An Interactive Stance Classification Method Incorporating Background Knowledge

doi:10.13209/j.0479-8023.2019.096

Acta Scientiarum Naturalium Universitatis Pekinensis ›› 2020, Vol. 56 ›› Issue (1): 16-22.DOI: 10.13209/j.0479-8023.2019.096

Previous Articles Next Articles

An Interactive Stance Classification Method Incorporating Background Knowledge

LIU Changjian¹, DU Jiachen¹, LENG Jia¹, CHEN Di¹, MAO Ruibin², ZHANG Jun², XU Ruifeng^1,†

1. School of Computer Science and Technology, Harbin Institute of Technology (Shenzhen), Shenzhen 518055
2. Shenzhen Securites Information Co. Ltd., Shenzhen 518028

Received:2019-05-20 Revised:2019-09-27 Online:2020-01-20 Published:2020-01-20
Contact: XU Ruifeng, E-mail: xuruifeng(at)hit.edu.cn

一种融入背景知识的交互文本立场分析方法

刘常健¹, 杜嘉晨¹, 冷佳¹, 陈荻¹, 毛瑞彬², 张俊², 徐睿峰^1,†

1. 哈尔滨工业大学(深圳)计算机科学与技术学院, 深圳 518055 2. 深圳证券信息有限公司, 深圳 518028

通讯作者: 徐睿峰, E-mail: xuruifeng(at)hit.edu.cn
基金资助:
国家自然科学基金(U1636103, 61632011, 61876053)、深圳市基础研究项目(JCYJ20180507183527919, JCYJ20180507183608379)、深圳市技术攻关项目(JSGG20170817140856618)和深圳证券信息联合研究计划资助

Abstract

Abstract:

This paper proposes a stance classification method on interactive text by incorporating background knowledge. This method retrieves relevant background knowledge texts from Wikipedia by using the interactive text as query. The retrieved background knowledge texts are encoded and then ultilized to learn the representation of relavent background knowledge through deep memory network for improving the representation learning of interactive text. The experimental results on three English online debate datasets show that the performance of interactive stance classification can be effectively improved by incorporating background knowledge through choosing the appropriate number of background knowledge embedding layers and the connection method of background knowledge embedding layer.

Key words: stance classification, interactive text, background knowledge, deep memory network

摘要：

提出一种融入背景知识的交互文本立场分析方法。该方法以交互文本作为查询, 从维基百科中检索相关的背景知识文本, 然后对背景知识文本进行编码, 并通过深度记忆网络获取相关的背景知识特征, 以此来增强交互文本的表示学习。在3个英文在线辩论数据集上的实验结果表明, 通过选取适当的背景知识嵌入层数以及背景知识嵌入层连接方式, 可以有效地提高交互文本立场分析性能。

关键词: 立场分析, 交互文本, 背景知识, 深度记忆网络

LIU Changjian, DU Jiachen, LENG Jia, CHEN Di, MAO Ruibin, ZHANG Jun, XU Ruifeng. An Interactive Stance Classification Method Incorporating Background Knowledge[J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2020, 56(1): 16-22.

刘常健, 杜嘉晨, 冷佳, 陈荻, 毛瑞彬, 张俊, 徐睿峰. 一种融入背景知识的交互文本立场分析方法[J]. 北京大学学报自然科学版, 2020, 56(1): 16-22.

Add to citation manager EndNote|Ris|BibTeX

URL: https://xbna.pku.edu.cn/EN/10.13209/j.0479-8023.2019.096

https://xbna.pku.edu.cn/EN/Y2020/V56/I1/16

An Interactive Stance Classification Method Incorporating Background Knowledge

一种融入背景知识的交互文本立场分析方法

PDF

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 0

Recommended Articles

Metrics