measuring and predicting departures from routine in human mobility

46
Measuring and Predicting Departures from Routine in Human Mobility Dirk Gorissen | @elazungu PyData London - 23 February 2014

Upload: dirk-gorissen

Post on 26-Jun-2015

275 views

Category:

Technology


1 download

DESCRIPTION

Understanding human mobility patterns is a significant research endeavor that has recently received considerable attention. Developing the science to describe and predict how people move from one place to another during their daily lives promises to address a wide range of societal challenges: from predicting the spread of infectious diseases, improving urban planning, to devising effective emergency response strategies. This presentation will discuss a Bayesian framework to analyse an individual’s mobility patterns and identify departures from routine. It is able to detect both spatial and temporal departures from routine based on heterogeneous sensor data (GPS, Cell Tower, social media, ..) and outperforms existing state-of-the-art predictors. Applications include mobile digital assistants (e.g., Google Now), mobile advertising (e.g., LivingSocial), and crowdsourcing physical tasks (e.g., TaskRabbit).

TRANSCRIPT

  • 1. Measuring and Predicting Departures from Routine in Human Mobility Dirk Gorissen | @elazungu PyData London - 23 February 2014

2. Me?www.rse.ac.uk 3. Human Mobility - Credits University of Southampton BAE Systems ATC James McInerney Sebastian Stein Alex Rogers Nick JenningsDave NicholsonReference: J. McInerney, S. Stein, A. Rogers, and N. R. Jennings (2013). Breaking the habit: measuring and predicting departures from routine in individual human mobility. Pervasive and Mobile Computing, 9, (6), 808-822. Submitted KDD paper 4. Beijing Taxi rides Nicholas Jing Yuan (Microsoft Research) 5. Human Mobility London in Motion - Jay Gordon (MIT) 6. Human Mobility: Inference Functional Regions of a city Nicholas Jing Yuan (Microsoft Research) 7. Human Mobility: Inference Jay Gordon (MIT) 8. Human Mobility: Inference Cross cuts many fields: sociology, physics, network theory, computer science, epidemiology, MIT PNAS 9. Project InMind Project InMind announced on 12 Feb $10m Yahoo-CMU collaboration on predicting human needs and intentions 10. Human Mobility Human mobility is highly predictable Average predictability in the next hour is 93% [Song 2010] Distance little or no impact High degree of spatial and temporal regularity Spatial: centered around a small number of base locations Temporal: e.g., workweek / weekendwe find a 93% potential predictability in user mobility across the whole user base. Despite the significant differences in the travel patterns, we find a remarkable lack of variability in predictability, which is largely independent of the distance users cover on a regular basis. 11. Temporal Regularity [Herder 2012] [Song 2010] 12. Spatial Regularity [Herder 2012] [Song 2010] 13. Breaking the Habit However, regular patterns not the full story travelling to another city on a weekend break or while on sick leaveBreaks in regular patterns signal potentially interesting events Being in an unfamiliar place at an unfamiliar time requires extra context aware assistance E.g., higher demand for map & recommendation apps, mobile advertising more relevant, Predict future departures from routine? 14. Applications Optimize public transport Insight into social behaviour Spread of disease (Predictive) Recommender systems Based on user habits (e.g., Google Now, Sherpa)Context aware advertising Crime investigation Urban planning Obvious privacy & de-anonymization concerns -> Eric Drass talk 15. Human Mobility: Inference London riots commute 16. Modeling Mobility Entropy measures typically used to determine regularity in fixed time slots Well understood measures, wide applicability Break down when considering prediction or higher level structureModel based Can consider different types of structure in mobility (i.e., sequential and temporal) Can deal with heterogeneous data sources Allows incorporation of domain knowledge (e.g., calendar information) Can build extensions that deal with trust Allows for prediction Bayesian approach distribution over locations enables use as a generative model 17. Bayes Theorem 18. Bayesian Networks Bottom up: Grass is wet, what is the most likely cause? Top down: Its cloudy, what is the probability the grass is wet? 19. Hidden Markov Model Simple Dynamic Bayesian Network Shaded nodes are observed 20. Probabilistic Models Model can be run forwards or backwards Forwards (generation): parameters -> dataE.g., use a distribution over word pair frequencies to generate sentences 21. Probabilistic Models Model can be run backwards Backwards (Inference): data -> parameters 22. Building the model We want to model departures from routine Assume assignment of a person to a hidden location at all time steps (even when not observed) Discrete latent locations Correspond to points of interest e.g., home, work, gym, train station, friend's house 23. Latent Locations Augment with temporal structure Temporal and periodic assumption to behaviour e.g., tend to be home each night at 1am e.g., often in shopping district on Sat afternoon 24. Add Sequential Structure Added first-order Markov dynamics e.g., usually go home after work can extend to more complex sequential structures 25. Add Departure from Routine zn = 0 : routine zn = 1 : departure from routine 26. Sensors Noisy sensors, e.g., cell tower observations observed: latitude/longitude inferred: variance (of locations) 27. Reported Variance E.g., GPS observed: latitude/longitude, variance 28. Trustworthiness E.g., Eyewitness observed: latitude/longitude, reported variance inferred: trustworthiness of observation single latent trust value(per time step & source) 29. Full Model 30. Inference 31. Inference is Challenging Exact inference intractable Can perform approximate inference using: Expectation maximisation algorithm Gibbs sampling, or other Markov chain Monte Carlo Fast But point estimates of parametersFull distributions (converges to exact) But slowVariational approximation Full distributions based on induced factorisation of model And fast 32. Variational Approximation Advantages Straightforward parallelisation by user Months of mobility data ~ hours Updating previous day's parameters ~ minutes Variational approximation amenable to fully online inferenceM. Hoffman, D. Blei, C. Wang, and J. Paisley. Stochastic variational inference. arXiv:1206.7051, 2012 33. Model enables Inference Exploration/summarisation location departures from routine noise characteristics of observations trust characteristics of sensorsparameters have intuitive interpretationsPrediction Future mobility (given time context) Future departures from routine 34. Performance Nokia Dataset (GPS only) [McInerney 2012] 35. Performance 36. Performance Synthetic dataset with heterogeneous, untrustworthy observations. Parameters of generating model learned from OpenPaths dataset 37. Performance 38. Implementation Backend inference and data processing code all python UI to explore model predictions & sanity check numpy scipy matplotlib flask d3.js leaflet.js kockout.jsFuture Gensim, pymc, bayespy, Probabilistic programming 39. Map View: Observed 40. Map View: Inferred 41. Departures from Routine: Temporal 42. Departures from Routine: Spatial 43. Departures from Routine: Combined 44. Departures from Routine 45. Conclusion & Future Work Summary Novel model for learning and predicting departures from routineLimitations Need better ground truth for validation Finding ways to make the model explain why each departure from routine happened. Needs more data (e.g., from people who know each other, using weather data, app usage data, ).Future Work Incorporating more advanced sequential structure into the model e.g., hidden semi-Markov model, sequence memoizerSupervised learning of what interesting" mobility looks like More data sources Online inference Taxi drivers 46. Questions? Thank you. [email protected]| @elazunguReference: J. McInerney, S. Stein, A. Rogers, and N. R. Jennings (2013). Breaking the habit: measuring and predicting departures from routine in individual human mobility. Pervasive and Mobile Computing, 9, (6), 808-822.