GraphAcademy

Graph Data Science

node similarity

An algorithm that scores how alike two nodes are by comparing the neighbours they share.

Example

Ava and Ben both bought the Mug, and one other product each.

cypher
CALL gds.nodeSimilarity.stream('purchases')
YIELD node1, node2, similarity
RETURN gds.util.asNode(node1).name AS a,
       gds.util.asNode(node2).name AS b,
       similarity

They share one product out of the three between them, so the score is 0.33. Sharing all three would score 1, and sharing none would leave the pair out of the results.

Lessons that use this term

The lesson and course links below open in a new tab.

15 lessons use this term. Results are ordered by term density.

Get started with Graph Data Science43 mentions

  • Challenge: Weighted graph projection and analysisWorking with algorithmsused Node Similarity in the previous lesson to find similar movies and write relationships back to the10 mentions
  • Practice bipartite projectionsGDS basic conceptsthe Movies dataset to create various labelled bipartite projections and run Node Similarity on them—an algorithm specifically designed to work with bipartite14 mentions
  • Projecting bipartite and multipartite graphsGDS basic conceptsSimilarity compares nodes based on their shared neighbours—perfect for bipartite graphs where one set of nodes connects to6 mentions
  • Challenge: Multipartite projectionsGDS basic conceptsthat your users-genres projection is in place, you can run node similarity on it to create SIMILAR relationships between users who share genre1 mention
  • Projection modeling for algorithmsEssential projection techniques(Node Similarity) works best2 mentions
  • Algorithms overviewWorking with algorithmsGDS algorithms: degree centrality and PageRank on monopartite graphs, and node similarity on bipartite6 mentions
  • Module recap and what's nextGDS basic conceptsbasic algorithms (degree centrality, PageRank, node similarity1 mention
  • Module recap and what's nextWorking with algorithmsnetworks (Dijkstra) Similarity - Find structurally similar nodes (Node Similarity, FastRP) Embeddings - Create vector representations of nodes (FastRP,1 mention
  • Module recapEssential projection techniquesnetworks (Dijkstra) Similarity - Find structurally similar nodes (Node Similarity, FastRP) Embeddings - Create vector representations of nodes (FastRP,1 mention
  • Projecting monopartite graphsGDS basic conceptsYou're using algorithms designed for bipartite graphs (Node Similarity, some recommendation algorithms) The relationship between different types is what you're analysing You need to1 mention

Graph Data Science in Practice18 mentions

Analyze Graph Data with Python1 mention

Related terms

bipartite graph
A graph with two kinds of node, where every relationship joins one kind to the other and never two of the same kind.13 shared lessons
PageRank
A centrality algorithm that scores a node by the number of nodes pointing at it and by how important those nodes are.13 shared lessons
monopartite graph
A graph in which every node is the same kind of thing, so relationships connect like to like.9 shared lessons
centrality
How important a node is within a graph. Each centrality algorithm defines importance differently.9 shared lessons
FastRP
A node embedding algorithm, short for Fast Random Projection. It builds each node's vector by combining random vectors drawn from the nodes around it.6 shared lessons
multipartite graph
A graph with three or more kinds of node, where relationships only ever join nodes of different kinds.5 shared lessons
Dijkstra's algorithm
An algorithm that finds the cheapest route between two nodes.6 shared lessons
pathfinding
A family of algorithms that find routes through a graph. What counts as the best route differs by algorithm.6 shared lessons

All glossary terms