Data analysis, preprocessing, and implementing models such as Text-CNN, GRU, and PhoBERT, tailored for Vietnamese datasets. The project integrates multiple datasets, including ViHSD, to enhance model performance and provide robust detection capabilities.