Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Smart assessment and forecasting framework for healthy development index in urban cities
https://orcid.org/0000-0001-8868-6913
https://orcid.org/0000-0001-5735-2507
Abstract With the sustainable development being a consensus in human society, assessment of healthy development index in urban cities (HDI-UC) has been a hot concern in academia. Existing research works had proposed some assessment models from the perspective of sociology. However, these approaches can just assess HDI-UC values of current year and past years, as they relied on complete index systems. They failed to possess the ability to directly assess HDI-UC values in future years. To bridge such gap, this paper proposes a smart assessment and forecasting framework for HDI-UC. On the one hand, an assessment model proposed in a previous study is introduced as the basic model. On the other hand, the Gaussian process regression is utilized to model the evolving HDI-UC sequence, so that HDI-UC values in future years can be directly forecasted according to historical ones. A case study is conducted on real-world statistical data collected from some regions of China to illustrate assessment process of the framework. Besides, another group of experiments are also carried out to evaluate forecasting performance. The simulation results show that prediction error of SAF-HDI is around 5 %, which is within an acceptable range.
Highlights Smart assessment and forecasting framework for index of cities A machine learning method for rapid forecasting Simulative analysis via computer programming
Smart assessment and forecasting framework for healthy development index in urban cities
https://orcid.org/0000-0001-8868-6913
https://orcid.org/0000-0001-5735-2507
Abstract With the sustainable development being a consensus in human society, assessment of healthy development index in urban cities (HDI-UC) has been a hot concern in academia. Existing research works had proposed some assessment models from the perspective of sociology. However, these approaches can just assess HDI-UC values of current year and past years, as they relied on complete index systems. They failed to possess the ability to directly assess HDI-UC values in future years. To bridge such gap, this paper proposes a smart assessment and forecasting framework for HDI-UC. On the one hand, an assessment model proposed in a previous study is introduced as the basic model. On the other hand, the Gaussian process regression is utilized to model the evolving HDI-UC sequence, so that HDI-UC values in future years can be directly forecasted according to historical ones. A case study is conducted on real-world statistical data collected from some regions of China to illustrate assessment process of the framework. Besides, another group of experiments are also carried out to evaluate forecasting performance. The simulation results show that prediction error of SAF-HDI is around 5 %, which is within an acceptable range.
Highlights Smart assessment and forecasting framework for index of cities A machine learning method for rapid forecasting Simulative analysis via computer programming
Smart assessment and forecasting framework for healthy development index in urban cities
https://orcid.org/0000-0001-8868-6913
https://orcid.org/0000-0001-5735-2507
Abstract With the sustainable development being a consensus in human society, assessment of healthy development index in urban cities (HDI-UC) has been a hot concern in academia. Existing research works had proposed some assessment models from the perspective of sociology. However, these approaches can just assess HDI-UC values of current year and past years, as they relied on complete index systems. They failed to possess the ability to directly assess HDI-UC values in future years. To bridge such gap, this paper proposes a smart assessment and forecasting framework for HDI-UC. On the one hand, an assessment model proposed in a previous study is introduced as the basic model. On the other hand, the Gaussian process regression is utilized to model the evolving HDI-UC sequence, so that HDI-UC values in future years can be directly forecasted according to historical ones. A case study is conducted on real-world statistical data collected from some regions of China to illustrate assessment process of the framework. Besides, another group of experiments are also carried out to evaluate forecasting performance. The simulation results show that prediction error of SAF-HDI is around 5 %, which is within an acceptable range.
Highlights Smart assessment and forecasting framework for index of cities A machine learning method for rapid forecasting Simulative analysis via computer programming
Smart assessment and forecasting framework for healthy development index in urban cities
Li, Qiao (Autor:in) / Liu, Lian (Autor:in) / Guo, Zhiwei (Autor:in) / Vijayakumar, Pandi (Autor:in) / Taghizadeh-Hesary, Farhad (Autor:in) / Yu, Keping (Autor:in)
Cities ; 131
26.08.2022
Aufsatz (Zeitschrift)
Elektronische Ressource
Englisch
Urban Computing: The Technological Framework for Smart Cities
| Springer Verlag | 2021
Developing an assessment governance framework for urban digital twins: Insights from smart cities
| Elsevier | 2025