SaguaroCapital/sentiment-analysis-in-commodity-market-gold

这是一个关于商品市场新闻情感分析的数据集，包含了2000年至2021年间的10,000多条新闻标题，这些标题被手动注释为多个维度上的不同类别。数据集由Ankur Sinha和Tanmay Khandait整理，并在Kaggle上发布，遵循CC BY-NC-ND 4.0许可。数据集的主要用途是情感分类，特别是对商品（如黄金）价格方向、资产比较、过去和未来信息以及价格情感的分析。数据集的结构包括日期、URL、新闻标题以及多个分类标签，如价格方向、资产比较、过去信息、未来信息和价格情感。数据集分为训练集和测试集，比例为80%/20%。

This is a news dataset for sentiment analysis in the commodity market, specifically for gold. The dataset contains over 10,000 news headlines spanning more than 20 years (2000-2021), manually annotated by experts across multiple dimensions including price direction, asset comparison, past and future information, and price sentiment. The purpose of the dataset is to help machine learning models understand commodity news, providing additional input for short-term and long-term price forecasting models, and creating news-based indicators for commodities. The dataset includes 10 columns and 10570 rows, divided into train and test sets in an 80%/20% ratio.