Data-driven urban human activity mining has become a hot topic of urban dynamic modeling and analysis. Semantic activity chain modeling with activity purpose provides scientific methodological support for the analysis and decision-making of human behavior, urban planning, traffic management, green sustainable development, etc. However, the spatial and temporal uncertainty of the ubiquitous mobile sensing data brings a huge challenge for modeling and analyzing human activities. Existing approaches for modeling and identifying human activities based on massive social sensing data rely on a large number of valid supervised samples or limited prior knowledge. This paper proposes an effective methodology for building human activity chains based on mobile phone signaling data and labeling activity purpose semantics to analyze human activity patterns, spatiotemporal behavior, and urban dynamics. We fully verified the effectiveness and accuracy of the proposed method in human daily activity process construction and activity purpose identification through accuracy comparison and spatial-temporal distribution exploration. This study further confirms the possibility of using big data to observe urban human spatiotemporal behavior.
Stored only when both papers are in the corpus — 22 citations exist upstream.
Imported from OpenAlex (CC0) — nothing here is generated.