Linguistic Benchmarks of Online News Article Quality

Arapakis, Ioannis; Peleja, Filipa; Cambazoglu, Barla Berkant; Magalhães, João

doi:10.18653/v1/p16-1178

Cited by 13 publications

(17 citation statements)

References 26 publications

(26 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the works that study bias, Arapakis et al (2016) collected a dataset of 561 news articles, each being labeled with 14 qualitative aspects along with article's subjectivity. Another dataset, the multi-perspective question answering (MPQA) corpus (Wiebe et al, 2005), contains 692 news articles, each with a label of its subjectivity.…”

Section: Political Bias Datasetsmentioning

confidence: 99%

Viable Threat on News Reading: Generating Biased News Using Natural Language Models

Gupta¹,

Nguyen

Yamagishi

et al. 2020

Proceedings of the Fourth Workshop on Natural Language Processing and Computational Social Science

View full text Add to dashboard Cite

Recent advancements in natural language generation has raised serious concerns. Highperformance language models are widely used for language generation tasks because they are able to produce fluent and meaningful sentences. These models are already being used to create fake news. They can also be exploited to generate biased news, which can then be used to attack news aggregators to change their reader's behavior and influence their bias. In this paper, we use a threat model to demonstrate that the publicly available language models can reliably generate biased news content based on an input original news. We also show that a large number of high-quality biased news articles can be generated using controllable text generation. A subjective evaluation with 80 participants demonstrated that the generated biased news is generally fluent, and a bias evaluation with 24 participants demonstrated that the bias (left or right) is usually evident in the generated articles and can be easily identified.

show abstract

Section: Political Bias Datasetsmentioning

confidence: 99%

Viable Threat on News Reading: Generating Biased News Using Natural Language Models

Gupta¹,

Nguyen

Yamagishi

et al. 2020

Proceedings of the Fourth Workshop on Natural Language Processing and Computational Social Science

View full text Add to dashboard Cite

show abstract

“…Beautiful phrasing: High-quality articles are often written using beautiful language [2,22], meaning more unusual phrasing and creative words that lead to more positive feedback [34]. We used Term Frequency-Inverse Document Frequency (tf•idf) as a proxy, because the use of rarer words can be seen as an indicator for the use of beautiful language.…”

Section: Stylementioning

confidence: 99%

“…Genre: The topic of the article reflects certain characteristics of its nature and it has been greatly investigated in previous work [2,22].…”

Section: Context Featuresmentioning

confidence: 99%

Opening the black box of perceived quality: Predicting endorsement on a blog site

Sotirakou

Trilling

Germanakos

et al. 2019

IEEE/WIC/ACM International Conference on Web Intelligence

View full text Add to dashboard Cite

Uncovering their readers' perceptions is of key importance for every news media organization to find methods to improve the quality of their product. It has the potential to facilitate journalists' work in attracting attention and gaining a loyal audience. Discovering which elements of a news story influence readers' perceptions has been a cross-disciplinary research goal for the past years, because it can play a crucial role in news dissemination and consumption in the digital age. Drawing upon literature in the various areas such as journalism, psychology, computer science, and AI, this paper proposes a machine learning approach that explores three dimensions of article features that can help predicting the online behavior of the reader. Results show that how the story is written, the topic, and certain aspects of the author's online reputation can affect reader endorsements and the perceived quality of an article. CCS CONCEPTS • Computing methodologies → Natural language processing; • Applied computing → Document management and text processing.

show abstract

“…Bias Analysis The analysis of media bias has been a subject of investigation for decades (Groseclose and Milyo, 2005;Fang et al, 2012;Arapakis et al, 2016). Various aspects of bias have been studied from different perspectives.…”

Section: Related Workmentioning

confidence: 99%

“…Bias Datasets To study the bias in the newspaper domain, several developed corpora include one or more label types related to bias. For example, the news quality corpus created by Arapakis et al (2016) comprises 561 articles, each of which being labeled with 14 different quality aspects including article's subjectivity. Also, the MPQA corpus contains a label for the subjectivity of its 692 news articles (Wiebe et al, 2005).…”

Section: Related Workmentioning

confidence: 99%

Learning to Flip the Bias of News Headlines

Chen¹,

Wachsmuth²,

Khatib³

et al. 2018

Proceedings of the 11th International Conference on Natural Language Generation

View full text Add to dashboard Cite

This paper introduces the task of "flipping" the bias of news articles: Given an article with a political bias (left or right), generate an article with the same topic but opposite bias. To study this task, we create a corpus with bias-labeled articles from allsides.com. As a first step, we analyze the corpus and discuss intrinsic characteristics of bias. They point to the main challenges of bias flipping, which in turn lead to a specific setting in the generation process. The paper in hand narrows down the general bias flipping task to focus on bias flipping for news article headlines. A manual annotation of headlines from each side reveals that they are self-informative in general and often convey bias. We apply an autoencoder incorporating information from an article's content to learn how to automatically flip the bias. From 200 generated headlines, 73 are classified as understandable by annotators, and 83 maintain the topic while having opposite bias. Insights from our analysis shed light on how to solve the main challenges of bias flipping.

show abstract

Linguistic Benchmarks of Online News Article Quality

Cited by 13 publications

References 26 publications

Viable Threat on News Reading: Generating Biased News Using Natural Language Models

Viable Threat on News Reading: Generating Biased News Using Natural Language Models

Opening the black box of perceived quality: Predicting endorsement on a blog site

Learning to Flip the Bias of News Headlines

Contact Info

Product

Resources

About