Pajek Large Networks Paper
Pajek Large Networks Paper
PAJEK
ANALYSIS AND VISUALIZATION
OF LARGE NETWORKS
Vladimir Batagelj, Andrej Mrvar
ISSN 1318-4865
1 Introduction
2 Applications
There exist several sources of large networks that are already in machine-
readable form. Pajek provides tools for analysis and visualization of such
networks and is applied by researchers in different areas: social network analy-
sis [11], chemistry (organic molecule), biomedical/genomics research (protein-
receptor interaction networks) [59], genealogies [57,28], Internet networks [22],
citation networks [42], diffusion networks (AIDS, news), analysis of texts [17],
data-mining (2-mode networks) [14], etc. Although it was developed primarily
for analysis of large networks it is often used also for, especially visualization
of, small networks.
In last months (end of 2002) we had over 500 downloads of Pajek per
month.
Pajek is also used at several universities: Ljubljana, Rotterdam, Stanford,
Irvine, The Ohio State University, Penn State, Wisconsin/Madison, Vienna,
Pajek, Analysis and Visualization of Large Networks 3
3 Algorithms
To support the design goals we implemented several algorithms known from
the literature (see section 4.2), but for some tasks new, efficient algorithms,
suitable to deal with large networks, had to be developed. They mainly pro-
vide different ways to identify interesting substructures in a given network.
uRv ≡ v cites u
where
1 u=s
n− (u) =
n− (v)
P
v:vRu otherwise
4 Vladimir Batagelj and Andrej Mrvar
and
+ 1 u=t
n (u) = P +
v:uRv n (v) otherwise
This is the basis of an efficient algorithm for computing n(u, v) – after the
topological sort [TF 2.2] of the st-digraph we can compute, using the above
relations in topological order, the weights in time of order O(m), m = |R|.
The topological order ensures that all the quantities in the right sides of the
above equalities are already computed when needed.
The Hummon and Doreian indices are defined as follows:
• search path link count (SPLC) method: wl (u, v) equals the number of
“all possible search paths through the network emanating from an origin
node” through the arc (u, v) ∈ R.
• search path node pair (SPNP) method: wp (u, v) “accounts for all con-
nected vertex pairs along the paths through the arc (u, v) ∈ R”.
We get the SPLC weights by applying the SPC method on the network
obtained from a given standardized network by linking the source s by an arc
Pajek, Analysis and Visualization of Large Networks 5
to each nonminimal vertex from U ; and the SPNP weights by applying the
SPC method on the network obtained from the SPLC network by additionally
linking by an arc each nonmaximal vertex from U to the sink t.
The values of counters n(u, v) form a flow in the citation network – the
Kirchoff ’s vertex law holds: For every vertex u in a standardized citation
network incoming flow = outgoing flow :
X X
n(v, u) = n(u, v) = n− (u) · n+ (u)
v:vRu v:uRv
The weight n(t, s) equals to the total flow through network and provides a
natural normalization of weights
n(u, v)
w(u, v) = ⇒ 0 ≤ w(u, v) ≤ 1
n(t, s)
In large networks the values of weights can grow very large. This should
be considered in the implementation of the algorithms.
In Figure 2 the main subnetwork obtained as an edge-cut at level 0.001
of the citation network (n = 4470, m = 12731) on SOM (self-organizing
maps) literature is presented. The picture is exported in SVG with addi-
tional Javascript support that provides the user with options to inspect the
subnetwork at different predetermined levels.
E.Arkin
J.Mitchell I.Tollis
M.Bern A.Garg
L.Vismara
D.Eppstein
G.diBattista
M.Goodrich
R.Tamassia
G.Liotta
D.Dobkin S.Suri J.O’Rourke
J.Vitter
J.Hershberger
B.Chazelle F.Preparata
R.Seidel B.Aronov L.Guibas J.Snoeyink
H.Edelsbrunner
M.Sharir P.Agarwal
R.Pollack
J.Pach D.Halperin P.Gupta
M.Smid R.Janardan
E.Welzl M.Overmars P.Bose
J.Boissonnat
M.vanKreveld O.Devillers
J.Matousek J.Majhi
M.Yvinec
C.Yap M.deBerg
J.Schwerdt
O.Schwarzkopf G.Toussaint M.Teillaud
J.Czyzowicz
J.Urrutia
C.Icking
R.Klein
core number of vertex v is the highest order of a core that contains this vertex.
The degree deg(v) can be: in-degree, out-degree, in-degree + out-degree, etc.,
determining different types of cores.
In Figure 3 an example of cores decomposition of a given graph is pre-
sented. From this figure we can see the following properties of cores:
• The cores are nested: i < j =⇒ Hj ⊆ Hi
• Cores are not necessarily connected subgraphs.
Our algorithm for determining the cores hierarchy is based on the follow-
ing property [16]:
If from a given graph G = (V, E) we recursively delete all vertices,
and edges incident with them, of degree less than k, the remaining
graph is the k-core.
Its outline is given in Algorithm 1. In the refinements of the algorithm we
have to provide efficient implementations of sorting the degrees and their
reordering. Since the values of degrees are in the range 0..n − 1 we can order
them in O(n) using a variant of bin sort; and the update of the ordering can
be done in a constant time. For details see [18].
The cores, because they can be determined very efficiently, are one among
few concepts that provide us with meaningful decompositions of large net-
works. We expect that different approaches to the analysis of large networks
Pajek, Analysis and Visualization of Large Networks 7
can be built on this basis. For example: we get the following bound on the
chromatic number of a given graph G
χ(G) ≤ 1 + core(G)
Cores can also be used to localize the search for interesting subnetworks in
large networks since: if it exists, a k-component is contained in a k-core; and
a k-clique is contained in a k-core.
The notion of core can be generalized to networks. Let N = (V, E, w) be a
network, where G = (V, E) is a graph and w : E → IR is a function assigning
values to edges. A vertex property function on N, or a p-function for short, is
a function p(v, U ), v ∈ V , U ⊆ V with real values. Let adjU (v) = adj(v) ∩ U .
Besides degrees, here are some examples of p-functions:
X
pS (v, U ) = w(v, u), where w : E → IR+ 0
u∈adjU (v)
Junius/Zrieva/ Lorenzo/Ragnina/
Damianus/Georgio/ Margarita/Bona/
Legnussa/Babalio/ Slavussa/Mence/
Nicolinus/Gondola/
Franussa/Bona/ Sarachin/Bona/
Nicoletta/Gondola/
Marin/Gondola/ Marinus/Bona/
Magdalena/Grede/ Phylippa/Mence/
Fig. 6. Triads
3.4 Triads
Let G = (V, R) be a simple directed graph without loops. A triad is a sub-
graph induced by a given set of three vertices. There are 16 nonisomorphic
(types of) triads [55, page 244]. They can be partitioned into three basic
types (see Figure 6):
• the null triad 003;
• dyadic triads 012 and 102; and
• connected triads: 111D, 201, 210, 300, 021D, 111U, 120D, 021U, 030T,
120U, 021C, 030C and 120C.
10 Vladimir Batagelj and Andrej Mrvar
types partition the set of all triads, the idea of the algorithm is as follows:
• count all dyadic T2 and all connected T3 triads with their subtypes;
• compute the number of null triads T1 = T − T2 − T3 .
In the algorithm we have to assure that every non-null triad is counted ex-
actly once while scanning the set of arcs. A set of three vertices {v, u, w}
can be in general selected in 6 different ways (v, u, w), (v, w, u), (u, v, w),
(u, w, v), (w, v, u), (w, u, v). We solve the isomorphism problem by introduc-
ing the canonical selection that contributes to the triadic count; the other,
noncanonical selections need not to be considered in the counting process.
Every connected dyad forms a dyadic triad with every vertex both mem-
bers of the dyad are not adjacent to. Let R̂ = R ∪ R−1 . Each pair of vertices
(v, u), v < u connected by an arc contributes
n − |R̂(u) ∪ R̂(v) \ {u, v}| − 2
triads of type 3 – 102, if u and v are connected in both directions; and
of type 2 – 012 otherwise. The condition v < u determines the canonical
selection for dyadic triads. A selection (v, u, w) of connected triad is canonical
iff v < u < w.
The triads isomorphism problem can be efficiently solved by assigning to
each triad a code – an integer number between 0 to 63 obtained by treating
the out-diagonal entries of triad adjacency matrix as a binary number. Each
triad code corresponds to a unique triad type that can be determined from
a precomputed table.
For a connected triad we can always assume that v is the smallest of its
vertices. So we have to determine the canonical selection from the remaining
two selections (v, u, w) and (v, w, u). If v < w < u and v R̂w then the selection
(v, w, u) was already counted before. Therefore we have to consider it as
canonical only if it is not v R̂w.
In an implementation of the algorithm we must also take care about the
range overflow in the case of T and T1 .
The total complexity of the algorithm is O(∆m)ˆ and thus, for graphs with
ˆ ˆ
small maximum degree ∆ << n, since 2m ≤ n∆, of order O(n).
KLEITMAN, DANIEL J.
ARONOV, BORIS
LINIAL, NATHAN PACH, JANOS
POLLACK, RICHARD M.
HENNING, MICHAEL A.
FRANKL, PETER SPENCER, JOEL H.
ALON, NOGA OELLERMANN, ORTRUD R.
LOVASZ, LASZLO
KOMLOS, JANOS GODDARD, WAYNE D.
FUREDI, ZOLTAN
TUZA, ZSOLT ALAVI, YOUSEF
BABAI, LASZLO
SZEMEREDI, ENDRE CHARTRAND, GARY
BOLLOBAS, BELA HARARY, FRANK
AJTAI, MIKLOS KUBICKI, GRZEGORZ
SCHWENK, ALLEN JOHN
RODL, VOJTECH
NESETRIL, JAROSLAV ROSA, ALEXANDER
GYARFAS, ANDRAS
SCHELP, RICHARD H. STINSON, DOUGLAS ROBERT
LEHEL, JENO CHEN, GUANTAO
MULLIN, RONALD C.
FAUDREE, RALPH J.
COLBOURN, CHARLES J.
JACOBSON, MICHAEL S.
PHELPS, KEVIN T.
Undirected graphs
We call a triangle a subgraph isomorphic to K3 . A subgraph H = (V 0 , E 0 )
of G = (V, E) is triangular if each its vertex and each its edge belongs to at
least one triangle in H.
A sequence (T1 , T2 , . . . , Ts ) of triangles of G (vertex) triangularly connects
vertices u, v ∈ V iff u ∈ T1 and v ∈ Ts or u ∈ Ts and v ∈ T1 and V (Ti−1 ) ∩
V (Ti ) 6= ∅, i = 2, . . . s. Such sequence is called a triangular chain. It edge
triangularly connects vertices u, v ∈ V iff a stronger version of the second
condition holds E(Ti−1 ) ∩ E(Ti ) 6= ∅, i = 2, . . . s.
A pair of vertices u, v ∈ V is (vertex) triangularly connected iff u =
v, or there exists a chain that triangularly connects u and v. Triangular
connectivity is an equivalence relation on the set of vertices V ; and nontrivial
triangular connectivity components are exactly maximal connected triangular
subgraphs.
A pair of vertices u, v ∈ V is edge triangularly connected iff u = v, or
there exists a chain that edge triangularly connects u and v. Edge triangular
connectivity components determine an equivalence relation on the set of edges
E. Each nontriangular edge is in its own component.
12 Vladimir Batagelj and Andrej Mrvar
For each type we get the corresponding triangular network Ncyc , Ntra ,
Nin and Nout . Also procedures for determining the networks are similar to
undirected case. For example, for the cyclic network Ncyc = (V, Acyc , wcyc )
we have for (u, v) ∈ Acyc
ES = {{u, v} : u 6= v ∧ (u, v) ∈ A}
serial
publication
American Library Directory
transaction log
periodical
suggestion box
review frequency
charge
series
issue
library colophon
call number Library Literature
journal layout
fixed location publishing
printing
blanket order American Library Association /ALA/
title page
Books in Print /BIP/
vendor
homepage International Standard Book Number /ISBN/
entry published price
round table
dummy edition
librarian condition
catalog plate fiction
Oak Knoll
bibliographic record imprint abstract
dust jacket work
book
bibliography editor
half-title
cover
page
Pajek
v iff A(Ti−1 ) ∩ A(Ti ) 6= ∅, i = 2, . . . s holds; such sequence is called an arc
cyclic triangular chain.
Again, we can introduce two types of cyclic triangular connectivity:
A pair of vertices u, v ∈ V is (vertex) cyclic triangularly connected iff
u = v, or there exists a cyclic triangular chain that connects u to v.
A pair of vertices u, v ∈ V is arc cyclic triangularly connected iff u = v,
or there exists an arc cyclic triangular chain that connects u to v.
Cyclic triangular connectivity is an equivalence relation on the set of
vertices V ; and the arc cyclic triangular connectivity components determine
an equivalence relation on the set of arcs A.
There exists also a parallel to unilateral connectivity. The vertex v ∈ V
is transitively triangularly reachable from the vertex u ∈ V iff u = v, or there
exists a walk from u to v in which each arc is transitive – is a base of some
transitive triangle.
Transitive arcs are essentially reinforced arcs. If we remove from a graph
G = (V, A) a transitive arc the reachability relation in V does not change.
In Figure 8 the edge-cut at level 11 of transitive network of ODLIS dic-
tionary graph [45] is presented.
These notions can be generalized to short cycle connectivity [20].
14 Vladimir Batagelj and Andrej Mrvar
Pr({u, v} ∈ G) = p
But, for large and very sparse networks this is too slow. A faster procedure
can be built on the following idea: move by random steps over the M = n2
cells and mark the touched cells.
How to select the length of the random step? For our Bernoulli model
we have Pr(step = s) = q s−1 p, s = 1, 2, 3, . . . and F (s) = Pr(step < s) =
Ps−1 t−1
t=1 q p = 1−q s−1 . Therefore we get the random step s from the equation
F (s) = random
log(1 − random)
s = F −1 (random) = 1 + b c
log q
This is the basis of the fast random graph generation procedure presented in
Algorithm 2. The expected number of steps of this procedure is M p.
edges are added to the network N . The endpoints of the edges are randomly
selected among all vertices according to the probability
indeg(v) outdeg(v) 1
Pr(v) = α +β +γ
|E| |E| |V |
P
where α + β + γ = 1. It is easy to check that v∈V Pr(v) = 1. The time
complexity of this procedure is O(m).
w(u, v) (u, v) ∈ A
n
auv =
0 otherwise
For direct analysis of 2-mode networks we can use eigen-vector approach,
clustering and blockmodeling. But most often we transform a 2-mode net-
work into an ordinary (1-mode) network N1 = (U, E1 , w1 ) or/and N2 =
(V, E2 , w2 ), where E1 and w1 are determined by the matrix A(1) = AAT ,
(1) (1) (1)
auv = z∈V auz · aTzv . Evidently auv = avu . There is an edge {u, v} ∈ E1 in
P
(1)
N1 iff adj(u) ∩ adj(v) 6= ∅. Its weight is w1 (u, v) = auv . The network N2 is
determined in a similar way by the matrix A(2) = AT A. The networks N1
and N2 are analyzed using standard methods.
3.8 Normalizations
wuv wuv
Geouv = √ GeoDeguv = p
wuu wvv degu degv
wuv wuv
Inputuv = Outputuv =
wvv wuu
wuv wuv
Minuv = Maxuv =
min(wuu , wvv ) max(wuu , wvv )
wuv n wuv
wuu ≤ wvv wuu ≤ wvv
n
MinDiruv = wuu MaxDiruv = wvv
0 otherwise 0 otherwise
and for directed networks as some mean value of the row and column sum,
for example
1 X X
wvv = ( wvu + wuv )
2 u u
Usually we assume that the network does not contain any isolated vertex.
After a selected normalization the important parts of network are ob-
tained by edge-cutting the normalized network at selected level t and pre-
serving components with at least k vertices.
Pajek, Analysis and Visualization of Large Networks 17
3.9 Blockmodeling
Pajek - shadow 0.00,1.00 Sep- 5-1998 Pajek - shadow 0.00,1.00 Sep- 5-1998
World trade - alphabetic order World Trade (Snyder and Kick, 1979) - cores
afg uki
alb net
alg bel
arg lux
aus fra
aut ita
bel den
bol jap
bra usa
brm can
bul bra
bur arg
cam ire
can swi
car spa
cha por
chd wge
chi ege
col pol
con aus
cos hun
cub cze
cyp yug
cze gre
dah bul
den rum
dom usr
ecu fin
ege swe
egy nor
els irn
eth tur
fin irq
fra egy
gab leb
gha cha
gre ind
gua pak
gui aut
hai cub
hon mex
hun uru
ice nig
ind ken
ins saf
ire mor
irn sud
irq syr
isr isr
ita sau
ivo kuw
jam sri
jap tha
jor mla
ken gua
kmr hon
kod els
kor nic
kuw cos
lao pan
leb col
lib ven
liy ecu
lux per
maa chi
mat tai
mex kor
mla vnr
mli phi
mon ins
mor nze
nau mli
nep sen
net nir
nic ivo
nig upv
nir gha
nor cam
nze gab
pak maa
pan alg
par hai
per dom
phi jam
pol tri
por bol
rum par
rwa mat
saf alb
sau cyp
sen ice
sie dah
som nau
spa gui
sri lib
sud sie
swe tog
swi car
syr chd
tai con
tha zai
tog uga
tri bur
tun rwa
tur som
uga eth
uki tun
upv liy
uru jor
usa yem
usr afg
ven mon
vnd kod
vnr brm
wge nep
yem kmr
yug lao
zai vnd
dom
maa
mon
maa
dom
mon
cam
mex
som
yem
mex
cam
som
yem
wge
wge
kuw
swe
swe
kuw
brm
mor
rum
rum
mor
brm
dah
den
ege
gab
gha
gua
hon
hun
kmr
mat
nau
nep
pan
uga
den
ege
hun
gua
hon
pan
gha
gab
mat
dah
nau
uga
nep
kmr
aus
can
cha
chd
con
cub
ecu
egy
jam
ken
kod
mla
nze
pak
rwa
sau
sen
spa
sud
upv
usa
ven
vnd
yug
usa
can
spa
aus
yug
egy
cha
pak
cub
ken
sud
sau
mla
ven
ecu
nze
sen
upv
jam
chd
con
rwa
kod
vnd
cos
cyp
cze
cze
cos
cyp
arg
bra
bur
gre
nor
par
per
por
swi
uru
bra
arg
swi
por
gre
nor
uru
per
par
bur
afg
aut
car
eth
kor
net
tha
tog
tun
usr
vnr
net
usr
aut
tha
kor
vnr
tog
car
eth
tun
afg
alb
alg
bel
bol
bul
gui
hai
ind
jap
lao
leb
nig
phi
pol
saf
syr
bel
jap
pol
bul
leb
ind
nig
saf
syr
phi
alg
hai
bol
alb
gui
lao
chi
col
els
ice
ins
ivo
lux
mli
nic
sie
uki
zai
uki
lux
els
nic
col
chi
ins
mli
ivo
ice
sie
zai
fra
tur
fra
tur
ire
irn
irq
jor
nir
ire
irn
irq
nir
jor
fin
isr
ita
sri
tai
ita
fin
isr
sri
tai
lib
lib
liy
liy
tri
tri
Fig. 10. Orderings
R(Ci , Cj ) = R ∩ Ci × Cj
Each such block consists of units belonging to clusters Ci and Cj and all arcs
leading from cluster Ci to cluster Cj . If i = j, a block R(Ci , Ci ) is called a
diagonal block.
18 Vladimir Batagelj and Andrej Mrvar
Since the set of units U is finite, the set of feasible clusterings Φ is also finite.
Therefore the set Min(Φ, P ) of all solutions of the problem (optimal cluster-
ings) is not empty. In theory, the set Min(Φ, P ) can be determined by the
complete search – but it turns out that most cases of the clustering problem
are N P hard. The blockmodeling problems are usually solved using local
optimization methods based on moving a unit from one cluster to another or
interchanging two units between two clusters.
One of the possible ways of constructing a criterion function that directly
reflects the considered equivalence is to measure the fit of a clustering to
Pajek, Analysis and Visualization of Large Networks 19
an ideal one with perfect relations within each cluster and between clusters
according to the considered equivalence.
Given a clustering C = {C1 , C2 , . . . , Ck }, let B(Cu , Cv ) denote the set of
all ideal blocks corresponding to block R(Cu , Cv ). Then the global error of
clustering C can be expressed as
X
P (C) = min d(R(Cu , Cv ), B)
B∈B(Cu ,Cv )
Cu ,Cv ∈C
where the term d(R(Cu , Cv ), B) measures the difference (error) between the
block R(Cu , Cv ) and the ideal block B. d is constructed on the basis of
characterizations of types of blocks. The function d has to be compatible
with the selected type of equivalence. Determining the block error, we also
determine the type of the best fitting ideal block (the types are ordered).
The criterion function P (C) is sensitive iff P (C) = 0 ⇔ C determines an
exact blockmodeling. For all presented block types sensitive criterion func-
tions can be constructed. Once a clustering C and types of blocks are de-
termined, we can also compute the values of connections by using averaging
rules.
In Figure 13 a symmetric acyclic (edge connected inside clusters, acyclic
reduced graph) blockmodel [27] of Student Government at the University of
Ljubljana [35] is presented. The obtained clustering in 4 clusters is almost
exact. The only error is produced by the arc (a3, m5).
20 Vladimir Batagelj and Andrej Mrvar
4 Implementation
4.1 Data structures
In Pajek analysis and visualization are performed using 6 data types:
• network (graph),
• partition (nominal or ordinal properties of vertices),
• vector (numerical properties of vertices),
• cluster (subset of vertices),
• permutation (reordering of vertices, ordinal properties), and
• hierarchy (general tree structure on vertices).
In the near future we intend to extend this list with a support of multiple
networks and partitions of edges.
The power of Pajek is based on several transformations that support
different transitions among these data structures. Also the menu structure
(see Figure 14) of the main Pajek’s window is based on them. Pajek’s main
window uses a ‘calculator’ paradigm with list-accumulator for each data type.
The operations are performed on the currently active (selected) data and are
also returning the results through accumulators.
The values of vectors can be used to determine several elements of network
display such as: X, Y, Z coordinates and the size of the vertex shape. The
partition can be graphically represented by the color and shape of vertices.
Also the values of edges can be represented by the thickness and/or color.
Pajek, Analysis and Visualization of Large Networks 21
4.4 Interfaces
Pajek supports also some non-native input formats: UCINET DL files [53];
Vega graph files [54]; chemical MDLMOL [41] and BS; and genealogical GED-
COM [30].
The layouts can be exported in the following output graphic formats that
can be examined by special 2D and 3D viewers: Encapsulated PostScript
(EPS) [31], Scalable Vector Graphics (SVG) [1], VRML [24], MDLMOL/
chime [41], and Kinemages (Mage) [49].
The main window menu Tools provides export of Pajek’s data to statisti-
cal program R [48,21]. In the Tools menu, the user can prepare calls to her/his
favorite viewers and other tools. It is also possible to run Pajek (+macros)
from other programs (R, Ucinet, and others).
5 Examples
6 Software
6.1 Architecture
6.2 Availability
Pajek is still under development. The latest version is freely available, for
noncommercial use, at its home page:
http://vlado.fmf.uni-lj.si/pub/networks/pajek/
24 Vladimir Batagelj and Andrej Mrvar
References
13. Batagelj V., Mrvar A. (2002) Pajek - Analysis and Visualization of Large Net-
works. In: Mutzel P., Jünger M., Leipert S. (Eds.) GD’01, Vienna, Austria.
September 23-26, 2001 LNCS 2265. Springer-Verlag, 477-478.
14. Batagelj V., Mrvar A. (2002) Density based approaches to Reuters terror news
network analysis. submitted.
15. Batagelj V., Pisanski T. (1989) Xgraph project documentation.
16. Batagelj V., Mrvar A., Zaveršnik M. (1999) Partitioning Approach to Visual-
ization of Large Graphs. In: Kratochvil J. (Ed.) GD’99, Štiřin Castle, Czech
Republic. LNCS 1731. Springer-Verlag, 90-97.
17. Batagelj V., Mrvar A., Zaveršnik M. (2002) Network analysis of texts. Language
Technologies, Ljubljana, p. 143-148.
18. Batagelj V., Zaveršnik M. (2001) An O(m) Algorithm for Cores Decomposition
of Networks. Submitted.
19. Batagelj V., Zaveršnik M. (2002) Generalized Cores. Submitted.
http://arxiv.org/abs/cs.DS/0202039
20. Batagelj, V. and Zaveršnik, M. (2002) Triangular connectivity and its general-
izations, in preparation.
21. Butts, C.T. (2002) sna: Tools for Social Network Analysis.
http://cran.at.r-project.org/src/contrib/PACKAGES.html#sna
22. Caida: Internet Visualization Tool Taxonomy.
http://www.caida.org/tools/taxonomy/visualization/
23. Cormen T.H., Leiserson C.E., Rivest R.L., Stein C. (2001) Introduction to
Algorithms, Second Edition. MIT Press.
24. Cosmo Player (2002) http://ca.com/cosmo/
25. de Nooy W., Mrvar A., Batagelj V. (2002) Exploratory Social Network Analysis
With Pajek. to be published by the Cambridge University Press.
26. Doreian P., Mrvar A. (1996) A Partitioning Approach to Structural Balance.
Social Networks, 18. 149-168
27. Doreian, P., Batagelj, V., Ferligoj, A. (2000) Symmetric-acyclic decompositions
of networks. J. classif., 17(1), 3-28.
28. Dremelj P., Mrvar A., Batagelj V. (2002) Analiza rodoslova dubrovačkog vlas-
teoskog kruga pomoću programa Pajek. Anali Dubrovnik XL, HAZU, Zagreb,
Dubrovnik, 105-126 (in Croat).
29. Garfield E, Sher IH, and Torpie RJ.: The Use of Citation Data in Writing
the History of Science. Philadelphia: The Institute for Scientific Information,
December 1964. http://www.garfield.library.upenn.edu/papers/
useofcitdatawritinghistofsci.pdf
30. GEDCOM 5.5.
http://homepages.rootsweb.com/~pmcbride/gedcom/55gctoc.htm
31. Ghostscript, Ghostview and GSview http://www.cs.wisc.edu/~ghost/
32. Gibbons A. (1985) Algorithmic Graph Theory. Cambridge University Press.
33. Graph Drawing Contest 1997. http://vlado.fmf.uni-lj.si/pub/gd/gd97.htm
34. Grossman J. (2002) The Erdős Number Project.
http://www.oakland.edu/~grossman/erdoshp.html
35. Hlebec, V. (1993) Recall versus recognition: Comparison of two alternative
procedures for collecting social network data. Metodološki zvezki 9, Ljubljana:
FDV, 121-128.
36. Hummon, N.P. & Doreian, P. (1989) Connectivity in a citation network: The
development of DNA theory. Social Networks, 11, 39–63.
26 Vladimir Batagelj and Andrej Mrvar