Antisemitism after 10/7 in posts on X containing the keyword "Israel" (November–December 2023)

A Dataset by the Social Media & Hate Research Lab, Institute for the Study of Contemporary Antisemitism (ISCA), Borns Jewish Studies Program. DATASET OVERVIEW This dataset contains 737 English-language tweets mentioning “Israel,” collected between November and December 2023. It includes topics related to Jews, Israel, and antisemitism, with dual annotations for both antisemitic content and tweets calling out antisemitism. DATA COLLECTION METHODOLOGY Data were collected from the Twitter/X Archive using the keyword “Israel” during November and December 2023. Due to API restrictions, we collected 100 posts three times per day. Technical issues with our data collection system resulted in occasional downtime; however, we maintained consistent data collection for the first three weeks of both months (November 1–21 and December 1–21). November 2023: 4,369 posts collected; 500 randomly sampled for analysis.December 2023: 4,792 posts collected; 500 randomly sampled for analysis. After excluding deleted tweets or unavailable accounts: November 2023 included 381 tweets, of which 115 were classified as antisemitic (30.18%).December 2023 included 356 tweets, of which 99 were classified as antisemitic (27.81%). In total, 737 tweets were annotated, and 214 (29.04%) were classified as antisemitic. Additionally, 20 tweets in November (5.25%) and 24 in December (6.74%) were annotated as calling out antisemitism. ANNOTATION PROCESS Annotation was conducted via the AnnotHate Portal (annotate.osome.iu.edu; see Jikeli, Soemer, & Karali, 2024), which allows annotators to review posts in full conversational and visual context, including images, threads, and quoted material. Each tweet was annotated by two independent annotators affiliated with the Social Media & Hate Research Lab at Indiana University. Annotators underwent extensive training in discourse analysis, antisemitism typology, and multimodal contextualization. A second annotation round was conducted to specifically identify tweets that call out or condemn antisemitism, aiming to reduce false positives in deep-learning and hate-speech detection models. FILE DESCRIPTION The dataset is provided in CSV format, where each row represents a single tweet. Columns include: id: Unique identifier for each tweet text: Full, unprocessed tweet text created_at: Timestamp of publication Antisemitism: Binary label indicating whether the tweet is antisemitic (1) or not (0) calling_out: Binary label indicating whether the tweet explicitly calls out or condemns antisemitism (1) or not (0) ACCESS TO FULL DATASET In accordance with Twitter/X data-sharing policies, the public release includes only tweet IDs and annotations. We are happy to provide the full dataset, including tweet text, upon request for non-commercial research purposes.For access inquiries, please contact us via mail. ACKNOWLEDGEMENTS We gratefully acknowledge the Advanced Cyberinfrastructure Curriculum Fellows Program for their support, with special thanks to Yu Ma and Tony Walker for their exceptional resources and assistance.This work utilized Jetstream2 at Indiana University through allocation HUM200003 from the Advanced Cyberinfrastructure Coordination Ecosystem: Services & Support (ACCESS) program, supported by the U.S. National Science Foundation (grants #2138259, #2138286, #2138307, #2137603, and #2138296). REFERENCES Jikeli, Gunther, Soemer, Katharina & Karali, Sameer (2024). Annotating live messages on social media. Testing the efficiency of the AnnotHate – live data annotation portal. Journal of Computational Social Science 7, 571–585.