A nonlinear, data-driven, ANNs-based approach to culture-led development policies in rural areas: The case of Gjakove and Peć districts, Western Kosovo

We develop a computational approach to the analysis of cultural vibrancy and to the role of the cultural and creative sectors in the socio-economic organization of two districts of Western Kosovo, Gjakove and Peć. Our analysis is built on a geolocalized mapping of the cultural activities and facilities, and on the main socio-economic variables for the two districts, and makes use of innovative data analysis techniques: Theory of Impossible Words (TIW), the Topological Weighted Centroid (TWC), and the AutoCM ANN. We find that the dynamics of cultural vibrancy of the territory is mainly driven by the competing attraction pulls of the nearby countries of Serbia and Albania, that also form the region's main and often conflicting ethnicities, and that such dynamics are likely to further polarize in the future. We also find that the cultural system plays a marginal role in the territory's socio-economic organization. This situation makes a case for a more active role of cultural policy in shaping future local developmental models in rural areas and in acting as an agent of social cohesion.