Nonlinear effects of built environment on living street vitality considering spatial heterogeneity: Evidence from Xiamen Island, China

Abstract

Understanding the intricate relationship between the built environment and living street vitality is crucial for guiding the revitalization of street life. Traditional linear models often oversimplify this relationship, while machine learning models fail to account for spatial heterogeneity. To overcome these limitations, this study proposes a hybrid framework, geographically weighted-gradient boosting decision trees (GW-GBDT), which integrates geographically weighted regression (GWR) with gradient boosting decision trees (GBDT). Using Xiamen Island as a case study, we examined the nonlinear effects of built environment features on both the intensity and stability of living street vitality across four dimensions, including transport accessibility, built environment quality, daily service facilities, and morphological features. Results reveal that daily service facility density and floor area ratio are the most significant characteristics shaping intensity of vitality, while daily service facility diversity and green view ratio play dominant roles in stability of vitality. Furthermore, built environment features exhibit spatially varying threshold effects across different functional zones. These findings offer actionable insights for urban planners, emphasizing the need for localized, context-specific strategies to enhance living street vitality.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 2 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.