Int
ern
at
i
onal
Journ
al of Ele
ctrical
an
d
Co
mput
er
En
gin
eeri
ng
(IJ
E
C
E)
Vo
l.
10
,
No.
5
,
Octo
be
r
2020
,
pp.
4569
~
4580
IS
S
N: 20
88
-
8708
,
DOI: 10
.11
591/
ijece
.
v10
i
5
.
pp
4569
-
45
80
4569
Journ
al h
om
e
page
:
http:
//
ij
ece.i
aesc
or
e.c
om/i
nd
ex
.ph
p/IJ
ECE
Data sci
ence for
d
igital c
ultu
re
i
mp
rovement in
h
igher
educatio
n usin
g
K
-
m
eans
clusteri
ng and tex
t anal
ytics
Dian S
a’ad
il
la
h Ma
yla
w
ati
1
,
Tedi
Pria
tna
2
, Hamd
an Su
gi
lar
3
, Muh
am
mad
Ali
Ra
m
dhani
4
1
Depa
rtment of I
nform
at
ic
s,
UIN
Sunan
Gunung
Djat
i
Bandung, I
ndonesia
1
Facul
t
y
of
Infor
m
at
ion
and
Com
m
unic
at
ion
Tech
nolog
y
,
Univ
ersi
ti
Te
knik
al Mal
a
y
sia
Mel
aka,
Ma
lay
s
ia
2
Depa
rtment of I
slamic
Edu
catio
n,
UIN
Sunan
G
unung
Djati
Ban
dung,
Indon
esia
3
Depa
rtment of
Mathe
m
at
i
cs
Ed
uca
t
ion, UIN Sunan
Gunung Dj
a
ti
B
andung, I
ndo
nesia
4
Depa
rtment of I
nform
at
ic
s,
UIN
Sunan
Gunung
Djat
i
Bandung, I
ndonesia
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
A
pr
1,
2020
Re
vised
A
pr 13
,
2020
Accepte
d
Apr 23
, 202
0
Thi
s
stud
y
ai
m
s
to
inve
stig
at
e
the
m
ea
n
ingfu
l
patter
n
tha
t
c
an
be
used
to
improve
di
git
al
culture
i
n
highe
r
educat
ion
b
ase
d
on
par
amet
ers
of
the
te
chno
lo
g
y
a
cc
ep
ta
n
ce
m
odel
(TAM).
The
m
et
hodolo
g
y
used
is
the
da
ta
m
ini
n
g
te
chn
ique
wi
th
K
-
m
ea
ns
al
g
orit
hm
and
te
x
t
ana
l
y
t
ic
s.
The
expe
rimen
t
using
qu
estionnaire
data
with
2887
r
esponde
nts
i
n
Univer
sita
s
Isla
m
Nege
ri
(UIN
)
Sunan
Gunun
g
Djat
i
Bandun
g.
The
data
ana
l
y
sis
and
clus
te
ring
resul
t
show
tha
t
th
e
per
ceive
d
usefu
lne
ss
and
beha
vior
al
intent
ion
t
o
use
info
r
m
at
ion
s
y
st
ems
are
above
the
no
rm
al
val
u
e,
while
th
e
per
c
ei
ved
ea
se
of
use
and
actua
l
s
y
stem
use
is
quit
e
low.
Strengt
hen
ed
wi
th
te
xt
anal
y
tic
s,
thi
s
rese
ar
ch
found
tha
t
th
e
EDA
and
K
-
m
ea
ns
result
in
har
m
on
y
wit
h
the
hope
or
desire
of
a
ca
d
e
m
ic
soci
e
t
y
the
informat
ion
s
y
stem
imple
m
ent
at
ion
.
Thi
s
rese
arc
h
al
so
found
how
important
the
socializat
ion
and
guida
nc
e
of
informati
on
s
y
stems
,
espe
cia
l
l
y
the
new
one
informati
on
s
y
ste
m
,
in
orde
r
to
improve
digi
tal
cul
ture
in
highe
r educ
at
ion
.
Ke
yw
or
d
s
:
Cl
us
te
rin
g
Data sci
ence
Digital
cu
lt
ure
Higher
educat
ion
K
-
m
eans alg
ori
th
m
Text a
naly
ti
cs
Wor
d
cl
ou
d
Copyright
©
202
0
Instit
ut
e
o
f Ad
vanc
ed
Engi
n
ee
r
ing
and
S
cienc
e
.
Al
l
rights re
serv
ed
.
Corres
pond
in
g
Aut
h
or
:
Dian Sa’a
dill
ah
Ma
yl
awati
,
Dep
a
rtm
ent o
f
Infrom
at
ic
s,
Faculty
o
f
Scienc
e an
d
Tec
hnol
og
y,
UIN Su
nan Gu
nung
Dj
at
i B
a
ndun
g,
A.
H. Nas
utio
n St
reet,
105 B
andu
ng, In
done
sia
Ce
ntre fo
r Adv
anced C
om
pu
ti
ng Tec
hnology
, F
ac
ulty
o
f
Inform
at
ion
and
Com
m
un
ic
at
io
n
Tec
hnol
og
y,
Un
i
ver
sit
i Te
knikal M
al
ay
sia
Mel
a
ka,
Hang T
ua
h
Jay
a Street
, 7
6100
Dur
ia
n
T
ungg
al
, Mel
aka,
Ma
la
ysi
a
.
Em
a
il
:
diansm@ui
ns
gd.ac
.id
1.
INTROD
U
CTION
In
t
he
te
ch
no
l
og
y
disruptio
n
e
ra,
m
any
so
ftwar
e/
a
pp
li
cat
ion
s/
in
form
at
i
on
syst
em
s
bu
il
t
to
help
hu
m
an
act
ivit
ie
s.
Tra
gical
ly
,
37%
of
1,800
s
of
t
war
e
a
re
wa
ste
d,
a
nd
47%
of
t
hem
are
so
f
tware
in
the
fiel
d
of
edu
cat
io
n
[
1]
.
This
fact
occ
ur
s
beca
us
e
m
any
factors
s
uch
as
unf
ulfill
ed
use
r
requirem
ents;
there
a
re
s
oft
war
e
error
s
,
fa
ults,
and
fail
ur
es;
s
of
t
war
e
qu
al
it
y
is
no
t
fu
lfil
le
d;
no
inno
vation;
does
no
t
app
ly
the
co
nc
ept
of
hu
m
an
and
c
om
pu
te
r
interact
ion
pro
per
ly
;
diff
ic
ult
to
us
e
;
no
t
accor
ding
to
m
ark
et
need
s
(
no
t
up
-
to
-
date);
to
the
la
ck
of
unde
rstan
ding
of
the
us
e
of
t
echnolo
gy
due
to
it
s
rap
id
de
velo
pm
ent
so
that
tren
ds
ca
nnot
be
fo
ll
owe
d;
et
cet
era.
Of
c
ours
e,
this
is
an
obsta
cl
e
f
or
hi
gher
ed
ucati
on
to
ac
hie
ve
Te
chno
U
niversi
ty
[
2]
,
Digital
C
amp
us
[
3,
4]
,
Smart
Camp
us
[5
-
8]
,
Gre
en
C
amp
us
[9]
,
a
s
well
as
va
rio
us
oth
e
r
te
rm
s
in
the
era
of
dig
it
al
te
chnol
og
y
-
base
d
e
du
cat
ion
.
The
refor
e
,
no
m
at
te
r
how
s
ophisti
ca
te
d
the
te
c
hnol
og
y
is
offe
red,
w
hen
the
ap
plica
ti
on
program
is
not
us
ed
as
plan
ned,
it
will
no
t
hav
e
si
gnific
ant
i
m
plica
ti
on
s
for
hum
an
act
ivit
y.
On
e
of
the
m
ain
prob
le
m
s
of
the
fail
ure
of
i
m
ple
m
enting
di
gital
syst
e
m
s
i
s
that
they
a
re
no
t
rea
dy
to
ac
cept
te
chnolo
gical
c
hanges
so
quic
kly
that
the
us
e
of
te
ch
nolo
gy
is
not
culti
vate
d
a
nd
does
not
beco
m
e
a
nece
ssit
y
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
10
, No
.
5
,
Oct
ob
e
r 2
020
:
45
69
-
45
80
4570
to
suppo
rt
it
s
act
iv
it
ie
s.
As
a
resu
lt
,
m
any
app
li
cat
io
ns
are
no
t
use
d
pro
pe
rly
or
rely
on
so
m
e
peo
ple
to
us
e
them
.
Digital
culture
in
highe
r
ed
ucati
on
m
us
t
be
est
a
blishe
d
to
sup
port
ac
adem
ic
act
iviti
es
m
or
e
eff
ect
i
vely
and
e
ff
ic
ie
ntly
.
Howe
ver,
not
al
l
academ
ic
ian
s
in
higher
ed
ucati
on
awa
re
about
the
be
ne
fits
of
dig
it
al
li
te
racy
with
va
rio
us
in
flue
nce
facto
rs
,
su
c
h
as
la
ck
of
s
ocial
iz
at
ion
,
not
easy
to
us
e,
ina
de
qu
at
e
infr
ast
ru
ct
ur
e
,
and
s
o
on.
Seei
ng
va
riou
s
issues
relat
ed
to
t
he
a
wareness,
abili
ti
es,
an
d
cultu
re
of
the
aca
d
em
ic
com
m
un
it
y
in
us
in
g
dig
it
al
syst
e
m
s
in
hi
gh
e
r
e
duc
at
ion
,
a
com
prehensi
ve
a
nd
in
-
de
pth
st
ud
y
of
t
he
fact
ors
infl
uen
ci
ng
t
he
us
e
of
dig
it
al
syst
e
m
s
is
al
read
y
i
n
pr
ogress.
T
he
stu
dy
ca
n
rev
eal
th
e
us
a
bili
ty
of
softw
are,
bo
t
h
in
te
rm
s
of
so
ft
war
e/
a
ppli
ca
ti
on
s a
nd in
term
s o
f users
.
Currentl
y,
dat
a
sci
ence
is
a
popu
la
r
te
ch
nique
to
proc
ess
data
eff
ic
ie
ntly
[10]
,
wit
h
a
big
data
char
act
e
r
s
uch
as
vo
l
um
e,
va
riet
y,
velocit
y,
an
d
s
o
on
[
11
-
13]
.
Data
s
ci
ence
te
ch
nique
al
lo
ws
pro
cessi
ng
var
i
ou
s
ty
pes
of d
at
a
at
o
nce
e
ven
in
la
rg
e
a
m
ou
nts
beca
use
data
sci
e
nce co
m
bin
es
the s
ta
ti
sti
ca
l
proce
ss
with
data
m
ining
or
m
achine
le
ar
ni
ng
.
Where
dat
a
m
ining
is
a
c
om
pu
ta
ti
on
al
te
chn
i
qu
e
to
fi
nd
insi
gh
t
know
le
dg
e
from
la
rg
e
data
[
14
]
.
I
n
data
sci
ence,
the
re
is
the
expl
or
at
or
y
da
t
a
anal
ysi
s
(E
DA)
te
c
hn
i
qu
e
that
pr
epar
es
and
proces
ses
the
data
sta
ti
sti
cal
ly
[15]
.
St
at
ist
ic
al
data
pr
oces
sin
g
te
ch
niques,
know
n
as
ex
plorat
ory
data
analy
sis
(EDA
),
are
co
ns
ide
r
ed
capa
ble
of
pr
e
par
i
ng
da
ta
bef
ore
it
is
pr
ocesse
d
pro
perl
y.
EDA
can
r
edu
c
e
redu
nd
a
nt
data
,
is
con
si
der
e
d
no
t
to
a
ff
ect
th
e
resu
lt
s,
c
om
plete
the
m
issi
n
g
val
ue,
c
om
pl
et
e
the
data,
so
that
oth
e
r
thin
gs
th
at
m
axi
m
iz
e
t
he
data
to
be
cl
eaner
are
ca
r
ried
out
la
te
r.
Currentl
y,
ED
A
is
al
so
de
ve
lop
in
g
us
in
g
pr
e
dicti
on
m
od
el
s
a
nd
m
achine
le
ar
nin
g
[
16]
.
O
ne
of
the
m
os
t
popula
r
m
achine
l
earn
i
ng
/
data
m
ining
al
gorithm
s that can
be
us
e
d
i
n t
he
E
D
A proc
ess is the
K
-
m
e
ans
cl
us
te
rin
g al
gorithm
[17
-
21]
.
Ther
e
are
se
ve
ral
relat
ed
pre
vi
ou
s
ki
nd
s
of
r
esearch
that
use
data
sc
ie
nce
and
da
ta
m
ining
te
c
hn
i
qu
e
to
pr
ocess
data
,
am
on
g
oth
e
rs
:
(1)
data
sci
en
ce
an
d
harness
ing
a
naly
ti
cs
use
d
t
o
get
a
m
e
anin
gful
asses
s
m
ent
for
le
ar
ning
ac
ti
viti
es
[22]
;
(2)
e
du
cat
io
nal
da
ta
sci
ence
wa
s
us
e
d
t
o
e
valu
at
e
stud
e
nts’
usa
ge
i
n
the
m
a
ssiv
e
op
e
n
on
li
ne
c
ourse
[23]
;
(3)
da
ta
sci
en
ce
ap
proac
h
al
so
us
e
d
f
or
i
den
ti
fyi
ng
the
cr
ucial
fa
ct
or
s
f
or
assess
m
ent
of
an
i
nter
national
stu
den
t
w
it
h
predict
in
g
t
he
e
xam
resu
lt
[24]
;
(
4)
data
m
ining
a
ppr
oa
ch
with
K
-
m
eans
a
nd
Naï
ve
Ba
ye
s
a
lgorit
hm
al
so
us
e
d
f
or
unde
r
sta
nd
i
ng
t
he
di
gital
le
arn
in
g
s
ources
[25]
;
(5)
K
-
m
eans
al
go
rithm
was
pro
ve
n
as
accurate
eval
ua
ti
on
f
or
le
ar
nin
g
e
valuati
on
base
d
on
b
rai
nwave
-
based
e
m
ot
ion
[
26
]
;
(6
)
the
re
is researc
h
that evaluat
es tea
cher’s expe
rienc
e in d
igit
al
co
nt
ent ev
al
uatio
n usin
g
qual
it
at
i
ve
them
at
ic
an
al
ysi
s
with
K
-
m
eans
cl
us
te
r
a
naly
sis
[27]
;
an
d
(
7) the
le
arn
i
ng
be
hav
i
or
p
at
te
r
n
of
t
he
di
gital
te
xtbook w
as
a
na
ly
zed
us
in
g
cl
us
te
ri
ng
m
e
tho
d
us
in
g
K
-
m
eans
al
go
rithm
[28]
.
B
ased
on
m
any
pr
e
vious
ki
nds
of
resea
rc
h
th
at
us
e
data
sci
ence
with
data
m
ining
,
es
pecial
ly
the
K
-
m
eans
cl
us
te
rin
g
al
gorithm
,
to
eva
luate
dig
it
al
l
earn
i
ng
,
this
stud
y
aim
s
to
inv
est
igate
,
interpr
et
,
a
nd
fin
d
the
m
eani
ngf
ul
patte
rn
of
dig
it
al
cultur
e
in
hig
he
r
ed
uc
at
ion
us
in
g
K
-
m
ean
s
al
gorithm
.
The
i
nter
pr
et
a
ti
on
res
ult
can
be
a
m
eaningfu
l
knowle
dg
e
in
res
po
nd
i
ng,
dev
el
op
i
ng, a
nd im
pr
ovin
g digit
al
cu
l
ture
in hig
her ed
ucati
on.
2.
RESEA
R
CH MET
HO
D
2.1.
Rese
arch
act
i
vities
The
case
of
t
hi
s
researc
h
is
UIN
S
una
n
G
unung
D
j
at
i
B
andu
ng
that
is
on
e
of
t
he
higher
ed
ucati
on
with
a
visio
n
t
o
beco
m
e
a
superi
or
a
nd
c
om
pet
it
ive
ca
m
pu
s
th
rou
gh
t
he
us
e
of
te
ch
nolo
gy.
No
le
ss
than
58
inf
or
m
at
ion
syst
e
m
s
in
the
UIN
Suna
n
G
unung
Dj
at
i
Ba
ndung
e
nv
i
ronm
ent
that
su
pp
or
t
al
l
edu
cat
io
n
act
ivit
ie
s,
ran
ging
from
adm
issi
on
s
,
Aca
dem
ic
serv
ic
es
adm
inist
rati
on
syst
em
s
,
Finan
ci
al
inf
orm
ation
syst
e
m
s,
e
m
plo
ye
e
in
form
at
i
on
syst
em
s
,
e
-
Lib
ra
ry,
e
-
Lea
rn
i
ng,
syst
em
s
reg
ist
rati
on
of
assem
blies
,
Helpd
es
k
Syst
e
m
s,
and
var
i
ou
s
i
nfor
m
at
ion
syst
em
s
and
oth
e
r
a
pp
l
ic
at
ion
s.
Howe
ver,
it
tur
ns
ou
t
that
awar
e
ne
ss
an
d
the
nee
d
f
or
th
e
us
e
of
t
he
in
form
ation
syst
e
m
pr
ovide
d
a
re
not
eve
nly
distrib
uted
t
hro
ug
hout
the
ac
adem
ic
com
m
un
it
y, there ar
e
sti
ll
those w
ho r
el
y
on
each
oth
e
r,
e
ve
n
in
dif
fer
e
nt.
The
re
searc
h
act
ivit
y
dep
ic
te
d
in
Fi
gure
1
sta
rts
f
r
om
li
te
ratur
e
stu
dies
relat
ed
to
data
sci
ence
,
data
m
ining
,
a
nd
t
he
K
-
m
eans
al
gorithm
,
wh
ic
h
the
n
c
ompil
es
quest
ions
for
quest
io
nna
ires
to
be
distri
bu
te
d
to
sta
keholde
r
s
in
var
i
ou
s
fa
culti
es,
stud
y
pro
gr
am
s/
dep
artm
ents,
to
un
it
s
in
the
en
vi
ronm
ent.
UI
N
Su
na
n
Gun
ung
D
j
at
i
Ba
ndung.
The
qu
est
io
nnai
re
data
is
then
proces
sed
us
in
g
EDA
a
nd
cl
ust
ering
t
he
K
-
m
eans
al
g
ori
thm
.
The
resu
lt
s
of
E
D
A
an
d
K
-
m
eans
data
processi
ng
a
re
analy
zed,
stu
died
,
an
d
interpret
ed
to
find
a
m
eaning
f
ul
pa
tt
ern
s
o
t
hat
c
an
pro
du
ce
a
r
ecom
m
end
at
io
n
m
od
el
f
or
str
at
egies
to
stren
gth
e
n
dig
it
al
c
ultur
e
in
the
academ
i
c
com
m
un
it
y
of
S
unan
G
unung
Dj
at
i
U
ni
ver
sit
y,
Ba
nd
ung.
A
nd
this
r
esearch
us
e
Py
thon
as
pro
gr
am
m
ing
l
angua
ge fo
r
E
DA,
K
-
m
eans
, a
nd text a
naly
ti
cs
[29]
.
The
ex
pe
rim
e
nt
of
t
his
stu
dy
util
iz
ed
th
e
Goo
gle
cola
borato
ry
(Go
ogle
colab
)
wit
h
Pyt
hon
as
pro
gr
am
m
ing
l
angua
ge
for
E
DA
an
d
te
xt
a
naly
ti
cs.
For
t
he
K
-
m
eans
cl
us
te
rin
g,
this
s
tud
y
us
e
d
O
ra
ng
e
as
data
m
ining
too
ls
.
The
vis
ualiz
at
ion
of
EDA,
K
-
m
eans
cl
us
te
rin
g,
and
te
xt
a
naly
ti
cs
is
pr
ovi
ded
by
the Pyt
hon an
d Ora
nge li
brary
.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N: 20
88
-
8708
Da
t
a
sci
e
nce f
or
dig
it
al c
ultu
re im
pr
ove
me
nt
in
hi
gh
e
r e
ducation
…
(
Dian
Sa’
ad
il
la
h
M
aylaw
ati
)
4571
Figure
1
.
Re
se
arch
act
ivi
ti
es
2.2.
Data
c
ollec
ting
Data
colle
ct
io
n
was
car
ried
ou
t
by
distrib
ut
ing
quest
io
nn
ai
res
us
i
ng
G
oogle
form
s
wit
h
a
total
of
60
quest
ions
distrib
uted
to
the
ranks
of
the
Re
ct
or
at
e,
Dea
n,
Se
na
te
,
Burea
u,
In
sti
tuti
on,
a
nd
SP
I
(as
po
li
cy
m
aker
s)
,
9
Fac
ulti
es
an
d
Post
gr
a
du
at
e
(in
volvi
ng
stu
de
nts,
le
ct
ur
er
s,
a
nd
e
ducat
ion
sta
f
f
from
15
stud
y
pr
ogram
s
in
Po
st
gr
a
du
at
e,
5
m
ajo
rs
i
n
the
Fac
ulty
of
Us
hulu
dd
i
n
,
5
m
ajo
rs
in
th
e
Faculty
of
T
arbiy
ah
and
Teache
r
T
rainin
g,
6
st
udy
program
s
in
the
Faculty
of
Sh
a
ria
an
d
La
w,
4
m
ajo
rs
i
n
the
Faculty
of
Da'wa
h
and
Com
m
un
ic
at
ion
,
3
m
ajo
rs
in
t
he
Fa
culty
of
A
dab
an
d
Hu
m
aniti
es,
1
de
pa
rtm
ent
in
the
Faculty
Psycho
l
ogy,
7
m
ajo
rs
in
the
Faculty
of
Scie
nce
and
Tec
hnology,
3
m
a
j
ors
in
the
Fa
culty
of
So
ci
a
l
and
Po
li
ti
cal
Scie
nces,
an
d
4
st
udy
pro
gr
am
s
i
n
the
Fac
ulty
of
Ec
onom
ic
s
and
Islam
ic
B
us
iness
),
11
T
echn
ic
al
Ser
vice
U
nits,
5
Gen
e
ral
an
d
Ma
ha
d
S
er
vi
ce
Un
it
s.
T
he
quest
ionnaire
was
pr
e
pa
red
with
the
co
nc
ept
of
the
Tec
hnolog
y
Acce
ptance
Mod
el
,
am
on
g
othe
rs:
pe
rceived
ease
of
use
,
per
cei
ved
us
ef
uln
e
ss,
be
hav
i
or
intenti
on to
u
s
e, and act
ual s
yst
e
m
u
se
.
2.3.
Explor
ator
y
d
ata an
aly
sis
Ex
plo
rat
or
y
D
at
a
An
al
ysi
s
is
a
pr
oce
dure
to
analy
ze
data
easi
ly
,
accurate,
pr
eci
se
with
m
at
hem
atical
sta
ti
sti
cs
as
an
ou
t
pu
t,
w
her
e
the
proce
ss
is
autom
at
ic
ally
by
m
achine
[3
0,
31]
.
Ba
sic
al
ly
,
EDA
pro
vid
es
a
su
m
m
ary
of
nu
m
erical
data
su
c
h
as
a
ve
ra
ge,
m
edian,
m
axim
u
m
value,
m
ini
m
u
m
value,
a
nd
qua
rtil
e.
ED
A
aim
s
to
su
gges
t
hypo
t
heses
a
bout
the
ca
us
e
s
of
obse
rv
e
d
phen
om
ena,
to
a
ssess
ass
um
pti
on
s
on
wh
ic
h
t
o
ba
se
sta
ti
sti
cal
con
c
lusio
ns
,
to
s
up
port
the
sel
ect
ion
of
a
ppr
opri
at
e
sta
ti
st
ic
al
t
echn
i
qu
e
s,
a
nd
to
pro
vid
e
a
ba
sis
for
furthe
r
data
co
ll
ec
ti
on
.
ED
A
resu
lt
s
are
usu
al
ly
visu
al
iz
ed
us
in
g
gr
a
phic
al
te
chn
iq
ues,
su
c
h
as
square
plo
ts
,
histo
gr
am
s,
Pareto
diag
ram
s,
distrib
utio
n
pl
ots,
m
ulti
di
m
ensio
nal
scal
ing,
pr
inci
pal
com
po
ne
nt
an
al
ysi
s,
and
interact
ive
versi
on
of
t
he
plo
t.
I
n
data
m
ining
or
m
ach
ine
le
a
rn
i
ng
t
echn
i
qu
e
s,
E
D
A
is
us
ually
use
d
i
n
the
pr
e
-
pr
oce
s
sing
process
t
o
vis
ualiz
e,
find
m
issi
ng
,
a
nd
al
s
o
to
lo
ok
f
or
c
orrelat
ion
s
betwee
n
data
or
var
ia
bles.
Be
cause
the
pre
-
processin
g
phase
is
i
m
po
rtant
fo
r
data
sel
ect
io
n,
data
cl
eanin
g
to
i
m
pr
ove
qual
it
y,
data tra
ns
f
or
m
at
ion
, a
nd
data
reducti
on to
ru
n
a
n
e
ff
ic
ie
nt
m
ining
process
.
2.4.
K
-
me
an
s
a
l
gori
th
m
Data
m
ining
i
s
a
te
chn
iq
ue
for
fin
ding
i
m
po
rtant
inf
orm
at
ion
or
i
ns
igh
t
knowle
dg
e
from
big
data
[
32
]
.
Whe
re,
data
m
inin
g
has
f
our
m
ain
a
ppro
ac
hes
,
a
m
on
g
ot
her
s
,
cl
assifi
cat
ion
(
cl
assifi
cat
ion
)
wh
ic
h
is
su
pe
rv
ise
d
le
arn
in
g,
cl
us
te
rin
g
w
hich
i
s
un
s
uper
vise
d
le
arn
i
ng,
as
so
ci
at
ion
ru
le
,
and
sem
i
-
su
pe
rv
ise
d
le
arn
in
g
t
hat
c
om
bin
es
cl
assifi
cat
ion
a
nd
cl
us
te
rin
g.
Data
m
ining
is
us
e
d
to
fi
nd
hi
dd
e
n
in
form
at
ion
that
is
i
m
po
rtant
an
d
can
be
use
d
t
o
predict
an
d
su
pp
or
t
decisi
on
m
aking
.
Cl
us
te
rin
g
is
not
us
ed
to
pr
e
di
ct
li
ke
cl
assifi
cat
ion
,
bu
t
cl
us
te
ri
ng
will
pro
du
ce
th
e
insig
ht
of
dat
a
that
prob
le
m
at
ic
and
analy
z
ed
a
nd
inter
pre
te
d
by
hu
m
an
[
19,
33]
.
K
-
m
eans
is
on
e
of
the
m
os
t
widely
us
e
d
cl
us
te
rin
g
al
gorithm
s
that
find
m
ini
m
u
m
distance
values
in
t
he
s
a
m
e
cl
us
te
r
[34
-
36
]
.
K
-
m
eans
is
a
sim
ple
alg
ori
thm
with
f
ast
processi
ng
tim
e
and
pr
oduc
es
an
op
ti
m
al
cl
us
te
r.
T
he K
-
m
eans
al
gorithm
is as foll
ow
s:
1.
Determ
ine the
nu
m
ber
of clus
te
rs.
2.
In
it
ia
te
t
he
ce
nt
ro
id
v
al
ue
f
or
each clu
ste
r
(
1
,
2
,
…
,
ℝ
)
rand
om
l
y.
3.
Re
peat the cal
c
ulati
on
with
th
e f
or
m
ula
(1)
a
nd
(
2)
unti
l co
nv
e
r
gen
t.
St
art
Observing avail
able
ICT i
n a
case s
tud
y environment
Preparing th
e
questionnaire
Di
stribut
ing the questionnai
res to
respondent
s
Data processi
ng using EDA
Data mi
ning (cl
ustering) using K-
Me
ans Al
gorithm
Ev
aluating the cluster resul
t using si
lhouette coeffici
ent
Analysing and i
nter
pret
ing EDA
and K-M
eans res
ult
Finding the me
aningful patt
ern/ model
for
streng
thening of digital
cultur
e
in
the academic community
End
Ques
tionnaire
Database
EDA
Result
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
10
, No
.
5
,
Oct
ob
e
r 2
020
:
45
69
-
45
80
4572
4.
Fo
r
eac
h
i,
cal
c
ulate
:
(
)
∶
=
a
rg
‖
(
)
−
‖
2
…
(1)
5.
And for eac
h
j
,
cal
culat
e:
∶
=
∑
1
{
(
)
=
}
(
)
=
1
∑
1
{
(
)
=
}
=
1
…
(2)
2.5.
Te
xt
analy
tics
Text
analy
ti
cs
is
a
te
chn
iq
ue
t
o
fi
nd
m
eaningfu
l
know
le
dge
from
te
xt
data
[37
]
.
Te
xt
anal
yt
ic
s
is
no
t
al
ways
cal
le
d
te
xt
m
ining
,
because
te
xt
m
ining
al
ways
con
ta
ins
th
e
m
ining
proce
s
s
inside
it
,
suc
h
as
cl
assifi
cat
ion
,
cl
us
te
rin
g,
or
associat
ion
ru
l
e
fo
r
te
xt
data
.
But,
te
xt
m
in
ing
is
a
par
t
of
te
xt
analy
ti
cs
al
so
.
Anothe
r
te
xt
analy
ti
cs
te
chn
iq
ue
su
c
h
as
senti
m
ent
analy
sis
[3
8
-
40
]
,
op
ini
on
m
ini
ng
[4
1
]
,
so
ci
a
l
m
edia
analy
sis
[4
2
-
4
5
]
,
so
ci
al
netw
orks
[4
6
]
,
a
nd
web
sc
rap
i
ng
and
c
raw
li
ng
[
47
]
.
Se
ver
al
li
te
ratur
e
sai
d
th
at
te
xt
analy
ti
cs
is
a
pa
rt
of
natur
al
l
angua
ge
proce
ssing
(
NLP)
[48
]
,
bec
ause
not
al
l
NLP
us
e
t
ext
data
as
la
ngua
ge
database
,
it
can
be
voic
e/
sound,
im
age,
an
d
vi
deo.
I
nform
at
ion
retriev
al
[
49
]
,
sem
ant
ic
search
e
ng
i
ne
[5
0
]
,
te
xt
si
m
il
arity
or
st
rin
g
m
at
c
hing
[5
1
]
,
an
d
te
xt
su
m
m
ariz
at
ion
[52
]
are
a
ty
pe
of
NL
P
that
com
m
on
ly
us
es
te
xt d
at
a.
2.6.
Sil
ho
ue
tt
e
c
oe
ff
ic
ie
nt
The
cl
us
te
rin
g
resu
lt
sho
uld
be
m
easur
ed
t
o
ens
ur
e
that
the
resu
lt
in
g
patte
rn
is
go
od
e
no
ugh.
T
he
re
are
inte
rn
al
a
nd
e
xter
nal
m
e
asur
em
ents.
E
xter
nal
m
easur
e
m
ents
li
ke
Ja
ccard
Coe
ff
ic
i
ent
[5
3
]
,
P
ur
it
y
[5
4
]
,
Pr
eci
sio
n
an
d
Re
cal
l
[55
]
,
F
-
Me
asu
re
[
56
]
,
and
s
o
on
.
Wh
e
reas,
i
nter
nal
m
easur
e
m
ents
su
c
h
as
Z
-
Sc
or
e
Inde
x
[5
7
]
,
Ga
m
m
a
and
So
m
er’
s
Gam
m
a
[5
8
]
,
Sil
houette
coeffic
ie
nt
[59
]
,
Be
ta
CV
an
d
D
unn
in
dex
[6
0
]
,
and
s
o
on
.
Sil
h
ou
et
te
coe
ff
ic
i
ent
is
widely
us
ed
to
evaluate
the
resu
lt
s
of
cl
us
te
rin
g.
Sil
houette
coe
ff
ic
i
ent
is
a
m
e
tric
that
m
easur
es
cl
us
te
r
se
par
at
io
n
a
nd
com
pactnes
s
at
the
sam
e
ti
m
e
[59,
61
-
63
]
.
F
or
m
ula
(3)
i
s
use
d
to calc
ulate
the
av
e
rag
e
d
ist
a
nc
e in a
cl
us
te
r
and the m
ini
m
um
d
ist
ance b
e
tween
obj
ect
s
to othe
r
cl
us
te
rs
,
=
1
∑
−
{
,
}
′
=
1
(3)
w
he
re,
is t
he
a
ver
a
ge dist
anc
e of
obj
ect
s i
n a cl
us
te
r, i
.e.
(f
or
m
ula (
4)):
=
∑
≠
1
,
∈
|
−
|
|
|
(4)
and
is a distan
ce betwee
n
t
he
obj
ect
wit
h n
earest cent
ro
i
d ce
nter
.
cal
culat
ed by t
he f
or
m
ula (5)
:
=
min
{
|
−
|
,
=
1
,
2
,
…
,
,
≠
1
}
(5)
Sil
houette
Coe
ff
ic
ie
nt
val
ues
range
betwee
n
1
to
-
1
(
−
1
≤
ℎ
≤
1
),
w
he
re
1
m
eans
the
gr
ouping
s
olu
ti
on
is
"c
or
rect"
an
d
-
1
m
eans
t
he
gro
uping
so
l
ution
is
"wro
ng".
H
oweve
r,
a
cco
r
din
g
to
the
res
ults
of
c
lusterin
g,
it
do
es
no
t
offer
a
gu
a
ra
ntee
of
a
ccur
acy
,
but
m
any
inter
pr
et
at
ion
s
of
t
he
res
ults
of
cl
us
te
rin
g.
S
o,
there
is
no
guara
ntee
that
the
Sil
ho
uette
Coef
fici
ent
val
ue
cl
ose
to
1
al
ways
has
th
e
rig
ht
cl
us
te
r
a
nd m
a
ny inter
pr
et
at
io
ns
, a
nd
vice
ve
rsa.
3.
RESU
LT
S
A
ND
DI
SCUS
S
ION
S
3.1.
Data
c
ollec
tio
n
D
a
t
a
c
ol
l
e
c
t
i
o
n
w
h
i
c
h
w
a
s
s
uc
c
e
s
s
f
u
l
l
y
o
b
t
a
i
n
e
d
i
s
a
t
o
t
a
l
l
y
o
f
2
8
8
7
d
a
t
a
f
r
o
m
3
3
8
L
e
c
t
u
r
e
r
s
,
2
0
0
E
d
u
c
a
t
i
o
n
a
l
P
e
r
s
o
n
n
e
l
,
a
n
d
2
3
4
9
S
t
u
d
e
n
t
s
o
f
U
I
N
S
u
n
a
n
G
u
n
u
n
g
D
j
a
t
i
B
a
n
d
u
n
g
.
T
h
i
s
d
a
t
a
a
l
r
e
a
dy
f
u
l
f
i
l
l
e
d
1
0
%
o
f
t
h
e
p
o
p
u
l
a
t
i
o
n
.
H
o
w
e
v
e
r
,
t
o
m
e
e
t
t
he
q
u
a
l
i
t
y
o
f
t
h
e
r
e
s
u
l
t
,
t
h
e
m
i
s
s
i
n
g
v
a
l
u
e
a
n
d
o
u
t
l
i
e
r
a
r
e
d
e
c
i
d
e
d
t
o
b
e
d
e
l
e
t
e
d
.
S
o
,
t
o
t
a
l
d
a
t
a
t
h
a
t
u
s
e
d
a
r
e
2
3
6
5
r
e
s
p
o
n
d
e
n
t
d
a
t
a
w
i
t
h
2
9
8
L
e
c
t
ur
e
r
,
1
2
8
E
d
u
c
a
t
i
o
n
a
l
P
e
r
s
o
n
n
e
l
,
a
n
d
1
9
3
9
S
t
u
d
e
n
t
.
W
h
i
l
e
t
h
e
t
ot
a
l
f
e
m
a
l
e
i
s
1
3
4
8
r
e
s
p
o
n
d
e
n
t
s
a
n
d
1
0
7
1
r
e
s
p
o
n
d
e
n
t
s
a
r
e
m
a
l
e
.
T
h
e
q
u
e
s
t
i
o
n
s
a
r
e
c
o
l
l
e
c
t
e
d
b
a
s
e
d
o
n
p
a
r
a
m
e
t
e
r
s
o
f
t
h
e
t
e
c
h
n
o
l
o
g
y
a
c
c
e
p
t
a
n
c
e
m
o
d
e
l
(
T
A
M
)
,
s
u
c
h
a
s
p
e
r
c
e
i
v
e
d
u
s
e
f
u
l
n
e
s
s
(
P
U
)
,
P
e
r
c
e
i
v
e
d
e
a
s
e
o
f
u
s
e
(
P
E
U
)
,
B
e
h
a
v
i
o
u
r
a
l
i
n
t
e
nt
i
o
n
t
o
u
s
e
(
B
I
U
)
,
a
n
d
A
c
t
u
a
l
s
y
s
t
e
m
u
s
e
(
A
S
U
)
[6
4
,
6
5
]
.
T
A
M
a
l
s
o
a
l
r
e
a
d
y
u
s
e
d
t
o
e
va
l
u
a
t
e
i
n
f
o
r
m
a
t
i
o
n
t
e
c
h
n
o
l
o
g
y
i
n
h
i
g
h
e
r
e
d
u
c
a
t
i
o
n
,
s
u
c
h
a
s
e
-
l
e
a
r
ni
n
g
o
r
l
e
a
r
n
i
n
g
m
a
n
a
g
e
m
e
nt
s
y
s
t
e
m
s
[66
,
6
7
]
.
T
h
i
s
r
e
s
e
a
r
c
h
h
a
s
6
q
u
e
s
t
i
o
n
s
r
e
l
a
t
e
d
t
o
P
U
,
1
1
q
u
e
s
t
i
o
n
s
f
o
r
P
E
U
,
1
0
q
u
e
s
t
i
o
n
s
f
o
r
B
I
U
,
a
n
d
1
2
q
u
e
s
t
i
o
n
s
f
o
r
A
S
U
.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N: 20
88
-
8708
Da
t
a
sci
e
nce f
or
dig
it
al c
ultu
re im
pr
ove
me
nt
in
hi
gh
e
r e
ducation
…
(
Dian
Sa’
ad
il
la
h
M
aylaw
ati
)
4573
3.2.
Resul
t of
ex
plorat
ory d
ata a
na
lysis
and
data cl
ust
eri
n
g
3.2.1.
E
xp
l
orato
r
y
data a
na
l
ys
is resul
t
T
h
e
r
e
s
u
l
t
o
f
e
x
p
l
o
r
a
t
o
r
y
d
a
t
a
a
n
a
l
y
s
i
s
(
E
D
A
)
s
h
o
w
s
s
e
v
e
r
a
l
c
o
n
c
l
u
s
i
o
n
s
r
e
l
a
t
e
d
t
o
t
h
e
i
m
pl
e
m
e
nt
a
t
i
o
n
o
f
i
n
f
o
r
m
a
t
i
o
n
t
e
c
h
n
o
l
o
g
y
i
n
U
I
N
S
u
n
a
n
G
u
n
u
n
g
D
j
a
t
i
B
a
n
d
u
n
g
a
n
d
c
a
n
b
e
u
s
e
d
a
s
a
b
a
s
i
s
f
o
r
e
n
h
a
n
c
i
n
g
d
i
g
i
t
a
l
c
u
l
t
u
r
e
i
n
h
i
g
h
e
r
e
d
u
c
a
t
i
o
n
.
T
h
i
s
a
n
a
l
y
s
i
s
i
s
b
a
s
e
d
o
n
t
h
e
r
e
s
u
l
t
o
f
E
D
A
a
n
d
t
h
e
c
o
r
r
e
l
a
t
i
o
n
b
e
t
w
e
e
n
p
a
r
a
m
e
t
e
r
s
t
h
a
t
v
i
s
u
a
l
i
z
e
d
i
n
F
i
g
u
r
e
2
.
T
h
e
a
n
a
l
y
s
i
s
r
e
s
u
l
t
s
,
a
m
o
n
g
o
t
h
e
r
s
:
a.
Ov
e
rall
,
the
pe
rceive
d
us
ef
ul
ness
of
inf
orm
at
ion
syst
em
s
i
s
ab
ov
e
ave
ra
ge
with
value
is
3.4
3.
Pe
rceive
d
us
ef
uln
e
ss
i
ndic
at
es
the
le
ve
l
of
c
onfide
nce
i
n
in
div
i
du
al
s
th
at
te
chnolo
gy
ca
n
im
pr
ov
e
the
ir
perform
ance
[6
8
]
.
The
refo
re
,
the
res
ult
va
lue
of
pe
rceiv
ed
us
ef
ulne
ss
in
U
I
N
S
unan
G
unung
D
j
at
i
Ba
ndung
can
be
co
nclu
ded
th
at
academ
ic
so
ci
et
y
awar
e
of
the
use
f
uln
ess
of
dig
it
al
te
ch
no
l
og
y
t
hat
ca
n
m
ake
their
act
i
viti
es
m
or
e
effi
ci
ent
and
eff
e
ct
ive.
This
val
ue
is
su
pp
or
te
d
by
65.
62
%
r
esp
onde
nts
know
about
the
te
rm
of
the
in
form
at
ion
syst
em
,
69%
knows
a
bout
the
be
ne
f
it
s
of
inf
or
m
ation
te
ch
nolo
gy/
inf
or
m
at
ion
syst
e
m
, 6
2.41% r
esp
onde
nts kn
ow
th
e m
a
in websit
e that
pro
vid
es the
up
-
to
-
date in
form
at
i
on
about
academ
ic
act
ivit
ie
s
and
ne
ws.
H
ow
e
ver,
the
in
for
m
at
ion
sh
ari
ng
about
the
in
form
ation
syst
em
is
low,
only
42%
.
This
fact
sho
ws
that
to
day,
the
academ
ic
so
ci
et
y
has
be
en
awa
re
that
their
academ
i
c
act
ivit
ie
s
can
no
t
b
e
sepa
rat
ed
f
ro
m
te
chnolo
gy
suppo
rt.
Ther
e
f
or
e,
t
o
i
m
pr
ove
di
gital
culture
in
higher
edu
cat
io
n,
t
he
te
chnolo
gy
not
al
ways
us
e
d
e
ver
y
ti
m
e
bu
t
t
he
pe
rce
ption
of
th
e
us
e
fu
l
ne
ss
of
te
c
hnolog
y
m
us
t
al
ways
be
m
a
intai
ned
[
69
]
.
This
can
be
reali
zed
if
the
te
ch
nolo
gy
that
avail
able
s
upport
t
he
nee
ds
of u
se
r
s,
so tha
t t
he
use
f
uln
es
s can
b
e
f
el
t.
Figure
2
.
Co
rr
e
la
ti
on
b
et
wee
n TAM
par
am
et
e
rs
b.
The
per
cei
ved
ease
of
us
e
of
t
he
inf
orm
at
ion
syst
em
wh
ic
h
avail
a
ble
in
U
IN
Sun
an
G
unun
g
D
ja
ti
Ba
ndung
is
qu
it
e
low,
belo
w
the
norm
al
value,
it
is
2.6
7.
Perceive
d
ease
of
us
e
i
nd
ic
at
es
the
use
of
a
n
inf
or
m
at
ion
syst
e
m
wh
et
her
easy
to
us
e
or
unde
rsta
nd
[70
]
.
T
he
r
esult
value
m
eans
that
m
a
ny
respo
nd
e
nts
fe
el
diff
ic
ulty
in
us
in
g
the
syst
em
.
The
fact
sh
ow
s
that
(
31.
07%
res
pondent
)
is
on
ly
25
%
of
al
l
inform
at
ion
syst
em
wh
ic
h
pro
vid
e
the
m
anu
al
guide
be
cause
of
on
ly
a
half
of
syst
em
that
pr
ovide
s
the
com
plete
m
anu
al
guide
and
c
onduct
ed
the
s
ocial
iz
at
ion
or
trai
ning
ho
w
to
us
e
the
syst
e
m
.
The
n,
42.
16%
of
re
sp
onde
nt
agr
ee
that
half
of
the
in
form
at
ion
syst
em
e
asy
to
us
e,
44.
95%
res
ponde
nt
agr
ee
that
half
of
syst
em
us
er
interfa
ce
is
int
eresti
ng
s
o
eas
y
to
un
der
sta
nd,
a
nd
45.92%
res
pondent
fe
el
that
only
a
half
of
syst
em
that
f
ulfill
s
the
requirem
ents
of
th
e
proce
s
s
business
t
hro
ugh
t
he
functi
ons
t
ha
t
avail
able.
M
ost
of
them
decide
to
us
e
the
syst
e
m
al
tho
ugh
t
hey
n
ee
d
m
or
e
tim
e
for
underst
an
ding
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
10
, No
.
5
,
Oct
ob
e
r 2
020
:
45
69
-
45
80
4574
the
syst
e
m
,
e
ven
if
the
re
is
a
new
syst
e
m
that
release
d
(
58.56%
).
This
re
su
lt
con
cl
udes
th
at
the
so
ci
al
iz
at
ion
,
trai
ni
ng
,
an
d
m
anu
al
gu
i
de
for
the
syst
em
is
i
m
po
rtant
to
i
m
pr
ove
the
pe
rceive
d
ease o
f
us
e
of
t
he
in
f
orm
ation
syst
e
m
in
higher
ed
ucati
on.
T
he
i
m
pact
is
the
dig
it
al
culture
in
higher
ed
ucati
on
will
b
e im
pr
ov
ed
to
o.
c.
The
be
ha
vior
of
inte
ntion
is
def
ine
d
as
a
n
effo
rt
or
str
ong
de
sir
e
fro
m
the
us
ers
to
try
and
us
e
the
syst
em
[71
]
.
In
U
IN
S
un
a
n
Gun
ung
D
j
at
i
Ba
ndung,
the
be
hav
i
or
of
in
te
ntion
t
o
us
e
t
he
in
f
or
m
at
ion
syst
e
m
is
abo
ve
the
norm
al
value,
ar
ound
3.28.
It
m
eans
th
at
dig
it
al
cul
tu
re
in
UIN
S
un
an
G
unun
g
D
ja
ti
B
a
n
d
u
n
g
a
l
r
e
a
d
y
a
w
a
k
e
n
e
d
.
I
t
i
s
p
r
o
v
e
n
b
y
a
l
m
o
s
t
5
0
%
o
f
r
e
s
p
o
n
d
e
n
t
s
'
s
up
p
o
r
t
a
n
d
e
n
t
h
u
s
i
a
s
m
t
o
t
r
y
a
n
d
u
s
e
t
h
e
n
e
w
i
n
f
o
r
m
a
t
i
o
n
s
y
s
t
e
m
i
f
i
t
i
s
r
e
l
e
a
s
e
d
.
B
e
c
a
u
s
e
5
2
.
2
6
%
o
f
r
e
s
p
o
n
d
e
n
t
f
e
e
l
t
h
a
t
u
s
i
n
g
t
h
e
i
n
f
o
r
m
a
t
i
o
n
s
y
s
t
e
m
c
a
n
s
up
p
o
r
t
t
h
e
i
r
a
c
a
d
e
m
i
c
a
c
t
i
v
i
t
i
e
s
e
f
f
i
c
i
e
nt
l
y
a
nd
e
f
f
e
c
t
i
v
e
l
y
.
T
h
i
s
u
n
d
e
r
s
t
a
n
d
i
n
g
m
u
s
t
b
e
c
o
n
t
i
nu
a
l
l
y
d
e
v
e
l
o
p
e
d
a
n
d
m
a
i
nt
a
i
n
e
d
t
o
i
m
p
r
o
v
e
d
i
g
i
t
a
l
c
u
l
t
u
r
e
i
n
h
i
g
h
e
r
e
d
u
c
a
t
i
o
n
,
be
cau
se
the
be
hav
i
or
al
intenti
on
to u
s
e
inf
or
m
at
ion
te
chnolo
gy
ca
n
be
the h
a
bit
o
f
the
dig
it
al
us
e
r
[72
]
,
es
pecial
ly
academ
ic
so
ci
et
y i
n
us
i
ng aca
dem
ic
inf
or
m
at
ion
syst
e
m
s such as
the lea
rn
i
ng m
anag
em
ent syst
e
m
[73
]
.
d.
Fo
r
the
whole
inf
or
m
at
ion
syst
e
m
that
avail
a
ble
in
UIN
S
unan
G
unun
g
D
j
a
ti
Ba
nd
un
g,
th
e
act
ual
syst
em
us
es
is
sti
ll
qui
te
low
belo
w
t
he
norm
al
value,
it
is
2.6
5.
T
his
val
ue
m
eans
in
the
im
ple
m
entat
ion
of
t
he
inf
or
m
at
ion
syst
e
m
s
is
sti
l
l
l
ow,
su
c
h
as
44
.27%
of
res
pond
e
nts
ag
ree
that
on
ly
half
of
the
syst
e
m
th
at
fu
lfil
ls
the
use
r
or
busines
s
process
re
qu
i
r
e
m
ents,
only
29.
69%
of
res
ponde
nts
that
re
m
e
m
ber
the
li
nk
address
to
acce
ss
the
syst
em
t
hat
they
nee
d.
Eve
n
th
ough,
78.14%
of
res
po
ndents
deci
de
to
us
e
a
dig
it
al
syst
e
m
to
su
pp
or
t
thei
r
aca
de
m
ic
act
ivities.
This
re
su
lt
pro
ves
that
it
is
i
m
po
rtant
to
de
sign
t
he
system
that
involvin
g
us
ers
i
n
it
s
de
velo
pm
ent
to
fu
lfil
l
their
r
equ
i
rem
ents.
Be
cause
eve
ry
ty
pe
of
us
er
ha
s
a
diff
e
re
nt
requi
rem
ent
that
m
us
t
be
accom
m
od
at
ed,
anal
yz
ed,
an
d
sel
e
ct
ed
so
as
t
o
be
st
m
eet
the
need
s
of
al
l
us
er
s.
T
he
good
in
for
m
at
ion
syst
e
m
/
so
ftwa
re
desi
gn
will
pr
od
uc
e
a
good
qu
al
i
ty
of
inf
or
m
at
i
on
syst
e
m
[74
]
, t
he
n
the
d
i
gital
cu
lt
ur
e
w
il
l be
im
pr
ov
e
d
if all
they
n
ee
d
a
re a
ccom
m
od
at
ed.
3.2.2.
Resul
t a
nd
inte
rpret
at
ion
of t
he
K
-
means
a
l
go
ri
t
hm
Figures
3
-
5
vi
su
al
iz
ed
the
resu
lt
of
the
K
-
m
eans
al
gorithm
in
cl
ust
ering
the
ty
pe
of
data
(the
cl
us
te
ri
ng
resu
lt
an
d
the
exam
ple
of
a
cl
us
te
r
m
e
m
ber
).
The
cl
us
te
r
s
are
form
ed
due
to
the
si
m
il
a
rity
of
data
char
act
e
risti
cs.
Ba
sed
on
the
sil
ho
uette
coeffic
ie
nt
val
ue
with
E
uclid
ean
distance
,
the
best
cl
us
te
r
for
this
researc
h
data
is
two
cl
us
te
rs
,
with
1124
in
C
luster
1
(C
1
-
Bl
ue)
a
nd
12
41
in
Cl
us
te
r
2
(
C2
-
Re
d).
T
he
refor
e
,
Figure
3
visu
al
iz
e
the
resu
lt
with
tw
o
cl
us
t
er
-
base
d
on
Re
sp
on
de
nt
(Lectur
e
r
=
1;
Ed
uc
at
ion
al
Pers
on
nel
=
2;
Stud
e
nt
=
3),
Fig
ur
e
4
is
base
d
on
Ge
nder
(Mal
e
=
1;
Fem
al
e
=
2),
w
hile
Fi
gure
5
bas
ed
on
Ag
e
.
The
init
ia
ti
on
of
th
e
centr
oid
center
is
assig
ne
d
ra
ndom
ly
.
K
-
m
eans
al
gor
it
h
m
us
es
ra
ndom
initial
iz
ation
wit
h
300
m
axi
m
um
it
erati
on
s.
Actuall
y,
the
K
-
m
eans
cl
us
t
erin
g
res
ult
is
no
t
reli
able
enou
gh,
the
re
are
m
any
me
m
ber
s
of
t
he
cl
us
te
r
(
bo
t
h
blu
e
cl
us
te
r
and
red
cl
ust
er
)
that
far
apa
rt
fr
om
the
centr
oid
.
Ma
ny
m
e
m
ber
s
that
al
s
o
ha
ve
the
sim
il
arit
y,
wh
e
reas
t
hey
are
in
a
dif
fer
e
nt
cl
us
te
r
.
T
he
Sil
houette
coe
ff
ic
ie
nt
value
of
t
his
cl
us
te
r
is
to
o
low
(
0.1
25)
s
o
that
the
cl
us
t
er
is
quit
e
diffi
cult
to
be
inte
rpreted.
H
ow
e
ver,
the
cl
us
te
r
re
gion/
area
is
sti
ll
qu
it
e cle
arly
se
par
at
e
d
(
vis
ualiz
ed
with
a
bac
kgr
ound c
olo
r
)
.
Figure
3
.
Cl
us
t
er
re
su
lt
based
on r
es
ponde
nt
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N: 20
88
-
8708
Da
t
a
sci
e
nce f
or
dig
it
al c
ultu
re im
pr
ove
me
nt
in
hi
gh
e
r e
ducation
…
(
Dian
Sa’
ad
il
la
h
M
aylaw
ati
)
4575
Figure
4
.
Cl
us
t
er
re
su
lt
based
on g
e
nder
Figure
5
.
Cl
us
t
er
re
su
lt
based
on ag
e
Wh
e
n
e
xam
ined furthe
r,
t
her
e
are
se
ve
ral co
nclusi
on
s
that
can
be ob
ta
i
ne
d,
am
ong othe
r
s:
a.
The
C1
is
a
gr
oup
of
res
pond
ents
that
ha
ve
per
cei
ved
ease
of
use
at
the
lowest
le
vel
(
unde
r
2),
it
m
ea
ns
that
47.
53%
of
re
spo
nd
e
nts
sti
ll
feel
need
the
m
or
e
ef
for
t
to
use
/
ada
pt
/
le
arn
t
he
in
f
or
m
at
ion
syst
e
m
because
of
la
ck
of
the
in
f
or
m
at
ion
an
d
so
ci
al
iz
at
ion
of
syst
em
.
On
t
he
ot
her
hand,
56.
47%
of
the
res
ponde
nt
(C2)
feel
nor
m
al
or
eve
n
fe
el
easy
in
us
e
the
syst
em
.
But,
C2
ha
s
the
hi
gh
est
per
cei
ve
d
us
ef
uln
e
ss.
It
m
eans
that
m
os
t
of
the
res
po
nd
e
nts
know
a
nd
un
der
sta
nd
the
ben
e
fits
of
dig
it
al
te
ch
no
l
ogy
to s
upport aca
dem
ic
acti
viti
es
in h
i
gh
e
r
e
duc
at
ion
.
b.
Ba
sed
on
th
e
cl
us
te
ri
ng
re
su
lt
in
Fig
ur
e
3,
c
om
par
ed
with
the
stu
den
t,
m
os
t
of
the
le
ct
ure
r
a
nd
edu
cat
io
nal
pe
rson
nel
are
i
n
C2.
It
m
eans
that
the
le
c
turer
an
d
e
du
cat
ion
al
pe
rs
onnel
ca
n
le
a
r
n
the
syst
em
eas
ie
r
that
stu
dent
.
This
fact
shows
that
the
s
ocial
iz
at
ion
h
as
not
bee
n
e
ve
nly
distrib
uted
,
it
sh
ou
l
d
be
a
ssu
m
ed
that
so
ci
al
iz
at
ion
/
trai
nin
g
on
the
us
e
of
the
sys
tem
is
m
os
t
ly
carried
out
by
le
ct
ur
ers
and e
du
cat
io
nal
pers
onnel tha
n
t
he st
ud
e
nt.
c.
Ba
sed
on
ge
nder
in
Fi
gure
4,
the
re
is
no
sign
ific
a
nt
cl
ust
er
dif
fer
e
nce.
Moreove
r,
th
ere
are
to
o
m
a
ny
m
e
m
ber
s
of
C
1
an
d
C2
t
hat
in
the
sam
e
cl
us
te
r
reg
i
on.
H
oweve
r,
it
is
show
n
that
gend
er
does
no
t
af
f
ect
the use
of
tec
hnol
og
y a
nd
digi
ta
l cult
ur
e in
hi
gh
e
r
e
du
cat
io
n.
d.
Be
cause
of
t
he
high
a
ge
var
i
at
ion
(
vis
ualiz
ed
in
Fig
ur
e
5),
it
ap
pea
rs
th
at
the
cl
us
te
rs
form
ed
are
not
com
pact.
How
ever,
w
hat
ca
n
be
ob
ta
in
ed
from
cl
us
te
r
res
ults
base
d
on
a
ge
that
i
n
the
age
ra
nge
a
bove
30
ye
ars
,
m
or
e
in
C2.
This
sh
ows
that
they
can
us
e
the
inform
ation
syst
e
m
well
and
feel
the
ben
efit
of
the in
fo
rm
at
io
n
syst
em
b
ecau
se of the
s
uppo
rt of
good s
oci
al
iz
at
ion
and tr
ai
nin
g sy
ste
m
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
10
, No
.
5
,
Oct
ob
e
r 2
020
:
45
69
-
45
80
4576
3.3.
Result
of
text
analy
tics
3.3.1.
Te
s
t
pre
-
pr
ocessin
g
Text
pr
e
-
proce
ssing
is
an
im
portant
phase
in
te
xt
analy
tics,
inclu
ding
in
te
xt
m
ining
and
natu
ral
la
nguag
e
pro
c
essing.
Te
xt
pre
-
processi
ng
i
s
a
phase
t
o
prepa
re
te
xt
dat
a
unti
l
read
y
t
o
process
i
n
the
nex
t
ph
a
se
an
d
ens
ure
the
qual
it
y
of
te
xt
data
[7
5
,
76
]
,
ei
ther
in
t
he
input
proce
s
s
or
res
ult
proc
ess.
N
ot
al
l
pr
oc
ess
in
the
te
xt
pre
-
proce
ssin
g
is
us
e
d,
s
om
eti
m
es
it
is
in
accord
a
nce
wit
h
th
e
need
s
of
the
researc
h.
G
en
erall
y,
there
are
to
ke
nizing,
lo
wer
case
(case
f
old
in
g),
rem
ov
e
re
gu
la
r
ex
pr
essi
on,
sto
p
-
w
ord
rem
ov
i
ng,
an
d
stemm
ing
pro
cess
in
t
he
te
xt
pre
-
proces
s
ing
proces
s
[77
]
.
E
ver
y
la
ngua
ge
has
dif
fer
e
nt
cha
ract
erist
ic
s,
structu
res,
an
d
gr
am
m
a
r,
inc
lud
in
g
t
he
I
nd
on
e
sia
n
la
ng
ua
ge.
T
his
resea
rch
us
es
the
I
ndonesi
an
la
ngua
ge.
The
te
xt d
at
a is
co
ll
ect
ed
is c
onta
ined
the
m
e
ssage, i
m
pr
ession, a
nd ho
pe fr
om
2
887 res
pond
e
nts.
3.3.2.
Resul
t a
nd
inte
rpret
at
ion
of t
e
xt ana
lytics
Figure
6
il
lustr
at
ed
the
w
ord
cl
oud
base
d
on
the
f
re
qu
e
ncy
of
w
ords.
As
s
how
n
in
Fig
ur
e
7
t
op
15
of
words
that
ap
pear
e
d
from
te
xt
data
are:
l
ebih
(m
or
e),
di
gital
,
apli
kasi
(a
pp
li
cat
io
n)
,
m
ahas
isw
a
(c
ollege
stud
e
nt),
sy
ste
m
(syste
m
),
semog
a
(
hope/
wish
)
,
baik
(go
od
/
well
)
,
UIN,
on
li
ne
,
yg
(a
bbre
viati
on
of
yang
-
prep
os
it
ion
in
I
ndones
ia
n
la
ng
uag
e
),
bany
ak
(a
l
ot
of
/
m
any/
m
uch
),
ti
dak
(no),
nya
(
pos
sessive
pro
noun
in
I
ndon
e
sia
n
la
ngua
ge),
s
os
iali
sa
i
(so
ci
al
iz
at
ion),
an
d
inf
or
m
ation
(in
f
or
m
asi
).
Actuall
y,
the
wor
d
su
c
h
as
yg
,
ti
dak,
n
ya
,
ti
da
k
,
and
m
any
m
ore
wh
ic
h
are
a
bbre
viate
d
a
nd
i
nclu
ded
i
n
the
stop
-
w
ord
cat
e
gory,
un
s
ucces
sf
ully
rem
ov
ed.
And
al
so
,
seve
ra
l
aff
ixes
in
t
he
stemm
ing
proces
s
is
not
change
d.
It
ha
pp
e
ns
because
this e
xperim
ent u
ses t
he
Sa
strawi li
brary
for
Pyt
ho
n wit
hout
im
pr
ovem
ent f
or this
case
[7
8
]
.
Ba
sed
on
the
te
xt
analy
ti
cs
resu
lt
a
bout
t
he
m
essage
an
d
hope
of
res
ponde
nt
incl
ud
i
ng
le
ct
ur
e
r,
edu
cat
io
nal
pe
r
so
nnel
,
and st
udent,
it
can
b
e
con
cl
ud
e
d
t
hat:
a.
In
acc
orda
nce
with the res
ult of
pe
rceive
d
ea
se of use (
PEU) and act
ual syst
e
m
u
se (ASE
)
w
hich
are
l
ow
,
the
res
pondent
(esp
eci
al
ly
stud
e
nt)
hopes
t
hat
so
ci
al
iz
at
ion
or
trai
ning
of
in
f
or
m
at
ion
syst
e
m
m
us
t
be
c
om
pr
ehe
ns
ive
and
m
assive.
This
can
im
pr
ov
e
t
he
di
gital
culture
in
higher
e
du
c
at
ion
that
introd
uce
s
the
syst
em
(o
r
ne
w
syst
em
)
com
plete
ly
a
nd
th
oroug
hly
for
al
l
en
d
-
use
rs.
N
ot
only
certai
n
gro
ups
,
because
each
us
er
has
a
different
le
vel
of
unde
rstan
ding
and
a
djust
m
en
t
about
the
in
f
or
m
at
ion
syst
em
.
The
dig
it
al
nat
ives
who
bor
n
m
or
e
than
19
80
an
d
fam
ilia
r
with
dig
it
al
te
chnolo
gy
al
le
ged
ly
faster
i
n
unde
rstan
ding
new syst
em
s o
r
te
ch
nolo
gy th
an
im
m
igran
ts
natives
[79
]
.
b.
The
so
ci
al
iz
at
ion
m
us
t
be
suppo
rted
by
a
m
anu
al
book
wh
ic
h
com
plete
s
for
each
in
f
or
m
at
ion
syst
em
.
The
fact
of
the
su
r
vey
sho
ws
that
m
anu
al
books
no
t
al
l
av
ai
la
ble
fo
r
eac
h
syst
em
,
and
al
so
incom
plete
instru
ct
io
ns
f
or
us
e
i
n
t
he
s
yst
e
m
.
Even
thou
gh
the
m
a
nu
al
bo
ok
is
a
vaila
ble,
but
it
has
not
s
ha
re
d/
inf
or
m
/
so
ci
al
i
zed
well
,
s
o
t
hat
not
al
l
end
-
us
e
r
ge
t
the
m
anu
al
book
or
ca
n
sea
rch
the
m
anu
al
book
easi
ly
.
Especia
ll
y
fo
r
t
he
st
udent,
in
acc
orda
nce
with
the
K
-
m
eans
cl
us
te
ri
ng
res
ult,
m
os
t
of
the
stu
den
t
s
are in
C1 w
ho
feel nee
d
m
or
e
effo
rt to use t
he
syst
e
m
b
ecau
se of the
lack
of s
ocial
iz
at
i
on.
c.
Gen
e
rall
y,
res
pondents
ho
pe
that
the
in
for
m
at
ion
syst
em
,
buday
a
dig
it
al
/
dig
it
al
cultu
r
e,
syste
m
dig
it
al/
dig
it
al
syst
e
m
,
on
li
ne
syst
em
in
UI
N
Sunan
Gun
ung
Dj
at
i
Ba
ndun
g
le
bih
ba
ik
/
be
tt
er
than
befor
e
.
This
hope
pro
ves
that
dig
it
al
culture
in
U
I
N
Suna
n
G
unung
D
j
at
i
Ba
ndun
g
al
read
y
a
wak
e
ne
d.
It
is
in
accor
da
nce
wi
th
BIU
res
ults
that
above
aver
a
ge,
i
nter
est
,
support
,
and
desire
t
o
us
e
in
form
ation
te
chnolo
gy
are
quit
e
high
.
T
his
need
s
to
be
sup
ported
by
syst
e
m
fu
nc
ti
on
s
t
hat
m
ee
t
the
ne
eds
of
academ
ic
so
ci
et
y,
tho
r
ough
so
ci
al
iz
at
ion
,
and
a
c
om
plete
m
anu
al
book.
S
o
that
digi
ta
l
culture
will
furthe
r
i
m
pr
ov
e
becau
se
th
e
syst
e
m
is
easy
to
us
e,
acc
ord
ing
to
the
needs
of
aca
dem
ic
so
ci
et
y,
and
ha
s
a d
irect
im
pact on
perform
ance b
eca
us
e
w
ork becom
es m
or
e eff
ect
i
ve
a
nd
eff
ic
ie
nt.
Figure
6
.
Wo
r
d
cl
oud res
ult
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N: 20
88
-
8708
Da
t
a
sci
e
nce f
or
dig
it
al c
ultu
re im
pr
ove
me
nt
in
hi
gh
e
r e
ducation
…
(
Dian
Sa’
ad
il
la
h
M
aylaw
ati
)
4577
Figure
7
.
To
p 15
f
reque
nt of
words
4.
CONCL
US
I
O
N
This
researc
h
is
co
nducte
d
com
pr
ehe
ns
i
vely
in
orde
r
to
e
valuate
the
i
nfor
m
at
ion
te
c
hnology
i
m
ple
m
entat
io
n
i
n
higher
e
du
cat
io
n.
T
he
quest
io
nn
ai
re
data
from
le
ct
ur
er
s,
stu
de
nts,
a
nd
e
duc
at
ion
al
per
s
onnel
are
analy
zed
us
i
ng
exp
lo
rato
ry
da
ta
analy
sis
(ED
A
),
K
-
m
ean
cl
us
te
rin
g
al
gorithm
,
and
al
so
te
xt
analy
ti
cs.
The
resu
lt
of
the
e
xp
e
rim
ent
of
EDA
a
nd
K
-
m
eans
al
gorithm
sh
ows
that
to
i
m
pr
ov
e
the
dig
it
al
culture
with
hi
gh
per
cei
ved
e
ase
of
us
e
an
d
act
ual
syst
e
m
us
e
of
inf
orm
a
ti
on
te
ch
no
l
ogy
sh
ould
be
s
uppo
rted
with
com
plete
and
c
om
pr
eh
ensive
s
ocial
iz
at
ion
,
a
nd
al
s
o
pro
vid
e
the
m
anu
al
guide
for
each
in
for
m
at
ion
syst
e
m
.
This
resu
lt
in
accor
da
nce
with
the
hope
of
en
d
-
use
r
that
need
in
f
or
m
at
ion
,
knowle
dge,
an
d
guideli
ne
for
a
n
in
form
a
ti
on
syst
em
th
at
they
us
e
d.
Digital
cultu
re
thr
ough
beh
a
vioral
intenti
on
use
of
in
f
orm
at
ion
syst
e
m
that
alr
eady
a
wak
e
ne
d
s
hould
be
m
ai
ntained
a
nd
i
m
pr
ove
with
the
qual
it
y
of
inf
or
m
at
ion
syst
e
m
wh
ic
h fu
lfil
ls t
he user
r
e
quire
m
ents.
Fo
r
furthe
r
w
orks,
it
needs
to
pr
e
par
e
the
data
bette
r
so
th
at
it
can
produce
the
reli
able
cl
us
te
r,
al
tho
ug
h
cl
us
te
rin
g
is
no
t
us
e
d
to
pre
dict,
it
can
pro
duce
a
m
or
e
accurate
interp
retat
ion
i
f
the
data
pr
e
pa
red
is
bette
r.
The
ot
he
r
cl
us
te
rin
g
m
et
hods
ca
n
be
us
e
d
to
get
a
be
tt
er
cl
us
te
r.
A
nd
al
so,
it
can
us
e
t
he
cl
a
ssifi
cat
io
n
appr
oach
to
pr
edict
the
ty
pe
of
res
ponde
nt
and
the
res
ult
can
be
us
e
d
a
s
decisi
on
sup
port
by
poli
cym
aker
in
higher
educat
ion rela
te
d t
o
i
nfo
rm
ation
tec
hnol
og
y a
nd
digi
ta
l cult
ur
e im
pr
ovem
ent.
ACKN
OWLE
DGE
MENTS
The
re
searc
hers
woul
d
li
ke
to
ap
p
reciat
e
and
m
any
than
ks
to
Re
ct
or
a
nd
t
he
aca
dem
ic
so
ci
et
y
of
UIN Su
nan Gu
nung
Dj
at
i B
a
ndun
g who s
upport t
his r
esea
rc
h.
REFERE
NCE
S
[1]
1E
Repor
t
,
“
The Re
a
l
Cost
of
Un
used
Software
,
”
1E
Company
,
20
15.
[2]
D.
Jam
al
uddin,
M.
A.
Ramdhani,
T.
Pria
tna,
an
d
W
.
Darm
al
aksa
na,
“
Techno
Univer
sit
y
to
incr
ea
se
the
qua
li
t
y
of
isla
m
ic
highe
r
e
duca
t
ion
in
Indo
nesia
,
”
Inte
rnat
i
onal
Journal
of
Civ
il
Engi
ne
erin
g
and
Technol
ogy
,
vol
.
10,
no.
1,
pp.
1264
-
1273
,
2019.
[3]
D.
Kurniadi,
“
Pera
nc
anga
n
Ars
i
te
ktur
Sis
t
em
E
-
ac
ad
emic
deng
a
n
Kons
ep
Kam
pus
Digit
al
Meng
gunaka
n
Unifi
ed
Software
Deve
l
opm
ent
Proce
ss
(US
D
P)
-
Archi
te
ct
ur
al
Design
of
E
-
ac
ad
emic
S
y
stems
with
Digit
al
Campus
Conce
pts
Us
ing
the
Unifie
d
Software
Dev
el
op
m
ent
Proce
ss
(
US
DP
),
”
Jurnal
Wawasan
Ilmia
h
Manaje
men
d
an
Tekni
k
Informati
ka
,
vo
l. 5, no. 10
,
Mar
.
2014
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
10
, No
.
5
,
Oct
ob
e
r 2
020
:
45
69
-
45
80
4578
[4]
M.
E.
Mahm
ud,
“
Mewujudka
n
Sekola
h
At
au
K
ampus
Digit
al
-
C
rea
t
ing
a
School
or
Campus
Dig
it
al
,
”
Dinamika
Ilmu
,
vol
.
1
,
no
1,
2011
.
DO
I:
do
i.
org/10
.
21093/d
i.
v11i1
.
46
[5]
A.
Abuarqoub
e
t
al.
,
“
A
surve
y
on
int
ern
et
of
th
ings
ena
bl
ed
sm
art
c
ampus
applications,
”
in
AC
M
Inte
rnationa
l
Confe
renc
e
on
Fut
ure
Net
work
s
and
Distribute
d
Syste
ms
(
ICF
NDS
2017
)
,
Ca
m
bridge
,
Unite
d
Kingdom
,
pp.
1
-
7
,
2017.
DO
I:
do
i.org/10.
1145/310
2304.
3
109810
[6]
X.
Dong,
X.
Ko
ng,
F.
Zha
ng,
Z
.
Chen,
and
J.
Kang,
“
OnCampus:
a
m
obil
e
pla
tform
towar
ds
a
sm
art
ca
m
pus,”
Springerplus
,
vo
l.
5
,
no
.
974
,
201
6.
DO
I:
do
i.
org/
10.
1186/s40064
-
016
-
2608
-
4
[7]
M.
R.
Vee
rama
nic
kam
and
M.
Mohana
pri
y
a
,
“
IOT
ena
ble
d
Fut
ure
s
Sm
art
Cam
pus
with
eff
ec
tive
E
-
Learni
ng :
i
-
Campus
,
”
Journal
of
Engi
n
ee
ri
ng
Technol
og
y
(
JE
T)
, v
ol.
3
,
n
o.
4,
pp
.
81
-
87
,
Ap
ril
20
16
.
[8]
A.
Alghamdi
an
d
S.
Shetty
,
“
Surve
y
towar
d
a
s
m
art
ca
m
pus
using
t
he
interne
t
o
f
thi
ngs,”
in
Proce
ed
ings
-
2016
IEE
E
4th
Inte
rn
ati
onal
Con
fe
ren
ce
on
Fu
ture
In
t
erne
t
o
f
Things
a
nd
Cloud
(
Fi
Clo
ud)
,
Vienna
,
pp.
235
-
239,
2016
.
[9]
H.
I.
W
ang,
“
C
onstruct
ing
the
gre
en
ca
m
pus
withi
n
the
inter
net
of
thi
ngs
arc
hitect
ur
e,”
Int.
J.
Distrib.
Sens.
Net
works
,
vo
l. 1
0,
no
.
3
,
pp
.
1
-
8
,
2014.
[10]
B.
Manoj,
K.
V.
K.
Sasikant
h,
M.
V.
Subbara
o,
and
V.
Jy
oth
i
Praka
sh,
“
Anal
y
sis of
dat
a
scie
n
ce
wi
th
the
use
of
big
dat
a
,
”
Int. J. Adv
.
Tr
ends
Comput
.
Sc
i. E
ng
.
,
vol.
7,
no
.
6
,
pp
.
87
-
90,
2018
.
[11]
P.
S.
Arocki
a,
S
.
S.
Varne
kha
,
a
nd
K.
A.
Vene
shia,
“
The
17
V’
s
of
Big
Data
,
”
Int.
R
es.
J.
Eng.
Technol
.
,
vol.
4
,
no.
9
,
pp
.
3
-
6
,
2
017.
[12]
S.
Sagirogl
u
an
d
D.
Sinan
c,
“
Big
data:
A
re
vie
w,”
in
Proceedi
ngs
of
the
2
013
Inte
rnat
ion
al
Confe
r
enc
e
on
Coll
a
boration
T
ec
hnolog
ie
s and
Syste
ms
,
CT
S
20
13
,
San
Diego,
CA,
pp.
42
-
47,
2
013.
[13]
K.
Borne,
“
To
p
10
Li
st
–
The
V’s
of
Big
Data,”
Data
Sci
en
ce
C
ent
ral
,
2014.
[Onl
i
ne]
.
Avail
ab
le:
htt
ps://
ww
w.dat
asc
ie
n
cece
ntr
al.com
/profi
le
s/blog
s/top
-
10
-
li
st
-
th
e
-
v
-
s
-
of
-
big
-
d
at
a
.
[14]
H.
Jiawe
i,
M.
Kam
ber
,
J.
Han,
M.
Kam
ber
,
and
J.
Pei,
“
Dat
a
Mining:
Conc
ept
s
and
Te
chn
ique
s
,”
Morgan
Kaufmann
,
2
nd
E
d.
,
2
006.
[15]
M.
Abza
lov
,
“
Expl
ora
tor
y
data a
naly
s
is,” i
n
Mod
ern
Approache
s
in
Soli
d
Earth
Sc
ie
nc
es
,
2016
.
[16]
G.
Saporta
,
“
50
Yea
rs
of
Data
Anal
y
sis:
From
Expl
ora
tor
y
D
ata
Anal
y
s
is
to
Predictive
Mode
ling
and
Mac
hine
Le
arn
ing,”
Dat
a
Anal.
Appl.
1
Clust.
R
egr
essio
n,
Model
.
Fore
c
ast.
Da
ta
Min.
,
ISTE
-
W
il
e
y
,
Da
ta
Ana
l
y
s
is
and
Applic
a
ti
ons,
97
8
-
1
-
78630
-
382
-
0.
ffh
al
-
0247
074
0f,
2019
.
[17]
A.
Kaz
emi
and
G.
Khodaba
ndehl
oui
e,
“
A
new
ini
ti
alisati
on
m
et
hod
for
k
-
me
ans
al
gori
thm
in
the
cl
ust
eri
n
g
proble
m
:
da
ta a
n
aly
s
is,”
Int
.
J. Data
Ana
l. Tec
h.
Strate
g.
,
vo
l. 10, no. 3, pp. 291
-
3
04,
2018
.
[18]
I.
D.
Dinov,
“
K
-
Mea
ns Cluste
r
in
g,
”
in
Data
Sc
ience
and
Pred
ic
t
i
ve
Anal
y
tics
,
Spr
inge
r, pp.
443
-
4
73
,
2018
.
[19]
P.
Nerurka
r,
A.
Shirke,
M.
Chan
dane
,
and
S.
Bhi
rud,
“
Empiric
a
l
ana
l
y
sis
of
data
cl
uster
ing
al
gor
i
thms
,
”
Proce
di
a
Comput.
Sc
i.
,
vo
l.
125
,
pp
.
770
-
7
79,
2018
.
[20]
H.
Anderson
and
G.
As
col
i,
“
Expl
ora
tor
y
D
at
a
Ana
l
y
s
is
of
Autobiogra
phi
c
al
Mem
or
y
T
re
nds,”
J.
S
tude
n
t
-
Sci
en
ti
sts’ R
es.
,
vol.
1
,
2019
.
[21]
J.
T.
Chi
,
E
.
C
.
Chi,
and
R.
G.
Bara
niuk
,
“
K
-
pod:
A
m
et
hod
for
K
-
m
ea
ns
cl
uste
ring
of
m
issing
dat
a
,
”
Am.
S
ta
t
.
,
vol.
70
,
no
.
1
,
pp
.
91
-
9
9
,
2016
.
[22]
D.
Ife
nthaler
,
S.
Greif
f
,
and
D.
Gibson,
“
Making
use
of
dat
a
for
assess
m
ent
s:
Harne
ss
ing
anal
y
ti
cs
and
d
ata
scie
nc
e,”
in
Int
e
rnational
handb
ook
of
IT i
n
primar
y
and
se
conda
ry
educ
a
ti
on
(
2
e
d.
)
,
Springer
,
20
18.
[23]
C.
Rom
ero
and S
.
Ventur
a
,
“
Ed
uca
t
iona
l
data
sc
ie
nc
e
in
m
assive
open
onl
ine
cou
rses,”
Wi
le
y
Int
e
rdiscip.
R
ev.
Dat
a
Min.
Know
l. Discov
.
,
vo
l. 7, no.
1,
p
p
.
e1187
,
20
17.
[24]
L.
Gaid
and
C.
H.
Yu,
“
A
dat
a
scie
nc
e
appr
o
ac
h
to
ide
nt
if
y
cru
cial
factors
of
pre
dic
t
ing
tes
t
per
form
anc
e
in
Program
for
Inte
rna
t
iona
l
Stud
e
nt
As
sessment,
”
SCCUR
South
ern
Cali
fornia
Confe
renc
es
for
Undergr
aduate
Re
search
,
2019.
[25]
H.
Prahe
rdhiono
,
E.
P
.
Adi,
and
R.
N.
Devi
ta
,
“
Understa
nding
o
f
Digit
a
l
Learni
ng
Source
s
with
the
Heut
agog
y
Approac
h
using
the
K
-
Mea
ns
an
d
Naive
B
a
y
es
Methods,
”
in
20
18
4th
Int
ernational
Confe
ren
ce
on
Educ
a
ti
on
a
nd
Technol
ogy
(
ICET)
,
pp.
23
-
27
,
2018
.
[26]
T.
M.
L
i,
H
.
H.
Cho,
H.
C
.
Ch
a
o,
T
.
K.
Shih,
a
nd
C.
F.
L
ai
,
“
An
accurate
br
ai
n
wave
-
base
d
emotion
cl
uster
ing
f
or
le
arn
ing
ev
al
u
at
i
on,
” in
In
te
rnati
onal
Symposium
on
Eme
rging
Te
chnol
ogi
es
for Educat
ion
,
pp
.
22
3
-
233
,
2017
.
[27]
M.
K.
Kim
,
K.
Xie,
and
S.
-
L
.
C
heng,
“
Buil
d
ing
te
a
che
r
competenc
y
for
dig
it
a
l
c
onte
nt
evalua
t
io
n,
”
Tea
ch.
Tea
c
h.
Educ
.
,
vo
l. 66, p
p.
309
-
324
,
201
7.
[28]
C.
Yin,
Z.
Ren
,
A.
Pol
y
zou
,
an
d
Y.
W
ang,
“
L
ea
rning
Beha
v
io
ral
Pat
te
rn
Anal
y
sis
Based
on
Digit
al
T
ext
boo
k
Rea
ding
Logs,
”
i
n
Inte
rnat
ional
Confe
renc
e
on
Hum
an
-
Comput
er
Inte
rac
ti
on
,
p
p.
471
-
480
,
201
9
.
[29]
D.
Sarka
r,
"
Te
x
t
Anal
y
ti
cs
with
P
y
tho
n
-
A
Prac
ti
c
al
Real
-
W
orld
Approac
h
to
Gaini
ng
Act
iona
bl
e
Insights
from
Your Dat
a
,"
A
Press
,
2016.
[30]
J.
W
.
Tuk
e
y
,
“
T
he
Future
of
Da
t
a
Anal
y
sis,
”
Ann
.
Math
.
S
tat.
,
vol
.
33
,
n
o.
1
,
pp
.
1
-
67
,
1962
.
[31]
D.
C.
Hoagli
n
,
“
John W
.
Tukey
and
Dat
a
Ana
l
y
s
is,”
S
tat
ist
ic
al
Sc
ie
nc
e
,
vol. 18, n
o.
3
,
pp
.
311
-
31
8
,
2004
.
[32]
H.
Jiawe
i,
M.
Kam
ber
,
J.
Han,
M.
Kam
ber
,
and
J.
Pei,
"
Da
ta
Mining:
Con
ce
pts
and
T
ec
h
nique
s
,"
Morgan
Kaufmann
,
3
rd
Ed
.
,
2012.
[33]
P.
Nerurka
r,
A.
Shirke,
M.
Chanda
ne,
and
S.
B
hirud,
“
A
novel
heur
isti
c
for
ev
olut
iona
r
y
cl
ust
eri
ng,
”
Proce
d
ia
Comput.
Sc
i.
,
vo
l.
125
,
pp
.
780
-
7
89,
2018
.
[34]
J.
A.
Har
ti
g
an
a
nd
M.
A.
W
ong,
“
Algorit
hm
A
S
136:
A
K
-
Me
ans
Cluste
r
ing
Algorit
hm
,”
Jou
rnal
of
the
Royal
Stat
isti
cal Soc
i
ety.
S
erie
s C
,
(
App
li
ed
Statis
ti
cs)
,
v
ol.
28
,
no
.
1
,
pp
.
100
-
108
,
1979
.
[
35
]
C.
Slamet,
A
.
R
ahman,
M.
A.
R
amdhani,
and
W
.
Dharm
al
aksa
n
a
,
“
Cluste
ring
th
e
ver
ses
of
the
h
ol
y
qur’
an
using
K
-
m
ea
ns a
lgorit
hm
,
”
Asian
Jour
nal
of
Informati
on
Technol
og
y,
vol.
15
,
no
.
24
,
p
p.
5159
-
5162
,
2
016
Evaluation Warning : The document was created with Spire.PDF for Python.