TELKOM
NIKA Indonesia
n
Journal of
Electrical En
gineering
Vol.12, No.5, May 2014, pp
. 3303 ~ 33
1
2
DOI: http://dx.doi.org/10.11591/telkomni
ka.v12i5.4927
3303
Re
cei
v
ed O
c
t
ober 2
4
, 201
3; Revi
se
d Novem
b
e
r
28, 2013; Accept
ed De
cem
b
e
r
16, 2013
Visualization Analysis of Dynamic Evo
l
ution of the
Theme in Improvisation Studies
Peng-bin G
a
o*, Wei
w
e
i
-Wu, Bo Yu
Schoo
l of man
agem
ent, Harb
in Institute of
T
e
chn
o
lo
g
y
, Har
b
in, 15
00
01, P. R. China
*Corres
p
o
ndi
n
g
author, e-ma
i
l
: gaop
en
gbi
nh
i
t
@163.com
A
b
st
r
a
ct
T
he fiel
d of i
m
pr
ovisati
on
i
s
compos
ed
o
f
a mult
ip
licity
of topics l
e
a
d
in
g to a v
a
st array of
ma
na
ge
me
nt li
terature. How
e
ver, the resear
ch does
not
pr
ovid
e a chro
n
o
lo
gica
l pi
ctur
e of the topics
it
addr
esses, ma
king it difficu
lt to devel
op a
n
overvi
ew
of the evol
ution
and tren
ds
in
the literatur
e. T
o
addr
ess th
is is
sue, co-w
ord
a
nalysis
w
a
s
e
m
p
l
oy
ed t
o
rev
eal
patter
n
s
an
d tren
ds
in
the
improvis
atio
n fi
el
d
by me
asuri
ng
t
he associ
atio
n strengths
of ke
yw
ords of re
lev
ant d
o
cu
me
nts
.
Data w
e
re
col
l
ected
fro
m
W
e
b
of Know
le
dge
datab
ase for t
h
e per
iod
19
97-
201
2. Usi
ng th
e co-occ
urrenc
e matrix of k
e
y
w
ords, the res
u
lts
of mu
ltivari
a
te
statistical tech
niq
ues
sh
ow
that the i
m
pr
ovi
s
ation res
ear
c
h
invo
lves
ma
ny fields i
n
cl
ud
in
g
inn
o
vatio
n
, strategy, lear
nin
g
,
change, le
ad
ershi
p
, meta
ph
or, entrepre
n
e
u
rshi
p, capab
ili
ty.In order to trace
the dy
mamic c
han
ges
of the
i
m
pr
ovis
ati
o
n
fi
eld, th
e w
hol
e
peri
od w
a
s furt
her se
par
ated
i
n
to thre
e p
e
rio
d
s:
199
7-20
02, 2
0
03-2
007
an
d 2
008-
201
2.
T
h
e
strategic d
i
a
g
r
am
an
d soci
al
netw
o
rk an
aly
s
is w
a
s used
to
trace the
dy
na
mic
cha
n
g
e
s
o
f
the i
m
provis
a
t
ion r
e
sear
c
h
,
and
resu
lts sh
ow
that i
m
pr
ov
isatio
n fie
l
d
ha
s
some estab
lish
ed rese
arch th
emes an
d it als
o
chan
ges ra
pi
dly to e
m
brac
e
new
themes.
Ke
y
w
ords
: i
m
provis
ation,
bi
blio
metric stu
d
y
, co-w
ord a
n
a
lysi
s, mu
ltivar
iate
statisti
c
a
l analysis,
strat
egic
dia
g
ra
m, socia
l
netw
o
rk analy
s
is, emer
gi
ng trends
Copy
right
©
2014 In
stitu
t
e o
f
Ad
van
ced
En
g
i
n
eerin
g and
Scien
ce. All
rig
h
t
s reser
ve
d
.
1. Introduc
tion
The
word im
provisation
can be i
n
terp
reted to me
an
unfore
s
e
en
or to ta
ke a
c
t
i
on in the
moment. As
a commo
n p
henom
eno
n
of jazz a
nd t
heatre, thi
s
i
dea
ha
s be
e
n
a topi
c to
a
ttract
many schol
ars of ma
nag
e
m
ent fields to
study it.
A major mil
e
stone for research in improvisation
occurre
d
at the Acad
emy of Manage
m
ent meeti
ng
held in 199
5
in Vancouve
r
. Hatch, Barrett
and some oth
e
r sch
o
lars e
x
plore the u
s
e of jazz as a
metapho
r for unde
rsta
ndin
g
org
ani
zatio
nal
and imp
r
ovisation, these
motivated se
veral re
se
arch studie
s
wh
ich, in 19
98,
resulted in
a
spe
c
ial i
s
sue
of Organi
zat
i
on Scien
c
e
devoted to
o
r
gani
zatio
nal
improvi
s
ation
.
Since then,
a
strea
m
of arti
cle
s
ha
s pou
red into the liter
atu
r
e on i
s
sues rangi
ng from many fields.
Past re
sea
r
ch sugg
est
s
that improvisa
t
i
on has be
e
n
beco
m
e a hot research
field in
recent yea
r
s
and m
any
works
sh
ould
to be
don
e to
study d
eepl
y into this to
pic.
Cunh
a e
t
al.
(199
9) reviee
d the gro
w
ing
body of litera
t
ure on o
r
ga
n
i
zation
al improvisation in o
r
de
r to pre
s
e
n
t
an en
comp
a
ssi
ng and
systemati
c
perspective on th
is
co
ncept. An integrative
definition of its
con
s
tru
c
t
wa
s p
r
e
s
e
n
ted
togethe
r
wi
th a n
e
w
way of
mea
s
uri
n
g
this
phen
omen
on
in
orga
nizationa
l setting
s. Th
e arti
cle fu
rt
her
explor
ed
this
con
s
tru
c
t by p
r
e
s
en
ting its tri
gge
rs,
necessa
ry co
ndition
s, influenci
ng
facto
r
s and m
a
jor o
u
tcome
s
[1]. Li et al. (2011
) su
gge
sted t
hat
Improvisation
is related to a host of outcome
variabl
e
s
, includi
ng e
n
trep
ren
eurship, new prod
uct
developm
ent
and in
novatio
n [2]. Based
on the
releva
nt literatures from
the
year 1990 until
20
10,
Hua
ng et al. (2012
) systematically reviewe
d
the studie
s
of orga
nizationa
l improvisati
on,
inclu
d
ing its
definition
s
, chara
c
te
risti
c
s, catego
ries,
measurement
s, trigge
r, influen
cing fa
cto
r
s
and the
out
comes.
The
result
s indi
cat
ed that the
o
ry-buildin
g reli
ed mo
re
on
metapho
rs, the
con
c
e
p
ts were ambig
uou
s,
the system
wa
s in
compl
e
te
and
the empiri
cal stu
d
ies we
re
scarce
[3]. Although there h
a
ve b
een
several a
ttempts to ge
nerali
z
e fin
d
i
ngs i
n
imp
r
ov
isation lite
r
at
ure,
they used lite
r
ature synthe
sis te
chni
que
s, whi
c
h a
r
e
more d
epe
nd
ent on subje
c
tive analysis
an
d
coul
dn’t discl
o
se the multi
p
licit
y of improvisation research.
Specifically, the aim of this arti
cle is
to use co
-word analy
s
i
s
for dete
c
ting and
visuali
z
ing
co
nce
p
tual sub
domain
s
. Qu
antitative
and
qualitative m
easure
s
a
r
e
use
d
in o
r
de
r to
identify the m
o
st p
r
omi
nent
theme
s
. Th
e
study
also in
corpo
r
ate
s
bi
bliometri
c
m
a
ps to
sho
w
, i
n
a
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 5, May 2014: 3303 – 33
12
3304
visual
way, th
e a
s
soci
ation
s
b
e
twe
en th
e mai
n
them
e
s
. At the
sam
e
time, lo
ngit
udinal
map
s
are
use
d
to analy
z
e the chan
g
e
s of theme
s
and fore
ca
st emerging tre
nds for
a subj
ect domai
n.
2. Rese
arch
Metho
d
2.1. Co-
w
o
r
d
Analy
s
is
The
co-wo
r
d
analysi
s
d
r
aws up
on t
he a
s
sumpti
on that a
d
o
cum
ent’s keywords
con
s
titute an
adeq
uate d
e
s
cription
of its content.
T
w
o
keyword
s
co
-o
ccurring
within the
same
pape
r a
r
e a
n
indication of
a link
betwe
en the topi
cs to whi
c
h the
y
refer [4]. T
he p
r
esen
ce
of
many co
-o
ccurren
c
e
s
aro
und the sa
m
e
word or pa
ir of word
s p
o
ints to the locu
s of strat
egic
allian
c
e
withi
n
pa
pers th
a
t
may co
rrespond
to
a re
sea
r
ch
the
m
e.
Co
-word analysi
s
reve
als
pattern
s a
n
d
trend
s i
n
a
spe
c
ific di
sci
pline
by
mea
s
uri
ng th
e a
s
so
ciation
stre
ngths of terms
rep
r
e
s
entativ
e of relevant
publication
s
produ
ce
d
in
this are
a
. The main feat
ure of co-wo
r
d
analysi
s
i
s
th
at it visuali
z
e
s
the i
n
telle
ctual st
ru
cture
of one
sp
ecifi
c
di
scipline
in
to map
s
of th
e
con
c
e
p
tual
space of thi
s
f
i
eld, an
d that
a time
-serie
s of
su
ch
ma
ps
pro
d
u
c
e
s
a tran
ce
of t
he
cha
nge
s i
n
th
is
con
c
e
p
tual
sp
ace [5]. In
this stu
d
y, bi
bliometri
c
s
software
Bibex
cel
wa
s u
s
ed
to
cal
c
ulate th
e
numbe
r of ti
mes t
w
o
key
w
ords ap
pea
r togethe
r in t
he same
pub
lication. T
h
u
s
we
have formed
a co-o
ccurrence matrix
of keyw
ord
s
. For sub
s
eque
nt analysis, in order to
stand
ardi
ze
t
he d
a
ta, avoi
d po
ssible
scale effe
cts,
a
nd
red
u
ce the
num
ber of
ze
ros in
the
mat
r
ix,
the raw
co
-cit
ation matrix wa
s co
nverte
d into a
matri
x
of Pearson’
s co
rrelation
coeffici
ents.
2.2. Multi
v
ari
a
te Statistical Anal
y
s
is
The correlati
on coefficient
s were
analy
z
ed
usin
g th
e statisti
cal
pro
c
ed
ures o
f
cluste
r
analysi
s
, mul
t
idimensi
onal
scalin
g (M
DS), and facto
r
analysi
s
. Hi
era
r
chical clu
s
terin
g
involves
cre
a
ting cl
ust
e
rs that are hiera
r
chi
c
ally nested wi
thi
n
clu
s
ters at earlie
r iterati
ons, in that each
clu
s
ter can b
e
includ
ed a
s
a member of
a larger, mo
re comp
re
hen
sive clu
s
ter a
t
a higher lev
e
l
of similarity. Among ag
glo
m
erative hie
r
archical
met
hod
s, we sel
e
ct the Wa
rd Method. T
h
is
pro
c
ed
ure is
desi
gne
d to optimize the m
i
nimum vari
a
n
ce
within cl
u
s
ters, and it works by joini
n
g
those
gro
u
p
s
or cl
uste
rs th
at result in th
e mini
mum i
n
cre
a
se in the
varian
ce [6]. The
correlati
on
data we
re al
so an
alyze
d
usin
g the mu
ltidimens
i
onal
scali
ng procedure, a dim
ensi
on re
du
ction
techni
que th
at aims
at fitting the ori
g
inal data
i
n
to a lo
w-dim
ensi
onal
sp
a
c
e
su
ch that
the
distortio
n
of
the simila
ritie
s
a
nd dissimi
l
aritie
s am
on
g the
ori
g
inal
data
caused
by redu
ction
in
dimen
s
ion
a
lity is mini
m
i
zed
[7]. T
w
o-dime
nsio
nal
solution
s
were ex
plore
d
in
the
multidimen
sio
nal scali
ng
with the
procedure of
AL
SCAL. Fu
rth
e
rmo
r
e, a
n
e
x
plorato
r
y facto
r
analysi
s
wa
s
con
d
u
c
ted to
asse
ss the
u
nderlyin
g di
m
ensi
o
n
s
am
o
ng the
jou
r
n
a
ls. Th
e p
r
in
ci
pal
comp
one
nts analysi
s
wa
s
used
to extract
fa
cto
r
s.
K
a
ise
r
’s criteri
on a
nd the
scree
test
were
comp
ared to
determine
the extra
c
ted
numb
e
r of
f
a
ctors. After the extracti
on, facto
r
s
were
rotated u
s
ing
the procedu
re of Varima
x rotation. Factor an
alysi
s
can b
e
use
d
to complem
ent
multidimen
sio
nal scalin
g a
nd clu
s
te
ring
displ
a
ys an
d
sho
w
an
entity’s co
ntributi
on to more th
an
one spe
c
ialty. Unlike
clu
s
t
e
r an
alysi
s
, which o
n
ly
assi
gns a
n
entity to one cl
uste
r, the entity can
load o
n
mo
re than o
ne f
a
ctor in a fa
ctor
analy
s
is.
Therefore, the inte
rrel
ationship
s
bet
ween
spe
c
ialtie
s ca
n be ea
sily re
vealed from a
different perspective [8].
2.3. Strate
gic Diagram Analy
s
is
Strategic di
ag
ram devel
ope
d by
the co
-word a
nalysi
s
has a m
e
rit, whi
c
h can ide
n
tify the
evolving tren
ds and relati
onal patterns between th
e topics rep
r
ese
n
ted by clusters [9]. In a
strategi
c dia
g
r
am, X-axis
stand
s for ce
nt
rality and Y-axis stan
d for d
ensity.
Den
s
ity is used to measure the stren
g
th of re
lation
s that make terms in a cl
uster. We
define the de
nsity as follo
wing.
1
)
(
0,
0
N
r
k
D
N
i
N
i
j
j
ij
Whe
r
e
D (k
)
is the
den
sity of clu
s
te
r
k
,
N
i
s
the
num
ber
of keywo
r
ds in
clu
s
ter
k
, an
d
r
ij
is the
relation valu
e
betwee
n
wo
rd i and wo
rd j
which are b
o
t
h within the cluster
k
.
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Visuali
z
ation
Analysis of Dynam
ic Evolu
t
ion of
the Them
e in Im
provisation… (P
eng-bin Ga
o)
3305
Centrality is
use
d
to me
a
s
ure the
extent to
which
a cl
uste
r i
s
con
n
e
c
ted
wi
th other
clu
s
ters. We
define the ce
ntrality as followin
g
.
N
N
M
r
k
C
N
i
N
M
j
ij
)
(
)
(
00
Whe
r
e
C(
k)
i
s
the centralit
y of the clust
e
r
k
,
M
is th
e
numbe
r of al
l keywo
r
d
s
which a
r
e
sele
cted
for c
l
us
tering,
N
is the num
ber of key
w
ords in clu
s
te
r
k
, and
r
ij
is the relation valu
e
between
word
i
within the
cl
uster
k
and word
j
without the c
l
us
ter
k
.T
he
Strategi
c Diag
ram and its
mea
n
ing a
r
e
s
h
ow
n
in
F
i
gu
r
e
1
.
Figure 1. Strategic Di
agra
m
and Its Me
aning
2.4. Social Ne
w
o
rk
Analy
s
is
Social n
e
two
r
k a
nalysi
s
(SNA) is th
e mappin
g
an
d
measurin
g o
f
relation
ship
s amo
n
g
comp
one
nts i
n
a
system.
A netwo
rk in
SNA con
s
ist
s
of a
set of
node
s a
nd li
nks. The
nod
es
rep
r
e
s
ent the
comp
one
nts and the lin
ks sta
nd for
relation
ship
s
betwe
en the
node
s. In this
pape
r, we
structure the ke
yword
s
net
work
of
resea
r
ch on treatm
ent adhe
ren
c
e, in which the
node
s a
r
e th
e key
w
ords
while th
e lin
ks
rep
r
e
s
e
n
t the co
-o
ccu
rre
nce of the
s
e
keyword
s
.
T
o
unde
rsta
nd
t
he stru
cture of
the keyword
network
in literatu
r
e
on treatment
adh
ere
n
ce,
we
evaluate th
e l
o
catio
n
of
ke
yword
s
i
n
the
network
by
measuri
ng th
e centrality of
ea
ch
nod
e a
n
d
the netwo
rk centrali
zatio
n
.The co
mm
unicati
on bet
wee
n
two n
ode
s in a n
e
twork can
be
facilitated, blocked, di
storted or fal
s
ified
by
a node fall
ing bet
ween t
hem
, and therefore the
node
betwe
en the
other two no
des h
a
s
a po
tential to c
ont
rol their
com
m
unication.
Whe
n
a pa
rticula
r
node in
a gro
up is
strategi
cally located
on the
shorte
st com
m
uni
cation path
co
nne
cting pai
rs of
others, that n
ode i
s
in a
ce
ntral
po
sition.
The centrality is define
d
in
terms
of the
degree to
whi
c
h
a node fall
s o
n
the sho
r
test
path betwe
e
n
other
s, and
named a
s
bet
wee
nne
ss ce
ntrality [10].
3. Data
Colle
ction and Pr
epara
t
ion
3.1.
Data Col
l
ection
To retrieve
sufficient ‘‘im
provisation’
rela
ted papers,
the Web of
Science
li
terature
databa
se
is i
n
itially used f
o
r p
ape
r
retri
e
val. In
orde
r to have
sufficient
cove
rag
e
of the
pa
pe
rs,
the following
query ha
s be
en tried: improvisation o
r
improvi
s
ation
a
l or improvi
s
e or improvi
s
i
ng
in the topic. A total of 212 papers wa
s retriev
ed from
the database
coverin
g
the perio
d of 199
7-
2012 a
nd were
sele
cted
as the co-word an
alysi
s
sa
mple. In
Figure
2. the dist
ributio
n of
document
s (Article, Pro
c
e
eding Pa
per
and Review)
from man
age
ment, busi
n
e
ss
and e
c
o
n
o
m
ic
field per yea
r
is sh
own.
From
ea
ch of
these
pap
ers,
autho
r key
w
ords
and
ke
yword
s
plus were sele
cte
d
.
Due
to
the fact that different wo
rds can be u
s
ed for de
scri
bing the sam
e
con
c
e
p
t, it
is ne
ce
ssary
to
stand
ardi
ze
words.
Fo
r ex
ample,
(1) pl
ural fo
rm
s a
r
e sta
nda
rdi
z
e
d
to thei
r
sing
ular fo
rm;
(2
)
firm
perfo
rman
ce,
task pe
rformance, ne
w ventur
e pe
rforma
nce, orga
nizationa
l
perfo
rma
n
ce,
busi
n
e
s
s pe
rforma
nce, j
ob p
e
rfo
r
ma
nce, fi
n
a
n
c
i
a
l pe
rform
a
nce
are
standa
rdi
z
ed
to
perfo
rman
ce;
(3
)
org
ani
zation
al m
e
mory, worki
ng m
e
mo
ry, tran
sa
ctive
memo
ry
are
stand
ardi
ze
d
to memory; (4) tra
n
sfo
r
ma
tional l
ead
ership, strate
gic
leade
rship a
r
e stand
ardize
d
Qua
d
rant
C
e
nt
ral
an
d
de
vel
o
ped
Qua
d
rant
C
e
nt
ral
an
d
u
n
d
evel
ope
d
Qua
d
rant
Per
phe
ral
an
d
devel
ope
d
Qua
d
rant
Per
phe
ral
an
d
un
de
vel
o
ped
Densit
y
Cen
t
ralit
y
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 5, May 2014: 3303 – 33
12
3306
to leade
rshi
p; (5
) kno
w
ledge i
n
ten
s
ive entre
pr
e
neurshi
p
, int
e
rnatio
nal
e
n
trep
ren
eurship,
techn
o
logy e
n
trep
ren
eurship are
sta
ndardiz
ed to
entrep
r
en
eu
rshi
p; and (6) internatio
nal
strategy,
marketing
st
rate
gy, techol
ogy
led
stra
tegy, develo
p
men
t
strate
gy,
busine
s
s
strate
gy
are
stan
da
rdi
z
ed
to
strate
gy. At last, 7
45
keywor
d
s
were
colle
cte
d
an
d fre
que
ncy di
strib
u
tion of
keyword
s
i
s
shown in Fi
gure 3. As
sho
w
n in Ta
ble 1.
50 keywo
r
d
s
with fre
quen
cy more tha
n
5
were ch
osen
as the rese
arch sample fo
r co-wo
r
d an
al
ysis.
Figure 2. Do
cuments Pu
bli
s
he
d from 19
97 to
2012
Figure 3. Fre
quen
cy Di
st
ri
bution of Keywords
Table 1. Top
50 Hig
h
Fre
q
uen
cy Keywo
r
ds
No.
K
e
y
w
or
d
Freque
nc
y
No.
K
e
y
w
or
d
Freque
nc
y
1
Improvisation
94
26
Cr
eativity
12
2
Performance
48
27
D
y
namic capabilit
y
10
3
Product develop
ment
41
28
Sy
s
t
e
m
10
4
Innovation
40
29
F
l
ex
ibility
10
5
Orga
nizational improvisation
36
30
Sensemaking
9
6
Kno
w
ledge
36
31
Identit
y
8
7
Strateg
y
34
32
Capability
8
8
Jazz
32
33
Orga
nizational learning
8
9
Orga
nization
32
34
Decision making
8
10
Management
31
35
Communication
8
11
Environment
26
36
Net
w
ork
8
12
Fi
rm
23
37
Antecedents
8
13
Perspective
22
38
Field
8
14
Learning
18
39
Leadership
7
15
Memory
18
40
Success
7
16
Model
17
41
Orga
nizational change
7
17
Metaphor
17
42
Experience
7
18
Entrepre
neurship
15
43
Absorptive capacity
7
19
Evolution
15
44
Team
7
20
Competitive advantage
14
45
Market orientatio
n
7
21
Technolog
y
14
46
Transform
ation
6
22
Industr
y
13
47
Uncertaint
y
6
23
Bricolage
13
48
Information tech
nolog
y
6
24
Wor
k
13
49
Complexit
y
6
25
Ti
me
13
50
Impact
6
3.2. Matrix G
e
nera
tion
Specifically b
i
bliometri
c
s software Bi
be
xcel
was u
s
e
d
to cal
c
ulate
the numb
e
r
of times
two keywo
r
d
s
appe
ar to
get
her i
n
the
sa
me pu
blicatio
n. Thu
s
, we
have form
ed
a co
-o
ccurre
nce
matrix of 50×50 key
w
ords.
In the cell of keywo
r
d
X
and key
w
ord
Y
we put the co-o
ccu
rre
n
c
e
freque
ncy of
X
and
Y.
The diago
nal value
s
of the matrix we
re treated
as mi
ssi
ng data. T
h
e
matrix wa
s tran
sform
ed i
n
to a co
rrela
t
ion matr
ix b
y
using Pea
r
son’
s correla
t
ion coeffici
e
n
t
indicating the
similarity and
dissimil
a
rity of
each
keyword pai
r, whi
c
h is sh
own in Table 2.
0
5
10
15
20
25
30
Do
c
u
m
e
nt
s
19
97
19
98
19
99
20
00
20
01
20
02
20
03
20
04
20
05
20
06
20
07
20
08
20
09
20
10
20
11
20
12
Y
ear
54
8
92
42
25
13
1
3
0
10
0
20
0
300
40
0
50
0
60
0
1
2
3-
5
4
-
1
0
1
1
-
20
mo
re
th
a
n
21
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Visuali
z
ation
Analysis of Dynam
ic Evolu
t
ion of
the Them
e in Im
provisation… (P
eng-bin Ga
o)
3307
Table 2. The
Ra
w Co
-citati
on Matrix
and
Correlatio
n Matrix (sectio
n
)
1 2 3 4 5 6 7 8 9
10
Absorptive
capacity
0
(1.00)
0.503
0.389
0.293
0.186
0.468
0.192
0.374
0.463
0.713
Antecedents
1
0
(1.00)
0.364
0.358
0.613
0.64
0.551
0.654
0.505
0.650
Bricolage
1
0
0
(1.00)
0.573
0.321
0.371
0.511
0.421
0.648
0.397
Capability
0
0
0
0
(1.00)
0.387
0.346
0.509
0.527
0.512
0.268
Communication
0 1 0 0
0
(1.00)
0.44
0.511
0.645
0.346
0.378
Competitive
advantage
2 2 1 1 0
0
(1.00)
0.584
0.415
0.518
0.618
Complexit
y
0 0 0 0 0 0
0
(1.00)
0.566
0.499
0.337
Creativit
y
0 2 0 0 1 0 1
0
(1.00)
0.502
0.426
Decision
making
0 1 1 1 1 1 0 0
0
(1.00)
0.425
D
y
namic
capabilit
y
3 1 1 0 0 2 0 0 0
0
(1.00)
Note. Th
e d
a
t
a above
diag
onal lin
e me
a
n
the
co
rrel
a
tion coefficie
n
t
and the
dat
a blo
w
di
ago
nal
line mean the
frequen
cy
4. Results a
nd Analy
s
is
4.1. Results of Multiv
aria
te Sta
t
istical
Analy
s
is
A hierarchi
c
al
cluste
r analy
s
is
with Wa
rd
’s
method a
n
d
multidimen
sion
al scaling
(MDS)
with ALS
C
AL metho
d
were
ca
rri
ed
out, and
the
re
sults were sho
w
n i
n
Figure 4
an
d 5,
respe
c
tively. The st
re
ss v
a
lue (0.198
9
0
, lowe
r than
an acce
ptabl
e value 0.2
)
and R2 (0.75
809
for two
-
dim
e
n
s
ion
s
) sho
w
e
d
an
outsta
n
d
ing fit for th
e data. T
he
result
s of fa
ctor a
nalysi
s
were
sho
w
n in Ta
b
l
e 3 and 4.
Clu
s
ter a
naly
s
is
and m
u
ltidimen
sion
al
scaling
rea
c
h
a uniform co
nclu
sio
n
. As
a re
sult,
five large the
m
e group
s e
m
erg
ed fro
m
right to le
ft o
n
the ho
rizont
al axis. Them
e 1 focu
s o
n
the
resea
r
ch ab
out strate
gy and innov
ation [
11-1
2
], which inv
o
lves produ
ct developm
ent,
kno
w
le
dge, p
e
rform
a
n
c
e, environ
ment,
man
agem
en
t, model, m
a
rket
ori
entati
on, comp
etitive
advantag
e, unce
r
tainty, industry, su
cce
ss an
d fl
exibility. Theme 2 focus on the
rese
arch ab
out
learni
ng [1
3], lead
ership
and
ch
ange,
whi
c
h
emp
hasi
z
e
s
th
e
importa
nce o
f
field, syste
m
,
compl
e
xity, experien
c
e,
organi
zation
a
nd te
chn
o
log
y
. Theme
3 f
o
cu
s
on th
e
resea
r
ch a
b
o
u
t
metapho
r [14
-
15], memory
and cre
a
tivity, and includ
es jazz, time, work, sen
s
e
m
akin
g, ident
ity,
team a
nd
co
mmuni
cation.
Them
e
4 fo
cu
s o
n
th
e rese
arch
abo
ut entr
epren
eurship
[16]
an
d
evolution, a
nd e
s
pe
ciall
y
involves
netwo
rk
i
ng,
bricolag
e, firm, deci
s
ion
makin
g
a
nd
transformation. Theme
5 focus on
the
research about c
apability,
such as ab
sorptive capabilit
y,
dynamic
ca
pability, and organi
zatio
nal improv
i
s
ation unde
r the context of information
techn
o
logy.
Figure 4. Multidimen
sion
al Scaling M
a
p
T
hem
e
1
T
hem
e
2
T
hem
e
5
T
hem
e
4
T
hem
e
3
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 5, May 2014: 3303 – 33
12
3308
Figure 5. Hierarchical Clu
s
t
e
r Analysi
s
Table 3. Rota
ted Comp
one
nt Matrix of Factor An
alysi
s
Fac
t
or
1
2
3
4
5
Fac
t
or
1
2
3
4
5
Management
0.922
0.163
0.232
-0.050
0.108
Decision
making
0.437
0.747
-0.245
0.013
-0.047
F
l
ex
ibility
0.921
-0.061
-0.093
-0.120
0.107
Net
w
ork
0.056
0.709
-0.298
0.464
0.100
Competitive
advantage
0.899
0.089
0.006
0.074
-0.205
Transform
ation
-0.422
0.668
-0.196
0.091
0.053
Market
Orientation
0.895
-0.134
-0.160
0.002
-0.244
Capability
0.217
0.627
-0.036
0.290
0.387
Antecedents
0.885
-0.164
0.334
-0.047
-0.022
Strateg
y
0.443
0.594
0.452
-0.262
0.149
Environment
0.882
0.139
0.317
0.053
0.149
Technolog
y
0.394
0.571
0.36
0.204
0.3
Uncertaint
y
0.875
0.265
-0.051
-0.103
-0.026
Jazz
0.405
-0.104
0.847
-0.047
0.028
Product
development
0.866
0.068
0.341
0.036
-0.155
Ti
me
-0.068
0.142
0.82
-0.432
0.162
Improvisation
0.852
-0.06
0.159
-0.087
-0.003
Metaphor
0.016
-0.159
0.816
0.240
0.332
Innovation
0.838
0.179
0.408
0.06
0.029
Sensemaking
0.058
-0.365
0.786
0.112
-0.094
Performance
0.813
0.206
0.434
0.104
-0.174
Team
0.537
-0.279
0.666
-0.321
-0.001
Kno
w
ledge
0.729
0.252
0.303
0.249
-0.243
Experience
0.186
0.459
0.644
0.137
0.146
Orga
nization
0.725
0.149
0.205
0.379
0.272
Meomor
y
0.619
-0.055
0.629
-0.017
0.204
D
y
namic
capability
0.723
0.103
-0.089
0.022
-0.539
Cr
eativity
0.46
-0.073
0.627
0.155
0.357
Model
0.699
0.377
0.196
0.364
-0.234
Wor
k
0.298
0.025
0.578
0.031
-0.219
Success
0.680
0.210
0.066
-0.254
0.236
Orga
nizational
learning
0.265
0.077
0.140
0.838
-0.127
Impact
0.676
0.157
0.388
-0.034
-0.076
Field
-0.040
0.292
-0.066
0.781
0.138
Communica-
tion
0.618
-0.159
0.544
0.147
0.268
Sy
s
t
e
m
-0.222
0.209
0.084
0.731
0.146
Perspective
0.596
0.244
0.336
-0.591
0.024
Identit
y
-0.135
-0.205
0.532
0.714
0.172
Complexit
y
0.552
0.228
0.038
0.470
0.499
Learning
0.537
-0.042
-0.065
0.709
-0.18
Evolution
0.108
0.967
0.006
0.100
-0.064
Orga
nizational
improvisation
0.285
0.048
0.380
-0.496
-0.489
Entrepre
neur
-
ship
-0.017
0.927
-0.2
0.021
-0.043
Absor
p
tive
capability
0.339
0.160
-0.087
0.075
-0.857
Fi
rm
0.260
0.894
0.099
-0.069
-0.141
Leadership
0.458
0.268
0.233
0.280
0.660
Bricolage
-0.130
0.88
-0.113
0.325
0.121
Information
technolog
y
0.246
-0.439
-0.219
0.088
-0.614
Industr
y
0.215
0.749
0.136
-0.359
0.083
Orga
nizational
change
0.487
0.271
0.212
0.188
0.489
Them
e 1
The
m
e 2
Them
e 5
Them
e 4
Them
e 3
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Visuali
z
ation
Analysis of Dynam
ic Evolu
t
ion of
the Them
e in Im
provisation… (P
eng-bin Ga
o)
3309
Table 4. Total
Varian
ce Explaine
d of Factor Analy
s
is
Initial Eigenvalue
s
Extraction Sums
of
Squared Lo
ading
s
Rotation Sums of
Squared Lo
ading
s
Total
% of Variance
C
umulative %
Total
%
of Varianc
e
C
umulative %
Total
%
of Variance
Cumulative %
1 18.865
37.729
37.729
18.865
37.729
37.729
15.725
31.450
31.450
2 8.432
16.863
54.592
8.432
16.863
54.592
8.239
16.478
47.928
3 5.910
11.819
66.412
5.910
11.819
66.412
7.422
14.844
62.773
4 4.828
9.656
76.067
4.828
9.656
76.067
5.364
10.729
73.501
5 2.634
5.267
81.335
2.634
5.267
81.335
3.917
7.833
81.335
Based
on
the
co
rrelation
m
a
trix, we
co
n
ducte
d a fa
ct
or a
nalysi
s
wi
th a Vari
max
rotation
to extract the
key con
c
e
p
tual theme
s
i
n
the impr
ovi
s
ation field.
Table 4
sho
w
s t
hat
six
f
a
ct
or
s
are extracte
d
with 81.33
5
%
of the explained va
rian
ce. The
re
sul
t
s of factor
a
nalysi
s
different
from the
out
comes of two
approa
che
s
a
bove, but th
e
facto
r
1, 2,
3
,
4 an
d 5
ba
sically reflect t
h
e
same
re
sea
r
c
h
st
ru
ct
ur
e.
4.2. Results of Stra
tegic
Diagram An
aly
s
is
Based o
n
the comp
utatio
nal formula
of
density and cent
rality, we can obt
ain the
strategi
c dia
g
r
am
s ba
sed o
n
times cited
of diffe
rent pe
riod
s, whi
c
h a
r
e sh
own in F
i
gure 6
-
9.
Figure 6. Strategic
Di
agra
m
s of 1997
-2
012
Figure 7. Strategic
Di
agra
m
s of 1997
-2
002
Figure 8. Strategic
Di
agra
m
s of 2002
-2
007
Figure 9. Strategic
Di
agra
m
s of 2008
-2
012
Figure 6
sho
w
s th
e results of the
wh
o
l
e per
i
od (19
97-2
012
).
Be
cau
s
e of
its strategi
c
positio
n (u
pp
er-right
quad
rant), the
m
e
1
wa
s
id
enti
f
ied as th
e
motor-t
heme
of the pe
ri
od.
Similarly, because of its hi
gh/m
ediu
m
centrality and l
o
w de
nsity (l
owe
r
-right qu
adra
n
t) them
e 4
and 5 we
re
rega
rd
ed as
gene
ral ba
si
c theme
s
with
stron
g
external inte
rco
nne
ction but
low
con
c
e
p
tual d
e
velopme
n
t. Ho
wever, the
strate
gic
po
sition of theme
2 and
3 (l
ower-l
eft quad
ra
nt),
whi
c
h
had
a
low d
e
n
s
ity and
lo
w
ce
ntrality,
i
ndicated that they were eith
er eme
r
gin
g
or
disa
ppe
arin
g
themes. Fi
gu
re 7
sh
ows t
he results of
the first p
e
ri
od (1997
-2
00
2), in thi
s
sta
ge,
t
h
em
e1
t
heme
2
th
eme3
the
m
e4
t
h
em
e5
0
5
10
15
20
Dens
i
t
y
.4
.45
.5
.55
.6
C
e
n
t
r
a
lit
y
th
em
e1
th
em
e2
t
h
em
e3
t
h
em
e4
0
2
4
6
8
Dens
i
t
y
.2
2
.2
4
.2
6
.28
.3
C
e
n
t
r
a
lit
y
th
em
e1
th
em
e2
the
m
e3
th
em
e4
th
eme5
0
2
4
6
8
De
nsi
t
y
.1
.15
.2
.2
5
C
e
n
t
r
a
lit
y
th
eme1
th
em
e2
th
eme3
t
heme
4
th
eme5
0
5
10
15
De
nsi
t
y
.2
5
.3
.3
5
.4
.4
5
C
e
n
t
r
a
lit
y
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 5, May 2014: 3303 – 33
12
3310
theme 1was
still the motor-theme. The
m
e 3 and 4
were re
ga
rded
as gen
eral ba
sic them
es, a
n
d
theme 2 wa
s emergin
g
theme. At the
same
time, theme 5 wa
s
not appea
rin
g
. Figure 8 shows
the results of the second p
e
riod
(20
03-2
007), this
sta
ge wa
s simil
a
r with the re
sult of the who
l
e
perio
d, which
indi
cated
th
at the p
o
sitio
n
of the
m
e
2
and
4
wa
s
cha
nge
d, an
d them
e 5
was
appe
arin
g fro
m
the first stage to the se
con
d
stag
e. Figure 9 sho
w
s the re
sults of the second
perio
d (2008
-201
2), the
p
o
sition
of the
m
e 4 cha
n
g
ed from l
o
we
r-right qu
ad
rant to lower-l
e
ft
quad
rant.
The re
sult
s a
bove sh
ow t
hat the po
siti
on of theme
1, 3 and 5
wa
s stability
and the
postin
g
of th
eme 2
and
4
wa
s chan
ge
d greatly, wh
ic
h in
dicates
that the hot t
opic
of different
perio
d wa
s sli
ghtly different.
4.3. Results of Social Ne
tw
o
r
k Analy
s
is
In orde
r to g
r
asp the ove
r
all co-wo
r
d
analysi
s
, we
analyzed
key
w
ords
ba
sed
on the
whol
e p
e
rio
d
(19
9
7
-
20
12).
Then
we div
i
ded th
e
whol
e pe
riod
into
three
pa
rts,
so that
we
ca
n
identify the dynamic
cha
n
ges d
u
ri
ng th
ese th
ree p
e
riods. In the figure
s
of the
co-wo
r
d n
e
twork,
the
si
ze
of d
o
ts
m
ean
s
t
he scale of degree ce
ntrality, and the
si
ze
of lin
e
s
m
ean
s th
e
tie
stren
g
th of th
e keywords,
and different
colo
r me
a
n
s
the theme. T
he co
-word n
e
tworks of fo
ur
perio
ds
we
re
sho
w
n in Fig
u
re 10
-1
3.
Figure 10. Co
-wo
r
d
Network of 1997
–20
12
Figure 11. Co
-wo
r
d
Network of 1997
–20
02
A
B
S
O
R
PT
I
V
E
C
A
PA
C
I
T
Y
ANT
E
C
E
D
E
NT
S
BR
I
C
O
L
A
G
E
CA
P
A
B
I
L
I
T
Y
CO
M
M
U
N
I
CA
T
I
O
N
C
O
M
P
E
T
IT
IV
E
A
D
V
A
N
T
A
G
E
CO
M
P
L
E
X
I
T
Y
C
R
E
A
TI
V
I
TY
DE
C
I
S
I
O
N
M
A
K
I
NG
D
Y
N
A
M
I
C CA
P
A
B
I
L
I
T
Y
EN
T
R
EP
R
E
N
E
U
R
SH
I
P
EN
V
I
R
O
N
M
EN
T
EV
O
L
U
T
I
O
N
EX
P
E
R
I
EN
C
E
FIE
L
D
FIR
M
F
L
E
X
IB
IL
IT
Y
I
D
E
N
TI
TY
IMP
A
C
T
IMP
R
OV
I
S
A
T
ION
IN
D
U
S
T
R
Y
I
N
F
O
R
M
AT
I
O
N T
E
C
H
NO
L
O
GY
I
NNO
VA
T
I
O
N
JA
Z
Z
KN
OW
L
E
D
G
E
LEA
D
ERSH
I
P
L
E
AR
NI
NG
M
A
NAGE
M
E
NT
MA
R
K
E
T
OR
IE
N
T
A
T
ION
ME
MO
R
Y
ME
T
A
P
H
OR
MO
D
E
L
NE
T
W
O
R
K
O
R
GANI
Z
A
T
I
O
N
O
R
GA
NI
Z
A
T
I
O
N
A
L
C
H
ANGE
OR
G
A
N
I
Z
A
T
I
ON
A
L
IM
P
R
OV
IS
A
T
ION
O
R
GA
NI
Z
A
T
I
O
N
A
L
L
E
AR
NI
NG
PE
R
F
O
R
M
A
N
C
E
PE
R
S
PE
C
T
I
V
E
PR
O
D
U
C
T
D
E
V
E
L
O
PM
E
N
T
SEN
S
EM
A
K
I
N
ST
RA
T
E
G
Y
SU
C
C
ESS
SY
ST
EM
TE
A
M
T
E
C
H
NO
L
O
GY
TI
M
E
TR
A
N
S
F
O
R
M
A
TI
O
N
UN
C
E
R
T
A
I
N
T
Y
WOR
K
ANT
E
CE
DE
NT
S
C
O
MM
U
N
IC
A
T
IO
N
CO
M
P
L
E
X
I
T
Y
C
R
E
A
TI
V
I
TY
DE
C
I
S
I
O
N
M
A
K
I
N
G
EN
V
I
R
O
N
M
EN
T
EV
O
L
U
T
I
O
N
EX
P
E
R
I
E
N
C
E
FI
R
M
F
L
E
X
I
B
IL
IT
Y
I
D
E
N
TI
TY
I
M
P
R
O
V
IS
A
T
IO
N
IN
D
U
S
T
R
Y
IN
N
O
V
A
T
I
ON
JA
ZZ
KN
O
W
L
E
D
G
E
LEA
R
N
I
N
G
MA
N
A
G
E
ME
N
T
ME
M
O
R
Y
ME
T
A
P
H
O
R
MO
D
E
L
NE
T
W
O
R
K
O
R
GA
NI
Z
A
T
I
O
N
PE
R
F
O
R
M
A
N
C
E
PE
RSPE
C
T
I
V
E
PR
O
D
U
C
T
D
E
V
E
LO
PM
E
N
T
ST
RA
T
E
G
Y
SU
C
C
ES
S
SY
ST
E
M
TE
C
H
N
O
L
O
G
Y
TI
M
E
T
R
A
N
S
F
OR
MA
T
I
ON
UNC
E
R
T
A
I
N
T
Y
WO
R
K
Them
e 4
Them
e 1
Them
e 2
T
hem
e
3
Them
e 5
Them
e 4
Them
e 1
Them
e 2
T
hem
e
3
Them
e 5
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Visuali
z
ation
Analysis of Dynam
ic Evolu
t
ion of
the Them
e in Im
provisation… (P
eng-bin Ga
o)
3311
Figure 12. Co
-wo
r
d
Network of 2003–
20
07
Figure 13. Co
-wo
r
d
Network of 2008
-20
1
2
The
results
of figure
s
a
bov
e indi
cate th
a
t
im
provisatio
n re
se
arch
often involve
s
th
eme 1,
2, 3 and
4 in
1997
-20
02,
and involve t
heme
1, 2,
3, 4, 5 in 2
003
-200
7 an
d 2
0
08-2
012,
whi
c
h
indicate that t
heme
1, 2, 3
and
4 a
r
e
hot
topics
all the
time. At the
same
time, th
e keywo
r
d
s
o
f
theme 1
hav
e the hig
h
centrality (bi
g
circle
point) and h
a
ve
many rel
a
tio
n
shi
p
with
other
keyword
s
. Th
e re
sults of
social n
e
two
r
k analysi
s
are
similar to th
e re
sults of
strategic
diag
ram
analysi
s
.
5. Conclusio
n
Based
on
co
-wo
r
d
an
alysi
s
, multivari
a
te stati
s
tical
a
nalysi
s
strate
gic
diag
ram
analysi
s
and soci
al n
e
twork a
nalysis, this
stud
y produ
c
ed
clear, cohe
ren
t
and rea
s
o
n
able result
s. The
resea
r
ch of i
m
provi
s
ation
involves ma
n
y
fields
in
clud
ing entrep
r
en
eurship, le
arn
i
ng, innovatio
n,
metapho
r, strategy, etc. At the
sam
e
time, the hot topic of diffe
rent peri
o
d
s
can be diffe
re
nt. In
the future, we
can enla
r
ge
the sou
r
ces o
f
public
ation
and use othe
r bibliomet
r
ics app
roa
c
h
e
s to
learn m
o
re a
bout the intell
ectual
st
ru
ctu
r
e of improvi
s
ation re
sea
r
ch.
A
B
S
O
R
P
T
I
V
E
CAP
A
CI
T
Y
AN
T
E
CE
D
E
N
T
S
BR
I
C
O
L
A
G
E
CA
P
A
B
I
L
I
T
Y
C
O
M
M
U
N
IC
A
T
IO
N
C
O
MP
E
T
IT
IV
E
A
D
V
A
N
T
A
G
E
CO
M
P
L
E
X
I
T
Y
C
R
E
A
TI
V
I
TY
DE
C
I
S
I
O
N
M
A
K
I
N
G
DY
N
A
M
I
C
C
A
PA
B
I
L
I
T
Y
EN
T
R
EP
R
E
N
E
U
R
SH
I
P
EN
V
I
R
O
N
M
EN
T
EV
O
L
U
T
I
O
N
EX
P
E
R
I
E
N
C
E
FI
E
L
D
FI
R
M
F
L
E
X
I
B
IL
IT
Y
ID
E
N
T
I
T
Y
IMP
A
C
T
IMP
R
OV
IS
A
T
ION
IN
D
U
S
T
R
Y
I
N
F
O
R
M
A
T
I
O
N T
E
C
H
NO
L
O
GY
I
NNO
V
A
T
I
O
N
JA
Z
Z
KN
O
W
L
E
D
G
E
LE
A
D
ER
SH
I
P
LE
A
R
N
I
N
G
MA
N
A
G
E
ME
N
T
M
A
R
K
E
T
OR
IE
N
T
A
T
ION
ME
M
O
R
Y
ME
T
A
P
H
OR
MOD
E
L
NE
T
W
O
R
K
OR
G
A
N
I
Z
A
T
I
ON
O
R
GA
NI
Z
A
T
I
O
N
A
L
C
H
A
NGE
O
R
G
A
N
I
Z
A
T
I
ON
A
L
IM
P
R
O
V
IS
A
T
I
O
N
O
R
GAN
I
Z
A
T
I
O
N
AL
L
E
A
R
NI
NG
PER
F
O
R
M
A
N
C
E
PE
R
S
PE
C
T
I
V
E
PR
O
D
U
C
T
DE
V
E
L
O
PM
E
N
T
SEN
S
EM
A
K
I
N
G
ST
RA
T
E
G
Y
SU
C
C
ESS
SY
ST
EM
TE
A
M
T
E
CHNO
L
O
GY
TI
M
E
TR
A
N
S
F
O
R
M
A
TI
O
N
UN
C
E
R
T
AI
N
T
Y
WO
R
K
A
B
S
O
R
P
TI
V
E
C
A
P
A
C
I
TY
ANT
E
CE
D
E
N
T
S
BR
I
C
O
L
AG
E
CA
P
A
BI
L
I
T
Y
C
O
M
M
U
N
IC
A
T
IO
N
C
O
MP
E
T
IT
IV
E
A
D
V
A
N
T
A
G
E
CO
M
P
L
E
X
I
T
Y
C
R
E
A
TI
V
I
TY
DE
C
I
S
I
O
N
M
A
K
I
N
G
DY
N
A
M
I
C
C
A
PA
B
I
L
I
T
Y
EN
T
R
E
P
R
E
N
E
U
R
S
H
I
P
EN
V
I
RO
N
M
E
N
T
EV
O
L
U
T
I
O
N
EX
PERI
E
N
C
E
FI
E
L
D
FIR
M
F
L
E
X
IBIL
IT
Y
I
D
E
N
TI
TY
IM
P
A
C
T
IM
P
R
O
V
I
S
A
T
ION
IN
D
U
S
T
R
Y
I
N
F
O
R
M
AT
I
O
N T
E
CH
N
O
L
O
GY
I
NNO
VAT
I
O
N
JA
Z
Z
KN
O
W
L
E
D
G
E
LE
A
D
E
R
S
H
I
P
L
E
A
R
NI
NG
MA
N
A
G
E
ME
N
T
MA
R
K
E
T
O
R
IE
N
T
A
T
IO
N
ME
M
O
R
Y
ME
T
A
P
H
OR
MO
D
E
L
NE
T
W
O
R
K
O
R
GA
NI
Z
A
T
I
O
N
O
R
G
A
NI
Z
A
T
I
O
N
AL
CH
A
NGE
OR
G
A
N
I
Z
A
T
I
ON
A
L
IMP
R
OV
IS
A
T
I
O
N
O
R
GA
NI
Z
A
T
I
O
N
AL
L
E
AR
NI
NG
PERF
O
R
M
A
N
C
E
PERSPE
C
T
I
V
E
PRO
D
U
C
T
D
E
V
E
LO
PM
EN
T
SE
N
S
E
M
A
K
I
N
G
ST
RA
T
E
G
Y
SU
C
C
E
SS
SY
S
T
E
M
TE
A
M
TE
C
H
N
O
L
O
G
Y
TI
M
E
TR
A
N
S
F
O
R
M
A
TI
O
N
UNC
E
R
T
A
I
N
T
Y
WO
R
K
Them
e 4
Them
e 1
Them
e 2
T
hem
e
3
Them
e 5
Them
e 4
Them
e 1
Them
e 2
T
hem
e
3
Them
e 5
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 5, May 2014: 3303 – 33
12
3312
Ackn
o
w
l
e
dg
ements
This wo
rk was su
ppo
rte
d
by
the
National Natu
ral
Scie
nce Found
ation of
Chi
na
(No.7
097
208
9; No.710
020
61; No. 712
7
2175
).
Referen
ces
[1]
Cun
ha MPE, Cun
ha JVD, Kamoch
e K. Organ
izat
io
na
l
improvisati
on: W
hat, w
h
en,
h
o
w
an
d w
h
y.
In
te
rn
a
t
i
o
n
a
l
Jo
u
r
na
l
o
f
Ma
nag
em
en
t R
e
vi
ews
.1999; 1(3): 299-
341.
[2]
Li H
D
, Li
n Z
Y
. Organiz
a
tio
n
a
l
Improvis
atio
n
:
a dom
ai
n re
mains to
b
e
fu
rther studi
ed.
An Academ
ic
Editio
n of Man
age
ment.
201
1
;
3(2): 45-58.
[3]
Jishe
ng HY
P. Organiz
a
tion
al
Impr
ovisati
on:
status qu
o an
d
future.
Ch
ines
e Jour
nal
of M
ana
ge
me
nt.
201
2; 9(7): 108
4-10
91.
[4]
Cambros
i
o A,
Limo
ges C,
C
ourtia
l
JP, et
al.
Histor
i
cal
s
c
ientom
etrics?
Mapp
in
g ov
er
70
ye
ars
of
biol
og
ical saf
e
ty rese
arch
w
i
t
h
co
w
o
rd a
nal
ys
is.
Sciento
m
etr
i
cs.
1993; 2
7
(2
): 119-14
3.
[5]
Ding
Y, C
h
o
w
dhur
y G G, F
o
o S. Bi
bli
o
metr
ic cartogr
ap
h
y
of inform
ation
retrieval
res
ear
ch b
y
usi
n
g
co-
w
ord ana
l
y
s
i
s.
Informati
on process
i
ng
& ma
na
ge
me
nt.
200
1; 37(6): 81
7-84
2.
[6]
Szekel
y GJ, R
i
zzo ML. H
i
erar
chical
clust
e
rin
g
via
jo
int b
e
tw
e
e
n
-
w
ith
i
n
di
stances: E
x
ten
d
in
g W
a
rd'
s
minimum var
i
an
ce method.
Journ
a
l of classi
fication.
20
05;
22(2): 15
1-1
8
3
.
[7]
Liu Z
.
Visu
aliz
ing th
e i
n
tell
e
c
tual structure
in
ur
ban
stud
ies: A jo
urn
a
l
co-citatio
n a
n
a
l
y
s
is (19
92-
200
2).
Sciento
m
etrics.
20
05;
62(3): 38
5-4
0
2
.
[8]
W
h
ite HD, McCain KW
. Visu
alizi
ng
a disci
pl
ine: An
a
u
thor
co-citatio
n an
al
ys
is
of informat
i
on sci
enc
e
,
197
2-19
95.
Jo
urna
l of the America
n
Society
for Informati
on
Scienc
e
.19
98; 49 (4): 327-
35
5.
[9]
Bangr
ae
L, Yo
ng-il
J. Ma
ppi
n
g
Kor
ea’s
n
a
ti
ona
l R&
D d
o
m
a
in
of ro
bot te
chno
log
y
b
y
u
s
ing
the c
o
-
w
o
rd a
nal
ys
is.
Sciento
m
etrics
. 2008; 77(
1):.3–19.
[10]
F
r
eeman L
C
. A set of measures of
centralit
y
base
d
on b
e
t
w
een
ness. Soci
o
metry.
1977;
40(1): 35-
41.
[11]
Moorma
n C,
Miner
AS. T
he conv
erg
ence
of
pl
an
nin
g
an
d
e
x
ecuti
on: I
m
provisati
o
n
i
n
n
e
w
pro
duct
deve
l
opm
ent.
Journ
a
l of Mark
eting.
1
9
9
8
; 62(
7):.1–20.
[12]
Moorma
n C,
Miner AS. Or
gan
izatio
na
l
i
m
provisati
on and org
aniz
a
ti
ona
l
memor
y
.
Acade
my
of
Mana
ge
me
nt Review
. 199
8; 23(4):.69
8
–
723
.
[13]
Miner AS, Ba
ssoff P, Moorman C. Orga
nizati
ona
l imp
r
ovisati
on a
n
d
learn
i
n
g
: A field stu
d
y
.
Administr
a
tive
Scienc
e Quarterl
y
.
2
0
0
1
; 46(
6):.304–
33
7.
[14]
KE W
e
ick. Introductor
y
Essa
y: Improv
isat
io
n as
a Mi
ndset
for Organ
izati
ona
l An
al
ysis.
Organi
z
a
t
i
o
n
Scienc
e.
19
98; 9(5): 543-
55
5.
[15]
Kamoch
e K, Pina e Cu
nh
a M,
Viera da
Cun
ha J. T
o
w
a
rds
a T
heor
y
of Organ
izatio
na
l Improvis
ation
:
Look
ing B
e
yon
d
the Jazz Metaph
or.
Journ
a
l
of Manag
e
m
e
n
t
Studies.
200
3
;
40(12): 20
23–
205
1.
[16]
Baker T
,
Mine
r AS, Eesle
y
DT
.
Improvisin
g fi
rms:
Bricol
age, accou
n
t givin
g
and
im
provis
ation
a
l
compete
n
cies i
n
the foun
din
g
process.
Res
e
arch Pol
i
c
y
. 2
0
03; 32(2): 2
55–
276.
Evaluation Warning : The document was created with Spire.PDF for Python.