Determining Code Words in Euphemistic Hate Speech Using Word Embedding Networks

Magu, Rijul; Luo, Jiebo

doi:10.18653/v1/w18-5112

Cited by 39 publications

(44 citation statements)

References 10 publications

(5 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The automatic identification of hate speech has been mostly formulated as a natural language processing problem (e.g. Mishra et al, 2018;Gunasekara and Nejadgholi, 2018;Kshirsagar et al, 2018;Magu and Luo, 2018;Sahlgren et al, 2018). The signal from text, however, sometimes is not sufficient for determining whether a piece of content (such as a post) on the social network platforms constitutes hate speech.…”

Section: Introductionmentioning

confidence: 99%

Exploring Deep Multimodal Fusion of Text and Photo for Hate Speech Classification

Yang¹,

Xu²,

Ghosh³

et al. 2019

Proceedings of the Third Workshop on Abusive Language Online

View full text Add to dashboard Cite

Interactions among users on social network platforms are usually positive, constructive and insightful. However, sometimes people also get exposed to objectionable content such as hate speech, bullying, and verbal abuse etc. Most social platforms have explicit policy against hate speech because it creates an environment of intimidation and exclusion, and in some cases may promote real-world violence. As users' interactions on today's social networks involve multiple modalities, such as texts, images and videos, in this paper we explore the challenge of automatically identifying hate speech with deep multimodal technologies, extending previous research which mostly focuses on the text signal alone. We present a number of fusion approaches to integrate text and photo signals. We show that augmenting text with image embedding information immediately leads to a boost in performance, while applying additional attention fusion methods brings further improvement.

show abstract

Section: Introductionmentioning

confidence: 99%

Exploring Deep Multimodal Fusion of Text and Photo for Hate Speech Classification

Yang¹,

Xu²,

Ghosh³

et al. 2019

Proceedings of the Third Workshop on Abusive Language Online

View full text Add to dashboard Cite

show abstract

“…Kulkarni et al (2015) also show that longitudinal analysis can be efficiently processed with this method, which can help researchers detect and understand the change in the meaning of different expressions (and the cultural change behind it). The latest publications imply that word embedding models can detect such smooth distinctions as the identification of euphemistic coded words used in hate speeches (Magu and Luo 2018), or the detection of sarcasm (Joshi et al 2016). Different computational methods, such as GloVe, fastText, and Word2vec, are available.…”

Section: Unsupervised Methodsmentioning

confidence: 99%

The Potential of Automated Text Analytics in Social Knowledge Building

Németh

Koltai

2021

Computational Social Sciences

View full text Add to dashboard Cite

There are still many sociologists who are skeptical of the findings of big data-based analysis of social-data, questioning the potential of this knowledge production and its contribution to the scientific discourse of sociology.The chapter shows that this tension can be addressed through the redefinition of the research methodological basis of sociology, by the organic incorporation of data science know-how into its methods; the combined application of qualitative and quantitative analysis; and, the use of knowledge-driven science instead of the data-driven approach.The theoretical, methodological, and topical pathways between traditional and computational sociology emerge gradually along the chapter, which also includes plenty of illustrative examples of research situated at the interplay between sociology and data science. As our overview shows, there are new possibilities for sociological research, which are, in some sense, just by-products of information science. We introduce recently developed methods, which can be applied to specific sociological problems outside the scope of business applications. We present sociological topics not yet studied in this area and show new insights the approach can offer to classical sociological questions. As our aim is to encourage sociologists to enter this field, we discuss the new methods on the base of the classic quantitative approach, using its concepts and terminology and addressing the question of how traditionally trained sociologists can acquire new skills.

show abstract

“…Euphemisms and dysphemisms have been studied in linguistics and related disciplines (e.g., (Allan and Burridge, 1991;Pfaff et al, 1997;Rawson, 2003;Allan, 2009;Rababah, 2014)), but they have received little attention in the NLP community. Magu and Luo (2014) recognized code words in "euphemistic hate speech" by measuring cosine distance between word embeddings. But their code words conceal references to hate speech rather than soften them (e.g., the code word "skypes" covertly referred to Jews), which is different from the traditional definition of euphemisms that is addressed in our work.…”

Section: Related Workmentioning

confidence: 99%

“…Euphemisms are related to politeness, which plays a role in applications involving dialogue and social interactions (e.g., (Danescu-Niculescu-Mizil et al, 2013)). Dysphemisms can include pejorative and offensive language, which relates to cyberbullying (Xu et al, 2012;Van Hee et al, 2015), hate speech (Magu and Luo, 2014), and abusive language (Park et al, 2018;Wiegand et al, 2018). Recognizing euphemisms and dysphemisms for controversial topics could be valuable for stance detection and argumentation in political discourse or debates (Somasundaran and Wiebe, 2010; Walker et al, 2012;Habernal and Gurevych, 2015).…”

Section: Introductionmentioning

confidence: 99%

Proceedings of the Second Workshop on Figurative Language Processing

2020

View full text Add to dashboard Cite

Detecting sarcasm and verbal irony is critical for understanding people's actual sentiments and beliefs. Thus, the field of sarcasm analysis has become a popular research problem in natural language processing. As the community working on computational approaches for sarcasm detection is growing, it is imperative to conduct benchmarking studies to analyze the current state-of-the-art, facilitating progress in this area. We report on the shared task on sarcasm detection we conducted as a part of the 2nd Workshop on Figurative Language Processing (FigLang 2020) at ACL 2020. Referenceskalaivani A and Thenmozhi D. 2020. Sarcasm identification and detection in conversion context using BERT. In Proceedings of the Second Workshop on Figurative Language Processing,

show abstract

Determining Code Words in Euphemistic Hate Speech Using Word Embedding Networks

Cited by 39 publications

References 10 publications

Exploring Deep Multimodal Fusion of Text and Photo for Hate Speech Classification

Exploring Deep Multimodal Fusion of Text and Photo for Hate Speech Classification

The Potential of Automated Text Analytics in Social Knowledge Building

Proceedings of the Second Workshop on Figurative Language Processing

Contact Info

Product

Resources

About