[P] Knowledge Graph Node Similarity calculation

Hey everyone, I've been trying to calculate the similarity score between two different nodes in a Knowledge Graph that I've created. Basically I want the most relevant categories (in the KG) for a set of keywords (also in the KG). I was thinking of just doing a cosine similarity on top of the individual node embeddings for each of the keywords and returning all the related categories above an arbitrary threshold (for e.g for keyword X return all of the …

graph hey keywords knowledge knowledge graph machinelearning node set thinking