This study explores the intricate relationship between human sentiment on social media data, herein tweet posts on X platform, urban building characteristics, and the socio-spatial dynamics of New York City (NYC) boroughs. Leveraging Natural Language Processing (NLP) techniques, particularly sentiment analysis, augmented by the capabilities of transformer deep learning models, RoBERTa, the study places particular emphasis on the term ‘Stay-at-Home’ to encapsulate the pronounced shift in building occupancy during the pandemic's inaugural year. This focus intertwines with pivotal terms like ‘Energy Bill’ and ‘HVAC’, shedding light on their interconnected implications. The sentiment analysis leverages data from New York City's PLUTO and the Department of Energy's LEAD databases to emotional disparities connected to urban building characteristics as well as demographic and socioeconomic factors. This analytical approach unravels prevailing public emotions and extends the discussion to include energy justice concerns, viewing them through the lens of the city's built infrastructure. The research uncovers profound disparities in the built environment and the allocation of resources in NYC, highlighting the critical need to embrace a spatial justice framework for a sustainable future. This research can aid designers, planners, and policymakers in their efforts to promote equitable and inclusive urban development.
Read full abstract