Travel behavioral analysis is crucial to the demand computation, which underpins traffic planning and optimization of transportation management and control. While traditional methods seek to descriptively learn the travel patterns through behavioral models as well as big data from multiple sources, they perform weak predictive ability of system's evolutionary features. To model and analyze the dynamics of such cyber physical social systems, this paper proposes an interpretable analytical method for travel behaviors based on knowledge automation (named as knowledge refinement). The new approach provides a feasible and reliable methodology for the analytics of heterogenous human participated systems, and is validated using one month's data of public transportation in Beijing. Results indicate that the approach not only has the state-of-the-art “reproductive” ability of actual traffic flows compare with traditional methods, but also can further reveal systemic dynamics with semantics such as the stable or divergent regions.
Knowledge Refinement: An Interpretable Analytics for Travel Behaviors Based on Knowledge Automation
24.09.2024
1358931 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Automated Knowledge-Base Refinement
British Library Conference Proceedings | 1993
|Interpretable multi-hop knowledge reasoning for gastrointestinal disease
Springer Verlag | 2025
|Knowledge Sharing Behaviors in Knowledge Management System
Springer Verlag | 2012
|Knowledge Interactions and Travel Behavior
British Library Conference Proceedings | 2009
|