GraphBI: Expanding Analytics to All Data Through the Combination of GenAI, Graph, & Visual…

GraphBI: Expanding Analytics to All Data Through the Combination of GenAI, Graph, & Visual Analytics

MLOps Podcast #310 with Paco Nathan, Principal DevRel Engineer at Senzing & Weidong Yang, CEO of Kineviz.

Take a deep dive into the challenges inherent in realizing the value of big data. In MLOps Podcast no. 310, Paco Nathan, Principal DevRel Engineer at Senzing and Weidong Yang, CEO of Kineviz talk about their experiences engaging with big data and graph technology, and how an emerging GraphBI strategy can enable visual analysis of connections in all types of data.

Basic GraphBI principles emerge through a wide-ranging discussion of human language and thought processes across cultures. A GraphBI environment should let you manage both structured and unstructured data, use GenAI for data pre-processing, perform iterative analytics using a BI-focused graph grammar, and seamlessly communicate insights.

For exploring unstructured data, graphs based on facts alone aren’t enough. You also need to preserve original context and nuance, because you’ll need it to fully understand documents later in the analytic process. That’s where GenAI for data pre-processing comes in: it can deliver a graph-based knowledge map that does retain crucial context, yet relaxes the constraints inherent in detailed knowledge graphs you get using black-box RAG (Retrieval-Augmented Generation) and Graph RAG processes.

“… I love how you all are approaching this. You have a very powerful view of … relaxing the constraints up front, but then having the context propagated through.” — Paco Nathan

To build a graph, how much detail should we start with? Usually at first you are simply trying to find the right questions to ask — something you’ll discover as you interact with the graph.

“The exploratory process comes up with the questions. Knowing what question to ask is often 80 -90% of the work.” — Weidong Yang

With GraphBI, the goal is to support visual analytics– the process of visually transforming information for clear, concise presentation of insights.

Listen to the podcast for more on key issues in delivering graph technology and visual analytics. 👉 https://podcasts.apple.com/us/podcast/graphbi-expanding-analytics-to-all-data-through-the/id1505372978?i=1000705418809