School of Public Affairs, Zhejiang University, Zhejiang 310027, China.
School of Management, Zhejiang University, Zhejiang 310027, China.
Int J Environ Res Public Health. 2020 Nov 5;17(21):8167. doi: 10.3390/ijerph17218167.
Mental health is the foundation of health and happiness as well as the basis for an individual's meaningful life. The environmental and social health of a city can measure the mental state of people living in a certain areas, and exploring urban dwellers' mental states is an important factor in understanding and better managing cities. New dynamic and granular urban data provide us with a way to determine the environmental factors that affect the mental states of urban dwellers. The characteristics of the maximal information coefficient can identify the linear and nonlinear relationships so that we can fully identify the physical and social environmental factors that affect urban dwellers' mental states and further test these relationships through linear and nonlinear modeling. Taking the Greater London as an example, we used data from the London Datastore to discover the environmental factors that had the highest correlation with urban mental health from 2015 to 2017 and to prove that they had a high nonlinear correlation through neural network modeling. This paper aimed to use a data-driven approach to find environmental factors that had not yet received enough attention and to provide a starting point for research by establishing hypotheses for further exploration of the impact of environmental factors on mental health.
心理健康是健康和幸福的基础,也是个人有意义生活的基础。城市的环境和社会健康可以衡量生活在特定地区的人们的精神状态,探索城市居民的精神状态是了解和更好管理城市的重要因素。新的动态和粒度城市数据为我们提供了一种确定影响城市居民精神状态的环境因素的方法。最大信息系数的特征可以识别线性和非线性关系,从而可以充分识别影响城市居民精神状态的物理和社会环境因素,并通过线性和非线性建模进一步检验这些关系。以大伦敦为例,我们使用来自伦敦数据库的数据,从 2015 年到 2017 年发现与城市心理健康相关性最高的环境因素,并通过神经网络建模证明它们具有高度的非线性相关性。本文旨在通过数据驱动的方法找到尚未受到足够重视的环境因素,并通过建立假设为进一步探索环境因素对心理健康的影响提供研究起点。