Proliferation of Short Term Rental (STR) in cities has generated considerable debate as it was found associated with negative externalities, such as gentrification. Nonetheless, it signals urban qualities working as attractors at different geographical scales. STRs’ relation with urban form remains largely understudied. In this paper, we explore how urban form relates to STRs registered by the Airbnb platform in Amsterdam (NL). First, we identify urban types (homogenous patterns of form) through an ‘urban morphometric’ approach. Second, we assess the relation between urban types and density of Airbnbs via a composite machine learning (ML) technique. Third, we provide profiles of the urban types most strongly associated with it. Fifteen urban types explain up to 44% of Airbnb density’s variance. Compact and diverse urban types relate more strongly with Airbnbs. Conversely, repetitive, sparse and uniform urban types are inversely related. The proposed morphometric-based method is robust, replicable and scalable, offering a novel way to study the intricate relation between urban form, STRs and, in fact, any other measurable urban dynamics at an unprecedented scale. By identifying spatial features related to urban attractiveness, it can inform evidence-based design codes incorporating place-making qualities in existing and new neighbourhoods.
|Number of pages||15|
|Journal||Environment and Planning B: Urban Analytics and City Science|
|Publication status||Published - Feb 2023|
- machine learning
- urban morphology
- urban morphometrics