While there is increasing global attention to data privacy, most of their current theoretical understanding is based on research conducted in a few countries. Prior work argues that people's cultural backgrounds might shape their privacy concerns; thus, we could expect people from different world regions to conceptualize them in diverse ways. We collected and analyzed a large-scale dataset of tweets about the #CambridgeAnalytica scandal in Spanish and English to start exploring this hypothesis. We employed word embeddings and qualitative analysis to identify which information privacy concerns are present and characterize language and regional differences in emphasis on these concerns. Our results suggest that related concepts, such as regulations, can be added to current information privacy frameworks. We also observe a greater emphasis on data collection in English than in Spanish. Additionally, data from North America exhibits a narrower focus on awareness compared to other regions under study. Our results call for more diverse sources of data and nuanced analysis of data privacy concerns around the globe.

The spread of hate speech and hateful imagery on the Web is a significant problem that needs to be mitigated to improve our Web experience. This work contributes to research efforts to detect and understand hateful content on the Web by undertaking a multimodal analysis of Antisemitism and Islamophobia on 4chan's /pol/ using OpenAI's CLIP. This large pre-trained model uses the Contrastive Learning paradigm. We devise a methodology to identify a set of Antisemitic and Islamophobic hateful textual phrases using Google's Perspective API and manual annotations. Then, we use OpenAI's CLIP to identify images that are highly similar to our Antisemitic/Islamophobic textual phrases. By running our methodology on a dataset that includes 66M posts and 5.8M images shared on 4chan's /pol/ for 18 months, we detect 573,513 posts containing 92K Antisemitic/Islamophobic images and 246K posts that include 420 hateful phrases. Among other things, we find that we can use OpenAI's CLIP model to detect hateful content with an accuracy score of 0.84 (F1 score = 0.58). Also, we find that Antisemitic/Islamophobic imagery is shared in 2x more posts on 4chan's /pol/ compared to Antisemitic/Islamophobic textual phrases, highlighting the need to design more tools for detecting hateful imagery. Finally, we make publicly available a dataset of 420 Antisemitic/Islamophobic phrases and 92K images that can assist researchers in further understanding Antisemitism/Islamophobia and developing more accurate hate speech detection models.

No abstract

The spread of hate speech and hateful imagery on the Web is a significant problem that needs to be mitigated to improve our Web experience. This work contributes to research efforts to detect and understand hateful content on the Web by undertaking a multimodal analysis of Antisemitism and Islamophobia on 4chan’s /pol/ using OpenAI’s CLIP. This large pre-trained model uses the Contrastive Learning paradigm. We devise a methodology to identify a set of Antisemitic and Islamophobic hateful textual phrases using Google’s Perspective API and manual annotations. Then, we use OpenAI’s CLIP to identify images that are highly similar to our Antisemitic/Islamophobic textual phrases. By running our methodology on a dataset that includes 66M posts and 5.8M images shared on 4chan’s /pol/ for 18 months, we detect 173K posts containing 21K Antisemitic/Islamophobic images and 246K posts that include 420 hateful phrases. Among other things, we find that we can use OpenAI’s CLIP model to detect hateful content with an accuracy score of 0.81 (F1 score = 0.54). By comparing CLIP with two baselines proposed by the literature, we find that CLIP outperforms them, in terms of accuracy, precision, and F1 score, in detecting Antisemitic/Islamophobic images. Also, we find that Antisemitic/Islamophobic imagery is shared in a similar number of posts on 4chan’s /pol/ compared to Antisemitic/Islamophobic textual phrases, highlighting the need to design more tools for detecting hateful imagery. Finally, we make available (upon request) a dataset of 246K posts containing 420 Antisemitic/Islamophobic phrases and 21K likely Antisemitic/Islamophobic images (automatically detected by CLIP) that can assist researchers in further understanding Antisemitism and Islamophobia.