Innovation on tradition: a big data analysis

Motivated by the increasing demand for tools and methodologies to manage the ongoing digital transformation across various sectors ef- fectively, this thesis delves into innovative techniques tailored specifi- cally for ecclesiastical cultural heritage within the broader framework of Italian cultural assets. Keeping in mind the challenges and oppor- tunities presented by this domain, which is traditionally less associ- ated with these topics, the research aims to bridge the gap between advanced technological applications and cultural preservation. The study builds upon the assumption that the true value of big data lies not in its defining volume, but in the ability to extract action- able insights and generate meaningful connections. To achieve this, the research leverages state-of-the-art machine learning techniques, focusing on natural language processing (NLP) for extracting entities and relationships from texts and constructing knowledge graphs, and graph neural networks (GNNs) applied to real estate assets, namely churches. By employing these methodologies, the work uncovers hid- den patterns and relationships within ecclesiastical cultural heritage data, creating new perspectives on its future organization and man- agement. Therefore, this thesis highlights the transformative poten- tial of the technologies as mentioned earlier, paving the way for fur- therresearchandpracticalapplications, emphasizingtheircrucialrole in knowledge discovery and preservation of cultural assets for future generations.