Comparison of two network-theory-based methods for detecting functional regions

Background: Functional regions are abstract, uniformly defined territorial units that form an important basis for many development strategies of a country or a region. Objectives: This study analyses the application of network theory to the detection of such regions. Methods/Approach: Functional regions are analysed using two methods based on the graph theory: the Walktrap algorithm and the chain approach. The quality of the two regionalization methods is analysed using the fuzzy set theory with the revised method. Slovenia was used as a case study. Results: The Walktrap algorithm generated eight functional regions; seven of them corresponded to those identified in previous studies. The only difference occurred in the northwestern mountainous part of Slovenia. The chain approach led to similar results, although it resulted in a huge functional urban region of the capital Ljubljana. Conclusions: The results show that the Walktrap algorithm calculates regions that are more closed, where more workers find work in the home region, than the chain approach.