Internati
o
nal
Journal of Ele
c
trical
and Computer
Engineering
(IJE
CE)
V
o
l.
6, N
o
. 5
,
O
c
tob
e
r
201
6, p
p
. 2
488
~249
8
I
S
SN
: 208
8-8
7
0
8
,
D
O
I
:
10.115
91
/ij
ece.v6
i
5.1
213
6
2
488
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
The Analysis of Student Colla
borative Work Inside Social
Learning Network Analysis
Based on Degree and
Eigenvector Centrality
Andi
Besse
Fi
rdausi
ah Man
s
ur, Nor
a
z
a
h
Yus
o
f, Ahm
a
d
H
o
irul
Bas
o
ri
Faculty
of
Computing and
Inform
atio
n
Technolog
y
R
a
bigh
, King
Abdulaziz Un
iv
ersity
, Kingdom o
f
Saudi Ar
abia
Article Info
A
B
STRAC
T
Article histo
r
y:
Received
J
u
n 10, 2016
Rev
i
sed
Au
g
18
, 20
16
Accepted Aug 30, 2016
S
o
cial l
earning
network anal
ys
is
is
a potential approa
ch to
anal
yz
e th
e
behaviour o
f
stu
d
ents in
collabo
r
ativ
e work. Ho
wever, most of the previous
works focus on as
y
n
chronous dis
c
ussion forum as the learning
activity
.
Ver
y
few of them
are tr
ying to
anal
yz
e th
e s
t
udents
'
co
llabor
at
ive
work while
using wiki e-learning.
This p
a
per
proposes
the degr
ee cen
t
rality
and
eigenv
ector
method for iden
tif
y
i
ng the
co
llaborative work of
stu
d
ents while
in wiki e-learnin
g
. The log data
of
the Moodle e-
learn
i
ng s
y
stem is observed
that
records
the
students' a
c
tiv
iti
es and
act
ions w
h
ile using
wiki
.
The
result
s
hows
that there is
a clos
e s
i
m
ilarit
y
betw
een th
e degree c
e
ntr
a
l
i
t
y
and the
eigenv
ector
.
The
res
u
lt
als
o
rev
e
als
th
e students
who obtain h
i
gh
outdegr
ee
values. Furth
e
r
m
ore, Agent_1
and Agent_12 r
e
present
the stu
d
ents who
obtain
e
d high outdegree valu
es, which
mean these two nodes are acting as
source providers
that ab
le to sup
p
ly
information
and knowledge through the
network.
This r
e
sult also
strength
e
ned
b
y
va
lue of
clos
en
es
s
and b
e
tweenn
e
s
s
where Agent_1
and Agent_12
lead
ing on th
is
m
eas
urem
ent. The
hig
h
closeness value of Agent_1
and Agen
t_12 w
ill
lead in
to fast spreadin
g
information since they
hav
e
fastest rout
e and has
the most direct r
oute to th
e
other node inside the network
,
thus
collabor
ativ
e work is
eas
y
to b
e
initi
ali
zed
b
y
thes
e Agen
ts
.
This
work
has successfully iden
tifi
e
d
collaborative work of student. This fi
nding is believed to brin
g enormous
benefit on
the e-
learning
s
y
stem improvement in
the futur
e
.
Keyword:
Aut
h
o
r
i
n
g t
ool
s an
d m
e
t
hods
Co
llab
o
rativ
e l
earn
i
n
g
Co
op
erativ
e learn
i
n
g
Peda
go
gi
cal
i
s
sues
Copyright ©
201
6 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
:
Andi Besse
Firdausiah Ma
ns
ur,
Depa
rtem
ent of Com
puter Sci
e
nce,
Facul
t
y
o
f
C
o
m
put
i
ng an
d
I
n
fo
rm
ati
on Tec
h
n
o
l
o
gy
R
a
bi
g
h
Ki
n
g
Ab
d
u
l
azi
z Uni
v
ersi
t
y
,
Rabigh,
Ma
kkah, Saudi Ara
b
ia.
Em
a
il: ab
m
a
n
s
u
r
@k
au
.edu
.sa
1.
INTRODUCTION
Social network analysis
(SNA) is one of the widely us
ed ap
pr
oac
h
t
o
anal
y
ze t
h
e
beha
vi
o
u
r
o
f
certain
co
mm
u
n
ity o
f
so
cial network. Th
is is d
u
e
t
o
th
e
fact
that, it can re
prese
n
t the s
o
c
i
al relation bet
w
een
people [1],[2]. Teachers
a
n
d
research
e
r
s
often em
ployed the social net
w
or
k a
n
alysis (SNA) to dete
rmine the
lev
e
l o
f
p
a
rticip
an
ts, to
id
en
ti
fy th
e cen
tral acto
r
s,
o
r
to
rec
o
gnize ot
her st
ruct
ural cha
r
ac
teristics of the social
learn
i
ng
i
n
teractio
n
s
[3
].
W
i
k
i
is a co
llab
o
rativ
e too
l
t
h
at offers
u
s
ers to
work
t
o
g
e
th
er
with
in th
e
sam
e
sys
t
e
m
t
o
in
crease the
resu
lt an
d prod
u
c
tiv
ity of t
h
e tea
m
wo
rk
[4]. Typ
i
cally, it
recorded the
page
of the
or
i
g
in
al
co
n
t
r
i
bu
to
r
of
a
piece of writing and those
who
ha
ve e
d
ited and m
a
de am
endm
ents. Se
veral social learning activities, whic
h
are co
nsi
d
e
r
ed
as col
l
a
bo
rat
i
v
e wo
rk are e
d
i
t
i
ng,
upl
oadi
ng
,
com
m
e
nt
i
ng and t
a
g
g
i
n
g [
5
]
.
The ot
he
r w
o
r
k
o
n
W
i
k
i
is pr
opo
sed
b
y
Twu
in
2
010
. H
e
stud
i
e
s th
e in
ter
acti
o
n am
o
n
g
studen
t
s in Ch
i
n
ese ESL Classroo
m.
Th
eir m
e
th
o
d
s are
d
i
v
i
d
e
d
i
n
to
two
parts:
Attitu
d
e
towa
rd
s
W
i
k
i
an
d
W
i
k
i
in
teraction
[6
]. Furth
e
rm
o
r
e, th
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
The
An
al
ysi
s
o
f
St
u
d
ent
C
o
l
l
a
bor
at
i
ve W
o
rk
Insi
de
Soci
a
l
L
e
ar
ni
n
g
...
. (
A
ndi
Besse Fi
r
d
ausi
a
h M
a
n
s
ur
)
2
489
W
i
k
i
p
e
d
i
a syste
m
h
a
s b
een
an
alyzed
u
s
i
n
g
cen
trality
m
e
a
s
u
r
em
en
t b
y
Ko
rfiatis et. al [7
]. Th
ey h
a
v
e
u
s
ed
in
d
e
g
r
ee cen
t
rality
to
m
easu
r
e h
o
w m
u
ch
th
e ch
ang
e
th
at co
n
t
ribu
ted
b
y
an
acto
r
th
at ed
i
t
ed
b
y
an
o
t
h
e
r
acto
r
.
They also utilized fuzzy ope
rator to aggre
g
a
t
e the differe
nc
e between
the
recent ve
rsion
of the article and the
su
b
m
itted
o
n
e
[7
].
W
i
k
i
Mo
od
le h
a
s sev
e
ral
featu
r
es in
M
o
od
le e-learn
i
ng
syste
m
su
ch
as Add
a n
e
w to
p
i
c,
Ed
it o
r
Up
d
a
te, Co
mm
en
t, Li
n
k
and
View. It is ab
le to
g
i
ve stu
d
e
n
t
s and
lectu
r
ers
flex
i
b
ilit
ies to
co
n
t
rol th
eir
ow
n Wi
ki
.
M
o
st
of t
h
e p
r
evi
o
us resea
r
c
h
f
o
cu
sed
on a
n
al
y
z
i
ng t
h
e c
ont
e
n
t
of a
s
y
n
c
hr
o
n
o
u
s
di
scu
ssi
on s
u
c
h
as
d
i
scu
ssi
on
fo
ru
m
s
an
d
m
e
ss
ag
ing
.
On
the o
t
h
e
r h
a
n
d
, the works on
Wik
i
still d
i
d
n
’
t
to
u
c
h
Moo
d
l
e
W
i
k
i
(speci
fi
cal
l
y
fo
r e-l
e
a
r
ni
ng
), t
h
ey
o
n
l
y
f
o
c
u
s
i
ng
o
n
reg
u
l
a
r
W
i
ki
s
u
c
h
as
W
i
ki
pe
di
a.
B
e
si
des t
h
at
,
m
o
st
of
t
h
e
work foc
u
ses
on the i
nde
gree
centrality. Howeve
r, t
h
ey
are still lacking i
n
term
of the
outde
g
ree m
e
thod
and
othe
r ce
ntrality m
e
thods
s
u
ch
as close
n
ess a
n
d
betwee
nne
ss. The
r
efore
,
the
obj
ective
of this pape
r is t
o
fi
ll in
t
hose rese
arc
h
gaps by
f
o
cu
si
ng o
n
i
m
pl
em
ent
i
ng SN
A
t
o
anal
y
ze t
h
e part
i
c
i
p
at
i
o
n
and i
n
t
e
ract
i
on
of
stude
nts i
n
side
Moodle
Wi
ki
by
using SNA m
e
thod s
u
c
h
a
s
De
gree
Cent
rality, Closenes
s, Betwee
nnes
s and
Eig
e
nv
ector
.
So
cial Netwo
r
k
An
alysis (SNA) is th
e stu
d
y
o
f
th
e st
ruc
t
ure of social interac
t
i
o
ns [
8
]
.
Furt
he
rm
ore,
Social learning network analy
s
is is th
e
analy
s
is of s
o
cial networks
in
the
elearni
ng
domain. Because
social
l
earni
n
g
net
w
o
r
k
i
s
a
ne
w
wa
y
of
com
m
uni
cat
i
on
net
w
or
k,
i
t
can i
n
fl
ue
nc
e t
h
e t
eac
hi
n
g
and
l
ear
ni
n
g
pr
oces
s
[9]
.
P
r
e
v
i
o
us r
e
searche
r
s
ha
v
e
pr
o
pose
d
a
u
t
o
m
a
ti
c doc
um
ent
t
e
xt
anal
y
s
i
s
(A
TA
) i
n
t
h
e
st
ude
nt
t
e
xt
. T
h
e t
e
xt
of t
h
e st
ude
nt
’
s
m
e
ssage was
abl
e
t
o
ex
pl
ai
n t
h
e
em
ot
i
ons
of
t
h
e
t
u
t
o
rs
and
st
u
d
e
n
t
s
d
u
ri
ng
t
h
e
pr
oc
ess o
f
interaction
[10]-[13]. The
other re
searc
h
a
n
alyse te
xt
m
e
ssages
f
r
o
m
st
ude
nt
s
cat
egorize
d
and analyzed
according to the content of the
m
e
ssage. This
m
e
thod ap
pe
ars to produce
em
otiona
l behavior from
students
an
d
tu
t
o
rs
d
u
rin
g
th
e in
teractio
n
.
In
add
ition
,
th
is no
d
e
correspo
n
d
s
to
a h
u
m
an
, an
ag
en
t o
r
an
actor in
the
co
mm
u
n
ity. Sev
e
ral tech
n
i
qu
es in SNA
are th
e
d
e
g
r
ee cen
trality,
b
e
tween
n
e
ss an
d clo
s
en
ess. Th
ese
tech
n
i
qu
es can b
e
u
tilized
for b
e
h
a
v
i
ou
r iden
tificatio
n
.
Th
ey also
ab
le to
rev
eal b
e
h
a
v
i
ou
r
o
f
u
s
ers in
sid
e
soci
al
l
ear
ni
n
g
net
w
or
k [1
4]
-
[
18]
.
Th
e d
e
g
r
ee cen
t
rality co
m
p
oses o
f
th
e ou
tdeg
ree and
th
e in
d
e
gree. Th
e
ou
td
eg
ree is in
terpreted
as a
num
ber
of i
n
f
o
rm
ati
on o
r
kn
o
w
l
e
d
g
e t
h
at
bei
ng
sp
read
f
r
om
cert
a
i
n
n
o
d
e (
v
ert
e
x) t
o
t
h
e
o
t
her
no
de
(o
ut
g
o
i
n
g
edge
). Meanwhile, the inde
gree is foc
u
sed on calculatin
g the num
b
er of inform
ation that a node receive
d
(i
n
goi
ng e
d
g
e
)
.
In a
d
di
t
i
on, c
l
osenes
s i
s
a t
echni
que t
o
measure t
h
e time neede
d
to
spread
the in
fo
rmatio
n
from
an initial
node t
o
anot
he
r node
by cons
idering the s
h
ortest path. Be
tweenness
is a
m
e
thod t
o
determ
ine
t
h
e n
o
d
e t
h
at
co
nt
r
o
l
t
h
e c
o
m
m
uni
cat
i
on fo
r
ot
he
r
n
o
d
e i
n
si
de t
h
e
net
w
or
k
(can
be cal
l
e
d a
s
a
h
u
b
)
.
Furt
herm
ore,
Ei
gen
v
ect
o
r
re
prese
n
t
s
t
h
e c
o
nnect
i
o
n
of a
no
de t
o
ot
he
r wel
l
con
n
ect
ed
no
de. T
h
i
s
m
eans, a
no
de t
h
at
has
hi
g
h
Ei
gen
v
ect
or ha
s t
h
e pot
e
n
t
i
a
l
t
o
spread
i
n
fo
rm
ati
on f
a
st
and sm
oot
h i
n
si
de t
h
e ne
t
w
o
r
k
.
Furt
herm
ore, t
h
e ot
her re
sear
cher
pr
o
pose
d
C
o
m
put
er-
A
i
d
ed C
o
l
l
a
b
o
rat
i
ve Deci
si
o
n
M
a
ki
n
g
(C
A-C
D
M
)
t
h
at
can hel
p
st
ude
nt
t
o
o
p
t
i
m
i
ze
thei
r cri
t
i
cal
t
h
i
nki
ng a
nd
deci
si
on
pr
ocess d
u
ri
ng t
h
e st
udy
[1
9]
.
W
h
i
l
e
t
h
e ot
he
r
t
r
y
t
o
m
easure t
h
e col
l
a
bor
at
i
on an
d co
m
m
uni
cat
i
on of
stud
en
t th
ro
ugh
so
cial netw
or
k
an
alysi
s
. Th
ey
recomm
end that lecturer can
provid
e m
o
re
o
r
ga
ni
zed
eve
n
t
s
t
h
at
st
re
g
h
t
h
e
n
st
ude
nt
c
o
l
l
a
bo
rat
i
o
n [
2
0]
.
2.
R
E
SEARC
H M
ETHOD
In
t
h
is research th
ere are t
w
o
main
p
h
a
ses,
first th
e
d
a
ta co
llectio
n
wh
ich is
ob
tain
ed
fr
om
e-
lear
n
i
ng
lo
g
d
a
ta, th
en
th
e second
p
a
rt
is so
cial n
e
twork
an
alysis fo
r
wik
i
. Th
e d
a
ta for an
alysis is th
e lo
g
activ
ities o
f
t
h
e M
o
odl
e
W
i
ki
co
nd
uct
e
d
i
n
sem
e
st
er 20
10/
20
1
1
f
o
r t
h
e co
urse
nam
e
d "I
nst
r
um
ent
a
t
i
on i
n
A
n
al
y
t
i
cal
Ch
em
is
try" with
th
e course co
d
e
o
f
SSC
2
2
1
3
-0
1.
Th
ere
are sev
e
ral activ
ities in
Mo
odle W
i
k
i
su
ch
as "
Add
new Wiki Topic
", "
Ed
it co
n
t
en
t
", "
View content
" and "
Add Li
nk
". Th
is activ
ity
lo
g
o
n
Mood
le
Wik
i
is
capt
u
red
f
o
r
t
h
ree m
ont
hs
pe
r
i
od.
The
dat
a
fr
om
M
o
o
d
l
e
W
i
ki
i
s
cl
assi
fi
ed
ba
sed
o
n
th
e user
p
a
rticip
ation
to
th
e E-Learnin
g
syste
m
.
This data is then conve
rted int
o
adj
acency
matrix in orde
r to m
easure the so
cial interaction
betwee
n the
users
.
The a
d
jacency
matrix for t
h
e
data is
prese
n
t
e
d in Ta
ble 1.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
I
J
ECE
Vo
l. 6
,
N
o
. 5
,
O
c
tob
e
r
20
16
:
248
8
–
24
98
2
490
Tabl
e 1. A
d
jac
e
ncy
m
a
t
r
i
x
fo
r W
i
ki
Dat
a
Th
e
d
a
ta fro
m
th
e adj
acen
c
y matrix
is th
en
b
e
ing
co
m
p
u
t
ed
fo
r t
h
e d
e
gree cen
trality, th
e clo
s
en
ess
an
d
th
e
b
e
tween
n
e
ss cen
tral
ity. Tab
l
e 2
sh
ows th
e d
e
gree cen
trality
wh
ich
p
r
esen
t
s
th
e in
d
e
gree an
d
out
degree
results. The inde
gree and outde
g
ree m
easure
m
e
n
t
are
n
ecessary in
o
r
d
e
r to
see th
e activ
ity lev
e
l of
each stude
n
t. T
h
e indegree is
a calculati
on of the num
b
er of directe
d
bi
nd
to
the node. On the
other
ha
nd, the
out
deg
r
ee i
s
t
h
e num
ber o
f
bi
nds t
h
at
t
h
e n
o
d
e di
rect
s t
o
ot
hers
. I
n
an e
x
a
m
pl
e of a fri
en
dshi
p rel
a
t
i
o
ns
, t
h
e
in
d
e
g
r
ee m
a
y
b
e
in
terp
reted
as a form
o
f
at
tractiv
en
ess
wh
ile th
e o
u
t
d
e
gree is in
terpret
e
d
as so
ciab
ility. Th
e
deg
r
ee ce
nt
ral
i
t
y
of st
ude
nt
i
n
si
de s
o
ci
al
l
ear
ni
n
g
net
w
or
k c
a
n
be c
o
m
put
ed t
h
r
o
u
g
h
E
q
u
a
t
i
on
1.
Fo
r gr
ap
h G
:
=(
V,E
) with
n
v
e
rtices, th
e
d
e
gree cen
trality C
D
(
v
) f
o
r ve
rtex
v
is:
(1
)
In T
a
bl
e
1, t
h
e
sym
bol
U i
s
s
y
m
bol
i
zed as
Age
n
t
_
1.
It
ca
n be
seen
t
h
at
Age
n
t
_
1
2
(
U
1
2
)
,
A
g
e
n
t
_
1
(U
1) a
n
d A
g
e
n
t
_
21
(U
2
1
)
h
a
ve hi
gh
o
u
t
d
egree
val
u
es
.
These m
ean t
h
at
t
hose st
u
d
e
nt
s gi
ve si
g
n
i
fi
can
t
cont
ri
b
u
t
i
on t
o
ot
her
Wi
ki
us
ers i
n
t
e
rm
s of
edi
t
i
ng, a
d
di
n
g
,
or e
v
en
u
p
d
a
t
i
ng t
h
e
W
i
ki
.
On t
h
e ot
h
e
r
han
d
,
Age
n
t
_
7,
A
g
e
n
t
_
6,
Age
n
t
_
5,
Age
n
t
_
3 a
nd
Age
n
t
_
1
0
rece
i
v
e hi
g
h
i
n
de
g
r
ee val
u
es. T
h
i
s
m
ean t
h
at
t
h
e
W
i
ki
page
t
h
at
bei
n
g
edi
t
e
d
by
t
h
es
e st
u
d
ent
s
ha
ve
bee
n
e
d
i
t
e
d
by
t
h
e m
o
st
use
r
s
.
Tab
l
e
2
.
Deg
r
ee Cen
t
rality measu
r
e fo
r
W
i
ki d
a
ta
Na
m
e
Indegree
Outdegree
Na
m
e
Indegree
Outdegree
Agent_1
1.
00
26.
00
Agent_20
1.
00
0
Agent_2
3.
00
0
Agent_21
1.
00
11.
00
Agent_3
5.
00
0
Agent_22
1.
00
0
Agent_4
3.
00
0
Agent_23
1.
00
0
Agent_5
4.
00
0
Agent_24
1.
00
0
Agent_6
4.
00
0
Agent_25
1.
00
0
Agent_7
6.
00
0
Agent_26
1.
00
0
Agent_8
2.
00
0
Agent_
27
1.
00
0
Agent_9
3.
00
0
Agent_
28
1.
00
0
Agent_10
4.
00
0
Agent_
29
1.
00
0
Agent_11
1.
00
0
Agent_
30
1.
00
0
Agent_12
1.
00
29.
00
Agent_31
1.
00
0
Agent_13
2.
00
0
Agent_
32
1.
00
0
Agent_14
1.
00
2.
00
Agent_
33
1.
00
0
Agent_15
1.
00
0
Agent_
34
1.
00
0
Agent_16
1.
00
0
Agent_
35
1.
00
0
Agent_17
1.
00
0
Agent_
36
1.
00
0
Agent_18
1.
00
0
Agent_
37
1.
00
0
Agent_19
1.
00
0
Agent_
38
1.
00
0
Agent_
39
1.
00
0
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
The
An
al
ysi
s
o
f
St
u
d
ent
C
o
l
l
a
bor
at
i
ve W
o
rk
Insi
de
Soci
a
l
L
e
ar
ni
n
g
...
. (
A
ndi
Besse Fi
r
d
ausi
a
h M
a
n
s
ur
)
2
491
Tab
l
e
3
sho
w
s th
e cl
o
s
en
ess
an
d between
n
e
ss cen
t
r
ality.
W
i
t
h
in a
g
r
aph
,
t
h
e
b
e
tweenn
ess is the
cen
trality
m
e
a
s
u
r
e
o
f
a
v
e
rtex
.
Vertices hav
e
h
i
g
h
b
e
tween
n
e
ss v
a
l
u
e
wh
en
th
ey
o
c
cu
r
o
n
m
a
n
y
sho
r
test
pat
h
s
.
Fo
r t
h
e
gra
p
h
t
h
at
has vert
i
ces
t
h
en t
h
e bet
w
een
ness
for
vert
e
x
can be
m
easured
u
s
i
n
g se
veral
st
eps
bel
o
w:
1
.
Fo
r each
p
a
ir
of
, calcu
late all sho
r
test
p
a
th
s
b
e
tween
th
em
.
2.
For
eac
h
pai
r
o
f
, det
e
rm
i
n
e t
h
e fract
i
o
n
o
f
s
h
ort
e
st
pat
h
s
t
h
a
t
pass t
h
r
o
ug
h t
h
e
vert
e
x
i
n
.
3.
Sum
t
h
i
s
fract
i
o
n
o
v
e
r
al
l
pai
r
s o
f
, refe
r t
o
E
quat
i
o
n
2.
(2
)
whe
r
e
i
s
t
h
e num
ber o
f
sh
ort
e
st
pat
h
s f
r
o
m
and
i
s
t
h
e num
ber o
f
sh
ort
e
st
pat
h
s f
r
o
m
that go
by
.
Furt
herm
ore, t
h
e close
n
ess
for a ve
rtex
is the
reciprocal of the
sum
of
geodesic dista
n
ces
to
all o
t
h
e
r
v
e
r
tices of
V(
Sab
i
du
ssi,
19
66
). Instead
of
u
s
ing
geodesic
distance t
o
m
easure close
n
ess,
Noh a
nd
Rieg
er (200
3) p
r
op
o
s
ed
th
e rando
m
-
walk
cen
trality
th
at
is a
measu
r
e of th
e sp
eed
with
wh
ich
rando
m
l
y
walking m
e
ssages
reach a
ve
rtex
from
elsewhe
r
e i
n
the
network a s
o
rt
of ra
ndom
-walk
versi
o
n
of cl
ose
n
es
s
cen
trality (Noh
and
Rieg
er,
2
004
). In
essen
ce, it
m
easu
r
es th
e h
a
rm
o
n
i
c
m
ean
len
g
t
h o
f
p
a
t
h
s end
i
n
g
at
a
v
e
rtex
.
is small wh
en
th
ere
are m
a
n
y
sh
ort p
a
th
s
conn
ectin
g
it to o
t
h
e
r
vertices (Steph
en
son
and
Zelen
,
1
989
).
h
e
eq
u
a
tio
n
o
f
clo
s
en
ess h
a
s been imp
r
ov
ed to ov
erco
m
e
n
e
two
r
k vu
ln
erab
ility. Th
is con
d
ition
is
usef
ul
a
n
d
m
a
kes cl
ose
n
ess c
o
m
put
at
i
on f
o
r
di
sco
n
n
ect
ed
g
r
ap
h
3
bec
o
m
e
easy
.
(3
)
Tabl
e 3
pre
s
ent
s
t
h
e fi
n
d
i
ngs
of t
h
e be
t
w
een
ness a
n
d the closenes
s
calculation based on the
adjace
ncy
m
a
tri
x
i
n
Tabl
e
1.
Ag
ent
_1
(St
u
dent
_1
) a
n
d
A
g
ent
_
1
2
(St
u
d
e
nt
_
1
2
)
have
t
h
e
hi
g
h
est
cl
o
s
enes
s
values
whic
h
mean that these stude
nt
s are
particula
r
ly important beca
us
e
their distanc
e
s are closest to ot
her
stu
d
e
n
t
s
in
th
e n
e
two
r
k
.
In
o
t
h
e
r word
s,
th
e clo
s
est
stud
en
t
d
i
stan
ce, t
h
e fastest in
form
at
i
o
n
will b
e
d
e
li
v
e
red.
Fu
rt
h
e
rm
o
r
e, betweenn
e
ss also
sho
w
a co
mp
ellin
g resu
lt
wh
ere
Ag
en
t
_
1
2
and Ag
en
t_2
1
b
e
co
m
e
th
e m
a
in
act
or
hu
b
w
ho
con
n
ect
s
o
m
e
no
des t
o
ot
her
no
des
.
T
h
e r
o
l
e
of act
or
h
u
b
i
s
cruci
a
l
f
o
r t
h
e i
n
fo
rm
at
i
on fl
o
w
i
n
si
de
t
h
e net
w
or
k.
Tab
l
e
3
.
Cl
o
s
en
ess and
Between
n
e
ss Cen
t
rality
measu
r
e
for
W
i
k
i
d
a
ta
Na
m
e
Closeness
Betwenness
Na
m
e
Closeness
Betwenness
Agent_1
0.
011
5.
000
Agent_20
0.
000
0.
000
Agent_2
0.
00
0.
000
Agent_21
0.
001
12.
000
Agent_3
0.
00
0.
000
Agent_22
0.
000
0.
000
Agent_4
0.
000
0.
000
Agent_23
0.
000
0.
000
Agent_5
0.
000
0.
000
Agent_24
0.
000
0.
000
Agent_6
0.
000
0.
000
Agent_25
0.
000
0.
000
Agent_7
0.
000
0.
000
Agent_26
0.
000
0.
000
Agent_8
0.
000
0.
000
Agent_
27
0.
000
0.
000
Agent_9
0.
000
0.
000
Agent_
28
0.
000
0.
000
Agent_10
0.
000
0.
000
Agent_
29
0.
000
0.
000
Agent_11
0.
000
0.
000
Agent_
30
0.
000
0.
000
Agent_12
0.
016
31.
000
Agent_31
0.
000
0.
000
Agent_13
0.
000
0.
000
Agent_
32
0.
000
0.
000
Agent_14
0.
001
1.
500
Agent_
33
0.
000
0.
000
Agent_15
0.
000
0.
000
Agent_
34
0.
000
0.
000
Agent_16
0.
000
0.
000
Agent_
35
0.
000
0.
000
Agent_17
0.
000
0.
000
Agent_
36
0.
000
0.
000
Agent_18
0.
000
0.
000
Agent_
37
0.
000
0.
000
Agent_19
0.
000
0.
000
Agent_
38
0.
000
0.
000
Agent_
39
0.
000
0.
000
:(
,
)
GV
E
n
()
B
Cv
v
(,
)
ver
t
i
ces
s
t
(,
)
ver
t
i
c
e
s
s
t
(,
)
q
u
es
t
i
o
n
h
er
e
ver
t
e
x
v
(,
)
v
e
r
t
i
ces
s
t
()
()
st
B
sv
t
V
st
v
Cv
s
t
to
s t
()
st
v
to
s t
()
ver
t
e
x
v
C
Cv
v
ii
(,
)
\
()
2
G
dv
t
C
tV
v
Cv
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
I
J
ECE
Vo
l. 6
,
N
o
. 5
,
O
c
tob
e
r
20
16
:
248
8
–
24
98
2
492
In add
itio
n, this sectio
n in
ten
d
s
to rev
eal
m
o
re
th
e b
e
h
a
v
i
ou
r of co
llabo
rativ
e work of stud
en
t
b
y
measu
r
ing
th
e
p
opu
larity o
f
stu
d
e
n
t
using
ei
g
e
nv
ector
cen
t
rality. Eig
e
nv
ecto
r cen
t
rality is ab
le to
m
easu
r
e
"
Ho
w
well is th
is p
e
rson
con
n
ected
t
o
o
t
her well con
n
e
cted
peop
le
", so
t
h
e hi
g
h
est
ei
gen
v
ect
or c
a
n be
con
s
i
d
ere
d
as
t
h
e at
t
r
act
i
v
e
pers
o
n
i
n
si
de t
h
e net
w
o
r
k(C
h
el
i
o
t
i
s
, 20
0
6
)
.
Sym
m
et
ri
ze i
s
a m
e
t
hod t
o
c
h
an
ge
"d
irected
"
or "asy
mm
e
t
ric" n
e
twork
d
a
ta in
t
o
"und
irect
ed"
or "sy
m
m
e
t
r
i
c
" dat
a
. T
h
ere a
r
e vari
ous
m
e
t
h
ods
t
o
sy
mm
e
t
rize d
a
ta. In
th
is stud
y, all sy
mmet
r
ize d
a
ta are to co
v
e
r th
e ev
al
u
a
tio
n
o
f
co
llab
o
rativ
e wo
rk
in
sid
e
Mo
od
le
W
i
k
i
.
Th
e d
a
ta is ob
tain
ed
fro
m
t
h
e
W
i
k
i
in
teractio
n
presen
ted
in
Tab
l
e
1
.
Th
e m
easu
r
emen
t o
f
eig
e
nv
ector cen
t
rality can
b
e
referred
to th
e
Equ
a
tio
n 4.
0
(4
)
whe
r
e
A is a
d
jacency m
a
trix from
graph,
λ
i
s
ei
ge
n
val
u
e a
n
d
v
i
s
ei
gen vect
or.
The sy
m
m
et
ri
ze m
e
t
hod
i
s
de
scri
be
d as
f
o
l
l
o
w:
M
a
x
i
mu
m
If there a
r
e two actors A a
nd
B, then
th
e stro
ng
est tie a
m
o
n
g
th
em
is ch
o
s
en
to
b
e
rep
r
esen
tativ
e o
f
a
tie f
o
r
A
and
B. Th
e
d
a
ta o
f
wik
i
th
at h
a
s b
e
en
adju
sted
in
t
o
m
a
tr
ix
ad
j
acen
c
y on
Tab
l
e1
is an
alyzed
th
ro
ugh
Eig
e
nv
ector cen
t
rality u
s
in
g sy
mmetrize
max
i
m
u
m
.
Th
e
d
i
stribu
tio
n
o
f
Ei
g
e
nv
ector cen
trality sco
r
es i
s
prese
n
t
e
d
i
n
T
a
bl
e 4
wi
t
h
m
e
an
0.
10
2.
Tab
l
e 4
.
Distri
b
u
tion
score for
Eig
e
nv
ector
cen
t
rality-MAX
Measures
Value
M
ean
0.
102
Std.
Dev. 0.
123
M
i
n. 0.
017
M
a
x. 0.
624
Th
e relation
s
hip
b
e
tween
Ag
en
t throug
h
Eig
e
nv
ect
or cen
t
rality was illu
strated
clearly th
rou
gh
C
once
n
t
r
i
c
Di
a
g
ram
as sho
w
n
i
n
Fi
g
u
re
1.
A
g
ent
_1
be
c
o
m
e
s
centre of other
Agent
.
This
means that Agent_1
can be co
nsi
d
e
r
ed as wel
l
-
c
o
nnect
e
d
pe
opl
e
i
n
t
h
e real
wo
rl
d. T
h
i
s
ki
n
d
of pe
rs
on i
s
pa
rt
i
c
ul
arl
y
essen
t
i
a
l
t
o
make sure that
the colla
borativ
e
work
become a success
or
not.
Fig
u
re 1
.
Con
c
en
tric d
i
agram
for
Ei
g
e
nv
ector cen
t
rality with
Max
i
m
u
m
S
y
mmetrize
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
The
An
al
ysi
s
o
f
St
u
d
ent
C
o
l
l
a
bor
at
i
ve W
o
rk
Insi
de
Soci
a
l
L
e
ar
ni
n
g
...
. (
A
ndi
Besse Fi
r
d
ausi
a
h M
a
n
s
ur
)
2
493
M
i
n
i
mu
m
Exem
pl
i
f
i
e
s t
h
e po
wer
of t
h
e sym
m
et
ri
c t
i
e
am
ong
A an
d
B
as bei
ng t
h
e feebl
e
o
f
t
h
e t
i
e
s AB
or B
A
o
r
g
e
n
e
rally is
k
nown as weak
est lin
k.
T
h
e s
econd e
v
aluati
on of the c
o
lla
borative
work i
n
side M
o
odle
wiki is
th
ro
ugh
Eig
e
nv
ector with
Min
i
m
u
m
Sy
mme
trize. The d
i
strib
u
tion o
f
Eig
e
n
v
ect
o
r
cen
t
rality
sco
r
e is
descri
bed
i
n
T
a
bl
e 5
wi
t
h
m
e
an
0.
03
6.
Tab
l
e
5
.
Score
for Ei
g
e
nv
ector cen
t
rality v
e
cto
r
with
Min
i
mu
m
Sy
mmetriz
e
Measures
Value
M
ean
0.
036
Std.
Dev. 0.
156
M
i
n. 0
M
a
x. 0.
814
Th
is Eigenv
ect
o
r
con
cen
t
r
ic is d
e
scrib
e
d
clearly in
Figure
2
.
In th
is
figu
re, Ag
en
t_1
2
beco
m
e
s th
e
center
of the
ot
hers
age
n
t a
n
d
acts as the m
o
s
t
well conne
cted
people.
Fig
u
re
2
.
Con
c
en
tric
d
i
agram
for Ei
g
e
nv
ector cen
t
rality with
Min
i
m
u
m
Symme
trize
A
v
er
ag
e
Dem
onst
r
at
es
t
h
e
po
wer
o
f
t
h
e sy
m
m
e
t
r
i
c
t
i
e
am
ong
A a
n
d
B
as t
h
e
pl
ai
n ave
r
a
g
e
of
t
h
e t
i
e
s
A
B
an
d BA.
Th
e
th
ird ev
al
u
a
tion
o
f
t
h
e co
llab
o
rativ
e
wo
rk
in
sid
e
M
o
od
le w
i
k
i
is t
h
roug
h Eig
e
nv
ector
w
ith
Av
erag
e
Symmetrize.
Th
e distrib
u
tion
o
f
Eig
e
nv
ector
ce
n
t
rality sco
r
e i
s
d
e
scrib
e
d
i
n
Tab
l
e
6 with m
ean
0.
10
3.
Tab
l
e 6
.
Score for
Ei
g
e
nv
ector
cen
t
rality
v
e
cto
r
with
Av
erag
e
Symm
e
t
rize
Measures
Value
M
ean
0.
103
Std.
Dev. 0.
123
M
i
n. 0.
016
M
a
x. 0.
617
Figure 3 prese
n
ts
the positions of
the
a
g
e
n
t
am
ong ot
her a
g
ents
.
Age
n
t_1 is locate
d
at
the cent
r
e
of
the conce
n
tric
diagram
,
and
Age
n
t_3 is
placed in t
h
e
third layer. Mea
n
while,
Ag
ent
_
7 is positioned in the
fo
urt
h
lay
e
r.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
I
J
ECE
Vo
l. 6
,
N
o
. 5
,
O
c
tob
e
r
20
16
:
248
8
–
24
98
2
494
Fig
u
re
3
.
Con
c
en
tric
d
i
agram
o
f
th
e Ei
g
e
nv
ecto
r cen
t
rality with
Av
erag
e
Sy
mm
e
t
rize
Lo
wer Half
Utilizes o
n
e
h
a
lf v
a
l
u
es
o
f
th
e m
a
trix
fro
m
th
e
o
t
h
e
r h
a
lf.
If
th
e
"send
e
rs"
b
e
co
m
e
th
e
fo
cu
s of
th
e
n
e
two
r
k
p
r
op
erties, th
en
it will set
t
h
e lo
wer h
a
lf
eq
u
a
ls to
th
e
u
p
p
e
r h
a
lf.
Howev
e
r, if th
e fo
cus o
f
the
net
w
or
k ha
s c
h
an
ge
d t
o
"rec
ei
ver", t
h
en t
h
e up
per
hal
f
e
qual
s
t
o
t
h
e l
o
wer
hal
f
. T
h
e
fo
urt
h
sy
m
m
e
tri
ze o
f
Eig
e
nv
ector cen
t
rality is th
e Lo
wer h
a
l
f
. Tab
l
e 7
d
i
splays th
e
m
eas
u
r
em
en
t v
a
lu
es for th
e Eig
e
n
v
ect
o
r
d
i
stribu
tio
n and
Fi
g
u
re
4
illu
strates th
e ag
en
t b
e
h
a
v
i
o
u
r i
n
sid
e
th
e con
c
en
tric d
i
agram
.
Tab
l
e
7
.
Score
for Ei
g
e
nv
ector cen
t
rality with
Lower
h
a
lf Sy
mmetrize
Measures
Value
M
ean
0.
067
Std.
Dev. 0.
145
M
i
n. 0
M
a
x. 0.
623
Fig
u
re
4
.
Con
c
en
tric
d
i
agram
for Ei
g
e
nv
ector cen
t
rality with
Lower
h
a
lf Sy
mmetrize
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
The
An
al
ysi
s
o
f
St
u
d
ent
C
o
l
l
a
bor
at
i
ve W
o
rk
Insi
de
Soci
a
l
L
e
ar
ni
n
g
...
. (
A
ndi
Besse Fi
r
d
ausi
a
h M
a
n
s
ur
)
2
495
Up
pe
r H
a
lf
Ev
alu
a
tes th
e
valu
es in
un
it AB an
d
BA, and retu
rn
s a
v
a
lue b
a
sed
on
th
e
test fu
n
c
tion
.
Fo
r ex
am
p
l
e,
i
f
Up
per > Lo
wer an
d AB
= 2, B
A
= 6, t
h
e
n
t
h
e appl
i
cat
i
on
wo
ul
d sel
e
c
t
val
u
e "6," si
n
ce up
per val
u
e
(AB
)
was no
t larg
er th
an
th
e lo
wer v
a
lu
e (B
A). Th
e last sy
mmetrize th
at will
b
e
u
s
ed
in
th
i
s
ev
alu
a
tion
is u
p
p
e
r
hal
f
. T
h
e res
u
l
t
of Ei
gen
v
ect
or u
s
i
n
g t
h
i
s
m
ode i
s
pre
s
ent
e
d i
n
Tabl
e 8. T
h
e det
a
i
l
s
of t
h
e i
n
t
e
ract
i
on a
m
ong
agent
s
are
p
o
rt
ray
e
d i
n
Fi
gu
re
5 a
n
d
6.
Tab
l
e 8
.
Score for
Ei
g
e
nv
ector
cen
t
rality
with
Upp
e
r h
a
lf
Sy
mmetrize
Measures
Value
M
ean
0.
085
Std.
Dev. 0.
136
M
i
n. 0.
004
M
a
x. 0.
701
Fi
gu
re
5
p
o
rt
r
a
y
s
t
h
e l
a
y
e
re
d
di
ag
ram
i
n
whi
c
h
Age
n
t
_
1 i
s
at
t
h
e ce
nt
re,
f
o
l
l
o
wed
by
A
g
ent
_
7
,
Age
n
t4, Age
n
t
_
3, Age
n
t_5,
Age
n
t_6 a
n
d s
o
on. The po
si
tion
of t
h
e agent is determ
ined accordi
n
g to the
Eigenvector va
lue.
Fig
u
re
5
.
Con
c
en
tric
d
i
agram
for Ei
g
e
nv
ector cen
t
rality with
Upp
e
r
h
a
lf Sy
mmetrize
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
I
J
ECE
Vo
l. 6
,
N
o
. 5
,
O
c
tob
e
r
20
16
:
248
8
–
24
98
2
496
Fig
u
re 6
.
Sp
ring
d
i
agram
fo
r Eig
e
nv
ector
cen
t
rality
with
Up
p
e
r h
a
lf Symmetrize
Fi
gu
re 6 s
h
ow
s t
h
e spri
ng m
odel
o
f
Ei
ge
n
v
ect
or val
u
e t
h
r
o
u
g
h
U
ppe
r sy
m
m
e
t
r
i
ze. Fro
m
Fi
gure
6
,
Ag
en
t_1
is seen
as th
e m
o
st well co
nn
ect
ed
nod
e th
at ab
le to
m
a
in
tai
n
relation
s
h
i
p
s
with
Ag
en
t_12
and
Ag
en
t_21
. M
o
reov
er, th
e seco
nd
b
i
gg
est Ei
g
e
nv
ector
v
a
lue in
th
is symmetrize
m
o
d
e
is
Ag
en
t_7
.
Th
is
is d
u
e
to
th
e
fact th
at
it ab
les to
con
n
ect with
Ag
en
t
_
1 and
Ag
en
t
_
1
2
.
3.
R
E
SU
LTS AN
D CONC
LUSION
This section s
u
mmarizes
the analyzes result
of
the evaluations c
o
nducted
o
n
th
e co
llab
o
rativ
e work
in
sid
e
Mood
le
wik
i
. Th
e m
easu
r
em
en
t app
r
o
ach
es b
e
i
n
g
co
n
s
i
d
ered
are th
e Degree Centrality, Between
n
e
ss,
Clo
s
en
ess, and Eig
e
nv
ector cen
t
rality. Th
e resu
lt and
an
alysis o
f
th
ese ap
pro
ach
es are
p
r
esen
ted
i
n
Tab
l
e 9
.
In
t
h
is tab
l
e, th
e sym
b
o
l
"Agen
t
" is u
s
ed
t
o
represen
t th
e stu
d
e
n
t
's n
a
m
e
.
Th
e
d
a
ta fo
r the Deg
r
ee cen
t
rality,
C
l
oseness
,
a
n
d
B
e
t
w
een
ness
are
obt
ai
ne
d
fr
om
Tabl
e 1 a
n
d
2.
Table
9. T
h
e
re
sult and a
n
alysis of the a
p
proaches
us
ed
to ev
alu
a
te t
h
e co
llab
o
rativ
e
wo
rk in
W
i
k
i
M
o
odle
Eigenvector Centrality
Degree Ce
ntrality Closeness
Betweenness
In the Eigenvector centralit
y,
Agent_1 and
Agent_12 ar
e
r
ecognized as the
m
o
st
well
connected to other
well
connected node ins
i
de networ
k
Agent_12 lead
ing the
outdegr
ee values
by
29 an
d
followed
by
Agent_1 and
Agent_21.
Agent_12 o
b
tains t
h
e highest
score f
o
r closeness with
0.
016 a
nd the
n
p
u
r
s
ued by
Agent_1.
Agent_12 receives the highest
scor
e by
31.
Nex
t
is Agent_21
with the scor
e 15 and finally
,
Agent_1 with the scor
e 5
Table 9 illustrates the res
u
lts from
differe
nt SNA m
e
thod to a
n
aly
ze the interaction
inside
W
i
ki
M
o
o
d
l
e
. F
r
o
m
overal
l
o
b
se
rvat
i
o
n,
A
g
ent
_1 a
n
d A
g
e
n
t
_
12
ha
ve pl
ay
e
d
a vi
t
a
l
r
o
l
e
f
o
r t
h
e net
w
o
r
k
.
B
a
se
d
on t
h
eir Ei
genvector val
u
e, t
h
ese two a
g
ents are the
m
o
st well connecte
d
node. T
h
is
mean
s th
at th
ese two
st
ude
nt
s a
r
e
p
o
p
u
l
a
r
an
d a
r
e
abl
e
t
o
m
a
nage t
h
e
c
o
l
l
a
bo
rat
i
v
e
wo
r
k
.
F
u
rt
herm
ore,
A
g
ent
_
1
a
n
d
A
g
ent
_
1
2
also
ob
tain
h
i
gh
ou
td
eg
ree v
a
lu
e, wh
ich
m
ean
th
at
these t
w
o st
ude
nts are acting as the
source
provi
der who
sup
p
l
y
i
n
f
o
rm
at
i
on an
d kn
o
w
l
e
d
g
e t
h
r
o
u
g
h
t
h
e net
w
or
k.
These resul
t
s
al
so st
rengt
he
ned
by
t
h
e val
u
e o
f
closenes
s and betwee
nne
ss
whe
r
e Age
n
t_1 and
Age
n
t_
12 als
o
leadi
n
g on this m
e
a
s
urem
ent. The
high
cl
osenes
s val
u
e of A
g
e
n
t
_
1 a
nd
Ag
ent
_
1
2
l
ead i
n
t
o
fast
sprea
d
i
n
g i
n
fo
r
m
at
i
on si
nce t
h
ey
ha
ve fast
e
s
t
ro
ut
e
and
have t
h
e
m
o
st di
rect
ro
ut
e t
o
t
h
e ot
he
r no
de i
n
si
de t
h
e net
w
o
r
k
.
The anal
y
s
i
s
res
u
l
t
of t
h
e Ei
ge
nvect
or
cen
trality h
a
s
relatio
n
s
with
th
e so
cial learn
i
ng
n
e
t
w
o
r
k
an
alysis for iden
tifyin
g
co
llab
o
rativ
e
work
in
sid
e
Mo
od
le W
i
k
i
. Accord
ing
to
Ch
elio
tis,
G. (2
010
),
"Ho
w
well is th
is p
e
rso
n
conn
ected
to
o
t
h
e
r
well co
nn
ected
p
e
op
le
?
"
is th
e fund
am
en
tal
o
f
Ei
g
e
nv
ect
o
r
cen
trality
m
e
asu
r
em
en
t.
W
i
th
th
is prin
ci
p
l
e, th
e analysis resu
lt
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
The
An
al
ysi
s
o
f
St
u
d
ent
C
o
l
l
a
bor
at
i
ve W
o
rk
Insi
de
Soci
a
l
L
e
ar
ni
n
g
...
. (
A
ndi
Besse Fi
r
d
ausi
a
h M
a
n
s
ur
)
2
497
also recognized that Age
n
t_1 and Ag
ent
_
1
2
becom
e
t
h
e k
e
y
poi
nt
o
n
i
n
f
o
rm
at
i
on di
st
ri
but
i
o
n si
nce t
h
ey
are
th
e m
o
st well
co
nn
ected
nod
e. Fi
n
a
lly, th
e resu
lts
fro
m
th
e Deg
r
ee cen
t
rality, Clo
s
en
ess, Between
n
e
ss,
Hierarc
h
ical a
n
d MST cl
ustering al
so supp
or
t th
e
pr
in
cip
l
e, in
wh
ich w
e
can
clai
m
th
at A
g
en
t
_
1 and
Age
n
t
_
1
2
a
r
e t
h
e ce
nt
ral
no
de
s i
n
si
de
t
h
e
net
w
o
r
k
t
h
at
h
o
l
d
a vi
t
a
l
r
o
l
e
f
o
r
i
n
f
o
rm
at
i
on ci
r
c
ul
at
i
on
am
ong t
h
e
no
des
.
ACKNOWLE
DGE
M
ENTS
A
u
t
h
or
s ar
e
g
r
atef
u
l
to Facu
lty o
f
C
o
m
p
u
ting
an
d In
for
m
at
io
n
Techno
logy Rab
i
gh
,
K
i
ng
A
bdu
laziz
U
n
i
v
er
sity,
K
i
n
gdo
m
o
f
Saudi A
r
ab
ia
REFERE
NC
ES
[1]
J.
Ji,
et al.
, “Maximizing th
e Community
Cov
e
rage of
Influen
c
e through a Social Network
,
”
AISS: Advances in
Information Scien
ces and Service Sciences
, vol/issue: 3(9), pp. 339~346, 2011.
http://dx.doi.org/10.4156/
aiss.vol3.issue9.44
.
[2]
B. Zhou
and
C.
W
u
, “
S
em
antic
M
odel
for Social Networ
king
Federation,”
Aiss: Advances in
Info
rmation Sciences
and Service Sciences
,” vol/issue:
3(11), pp
. 213~2
23, 2011
.
[3]
Erlin B.
,
et al.
, “Analy
zing On
line As
y
n
chro
no
us Discussion
Usi
ng Content and Social Network Analy
s
is,” in
IEEE Nin
t
h International
Conferenc
e
On In
tell
igent
Systems Design And Applicat
ions
, 20
09.
http://dx.doi.org/
10.1109/isda.20
09.40.
[4]
F. Ortiza
and
R. Fraile, “Social Network Featur
ing
Enter
t
ainment, Cultur
e
And Technolo
g
y
,” in
Span
ish
Univer
s
iti
es
: T
h
e Infocampus
Pr
oject T
h
e
Open Infor
m
ation Sys
t
ems
Jour
nal
, vol. 3
,
pp. 48-53, 200
9.
http://dx.doi.org/10.2174/187413
3900903010048.
[5]
S. J. Kepp and H. Scho
rr, “Analy
zing Collaborative Learn
i
ng Activities
in Wikis Using Social Network Analy
s
is,”
in
Chi, Spotlight on Work in
Progress~Session
2, April 4-9, Bosto
n
, Ma
, Usa.
AC
M,
2009
.
[6]
H. L. Twu, “
A
Predictiv
e Stud
y of W
i
ki Interaction: Ca
n Attitu
de toward W
i
ki Predict W
i
ki Int
e
rac
tion in High-
Context Cultur
e
s Groups?
”
Journ
a
l of Educationa
l Technology Developmen
t and Exchange
, vol/iss
u
e: 3(1), pp. 57-
68, 2010
.
[7]
N.
T.
Korfiatis
,
et al.
, “Ev
a
luating Authoritativ
e Sources Using So
cial Networks: An Insight from Wikipedia,”
Online In
for
m
ati
on Re
view
, vo
l/issue: 30(3), pp.
252 – 262
, 2006
.
[8]
S. Wasserman and K. Fa
ust, “Social
Network Analy
s
is,”
Cambridge University Press, 1994.
http://dx.doi.org/10.1017/CBO9780511815478.
[9]
M
.
F
.
D
e
laat,
et al.
, “Online teaching in networke
d learning com
m
unities: A m
u
lti-m
e
thod approach to study
ing the
role of
th
e t
e
a
c
h
e
r,”
Int
e
rnationa
l Journal
of
Web
-
Based Communities
, 2006
. Doi:10.1007/s11251-
006-9007-0.
[10]
C. S
p
adav
ecch
ia
and C.
Giovann
ell
a
, “
M
onitorin
g
Lea
r
ning Ex
pe
rienc
e
s
and S
t
yl
es
: The
S
o
cio-E
m
otional L
e
ve
l
,
”
in
10th I
EEE In
ternat
ional Conf
erence
on
Advanc
ed Le
arning Techn
o
logies
, 201
0.
http://dx.doi.org/
10.1109/icalt.20
10.129.
[11]
G. Cheliotis, “Social Network A
n
aly
s
is (SNA): Including
a
Tutor
i
al on
C
oncepts and
Methods,” Communications
and New Med
i
a
,
Nation
a
l
Univer
sit
y
of Sing
apor
e, 2006
. [A
ccess
e
d 27 Janu
ar
y
20
13]
.
[12]
P
.
Drazdi
l
ova,
et al.
, “Analy
sis
of Relations in
E-Learning
,”
in
IEEE/WIC/ACM
Internationa
l C
onference on W
e
b
Intelligen
ce and
Intell
igen
t Ag
ent Technolog
y
, 20
08. Doi: 10.1109
/WIIAT.2008.25
0.
[13]
Erlin B.
,
et al.
, “Integrating Content Analy
s
is and Social
Ne
twork Analy
s
is for Analy
z
ing Asy
n
chrono
us
Disc
ussion Forum,
”
in
IEEE,
20
08. http://dx.doi
.org/10.1109/itsim.2008.4631996.
[14]
I. Ham
u
lic and
N. Bijed
i
c, “
S
oc
ial Network An
al
y
s
is in
Virtual Learn
i
ng Community
at Faculty
of Informatio
n
Techno
logies (F
it), Mostar,”
Elsevier-Procedia
Social and Behavioral Scien
c
es
, vol. 1
,
pp.
2269–2273, 200
9.
http://dx.doi.org/10.1016/
j.sbspro.2009.01
.399.
[15]
A. Martin
ez
,
et al.
, “Combining Qualitativ
e Ev
aluati
on And Social Network An
aly
s
is For Th
e Stu
d
y
Of C
l
assroom
S
o
cial In
ter
a
c
tions
,”
Comp
uter And
Education
,
Elsevier, vol. 41, pp
. 353-368,
2003.
http://dx.doi.org/10.1016/
j.compedu.2003.06
.001.
[16]
J. D. Noh
and H
.
Rieg
er,
Phys
. R
ev.
L
e
t
t
., pp
. 92
,
2004.
[17]
G. Sabidussi, “The Centra
lity
Index of
a Graph,”
Ps
ychometr
i
ka
, vol. 31, pp. 581-603, 1966.
http://dx.doi.org/10.1007/BF02289527.
[18]
K. A. Stephenso
n
and M.
Zelen
,
“Rethinking
Cen
t
rality
: Methods and Examples
,”
Social N
e
tworks
, vol. 11, 1989
.
http://dx.doi.org/10.1016/0378-8
733(89)90016-6.
[19]
U.
Asgher,
et a
l
., “Analy
sis and
Modeling of A
cademia'
s
Colla
borativ
e Decisio
n
Support S
y
stem Based on Key
P
e
rform
ance Ind
i
ca
tors
and Degr
ee of C
e
rta
i
nt
y,
”
the 6th
International Conferen
ce on Applied Hu
man Factors and
Ergonomics (
A
HFE 2015)
and
th
e A
ffiliated Conferences, AHFE
,
vol. 3
,
pp
. 4084-
4089, 2015
.
[20]
G.
Han,
et a
l
.
,
“
U
s
i
ng S
o
cial Ne
twork Anal
y
s
is
t
o
M
eas
ure S
t
ud
ent Collabor
atio
n in an Und
e
rgr
a
duate Capston
e
Course,”
North
American Co
lleg
e
s and Teachers
of Agriculture (
N
ACTA)
, vol/issue: 60(2)
, 2016
.
Evaluation Warning : The document was created with Spire.PDF for Python.