A Survey on Combating Hate Speech by Detection and Prevention
Abstract: With the rise in social media (SM) platforms that offer easy access, community formation, and online debate, the issue of hate speech has risen rapidly.
The hate detection, and countering it becomes a growing challenge to society, researchers, companies, and policymakers.
Hate speech is in the form of text or multimodal such as memes, GIFs, audio, or video.
The scientific study of hate speech from a computer science view has gained attention in recent years.
Mostly it is considered a supervised task where the annotated corpora and shared resources play a big role.
To combat it, SM, employing modern AI tools is getting attention.
This survey comprehensively examines the work done to combat hate in the English language so far.
This structures the state-of-the-art methodologies employed for unimodal identification, studies conducted in multimodal hate identification, the role of Explainable AI, prevention of hate speech through style transfer, and counter-narrative generation for the English language.
The efficacy and limitations are also discussed.
Compared with the earlier surveys this paper concisely gives a well-organized presentation of the methods to combat hate.
Paper Type: Long
Research Area: NLP Applications
Research Area Keywords: Hate speech detection, Multimodal meme identification, Style transfer, Counter Narrative generation
Contribution Types: Surveys
Languages Studied: English
Submission Number: 5603
Loading