Building your Enterprise Knowledge Graph one data product at a time
A Talk by Andrea Gioia (Quantyca SPA, Blindata SRL, DAMA, Milan, Italy)
About this Talk
Creating an Enterprise Knowledge Graph is a complex challenge that involves the entire organization. A top-down approach, where a central team of information architects designs the ontology and then one or more data steward teams link it to the managed data assets, simply doesn’t scale. Centralized modeling teams quickly become a bottleneck, leaving most new data assets disconnected from the knowledge graph turning it into a cathedral in the desert.
This masterclass aims to present a new distributed, iterative, and value-driven approach to knowledge modeling, taking participants from business cases and raw data into the creation of a full knowledge graph.
We will begin by introducing the concept of a data product and how it enables the construction of modular, distributed, and scalable data architectures over time. We will then explore how the principles used to manage data as products in distributed environments (data mesh) can also be applied to knowledge management (knowledge mesh). Specifically, we'll look at how to build an enterprise ontology by composing simpler ontologies, each managed autonomously as a product by different teams.
Once participants are comfortable with the concepts of data mesh, knowledge mesh, and their foundational principles, we will explore how to map data products (data plane) with concepts defined in corporate ontologies (knowledge plane) to build a virtual knowledge graph. To achieve this, we will dive into both the technological and organizational elements required to execute this mapping efficiently and effectively.
This masterclass is designed for data and knowledge engineers of all levels. Participants will learn how to define and implement a sociotechnical architecture to manage the entire information ecosystem—data, metadata, and knowledge—in a sustainable and distributed manner. From there they will learn how to query the generated knowledge graph and leverage it to augment GenAI models.
The end result will be that the participants will be able to set up within their organization an operating model to build a fully functional knowledge graph in an incremental and value-driven way, one data product at a time.
Key Topics
- Managing data as a product (data mesh)
- Managing knowledge as a product (knowledge mesh)
- Mapping data to knowledge (technical perspective)
- Mapping data to knowledge (organizational perspective)
- Augmenting GenAI models with ontology-based data access (OBDA)
Target Audience
- Data Scientists and Machine Learning Engineers
- Data Engineering
- Knowledge Engineer
- Data Analysts
- Managers of the above
Goals
This workshop aims to equip participants with the skills to develop and scale a modular, distributed enterprise knowledge graph using an incremental, value-driven approach that helps avoid common pitfalls of centralized knowledge modeling.
Session outline:
- A brief introduction to data products and data mesh
- Introduction to managing knowledge as a product in a distributed environment (knowledge mesh)
- How to technically map data products on ontology concepts (DPDS and DPROD)
- How to structure team topologies to manage data & knowledge mesh
- Methods to augment GeAI with the implemented virtual knowledge graph
Format
This class is held in a lecture format, with a lot of practical examples.
Level
Beginner - Intermediate
Prerequisite Knowledge
There are no specific prerequisite
Exclusive Offers
Register for this Talk to unlock these exclusive offers.
Holiday Special: Unlock the Secrets of Data Products for Just $9.99! 🎁📚
Offer
Holiday Special: Unlock the Secrets of Data Products for Just $9.99! 🎁📚
Offer
We’re a cloud tech company that provides organisations around the world with computing infrastructure and software to help them innovate, unlock efficiencies and become more effective. We also created the world’s first – and only – autonomous database to help organise and secure our customers’ data.
