| 摘要: |
| 本研究使用腾讯云自然语言处理技术
对微博数据中的居民情绪进行了识别，得到了不
同微博发送点位的情绪值，对不同类型情绪的
空间分布特征进行了分析，并采用回归方法分
析了社区的平均微博情绪指数与社区绿地率、
职住比之间的关系。研究主要有如下发现：居
民的情绪整体偏向积极，积极情绪主要分布在
景点和休闲相关设施的周边；当社区内科教文
化、公司企业、医疗设施类POI密度增加到一定
阈值后，居民情绪呈下降趋势；当社区绿地率在
39.3%附近时，社区平均情绪指数情况较低，而
职住关系指数与微博情绪指数呈反比关系。 |
| 关键词: 健康城市，情绪分布，绿地率，职住关
系，建成环境 |
| DOI：10.13791/j.cnki.hsfwest.20230204 |
| 分类号: |
| 基金项目:国家科技基础性工作专项（2013FY11250）；国家自
然科学基金青年科学基金项目（51208389） |
|
| Research on the Relationship Between Residents’ Emotions and Built Environment Basedon Weibo Data: Taking Wuhan City as an Example |
|
LIU Minghao,LI Kun,LI Chenhui
|
| Abstract: |
| Emotions and moods are essential components of mental health indicators, reflecting
a person’s psychological state and well-being. However, mental health issues are prevalent in our
country, with many people not meeting mental health standards，which calls for attention to be given
to mental health, and one way to do that is through creating a pleasant built environment that can
significantly improve people’s emotions and consequently improve their mental health status. While
a pleasant built environment can be beneficial, its traditional research methods, like questionnaires,
are often time-consuming, laborious, and non-immediate. However, with the development of big data
technology, social media information has facilitated the study of the relationship between residents’
moods and the built environment, which has some immediacy. This method has opened up new
possibilities for studying the built environment’s impact on emotions and moods. To conduct this
study, it firstly obtained relevant built environment data and sentiment values of the microblog text.
It used Tencent Cloud natural language processing technology to identify the residents’ sentiment
in the microblog data of Wuhan city in July 2022 and derive the sentiment values of different
microblog sending locations. It also used the POI and AOI of Gaode Map with Google satellite
map information to obtain the green space rate and building information. Secondly, it analyzed
the spatial distribution characteristics of different types of emotions and summarized the areas of
concentration of positive and negative emotions, respectively. In addition, we used the most basic
administrative unit in China’s administrative division, i.e., community, as the unit of analysis and
used the geographic probe to determine the built environment characteristics that have a relationship
with the microblogging sentiment index. It classified the unit of analysis into seven classes according
to the quantile of the built environment. This helped us determine the built environment’s impact
on emotions and moods. The results of our study revealed that Wuhan residents’ emotions are
positive overall, with positive emotions mainly distributed around attractions and leisure-related
facilities. However, when the density of science, education, culture, corporate enterprises, and
medical facilities POIs in the community is increased to a certain threshold, residents’ emotions tend
to decline. This highlights the importance of having a balance between different types of facilities
in a community. Furthermore, it constructed a mathematical model of sentiment, community
green space rate, and occupational and residential relationship index. This study highlights the
importance of studying the relationship between the built environment and emotions and provides
recommendations for improving emotional positivity in communities. The findings showed that
when the community green space rate was around 39.3%, the average community sentiment index
situation was low, while the occupational and residential relationship index was inversely related to
the microblogging sentiment index. This highlights the importance of green spaces in communities
and the need to balance occupational and residential areas. To improve residents’ emotional
positivity, it recommends increasing the green space rate in areas with high green space rates, such
as parks to improve emotional positivity. In addition, the daytime vitality of suburban communities
should be increased by balancing transportation, reducing commuting distances, and increasing
infrastructure in lagging communities to improve the job-to-residence ratio to increase residents’emotional positivity. At the policy level, the number of POIs in the central city can be reduced by “reducing the burden” in order to relieve the pressure on
students and increase their emotional value. In conclusion, emotions and moods are essential components of mental health indicators, and a pleasant built
environment can significantly improve people's emotions and mental health. |
| Key words: Healthy City Sentiment Distribution Green Space Rate Work-Life Relationship Built Environment |
