Get started with Graph Data Science
Learn Neo4j Graph Data Science fundamentals hands-on: project graphs, run and configure graph algorithms, and interpret results on a real movie dataset.
In this 3-hour course, you will learn
The Neo4j Go to glossary for graph data science (opens in a new tab)Analysing data through the structure of its connections. Also the name of the Neo4j library that implements it. (GDS) library brings graph algorithms to the data in your database. This course teaches you the fundamentals of graph data science: how to project a graph for analysis, how to choose and configure graph algorithms, and how to interpret the results they produce.
Graph algorithms answer questions that ordinary queries cannot express well. A Full definition for Cypher (opens in a new tab)Neo4j's implementation of GQL, the ISO standard query language for graph databases. It is declarative: you describe the pattern to find, and the database decides how to find it. query finds the records that match a Full definition for pattern (opens in a new tab)A graph structure written in Cypher, such as a node joined to another node by a relationship.; a graph algorithm considers the structure of the whole graph at once to measure influence, detect communities, or score similarity. In around three hours you will go from your first Full definition for projection (opens in a new tab)An in-memory copy of part of your database that graph algorithms run against. You choose which nodes and relationships it holds. to running any algorithm in the GDS catalog against projections you design yourself.
Graph Projections
Build in-memory graph projections from Neo4j data, selecting nodes and relationships to expose your dataset to the Graph Data Science library.
Algorithm Execution
Run graph algorithms such as PageRank and community detection on projected graphs and retrieve results using Cypher queries.
Algorithm Configuration
Configure algorithm parameters — including concurrency, iteration limits, and write modes — to tune performance and control how results are stored.
Relationship Aggregation
Apply relationship aggregation strategies to combine parallel relationships, controlling how weights and properties are merged during projection.
Projection Modeling
Model graph projections with native and Cypher projections, choosing the right approach to match your algorithm's structural and property requirements.
4 modules, 3 hours.
- Understand the GDS workflow5 min
- Project a graph5 min
- Projecting monopartite graphs5 min
- Practice monopartite projections5 min
- Challenge: Monopartite projection5 min
- Projecting bipartite and multipartite graphs5 min
- Practice bipartite projections5 min
- Challenge: Multipartite projections5 min
- List and drop graphs5 min
- Challenge: Clean your catalog5 min
- Module recap and what's next5 min
- Algorithms overview5 min
- Understand the five execution modes5 min
- Use GDS utility functions5 min
- Understand GDS documentation5 min
- Configure your algorithms5 min
- Running algorithms5 min
- Configuring projections for undirected relationships5 min
- Configuring projections with unweighted relationships5 min
- Challenge: Weighted graph projection and analysis5 min
- Module recap and what's next5 min
Who this course is for
Analysts and data scientists who want to move beyond querying and start analyzing the structure of their graphs, and the developers and data engineers who support those workloads. You need no prior experience with the Go to glossary for graph data science (opens in a new tab)Analysing data through the structure of its connections. Also the name of the Neo4j library that implements it. library — every concept, from Full definition for projection (opens in a new tab)An in-memory copy of part of your database that graph algorithms run against. You choose which nodes and relationships it holds. to execution modes, is introduced from the beginning. You should be comfortable with the basics of Neo4j and Full definition for Cypher (opens in a new tab)Neo4j's implementation of GQL, the ISO standard query language for graph databases. It is declarative: you describe the pattern to find, and the database decides how to find it. first; the prerequisite courses are listed beside this overview.
What you'll do
This is a hands-on course. When you enrol, GraphAcademy automatically creates a
movie recommendationssandbox within Neo4j Sandbox, a cloud-hosted Neo4j instance pre-loaded with the Go to glossary for graph data science (opens in a new tab)Analysing data through the structure of its connections. Also the name of the Neo4j library that implements it. library and a movie dataset of films, people, genres, and user ratings.You will run every Full definition for projection (opens in a new tab)An in-memory copy of part of your database that graph algorithms run against. You choose which nodes and relationships it holds. and algorithm yourself against that sandbox, through guided practice lessons and challenges that check your work against the database. By the end you will have projected Full definition for monopartite graph (opens in a new tab)A graph in which every node is the same kind of thing, so relationships connect like to like., Full definition for bipartite graph (opens in a new tab)A graph with two kinds of node, where every relationship joins one kind to the other and never two of the same kind., and Full definition for multipartite graph (opens in a new tab)A graph with three or more kinds of node, where relationships only ever join nodes of different kinds., run algorithms in all five execution modes, aggregated parallel Full definition for relationship (opens in a new tab)A named, directed connection between two nodes. Every relationship has a type, a start node and an end node. into weighted ones, and reshaped a graph so its structure matches the question you are asking.
Where to go next
Path Finding with GDS applies the library to shortest path problems on weighted and unweighted graphs, and Understand centrality algorithms teaches you which Full definition for centrality (opens in a new tab)How important a node is within a graph. Each centrality algorithm defines importance differently. measure answers which question. If you want to run graph algorithms without managing Go to glossary for graph data science (opens in a new tab)Analysing data through the structure of its connections. Also the name of the Neo4j library that implements it. yourself, Aura Graph Analytics fundamentals introduces Neo4j's session-based graph compute service, where the skills you learn here transfer directly.
This course is also the starting point for the Neo4j Graph Data Science certification, which validates your graph data science skills with a credential you can share.