Limits...
Tracing the Attention of Moving Citizens

View Article: PubMed Central - PubMed

ABSTRACT

With the widespread use of mobile computing devices in contemporary society, our trajectories in the physical space and virtual world are increasingly closely connected. Using the anonymous smartphone data of 1 × 105 users in a major city of China, we study the interplay between online and offline human behaviors by constructing the mobility network (offline) and the attention network (online). Using the network renormalization technique, we find that they belong to two different classes: the mobility network is small-world, whereas the attention network is fractal. We then divide the city into different areas based on the features of the mobility network discovered under renormalization. Interestingly, this spatial division manifests the location-based online behaviors, for example shopping, dating, and taxi-requesting. Finally, we offer a geometric network model to help us understand the relationship between small-world and fractal networks.

No MeSH data available.


Two universal classes of behaviors and the transition of degree correlations.(A) The number of boxes N(lB) is a power law function of box length lB in the attention network, and these two variables show an exponential relationship in the mobility network. (B) In the mobility network, the degree correlation (measured by the Pearson correlation coefficient Cor(knn, k)) decreases from positive to negative when lB = 4, while the correlation remains negative in the attention network. Panel (C,D) show the transition of degree correlation in details. For the mobility network (C), the slope of data points is positive when , and the slope turns negative when . Meanwhile, the correlation is always negative in the attention network (D).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC5016808&req=5

f2: Two universal classes of behaviors and the transition of degree correlations.(A) The number of boxes N(lB) is a power law function of box length lB in the attention network, and these two variables show an exponential relationship in the mobility network. (B) In the mobility network, the degree correlation (measured by the Pearson correlation coefficient Cor(knn, k)) decreases from positive to negative when lB = 4, while the correlation remains negative in the attention network. Panel (C,D) show the transition of degree correlation in details. For the mobility network (C), the slope of data points is positive when , and the slope turns negative when . Meanwhile, the correlation is always negative in the attention network (D).

Mentions: The online and offline networks show very different properties in the renormalization process (Fig. 2B). In the attention network, the number of boxes N(lB) scales to the length of box lB, following a power-law function, . However, in the mobility network, N(lB) and lB display an exponential relationship N(lB) ~ e−β. According to Song et al.7, the mobility network is small-world and the attention network is fractal. This findings imply that online and offline human behaviors are governed by different mechanisms.


Tracing the Attention of Moving Citizens
Two universal classes of behaviors and the transition of degree correlations.(A) The number of boxes N(lB) is a power law function of box length lB in the attention network, and these two variables show an exponential relationship in the mobility network. (B) In the mobility network, the degree correlation (measured by the Pearson correlation coefficient Cor(knn, k)) decreases from positive to negative when lB = 4, while the correlation remains negative in the attention network. Panel (C,D) show the transition of degree correlation in details. For the mobility network (C), the slope of data points is positive when , and the slope turns negative when . Meanwhile, the correlation is always negative in the attention network (D).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC5016808&req=5

f2: Two universal classes of behaviors and the transition of degree correlations.(A) The number of boxes N(lB) is a power law function of box length lB in the attention network, and these two variables show an exponential relationship in the mobility network. (B) In the mobility network, the degree correlation (measured by the Pearson correlation coefficient Cor(knn, k)) decreases from positive to negative when lB = 4, while the correlation remains negative in the attention network. Panel (C,D) show the transition of degree correlation in details. For the mobility network (C), the slope of data points is positive when , and the slope turns negative when . Meanwhile, the correlation is always negative in the attention network (D).
Mentions: The online and offline networks show very different properties in the renormalization process (Fig. 2B). In the attention network, the number of boxes N(lB) scales to the length of box lB, following a power-law function, . However, in the mobility network, N(lB) and lB display an exponential relationship N(lB) ~ e−β. According to Song et al.7, the mobility network is small-world and the attention network is fractal. This findings imply that online and offline human behaviors are governed by different mechanisms.

View Article: PubMed Central - PubMed

ABSTRACT

With the widespread use of mobile computing devices in contemporary society, our trajectories in the physical space and virtual world are increasingly closely connected. Using the anonymous smartphone data of 1 × 105 users in a major city of China, we study the interplay between online and offline human behaviors by constructing the mobility network (offline) and the attention network (online). Using the network renormalization technique, we find that they belong to two different classes: the mobility network is small-world, whereas the attention network is fractal. We then divide the city into different areas based on the features of the mobility network discovered under renormalization. Interestingly, this spatial division manifests the location-based online behaviors, for example shopping, dating, and taxi-requesting. Finally, we offer a geometric network model to help us understand the relationship between small-world and fractal networks.

No MeSH data available.