Latest Neo4j release aims to simplify graph technology
Prebuilt algorithms that reduce the complexity of queries and connectivity to any data source are designed to make graph-based analysis accessible to a broader audience.
Neo4j on Wednesday launched Aura Graph Analytics, a new serverless platform that uses graph technology to uncover patterns and relationships among data points and can connect to any data source.
Now generally available on a pay-as-you-go basis, Aura Graph Analytics is designed to work on any cloud with all databases, data warehouses, data lakes and data lakehouses without forcing users to move data into Neo4j's graph database platform.
In addition, the graph database specialist's offering aims to make graph analytics easier to use by removing the need for custom queries -- specialized algorithms designed to discover specific patterns or relationships -- and instead comes with 65 prebuilt algorithms that reduce the need for graph expertise.
Graph technology is an approach to storing and analyzing interconnected data. With graph technology, which simultaneously connects multiple related data points to one another rather than just one at a time, users can discover complex patterns and networks that might not otherwise be found.
As a result, graph technology is valuable for applications such as fraud detection, recommendation engines and generative AI (GenAI) development, and applications that rely on networks and patterns, according to William McKnight, president of McKnight Consulting Group.
The launch marks a significant milestone similar to what leaders are doing in other realms -- making powerful analytics more accessible, enabling users to apply advanced algorithms to diverse data sets, drive better insights and enhance efficiency while integrating with AI and machine learning workflows.
William McKnightPresident, McKnight Consulting Group
"Graph is immensely important for applications that involve relationships, networks, patterns and even establishing the relative importance of elements of the business," he said. "Graph analytics enhances AI decision-making by revealing complex connections and patterns, providing deeper insights with added context."
Because Neo4j's Aura Graph Analytics aims to make graph technology more widely used, it is a valuable offering, McKnight continued.
"The launch marks a significant milestone similar to what leaders are doing in other realms -- making powerful analytics more accessible, enabling users to apply advanced algorithms to diverse data sets, drive better insights and enhance efficiency while integrating with AI and machine learning workflows," he said.
New capabilities
Until recently, graph technology was for specialized applications that rely on relationships between data. In addition, it was complex, requiring users to learn graph-specific query languages such as Cypher and GraphQL. Its use was therefore limited, and it remained a niche within data management and analytics.
GenAI development is changing that.
For GenAI applications to understand an enterprise's operations, they need to be trained using a combination of that enterprise's relevant proprietary data and GenAI model technology. In addition, the relevant data has to be high-quality and voluminous for applications to be accurate and reduce the occurrences of AI hallucinations.
One way many enterprises are searching for and discovering enough relevant high-quality data is by vectorizing data, which is the process of assigning numerical values to data to make similar data discoverable. Another is graph technology, which uses neural networks to discover relationships between data points.
As a result, graph technology is taking on greater importance, according to McKnight.
"Graph technology can … transform data into machine learning-ready features, boosting model accuracy and doubling insight efficacy through real-time adaptability and advanced algorithms," he said.
Amid its growing importance, Neo4j is attempting to make graph-based analysis more accessible with the release of Aura Graph Analytics.
In 2023, the vendor targeted performance speed to meet the needs of developers running high-volume workloads such as those required for GenAI development. Now, Neo4j is targeting more widespread use.
The 65 prebuilt algorithms reduce the need to know specialized query languages, lowering the need to write code by 75%, according to Neo4j. Meanwhile, the ability to work with any data source fully eliminates the need to develop complex -- and expensive -- extract, transform and load (ETL) pipelines.
In addition to reducing coding requirements and eliminating ETL workloads, Aura Graph Analytics benefits include the following, according to the vendor:
Increased model accuracy by automatically transforming graph structures into graph embeddings that can be used to inform machine learning and GenAI models.
Real-time updates that capture changes in data as they occur to further increase model accuracy.
A fully managed offering that removes administrative burdens and costs associated with server provisioning and maintenance.
Native integration with Snowflake coming in June.
Customers requesting a less complex means of using graph technology provided the motivation for developing Aura Graph Analytics, according to Sudhir Hasbe, Neo4j's chief product officer.
"The impetus came directly from customers who wanted the power of graph without the complexity," he said. "We built it to eliminate the barriers … so anyone can tap into its unique insights and improve decision-making."
In addition, expanding Neo4j's target market to more potential customers by enabling the vendor's graph algorithms to work with any data source played a role, Hasbe added.
Combined, Aura Graph Analytics serves to simplify graph analytics and potentially improve graph-driven outcomes, according to McKnight.
"Neo4j's Aura Graph Analytics … makes it easier for users to apply graph analytics without needing extensive expertise or infrastructure setup," he said. "The platform will also deliver significantly better analytical results, with improved accuracy and insights that drive better business decisions."
However, despite its benefits, Aura Graph Analytics doesn't fully eliminate the complexities of graph analytics, McKnight continued. To further simplify using its tools, Neo4j could add support for more graph query languages and integrate natively with cloud data warehouses beyond Snowflake, he said.
Looking ahead
With Aura Graph Analytics now generally available, Neo4j plans to continue trying to make graph technology more accessible, according to Hasbe.
The vendor's roadmap includes adding integrations with GenAI providers, adding more self-service analytics capabilities and reducing the complexity of its platform. In addition, Neo4j plans to add GenAI capabilities to its own platform to make it easier to use and more efficient, Hasbe said.
McKnight, meanwhile, suggested that Neo4j could potentially attract new customers by more clearly demonstrating graph technology's role in AI development and further addressing ease of use.
"Refining its AI/ML messaging, helping customers tame the potential chaos of having many more graphs in play, improving ease of use and publishing benchmarks could help drive growth and deepen its value proposition," he said. "These can help Neo4j reach a wider audience."
Eric Avidon is a senior news writer for Informa TechTarget and a journalist with more than 25 years of experience. He covers analytics and data management.
