The emerging structure of Russian urban system : a classification based on Self-Organizing Maps

After the collapse of the Soviet regime and the end of centralized policies on urban planning, Russian cities underwent deep transformations. The complex socio-economic, spatial and political trends are leading towards an increasing unevenness among Russian cities. This topic has been poorly investigated by the current urban research. The paper aims to analyze and classify the cities belonging to the Russian urban system on the base of some socio-economic indica-tors captured at two different temporal thresholds. By means of the Neural Networks Self-Organizing Maps (SOM), the study has investigated and singled out groups of cities with high internal resemblance. Due to SOM implementation it has been possible to identify twenty five groups of cities with similar socio-economic trends, where each group is char-acterized by an appropriate profile (a codebook). Moreover the empirical results have allowed to identify a new urban ranking, structured in four layers: "urban engine", "strong cities"; "dynamic cities" and "weak cities". The outcomes should support the definition of appropriate urban develop-ment strategies.