diff --git a/site/content/_index.md b/site/content/_index.md new file mode 100644 index 0000000000..b95a8b3a4e --- /dev/null +++ b/site/content/_index.md @@ -0,0 +1,116 @@ +--- +title: Arango Documentation +menuTitle: Home +weight: 1 +description: >- + Arango provides the trusted data foundation for the next wave of AI grounded + in business context +#aliases: +# - data-science/overview +--- +## User manuals by product + +{{< cards >}} + +{{% card title="ArangoDB" link="arangodb/" icon="avo-core.svg" %}} +Native multi-model database system that unifies graph, document, +key-value, vector, and full-text search with one query language. +{{% /card %}} + +{{% card title="Arango Data Platform" link="data-platform/" icon="avo-middle.svg" %}} +Adds platform services for scalability, reliability, governance, and a graph exploration tool. +{{% /card %}} + +{{% card title="AI Suite" link="ai-suite/" icon="avo-full.svg" %}} +Supercharge your Data Platform with GraphRAG, GraphML, +and queries generated from natural language for AI-powered insights. +{{% /card %}} + +{{% card title="Arango Managed Platform (AMP)" link="amp/" %}} +Arango's fully-managed cloud offering for a faster time to value, +formerly known as ArangoGraph Insights Platform. +{{% /card %}} + +{{< /cards >}} + +## From graph to AI + +### Data Persistence + +ArangoDB is a scalable database system that you can use to store +[JSON documents](arangodb/3.12/concepts/data-structure/documents/_index.md), +which allows a flexible data structure for each record. ArangoDB natively supports +[graphs](arangodb/3.12/graphs/_index.md), 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](arangodb/3.12/aql/_index.md). +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](data-platform/graph-visualizer.md). +It is part of the [Arango Data Platform](data-platform/_index.md) 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](arangodb/3.12/aql/graphs/_index.md). +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](ai-suite/graph-analytics.md) solution as part +of the [Arango AI Data Platform](data-platform/features.md) 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](ai-suite/_index.md) as part of the +Arango AI Data Platform. See [Arango GraphML](ai-suite/graphml/_index.md). + +### 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](ai-suite/graphrag/_index.md) included in the +[AI Suite](ai-suite/_index.md) 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. diff --git a/site/content/ai-suite/_index.md b/site/content/ai-suite/_index.md index c149b1fea4..4f23bcaa4c 100644 --- a/site/content/ai-suite/_index.md +++ b/site/content/ai-suite/_index.md @@ -37,3 +37,9 @@ Alongside these components, you also get the following additional features: AI Suite and build your own integrations. See the [Protocol Documentation](https://arangoml.github.io/platform-dss-api/GenAI-Service/proto/index.html) for more details. + +## Sample datasets + +If you want to try out ArangoDB's data science features, you may use the +[`arango-datasets` Python package](../arangodb/3.12/components/tools/arango-datasets.md) +to load sample datasets into a deployment. \ No newline at end of file diff --git a/site/content/ai-suite/graph-to-ai.md b/site/content/ai-suite/graph-to-ai.md deleted file mode 100644 index f5a4730563..0000000000 --- a/site/content/ai-suite/graph-to-ai.md +++ /dev/null @@ -1,132 +0,0 @@ ---- -title: From Graph to AI -menuTitle: From Graph to AI -weight: 25 -description: >- - ArangoDB's set of tools and technologies enables analytics, machine learning, - and AI applications powered by graph data -aliases: - - data-science/overview ---- - -{{< tip >}} -The Arango Data Platform & AI Suite are available as a pre-release. To get -exclusive early access, [get in touch](https://arango.ai/contact-us/) with -the Arango team. -{{< /tip >}} - -ArangoDB provides a wide range of functionality that can be utilized for -data science applications. The core database system includes multi-model storage -of information with scalable graph and information retrieval capabilities that -you can directly use for your research and product development. - -ArangoDB also offers dedicated AI Suite, using the database core -as the foundation for higher-level features. Whether you want to turbocharge -generative AI applications with a GraphRAG solution or apply analytics and -machine learning to graph data at scale, ArangoDB covers these needs. - - - -## From graph to AI - -This section classifies the complexity of the queries you can answer with -ArangoDB and gives you an overview of the respective feature. - -It starts with running a simple query that shows what is the path that goes from -one node to another, continues with more complex tasks like graph classification, -link prediction, and node classification, and ends with generative AI solutions -powered by graph relationships and vector embeddings. - -### Graph Queries - -When you run an AQL query on a graph, a traversal query can go from a node to -multiple edges, and then the edges indicate what the next connected nodes are. -Graph queries can also determine the shortest paths between nodes. - -Graph queries can answer questions like _**Who can introduce me to person X**_? - -![Graph Query](../images/graph-query.png) - -See [Graphs in AQL](../arangodb/3.12/aql/graphs/_index.md) for the supported graph queries. - -### Graph Analytics - -Graph analytics or graph algorithms is what you run on a graph if you want to -know aggregate information about your graph, while analyzing the entire graph. - -Graph analytics can answer questions like _**Who are the most connected persons**_? - -![Graph Analytics](../images/graph-analytics.png) - -ArangoDB offers _Graph Analytics Engines_ to run algorithms such as -connected components, label propagation, and PageRank on your data. This feature -is available for the Arango Managed Platform (AMP). See -[Graph Analytics](graph-analytics.md) for details. - -### GraphML - -When applying machine learning on a graph, you can 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?**_ - -![Graph ML](../images/graph-ml.png) - -For ArangoDB's enterprise-ready, graph-powered machine learning offering, -see [Arango GraphML](graphml/_index.md). - -### GraphRAG - -GraphRAG is ArangoDB's 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. - -The overall process of GraphRAG involves the following: -- **Creating a Knowledge Graph** from raw text data. -- **Identifying and extract entities and relationships** within the data. -- **Storing the structured information** in ArangoDB. -- **Clustering each closely connected set of entities into semantic contexts** - via topology-based algorithms and summarization. -- **Using such semantically augmented structured representation** as the - foundation for efficient and accurate information retrieval via lexical and - semantic search. -- **Integrating retrieval methods with LLMs (privately or publicly hosted)** - to augment responses using both structured and unstructured data, providing - accurate responses with the desired format and degree of detail for each query. - -To learn more, see the [GraphRAG](graphrag/_index.md) documentation. - -## Knowledge Graphs - -A knowledge graph can be thought of as a dynamic and interconnected network of -real-world entities and the intricate relationships that exist between them. - -Key aspects of knowledge graphs: -- **Domain-specific knowledge**: You can tailor knowledge graphs to specific - domains and industries. -- **Structured information**: Makes it easy to query, analyze, and extract - meaningful insights from your data. -- **Accessibility**: You can build a Semantic Web knowledge graph or using - custom data. - -LLMs can help distill knowledge graphs from natural language by performing -the following tasks: -- Entity discovery -- Relation extraction -- Coreference resolution -- End-to-end knowledge graph construction -- (Text) Embeddings - -## Sample datasets - -If you want to try out ArangoDB's data science features, you may use the -[`arango-datasets` Python package](../arangodb/3.12/components/tools/arango-datasets.md) -to load sample datasets into a deployment. diff --git a/site/content/ai-suite/graphrag/_index.md b/site/content/ai-suite/graphrag/_index.md index b284f820e1..55b66214f3 100644 --- a/site/content/ai-suite/graphrag/_index.md +++ b/site/content/ai-suite/graphrag/_index.md @@ -3,7 +3,7 @@ title: GraphRAG menuTitle: GraphRAG weight: 5 description: >- - ArangoDB's GraphRAG solution combines graph-based retrieval-augmented generation + Arango's GraphRAG solution combines graph-based retrieval-augmented generation with Large Language Models (LLMs) for turbocharged AI solutions aliases: llm-knowledge-graphs @@ -14,9 +14,30 @@ exclusive early access, [get in touch](https://arango.ai/contact-us/) with the Arango team. {{< /tip >}} +## What are knowledge graphs? + +A knowledge graph can be thought of as a dynamic and interconnected network of +real-world entities and the intricate relationships that exist between them. + +Key aspects of knowledge graphs: +- **Domain-specific knowledge**: You can tailor knowledge graphs to specific + domains and industries. +- **Structured information**: Makes it easy to query, analyze, and extract + meaningful insights from your data. +- **Accessibility**: You can build a Semantic Web knowledge graph or using + custom data. + +LLMs can help distill knowledge graphs from natural language by performing +the following tasks: +- Entity discovery +- Relation extraction +- Coreference resolution +- End-to-end knowledge graph construction +- (Text) Embeddings + ## Transform unstructured documents into intelligent knowledge graphs -ArangoDB's GraphRAG solution enables organizations to extract meaningful insights +Arango's GraphRAG solution enables organizations to extract meaningful insights from their document collections by creating knowledge graphs that capture not just individual facts, but the intricate relationships between concepts across documents. This approach goes beyond traditional RAG systems by understanding document @@ -30,23 +51,22 @@ conceptual understanding. ## Key benefits for enterprise applications -- **Cross-document relationship intelligence**: -Unlike traditional RAG systems that treat documents in isolation, ArangoDB's GraphRAG -pipeline detects and leverages references between documents and chunks. This enables -more accurate responses by understanding how concepts relate across your entire knowledge base. - -- **Multi-level understanding architecture**: -The system provides both detailed technical responses and high-level strategic insights -from the same knowledge base, adapting response depth based on query complexity and user intent. +- **Cross-document relationship intelligence**\ + Unlike traditional RAG systems that treat documents in isolation, Arango's GraphRAG + pipeline detects and leverages references between documents and chunks. This enables + more accurate responses by understanding how concepts relate across your entire knowledge base. -- **Reference-aware knowledge graph**: -GraphRAG automatically detects and maps relationships between document chunks while -maintaining context of how information connects across different sources. +- **Multi-level understanding architecture**\ + The system provides both detailed technical responses and high-level strategic insights + from the same knowledge base, adapting response depth based on query complexity and user intent. -- **Dynamic knowledge evolution**: -The system learns and improves understanding as more documents are added, with -relationships and connections becoming more sophisticated over time. +- **Reference-aware knowledge graph**\ + GraphRAG automatically detects and maps relationships between document chunks while + maintaining context of how information connects across different sources. +- **Dynamic knowledge evolution**\ + The system learns and improves understanding as more documents are added, with + relationships and connections becoming more sophisticated over time. ## What's next diff --git a/site/content/arangodb/_index.md b/site/content/arangodb/_index.md index 56103ad6a4..c77041e771 100644 --- a/site/content/arangodb/_index.md +++ b/site/content/arangodb/_index.md @@ -9,15 +9,13 @@ aliases: - introduction - introduction/about-arangodb --- -![ArangoDB Overview Diagram](../images/arangodb-overview-diagram.png) - ArangoDB combines the analytical power of native graphs with an integrated -search engine, JSON support, and a variety of data access patterns via a single, -composable query language. +search engine, JSON support, vector indexes, and a variety of data access +patterns via a single, composable query language. ArangoDB is available in a community and a commercial [edition](3.12/features/_index.md). -You can use it for on-premises deployments, as well as a fully managed -cloud service, the [Arango Managed Platform (AMP)](../amp/_index.md). +You can use it for on-premises deployments, self-managed cloud deployments, +as well as a fully managed cloud service, the [Arango Managed Platform (AMP)](../amp/_index.md). ## What are Graphs? diff --git a/site/content/images/avo-core.svg b/site/content/images/avo-core.svg new file mode 100644 index 0000000000..5d7bc8c7ab --- /dev/null +++ b/site/content/images/avo-core.svg @@ -0,0 +1,6 @@ + + + + + + diff --git a/site/content/images/avo-full.svg b/site/content/images/avo-full.svg new file mode 100644 index 0000000000..0e11a25947 --- /dev/null +++ b/site/content/images/avo-full.svg @@ -0,0 +1,6 @@ + + + + + + diff --git a/site/content/images/avo-middle.svg b/site/content/images/avo-middle.svg new file mode 100644 index 0000000000..00c9d91cbd --- /dev/null +++ b/site/content/images/avo-middle.svg @@ -0,0 +1,6 @@ + + + + + + diff --git a/site/themes/arangodb-docs-theme/static/css/theme.css b/site/themes/arangodb-docs-theme/static/css/theme.css index 1dd47fc6ef..8eeed2df74 100644 --- a/site/themes/arangodb-docs-theme/static/css/theme.css +++ b/site/themes/arangodb-docs-theme/static/css/theme.css @@ -478,24 +478,16 @@ a.section-link { } .card-head { - position: relative; z-index: 2; - display: flex; margin-bottom: 10px; - flex-flow: column; min-height: 40px; - justify-content: center; - align-items: flex-start; } .card-icon { - width: 40px; - height: 40px; - -o-object-fit: contain; + width: 30px; object-fit: contain; - position: absolute; - top: 0; - left: 0; + float: right; + margin-left: 10px; } .card-title {