Automatic Detection of Multilingual Misogynistic Content in Social Media Data Based on Machine Learning Approach
Pravin D. Kaware, Anjali B. Raut · 2024
Global communication has changed due to social media's rapid ascent. Because social media is a global stage where people from all walks of life may communicate in their own language, it is important to tackle misogynistic content detection on these platforms in a multilingual manner. Hence, this paper presents the development of an automatic detection of misogynistic content in social media data that uses machine learning (ML) and natural language processing (NLP) techniques. The model is based on a multilingual annotated corpus of aggressive and sexist content in Hindi and Indian English based on TRAC-2 benchmark dataset. To train machine learning classifiers, the four ML methods are utilized in this research based on hybrid linguistic features. Finally, test results were assessed to construct the best classifier model in both languages. The proposed methodology outperforms than the existing state of the art model, demonstrating it successfully handles social media content.