Mining scholarly data for fine-grained knowledge graph construction

Knowledge graphs (KG) are large networks of entities and relationships, typically expressed as RDF triples, relevant to a specific domain or an organization. Scientific Knowledge Graphs (SKGs) focus on the scholarly domain and typically contain metadata describing research publications such as authors, venues, organizations, research topics, and citations. The next big challenge in this field regards the generation of SKGs that also contain an explicit representation of the knowledge presented in research publications. In this paper, we present a preliminary approach that uses a set of NLP and Deep Learning methods for extracting entities and relationships from research publications, and then integrates them in a KG. More specifically, we i) tackle the challenge of knowledge extraction by employing several state-of-the-art Natural Language Processing and Text Mining tools, ii) describe an approach for integrating entities and relationships generated by these tools, iii) analyze an automatically generated Knowledge Graph including 10 425 entities and 25 655 relationships derived from 12 007 publications in the field of Semantic Web, and iv) discuss some open problems that have not been solved yet