GenChaR
收藏资源简介:
GenChaR数据集是由香港理工大学创建的一个用于股票图表描述任务的新数据集。该数据集包含1972对图表-描述对,旨在通过自动生成描述文本,帮助理解股票市场的动态。数据集的内容主要来源于ElliottWave-Forecast公司发布的分析文章,经过筛选和清理后,确保每对图表-描述对的质量。数据集的创建过程包括样本收集、图表注释和数据清洗,确保数据的准确性和可用性。该数据集主要应用于金融领域的技术分析,旨在通过图像描述生成技术,为投资者提供更直观的市场分析和决策支持。
The GenChaR dataset is a novel dataset developed by The Hong Kong Polytechnic University for stock chart description tasks. It contains 1972 chart-description pairs, aiming to facilitate the understanding of stock market dynamics via automatically generated descriptive texts. The content of the dataset is primarily sourced from analytical articles published by ElliottWave-Forecast, which underwent filtering and cleaning procedures to guarantee the quality of each chart-description pair. The dataset creation process includes sample collection, chart annotation, and data cleaning, ensuring the accuracy and usability of the data. This dataset is mainly applied to technical analysis in the financial sector, with the goal of providing investors with more intuitive market analysis and decision support through image captioning technology.




