Artificial Intelligence
Natural language processing
Semantic web

Extraction of composition relations within archaeological documents

Student: Erwan Marchand

Supervisor: Michel Gagnon

Co-supervisor(s): Amal Zouaq

The Quebec Ministry of Culture and Communications has many archaeological reports. These documents contain a lot of interesting information, for educational purposes for example, but these are lost in these unused reports. The objective of the research project is therefore to develop a tool capable, at first, of locating the interesting entities within the text and then identifying the relationships between them. As part of this project, only the composition relationship between artifacts and materials, very present in these texts, will be considered. The objective of the work is, using the natural language processing and the extraction of entities and relations, to be able to detect within text the presence of relations of compositions between artifacts and materials. The purpose of this work is, afterwards, to be able to populate an existing ontology to allow the easier exploitation of this informations.