|
| Fig. 1. Social Network in Geo-Social Simulation |
Human mobility and social networks have received considerable
attention from researchers in recent years. What has been sorely missing is a
comprehensive data set that not only addresses geometric movement patterns derived from
trajectories, but also provides social networks and causal links as to why movement
happens in the first place. To some extent, this challenge is addressed by studying
location-based social networks (LBSNs). However, the scope of real-world LBSN data sets
is constrained by privacy concerns, a lack of authoritative ground-truth, their
sparsity, and small size. To overcome these issues we have infused a novel
geographically explicit agent-based simulation framework to simulate human behavior and
to create synthetic but realistic LBSN data based on human patterns-of-life (i.e., a
geo-social simulation). Such data not only captures the location of users over time, but
also their motivation, and interactions via temporal social networks.
|
| Fig. 2. Spatial Network in Geo-Social Simulation
|
This framework models human patterns of life in urban environments based on
well-established social science theory. We simulated people living and working in
places, visiting restaurants to eat, going to bars, coffee shops, and places of
worship to socialize and effectively create an evolving social network. Our
framework is unique in the sense that we combine temporal social networks and
patterns of daily life, whereas many existing frameworks consider only one of them.
Figures 1 and 2 show a snapshot of our simulation. In an artificially generated
world, 5,000 agents go to work are created at initialization. The spatial network in
Figure 2 all agents (blue and red dots depending on whether agents are hungry or
not) in their home places outside the virtual city while the social network of
agents is shown Figure 1. Each color in Figure 1 represents a social interest. We
observe that some groups of agents having the same interest (e.g., blue or yellow)
form social clusters. The following video captures animation of the simulation
including evolution of social networks.
Once the simulation is started, we observe agents commuting to work, going to
restaurants to eat and going to pubs to meet friends. When the simulation begins,
there are no agent social networks. However, as the simulation runs, the agents make
friendships. We observe social ties quickly emerging between agents. Shortly after
the simulation has begun, the agents have self-organized into coherent social
networks. According to our model of social interaction, social ties between agents
appear and strengthen as they meet in the physical world. Every day at midnight, all
social network edges are decreased by a factor, leading to weak links to disappear,
leading to friendless agents becoming ejected from the large social network.
