Graduate Seminar: Mark He
Intertemporal Community Detection in Human Mobility Networks: Case Studies of Bikeshare Systems in US Cities
We introduce a community detection method that finds clusters in network time-series by introducing a method that finds significantly interconnected nodes across time that are either increasing, decreasing, or constant in connectivity. Significance of nodal connectivity within a set are judged as by the Weighted Configuration Null Model at each time-point, then a novel significance-testing scheme is used to determine assess connectivity at all time points and the direction of its time-trend. We apply this method to Bikeshare networks in New York City and Chicago to find patterns in human mobility among urban zones. Results show stark geographical patterns in clusters that are growing and declining in relative bike-share usage across time and potentially elucidate latent economic or demographic trends.