Small Area Analysis of Daily Urban Mobility Patterns

Presented at the RGS-IBG conference. Proposes methods of augmenting Census data with 'Big Data' from social media.

Transcript

1. Small Area Analysis of Daily Urban Mobility Patterns Mark Birkin, Robin Lovelace, Nick Malleson, Philip Cross
2. Overview • Background & Research Questions • Data Sources • Methods • Results • Discussion
3. Background • Inexorable rise of “Big Data” and its assimilation within the academic community – Store loyalty cards – Mobile telephones – Smart ticketing – Financial transactions – Social media – Internet use
4. 2014-8-28
5. Background – Retail Example Number of recorded transactions by district for Newquay store. Data shown for selected weeks in August 2010, October 2010 and January 2011.
6. Background – Search Engine Example • Google flu trends… • http://www.google.org/flutrends/about/how.html • Ginsberg J, Mohebbi M, Patel R, Brammer L, Smolinski M, Brilliant L (2009) Detecting influenza epidemics using search engine query data, Nature, 457, doi: 10.1038/nature07634
7. Social Media Data (Twitter) Java, A., Song, X., Finin, T., Tseng, B. (2007) Why we twitter: understanding microblogging usage and communities, Proceedings of the 9th WebKDD and 1st SNA-KDD 2007 workshop on Web mining and social network analysis Sakaki, T., Okazaki, M., Matsuo, Y. (2010) Earthquake shakes Twitter users: real-time event detection by social sensors, Proceedings of the 19th international conference on World wide web Tumasjan, A., Sprenger, T., Sandner, P., Welpe, I. (2010) Predicting Elections with Twitter: What 140 Characters Reveal about Political Sentiment, Proceedings of the Fourth International AAAI Conference on Weblogs and Social Media. Peter Sheridan Dodds, P., Harris, K., Kloumann, I., Bliss, C., Danforth, C. (2011), Temporal Patterns of Happiness and Information in a Global Social Network: Hedonometrics and Twitter, PLoS One, DOI: 10.1371/journal.pone.0026752
8. 2014-8-28 Hurricane "Sandy" tweets. Blue: "Donate". Red: "Closed"
9. 2014-8-28
10. 2014-8-28 Telefonica data - it's not just Twitter
11. Journey-to-Work Data
12. Research Question(s) • To what extent are Census Journey-to-Work flows replicated by social media data – Can we generate mutually consistent models from the two datasets? – What can social media tell us beyond the workplace data?
13. Data • Twitter
14. Data
16. Data
17. Method • 4 stage process to identify 'work tweets' – Geographical filter – Temporal filter – Textual filter – Repetition • Space-time analysis ('spacetime', clustering) • NLP • Anchor-points
18. Method: Space-time analysis
20. Method: NLP Birkin, M., Harland, K., & Malleson, N. (2013). The classification of space-time behaviour patterns in a British city from crowd-sourced data. In Lecture Notes in Computer Science (Vol. 7974 LNCS, pp. 179–192). doi:10.1007/978-3-642-39649-6-13
21. Method: Anchor Points Malleson, N., & Birkin, M. 2014. New Insights into Individual Activity Spaces using Crowd-Sourced Big Data.
22. Results: Maps
23. Results: Models Twitter on x; census on y
24. Results: Extensions • Activities • Spheres of influence – Interaction with local surroundings • Up-to-date • Day-to-day • Seasonal
27. Discussion and Conclusion • Jury still out on utility of social media data • Further refinements – Spatial extension – Triangulation vs mobile phones etc (NARSC paper) – Reweight (MSM, geodems...) – More data (masts, interpolation) • In conclusion: – social media are growing and increasingly available – still much work to do on validation – already providing insights into flow patterns