计算机工程与应用2019，Vol.55Issue(20)：107-113,7.DOI:10.3778/j.issn.1002-8331.1806-0438
基于双重注意力机制的远程监督中文关系抽取
Distant Supervision Chinese Relation Extraction Based on Dual Attention Mechanism
摘要
Abstract
Compared with the traditional supervised Chinese relation extraction, the method based on distant supervision can greatly avoid the shortage of training corpus, so it has received extensive attention. However, the performance of the methods based on distant supervision is seriously constrained by the wrong labels introduced in the process of construct-ing corpus. Therefore, in order to alleviate the impact of noisy data, a relation extraction model based on dual attention mechanism is proposed in this paper. The model can obtain the context semantic information of training instances by bidi-rectional gated recurrent unit network, and focus on the important semantic features in the instances through the character-level attention mechanism. At the same time, the instance-level attention mechanism is introduced to calculate the correla-tion between instance and the corresponding relation in multiple instances in order to reduce the weight of noisy data. The experimental results on the Chinese character relationship corpus based on hudong encyclopedia show that the model com-pared to the single attention mechanism models can effectively utilize the semantic information contained in the instances and reduce the influence of the wrong label instance, and get higher accuracy.关键词
中文关系抽取/远程监督/双重注意力机制/双向门限循环单元(BI-GRU)/互动百科Key words
Chinese relation extraction/distant supervision/dual attention mechanism/Bidirectional Gated Recurrent Uni(t BI-GRU)/hudong encyclopedia分类
信息技术与安全科学引用本文复制引用
车金立,唐力伟,邓士杰,苏续军..基于双重注意力机制的远程监督中文关系抽取[J].计算机工程与应用,2019,55(20):107-113,7.基金项目
国家自然科学基金(No.51575523) （No.51575523）
军内科研基金. （）
