首页 关于本刊 编 委 会 期刊动态 作者中心 审者中心 读者中心 下载中心 联系我们 English
 自动化学报  2018, Vol. 44 Issue (5): 811-818 PDF

1. 北京理工大学计算机学院 北京 100081;
2. 北京理工大学自然语言处理实验室 北京 100081

Causality Extraction With GAN
FENG Chong1,2, KANG Li-Qi1,2, SHI Ge1,2, HUANG He-Yan1,2
1. School of Computer Science and Technology, Beijing Institute of Technology, Beijing 100081;
2. Natural LanguageProcessing Laboratory, Beijing Institute of Technology, Beijing 100081
Manuscript received : August 31, 2017, accepted: January 8, 2018.
Corresponding author. FENG Chong  Associate professor at the College of Computer Science and Technology, Beijing Institute of Technology. He received his Ph. D. degree from the Department of Computer Science, University of Science and Technology of China in 2005. His research interest covers natural language processing, information extraction, and machine translation. Corresponding author of this paper
Recommended by Associate Editor LI Li
Abstract: Causality extraction is of important practical value in tasks such as event prediction, scenario generation, question answering, and textual implication; but most of the existing causality extraction methods require artificial definition of patterns and constraints and are heavily dependent on knowledge base. In this paper, the bidirectional gated recurrent units networks (BGRU) with attention mechanism are merged with confrontational learning by leveraging the confrontational learning characteristics of generative adversarial networks (GAN). Through redefining the generator and discriminator, the basic causality extraction network can construct a confrontation with the discriminator, and then obtain a high distinguishing feature from the causality interpretation information. Our experiments show that our approach leads to an improved performance over strong baselines.
Key words: Causality extraction     generative adversarial network (GAN)     attention mechanism     adversarial learning

 ${\pmb h^*} = {\rm tanh}({\pmb r})$ (4)

2.1.2 判别模型

2.2 训练过程

3 实验

3.1 数据集

3.2 BGRU因果关系抽取模型预训练

3.3 对抗训练

3.4 带注意力机制的对抗训练实验

3.5 与已有因果关系抽取方法的对比实验

 图 4 不同模型的对比实验 Figure 4 Comparative experiment of different models

4 结论