Back
Efficient Graphlet Kernels for Large Graph Comparison
State-of-the-art graph kernels do not scale to large graphs with hundreds of nodes and thousands of edges. In this article we propose to compare graphs by counting {it graphlets}, ie subgraphs with $k$ nodes where $k in { 3, 4, 5 }$. Exhaustive enumeration of all graphlets being prohibitively expensive, we introduce two theoretically grounded speedup schemes, one based on sampling and the second one specifically designed for bounded degree graphs. In our experimental evaluation, our novel kernels allow us to efficiently compare large graphs that cannot be tackled by existing graph kernels.
@inproceedings{5664, title = {Efficient Graphlet Kernels for Large Graph Comparison}, journal = {Proceedings of the Twelfth International Conference on Artificial Intelligence and Statistics (AIStats 2009)}, booktitle = {JMLR Workshop and Conference Proceedings Volume 5: AISTATS 2009}, abstract = {State-of-the-art graph kernels do not scale to large graphs with hundreds of nodes and thousands of edges. In this article we propose to compare graphs by counting {it graphlets}, ie subgraphs with $k$ nodes where $k in { 3, 4, 5 }$. Exhaustive enumeration of all graphlets being prohibitively expensive, we introduce two theoretically grounded speedup schemes, one based on sampling and the second one specifically designed for bounded degree graphs. In our experimental evaluation, our novel kernels allow us to efficiently compare large graphs that cannot be tackled by existing graph kernels.}, pages = {488-495}, editors = {Van Dyk, D. , M. Welling}, publisher = {MIT Press}, organization = {Max-Planck-Gesellschaft}, school = {Biologische Kybernetik}, address = {Cambridge, MA, USA}, month = apr, year = {2009}, slug = {5664}, author = {Shervashidze, N. and Vishwanathan, SVN. and Petri, TH. and Mehlhorn, K. and Borgwardt, KM.}, month_numeric = {4} }