NLP Tools for Knowledge Extraction from Italian Archaeological Free Text

This paper deals with the development of advanced tools and technologies for creating relevant information and suitable metadata out of textual documentation produced by Italian archaeological research. A set of Natural Language Processing tools were developed to recognize and annotate various archaeological entities in Italian language textual reports. The CIDOC CRM is the ontology chosen for encoding resulting output, allowing for a maximum degree of standardisation of the produced metadata to guarantee interoperability with archaeological information already existing in other semantically enabled digital archives. The work took place as part of the development for the TEXTCROWD platform for the European Open Science Cloud for Research Pilot Project.