SAN FRANCISCO – October 11, 2023 – ArangoDB, the company behind the most complete and scalable graph data and analytics platform, announced the GA release of ArangoGraphML, a fully managed and intuitive graph machine learning platform.
In today’s unpredictable times, businesses must balance cost-efficiency, customer satisfaction, and revenue growth while protecting themselves from fraud and security risks. Deeply connected graph data and machine learning provide the key to unlocking valuable insights for optimal decision-making in these areas. However, the adoption of graph-powered ML has historically been hindered by overwhelming complexity, poor accuracy, and the need for specialist expertise.
ArangoGraphML bridges this gap by providing enterprise-ready, graph-powered machine learning as-a-service – helping both experts and non-experts extract deeper insights and predictions from their data. ArangoGraphML is available as part of the ArangoGraph cloud-based graph data and analytics platform. With this release, ArangoGraph offers users a comprehensive set of data management and analytical tools – all unified by a powerful query language that natively integrates graph, JSON, full-text search, and now machine learning.
What others have to say about GraphML and ArangoDB:
“GraphML was the domain of highly skilled PhD Data Scientists – you have made it accessible to a far broader audience.’’ – Erik J. Larson, author of “The Myth of Artificial Intelligence.”
“It systematizes at scale, adding graph features and embeddings to your typical transaction-level models.” – Satheesh Ramachandran, Head of AI and Analytics Products, Charles Schwab, Fraud Detection.
“My team has been really impressed by the GraphML approach that ArangoDB has taken. Their low-code Python strategy also makes it easy for us to try new machine learning methods and evaluate GraphML concepts using the built-in ArangoDB database integration.” – Gareth Block, CEO & Founder, ThirdInsight.
The ArangoGraphML platform is a managed service that provides ease of use, high performance, efficiency, and accuracy. Given the edge every company is trying to get in these times, the accuracy uplift can make a significant impact. “We are very excited about the ArangoGraphML GA release that has been years in the making. The feedback has been very positive, and we feel the power of GraphDB and ML will drive huge business value for real-world use cases for our customers” said Shekhar Iyer, CEO of ArangoDB.
Leveraging cutting-edge advancements in graph machine learning, particularly Graph Neural Networks (GNNs), ArangoGraphML integrates these techniques into a seamless end-to-end ML lifecycle. Using the universally-recognized Jupyter Notebooks as an intuitive interface, it oversees every step, from data preparation to training to predictions. With its ease of use and robust capabilities, ArangoGraphML is set to redefine how enterprises approach machine learning with graph data.
Jörg Schad, ArangoDB CTO, said: “Graph ML offers the next level in Machine Learning and Data Science by considering the Data not as isolated data points but in context. This allows faster and more precise insights. Given our long experience with graph and our mission to simplify the access to the Power of Graph, we are happy to take the same approach to the Machine Learning ecosystem. Building production-grade Machine Learning pipelines is challenging, and ArangoGraphML contains best practices and combines some of the leading Open Source Machine tools into a complete end-to-end platform.”
Dive into a comprehensive graph machine learning journey with ArangoGraphML, or initiate your exploration with a free trial of ArangoGraph today.
Learn More:
- Join Shekhar Iyer, ArangoDB CEO, at the ArangoDB annual Summit event to hear more about ArangoGraphML and other special announcements.
- Check out arangodb.com
- Activate ArangoGraph, our cloud platform, and take it for a spin: Sign up for a free trial and join our next Cloud Lab to walk you through your first deployment.
- Learn more about ArangoGraphML or use the Jupyter Notebooks-as-a-service.