Nonnegative Matrix Factorisation of Bike Sharing System Temporal Network

Invalid Date (modified: Oct 14, 2016) NIPS 2016 workshop MLITS submission readers: everyone
  • Abstract: In recent years, bike sharing systems have become very popular in many major cities. Thanks to the data they generate, their activity can be tracked down, giving an overall view of how human activities are spread over time and space. We propose in the present article a novel method to extract mobility patterns that occur in such large-scale transportation systems. The trips made by the users are first represented as flows between the different stations of the system, describing a network whose structure evolves over time. A decomposition technique is then proposed using non-negative matrix factorisation, to express the resulting temporal networks as a mixture of sub-networks, each of them characterising the different behaviours of users over time and space. This method is applied on the Lyon's bike sharing system, and it is emphasised that key spatio-temporal elements of urban activity are retrieved, capturing known phenomena such as commuting. This approach could be easily extended to large-scale transportation systems exhibiting a network structure, paving the way to an unsupervised modelling of mobility patterns.
  • Conflicts:
0 Replies