Predicting comfort levels in cities is challenging due to the many metric assessment.To overcome these challenges,much research is being done in the computing community to develop methods capable of generating outdoor...Predicting comfort levels in cities is challenging due to the many metric assessment.To overcome these challenges,much research is being done in the computing community to develop methods capable of generating outdoor comfort data.Machine Learning(ML)provides many opportunities to discover patterns in large datasets such as urban data.This paper proposes a data-driven approach to build a predictive and data-generative model to assess outdoor thermal comfort.The model benefits from the results of a study,which analyses Computational Fluid Dynamics(CFD)urban simulation to determine the thermal and wind comfort in Tallinn,Estonia.The ML model was built based on classification,and it uses an opaque ML model.The results were evaluated by applying different metrics and show us that the approach allows the implementation of a data-generative ML model to generate reliable data on outdoor comfort that can be used by urban stakeholders,planners,and researchers.展开更多
基金This work has been supported by the European Commission through the H2020 project Finest Twins(grant No.856602).
文摘Predicting comfort levels in cities is challenging due to the many metric assessment.To overcome these challenges,much research is being done in the computing community to develop methods capable of generating outdoor comfort data.Machine Learning(ML)provides many opportunities to discover patterns in large datasets such as urban data.This paper proposes a data-driven approach to build a predictive and data-generative model to assess outdoor thermal comfort.The model benefits from the results of a study,which analyses Computational Fluid Dynamics(CFD)urban simulation to determine the thermal and wind comfort in Tallinn,Estonia.The ML model was built based on classification,and it uses an opaque ML model.The results were evaluated by applying different metrics and show us that the approach allows the implementation of a data-generative ML model to generate reliable data on outdoor comfort that can be used by urban stakeholders,planners,and researchers.
