Moral Foundations Twitter Corpus: A collection of 35k tweets annotated for moral sentiment

Hoover, Joe; Portillo-Wightman, Gwenyth; Yeh, Leigh; Havaldar, Shreya; Davani, Aida Mostafazadeh; Li, Ying; Kennedy, Brendan J.; Atari, Mohammad; Kamel, Zahra; Mendlen, Madelyn; Moreno, Gabriela; Park, Christina; Chang, Tingyee E.; Chin, Jenna K.; Leong, Christian; Leung, Jun Yen; Mirinjian, Arineh; Dehghani, Morteza

doi:10.31234/osf.io/w4f72

Cited by 29 publications

(50 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…For example, developmental research has shown that children and adults use similar moral reasoning in both hypothetical and real-life dilemmas (Walker et al, 1987). Furthermore, moral values as assessed by in-lab surveys have been shown to map closely onto how people discuss morally relevant events on social media (Hoover et al, 2020). Additionally, moral concerns as measured in lab-based judgments encompass the vast majority of moral events that participants describe in ecological momentary assessments throughout their days (Hofmann et al, 2014).…”

Section: Discussionmentioning

confidence: 99%

Against Empathy Bias: The Moral Value of Equitable Empathy

2021

View full text Add to dashboard Cite

Empathy has long been considered central to living a moral life. However, mounting evidence has shown that people’s empathy is often biased toward (i.e., felt more strongly for) others that they are close or similar to, igniting a debate over whether empathy is inherently morally flawed and should be abandoned in efforts to strive toward greater equity. This debate has focused on whether empathy limits the scope of our morality, but little consideration has been given to whether our moral beliefs may be limiting our empathy. Across two studies conducted on Amazon’s Mechanical Turk ( N = 604), we investigated moral judgments of biased and equitable feelings of empathy. We observed a moral preference for empathy toward socially close over distant others. However, feeling equal empathy for all people is seen as the most morally and socially valuable approach. These findings provide new theoretical insight into the relationship between empathy and morality, and they have implications for navigating toward a more egalitarian future.

show abstract

Section: Discussionmentioning

confidence: 99%

Against Empathy Bias: The Moral Value of Equitable Empathy

2021

View full text Add to dashboard Cite

show abstract

“…Previous works have also contributed to the detection of moral sentiments. Johnson and Goldwasser (2018) showed that policy frames (Boydstun et al, 2014) help in moral foundation prediction, Hoover et al (2020) proposed a dataset of 35k tweets annotated for moral foundations, used background knowledge for moral sentiment prediction, Xie et al (2019) proposed a text based framework to account for moral sentiment change, and used pretrained distributed representations of words to extend the Moral Foundations Dictionary (Graham et al, 2009) for detecting moral rhetoric.…”

Section: Related Workmentioning

confidence: 99%

“…These are referred to as Moral Foundations (MFs), each with a positive and a negative polarity, and include Care/Harm, Fairness/Cheating, Loyalty/Betrayal, Authority/-Subversion, and Purity/Degradation (Table 1 provides details). Identifying MF in text is a relatively new challenge and past work has relied on lexical resources such as the Moral Foundation Dictionary (Graham et al, 2009;Fulgoni et al, 2016;Xie et al, 2019) and annotated data (Johnson and Goldwasser, 2018;Hoover et al, 2020).…”

Section: Introductionmentioning

confidence: 99%

Identifying Morality Frames in Political Tweets using Relational Learning

Roy¹,

Pacheco²,

Goldwasser³

2021

Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing

View full text Add to dashboard Cite

Extracting moral sentiment from text is a vital component in understanding public opinion, social movements, and policy decisions. The Moral Foundation Theory identifies five moral foundations, each associated with a positive and negative polarity. However, moral sentiment is often motivated by its targets, which can correspond to individuals or collective entities. In this paper, we introduce morality frames, a representation framework for organizing moral attitudes directed at different entities, and come up with a novel and highquality annotated dataset of tweets written by US politicians. Then, we propose a relational learning model to predict moral attitudes towards entities and moral foundations jointly. We do qualitative and quantitative evaluations, showing that moral sentiment towards entities differs highly across political ideologies.

show abstract

“…The computational methods using MFT tend to rely on supervised approaches to predicting the moral sentiment reflected in text (Garten et al, 2016;Mooijman et al, 2018;Xie et al, 2020). Other related work has characterized moral biases in language models Xie et al, 2019), and contributed new datasets for tasks such as automatic ethical judgment and inference of sociomoral norms (Hoover et al, 2020;Lourie et al, 2020;Forbes et al, 2020). Existing work has also studied moral sentiment change over time (Xie et al, 2019) showing how word embeddings capture hidden moral biases underlying different concepts (e.g., slavery) in history.…”

Section: Related Work On Textual Inference Of Moral Sentiment In Nlpmentioning

confidence: 99%

An unsupervised framework for tracing textual sources of moral change

Ramezani¹,

Zhu

Rudzicz

et al. 2021

Findings of the Association for Computational Linguistics: EMNLP 2021

View full text Add to dashboard Cite

Morality plays an important role in social wellbeing, but people's moral perception is not stable and changes over time. Recent advances in natural language processing have shown that text is an effective medium for informing moral change, but no attempt has been made to quantify the origins of these changes. We present a novel unsupervised framework for tracing textual sources of moral change toward entities through time. We characterize moral change with probabilistic topical distributions and infer the source text that exerts prominent influence on the moral time course. We evaluate our framework on a diverse set of data ranging from social media to news articles. We show that our framework not only captures fine-grained human moral judgments, but also identifies coherent source topics of moral change triggered by historical events. We apply our methodology to analyze the news in the COVID-19 pandemic and demonstrate its utility in identifying sources of moral change in high-impact and real-time social events. 1 The 10 foundation categories follow from the Moral Foundations Theory including 5 opposing pairs:Care(+)/Harm( ), Fairness(+)/Cheating( ), Loyalty(+)/Betrayal( ), Authority(+)/Subversion( ), and Sanctity(+)/Degradation( ).

show abstract

Moral Foundations Twitter Corpus: A collection of 35k tweets annotated for moral sentiment

Cited by 29 publications

References 5 publications

Against Empathy Bias: The Moral Value of Equitable Empathy

Against Empathy Bias: The Moral Value of Equitable Empathy

Identifying Morality Frames in Political Tweets using Relational Learning

An unsupervised framework for tracing textual sources of moral change

Contact Info

Product

Resources

About