Enterprise Analytics & Graph-Native ML at Scale

Neo4j Graph Data Science

Enterprise graph analytics and graph-native machine learning at scale: adding contextual insights to your data science efforts and accelerating your path to production value.

What’s important? What’s unusual? What’s next?

Turn Connections Into Answers for Your Critical Business Questions

Predictive signals get lost in big data noise.

Connected data is powerful, but traditional approaches to data make it impossible to understand and use those connections. The bigger your data set, the more difficult it is to know where to even get started.

Neo4j Graph Data Science (GDS) uses connections within your big data to answer questions critical to your business. With scalable and performant GDS, unlock hidden insights in your data to make better predictions and accelerate innovation.

Graph Data Science
 Features and Benefits

Neo4j GDS delivers the most advanced, integrated framework available today, enabling enterprises to easily implement the latest connected data techniques to extract insights, train machine learning (ML) models, and deploy projects in production.

Best-In-Class Data Science

  • Access 65+ robust, scalable algorithms and supervised ML

  • Empower your team with state-of-the-science techniques

  • Make better predictions by leveraging data relationships

The Industry’s Only In-Graph Machine Learning

  • Unified model training and deployment in single environment

  • Automated MLOps so you stay focused on insights not infrastructure

  • Proven in production by the world’s biggest and best brands

Fastest Path to Bring Graph Analytics to Production

  • Dedicated data science workspace to quickly develop sophisticated AI

  • Powered by the world’s most popular and proven graph database

  • Complete workflow for data ingest, analysis, and management

Get Started

Pick a Solution that Fits Your Needs


No need to download software. Test drive Neo4j GDS on sandbox with preloaded data and a guide.


Get Neo4j desktop to try all the algorithms and ML methods with GDS Community Edition.


Sign up for the early access program

Opening the Door to New Data Possibilities

Banking and Financial Services

Find fraudsters by analyzing behavior between accounts and users, and predict fraudulent activity with greater accuracy.

Media and Publishing

Improve content recommendations to users with the predictive power of graphs, increasing conversion and engagement.


Map journeys to understand your customers, helping you reduce churn and bolster marketing strategy

Life Sciences

Leverage connections within biological and chemical data or between patients, diseases, and treatments for leaner experimentation.


Resolve entities for targeting, segmentation, and churn prediction or fine-tune recommendations via ontologies and customer behavior.


Detect threats, laundering schemes, and tax fraud; boost supply chain by predicting faulty components and diagnosing failures.

Data Science Algorithms for
Analytics Success

View Algorithms

  • Community Detection
    Communities are clusters within your graph.


    Community Detection algorithms cluster your graph based on relationships to find communities where members have more significant interactions. Detecting communities helps predict similar behavior, find duplicate entities, or simply prepare data for other analyses.

    Case Study

    Meredith Corp uses Community Detection to group anonymous web traffic into unique identities based on behavior patterns and identifiers. These disambiguated profiles have resulted in better personalization for increased traffic and more accurate profiles to target for increased ad revenue.

  • Centrality/Importance
    Centrality metrics like PageRank help you identify what’s important.


    Centrality algorithms reveal which nodes are important based on graph topology. They identify influential nodes based on their position in the network and are used to infer group dynamics such as credibility, rippling vulnerability, and bridges between groups.

    Case Study:

    Boston Scientific uses Centrality in their supply chain analytics to find faulty components and prevent device failures. As a result, they are able to find root causes of defects more quickly and manufacturing processes are more efficient.

  • Similarity
    Similarity identifies repeating patterns in your graph.


    Similarity algorithms employ set comparisons to score how alike individual nodes are based on their neighbors or properties. This approach is used in applications such as personalized recommendations and developing categorical hierarchies.

    Case Study:

    NASA finds similarities between employees’ skill sets and role responsibilities to not only help teams build critical skills and competencies to advance their careers, but also to identify hidden skills within the company. As a result, they better understand what skills the broader organization already possesses and what skills are lacking in order to inform future HR initiatives.

  • Pathfinding and Search
    Pathfinding algorithms find the best routes within your connected data.


    Pathfinding algorithms are foundational to graph analytics and find the most efficient or shortest paths to traverse between nodes. They can be used to understand complex dependencies and evaluate routes for uses such as physical logistics and least-cost call or IP routing.

    Case Study:

    OrbitMI's SaaS platform is powered by GDS pathfinding algorithms, which calculate maritime routes based on factors – such as weather, canal details, and other constraints – to find the most optimal route, which is not always the shortest one. These optimized routes save their clients time, money, and fuel.

  • Node Embeddings
    Node Embeddings are powerful predictive features.


    Node Embedding algorithms transform the topology and features of your graph into fixed-length vectors that represent each node. They capture the complexity and structure of a graph and transform it for use in various ML tasks.

    Case Study:

    gov.uk maps user journeys on their website into their knowledge graph and runs a Node Embedding algorithm to distill these journeys into vectors on each node. These embeddings are then used to make real-time content recommendations.

  • Graph-Native Machine Learning

    Learn from your graph to predict missing data or future connections.


    Graph native machine learning techniques like Link Prediction and Node Classification can fill in the blanks in your data and predict changes in the structure of your graph. They enable use cases such as fraud detection, drug discovery, entity resolution, and more.

    Case Study:

    A team at Georgia Tech used link prediction to discover new covid treatments by analyzing biomedical literature on previous coronaviruses, such as SARS, via text mining. Using link prediction, they were able to identify drug candidates that had potential to treat COVID-19 and its side effects.

The Enterprise Difference

Try out all the algorithms, embeddings, and ML techniques in Community Edition, and graduate to Enterprise when you’re ready for production scale.

Community Edition

Enterprise Edition

Algorithms, Embeddings & ML Models Yes Yes
Pregel API Yes Yes
Parallelization ≤ 4 cpus Unlimited
Fine Grained Security No Yes
Low Memory Analytics Graph Format No Yes
Model Catalog 3 Models Unlimited
Model Persistence & Publication No Yes

Join the Leading Organizations Using Neo4j GDS

Test drive Neo4j GDS on Sandbox. No download or data required.

Contact Us