Offensive Language Detection on Video Live Streaming Chat

Gao, Zhiwei; Yada, Shuntaro; Wakamiya, Shoko; Aramaki, Eiji

doi:10.18653/v1/2020.coling-main.175

Cited by 13 publications

(6 citation statements)

References 6 publications

(7 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Significant efforts in the domain of abuse detection have been devoted to abusive text [5,6], image [11], video [3,4] and comments [7,8]. Existing approaches that are employed to perform abuse detection vary from using lexical features together with meta data [12], to machine/deep learning models, and their derivatives [13,14].…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Multilingual and Multimodal Abuse Detection

Sharon¹,

Shah²,

Mukherjee³

et al. 2022

Preprint

View full text Add to dashboard Cite

The presence of abusive content on social media platforms is undesirable as it severely impedes healthy and safe social media interactions. While automatic abuse detection has been widely explored in textual domain, audio abuse detection still remains unexplored. In this paper, we attempt abuse detection in conversational audio from a multimodal perspective in a multilingual social media setting. Our key hypothesis is that along with the modelling of audio, incorporating discriminative information from other modalities can be highly beneficial for this task. Our proposed method, MADA, explicitly focuses on two modalities other than the audio itself, namely, the underlying emotions expressed in the abusive audio and the semantic information encapsulated in the corresponding textual form. Observations prove that MADA demonstrates gains over audio-only approaches on the ADIMA dataset. We test the proposed approach on 10 different languages and observe consistent gains in the range 0.6%-5.2% by leveraging multiple modalities. We also perform extensive ablation experiments for studying the contributions of every modality and observe the best results while leveraging all the modalities together. Additionally, we perform experiments to empirically confirm that there is a strong correlation between underlying emotions and abusive behaviour. 1

show abstract

Section: Related Workmentioning

confidence: 99%

“…Abuse detection in visual [3,4] and textual [5,6,7,8] domains has been well explored. However, abuse detection in spoken audio is relatively under explored.…”

Section: Introductionmentioning

confidence: 99%

Multilingual and Multimodal Abuse Detection

Sharon¹,

Shah²,

Mukherjee³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Work has also been undertaken on data from comments on news media (Glavaš et al, 2020;Razo and Kübler, 2020;Wang et al, 2020;Zampieri et al, 2020), chatrooms and discussion forums (Gao et al, 2020;Wang et al, 2020), Wikipedia discussions (Fortuna et al, 2020;Glavaš et al, 2020;Gröndahl et al, 2018;Nejadgholi and Kiritchenko, 2020;Pavlopoulos et al, 2020), and message services such as WhatsApp (Saha et al, 2021).…”

Section: Related Workmentioning

confidence: 99%

ConvAbuse: Data, Analysis, and Benchmarks for Nuanced Abuse Detection in Conversational AI

Curry,

Abercrombie,

Rieser

2021

Preprint

View full text Add to dashboard Cite

We present the first English corpus study on abusive language towards three conversational AI systems gathered 'in the wild': an opendomain social bot, a rule-based chatbot, and a task-based system. To account for the complexity of the task, we take a more 'nuanced' approach where our ConvAI dataset reflects fine-grained notions of abuse, as well as views from multiple expert annotators. We find that the distribution of abuse is vastly different compared to other commonly used datasets, with more sexually tinted aggression towards the virtual persona of these systems. Finally, we report results from bench-marking existing models against this data. Unsurprisingly, we find that there is substantial room for improvement with F1 scores below 90%.Warning: This paper contains examples of language that some people may find offensive or upsetting.

show abstract

“…In the field of linguistic studies and discourse analysis, semiotics has been studied in several contexts of live-streaming videos. Text communication and conversation [17]- [19] and viewer comments [20], [21] have been analyzed in the live streaming video. Visual and spatial analytics, techniques, and algorithms of cognitive artificial intelligence, as well as immersive digital simulations, have also been examined in live shopping within the metaverse [22].…”

Section: Introductionmentioning

confidence: 99%

A Semiotic Analysis of the Hosts’ Appearance and Product Displays in Live Shopping

Survey Sijabat,

Margaretha Febriany Narahawarin,

Nurul Istiqomah

2023

TSSJ

View full text Add to dashboard Cite

This study aims to analyze the host's appearance and product display in live shopping from a semiotic perspective. The research method used was qualitative with observation techniques and in-depth interviews. The data were obtained by observing live shopping broadcasts and conducting interviews with 30 live shopping viewers. The data were analyzed using semiotic analysis and qualitative data analysis techniques. The findings showed that the hosts’ appearance and product display in live shopping has a complex meaning that relies on the social and cultural context. The use of filters on social media and technology on smartphones influences how hosts and products are displayed in live shopping, which changes how the audience sees the products being sold. The audience's attention and purchase decisions may be influenced by the hosts' and the product's appearance. Additionally, the appearance created by social media filters and smartphone technology can influence the brand image and aesthetic value of the offered products.

show abstract

Offensive Language Detection on Video Live Streaming Chat

Cited by 13 publications

References 6 publications

Multilingual and Multimodal Abuse Detection

Multilingual and Multimodal Abuse Detection

ConvAbuse: Data, Analysis, and Benchmarks for Nuanced Abuse Detection in Conversational AI

A Semiotic Analysis of the Hosts’ Appearance and Product Displays in Live Shopping

Contact Info

Product

Resources

About