DerivedFunction/derivative-type-classifier-dataset

这是一个合成衍生品分类数据集，包含金融披露的文本段落和代表衍生品信息类别的整数标签。数据集旨在解决金融领域标注数据不足的问题，通过合成数据生成技术创建了一个大而多样化的数据集，用于训练和评估NLP模型。数据集没有进行划分，数据是使用Python代码合成的，基于模板和规则自动标注。虽然数据是合成的，可能无法完全捕捉现实世界金融披露的复杂性和细微差别，但经过训练的模型在部署前应在现实世界数据上进行评估。

This is a synthetic derivative classification dataset containing text paragraphs of financial disclosures and integer labels representing categories of derivative information. The dataset is created to address the limited availability of labeled data in the financial domain, using synthetic data generation to create a large and diverse dataset for training and evaluating NLP models. The dataset is not split and the data is synthetically generated using Python code, labeled automatically based on templates and rules. Although the data is synthetic and may not fully capture the complexity and nuances of real-world financial disclosures, models trained on this data should be evaluated on real-world data before deployment.