The Intelligent Twin: Understanding the Past, Navigating the Future with AI

This resource provides an overview of how Artificial Intelligence (AI) and semantic technologies can support the development of Intelligent Digital Twins for cultural heritage. It explores semantic search systems, vector embeddings, predictive modelling, and simulation environments capable of transforming fragmented datasets into accessible and meaningful knowledge. Particular emphasis is placed on natural language querying, similarity-based discovery of hidden relationships between cultural heritage objects, and the role of structured, interoperable data infrastructures in making heritage information more intuitive and accessible for both experts and non-technical users.

Learning Outcomes

After completing this module, learners will be able to: 

• explain how knowledge graphs, natural language querying, and vector embeddings enable more intuitive and semantic access to cultural heritage data.

• describe the challenges of managing cultural heritage data, including fragmentation, complexity, interoperability, and accessibility barriers. 

• analyze the differences between keyword-based search and meaning-based semantic search using vector embeddings. 

• evaluate how AI techniques such as similarity search, predictive modeling, and simulation can support cultural heritage conservation, restoration planning, and future scenario analysis.