Semantic co-location model: a novel approach to explore the urban spatial structure with a social phenomenon

This study explores the intersection of spatial co-location patterns and social phenomena through an innovative analysis of Twitter data, addressing a gap in existing spatial co-location research that predominantly focuses on geographical phenomena. Spatial co-location pattern analysis is fundamental to understanding spatial data and enhancing geographic context-awareness in applications. While traditional studies have concentrated on identifying spatial proximity of physical features to discern interactions among geographical phenomena, this research integrates social phenomena, acknowledging the intrinsic relationship between geographic and social dynamics. Through the analysis of georeferenced Twitter data, this study identifies spatial features associated with social interactions and activities, providing a comprehensive understanding of social-spatial interplay. The research introduces an innovative Semantic Co-Location (SCL) model to analyze spatial co-location patterns from individual tweets at aggregated spatial levels. This includes developing spatial co-location mining techniques, analyzing topical categories of spatial co-location based on contextual information, and uncovering previously unknown patterns that expand current research boundaries. The findings advance our understanding of urban discourse and illuminate the relationship between place and people, specifically within spatial and social networks.