Automatic Knowledge Extraction from a Digital Library and Collaborative Validation

Abstract

The automatic extraction of knowledge from a Digital Library is a crucial task in the world of Digital Humanities by enabling the discovery of information on a scale that is not achievable by human experts alone. However, the automation of information extraction processes brings the typical problems of a fully automated process, namely data quality and explainability of the results. This paper proposes a system that allows domain experts to collaboratively validate information previously automatically extracted from a Digital Library (DL), supporting an incremental data quality improvement approach, specifically through entity linking. Furthermore, rather than seeing just the results of the extraction process, the domain experts can trace the origin of where the AI recognized a specific entity (i.e. a “snippet” of text or an image).

In order to allow the domain experts to contextualize the information they need to validate (i.e. topics, descriptions, etc.) leveraging the Knowledge Graph potential, in the proposed use case, the validation is integrated into the interface designated for the DL semantic exploration.

This work has been partly supported by projects ARCA (POR FESR Lazio 2014–2020 - Avviso pubblico “Creatività 2020”, domanda prot. n. A0128-2017-17189) and SCIBA (Regione Lazio and MIUR - Determinazione n. G07413 del 16/06/2021). Miguel Ceriani acknowledges funding from the Italian Ministry of University and Research (MIUR), within the European program PON R &I 2014–2020 – Attraction and International Mobility (AIM) – project no. COD. AIM1852414, activity 2, line 1.