Graph Kernels: Crossing Information from Different Patterns using Graph Edit Distance
Résumé
Graph kernels allow to define metrics on graph space and constitute thus an efficient tool to combine advantages of structural and statistical pattern recognition fields. Within the chemoinformatics frame- work, kernels are usually defined by comparing number of occurences of patterns extracted from two different graphs. Such a graph kernel con- struction scheme neglects the fact that similar but not identical patterns may lead to close properties. We propose in this paper to overcome this drawback by defining our kernel as a weighted sum of comparisons be- tween all couples of patterns. In addition, we propose an efficient com- putation of the optimal edit distance on a limited set of finite trees. This extension has been tested on two chemoinformatics problems.
Origine | Fichiers produits par l'(les) auteur(s) |
---|
Loading...