Open Collections

Abstract

Street vitality is an important indicator of urban attractiveness and sustainable development, and it has become a central topic in contemporary urban planning and research. Using the PRISMA methodology, this review systematically examines four major technologies including machine learning (ML), space syntax, GPS, and sensors, together with six categories of data that are commonly used in street vitality studies. The analysis traces the methodological development of these approaches and identifies application trends across both macro and micro spatial scales. ML has become the leading technology in this field, showing strong performance in dynamic modeling, pattern recognition, and the integration of multiple data sources. GPS provides high temporal accuracy for tracking mobility and identifying spatiotemporal dynamics. UAVs and sensor networks make it possible to observe environmental and behavioral responses in real time. When combined, these technologies support four main research themes: the built environment and vitality, pedestrian mobility and urban dynamics, spatial and visual characterization, and social interaction. Other complementary data sources, including social media, online maps, surveys, and government statistics, expand analytical coverage and improve contextual interpretation across different spatial and cultural settings. The review emphasizes the need to connect advanced technologies and diverse data sources with broader concerns of governance, ethics, and civic participation, while maintaining a focus on methodological and data-based synthesis. By clarifying the technological pathways and data foundations of street vitality research, this study provides a structured reference for researchers, urban designers, and policymakers who aim to develop evidence-based and socially responsive frameworks for urban space evaluation and planning.