POLO: Program for Optimal Linear Ordering

Linear ordering is to transform a set of data objects to a one dimensional order, which preserves multivariate or graph patterns as much possible. Linear ordering is widely used to discover patterns. When applied in geographical analysis, a linear ordering can be used to facilitate the detection of space-time (Guo et al. 2006), space-multivariate, and spatial interaction patterns (Guo 2007). This program (POLO) implements a linear ordering method based on the complete linkage hierarchical clustering and an optimal cluster ordering method. The method outperforms other existing ordering methods (see the evaluation provided in (Guo and Gahegan 2006)). 

Related Publication: 
  • Guo, D. and M. Gahegan (2006). "Spatial Ordering and Encoding for Geographic Data Mining and Visualization", Journal of Intelligent Information Systems, 27(3), pp.243-266.
  • Guo, D., J. Chen, A. M. MacEachren, and K. Liao (2006), "A Visualization System for Spatio- Temporal and Multivariate Patterns (VIS-STAMP)", IEEE Transactions on Visualization and Computer Graphics, 12(6), pp. 1461-1474. 
  • Guo, D. (2007). "Visual Analytics of Spatial Interaction Patterns for Pandemic Decision Support",International Journal of Geographical Information Science, 21(8), pp. 859-877.

Java is needed to run the following software. You can verify if Java is already installed on your computer at this link: http://java.com/en/download/installed.jsp.

AttachmentSize
linearordering.jar (java package)4.96 MB
polo.zip (source code)68.6 KB
testdata.csv (sample data - multivariate data)2.16 KB
testmatrixdata.csv (sample data - matrix data)151 bytes
linearorderingmanual.pdf (manual)167.36 KB