Assessing named entity recognition efficacy using diverse geoscience datasets

The development of Knowledge Graphs (KGs) significantly relies on the advancements in Named Entity Recognition (NER), which is often hindered by the limited availability of specialised, labelled datasets. Geoscience researchers are exploring innovative strategies for NER due to the lack of a robust labelled terms corpus. In this work, the efficacy of NER in the automatic generation of KGs is examined, and opportunities for further research are identified.