====== Neo4j ======

Neo4j is the world's leading graph database management system, developed by Neo4j, Inc. It uses a native property graph model with nodes, relationships (edges), and properties to store and query highly connected data. Neo4j provides ACID-compliant transactions, native graph storage, and the Cypher declarative query language. As of early 2026, Neo4j has surpassed **$200 million in annual revenue** and serves as the most widely deployed graph database globally. ((Source: [[https://neo4j.com/press-releases/neo4j-revenue-milestone-2024|Neo4j Surpasses $200M in Revenue]]))

===== History =====

Neo4j's origins trace back to **2000**, when founders **Peter Neubauer** and **Johan Svensson** encountered limitations with relational databases while building a media management system in Sweden. They modeled data as interconnected networks rather than tables, which improved flexibility despite initial performance challenges. ((Source: [[https://docs.nebula-graph.io/3.5.0/1.introduction/0-1-graph-database/|NebulaGraph - History of Graph Databases]]))

Johan Svensson implemented a native graph storage engine, and the intellectual property was spun out into **Neo4j, Inc. in 2007**.

Key milestones:

  * **2000** -- Property graph model conceived for media management
  * **2007** -- Neo4j, Inc. founded as a standalone company
  * **2009** -- XPath-like query syntax developed (precursor to Cypher and Gremlin)
  * **2010** -- Term "Property Graph" coined by employee Marko Rodriguez
  * **2011** -- Neo4j 1.4 released publicly; Cypher v1 introduced (read-only)
  * **2012** -- Neo4j 1.8 added Cypher modifications
  * **2013** -- Neo4j 2.0 introduced labels and indexes
  * **2015** -- openCypher project launched for broader Cypher adoption
  * **2016** -- Neo4j 3.0 released
  * **2018** -- Enterprise Edition core closed-source from version 3.5
  * **2021** -- $325 million Series F at $2 billion+ valuation
  * **2025** -- Revenue surpassed $200 million; IPO preparations began ((Source: [[https://en.wikipedia.org/wiki/Neo4j|Wikipedia - Neo4j]]))

===== Cypher Query Language =====

Cypher is Neo4j's **declarative graph query language**, inspired by SQL's principle of separating "what" from "how." It uses an intuitive ASCII-art syntax for pattern matching, with arrows and brackets representing relationships:

  MATCH (p:Person)-[:KNOWS]->(f:Person)
  WHERE p.name = 'Alice\n  RETURN f.name

Cypher was first introduced in 2011 as a read-only language and has evolved through multiple versions to support full CRUD operations, aggregations, subqueries, and graph algorithms. ((Source: [[https://docs.nebula-graph.io/3.5.0/1.introduction/0-1-graph-database/|NebulaGraph - Graph Database Introduction]]))

In 2015, Neo4j launched the **openCypher** initiative, making Cypher available as an open standard adopted by other graph databases including Memgraph, Amazon Neptune, and SAP HANA Graph.

===== Use Cases =====

Neo4j excels in scenarios involving highly connected data:

  * **Knowledge Graphs** -- modeling interconnected entities for semantic search, AI/ML feature engineering, and enterprise data integration. Neo4j has become a foundational technology for GenAI applications, backing LLMs with structured knowledge graphs for retrieval-augmented generation (RAG).
  * **Fraud Detection** -- traversing relationship patterns in financial networks to identify anomalous connections, money laundering rings, and suspicious transaction chains. Major banks and financial institutions use Neo4j for real-time fraud analysis.
  * **Recommendation Engines** -- analyzing user-item-category connections for personalized suggestions, leveraging native graph traversal for collaborative and content-based filtering.
  * **Network and IT Operations** -- mapping infrastructure dependencies, root cause analysis, and impact assessment.
  * **Identity and Access Management** -- modeling complex permission hierarchies and organizational structures.
  * **Supply Chain Management** -- tracking goods, suppliers, and logistics relationships.

===== Neo4j Aura =====

Neo4j Aura is the fully managed **cloud database-as-a-service** offering, available on AWS, GCP, and Azure. ((Source: [[https://neo4j.com/press-releases/neo4j-announces-seriesf-funding/|Neo4j Series F Announcement]]))

  * **AuraDB Free** -- limited-size graph database for learning and prototyping
  * **AuraDB Professional** -- auto-scaling production databases with pay-per-use pricing
  * **AuraDB Enterprise** -- dedicated infrastructure with advanced security, SSO, VPC peering, and SLA guarantees
  * **Aura Graph Analytics** -- fully managed graph analytics as a service
  * **Aura Agent** -- a unified console for managing database instances

Aura eliminates the operational overhead of database management, backups, and scaling while providing the full Neo4j feature set.

===== Graph Data Science Library =====

The **Neo4j Graph Data Science (GDS)** library provides a comprehensive suite of graph algorithms for analytics and machine learning:

  * **Centrality algorithms** -- PageRank, Betweenness Centrality, Degree Centrality
  * **Community detection** -- Louvain, Label Propagation, Weakly Connected Components
  * **Pathfinding** -- Dijkstra, A*, Breadth-First Search, Minimum Spanning Tree
  * **Similarity** -- Node Similarity, K-Nearest Neighbors
  * **Link prediction** -- predicting future connections in graphs
  * **Graph embeddings** -- Node2Vec, FastRP for machine learning feature generation
  * **ML pipelines** -- end-to-end machine learning workflows within the graph

GDS is available both as a self-managed library and through **Aura Graph Analytics**. It also integrates with **Snowflake AI Data Cloud** and **Microsoft Fabric** through dedicated partnerships.

===== Funding and Company =====

Neo4j has raised over **$580 million** in total funding:

  * **Series E** -- $80 million
  * **Series F** -- $325 million (June 2021), led by Eurazeo and GV, the largest single investment round in database history at that time ((Source: [[https://neo4j.com/news/neo4j-announces-325-million-series-f-investment-the-largest-in-database-history/|Neo4j - $325M Series F]]))
  * Valued at over **$2 billion** following the Series F
  * IPO preparations reportedly began in late 2024, targeting a Nasdaq listing

The company is headquartered in San Mateo, California, with offices worldwide. Neo4j's Community Edition is available under a modified GPL license, while the Enterprise Edition requires a commercial license.

===== See Also =====

  * [[memgraph|Memgraph]]
  * [[graph_databases|Graph Databases]]
  * [[knowledge_graphs|Knowledge Graphs]]

===== References =====