• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.05a
• /
• pp.451-454
• /
• 2020

딥러닝 기술을 이용한 정치적 성향의 편향성 분류를 위하여 신문 뉴스 기사를 수집하고, 머신러닝을 위한 학습 데이터를 구축하였다. 학습 데이터의 구축은 보수 성향과 진보 성향을 대표하는 6개 언론사의 뉴스에서 정치적 성향을 이진 분류 데이터로 구축하였다. 뉴스 기사의 수집 방법으로 최근 이슈들 중에서 정치적 성향과 밀접하게 관련이 있는 키워드 15개를 선정하고 이에 관한 뉴스 기사들을 수집하였다. 그 결과로 11,584개의 학습 및 실험용 데이터를 구축하였으며, 정치적 편향성 분류를 위한 머신러닝 모델을 설계하였다. 머신러닝 기법으로 학습 및 실험을 위해 형태소 단위의 임베딩을 이용하여 문장 및 문서 임베딩으로 확장하였으며, SVM(Support Vector Machine)을 이용하여 정치적 편향성 분류 실험을 수행한 결과로 75%의 정확도를 달성하였다.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.1
• /
• pp.1-8
• /
• 2021

In the political field of news articles, there are polarized and biased characteristics such as conservative and liberal, which is called political bias. We constructed keyword-based dataset to classify bias of news articles. Most embedding researches represent a sentence with sequence of morphemes. In our work, we expect that the number of unknown tokens will be reduced if the sentences are constituted by subwords that are segmented by the language model. We propose a document embedding model with subword tokenization and apply this model to SVM and feedforward neural network structure to classify the political bias. As a result of comparing the performance of the document embedding model with morphological analysis, the document embedding model with subwords showed the highest accuracy at 78.22%. It was confirmed that the number of unknown tokens was reduced by subword tokenization. Using the best performance embedding model in our bias classification task, we extract the keywords based on politicians. The bias of keywords was verified by the average similarity with the vector of politicians from each political tendency.

• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.16-21
• /
• 2022

사전학습 말뭉치는 위키백과 문서 뿐만 아니라 인터넷 커뮤니티의 텍스트 데이터를 포함한다. 이는 언어적 관념 및 사회적 편향된 정보를 포함하므로 사전학습된 언어 모델과 파인튜닝한 언어 모델은 편향성을 내포한다. 이에 따라 언어 모델의 중립성을 평가할 수 있는 지표의 필요성이 대두되었으나, 아직까지 언어 인공지능 모델의 정치적 중립성에 대해 정량적으로 평가할 수 있는 척도는 존재하지 않는다. 본 연구에서는 언어 모델의 정치적 편향도를 정량적으로 평가할 수 있는 지표를 제시하고 한국어 언어 모델에 대해 평가를 수행한다. 실험 결과, 위키피디아로 학습된 언어 모델이 가장 정치 중립적인 경향성을 나타내었고, 뉴스 댓글과 소셜 리뷰 데이터로 학습된 언어 모델의 경우 정치 보수적, 그리고 뉴스 기사를 기반으로 학습된 언어 모델에서 정치 진보적인 경향성을 나타냈다. 또한, 본 논문에서 제안하는 평가 방법의 안정성 검증은 각 언어 모델의 정치적 편향 평가 결과가 일관됨을 입증한다.

• Korean journal of communication and information
• /
• v.71
• /
• pp.274-298
• /
• 2015

Doctoral Candidate, Department of Communication, Pusan National University This study analyzed the political and social significance of the disaster accident news with the frame and bias concept. In particular, this study confirmed theoretically how domestic media biased frame when it presents problem definition, causing interpretation, moral evaluation, and post-prescription on the ferry Sewol accident, In addition, the bias of the frame was analyzed comparing what is the difference between the conservative newspapers and liberal newspapers. Findings are as follows. First, in diagnosis of ferry Sewol accident, news slanted fragmentation frame>personalization frame>authority-disorder frame. The Chosun Ilbo focus on fragmentation bias, meanwhile Hankyoreh focus on the authority disorder relatively. Second, in accident evaluation, responsibility frame> moral frame> problem-solution frame. The Chosun Ilbo focus on responsibility frame and moral frame. But Hankyoreh focus on responsibility frame and problem-solution frame. Third, in the matter of responsibility, government frame>personal frame>organizational frame. Chosun Ilbo biased responsibility of the government and individuals, while the Hankyoreh is relatively more emphasis on government responsibility and the responsibility of the organization also showed. Fourth, in problem solving, thematic frame and episodic frame bias appeared as rough and level. Chosun Ilbo showed episodic frame, Hankyoreh showed thematic frame. News frame and bias as well as ideological differences of media on ferry Sewol accident was discussed in the context of the social dimension.

• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.533-538
• /
• 2023

문서에서 저자의 의도와 주제, 그 안에 포함된 감성을 분석하는 것은 자연어 연구의 핵심적인 주제이다. 이와 유사하게 특정 글에 포함된 정치적 문화적 편향을 분석하는 것 역시 매우 의미 있는 연구주제이다. 우리는 최근 발생한 한 사건에 대하여 여러 신문사와 해당 신문사에서 생산한 기사를 중심으로 해당 글의 정치적 편향을 정량화 하는 방법을 제시한다. 그 방법은 선택된 주제어들의 문장 공간에서의 거리를 중심으로 그래프를 생성하고, 생성된 그래프의 기계학습을 통하여 편향과 특징을 분석하였다. 그리고 그 그래프들의 시간적 변화를 추적하여 특정 신문사에서 특정 사건에 대한 입장이 시간적으로 어떻게 변화하였는지를 동적으로 보여주는 그래프 애니메이션 시스템을 개발하였다. 실험을 위하여 최근 이슈에 대하여 12개의 신문사에서 약 2000여 개의 기사를 수집하였다. 그 결과, 약 82%의 정확도로 일반적으로 알려진 정치적 편향을 예측할 수 있었다. 또한, 학습 데이터에 쓰이지 않은 신문기사를 활용하여도 같은 정도의 정확도를 보임을 알 수 있었다. 우리는 이를 통하여 신문기사에서의 정치적 편향은 작성자나 신문사의 특성이 아니라 주제어들의 문장 공간에서의 거리 관계로 특성화할 수 있음을 보였다. 할 수 있다.

• The Journal of the Korea Contents Association
• /
• v.21 no.5
• /
• pp.1-10
• /
• 2021

This study empirically confirmed 'the political bias of the YouTube recommendation algorithm' and 'the selective exposure of user' to verify the Filter Bubble phenomenon of YouTube. For the experiment, two new YouTube accounts were opened and each account was trained simultaneously in a conservative and a liberal account for a week, and the "Recommended" videos were collected from each account every two days. Subsequently, through the text mining method, the goal of the research was to investigate whether conservative videos are more recommended in a righties account or lefties videos are more recommended in a lefties account. And then, this study examined if users who consumed political news videos via YouTube showed "selective exposure" received selected information according to their political orientation through a survey. As a result of the Text Mining, conservative videos are more recommended in the righties account, and liberal videos are more recommended in the lefties account. Additionally, most of the videos recommended in the righties/lefties account dealt with politically biased topics, and the topics covered in each account showed markedly definitive differences. And about 77% of the respondents showed selective exposure.

• Journal of the Korean Society for Library and Information Science
• /
• v.32 no.3
• /
• pp.113-126
• /
• 1998

This paper examines the influence of and in the process of collection building in public libraries. These events resulted in depriving library workers of their voluntary spirit. And these events have also become a main cause of ideological prejudices as reflected in collection building.

• The Journal of the Korea Contents Association
• /
• v.17 no.11
• /
• pp.554-566
• /
• 2017

Through the Broadcasting Act, Republic of Korea regulates the broadcasting system to remain neutral regarding particular party or candidate. However, as MBC and JTBC reports the issue of President Park's impeachment in bisected way - conservative and progressive - the controversy aroused. This research paper comparatively analyzed each broadcasting company's evening news by focusing on quantity aspect, reporting tendency regarding trend of public opinion and mass rally and the news frame. Our research showed that both JTBC and MBC had partially reported; JTBC on pro-impeachment rally's side which was called candlelight rally, and MBC on anti-impeachment rally's side, called Korean National Flag rally. Regarding the way how they reported the impeachment, JTBC reported much more in depth than MBC, and MBC reported the process emotionally, standing for President Park.

• Journal of Digital Convergence
• /
• v.12 no.11
• /
• pp.171-181
• /
• 2014

As the use of mass media in modern politics grows, its influential power is getting larger than before. Therefore, fairness of broadcasting is identified as a very important factor in the current law. In particular, whether the mass media has balanced attitude toward election issues has been a critical point, which maked the current law have separate provisons to deal with it. As for the fairness and bias, most existing studies had focused on how long the media dealt with the specific political issues, which leads to only quantitative analysis. Also, most analysis of the contents had been based on very personal judgement and evaluation of researchers rather than following the criteria which is based on scientific method. This study introduced the AHP analysis method to compare the quantitative data and qualitative data altogether, which aims to develope the indicator for weighted measures and measurement of the fairness. Research findings reveals that each broadcaster has, MBC was highly biased and KBS and SBS followed that. Compared with existing studies regarding the political fairness of the media.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.8
• /
• pp.999-1005
• /
• 2020

While search portals' 'Portal News' account for the largest portion of aggregated news outlet, its neutrality as an outlet is questionable. This is because news aggregation may lead to prejudiced information consumption by recommending biased news articles. In this paper we introduce a new method of measuring political bias of news articles by using deep learning. It can provide its readers with insights on critical thinking. For this method, we build the dataset for deep learning by analyzing articles' bias from keywords, sourced from the National Assembly proceedings, and assigning bias to said keywords. Based on these data, news article bias is calculated by applying deep learning with a combination of Convolution Neural Network and Recurrent Neural Network. Using this method, 95.6% of sentences are correctly distinguished as either conservative or progressive-biased; on the entire article, the accuracy is 46.0%. This enables analyzing any articles' bias between conservative and progressive unlike previous methods that were limited on article subjects.