Discovering Collective Group Relationships
In many real-world situations, individual components of complex systems tend to form groups to interact collectively. The grouping effectuates collective relationships. On the other hand, collective relationshsips stimulate individual components to form groups. To gain clear understanding of the structure and functioning of these systems, it is necessary to identify both group formation and collective relationships at the same time. In this paper, we define the notation of collective group relationships (CGRs) between two sets of individual components and propose a method to discover CGRs from heterogeneous datasets. The method integrates canonical correlation analysis (CCA) with graph mining to find top-k CGRs. Several experimental studies are conducted on both synthetic and real-world datasets to demonstrate the effectiveness and efficiency of the proposed method.
KeywordsCollective group relationships group pair canonical correlations quasi-cliques
Unable to display preview. Download preview PDF.
- 6.Danon, L., Díaz-Guilera, A., Duch, J., Arenas, A.: Comparing community structure identification. Journal of Statistical Mechanics: Theory and Experiment P09008 (2005)Google Scholar
- 10.Lee, W., Lee, D., Lee, Y., Pawitan, Y.: Sparse Canonical Covariance Analysis for High-throughput Data. Statistical Applications in Genetics and Molecular Biology 10(1): Article 30 (2011)Google Scholar
- 15.Parkhomenko, E., Tritchler, D., Beyene, J.: Sparse Canonical Correlation Analysis with Application to Genomic Data Integration. Statistical Applications in Genetics and Molecular Biology, 8(1), Article 1 (2009)Google Scholar
- 18.Smyth, G.K.: Limma: linear models for microarray data. Statistics for Biology and Health. Bioinformatics and Computational Biology Solutions using R and Bioconductor. pp. 397-420. Springer (2005)Google Scholar
- 19.Tang, L., Liu, H., Zhang, J., Nazeri, Z.: Community Evolution in Dynamic Multi-Mode Networks. In: 14th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), Las Vegas, USA, pp. 677–685 (2008)Google Scholar
- 24.Yeung, K.Y., Medvedovic, M., Bumgarner, R.E.: From co-expression to co-regulation: how many microarray experiments do we need? Genome Biology, 5(7), Article R48 (2004)Google Scholar