An Evaluation of Automation on Misogyny Identification(AMI) and Deep-Learning Approaches for Hate Speech -Highlight on Graph Convolutional Networks and Neural Networks
Kedi Li · 2022 International Conference on Computers, Information Processing and Advanced Education (CIPAE) · 2022
The increasing misogynistic perspectives and the popularized abusive languages toward women appeared on social media worldwide in recent decades. People took advantages from the anonymous accounts spreading hatred and linguistic violence toward women, especially marginalized female groups. To indent this issue from a holistic lens and detect misogynistic aggression during this pandemic recession period, researchers specifically categorized abusive languages into several stratified groups, and utilized deep learning techniques to distinguish the differences between normal languages and unfriendly languages, with the creation of data-sets. To clarify the target, the paper will strictly target Spanish, English, and Hindi languages, and the social media users are concentrated on twitter and Facebook. Most of the research relies on systems of misogyny identification (AMI) and Natural Languages Processing technology (NLP). Besides, this paper will address external explanation and demonstration on Graph attention networks and Semi-supervised classification with graph convolutional networks. To interpret to what extent the misogyny issue can be explained by deep learning and relevant subfields, this paper will further provide a more insightful comprehension for the audience with understanding in depth.