A density-based spatial cluster analysis supporting the building stock analysis in historical towns

Abstract

The paper presents the application of a spatial cluster approach supporting the building-stock analysis of historic towns. This method was applied in the town of Calavino, within the Municipality of Madruzzo, an historic settlement located in the Province of Trento and characterized by high natural and heritage values. The proposed data mining approach shows as several buildings can not be classified in clusters for the historic centre of Calavino when physical features are combined with other variables (e.g. building function, owner, age class, shape and physical features, heritage values and conservation state, etc.). After this analysis, detailed energy audits and dynamic simulations can be addressed on specific "building typologies".