Breaking the habit: Measuring and predicting departures from routine in individual human mobility

Abstract

a b s t r a c tResearchers studying daily life mobility patterns have recently shown that humans are typically highly predictable in their movements. However, no existing work has examined the boundaries of this predictability, where human behaviour transitions temporarily from routine patterns to highly unpredictable states. To address this shortcoming, we tackle two interrelated challenges. First, we develop a novel information-theoretic metric, called instantaneous entropy, to analyse an individual's mobility patterns and identify temporary departures from routine. Second, to predict such departures in the future, we propose the first Bayesian framework that explicitly models breaks from routine, showing that it outperforms current state-of-the-art predictors.

Data as Monads: How Digital Data can be Understood as the Sum of the Components in the Process of Locating it

“…This is, for example, seen in the variety of themes that are being studied such as; productivity and information diffusion (Aral et al, 2007), collaborative networks (Sonnenberg et al, 2000), interdependence and trust (Tomkins, 2001), prediction of habitual and non-habitual actions (McInerney et al, 2013), behaviour contagion (Centola, 2010) as calls, e-mail and message logs, geo-location, Bluetooth, online social networks, sensory networks, search histories and more. While some find the storing and usage of these large amounts of obtained digital data concerning and call for increased security (Perrig, 2002), others find that the 'real life' aspect of digital data has become superfluous and that computational research of sociality should instead be done in artificial society simulations built from digital data (Gilbert, 2005).…”

Neighbour discovery in opportunistic networks

“…However, these techniques often only predict social encounters, when what is necessary are methods which can predict regular encounters with strangers as well. It may also be desirable to detect departures from regular encounter patterns [33].…”

Cepr

“…In the former one, the probability of exploration was assumed to depend on the number of each user's visited locations, while in the latter one, the exploration probability was simply supposed invariant. Additionally, Exploration Prediction is also related to detecting the deviations from routines [McInerney et al 2013a], where the author studied how to predict future deviations from routines (including the visit to novel locations and the visit to familiar locations at unusual times) based on time and the immediately preceding state (deviated or not). Our work is different from this existing work in the following ways.…”