Construction of Chinese Obstetrics Knowledge Graph Based on the Multiple Sources Data

With the development of information technology, a large amount of information data has been accumulated in the field of obstetric. The effective ways to manage and apply these data are to construct a professional medical knowledge graph. In this paper, the Chinese Obstetric Knowledge Graph (COKG) based on multiple data sources of the obstetric professional thesauruses, clinical pathways, diagnosis, and treatment norms is constructed by the semi-automated method. The framework of concept classification and the related description are established. Thus COKG conceptual layer is also built. Based on traditional models of BI-LSTM-CRF and PCNN, and the guidance of medical experts, the data layer of COKG was founded by more than 2 million unstructured text words via artificially calibrating. Finally, COKG, which included 2343 diseases and 15249 named entity relationships, is constructed by knowledge fusion of multi-source data. The constructed COKG can provide structured knowledge support for medical question-answering systems, intelligent assisted diagnosis and treatment.