【PaperReading】GraphIE：A Graph-Based Framework for Information Extraction

镰刀韭菜

已于 2022-10-14 13:09:52 修改

阅读量1.3k

点赞数

CC 4.0 BY-SA版权

分类专栏： PaperReading 文章标签：知识图谱信息提取图计算 GraphIE CRF

于 2020-04-27 16:25:12 首次发布

本文链接：https://blog.youkuaiyun.com/ARPOSPF/article/details/105235605

PaperReading 专栏收录该内容

65 篇文章 ¥49.90 ¥99.00

订阅专栏

超级会员免费看

GraphIE是一个图基信息提取框架，通过图卷积在文本单位间传播信息，改善了非本地和非顺序依赖性建模，从而在文本、社交媒体和视觉信息提取任务上超越了传统的序列标记模型。该框架利用图结构捕获依赖关系，通过图模块和循环神经网络增强局部上下文表示，提高预测准确性。实验结果显示，GraphIE在多个任务中显著优于基于BiLSTM和CRF的序列模型。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

GraphIE：A Graph-Based Framework for Information Extraction

GraphIE：基于图的信息提取框架

Yujie Qian,Enrico Santus,Zhijing Jin,Jiang Guo,Regina Barzilay

Abstract
Most modern Information Extraction (IE) systems are implemented as sequential taggers and only model local dependencies. Non-local and non-sequential context is, however, a valuable source of information to improve predictions. In this paper, we introduce GraphIE, a framework that operates over a graph representing a broad set of dependencies between textual units (i.e. words or sentences). The algorithm propagates information between connected nodes through graph convolutions, genera

了解本专栏