[Objective/Significance]Based on the multi-source data in PubMed,OMIM and other medical databases,the construction scheme of disease knowledge graph is designed to provide reference and basis for biological experimental research,diagnosis and treatment of diseases.[Methods/Processes]Firstly,SPO triples are extracted by SemRep,and knowledge fusion is carried out by data processing methods such as entity alignment and relationship mapping.Then,knowledge storage and visual display are realized by Neo4j graph database.Taking polycystic ovary syndrome as an example,61589 SPO triples,34697 entities and 27 semantic relationships are finally obtained,and 7 semantic patterns are summarized.[Limitations]In the process of data processing,manual examination is involved,but due to the large amount of data,there may be some errors in the examination process.[Results/Conclusions]This study improves the existing knowledge fusion method and verifies the feasibility of the disease knowledge graph construction scheme.It lays a foundation for the follow-up exploration of knowledge discovery in medical field based on disease knowledge graph.
维普
万方数据