Detection and Classification of Sexist and Misogynist Posts and Rape Threats Against Women on Social Media Networks in Code-Mixed or Code-Switched Language.
The project will involve several stages, including data collection (code-mixed and codeswitched posts that are sexist, misogynist and delivering rape threats to women) and preprocessing, feature extraction, model training and evaluation, and model deployment. Classification of these posts based on Expert-opinion and AI (involving various machine learning algorithms, such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs), and evaluate their performance in terms of accuracy, precision, recall, and F1 score).
1- The automated scraper extracts the comments with keywords of curses from reddit.