Directed Laplacian Kernels for Link Analysis
Application of kernel methods to link analysis is presented. Novel kernels based on directed graph Laplacians are proposed and their application as measures of relatedness between nodes in a directed graph is presented. The kernels express relatedness and take into account the global importance of the nodes in a citation graph. Limitations of existing kernels are given with a discussion how they are addressed by directed Laplacian kernels. Links between the kernels and PageRank ranking algorithm are also presented. The proposed kernels are evaluated on a dataset of scientific bibliographic citations.
KeywordsLink Analysis Directed Graph Undirected Graph Transition Probability Matrix Adjacency Matrice
Unable to display preview. Download preview PDF.
- Brin, S., Page, L., Motwani, R., Winograd, T.: The PageRank citation ranking: bringing order to the Web. Technical Report 1999-0120, Computer Science Department, Stanford University (1999)Google Scholar
- Ito, T., Shimbo, M., Kudo, T., Matsumoto, Y.: Application of kernels to link analysis. In: Proceedings of KDD, pp. 586–592 (2005)Google Scholar
- White, S., Smyth, P.: Algorithms for estimating relative importance in networks. In: Proceedings of SIGKDD, pp. 266–275 (2003)Google Scholar
- Schölkopf, B., Smola, A.: Learning with Kernels. MIT Press, Cambridge (2001)Google Scholar
- Kandola, J., Shawe-Taylor, J., Cristianini, N.: Learning semantic similarity. In: NIPS 15, MA, pp. 657–664. MIT Press, Cambridge (2003)Google Scholar
- Kondor, R.I., Lafferty, J.: Diffusion kernels on graphs and other discrete structures. In: Proceedings of MGTS at ECML/PKDD, pp. 315–322 (2002)Google Scholar
- Shawe-Taylor, J., Cristianini, N.: Kernel Methods for Pattern Analysis. Cambridge University Press, Cambridge (2004)Google Scholar
- Smola, A.J., Kondor, R.: Kernels and Regularization on Graphs. In: Proceedings of Conference on Learning Theory, COLT/KW (2003)Google Scholar