Understanding Predictability and Exploration in Human Mobility

TL;DR

This paper investigates factors affecting the accuracy of human mobility prediction models, highlighting the roles of data resolution, exploration, and temporal factors, and providing insights into the limits of predictability.

Contribution

It identifies key factors influencing prediction accuracy, such as data resolution and exploration, and quantifies the impact of new location visits on predictability.

Findings

Higher accuracy in time-bin location prediction compared to next-place prediction

Data resolution significantly affects prediction accuracy

20-25% of transitions are to new locations, with 70% visited only once

Abstract

Predictive models for human mobility have important applications in many fields such as traffic control, ubiquitous computing and contextual advertisement. The predictive performance of models in literature varies quite broadly, from as high as 93% to as low as under 40%. In this work we investigate which factors influence the accuracy of next-place prediction, using a high-precision location dataset of more than 400 users for periods between 3 months and one year. We show that it is easier to achieve high accuracy when predicting the time-bin location than when predicting the next place. Moreover we demonstrate how the temporal and spatial resolution of the data can have strong influence on the accuracy of prediction. Finally we uncover that the exploration of new locations is an important factor in human mobility, and we measure that on average 20-25% of transitions are to new places, and approx. 70% of locations are visited only once. We discuss how these mechanisms are important factors limiting our ability to predict human mobility.