Kernel Node Embeddings
This work addresses the need for improved node embeddings in graph analysis, but it is incremental as it combines existing approaches without introducing a fundamentally new paradigm.
The paper tackles the problem of learning node representations for network analysis by proposing a weighted matrix factorization model that incorporates random walk information, and it shows that this model outperforms baseline algorithms in link prediction and node classification tasks on real-world networks.
Learning representations of nodes in a low dimensional space is a crucial task with many interesting applications in network analysis, including link prediction and node classification. Two popular approaches for this problem include matrix factorization and random walk-based models. In this paper, we aim to bring together the best of both worlds, towards learning latent node representations. In particular, we propose a weighted matrix factorization model which encodes random walk-based information about the nodes of the graph. The main benefit of this formulation is that it allows to utilize kernel functions on the computation of the embeddings. We perform an empirical evaluation on real-world networks, showing that the proposed model outperforms baseline node embedding algorithms in two downstream machine learning tasks.