Authors: Dan Cheng*, George Mason University, Dieter Pfoser, George Mason University
Topics: Urban Geography, Spatial Analysis & Modeling, Geographic Information Science and Systems
Keywords: Social Media, Urban Mobility, Trajectory, Network Decomposition
Session Type: Paper
Presentation File: No File Uploaded
Geo-tagged location information becomes a powerful data source to help with the geospatial science area in the past decade. Given the important role of online check-in information from social media data, this research aims to extract meaningful movement information on individuals and understand travel patterns in the urban area. In this study, we firstly analyze the geo-tagged Twitter trajectories in the Washington DC metropolitan area and develop a heuristic algorithm to generate a movement network that summarizes human mobilities. Secondly, in order to understand the travel behavior of people, we reveal the structure of the movement network through the k-core decomposition method and extract a hierarchical network structure with three layers: core, bridge and periphery layer. Through the analysis of the three network subsets, we have a better understanding of the global connections and local connections between individuals and locations. Last but not the least, we conduct a community detection of our movement network in order to understand the structure of the network and visualize the relationship between different communities through an interactive web mapping application we developed. The findings in this study will provide valuable predictions of traffic volume on the road network or the metro network and could be helpful to improve public transportation services in the metropolitan area of large cities. This method will offer a new way to take advantage of social media data in the current big data era.
To access contact information login