NUIG at SemEval-2020 Task 12: Pseudo Labelling for Offensive Content Classification

Suryawanshi, Shardul; Arčan, Mihael; Buitelaar, Paul

doi:10.18653/v1/2020.semeval-1.208

Cited by 19 publications

(43 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the coding scheme, we first presented examples and then asked the coders to annotate to what extent each video used cartoons using a 4-point Likert-type scale (0 = never ; 1 = a few ; 2 = many ; 3 = almost always ). We included both Internet memes and animated GIFs as memes (Suryawanshi et al, 2020; Tuters & Hagen, 2020), and the coders also saw examples of memes for reference before annotating the use of memes using the same 4-point Likert-type scale. For funny sounds, we presented the coders with a list of funny sounds 8 in the training phase.…”

Section: Methodsmentioning

confidence: 99%

Unpacking Multimodal Fact-Checking: Features and Engagement of Fact-Checking Videos on Chinese TikTok (Douyin)

Shen

2023

Social Media + Society

View full text Add to dashboard Cite

As fact-checking videos increasingly circulate on video-sharing platforms, more research is needed to understand the prevalent features of such videos and how they are associated with audience engagement. Drawing from the literature on fact-checking, communication, marketing, and computer science, we identified eight audiovisual features as well as seven persuasive strategies that are most relevant to fact-checking videos. Using a hybrid video analysis framework combining both automated and manual content analysis, we examined 4,309 fact-checking videos on Douyin, the Chinese version of TikTok. We found that fact-checking videos on Douyin tended to have higher brightness, less cool color dominance, and faster tempo than non-fact-checking videos from the same accounts and Douyin Trending videos, and frequently used persuasive strategies like clickbait and humor. Through feature clustering, we established three types of fact-checking videos on Douyin—long storytelling cartoons, short stimulating videos, and short authoritative videos. We found that several audiovisual features and persuasive strategies were associated with audience engagement, such as likes, comments, and reshares. This study sheds light on the common practices of fact-checking videos in Chinese cyberspace, extends the current image-as-data paradigm to fact-checking videos, and helps fact-checkers make evidence-based decisions on content creation.

show abstract

Section: Methodsmentioning

confidence: 99%

Unpacking Multimodal Fact-Checking: Features and Engagement of Fact-Checking Videos on Chinese TikTok (Douyin)

Shen

2023

Social Media + Society

View full text Add to dashboard Cite

show abstract

“…The results show that the model's performance has increased by around 4% with the addition of the additional collection of features. Suryawanshi et al (2020) created a multimodal dataset on the 2016 U.S. presidential election that included 743 offensive and nonoffensive memes. They combined the multimodal features using the early fusion strategy.…”

Section: Related Workmentioning

confidence: 99%

Hate speech detection in social media: Techniques, recent trends, and future challenges

Rawat,

Kumar,

Samant

2024

WIREs Computational Stats

View full text Add to dashboard Cite

The realm of Natural Language Processing and Text Mining has seen a surge in interest from researchers in hate speech detection, leading to an increase in related studies. This analysis aims to create a valuable resource by summarizing the methods and strategies used to combat hate speech in social media. We perform a detailed review to achieve a deep knowledge of the hate speech detection landscape from 2018 to 2023, revealing global incidents of hate speech in 2022–2023. Sixty‐six relevant articles were selected for this review. Existing studies were analyzed and categorized into five method categories: Machine Learning, Deep Learning, Ensemble models, Graph Neural Networks, and Graph Convolutional Networks. These advancements can aid social networking services in identifying hate messages before being posted, reducing the risk of harassment. The review also covers available hate speech datasets and highlights research challenges, but it is clear that a definitive solution to this problem is yet to be found. Future research directions are recommended to address the ongoing challenges in Hate Speech Detection.This article is categorized under: Applications of Computational Statistics > Computational Linguistics Statistical Learning and Exploratory Methods of the Data Sciences > Knowledge Discovery Statistical Learning and Exploratory Methods of the Data Sciences > Classification and Regression Trees (CART) Statistical Learning and Exploratory Methods of the Data Sciences > Text Mining

show abstract

“…Hateful meme classification is an emerging multimodal task made popular by the availability of several recent hateful memes datasets (Kiela et al, 2020;Suryawanshi et al, 2020;Gomez et al, 2020). For instance, Facebook had organized the Hateful Memes Challenge, which encouraged researchers to submit solutions to perform hateful memes classification (Kiela et al, 2020).…”

Section: Hateful Meme Detectionmentioning

confidence: 99%

“…Existing studies have explored classic twostream models that combine the text and visual features learned from text and image encoders using attention-based mechanisms and other fusion methods to perform hateful meme classification (Zhang et al, 2020;Kiela et al, 2020;Suryawanshi et al, 2020). Another popular line of approach is finetuning large scale pre-trained multimodal models for the task (Lippe et al, 2020;Zhu, 2020;Zhou and Chen, 2020;Muennighoff, 2020;Velioglu and Rose, 2020;Pramanick et al, 2021b;Hee et al, 2022).…”

Section: Hateful Meme Detectionmentioning

confidence: 99%

“…Memes are presented as images with accompanying text, which are usually intended to be funny or satirical in nature. However, malicious online users generate and share hateful memes under the guise of humor (Kiela et al, 2020;Pramanick et al, 2021a;Gomez et al, 2020;Suryawanshi et al, 2020). Many hateful memes had been created to attack and ridicule people based on specific characteristics such as race, ethnicity, and religion.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Prompting for Multimodal Hateful Meme Classification

Cao¹,

Lee²,

Chong³

et al. 2023

Preprint

View full text Add to dashboard Cite

Hateful meme classification is a challenging multimodal task that requires complex reasoning and contextual background knowledge. Ideally, we could leverage an explicit external knowledge base to supplement contextual and cultural information in hateful memes. However, there is no known explicit external knowledge base that could provide such hate speech contextual information. To address this gap, we propose PromptHate, a simple yet effective prompt-based model that prompts pre-trained language models (PLMs) for hateful meme classification. Specifically, we construct simple prompts and provide a few in-context examples to exploit the implicit knowledge in the pre-trained RoBERTa language model for hateful meme classification. We conduct extensive experiments on two publicly available hateful and offensive meme datasets. Our experimental results show that PromptHate is able to achieve a high AUC of 90.96, outperforming state-of-the-art baselines on the hateful meme classification task. We also perform fine-grained analyses and case studies on various prompt settings and demonstrate the effectiveness of the prompts on hateful meme classification.

show abstract

NUIG at SemEval-2020 Task 12: Pseudo Labelling for Offensive Content Classification

Cited by 19 publications

References 18 publications

Unpacking Multimodal Fact-Checking: Features and Engagement of Fact-Checking Videos on Chinese TikTok (Douyin)

Unpacking Multimodal Fact-Checking: Features and Engagement of Fact-Checking Videos on Chinese TikTok (Douyin)

Hate speech detection in social media: Techniques, recent trends, and future challenges

Prompting for Multimodal Hateful Meme Classification

Contact Info

Product

Resources

About