Deciphering Predictability Limits in Human Mobility

Abstract

Human mobility has been studied from different perspectives. One approach addresses predictability, deriving theoretical limits on the accuracy that any prediction model can achieve in a given dataset. This approach focuses on the inherent nature and fundamental patterns of human behavior captured in the dataset, filtering out factors that depend on the specificities of the prediction method adopted. In this paper, we revisit the state-of-the-art method for estimating the predictability of a person's mobility, which, despite being widely adopted, suffers from low interpretability and disregards external factors that have been suggested to improve predictability estimation, notably the use of contextual information (e.g., weather, day of the week, and time of the day). We also conduct a thorough analysis of how this widely used method works, by looking into two different measures (one proposed by us) which are easier to understand and, as shown, capture reasonably well the effects of the original technique. Additionally, we investigate strategies to incorporate different types of contextual information into predictability estimates, and show that the benefits vary depending on the underlying prediction task. Finally, we propose and evaluate alternative estimates of predictability which, while being much easier to interpret, provide comparable results to the state-of-the-art.

Understanding individuals' proclivity for novelty seeking

“…Regardless of the applied methods (e.g., Markov chains, Naive Bayes, neural networks), the type of prediction (i.e., next-cell or next place) or the used data sets (e.g., GPS, CDR, surveys), the accuracy of prediction never reaches the coveted 100%. The reasons are manyfold: the lack of ground truth data, human beings' complex nature and behavior, and the difficulty to forecast visits to non-routinary areas and discoveries of new places [8].…”

On estimating the predictability of human mobility: the role of routine

“…For instance, Smith et al [11] showed that the limits of predictability can be refined if we exclude from the possible next locations those that are far away from the user's current position. Teixeira et al [13] quantified the impact of context on predictability estimates, showing that context does not always increase predictability.…”

“…Although previous work [10] analyzed individual human mobility in terms of exploration and preferential returns, Song et al's work and subsequent studies derived from it [4,5,[11][12][13] studied the predictability of a person's mobility considering one's mobility as a single monolithic entity. In this paper, we propose to study predictability in terms of two components, and we argue that separately studying such components can reveal important insights into the predictability of one's mobility.…”

“…5, we zoom in on the routine component to try to understand what makes a person's routine easier or harder to predict. To do that, we rely on our previously proposed metrics [13], namely regularity and stationarity, to try to understand what affects a person's routine. We also propose a novel metric, called diversity of trajectories which, together with regularity and stationarity, paint a clearer picture of what affects the predictability of the routine component of one's mobility.…”

“…We evaluate these three metrics by building regression models that use them as proxies to understand the predictability of an individual's routine. Our study relies on the analysis of two datasets of different spatial and temporal granularities, as these properties have been shown to influence predictability [11,13].…”