Skip to content
This repository has been archived by the owner on Jun 18, 2021. It is now read-only.

Relations relevance

Jump to bottom
Jesse Himmelstein edited this page Jun 13, 2013 · 3 revisions

Computation of a measure dealing with relevance of links between nodes

Idea : 2 guys are close if they have many common friends. In a more formalized way, and with the idea of directionality : a node N1 has a strong relation to N2 if there are many nodes Ci for which :

  • A links to Ci
  • Ci links to B

The number of links from A must be taken into account. Then the relevance of the link A→B is (number of links from A divided by the number of "Ci" nodes. From a probabilistic point of view, the relevance of the link A→B is proportional to the probability of accessing B from A in 2 steps during a random walk in Knownodes.

knownodes relevance computation

Queries

Take a look at the import/cypherQueries.cypher file to see how we were able to calculate these relevance values using Neo4j.

Clone this wiki locally