Methodological Advances in the Study of Online Antisemitism:
Keywords:
multimodal analysis · computational social science · political communication · algorithmic social media · short-form video
Studying the extent of online hate speech and antisemitism is essential to understanding the dynamics of digital discourse, which can help predict ongoing societal changes beyond the digital realm. However, the underlying factors that make antisemitic narratives so persuasive are frequently overlooked. This project focuses on emotive layers embedded in digital media that influence the spread of hate speech and antisemitism. It examines how frequently users devalue Jews and Israel, contrasting these expressions with those directed at opposing entities, including instances of sympathy for, or even endorsement of, violent actors such as Hamas.
Beyond the Written Word…
Beyond the written word, the project analyzes the multifaceted layers of political communication by utilizing cutting-edge technology that incorporates both audio and visual signals into its analysis, aiming to understand and scale the dynamics that create fertile ground for hostility in digital realms.
Key questions driving the project include:
- What methodological advances are necessary to analyze online hate and measure multimodal content effectively?
- How can these dynamics be robustly measured at scale and tracked longitudinally?
- What recurring visual frames appear in today’s increasingly dominant short-form video content?
- Which metrics best explain why polarizing narratives spread widely on Web 2.0 platforms?
These questions are addressed through a series of high-quality NLP-based research papers.
The paper series conducts empirical research on how political narratives circulate online by analyzing international news content, particularly on YouTube—the second most visited website worldwide. It focuses on state-funded and state-supported outlets, such as Al Jazeera, the BBC, Deutsche Welle, and TRT World. Using a longitudinal approach, the project measures the proportionality and dynamics of content related to the aftermath of the Israel–Hamas war since October 7, 2023.
Activities (Talks & Conferences)
2026
- Analyzing Multimodal Political Communication in State-Funded News
Antisemitism & AI Forum, Contemporary Antisemitism, Haifa, Israel.
2025
Affect Mobilization on YouTube: Emotional Toning in State-Funded News Outlets Covering the Israel–Hamas War
ISCA Early Career Speaker Series, Indiana University, USA.Analyzing Polarization in Online Discourse on the 2023–2024 Israel–Hamas War
5th Workshop on Computational Linguistics for the Political and Social Sciences (CPSS), University of Hildesheim, Germany.Analyzing State Media and Online Commentary After October 7: A Longitudinal Study of Sentiment and Narrative
Touro University, New York City, USA.Emotion in Motion: Shifting Narratives and Sentiment in State Media and Social Discourse After October 7
Brock University, St. Catharines, Canada.Antisemitic Discourses and Discourses on Antisemitism Post–October 7
Jewish Community Centre, London, UK.Tracking Sentiment Shifts and Linguistic Patterns in Online Discourse After October 7
NLP Lab, Luddy School of Informatics, Indiana University, USA.
Empirical Studies
Study 4: How to Investigate Short‑Form Video Content — A Pipeline for Multimodal Analysis (under review)
This study shifts the analytical focus to so-called short videos by examining audio and visual layers of news content. It introduces a multimodal pipeline that combines automatic transcription, aspect-based sentiment analysis (ABSA), and visual scene classification to analyze variation in sentiment and framing across outlets and over time.
Link: (forthcoming)
Study 3: Mapping Affective Polarization in YouTube Shorts — A Data‑Driven Analysis of Political Communication During the 2023–2024 Israel–Hamas War (under review)
This study examines affective polarization in user responses to short-form news content published by state-funded media outlets. The analysis reveals systematic differences in sentiment toward key geopolitical and ideological entities, with disproportionately negative evaluations of Israel and Zionism, contrasted with more positive or sympathetic evaluations of Palestine and Palestinians. These patterns indicate polarized evaluative dynamics in audience reactions and discuss how affective narratives and emotionally charged language shape political communication in Web 2.0 environments.
Link: (forthcoming)
Study 2: Analyzing Polarization in Online Discourse on the 2023–2024 Israel–Hamas War
Proceedings of the 21st Conference on Natural Language Processing (KONVENS 2025): Workshops, pages 7–16, Hannover, Germany.
Download paper: https://aclanthology.org/2025.konvens-2.1/
This study applies large-scale sentiment analysis to track longitudinal changes in user attitudes over the course of one year. The analysis shows that aggregate sentiment trends reflect reactions to geopolitical developments, while also demonstrating that meaningful interpretation requires pairing automated analysis with domain expertise.
Study 1: Investigating Polarization in YouTube Comments via Aspect‑Based Sentiment Analysis
Proceedings of the 15th International Conference on Recent Advances in Natural Language Processing – NLP in the Generative AI Era (RANLP 2025).
Download paper: https://aclanthology.org/2025.ranlp-1.83/
This study uses aspect‑based sentiment analysis (ABSA) to analyze polarization in online discourse based on more than three million user comments and replies from YouTube Shorts. A manually annotated subset is used to train and evaluate a domain‑adapted ABSA model. The results show that fine‑grained sentiment analysis provides reliable insights into how politically and ideologically charged topics are discussed over time and across outlets.
Further Research Outputs
The project is in the process of releasing a curated dataset that supports replication and enables further research on short‑form news coverage and online discourse.
Status: Documentation and schema finalization in progress.
Contact & Inquiries
This research project is led by Daniel Miehling (PhD).
- Email: damieh@iu.edu
- GitHub: https://damieh1.github.io/
If you use or cite this project, a short note is always greatly appreciated.