Global Vertex Similarity for Large-Scale Knowledge Graphs

Here we provide resources for the paper Global Vertex Similarity for Large-Scale Knowledge Graphs. The work aims to find similar entities (represented by vertices/nodes) in a knowledge graph based on the common neighbours that they share. We provide the code we have used, as well as the dataset for evaluation from Last.fm and BestSimilar.

Data

• We provide the dataset for BestSimilar and Last.fm used for evaluation in the following Github repository

Code

• Code for SimRank and for other graph algorithms in Spark is available in the the following Github repository