Mining official data

In statistics, the term “official data” denotes data collected in censuses and statistical surveys by National Statistics Institutes (NSIs), as well as administrative and registration records collected by government departments and local authorities. They are used to produce “official statistics” for the purpose of making policy decisions, and to facilitate the appreciation of economic, social, demographic, and other matters of interest to governments, government departments, local authorities, businesses and to the general public. For instance, population and economic census information is of great value in planning public services (education, fund allocation, public transport), as well as in private businesses (placing new factories, shopping malls, or banks, as well as marketing particular products). Moreover, survey data on specific topics, such as labour force, time use, household budget, are regularly collected by NSIs to keep updated information on some economic and social phenomena. The application of data mining techniques to official data has great potential in supporting good public policy and in underpinning the effective functioning of a democratic society. Nevertheless, it is not straightforward and requires challenging methodological research, which is still in the initial stages. This special issue includes six papers which constitute updated and extended versions of papers selected from those presented at the Workshop on Mining Official Data, chaired by the guest editors of this issue in Helsinki in August 2002. The workshop was organized under the auspices of the European project KDNet (The Knowledge Discovery Network of Excellence) and within the framework of the 13th European Conference on Machine Learning (ECML’02) and the 6th European Conference on Principles and Practice of Knowledge Discovery in Databases (PKDD’02). Different directions can be distinguished in the approach of the problem of mining official data. In this issue, emphasis is placed on the following topics: Geo-referenciation. The practice of geo-referencing census data has increasingly spread over the last few decades and the techniques for attaching socio-economic data to specific locations have markedly improved at the same time. In the UK, for instance, household expenditure data are provided for each enumeration district (ED), the smallest areal unit for which census data are published. At the same time, vectorized boundaries of the 1991 census EDs enable the investigation of socio-economic phenomena in association with the geographical location of EDs. These advances cause a growing demand for more powerful data analysis techniques that can link population data to their spatial distribution. In this context, a European project, SPIN, has been developed to address problems concerning geo-referenciation. SPIN’s