A network model of happiness at destinations

Understanding the antecedents and consequences of happiness at destinations is critical for building livable and sustainable communities for residents and tourists. Big data and social signals provide new opportunities to unpack the driving forces of happiness. For this study, geotagged social media data, physical environment data, and economic data are utilized to shed light on how neighborhood factors shape happiness. An interdisciplinary approach is adopted to integrate natural language processing, spatial analysis, network science, and statistical modeling. The results indicate that (1) crimes are negatively associated with neighborhood happiness; (2) visitor check-in activity mediates the relationship between places of interest and neighborhood happiness; (3) happy neighborhoods with similar happiness levels share higher numbers of common happy visitors, which implies that happy neighborhoods share attributes that attract happy visitors. This research contributes to theories regarding how neighborhood attributes may shape happiness, and demonstrates how big data can be used to characterize human-environment relationships for happiness-related research. Planners and tourism stakeholders can improve neighborhood happiness by engaging with residents and tourists to evaluate the current physical conditions of neighborhoods and develop context-sensitive plans and projects.