Weisfeiler-Lehman Graph Kernels
Code
This Downloadzip (ZIP, 580 KB)vertical_align_bottom archive contains Matlab scripts to compute various graph kernels for graphs with unlabeled or categorically labeled nodes, such as the random walk, shortest path, graphlet, several instances of Weisfeiler-Lehman or other subtree kernels. For a more detailed list of available kernels please consult the DownloadREADME (3 KB)vertical_align_bottom in the archive.
Datasets
Download the graph data sets MUTAG, NCI1, NCI109, ENZYMES, and D&D (all in Matlab format) as a Downloadzip (ZIP, 10.1 MB)vertical_align_bottom or a Downloadtar.gz (GZ, 10.1 MB)vertical_align_bottom archive. The archive also contains a README describing the data sets and explaining their usage. Download the Matlab scripts for the WL subtree, edge and shortest path kernels as a Downloadzip (ZIP, 8 KB)vertical_align_bottom or a Downloadtar.gz (GZ, 4 KB)vertical_align_bottom archive. This archive also contains a README explaining the usage of the scripts.
Publication
Weisfeiler-Lehman Graph Kernels
Nino Shervashidze, Pascal Schweitzer, Erik Jan van Leeuwen, Kurt Mehlhorn and Karsten M. Borgwardt
Journal of Machine Learning Research (JMLR) 2011, 12: 2539−2561
external pageOnlinecall_made | ETH Research Collection | Project page including code
Contact Karsten Borgwardt for questions.