当前位置: X-MOL 学术Journal of Quantitative Linguistics › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Quantifying Perceived Political Bias of Newspapers through a Document Classification Technique
Journal of Quantitative Linguistics ( IF 0.7 ) Pub Date : 2020-06-16 , DOI: 10.1080/09296174.2020.1771136
Hyungsuc Kang 1 , Janghoon Yang 1
Affiliation  

ABSTRACT

Even though a certain degree of political bias is unavoidable in the media, strong media bias is likely to have an impact on society, especially on the formation of public opinion. This research proposes a data-driven method for quantifying political bias of media contents. With a document classification technique called doc2vec and social data from Facebook posts, a model for analysing the bias is developed. By applying the model to contents of major South Korean newspapers, this paper demonstrates quantitatively that significant political bias exists in the newspapers in line with the perceived political bias.



中文翻译:

通过文件分类技术量化报纸的政治偏见

摘要

尽管媒体不可避免地存在一定程度的政治偏见,但强烈的媒体偏见很可能对社会产生影响,尤其是对舆论的形成产生影响。本研究提出了一种数据驱动的方法来量化媒体内容的政治偏见。使用称为 doc2vec 的文档分类技术和来自 Facebook 帖子的社交数据,开发了用于分析偏差的模型。通过将模型应用于韩国主要报纸的内容,本文定量地证明了报纸中存在与感知到的政治偏见一致的重大政治偏见。

更新日期:2020-06-16
down
wechat
bug