Abstract: In this paper, we propose and evaluate a data augmentation technique for improving the classification of text data. Using a large language model (LLM), samples of real-world data and a ...