1:8 R. Angles and C. Gutierrez
transport networks (train, plane, water, telecommunications) [Mainguenaud 1995],
and spatially embedded networks like highway, public transport [G
¨
uting 1994]. Sev-
eral of these applications are now in the field of GIS and spatial databases;
—the limited expressive power of current query languages motivated the search for
models that allowed better representation of more complex applications [Paredaens
et al. 1995];
—limitations at the time of knowledge representation systems [Kunii 1987], and
the need for intricate but flexible knowledge representation and derivation tech-
niques [Paredaens et al. 1995];
—after observing that graphs have been an integral part of the database design pro-
cess in semantic and object-oriented db-models, the idea of having a model where
both data manipulation and representation are graph-based emerged [Gyssens et al.
1990a];
—the need for improving the functionality offered by object-oriented db-models
[Poulovassilis and Levene 1994]; for example, CASE, CAD, image processing, and
scientific data analysis software all fall into this category;
—graphical and visual interfaces, geographical, pictorial, and multimedia systems
[Gyssens et al. 1990b; Consens and Mendelzon 1993; Sheng et al. 1999];
—software systems and integration [Kiesel et al. 1996];
—the appearance of on-line hypertext evidenced the need for other db-models, like
the ones suggested by Tompa [1989], Watters and Shepherd [1990], and Amann and
Scholl [1992]; together with hypertext, the Web created the need for a more apt model
than the classical ones for information exchange.
Complex Networks. Several areas have witnessed the emergence of huge data net-
works with interesting mathematical properties, called complex networks [Newman
2003; Albert and Barab
´
asi 2002; Dorogovtsev and Mendes 2003]. The need for database
management for certain types of these networks has been recently highlighted [Olken
2003; Jagadish and Olken 2003; Tsvetovat et al. 2004; Graves et al. 1995b]. Although
it is not yet evident if we can group these databases into one category, we will present
them in this manner. As in Newman’s [2003] survey, we will subdivide this category into
four subcategories: social networks, information networks, technological networks and
biological networks. In the following text, we describe each subcategory via an example.
—In social networks [Hanneman 2001], nodes are people or groups, while links show
relationships or flows among nodes. Some examples are friendships, business rela-
tionships, sexual contact patterns, research networks (collaboration, coauthorship),
communication records (mail, telephone calls, email), computer networks [Wellman
et al. 1996], and national security [Sheth et al. 2005]. There is growing activity in the
area of social network analysis [Brandes 2005], and also in visualization and data
processing techniques for these networks.
—Information networks model relations representing information flow, such as cita-
tions among academic papers [de S. Price 1965], World Wide Web (hypertext, hy-
permedia) [Florescu et al. 1998; Kumar et al. 2000; Broder et al. 2000], peer-to-peer
networks [Nejdl et al. 2003], relations among word classes in a thesaurus, and pref-
erence networks.
—In technological networks, the spatial and geographical aspects of the structure are
dominant. Some examples are the Internet (as a computer network), electric power
grids, airline routes, telephone networks, delivery network (post office), and Geo-
graphic Information Systems (GIS) are today covering a big part of this area (roads,
railways, pedestrian traffic, rivers) [Shekhar et al. 1997; Medeiros and Pires 1994].
ACM Computing Surveys, Vol. 40, No. 1, Article 1, Publication date: February 2008.