Understanding predictability and exploration in human mobility
Predictive models for human mobility have important applications in many fields including traffic control, ubiquitous computing, and contextual advertisement. The predictive performance of models in literature varies quite broadly, from over 90% to under 40%. In this work we study which underlying f...
| Main Authors: | , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
Springer
2018
|
| Online Access: | http://hdl.handle.net/1721.1/114665 https://orcid.org/0000-0002-8482-0318 |
| _version_ | 1826192933773115392 |
|---|---|
| author | Lehmann, Sune Cuttone, Andrea Gonzalez, Marta C. |
| author2 | Massachusetts Institute of Technology. Department of Civil and Environmental Engineering |
| author_facet | Massachusetts Institute of Technology. Department of Civil and Environmental Engineering Lehmann, Sune Cuttone, Andrea Gonzalez, Marta C. |
| author_sort | Lehmann, Sune |
| collection | MIT |
| description | Predictive models for human mobility have important applications in many fields including traffic control, ubiquitous computing, and contextual advertisement. The predictive performance of models in literature varies quite broadly, from over 90% to under 40%. In this work we study which underlying factors - in terms of modeling approaches and spatio-temporal characteristics of the data sources - have resulted in this remarkably broad span of performance reported in the literature. Specifically we investigate which factors influence the accuracy of next-place prediction, using a high-precision location dataset of more than 400 users observed for periods between 3 months and one year. We show that it is much 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 have strong influence on the accuracy of prediction. Finally we reveal 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. Keywords: human mobility; next-location prediction; predictability |
| first_indexed | 2024-09-23T09:31:05Z |
| format | Article |
| id | mit-1721.1/114665 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T09:31:05Z |
| publishDate | 2018 |
| publisher | Springer |
| record_format | dspace |
| spelling | mit-1721.1/1146652022-09-26T11:59:39Z Understanding predictability and exploration in human mobility Lehmann, Sune Cuttone, Andrea Gonzalez, Marta C. Massachusetts Institute of Technology. Department of Civil and Environmental Engineering Cuttone, Andrea Gonzalez, Marta C. Predictive models for human mobility have important applications in many fields including traffic control, ubiquitous computing, and contextual advertisement. The predictive performance of models in literature varies quite broadly, from over 90% to under 40%. In this work we study which underlying factors - in terms of modeling approaches and spatio-temporal characteristics of the data sources - have resulted in this remarkably broad span of performance reported in the literature. Specifically we investigate which factors influence the accuracy of next-place prediction, using a high-precision location dataset of more than 400 users observed for periods between 3 months and one year. We show that it is much 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 have strong influence on the accuracy of prediction. Finally we reveal 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. Keywords: human mobility; next-location prediction; predictability 2018-04-12T14:58:16Z 2018-04-12T14:58:16Z 2018-01 2017-08 2018-01-12T05:27:21Z Article http://purl.org/eprint/type/JournalArticle 2193-1127 http://hdl.handle.net/1721.1/114665 Cuttone, Andrea et al. "Understanding predictability and exploration in human mobility." EPJ Data Science 2018, 7 (January 2018): 2 © 2018 The Author(s) https://orcid.org/0000-0002-8482-0318 en http://dx.doi.org/10.1140/epjds/s13688-017-0129-1 EPJ Data Science Creative Commons Attribution http://creativecommons.org/licenses/by/4.0/ The Author(s) application/pdf Springer Springer Berlin Heidelberg |
| spellingShingle | Lehmann, Sune Cuttone, Andrea Gonzalez, Marta C. Understanding predictability and exploration in human mobility |
| title | Understanding predictability and exploration in human mobility |
| title_full | Understanding predictability and exploration in human mobility |
| title_fullStr | Understanding predictability and exploration in human mobility |
| title_full_unstemmed | Understanding predictability and exploration in human mobility |
| title_short | Understanding predictability and exploration in human mobility |
| title_sort | understanding predictability and exploration in human mobility |
| url | http://hdl.handle.net/1721.1/114665 https://orcid.org/0000-0002-8482-0318 |
| work_keys_str_mv | AT lehmannsune understandingpredictabilityandexplorationinhumanmobility AT cuttoneandrea understandingpredictabilityandexplorationinhumanmobility AT gonzalezmartac understandingpredictabilityandexplorationinhumanmobility |