A Parallel Spatial Co-location Pattern Mining Approach Based on Ordered Clique Growth
Co-location patterns or subsets of spatial features, whose instances are frequently located together, are particularly valuable for discovering spatial dependencies. Although lots of spatial co-location pattern mining approaches have been proposed, the computational cost is still expensive. In this paper, we propose an iterative mining framework based on MapReduce to mine co-location patterns efficiently from massive spatial data. Our approach searches for co-location patterns in parallel through expanding ordered cliques and there is no candidate set generated. A large number of experimental results on synthetic and real-world datasets show that the proposed method is efficient and scalable for massive spatial data, and is faster than other parallel methods.
KeywordsSpatial data mining Co-location patterns Ordered clique Parallel algorithm MapReduce
This work is supported by the National Natural Science Foundation of China (61472346, 61662086, 61762090), the Natural Science Foundation of Yunnan Province (2015FB114, 2016FA026), and the Project of Innovative Research Team of Yunnan Province.
- 3.Yoo, J.S., Shekhar, S.: A partial join approach for mining co-location patterns. In: The 12th Annual ACM International Workshop on Geographic Information Systems, pp. 241–249 (2004)Google Scholar
- 5.Xiao, X., Xie, X., Luo, Q., Ma, W.: Density based co-location pattern discovery. In: 16th ACM SIGSPATIAL, pp. 1–10 (2008)Google Scholar
- 6.Lin, Z., Lim, S.J.: Fast spatial co-location mining without cliqueness checking. In: International Conference on Information and Knowledge Management, pp. 1461–1462 (2008)Google Scholar
- 7.Yoo, J.S., Boulware, D., Kimmey, D.: A parallel spatial co-location mining algorithm based on MapReduce. In: IEEE International Congress on Big Data, pp. 25–31 (2014)Google Scholar