Biomedical text named entity recognition method and system based on deep learning

The invention belongs to the field of artificial intelligence and natural language processing, and particularly relates to a biomedicine text named entity recognition method and system based on deep learning. The method comprises the following steps: acquiring biomedical text training data with a genome variation entity label, and enhancing the biomedical text training data to obtain enhanced data; performing word segmentation processing on the enhanced data according to an improved word segmentation tag method to obtain a word segmentation sequence; performing feature extraction on the word segmentation sequence through a BioBERT layer to obtain a word vector sequence; inputting the word vector sequence into a stacked BILSTM network to extract text position information to obtain a feature vector sequence; the attention layer adopts an improved scoring function to obtain semantic features of the feature vector sequence; respectively sending the semantic features into four task modules, calculating task loss,