Arango Documentation
Arango provides the trusted data foundation for the next wave of AI grounded in business context
User manuals by product
Native multi-model database system that unifies graph, document, key-value, vector, and full-text search with one query language.
Adds platform services for scalability, reliability, governance, and a graph exploration tool.
Supercharge your Data Platform with GraphRAG, GraphML, and queries generated from natural language for AI-powered insights.
Arango’s fully-managed cloud offering for a faster time to value, formerly known as ArangoGraph Insights Platform.
From graph to AI
Data Persistence
ArangoDB is a scalable database system that you can use to store JSON documents, which allows a flexible data structure for each record. ArangoDB natively supports graphs, letting you connect documents with edges to express relationships between records and build complex information networks.
Data Retrieval
You can query your data in various ways using the core database system. The native support for multiple data models lets you access information in different ways with a single query language called AQL. It has built-in support for aggregation, vector and full-text search, geo-spatial queries, and more.
Data Exploration
You can visually explore and interact with your ArangoDB graphs through an intuitive web interface called the Graph Visualizer. It is part of the Arango Data Platform that builds on ArangoDB, extending it to a Kubernetes-native environment that unifies data management, monitoring, and automation.
Graph Queries
Utilizing connected data starts with running simple graph queries. Using ArangoDB and its query language, you can determine the shortest paths between nodes as well as execute graph traversals. A traversal starts at a given node of a graph and follows the directly connected edges. The edges indicate what the next connected nodes are, and this discovery of neighbors can repeat.
Graph queries can answer questions like
Who can introduce me to person X?
Graph Analytics
The next level of utilizing connected data in terms of complexity is to use graph analytics or graph algorithms to aggregate information about a graph. Unlike with graph queries, this involves the entire graph at once.
Graph analytics can answer questions like
Who are the most connected persons?
Arango offers a Graph Analytics solution as part of the Arango AI Data Platform to run algorithms such as connected components, label propagation, and PageRank on your data.
GraphML
For higher-level insights, you can use advanced graph-based data science. Applying machine learning on graphs lets you predict connections, get better product recommendations, and also classify nodes, edges, and graphs.
GraphML can answer questions like:
- Is there a connection between person X and person Y?
- Will a customer churn?
- Is this particular transaction anomalous?
Arango’s enterprise-ready, graph-powered machine learning capabilities are included in the AI Suite as part of the Arango AI Data Platform. See Arango GraphML.
GraphRAG
Generative AI often struggle with hallucinations because the connectedness of data is not properly or cleanly represented. GraphRAG is a technique that turbocharges GenAI applications using the power of graph relationships and vector embeddings.
Arango’s GraphRAG included in the AI Suite is a turn-key solution to transform your organization’s data into a knowledge graph and let everyone utilize the knowledge by asking questions in natural language.
It automatically creates a knowledge graph from raw text by identifying and extracting entities and relationships within the data, groups and summarizes semantically similar entities, and stores everything in ArangoDB. When you ask a question, the large language model (LLM) is supplied with additional context from the knowledge graph, using lexical and semantic search. This enables accurate, context-aware intelligence grounded in enterprise data.
