Weisfeiler-Lehman Graph Kernels

Nino Shervashidze, Pascal Schweitzer, Erik Jan van Leeuwen, Kurt Mehlhorn, Karsten Borgwardt

Enlarged view: Logo of JMLR

Weisfeiler-Lehman Graph Kernels

Code

This Download zip (ZIP, 580 KB) 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 Download README (3 KB) in the archive.

Datasets

Download the graph data sets MUTAG, NCI1, NCI109, ENZYMES, and D&D (all in Matlab format) as a Download zip (ZIP, 10.1 MB) or a Download tar.gz (GZ, 10.1 MB) 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 Download zip (ZIP, 8 KB) or a Download tar.gz (GZ, 4 KB) 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 page Online | ETH Research Collection | Project page including code

    

Contact  Karsten Borgwardt for questions.

JavaScript has been disabled in your browser