Copyright

Elisabeth Steffen, Milena Pustet, Helena Mihaljević

Published On

2024-06-21

Page Range

pp. 205–236

Language

  • English

Print Length

236 pages

8. Algorithms Against Antisemitism?

Towards The Automated Detection of Antisemitic Content Online

Chapter of: Antisemitism in Online Communication: Transdisciplinary Approaches to Hate Speech in the Twenty-First Century(pp. 205–236)
The proliferation of hateful and violent speech in online media underscores the need for technological support to combat such discourse, create safer and more inclusive online environments, support content moderation and study political-discourse dynamics online. Automated detection of antisemitic content has been little explored compared to other forms of hate-speech. This chapter examines the automated detection of antisemitic speech in online and social media using a corpus of online comments sourced from various online and social media platforms. The corpus spans a three-year period and encompasses diverse discourse events that were deemed likely to provoke antisemitic reactions. We adopt two approaches. First, we explore the efficacy of Perspective API, a popular content- moderation tool that rates texts in terms of, e.g., toxicity or identity-related attacks, in scoring antisemitic content as toxic. We find that the tool rates a high proportion of antisemitic texts with very low toxicity scores, indicating a potential blind spot for such content. Additionally, Perspective API demonstrates a keyword bias towards words related to Jewish identities, which could result in texts being falsely flagged and removed from platforms. Second, we fine-tune deep learning models to detect antisemitic texts. We show that OpenAI’s GPT-3.5 can be fine-tuned to effectively detect antisemitic speech in our corpus and beyond, with F1 scores above 0.7. We discuss current achievements in this area and point out directions for future work, such as the utilisation of prompt-based models.

Contributors

Elisabeth Steffen

(author)
https://orcid.org/0000-0003-0170-968X

Elisabeth Steffen is a researcher at the Berlin University of Applied Sciences (HTW Berlin). Her research interests include the intersections between natural language processing, supervised machine learning algorithms and cultural, social and political studies. This interdisciplinary focus is also reflected in her academic education as both a computer scientist (BA in Computer Science and Business Administration, HTW– University of Applied Sciences Berlin) and anthropologist of contemporary European societies (MA in European Ethnology, Humboldt University Berlin). Currently, she works as a research assistant in the project Digitaler Hass on conspiracy theories in the context of the COVID-19 pandemic.

Milena Pustet

(author)
https://orcid.org/0000-0003-3825-6530

Milena Pustet (BSc, University of Potsdam) is a research assistant at the Berlin University of Applied Sciences (HTW Berlin), specialising in data science, machine learning and natural language processing. As a member of the research projects Digitaler Hass and Decoding Antisemitism, which aim to better understand and combat online hate speech and antisemitism, her research focuses on understanding social phenomena online, particularly antisemitic and conspiracy-related discourses on social media.

Helena Mihaljević

(author)
https://orcid.org/0000-0003-0782-5382

Prof. Helena Mihaljević holds a chair in Data Science at the Berlin University of Applied Sciences (HTW Berlin). Her expertise lies in analysing data and technology, utilising cutting-edge methods in data science, machine learning and natural language processing. She is an accomplished researcher with a passion for interdisciplinary projects. She has contributed to a variety of projects, including the algorithmic detection of conspiracy theories and antisemitic hate speech. In her current role as the principal investigator for the research project Digital Hate, she focuses on studying conspiracy narratives related to the COVID-19 pandemic. She is also a Co-Investigator on the international research project Decoding Antisemitism, coordinated by the Technische Universität Berlin. Prior to her role at HTW Berlin, Prof. Mihaljević served as a Senior Data Scientist and gained substantial experience in scientific information infrastructure. She earned her PhD in Mathematics with a focus on the topological dynamics of entire transcendental functions.