I focused on analyzing the "Stress Detection from Social Media Articles" dataset obtained from Kaggle. I utilized GPT models with three distinct techniques: Zero-Shot Prompting, Few-Shot Prompting, and Fine-tuning the GPT2 Model. These techniques were employed to effectively classify stress-related posts within the dataset, aiming to enhance understanding and detection of stress signals in social media discourse.
🗝 Keywords - LLMs for Text Classification