Joint extraction of entity and relation based on fine-tuning BERT for long biomedical literatures.
基於微調 BERT 進行長篇生物醫學文獻的實體與關係聯合提取。
Bioinform Adv 2024-12-24
A Combined Manual Annotation and Deep-Learning Natural Language Processing Study on Accurate Entity Extraction in Hereditary Disease Related Biomedical Literature.
在遺傳疾病相關生物醫學文獻中,準確實體提取的手動標註和深度學習自然語言處理研究。
Interdiscip Sci 2024-02-10
Optimized biomedical entity relation extraction method with data augmentation and classification using GPT-4 and Gemini.
使用 GPT-4 和 Gemini 的數據增強與分類優化生物醫學實體關係提取方法。
Database (Oxford) 2024-10-09
這篇論文探討了一種利用大型語言模型來增強生物醫學關係提取的方法,特別是Gemini和GPT-4。研究中使用GPT-4生成增強的訓練數據,並將這些數據與不同模型的輸出結合,透過集成學習提升預測準確性。此外,Gemini的回應也用來微調BioNLP-PubMed-Bert分類模型。結果顯示,在BioCreative VIII Track 01挑戰的評估數據集上,精確度、召回率和F1分數都有所提升。想了解更多,可以透過提供的網址查詢。
PubMedDOI
Evaluating Medical Entity Recognition in Health Care: Entity Model Quantitative Study.
評估醫療實體識別在健康照護中的應用:實體模型定量研究。
JMIR Med Inform 2024-10-17
PLRTE: Progressive learning for biomedical relation triplet extraction using large language models.
PLRTE:使用大型語言模型進行生物醫學關係三元組提取的進步學習。
J Biomed Inform 2024-10-19
An Accurate and Efficient Approach to Knowledge Extraction from Scientific Publications Using Structured Ontology Models, Graph Neural Networks, and Large Language Models.
一種準確且高效的科學文獻知識提取方法:使用結構化本體模型、圖神經網絡和大型語言模型。
Int J Mol Sci 2024-11-09