文章基本信息

标题：Automatic political discourse analysis with multi-scale convolutional neural networks and contextual data
本地全文：下载
作者：Aritz Bilbao-Jayo ; Aitor Almeida
期刊名称：International Journal of Distributed Sensor Networks
印刷版ISSN：1550-1329
电子版ISSN：1550-1477
出版年度：2018
卷号：14
期号：11
页码：1
DOI：10.1177/1550147718811827
出版社：Hindawi Publishing Corporation
摘要：In this article, the authors propose a new approach to automate the analysis of the political discourse of the citizens and public servants, to allow public administrations to better react to their needs and claims. The tool presented in this article can be applied to the analysis of the underlying political themes in any type of text, in order to better understand the reasons behind it. To do so, the authors have built a discourse classifier using multi-scale convolutional neural networks in seven different languages: Spanish, Finnish, Danish, English, German, French, and Italian. Each of the language-specific discourse classifiers has been trained with sentences extracted from annotated parties’ election manifestos. The analysis proves that enhancing the multi-scale convolutional neural networks with context data improves the political analysis results.
关键词：Supervised classification; convolutional neural networks; online political discourse; sentence classification