GraphAcademy

Graph Data Science

Louvain algorithm

A community detection algorithm that repeatedly merges nodes into groups for as long as merging raises modularity.

Also written: Louvain

Example

Two triangles, joined by a single relationship between Cleo and Dan.

cypher
CALL gds.louvain.stream('people')
YIELD nodeId, communityId
RETURN gds.util.asNode(nodeId).name AS person, communityId

Louvain returns two communities, one for each triangle.

Lessons that use this term

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

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

Graph Data Science in Practice54 mentions

  • Louvain Community DetectionCommunity Detection for Fraudthe algorithm that we'll use to find communities of connected suspects: Louvain24 mentions
  • Reducing Search Space with LouvainCommunity Detection for FraudYou've learned how Louvain finds communities by optimizing12 mentions
  • Degree Centrality & WCCCommunity Detection for FraudLouvain helped us explore the fraud7 mentions
  • Applying AlgorithmsGDS FoundationsLouvain groups nodes by maximising "modularity"--connection density within groups3 mentions
  • Algorithm CategoriesGDS Foundations| Louvain, Leiden, WCC | | Similarity | Which nodes are1 mention
  • Workshop Complete!Community Detection for FraudUsed Louvain to reduce a 204,000-user search space by 98% Applied WCC for deterministic, explainable community assignment -1 mention
  • Fraud DetectionCommunity Detection for FraudLesson 2: Learn how Louvain community detection works Lesson 3: Run Louvain to reduce your search space by 98% Lesson 4: Learn Degree2 mentions
  • Building Fraud CommunitiesCommunity Detection for Fraudwas valuable for exploration, but it has limitations for formal fraud3 mentions
  • Execution Modes and ConfigurationGDS Foundations1 mention

Get started with Graph Data Science53 mentions

  • Running algorithmsWorking with algorithmsthis lesson, you'll run degree centrality, PageRank and Louvain community detection on the actor collaboration network, experimenting with algorithm-specific18 mentions
  • Configuring projections for undirected relationshipsWorking with algorithmsHow PageRank behaves on undirected graphs How Leiden differs from Louvain When to use undirected10 mentions
  • Algorithms overviewWorking with algorithmsof the most common algorithms are Louvain and Leiden —but there are many9 mentions
  • Configuring projections with unweighted relationshipsWorking with algorithmsis similar to Louvain3 mentions
  • Module recap and what's nextWorking with algorithmsBetweenness) Community Detection - Find natural clusters and groups (Louvain, Leiden, Label Propagation) Pathfinding - Discover optimal routes through networks (Dijkstra) Similarity -2 mentions
  • Understand the five execution modesWorking with algorithmsrun Louvain in stats mode on the actor collaboration network —without any YIELD or5 mentions
  • Understand GDS documentationWorking with algorithmsrequired retaining the directionality of relationships, you could switch to Louvain, which can run on directed relationship2 mentions
  • Projection modeling for algorithmsEssential projection techniquesdetection (Louvain, Leiden) works best1 mention
  • Configure your algorithmsWorking with algorithms1 mention
  • Module recapEssential projection techniquesBetweenness) Community Detection - Find natural clusters and groups (Louvain, Leiden, Label Propagation) Pathfinding - Discover optimal routes through networks (Dijkstra) Similarity -1 mention
  • Use GDS utility functionsWorking with algorithmsLouvain community detection and extract both the actor name and the community they belong1 mention

Analyze Graph Data with Python11 mentions

  • Louvain Community DetectionGDS Python ClientYou may have already used Louvain in the Graph Data Science in Practice workshop to detect fraud rings by optimizing7 mentions
  • Workshop Complete!Aura Graph Analyticswith PageRank and Betweenness Centrality Detected research communities with Louvain Generated FastRP embeddings for machine learning1 mention
  • Introduction to the Python GDS ClientGDS Python Clientinto memory Running algorithms like PageRank, Betweenness Centrality, Louvain, and FastRP Processing results as DataFrames Cleaning up projections when we're1 mention
  • Scaled Properties and FastRP EmbeddingsGDS Python Clientwith Louvain from Lesson2 mentions

AI on Your Lakehouse: Context Comes in Shapes, Not Queries1 mention

  • Port the PatternPort the PatternDetection](/courses/gds-community-detection/) to go deeper on Leiden and Louvain1 mention

Related terms

community detection
A family of algorithms that group nodes by how they connect. Each algorithm defines a community differently.19 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.17 shared lessons
centrality
How important a node is within a graph. Each centrality algorithm defines importance differently.12 shared lessons
modularity
A score for how much more densely connected the nodes within a group are than they would be by chance.9 shared lessons
degree centrality
A score for each node equal to its number of outgoing relationships.12 shared lessons
undirected relationship
A relationship with no direction. Neo4j GDS represents them as parallel relationships in opposite directions.11 shared lessons
graph data science
Analysing data through the structure of its connections. Also the name of the Neo4j library that implements it.21 shared lessons
projection
An in-memory copy of part of your database that graph algorithms run against. You choose which nodes and relationships it holds.24 shared lessons

All glossary terms