Here are some resources mentioned during the session:

• Presentation Slides
• Follow Paco on LinkedIn, read his bio on Sessionize or see more resources on Derwen

Are you interested in joining our communities?

• GraphGeeks.org Discord
• Graph RAG Discord
• LinkedIn ERKG Discussion Group
• Consult with a Senzing Entity Resolution Expert

We’d love to see you at our next event! Keep an eye out for our upcoming webinars by subscribing to our mailing list.

ABOUT THIS WEBINAR

Graph RAG has recently emerged as a significant buzzword in the tech industry, driven by the growing popularity of using knowledge graphs to "ground" large language models (LLMs) with domain-specific facts. This approach enhances the overall quality of AI-generated responses by reducing "hallucinations" and enables faster data updates. It also helps to minimize the need for—and costs associated with—fine-tuning models.

In this talk, we will delve into the origins and architecture of Graph RAG, exploring the different variations of the technology. For instance, the term "graph" in Graph RAG can refer to at least six different concepts. We will review popular open-source libraries, tutorials, and best practices for implementing Graph RAG.

We'll discuss the implications of graph construction and updating practices in an enterprise environment, including techniques such as entity resolution and entity linking, and how these affect downstream AI applications. We'll also provide recommendations for valuable resources, such as open-source projects, online community forums for Graph RAG developers, relevant conferences, and top recent books for those interested in a hands-on deep dive into the technology.

Finally, we will examine generalized architectures for building and updating knowledge graphs, incorporating both structured and unstructured data sources, and discuss the impact of entity resolution on downstream AI applications.

Key takeaways

• Understand the various meanings and applications of "Graph RAG" technology.
• Explore how Graph RAG enhances AI by grounding LLMs with domain-specific knowledge.
• Gain access to hands-on tutorials and resources for further self-paced learning.

Important Note: Prerequisites include at least introductory Python programming experience.