Multilabel classification on heterogeneous graphs with gaussian embeddings

Ludovic dos Santos (1) , Benjamin Piwowarski (2) , Patrick Gallinari (1)

1 MLIA - Machine Learning and Information Access
2 BD - Bases de Données

We consider the problem of node classification in heterogeneous graphs where both nodes and relations may be of different types and a different set of categories is associated to each node type. When graph node classification has mainly been addressed for homogeneous graphs, heterogeneous classification is a recent problem which has been motivated by applications in fields such as social networks where the graphs are intrinsically heterogeneous. We propose a transductive approach to this problem based on learning graph embeddings and model the uncertainty associated to the node representations using Gaussian embeddings. A comparison with representative baselines is provided on three heterogeneous datasets.

Domaines: Intelligence artificielle [cs.AI]

Ludovic dos Santos, Benjamin Piwowarski, Patrick Gallinari. Multilabel classification on heterogeneous graphs with gaussian embeddings. European Conference on Machine Learning and Principles and Practice of Knowledge Discovery, Sep 2016, Riva del garda, Italy. ⟨hal-01352911⟩