Detecting Harmful Memes and Their Targets

Pramanick, Shraman; Dimitrov, Dimitar; Mukherjee, Rituparna; Sharma, Shivam; Akhtar, Md. Shad; Nakov, Preslav; Chakraborty, Tanmoy

doi:10.48550/arxiv.2110.00413

Cited by 6 publications

(5 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Each word in a tweet was marked with either 0 or 1, where 1 means rationale. We have considered seven target classes (Religion, Sexual-Orientation, Attacking-Relatives-and-Friends, Organization, Community, Profession and Miscellaneous) as mentioned in [25] and [29]. Later expert annotators discussed each other and resolved the differences to create 300 gold standard samples with rationale and target annotations.…”

Section: Annotation Trainingmentioning

confidence: 99%

“Explain Thyself Bully”: Sentiment Aided Cyberbullying Detection with Explanation

Maity,

Jha,

Jain

et al. 2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Cyberbullying has become a big issue with the popularity of different social media networks and online communication apps. While plenty of research is going on to develop better models for cyberbullying detection in monolingual language, there is very little research on the code-mixed languages and explainability aspect of cyberbullying. Recent laws like "right to explanations" of General Data Protection Regulation, have spurred research in developing interpretable models rather than focusing on performance. Motivated by this we develop the first interpretable multi-task model called mExCB for automatic cyberbullying detection from code-mixed languages which can simultaneously solve several tasks, cyberbullying detection, explanation/rationale identification, target group detection and sentiment analysis. We have introduced BullyExplain, the first benchmark dataset for explainable cyberbullying detection in code-mixed language. Each post in BullyExplain dataset is annotated with four labels, i.e., bully label, sentiment label, target and rationales (explainability), i.e., which phrases are being responsible for annotating the post as a bully. The proposed multitask framework (mExCB) based on CNN and GRU with word and subsentence (SS) level attention is able to outperform several baselines and state of the art models when applied on BullyExplain dataset. 3 Disclaimer: The article contains offensive text and profanity. This is owing to the nature of the work, and do not reflect any opinion or stand of the authors.

show abstract

Section: Annotation Trainingmentioning

confidence: 99%

“Explain Thyself Bully”: Sentiment Aided Cyberbullying Detection with Explanation

Maity,

Jha,

Jain

et al. 2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…Each word in a tweet was marked with either 0 or 1, where 1 indicated it was a rationale. We considered seven target classes (Religion, Sexual-Orientation, Attacking-Relatives-and-Friends, Organization, Community, Profession, and Miscellaneous) as defined in (Mathew et al, 2020) and (Pramanick et al, 2021). Expert annotators engaged in discussions to resolve any differences and created 300 gold standard samples with rationale and target annotations.…”

Section: Data Annotationmentioning

confidence: 99%

GenEx: A Commonsense-aware Unified Generative Framework for Explainable Cyberbullying Detection

Maity,

Jain,

Jha

et al. 2023

Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing

View full text Add to dashboard Cite

With the rise of social media and online communication, the issue of cyberbullying has gained significant prominence. While extensive research is being conducted to develop more effective models for detecting cyberbullying in monolingual languages, a significant gap exists in understanding code-mixed languages and the need for explainability in this context. To address this gap, we have introduced a novel benchmark dataset named Bully-Explain for explainable cyberbullying detection in code-mixed language. In this dataset, each post is meticulously annotated with four labels: bully, sentiment, target, and rationales, indicating the specific phrases responsible for identifying the post as a bully. Our current research presents an innovative unified generative framework, GenEx, which reimagines the multitask problem as a text-to-text generation task. Our proposed approach demonstrates its superiority across various evaluation metrics when applied to the BullyExplain dataset, surpassing other baseline models and current state-of-the-art approaches. 1 Disclaimer: The article contains profanity, an inevitable situation for the nature of the work involved. These in no way reflect the opinion of authors.

show abstract

“…Notably, Facebook launched "The Hateful Meme Challenge, " releasing a dataset of over 10,000 memes annotated for hateful and non-hateful attributes [4]. Subsequently, further studies have emerged, exploring various applications of meme analysis, such as offensive content detection [26], sentiment analysis [22], and identification of victims and roles [3,15].…”

Section: Introductionmentioning

confidence: 99%

MATK: The Meme Analytical Tool Kit

Hee,

Kumaresan,

Hoang

et al. 2023

Proceedings of the 31st ACM International Conference on Multimedia

View full text Add to dashboard Cite

The rise of social media platforms has brought about a new digital culture called memes. Memes, which combine visuals and text, can strongly influence public opinions on social and cultural issues. As a result, people have become interested in categorizing memes, leading to the development of various datasets and multimodal models that show promising results in this field. However, there is currently a lack of a single library that allows for the reproduction, evaluation, and comparison of these models using fair benchmarks and settings. To fill this gap, we introduce the Meme Analytical Tool Kit (MATK), an open-source toolkit specifically designed to support existing memes datasets and cutting-edge multimodal models. MATK aims to assist researchers and engineers in training and reproducing these multimodal models for meme classification tasks, while also providing analysis techniques to gain insights into their strengths and weaknesses. To access MATK, please visit https://github.com/Social-AI-Studio/MATK. CCS CONCEPTS• Computing methodologies → Natural language processing; Computer vision representations.

show abstract

Detecting Harmful Memes and Their Targets

Cited by 6 publications

References 17 publications

“Explain Thyself Bully”: Sentiment Aided Cyberbullying Detection with Explanation

“Explain Thyself Bully”: Sentiment Aided Cyberbullying Detection with Explanation

GenEx: A Commonsense-aware Unified Generative Framework for Explainable Cyberbullying Detection

MATK: The Meme Analytical Tool Kit

Contact Info

Product

Resources

About