing, Elsev
i
er. Volume
5
,
Is
sue 4, December 2014, Pages 31
3–322.
[23]
A,Serrano-Lagu
na, A.B, Fern
ández-Manjón
. (
2014). Apply
i
n
g
learning
an
aly
t
ics to simplify
serious games
deplo
y
m
e
nt in
the classroom.
Proceedings of
the
2014 I
EEE Global Engineer
ing Edu
cation Confer
ence
(EDUCON) Pages 872-877.
Evaluation Warning : The document was created with Spire.PDF for Python.