2023-10-11 –, News Room
The rapid digitisation of cultural heritage data has resulted in vast repositories of information that encompass diverse entities such as artworks, historical figures, institutions and landmarks. However, the effective utilisation of this data is hindered by the challenge of entity disambiguation—accurately identifying and resolving references to entities with similar names or ambiguous contexts, and linking them to external knowledge bases such as Wikidata.
This presentation aims to address this issue by leveraging the power of deep learning models for entity disambiguation in cultural heritage data. The proposed approach combines state-of-the-art deep learning techniques with advanced natural language processing algorithms to improve the accuracy and efficiency of entity disambiguation. By harnessing the contextual information embedded within the textual descriptions, metadata, and interconnections of cultural heritage data, the models can discern subtle semantic cues that aid in disambiguation.
Key points to be discussed in the presentation include an overview of the prevalent challenges in entity disambiguation within cultural heritage data, the design and architecture of the deep learning approach to the problem, and the experimental evaluation of their performance. Overall, this presentation highlights the potential of deep learning models to advance entity disambiguation in cultural heritage data. By resolving ambiguities and connecting related entities more accurately, these models contribute to a deeper understanding and broader accessibility of our rich cultural heritage.
Orfeas Menis - Mastromichalakis is a graduate of the School of Electrical and Computer Engineering of the National Technical University of Athens (NTUA). He is a Ph.D. student at the Artificial Intelligence and Learning Systems (AILS) Laboratory of NTUA, and he is also a Master's student at the "Science, Technology, and Society" MsC program of the Department of History and Philosophy of the National and Kapodistrian University of Athens. His research interests include Explainable Artificial Intelligence (XAI), Natural Language Processing (NLP), as well as the social impact and ethics of AI.