Research on Medical Named Entity Recognition Technology Based on Prompt BioMRC Model for Deep NLP Algorithm
International Journal of Big Data Intelligent Technology · 2025
Biomedical science has seen remarkable technological progress, resulting in vast amounts of research findings and clinical information.However, processing this unstructured data manually presents significant challenges.Natural language processing (NLP), particularly biomedical entity recognition (BioNER), offers promising solutions in this field.BioNER focuses on identifying medical and biological entities within scientific texts but encounters several obstacles.This paper examines common deep learning approaches and introduces two new models: a character-enhanced recognition system (Bert CNN CapsNET) utilizing capsule networks, and a biomedical entity recognition system (BioBERT GAT-NET) that incorporates grammatical structures and word relationships.Testing confirms that both models substantially improve entity identification accuracy.Nevertheless, BioNER technology continues to develop.Future research directions include analyzing combined data types, recognizing entities across different languages, applying knowledge from one domain to another, adapting to specific fields, and developing more transparent and reliable models.These advancements will enhance performance and facilitate broader implementation in both research and clinical environments.