An Entity Alignment Model for Echinococcosis Knowledge Graph

Knowledge graph is a structured knowledge representation method that can be used to integrate knowledge in the field of echinococcosis, providing strong support for the prevention and treatment of echinococcosis. Echinococcosis, a zoonotic disease, is caused by the larval stage of tapeworms. It is of significant importance in terms of prevention, control strategies, and reducing the impact of the disease. In recent years, the study of RNA, genes, proteins, and therapeutic drugs related to echinococcosis has been undertaken by researchers. However, this research information has been widely disseminated across various repositories, dispersed data makes it difficult for researchers to understand the research status of echinococcosis. To address this issue, the EKGEA model has been proposed, which incorporates entity alignment technology into the field of echinococcosis to facilitate the fusion of echinococcosis knowledge graphs. A large amount of echinococcosis information is collected from multiple sources, including attribute information and relationship information. Based on these two types of data, two different language knowledge graphs for echinococcosis have been constructed. Additionally, the EKGEA model utilizes BioBert to construct attribute embeddings and makes use of BiLSTM to capture relation embeddings. Through graph alignment, a more complete echinococcosis knowledge graph containing English information can be obtained. Compared with four state-of-the-art entity alignment algorithms, EKGEA demonstrates superior performance on the echinococcosis dataset. The information of the knowledge graph can be obtained from the Echinococcosis Database (EchiDB).