More Like this

1. News source bias and sentiment on social media

   https://doi.org/10.1371/journal.pone.0305148  

   Knutson, Brian; Hsu, Tiffany W; Ko, Michael; Tsai, Jeanne L (October 2024, PLOS ONE)

   Candido, Silvio_Eduardo Alvarez (Ed.)

   As social media becomes a key channel for news consumption and sharing, proliferating partisan and mainstream news sources must increasingly compete for users’ attention. While affective qualities of news content may promote engagement, it is not clear whether news source bias influences affective content production or virality, or whether any differences have changed over time. We analyzed the sentiment of ~30 million posts (ontwitter.com) from 182 U.S. news sources that ranged from extreme left to right bias over the course of a decade (2011–2020). Biased news sources (on both left and right) produced more high arousal negative affective content than balanced sources. High arousal negative content also increased reposting for biased versus balanced sources. The combination of increased prevalence and virality for high arousal negative affective content was not evident for other types of affective content. Over a decade, the virality of high arousal negative affective content also increased, particularly in balanced news sources, and in posts about politics. Together, these findings reveal that high arousal negative affective content may promote the spread of news from biased sources, and conversely imply that sentiment analysis tools might help social media users to counteract these trends. 

   more » « less
2. Public Sentiment on Chinese Social Media during the Emergence of COVID19

   https://doi.org/10.51685/jqd.2021.013  

   Lu, Yingdan; Pan, Jennifer; Xu, Yiqing (April 2021, Journal of Quantitative Description: Digital Media)

   When COVID-19 first emerged in China, there was speculation that the outbreak would trigger public anger and weaken the Chinese regime. By analyzing millions of social media posts from Sina Weibo made between December 2019 and February 2020, we describe the contours of public, online discussions pertaining to COVID-19 in China. We find that discussions of COVID-19 became widespread on January 20, 2020, consisting primarily of personal reflections, opinions, updates, and appeals. We find that the largest bursts of discussion, which contain simultaneous spikes of criticism and support targeting the Chinese government, coincide with the January 23 lockdown of Wuhan and the February 7 death of Dr. Li Wenliang. Criticisms are directed at the government for perceived lack of action, incompetence, and wrongdoing—in particular, censoring information relevant to public welfare. Support is directed at the government for aggressive action and positive outcomes. As the crisis unfolds, the same events are interpreted differently by different people, with those who criticize focusing on the government’s shortcomings and those who praise focusing on the government’s actions. 

   more » « less
3. Understanding the role of media in the formation of public sentiment towards the police

   https://doi.org/10.1038/s44271-024-00059-8  

   Succar, Rayan; Ramallo, Salvador; Das, Rishita; Ventura, Roni Barak; Porfiri, Maurizio (December 2024, Communications Psychology)

   Abstract Public sentiment towards the police is a matter of great interest in the United States, as reports on police misconduct are increasingly being published in mass and social media. Here, we test how the public’s perception of the police can be majorly shaped by media reports of police brutality and local crime. We collect data on media coverage of police brutality and local crime, together with Twitter posts from 2010-2020 about the police in 18 metropolitan areas in the country. Using a range of model-free approaches building on transfer entropy analysis, we discover an association between public sentiment towards the police and media coverage of police brutality. We cautiously interpret this relationship as causal. Through this lens, the public’s sentiment towards the police appears to be driven by media-projected images of police misconduct, with no statistically significant evidence for a comparable effect driven by media reports on crimes. 

   more » « less
4. Leveraging Personalized Sentiment Lexicons for Sentiment Analysis

   https://doi.org/10.1145/3409256.3409850  

   Seyler, Dominic; Shen, Jiaming; Xiao, Jinfeng; Wang, Yiren; Zhai, ChengXiang (September 2020, ICTIR '20: Proceedings of the 2020 ACM SIGIR on International Conference on Theory of Information Retrieval)

   null (Ed.)
5. Weakly-Supervised Aspect-Based Sentiment Analysis via Joint Aspect-Sentiment Topic Embedding

   https://doi.org/10.18653/v1/2020.emnlp-main.568  

   Huang, Jiaxin; Meng, Yu; Guo, Fang; Ji, Heng; Han, Jiawei (January 2020, EMNLP'20: 2020 Conf. on Empirical Methods in Natural Language Processing, Nov. 2020)

   null (Ed.)

   Aspect-based sentiment analysis of review texts is of great value for understanding user feedback in a fine-grained manner. It has in general two sub-tasks: (i) extracting aspects from each review, and (ii) classifying aspect-based reviews by sentiment polarity. In this pa-per, we propose a weakly-supervised approach for aspect-based sentiment analysis, which uses only a few keywords describing each aspect/sentiment without using any labeled examples. Existing methods are either designed only for one of the sub-tasks, neglecting the benefit of coupling both, or are based on topic models that may contain overlapping concepts. We propose to first learn sentiment, aspectjoint topic embeddings in the word embedding space by imposing regularizations to encourage topic distinctiveness, and then use neural models to generalize the word-level discriminative information by pre-training the classifiers with embedding-based predictions and self-training them on unlabeled data. Our comprehensive performance analysis shows that our method generates quality joint topics and outperforms the baselines significantly (7.4%and 5.1% F1-score gain on average for aspect and sentiment classification respectively) on benchmark datasets. 

   more » « less