Towards Knowledge Graph Construction using Semantic Data Mining

Over the last few years, constructing knowledge graphs for new domains and linking them to existing ones has gained a lot of attention, especially in domains that have experienced a tremendous increase in available data such as biodiversity research. To this end, in this paper, we introduce a new semantic data mining-based approach to support the (semi-)automatic generation of a biodiversity knowledge graph. The proposed approach exploits and links information from several biodiversity-related resources including the Encyclopedia of Life (EOL), the Global Biodiversity Information Facility (GBIF), and the Global Biotic Interactions (GLOBI), as well as more general information like the GeoNames database, and the ORCID repository. In particular, we adopt a data mining technique to extract association rules that support the construction of an initial species interactions knowledge graph. We then make use of available biodiversity resources to enrich the knowledge graph. We believe that this graph will support scientists from the biodiversity domain to gain new insights and enrich the data interoperability.