Using Self-Organizing Map and Heuristics to Identify Small Statistical Areas Based on Household Socio-Economic Indicators in Turkey's Address Based Population Register System

Census operations are very important events in the history of a nation. These operations cover every bit of land and property of the country and its citizens. The publication of census based on spatial units is one of the important problems of national statistical organizations, which requires determination of small statistical areas (SSAs) or so called census geography. Since 2006, Turkey aims to produce census data not as “de-facto” (static) but as “de-jure” (real-time) by the new Address Based Register Information System (ABPRS). Besides, by this new register based census, personal information is matched with their address information and censuses gained a spatial dimension. However, as Turkey lacks SSA's, the data cannot be published in smaller spatial granularities. In this study, it is aimed to employ a spatial clustering and districting methodology to automatically produce SSAs which are basically built upon the ABPRS data that is geo-referenced with the aid of geographical information systems (GIS). For its realization, simulated annealing on k-means clustering of Self-Organizing Map (SOM) unified distances is employed to produce SSA's for ABPRS. This method is basically implemented on block datasets having either raw census data or socio-economic status (SES) indices obtained from census data. The resulting SSA's are evaluated for the case study area.