The research on the algorithmics of nonuniform networks undertaken in this project will result in novel methods of analysing network structure, such as algorithms for clustering and sampling large nonuniform networks, and algorithms for computationally demanding network problems that are optimised to perform efficiently on certain natural classes of input instances. Such techniques are straightforward to apply in several engineering, e.g. telecommunication, applications, but as the study of nonuniform network models is highly multidisciplinary, potential application areas for the research results also exist in several other fields, including biochemistry, genetics, and epidemiology.

Also, as better tools are developed to recognise and classify structural properties of natural networks such as traffic or communication networks, or dependencies within a genome, mathematical models may be constructed to generate random instances of desired size that display similar characteristics and can be used to study the behavior of algorithms and any phenomenon that can be modeled either stochastically or deterministically. Generating realistic problem instances helps to analyse the performance of existing network algorithms with respect to their practical purpose, alongside with analytical methods that take the structural properties of the true-world applications into account.

The results of the research will be reported in the international open scientific literature, and will be collected in 2006 in Satu Elisa Schaeffer's D.Sc. thesis. Dissemination of the work will also be achieved by research visits and active participation in relevant workshops and conferences. We are also keen to establish international and multidisciplinary collaboration contacts to strengthen both the foundations and the applicability of the research results. Along with the research publications, we shall also produce data sets and small-scale experimentation software which may be of interest to other researchers as well.