ENTITY RELATION MINING METHOD BASED ON BIOMEDICAL LITERATURE

摘要

Disclosed in the present invention are an entity relation mining method based on biomedical literature, comprising the following steps: querying disease-related biomedical literature in a public database, and performing data preprocessing to obtain biomedical text data; performing biomedical named entity recognition on the obtained biomedical text data in combination with a regular matching template and a deep learning model; and mining entity relations on the basis of the entity recognition result by transfer learning and reinforcement learning methods. According to the present invention, by obtaining disease-related biomedical literature from a network, extracting abstracts and the titles and performing entity recognition and relation mining, biomedical named entities in the literature can be effectively recognized and hidden relations among various entities can be mined.