Abstract
Understanding the spatio-temporal dynamics of cities is important for many applications including urban planning, zoning, and real-estate construction. So far, much of this understanding came from traditional surveys conducted by persons or by leveraging mobile data in the form of Call Detailed Records. However, the high financial and human cost associated with these methods make the data availability very limited. In this paper, we investigate the use of large scale and publicly available user contributed content, in the form of social media posts to understand the urban dynamics of cities. We build activity time series for different cities, and different neighborhoods within the same city to identify the different dynamic patterns taking place. Next, we conduct a cluster analysis on the time series to understand the spatial distribution of patterns in the city.
R. Zegour was intern at QCRI during this work.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Link anonymized for the submission.
- 2.
- 3.
- 4.
- 5.
References
Arribas-Bel, D., Kourtit, K., Nijkamp, P., Steenbruggen, J.: Cyber cities: social media as a tool for understanding cities. Appl. Spat. Anal. Policy 8(3), 231–247 (2015). https://doi.org/10.1007/s12061-015-9154-2
Batty, M.: Cities and Complexity: Understanding Cities with Cellular Automata, Agent-Based Models, and Fractals. The MIT Press, Cambridge (2007)
CIESIN, SEDAC: Gridded population of the world, version 4 (GPWV4): population density. Technical report, Center for International Earth Science Information Network - CIESIN - Columbia University. NASA Socioeconomic Data and Applications Center (SEDAC)
Çolak, S., Lima, A., González, M.C.: Understanding congested travel in urban areas. Nature Commun. 7 (2016)
Cranshaw, J., Schwartz, R., Hong, J.I., Sadeh, N.: The livehoods project: utilizing social media to understand the dynamics of a city (2012)
González, M.C., Hidalgo, C.A., Barabási, A.L.: Understanding individual human mobility patterns. Nature 453(7196), 779–782 (2008)
Grauwin, S., Sobolevsky, S., Moritz, S., Gódor, I., Ratti, C.: Towards a comparative science of cities: using mobile traffic records in New York, London, and Hong Kong. In: Helbich, M., Jokar Arsanjani, J., Leitner, M. (eds.) Computational Approaches for Urban Environments. GE, vol. 13, pp. 363–387. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-11469-9_15
Keogh, E., Ratanamahatana, C.A.: Exact indexing of dynamic time warping. Knowl. Inf. Syst. 7(3), 358–386 (2005)
Ketchen Jr., D.J., Shook, C.L.: The application of cluster analysis in strategic management research: an analysis and critique. Strat. Manag. J. 17, 441–458 (1996)
Krumm, J., Kun, A.L., Varsanyi, P.: TweetCount: urban insights by counting tweets. In: Proceedings of the 2017 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2017 ACM International Symposium on Wearable Computers, pp. 403–411. ACM (2017)
MacQueen, J., et al.: Some methods for classification and analysis of multivariate observations. In: Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, Oakland, CA, USA, vol. 1, pp. 281–297 (1967)
Morenoff, J.D., Sampson, R.J., Raudenbush, S.W.: Neighborhood inequality, collective efficacy, and the spatial dynamics of urban violence. Criminology 39(3), 517–558 (2001)
Noulas, A., Scellato, S., Mascolo, C., Pontil, M.: Exploiting semantic annotations for clustering geographic areas and users in location-based social networks. Soc. Mob. Web 11(2) (2011)
Nyhan, M., et al.: Exposure track: The impact of mobile-device-based mobility patterns on quantifying population exposure to air pollution. Environ. Sci. Technol. 50(17), 9671–9681 (2016)
Pham, D.T., Dimov, S.S., Nguyen, C.D.: Selection of k in k-means clustering. Proc. Inst. Mech. Eng. Part C J. Mech. Eng. Sci. 219(1), 103–119 (2005)
Ratti, C., Frenchman, D., Pulselli, R.M., Williams, S.: Mobile landscapes: using location data from cell phones for urban analysis. Environ. Plan. B Plan. Des. 33(5), 727–748 (2006)
Reades, J., Calabrese, F., Sevtsuk, A., Ratti, C.: Cellular census: explorations in urban data collection. IEEE Pervasive Comput. 6(3), 30–38 (2007). https://doi.org/10.1109/MPRV.2007.53
Sakoe, H., Chiba, S.: Dynamic programming algorithm optimization for spoken word recognition. IEEE Trans. Acoust. Speech Signal Process. 26(1), 43–49 (1978)
Steiger, E., Westerholt, R., Resch, B., Zipf, A.: Twitter as an indicator for whereabouts of people? Correlating twitter with UK census data. Comput. Environ. Urban Syst. 54, 255–265 (2015)
Theobald, D.M.: Land-use dynamics beyond the american urban fringe. Geogr. Rev. 91(3), 544–564 (2001)
Tibshirani, R., Walther, G., Hastie, T.: Estimating the number of clusters in a data set via the gap statistic. J. R. Stat. Soc. Ser. B (Stat. Methodol.) 63(2), 411–423 (2001)
Toole, J.L., Ulm, M., González, M.C., Bauer, D.: Inferring land use from mobile phone activity. In: Proceedings of the ACM SIGKDD International Workshop on Urban Computing, pp. 1–8. ACM (2012)
UNHABITAT: World cities report 2016: urbanization and development. Technical report, UN HABITAT (2016)
Wang, X., Mueen, A., Ding, H., Trajcevski, G., Scheuermann, P., Keogh, E.: Experimental comparison of representation methods and distance measures for time series data. Data Min. Knowl. Discov. 26, 275–309 (2013)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Appendix A Related Work
Appendix A Related Work
Social Media and Urban Studies. Using social media data to characterize urban dynamics in neighborhoods is not new [1, 5, 10, 13, 19]. Therefore, we analyze in the following major works and try to contrast them to our own. In a study by [13], authors looked at modeling human activity and geographical areas via spectral clustering of Foursquare data. The main idea is to split a region into equally sized rectangles, and then characterize each rectangle with a vector containing the total number Foursquare places belonging to different categories. While this approach is interesting, it completely dismisses the temporal aspect of human activity and focuses solely on the geographical distribution and popularity of places (e.g. shops, restaurants, schools, etc.) Our framework’s main contribution is to allow a spatio-temporal analysis of human dynamics in cities. The Livehoods Project is another influential work in this area [5]. It aims at using Foursquare data for explaining dynamics in cities, and use for that spectral clustering on types of Foursquare venues present in different areas of a city. However, similar the work by [13], this project ignores the temporal aspect in the data, which we believe is very important in modeling human dynamics. Indeed, two areas with similar facilities may show different temporal behaviors (e.g. people may stay late at night in one, but not in another.) More recently, [10] demonstrated that using tweet counts could help identifying the land use profile of a neighborhood in a city. Authors used temporal features such as average number of tweets observed within different hours of the day to train a classifiers to label neighborhoods. The main weakness of this approach resides in the many heuristics introduced in defining land use profiles and selecting the ones to test against. Indeed, while land use has already a well-known classification into commercial, business, residential, industrial, etc. Authors provide their own definition of land use profiles on the only basis of resident and business count. A typical profile in their case would be a one that represents a neighborhood (a square cell) with low number of residents and high number of businesses. Our work is different from this one in two aspects. First we introduce a clustering approach using dynamic time warping on weekly time-series. Second, we study well-defined neighborhoods (represented as polygons) that correspond to actual administrative zones.
Using Mobile Data for Urban Analytics. The wide-spread and adoption of mobile phone technologies have allowed telecommunication companies to gather massive data sets about the spatio-temporal daily activities of people that yield better understanding of how our cities function [2]. These rich mobile phone datasets have unlocked the potential for several urban related applications such as traffic congestion [4], human mobility patterns [6], exposure to air pollution [14], and sensing urban dynamics [16] to name but a few. One of the closest work to ours is the one done by MIT’s Senseable City LabFootnote 5 in which they partnered with different telecommunication companies based in Europe and used the call data records (CDRs) to profile cities and neighborhoods using typical weekly signatures (TWS) in the format of time-series. Reades et al. [17] present a nice overview of their related urban computing projects in. Following the same line of research, Grauwin et al. [7] proposed a framework for comparative science of cities by comparing the spatio-temporal dynamics of neighborhoods, represented as time-series featuring different mobile phone activities such as: calls, messages, and data usage. Toole et al. [22] propose also to characterize neighborhoods using mobile data based time-series that are used to classify land use of different neighborhoods in Boston. The main finding is that time-series of mobile activity can be used to figure out what type of urban activities are taking place in different areas by inferring the land use of those areas. Examples of land use classes considered in the study are: residential, commercial and industrial.
Rights and permissions
Copyright information
© 2018 Springer Nature Switzerland AG
About this paper
Cite this paper
Abbar, S., Zanouda, T., Al-Emadi, N., Zegour, R. (2018). City of the People, for the People: Sensing Urban Dynamics via Social Media Interactions. In: Staab, S., Koltsova, O., Ignatov, D. (eds) Social Informatics. SocInfo 2018. Lecture Notes in Computer Science(), vol 11186. Springer, Cham. https://doi.org/10.1007/978-3-030-01159-8_1
Download citation
DOI: https://doi.org/10.1007/978-3-030-01159-8_1
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-01158-1
Online ISBN: 978-3-030-01159-8
eBook Packages: Computer ScienceComputer Science (R0)