KernelNE is a method for learning node embeddings by incorporating kernel functions with models relying on weighted matrix factorization, encoding random walk-based structural information of the graph.

Code

An implementation of the project in C++ can be reached at the Github repository.