Mining a Stroke Knowledge Graph from literature
Stroke has an acute onset and a high mortality rate, making it one of the most fatal diseases worldwide. To aid research in finding effective prevention methods and treatments, we aim to build a stroke-oriented knowledge graph via literature mining that includes both Western biomedical knowledge, and Traditional Chinese medicine. We employed a suite of biomedical named entity recognition approaches to tag genes, diseases, drugs, symptoms, Chinese herbs, and other entities from a large set of domain-specific literature to made the graph’s nodes. Linking nodes by relying on an existing rule-based approach and a pre-trained BioBERT model, we extracted and classified relationships among stroke-related entities. We constructed a searchable knowledge graph, StrokeKG with 46,983 nodes of 9 types, and 157,302 relationships of 30 types , to facilitate browsing of a wide range of relationships to explore new directions for stroke research and ideas for drug repurposing and discovery. Besides, we marked the results that can be verified to get accurate nodes and edges to provide practical and reliable stroke-related knowledge.
StrokeKG freely available at http://114.115.208.144:7474/browser/