Detecting East Asian Prejudice on Social Media

By Bertie Vidgen, Austin Botelho, David A. Broniatowski, E. Guest, Helen Z. Margetts, Rebekah Tromble, Zeerak Talat, Scott A. Hale. 2020

115 citations recorded by the source corpus.

During COVID-19 concerns have heightened about the spread of aggressive and hateful language online, especially hostility directed against East Asia and East Asian people. We report on a new dataset and the creation of a machine learning classifier that categorizes social media posts from Twitter into four classes: Hostility against East Asia, Criticism of East Asia, Meta-discussions of East Asian prejudice, and a neutral class. The classifier achieves a macro-F1 score of 0.83. We then conduct an in-depth ground-up error analysis and show that the model struggles with edge cases and ambiguous content. We provide the 20,000 tweet training dataset (annotated by experienced analysts), which also contains several secondary categories and additional flags. We also provide the 40,000 original annotations (before adjudication), the full codebook, annotations for COVID-19 relevance and East Asian relevance and stance for 1,000 hashtags, and the final model.

Subfields: Artificial Intelligence, Information Systems, Sociology and Political Science, AI Ethics, Fairness & Society

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Information Systems

In Sociology and Political Science

In AI Ethics, Fairness & Society

Citations in the corpus

HateCheck: Functional Tests for Hate Speech Detection Models (2020, methodology): “This could be addressed by “live” datasets, such as dynamic adversarial benchmarks (Nie et al., 2020; Vidgen et al., 2020b; Kiela et al., 2021).”
Learning from the Worst: Dynamically Generated Datasets to Improve Online Hate Detection (2021, background): “Benchmark datasets Several benchmark datasets have been put forward for online hate classification (Waseem and Hovy, 2016; Waseem, 2016; Davidson et al., 2017; Founta et al., 2018; Mandl et al., 2019; Zampieri et al., 2019, 2020; Vidgen et al., 2020, 2021).”
A Framework for the Computational Linguistic Analysis of Dehumanization (2020, background): “For example, Asians have faced increased prejudice and dehumanization since the beginning of the COVID-19 pandemic (Van Bavel et al., 2020; Vidgen et al., 2020; Ziems et al., 2020).”
Predicting Anti-Asian Hateful Users on Twitter during COVID-19 (2021, background): “gate online hate towards Asians during COVID-19, several studies have introduced new datasets and methods for hate detection towards Asians (Ziems et al., 2020; Vidgen et al., 2020).”
Confronting Abusive Language Online: A Survey from the Ethical and Human Rights Perspective (2020, background): “At the present time, the most achievable sense of explanation is to provide the user with a reliability score of models’ predictions (Miok et al., 2020; Vidgen et al., 2020b).”
Introducing CAD: the Contextual Abuse Dataset (2021, background): “In a classiﬁcation study of prejudice against East Asia, Vidgen et al. (2020) ﬁnd that 27% of classiﬁcation errors are due to annotation mistakes.”
ConvAbuse: Data, Analysis, and Benchmarks for Nuanced Abuse Detection in Conversational AI (2021, background): “Abusive language detection has received extensive attention for social media, (see e.g. Vidgen et al., 2020a), but far less within the context of conversational systems.”