Semantic Data New York 2025

Background

Over the past five years, Yale University has built an innovative knowledge graph that connects its libraries, archives and museums in a single discovery platform, advancing the university’s core missions of teaching and research excellence. 

Without changing data creation practices or introducing new systems of record, Yale was able to put into production a freely accessible graph of more than 2.5 billion triples, around 42 million records, using existing semantic standards for interoperability. 

In doing so, Yale had to call into question some of the core principles of linked open data and, in response, focused on practical solutions through data and platform usability. 

Because expectations around usability are changing rapidly Yale is investing in AI-powered solutions that provide additional functionality and ease of use. 

What Yale built and did – and the lessons learned

• What LUX is - and how and why we built it. 

• Lessons learned about data transformation and enrichment.

• Having documents and the graph coexist.

• How generative AI can improve end user access to the knowledge we manage and preserve. 

Yale’s system manages cultural data. However, the best-in-class practices we have developed can easily be applied across domains and technologies. Beyond the technology and functional improvements, teams dramatically improved internal operations, moving from competition to collaboration and working together rather than across siloed databases.