双关系预测与特征融合的实体关系抽取模型
Entity relation extraction model with dual relation prediction and feature fusion
沈健 1夏鸿斌 2刘渊2
作者信息
- 1. 江南大学 人工智能与计算机学院, 江苏 无锡 214122
- 2. 江南大学 人工智能与计算机学院, 江苏 无锡 214122;江苏省媒体设计与软件技术重点实验室, 江苏 无锡 214122
- 折叠
摘要
现有分阶段解码的实体关系抽取模型仍存在着阶段间特征融合不充分的问题,会增大曝光偏差对抽取性能的影响.为此,提出一种双关系预测和特征融合的实体关系抽取模型(entity relation extraction model with dual relation prediction and feature fusion,DRPFF),该模型使用预训练的基于Transformer的双向编码表示模型(bidirectional encoder representation from transformers,BERT)对文本进行编码,并设计两阶段的双关系预测结构来减少抽取过程中错误三元组的生成.在阶段间通过门控线性单元(gated linear unit,GLU)和条件层规范化(conditional layer normalization,CLN)组合的结构来更好地融合实体之间的特征.在NYT和WebNLG这 2 个公开数据集上的试验结果表明,该模型相较于基线方法取得了更好的效果.
Abstract
The staged decoding entity relation extraction model still has an insufficient feature fusion problem between stages,which increases the impact of exposure bias on the extraction performance.Herein,we propose a new entity rela-tion extraction model with dual relation prediction and feature fusion(DRPFF).DRPFF uses a pretrained model of bid-irectional encoder representation from transformers to encode texts,and a two-stage dual relation prediction structure is developed to reduce the false triples'generation.Between stages,a structure combining gated linear units and condition-al layer normalization is utilized to fuse features better between entities.Experimental findings on two public datasets,NYT and WebNLG,demonstrate that the presented method has better results than the baseline methods.
关键词
实体关系抽取/关系三元组/预训练模型/双关系预测/指针网络/特征融合/门控线性单元/条件层规范化Key words
entity relation extraction/relational triple/BERT pretrained model/dual relation prediction/pointer net-work/feature fusion/gated linear unit/conditional layer normalization引用本文复制引用
出版年
2024
国家科技期刊平台
NSTL
万方数据