Quantifying Content Polarization on Twitter

Yang, Muheng; Wen, Xidao; Lin, Yu‐Ru; Deng, Lingjia

doi:10.1109/cic.2017.00047

Cited by 22 publications

(16 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In our study, we first identify two groups of users from Twitter. Our data collection extends a Twitter corpus collected in prior work (Yang et al, 2017), where the authors identified two groups of users-Trump-supporters and Clinton-supporters-who are likely to have distinct political and ideological positions. The two groups consist of those "exclusive followers"-i.e., Twitter users who followed only one presidential election candidate but not the other.…”

Section: Methodsmentioning

confidence: 99%

The dynamics of Twitter users’ gun narratives across major mass shooting events

Lin

Chung

2020

Humanit Soc Sci Commun

Self Cite

View full text Add to dashboard Cite

This study reveals a shift of gun-related narratives created by two ideological groups during three high-profile mass shootings in the United States across the years from 2016 to 2018. It utilizes large-scale, longitudinal social media traces from over 155,000 ideology-identifiable Twitter users. The study design leveraged both the linguistic dictionary approach as well as thematic coding inspired by Narrative Policy Framework, which allows for statistical and qualitative comparison. We found several distinctive narrative characteristics between the two ideology groups in response to the shooting events-two groups differed by how they incorporated linguistic and narrative features in their tweets in terms of policy stance, attribution (how one believed to be the problem, the cause or blame, and the solution), the rhetoric employed, and emotion throughout the incidents. The findings suggest how shooting events may penetrate the public discursive processes that had been previously dominated by existing ideological references and may facilitate discussions beyond ideological identities. Overall, in the wake of mass shooting events, the tweets adhering to the majority policy stance within a camp declined, whereas the proportion of mixed or flipped stance tweets increased. Meanwhile, more tweets were observed to express causal reasoning of a held policy stance, and a different pattern in the use of rhetoric schemes, such as the decline of provocative ridicule, emerged. The shifting patterns in users' narratives coincide with the two groups distinctive emotional response revealed in text. These findings offer insights into the opportunity to reconcile conflicts and the potential for creating civic technologies to improve the interpretability of linguistic and narrative signals and to support diverse narratives and framing.

show abstract

Section: Methodsmentioning

confidence: 99%

The dynamics of Twitter users’ gun narratives across major mass shooting events

Lin

Chung

2020

Humanit Soc Sci Commun

Self Cite

View full text Add to dashboard Cite

show abstract

“…Group membership is defined as an exclusive follower; i.e., Twitter users who followed only one presidential election candidate but not the other. In their study, Yang et al (2017) have validated the concept that exclusive followers make good proxies for group affiliations. Our final raw corpus used for this study consists of over 7 million tweets.…”

Section: Data Source: Twittermentioning

confidence: 83%

“…We focus on Twitter because it is a widely-used space for people to express their views on social topics. For our study, we rely on a prior work (Yang et al, 2017) that collected data from publicly posted tweets using official Twitter APIs during a time-frame close to the 2016 U.S. Presidential election. Two groups of users are identified -Clinton-supporters (Blue) and Trump-supporters (Red) -that are likely to have distinct political and ideological preferences.…”

Section: Data Source: Twittermentioning

confidence: 99%

Inflating Topic Relevance with Ideology: A Case Study of Political Ideology Bias in Social Topic Detection Models

Guo¹,

Hwa²,

Lin³

et al. 2020

Proceedings of the 28th International Conference on Computational Linguistics

Self Cite

View full text Add to dashboard Cite

We investigate the impact of political ideology biases in training data. Through a set of comparison studies, we examine the propagation of biases in several widely-used NLP models and its effect on the overall retrieval accuracy. Our work highlights the susceptibility of large, complex models to propagating the biases from human-selected input, which may lead to a deterioration of retrieval accuracy, and the importance of controlling for these biases. Finally, as a way to mitigate the bias, we propose to learn a text representation that is invariant to political ideology while still judging topic relevance.

show abstract

“…Another negative consequence that has been cited more specifically in the literature is the polarisation of political discussions in social media when people are trapped in a bubble that prevents them from receiving outsider information (Bakshy et al, 2015;Foth et al, 2016;Lahoti, Garimella, & Gionis, 2018;Quraishi, Fafalios, & Herder, 2018;Thonet et al, 2017;Yang et al, 2017). Previous literature has not found a significant relationship between exposure to an opposing political view and a change in people's political opinion (Bail et al, 2018).…”

Section: Impacts Of Filter Bubblesmentioning

confidence: 99%

A Conceptual Tool to Eliminate Filter Bubbles in Social Networks

Amrollahi¹

2021

AJIS

View full text Add to dashboard Cite

Reliance on social media as a source of information has lead to several challenges, including the limitation of sources to viewers’ preferences and desires, also known as filter bubbles. The formation of filter bubbles is a known risk to democracy. It can bring negative consequences like polarisation of the society, users’ tendency to extremist viewpoints and the proliferation of fake news. Previous studies have focused on specific aspects and paid less attention to a holistic approach for eliminating the notion. The current study, however, aims to propose a model for an integrated tool that assists users in avoiding filter bubbles in social networks. To this end, a systematic literature review has been undertaken, and initially, 571 papers in six top-ranked scientific databases have been identified. After excluding irrelevant studies and performing an in-depth analysis of the remaining papers, a classification of research studies is proposed. This classification is then used to introduce an overall architecture for an integrated tool that synthesises all previous studies and offers new features for avoiding filter bubbles. The study explains the components and features of the proposed architecture and concludes with a list of implications for the recommended tool.

show abstract

Quantifying Content Polarization on Twitter

Cited by 22 publications

References 17 publications

The dynamics of Twitter users’ gun narratives across major mass shooting events

The dynamics of Twitter users’ gun narratives across major mass shooting events

Inflating Topic Relevance with Ideology: A Case Study of Political Ideology Bias in Social Topic Detection Models

A Conceptual Tool to Eliminate Filter Bubbles in Social Networks

Contact Info

Product

Resources

About