Toward establishing a knowledge graph for drought disaster based on ontology design and named entity recognition

Drought disasters have caused serious impacts on the social economy and ecological environment, which are continuously and increasingly exacerbated by climate warming and other factors. Drought disaster management usually involves processing a mass of isolated data from many fields expressed in different terminologies and formats. These heterogeneous data or so-called data silos have greatly hindered drought disaster management in an information-rich manner. Establishing a drought disaster knowledge graph can facilitate the reuse of these heterogeneous data and provide references for drought disaster management, and ontology design and named entity recognition are the two major challenges. Therefore, in this study, we first designed a drought disaster ontology by recognizing the major concepts in the drought disaster field and their relationships, which was implemented with an ontology modeling language. We next constructed a drought disaster corpus and an integrated entity recognition model that was built by integrating multiple deep learning methods. Finally, we applied the integrated entity recognition model to extract information from the CNKI literature database. The integrated model shows satisfactory results in drought disaster named entity recognition. We thus conclude that combining ontology and deep learning technology toward establishing a knowledge graph for drought disasters is promising.