Search for a command to run...
It has been recently claimed that human movement is highly predictable. While an upper bound of 93% predictability was shown, this was based upon human movement trajectories of very high spatiotemporal granularity. Recent studies reduced this spatiotemporal granularity down to the level of GPS data, and under a similar methodology results once again suggested a high predictability upper bound (i.e. 90% when movement was quantized down to a spatial resolution approximately the size of a large building). In this work we reconsider the derivation of the upper bound to movement predictability. By considering real-world topological constraints we are able to achieve a tighter upper bound, representing a more refined limit to the predictability of human movement. Our results show that this upper bound is between 11-24% less than previously claimed at a spatial resolution of approx. 100m×100m, with a greater improvement for finer spatial resolutions. This indicates that human mobility is potentially less predictable than previously thought. We provide an in-depth examination of how varying the spatial and temporal quantization affects predictability, and consider the impact of corresponding limits using a large set of real-world GPS traces. Particularly at fine-grained spatial quantizations, where a significant number of practical applications lie, these new (lower) upper limits raise serious questions about the use of location information alone for prediction, contributing more evidence that such prediction must integrate external variables.