Data product catalog: from data contracts to knowledge graph
The management of data as a product is at the core of all modern approaches to data management. The objective is to build modular data architectures capable of effectively handling their complexity and evolving sustainably over time. While data contracts are crucial for achieving interoperability among data products, they alone fall short in ensuring their composability. To this end, a data product catalog is necessary.
In this talk, we will explore what a data product catalog is and how it leverages metadata contained in data contracts to make data products addressable, discoverable, and understandable for consumers or in simpler terms, to make them effortlessly composable to implement an unbouded set of use cases. Specifically, we will see how the data product catalog can link the data and metadata of each data product to a semantic data model (domain ontology), thus constructing an enterprise knowledge graph essential for enabling self-service analytics and generative AI use cases.