Abstract
Introduced in 2007, affinity propagation (AP) is a relatively new machine learning algorithm for unsupervised classification that has seldom been applied in geospatial applications. One bottleneck is that AP could hardly handle large data, and a serial computer program would take a long time to complete an AP calculation. New multicore and manycore computer architectures, combined with application accelerators, show promise for achieving scalable geocomputation by exploiting task and data levels of parallelism. This chapter introduces our recent progress in parallelizing the AP algorithm on a graphics processing unit (GPU) for spatial cluster analysis, the potential of the proposed solution to process big geospatial data, and its broader impact for the GIScience community.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Sample AP datasets: http://www.psi.toronto.edu/affinitypropagation/vsh/.
- 2.
- 3.
- 4.
- 5.
References
Anselin L (1995) Local indicators of spatial association—LISA. Geogr Anal 27:93–115
Bodenhofer U, Palme J, Melkonian C, Kothmeier A (2015) APCluster: an R package for affinity propagation clustering. https://cran.r-project.org/web/packages/apcluster/vignettes/apcluster.pdf
Chehdi K, Soltani M, Cariou C (2014) Pixel classification of large-size hyperspectral images by affinity propagation. J Appl Remote Sens 8(1), 083567: 1–14
Dueck D (2009) Affinity propagation: clustering data by passing messages. Dissertation, University of Toronto
Frey BJ, Dueck D (2007) Clustering by passing messages between data points. Science 315:972–976
Frey BJ, Dueck D (2008) Response to comment on “Clustering by passing messages between data points”. Science 319:726
Frey BJ, Mohammad N, Morris QD, Zhang W, Robinson MD, Mnaimneh S, Chang R, Pan Q, Laurin N, Sat E, Rossant J, Bruneau BG, Aubin JE, Blencowe BJ, Hughes TR (2005) Genome-wide analysis of mouse transcripts using exon tiling microarrays and factor graphs. Nat Genet 37(9):991–996
Guo D, Mennis J (2009) Spatial data mining and geographic knowledge discovery—an introduction. Comput Environ Urban Syst 33(6):403–408
Han J, Kamber M, Tung AKH (2001) Spatial clustering methods in data mining: a survey. In: Miller HJ, Han J (eds) Geographic data mining and knowledge discovery. Research monographs in GIS. Taylor and Francis, London, pp 201–232
Jacquez GM (2008) Spatial cluster analysis. In: Fotheringham S, Wilson J (eds) The handbook of geographic information science. Blackwell, Oxford, pp 395–416
Kwan M, Xiao N, Ding G (2014) Assessing activity pattern similarity with multidimensional sequence alignment based on a multiobjective optimization evolutionary algorithm. Geogr Anal 46:297–320
Lu Y (2000) Spatial cluster analysis for point data: location quotients versus kernel density. University Consortium for Geographical Information Science Summer Assembly, Portland, Oregon. http://dusk.geo.orst.edu/ucgis/web/oregon/papers/lu.htm
Napoleon D, Praneesh M, Subramanian MS, Sathya S (2012) Manhattan distance based affinity propagation technique for clustering in remote sensing images. Int J Adv Res Comput Sci Softw Eng (IJARCSSE) 2(3):326–330
National Research Council (2006a) Beyond mapping: meeting national needs through enhanced geographic information science. National Academies Press, Washington, DC
National Research Council (2006b) Priorities for GEOINT research at the National Geospatial-Intelligence Agency. National Academies Press, Washington, DC
Tobler W (1979) Cellular geography. In: Gale. S, Olsson G (eds) Philosophy in geography. Reidel, Dortrecht, pp 379–386
Xia HY, Chen XY, Guo P (2009) A shadow detection method for remote sensing images using affinity propagation algorithm. In: Proceedings 2009 IEEE international conference on systems, man and cybernetics, pp 3116–3121. doi:10.1109/ICSMC.2009.5346147
Yang C, Bruzzone L, Sun F, Lu L, Guan R, Liang Y (2010) A fuzzy-statistics-based affinity propagation technique for clustering in multispectral images. IEEE Trans Geosci Remote Sens 48(6):2647–2659
Acknowledgments
This research was partially supported by the National Science Foundation (NSF) through the award NSF SMA-1416509 “IBSS: Spatiotemporal Modeling of Human Dynamics across Social Media and Social Networks” and the National Institutes of Health (NIH) through the award NIH 1R21CA182874-01 “Reducing Physician Distribution Uncertainty in Spatial Accessibility Research.” Any opinions, findings, recommendations, or conclusions expressed in this material are those of the author and do not necessarily reflect the views of the NSF or NIH.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Appendices
Appendix 1
CUDA/GPU program that reflects the preceding work flow for AP implementation in comparison with the original serial C program
Appendix 2
The host program and device program in CUDA/GPU code
Rights and permissions
Copyright information
© 2017 Springer International Publishing Switzerland
About this paper
Cite this paper
Shi, X. (2017). Parallelizing Affinity Propagation Using Graphics Processing Units for Spatial Cluster Analysis over Big Geospatial Data. In: Griffith, D., Chun, Y., Dean, D. (eds) Advances in Geocomputation. Advances in Geographic Information Science. Springer, Cham. https://doi.org/10.1007/978-3-319-22786-3_32
Download citation
DOI: https://doi.org/10.1007/978-3-319-22786-3_32
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-22785-6
Online ISBN: 978-3-319-22786-3
eBook Packages: Earth and Environmental ScienceEarth and Environmental Science (R0)