原标题：A joint model for entity and relationextraction based on BERT
作者：Bo Qiao、ZhuoyangZou、Yu Huan、Kui Fang、Xinghui Zhu、Yiming Chen
关键词：Agricultural knowledgegraph, Named entity recognition, Relation extraction, Joint extraction, BERT
中文摘要：
近年来，随着知识图谱在许多特定领域取得了显著成果，目前已成为互联网和人工智能领域发展的核心驱动力之一。然而，在农业领域还没有成熟的知识图谱，因此对农业知识图谱构建技术的研究具有重要意义。命名实体识别和关系抽取是构建知识图谱的关键步骤。本文在联合提取模型LSTM-LSTM-Bias的基础上，引入BERT预训练语言模型，提出了农业实体关系联合提取模型BERT-BILSTM-LSTM，将该模型应用于标准数据集NYT和自建立的农业数据集AgriRelation上，实验结果表明，该模型能有效提取农业实体与实体之间的关系。
英文摘要：
In recent years, as the knowledge graph hasattained significant achievements in many specific fields, which has become oneof the core driving forces for the development of the internet and artificialintelligence. However, there is no mature knowledge graph in the field ofagriculture, so it is a great significance study on the construction technologyof agricultural knowledge graph. Named entity recognition andrelation extraction are key steps in the construction of knowledge graph. Inthis paper, based on the joint extraction model LSTM-LSTM-Bias brought in BERTpre-training language model to proposed a agricultural entity relationshipjoint extraction model BERT-BILSTM-LSTM which is applied to the standard dataset NYT and self-built agricultural data set AgriRelation. Experimental resultsshowed that the model can effectively extracted the relationship betweenagricultural entities and entities.
原文链接：
https://link.springer.com/content/pdf/10.1007/s00521-021-05815-z.pdf
文献总结：
