Voices from the city

Sentiment across public open space types in Hong Kong before and during Covid-19

Bibliographic Data

ID24227232
AuthorsYan Qian (0000-0001-8340-354X), Sifan Cheng (0000-0002-2303-7151), Yunmi Park (0000-0002-7924-6518)
Year2026
Volume179
Pages107652
Publication date2026-12-01
Open AccessNo
TypeARTICLE
VenueCities (JOURNAL)
Journal identifiersISSN: 0264-2751 • E-ISSN: 1873-6084
PublisherElsevier BV (PUBLISHER)
DOI10.1016/j.cities.2026.107652
OpenAlexW7221785983
LanguageEN
References cited84

Understanding public sentiment toward public open spaces (POSs) is essential for sustainable urban planning, as these spaces support ecological, social, cultural, and recreational functions. However, despite the diversity of POSs, there has been limited research to systematically examine how public sentiment varies across different POS types. In addition to commercial and recreational features, environmental features such as greenery and waterfronts play a key role in shaping public perceptions. Meanwhile, the COVID-19 pandemic has altered how these spaces are used and may have influenced emotional responses to environmental features. Against this background, this study examined the relationships between environmental features and public sentiment across three POS types (active, passive, and mixed), applying BERT-based sentiment analysis to more than 1.75 million geotagged tweets posted in Hong Kong between 2019 and 2023. The results show that (1) public sentiment toward POSs was generally positive, with active POSs eliciting stronger positive sentiment than passive ones; (2) greenery was positively associated with sentiment across all POS types and periods; (3) during the pandemic, the positive association between sentiment and commercial/recreational features weakened in mixed POSs, while there were directional changes in several other associations: sentiment became positively associated with transportation and art/culture in mixed POSs, whereas sentiment became negatively associated with waterfront features in mixed and passive POSs. These findings reveal how the associations between public sentiment and environmental features vary across POS types and in relation to the pandemic, offering empirical support for designing resilient and emotionally supportive POSs in the post-pandemic era.

COVID-19 pandemic · Hong Kong · pre- and post-pandemic comparison · Public open space · urban residents' perceptions · Place Attachment and Urban Studies · Sentiment Analysis and Opinion Mining · Urban Green Space and Health

Tools

Open DOIOpen Access