Neo4j and AWS strengthen their collaboration

The collaboration between Neo4j and Amazon Web Services (AWS) has existed since 2013. After 10 years, the collaboration is now being extended to a Strategic Collaboration Agreement over several years. Among other things, AWS wants to meet the demand for long-term storage for the validation of specific company data and domains for Large Language Models (LLMs).

Neo4j and AWS: The collaboration

Neo4j is one of the leading providers of graph databases and graph analytics. Neo4j's solutions enable relationships and patterns in billions of pieces of data to be recognized quickly and easily. This data structure is used, for example, for fraud detection or network management.

AWS is particularly interested in the native graph data storage and graph analytics that Neo4j offers. Neo4j therefore also offers Neo4j Aura Professional in the AWS Marketplace alongside Neo4j Aura Enterprise. This is a cloud-networked graph database that is fully managed and therefore enables a quick introduction to generative AI.

Neo4j integrates the applications directly into Amazon Bedrock. Bedrock was launched just a few weeks ago and combines numerous Foundation Models (FMs) from leading AI companies whose applications and functions ensure the secure and flexible use of generative AI. The graph databases and graph analytics from Neo4j expand this offering.

The technical options of Neo4j's graph databases

AWS hopes that the collaboration with Neo4j will solve certain problems or rather improve existing functions. Neo4j's applications are important in these areas, among others:

- AI hallucinations: When an AI makes false statements, we speak of a hallucination. The AI is "making things up". Retrieval Augmented Generation (RAG) is intended to reduce or, at best, completely eliminate these hallucinations. This works through virtual assistants that are based exclusively on internal, curated data from a company. Data limitation reduces hallucinations and makes it easy to draw conclusions about the source.

- Personalized content: Knowledge graphs enable the creation of extensive basic model collections. It is not necessary to develop new models. Instead, the FMs serve as a basis on which new developments can be built.

- Holistic answers: Search functions are always popular and the more precise the search, the better. Neo4j offers features that allow real-time search to be enhanced with vector embeddings of unstructured data such as images or text. Neo4j's knowledge graphs are so detailed that customers can search for products using clear search terms such as categories or ID as well as descriptions or images.

- Quick start for knowledge graphs: Amazon Bedrock's generative AI also helps to structure data. This enables faster access to the knowledge graphs and decisions can be made in real time.

Neo4j plans to present the new features and integrations at AWS re:Invent from November 27 to 30 in Las Vegas.

Get to know AWS with skill it

AWS offers a huge number of applications for all kinds of processes. Getting started can be overwhelming at first. We are happy to help you find your way around AWS. In our training AWS Technical Essentials we give you an overview of the most important applications and functions in AWS. In the training AWS Cloud Practitioner Essentials we prepare you to become a certified Cloud Practitioner. With this basic knowledge, nothing stands in the way of your path through the numerous functions of AWS.

‍