Happy neighborhoods: Investigating neighborhood conditions and sentiments of a shrinking city with Twitter data

Abstract

Planning interventions have been applied to improve the well‐being, hereafter happiness, of residents. The happiness in shrinking cities, in particular, becomes more critical since urban decline tends to induce an unequal and uneven distribution of care under a limited budget and human resources. Using geo‐tagged Twitter, census, and geospatial data on Detroit, Michigan, which is one of the well‐known shrinking cities in the U.S., the spatial distribution of sentiments, topics of tweets appeared, and the association between neighborhood conditions and the level of happiness were examined. The outcomes indicate that people in Detroit are posting happy tweets more than negative tweets. The downtown area holds both positive and negative hotspots, which are clustered around sports arenas and bars, respectively. Neighborhoods with young and well‐educated residents, situated close to amenities (i.e., recreation facilities, colleges, and commercial areas), and less crime tend to be happier. The use of SNS data could serve as a meaningful social listening tool to reconcile the declining urban vitality of neighborhoods since people interact with those spaces. Negative sentiments are attached to specific neighborhoods with certain conditions so that regeneration efforts should take place in neighborhoods with a higher priority.

Socially Enhanced Situation Awareness from Microblogs Using Artificial Intelligence: A Survey

“…Similarly, Park et al [103] examined the spatial distribution of sentiments and the level of happiness using geo-tagged Twitter data, census data and geospatial data on one of the shrinking cities in the United States-Detroit. The authors collected username, text, geo-coordinates and used the Complete Automation Probability (CAP) technique to filter out bot-involved tweets.…”

Sources and Applications of Emerging Active Travel Data: A Review of the Literature

“…Through mining techniques, researchers can extract user perceptions on certain topics, whereby user locations can be inferred from geotags [108]. These data have long been acquired from surveys, which require effort in recruiting the sample and may be hindered by low response rates [109]. Thus, transport policies can harvest information from social media to monitor traffic in real time, model travel behavior and demand, and qualitatively analyze facilities' service qualities [110].…”

“…Despite their benefits, social media data are subject to age group bias and inconsistencies in the data collection [109].…”

Inequalities in Skills at the End of Education

“…One possibility would be to use dated and geocoded information generated by social networks. For example, using data from Twitter, Park et al (2021) show that it is possible to identify areas of a city that generate feelings of happiness or dissatisfaction. Using data from the platform Yelp, Glaeser et al (2018) show that the information generated by social networks can not only provide a better understanding of gentrification phenomena but can also predict them, almost in real time.…”