Chinese Entity Relations Classification Based on BERT-GRU-ATT

• Published in: Ji suan ji ke xue Vol. 49; no. 6; pp. 319 - 325
• Main Authors: Zhao, Dan-Dan, De-Huang, Meng, Jia-Na, Dong, Yu, Zhang, Pan
• Format: Journal Article
• Language: Chinese
• Published: Chongqing Guojia Kexue Jishu Bu 01.06.2022; Editorial office of Computer Science
• Subjects: chinese entity relations classification|pre-training model|gate recurrent unit|self-attention mechanism; Classification; Coders; Knowledge representation; Natural language processing; Segmentation; Semantic web; Sentences
• ISSN: 1002-137X
• DOI: 10.11896/jsjkx.210600123

As the basic task of natural language processing,entity relations classification plays a critical role in tasks such as knowledge graphs,intelligent question answering,semantic web construction and so on.This paper constructs the BERT-GRU-ATT model to classify Chinese entity relations.In order to eliminate the influence of Chinese word segmentation ambiguity on entity relations classification,the pre-training model BERT(bi-directional encoder representations from transformers) is introduced as the embedding layer to better obtain the context information of Chinese characters.Then gate recurrent unit(GRU) is used to capture the long-distance dependence of entities in sentences and self-attention mechanism(ATT) is used to strengthen the weight of characters that contribute significantly to relations classification,so as to obtain better results of entity relations classification.In order to enlarge the Chinese entity relations classification corpus,we translate the SemEval2010_Task8 English entity relations eva