2 October 2019 in Frontiers in Environmental Science
Urban green spaces (UGS) are crucial providers of cultural ecosystem services such as recreation. So that urban residents can benefit from UGS recreation, there is a need for good access to these, in particular for elderlies with reduced mobility. Recent research on accessibility has often neglected to consider real distances to green spaces and emerging barriers as well as the demand by specific user groups for UGS accessibility. In the light of demographic aging and ongoing urbanization, this study investigates in an explorative case study the potential and actual access to UGS for elderlies living in care facilities in Salzburg (Austria). By connecting a GIS network-analysis with a survey among elderlies of four care facilities, the results give insights into the supply and demand concerning UGS accessibility. The supply analysis showed that the majority of UGS are situated between 500 and 1,000 m. Due to barriers identified by the elderlies hindering the pathways to the UGS, time losses occur and the security of the old-aged is under risk. The demand side showed that despite the supply of UGS within 1,000 m, elderlies mostly visit UGS outside the service area. More research is needed to include elderlies' supply and demand regarding UGS qualities when analyzing UGS accessibility by considering greater sample sizes.
1 September 2018 in Computers, Environment and Urban Systems
30 May 2024 in AGILE: GIScience Series
This paper investigates the use of machine learning (ML) models to predict housing prices. A well-performing housing price model was trained, which can seamlessly be integrated into public sector processes to increase market transparency and is based on modern ML and feature engineering methods. For these models, particular consideration was given to the spatial component. The research uses the Design Science Research approach, with a case study carried out in the city of Duisburg, Germany. The ML models developed showed better performance than traditional models. The models were embedded in official processes using Shapley values to increase interpretability. The study concludes that ML models can contribute to increased market transparency in the real estate sector.
10 June 2026 in AGILE: GIScience Series
Location is a key feature of urban spaces. It influences housing and land markets, social segregation, and decisions in urban planning and administration. Hedonic price models and, increasingly, machine learning (ML) methods are traditionally used for quantitative location valuation. At the same time, transformer architectures and their more recent adaptations for tabular data mark a methodological milestone. However, to date, tabular transformers have not been widely adopted in spatial analyses and, in particular, have not been systematically evaluated for location value assessments. This study addresses this gap by comparing classic ML methods with two tabular transformer approaches: TabNet and Mitra. In a case study, rents are used as the target variable, with property-specific characteristics being normalised using spatial regressions. Location-describing features include accessibility to POIs and areal variables such as imperviousness, noise, and population density. While a stacked ensemble of classical ML models achieved the lowest RMSE, TabNet and Mitra performed within a narrow margin, demonstrating that they can be used to map spatial patterns.