1

While there has been a lot of talk in how to define the similarity between nodes in the embedding space, but I don't seem to come across any talking about defining the similarity between nodes in the original, non-embedded graph. Any suggestions as of how to define such?

Student
  • 419
  • 2
  • 9

1 Answers1

0

If the nodes you mention have inherent features, you could encode the features in a way that is similar. For example, nodes involving user data could have features like location, gender, etc.

And if you mean similarity in terms of connected/neighboring edges, you could probably define your own node feature vector such as one-hot encoding to the nodes it is connected to ( like an adjacency matrix).